The Chemokine Receptor CCR5, a Therapeutic Target for HIV/AIDS Antagonists, Is Critical for Recovery in a Mouse Model of Japanese Encephalitis

Japanese encephalitis is a severe central nervous system (CNS) inflammatory disease caused by the mosquito-borne flavivirus, Japanese encephalitis virus (JEV). In the current study we have investigated the immune responses against JEV in mice lacking expression of the chemokine receptor CCR5, which functions in activation and chemotaxis of leukocytes during infection. We show that CCR5 serves as a host antiviral factor against Japanese encephalitis, with CCR5 deficiency markedly increasing mortality, and viral burden in the CNS. Humoral immune responses, which are essential in recovery from JEV infection, were of similar magnitude in CCR5 sufficient and deficient mice. However, absence of CCR5 resulted in a multifaceted deficiency of cellular immune responses characterized by reduced natural killer and CD8+ T cell activity, low splenic cellularity, and impaired trafficking of leukocytes to the brain. Interestingly, adoptive transfer of immune spleen cells, depleted of B lymphocytes, increased resistance of CCR5-deficient recipient mice against JEV regardless of whether the cells were obtained from CCR5-deficient or wild-type donor mice, and only when transferred at one but not at three days post-challenge. This result is consistent with a mechanism by which CCR5 expression enhances lymphocyte activation and thereby promotes host survival in Japanese encephalitis.


Introduction
The migration of leukocytes in lymphoid organs and to sites of inflammation is coordinated by an array of chemokines that bind to specific receptors on immune cells (reviewed in [1]). Of these, the chemokine receptor, CCR5, is expressed on natural killer (NK) cells, macrophages, and CD4 + and CD8 + T cells. In these cell types, it regulates chemotaxis and cell activation through interaction with the chemokine ligands CCL3, CCL4 and CCL5, which are up-regulated at the site of infection (reviewed in [2]). Understanding the role of CCR5 in the control of pathogen infections has important implications for human health beyond that of other chemokine ligand/receptor interactions, in view of the discovery that CCR5 is a major co-receptor for HIV-1 (reviewed in [3]). Therefore, the chemokine receptor is an important target for therapeutic intervention against HIV/AIDS, and recent clinical trials investigating the efficacy of CCR5 antagonists in patients with HIV/AIDS have provided promising results (reviewed in [4]). However, it has been argued that if CCR5 had a protective role against another group of pathogens, for instance in flaviviral encephalitis, it follows that a therapeutic treatment, which aims to block the receptor, could exacerbate the diseases caused by these pathogens [5,6].
CCR5 was the first chemokine receptor recognized to play a critical role in recovery from flavivirus encephalitis in a study that showed that absence of CCR5 prevented efficient leukocyte trafficking to the brain and viral clearance in mice infected with West Nile virus (WNV) [7]. The important role of CCR5 in the human host response against West Nile encephalitis was supported by a retrospective cohort study involving persons homozygous for CCR5D32 [8], a loss-of-function mutation found in 1-2% of Caucasians [2]. Compared to individuals without the mutation, persons carrying a homozygous CCR5D32 allele had an increased risk for symptomatic WNV infection. This finding was corroborated with a large-scale database study, which associated homozygosity for CCR5D32 with an increased risk of early and late clinical manifestation following WNV infection [9]. CCR5D32 homozygosity has also been associated with severe tick-borne encephalitis symptoms [10] caused by infection with tick-borne encephalitis virus, and a severe case of yellow fever virus-associated viscerotropic disease [11]. Tick-borne encephalitis and yellow fever viruses are also members of the Flavivirus genus, raising the question of generality of CCR5 as an important host factor in recovery from flaviviral infections. Confirmation of a broader link between CCR5 deficiency and augmented incidence and severity of flaviviral disease would add to the concern of potential adverse outcomes associated with CCR5 antagonist use, in view of the large number of human infections inflicted by the different pathogenic members of the Flavivirus genus, and their widespread global distribution (reviewed in [12]. Here we have investigated the role and mechanism of CCR5 in recovery from infection in a mouse model of Japanese encephalitis. Japanese encephalitis virus (JEV) is closely related to WNV, and in terms of human disease incidence and severity the most important member of a serocomplex of mosquito-borne, encephalitic flaviviruses (reviewed in [13]). It is the leading cause of viral encephalitis in Asia, annually accounting for 30,000 to 50,000 cases and ,10,000 deaths. Approximately 3 billion people in the Asia-Pacific region are at risk of infection with JEV. Host immune factors are thought to be the dominant determinants of disease outcome in Japanese encephalitis (reviewed in [14,15]), with intact type I interferon and vigorous humoral immune responses essential for recovery, and CD8 + T cell immunity providing a subsidiary contribution to controlling JEV infection [16,17]. With the use of CCR5-deficient mice we show in this study that the chemokine receptor is an additional important host factor involved in reducing disease severity with Japanese encephalitis.

Ethics statement
All animal experiments were approved by and conducted in accordance with the Australian National University (ANU) Animal Ethics Committee.

Virus and cells
African green monkey kidney (Vero) cells were obtained from the American Type Culture Collection. Working stocks of JEV (strain Nakayama) were infected Vero cell culture supernatants (2610 8 PFU/ml) stored in single-use aliquots at 270uC. Vero cells and YAC-1 cells (Moloney leukemia virus-induced T cell lymphoma) [18] were grown at 37uC in Eagle's minimal essential medium plus nonessential amino acids (MEM) supplemented with 5% fetal bovine serum (FBS). Virus titration was by plaque assay on Vero cell monolayers, as described previously [19].

Mice
Congenic CCR5 2/2 mice (B6.129P2-Ccr5 tm1Kuz /J) [20] and their wild-type controls were bred under specific-pathogen-free conditions and supplied by the Animal Breeding Facility at the John Curtin School of Medical Research, ANU, Canberra. Female mice were used in all experiments.

Mouse inoculation and tissue collection
Mice were infected intravenously (i.v.) via the lateral tail vein with a single injection of 1610 3 PFU of JEV in 100 ml Hanks' balanced salt solution containing 20 mM HEPES (pH 8.0) and 0.2% bovine serum albumin (HBSS-BSA). Mice were monitored twice daily, and changes in 5 parameters including hair coat, posture, breathing pattern, activity, and movement were scored as normal (0), mild (1), moderate (2) or severe (3). Severely moribund mice were euthanized if the overall clinical score was .7, and/or when hindlimb paralysis was observed. For tissue processing, mice were euthanized at the time points indicated and a sterile midline vertical thoracoabdominal incision was made to expose the internal organs. After cardiac puncture for blood collection, animals were perfused with 10 ml sterile PBS. The brain and spinal cord were excised intact and collected for virus titration. For determination of virus titers, sample tissues were snap-frozen on dry ice. One-half of the brain samples were placed in cell culture medium and homogenized for lymphocyte isolation.

Real-time RT-PCR
For determination of viral burden in mouse serum and spleen samples, total RNA in 50 ml splenic homogenates (10% [wt/vol]) and 50 ml serum was extracted using Trizol as described previously [21], and virion RNA content, expressed in genome equivalents, was determined by quantitative reverse transcription (RT)-PCR. For a genome copy standard, JEV RNA extracted from a Vero cell-grown virus stock and quantitated by spectrophotometry was used. RT was performed at 43uC for 90 min in a 10 ml mixture containing 2 ml sample RNA, Expand reverse transcriptase (Roche), RNase inhibitor (Invitrogen), 10 mM deoxynucleoside triphosphate, 10 pmol downstream primer (59-TTGACCGTTGTTACTGCAAGGC-39), 10 mM dithiothreitol, and the manufacturer's recommended buffer condition. Real-time PCR was performed using IQSybr qPCR mixture (Bio-Rad) and 0.2 nM downstream and upstream primers (59-GCTGGATT-CAACGAAAGCCACA-39) under cycling conditions of 95uC for 3 min for 1 cycle and 95uC for 30 sec, 63uC for 30 sec and 72uC for 60 sec for 40 cycles. Each sample was tested in duplicate, and genome copy numbers were determined by extrapolation from a standard curve generated within each experiment. The detection limit of the assay was 4610 3 RNA copies/ml.

Lymphocyte isolation from brain
Homogenized brain samples were digested with 2 mg/ml collagenase type I (Gibco-Life Technologies) in MEM plus 5% FBS for 45 min at 37uC with intermittent shaking and then centrifuged at 4006 g for 10 min. Pellets were suspended in 2 ml 90% Percoll (GE Healthcare) in MEM plus 5% FBS and overlaid gently with 60%, 40% and 10% Percoll in MEM plus 5% FBS. The gradients were centrifuged at 8006g for 45 min at 25uC, and lymphocytes collected from the 40 to 60% interface were subsequently analyzed by flow-cytometry.

NK cell cytotoxic assay
Mice were infected with 10 3 PFU of JEV, i.v., and sacrificed at days 4 and 7 post-infection (pi). Spleens were collected and tested for NK cell cytotoxicity in a standard 51 Cr-release assay [23], using uninfected YAC-1 cells as targets. Splenocytes from uninfected mice were used as controls. P values were calculated from a four-point logarithmic regression curve, interpolated at an effector-to-target (e/t) cell ratio of 30.

Transfer experiments
Eight-week-old CCR5 2/2 or CCR5 +/+ mice were infected with 1610 3 PFU of JEV, i.v., and sacrificed a week later for aseptic removal of spleens. Single-cell splenocyte suspensions were prepared by pressing the spleen tissue gently through a fine metal mesh tissue sieve. Erythrocyte lysis was by suspension of the splenocyte pellet in 4.5 ml distilled water, followed immediately by the addition of 0.5 ml of 106PBS. Lysed cells were discarded after centrifugation at 4006g for 5 min. Splenocytes were resuspended in 100 ml PBS and injected through the lateral tail vein into 8week-old CCR5 2/2 recipient mice. Recipient mice were challenged one or three days later with 1610 3 PFU JEV via footpad injection. For B cell depletion, splenocytes were incubated with anti-CD19 magnetic beads (Miltenyi Biotec) and loaded onto magnetic columns according to the supplier's instructions. Effluent from the columns was collected, and cells were pelleted by centrifugation at 4006g for 5 min. The efficiency of B cell depletion was 99% as assessed by FACS analysis.

Serological tests
For titration of JEV-specific antibody isotypes in mouse serum, ELISAs were performed with HRP-conjugated rabbit anti-mouse IgM, IgG1 and IgG2b (Serotec), and the peroxidase substrate 2,29azino-di(3-ethyl-benzthiazoline sulfonate). The JEV Nakayama strain was used for ELISA antigen production as described previously for Murray Valley encephalitis virus [24]. For determination of ELISA endpoint titers, absorbance cut-off values were established as the mean absorbance of eight negative-control wells containing sera of naive mice plus 3 standard deviations (SD). Absorbance values of test sera were considered positive if they were equal to or greater than the absorbance cut-off, and endpoint titers were calculated as the reciprocal of the last dilution giving a positive absorbance value. Neutralization titers, measured in a 50% plaque reduction neutralization test (PRNT 50 ), were determined as previously described [25].

CCR5 is required for protection from Japanese encephalitis
To assess the impact of CCR5 on the pathogenesis of Japanese encephalitis, mortality was recorded in 8-week-old congenic CCR5 2/2 and CCR5 +/+ wild-type mice after i.v. challenge with 10 3 PFU of JEV. Moribund mice in both groups presented with similar clinical signs starting with generalized piloerection, paresis and rigidity, invariably progressing to severe neurological signs demonstrated by postural imbalance, ataxia and generalized tonicclonic seizures. Median survival time amongst mice that succumbed to infection did not differ between the two groups (11.762.3 days for CCR5 2/2 and 11.162.1 days for CCR5 +/+ mice). Nevertheless, absence of CCR5 resulted in a significant increase in mortality: 64% in CCR5 2/2 versus 28% in CCR5 +/+ mice ( Figure 1).
To address the mechanism for increased susceptibility of mice to JEV infection in the absence of CCR5, viral titers in CCR5 2/2 and control mice infected with 10 3 PFU, i.v., were determined by real-time RT-PCR in serum and spleen, and by plaque assay in brain and spinal cord. Both groups displayed similar viral burden in serum from day 2 to day 6 pi, with peak viremia occurring on day 2 pi (Figure 2A). Similarly, viral load in spleen did not differ significantly between the two groups ( Figure 2B). On the other hand, viral spread into and/or clearance from the CNS was significantly affected by the absence of CCR5: compared to control mice, viral load in brains of CCR5 2/2 mice was between 10-and 10,000-fold higher on days 8 and 10 pi, respectively, with the proportion of mice showing detectable virus titers markedly greater in CCR5 2/2 mice than that in the control group ( Figure 2C). In addition, viral spread in the CNS was more pronounced in CCR5 2/2 mice, with viral titers in spinal cord exceeding those in CCR5 +/+ mice by four log on day 10 pi ( Figure 2D).
Together, these data suggest that CCR5 is critically required for recovery from JEV infection.

Humoral immunity in the absence of CCR5
We have previously established a pivotal role of antibody in recovery from JEV infection [16]. Given that CCR5 is expressed on CD4 + T cells, which are required for efficient generation and maintenance of humoral immunity against JEV, we investigated whether a deficiency in CCR5 would result in poorer JEV-specific antibody responses. However, the kinetics and magnitude of the IgM response was similar in CCR5 2/2 and control mice, appearing first on day 4 pi, and peaking on day 8 pi ( Figure 3A). Class switching to IgG production was also comparable between both groups, initially detected on day 8 pi, and increasing on day 10 pi ( Figure 3B). Assessment of functional activity of the antibody responses by neutralization assay revealed no significant difference between the two groups, reaching mean PRNT 50 titers of 660 (range, 80 to 1280) in CCR5 2/2 mice and 570 (range, 160 to 1280) in CCR5 +/+ mice on day 10 pi ( Figure 3C).
These data indicate that a deficiency in CCR5 does not compromise the ability of mice to prime JEV-specific B cell immune responses.
Blunted NK and CD8 + T cell responses in CCR5 2/2 mice NK cells form an important part of the cellular arm of the host's innate immunity, and function by killing virally infected cells and by release of the cytokines, IFN-c and TNF-a. NK cell activity in JEV infected CCR5 2/2 and control mice were determined by Figure 1. Susceptibility of CCR5 +/+ and CCR5 2/2 mice to infection with JEV. Groups of 8-week-old mice were infected i.v. with 10 3 PFU of JEV. Morbidity and mortality were recorded daily, and surviving mice were monitored for 28 days. The data shown were constructed from two independent experiments. The significance of differences in mortality between wild-type (n = 28) and knockout mice (n = 42) was determined by using the log-rank test (**, P,0.01). doi:10.1371/journal.pone.0044834.g001 measuring cytolytic activity against YAC-1 cells and by intracellular staining of IFN-c. At the peak of NK cell activity (day 4 pi), lysis of YAC-1 cells was significantly reduced in CCR5 2/2 mice compared to CCR5 +/+ mice ( Figure 4A and 4B). This was substantiated by a significantly reduced number of activated, IFNc-expressing, NK cells in spleen of CCR5 2/2 relative to control mice at day 4 pi ( Figure 4C).
CD8 + T cells form the dominant cytotoxic arm of the adaptive immune system. We previously reported a subsidiary role of CD8 + T cells in recovery from JEV infection [16], and hypothesized that a defective CD8 + T cell response could, at least in part, account for the increase in susceptibility of CCR5 2/2 mice to JEV infection. The primary murine CD8 + T cell responses against JEV and other flaviviruses peak at 7 days pi, and thereafter markedly decline [16,26,27]. Figure 4D demonstrates a significantly blunted JEVimmune CD8 + T cell response in CCR5 2/2 relative to control mice, determined by intracellular cytokine staining following ex vivo stimulation of CD8 + lymphocytes with a JEV NS4B proteinderived H-2D b -binding peptide. Relative to CCR5 +/+ mice, CCR5 2/2 mice showed a 61% and 74% reduction in the number of IFN-c-secreting CD8 + T cells, and a 60% and 64% reduction in the number of TNF-a-secreting CD8 + T cells in spleen on days 4 and 7 pi with JEV, respectively ( Figure 4E). This was also reflected in a significantly smaller number of dual-functional JEVimmune CD8 + T cells (defined as the number of CD8 + T cells expressing both IFN-c and TNF-a following stimulation with cognate peptide), which was reduced by 3-fold and 4-fold on days 4 and 7 pi, respectively, in the absence of CCR5 expression ( Figure 4F).
Collectively, these data indicate that a deficiency in CCR5 results in blunted NK and CD8 + T cell responses after JEV infection.
Impaired leukocyte proliferation and trafficking in CCR5 2/2 mice CCR5 expression drives the migration of leukocytes into the CNS after WNV infection, and thereby contributes to viral clearance and recovery from infection [7]. To investigate whether this is also evident in our mouse model of Japanese encephalitis, we isolated leukocytes from brains of JEV infected CCR5 +/+ and CCR5 2/2 mice by density gradient centrifugation and quantified leukocyte subpopulations by flow-cytometry. On day 7 pi, CCR5 2/2 mice showed a significant (,50%) reduction in infiltration of leukocytes into the brain relative to control mice, and this effect was found for all CCR5-expressing subpopulations, viz. NK cells, F480 + /CD45 hi macrophages, CD8 + T cells and CD4 + T cells ( Figure 5A-D). This difference ceased to be significant by day 10 pi. These data show that in the case of Japanese encephalitis, the absence of CCR5 resulted in delayed trafficking of leukocytes into the brain, despite the much higher virus titers in the CNS of infected CCR5 2/2 than CCR5 +/+ mice.
Given the role of CCR5 expression in augmented leukocyte trafficking into the brain, we next assessed the proportion of cells of the different leukocyte subpopulations that express the chemokine receptor using CCR5 +/+ mice that were infected with JEV. On day 7 pi, 15% of NK cells, 12% of F4/80 + /CD45 hi infiltrating macrophages, 8% of CD8 + T cells and 20% of CD4 + T cells were CCR5-positive ( Figure 5E). This proportion increased on day 10 pi in the case of CD8 + T cells (19%) and F4/80 + / CD45 hi infiltrating macrophages (16%), but decreased for NK and CD4 + T cells. Similar numbers of infiltrating CCR5-positive cells as a proportion of the different leukocyte subpopulations in the brain have been observed in mice with West Nile encephalitis [7].
To assess whether impaired trafficking of leukocytes, per se, or a failure of leukocyte expansion in the periphery accounted for the reduced numbers of immune cells in the brain of infected CCR5 2/2 relative to control mice, we quantified splenic cellularity in CCR5 2/2 and CCR5 +/+ mice at various time points after JEV infection. As expected, given that spleen serves as the main immune-responsive lymphoid organ after an i.v. challenge, total splenocyte numbers in CCR5 +/+ mice increased following infection with JEV, with the peak expansion on day 7 pi (8610 7 cells/spleen; Figure 6A). This contrasted with a significantly lesser number of splenocytes in JEV infected mice lacking CCR5 expression, although the kinetics of the response was similar between the two groups of mice ( Figure 6A). The deficit in cell numbers was found across the different CCR5encoding splenocyte subpopulations, showing a reduction of 43% for NK cells, 33% for F4/80 + macrophages, 34% for CD4 + T cells, and 39% for CD8 + T cells in CCR5 2/2 mice relative to control animals on day 7 pi (Figure 6B-E). The percentage reduction of T cell numbers and activation phenotype (Figure 4) in spleen on day 7 pi correlated closely with the differential leukocyte migration into the brains of CCR5 2/2 and wt mice ( Figure 5).
Together these data imply that the impaired trafficking of leukocytes into the CNS after JEV infection in CCR5 2/2 mice may, at least in part, be a direct result of a deficit in leukocyte numbers in the periphery.
Early transfer of CCR5-deficient splenocytes increases survival of recipient CCR5 2/2 mice challenged with JEV Our mechanistic studies on the role of CCR5 in recovery from JEV infection showed deficiencies in leukocyte trafficking to the CNS, blunted NK and CD8 + T cell responses, and low splenic cellularity, while a previous investigation on WNV concluded that CCR5 increased host survival predominantly by promoting leukocyte migration to the infected brain [7]. To test whether defective immune cell priming and expansion significantly contributed to the increased susceptibility of CCR5 2/2 mice to JEV infection, we adoptively transferred splenocytes from JEV infected CCR5 2/2 and CCR5 +/+ mice to CCR5 2/2 recipients lethally challenged with the virus. Complete protection was achieved when total immune splenocytes from either donor strain were transferred at day 1 pi with JEV ( Table 1). In contrast, no increase in survival was observed when the transfers were performed at 3 days pi, independent of whether the donor mice were CCR5 sufficient or deficient. This suggested that following the establishment of a CNS infection, viral clearance and protection mediated by the adaptive immune responses is ineffective in the case of Japanese encephalitis. This conclusion is not limited to the CCR5-deficient recipient mice, since a similar lack of protective value against JEV of late transfer (day 3 postchallenge) of immune spleen cells was also found in wild-type B6 recipients (unpublished data).
Given the dominant role of B cells (which do not express CCR5) in recovery from Japanese encephalitis [16], we next performed adoptive transfer experiments of B cell-depleted splenocytes from CCR5 2/2 and CCR5 +/+ donor mice to CCR5 2/2 recipients at one day after JEV infection. Interestingly, even B cell-depleted splenocytes from both donor mice provided significant, albeit not complete, protection, increasing the survival rate of recipient mice to that of untreated CCR5 +/+ mice infected with JEV (Table 1).
Together, these data confirm the important role of B cells in protection against Japanese encephalitis, and indicate that CCR5 expression was not required for the disease-ameliorating effect of B cell-depleted lymphocytes primed against JEV. The data also suggest that in adoptive transfer of immune T cells control of JEV infection likely occurred extraneurally, since protection was seen only when the cells were provided early pi, and that accordingly migration of the lymphocytes to peripheral sites of infection did not require expression of CCR5.

Discussion
The role of the chemokine receptor CCR5 in infection is variable in terms of its impact on pathogenesis and disease outcome. An important disease-ameliorating contribution of CCR5 has been documented in settings of trypanosomiasis [28], toxoplasmosis [29], influenza [30,31], parainfluenza [32], West Nile encephalitis [7], tick-borne encephalitis [10], genital herpes [33], and chlamydia infection [34], while no significant effect of CCR5 deficiency on the outcome of other microbial infections could be found [35][36][37][38][39]. Mechanistically, the protective value of CCR5 has been largely attributed to its regulatory effect on leukocyte trafficking to the site of infection, although additional functions of CCR5 in the immune response involving antigen recognition [36,40,41], priming and proliferation of lymphocytes  [33,36,37,[40][41][42][43][44], and T cell memory formation [32,45] have also been recognized. In contrast, a disease-potentiating effect of CCR5 has been found in experimental cerebral malaria [46], schistosomiasis [47], leishmaniasis [48], cryptococcosis [49], herpes keratitis [50], mouse hepatitis virus-induced multiple sclerosis [51], and HIV/AIDS [52], mostly as a consequence of increased immunopathology, and in the case of HIV/AIDS due to usage of CCR5 as an alternative receptor for virus infection (reviewed in [3]). This dichotomy in the role of CCR5 in the immune response to infection with different pathogens prevents generalization of the impact of the chemokine receptor in disease prognosis.
In the present study we establish a critical role of CCR5 in recovery from Japanese encephalitis, which was reflected in increased viral burden in the CNS, but not extraneural tissues, and increased mortality in the absence of CCR5 expression. It was previously shown that type I interferon and antibody, but not T cells, are key to control of virus replication in peripheral tissues in the mouse model of Japanese encephalitis, while T cells play a role in reducing viral burden and dissemination in the CNS [16]. CCR5-deficient mice displayed a wide-ranging debilitation of JEV-specific immune responses; these included poor NK cell activity, a suboptimal JEV-immune CD8 + T cell response, low splenic cellularity, and impaired trafficking of leukocytes (NK cells, macrophages, and CD8 + and CD4 + T cells) to the brain. These factors most likely acted in concert to create a lethal combination of defective viral clearance from peripheral tissues and the CNS, thereby increasing the susceptibility of the host to JEV infection. In addition, recent findings suggest that CCR5, which is expressed in neurons and is up-regulated in the brain in pathological conditions, could have a direct neuro-protective function by increasing neuronal survival (reviewed in [2]). Our results are consistent with the previous finding of increased susceptibility of CCR5-deficient mice in a model of West Nile encephalitis [7], although mechanistic differences in the role of CCR5 in recovery from WNV and JEV infection are apparent. While a significant ($50%) overall reduction in migration of NK cells, CD4 + and CD8 + T cells, and activated infiltrating macrophages to the brain of JEV and WNV infected CCR5 2/2 mice was seen in both investigations, this difference relative to CCR5 +/+ control mice narrowed between days 7 and 10 pi in the case of JEV infection, but became more pronounced for WNV. More importantly, the WNV study found similar splenic immune responses in CCR5 sufficient and deficient mice, whereas we observed markedly reduced numbers of lymphocytes in spleens of mice lacking CCR5 expression. It is likely that the latter was a major factor contributing to the severe phenotype in JEV infected CCR5-deficient mice, since adoptive transfer of B cell-depleted immune splenocytes corrected the phenotype, independent of whether the cells were derived from CCR5 2/2 or CCR5 +/+ donor mice. Low splenocyte numbers in the absence of CCR5 may reflect restricted leukocyte proliferation secondary to poor early priming, increased cell turnover to other lymphoid organs, and/or increased T cell death. However, since the spleen is the main responsive peripheral lymphoid organ following an i.v. challenge, defective proliferation most likely accounts for the diminished splenocyte numbers in CCR52/2 mice. Moreover, other studies have also observed restricted proliferation of splenocytes, or specific subsets thereof, in the absence of CCR5, both in physiologic [53] and pathologic conditions [33,36]. The latter studies deal with virus challenge models with a resultant impairment in control of infection.
Concomitant with the limited number of leukocytes in spleens of CCR5-deficient mice infected with JEV, the functional activities of NK and CD8 + T cell responses were blunted, while the antibody response against JEV did not significantly differ from that elicited in infected wild-type mice. This finding was consistent with several other studies that demonstrated a critical role of CCR5 expression on naïve T cells in activation of CD8 + T cell responses by guiding the cells to and enhancing interaction with antigen-presenting dendritic cells in immunogen-draining lymph nodes [40][41][42][43][44]. We have previously found that co-transfer of immune CD4 + and CD8 + T cells, but not individual transfer of either lymphocyte subpopulations, was protective in lethal infection with JEV [16]. Thus, the suboptimal stimulation of virus-specific T cell responses together with their reduced migration into the CNS most likely accounted for the increased susceptibility of CCR5-deficient mice to Japanese encephalitis.
It is unclear whether the reduced NK cell activity in CCR5deficient relative to wild-type mice contributed to the increased disease severity with Japanese encephalitis, as has been proposed in studies on herpes simplex virus in CCR5 2/2 mice [33,36]. Flaviviruses are in general poor inducers of NK cells, and flavivirus infection reduces the susceptibility of target cells to NK cell lysis [23,54]. The latter is thought to be a consequence of virus-induced up-regulation of MHC-I on the surface of flavivirus-infected cells [55][56][57][58]. Therefore, it appears that NK cell responses do not markedly enhance recovery from flavivirus infection. Consistent with this interpretation, in vivo depletion of NK cells using monoclonal antibody anti-NK1.1 did not result in increased mortality or increased viral burden in the CNS in adult mice infected with JEV (our unpublished result) or WNV [59].
In conclusion, our study established CCR5 as a host factor against severe disease in a mouse model of Japanese encephalitis.