|Home | About | Journals | Submit | Contact Us | Français|
Type I IFNs are potent anti-viral cytokines that contribute to the development of the adaptive immune response. To determine the role of type I IFNs in this process in an infectious disease model, mice deficient in the type I IFN receptor (CD118-/-) were ocularly infected with HSV-1 and surveyed at times post infection in the nervous system and lymph node for virus and the host immune response. Virus titers were elevated in the trigeminal ganglia and brain stem with virus disseminating rapidly to the draining lymph node of CD118-/- mice. T cell and plasmacytoid dendritic cell infiltration into the brain stem was reduced in CD118-/- mice following infection which correlated with a reduction in CXCL10 but not CXCL9 expression. In contrast, CXCL1 and CCL2 levels were up-regulated in the brain stem of CD118-/- mice associated with an increase in F4/80+ macrophages. By day 5 post infection, there was a significant loss in T cell, NK cell, and plasmacytoid dendritic cell numbers in the draining lymph nodes associated with an increase in apoptotic/necrotic T cells and an appreciable lack of HSV-specific CD8+ T cells. The adoptive transfer of HSV-specific TCR transgenic CD8+ T cells into CD118-/- mice at the time of infection modestly reduced viral titers in the nervous system suggesting in addition to the generation of HSV-specific CD8+ T cells, other type I IFN-activated pathways are instrumental in controlling acute infection.
IFN-α/β (type I IFNs) are cytokines produced by numerous cell types typically associated with anti-viral or anti-proliferative characteristics and more recently, appreciated for their potential application in controlling autoimmune processes including multiple sclerosis (1, 2). Type I IFNs signal through a single heterodimer receptor (CD118) composed of an alpha and beta chain which upon activation elicits a signaling cascade leading to the induction of a number of IFN responsive genes (3). The absence of one of the components within the signaling cascade signal transducer and activator of transcription 1 (STAT1)3 increases susceptibility to viral pathogens in humans (4) and mice (5). Relative to HSV-1 infection, STAT1 deficient mice are profoundly sensitive to infection underscored by avirulent HSV-1 mutants that replicate and spread in STAT1 deficient animals but not in fully competent wild type (WT) mice (6). Consequently, it is not surprising HSV-1 encodes for proteins that repress CD118 signaling (7) or target downstream effector pathways that promote an anti-viral state in the host cell (8).
In addition to targeting the type I IFN pathway, HSV-1 interferes with the MHC class I (9-11) and class II (12) processing pathways, disrupts TCR signaling (13), attenuates CTL cytolytic activity (14), and induces CD4+ T cell apoptosis (15). Active infection of immature dendritic cells (DCs) results in asynchronous downregulation of co-stimulatory and adhesion molecules including CD40, CD54, and CD80 which may be driven by the HSV-encoded virion host shutoff protein (16-19). Similar to T cells, DCs also reportedly undergo apoptosis following HSV-1 infection (20, 21). Collectively, disruption in the capacity to process and present antigen, suppress the expression of appropriate co-stimulatory molecules, and block DC maturation all contribute to the success of HSV-1 in eluding detection and countering the host immune response. Type I IFNs provide direct support of the host adaptive immune system at the level of T cells (22, 23) and DCs (24-26) and therefore, antagonize the action of HSV-1-encoded proteins.
Previous investigations have used CD118 deficient (CD118-/-) mice to demonstrate the sensitivity of these animals to HSV-1 infection based on virus replication, dissemination of the virus in the host or to illustrate the importance of virally-encoded proteins (6, 27, 28). However, no studies have addressed the innate or adaptive immune response within organized lymphoid tissue or infected tissue in CD118-/- mice in response to acute HSV-1 infection. The current study has focused on characterizing changes within the draining lymph nodes (mandibular, MLN) and nervous system relating viral loads to infiltrating and resident leukocyte populations and the cytokines/chemokines expressed following infection.
C57BL/6J (WT) mice were purchased from The Jackson Laboratory. Mice deficient in the type I IFN receptor (CD118-/-) (29) or HSV glycoprotein B (gB)T-I.1 TCR transgenic mice (30) on a WT background were maintained at Dean McGee Eye Institute. Animal treatment was consistent with the National Institutes of Health Guidelines on the Care and Use of Laboratory Animals. All procedures were approved by the University of Oklahoma Health Sciences Center and Dean A. McGee Eye Institute Institutional Animal and Care Use Committee. HSV-1 (strain McKrae) was grown and maintained as previously described (31).
Male and female WT and CD118-/- mice (6-10 weeks of age) were anesthetized by i.p. injection with xylazine (6.6 mg/kg) and ketamine (100 mg/kg) followed by scarification of the cornea using a 25 5/8-gauge needle. The tear film was then blotted, and the cornea was topically inoculated with 1,000 PFU of HSV-1 in 3 μl of RPMI-1640 medium. HSV-1 viral titers were determined in the designated tissue at times post infection (pi) by plaque assay as previously described (32).
At the indicated time pi, mice were exsanguinated and the MLN, thymus, spleen, trigeminal ganglia (TG), and brain stem (BS) were removed, processed, labeled with antibodies, analyzed using a Coulter Epics XL flow cytometer (Beckman Coulter), and the absolute number of cells residing in the indicated tissue or organ was determined as previously described (33).
Single cell suspensions from the MLN were generated by passing the tissue thru a 70 μm cell strainer (BD Falcon). One million cells were labeled with 1-2 μg of the PE-conjugated HSV peptide-specific gB498-505 (SSIEFARL) MHC tetramer (MHC Tetramer Lab, Baylor College of Medicine) for 60 min on ice in the dark. The cells were washed (300 × g, 5 min at 4° C) and labeled with 1-2 μg FITC-conjugated anti-CD8 and PE-Cy5-conjugated anti-CD45. Following a 30 min incubation on ice in the dark, cells were washed again and resuspended in 1% paraformaldehyde. After a 60 min incubation at 4° C, the cells are washed again and resuspended in 1X PBS. Cells were subsequently analyzed by flow cytometry as described (33).
At the indicated time prior to or pi, the TG, BS, and MLN were removed from the exsanguinated mice and placed in 500 μl of 1X PBS containing a protease inhibitor mixture (Calbiochem) on ice. Following homogenization with a tissue miser (Fisher Scientific), the homogenates were clarified by centrifugation at 10,000 × g for 1 min. The levels of CXCL1, CXCL9, CXCL10, IFN-γ, and IL-2 were determined by ELISA according to the manufacturer's instructions (Quantikine immunoassay; R&D Systems).
Apoptotic cells from the MLN of WT and CD118-/- mice were determined using a commercially available kit containing PE-conjugated annexin V (BD Pharmingen) and used according to the manufacturer's instructions. Uninfected MLN served as background controls.
One million MLN cells from WT and CD118-/- mice were placed in 1.0 ml of RPMI-1640 containing 10% FBS and added to 24 well cultures plates containing 10 μl of HSV-1 (multiplicity of infection = 0.01). Twenty four hr following incubation in 5% CO2, 95% air at 37° C, the cultures were frozen/thawed twice, and clarified (10,000 × g, 1 min) supernatant was assayed for infectious virus by plaque assay.
Spleen cells from naive HSV gB-specific T cell receptor (TCR) transgenic mice (34) were highly enriched (> 95%) for CD8+ T cells using MACS columns (Miltenyi Biotec). At the time of infection, 3 × 106 enriched HSV gB-specific CD8+ T cells were introduced intravenously into CD118-/- mice. Five days pi, the mice were euthanized and assessed for viral content of infected TG and BS or gB-specific CD8+ T cells in the MLN by tetramer staining and flow cytometry. CD118-/- and WT mice that did not receive cells served as negative and positive controls respectively.
Statistical analysis was conducted using the GBSTAT program (Dynamic Microsystems). Student's t test was used to determine significant (p<.05) differences between WT and CD118-/- groups. For adoptive transfer experiments, ANOVA and Tukey's t test were used to determine significant (p<.05) differences.
A previous study reported the dissemination of HSV-1 in CD118-/- mice following inoculation of mice in the footpad or cornea using a luciferase-engineered recombinant virus and bioluminescence imaging (28). To more fully understand the contribution of the type I IFN pathway in resistance to HSV-1 infection, we sought to evaluate the host innate and adaptive immune response relative to virus replication and spread. Following ocular infection, CD118-/- mice harbored significantly more virus in the TG and BS in comparison to WT controls at day 3 and 5 pi (Fig. 1A & B). Moreover, virus rapidly disseminated to the MLN evident by detection as early as day 2 pi. (Fig. 1C). Likewise, virus was detected in 50% of spleen samples from CD118-/- mice by day 3 pi with 7/8 CD118-/- spleen samples with detectable virus by day 5 pi in comparison to 0/8 WT spleen samples at day 3 pi and 1/5 WT spleen samples at day 5 pi (Fig. 1D). Thymus samples were equally telling with 7/7 thymus samples from CD118-/- mice possessing HSV-1 by day 5 pi compared to 0/7 thymus from WT mice (Fig. 1E). No virus was detectable in the thymus of CD118-/- mice at earlier time points.
Since virus was detected early within the MLN of the CD118-/- mice following ocular infection, the draining lymph node was evaluated for predicted changes in the cellular constituency populating the organ. There was no significance change in the total number of CD45+ leukocytes or subpopulations including NK cells, neutrophils, macrophages, DC, or T cells residing in the MLN prior to or 3 days pi comparing WT to CD118-/- mice (Fig. 2). However, by day 5 pi, there was a massive loss of nearly all cell populations found in the MLN of CD118-/- mice in comparison to WT animals (Fig. 2). The loss was most pronounced in the CD4+ and CD8+ T cell populations. Consistent with this finding, there was a significant increase in the percentage of late apoptotic/necrotic CD4+ and CD8+ T cells in the MLN of CD118-/- mice by day 5 pi in comparison to WT animals (Fig. 3). At day 3 pi, there were between 30-50% more T cells undergoing apoptosis in the CD118-/- mouse MLN in comparison to WT MLN T cells which may have contributed to the significant increase in late apoptotic cells observed by day 5 pi in CD118-/- MLN (Fig. 3). Even though there was a significant loss of DC in the MLN of HSV-1-infected CD118-/- mice by day 5 pi (Fig. 2), the percentage of cells positive for the co-stimulatory molecule CD80 was elevated in comparison to WT mice at this time point (Fig. 4A).
In addition to an increase in apoptotic/necrotic events in the MLN as a means to explain the reduction in leukocyte numbers in HSV-1-infected CD118-/- mice, a number of other aspects may factor into the mechanism associated with cell loss including soluble mediators that contribute to the anti-viral state in the lymph node environment or facilitate growth and differentiation of lymphocytes. To address the first possibility, single cell suspensions of MLN cells were evaluated for susceptibility to HSV-1 infection in culture. MLN cell cultures from WT and CD118-/- mice produced similar levels of HSV-1 (3.59 ± 0.14 HSV-1 log PFU/ml compared to 3.81 ± 0.13 HSV-1 log PFU/ml respectively) negating the possibility that MLN cells from CD118-/- mice were more susceptible to infection due to the absence of the type I IFN pathway. To address the second point, MLN from infected and uninfected mice were assayed for cytokine content. Contrary to the predicted outcome, MLN from CD118-/- contained significantly more IFN-γ compared to the WT MLN at times pi (Fig. 4B). There was also an increase in IFN-responsive chemokines including CXCL9 and CXCL10 (Fig. 4B). In contrast, IL-2 levels were reduced in the MLN from CD118-/- mice commensurate with the loss of T cells by day 5 pi (Fig. 4B). Consequently, the crash in T cell numbers in the MLN of CD118-/- mice may be due to a combination of virus present in the tissue resulting in an increase in apoptotic T cells and a reduction in IL-2 levels which exacerbate the disruption.
Since CD118-/- mice were found to possess significantly more virus in the nervous system at times pi and chemokines are expressed within the nervous system in response to infection (35, 36), levels of select chemokines were evaluated in the TG and BS of WT and CD118-/- mice. In the TG, CXCL1 levels were elevated whereas CXCL10 levels were reduced in CD118-/- mice at day 5 pi. In uninfected mice or at day 3 pi, the chemokine levels were found to be similar comparing the two genotypes (Fig. 5A). In contrast, no significant change in the levels of two other prominent chemokines CCL2 or CXCL9 were found to be different (Fig. 5A). By comparison, CXCL1 and CCL2 levels were elevated in the BS of CD118-/- mice by day 5 pi whereas day 3 pi and in uninfected mice, similar levels were found comparing WT to CD118-/- mice (Fig. 5B). The results measuring CXCL9 and CXCL10 levels showed a striking difference in BS CXCL10 with reduced expression in CD118-/- mice by day 5 pi (Fig. 5B). Similar levels were found at day 3 pi or in uninfected mice comparing WT to CD118-/- animals (Fig. 5B). Likewise, there were no significant differences in CXCL9 expression in the BS comparing WT to CD118-/- mice prior to or following infection (Fig. 5B).
Since the results show selective changes in specific chemokines in the peripheral and central nervous system comparing WT to CD118-/- mice following HSV-1 infection, the influence these changes may have on the recruitment of leukocytes was next investigated. Analysis of leukocyte populations (gated on CD45hi expressing cells) residing in the TG including DC (both B220+CD11c+ and B220-CD11c+), macrophages (F4/80+Gr1-), neutrophils (F4/80-Gr1+), NK cells (NK1.1+CD3-), CD4+ T cells (CD3+CD4+), and CD8+T cells (CD3+CD8+) found no difference in the total influx of leukocytes prior to or after infection comparing WT to CD118-/- mice (data not shown) but showed a trend (p = 0.06) in an increase in macrophages and a decrease in CD8+ T cells by day 5 pi in the CD118-/- mice (Fig. 6A). Similar to the results in the TG, there was no significant difference in the total leukocyte population residing in the BS of WT or CD118-/- mice prior to or after infection (data not shown). However, there was a significant increase in the number of macrophages and a decrease in the number of plasmacytoid DC (pDC, B220+CD11c+) and CD8+ T cells recruited or retained in the BS of CD118-/- mice by day 5 pi (Fig. 6B). CD4+ T cell numbers were also found to be reduced in the BS of CD118-/- mice but similar to TG samples, the difference did not reach significance (p>.05).
CD8+ T cells have previously been reported to participate in the clearance of HSV-1 during acute infection or prevention of latent virus from reactivation (37-40). Since type I IFNs have previously been reported to facilitate CD8+ T cell clonal expansion in a STAT-4-dependent manner (41), the level of HSV gB-specific CD8+ T cell numbers was evaluated in the MLN of WT and CD118-/- mice following ocular HSV-1 infection. The day 5 pi time point was chosen since an earlier time point (i.e., day 3 pi) found no detectable HSV gB-specific CD8+ T cells in the MLN. At day 5 pi, the results show the presence of the HSV-specific CD8+ T cells in the MLN of infected CD118-/- mice was noticeably reduced in comparison to the population residing in the MLN of WT mice (Fig. 7A). Specifically, between 3.5-5.1% of the CD8+ T cells recovered from the MLN of WT mice were specific for HSV gB compared to 0.6-0.7% of the CD118-/- MLN population.
To further define the importance of this population of cells relative to HSV-1 infection in CD118-/- mice, adoptive transfer experiments were undertaken in which HSV gB-specific TCR transgenic CD8+ T cells were transferred into CD118-/- mice and the recipients were evaluated for sensitivity to infection. The transfer of transgenic TCR T cells into CD118-/- significantly reduced the viral load in the TG and BS in comparison to CD118-/- mice that did not receive cells (Fig. 7B). However, the virus titer was still significantly elevated in comparison to WT mice (Fig. 7B). It should also be noted the number of HSV gB-specific CD8+ T cells, total CD4+ T cells, and total CD8+ T cells were restored in the CD118-/- mouse recipients of the HSV gB-specific CD8+ T cells (Figs. 7C & 7D). Collectively, while HSV-specific CD8+ T cells do contribute toward resistance to HSV-1 infection in CD118-/- mice, additional factors are also required to restore the phenotype consistent with that of WT mice.
Type I IFNs contribute to the development of the immune response to infectious pathogens at various levels. Regarding the innate immune response, IFN-α promotes NK cell proliferation and cytotoxicity (42) in a STAT-4-independent, STAT-1-dependent fashion requiring IL-15 (43-45). In the present study, there was a modest loss of NK cell numbers by day 5 pi in the MLN of CD118-/- mice. However, at day 3 pi equivalent numbers of NK cells were measured in the MLN comparing WT to CD118-/- mice. At that time point, significant levels of HSV-1 were recovered in the MLN associated with the expression of IFN-γ in the CD118-/- mice. Although not studied, it is tempting to speculate the likely source of the IFN-γ was NK cells (46). Since IFN-α drives the expression of IL-10 thru IFN regulatory factor-1 and the ISGF3 complex (47, 48) and IL-10 is a feedback inhibitor of IFN-γ production (49), the absence of an intact type I IFN pathway would alleviate negative pressure on IFN-γ expression by NK cells. One likely outcome in this scenario is the up-regulation of chemokines responsive to IFN-γ including CXCL9 and CXCL10. Indeed, these chemokines were significantly elevated in the MLN of CD118-/- mice but there was not a corresponding increase in the number of cells responsive to these chemokines including NK cells and pDCs. It is possible HSV-1 within the MLN is capable of lysing the cells at a steady-state rate and thus, maintaining a level similar to that in WT mice.
Type I IFNs are also instrumental driving the production of chemokines thru toll-like receptor-dependent and -independent pathways (50-52). The present study found a correlation between T cell infiltration in the nervous system and local expression of CXCL10. Specifically, CD118-/- mice were found to express significantly less CXCL10 but not CXCL9 in the TG and BS by day 5 pi and such expression was associated with reduced CD4+ and CD8+ T cell infiltration. Since a recent study by our group has found no deficiency in T cell recruitment to the nervous system in HSV-1 infected mice deficient in CXCL10 expression (33), it is more likely the deficiency in T cell recruitment to the nervous system in CD118-/- mice is due to the massive loss of T cells proliferating in the MLN of these animals. Of interest, elevated expression of CXCL1 and CCL2 in the BS was associated with an increase in macrophage influx in CD118-/- mice at a time when the animals present with edematous heads and eyes. In fact, CD118-/- mice rarely live pass 5 days pi which is likely due to the inflammatory response elicited by the pathogen including cytokines and proteases emanating from tissue macrophages and neutrophils that infiltrate the brain following infection (53).
At the crossroads between innate and adaptive immunity resides the DC, the principal antigen presenting cell that greatly influences the direction of the T cell response thru the production of soluble factors including IFN-α (54). Different DC subtypes are known to promote Th1 or Th2 responses (55) in which type I IFNs can contribute toward Th1 (16) or Th2 (56) development. In the case of the infected CD118-/- mice, CD11c+ cells expressed elevated levels of the co-stimulatory molecule, CD80. In contrast with our results, HSV-1 has been found to block DC maturation and stimulatory capacity (57, 58). However, mice deficient in IFN-β reportedly up-regulate CD80 expression (59). The mechanism driving the up-regulation of CD80 expression is unknown but could include the increase in IFN-γ (60) observed in the MLN of CD118-/- mice. Ironically, the elevated IFN-γ level in the MLN may be counterproductive to the overall immune status of the host. Specifically, it has recently been shown IFN-γ-induced nitric oxide generated in response to apoptotic cells leads to cell-mediated immunosuppression (61). As an increase in apoptotic T cells is noted in the CD118-/- mice that manifest in significantly elevated levels of late apoptotic/necrotic T cells by day 5 pi, apoptotic-driven suppression of HSV-specific T cell clonal expansion may explain the deficiency in HSV gB-specific CD8+ T cells in the MLN of CD118-/- mice.
Type I IFNs facilitate the development of T cell immunity including TAP expression (62) which would greatly benefit the host and CD8+ T cell responses thru cross-priming (63). In mice infected with lymphocytic choriomeningitis virus, the loss of type I IFN signaling greatly diminishes clonal expansion of CD8+ T cells as a result of a defect in cell survival (64). These results were found to be more pronounced in lymphocytic choriomeningitis virus-infected mice as opposed to other pathogens including vaccinia virus, vesicular stomatitis virus, or listeria monocytogenes suggesting the dependence on type I IFNs by CD8+ T cells is influenced by the insulting agent (65). In the present study, CD8+ T cells were significantly affected by the absence of CD118 in terms of a reduction in cell number associated with an increase in late apoptosis/necrosis recovered in the MLN, lack of or diminished clonal expansion in the MLN, and a reduction in recruitment to the nervous system. The relevance of HSV-specific CD8+ T cells in the resistance to HSV-1 infection is underscored by the reduction in infectious virus recovered in the TG and BS of CD118-/- recipients of HSV-specific TCR transgenic CD8+ T cells and a partial restoration in the number of CD4+ and CD8+ T cells residing in the MLN. However, the viral titers did not achieve levels found in the immunocompetent WT mice suggesting other pathways independent of type I IFN contribute to the control of HSV-1 which may include IFN-γ-activated mechanisms (32).
Unlike β- and γ-herpesviruses, it was originally thought primary HSV-1 infection does not cause viremia in healthy patients (66, 67). With the advent of PCR, a more recent study reported 34% of young patients presenting with primary herpetic gingivostomatitis had detectable viral transcripts in their PBMCs as determined by PCR (68). In comparing WT to CD118-/- mice, infectious virus was recovered in 20% of WT spleen samples screened in comparison to 88% of spleen samples screened from CD118-/- mice. The dissemination of virus to the spleen strongly suggests a blood-born route. In fact, we have previously reported CD11b+ cells are actively infected with HSV-1 following ocular infection (69). If passenger leukocytes disseminate HSV-1 to various tissues and the incidence of spread is more pronounced in the host with a compromised type I IFN pathway, it is likely such a route may significantly contribute to the incidence of HSV-mediated encephalitis and meningitis (70).
The authors would like to thank Gabby Nguyen and Candace Beverly for their technical help. The authors are also grateful to Drs. Helen Rosenberg and Frank Carbone for the CD118-/- and gBT-I.1 transgenic mice respectively.
Principal support for this work is by NIH R01 AI053108 (DJJC). Additional support includes P20 RR017703, NEI core grant EY12190, and an unrestricted grant from Research to Prevent Blindness.
3Abbreviations used in this paper: STAT-1, signal transducer and activator of transcription 1; WT, wild type; DC, dendritic cell; MLN, mandibular lymph nodes; pi, post infection; TG, trigeminal ganglia; BS, brain stem; pfu, plaque forming unit; gB, glycoprotein B; TCR, T cell receptor; DC, dendritic cells
Disclosures The authors have no financial conflict of interest.