virologic, antigenic and genetic characterization of
TRANSCRIPT
Virologic, antigenic and geneticcharacterization of chicken anemia virus (CAV)and development of a new serologic diagnosticmethod
その他（別言語等）のタイトル
鶏貧血ウイルスのウイルス学的，抗原学的および遺伝学的特徴づけと新しい血清学的検査法の開発
著者（英） Trinh Quang Dai学位名 博士（畜産衛生学）学位授与機関 帯広畜産大学学位授与年度 2015学位授与番号 10105甲第66号URL http://id.nii.ac.jp/1588/00001382/
Virologic, antigenic and genetic characterization of
chicken anemia virus (CAV)
and development of a new serologic diagnostic method
2015
Trinh Quang Dai
Doctoral Program of Animal and Food Hygiene
Graduate School of Animal Husbandry
Obihiro University of Agriculture and Veterinary Me dicine
鶏貧血ウイルスのウイルス学的、抗原学的鶏貧血ウイルスのウイルス学的、抗原学的鶏貧血ウイルスのウイルス学的、抗原学的鶏貧血ウイルスのウイルス学的、抗原学的
および遺伝学的特徴づけと新しい血清学的検査および遺伝学的特徴づけと新しい血清学的検査および遺伝学的特徴づけと新しい血清学的検査および遺伝学的特徴づけと新しい血清学的検査法の開発法の開発法の開発法の開発
平成平成平成平成 27年年年年
（（（（2015））））
帯広畜産大学大学院畜産学帯広畜産大学大学院畜産学帯広畜産大学大学院畜産学帯広畜産大学大学院畜産学研究科研究科研究科研究科
博士博士博士博士後期後期後期後期課程課程課程課程 畜産衛生学専攻畜産衛生学専攻畜産衛生学専攻畜産衛生学専攻
ツツツツウウウウインインインイン ククククワンワンワンワン ダイダイダイダイ
i
Contents
Contents ................................................................................................................................. i
Abbreviations ...................................................................................................................... iii
General introduction ........................................................................................................... 1
Chapter I ............................................................................................................................ 16
Characterization of monoclonal antibodies to chicken anemia virus ......................................
and epitope mapping on its viral protein, VP1 .................................................................... 16
1.1. Introduction ................................................................................................... 16
1.2. Materials and methods ................................................................................... 17
1.3. Results ........................................................................................................... 26
1.4. Discussion ...................................................................................................... 30
1.5. Summary ........................................................................................................ 35
Chapter II ........................................................................................................................... 47
Development of a blocking latex agglutination test for the detection of .................................
antibodies to chicken anemia virus ...................................................................................... 47
2.1. Introduction ................................................................................................... 47
2.2. Materials and methods ................................................................................... 48
2.3. Results ........................................................................................................... 53
2.4. Discussion ...................................................................................................... 55
2.5. Summary ........................................................................................................ 58
Chapter III ......................................................................................................................... 66
Isolation and preliminary characterization of chicken anemia virus circulating
in Vietnam ............................................................................................................................ 66
3.1. Introduction ................................................................................................... 66
3.2. Materials and methods ................................................................................... 67
3.3. Results ........................................................................................................... 72
ii
3.4. Discussion ...................................................................................................... 75
3.5. Summary ........................................................................................................ 78
General discussion 86
General conclusion ............................................................................................................ 91
Acknowledgements 93
References ........................................................................................................................... 95
iii
Abbreviations
A AGV2 avian gyrovirus 2
AIV avian influenza virus
aa amino acid
B BFDV beak and feather disease virus
b–LAT blocking latex agglutination test
BSA bovine serum albumin
C CAA chicken anemia agent
CAV chicken anemia virus
CIAV chicken infectious anemia virus
CPE cytopathic effect
D DVDC DABACO Veterinary Diagnosis Centre
dpi day post–inoculation
DuCV duck circovirus
E ELISA enzyme–linked immunosorbent assay
G GM growth medium
H HGyV human gyrovirus
hpi hour post infection
I IBDV infectious bursal disease virus
IFAT indirect fluorescent antibody tests
IP immunoprecipitation
L LBM live–bird markets
M MDV Marek’s disease virus
mAb monoclonal antibody
N NDV Newcastle disease virus
iv
NIAH National Institute of Animal Health
NIVR National Institute of Veterinary Research
O ORF open reading frame
P PBS phosphate buffered saline
PCR polymerase chain reaction
PCV1 porcine circovirus type 1
PCV2 porcine circovirus type 2
PiCV pigeon circovirus
S SPF specific pathogen–free
SwCV swan circovirus
T TCID50 50% tissue culture infective dose
V VNT virus neutralization test
1
General introduction
Historical background
Chicken anemia virus (CAV), also called CIAV, was first isolated from the diseased
chickens of the commercial flocks with MDV vaccine failures that were caused by the
contaminated reticuloendotheliosis virus during investigation of this problem in Japan.
CAV was tentatively called as CAA (chicken anemia agent) as a virus–like agent at that
time. The agent passed through a 25 nm pore size membrane, and was resistant to organic
solvents (ether and chloroform) and heating at 85℃ for 15 min. Moreover, CAA was
transmissible to one−day−old SPF chickens, which resulted in severe anemia and death,
and neutralized with sera of CAA–inoculated chicks. CAA could grow only in SPF
chickens without antibodies to CAA (Yuasa et al., 1979). However, it could not proliferate
in conventional monolayer primary cell cultures derived from kidney, thymus bursa of
Fabricius, or bone marrow of chickens, and skin, muscle, liver, or brain of chick embryos
(Yuasa, 1983). Thus, CAA was strongly suspected to be a new virus distinct from known
viruses of chickens. However, CAA was lacking in conclusive evidences as a virus since
no nucleic acid type was determined, and viral particles were not detectable by electron
microscopy at that time (Yuasa et al., 1979).
In 1983, the first evidence to support the growth of CAA in vitro was described in
certain chicken lymphoblastoid cell lines, such as cell lines established from Marek’s
disease (MD) lymphoma, MDCC−MSB1 (MSB1) and MDCC−JP2, and an avian
lymphoid leukosis (LL) cell line, LSCC−1104B1, (Yuasa, 1983; Yuasa et al., 1983a).
However, the propagation of CAA was not observed in the two MD lymphoma cell lines,
MDCC–RP1 and MDCC–BP1, or in the two LL cell lines, LSCC–1104X5 and LSCC–
2
TLT. This breakthrough provided remarkable progress in CAA research, especially on its
molecular biology. First, conventional type–diagnostic techniques including viral
infectivity and serological assays were developed. Subsequently, successful purification of
CAA from the infective MDCC-MSB1 cells (MSB1 cells) culture fluids was achieved, and
various virological and molecular biological aspects of CAA were clarified using the
purified samples (Schat and van Santen, 2008).
Goryo et al. (1987) showed the virus particles which were spherical or hexagonal in
shape, varying from 18 to 22 nm (19.1 ± 0.2 nm) in diameter by negative electron
microscopy with phosphotungstic acid staining. At a later time, Gelderblom et al. (1989)
also described the more detailed morphology of the virus particles with a mean diameter of
25 nm (23 to 28 nm). The virus particle was composed of 32 hollow morphological units
representing a regular T=3 icosahedron by negative electron microscopy with uranyl
acetate staining. Furthermore, they indicated that the viral DNA was circular and single–
stranded with the size of 2.1744±0.148 kb. According to those findings, the term “chicken
anemia virus (CAV) or chicken infectious anemia virus (CIAV)” was proposed instead of
chicken anemia agent (CAA) (Gelderblom et al., 1989; Schat and van Santen, 2008).
The second report on the presence of CAV in chicken flocks was provided in Germany
(von Bülow et al., 1983), and subsequently it has been reported in many countries in all
continents. However, the presence of CAV was retrospectively confirmed in serum samples
collected from chickens at least since 1959 in USA (Toro et al., 2006).
Virus taxonomy
The family Circoviridae contains two genera, Circovirus, and Gyrovirus. Recently,
another related virus, Cyclovirus, was found in human’s and chimpanzee’s feces (Li et al.,
3
2010), which may be a member of the family Circoviridae but has not been approved as a
species so far (King et al., 2011). The genus Circovirus consists of several species
infecting mammalian and avian species, such as PCV1 and PCV2, DuCV, PiCV, BFDV or
SwCV, and can be the causative to induce diseases in host animals. On the other hand,
CAV has been only the member of the genus Gyrovirus (King et al., 2011). Recently, a new
virus that possesses a distant relationship with CAV, designated as avian gyrovirus 2
(AGV2), was discovered in chickens in Brazil and Netherland, which is considered as a
new member of the genus Gyrovirus (dos Santos et al., 2012). In addition, some new virus
genomes partially similar to CAV genomes were detected in humans, posing a new
member of the genus Gyrovirus, named human gyrovirus (HGyV) (Biagini et al., 2013;
Maggi et al., 2012; Phan et al., 2012).
Viral morphology, genomes and proteins
Virion of CAV with a buoyant density in CsCl of 1.33–1.35g/cm3 is not enveloped,
and icosahedral with an average diameter of around 19.1 nm–26.5 nm, which is similar to
the other species in the family Circoviridae, for instance, 17.0 nm–20.7 nm (PCV1), 20.5
(PCV2), 12.0 nm–20.7 nm (BFDV). Cryo–electron microscopy indicated that the capsid
structure exhibits distinctive surface structure, and a structure comprising 60 subunits
(T=1) arranged in 12 trumpet–shaped petameric rings (King et al., 2011).
The viral genome consists of a negative–sense, circular, single–stranded of 2.3–kb
DNA. CAV genome was constructed by an enhancer region and 3 partially overlapping
ORFs (ORF1, ORF2, ORF3) which encode 3 viral proteins, VP1 (52 kDa), VP2 (24 kDa),
and VP3 (14 kDa) (Schat and van Santen, 2008).
4
Among the 3 viral proteins, VP1 is the only structural protein known to form the viral
capsid, which can be detected in highly purified virus particles (Todd et al., 1990a). This
protein plays an important role for inducing neutralizing antibodies in infected chickens.
However, the information of neutralizing epitopes on VP1 remains unclear, since no
epitopes have been mapped on this capsid protein.
Although the non–structural VP2 protein function is unknown, it has been proposed
that VP2 may act as a scaffold protein during virion assembly to facilitate the correct
conformation of VP1 (Noteborn et al., 1998). The study of Koch et al. (1995) could
support this hypothesis, since insect cells expressing both VP1 and VP2 using the
baculovirus expression system induced neutralizing antibodies when chickens were
inoculated with the insect cells with both viral proteins, and the inoculated chickens could
protect their progeny from CAV infection, while the insect cells containing either VP1 or
VP2 could not. However, kinetics of viral proteins expression in infected MSB1 cells
showed that both VP2 and VP3 could be detected as early as 12 hpi, whereas VP1 was first
detectable at 30 hpi (Douglas et al., 1995). Further studies are definitely needed to fully
understand the function of VP2.
The non–structural VP3 protein, also called apoptin, induces apoptosis in chicken
thymocytes and chicken lymphoblastoid T cell lines (Noteborn et al., 1994). VP3 was also
found to have ability to induce apoptosis in several human cancer cells. This finding
together with the animal experiments with recombinant VP3 proteins suggests that VP3
might be considered as a new therapy for human cancers (Rollano Penaloza et al., 2014;
Schat and van Santen, 2008).
5
Genetic variation
In general, there is no significant difference in CAV genome sequences irrespective of
geographic origin of the viruses. Comparison of the complete or partial genome sequences
available in GenBank resulted in very little difference among them in both nucleotide and
predicted aa levels. The diversity of nucleotide sequences of CAV is usually around 5%,
and the largest distance was found between some Australian isolates and others (Ducatez et
al., 2006; Eltahir et al., 2011; Islam et al., 2002; Krapez et al., 2006; Zhang et al., 2013).
Phylogenetic analysis of nucleotide sequences of CAV indicated the presence of 3 or 4
genotypes (Ducatez et al., 2006; Eltahir et al., 2011).
At the aa level, VP2 and VP3 are conservative among strains, while VP1 seems to be
more diverse. The presence of a hypervariable region at the positions from 139 to 151 of
VP1 aa residues was reported (Renshaw et al., 1996). The replication of CAV strains cells
with aa profile Q139, and/or Q144 in VP1 in MSB1 were affected more than that of the
strains with different aa at these position. This result was confirmed by chimeric CAVs
containing regions with or without aas Q139/Q144. However, some research groups
showed a good growth of their field CAV strains with the aa profile (Q139/Q144) in MSB1
cells (Connor et al., 1991; Krapez et al., 2006). Therefore, site–directed mutagenesis might
be needed to confirm the effect of these aas in the propagation of CAV in cell cultures. In
addition, several studies on the hypervariable region of VP1 indicated the correlation
between the aa profile (positions 75, 97, 139, and 144) in the region and the clustering of
CAV strains in phylogenetic analysis (Hailemariam et al., 2008; Islam et al., 2002; van
Santen et al., 2001). The common signature aa profile (I/T75, L97, Q139, and Q144) could
be identified only in cluster II, while another major aa profile (V75, M97, K139, and E144)
6
was found in clusters I and III (Ducatez et al., 2006; Islam et al., 2002). However, their
virological significance of the correlation remains unclear.
Antigenicity and strain variation
Since the first description of CAV in Japan, the virus has been isolated in chickens of
all continents indicating the woldwide distribution of this virus in poultry industry (Schat
and van Santen, 2008; Yuasa et al., 1979). There was no difference in the antigenicity of
CAV isolates when they were tested by using chicken polyclonal antibodies (McNulty et
al., 1990a; Yuasa and Imai, 1986). Although there were few reports on the experimental
condition to study the difference of antigenicity, CAV was generally considered to belong
to a single serotype (Schat and van Santen, 2008). In other studies, the USA isolate CAIV–
7, in despite of its CAV–like pathogenic and physicochemical characteristics, showed the
evidence of antigenicity distinct from a CAV presentative Del–ros strain (Spackman et al.,
2002a, 2002b). However, since the confirmation by sequencing CAIV–7 has not been done,
the presence of the new serotype of CAV remains unclear.
Host range and pathogenicity
Chickens are a major host of CAV, although turkeys and Japanese quails might be also
infected with a circovirus similar to CAV (Schat and van Santen, 2008). All ages of
chickens are susceptible to CAV infection; however, many factors including the age of
chickens, level of maternal antibodies, viral load, route of infection, virulence of the virus,
or the presence of other pathogens result in the consequences of infection in infected
chickens (Tan and Tannock, 2005; Yuasa and Imai, 1986; Yuasa et al., 1980a). However, it
has been regarded that there is not substantial difference in pathogenicity among strains
7
isolated in different areas of the world (Natesan et al., 2006; Schat and van Santen, 2008;
Yuasa and Imai, 1986).
In the field, usually, outbreaks associated with CAV infection due to vertical
transmission are observed in the progeny of young breeder chickens lacking antibodies to
CAV. The hatched chicks develop clinical diseases and show increased mortality beginning
at around 2 weeks old after hatching, with a peak at around 3 weeks old (Chettle et al.,
1989; Yuasa et al., 1987). Typical clinical signs include anorexia, depression, and
discoloration of skin and muscle due to severe anemia. The diseased chicks have a low
hematocrit value less than 27% compared with a normal value of 30% or higher. Typical
gross lesions including intramuscular and subcutaneous hemorrhages, whitish−yellow or
pink bone marrow, severe atrophy of thymus, atrophy of bursa of Fabricius, hemorrhages
in proventriculus, and swelling of liver are found in the diseased young chicks. These
clinical signs and lesions can be experimentally reproduced when 1−day−old chicks were
inoculated with CAV (Taniguchi et al., 1982; Yuasa et al., 1979). In the field cases,
mortality is usually around 10−20%, and can be up to 60% probably due to secondary
infections with other viruses, bacteria or fungi (Schat and van Santen, 2008).
The age resistance of chickens to the clinical diseases caused by CAV was
experimentally demonstrated. In the experiment conditions, chickens of older than 2
weeks of age infected with CAV did not show any clinical signs, although they remain
susceptible to CAV infection (Yuasa and Imai, 1986; Yuasa et al., 1983b). CAV infection in
older chickens (subclinical diseases) may induce immunosuppression. Consequently, CAV
infection may enhance susceptibility of the infected chickens to other pathogens. In
addition, reduction in response to several vaccines of important avian infectious diseases
such as MD and Newcastle disease (ND) has also been reported (Adair, 2000).
8
When one–day–old chicks were inoculated with CAV, the virus caused clinical
diseases, and was recovered from all the organs examined (thymus, liver, spleen, bursa of
Fabricius, and bone marrow) until 28 dpi; except for the fecal samples and brain in which
the virus was detectable up to 49 dpi when the experiment was finished, although virus
neutralizing (VN) antibodies to CAV were first detectable at 21 dpi. By contrast, CAV–
inoculated older chickens (4 and 7 weeks old) with subclinical diseases rapidly developed
VN antibodies that were detectable at 7 dpi, and the virus was rapidly eliminated from the
chickens (Yuasa et al., 1983b). However, CAV could cause typical clinical diseases in older
chickens when development of VN antibodies to CAV was suppressed by simultaneous
infection with IBDV or by bursectomy (Yuasa et al., 1980b, 1988). Thus, VN antibody
development might be the key factor to gain the age resistance in older immunocompetent
chickens. However, since CAV caused a depletion of T cells and also affected macrophage
function in 3–week–old chickens with the development of the neutralizing antibodies
(Adair, 2000; McConnell et al., 1993), cells susceptible to CAV seem to still persist in
those chickens.
One of the main effects of CAV may be on the immune organs of infected chickens.
The experimental infection of CAV in one−day−old chicks showed that the cells highly
susceptible to the early stage of CAV infection were hemocytoblasts in the bone marrow
(Smyth et al., 1993; Taniguchi et al., 1983). The destruction of these cells causes severe
depletion of blood cells, which results in anemia and decrease of leukocytes. However, the
activity of these cells recovered at around 16−18 dpi, and the normal condition of bone
marrows returned at around 32 dpi. CAV infection also affects the lymphoid tissues, where
T lymphocyte progenitor cells are the major target of the virus. These cells in the thymus
seem to be highly susceptible to CAV. Therefore, the severe depletion of T cells in the
9
thymus is observed during the infection. However, the mechanism of CAV infection in T
cells of the spleen remains unknown, although the infected cells were detectable in the
spleen at the same time as in the thymus under the inoculation conditions (Adair, 2000;
Smyth et al., 1993; Taniguchi et al., 1983).
Transmission
There are 2 major pathways for spread of CAV, vertical and horizontal transmissions.
A large amount of CAV is excreted into feces of the infected chickens. Therefore,
contaminated feces are probably the main source of infection. Horizontal transmission of
CAV mostly occurs via oral route through direct or indirect contact with CAV. CAV
infection through horizontal transmission mostly occurred in chickens after dissapearance
of maternal antibody at around 2−4 weeks of age. Then, seroconversion usually occurred at
around 8−12 weeks of age due to the horizontal infection. The horizontal infection
normally results in subclinical diseases in chickens (McNulty et al., 1988; Todd et al.,
2001)
Vertical transmission through the egg from infected breeder chickens is the major
transmission route to cause clinical diseases in the progeny. CAV infection occurring in
antibody–negative breeder chickens by horizontal infection or by the contaminated semen
can result in vertical transmission (Hoop, 1993; Yuasa et al., 1987). In experimental
conditions, vertical transmission can occur only for 8−14 days following infection in
breeder chickens (Hoop, 1992; Yuasa and Yoshida, 1983). In the fields, vertical
transmission may occur for around 3−9 weeks after the exposure to CAV, and the peak is
usually around 1−3 weeks. Immune response developed in the infected breeder chickens
can protect their progeny from vertical transmission (Schat and van Santen, 2008).
10
Brentano et al. (2005) demonstrated the presence of CAV genes in the organs of progeny of
the breeder chickens with high titers of VN antibody using nested PCR, although virus
isolation was not successful. This research results raised the question of CAV persistence in
the antibody–positive chickens.
Economic significance
CAV is considered as an important disease in chickens associated with economic
significance in poultry industry. First, in flocks with clinical diseases, losses are from
increased mortality of around 10−20%, and can reach to 60% by secondary infection of
other pathogens such as other viruses, bacteria or fungi. Poor growth and the costs of
antibiotics used, disinfectant, etc. are also counted into the losses (McNulty, 1991). The
loss of net income from 17.3% to 19.6% due to CAV infection has been reported (McIlroy
et al., 1992). In another report, 14−24% reduction of weight and changes in feed
conversion ratios in the infected flocks with clinical signs were recorded (Davidson et al.,
2004).
Second, CAV infections in SPF chickens were reported, which resulted in serious
problems in SPF chicken producing industry. When seroconversion in SPF flocks occurs,
eggs are no longer to be SPF, and these eggs cannot be accepted for vaccine production
usage in many countries (Schat and van Santen, 2008). Moreover, since the virus is
extremely resistant to chemical and physical agents (Yuasa et al., 1979), the clearance of
CAV from the SPF chicken farms is not feasible and costly.
Third, the impact of subclinical diseases in the infected flocks has been discussed.
Study in North Ireland showed a decrease of 13% of net income in the CAV-infected flocks
in comparison to antibody–negative flocks (McNulty et al., 1991), while some other
11
studies showed no effects of subclinical disease in chicken flocks in USA and Denmark
(Goodwin et al., 1993; Jorgensen et al., 1995). However, the impact of subclinical diseases
may indirectly impair immune system, which leads to the reduction of immune response to
vaccines. Thus, it may be difficult to estimate the real loss of the subclinical diseases in
chicken flocks (Schat and van Santen, 2008).
Diagnosis
CAV can be isolated from many tissues such as liver, spleen or bone marrow, buffy
coat cells, and rectal contents of diseased chickens (Yuasa et al., 1983b). The virus can be
isolated using one−day−old SPF chicks, or susceptible cell cultures described above.
Intramuscular or intraperitoneal innoculation into susceptible chicks is considered as the
most specific and sensitive routes for CAV isolation. At 2−3 weeks post virus inoculation,
anemia can be detected through hematocrit values, which is usually below 27%. However,
confirmation of the presence of virus by PCR or immunohistochemistry using the lesions
of the diseased chickens is also important (Schat and van Santen, 2008).
Cell cultures are widely used for CAV isolation due to their convenient handle. Two
cell lines, MDCC–CU147 and MSB1 cells, have been preferably used (Calnek et al., 2000;
Yuasa, 1983). However, cells inoculated with suspected samples requires subcultures every
2−4 days for 7−10 passages or until CPE observed. However, Renshaw et al. (1996)
reported that some CAV strains did not replicate well in MSB1 cells. The reason why some
strains could not grow in MSB1 cells remains unclarified.
Molecular techniques (PCR, nested PCR or real–time PCR) are widely used for the
detection or quantitation of CAV genes in infected chickens and cell cultures, etc (Cardona
12
et al., 2000b; Markowski–Grimsrud et al., 2002; Noteborn et al., 1992; Soine et al., 1993).
These techniques are usually more sensitive than cell culture assays.
For detection of antibody to CAV, three serological tests are routinely available:
ELISA, IFAT, and VNT. All of these tests require expensive equipment. While indirect and
blocking ELISAs are well–suited to routine screening of a large number of samples, neither
of VNT nor IFAT is suitable to test a large number of samples (Schat and van Santen,
2008). Moreover, none of the 3 tests is applicable to conduct in the field without well–
equipped conditions. Therefore, a simple, rapid and reliable test that can be applied to
detect CAV antibody especially in the field conditions, where specific equipment for
diagnosis cannot be available, is useful for control of CAV infection.
Preventive and controlling strategies
Due to the high prevalence of CAV in the field and its economic impact on poultry
industry, prevention and control of CAV infection are important for poultry industry.
Controlling strategies including management procedures and vaccination are recommended
to minimize the impact of CAV infection on poultry industry. Commercial CAV live
vaccines are currently available; they can be applied to breeder chickens from 6 weeks of
age until 6 weeks before the first egg collection to protect vertical transmission of CAV to
their progeny during a laying period.
Vaziry et al. (2011) studied on chickens vaccinated at one−day−old, and the results
suggested that the attenuated vaccine strain could persist in the thymus and spleen in some
birds resulting in a low humoral immune response, and altering thymopoiesis. This finding
raised the possibility that the vaccine strain might play an important role in subclinical
diseases and reduce responsiveness to other avian pathogens.
13
A mutant CAV vaccine strain has been described to offer a method to protect newly
hatched chicks against CAV (Kaffashi et al., 2008). In that study, CAV with a mutation of
aa E186G in VP2 induced high neutralization titers and caused little damage to the thymus
in inoculated one−day−old chicks. However, it is necessary to conduct additional studies
for testing the stability of this mutant virus and the field application as a vaccine to young
chicks.
There have been a few reports on development and evaluation of DNA vaccines and
inactivated vaccines against CAV. Unfortunately, these vaccines have not been
commercially available to protect CAV infection (Moeini et al., 2011; Pages–Mante et al.,
1997; Sawant et al., 2015).
Improved management and hygiene procedures can protect breeder flocks from early
exposure to CAV, and reduce seroconversion rates in flocks. However, the late exposure
may cause problems in non−immune flocks when flocks are exposed to CAV during a
laying period. Monitoring the CAV antibody status in breeder flocks should be conducted
to avoid the vertical transmission before the flocks enter a laying period and also to
evaluate the efficacy of vaccines (Schat and van Santen, 2008).
Objectives and chapter structures
As mentioned above, there are still many questions on the CAV properties to be
elucidated: for example, elucidation of synthesis mechanism and function of viral proteins,
synthesis mechanism of virus particles, and characteristics of neutralizing epitopes
associated with pathogenicity and development of immunity, or persistence of virus in
chickens in the presence of neutralizing antibody. Thus, additional studies are needed to
14
better understand pathobiology of CAV, which could also provide valuable information on
virus epidemiology, more effective vaccines, or more sophisticated diagnostic methods.
Among the 3 viral proteins, VP1, the only capsid protein, is the major protein to generate
neutralizing antibody against CAV. Therefore, VP1 is considered as a key target to study
pathogenesis and antigenicity of CAV, and to use as immunogen of subunit vaccines or
antigens of diagnostic kits, etc. However, there is a lack of information on the importance
of differences in aa sequence for the tertiary structure and antigenicity of VP1 in addition
to the biological function (Schat, 2009). One of my study aims is to characterize VP1 using
mAb strategy. In addition, monitoring of CAV antibodies is important in poultry production
to protect young chicks from vertical transmission of CAV, and to ensure that SPF chicken
flocks are free of CAV infection. Currently, 3 tests (VNT, IFAT, ELISA) are available to
detect CAV antibodies; however, these tests still have several limits in the field application.
In this study, I developed a highly sensitive and specific latex agglutination test, b–LAT, for
the detection of CAV antibodies in chickens. This test was also applied to detect the
antibodies in chickens in Vietnam, where the presence of CAV has not been reported, for
evaluation of its field use.
Chapter I describes the production and characterization of mAbs against CAV and
epitope mapping on its viral protein 1 (VP1). Three of 4 mAbs established against CAV
showed neutralization activity and recognized VP1 capsid protein. Using the neutralizing
mAbs and escape mutants generated by using each neutralizing mAb, the CAV and its VP1
were precisely analyzed. In this chapter, I describe antigenic properties of CAV strains and
an association with their genetic background, and the first finding of neutralizing epitopes
on VP1.
15
Chapter II describes the development of a new serological test, b–LAT utilizing a
neutralizing mAb, for the detection of antibodies against CAV in chickens. The specificity
of the b–LAT was evaluated using sera from SPF chickens, and chicken positive sera to
other avian viruses. The results of b–LAT showed a high agreement with those of VNT
known to be most sensitive and specific in the detection of CAV antibodies.
Chapter III describes the first detection of CAV infection in chickens in Vietnam.
Virologic, genetic and serologic examination showed a high prevalence of CAV infection
in chickens. The characterization of Vietnamese CAVs was also described.
16
Chapter I
Characterization of monoclonal antibodies to chicken anemia virus
and epitope mapping on its viral protein, VP1
1.1. Introduction
All CAV isolates belong to one serotype (McNulty et al., 1990a; Yuasa and Imai,
1986), and the aa composition of VP1 is considered to be highly conserved, although a
hypervariable region within VP1 (aa positions 139 to 151) was also reported (Renshaw et
al., 1996). However, in a phylogenetic analysis of full–length deduced VP1 aa sequences, 3
distinct clusters (genetic groups) were reported, and a common signature aa profile (I/T75,
L97, Q139, and Q144) could be identified only for cluster II, and another major aa profile
(V75, M97, K139, and E144) was also found in clusters I and III (Ducatez et al., 2006;
Islam et al., 2002). However, there is a lack of information related to the antigenicity of
CAV strains belonging to these genetic groups. Therefore, antigenicity of these genetic
groups is needed to be clarified, which could provide valuable information for control and
prevention of CAV infection.
There have been some reports on the production of mAbs to CAV, particularly
focusing on those with neutralizing activity. McNulty et al. (1990b) produced mAbs to the
CAV strain Cux-1 which were categorized into three fluorescent staining patterns to
antigens in the infected MSB1 cells: Type 1, fine granular nuclear staining; Type 2, large,
spherical nuclear inclusions; Type 3, similar to Type 1, but much more intense staining and
occurring in a higher proportion of nuclei. However, only Type 1 mAbs showed
neutralization activity, but not all (3 of the 4 mAbs). The epitopes recognized by these
mAbs were not analyzed. Immunofluorescent staining with the mAbs indicated antigenic
17
differences among the 5 CAV strains tested (McNulty et al., 1990b). In another study, 8
mAbs were generated but those lacked virus–neutralizing activity (Chandratilleke et al.,
1991). Recently, one VP1–specific mAb was established by immunization of mice with
truncated recombinant VP1; however, its virus–neutralizing activity was not evaluated
(Lien et al., 2012). Thus, the neutralizing epitopes of CAV remain poorly understood. Scott
et al. (1999) reported that most of the molecularly cloned viruses derived from the Cux–1
strain after 310 cell culture passages showed weak reactivity to the neutralizing 2A9 mAb
(McNulty et al., 1990b) compared with the low–passaged cloned viruses, and the aa at
position 89 in VP1 appeared to be crucial for determining its reactivity with the mAb 2A9.
MAbs could be very useful and powerful tools for understanding the pathogenesis,
isolate characterization, and epidemiology, or for improving CAV diagnosis. In this
chapter, I describe the production and characterization of mAbs to CAV and the expected
epitopes recognized by neutralizing mAbs. To the best of my knowledge, this is the first
report on epitope mapping of VP1 using neutralizing mAbs. Furthermore, we also
genetically characterized two antigenic CAV groups that were differentiated by the mAbs.
1.2. Materials and methods
Cell Culture
MSB1 cells, floating cells, were cultured in GM consisting of RPMI 1640 medium
(Nissui Pharmaceutical Co., Ltd.) supplemented with 10% FBS and 10% Daigo’s GF 21
growth factor (Wako Junyaku) in a humidified incubator with 5% CO2 at 39.5°C.
18
Virus and virus titration
The following CAV strains were used: A1/76, A2/76, AO/77, G1/74, G3/78, G5/79,
G6/79, KY/80, and NI/77 (Yuasa and Imai, 1986); HY/80, G7/91, IBA/94, NI/92, and
HK1/13, which were isolated from the diseased chicks infected with CAV in Japan
(unpublished data); CAA82–2 (Otaki et al., 1987); and a CAV vaccine strain 26P4 which
was obtained from a vaccine (Intervet).
Titers of CAV were determined as described by Imai and Yuasa (1990). Briefly, 20 µl
of a 10–fold serially diluted virus solution was added to wells of a 96–well microplate
containing 200 µl MSB1 cells (2 × 105 cells/ml) in GM. Four wells were used for each
virus dilution. The inoculated cells were passaged every 3 days, in which 40 µl of the cell
suspension was transferred to a new well including 200 µl of GM. The wells without virus
growth were determined after 8 passages. The cultures showing red color (no cell growth)
due to CPE were regarded as CAV–positive (Yuasa, 1983). Virus titers were quantified as
the TCID50 by the Behrens–Kärber method (Behrens and Kärber, 1934).
Mouse immunization and mAb production
A2/76 propagated in MSB1 cell cultures was partially purified and concentrated as
described previously (Imai et al., 1991), and then used as the inoculum for 4 BALB/c mice
(female, 6 weeks of age). The virus titer of the inoculum was approximately 1010
TCID50/ml. Each mouse was immunized with 3 intraperitoneal injections of 0.1 ml of the
inoculum emulsified in Freund’s adjuvant (Sigma-Aldrich) every 3 to 4 weeks. The
antibody titer to CAV in mouse sera was measured by IFAT. One mouse showing the
highest fluorescent antibody titer was intravenously injected with 0.1 ml of the inoculum.
Four days later, spleen cells were fused with P3X63Ag8U.1 myeloma cells in the presence
19
of polyethylene glycol, and the fused cells were selected and cultivated in GM
supplemented with hypoxanthine, aminopterin, thymidine, endothelial cell growth
supplement (Becton Dickinson), and insulin–transferrin–selenium–S supplement (Life
Technologies) according to standard procedures. Antibody–positive hybridoma cells were
selected by IFAT and cloned 2 or 3 times by limiting dilution.
Ascitic fluids containing mAb was obtained by intraperitoneal injection with
approximately 107 hybridoma cells into a BALB/c mouse that had been primed with
incomplete Freund’s adjuvant, as described previously (Harlow and Lane, 1988). Isotypes
of mAbs were determined in an ELISA using a commercial kit (mouse monoclonal
antibody isotyping reagents; Sigma–Aldrich). Ascitic fluids containing each mAb were
used as mAb in most experiments.
An IgG fraction of the ascites including mAb was precipitated by 33% saturated
ammonium sulfate and dialyzed against PBS. Protein concentration of the semi–purified
IgG mAb was determined using the Lowry method (Lowry et al., 1951).
All mouse studies were conducted in compliance with the institutional rules for the
care and use of laboratory animals, and using protocols approved by the relevant
committee at the institution.
Immunoprecipitation (IP) and Western blotting (WB)
A2/76–infected and uninfected MSB1 cells (negative control) were harvested at 48
hpi. IP was conducted according to the instructions of a commercial kit
(Immunoprecipitation Kit, Roche). Briefly, the cells (107 cells/ml) were lysed in 50 mM
Tris–HCl (pH 7.5) containing 150 mM NaCl, 1% nonidet P40, and 0.5% sodium
deoxycholate. The lysed cells were labeled with biotin–7–NHS (EZ–LinkTM Sulfo–NHS–
20
LC–Biotinylation Kit, Thermo), according to manufacturer instructions. The lysates were
precleaned by incubation with Protein G–agarose and mouse IgG–agarose (Sigma–
Aldrich). The supernatant samples collected by centrifugation were incubated with mAb
for at least 3 h at 4°C, and then, 50 µl of Protein G–agarose was added to the mixture
followed by incubation for at least 3 h at 4°C to form the complexes containing antigens,
mAb and Protein G–agarose. After centrifugation of the complex, 50 μl gel loading buffer
(0.06 M Tris–HCl, pH 6.8; 10% (w/v) glycerol; 2% (w/v) SDS; 0.005% (w/v)
bromophenol blue) was added per the complex pellet. The immunoprecipitated complexes
were boiled for 5 min and quenched on ice. The mAb to influenza A virus nucleoprotein
(Serotec Ltd.) was used as a negative control.
The obtained samples were then applied to 17% low bis– polyacrylamide slab gels
according to the method described by Hirano (1989). Transfer of the proteins from the gel
to a nitrocellulose membrane (0.22–µm pore size; Bio–Rad Laboratories) was conducted
using a semi–dry apparatus in transfer buffer (48 mM Tris base, 39 mM glycine, 20%
methanol, 1.3 mM SDS). Nonspecific binding sites on the membrane were blocked by
incubation with 3% BSA in PBS.
Biotin–labeled viral proteins were detected by a streptavidin–horseradish conjugate
(Sigma–Aldrich), and visualized with a chemiluminescent substrate using LAS–3000 (Fuji
Film). A molecular–weight standard (Precision Plus ProteinTM WesternCTM Standards;
Bio–Rad Laboratories) was incubated with Precision StreptTactin–HRP conjugate (Bio–
Rad Laboratories) and visualized as described above.
21
Expression of VP1 recombinant protein in mammalian cells
The full-length of the gene coding the VP1 protein was amplified using the following
primers: CAV–VP1–F EcoRI: 5’−GCGGAATTCATGGCAAGACGAGCTCGCAGA−3’
and CAV–VP1–R XhoI 5’−AATCTCGAG TCAGGGCTGCGTCCCCCAGTA−3’. The
PCR product was digested with restriction enzymes, EcoRI and XhoI, and purified using a
GENECLEAN® II Kit (MP Biomedicals). The purified VP1 gene was ligated into the
pcDNA3.1 (+) vector (Invitrogen) using a DNA ligation kit (Takara) according to the
manufacture’s instruction, and transformed into DH5α competent cells (Takara). After
culturing the bacteria overnight at 37°C, plasmid DNA was extracted by a Miniprep kit
(QIAGEN). The constructed plasmid, pcDNA3.1 (+)−VP1, was characterized by
restriction enzymes digession (EcoRI and XhoI) or sequencing. For the purpose of cell
transfection, the plasmid DNA was purified from bacteria using a EndoFree Plasmid Kit
(QIAGEN).
To express the VP1 protein in mamalian cells, 0.3 µg of pcDNA3.1 (+)−VP1 plasmids
per well were transfected into COS7 cells cultivated in a Lab–Tek® Chamber Slide
(NUNC) by TransIT–LT1 Transfection Reagent (Mirus Bio) according to the
manufacture’s instruction. Mock cells were transfected with pcDNA3.1 (+) plasmid alone.
At 36 h post transfection, the cells were fixed with acetone for 10 min and subjected to the
IFAT with mAbs as described below.
IFAT
IFAT using MSB1 cells was performed to detect CAV antigens or antibodies according
to the method described by Yuasa et al. (1985). Briefly, A2/76–infected MSB1 cells were
smeared onto a glass microscope slide, dried, and fixed with acetone for 10 min. The
22
antigen slides were incubated with the culture supernatant of hybridoma cells, ascitic fluids
containing mAbs, semi-purified mAb at approximately 3 µg/ml, rabbit antiserum to VP1
peptide (1:200), or chicken antiserum to A2/76 (1:40), and then with FITC–conjugated
rabbit anti–mouse IgG (Rockland), goat anti–rabbit IgG (Sigma–Aldrich), or rabbit anti–
chicken IgG (Rockland), respectively at 37°C for 30 min after washing with PBS (pH 7.4).
Anti–VP1 peptide serum was prepared by immunizing a rabbit with a peptide
(CWDVNWANSTMYWESQ; QIAGEN) specific to CAV VP1. The fluorescent signal was
observed under a fluorescence microscope (Biorevo BZ–9000, Keyence). DAPI (Sigma–
Aldrich) was used to counterstain the cell nuclei.
Viral antigen expression kinetics
The A2/76–infected cells were prepared on a glass microscope slide at 6, 12, 24, 36,
60, and 72 hpi, as described above. The antigen slides were then used for IFAT with the
ascetic fluids containing mAbs to examine the kinetics of viral antigen expression. The
fluorescent signals were observed under the fluorescence microscope.
Co–staining
Co–staining of A2/76–infected MSB1 cells with mAbs was conducted to examine the
localization of antigens recognized by each mAb. Briefly, the antigen slides were incubated
with a combination of each of 2 ascitic fluids containing mAbs (1:100) as primary
antibodies at 37°C for 30 min, and then with a combination of 2 isotype–conjugates [rabbit
anti–mouse IgG1–Rhodamine (Rockland), goat anti–mouse IgG2a–FITC (Southern
Biotechnology), or goat anti–mouse IgG2b–Rhodamine (Santa Cruz Biotechnology)] after
washing with PBS. DAPI was also used to counterstain the cell nuclei. The localization of
23
antigens detected by mAbs was analyzed using a confocal microscope (Leica
Microsystems).
Blocking IFAT
The A2/76–infected MSB1 cells prepared as described above were reacted with
mAbs at 5 µg/ml (MoCAV/F2, F8, or F11) or 200 µg/ml (MoCAV/E6) for 30 min at 37°C.
After washing with PBS, mAbs that were labeled with R-phycoerythrin fluorescence using
a Zenon® mouse IgG labeling kit (Life Technologies) were reacted for 30 min. After
washing, the fluorescent signal was observed under the fluorescence microscope.
VNT
A VNT was performed according to the microtest method described by Imai and
Yuasa (1990), which is based on two main methods: an α–neutralization procedure
(constant–mAb, diluted–virus) and a β–neutralization procedure (constant–virus, diluted–
mAb). Briefly, in the α–procedure, 10–fold stepwise dilutions of CAV were mixed with
ascitic fluids containing mAb (1:100) or GM (virus control), and then the mixtures were
incubated overnight at 4°C. Afterward, 20 µl of each mixture was inoculated to each of 4
wells with 200 µl of MSB1 cells (2 × 105 cells/ml). The inoculated cells were passaged
every 3 days. The virus titer of the mixture was deterimed as described above, and the
neutralizing index was calculated based on the differences of virus titers (log10 TCID50)
between the mixtures with mAb and the virus control.
In the β–procedure, serial 2–fold dilutions of ascitic fluids containing mAb,
beginning with a 1:100 dilution for the A2/76 strain or with a 1:2 dilution for the escape
mutants, were mixed with an equal amount of CAV containing 200 TCID50/0.1ml. The
24
mixture was incubated overnight at 4°C and then inoculated into the wells containing cells,
followed by cell passaging as described above. Endpoint titers corresponding to 50%
neutralization were calculated by the Behrens–Kärber method. The reciprocal of the
highest dilution of mAb neutralizing 50% of CAV was taken as the antibody titer.
Selection of escape mutants
The undiluted virus stock of the A2/76 strain (approximately 107 TCID50/ml) was
mixed with ascitic fluids containing mAbs (1:10). Original antibody titers of the 3 mAbs
used are shown in Table 1.1. After incubation for 1.5 h at 37°C followed by overnight
incubation at 4°C, the mixture was inoculated into 7 test tubes containing MSB1 cells (2 ×
105 cells/ml), and the inoculated cells were passaged every 3 days up to 8 times. The
viruses that were not neutralized, indicated by the red color of the culture, were cloned 2 or
3 times by limiting dilutions using MSB1 cells.
DNA extraction and PCR
Viral DNA was extracted from CAV–infected MSB1 cell culture fluids using a
QIAamp DNA blood Mini kit (QIAGEN).
Primers for amplification and sequencing of full–length of coding region of CAV VP1,
VP2, VP3 genes described by Zhang et al. (2013) were used. The internal primers were
selected based on the Cux–1 sequence [GenBank accession No. M55918, Noteborn et al.
(1991)] and the details are as follows:
Name of primers Sequences of primers Position
CAV–CQ1F 5’–CAATCACTCTATCGCTGTGT –3’ 608–628
CAV–CQ1R 5’–TTCGTCCATCTTGACTTTCT–3’ 47–67
CAV–CQ2F 5’–GGCTACTATTCCATCWCCATTCT–3’ 14–37
CAV–CQ2R 5’–GCTCGTCTTGCCATCTTACA–3’ 848–879
25
VP1–F full–EcoRI 5’–GCGGAATTCATGGCAAGACGAGCTCGCAGA–3’ 853–874
VP1–1246R 5’–AGACCCGTCCGCAATCAACTC–3’ 1226–1246
VP1–658F 5’–GACCCGACGAGCAACAGTACC–3’ 1658–1678
VP1 full–R–XhoI 5’–AATCTCGAG TCAGGGCTGCGTCCCCCAGTA –3’ 2181–2202
VP2 full–F 5’–GAGCGCACATACCGGTCGG –3’ 333–352
VP2 full–R 5’–CGAAGTCGCTTGAGGTGGTGC–3’ 914–934
The PCR amplification was carried out using TaKaRa Ex Taq (Takara Bio Inc) using
the following cycling profile: initial denaturation of 94°C for 5 min, followed by 35 cycles
of denaturation, annealing and extension at 94°C for 30 s, 50°C to 65°C for 30 s which
depends on specific primers and 72°C for 1 min, respectively, and the final extension was
carried out at 72°C for 10 min. The PCR products were then analyzed by 2% agarose gel
electrophoresis and imaged with UV light.
Sequencing and phylogenetic analysis
VP1, VP2, and VP3 genes of escape mutants of A2/76 selected by mAbs, and VP1
gene sequences of CAV strains used in this study, except A2/76, G6/79, 26P4, and CAA
82–2, were determined by direct sequencing using a BigDye Terminator v3.1 cycle
sequencing kit according to the manufacturer’s instructions (Life Technologies).
Nucleotide sequencing was performed using an Applied Biosystems 3500 Genetic
Analyzer (Life Technologies). VP1 gene sequences of A2/76, G6/79, 26P4, and CAA 82–2,
and VP2 and VP3 genes of A2/76 were obtained from GenBank.
Nucleotide sequences obtained were analyzed using GENETYX ver. 10 software
(GENETYX Corp.) and compared with other available sequences using the BLAST
program. The nucleotides and translated aa sequences were aligned with Clustal W
(Thompson et al., 1994). A phylogenetic tree of the VP1 gene was constructed using the
maximum likelihood method based on the Poisson correction model, supported by 500
26
bootstrap replicates. The initial trees for the heuristic search were obtained automatically
by applying neighbor–joining and BioNJ algorithms to a matrix of pairwise distances
estimated using the maximum composite likelihood approach, and then the topology with a
superior log–likelihood value was selected. The tree was drawn to scale, with branch
lengths corresponding to the number of substitutions per site. All positions containing
alignment gaps and missing data were eliminated in complete deletion (complete deletion
option). Evolutionary analyses were conducted in MEGA 5 software (Tamura et al., 2011).
1.3. Results
Establishment of hybridomas secreting mAb to CAV
Using IFAT, the hybridomas secreting antibody to CAV were examined. As the result, four
hybridomas secreting CAV antibodies were established from a mouse immunized with
CAV, and were designated as MoCAV/F2 (IgG1), MoCAV/F8 (IgG1), MoCAV/F11
(IgG2b), and MoCAV/E6 (IgG2a). Three of the mAbs (MoCAV/F2, MoCAV/F8, and
MoCAV/F11) showed neutralizing activity at titers ranging from 1:12,800 to 1:25,600
(Table 1.1), but MoCAV/E6 did not.
The immunofluorescent staining patterns of the A2/76–infected MSB1 cells with the
mAbs could be largely classified into 2 types when observed within 36 hpi, as shown in
Figs. 1.1a and 1.2. Diffused, irregularly shaped granular staining with MoCAV/F2,
MoCAV/F8, and MoCAV/F11 was observed in the enlarged infected cells, whereas
scattered, spherically shaped antigens of various sizes were observed in the infected cells
reacted with MoCAV/E6.
Immunoprecipitation showed that MoCAV/F2, MoCAV/F8, and MoCAV/F11
precipitated a protein band of an estimated size of 50 kDa, corresponding to the VP1
27
protein (50 kDa) in the infected MSB1 cell lysates (Fig. 1.1c); however, MoCAV/E6 did
not precipitate this protein and also failed to precipitate any other viral proteins. An mAb to
the nucleoprotein of influenza A virus was used as a control, and did not precipitate CAV
proteins.
The VP1 recombinant proteins expressed in COS7 cells using a pcDNA3.1 (+) vector
were reacted with neutralizing mAbs as well as anti–VP1 peptide antibodies (Fig. 1.1d),
while they were not with MoCAV/E6 (data not shown).
Viral protein expression in A2/76–infected MSB1 Cells
The kinetics of the expression of viral antigens was examined using 3 neutralizing
mAbs and MoCAV/E6 in infected MSB1 cells fixed at different time points (6, 12, 24, 36,
60, and 72 hpi).
Although there was no fluorescent signal observed at 6 hpi (data not shown), positive
immunofluorescent staining was observed with all of the mAbs tested at 12 hpi; however,
only MoCAV/E6 showed clearer and stronger fluorescence compared with the other mAbs
(Fig. 1.1a).
The irregular–shaped small granules detected by the 3 neutralizing mAbs became
stronger and clearer at 24 hpi than at 12 hpi, and reached a maximal level at 36 hpi, when
they were distributed all over the cells. However, the staining pattern markedly changed
toward misshapen fluorescent staining of various sizes at 60 hpi (Fig. 1.1a) and 72 hpi
(data not shown), which was most likely due to CPE. Many infected cells with misshapen
antigens seemed not to be intact. The number of DAPI–positive cells with typical staining
patterns reduced over the time course. An anti–VP1 peptide antibody confirmed the
presence of VP1 antigen in the infected MSB1 cells (Fig.1.1a).
28
The intensity of fluorescent signals detected by MoCAV/E6 peaked at 24 hpi and 36
hpi. MoCAV/E6 did not change its fluorescent staining pattern (scattered, spherical
structures of various sizes) during the observation period, although the number of infected
cells reduced and the signal became weak at later time points of 60 hpi (Fig. 1.1a) and 72
hpi (data not shown).
Co–staining of A2/76–infected MSB1 cells with mAbs
Co–staining patterns of A2/76–infected MSB1 cells with MoCAV/E6 and neutralizing
mAbs were analyzed with a confocal microscope. Antigens detected by MoCAV/F11 and
MoCAV/E6 were localized in the nuclei of infected cells as indicated by DAPI
counterstaining (Fig. 1.1b). The merged image of antigens detected by both mAbs
indicated that the antigen signals seemed to partially overlap. The same results were
obtained in combinations of MoCAV/E6 with other mAbs (data not shown).
Blocking IFAT
As shown in Fig. 1.2, bindings of fluorescein–conjugated MoCAV/F2 and MoCAV/F8
were mutually competitively blocked, whereas the F11 and E6 competitors did not block
the binding of the conjugated MoCAV/F2 and MoCAV/F8. On the other hand, fluorescein–
conjugated MoCAV/F11 and MoCAV/E6 were not blocked by any competitor, except for
homologous mAbs.
Reactivity of neutralizing mAbs to heterologous CAV strains in the VNT
As shown in Table 1.2, MoCAV/F2 and MoCAV/F8 neutralized all of the CAV strains
examined. By contrast, MoCAV/F11 could not neutralize G3/78, G5/79, G6/79, NI/77, and
29
HY/80. Thus, the CAV strains were antigenically divided into 2 distinct groups based on
MoCAV/F11 reactivity; mAb antigenic Group 1 included CAVs recognized by
MoCAV/F11, and Group 2 included CAVs not recognized by this mAb.
Phylogenetic analysis
Phylogenetic analysis of full–length deduced VP1 aa sequences of the 2 mAb
antigenic group strains in comparison with other strains available in GenBank resulted in 3
distinct clusters (clusters I, II, and III) (Fig. 1.3a). The aa profiles in VP1 sequences of
CAV strains are shown in Fig. 1.3b.
MAb antigenic Group 2 strains (G5/79, G6/79, NI/77, and HY/80) with the aa profile
of I75, L97, Q139, Q144, which were not neutralized with MoCAV/F11, fell into cluster II,
and they formed a single genetic group with 5 other strains (TR20, Arg729, 704, CAV–E,
and SMSC–1) with the same profile. Although the G3/78 strain had the same profile, it was
not used in phylogenetic analysis since its complete sequence was not conclusively
determined due to the appearance of double peaks in some positions. Only 6 of all the
strains including a recent Japanese isolate HK1/13 classified in cluster II had a different
profile (I/V75, L/M97, N/Q139, H/Q144). Cluster II formed a larger genetic group than the
other clusters.
Among antigenic Group 1 strains that were neutralized by MoCAV/F11, the G1/74 and
KY/80 strains (V75, M97, K139, E144) were classified into cluster I with the common
profile of V75, M97, K139, E/D/N144. The other Group 1 strains (A2/76, CAE26P4, 82–2,
AO/77, A1/76, G7/91, IBA/94, and NI/92) were classified in cluster III with the common
profile of V75, M97, K139, E144.
30
Antigenic and genetic characterizations of escape mutants
One escape mutant was selected for each of MoCAV/F2, MoCAV/F8, and
MoCAV/F11, designated as EsCAV/F2, EsCAV/F8, and EsCAV/F11, respectively.
In IFAT of the MSB1 cells infected with escape mutants, MoCAV/F11 reacted to both
EsCAV/F2 and EsCAV/F8, while both MoCAV/F2 and MoCAV/F8 recognized only
EsCAV/F11. Chicken polyclonal antibody against A2/76 reacted with all of the escape
mutants (Fig. 1.4).
MoCAV/F8 and MoCAV/F11 did not neutralize their corresponding escape mutants,
EsCAV/F8 and EsCAV/F11 (Table 1.1). Moreover, MoCAV/F8 did not neutralize
EsCAV/F2. By contrast, MoCAV/F2 neutralized both EsCAV/F2 and EsCAV/F8 with low
dilutions of 1:32 and 1:152, respectively. To confirm the reactivity of the 2 mAbs
(MoCAV/F2 and F8) against escape mutants (EsCAV/F2 and EsCAV/F8) in VNT, the
mAbs diluted at 1:10 were examined for the escape mutants using IFAT. As the result,
neither MoCAV/F2 nor MoCAV/F8 reacted to the mutants (data not shown).
Comparison of the VP1 aa sequence of the parent virus (A2/76) with those of the 3
escape mutants revealed the deletion of threonine (T) and alanine (A) at positions 89 and
90 (T89+A90) in EsCAV/F2, a single aa change of isoleucine (I) to T at position 261
(I261T) in EsCAV/F8, and a glutamic acid (E) to glycine (G) change at position 144
(E144G) in EsCAV/F11 (Table 1.1). The 3 escape mutants did not show any aa changes in
VP2 and VP3 in comparison to the parent virus (data not shown).
The titer of EsCAV/F2 was reduced by 1.75 log compared with that of the parent
A2/76 strain in MSB1 cells (Table 1.1).
1.4. Discussion
Three mAbs (MoCAV/F2, MoCAV/F8, MoCAV/F11) possessed neutralizing activity
31
against A2/76 and were directed against VP1 in immunoprecipitation analysis, while
MoCAV/E6 did not show neutralizing activity and did not immunoprecipitate any of the
VPs (Fig. 1.1c). This result was confirmed using COS7 cells expressing the VP1
recombinant proteins (Fig. 1.1d), while MoCAV/E6 did not react to the recombinant
proteins (data not shown). Moreover, MoCAV/E6 showed a different staining pattern from
the other mAbs as shown in Figs. 1.1a and 1.2.
A co–staining study showed partial co–localizations of the antigens detected by
MoCAV/E6 and VP1 neutralizing mAbs (Fig. 1.1b). However, the time–course staining
pattern of MoCAV/E6 was different from that of VP1 neutralizing mAbs (Fig. 1.1a). Thus,
MoCAV/E6 seemed not to recognize VP1, although further study should be performed to
define the target protein of MoCAV/E6.
Douglas et al. (1995) reported that fluorescent VP1 antigens in infected MSB1 cells
were not detected using a non–neutralizing mAb against VP1 (1H1) until 30 hpi, although
both VP2 and VP3 antigens were detectable as early as 12 hpi. By contrast, the present
study showed that neutralizing mAbs (VP1–specific) could detect antigens as early as 12
hpi, indicating that VP1 proteins are produced early in contrast to the previous report,
although the fluorescent intensity was weak and the number of positive cells was limited at
this time point (Fig. 1.1a). The reason for this discrepancy between studies is unclear.
Although the MSB1 cells were similarly used in both studies, the CAV strains tested were
different. However, since the CAV strains used in the 2 studies, A2/76 and Cux–1, do not
appear to have different biological properties, the mAbs used in the experiment might have
affected the results. Our mAbs (MoCAV/F2, F8, and F11) showed neutralization activity,
whereas the mAb 1H1 lacks this activity (McNulty et al., 1990b). Yuasa et al. (1985) found
that fluorescent granules were detectable in a few infected MSB1 cells that reacted with
32
polyclonal antibodies to CAV as early as 12 hpi, and the number of fluorescent positive
cells increased gradually by 24 hpi, which is consistent with our results. In addition, anti–
VP1 peptide antibody results supported the early detection of VP1 antigens recognized by
the neutralizing mAbs (Fig. 1.1a).
Although several mAbs against CAV or recombinant VP1 proteins have been
developed (Chandratilleke et al., 1991; Lien et al., 2012; McNulty et al., 1990b), there is a
lack of important information related to the antigenicity of VP1, especially with respect to
neutralizing epitopes. We identified the neutralizing epitopes on VP1 in the current study.
Blocking IFAT showed that the binding of MoCAV/F2 and MoCAV/F8 to the A2/76–
infected cells was blocked by mutual competition between the mAbs (Fig. 1.2). In addition,
the 2 mutants (EsCAV/F2 and EsCAV/F8) reacted with neither MoCAV/F2 nor MoCAV/F8
(Fig. 1.4). These results suggested that these mAbs recognize the same epitope on VP1.
However, VP1 aa analysis of these mutants revealed two different mutations in the
epitopes; namely, the deletion of T89+A90 in VP1 of EsCAV/F2, and I261T in VP1 of
EsCAV/F8 (Table 1.1). Kaverin et al. (2002) revealed the 2 antigenic sites on the
hemagglutinin molecule of the H5 subtype of avian influenza virus that have distinct aa
positions recognized by analyzed escape mutants but are topographically close in the
three–dimensional structure, and partially overlap in reaction with mAbs. Therefore, it is
likely that the antigenic sites, including aas T89+A90 and I261, are topographically close
in the VP1 structure.
Scott et al. (2001) reported a variant CAV, P310 2A9–resist, that resists neutralization
by mAb 2A9 (McNulty et al., 1990b), which was selected from a Cux–1 virus strain that
had been passaged 310 times in MSB1 cells. Fluorescent VP1 antigens were not detectable
by the mAb in the cells infected with P310 2A9–resist virus, even at low antibody dilutions
33
(1:100), whereas the low–passage virus produced positive staining at high dilutions
(≥1:80,000). Therefore, the authors suggested that the aa substitution at position 89 of VP1
was a key determinant of mAb 2A9 reactivity, because P310 2A9–resist virus, which has
A89 instead of T89, produced no immunofluorescence (1:100). Although mAb 2A9
neutralized the P310 2A9–resist virus at very low dilutions (1:5), whether the mAb could
react to antigens at lower dilutions than 1:100 in IFAT was not evaluated. In this study, a
similar phenomenon was observed. Although MoCAV/F2 did not neutralize EsCAV/F2 at
low dilutions (1:100), unexpectedly, it was neutralized at even lower dilutions (1:32) (Table
1.1). Thus, these results suggest that the aa change of T89A in VP1 is not necessarily a key
determinant of MoCAV/F2 reactivity, since EsCAV/F2 lacks the aas T89+A90.
These results raise questions as to why MoCAV/F2 could neutralize EsCAV/F2 at a
high mAb concentration (Table 1.1). One possible explanation for this phenomenon may be
that the epitope was not mutated completely; therefore, the corresponding mAb was still
able to bind to it but with weaker affinity. However, MoCAV/F2 also neutralized
EsCAV/F8, which possesses T89+A90, at a high mAb concentration (Table 1.1). Although
the reason for this phenomenon is unclear, the results suggest that complete binding of
MoCAV/F2 to the epitope might require the coexistence of an antigenic site including I261,
which is recognized by MoCAV/F8. On the other hand, we could not explain why
MoCAV/F2 did not recognize the antigens in the infected cells with EsCAV/F2 even when
using very low dilutions (e.g., 1:10), but this could be due to the low sensitivity of IFAT.
MoCAV/F8 could not only neutralize EsCAV/F8 but also EsCAV/F2 (Table 1.1). This
unexpected phenomenon may indicate that the binding of MoCAV/F8 to the epitope
requires coexistence of the epitope recognized by MoCAV/F2.
34
However, the possibility of a single epitope recognized by these mAbs cannot be
denied considering the similar reactivity of these mAbs to the VP1 protein, although the
escape mutants of MoCAV/F2 and MoCAV/F8 displayed different aa mutations in VP1. In
this study, only one escape mutant for each neutralizing mAb was examined. More escape
mutants should be examined to obtain further information about antigenic epitopes in VP1.
Wang et al. (2009) reported that the structure protein genes of VP1 had undergone
positive selection, and 8 positively selected aa sites (75, 125, 139, 141, 144, 287, 370, 447)
were identified. In this study, only EsCAV/F11 showed an aa change (E144G)
corresponding to one of the positively selected aa sites of VP1 reported by Wang et al.
(2009). However, I think that it is likely that the escape mutants were selected by antibody
selection pressure rather than the positive selection.
Renshaw et al. (1996) indicated that one or both of the aa differences at positions 139
and 144 affected the rate of replication or the spread of infection in MSB1 cells (sublines L
and S). In this study, the aa change E144G did not affect the replication rate of CAV in
MSB1 cells (Table 1.1). Therefore, both of the aa changes at positions 139 and 144 might
be required to affect the replication of CAV in MSB1 cells.
In this study, the CAV strains examined were phylogenetically grouped into 3 main
clusters (Fig. 1.3a and b). In cluster II, corresponding to genotype II described by Islam et
al. (2002), 26 of the 32 strains had the aa profile I75, L97, Q139, Q144, including 4
Japanese CAV strains of mAb antigenic Group 2 that were not neutralized by MoCAV/F11.
A2/76, with the V75, M97, K139, E144 profile, lost the neutralizing epitope recognized by
MoCAV/F11 owing to the aa change E144G (Table 1.1). However, this change was not
observed in the mAb antigenic Group 2 strains with the profile of I75, L97, Q139, Q144,
which naturally lack the epitope recognized by MoCAV/F11. MoCAV/F11 reacted to the
35
recent Japanese isolate HK1/13 (cluster II), with the V75, L97, N139, Q144 profile, in
IFAT (data not shown). Thus, the aa change at position 144 is not necessarily associated
with the loss of binding ability to MoCAV/F11. However, further studies are required to
determine whether all CAV strains with the profile of I75, L97, Q139, Q144 naturally lack
the antigenic site(s) recognized by MoCAV/F11.
In conclusion, I established 4 CAV mAbs, 3 of which (MoCAV/F2, MoCAV/F8, and
MoCAV/F11) had neutralizing activity and recognized the CAV VP1 protein. Analysis of
the escape mutants of the neutralizing mAbs revealed at least 2 neutralizing epitopes on the
CAV VP1 protein, which have not been reported previously, to my knowledge. As the
reactivity of MoCAV/F2 and MoCAV/F8 to VP1 was similar, the existence of a single
epitope recognized by these mAbs cannot be denied. The CAV strains evaluated could be
differentiated into 2 distinct antigenic groups by MoCAV/F11, which could be associated
with specific aa profiles of VP1. Mutations in VP1 are known to affect pathogenicity in
chickens or viral replication in cells. However, there is no consistent molecular biological
evidence to explain the events, and there are still many aspects that remain unresolved with
respect to CAV biology.
1.5. Summary
Three (MoCAV/F2, MoCAV/F8, MoCAV/F11) of 4 mouse monoclonal antibodies
(mAbs) established against the A2/76 strain of chicken anemia virus (CAV) showed
neutralization activity. Immunoprecipitation showed a band at approximately 50 kDa in
A2/76–infected cell lysates by neutralizing mAbs, corresponding to the 50–kDa capsid
protein (VP1) of CAV, and the mAbs reacted with recombinant VP1 proteins expressed in
COS7 cells. MoCAV/F2 and MoCAV/F8 neutralized the 14 CAV strains tested, whereas
36
MoCAV/F11 did not neutralize 5 of the strains, indicating distinct antigenic variation
among the strains. In blocking immunofluorescence tests with the A2/76–infected cells,
binding of MoCAV/F11 was not inhibited by the other mAbs. MoCAV/F2 inhibited the
binding of MoCAV/F8 to the antigens and vice versa, suggesting that the 2 mAbs
recognized the same epitope. However, mutations were found in different parts of VP1 of
the escape mutants of each mAb: EsCAV/F2 (deletion of T89+A90), EsCAV/F8 (I261T),
and EsCAV/F11 (E144G). Thus, the epitopes recognized by MoCAV/F2 and MoCAV/F8
seemed to be topographically close in the VP1 structure, suggesting that VP1 has at least 2
different neutralizing epitopes. However, MoCAV/F8 did not react to EsCAV/F2
(containing the epitope recognized by this mAb) or to EsCAV/F8, suggesting that binding
of MoCAV/F8 to the epitope requires coexistence of the epitope recognized by
MoCAV/F2. In addition, MoCAV/F2, with a titer of 1:12,800 to the parent strain,
neutralized EsCAV/F2 and EsCAV/F8 with low titers of 32 and 152, respectively. The
similarity of the reactivity of MoCAV/F2 and MoCAV/F8 to VP1 may also suggest the
existence of a single epitope recognized by these mAbs.
37
Table 1.1. Characterization of escape mutants
Escape mutant Amino acid
change
Titer of virus
(TCID50/ml)
Neutralizing antibody titers of
mAb to escape mutants1)
F2 F8 F11
EsCAV/F2 Deletion of aa
T89+A90 105.25 32 <2.8 6,888
EsCAV/F8 I261T 107.25 152 <2.8 46,340
EsCAV/F11 E144G 106.5 9,741 19,483 <2.8
Parent virus
A2/76 107.0 12,800 25,600 25,600
1) The reciprocal of the highest dilution of mAb neutralizing 50% of the virus was taken as
the antibody titer.
38
Table 1.2. Virus neutralization test1) with mAbs against various CAV strains
Various Japanese CAV strains
G1 /74
AO /77
CAA 82–2
A1 /76
IBA /94
KY /80
G7 /91
NI /92
26P4 G3 /78
G5 /79
G6 /79
NI /77
HY /80
Genetic cluster2)
I III NI4) II
Amino acid profile3)
V75/M97/K139/E144 NI I75/L97/Q139/Q144
Neutralizing index of mAbs against various CAV strains
F2 4.5 4.5 4.5 3.5 2.0 >5.5 3.0 2.5 >4.0 3.0 3.0 4.0 4.0 4.0
F8 >3.5 3.5 4.5 4.0 2.5 >3.5 4.0 3.5 >5.0 >3.5 2.5 >4.0 4.5 4.0
F11 >4.0 5.5 3.5 >5.0 >4.0 >5.5 3.0 >4.0 >6.0 0.5 0.0 0.0 0.5 0.0
1) Virus neutralization test was performed using the α–neutralization procedure described in the Materials and methods; 2) Genetic clusters were shown in Fig. 1.3a; 3) Amino acid profiles were shown in Fig. 1.3b; 4) NI: non-identified.
39
Fig. 1.1a. Viral protein expression kinetics in CAV A2/76–infected MSB1 cells.
Immunofluorescent antibody tests were conducted to detect antigens using semi-purified
mAbs (MoCAV/F2, F8, F11, E6) at 3 µg/ml and a rabbit serum anti–VP1 peptide (1:200).
Mouse normal ascitic fluid was used as a negative control (not shown). Cell nuclei were
counterstained with DAPI. Infected cells were collected at 12, 24, 36, and 60 hpi and used
as antigens. Scale bar = 10 µm.
MoCAV/F2
MoCAV/F11
MoCAV/E6
Anti-VP1 peptide
MoCAV/F8
DAPI
12 hpi 24 hpi 36 hpi 60 hpi
40
Fig. 1.1b. Co–staining of A2/76–infected MSB1 cells with mAbs. The infected cells were
stained with ascitic fluids containing a neutralizing mAb (MoCAV/F11) (1:100) and a non–
neutralizing mAb (MoCAV/E6) (1:100), and then with IgG isotype–specific secondary
antibodies labeled with rhodamine for MoCAV/F11, and with FITC for MoCAV/E6.
Infected cells collected at 36 hpi were used as antigens. Cell nuclei were counterstained
with DAPI. The fluorescent signals were observed under the confocal microscope.
MoCAV/F11 Merge
MoCAV/E6 DAPI
41
Fig
. 1.
1c.
Imm
unop
reci
pita
tion
ana
lysi
s of
A2
/76
-infe
cte
d M
SB
1 ce
lls.
The
inf
ecte
d or
uni
nfec
ted
cel
l ly
sate
s co
llect
ed
at
48
hpi w
ere
bio
tin-la
bele
d an
d im
mun
opre
cip
itate
d w
ith m
Abs
aga
inst
CAV
and
aga
inst
influ
enz
a A
vir
us n
ucle
opro
tein
(N
P).
The
imm
uno
pre
cipi
tate
d sa
mpl
es
we
re a
naly
zed
by S
DS
-PA
GE
, an
d th
en
the
bio
tin-l
abe
led
prot
ein
s w
ere
tra
nsf
err
ed
from
a g
el t
o
a n
itroc
ellu
lose
me
mb
rane
. B
iotin
-la
bele
d vi
ral
pro
tein
s w
ere
de
tect
ed
by a
str
ept
avi
din-
hors
era
dish
con
juga
te a
nd v
isua
lize
d
with
the
ch
emilu
min
esc
ent
sub
stra
te.
M:
mol
ecul
ar-
we
ight
sta
nda
rd.
Se
mi-p
urifi
ed
mA
bs
F2
(MoC
AV/F
2),
F8
(M
oCAV
/F8)
,
F11
(M
oCAV
/F11
), E
6 (M
oCAV
/E6)
, a
nd
NP
(ne
gativ
e c
ont
rol)
we
re u
sed
to i
mm
unop
reci
pita
te v
iral
prot
ein
s in
inf
ecte
d
cells; F2 M
SB1, F8 M
SB1, F11 M
SB1, E
6 M
SB
1,
and
NP
MS
B1
indi
cate
the
tre
atm
ent
of
unin
fe
cte
d ce
lls w
ith m
Abs
desc
ribe
d a
bove
.
42
MoC
AV/F
2
MoC
AV/F
8
MoC
AV/F
11
Ant
i-V
P1
pept
ide
VP
1-ce
lls
Moc
k-ce
lls
Nor
mal
Asc
itic
fluid
Fig
1.1
d. R
ea
ctiv
ity o
f ne
utra
lizin
g m
Abs
with
re
comb
ina
nt
VP
1 pr
ote
ins e
xpre
sse
d in
CO
S7
cells
. T
he f
ull
-le
ngth
of
the
VP
1
gene
wa
s cl
one
d in
to p
cDN
A3.
1 (+
) ve
ctor
. T
he c
onst
ruct
ed
pcD
NA
3.1
(+)-
VP
1 p
lasm
ids
we
re t
rans
fect
ed
int
o C
OS
7 ce
lls.
IFA
T w
ere
con
duct
ed
usin
g th
e V
P1
exp
ress
ed
cells
with
aci
tic f
luid
s co
ntai
nin
g m
Abs
(M
oC
AV/F
2, F
8, F
11)
(1:
100)
and
ant
i-
VP
1 p
ept
ide
ant
ibod
y (1
:200
) a
t 36
h p
ost
tra
nsfe
ctio
n. T
he m
Ab
MoC
AV/E
6 d
id n
ot r
ea
ct i
n t
he I
FAT
(d
ata
not
sho
wn)
.
Mou
se n
orm
al a
sciti
c flu
id w
as
use
d a
s a
ne
gativ
e c
ont
rol.
Mo
ck c
ells
tha
t w
ere
tra
nsfe
cte
d w
ith p
cDN
A3.
1 (+
) ve
ctor
we
re
als
o us
ed
as
nega
tive
con
trol
. Sca
le b
ar
= 1
0 µ
m.
43
Fig. 1.2. Blocking immunofluorescent antibody tests. Blocking tests were conducted
using semi-purified mAbs (MoCAV/F2, F8, F11 at 5 µg/ml, or E6 at 200 µg/ml) as
competitors and mAbs directly labeled with R-phycoerythrin fluorescein. Infected cells
collected at 36 hpi were used as antigens. Scale bar = 10 µm.
44
Fig. 1.3a. Phylogenic analysis of the complete deduced amino acid sequences of the CAV VP1 protein. (▲) indicated the CAV strains neutralized by MoCAV/F11 (G1/74, KY/80, AO/77, 26P4, A2/76, A1/76, CAA 82–2, G7/91, IBA/94, and NI/92); (■) indicated the CAV strains that did not neutralized by MoCAV/F11 (G5/79, G6/79, NI/77, and HY/80). (●) indicated the current Japanese CAV isolate (HK1/13). Sequences from GenBank are indicated with the country name followed by accession number. The phylogenetic tree was constructed using the maximum likelihood method based on the Poisson correction model for amino acids in MEGA5 software, supported by 500 bootstrap replicates. The scale bar shows the number of base substitutions per site. Three major clusters were identified and designated as clusters I, II, and III.
45
Fig. 1.3b. Comparison of the amino acid residues of VP1 of CAV strains in clusters I, II,
and III. Alignment was conducted with Clustal W. Numbering (70–150) was based on the
Cux–1 VP1 sequence (GenBank accession no. M55918, Noteborn et al., 1991). Sequences
of strains used in neutralization tests are underlined. The amino acids at positions 75, 97,
139, and 144 are boxed.
46
Fig. 1.4. Reactivity of mAbs in MSB1 cells infected with escape mutants.
Immunofluorescent antibody tests were conducted using semi-purified mAbs (MoCAV/F2,
F8, and F11) at 3 µg/ml, and chicken anti–A2/76 serum to detect antigens in the cells
infected with escape mutants collected at 36 hpi. Scale bar = 10 µm.
MoCAV/F2 MoCAV/F8 MoCAV/F11 Chicken
anti A2/76 serum
EsCAV/F2
EsCAV/F8
EsCAV/F11
47
Chapter II
Development of a blocking latex agglutination test for the detection of
antibodies to chicken anemia virus
2.1. Introduction
CAV is known to be ubiquitous among chicken flocks throughout the world (Schat and
van Santen, 2008). Usually, field flocks become horizontally infected with CAV without
clinical diseases after diminishing of maternal antibodies by about 3 weeks of age, and then
seroconversion occurred in most breeder chicken flocks from 8 to 12 weeks of age
(McNulty et al., 1988). However, the exposure of antibody–negative breeder flocks to CAV
during the laying period results in vertical transmission of the virus, and causes severe
disease in the progeny (Chettle et al., 1989; Hoop, 1993; Yuasa et al., 1987). Therefore,
vaccination of breeder flocks that failed to show seroconversion should be considered prior
to the collection of eggs. Even SPF chicken flocks reared under very strict hygienic
conditions have been reported to become infected with CAV (Cardona et al., 2000b;
McNulty et al., 1989; Yuasa et al., 1985). Therefore, ensuring that SPF chicken flocks that
supply eggs for vaccine production are free of CAV infection is extremely important (Todd
et al., 1990b), in turn highlighting the importance of monitoring seroconversion in
commercial and SPF breeder flocks.
Currently, 3 serological tests are routinely available for the detection of antibodies to
CAV: ELISA, IFAT, and VNT (Todd et al., 1990b, 1999; von Bülow et al., 1985; Yuasa et
al., 1983b, 1985). Of these, VNT has the highest sensitivity and specificity; however, the
test is time–consuming and laborious. In contrast, IFAT is a relatively simple test to
perform, but requires experienced personnel and frequently yields false–positive results,
48
particularly when sera are tested at lower dilutions (Otaki et al., 1991). Both VNT and
IFAT are unsuitable for testing a large number of samples. However, ELISA is well–suited
for the routine screening of a large number of samples (Lamichhane et al., 1992; Tannock
et al., 2003; Todd et al., 1990b, 1999), but requires approximately 2.5 to 5 h following the
overnight incubation of antigens or a mAb for its completion. Moreover, ELISA
necessitates the purification or semi–purification of antigens from infected cells. Compared
to IFAT, higher rates of false–positive reactions were obtained with certain commercial
ELISA kits (Michalski et al., 1996; Tannock et al., 2003). Therefore, the development of a
reliable, simple, and rapid test for the detection of antibodies to CAV is of utmost
importance.
In this study, the development of a highly sensitive and specific blocking latex
agglutination test (b–LAT) for the detection of antibodies to CAV in chickens is described.
The test is based on the ability of CAV–specific antibodies present in the test sera to block
the binding of CAV antigens, which were prepared from infected cell lysates, to latex beads
coupled with the mAb against CAV. The results were scored as antibody–positive (no
agglutination of mAb–beads) or antibody–negative (agglutination observed).
2.2. Materials and methods
Cell culture and virus
MSB1 cells were cultured in the GM RPMI–1640 (Nissui Pharmaceutical Co., Ltd.)
supplemented with 10% FBS and 10% Daigo’s GF21 growth factor (Wako Junyaku) in a
humidified incubator with 5% CO2 at 39.5°C.
The CAV strain A2/76 (Yuasa and Imai, 1986) was used in the current study. Viral
titers were determined by a microtest method (Imai and Yuasa, 1990), and was described in
49
detail in the subsection of “Cell culture” in the Materials and methods in chapter I. Briefly,
20 µl of a 10–fold serially diluted virus solution was added to wells of a 96–well
microplate containing 200 µl MSB1 cells (2 × 105 cells/ml) in GM. Four wells were used
for each virus dilution. The inoculated cells were passaged every 3 days. The wells without
virus growth were determined after 8 passages. The cultures showing red color (no cell
growth) due to CPE were regarded as CAV–positive (Yuasa, 1983). Virus titers were
quantified as the TCID50 by the Behrens–Kärber method (Behrens and Kärber, 1934).
IFAT
IFAT was performed for the detection of antibodies to CAV in chicken sera using
CAV–infected MSB1 cells, as detailed in Yuasa et al. (1985). In brief, the infected cells
were harvested at 36 h post–infection, and were smeared on to a microscope slide, dried,
and fixed with cold acetone for 10 min. The slides were incubated with chicken serum at
the dilution of 1:40 and then with FITC–conjugated rabbit anti–chicken IgG (Rockland) at
37ºC for 30 min each, followed by observation under a fluorescence microscope (Biorevo
BZ–9000, Keyence) for the measurement of fluorescent signal.
Preparation of CAV antigens
MSB1 cells infected with the CAV strain A2/76 were prepared as described
previously (Yuasa et al., 1985), with slight modifications. In brief, 2 ml CAV solution
(approximately 107 TCID50/ml) was mixed with 107 MSB1 cells, incubated at 39.5°C for 1
h, and suspended in GM at a concentration of 106 cells/ml.
The infected cells with CPE (enlargement of the infected cells) in 1,000 ml of cell
culture were harvested at 72 hpi by centrifugation at 1,500 ×g for 5 min, and washed with
50
PBS (pH 7.4). The supernatant was then removed, and the cells were resuspended in 1 ml
of PBS, followed by 3 freeze–thaw cycles. The cell lysate was subjected to sonication, and
then to centrifugation at 10,000 ×g for 15 min. The supernatant was collected as CAV
antigens, and the pellet of cellular debris was discarded. The titer of CAV in the antigen
solution was approximately 109.5 TCID50/ml. The cell lysates from uninfected MSB–1 cells
were prepared as negative antigens in the same manner. The infected and uninfected cell
lysate antigens were stored at −30°C until use.
Sensitization of latex beads
A neutralizing mAb (MoCAV/F11, IgG2b) (Trinh et al., 2015), which was described in
Chapter I, was used in the current study. The mAb was semi–purified using 50% saturated
aluminum sulfate. Protein concentration of the semi–purified mAb IgG, which was
determined by Lowry method (Lowry et al., 1951), was 6.9 mg/ml.
Polystyrene latex beads (1.0 µm) were coupled with mAb in accordance with the
manufacturer’s instructions (Polysciences, Inc.). In brief, 0.5 ml of a 2.6% (w/v)
suspension of the beads was coupled with 400 µg of mAb IgG. The coupled beads (mAb–
beads) were blocked with 1.0 ml of bovine serum albumin (BSA, 10 mg/ml) for 30 min at
room temperature. Then, the mAb–beads were suspended in 1 ml of storage buffer (1%
BSA, 5% glycerol, and 0.1% sodium azide in PBS), and stored at 4°C until use.
b−LAT
First, 2–fold serial dilutions of CAV antigens in PBS were incubated with mAb–beads
for determining the highest dilution that permitted complete agglutination (1 unit). In this
51
study, 8 units of antigen contained in 5–µl volume (the viral titer: approximately 106.6
TCID50) were used in a single test of b–LAT.
The b–LAT protocol is as follows:
1. A mixture containing 5 µl of CAV antigens and an equal volume of chicken serum
was incubated at room temperature for 15 min.
2. Subsequently, 5 µl of mAb–beads was mixed on a plastic surface with an equal
volume of the mixture of CAV antigens and chicken serum.
3. The resulting mixture was then spread as a circle with a diameter of approximately 1
cm, followed by gentle agitation for 5 min.
4. The results were scored as antibody–positive (no agglutination of mAb–beads) or
antibody–negative (agglutination observed).
Serum samples
A total of 94 serum samples collected from 4 different layer breeder flocks without
CAV problems in different areas of Japan, were used for the comparative evaluation of b–
LAT, VNT, and IFAT (Tables 2.1 and 2.3a). The characteristics of these flocks and the
results of antibody detection in these serum samples using VNT and IFAT have been
previously reported (Imai et al., 1993). In brief, sera were collected at a single time point
from flocks 2, 3, and 4, and periodically collected from the same individual chickens of
flock 1 at 19, 52, and 63 weeks of age.
CAV–induced diseases were observed among the progeny of breeder chickens in
broiler breeder (Farm 1) and layer breeder (Farm 2) farms, which are located in different
areas of Japan, in 2008 and 2013, respectively. CAV infection in the diseased chicks was
confirmed by clinical signs and gross lesions, followed by viral isolation and gene
52
detection using PCR from the livers of the diseased chicks, by methods described
previously (Imai et al., 1998; Yuasa et al., 1983a).
CAV vaccination was not performed in Farm 1, and sera were collected from the
breeder chickens of 2 flocks (flocks A, B) before and after the occurrence of CAV–induced
diseases. The CAV–induced diseases were recorded among the progeny of these breeder
chickens at the age of approximately 240 (flock A) and 218 (flock B) days. Sampling time
and the number of serum samples collected are shown in Tables 2.1 and 2.4.
Flock C of Farm 2 included 3 groups of breeder chickens of different ages (196, 448,
and 476 days) that had been vaccinated at the age of 70 days, despite which CAV–induced
diseases were recorded among their progeny. Sera from the breeder chickens were not
collected before or during the occurrence of the disease; therefore, identification of the age
group responsible for the vertical transmission of CAV is not possible. Sera were collected
from the breeder chickens, with the exception of the oldest age group, after CAV–induced
diseases were no longer observed among the progeny. Sampling time and the number of
serum samples collected are shown in Tables 2.1 and 2.4.
Sera from SPF chickens were kindly provided by NIAH, Japan, and the Advanced
Technology Development Center of Kyoritsu Seiyaku (Tsukuba, Japan). Antisera to CAV,
AIV subtype H9N2, NDV, and IBDV produced in SPF chickens were also provided by
NIAH. Sera collected from 10 weeks old breeder chickens inoculated with MDV vaccine
were also used in this study instead of antisera to MDV, since antisera were not available.
These sera were examined for the presence of antibodies to MDV using IFAT with
uninfected MSB1 cells, which are known to express MDV antigens (Schat and van Santen,
2008; Yuasa et al., 1985), fixed with acetone described above. The sera were positive for
antibodies to MDV (data not shown). The number of sera used is given in Table 2.2.
53
Data analysis
Data were analyzed using the chi–square test. Kappa value was determined using
Graphpad (http://graphpad.com/quickcalcs/kappa2/).
2.3. Results
Specificity of b–LAT
Prior to evaluating the specificity of b–LAT, nonspecific agglutination of mAb–beads
by the serum samples in the absence of CAV antigens was examined. Of the 152 undiluted
serum samples tested, 6 sera (3.9%) from SPF chicken showed nonspecific agglutination of
mAb–beads (Table 2.2); however, nonspecific reactions were completely eliminated upon
2–fold dilutions of the serum samples in PBS. Therefore, 2–fold dilutions of the sera were
employed in subsequent experiments.
All the serum samples from SPF chickens and the sera containing antibodies to AIV,
NDV, IBDV, and MDV showed negative results in b–LAT, while 5 chicken antisera to CAV
showed positive results (Table 2.2). In addition, b–LAT with the negative antigens prepared
from the uninfected cells showed negative results for chicken antisera to CAV (data not
shown).
Comparison of b–LAT with VNT and IFAT
To evaluate the usefulness of b–LAT in the detection of antibodies to CAV, a
comparison was made between b–LAT, VNT, and IFAT using sera from 94 layer breeder
chicken (Table 2.3a).
The total incidence of antibody to CAV, as determined using the 3 tests, was found to
be 78.7% (VNT), 72.3% (b–LAT), and 55.3% (IFAT). The incidence of antibody to CAV
54
was not significantly different between b–LAT and VNT, but showed statistically
significant differences between b–LAT and IFAT (P < 0.05).
Antibodies to CAV were detected in serum samples from flock 1 by the 3 tests in all
the chickens at 52 weeks of age, while the incidence of CAV antibody using IFAT (31.5%)
was significantly lower as compared to that using VNT (100%) and b–LAT (89.4%) in 63–
week–old chickens (P < 0.05). Similarly, in 48–week–old chickens of flock 4, the incidence
of antibody to CAV using IFAT (50%) was found to be significantly lower as compared to
that using VNT (100%) and b–LAT (77.7%) tests (P < 0.05).
The results of VNT and b–LAT tests showed 93.6% agreement (Kappa value = 0.82;
Table 2.3b). The sensitivity of b–LAT in comparison with VNT was 91.8% (95%
confidence interval [CI]: 83.4%–96.2%). In contrast, the results of IFAT and b–LAT
showed 78.7% agreement (Kappa value = 0.55; Table 2.3c). The sensitivity of IFAT in
comparison with b–LAT was 76.4% (95% CI: 65.1%–84.9%).
In contrast, the results of VNT and IFAT showed 76.5% agreement (Kappa value =
0.50; data not shown). The sensitivity of IFAT in comparison with VNT was 70.2% (95%
CI: 59.0%–79.4%; data not shown).
Use of b–LAT for the serological examination of breeder chicken flocks with CAV–
induced diseases among their progeny
As shown in Tables 2.1 and 2.4, in Farm 1 (flocks A and B), CAV vaccination was not
performed, and the CAV–induced disease was recorded among the progeny of these
breeder chickens at the age of approximately 240 (flock A) and 218 (flock B) days. Each
10 serum samples collected from breeder flocks A (103 days old) and B (116 and 180 days
old) prior to the incidence of CAV–induced disease among their progeny were all found to
55
be negative for CAV antibodies using b–LAT (Table 2.4). However, seroconversion to
CAV−positive was detected in each 10 serum samples collected from flocks A and B when
the chickens were examined at 270 days old (flock A) and 259 days old (flock B) after
CAV vertical transmission to their progeny ceased. The positive results obtained by b–LAT
were supported by IFAT analysis that indicated the high antibody incidence (higher than
70%) as shown in Table 2.4.
The serum samples of chickens inoculated with CAV vaccines at 70 days old in flock
C of Farm 2 were collected at 240 and 481 days old after the vertical transmission of CAV
to their progeny ceased (Tables 2.1 and 2.4). The results showed that the serum samples
were positive for antibodies to CAV by both b–LAT and IFAT.
2.4. Discussion
VNT is known to be the most specific, sensitive, and reliable serological test (Otaki et
al., 1991; Yuasa et al., 1983b) for the detection of antibodies to CAV. However, the test is
laborious and time–consuming, requiring as many as 7–9 passages of cells for completion
(Schat and van Santen, 2008). In contrast, IFAT is not as sensitive as VNT for the detection
of antibodies in older chickens (Imai et al., 1993) and showed nonspecific staining,
particularly with the use of a lower dilution of the serum (Otaki et al., 1991). Therefore,
well–trained observers are required for differentiating specific results from nonspecific
ones. In addition, both VNT and IFAT are unsuitable for testing a large number of serum
samples. ELISA has a distinct advantage in this aspect (Lamichhane et al., 1992; Tannock
et al., 2003; Todd et al., 1990b, 1999); however, ELISA is also laborious and time–
consuming both for setting–up and for completion. Commercial ELISA kits are available
for the detection of antibodies to CAV, albeit not in Japan; however, instances of false–
56
positive or false–negative results have been reported (Michalski et al., 1996; Tannock et
al., 2003). All these tests require specialized equipment or facilities.
In the present study, b–LAT was developed for the detection of antibodies to CAV in
order to overcome the drawbacks of the currently available serological tests. This test is
based on the principle that serum (antibody) from CAV–infected chicken blocks the
binding of CAV antigens to mAb–beads. The b–LAT test does not require specialized
equipment, and appears to be advantageous in terms of simplicity and speed as compared
to IFAT, VNT, or ELISA. The results of b–LAT are obtained within minutes. Therefore, b–
LAT is readily utilizable under field conditions.
Nonspecific reactions, often observed in serological tests performed for detecting
antibodies in sera, are likely to lead to erroneous diagnoses. In this study, a very low
incidence of nonspecific agglutination of mAb–beads in the absence of CAV antigens was
observed with the use of undiluted sera from SPF chicken. However, nonspecific
agglutination disappeared when 2–fold dilution of the chicken sera was used (Table 2.2). In
addition, nonspecific reaction and cross–reactivity were not observed upon analysis of sera
from SPF chicken and sera including antibodies to AIV, NDV, IBDV, or MDV using b–
LAT, with positive results obtained only with antisera to CAV. These results indicate the
high specificity of b–LAT for the detection of antibodies to CAV in chicken serum.
VNT and b–LAT showed significantly higher sensitivity for the detection of antibodies
to CAV as compared to that by IFAT, although this difference in sensitivity was observed
only with older chickens (Table 2.3a). The results of b–LAT and VNT were in good
agreement (93.6%) with a Kappa value of 0.82 (Table 2.3b), which could be weighted into
the category of almost perfect agreement (Kappa = 0.81–0.99; Viera and Garrett (2005)).
Because a neutralizing mAb was employed in b–LAT, the antibody detected in serum
57
samples by the test likely corresponds to the neutralizing antibody found in the sera of
CAV–infected chickens. Moderate agreement was observed between the results of b–LAT
and IFAT (Kappa value = 0.55; Table 2.3c).
The vertical transmission of CAV from breeder flocks to their progeny has been
known to play a major role in CAV infections in young chicks. Antibody–negative breeders
could be infected with CAV by horizontal transmission or the semen of infected cocks
during the laying period (Chettle et al., 1989; Hoop, 1993; Yuasa et al., 1987). Vertical
transmission of CAV was observed 8–14 days following the infection of hens under
experimental conditions (Hoop, 1992; Yuasa and Yoshida, 1983). In the present study, the
applicability of b–LAT in the diagnosis of field CAV cases was evaluated. CAV antibodies
were not detectable in sera collected from breeder chicken of flocks A and B in Farm 1
prior to the occurrence of CAV–induced diseases (Table 2.4); this observation also
indicates good health management programs in the farm. After the occurrence of CAV–
induced disease, the results in b–LAT clearly showed the seroconversion of tested breeder
chickens to CAV–positive, which was also supported by IFAT results. This finding
demonstrates the suitability of b–LAT for serological diagnosis in the field.
CAV vaccination of breeder flocks has been successfully employed for the prevention
of vertical transmission of the virus to progeny chicks (Schat and van Santen, 2008).
Although breeder chickens of Farm 2 were vaccinated at the age of 70 days, severe CAV–
induced diseases in their progeny resulted from the vertical transmission of the virus from
these breeders. This observation raises the question of why progeny chicks from the
vaccinated breeders remained susceptible to CAV infection. Two different scenarios, such
as antigenic mismatching of the vaccine strain to CAV isolate or failure of vaccination
procedure, could explain this situation. First, the antigenicity of the CAV strain that
58
infected the breeder chickens could have been different from that of the vaccine strain;
however, the reactivity of the CAV strain (HK1/13) isolated from the diseased chicks was
not different from the polyclonal antibody raised against the A2/76 strain (data not shown),
and amino acid properties of the strain (GenBank accession no. KJ126838) were
comparable with those of the other reported strains (Trinh et al., 2015). It has been reported
that antigenic differences were not observed among CAV isolates using chicken polyclonal
antibodies (McNulty et al., 1990a; Yuasa and Imai, 1986). Second, the vaccine was not
adequately inoculated using the route that the vaccine company recommends (personal
communication). Therefore, the incidence of CAV–associated diseases among the progeny
of the vaccinated breeders was most likely due to the failure of vaccination procedures.
In conclusion, it is emphasized that serological monitoring of breeder flocks for CAV
infection is important prior to the laying period in order to protect chicks from vertical
transmission of CAV and for ensuring the CAV–free status of SPF chicken flocks. The
results of b–LAT developed in the present study were in almost complete agreement
(93.6%, Kappa value = 0.82) with those of VNT, known to be the most specific and
sensitive test for the detection of antibodies to CAV, and moreover, could be obtained
within 5 min. Thus, the simple, rapid, highly specific, and sensitive b–LAT technique is
expected to have a potentially high application in CAV serology.
2.5. Summary
A b–LAT developed in this study was evaluated for the detection of antibodies against
CAV in chickens. Polystyrene latex beads were coupled with a neutralizing mAb to CAV
(mAb–beads), and when mixed with antigens prepared from the lysate of MSB1 cells
infected with CAV resulted in agglutination. A short pre–incubation of CAV antigens with
59
CAV–specific antiserum inhibited the agglutination of mAb–beads. The test results were
obtained within 5 min. The specificity of b–LAT was evaluated using sera from SPF
chickens and sera containing antibodies to AIV, NDV, IBDVs, and MDV; nonspecific
agglutination and cross–reactivity with antibodies to unrelated viruses were not observed.
The examination of 94 serum samples collected from commercial breeder chickens of
various ages (17–63 weeks) revealed good agreement (93.6%, Kappa value = 0.82)
between b–LAT and a VNT, known to be most sensitive and specific in the detection of
antibodies to CAV. These results indicate that b–LAT, a simple and rapid test, is a useful
and reliable tool in CAV serology.
60
Table 2.1. Field chicken serum samples used in this study
Serum samples Remarks
Source Age at sampling
No. of samples
Sera from 4 breeder farms without CAV problems in Japan
Flock 1 19, 52 and 63
weeks1) 10, 19,
19
No outbreak of CAV−induced diseases in the progeny of breeders
Flock 2 25 weeks 18
Flock 3 17 weeks 10
Flock 4 48 weeks 18
Sera from 2 breeder farms with CAV problems in Japan
Farm 1
Flock A 103 and 270
days 10 each
The outbreak in the progeny of breeders at 240 days of age
Flock B 116, 180 and
259 days 10 each
The outbreak in the progeny of breeders at 218 days of age
Farm 2
Flock C2)
240 and
481 days 10 each
Vaccination in breeders at 70 days of age Sampling after the outbreak in the progeny3)
1) Sera were periodically collected from the same individual chickens of Flock 1.
Detailed information about the flocks was described previously (Imai et al., 1993). 2) Flock C contained 3 groups of chickens with different ages (196, 448, and 476 days), and
sampling was conducted in 2 age groups except the oldest age group after CAV-induced
disease ceased. 3) It was not identified which age group was responsible for vertical transmission to the progeny.
61
Table 2.2. Evaluation of the specificity of b–LAT
Origin of serum No. of serum
samples
Non–specific agglutination 1) No. of antibody–
positive serum samples Dilution of serum
1:1 1:22)
SPF chicken serum 107 6 0 0
Chicken antiserum to AIV 10 0 0 0
Chicken antiserum to NDV 15 0 0 0
Chicken antiserum to IBDV 5 0 0 0
Chicken antiserum to CAV 5 0 0 5
Positive chicken serum to MDV3)
10 0 0 0
1) PBS was used instead of CAV antigens; 2) Dilution in PBS; 3) The serum samples were
collected from breeder chickens vaccinated with MDV vaccine.
62
Table 2.3a. Comparison of the incidence of CAV antibody in sera from field chicken using VNT, IFAT, and b–LAT
Chicken flocks
(Age of chicken)
Incidence of CAV antibody
VNT (%)1) IFAT (%)2) b–LAT (%)
Flock 1 (19 weeks old) 0/103) (0.0) 0/10 (0.0) 0/10 (0.0)
(52 weeks old) 19/19 (100.0) 19/19 (100.0) 19/19 (100.0)
(63 weeks old) 19/19 (100.0) a4) 6/19 (31.5) b 17/19 (89.4) a
Flock 2 (25 weeks old) 18/18 (100.0) 18/18 (100.0) 18/18 (100.0)
Flock 3 (17 weeks old) 0/10 (0.0) 0/10 (0.0) 0/10 (0.0)
Flock 4 (48 weeks old) 18/18 (100.0) a 9/18 (50.0) b 14/18 (77.7)a
Total 74/94 (78.7%) a5) 52/94 (55.3%) b 68/94 (72.3%) a 1) and 2) data from the report previously described (Imai et al., 1993) 3) No. of positives/no. of sera examined 4) Data within flocks followed by a different superscript letter were significantly different
(P < 0.05) 5) Data of the total incidence of CAV antibody followed by a different superscript letter were significantly different (P < 0.05)
63
Table 2.3b. Agreement in antibody detection between b–LAT and VNT
VNT b–LAT No. of serum
samples Agreement (%) Kappa value
＋1) ＋ 68
93.6 0.82 −2) − 20
＋ − 6
− ＋ 0
Total 94 1) ＋: Positive result; 2) −: Negative result
64
Table 2.3c. Agreement in antibody detection between b–LAT and IFAT
IFAT b–LAT No. of serum
samples
Agreement (%)
Kappa value
＋1) ＋ 50
78.7 0.55 −2) − 24
＋ − 2
− ＋ 18
Total 94 1) ＋: Positive result; 2) −: Negative result
65
Table 2.4. Detection of CAV antibodies in breeder chicken flocks with the outbreak of CAV−
induced diseases
Farm Chicken flocks1)
(Age at serum collection)
Vaccination
Antibody detection
Serum collection time
before or after the outbreak of
CAV−induced diseases b–LAT IFAT
1
Flock A (103 days) No 0/102) nt3) Before
(270 days) 10/10 7/10 After
Flock B (116 days) 0/10 nt Before
(180 days) 0/10 nt Before
(259 days) 10/10 10/10 After
2 Flock C
(240 and 481 days)
Yes
(70 days old) 20/20 20/20 After
1) CAV–induced diseases were observed in the progeny of the breeder chickens at the age of approximately 240 (Flock A) and 218 (Flock B) days. Flock C contained 3 groups of chickens with different ages (196, 448, and 476 days), and sampling was conducted in 2 age groups except the oldest age group after CAV-induced disease ceased as shown in Table. However, it was unidentified which age group was responsible for CAV vertical transmission. 2) No. of positives/ no. of sera examined 3) Not tested.
66
Chapter III
Isolation and preliminary characterization of chicken anemia virus
circulating in Vietnam
3.1. Introduction
It has been believed that there is no difference in antigenicity among CAV isolates,
suggesting that a single serotype was present among them (McNulty et al., 1990a; Yuasa
and Imai, 1986). However, the USA isolate CAIV–7 showed the antigenicity distinct from
a CAV representative Del–ros strain despite of its CAV–like pathogenic and
physicochemical characteristics (Spackman et al., 2002a and 2002b). Therefore, new
serotypes or subtypes of CAV that are present in the field might not be excluded.
In addition, Zhang et al. (2012) reported that a virus isolated from the human fecal
sample in China is likely to be originated from infected chickens because the sequence
identity seen between this isolate and CAV isolated from chicken meat ranged from 97.0%
to 99.7% in the genes coding 3 viral proteins (VP1, VP2 and VP3) of CAV. This suggests
that CAV might be transmitted to humans through consumption of infected chicken meat or
chicken products; however, human health under threat of CAV infection remains unclear.
Therefore, continuously monitoring of CAV infection is not only important for protection
of infection in chickens, but also useful to obtain information related to public health.
Vietnam is a developing country based on agriculture, in which livestock production
contributes about 26.3% of agricultural GDP (Vietnam general statistic office, 2013). With
increasing the rearing of animals, 315 million poultry, 26.2 million pigs and 7.7 million
cattle at the present time, the livestock products have been serving largely consumer
67
demand. However, the growth of livestock production has also caused the prevalence of
many infectious diseases including zoonotic and food–borne ones throughout the country.
In the recent years, several serious animal infectious diseases such as highly
pathogenic avian influenza or foot–and–mouth disease are considered as the major losses
for both livestock industrial sectors and small–scale stakeholders in Vietnam. Although
CAV is an economically important pathogen worldwide in poultry industry, there is almost
no information related to this topic in Vietnam. The isolation of CAV and detection of
antibody to the virus have not been described, as far as I know. However, CAV vaccination
is being conducted in some breeder farms. Therefore, there is the need to reveal actual
situation of CAV in poultry flocks in Vietnam, if any, which would help to devise a
suitable control strategy to prevent losses in poultry industry.
In this study, I describe the presence of CAV in chicken flocks and LBMs in Hanoi
and surrounding provinces. I attempted to isolate CAV, detect viral genes, and detect
antibodies to the virus from chicken samples. To the best of my knowledge, this is the first
description of the presence of CAV circulation in Vietnam.
3.2. Materials and methods
Samples
Antisera to CAV strains, A2/76, NI/77 and G6/79, produced in SPF chickens were
kindly provided by the NIAH, Japan.
Sera collected from field chickens were kindly provided by NIVR in Hanoi, Vietnam.
Seventy−four sera were collected from 16 flocks with 4–6–week–old chickens in the Hanoi
and Hanam Provinces in 2013. Sera were also collected from 237 chickens with older than
8 weeks old from 4 LBMs in Hanoi in 2013 (Table 3.1).
68
A total of 51 samples of spleen or liver were collected from chickens with younger
than 6 weeks old (Table 3.2), and 21 of which were collected from unknown diseased
chickens during 4 months from September to December, 2013, and were sent to the
DABACO Veterinary Diagnosis Centre (DVDC) Corporation in Bacninh Province. The
thirty tissue samples were collected from healthy broiler chickens of 4 farms located in
Hanam, Hanoi, Hungyen, and Vinhphuc Provinces in December, 2013. In each flock farm,
2–4 chickens were randomly selected for sampling. All of samples were collected from
chickens that originated from non–CAV vaccinated breeder chickens.
Tissue samples were homogenated in virus transfer medium, which contains
Dulbecco's modified Eagle's medium (Nissui Pharmaceutical Co., Ltd) including penicillin
G (final concentration of 1000 U/mL), streptomycin (1 mg/mL), gentamycin (100 µg/mL),
and amphotericin B (10 µg/mL), to prepare 10% homogenates. The homogenates were
stored at −20°C until use.
b–LAT
A b–LAT was conducted as described in chapter II to detect CAV antibody in chickens.
In brief, a mixture containing 5 µl of CAV antigens and an equal volume of Vietnamese
field chicken serum was incubated at room temperature for 15 min. Subsequently, 5 µl of
mAb–beads was mixed on a plastic surface with an equal volume of the mixture of CAV
antigens and chicken serum. The resulting mixture was then spread as a circle with a
diameter of approximately 1 cm, followed by gentle agitation for 5 min. The results were
scored as antibody–positive (no agglutination of mAb–beads) or negative (agglutination of
mAb–beads).
69
Virus isolation
Virus isolation was conducted in MSB1 cells as described (Yuasa et al., 1983a).
Briefly, 0.5 ml of MSB1 cells (2x105 cells/ml) was inoculated with 0.1 ml of a 10% liver or
spleen homogenate in a 24 well−plate. The inoculated cells were subcultured every 2−3
days for 10 passages, in which 0.1 ml of the cell suspension was transferred to a new well
including 0.5 ml of GM. The cultures showing red color (no cell growth) due to CPE were
regarded as CAV–positive (Yuasa, 1983). The isolation of CAV was verified using PCR
and IFAT as described below.
Virus titration
The virus titer of the CAV isolates were determined in MSB1 cells by a microtest
method (Imai and Yuasa, 1990) as described in the subsection of Cell culture in Materials
and methods of chapter I. Briefly, 20 µl of a 10–fold serially diluted virus solution was
added to wells of a 96–well microplate containing 200 µl MSB1 cells (2 × 105 cells/ml) in
GM. The inoculated cells were passaged every 3 days. The wells without virus growth
were determined after 8 passages. The cultures showing red color (no cell growth) due to
CPE were regarded as CAV–positive (Yuasa, 1983). Virus titers were quantified as the
TCID50 by the Behrens–Kärber method (Behrens and Kärber, 1934).
IFAT
IFAT (Yuasa et al., 1985) was used to confirm CAV isolation in MSB1 cells inoculated
with the homogenates as described above. In brief, MSB1 cells (1x106 cells/ml) were
inoculated with the supernatant of the MSB1 cells showing CPE. After 36 hpi, the infected
cells were harvested by centrifugation at a low speed, and were smeared on a microscope
70
slide, dried, and fixed with cold acetone for 10 min. The slides containing CAV–infected
cells were incubated with antisera to A2/76 at the dilution of 1:40, and then with FITC–
conjugated rabbit anti–chicken IgG (Rockland, Gilbertsville, PA) at 37ºC for 30 min each.
Observation was conducted under a fluorescence microscope (Biorevo BZ–9000, Keyence)
for the measurement of fluorescent signal.
VNT
VNT was performed according to the alpha–neutralization procedure (Imai and Yuasa,
1990). Briefly, 10–fold stepwise dilutions of CAV were mixed with chicken antiserum to
CAV (1:100) or GM (virus control), and then the mixtures were incubated overnight at
4°C. Afterward, 20 µl of each mixture was inoculated to each of 4 wells with 200 µl of
MSB1 cells (2 × 105 cells/ml). The inoculated cells were passaged every 3 days. The virus
titer of the mixture was determined as described above, and the neutralizing index was
calculated based on the differences of virus titers (log10 TCID50) between the mixtures with
chicken antiserum and the virus control.
Antisera to CAV isolated in Japan, A2/76, NI/77 and G6/79, were kindly provided by
NIAH.
DNA extraction, PCR and real–time PCR
DNA was extracted from 10% liver or spleen homogenates, or CAV–infected MSB1
cell culture fluids using a QIAamp DNA Mini kit (QIAGEN) in accordance with the
manufacture’s instructions. Extracted DNA was stored at ̶ 20℃ until use.
The extracted DNAs were examined for CAV DNA using PCR with partial VP1 gene
specific primers, CAV–954F: 5’–TCGGAAGAGACAGCGGTATCG–3’ and CAV–1246R:
71
5’–AGACCCGTCCGCAATCAACTC–3’(a product size of 292–bp). PCR amplification
was carried out using a TaKaRa Ex Taq kit (Takara Bio Inc.) using the following cycling
profile: initial denaturation of 94°C for 5 min, followed by 35 cycles of denaturation,
annealing and extension at 94°C for 30 sec, 50°C for 30 sec and 72°C for 1 min,
respectively, and the final extension was carried out at 72°C for 10 min. The PCR products
were then analyzed by 2% agarose gel electrophoresis and imaged under the UV.
Real−time PCR for detection of CAV VP2 gene using a commercial kit (The
PrimerDesign™ Kit for Chicken anemia virus, Genesig) was used to confirm the presence
of CAV DNA in samples.
Amplification of coding regions of CAV genes
The positive viral DNAs were amplified to obtain the complete nucleotide sequence of
CAV by two pairs of primers, CQ1F/R and CQ2F/R (Zhang et al., 2013). PCR
amplification was carried out in a 20 µl volume using a TaKaRa Ex Taq kit using the
following cycling profile: initial denaturation of 94°C for 5 min, followed by 35 cycles of
denaturation, annealing and extension at 94°C for 30 sec, 58°C for 30 sec and 72°C for 2
min 30 sec, respectively, and the final extension was carried out at 72°C for 10 min. PCR
products including 1,778 and 831 bp fragments were purified using a GENECLEAN® II
Kit (MP Biomedicals). The purified DNAs were then used as template for nucleotide
sequencing.
Nucleotide sequencing and phylogenetic analysis
Nucleotide sequences of Vietnamese CAV-positive DNAs were determined by using a
BigDye Terminator v3.1 cycle sequencing kit according to the manufacture’s instruction
72
(Life Technologies). Nucleotide sequencing was performed using Applied Biosystems
3500 Genetic Analyzer (Life Technologies).
Nucleotide sequences obtained in this study were analyzed using GENETYX ver. 10
software (GENETYX Corp., Tokyo, Japan) and compared with other sequences available
in GenBank using the BLAST program. The nucleotides and translated aa sequences were
aligned by Clustal W (Thomson et al, 1994). Phylogenetic trees were constructed using the
Maximum likelihood method and bootstrap analysis (500 replicates) using MEGA6
(Tamura et al., 2013).
3.3. Results
Serological surveillance of CAV infection in chickens
To examine the prevalence of CAV in Vietnam, 311 serum samples randomly collected
from 4–6–week–old chickens from 16 flocks located in Hanoi and Hanam Provinces, and
from chickens of older than 8 weeks of age from 4 LBMs in Hanoi, were analyzed by b–
LAT (Table 3.1).
Only 2.7% (2/74) of the serum samples from 4–6–week–old chickens and 70.4%
(167/237) of the samples from chickens of older than 8 weeks age were positive for
antibody to CAV. Totally, 54.3% (169/311) of chicken serum samples were positive for
antibody to CAV. In b–LAT, 4.5% (14/311) of chicken serum samples showed unclear
agglutination, which was regarded as “suspected case (result)”.
Detection of CAV genes
In PCR, 10 DNA samples obtained from the 51 tissue samples showed the positive
result for CAV genes (19.6%). These 10 positive cases included the 5 chicken samples
73
provided by DVDC in Bacninh Province (sample nos. BN1, BN2, BN5, BN11, and BN16),
which was recorded as unknown disease cases, and the remaining 5 samples (sample nos.
HN1, VP7, VP8, VP9, and VP10) from healthy chickens in 2 different provinces, Hanoi
and Vinhphuc (Table 3.2). However, PCR for the 4 DNA samples originated from DVDC
showed unclear results because weak bands of PCR products were observed in the stained
gel. These samples were confirmed to be positive for CAV genes by real–time PCR (data
not shown).
Isolation of CAV
The 3 homogenates, sample nos. VP7, VP8, and VP9 positive for CAV genes, were
applied to virus isolation using MSB1 cells. As the result, the 2 isolates were successfully
obtained and were designated as VP8/13 and VP9/13. The CAV isolation was confirmed
by PCR and IFAT (data not shown), and then these isolates were applied to virological
characterization. I failed to isolate CAV from the sample no. VP7, although attempts to
isolate CAV from this homogenate were repeated twice.
Both isolates grew well in MSB1 cells with the titers ranging from 7.0 to 7.25
TCID50/ml. Cross−VNT was conducted to examine antigenicity of the Vietnamese isolates
in comparison with that of the reference CAV strains (A2/76, NI/77, and G6/79). The
neutralizing index of CAV antisera against VP8/13 and VP9/13 ranged from 4.5–5.0
(Table 3.3).
Genetic and phylogenetic analysis
The full–length nucleotide sequences (1,823 bp) of the coding region for VP1 (1,350
bp), VP2 (651 bp), and VP3 (366 bp) genes of 6 out of the 10 PCR positive samples were
74
obtained by direct sequencing. These sequences contained no insertions or deletions.
Sequence information was deposited in GenBank under the accession numbers from
KP780287 to KP780292 (Table 3.4).
Genetic analysis of the 6 Vietnamese CAV gene sequences showed 96.21 % to
100.0 % in homology. The lowest nucleotide identity (96.21%) was observed between the
nucleotide sequences of VP7 and VP8 or VP9, and the highest nucleotide identity (100.0%)
was between BN1 and HN1, and VP8 and VP9. Among the 6 sequences, BN1, HN1, VP8,
and VP9 share the highest nucleotide identities with the Taiwanese CAV gene sequence,
isolate 7 (Accession number KJ728818), while VP7 and VP10 were most closely related to
the Taiwanese CAV gene sequence, isolate 18 (Accession number KJ728827) (Table 3.4).
Alignment of the aa sequence of VP1, which encodes the capsid protein, of the 6
Vietnamese CAV sequences with other sequences available in GenBank database was
conducted. The 12 aa positions (22, 75, 97, 125, 139, 144, 287, 290, 370, 376, 394, 413),
which are the most variable aas between CAV VP1 sequences, were detected in these 6
Vietnamese sequences without any specific aa difference in comparison with those of the
reported sequences. The VP1 aa sequences of BN1, HN1, VP8, and VP9 possessed a V75,
M97, K139, E144 aa profile, whilst VP7 and VP10 had a distinguished aa profile of I75,
L97, Q139, Q144. All the 6 sequences contained Q at position 394 (Q394). In addition, all
of the Vietnamese CAVs have aa profiles that are different from the profile of vaccine
strains of 26P4, Cux-1, and Del–ros (Table 3.5).
Phylogenetic analysis of the full–length gene of the coding regions of Vietnamese
CAVs indicated that the Vietnamese sequences were separated into 2 distinct genotypes.
The 4 Vietnamese sequences, BN1, HN1, VP8, and VP9, fell into Genotype III, and
formed a subgroup with the isolates from Taiwan (isolates 7 and 13) and China (GD–J–
75
12). On the other hand, the 2 sequences, VP7 and VP10, were classified within Genotype II
with the other isolates reported in different geographic areas including Asia, North and
South America, and Oceania (Fig. 3.1).
3.4. Discussion
It has been reported that CAV causes economic losses in poultry industry in many
countries (McNulty, 1991). Therefore, revealing the prevalence of CAV in the field is
important for creating effective prophylactic strategy to minimize the risk of CAV infection
for poultry industry. In Vietnam, information about the prevalence of CAV has not been
reported. However, in several large scales of breeder chicken farms, vaccination is being
currently conducted in order to prevent vertical transmission. In this study, we first
demonstrated the presence of CAV in Vietnam by serological, virological and genetical
analysis.
The evidence of CAV infection was shown during a small serological survey with
chicken sera collected from LBMs and field flocks, using the b–LAT described in chapter
II. In the field chicken flock conditions, natural CAV infection usually occurs in chickens
due to horizontal transmission after maternal antibodies diminished around 2 to 4 weeks of
age, and seroconversion in most of the infected chickens needs several weeks (McNulty et
al., 1988; von Bülow, 1988). The present result showed that only a small number of serum
samples (2 of 74) collected from 4 to 6 weeks old chickens in the field were positive in b–
LAT (Table 3.1), which may indicate the early phase exposures to CAV in the flocks. In
contrast, most of serum samples collected from older than 8 weeks old chickens gathered
in LBMs possessed antibodies. It was suggested that CAV transmission among field
chickens in Vietnam appears to occur in a similar way to that in other countries.
76
Furthermore, circulation of CAV in Vietnam was also demonstrated either by the detection
of viral DNA or by virus isolation from chicken tissue samples in 3 out of the 5 examined
provinces (Bacninh, Hanoi, Vinhphuc). These results confirmed the findings obtained by
antibody detection; however, not all of the samples obtained within flock were positive for
CAV gene (data not shown). Since 4–6 weeks old chickens were exposed to CAV when
maternal antibody disappeared, probably by around 3 to 4 weeks of age; therefore, the
number of CAV DNA present in most of the samples of those chickens might be under the
limit of detection.
Investigation on the antigenicity of CAVs isolated in Japan and the U.K. by cross−
IFAT or VNT with chicken antisera to CAV showed no antigenic differences among them
(McNulty et al., 1990a; Yuasa and Imai, 1986). However, immunofluorescent staining with
mAbs demonstrated antigenic differences between some isolates which were
indistinguishable using antisera (McNulty et al., 1990b). In addition, an evidence of a
second serotype of CAV was provided in USA. The isolate CIAV–7 possessed CAV
characteristics such as a small size, high resistance to chemical agents, and inducing
similar syndrome with CAV in chickens (Spackman et al., 2002a and 2002b). However,
the diseases induced by this strain were much milder compared with the reference CAV
strain. Cross–reactivity of chicken antisera and genetic similarity between the reference
and CIAV–7 strains were not found. Thus, the presence of the second serotype of CAV has
not been generally accepted. In the present study, we characterized the first CAVs isolated
from chickens in Vietnam. No antigenic difference among the isolates in Vietnam and
Japan was recognized by VNT with antisera to the reference CAV strains (Table 3.3). This
result would imply that CAV belongs to a single serotype irrespective of geographic
origins.
77
On the other hand, the correlation between the VP1 aa substitutions and growth of
CAV in cell cultures was reported, in which CAVs with the aa Q139 and Q144 profile
poorly grew in cell cultures (Renshaw et al., 1996). However, other research groups
observed a good growth of the strains with the same aa profile tested (Connor et al., 1991;
Krapez et al., 2006). The reason of this discrepancy seen in virus growth of CAV is
unclear. In the present study, I failed to isolate CAV with the aa Q139 and Q144 profile,
while the other 2 CAVs with the aa K139 and E144 profile were successfully isolated in
cell cultures.
Yamaguchi et al. (2001) reported that Q394 of VP1 aa sequence was considered as a
major determinant of viral pathogenicity of CAV, since change of VP1 aa position 394
from Q to H reduced pathogenicity. All of the 6 CAV VP1 aa sequences identified in
Vietnam showed different aa profiles from that of the attenuated vaccine strains (26P4,
Cux-1, Del-ros) as shown in Table 3.5, but the all had Q394. Therefore, the Vietnamese
CAVs may be virulent. Animal study is needed to evaluate the virulence of Vietnamese
CAV strains in chickens.
Genetic analysis of the entire coding region of VP1, VP2, and VP3 genes of the 6
Vietnamese CAV strains revealed high homology (higher than 96%) among them. The
Vietnamese CAVs shared the highest nucleotide identity (higher than 99%) with the
Taiwanese CAVs (Table 3.4). In phylogenetic analysis of these sequences and other CAV
sequences available in GenBank, the presence of 3 genotypes was observed. While
genotype I consists of the sequences of CAVs originated from Australia, genotypes II and
III include the sequences that have a worldwide distribution. Six CAV gene sequences
obtained from Vietnamese chickens were classified into genotypes II and III, indicating the
genetic diversity of CAV circulating in Vietnam. The 4 CAV gene sequences belonging to
78
genotype III were originated from the chickens in 3 different provinces, while the other 2
sequences that were fallen into genotype II were found only in Vinhphuc Province. These
results may suggest that genotype III CAV is widespread in chicken flocks in Vietnam.
However, due to the limited sample size available in the current study, further systematic
surveillances of CAV in chicken flocks may be needed to fully understand the exact
distribution of CAV and its genotypes in Vietnam.
In conclusion, this study provided the first demonstration of the presence of CAV in
Vietnam by the detection of CAV antibodies, and CAV genes, and isolation of virus in
field samples. The circulation of virus was confirmed in 3 out of 4 provinces examined.
There was no difference in the antigenicity of Vietnamese isolates in comparison to that of
the reference CAV strains. Molecular charactization of revealed Vietnamese CAV
sequences fell into 2 different genotypes of CAV, which are the most widely distributed
throughout the world. These results emphasized that CAV is circulating and might be
affecting poultry industry in Vietnam; however, further study is needed to provide actual
data to describe how CAV affects Vietnamese chickens.
3.5. Summary
CAV is a ubiquitous and economical important pathogen causing severe anemia in
young chicks. Although CAV has been detected in many countries with poultry industry,
there has been no information about CAV in Vietnam. In this study, the first detection and
characterization of CAV in chickens in Vietnam was described. CAV antibody was detected
in 70.4% of the samples from chickens of older than 8 weeks age, and that of 2.7% in the
samples from chickens of 4–6 weeks of age. CAV genes were detected in 10 out of the 51
tissue samples from chickens in 4 different provinces in Northern Vietnam by PCR. Result
79
of VNT with antisera to the reference CAV strains showed no antigenic differences
between the 2 Vietnamese isolates that were obtained from MSB1 cells inoculated with
homogenate of tissue samples. The full coding region of 3 viral proteins, VP1, VP2 and
VP3 (1,823 bp) of the 6 CAVs was sequenced and characterized. Phylogenetic analysis
revealed that Vietnamese CAVs were classified into 2 distinct genotypes II and III showing
worldwide distribution. In addition, amino acid profile of all Vietnamese CAVs contains
Q394 that was reportedly associated with virulence.
80
Table 3.1. Detection of CAV antibodies in the field chickens using b–LAT
Location
(Province)
Serum samples from
Age (weeks)
No. of samples
No. of antibody positive samples (%)
Positive Suspected1)
Hanoi and Hanam
Chicken flock
4–6 74 2 (2.7) 0 (0.0)
Hanoi LBM 2) > 8 237 167 (70.4) 14 (5.9)
Total 311 169 (54.3) 14 (4.5) 1) Suspected: Samples showed unclear agglutination in b–LAT. 2) LBM: Live bird market
81
Table 3.2. Detection of CAV genes in tissue samples collected from the field chickens by PCR
Location (Province)
No. of flocks No. of samples Chicken status3)
No. of positive samples
Hanam 3 9 Healthy 0
Hanoi 2 6 Healthy 1
Hungyen 2 4 Healthy 0
Vinhphuc 4 11 Healthy 4
DVDC 1) in Bacninh
NI 2) 21 Unknown diseased
5
Total 51 10 (19.6%) 1) DVDC: DABACO Veterinary Diagnosis Centre; 2)
NI: Not identified; 3) Chicken are less
than 6 weeks of age.
82
Table 3.3. VNT1) with antisera to Japanese CAVs against Vietnamese isolates
Isolates Neutralizing index of antisera to CAV against Vietnamese isolates
Anti–A2/76 Anti–NI/77 Anti–G6/78
VP8/13 4.5 5.0 4.5
VP9/13 4.75 5.0 4.5
1) VNT was performed using the α–neutralization procedure described in the Materials and methods.
83
Table 3.4. Comparison of nucleotide sequence of Vietnamese sequences with that of other sequences available in GenBank
Samples ID1)
Length (bp)2) Accession
no.
Viruses with the highest identity of nucleotide
Strain %
identity Accession
no.
BN1 1,823 KP780287 Taiwan/isolate 7 99.72 KJ728818
HN1 1,823 KP780288 Taiwan/isolate 7 99.72 KJ728818
VP7 1,823 KP780289 Taiwan/isolate 18 99.34 KJ728827
VP8 1,823 KP780290 Taiwan/isolate 7 99.67 KJ728818
VP9 1,823 KP780291 Taiwan/isolate 7 99.67 KJ728818
VP10 1,823 KP780292 Taiwan/isolate 18 99.34 KJ728827 1) BN1 is from a chicken with unknown diseases and others from healthy chickens. 2) The full–length coding regions of CAV genes were compared.
84
Table 3.5. Amino acid profile of VP1 of Vietnamese CAVs
Sequences Genotypes Amino acid position in VP11)
22 75 97 125 139 144 287 290 370 376 394 413
Germany/M55918/CUX1 III H V M I K D A A S L Q A
Netherland/D10068/26P42) III . . . . . E T . . . . .
USA/AF313470/Del–Ros III . . . . . E S . G . . S
Germany/M81223/Cux1 III . . . . . . . . . . . .
China/KF224934/GD–J–12 III . . . L . E S . G I . S
China/JQ690762/Human III . . . . Q Q . . . . . .
Japan/AB031296/A2 III . . . . . E S . G . . .
Japan/E51057/Att–CAV III . . . L . E S . G I H S
Malaysia/AF390038/3–1 III . . . . . E D . . . . .
Taiwan/KJ728823/isolate 13 III . . . L . E S . G I . S
Taiwan/KJ728818/isolate 7 III . . . L . E S . G I . S
USA/L14767/CIA–1 III N I L . Q Q . . . . . .
Vietnam/KP780287/BN13) III . . . L . E S . G I . S
Vietnam/KP780288/HN1 III . . . L . E S . G I . S
Vietnam/KP780290/VP8 III . . . L . E S . G I . S
Vietnam/KP780291/VP9 III . . . L . E S . G I . S
Australia/EF683159/3711 I . . . . . E S . G . . S
Australia/AF227982/CAU269/7 I . . . . . E T . R . . S
Vietnam/KP780289/VP7 II Q I L . Q Q . . . . . .
Vietnam/KP780292/VP10 II Q I L . Q Q . . . . . .
Australia/U65414/704 II . I L . Q Q T P . . . .
Chile/JQ308214/CL52 II . I L . Q Q T P T . . .
Japan/AB119448/G6 II . I L . Q Q T P . . . .
Malaysia/AF285882/SMSC–1 II . I L . Q Q T P . . . .
Taiwan/KJ728827/isolate 18 II Q I L . Q Q . . . . . .
USA/AF311900/98D06073 II Q I L . Q Q . . . . . .
1) Position was based on the amino acid sequence of the VP1 of Cux-1 strain (accession no. M55918, Noteborn et al. (1991)); 2) The vaccine strains were in bold; 3) The Vietnamese CAV sequences obtained in this study were underlined.
85
Fig. 3.1. Phylogenetic tree of the full–length gene of coding region (1,823 bp) of the Vietnamese
CAVs’ VP1, VP2, and VP3 protein compared to the sequences available in GenBank. Sequences
from GenBank are indicated with the country name followed by accession number. The
phylogenetic tree was constructed using the maximum likelihood method (500 bootstrap replicates)
and MEGA6 software. Number at each branch point indicate bootstrap values ≥50% in the
bootstrap interior branch test. The Vietnamese CAVs are marked with closed squares (■). Three
major genotypes were identified and designated as genotype I, II, and III.
86
General discussion
CAV was first reported in chickens in Japan in 1979 (Yuasa et al., 1979), and then the
presence of CAV has been confirmed in chickens in most countries with poultry industry.
CAV can be transmitted vertically from the parents to their progeny or horizontally by
contact exposure with infected chickens or fomites contaminated with the virus. Although
vertical transmission with CAV is known to cause severe clinical diseases in young chicks,
CAV may also cause subclinical infection related to immunosuppression in older chickens.
Both clinical and subclinical infections may cause direct or indirect economic losses in
poultry industry (McNulty, 1991). Even SPF chicken flocks reared under very strict
hygienic conditions have reportedly become infected with CAV (Cardona et al., 2000a;
Goryo et al., 1985; McNulty et al., 1989; Yuasa et al., 1985). Consequently, the eggs from
the flocks infected with CAV are no longer SPF. Australia, Europe, and USA do not accept
these eggs for production of vaccines for human use for such as measles and mumps
vaccines (Schat and van Santen, 2008). However, since CAV has a widespread distribution
and high resistance to inactivation, reduced virus exposure requires a well-established
biosecurity system and/or effective vaccination. Therefore, there is a need to understand
biological properties of CAV and interaction between virus and hosts, which will enable us
to have appropriate strategies such as a suitable biosecurity system, effective vaccines and
vaccination procedures, or sensitive diagnostic kits for control and prevention of this
disease.
It is known that there are 3 viral proteins of CAV, VP1, VP2 and VP3. Of which, VP1
is the only capsid protein that is crucial for producing neutralizing antibodies against CAV
in chickens (Schat and van Santen, 2008; Todd et al., 1990a); therefore, VP1 is the target to
87
study pathogenicity and antigenicity of CAV, and to use as immunogen of subunit vaccine,
or to develop diagnostic kits. However, there are many questions remaining unclear related
to the appearance time of VP1 in infected MSB1 cells, or neutralizing epitopes on VP1. In
the present study, I studied about neutralizing epitopes on VP1 by using neutralizing mAb
strategy. As the results, 3 neutralizing mAbs were found to be specific to VP1 protein by
using IP and recombinant VP1 protein expressed in mammalian cells. The mAbs could
detect the VP1 synthesized in the infected MSB1 cells as early as 12 day post viral
inoculation. This is the novel finding in contrast to the previous report in which VP1 was
first detectable at 30 hpi (Douglas et al., 1995). This is also the first demonstration of
neutralizing epitopes located on VP1. Sequence analysis of the 3 escape mutants
established from the neutralizing mAbs revealed mutations in different parts of VP1, aas
T89+A90, E144, and I261. However, the similarity in the reactivity of MoCAV/F2 and
MoCAV/F8 to the infected MSB1 cells in blocking IFAT, and to their corresponding escape
mutants (EsCAV/F2 and F8) in IFAT and VNT may suggest the existence of a single
epitope recognized by the 2 mAbs. Therefore, my study indicated that VP1 has at least 2
different neutralizing epitopes that have not been reported previously, to the best of my
knowledge. Mapping the of neutralizing epitopes might be very important in the light of
attempts for future to obtain insight for CAV biology, or to prepare polypeptides, or
recombinant proteins for research or development of new diagnostic kits. It is also
important to provide neutralizing epitope sites for development of subunit vaccine, or DNA
vaccine, since current commercial live vaccines for CAV can not be applied to breeder
chickens during a laying pariod.
As with epitope mapping, the application of mAbs was also considerably useful in the
virological research field with antigenic variation and epidemiology (McCullough and
88
Spier, 2009). In the current study, the CAV strains evaluated could be differentiated into 2
distinct antigenic groups by one mAb (MoCAV/F11), which could be associated with
specific aa profiles at position I75, L97, Q139, and Q144 of VP1. However, in this study, I
only examined the CAV strains originated from Japanese chickens, and 1 vaccine strain
(26P4, Netherland). Thus, additional analysis with other strains from different geographic
areas is necessary to confirm the presence of different antigenic groups of CAV which is
known to be a single serotype. Further study is needed to understand the antigenic
properties of these antigenic groups, which could be important for vaccination against
CAV.
Scott et al. (1999) stated that 5 out of the 6 clones possessed aa change at position 89
(T89A) with other 5 aa changes in VP1 showed the reduction of pathogenicity in chickens
compared to the clones without aa change at position 89. Further investigation of aa T89
with chimeric approach confirmed that aa change T89A combining with other aa changes
in VP1 caused virus attenuation (Todd et al., 2002). In addition, aa Q394 in VP1 was
considered as a major determinant of pathogenicity (Yamaguchi et al., 2001). Chowdhury
et al. (2003) reported that their isolates which received 60 and 120 passages in MSB 1 cells
showed a significant reduction of pathogenicity in chickens compared to the low passage
ones; however, these highly passaged isolates showed several aa changes different from
T89 and Q394. Therefore, alternative aa changes occurring during CAV passages in cell
cultures might result in virus attenuation. In this study, since EsCAV/F2 possessed the
deletion of 2 aa T89A90 in VP1, it should be evaluated whether EsCAV/F2 shows lower or
no pathogenicity to chickens in pathogenicity test using chickens.
In the field, serological monitoring of breeder flocks for CAV infection is quite
important prior to the laying period in order to protect chicks from vertical transmission of
89
CAV and to ensure the CAV–free status of SPF chicken flocks. For detection of CAV
antibody in chicken sera, there are several established serological tests (VNT, IFAT, or
ELISA); however, they still have some limits in use or are not available in several
countries, for instance, commercial ELISA kits are not available in Japan. In addition, it is
well–known that applications of mAbs have increased the accuracy and rapidity of
diagnostic tests. The increase of efficiency of diagnostic could result in more effective
treatment results, or appropriate control and prevention strategies. For those reasons, I
developed the blocking latex agglutination test, b–LAT, using the neutralizing mAb
(MoCAV/F11), which might be good for field application. The results of b–LAT obtained
in the present study were in almost complete agreement with those of VNT, known to be
the most specific and sensitive test for the detection of antibodies to CAV (Otaki et al.,
1991; Yuasa et al., 1983b), and moreover, the result could be obtained within 15 minutes.
Since there is only one serotype present among CAV isolates (McNulty et al., 1990a; Yuasa
and Imai, 1986), the b-LAT can be used to detect CAV antibody in chickens with either
CAV infection or vaccination. However, the b-LAT as well as other current serological tests
cannot differentiate CAV natural infection from vaccination. Thus, the simple, rapid, highly
specific, and sensitive b–LAT technique is expected to have a potentially high application
in CAV serology.
In Vietnam, there has been no report on the circulation of CAV in chicken flocks,
although vaccination is being applied in several breeder farms with large–scale of poultry.
The b–LAT was used to detect CAV antibodies in field sera from Vietnamese chickens. The
result of antibodies detection showed the high prevalence of CAV infection in Vietnamese
chickens. These results may reflect the potential field application of the test. Indeed, I
confirmed the results of b–LAT (the presence of CAV in Vietnamese chickens) using virus
90
isolation (2 isolates) and viral gene detection (10 positive samples). Molecular
characterization of Vietnamese CAV gene sequences revealed 2 different genotypes of
CAV that are distributed worldwide. These results emphasized that CAV is circulating and
probably affecting poultry industry in Vietnam; however, further study is needed to provide
detailed data for describing how CAV affects Vietnamese chickens. In addition, the results
of virological characterization showed no antigenic difference among 2 Vietnamese
isolates and the reference CAV strains. Thus, I could emphasize that CAV belongs to a
single serotype.
Although the high prevalence of CAV has been confirmed in all the types of chickens
such as broiler, layer, or breeder chickens examined, it is surprising that CAV antibodies
were detected in SPF chicken flocks which were reared in a very strict condition (Cardona
et al., 2000a). CAV infections were often found in SPF flocks during the first laying cycles
(Schat and van Santen, 2008). Cardona et al. (2000b) detected CAV DNA in the gonadal
tissues of SPF chickens with or without neutralizing antibodies. CAV DNA was also
detected in the gonads in neutralizing antibody–positive hens, and the viral DNA was
confirmed to be transmitted from these hens to their embryos (Brentano et al., 2005).
Taken together, the persistence of CAV in chickens was suspected. However, since the
virus was not successfully recovered from these chickens, there is a need to do additional
studies to prove the persistence of CAV in chickens. One possible strategy which can be
used to investigate the persistent infection of CAV is the use of an in vitro model. In the
preliminary experiments, CAV–infected MSB1 cells were serially passaged in the medium
supplemented with chicken polyclonal antibody to CAV. Then the presence of virus, and/or
viral DNA was detectable in the cultured cells up to 15th passage.
91
General conclusion
In this study, I established 4 CAV mAbs, 3 of which (MoCAV/F2, MoCAV/F8, and
MoCAV/F11) had neutralizing activity and recognized the VP1 as shown in IP and VP1
recombinant protein expression assays. Analysis of the 3 escape mutants generated from
the neutralizing mAbs revealed 3 different patterns of aa change (T89+A90, E144, and
I261) on VP1. However, the similarity in the reactivity of MoCAV/F2 and MoCAV/F8 to
the infected MSB1 cells in blocking IFAT, or to their corresponding escape mutants
(EsCAV/F2 and F8) in IFAT and VNT may suggest the existence of a single epitope
recognized by these mAbs. Therefore, this study showed that VP1 has at least 2 different
neutralizing epitopes that have not been reported previously, to the best of my knowledge.
In addition, the CAV strains evaluated could be differentiated into 2 distinct antigenic mAb
groups by MoCAV/F11, which could be associated with specific aa profiles of VP1 (I75,
L97, Q139, Q144). Mutations in VP1 have been reported to affect pathogenicity or viral
replication in in chickens. However, there have not been consistent molecular biological
evidences to explain the events, and there are still many aspects that remain unresolved
with respect to CAV biology (Chapter I).
In this study, the neutralizing mAb (MoCAV/F11) was also used to develop a new
diagnostic method, b–LAT, for detection of CAV antibodies in chicken sera. The results of
b–LAT developed in the present study were in almost complete agreement (93.6%, Kappa
value = 0.82) with those of VNT, known to be the most specific and sensitive test for the
detection of antibodies to CAV, and moreover, could be obtained within 15 min. Thus, the
simple, rapid, and highly specific and sensitive b–LAT is expected to have a high
application potential in CAV serology (Chapter II).
92
This study first demonstrated the presence of CAV in Vietnamese chickens using virus
isolation, PCR, and b–LAT. Two isolates and 10 viral gene–positive samples were obtained
from the 51 chicken tissue samples. Genetical and antigenic analysis indicated that
characteristics of CAV prevalent in Vietnam were not substantially different from those of
the known CAV strains. This study provides a basic information for future epidemiological
research on CAV in the country. However, further studies are needed to fully understand
the exact distribution of CAV in Vietnam (Chapter III).
93
ACKNOWLEDGEMENTS
I would like to express my heartfelt gratitude to my supervisor Professor Dr. Kunitoshi
Imai, you have been a tremendous mentor for me. I would like to thank you for
encouraging my research and for allowing me to grow up in science. Your advices through
my study have been priceless.
I would like to express my special appreciation and thanks to my co-supervisor
Professor Dr. Haruko Ogawa, you have given me special support during my research.
Working with you, I can learn a lot.
I would like to express my special appreciation and thanks to my co-supervisor
Professor Dr. Ikuo Igarashi, you have given me special support during my research.
I would also like to thank my committee members, Professor Dr. Xuenan Xuan,
Associate Professor Dr. Yoshifumi Nishikawa for serving as my committee members even
at hardship. I also want to thank you for spending time to review my thesis, and for your
valuable comments and suggestions.
I would like to thank members and students of the Diagnostic Center of Animal Health
and Food Safety. I am grateful for the chance to be a part of the lab. Thank you for
welcoming me as a friend and helping me during my PhD course.
I would like to thank Graduate School of Animal Husbandry, Obihiro University of
Agriculture and Veterinary Medicine to provide me the chance to study here. I would also
like to thank all of my friends in the University who helped me over the years.
I would like to thank my office, Vietnam National Institute of Veterinary Research, to
allow me to join this PhD course.
94
A special thanks to my family. It is difficult to express how grateful I am to my
mother, father, mother-in law, father-in-law, for all of the things you do for me. Your prayer
for me was what sustained me thus far. At the end I would like express appreciation to my
beloved wife and two sons who are always with me throughout the time.
95
References
Adair, B. M. (2000). Immunopathogenesis of chicken anemia virus infection. Dev Comp
Immunol 24, 247–255.
Behrens, B. and Kärber, G. (1934). Wie sind Reihenversuche für biologishe Auswertungen
am zweckmäßigsten anzuordnen (in German). Naunyn Schmiedebergs Arch Exp
Pathol Pharmakol 18, 379-388.
Biagini, P., Bedarida, S., Touinssi, M., Galicher, V. and de Micco, P. (2013). Human
gyrovirus in healthy blood donors, France. Emerg Infect Dis 19, 1014–1015.
Brentano, L., Lazzarin, S., Bassi, S. S., Klein, T. A. and Schat, K. A. (2005). Detection of
chicken anemia virus in the gonads and in the progeny of broiler breeder hens with
high neutralizing antibody titers. Vet Microbiol 105, 65–72.
Calnek, B. W., Lucio–Martinez, B., Cardona, C., Harris, R. W., Schat, K. A. and Buscaglia,
C. (2000). Comparative susceptibility of Marek's disease cell lines to chicken
infectious anemia virus. Avian Dis 44, 114–124.
Cardona, C., Lucio, B., O'Connell, P., Jagne, J. and Schat, K. A. (2000a). Humoral immune
responses to chicken infectious anemia virus in three strains of chickens in a closed
flock. Avian Dis 44, 661–667.
Cardona, C. J., Oswald, W. B. and Schat, K. A. (2000b). Distribution of chicken anaemia
virus in the reproductive tissues of specific–pathogen–free chickens. J Gen Virol 81,
2067–2075.
Chandratilleke, D., O'Connell, P. and Schat, K. A. (1991). Characterization of proteins of
chicken infectious anemia virus with monoclonal antibodies. Avian Dis 35, 854–862.
96
Chettle, N. J., Eddy, R. K., Wyeth, P. J. and Lister, S. A. (1989). An outbreak of disease due
to chicken anaemia agent in broiler chickens in England. Vet Rec 124, 211–215.
Connor, T. J., McNeilly, F., Firth, G. A. and McNulty, M. S. (1991). Biological
characterisation of Australian isolates of chicken anaemia agent. Aust Vet J 68, 199–
201.
Davidson, I., Kedem, M., Borochovitz, H., Kass, N., Ayali, G., Hamzani, E., Perelman, B.,
Smith, B. and Perk, S. (2004). Chicken infectious anemia virus infection in Israeli
commercial flocks: virus amplification, clinical signs, performance, and antibody
status. Avian Dis 48, 108–118.
dos Santos, H. F., Knak, M. B., de Castro, F. L., Slongo, J., Ritterbusch, G. A., Klein, T. A.,
Esteves, P. A., Silva, A. D., Trevisol, I. M., Claassen, E. A., Cornelissen, L. A., Lovato,
M., Franco, A. C., Roehe, P. M. and Rijsewijk, F. A. (2012). Variants of the recently
discovered avian gyrovirus 2 are detected in Southern Brazil and The Netherlands. Vet
Microbiol 155, 230–236.
Douglas, A. J., Phenix, K., Mawhinney, K. A., Todd, D., Mackie, D. P. and Curran, W. L.
(1995). Identification of a 24 kDa protein expressed by chicken anaemia virus. J Gen
Virol 76 ( Pt 7), 1557–1562.
Ducatez, M. F., Owoade, A. A., Abiola, J. O. and Muller, C. P. (2006). Molecular
epidemiology of chicken anemia virus in Nigeria. Arch Virol 151, 97–111.
Eltahir, Y. M., Qian, K., Jin, W., Wang, P. and Qin, A. (2011). Molecular epidemiology of
chicken anemia virus in commercial farms in China. Virol J 8, 145.
Gelderblom, H., Kling, S., Lurz, R., Tischer, I. and von Bulow, V. (1989). Morphological
characterization of chicken anaemia agent (CAA). Arch Virol 109, 115–120.
97
Goodwin, M. A., Smeltzer, M. A., Brown, J., Girshick, T., McMurray, B. L. and McCarter,
S. (1993). Effect of so–called chicken anemia agent maternal antibody on chick
serologic conversion to viruses in the field. Avian Dis 37, 542–545.
Goryo, M., Sugimura, H., Matsumoto, S., Umemura, T. and Itakura, C. (1985). Isolation of
an agent inducing chicken anaemia. Avian Pathol 14, 483–496.
Goryo, M., Suwa, T., Matsumoto, S., Umemura, T. and Itakura, C. (1987). Serial
propagation and purification of chicken anaemia agent in MDCC–MSB1 cell line.
Avian Pathol 16, 149–163.
Hailemariam, Z., Omar, A. R., Hair–Bejo, M. and Giap, T. C. (2008). Detection and
characterization of chicken anemia virus from commercial broiler breeder chickens.
Virol J 5, 128.
Harlow, E. and Lane, D. (1988). Antibodies: A Laboratory Manual. Cold Spring Harbor
Laboratory, 274–275.
Hirano, H. (1989). Microsequence analysis of winged bean seed proteins electroblotted
from two-dimensional gel. J Protein Chem 8, 115-130.
Hoop, R. K. (1992). Persistence and vertical transmission of chicken anaemia agent in
experimentally infected laying hens. Avian Pathol 21, 493–501.
Hoop, R. K. (1993). Transmission of chicken anaemia virus with semen. Vet Rec 133, 551–
552.
Imai, K., Maeda, M. and Yuasa, N. (1991). Immunoelectron microscopy of chicken anemia
agent. J Vet Med Sci 53, 1065–1067.
Imai, K., Mase, M., Yamaguchi, S., Yuasa, N. and Nakamura, K. (1998). Detection of
chicken anaemia virus DNA from formalin–fixed tissues by polymerase chain
reaction. Res Vet Sci 64, 205–208.
98
Imai, K., Mase, S., Tsukamoto, K., Hihara, H., Matsumura, T. and Yuasa, N. (1993). A long
term observation of antibody status to chicken anaemia virus in individual chickens of
breeder flocks. Res Vet Sci 54, 392–396.
Imai, K. and Yuasa, N. (1990). Development of a microtest method for serological and
virological examinations of chicken anemia agent. Nihon Juigaku Zasshi 52, 873–875.
Islam, M. R., Johne, R., Raue, R., Todd, D. and Muller, H. (2002). Sequence analysis of the
full–length cloned DNA of a chicken anaemia virus (CAV) strain from Bangladesh:
evidence for genetic grouping of CAV strains based on the deduced VP1 amino acid
sequences. J Vet Med B Infect Dis Vet Public Health 49, 332–337.
Jorgensen, P. H., Otte, L., Nielsen, O. L. and Bisgaard, M. (1995). Influence of subclinical
virus infections and other factors on broiler flock performance. Br Poult Sci 36, 455–
463.
Kaffashi, A., Shrestha, S. and Browning, G. F. (2008). Evaluation of chicken anaemia virus
mutants as potential vaccine strains in 1–day–old chickens. Avian Pathol 37, 109–114.
Kaverin, N. V., Rudneva, I. A., Govorkova, E. A., Timofeeva, T. A., Shilov, A. A.,
Kochergin–Nikitsky, K. S., Krylov, P. S. and Webster, R. G. (2007). Epitope mapping
of the hemagglutinin molecule of a highly pathogenic H5N1 influenza virus by using
monoclonal antibodies. J Virol 81, 12911–12917.
Kaverin, N. V., Rudneva, I. A., Ilyushina, N. A., Varich, N. L., Lipatov, A. S., Smirnov, Y.
A., Govorkova, E. A., Gitelman, A. K., Lvov, D. K. and Webster, R. G. (2002).
Structure of antigenic sites on the haemagglutinin molecule of H5 avian influenza
virus and phenotypic variation of escape mutants. J Gen Virol 83, 2497–2505.
King, A. M. Q., Lefkowitz., E., Adams., M. J. and Carstens., E. B. (2011). Virus Taxonomy:
Ninth Report of the International Committee on Taxonomy of Viruses: Elsevier.
99
Koch, G., van Roozelaar, D. J., Verschueren, C. A., van der Eb, A. J. and Noteborn, M. H.
(1995). Immunogenic and protective properties of chicken anaemia virus proteins
expressed by baculovirus. Vaccine 13, 763–770.
Krapez, U., Barlic–Maganja, D., Toplak, I., Hostnik, P. and Rojs, O. Z. (2006). Biological
and molecular characterization of chicken anemia virus isolates from Slovenia. Avian
Dis 50, 69–76.
Lamichhane, C. M., Snyder, D. B., Girschick, T., Goodwin, M. A. and Miller, S. L. (1992).
Development and comparison of serologic methods for diagnosing chicken anemia
virus infection. Avian Dis 36, 725–729.
Li, L., Kapoor, A., Slikas, B., Bamidele, O. S., Wang, C., Shaukat, S., Masroor, M. A.,
Wilson, M. L., Ndjango, J. B., Peeters, M., Gross–Camp, N. D., Muller, M. N., Hahn,
B. H., Wolfe, N. D., Triki, H., Bartkus, J., Zaidi, S. Z. and Delwart, E. (2010). Multiple
diverse circoviruses infect farm animals and are commonly found in human and
chimpanzee feces. J Virol 84, 1674–1682.
Lien, Y. Y., Huang, C. H., Sun, F. C., Sheu, S. C., Lu, T. C., Lee, M. S., Hsueh, S. C., Chen,
H. J. and Lee, M. S. (2012). Development and characterization of a potential
diagnostic monoclonal antibody against capsid protein VP1 of the chicken anemia
virus. J Vet Sci 13, 73–79.
Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951). Protein
measurement with the Folin phenol reagent. J Biol Chem 193, 265–275.
Maggi, F., Macera, L., Focosi, D., Vatteroni, M. L., Boggi, U., Antonelli, G., Eloit, M. and
Pistello, M. (2012). Human gyrovirus DNA in human blood, Italy. Emerg Infect Dis
18, 956–959.
100
Markowski–Grimsrud, C. J., Miller, M. M. and Schat, K. A. (2002). Development of
strain–specific real–time PCR and RT–PCR assays for quantitation of chicken anemia
virus. J Virol Methods 101, 135–147.
McConnell, C. D., Adair, B. M. and McNulty, M. S. (1993). Effects of chicken anemia
virus on cell–mediated immune function in chickens exposed to the virus by a natural
route. Avian Dis 37, 366–374.
McCullough., K. C. and Spier., R. E. (2009). Monoclonal antibodies in biology and
biotechnology: theoretical and practical aspects: Cambridge University Press.
McIlroy, S. G., McNulty, M. S., Bruce, D. W., Smyth, J. A., Goodall, E. A. and Alcorn, M.
J. (1992). Economic effects of clinical chicken anemia agent infection on profitable
broiler production. Avian Dis 36, 566–574.
McNulty, M. S. (1991). Chicken anaemia agent: a review. Avian Pathol 20, 187–203.
McNulty, M. S., Connor, T. J. and McNeilly, F. (1989). A survey of specific pathogen–free
chicken flocks for antibodies to chicken anaemia agent, avian nephritis virus and
group a rotavirus. Avian Pathol 18, 215–220.
McNulty, M. S., Connor, T. J., McNeilly, F., Kirkpatrick, K. S. and McFerran, J. B. (1988).
A serological survey of domestic poultry in the United Kingdom for antibody to
chicken anaemia agent. Avian Pathol 17, 315–324.
McNulty, M. S., Connor, T. J., McNeilly, F., McLoughlin, M. F. and Kirkpatrick, K. S.
(1990a). Preliminary characterisation of isolates of chicken anaemia agent from the
United Kingdom. Avian Pathol 19, 67–73.
McNulty, M. S., Mackie, D. P., Pollock, D. A., McNair, J., Todd, D., Mawhinney, K. A.,
Connor, T. J. and McNeilly, F. (1990b). Production and preliminary characterization of
monoclonal antibodies to chicken anemia agent. Avian Dis 34, 352–358.
101
McNulty, M. S., McIlroy, S. G., Bruce, D. W. and Todd, D. (1991). Economic effects of
subclinical chicken anemia agent infection in broiler chickens. Avian Dis 35, 263–268.
Michalski, W. P., O'Rourke, D. and Bagust, T. J. (1996). Chicken anaemia virus antibody
ELISA: problems with non–specific reactions. Avian Pathol 25, 245–254.
Moeini, H., Omar, A. R., Rahim, R. A. and Yusoff, K. (2011). Development of a DNA
vaccine against chicken anemia virus by using a bicistronic vector expressing VP1 and
VP2 proteins of CAV. Comp Immunol Microbiol Infect Dis 34, 227–236.
Natesan, S., Kataria, J. M., Dhama, K., Rahul, S. and Bhardwaj, N. (2006). Biological and
molecular characterization of chicken anaemia virus isolates of Indian origin. Virus
Res 118, 78–86.
Noteborn, M. H., de Boer, G. F., van Roozelaar, D. J., Karreman, C., Kranenburg, O., Vos,
J. G., Jeurissen, S. H., Hoeben, R. C., Zantema, A., Koch, G. and et al. (1991).
Characterization of cloned chicken anemia virus DNA that contains all elements for
the infectious replication cycle. J Virol 65, 3131–3139.
Noteborn, M. H., Verschueren, C. A., Van Roozelaar, D. J., Veldkamp, S., Van Der Eb, A. J.
and de Boer, G. F. (1992). Detection of chicken anaemia virus by DNA hybridization
and polymerase chain reaction. Avian Pathol 21, 107–118.
Noteborn, M. H., Todd, D., Verschueren, C. A., de Gauw, H. W., Curran, W. L., Veldkamp,
S., Douglas, A. J., McNulty, M. S., van der, E. A. and Koch, G. (1994). A single
chicken anemia virus protein induces apoptosis. J Virol 68, 346–351.
Noteborn, M. H., Verschueren, C. A., Koch, G. and Van der Eb, A. J. (1998). Simultaneous
expression of recombinant baculovirus–encoded chicken anaemia virus (CAV)
proteins VP1 and VP2 is required for formation of the CAV–specific neutralizing
epitope. J Gen Virol 79, 3073–3077.
102
Otaki, Y., Nunoya, T., Tajima, M., Tamada, H. and Nomura, Y. (1987). Isolation of chicken
anaemia agent and Marek's disease virus from chickens vaccinated with turkey
herpesvirus and lesions induced in chicks by inoculating both agents. Avian Pathol 16,
291–306.
Otaki, Y., Saito, K., Tajima, M. and Nomura, Y. (1991). Detection of antibody to chicken
anaemia agent: a comparison of three serological tests. Avian Pathol 20, 315–324.
Pages–Mante, A., Saubi, N., Artigas, C. and Espuna, E. (1997). Experimental evaluation of
an inactivated vaccine against chicken anaemia virus. Avian Pathol 26, 721–729.
Phan, T. G., Li, L., O'Ryan, M. G., Cortes, H., Mamani, N., Bonkoungou, I. J., Wang, C.,
Leutenegger, C. M. and Delwart, E. (2012). A third gyrovirus species in human faeces.
J Gen Virol 93, 1356–1361.
Renshaw, R. W., Soine, C., Weinkle, T., O'Connell, P. H., Ohashi, K., Watson, S., Lucio, B.,
Harrington, S. and Schat, K. A. (1996). A hypervariable region in VP1 of chicken
infectious anemia virus mediates rate of spread and cell tropism in tissue culture. J
Virol 70, 8872–8878.
Rollano Penaloza, O. M., Lewandowska, M., Stetefeld, J., Ossysek, K., Madej, M., Bereta,
J., Sobczak, M., Shojaei, S., Ghavami, S. and Los, M. J. (2014). Apoptins: selective
anticancer agents. Trends Mol Med 20, 519–528.
Sawant, P. M., Dhama, K., Rawool, D. B., Wani, M. Y., Tiwari, R., Singh, S. D. and Singh,
R. K. (2015). Development of a DNA vaccine for chicken infectious anemia and its
immunogenicity studies using high mobility group box 1 protein as a novel
immunoadjuvant indicated induction of promising protective immune responses.
Vaccine 33, 333–340.
Schat, K. A. (2009). Chicken anemia virus. Curr Top Microbiol Immunol 331, 151–183.
103
Schat, K. A. and van Santen, V. L. (2008). Chicken Infectious Anemia Virus and Other
Circovirus Infections. In Diseases of Poultry, 12th edn, pp. 211–235. Edited by A. M.
F. Y. M. Saif, J. R. Glisson, L. R. McDougald, L. K. Nolan, D. E. Swayne. Iowa, USA:
Blackwell publishing.
Scott, A. N., Connor, T. J., Creelan, J. L., McNulty, M. S. and Todd, D. (1999).
Antigenicity and pathogenicity characteristics of molecularly cloned chicken anaemia
virus isolates obtained after multiple cell culture passages. Arch Virol 144, 1961–1975.
Scott, A. N., McNulty, M. S. and Todd, D. (2001). Characterisation of a chicken anaemia
virus variant population that resists neutralisation with a group–specific monoclonal
antibody. Arch Virol 146, 713–728.
Smyth, J. A., Moffett, D. A., McNulty, M. S., Todd, D. and Mackie, D. P. (1993). A
sequential histopathologic and immunocytochemical study of chicken anemia virus
infection at one day of age. Avian Dis 37, 324–338.
Soine, C., Watson, S. K., Rybicki, E., Lucio, B., Nordgren, R. M., Parrish, C. R. and Schat,
K. A. (1993). Determination of the detection limit of the polymerase chain reaction for
chicken infectious anemia virus. Avian Dis 37, 467–476.
Spackman, E., Cloud, S. S., Pope, C. R. and Rosenberger, J. K. (2002a). Comparison of a
putative second serotype of chicken infectious anemia virus with a prototypical isolate
I. Pathogenesis. Avian Dis 46, 945–955.
Spackman, E., Cloud, S. S. and Rosenberger, J. K. (2002b). Comparison of a putative
second serotype of chicken infectious anemia virus with a prototypical isolate II.
Antigenic and physicochemical characteristics. Avian Dis 46, 956–963.
Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M. and Kumar, S. (2011).
MEGA5: Molecular Evolutionary Genetics Analysis Using Maximum Likelihood,
104
Evolutionary Distance, and Maximum Parsimony Methods. Mol Biol Evol 28, 2731–
2739.
Tamura, K., Stecher, G., Peterson, D., Filipski, A. and Kumar, S. (2013). MEGA6:
Molecular Evolutionary Genetics Analysis version 6.0. Mol Biol Evol 30, 2725–2729.
Tan, J. and Tannock, G. A. (2005). Role of viral load in the pathogenesis of chicken anemia
virus. J Gen Virol 86, 1327–1333.
Taniguchi, T., Yuasa, N., Maeda, M. and Horiuchi, T. (1982). Hematopathological changes
in dead and moribund chicks induced by chicken anemia agent. Natl Inst Anim Health
Q (Tokyo) 22, 61-69.
Taniguchi, T., Yuasa, N., Maeda, M. and Horiuchi, T. (1983). Chronological observations
on hemato–pathological changes in chicks inoculated with chicken anemia agent. Natl
Inst Anim Health Q (Tokyo) 23, 1–12.
Tannock, G. A., Tan, J., Mawhinney, K. A., Todd, D., O'Rourke, D. and Bagust, T. J.
(2003). A modified blocking ELISA for the detection of antibody to chicken anaemia
virus using an Australian strain. Aust Vet J 81, 428–430.
Thompson, J. D., Higgins, D. G. and Gibson, T. J. (1994). CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting,
position–specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673–
4680.
Todd, D., Creelan, J. L., Mackie, D. P., Rixon, F. and McNulty, M. S. (1990a). Purification
and biochemical characterization of chicken anaemia agent. J Gen Virol 71, 819–823.
Todd, D., Mackie, D. P., Mawhinney, K. A., Connor, T. J., McNeilly, F. and McNulty, M. S.
(1990b). Development of an enzyme–linked immunosorbent assay to detect serum
antibody to chicken anemia agent. Avian Dis 34, 359–363.
105
Todd, D., Mawhinney, K. A., Graham, D. A. and Scott, A. N. (1999). Development of a
blocking enzyme–linked immunosorbent assay for the serological diagnosis of chicken
anaemia virus. J Virol Methods 82, 177–184.
Todd, D., McNulty, M. S., Adair, B. M. and Allan, G. M. (2001). Animal circoviruses. Adv
Virus Res 57, 1–70.
Toro, H., Ewald, S. and Hoerr, F. J. (2006). Serological evidence of chicken infectious
anemia virus in the United States at least since 1959. Avian Dis 50, 124–126.
Trinh, D. Q., Ogawa, H., Bui, V. N., Baatartsogt, T., Kizito, M. K., Yamaguchi, S. and
Imai, K. (2015). Characterization of monoclonal antibodies to chicken anemia virus
and epitope mapping on its viral protein, VP1. J Gen Virol 96, 1086-1097.
van Santen, V. L., Li, L., Hoerr, F. J. and Lauerman, L. H. (2001). Genetic characterization
of chicken anemia virus from commercial broiler chickens in Alabama. Avian Dis 45,
373–388.
Vaziry, A., Silim, A., Bleau, C., Frenette, D. and Lamontagne, L. (2011). Chicken
infectious anaemia vaccinal strain persists in the spleen and thymus of young chicks
and induces thymic lymphoid cell disorders. Avian Pathol 40, 377–385.
Viera, A. J. and Garrett, J. M. (2005). Understanding interobserver agreement: the kappa
statistic. Fam Med 37, 360–363.
von Bülow, V. (1988). Unsatisfactory sensitivity and specificity of indirect
immunofluorescence tests for the presence or absence of antibodies to chicken
anaemia agent (CAA) in sera of SPF and broiler breeder chickens. Zentralbl
Veterinarmed B 35, 594–600.
106
von Bülow, V., Fuchs, B and Bertram, M. (1985). In vitro studies of the causative agent of
infectious anemia (CAA) in chickens: multiplication, titration, serum neutralization
test and the indirect immunofluorescence test. Zentralbl Veterinarmed B 32, 679–693.
von Bülow, V., Fuchs, B., Vielitz, E. and Landgraf, H. (1983). Early mortality syndrome in
chickens following double infection with Marek's disease virus and chicken anemia
agent. Zentralbl Veterinarmed B 30, 742–750.
Wang, D., Fan, W., Han, G. Z. and He, C. Q. (2009). The selection pressure analysis of
chicken anemia virus structural protein gene VP1. Virus Genes 38, 259–262.
Yamaguchi, S., Imada, T., Kaji, N., Mase, M., Tsukamoto, K., Tanimura, N. and Yuasa, N.
(2001). Identification of a genetic determinant of pathogenicity in chicken anaemia
virus. J Gen Virol 82, 1233–1238.
Yuasa, N. (1983). Propagation and infectivity titration of the Gifu–1 strain of chicken
anemia agent in a cell line (MDCC–MSB1) derived from Marek's disease lymphoma.
Natl Inst Anim Health Q (Tokyo) 23, 13–20.
Yuasa, N. and Yoshida, I. (1983). Experimental egg transmission of chicken anemia agent.
Natl Inst Anim Health Q (Tokyo) 23, 99–100.
Yuasa, N. and Imai, K. (1986). Pathogenicity and antigenicity of eleven isolates of chicken
anaemia agent (CAA). Avian Pathol 15, 639–645.
Yuasa, N., Taniguchi, T. and Yoshida, I. (1979). Isolation and Some Characteristics of an
Agent Inducing Anemia in Chicks. Avian Dis 23, 366–385.
Yuasa, N., Imai, K. and Tezuka, H. (1985). Survey of antibody against chicken anaemia
agent (CAA) by an indirect immunofluorescent antibody technique in breeder flocks in
Japan. Avian Pathol 14, 521–530.
107
Yuasa, N., Imai, K. and Nakamura, K. (1988). Pathogenicity of chicken anaemia agent in
bursectomised chickens. Avian Pathol 17, 363–369.
Yuasa, N., Noguchi, T., Furuta, K. and Yoshida, I. (1980a). Maternal Antibody and Its
Effect on the Susceptibility of Chicks to Chicken Anemia Agent. Avian Dis 24, 197–
201.
Yuasa, N., Taniguchi, T., Noguchi, T. and Yoshida, I. (1980b). Effect of infectious bursal
disease virus infection on incidence of anemia by chicken anemia agent. Avian Dis 24,
202–209.
Yuasa, N., Taniguchi, T., Goda, M., Shibatani, M., Imada, T. and Hihara, H. (1983a).
Isolation of chicken anemia agent with MDCC–MSB1 cells from chickens in the field.
Natl Inst Anim Health Q (Tokyo) 23, 75–77.
Yuasa, N., Taniguchi, T., Imada, T. and Hihara, H. (1983b). Distribution of chicken anemia
agent (CAA) and detection of neutralizing antibody in chicks experimentally
inoculated with CAA. Natl Inst Anim Health Q (Tokyo) 23, 78–81.
Yuasa, N., Imai, K., Watanabe, K., Saito, F., Abe, M. and Komi, K. (1987). Aetiological
examination of an outbreak of haemorrhagic syndrome in a broiler flock in Japan.
Avian Pathol 16, 521–526.
Zhang, X., Xie, Q., Ji, J., Chang, S., Liu, J., Chen, F., Ma, J. and Bee, Y. (2012). Complete
genome sequence analysis of a recent chicken anemia virus isolate and comparison
with a chicken anemia virus isolate from human fecal samples in China. J Virol 86,
10896–10897.
Zhang, X., Liu, Y., Wu, B., Sun, B., Chen, F., Ji, J., Ma, J. and Xie, Q. (2013). Phylogenetic
and molecular characterization of chicken anemia virus in southern China from 2011 to
2012. Sci Rep 3, 3519.
108
ABSTRACT
Virologic, antigenic and genetic characterization of chicken anemia virus
(CAV) and development of a new serologic diagnostic method
(鶏貧血ウイルスのウイルス学的、抗原学的および遺伝学的特徴づけと新しい血清学的検
査法の開発)
Chicken anemia virus (CAV), the member of the family Circoviridae, is a non-
enveloped, icosahedral virus with about 25 nm in diameter. The viral genome which is a
negative sense, single-stranded circular DNA of around 2.3 kb contains 3 overlapping open
reading frames (ORFs) encoding for the major structural (capsid) protein VP1 (ORF1), a
phosphatase protein VP2 (ORF2), and apoptosis−inducing protein VP3 (ORF3).
CAV has been reported in all continents since the first isolation in chickens in Japan.
When young chicks (<2 weeks old) were inoculated with CAV, they showed anorexia,
depression, or discoloration of skin and muscle due to anemia. Gross lesions in the
diseased chicks include intramuscular and subcutaneous hemorrhages, whitish−yellow
bone marrow, thymus and bursal atrophy, hemorrhages in proventriculus, and swelling of
liver. In the field, CAV−induced diseases are observed in the progeny of breeder hens
without immunity to CAV due to vertical transmission. CAV also causes subclinical
diseases in older chickens resulting in enhanced susceptibility to other avian pathogens
such as other viruses, bacteria, or fungi, and reduction in vaccinal response due to
immunosuppression. Therefore, CAV infection is of concern for economic importance in
poultry industry. Moreover, CAV infection in specific pathogen−free (SPF) flocks results in
109
serious problems in SPF facilities, because contaminated SPF chicken eggs cannot be used
for vaccine production.
It is known that VP1 (the only capsid protein) is the major protein associated with
generating neutralizing antibody against CAV. Therefore, VP1 is considered as a key target
to study pathogenicity and antigenicity of CAV, and to use as immunogen of subunit
vaccines or antigens of diagnostic kits. However, there is a lack of important information
about the property of VP1.
One of my study aims is to produce mouse monoclonal antibodies (mAbs) against
CAV and characterize CAV using the mAbs. Monitoring of CAV antibodies in chicken
flocks is important to protect young chicks from CAV vertical transmission, and also to
ensure that SPF flocks are free of CAV. Currently, virus neutralization tests (VNT), indirect
fluorescent antibody tests (IFAT), and enzyme-linked immunosorbent assays are available
to detect CAV antibodies in chickens. However, they possess some limits in the field
application, since they are generally time−consuming and laborious, and need specific
equipment or facilities. Accordingly, the second aim is to develop a simple, rapid and
highly sensitive and specific serological test, a blocking latex agglutination test (b−LAT).
This test was applied to detect CAV antibodies in chickens in Vietnam, where CAV has not
been reported, to evaluate its field application, and to investigate the presence of CAV.
Chapter I describes the production and characterization of mAbs against CAV, and
epitope mapping on VP1 (capsid protein). Three (MoCAV/F2, MoCAV/F8, MoCAV/F11)
of 4 mouse mAbs established against the A2/76 strain of CAV showed neutralization
activity. Immunoprecipitation showed that the neutralizing mAbs precipitated a protein
band of an estimated size of 50 kDa in A2/76−infected MSB1 cell lysates, corresponding to
the VP1 (50 kDa) protein, and the mAbs reacted with recombinant VP1 proteins expressed
110
in COS7 cells. The remaining mAb (MoCAV/E6) did not have the neutralization activity,
and did not detect any viral proteins in immunoprecipitation. Although the antigen staining
patterns by the neutralizing and non-neutralizing mAbs were different in IFAT, the antigens
recognized by these mAbs were partially overlapping. Viral antigens were detectable as
early as 12 h post−infection by the mAbs. In VNT, MoCAV/F11 could divide the 14 CAV
strains tested into 2 distinct mAb antigenic groups that could be associated with specific
amino acid profiles at position I75, L97, Q139, and Q144 of VP1. In blocking IFAT with
the infected MSB1 cells, binding of MoCAV/F11 was not inhibited by the other mAbs.
MoCAV/F2 inhibited the binding of MoCAV/F8 to the antigens and vice versa, which may
suggest that the 2 mAbs recognized the same epitope. However, different mutations in VP1
of the escape mutants generated from each neutralizing mAb were found: EsCAV/F2
(deletion of T89+A90), EsCAV/F8 (I261T) and EsCAV/F11 (E144G). Thus, the epitopes
recognized by MoCAV/F2 and MoCAV/F8 seemed to be topographically close in the VP1
structure, which may suggest that VP1 has at least 2 different neutralizing epitopes at this
time. Unexpectedly, however, MoCAV/F8 did not react to EsCAV/F2 containing the
epitope recognized by this mAb as well as EsCAV/F8, suggesting that binding of
MoCAV/F8 to the epitope requires coexistence of the epitope recognized by MoCAV/F2.
Also, unexpectedly, MoCAV/F2 with a titer of 1:12,800 to the parent CAV strain
neutralized EsCAV/F2 and EsCAV/F8 with low titers of 32 and 152, respectively. As
MoCAV/F2 and MoCAV/F8 reactivity to VP1 was similar each other, the existence of a
single epitope recognized by these mAbs cannot be excluded.
Chapter II describes the development of a new serological test, b-LAT, for the
detection of CAV antibodies in chickens. Polystyrene latex beads were coupled with the
neutralizing mAb (MoCAV/F11) (mAb–beads). Mixing of mAb−beads with antigens
111
prepared from the lysate of CAV–infected MSB1 cells resulted in agglutination. A short
pre-incubation (15 min) of the antigens with CAV–specific antiserum inhibited the
agglutination of mAb-beads (antibody-positive). The test results were obtained within 5
min. When SPF chicken sera and chicken positive sera to other avian viruses were tested in
b–LAT, nonspecific agglutination was not observed. The examination of 94 serum samples
collected from commercial breeder chickens of various ages (17–63 weeks) revealed a
good agreement (93.6%,κvalue = 0.82) between b–LAT and VNT, known to be most
sensitive and specific, in the detection of CAV antibodies. The b–LAT could detect CAV
antibodies in the filed breeder flocks after the outbreaks of CAV–induced diseases,
although the flocks had not possessed the antibodies before the outbreaks. These results
indicated that the simple, rapid, sensitive and specific b–LAT is a useful and reliable tool in
CAV serology.
Chapter III describes the first isolation and characterization of CAV in chickens in
Vietnam. b–LAT showed that 54.3% of the 311 chickens tested were antibody positive. By
PCR, CAV genes were detected in 19.6% of the 51 chicken tissue samples. Two CAV
isolates grew well in MSB1 cells and are not different in antigenicity from that of Japanese
CAV strains. Six CAV complete genome sequences (1,823 bp) indicated that the amino
acid substitution in VP1 contained Q394 that has been reported to be a genetic indicator for
high virulent strains. Vietnamese CAV sequences possessed the amino acid profile that is
different from that of vaccine strains. Phylogenetic analysis revealed that Vietnamese CAV
sequences were classified into 2 distinct genotypes II and III showing a worldwide
distribution.
In conclusion, analysis of CAV VP1 (capsid protein) using VP1−specific mAbs first
revealed the neutralizing epitopes on VP1 that is highly associated with the production of
112
neutralizing antibodies leading to the protection of chickens to CAV-induced diseases.
Although it remains unclear how virus particles are assembled, this study first
demonstrated that VP1 could be synthesized in the early stage of viral replication in the
infected cells. The CAV strains tested could be differentiated into two distinct mAb
antigenic groups, which could be associated with specific amino acid profiles of VP1.
Serological monitoring of breeder flocks for CAV infection prior to the laying period is
important to protect their progeny from CAV−induced diseases, and to ensure the CAV−
free status of SPF flocks. The results of b−LAT utilizing the mAb were in almost complete
agreement (93.6%,κvalue = 0.82) with those of VNT, known to be the most specific and
sensitive test, and could be rapidly obtained. Thus, b−LAT is expected to have a high
potential of its application in CAV serology. The study first demonstrated the presence of
CAV in Vietnam, but Vietnamese CAVs were not different from those of the known CAV
strains. There are still many things that must be elucidated in the pathobiology of CAV, and
the use of mAbs could be a very useful tool for better understanding of this purpose.
113
和文要約和文要約和文要約和文要約
鶏貧血ウイルスのウイルス学的、抗原学的および遺伝学的特徴づけと鶏貧血ウイルスのウイルス学的、抗原学的および遺伝学的特徴づけと鶏貧血ウイルスのウイルス学的、抗原学的および遺伝学的特徴づけと鶏貧血ウイルスのウイルス学的、抗原学的および遺伝学的特徴づけと
新しい血清学的検査法の開発新しい血清学的検査法の開発新しい血清学的検査法の開発新しい血清学的検査法の開発
鶏貧血ウイルス(CAV)は、サーコウイルス科ジャイロウイルス属に分類されている直
径約 25 nmの正二十面体対称型の非エンベロープウイルスである。ウイルス遺伝子は、
2.3 kb の(−)鎖の環状一本鎖 DNA であるが、3 つの部分的に重なるオープンリーデイング
フレイム (ORF 1, 2, 3)を含んでいる。これらの ORFはそれぞれ唯一の構造タンパク（カ
プシド蛋白）である VP1 (ORF1、52 kDa)、プロテインフォスファターゼ活性のある VP2
（ORF2、26 kDa）、アポトーシスを引き起こす蛋白である VP3 (ORF3、14 kDa)をコード
している。
CAV は 1979年に日本で分離されて以来、世界中に分布していることが明らかになっ
ている。2 週齢以下の鶏に CAV を接種した場合、死亡、沈鬱、食欲不振、貧血に起因す
る皮膚や筋肉の蒼白化が観察される。肉眼病変として、皮下や筋肉の出血、腺胃の出血、
骨髄の退色、胸腺の萎縮、肝腫大が観察される。野外では CAV が引き起こす疾病は、主
に産卵期に入って免疫をもたない種鶏が CAV の感染を受けた場合に、介卵感染により孵
化したひなで認められている。2 週齢以上の鶏では CAV は不顕性感染を引き起こすが、
免疫抑制による他の病原体（他のウイルス、細菌、真菌など）への感受性の亢進やワク
チン応答の減弱が報告されている。それ故、CAV は養鶏産業にとって経済的に重要な感
染症である。更に、SPF施設にとっても CAV 感染は大きな問題となる。何故ならば、汚
染された SPF卵はワクチン生産に使用できないからである。
VP1 は鶏の中和抗体産生と密接な関連がある主要なウイルス蛋白である。それ故、
VP1 は CAV の病理発生、病原性や抗原性の研究、およびサブユニットワクチンや診断キ
114
ットの抗原として使用が期待されている。しかしながら、VP1の性状に関する情報は少な
い。本研究の目的のひとつは、マウスモノクローナル抗体 (mAb) の産生とそれを用いた
CAV の性状解明である。
コマーシャルおよび SPF種鶏群のおける CAV 抗体のモニタリングは、CAV の介卵感
染を未然に阻止する上で重要である。現在、血清学的診断法として中和試験 (VNT)、蛍光
抗体法 (IFAT)、ELISA が用いられている。しかし、いずれの方法も時間や労力、特別な設
備や施設が必要であるなどの欠点があり、特に農場などの現場で実施することが困難で
ある。それ故、本研究の二つ目の目的は、簡便且つ迅速、高感度および高特異性である
血清学的診断法を開発することであるが、今回ブロッキングラテックス凝集試験 (b−LAT)
を開発した。また、b−LAT の有用性を評価する目的で CAV の存在が報告されていないベ
トナムの鶏からの抗体検出を試みた。
第１章では、CAV A2/76株に対する mAbの確立とその特徴と VP1 (カプシド蛋白) 上
に存在するエピトープを初めて明らかにした（エピトープマッピング）。確立した 4 つ
mAb のうち３つの mAb（MoCAV/F2, MoCAV/F8, MoCAV/F11）が中和活性を示した。免
疫沈降試験において、これら中和活性 mAb は感染 MSB1細胞のライセートとの反応にお
いて 50 kDaのバンドを沈降させ、検出されたバンドの分子量は報告された VP1のそれと
一致した。また、それらの mAbは組換え VP1を発現している COS7 細胞とも反応した。
中和活性を示さなかった残りの mAb (MoCAV/E6) は、免疫沈降試験においてウイルス蛋
白を沈降しなかったことから、この mAb の標的ウイルス蛋白を明らかにできなかった。
中和活性および非中和活性 mAb の感染細胞内の蛍光抗原の染色パターンは異なっていた
が、一部抗原部位が重なっていた細胞も少数認められた。CAV 感染細胞では、VP1 抗原
は CAV 接種後 12 時間ほどで検出されたことから、ウイルスの構成蛋白である VP1 の感
染後早期の産生が初めて明らかにされた。MoCAV/F11 を用いた中和試験によって、14 株
115
の CAV は２つの抗原グループに分けられたが、これらの抗原グループは、VP1 の特定の
アミノ酸プロファイル (I75, L97, Q139, Q144) との関連が示唆された。
MoCAV/F11のウイルス抗原への結合は、他の mAbによって阻害されなかった。一方、
MoCAV/F2 の結合は MoCAV/F8 によって阻害され、逆に、MoCAV/F8 の結合も
MoCAV/F2 によって阻害された。この結果は、これらの２つの mAb が認識するエピトー
プが同一であることを示唆している。しかし、中和活性 mAb を用いて作製されたエスケ
ープミュータント（EsCAV/F2、EsCAV/F8、EsCAV/F11）の VP1 のアミノ酸解析によって、
EsCAV/F2 では T89と A90の欠損、EsCAV/F8では I261Tのアミノ酸置換、 EsCAV/F11で
は E144Gのアミノ酸置換がそれぞれ認められたことから、MoCAV/F2 および MoCAV/F8
が認識する中和エピトープは、位相幾何学的に近接していることが示唆され、VP1上には
少なくとも 2 つの中和エピトープの存在が想定された。しかしながら、予想に反して
MoCAV/F8 は、この mAb が認識するエピトープを持っているエスケープミュータント
EsCAV/F2 を中和できなかった。この結果は MoCAV/F8 のエピトープへの結合は
MoCAV/F2 が認識するエピトープの共存が必要であることを示している。一方、
MoCAV/F2の A2/76株（免疫原）に対する中和抗体価は 1:12,800であるが、予想に反して
MoCAV/F2 はエスケープミュータントである EsCAV/F2 と EsCAV/F8を中和した（それぞ
れの中和抗体価は 1:32および 1:152）。このように MoCAV/F2および MoCAV/F8 の反応性
から、これらの mAbが VP1上の単一のエピトープを認識している可能性が示唆された。
本研究において、中和抗体の産生や病原性に関与すると考えられている VP1（カプシド
蛋白）上の中和エピトープの存在が初めて明らかにされた。また、VP1の産生が感染細胞
内で早期起こることが示された。今回の知見は、CAV の感染増殖機序の解明に資するこ
とが期待される。
第 II 章では、新しく開発した血清学的診断法である b−LAT について記述する。ポリ
スチレンラテックスビーズに中和 mAb（MoCAV/F11）を結合させたビーズ (mAb–bead) と
116
CAV 感染 MSB1 細胞から作製したウイルス抗原を混合すると mAb–beadの凝集が 5 分以
内に観察された。ウイルス抗原と鶏血清を 15 分間反応後、混合液に mAb–beadを加えて
5 分間混和した時に、凝集像が観察されなかった場合を抗体陽性、凝集陽性を抗体陰性と
判定した。b−LAT を用いて野外の 17〜63週齢の種鶏血清 94サンプルを調べたところ、最
も特異的且つ高感度である中和試験 (VNT) の結果と高い一致率（93.6%、κvalue = 0.82）
を示した。b−LAT は野外種鶏 3 群において CAV 感染後の抗体陽転を検出することができ
た。本研究で開発した新しい血清学的診断法である b−LAT は、CAV 抗体検出において簡
便且つ迅速、高感度・高特異性を示したことと分析機器が不要なことから、農場などの
野外において実施可能であることが示された。
第 III 章では、ベトナムの鶏からの CAV 初分離とその特徴について記載されている。
b–LAT を用いて鶏血清 311サンプルを調べたところ、54.3%が抗体陽性であった。51羽の
鶏の肝臓および脾臓検体から、PCRにより CAV 遺伝子の検出を行ったところ 19.6%のサ
ンプルが CAV 遺伝子陽性を示した。また、これらのサンプルから 2株の CAV が分離され
たが、抗原性や遺伝学的特徴は参照ウイルス株と同一であった。分離株の VP1 アミノ酸
配列の 394 番目がグルタミンであった。この変異は病原性発現と関連する”genetic
indicator”として報告されている。系統樹解析から、分離株は世界の多くのウイルス株が
入っている 2つの遺伝子グループ (genotype II および III) に分類された。ベトナムにおけ
る CAV の存在はこれまで報告がなく、今回、初めてベトナムにおける CAV の存在が明ら
かにされた。
CAV の病理生態学については、解明しなければならない不明点が多く残されている
が、本研究において、CAV の感染増殖機序や病理発生を明らかにするうえで重要な知見
が得られた。さらに、mAb を利用した抗体を簡便且つ迅速、高感度・高特異的検出でき
る新しい診断法が開発されたことから、得られた成果が CAV の感染状況の把握や CAV 感
117
染症のより有効な制御法の確立に資することが期待される。また、mAb が CAV 感染症に
関する研究において、有用な解析アイテムとなりうることが判明した。