Exogenous OX40 Stimulation during Lymphocytic Choriomeningitis Virus Infection Impairs Follicular Th Cell Differentiation and Diverts CD4 T Cells into the Effector Lineage by Upregulating Blimp-1

Tobias Boettler, Youn Soo Choi, Shahram Salek-Ardakani, Yang Cheng, Friedrich Moeckel, Michael Croft, Shane Crotty and Matthias von Herrath
J Immunol November 15, 2013, 191 (10) 5026-5035; DOI: https://doi.org/10.4049/jimmunol.1300013

Abstract

T cell costimulation is a key component of adaptive immunity to viral infection but has also been associated with pathology because of excessive or altered T cell activity. We recently demonstrated that the TNFR family costimulatory molecule OX40 (CD134) is critically required to sustain antiviral T cell and Ab responses that enable control of viral replication in the context of chronic lymphocytic choriomeningitis virus (LCMV) infection. In this study, we investigated whether reinforcing OX40 stimulation through an agonist Ab had the potential to prevent LCMV persistence. We observed that anti-OX40 injection early after LCMV clone 13 infection increased CD8 T cell–mediated immunopathology. More strikingly, OX40 stimulation of virus-specific CD4 T cells promoted expression of the transcriptional repressor Blimp-1 and diverted the majority of cells away from follicular Th cell differentiation. This occurred in both acute and chronic infections, and resulted in dramatic reductions in germinal center and Ab responses to the viral infection. The effect of the OX40 agonist was dependent on IL-2 signaling and the timing of OX40 stimulation. Collectively, our data demonstrate that excessive OX40 signaling can result in deleterious consequences in the setting of LCMV infection.

Introduction

T cell responses to persistent viruses are often functionally impaired. Although this protects the host from overwhelming immunopathology, it is thought to be a contributing factor to the establishment of persistent infection (1, 2). It has been demonstrated that the enhancement of antiviral T cell responses through blockade or genetic deletion of inhibitory pathways can facilitate rapid clearance of an otherwise protracted viral infection in the murine lymphocytic choriomeningitis virus (LCMV) cl13 system (1, 3–5). More recently, the importance of immune-stimulatory pathways has been appreciated. IL-6, IL-21, and the costimulatory molecule OX40 have each been shown to be required to sustain immune system pressure on viral replication and pathogen control (6–10).

OX40 (CD134) is an inducible costimulatory receptor that belongs to the TNFR superfamily. It is primarily expressed on activated T cells, and OX40–OX40L interactions promote survival but also division and cytokine production of T cells in various settings (11). Therapeutic stimulation of the OX40 receptor through an agonistic mAb has been shown to enhance Ag-specific T cell responses in animal models, as well as in humans (12, 13). The immune-stimulating capacities of therapeutic OX40 interventions have been used to strengthen vaccine-induced T cell responses and also to promote antitumor immunity (14–16). Moreover, OX40 signaling has been suggested to be involved in the development of follicular Th cell (Tfh) responses through association with induction of CXCR5 (17–20), and the importance of humoral immune responses in controlling persistent viruses is increasingly appreciated (9, 10, 21–23). Thus, reagents that trigger OX40 signaling might constitute an interesting approach to boost cellular and humoral immunity that could combat persistent or chronic viral infection. To study the effects of exogenous OX40 stimulation in this scenario, we used the LCMV clone 13 model where high viral titers are maintained for several weeks postinfection (p.i.) of mice. Previous studies of acute or latent viruses such as vaccinia virus and CMV have shown that targeting OX40 can promote beneficial effects in both cytotoxic and helper arms of the adaptive immune response leading to curtailed viral replication (12, 24, 25). In this article, we describe the unexpected observation that augmenting OX40 signaling with an agonist Ab during the early stages of LCMV infection profoundly diverted the CD4 T cell response away from Tfh differentiation and also exacerbated CD8 T cell immunopathology. We demonstrate that agonistic OX40 signaling at an early time drives Blimp-1 expression in LCMV-specific CD4 T cells and Th1 biased CD4 T cell differentiation. Because Blimp-1 antagonizes development of Tfh, enforcing OX40 signaling above endogenous levels then becomes deleterious, severely hampering the induction of humoral immunity against LCMV.

Materials and Methods

Mice and viruses

All animals were housed at the La Jolla Institute for Allergy and Immunology vivarium under specific pathogen-free conditions. C57BL/6 mice were purchased from The Jackson Laboratory. Wild type (WT) and OX40−/− P14 CD8 TCR transgenic (TCRtg) mice (LCMV-GP33–41–specific) and WT, CD25−/−, and Blimp-1-YFP reporter Smarta CD4 TCRtg mice (LCMV-GP61–80–specific) were bred in-house on a C57BL/6 background (26, 27). LCMV infection of 5- to 8-wk-old mice was performed either i.v. with 2 × 106 PFU LCMV cl13 or i.p. with 2 × 105 PFU LCMV Armstrong or 2 × 103 PFU LCMV cl13 as indicated. A total of 10 × 105 PFU and 5 × 105 PFU were used for day 2 and 3 experiments, respectively. All experiments involving mice were reviewed and approved by the La Jolla Institute’s Animal Care Committee (AP152-MvH6).

Cell transfer

Splenocytes from TCRtg Smarta and P14 mice (LCMV-GP61–80–specific and LCMV-GP33–41–specific, respectively) on a WT background were harvested, and negative CD4 (Smarta) and CD8 (P14) isolation was performed using purified rat Abs against CD8/CD4, B220, CD11b, CD11c, I-A/I-E, and CD16/32 followed by incubation with sheep anti-rat Dynal Beads (Invitrogen). Isolation of CD25−/− Smarta cells was performed by cell sorting via FACSAria, and CD44−CD62L+CD4+CD19− cells were isolated to exclude activated cells because CD25−/− mice are prone to autoimmunity. Equal numbers of transgenic cells (5000) were then transferred into naive WT mice before LCMV infection for analysis on day 7 or later in both Armstrong and cl13 infection. For analysis of transferred cells on days 2 and 3 p.i., 1 × 106 and 0.5 × 106 Smarta cells were transferred as previously described (10, 26).

In vivo Ab treatment

Mice received a single i.p. injection of 100 μg agonistic OX40 Ab (clone OX86; Biolegend and from M.C.) 1 d p.i. or at later time points p.i. as indicated. CD8 depletion experiments were performed by i.p. administration of 500 μg anti-CD8 Ab (clone 2.43; BioXCell) on days 1 and 3 p.i. Fas ligand was blocked by i.p. administration of 200 μg anti-FasL (clone MFL4; eBioscience) on days 1 and 3 p.i.

Flow cytometry

Splenic single-cell suspensions were prepared by mashing spleen through a cell strainer into a petri dish filled with ACK-lysis buffer. After 90 s of RBC lysis, cells were washed three times and subsequently counted (countess cell counter; Invitrogen). Surface staining was performed using either fluorescently labeled or biotinylated Abs against CD8, CD4, CD19, B220, CXCR5 (26), PD-1, CD150 (SLAM), KLRG-1, CD127, FAS, IgD, CD45.1, CD45.2, and peanut agglutinin (PNA). PE- or allophycocyanin-labeled streptavidin was used to stain biotinylated Abs. PE-labeled GP66-tetramers (National Institutes of Health tetramer core facility) or GP33-pentamers (Proimmune) were used according to manufacturer’s instruction. Bcl-6 (clone K112-91) and T-bet (clone 4B10) were stained intracellularly after permeabilization with FoxP3 staining buffer set (eBioscience). To exclude dead cells, we used LIVE/DEAD (Invitrogen) viability dye.

Intracellular cytokine staining was performed after 5 h of in vitro stimulation with CD8- (GP33, GP276) and CD4-restricted (GP61) LCMV-epitopes (10 μg/ml) at 37°C in supplemented RPMI 1640 medium (Invitrogen). Permeabilization was performed using Cytofix/Cytoperm (BD) and followed by incubation with fluorescently labeled Abs against IL-2, IFN-γ, and TNF. Acquisition was done on an LSRII (BD) using DIVA software (BD), and postacquisition analysis was performed using FlowJo software (Tree Star).

Plaque assay

Vero cells (ATCC #CRL-1587) were cultured in supplemented DMEM (Invitrogen) at 37°C and plated in six-well plates. On the next day, vero cell monolayers were exposed to serum or tissue homogenate of selected mice in 1/10 dilutions and overlaid with agarose and 2× media. Five (organs) or 6 (serum) d p.i., cells were fixed, stained with crystal violet, and PFUs were counted.

Immunofluorescence

Spleen sections were embedded in optimal cutting temperature (Tissue Tek) and stored at −80°C. Samples were cut on a cryostat, and 6-μm-thick samples were fixed with acetone, blocked, and stained with CD45.1-PE and B220-FITC. After incubation, samples were washed and stained with a purified rabbit anti-PE Ab. The last staining step was performed with a goat anti-rabbit IgG-AF594 and a donkey anti-rat IgG AF488. Sections were washed and mounted with Prolong Gold antifade mounting medium with DAPI (Invitrogen).

LCMV-Ab ELISA

Coating of ELISA plates (Nunc) was performed overnight with LCMV-infected cell lysate. Wells were blocked and LCMV-specific IgG was detected in prediluted serum samples using HRP-labeled goat anti-mouse IgG (Invitrogen). SureBlue Reserve TMB Kit (KPL) was used as substrate, and absorbance was analyzed with a Spectra Max M2e (Molecular Devices).

Statistics

Statistical analyses were performed using GraphPad Prism 5 software (GraphPad). The p values were calculated using the unpaired, two-tailed Student t test. All error bars throughout the article are SD. The p values <0.05 were considered significant; *p < 0.05, **p < 0.01, ***p < 0.001.

Results

Impaired antiviral immune responses and loss of viral control after early OX40 stimulation

To analyze whether exogenous OX40 stimulation has the potential to prevent the development of T cell exhaustion and accelerate virus clearance in LCMV cl13 infection, we infected C57BL/6 mice i.v. with 2 × 106 PFU LCMV cl13. One day p.i., we i.p. administered a single injection of 100 μg agonistic OX40 Ab (clone OX86). Because LCMV cl13–infected mice experience development of severe lymphopenia caused by immunopathology, we analyzed T and B cell populations 14 d post LCMV cl13 infection in anti-OX40–treated and control animals. Endogenous OX40 signaling promotes the clonal expansion and survival of both CD4 and CD8 T cells in many immune responses, including during LCMV cl13 infection, but unexpectedly, we found that agonistic anti-OX40 treatment dramatically aggravated LCMV cl13–induced lymphopenia as shown by a profound reduction of T and B cell numbers (Fig. 1A–C). Similar to our observations on bulk T cell populations, we detected significantly lower numbers of virus-specific CD8 and CD4 T cells in anti-OX40–treated mice (Fig. 1D–F). Strikingly, although OX40 agonist treatment resulted in a transient increase of IFN-γ–producing CD4 and CD8 T cells around day 7 p.i., it caused a sustained reduction of functional effector T cells at all later times (Fig. 1E, 1F).

FIGURE 1.
FIGURE 1.

Impaired antiviral immune responses after OX40 stimulation. (A–F) C57BL/6 mice were challenged i.v. with 2 × 106 PFU/ml LCMV cl13. Mice received a single injection of 100 μg agonistic OX40 Ab (clone OX86) or rat IgG1 (placebo group) 1 d p.i. (A–C) Bar graphs show number of CD8 T cells, CD4 T cells, and B cells per spleen in naive and untreated (gray bars), infected and IgG-treated (black bars), and infected and α-OX40–treated (white bars) mice 14 d p.i. (n = 4–6 per group). (D) GP33 pentamer staining of splenocytes from IgG and α-OX40–treated mice 14 d and 70 d p.i. FACS plots are gated on CD8+CD19− cells. (E and F) Intracellular cytokine staining (ICCS) was performed on splenocytes of IgG- (black bars) and α-OX40–treated (white bars) mice on days 7, 14, 28, and 70 p.i. (n = 4–6 per group and time point) after in vitro stimulation of splenocytes with the GP33 peptide (E) and with the GP61 peptide (F). ICCS FACS plots from day 14 p.i. analyses are gated on CD8+CD19− cells (E) and CD4+CD19− cells (F). Data are derived from two independent experiments per time point (1 experiment for day 70). *p < 0.05, **p < 0.01, ***p < 0.001.

Next, we sought to analyze how the defective T cell response would affect the establishment of LCMV-specific Ab responses, because CD4 T cells are critically required for the development of humoral immunity (28). We found that OX40 stimulation heavily impaired the generation of germinal centers as characterized by the absence of Fas+ PNA+ IgD− germinal center B cells (Fig. 2A), and dramatically reduced anti-LCMV serum IgG titers by day 70 p.i. (Fig. 2B). Thus, although endogenous OX40 signals are critical for Ab responses to LCMV cl13 infection (10), enhancing OX40 engagement surprisingly had powerful negative effects on humoral immunity.

FIGURE 2.
FIGURE 2.

Impaired generation of LCMV-specific Ab responses and loss of viral control after OX40 stimulation. (A–D) C57BL/6 mice were infected with LCMV cl13 and treated with 100 μg rat IgG1 or α-OX40 1 d p.i. (A) Germinal center B cells (Fas+PNA+IgD−) were analyzed 7 and 14 d p.i. (n = 4–5 per group) in IgG- and α-OX40–treated mice. FACS plots are gated on CD19+ cells. (B) LCMV-specific IgG Ab titers were analyzed by ELISA on days 28 and 70 p.i. (n = 4–5 per group). (C) Serum and (D) kidney samples of treated and untreated mice were harvested on day 70 p.i., and viral loads were determined by plaque assay. Data are derived from a total of two to three independent experiments (one experiment for day 70). ***p < 0.001.

Because LCMV-specific T cells and Ab responses are required for the long-term control of LCMV cl13 infection, we assessed LCMV titers by plaque assay on day 70 p.i. Although control animals eventually control viral replication, LCMV titers remained high in the anti-OX40–treated groups (Fig. 2C, 2D), consistent with the defective T cell and Ab responses observed.

Delayed OX40 stimulation does not affect LCMV-specific immunity and virus control

Next, we assessed whether the effect of anti-OX40 was time dependent. We previously found that OX40 was strongly expressed on LCMV-specific CD4 T cells for at least 20 d throughout the course of LCMV cl13 infection. In contrast, OX40 was only transiently expressed on LCMV-specific CD8 T cells during the initial 5 d of infection (10). Thus, instead of treatment at day 1, anti-OX40 was given 7 d p.i. with LCMV cl13. Interestingly, our experiments revealed that delayed OX40 stimulation did not affect LCMV-specific immunity. Indeed, OX40 agonist treatment on day 7 post LCMV cl13 infection did not significantly impact the numbers of LCMV-specific IFN-γ–producing CD8 and CD4 T cells (Fig. 3A, 3B), Fas+ PNA+ IgD− germinal center B cells (Fig. 3C), or anti-LCMV serum IgG titers (Fig. 3D). Consequently, anti-OX40 treatment on day 7 did not impact viral titers in treated mice (Fig. 3E).

FIGURE 3.
FIGURE 3.

Delayed OX40 stimulation does not alter LCMV-specific immunity and viral control. (A–F) C57BL/6 mice were challenged i.v. with 2 × 106 PFU/ml LCMV cl13. Mice received a single injection of 100 μg agonistic OX40 Ab (clone OX86) or rat IgG1 (placebo group) 7 d p.i. (A and B) Intracellular cytokine staining (ICCS) was performed on splenocytes of IgG- (black bars) and α-OX40–treated (white bars) mice on days 30 and 45 p.i. (n = 4–7 per group and time point) after in vitro stimulation of splenocytes with the GP33 peptide (A) and with the GP61 peptide (B). (C) Germinal center B cells (Fas+PNA+IgD−) were analyzed 45 d p.i. (n = 5–7 per group) in IgG and α-OX40–treated mice. (D) LCMV-specific IgG Ab titers were analyzed by ELISA on day 45 p.i. (n = 5–7 per group). (E) Serum and (F) kidney samples of treated and untreated mice were harvested on days 30 and 45 p.i., and viral loads were determined by plaque assay. Data are derived from two independent experiments.

OX40 triggers T cell–mediated immunopathology

LCMV cl13–induced lymphopenia is predominantly mediated by CD8 T cell immunopathology (29, 30). Thus, we hypothesized that exacerbated lymphopenia in anti-OX40–treated mice was attributed to overstimulation of the CD8 T cell compartment. Notably, we have previously shown that OX40 is strongly expressed on LCMV-specific CD8 T cells early p.i., suggesting that they might be direct targets of the agonist Ab (10). In line with this, the more severe lymphopenia in both CD4 T cell and CD19+ B cell compartments triggered by anti-OX40 (Fig. 1) could be prevented by CD8 depletion (Fig. 4A, 4B). To address whether direct OX40 signaling causes overactivation of CD8 T cells, we cotransferred OX40+/+ and OX40−/− P14 TCR-transgenic CD8 T cells into WT hosts and analyzed CD107a expression 7 d post LCMV infection in IgG- and anti-OX40–treated mice directly ex vivo. OX40+/+ P14 cells displayed higher CD107a expression than OX40−/− P14 cells, suggesting that OX40 engagement from the endogenous ligand and from exogenous anti-OX40 treatment directly stimulated LCMV-specific CD8 T cells to degranulate, and this might have contributed to immunopathological events (Fig. 4C).

FIGURE 4.
FIGURE 4.

OX40-triggered lymphopenia is mediated by CD8 T cells. (A and B) C57BL/6 mice were infected with LCMV cl13 and treated with 100 μg rat IgG1 or α-OX40 1 d p.i. CD8 T cells were depleted by i.p. administration of 500 μg anti-CD8 Ab on days 1 and 3 p.i. CD4 T cell and B cell numbers were assessed by flow cytometry 14 d p.i. (C) Two thousand naive LCMV-GP33–41–specific cells from CD8 TCRtg (P14) OX40+/+ (WT B6, CD45.1+/+) and OX40−/− (CD45.1+CD45.2+) mice were transferred into CD45.2+/+ recipients before LCMV cl13 infection. Mice were treated with 100 μg rat IgG1 or α-OX40 1 d p.i., and CD107a expression was assessed directly ex vivo 7 d p.i. Representative FACS plot gated on live CD8+CD45.1+ cells. (D) OX40 expression on different cell subsets 2 d post LCMV cl13 infection. (E) Longitudinal analysis of Treg frequencies (CD4+CD127−FoxP3+) during LCMV cl13 infection in mice treated with IgG1 (circles) or 100 μg α-OX40 1 d p.i. (squares). (F) Expression levels of CTLA-4 (intracellular and extracellular), CD39, Neuropilin-1 and PD-L1 on Tregs 7 d p.i. in mice treated with 100 μg rat IgG1 or α-OX40 1 d p.i. (G) FoxP3 MFI on Tregs 7 d p.i. in mice treated with 100 μg rat IgG1 or α-OX40 1 d p.i. Data are derived from at least two independent experiments with four to six mice per group. **p < 0.01, ***p < 0.001.

To determine whether other cell populations might be targets of anti-OX40, we additionally analyzed OX40 expression on CD4 T cells, NK cells, dendritic cells, and neutrophils 48 h after cl13 infection (Fig. 4D). Interestingly, LCMV cl13 infection resulted in an upregulation of OX40 on regulatory T cells (Tregs; Fig. 4D). We previously published that OX40 agonist treatment can expand CD4 T cells with a regulatory phenotype that are able to prevent autoimmune diabetes in mice (31). Thus, to address the role of Tregs in anti-OX40–treated mice during LCMV infection, we longitudinally analyzed Treg numbers (Fig. 4E). Interestingly, we observed that Tregs were significantly reduced in mice treated with anti-OX40 early after LCMV cl13 infection. OX40 signals have been reported to cause Treg exhaustion that is characterized by lower levels of FoxP3 (32). To determine whether the Tregs in anti-OX40–treated mice displayed an exhausted phenotype, we analyzed several molecules that have been implicated in the suppressive activities of Tregs, such as PD-L1, CTLA-4, CD39, and Neuropilin-1 (33–35). These experiments revealed that Tregs from anti-OX40–treated mice were indistinguishable from Tregs derived from IgG-treated mice regarding FoxP3, PD-L1, CTLA-4, CD39, and Neuropilin-1 expression levels (Fig. 4F, 4G). Collectively, our findings suggest that CD8-mediated immunopathology after anti-OX40 treatment is likely a combined consequence both of direct activation of LCMV-specific CD8 T cells and reduced numbers of Tregs.

In contrast with LCMV cl13 infection, immunopathology and lymphopenia are typically not observed during LCMV Armstrong infection (4). Thus, we assessed the effects of anti-OX40 treatment after acute LCMV Arm infection. Analysis was performed 20 d p.i., because this is a time point at which global and virus-specific lymphopenia was most severe in LCMV cl13 infection (Fig. 1). Anti-OX40 did not induce lymphopenia with LCMV Arm infection (Fig. 5A, 5B). Similarly, there was minimal change in virus-specific CD8 and CD4 T cell numbers (Fig. 5C, 5D). Unexpectedly, a pronounced reduction in germinal center B cells and LCMV-specific Abs occurred in acute LCMV Arm infection (Fig. 5E, 5F), indicating that the anti-OX40–triggered loss of humoral immunity was not a consequence of immunopathology. Despite the loss of a functional germinal center response, LCMV Armstrong–infected mice treated with anti-OX40 did not display prolonged viremia, because LCMV Arm is primarily controlled by CD8 T cells (Fig. 5G).

FIGURE 5.
FIGURE 5.

Impaired humoral immunity after OX40 stimulation during LCMV-Armstrong despite absence of enhanced immunopathology. (A–G) C57BL/6 mice were infected with LCMV Armstrong and treated with 100 μg rat IgG1 or α-OX40 1 d p.i. Mice were euthanized 20 d p.i. CD4 and CD8 T cell numbers were determined by flow cytometry (A, B). (C and D) Intracellular cytokine staining (ICCS) was performed after in vitro stimulation of splenocytes with the GP33 peptide (CD8 T cells) or GP61 peptide (CD4 T cells). FACS plots are gated on CD8+CD19− cells (C) and CD4 T cells (D). (E) GC B cells were analyzed in IgG- and α-OX40–treated mice, FACS plots are gated on CD19 B cells. (F) Anti-LCMV IgG responses were analyzed by ELISA. (G) Serum samples of treated and untreated mice were harvested on day 20 p.i., and viral loads were determined by plaque assay. Data are derived from at least two independent experiments with four to six mice per group. ***p < 0.001.

Loss of follicular helper cell responses in anti-OX40–treated mice

The observation that early anti-OX40 treatment profoundly inhibited LCMV-specific humoral immunity with both acute and persistent infection was particularly surprising because OX40 deficiency results in a loss of germinal centers and anti-LCMV IgG during chronic LCMV infection (10). Given the profound reduction in germinal center B cells and LCMV-specific Ab titers observed in this study, we hypothesized that enhanced OX40 signaling negatively impacted the Tfh population. Tfh constitute a distinct CD4 T cell lineage that is committed to the establishment of germinal center responses and is characterized by the expression of Bcl-6 and CXCR5 (28, 36). Interestingly, recent work has highlighted the importance of Tfh responses in the context of persistent viral infection (9, 10, 22, 37). Thus, we analyzed the Tfh response in anti-OX40–treated mice. We stained for CXCR5 and the Tfh transcription factor Bcl-6 on day 7 p.i. on previously transferred TCRtg CD4 (Smarta) T cells and endogenous virus-specific CD4 T cells (Fig. 6A, 6B). In line with the poor Ab response, the virus-specific Bcl-6+CXCR5+ Tfh cell compartment was substantially reduced in mice receiving anti-OX40 (6.8 ± 0.7%, n = 4) compared with the control group (46.1 ± 1.6%, n = 5; Fig. 6A). Moreover, immunofluorescent staining of spleen sections revealed a dramatic reduction of transferred Smarta cells in splenic B cell zones, suggesting impaired T cell–B cell interactions in treated mice compared with control animals (Fig. 6C, 6D).

FIGURE 6.
FIGURE 6.

OX40 stimulation inhibits Tfh formation. Naive congenic CD4 T cells from TCRtg Smarta mice were transferred into C57BL/6 mice before LCMV cl13 infection and treatment with 100 μg rat IgG1 or α-OX40 1 d p.i. (A and B) Tfh responses of IgG- and α-OX40–treated mice were analyzed on congenic Smarta cells (A) and endogenous GP66-tetramer+ cells (B) by CXCR5 surface staining and intracellular staining for Bcl-6 on day 7 post LCMV cl13 infection. (C and D) Spleens were harvested on day 7 p.i. and embedded in optimal cutting temperature (OCT) compound for cryosectioning. Immunofluorescence analysis was performed on cryostat sections using Abs to B220 (for B cells) and CD45.1 (for transferred Smarta CD4 T cells) as described in Materials and Methods. Representative stainings from two independent experiments with three to five mice per group are displayed. Low magnification (original magnification ×5) in left panels, high magnification (original magnification ×20) in right panels.

Early anti-OX40 treatment drives Blimp-1 expression in LCMV-specific CD4 T cells

The transcription factors Bcl-6 and Blimp-1 are functional antagonists in CD4 T cell differentiation regulating divergence into Tfh versus effector pathways, respectively (36). We hypothesized that exogenous OX40 stimulation during LCMV infection might have driven more T cells to express Blimp-1 and then become effector Th cells rather than Tfh cells. To analyze how OX40 engagement affects Blimp-1 expression by virus-specific CD4 T cells, we transferred 5000 congenic, naive Blimp-1–YFP reporter Smarta T cells into B6 mice before LCMV cl13 infection. Significantly, Blimp-1 and the Th1-associated surface marker SLAM were strongly expressed by the vast majority of virus-specific CD4 T cells after injection of anti-OX40 (85.8 ± 0.85%, n = 7), compared with treatment with a control Ab (37.7 ± 3.88%, n = 7; Fig. 7A). Interestingly, this was also observed in acute LCMV Armstrong infection (2 × 105 PFU) given i.p., as well as low-dose LCMV cl13 infection (2 × 103 PFU i.p.), suggesting that Blimp-1 induction by OX40 is a broad phenomenon with LCMV infections not determined by the strain, inoculation dose, and route of infection (Fig. 7B). However, when we administered anti-OX40 on days 7 or 14 post LCMV infection, instead of day 1, Blimp-1 expression on virus-specific CD4 T cells was unaltered (Fig. 7C). This indicated that CD4 T cell fate commitment was perhaps already established by day 7 (38).

FIGURE 7.
FIGURE 7.

OX40 stimulation drives Blimp-1 expression. (A–E) Five thousand naive congenic CD4 T cells from TCRtg Smarta Blimp-1–YFP reporter mice were transferred into C57BL/6 mice before LCMV infection. (A and B) Blimp-1 expression and the Th1-associated surface marker SLAM (CD150) were analyzed on transferred Smarta cells 7 d p.i. in IgG- and α-OX40–treated mice. (B) The same analysis was performed in mice that have been infected i.v. with high-dose LCMV cl13, or i.p. with low-dose LCMV cl13 or LCMV Armstrong. Data are representative of two to three independent experiments with three to five mice per group. (C) A total of 100 μg IgG or α-OX40 was administered 1, 7, or 14 d post LCMV cl13 infection. Analysis of Blimp-1 expression on adoptively transferred Smarta Blimp-1–YFP cells was performed on days 7, 14, and 28, respectively. (D and E) After adoptive transfer of naive Smarta Blimp-1–YFP reporter cells and LCMV cl13 infection, mice were treated with 100 μg rat IgG1 or α-OX40 either alone or in combination with CD8 depletion or Fas ligand blockade as indicated. Transferred Smarta cells were assessed for Blimp-1 expression and Tfh markers (CXCR5, PD-1) 7 d p.i. Data are representative of two independent experiments with two to four mice per group. ***p < 0.001.

Both GC B cells and Tfh cells express high levels of the death receptor Fas (CD95). In addition, anti-OX40 treatment resulted in upregulation of Fas on both CD4 and CD8 T cells (data not shown). To evaluate whether this contributed to poor Tfh development, we administered a blocking Ab to Fas ligand. Analysis of the virus-specific CD4 T cells 7 d p.i. revealed that this did not rescue the Tfh population, suggesting that Fas-mediated depletion was not the underlying cause for the loss of Tfh cells in anti-OX40–treated mice (Fig. 7D, 7E). Similarly, we performed CD8 depletion experiments that essentially yielded the same results (Fig. 7D, 7E). These data then suggested that the loss of Tfh cells resulted from an early action of OX40 signaling leading to skewed differentiation into Th1 effectors because of Blimp-1 expression.

Altered CD4 T cell differentiation after OX40 stimulation

In response to LCMV infection, virus-specific CD4 T cells can be divided into two distinct populations, a Th1 population and a Tfh population (Fig. 8A) (39). Th1 cells are typically characterized by the expression of the transcription factor T-bet (40), the signaling leukocyte activation molecule CD150 (SLAM) (41), and secretion of IFN-γ (42), whereas Tfh cells primarily express CXCR5 and their transcription factor Bcl-6 (28). To more closely define phenotypic and functional properties of the virus-specific CD4 T cells after anti-OX40 treatment, we examined the Th1 and Tfh transcription factors T-bet and Bcl-6, respectively, as well as IFN-γ and IL-2 production. CD4 T cells from anti-OX40–treated mice were phenotypically and functionally almost indistinguishable from Th1 cells in control animals as they expressed low levels of Bcl-6 (Fig. 8B), high levels of T-bet (Fig. 8C), and produced similar amounts of IFN-γ in response to in vitro peptide stimulation (Fig. 8D). Intriguingly, IL-2 expression by the anti-OX40–treated cells was substantially higher than Th1 cells from untreated mice (Fig. 8E). Increases in IL-2 production have been observed after OX40 engagement on CD4 T cells (43). Importantly, the loss of Bcl-6 expression and upregulation of T-bet was observed on both transferred and endogenous LCMV-specific CD4 T cells in response to early anti-OX40 administration (Fig. 8F). Collectively, these results demonstrate that anti-OX40 treatment potently drives Th1 cell differentiation by upregulating Blimp-1 and T-bet expression at the expense of Tfh cell responses.

FIGURE 8.
FIGURE 8.

Altered CD4 T cell differentiation after OX40 stimulation. (A–E) Five thousand naive congenic CD4 T cells from TCRtg Smarta Blimp-1–YFP reporter mice were transferred into C57BL/6 mice before LCMV infection and analyzed 7 d p.i. T-bet and Bcl-6 expression, as well as IFN-γ and IL-2 production (by ICCS), were analyzed on transferred Smarta cells within the Th1 gate (SLAM+CXCR5−, blue) or Tfh gate (SLAM-CXCR5+, red) of IgG-treated mice and on transferred Smarta cells of α-OX40–treated mice (white). (F) Plots are gated on CD4+GP66-tetramer+ T cells. CD45.1+ cells are transferred Smarta cells, CD45.1− cells are endogenous LCMV-specific CD4 T cells. T-bet and Bcl-6 expression are identically altered in endogenous and transferred LCMV-specific CD4 T cells after OX40 stimulation. (G) Two thousand naive LCMV-GP33–41–specific cells from CD8 TCRtg mice (P14) were transferred into CD45.1/2 mismatched recipients before LCMV cl13 infection. Mice were treated with 100 μg rat IgG1 or α-OX40 1 d p.i. Bcl-6, T-bet, PD-1, SLAM, CD127, and KLRG-1 expression were analyzed 7 d post LCMV infection. Data are representative of two to three independent experiments with three to four mice per group. **p < 0.01, ***p < 0.001.

Next, we analyzed whether early OX40 stimulation would also affect the differentiation fate of LCMV-specific CD8 T cells. In contrast with the pronounced effect observed in the CD4 T cell population, LCMV-specific CD8 T cells remained unaffected after OX40 stimulation as shown by unaltered expression of Bcl-6, T-bet, PD-1, SLAM, CD127, and KLRG1 on GP33-specific CD8 T cells 7 d p.i. (Fig. 8G).

IL-2 signaling is required for OX40-mediated effector differentiation

Several factors are known to codetermine the differentiation fate of a naive CD4 T cell in response to LCMV infection. IL-2, in particular, is known to inhibit Tfh differentiation and promote effector Th cell differentiation through induction of STAT5 activation and Blimp-1 expression (26, 27). Indeed, IL-2 was dose limiting for Tfh versus Th1 differentiation (27). Given that increased IL-2 expression was observed by LCMV-specific CD4 T cells in anti-OX40–treated mice, we explored this pathway in greater detail. Tfh cell precursors can be distinguished from effector Th precursors through differential expression of Bcl-6, CXCR5, and the α unit of the IL-2R (CD25) as early as 2 d p.i. (26). Thus, we analyzed the OX40 expression levels on both populations 48 h after LCMV infection. Although both populations expressed OX40, we observed significantly higher levels on the effector Th precursors (Fig. 9A), suggesting that during CD4 T cell differentiation, OX40 may be of particular importance in the generation of non-Tfh effector responses.

FIGURE 9.
FIGURE 9.

Involvement of IL-2 in OX40-mediated Th1 effector differentiation. (A) A total of 500,000 naive congenic CD4 T cells from TCRtg Smarta Blimp-1–YFP reporter mice were transferred into C57BL/6 mice before LCMV infection and analyzed 2 d p.i. OX40 expression was analyzed on effector Th1 precursors expressing the IL-2R α-chain (CD25) and Blimp-1 and on Tfh precursors expressing CXCR5. (B) Naive congenic CD4 T cells from TCRtg Smarta mice were transferred into C57BL/6 mice before LCMV cl13 infection and treatment with 100 μg rat IgG1 or α-OX40 1 d p.i. Expression of CD25 was analyzed on CD4 Smarta cells in both groups on day 7 p.i. (C–F) Naive WT or CD25−/− CD4 T cells from TCRtg Smarta mice were transferred into congenically mismatched C57BL/6 mice before LCMV cl13 infection and treatment with 100 μg rat IgG1 or α-OX40 1 d p.i. Representative FACS plot gated on CD4 Smarta T cells. On day 7 p.i., Th1 and Tfh surface markers (C), transcription factors (D, F), as well as IFN-γ (E) were analyzed in WT and CD25−/− CD4 Smarta T cells. Data are representative for 2 independent experiments (n = 3–5 per group). *p < 0.05, **p < 0.01, ***p < 0.001.

Because OX40 signaling has the ability to enhance IL-2 production, we sought to analyze whether IL-2R signaling was involved in the loss of Tfh responses after OX40 stimulation. Notably, we found that anti-OX40 treatment resulted in upregulation of the α unit of the IL-2R on virus-specific CD4 T cells (Fig. 9B). Therefore, we isolated naive CD4 T cells from WT and IL-2Rα (CD25)–deficient Smarta transgenic mice and cotransferred those cells into WT hosts to determine the biological impact of IL-2 signaling on the altered T cell differentiation after anti-OX40 treatment. In mice receiving WT Smarta T cells, anti-OX40 treatment resulted in the loss of Tfh cells expressing Bcl-6 and CXCR5 (Fig. 9C–E). We found that the vast majority of CD25−/− Smarta T cells in control mice acquired a Tfh phenotype (CXCR5+ Bcl-6+), consistent with previous findings (27, 44). After anti-OX40 treatment, we observed an increase of Th1-associated markers T-bet and SLAM, a loss of Tfh markers CXCR5 and Bcl-6, and an increase in IFN-γ production by CD25−/− Smarta T cells. However, the effects of anti-OX40 treatment on CD25−/− cells was significantly less than for WT T cells (Fig. 9C–F), suggesting that IL-2 contributed to the altered T cell differentiation after anti-OX40 treatment. Collectively, these data suggest that exogenous OX40 stimulation enhances IL-2 expression and uses the IL-2–STAT5–Blimp-1 axis to hamper Tfh differentiation and enhance Th1 development in the context of LCMV infection.

Discussion

In summary, we demonstrate that exogenously targeting the costimulatory receptor OX40 during infection with LCMV results in skewed T cell differentiation, CD8-mediated immunopathology, and loss of LCMV-specific Ab responses. Interestingly, although the impairment of humoral immunity was observed during both acute and persistent infection, the severe immunopathology was largely restricted to persistent LCMV infection. Thus, although endogenous OX40 signaling is required to sustain LCMV-specific immune responses that eventually facilitate control of LCMV cl13 infection (10), providing further stimulation via this receptor results in deleterious consequences.

Although endogenous OX40–OX40L interactions or exogenous treatment with agonists to OX40 can promote clonal expansion and survival of CD8 T cells in a number of settings, including responses to LCMV and several other viruses and to tumors, our findings demonstrate that the lymphopenia during chronic LCMV infection after anti-OX40 treatment was a consequence of CD8-mediated immunopathology. Interestingly, OX40–OX40L interactions have previously been demonstrated to mediate immunopathology in response to respiratory influenza infection (45), suggesting that if excessive OX40 stimulation occurs in certain contexts, this can result in overexuberant CD8 T cell activity that does not favor the host. Besides direct stimulation of CD8 T cell activity, early OX40 stimulation in the context of LCMV cl13 infection also resulted in a significant reduction in the percentage of Foxp3+ Tregs. Thus, our findings suggest that CD8-mediated immunopathology in OX40-treated mice is a consequence of direct OX40 signaling on CD8 T cells, as well as a reduction of Treg frequency resulting in a failure to control the overwhelming T cell response. Recent studies have shown OX40 can influence the activity of certain inflammatory cell types other than T cells. Indeed, under certain conditions, OX40 can regulate neutrophil activity and impact the NK/NKT-plasmacytoid dendritic cell axis (11, 46, 47). Although we have not seen relevant levels of OX40 expressed on any of these cells early after LCMV cl13 infection, we cannot exclude the possibility that innate cells contributed to the immunopathology in LCMV cl13 infection after OX40 stimulation.

In contrast with the immunopathology we observed in anti-OX40–treated mice, the loss of humoral immunity could not be attributed to the action of CD8 T cells. A profound reduction of Tfh cells was observed in acute LCMV Armstrong and persistent cl13 infection, even in the absence of CD8 T cells. Our findings clearly demonstrate that anti-OX40 profoundly altered the differentiation fate of LCMV-specific CD4 T cells. There are several known pathways by which OX40 stimulation promotes the development of effector CD4 T cells (43, 48–50). One is the induction of antiapoptotic Bcl-2 family members and cell-cycle proteins that contribute to clonal expansion and survival (51, 52). Very recently it has been shown that anti-OX40 treatment can induce the upregulation of eomesodermin on CD4 T cells, which results in terminal differentiation and enhanced cell lytic capabilities of these cells in a tumor environment (53). OX40 can also promote IL-2 production and IL-2R expression (43, 49, 50). Our data now imply that Blimp-1 induction by autocrine IL-2 is another mechanism by which early and strong OX40 signaling favors effector generation. We demonstrate that OX40 stimulation resulted in a higher frequency of IL-2–secreting CD4 T cells in LCMV-infected mice, consistent with enhanced Th1 polarizing conditions, as IL-2+ Th1 cells are regarded as “multifunctional” or highly polarized Th1 cells. In addition, IL-2Ra expression on LCMV-specific CD4 T cells was dramatically increased in anti-OX40–treated animals. Given recent findings that IL-2R signaling potently interferes with Tfh development (27), we tested the importance of IL-2 signaling pathways downstream of OX40 engagement. We indeed observed that Tfh differentiation was partially restored in anti-OX40–treated mice when the virus-specific CD4 T cells were CD25−/−.

Our findings in this study are surprising in the context of the previous observation that endogenous OX40 signaling is critically required for sustaining adaptive immune responses in the context of LCMV infection (10). Antiviral T cell and Ab responses were profoundly reduced in OX40-deficient mice, and although those mice displayed reduced signs of immunopathology (because of weaker CD8 responses), they were incapable of controlling LCMV cl13 replication. Although the molecular machinery downstream of OX40 is likely similar in both cases, the dramatic effects caused by agonist targeting of OX40 in this viral infection demonstrate how tightly OX40 usage is controlled in vivo as both the absence of this molecule and enforced triggering of this molecule resulted in negative consequences for the host. Interestingly, the deleterious activity of anti-OX40 we report in this article has not been seen in other settings (14, 16, 24, 54). In the context of MCMV infection, which causes persistent viral infection with lower viremia than LCMV, early exogenous OX40 stimulation profoundly enhanced antiviral CD8 T cell responses, but immunopathology was not observed (12). It is likely that skewing toward effector Th1 differentiation and inhibition of Tfh differentiation seen in this study with LCMV requires cofactors present in the inflammatory environment when OX40 is engaged. Exactly what these may be is unknown but of obvious importance for future agonist therapeutics that might target OX40. Furthermore, the striking outcomes observed in this study help illuminate the multiple downstream signaling pathways that can be regulated by OX40. Moreover, a virulent virus-induced inflammatory environment also likely varies compared with the environment induced by nonvirulent viruses, or adjuvants used for vaccination, or, for example, in the context of tumor growth. These are all scenarios where reagents to OX40 might be clinically useful. Therapeutic ligation of OX40 using an mAb with agonistic properties has already been tested in a phase I cancer trial (ClinicalTrial.gov Identifier: NCT01303705) with no reported adverse activity (AgonOx; Providence Portland Medical Center). This reagent was also safe with no major toxicity when injected into nonhuman primates after immunization with SIV gp130, and it enhanced both T cell and Ab responses to gp130 (55). However, it will be of great importance to precisely characterize those factors that codetermine the fate of a T cell after OX40 engagement because the possible deleterious effects may constitute an obstacle for successful clinical targeting in humans whether in cancer or vaccination for infectious disease.

Disclosures

The authors have no financial conflicts of interest.

Footnotes

  • This work was supported by National Institutes of Health Grants AI089624-02 and AI068818-04 (to M.v.H.); CA91837, AI67341, and AI49453 (to M.C.); and AI072543 and AI063107 (to S.C.); and German Research Foundation (Deutsche Forschungsgemeinschaft) Research Fellowships BO 3361/1-1 and BO 3361/1-2 (to T.B.).

  • Abbreviations used in this article:

    LCMV
    lymphocytic choriomeningitis virus
    p.i.
    postinfection
    PNA
    peanut agglutinin
    TCRtg
    TCR transgenic
    Tfh
    follicular Th cell
    Treg
    regulatory T cell
    WT
    wild type.

  • Received January 7, 2013.
  • Accepted September 10, 2013.

References

  1. ↵
    1. Barber D. L.,
    2. E. J. Wherry,
    3. D. Masopust,
    4. B. Zhu,
    5. J. P. Allison,
    6. A. H. Sharpe,
    7. G. J. Freeman,
    8. R. Ahmed
    . 2006. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 439: 682–687.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Wherry E. J.,
    2. S. J. Ha,
    3. S. M. Kaech,
    4. W. N. Haining,
    5. S. Sarkar,
    6. V. Kalia,
    7. S. Subramaniam,
    8. J. N. Blattman,
    9. D. L. Barber,
    10. R. Ahmed
    . 2007. Molecular signature of CD8+ T cell exhaustion during chronic viral infection. Immunity 27: 670–684.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Tinoco R.,
    2. V. Alcalde,
    3. Y. Yang,
    4. K. Sauer,
    5. E. I. Zuniga
    . 2009. Cell-intrinsic transforming growth factor-beta signaling mediates virus-specific CD8+ T cell deletion and viral persistence in vivo. Immunity 31: 145–157.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Ejrnaes M.,
    2. C. M. Filippi,
    3. M. M. Martinic,
    4. E. M. Ling,
    5. L. M. Togher,
    6. S. Crotty,
    7. M. G. von Herrath
    . 2006. Resolution of a chronic viral infection after interleukin-10 receptor blockade. J. Exp. Med. 203: 2461–2472.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Brooks D. G.,
    2. M. J. Trifilo,
    3. K. H. Edelmann,
    4. L. Teyton,
    5. D. B. McGavern,
    6. M. B. Oldstone
    . 2006. Interleukin-10 determines viral clearance or persistence in vivo. Nat. Med. 12: 1301–1309.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Elsaesser H.,
    2. K. Sauer,
    3. D. G. Brooks
    . 2009. IL-21 is required to control chronic viral infection. Science 324: 1569–1572.
    OpenUrlAbstract/FREE Full Text
    1. Fröhlich A.,
    2. J. Kisielow,
    3. I. Schmitz,
    4. S. Freigang,
    5. A. T. Shamshiev,
    6. J. Weber,
    7. B. J. Marsland,
    8. A. Oxenius,
    9. M. Kopf
    . 2009. IL-21R on T cells is critical for sustained functionality and control of chronic viral infection. Science 324: 1576–1580.
    OpenUrlAbstract/FREE Full Text
    1. Yi J. S.,
    2. M. Du,
    3. A. J. Zajac
    . 2009. A vital role for interleukin-21 in the control of a chronic viral infection. Science 324: 1572–1576.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Harker J. A.,
    2. G. M. Lewis,
    3. L. Mack,
    4. E. I. Zuniga
    . 2011. Late interleukin-6 escalates T follicular helper cell responses and controls a chronic viral infection. Science 334: 825–829.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Boettler T.,
    2. F. Moeckel,
    3. Y. Cheng,
    4. M. Heeg,
    5. S. Salek-Ardakani,
    6. S. Crotty,
    7. M. Croft,
    8. M. G. von Herrath
    . 2012. OX40 facilitates control of a persistent virus infection. PLoS Pathog. 8: e1002913.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Croft M.
    2010. Control of immunity by the TNFR-related molecule OX40 (CD134). Annu. Rev. Immunol. 28: 57–78.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Humphreys I. R.,
    2. A. Loewendorf,
    3. C. de Trez,
    4. K. Schneider,
    5. C. A. Benedict,
    6. M. W. Munks,
    7. C. F. Ware,
    8. M. Croft
    . 2007. OX40 costimulation promotes persistence of cytomegalovirus-specific CD8 T Cells: A CD4-dependent mechanism. J. Immunol. 179: 2195–2202.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Sugamura K.,
    2. N. Ishii,
    3. A. D. Weinberg
    . 2004. Therapeutic targeting of the effector T-cell co-stimulatory molecule OX40. Nat. Rev. Immunol. 4: 420–431.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Goulding J.,
    2. V. Tahiliani,
    3. S. Salek-Ardakani
    . 2011. OX40:OX40L axis: emerging targets for improving poxvirus-based CD8(+) T-cell vaccines against respiratory viruses. Immunol. Rev. 244: 149–168.
    OpenUrlCrossRefPubMed
    1. Hirschhorn-Cymerman D.,
    2. G. A. Rizzuto,
    3. T. Merghoub,
    4. A. D. Cohen,
    5. F. Avogadri,
    6. A. M. Lesokhin,
    7. A. D. Weinberg,
    8. J. D. Wolchok,
    9. A. N. Houghton
    . 2009. OX40 engagement and chemotherapy combination provides potent antitumor immunity with concomitant regulatory T cell apoptosis. J. Exp. Med. 206: 1103–1116.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Salek-Ardakani S.,
    2. M. Moutaftsi,
    3. A. Sette,
    4. M. Croft
    . 2011. Targeting OX40 promotes lung-resident memory CD8 T cell populations that protect against respiratory poxvirus infection. J. Virol. 85: 9051–9059.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Brocker T.,
    2. A. Gulbranson-Judge,
    3. S. Flynn,
    4. M. Riedinger,
    5. C. Raykundalia,
    6. P. Lane
    . 1999. CD4 T cell traffic control: in vivo evidence that ligation of OX40 on CD4 T cells by OX40-ligand expressed on dendritic cells leads to the accumulation of CD4 T cells in B follicles. Eur. J. Immunol. 29: 1610–1616.
    OpenUrlCrossRefPubMed
    1. Walker L. S.,
    2. A. Gulbranson-Judge,
    3. S. Flynn,
    4. T. Brocker,
    5. C. Raykundalia,
    6. M. Goodall,
    7. R. Förster,
    8. M. Lipp,
    9. P. Lane
    . 1999. Compromised OX40 function in CD28-deficient mice is linked with failure to develop CXC chemokine receptor 5-positive CD4 cells and germinal centers. J. Exp. Med. 190: 1115–1122.
    OpenUrlAbstract/FREE Full Text
    1. Yu D.,
    2. C. G. Vinuesa
    . 2010. The elusive identity of T follicular helper cells. Trends Immunol. 31: 377–383.
    OpenUrlCrossRefPubMed
  20. ↵
    1. Akiba H.,
    2. H. Oshima,
    3. K. Takeda,
    4. M. Atsuta,
    5. H. Nakano,
    6. A. Nakajima,
    7. C. Nohara,
    8. H. Yagita,
    9. K. Okumura
    . 1999. CD28-independent costimulation of T cells by OX40 ligand and CD70 on activated B cells. J. Immunol. 162: 7058–7066.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    1. Bergthaler A.,
    2. L. Flatz,
    3. A. Verschoor,
    4. A. N. Hegazy,
    5. M. Holdener,
    6. K. Fink,
    7. B. Eschli,
    8. D. Merkler,
    9. R. Sommerstein,
    10. E. Horvath,
    11. et al
    . 2009. Impaired antibody response causes persistence of prototypic T cell-contained virus. PLoS Biol. 7: e1000080.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Fahey L. M.,
    2. E. B. Wilson,
    3. H. Elsaesser,
    4. C. D. Fistonich,
    5. D. B. McGavern,
    6. D. G. Brooks
    . 2011. Viral persistence redirects CD4 T cell differentiation toward T follicular helper cells. J. Exp. Med. 208: 987–999.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Streeck H.,
    2. M. P. D’Souza,
    3. D. R. Littman,
    4. S. Crotty
    . 2013. Harnessing CD4⁺ T cell responses in HIV vaccine development. Nat. Med. 19: 143–149.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Humphreys I. R.,
    2. C. de Trez,
    3. A. Kinkade,
    4. C. A. Benedict,
    5. M. Croft,
    6. C. F. Ware
    . 2007. Cytomegalovirus exploits IL-10-mediated immune regulation in the salivary glands. J. Exp. Med. 204: 1217–1225.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Salek-Ardakani S.,
    2. R. Flynn,
    3. R. Arens,
    4. H. Yagita,
    5. G. L. Smith,
    6. J. Borst,
    7. S. P. Schoenberger,
    8. M. Croft
    . 2011. The TNFR family members OX40 and CD27 link viral virulence to protective T cell vaccines in mice. J. Clin. Invest. 121: 296–307.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Choi Y. S.,
    2. R. Kageyama,
    3. D. Eto,
    4. T. C. Escobar,
    5. R. J. Johnston,
    6. L. Monticelli,
    7. C. Lao,
    8. S. Crotty
    . 2011. ICOS receptor instructs T follicular helper cell versus effector cell differentiation via induction of the transcriptional repressor Bcl6. Immunity 34: 932–946.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Johnston R. J.,
    2. Y. S. Choi,
    3. J. A. Diamond,
    4. J. A. Yang,
    5. S. Crotty
    . 2012. STAT5 is a potent negative regulator of TFH cell differentiation. J. Exp. Med. 209: 243–250.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Crotty S.
    2011. Follicular helper CD4 T cells (TFH). Annu. Rev. Immunol. 29: 621–663.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Wu-Hsieh B. A.,
    2. J. K. Whitmire,
    3. R. de Fries,
    4. J. S. Lin,
    5. M. Matloubian,
    6. R. Ahmed
    . 2001. Distinct CD8 T cell functions mediate susceptibility to histoplasmosis during chronic viral infection. J. Immunol. 167: 4566–4573.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Crotty S.,
    2. M. M. McCausland,
    3. R. D. Aubert,
    4. E. J. Wherry,
    5. R. Ahmed
    . 2006. Hypogammaglobulinemia and exacerbated CD8 T-cell-mediated immunopathology in SAP-deficient mice with chronic LCMV infection mimics human XLP disease. Blood 108: 3085–3093.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Bresson D.,
    2. G. Fousteri,
    3. Y. Manenkova,
    4. M. Croft,
    5. M. von Herrath
    . 2011. Antigen-specific prevention of type 1 diabetes in NOD mice is ameliorated by OX40 agonist treatment. J. Autoimmun. 37: 342–351.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Xiao X.,
    2. W. Gong,
    3. G. Demirci,
    4. W. Liu,
    5. S. Spoerl,
    6. X. Chu,
    7. D. K. Bishop,
    8. L. A. Turka,
    9. X. C. Li
    . 2012. New insights on OX40 in the control of T cell immunity and immune tolerance in vivo. J. Immunol. 188: 892–901.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    1. Borsellino G.,
    2. M. Kleinewietfeld,
    3. D. Di Mitri,
    4. A. Sternjak,
    5. A. Diamantini,
    6. R. Giometto,
    7. S. Höpner,
    8. D. Centonze,
    9. G. Bernardi,
    10. M. L. Dell’Acqua,
    11. et al
    . 2007. Expression of ectonucleotidase CD39 by Foxp3+ Treg cells: hydrolysis of extracellular ATP and immune suppression. Blood 110: 1225–1232.
    OpenUrlAbstract/FREE Full Text
    1. Wing K.,
    2. Y. Onishi,
    3. P. Prieto-Martin,
    4. T. Yamaguchi,
    5. M. Miyara,
    6. Z. Fehervari,
    7. T. Nomura,
    8. S. Sakaguchi
    . 2008. CTLA-4 control over Foxp3+ regulatory T cell function. Science 322: 271–275.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Francisco L. M.,
    2. V. H. Salinas,
    3. K. E. Brown,
    4. V. K. Vanguri,
    5. G. J. Freeman,
    6. V. K. Kuchroo,
    7. A. H. Sharpe
    . 2009. PD-L1 regulates the development, maintenance, and function of induced regulatory T cells. J. Exp. Med. 206: 3015–3029.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    1. Johnston R. J.,
    2. A. C. Poholek,
    3. D. DiToro,
    4. I. Yusuf,
    5. D. Eto,
    6. B. Barnett,
    7. A. L. Dent,
    8. J. Craft,
    9. S. Crotty
    . 2009. Bcl6 and Blimp-1 are reciprocal and antagonistic regulators of T follicular helper cell differentiation. Science 325: 1006–1010.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Petrovas C.,
    2. T. Yamamoto,
    3. M. Y. Gerner,
    4. K. L. Boswell,
    5. K. Wloka,
    6. E. C. Smith,
    7. D. R. Ambrozak,
    8. N. G. Sandler,
    9. K. J. Timmer,
    10. X. Sun,
    11. et al
    . 2012. CD4 T follicular helper cell dynamics during SIV infection. J. Clin. Invest. 122: 3281–3294.
    OpenUrlCrossRefPubMed
  38. ↵
    1. Choi Y. S.,
    2. J. A. Yang,
    3. I. Yusuf,
    4. R. J. Johnston,
    5. J. Greenbaum,
    6. B. Peters,
    7. S. Crotty
    . 2013. Bcl6 expressing follicular helper CD4 T cells are fate committed early and have the capacity to form memory. J. Immunol. 190: 4014–4026.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    1. Hale J. S.,
    2. B. Youngblood,
    3. D. R. Latner,
    4. A. U. Mohammed,
    5. L. Ye,
    6. R. S. Akondy,
    7. T. Wu,
    8. S. S. Iyer,
    9. R. Ahmed
    . 2013. Distinct memory CD4+ T cells with commitment to T follicular helper- and T helper 1-cell lineages are generated after acute viral infection. Immunity 38: 805–817.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Szabo S. J.,
    2. S. T. Kim,
    3. G. L. Costa,
    4. X. Zhang,
    5. C. G. Fathman,
    6. L. H. Glimcher
    . 2000. A novel transcription factor, T-bet, directs Th1 lineage commitment. Cell 100: 655–669.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Howie D.,
    2. S. Okamoto,
    3. S. Rietdijk,
    4. K. Clarke,
    5. N. Wang,
    6. C. Gullo,
    7. J. P. Bruggeman,
    8. S. Manning,
    9. A. J. Coyle,
    10. E. Greenfield,
    11. et al
    . 2002. The role of SAP in murine CD150 (SLAM)-mediated T-cell proliferation and interferon gamma production. Blood 100: 2899–2907.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Lederer J. A.,
    2. V. L. Perez,
    3. L. DesRoches,
    4. S. M. Kim,
    5. A. K. Abbas,
    6. A. H. Lichtman
    . 1996. Cytokine transcriptional events during helper T cell subset differentiation. J. Exp. Med. 184: 397–406.
    OpenUrlAbstract/FREE Full Text
  43. ↵
    1. Gramaglia I.,
    2. A. D. Weinberg,
    3. M. Lemon,
    4. M. Croft
    . 1998. Ox-40 ligand: a potent costimulatory molecule for sustaining primary CD4 T cell responses. J. Immunol. 161: 6510–6517.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Ballesteros-Tato A.,
    2. B. León,
    3. B. A. Graf,
    4. A. Moquin,
    5. P. S. Adams,
    6. F. E. Lund,
    7. T. D. Randall
    . 2012. Interleukin-2 inhibits germinal center formation by limiting T follicular helper cell differentiation. Immunity 36: 847–856.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Humphreys I. R.,
    2. G. Walzl,
    3. L. Edwards,
    4. A. Rae,
    5. S. Hill,
    6. T. Hussell
    . 2003. A critical role for OX40 in T cell-mediated immunopathology during lung viral infection. J. Exp. Med. 198: 1237–1242.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Baumann R.,
    2. S. Yousefi,
    3. D. Simon,
    4. S. Russmann,
    5. C. Mueller,
    6. H. U. Simon
    . 2004. Functional expression of CD134 by neutrophils. Eur. J. Immunol. 34: 2268–2275.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Diana J.,
    2. T. Griseri,
    3. S. Lagaye,
    4. L. Beaudoin,
    5. E. Autrusseau,
    6. A. S. Gautron,
    7. C. Tomkiewicz,
    8. A. Herbelin,
    9. R. Barouki,
    10. M. von Herrath,
    11. et al
    . 2009. NKT cell-plasmacytoid dendritic cell cooperation via OX40 controls viral infection in a tissue-specific manner. Immunity 30: 289–299.
    OpenUrlCrossRefPubMed
  48. ↵
    1. Gramaglia I.,
    2. A. Jember,
    3. S. D. Pippig,
    4. A. D. Weinberg,
    5. N. Killeen,
    6. M. Croft
    . 2000. The OX40 costimulatory receptor determines the development of CD4 memory by regulating primary clonal expansion. J. Immunol. 165: 3043–3050.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    1. Huddleston C. A.,
    2. A. D. Weinberg,
    3. D. C. Parker
    . 2006. OX40 (CD134) engagement drives differentiation of CD4+ T cells to effector cells. Eur. J. Immunol. 36: 1093–1103.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Williams C. A.,
    2. S. E. Murray,
    3. A. D. Weinberg,
    4. D. C. Parker
    . 2007. OX40-mediated differentiation to effector function requires IL-2 receptor signaling but not CD28, CD40, IL-12Rbeta2, or T-bet. J. Immunol. 178: 7694–7702.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Rogers P. R.,
    2. J. Song,
    3. I. Gramaglia,
    4. N. Killeen,
    5. M. Croft
    . 2001. OX40 promotes Bcl-xL and Bcl-2 expression and is essential for long-term survival of CD4 T cells. Immunity 15: 445–455.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Song J.,
    2. T. So,
    3. M. Cheng,
    4. X. Tang,
    5. M. Croft
    . 2005. Sustained survivin expression from OX40 costimulatory signals drives T cell clonal expansion. Immunity 22: 621–631.
    OpenUrlCrossRefPubMed
  53. ↵
    1. Hirschhorn-Cymerman D.,
    2. S. Budhu,
    3. S. Kitano,
    4. C. Liu,
    5. F. Zhao,
    6. H. Zhong,
    7. A. M. Lesokhin,
    8. F. Avogadri-Connors,
    9. J. Yuan,
    10. Y. Li,
    11. et al
    . 2012. Induction of tumoricidal function in CD4+ T cells is associated with concomitant memory and terminally differentiated phenotype. J. Exp. Med. 209: 2113–2126.
    OpenUrlAbstract/FREE Full Text
  54. ↵
    Pan, P. Y., Y. Zang, K. Weber, M. L. Meseck, and S. H. Chen. 2002. OX40 ligation enhances primary and memory cytotoxic T lymphocyte responses in an immunotherapy for hepatic colon metastases. Mol. Ther. 6: 528–536.
  55. ↵
    1. Weinberg A. D.,
    2. C. Thalhofer,
    3. N. Morris,
    4. J. M. Walker,
    5. D. Seiss,
    6. S. Wong,
    7. M. K. Axthelm,
    8. L. J. Picker,
    9. W. J. Urba
    . 2006. Anti-OX40 (CD134) administration to nonhuman primates: immunostimulatory effects and toxicokinetic study. J. Immunother. 29: 575–585.
    OpenUrlCrossRef
View Abstract
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section