|Home | About | Journals | Submit | Contact Us | Français|
Chronic infections with viruses such as hepatitis B virus, hepatitis C virus, and HIV constitute a major global public health problem. Studies of chronic viral infections in humans and mice show that persistent antigenic stimulation induces dysregulation of T cell responses; virus-specific T cells either undergo clonal deletion or lose their ability to display the full spectrum of effector functions, a condition termed functional exhaustion. The ability to generate and retain sufficient numbers of functionally competent T cells, therefore, becomes vitally important in controlling chronic viral infections. Our understanding of the mechanisms governing T cell homeostasis during chronic viral infections, however, is poor. The phosphatidylinositol 3-kinase (PI3K)/Akt signaling pathway controls cell fate decisions in many cell types by modulating the activity of downstream effectors, including the FOXO family of transcription factors. We have observed dynamic, in vivo alterations in the phosphorylation levels of three key proteins (Akt, FOXO1/FOXO3 [FOXO1/3], and mammalian target of rapamycin [mTOR]) involved in this signaling cascade and have identified the transcription factor FOXO3 as a negative regulator of the magnitude and effector function of CD8 T cells during chronic lymphocytic choriomeningitis virus (LCMV) infection in mice. We report that ablation of FOXO3 in T cells reduced apoptosis, increased the abundance of polyfunctional virus-specific CD8 T cells, and improved viral control. Thus, FOXO3 is a promising candidate target for immunotherapies of chronic viral infection.
More than 500 million people in the world are currently afflicted with chronic infections with viruses such as hepatitis B virus, hepatitis C virus, and HIV (24, 37, 41). A common denominator underlying viral persistence in these chronic viral infections (CVIs) is the dysregulation of virus-specific T cell responses; virus-specific T cells either undergo clonal deletion or lose their ability to express the full spectrum of effector functions, a condition termed functional exhaustion (8, 22, 29, 33, 41, 53, 56, 59). During CVIs, there is a continuum of T cell proliferation and apoptosis, and the balance between these cellular processes controls the abundance of virus-specific CD8 T cells (18, 48, 49, 53, 60). However, the homeostatic mechanisms that control the number of virus-specific T cells under conditions of protracted antigenic stimulation are poorly defined. This is an important issue, because the magnitude of the T cell response is a critical factor in determining viral control.
The FOXO family (FOXO1, FOXO3) of transcription factors plays a crucial role in regulating multiple facets of T cell homeostasis, including apoptosis, proliferation, trafficking, and differentiation (16, 20, 21, 34, 39, 47). In particular, FOXO3 is known to promote the apoptosis of T cells and to limit the clonal expansion of CD8 T cells during an acute viral infection (14, 47). Additionally, FOXO3 has been implicated in vitro as a target for modulating T cell dysregulation in human T cells during HIV infection (52). The principal objective of this study was to gain insight into the role of FOXO3 in orchestrating the dynamics of CD8 T cell responses during a chronic lymphocytic choriomeningitis virus (LCMV) (clone 13) infection in mice.
The derivation of mice carrying floxed FOXO3 alleles has been described elsewhere (36, 50). Mice carrying floxed FOXO3 alleles were bred with CD4-Cre mice at the University of Wisconsin—Madison (UW-Madison) to generate T-cell-specific FOXO3-deficient (FOXO3L) mice. Wild-type (WT) littermates were used as controls with the FOXO3L mice. Mice used in experiments were between the ages of 6 and 8 weeks, and all experiments were performed in accordance with the protocols approved by the University of Wisconsin School of Veterinary Medicine Institutional Animal Care and Use Committee (IACUC). Mice were infected with 2 × 106 PFU of the LCMV clone 13 strain by intravenous (i.v.) injection. Tissue viral titers were quantified by plaque assays with Vero cell monolayers (2).
Single-cell suspensions of splenocytes were stained with antibodies for surface markers, including CD8, CD44, CD122, CD127, CD62L, LFA-1, and KLRG-1 (BD Biosciences, Franklin Lakes, NJ; eBioscience, San Diego, CA; or Southern Biotech, Birmingham, AL) in conjunction with major histocompatibility complex class I (MHC-I) tetramers (Db) specific for the class I-restricted LCMV epitopes, NP396, GP33, and GP276 (30). Cells were fixed in 2% paraformaldehyde (PFA) and were acquired in a FACSCalibur or LSR II flow cytometer (BD Biosciences, Franklin Lakes NJ). To quantify intracellular cytokine production, splenocytes were incubated for 5 h at 37°C with LCMV epitope peptides in the presence of brefeldin A. After stimulation, cells were first incubated with antibodies for surface markers. Next, cells were permeabilized and stained for intracellular cytokines (gamma interferon [IFN-γ], interleukin 2 [IL-2], and tumor necrosis factor alpha [TNF-α]) using the Cytofix/Cytoperm kit (BD Biosciences, Franklin Lakes, NJ). The percentages of cytokine-producing cells were quantified by flow cytometry.
Splenocytes were stained for cell surface markers as described above. After cell surface staining, cells were fixed and permeabilized using Phosflow lysis and Phosflow PermWash I reagents (BD Biosciences, Franklin Lakes, NJ) according to the manufacturer's recommendations. Next, cells were blocked for 30 min on ice in blocking buffer (10% normal goat serum in 2% bovine serum albumin [BSA]–phosphate-buffered saline [PBS]) and were subsequently stained with phosphospecific antibodies (against P-Akt [T308], P-FOXO1/3 [T24, T32], or P-mTOR [S2448]; Cell Signaling Technology, Danvers, MA). As negative controls for staining, antibodies were blocked by preincubation with their specific antigenic peptides for 1 h at room temperature before being added to the cells. Following incubation with the antibody or the peptide-blocked antibody, cells were washed twice and were incubated with a secondary antibody (Alexa 488-conjugated goat anti-rabbit antibody; Sigma-Aldrich, St. Louis, MO) for 40 min. Cells were washed and fixed with 2% PFA. The levels of phosphospecific staining were quantified by flow cytometry. Specific levels of staining (expressed as the corrected mean fluorescence intensity [MFI]) were calculated as the difference between the observed MFI for the phosphospecific protein and the observed MFI for the peptide-blocked control, divided by the observed MFI for the peptide-blocked control.
Splenocytes were stained for surface markers and MHC-I tetramers as described above. After surface staining, cells were fixed and permeabilized using fluorescence-activated cell sorting (FACS) lysing solution and FACS permeabilization solution 2 reagents (BD Biosciences, Franklin Lakes, NJ) and were subsequently incubated with antibodies against Ki67 or granzyme B (BD Biosciences, Franklin Lakes, NJ) for 45 min at room temperature. Virus-specific CD8 T cells staining positive for Ki67 or granzyme B were visualized using a FACSCalibur flow cytometer. Data are expressed either as a percentage of antigen-specific CD8 T cells positive for the respective protein or as the MFI for the indicated protein.
Splenocytes from WT and FOXO3L mice were isolated and stained with anti-CD8 and MHC-I tetramers on the indicated days postinfection (p.i.) as described above, except that no red blood cell lysis was performed. Annexin V staining (BD Biosciences, Franklin Lakes, NJ) was then carried out according to the manufacturer's protocol, except that all staining was performed on ice. The percentage of cells high in annexin V among antigen-specific CD8 T cells was determined by flow cytometry.
FOXO3, in its active hypophosphorylated state, localizes to the nucleus and drives the transcription of genes that oppose cell cycle entry (e.g., p27Kip1) or promote cellular apoptosis (e.g., BIM) (15, 20, 23, 39, 46, 47). Since FOXO3 is part of a phosphatidylinositol 3-kinase (PI3K)-Akt signaling cascade, its hyperphosphorylation by kinases, including Akt, induces the relocalization of FOXO3 to the cytoplasm, thereby preventing the transcription of its target genes (7, 11, 27, 40). In T cells, signaling via the T cell receptor (TCR), CD28, and IL-2/7/15 receptors activates the PI3K/Akt pathway, leading to the phosphorylation of FOXOs, but the dynamics of Akt/FOXO phosphorylation in antigen-specific CD8 T cells in vivo during CVIs remains unknown (5, 9, 10, 13, 19, 42, 43, 54). For mice infected with LCMV clone 13, we used flow cytometry to define the phosphorylation kinetics of PI3K-associated signaling proteins, including Akt (T308), FOXO1/FOXO3 (FOXO1/3) (T24/T32), and mammalian target of rapamycin (mTOR) (S2448), in LCMV-specific CD8 T cells directly ex vivo (Fig. 1). The phosphorylation levels of Akt, FOXO1/3, and mTOR were readily detected in LCMV-specific CD8 T cells and displayed dynamic alterations during the course of a chronic LCMV infection; high levels of phosphorylation were detected at day 8 p.i., but levels declined progressively thereafter, a finding congruent with the circulating viral loads in LCMV clone 13-infected mice.
To explore whether FOXO3 regulates CD8 T cell responses by T-cell-intrinsic mechanisms during the early stages of chronic LCMV infection, we infected FOXO3L mice and WT littermates with LCMV clone 13, and LCMV-specific CD8 T cells were quantified at the peak of the expansion phase (day 8 p.i.) (4) using MHC-I tetramers. The low frequencies of NP396-specific CD8 T cells in WT mice were reminiscent of clonal deletion following infection with LCMV clone 13 (Fig. 2A). The frequencies and total numbers of NP396-specific CD8 T cells in FOXO3L mice were similar to those in WT mice, suggesting that FOXO3 deficiency did not mitigate deletion of these cells during a chronic LCMV infection (Fig. 2A and andB).B). However, FOXO3 deficiency caused marked enhancements in the numbers of GP33- and GP276-specific CD8 T cells (Fig. 2A). Due to the low frequency and poor tetramer binding of NP396-specific CD8 T cells, studies from here on were focused on GP33- and GP276-specific CD8 T cells. Next, to determine whether the augmented accumulation of LCMV-specific CD8 T cells in FOXO3L mice resulted from altered proliferation and/or apoptosis, we measured Ki67 expression and annexin V binding directly ex vivo, as described previously (47). The increased expansion of LCMV-specific CD8 T cells cannot be explained by enhanced proliferation, since the percentages of Ki67-positive cells in WT and FOXO3L mice were similar (Fig. 2B). Instead, the enhanced number of LCMV-specific CD8 T cells in FOXO3L mice was linked to a significant reduction in apoptosis (Fig. 2C). Taken together, the data in Fig. 2 suggested that FOXO3 limited the accumulation of CD8 T cells during a CVI by promoting apoptosis. However, FOXO3 deficiency did not affect the expression of activation markers/cytokine receptors, including CD44, CD62L, LFA-1, CD122, and CD27, on LCMV-specific CD8 T cells (Fig. 3A). The population of effector CD8 T cells elicited by acute viral infections consists of at least two subsets based on the differential expression of CD127 and KLRG-1: the short-lived effector cells (SLECs) (KLRG-1HI CD127LO) and memory precursor effector cells (MPECs) (KLRG-1LO CD127HI) (19, 44). We observed no difference in the percentages of these two effector subsets among LCMV-specific CD8 T cells between WT and FOXO3L mice (Fig. 3B), suggesting that FOXO3 deficiency did not modulate the differentiation of effector subsets following a CVI. During CVIs, elevated levels of PD-1 and LAG-3 have been linked to impaired CD8 T cell effector function (functional exhaustion) (6, 8, 32, 55), but we found no alteration in the levels of either of these proteins on effector CD8 T cells in FOXO3L mice (Fig. 3C).
Next, we investigated whether FOXO3 deficiency affected the effector function of CD8 T cells. For this purpose, we used intracellular cytokine staining (ICCS) to measure antigen-induced production of cytokines (IFN-γ, TNF-α, and IL-2) (Fig. 4). Notably, the frequencies and numbers of IFN-γ-producing LCMV-specific CD8 T cells were significantly greater in the spleens of FOXO3L mice than in those of WT mice. A small percentage of LCMV-specific IFN-γ-producing cells also produced TNF-α and IL-2 in both FOXO3L and WT mice (Fig. 4A). However, the numbers and frequencies of these polyfunctional triple-cytokine-producing (IFN-γ, TNF-α, and IL-2) CD8 T cells in FOXO3L mice were similar to those in WT mice. Likewise, quantification of granzyme B, as a surrogate marker of lytic effector function, in LCMV-specific CD8 T cells showed no differences between WT and FOXO3L mice (Fig. 4B). It is well established that maintenance of CD8 T cell responses and viral control following an LCMV clone 13 infection requires CD4 T cell help (28, 59). Therefore, we investigated whether FOXO3 deficiency affected CD4 T cell responses to the I-Ab-restricted LCMV CD4 epitope GP66 at day 8 p.i. Interestingly, the numbers of IFN-γ-producing GP66-specific CD4 T cells in the spleens of FOXO3L mice were not significantly different from those in WT mice (Fig. 4C). Despite a substantial enhancement in the accumulation of functional CD8 T cells in FOXO3L mice, we observed no differences in viral control in the serum, liver, or lung at day 8 p.i. (Fig. 4D).
Next, we questioned whether FOXO3 regulated T cell responses during the chronic phase of LCMV infection (>day 8 p.i.). Strikingly, at day 30 p.i., the spleens of FOXO3L mice contained significantly higher numbers of LCMV-specific CD8 T cells than those of their WT counterparts (Fig. 5A). As on day 8 p.i. (Fig. 2), this increase in the magnitude of LCMV-specific CD8 T cells was the result of diminished apoptosis and not of increased proliferation (Fig. 5B). FOXO3 deficiency did not influence the expression of CD44, CD62L, CD27, CD122, or CD127 on the cell surface (Fig. 5C). Interestingly, however, the frequencies of the SLEC and MPEC subsets in FOXO3L mice were both increased over those of their WT counterparts (Fig. 5D), and the percentages of early effectors (EEs) (KLRG-1LO CD127LO) were reduced in FOXO3L mice. We theorize that the increased percentages of SLECs and MPECs among FOXO3L CD8 T cells might be due to reduced apoptosis of these effector subsets following differentiation from EEs. However, we cannot exclude the possibility that FOXO3 might also regulate the differentiation of effector CD8 T cells during a CVI.
Next, we enumerated the functional LCMV-specific CD8 T cells in the spleens of WT and FOXO3L mice. At 30 days p.i., we observed a significant augmentation in the number of IFN-γ-producing CD8 T cells in FOXO3L mice (Fig. 6A). As shown in Fig. 6A, only a small fraction of IFN-γ-producing CD8 T cells from WT mice were also able to produce TNF-α and IL-2; the inability to produce all three cytokines is a hallmark of functional exhaustion (45, 53). In contrast, there were substantial increases in the frequency and total number of triple-cytokine-producing cells in FOXO3L mice (Fig. 6A), suggesting that the FOXO3 deficiency might have muted the functional exhaustion of CD8 T cells. The improved functionality of CD8 T cells in FOXO3L mice was, however, not associated with alterations in the expression of the inhibitory receptors PD-1 and LAG-3 (Fig. 6B). Interestingly, levels of granzyme B in LCMV-specific CD8 T cells from FOXO3L mice were lower than those in their WT counterparts (Fig. 6C). In contrast to the enhanced CD8 T cell responses, the abundance of IFN-γ-producing GP66-specific CD4 T cells was minimally affected by FOXO3 deficiency (Fig. 6D). Notably, however, viral burdens in the circulation and livers of FOXO3L mice were significantly lower (~10-fold) than those in WT mice (Fig. 6E). Since CD4 T cell responses were minimally affected by FOXO3 deficiency, improved viral control in FOXO3L mice can be attributed primarily to increased numbers of polyfunctional LCMV-specific CD8 T cell responses. Consistent with our data, maintenance of polycytokine-producing CD8 T cells has been linked with improved HIV and hepatitis C virus (HCV) control in humans (3, 12, 17, 26, 35, 51). Collectively, the data in Fig. 5 and and66 suggested that FOXO3 regulated CD8 T cell numbers, the composition of the effector subsets, the effector function of CD8 T cells, and viral control during the chronic phase of a viral infection.
Clearance of infectious LCMV clone 13 occurs in most tissues by day 60 to 90 p.i., and it was important to investigate whether FOXO3 regulated the establishment of “CD8 T cell memory” after a chronic viral infection (1, 2). At day 120 p.i., the numbers of LCMV-specific CD8 T cells were sustained at significantly higher levels in FOXO3L mice than in WT mice (Fig. 7A). Also, note that FOXO3 deficiency did not alter homeostatic proliferation (Fig. 7B) or the expression of adhesion molecules, cytokine receptors, and differentiation/activation markers on the surfaces of LCMV-specific memory CD8 T cells (Fig. 7C). The percentages of CD62LHI (central memory) and CD27HI cells among LCMV-specific CD8 T cells in FOXO3L mice were comparable to those in WT mice, which is suggestive of normal memory differentiation. Furthermore, the expression of inhibitory receptors PD-1 and LAG-3 on memory CD8 T cells was similar in FOXO3L and WT mice (Fig. 7D). The numbers of polyfunctional LCMV-specific memory CD8 T cells were prominently elevated in the spleens of FOXO3L mice relative to those in WT mice (Fig. 7E). In contrast, the numbers of GP66-specific CD4 T cells in the spleens of FOXO3L mice were similar to those in WT mice (Fig. 7F). Taken together, the data in Fig. 7 demonstrated that FOXO3L plays a T-cell-intrinsic role in restricting the number of high-quality memory CD8 T cells following a chronic LCMV infection.
The central finding in this study is that FOXO3 limited the accumulation of virus-specific CD8 T cells during a chronic LCMV infection by inducing apoptosis and not by increasing proliferation. By virtue of the increase in the number of functional virus-specific CD8 T cells, viral control was more effective in FOXO3L mice than in WT mice during the chronic phase of LCMV infection. This idea is consistent with the report that effective viral control depends on the ratio of the number of virus-infected cells to the number of antiviral CD8 T cells (25). In FOXO3L mice, not only were LCMV-specific CD8 T cells increased in number; they displayed polyfunctionality. Therefore, the increase in the abundance of nonexhausted polycytokine-producing CD8 T cells might have contributed to improved LCMV control in FOXO3L mice (3, 12, 17, 26, 35).
Other signaling intermediates, including the suppressor of cytokine signaling 3 (SOCS3), have been implicated in downregulating the function of CD8 T cells during a CVI (38). However, we found that the levels of SOCS3 mRNA in FOXO3L CD8 T cells were only modestly lower than those in WT cells (data not shown). In agreement with our findings with FOXO3L mice, IL-7 treatment also increased the abundance of polycytokine-producing CD8 T cells and accelerated viral control in LCMV clone 13-infected mice (31, 38). We have preliminary evidence that IL-7 treatment led to increased phosphorylation of FOXO1/3 in LCMV-specific CD8 T cells in vivo (not shown), and it is likely that at least some of the effects of IL-7 on CD8 T cells are mediated by FOXO3 inactivation.
How FOXO3 regulates CD8 T cell apoptosis in vivo is unclear. FOXO3 is known to induce the expression of BIM, a proapoptotic member of the Bcl-2 family, and lower levels of BIM in FOXO3-deficient CD8 T cells might have inhibited apoptosis (15, 23, 46, 57). We found that BIM mRNA levels in FOXO3L CD8 T cells were ~2-fold lower than those in WT CD8 T cells (data not shown). Additionally, FOXO3 could promote the apoptosis of CD8 T cells by inducing the proapoptotic molecule PUMA (58). It is less likely that FOXO3 induces apoptosis by Fas ligand-dependent mechanisms, because accumulation of virus-specific CD8 T cells is minimally affected by Fas deficiency during a chronic LCMV infection (60). Nevertheless, further studies to determine how FOXO3 regulates T cell apoptosis during a CVI are warranted.
The results of this study have provided important insights into the mechanisms of T cell homeostasis by identifying the transcription factor FOXO3 as a potential negative regulator of virus-specific CD8 T cell responses during a CVI. This finding has implications for the exploration of new therapeutic avenues to enhance CD8 T cell responses by targeting FOXO3 in patients with CVIs.
We thank Ron Depinho for the generous gift of the FOXO3L mouse.
This work was supported by PHS grant AI048785 from the NIH to M. Suresh. Jeremy A. Sullivan was supported by a postdoctoral fellowship from the American Heart Association (10POST4580038).
Published ahead of print 6 June 2012