thymidine kinase-deficient herpes simplex virus type 2 genital infection in guinea pigs

JOURNAL OF VIROLOGY, Aug. 1985, p. 322-3280022-538X/85/080322-07$02.00/0Copyright © 1985, American Society for Microbiology

Thymidine Kinase-Deficient Herpes Simplex Virus Type 2 GenitalInfection in Guinea Pigs

LAWRENCE R. STANBERRY,l* SAUL KIT,2 AND MARTIN G. MYERS'Division of Infectious Diseases, Children's Hospital Research Foundation, University of Cincinnati College of Medicine,

Cincinnati, Ohio 45229,1 and Division of Biochemical Virology, Baylor University College of Medicine, Houston,Texas 770302

Received 27 September 1984/Accepted 15 April 1985

In guinea pigs, thymidine kinase-producing strains of herpes simplex virus type 2 replicated to high titer inthe vagina and spinal cord, and animals developed severe clinical disease. Infection with thymidinekinase-deficient virus resulted in similar vaginal virus titers; however, animals exhibited little or no clinicalillness and only low titers of virus were detected in spinal cord homogenate cultures. Neural and extraneurallatent infection as well as recurrent infection were noted in animals inoculated with either thymidinekinase-producing or -deficient viruses. These data suggest that neural pathways are important in thepathogenesis of genital herpes and that virus-coded thymidine kinase may influence virulence but is notrequired for latency.

During the course of initial herpes simplex virus (HSV)infection, specific immune responses occur which limit pro-ductive viral replication but permit the establishment of alatent viral state. Thus, most infections with HSV appearself-limited but, in fact, the virus persists within the host (3,37). Although viral latency is incompletely understood, it isthought that recurrent herpetic infection is a consequence ofreactivation of virus latent within host cells (25, 28, 38, 44).Although HSV disease is rarely life-threatening, its recurrentnature and the associated discomfort cause many patients toseek medical treatment.One prescription drug, acyclovir, is presently available for

the treatment of initial oral and genital herpes. Acyclovir andseveral experimental antiviral agents are selectivelyphosphorylated by HSV-coded thymidine kinase (TK). Thetriphosphate form of acyclovir competitively inhibits theutilization of normal substrates by the viral DNA polymer-ase and is also incorporated terminally at the 3' ends ofgrowing DNA chains stopping DNA synthesis at that point.Treatment with acyclovir, however, has resulted in theemergence of drug-resistant HSV mutants, some of whichare deficient in viral TK (TK-) (6, 8, 35). The risingincidence of HSV type 2 (HSV-2) genital infections (5) andthe increasing use of antiviral agents for the treatment ofgenital herpes are likely to result in increased human expo-sure to TK- mutants of HSV-2.HSV-coded TK is distinct from mammalian cell TK (17)

and may facilitate HSV replication in nondividing cells (16).Because HSVs are neurotropic viruses normally capable ofboth productive and latent infection in nondividing neuralcells, it has been hypothesized that TK- virus may be lessneurovirulent than wild-type TK-producing (TK+) strains(11). For example, HSV type 1 (HSV-1) TK- mutantsproduce less clinical disease and lower viral titers in neuraltissues, and they may be incapable of producing latentganglionic infection in animal models of nongenital HSVinfections (11, 14, 21, 30, 41). Viral TK therefore wouldappear to influence HSV neurovirulence. Although acuteand subsequently latent HSV-2 infection of neural tissues

* Corresponding author.

occurs as a consequence of initial genital infection (29, 33,36), the importance of neural infection in initial and recurrentgenital herpes is incompletely understood and complicatedby the observation that latent virus may also reside inextraneural tissues, such as skin at the site of initial infection(15, 34, 36, 43). The role of viral TK in initial genitalinfection, as well as in persistent infection at neural andextraneural sites, has not been established. In this study weexamined the influence of viral TK on the pathophysiologyof HSV-2 genital infection.

MATERIALS AND METHODS

Animals. Female strain 2 (Children's Hospital ResearchFoundation Vivarium, Cincinnati, Ohio) or Hartley guineapigs (Charles River Breeding Laboratories, Wilmington,Mass.) weighing ca. 150 to 200 g were used in these studies.Their vaginal closure membrane was ruptured and the vagi-nal vault was swabbed with a premoistened calcium alginate-tipped swab (Spectrumn Diagnostics, Glenwood, Ill.). Onehour later the animals were inoculated with 0.1 ml of virusinstilled into the vaginal vault by use of a syringe and a20-gauge plastic catheter (Abbott Hospitals, North Chicago,Ill.).

Cells and viruses. First- through third-passage rabbit kid-ney (RK) cells were prepared from the kidneys of 3-week-old, Pasteurella-free, female New Zealand white rabbits(Hazleton-Dutchland Inc., Denver, Pa.) and maintained withEagle basal medium containing heat-inactivated (56°C, 30min) fetal bovine serum, penicillin (100 U/ml), streptomycin(50 ,ug/ml), and L-glutamine (2 mM). The preparation ofhuman foreskin fibroblast cells and fetal guinea pig tissuecultures has been previously described (27). Clarified poolsof the Maclntyre strain of HSV-1 (ATCC VR-539), the MS(ATCC VR-540) and 333 (19) strains of HSV-2, and thebromodeoxyuridine-resistant mutant of strain 333 (20) wereprepared in RK cells and stored frozen at -70°C. Thebromodeoxyuridine-resistant mutant of strain 333 fails toinduce detectable TK activity (TK-) in TK-deficient cellsbut enhances the resident TK activity of biochemicallytransformed cells (7, 18).

Virus detection. Swab samples of vaginal secretions were

322

Vol. 55, No. 2

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.

TK- HSV-2 GENITAL INFECTION IN GUINEA PIGS 323

collected with a premoistened calcium alginate-tipped swab.The swab samples were placed in 1 ml of Eagle basalmedium containing penicillin (100 U/ml), streptomycin (50,ug/ml), gentamicin (50 ,ug/ml), and amphotericin B (2.5,ug/ml). During initial infection, groups of animals weresacrificed daily and their lumbosacral dorsal root ganglia andspinal cords were aseptically removed. Ten percent (wt/vol)homogenates were prepared from the tissue samples. Swabspecimens and supernatants from the tissue homogenateswere stored at -70°C until assayed. Virus was titrated byplaque assay in RK cells employing a 1% methylcelluloseoverlay. Five weeks after intravaginal inoculation, animalswere killed and the external genital skin, uterine cervix, andlumbosacral dorsal root ganglia were aseptically removed. Aportion of each tissue was homogenized, and the clarifiedsupernatant was cultured on RK cells for 10 days to evaluatefor the presence of infectious HSV-2. A second portion ofeach tissue was finely minced and explanted ontosemiconfluent RK cell monolayers. Additional RK cellswere periodically added to the explant cultures and observedfor cytopathic effect for at least 6 weeks. The remainingtissue, stored at -70°C, was analyzed for HSV DNA by thedot-blot DNA hybridization technique we have previouslydescribed (27). Briefly, DNA extracted from infected anduninfected animal tissues was spot hybridized on nitro-cellulose filters with 32P-labeled HSV-2 DNA probes pre-pared from nucleocapsids (4, 27, 39). Hybridization wasconducted in 50% formamide at 42°C for 16 to 24 h (27). Theextent of binding was analyzed by autoradiography.

Infectious virus was defined as virus recovered fromcell-free homogenate cultures. Latent virus was defined byfailure to recover HSV from cell-free homogenate cultures

Aco

h. ca _

._.._

cli = 0

I 1- 0)C) o

1

C0

02

B0)

1- C

c , 0

I. -'°P 0

.

C0z

0'->co £C

6

4

2

0

Inoculum (Log10 pfu)

A TK +/4.5 0 TK-/4.5O TK +/5.5 a TK-/.5* Tk +/6.5 0 TK-/6.5

0 1 2 3 4 5 6 7 8 9 10

Area

333/TK- MS/TK+ = 31.3

6 - 333/TK+ -36.53334TK-=35.5

MSITK+ 3 K/IC

2-

0 ' I I I I I I I

0 1 2 3 4 5 6 7 8 9 10

Days After Inoculation

FIG. 1. Course of vaginal HSV-2 replication after intravaginalviral inoculation. Vaginal viral replication was measured by plaquetitration of vaginal swab samples. (A) Six weanling strain 2 guineapigs infected at a variety of inocula of either wild-type (TK+) strain333 or its TK- mutant. (B) Eighteen weanling Hartley guinea pigsinoculated with -6.5 logl0 PFU of either TK+ strain MS, TK+ strain333, or TK- strain 333 HSV-2.

cn

oCl)

0._4

-J

co0)

4.0 -

3.0 -

2.0 -

1.0 -

0.0 4

AreaMS/TK:333/TK+333/TK-

.r I I I. . I

0 1 2 3 4 5 6 7 8 9 10 11 12 13Days After Inoculation

FIG. 2. Clinical course of external genital skin disease in 18weanling Hartley guinea pigs produced by intravaginal inoculationof -6.5 logl0 PFU of either TK+ strain MS, TK+ strain 333, or TK-strain 333 HSV-2. External genital skin disease was quantitated bya skin lesion score (36). Severity of skin disease, as defined by thearea under the lesion score-day curve, was significantly less severein TK- (333/TK-) infected animals than in either TK+ (333/TK+ orMS/TK+) group (P < 0.05 by Mann-Whitney U test). Skin diseasewas similar in the two TK+-infected groups.

when virus or the viral genome could be detected withintissues by explant cocultivation or nucleic acid hybridizationtechniques.

Clinical disease. During the acute infection, animals wereevaluated daily for genital skin disease, urinary retention,hindlimb paralysis, and death. The genital skin disease wasquantitated under a lesion score system described previously(36). After recovery from the primary infection, animalswere periodically examined for evidence of recurrent vesicu-lar lesions on the external genital skin.TK assay. HSV isolates were characterized by enzyme

assays of cytosol extracts of infected cells and by plaqueautoradiography (18, 20, 22, 40). The plaque autoradi-ographic experiments were carried out essentially as de-scribed by Tenser et al. (40), except that monolayer culturesof TK- rabbit skin cells [RAB(BU)] (17) in 100-mm petridishes were used and the cultures were infected with 100 to1,000 PFU of HSV-2 per dish. Two to three days afterinfection, cells were labeled for 6 h with [14C]thymidine (3,uCi per dish), stained with crystal violet (0.1%), and ex-posed to X-ray film for 3 to 7 days at -70°C. After develop-ment of the films, TK+ plaques showed dark rims due toisotope incorporation, whereas TK- plaques were unla-beled.

RESULTSPrimary infection. Intravaginal inoculation of HSV-2 re-

sulted in viral replication in vaginal tissues. Vaginal HSV-2

TABLE 1. Incidence of clinical disease in HSV-2-infectedweanling female guinea pigsa

Incidence (%) with the following virus:Clinical feature

MS/TK+ 333/TK+ 333/TK-

Any genital skin lesions 24/25 (96) 14/14 (100) 11/26 (42)Urinary retention 21/25 (84) 11/14 (79) 0/26 (0)Hindlimb paralysis 3/25 (12) 3/14 (21) 0/26 (0)Death 6/25 (24) 4/14 (29) 0/26 (0)

a Animals were defined as infected if virus was recovered from at least twovaginal swab samples collected at 24-h intervals beginning 24 h after intrava-ginal HSV-2 inoculation (4.5 to 6.5 log1o PFU). Clinically significant urinaryretention was said to be present if, on at least two occasions 24 h apart, morethan 2 ml of urine could be expressed by Credd's maneuver on the bladder.

VOL. 55, 1985 _= 22.3 MS/TK+-17.3= 5.8

333/TK+

333/TK-

%R. %O

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.

324 STANBERRY, KIT, AND MYERS

-

cso00I 0.

c 'c

c oT-nX) o)_

2cn*

%o

6-

4 -

2 -

0

AreaMS/TK+ =1333/TK+ = 2333/TK- =

0 1

18.7>4.612.6

MS/

333,

2 3 4

Days After InoculaFIG. 3. HSV-2 titers in lumbosacral spinal core

tion of spinal cord homogenate samples (two apoint). Hartley guinea pigs were intravaginally inoclogl0 PFU of either TK+ strain MS, TK+ strain 33333 HSV-2. Comparison of the area under the mezcurve demonstrated significantly less virus in spinaTK--infected guinea pigs than in either TK+ grouMann-Whitney U test). The two TK+-infected gramounts of virus in spinal cord samples.

333/TK+ replication, as reflected by the titer of virus in vaginal swabsamples, was greatest in the first 2 days after inoculation andthen declined steadily to low or undetectable levels by day10. The onset, magnitude, and duration of vaginal virusreplication in strain 2 animals were comparable after inocula-tion with either TK+ or TK- strain 333 HSV-2 delivered at

TK- inocula ranging from 4.5 to 6.5 log10 PFU (Fig. la). Thepattern of vaginal HSV-2 replication after intravaginal inocu-lation of Hartley guinea pigs with 6.5 log1o PFU of the MS

5 6 7 strain was similar to that observed for TK+ and TK- HSV-2333 strain (Fig. lb).

tion In contrast to vaginal virus shedding, clinically apparentJ by plaque titra- initial genital infection due to TK- HSV-2 was less severenimals per time and of shorter duration than infection produced by eitherulated with -6.5 TK+ strain (Fig. 2). First lesions with all three strains were13, or TK- strain observed on day 4 after inoculation at the time when vaginalan virus titer-day virus titers were beginning to decline. In the TK+-infectedL1 cord samples of animals, lesions began as erythematous macules which rap-p (P < 0.005 by idly progressed to vesicles and then to a severe ulcerativeoups had similar stage. The majority of these animals had concomitant uri-

nary retention; 10 to 20% developed hindlimb paralysis, andca. 25% of the guinea pigs died (Table 1). Surviving guineapigs demonstrated complete resolution of illness by the endof week 2. In contrast, TK--infected animals seldom exhib-ited overt disease, and those that did develop lesions exhib-

A

gpCMV gpHLV

Dorsal RootB Ganglia

TK+~ *S

InfectedTK .Infected

Controls tI :

HSV-2DNA Uninfected1 6.5ng Genital Skin

UninfectedDorsal Root Ganglia

FIG. 4. Dot-blot hybridization of DNA extracted from virus and tissue samples and hybridized with a 32P-labeled HSV-2 DNA probe. Theprobe was prepared by nick translation of DNA derived from MS strain HSV-2 nucleocapsids prepared in infected RK cells (39). DNAsamples to be hybridized were extracted from tissue specimens (-20 mg wet weight) or virus-infected cell cultures by protease K digestion,phenol-chloroform extraction, and ethanol precipitation. The samples containing ca. 40 ,ug of DNA were suspended in 20 ,ul of buffer, and1:3 and 1:6 dilutions were prepared in buffer. Samples (10 ,u1) of each dilution containing ca. 10, 3.3, and 1.7 ,ug of DNA were NaOH treated,bound to nitrocellulose filters, and hybridized with probe DNA (5 x 107 CPM/,ug) in 50% formamide at 42°C for 16 to 24 h (27). (A) Dot blothybridization of four viral DNA samples: HSV-1 (6.5 log10 PFU per culture) and HSV-2 (7.5 logl0 PFU per culture) grown in human foreskinfibroblasts, and two caviid viruses, guinea pig cytomegalovirus (gpCMV) (4.2 log10 PFU per culture) and guinea pig herpes-like virus (gpHLV)(6.2 logl0 PFU per culture) grown in fetal guinea pig cells. (B) Dot-blot hybridization of DNA extracted from guinea pig tissues. Samples arefrom uninfected, 333/TK+-infected or 333/TK-infected guinea pigs.

HSV-1 HSV-2

Genital Skin Uterine Cervix

_.... . ...............

J. VIROL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.


ited only a small number of discrete vesicles which did notprogress to the severe stage. None of the TK--infectedanimals died.The reduced clinical severity of TK- HSV-2 genital infec-

tion, despite equivalent vaginal virus titers, prompted us toinvestigate the magnitude of viral replication at other sites.Both the magnitude and the duration of productive TK-HSV-2 replication in spinal cord homogenates were less thanthose observed for TK+ viruses (Fig. 3). Additionally, viruswas detected by day 3 in homogenate cultures oflumbosacral dorsal root ganglia, and the titers were lower inTK--infected animals (mean titer, 1.4 log1o PFU/g of tissue)than in 333 TK+-infected animals (mean titer, 2.7 log1oPFU/g of tissue).

Recurrent infection. After recovery from initial infection,survivors were examined for evidence of recurrent herpeticdisease. Brief, isolated vesicular recurrences appeared onthe external genital skin and were observed in both TK+-and TK--infected animals. The lesions were 1 to 2 mm indiameter with an erythematous base. Recurrences were notassociated with systemic illness, lasted ca. 24 h, and werenot observed in mock-infected guinea pigs. In addition,guinea pigs were noted to shed HSV-2 from the vagina in theabsence of demonstrable lesions (asymptomatic shedding).

Latent infection. Approximately 5 weeks after intravaginalHSV-2 inoculation, the genital skin, uterine cervix, andlumbosacral dorsal root ganglia were examined for evidenceof persistent viral infection. To evaluate for the presence ofinfectious HSV-2, a portion of each tissue was homogenizedand the supernatant was cultured for 10 days on RK cells.No infectious virus ws recovered from any of the tissuehomogenate cultures. The presence of persistent HSV DNAwas sought in tissues by dot-blot DNA hybridization. Theblot pattern in Fig. 4a shows homology between the HSV-2DNA probe and both HSV-1 and HSV-2 DNA extractedfrom human fibroblasts. There was no detectable homologyfor guinea pig cellular DNA, or guinea pig cytomegalovirus(ATCC VR-682) and guinea pig herpes-like virus (ATCCVR-543) DNAs. In the accompanying blot (Fig. 4b), theprobe bound to extracted HSV-2 DNA (positive control) butnot to DNAs from uninfected tissue controls. This probealso bound DNA extracted from dorsal root ganglia, externalgenital skin, and uterine cervix of six infected animals,demonstrating the presence of HSV DNA in the tissues ofboth TK+- and TK--infected animals.Because the probe we utilized cannot discriminate the

TK+ or TK- phenotypes, the possibility existed that thelatent virus in TK--infected animals represented a wild-typerevertant. To address this question, latently infected tissues

TABLE 2. Recovery of latent HSV-2 from guinea pig tissuesa

Recovery (%) of the following virus:Tissue

MS/TK+ 333/TK+ 333/TK-

Dorsal root ganglia 41/61 (67) 8/8 (100) 6/8 (75)Genital skin 16/34 (47) 6/8 (75) 6/8 (75)Uterine cervix 10/20 (50) 5/8 (63) 5/8 (63)

a Animals were sacrificed and tissue specimens were aseptically collected35 to 67 days after intravaginal HSV-2 inoculation. A portion of the tissue washomogenized, and the supernatant was cultured on RK cells for infectiousvirus. All homogenate cultures were negative. The remaining tissue was finelyminced and the explants were cocultivated with RK cells. Explant cultureswere maintained for 6 weeks. Selected isolates from 333 TK+- and 333 TK--infected animals were then confirmed to be the TK phenotypes of theinoculum virus by enzymatic assay and plaque autoradiography (18, 20, 42).

TABLE 3. Induction of TK activity after infection of LM(TK-)cells with HSV-2 (333) strains reisolated from virus-infected

guinea pigs (gps)a

Expt no. Source of reisolated virus activityb

1 No virus 0.03TK+ virus used for intravaginal inoculation 6.0

of gpDRGC of gp no. 254 inoculated with TK 0.03HSV-2

Cervix of gp no. 254 inoculated with TK- 0.03HSV-2

Cervix of gp no. 257 inoculated with TK+ 9.4HSV-2

DRG of gp no. 257 inoculated with TK+ 6.4HSV-2

2 No virus 0.02TK+ virus used for intravaginal inoculation 10.3

of gpDRG of gp no. 251 inoculated with TK+ 19.1HSV-2

DRG of gp no. 253 inoculated with TK+ 11.8HSV-2

Cervix of gp no. 253 inoculated with TK+ 18.2HSV-2

Cervix of gp no. 256 inoculated with TK- 0.04HSV-2

a Confluent monolayer cultures of LM(TK-) cells were infected at amultiplicity of infection of about 2 to 8 PFU per cell for 6 h. TK was extractedfrom the cytosol fraction and assayed as described previously (18-20).

b Picomoles of [3H]dTMP formed from [3HJthymidine in 10 min at 38°C permicrogram of protein.

c DRG, Dorsal root ganglia.

were finely minced and cocultivated with low passage RK.Latent virus often required up to 6 weeks of cocultivationbut was detected in the lumbosacral dorsal root ganglia,external genital skin, and uterine cervical specimens ofguinea pigs infected with either TK+ or TK- viruses (Table2). HSV isolates were then characterized by enzyme assaysof cytosol extracts of infected cells and by thymidine plaqueautoradiography. The viruses reisolated from guinea pigsthat had been infected with parental TK+ HSV-2 inducedhigh levels of TK activity, but the viruses reisolated fromanimals infected with the TK- mutant did not have detect-able TK-inducing activity in TK-deficient mouse fibroblast[LM(TK-)] cells (Table 3). A typical thymidine plaqueautoradiographic experiment is shown in Fig. 5. The experi-ment shows that all of the plaques formed by inoculatingTK+ HSV-2 (333) onto TK- RAB(BU) cells incorporatedlabeled thymidine into DNA, but that none of the plaquesformed in the RAB(BU) cells by virus reisolated from thedorsal root ganglia or cervix ofTK- HSV-2-infected animalsincorporated labeled thymidine into DNA. HSV-2 isolatesfrom five dorsal root ganglia, two cervix, and four genitalskin specimens from six guinea pigs infected with TK+viruses were TK+. Isolates from four dorsal root ganglia,three cervix, and four genital skin specimens from eightanimals infected with TK- virus were TK- (Table 2).

DISCUSSION

Genital infection produced by TK- HSV-2 was clinicallymild despite vaginal virus replication comparable to thatproduced by TK+ viruses. The milder illness, however, wasassociated with reduced viral replication in lumbosacral

VOL. 55, 1985

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.


HSV-2A B

LbM

#4

FIG. 5. Thymidine plaque autoradiography, showing the TK-phenotype of latent HSV-2 recovered from guinea pig tissues. (Aand C) Crystal violet-stained monolayers of mock-infected andvirus-infected TK-deficient rabbit skin [RAB(BU)] cells. (B and D)Autoradiogram obtained after monolayers were incubated with[14C]thymidine for 6 h. (A and B) Top panel, TK+ HSV-2 (333)-infected RAB(BU) cells (positive control); middle panel, mock-infected cells (negative control); bottom panel, HSV-2 recovered byexplant cocultivation of the uterine cervix of a TK- HSV-2 (333)-infected guinea pig. (C and D) HSV-2 recovered by explantcocultivation of the dorsal root ganglia (top panel) and the uterinecervix (bottom panel) removed from TK- HSV-2 (333)-infectedguinea pigs.

dorsal root ganglia and spinal cord. These data imply that inproducing genital infection, TK- HSV-2 is less virulent thanwild-type (TK+) HSV-2. Similarly, TK- HSV-1 has beenreported to have reduced pathogenicity in animal models ofherpes encephalitis (11), herpes keratitis (30, 41), and herpeslabialis (21). Additionally, studies with temperature-sensi-tive mutants of HSV-2, some of which were TK-, showedthat these mutants were less pathogenic in a model of genitalinfection but also apparently replicated less well in vaginaltissue than did the wild-type parental strain (1). The mecha-nism by which virus-coded TK influences the pathogenicityofHSV may relate to reduced virulence in neural tissues (11,21, 23, 30, 41). We have previously postulated that virus inlumbosacral dorsal root ganglia or spinal cord or both is thesource of HSV-2 responsible for genital skin lesions duringinitial infection (36). Thus, after vaginal replication, virustransmitted via peripheral nerves replicates in central neuraltissues and then is transported via peripheral nerves togenital skin where local HSV replication produces vesicularskin lesions. Reduced amounts of TK- HSV-2 in neuraltissues could be a consequence of reduced transmission ofvirus to the neural sites or reduced replication in neuraltissues. In either case, the correlation between low TK-

HSV-2 titers in neural tissues and mild genital skin diseasesupports the hypothesis that replication of HSV in neuraltissue is an essential component in the pathogenesis ofgenital herpes.

Latent HSV within the host is felt to be the source of virusresponsible for recurrent herpetic disease (25, 28, 38, 44). Inthe past 50 years, both direct and indirect evidence has beenamassed to support the suggestion of Goodpasture (13) thatneurons in the sensory ganglia harbor latent virus. Morerecently, it has been demonstrated that HSV may establish a

latent infection in extraneural tissues, including the guineapig vagina and footpad (34) and the mouse vulvovagina (43)and ear pinna (15). Our observation that latent virus mayalso be recovered from external genital skin and uterinecervix after recovery from HSV-2 genital infection suggeststhat extraneural latent virus may play a role in thepathophysiology of recurrent HSV infection and possibly inthe evolution of cervical carcinoma (2, 24). Currenty, it isunknown whether there are differences between virus latentin skin and virus latent in ganglia, or whether the latent virusin extraneural tissue resides in neural or nonneural elements.The relative importance of virus latent in neural andextraneural sites in producing recurrent genital disease andcervical carcinoma remains to be defined.

HSV-2C D

J. VIROL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.


The ability of a TK- HSV mutant to establish a latentinfection appears related to whether or not the mutantreplicates in neural tissues. Thus, studies in which viruscould not be detected in ganglia during acute infection alsoreported no recovery of latent TK- HSV (14, 23, 30). Incontrast, when TK- virus has been recovered from gangliaduring acute infection, as we observed, it has often beenpossible to demonstrate latent TK- HSV infection (10, 11,41, 42). These studies suggest that viral replication at neuralsites may be necessary for the establishment of latency.Although viral TK- is important for replication in neural

tissues, it is not an obligatory requirement for establishment,maintenance, or reactivation of latent HSV. For example,Field and Lay (10) have recently characterized latent infec-tions in mice inoculated with an HSV-1 strain that wasclinically resistant to acyclovir. They reported that althoughlatency was uncommonly established, latent TK- HSV-1could be isolated from some ganglia. Similarly, our studies ofgenital HSV-2 infection of guinea pigs demonstrate that viralTK is not an essential requirement for latent HSV-2 infectionin either neural or extraneural tissues. In addition, our datasuggest that only low levels of virus replicating in neuraltissues are necessary to establish latent infection. Such ahypothesis is supported by the observation that althoughacyclovir treatment of infected mice reduced HSV replica-tion in neural tissues, the establishment of latent infectionwas not prevented (9).The frequency with which we detected latent TK- virus

was higher than that reported by other investigators (10, 11,41, 42). Most studies that have explored the role of TK inlatency have utilized HSV-1 mutants with eye or skininoculation of mice (10, 11, 14, 21, 30, 41, 42). Severalfactors reported to be important in the establishment oflatency may have contributed to the differences between ourfindings and those of others. These factors include virusstrain (12, 31), route of inoculation (26, 31), and animalspecies (32). A possible explanation for the higher frequencywith which we detected latent virus compared with investi-gators in other laboratories is the difference in methods usedto search for virus. We used sensitive DNA hybridizationtechniques to probe for ISV DNA. Additionally, becauseTK- viruses do not replicate in nondividing cells as well asTK+ HSV (16), to facilitate recovery of latent TK- virus,cells in the log phase of growth were frequently added to theflasks used for explant cocultivation. With these techniques,the frequency of detection of latent TK- HSV-2 from guineapig tissues was similar to that of TK+ virus. Other investiga-tors have reported the recovery of TK+ virus after infectionwith a TK- HSV mutant (1, 30). In the present study,however, the enzyme assays (Table 3) and plaque autoradi-ographic studies (Fig. 5) demonstrated that the virusesreisolated from the TK- HSV-2-infected guinea pigs had nodetectable TK activity and lacked the ability to induce TKactivity. Thus, the recovery of latent TK- HSV-2 in ourexperiments is not attributable to residual enzyme-inducingactivity in the TK- HSV-2, the presence of TK+ virus in thevirus pools used to infect the guinea pigs, or formation invivo of TK+ revertants.These studies suggest that viral replication in neural

tissues may play an important role in the pathogenesis ofsymptomatic HSV-2 genital infection. The absence of viralTK may alter the neurovirulence, and hence the pathogen-esis, of acute HSV-2 genital infection, but may affect virallatency only indirectly. Viral TK, however, is not an es-sential requirement for latent infection in either neural orextraneural tissues.

ACKNOWLEDGMENTS

This work was supported in part by funds from the Children'sHospital Medical Center, Cincinnati, Ohio, now celebrating its cen-tennial anniversary. L.R.S. is the recipient of a John A. and GeorgeL. Hartford Fellowship.We thank Jackie Brody for technical assistance and Vicki Rieder

for manuscript preparation.

LITERATURE CITED1. Anderson, C. A., M. J. August, and G. D. Hsiung. 1980.

Pathogenicity of wild-type and temperature-sensitive mutants ofherpes simplex virus type 2 in guinea pigs. Infect. Immun.30:159-169.

2. Aurelian, L., I. I. Kessler, N. B. Rosenshein, and G. Barbour.1981. Viruses and gynecologic cancers: herpesvirus protein(ICP 10/AG-4), a cervical tumor antigen that fulfills the criteriafor a marker of carcinogenicity. Cancer 48:455-471.

3. Baringer, J. R., and P. Swoveland. 1973. Recovery of herpessimplex virus from human trigeminal ganglions. N. Engl. J.Med. 288:648-650.

4. Brandsma, J., and G. Miller. 1980. Nucleic acid spot hybridiza-tion: rapid quantitative screening of lymphnode cell lines forEpstein-Barr viral DNA. Proc. Natl. Acad. Sci. U.S.A. 77:6851-6855.

5. Centers for Disease Control. 1982. Genital herpes infection-U.S., 1966-1979. Morbid. Mortal. Weekly Rep. 31:137-139.

6. Crumpacker, C. S., L. E. Schnipper, S. I. Marlowe, P. N.Kowalsky, B. J. Hershey, and M. J. Levin. 1982. Resistance toantiviral drugs of herpes simplex virus isolated from a patienttreated with acyclovir. N. Engl. J. Med. 306:343-346.

7. Donner, L., D. R. Dubbs, and S. Kit. 1977. Chromosomal site(s)of integration of herpes simplex virus thymidine kinase gene inbiochemically transformed human cells. Int. J. Cancer 20:256-267.

8. Field, H. J. 1982. Development of clinical resistance toacyclovir in herpes simplex virus-infected mice receiving oraltherapy. Antimicrob. Agents Chemother. 21:744-752.

9. Field, H. J., S. E. Bell, G. B. Elion, A. A. Nash, and P. Wildy.1979. Effect of acycloguanosine treatment on acute and latentherpes simplex infections in mice. Antimicrob. Agents Chemo-ther. 15:554-561.

10. Field, H. J., and E. Lay. 1984. Characterization of latentinfections in mice inoculated with herpes simplex virus which isclinically resistant to acyclovir. Antiviral Res. 4:43-52.

11. Field, H. J., and P. Wildy. 1978. The pathogenicity of thymidinekinase-deficient mutants of herpes simplex virus in mice. J.Hyg. 81:267-277.

12. Gerdes, J. C., and D. S. Smith. 1983. Recurrence phenotypeswith establishment of latency following rabbit keratitis pro-duced by nmultiple herpes simplex virus strains. J. Gen. Virol.64:2441-2454.

13. Goodpasture, E. W. 1929. Herpetic infection with especialreference to involvement of the nervous system. Medicine8:223-243.

14. Gordon, Y., D. H. Gilden, Y. Shtram, Y. Asher, E. Tabor, M.Wellish, M. Devlin, D. Snipper, J. Hadar, and Y. Becker. 1983.A low thymidine kinase-producing mutant of herpes simplexvirus type 1 causes latent trigeminal ganglia infections in mice.Arch. Virol. 76:39-49.

15. Hill, J. J., D. A. Harbour, and W. A. Blyth. 1980. Isolation ofherpes simplex virus from the skin of clinically normal miceduring latent infection. J. Gen. Virol. 47:205-207.

16. Jamieson, A. T., G. A. Gentry, and J. H. Subak-Sharpe. 1974.Induction of both thymidine and deoxycytidine kinase activityby herpesviruses. J. Gen. Virol. 24:465-480.

17. Kit, S. 1979. Viral-associated and induced enzymes, p. 501-585.In A. C. Sartorelli (ed.), Pharmacology and therapeutics.Pergamon, Oxford, England.

18. Kit, S., and D. R. Dubbs. 1977. Regulation of herpesvirusthymidine kinase activity in LM (TK-) cells transformed byultraviolet light irradiated herpes simplex virus. Virology76:331-340.

VOL. 55, 1985

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.


19. Kit, S., M. Kit, H. Qavi, D. Trkula, and H. Otsuka. 1983.Nucleotide sequence of the herpes simplex virus type 2 (HSV-2)thymidine kinase gene and predicted amino acid sequence ofthymidine kinase polypeptide and its comparison with theHSV-1 thymidine kinase gene. Biochim. Biophys. Acta741:158-170.

20. Kit, S., and H. Qavi. 1983. Thymidine kinase (TK) inductionafter infection of TK-deficient rabbit cell mutants with bovineherpesvirus type 1 (BHV-1): isolation of TK- BHV-1 mutants.Virology 130:381-389.

21. Klein, R. J., A. E. Friedman-Kien, and E. DeStefano. 1981.Pathogenesis of experimental skin infections induced by drug-resistant herpes simplex virus mutants. Infect. Immun.34:693-701.

22. Leung, W.-C., D. R. Dubbs, D. Trkula, and S. Kit. 1975.Mitochondrial and herpesvirus-specific deoxypyrimidinekinases. J. Virol. 16:486-497.

23. McDermott, M. R., J. R. Smiley, P. Leslie, J. Brais, H. E.Rudzroga, and J. Bienenstock. 1984. Immunity in the femalegenital tract after intravaginal vaccination of mice with anattenuated strain of herpes simplex virus type 2. J. Virol.51:747-753.

24. McDougall, J. K., D. A. Ga.lloway, and C. M. Fenoglio. 1980.Cervical carcinoma: detection of herpes simplex virus RNA incells undergoing neoplastic change. Int. J. Cancer 26:1-8.

25. McLennan, J. L., and G. Darby. 1980. Herpes simplex viruslatency: the cellular location of virus in dorsal root ganglia andthe fate of the infected cell following virus activation. J. Gen.Virol. 51:233-243.

26. Meignier, B., B. Norrild, and B. Roizman. 1983. Colonization ofmurine ganglia by a superinfecting strain of herpes simplexvirus. Infect. Immun. 41:702-708.

27. Myers, M. G., L. R. Stanberry, and B. J. Edmond. 1985.Varicella zoster virus infection of strain 2 guinea pigs. J. Infect.Dis. 151:106-113.

28. Nesburn, A. B., M. T. Green, M. Radnoti, and B. Walker. 1977.Reliable in vivo model for latent herpes simplex virus reactiva-tion with peripheral virus shedding. Infect. Immun. 15:772-775.

29. Overall, J. C., Jr., E. R. Kern, R. L. Schlitzer, S. B. Friedman,and L. A. Glasgow. 1975. Genital Herpesvirus hominis infectionin mice. I. Development of an experimental model. Infect.Immun. 11:476-480.

30. Price, R. W., and A. Khan. 1981. Resistance of peripheralautonomic neurons to in vivo productive infection by herpessimplex virus mutants deficient in thymidine kinase activity.Infect. Immun. 34:571-580.

31. Richards, J. R., E. R. Kern, J. C. Overall, Jr., and L. A.Glasgow. 1981. Differences in neurovirulence among isolates ofherpes simplex virus types 1 and 2 in mice using four routes of

infection. J. Infect. Dis. 144:464-470.32. Roizman, B., B. Meignier, B. Norrild, and J. L. Wagner. 1984.

Bioengineering of herpes simplex virus variants for potential useas live vaccines, p. 275-281. In R. M. Channock and R. A.Lerner (ed.), Modern approaches to vaccines. Molecular andchemical basis of virus virulence and immunogenicity. ColdSpring Harbor Laboratory, Cold Spring Harbor, N.Y.

33. Scriba, M. 1975. Herpes simplex virus infection in guinea pigs:an animal model for studying latent and recurrent herpessimplex virus infection. Infect. Immun. 12:162-165.

34. Scriba, M. 1977. Extraneural localization of herpes simplexvirus in latently infected guinea pigs. Nature (London) 267:529-531.

35. Sibrack, C. D., L. T. Gutman, C. M. Wilfert, C. McLaren,M. H. St. Clair, P. M. Keller, and D. W. Barry. 1982. Pathoge-nicity of acyclovir-resistant herpes simplex virus type 1 from animmunodeficient child. J. Infect. Dis. 146:673-682.

36. Stanberry, L. R., E. R. Kern, J. T. Richards, T. M. Abbott, andJ. C. Overall, Jr. 1982. Genital herpes in guinea pigs: pathogen-esis of the primary infection and description of recurrent dis-ease. J. Infect. Dis. 146:397-404.

37. Stevens, J. G., and M. L. Cook. 1971. Latent herpes simplexvirus in spinal ganglia of mice. Science 173:843-845.

38. Stevens, J. G., M. L. Cook, and M. C. Jordan. 1975. Reactiva-tion of latent herpes simplex virus after pneumococcal pneumo-nia in mice. Infect. Immun. 11:635-639.

39. Straus, S. E., H. S. Aulakh, W. T. Ruyechan, J. Hay, T. A.Casey, G. F. Vande Woude, J. Owens, and H. A. Smith. 1981.Structure of varicella-zoster DNA. J. Virol. 40:516-525.

40. Tenser, R. B., J. C. Jones, S. J. Ressel, and F. A. Fralish. 1983.Thymidine plaque autoradiography of thymidine kinase-positiveand thymidine kinase-negative herpesviruses. J. Clin.Microbiol. 17:122-127.

41. Tenser, R. B., R. L. Miller, and F. Rapp. 1979. Trigeminalganglion infection by thymidine kinase-negative mutants ofherpes simplex virus. Science 205:915-917.

42. Tenser, R. B., S. Ressel, and M. E. I)unstan. 1981. Herpessimplex virus thymidine kinase expression in trigeminalganglion infection: correlation of enzyme activity with ganglionvirus titer and evidence of in vivo complementation. Virology112:328-341.

43. Walz, M. A., R. W. Price, K. Kayashi, B. J. Katz, and A. L.Notkins. 1977. Effect of immunization on acute and latentinfections of vaginouterine tissue with herpes simplex virustypes 1 and 2. J. Infect. Dis. 135:744-752.

44. Walz, M. A., R. W. Price, and A. L. Notkins. 1974. Latentganglionic infection with herpes simplex virus types 1 and 2:viral reactivation in vivo after neurectomy. Science 184:1185-1187.

J. VIROL.

Dow

nloa

ded

from

http

s://j

ourn

als.

asm

.org

/jour

nal/j

vi o

n 04

Feb

ruar

y 20

22 b

y 20

01:e

e0:4

e5f:

68b0

:80e

f:d5

e4:d

403:

d3e0

.

thymidine kinase-deficient herpes simplex virus type 2 genital infection in guinea pigs

Documents