part 1 overview: the present state of veterinary vaccine

PART 1

Overview: the present state of veterinaryvaccine development

uctirn

of irnUn pro

Infectious disease continues to be one ofthe most important constraints on theefficient production of farm livestock inboth developing and developed countries.While vaccination and the therapeutic orprophylactic use of drugs both play animportant role in animal disease control,vaccination is increasingly being viewedas the more sustainable option. This viewis influenced not only by the potential thatvaccination offers for greater economicefficiency but also by the concerns thathave been raised about the selection ofdrug-resistant pathogens and the potentialharmful effects of drug residues in animalproducts and the environment. Vaccinationhas had a major impact on the control ofepidemic viral diseases of livestock suchas foot-and-mouth disease and rinderpest.However, there are many other importantdiseases for which efforts to developeffective methods of vaccination have beenunsuccessful.

The advent of recombinant DNAtechnology in the early 1980s createdexciting new opportunities to producevaccines based on the use of expressedproducts of cloned genes. However, only afew such vaccines have been successfullydeveloped. In retrospect, it can be seenthat the immediate expectations of the newtechnology were unrealistically high, giventhe limited knowledge of the immunologyof many of the target diseases and of howantigens are processed and recognized bythe immune system. In the interveningdecade there have been major advances inimmunology which, coupled with further

W.I. Morrison

unity by veteirp'.ylactic

developments in the application of DNAtechnology, now provide a strong con-ceptual framework for the rational devel-opment of new vaccines.

This chapter will consider recent de-velopments in immunology that arepertinent to understanding how theimmune system controls infectionsand will discuss their implications forcontemporary approaches to vaccinedevelopment.

APPROACHES TO VACCINE DEVELOPMENT

Most of the current veterinary vaccines arebased on the use of either killed organismsor their products or live attenuatedorganisms. The development of thesevaccines has not relied on knowledge ofthe immune responses that mediateimmunity. Significant advances have beenmade primarily by the development of newculture techniques, improved attenuationprocedures and better adjuvants. Whilethere is some scope for further efforts todevelop vaccines along these lines, thereare many diseases for which the moreempirical methods are unlikely to besuccessful.

Two main approaches to vaccine designcan be considered using modern moleculartechnologies, namely the production ofattenuated mutant organisms by deliber-ate molecular manipulation and theidentification of antigenic components ofpathogens that can be used to induceprotective immune responses (thesecomponents are known as subunitvaccines). Unlike the traditional vaccine

Vaccine manual 5

strategies, the ability to exploit these newapproaches to vaccine development isdependent on an intimate knowledge ofthe molecular structure of the targetpathogens and an understanding of themechanisms of immunity.

PROCESSING AND RECOGNITION OF

ANTIGENS

Immune effector mechanismsStudies in laboratory animal modelsystems have demonstrated that theimmune system can respond in a numberof different ways to control an infection.The type of response elicited by a pathogendepends largely on the nature of theorganism and its site of replication withinthe host. In general, organisms thatestablish systemic infections and replicateextracellularly are controlled by antibodyresponses.

Secretory antibody responses also playan important role in the control of infec-tions at mucosal surfaces. By contrast, cell-mediated immune responses are generallymore important in controlling organismswhich replicate intracellularly. The T lym-phocytes that participate in cell-mediatedimmune responses may exert their effectorfunction in a number of different ways.They may kill infected cells or releasecytokines, which inhibit growth of intra-cellular organisms or which recruit and.activate accessory cells such as macro-phages, to perform these functions. Indeed,a number of different mechanisms mayoperate against the same organism. Stimu-lation of T lymphocyte responses is alsoessential to provide help, again in the formof cytokines, for the production of antibodyby B lymphocytes. Hence, T lymphocyteshave a pivotal role in the induction ofvirtually all specific immune responses.The way in which antigen is processedand recognized by the immune systemdetermines the type of T cell response thatis induced.

Recognition of antigen by B and Tlymphocytes

The antigen recognition structures on Band T lymphocytes, namely immuno-globulin (Ig) and the T cell receptor (TCR)are both generated by a process of generearrangement whereby each of the twochains that make up the molecules isproduced by bringing together two or threevariable sequences, from a pre-existinglibrary of variable genes, with a constantsequence to form a functional gene (Cooperand Burrows, 1989; Davis and Bjorkman,1988). This mechanism, together with thefurther diversity created by a combinationof different variable regions in Ig heavyand light chains and TCR alpha and betachains facilitates the generation of a verylarge repertoire of B lymphocytes andT lymphocytes, each with a unique antigenrecognition specificity.

Despite the similarity in structure of Igand TCR molecules, B and T lymphocytesdiffer fundamentally in the way theyrecognize antigen. Immunoglobulins,whether on the surface of B lymphocytesor as secreted antibodies interact directlywith foreign antigen, usually in the formin which it is initially encountered by thehost, i.e. as an intact organism or asmolecules released from the organism.Recognition of antigens by antibody is,therefore, often dependent on the con-formational integrity of the molecules. Bycontrast, T lymphocytes only recognizeantigens after they have been degradedand presented on the surface of other cells(Brodsky and Guagliardi, 1991). Theseantigen-bearing cells rnay be cells infectedwith foreign organisms or "professional"antigen-presenting cells, such as macro-phages and dendritic cells, which haveingested antigen. The processed antigen,which is in the form of short peptides ofless than 20 amino acids, is associated withmajor histocompatibility complex (MHC)molecules on the surface of the antigen-

6 The induction of immunity by veterinary immunoprophylactics

presenting cell (Bodmer, 1984). The T cellreceptor does not react with antigen alonebut rather recognizes a combination of theantigenic peptide and the associated MHCmolecule (Townsend and Bodmer, 1989).

The special role of MHC moleculesThe MHC consists of a set of closely linkedgenes, many of which encode moleculesinvolved in antigen processing andpresentation (Trowsdale, 1993). There aretwo main types of MHC molecules, namelyclass I and class II. The most strikingfeature of these MHC molecules is the highdegree of polymorphism they displayamong individuals of a species (Bodmer,1984; Trowsdale, 1993). Class I moleculesare heterodimers composed of a poly-morphic heavy chain and a non-poly-morphic light chain (j32-microglobulin), thelatter encoded outside the MHC. They areexpressed on most cells of the body. ClassII molecules are also heterodimers, bothpolypeptides being encoded within theMHC, but their expression in healthyanimals is confined mainly to "profes-sional" antigen-presenting cells, namelymacrophages, dendritic cells and B lym-phocytes. In most mammalian speciesexamined, each class is encoded by twogene loci and, in each instance, both allelesare expressed.

While it has been recognized since theearly 1970s that class I and class IImolecules are involved in presentingantigen to T cells, the precise molecularand structural basis of presentation wasnot elucidated until the late 1980s. A keyevent was the resolution of the structure ofclass I molecules by X-ray crystallography(Bjorkman et al., 1987). This revealed aprominent cleft in the membrane-distalpart of the molecule, which subsequentstudies have shown to be the site of antigenbinding. A similar structure has beendescribed for class II MHC molecules(Brown et al., 1993; Stern et al., 1994)

although the antigenic peptides thatassociate with class II are longer (13 to 17amino acids) than those bound to class I(eight to ten amino acids). Much of thesequence polymorphism in MHCmolecules occurs in and around thepeptide-binding region and, although thisvariation does not affect the overallstructure of the peptide-binding groove, itresults in subtle differences that influencethe nature of the peptides that eachmolecule will bind. Thus, each individualclass I molecule tends to bind a differentrepertoire of peptides (Sette et al., 1987;Rothbard and Gefter, 1991), so that T cellsfrom animals expressing different MHCmolecules will often recognize differentepitopes from the same pathogen, and insome cases these epitopes may be ondifferent proteins.

Although this variation might be expect-ed to result in quantitative differences inthe immune response to pathogens, andhence differences in susceptibility todisease, there are relatively few well-documented examples of strong asso-ciations of MHC with susceptibility toinfectious disease in outbred species. Thisprobably reflects the large number ofpotential T cell epitopes in most pathogensand the fact that most animals in anoutbred population will be heterozygousand therefore will express several class Ior class II molecules. Variations in theepitopes that are selected in individualanimals may, however, affect the strainspecificity of the T cell response if some ofthe epitopes vary between pathogen strains(Vitiello and Sherman, 1983).

As already indicated, the TCR recognizesa combination of self MHC molecule andbound peptide. Since the region of theMHC molecule recognized by theTCR, i.e. the peptide-binding groove, ispolymorphic, each T cell will only re-cognize an antigenic peptide associatedwith a particular MHC molecule. This

Vaccine manual 7


phenomenon, known as MHC-restriction(Doherty, Blanden and Zinkernagel, 1976),has important practical implications forstudies of T cell responses in outbredanimals; T cells from one animalwill recognize antigen presented on theanimal's own antigen-presenting cells butwill not recognize the same antigen onpresenting cells from another animal of adifferent MHC phenotype.

Until recently, the capacity of a givenMHC molecule to bind a large number ofdifferent antigenic peptides was difficultto explain. However, studies carried outover the last four years, involving theisolation and sequencing of peptidesbound to class I molecules, have helped toresolve this issue. The heterogeneousmixture of peptides isolated from anindividual class I molecule were found tobe conserved at one or two residues,usually at positions 2 and 9, and theseconserved amino acids were shown to beessential for binding the peptides to therespective MHC molecule (Matsumura etal., 1992; Rammensee, Falk and Rotzschke,1993). Thus, the amino acids at thesepositions represent an MHC binding motif,the antigenic specificity of the peptidebeing determined by amino acids at otherpositions. Information on the binding motifof a particular MHC molecule can be usedto predict possible T cell epitopes withinproteins of known amino acid sequences.There is some evidence that similarcharacteristics determine the binding ofpeptides to class II molecules (Rudenskyal., 1992) although this has proved difficultto substantiate.

T cell subpopulations recognize antigenprocessed in different subcellam

compartmentsThe identification of the cell surfacemolecules, CD4 and CD8, as markers forthe two major subpopulations of T cells inmammals was followed by the dem-

onstration that CD4+ T cells recognizeantigen presented by class II MHCmolecules whereas CDS+ T cells recognizeantigen presented by class I MHCmolecules. These interactions areaccompanied by binding of the CD4 andCD8 molecules to conserved regions onthe presenting class II and class I MHCmolecules, respectively. CD4+ T cells wereshown to mediate help for B cell responsesand delayed-type hypersensitivityreactions whereas CD8+ T cells wereresponsible for cell-mediated cytotoxicityof virus-infected cells. However, for sometime the factors that determined whetheran antigen was presented by class I or classII MHC molecules were unclear. This wasresolved by studies which showed thatclass I and class II molecules bind peptidesgenerated within different subcellularcompartments of antigen-presenting cells.Antigens derived from organisms thatreplicate in the cytoplasm of cells aredegraded by proteases within the cytosoland the resultant antigenic peptides aretransported into the endoplasmic reti-culum where they associate with newlysynthesized class I MHC moleculesdestined for the cell surface (van Meek andNathenson, 1992). By contrast, organismsor proteins taken into antigen-presentingcells by phagocytosis or endocytosisundergo enzymatic degradation withinendosomes and associate within anendosomal compartment with newlysynthesised class II MHC moleculestransported from the Golgi apparatusbefore being expressed on the cell surface(Brodsky, 1992; Unanue, 1992).

These alternative routes of antigenprocessing are known as the endogenousand exogenous pathways, respectively.Clearly, processing of antigens by theendogenous pathway and presentation byclass I will be confined to organisms, suchas viruses and some bacteria and protozoa,that replicate intracellularly (in the

cytoplasm). The killing of such infectedcells by CD8+ T lymphocytes can occurearly in the replication cycle and thusrepresents an effective means of limitingmultiplication of the pathogens.

Heterogeneity of CD4+ T lymphocytefunction

Mosmanm et al. (1986), working withmouse T cell clones, described two typesof CD4+ T cell, termed Th1 and Th2, thatwere distinguished by the cytokines theyproduced. This and subsequent studiesestablished that activated Th1 cells secreteinterleukin 2 (IL-2) and interferon-y (IFN-y) but not IL-4, IL-5, IL-10 and IL-13,whereas the converse applies to activatedTh2 cells (Mosmann and Coffman, 1989).A similar dichotomy has been reported forhuman T cells (Weiranga et al., 1990).

Studies of murine models of parasiticinfections have proved invaluable inelucidating the biological significance ofthe differences in Th1 and Th2 cells.Infection of mice with Leishmania major, orimmunization with Leishmania antigens,can induce either Th1 or Th2 T cellresponses, depending on the strain ofmouse and route of immunization. Th1responses result in the control of infectionand immunity, whereas Th2 responses leadto enhanced disease (Liew, 1990). This isjust one of a growing number of examplesin which the outcome of infection isstrongly influenced by the cytokine profileof the responding T cells. The induction ofTh1 responses in mice is associated withthe activation of macrophages, the pro-duction of antibody of the IgG,,, isotypeand the detection of delayed-type hyper-sensitivity reactions, while Th2 responsesgive rise to eosinophilia and production ofantibody of the IgG, and IgE isotypes. Thelatter characteristics feature prominentlyin many helminth infections and there isevidence that Th2 responses are beneficialfor the control of enteric nematode

infections (Else and Grencis, 1991; Urbanet al., 1992).

Studies of the biological activities of thecytokines produced by Th1 and Th2 cellshave demonstrated strong cross-regulationoperating between the subsets: IFN-yproduced by Th1 cells inhibits theinduction of Th2 responses and both IL-4and IL-10 have inhibitory effects on theinduction of Th1 responses (Mosmann etal., 1991; Fitch et al., 1993). Thus, the initialinduction of a strong Th1 or Th2 responsewill tend to inhibit responses by thereciprocal subset. Nevertheless, in someinfections, notably Schistosoma mansoni inmice, an initial Th1 response is followedby a switch to a Th2 response (Pearce et al.,1991). The switch is believed to reflect thepresence of antigens in the parasite eggswith a strong propensity for inducing Th2responses. This may represent a deliberatestrategy by the parasite to favour survivalof the adult worms. The parameters thatdetermine whether an antigen willstimulate a strong Th1 or Th2 response arepoorly understood. However, there isevidence that the biochemical nature ofthe antigen and the type of antigen-presenting cell in which it is presented toT cells are important contributory factors.

IMPLICATIONS FOR VACCINE DEVELOPMENT

Induction of antibody responsesIn the early 1980s, a series of studiesdemonstrated that antibodies raisedagainst intact proteins recognized shortpeptide fragments of the proteins (Geyson,Meloen and Barteling, 1984). Theseobservations encouraged the belief that itwould be possible to use synthetic peptidesfor vaccination. A large number of studiesaimed at stimulating immunity withpeptides were undertaken, in whichanimals were immunized with syntheticpeptides representing B cell epitopes,conjugated either to other peptides fromthe same pathogen or to unrelated proteins,

Vaccine manual 9

to provide the necessary T helper cellepitopes. With a few exceptions, theseattempts at immunization were unsuc-cessful. In many instances, the syntheticpeptides completely failed to induceantibodies against the parent protein ororganism while, in other cases, antibodyresponses did occur but were at best onlypartially effective.

Subsequent studies of protein structurehave highlighted the fact that so-calledlinear epitopes have a degree of confor-mation and that this conformation maydiffer subtly from that adopted by therespective synthetic peptides. The processof conjugating a peptide to a carriermolecule may also affect the conformationof the peptide, resulting in antibodies oflow avidity for the pathogen in question.

One pathogen for which a degree ofsuccess has been achieved by immun-ization with a synthetic peptide is foot-and-mouth disease virus. Immunizationwith a peptide consisting of two linkedpeptide sequences representing residues141-158 and 200-213 of the VP1 capsidprotein resulted in a proportion of immun-ized cattle (Di Marchi et al., 1986). Struc-tural studies of the virus demonstrated thatthe 141-158 component of the peptidecorresponds to a superficial loop on thesurface of the virus particle (Acharya et al.,1989) and suggested that the peptidesuccessfully reproduces this loop structure.Nevertheless, it is still unclear whether thefailure to achieve protection in all animalsimmunized with the peptide was due tosubtle conformational differences in theantibody recognition site or to inadequa-cies in the helper T cell response.

Another factor that may limit the successof immunization with peptides is the useof an unrelated carrier protein to stimulateT cell help for antibody production. Noanamnestic T cell response will occurfollowing the challenge of immunizedanimals with the pathogen, and immunity

will therefore rely largely on the pre-existing antibody induced by vaccination.

The use of intact, purified or recombinantproteins for immunization can overcomesome of the problems associated withimmunization with peptides. Such proteinsare likely to contain several T cell epitopesand, if appropriately produced, shouldhave the correct conformation for recog-nition by antibodies. However, the latter isnot always true. For example, the integrityof antibody epitopes on individual viralcapsid proteins may be dependent on thestructural interaction with other proteincomponents of the capsid. This is the casefor several of the neutralizing epitopes onthe surface of foot-and-mouth diseasevirus; of the three capsid proteins only one(VP1) retains any immunogenicity follow-ing purification, and it is much lessimmunogenic than killed intact virus(Bachrach et al., 1975).

The methods by which recombinantproteins are produced can also affect theintegrity of epitopes recognized by anti-bodies. Thus, if glycosylation is requiredfor antibody recognition, production of theproteins in bacteria will be inappropriateand proteins produced in insect cells maybe defective because of differences in thesugar side chains added by these cellscompared with mammalian cells. Differ-ences in the folding of proteins producedin bacteria rnay also result in the disruptionof some B cell epitopes.

The need for nnultiple T cell epitopesSince T lymphocytes recognize smallprocessed fragments of antigens, theconformational structure of antigens isgenerally not a constraint for the inductionof T cell responses. However, becauseT cells from animals of different MHCtypes tend to recognize different peptidesequences within an antigen, immun-ization with short polypeptides containingonly one or two potential T cell epitopes is

10 The induction of immunity by veterinary immunoprophvlactics

likely to induce a response in only aproportion of animals. This will be trueboth for T helper cell responses forantibody production and for effector T cellresponses. This further strengthens theargument for using one or more intactproteins in subunit vaccines so that therewill be sufficient n.umbers of potentialT cell epitopes to ensure that the majorityof individuals within an outbred popu-lation will respond to the antigens. In somecircumstances, there may be a case forexcluding particular T cell epitopes from avaccine construct. For example, if anantigen contains a particularly dominantT cell epitope that is variable betweenstrains of a pathogen, exclusion of such anepitope might result in a response that isless strain-specific.

Constraints on stimulation of CD8+ T cellresponses

Because of the need for antigen to beprocessed by the endogenous pathwayfor recognition by CD8+ T cells, the immun-ization of animals with killed organismsor their component proteins generally failsto induce CDS+ T cell responses. Therefore,alternative antigen delivery systems mustbe considered when developing subunitvaccines required to stimulate CD8+ T cellresponses. These could include the use ofvirus vectors or vaccination with "naked"DNA, both of which result in theexpression of proteins within the cellcytosol. Recent studies have also providedevidence that the active component of theadjuvant saponin, when used to prepareantigen-complexed structures known asimmunostimulating complexes (ISCOMs),facilitates the transfer of antigen across cellmembranes and the induction of CD8+T cell responses (Takahashi et al., 1990).

Influencing the cytokine responseThe recognition that the cytokines pro-duced by CD4+ T cells to a large extent

determine their function and that antigenicstimulation may result in the activation ofT cells producing different patterns ofcytokines is of major importance whenconsidering immunization strategies. It is,therefore, desirable to know whether theprotective responses against the targetpathogens involve Th1 or Th2 CD4+ T cellresponses. However, since the early eventsin antigen processing that result in a biasto Thl or Th2 responses are still incom-pletely understood, strategies for preferen-tial induction of one or other response arenot yet well established. The use of adju-vants that give a bias in the response, forexample the induction of Th1 responsesby rnycobacteria, is one approach that canbe pursued. Experiments in mice involvingadministration of recombinant cytokinesor cytokine-specific antibodies at the timeof immunization have implicated IL-4 andIL-12 as promoters of Th2 or Th1 cellresponses, respectively (Swain et al., 1991;Locksley, 1993). These findings indicatethat, by administering cytokines withantigen or by including cytokine genes inmolecular vaccine constructs, it may bepossible to influence the cytokine profileof T cell responses.

More than one mechanism of immunityIt is becoming increasingly apparent thatimmunity against a given pathogen maybe achieved by alternative immune mech-anisms. This is clearly the case withcomplex protozoan parasites which under-go differentiation through several develop-mental stages that differ antigenically andreplicate in different cell types. Thus, withmalaria parasites, antibody against theinfective sporozoite stage can blockinfection, class I MHC-restricted cytotoxicT cell responses are generated against thehepatic intracellular stages and other, asyet poorly understood, cell-mediatedmechanisms operate against the intra-erythrocytic stages.

Vaccine manual11

Similarly, different mechanisms mayoperate against migratory helminthparasites. However, there is also evidencethat more than one immune mechanismmay be effective in the control of virusinfections. For example, it is well es-tablished that maternally derived antibodyprotects offspring from infection with themorbilliviruses (measles, rinderpest) andalso interferes with vaccination (Albrechtet al., 1977). Yet a proportion of cattlesuccessfully immunized against rinderpestvirus with a recombinant vaccinia virusexpressing the F glycoprotein were foundto produce little or no rinderpest-specificantibody (Yilma et al., 1988; Belsham et al.,1989), indicating that immunity must havebeen mediated by T cell responses. Similarresults have been obtained with infectiousbursal disease virus in chickens, using arecombinant avipox virus expressing theVP2 protein, although the immunityachieved with the recombinant wasincomplete (Bayliss et al., 1991).

Clearly, if sufficient antibody of appro-priate specificity and biological activity ispresent in an immunized animal to preventinfection with the respective virus, cell-mediated effector mechanisms will not berequired for protection. However, if someof the challenge virus escapes initialneutralization by antibody, it is likely thatcell-mediated immune responses will bebeneficial, if not essential, for clearance ofthe infection. This will be particularly truefor viruses that spread by cell to cellcontact. There may also be some flexibilityin the type of T cell response that isemployed. In mice immunized againstinfluenza A virus, the transfer of eitherCD4+ or CD8+ T cells into native recipientsresulted in the clearance of challenge virus,the important common feature being thecytokines produced by the two cell types(Lukacher et al., 1986). With viruses suchas that which causes foot-and-mouthdisease, and for which antibody responses

are undoubtedly of major importance inmediating immunity, T cell-mediatedresponses may provide an additionalmechanism for clearing the virus. Yet therole of T cell effector mechanisms in foot-and-mouth disease has been largelyignored.

These observations indicate that theinduction of highly effective immunitywith subunit vaccines may necessitate theinclusion of more than one antigeniccomponent in a vaccine and the use of anantigen delivery system that is effective atinducing both humoral and cell-mediatedimmune responses. The poor efficacy ofsome of the currently used killed vaccinesmay be due in part to their limited abilityto stimulate T cell-mediated componentsof the protective immune responses.

Pathogens that subvert host immuneresponses

A common objective in the developmentof a vaccine is to mimic the immuneresponses that occur during recovery fromnatural infection. While this is an ap-propriate approach for many pathogens,some organisms have evolved stratagemsfor modifying host immune responses inorder to establish persistent infections.Helminth and protozoan parasites havebeen particularly adept at developing avariety of escape mechanisms. One of thestrategies employed by parasites such asLeishmania sp. and Schistosoma sp. is todirect the T cell response to producecytokines that are inappropriate forparasite clearance (Sher et al., 1992). In thecase of Schistosoma rnansoni, this results inthe development of characteristic egggranulomas that are responsible for clinicaldisease. Nevertheless, schistosome-infected (SCID) mice, which are unable tomount a granulomatous response, developsevere hepatitis that is believed to be dueto release the of parasite proteases (Amiraet al., 1993). Thus, the "deviated" immune


Vaccine manual

response not only favours parasitepersistence but may also have a role inprotecting and ensuring the survival of thehost, albeit with some pathology. Anyapproach to designing a vaccine for suchparasites must be based on an under-standing of which components of the host'simmune response are responsible forprotection as well as on ensuring theavoidance of immune responses thatpotentiate disease.

Bloodsucking ectoparasites have devel-oped a number of mechanisms to avoidthe clotting of imbibed blood and tominimize adverse effects of inflammatorymediators released at the site of feeding. Inthe case of ticks, which feed continuouslyfor several days, significant hyper-sensitivity reactions are induced bysalivary proteins. While in previouslyexposed animals this results in a reductionin the number of ticks that engorge, suchanimals still carry significant tick burdens(Willadsen, 1980). Over the last ten years,researchers in Australia have developedan alternative strategy for vaccinationagainst the one-host tick Boophilus microplusbased on immunization with proteins fromthe tick gut, to which the host is notnormally exposed (Willadsen, McKennaand Riding, 1988; Willadsen et al., 1989).Antibody induced by these "concealed"antigens, when ingested by the tick, causesdamage to the gut wall and results notonly in tick mortality but also in a markedlyreduced fecundity of surviving ticks. Asimilar approach is also being pursued forvaccination against the bloodsuckingnematode Haemonchus contortus.

Several of the large DNA viruses haveevolved molecular mechanisms that couldpotentially modify antiviral immuneresponses. The vaccinia virus genomecontains a number of genes with homologyto host receptors for the cytokines IL-1,IL-6 and IFN-7. Some of these genes areexpressed as soluble proteins (virokines)

that are able to bind the respectivecytokines (Smith, 1993). There is alsoevidence that serine protease inhibitorsexpressed by vaccinia may inhibitintracytosolic processing of antigensdestined for association with class I MHC(Townsend et al., 1988). Several of theherpesviruses and adenoviruses have beenshown to express genes that in_hibit theassembly of class I MHC molecules (Lippéet al., 1991; Hill et al., 1994). Identificationof the precise role of these genes indetermining the virulence of the virusesand their immunogenicity is relevant notonly for understanding the pathogenesisof disease caused by the viruses but also inconsidering the use of animal poxvirusesand herpesviruses as vaccine vectors.

NEW APPROACHES TO VACCINE DESIGN

Molecularly defined attenuationThe simplest way of producing a vaccinethat mimics the immune responsesinduced by natural infection is to selectan attenuated mutant. The use of liveorganisms also has the advantage ofproviding longer-lasting immunity thancan be attained with killed antigen. Manyof the traditional vaccines are based on theuse of attenuated organisms that wereeither identified by chance or selected bythe prolonged culture of the organisms.Advances in knowledge of the molecularstructure of viruses and of the functionof individual genes in replication andassembly now provide the opportunity toproduce targeted mutations that result inaltered virulence. The deletion of wholegenes negates the possibility of reversionto virulence as a consequence of pointmutations. Moreover, viruses withmutations in several genes can beproduced. A strain of pseudorabies virus(Aujeszky's disease virus), in which thegenes encoding the GI glycoprotein andthymidine kinase have been disrupted, hasbeen shown to be avirulent and to

stimulate immunity against challenge withnative virus (Moormann et al., 1990). Thisvirus is being marketed as an attenuatedvaccine.

A similar approach is being used withbacteria by targeting genes involved inbacterial metabolism. The most thoroughlyinvestigated system is Salmonella spp., inwhich mutation of a number of genes hasbeen shown to result in attenuatedorganisms that retain immunogenicity(Dougan, Hormaeche and Maskell, 1987).Again, the incorporation of double or triplemutations minimizes the risk of reversionto virulence. The best characterized of theseare the aro mutants (Dougan et al., 1988), inwhich the mutations interrupt the pathwayfor biosynthesis of aromatic metabolitesresulting in organisms that are dependenton nutritional elements not available inmammalian tissues. Such aro mutants havebeen used successfully to vaccinate calvesagainst Salmonella typhimurium (Jones et al.,1991).

Immunization with recombinant proteinsThe use of purified recombinant proteinsfor vaccination can be considered wherespecific antigens have been identified asthe targets for protective antibody and / orCD4+ T cell responses. However, as alreadyindicated, the expression system used toproduce the protein should not alterimportant antibody recognition sites on theprotein. Generally, recombinant proteinsneed to be administered in a potentadjuvant.

The human hepatitis B vaccine, whichconsists of recombinant viral core protein,is the most notable success in this area(McAleer et al., 1984); immunity has beenachieved with antigen produced inEscherichia coli or in yeast. Recently, avaccine for Boophilus micro plus tickinfestation in cattle, based on theproduction in E. coli of an antigen that isnormally expressed in the tick gut, has

been put into commercial production inAustralia (Willadsen et al., 1989; Rand etal., 1989). A recombinant protein from thetapeworm Taenia ovis has been shown tobe effective at immunizing sheep againsttapeworm infestation (Johnson et al., 1989),and immunization of cattle with arecombinant sporozoite surface antigenfrom Theileria parva has been found toprotect a proportion of animals againstexperimental challenge (Musoke et al.,1992).

Production of virus capsidsThe capacity of the hepatitis B core proteinto assemble spontaneously into virus-likeparticles (McAleer el al., 1984) may partlyaccount for the success of this vaccine, sinceassembly should allow the protein to adopta conformation similar to that of the nativevirus particle.

In the case of non-enveloped viruses inwhich the capsid is composed of severalproteins, the surface conformation adoptedby each protein in the virus particle isdependent on interactions with the otherproteins. Thus, the isolated proteins areoften poorly immunogenic. This problemmay be overcome if non-infectious viralcapsids can be produced by expressing allof the capsid proteins in a single construct.This has been achieved with bluetonguevirus (Roy, 1992). Bluetongue virus, theprototype of the orbiviruses, is made upof seven proteins, three of which are with-in an inner core surrounded by twoconcentric protein layers formed by VP3and VP7. The remaining two proteins, VP2and VP5, are attached to the outer VP7layer. Empty virus-like particles have beenproduced by expressing the four mainstructural proteins (VP2, VP3, VP5 andVP7) in a single recombinant baculovirus.These particles closely resemble nativevirus particles in structure. Immunizationexperiments in sheep have shown that theyprotect against challenge with bluetongue


virus of the homologous serotype and thatthey are much more immunogenic thanpurified VP2 or VP5 proteins (Roy, Frenchand Erasmus, 1992).

Similar studies are under way with foot-and-mouth disease virus.

Live vectors

The potential use of viruses or bacteria aslive vectors for vaccination has been amajor focus of experimental investigationover the last ten years. Vectors may benaturally occurring apathogenic organismsor attenuated mutants produced by geneticmanipulation. Such vectors offer many ofthe advantages of live vaccines and canpotentially be used to express antigensfrom more than one pathogen. The virusvectors in particular provide a means ofinducing CD8+ T cell responses, which arenot readily generated by killed vaccines.

Initial studies of virus vectors werefocused on vaccinia, which had alreadybeen used with great success in smallpoxvaccination programmes and was knownto have a wide host range (Mackett andSmith, 1986). Being a poxvirus, it also hada large genome in which non-essentialregions could be identified for the insertionof foreign genes. Recombinant vacciniaviruses expressing viral genes have beenused successfully in experimental studiesto immunize cattle against rinderpest andvesicular stomatitis and dogs and foxesagainst rabies (Yilma et al., 1988; Belshamet al., 1989; Blancou et al., 1986). Fieldvaccination of foxes against rabies, usingbaits containing a vaccinia recombinant,has also been successful. Despite thesesuccesses and the demonstrated safety ofthe vaccines employed, concern has beenexpressed about the widespread release ofvaccinia now that smallpox vaccination hasstopped. Consequently, attention hasturned to animal poxviruses.

Capripox virus, which infects cattle,sheep and goats, is found throughout large

areas of Asia and Africa where it causessporadic outbreaks of disease. Anattenuated strain of the virus that had beenused locally in Africa to vaccinate againstthe disease (Kitching, Hammond andTaylor, 1987) has been developed as avector, and recombinant viruses expressingrinderpest virus glycoproteins have beenshown to protect cattle against challengewith rinderpest virus (Romero et al., 1993).A disadvantage of the capripox virus isthat it cannot be used outside the areaswhere infection is endemic owing to itsnotifiable status. Nevertheless, there aremany potential applications of this vectorin Africa and Asia.

Avipox and canarypox viruses have beendeveloped as potential vaccine vectors andthe former has been used successfully tovaccinate chickens against Newcastledisease and infectious bursal disease(Boursnell et al., 1990a and 1990b; Baylisset al., 1991). These viruses can also beconsidered as vectors for mammals, sincethey produce abortive infections inmammalian cells, i.e. the synthesis of virusproteins occurs but no infectious virus isproduced; there is therefore no risk of thevirus spreading to other animals.

The herpesviruses have large genomeswhich, like that of the poxviruses, canaccommodate large inserts of foreign DNA.The potential of a number of animalherpesviruses to serve as vaccine vectorsis currently being explored. These includepseudorabies virus, bovine herpesvirus 1,equine herpesvirus 1 and turkeyherpesvirus. Adenoviruses are also beinginvestigated, particularly in the context ofdelivering antigens to mucosal surfaces.

The coexpression of cytokine genes andforeign antigens in virus vectors offersa potential means of enhancing ormodulating immune responses to theexpressed antigens. Experiments withvaccinia virus in mice have yieldedpromising results with a number of

Vaccine manual 15

cytokines (Ramshaw et al., 1992). Thisapproach may be of particular value inproviding a bias towards Th1 or Th2 T cellresponses.

Attenuated mutant strains of severaldifferent Salmonella species have beenstudied extensively for their potential asvaccine vectors (Dougan, Hormaeche andMaskell, 1987; Dougan et al., 1988). Foreignantigens can be expressed from plasmidsor following the integration of the genesinto the bacterial genome. Much of thiswork has been carried out in mice and hasfocused on the induction of immuneresponses to other bacterial antigens andon the capacity to stimulate immunity atgut mucosal surfaces. Mice immunizedwith a Salmonella recombinant expressinga malaria circumsporozoite antigen werefound to be protected against malaria inthe absence of antibodies (Sadoff et al.,1988). Since this immunity is mediated byCD8+ T cells, this finding suggests thatSalmonella spp. may be able to introduceantigens into the endogenous processingpathway.

A variety of other bacteria are beinginvestigated as potential vaccine vectors.Because of its potent capacity to stimulateTh1 T cell responses, the mycobacteriumbacillus Calmette-Guérin (BCG) may haveparticular utility as a vector. A recombinantBCG expressing a surface antigen from theprotozoan parasite Leishmania major hasbeen shown to protect mice againstcutaneous leishmaniasis (Connell et al.,1993), a disease that is known to requireTh1 responses for immunity.

Immunization with nucleic acidsThe recent discovery that the injection ofDNA into animals can result in immuneresponses to proteins encoded by the DNAhas opened up a completely new approachto the development of subunit vaccines.Vaccination of mice with plasmids contain-ing the influenza haemagglutinin gene has

been shown to result in protection againstvirus challenge (Ulmer et al., 1993; Fynanet al., 1993). The injected genes appear tobe expressed within host cells withoutintegration of plasmid into chromosomalDNA. While studies of the transfectionefficiency of injected DNA have shown thatmuscle is 100 to 1 000 times more permis-sive than other tissues, protection has alsobeen achieved following inoculation by theintravenous, intranasal and intratrachealroutes and by delivering the DNA intra-dermally using a "gene gun". Theimmunization of mice with DNA inducesantibody responses as well as CD4÷ andCD8+ T cell responses. The induction ofimmune responses to bovine herpesvirus1 in cattle has been investigated in onestudy (Cox, Zamb and Babiuk, 1993); aplasmid containing the gII viral genestimulated neutralizing antibodies andresulted in a marked reduction in nasalshedding of the virus following challenge.

Targeting of antigen at nnucosal surfacesThe stimulation of protective immuneresponses at mucosal surfaces presents aparticularly challenging problem becauseof the requirement for such responses tobe induced locally. Non-viable antigenadministered orally is susceptible toproteolytic degradation and is absorbed inonly small amounts through the gut wall.Moreover, such "dietary" antigens tend toinduce tolerance rather than an activeimmune response. One way of overcomingthis problem is to use bacterial or viralvectors, such as Salmonella spp. oradenovirus, which replicate in thealimentary tract.

Other strategies that facilitate the uptakeof protein in the gut and the induction ofan active response are currently beinginvestigated. One of these involves the useof cholera toxin, which is a multimericprotein composed of a single A subunitand a pentameric B subunit; binding to

16 The induction of immunity by veterinary immttnoprophylactics

epithelial cells occurs by interaction of theB subunit with GMi-gangliosides whiletoxicity is mediated by ADP-ribosyl-transferase activity of the A subunit(Spangler, 1992). Oral administration ofeither subpathogenic doses of intact toxinor purified B subunits results in inductionof a specific immune response. Moreover,protein antigens administered along with,or covalently linked to, the toxin or its Bsubunit elicit strong antibody and CD4+ Tcell responses (Dertzbaugh and Elson,1993). Whether or not the B subunit retainsthe full adjuvant activity of the intact toxinhas not been fully resolved.

Another approach to delivering antigensto the gut is to encapsulate antigen inmicrospheres with an outer biodegradablepolymer coat that protects the antigenagainst rapid enzymatic degradation.Microspheres with a diameter of 5 to 10 p.mhave been shown to be taken up efficientlyby the Peyer's patches and to induce IgGand IgA antibody responses in mice.

CONCLUDING REMARKS

In the last decade, a clear picture hasemerged of how antigens are processedand recognized by the immune system andsignificant progress has been made inunderstanding the immune mechanismsthat operate against different types ofpathogens. This new knowledge, togetherwith further advances in technologies forgenetic manipulation, has led to a varietyof new approaches to the attenuation ofpathogenic organisms and to the design ofantigen delivery systems appropriate forinducing particular immune responses.There has also been progress in other areasof research, not discussed here, such as thetargeting of antigens at particular cell typesin the immune system and studies todevelop new adjuvants and understandtheir mode of action. These developmentsoffer exciting potential for the productionof a new generation of vaccines.

BIBLIOGRAPHY

Acharya, R., Fry, E., Stuart, D., Fox, G.,Rowlands, D. & Brown, F. 1989. Thethree-dimensional structure of foot-and-mouth disease virus at 2.9 A resolution.Nature, 337: 709-716.

Albrecht, P., Ennis, F.A., Saltzman, E.J. &Krugman, S. 1977. Persistence of internalantibody in infants beyond 12 months:mechanism of measles vaccine failure.J. Paediat., 91: 715-719.

Amira, P., Locksley, R.M., Parslow, T.,Sadick, M., Rector, E., Ritter, D. &McKerrow, J. 1993. Tumour necrosisfactor-7 restores granulomas and inducesegg laying in schistosome-infected(SCID) mice. Nature, 356 :604-607.

Bachrach, H.L., Moore, D.M., McKercher,P.D. & Polatnick, J. 1975. Immune andantibody responses to an isolated capsidprotein of foot-and-mouth disease virus.J. Immunol., 115: 1636-1641.

Bayliss, C.D., Peters, R.W., Cook, J.K.A.,Reece, R.L., Howes, K., Binns, M.M. &Boursnell, M.E.G. 1991. A recombinantfowlpox virus that expresses the VP2antigen of infectious bursal disease virusinduces protection against mortalitycaused by the virus. Arch. Virol., 120:193-205.

Belsham, G.J., Anderson, E.C., Murray, P.K.,Anderson, J. & Barrett, T. 1989. Immuneresponse and protection of cattle andpigs generated by a vaccine virusrecombinant expressing the F proteinof rinderpest virus. Vet. Rec., 124: 655-658.

Bjorkman, P.J., Saper, M.A., Samraoui, B.,Bennett, W.S., Strominger, J.L. & Wiley,D.C. 1987. The foreign antigen site andT cell recognition regions of class Ihistocompatibility antigens. Nature, 329:512-518.

Blancou, J., Kieny, M.P., Lathe, R., Lecocq,J.P., Pastoret, P.P., Soulebot, J.P. &Desmettre, P. 1986. Oral vaccination of

Vaccine manual 17

18 The induction of immunity by veterinary immunoprophylactic,v

the fox against rabies using a liverecombinant vaccinia virus. Nature, 322:373-375.

Bodmer, W. 1984. The major histocom-patibility system: structure and function:summary and synthesis. In Y. Yamamura& T. Tada, eds. Progress in ImmunologyV. Fifth Int. Congr. Immunology, P. 959-970. Tokyo, Academic Press Japan.

Boursnell, M.E.G., Green, P.P., Campbell,J.I.A., Deuter, A., Peters, R.W., Tomley,F.M., Samson, A.C.R., Emmerson, P.T.& Binns, M.M. 1990a. A fowlpox virusvaccine vector with insertion sites inthe terminal repeats: demonstration ofits efficacy using a fusion gene ofNewcastle disease virus. Vet. Microbiol.,23: 305-316.

Boursnell, M.E.G., Green, P.F., Campbell,J.I.A., Deuter, A., Peters, R.W., Tomley,F.M., Samson, A.C.R., Chambers, P.,Emmerson, P.T. & Binns, M.M. 1990b.Insertion of the fusion gene fromNewcastle disease virus into a non-essential region in the terminal repeatsof fowlpox virus and demonstration ofprotective immunity induced by therecombinant. J. Gen. Virol., 71: 621-628.

Brodsky, F.M. 1992. Antigen processing andpresentation: close encounters in theendocytic pathway. Trends Cell Biol., 2:109-115.

Brodsky, F.M. & Guagliardi, L.E. 1991. Thecell biology of antigen processing andpresentation. Annu. Rev. Immunol., 9:707-744.

Brown, J.H., Jardetzky, T.S., Gorga, J.C.,Stern, L.J., Urban, R.G., Strominger,J.L. & Wiley, D.C. 1993. Three-dimensional structure of the human classII histocompatibility antigen HLA-DR1.Nature, 364: 33-39.

Connell, N.D., Medina-Acosta, E., McMaster,W.R., Bloom, B.R. & Russell, D.G. 1993.Effective immunization against cutane-ous leishmaniasis with recombinantbacille Calmette-Guérin expressing the

Leislimania surface proteinase gp63. Proc.Natl Acad. Sci. USA, 90: 11473-11477.

Cooper, M.D. & Burrows, P.D. 1989. B celldifferentiation. In T. Honjo, F.W. Alt &T.H. Rabbits, eds. Immunoglobulin genes,p. 1-21. San Diego, Calif., USA, Academic.

Cox, J.M., Zamb, T.J. & Babiuk, L.A. 1993.Bovine herpesvirus 1: immune responsesin mice and cattle injected with plasmidDNA. J. Virol., 67(9): 5664-5667.

Davis, M.M. & Bjorkman, P.J. 1988. T-cellantigen receptor genes and T cell re-cognition. Nature, 334: 395-401.

Dertzbaugh, M.T. & Elson, C.O. 1993.Comparative effectiveness of choleratoxin B subunit and alkaline phosphataseas carriers for oral vaccines. Infect.Immunol., 61: 48-55.

Di Marchi, R., Brooke, G., Gale, C.,Cracknell, V., Doel, T. & Mowat, N.1986. Protection of cattle against foot-and-rnouth disease by a syntheticpeptide. Science, 232: 639-641.

Doherty, P.C., Blanden, R.V. & Zinkernagel,R.M. 1976. Specificity of virus-immuneeffector T cells for H-2K and H-2Dcompatible interactions: implications forH-antigen diversity. Transplant. Rev., 29:89-124.

Dougan, G., Hormaeche, C.E. & Maskell,D.J. 1987. Live oral Salmonella vaccines:potential use of attenuated strains ascarriers of heterologous antigens to theimmune system. Parasite Immunol., 9:151-160.

Dougan, G., Chatfield, S., Pickard, D., Bester,J., O'Callaghan, D. & Maskell, D.J. 1988.Construction and characterisation ofvaccine strains of Salmonella harbouringmutations in two different aro genes. J.Inf. Dis., 158: 1329-1335.

Else, K.J. & Grencis, R.K. 1991. Cellularimmune responses to the murine nema-tode parasite Trichuris muris. I. Differ-ential cytokine production during acuteor chronic infection. Immunology, 72:508-513.

Fitch, F.W., McKisic, M.D., Lancki, D.W.& Gajewski, T.F. 1993. Differentialregulation of murine lymphocyte subsets.Annu. Rev. Immunol., 11: 29-48.

Fynan, E.F., Webster, R.G., Fuller, D.H.,Haynes, J.R., Santoro, J.C. & Robinson,H.L. 1993. DNA vaccines: protectiveimmunizations by parenteral, mucosal,and gene-gun inoculations. Proc. NatlAcad. Sci. USA, 90: 11478-11482.

Geyson, H.M., Meloen, R.H. & Barteling,S.J. 1984. Use of peptide synthesis toprobe viral antigens for epitopes to aresolution of a single amino acid. Proc.Natl Acad. Sci. USA, 81: 3998-4002.

Hill, A.B., Barnett, B.C., McMichael, A.J.& McGeoch, D.J. 1994. HLA class Imolecules are not transported to thecell surface in cells infected with herpessimplex virus types 1 and 2. J. Immunol.,152: 2736-2741.

Johnson, K.S., Harrison, G.B.L., Lightowlers,M.W., O'Hoy, K.L., Cougle, W.G.,Dempster, R.P., Lawrence, S.B., Vinton,J.G., Heath, D.D. & Rickard, M.D. 1989.Vaccination against ovine cysticerosisusing a defined recombinant antigen.Nature, 338: 585-587.

Jones, P.W., Dougan, G., Hayward, N.,MacKenzie, N., Collins, P. & Chatfield,S.N. 1991. Oral vaccination of calvesagainst experimental salmonellosis usinga double aro mutant of Salmonellatyphimurium. Vaccine, 9: 29-34.

Kitching, R.P., Hammond, J.M. & Taylor,W.P. 1987. A single vaccine for the controlof capripox infection in sheep and goats.Res. Vet. Sci., 42: 53-60.

Liew, F.Y. 1990. Regulation of cell-mediatedimmunity to leishmaniasis. In S.H.E.Kaufmann, ed. T-cell paradigms inparasitic and bacterial infections. Curr.Top. Microbiol. Immunol., 155: 53-64.

Lippé, R., Luke, E., Kuah, Y.T., Lomas, C.& Jeff eries, W.A. 1991. Adenovirusinfection inhibits the phosphorylationof major histocompatibility complex class

I proteins. J. Exp. Med., 174: 1159-1166.Locksley, R. 1993. Interleukin 12 in host

defense against microbial pathogens.Proc. Natl Acad. Sci. USA, 90: 5879-5880.

Lukacher, A.E., Morrison, L.A., Braciale,V.L. & Braciale, T.J. 1986. T lymphocytefunction in recovery from experimentalviral infection: the influenza model. InR.M. Steinman & R.J. North, eds.Mechanisms of Host Resistance to InfectiousAgents, Tumours and Allografts: A Con-ference in Recognition of the TrudeauInstitute Centenial, p. 233-254. New York,The Rockefeller University Press.

Mackett, M. & Smith, G.L. 1986. Vacciniavirus expression vectors. J. Gen. Virol.,67: 2067-2082.

Matsumura, M., Fremont, D.H., Peterson,P.A. & Wilson, I.A. 1992. Emergingprinciples for the recognition of peptideantigens by MHC class I molecules.Science, 257: 927-934.

McAleer, W.J., Buynak, E. I., Margetter, R.Z.,Wampler, E. & Miller, W.J. 1984. Humanhepatitis B vaccines from recombinantyeast. Nature, 307: 178-180.

Moormann, R.J.M., de Rover, T., Briaire,J., Peeters, B.P.H., Gielkens, A.L.J. &van Oirschot, J.T. 1990. Inactivation ofthe thymidine kinase gene of a gl deletionmutant of pseudorabies virus generatesa safe but still highly immunogenicvaccine strain. J. Gen. Virol., 71: 1591-1595.

Mosmann, T.R. & Coffman, R.L. 1989. Th1or Th2 cells: different patterns of lympho-kine secretion lead to different functionalproperties. A1111 u. Rev. Immunol., 7: 45-173.

Mosmann, T.R., Cherwinski, H., Bond,M.W., Giedlin, M.A. & Coffman, R.L.1986. Two types of murine helper T cellclone. I. Definition according to profilesof lymphokine activities and secretedproteins. J. Immunol., 136: 2348-2357.

Mosmann, T.R., Schumacher, J.H., Street,N.F., Budd, R., O'Garra, A., Fong, T.A.T.,Bond, M.W., Moore, Sher, A.& Fiorentino, D.F. 1991. Diversity of

Vaccine manual 19

20 The induction of inununity by veterinary immunoprophylactics

cytokine synthesis and function of mouseCD4+ T cells. Immunol. Rev., 123: 209-229.

Musoke, A.J., Morzaria, S., Nkonge, C.,Jones, E. & Nene, V. 1992. A recombinantsporozoite surface antigen of Theileriaparva induces protection in cattle. Proc.Natl Acad. Sci. USA, 89: 514-518.

Pearce, E.J., Caspar, P., Grzych, J.-M., Lewis,F.A. & Sher, A. 1991. Downregulationof Thl cytokine production accompaniesinduction of Th2 responses by a parasitichelminth, Schistosoma mansoni. J. Exp.Med., 173: 159-166.

Rammensee, H.-G., Falk, K. & Rotzschke,0. 1993. Peptides naturally presentedby MHC class I molecules. Annu. Rev.Immunol., 11: 213-244.

Ramshaw, I., Ruby, J., Ramsay, A., Ada, G.& Karupiah, G. 1992. Expression ofcytokines by recombinant vacciniaviruses: a model for studying cytokinesin virus infections in vivo. Immunol. Rev.,127: 157-182.

Rand, K.N., Moore, T., Sriskantha, A.,Spring, K., Tellam, R., Willadsen, P.& Cobon, G.S. 1989. Cloning andexpression of a protective antigen fromthe cattle tick Boophilus microplus. Proc.Natl Acad. Sci. USA, 86: 9657-9661.

Romero, C.H., Barrett, T., Evans, S.A.,Kitching, R.P., Gershon, P.D., Bostock,C. & Black, D.N. 1993. A single capripoxrecombinant vaccine for the protectionof cattle against rinderpest and lumpyskin disease. Vaccine, 11: 732-742.

Rothbard, J.B. & Gefter, M.L. 1991.Interactions between immunogenicpeptides and MHC proteins. Annu. Rev.Immunol., 9: 527-565.

Roy, P. 1992. Bluetongue virus proteins. J.Gen. Virol., 73: 3051-3064.

Roy, P., French, T.J. & Erasmus, B.J. 1992.Protective efficacy of virus-like particlesfor bluetongue disease. Vaccine, 10: 28-32.

Rudensky, A.Y., Preston-Hurlburt, P., Al-Ramadi, B.K., Rothbard, J. & JanewayJr, C.A. 1992. Truncation variants of

peptides isolated from MHC class IImolecules suggest sequence motifs.Nature, 359: 429-431.

Sadoff, J.C., Ballou, W.R., Baron, L.S.,Majarian, W.R., Brey, R.N., Hockmeyer,W.T., Young, J.F., Cryz, S.J., Ou, J.,Lowell, G.H. & Chulay, J.D. 1988.Oral Salmonella typhimurium vaccineexpressing circumsporozoite proteinprotects against malaria. Science, 240:336-338.

Sette, A., Buus, S., Colon, S., Smith, J.A.,Miles, C. & Grey, H.M. 1987. Structuralcharacteristics of an antigen requiredfor its interaction with Ia and recognitionby T cells. Nature, 328: 395-399.

Sher, A., Gazzinelli, R.T., Oswald, I.P.,Clerici, M., Kullberg, M., Pearce, E.J.,Berzofsky, J.A., Mosmann, T.R., James,S.L., Morse, H.C. & Shearer, G.M. 1992.Role of T-cell derived cytokines in thedownregulation of immune responsesin parasitic and retroviral infection.Immunol. Rev., 127: 183-204.

Smith, G.L. 1993. Vaccinia virus glycoproteinsand immune evasion. J. Gen. Virol., 74:1725-1740.

Spangler, B.D. 1992. Structure and functionof cholers toxin and the related Escherichiacoli heat-labile enterotoxin. Microbiol.Rev., 56: 622-647.

Stern, L.J., Brown, J.H., Jardetzky, T.S.,Gorga, J.C., Urban, R.G., Strominger,J.L. & Wiley D.C. 1994. Crystal structureof the human class II MHC protein HLA-DR1 complexed with an influenza viruspeptide. Nature, 368: 215-221.

Swain, S.L., Bradley, L.M., Croft, M.,Tonkonogy, S., Atkins, G., Weinberg,A.D., Duncan, D.D., Hendrick, S.M.,Dutton, R.W. & Huston, G. 1991. HelperT cell subsets: phenotype function andthe role of lymphokine in regulatingtheir development. Immunol. Rev., 123:115-144.

Takahashi, H., Takeshita, T., Morein, B.,Putney, S., Germain, R.N. & Berzofsky,

J.A. 1990. Induction of CD8+ cytotoxicT cells by immunisation with purifiedHIV-1 envelope protein in ISCOMs.Nature, 344: 873-875.

Townsend, A. & Bodmer, H. 1989. Antigenrecognition by class I-restrictedT lymphocytes. Annu. Rev. Immunol., 7:601-624.

Townsend, A., Bastin, J., Gould, K.,Brownlee, G., Andrew, A., Boyle D.B.,Chan, Y. & Smith, G. 1988. Defectivepresentation to class I-restricted cytotoxicT lymphocytes in vaccinia-infected cellsis overcome by enhanced degradationof antigen. J. Exp. Med., 166: 1211-1224.

Trowsdale, J. 1993. Genomic structure andfunction in the MHC. Trends Gen., 9:117-122.

Ulmer, J.B., Donnelly, J.J., Parker, S.E.,Rhodes, G.H., Feigner, P.L., Dwarki,V.J., Gromkowski, S.H., Deck, R.R.,DeWitt, C.M., Friedman, A., Hawe, L.A.,Leander, K.R., Martinez, D., Perry, H.C.,Shiver, J.W., Montgomery, D.L. & Liu,M.A. 1993. Heterologous protectionagainst influenza by injection of DNAencoding a viral protein. Science, 259:1745-1749.

Unanue, E.R. 1992. Cellular studies on antigenpresentation by class II MHC molecules.Curr. Opin. Immunol., 4: 63-69.

Urban, J., Madden, K., Svetic, A., Cheever,A., Trotta, P., Gause, W. & Finkelman,F. 1992. The importance of Th2 cytokinesin protective immunity to nematodes.Immunol. Rev., 127: 205-220.

van Bleek, G.M. & Nathenson, S.G. 1992.Presentation of antigenic peptides byMHC class I molecules. Trends Cell Biol.,2: 202-207.

Vitiello, A. & Sherman, L.A. 1983. Re-cognition of influenza-infected cells bycytolytic T lymphocyte clones: de-terminant selection by class I restrictionelements. J. Immunol., 131: 1635-1640.

Weiranga, E.A., Snook, M., deGroot, C.,Chretien, I., Bos, J., Jansen, H. &

Kapsenberg, M. 1990. Evidence forcompartmentalisation of functionalsubsets of CD4 T lymphocytes in atopicpatients. J. Immunol., 144: 4651-4656.

Willadsen, P. 1980. Immunity to ticks. Adv.Parasitol., 18: 293-313.

Willadsen, P., McKenna, R.V. & Riding,G.A. 1988. Isolation from the cattle tickBoophilus microplus of antigenic materialcapable of eliciting a protective immuno-logical response in the bovine host. Int.J. Parasitol., 18: 183-191.

Willadsen, P., Riding, G.A., McKenna, R.V.,Kemp, D.H., Tellam, R.L., Nielsen, J.N.Lahnstein, J., Cobon, G.S. & Gough,J.M. 1989. Immunologic control of aparasitic arthropod: identification of aprotective antigen from Boophilusmicroplus. J. Immunol., 143: 1346-1351.

Yilma, T., Hsu, D., Jones, L., Owens, S.,Grubman, M., Mebus, C., Yamanaka,M. & Dale, B. 1988. Protection of cattleagainst rinderpest with vacciniarecombinants expressing the HA or Fgene. Science, 242: 1058-1061.

Vaccine manual 21

GensilleaNy jnes[red vaccines

Genetic engineering is a term used todescribe experimental or industrialapproaches to modifying the genome of aliving cell without normal sexual or asexualtransmission of genetic material. In abroader sense, it is used to describe geneticmanipulation of the genome of a virus andis often used synonymously with the termrecombinant DNA technology. The abilityto manipulate the genome of a living cellor virus was made possible by a series ofdiscoveries made 20 to 30 years ago.

First was the recognition of bacterialrestriction enzymes by Arber and Dussoix(1962). These are endonucleases thatrecognize specific DNA sequences andcleave the DNA only at those sites(Meselson and Yuan, 1968). Coupled withother enzymes involved in DNA synthesissuch as ligases (DNA-joining enzymes), itbecame feasible to introduce deliberatelyspecific foreign DNA sequences into DNAmolecules which are then described asrecombinant DNA. When the recombinantDNA replicates, the foreign DNA is alsoreplicated. If the foreign DNA is insertedinto a bacterial plasmid or phage DNAwhich undergoes rapid multiplicationtogether with the host bacteria, it is possibleto generate large quantities of the foreignDNA. This process is called molecularcloning (Sambrook, Fritsch and Maniantis,1989).

Another crucial discovery was theexistence of reverse transcriptase, anenzyme present in oncogenic RNA virusesand which converts RNA into DNA(Baltimore, 1970). This allowed theconversion of cellular or viral messengerRNAs into DNA so that they could then

B.W.J. Mahy

be molecularly cloned and expressed toproduce the polypeptide or proteinspecified by the original messenger RNA.Bacterial plasmids can serve both ascloning and expression vectors whenintroduced into appropriate host cells and,in the last few years, a range of expressionvectors have become available that willreplicate in bacteria, yeast, insect or animalcells to produce large quantities of adesired polypeptide or protein.

Such expressed proteins can be purifiedto serve as a vaccine antigen, as in thehighly successful vaccine against hepatitisB prepared in yeast (Saccharomycescerevisiae), which was the first geneticallyengineered vaccine licensed for use inhumans. Alternatively, if the expressionvector will replicate in animal or humancells, the expressed antigen may immunizethe recipient host without the need forpurification. This is the basis for thedevelopment of a wide range of candidatevaccines for both human and veterinaryuse which depend on recombinant geneexpression in an appropriate viral orbacterial vector. Such live recombinantvector vaccines offer great promise but, todate, have not been licensed for use inhuman subjects.

Recently, the discovery that plasmidDNA sequences can be expressed directlyafter injection into mouse muscle (Wolff etal., 1990) without the need for a viral orbacterial expression vector has opened upan exciting range of possibilities for thedesign of vaccines for human or veterinaryuse through polynucleotide vaccination(Cohen, 1993; Ulmer et al., 1993; Ulmer,Donnelly and Liu, 1994). In addition, since

Vaccine manual 23

the encoded antigens are expressedintracellularly in the recipient host, theyare introduced into class I MHC moleculesand this results in the priming of cytotoxicT cells (Townsend and Bodmer, 1989;Barber and Parham, 1994) so that bothcellular and humoral immune responsesare stimulated (Ulmer et al., 1993).

Finally, the ability to determine thenucleotide sequence of the genomes ofDNA or RNA animal viruses has led to aconsiderable body of research on thestructure of the antigens they encode. Inaddition to expressing the genes in variousvectors, attempts have been made tosynthesize in vitro various polypeptideswhich correspond to these antigenicdeterminants or epitopes and to use thepeptides themselves as candidate immuno-gens. Unfortunately, this approach has hadonly limited practical success, but someimportant early work was carried out withfoot-and-mouth disease (Bittle et al., 1982;Di Marchi et al., 1986).

In this brief overview, the developmentand future prospects for geneticallyengineered veterinary vaccines will bedescribed. The various types of vaccinewill be considered separately according tothe listing given in Table 1.

GENETICALLY ENGINEERED ATTENUATION

Many conventional vaccines currentlyemployed in human or veterinarymedicine contain live micro-organisms orvirus that has been attenuated, for exampleby multiple passages in cell culture, to thepoint that it will multiply in the recipienthost but no longer induce disease. Livevaccines offer several potential advantagescompared with killed vaccines in that theyusually induce a rapid immune responseof long duration and also induce localimmunity which may be critical inproviding protection against subsequentinfection by virulent micro-organisms.However, conventional methods of

TABLE 1

Prospective genetically engineeredvaccine types

Attenuated live

Defective live

Chimeric live

Pseudorabies (Aujeszky'sdisease)

ALVAC-FL (feline leukaemia)

Capripox, rinderpest, lumpyskin disease

attenuation rely on spontaneous, randommutations occurring during multiplepassages, and the basis for the attenuationis usually not known. An alternativeapproach is to use genetic engineering todefine specific genes or regions of thegenome which are responsible forvirulence, then delete these in order toobtain a replicating, non-pathogenic virusfor use as the immunogen.

This approach has been particularlyfruitful with bacteria and DNA viruseswith large genomes, such as herpesvirusesand poxviruses. Since this type ofattenuation involves the removal of partsof the genome, such bacteria and virusesmay then be used to vector genes encodinga variety of foreign antigens, creatingchimeric vaccines.

The experimental protocol for genedeletion and insertion usually requiressome means of identification of theprogeny which contain the desiredgenotype. In the original experiments todevelop modified poxviruses for foreigngene insertion, the thymidine kinase (TK)gene was deleted and a positive selectionof progeny which lack the gene was madeby growing the virus in a TK cell line(lacking a cellular thymidine kinase) using

24 Genetically engineered vaccines

Type Examples

Subunit Bluetongue

DNA polynucleotide Avian influenza

Polypeptide Foot-and-mouth disease

a medium containing 5-bromodeoxy-uridine (BudR). Any progeny virus whichcontains the TK gene will convert BudRinto a phosphorylated form which is toxicto the host cell. Consequently, onlyrecombinant viruses with the TK deletionwill survive (Mackett and Smith, 1986). Asimilar system can be used to selectherpesviruses from which the TK gene hasbeen deleted, for example bovine herpes-virus 1 (Bello, Whitbeck and Lawrence,1992; Kit, Kit and McConnell, 1986). Adeletion mutant of pseudòrabies virus(suid herpesvirus 1) has been used succes-sfully to vaccinate pigs against Aujeszky'sdisease. This genetically engineered virus,strain 783, contains three deletions: onethat prevents expression of pseudorabiesvirus glycoprotein 1 (2 055 base pairs), onethat inactivates the viral thymidine kinase(19 base pairs) and one of 100 base pairsthat lies in the promoter-enhancer regionof the immediate early gene (Glazenburget al., 1994).

A more general method for gene deletionand the detection of progeny virus withthe required genetically engineereddeletions is to use a h.eterologous reportergene contained in a plasmid transfervector. The reporter gene is inserted into aplasmid between flanking sequences of thegene to be deleted. When the virus genomeDNA is cotransferred into cells along withthe plasmid containing the reporter gene,recombination occurs between the homolo-gous (flanking) sequences of the genomeDNA and those in the transfer vector. Thisyields some progeny containing substitu-tion of the virus gene by the reporter gene,and these can be detected in a variety ofways depending on which reporter gene isused. The most convenient plasmidtransfer vectors contain a chromo-genic reporter such as p-galactosidasewhich allows an easy visual selection ofplaques containing recombinant virus(Chakrabarti, Brechling and Moss, 1985).

In the early development of poxvirusesas gene vectors, it was estimated thatinfectious virus could be producedcontaining at least 25 000 base pairs offoreign DNA (Smith and Moss, 1983). Morerecently, a highly attenuated candidatevaccinia virus vector vaccine (NYVAC) hasbeen constructed by the deletion of 18genes which are involved in pathogenesisand host range (Tartaglia et al., 1992). Theresultant virus replicates well in Vero cellsor chick embryo fibroblasts but poorly incells of human, equine, murine or swineorigin, and provides a safer alternative tovaccinia virus as a potential gene vectorvaccine (Tartaglia et al., 1994).

The potential of NYVAC as a vectorvaccine against pseudorabies virus hasbeen demonstrated. NYVAC recombinantsexpressing pseudorabies virus glyco-protein gp50 protected pigs against chal-lenge with live virulent virus (Brockmeieret al., 1993).

GENETICALLY ENGINEERED DEFECTIVE

VACCINES

Recently, the potential of non-replicating(defective) poxvirus as vectors for foreigngenes has been explored. Avipoxvirusesdo not replicate when introduced intomammalian hosts, but they may still act asgene expression vectors in mammals aswell as birds, since replication is blockedat a stage after early gene expression, butbefore DNA replication. The basis for theavian poxvirus vector was an attenuatedlicensed vaccine for canaries (ALVAC)(Taylor et al., 1991 and 1992; Cadoz et al.,1992). When the glycoprotein gene of rabiesvirus was introduced into ALVAC byrecombination, a highly effective vaccinepreparation was obtained (ALVAC-RG)which induced high levels of immunity inhumans (Cadoz et al., 1992) and provideda high level of protection in dogs againstchallenge with virulent rabies virus, eventhough the ALVAC-RG virus did not

Vaccine manual25


replicate in the dogs or human subjects(Taylor et al., 1994). The use of ALVAC as avector for the env and gag genes of felineleukaemia virus (ALVAC-FL) has also beendescribed (Tartaglia et al., 1993).

Another highly attenuated poxvirus thathas been well characterized is modifiedvaccinia virus Ankara (MVA). This viruswas passaged more than 500 times in chickembryo fibroblasts, and is unable toreplicate in mammalian cells (Meyer, Sutterand Mayr, 1991). Replication is blocked atthe assembly stage so that DNA replicationand late protein synthesis occur and largeamounts of foreign gene products areproduced, similar in amount to wild-typevirus (Sutter and Moss, 1992).

It is clear that such non-replicating geneexpression vectors provide a safe andeffective means to immunize a variety ofmammalian species, including humans,since the virus cannot spread beyond theinitially infected cells.

CHIMERIC LIVE VACCINES

Candidate live recombinant vaccinesexpressing foreign genes have beendeveloped from a variety of poxvirusvectors as well as adenoviruses andSalmonella spp. For the most part,adenoviruses are currently being exploredas potential vectors for human mucosalvaccination and so will not be consideredfurther here. However, there have beensome studies in animal species, especiallycanines, and an excellent review on thesubject has been published (Graham andPrevec, 1992). By the same token, theprincipal interest in Salmonella spp. is inthe development of multivalent oralvaccines for human use. Although there isalso considerable potential for Salmonella-based vectors in veterinary medicine,particularly in poultry, the subject is stillrelatively unexplored. A useful review onSalmonella-based vaccines has appearedrecently (Chatfield, Dougan and Roberts,

1994). For these reasons, the followingdiscussion will be confined to someremarkable examples of poxvirus vectorsfor use in veterinary medicine. A moredetailed account of the earlier studies inthis area was published by Esposito andMurphy (1989).

One of the first vaccinia virus recombin-ants of veterinary interest to be developedwas a potential vaccine against vesicularstomatitis (VSV), w.hich is a contagiousdisease of horses, cattle and pigs that cancause serious economic losses in North andSouth America because the lesions may beconfused with those of foot-and-mouthdisease (FMD). Recombinant vaccinia viruswas prepared which expressed theG glycoprotein of VSV at the surface ofinfected cells. Some protection againstintralingual challenge with virulent VSVwas observed in cattle 44 days afterreceiving the recombinant vaccine on days1 and 29 (Yilma, 1994).

When the glycoprotein gene of anotherrhabdovirus, rabies, was expressed in avaccinia virus recombinant, goodprotection against challenge was found ina variety of animals such as mice, rabbits,skunks, raccoons and foxes (Wiktor et al.,1984; Blancou et al., 1986; Rupprecht et al.,1988). It was possible to im.munizeraccoons and foxes, but not dogs or skunks,by oral administration of the vaccinia-rabies G protein recombinant (VRG). Thishas led to a series of detailed studies of thepotential for oral vaccination of raccoonsand foxes by distribution of baitscontaining VRG in the wild. Considerablesuccess has been achieved with rabiesvaccination of foxes in Europe using thismethod, and carefully controlled studiesof the use of such baits to vaccinateraccoons are under way at several sites inNorth America (Rupprecht et al., 1988 and1993; Rupprecht, Hanlon and Koprowski,1992; Hanlon et al., 1993). The principalconcern in the use of such VRG-containing

baits in the field is the possible risk tohumans who might become accidentallyinfected with the vaccinia recombinant. Inexceptional circumstances, for example ifan immunocomprornised person wasinfected, there might be a risk of a seriousvaccinia infection. Alternative poxvirusvectors such as raccoon pox (Esposito,Chandler and Baer, 1989) have been usedto create a potentially safer vaccine butinsufficient studies have been made of itseffects in other species. For this reason thecurrent trials with VRG are being carefullymonitored and the risk of human exposureappears to be extremely low. Theconsiderable potential for poxvirus-vectored rabies vaccines to control theunprecedented rabies epidemic in raccoonsin North America (Rupprecht and Smith,1994) needs to be carefully evaluated withregard to risks and benefits. It seems likely,however, that VRG vaccines will soon belicensed for use in the United States.

Unfortunately, a VRG recombinantvaccine that had not been sanctioned bythe Argentine authorities was used inArgentina in a field experiment involvingcattle (Koprowski et al., 1957). The potentialfor human exposure to vaccinia virus hadnot been adequately explained to thosewho handled the inoculated animals; forexample, it is known that vaccinia viruscan be isolated from the scabs present oncattle at the site of recombinant virusinoculation (Gillespie et al., 1986). Eventhough none of the cattle handlers sufferedany disease, this incident resulted in aconsiderable setback to the introduction ofrecombinant poxvirus vaccines foragricultural purposes, especially in SouthAmerica (Crawford, 1987).

One of the principal advantages ofvaccinia virus, which contributed to thesuccess of the smallpox eradicationcampaign, is its relative stability. Vacciniavirus preparations will remain viable inconditions which would destroy the

infectivity of live vaccines againstmorbilliviruses such as measles orrinderpest. A highly effective vacciniarecombinant virus expressing the F and Hgenes of rinderpest has been described foruse in cattle (Giavedoni et al., 1991), andthis virus will also protect goats againstthe related morbillivirus, peste des petitsruminants (PPR) virus (Jones et al., 1993).Recently, an alternative chimeric vaccinecandidate was described which is basedon capripoxvirus, a natural virus infectionof cattle, sheep and goats. Live attenuatedcapripoxvirus preparations are already inuse as effective vaccines against sheep andgoat pox and lumpy skin disease of cattle,so they are ideal vectors for recombinantvaccines in these species. When the F geneof rinderpest virus was introduced intothe capripox vaccine virus at the TK locus,the resultant recombinant virus protectedcattle against rinderpest following a lethalchallenge and also immunized the cattleagainst lumpy skin disease (Romero et al.,1993).

In addition to these examples, re-combinant vaccines against other bovinediseases have been described. Theprotection of sheep against bovineleukaemia virus using a vacciniarecombinant expressing the gp51 poly-peptide has been described, and this maywell be effective in cattle also (Daniel et al.,1993). The use of such poxvirus-vectoredrinderpest vaccines could be crucial in thecampaign to eradicate this plague from theremaining pockets of disease in Africa andAsia.

The potential of fowl poxviruses as genevectors for vaccines against a wide rangeof poultry diseases was recognized morethan ten years ago (Boyle and Coupar,1988). The difficulty of producing suchgenetically engineered vaccines at aneconomic cost has hindered theirdevelopment, however. For vaccinesto be effective in the poultry production

Vaccine manual27


industry, they need to be administered byone day of age using mass deliverymethods, for example inclusion in thedrinking water. There are numerousexamples of potential recombinantvaccines against diseases such as avianinfluenza, Newcastle disease and Marek'sdisease, but all require wing-webinoculation for maximum efficacy and thismay not be practical. An excellent reviewof potential recombinant vaccines forpoultry appeared recently (Boyle andHeine, 1993).

SUBUNIT VACCINES

Live virus vaccines usually induce highlevels of immunity which persist in thehost. The major concern regarding theiruse is the possibility of reversion tovirulence, and for diseases such as foot-and-mouth disease only inactivated virusis suitable for use as a vaccine.Nevertheless, problems have occurredowing to incomplete virus inactivation,with resultant outbreaks of disease.

An alternative approach is to usesubviral components as immunogens inthe form of subunit vaccines. Generally,the isolated components of a virus maynot be effective in stimulating anappropriate immune response but, ifpresented in conjunction with anappropriate adjuvant such as alumina gelor 6D-stearoyL-murarnyL-dipeptide, abetter response can be generated.Potentiation of the vaccine by addition ofcytokines or lymphokines is also beingstudied (Hughes and Babiuk, 1994). Formembrane proteins of enveloped viruses,an immunostimulating complex (ISCOM)was developed which, when mixed withthe antigen, presents it to the immunesystem in the form of micelles (Morein etal., 1984; Morein and Simons, 1985).

The advent of expression vectors suchas baculoviruses, which can produceenormous quantities of a desired protein

in cell culture, has greatly enhanced thepossibilities for subunit vaccine devel-opment. For example, the baculovirusAutographa cahfornica has been developedto express a variety of foreign genes undercontrol of the strong polyhedrin promoter(Miller, 1988). The polyhedrin gene isreplaced by the gene encoding the desiredprotein., and the recombinant baculovirus-infected Spodoptera frugiperela cells mayproduce up to 50 percent of total cellprotein as the foreign gene product.

This system has been used to engineermultiple-component proteins of blue-tongue virus and, when produced inappropriate quantities, the proteins self-assemble to form non-infectious virus-likeparticles (Marshall and Roy, 1990; Pearsonand Roy, 1993). A further development ofthis system is to introduce other virusantigens into the bluetongue virus subunitsduring self-assembly to create proteinchimeras which will immunize againstmore than one disease.

This approach offers considerablepotential for the future development ofveterinary vaccines.

DNA POLYNUCLEOTIDE VACCINES

Wolff et al. (1990) reported that the directintroduction of plasmid DNA into mousemuscle resulted in expression of encodedproteins in the rnuscle cells. Although theexact process by which transcription occursin this system has not been completelyelucidated, it seems that muscle injectionis more effective at inducing an immuneresponse than other routes or tissues thathave been tried.

Recently, the application of a biolisticgun (gene gun) that shoots gold particlescoated with DNA directly into the muscle,has been described (Williams et al., 1991;Fynan et al., 1993).

This field is just opening up, but theeffectiveness of naked DNA in provokingimmunity to rabies in rnice and virulent

influenza in chickens has been reported(Xiang et al., 1994; Robinson, Hunt andWebster, 1993). The state of the art of DNAvaccination has been well summarized byUlmer, Donnelly and Liu (1994), Ulmer etal. (1993) and Cohen (1993).

POLY7PY:D7 VACCINES

When predicted amino acid sequencesfrom a wide range of pathogenic virusesbecame available in the mid-1970s, thepossibility of using chemically synthesizedpolypeptides as immunogens wasexplored. Previous studies on proteinssuch as lysozyme had established thatimmune responses could be elicited byshort peptides containing about 20 aminoacids. It was also clear that peptidescorresponding to the aminoterminus orcarboxyterminus were frequently antigenic(i.e. elicited antibodies that reacted withthe original protein). Numerous examplesof viral peptides that served asimmunogens were reported (Geysen,Barteling and Meloen, 1985; Shinnick et al.,1983). When it was found that guinea pigscould be protected from foot-and-mouthdisease by a short peptide of 19 aminoacids corresponding to a portion of theVP1 protein of the virus (Bittle et al., 1982),peptides were hailed as the next generationof foot-and-mouth disease vaccines(Brown, 1985). Subsequently, Di Marchi etal. (1986) showed that it was possible toprotect cattle from a virulent intralingualchallenge with foot-and-mouth diseasevirus using a peptide of 42 amino acids,but very large quantities of the peptidewere required. In general, the peptideswere found to stimulate a remarkablystrong humoral response even at low dosesof peptide, but protection against challengedid not occur except at high doses (5 mgper animal).

Despite the great amount of work donein attempting to improve on these results,for example by coupling the peptide with

predicted amino acid sequences designedto stimulate cell-mediated immunity, thepotential for such simple peptides asvaccines has not been realized.

BIBLIOGRAPHY

Arber, W. & Dussoix, D. 1962. Host specificityof DNA produced by Escherichia coll.1. Host-controlled modification ofbacteriophage. J. Mol. Biol., 5: 18.

Baltimore, D. 1970. RNA-dependent DNApolymerase in virions of RNA tumorviruses. Nature, 226: 1209-1211.

Barber, L.D. & Parham, P. 1994. The essenceof epitopes. J. Exp. Med., 180: 1191-1194.

Bello, L.J., Whitbeck, J.C. & Lawrence, W.C.1992. Bovine herpesvirus 1 as a live virusvector for expression of foreign genes.Virology, 190: 666-673.

Bittle, J.L., Houghten, R.A., Alexander, H.,Shinnick, T.M., Sutcliffe, J.G. & Lerner,R.A. 1982. Protection against foot-and-mouth disease by immunization with achemically synthesized peptide predictedfrom the viral nucleotide sequence.Nature, 298: 30-33.

Blancou, J., Kieny, M.P., Lathe, R., Lecocq,J.P., Pastoret, P.P., Soulebot, J.P. &Desmettre, P. 1986. Oral vaccination ofthe fox against rabies using a liverecombinant vaccinia virus. Nature, 322:373-375.

Boyle, D.B. & Coupar, B.E.H. 1988.Construction of recombinant fowlpoxviruses as vectors for poultry vaccines.Virus Res., 10: 343-356.

Boyle, D.B. & Heine, H.G. 1993. Recombinantfowlpox virus vaccines for poultry.Immunol. Cell Biol., 71: 391-397.

Brockmeier, S.L., Lager, K.M., Tartaglia,J., Riviere, M., Paoletti, E. & Mengeling,W.I. 1993. Vaccination of pigs againstpseudorabies with highly attenuatedvaccinia (NYVAC) recombinant viruses.Vet. Microbiol., 38: 41-58.

Brown, F. 1985. Peptides as the next

Vaccine manual 29


creneration of foot-and-mouth diseasevaccines. BioTechnology, 3: 445-447.

Cadoz, M., Strady, A., Meigner, B., Taylor,J., Tartaglia, J., Paolelti, E. & Ploticin,S. 1992. Immunization with canarypoxvirus expressing the rabies glycoprotein.Lancet, 339: 1429-1432.

Chakrabarti, S., Brechling, K. & Moss, B.1985. Vaccinia virus expression vector:coexpression of P-galactosidase providesvisual screening of recombinant virusplaques. Mol. Cell. Biol., 5: 3403-3409.

Chatfield, S.N., Dougan, G. & Roberts, M.1994. Progress in the development ofmultivalent oral vaccines based on liveattenuated Salmonella. In E. Kurstak, ed.Modern vaccinology, p. 55-85. New York,Plenum Medical.

Cohen, J. 1993. Naked DNA points way tovaccines. Science, 259: 1691-1692.

Crawford, M. 1987. Argentine experimentdid not break U.S. biotechnology rules.Science, 235(4786): 276.

Daniel, R.C.W., Gatei, M.H., Good, M.F.,Boyle, D.B. & Lavin, M.F. 1993. Re-combinant viral vaccines for enzooticbovine leucosis. Immunol. Cell Biol., 71:399-404.

Di Marchi, R., Brooke, G., Gale, C.,Cracknell, V., Doel, T. & Mowat, N.1986. Protection of cattle against foot-and-mouth disease by a syntheticpeptide. Science, 232: 639-641.

Esposito, J.J. & Murphy, F.A. 1989. Infectiousrecombinant vectored virus vaccines.Adv. Vet. Sci. Comp. Med., 33: 195-247.

Esposito, J.J., Chandler, F.W. & Baer, G.M.1989. Oral immunization of animals withraccoon poxvirus expressing rabies virusglycoprotein. Vaccine, 89: 1-6.

Fynan, E.F., Webster, R.G., Fuller, D.H.,Hanes, J.R., Santoro, J.C. & Robinson,H.L. 1993. DNA vaccines: protectiveimmunization by parenteral, mucosaland gene-gun inoculations. PNAS, 90:11478-11482.

Gao, Y., Leary, T.P., Eskra, L. & Splitter,

G.A. 1994. Truncated bovineherpesvirus-1 glycoprotein 1 (gpl)initiates a protective local immuneresponse in its natural host. Vaccine,12: 145-152.

Geysen, H.M., Barteling, S.J. & Meloen,R.H. 1985. Small peptides induce anti-bodies with a sequence and structuralrequirement for binding antigen com-parable to antibodies raised against thenative protein. PNAS, 82: 178-182.

Giavedoni, L., Jones, L., Mebus, C. & Yilma,T. 1991. A vaccinia virus double re-combinant expressing the F and H genesof rinderpest virus protects cattle againstrinderpest and causes no pock lesions.PN AS, 88: 8011.

Gillespie, J.H., Geissinger, C., Scott, F.W.,Higgins, W.P., Holmes, D.F., Perkus,M., Mercer, S. & Paoletti, E. 1986.Response of dairy calves to vacciniaviruses that express foreign genes. J.Clin. Microbiol., 23: 283-288.

Glazenburg, K.L., Elgersma-Hooisma, M.,Briaire, J, Voermans, J., Kimman, T.G.,Gielkens, A.L.J. & Moormann, R.J.M.1994. Vaccine properties of pseudorabiesvirus strain 783 are not affected by adeletion of 71 base pairs in the promoter/enhancer region of the viral immediateearly gene. Vaccine, 12: 1097-1101.

Graham, F.L. & Prevec, L. 1992. Adenovirus-based expression vectors and recombin-ant vaccines. In R. Ellis, ed. Vaccines:new approaches to immunological problems,p. 363-390. Stoneham, Mass, USA,Butterworth-Heinemann.

Hall, W.W., Kubo, T., Ijichi, S. Takahashi,H. & Zhu, S.W. 1994. Human T cellleukemia / lymphoma virus, type II(HTLV-11): emergence of an importantnewly recognized pathogen. Virology,5: 165-178.

Hanlon, C.A., Buchanan, J.R., Nelson, E.,Niu, H.S., Diehl, D. & Rupprecht, C.E.1993. A vaccinia-vectored rabies vaccinefield trial: ante- and post-mortem

biomarkers. Rev. sci. tech. Off. int. Epizoot.,12(1): 99-107.

Hughes, H.R.A. & Babiuk, L.A. 1994.Potentiation of vaccines through effectiveadjuvant fomulations and manipulationof the immune response. In E. Kurstak,ed. Modern vaccinology, p. 87-118. NewYork, Plenum Medical.

Jones, L., Giavedoni, L., Saliki, J.T., Brown,C., Mebus, C. & Yilma, T. 1993.Protection of goats against peste despetits ruminants with a vaccinia virusdouble recombinant expressing the Fand H genes of rinderpest virus. Vaccine,11: 961-964.

Kit, S., Kit, M. & McConnell, S. 1986.Intramuscular and intravaginal vaccin-ation of pregnant cows with thymidinekinase-negative, temperature-resistantinfectious bovine rhinotracheitis virus(bovine herpesvirus 1). Vaccine, 4: 55-61.

Koprowski, H., Celis, E., Curtis, P.,Dietzschold, B., Rupprecht, C., Tollis,M. & Wunner, W. 1987. Rabies experi-ment. Nature, 326: 636.

Mackett, M. & Smith, G.L. 1986. Vacciniavirus expression vectors. J. Gen. Virol.,67: 2067-2082.

Marshall, J.A. & Roy, P. 1990. High levelexpression of two outer cepside proteinsof bluetongue virus serotype 10. VirusRes., 15: 189-196.

Meselson, M. & Yuan, R. 1968. DNArestriction enzyme from E. coli. Nature,217: 1110.

Meyer, H., Sutter, G. & Mayr, A. 1991.Mapping of deletions in the genome ofthe highly attenuated vaccinia virusMVA and their influence on virulence.J. Gen. Virol., 72: 1031-1038.

Miller, L.K. 1988. Baculoviruses as geneexpression vectors. Annu. Rez2. Microbiol.,42: 177-199.

Morein, B. & Simons, K. 1985. Subunitvaccines against enveloped viruses:virosomes, micelles and other proteincomplexes. Vaccine, 3: 83-93.

Morein, B., Sundquist, B., Heiglund, S.,Dalsgaard, K. & Osterhaus, A. 1984.ISCOM, a novel structure for antigenicpresentation of membrane proteins fromenveloped viruses. Nature, 308(5958):457-460.

Moss, B. & Flexner, C. 1987. Vaccinia virusexpression vectors. Annu. Rev. Immunol.,5: 305-324.

Pearson, L.D. & Roy, P. 1993. Geneticallyengineered multi-component virus-likeparticles as veterinary vaccines. Immunol.Cell Biol., 71: 381-389.

Pensaert, M., Gielkens, A.L.J., Lomniczi,B., Kimman, T.G., Vannier, P. & Eloit,M. 1992. Round table on control ofAujeszky's disease and vaccine devel-opment based on molecular biology. Vet.Microbiol., 33: 53-67.

Robinson, H.L., Hunt, L.A. & Webster, R.G.1993. Protection against a lethal influenzavirus challenge by immunization witha haemagglutinin-expressing plasmidDNA. Vaccine, 11: 957-960.

Romero, C.H., Barrett, T., Evans, S.A.,Kitching, R.P., Gershon, P.D., Bostock,C. & Black, D.N. 1993. Single capri-poxvirus recombinant vaccine for theprotection of cattle against rinderpestand lumpy skin disease. Vaccine, 11:737-742.

Rupprecht, C.E. & Smith, J.S. 1994. Raccoonrabies: the emergence of an epizooticin a densely populated area. Semin.Virol., 5: 155-164.

Rupprecht, C.E., Hanlon, C.A. & Koprowski,H. 1992. Oral wildlife rabies vaccination:Development of a recombinant virusvaccine. In Trans. 57th N.A. Wildl. Nat.Res. Conf, p. 439-452.

Rupprecht, C.E., Hamir, A.N., Johnson, D.H.& Koprowski, H. 1988. Efficacy of avaccinia-rabies glycoprotein recombinantvirus vaccine in raccoons (Procyon lotor).Rev. Inf. Dis., 10: 803-809.

Rupprecht, C.E., Hanlon, C.A., Cummins,L.B. & Koprowski, H. 1992. Primate

Vaccine manual 31

32 Genetically enqineered vaccines

responses to a vaccini a-rabies glycoproteinrecombinant virus vaccine. Vaccine, 10:368-374.

Rupprecht, C.E., Hanlon, C.A., Niezgoda,M., Buchanan, J.R., Diehl, D. &Koprowski, H. 1993. Recombinant rabiesvaccines: efficacy assessment in free-ranging animals. Onderstepoort J. Vet.

Res., 60: 463-468.Sakaguchi, M., Hirayama, Y., Maeda, H.,

Matsuo, K., Yamamoto, M. & Hirai, K.1994. Construction of recombinantMarek's disease virus type 1 (MDV1)expressing the Escherichia coli lacZ geneas a possible live vaccine vector: theUS10 gene of MDV1 as a stable insertionsite. Vaccine, 12: 153-957.

Sambrook, J., Fritsch, E.F. & Maniantis, J.1989. Molecular cloning: a laboratorymanual, 2nd ed. Cold Spring Harbor,New York, Cold Spring Harbor Press.

Shinnick, T.M., Sutcliffe, J.G., Green, N.& Lerner, R.A. 1983. Synthetic peptideimmunogens as vaccines. Annu. Rev.Microbial., 37: 425-426.

Smith, G.L. & Moss, B. 1983. Infectiouspoxvirus vectors have capacity for atleast 25 000 base pairs of foreign DNA.Gene, 25: 21-28.

Sutter, G. & Moss, B. 1992. Non-replicatingvaccinia vector efficiently expressesrecombinant genes. PNAS, 89: 10847-10851.

Tartaglia, J., Cox, W.I., Pincus, S. & Paoletti,E. 1994. Safety and immunogenicity ofrecombinants based on the geneticallyengineered vaccine strain, NYVAC. Dev.Biol. Stand., 82: 125-129.

Tartaglia, J., Jarrett, O., Neil, J.C., Desmetre,P. & Paoletti, E. 1993. Protection of catsagainst feline leukemia virus by vac-cination with a canarypox virus re-combinant, ALVAC-FL. J. Virol., 67:2370-2375.

Tartaglia, J., Perkus, M.E., Taylor, J., Norton,E.K., Audonnet, J.-C., Cos, W.I., Davis,S.W., VanderHoeven, J., Meignier, B.,

Riviere, M., Languet, B. & Paoletti, E.1992. A highly attenuated strain ofvaccinia virus. Virology, 188: 217-232.

Taylor, J., Trimarchi, C., Weinberg, R.,Languet, B., Guillemin, F., Desmetre,P. & Paoletti, E. 1991. Efficacy studieson a canary-pox-rabies recombinantvirus. Vaccine, 9: 190-193.

Taylor, J., Weinberg, R., Tartaglia, J.,Richardson, C., Alkhatib, G., Briedis,D., Appel, M., Norton, E. & Paoletti,E. 1992. Non-replicating viral vectorsas potential vaccines: recombinantcanarypox virus expressing measles virusfusion (F) and hemagglutinin (HA)glycoproteins. Virology, 187: 321-328.

Taylor, J., Tartaglia, J., Riviere, M., Duret,C., Languet, B., Chappuis, G. & Paoletti,E. 1994. Applications of canarypox(ALVAC) vectors in human and veteri-nary vaccination. Dev. Biol. Stand., 82:131-135.

Townsend, A. & Bodmer, H. 1989. Antigenrecognition by class I-restrictedT lymphocytes. Annu. Rev. Immunol., 7:601-624.

Ulmer, J.B., Donnelly, J. & Liu, M.A. 1994.Vaccination with polynucleotides: anovel means of generating immunity.In E. Kurstak, ed. Modern vaccinology,p. 13-23. New York, Plenum Medical.

Ulmer, J.B., Donnelly, J.J., Parker, S.E.,Rhodes, G.H., Feigner, P.L., Dwarki,V.J., Gromkowski, S.H., Deck, R.R.,DeWitt, C.M., Friedman, A., Hawe, L.A.,Leander, K.R., Martinez, D., Perry, FI.C.,Shiver, J.W., Montgomery, D.M. & Liu,M.A. 1993. Heterologous protectionagainst influenza by injection of DNAencoding a viral protein. Science, 259:1745-1749.

van Oirschot, J.T. 1994. Vaccination in foodanimal populations. Vaccine, 12: 415-418.

Vannier, P., Hutet, E., Bourgueil, E. &Cariolet, R. 1991. Level of virulent virusexcreted by infected pigs previouslyvaccinated with different glycoprotein-

Vaccine manual 33

deleted Aujeszky's disease vaccines. Vet.Microbiol., 29: 213-223.

Wiktor, T.J., Macfarlan, R.I., Reagan, K.J.,Dietzschold, B., Curtis, P.J., Wunner,W.H., Kieny, M.P., Lathe, R., Lecocq,J.P., Mackett, M., Moss, B. & Koprowski,H. 1984. Protection from rabies by avaccinia virus recombinant containingthe rabies virus glycoprotein gene. PNAS,81: 7194-7198.

Williams, R.S., Johnson, S.A., Reidy, M.,Devil, M.J., McElligott, S.G. & Sanford,J.C. 1991. Introduction of foreign genesinto tissues of living mice by DNA-coatedmicroprojectiles. PNAS, 88: 2726-2730.

Wolff, J.A., Malone R.W., Williams, P.,Chong, W., Acsadi, G., Jani, A. &Feigner, P.L. 1990. Direct gene transferinto mouse muscle in vivo. Science, 247:1465-1468.

Xiang, Z.Q., Spitalnik, S., Tran, M., Wunner,W.H., Cheng, J. & Ertl, H.C. 1994.Vaccination with a plasmid vectorcarrying the rabies virus glycoproteingene induces protective immunityagainst rabies virus. Virology, 199:132-140.

Yilma, T. 1994. Genetically engineeredvaccines for animal viral diseases. J. Am.Vet. Med. Assoc., 204: 1606-1615.

Vaccination is a major weapon in thecontrol of many viral diseases of humansand their domestic and pet animals. Thereis no doubt that vaccines have made anenormous impact on the health andconsequently the productivity of therecipients. Although this chapter concen-trates on vaccines which are used indomestic livestock, where necessary itdraws analogies with information involv-ing other species. In addition, an overviewon the present state of vaccine devel-opment must necessarily review what hasgone before and, with advantage, shouldalso look to the future. Consequently, thischapter contains references to productsthat were made in the past, including theway in which they were improved toprovide the vaccines in current use, andoutlines the possibilities of entirely newproducts which molecular biology offers.

Although it is always dangerous to makepredictions, one can be certain that themethods for preparing vaccines willimprove. It is also possible and even likelythat the methods will change radically aswe learn more about the immune responseat the molecular level and apply re-combinant DNA technology to produceand present to the host only those parts ofthe virus that are required to provideprotection.

BACKGROUND

Vaccination as we know it today essentiallybegan late in the eighteenth century withthe observations of Jesty (see Wallace, 1981;Jenner, 1798) and many others thatmilkmaids rarely contracted smallpox. Theimmunity was attributed to the fact that

acciiF. Brown

milkmaids contracted cowpox duringmilking which protected them against thevirulent smallpox. Many of the vaccines inuse at the present time are based on thesame principle, namely that of a weakenedor attenuated strain of a virus multiplyingsufficiently in the host to elicit a protectiveimmune response without causing clinicaldisease. Vaccines based on this principleare highly effective, as shown by theeradication of smallpox; the control ofmany human diseases such as poliomyelitis, measles and mumps; and theeffectiveness in domestic animals ofvaccines against Marek's disease,Newcastle disease and rinderpest.

The second part of our armoury consistsof inactivated vaccines, prepared bygrowing large amounts of the viruses intissue culture cells or sometimes in theintact animal and then inactivating themeither chemically, with agents such asformaldehyde, phenol, [3-propiolactone oran imine, or physically, with ultravioletlight. Inactivated vaccines against foot-and-mouth disease, Newcastle disease,poliomyelitis and rabies have been usedextensively and with great success.

Together, these vaccines have made veryimportant contributions to animal healthworldwide. One of the early vaccines usedon a large scale was described byRosenbusch, Decamps and Gelormini(1948). This vaccine, used against foot-and-mouth disease, was prepared byformaldehyde inactivation of extracts offresh tongue lesions of artificially infectedsusceptible cattle. Millions of cattle wereimmunized with vaccines prepared by thismethod. It is of interest, in view of later

Vaccine manual35

observations and concern about theinnocuity of vaccines prepared byinactivation with formaldehyde, that theauthors used an alkaline glycocoll buffer.Recent studies have shown that the virusis very rapidly inactivated at pH 8. Thisresults from the degradation of the RNAwithin the virus particle by the RNApolymerase (Newman et al., 1994). Rabiesvaccines prepared from the brains of sheepand goats artificially infected with the virusare still widely used in humans despite theevidence that such products causeencephalitic reactions.

The discovery of antibiotics in the late1930s and early 1940s revolutionized theway in which vaccines could be prepared.Viruses had been grown outside the animalbody as early as 1930 but these experimentshad been done on a small scale. To expandthese laboratory observations to aproduction scale was clearly difficult withthe technology available at that timebecause the problem of ensuringbacteriological sterility would have beenoverwhelming. The discovery of antibioticsovercame the problem.

The development of large-scale in vitrocultivation of foot-and-mouth disease virusby Frenkel (1947) was a major technicalachievement. This led to vaccinationprogrammes which have been used toprotect millions of animals (andincidentally predated the much publicizedpolio vaccination programmes by severalyears).

ATTENUATED VACCINES

Jermer's vaccine against smallpox, referredto above, was quickly applied in manycountries but it was not until almost acentury later, in the 1880s, thatimmunization against other infectionsbegan. By that time the germ theory ofinfection was starting to be accepted and,in rapid succession between 1880 and 1885,Pasteur and his colleagues demonstrated

the efficacy of vaccines against chickencholera, anthrax and rabies (Pasteur, 1880and 1985; Pasteur, Chamberland and Roux,1881). All three vaccines had been preparedby attenuating the naturally occurringwild-type agents in the laboratory.

Vaccines for several viral diseases havebeen prepared by growing the wild-typeagents in non-natural hosts or in tissueculture cells. For success to be achieved,careful selection of appropriate strains isrequired. Even then, the selection has to beaccompanied by an enormous amount ofluck to achieve the correct balance betweenloss of virulence and the ability to multiplyin the host without causing clinical disease.Nevertheless, highly successful vaccinesagainst Newcastle disease and rinderpesthave been prepared and used in millionsof birds and animals. The selection of anaturally occurring mutant has also beensuccessful in the case of Marek's disease.The virus infecting turkeys was found tobe avirulent for chickens and has been usedin millions of birds.

CURRENT METHODS OF INACTIVATION

Most inactivated vaccines are produced byreacting the viruses with formaldehyde.This reagent has been used almostexclusively for producing diphtheria andtetanus toxoids since early this century.The reagent was used by Römer inexperiments to produce a vaccine againstpoliomyelitis as early as 1911, but withoutsuccess. However, it was used successfullyby Vallée, Carré and Rinjard (1925) toproduce an experimental foot-and-mouthdisease vaccine. These researchers, locatedin Paris, were probably influenced in theirchoice of reagent by Ramon, who was atthat time director of the Pasteur Institutein that city.

Serious concerns about the use offormaldehyde to inactivate poliomyelitisvirus completely were voiced followingclinical trials of a polio vaccine prepared

36 Viral vaccines

in this way by Brodie and Park (1936) inthe 1930s. Similar doubts were alsoexpressed later about the foot-and-mouthdisease vaccines prepared by the samemethod. It was generally considered thatthe preliminary step of adsorbing the virusto alum before adding the formaldehydewas necessary in the inactivation pro-cedure. In fact, the adsorption step mayhave been responsible for masking thefact that inactivation of the virus wasincomplete.

The Cutter incident with an early poliovaccine in the mid-1950s only served toemphasize these concerns and categoricalevidence that foot-and-mouth diseasevaccines prepared by treatment of the viruswith formaldehyde were not innocuouswas provided by Brown et al. (1963) andGraves (1963). But it needed the outbreaksin Normandy and Brittany in France in1981, when the causal agent was identifiedby molecular methods as a virus isolated16 years previously, to convince manu-facturers that formaldehyde inactivationwas not safe (King et al., 1981). Subsequentevidence, provided by Beck andStrohmaier (1987) using nucleic acidsequencing to identify the viruses causingseveral outbreaks in western Europe,pinpointed either spills from factories orvaccines inactivated with formalin as themajor causes.

In view of these findings, it is somewhatperplexing, at least to the author, thatalternative inactivating agents were notused more generally before these incidents.Firm laboratory evidence had beenprovided several years previously thatacetylethyleneimine was a better inactivant(Brown and Crick, 1958; Brown et al., 1963)and this had been confirmed on alarge industrial scale by the WellcomeLaboratories. In recent years, bis-ethyl-eneimine has replaced acetylethyleneiminebecause it is safer to handle on a largescale (Bahnemann, 1990).

THE NEW TECHNOLOGY

During the past 40 years, the concepts ofmolecular biology have been embraced bythe entire field of biology and medicine.This has led to a great increase in ourknowledge, not only of the molecularstructure of many infectious agents butalso of those elements which conferprotective immunity. More recently, thesame concepts have been applied to studiesof the immune responses which protectagainst infection. The objectives of thisapproach, while clearly of fundamentalimportance, also have great practicalimplications for vaccine production, sincethey will eventually provide productswhich are safer and more effective andwhich can be designed to elicit only thoseresponses that are required for protection.

Structure of virusesThe concept that the entire organism is notrequired to elicit protective immunity wasdemonstrated a century ago when it wasshown that antiserum produced in animalsagainst the toxins secreted by the agentscausing diphtheria and tetanus wouldpassively protect against these diseases.The subsequent extension of this work,which showed that these toxins wouldproduce active immunity after suitableinactivation, provided the first cleardemonstrations that vaccination withsubunits was a practical proposition (Glennand Hopkins, 1923; Ramon, 1923).

The advances made with virusesdepended in the first place on thetechniques which became available to growthem in tissue culture in quantitiessufficient to allow their purification andcharacterization. By growing the virusesin the presence of radioactive precursorsof nucleic acids and proteins, theindividual constituents can be labelled.These analyses have shown that there aremany distinct groups of viruses, differingin size and shape, ranging from small

Vaccine manual 37

spherical particles of 30 nm in diameter(such as those of foot-and-mouth diseaseand swine vesicular disease) to bullet-shaped particles measuring 140 x 70 nm(such as that causing rabies) and largespherical particles measuring 300 nm indiameter (e.g. smallpox virus).

In addition, viruses can be groupedaccording to whether they contain DNAor RNA and whether they possess a lipidenvelope. For example, rabies viruspossesses such an envelope. By dissolvingthis envelope in a lipid solvent or a milddetergent, subunits of the virus can beisolated and their potential for elicitingprotective immunity tested. More vigorousmethods are required to disrupt thoseviruses not possessing a lipid envelope.

During the same period, major advanceswere being made in our knowledge of virusmultiplication. These studies haveprovided genetic maps of many groups ofviruses, at the same time identifying thegenes coding for the proteins of im-munological significance. Consequently,when the methods for ligating DNAmolecules were described early in the1970s, the knowledge required for makingsynthetic vaccines was already available.

The steps involved in the provision ofengineered vaccines are:

identification of the immunogenicprotein;identification of the gene coding forthis protein;expression of the gene in a suitablevector to provide either a live or adead vaccine.

i)Identification of immunogenic proteins.The identification of immunogenicproteins is made by dissection of the virusinto subunits which are then tested forimmunogenicity. With lipid-containingviruses this is a relatively simpleprocedure, since the envelope can bedissolved in a solvent such as ether or

chloroform or a mild detergent such asTween, releasing the subunits withoutdenaturing the pro teins. The separatedsubunits are also analysed for the proteinsthey contain. The method which hasproved most valuable for analysingproteins is polyacrylamide gel elec-trophoresis (PAGE), which separates themon the basis of their relative molecularweights. Coupled with examination of thebiologically active subunits in the electronmicroscope, the architecture of a virus canbe derived with this method.

ii)Identification of the gene coding for theimmunogenic protein. The genome of avirus codes for several proteins in additionto the structural proteins which areinvolved in the immune response. Therelevant gene or genes are identified byexpressing the viral nucleic acid in asuitable expression system in vitro andthen precipitating the products withneutralizing antibody. In this way, the genecoding for the product reacting with theneutralizing antibody can be identified.

iii) Expression of genes. Once a relevantgene has been identified, several vectorsystems are available for its expression. Ifthe gene is part of a DNA virus genome, itcan be ligated directly into the DNA of thevector. If, however, the gene is part of anRNA virus genome, it must first betranscribed into DNA before it can beinserted into the vector DNA.

Several vectors are suitable and thesecan be used in two ways. In the first, therelevant gene is ligated into the DNA of abacterium or virus which has been used,or has potential for use, as an attenuatedvaccine. Examples of these are the strainbacillus Calmette-Guérin (BCG) (Bloom etal., 1990), avirulent strains of Salmonellatyphimurium (Dougan and Tite, 1990),adenovirus (Graham and Prevec, 1992), the

38 Viral vaccines

poxviruses notably vaccinia virus(Mackett, 1990) and some members ofthe herpesvirus family, for examplepseudorabies virus (Kit et al., 1991).

Recently, this approach was adapted byAlmond and Burke (1990) to make use ofthe attenuated strains of poliovirus, anRNA virus, taking advantage of the factthat the DNA complementary to the virusRNA is infectious (Racaniello andBaltimore, 1981). This DNA will infect cellsto produce poliovirus particles. The DNAcorresponding to the gene of interest isinserted into the complementarypoliovirus DNA, which then replicates toproduce polio virus particles containingthe inserted gene product. Therecombinant organisms obtained in thisway have the potential to be used as livevaccines.

In the second approach, the gene isexpressed in Escherichia coli, Saccharomycescerevisiae, the baculovirus Autographacalifornica or in mammalian cells, and thegene product is purified for use as a killedvaccine (Brown, 1984; Matsuura et al.,1987). The initial early preference forE. coli has been largely superseded becausethe bacterial vector does not glycosylatethe expressed protein. Since many of theproteins of immunogenic importance areglycosylated (e.g. the surface projectionsof rabies virus) this was a serious and oftencritical disadvantage. Consequently, morerecent developments have been made witheukaryotic vectors.

Recombinant attenuated vaccinesThe use of attenuated viruses and bacteriathat are already accepted vaccines to carryforeign genes is clearly a concept withmany attributes. For example, the greatadvantages of vaccinia virus (Mackett,1990) as a vector are its extensive andsuccessful use against smallpox and in theeventual eradication of the disease,together with the fact that its DNA can

accommodate several foreign genes.Similar considerations apply to BCG(Bloom et al., 1990) which has been used intuberculosis prophylaxis for many decadeswith an impressive safety record, eventhough there is some debate about itsefficacy. In contrast, poliovirus, with itssmaller genome and more structurallyconstrained particle, would be unlikely toaccommodate more than one foreign gene.

The use of Salmonella typhimurium(Dougan and Tite, 1990) as a carrier is moredebatable as it does not possess the proventrack record of vaccinia virus and BCG.However, its potential for oral immuni-zation demands that its qualities as a vectorshould be investigated extensively. Strainsof adenovirus (Graham and Prevec, 1992)which have been used as vaccines shouldalso prove valuable since they can bedelivered orally, provided the structuralconstraints imposed by the architecture ofthe particle do not prove too stringent.Moreover, research done with pseudo-rabies virus (Kit et al., 1991), a member ofthe herpesvirus family, indicates that thisvirus could be a valuable vector for theimmunization of animals.

The vaccinia virus system has beenstudied more extensively than the othercandidates, and field trials with the rabiesvirus glycoprotein recombinant haveproved very successful. Of particularinterest in the control of rabies has beenthe successful vaccination of the fox in thewild by using the recombinant vaccine inbait. Experiments with rinderpest virusrecombinant in the natural host animalhave also been very successful (Yilma etal., 1988). A major issue of debate in theuse of such recombinant, however, hasbeen the question of their safety. Theinnocuity of an organism and in thisinstance a new recombinant virus for.thelaboratory mouse can be no guarantee ofits safety for other species.

It is also well recognized that any

Vaccine manual 39

product which is to be inoculated into ahealthy animal should ideally not causeany side effects. This position is difficult toachieve and there seems to be no idealsolution. Another issue which has causedconcern is whether insertion of a foreigngene alters the tissue tropism of theorganism, or even its host range. Againthere seems to be no easy solution,particularly since we have little knowledgeof the factors which determine hostspecificity.

Killed vaccinesProteins produced in in vitro systems areessentially equivalent to the subunitvaccines used experimentally several yearsago. The crucial problem is to present theexpressed proteins to the immune systemin the configuration which they have whenthey form part of the intact organism. Boththe subunits derived from the intactorganism and the corresponding gene-tically engineered proteins are much lessimmunogenic than the intact organism.This is particularly true when the subunitsare derived from viruses which do notcontain a lipid envelope because theirrelease from the intact organisrn requiresmuch harsher conditions.

There are, however, lessons to be learnedfrom the experience with the hepatitis Bvirus surface antigen (Valenzuela et al.,1983). The protein expressed in yeast cellsaggregates to form particles which aresimilar to the 22 nrn particles found in theblood of carriers of the disease. These yeast-expressed particles are much more im-munogenic than the monomeric proteinand they form the basis for the firstgenetically engineered vaccine. Otherevidence also indicates that presentationof an immunogenic protein in a multimericform on a particulate structure greatlyenhances its activity.

Most proteins expressed in vitro have alow immunogenicity. By presenting them

in a multimeric form, most clearly demons-trated by Morein's experiments withimmunostimulating complexes (ISCOMs),their immunogenicity is greatly enhanced(Morein et al., 1990). The principle in thispresentation system is the creation of aparticle comprising several copies of theprotein. The particles are composed ofsaponin, cholesterol, phosphatidyl cholineand the protein under study, the constit-uents being held together by hydrophobicinteractions. The ISCOMS are seen as cage-like structures in the electron microscope,about 40 nm in diameter and comprising12 nm morphological subunits. Proteinspresented in this way not only elicit highlevels of antibody but they also stimulatecell-mediated immunity. Moreover, pre-liminary results indicate that intranasaladministration of ISCOMs containinginfluenza and measles haemagglutininsprotect mice against experimentalch.allenge.

Peptides as immunogensProducing a protein, either in vitro or in

vivo, which has the same amino acidsequence as in the parent organism doesnot provide any information about theepitopes or short amino acid sequenceswhich elicit the protective immuneresponse. Indeed, the methods which havebeen used so far to express immunogenicproteins do not even attempt to organizethe structure of th.e protein into theconfig-uration which it has on the parentorganism. It is fortuitous if a vaccineproduced in vitro or in vivo is effectivebecause it is improbable that a proteinproduced in a foreign environment willfold in the way it does in the parentorganism. All that has been done so far isto produce immunogenic proteins in amilieu which is different from the nativeorganism. This is not to deny that this is aremarkably sophisticated biochemicalachievement but it does not provide any

40 Viral vaccines

information about how to achieve thespecific immune responses we seek fromrationally designed vaccines. Theseanswers can only be obtained by furtherdissection of the proteins into activepeptide fragments.

There are several reasons for pursuingsuch a strategy (Brown, 1990):

chemical synthesis allows theproduction of stable products whichdo not have the problems associatedwith materials produced in cells;

the approach offers the advantagesassociated with the handling of smallmolecules compared with much largerprotein molecules;the information which is accumulatingon the interaction of peptides withMEIC molecules (Bjorkman et al., 1987;J.H. Brown et al., 1993) suggests thatthe immune response to proteins willsoon be understood at the level of shortamino acid sequences.

These considerations indicate that it willbe possible to design peptides which willelicit the required immune responses forany pathogen. Clearly, there is still muchto be learned about the componentsrequired in a totally synthetic vaccine.Immunodominance plays a major part inB cell recognition and this factor is nowbeing seen as important in T cellrecognition also. Optimal partnerships ofB and T cell epitopes are the clear goal.

Genetic immunizationDuring the past three years, a quiterevolutionary conceptual advance hasbeen made in vaccination. It had alwaysbeen held that it would be necessary topresent the immunogenic protein (orcarbohydrate-protein complex) to the hostin order to elicit the appropriate immuneresponse. Recent experiments withseveral viruses have shown that immuneresponses can be obtained by injecting theDNA coding for the protective antigens

directly into the muscle tissue of mice(Ulmer et al., 1993). In the case of influenza,protective immunity was achieved. Theproteins coded by the inserted genes areexpressed in the muscle and skin cells andare then presented to the immune system.It appears that most of the DNA isdegraded but, nevertheless, enoughremains for the expression of sufficientprotein to stimulate the appropriateimmune response. It would be interestingto compare the level of the immuneresponse elicited by this method with thatproduced by the direct inoculation of theprotein for which the DNA codes.

Unlike recombinant proteins made intest tubes, if the proteins were madedirectly in living animals they would nothave to go through the extensivepurification steps which are currentlydemanded. Provided the immune responseis sufficient to afford protection, thismethod appears to provide a major steptowards the provision of new vaccines.

SUMMARY

The study of vaccines and vaccinationbased on the concepts of molecular biologyhas provided the expectation that newvaccines will be produced which elicitprotective immune responses without thedisadvantageous side effects of the currentproducts. As we learn more about thoseparts of a virus which are required toprovide immunity and the way in whichto present them to the host, it is anticipatedthat vaccines will consist of small frag-ments of the agent, synthesized eitherchemically or biochemically, without theneed to grow the agent itself. More specific-ally, however, it must be recognized thateach disease is a separate challenge andthe protective immunity which eachdemands should be considered according-ly. The crucial issue is to identify theimmune response which correlates withprotection. It should not be forgotten how

Vaccine manual 41

effective the empirical approach has beenand we should not ignore the lessons thathave been learned from these successes.

BIBLIOGRAPHY

Almond, J.W. & Burke, K.L. 1990. Poliovirus as a vector for the presentation offoreign antigens. Semin. Virol., 1: 120.

Bahnemann, H.B. 1990. Inactivation of viralantigens for vaccine preparation withparticular reference to the applicationof binary ethylenimine. Vaccine, 8:299-303.

Beck, E. & Strohmaier, K. 1987. Subtypingof European foot-and-mouth diseasevirus strains by nucleotide sequencedetermination. J. Virol., 61: 1621-1629.

13j orl:man, P.J., Saper, M.A., Samraoui, .3.,Bennett, W.S., Strominger, J.L. & Wiley,

, D.C. 1987. Structure of the human classI histocompatibility antigen, HLA-A2.Nature, 329: 506-512.

Bloom, B.R., SnaPper, S.B., Kieser, T. &Jacobs, W.R. 1990. Development ofrecombinant BCG vaccines. Semin. Viral.,1.: 21-27.

Brodie, M. & Park, W.H. 1936. Activeimmunisation against polio myelitis. Am.J. Publ. Health, 26: 119-125.

Brown, F. 1984. Synthetic viral vaccines. Annu.Rev. Microbial., 38: 221-235.

Brown, F. 1990. The potential of peptides asvaccines. Semin. Viral., 1: 67-74.

Brown, F. & Crick, J. 1958. Application ofagar gel diffusion analysis to a study ofthe antigenic structure of inactivatedvaccines prepared from the virus of foot-and-mouth disease. J. Immunol., 82:444-447.

Brown, F., Hyslop, N.St.G., Crick, J. &Morrow, A.W. 1963. The use ofacetylethyleneimine in the productionof inactivated foot-and-mouth diseasevaccines. J. Hyg., 61: 337-344.

Brown, F., Dougan, G., Honey, E.M., Martin,S.J., Rima, B.K. & Trudgett, A., eds.

1993. Vaccine design. Chichester, UK,Wiley.

Brown, J.H., Jardetzky, T.S., Gorga, J.C.,Stern, L.J., Urban, R.G., Strominger,J.L. & Wiley, D.C. 1993. Three-dimensional structure of the human classII histocompatibility antigen, HLA-DR1.Nature, 364: 33-39.

Dougan, G. & Tite, J. 1990. Live bacterialvaccines for delivering antigens to themammalian immune system. Semin.Viral., 1: 29-37.

Frenkel, H.S. 1947. La culture du virus dela fièvre aphteuse sur l'épithelium dela langue des bovidés. Bull. Off. int.Epizoot., 28: 155-162.

Glenn, A.T. & Hopkins, B.E. 1923. Diphtheriatoxoid as an immunising agent. Brit. J.Ex per. Pathol., 4: 283-288.

Graham, F.L. & Prevec, L. 1992. Adenovirus-based expression vectors and recom-binant vaccines. In R.W. Ellis, ed.Vaccines: new approaches to immunologicalproblems, p. 363-390. Stoneham, Mass.,USA, Butterworth-Heinemann.

Graves, J.H. 1963. Formaldehyde inactivationof foot-and-mouth disease virus asapplied to vaccine preparation. Am. J.Vet. Res., 24: 1131-1135.

enner, E. 1798. An inquiry into the causesand effects of the variolae vaccine, a diseasediscovered in some of the western countiesof England, particularly Gloucestershire,and known by the name of smallpox. London,Samson Low.

King, A.M.Q., Underwood, B.O., McCahon,D., Newman, J.W.I. & Brown, F. 1981.Biochemical identification of virusescausing the 1981 outbreaks of foot-and-mouth disease in the U.K. Nature, 293:479-480.

Kit, M., Kit, S., Little, S.,Di Marchi, R. &Gale, C. 1991. Bovine herpesvirus-1(infectious bovine rhinotracheitis virus)based viral vector which expresses foot-and-mouth disease epitopes. Vaccine, 9:564-572.

42 Viral vaccines

Mackett, M. 1990. Vaccinia virus as a vectorfor delivering foreign antigens. Semin.Virol., 1: 39-47.

Matsuura, Y., Possee, R.D., Overton, H.A.& Bishop, D.H.L. 1987. Baculovirusexpression vectors: the requirements forhigh level expression of proteins, in-cluding glycoproteins. J. Gen. Virol., 68:1233-1250.

Morein, B., Fossum, C., Lovgren, K. &Höglund, S. 1990. The ISCOM - amodern approach to vaccines. Semin.Viral., 1: 49-55.

Newman, J.F.E., Piatti, P.G., Gorman, B.M.,Burrage, T.G., Ryan, M.D., Flint, M. &Brown, F. 1994. Foot-and-mouth diseasevirus particles contain replicase protein3D. Proc. Natl Acad. Sci. USA, 91: 733-737.

Pasteur, L. 1880. De l'atténuation du virusdu choléra des poules. CR Acad. Sci.,673-680.

Pasteur, L. 1885. Méthode pour prévenir larage après morsure. CR Acad. Sci.,765-772.

Pasteur, L., Chamberland, C.-E. & Roux, E.1881. Sur la vaccination charbonneuse.CR Acad. Sci., 92: 1378-1381

Racaniello, V.R. & Baltimore, D. 1981.Cloned poliovirus complementary DNAis infectious in mammalian cells. Science,214: 916-919.

Ramon, G. 1923. Sur le pouvoir floculantsur les propriétés immunisantes d'unetoxine diphtérique, rendue anatoxique(anatoxine). CR Acad. Sci., 177: 1338-1340.

Römer, P.H. 1981. Die epidemische Kinder-la7rmung (Heine-Medinsche Krankheit), p.49. Berlin, Germany, Springer.

Rosenbusch, C.T., Decamps, A. & Gelormini,N. 1948. Intradermal foot-and-mouthdisease vaccine: results obtained fromthe first million head of cattle vaccinated.J. Am. Vet. Med. Assoc., 112: 45-47.

Ulmer, J.B., Donnelly, J.J., Parker, S.E.,Rhodes, G.H., Feigner, P.L., Dwarki,V.J., Gromkowski, S.H., Deck, R.R.,

DeWitt, C.M., Friedman, A., Hawe, L.A.,Leander, K.R., Martinez, D., Perry, H.C.,Shiver, J.W., Montgomery, D.L. & Liu,M.A. 1993. Heterologous protectionagainst influenza in injection of DNAencoding a viral protein. Science, 259:1745-1749.

Valenzuela, P., Medina, A., Rutter, W.J.,Ammerer, G. & Hall, B.D. 1983.Synthesis and assembly of hepatitis Bvirus surface antigen particles in yeast.Nature, 298: 347-350.

Vallée, H., Carl* H. & Rinjard, R. 1925.On immunisation against foot-and-mouth disease. Rech. Med. vet., 101:297-299.

Wallace, E.M. 1981. The first vaccinator:Benjamin Jesty of Yetrninster and WorthMatravers and his family. Wareham,Dorset, UK, Anglebury Bartlett.

Yilma, T., Hsu, D., Jones, L., Owens, S.,Grubman, M., Mebus, C., Yamanaka,M. & Dale, B. 1988. Protection of cattleagainst rinderpest with vaccinia virusrecombinants expressing the HA or Fgene. Science, 242: 1058-1061.

Vaccine manual 4,3

Bacterial vaccines can be divided into twocategories:

Inactivated vaccines which arecomposed of whole bacterial cellsand/ or their metabolic products or,alternatively, a purified fraction ofthe cell.Live vaccines which are composedof attenuated strains of the parentvirulent organisms.

Although in the developed world,particularly with medical vaccines,purified and characterized imrnunogensare preferred from both a safety and qualitycontrol viewpoint (Walker, 1991), in non-domestic animals economic considerationsdictate that the product, in addition tobeing efficacious and safe, should be cost-effective in an animal of finite worth.Under these circumstances, cruder, lesspurified products of proven efficacycontinue to be employed. In addition,protection by vaccination is often multi-factorial, involving a number of antigens,while the use of single entities producedeither by conventional purification orgenetic engineering should be treated withsome caution (Walker, 1992).

Both types of vaccine can either beinjected parenterally or administeredorally. Inactivated vaccines are invariablyadministered by injection, although largedoses administered orally have also beenused successfully. Living vaccines areadministered parenterally or, whereprotection is against enteric diseases,orally. The use of the oral route as a meansof stimulating mucosal immunity ingeneral is being increasingly exploited,particularly using genetically engineered

P.D. Walker

zlnes

organisms such as the aro mutants ofSalinonella spp. as cloning vectors (Hoisethand Stocker, 1981; Hone et al., 1991). Suchvaccines stimulate secretory, humoral andcell-mediated responses. Vaccines aredesigned to stimulate the normalimmunological mechanisms of the host.Following parenteral injection of vaccines,antibodies, mainly IgG, appear in thebloodstream. Vaccines administered orallystimulate local production of antibody(mainly IgA) at the mucosal surface. Ingeneral, antigenicity is a function ofmolecular size. For this reason, bacterialproteins are good antigens, although longchain capsular polysaccharides are alsoantigenic and have formed the basis ofsuccessful vaccines, for example in pro-tection against haemorrhagic septicaemiain cattle (Nagy and Penn, 1976).

In developing a vaccine strategy forthe control of disease, the vaccine manu-facturer has to be mindful that there aretwo stages in infectious disease, i.e.establishment of the organism in the hostfollowed by the clinical signs of infection.Although, in practice, these two stagesquickly pass from one to the other, fromthe point of view of the vaccine manu-facturer, it is useful to distinguish betweenthem. Vaccines can be designed that eitherprevent establishment of the organism inthe host or prevent the effects of infection.The problems posed to the manufacturerdiffer depending on which approach istaken (Walker and Foster, 1981).

The effects of infection are caused mainlyby toxins elaborated by the organismswhich have specific pharmacologicaleffects. Indeed, the results of infection can

Vaccine manual 45

be reproduced by injecting sterile filtratesof the organisms concerned. Bacterialtoxins are usually secreted into the culturemedium and the manufacturer is thereforefaced with the problem of producing largequantities of toxins for further processing.These are usually separated from thebacterial cell mass.

On the other hand, if the approach is toprevent establishment of infection then itis the antigens of the bacterial cell whichthe manufacturer needs to consider. Forexample it has been shown that in manydiseases adherence of the organism toepithelial tissues by specific adherencemechanisms followed by multiplication ofthe attached organisms is important(Walker and Nagy, 1980). In this con-nection, the manufacturer must identifythese particular bacterial antigens andadjust the growth conditions to ensure theproduction of large quantities of theappropriate antigen. After inactivation, theorganisms are harvested for furtherprocessing. Whole culture vaccines maystimulate both types of immunity.

This overview deals with both types ofprotection using illustrations fromveterinary vaccines which are currently inuse for sheep, cattle, pigs and dogs orwhich are under development. A briefsection on fish vaccines is included, asaquaculture may be of increasing relevanceto developing countries in the future.

PROTECTION AGAINST THE EFFECTS

OF DISEASE

The classic example of the use of this typeof vaccine protection is that of clostridialtoxoid vaccines. The most economicallyimportant diseases in sheep, cattle and pigsare caused by clostridial infections (Sterneand Batty, 1975). As a result of the stimulusto research of the potential threat of gasgangrene during the First and SecondWorld Wars, a vast amount of literatureexists on the toxins produced by these

organisms (Oakley, 1943; Oakley andWarrack, 1953; Brooks, Sterne andWarrack, 1957). The major toxins res-ponsible for disease have been identifiedand the use of toxoid vaccines to controlclostridial diseases is widely practised(Sterne et al., 1962). Bacterial toxins areinactivated to produce bacterial toxoids insuch a way that toxicity is lost butantigenicity retained. The usual method oftreatment is with formaldehyde (Ramon,1924).

Toxoids are normally adsorbed on to anadjuvant, usually a mineral salt such asaluminium hydroxide, aluminium phos-phate or potassium aluminium sulphate.By far the most potent adjuvants aremineral oils or their derivatives (seeTable 2), which enhance immunogenicityby at least an order of magnitude overconventional adjuvants (Freund, Casals-Ariet and Genghof, 1940). However, owingto their production of severe local reactionswhich may persist indefinitely, they arecompletely unacceptable for human vac-cination and even in veterinary medicine,despite their effectiveness, their use in thedeveloped world is severely limited owingto potential carcass blemish and its effecton the export of meat. Attempts have beenmade to overcome this problem, forexample by the use of the interperitonealroute (Thomson et al., 1969). However, inthe developing world it is questionablewhether these issues are important. Whereproduction is for a domestic market, thebenefits of a more effective and lastingimmunity, with all the advantages ofreduced handling of animals, must surelyoutweigh any cosmetic considerations.Furthermore, it is quite possible to identifyan injection site such as behind the ear oron top of the skull which is removed atslaughter and will not affect theappearance of the meat.

Adult animals are normally given twoinjections of toxoid separated by at least 28

46 Bacterial vaccines

Vaccine manual

TABLE 2

Arithmetic mean antitoxin responses of lambs to 2x2 ml doses of a multicomponent clostridialvaccine precipitated with potash alum and to a I ml dose adsorbed with Freund's adjuvant

days for a primary immunization course.Thereafter, further boosting to theimmunity can be given at critical periodsof life. One such period coincides with the14 days preceding parturition to ensurethat the maximum amount of antibody istransferred from the serum via thecolostrum and milk for the protection ofthe newborn. Following the disappearanceof such protection, the young animalshould undergo prophylactic vaccination.

The range of clostridial diseases againstwhich vaccination is practised is shown inTable 3. Both single and combined vac-cines are available, with individual manu-facturers responding to the needs of themarket place. Based on experience, manu-facturers have been able to formulatemulticomponent products so that there isa balanced response to all the majorantigens despite antigenic competition(Table 4). It is clear from the table that,although individual antigenic responsesmay be lower with a multicomponent thanwith an equivalent monocomponentvaccine, the responses to the multi-component vaccine are above the mini-mum responses required by the regulatoryauthorities.

In the United Kingdom, in order toprovide effective and safe vaccines,manufacturers have had to comply withthe conditions laid down in the BritishPharmacopoeia (Veterinary) (HMSO, 1985)

whose requirements have subsequentlybeen incorporated in the EuropeanPharmacopoeia. These regulations wereoriginally formulated after consultation

TABLE 3

Principal Clostridia spp. causing diseasesin animals

47

Weeks aftersecondinnoculation

Adjuvant Clostr.dium

perfringens septicum novyi tetani

beta epsilon

Two Alum 25.4 2.2 1.9 10.2 16.0Freund's 46.6 12.9 5.7 13.2 27.8

Ten Alum 1.1 0.6 0.38 0.7Freund's 5.6 3.1 0.6 1.3 6.7

Species Disease

Cl. perfringens Wound infections, gangrenoustype A mastitis, enterotoxaemia in

nursing lambs in Californiaand Oregon ("yellow lamb")

Cl. perfringens Enterotoxaemias in varioustypes B, C and D animals including calves,

sheep, goats, piglets andfoals

Cl. septicum Malignant oedema of horses,cattle, sheep and swine

CI. chauvoei Blackleg in sheep and cattle

Cl. novyi Gas gangrene in cattle andtype A sheep

Cl. novyi Black disease in sheep,type B sudden death in cattle and

pigs

CI. haemolyticum Bacillary haemoglobinuria(Cl. novyi type D) in cattle

Cl. sordellii Gas gangrene

Cl. tetan! Tetanus in all speciesof domestic animals

Cl. botulinum Botulism in sheep, cattle,type C and D dogs, chickens and wild

duck

TABLE 4

Arithmetic mean antitoxin response of groups of rabbits in tests on 10 consecutive batches ofcombined vaccines compared with tests on 12 consecutive single batches

with major manufacturers who, based ontheir own experiences of vaccines that wereeffective in the field, were able to suggestconsensus criteria for the release ofproducts. In the case of clostridial vaccines,with the exception of Clostridium chauvoei,in order to release a product a manu-facturer has to show that, after twoinjections given according to the recom-mended dosage regime in the field, groupsof rabbits respond with certain minimumantitoxin titres. Subsequent formal studieshave shown that vaccines producingadequate titres in rabbits produce anadequate serological response in the targetspecies (Frerichs and Gray, 1975).

Nevertheless, these criteria have somedrawbacks. For example, they onlymeasure antibodies to the major lethalcomponents and not to other toxins andcellular elements which may be importantin protection. Thus, they do not distinguishbetween whole-culture and toxoid vac-cines, for example. A further complicationis the differences in response by individualbreeds of rabbit, which can vary threefoldin their response to the same vaccinepreparation (Walker and Batty, 1985). Inthe case of Cl. chauvoei, groups of guinea

pigs are vaccinated with two doses of thevaccine according to the recommendeddosage regime and challenged with avirulent culture 14 days after the secondinjection. All vaccinates should surviveand all controls die within 48 hours.

The problem for the manufacturer is thento produce large quantities of toxins thatcan be converted into toxoids and, fol-lowing suitable processing, incorporatedinto vaccines.

Component manufactureAs a result of developments in fermentertechnology, some manufacturers are ableto grow clostridial species in fermentationculture to produce high yields of clostridialtoxins suitable for toxoiding (Walker andFoster, 1981). Over a period of time,manufacturers have carefully selectedthose strains which give maximum toxinproduction in the growth media they havedeveloped. They have, in effect, cloned theirstrains to give stable high-yielding toxinmutants which enable them to produceconsistently high yields of toxins. Strainimprovement by mutation and selectionforms the basis of a continuing researchprogramme by most manufacturers.

Vaccinesused

Responses in IU per ml to Clostridium spp.

Cl. zvelchiibeta

Cl.epsilon

Cl. septicum Cl. oedematiens Cl. tetani

Br. Vet.Codexstandards

Mean responsesto 10 combinedbatches

Mean responses to12 singlebatches

10

55

141

5

6

15

2,5

9,2

9

10

16

122

3.5

62

Note: IU international units.


The majority of strains are stable whenstored as freeze-dried cultures and thisenables large batches of master seedcultures to be stored. By the use of a freshaliquot of the master seed to prepareworking seeds, any deterioration of thestrain owing to passage in liquid mediacan be avoided. Where storage by freeze-drying is not satisfactory, cultures can bestored in liquid nitrogen and freshampoules opened as required.

As bacterial toxins are not generallyproduced in significant quantities insynthetic media, it is necessary to providea source of peptides. The manufacturer hasto select a source of protein for digestionwhich is readily available, for examplemeat, casein or soybean. Digests of meatare usually prepared with papain ortrypsin and digests of casein with acid ortrypsin. For maximum toxin production,digest media need to be supplementedwith various amino acids and vitamins;yeast extract provides a convenient,reliable and cheap source for the latter. Inthe case of certain toxins, it is necessary toregulate carefully the amounts of variousinorganic ions present, for example iron intetanus toxin production (Mueller andMiller, 1954). Glucose is the normal energysource although maltose and sucrose arealso used (Moore, 1968; Walker, Harris andMoore, 1971).

Manufacturers have invested heavily indeveloping media and growth conditionsto give maximal toxin production(Thomson, 1979). Published data areavailable for the growth and toxinproduction of the Harvard strain of Cl.tetani using Mueller medium (Mueller andMiller, 1954). This consists of a calciumcaseinate digest and acid hydrolysed caseinbase supplemented with bullock's heartinfusion broth, glucose, vitamins, cystineand iron powder.

The oxidation-reduction potential offreshly prepared digest media after

autoclaving is usually sufficiently low topermit the growth of almost all species ofClostridia without further additions,particularly when combined with the useof 5 to 10 percent inocula. If necessary,reducing agents such as cysteine hydro-chloride can be added prior to inoculation.There is therefore no special requirementfor the introduction of gases into theculture.

Clostridia spp. are for the most partsaccharolytic, acid-producing organismsand automatic pH control and addition ofcarbohydrate results in considerableimprovements in yield and reproducibilityof culture. Nevertheless, it is possible tocontrol these processes manually, albeit atthe expense of some efficiency. By usingbottled media and working in an incubator,it is possible to control pH and makeperiodic additions to containers using apH meter and appropriate solutions.Where the costs of fermenters are una-cceptable and adequate labour is available,this can be an effective method of produc-tion, particularly for developing countries.

The optimum pH range for maximumtoxin production varies with the toxinconcerned. For example, maximum pro-duction of epsilon toxin by Cl. perfringens(welchii) type D occurs in the alkaline rangewhile the pH optima for other clostridialtoxins are lower.

Toxin production during growth can bemonitored by a variety of techniques.These include, where present, the mea-surement of enzymic activity, for examplelecithinase, haemolysis and proteolysis or,alternatively, in vivo testing in guinea pigs,mice and rabbits, using death or skinlesions as the indicator. Values of toxin areexpressed in terms of indicator effects, forexample minimum lethal dose, minimumhaemolytic dose or, alternatively, in termsof unit equivalents of standard antitoxin.The Lf dose of a toxin is that dose of toxinwhich, when mixed with one unit of

Vaccine manual 49


antitoxin, flocculates in the least time, i.e.it is the first mixture to flocculate, whilethe L+ is the smallest dose of toxin which,mixed with one unit of antitoxin, kills 50percent of injected mice of a designatedweight within a designated time (Batty,1971).

The length of incubation is determinedby the particular toxin being produced. Cl.perfringens (welchii) beta toxin is destroyedby the proteolytic enzymes which theorganism produces if incubation continuesfor too long while others such as epsilontoxin are much more stable.

Bacterial toxins are invariably inactiv-ated with formaldehyde (Ramon, 1924).Although the principal action of form-aldehyde on a toxin is to remove toxicity,this is only achieved at the expense of someloss of immunogenicity. For this reason,toxoids are normally standardized in termsof a total combining power (TCP) test. TheTCP test involves a partial neutralizationof a fixed dose of antitoxin with a series ofvarying doses of the toxoid being tested.The unreacted antitoxin is then mixed witha fixed dose of toxin equivalent to half thedose of antitoxin used and the whole seriesof mixtures are injected into mice whichare observed for two days. The toxoidpresent in the mixture that kills half themice into which it is injected is equivalentto the fixed dose of toxin used (Batty, 1971).This is necessary because, after inactiv-ation, it is no longer possible to measurethe activity of the toxoids in terms ofindicator effects, although it is possible tocarry out an Lf test with certain toxoids.Where there are no appropriate facilitiesto carry out such tests, developingcountries may well have to rely on datafrom the original testing, for exampleminimum lethal doses.

HarvestingFor veterinary purposes, crude toxoids arenormally used. Crude toxoids are good

antigens and are extremely stable under avariety of environmental conditions. Thepreparation of such toxoids only involvesseparation from the bacterial cell mass bycentrifugation (Walker and Foster, 1981).

Blending

Bacterial toxoids are normally blended intomulticomponent vaccines. Very carefulformulation is necessary if the host is toproduce a balanced response against allcomponents and the vaccine is to pass thecriteria laid down by regulatoryauthorities. Toxoids are normally blended,based on TCP value or some equivalent,and the appropriate unit equivalents addedto the final vaccine. In practice, themanufacturer will take into account thetesting of the toxin before inactivation (Lf,L+, Lv, Lh, etc.) as well as the TCP and,where appropriate, the Lf of the toxoid.Bearing in mind the loss of immuno-genicity on inactivation if the toxoidingprocess is allowed to proceed too far, theratio of TCP to Lf declines. A fine balancehas to be struck between a high TCP-Lfratio, on the one hand, and minimumresidual toxicity and freedom from anytendency to reversal, on the other.

After blending and the addition of theadjuvant, the vaccines are ready for issuewhen certain statutory regulatory testshave been completed.

Use of vaccinesThe aim of the manufacturer is to provideprotection both to the adult animal and tothe offspring by the passive transfer ofantibody. In the case of the adult animal,providing it has been given primary andsecondary injections with appropriatebooster doses, it should remain protectedfor life. In an adequately primed animal,any subsequent infection will result in ananamnestic response which will affordimmediate protection (Chadnik, Watsonand Hepple, 1959). In the case of the

offspring, the length of protection will bedetermined by the persistence of antibodytitres which will, in turn, be determinedby the amount and titre of the colostralantibody ingested. As the decay ofpassively transferred antibodies is known(roughly 16 percent per week), it isrelatively easy to predict how longprotective levels are maintained. Protectiveantibody levels are known for severalantitoxins, for example 0.1 unit forCl. perfringens epsilon and tetanus and0.5 units for Cl. perfringens beta. Tables 5and 6 illustrate this by reference to twocommercial vaccines, A and B.

Table 5 shows the difference betweenthe performance of the two commercialvaccines, A and B, when the Cl. perfringensepsilon antitoxin titres to the first andsecond injections in previously unim-munized sheep are compared. It can beseen with vaccine A that after a singleinjection 90 percent of the animals gave ameasurable titre compared with 65.5

Note: GMT geometric mean titre.Source: Adapted from Kerry and Craig, 1979.

percent with vaccine B. The groupgeometric mean (GMT) titre achieved aftera single injection with vaccine A is abouteight times that of vaccine B. Thesedifferences in the titres are also seen afterthe second injection (Table 5), and withvaccine B two animals still have nomeasurable titre. The responses in the ewes(Table 6) are reflected in the antitoxin titresobtained in the lambs subsequent toparturition and, hence, the length ofprotection afforded to the lambs duringthe early weeks of life (Table 7).

It can thus be seen that, even with twovaccine preparations released on the basisof the Codex requirements, there aredifferences in potential protection of boththe adult and the newborn.

While toxoid vaccines are normallyeffective, there are examples of breakdownin the field, usually in cases where highlevels of challenge lead to large quantitiesof toxins which overwhelm the antitoxicimmunity. For this reason some manu-

TABLE 5

Distribution of Cl. perfringens epsilon antitoxin titres in the sera of 8-month sheep immunizedwith two commercial multicomponent clostridia! vaccines, A and B

Serumtitres(IU m1-1)

No. of sera (%)

Vaccine A Vaccine B

42 days affer first injection5 13 (43) 1(3.0)1-5 9 (30) 8 (28.0)0.1-1.0 5 (17) 10 (34.5)<0.1 3 (10) 10 (34.5)Total 30 29

Group GMT 1.84 0.21

14 days after second injection10 13 (43) 2 (6.5)1-10 16 (53) 16 (59.0)0.1-1.0 1 (4) 8 (28.0)<0.1 2 (6.5)Total 30 29

Vaccine manual51

Group GMT 9.04 1.39

TABLE 6

Distribution of C./. perfringens epsilon antitoxin titres in the sera of ewes immunized with twocommercial multicomponent clostridia! vaccines, A and B, following the second injection and

bled 2 to 5 days after lambing

TABLE 7

Distribution of Cl. perfringens epsilon antitoxin titres in the sera of 2- to 5-day lambs born toewes immunized with two commercial multicomponent clostridia! vaccines, A and B

Note: GMT = geometric mean titre.Source: Adapted from Kerry and Craig, 1979.

facturers do not remove cells from all thecomponents and leave them as anaculturesso as to obtain both antibacterial andantitoxic immunity.

A typical example of this was seen inDenmark in vaccination against the diseasepiglet enteritis, caused by Cl. perfringenstype C. This disease is a haemorrhagicnecrotizing enteritis of young pigletswhich, over a period of ten years, hasspread from small foci to cover large areasof Jutland and Zeeland. Initial studies onthis disease by Hogh (1974) showed thatthe causative organism was Cl. perfringens

type C and that the disease could becontrolled by administration of high-titrehorse antisera containing beta antitoxinand also by vaccination with vaccinescontaining beta toxoid. Using a vaccinewhich contained both cells and toxoid(Xento, Wellcome), successful protectionwas demonstrated (Hogh, 1974).

Some years later a combined Escherichiacoli/C1. perfringens type C toxoid vac-cine (Xentocol, Wellcome), specificallydesigned for the market to control both E.coli diarrhoea and Cl. perfringens type Centeritis, was introduced. In many herds

5 19 (61) 4 (13)2-5 6 (19) 8 (26)1.0-2.0 4 (13) 3 (10)0.5-1.0 1 (3) 4 (13)<0.5 1 (3) 12 (39)Total 31 31

Serumtitres(ILI ,n11)

No. of sera (%)

Vaccine A Vaccine B

10 13 (38) 3 (9)1-10 19 (56) 20 (61)0.1 -1 .0 2 (6) 7 (21)Total 34 33

Group GMT 6.72 1.66

Note: GMT geometric mean titre.Source: Adapted from Kerry and Craig, 1979.


Group GMT 7.48 1.25

Serum No. of sera (%)titres(/U m/-1) Vaccine A Vaccine B

the vaccine was used successfully tocontrol E. coli and piglet enteritis. However,in a number of problem herds the vaccinewas totally ineffective in controlling pigletenteritis.

Table 8 shows the results of a comparativetrial that was carried out using two productformulations and a control group. It can beseen that removal of the cells of Cl.perfrin gens resulted in a considerable lossof efficacy. The results indicated that,under conditions of high challenge, bothantibacterial and antitoxic immunity wererequired to contain infection (Hogh, 1988).Electron microscopy studies on the diseaseshowed a pattern of adhesion of Cl.perfrin gens type C to the jejunal villi inpiglets identical to that of E. coli infection inthe same species (Walker and Nagy, 1980).The initial phases of the disease werecharacterized by colonization of the villiowing to adhesion and multiplication of theorganisms. Unlike E. coli, the adhesivefactors are not characterized but clearly werepresent on the cells of Cl. perfringens in Xentovaccine and probably play the same role asthe "K" antigens of E. colt.

The lessons of these results are clear.Protection is multifactorial and, underextreme conditions of challenge, vaccinesbased on more simplified entities breakdown.

TABLE 8

Comparison of a perfringens beta toxoid vaccines and an identical vaccine containingCl. perfringens type C cells in pigs on s problem farm in northern Jutland

The combination of Cl. perfringens typeC and E. coli, i.e. a combination of anaerobicand aerobic vaccines, represents a furtherdevelopment in the marketplace. This is aresponse both to the needs of the marketand the desire to obtain a competitive edge.In this respect it is worth noting two otherdevelopments in this area.

A combination clostridial and Pasteurellavaccine (Heptavac-P) was introduced byHoechst (United Kingdom) to providesimultaneous protection against clostridialand respiratory diseases, particularly inyoung lambs. A somewhat differentapproach is the combination of clostridialvaccines with an anthelmintic, levamosole,introduced by the Tasman VeterinaryLaboratories (New Zealand). This combi-nation allows simultaneous protectionagainst clostridial infections and parasitesalthough, as dosing with anthelmintics isvery dependent on the weight of theanimal, dosage regimes can be difficult.

PROTECTION AGAINST ESTABLISHMENT OF

INFECTION

The importance of the cellular componentas a supplement in traditional toxoidvaccines has already been discussed.Cellular antigens are important in theestablishment of an organism in the host.Antibodies interfering with establishment

Vaccine Necrotic enteritis

Litters Piglets No. dead

Cl. perfringensbeta toxold

perfringensbeta toxoidplus cells

21

22

239

249

83

21

34.7

6.8

Control 20 217 93 42.9

Vaccine manual 53

will prevent infection. In general, cellularantigens can be divided into cell-wall andcapsular types. Until fairly recently, mostwhole-culture vaccines were relativelyundefined and consisted of suitablyinactivated cells standardized on the basisof cell numbers (approximately 1010organisms per ml), as measured by opacity.Such vaccines undoubtedly act bystimulating bactericidal antibodiesdirected against the cell and, as such, tendto be sero-specific. When the number ofimportant serotypes in a species is limited,the formulation of vaccines is relativelysimple. However, in species where a largenumber of serotypes are important, it isnecessary to blend a multiplicity ofserotypes into the final vaccine.

Toxicity resulting from endotoxincomponents of the cell wall of Gram-negative bacteria which can result inadverse reactions in the adult and abortionin pregnant animals becomes important asthe numbers of organisms are increased.In the case of E. coli vaccines forimmunization against diarrhoea inanimals, these consisted of a mixture ofthe predominant "0" serotypes prevalentin the disease. In the case of Pasteurellavaccines for the treatment of respiratorydiseases, these again consisted ofinactivated cells of the predominantserotypes of Pasteurella haemolytica andPasteurella multocida.

Similar inactivated whole-culturevaccines have also been used for the controlof leptospiral infections. Leptospira spp. arewell known as causal agents of disease inthe dog (Broughton and Scarnell, 1985).The importance of Leptospira spp. as acausative agent of disease, particularly incattle, has been recognized in recent years.The problem has become acute owing tothe transfer of leptospirosis from cattle tohumans during handling. The organism isshed in urine and is a potential source ofinfection. Vaccines have been developed

from inactivated whole cultures to preventshedding of the organism into the environ-ment (Broughton et al., 1984). Results showthat, in unvaccinated cattle, leptospiralorganisms are isolated frequently from theurine whereas, in vaccinated cattle, thereis virtually no excretion (Walker, 1987).

While the transfer of E. coli bactericidalantibodies via the colostrum and milk tothe suckling animal undoubtedly affordssome protection, the approach to vac-cination against E. coli was revolutionizedby the identification of the antigensresponsible for adhesion of the organismto the intestinal villi. The adhesion oforganisms to the intestinal villi is the firststage in E. coli infection and, if this stagecan be blocked by antibodies, infection isprevented. K88ab and K88ac wereidentified as important adhesins for thedisease in piglets (Smith and Lingood,1971) and early work in pigs showed thatexperimental vaccines prepared frompurified K88 antigens were effective inpreventing disease in piglets after oralchallenge (Jones and Rutter, 1972).

The first available commercial vaccineswere based on partially purified K88antigens of E. coli (Nagy et al., 1978). Duringgrowth, the 1(88 antigen is secreted intothe growth medium and can be selectivelyremoved from the supernatant byadsorption on to aluminium hydroxide,leaving behind endotoxin in thesupernatant.

Rather than depend on anti-adhesionalone, the early vaccines comprised cellsof E. coli supplemented with measuredquantities of K88ab and K88ac. In this way,both the anti-adhesive and bactericidaleffects of antibodies were present. Sows orgilts were given two injections of thevaccine, with the second dose timed to begiven two weeks prior to parturition, toachieve maximum levels of antibody inthe colostrum and milk for a passivetransfer to the piglets. To piglets of


vaccinated gilts following oral challenge,such vaccines gave significant protectionagainst mortality, diarrhoea and excretion(Nagy et al., 1978). While a conventionalE. coli K88-riegatiye vaccine affordedsignificant protection over the controls, thevaccine supplemented with measuredquantities of K88 antigen is even moresignificantly protective.

As time passed, further adhesins such asK99 and 987P were recognized as beingimportant and it became necessary toformulate multicomponent vaccines(Orskov et al., 1975; Moon et al., 1977; Smithand Huggins, 1978). The protection afford-ed by such vaccines is shown in Tables 9and 10. Experiments similar to those forthe original K88 vaccines were performedand it can be seen that significant protect-ion is afforded against mortality (Table 9)and diarrhoea (Table 10) with all fourserotypes (Nagy and Walker, 1983; Nagy,Mackenzie and Painter, 1985).

The identification of the surface antigensiesponsible for adhesion in E. coli repre-sents a move away from the traditionalvaccine to a more defined product. Otherenvelope antigens which may be importantin infection have been the subject ofincreasing interest.

Treponema hyodysenteriae is the aetio-logical agent of swine dysentery, a diseaseoccurring in post-weaning pigs, ususallyin fattening units where pigs are kept inlarge numbers (Alexander and Taylor,1969; Taylor and Alexander, 1971; Harriset al., 1972; Hughes, Olander and Williams,1975). After infection with the organism,swine develop a mucohaemorrhagicdiarrhoea characterized by extensivenecrosis of the mucosal epithelium of thecolon and caecum. This results in dehy-dration, emaciation and rapid weight lossfollowed, in severe cases, by death. Recentstudies have shown that, following asingle-dose parenteral immunization withan inactivated culture of T. hyodysenteriae

emulsified in oil, swine are protectedagainst experimental challenge.

A field trial indicated that this vaccineconfers significant benefits. In one fatteningunit, swine dysentery was reduced by 50percent and the severity of infection by 50percent, while vaccinated piglets showeda 10 percent increase in weight gain overcontrols in the same period after a singleinjection of the vaccine (Fernie, Ripley andWalker, unpublished results). Asmentioned previously, there is increasingresistance to oil-based vaccines because ofcarcass blemish and its effect on the exportof meat. For this reason, the vaccine hasnot been commercially developed.

To obtain further data on the protectiveantigens with a view to formulating amore non-reactive product, research hascentred on the envelope proteins ofT. hydosyenteriae. Using sodium dodecylsulphate-polyacrylamide gel electro-phoresis (SDS-PAGE) and immunoblot-ting, it has been shown that polypeptideswith a molecular weight of between 30 000and 36 000 were the predominant onesdetected by porcine immune serum andwere not present on non-pathogenic strainsof this organism. These unique antigensmay play a role in the virulence ofT. hyodysenteriae and form the basis ofa new effective non-oil-based vaccine(Chatfield et al., 1988).

A similar outer membrane protein (68kDa) was demonstrated to be an importantantigen in protection against atrophicrhinitis in pigs caused by Bordetellabronchiseptica. There is a strong correlationbetween antibody titres to this antigen andprotection in pigs. The antigen is presentin virulent strains of B. bronchiseptica butabsent in non-virulent strains. Monoclonalantibody produced against this antigenpassively protects mice from aerosolchallenge with the organism (Montaraz,Novotny and Ivanyi, 1985; Kobisch andNovotny, 1990).

Vaccine manual 55

TABLE 9

Mortality rates 24 hours aftel chcillenge in Iiiz Svoin voccinaied ancl non-vaccincieclilsafiel orogasivic c'cdlençc wiTh 11310 Escherichic colt plus iypes 24, hams (-Ater

In an attempt to improve the per-formance of Pasteurella vaccines, con-siderable work has been carried out usingthe specific pathogen-free (SPF) lambmodel developed at the Mordun Institute(Gilrnour et al., 1975). Such lambs are highlysusceptible to challenge with Pasteurellaspp. when given in conjunction with PI 3virus. Even in the absence of death, lunglesions develop after a few days and canbe scored at post mortem examination.

For a number of the A biotypes,

number of litters used.b Vaccine 5:1; partially purified K88ab, K88ac, K99, 987P of E. coli (100 units each).

TABLE 10

Diarrhoea rates 24 il01.11S 6101 ohallenga in litiem from vaccinated and non-voccincied ailts

n number of litters used.'Vaccine 5:1; partially purified K88ab, KS8ac, K99, 987P of E. coli (100 units each).

significant protection has been shown inthis model using heat-killed cells or sodiumsalicylate extracts of the organisms asvaccines. However, vaccines containingextracts or heat-killed cells of serotype A2,which is responsible for the majority ofoutbreaks in sheep, have been less effective(Gilmour et al., 1983; Donachie et al., 1986).

Studies have indicated that organismsgrown in vivo produce additional antigensto those grown in vitro. Pasteurellahaeinolytica A2 cells recovered from the

Vaccine E. coli challenge cultures

K88ab K88ac K99(diedlchallenged) (/o)

987P

None 19/36 (53) 38/58 (66) 10/34 (29) 33/41 (80)(n=5)° (n=10) (n=5) (n=7)

5:1° 0/39 (0) 3/36 (8) 2/37 (5) 3/35 (9)(n=5) (n=5) (n=5) (n=5)

Reduction inmortalityrates (%) 100 88 83 90

clflG crogasivic challenge vviih 1010 Escherichia c,oli plus iypes 24 hOUIS aiier

Vaccine E. coli challenge cultures

K88ab K88ac K99(diarrhoea/challenged) (%)

987P

None 46/54 (85) 47/58 (81) 34/54 (62.9) 50/52 (96)(n=7)a (n=10) (n=8) (n=9)

5:1° 1/51 (1,9) 2/53 (3.7) 20/57 (35.1) 22/48 (46)(n=7) (n=7) (n=7) (n=7)

Reduction indiarrhoea (%) 97.8 95,4 44.1 52,1


pleural fluid of lambs with pasteurellosiswere found to express previouslyunidentified proteins, and two of thesewith molecular weights of 70 and 100kilodaltons (kDa) could be induced in vitroby restricting iron availability in thegrowth medium.

It was subsequently shown thatsignificant protection against P. hacmolyticaA2 infection was conferred on SPF lambsby a vaccine containing extracts of P.haemolytica A2 cells grown under ironrestriction, compared with a vaccine madein exactly the same way from cells grownin an iron-replete medium. This wasconfirmed by comparing the antibodyresponses seen previously in SPF lambswhich were immune after recovery fromA2 infection. Using immunoblotting, itwas demonstrated that there were strongresponses to iron-limiting proteins of 70and 35 kDa and a weaker response at100 kDa (Gilmour et al., 1991).

In the case of Pasteurella multocida, analternative approach has been toconcentrate on the toxins produced by theorganism. Some success has been reportedin immunizing pigs against experimentallyinduced progressive atrophic rhinitiscaused by this organism, using a formalin-detoxified purified toxin. This approachmay well be valid for control of disease inlambs (Chantler, Rutter and Mackenzie,1986; Fogel, Neilsen and foral, 1989).

Surface appendages known as pili(fimbria) have also been shown to beimportant in the pathogenesis andprotection against ovine footrot. Ovinefootrot results in separation of the hornfrom the underlying soft tissues of the foot(Beveridge, 1941). Separation of the hornleads to abscess formation and anextremely painful condition in which thesheep have difficulty in walking andgrazing. Although the flora of the foot isvery complex, it has been shown that thekey triggering organism for infection is

Bacteroides nodosus (Egerton, Roberts andParsonson, 1969). A culture of scrapingsfrom the infected area reveals a verydistinctive rhizoidal colony (Thorley, 1976),the growth of which is highly pilated. Onsubculture in the laboratory a number ofnon-pilated smooth colonies are generated,and it has been demonstrated that the mosteffective vaccines are those made frompilated cultures (Egerton and Thorley,1981; Thorley and Egerton, 1981). Theorganism is divisible into a number ofserotypes based on pili agglutination and,in Australia, a number of serotypes havebeen identified as being important(Claxton, 1981). Multivalent vaccinesincorporating whole cultures of a mixtureof serotypes have been shown to be highlyeffective both in experiments involvingchallenge and in the field.

In summary, progress has been made inrecent years to identify those antigenswhich are important in the pathogenicprocess and to ensure that, underfermentation conditions, these antigens areproduced for vaccine production. Duringproduction, in addition to controlling pHand temperature, it is necessary to be ableto measure and control aeration of theculture. The degree of aeration isdependent on the growth rate of theorganism. The growth of E. coli strainsrequires high rates of oxygenation whereasstrains of Leptospira spp., which grow muchmore slowly, have relatively low oxygenrequirements. For several species, it isnecessary to supplement the air or oxygenwith carbon dioxide. Provision must bemade for the sterilization of the incomingair and outgoing effluent; the latterpresents a number of problems when highaeration rates are necessary.

The majority of these problems areminimized by growing the organism onsolid surfaces, although this is achievedonly at the expense of efficiency ascompared with fermenter-grown material.

Vaccine manual 57

Individual bottles of nutrient agar areinoculated with the appropriate organismand incubated for one to three days.Following incubation and growth, theorganisms can be removed from the surfaceof the culture and put into a resuspendingfluid by agitation. Where the antigens aresecreted into the supernatant fluid, therehas been some attempt to purify them butthe marketplace does not allow theextensive purification required for itshuman counterpart.

LIVE VACCINES

For many diseases, investigators believethat cell-mediated immunity (CMI) isnecessary to provide solid long-termprotection against infection. This has ledto the use of live attenuated strains asvaccines because, by comparison, killedvaccines generally do not elicit such strongCMI responses.

Live attenuated vaccines in theveterinary field include Salmonella dublin,S. cholera suis, Brucella abortus and anthraxspore vaccine. The attenuated S. dublin andS. cholera suis strains are avirulent roughvariants of the smooth virulent strain and,as such, are unable to synthesize a numberof polysaccharide side chains required forvirulence. However, in the present climateof legislation, a more rationally modifiedstrain harbouring well-defined, attenu-ating genetic lesions is to be preferred.Examples of the latter are strains har-bouring defined mutations in genes in theprechorismate pathway rendering themdeficient in the ability to synthesize certainessential aromatic compounds. In addition,it is preferable that strains used for livevaccines should harbour at least two stableattenuating mutations which map forseparate regions of the chromosome, thuslimiting the danger of reversion tovirulence. It has been shown that suchstrains, for example aroA, aroC or aroD, areadmirably suitable for this purpose

(Hoiseth and Stocker, 1981; Hone et al.,1991).

In the case of enteric pathogens, mucosalmmunity is regarded as an importantfactor in protection. In this respect, orallydelivered vaccines are superior and theuse of live oral vaccines to protect againstsalmonellosis in cattle provides a goodexample of modern developments.

Using a double aro mutant of Salmonellatyphimurium to immunize cattle orally, ithas been shown that there is significantprotection against a live challenge withthis organism. Of eight cattle immunizedorally with approximately 10" of themutant seven days after birth using anantacid mixture, seven were completelyprotected against a challenge of appro-ximately 108 organisms three weeks later.All four controls showed significant symp-toms and, along with the remaining vac-cinate, had to be killed (Jones et al., 1991).

More recently, this approach has beenextended to the use of S. typhimurium G30as a vector to express the fimbrial antigensK88 and K99 of enterotoxigenic E. coli(Stevenson and Manning, 1985; Attridge etal., 1988). Oral administration of 1011vaccine organisms to adult pigs elicitedsignificant serum antibody responses tothe respective fimbrial antigens (Moranaet al., 1994).

Live attenuated organisms have alsobeen used in the case of Pasteurella spp.Passaged in the laboratory, cultures ofPasteurella spp. disassociate, throwing offmutants that have lost their ability toproduce capsular polysaccharides. Suchmutants are non-pathogenic but highlyimmunogenic when injected subcuta-neously into the animal. Combined vac-cines comprising mixtures of the livingattenuated strains can be formulated andare highly effective against challenge inthe SPF lamb model (Walker, 1987). Again,these studies are limited by the use of anaturally occurring avirulent strain. Such


strains will not be favoured because ofpotential reversion and it will be necessaryto develop rationally attenuated strainsharbouring well-defined attenuatingcrenetic lesions such as the aro mutants forSalmonella spp.

Although reservations regarding the useof naturally occurring mutants or strainsattenuated in the laboratory as vaccineshave been stressed, the contribution thatsuch vaccines have made in controllingeconomically important diseases in thefield should not be forgotten. In the case ofanthrax, the live attenuated spore vaccinederived from an avirulent non-capsulatedvariant of Bacillus anthracis has had apronounced effect in limiting economiclosses resulting from anthrax and anindirect effect in reducing human incidence(Sterne, Nichol and Lambrechts, 1942;Turnbull, 1991). Similarly, the B. abortusS19 strain attenuated by temperature hasbeen highly effective in controllingbrucellosis.

BIBLIOGRAPHY

Alexander, T.J.L. & Taylor, D.J. 1969. Theclinical signs, diagnosis and control ofswine dysentery. Vet. Rec., 85: 59-63.

Attridge, S.R., Hackett, J., Morana, R. &Whyte, P. 1988. Towards a live oralvaccine against enterotoxigenic Escher-ichia coli of swine. Vaccine, 6: 387-389.

Batty, I. 1971. In J.R. Norris & D.W. Ribbons,eds. Toxin-antitoxin assay: methods inmicrobiology, Vol. 5A, p. 255-280. London,Academic.

Beveridge, W.I.B. 1941. Footrot in sheep: atransmissible disease due to infectionwith Fusiformis nodosus. Studies on itscause, epidemiology and control.Commonzyealth Aust. Council Sci. Ind. Res.Bull., 140: 1-56.

Brooks, M.E., Sterne, M. & Warrack, G.H.1957. A re-assessment of the criteria usedfor type differentiation of Clostridium

perfringens. J. Pathol. Bacteriol., 74: 185-195.Broughton, E.S. & Scarnell, J. 1985.

Prevention of renal carriage of lepto-spirosis in dogs by vaccination. Vet-. Rec.,117: 307-311.

Broughton, E.S., Marshall, R.B., Little,T.W.A., Hathaway, S.C., Mackintosh,C.G. & Hellstrom, J.S. 1984. Leptospirainterrogans serovar. Hardjo vaccines incattle: immunogenicity of vaccinesprepared from cultures grown in aprotein-free medium. Preventive Vet.Med., 2: 423-433.

Chadnik, K.S., Watson, A.R.A. & Hepple,J.R. 1959. Active immunisation of sheepand horses against tetanus withaluminium hydroxide adsorbed toxin.Vet. Rec., 43: 904-909.

Chantler, N., Rutter, J.M. & Mackenzie, A.1986. Partial purification of an oesteolytictoxin from Pasteurella multocida. J. Gen.Microbiol., 32: 1089-1097.

Chatfield, S.N., Fernie, D.S., Penn, C. &Dougan, G. 1988. Identification ofthe major antigens of Treponema hyo-dysenteriae and comparison with thoseof Treponema innocens. Infect. Immunol.,56: 1070-1075.

Claxton, P.D. 1981. Studies on Bacteroidesriodosus vaccines with particular referenceto factors associated with their efficacyin protecting against ovine virulentfootrot. University of Sydney, Australia.(Ph.D. thesis)

Donachie, W., Burrells, C., Sutherland, A.D.,Gilmour, J.S. & Gilmour, N.J.L. 1986.Immunity of specific pathogen-freelambs to challenge with an aerosol ofPasteurella haemolytic biotype A serotype2. Pulmonary antibody and cell responseto primary and secondary infections.Vet. Immunol. Immunopathol., 11: 265-279.

Egerton, J.R. & Thorley, C.M. 1981. Effectof alum-precipitated or oil adjuvantBacteroides nodosus vaccines on theresistance of sheep to experimentalfootrot. Res. Vet. Sci., 30: 28-31.

Vaccine manual 59

Egerton, J.R., Roberts, E.S. & Parsonson,I.M. 1969. The aetiology and patho-genesis of ovine footrot, I. A histologicalstudy of the bacterial invasion. J. Comp.Pathol., 79: 207-216.

Fernie, D.S., Ripley, P.H. & Walker, P.D.1983. Swine dysentery: Protection againstexperimental challenge following singledose, parenteral immunisation withinactivated Treponema hyodysenteriae. Res.Vet. Sci., 35: 217- 221.

Fogel, N.T., Neilsen, J.P. & Joral, S.E. 1989.Protection against atrophic rhinitis byvaccination with Pasteurella multocidatoxin purified by monoclonal antibodies.Vet. Rec., 125: 7-11.

Frerichs, G.N. & Gray, A.K. 1975. The relationbetween the rabbit potency test and theresponse of sheep to sheep clostridialvaccines. Res. Vet. Sci., 18: 70-75.

Freund, J., Casals-Ariet, J. & Genghof, D.S.1940. The synergistic effect of paraffin-oil combined with heat-killed tuberclebacilli. J. Immunol., 38: 67-79.

Gilmour, N.J.L., Thomson, D.A., Smith,W.D. & Angus, K.W. 1975. Experimentalinfection of lambs with an aerosol ofPasteurella haemolytica. Res. Vet. Sci., 18:340-341.

Gilmour, N.J.L., Martin, W.B., Sharp, J.M.,Thomson, D.A., Wells, P.W. &Donachie, W. 1983. Experimental im-munisation of lambs against pneumonicpasteurellosis. Res. Vet. Sci., 35: 80-86.

Gilmour, N.J.L., Donachie, W., Sutherland,A.D., Gilmour, J.S., Jones, G.E. & Quirie,M. 1991. Vaccine containing iron-regulated proteins of Pasteurellahaemolytica A2 enhances protectionagainst experimental pasteurellosis inlambs. Vaccine, 9: 137-140.

Harris, D.L., Glock, R.D., Christensen, C.R.& Kinyon, J.M. 1972. Swine dysentery:inoculation of pigs with Treponemahyodysenteriae (new species) and repro-duction of the disease. Vet. Med. SmallAnim. Clin., 67: 61-64.

HMSO. 1985. British Pharmacopoeia(Veterinary). London.

Hogh, P. 1974. Porcine infectious enteritiscaused by Clostridium perfringens. RoyalVeterinary and Agriculture University,Copenhagen. (dissertation)

Hogh, P. 1988. Vaccinationsforsog med fireforskellige tarmbrandsvacciner. DanskVet. Tidsskr., 71: 1-9.

Hoiseth, S.K. & Stocker, B.A.D. 1981.Aromatic-dependent Salmonella typhi-murium are non-virulent and effectiveas live vaccines. Nature, 291: 238.

Hone, D.M., Harris, A.M., Chatfield, S.,Dougan, G. & Levine, M.M. 1991.Construction of genetically defineddouble aro mutants of Salmonella typhi.Vaccine, 9: 810-816.

Hughes, R., Olander, H.J. & Williams, C.B.1975. Swine dysentery: pathogenicityof Treponema hyodysenteriae. Am. J. Vet.Res., 36: 971-977.

Jones, G.W. & Rutter, J.M. 1972. Role ofK88 antigen in the pathogenesis ofneonatal diarrhoea caused by Escherichiacoli in piglets. Infect. Immuna,6,918-927.

Jones, P.W., Dougan, G., Hayward, C.,Mackensie, N., Collins, P. & Chatfield,S.N. 1991. Oral vaccination of calvesagainst experimental salmonellosis usinga double aro mutant of Salmonellatyphimurium. Vaccine, 9: 29-36.

Kerry, J.B. & Craig, G.R. 1979. Field studiesin sheep with multicoMponent clostridialvaccines. Vet. Rec., 105: 551-554.

Kobisch, M. & Novotny, P. 1990.Identification of a 68-kilodalton outermembrane protein as the major protectiveantigen of Bordetella bronchiseptica byusing specific pathogen-free piglets.Infect. Immunol., 58: 352.

Montaraz, J.A., Novotny, P. & Ivanyi, J.1985. Identification of a 68 kilodaltonprotective protein antigen from Bordetellabronchiseptica. Infect. Immunol., 47:744-751.

Moon, H.W., Nagy, B., Isaacson, R.E. &


Orskov, I. 1977. Occurrence of 1(99antigen on Escherichia coli isolated frompigs and colonisation of pig ileum by1(99 and enterotoxigenic E. coli fromcalves and pigs. Infect. Immunol., 15:614-620.

Moore, W.B. 1968. Solidified media suitablefor the cultivation of Clostridium novyitype B. J. Gen. Microbiol., 53: 415-423.

Morana, R., Morana, J.K., Cosidine, J.A.,Hackett, L., van den Bosch Beyer, L. &Attridge, S.R. 1994. Construction of K88and K99 expressing clones of Salmonellatyphimurium G30: immunogenicityfollowing oral administration to pigs.Vaccine, 12: 513-517.

Mueller, J.H. & Miller, P.A. 1954. Variablefactors influencing the production oftetanus toxin. J. Bacteriol., 67: 271-277.

Nagy, L.K. Sr Penn, W. 1976. Protection ofcattle against experimental haemorrhagicsepticaemia by the capsular antigens ofPasteurella multocida types B and E. Res.Vet. Sci., 20: 249-253.

Nagy, L.K. & Walker, P.D. 1983. Multi-adhesin vaccine for the protection ofthe neonatal piglet against E. coliinfections. Proc. Int. Symp. EntericInfections in Man and Animals: Standardof Immunology, Dublin, Ireland. Dev.Biol. Stand., 53: 189-197.

Nagy, L.K., Mackenzie, T. & Painter, K.R.1985. Protection of the nursing pig againstexperimentally induced enteric coli-bacillosis by vaccination of the dam withfimbrial antigens of E. coli (K88, K99,987P). Vet. Rec., 117: 408-413.

Nagy, L.K., Walker, P.D., Bhogal, B.S. &Mackenzie, T. 1978. Evaluation ofEscherichia coli vaccines against ex-perimental enteric colibacillosis. Res. Vet.Sci., 24: 39-45.

Oakley, C.L. 1943. The toxins of Clostridiumwelchii. Bull. Hyg., 18: 781-806.

Oakley, C.L. & Warrack, G.H. 1953. Routinetyping of Cl. welchii. J. Hyg., 51: 102-108.

Orskov, I., Orskov, F., Smith, H.W. & Sojka,

W.J. 1975. The establishment of K99, athermolabile transmissible Escherichia coliK antigen previously called "KCO"possessed by calf and lamb entero-pathogenic strains. Acta Pathol. Microbiol.Scand., 83B: 31-36.

Ramon, G. 1924. Sur la toxine et surl'anatoxine diphtériques. Annu. Inst.Pasteur, 38: 1-10.

Smith, H.W. & Huggins, M.B. 1978. Theinfluence of plasmid-determined andother characteristics of entero-pathogenicEscherichia coli on their ability to pro-liferate on the alimentary tract of piglets,calves and lambs. J. Med. Microbiol., 11:471-492.

Smith, H.W. & Lingood, M.A. 1971.Observations on the pathogenic pro-perties of the K88 and ENT plasmids ofEscherichia coli with particular referenceto porcine diarrhoea. J. Med. Microbiol.,4: 467-485.

Sterne, M. & Batty, I. 1975. PathogenicClostridia. London, Butterworth.

Sterne, M., Nichol, J. & Lambrechts, M.C.1942. The effect of large-scale activeimmunisation against anthrax. J. S. Afr.Vet. Med. Assoc., 13: 53-63.

Sterne, M., Batty, I., Thomson, A. &Robertson, J.M. 1962. Immunisation ofsheep with multicomponent clostfidialvaccines. Vet. Rec., 74: 909-913.

Stevenson, G. & Manning, P.A. 1985.Galactose epimeraseless (Gal E) mutantG30 of Salmonella typhimurium is a goodlive oral vaccine carrier for fimbrialantigens. FEMS Microbiol. Lett., 28:317-321.

Taylor, D.J. & Alexander, I.J.L. 1971. Theproduction of dysentery in swine byfeeding cultures containing a spirochaete.Br. Vet. J., 127: 58-61.

Thomson, R.O. 1979. Secondary productsof metabolism. In A.H. Rose, ed. Bacterialtoxins economic microbiology, Vol. 3, p.435-466. London, Academic.

Thomson, R.O., Batty, I., Thomson, A.,

Vaccine manual61


Kerry, J.B., Epps, H.B.G. & Foster, W.H.1969. The immunogenicity of a multi-component clostridial oil emulsionvaccine in sheep. Vet. Rec., 85: 81.

Thorley, C.M. 1976. A simplified methodfor the isolation of Bacteroides nodosusfrom ovine footrot and studies on itscolonial morphology and serology. J.Appl. Bacteriol., 40: 301-309.

Thorley, C.M. & Egerton, J.R. 1981.Comparison of alum-absorbed oilemulsion vaccines containing eitherpilate or non-pilate Bacteroides nodosuscells in inducing and maintainingresistance of sheep to experimentalfootrot. Res. Vet. Sci., 30: 32-37.

Turnbull, P.C.B. 1991. Anthrax vaccines:past, present and future. Vaccine, 9:533-539.

Walker, P.D. 1987. Current trends andadvances in bacterial vaccines today. J.Appl. Bacteriol., 62: 1-23.

Walker, P.D. 1991. Editorial: Proceedingsof the International Conference onPrevention of Hib Meningitis in the 1990s.Vaccine, 9 (Suppl.) S.3.

Walker, P.D. 1992. Bacterial vaccines: oldand new, veterinary and medical. Vaccine,10: 977-986.

Walker, P.D. & Batty, I. 1985. The use andstandardisation of multicomponentveterinary vaccines in 1986. Proc. IABSCongr. Use and Standardisation ofCombined Vaccines, Amsterdam, theNetherlands. Dev. Biol. Stand., 65: 227-236.

Walker, P.D. & Foster, W.H. 1981. Bacterialvaccine production. In J.R. Norris & M.H.Richmond, eds. Essays in appliedmicrobiology. Chichester, UK, Wiley.

Walker, P.D. & Nagy, L.K. 1980. Adhesionof organisms to animal tissues. In R.C.W.Berkeley, J.M. Lynch, J. Melling, P.R.Rutter & B. Vincent, eds. Microbialadhesion to surfaces, p. 473-494. Chichester,UK, Ellis Horwood.

Walker, P.D., Harris, E. & Moore, W.B. 1971.

The isolation of Clostridia from animaltissues isolation of anaerobes. In D.A.Shapton & R.G. Board, eds. Soc. Appl.Bacteriol. Tech. Series No. 5, p. 25-38.London, Academic.

Compared with other sectors of micro-biology, it is unfortunate to have to admitthat the range of available vaccines formycoplasmal diseases does not provideentire satisfaction for a number of reasons,among which are their poor safety recordfor recipient animals and a weak, short-lived resistance to natural disease aftervaccination.

This disappointing situation can beexplained on the following grounds:

e Imprecise knowledge of the actualpathogenic processes in mycoplasmaldiseases.

Insufficient knowledge of the preciserole of the different antigens present inthe mycoplasmal organism in both thepathogenesis and immune mecha-nisms; for example, in the case ofMyco plasma mycoides subsp. mycoidesSC,' the function of galactan, thepolyosidic outer capsule, is unclear.

A lack of progress in developingsimple, cheap and reliable potency testsfor vaccines, taking account of thestrong host species specificity of themycoplasms.An almost total ignorance of the natureof the protective mechanisms inresistant animals, either after recoveryfrom the natural disease or aftervaccination. In no case is there a strictcorrelation between protection andthe presence of circulating antibodies.The situation in contagious bovine

In this document, Mycoplasma mycoides subsp.mycoides (SC, bovine biotype) will be referred to asMycoplasma mycoides (abbreviated as M. mycoides).

A. Provost

pleuropneumonia (CBPP) is such thatcomplement fixing, agglutinating,growth and / or metabolism inhibitingantibodies to M. mycoides are no proofof resistance to a challenge. Conversely,resistance can occur without them.The same events exist in enzooticpneumonia of pigs, owing toM. hyopneumoniae, and in M. galli-septicum infection of poultry. A possibleexception, found in humans, is thepresence of growth-inhibiting anti-bodies to mycoplasms in bronchialwashings. This is an avenue whichshould be pursued in animal myco-plasmosis. Cellular-type immunityproves equally ineffective for anycorrelation with protection.

0 Only partial protection against naturalor experimental challenge can beinduced partial being understood intwo ways: either only a reducedproportion of the total number ofanimals are protected, or the protectionleads to an abortive or less severedisease. This occurrence may happenfrequently following the use of in-activated vaccines but is also en-countered with live vaccines.

Unforeseeable reactions followingvaccination, either locally at the site ofinjection (called Willem's reaction inthe case of CBPP live vaccine) or by thepotentiation of other diseases after theuse of an inactivated vaccine (asobserved in human primary atypicalpneumonia and in enzootic pneumoniaof pigs).

In the field of animal diseases, there isonly one condition for which vaccination

Vaccine manual 63

ycoeis vac.., li

64 Mycoplasmal vaccines

is currently in use, namely CBPP. Experi-mental vaccines have been producedagainst swine pneumonia but otherstrategies are also used to control thedisease. An inactivated vaccine againstcontagious caprine pleuropneumonia(CCPP) has been studied in Kenya and isreported to have given valuable results inthe field; it is not known, however, whetherit is produced on a large scale. Livevaccines have been developed for avianmycoplasmosis and have been producedcommercially on a temporary basis but,owing to some disappointments (includinga lack of safety for non-vaccinated birds),their use tends to be ignored and control isconfined to a combination of means centredon the treatment of fertile eggs (dipping,heating) to evade vertical transmission ofthe agent.

CBPP VACCINE: GENERAL CONSIDERATIONS

Types of vaccineVaccination against CBPP has been im-plemented by the following methods:

Willem's method, using the pleuritic orpulmonary "lymph", or the secondarylymph resulting from a subcutaneousWillem's reaction induced artificially.The traditional African procedure(subcutaneous inoculation of amacerated lung lesion into the nasalmucosa) is an improvised and relatedvariation of the procedure rediscoveredin Europe in the 1850s.

° Vaccines prepared from broth cultures ofnaturally attenuated strains (e.g. theAustralian strain V, and the Sudanesestrain F), some of which are freeze-dried.

°Cultures of M. mycoides attenuated bysubculture, varying in number of serialpassages according to the virulence ofthe original strain, the culture medium,the sensitivity of local cattle and theinoculation site. These vaccines havebeen called incorrectly "Bennett

vaccines". At present, strain T1 is oneof the two available. There have beenrecent advances in freeze-drying,prolonging the storage life of vaccineand facilitating its distribution in thefield.

Egg culture vaccines, employedsuccessfully during the 1960s onaccount of their good immunogenicityand availability in a freeze-dried form.However, they have been heldresponsible for pulmonary compli-cations in certain circumstances andhave now been abandoned.

Inactivated vaccines of various types areof only poor immunogenicity or lead toan unacceptable reaction at the site ofinoculation.

Fundamental rulesVaccines against CBPP need to complywith three fundamental rules, derived fromexperience:

o Viability. Only vaccines prepared fromthe specific live mycoplasma possessadequate immunogenicity, engender-ing a resistance of the premunitiontype. The vaccine strain can be recov-ered from lymph nodes several monthsafter vaccination and antigens maypersist for up to 204 days.

o Virulence. Following the observationsof scientists working in 1906, whichwere confirmed in Kenya in 1921 and afew years later in the Sudan in 1929,

subcultures in broth lead to a certaindecrease in the pathogenicity of M.mycoides. For any given strain, thereare a certain number of in vitro sub-cultures which will attenuate theculture.

°Attenuation is characterized by theability of the particular strain toproduce at the site of injection only aminor reaction (localized oedema,which heals spontaneously) after sub-cutaneous or intradermal inoculation.

The actual technique of inoculationvaries from one laboratory to another.This attenuation is not fixed, for furthersubcultures lead to a progressivedecrease in virulence, with the disap-pearance of immunogenic properties.Therefore, it is necessary to reach adelicate compromise between retentionof immunogenicity and minimal localreactions acceptable to cattle owners.Subculturing should be avoided.The attenuation of cultures to a greateror lesser extent by in vitro subculturinghas provided the foundation for vac-cination policy during the past 80 years.It is only the approach to the problemthat has varied.Vaccination route. It is usual to followthe rule of the permissible or acceptablevaccination site. This requires denseconnective tissue of a low reactivity tolimit adverse local and generalizedreactions which may sometimes befatal. For this reason, it has long beenadvisable for the vaccine to be inoculat-ed into the skin or the subcutaneoustissue of the tip of the tail. Nevertheless,for practical purposes and to avoid lossof the tail after a severe reaction to thevaccine, the current method is toinoculate the vaccine subcutaneouslyon the flank posterior to the shoulderblade, necessarily using a vaccine ofacceptable or no residual virulence.

Occurrence of local postvaccinal reactionsAt the end of the last century it wasbelieved that, for a vaccination to beeffective, a local lesion was needed acircumscribed Willem's reaction. It wasthen thought inevitable that there wouldbe a certain number of vaccination com-plications, ranging from extensive post-vaccinal swelling to death, which couldaffect 2 to 3 percent of the vaccinatedanimals.

Towards 1930, scientists first started to

claim that there was no relationshipbetween the residual pathogenicity of avaccine strain of M. mycoides and itsimmunogenicity. The years that followedseemed to disprove this. Most researchersand field observers shared the view thatprotection was transient and weak in theabsence of a local postvaccinal reaction.

This fact was widely recognized at thetime, but was difficult to explain. Amongthe factors involved might have been poorcultures, faulty storage of vaccine andonly partial vaccination in a heavilycontaminated herd. The local reaction wasproof of the viability of the vaccine and, inits absence, there were doubts about thevaccine's efficacy. To this may be addedthe lack of uniformity in methods ofexperimental infection of cattle, leading todiffering evaluations of the resistanceconferred by one or other of the variousprocedures.

However, research conducted inAustralia between 1960 and 1970 clearlydemonstrated that there was no correlationbetween the intensity of the vaccinalreaction and the quality of protection.Excellent resistance to experimentalinfection developed without any localreaction provided that the minimumrequired number of live mycoplasmaswere inoculated (107 per dose for strain T1)

If untoward local reactions of theWillem's type occur (i.e. extensive oedemastarting 10 to 12 days after vaccination)antibiotic treatment (spiramycine, tylosin,tilmicosin) can be advocated.

General vaccination precautionsVaccination against CBPP is neverharmless. Apart from the local, and evengeneralized, reactions which may beproduced, vaccination may potentiatelatent infections or infestations in cattleexisting under the precarious physiologicalbalance prevalent in tropical conditions(e.g. trypanosomiasis and piroplasmosis).

Vaccine manual65

Factors to take into account:Breed (where the term "breed" is often

used incorrectly for species). In general,cattle breeds are more susceptible tovaccination accidents than zebu, at leastin Africa. Dairy breeds are usually moresusceptible than beef breeds butexceptions occur in both cases.

Group. Within the same species andsame breed, there are individualvariations in sensitivity to vaccination(and also to infection). Such variationsare a hindrance to vaccinationcampaigns because of possibleunpredictable reactions.Sex. Pregnant females in the terminalstage of gestation should not bevaccinated.

Age. The inoculation of calves underthree months is usually followed byinsignificant local reactions but suchanimals may develop arthritis,synovitis and heart valve lesions whichcan be fatal, at least with the mostvirulent strains. In practice, calvesshould not be vaccinated under six toeight months of age.

The level of protection inducedThe vaccination of cattle against CBPPusually elicits temporary serologicalconversion, detectable by various standardserological tests. It takes approximately sixto eight weeks to return to a serologicallynegative level.

The establishment of resistance is slow,at three to four weeks, and the level ofresistance conferred by a CBPP vaccine canbe demonstrated by testing a group ofanimals but not individuals. The effect ofvaccination is to raise and consolidate theaverage level of resistance in a herd ratherthan to provide individual protection.Clearly, it cannot protect 100 percent ofvaccinated animals. Among those vac-cinated, there is a variation in resistance toinfection, ranging from excellent immunity

(mycoplasmocidal antibodies) to simpleclinical resistance accompanied by en-capsulated lung lesions from which M.mycoides can be recovered.

With any CBPP vaccination, there is therisk of creating semi-resistant cattle whichcan become chronic carriers followingnatural infection, thus promoting thecontinuance of the disease. Hence theimportance of the rule of disease controlwhich states that CBPP vaccination mustbe extensive in an area and continuous in timeto be effective. By contrast, partial andunrenewed vaccination of a susceptibleherd may simply maintain the disease at alow level. Failure to observe this rule hasresulted in numerous disappointments inthe past.

In practice, there is considerable dis-crepancy between resistance as assessedin the laboratory and that which operatesin the field. A vaccine capable of protectingonly 70 percent of cattle experimentallymay still give excellent results in the field.

CBPP VACCINE: GENERAL MANUFACTURING

AND CONTROL PROCEDURES

Source materialsStrain of M. mycoides. The strains of M.mycoides used for vaccine production areidentified by documents which provideinformation on their origin and themanipulations they have undergone. Thestrain selected must have been shown tobe suitable for the region and the type ofbovines for which it is intended, whenadministered by the inoculation routeprescribed by the manufacturer. Beforeuse, the strain must first have been testedto show that the resulting vaccine is safeand confers a protection lasting at leastone year in cattle. One such vaccine withthese characteristics is the strain T1-44.

Strain T1 was isolated in 1961 at the EastAfrican Veterinary Research Organisation,from a clinical case of CBPP in what is nowthe United Republic of Tanzania. From the


outset, it appeared to be of only moderatevirulence. It has been cultured in embryon-ated hens' eggs, first on the chorio-allantoicmembrane for six passages and then in theyolk sac. Attenuated from the ninth eggpassage, it has been used mainly in EastAfrica for preparing freeze-dried eggculture vaccines with good immunogenicqualities.

Unfortunately, following the use of eggculture vaccines there were a number ofcases of postvaccinal lung lesions andnumerous local reactions, leading to thesuspicion that such vaccines may bedangerous. Their use was thereforesuspended in 1962. Since then, it has beenshown that the lung lesions may have beencaused by biologically active substancespresent in the egg material. The intrinsicqualities of strain T, were not doubted,however, so it was decided to prepareliquid vaccines from broth cultures of the44th egg passage of the strain. This materialis stored in freeze-dried form.

In parallel with these studies, thepossibility of combining the strain T, withcell-culture rinderpest vaccine to obtain acombined vaccine has also been investigat-ed. Mutants of the strain T1-44 that areresistant to streptomycin (present in theculture fluid of rinderpest virus) have beendeveloped at the Institut d'Elevage et deMédecine Vétérinaire des Pays Tropicaux(IEMVT) Laboratory at Maisons-Alfort inFrance and the Farcha Laboratory in Chad.The Maisons-Alfort mutant (T1-SR),obtained by three passages in a liquidmedium containing increasing concentra-tions of streptomycin, has been studiedunder experimental conditions at Dakar,Senegal, for its immunogenic properties.

Thus, two lines of T1 are now available:the original strain T1-44 and the strain T,-SR, which has undergone some additionalpassages in broth. Streptomycin appearsto act as a growth factor for the latterstrain, although there is no question of

streptomycin-dependence. The immuno-genic properties of both strains areidentical, but the strain T1-SR has producedfewer local postvaccinal reactions inhumpless cattle breeds.

Both strains may be obtained from theDirector of CIRAD-IEMVT, 10 rue PierreCurie, 94704 Maisons-Alfort, Cedex,France. This is the FAO reference centrefor mycoplasmas of ruminants.

Culture medium. The culture medium usedcontains serum or some other product ofanimal origin, which must be:

either heated at 56°C for at least 30minutes before being added to themedium, or treated by some otherprocedure that is equally effective indestroying contaminant microbialagents;

submitted to appropriate tests toshow that the material is free fromcontaminants.

In general, culture media contain:*a basic medium consisting of meatextract (e.g. infusion or papain digestof beef heart) and/ or peptone (trypose,tryptone);an extract or autolysate of brewer'syeast (commercially available orprepared on the spot), which providesgrowth factors (group B vitamins);

*glycerol, oleic acid, palmitic acid,glucose, sometimes a buffering system,a small amount of DNA and, if thevaccine is produced in a fermenter, alsoan antifoaming agent;

*10 percent blood serum, preferablyfrom adult horses, which must complywith the standard mentioned above.

If the strain T1-SR is used, streptomycinis added at 200 mg per litre of medium.The final pH is adjusted to 8.1. A satisfact-ory formulation is Medium F-66 of theFarcha Laboratory.

After clarification, it is advisable tosterilize the medium by membrane filtra-

Vaccine manual 67


tion rather than by autoclaving. It shouldbe used as soon as possible after pre-incubation for 24 hours at 37°C to verifythe absence of bacterial contamination.

Seed lot system. Vaccine production isfounded on a seed lot system. A seed lotmay be part of a culture which has servedfor vaccine production, has been shown tobe safe for cattle when administered undernatural conditions and is capable ofconferring an immunity that lasts for atleast a year. A seed lot obtained from sucha culture must not be submitted to morethan three additional passages.

Seed lots' are freeze-dried and stored ata temperature not higher than -20°C.

Tests on seed lots. Each seed lot must befree from extraneous agents and mustsatisfy the standards of identity and safetyin laboratory animals. In addition, avaccine prepared from the seed lot mustbe tested for safety and efficacy in cattle.

Cattle used for this testing must berepresentative of the cattle to be vaccinated;they must come from an area free fromCBPP and must be at least 24 months oldand in good health. The cattle should beidentified in an indelible manner and keptin quarantine under veterinary supervisionfor at least four months. If any clinical signsof disease are present, the animal con-cerned should not be used until the causeof the abnormality has been determinedand recognized as having no effect on thecorrect conduct of the tests. Serum shouldbe taken from each animal at the start andend of the quarantine period to be testedfor CBPP antibodies. Only serologicallynegative cattle should be used.

Safety testing. A quantity of vaccine

A seed lot prepared jointly by the IEMVT and thePan African Veterinary Vaccine Centre (PANVAC)is now stored at both centres for free distribution tovaccine manufacturers.

containing at least ten times the numberof mycoplasmas present in a recom-mended dose should be inoculatedsubcutaneously on the flank posteriorto the scapula in each of at least tencattle. The seed batch is satisfactory ifthe cattle fail to develop any un-expected clinical reaction during aperiod of observation lasting for at leastfour weeks and if no lung lesion isfound on post mortem examination atthe end of this period.

Efficacy testing. A quantity of vaccinecontaining the same number of bacteriaas the recommended dose should beinoculated subcutaneously on the flankposterior to the scapula in each of atleast ten cattle. A similar number ofcattle should be kept as a control group.After a minimum of two months haveelapsed, all cattle should be submittedto challenge infection by contact withinfected cattle (i.e. cattle infected bythe introduction of lung lesion homo-genate through an endobronchial tube).At least one infected animal should beused for each group of three cattleunder test. The duration of the test isthree months and any donor animalinfected by intubation that dies duringthis period must be replaced. The seedbatch passes the test if the animalsinoculated with vaccine prepared fromthe batch do not develop a clinical or aserological reaction (i.e. an anamnesticresponse), post challenge, if they arefree from lesions of pleuropneumoniaat slaughter and if the tracheobronchiallymph nodes are free from M. mycoides.At least 80 percent of the unvaccinatedcontrol animals should have typicallesions at post mortem examination.

Production precautionsInoculation and culture. The sample of thevaccine strain used for inoculating theculture vessels or the fermenter should not

have been submitted to more than twopassages in broth from the seed lot.

It is best to inoculate the productionmedium, pre-incubated at 37°C, with aculture of the vaccine strain in the log-arithmic phase of growth (36 to 40 hoursculture from agitated medium), using onepart of culture to ten parts of the culturemedium.

After a stationary phase of incubation at37°C, the medium is aerated (by a magneticstirrer if flasks are used, or by injection ofair if a fermenter is used).

The culture is stopped at peak growth,usually at 65 to 77 hours after inoculation.This time has to be determined preciselyby each producing laboratory, with regardto local conditions, by establishing agrowth curve for a culture of the vaccinestrain in medium and with vessels identicalto those used for production. This recom-mendation is important because con-tinuation of culture after the growth peakleads to a considerable fall in pH and adiminished number of live mycoplasma.

Harvesting and freeze-drying. After aculture has been stopped, it is essential toharvest immediately, add stabilizer (for thepreparation of the final product in bulk),fill the final containers and freeze-dry theproduct. Storage of the harvest or the finalproduct in bulk, even in a cold chamber,has an adverse effect on the number of livemycoplasma in the vaccine. Theseoperations are depicted in Figure 1.

The bulk suspension is prepared from asingle harvest. Once transferred to the tankfrom which the final containers are filled,it becomes the final product in bulk. Thisfinal product is transferred quickly to thefinal containers and prepared for freeze-drying. At this point it becomes a batch ofvaccine.

Distribution into the final containers andthe subsequent freeze-drying must be doneas soon as possible after the final bulk

product has been prepared. After freeze-drying of the vaccine, the containers arehermetically sealed under vacuum orunder dry nitrogen free from oxygen. Incase of failure, all containers sealed undervacuum should be tested for airtightnessand any defective container rejected.

If part of a vaccine batch is to be freeze-dried at some other time, it is best to freezethe vaccine within the final containers.After freeze-drying, it should be submittedto the control tests described below.

Freeze-drying procedure. A freeze-dryingprocess of long duration is used where thevaccine is not exposed to temperaturesabove 0°C while it still contains more than3 percent residual moisture. This can onlybe done in freeze-driers of the shelf type,not in the centrifuge type. Under theseconditions and with a satisfactorystabilizer, the loss of viable units shouldnot exceed 1 log10. In other words, with aharvest containing 1010 viable units per ml,concentrated freeze-dried vaccines areobtained, and their storage is considerablysimplified. Operating with modern freeze-driers, less than 1.5 percent residualmoisture can in fact be attained.

Control tests

Samples are taken from each vaccine batchor each distribution batch for the testsgiven below. The tests are performed onthe vaccine reconstituted to a volumewhich will depend on the initialconcentration of the final product in bulk.After reconstitution, samples should bestored at 4°C and tested within an hour.

Identity tests. A test of identity has to beperformed on at least one labelledcontainer of each distribution batch byusing a technique suitable for identifyingthe presence of M. mycoides.

A simple and satisfactory test is growthinhibition on agar gel, using filter paper

Vaccine manual 69

70 Mycoplasnial vaccines

-40°

Day

0 1 2 3 4 5

(-1 »

Magneticstirring

5litres)

72 hrs for Ti65 hrs for T1-SR

Stirring

6 7

Harvest

Stirring

Preparationof vaccine

(monovalent or combined)

Freeze-drying

FIGURE 1

Procedure for producing CBPP vaccine from strains T, and T1-SR

discs impregnated with a serum containinga high titre of antibodies to M. mycoidessubsp. mycoides (SC bovine biotype).Another procedure is to place antiserumin wells punched into the agar culture. Azone of inhibition measuring at least 2 mmwill be present after 48 to 72 hours ofincubation. This test has the advantage ofidentifying other mycoplasmas (whichgrow within the inhibition zone) and/ orbacteria present as contaminants. Morerefined tests (Western Blot, dot immuno-binding) can also be considered.

Sterility tests. Freedom of the vaccine frombacteria and fungi is assessed by tests madeaccording to the recommendations of theWorld Health Organization's Standards forBiological Substances No. 6 (generalstandards concerning the sterility ofbiological substances).

Determination of the myco plasma content.An organism count should be made oneach batch of vaccine (or each distributionbatch) to determine the number of viableM. mycoides per dose for cattle. This assayis done by the following procedures:

An initial titration is performed on amixture of at least five bottles selectedat random. This mixture can also servefor testing safety and efficacy in cattleif necessary.

e Second and third titrations are eachperformed on the mixed contents ofthree bottles.

e The titre of the vaccine is the geometricmean of the three titres thus obtained.

Each titration uses a system of vaccinedilutions and the appropriate number oftubes of growth medium sufficient toestimate the 50 percent end point by astandard statistical procedure. A series oftenfold dilutions, with ten tubes of mediumfor each dilution, or any other dilutionsystem which provides equal precision, isused. The tubes are incubated for at least

ten days at 37°C and are shaken lightlyeach day to aerate the medium. Theminimum titre per vaccine dose must be atleast 10 viable mycoplasmas.

It is recommended that productionlaboratories take into account possibleproblems arising from local transportconditions and therefore supply vaccineswith titres of at least 108 mycoplasmas perdose.

Safety tests. These tests are performed oneach batch of vaccine by inoculating guineapigs and mice. The procedure generallyused consists of injecting 0.5 ml ofreconstituted and undiluted vaccineintraperitoneally into each of two guineapigs, the same dose intramuscularly intotwo more guinea pigs and 0.1 ml intra-peritoneally into six mice. The animals arekept under observation for three weeksand then killed for examination. The batchof vaccine is considered satisfactory if theanimals remain healthy and if they haveno pathological lesion post mortem.

Tests for safety and efficacy in cattle.These tests are done with a mixture ofvaccine from at least five bottles selectedat random and as stipulated in the sectionSeed lot system, p. 68.

A safety test is obligatory for eachvaccine batch. Because of the complexityand cost of an efficacy test, this may beperformed occasionally (at least once ayear) for a given batch of tested seedculture or for a vaccine derived fromculture which has undergone no more thanthree passages from the seed culture,provided that all the arrangements forbiological security of the installation andmanipulations have been observed.

Stability test. It is desirable to conduct atest of stability on samples of vaccinebatches to obtain experimental data topredict the effect of storage temperature

Vaccine manual 71

and the expiry date. This data should begiven on the label and in the accompanyinginformation leaflet.

In the absence of data on long-termstability, it is recommended thataccelerated degradation tests should beconducted on the vaccine to arrive atrecommendations for storage conditionsand expiry date. The information providedby such tests must be confirmed as soon aspossible by observations on vaccine storedunder the recommended conditions.Storage under these conditions mustguarantee that the titre of mycoplasmas inthe vaccine up to the expiry date is not lessthan 107 viable mycoplasmas per vaccinedose.

Stability tests may also be made toprovide a foundation for the informationaccompanying each container. These testsestimate the fall in titre of mycoplasmasonce the product has been reconstitutedfor use and after storage of the intactproduct under the temperature conditionswhich may be encountered in the field.Information should also be provided forvaccine kept in ice boxes (00 to 10°C) andvaccine kept at different ambienttemperatures between 20° and 30°C (oreven 40°C).

Storage conditionsWhile awaiting dispatch from the pro-duction establishment or storage depot, thefinal containers of vaccine should prefer-ably be stored in the dark and at atemperature of -20°C or lower withinrooms provided with continuoustemperature-recording equipment.

During storage and transport of thevaccine, it is strongly recommended thattemperature indicators be used (heat-sensitive strips or paint) to check quicklythe temperature at which the vaccine hasbeen exposed.

The expiry date should not exceed twoyears from the last determination of the

mycoplasma count, provided that thevaccine has been stored continuously ata temperature of -20°C until delivery.Producers with large stocks of vaccineshould retest the titre before extending theauthorized storage period. On the otherhand, no vaccine should be used later thansix months after its supply by the pro-duction establishment or storage depot,and then only if the containers have beenstored at temperatures of less than +10°C.

The half-life of vaccine freeze-dried to aresidual moisture content of 1.5 percent isas follows:

4 days at 45°C;*1.6-2.8 weeks at 37°C;2.8-4.6 weeks at 28°C;

030-40 weeks at 4°C.These figures mean that, when tested

during production, a vaccine that containsa concentration of 108 viable units per dose(which is on the low side and encounteredin batches of only moderate quality) canbe kept for 2 to 4 months at 28°C, 1.5 to 3months at 37°C and 12 to 14 days at 45°C.

The storage life of reconstituted ready-to-use vaccine must not exceed the timeduring which the mycoplasmal count isequal to or greater than 107 viableorganisms per vaccine dose at the specifiedtemperatures, this having been verifiedexperimentally.

It has been shown experimentally thatreconstitution of freeze-dried vaccine inthe field to give the number of dosesspecified by the producer is best donewith a molar solution of magnesiumsulphate. Under these conditions there isno fall in mycoplasma titre for four hoursat 37°C. In no case should the vaccine bereconstituted with distilled water,tap water or surface water, even afterboiling.

CONCLUSIONS

As stated in the introduction, any im-provement in mycoplasmal vaccines is


hampered by little or no information onthe function of the essential immunogensof the organism, the lack of a preciseunderstanding of the pathogenesis ofmycoplasmal diseases and the insufficientknowledge of factors governing resistanceto challenge or natural disease. This meansthat subunit vaccines, and even possiblyrecombinant vaccines, are still far fromsuccessful development.

Some simpler avenues appear worthexploring. New adjuvants, namely im-munostimulating complexes (ISCOMs),could be studied to enhance the im-munogenicity of inactivated vaccines; suchsteps are under way for CBPP. Anotherapproach would be to explore the value oflocal immunity. In the case of CBPP, "local"means the application of the immunizingagent to the nasal mucosa by means of avaccine spray. Preliminary tests havealready been shown to be successful insome cases of enzootic pneumonia, CCPPand M. pneumoniae in hamsters.

Similar experiments conducted 20 yearsago in Nigeria and Chad proved to be verysuccessful with CBPP. Such an approachwould now seem to be worth pursuingfurther.

BIBLIOGRAPHY

FAO. 1971. Contagious bovine pleuropneumoniaa review. FAO Agricultural Studies

No. 86. Rome.Gilbert, F.R. & Windsor, R.S. 1971. The

immunising dose of T, strain Myco plasmamycoides against contagious bovinepleuropneumonia. Trop. Anim. Health.,3: 71-76.

Hudson, J.R. & Turner, A.H. 1963.Contagious bovine pleuropneumonia:a comparison of the efficacy of two typesof vaccine. Aust. Vet. J., 39: 373-385.

Litamoi, J., Palya, V.J., Sylla, D. &Rweyemamu, M.M. 1994. Quality controltesting of contagious bovine pleuro-

pneumonia (CBPP) live attenuated vaccinestandard operating procedures. Addis

Ababa, PANVAC.Office International des Epizooties. 1992.

Manual of standards for diagnostic testsand vaccines, 2nd ed. Paris.

Provost, A., Perreau, P., Breard, A., Le Goff,C., Martel, J.L. & Cottew, G.S. 1987.Contagious bovine pleuropneumoniaa review. Rev, sci. tech. Off int. Epiz.,6(3): 625-679.

Vaccine manual 73

Notaz a0 and ricke

THE NEED FOR VACCINES

Tick-borne protozoal and rickettsialdiseases affect most domestic animalspecies and have been a major constraintto the development of livestock industriesin developing countries in the tropics andsubtropics. Although tick-borne diseases(TBDs) are important in all domesticanimals, this review will be concernedmainly with vaccines against the eco-nomically important tick-transmitteddiseases of cattle caused by the protozoalparasites Babesia bovis and Babesia bigemina(babesiosis, red water or tick fever),Theileria parva (East Coast fever) andTheileria annulata (tropical theileriosis), aswell as the rickettsial organisms Anaplasmamarginale (anaplasmosis) and Cowdriaruminantium (cowdriosis, heartwater).Vaccines against coccidiosis have beenreviewed by Danforth and Augustine(1989).

Babesiosis caused by B. bovis and B.bigemina is present in many countriesbetween 400 N and 32.0S (McCosker, 1981).The most important vectors are ticks of thegenus Boophilus (Friedhoff, 1988).

East Coast fever is a disease limited tocountries in eastern, central and southernAfrica where the principal vector, Rhipi-cephalus appendiculatus, is present (Norval,Perry and Young, 1992). In these countries,it is of major importance because of thehigh morbidity and mortality it causes.Tropical theileriosis is transmitted by anumber of species of the genus Hyalommawhich have a wide distribution along theMediterranean littoral, in the Near Eastand throughout Asia (Pipano, 1989).

Anaplasmosis has the widest distri-

A.J. de Vos, E. Pipano, F. Musisi and W.K. Jorgensen

bution of all economically important TBDsand is transmitted biologically by at least20 tick species. Mechanical transmissionby biting flies and unsanitary veterinarypractices are also important in certain areas(Palmer, 1989).

Heartwater is important in cattle and isalso a major cause of losses in sheep andgoats. This disease is transmitted by ticksof the genus Amblyomma and occurs mainlyin sub-Saharan Africa, although it hasalso been introduced to islands in theAtlantic and Indian Oceans as well as theCaribbean. It is regarded as the mostimportant TBD in South Africa and, nextto East Coast fever, the most importantone in Africa (Uilenberg, 1983;Bezuidenhout, 1989). The presence ofheartwater and its vector in the Caribbeanposes a threat to livestock production inthe tropical and subtropical parts of theAmerican mainland (Barre et al., 1987).

Up to 500 million cattle the world overare exposed to one or more of the TBDsbut this figure is not a true reflection of thenumber at risk to disease. Indigenousbreeds of cattle often have a certain degreeof natural resistance to these diseases andthe consequences of infection are not asserious as when exotic Bos taurus breedsare involved. In addition, a state of enzooticstability frequently develops whereby localcattle become naturally infected at an earlyage when there is some passively acquiredor innate immunity. These cattle areresistant to subsequent challenge (de Vos,1992).

Cattle, particularly of exotic breeds, areat risk to TBDs, especially under thefollowing circumstances.

Vaccine manual75

The importation of susceptible cattleinto endemic regions. Mortality ratesunder these conditions can exceed 50percent in regions endemic forbabesiosis (McCosker, 1981) and thesame may apply in areas infected withEast Coast fever and heartwater.

The spread of ticks and TBDs intopreviously uninfected areas. Anexample of this was the loss of overone million cattle owing to TBDsfollowing the disruption of tick controlmeasures in Zimbabwe (Lawrence,Foggin and Norval, 1980).

The introduction of TBDs into adisease-free vector population ashappened when East Coast fever wasintroduced into southern Africa in 1896.This introduction resulted in the lossof an estimated 1.4 million cattle andtook 50 years to bring under control(Lawrence, 1992).

Increases in tick transmission rates inendemic areas caused by, amongothers, ecological factors. Mostoutbreaks of babesiosis in Australiaoccur in cattle bred in the endemic area(Callow and Dalgliesh, 1980).

Immunization and vector control are twooptions which need to be considered in allthese situations. Provided there aresuitable facilities and assuming com-mercial vaccines are not available, the localproduction of vaccine may be the onlysustainable means of control in the countryconcerned.

HISTORY OF VACCINE PRODUCTION AND

FUTURE PROSPECTS

B. bovis and B. bigeminaMethods used to immunize cattle againstbabesiosis have been described orreviewed (Callow and Dalgliesh, 1980;Pipano, Frank and Shkap, 1991; de Vosand Jorgensen in FAO, in press). In manycountries, early attempts to vaccinate cattleagainst B. bovis and B. bigemina infections

involved the use of blood from naturallyor artificially infected carriers. This carrier-donor method, also known as premunition,has several major limitations, includingunreliable potency, unpredictable reactionsand the risk of contamination (Callow,1984; de Vos and Jorgensen, 1992). Duringthe past 30 years, more sophisticatedtechniques were developed to produce livevaccines from splenectomized donors,which had predicably low levels ofvirulence (Callow and Dalgliesh, 1980).The inherent disadvantages of vaccinecontaining parasites derived from theblood of animals are well documented,including the risk of reactions, conta-mination, sensitization against bloodgroups and the need for cold chaintransportation (Callow and Dalgliesh, 1980;Wright, 1991). In spite of these disad-vantages, virtually all operational centresproducing Babesia vaccines are at presentusing variants of this technique to producechilled or frozen live vaccines. In vitroculture methods reviewed by Pudney(1992) have been used to produce B.bigemina parasites for vaccine (Jorgensenet al., 1992) and are also suitable forproducing B. bovis parasites. However,these culture techniques are still at thedevelopmental stage and not widely usedfor the production of vaccine.

Other products used in attempts toimmunize cattle against babesiosis includeantigens extracted from parasites orparasitized blood. Cell culture-derivedexoantigens of B. bovis and B. bigemina havebeen extensively studied and proposed foruse as vaccine in developing countries(Montenegro-James, Kakoma and Ristic,1989). Unfortunately, the level andduration of protection conferred by theseantigens against heterologous challengeare less than those of live vaccines (Timmset al., 1983).

Progress to overcome the limitations ofvaccines based on blood or blood extracts

76 Protozoal and rickettsial vaccines

through the development of synthetic orrecombinant Babesia vaccines, werereviewed by Wright et al. (1992) andDalgliesh (1993). Preliminary resultssuggest that vaccines based on singleantigens do not confer the desired level orduration of protection. It is likely that avaccine containing several recombinantantigens will be needed to induce adequateprotection (Pipano, Frank and Shkap,1991). Even a multicomponent recom-binant vaccine may not provide long-termprotection against field strains of Babesiawhich have been shown to be capable ofconsiderable genetic variation (Dalrympleet al., 1992; de Vos and Jorgensen, 1992).As yet, no recombinant vaccine for bovinebabesiosis has been registered for use inany country and it seems unlikely that sucha vaccine will be available in the nearfuture. When one does become available,it will almost certainly be marketedinternationally and in competition withvaccines produced locally.

T. parva and T. annulata

As reviewed by Purnell (1977), Irvin andMorrison (1989), Pipano (1989) andLawrence (1992), a variety of products andprocedures have been or are being used toimmunize cattle against East Coast feverand tropical theileriosis. Early attemptsinvolved the use of infected blood andtissues. In the case of East Coast fever, theresults were very variable but the blood ofdonor calves infected with low-virulencestocks of T. annulata was used for manyyears as vaccine (Pipano, 1989). Subse-quently, a vaccination technique wasdeveloped which involved infecting cattlewith sporozoites from ticks and thenmitigating the clinical responses bychemotherapy (FAO, 1984; Irvin andMorrison, 1989; Dolan and McKeever,1993). This "infection and treatment"method using T. parva sporozoites is usedin some countries in East and central

Africa, mainly on an experimental or fieldtrial basis. A ground-up suspension ofinfected adult R. appendiculatus (ground-up ticks supernatant [GUTS]) is used toinitiate the infection which is then control-led by treatment with oxytetracycline.Depending on the T. parva stocks used,sporozoite vaccine has proved effectiveunder laboratory conditions and in fieldtrials, providing long-lasting immunityagainst homologous and heterologouschallenge (Burridge et al., 1975a and 1975b).However, antigenic heterogeneity ofT.parva isolates has been a major obstacleto the widespread application of thismethod of immunization (Irvin andMorrison, 1989). The use of a mixture ofstocks is necessary to provide wide-spectrum cover against field challenge,especially where T. parva lawrencei isinvolved. In general, good protection isafforded by a vaccine made up of T. parva(Muguga), T. parva (Kiambu 5) and buffalo-derived T. parva (Serengeti transformed).However, because animals immunized inthis way may become carriers to theinfection and thus may transmit theparasite to the resident population, thereis an understandable reluctance toimplement immunization programmeswith stocks exotic to the region or country(Irvin and Morrison, 1989).

Despite this drawback as well as the lackof attenuated, immunogenic stocks andhigh cost, the "infection and treatment"method is still the principal means ofimmunizing cattle against T. parva. T. parva(Boleni) has been used in extensive fieldtrials in Zimbabwe. It is of low patho-genicity and treatment of reactions is rarelynecessary. An application has now beenmade to have vaccine containing thisstock registered for more general use inZimbabwe (Pegram, personal communi-cation, 1994). There is evidence that T. parva(Boleni) will also protect cattle againstsome highly pathogenic isolates but this

Vaccine manual 77

feature has not been exploited in practicaldisease control strategies in other parts ofAfrica. The infection and treatment methodhas also been used to prepare a T. annulatasporozoite vaccine (Pipano, 1989) but ithas not been evaluated under field condi-tions. Immunological differences betweenfield isolates of T. annulata do not appearto be as important as in the case of T. parva.

In vitro techniques for the culture ofschizonts of Theileria spp. were reviewedby Brown (1979; 1981) and Pipano (1989).T. annulata schizonts can readily bepropagated in culture and the patho-genicity of the parasites decreases duringprolonged maintenance in culture. Thisprovided the basis for developing a safe,culture-derived vaccine for the control oftropical theileriosis (Pipano, 1989; Singh,1990). This vaccine is currently in use or indifferent stages of development in severalcountries in Asia and around the Medi-terranean (Dolan and McKeever, 1993).Unfortunately, attempts to develop asimilar vaccine for T. parva gave disap-pointing results. Histocompatability is seenas one of the main obstacles to thesuccessful transfer of T. parva schizontsfrom infected cultures to recipient hosts.

The hazards involved in using thesporozoite vaccine and the stock-specificnature of the immunity conferred have ledto intensive efforts to identify parasiteantigens which may be targets for pro-tective immune responses, particularlyagainst T. parva infection (Dolan andMcKeever, 1993). Of sporozoite antigensidentified as candidates for inclusion in avaccine, a gene that codes for a 67kilodalton (kDa) major surface protein hasbeen isolated, sequenced and expressed. Itprovides partial protection in cattle againsthomologous challenge (Musoke et al.,1992). Two sporozoite antigens ofT. annulata defined by different mono-clonal antibodies have also been studiedas vaccine candidates. These antigens have

been expressed in Escherichia coli andstimulate the production of neutralizingantibody in cattle (Hall and Baylis, 1993).

Anaplasma marginaleA variety of procedures have been used toimmunize cattle against anaplasmosis(McHardy, 1984; Palmer, 1989). Theseprocedures involve the use of live orga-nisms such as attenuated A. marginale orA. centrale and killed organisms or extractsof infected blood.

A stock of A. inarginale has been at-tenuated by irradiation and passage insheep and deer (Palmer, 1989). Vaccinecontaining this stock was marketed asAnavac and used to immunize cattle inLatin America and the United States(Correr, Johnson and Wagner, 1985). Itprovided substantial protection againstchallenge with most heterologous isolates.In general, use of the vaccine had no orvery little clinical effect although severemorbidity and mortality have been re-ported, especially when older bulls andcows were immunized (Palmer, 1989). Useof attenuated A. marginale is the mosteffective means of immunization againstanaplasmosis but this vaccine is apparentlyno longer available. Attempts to attenuatea stock of A. marginale by passage in sheepin Australia were unsuccessful (Jorgensenet al., 1993).

Infection with A. centrale, an organismoriginally isolated in South Africa(Potgieter, 1979), provides partial cross-immunity against A. marginale challenge.It is usually mildly pathogenic, particularlyin young animals. These features have ledto the use of A. centrale as vaccine in anumber of countries in Africa, Asia, SouthAmerica and in Australia. Cross-immunitybetween A. centrale and A. marginale isadequate if challenge is moderate as inAustralia (Callow and Dalgliesh, 1980), butmay be insufficient against virulentheterologous challenge (de Vos and


Jorgensen in FAO, in press; Palmer, 1989).Despite this limitation and the risk ofreactions in older animals, use of A. centraleappears to be the only choice available tomost developing countries where ana-plasmosis is endemic. An advantage of A.centrale is its apparent non-transmissibilityby the common vectors of A. marginale,with the exception of the African tickRhipicephalus simas (Potgieter and vanRensburg, 1987).

A non-living vaccine based on a lyophili-zed preparation of A. marginale organismsadministered with adjuvant has beenavailable for many years as Anaplaz (FortDodge Laboratories) in the United Statesand Latin America (Brock, Kliewer andPearson, 1965). It induces partial protectionagainst virulent heterologous challenge butthe level of protection is less than thatprovided by live A. marginale vaccine(McHardy, 1984; Palmer, 1989). Neonataliso-erythrolysis has also been reportedfollowing use of this vaccine owing to theinduction of iso-antibodies to blood groupantigens (Palmer, 1989). Attempts todevelop a vaccine containing highlypurified Ana plasma spp. particles have beensuccessful (Luther et al., 1989; Montenegro-James et al., 1991) and a vaccine is availablein Louisiana, in the United States, whichdoes not induce sensitization against bloodgroups. In general, the limitations of non-living Ana plasma vaccines (difficulty ofmanufacturing them, their relative inef-fectiveness and the risk of contaminationwith erythrocyte stroma) render pro-duction of this type of vaccine impracticalin most developing countries. However,some of these vaccines are or werecommercially available in Central andSouth American countries.

The ability to induce protective im-munity with killed Ana plasma organismsled to further studies on purified antigens.Two surface proteins (36 and 105 kDa)induce a protective immune response in

calves to homologous and heterologouschallenge (Palmer, 1989; Tebele, McGuireand Palmer, 1991; McGuire et al., 1992).Future work will include the evaluation ofrecombinants of these proteins (McGuireet al., 1992).

Cowdria ruminantiumAnimals which recover from heartwateracquire an immunity to the disease. Thisknowledge has formed the basis of an"infection and treatment" method ofimmunization involving the inoculation ofanimals with Cowdria sp. and treatmentof ensuing reactions with tetracycline(Uilenberg, 1983). This procedure is usedroutinely to immunize cattle in SouthAfrica (Oberem and Bezuidenhout, 1987;Bezuidenhout, 1989). It has been standard-ized to include the use of blood of sheepacutely infected with a stock of Cowdria sp.known to cause well-defined febrilereactions, thus allowing the optimum timeof treatment to be determined. Because ofthe limited viability of the organisms inthe unfrozen state, cryopreservation isnecessary (Oberem and Bezuidenhout,1987). The innate resistance of younganimals to the clinical effects of Cowdriainfection are well documented and havebeen exploited by recommending immun-ization at a young age. Disadvantages ofthis procedure include the risk of reactionsin older animals, the large volume of theinoculum required (5 to 10 ml) and theneed to use the intravenous route ofinoculation (Uilenberg, 1983).

A supernatant of ground-up infectednymphal Amblyomma hebraeum ticks(GUTS) has also been used to immunizeanimals against Cowdria sp. (Oberem andBezuidenhout, 1987; Bezuidenhout, 1989).Details of the production of this vaccine,including the infection of larval ticks,the preparation of infective nymphsupernatant and the dilution and storageof the supernatant were reviewed by

Vaccine manual 79


Bezuidenhout (1989). The immunity con-ferred by this vaccine was similar to that ofthe blood vaccine but an unacceptably highnumber of animals, particularly goats,developed severe allergic or shock reactions(van der Merwe, 1987).

Considerable progress has also beenmade in recent years with the developmentof in vitro methods for propagating Cowdriasp. and a great number of stocks have beengrown successfully (Bezuidenhout andBrett, 1992). One roller bottle (surface area800 cm' ) with bovine or ovine endothelialcells can yield enough organisms to infect20 000 animals. There is also evidence ofattenuation in vitro without loss of im-munogenicity (Jongejan, 1991). Futurework will be aimed at optimizing cultureconditions for organism yield, improvingviability following freezing and thawing(Bezuidenhout and Brett, 1992) andovercoming the effect of antigenic dif-ferences between isolates (Jongejan et al.,1993).

Attempts are being made to identifyand express immunodominant Cowdriaproteins (Barbet et al., 1992). The ultimategoal of the Heartwater Research Project atthe University of Florida, United States, isto develop a recombinant vaccine deliveredby a live virus vector.

PRODUCTION OF VACCINES

The following methods of vaccine pro-duction are at present being used indeveloping countries or are likely to bewithin the means of countries wherereliance on enzootic stability is not anoption and immunization is the only viablealternative. As discussed in the previoussection, the vaccines all have the draw-backs inherent in the production and useof live products.

Babesia bovis and Babesia bigeminaSelection, attenuation and storage of stocks. Itis tempting to use Babesia isolates from the

country in which the vaccine is to beproduced. Such isolates can be obtainedby feeding infected Boophilus spp. ticks ona susceptible bovine host or by inoculatingit with blood from a long-standing carrier(Callow, 1984). The major limitation ofusing a carrier is the risk of contamination.Techniques for collecting, purifying andattenuating B. bovis and B. bigemina stockshave been described (de Vos and Jorgensenin FAO, in press).

Attempts to develop attenuated stocksof B. bovis from local isolates are not alwayssuccessful (de Vos and Jorgensen, 1992).For this reason, it may be worthwhile toimport a stock that is known to beprotective, of low virulence and free fromcontaminants. Australian stocks have beenshown to be protective in several countriesand vaccines containing these stocks havebeen used with beneficial results in Africa,Asia and South America (de Vos andJorgensen, 1992).

The attenuation of a B. bovis stock isusually achieved by rapid syringe passagethrough splenectomized calves (Callowand Dalgliesh, 1980; de Vos and Jorgensenin FAO, in press; Pipano, Frank and Shkap,1991). The mechanism by which at-tenuation is achieved is not fully under-stood but appears to be the result of aselective enrichment of avirulent parasitesubpopulations. Other methods used toattenuate B. bovis include irradiation(Wright, Goodger and Mahoney, 1980), invitro culture techniques (Yunker, Kuttlerand Johnson, 1987) and the use of alterna-tive hosts, but these methods give lessreliable results than passage in splenectom-ized calves. Rapid passage of B. bigemina insplenectomized calves is not recommended.A stock of this parasite with generally lowvirulence may be obtained after a series ofslow passages in non-splenectomized calves(Callow and Dalgliesh, 1980).

Vaccine stocks of B. bovis and B. bigeminacan be stored for lengthy periods as

cryopreserved stabilates of infected blood.Dimethyl sulphoxide (DMSO) (Mellors etal., 1982), glycerol (Dalgliesh, Jorgensenand de Vos, 1990) and polyvinyl pyr-rolidone (PVP) (de Vos and Jorgensen inFAO, in press) have all been used suc-cessfully to cryopreserve Babesia spp.parasites.

Supply of infective material using donorcalves. Cryopreserved master seed of anattenuated stock is used to initiate aprimary infection in a susceptible, sple-nectomized donor. The initial B. bovisparasitaemia is often low but adequate toinitiate a series of four to five passages,thus allowing a continuous supply ofvaccine over a three- to five-week period.B. bovis vaccine stocks should not bepassaged more than a total of 30 times toreduce the risk of losing immunogenicity(Callow and Dalgliesh, 1980; Bock et al., inpress; de Vos and Jorgensen, 1992). Inoculacontaining about 1 x 1010parasites are idealfor passaging and will result inparasitaemias in excess of 1 x 108 permillilitre within four to five days. Up to25 000 doses of vaccine can be obtainedfrom a calf weighing 150 to 200 kg.B. bigemina-infected blood is obtained fromindividually infected calves; serial passingis not recommended because of possibleselection for virulence.

Supply of infective material using parasitesgrown in vitro. Available technology allowsB. bovis and B. bigemina to be maintained incontinuous culture (Pudney, 1992). Theseculture systems have the advantage overthe calf-donor system of minimizing therisk of contamination and limiting thenumber of animals required. A relativelyrobust method for producing B. bigeminavaccine in sealed flask suspension cultureswas developed by Jorgensen et al. ( 1992)and may be a viable option in countrieswhere insufficient numbers of donors areavailable for the production of vaccine invivo. Some studies found that neither the

virulence nor the immunogenicity ofBabesia vaccine stocks were appreciablymodified by maintenance in culture(Timms et al., 1983; Jorgensen, de Vos andDalgliesh, 1989). However, recent workusing the polymerase chain reaction (PCR)of polymorphic genetic markers has shownthat proportions of B. bovis subpopulationsdo change with culture in vitro (Bock et al.,in press; Lew, unpubl.).

Theileria parvaIsolation, storage and characterization of stocks.Vaccine should be prepared using masterseed of well-characterized T. parva stocksto account for the antigenic heterogeneityof this species. If there is a requirement forlocal stocks, these can be obtained by usingbait or naturally infected cattle or bycollecting and feeding infected ticks fromthe field. Procedures used to isolate stocksof T. parva have been documented in theliterature (FAO, 1984; Purnell, 1977).However, considerable expertise is re-quired in the isolation and characterizationof isolates. Species identification is basedon examination of some or all of thefollowing: geographical distribution,vector specificity, morphology, hostspecificity, pathogenicity, serology, cross-immunity and DNA probes. Stock char-acterization traditionally involves cross-immunity studies, among others. Severalmonoclonal antibodies and DNA probeshave also been used to divide stocks of T.parva into groups (Minami et al., 1983;Conrad et al., 1987; Allsopp and Allsopp,1988) but more work is necessaryto determine whether polymorphismsdetected correlate with differences in cross-protection (Irvin and Morrison, 1989).

Preparation of ground-up tick supernatant.Details of the preparation of GUTS areavailable in the literature (FAO, 1984). Theprocess involves. the inoculation of cattlewith T. parva seed material. Approximatelyten days after inoculation, the cattle are

Vaccine manual 81

infested with non-infected nymphal ticks.Engorged ticks are harvested and allowedto moult. The level of infection can bedetermined by examining the number ofinfected acini in the salivary glands of someof the prefed adult ticks. The remainder ofthe adult ticks are fed on the ears of rabbits,detached after four days, washed, di-sinfected and ground up in a suitablemedium. Ground-up tick tissue is centri-fuged at a low speed and the supernatantcollected, mixed with glycerolized me-dium, equilibrated and cryopreserved. Onelitre of frozen vaccine can be prepared from10 000 adult ticks and this will provide1 000 1-ml doses of vaccine.

Theileria annulataIsolation of stocks. Isolation of T. annulataparasites from the field was reviewed byPipano (1989). Isolates can be obtained byinoculating cattle with the blood of infectedanimals or by exposing the cattle toinfected ticks. An alternative and moreelegant technique is to infect peripheralblood leucocytes in vitro with sporozoitesobtained from macerated infected ticks.

Propagation of schizonts in vitro. Techni-ques used to initiate and maintain T.annulata cultures are well documented(Brown, 1979; Pipano, 1989). Briefly,lymphocytes are collected from the bloodor organs of infected cattle and grown inmonolayer or suspension cultures. Alter-natively, normal peripheral blood lympho-cytes can be infected in vitro with sporo-zoites obtained from infected ticks (Brown,1979). Standard culture procedures areused and a wide variety of culture mediaare suitable. Yields of up to 9 x 10 schizont-infected lymphocytes can be obtained froma stationary culture vessel containing100 ml of rnedium, and even more fromroller bottles.

Continuous propagation of T. annulatain lymphoid cell cultures eliminates othertick-borne agents such as Babesia spp.,

Anaplasma spp. and Eperythozoon spp.Theoretically, however, some viral in-fections can be introduced into culturesfrom ticks or cattle used as a source ofinfected lymphocytes.

Partial or complete attenuation ofT. annulata is achieved by prolongedcultivation in vitro (Pipano, 1989). Com-plete attenuation (no clinical manifestationand no detectable tissue schizonts orerythrocytic merozoites) has been reportedafter 600 to 900 days in culture (Pipano,1989; Singh, 1990). Partially attenuatedschizonts have also been used for vac-cination (Zablotsky, 1988).

When the desired degree of attenuationis reached, schizont-infected cells can becryopreserved with a high degree ofsurvival upon reconstitution (Wathanga,Jones and Brown, 1986). DMSO or glycerolcan be used as cryopreservative. Masterseed stock is usually prepared and used toinitiate a series of passages in culture toyield parasites for vaccine. There is littleinformation on the selective effect oflong-term passaging of schizonts onimmunogenicity but there is evidence thatattenuated schizonts are less infective andless protective than virulent schizonts.Therefore, it is recommended that pro-duction cycles be restarted periodicallyfrom the same master seed (Pipano, 1989).

Anaplasma spp. using A. centrale asvaccine

Selection and storage of stock. Only one stockof A. centrale exists. The history of itsisolation in South Africa in 1911 wasrecorded by Potgieter (1979). This stockhas been used to initiate vaccine pro-duction programmes in several countries(Potgieter, 1979; Callow and Dalgliesh,1980; Pipano et al., 1986; Abdala et al., 1990).Master seed of A. centrale is prepared andstored in the same way as that of Babesiaspp.

Supply of infective material. Infective blood


containing suitable numbers of A. centraleorganisms for the production of vaccinecan be obtained in much the same way asBabesia parasites. Blood from acutelyinfected splenectomized calves is mostsuitable for the purpose although, in somelaboratories, the blood of carrier-donorsshowing quantifiable rickettsemia is usedand reported to be effective (Potgieter,1979). A. centrale vaccine can be preparedin frozen or chilled forms in the same wayas Babesia vaccines (de Vos and Jorgensenin FAO, in press). In some countries,for example Australia, mixed Babesia/Ana plasma vaccines are produced (Callowand Dalgliesh, 1980).

Cowdria ruminantiumSelection and storage of stocks. The isolationand characterization of Cowdria stocks havebeen reviewed in the literature (Uilen-berg, 1983; FAO, 1984; Oberem andBezuidenhout, 1987; Bezuidenhout, 1989).Isolates can be obtained from both thevertebrate host and the tick vector bysubinoculation of infected blood, organemulsions or supernatant of ground-up,engorging Amblyomma ticks. Stocks varyconsiderably in virulence and there is someevidence of immunological differences(Jongejan et al., 1993). However, immunitydoes not depend on the virulence of theisolate. It is important to select a stockwhich is as mild as possible but whichshows adequate cross-protection againstlocal isolates. Furthermore, it must also besusceptible to treatment with availableanti-Cowdria drugs. Only one stock (Ball 3)is used in South Africa. It generally causesa marked febrile response some daysbefore the onset of other clinical signs.

Stabilate of the stock used for productionof vaccine is produced in the same way asthe vaccine itself (see below) and stored at-70° or -196°C. Infective material has alsobeen freeze-dried successfully if keptbelow -18°C (du Plessis et al., 1990).

Production of infective material from donorsheep. The method used in South Africahas been described in some detail (FAO,1984; Oberem and Bezuidenhout, 1987;Bezuidenhout, 1989). Briefly, susceptiblesheep are inoculated with thawed stabilateand bled nine to ten days later at the heightof the febrile reaction. The stabilate isusually prepared from the blood ofreacting sheep but tick-derived stabilateswill fulfil the same purpose. Similarly,goats and cattle can also be used as donors.Blood from reacting animals is collecteddirectly into an equal volume of buffer towhich DMSO has been added. The dilutedblood is then snap-frozen and stored inliquid nitrogen. Up to 4 000 doses ofvaccine can be prepared from one sheep.

VACCINE DISPENSING AND DISPATCH

Dilution of vaccine concentrateSome vaccine starter materials, notablyTheileria GUTS and blood infected withBabesia spp. and Theileria spp., are veryconcentrated and can be diluted to increasethe vaccine yield. The appropriate mediumto be used as diluent depends on the startermaterial and whether the vaccine is to befrozen or not. Details of media used in theproduction of TBD vaccines have beenreviewed (FAO, 1984; Bezuidenhout, 1989;Pipano, 1989; de Vos and Jorgensen inFAO, in press).

Viability of Babesia and Anaplasmaorganisms can be maintained in chilledform by storing infected blood at 2° to 4°C.Even at this temperature, infectivity is lostexponentially, declining by about one logunit during the first week of storage(Callow, 1984). Chilled Babesia andAna plasma vaccines usually contain 0.5 to1 x 10 organisms per dose or about 100times the minimum infective numberrequired for subcutaneous inoculation. Therequired number of organisms is obtainedby diluting the infected blood with asuitable medium. Chilled T. annulata

Vaccine manual 83


vaccine can also be prepared from culturematerial, with the recommended numberof infected cells being 0.5 x 10 (Pipano,1989).

The minimum volume of a dose ofvaccine should preferably not be less than2 mito allow for inaccuracies in vaccinationequipment and procedures. An exeption isT. parva sporozoite vaccine where therecommended dose is 1 ml to allow forinoculation over the parotid lymph node.

Vaccine dispensingTechniques for producing frozen vaccinesusing DMSO as a cryoprotectant weredescribed by Mellors et al. (1982) andAbdala et al. (1990). Vaccine with glycerolas cryoprotectant has also been produced(Dalgliesh, Jorgensen and de Vos, 1990)and is known to remain viable for at leasteight hours after thawing. Frozen vaccineis dispensed in cryovials suitable for low-temperature storage.

Dispatch and cold chainThe types of transport networks and theirefficiency vary greatly between countriesand the choice depends largely on avail-ability. Rapid, reliable means of communi-cation and transport are desirable forfrozen vaccine and essential for chilledproducts. Frozen vaccine can be trans-ported in vacuum-insulated containerswith a refrigerant (liquid N2 or solid CO2).Vaccine should not be thawed until justbefore use and, once thawed, must not berefrozen. Chilled Babesia, Ana plasma andTheileria vaccines have a very short shelf-life, even when packed in ice, and maybecome ineffective if transport to thedestination exceeds 48 hours.

QUALITY ASSURANCE

Because of the live nature of TBD vaccines,quality control is essential to ensurepotency and to minimize the risk ofcontamination. Most of the early Babesia

and Anaplasma vaccines produced by thecarrier-donor method and the T. parvavaccines produced with tissue suspensionshad variable potencies, thus bringing thesevaccines into general disrepute. In somecountries in the tropics and subtropics, therisk of contamination is also very real.Quality assurance therefore needs to be anintegral part of any TBD vaccine pro-duction programme covering all operatingprocedures.

The standards of quality assurance ofBabesia, Ana plasma and Theileria vaccineshave been addressed by the Office inter-nationale des épizooties (OIE, 1991).However, these standards only cover themain aspects. Quality control of T. annulatavaccine in Israel is based on the proceduresfor testing live vaccines described in theUnited States Code of Federal Regulationsfor Animals and Animal Products. InAustralia, procedures for the productionof Babesia and Ana plasma vaccines are beingmodified to comply with the AustralianCode of Good Manufacturing Practice forVeterinary Preparations. Compliance withthese or similar codes may not be possiblein all developing countries. However, thepotential consequences of providingsubstandard or contaminated vaccineshould be considered. The contaminationof one batch of Babesial Anaplasma vaccinein Australia with enzootic bovine leucosis(Rogers et al., 1988) resulted in the paymentof USS1.5 million in compensation.

Quality control should be performed atboth the preproduction and postprod-uction stages. Preproduction qualitycontrol is particularly important in the caseof chilled vaccine but is, in itself, notadequate to guarantee potency and purity.It should include facilities, the docu-mentation of standards and procedures forobtaining and quarantining suitable donoranimals and the preproduction testing ofanimals, master seed and other startingmaterials used in the production of vaccine.

It should also include environmentalmonitoring to ensure that there is only aminimal risk of infectious challenge in theregion where the production facility islocated.

Postproduction quality control shouldbe aimed at determining potency, purityand, where relevant, virulence. Potencyand virulence are monitored by inoculatinggroups of susceptible cattle and thenmonitoring the reactions. Dalgliesh,Jorgensen and de Vos (1990) described amethod for testing the potency of frozenBabesia vaccine. Similar procedures havebeen described for Theileria spp. (FAO,1984) and Cowdria sp. (Bezuidenhout,1989). Each batch should also be tested forfreedom from contaminants, with thechoice of tests depending on the diseasessuspected and the type of vaccine con-cerned. Tick-derived sporozoite vaccinesnaturally do not need to be scrutinized asintensely as vaccines produced from theblood of infected animals but there is arisk of other TBDs, including CrimeanCongo haemorrhagic fever.

FACTORS CRITICAL TO THE PRODUCTION OF

TBD VACCINES IN DEVELOPING COUNTRIES

Availability of stocks known to beprotective against local challenge andshowing other desirable traits.Availability of disease-free donor ani-mals as a source of infected blood,infected ticks and other starter materials.This may require the animals to be bredunder tick- and arthropod-free con-ditions for the specific purpose of vaccineproduction.Availability of suitable facilities tomaintain disease-free animals and toperform the necessary procedures (sple-nectomies, feeding of ticks, collection ofblood).Laboratory facilities and expertise tomonitor donor cattle and starter ma-terials (infected blood, culture material,

infected ticks) for quality controlpurposes.Suitable laboratory facilities and services,including reliable electrical supplies andlow-temperature storage facilities.Access to foreign capital for the purchaseof equipment, reagents and other sup-plies needed for production and qualitycontrol.Access to reliable transport networks.A core of dedicated, trained staff.Access to training opportunities, pro-cedures and proven master seed.

KEY LABORATORIES PRODUCING TBD

VACCINES

The following laboratories are involved inthe production of TBD vaccines and maybe able to provide suitable master seed,details of production procedures andquality control as well as training op-portunities.

Babesia spp.

Tick Fever Research Centre, Wacol,Queensland, AustraliaKimron Veterinary Institute, Bet Dagan,IsraelVeterinary Research Institute, Onderste-poort, South AfricaCentral Veterinary Laboratory, Li-longwe, MalawiInstituto Nacional de Tecnologia Agro-pecuaria, Rafaela, ArgentinaMiguel C. Rubino Laboratory, Pando,Uruguay

Theileria parva

Central Veterinary Laboratory, Li-longwe, MalawiILRAD, Nairobi, Kenya

Theileria annulata

Kimron Veterinary Institute, Bet Dagan,IsraelAnimal Disease Research Laboratory,NDDB, Anand, India

Vaccine manual 85


AnaplasmaSee Babesia spp. list.

Cowdria spp.Veterinary Research Institute, Onderste-poort, South Africa

BIBLIOGRAPHY

Abdala, A.A., Pipano, E., Aguirre, D.H.,Gaido, A.B., Zurbriggen, M.A.,Mangold, A.J. & Guglielmone, A.A.1990. Frozen and fresh Anaplasma centralevaccines in the protection of cattle againstAnaplasma marginale infection. Rev.Elevage Méd. Vét. Pays Trop., 43:155-158.

Allsopp, B.A. & Allsopp, M.T.E.P. 1988.Theileria parva: genomic DNA studiesreveal intraspecific sequence diversity.Mol. Biochem. Parasitol., 28: 77-84.

Barbet, A.F., Mahan, S., Allred, D., McGuire,T.C., Palmer, G.H. & Yunker, C.E. 1992.Molecular biology of rickettsiae: geneorganisation and structure in Ana plasmamarginale and Cowdria ruminantium. InT.T. Dolan, ed. Recent developments inthe control of anaplasmosis, babesiosis andcowdriosis, p. 87-91. Nairobi, ILRAD.

Barre, N., Uilenberg, G., Morel, P.C. &Camus, E. 1987. Danger of introducingheartwater onto the American mainland:potential role of indigenous and exoticAmblyomma ticks. Onderstepoort J. Vet.Res., 54: 405-417.

Bezuidenhout, J.D. 1989. Cowdria vaccines.In I.G. Wright, ed. Veterinary protozoanand hemoparasite vaccines, p. 31-42. BocaRaton, Fla., USA, CRC.

Bezuidenhout, J.D. & Brett, S. 1992. Progresswith the cultivation of Cowdriaruminantium in endothelial cells. In T.T.Dolan, ed. Recent developments in thecontrol of anaplasmosis, babesiosis andcowdriosis, p. 141-147. Nairobi, ILRAD.

Bock, R.E., de Vos, A.J., Kingston, T.G.,Shiels, I.A. & Dalgliesh, R.J. 1992.Investigations of breakdowns inprotection provided by living Babes ja

bovis vaccine. Vet. Parasitol., 43: 45-56.Bock, R.E., de Vos, A.J., Lew, A., Kingston,

T.G. & Fraser, I.R. Studies on failure ofT strain live Babesia bovis vaccine. Aust.Vet. J. (in press)

Brock, W.E., Kliewer, I.O. & Pearson, C.C.1965. A vaccine for anaplasmosis. J. Am.Vet. Med. Assoc., 147: 948-951.

Brown, C.G.D. 1979. Propagation of Theileria.In K. Maramorosch & H. Hirumi, eds.Practical tissue culture applications, p. 223-254. New York, Academic.

Brown, C.G.D. 1981. Application of in vitrotechniques to vaccination againsttheileriosis. In A.D. Irvin, M.P.Cunningham & A.S. Young, eds.Advances in control of theileriosis, p.104-119. The Hague, Martinus Nijhoff.

Burridge, M.J., Morzaria, S.P., Cunningham,M.P. & Brown, C.G.D. 1972. Durationof immunity to East Coast fever (Theileriaparva) infection in cattle. Parasitol., 64:511-515.

Callow, L.L. 1984. Animal health in Australia,Vol. 5. Protozoal and rickettsial diseases,p. 1-264. Canberra, Australian Govern-ment Publishing Service.

Callow, L.L. & Dalgliesh, R.J. 1980. Thedevelopment of effective, safevaccination against babesiosis andanaplasmosis in Australia. In L.A.Y.Johnston & M.G. Cooper, eds. Ticks andtick-borne diseases, p. 4-8. Sydney,Australian Veterinary Association.

Conrad, P.A., Jams, K., Brown, W.C.,Sohanpal, B. & Ole-Moi Yoi, 0. 1987.DNA probes detect genomic diversityin Theileria parva stocks. Mol. Biochem.Parasitol., 25: 213-226.

Correr, D.F., Johnson, J.S. & Wagner, G.G.1985. Demonstration of vaccine-inducedimmunity to anaplasmosis withoutinduction of persistent post-vaccinalcomplement-fixing and agglutinatingantibodies in yearling steers. Am. J. Vet.Res., 46: 583-586.

Dalgliesh, R.J. 1993. Immunity, immuno-

Vaccine manual

pathology and molecular biology inbabesiosis. In K.A. Warren & N. Agabian,eds. Immunology and molecular biology ofparasitic infections, p. 350-381. New York,Blackwell Scientific.

Dalgliesh, R.J., Jorgensen, W.K. & de Vos,A.J. 1990. Australian frozen vaccinesfor the control of babesiosis and ana-plasrnosis in cattle - a review. Trop.Anim. Health Prod., 22: 44-52.

Dalrymple, B.P., Jorgensen, W.K., de Vos,A.J. & Wright, I.G. 1992. Analysis ofthe composition of samples of Babesiabovis and the influence of differentenvironmental conditions on geneticallydistinct subpopulations. Int. J. Parasitol.,22: 31-737.

Danforth, H.D. & Augustine, P.C. 1989.Coccidia vaccines. In I.G. Wright, ed.Veterinary protozoan and hemoparasitevaccines, p. 165-175. Boca Raton, Fla.,USA, CRC.

de Vos, A.J. 1992. Distribution, economicimportance and control measures forBabesia and Anaplasma. In T.T. Dolan,ed. Recent developments in the control ofanaplasmosis, babesiosis and cowdriosis,p. 3-12. Nairobi, ILRAD.

de Vos, A.J. & Jorgensen, W.K. 1992.Protection of cattle against babesiosisin tropical and subtropical countries witha live, frozen vaccine. In B.H. Fivaz,T.N. Petney & I.G. Horak, eds. Tick vectorbiology - medical and veterinary aspects,p. 159-174. Berlin, Springer.

Dolan, T.T. & McKeever, D.J. 1993. Currentand future vaccines against theileriosis.In R. Pandey, S. Hoglund & G. Prasad,eds. Veterinary vaccines, p. 318-337. Berlin,Springer.

du Plessis, J.L., van Gas, L., Labuschagne,F.J. & Wijma, S. 1990. The freeze-dryingof Cowdria ruminantium. OnderstepoortJ. Vet. Res., 57: 145-149.

FAO. 1984. Tick and tick-borne disease control.A practical field manual, Vol.II. Tick-bornedisease control, p. 300-621. Rome.

FAO. Babesia bovis, Babesia bigemina andAna plasma marginale vaccines. In Ticksand tick-borne disease control - a practicalfield manual. Rome. (in press)

Friedhoff, K.T. 1988. Transmission of Babesia.In M. Ristic, ed. Babesiosis of domesticanimals and man, p. 23-52. Boca Raton,Fla., USA, CRC.

Hall, R. & Baylis, H.A. 1993. Tropicaltheileriosis. Parasitol. Today, 9: 310-312.

Irvin, A.D. & Morrison, W.I. 1989. Vaccinesagainst Theileria parva. In I.G. Wright,ed. Veterinary protozoan and hemoparasitevaccines, p. 115-130. Boca Raton, Fla.,USA, CRC.

Jongejan, F. 1991. Protective immunity toheartwater (Cowdria ruminantium infe-ction) is acquired after vaccination within vitro attenuated rickettsia. Infect.Immun., 59: 729-731.

Jongej an, F., Vogel, S.W., Gueye, A. SEUilenberg, G.1993). Vaccination againstheartwater using in vitro attenuatedCowdria ruminatium organisms. Rev.Elevage Méd. Vét. Pays Trop., 46: 223-227.

Jorgensen, W.K., de Vos, A.J. & Dalgliesh,R.J. 1989. Comparison of immuno-genicity and virulence between Babesiabigemina parasites from continuousculture and from a splenectomized calf.Aust. Vet. J., 61: 371-372.

Jorgensen, W.K., Waldron, S.J., McGrath,J., Roman, R.J., de Vos, A.J. & Williams,K.E. 1992. Growth of Babesia bigeminaparasites in suspension cultures forvaccine production. Parasitol. Res., 78:423-426.

Jorgensen, W.K., Bock, R.E., de Vos, A.J. &Shiels, I.A. 1993. Sheep-adapted Ana-plasma marginale maintains virulence forcattle. Aust. Vet. J., 70: 192-193.

Lawrence, J.A. 1992. History of bovinetheileriosis in southern Africa. In R.A.I.Norval, B.D. Perry & A.S. Young, eds.The epidemiology of theileriosis in Africa,p. 1-39. London, Academic.

Lawrence, J.A., Foggin, C.M. & Norval, R.A.I.1980. The effects of war on the controlof diseases of livestock in Rhodesia(Zimbabwe). Vet. Rec., 107: 82-85.

Luther, D.G., Hart, L.T., Morris, N.G.,McRae, B. & Todd, W.J. 1989. Fieldtrial of an experimental anaplasmosisvaccine in Louisiana. Proc. Annu. Meet.US Anim. Health Assoc., 93: 35-38.

McCosker, P.J. 1981. The global importanceof babesiosis. In M. Ristic, ed. Babesiosis,p. 1-24. New York, Academic.

McGuire, T.C., Barbet, A.F., Tebele, N.,McElwain, T.F. & Palmer, G.H. 1992.Progress in development of subunitvaccines for anaplasmosis. In T.T. Dolan,ed. Recent developments in the control ofanaplasmosis, babesiosis and cowdriosis, p.99-104. Nairobi, ILRAD.

McHardy, N. 1984. Immunisation againstanaplasmosis - a review. Prey. Vet. Med.,2: 135-146.

Mellors, L.T., Dalgliesh, R.J., Timms.,Rodwell, B.J. & Callow, L.L. 1982.Preparation and laboratory testing offrozen vaccine containing Babesia bovis,Babesia bigemina and Anaplasma centrale.Res. Vet. Sci., 32:194-197.

Minami, T., Spooner, P.R., Irvin, A.D.,Ocama, J.G.R., Dobbelaere, D.A.E. &Fujinaga, T. 1983. Characterisation ofstocks of Theileria parva by monoclonalantibody profiles. Res. Vet. Sci., 35: 334-340.

Montenegro-James, S., Kakoma, I. & Ristic,M. 1989. Culture-derived Babesia exo-antigens as immunogens. In I.G. Wright,ed. Veterinary protozoan and hemoparasitevaccines, p. 61-97. Boca Raton, Fla., USA,CRC.

Montenegro-James, S., James, M.A., ToroBenitez, M., Leon, E., Baek, B.K. &Guillen, A.T. 1991. Efficacy of purifiedAnaplasma marginale initial bodies as avaccine against anaplasmosis. Parasitol.Res., 77: 93-101.

Musoke ,A.J., Morzaria, S.P., Nkonge, C.,

Jones, E. & Nene, V.1992. A recombinantsporozoite surface antigen of Theileriaparva induces protection in cattle. Proc.Nat. Acad. Sci. USA, 89: 514-518.

Norval, R.A.I., Perry, B.D. & Young, A.S.1992. The epidemiology of theileriosis inAfrica, p. 1-481. London, Academic.

Oberem, P.T. & Bezuidenhout, J.D. 1987.The production of heartwater vaccine.Onderstepoort J. Vet. Res., 54: 485-488.

OIE. 1991. Manual of recommended diagnostictechniques and requirements for biologicalproducts, Vol. III. Paris.

Palmer, G.H. 1989. Ana plasma vaccines. InI.G. Wright, ed. Veterinary protozoan andhemoparasite vaccines, p. 1-29. Boca Raton,Fla., USA, CRC.

Pipano, E. 1989. Vaccination against Theileriaannulata theileriosis. In I.G. Wright, ed.Veterinary and hemoparasite vaccines,p. 203-234. Boca Raton, Fla., USA, CRC.

Pipano, E., Frank, M. & Shkap, V. 1991.Current methods for the control of tickfevers in cattle. Is. J. Vet. Med., 46: 79-88.

Pipano, E., Krigel, Y., Frank, M., Markovics,A. & Mayer, E. 1986. Frozen Ana plasmacentrale vaccine against anaplasmosisin cattle. Br. Vet. J., 142: 553-556.

Potgieter, F.T. 1979. Epizootiology andcontrol of anaplasmosis in South Africa.J. S. Afr. Vet. Assoc., 50: 367-372.

Potgieter, F.T. & van Rensburg, L.T. 1987.Tick transmission of Anaplasma centrale.Onderstepoort J. Vet. Res., 54: 5-7.

Pudney, M. 1992. Cultivation of Babesia. InT.T. Dolan, ed. Recent developments inthe control of anaplasmosis, babesiosis andcowdriosis, p. 129-140. Nairobi, ILRAD.

Purnell, R.E. 1977. East Coast fever: somerecent research in East Africa. Adv.Parasitol., 15: 83-132.

Radley, D.E., Brown, C.G.D., Burridge, M.J.,Cunningham, M.P., Kirimi, I.M.,Purnell, R.E. & Young, A.S. 1975a. EastCoast fever. 1. Chemoprophylacticimmunisation of cattle against Theileria


parva (Muguga) and five Theileria strains.Vet. Parasitol., 1: 5-41.

Radley, D.E., Brown, C.G.D, Cunningham,M.P., Kimber, C.D., Musisi, F.L., Payne,R.C., Purnell, R.E. & Stagg, S.M. 1975b.East Coast fever. 3. Chemoprophylacticimmunisation of cattle using oxytetra-cycline and a combination of Theileriastrains. Vet. Parasitol., 1: 51-60.

Rogers, R.J., Dimmock, C.K., de Vos, A.J.& Rodwell, B.J. 1988. Bovine leucosisvirus contamination of a vaccine pro-duced in vivo against bovine babesiosisand anaplasmosis. Aust. Vet. J., 65:285-287.

Singh, D.K. 1990. Methods currently usedfor the control of Theileria annulata: theirvalidity and proposals for future controlstrategies. Parasitologia, 32: 33-40.

Tebele, N., McGuire, T.C. & Palmer, G.H.1991. Induction of protective immunityusing Ana plasma marginale initial bodymembranes. Infect. Immun., 59: 3199-3204.

Timms, P., Dalgliesh, R.J., Barry, D.N.,Dimmock, C.K. & Rodwell, B.J. 1983.Babesia bovis: comparison of culture-derived parasites, non-living antigen andconventional vaccine in the protectionof cattle against heterologous challenge.Aust. Vet. J., 60: 75-77.

Uilenberg, G. 1983. Heartwater (Cowdriaruminantium infection): current status.Adv. Vet. Sci. Comp. Med., 27: 427-480.

van der Merwe, L. 1987. The infection andtreatment method of vaccination againstheartwater. Onderstepoort J. Vet. Res.,54: 489-491.

Wathanga, J.M., Jones, T.W. & Brown,C.G.D. 1986. Cryopreservation ofTheileria infected lymphoblastoid cellswith functional assessment of viability.Trop. Anim. Health Prod., 18: 191-197.

Wright, I.G. 1991. Towards a synthetic Babesiavaccine. Int. J. Parasitol., 21: 156-159.

Wright, I.G., Goodger, B.V. & Mahoney,D.F. 1980. The irradiation of Babesia bovis.I. The difference in pathogenicity

between irradiated and non-irradiatedpopulations. Z. Parasitol., 63: 47-57.

Wright, I.G., Casu, R., Commins, M.A. &Dalrymple, B.P. 1992. The developmentof a recombinant Babesia vaccine. Vet.Parasitol., 44: 3-13.

Yunker, C.E., Kuttler, K.L. & Johnson, L.W.1987. Attenuation of Babesia bovis by invitro cultivation. Vet. Parasitol., 24: 7-13.

Zablotsky, V.T. 1988. Response of cattle ofdifferent ages to Theileria annulatavaccine, II. Reaction and immune res-ponse of calves of different breeds totheileriosis vaccine. Trudy InstituteEksperimental'noi Veterinarii, 66: 62-75.(in Russian)

Vaccine manual 89

,.[1E1 L-,S

THE NECESSITY FOR ANTIPARASITE VACCINES

Diseases caused by multicellular parasitesare a major cause of mortality andmorbidity in humans and domesticanimals throughout the world. Jenner'sdiscovery that cowpox virus was effectivein immunization against smallpox sti-mulated research into vaccination againstmost major pathogens including multi-cellular parasites. Unfortunately, thesuccesses achieved against viruses andbacteria have not been paralleled in thecontrol of diseases caused by multicellularparasites. At present these are controlledby chemotherapy, although i t is nowapparent that resistance is an inevitableconsequence of prolonged chemical ap-plication. Anthelmintic resistance has beenreported against all the broad-spectrumagents currently available (Prichard, 1990).For example, benzimidazole resistance hasbeen found in Haemonchus contortus,Ostertagia circumcincta and Trichostrongyluscolubriformis nematode infections in sheepand goats throughout the world, whilelevamisole and morantel resistance hasbeen recorded with increasing frequencyin Australasia and South Africa (Prichard,1990; Jackson, 1993). Of concern is the factthat resistance to the relatively recentlydeveloped ivermectin has been experi-mentally selected in H. con tortus (Egerton,Suhayda and Eary, 1988) and has now beenreported in the field in H. con tortus in bothsheep (van Wyck and Malan, 1988) and

The authors would like Lo thank Professor AndyTait, Dr Eileen Devaney and Dr Bob Coop for theirsupport in the preparation of this chapter.

Cd p VIC C

J.B. McKeand and D.P. Knox

goats (Jackson et al., 1991). There iswidespread acaricide resistance among theimportant cattle ticks, Boophilus microplusand Hyalomma anatolicum anatolicum(Wharton, 1976) and an increasingincidence of insecticide resistance isevident in the sheep blowfly, Lucilia cuprina(Arundel and Sutherland, 1988).

Although the rapid spread of drugresistance in parasite populations can to adegree be reduced by strategic manage-ment procedures, these can impose consi-derable restraints on land use, especiallyon marginal pastures. The cost of devel-oping novel chemical control agents isprohibitive and considerable effort is nowbeing directed towards prolonging theefficacy of the currently available drugs,with a particular emphasis on minimizingthe effects of drug resistance. Furthermore,there is now widespread public concernregarding the effects° of chemical residuesin both animal products and the environ-ment. There is clearly a requirement todevelop novel control methods for pa-rasitic disease.

Recently, biological control strategies,such as the use of nematophagous fungi,have been evaluated for the control ofgastrointestinal nematodes of domesticanimals (Waller and Larsen, 1993) but mostresearch has been directed at the devel-opment of antiparasite vaccines. Despitedecades of intensive work, only twocommercially available multicellularparasite vaccines exist. These are based onradiation-attenuated larvae for the controlof Dictyocaulus viviparus in cattle (Jarrett etal., 1958) and Dictyocaulus filaria in sheep(Sharma, Bhat and Dhar, 1988). It would

Vaccine manual 91

92 Multicellular parasite vaccines

appear that the antigenic complexity ofmulticellular parasites has provided aneffective barrier to the development ofuseful vaccines. To compound this, it isoften the case that protective antiparasiteimmune responses are absent or deficientin immature animals. The reasons for thisapparent unresponsiveness are unclearbut, given that production lossesprimarily accrue owing to the infection ofyoung stock, studies must be under-taken to identify the cause of this and todevelop vaccines which are effective in thisgroup.

Recent advances in immunology, proteinchemistry and recombinant DNA techni-ques have enabled a more accurate iden-tification of potential host-protectiveimmune responses and the parasite anti-gens which stimulate them. In order tounderstand the problems and difficultiesassociated with the development ofvaccines against multicellular parasites,this review provides a historical accountof the work that has been undertaken andthen continues with a description of recentadvances which suggest that the devel-opment of subunit parasite vaccines maybe an achievable goal. Numerous kinds ofparasite preparations have been used toimmunize animals. These range from theuse of attenuated whole parasites, throughcrude extracts of dead parasites to specificpurified molecules.

VACCINATION WITH RADIATION-ATTENUATED

PARASITES

Parasites attenuated by irradiation do notreach patency but can stimulate protectiveimmune responses without significantparasite-induced pathology. Immunizationwith irradiated third-stage larvae (L3) ofthe bovine lungworm, D. viviparus, is ahighly successful example of vaccinationusing this method of attenuation. Naturalinfection with this parasite induces highlevels of resistance in calves (Poynter and

Cauthen, 1942; Jarrett, McIntyre andUrquhart, 1954). Vaccination studies usingtwo doses of 10 000 40 kilorad- (400 joulesper kilogram-) irradiated L3 showed thatcalves could be significantly protectedagainst challenge using attenuated larvae(Jarrett et al., 1958). This vaccine wasdeveloped commercially and has beenused successfully for the controlof dictyocaulosis in the United Kingdomand other parts of Europe ever since.This pioneering breakthrough led toseveral attempts to develop similarvaccines against other important helminthspecies.

In India, a successful irradiated larvalvaccine was developed for the control ofthe pathogenic ovine lungworm, D. filaria(Sharma, Bhat and Dhar, 1988). Vaccinationof dogs against the intestinal nematodeAncylos toma caninum was similarly ef-fective (Miller, 1971). Unfortunately, thisvaccine failed commercially because of itsrespiratory side-effects, short shelf-life and,in some cases, lack of sterile immunity(Miller, 1978). Schistosoma spp. are trema-todes of importance in humans andanimals in tropical regions and successfulimmunization has been achieved usingirradiated cercariae in laboratory models(Simpson et al., 1985). Furthermore, zebucattle were significantly protected against

bovis challenge following vaccinationwith irradiated schistosomula or cercariaeof the homologous species (Bushara et al.,1978).

Other attempts to immunize animalswith irradiated parasites have met withless success. Mature sheep (> six months)were protected against homologouschallenge using irradiated L3 of eitherH. contortus (Urquhart et al., 1966b) or

colubriformis (Gregg and Dineen, 1978).However, as in natural infection, immaturesheep (< three months) failed to developimmunity following vaccination (Urquhartet al., 1966a; Gregg et al., 1978).

VACCINATION USING MATERIAL

FROM DEAD PARASITES

Attempts to vaccinate against parasitichelminths using somatic extracts fromdead organisms were first made in the1930s but achieved limited success(Chandler, 1932). However, some subse-quent attempts have given protection: forexample, extracts of fourth-stage larvae ofT. colubriformis induced significant levelsof protection in guinea pigs againsthomologous challenge (Rothwell andLove, 1974) and worm burdens in sheepimmunized with sonicates of adult stageH. con tortus were significantly lower (63percent) than burdens found in challengecontrols (Adams, 1989). Somatic extractsderived from onchospheres of the cestodesEchinococcus sp. and Taenia sp. were alsofound to be potent sources of host-protective antigens (Rickard and Williams,1982; Xilinas, Papavasiliou and Marselou-Kinti, 1976). These results appear to be theexception rather than the rule, since in mostother species success has been limitedwhen somatic helrninth extracts were usedfor immunization (Clegg and Smith, 1978).

Similar to the situation with helminths,somatic extracts of ectoparasites haveinduced variable degrees of protectiveimmunity (Willadsen, 1980). Salivary glandextracts induced protective immunity toB. microplus (Brossard, 1976) andAmblyomma maculatum (McGowan et al.,1981) in cattle and also in Dermacentorandersoni (Wikel, 1981) in guinea pigs.Subsequently, Opdebeeck et al. (1988)vaccinated calves with crude soluble andmembrane extracts from the midgut ofpartially engorged female B. micro plus andobserved up to a 98 percent reduction intick numbers following challenge. Sheepproduce antibodies to L. cuprina, theprimary agent responsible for fly strike inAustralia (O'Donnell et al., 1980) andvaccination of sheep with soluble extractsof third-instar larvae stimulated antibodies

which inhibited larval growth anddevelopment in vitro (O'Donnell et al.,1981). Larvae which were harvested fromsheep previously vaccinated with solublecomponents of second-instar larvae were58 percent smaller than larvae from controlsheep (Johnston, Kemp and Pearson, 1986).

VACCINATION WITH MATERIAL

FROM LIVE PARASITES

Although some studies using crudeextracts of dead parasites gave protection,few of these studies were taken furtherowing to the difficulty in defining theprotective components within these com-plex extracts, and attention increasinglyfocused on less complex parasite com-partments which may comprise compo-nents essential to parasite survival withinthe host.

Sarles and Taliaferro (1936) firstsuggested that products released fromhelminths in vivo were involved in hostimmunity' when they observed immuneprecipitates at the mouth, excretory poreand anus of Nippostrongylus brasiliensisparasites which had been incubated withserum from immune rats. It was thoughtthat the antigen/ antibody complexesformed at these parasite orifices might actby blocking the activity of secretionsessential to parasite survival. Severalattempts to immunize laboratory animalswith these excretory-secretory (ES)antigens have produced encouragingresults. For example, protective immunitywas achieved in mice using the ES productsof first-stage larvae of Trichinella spiralis(Vernes, 1976) as well as in guinea pigsusing the ES products of fourth-stage T.colubriformis (Rothwell and Love, 1974), theES products of third- and fourth-stageAscaris suum (Stromberg and Soulsby,1977) and the ES products of adultD. viviparus (McKeand et al., 1995).

Immunization using ES materials has notalways been successful, however, as

Vaccine manual 93


demonstrated by Neilson (1975) whoimmunized lambs with the ES products ofL3 and L4 H. con tortus and found nodifferences in faecal egg counts or wormburdens between vaccinates and challengecontrols. The lambs used in this studyhowever, were less than three months oldand may have been immunologicallyunresponsive. Alternatively, the result mayhave reflected differences in the immuno-genicity of ES in laboratory animalscompared with that in natural hosts.

Materials released from ectoparasiteshave also been proposed as major immuno-gens. Fly strike is initiated by enzymessecreted by the fly larvae on to the sheep'sskin and sheep exposed to several infesta-tions strongly recognized antigens derived,and probably released from, the larval gutand salivary glands (Skelly and Howells;1987). It has been postulated that inflam-mation induced by larval ES productsplays a central role in wound formationand exudation and that these, in turn, may-reduce larval establishment (Sandeman,1990). More refined vaccination studies inwhich the potential of secreted larvalenzymes as vaccine candidates have beenindicated are outlined later.

Despite all these studies, until recentlyno one has been able to identify and isolateprotective components within thesefractions and incorporate them intocommercially viable vaccines. With theadvent of improved immunochemicalmethods for identifying protective antigensand the ability to produce these in largequantities using recombinant DNA tech-nology, it now appears possible to developsubunit parasite vaccines. The likely stepsin subunit parasite vaccine developmentare summarized (Figure 2).

DEVELOPMENT OF SUBUNIT MULTICELLULAR

PARASITE VACCINES

Historically, the selection of protectiveantigens has been based on extracting

native antigen on a small scale fromparasites harvested from donor animals orfrom parasites maintained in vitro. Thiswould clearly be unsuitable for commercialvaccine production. Emery and Wagland(1991) estimated that, to obtain sufficientnematode antigen to vaccinate one sheep,parasites would have to be harvested fromthree donor animals! The problem ofantigen production can now be resolvedby using peptide synthesis and / or re-combinant DNA technology. The mainchallenges are to identify appropriateindividual protective antigens suitable forincorporation into subunit vaccines whilemaintaining their immunogenicity.

Antigen selectionWhile many parasite proteins are antigenic,the immune responses to these antigensmay not be protective (O'Donnell et al.,1989). Parasite antigens can be classifiedas conventional or covert. Conventionalantigens are recognized by the host duringthe course of a natural infection arid, asvaccine components, would augmentnatural immunity. However, mutations inthe genes coding for these antigens may beselected by the pressure associated withimmunity in vaccinated hosts and couldrestrict their long-term use. On the otherhand, covert antigens are hidden from thehost immune system during naturalinfection and, in the absence of selection,are less likely to exhibit antigenic variation.The main disadvantage of using covertantigens is that repeated vaccination maybe necessary as specific host immuneresponses would not be boosted duringsubsequent natural infections. However, ifanimals are protected during the verysusceptible period in early life, as forexample in haemonchosis in sheep, subse-quent exposure to natural challenge rnaymaintain protective levels of immunity.

Antigens can be selected on either anempirical or a rational basis. The empirical

Antigen selection

11Pr

Rational or empirical approach

111F

Vaccination studies using native antigen

1PirProduction of recombinant antigen

lirVaccination studies with recombinant antigen

VrIncorporation of optimal delivery system adjuvant, vector, etc.

VrField evaluation and commercial development

FIGURE 2

Steps in the development of subunit parasite vaccines

Vaccine nual 95


approach relies on the fractionation ofparasite material followed by successiveimmunization studies. These extractsprovide the starting material forprogressively more refined fractionationand the ultimate definition of specificprotective components. Fractionation hasnormally been achieved using con-ventional protein chemistry techniquessuch as gel filtration, ion exchange chroma-tography and polyacrylamide gel electro-phoresis. Empirical evaluation is, however,extremely time-consuming and costly so amore rational approach, where moleculesare selected on the basis of theircontribution to parasite survival, has manyadvantages.

Despite its drawbacks, the empiricalapproach has resulted in the production ofsome viable vaccine candidates. Forexample, in Australia antigen fractionationhas led to the identification of tropomyosinas a candidate antigen for stimulatingprotective immunity in sheep againstT. colubriformis and H. contortus.Immunization with whole wormhomogenates from fourth-stage larvae ofT. colubriformis was observed to accelerateexpulsion of homologous challenge inguinea pigs (Rothwell and Love, 1974;Rothwell and Griffiths, 1977; Rothwell,1978). Subfractions of this homogenate,obtained by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE), were later shown to induce hostprotective immunity (O'Donnell et al.,1985). These subfractions, however, stillcomprised rather complex proteinmixtures, so different extraction pro-cedures were then evaluated to reducetheir complexity (O'Donnell et al., 1989).Using phosphate-buffered saline/ sodiumdeoxycholate (PBS / NaDOC) to makesomatic extracts of T. colubriformis larvae,a simple antigen mixture was producedwhich -consisted of four protein com-ponents, one of which, a 41 kilodalton

(kDa) protein, induced 43 to 51 percentprotection in guinea pigs (O'Donnell et al.,1989). Partial amino acid sequence analysisof this protein indicated that it wastropomyosin. Progress towards arecombinant vaccine based on parasitetropomyosin is outlined in the sectionMulticellular parasite vaccines: recentdevelopments, p. 100.

A slightly different approach has beenused to isolate protective antigen fractionsfrom Oesophagostomum radiatum, a patho-genic nematode found in the large intestineof calves. Animals acquire strong resistanceto reinfection (Roberts, Elek and Keith,1962) and protective immunity was stimul-ated using excretory gland homogenatesof adult parasites (Keith and Bremner,1973). The adult O. radiatum extracts wereresolved into four fractions by gel filtration(East, Berrie and Fitzgerald, 1989) andELISA analyses showed that antibodiesfrom naturally infected calves predomin-antly reacted with the void volume fractionwhich comprised high molecular weightcomponents. Vaccination of calves withthis fraction significantly reduced wormestablishment and faecal egg output (East,Berrie and Fitzgerald, 1988). High mole-cular weight antigens have also been usedto vaccinate against Nematospiroides dubiusin mice (Monroy and Dobson, 1987), T.colubriformis (O'Donnell et al., 1985) and H.contortus (Neilson and Van de Walle, 1987)in sheep, with varying degrees of success.

By using antibodies or cells derived frominfected and / or immune animals to probecrude antigen preparations, the majorimmunogens can be identified. Thisstrategy can be augmented by comparingthe responses of individual hosts whichare defined as susceptible or resistant toinfection. Monoclonal antibodies derivedfrom immune mice were used to defineand purify the major immunogens of thesmall intestinal nematode, T. spiralis. Thiswork led to the purification to

homogeneity of the first single, protectiveantigen of a nematode, a 48 kDa protein(Silberstein and Despommier, 1984).Initially, three proteins were isolated fromextracts of infective larvae by immuno-affinity chromatography (Silberstein andDespommier, 1984). When used to im-munize mice, one of these proteins gavesignificant protection at very low (0.1D1 i.tg) doses. Unfortunately, this proteinwas ineffective in the natural pig host(Gamble, Murrell and Marti, 1986).

In another study, sera from infectedsheep, which were defined as resistant orsusceptible to O. circumcincta, were usedto identify a 31 kDa molecule present inTriton X-100 extracts of L3. This moleculewas recognized preferentially by resistantanimals as early as three weeks afterexperimental infection and was presentin and secreted by third-stage larvae(McGillivery et al., 1990). Lambs im-munized with the purified protein weresignificantly protected against homologouschallenge as judged by faecal egg countsand adult worm burdens compared withchallenge controls (McGillivery et al., 1992).Homologous antigens have alsobeen identified in T. colubrtformis andH. con tortus (McGillivery et al., 1990).

Recently, more rational approaches tovaccine design have produced encouragingresults. One approach has been to identifyand isolate antigens from the intestinalluminal surface of blood-feeding parasites.It has been suggested that the induction ofsystemic antibody responses against thegut antigens of blood-sucking parasitesmay cause sufficient damage to impairparasite survival. This response is"artificial" as these gut antigens are notrecognized by the host during the courseof natural infection, i.e. the antigens arecovert. Lambs have been immunizedsuccessfully against haemonchosis, usingantigens fractionated from the intestinalluminal surface of adult H. contortus

(Munn, Greenwood and Coadwell, 1987;Munn and Smith, 1990). A major im-munogen of this preparation was anintegral membrane glycoprotein of mole-cular weight 110 kDa, termed H11 (Munn,1988; Smith and Munn, 1990). Immuniz-ation of sheep with microgram amounts ofessentially pure H11 stimulated substantialprotection against challenge infection inseveral breeds of sheep (Munn, 1988;Tavernor et al., 1992a and 1992b; Munnet al., 1993a). This immunity was closelycorrelated with H11-specific circulatingIgG levels (Munn and Smith, 1990) andvaccination was equally successful inyoung lambs (< two months, Tavernoret al., 1992a) as in older lambs (> eightmonths, Tavernor et al., 1992b). Withrespect to the age-related unresponsive-ness of young lambs to vaccination withirradiated L3 (Urquhart et al., 1966b), theresults obtained with the H11 antigenwould suggest that this would be an idealvaccine candidate for the control of thisdisease. Sequence analysis of full-lengthcDNAs encoding H11 has shown theprotein to be homologous to mammalianamino peptidases. Amino peptidase acti-vity was subsequently found in native pureH11 and enzyme activity was inhibited byimmunoglobulin from H11-vaccinatedsheep (Munn et al., 1993b). A full-length,enzymically active H11 cDNA has beenproduced in the baculovirus expressionsystem (Munn et al., 1993b) and its abilityto protect lambs against haemonchosis iscurrently being evaluated. Recent work hasindicated that integral gut membranepreparations from adult H. contortuscontain other protective antigens apartfrom H11. For instance, substantial pro-tection against H. con tortus was induced inlambs hyperimmunized with an integralmembrane extract in which the character-istic H11 doublet was not usually observed(Smith, 1993).

A similar approach has led to the

Vaccine manual 97


isolation of several protective antigensfrom blood-feeding arthropods. Forexample, cattle are protected against thetick B. microplus by using componentsisolated from the tick gut (Johnston, Kempand Pearson, 1986; Kemp et al., 1986;Willadsen, McKenna and Riding, 1988;Opdebeeck et al., 1988). The number of ticksengorging, their average weight and egg-laying ability were all diminished in thoseticks feeding from immunized cattle(Willadsen and McKenna, 1991). TritonX-100 extracts of the gut of first-instarlarvae of L. cuprina were used to vaccinatesheep, and challenge larvae obtained fromthese sheep had their growth reduced bymore than 40 percent when compared withlarvae from challenge controls. In addition,immune sera from vaccinated sheepinhibited the growth of larvae in vitro bymore than 50 percent indicating that theeffect was antibody-mediated (East et al.,1993). Gut membrane preparations havealso given promising results with suckinglice, flies and mosquitoes (Schlein andLewis, 1976).

The lumen of the blowfly gut is linedwith a periotrophic membrane which, inaddition to being essential to digestion,acts as a molecular sieve preventing intactantibody molecules coming into contactwith underlying gut epithelial cells (Eastet al., 1993). This range of functions, crucialto insect survival, attracted attention to theperiotrophic membrane as a source ofpotential protective antigens. Periotrophicmembranes, harvested in vitro fromL. cuprina larvae, were used to immunizesheep and resulted in a 30 percentreduction in the weight of establishedlarvae compared with those obtained fromchallenge control animals (East et al., 1993).

Another source of potential vaccinecandidates are ES products which arethought to assist in tissue invasion, parasitefeeding and evasion of host effectormechanisms. ES products may be derived

from the nematode cuticle or fromspecialized excretory-secretory organs andrepresent the major antigenic andfunctional challenge to the host (MaizeIsand Selkirk, 1988; Lightowlers and Rickard,1988). In terms of so-called functionalmolecules, several classes of enzyme havebeen identified in ES products. Theseinclude proteinases which may facilitatepenetration of host tissues (Matthews,1982; Dalton and Heffernan, 1988) or act asanticoagulants (Hotez and Cerami, 1983),acetylcholinesterases (AChE) (Rhoads,1984) and superoxide dismutases (Rhoads,1983; Knox and Jones, 1992). Some of thesehave the potential to modulate hostimmune responses, for example secretedproteinases have been ascribed a role incleaving surface-bound immunoglobulinin several helminth systems (Auriault etal., 1981; Chapman and Mitchell, 1982;Smith et al., 1993). AChE may interfere withhosts' immune mechanisms by breakingdown host acetylcholine (ACh), which isknown to stimulate potential effectormechanisms such as neutrophil-mediatedantibody-dependent cellular cytotoxicity(Gale and Zhigelboim., 1974), neutrophilchemotaxis (Hill et al., 1975) and mast cellhistamine release (Kaliner and Austen,1975). Although not an enzyme, a y-interferon analogue has been identified inT. colubriformis ES products (Dopheide etal., 1991). Owing to these putativefunctional roles, ES products have receivedclose attention as a source of protectiveantigens.

Secretions produced during the moultfrom third- to fourth-stage larvae of Ascarissuum induced significant protectionagainst challenge in guinea pigs, whilesoluble proteins produced during cultureof L2, L3, L4 and adult worms wereineffective (Stromberg and Soulsby, 1977).Metabolites secreted by H. con tortus duringthe moult from the third to fourth larvalstages were also effective in sheep (Ozerol

and Silverman, 1970) and the ES productsof adult D. viviparus induced significantlevels of protection in immunized guineapigs (McKeand et al., 1995).

Attention is now focusing on definingthe precise nature of the polypeptideswhich comprise nematode ES products.The application of molecular biology andincreasingly specific immunologicalmethods to the study of parasite ESproducts has enabled the definition of anumber of novel proteins with protectiveproperties. Recently, the dominantglycoprotein present in ES productsderived from the parasitic stages ofT. colubriformis has been cloned andsequenced and shown to induce protectiveimmunity and have significant homologyto the porcine intestinal peptide, valosin(Savin et al., 1990). Valosin can modulateseveral aspects of gastrointestinal functionand, by secreting a homologue protein, theparasite itself may alter its localenvironment.

AChE is secreted by several nematodespecies and is immunogenic in a numberof host/ parasite systems (Lee, 1970;Edwards, Burt and Ogilvie, 1971; Jones andOgilvie, 1972; Ogilvie et al., 1973; McKeandet al., in press). For example, sheep infectedwith T. colubriformis produce antibodies toworm AChE; however, AChE purifiedfrom this parasite failed to induce aprotective immune response in vaccinatedguinea pigs (Rothwell and Merritt, 1975).In the D. viviparus system, calves naturallyor experimentally infected with thisparasite also produce antibodies whichbind and inhibit parasite AChE activity(McKeand et al., in press) and D. viviparusES preparations enriched for AChE activityprotected guinea pigs against subsequentchallenge compared with adjuvant controls(McKeand, 1992).

Glutathione S-transferase (GST) appearsto be one of the major detoxificationenzymes of parasitic helminths and has

received much attention as a candidateprotective antigen (Brophy and Barrett,1990). GST is recognized by antibodiesfrom mice infected with the trematodeSchistosoma japonicum, and the devel-opment of protective immunity in variousmice strains has been positively correlatedwith the levels of specific anti-GSTantibody (Smith et al., 1986). RecombinantGST, prepared from S. japonicum and S.mansoni, was subsequently observed toconfer partial, but not significant, pro-tection in rats and hamsters (Balloul et al.,1987). However, vaccination of geneticallysusceptible mice with a j3-galactosidasefusion protein of this enzyme gave signi-ficant protection against infection with S.japonicum (Smith et al., 1986). Affinity-purified GST from Fasciola hepatica did notsignificantly protect rats against challenge(Howell, Board and Boray, 1988), but sheepimmunized with purified GST weresignificantly protected (57 percent re-duction in fluke numbers) (Sexton, Milnerand Paraccio, 1990). The genes encoding F.hepatica GSTs have now been isolated witha view to testing a recombinant vaccine(Wijffels, Sexton and Salvatore, 1992). Incontrast to these trematode systems,antisera raised against GST purified fromadult H. con tortus, while inhibiting enzymeactivity in vitro, failed to affect survival ofthe parasite in vivo (Sharp et al., 1991).

Proteinases are secreted by L. cuprinalarvae on to the sheep's skin to enable thelarvae to feed and it is these enzymes thatare thought to initiate wound formation. Ithas been observed that proteinases re-leased by these flies can be inhibited bysheep plasma proteinase inhibitorsand that the proteinase inhibitorsu2-macroglobulin and antithrombin IIIreduced larval growth in vitro (Bowles,Feehan and Sandeman, 1990). Further-more, antiproteinase antibodies inhibitedlarval growth in vitro (Sandeman, 1990). Incommon with many helminth parasites,

Vaccine manual 99


blowfly larvae produce a multiplicity ofsecreted proteinases, each of which mayneed to be inhibited by the host immuneresponse to impair larval growthadequately (East and Eisemann, 1993).

Surface antigens of parasitic nematodesare often highly antigenic (Maizels andSelkirk, 1988). Cetyltrimethylammoniumbromide (CTAB) surface extracts ofT. spiralis larvae induced significant levelsof protection against reinfection in mice(Grencis et al., 1986). However, sheepimmunized with surface extracts ofH. contortus L3 were as susceptible aschallenge control animals (Turnbull et al.,1992) and cuticle collagens from the thirdand fourth stages of the same parasitefailed to induce significant protectionagainst homologous challenge (Boisvenueet al., 1991).

MULTICELLULAR PARASITE VACCINES:

RECENT DEVELOPMENTS

The isolation of protective antigens insufficient quantities for practical use invaccine development has, until recently,been restricted by a lack of available nativeparasite material. The development ofrecombinant DNA techniques for the invitro expression of foreign genes ineukaryotic and prokaryotic cells hasprovided an alternative strategy which hasrecently stimulated extensive research intoantiparasite vaccines. While there are nowmany examples of expression systems forparasite proteins, there are only a fewreports of recombinant parasite proteinswhich induce significant levels of pro-tective immunity. The steps in recombinantvaccine development are illustrated by thefollowing successful examples.

Immunization of sheep using T. colubri-formis tropomyosin recombinants. Asdiscussed earlier, the T. colubriformismuscle protein, tropomyosin, was shownto be a potentially useful protective antigen

in the guinea pig when given in nativeform (O'Donnell et al., 1989). Usingoligonucleotide probes based on partialamino acid sequence data of tropomyosin,the gene encoding this protein was isolatedfrom a cDNA expression library preparedfrom fourth larval-stage mRNA (Cobon etal., 1989). A 27 kDa subunit ofT. colubriformis tropomyosin, expressed asa P-galactosidase fusion protein inEscherichia coli, produced accelerated wormexpulsion following challenge in guineapigs. When the T. colubriformis tropom-yosin DNA was used as a hybridizationprobe, the gene encoding a related antigenwas isolated from mRNA prepared fromadult H. con tortus. The expressed geneproduct significantly protected immunizedsheep against H. contortus challenge(Cobon et al., 1989).

Immunization of sheep with recombinantsof T. ovis onchosphere ES antigens. Asimilar approach has shown promise inexperiments with Cysticercus avis, theintermediate stage of T. avis, in sheep.Previous experiments indicated that theES products of hatched and activatedT. avis onchospheres contained potenthost-protective antigens. Several of the EScomponents, ranging from 47 to 52 kDa insize, were found to be recognized stronglyby sera from resistant sheep (Rickard andBell, 1971a and 1971b; Rickard and Adolph,1977) and lambs immunized with apolyacrylamide gel-purified preparation ofthese components were significantly (98percent) protected against challenge(Johnson et al., 1989). Rabbit antibodies,specific for the 47 to 52 kDa region, wereeluted from Western blots and used toprobe a cDNA expression library preparedfrom rnRNA extracted from hatched andactivated onchospheres (Johnson et al.,1989). P-galactosidase fusion proteins,prepared from selected immunopositives,were found to be antigenic in sheep but

were not protective. Subsequently, one ofthe proteins was prepared as a fusion withS. japonicum glutathione S-transferase andwas found to protect sheep significantlyagainst challenge (Johnson et al., 1989).

Recombinant proteins have also beenassessed in other Taenia species. Threefusion proteins, derived from T. taeniae-formis onchosphere cDNA and expressedin a pGEX plasmid vector, gave a 95percent reduction in total metacestoderecoveries in Wistar rats (Ito et al., 1991)

Immunization of ruminants using re-combinant forms of membrane gut proteins0f B. microplus. As discussed previously,in infections in which immunity developsslowly or poorly, artificial immunizationwith hidden, or covert, antigens may beused to circumvent a lack of responsive-ness. Recent developments using re-combinant gut antigens from the blood-sucking tick B. microplus in cattle have beenpromising.

Cattle acquire limited levels of resistancefollowing prolonged natural exposure toB. microplus (Wagland, 1975). Protectiveimmunity has been achieved using anti-gens fractionated from gut membranepreparations of engorged female ticks(Willadsen, McKenna and Riding, 1988),and a membrane-bound glycoprotein(Bm86) expressed on the surface of tickgut digestive cells was shown to be a highlyeffective immunogen (Willadsen et al.,1989). The gene coding for this antigenwas subsequently isolated from a cDNAexpression library prepared from adultB. micro plus and a fusion comprising 599amino acids of Bm86 and 651 amino acidsof P-galactosidase was expressed in E. colias inclusion bodies (Rand et al., 1989). Ticksengorging on cattle vaccinated with theseinclusion bodies were significantlydamaged as a result of the immuneresponse to the cloned antigen (Rand et al.,1989).

The above examples have attractedconsiderable commercial support and arecurrently undergoing development.

PROBLEMS OF SUBUNIT VACCINE

DEVELOPMENT

Laboratory models versus natural host/parasite systems. Potentially useful anti-gens are often selected on the basis ofprotection trials conducted in laboratoryanimal models. In general terms, lab-oratory animals expel the parasite morerapidly than does the natural host andinfection is often terminated prior topatency. Protection observed in modelsystems is therefore often only an accelera-tion of an already efficient antiparasiteresponse and does not necessarily reflectutility in the natural host. Caution musttherefore be taken when extrapolatingfrom the laboratory model to the definitivehost.

Recombinants expressed in the wrongconformation. The protective properties ofa recombinant protein will be influencedby the tertiary structure which can, in turn,be altered depending on the fusion partner.It also depends on whether or not thenative protein is glycosylated. This caution-ary note is exemplified by work on theprotective onchosphere antigens of T. ovisin which protective immunity was onlyattained after expression of the recombin-ant molecule as a GST fusion protein(Johnson et al., 1989).

Solubilization of the relevant antigens. Thesolvent used to extract parasite proteinscan be crucial to the isolation of theappropriate protective antigens. Forexample, work performed on the cestodeTaenia pisiformis indicated that sodiumdeoxycholate, but not PBS, extracts madefrom onchospheres of the parasite confer-red high levels of protection to rabbits(Rajasekariah, Rickard and O'Donnell,

Vaccine manual 101

1985). Similarly, DOC-PAGE, but not SDS-PAGE, gel cut-outs of T. taeniaeformisonchosphere antigens protected miceagainst challenge (Lightowlers, Rickardand Mitchell, 1984). In comparison, sheephave been successfully immunized againstchallenge with Echinococcus gran ulosususing onchosphere antigens which hadbeen solubilized in SDS (Dempster et al.,1992) and against T. ovis using homogen-ized SDS-polyacrylamide gels containinghomologous antigen (Harrison et al., 1993).The variation produced by the differentextraction buffers may be caused by thedisruption of conformational epitopes orby the lack of solubilization of effectivecomponents.

Presence of non-peptide epitopes. B cellepitopes are often conformational and maynot be peptide in nature. These epitopeswill be difficult to reproduce usingconventional molecular biology or peptidesynthesis techniques. As the antigenbinding site of an antibody molecule iscomplementary in physical structure to theantigen against which it is raised, theantigen binding site can, itself, be used asantigen to raise anti-idiotype antibodieswhich mimic the shape of the originalantigen. For example, an anti-idiotypevaccine which mimics the glycan com-ponent of a 38 kDa surface component ofS. mansoni stimulated protective immunityin rats against homologous challenge(Grzych et al., 1985). This approach is verylaborious, however, and would probablyonly be considered if the protective epitopewas non-peptide or demonstrably confor-mational and could not be produced bythe variety of recombinant DNA ap-proaches now available. In addition tocontaining carbohydrate epitopes, glyco-sylation is also likely to affect the protectiveproperties of proteins by modifying proteinfolding, so that recombinants will be moreappropriately expressed in systems which

glycosylate the recombinant protein suchas mammalian cell lines and viral vectors.

Accessibility of antigen to the immunesystem. In order to have an effect on aninvading parasite, host effector mechan-isms must be directed against antigenswhich are accessible to this response. Forexample, Haemonchus spp. and Boophilusspp. ingest host blood so that antigens onthe gut surface are bathed in bloodcomponents such as antibody. The extentto which this approach can be applied tonon-bloodsucking parasites such asOstertagia spp. and Trichostrongylus spp.remains to be defined. Furthermore,because covert antigens will not be exposedduring subsequent natural infection it ispresumed that booster vaccination wouldbe required until natural immunitydevelops or live vaccine vectors may beused to enable persistence of the vaccinecomponents within the host.

Genetic variation in the vaccinatedpopulation. Once a vaccine has beendeveloped, it is essential that it is effectivein most, if not all, of the target population.The major histocompatibility complex(MHC) and other, as yet undefined,background genes within a populationinfluence the capability of individuals torespond immunologically to specificantigens. This in turn will establish thesuccess or failure of any vaccine (Kennedy,1990). For example, a variable responsive-ness was encountered when syntheticpeptides of the circumsporozoite proteinof Plasmodium falciparum were used tovaccinate mice and it was observed thatboth antibody and T cell responses to thisantigen were H-2b-restricted (Del Giudiceet al., 1986; Togna et al., 1986). Thus, it canbe anticipated that subunit vaccines whichstimulate elements of the natural immuneresponse to the parasite will have tocomprise a variety of peptides to maintain


long-term field utility. Furthermore, theinduction of natural immune responseswill be profoundly influenced by the waythe antigen(s) is / are presented to theimmune system.

Assisting antigen presentationThe rapid developments in recombinantDNA technology and protein chemistryshould provide antigens in quantity forexperimental immunization studies. Indi-vidual molecules are likely to have reducedimmunogenicity and will have to bepresented to the host immune system inthe context of an adjuvant or a live vaccinevector. Moreover, the method of antigenpresentation can be optimized if therelevant effector arms of the immuneresponse are defined.

AdjuvantsAdjuvants non-specifically stimulateimmune responses to antigen. The under-standing of how adjuvants work hasincreased rapidly and many of the adju-vants now available are recognized tostimulate different types of responses (seechapter by Bomford, Adjuvants in vet-erinary vaccines, p. 277, and Bomford,1989).

Adjuvant selection, and hence vaccineefficacy, can be improved if the preciseimmunological mechanisms involved inparasite killing or expulsion are defined.For example, al-uminium hydroxide pre-ferentially stimulates T helper type 2 (Th2)lymphocytes so that a humoral response,particularly IgG, and IgE, is favoured(Smith, 1992). In contrast, Freund's com-plete adjuvant, which contains bacterialcomponents, is a strong promoter ofcellular responses. Freund's incompleteadjuvant, which lacks the mycobacterialcomponent, stimulates only a humoralresponse. Unfortunately, Freund's adju-vants have several adverse side-effects andare not acceptable for use in veterinary

vaccines. However, because of theirgeneral efficacy, they continue to be usedin experimental trials to establish antigenutility, after which less toxic adjuvantmixes may be attempted.

A novel adjuvant which is now licensedfor use in domestic animals is thatbased on imrnunostimulating complexes(ISCOMs). These complexes are mostreadily formed with antigens whichpossess a hydrophobic transrnembraneregion (Bomford, 1989), suggesting thatthey may be particularly useful for theformulation of helminth vaccines based onintegral membrane proteins. So far, therehas been little work detailing the potentialof helminth or ectoparasite subunit vac-cines in the context of any novel adjuvantpreparations.

Vectors and antigen deliveryThe chronicity associated with manyparasitic infections suggests that antigenmay need to be administered over aprolonged period to stimulate protectiveimmunity. Antigens administered orallywith an avirulent virus or bacterial vectorwill have the potential to stimulate avigorous cell-mediated and persistentimmune response (Murray, 1989). Thesevectors also have the potential to stimulateIgA precursor B cells in gut-associatedlymphoid tissue, a possibility for over-coming immune unresponsiveness at thissite. Virus vectors, such as vaccinia andherpesviruses, have the advantage that theantigen(s) are processed in their nativeform (Murray, 1989). So far, vaccinia hasbeen used successfully as a vector for acloned 28 kDa surface antigen of S. mansoni(Simpson and Cioli, 1987). Enterobacteria,such as E. coli and Salmonella spp., havebeen used as vaccine vectors in non-parasite systems, although use of the lattermay lead to public concern.

Vectors may be designed so that theantigens they encode are expressed in

Vaccine manual 103

combination with the appropriaterecombinant cytokine(s) in order tostimulate the desired immune effector arm.For gastrointestinal nematodes, antigenand cytokine genes could be expressed inparallel in a bacterial or viral vectorsuitable for oral vaccine delivery.

Bacterial toxins also have considerablepotential for the oral presentation ofantigen. For example, cholera toxin has ahigh affinity for specific receptors on thesurface of the intestinal epithelium andstimulates both secretory IgA and serumIgG antibody responses with prolongedimmunological memory (McGhee et al.,1992). These responses would be appro-priate in vaccination against gastrointe-stinal nematodes, although the problemsof toxicity may have to be overcome.

CONCLUSIONS

The optimism express.ed when recornbin-ant DNA technology was first developedperhaps overestimated the capabilities ofthese techniques and failed to recognizethe complex nature of the host-parasiteinteraction and the difficulty in maintain-ing immunogenicity. A major problem hasbeen the inability to identify the relevantantigens from complex preparations in theface of multiple immune responses, not allof which are protective. These responsesstill have to be unravelled for all multi-cellular parasitic infections. A substantialamount of parasitological research is nowdirected at defining the precise immuneresponses relevant to the elimination, orotherwise, of parasites. Although beyondthe scope of this vaccine review, it appearsthat, of the many immune effector mecha-nisms involved in the host-parasite inter-action, only some are relevant to thedevelopment of protective immunity.Nearly all the studies described in thisreview selected antigens on the basis ofhumoral immunity, and there has beenlittle detail given regarding the essential

cellular arm of the immune response. Interms of vaccine development, this reflectsthe route that research has taken in thepast. More recently, detailed analyses ofthe cellular responses against helminthshave identified responses which areprotective or those which may be immuno-pathological or immunorepressive (Smith,1992). By measuring mitogen- or antigen-specific in vitro propagation of immunecell lines, including T cell subsets, ormeasuring the types of cytokines releasedor expresssed by these cells, cellularresponses can be dissected for each parasiteinfection. Parasite antigens also triggermast cell degranulation with the associatedrelease of mast cell proteinases, histamineand various cytokines. In several cases,these responses appear to be central toworm expulsion, and assays which detectantigen-specific mast cell proteinaserelease may be used to identify whichantigens stimlate these cells (Jones,Huntley and Emery, 1992). Thus, it can beseen that basic parasite immunologyshould go hand in hand with vaccineresearch in order to define which kind ofresponses should be aimed for whenidentifying vaccine candidates.

As outlined above, delivery systems andadjuvants can now be selected on the basisthat they can optimize or augment therequired effector arm of the immunesystem. Nevertheless, by obtaining in-formation on putative functions for specificparasite components or realizing theircontribution to parasite survival, rationalsubunit vaccines may be designed withoutthe need for a full understanding of thehost parasite interaction.

Pioneering advances have been made inthe last decade, most notably in thedefinition of candidate antigens forprotection against T. ovis and H. con tortusin sheep, but the identification of protectiveantigens and evidence for their efficacy inimmunization trials is only the start of the


process. Field trials and the incorporationof the antigens into a commercially viablevaccine are expensive and time-consumingand will rely heavily on commitment fromlarge pharmaceutical companies whichmust accept the vaccine as a potentialmarketable commodity. It should be notedthat a vaccine which only reduces nema-tode egg output might reduce pasturecontamination to a level where productionlosses are balanced by stock managementsavings. Thus, at least with veterinaryvaccines, it may not be imperative toinduce sterile immunity completely and,when considering the most appropriatevaccine development strategy, full accountmust be taken of the relative prevalenceand epidemiology of the parasite.

BIBLIOGRAPHY

Adams, D.B. 1989. A preliminary evaluationof factors affecting an experimentalsystem for vaccination-and-challengewith Haemonchus contortus in sheep. Int.J. Parasitol., 19: 169-175.

Arundel, J.H. & Sutherland, A.K. 1988.Blowflies of sheep. In Animal Health inAustralia, 10: 35-60. Canberra, AustralianGovernment Publishing Service.

Auriault, C., Ouissi, M.A., Torpier, G., Elsen,H. & Capron, A. 1981. Proteolyticcleavage of IgG bound to the Fc receptorof Schistosoma rnansoni schistosomula.Parasite Immunol., 3: 33-44.

Balloul, J.M., Grzych, J.M., Pierce, R.J. &Capron, A. 1987. A purified 28 000 daltonprotein from Schistosoma mansoni adultworms protects rats and mice againstexperimental schistosomiasis. J.

Immunol., 138: 3448-3453.Boisvenue, R.J., Stiff, M.I., Tonkinson, L.V.

& Cox, G.N. 1991. Protective studiesin sheep immunized with cuticularcollagen proteins and peptides ofHaemonchus contortus. Parasite Immunol.,13: 227-240.

omford, R. 1989. Adjuvants for antiparasitevaccines. Parasitol. Today, 5: 41-46.

Bowles, V.M., Feehan, J.P. & Sandeman,R.M. 1990. Sheep plasma proteinaseinhibitors influencing activity andgrowth of Lucilia cuprina larvae in vitro.Int. J. Parasitol., 20: 169-174.

Brophy, P.M. & Barrett, J. 1990. Glutathionetransferase in helminths. Parasitol., 100:345-349.

Brossard, M. 1976. Relations immunologiquesentre bovins et tiques: Boophilus micro plus.Acta Trop., 33: 15-36.

Bushara, H.O., Hussein, M.F., Saad, A.M.,Taylor, M.G., Dargie, J.D., De. C.Marshall, T.F. & Nelson, G.S. 1978.Immunization of calves againstSchistosoma bovis using irradiatedcercariae or schistosomula of S. bovis.Parasitol., 77: 303-311.

Chandler, A.C. 1932. Experiments onresistance of rats to superinfection withthe nematode Nippostrongylus muris.Am. J. Hyg., 16: 780-782.

Chapman, C.B. & Mitchell, G.F. 1982.Proteolytic cleavage of immunoglobulinby enzymes released by Fasciola hepatica.Vet. Parasitol., 11: 165-178.

Clegg, J.A. & Smith, M.A. 1978. Prospectsfor the development of dead vaccinesagainst helminths. Adv. Parasitol., 16:165-217.

Cobon, G.S., Austen, A., O'Donnell, I.J.,Frenkel, M.J., Kennedy, W.P., Savin,K.W. & Wagland, 7.M. 1989. Vaccinesagainst animal parasitic nematodes.Patent: W089 / 00163.

Dalton, J.P. & Heffernan, M. 1988. Thiolproteases released in vitro by Fasciolahepatica. Mol. Biochem. Parasitol., 35: 161-166.

Del Giudice, G., Cooper, J.A., Merino, J.,Verdini, A.S., Pessi, A., Togna, A.R.,Engers, H.D., Corradin, G. & Lambert,P.-H. 1986. The antibody response inmice to carrier-free synthetic polymersof Plasmodium falciparum circum-

Vaccine manual 105

sporozoite repetitive epitope is I-Ab-restricted: possible implications formalaria vaccines. J. Immunol., 137:2952-2955.

Dempster, R.P., Harrison, G.B.L., Berridge,M.V. & Heath, D.D. 1992. Echinococcusgranulosus: use of an intermediate hostmouse model to evaluate sources ofprotective antigens and a role forantibody in the immune response. Int.J. Parasitol., 22: 435-441.

Dopheide, T.A.A., Tachedjian, M., Phillips,C., Frenkel, M.J., Wagland, B.M. 8:Ward, C.W. 1991. Molecularcharacterisation of a protective, 11-kDaexcretory-secretory protein from theparasitic stages of Trichostrongyluscolubriformis. Mol. Biochem, Parasitol., 45:101-108.

East, I.J. & Eisemann, C.H. 1993. Vaccinationagainst Lucilia cuprina: the causativeagent of sheep blowfly strike. Immunol.Cell Biol., 71: 453-462.

East, I.J., Berrie, D.A. & Fitzgerald, C.J.1988. Oesophagostomum radiatum:successful vaccination of calves withan extract of in vitro cultured larvae.Int. J. Parasitol., 18: 125-127.

East, I.J., Berrie, D.A. & Fitzgerald, C.J.1989. Oesophagostomum radiatum:successful vaccination of calves withhigh molecular weight antigens. Int. J.Parasitol., 19: 271-274.

East, I.J., Fitzgerald, C.J., Pearson, R.D.,Donaldson, R.A., Vtiocolo, T., Caclogan,L.C., Tellam, E isernartn,1993. Lucilia cuprina: inhibition of larvalgrowth induced by immunization of hostsheep with extracts of larval periotrophicmembrane. Int. J. Parasitol., 23: 221-229.

Edwards, A.J., Burt, J.S. & Ogilvie, 31971. The effect of immunity upon someenzymes of the parasitic nematode,Nippostrongylus brasiliensis. Parasitol., 62:339-347.

Egerton, J.S., Suhayda, D. & Eary, C.H.1988. Laboratory selection of Haemonchus

contortus for resistance to ivermectin.Parasitol., 74: 614-617.

Emery, D.L. & Wagland, B.M. 1991. Vaccinesagainst gastrointestinal parasites ofruminants. Paras itol. Today, 7: 347-349.

Gale, R.P. & Zighelboim, J. 1974. Modulationof polymorphonuclear leukocyte-mediated antibody-dependent cellularcytotoxicity. J. Immunol., 113: 1793-1800.

Gamble, J.R., Murrell, K.D. & Marti, H.P.1986. Inoculation of pigs againstTrichinella spiralis, using larval excretory-secretory antigens. Am. J. Vet, Res., 47:2396-2399.

Gregg, P. & Dineen, H.P. 1978. The responseof sheep vaccinated with irradiatedTrichostrongylus colubriformis larvae. Vet.Parasitol., 4: 49-53.

Gregg, P., Dineen, J.K., Rothwell, T.L.W.& Kelly, J.D. 1978. The effect of age onthe response of sheep to vaccinationwith irradiated Trichostrongyluscolubriformis larvae. Vet. Parasitol., 4:35-48.

Grencis, R.K., Crawford, C., Pritchard, D.I.,Behnke, J.M. & Wakelin, D. 1986.Immunization of mice with surfaceantigens from the muscle larvae ofTrichinella spiralis. Parasite Immunol., 8:587-596.

Gryzch, J.M., Capron, M., Lambert, P.H.,Dissous, C., Torres, S. & Capron, A.1985. An anti-idiotype vaccine againstexperimental schistosomiasis. Natu re,316: 74-76.

Harrison, G.B.L., Heath, D.D., Dempster,R.P., Lawrence, S.B., Lightowlers, M.W.

Riclr.ard, M. 1993. Identification ofhost-protective antigens of Taenia avisoncospheres. Int. J. Parasitol., 23: 41-50.

Hill, H.R., Estensen, R.D., Quie, P.G., Hogan,N.A. & Goldberg, N.D. 1975. Modulationof human neutrophil chemotactic re-sponses by cyclic 3', 5'-guanosine mono-phosphate and cyclic 3', 5'-adenosineatonophosphate. Metabolism, 24: 447-456.

Hotez, P. & Cerami, A. 1983. Secretion of a


proteolytic anticoagulant by Ancylostomahookworms. J. Exp. Med., 157: 1594-1603.

Howell, M.J., Board, P.G. & Boray, J.C.1988. Glutathione S-transferases. J.Parasitol., 74: 715-718.

Ito, A., Bogh, H.O., Lightowlers, M.W.,Mitchell, G.F., Takami, T., Kamiya, M.,Onitake, K. & Rickard, M.D. 1991.Vaccination against Taenia taeniaeformisinfection in rats using a recombinantprotein and preliminary analysis of theinduced antibody response. Mol. Biochem.Parasitol., 44: 43-52.

Jackson, F. 1993. Anthelmintic resistance -the state of play. Br. Vet. J., 149: 123-138.

Jackson, F., Coop, R.L., Jackson, E., Little.,S. & Russell, A.J.F. 1991. Anthelmintic-resistant nematodes in goats. Vet. Rec.,129: 39-40.

Jarrett, W.F.H., M cIntyre, W.I.M. &Urquhart, G.M. 1954. Husk in cattle. Areview of a year's work. Vet. Rec., 66:665-676.

Jarrett, W.F.H., Jennings, F.W., Martin, B.,McIntyre, W.I.M., Mulligan, W., Sharp,N.C.C. & Urquhart, G.M. 1958. A fieldtrial of a parasitic bronchitis vaccine.Vet. Rec., 70: 451-454.

Johnson, K.S., Harrison, G.B.L., Lightowlers,M.W., O'Hoy, K.L., Cougle, W.G.,Dempster, W.G., Lawrence, R.P.,Vinton, J.G., Heath, D.D. & Rickard,M.D. 1989. Vaccination against ovinecysticercosis using a defined recombinantantigen. Nature, 338: 585-587.

Johnston, L.A.Y., Kemp, D.H. & Pearson,R.D. 1986. Immunisation of cattle againstBoophilus micro plus using extracts derivedfrom adult female ticks. Int. J. Parasitol.,16: 27-35.

Jones, V.E. & Ogilvie, f.M. 1972. Protectiveimmunity to Nippostrongylus brasiliensisin the rat. III. Modulation of wormacetylcholinesterase by antibodies.Immunology, 22: 119-129.

Jones, W.O., Huntley, J.F. & Emery, D.L.1992. Isolation and degranulation of

mucosal mast cells from the smallintestine of parasitised sheep. Int. J.

Parasitol., 22: 519-521.Kanner, M. & Austen, F.K. 1975. Immunol-

ogic release of chemical mediators fromhuman tissues. Annu. Rev. Pharmacol.,15: 177-189.

Keith, R.K. & Bremner, K.C. 1973. Im-munisation of calves against the nodularworm Oesophagostomum radia tum. Res.Vet. Sci., 15: 23-24.

Kemp, D.H., Agbebe, R.I.S., Johnston, L.A.Y.& Gough, J.M. 1986. Immunisation ofcattle against Boophilus micro plus usingextracts derived from adult female ticks:feeding and survival of the parasite onvaccinated cattle. Int. J. Parasitol., 16:115-120.

Kennedy, M.W. 1990. Resistance to parasiticnematodes - how is the MHC involved?Parasitol. Today, 6: 374-375.

7:I-Lox, D.P. & Jones, D.G. 1992. A comparisonof superoxide dismutase (SOD,EC:1.15.1.1) distribution in gastro-intestinal nematodes. In t. J. Parasitol.,22: 209-214.

Lee, D.L. 1970. The fine structure of theexcretory system in adult Nippostrongylusbrasiliensis (Nernatoda) and a suggestedfunction for the "excretory glands".Tissue Cell, 2: 225-231.

Lightowlers, M.W. & Rickard, M.D. 1988.Excretory-secretory products of helminthparasites: effects on host immuneresponses. Parasitol., 6: S123-S166.

Lightowlers, M.W., Rickard, M.D. &Mitchell, G.F. 1984. Immunizationagainst Taenia taeniaeformis in mice:identification of onchospheral antigenson polyacrylamide gels by Westernblotting and enzyme immunoassay. Int.J. Parasitol., 4: 297-306.

Maizels, R.M. & Selkirk, M.E. 1988. Immuno-biology of nematode antigens. In P.T.Englund & A. Sher, eds. I: Biology ofparasitism, p. 285-308. New York, A.R.Liss.

Vaccine manual 107

Matthews, B.E. 1982. Skin penetration byNecator americanus larvae. Z. Parasitenk.,68: 81-86.

McGhee, J.R., Mestecky, J., Dertxbaugh,M.T., Eldridge, J.H., Hirasawa, M. &Kiyono, FI. 1992. The mucosal immunesystem: some fundamental concepts tovaccine development. Vaccine, 10: 75-89.

McGillivery, D.J., Yong, W.K., Riffkin, G.G.& Adler, B. 1990. The distribution andlocalisation of the stage-specific GP31antigen from infective Ostertagiacircumcincta larvae. Int. J. Parasitol., 20:87-93.

McGillivery, D.J., Yong, W.K., Adler, B. &Riffkin, G.G. 1992. A purified stage-specific 31 kDa antigen as a potentialprotective antigen against Ostertagiacircumcincta infection in lambs. Vaccine,10: 607-613.

McGowan, M.J., Barker, R.W., Homer, J.T.,McNew, R. & Holscher, K.H. 1981.Success of ticks feeding on calvesimmunised with Ambylomma americanumextract. J. Med. Entomol., 18: 328-332.

McKeand, J.B. 1992. Aspects of theimmunobiology of Dictyocaulus viviparusinfection. University of Glasgow, UK.(Ph.D. thesis)

McKeand, J.B., Knox, D.P., Duncan, J.L. &Kennedy, M.W. 1995. Protectiveimmunisation of guinea pigs againstDictyocaulus viviparus using excretory-secretory products of adult parasites.Int. J. Parasitol., 25(1): 95-104.

McKeand, J. ., Knox, D.P., Duncan, J.L. &Kennedy, M.W. The immunogenicityof the cattle lungworm, Dictyocaulusviviparus. Int. J. Parasitol. (in press)

Miller, T.A. 1971. Vaccination against thecanine hookworm diseases. Adv.Parasitol., 9: 153-183.

Miller, T.A. 1978. Industrial developmentand field use of the canine hookwormvaccine. Adv. Parasitol., 16: 333-342.

Monroy, F.G. & Dobson, C. 1987. Micevaccinated against Nematospiroides dubius

with antigens isolated by affinitychromatography from adult worms.Immunol. Cell Biol., 65: 223-230.

Munn, E.A. 1988. Production and use ofanthelmintic agents and protectiveproteins. Patent: W088 / 00835.

Munn, E.A. & Smith, T.S. 1990. Productionand use of anthelmintic agents andprotective proteins. Patent: W090 / 11086.

Munn, E.A., Greenwood, C.A. & Coadwell,W.J. 1987. Vaccination of young lambsby means of a protein fraction extractedfrom adult Haemonchus contortus.Parasitol., 94: 385-397.

Munn, E.A., Smith, T.S., Graham, M.,Greenwood, C.A., Tavernor, A.S. &Coetzee, G. 1993a. Vaccination of merinolambs against haemonchosis withmembrane-associated proteins from theadult parasite. Parasitol., 106: 63-66.

Munn, E.A., Graham, M., Smith, T.S.,Newton, S., Knox, D.P., Oliver, J. &Smith, F. 1993b. Cloning and iden-tification of H11, a highly protectiveantigen from Haemonchus contortus. Proc.Keystone Symp. Molecular Parasitology,Colorado, USA.

Murray, P.K. 1989. Molecular vaccines againstanimal parasites. Vaccine, 7: 291-300.

Neilson, J.T.M. 1975. Failure to vaccinatelambs against Haemonchus contortus withfunctional metabolic antigens identifiedby immunoelectrophoresis. Int. J.Parasitol., 5: 427-430.

Neilson, J.T.M. & van de Walle, M.J. 1987.Partial protection of lambs againstHaemonchus con tortus by vaccination witha fractionated preparation from theparasite. Vet. Parasitol., 23: 211-221.

O'Donnell, I.J., Green, P.E., Connell, J.A.& Hopkins, P.S. 1980. ImmunoglobulinG antibodies to the antigens of Luciliacuprina in sera of fly-struck sheep. A us t.J. Biol. Sci., 33: 27-34.

O'Donnell, I.J., Green, P.E., Connell, J.A.& Hopkins, P.S. 1981. Immunisationof sheep with larval antigens of Lucilia


cuprina. Aust. J. Biol. Sci., 34: 411-417.O'Donnell, I.J., Dineen, J.K., Rothwell,

T.L.W. & Marshall, R.C. 1985. Attemptsto probe the antigens and protectiveimmunogens of Trichostrongyluscolubriformis in immunoblots with serafrom infected and hyperimrnune sheepand high and low responder guinea pigs.Int. J. Parasitol., 15: 129-136.

O'Donnell, I.J., Dineen, j.K., Wagland, B.M.,Letho, S., Dopheide, T.A.A., Grant, W.N.& Ward, C.W. 1989. Characterizationof the major immunogen in the excretory-secretory products of exsheathedthird-stage larvae of Trichostrongylus cola-briformis. Int. J. Parasitol., 19: 793-802.

Ogilvie, B.M., Rothwell, T.L.W., Bremner,K.C., Schnitzerling, H.J., Nolan, J. &Keith, R.K. 1973. Acetylcholinesterasesecretion by parasitic nematodes. I.Evidence of secretion by a number ofspecies. Int. J. Parasitol., 3: 589-597.

Opdebeeck, J.P., Wong, J.Y.M., Jackson,L.A. & Dob son, C. 1988. Hereford cattleimmunised and protected againstBoophilus micro plus with soluble andmembrane-associated antigens from themidgut of ticks. Parasite Immunol., 10:405-410.

Ozerol, N.H. & Silverman, P.H. 1970. Furthercharacterisation of active metabolitesfrom histotrophic larvae of Haemonchuscontortus cultured in vitro. J. Parasitol.,56: 1199-1205.

Poynter, D.A. & Cauthen, G.E. 1942.Experiments on the life history of thecattle lungworm, Dictyocaulus viviparus.Am. J. Vet. Res., 3: 395-400.

Prichard, R.K. 1990. Anthelmintic resistancein nematodes, recent understanding andfuture directions for research. Int. J.

Parasitol., 20: 515-523.Rajasekariah, G.R., Rickard, M.D. &

O'Donnell, I.J. 1985. Taenia pisiformis:protective immunisation of rabbits withsolubilised oncosheral antigens. Exp.Parasitol., 59: 321-327.

Rand, K.N., Moore, T., Sriskantha, A.,Spring, K., Tellam, R., Willadsen, P.& Cobon, G.S. 1989. Cloning andexpression of a protective antigen fromthe cattle tick Boophilus microplus. Proc.Natl. Acad. Sci., 86: 9657-9661.

Rhoads, M.L. 1983. Trichinella spiralis:identification and purification ofsuperoxide dismutase. Exp. Parasitol.,56: 41-54.

Rhoads, M.L. 1984. Secretory cholinesterasesof nematodes: possible functions in thehost-parasite relationship. Trop. Vet., 2:3-10.

Rickard, M.D. & Adolph, A.J. 1977.Vaccination of lambs against infectionwith Taenia ovis using antigens collectedduring short-term in vitro incubationof activated oncospheres. Parasitol., 75:183-188.

Rickard, M.D. & Bell, K.J. 1971a. Inductionof immunity in lambs to larval cestodeantigens. Nature, 232: 120.

Rickard, M.D. & Bell, K.J. 1971b. Successfulvaccination of lambs against infectionwith Taenia ovis using antigens producedduring in vitro cultivation of the larvalstages. Res. Vet. Sci., 12: 401-402.

Rickard, M.D. & Williams, J.F. 1982.Hydatidosis/ cysticercosis: immunemechanisms and immunisation againstinfection. Adv. Parasitol., 21: 230-296.

Roberts, F.H.S., Elek, P. & Keith, R.K. 1962.Studies of resistance in calves toexperimental infections with the nodularworm Oesophagostomum radiatum. Aust.J. Agric. Res., 13: 551-573.

Rothwell, T.L.W. 1978. Vaccination againstthe nematode Trichostrongylus colubri-formis. III. Some observations on factorsinfluencing immunity to infection invaccinated guinea pigs. Int. J. Parasitol.,8: 33-37.

Rothwell, T.L.W. & Griffiths, D.A. 1977.Comparison of the kinetics of expulsionof Trichostrongylus colubriformis frompreviously uninfected, reinfected and

Vaccine manual 109

110 Mititicellular parasite vaccines

vaccinated guinea pigs. J. Parasitol., 63:761-762.

ReiTawell, T.L.W. & Love, R.J. 1974.Vaccination against the nematode Tricho-strongylus colubriformis. I. Vaccinationof guinea pigs with worm homogenatesand soluble products released duringin vitro maintenance. Int. J. Parasitol., 4:293-299.

Rothwell, T.W.L. & Merritt, G.C. 1975.Vaccination against the nematode Tricho-strongylus colubriformis. II. Attempts toprotect guinea pigs with worm acetyl-cholinesterase. Int. J. Parasitol., 5: 453-460.

Sandeman, R.M. 1990. Prospects for thecontrol of sheep blowfly strike byvaccination. Int. J. Parasitol., 20: 537-541.

Sanes, M.P. & Taliaferro, W.A. 1936. Thelocal points of defence and the passivetransfer of acquired immunity to Nippo-strongylus brasiliensis in rats. J. Inf. Dis.,59: 207-220.

Savin, K.W., Dopheide, T.A.A., Frenkel,M.J., Wagland, B.M., Grant, W.A. &Ward, C.W. 1990. Characterization,cloning and host-protective activity ofa 30-kilodalton glycoprotein secretedby the parasitic stages of Trichostrongyluscolubriformis. Mol. Biochem. Parasitol., 41:167-176.

Schlein, Y. & Lewis, C.T. 1976. Lesions inhaematophagous flies after feeding onrabbits immunised with fly tissues.Physiol. Entomol., 1: 55-59.

Sexton, J.L., Milner, A.R. & Paraccio, M.1990. Glutathione S-transferase. Novelvaccine against Fasciola hepatica infectionin sheep. J. Immunol., 145: 3905-3910.

Sharma, R.L., Bhat, T.K. & Dhar, D.N. 1988.Control of sheep lungworm in India.Paras itol. Today, 4: 33-36.

Sharp, P.J., Smith, D.R.T., Bach, W.,Wagland, B.M. & Cobon, G.S. 1991.Purified glutathione S-transferases fromparasites as candidate protectiveantigens. Int. J. Parasitol., 21: 839-846.

Silberstein, D.S. & Despommier, D.D. 1984.

Antigens from Trichinella spiralis thatinduce a protective response in themouse. J. Immunol., 132: 898-904.

Simpson, A.J.G. & Cioli, D. 1987. Progresstowards a defined vaccine for schisto-somiasis. Paras itol. Today, 3: 24-26.

Simpson, A.J.G., Hackett, F., Walker, T.,De Rossi, R. & Smithers, S.R. 1985.Antibody response against schisto-somulum surface antigens and protectiveimmunity following immunization withhighly irradiated cercariae of Schistosomamansoni. Parasite Immunol., 7: 133-152.

Skelly, P.J. & Howells, A.J. 1987. Thehumoral immune reponse of sheep toantigens from larvae of the sheep blowfly(Lucilia cuprina). Int. J. Parasitol., 17:1081-1087.

Smith, N.C. 1992. Concepts and strategiesfor anti-parasite immunoprophylaxis andtherapy. Int. J. Parasitol., 22: 1047-1082.

Smith, W.D. 1993. Protection in lambsimmunised with Haemonchus contortusgut membrane proteins. Res. Vet. Sci.,54: 94-101.

Smith, T.S. & Munn, E.A. 1990. Strategiesfor vaccination against gastro-intestinalnematodes. Rev. sci. tech. Off. int. Epiz.,9: 577-595.

Smith, D.B., Davern, K.M., Board, P.G.,Tiu, W., Garcia, E.G. & Mitchell, G.F.1986. Mr 26000 Da antigen of Schistosomajaponicum recognised by WEH1 129/Jmice is a gluathione S-transferase. Proc.Natl Acad. Sci., 83: 8703-8707.

Smith, A.M., Dowd, A.J., McGonigle, S.,ftCeegan, P.S., Brennan, G., Trudgett,

J.P. 1993. Purification of acathep sin L-like proteinase secreted byadult Fasciola hepatica. Mol. Biochem.Parasitol., 62: 1-8.

Stromberg, B.E. & Soulsby, E.J.L. 1977.Ascaris suum: immunization with solubleantigens in the guinea pig. Int. J. Parasitol.,7: 287-291.

Tavernor, A.S., Smith, T.S., Langford, C.F.Munn, E.A. & Graham, M. 1992a.

Vaccination of Dorset lambs againsthaemonchosis. Parasite Immunol., 14:645-655.

Tavernor, A.S., Smith, T.S., Langford, C.F.,Munn, E.A. & Graham, M. 1992b.Immune reponse of Clun Forest tovaccination with membrane glyco-proteins from Haemonchus contortus.Parasite Immunol., 14: 671-675.

Togna, A.R., Del Giudice, G., Verdini, A.S.,Bonelli, F., Pessi, A., Engers, H.D. &Corradin, G. 1986. Synthetic Plasmodiumfalciparum circumsporozoite peptideselicit heterogeneous L3T4+ T cell proli-ferative responses in H-2b mice. J.Immunol., 137: 2956-2960.

Turnbull, I.F., Bowles, V.M., Wiltshire, C.J.,Brandon, M.R. & Els, N.T. 1992. Systemicimmunization of sheep with surfaceantigens from Haemonchus contortuslarvae. Int. J. Parasitol., 22: 537-540.

Urquhart, G.M., Jarrett, W.F.H., Jennings,F.W., McIntyre, W.I.M. & Mulligan,W. 1966a. Immunity to Haemonchuscontortus infection: failure of x-irradiatedlarvae to immunize young lambs. Am.J. Vet. Res., 27: 1641-1643.

Urquhart, G.M., Jarrett, W.F.H., Jennings,F.W., McIntyre, W.I.M. & Mulligan,W. 1966b. Immunity to Haemonchuscon tortus infection: relationship betweenage and successful vaccination withirradiated larvae. Am. J. Vet. Res., 27:1645-1648.

van Wyck, J.A. & Malan, F.S. 1988. Resistanceof field strains of Haemonchus contortusto avermectin, closantel, rafoxanide andthe benzimidazoles in South Africa. Vet.Rec., 123: 226-228.

Vernes, A. 1976. Immunization of the mouseand guinea pig against Trichinella spiralis.In van den Bossche, ed. Biochemistry ofparasites and host/parasite relationships,p. 319-324. Amsterdam, Elsevier /NorthHolland Biomedical.

Wagland, B.M. 1975. Host resistance to thecattle tick (Boophilus micro plus) in cattle.

Aust. J. Agric. Res., 26: 1073-1080.Waller, P.J. & Larsen, M. 1993. The role of

nematophagous fungi in the biologicalcontrol of nematode parasites oflivestock. Int. J. Parasitol., 23: 539-546.

Wharton, R.H. 1976. Tick-borne livestockdiseases and their vectors, 5. Acaricideresistance and alternative methods oftick control. World Anim. Rev., 20: 8-15.

Wijffels, G.L., Sexton, J.L. & Salvatore, L.1992. Primary sequence heterogeneityand tissue expression of glutathione S-transferases of Fasciola hepatica. Exp.Parasitol., 74: 87-99.

Wikel, S.K. 1981. The induction of hostresistance to tick infestation with salivarygland antigen. Am. J. Trop. Med. Hyg.,30: 284-288.

Willadsen, P. 1980. Immunity to ticks. Adv.Parasitol., 18: 293-313.

Willadsen, P. & McKenna, R.V. 1991.Vaccination with concealed antigens:myth or reality? Parasite Immunol., 13:605-616.

Willadsen, P., McKenna, R.V. & Riding,G.A. 1988. Isolation from the cattle tick,Boophilus microplus, of antigenic materialcapable of eliciting a protective immuno-logical response in the bovine host. Int.J. Parasitol., 18: 183-189.

Willadsen, P., T.tiding, G.A., McKenna, R.V.,Kemp, D.H., Tellarn, R.L., Nielsen, J.N.,Lahnstein, J., Cobon, G.S. & Gough,J.M. 1989. J. Immunol., 143: 1346-1351.

Xilinas, M.E., Papavasiliou, L.T. & Marselou-Kinti, 0. 1976. Active immunisation ofmice against Echinococcus granulosus.Hellenica Acta Microbiol., 21: 86-90.

Vaccine manual111

Vaccination of poultry presents a numberof special problems. The first is the limitedlife of the meat-producing chicken, whichis often as short as 35 days from hatchingto slaughter. These birds have a lowindividual value and any vaccine musttherefore be relatively cheap. They are keptin very large numbers at a very highdensity. Houses often contain 10 000 to30 000 birds and sites often have between50 000 and one million birds. The problemis compounded because these birds haveprobably originated from eggs producedby a number of breeding farms. Thesebreeding farms may well be in a differentregion (or evert country) and may have adifferent antibody status and diseasehistory. Therefore, the chickens hatchingfrom these eggs will have considerablevariations in maternal antibodies and, insome cases, chicks will have been verticallyinfected with viruses such as chick anaemiavirus (CAV) or avian encephalomyelitis.This problem is becoming more acute ashygiene and disease precautions areincreased in the breeding farms, forexample to control Salmonella spp. If all thebreeding farms are infected with a virussuch as CAV before they start producingeggs, it is not a problem. However, if someare not infected, then two problems canarise: the first is if the chicks are placed inan area where CAV is endemic, in whichcase the susceptible chicks originating fromparents without antibody become infected;the second is when one of the parent flocksbecomes infected. This flock will produceinfected and diseased chicks and they willinfect the susceptible chicks in the hatd-teries.

Poultry are infected with a number of

Poultry vaccines (1)

Convntonci vaccinesJ.B. McFerran

immunosuppressive viruses, of which themost important is infectious bursal disease(ISD) virus, which has virtually a world-wide distribution. These viruses have threeimportant effects: they can cause diseasein their own right; they can compromisethe birds' immunological response to othervaccines; and they can reduce the birds'resistance to other organisms so that asevere reaction to vaccination with anattenuated vaccine can occur as a result ofother agents combining with the vaccineto induce disease.

The disadvantages of large-scale poultryproduction can in part be overcome byhaving an "all-in all-out" system, where asite is completely depopulated and isproperly cleaned and disinfected beforethe next intake of day-old chicks. It isrelatively easy to clean and disinfect (e.g.by formalin fumigation) environmentallycontrolled houses with impervious wallsand concrete floors but it is virtuallyimpossible to disinfect open-sided houseswith earth floors. In addition, commercialpressures often lead to bad microbiologicalpractices. It is not uncommon for organi-zations which do depopulate and cleanhouses to have large sites where the lasthouses to be depopulated are still beingcleaned when chicks are filling the firsthouses. Even worse, in some areas litter isreused many times and young chicks arebrooded in houses still containing olderbirds.

Most commercial poultry organizationsare highly integrated with considerablemovement between farms. Thus, layingflocks will have regular egg collections,with transport moving from farm to farm;

Vaccine manual 113

114 Conventional vaccines

breeders and broilers are often weighedregularly by the organization's advisorystaff; there is movement and contact bycatching and vaccination teams, etc. Whilea reasonable disinfection of staff can becarried out, it is more difficult to ensurethat tools and equipment (e.g. weighingscales, electrician's tools, egg trolleys) aresafe. These factors make the spread ofdisease relatively easy.

From a disease control viewpoint thereare three main types of poultry: i)commercial flocks, usually kept underintensive husbandry methods; ii) localsmall backyard flocks, which in most areasare unvaccinated; and iii) hobby birds. Thelatter range from Psittacines (especiallydangerous because they can be infectedwith virulent Newcastle disease virus) toracing pigeons. Wild birds also pose aproblem as they are capable of carryinginfection long distances and may indeedbe hosts of certain viruses. For instance,wild ducks are often infected withparamyxoviruses, including Newcastledisease, myxoviruses and the egg dropsyndrome, adenovirus.

Humans are probably one of the mostimportant vectors for spreading infection,second only to the bird. It is possible toreduce the risk of introducing disease bykeeping all but essential people out ofpoultry houses and by ensuring that thosewho enter take precautions (e.g. by usingspecial outer clothing kept on the farm foreach essential visitor).

It is unlikely that organisms will spreadfrom hobby birds or backyard flocks byaerial routes because the number of birdsinvolved is too small. Some viruses canspread from commercial farm tocommercial farm by wind owing to thelarge concentration of birds (e.g. Newcastledisease and Marek's disease viruses). Otherroutes are also important, such as un-treated water (e.g. spreading influenza,Newcastle disease, egg drop syndrome,

salmonellosis), feed (e.g. paramyxovirustype 1 [pigeon Newcastle disease], Salmo-nella spp., IBD) and transport.

Given the above considerations, it canbe seen that the chances of poultry diseasespreading are very high. In addition, thepoultry industry is truly international, withbirds (and advisory staff) moving freelybetween countries. The speed of the spreadof "new" diseases in recent years under-lines this. Diseases such as runting-stunting, egg drop syndrome, turkeyrhinotracheitis virus, virulent forms ofNewcastle disease, Marek's disease andIBD as well as antigenic variants ofinfectious bronchitis (IB) have all spreadrapidly around the world. The emergenceof other new diseases such as chickanaemia, may be the result of changes inmanagement rather than the spread ofvirus, but this is unclear at present.

VACCINATION

Clearly, vaccination is essential tocommercial poultry production. Howeverwith the birds' short life, high-densityconditions and variable antibody levels, itis very difficult to achieve full protectionand it is essential that every effort be madeto prevent the introduction of infection orcarryover of infectious agents from flockto flock. To some extent, the degree ofresistance induced by any vaccine is subjectto the level of challenge present and, ifthat challenge is great enough, vaccinesare unlikely to give protection.

The ideal vaccine should:be able to be given by mass application;

.be cheap;be safe;

° be capable of inducing a strongimmunity even in the presence ofmaternal antibody;be capable of inducing both local andgeneral immunity;induce an immune response that canbe distinguished from that induced by

infection or maternal derived antibody;require only one dose to give lifelongimmunity;

*be packaged in different sizes (e.g.while 1 000-dose vials are appropriatefor use in commercial flocks, they arenot appropriate for use in a smallbackyard flock).

Such a vaccine does not exist at present.Successful vaccination against many

organisms depends on stimulating bothlocal immunity at a mucosal surface andgeneral humoral antibody. The localimmunity is of major importance for thoseagents which affect mucosal surfaces forexample IB and is of special importancein meat-producing birds which are housedat very high densities, have a short lifespan and require quick immunity.

SerologyThere are a wide range of tests used tomonitor vaccination and also for diagnosis.The modern tests, for example ELISA tests,are expensive and it is important to usethem appropriately and not just to ac-cumulate data. They can be used tomonitor the efficiency of vaccination whereimmunity (general) is measured by anantibody response. Titres should not onlybe high but, even more important, be in aclose range. It is important to rememberthat if 10 percent of the birds have lowtitres or no antibody, this can be translatedinto 2 000 to 3 000 birds dying in a house.There is no easy method of measuring localimmunity. This is unfortunate, since localimmunity is often very important, as forexample in IB. Other uses for serology areto determine if a flock has been infected.Thus, if birds in the rearing period developantibody to CAV, there is no need tovaccinate. As already discussed, it is theflocks without antibody that pose theproblems.

Determination of antibody titres inbreeding flocks can allow the breeding

flocks to be batched into groups so that theprogeny will have similar titres and, con-sequently, the timing of vaccination can bebetter judged. This has been used in thecontrol of IBD. The difficulty is thatthere is often a wide variation in antibodytitres in birds in any one flock.

Vaccine administrationThere are a number of commonly usedmethods of vaccination. Meat-producingchickens are normally handled twice atbirth and at death. Therefore, vaccinationis either done individually in the hatchery(e.g. Marek's vaccination) or else massapplication methods are used. The easiestmethod is to give the vaccine in thedrinking-water, yet this is most difficult todo properly. There is always a percentageof birds that will not drink. Although thispercentage can be reduced by ensuring thebirds are thirsty, they must not be stressedand the water must be turned onimmediately afterwards, with no dangerof airlocks occurring. Furthermore, thevaccine may not be uniformly available inthe drinking-water because of physicalproblems such as uneven troughs, orvaccine may be lost as a result of adherenceto rusty containers or feed which has falleninto the drinkers. Vaccine can also beinactivated by chemicals (e.g. chlorine) inthe water or by heat. Such problems can inpart be overcome by adding protein (e.g.dried skim mill< powder) to the vaccineand by giving the vaccine in two parts(McFerran et al., 1972). It is also essential tokeep the birds at a comfortable airtemperature. This may require raising thetemperature in cold areas or vaccinatingearly in the morning in hot regions.

Partly to overcome the problems of wateradministration and partly to induce localimmunity, spraying has been widely used.The use of a coarse spray (in reality a masseye and nose drop application) toimmunize day-old chicks which possess

Vaccine manual 115

maternally derived antibody againstdiseases such as IB, has been routine inhatcheries in some areas for 25 years. Thishas given excellent immunity, lasting thelife of the meat-producing chicken. Evenin the case of emergency vaccinations ofantibody-free chicks with Newcastledisease virus, excellent immunity wasprovided with acceptable side-effects(McFerran, 1992). Severe side-effects didoccur when the spray was too fineand virus penetrated the lower respira-tory tract. Results similar to those givenby coarse spray have been obtained usingthe labour-intensive method of beakdipping.

Aerosols, generated using an electricallypowered sprayer, or a fine spray from ahand operated pump have been used torevaccinate birds. In some instances, thishas been done every 12 weeks to ensurethat virtual continuous local immunity ismaintained at mucosal surfaces.

One constraint on the use of sprayvaccines in some areas is that other micro-organisms infecting the birds, such asMyco plasma spp. and Haemophilitis spp., canproduce severe side-effects. Therefore, incertain organizations or regions it isinadvisable to use spray vaccination.

In theory, the large cleft in the oralpalates of fowl and turkeys should alloworally administered vaccine to come incontact with the upper respiratory tractand achieve the same results as sprayvaccination. However, in practice, sprayvaccination is much more effective instimulating local immunity in therespiratory tract as well as being lesslabour-intensive. In part, the increasedimmunity following spray vaccination maybe the result of stimulation of theHardarian gland by the spray-appliedvaccine.

Injection (including wing-web stabbing)is still widely used. Virtually all egg-producing birds and many meat-produ-

cing chicks are vaccinated at one day ofage with Marek's vaccine. This is a quickand cheap method of application. Break-downs occur when the vaccine is depositedin the feathers rather than under the skinin a quest for speed and when a blockedneedle or bacterial contamination of theapparatus goes unnoticed.

Inactivated vaccines, often with an oiladjuvant, are widely used. They tend to beexpensive to buy and deliver but have beenvery successful in controlling disease inadults. While they are of value in pro-tecting chicks against disease throughmaternally derived antibody, recentexperience has shown (as in the case ofIBD) that virulent strains can breakthrough the maternal antibody protectionand active immunity must therefore beinduced.

SOME CONVENTIONAL VACCINES

There are a wide variety of conventionalvaccines available. Brief descriptions ofsome are given here, concentrating onunusual aspects or problems and potentialfuture developments.

Marek's diseaseMarek's disease has been controlled byusing type 1 attenuated vaccines (eitherartificially attenuated or naturallyoccurring strains) or the type 3 turkeyherpesvirus (THV). In some areas, a type 2naturally occurring attenuated strain wasnecessary. In some genetic lines, if the birdsare infected with both the serotype 2 andthe lymphoid leukosis virus early in life,there is an enhancement of lymphoidleukosis. In the 1980s, very virulent strainsof Marek's disease arose, and control wasachieved using a mixture of vaccine eithertypes 2 and 3 or types 1 and 3. Thereappears to be a true synergism betweendifferent strains, especially betweenserotypes 2 and 3, and polyvalent vaccineshave been found to provide better


protection against tumour development(Witter et al., 1984). However even thesevaccines have not been totally successfuland preliminary results suggest that, inorder to ensure maximum resistance toMarek's disease, it may be necessary toselect a vaccine appropriate for thepredominant B-haplotype of the chickenflock (Bacon and Witter, 1993).

Infectious bronchitisThe envelope of IB virus possesses surfaceprojections comprising two glycopoly-peptides, S, and S,. While both areimportant in the immune response, itappears that the S, glycopolypeptide is themajor antigenic component.

Protection is given against the homo-logous serotype as well as, to a varyingdegree, against other serotypes. This hasresulted in the use of a number of differentattenuated vaccines in some areas. Re-combination among field and vaccinestrains possibly occurs and is responsiblefor producing new "variant" strains(Kusters et al., 1990). Because of thepossibility of recombination and the factthat attenuated strains can revert back tovirulence, a new vaccine serotype shouldnot be introduced into an area until thereis clear evidence that its use that isessential.

There have been two approaches toimmunizing egg-producing birds. The firstis to prime the birds with an attenuatedvaccine (around four weeks when maternalantibody levels have fallen) and then withan inactivated oil-adjuvanted vaccine or aless attenuated vaccine around 18 weeksto produce high levels of circulatinghumoral antibody. The second is to primethe birds and then commence sprayvaccination at 18 weeks, followed there-after every eight to ten weeks with anattenuated vaccine. The aim here is to keepa high degree of local immunity in the gutand respiratory tract.

Infectious bursa! diseaseBoth attenuated and inactivated vaccinesexist fro IBD. Attenuated vaccines havebeen divided into three broad categoriesbased on virulence: i) virulent or "hot"; ii)intermediate; and iii) "mild" or avirulent.There is some discussion as to whethersome of the "hot" vaccines are any morevirulent than the intermediate ones. Thesuggestion is that some of the "hot" strainsspread better between birds and cantherefore infect birds that were not initiallyvaccinated either because they did notdrink or because they have high maternalantibody levels. It has been suggested thatvirulent viruses can overcome virus-neutralizing antibody titres of 1.500,intermediate vaccines can overcome titresof 1.250 and avirulent viruses can overcometitres of 1.100 or less (Lukert and Saif, 1991).However, given the wide variation of titresin broiler chickens, these figures are at bestonly a guide as to the optimum time tovaccinate.

In view of the wide antigenic variationwithin IBD viruses (McFerran et al., 1980)it is surprising that only one subtypevaccine (the Delaware) has been necessary.There has been a major upsurge invirulence of IBD viruses throughout theworld, and the policy adapted in manyareas of only vaccinating the breeders andrelying on maternal antibody to protectthe broilers has proved flawed. In manyareas, it has been necessary to useintermediate or hot vaccines in broilerflocks to control the losses. Even whenflocks are clinically normal, IBD virus canattack the bursa and cause lesions. In thesecircumstances, the subclinical infectionscan cause up to a 10 percent reduction inincome (McIlroy, Goodall and McCracken,1989).

Mycoplasma gallisepticumAlthough Mycoplasma gallisepticum hasbeen eradicated from commercial poultry

Vaccine manual 117


in many areas, it is still widely distributed,especially in areas where multi-age sitesand open-sided houses with earth floorsmake eradication very difficult. Underthese conditions, vaccination of com-mercial egg-producing birds to control fallsin egg production is indicated. Bacterins,which are inactivated suspensions of wholeM. gallisepticum (Mg') organisms in oilemulsion, are apparently effective incontrolling disease but have little benefitin eliminating infection in multi-age sites.Recently, inactivated M. gallisepticumbacterin mixed with 0.2 percent iotacarrageenan as an adjuvant inducedresistance to air sacculites in chickenchallenged with the virulent MGR strain(Elfaki et al., 1992).

Live vaccines are also used to control M.gallisepticum. The F strain is a naturallyoccurring strain of relatively low virulencefor chickens but of a higher virulence forturkeys. It will displace more virulent fieldstrains from multi-aged sites but vac-cinated birds become permanent carriers(Kleven, Khan and Yamamoto, 1990). Atemperature-sensitive mutant which isavirulent has recently been described(Whithear et al., 1990).

Fowl choleraFowl cholera is caused by Pasteurellamultocida, which can be serologicallysubdivided by differences in capsularantigens and cell wall lipopolysaccharideantigens. Using a passive haemagglutinintest, five capsular serogroups (A,B,D,E andF) can be identified; however, thesecapsular antigens do not appear to play arole in protection. P. multocida can also besubdivided into 16 serotypes using a heat-stable somatic antigen (Brogdens andRebers, 1978). Both inactivated and liveP. multocida vaccines are available. Inactiv-ated vaccines are whole cell products of anumber of strains (often including "auto-genous" local strains) emulsified in an oil

adjuvant. The attenuated strains have arange of virulence, although two doses ofinactivated vaccine during rearing usuallyprovide adequate immunity as long as theappropriate strains are included in thevaccine. Attenuated vaccines are widelyused in some areas, especially in meatturkeys. Live vaccines often cause chronicfowl cholera. This effect can be minimizedby first giving an inactivated vaccine.

Infectious coryzaInfectious coryza is caused by Haemophilusparagallinarum, the isolates of which it ispossible to divide into three serogroupswith at least seven serovars by using theHI test (Kume et al., 1983). Vaccines consistof inactivated whole-cell bacterinsemulsified in oil adjuvant or absorbed onto aluminium hydroxide. The immunityinduced by these vaccines is serogroup-specific. Breeders and commercial layersare routinely vaccinated in many countries,using two injections during rearing at leastfour weeks apart. Using chemicalmutogens, Blackall et al., (1993) haveproduced non-pathogenic mutants whichmay be of value as attenuated vaccines.

CoccidiosisA widespread and serious disease causedby Eimeria spp. is coccidiosis. It has beencontrolled in the past by the continuousadministration of chemotherapeutic agentsto the flock's feed. However, problemswith the development of resistance andthe need to withdraw the agents for aspecified period before slaughter have ledto the development of vaccines.

At present, vaccines are used mainly inlayers and breeders. They containattenuated oocysts of multiple species ofEimeria. Attenuation has been achieved bya number of methods such as adaptationto growth in avian embryos or by selectingstrains with a shortened prepatent periodin the precocious strains.

A second approach has been to definecoccidial immunogens which _could beused as subunit vaccines. There are,however, many species of Eimeria, withlittle or no cross-immunity betweenspecies.

NEW TECHNICAL DEVELOPMENTS

In 1982, Sharma and Burmester inoculatedembryos with THV in an attempt tocircumvent the massive Marek's diseasechallenge that day-old chicks faced onsome sites. These chickens developed apersistent THV infection and were resistantto early challenge with pathogenic Marek'sdisease. The vaccine was inoculated intothe allantoic sac; and it was found that thebest results were obtained in 17- to 18-day-old embryos. Inoculation at 18 days wouldcoincide with the transfer of eggs from thesetting to the hatching machines.

Sharma (1985) extended this work to theinvestigation of protection against IBD. Hefound that low virulent vaccine strainscould induce immunity without effectinghatchability or survival of the chicks.However, maternal antibody didneutralize strains of low virulence.Recently, the use of an IBD commercialvaccine-antiserum mixture to inoculate 18-day-old embryonated eggs has beendescribed. This preparation inducedimmunity in chicks produced from eggswith maternal antibody to IBD (Whitfill etal . , 1993). Embryos originating fromspecific pathogen-free (SPF) turkeys weresuccessfully inoculated with marble spleendisease and Newcastle disease vaccinesand the poults were resistant to subsequentchallenge (Ahmad and Sharma, 1993).

This work has been developedcommercially and an automatic egginjection system is now in operation(Marchant, 1993), injecting THV and SB-1Marek's vaccine into embryonated chickeggs. The machine can inject 20 000 to30 000 eggs per hour. The success of these

approaches will depend on whether theeconomic advantages claimed are realized.This is especially true where hygienestandards are high and day-old chicks donot face a massive challenge.

One different approach to embryonatedegg inoculation is to inject an immuno-stimulant containing lymphokines. It isclaimed that this has enhanced the efficien-cy of Marek's disease vaccination and, inaddition, resulted in lower mortality,reduced feed conversion and lowerproduction costs for commercial broilers(Miles et al., 1993).

Genetic engineeringInitially, organisms were modified by eggpassage (e.g. IB), by cell culture passage orby identifying a naturally occurring non-pathogenic strain (e.g. Rispen's vaccineagainst Marek's disease). The attenuationprocess can be enhanced by using chemicalmutagens, and mutants can be selectedusing a range of techniques such as plaquepurification, temperature sensitivity,colony size or an appropriate monoclonalantibody. Some proteins are only expressedwhen bacterial cells are grown in vivo, and itis suggested that this is the reason whyPasteurella multocida grown in vivo givesbetter vaccines. An attempt to clone thegenes responsible for expressing theseproteins is in progress and may give rise tobetter inactivated or attenuated vaccines(Jost et al., 1993). It is now possible tomodify the genorne by chemical means, asin the Aujeszky's disease vaccine virus inwhich the thymidine kinase gene has beendeleted.

It is possible to insert genes into vectors.These vectors can be organisms which donot replicate in the host, in which case anadjuvant is indicated. Thus, a recombinantbaculovirus (expressing the haemag-glutinin and neuraminidase protein ofNewcastle disease virus), combined withan oil adjuvant, is effective in protecting

Vaccine manual119


against Newcastle disease challenge (Nagyet al., 1991).

The alternative approach is to incor-porate one or more genes into a vectorwhich does replicate in the host. Avianpoxviruses, avian herpesviruses (serotype2 and turkey herpesvirus) and avianadenoviruses have been investigated. Anexample is the incorporation of the genecoding for the haemagglutinin of avianinfluenza (H5N2) into a poxvirus (Beard,Schnitzlein and Tripathy, 1991). The genecoding for the fusion protein (F) ofNewcastle disease virus has also beenincorporated into poxvirus (Iritani et al.,1991). This gives immunity but does notgive rise to H1 antibodies, thus allowing atest to detect infection by the Newcastledisease virus. Antibody to Newcastledisease did not prevent the developmentof immunity, but antibody to fowl poxprevented the poxvirus growing. Thus, thisvaccine could be used to vaccinate duckswith maternal antibody to Newcastledisease. Similarly, a fowl poxvirusexpressing the turkey rhinotracheitisfusion glycoprotein gave partial protectionto challenge by turkey rhinotracheitis virus(Quinzhong et al., 1993).

A promising approach is to use a turkeyherpesvirus (THV) as the vector. Arecombinant THV expressing theNewcastle disease virus fusion proteingave good protection against both Marek'sand Newcastle diseases (Morgan et al.,1992). THV is used widely to vaccinateday-old chicks. It is universal and safe(unlike fowl pox which is absent in someareas and which may cause local reactions),it does not spread horizontally and, mostimportant, it gives a persistent viraemialasting several weeks. This approach maywell overcome the problems of maternalimmunity. Thus, if genes giving resistanceto a number of diseases were incorporated,they could stimulate the immune responseto each agent at the appropriate time as

maternal antibody wanes. This wouldovercome the hurdle of varying levels ofantibody to different viruses and also tothe same virus, owing to the fact that theflock is made up of the progeny of anumber of breeding rllocks.

There is always the problem of thematernal antibody to the vector preventingthe vector growing. In the case of THV thiscould be overcome by using the THV inalternate generations (it does not occur asa natural infection in chickens), as isalready the practice.

While it is possible to express the genesencoding protein antigens in vitro for useas subunit vaccines, and also to synthesizepeptide vaccines chemically, thesesolutions may present problems regardingthe cost of production and administrationof these vaccines, difficulties in obtainingthe correct folding of the protein and thelimited number of antigenic determinants,which could mean that strains of agentscould arise which are not neutralized.

Although, in theory, anti-idiotypevaccines are possible, in practice they havebeen poorly immunogenic and show littleimmediate promise. A different approachis the use of cytokines as adjuvants. Thesecan be incorporated in the injectible vaccinewhich may also include slow-releasemechanisms and immunostimulatingcomplexes (ISCOMs). Such an approachhas been taken with a subunit vaccineagainst avian influenza in which nucleo-protein with residual haemagglutininactivity was incorporated into ISCOMs(Sivanandan et al., 1993).

Liposome-adjuvanted egg drop syndrome1976 vaccines appeared as efficacious as oil-emulsion vaccines without the side-effectsof the oil (Jing-Sheng and Yi-Zhu, 1993).

Using trivalent avian influenza antigens,Fatunmbi et al. (1992) found that chargedliposomal avridine adjuvant produced abetter antibody response than otheruncharged liposomal avridine or oil-

emulsion adjuvants. The positively chargedliposornal avridine adjuvant was superiorto negatively charged adjuvant.

BIBLIOGRAPHY

Ahmad, J. & Sharma, J.M. 1993. Protectionagainst hemorrhagic enteritis and New-castle disease in turkeys by embryo vac-cination with monovalent and bivalentvaccines. Avian Dis., 37: 485-491.

Bacon, L.D. & Witter, R.C. 1993. Influenceof B-haplotype on the relative efficacyof Marek's disease vaccines of differentserotypes. Avian Dis., 37: 53-59.

Beard, C.W., Schnitzlein, W.M. & Tripathy,D.N. 1991. Protection of chickens againsthighly pathogenic avian influenza virus(H5N2) by recombinant fowlpox viruses.Avian Dis., 35: 356-359.

Blackall, P.J., Graydon, R.J., Rafiee, M. &Tinworth, D. 1993. Towards a liveinfectious coryza vaccine. In Proc. X WorldVet. Poultry Assoc. Congr., Sydney,Australia.

Brogdens, K.A. & Rebers, P.A. 1978. Serolo-gical examination of the Westphal-typelipopolysaccharides of Pasteurellamultocida. Am. J. Vet. Res., 39: 1680-1682.

Elfaki, M.G., Kleven, S.H., Jin, L.H. &Ragland, W.L. 1992. Sequential intra-coelomic and intrabursal immunisationof chickens with inactivated Myco plasmagallisepticum bacteria and iota carageenanadjuvant. Vaccine, 10: 655-662.

Fatunmbi, 0.0., Newman, J.A., Sivanandan,V. & Halvorson, D.A. 1992. Enhance-ment of antibody response of turkeysto trivalent avian influenza vaccine bypositively charged liposomal avridineadjuvant. Vaccine, 10: 623-626.

Iritani, Y., Aoyama, S., Takigami, S.,Hayashi, Y., Ogawa, R., Yanagida, N.,Saeki, S. & Kamogowa, K. 1991. Anti-body response to Newcastle disease virus(NDV) by recombinant fowlpox virus(FPV) expressing a hemagglutinin-

neuraminidase of NDV into chickensin the presence of antibody to NDV orFPV. Avian Dis., 35: 659-661.

Jing-Sheng, L. & Yi-Zhu, W. 1993. Studieson the liposorne-adjuvanted antigensof the inactivated EDS.76 virus. In Proc.X World Vet. Poultry Assoc. Congr.,Sydney, Australia.

Jost, H., Homchampa, P., Ruffolo, C. &Adler, B. 1993. Molecular approachesfor studying the virulence genes andantigens of P. multocida; strategies fornovel fowl cholera vaccines. In Proc. XWorld Vet. Poultry Assoc. Congr., Sydney,Australia.

Kleven, S.H., Khan, M.I. & Yamamoto, R.1990. Fingerprinting of Myco plasmagallisepticum strains isolated from multi-age layers vaccinated with live F strain.Avian Dis., 34: 984-990.

v>ume, K., Sawata, A., Nakai, T. &Matsumoto, M. 1983. Serological classi-fication of Haemophilus paragallinarumwith a hemagglutinin system. J. Clin.Microbiol., 17: 958-964.

Kusters, J.G., Jager, E.J., Niesters, H.G.M.& van der Zeijst, B.A.M. 1990. Sequenceevidence for RNA recombination in fieldisolates of avian coronavirus infectiousbronchitis virus. Vaccine, 8: 605-608.

Lukert, P.D. & Saif, Y.M. 1991. Infectiousbursal disease. In B.W. Calnek, ed.Diseases of poultry, p. 342-385. Ames,USA, Iowa State University Press.

Marchant, J. 1993. Embrex targeting a nichemarket. Animal Pharmacol., 269: 14-15.

McFerran, J.B. 1992. Control of Newcastledisease in Northern Ireland (NDV vacc.).In Proc. Workshop of Avían Paramyxo-viruses, p. 238-249. Rauischolzhasen,Germany, Commission of the EuropeanCommunities.

McFerran, J.B., Young, J.A., Clarke, J.K.Wright, C.L. 1972. Observations con-cerning the use of living infectiousbronchitis vaccine under field conditions.Vet. Rec., 90: 527-530.

Vaccine manual 121


McFerran, J.B., McNulty, M.S., McKillop,E.R., Connor, T.J., McCracken, R.M.,Collins, D.S. & Allan, G.M. 1980.Isolation and serological studies withinfectious bursal disease viruses fromfowl, turkeys and ducks: demonstrationof a second serotype. Avian Pathol., 8:205-212.

McIlroy, S.G., Goodall, E.A. & McCracken,R.M. 1989. Economic effects of subclinicalinfectious bursal disease on broilerproduction. Avían Pathol., 18: 465-473.

Miles, A.M., Doelling, V.W., Phelps, P.V.,Ricks, C.A., Tyczkowski, J.K., Whitfell,C.E. & Gildersleeve, R.P. 1993. Efficacyof an immunostimulant administeredin ovo. In Proc. X World Vet. Poultry Assoc.Congr., Sydney, Australia.

Morgan, R.W., Gelb, J., Schreurs, C.S.,Luttiken, D., Rosenberger, J.K. &Sandermeijer, P.J.A. 1992. Protectionof chickens from Newcastle and Marek'sdisease with a recombinant herpesvirusof turkeys vaccine expressing theNewcastle disease fusion protein. AvíanDis., 36: 858-870.

Nagy, E., Krell, P.J., Dulac, G.C. &Derbyshire, J.B. 1991. Vaccinationao-ainst Newcastle disease with a recom-binant baculovirus hemagglutinin-neuraminidase subunit vaccine. AvíanDis., 35: 585-591.

Quinzhong, Yu., Barrett, T., ri own, D.X.,Cook, J.K.A., Green, P., Skinner, M. &Cavanagh, D. 1993. Protection againstturkey rhinotracheitis pneumovirus(TRT) induced by a fowl pox recombinantexpressing the TRTV fusion glycoprotein(F). In Proc. X World Vet. Poultry Assoc.Congr., Sydney, Australia.

Sharma, J.M. 1985. Embryo vaccination withinfectious bursal disease virus alone orin combination with Marek's diseasevirus. Avían Dis., 29: 1155-1169.

Sharma, J.M. & Burmester, B.R. 1982.Resistance to Marek's disease at hatchingin chickens vaccinated as embryos with

turkey herpesvirus. Avían Dis., 26:134-149.

Sivanandan, V., Kodihalli, S., Nagaraja,K.V., Halvorson, D.A. & Newman, J.A.1993. ISCOM adjuvanted cross-protectivesubunit vaccine against avian influence.In Proc. X World Vet. Poultry Assoc. Congr.,Sydney, Australia.

Whitfill, C.E., Ricks, C.A., Haddad, E.E. &Avakian, A.P. 1993. In ovo administrationof a novel infectious bursal diseasevaccine in broiler chickens. In Proc. XWorld Vet. Poultry Assoc. Congr., Sydney,Australia.

Whithear, K.G., Soeripto, Harrigan, K.E.& Ghiocas, E. 1990. lmmunogenicity ofa temperature-sensitive mutant ofMycoplasina gallisepticurn vaccine. Aust.Vet. J., 67: 168-174.

Witter, R.L., Sharma, J.M., Lee, L.F., Opitz,H.M. & Henry, C.W. 1984. Field teststo test the efficacy of polyvalent Marek'sdisease vaccine in broilers. Avían Dis.,28: 44-60.

The poultry industry in developingcountries, especially the village poultryindustry, encounters different problems tothose found in the developed countriesand, consequently, the requirements forvaccine by this segment of the market arevery different. Certainly, there are elementsof the international commercial industrywith poultry enterprises close to the largercities in many developing countries, andthese enterprises have access to interna-tional breeding stock and internationalexpertise as well as the opportunity andfunding to draw vaccines of high qualityfrom the international market. The vaccinesused on these poultry populations in adeveloping country will seldom have beenproduced in that country. Instead, they willhave been produced to meet the higheststandards as discussed elsewhere in thismanual and the manufacturers will haveensured as far as possible the absolutesafety and adequate efficacy of thevaccines.

Village chickens comprise the major partof the poultry industry in many developingcountries. Village flocks are small, of mixedage and poorly housed or even unhoused.The chickens gain much of their nourish-ment by scavenging in the village environ-ment and by consuming supplements ofhousehold scraps if these are madeavailable. The village flocks are poorlyproductive, as the major call on energy isfor reproduction and maintenance of thepopulation. Extremely high mortality ratesensure that few surplus eggs and birdsbecome available for sale, barter orconsumption. Nearly all the eggs are

required for brooding. Two elementsresponsible for these extreme mortalitieshave been cited. First is the high wastagethat occurs during the brooding season,which is probably caused by a combinationof starvation, predation and infectiousdisease but is indirectly the result of poorhusbandry and neglect. The second causeof attrition is more directly caused byinfectious disease. Outbreaks of diseasecausing extremely high rates of mortalityin village chickens have been reportedfrom many developing countries. Wherethese diseases have been identified, themajor culprits are Newcastle disease andfowl cholera, which are preventable andreadily controlled if effective vaccines canbe applied. It is the lack of suitable vaccinesand appropriate methods of applicationthat allows Newcastle disease and fowlcholera to plunder the village flocks. Ifthese diseases could be controlled, villagerswould have the incentive to make thechanges in husbandry that would controlbrooding losses.

Suitable vaccines are the key to a "newscience" of village chicken keeping. Ananalysis of the special requirements ofvillage chickens indicates that their needsare not being served by the conventionalvaccines available on the internationalmarket. A start has been made on devel-oping vaccines against Newcastle diseasethat are specifically suited for delivery toand use in villages, (Copland, 1987;Spradbrow, 1992 and 1993 /94). However,it is important that regulatory authoritiesrecognizse the special needs of villagechickens and that well-intentioned regu-

Vaccine manual 123

Poultry vaccines (2)

c-Lavemerds for A' c)],giss cickR.P. Spradbrow

124 Special requirements for village chickens

lations do not impose further barriers tothe provision of suitable vaccines to villagechickens.

Village flocks require special vaccines.The transport of vaccines to villages andtheir storage within these villages mustusually be accomplished without refriger-ation. It will still not be feasible in theforseeable future to establish adequate coldchains; therefore, the vaccines themselvesmust be thermostable. As long as villagersthink that their flocks will survive thefrequent outbreaks of infectious disease towhich they are exposed, adequate housingwill remain a rarity. This means catchingchickens for individual vaccination willoften be impossible. There seems to be noalternative to the oral application ofvaccine, usually on food but also in water,in areas where the sources of drinkingwater can be controlled. When possible,individual applications of vaccine to theeye, nose or mouth is a more reliablemethod of inducing immunity with at-tenuated vaccines. Vaccines for use invillage chickens also need to be extremelycheap. It may be necessary to reconsidersome of the elements that add value toconventional vaccines formulation,packaging, substrate, testing. Nor is itnecessary to demand an extremely highdegree of efficacy of a vaccine for use invillage chickens. Absolute protection mayrequire unrealistically large amounts ofvaccine. The cost-effective compromisemay be a moderate level of protectionachieved with an inexpensive vaccine. Infew places have conventional vaccinescontributed to the protection of villagechickens.

Many developing countries would preferto produce their own veterinary vaccinesfor local use, especially the relativelysimple avian vaccines that can be made inembryonated eggs. There are severalreasons for this. One is the desire fornational self-reliance and the belief that

self-sufficiency in avian vaccines caneventually lead to independence in otherareas of vaccine production. The over-riding imperative is the need to conserveforeign exchange.

The purchase and transport of importedvaccines is a continuing drain on funds.On the negative side, a wasteful regionalovercapacity for vaccine production canbe reached, while the purchase of vaccine-producing equipment of any sophisticationis also an expensive undertaking. Wherevaccine is produced nationally, con-sideration should be given to establishingindependent regional centres for testing.

Many of the standards developed forthe testing of conventional avian vaccinesare not appropriate for the testing ofvaccines for village chickens. Tests forsafety should be undertaken in localchickens and tests for efficacy should alsobe made locally, using indigenous chickensand challenge organisms of local origin. Itshould be recognized that safety andfreedom from nominated adventitiousagents are not synonomous. The first factoris essential, the second may be a luxurythat can be postponed for vaccines forvillage use.

Several procedures that are used in theproduction of conventional vaccines mightbe questioned when vaccines for villagechickens are being produced:

Is specific pathogen-free substrate required?Genuine specific pathogen-free flocks arenot readily established in developingcountries. The costs of initiation, main-tenance and testing are enormous and theadvantages for village-style vaccines areminimal. Some of the flocks claimed to bespecific pathogen-free do not deserve thatdesignation. Some receive live vaccinesand the disease status of some is monitoredonly by the absence of clinical disease.Many are more realistically described asisolated flocks or minimal disease flocks.Vaccine seeds should be specific pathogen-

Place of production

Thermostability

Market

Seed material

Vaccine

Lyophilization

Efficacy

Delivery

Packaging units

free and production of vaccine should usea seed lot system. This will allow aneventual increase in the standards ofvaccine. However, rural poultry should notat this point be deprived of vaccinesbecause of an absence of specific pathogen-free flocks.

Is freeze-drying necessary? Most con-ventional attenuated vaccines produced foruse in commercial poultry are lyophilized.Compared with liquid vaccines, thelyophilized product has greater stabilityand is more convenient to store andtransport. However, freeze-drying andsuitable packaging of the dried productadd to the cost of a vaccine. In somedeveloping countries, avian vaccines areproduced as "wet" vaccines. This may notbe a disadvantage if vaccine is to be usedsoon after manufacture and if localproduction reduces the need to transportvaccines over long distances. Nevertheless,thermostability becomes an importantcharacter of the vaccine strain.

The sophisticated packaging of con-ventional vaccines is not necessary forproducts to be used in village chickens.There are frequent calls for the production

TABLE 11

Comparison of poultry vaccine requirements for commercial and village use

Few large laboratories

Not essential

International

Specific pathogen-free


Desirable

Extreme efficacy reqired

Individual vaccination or by aerosol,spray or drinking water

Large dose (multidose vials)

of small-dose packs because village flocksare small and the supply of vaccine inpackages that contain many doses is seento be wasteful. The conventional 1 000-dose vial is admittedly an embarrassmentfor the owner of a flock of 20 chickens andthere is an obvious need to share vaccineamong village flocks. There is also a needto develop a philosophy of overdosing,rather than discarding unused vaccine.Lentogenic Newcastle disease vaccines arenot harmful to chickens when given indoses larger than those recommended bythe manufacturer. The problem in villagesis not the 1 000-dose content of a singlevial of vaccine, rather it is that of obtaininga single vial of vaccine in systems that aredesigned for bulk distribution to a com-mercial industry.

The logistics of delivering vaccine tovillages is as important a problem asdeveloping a suitable vaccine strain. Thecentralized preparation of food-basedvaccines and their delivery to villages arenot feasible in most countries. Each chickenrequires 7 to 10 g of vaccine-coated food.In most areas, food and vaccine will needto be mixed in the villages, which will

Small local laboratoriés

Very desirable

Local


Not necessarily specific pathogen-free

Not necessary

Moderate efficacy acceptable

Delivery on food advantageous

Small-dose vials preferable

Vaccine manual 125

Commercial chickens Village chickens

126 Special requirements for village chickens

require the regular delivery of smallquantities of vaccine. Recent experimentsindicate that it may be possible to in-corporate in a single pellet a quantity ofNewcastle disease virus vaccine sufficientfor a single chicken. This would helpovercome the problems of central pro-duction and would ease the problem oftransport.

Table 11 indicates some of the require-ments that should be considered whencomparing conventional vaccines andvaccines for village chickens.

BIBLIOGRAPHY

Copland, J.W., ed. 1987. Newcastle disease inpoultry. A new food pellet vaccine. ACIARMonograph No. 5. Canberra, AustralianCentre for International AgriculturalResearch. 119 pp.

Spradbraw, P.S., ed. 1992. Newcastle diseasein village chickens. Control with thermo-s table oral vaccines. In ACIAR Proc. 39.Canberra, Australian Centre for Interna-tional Agricultural Research. 189 pp.

Spradbrow, P.S. 1993 / 94. Newcastle diseasein village chickens. Poultry Sci. Rev., 5:57-96.

With fish being the primary source ofanimal protein in many countries,aquaculture is growing rapidly worldwide,and stress and diseases that accompanyintensive fish culture have led to treatmentwith antibiotics and chemicals. However,as concern over pollution associated withchemical treatments and the emergence ofmultiple resistance to antibiotics make thecontrol of infections more and moredifficult, the emphasis should be on diseaseprevention by means of optimal husbandryand biological control methods, such asvaccination and immunostimulants.

FISH DISEASES

Fish in culture suffer a variety of diseases,including bacterial, viral, parasitic andfungal infections, some of which are shownin Table 12. The majority of infectionsdescribed for fish in culture are bacterial.Vibriosis, associated with a number ofmarine vibrios, is the most widespreadworldwide and not only causes hugeeconomic losses to the marine fish cultureindustry but is the main disease affectingthe farming of prawns in Southeast Asiaand Japan. Furunculosis, hitherto thoughtof as a freshwater disease of wildsalmonids, is the most serious diseaseaffecting cultured salmonids, particularlyin the seawater phase. Acute furunculosisis a common problem when Atlanticsalmon smolts are transferred to the sea, atwhich time losses of up to 30 percent of thestock may occur. The disease is oftencomplicated by simultaneous sea liceinfection and pancreatic disease. Un-fortunately, once furunculosis is intro-duced to a site it is very difficult to

vaccinesA. Adams, K.D. Thompson and R.J. Roberts

eradicate and is likely to recur inter-mittently among the ongrowing stock.Motile aeromonads appear to be the mostimportant bacterial pathogens offreshwater fish in tropical countries. Thesemicro-organisms have been reported tocause mass mortality in Indian major carpand can be detected during a variety ofinfections, including haemorrhagicsepticaemia, asymptomatic septicaemia,epizootic ulcerative disease (EUS), tail rotand fin rot.

The exact aetiology of some of the newlyemerging diseases, such as EUS, have notbeen unequivocally determined. In thecase of EUS, bacterial, viral and fungalparticipants are all associated with thedisease complex but the principal initiatingfactor has yet to be determined (Roberts,Willoughby and Chinabut, 1993).

As the aquaculture industry grows andnew species of fish are intensively cultured,the range of diseases and affected speciesis similarly expanding. For example,pasteurellosis was only recently describedin Europe and rickettsiosis in Chile, Irelandand Taiwan Province of China, but nowboth are recognized as highly significantpathogens.

COMMERCIAL VACCINES

Of the many infectious diseases affectingfish there are to date only five for whicheffective commercial vaccines have beenproduced. Three of these are bacterial

The authors wish to express their thanks to all those

who provided information for this paper, in particular

J. Turnbull, H. Rodger and J. McGeorge for theiradvice on fish diseases.

Vaccine manual 127

TA

BLE

12

Fis

h di

seas

es

(con

t.)

Dis

ease

Cau

sativ

e ag

ent

Maj

or f

ish

spec

ies

affe

cted

Cou

ntry

/reg

ion

Bac

teria

l dis

ease

Ent

eric

red

mou

thY

ersi

nia

ruck

eri

Sal

mon

ids,

prim

arily

rai

nbow

trou

tN

orth

Am

eric

a, E

urop

e, S

outh

Am

eric

aV

ibrio

sis

- V

ibrio

sis

- H

itra

dise

ase

Vib

rio a

ngui

llaru

m, V

ord

alii

V s

alm

onic

ida

Wid

espr

ead

in m

arin

e fis

h sa

lmon

ids

Atla

ntic

sal

mon

Wor

ldw

ide,

Jap

an, N

orth

Am

eric

aN

orw

ay, F

aroe

Isla

nds

Fur

uncu

losi

sA

erom

onas

sal

mon

icid

aS

alm

onid

sE

urop

eB

acte

rial

kidn

eydi

seas

eR

enib

acte

rium

sal

mon

inar

umS

alm

onid

sN

orth

Am

eric

a, E

urop

e, J

apan

, Chi

leE

nter

ic s

eptic

aem

iaE

dwar

dsie

lla ic

talu

riE

. tar

daC

affis

hE

el, h

iram

eS

outh

east

ern

Uni

ted

Sta

tes,

Can

ada

Japa

nM

otile

aero

mon

adse

ptic

aem

iaA

erom

onas

hyd

roph

ila, A

. cav

iae,

A. s

obria

Caf

fish,

cyp

rinid

s, s

alm

onid

sA

sia,

Eur

ope,

Uni

ted

Sta

tes

Pas

teur

ello

sis

Pas

teur

ella

pis

cici

daA

yu, y

ello

w ta

il, s

ea b

ream

, sea

bas

s, c

arp

Uni

ted

Sta

tes,

Jap

an, E

urop

e, T

aiw

anP

rovi

nce

of C

hina

Bac

teria

l col

d-w

ater

dis

ease

Lyto

psy

chro

philu

sS

alm

onid

sU

nite

d S

tate

s, E

urop

e, J

apan

Str

epto

cocc

us in

fect

ions

Str

epto

cocc

us s

pp.

Yel

low

tail,

rai

nbow

trou

t, ay

u, ti

lapi

a,ba

ss, b

ream

Japa

n, U

nite

d S

tate

s, T

aiw

an P

rovi

nce

of C

hina

Tub

ercu

losi

sM

ycob

acte

rium

mar

inum

,M

. for

tuitu

m, M

. che

lona

eSn

akeh

ead,

trop

ical

aqu

ariu

m fi

sh,

sea

bass

, wid

e va

riety

of o

ther

spe

cies

Sou

thea

st A

sia,

Jap

an, E

urop

e

Noc

ardi

osis

Noc

ardi

a as

tero

ides

, N. k

ampa

chT

ropi

cal a

quar

ium

fish

, yel

low

tail,

rain

bow

trou

t and

bro

ok tr

out

Spa

in, J

apan

, Can

ada

Sal

mon

id r

icke

itsia

l sep

ticae

mia

Pis

cific

ketts

ia s

alm

on's

Sal

mon

ids

Chi

le, T

aiw

an P

rovi

nce

of C

hina

, Ire

land

Epi

thel

iocy

stis

Chl

amyd

ia-li

ke o

rgan

ism

sW

ide

varie

ty o

f spe

cies

Nor

th A

mer

ica,

Sou

thea

st A

sia,

Eur

ope,

Sou

th A

fric

aC

lost

ridia

linf

ectio

nsC

lost

ridiu

m b

otul

inum

Sal

mon

ids

Eur

ope,

Uni

ted

Sta

tes

Col

umna

risdi

seas

eF

lexi

bact

er c

olum

nar's

, F. m

ariti

mus

All

fres

hwat

ersp

ecie

s, b

ream

, bas

s,tu

rbot

, sal

mon

Nor

th A

mer

ica,

Asi

a, E

urop

e, J

apan

Ent

roco

ccus

infe

ctio

nE

ntro

cocc

us s

erro

lcid

aY

ello

w ta

ilJa

pan

Bac

teria

l gill

dis

ease

Cyt

opha

ga s

pp.,

Fle

xiba

cter

spp

.F

lavo

bact

eriu

m b

ronc

hiop

hilia

Wid

e va

riety

of s

peci

esN

orth

Am

eric

a, J

apan

, Eur

ope

Ulc

erat

ive

sept

icae

mia

Pse

udom

onas

sp.

Eel

s an

d ot

hers

Japa

n

TA

BLE

12

(con

tinue

d)n

Dis

ease

Cau

sativ

e ag

ent

Maj

or f

ish

spec

ies

affe

cted

Cou

ntry

/reg

ion

Vira

l dis

ease

sIn

fect

ious

pan

crea

tic n

ecro

sis

Birn

aviru

s (d

s R

NA

)S

alm

onid

s, s

ea b

ass,

sea

bre

am, t

urbo

t,P

acifi

c co

dE

urop

e

Vira

l hae

mor

rhag

ic s

eptic

aem

iaR

habd

oviru

sS

alm

onid

sJa

pan,

Nor

th A

mer

ica,

Eur

ope

Infe

ctio

us h

aem

orrh

agic

nec

rosi

sR

habd

oviru

sS

nake

head

, car

p, b

arbs

Japa

n, T

aiw

an P

rovi

nce

of C

hina

,C

anad

a, N

orth

Am

eric

a

Infe

ctio

us s

alm

on a

naem

iaU

nkno

wn

(put

atat

ive

viru

s)A

tlant

ic s

alm

onN

orw

ay

Par

asiti

c di

seas

esS

ea li

ceL

epeo

phth

eiru

ssa

lmon

isM

arin

e-cu

lture

d sa

imon

ids

Nor

ther

n ci

rcum

pola

r (N

orw

ay, J

apan

,S

cotla

nd, I

rela

nd, C

anad

a)

Pro

lifer

ativ

e ki

dney

dis

ease

Uni

dent

ified

myx

ospo

rean

extr

aspo

rgon

ic s

tage

, PK

X

Fre

shw

ater

sal

mon

ids

Uni

ted

Kin

gdom

, Eur

ope,

Uni

ted

Sta

tes

Cos

tiasi

sIc

hthy

obod

o ne

cato

rF

resh

wat

er, n

on-h

ost-

spec

ific

finge

rling

fish

espe

cial

ly a

ffect

ed. A

lso

asa

ltwat

er fo

rm

Wor

ldw

ide,

2°-

30°C

Whi

te s

pot

lcht

hyop

hthi

rius

o F

resh

wat

er, e

spec

iaily

you

ng fi

sh;

e.g.

cyp

rinid

s, ti

lapi

ids,

sal

mon

ids,

icta

lurid

s

Wor

ldw

ide,

4°-

25°C

Tric

hodi

nids

Tri

chod

inas

p., T

ripar

tiella

sp,

and

oth

ers

Fre

shw

ater

and

mar

ine

non-

spec

ific

salm

onid

s fla

tfish

(e.

g. tu

rbot

) in

cul

ture

Wor

ldw

ide

Myx

ospo

rean

sA

ran

ge o

f pat

hoge

nic

spec

ies,

e.g.

Myx

obol

us s

ap.,

Sph

aero

spor

a sp

p.,

Kud

oa s

op.

Fre

shw

ater

and

mar

ine.

All

cultu

red

fish

Wor

ldw

ide

Mic

rosp

orea

nsP

leis

toph

ora

sp.,

Glu

gea

sp. a

nd o

ther

sF

resh

wat

er a

nd m

arin

e. O

ne r

epor

ted

prob

lem

in fl

affis

h

Fun

gal d

isea

ses

lcht

hyop

honi

asis

lcht

hyop

honu

sspp

.F

resh

wat

er a

nd m

arin

e sp

ecie

sW

orld

wid

e

Bra

nchi

omyc

osis

Bra

nchi

omyc

essa

ngui

nis,

B,

dem

igra

nC

yprin

ids,

eel

s, fr

eshw

ater

tenc

h,st

ickl

ebac

k

Indi

a, J

apan

, Eas

tern

Eur

ope

Sap

role

gnia

sis

Sap

role

gnia

par

as/ti

ca -

dic

lina

com

plex

Col

d, fr

eshw

ater

sal

mon

ids,

cat

fish

Nor

ther

n E

urop

e, U

nite

d S

tate

s

Asp

ergi

llom

ycos

isA

sper

gillu

s sp

p.T

ilapi

aW

orld

wid

e

Epi

zoot

ic u

lcer

ativ

e sy

ndro

me

Unk

now

n (p

utat

ive

fung

us)

Fre

shw

ater

and

bra

ckis

h sp

ecie

sA

ustr

alia

, Sou

thea

st A

sia

diseases affecting salmonids enteric redmouth (ERNI), vibriosis and furunculosiswhile the others are a bacterial disease incatfish enteric septicaemia of catfish(ESC) and the viral infection, springviraemia of carp (SVC), as shown inTable 13.

Fish vaccines first became available in1976 when a commercial ERM vaccine wasdeveloped and registered in the UnitedStates (Tebbit, Erikson and Vande Water,1981). By 1982, Busch stated that vac-cination was the single most effectivecontrol for ERM. The disease is closelystress-related and the causative agent,Yersinia ruckeri, is an important primarypathogen for intensively cultured rainbowtrout (Busch, 1982).

A commercial Vibrio vaccine wasdeveloped in 1980. This simple, killedwhole-cell vibrio vaccine is highly effectiveagainst the marine pathogens Vibrioanguillarum and Vibrio ordalii in salmonids(Fryer, Rehovec and Garrison, 1978). Theprotective heat-stable lipopolysaccharide(LPS) antigens, derived from cell walls,which make the Vibrio vaccine so successfulin salmonids have also proved effectivefor other farmed species, for example cod,eel, ayu (Kawai, Kusuda and Itami, 1981)with little or no modification to thevaccine's original design. More recently,with the upsurge in farming of sea bassand bream in the Mediterranean, thecontrol of vibriosis has become importantin these species too. Vaccines containingadditional serotypes such as serotype IIIappear to be necessary for full protection(Vigneulle et al., 1993). Generally, vibriosisin fish accompanies some other stress orphysical trauma but some strains of V.anguillarum or V. salmonicida appear to behighly infectious primary pathogens. InNorway, where V. salmonicida causes Hitradisease or cold-water vibriosis, a combinedvaccine is available because there is nocross-protection between the pathogens.

As shown in Table 13, a triple vaccine isalso available against furunculosis and thetwo Vibrio species.

Since the initial work of Duff (1942),many years of research have been dedi-cated to the development of a vaccineagainst furunculosis, the most economic-ally important disease affecting salmonids.Aeromonas salmonicida is the causative agentand outbreaks of the disease are stress-associated, with high mortalities amongsalmon. Until recently, only limited successhad been achieved with vaccination. Incontrast to the early simple, killed whole-cell vaccines, the organisms for the newgeneration of furunculosis vaccines arecultured under iron-limited conditionswhich closely mimic the situation found infish. This allows expression of the essentialprotective antigens, which include LPS andIROMPS (iron regulatory proteins) (Hirstand Ellis, 1994). The use of adjuvants,which enhance the immune response tothe vaccine, also play an important role inthe success of these vaccines while adju-vants do not appear to be necessary for theLPS-based ERM and Vibrio vaccines.

ESC is the major cause of mortality infarmed catfish in the southern UnitedStates and Asia. It was first identified in1976 and is caused by the bacteriumEdwardsiella ictaluri. The commercialvaccine available for ESC is based on abacterin preparation and involves a two-step vaccination programme. Catfish fry(eight to ten days old) are immersed in thevaccine prior to ponding. They are thengiven an oral boost of vaccine 30 daysbefore the start of the ESC season. Thevaccination of fish by this method hasproved effective in controlling ESCoutbreaks.

The one commercial viral vaccine forSVC has been available in the CzechRepublic and Slovenia since 1981 andalso includes an oil-based adjuvant foradministration by injection. SVC is caused

130 Fish vaccines

TA

BLE

13

Com

mer

cial

ly a

vaila

ble

fish

vacc

ines

Dis

ease

Man

ufac

ture

rC

ount

ry o

f or

gin

Prod

uct n

ame/

met

hod

of a

dmin

istr

atio

n

Fur

uncu

losi

sA

qua

heal

thU

nite

d K

ingd

omF

urog

en/in

ject

ion

Fur

uncu

losi

sA

qua

heal

thU

nite

d K

ingd

omF

urog

en B

/imm

ersi

on

Vib

riosi

sA

quac

ultu

re V

acci

nes

Ltd

Uni

ted

Kin

gdom

Aqu

aVac

Vib

rio/im

mer

sion

Vib

riosi

sB

iom

ed In

c,U

nite

d S

tate

sB

ioje

c 13

00 in

ject

ion

or im

mer

sion

(no

n-ad

juva

nted

)

Fur

uncu

losi

sA

quac

ultu

re V

acci

nes

Ltd

Uni

ted

Kin

gdom

Aqu

aVac

Fur

ovac

-imm

ersi

on/im

mer

sion

Fur

uncu

losi

sA

quac

ultu

re V

acci

nes

Ltd

Uni

ted

Kin

gdom

Aqu

aVac

Fur

ovac

-5/in

ject

ion

or im

mer

sion

(no

nad

juva

nted

)

Fur

uncu

losi

sA

quac

ultu

re V

acci

nes

Ltd

Uni

ted

Kin

gdom

Aqu

aVac

Fur

ovac

-5/in

ject

ion

(oil-

base

d ad

juva

nt)

Fur

uncu

losi

sB

iom

ed In

c.U

nite

d S

tate

sB

ioje

c 15

00/in

ject

ion

Fur

uncu

losi

sA

poth

eker

nes

Labo

rato

rium

Nor

way

Apo

ject

1-F

ural

/inje

ctio

n (o

il-ba

sed)

Fur

uncu

losi

s an

d co

ld-w

ater

vib

riosi

sA

poth

eker

nes

Labo

rato

rium

Nor

vvay

Apo

ject

2-F

ural

/inje

ctio

n (o

il-ba

sed)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis

and

vibr

iosi

sA

poth

eker

nes

Labo

rato

rium

Nor

way

Apo

ject

3-F

ural

/inje

ctio

n (o

il-ba

sed)

Col

d-w

ater

vib

riosi

s an

d vi

brio

sis

Apo

thek

erne

s La

bora

toriu

mN

orw

ayA

pova

x/im

mer

sion

(w

ater

-bas

ed)

Fur

uncu

losi

sE

wos

Aqu

a A

.S.

Nor

way

Lipo

gen

Mon

o/in

ject

ion

(oil-

base

d)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis

and

vibr

iosi

sE

wos

Aqu

a A

.S.

Nor

way

Lipo

gen

Trip

le/in

ject

ion

(oil-

base

d ad

juva

nt)

Fur

uncu

losi

sE

wos

Aqu

a A

.S.

Nor

way

Fur

ogen

/imm

ersi

on (

wat

er-b

ased

)

Fur

uncu

losi

sE

wos

Aqu

a A

.S.

Nor

way

Fur

ogen

/imm

ersi

on (

wat

er-b

ased

)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis

and

vibr

iosi

sE

wos

Aqu

a A

.S.

Nor

way

Fur

ogen

B/d

ip (

wat

er-b

ased

)

Fur

uncu

losi

sE

wos

Aqu

a A

.S.

Nor

way

Ora

vacc

F V

et/O

ral (

wat

er-b

ased

)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis

and

vibr

iosi

sIn

terv

et N

orbi

o A

.S.

Nor

way

Nor

vax

trip

le/in

ject

ion

(oil-

base

d)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis,

vib

riosi

sIn

terv

et N

orbi

o A

.S.

Nor

way

Nor

vax

trip

le+

IPN

/inje

ctio

n (o

il-ba

sed)

+in

fect

ious

pan

crea

tic n

ecro

sis

Col

d-w

ater

vib

riosi

s an

d vi

brio

sis

Inte

rvet

Nor

bio

A.S

.N

orw

ayN

orva

x vi

brio

sis/

dip

(wat

er-b

ased

)

Fur

uncu

losi

s, c

old-

wat

er v

ibrio

sis

and

vibr

iosi

sB

iom

ed In

c.U

nite

d S

tate

sB

ioje

ct 1

900/

inje

ctio

n (o

il-ba

sed

adju

vant

)

Ent

eric

sep

ticae

mia

of c

affis

hB

iom

ed In

c.U

nite

d S

tate

sB

iom

ed E

SC

/imm

ersi

on (

wat

er-b

ased

)

Ent

eric

red

mou

thB

iom

ed In

c.U

nite

d S

tate

sB

ioje

ct 1

100

imm

ersi

on (

wat

er-b

ased

)

Ent

eric

red

mou

thA

quac

ultu

re V

acci

nes

Ltd

Uni

ted

Kin

gdom

Aqu

a V

ac E

RM

/imm

ersi

on

Ent

eric

red

mou

thA

qua

heal

thU

nite

d K

ingd

omE

rmog

en/im

mer

sion

Spr

ing

vira

emia

of c

arp

Bio

veta

Cze

ch R

epub

licIn

ject

ion

by Rhabdovirus carpio (two serotypes) andthe disease is widespread in areas of carpculture.

AdjuvantsMost water-based vaccines for salmonidscontain either glucan- or aluminium-basedadjuvants which are partially effective andhave few side-effects. The latest trend insalmonid vaccines is towards oil-basedadjuvants which have been found to bemore effective. These do, however, need tobe injected into the fish and have manyside-effects, for example suspectedreduced growth rate, reduction of fertilityin broodstock, reduced carcass qualityowing to pigmentation at the site ofinjection, difficulty in administrationowing to high viscosity and an increasedhazard to human operators if accidentallyself-injected. The use of mineral oiladjuvants has been accepted in Norway,but a licence for their use has not yet beengranted in the United Kingdom.

Index of efficacyThe effectiveness of fish vaccines iscalculated in terms of relative percentsurvival (RPS) using the following formula:

percent vaccinate mortalityRPS = 1 x100

percent control mortality

Amend and Fender (1976) describe theadditional considerations which need tobe met.

During the development of these com-mercial vaccines many parameters im-portant to the success of vaccination havebeen determined, for example the route ofadministration, the length of protectionobtained, the size of the fish and thetemperature dependence of the immuneresponse (Johnson and Amend, 1983a and1983b; Johnson, Flynn and Amend, 1982aand 1982b; Tatner and Horne, 1985).

In general, intraperitoneal injectionappears to be the most effective route ofvaccination, although not the mostpractical. Much effort is therefore beingput into the development of immersionand oral vaccines.

DEVELOPMENT OF VACCINES

The primary considerations for anysuccessful vaccine for aquaculture are cost-effectiveness and safety. To accomplishthis the vaccine must provide long-termprotection against the disease under theintensive rearing conditions found oncommercial fish farms. Consideration mustbe given to all the serotypic variants of thedisease agent, the time and age at whichthe animal is most susceptible to disease,the route of administration and the methodof vaccine preparation (i.e. killed, at-tenuated, subunit, recombinant).

All the commercial vaccines currentlyavailable comprise inactivated (killed)disease agents. When that approach failedin the development of vaccines, par-ticularly viral vaccines, live attenuatedvaccines were developed.

Whenever a live vaccine is used there isalways concern that the attenuated strain(usually the result of a gene deletion) mayback-mutate and revert to the virulent wildtype. Many of the successful vaccinesagainst viral diseases of humans (e.g.rubella, measles, poliomyelitis) and indomestic animals (e.g. rabies, distemper)are attenuated organisms. The licensing ofsuch vaccines may, however, prove to bevery difficult in aquaculture. An alternativeapproach has been to prepare subunitvaccines using recombinant technology,where the specific components of thedisease-causing agents are isolated and,following amplification, used in vaccines.To increase the amount of antigenavailable, amplification is acheived bycloning the genes coding for specificantigens and incorporating them into

132 Fish vaccines

bacterial DNA, where they are expressed.Using fermentation technology for thegrowth of bacterial cells, expressed"foreign" proteins (antigens) can beproduced in bulk.

Bacterial vaccinesNumerous fish vaccines are currently beingdeveloped, although some of them withonly limited success. For example,vaccination would be an ideal approachfor controlling bacterial kidney disease(BKD) because it is widespread and itscontrol by means of chemotherapy isunsatisfactory. Experimental vaccination ofcoho salmon (Oncorhynchus kisutch) andsockeye salmon (Oncorhynchus nerka)indicated that agglutinating antibodieswere produced but the response was slowto develop and these antibodies werenot protective (Evelyn, 1971; Evelyn,Ketcheson and Prosperi-Porta, 1984;Baudin-Laurencin, Vigneulle and Mevel,1977). Kaattari et al. (1987 and 1988)investigated potential BKD vaccinescontaining the highly immunogenic Vibriovaccine components as an adjuvant;however, the results were inconsistent.Paterson, Desautels and Weber (1981)and McCarthy, Croy and Amend (1984)achieved the most promising results inAtlantic salmon and rainbow trout, whereprotection was conferred against a naturalchallenge and experimental challenge,respectively. One vaccine was adjuvantedwith Freund's complete adjuvant while theother was pH-lyzed and non-adjuvanted.Other researchers (Bruno and Munro, 1984;Sakai, Atsuta and Kobayashi, 1993) havebeen unable to reproduce these results, andso the search for a BKD vaccine continues.The preparations tested so far appear to beinadequate. This may be because theycontain inappropriate antigens and, which-ever protective antigens are present, areonly weakly immunogenic or responses tothem are suppressed by other antigens.

It has been reported that certain com-ponents of the extracellular proteins (ECP)from Renibacterium salmoninarum, forexample the 57 kDa protein, are immuno-suppressive (Turaga, Wiens and Kaattari,1987). These antigens must therefore beomitted (or modified) from any successfulvaccine preparation. Hastings and Ellis(1988) showed that rainbow trout onlyresponded to five out of 30 ECP com-ponents from A. salmonicida, while rabbitsresponded to 15. Thus, it is important toestablish which antigens the fish respondto and whether these are protective.Current research in the United Kingdomand Canada is based on the purification ofantigens from R. salmoninarum and itsextracellular products, and includesinvestigation into their effects on theimmune system prior to their incorporationinto a recombinant vaccine.

The other vaccines currently beingdeveloped against bacterial fish pathogensinclude atypical A. salmonicida, A. hydro-phila, Edwardsiella ictaluri, Flexibactercolumnaris, Streptococcus sp. and Pasteurellapiscicida.

Atypical furunculosis (caused byatypical non-pigmented A. salmonicida) isthe main bacterial disease affectingIcelandic fish farming. An experimentalvaccine given by injection has beendeveloped (B. Gudmundsdottir, personalcommunication). The protective antigenwas identified by passive immunization ofAtlantic salmon with antisera againstpurified antigen raised in rainbow troutand rabbits. A method to culture thebacterium with the maximal expression ofprotective antigen has also been developed.The vaccine results in 70 to 100 percentRPS when administered with mineral oilas an adjuvant to Atlantic salmon parr.Protection appears to last for 12 monthsand there is a good correlation betweenantibody production and protection. Fieldtrials have not yet been completed.

Vaccine manual 133

The antigenic diversity of A. hydrophilais the major limitation in the developmentof an effective vaccine and it seems likelythat a polyvalent preparation will benecessary. Although there is some doubtas to whether these bacteria ever act as aprimary pathogen, they do make asignificant contribution to the diseaseprocess in the fish they invade. As part ofthe development of a vaccine, researchgroups in Southeast Asia are currentlyinvestigating the immune response of carpand catfish to A. hydrophila antigens.

Moore, Eimers and Cardell (1990)demonstrated the feasibility of immuniz-ing channel catfish against columnarisdisease by immersion vaccination withformalin-inactivated Flexibac ter columnaris.However, a commercial vaccine has notyet been developed.

Streptococcal and Pasteurella spp. in-fections, hitherto reported principally inJapanese marine cultured fish populations,have recently become a problem inMediterranean countries. The Japaneseresearchers Iida, Wakabayashi and Egusa(1982) and Sakai et al. (1987) reported thatexperimental vaccination providedprotection from streptococcal infections,but a commercial vaccine is not yetavailable. There appear to be twoserotypes. A cocktail immersion vaccine toprevent Streptococcus sp. and V. anguillarumis currently being tested in Japan and fieldtrials are also being carried out in Italy(Ghittino, personal communication) on apotential Streptococcus vaccine.

Pasteurellosis has been of huge economicsignificance to yellow tail culture in Japanand, in 1991, was almost simultaneouslyreported as affecting sea bream in Italy,France and Greece. There appears to behomogeneity within this species ofbacterium, although little has beenpublished on its virulence factors andprotective antigens. A Pasteurella vaccineis currently being field tested and re-

searchers in Japan, the United Kingdomand Italy are continuing with developmentwork to determine an immune responseto the bacterium and investigate thepathogenesis of the disease.

Viral vaccinesThe development of a successful vaccineby culturing the causative agent underdefined conditions and subsequentlyinactivating it is not always feasible. It maybe necessary to select protective antigensand engineer a recombinant vaccinegenetically. This appears to be the case formost fish viral vaccines under devel-opment and requires the molecular cloningand expression of the viral genes in thebacterium Escherichia coli or the yeast,Saccharomyces cerevisae.

Viral haemorrhagic septicaemia (VHS)is a fish rhabdovirus responsible for severelosses in many continental European troutfarms. It is an economically devastatingdisease for the aquaculture industry, sinceit may affect all age groups. In the past,inactivated vaccines have been developed;however, these were immunogenic onlywhen given by injection (de Kinkelin,1988). Following the failure of inactivatedvaccines, several live vaccines weredeveloped and appeared to provideprotection against some of the serotypes.The vaccine strains were attenuated bysuccessive passage in cell lines, butsignificant mortality associated with thevirus was still observed and 2 to 13 percentof the fish died as a result of the vaccinationalone (de Kinkelin and Bearzotti-Le Berre,1981; Bernard, de Kinkelin and Bearzotti-Le Berre, 1983).

More recently, a subunit vaccine for VHShas been developed using recombinantDNA technology. The protective epitopesof VHS appear to be located on the surfaceglycoprotein of the virus, and antibodydirected against the viral glycoprotein-neutralized viral infectivity (de Kinkelin,

134 Fish vaccines

Vaccine manual 135

Bernard and Hattenberger-Baudovy, 1984;Lorenzen, Oleson and Vestergaard-Jorgensen, 1990). The gene coding for theVHS glycoprotein was cloned and expressedin bacteria (Thiery et al., 1990) and yeast(Lorenzen, 1991) to produce the glycoproteininexpensively and in large quantities.Preliminary results indicate that thismaterial induced protection followinginjection but no studies on immersionvaccination have been reported (Jorgensen,1992). Thus, for VHS, a recombinant subunitvaccine may be a promising way of pro-ducing an inexpensive safe vaccine in fish.

Recent results obtained by Lecoq-Xhonneux et al. (1993), however, suggestthat further research is needed prior tocommercial exploitation, since the yield ofantigens expressed was low and there wasa problem with the route of delivery of thevaccine (i.e. it is only effective whenintroduced intraperitoneally, not orally).

Several research groups are activelydeveloping vaccines for infectiouspancreatic necrosis virus (IPN). This is avery important pathogen of farm-rearedsalmonids. Mortality is highest in youngfish and survivors become life-longcarriers, thereby maintaining the virus inthe population by the continual sheddingand transmission of the disease. The virusis serotypically heterogenous and hastwo major serotypes, the first of whichcomprises nine subtypes which are allpathogenic for salmonid fish.

Several inactivated IPN vaccines havebeen tested but only vaccine administeredby injection-induced protection. Thepreferred route of immersion (sinceinfected fish are usually small) wasineffective (Dorson, 1977). This was alsofound when individual virion poly-peptides were tested. The virion proteinsappeared to have lost their antigenicityfollowing disruption with sodium dodecylsulphate, urea and acetic acid (Hill, Dorsonand Dixon, 1980).

Studies by Bootland, Dobos andStevenson (1990) suggest that the age andsize of the fish at the time of immunizationare important factors in the developmentof protective immunity. They reported thatonly fry immunized at two to three weeksafter hatching were protected. It appearsthat very young fish are capable ofresponding to vaccination but that growthrates must also be considered. These fishresponded while in a slow weight-gainingphase.

Two attenuated IPN vaccines have alsobeen tested and these appear to haveprovided protection when applied byimmersion (Dorson, Castric and Torchy,1978; McAllister, 1984). Further devel-opment was not pursued. Instead, as withother viral vaccines, many researchers arefollowing the route of recombinant subunitvaccines for IPN (Havarstein et al., 1990;Lawrence et al., 1989; Hah, Park and Jeong,1992). In these studies, the major capsidprotein VP2 has been identified as thevirion protein responsible for inducingprotective immunity in fish.

The drawbacks of killed and attenuatedvaccines (Leong, Fryer and Winton, 1988)have led research efforts to develop aneffective subunit vaccine for infectioushaematopoietic necrosis (IHN) virus(Leong et al., 1992). During these studies,monoclonal antibodies to the virion proteinswere produced, enabling characterizationof the fish immune response to viral antigensand the development of in situ hybridizationprobes to detect the virus in vaccines. IHNis caused by a rhabdovirus and produces asevere disease among fry and juveniles ofsusceptible species of salmonid fish.Engelking and Leong (1989) demonstratedthat the viral glycoprotein purified fromone isolate could induce protectiveimmunity in fish to a wide variety of IHNvirus isolates from different geographicallocations and different fish species. Inconsequence, the viral glycoprotein gene

from one strain was cloned and a re-combinant vaccine prepared. Furtherstudies have identified a specific immuno-dominant region in the middle of the gene(Xu et al., 1991; Mourich and Leong, 1991).A field trial (immersion immunization) ofthe IHN subunit vaccine was undertakenin Idaho, United States, and showed it tobe very effective in inducing protectiveimmunity. A commercial vaccine, howeveris not currently available.

Two other viral vaccines currently underdevelopment for non-salmonid species arethose for SVC and channel catfish virus(CCV). Despite a commercial inactivatedSVC vaccine being available in the CzechRepublic and Slovenia, attenuated vaccineshave also been developed (Fijan et al., 1977)and appear to offer protection. However,although the live vaccine in particularprovided excellent protection, the vaccin-ated fish appeared to become asymptom-atic carriers of the virus. When these fishwere reared beside unvaccinated stock, anSVC outbreak occurred. A subunit vaccineis therefore an attractive alternative. Areliable challenge model for the diseasealso needs to be developed so that vaccinepotency testing can be performed understandard conditions.

CCV, caused by Herpesvirus ictaluri, is anacute, highly infectious disease of juvenilechannel catfish (Ictalurus punctatus). Theonset of the disease is sudden and 100percent mortality can occur within ten days.Once fish are infected with the virus,survivors become carriers for their entirelives. At present, CCV outbreaks areprevented by management strategies suchas the use of resistant strains of channelcatfish and virus-free broodstock for thestocking of fish farms. Vaccination wouldseem to be the ideal strategy in areas whereCCV is endemic. An attenuated livevaccine was developed by serial passagein a tissue culture cell line derived from thewalking catfish (Noga and Hartman, 1981).

This was protective both by injection andimmersion. Booster vaccination appeared toincrease the efficacy of the vaccine greatly(Walczak, Noga and Hartman, 1981). Thisvaccine is not available commercially andhas not been licensed, probably because ofconcerns regarding the risk of reversion tovirulence and the possible establishmentof carriers among vaccinated fish. If carrierfish are generated as a result of vaccination,these fish can no longer be certified asvirus-free broodstock. Attempts to developa killed CCV vaccine have been hinderedby its poor immunogenicity (Plumb, 1973).A subunit vaccine for CCV was recentlydeveloped (Awad, Nusbaum and Brady,1989). The envelope components of thevirus appear to be capable of inducingprotective immunity and, once these havebeen determined more specifically, aneffective recombinant vaccine may bepossible.

Parasitic vaccinesMonoclonal antibody (MAb) probes areproving to be useful tools in the devel-opment of vaccines against sea lice(Lepeophtheirus salmonis) and the protozoaninfection proliferative kidney disease(PKD) in salmonids. A vaccine against sealice remains a high priority for theaquaculture industry. Monoclonal anti-bodies were used to produce extracts fromlice and to select individual antigens froma sea louse recombinant DNA library(Andrade-Salas et al., 1993). These clonesform the basis of experimental vaccineswhich are currently under investigation inthe United Kingdom.

PKD is the most economically damagingdisease affecting the trout industry inEurope. MAb probes have been producedagainst PKX, the causative agent, in aneffort to map out the antigens on theparasite's surface (Adams, Richards andMarin de Mateo, 1992) and have shownthat PKX from different species of fish and

136 Fish vaccines

different geographical locations sharecommon antigens (Marin de Mateo et al.,1993). At present, studies are under way toculture the parasite and investigateprotective antigens.

A successful parasite vaccine has beendescribed by Woo and Li (1990) workingin Canada. Live attenuated Cryptobiasalmositica, a pathogenic haemoflagellatecausing cryptobiosis, was injected intra-peritoneally into rainbow trout but thestrategy for the delivery of the vaccine hasstill to be resolved.

Numerous other parasites causesignificant losses to the aquacultureindustry, for example Icthyobodo necatrix,Ichthyopthirius multifiliis and Trichodina sp.,but their control by vaccination has not yetbeen attempted.

Future prospectsFish vaccines have become much moresophisticated in recent years, with the trendbeing for the development of subunitrecombinant vaccines in preference to theoriginal killed whole-cell preparations.This has been necessary because thesimpler approach did not succeed for manyof the important diseases and attempts atattenuated vaccines in general have notbeen encouraging from a safety pointof view. The cost of producing such"high-tech" vaccines must however beconsidered and the costs and benefitsweighed up prior to commercialization.

Direct DNA vaccination has recentlybeen successfully performed in cattleagainst infectious bovine rhinotracheitisvirus (Cox, Zamb and Basiuk, 1993).Muscles can apparently take in naked DNAand then express it for long periods (Wolffet al., 1992). DNA vaccination may proveto be a more cost-effective method ofvaccinating fish against viral infections inthe future.

As the range of cultured speciesincreases, there is also a need for basic

research on the immune responses of newspecies as well as for studies on geneticallyresistant species in parallel with vaccinedevelopment.

Immunostimulants. An approach recentlytaken by the aquaculture industry is theuse of immunostimulants, either bythemselves to counteract stress-inducedimmunosuppression in fish, or in vaccinesas adjuvants. They may serve to boost theimmune system in the short term andtherefore have great potential. One groupreceiving a great deal of attention at presentis the P-glucans. They apparently functionas immunostimulants of the non-specificdefence mechanism of fish (Robertsen,Engstad and Jorgensen, 1994).

Glucans are major structural poly-saccharides from fungal and yeast cellwalls, composed of glucose units whichare held together through 13-1,3 and 3-1,6bonds (Rosenberg, 1976; Duffus, Levi andManners, 1982). Both soluble P-glucans(e.g. scleroglucan, schizophyllan andlentinan) and microparticulated P-glucansfrom yeast (M-Glucan, Macroguard®) havebeen shown to function as immuno-stimulators in fish (Robertsen, Engstad andJorgensen, 1994). Intraperitoneal injectionsof glucans have resulted in enhanceddisease protection of carp to Edwardsiellatarda infection (Yano, Mangindaan andMatsuyama, 1989; Yano, Matsuyama andMangindaan, 1991) and yellow tail toStreptococcus spp. (Matsuyama, Mangin-daan and Yano, 1992). Yeast glucan hasbeen found to increase disease resistancein Atlantic salmon to Vibrio anguillarum,V. salmonicida, Yersinia ruckeri and Aero-monas salmonicida (Robertsen et al., 1990;Robertsen, Engstad and Jorgensen, 1994)and in channel catfish to Edwardsiellaictaluri (Chen and Ainsworth, 1992).

The exact mode of action of 'glucansremains unclear, but enhanced protectionagainst microbial pathogens, observed

/accine manual 137

after administering glucans to fish, cor-relates with increased blood lysozyme andcomplement activities and enhancedphagocytosis and killing of bacteria byheadkidney macrophages (Yano, Mangin-daan and Matsuyama, 1989; Engstad,Robertsen and Frivold, 1992; Chen andAinsworth, 1992; Matsuyama, Mangin-daan and Yano, 1992; Jorgensen et al., 1993).It is believed that such macrophages mayhave an essential role to play in immuno-stimulation, since it has been shown thatAtlantic salmon macrophages possessreceptors for f3-glucan (Robertsen, Engstadand Jorgensen, 1994).

Robertsen, Engstad and Jorgensen (1994)suggested that non-specific defence maybe more effective against opportunisticpathogens than specific bacteria. Glucans,therefore, probably benefit most duringstress-related conditions, such as handling,transportation or smoltification, whenan increased susceptibility to diseaseoccurs. Glucan-supplemented diets andfurunculosis vaccines are now com-mercially available. However, the dose, thetype of glucan and the route of admini-stration must all be considered whenexamining the effects of glucans on diseaseresistance (Ainsworth, Mao and Boyle,1994). Further studies are required tostandardize these parameters.

BIBLIOGRAPHY

Adams, A., Richards, R.H. & Marin deMateo, M. 1992. Development of mono-clonal antibodies to PKX, the causativeagent of proliferative kidney disease. J.Fish Dis., 15: 515-520.

Ainsworth, A.J., Mao, C.P. & Boyle, C.R.1994. Immune response enhancementin channel catfish, Ictalurus puncta tus,using f3-glucan from Schizophyllumcommune. In J.S. Stolen & T.C. Fletcher,eds. Modulators offish immune responses,Vol. 1, p. 67-81. Fair Haven, N.J., USA,

SOS Publications.Amend, D.F. & Fender, D.C. 1976. Uptake

of bovine serum albumin by rainbowtrout from hyperosmotic solution: amodel for vaccinating fish. Science, 192:793-794.

Andrade-Salas, O., Sommerville, C.,Wootten, R., Turnbull, T., Melvin, W.,Amezaga, T. & Labus, M. 1993. Immuno-histochemical screening and selectionof monoclonal antibodies to salmonlouse, Lepeophtheirus salmonis (Kroyer,1837). In E. Boxshall & D. Defaye, eds.Pathogens of wild farmed fish: sealice, p.323-331. Chichester, UK, Horwood (Ellis).

Awad, M.A., Nusbaum, K.E. & Brady, Y.J.1989. Preliminary studies of a newlydeveloped subunit vaccine for channelcatfish disease. J. Aquat. Anim. Health,1: 233-237.

Baudin-Laurencin, F., Vigneulle, M. &Mevel, M. 1977. Premières observationssur la corynebactériose des salrnonidesen Bretagne. Bull. Off. int. Epiz., 87:505-507.

Bernard, J., de Kinkelin, P. & Bearzotti-LeBerre, M. 1983. Viral haemorrhagicsepticaemia of trout: relation betweenthe G polypeptide, antibody productionand protection of the fish followinginfection with the F25 attenuated variantstrain. Infect. Immun., 39: 7-14.

Bootland, L.M., Dobos, P. & Stevenson,R.M.W. 1990. Fry age and size effectson immersion immunization of brooktrout, Salvelinus fontalis Michelle, againstinfectious pancreatic necrosis virus. J.Fish Dis., 13: 113-125.

Bruno, D.W. & Munro, A.L.S. 1984.Epidemiological studies of bacterialkidney disease in Scotland. In Int. Sem.Fish Pathology for the 20th Anniversary ofthe Japanese Society of Fish Pathology, p.57-58. Tokyo.

Busch, R.A. 1982. Enteric redmouth disease(Yersinia ruckeri). In D.P. Anderson, M.Dorson & P. Dubourget, eds. Antigens

138 Fish vaccine,

offish pathogens, p. 201-223. Lyon, France,Marcel Merieux.

Chen, D. Sr Ainsworth, A.J. 1992. Glucanadministration potentiates immunedefence mechanisms of channel catfish,Ictalurus punctatus. J. Fish Dis., 15: 295-304.

Cox, G.J.M., Zainb, T.J. & Basiuk, L.A. 1993.Bovine herpes virus I: immune responsesin mice and cattle injected with plasmidDNA. J. Virol., 67(9): 5664-5667.

de Kinkelin, P. 1988. Vaccination againstviral haemorrhagic septicaemia. In A.E.Ellis, ed. Fish vaccination, p. 172-192.London, Academic.

de Kinkelin, P. & Bearzotti-Le Berre, M.1981. Immunization of rainbow troutagainst viral haemorrhagic septicaemia(VHS) with a thermoresistant variantof the virus. Dev. Biol. Stand., 49: 431-439.

de Kinkelin, P., Bernard, J. & Hattenberger-Baudovy, A.M. 1984. Immunizationagainst viral diseases occurring in cold water.

Symposium on Fish Vaccination, Paris,Office International des Epizooties.

Dorson, M. 1977. Vaccination trials ofrainbow trout fry against infectiouspancreatic necrosis. Bull. Off. int. Epiz.,57: 405-406.

Dorson, M., Castric, J. & Torchy, C. 1978.Infectious pancreatic necrosis virus ofsalmonids: biological and antigenicfeatures of a pathogenic strain and anon-pathogenic variant selected in RTG-2 cells. J. Fish Dis., 1: 309-320.

Duff, D.C.B. 1942. The oral immunisationof trout against Bacterium salmonicida.J. Immunol., 44: 87-94.

Duffus, I.H., Levi, C. & Manners, D.J. 1982.Yeast cell-wall glucans. Adv. Microbiol.Physiol., 23: 151-181.

Engelking, H.M. & Leong, J.C. 1989. Theglycoprotein of infectious hematopoieticnecrosis virus elecit neutralizing anti-body and protective responses. VirusRes., 13: 213-230.

Engstad, R.E., Robertsen, B. & Frivold, E.1992. Yeast glucan induces increase inlysozyme and complement-mediatedhaemolytic activity in Atlantic salmonblood. Fish Shellfish Immunol., 2: 287-297.

Evelyn, T.P.T. 1971. The agglutinin responsein sockeye salmon vaccinated intra-peritoneally with a heat-killed prepar-ation of the bacteria responsible forsalmonid kidney disease. J. Wildl. Dis.,7: 328-335.

Evelyn, T.P.T., Ketcheson, J.E. & Prosperi-Porta, L. 1984. On the feasibility ofvaccination as a means of controllingbacterial kidney disease in Pacificsalmon. In Abstr.: Int. Conf Biology ofPacific Salmon, 5-12 September 1984.Victoria / Agassiz, British Columbia,Canada.

Fijan, N., Petrinec, Z., Stanel, Z., Kezic, N.& Teskeredzic, E. 1977. Vaccination ofcarp against spring viraemia. Compari-son of intraperitoneal and peroralapplication of live virus to fish kept inponds. Bull. Off int. Epiz., 92: 1055-1068.

Fryer, J.L., Rehovec, J.S. & Garrison, R.L.1978. Immunisation of salmonids forcontrol of vibriosis. Mar. Fish. Rev., 40:20-23.

Hah, Y.C., Park, J.W. Sr Jeong, G. 1992.Neutralize epitope of DRT serotype ofinfectious pancreatic necrosis virus(IPNV) isolated in Korea. In Proc. Int.Symp. Infectious Viruses in Fish, 9-11October 1992, p. 13-22. Seoul.

Hastings, T.S. & Ellis, A.E. 1988. The humoralimmune response of rainbow trout andrabbits to Aeromonas salmonicida extra-cellular products. J. Fish Dis., 11: 147-160.

Harvarstein, L.S., Kalland, K.H., Christie,K.E. & Endressen, C. 1990. Sequence ofthe large double-stranded RNA segmentof the NI strain of infectious pancreasnecrosis virus: a comparison withoutBirnaviridae. J. Gen. Virol., 71: 299-308.

Vaccine manual 139

140 Fish vaccines

Hill, B., Dorson, M. & Dixon, P.F. 1980.Studies on immunization of trout againstIPN. In W. Ahne, ed. Fish diseases. ThirdCOPRAQ-Session, p. 29-36. Berlin,Springer.

Hirst, I.D. & Ellis, A.E. 1994. Iron-regulatedouter membrane protein of Aeromonassalmonicida are important protectiveantigens in Atlantic salmon againstfurunculosis. Fish Shellfish Immunol., 4(1):29-46.

Iida, T., Wakabayashi, H. & Egusa, S. 1982.Vaccination for control of streptococcaldisease in cultured yellowtail. FishPathol., 16: 201-206.

Johnson, K.A. & Amend, D.F. 1983a. Efficacyof Vibrio anguillarurn and Yersinia ruckeribacterins applied by oral and analintubation of salmonids. J. Fish Dis., 6:473-476.

Johnson, K.A. & Amend, D.F. 1983b.Comparison of efficacy of several deliv-ery methods using Yersinia ruckeribacterins on rainbow trout, Salmogairdneri Richardson. J. Fish Dis., 6:

331-336.Johnson, K.A., Flynn, J.K. & Amend, D.F.

1982a. Duration of immunity insalmonids vaccinated by directimmersion with Yersinia ruckeri andVibrio anguillarum. J. Fish Dis., 5: 207-213.

Johnson, K.A., Flynn, J.K. & Amend, D.P.1982b. Onset of immunity in salmonidfry vaccinated by direct immersion inVibrio anguillarum and Yersinia ruckeri.J. Fish Dis., 5: 197-205.

Jorgensen, P.E. 1992. Recent advances insurveillance and control of viral haemor-rhagic septicaemia (VHS) in trout. In T.Kimura, ed. Proc. OJI Int. Symp. SalmonidDiseases, p. 60-71. Sapporo, Japan,Hokkaido University Press.

Jorgensen, J.B., Sharp, G.J.E., Secombes,C.J. & Robertsen, B. 1993. Effects of ayeast cell wall glucan on the bactericidalactivity of rainbow trout macrophages.Fish Shellfish Immunol., 3(4): 267-277.

Kaattari, S.L., Holland, N., Turaga, P. &Wiens, G. 1987. Development of avaccine for bacterial kidney disease.Bonneville Power Administration Project86-45. In Annual Report 1986. Portland,Oreg., USA, Bonneville Power Admini-stration.

Kaattari, S.L., Chen, D., Turaga, P. & Wens,G. 1988. Development of a vaccine forbacterial kidney disease. BonnevillePower Administration Project 86-46. InAnnual Report 1987. Portland, Oreg., USA,Bonneville Power Administration.

Karunasagar, I., Rosalind, G. & Karunasagar,I. 1991. Immunological response of theIndian major carps to Aeromonashydrophila vaccine. J. Fish Dis., 14: 413-417.

Kawai, K., Kusuda, R. & Itami, T. 1981.Mechanisms of protection in ayu orallyvaccinated for vibriosis. Fish Pathol., 15:257-262.

Lawrence, W.R., Nagy, E., Duncan, R., Krell,P. & Dobos, P. 1989. Expression in E.coli of the major outer capsid protein ofinfectious pancreatic necrosis virus. Gene,79: 369-374.

Lecoq-Xhonneux, F., Thiery, M., Dheur, I.& de Kinkelin, P. 1993. A recombinantviral haemorrhagic septicaemia virus(VHS) glycoprotein expressed in insectcells induces protective immunity inrainbow trout. Paper presented atEuropean Association of FishPathologists 6th Int. Conf., Diseases ofFish and Shellfish, Brest, France.

Leong, J.C. & Fryer, J.L. 1993. Viral vaccinesfor aquaculture. Ann. Rev. Fish Dis.,p. 225-240.

Leong, J.C., Fryer, J.L. & Winton, J.R. 1988.Vaccination against infectious hemo-topoietic necrosis virus. In A.E. Ellis,ed. Fish vaccination, p. 193-202. London,Academic.

Leong, J.C., Anderson, E., Bootland, L., Chen,L., Drolet, B., Engelking, H.M., Mason,C., Mourich, D., Johnson, K., Trobridge,C. & Wirkkula, J. 1992. Biotechnological

approaches to the development ofsalmonid fish vaccines. In T. Kimura,ed. Proc. OJI Int. Symp. Salmonid Diseases,p. 250-255. Sapporo, Japan, HokkaidoUniversity Press.

Lorenzen, N. 1991. Viral haemorrhagicsepticaemia virus of rainbow trout:immunological investigations in relationto vaccine development. University ofCopenhagen. (Ph.D. thesis)

Lorenzen, N., Oleson, J.F. & Vestergaard-Jorgensen, P.E. 1990. Neutralisation ofEgtved virus pathogenicity to cellcultures and in fish by monoclonalantibodies to the viral G protein. J. Gen.Virol., 71: 561-567.

Marin de Mateo, M., Adams, A., Richards,R.H., Castagnaro, M. & Hedrick, R.P.1993. Monoclonal antibody and lectinprobes recognise developmental andsporogonic stages of PKX, the causativeagent of proliferative kidney disease inEuropean and North American salmonidfish. Dis. Aquat. Organ., 15: 23-29.

Matsuyama, H., Mangindaan, R.E.P. & Yano,T.1992. Protective effects of shizophyllanand scleroglucan against Streptococcussp. infection in yellow tail (Seriolaquinqueradiata). Aquaculture,101: 197-203.

McAllister, P.E. 1984. Infectious pancreaticnecrosis immunisation. (AmericanFisheries Society) Fish Health Newsl.,12: 6.

McCarthy, D.H., Croy, T.R. & Amend, D.F.1984. Immunisation of rainbow trout,Salmo gairclneri Richardson, againstbacterial kidney disease: preliminaryefficacy evaluation. J. Fish Dis., 7: 65-71.

Moore, A.A., Eimers, M.E. & Cardell, M.A.1990. Attempts to control Flexibactercolumnaris epizootics in pond-rearedchannel catfish by vaccination. J. Aquat.Anim. Health, 2: 109-111.

Mourich, D.V. & Leong, J.C. 1991. Mappingof the immunogenic regions of the IHNVglycoprotein in rainbow trout and mice.In Proc. 2nd Int. Symp. on Viruses of Lower

Vertebrates, p. 93-100. Corvallis, Oreg.,USA,

Noga, E.J. & Hartman, J.X. 1981.Establishment of walking catfish (Clariasbatrachus) cell lines and developmentof a channel catfish (Ictalarus punctatus)virus vaccine. Cana. J. Fish, Aquat. Sci.,38: 925-930.

Paterson, W.D., Desautels, D. St Weber,J.M. 1981. The immune response ofAtlantic salmon, Salmo salar L., to thecausative agent of bacterial kidneydisease, Renibacterium salmoninarum. J.Fish Dis., 4: 99-111.

Plumb, J.A. 1973. Neutralisation of channelcatfish virus by serum of channel catfish.J. Wadi. Dis., 9: 324-330.

iloberts, R.J., Willoughby, L.G. & Chinabut,S. 1993. Mycotic aspects of epizooticulcerative syndrome (EUS) of Asianfishes. J. Fish Dis., 16: 169-183.

Robertsen, B., Engstad, R. & Jorgensen,J.B. 1994. j3-glucans as immunostimulantsin fish. In J.S. Stolen & T.C. Fletcher,eds. Modulators offish immune responses,Vol. 1, p. 83-99. Fair Haven, N.J., USA,SOS Publications.

Robertsen, B., Rorstad, G., Engstad, R. &Raa, J. 1990. Enhancement of non-specificdisease resistance in Atlantic salmon,Salmo salar L., by a glucan from Saacharo-myces cervisae cell walls. J. Fish Dis., 13:391-400.

Rosenberg, R.F. 1976. The cell wall. In J.E.Smith & D.R. Barry, eds. The filamentousfungi, Vol 2, Biosynthesis and metabolism,p. 328-334. London, Arnold (Edward).

Sakai, M., Atsuta, S. & Kobayashi, M. 1993.The immune response of rainbow trout(Oncorhynchus mykiss) injected with fiveRenibacterium salmoninarum bacterins.Aquaculture, 113: 11-18.

Sakai, M., Kubota, R., Atsuta, S. &Kobayashi, M. 1987. Vaccination ofrainbow trout Salmo gairdneri againstbeta-hemolytic streptococcal disease.Nippon Suisan Gakkaishi, 53: 1373-1376.

Vaccine manual 141

Tatner, M.F. & Horne, M.T. 1985. The effectsof vaccine dilution length of immersiontime and booster vaccinations on theprotection levels induced by directimmersion, vaccination of brown trout,Salmo trutta, with Yersinia ruckeri (ERM)vaccine. Aquaculture, 46: 11-18.

Tebbit, G.L., Erikson, J.D. & Vande Water,R.B. 1981. Development and use ofYersinia ruckeri bacterins to control entericredmouth disease. In W.V. Leetown, ed.International Symposium on Fish Biologies:Serodiagnos tics and Vaccines. Dev. Biol.Stand., 49: 395-401.

Thiery, M., Lecoq-Xhonneux, F., Dheur, I.,Renard, A. & de Kinkelin, P. 1990.Molecular cloning of the mRNA codingfor the G protein of the viral haemor-rhagic septicaemia (VHS) of salmonids.Vet. Microbioassay, 23: 221-226.

Turaga, P., Wiens, G. & Kaattari, S.L. 1987.Bacterial kidney disease: the potentialrole of soluble protein antigen(s). J. FishBiol., 31(supplement A):191-194.

Vigneulle, M., Brevil, G., Ceschia, G. &Blanch, A. 1993. Vaccination trials ofsea bass (Dicentrarchus labrax) againsttwo serotypes of Vibrio anguillarum.European Association of Fish Pathol-ogists 6th Int. Conf., Diseases of Fishand Shellfish, Brest, France, AbstractNo. 66.

Walczak, E.M., Noga, E.J. & Hartman, J.X.1981. Properties of a vaccine for channelcatfish virus disease and a method ofadministration. Dev. Biol. Stand., 49:419-429.

Wolff, A., Ludtke, J.J., Acsadi, G., Williams,P. & Jani, A. 1992. Hum. Mol. Genet., 1:363-369.

Woo, P.T.K. & Li, S. 1990. In vitro attenuationof Cryptobia salmositica and its use as alive vaccine against crytobiosis in Onco-rhynchus mykiss. J. Parasitol., 76: 752-755.

Yano, T., Mangindaan, R.E.P. & Matsuyama,H. 1989. Enhancement of the resistanceof carp Cyprinus carpio to experimental

Edwardsiella tarda infection, by some P-1,3 glucans. Nippon Suisan Gakkaishi, 55:1815-1819.

Yano, T., Matsuyama, H. & Mangindaan,R.E.P. 1991. Polysaccharide-inducedprotection of carp, Cyprinus carpio L.,against bacterial infection. J. Fish Dis.,14: 577-582.

Xu, L., Mourich, D.V., Engelking, H.M.,Ristou, S., Arnzen, J. & Leong, J.C.1991. Epitope mapping and charac-terization of the infectious hematopoieticnecrosis virus glycoprotein, using fusionproteins synthesized in Escherichia coli.J. Virol., 65: 1611-1615.

142 Fish vaccines

Harmonization and standardization ofrequirements for veterinary vaccines havetwo important goals:

*to keep the quality of vaccines at a highlevel;to prevent or reduce potential tradebarriers.

Specialized groups established by go-vernments or groups of governments, suchas the European Union (EU), formulatelegislation and requirements on the basisof the latest technology to ensure the highquality and safety of these products.

Every new vaccine developed in acountry or imported from another has tobe registered according to the legislationand requirements in force in the countrywhere the vaccine will be used.

The registration file must report on theresults of experiments directly related tothe product itself (Pensaert, 1992) andinclude:

*an analytical section, characterizingthe different components and thesubstrates on which the antigens areproduced;a toxico-pharmacological note, indicat-ing the absence of remaining patho-genicity or abnormal toxicity;a section on the product's efficacy,describing clinically oriented ex-periments with relation to immunityand protection (degree, duration, etc.).

The analytical section should include thecontrols which have been carried out bothon the product during manufacturing and

eand internati nai o

quaiity

ciIÎoriFI Epizooiiiesnizati ns in vacie

ndarolizationM. Truszczynski and J. Blancou

on the final product. These controls areperformed in the producers' laboratory orin a laboratory of their choice.

The registration files are examined byexperts appointed by the ministry of healthand, particularly, the national commissionfor veterinary drugs. The expert whoexamines the analytical section mayrequire additional controls if thosedescribed in the original file are in-complete. These additional controls mayhave to be carried out in an officiallyaccepted laboratory, for instance if thevaccine originates from a country for whichthere is little information on the quality ofthe local control system.

When the registration data are adequate,the commission for veterinary drugs givesa favourable assessment to the minister ofhealth and the product is registered formarketing.

After registration, every productionbatch of the registered vaccine must becontrolled by the producer, both duringmanufacturing and at the final productstage, and these controls are the respon-sibility of an industrial specialist. Furtherofficial controls are not required for themajority of bacterial vaccines although, inmany cases, the bacterial strain used in thevaccine is controlled annually by theappropriate government institute. How-ever, all viral vaccines and also somebacterial vaccines (e.g. for brucellosis) mustundergo an official batch control bythe national veterinary institute or an

Vaccine manual 143

r

equivalent institution before release. Everyimported vaccine must undergo an officialbatch control in an officially acceptedlaboratory chosen freely by the importer.Additionally, a control is prescribed by thenational veterinary institute.

For batch controls, representativesamples are collected under the respon-sibility of the laboratory and according toinformation supplied by the producer withregard to the number of doses per batch,the homogeneity of the batch, conditionsof storage, etc.

In the laboratory, the samples areexamined to identify the active com-ponent(s), to ascertain the absence of con-taminating agents and to ensure theproduct's sterility, safety and potency. Forthese controls, the relevant monograph ofthe European Pharmacopoeia (see p. 149)is followed. If there is no monograph for aparticular vaccine, the tests described inthe original registration file must beapplied. For inactivated vaccines, thepotency test may be performed on thetarget species or on a laboratory animalspecies but, in the latter case, the relationbetween the potency in the laboratory andthat in the target species must be demon-strated by the producer. For live vaccines,potency is generally limited to quantitationof the organism or of the virus, usingmethods described in the original registra-tion file. If the results are satisfactory, thebatch will be given an official batch numberand be released on to the market.

Depending on the country or group ofcountries, there are varying degrees ofdifference in their legislation. With this inmind and in the spirit of internationalharmonization aimed at reducing orpreventing potential trade barriers, theUnited States Animal Health Institute(ALII) has proposed the formation of aninternational working group consisting ofitself, the European Federation for AnimalHealth (FEDESA), the EU's Committee for

Veterinary Medicinal Products, the UnitedStates Department of Agriculture andinternational industry and governmentofficials (Draayer, Hilsabeck and Miller,1992; Folkers, 1992; Watson, 1992). It issuggested that annual meetings be held inconjunction with either FEDESA or AHImeetings. Issues to be dealt with includethe standardization of test requirements,time-frames for implementation and otherissues that could reduce or prevent tradebarriers without adversely affectingproduct quality.

The main purpose of the working groupwould be to harmonize European andUnited States legislation as a necessary steptowards international standardization andthe prevention or reducion of potentialtrade barriers.

There are, in addition, internationalorganizations acting in parallel also withthe goal of international harmonizationand standardization of vaccines butindependently from governments and inmore general terms. These organizationsinclude the Office international desEpizooties (OIE), the World Health Orga-nization (WHO) and the Food and Agri-culture Organization of the United Nations(FAO) as well as the Pharmacopoeias (e.g.the European, United States and JapanesePharmacopoeias).

OFFICE INTERNATIONAL DES EPIZOOTIES

The OIE, whose purpose is to harmonizeand coordinate animal health activitiesat the international level, also contributesto the standardization of vaccine quality.The three principal aims of the OIE(Truszczynski and Blancou, 1992) are:

e The provision of information on animalhealth worldwide.

°International coordination of researchinto and control of certain animaldiseases.

e The harmonization of import andexport regulations for animals and

144 Role of OIE and international organizations in vaccine quality standardization

animal products at the internationallevel.

A measure of the development andgrowing international recognition of theOIE is that it was established by 28countries in 1924 while it now has 136member countries. Seventeen are from theAmericas, 42 from Africa, 43 from Europeand 34 from Asia. The OIE operates underthe authority of its International Com-mittee, formed by the delegates of themember countries, under the leadership ofan elected president.

The Central Bureau, located at OIE'sheadquarters in Paris (12 rue de Prony), isheaded by the Director-General, at presentDr J. Blancou. The Central Bureau imple-ments decisions of the InternationalCommittee and the different Commissions.

There are five Regional Commissions,covering Africa, the Americas, Asia, theFar East and Oceania, Europe and theMiddle East.

There are also Specialist Commissions forThe International Animal Health Code;Standards; Foot-and-Mouth Disease andother Epizootics; and Fish Diseases, includ-ing those of crustaceans and molluscs. Inaddition, there are three working groups:Animal Health Information Systems,Veterinary Drug Registration and Bio-technology.

The International Animal Health CodeCommission draws up animal healthrecommendations for the import andexport of animals and animal products andcontributes through the Animal Health Code,an important OIE publication, to inter-national harmonization in this area. TheStandards Commission establishes standardsfor diagnostic methods, including diagnos-tic biologicals, and for vaccines. The Footand Mouth Disease and other EpizooticsCommission contributes to the developmentand standardization of vaccines againstfoot-and-mouth disease and strategies forthe eradication or control of this as well as

other diseases. The Fish Diseases Commissionestablishes standards for diagnosticmethods and vaccines for fish diseases.

From this short description it can beconcluded that all the Specialist Com-missions are interested in the internationalharmonization and standardization notonly of vaccines but also of diagnosticmethods, including diagnostic biologicals.These activities relate to those importantdiseases included in the Lists A and Bprepared by the OIE (Truszczynski andBlancou, 1992).

List A includes contagious diseaseswhich spread rapidly and the scope ofwhich extends beyond national borders.These diseases have particularly serioussocio-economic or public health con-sequences and are of major importance inthe international trade of animals andanimal products.

List B includes contagious diseases whichare considered of socio-economic and/ orpublic health importance within countriesand which, naturally, are also of signif-icance to the international trade of animalsand animal products.

Among these Specialist Commissions,the one most closely connected with thestandardization of diagnostic methods andvaccine quality is the Standards Com-mission, whose role includes participationin the standardization of biologicals,including vaccines used for prophylacticpurposes for List A and B diseases.

To accomplish this goal, the StandardsCommission undertakes the followingactivities:

° regular updating of the OIE manual ofstandards for diagnostics tests andvaccines;

the organization of reference labor-atories for several diseases in Lists Aand B;

e the establishment of internationalstandards for diagnostic tests andvaccines.

Vaccine manual 145


OIE manual of standards for diagnostic testsand vaccines

The purpose of the OIE manual (OIE, 1992)is to provide a uniform approach to thediagnosis of important animal diseases andto the production and control of biologicalproducts, mainly vaccines used in thecontrol of List A and B diseases. This isimplemented by the presentation in themanual of standard methods for laboratorydiagnosis of diseases and for the pro-duction and control of vaccines and otherbiological products for veterinary use inlaboratories all over the world.

The first edition of the manual waspublished in three volumes in 1989-1991,each covering about 30 List A or B diseases.After revision and additional editorial work,a single combined volume was publishedas the second edition of the manual in 1992.It contains information on 15 List A diseasesand 75 List B diseases. In future, theStandards Commission plans to produce anew updated edition every four years. The1992 edition of the manual has beendistributed worldwide, recommending:

e "prescribed tests" which should beused for diagnosis; andrequirements for vaccines, in relationto List A and B diseases.

This will undoubtedly help lead to aninternationally unified approach to thediagnosis and quality control of vaccines.

Organization of OIE reference laboratoriesfor List A and B diseases

The standardization of "prescribed tests"and the production and quality control ofveterinary vaccines can only be achievedwhen the necessary standards are madeavailable. The goal of standardization andavailability of standards is intended to beachieved through the OIE-organizedreference laboratories.

These laboratories must fulfil a specificfunction or range of functions at aninternationally recognized level related to

the standardization of diagnostic methodsand vaccines applied in the prophylaxisand control of List A and B diseases,especially those of greatest importance andthose causing high economic losses. Thedesignation of these laboratories is pro-posed by the OIE Standards Commissionand ratified by the OIE InternationalCommittee. Designation does not implythat financial support will be given by theOIE.

The functions and responsibilities of theOIE reference laboratories are as follows:

e the provision of a centre of excellencein a designated activity;the standardization of methodology;

e the storage and distribution of standardantisera, antigens and other importantreagents;the development of new methods;the provision of consultant assistanceto the 01E;training in a designated activity;

e the organization of scientific meetingson behalf of the OIE;

the coordination of collaborativestudies;the provision of assistance to the OIEin collecting and disseminating ne-cessary information.

The Director-General of the OIE sentletters to selected laboratories around theworld inviting them to become OIEreference laboratories for the more im-portant diseases of Lists A and B. Becauseof the necessity of some regionalization, inseveral cases a number of referencelaboratories for the same disease wereorganized in different countries or regions.At the beginning of 1994, 101 OIE referencelaboratories were designated for 37 List Aand B diseases or groups of diseases. Themajority of these laboratories developedsignificant activities in accordance withtheir responsibilities, and this is reflectedin annual reports which they submit to theOIE.

OIE requirements for vaccinesAs mentioned above, these requirementsare part of the OIE manual's chapters onthe List A and 13 diseases for whichvaccines are available. The chapters containgeneral information, indicating recom-mended vaccines, data on seed manage-ment (including characteristics of thevaccinal strains, culture, validation as avaccine), manufacture, in-process controland batch control (including sterility,safety and potency tests).

The Standards Commission is aware thatthe manual's coverage of standards forvaccines is less comprehensive than it is ofdiagnostic methods. Nevertheless, somechapters are exemplary and provide a goodmodel for others to follow in the pre-paration of the next edition.

The Commission will, however, avoidany involvement in product licensingprocedures. It will be essential to ensurethat the OIE manual keeps up to date withdevelopments in vaccine technologythrough current advances in microbiology,immunology and biotechnology.Following from this, the importance ofconsistency between the OIE manual, EUregulations and the European Pharma-copoeia must also be emphasized.

In future, the OIE reference laboratorieswill prepare for distribution the standardsthat are essential in vaccine production andcontrol. The present role of the OIE is,however, restricted to the distribution ofinformation contained in the manual andwhich is undoubtedly contributing tointernational harmonization in the area ofvaccine production.

OIE COOPERATION WITH OTHER

INTERNATIONAL ORGANIZATIONS

FAO

FAO plays an important role in improvingthe quality standards of vaccines. Its maininterest is in developing countries andactivities are concentrated on rinderpest

eradication campaigns (Pan-AfricanRinderpest Campaign [PARC], WesternAsia Rinderpest Eradication Campaign[WAREC] and South Asia RinderpestEradication Campaign [SAREC]) in 53countries on three continents; foot-and-mouth disease control in Asia, Africa andSouth America; and Newcastle disease ofpoultry in the same areas.

During the last decade, through nationaland regional projects, FAO has providedexpertise to individual laboratories andassistance in the setting up of networks.FAO has also cooperated with inter-national organizations (WHO, OIE) in thestandardization of particular products andthe holding of regional training courses onvaccines and their associated technologies.

The FAO Expert Consultation on QualityControl of Veterinary Vaccines in Devel-oping Countries, held in Rome, Italy, inDecember 1991 (FAO, 1991; Rweyemamu,SyIla and Palya in FAO, 1993), recommend-ed that closer cooperation be establishedwith other international organizations todevelop a more coherent approach onguidelines for vaccine quality control.FAO, WHO and OIE should work jointlywith developing countries to agree onpriorities and carry forward regionalparticipation in standardization, includingthe development of appropriate laboratoryfacilities and post-licensing surveillance.

This should include participation at thenational level as well as within regionsand would require those countries with noprocedures to begin standardization andregulation, unless a decision has been takento accept standards agreed by othercountries or regions. Regional repositoriesof master seed stocks of vaccine strains,cell lines and challenge strains should beestablished to aid in attaining uniformquality of vaccines.

The consultation recognized the valueof the activities of the Pan African Vet-erinary Vaccine Centre (PANVAC) and

Vaccine manual 147

recommended that FAO, together with theOrganization of African Unity (OAU) andthe Interafrican Bureau for Animal Re-sources (IBAR), should solicit appropriateregional and international support for theconversion of PANVAC to the status of along-term programme institute, withappropriate facilities and support.

It was also stated that tests required bynational authorities to license or releasevaccines should be identical with thoserecommended by the appropriate inter-national organizations. The meetingrecommended the preparation of a manualon veterinary vaccines, providing guide-lines for the production of veterinaryvaccines in developing countries, as aproject to be implemented by FAO andthe other international organizationsconcerned.

World Health OrganizationWHO, located in Geneva, Switzerland, is aspecialized United Nations agency whoseprimary responsibility is in internationaland public health matters. However, italso plays an essential role in the standard-ization of vaccine quality.

The WHO Expert Committee onBiological Standardization is mainlyresponsible for human pharmaceuticalproducts (vaccines) but also for someveterinary biologicals and products ofinterest to animals as well as humans. Thecommittee has published guidelines for thepreparation and establishmenSof referencematerials and reference reagents forbiological substances (WHO, 1978) anddefinitions of these materials and reagentshave been formulated. According to thesedefinitions, an international biologicalstandard is a biological substance to whichWHO has assigned an international uniton the basis of data obtained in aworldwide study.

The prime function of an internationalbiological standard is to enable the activity

of a sample of unknown potency to bemeasured in a biological system andexpressed conveniently in internationalunits. The international biological referencepreparation is a biological substance whichmay be used for a purpose similar to thatof a standard but which has beenestablished without a full collaborativestudy, or after such a study has shown thatit is not appropriate to establish it as aninternational standard.

Besides this activity, the WHO ExpertCommittee on Biological Standardizationestablishes more general requirements forbiological products, among them vaccinesfor veterinary use. A good example is inthe manufacturing and control require-ments for rabies vaccines for veterinaryuse (WHO, 1981).

With regard to products for veterinaryuse or of animal origin, the committee hasestablished requirements for tuberculins(human and bovine), anthrax spore vaccine(live vaccine for veterinary use), immunesera of animal origin, rinderpest cell culturevaccine (live) and rinderpest vaccine (live),Brucella almrtus Strain 19 vaccine andBrucella melitensis Strain Rev.1 vaccine (livefor veterinary use).

Requirements are revised regularly formost products as major advances intechnology for manufacturing or qualitycontrol are reported and accepted by theinternational scientific community.

In addition to specific products, theExpert Committee on BiologicalStandardization also issues general recom-mendations on biological substances andtest systems, for example on the nationalcontrol of vaccines and sera andrequirements for immunoassay kits. It alsoregularly reviews and updates the list ofinternational biological reference prep-arations and reagents.

Technical units such as the VeterinaryPublic Health (VPH) Unit of WHO areassociated with the committee's work each


time a substance or a subject falling withina specific technical unit's domain ofexpertise comes under discussion.

European PharmacopoeiaOne effective approach to vaccine qualitystandardization is based on the EuropeanPharmacopoeia (Artiges, 1992). The 19signatory countries the 12 (in 1992) EUMember States, countries of the EuropeanFree Trade Association (EFTA) and Cyprus

are committed to setting up commonmonographs and making them official intheir territories. The European Pharmaco-poeia has its own secretariat, admini-stratively attached to the Council of theEU, and a laboratory, both of which arelocated in Strasbourg.

Monographs and general analyticalmethods are prepared by specializedexpert groups made up of scientists fromuniversities and national control labor-atories, etc. Before their final adoption, thetexts, which in most cases have been theobject of interlaboratory cooperative tests,are published in public inquiry form in theEuropean Pharmacopoeia's Pharmeuropa,which appears four times a year in Englishand French.

The close cooperation between theEuropean Pharmacopoeia and thosecountries retaining their own nationalpharmacopoeias should be mentioned, i.e.Germany, Austria, Belgium, the UnitedKingdom, France, Italy, Switzerland andthe Nordic countries. The latter havebrought all their general methods in linewith those of the European Pharma-copoeia.

To date, all general analytical methodsare harmonized: there are 17 generalmonographs defining the main pharma-ceutical forms and about 800 monographscovering starting materials and certainbiological preparations such as immuno-sera and vaccines, both human andveterinary.

The second edition of the EuropeanPharmacopoeia has been published since1980 in the form of fascicles. Publication isannual (between June and September) andimplementation of the standards in allmember countries is effective on 1 Januaryof the following year.

Operationally, the roles of the pharma-copoeias and the regulatory governmentalagencies (mentioned earlier) are com-plementary:

*the regulatory agencies approve newdrugs and vaccines on the basis ofproven safety and efficacy andapproved specifications, tests andmethods of analysis for eachapplication;the Pharmacopoeia establishes publicstandards that apply to anymanufacturer of a particular drugsubstance or pharmaceutical product.

In Europe, there is a close legalconnection between licensing systems andpharmacopoeias regarding their legal(Directive EEC/ 75 / 81 / 852) as well aspractical aspects (shared experts and aclose relationship between secretariats).

The European Pharmacopoeia providesdetailed quality standards for vaccinesthrough a series of monographs. These areassembled by an expert group and, oncepublished, are mandatory for the memberstates. There are currently 33 monographson veterinary vaccines and otherbiologicals. Among them are: anthraxspore live vaccine for veterinary use, avianinfectious bronchitis live vaccine (freeze-dried), canine distemper live vaccine(freeze-dried), Clostridium botulinumvaccine for veterinary use, C. chauvoeivaccine for veterinary use, C. perfringensvaccine for veterinary use, equineinfluenza vaccine (inactivated), foot-and-mouth disease vaccine (inactivated), swineerysipelas vaccine (inactivated) and swinefever live vaccine (freeze-dried).

The monographs on veterinary vaccines

Vaccine manual 49


provide information on identification tests,testing for contaminants and storage,etc. The European Pharmacopoeia alsoprovides a limited number of referencepreparations for use as workingstandards - among these preparations arabies vaccine and a bovine tuberculosispurified protein derivative (PPD). A B.melitensis, Rev.1 vaccine is also currentlyunder study.

Except in the case of biotechnologyproducts, the European Pharmacopoeiamaintains only limited liaison with othermajor pharmacopoeias such as those of theUnited States and Japan. For veterinaryvaccines the differences between thesepharmacopoeias are of considerablesignificance.

In general, the European Pharmacopoeiamonographs are more detailed than theOIE manual's requirements for biologicalproducts (OIE, 1992), but they only dealwith vaccines of significance to Europe.Although several of these vaccines arealso important to other continents, theEuropean Pharmacopoeia may be lessaccessible, particularly in developingcountries, than the OIE manual.

In the next edition of the OIE manual,the sections on vaccines will be revisedin the full light of standards alreadyestablished in the European and othermajor pharmacopoeias.

Pan American Health OrganizationThe Pan American Health Organization(PAHO) contributes towards the standard-ization and quality control of foot-and-mouth disease (FMD) and some othervaccines in South America (FAO, 1991).Through the work of PAHO, standards ofvaccine production and quality controlin particular have been significantlyimproved in the last decade, and vac-cination campaigns against FMD arehaving a significant effect on the incidenceof the disease.

European Union

Cooperation between OIE and the EU(DGXII - Science, Research and Devel-opment) in the control of veterinaryreagents and vaccines has been initiated(Lee Aileen, 1992).

BIBLIOGRAPHY

Artiges, A. 1992. The role of the EuropeanPharmacopoeia. Dev. Biol. Stand., 79:87-93.

Draayer, H.A., Hilsabeck, L.J. & Miller,R.H. 1992. The American manufacturer'sview on the EEC texts and harmo-nization. Dev. Biol. Stand., 79: 75-83.

FAO. 1991. Report of the FAO ExpertConsultation on Quality Control ofVeterinary Vaccines in DevelopingCountries, 2-6 December, Rome.

FAO. 1993. Regional harmonization of vaccinequality standards and stimulation ofvaccine technologies in developingcountries. In Proc. Expert Consultation.Quality control of veterinary vaccines indeveloping countries, p. 3-12. FAO AnimalProduction and Health Paper, No. 116.Rome.

Folkers, C. 1992. The FEDESA's point ofview. Dev. Biol. Stand., 79: 27-30.

Lee Aileen, M.T. 1992. Presentation of theEEC Directive/ 81 / 852 - Quality. Dev.Biol. Stand., 79: 39-41.

OIE. 1992. Manual of standards for diagnostictests and vaccines for Lists A and B diseasesof mammals, birds and bees. Paris.

Pensaert, M. 1992. Control of product batches(before and after registration) - theBelgian approach. Dev. Biol. Stand., 79:183-186.

Truszczynski, M. & Blancou, J. 1992. Therole of the Office international desEpizooties in the standardizationof biologicals. Dev. Biol. Stand., 79:95-98.

Watson, J. 1992. Technical requirements ofveterinary vaccines for registration -

Vaccine manual 151

FEDESA's point of view. Dev. Biol. Stand.,79: 51-63.

WHO. 1978. Guidelines for the preparation andestablishment of reference materials andreference reagents for biological substances.WHO Expert Committee on Biological

Standardization. WHO Technical ReportSeries, No. 626. Geneva.

WHO. 1981. WHO Expert Committee onBiological Standardization. WHOTechnical Report Series, No. 658.Geneva.

part 1 overview: the present state of veterinary vaccine

Documents