|Home | About | Journals | Submit | Contact Us | Français|
Previous studies from our laboratory have shown that dietary α-tocopherol (vitamin EVE) is essential for regulating the cytokine and chemokine response in the brain to herpes simplex virus-1 (HSV-1) infection. The timing of T cell infiltration is critical to the resolution of central nervous system HSV-1 infections. Specifically, the appearance of “neuroprotective” CD8+IFN-γ+ T cells is crucial. During CNS infection, CD8+ T cell priming and expansion in the draining lymph node, followed by recruitment and expansion occurs in the spleen with subsequent accumulation in the brain. Weanling male BALB/cByJ mice were placed on VE deficient (Def) or adequate (Adq) diets for 4 weeks followed by intranasal infection with HSV-1. VE Def mice had fewer CD8+IFN-γ+ T cells trafficking to the brain despite increased CD8+IFN-γ+ T cells and activated dendritic cells in the periphery. VE Def mice had increased T regulatory cells in the periphery and brain and the increase in Tregs decreases CD8+ T cell numbers in the brain. Our results demonstrate that adequate levels of VE are important for trafficking antigen-specific T cells to the brain and dietary VE levels modulate T regulatory and dendritic cells in the periphery.
Herpes simplex virus (HSV)-1 is a common human pathogen. It is the primary causative agent of HSV encephalitis (HSE); the most common fatal sporadic encephalitis in humans [1, 2]. Ninety percent of all HSE cases are caused by HSV-1 . HSE is a significant problem for the immunosuppressed, including people with HIV and those undergoing chemotherapy [4, 5].
Intranasal (i.n.) infection with HSV-1 in mice results in viral entry to the CNS along neuronal pathways of the olfactory and trigeminal nerves and mimics the hypothesized route of human HSE. [1, 7].
CD8+ T cells play a critical role in the response to HSV-1. While the entrance of lymphocytes into the brain parenchyma is tightly controlled by an intact blood brain barrier (BBB), insults, such as a viral infection, allow the BBB to become permeable and circulating lymphocytes can enter the brain parenchyma [8-10]. After HSV-1 infection of the CNS, CD8+ lymphocytes can be found in the brain [7, 11-13]. T cells are responsible for controlling viral replication, but are also partially responsible for the neuropathogenesis associated with HSE[14-16]. The timing of T cell infiltration into the brain is critical for controlling the spread of HSV-1 through the brain, but also correlates with the onset and extent of encephalitic symptoms . T cell production of IFN-γ has been demonstrated to be critical for the resolution of both peripheral and central HSV-1 infections [17-20] and inhibits apoptotic neuronal cell death during HSV-1 infection . These studies suggest that while inflammation associated with T cell infiltration can cause encephalitic symptoms, IFN-γ production is important for both clearing the infection and inhibiting neuronal cell death.
T cells circulate through the brain via the vasculature, and monitor the brain at the BBB, however, resident lymphocytes are not a feature of the brain . Antigen or virus travels via the cerebral spinal fluid to the lymph nodes in order to initiate the T cell response . Dendritic cells that cross the endothelium, as well as microglia, may also function as antigen presenting cells at the BBB[23, 24]. Intracerebral infection of mice with murine hepatitis virus results in CD8+ T cell priming and expansion in the draining lymph node (DLN), followed by recruitment and expansion in the spleen with subsequent accumulation in the brain. Data from both the ocular and foot pad models of HSV-1 infection demonstrate that activation of T cells occurs in the DLN and that Ag-specific T cells can be found not only at the site of infection, but also in the spleen [19, 26]. During i.n. HSV-1 infection, antigen-specific CD8+ T cells are found in the superficial cervical lymph node.
In addition to antigen-specific CD8+ T cells, T regulatory cells (Tregs) are induced during HSV-1 infections [28-32] and function to limit antigen specific CD8+T cell responses [32-34]. Originally identified by their role as immune suppressor cells important for maintaining tolerance and limiting autoimmune disease, Tregs are characterized by the co-expression of CD4 and CD25 and the presence of the forkhead/winged helix transcription factor 3 (FoxP3).
The lipid soluble antioxidant vitamin E is found in cellular membranes and prevents lipid peroxidation by scavenging free radicals . VE Def mice are more susceptible to HSV-1 infection, exhibiting greater symptoms of encephalitis and higher viral titers . The production of cytokines, chemokines and adhesion molecules was also significantly increased in HSV-1 infected VE Def mice compared to mice on a VE Adq diet. Studies in both humans and rodents have demonstrated effects of VE supplementation and deficiency on T cells[37, 38]. Therefore, we designed these studies to determine the impact of VE deficiency on the development and trafficking of the antigen specific CD8+ T cell response to HSV-1 and ultimately, their role in clearing HSV-1 in the brain.
VE deficiency resulted in alterations in HSV-specific CD8+ T cells and Tregs. VE deficiency increased the number of activated DC in both the DLN and the spleen following infection. We demonstrate that the increase in Tregs in the periphery inhibits CD8+ T cell migration to the brain. This study suggests that the level of dietary VE can affect the T cell mediated immune response during a primary HSV-1 infection.
Weanling BALB/cByJ male mice (Jackson Labs, Bar Harbor, ME) were fed ad libitum: 1) VE deficient diet (TD 88163) or 2) VE (dl-α-tocopheryl acetate) adequate diet (Harlan Teklad, Madison, WI). After 4 weeks on the diets, mice were lightly anesthetized with a ketamine (0.6mg/kg)/xylazine (0.35mg/kg) solution and infected i.n. with 1.5×106 PFU HSV-1 in 10μL total volume. Mice were housed in the University of North Carolina Animal Facility, an AAALAC accredited facility. Animals were maintained under protocols approved by the Institutional Animal Use and Care Committee.
Mice were perfused with 30mL PBS and the brain, cervical lymph nodes and spleen were collected. The tissues were gently dissociated into single cell suspensions with a Seward Model 80 Stomacher (Tekmar, Cincinnati, OH) and filtered through 70μ nylon mesh. For isolation of parenchymal lymphocytes from the brain, cells were pelleted by centrifugation, resuspended in 70% Percoll (Sigma, St. Louis, MO) and overlayed with 35% Percoll . The suspension was then centrifuged for 20 min at 2000×g. The mononuclear cells were collected from the 35%/70% interphase, washed 2X with HBSS, and resuspended in RPMI. All cell suspensions were then counted using a hemocytometer.
Cells (5×105 total) were incubated with anti-CD16/32 to block non-specific binding to Fc receptors. The number of CD8+ lymphocytes was determined by incubating 1×106 cells with a FITC-conjugated anti-CD3 mAb, APC-conjugated anti-CD4 mAB and PerCp-conjugated anti-CD8α mAb (BD Biosciences) following the manufacturer’s protocol.
For IFN-γ staining, lymphocytes were incubated for 4 hrs with HSV-infected P815 antigen presenting cells or UV-inactivated HSV-1 at an MOI of 10 PFU/ cell in the presence of Bredfilin A (Golgiplug, BD Biosciences). P815 cells (ATCC) were maintained in RPMI-1640. Surface staining was performed as stated above, cells were fixed, permeabilized and incubated with anti-IFN-γ (PE) antibody (BD Biosciences). Splenocytes from uninfected mice incubated in the same culture conditions served as negative controls for IFN-γ staining. Cells from infected mice were also incubated with uninfected P815 cells as a second control. Because HSV-1 tetramers are not available for BALB/c mice, this method allows us to identify antigen specific IFN-γ producing CD8+ T cells.
DC were identified by their expression of CD11b and CD11c. Activation was assessed using anti- CD40, CD80 and CD86 antibodies (eBiosciences).
For the identification of T regulatory cells, cells were incubated with anti-CD4 and anti-CD25 antibody (BD PharMingen), permeabilized, fixed and incubated with anti-FoxP3 antibody and IL-10 or the appropriate isotype control (eBioscience, San Diego, CA). Cells for IL-10 staining were restimulated ex vivo with UV-inactivated HSV and described above. Cells were analyzed using a FACSCalibur (BD Biosciences, MountainView, CA) or Accuri C6 flow cytometer (Ann Arbor, MI) and data analysis was performed using WinMDI 2.9 (Joseph Trotter, Scripps) or CFlow software (Ann Arbor, MI).
The anti-CD25 antibody (Rat IgG1 Clone PC61) or irrelevant IgG (Rat IgG1 to horseradish peroxidase) (BioXCell, New Lebonan, NH) were administered by i.p. injection 2 days prior to infection and on d 6 p.i. at 0.5 mg/mouse in 200 ul of PBS . To confirm depletion, splenocytes were stained with anti-CD25 clone 7D4, CD4 and FoxP3 Ab (BD Biosciences). For CD8 depletion, anti-CD8α antibody 53-6.7 (eBioscience) or irrelevant IgG were administered by i.p. injection 2 days prior to infection and on d 2, 4 and 6 p.i. at 0.1 mg/mouse in 200 ul total volume . To confirm depletion, splenocytes were stained with anti-CD8β, CD3, CD4 and CD8α Ab(BD Biosciences).
For viral titers from the forebrain and brainstem, DNA was extracted (DNeasy, Qiagen), quantified and HSV-1 genome was determined by qPCR with primers and probe specific for the HSV-1 ICP0 gene as previously described . DNA from uninfected tissue was extracted in parallel and served as a negative control.
Data were analyzed by two-tailed ANOVA. Post-hoc analyses were performed using the Fisher’s PLSD post-hoc. Statistical analyses were performed with JMP 6 software (SAS Institute Inc., Cary, NC). Data were considered statistically significant if P<0.05.
To determine if VE influences either the trafficking of T cells or the number of antigen specific CD8+IFN-γ+ T cells, we identified the T cell subsets infiltrating the brain. VE Def mice had significantly fewer CD8+ T cells (Fig. 1A) at d9p.i. Additionally, VE Def mice had significantly fewer CD8+IFN-γ+ T cells at d9 p.i. (Fig. 1B).
To determine if removal of CD8+ T cells from mice results in increased HSE, VE Adq mice were treated with anti-CD8+ antibody or an irrelevant IgG. At the time of infection the percentage of CD8β+ T cells in the spleens of antibody treated mice was reduced from 27.6% (±2.2) to 18% (±1.9%). Continued depletion of the CD8+ T cells was confirmed by the expression of CD8α and CD8β on CD3+ cells (Fig. 2A). At d9 p.i. the percentage of CD3+CD8β+ in both the brain and the spleen were significantly reduced in the antibody treated mice (Fig. 2A and B). Anti-CD8 treated mice have a significant increase in CD4+ T cells in the brain and in the spleen after infection (Fig. 2A and B). The depletion of CD8 T cells results in a trend toward an increase in HSE pathology (Fig 2C).
As the decrease in T cells in the brain of VE Def mice may be a result of a failure of the cells to develop in the periphery, we examined the number of CD8+ T cells in the DLN of the infected mice at d7 and d9 p.i. The VE Def mice had an increase percentage of CD8+IFN-γ+ T cells at d7 and 9 p.i. and an increased number at d9 compared to VE Adq mice (Fig. 3A and B).
To determine if this pattern was also present in the spleen, the numbers CD8+ T cells (Fig. 3C) and CD8+IFN-γ+ T cells (Fig. 3D) were examined. Although the total number of CD8+ T cells was not different between the mice on the two diets, at d7 p.i, the number of HSV-1 specific IFN-γ-producing CD8+ T cells was significantly higher in the VE Def compared with VE Adq mice.
Dendritic cells present antigen to naïve T cells and are required for the expansion of antigen-specific CD8+ T cells. The expression of CD40, CD80 and CD86 indicates the activation of DC and expression of co-stimulatory molecules that are required for optimal T cell activation [41, 42]. We examined CD11c+CD11bint cells in the DLN and spleen of HSV-1 infected mice. Compared to VE Adq mice, VE Def mice had an increase in the percentage of activated DC in the DLN (Fig 4A) and spleen (Fig 4C) as determined by an increase in CD40, CD80 and CD86 expression (Fig 4E, F, and G). In the spleen the number of activated DC was significantly increased in the Def mice and a trend toward an increase in the DLN was also noted (Fig 4B and D).
CCR5 is required for T cells to cross the BBB . We hypothesized that CD8+IFN-γ+ T cells in DLN and spleen may have decreased CCR5 in Def mice and therefore inhibit their trafficking to the brain. However, the number of CD8+IFN-γ+CCR5+ cells was increased in the DLN and spleen in VE Def mice (Fig 5A and 5B). Additionally, the mean fluorescence intensity (MFI) for CCR5 on CD8+IFN-γ+ T cells was not different between the groups in the spleen (Fig 5C) or DLN (data not shown).
It has been hypothesized that Tregs may limit the migration of effector cells from the secondary lymphoid organs to the site of inflammation/infection . We sought to determine if the level of VE altered the numbers of Tregs in the spleen after HSV-1 infection. By d9 p.i., Tregs were significantly higher in percentage and number in the spleen of VE Def compared to VE Adq mice (Fig. 6 A and B). In the brain, the number of Tregs at d9 was significantly higher in VE Def mice (Fig 6C). Following infection, cells that were FoxP3+ also produced IL-10 (Fig 6D). Thus, there was a clear difference in splenic and brain Treg numbers dependent on VE levels.
To test the hypothesis that the increase in peripheral Tregs may inhibit T cell trafficking to the brain, we depleted Tregs by treating the VE Def mice with the anti-CD25 Ab PC61. Previous studies in HSV-1 infected VE Adq mice have demonstrated that treatment with PC61 results in an increase in HSV-1 specific CD8+ T cells [28, 29] and are more effective at controlling HSV-1 replication . At the time of infection the percentage of CD4+CD25+ cells in the spleen was reduced by PC61 treatment (from 6.9% to 2.1%). In VE Def mice, PC61 treatment abrogated the increase in Tregs in both the spleen (Fig 7A) and DLN (data not shown) following HSV-1 infection. Following PC61 treatment, Tregs were roughly equivalent between the VE Adq and VE Def groups (Fig 7A). At d9 p.i., VE Def mice had 68% fewer CD8+IFN-γ+ T cells (Fig 7B) compared to VE Adq mice. Depletion of Tregs from the VE Def mice by PC61 treatment resulted in a restoration of CD8+IFN-γ+ T cells in the brain.
At d7 p.i. brain stem and forebrain were removed from VE Adq-IgG, VE Def-IgG and VE Def-PC61-treated mice. In the VE Def mice, restoring CD8+ T cell trafficking by PC61-treatment failed to significantly decrease viral titers (Fig 8A). In addition, VE Def-PC61-treated mice showed similar encephalitic symptoms as the VE Def mice (Fig 8B).
The timing of T cell infiltration is critical to the resolution of CNS HSV-1 infections. The appearance of “neuroprotective” CD8+IFN-γ+ T cells is crucial for the resolution of infection [12, 20, 27]. In the present study, VE Def mice had significantly fewer brain parenchymal CD8+ T cells, as well as fewer CD8+IFN-γ+ T cells at d9 p.i. Decreased CD8+IFN-γ+ T cells in HSV-1 infected restraint-stressed mice resulted in increased mortality [12, 27]. Previous studies from our lab have demonstrated increased HSE symptoms mortality in HSV-1 infected VE Def mice emphasizing the functional significance of decreased CD8+IFN-γ+ T cells in the brain. The removal of CD8+ T cells from HSV-1 infected mice by genetic knockout, antibody depletion or the elimination of CD8+ T cell function via lymphotoxin-α or granzyme A knockout results in an increase in the severity of HSV-1 [19, 45-47]. In the current study, depletion CD8+ T cells resulted in a trend toward more significant HSE pathology, similar to the VE Def mice. Anti-CD8 treated VE Adq mice had a significant increase in CD4+ T cells that infiltrate the brain. This compensatory increase, which is not seen in VE Def mice (data not shown), may protect the anti-CD8 treated mice from significant mortality .
In VE Def mice it appears that CD8+ cells are accumulating in the DLN and spleen and do not traffic to the brain. It is likely that the increase in CD8+IFN-γ+ T cells in the DLN is due to increased antigen-bearing activated DC as a result of increased viral titers in the brains of VE Def mice . However, it is not clear if the increase in CD8+IFN-γ+ T cells in the spleen is a result of antigen presentation by activated DC draining larger amounts of virus from the brain, or simply a result of proliferation by CD8+IFN-γ+ T cells. We were unable to find genomic DNA or mRNA of HSV-1 in the spleen, suggesting that the virus is not actively replicating in the spleen (data not shown). In the ocular model of HSV-1 infection, DLN and spleen have been identified as locations for the proliferation of HSV-1-specific CD8+ T cells  and similar to our findings, no replicating virus was found in the spleen. Thus far, HSV-1 antigen from CNS infections has not been identified in the spleen. When fluorescent beads are injected into the cerebral hemisphere of mice, the majority of bead-positive DC are found in the cervical and submandibular LN, however, a small percentage of DC in the spleen are bead positive . These data suggest that HSV-antigen from the brain may be presented in the spleen to induce antigen-specific T cells.
The decrease in CD8+ T cells in the brain of VE Def mice is particularly surprising in light of the increase in CD8+IFN-γ+ T cells in both the DLN and spleen. Data from our lab indicate that both chemokines and adhesion molecules are increased in the brains of VE Def mice , suggesting that the failure of the T cells to traffic to the brain is not due to a lack chemokine and adhesion molecule expression in the brain.
CCR5 promotes T cell trafficking across the blood brain barrier . While CCR5 is increased on T cells in humans following VE supplementation , little is known about the effect of VE Def. In the current model, the failure of CD8+IFN-γ+ T cells to traffic to the brain is not a result of CCR5 down-regulation, as the expression of CCR5 on these cells in both the DLN and spleen was similar between diets.
Tregs function to maintain immunological tolerance as well as reducing pathological immune responses, which may occur with viral infections . Recently, many studies have demonstrated that Tregs are increased during both HSV-1 and HSV-2 infections [28-32]. It is hypothesized that Tregs are upregulated during infection in order to limit damage to tissues by cytotoxic cells . In our study, Tregs were significantly increased in the spleens and brains of VE Def mice compared to VE Adq mice. The role of Tregs in the ocular/stromal keratitis model of HSV-1 has been extensively studied [28, 50, 51]. Previous studies have demonstrated that VE Adq mice depleted of Tregs prior to HSV-1 infection are more effective at controlling the virus by increasing cytotoxicity and proliferation of CD8+T cells and increasing the frequency of CD8+IFN-γ+ T cells . However, the induction of Tregs in the periphery and CNS of HSV-1 infection with the i.n. model has not been studied. In the current study, the VE Def mice have a high frequency of Tregs and lower CD8+IFN-γ+ T cells. However, in the spleens of VE Def mice there was a high frequency of CD8+IFN-γ+ T cells, while also increasing Tregs in the spleen. These findings seem paradoxical, however, we only measured the numbers of Tregs, not their suppressive capacity. Tregs isolated from HSV-infected rabbit conjunctiva are effective in inhibiting the proliferation of anti-CD3-stimulated CD8+ T cells, but not CD8+ T cells re-stimulated with HSV, while studies in mice of the HSK model have demonstrated that Tregs form the spleen have significant suppressive effect on HSV-peptide stimulated CD 8 T cells. In VE Def mice, it appears that Tregs at the site of infection are functioning differently than those in the spleen. It remains to be determined if Tregs isolated from the site of infection/inflammation are functionally different from those isolated from secondary lymphoid organs.
Inducible Tregs (iTregs) are induced from conventional CD4+CD25- cells in the presence of DC-presented antigen and TGF-β in peripheral lymphoid organs[53, 54]. The development of natural Tregs (nTregs) vs. IL-2/TGF-β inducible Tregs in HSV infection is unclear. However, Tregs induced in vitro from conventional CD4+ T cells are effective in suppressing CD8+ effector cells when transferred prior to HSV-1 infection. VE deficiency increases TGF-β and ROS, both of which have been shown to convert precursor cells to become Tregs [56, 57], while Toll-like receptor (TLR) signaling and ROS induce effector T cells [58, 59]. We are unaware of other studies which have examined the simultaneous development of CD8+ T cells and Tregs in a VE Def host. It is possible that the environment in the VE Def host alters the homeostatic regulation of Tregs and T cell effectors. DC-Ag-stimulation in VE Def mice in conjunction with TGF-β, TLR and ROS could lead to both increased CD8+IFN-γ+ T cells and Tregs. Studies addressing this possibility are underway in our lab.
Removal of Tregs by PC61 treatment restored CD8+IFN-γ+ T cells in the brain. These data suggest that Tregs are acting either to inhibit CD8+IFN-γ+ T cell function in the brain or limit their ability to traffic to the brain. We are unaware of any studies that demonstrate Treg cell ability to inhibit CD8+ T cell trafficking during a viral infection. However, several studies have demonstrated that Treg cells can inhibit the migration of CD4+ T cells to the antigen containing tissue [44, 60, 61].
The restoration of CD8+IFN-γ+ T cells trafficking to the brains of VE Def mice failed to significantly decrease viral titers and encephalitic symptoms as an increase in the number of CD8+IFN-γ+ T cells did not occur until d9 p.i.. At 7 d p.i the brains of BALB/c mice infected i.n. with HSV-1 have significant increases in 8-isoprostane, DNA damage and nitric oxide mRNA.. In addition, VE deficiency alone results in the death of neurons and amplifies neurotoxicity in the brain. Studies are underway in the lab to address earlier events in the brains of VE Def mice related to oxidative stress and microglia activation which may contribute more directly to HSE symptoms and mortality.
A second possibility suggests that the inability of VE Def mice to control HSV-1 infection is more complex than the failure of CD8+IFN-γ+ T cells to traffic to the brain. T cell activation, proliferation and function, including the production of perforin and granzyme, are important for clearing viral infections from neurons [47, 65].
In summary, VE deficiency changes the dynamics of the development of the antigen-specific CD8+ T cell response both in the periphery and the CNS via a combination of dendritic cell activation in the periphery and development of T regulatory cells.
This research was supported, in part, by a grant from the National Institute of Environmental Health Sciences (P30ES10126) and by grants NIH to the Clinical Nutrition Research Unit (DK56350) at UNC.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.