zona pellucida antiviral

8/6/2019 Zona Pellucida Antiviral

1/12

Vaccine 25 (2007) 679690

Innate antiviral resistance influences the efficacy of a recombinantmurine cytomegalovirus immunocontraceptive vaccine

Megan L. Lloyd, Sonia Nikolovski, Malcolm A. Lawson, Geoffrey R. Shellam

Discipline of Microbiology and Immunology, M502, School of Biomedical, Biomolecular and Chemical Sciences,

University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia

Received 25 April 2006; received in revised form 26 July 2006; accepted 13 August 2006

Available online 24 August 2006

Abstract

Recombinant betaherpesviruses are attractive vaccine candidates because of their persistence in the host. A recombinant murine

cytomegalovirus expressing the mouse ovarian glycoprotein zona pellucida 3 induces long lasting sterility in female BALB/c mice. Using

inbred mouse strains selected for their innate resistance or susceptibility to MCMV, we show that genetically determined innate resistance

to MCMV can reduce immunocontraceptive success. The Cmv1 locus that controls natural killer cell mediated responses to MCMV was

implicated in determining vaccine efficacy. However, the role of the H-2 haplotype was less clear. Interestingly, Mus domesticus from an

outbred colony of wild-derived micewere readily sterilised by vaccination, consistent with observations that strong innate immunity to MCMV

is not common in Australian wild mice.

2006 Elsevier Ltd. All rights reserved.

Keywords: Recombinant vaccine; Immunocontraception; Murine cytomegalovirus; Zona pellucida

1. Introduction

Characteristics such as lifelong latency [1,2] coupled with

the ability to stably express foreign antigens [35] have led to

the proposition that theHerpesviridae may be candidate vac-

cine vectors [6,7]. In our laboratory, a recombinant MCMV

constructed to express the murine fertility antigen, zona pel-

lucida 3 (ZP3) induced long term infertility (>250 days) in

female BALB/cmice aftera single intraperitonealinoculation

[8]. In addition, a recombinant MCMV expressing immun-

odominant peptides derived from either influenza A virus or

lymphocytic choriomeningitis virus induced an expansion ofthe peptide-specific CD8+ T-cell population without the use

of an adjuvant or the need for re-immunisation [9]. These

experiments demonstrate that MCMV based vaccines can

induce long lived immune responses, probably reflecting pro-

longed stimulation of the immune system by the engineered

antigen.

Corresponding author. Tel.: +61 8 9346 2510; fax: +61 8 9346 2912.

E-mail address: [email protected] (M.L. Lloyd).

The prototype immunocontraceptive virus expressing

murine ZP3 was constructed using the K181 (Perth) strain

of MCMV. The laboratory infection of MCMV has been

well characterised and inbred mouse strains are suscepti-

ble (BALB/c and A/J), moderately resistant (C56BL/6) or

resistant (CBA) to acute MCMV disease [1012]. The sur-

vival of MCMV-resistant mouse strains is largely attributed

to the ability of these mice to control viral growth in vis-

ceral organs [11]. This control has been linked to par-

ticular H-2 associated haplotypes, and to non-H-2 associ-

ated genes [10]. Resistance attributable to non-H-2 genes

is largely due to innate immune mechanisms principallymediated by natural killer (NK) cells [13,14]. A host resis-

tance locus, Cmv1, which controls NK cell mediated clear-

ance of MCMV has been identified within the natural

killer cell gene complex on chromosome 6 [14]. Mod-

erately MCMV-resistant C57BL/6 mice (Cmv1r) exhibit

low levels of MCMV replication in the spleen, while in

MCMV-susceptible (Cmv1s) strains such as BALB/c and

A/J, MCMV reaches 1000-fold higher titres in the spleen.

Resistance attributable to Cmv1r is mediated by the Ly49H

0264-410X/$ see front matter 2006 Elsevier Ltd. All rights reserved.

doi:10.1016/j.vaccine.2006.08.019
mailto:[email protected]://dx.doi.org/10.1016/j.vaccine.2006.08.019http://dx.doi.org/10.1016/j.vaccine.2006.08.019mailto:[email protected]


2/12

680 M.L. Lloyd et al. / Vaccine 25 (2007) 679690

activation receptor on NK cells of resistant C57BL/6 mice

[1517].

Because innate immunity can restrict viral growth in the

host [9,12], we hypothesized that this factor would impair the

ability of an MCMV-based immunocontraceptive vaccine to

induce infertility in MCMV-resistant mouse strains. Accord-

ingly, we have studied the ability of rK181-mZP3 to induceinfertility in inbred and congenic strains of mice differing

in innate resistance to MCMV. We show that immunocon-

traception after inoculation of 2 104 plaque forming units

of rK181-mZP3 is incomplete in the MCMV-resistant strains

C57BL/6 and CBA, while inoculation of the same dose in the

MCMV-susceptible BALB/c and A/J strains induces com-

plete and long-lasting infertility.

This immunocontraceptive virus was constructed to

enhance current methods for controlling population explo-

sions of free living wild mice [18]. It was important therefore

to assess vaccine efficacy in wild mice. Using female wild

mice from a specific pathogen free outbred colony, long last-

ing sterility was readily induced. This result is consistentwith a report that these mice have polymorphisms in the nat-

ural killer cell complex and do not exhibit significant innate

resistance to the viral vector, MCMV [19]. These findings

have implications for the use of viruses as therapeutic tools

to deliver medically relevant antigens in genetically diverse

populations, where innate antiviral resistance to the vector

may confound vaccination attempts.

2. Materials and methods

2.1. Animals

The highly inbred strains BALB/cArc (BALB/c),

C57BL/6J (C57BL/6), CBA/CaHArc (CBA), A/JArc (A/J),

B10.D2/N2Sn-H2dArc(B10.D2)and C.C3-H2k/LilMcdJArc

(BALB.H-2k) were obtained as specific pathogen free (SPF)

from the Animal Resources Centre (Murdoch, Western Aus-

tralia) and maintained under minimal disease conditions.

Wild mice (Mus domesticus/Arc) were obtained from an out-

bred SPF colony generated by the caesarean derivation of the

offspring of 15 breeding pairs of wild caught mice trapped

near Jerilderie, Deniliquin, Braidwood, Yenda, Gungahlin

and Canberra in south-eastern Australia [19]. The outbred

status was maintained by a modification of the Poiley proto-

col used to maintain uniform genetic variability based on the

co-efficient of inbreeding [20]. All mice used were 68 weeks

of age. BALB.B6-CT3, B6.BALB-Cmv1s and B6.BALB-

TC1 NK gene complex (NKC) congenic strains were bred

in the Animal Services Facility at the University of West-

ern Australia. BALB.B6-CT3 (CT3) mice were generated by

brothersister mating of F1 (BALB.B6-Cmv1rBALB/c)

mice to create a homozygous line that includes markers from

Cd94 toD6Mit2, including theLy49a and Cmv1ralleles from

C57BL/6 on a BALB/c genetic background [21]. B6.BALB-

Cmv1s (Cmv1s) mice contain the Cmv1s gene and do not

express the NK1.1 alloantigen. B6.BALB-TC1 (TC1) mice

are an intra NKC recombinant strain generated by a back-

cross of B6.BALB-Cmv1s onto C57BL/6J mice. TC1 mice

contain the Cmv1s gene, express the NK1.1 alloantigen and

both B6.BALB-Cmv1s and TC1 mice show levels of splenic

infection similar to the BALB/c parental strain [21]. These

mice were kindly made available by Dr. A Scalzo (Lions EyeInstitute, Perth, Australia). Mouse care was based on theAus-

tralian Code of Practice endorsed by the National Health and

Medical Research Council, and was approved by the Univer-

sity of Western Australias Animal Experimentation Ethics

Committee. The presence of known murinepathogens includ-

ing MCMV was excluded by regular testing of sentinel mice.

2.2. Viruses and in vivo growth

The origins of K181 (Perth) have been described else-

where [14]. The production and characterisation of recombi-

nant MCMV containing the murine ZP3 gene from BALB/c

mice inserted into the non-essential ie2 region of the genome,hereafter abbreviated as rK181-mZP3, has been described

elsewhere [8]. RM427+ was the parental virus from which

rK181-mZP3 was constructed and was provided by Dr. E.

Mocarski (Stanford University, California, USA). RM427+

contains LacZin the ie2 region of the genome. The origins of

this virus have been detailed elsewhere [22]. All virus stocks

were tissue culture derived and stored in MEM supplemented

with 2% fetal calf serum. To measure the in vivo growth of

the recombinant and control viruses, BALB/c, C57BL/6 and

CBAmicewereusedandfourmicepertimepointwereinocu-

latedwith2 104 plaqueforming units (pfu) of rK181-mZP3

or K181. Organs were assessed on days 7, 14, 21 and 35 postinoculation by quantitative PCR. Ovaries were frozen in Tis-

suetek OCT (optimal cutting temperature) compound. Blood

was collected by cardiac puncture.

2.3. Plaque assay

Tissue culture virus or clarified 10% organ homogenates

were diluted in MEM + 2% newborn calf serum and plated

onto confluent mouse embryo fibroblasts in 24 well trays.

After 1 h incubation in 5% CO2 the suspension was removed

and 1 ml of 0.5% methylcellulose/MEM + 2% NCS was

applied per well. The plates were incubated for 4 days at

37 C in 5% CO2 and stained overnight with 1 ml of 0.05%

methylene blue with 10% formaldehyde.

2.4. Detection of virus genome from organ DNA by real

time PCR

DNA was extracted from organ homogenates using

standard procedures [23]. A commercial glyceraldehyde-

3-phosphate dehydrogenase (GAPDH) real time PCR kit

(Applied Biosystems, CA) was used to evaluate DNA con-

centrations using the manufacturers suggested method. All

real time PCR was carried out using the Perkin-Elmer Gene


3/12

M.L. Lloyd et al. / Vaccine 25 (2007) 679690 681

Amp PCR System 960C. A standard curve was constructed

by analysingidenticalsamples of diluted DNA extracted from

salivary gland and lung by spectrophotometry and GAPDH

real time PCR, and a linear correlation was calculated. This

equation (y =2.0189 ln(x) + 39.734, R2 = 0.975), was used

to convert the critical threshold (CT) score obtained from

each sample to ng DNA per sample. Identical samples forindividual organs were assayed by quantitative PCR for the

K181 ie1 gene. Real time PCR of the MCMV ie1 gene was

carried out using theforward primer5 TACGGCTGTTTCA-

GATCTGAGTTT 3, the reverse primer 5 CCTACG-

TAGCTCATCCAGACTCTCT 3 and the probe 6FAM-

ACCCACACTTCATGCCCACTAATCTAGG-TAMRA. CT

scores were obtained and converted to copy number using a

standard curve (y =1.232 ln(x) + 39.40,R2 = 0.940 for DNA

extracted from salivary glands and y =1.394 ln(x) + 40.61,

R2 = 0.955) for DNA extracted from lung samples [24]. The

standard curves were obtained by seeding organ DNA with

K181 plasmid as previously described [25]. All samples were

converted to copy number per ng DNA using the concentra-tion of DNA determined by the GAPDH results. All PCRs

were performed at least twice.

2.5. Determination of antibody bound to ovaries by

direct immunofluorescence

Pooled ovaries from BALB/c, C57BL/6 and CBA mice

inoculated with 2 104 pfu rK181-mZP3 or K181 were

frozen in OCT compound. Frozen sections were cut (5 M)

and were stored at 80 C. Ovaries from mice infected with

K181 or rK181-mZP3 from an individual timepoint were

embedded at specific positions in the same mould (Cry-omold, Tissue-Tek, Sakura, USA) to ensure identical treat-

ment of these tissues. Slides were thawed, fixed in 100%

methanol at 20 C for 5 min and blocked with 10% rab-

bit serum for 30 min after which they were incubated with

a rabbitanti mouse Fab fragment labelled with fluorescein

isothiocianate (FITC) (DakoCytomation, Carpinteria, CA) at

37 C for 30 min at a 1:150 dilution. All slides were mounted

in 50% glycerol and immediately examined by fluorescence

microscopy. Photographs were taken at the same setting to

eliminate bias.

2.6. Breeding experiments

Groups of 68-week-old female mice were inoculated

with 2 104 pfu of virus or sham-inoculated with 100 l

diluent (mouse osmolarity buffered saline (MOBS) + 0.5%

heat inactivated FCS) by the intraperitoneal (i.p.) route. Male

mice were introduced immediately post inoculation at a ratio

of one male to one female (breeding pairs) or one male to two

females (breedingtrios) or were introduced 21 days post-viral

inoculation (breeding groups). The experimental design for

the breeding pairs and trios allows one litter to be born to

each female before the initiation of the immunocontracep-

tive effect and establishes the overall fertility of the male and

female mice being assessed. Where fertility was not estab-

lished, the number of fertile females is noted. Male mice

were housed in separate cages for several days before the

introduction of female mice to enhance testosterone produc-

tion. The number of pups born was recorded for 100 days.

Pups were culled on the day of birth. Excepting the A/J trios,

the individual groups inoculated with 2

105

pfu rK181-mZP3 and the BALB/c dose response, the breeding exper-

iments were performed at least twice and all experiments

contained a cohort of BALB/c mice that were inoculated with

rK181-mZP3 to act as a positive control. Due to the gener-

ally poor fecundity exhibited by the congenic mouse strains,

results from 2 (CT3) or 3 (BALB.K and TC1/Cmv1s) sepa-

rately performed experiments each consisting of six breed-

ing females per treatment have been presented together. All

congenic and wild mouse results are presented per fertile

female.

2.7. ZP3 ELISA

The ZP3 ELISA was a modification of the method pre-

viously described [8]. Briefly, ELISA plates were coated

with 75 ng crude protein extract diluted in Carb/Bicarb buffer

overnight at room temperature. Plates were then blocked

with mouse osmolarity buffered saline (MOBS) + 5% skim

milk powder at 4 C overnight. The plates were washed after

each step with MOBS + 0.05% Tween 20 and MOBS + 10%

Superblock (Pierce, IL, USA) was used as the antibody dilu-

ent. Hyperimmune serum generated by serial inoculation of

mice with rK181-mZP3 was used as a positive control and

normal mouse serum was used as a negative control on all

plates. Serum was incubated for 2 h at 37

C. IgG1 and IgG2ahorse radish peroxidase-labelled conjugates were used as

secondary antibodies (BD Pharmingen, San Jose, CA). A

Sigma FAST o-phenylenediamine dihydrochloride tablet set

was used to develop the colour (SigmaAldrich, St Louis,

MO) and ELISA plates were read at 450nm after 15 min

incubation at room temperature. Due to the polycolonal anti-

body response elicited in response to infection with MCMV

[26,27], antibody titresestimated using theZP3 ELISAcan be

falsely elevated. To compensate for this, sera were collected

from control mice inoculated with K181 as well as from test

mice inoculated with rK181-mZP3 at all times and from all

strains. Test and control sera were titrated in the same ELISA

run andindividual negative control values were calculated for

each strain/time. The reciprocal value of the dilution with an

OD less than the mean OD of the 1:50 dilution of the control

sera was determined to be the titre. Data is presented as mean

titre per group.

2.8. Statistical analysis

All statistical analysis was carried out using non para-

metric tests to ensure that the assumption of normality was

not violated. Accordingly, MannWhitney or KruskalWallis

tests were used to evaluate data where appropriate. Statistical


4/12


analysis was carried out using Analyse-it Software version

1.71 (Analyse-it for Microsoft Excel, Leeds, UK). All errors

reported are the standard error of the mean.

3. Results

3.1. BALB/c mice are made infertile by rK181-mZP3

Groups of six female BALB/c mice were inoculated i.p.

with 2104 pfu rK181-mZP3 or the control virus K181

and male mice were introduced immediately after inocu-

lation into breeding trios. BALB/c mice inoculated with

2104 pfu rK181-mZP3 showed a statistically significant

reduction in the mean number of pups born per female over

100 days (0.67 0.33, number of fertile females = 3) com-

pared with those inoculated with K181 (14.0 2.46, number

of fertile females = 6), p = 0.002. Fig. 1A shows the cumula-

Fig. 1. Effect of rK181-mZP3 on BALB/c and A/J mice. For the BALB/c

mice, six female mice were inoculated i.p. with virus, males were added

immediately and the number of pups born was recorded for 100 days. For

the A/J mice, nine female mice were inoculated i.p. with virus or were

sham-inoculated. Males were added to groups of three females 21 days later

and the number of pups born was recorded for 100 days. Unless otherwise

described, all inocula were 2 104 pfu MCMV. (A) BALB/c mice were

inoculated with rK181-mZP3 or a control virus. (B) BALB/c mice were

inoculated with 10-fold reductions in rK181-mZP3, from 2 104 to 20pfu

or were sham-inoculated with diluent. (C) A/J mice were inoculated with

rK181-mZP3, a control virus or were sham-inoculated.

tive number of pups per female born to BALB/c mice over

100 days.

To determine whether the immunocontraceptive effect

was dose-dependent in BALB/c mice, groups of females

were inoculated i.p. with 20, 2 102, 2 103, 2 104 pfu

of rK181-mZP3 and were set up in breeding pairs (Fig. 1B).

These results were confounded by several female mice fail-ing to deliver any pups in the groups inoculated with 2102,

2 103, 2 104 pfu of rK181-mZP3 (2, 4 and 3 females did

not deliver pups, respectively). As all control mice and mice

inoculated with the lowest dose of rK181-mZP3 (20 pfu)

were completely fertile, the mice having no litters were

included in the calculation. Control female mice (diluent) had

a mean number of pups per female of 22.8 2.50 and mice

inoculated with 20 pfu rK181-mZP3 had a mean number of

pups per female of 15.8 2.32 (p = 0.06). Mice inoculated

with 2 102 pfu had a mean number of pups per female of

9.3 4.06 (p = 0.03), mice inoculated with 2 103 pfu had a

mean number of pups per female of 4.2 2.99 (p = 0.009)

and mice inoculated with 2 104 pfu rK181-mZP3 had amean number of pups per female of 1.83 0.83 pups per

litter (p = 0.002). Statistical significance was determined by

MannWhitney testing compared with control mice.

3.2. The MCMV-sensitive A/J mouse strain is made

infertile by rK181-mZP3

A/J mice are also very susceptible to experimental infec-

tion with MCMV [13]. Groups of nine female mice were

inoculated with either 2 104 pfu rK181-mZP3 or the con-

trol virus RM427+, or were sham-inoculated. After 21 days

the mice were set up as breeding groups (three females to onemale). There was a marked reduction in pup numbers after

100 days in mice inoculated with rK181-mZP3 (mean num-

ber ofpupspergroup ofthree females = 3.3 1.63)compared

with those inoculated with RM427+ (mean number of pups

per group = 45.3 3.3) or sham-inoculated (mean number of

pups per group = 44.7 1.33), Fig. 1C.

3.3. There is a reduced immunocontraceptive effect in

C57BL/6 mice

Groups of six female C57BL/6 mice were inoculated i.p.

with 2 104 or 2 105 pfu rK181-mZP3 or with the con-

trol virus K181 and were set up in breeding trios. C57BL/6

mice inoculated with 2 104 pfu of rK181-mZP3 showed a

43% reduction in pups (mean pups per female 11.8 1.81,

p = 0.7) compared with the control mice (20.7 6.47), and

a 66% reduction in pups when inoculated with a 10-fold

higher dose (2 105 pfu) of rK181-mZP3 (mean pups per

female 7.0 1.48, p = 0.24, number of fertile females = 5).

The difference in fertility between mice inoculated with dif-

fering doses of rK181-mZP3 was not statistically significant

(p = 0.06). Fig. 2A shows the cumulative number of pups per

female born to CB7BL/6 mice over 100days. Sera were taken

at the completion of the breeding experiment and analysed by


5/12


Fig.2. Effect of rK181-mZP3 on inbred C57BL/6and CBA mice.Six female

mice were inoculated i.p. with virus, males were added immediately and the

number of pups born was recorded for 100 days. Unless otherwise stated, all

inocula were 2 104 pfu MCMV. (A) C57BL/6 mice were inoculated with

rK181-mZP3, 2 105 pfu rK181-mZP3 or a control virus. (B) CBA mice

were inoculated with rK181-mZP3, 2 105 pfu rK181-mZP3 or a control

virus.

ELISAfor ZP3antibodies (Fig.3). This shows that mice inoc-

ulated with 2 105 pfu rK181-mZP3 (rK181-mZP310)

had higher antibody titres than those inoculated with the

lower dose and that antibody levels were similar to thosedetected from the sera of BALB/c mice. Direct fluorescence

was carried out on frozen sections from these animals. An

increase in antibody binding to the follicular zona pellucida

was seen in mice inoculated with 2 105 pfu rK181-mZP3

(results not shown), reflecting the ELISA data.

3.4. There is a reduced immunocontraceptive effect seen

in CBA mice

Groups of six female CBA mice were inoculated with

2 104 or 2 105 pfurK181-mZP3 or thecontrol virus K181

and were set up in breeding trios. CBA mice inoculated with

2 104 pfu rK181-mZP3 showed no reduction in the mean

number of pups per female (18.2 1.89) compared with the

control group inoculated with K181 (20.0 1.73). However,

there was a 40% reduction in cumulative pups when a 10-fold

higher dose (2105 pfu) of recombinant virus was inoc-

ulated (Fig. 2B) resulting in a mean number of pups per

female of 11.8 3.06. This result was statistically significant

(p = 0.004 rK181-mZP3 10 against K181 control, p = 0.06

between the two doses of ZP3). Very little ZP3 antibody was

detected by the ELISA in serum taken at the completion of

the breeding experiment (Fig.3). Direct immunofluorescence

of ovarian tissue from these mice revealed that antibody was

Fig. 3. ZP3antibodiesin BALB/c, C57BL/6 andCBAmice at 100dayspost

inoculation detected by ZP3 ELISA. Serum was collected from female mice

at the completion of breeding experiments. Results are presented as mean

titre per strain per virus treatment. Error bars are standard error of the mean.

(A) IgG1. (B) IgG2a.

bound to the follicles of mice inoculated with either dose of

rK181-mZP3 (results not shown).

3.5. Detection of virus in vivo by real time PCR

It has been reported previously that the plaque assay is notsufficiently sensitive to accurately determine the amount of

recombinant virus present in organ homogenates [8]. Sub-

sequently real time PCR was used to measure the viral

copy number present per g of organ DNA in MCMV-

susceptible (BALB/c), moderately resistant (C57BL/6) or

resistant (CBA) mice. The results are presented (Fig. 4) as the

mean viral copy numberperg DNA detected from four inde-

pendent organ homogenates, independently analysed on two

occasions. The theoretical limit of detection of one viral copy

per g organ DNA is denoted by the horizontal broken line

and values below this line indicate that not all samples tested

had virus genome detected. The K181 genome was readily

detected in the salivary glands and lungs from BALB/c mice

at alltimestested, with viral copy numberin thesalivary gland

peaking at 21 days post inoculation and declining thereafter,

and the viral copy number slowly increasing in lung tissue

from BALB/c mice over time (Fig. 4A and C). The K181

genomewas consistentlydetectedat a low level in thesalivary

glands of C57BL/6 and CBA miceat mosttimes tested, with a

slightly elevated viral copy number being detected from CBA

salivary glands (Fig. 4A). The K181 copy number detected in

the lungs of C57BL/6 and CBA mice remained extremely low

at all times tested, and was virtually undetectable by 35 days

post inoculation (Fig. 4C). Very little rK181-mZP3 genome


6/12


Fig. 4. Virus detected in DNA extracted from organ tissue by real time

PCRin BALB/c, C57BL/6 andCBAmice inoculated with 2104 pfu K181

(control) or rK181-mZP3. Salivary glands and lungs were collected fromfour mice per strain per time point. Equivalent samples were analysed for

GAPDH and MCMV detection. Results are presented as the mean viral

copy number per g organ DNA. Error bars are standard error of the mean.

The broken line denotes the theoretical limit of detection of one viral copy

per microgram organ DNA. (A) Salivary glands (control mice). (B) Salivary

glands (rK181-mZP3). (C) Lungs (control mice). (D) Lungs (rK181-mZP3).

was detected in the salivary gland from any strain, with only

BALB/c mice consistently showing evidence of the presence

of viral genome (Fig. 4B). The rK181-mZP3 genome was

detected in the lung tissue of all mouse strains tested at a

very low level (Fig. 4D).

3.6. Antibody is bound to the ZP in vivo in susceptible

and resistant mouse strains

Ovaries were removed from adult BALB/c, B6 and CBA

mice at 7, 14, 21 and 35 days post inoculation with K181

or rK181-mZP3 and frozen sections were cut for analysis by

direct immunofluorescence. Fluorescence was not detectedin any strain at 7 days post inoculation, but was present in all

strains by 14 days post inoculation, indicating that antibody

was bound in vivo to the follicular zona pellucida (Fig. 5A).

There was no obvious difference in the fluorescence intensity

in sections from the different strains, and no obvious fluctu-

ation over time.

3.7. The detection of ZP3 antibody bound in vivo by

immunofluorescence is more sensitive than detection of

serum antibody by ZP3 ELISA

Serum was taken from the mice described above and was

tested for the presence of ZP3 antibodies by ZP3 ELISA(Fig. 5B and C). There was a clear increase in antibody titre

seen over time in all three mouse strains. In BALB/c mice,

both the IgG1 and the IgG2a antibody isotypes were present

by Day 35 at equivalent titres. In C57BL/6 and CBA mice

there appeared to be an IgG2a bias at early times post inoc-

ulation, with very little IgG1 being detected in CBA mice at

35 days post inoculation. At 14 days post inoculation, there

is very little ZP3 antibody detected in the serum from any

mouse strain indicating that immunofluorescence on frozen

sections of ovary affords more sensitive detection of in vivo

ZP3 antibody.

3.8. The efficacy of immunocontraception is not defined

by the H-2 haplotype

Both H-2 and non-H-2 genes are known to influence the

susceptibility of laboratory mouse strains to MCMV [10].

Accordingly, two congenic strains were assessed for their

sensitivity to rK181-mZP3. Most congenic experiments were

set up as breeding trios.

BALB.H-2k mice (K181-resistant H-2k allele on a sus-

ceptible BALB/c background) were inoculated with either

rK181-mZP3 or the control virus K181. Three separate

experiments were performed and the combined results

are presented in Fig. 6A. Due to the poor fecundity of

this strain overall, the data from three experiments per-

formed separately have been statistically analysed together

using KruskalWallis analysis. The mean number of pups

in females inoculated with the control virus (n =20) was

10.6 1.01, and for those inoculated with rK181-mZP3

(n = 20) it was 5.1 0.90 (Fig. 6A). The difference was sta-

tistically significant (p = 0.0002, KruskalWallis).

B10.D2 mice (K181-sensitive H-2d allele on a moderately

resistant B10 background) were inoculated with 2 104 pfu

rK181-mZP3, 2 105 pfu rK181-mZP3, the control virus

RM427+ or were sham-inoculated (Fig. 6B). The mean num-


7/12


Fig. 5. Detection of ZP3 antibody at early times post infection. (A) Direct immunofluorescence of ovaries from BALB/c (B/c), C57BL/6 (B/6) and CBA mice

infected with rK181-mZP3. Ovaries were harvested from mice from the previous experiment. Bars on Day 7 and 21 photographs denote 100 m and bars on

Day 14 photographs denote 50m. Frozen sections of ovaries were incubated with FITC-linked conjugate to detect antibody bound in vivo. Serum taken from

these mice was assayed by ZP3 ELISA, (B) IgG1. (C) IgG2a. *No antibodies detected. Error bars are standard error of the mean.


8/12


Fig. 6. Effect of rK181-mZP3 on congenic mouse strains. Female mice

were inoculated i.p. with virus,maleswere added immediatelyand thenum-

ber of pups born was recorded for 100 days. As these mice tended to bepoor breeders, results comprise two (CT3 or B10.D2) or three combined

experiments (TC1 or Cmv1s and BALB.H-2k) and only mice proven to be

fertile were included in the calculation (number of fertile mice per group

given in brackets). Unless otherwise described, all inocula were 2 104 pfu.

(A) BALB.H-2k mice were inoculated with rK181-mZP3 (n = 20) or a con-

trol virus (n = 20). (B) B10.H-2d mice were inoculated with rK181-mZP3

(n =12), 2105 pfu rK181-mZP3 (n = 12), a control virus or were sham-

inoculated with diluent. (C) TC1 or Cmv1s mice were inoculated with

rK181-mZP3 (n = 15), a control virus (n = 18) or were sham-inoculated

(n =10). (D) CT3 (Cmv1r) mice were inoculated with rK181-mZP3 (n = 12),

2 105 pfu rK181-mZP3 (n = 12) or with a control virus (n = 12).

ber of pups born to sham-inoculated mice was 18.2 2.46,

thenumber born to mice inoculated with the control virus was

17.7 1.63 and to mice inoculated with rK181-mZP3 was

14.8 1.65. The female mice inoculated with 2 105 pfu

rK181-mZP3(10ZP3)showed no decrease in fertility, with

a mean number of pups per female of 19.8 4.03 (p = 0.63

KruskalWallis). Combined results for two separate experi-ments are shown.

3.9. B10.D2 mice do not produce ZP3 antibody

To determine whether the lack of effect on fertility in

B10.D2 mice was due to a lack of suitable ZP3-specific

antibody, frozen sections of ovaries harvested from mice

at the end of the breeding experiment were examined for

the presence of antibody binding to follicular zona pellu-

cida. There was no antibody detected on the ovarian sections

from B10.D2 mice regardless of input dose of rK181-mZP3

(results not shown). In addition, testing of the serum from

these mice using the ZP3 ELISA showed no evidence of ZP3antibody (results not shown).

3.10. C57BL/6 mice with the Cmv1s allele become

infertile after inoculation with rK181-mZP3

Mice with the Cvm1s allele exhibited extremely poor

fecundity overall and any mice failing to litter during the

experiment were excluded. This experiment was performed

on three separate occasions and the combined results are pre-

sented (Fig. 6C). A total of 21 female mice were inoculated

with 2 104 pfu rK181-mZP3 of which, 14 were fertile; 21

were inoculated with 2 104

pfu K181, of which 18 were fer-tile, and 12 mice were sham-inoculated with 100l MOBS

diluent (two experiments only), of which 9 were fertile. Mice

inoculated with rK181-mZP3 had a mean number of pups

per fertile female of 6.2 0.99 compared with those inocu-

lated with the control virus, K181 (number of pups per fertile

female 11.5 1.04), and sham-inoculatedmice (12.2 2.55)

a statistically significant result (p = 0.007, KruskalWallis).

The number of pups born to mice inoculated with rK181-

mZP3 was significantly different to those born to mice that

were sham-inoculated (p = 0.02, MannWhitney) or inocu-

lated with a control virus (p = 0.003, MannWhitney).

3.11. BALB/c mice with the Cmv1r allele are not made

infertile by rK181-mZP3

Due to the poor fecundity of CT3 mice, two separate

experiments were performed and the combined results are

presented (Fig. 6D). Any mice that did not produce a litter for

the duration of the experiment were excluded. Twelve female

CT3 mice (BALB.Cmv1r) were inoculated with 2 104

rK181-mZP3 (all were fertile), 12 mice were inoculated with

2 105 pfu rK181-mZP3 (11 were fertile) and 12 were inoc-

ulated with 2 104 pfu K181 control virus (10 were fertile).

CT3 mice inoculated with K181 had 17.3 1.65 pups per


9/12


Fig. 7. Effect of rK181-mZP3 on wild-derived Mus domesticus. Groups of

six female mice were inoculated with rK181-mZP3, a control virus or were

sham-inoculated. Male mice were added immediately post inoculation and

breeding was allowed to proceed for 100 days. Also shown is the combined

data for breeding experiments in wild mice with rK181-mZP3 (number of

fertile female mice = 21).

female. This result was not significantly different from CT3

mice inoculated with rK181-mZP3 (12.8 2.06 pups perfemale, p = 0.12, MannWhitney). There was a reduction in

cumulative pups in the mice inoculated with a 10-fold higher

dose of rK181-mZP3 (6.9 1.08) and this result was sta-

tistically significant over 100 days (p = 0.07 compared with

2 104 pfu rK181-mZP3,p = 0.0001 compared with the con-

trol virus, MannWhitney). The KruskalWallis evaluation

of the combined results (indicating the overall significance

of the experiment) was p = 0.002.

3.12. Outbred wild-derived mice (Mus domesticus/Arc)

are rendered infertile by rK181-mZP3

Virally vectored immunocontraception has been explored

as a potential means to control plaguing populations of

the house mouse ( Mus domesticus) [18]. To this end, mice

from a colony of caesarean-derived wild caught mice were

used to conduct several breeding studies to assess the effect

of rK181-mZP3 on their fertility. Six female mice were

inoculated with rK181-mZP3, RM427+ (control virus) or

were sham-inoculated and were set up as breeding pairs

(Fig. 7). The mean number of pups per fertile female (5)

born to mice inoculated with rK181-mZP3 was 4.8 0.97

compared with the control virus RM427+ (5 were fertile,

19.3 2.43, p = 0.004, MannWhitney) or sham-inoculated

mice (17.2 2.01,p = 0.008MannWhitney), demonstrating

the efficacy of this virus as an immunocontraceptive agent.

To date 31 wild mice have been inoculated with rK181-

mZP3 andobserved for fertility over 100 days. The combined

number of pups per fertile female is presented in Fig. 7(5.4 0.60). This is an underestimation of the infertility

effect of rK181-mZP3 on wild mice as 10 of the 31 mice

failed to have a single litter and were eliminated from this cal-

culation. All females that were infertile for the duration of the

experiment had the original male replaced with one that had

been proven fertile. No litters were recorded indicating that

the infertility was likely to be related to the female mice. The

reduction in fertility in mice inoculated with rK181-mZP3

was highly statistically significant compared with mice inoc-

ulated with a control virus (p < 0.0001, KruskalWallis).

4. Discussion

As shown here, the success of the recombinant MCMV

immunocontraceptive vaccine is affected by the innate

genetic susceptibility of the recipient mouse strain to the viral

vector itself. Both A/J and BALB/c mice are highly suscep-

tible to lethal infection with MCMV [10] and are rendered

infertile after inoculation with 2 104 pfu of recombinant

MCMV (rK181-mZP3). BALB/c mice show a clear dose

response, with the immunocontraceptive effect diminishing

in mice inoculated with decreasing doses of the recombinant

virus. In C57BL/6 and CBA mice there is a scaled response

that reflects the relative LD50 values observed after infectionof these mouse strains with MCMV (Table 1), suggesting

that the innate resistance mechanisms providing protection

from lethal MCMV also contribute to the reduced levels of

infertility which are observed.

Congenic mouse strains were used to determine the effect

of particular innate immunemechanisms on the level of infer-

tility induced by the recombinant virus. CBA mice gave a

relatively poor response to the vaccine, suggesting that the H-

2khaplotype may reduce vaccine efficacy. Although BALB.K

mice (H-2k haplotype from C3H on the BALB/c genetic

Table 1

Summary of breeding experiments after inoculation with rK181-mZP3 compared with known H-2 associated resistance or susceptibility to MCMV

Mouse strain H-2 allele Relative LD50 Fertility trial with rK181-mZP3

2104 pfu 2 105 pfu

BALB/c d 1a Infertile (95%)b NDc

BALB.K k 10 Reduced fertility (52%) ND

C57BL/10 b 34 NDsee C57BL/6 ND

B10.D2 d 45 Fully fertile Fully fertile

CBA k 20 Fully fertile Reduced fertility (40%)

A/J a 0.5 Infertile (93%) ND

C57BL/6 b 34 Reduced fertility (43%) Reduced fertility (66%)

a The LD50 of BALB/c mice is arbitrarily designated as 1 and the relative LD50 of other mouse strains is calculated with respect to this value [11,37].b Number in brackets is % reduction in fertility over 100 days compared with control mice of the same strain.c ND not done.


10/12


background) showed a 52% reduction in litter output over

100 days (p = 0.0002), the immunocontraceptive effect was

clearly less pronounced in these mice compared to BALB/c

mice. Conversely, the fertility of B10.D2 mice (H-2d haplo-

type from DBA/2 on a C57BL10 genetic background) was

not affected by inoculation with either 2 104 or 2105 pfu

rK181-mZP3. Taken together, these experiments suggest thatthe H-2 haplotype of the recipient strain is not itself respon-

sible for vaccine failure. The reduced immunocontraceptive

response in BALB.K mice may reflect differences in the pre-

sentation of ZP3 epitopes to the immune system, rather than

innate resistance to the MCMV vector itself. The complete

resistance of the B10.D2 mice to the immunocontraceptive

virus is surprising given the success of immunocontraception

in the genetically similar C57BL/6 mice, however, there was

no antibody detected from these mice by eitherZP3 ELISA or

examination of frozen sections. The importance of antibody

production in an effective immunocontraceptive response has

previously been noted [8].

Resistance to MCMV in C57BL/6 mice is conferred by aninnate immune mechanism controlled by the Cmv1 genetic

locus via the regulation of NK cells [21]. The availability of

BALB/c and C57BL/6 congenic mouse strains differing only

at the Cmv1 locus allowed us to explore the effect of this

innate immune mechanism without altering the H-2 haplo-

type. CT3 mice have a BALB/c genetic background (H-2d)

with the Cmv1r allele from C57BL/6 [21] and the fertility

of these mice mimics that of the C57BL/6 mice. Conversely,

mice with a C57BL/6 (H-2b) genetic background having the

Cmv1s allele from BALB/c (TC1 or Cmv1s mice) mimicked

the immunocontraception seen in BALB/c mice. These com-

plementary results strongly suggest that Cmv1-associatedinnate immune control of MCMV affects the immunocon-

traceptive success of rK181-mZP3 independently of the H-2

haplotype [28]. A definitive experiment would be to deplete

Cmv1r mice of NK cells prior to infection with the recombi-

nant virus, however, the technical difficulty with maintaining

NK cell depletion in breeding mice over a long term experi-

mentis substantial. We would expect NK cell-depleted Cmv1r

mice to be analogous to Cmv1s mice and be susceptible to

the immunocontraceptive effect of rK181-mZP3.

The control of MCMV via Cmv1 in C57BL/6 mice

involves the activation Ly49H receptors on NK cells that

interact with the virally expressed m157 ligand. Recently,

it has been demonstrated that m157 mutations are present in

many strains of MCMV isolated from wild mice andthe inter-

action with Ly49H provides no additional protection to these

viruses [29]. Additionally, studies investigating the genetic

resistance to MCMV exhibited by the MA/My mouse strain

[30] have identified an NK-cell mediated control mechanism

that utilises the Ly49P receptor and requires the presence of

polymorphic H-2k genes on MCMV infected cells [31,32].

These intriguing studies emphasise the complexity of innate

MCMV defence.

To determine whether immunocontraceptive success

could be attributed to the viral load, quantitative PCR anal-

ysis was carried out on DNA extracted from salivary gland

and lung tissue. BALB/c had a higher (K181) MCMV viral

load, particularly in the salivary gland compared with that

seen in C57BL/6 and CBA mice. However, the recombi-

nant virus was highly attenuated in all three strains and little

variation was detected in the genome load in either organ.

The quantitative difference between the recombinant genomeand the non-recombinant genome detected in salivary glands

suggests that a functional attenuation has occurred in vivo,

although the in vitro growth of the recombinant virus is nor-

mal [8]. One explanation for this result is that the immune

response to mZP3 is responsible for enhanced clearance of

the recombinant virus. Other recombinant viruses generated

in our laboratory have exhibited in vivo attenuation even

when generated by bacterial artificial chromosome technol-

ogy [33].

In animal models of immunocontraception, the magni-

tude of a peripheral antibody response has been used to

evaluate the likelihood of successful contraception [34,35]

although it has been noted that a high peripheral antibodytitre does not necessarily predict mouse infertility [8,36].

Antibody to mZP3 was detected by ZP3 ELISA in serum

from BALB/c, C57BL/6 and CBA mice at 35 days post

inoculation. In ovaries taken from these mice antibody was

bound in vivo to the follicular zona pellucida by 14 days

post inoculation, demonstrating that this is a more sensitive

method of detecting ZP3-specific antibody. When serum was

collected from mice at the end of the breeding studies, com-

paratively little peripheral antibody was detected from the

serum of CBA mice and there was a trend towards higher

titresin mice inoculated with higher doses of the recombinant

virus. This finding together with the lack of antibody detectedin B10.D2 mice suggests that adequate antibody titres are

important in the successful induction of immunocontracep-

tion. Other components such as complement or cellular Fcreceptors may interact with zona pellucida-bound antibody

to enhance the infertility effect. These aspects are currently

being explored. Clearly however, the presence of antibody

bound in situ to ovarian zona pellucida is not related to the

Cmv1 locus, as both C57BL/6 and CBA mice have anti-

body bound to follicular zona pellucida, and the presence of

antibody is not in itself predictive of immunocontraceptive

success.

As the immunocontraceptive virus was developed to con-

trol populations of wild mice [7], the effect of innate resis-

tance to the viral vector exhibited by CBA and C57BL/6 mice

suggests that host resistance may reduce the efficacy of this

technology. Accordingly, a laboratory colony of outbred wild

mice was used to estimate the efficacy of rK181-mZP3. To

date, a total of 31 female mice have been vaccinated with the

recombinant virusand rendered infertile over 100 days.These

mice have polymorphisms in the natural killer cell complex

anddo notexhibitsignificant innateresistance to theviralvec-

tor, MCMV [19]. Interestingly, 16% of mice regained their

fertility after 100 days and proceeded to have several litters

(results not shown), although the time taken to regain fertility


11/12


may be sufficient to stop significant population increases by

compensatory breeding in the field [18].

In conclusion, these results establish the principle that

innate resistance to the viral vector used in a recombinant

vaccine can significantly reduce vaccine efficacy. To ensure

vaccine efficacy, it may be necessary to have an understand-

ing of underlying resistance to the viral vector.

Acknowledgements

Thisresearchwas funded by the Grains ResearchDevelop-

ment Corporation and the Pest Animal Control Cooperative

Research Centre. We would like to thank Simone Ross and

Helen Moulder for their expert assistance in our animal facil-

ity, Dr. Max Bulsara for his help with statistical analysis,

Liz Williams for assisting with the quantitative PCR and

Professor John Papadimitriou and Dr. Lee Smith for their

constructive comments throughout the preparation of this

manuscript.

References

[1] Kurz S, Steffens HP, Mayer A, Harris JR, Reddehase MJ. Latency ver-

sus persistence or intermittent recurrencesevidence for a latent state

of murine cytomegalovirus in the lungs. J Virol 1997;71(4):29807.

[2] Pollock JL, Virgin HW. Latency, without persistence, of murine

cytomegalovirus in the spleen and kidney. J Virol 1995;69(3):17628.

[3] Henry SC, Schmader K, Brown TT, Miller SE, Howell DN, Daley

GG, Hamilton JD. Enhanced green fluorescent protein as a marker for

localizing murine cytomegalovirus in acuteand latent infection. J Virol

Methods 2000;89(12):6173.

[4] Brune W, Hasan M, Krych M, Bubic I, Jonjic S, Koszinowski UH.Secreted virus-encoded proteins reflect murine cytomegalovirus pro-

ductivity in organs. J Infect Dis 2001;184(10):13204.

[5] Lagenaur LA, Manning WC, Vieira J, Martens CL, Mocarski ES.

Structure and function of the murine cytomegalovirus sgg1 gene: a

determinant of viral growth in salivary gland acinar cells. J Virol

1994;68(12):771727.

[6] Latchman DS. Gene delivery and gene therapy with herpes simplex

virus-based vectors. Gene 2001;264(1):19.

[7] Shellam GR. The potential of murine cytomegalovirus as a viral vector

for immunocontraception. Reprod Fertil Dev 1994;6(3):4019.

[8] Lloyd ML, Shellam GR, Papadimitriou JM, Lawson MA. Immuno-

contraception is induced in BALB/c mice inoculated with murine

cytomegalovirus expressing mouse zona pellucida 3. Biol Reprod

2003;68(6):202432.

[9] Karrer U, Wagner M, Sierro S, Oxenius A, Hengel H, Dumrese T, et al.Expansion of protective CD8+ T-cell responses driven by recombinant

cytomegaloviruses. J Virol 2004;78(5):225564.

[10] Grundy JE, Mackenzie JS, Stanley NF. Influence of H-2 and non-H-2

genes on resistanceto murine cytomegalovirus infection. Infect Immun

1981;32(1):27786.

[11] Allan JE, Shellam GR. Genetic control of murine cytomegalovirus

infection: virus titres in resistant and susceptible strains of mice. Arch

Virol 1984;81(12):13950.

[12] Lawson CM, Grundy JE, Shellam GR. Antibody responses to murine

cytomegalovirusin genetically resistantand susceptible strains of mice.

J Gen Virol 1988;69(Pt 8):198798.

[13] Scalzo AA, Fitzgerald NA, Simmons A, La Vista AB, Shellam GR.

Cmv-1, a genetic locus that controls murine cytomegalovirus replica-

tion in the spleen. J Exp Med 1990;171:146983.

[14] Scalzo AA, Fitzgerald NA, Wallace CR, Gibbons AE, Smart YC, Bur-

ton RC, Shellam GR. The effect of the Cmv-1 resistance gene, which

is linked to the natural killer cell gene complex, is mediated by natural

killer cells. J Immunol 1992;149(2):5819.

[15] Brown MG, Scalzo AA, Stone LR, Clark PY, Du Y, Palanca B,

Yokoyama WM. Natural killer gene complex (Nkc) allelic variabil-

ity in inbred mice: evidence for Nkc haplotypes. Immunogenetics

2001;53(7):58491.

[16] Daniels KA,DevoraG, LaiWC, ODonnell CL,BennettM, Welsh RM.

Murine cytomegalovirus is regulated by a discrete subset of natural

killer cells reactive with monoclonal antibody to Ly49H. J Exp Med

2001;194(1):2944.

[17] Lee SH, Girard S, Macina D, Busa M, Zafer A, Belouchi A, et al.

Susceptibility to mouse cytomegalovirus is associated with deletion of

an activatingnaturalkiller cellreceptorof the C-type lectin superfamily.

Nat Genet 2001;28(1):425.

[18] Singleton GR, Farroway LN, Chambers LK, Lawson MA, Smith AL,

Hinds LA. Ecological basis for fertility control in the house mouse

(Mus domesticus) using immunocontraceptive vaccines. Reproduction

2002;(Suppl. 60):319.

[19] Scalzo AA, Manzur M, Forbes CA, Brown MG, Shellam GR. NK

gene complex haplotype variability and host resistance alleles to

murine cytomegalovirusin wildmouse populations.Immunol Cell Biol

2005;83(2):1449.

[20] Poiley SM. A systematic method of breeder rotation for non-inbred

laboratory animal colonies. Proc Anim Care Panel 1960;10(4):15966.

[21] Scalzo AA, Brown MG, Chu DT, Heusel JW, Yokoyama WM, Forbes

CA. Development of intra-natural killer complex (NKC) recombinant

and congenic mouse strains for mapping and functional analysis of NK

cell regulatory loci. Immunogenetics 1999;49(3):23841.

[22] Saederup N, Lin YC, Dairaghi DJ, Schall TJ, Mocarski ES.

Cytomegalovirus-encoded beta chemokine promotes monocyte-

associated viremia in the host. Proc Natl Acad Sci USA 1999;96(19):

108816.

[23] SambrookJ, Fritch E, ManiatisT. Molecular cloning:a laboratory man-

ual. New York: Cold Spring Harbor Laboratory Press; 1989.

[24] Gorman S, Harvey NL, Moro D, Lloyd ML, Voigt V, Smith LM, et al.

Mixed infectionwith multiplestrainsof murine cytomegalovirusoccursfollowing simultaneous or sequential infection of immunocompetent

mice. J Gen Virol 2006;87:112332.

[25] Wheat RL, Clark PY, Brown MG. Quantitative measurement of infec-

tious murine cytomegalovirusgenomesin real-time PCR.J Virol Meth-

ods 2003;112(12):10713.

[26] Karupiah G, Sacks TE, Klinman DM, Fredrickson TN, Hartley JW,

Chen JH, Morse HC. Murine cytomegalovirus infection-induced poly-

clonal b cell activation is independent of cd4(+) t cells and cd40.

Virology 1998;240(1):1226.

[27] Price P, Olver SD, Gibbons AE, Shellam GR. B-cell activation follow-

ing murine cytomegalovirus infection: implications for autoimmunity.

Immunology 1993;78(1):1421.

[28] BubicI, WagnerM, KrmpoticA, SauligT,KimS, Yokoyama WM, etal.

Gain of virulence caused by loss of a gene in murine cytomegalovirus.

J Virol 2004;78(14):753644.[29] Voigt V, Forbes CA, Tonkin JN, Degli-Esposti MA, Smith HR,

Yokoyama WM, Scalzo AA. Murine cytomegalovirus m157 mutation

and variation leads to immune evasion of natural killer cells. Proc Natl

Acad Sci USA 2003;100(23):134838.

[30] Scalzo AA, Lyons PA, Fitzgerald NA, Forbes CA, Yokoyama WM,

Shellam GR. Genetic mapping of Cmv1 in the region of mouse chro-

mosome 6 encoding the NK gene complex-associated loci Ly49 and

musNKR-P1. Genomics 1995;27(3):43541.

[31] Dighe A, Rodriguez M, Sabastian P, Xie X, McVoy M, Brown MG.

Requisite H2k role in NK cell-mediated resistance in acute murine

cytomegalovirus-infected MA/My mice. J Immunol 2005;175:68208.

[32] Desrosiers MP, Kielczewska A, Loredo-Osti JC, AdamSG, Makrigian-

nisAP, Lemieux S, et al.Epistasis between mouseKlraand majorhisto-

compatibility complex class I loci is associated with a new mechanism


12/12


of natural killer cell-mediated innate resistance to cytomegalovirus

infection. Nat Genet 2005;37(6):5939.

[33] Redwood AJ, Messerle M, Harvey NL, Hardy CM, Koszinowski UH,

Lawson MA, Shellam GR. Use of a murine cytomegalovirus k181-

derivedbacterial artificialchromosomeas a vaccinevectorfor Immuno-

contraception. J Virol 2005;79(5):29983008.

[34] Fayrer-Hosken RA, Bertschinger HJ, Kirkpatrick JF, Grobler D, Lam-

berski N, Honneyman G, Ulrich T. Contraceptive potential of the

porcine zona pellucida vaccine in the African elephant (Loxodonta

africana). Theriogenology 1999;52(5):83546.

[35] Garrott RA, Cook JG, Bernoco MM, Kirkpatrick JF, Cadwell LL,

Cherry S, Tiller B. Antibody response of elk immunized with porcine

zona pellucida. J Wildl Dis 1998;34(3):53946.

[36] Hardy CM, Clydesdale G, Mobbs KJ. Development of mouse-

specific contraceptive vaccines: infertility in mice immunized with

peptide and polyepitope antigens. Reproduction 2004;128(4):395

407.

[37] Farrell HE, Shellam GR. Immunoblot analysis of the antibodyresponse

to murine cytomegalovirusin genetically resistantand susceptible mice.

J Gen Virol 1989;70(Pt 10):257386.

zona pellucida antiviral

Documents