This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Van de Walle, G. R.
Right arrow Articles by Osterrieder, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Van de Walle, G. R.
Right arrow Articles by Osterrieder, N.

 Previous Article  |  Next Article 

Journal of Virology, February 2008, p. 1714-1722, Vol. 82, No. 4
0022-538X/08/$08.00+0     doi:10.1128/JVI.02137-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

CCL3 and Viral Chemokine-Binding Protein gG Modulate Pulmonary Inflammation and Virus Replication during Equine Herpesvirus 1 Infection{triangledown}

Gerlinde R. Van de Walle, Kaori Sakamoto, and Nikolaus Osterrieder*

Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, New York 14853

Received 28 September 2007/ Accepted 28 November 2007


arrow
ABSTRACT
 
CCL3 is a proinflammatory chemokine that mediates many of the cellular changes occurring in pulmonary disease. Here, CCL3–/– mice were used to investigate the role of this chemokine during respiratory herpesvirus infection. Compared to wild-type mice, CCL3–/– mice infected with the alphaherpesvirus equine herpesvirus 1 (EHV-1) displayed reduced body weight loss but had higher pulmonary viral loads. Lungs from infected CCL3–/– mice suffered a milder interstitial pneumonia, and fewer immune cells were recovered from the pulmonary airways after infection. We could also demonstrate that herpesvirus-encoded chemokine-binding glycoprotein G (gG) was capable of inhibiting the chemotactic functions of CCL3. This CCL3-mediated chemotaxis, however, was restored in the presence of gG-specific antibodies, which puts into question the advertised use of gG deletion mutants as marker vaccines. In summary, we concluded that CCL3 is a major player in controlling herpesvirus replication in the target organ, the lung, and does so by evoking a strong inflammatory response. The immunomodulatory activity of CCL3 is balanced by the expression of viral gG, whose chemokine-binding activity is mitigated in secondary infections by the production of anti-gG antibodies.


arrow
INTRODUCTION
 
Equine herpesvirus 1 (EHV-1) is taxonomically classified as a herpesvirus, subfamily Alphaherpesvirinae, and is a close relative of the causative agents of cold sores and genital herpes in humans, herpes simplex virus types 1 and 2, respectively. EHV-1 is a major pathogen of horses worldwide, and upon uptake of virus present in infected droplets that are transmitted by direct contact or through aerosols, EHV-1 replicates mainly in the upper respiratory tract, causing respiratory disorders. Subsequently, after entering regional lymph nodes, the virus is transported to the endothelia of inner organs by mononuclear cells. The infection of endothelia and the ensuing inflammatory response lead to hypoxic damage of tissue, ultimately resulting in abortion, neonatal foal death, and nervous system disorders (2). A mouse model of EHV-1-induced disease was established several years ago and has been used extensively to examine EHV-1 infection, virulence, and pathogenesis (3, 28). Intranasal infection of mice with EHV-1 results in an acute, productive infection of the lungs, followed by dissemination of the virus to visceral organs and the brain (6). Besides extensive perivascular and peribronchial cuffing, an interstitial inflammatory infiltrate in the lungs can be observed during EHV-1 infection, consisting mainly of neutrophils, lymphocytes, and macrophages (5, 21, 22). Such infiltration in the bronchoalveolar compartment also occurs in the natural host, the horse, as dynamic leukocyte migration has been observed upon EHV-1 infection (17). Although the mechanism that regulates leukocyte trafficking into the lungs during EHV-1 infection has not been studied in great detail, it seems likely that chemokines and/or cytokines play an important role in the cellular events required for motility and adhesion. Studies using RNase protection analysis of bronchoalveolar lavage (BAL) cells of EHV-1-infected mice revealed an upregulation of several chemokines, including macrophage inflammatory protein 1{alpha} (MIP-1{alpha}), MIP-1β, MIP2, and cytokines like tumor necrosis factor alpha (14, 22). Chemokines consist of a family of cytokines that are 8 to 16 kDa in size and are generated very early in the inflammatory response to infection, resulting in the attraction of leukocytes to local sites of inflammation (4, 19). These proteins are divided into four subclasses based on the arrangement of the N-terminal two cysteines, CC, CXC, C, and CX3C (where X is any amino acid). Each of the numerous chemokines stimulates the migration of a unique set of immune cell types. Selectivity is ensured by restricted expression patterns on recipient leukocytes of chemokine receptors, which are members of the multiple hydrophobic G-protein-coupled receptor superfamily. However, most chemokines recognize several receptors and a single receptor can bind more than one chemokine (18).

CC chemokine ligand 3 (CCL3; or MIP-1{alpha}) is a member of the CC chemokine family and signals through the CC chemokine receptors CCR1 and CCR5. CCL3 can exert chemotactic activities toward a variety of immune cells in vitro, including polymorphonuclear leukocytes, monocytes, lymphocytes, macrophages, and bone marrow-derived dendritic cells (7, 20, 23, 24, 31). In vivo, CCL3 is an important mediator of inflammation upon infection with viruses like Coxsackie virus and influenza virus and has been reported to play an important role in mediating many of the cellular changes that occur in lung disease (10, 23). On the other hand, overproduction of CCL3 has been reported to correlate with protection against HIV-1 infection (32).

Since upregulation of CCL3 had been reported during EHV-1 infection (14, 22), we sought to investigate in more detail the role of CCL3 during EHV-1 infection in vivo. In addition, EHV-1 has recently been reported to encode a protein, glycoprotein G (gG), capable of binding to a broad range of chemokines of human and murine origin, including murine CCL3 (9). Therefore, we also wanted to evaluate the interference of viral chemokine-binding protein (vCKBP) gG with CCL3 during EHV-1 infection. Because reinfections are a very important feature of herpesvirus pathogenesis, we also evaluated the effect of anti-gG-specific antibodies on the proper function of this vCKBP.


arrow
MATERIALS AND METHODS
 
Mice. CCL3–/– knockout breeder mice (B6.129P2-Scya3tm1Unc) were purchased from the Jackson Laboratories (Bar Harbor, ME) and maintained as a colony at the transgenic mouse facility at Cornell University. Several mice of this colony were genotyped to demonstrate the knockout nature by using the PCR protocol provided by the supplier (data not shown). Wild-type C57BL/6J mice were obtained from Taconic Farms (Germantown, NY). Animals were housed in filter-covered isolator cages in the animal facility of the College of Veterinary Medicine at Cornell University (Ithaca, NY), which is accredited by the American Association for Assessment and Accreditation of Laboratory Animal Care. All experimental procedures were approved by the Cornell University Animal Care and Use Committee.

Viruses, cells, and antibodies. EHV-1 strain RacL11, the gG deletion mutant (vL11{Delta}gG), and the gG rescuant virus (vL11{Delta}gGR) were described previously (26). Virus stocks were prepared on rabbit kidney cells (RK13) maintained in minimal essential medium (MEM) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 0.1 mg/ml streptomycin.

Murine neutrophils were obtained by peritoneal lavage, and murine macrophages were obtained from femoral bone marrow cell cultures of wild-type BL/6 mice exactly as described before (8, 15).

Equine immunoglobulin G (IgG) antibodies were derived from convalescent-phase serum obtained from a horse at 14 days after experimental inoculation with 107 PFU/ml RacL11. This hyperimmune serum had an anti-EHV-1 titer of 192, as determined by a standard serum neutralization (SN) test (using a 50% endpoint of serum dilutions). As a negative control, serum from a horse with an anti-EHV-1 SN titer of 4 was used.

Chemotaxis assays. For chemotaxis assays, 12-well Costar Transwell plates were used (Corning Costar Co., Cambridge, MA) as described before (25). Recombinant murine CCL3 was obtained from R&D Systems (Minneapolis, MN) and used at a concentration of 10 ng/ml. Briefly, recombinant CCL3 was preincubated for 30 min at 37°C with 0.3 µg/ml recombinantly expressed EHV-1 gG (25) and applied to the lower chambers of the wells. Wells were covered with a polycarbonate membrane with a 3-µm pore size for murine neutrophils and a 5-µm pore size for murine macrophages. A 100-µl cell suspension containing 1 x 104 cells was added to the top chamber, and assay plates were incubated at 37°C for 45 min for neutrophils or 2 h for macrophages. Recombinant CCL3 and medium alone were included as positive and negative controls, respectively. In some experiments, recombinantly expressed gG was preincubated with different dilutions of EHV-1-hyperimmune serum (titer, 192) or negative control serum (titer, 4). The cells that migrated through the filter into the wells were stained, counted under a light microscope (Zeiss Axiovert 25), and expressed as percent chemotaxis based on the number of input cells.

Infection protocols. Animal experiments were conducted as described previously, with some modifications (27). In a first set of experiments, 6- to 10-week-old, female wild-type or CCL3–/– knockout mice (10 per group) were infected by the intranasal route with 1 x 104 PFU of EHV-1 strain RacL11. Virus was suspended in 20 µl of Dulbecco's modified Eagle's medium. Control mice (six per group) treated with Dulbecco's modified Eagle's medium alone were also included. Individual weights of mice were determined on the day of infection (day 0) until day 14 postinfection (p.i.). Two mice in each group were euthanized to collect the lungs at 2 and 4 days p.i., which were homogenized to determine virus titers by standard titration on RK13 cells. Due to the limited availability of transgenic mice of comparable age, an identical but separate experiment was conducted to evaluate pulmonary airway inflammation at days 1, 2, and 4 p.i. (described below).

In a second animal experiments, 6- to 10-week-old female wild-type or CCL3–/– knockout mice were infected as described above with 1 x 104 PFU but this time with the mutant vL11{Delta}gG, which is unable to express viral chemokine-binding gG. The rescuant virus, vL11{Delta}gGR, was used as a positive control, and mock-infected animals served as negative controls. Blood was collected at days 0 and 14 p.i. and pooled according to group, and the serum was used to determine EHV-1 SN and enzyme-linked immunosorbent assay (ELISA) titers. Following primary infection, the remaining animals were reinfected 4 weeks later with EHV-1 strain RacL11 at a dose of 1 x 105 PFU/mouse. Sampling was performed exactly as described above.

Analyses of airway inflammation. Pulmonary airway inflammation in CCL3–/– and wild-type mice was evaluated at days 1, 2, and 4 p.i. Infection was performed with 1 x 104 PFU/ml as described above, and BAL was performed with 1 ml of saline on at least two mice per group as described before (22, 25. Total cell counts of the BAL fluid samples were determined, and differential cell counts were performed exactly as described before (25). After BAL, lungs were removed, fixed in 4% paraformaldehyde, embedded in paraffin, and cut into 4-µm sections. Four hematoxylin-and-eosin (H&E)-stained lung sections from central and peripheral parts of different lung lobes were scored in a double-blinded fashion under a light microscope to determine the degree of inflammation in the interstitium, alveoli, and surrounding airways and vessels. A value of 0 (normal), 1 (minimal), 2 (mild), 3 (moderate), 4 (marked), or 5 (severe) was assigned to each histological site, and the scores were totaled for each animal. The average of scores from two animals per group was computed as the total lung inflammation score.

ELISA. Wells of a microtiter plate (Nunc, Rochester, NY) were coated with recombinant EHV-1 gG (5 µg/ml in coating buffer consisting of 15 mmol Na2CO3 and 35 mmol NaHCO3, pH 9.6) overnight at 4°C. Wells were blocked with 3% skim milk powder in phosphate-buffered saline (PBS) for 2 h at room temperature and subsequently washed with PBS (2.5 mmol NaH2PO4, 7.5 mmol Na2HPO4, 145 mmol NaCl, pH 7.2) containing 0.05% (vol/vol) Tween 20. To detect the presence of anti-gG-specific antibodies in mouse serum, a dilution series of serum was added to the wells, which were incubated for 2 h at 37°C, followed by peroxidase-conjugated anti-mouse IgG (diluted 1/5,000 in 0.3% skim milk powder in PBS; Jackson Laboratories, Bar Harbor, ME) for 1 h at 37°C. To determine the gG-specific antibody titer in horse serum, a dilution series of serum was added to the wells and they were incubated for 2 h at 37°C; anti-EHV-1 gG-specific rabbit serum (a kind gift from C. Hartley) (16) was added, and the wells were incubated for 1 h at 37°C; and then peroxidase-conjugated anti-rabbit IgG (diluted 1/2,000; Amersham Biosciences) was added, and the wells were incubated for 1 h at 37°C. Binding of peroxidase-conjugated antibodies was colorimetrically determined at 450 nm with a microplate reader (Bio-Tek, Richmond, VA) with BD OptEIA reagents A and B (BD Biosciences, San Jose, CA) as a substrate. In general, wells were washed three times after coating and blocking and nine times elsewhere.

Statistical analyses. Student's t test for paired data was used to test for significant differences. Data given are the means, and bars show standard deviations. Body weights were compared by using nonparametric Wilcoxon Mann-Whitney and Kruskal-Wallis tests. All statistical calculations were performed with SAS version 9.1. (SAS Corporation, Cary, NC).


arrow
RESULTS
 
CCL3 controls viral replication but contributes to EHV-1-induced pulmonary inflammation. Because upregulation of CCL3 in mice during EHV-1 infection had been observed previously (14, 21), we decided to study the role of the chemokine during EHV-1 infection in more detail by using mice deficient in CCL3. CCL3–/– knockout mice were generated in a BL/6 background by deletion of half of the CCL3 coding region and were shown not to have overt hematopoietic abnormalities (10). CCL3–/– knockout or wild-type BL/6 mice were infected with 1 x 104 PFU of EHV-1 strain RacL11 and monitored for 14 days. CCL3–/– mice lost weight at a significantly reduced rate compared to wild-type BL/6 mice, especially at early time points of infection (Fig. 1A). In contrast, virus titers recovered from the lungs of infected animals were significantly higher in CCL3–/– mice compared to BL/6 mice at day 2 p.i. (Fig. 1B, P < 0.05). By day 4 p.i., viral titers were drastically decreased, which indicated efficient clearance of the virus, and no significant difference between wild-type and knockout mice was evident (Fig. 1B).


Figure 1
View larger version (32K):
[in this window]
[in a new window]

  		FIG. 1. Infection of wild-type BL/6 or CCL3–/– mice with EHV-1. (A) Development of mean body weights after infection. CCL3–/– (closed symbols) or wild-type (WT) BL/6 (open symbols) mice (groups of 10) were treated intranasally with MEM (circles) or with 104 PFU of RacL11 (squares). Mean body weights were determined on the day of infection (day 0) up to day 14 p.i. Mean body weights and standard deviations are shown. Asterisks indicate statistically significant differences (P < 0.05) between inoculated wild-type and CCL3–/– mice. (B) Virus titers in lungs. Viral titers were determined in two mice of each group on days 2 and 4 p.i. Mean titers in lungs and standard deviations are shown. The limit of detection was 1 x 101 PFU/organ, and <1 indicates that no virus was recovered from the lungs. Asterisks indicate statistically significant differences (P < 0.05) between wild-type (white bars) and CCL3–/– (black bars) mice.

Lungs of infected mice were also examined histologically. As evidenced by H&E staining, lungs of wild-type mice exhibited interstitial pneumonia, vasculitis, and bronchiolitis during the early stages of infection, at days 1, 2, and 4 p.i. (Fig. 2B). Lungs of infected CCL3–/– mice also showed interstitial pneumonia, which was generally milder. Due to the high variability of lung sections within one and the same animal and between individual animals, the histological differences did not reach statistical significance (Fig. 2B). In contrast to wild-type mice, however, bronchiolitis was never recorded in infected CCL3–/– mice. In order to quantitatively evaluate inflammatory cells infiltrating the airways, BALs were performed on lungs of mice prior to sampling for histology. As shown in Fig. 2A, the total amount of BAL cells differed significantly between CCL3–/– and wild-type mice. Compared to the wild-type mice, fewer immune cells could be recovered from lungs of infected CCL3–/– mice at days 1, 2, and 4 p.i. (P < 0.05). No difference was observed between mock-infected mice of the two strains, indicating that CCL3–/– mice do not differ from wild-type mice in the number of resident immune cells in normal lungs (Fig. 3A). Analysis of the different cell types in the pulmonary airways of infected mice revealed that significantly more neutrophils were present in the lungs of infected wild-type compared to CCL3–/– mice at day 1 p.i. (Fig. 3B, P < 0.05). A significant difference in the amount of macrophages present in the lungs of the two mouse strains was only apparent at day 4 p.i. (Fig. 3C, P < 0.05), whereas at all of the time points tested, significantly more lymphocytes were present in the airways of infected wild-type mice (Fig. 3D, P < 0.05). Similar to the total number of cells, no difference was observed in the individual cell types between uninfected animals of the two mouse strains, indicating no strain-specific preference for a certain immune cell type (Fig. 3B to D).


Figure 2
View larger version (67K):
[in this window]
[in a new window]

  		FIG. 2. Histological analysis of lung sections. (A) A total lung inflammation score was determined for two mice in each group (four lung sections per mouse) and graded on a scale of 0+ (normal) to 5+ (severe). Lung sections of wild-type infected (white bars) or CCL3–/– inoculated mice (light gray bars) were analyzed on days 1, 2, and 4 p.i. (B) H&E-stained images showing histological features in the lungs of wild-type (WT) and CCL3–/– infected mice at days 1, 2, and 4 p.i. Bars, 100 µm. IP, interstitial pneumonia (m, mild); PE, perivascular edema; V, vasculitis; B, bronchiolitis.


Figure 3
View larger version (45K):
[in this window]
[in a new window]

  		FIG. 3. Quantitation of the influx of immune cells into the airways of infected wild-type (WT) BL/6 or CCL3–/– mice. Total numbers of immune cells from wild-type BL/6 (white bars) or CCL3–/– mice (gray bars) infected intranasally with RacL11 (striped bars) or mock infected (open bars). At different days p.i., inflammatory cells were harvested by BAL, washed, and counted (A). Absolute numbers of neutrophils (B), macrophages (C), and lymphocytes (D) in BAL cells are shown. Data are presented as the mean ± standard deviation, and asterisks indicate statistically significant differences (P < 0.05).

Viral chemokine-binding gG of EHV-1 interferes with CCL3 in vitro and in vivo. Recently, EHV-1 gG was shown to function as a vCKBP that is capable of interfering with neutrophil migration both in vitro and in vivo (25). As EHV-1 gG has been reported to bind to a broad range of chemokines, including murine CCL3 (9), we evaluated the effect of gG on CCL3 chemotaxis during EHV-1 pathogenesis. When 10 ng/ml recombinant CCL3 was used in a chemotaxis assay (25), approximately 20% of the murine neutrophils and 25% of the murine macrophages migrated along the cytokine gradient (Fig. 4A). However, when CCL3 was preincubated with recombinantly expressed gG, a significant reduction of cell migration could be observed in neutrophils (P < 0.01) and macrophages (P < 0.05), indicating that EHV-1 gG can indeed interfere with the proper function of CCL3 in vitro (Fig. 4A).


Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 4. (A) gG of EHV-1 inhibits CCL3-induced chemotaxis of murine immune cells in vitro. Chemotaxis assays with murine neutrophils and macrophages were performed with 10 ng/ml murine CCL3 preincubated with (gray bars) or without (white bars) 0.3 µg/ml recombinant gG. Data are expressed as the mean ± standard deviation of at least three independent experiments. Asterisks indicate statistically significant differences (*, P < 0.01; **, P < 0.05). (B) In vivo relevance of gG to pathogenesis in a wild-type BL/6 murine infection model. Wild-type BL/6 mice (groups of 10) were infected intranasally with 1 x 104 PFU of vL11{Delta}gGR (white bars) or vL11{Delta}gG (black bars). Mean body weights and titers in lungs at day 2 p.i., together with their standard deviations, are shown. Asterisks indicate statistically significant differences (P < 0.05) between vL11{Delta}gGR- and vL11{Delta}gG-inoculated BL/6 mice. (C) Histological analysis of lung sections. A total lung inflammation score was determined for two mice (four lung sections per mouse) in each group and graded on a scale of 0+ (normal) to 5+ (severe). Lung sections of mice infected with vL11{Delta}gGR (white bars) or vL11{Delta}gG (light gray bars) were analyzed on days 1, 2, and 4 p.i.

To assess the interference of gG with CCL3 in vivo, we repeated the murine infection study with an EHV-1 mutant, vL11{Delta}gG, in which the gene encoding gG is deleted. The rescuant virus, vL11{Delta}gGR, in which the expression of gG is restored, was used as a positive control (26). Throughout the whole observation period, wild-type BL/6 mice infected with the rescuant virus vL11{Delta}gGR lost more weight (statistically significant, P < 0.05, on day 2 p.i., Fig. 4B) and recovered less rapidly compared to those infected with the vL11{Delta}gG deletion mutant (data not shown). Also, virus titers were determined in the lungs of these BL/6-infected mice and the gG deletion mutant replicated to lower titers than the rescuant virus, indicating an impairment of virus replication in this mouse strain in the absence of gG (Fig. 4B, P < 0.05). Histopathological analysis of the lungs revealed modest differences between the two groups. However, due to the high variation between different sections, no unequivocal conclusions could be drawn (Fig. 4C). When evaluating the BAL cells in vL11{Delta}gG-infected versus vL11{Delta}gGR-infected BL/6 mice, significantly increased numbers of immune cells migrated into the airways in the absence of viral gG (Table 1). At day 1 p.i., this difference was mainly due to a nearly twofold increase in the number of macrophages in the lungs of vL11{Delta}gG-infected mice, whereas at day 2 p.i. neutrophils were primarily responsible for the increase in cell numbers in BAL fluid samples (Table 1). No apparent difference in lymphocyte counts could be observed between vL11{Delta}gG-infected and vL11{Delta}gGR-infected BL/6 mice, with the notable exception of day 4 p.i., when a 1.6-fold increase in lymphocytes was evident in the airways of vL11{Delta}gG-infected mice (Table 1). The statistically significant differences (P < 0.05) between the two groups that were evident in several types of immune cells on different days p.i. confirm that viral gG can modulate the migration of immune cells in the target organ, the lung, during EHV-1 infection. This interference is very finely regulated, and effects on different immune cell subsets appear to be dependent on the time point after infection.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Total and differential cell counts of leukocytes recovered by BAL after infection of mice with EHV-1 mutants

To determine whether gG was able to specifically interfere with the chemotactic functions of CCL3, the infection experiment was repeated with CCL3–/– mice instead of wild-type BL/6 mice. Interestingly, in CCL3–/– mice, no difference could be observed in the influx of immune cells into the lungs of mice infected with the vL11{Delta}gG versus the vL11{Delta}gGR virus (Table 1).

Antibodies against EHV-1 gG interfere with the function of gG as a vCKBP. It has been reported that gG is highly immunogenic, as evidenced by a strong gG-specific antibody response during EHV-1 infection of the natural host (11, 12). In order to evaluate the effect of anti-gG-specific antibodies on the proper function of gG as a vCKBP, we first repeated the in vitro chemotaxis assay with murine CCL3 and gG but in the presence of an EHV-1-specific equine hyperimmune serum (with an SN titer of 192). First, an inhibition ELISA showed the presence of gG-specific antibodies in this hyperimmune serum (Fig. 5A). Preincubation of recombinant gG with hyperimmune serum at a 20% concentration completely abolished the gG-mediated inhibition of CCL3-induced chemotaxis of murine macrophages (Fig. 4B). This inhibitory effect could be observed with a final concentration of horse hyperimmune serum as low as 2% (Fig. 5B). In contrast, the negative control serum was not able to interfere with the inhibitory effect of gG on macrophage migration (Fig. 5B) and the sera by themselves did not have any effect on macrophage migration either in the absence or in the presence of CCL3 (data not shown).


Figure 5
View larger version (22K):
[in this window]
[in a new window]

  		FIG. 5. (A) Presence of gG-specific antibodies in EHV-1-hyperimmune serum. The inhibition of binding to coated gG of polyclonal rabbit anti-gG antibodies in EHV-1-hyperimmune horse serum with an SN titer of 192 ({circ}) or preimmune serum ({square}) was determined by an inhibition ELISA as described in Materials and Methods. (B) Antibodies against gG interfere with the proper function of gG as a vCKBP in vitro. Chemotaxis assays with murine macrophages were performed with 10 ng/ml murine CCL3 preincubated with 0.3 µg/ml recombinant gG in the presence or absence of 20% EHV-1-hyperimmune horse serum or preimmune serum (pre serum). Data are expressed as the mean ± standard deviation of at least three independent experiments. Asterisks indicate no statistically significant differences in CCL-3-induced chemotaxis (P > 0.1).

The effect of anti-gG-specific antibodies was also investigated during EHV-1 pathogenesis in vivo. BL/6 mice were infected with the gG deletion mutant vL11{Delta}gG or the revertant virus vL11{Delta}gGR and reinfected with wild-type RacL11 4 weeks later. In both groups of infected animals, an antibody response was raised against EHV-1, as determined by an SN test at 14 days p.i. (Table 2). In addition, an ELISA was performed to detect the presence of anti-gG-specific antibodies. As expected, no anti-gG-specific antibodies could be detected in serum from mice infected with the vL11{Delta}gG mutant. In contrast, sera obtained from animals infected with the revertant virus showed antibody binding to coated gG (Table 2). After reinfection with RacL11, mice were monitored daily for clinical signs of infection and body weight loss to investigate the pathogenesis of EHV-1 infection in the presence of an immune response in general. Mice that had initially been mock infected and then reinfected with RacL11 were used as positive controls, and mice that were inoculated with medium only were used as negative controls. Mice that had initially been mock infected or infected with the gG-negative vL11{Delta}gG mutant started to lose weight as early as day 1 p.i. after reinfection with RacL11. The weight loss continued until day 3 p.i., after which the mice started to gain back weight (Fig. 6A). Mice that had been infected 4 weeks earlier with vL11{Delta}gG overall lost less weight after exposure to wild-type RacL11 virus, compared to primary infection (Fig. 6A). Mice previously infected with vL11{Delta}gGR, however, did not lose any weight at all upon reinfection and were indistinguishable from mock-infected animals, indicating efficient protection against weight loss in the presence of preexisting gG-specific antibodies (Fig. 6A). Also, virus titers were determined in the lungs of infected mice on days 2 and 4 p.i. At day 2 p.i., viral titers in the lungs of primary infected mice were statistically significantly higher than in those of reinfected mice (P < 0.05), with no difference in viral titers between the vL11{Delta}gG- and vL11{Delta}gGR-infected groups (P = 0.5) (Fig. 6B). These results indicate that an immune response established after primary infection is protective against viral replication but independent of the presence of gG-specific antibodies. By day 4 p.i., virus was no longer detectable in the lungs of immune animals, in contrast to the situation after primary infection (Fig. 6B). Histopathological analysis of the lungs showed interstitial pneumonia, lymphocytic vasculitis, and peribronchiolitis in the infected animals, with an average score that was slightly but not significantly higher in the nonimmune animals on days 1 and 4 p.i. (Fig. 6C).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Antibodies against EHV-1 in mouse sera


Figure 6
View larger version (28K):
[in this window]
[in a new window]

  		FIG. 6. Infection and reinfection of wild-type BL/6 mice with EHV-1. (A) Development of mean body weights. Six-week-old BL/6 mice were infected with 1 x 104 PFU of vL11{Delta}gG ({triangleup}) or vL11{Delta}gGR ({diamond}). One group was initially mock infected (x). Four weeks later, mice were reinfected with RacL11 at a dose of 1 x 105 PFU. Mock-infected animals of the same age were used as negative controls ({circ}). Mean body weights were determined on the day of reinfection (day 0) up to day 5 after reinfection. Mean body weights and standard deviations are shown. Asterisks indicate statistically significant differences between the different groups of mice. (B) Virus titers in lungs. Viral titers were determined in two mice of each group on days 2 and 4 after reinfection. Titers in lungs and standard deviations are shown. The limit of detection was 1 x 101 PFU/organ, and <1 indicates that no virus could be recovered from the lungs. Asterisks indicate statistically significant differences (P < 0.05) between primary (white bars) and vL11{Delta}gGR (black bars)- or vL11{Delta}gG (gray bars)-inoculated mice after reinfection with RacL11. (C) Histological analysis of lung sections. A total lung inflammation score was determined for two mice (four lung sections per mouse) in each group and graded on a scale of 0+ (normal) to 5+ (severe). Lung sections of naive mice (white bars) or mice that had previously been infected with vL11{Delta}gGR (dark gray bars) or vL11{Delta}gG (light gray bars) were analyzed on days 1, 2, and 4 p.i.

Finally, the proper function of gG as a vCKBP in the presence of gG-specific antibodies was evaluated by analyzing BAL cells of mice from the different groups. As shown in Fig. 7, the total number of BAL cells and the subtypes of immune cells did not differ significantly between primary infection and after immunization with vL11{Delta}gG, with the exception of lymphocyte infiltration at day 2 p.i. (Fig. 7D, P < 0.05). In contrast, more immune cells could be recovered from the lungs of mice immunized with the vL11{Delta}gGR mutant (Fig. 7A). This increase was accounted for by neutrophils, macrophages, and lymphocytes, especially at early times p.i. (Fig. 7C and D). These data indicated that the function of gG as a vCKBP is greatly diminished in the presence of an anti-gG-specific immune response, likely mediated mainly by antibodies, as significantly more immune cells homed to the lungs after infection when antibodies against gG were present.


Figure 7
View larger version (32K):
[in this window]
[in a new window]

  		FIG. 7. Quantitation of the influx of immune cells into the airways of wild-type BL/6 mice after reinfection. Total numbers of immune cells from wild-type BL/6 mice that were infected with 1 x 104 PFU of vL11{Delta}gGR (black bars) or vL11{Delta}gG (gray bars) or mock infected (white bars) and then reinfected with virulent strain RacL11 4 weeks later. At different days after reinfection, inflammatory cells were harvested by BAL, washed, and counted (A). Absolute numbers of neutrophils (B), macrophages (C), and lymphocytes (D) in BAL cells are shown. Data are presented as the mean ± standard deviation, and asterisks indicate statistically significant differences (P < 0.05).


arrow
DISCUSSION
 
In this study, we used a murine infection model with conventional BL/6 and CCL3–/– mice to investigate the contributions of the CC chemokine CCL3 during herpesvirus infection of the respiratory tract. The alphaherpesvirus EHV-1 was used, as upregulation of CCL3 expression in the lungs of EHV-1-infected mice had been previously reported (14, 22). In addition, we evaluated the chemokine-binding activities of herpesvirus gG and its functional relationship to CCL3 in virus-induced pathogenesis.

The infection experiments were performed in a BL/6 genetic background as this mouse strain is typically used to derive chemokine-deficient mice. It has been reported that EHV-1 infection in the C57BL/6J mouse strain can result in inconsistent respiratory and systemic symptoms (30). In our study, however, all wild-type BL/6 mice showed clinical signs typical of EHV-1 infection, such as weight loss, ruffled fur, and hunched posture, and the signs were similar to those observed in the well-established BALB/c mouse model (25, 26). Infection of BL/6 mice with 1 x 104 PFU of RacL11 resulted in weight loss starting as early as 1 day p.i., with a maximum weight loss of approximately 15% at around days 2 and 3 p.i. The BL/6 mice in this study, however, started regaining weight by day 4 p.i., which is slightly faster than observed in BALB/c mice, where infection with a similar dose usually results in a more protracted recovery and preinfection weights are not reached until about day 8 p.i. (25, 26). When comparing pulmonary viral loads, virus could be detected up to day 4 p.i. in the lungs of BL/6 mice, with a maximum replication titer of around 2 x 103 PFU/mg of lung tissue at day 2 p.i. These results are similar to what is seen in the BALB/c model (25). Taken together, the results show that although there are differences between the two strains in the time needed for recovery from weight loss, the general profiles upon infection with wild-type EHV-1 are comparable between BALB/c and BL/6 mice, and we concluded that BL/6 mice are useful for studying EHV-1 infection.

Next, BL/6 mice devoid of CCL3 (CCL3–/–) were used to evaluate the role of CCL3 during EHV-1 infection. We could demonstrate that CCL3 contributes to EHV-1-induced pneumonia and that it appears to be an important chemokine mediating migration of immune cells into the lungs early in infection. The primary immune reaction in murine lungs upon EHV-1 infection has been shown to include migration of innate immune effector cells, neutrophils and macrophages, which mainly serve to clear debris deposited as a consequence of infection and consequent tissue damage (5). The appearance of lymphocytes, on the other hand, has been linked to efficient elimination of virus and virus-infected cells (5). Here we observed significantly reduced numbers of lymphocytes in the airways of CCL3–/– mice during the early stages of EHV-1 infection, which might explain why virus titers were significantly higher in the absence of CCL3 in CCL3–/– mice compared to wild-type BL/6 mice.

Recently, alphaherpesvirus-encoded vCKBP gG was shown to be a negative virulence factor by regulating neutrophil trafficking in the lungs after intranasal EHV-1 infection of BALB/c mice (25, 26). As gG has also been reported to bind to murine CCL3 (9), we aimed to study in more detail the effect of gG on this chemokine. Similar to earlier observations (25), more leukocytes migrated into the airways of the lungs of wild-type BL/6-infected mice in the absence of gG. This difference in leukocyte migration between viruses expressing or not expressing the vCKBP, however, was no longer observed when CCL3–/– mice were infected, clearly suggesting that gG is able to interfere with CCL3 in vivo. Despite these convincing results, we cannot state that gG interference with CCL3 is solely responsible for the observed effects and modulation of pathogenesis and inflammation, as this vCKBP is known to bind a broad range of chemokines and has recently been shown to interfere in vitro with the function of murine KC, a potent chemoattractor of neutrophils (25).

During these animal experiments, however, we made an unexpected observation. The gG deletion mutant showed reduced virulence in wild-type BL/6 mice, which is in striking contrast to what has been observed before with the well-established BALB/c mouse model, where infection with the gG deletion mutant resulted in significantly more weight loss and higher viral titers (25, 26). As mentioned before, the BALB/c mouse model is most commonly used as other mouse strains were found to be less susceptible to EHV-1 infection (3, 29). However, several studies have shown that the specific murine immune system is very important for the outcome of infection with EHV-1. In a study by Alber and coworkers, the immune responses of BALB/c (H-2d) and C3H (H-2k) mice to EHV-1 infection were compared, and while the disease patterns were similar in the two mouse types, the immune responses generated were quite different. Complement-independent virus-neutralizing antibodies could be detected after a single exposure to EHV-1 in CH3 mice, and this mouse strain also generated higher T-cell proliferative responses than BALB/c mice (1). Our study confirms that the genetic background of the specific mouse strain used can influence the outcome and progression of EHV-1 infection and highlight the importance of choosing the most appropriate mouse strain. The choice of mouse strain largely depends on the specific question addressed and might imply the usefulness of using more than one mouse strain when specific interactions of EHV-1, and alphaherpesviruses in general, with the immune system are to be studied. Although the murine respiratory model of EHV-1 infection is not perfect, it still provides a valid and suitable laboratory animal model for investigating the clinical outcome and virological aspects of EHV-1-induced disease, especially in terms of more rationally based vaccine strategies in the horse and prediction of how immunomodulatory molecules might influence residual virulence and vaccine efficacy. However, care should be taken when extrapolating immunological parameters from the mouse to the natural target host, as is always the case. With respect to gG, it is important to evaluate which effects an EHV-1 gG deletion mutant will have compared to a wild-type virus in the natural host, the horse, and experiments have been initiated to investigate the behavior of such mutants in horses.

In a final part of this study, we reinfected mice previously infected with the vL11{Delta}gG deletion virus or the vL11{Delta}gGR rescuant virus to evaluate the effect of antibodies against gG and whether such antibodies have an impact on its ability to properly function as a vCKBP. First, we could observe with our murine model that previously infected immune mice were more resistant to infection with a high dose of EHV-1, as shown by a reduction of viral replication in the lungs of immune animals. These results are in agreement with previous studies with mice showing that secondary EHV-1 infections are less severe than a primary encounter with the pathogen (13). Secondly, the presence of gG-specific antibodies had a protective effect with respect to clinical signs (e.g., weight loss), which was not observed in immune animals vaccinated with the gG-negative virus and therefore lacking a gG-specific response. As it was observed that more immune cells could be recovered from the lungs of mice immunized with the revertant vL11{Delta}gGR virus, we attributed this protective effect to the elimination of gG's function as a vCKBP by anti-gG antibodies. This finding is of considerable importance, because EHV-1 gG deletion mutants are widely advocated as useful modified-live virus marker vaccines that allow distinction between vaccinated and infected horses (16). Based on our results, however, such an approach seems suboptimal since antibodies against gG control the vCKBP activity of gG and therefore might be important in preventing EHV-1 from evading the immune system. Our results support the view that vaccines should stimulate local antigen-specific immune defense mechanisms in order to protect against primary infections and should promote a protective response against viral proteins with immunomodulatory functions. Accordingly, thorough testing of engineered vaccines with respect to vaccine safety and efficacy in the natural host seems to become all the more important.


arrow
ACKNOWLEDGMENTS
 
We are grateful to Kathy Mott for excellent help in breeding the CCL3–/– mice. We thank Carol Hartley for supplying anti-EHV-1 gG antibodies. We also thank Maeva M. May and Gillian A. Perkins for help with sample collection.

This work was supported by the Harry M. Zweig Fund for Equine Research and an unrestricted grant from Pfizer Animal Health to N.O.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853. Phone: (607) 253-4045. Fax: (607) 253-3384. E-mail: no34{at}cornell.edu Back

{triangledown} Published ahead of print on 12 December 2007. Back


arrow
REFERENCES
 
    1
  1. Alber, D. G., J. Greensill, R. A. Killington, and A. Stokes. 1995. Role of T-cells, virus neutralising antibodies and complement-mediated antibody lysis in the immune response against equine herpesvirus type-1 (EHV-1) infection of C3H (H-2k) and BALB/c (H-2d) mice. Res. Vet. Sci. 59:205-213.[CrossRef][Medline]
    2. 2
  2. Allen, G. P., and J. T. Bryans. 1986. Molecular epizootiology, pathogenesis, and prophylaxis of equine herpesvirus-1 infections. Prog. Vet. Microbiol. Immunol. 2:78-144.[Medline]
    3. 3
  3. Awan, A. R., Y. C. Chong, and H. J. Field. 1990. The pathogenesis of equine herpesvirus type 1 in the mouse: a new model for studying host responses to the infection. J. Gen. Virol. 71(Pt. 5):1131-1140.[Abstract/Free Full Text]
    4. 4
  4. Baggiolini, M. 1998. Chemokines and leukocyte traffic. Nature 392:565-568.[CrossRef][Medline]
    5. 5
  5. Bartels, T., F. Steinbach, G. Hahn, H. Ludwig, and K. Borchers. 1998. In situ study on the pathogenesis and immune reaction of equine herpesvirus type 1 (EHV-1) infections in mice. Immunology 93:329-334.[CrossRef][Medline]
    6. 6
  6. Baxi, M. K., K. Borchers, T. Bartels, A. Schellenbach, S. Baxi, and H. J. Field. 1996. Molecular studies of the acute infection, latency and reactivation of equine herpesvirus-1 (EHV-1) in the mouse model. Virus Res. 40:33-45.[CrossRef][Medline]
    7. 7
  7. Bennouna, S., S. K. Bliss, T. J. Curiel, and E. Y. Denkers. 2003. Cross-talk in the innate immune system: neutrophils instruct recruitment and activation of dendritic cells during microbial infection. J. Immunol. 171:6052-6058.[Abstract/Free Full Text]
    8. 8
  8. Bennouna, S., W. Sukhumavasi, and E. Y. Denkers. 2006. Toxoplasma gondii inhibits Toll-like receptor 4 ligand-induced mobilization of intracellular tumor necrosis factor alpha to the surface of mouse peritoneal neutrophils. Infect. Immun. 74:4274-4281.[Abstract/Free Full Text]
    9. 9
  9. Bryant, N. A., N. Davis-Poynter, A. Vanderplasschen, and A. Alcami. 2003. Glycoprotein G isoforms from some alphaherpesviruses function as broad-spectrum chemokine binding proteins. EMBO J. 22:833-846.[CrossRef][Medline]
    10. 10
  10. Cook, D. N., M. A. Beck, T. M. Coffman, S. L. Kirby, J. F. Sheridan, I. B. Pragnell, and O. Smithies. 1995. Requirement of MIP-1{alpha} for an inflammatory response to viral infection. Science 269:1583-1585.[Abstract/Free Full Text]
    11. 11
  11. Crabb, B. S., C. M. MacPherson, G. H. Reubel, G. F. Browning, M. J. Studdert, and H. E. Drummer. 1995. A type-specific serological test to distinguish antibodies to equine herpesviruses 4 and 1. Arch. Virol. 140:245-258.[CrossRef][Medline]
    12. 12
  12. Crabb, B. S., and M. J. Studdert. 1993. Epitopes of glycoprotein G of equine herpesviruses 4 and 1 located near the C termini elicit type-specific antibody responses in the natural host. J. Virol. 67:6332-6338.[Abstract/Free Full Text]
    13. 13
  13. Field, H. J., A. R. Awan, and R. de la Fuente. 1992. Reinfection and reactivation of equine herpesvirus-1 in the mouse. Arch. Virol. 123:409-419.[CrossRef][Medline]
    14. 14
  14. Frampton, A. R., Jr., P. M. Smith, Y. Zhang, T. Matsumura, N. Osterrieder, and D. J. O'Callaghan. 2002. Contribution of gene products encoded within the unique short segment of equine herpesvirus 1 to virulence in a murine model. Virus Res. 90:287-301.[CrossRef][Medline]
    15. 15
  15. Geisel, R. E., K. Sakamoto, D. G. Russell, and E. R. Rhoades. 2005. In vivo activity of released cell wall lipids of Mycobacterium bovis bacillus Calmette-Guerin is due principally to trehalose mycolates. J. Immunol. 174:5007-5015.[Abstract/Free Full Text]
    16. 16
  16. Huang, J., C. A. Hartley, N. P. Ficorilli, B. S. Crabb, and M. J. Studdert. 2005. Glycoprotein G deletion mutants of equine herpesvirus 1 (EHV1; equine abortion virus) and EHV4 (equine rhinopneumonitis virus). Arch. Virol. 150:2583-2592.[CrossRef][Medline]
    17. 17
  17. Kydd, J. H., D. Hannant, and J. A. Mumford. 1996. Residence and recruitment of leucocytes to the equine lung after EHV-1 infection. Vet. Immunol. Immunopathol. 52:15-26.[CrossRef][Medline]
    18. 18
  18. Laing, K. J., and C. J. Secombes. 2004. Chemokines. Dev. Comp. Immunol. 28:443-460.[CrossRef][Medline]
    19. 19
  19. Rossi, D., and A. Zlotnik. 2000. The biology of chemokines and their receptors. Annu. Rev. Immunol. 18:217-242.[CrossRef][Medline]
    20. 20
  20. Schall, T. J., K. Bacon, R. D. Camp, J. W. Kaspari, and D. V. Goeddel. 1993. Human macrophage inflammatory protein alpha (MIP-1{alpha}) and MIP-1β chemokines attract distinct populations of lymphocytes. J. Exp. Med. 177:1821-1826.[Abstract/Free Full Text]
    21. 21
  21. Smith, P. M., S. M. Kahan, C. B. Rorex, J. von Einem, N. Osterrieder, and D. J. O'Callaghan. 2005. Expression of the full-length form of gp2 of equine herpesvirus 1 (EHV-1) completely restores respiratory virulence to the attenuated EHV-1 strain KyA in CBA mice. J. Virol. 79:5105-5115.[Abstract/Free Full Text]
    22. 22
  22. Smith, P. M., Y. Zhang, W. D. Grafton, S. R. Jennings, and D. J. O'Callaghan. 2000. Severe murine lung immunopathology elicited by the pathogenic equine herpesvirus 1 strain RacL11 correlates with early production of macrophage inflammatory proteins 1{alpha}, 1β, and 2 and tumor necrosis factor alpha. J. Virol. 74:10034-10040.[Abstract/Free Full Text]
    23. 23
  23. Standiford, T. J., S. L. Kunkel, N. W. Lukacs, M. J. Greenberger, J. M. Danforth, R. G. Kunkel, and R. M. Strieter. 1995. Macrophage inflammatory protein-1 alpha mediates lung leukocyte recruitment, lung capillary leak, and early mortality in murine endotoxemia. J. Immunol. 155:1515-1524.[Abstract]
    24. 24
  24. Standiford, T. J., M. W. Rolfe, S. L. Kunkel, J. P. Lynch III, M. D. Burdick, A. R. Gilbert, M. B. Orringer, R. I. Whyte, and R. M. Strieter. 1993. Macrophage inflammatory protein-1 alpha expression in interstitial lung disease. J. Immunol. 151:2852-2863.[Abstract]
    25. 25
  25. Van de Walle, G. R., M. L. May, W. Sukhumavasi, J. von Einem, and K. Osterrieder. 2007. Herpesvirus chemokine-binding glycoprotein G (gG) efficiently inhibits neutrophil chemotaxis in vitro and in vivo. J. Immunol. 179:4161-4169.[Abstract/Free Full Text]
    26. 26
  26. von Einem, J., P. M. Smith, G. R. Van de Walle, D. J. O'Callaghan, and N. Osterrieder. 2007. In vitro and in vivo characterization of equine herpesvirus type 1 (EHV-1) mutants devoid of the viral chemokine-binding glycoprotein G (gG). Virology 362:151-162.[CrossRef][Medline]
    27. 27
  27. von Einem, J., J. Wellington, J. M. Whalley, K. Osterrieder, D. J. O'Callaghan, and N. Osterrieder. 2004. The truncated form of glycoprotein gp2 of equine herpesvirus 1 (EHV-1) vaccine strain KyA is not functionally equivalent to full-length gp2 encoded by EHV-1 wild-type strain RacL11. J. Virol. 78:3003-3013.[Abstract/Free Full Text]
    28. 28
  28. Walker, C., D. N. Love, and J. M. Whalley. 1999. Comparison of the pathogenesis of acute equine herpesvirus 1 (EHV-1) infection in the horse and the mouse model: a review. Vet. Microbiol. 68:3-13.[CrossRef][Medline]
    29. 29
  29. Walker, C., P. Packiarajah, J. R. Gilkerson, D. N. Love, and J. M. Whalley. 1998. Primary and challenge infection of mice with equine herpesvirus 1, strain HSV25A. Virus Res. 57:151-162.[CrossRef][Medline]
    30. 30
  30. Walker, C., V. M. Perotti, D. N. Love, and J. M. Whalley. 1999. Infection with equine herpesvirus 1 (EHV-1) strain HVS25A in pregnant mice. J. Comp. Pathol. 120:15-27.[CrossRef][Medline]
    31. 31
  31. Wolpe, S. D., G. Davatelis, B. Sherry, B. Beutler, D. G. Hesse, H. T. Nguyen, L. L. Moldawer, C. F. Nathan, S. F. Lowry, and A. Cerami. 1988. Macrophages secrete a novel heparin-binding protein with inflammatory and neutrophil chemokinetic properties. J. Exp. Med. 167:570-581.[Abstract/Free Full Text]
    32. 32
  32. Zagury, D., A. Lachgar, V. Chams, L. S. Fall, J. Bernard, J. F. Zagury, B. Bizzini, A. Gringeri, E. Santagostino, J. Rappaport, M. Feldman, S. J. O'Brien, A. Burny, and R. C. Gallo. 1998. C-C chemokines, pivotal in protection against HIV type 1 infection. Proc. Natl. Acad. Sci. USA 95:3857-3861.[Abstract/Free Full Text]

Journal of Virology, February 2008, p. 1714-1722, Vol. 82, No. 4
0022-538X/08/$08.00+0     doi:10.1128/JVI.02137-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Van de Walle, G. R., Kaufer, B. B., Chbab, N., Osterrieder, N. (2009). Analysis of the Herpesvirus Chemokine-binding Glycoprotein G Residues Essential for Chemokine Binding and Biological Activity. J. Biol. Chem. 284: 5968-5976 [Abstract] [Full Text]  
  • Van de Walle, G. R., Jarosinski, K. W., Osterrieder, N. (2008). Alphaherpesviruses and Chemokines: Pas de Deux Not Yet Brought to Perfection. J. Virol. 82: 6090-6097 [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Van de Walle, G. R.
Right arrow Articles by Osterrieder, N.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Van de Walle, G. R.
Right arrow Articles by Osterrieder, N.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»