Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Dec 26.
Published in final edited form as: Cell Rep. 2019 Dec 3;29(10):3293–3302.e3. doi: 10.1016/j.celrep.2019.10.119

Discovery of small molecules for the reversal of T cell exhaustion

Brett S Marro 1,2, Jaroslav Zak 1,2, Reza Beheshti Zavareh 3, John R Teijaro 1, Luke L Lairson 3, Michael BA Oldstone 1,4,*
PMCID: PMC6931400  NIHMSID: NIHMS1060479  PMID: 31801090

SUMMARY

Inhibitory receptors (IR) function as critical regulators of immune responses by tempering T cell activity. In humans, several persisting viruses as well as cancers exploit IR signaling by upregulating IR ligands, resulting in suppression of T cell function (i.e., exhaustion). This allows escape from immune surveillance and continuation of disease. Here, we report the design, implementation and results of a phenotypic high-throughput screen for molecules that modulate CD8+ T cell activity. We identified 19 compounds from the ReFRAME drug repurposing collection that restored cytokine production and enhanced the proliferation of exhausted T cells. Analysis of our top hit ingenol mebutate, a protein kinase C (PKC) inducing diterpene ester, revealed a role for this molecule in overriding the suppressive signaling cascade mediated by IR signaling on T cells. Collectively, these results demonstrate a disease-relevant methodology for identifying modulators of T cell function and reveal new targets for checkpoint blockade therapy.

Graphical Abstract

graphic file with name nihms-1060479-f0005.jpg

eTOC BLURB

Discovery of pharmacologic drugs that target exhausted T cells is essential to overcome the limitations of current checkpoint blockade therapies. Marro et al. utilize a high-throughput screening method to identify small molecule modulators of T cells and describe a role for protein kinase C in resurrecting T cell effector activity.

INTRODUCTION

Immune surveillance for recognition and removal of unwanted virus infected cells and for detection and attack of malignant cells resides primarily with the activity of cytotoxic T lymphocytes (CTLs). To counteract this response, viruses and cancers reduce the function (exhaust) CTLs (Hashimoto et al., 2018; Kahan et al., 2015). This is achieved, in part, by upregulation of inhibitory “checkpoint” receptors (IRs) on surfaces of CTLs. The importance of this strategy in controlling T cell responses is illuminated by findings that neutralizing IRs such as PD-1 or CTLA-4 on exhausted T cells restored their effector responses (Barber et al., 2006; Brooks et al., 2006; Leach et al., 1996). The use of such checkpoint inhibitory therapies has led to remarkable clinical benefits in cancer patients (Brahmer et al., 2010; Hodi et al., 2010; Robert et al., 2011; Topalian et al., 2012). Recognition of the importance of this area of research led to awarding of the 2018 Nobel prize in Physiology or Medicine for this achievement (Allison and Honjo, 2018). However, responses in many patients remain limited, in part, due to insufficient restoration of T cell function (Sharma et al., 2017). Thus, the discovery of additional targets and pharmacologic drugs is required to overcome the limitations of current checkpoint blockade (Baumeister et al., 2016). Therapeutics with distinct properties could enhance the effectiveness of existing IR blockade agents or achieve responses in patients resistant to existing treatment modalities. Several recent reports examining the synergistic effects of antibody-based blockade strategies by targeting alternative IRs, cytokines or cytokine signaling pathways have sparked numerous clinical trials (Benci et al., 2016; Budhu et al., 2017; Fan et al., 2014; West et al., 2013). Discovery and utilization of low molecular weight therapeutics can complement, and in some cases replace, existing IR blockade biologics (Gotwals et al., 2017).

One strategy to identify new T cell-modifying drugs is through phenotypic screening of chemical libraries. Several approaches to screen for small molecule modulators of T cell activation have been described (Au - Chen et al., 2019; Chen et al., 2018; Deng et al., 2018; Fouda et al., 2017). However, these methods rely on artificial activation of T cells from naïve mice via antibody stimulation with CD3/CD28 molecules rather than antigen-experienced T cells exhibiting dysfunctional effector responses.

Functional exhaustion of virus-specific T cells was first described in mice infected with the Clone 13 (CL13) variant of lymphocytic choriomeningitis virus (Barber et al., 2006; Brooks et al., 2006; Ejrnaes et al., 2006; Zajac et al., 1998). CL13 causes a persistent viral infection resulting in varying degrees of suboptimal CD4 and CD8 T cell activity, characterized by reduced to absent cytotoxic capacity of anti-viral CD8 T cells, poor proliferative potential, decreased production of antiviral effector molecules such as IFN-γ and TNF-α, insufficient expression of several homeostatic cytokines and sustained expression of IRs such as PD-1, LAG-3, TIM-3 and the immunosuppressive cytokine IL-10 (reviewed (Hashimoto et al., 2018)). T cell exhaustion is progressive and thought to be driven by persistent antigen stimulation (Mueller and Ahmed, 2009). The importance of immunosuppressive pathways that maintain T cell dysfunction was initially demonstrated by the resurrection of T cell activity following PD-1 or IL-10 receptor blockade during persistent LCMV infection (Barber et al., 2006; Brooks et al., 2008; Brooks et al., 2006). Combined blockade of PD-1 and IL-10 receptor indicated that at least two separate pathways were involved as neutralizing both receptors achieved superior enhancement of T cell function and virus clearance compared to blocking the receptors individually (Brooks et al., 2008). These findings were extended to mouse models of cancer and confirmed that checkpoint receptors maintained T cell dysfunction (Iwai et al., 2002). Here, we report utilizing the in vivo LCMV-CL13 model to construct a platform for in vitro high-throughput screening to detect small molecules that reverse T cell exhaustion.

RESULTS

In vitro restoration of T cell function from exhausted virus-specific CD8+ T cells following anti-PD-Ll and anti-LAG-3 blockade

Following infection of C57BL/6 mice with LCMV-CL13, virus-specific CD8+ T cells gradually lose their capacity to express IFN-γ (Barber et al., 2006; Brooks et al., 2006; Zajac et al., 1998). Thus, IFN-γ serves a useful disease-linked biomarker of T cell dysfunction. To both bypass the requirement of intracellular cytokine staining (ICS) to detect IFN-γ in exhausted T cells cultured ex vivo from LCMV infected animals and enhance the IFN-γ signal, we use transgenic mice on the C57BL/6 background that contain a bicistronic IRES-YFP transgene placed downstream of the translational stop codon of the IFN-γ gene (Reinhardt et al., 2015). In these IFN-γ-YFP C57BL/6 (H-2Db) mice, YFP expression is directly under the control of the promoter/enhancer elements that regulate IFN-γ transcription. Using blocking antibodies to PD-L1 and LAG-3 as a positive control for assay optimization, splenocytes from LCMV-CL13 infected (2x106 PFU i.v.) IFN-γ-YFP mice were isolated at day 15 post-infection and dispensed into 384-well plates (106 cells/mL) containing a cocktail of all known immunodominant LCMV CD8-specific (LCMV-GP33-41, LCMV-GP276-286, LCMV-NP396-404) and CD4-specific (LCMV-GP61-80) epitope peptides (Lewicki et al., 1995; Varga and Welsh, 1998) (Figure 1A). At day 15 post-LCMV infection, LCMV-specific CD8+ T cells express exhaustion-associated genes including high levels of inhibitory receptors and this expression pattern closely resembles what is observed at day 30 post-LCMV infection (Figure S1) (Doering et al., 2012; Wherry et al., 2003; Wherry et al., 2007). Following a 5-day incubation period to maximize the intracellular YFP signal, YFP fluorescence in living cells was measured by flow cytometry (Figure 1A).

Figure 1. Biologic/chemical screening strategy using the LCMV-CL13 T cell exhaustion model in IFN-γ-YFP mice.

Figure 1

Open in a new tab

(A) Screening strategy using the LCMV system to identify compounds that potentiate IFN-γ-YFP expression. (B) Frequency and number of YFP-expressing splenocytes isolated from LCMV-Arm53b infected mice at day 15 p.i.. Cells were incubated ex vivo for 5 days in the presence of LCMV-specific peptides. (C) Frequency and number of YFP+7AAD− cells from LCMV-CL13 infected mice following a 5-day incubation in the presence of LCMV-specific peptides with or without neutralizing antibodies to the inhibitory receptors PD-L1 and LAG-3. (D) Frequency of YFP-expressing splenocytes from LCMV-CL13 infected mice following a 5-day incubation ex vivo in the presence of LCMV-specific peptides and neutralizing antibodies to the inhibitory receptors PD-L1 and LAG-3. (E) YFP expression in CD4+ and CD8+ T cells with or without addition of PD-L1 antibody. (F) Intracellular detection of IFN-γ and YFP within CD8+ T cells following a 5-day incubation period. Brefeldin A was added 6 hours prior to isolating cells and staining for intracellular IFN-γ. Supernatants were isolated and IFN-γ protein was examined by homogenous time resolved fluorescence (HTRF). Data was generated from two experiments using pooled splenocyte samples from at least 5 LCMV-13 infected IFN-γ-YFP mice per experiment. A 384-well assay format was used to generate data in panels B, C, D, E and F. Statistical significance in panels B, C & E was determined using an unpaired Student’s t test. Significance in panels D and F was determined using one-way ANOVA analysis and Tukey post-hoc test. Each data point represents an individual well value. All treatment groups in Panel D were compared to untreated cells to determine significance. Bars represent average ± SD. *** p < 0.001, **** p<0.0001

To establish the maximum efficacy of the assay, IFN-γ-YFP mice were infected with the ARM53b variant of LCMV that results in an acute infection that is resolved following the expansion of anti-viral T cells that act to purge virus from infected cells. As expected, addition of LCMV peptides stimulated production of YFP while no YFP was detected in control wells that did not receive LCMV peptides (Figure 1B). In contrast, inclusion of LCMV-specific peptides in cultures containing splenocytes from LCMV-CL13 infected IFN-γ-YFP mice resulted in minimal YFP expression, ensuring that T cell effector activity was, as expected, suppressed (Figure 1C). Addition of blocking antibodies to PD-L1 and LAG-3 dramatically increased the frequency and number of YFP expressing cells compared to DMSO control wells (Figure 1C). The LAG-3 neutralizing antibody was not used as a control in subsequent experiments since anti-PD-L1 alone induced significant YFP expression compared to control wells (Figure 1D). Immunophenotyping cells after 5 days of culture revealed CD8+ T cells, and not CD4+ T cells, as the dominant cellular subset producing YFP in response to anti-PD-L1 treatment within our culture system (Figure 1E). Both CD8+ T cell-specific IFN-γ expression and IFN-γ protein within the supernatants of cultures were significantly increased following blockade of PD-L1 under the established conditions using a 384-well plate format (Figure 1F). These reproducible results demonstrate a biologically relevant in vitro screening platform for identifying compounds that restore function in exhausted CD8+ T cells.

Screening a drug repurposing library to identify compounds that resurrect T cell activity

Upon establishing a robust high throughput assay, we screened the ReFRAME drug repurposing library composed of ~12,000 compounds. This library contains purchased or resynthesized FDA-approved/registered drugs (~35%), as well as investigational new drugs (INDs) currently or previously in clinical development (~65%) (Janes et al., 2018). For primary hit identification, we employed a gating scheme to identify compounds that increase the frequency of YFP-positive cells at non-toxic concentrations, as determined based on 7-AAD staining (Figure 1B). A gating scheme based on the YFP-negative cell population was used to detect potential false positive auto-fluorescent compounds (e.g., PHA-665752 in Figure S2). A median absolute deviation (MAD) plate-based z-score for the frequency of YFP+7AAD− cells in experimental wells compared to DMSO control wells was established for each 384-well plate and distribution of experimental z-scores was compared to a normal distribution (Figure 2A). As expected, virtually all positive control wells deviated from the respective normal quantiles. Compounds that exhibited extreme z-scores but identified as auto-fluorescent based on YFP-negative populations were removed from analysis (Figure 2A).

Figure 2. Identification of small molecules that resurrect IFN-γ expression in exhausted T cells.

Figure 2

Open in a new tab

(A) Quantile-quantile (Q-Q) plot comparing the theoretical normal distribution (horizontal axis) to the experimental z-scores of all data points in the primary assay (vertical axis). Examples of auto-fluorescing compounds are labeled within the Q-Q plot. (B) Normalized MAD Z scores of the frequency and number of YFP+7AAD− cells for ReFRAME library compounds after removal of auto-fluorescent compounds. Lead hits are highlighted in red. All compounds were screened at screened at 5μM concentration. (C) Frequency of drug classes by therapeutic indication among the 58 primary hits. (D) Selection criteria that resulted in the identification of 19 lead compounds. Statistical normalization and analysis were performed using R package cellHTS2 (see Materials and methods).

Primary hits were classified as having a MAD z-score greater than 3 for the frequency YFP+7AAD− cells. In high-throughput screening (HTS) assays, a MAD score greater than 3 is is less sensitive to outliers compared to the traditional z-score methods (Chung et al., 2008). Based on these hit criteria, 247 compounds (2.1% of total) screened as single replicates within the ReFRAME library were selected as primary hits (Figure 2B). A secondary assay in which cells were treated with compound in the absence of LCMV peptides was used to categorize specific versus non-specific T cell activating compounds. Of the 247 compounds retested as triplicates, 58 (0.48% of total) demonstrated reproducibility (≥2/3 were hits) and did not induce YFP expression in the absence of LCMV-derived peptides. The most common drug classes among the hits were anti-neoplastic, anti-infective and anti-hypertensive (Figure 2C). All 58 confirmed hits were retested in 8-point dose-response format (Figure 2D). Based on the observed concentration response activity, 19 compounds were classified as lead hits (Table S1). There was an enrichment for compounds known to target toll-like receptor (TLR), Janus kinase/signal transducers and activators of transcription (JAK/STAT), intercellular-adhesion molecule 1 (ICAM-1), vitamin D receptor (VDR) and protein kinase C (PKC) signaling pathways (Table S1). The diterpenoid ingenol mebutate (IngMb) was ranked as the top hit based on its calculated primary screen z score at 0.5 μM (Table S1).

Reactivation of dysfunctional LCMV-specific CD8+ T cells by ingenol mebutate

IngMb is a diacylglycerol (DAG) analog that activates classic and novel protein kinase C (PKC) isoenzymes (Kedei et al., 2004; Liang et al., 2013; Rosen et al., 2012). When applied topically, IngMb acts as a chemotherapeutic agent by causing necrosis of tumor cells but also promotes pro-inflammatory responses through induction of NF-κB (Emmert et al., 2016; Le et al., 2009). To determine if the increased YFP signal observed following IngMb treatment was derived from LCMV-specific CD8+ T cells, splenocytes from LCMV-infected IFN-γ-YFP mice were cultured in the presence of increasing concentrations of IngMb for 5 days. Treatment with IngMb resulted in a dose dependent increase in global YFP expression compared to DMSO treated cells (EC50: ~250nM) but had no stimulatory effect without addition of LCMV peptides (Figure 3A). T cell-specific fluorescent antibody staining revealed a statistically significant increase in the frequency and number of YFP-expressing CD3+ CD8+ T cells (Figure 3A) and LCMV-specific CD8+ T cells recognizing the immunodominant (H-2Db restricted) LCMV-GP33-41 epitope (Figure 3B).

Figure 3. IngMb modulates LCMV-specific CD8+ T cell activity.

Figure 3

Open in a new tab

(A) Dose-response analysis of IngMb’s effect on the frequency of YFP+7AAD− cells and CD3+CD8+YFP+ T cells using the LCMV-CL13 exhaustion assay with IFN-γ-YFP mice. Vertical dashed red line represents the EC50 for IngMb. (A) Frequency and number of LCMV-GP33-41-specific T cells expressing YFP following incubation with IngMb for 5 days. (C) Dose-response analysis of YFP+7AAD− cells and CD3+CD8+YFP+ T cells following incubation with IngMb derivatives. (D) Number of CD8+CD44+CD90.1+ P14 T cells and live CD8+ T cells following a 5-day incubation ex vivo in the presence of IngMb and the pan-PKC inhibitor sotrastaurin concentrations of 0.02 μM, 0.5 μM, 5 μM and 10 μM. (E) Number of CD3+CD8+ T cells expressing YFP following a 5-day incubation in the presence of IngMb and ionomcyin. Representative flow plots show YFP expression following restimulation of cells with LCMV-derived peptides. (F) Intracellular detection of IFN-γ protein in proliferating CD8+CD44+ T cells following a 5-day incubation period. Splenocytes were treated with CTV following their isolation from LCMV-CL13 infected mice at day 15 p.i.. (G) Detection of selected markers on CD8+CD44+ T cells following a 5-day incubation period. Dashed horizontal lines in A and C denote the mean of DMSO or anti-PD-L1 control groups while shaded regions denote the standard deviation of DMSO or anti-PD-L1 control groups. Statistical comparison of experimental groups in B, E and G was performed using one-way ANOVA analysis with Tukey’s multiple comparison test. Statistical comparison of experimental groups in D and F was performed using two-way ANOVA analysis with Sidak’s multiple comparisons test. MFI is median fluorescence intensity. Data was generated using pooled splenocyte samples from at least 5 LCMV-13 infected IFN-γ-YFP mice per experiment. Each data point represents an individual well value. Data presented as average ± SD. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Previous structure activity-relationship (SAR) studies demonstrated the importance of the 3-angelate ester and 20-hydroxyl group of IngMb for inducing oxidative burst, cytokine release and necrosis (Liang et al., 2013). Ingenol, which lacks the 3-angelate ester, as well as a second analogue containing a methyl group at the C20 position (20-O-methyl) failed to induce YFP expression in CD3+CD8+ T cells compared to IngMb (Figure 3C). Next, we utilized the selective pan-PKC inhibitor sotrastaurin to determine if the T cell-modifying activity of IngMb could be abolished. Addition of IngMb to splenocyte cultures significantly increased the number of TCR transgenic P14 CD8+ T cells recognizing the LCMV-GP33-41 epitope that were adoptively transferred into IFN-γ-YFP mice prior to LCMV-CL13 infection (Figure 3D). There was a dose-dependent suppression in the expansion of P14 CD8+ T cells following addition of sotrastaurin to IngMb treated cells (Figure 3D). Sotrastaurin did not increase T cell death within the splenocyte cultures since the total numbers of CD8+ T cells were similar at all tested concentrations (Figure 3D). Together, these results indicate that IngMb induces a PKC-dependent reactivation of LCMV-specific CD8+ T cells following their recognition of cognant LCMV-peptide within the in vitro LCMV system.

The structurally similar phorbol ester phorbol myristate acetate (PMA) together with the calcium ionophore ionomycin are used to bypass TCR stimulation and activate PKC. The result is pro-inflammatory cytokine expression in CD8+ T cells ex vivo (Sullivan et al., 2003). PMA added together with ionomycin induced a significant increase in the YFP signal from CD8+ T cells without addition of LCMV-peptides. In contrast, PMA treatment alone required LCMV-peptides for YFP expression (Figure S3). We tested whether IngMb and ionomycin could non-specifically stimulate CD8+ T cells from LCMV-infected mice without presence of LCMV-peptides. Similar to the YFP-potentiating effect observed with PMA and ionomycin, cells treated with IngMb and ionomycin but devoid of LCMV-peptides resulted in a dramatic increase in numbers of CD8+YFP+ T cells (Figure 3E). The reduced frequency of CD8+YFP+ T cells following IngMb and ionomycin treatment in the absence of LCMV peptides was due to non-YFP expressing T cells also expanding within our culture system (Figure 3E). No increase was observed with IngMb or ionomycin treatment alone (Figure 3E). These results demonstrated the TCR-independent activation of CD8+ T cells with IngMb and ionomycin, whereas IngMb treatment alone required TCR signaling to enhance CD8+ T cell proliferation and effector functions.

YFP has a relatively long half-life and can accumulate over time whereas IFN-γ is rapidly secreted from CD8+ T cells. To determine if IngMb treatment maintained the expression of IFN-γ protein throughout the 5-day incubation period, we stained activated CD8+ T cells for intracellular IFN-γ. CellTrace Violet (CTV) labeled splenocytes from LCMV-infected mice were treated with Brefeldin A for 12 h to block IFN-γ secretion beginning at 72 h 96 h and 120 h post-peptide stimulation. A statistically significant increase in the frequency and total number of proliferating IFN-γ+CD8+CD44+ T cells was observed at each time-point relative to DMSO control wells (Figure 3F). We noted increased expression of PD-1, effector differentiation transcription factors T-BET and BLIMP1, and the CD8 memory lineage transcription factor BCL6 within CD8+CD44+ cells exposed to IngMb (Figure 3G).

We next questioned whether IngMb restored IFN-γ expression in dysfunctional CD8+ T cells or enhanced the proliferation of less-exhausted CD8+ T cells capable of producing IFN-γ. Prior to ex vivo LCMV-peptide stimulation, ~35% of LCMV-GP33-41-specific P14 CD8+ T cells from LCMV-CL13-infected mice produced IFN-γ (Figure 4A). IFN-γ-producing P14 CD8+ T cells co-expressed the inhibitory receptors PD-1 and TIM-3 relative to P14 cells that did not express IFN-γ at 48 h post-peptide stimulation (Figure 4B). Next, we monitored the proliferation of IFN-f-expressing P14 cells at 24hr, 48hr and 72hr using CTV. A majority of CD44+PD-1+IFN-γ+ P14 cells had not divided within the first 48hr of LCMV-peptide stimulation (Figure 4C). However, by 72hrs, a significant increase in the total number of dividing P14 cells exposed to either IngMb or anti-PD-L1 was observed (Figure 4C). Together, these results indicate that IngMb or anti-PD-L1 treatment targeted a small population of less exhausted LCMV-specific CD8+ T cells and enhanced their proliferation following restimulation with LCMV-specific peptides.

Figure 4. RNA-seq analysis of LCMV-GP33-41 CD8+ T cells following in vitro IngMb treatement.

Figure 4

Open in a new tab

(A) Intracellular detection of IFN-γ in Live/Dead−CD90.1+CD8+ P14 T cells at 24hr, 48hr and 72hr post-LCMV peptide stimulation in presence of IngMb (250nM) or anti-PD-L1 (25μg/ml). FACS plot represents 24-hours post-peptide stimulation. (B) Surface expression of PD-1, TIM-3 and CD44 in Live/Dead−CD90.1+CD8+ IFN-γ+ vs. Live/Dead−CD90.1+CD8+ IFN-γ− P14 T cells at 48hr post-peptide stimulation in the presence of anti-PD-L1. (C) Frequency of proliferating Live/Dead−CD44+PD-1+CD90.1+CD8+ IFN-γ+ P14 T cells at 24hr, 48hr and 72hr post-peptide stimulation. Proliferating cells were identified by reduced CTV levels. Representative gating scheme used to identify proliferating P14+ cells expressing IFN-γ or naïve CD8+ T cells that do not express IFN-γ is shown in the top panel of Figure C. (D) Differentially expressed genes within LCMV-GP33-41 CD8+ T cells following 5 days in presence of IngMb compared to vehicle control. Red labels indicate genes upregulated by IngMb (log2FC > 1; FDR-adjusted p < 0.05) while blue labels indicate downregulated genes (log2FC < 1; FDR-adjusted p < 0.05). (E) Relative expression pattern of genes in gene sets significantly upregulated by IngMb as determined by GSEA. (F) Relative expression pattern and normalized counts of chemokine receptor genes. (G) Principle component analysis of IngMb, vehicle treated samples and T cell subsets from GSE88987 (Mognol et al., 2017). (H) Comparison of IngMb, vehicle treated samples to T cell subsets from GSE70813 (Mackay et al., 2016). Panels A & C represent data from cell cultures derived from 4 independent mice. Two-way ANOVA with Tukey’s multiple comparison test was used for statistical analysis in panel C. Statistical analysis of RNA-seq data was performed using R package DESeq2. Batch correction was applied to remove batch effect-specific variance in panels G, H. Heatmap scale indicates variance from row means. Data in panels A-B presented as average ± SEM. ** p < 0.01

Lastly, we examined the transcriptional changes induced by IngMb in peptide-stimulated flow cytometry-purified LCMV-GP33-41 specific CD8+ T cells by RNA-sequencing. IngMb treatment resulted in the upregulation of 1854 genes (log2FC > 1; FDR-adjusted p < 0.05) and downregulation of 2501 genes (log2FC < −1; FDR-adjusted p < 0.05) after 5 days of culture (Figure 4D). Gene set enrichment analysis (GSEA) showed genes upregulated in effector CD8+ T cells isolated from acute LCMV infection, rather than exhausted CD8+ T cells from a chronic LCMV infection, were similarly upregulated in LCMV-GP33-41-specific CD8+ T cells treated within IngMb (Figure 4E). Genes associated with early effector lineage CD8+ T cells during acute LCMV infection, and T cell receptor signaling genes were also enriched (Figure 4E). Chemokine receptor genes Ccr2, Ccr5, Cxcr3 and Cxcr6 were among the most significantly upregulated in LCMV-specific cells following IngMb treatment while the lymph node homing receptor Ccr7 was dramatically reduced (fold change < 1/30, Figure 4F).

Although IngMb induced genes associated with effector CD8+ T cells, the IngMb treated cells did not show hallmarks of terminal effector differentiation (Gzmbhi, Tcf7Lo, Klrg1hi) and a subset of effector genes including Gzma, Gzmb and Cc14 were downregulated compared to vehicle (Figure 4E). Principle component analysis utilizing publicly available data showed a shift from exhausted to effector phenotype imparted by IngMb. Interestingly, the IngMb-treated cells exhibited greater transcriptional resemblance to memory CD8+ T cells than effector cells within this dataset (Figure 4G, PC2) (Scott-Browne et al., 2016). Comparison with an independent public dataset on memory CD8+ T cells further showed a transcriptional overlap with non-lymphoid tissue memory subsets, driven by genes such as Cxcr6, Cd7 and Cd244 (Figure 4H). IngMb-treated cells shared some markers with splenic memory subsets (Figure 4H, middle cluster) (Mackay et al., 2016). Collectively, these data suggest an induction of a distinct transcriptional signature by IngMb with features of both effector and memory phenotypes (Mackay et al., 2016; Scott-Browne et al., 2016).

DISCUSSION

We report a robust phenotypic drug discovery platform for identifying molecules that revitalize dysfunctional T cells. We employed in vitro and in vivo biology of the LCMV system, in which T cell exhaustion (Wherry et al., 2007; Zajac et al., 1998) and use of molecules to restore T cell function (Barber et al., 2006; Brooks et al., 2006) were first described and then extended to persistent infections and cancers in humans. The model we report has several distinct advantages over other systems. First, the YFP exhaustion screen allows compounds to be examined in physiologically relevant conditions that closely mimic the immunosuppressive environment occurring in vivo during T cell exhaustion. This is highlighted by our results utilizing T cells exhausted in vivo and that in vitro blockade of inhibitory molecules such as PD-L1 and LAG-3 enhanced T cell activity. Second, use of YFP as a readout of T cell effector activity is markedly superior to intracellular cytokine staining for IFN-γ since YFP has a longer half life (>24hrs), limited toxicity, is not readily secreted from cells and reduces reagent costs compared to other approaches. This allows for assessment of YFP accumulation in T cells as they functionally recover in vitro. The identification of efficacious candidate molecules in vitro can be evaluated in vivo in both the LCMV-CL13 model and in animal models of cancer.

Small molecules are attractive therapeutic options for resurrecting T cell activity since they can access intracellular targets, cover a wider distribution within tumors and can have reduced manufacturing costs compared to current antibody-based therapies (Imai and Takaoka, 2006). Our initial approach was to examine a collection of ~12,000 repurposed molecules. Using our screening method, we rapidly identified 19 lead compounds all capable of restoring IFN-γ expression in exhausted CD8+ T cells. Because the ReFRAME library comprises clinically evaluated entities, lead compounds discovered have the potential for use in patients and might not require further structure optimization.

IngMb was identified as our most potent and efficacious hit. IngMb is a naturally occurring 3-monoester of ingenol and angelic acid, and is structurally closely related to phorbol esters (Rizk et al., 1985). Phorbol esters are classic skin tumor promoters whereas ingenol derivatives appear to be weakly tumor promoting, but induce strong, but transient, pro-inflammatory responses (Furstenberger et al., 1989; Kedei et al., 2004). The distinct responses induced by these compounds are likely due to the activation of different PKC isoforms (Kedei et al., 2004).

The essential role of PKCs in regulating cell proliferation, differentiation, activation, survival and death makes them attractive therapeutic targets for cancer therapy. Indeed, topical administration of IngMb promotes the regression of mouse and human intraepidermal tumors through induction of pro-inflammatory cytokine gene expression and recruitment of lymphoid cells (Lebwohl et al., 2012; Rosen et al., 2012). IngMb was also shown to expand CD8+ T cells within a more aggressive preclinical melanoma model with tumor regression occurring at primary and distant secondary sites (Le et al., 2009).

The PKC-θ isoform is a central signaling node within the supramolecular activation complex (c-SMAC) at the immunological synapse (IS) within CD8+ T cells (Lee et al., 2010; Sheppard et al., 2004). PD-1 signaling recruits the phosphatase SHP2 to its cytoplasmic domain and disrupts PKC-θ activity through dephosphorylation and inactivation of CD28 (Hui et al., 2017; Kong et al., 2011; Sheppard et al., 2004; Yokosuka et al., 2008). Our results indicate that IngMb is capable of bypassing the suppressive signaling cascade mediated by PD-1 to induce a burst of proliferation and IFN-γ expression in exhausted LCMV-specific CD8+ T cells in vitro. Since the activity of IngMb on LCMV-specific CD8+ T cells required LCMV peptides, an intact IS complex was required for downstream signaling. Transcriptional profiling of LCMV-specific CD8+ T cells treated with IngMb revealed a gene signature with similarities to both a memory T cell pool observed following clearance of LCMV (Scott-Browne et al., 2016) as well as tissue-resident memory T cell subsets (Mackay et al., 2016). Mice with defective TCR proximal signaling that fail to recruit PKC to the immunological synapse show deficiencies in memory T cell formation (Teixeiro et al., 2009).

In summary, our results demonstrate an effective methodology for identifying novel modulators of T cell function and describe a direct role for IngMb and PKC in resurrecting antigen-specific T cell effector activity within chronic infection and cancer models. It is still unknown whether IngMb enhances the long-term survival of exhausted CD8+ T cells and what subpopulations of CD8+ T cells may be more responsive to IngMb. Our current investigations focus on which PKC isozymes within exhausted CD8+ T cells are most responsive to IngMb and the effects of IngMb in vivo to resurrect CD8+ T cell function on elimination/control of persistent LCMV infection and elimination of tumor cells within the B16 melanoma model. Given that IngMb induced a transcriptional signature in CD8+ T cells that shared similarities with memory T cells, we envision IngMb as a potential late-stage intratumoral therapy designed to act synergistically with current checkpoint-blockade therapies. A population of memory-like T cells has recently shown to be highly responsive to anti-PD-L1 treatment within the LCMV-system and preclinical cancer models (He et al., 2016; Im et al., 2016; Kurtulus et al., 2019; Siddiqui et al., 2019). In addition, IngMb may also be useful for ex vivo reactivation and expansion of chimeric antigen receptor (CAR) T cells for use in solid tumor cancers where T cell exhaustion may limit CAR T cell effectiveness.

LEAD CONTACT AND MATERIALS AVAILABILITY

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact Michael B.A. Oldstone (mbaobo@scripps.edu). This study did not generate new unique reagents.

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Mice

Seven to 8 week old IFN-γ-YFP male mice on the C57BL/6 (H-2b) background were used. Mice were bred and maintained in pathogen-free conditions in the animal facility at The Scripps Research Institute (TSRI). All procedures involving mice were in accordance with TSRI Animal Research Committee guidelines.

Virus

Viral stocks of LCMV-Clone13 were generated by infecting BHK-21 monolayers at a multiplicity of infection (MOI) of 0.1 with a plaque-purified parental stock. Supernatants were isolated following a 60 hour incubation and virus titers measured by plaque assay on Vero cell monolayers (Lewicki et al., 1995; Oldstone et al., 1991). Stock viruses were titered to 4-5x107 PFU/ml. Mice were injected with 2x106 PFU LCMV i.v. (Brooks et al., 2008; Brooks et al., 2006; Lewicki et al., 1995).

METHOD DETAILS

YFP Exhaustion Screening Assay

Serum titers from LCMV-CL13 infected IFN-γ-YFP mice were measured by plaque assay at day 12 p.i. to confirm a productive infection. At day 15 p.i., spleens were harvested, digested and single cell suspensions prepared using a mixture of collagenase/Dnase (Roche) prior to homogenation on a 100 μM filter using a butt-end of a syringe. RBCs were lysed for 2 minutes per spleen in 1X RBC lysis buffer. Following RBC lysis, B cells were depleted by magnetic bead separation using a CD19-positive selection II kit (EasySep). Splenocytes were counted and resuspended to 1x106 cells/ml in complete T cell media (10% FBS, 1% PenStrep, 1% L-Glutamine, NEAA, Sodium Pyruvate, HEPES, 50 μM BME) supplemented with 2 μg/ml LCMV-specific CD8 peptides (GP33-41, NP396-404 and GP276-286) and 5 μg/ml CD4 peptide (GP61-80). Next, 50 μl cells were seeded into 384-well flat-clear bottom TC-treated plates (Greiner, Cat #781090) that were pre-spotted using an Echo Liquid Handler (Labcyte) with DMSO or ReFRAME compounds at 5 μM final concentration. Plates were placed at 37°c + 5% CO2 at a 20° angle to increase cell to cell contact. Following a 5-day incubation period, 7-AAD was added to each well (1:50 dilution) and plates were rested for 15 minutes. Cells were then analyzed on a ZE5 flow cytometer (Bio-Rad). To track GP-specific CD8+ T cells in vitro, 1,000 congenic Thy1.1+ CD8+ T cells (P14) from TCR tg mice that recognize the LCMV GP33-41 epitope (Pircher et al., 1989) were transferred into IFN-γ-YFP 1 day prior to LCMV infection.

Ex vivo intracellular staining

Splenocytes from LCMV-CL13 infected IFN-γ-YFP mice were seeded onto round-bottom 96 wells plates at 1x106 cells/well in complete T cell media (10% FBS, L-glutamine, Pen/Strep, NEAA, Sodium Pyruvate, HEPES, BME) supplemented with LCMV-specific peptides. Brefeldin A (Thermo Fisher, Cat #: B7450) was added at a 1:500 dilution to each well 6 hours prior to cell isolation to block IFN-γ secretion and plates were incubated overnight. Surface antigens were then stained for 30 minutes on ice, cells fixed and permeabilized using Cytofix/Cytoperm (BD Pharmingen, Cat #: 554714) and stained for intracellular proteins.

Flow Cytometry

LIVE/DEAD fixable blue (ThermoFisher, Cat #: L23962), Rat anti-mouse T-BET (Biolegend, 644806, 1:100), BLIMP1 (Biolegend, Cat #: 150008, 1:100), PD-1 (Biolegend, Cat #: 135219, 1:100), BCL6 (Biolegend, Cat #: 358512, 1:200), CD8 (Biolegend, Cat #: 100766, 1:200), CD44 (Biolegend, Cat #: 103057, 1:500), IFN-γ (Biolegend, Cat #: 505838, 1:400), TIM-3 (Biolegend, Cat #: 134013, 1:100) and TCF1 (BD Pharmingen, Cat #: 564217, 1:200) were utilized.

Homogenous Time Resolved Fluorescence (HTRF) Assay

A total of 16μl was isolated from each well of a 384-well plate was placed into a 384-well small volume microplate (Greiner). 4μl of pre-mixed HTRF reagents (Cisbio) were added to each well. Plates were then scanned using a CLARIOstar microplate reader (BMG LabTech). An HTRF ratio was calculated using the following formula: (signal 665nm)/(signal 620nm) x 104.

RNA isolation and sequencing

Splenocyte cultures from CL13 infected mice were obtained as described above. Cells were treated with IngMb mebutate (250 μM) or medium for 5 days, then LCMV-GP33-41 peptide-specific CD8+ T cells isolated from the cultures by flow cytometry and subjected to RNA isolation. RNA amplification was performed using SMART-Seq Ultra Low Input RNA kit v4 (TakaraBio, Mountain View, CA) and cDNA libraries sequenced using single-end 75 bp protocol on NextSeq500 (Illumina, San Diego, CA) to a median depth of 20 million reads. The data have been deposited to GEO with the accession number GSE130406.

RNA-sequencing analysis

Reads were mapped to the mouse genome and genic reads quantified using STAR version 2.5.2a (Dobin et al., 2013) and Ensembl version 87 GRCm38 genome and transcriptome annotations. Normalization, differential expression analysis, principal component analysis were performed using R package DESeq2 version 1.22.1 (Love et al., 2014); heatmaps were constructed using R package pheatmap version 1.0.10. Gene counts for publicly available studies GSE70813 and GSE88987 were obtained using ARCHS4 (Lachmann et al., 2018). Prior to comparison of these publicly available datasets to the RNA-seq dataset generated for this study, batch correction was performed using the R function removeBatchEffect from the package limma (Ritchie et al., 2015). Analyses were performed using R version 3.5.2 (Team, 2008).

Microarray analysis

Publicly available microarray data were retrieved using R package GEOquery version 2.52.0 (Davis and Meltzer, 2007), subjected to log transformation and heatmap plotted using R package pheatmap version 1.0.12 (Kolde, 2013).

QUANTIFICATION AND STATISTICAL ANALYSIS

All quantitative data collected from experiments is expressed as mean ± SD or SEM as indicated in the figure legend. Differences between groups were assayed with the statistical test indicated in the figure legend using GraphPad Prism. Significant differences were noted when p ≥ 0.05. Raw readings of percent YFP+7AAD− cells from ReFRAME library were subjected to plate effect normalization and by-plate variance adjustment using the R package cellHTS2 (Boutros et al., 2006). Median-adjusted MAD z-scores were then calculated using cellHTS2. For hit identification, we first selected wells having a MAD z-score ≥ −3 for percent YFP- and a MAD z-score ≥ 3 for % YFP+7AAD− (to identify compounds that significantly increase the YFP signal). Statistical analysis was performed using cellHTS2 version 2.46.0 and R version 3.5.2.

DATA AND CODE AVAILABILITY

The accession number for the RNA sequence data reported in this paper is “GSE130406.”

STUDY APPROVAL

All procedures involving mice were reviewed and approved with TSRI Animal Care and Use Committee.

Supplementary Material

1

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Antibodies
PE/Cy7 rat anti-mouse Bcl-6 Biolegend (BioLegend Cat# 358512, RRID:AB_2566196)
APC rat anti-mouse Blimp-1 Biolegend (BioLegend Cat# 150008, RRID:AB_2728187)
Brilliant Violet 711 rat anti-mouse CD44 Biolegend (BioLegend Cat# 103057, RRID:AB_2564214)
APC/Fire 750 rat anti-mouse CD8α Biolegend (BioLegend Cat# 100766, RRID:AB_2572113)
Brilliant Violet 785 rat anti-mouse IFN-γ Biolegend (BioLegend Cat# 505838, RRID:AB_2629667)
Brilliant Violet 605 rat anti-mouse PD-1 Biolegend (BioLegend Cat# 135219, RRID:AB_11125371)
PerCP/Cyanine5.5 mouse anti-mouse T-Bet Biolegend (BioLegend Cat# 644806, RRID:AB_1595488)
PE mouse anti-mouse TCF-1 BD Biosciences (BD Biosciences Cat# 564217, RRID:AB_2687845)
PE/Dazzle 594 rat anti-mouse Tim-3 Biolegend (BioLegend Cat# 134013, RRID:AB_2632737)
Biotinylated DbGP33-41 monomer NIH Tetramer Core N/A
InVivoMAb Anti-mouse PD-L1 BioXCell Cat#: BE0101
InVivoMAb Anti-mouse LAG-3 BioXCell Cat#: BE0174
Bacterial and Virus Strains
LCMV (Arm53b) Oldstone Laboratory N/A, generated in house
LCMV (Clone 13) Oldstone Laboratory N/A, generated in house
Chemicals, Peptides, and Recombinant Proteins
Brefeldin A ThermoFisher Scientific Cat#: B7450
CellTrace Violet Cell Proliferation Kit ThermoFisher Scientific Cat#: C34557
Collagenase D Roche Cat#: 11088866001
Ingenol-3-angelate Cayman Cat#: 16207
Ingenol Cayman Cat#: 14031
Ionomycin calcium salt Sigma-Aldrich Cat#: 10634
Phorbol 12-myristate 13-acetate Sigma-Aldrich Cat#: P8139
20-O-Methyl Parker Laboratory, Scripps Research N/A
RPMI 1640 ThermoFisher Scientific Cat#: 21870076
FBS Gemini Bio-Products Cat#: 100-106
LCMV GP33-41 Recombinant Peptide Anaspec Cat#: AS61296
LCMV NP396-404 Recombinant Peptide Anaspec Cat#: AS61700
LCMV GP276-286 Recombinant Peptide Anaspec Cat#: AS62539
LCMV GP61-80 Recombinant Peptide Anaspec Cat#: AS64560
Critical Commercial Assays
EasySep Mouse CD19 Positive Selection Kit II Stemcell Technologies Cat#: 18954
Fixation/Permeabilization Solution Kit BD Biosciences Cat#: 554714
FoxP3/Transcription Factor Staining Buffer Set ThermoFisher Scientific Cat#: 00-5523-00
LIVE/DEAD Fixable Blue Dead Cell Stain Kit ThermoFisher Scientific Cat#: L34962
Mouse IFN gamma HTRF Kit Cisbio Cat#: 62MIFNGPEG
SMART-Seq v4 Ultra Low Input RNA Kit for Sequencing Takara Cat#: 634888
Deposited Data
Raw and processed RNA-sequencing data NCBI GEO GSE130406
Experimental Models: Cell Lines
Vero-E6 Oldstone Laboratory Scripps Research
Experimental Models: Organisms/Strains
C57BL/6 TSRI N/A
Thy1.1P14 TSRI N/A
C. 129S4(B6)-Ifngtm3.1Lky/J (IFN-γ-YFP) Jackson Laboratory Cat#: 017580
Software and Algorithms
Flowjo Software (Build #: 10.6.1) FlowJo https://flowjo.com
GraphPad Prism 6 GraphPad Software https://graphpad.com
R version 3.5.2 (Team, 2008) http://www.R-project.org
STAR version 2.5.2a (Dobin et al., 2013) https://github.com/alexdobin/STAR
cellHTS2 R package version 2.48.0 (Boutros et al., 2006) http://bioconductor.org/packages/release/bioc/html/cellHTS2.html
DESeq2 R package version 1.22.1 (Love et al., 2014) https://bioconductor.org/packages/release/bioc/html/DESeq2.html
Pheatmap R package version 1.0.10 Raivo Kolde https://cran.r-project.org/web/packages/pheatmap/index.html
Limma R package version 3.40.6 (Ritchie et al., 2015) https://bioconductor.org/packages/release/bioc/html/limma.html
GEOquery R package version 2.52.0 (Davis and Meltzer, 2007) https://bioconductor.org/packages/release/bioc/html/GEOquery.html
ARCHS4 (Lachmann et al., 2018) https://amp.pharm.mssm.edu/archs4/
Open in a new tab

HIGHLIGHTS.

  • Physiologically relevant high-throughput assay for modulators of T cell exhaustion

  • Known and novel immunomodulatory compounds identified that rescue cytokine production

  • Ingenol mebutate targets protein kinase C to reactivate hypofunctional CD8 T cells

ACKNOWLEDGEMENTS

This work was supported by NIH grants R01 AI099699 & AI09484 (M.B.A.O.), R01 AI123210 & AI118862 (J.R.T), R21 AI141842 (L.L.L), Jeanette Bertea Hennings Foundation fellowship award (B.S.M).

Footnotes

DECLARATION OF INTERESTS

The authors have declared that no competing interests exist.

REFERENCES:

  1. Allison JP, and Honjo T (2018). The Nobel Prize in Physiology or Medicine 2018. NobelPrize.org. Nobel Media AB 2019. Thu. 31 Jan 2019 <https://www.nobelprize.org/prizes/medicine/2018/summary/>.
  2. Au - Chen EW, Au - Ke CY, Au - Brzostek J, Au - Gascoigne NRJ, and Au - Rybakin V (2019). Identification of Mediators of T-cell Receptor Signaling via the Screening of Chemical Inhibitor Libraries. JoVE, e58946. [DOI] [PubMed] [Google Scholar]
  3. Barber DL, Wherry EJ, Masopust D, Zhu B, Allison JP, Sharpe AH, Freeman GJ, and Ahmed R (2006). Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 439, 682–687. [DOI] [PubMed] [Google Scholar]
  4. Baumeister SH, Freeman GJ, Dranoff G, and Sharpe AH (2016). Coinhibitory Pathways in Immunotherapy for Cancer. Annu Rev Immunol 34, 539–573. [DOI] [PubMed] [Google Scholar]
  5. Benci JL, Xu B, Qiu Y, Wu TJ, Dada H, Twyman-Saint Victor C, Cucolo L, Lee DSM, Pauken KE, Huang AC, et al. (2016). Tumor Interferon Signaling Regulates a Multigenic Resistance Program to Immune Checkpoint Blockade. Cell 167, 1540–1554 e1512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Boutros M, Brás LP, and Huber W (2006). Analysis of cell-based RNAi screens. Genome Biology 7, R66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brahmer JR, Drake CG, Wollner I, Powderly JD, Picus J, Sharfman WH, Stankevich E, Pons A, Salay TM, McMiller TL, et al. (2010). Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates. J Clin Oncol 28, 3167–3175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Brooks DG, Ha SJ, Elsaesser H, Sharpe AH, Freeman GJ, and Oldstone MB (2008). IL-10 and PD-L1 operate through distinct pathways to suppress T-cell activity during persistent viral infection. Proc Natl Acad Sci U S A 105, 20428–20433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brooks DG, Trifilo MJ, Edelmann KH, Teyton L, McGavern DB, and Oldstone MB (2006). Interleukin-10 determines viral clearance or persistence in vivo. Nat Med 12, 1301–1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Budhu S, Schaer DA, Li Y, Toledo-Crow R, Panageas K, Yang X, Zhong H, Houghton AN, Silverstein SC, Merghoub Tv et al. (2017). Blockade of surface-bound TGF-beta on regulatory T cells abrogates suppression of effector T cell function in the tumor microenvironment. Sci Signal 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chen EW, Brzostek J, Gascoigne NRJ, and Rybakin V (2018). Development of a screening strategy for new modulators of T cell receptor signaling and T cell activation. Scientific Reports 8, 10046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chung N, Zhang XD, Kreamer A, Locco L, Kuan PF, Bartz S, Linsley PS, Ferrer M, and Strulovici B (2008). Median absolute deviation to improve hit selection for genome-scale RNAi screens. J Biomol Screen 13, 149–158. [DOI] [PubMed] [Google Scholar]
  13. Davis S, and Meltzer PS (2007). GEOquery: a bridge between the Gene Expression Omnibus (GEO) and BioConductor. Bioinformatics 23, 1846–1847. [DOI] [PubMed] [Google Scholar]
  14. Deng J, Wang ES, Jenkins RW, Li S, Dries R, Yates K, Chhabra S, Huang W, Liu H, Aref AR, et al. (2018). CDK4/6 Inhibition Augments Antitumor Immunity by Enhancing T-cell Activation. Cancer Discov 8, 216–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, and Gingeras TR (2013). STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Doering TA, Crawford A, Angelosanto JM, Paley MA, Ziegler CG, and Wherry EJ (2012). Network analysis reveals centrally connected genes and pathways involved in CD8+ T cell exhaustion versus memory. Immunity 37, 1130–1144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ejrnaes M, Filippi CM, Martinic MM, Ling EM, Togher LM, Crotty S, and von Herrath MG (2006). Resolution of a chronic viral infection after interleukin-10 receptor blockade. J Exp Med 203, 2461–2472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Emmert S, Haenssle HA, Zibert JR, Schon M, Hald A, Hansen MH, Litman T, and Schon MP (2016). Tumor-Preferential Induction of Immune Responses and Epidermal Cell Death in Actinic Keratoses by Ingenol Mebutate. PLoS One 11, e0160096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fan X, Quezada SA, Sepulveda MA, Sharma P, and Allison JP (2014). Engagement of the ICOS pathway markedly enhances efficacy of CTLA-4 blockade in cancer immunotherapy. J Exp Med 211, 715–725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Fouda A, Tahsini M, Khodayarian F, Al-Nafisah F, and Rafei M (2017). A Fluorescence-based Lymphocyte Assay Suitable for High-throughput Screening of Small Molecules. Journal of visualized experiments : JoVE. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Furstenberger G, Schurich B, Kaina B, Petrusevska RT, Fusenig NE, and Marks F (1989). Tumor induction in initiated mouse skin by phorbol esters and methyl methanesulfonate: correlation between chromosomal damage and conversion (‘stage I of tumor promotion’) in vivo. Carcinogenesis 10, 749–752. [DOI] [PubMed] [Google Scholar]
  22. Gotwals P, Cameron S, Cipolletta D, Cremasco V, Crystal A, Hewes B, Mueller B, Quaratino S, Sabatos-Peyton C, Petruzzelli Lv et al. (2017). Prospects for combining targeted and conventional cancer therapy with immunotherapy. Nature Reviews Cancer 17, 286. [DOI] [PubMed] [Google Scholar]
  23. Hashimoto M, Kamphorst AO, Im SJ, Kissick HT, Pillai RN, Ramalingam SS, Araki K, and Ahmed R (2018). CD8 T Cell Exhaustion in Chronic Infection and Cancer: Opportunities for Interventions. Annu Rev Med 69, 301–318. [DOI] [PubMed] [Google Scholar]
  24. He R, Hou S, Liu C, Zhang A, Bai Q, Han M, Yang Y, Wei G, Shen T, Yang X, et al. (2016). Follicular CXCR5-expressing CD8(+) T cells curtail chronic viral infection. Nature 537, 412–428. [DOI] [PubMed] [Google Scholar]
  25. Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, Gonzalez R, Robert C, Schadendorf D, Hassel JC, et al. (2010). Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med 363, 711–723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hui E, Cheung J, Zhu J, Su X, Taylor MJ, Wallweber HA, Sasmal DK, Huang J, Kim JM, Mellman I, et al. (2017). T cell costimulatory receptor CD28 is a primary target for PD-1-mediated inhibition. Science 355, 1428–1433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Im SJ, Hashimoto M, Gerner MY, Lee J, Kissick HT, Burger MC, Shan Q, Hale JS, Lee J, Nasti TH, et al. (2016). Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature 537, 417–421. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Imai K, and Takaoka A (2006). Comparing antibody and small-molecule therapies for cancer. Nat Rev Cancer 6, 714–727. [DOI] [PubMed] [Google Scholar]
  29. Iwai Y, Ishida M, Tanaka Y, Okazaki T, Honjo T, and Minato N (2002). Involvement of PD-L1 on tumor cells in the escape from host immune system and tumor immunotherapy by PD-L1 blockade. Proc Natl Acad Sci U S A 99, 12293–12297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Janes J, Young ME, Chen E, Rogers NH, Burgstaller-Muehlbacher S, Hughes LD, Love MS, Hull MV, Kuhen KL, Woods AK, et al. (2018). The ReFRAME library as a comprehensive drug repurposing library and its application to the treatment of cryptosporidiosis. Proc Natl Acad Sci U S A 115 10750–10755. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kahan SM, Wherry EJ, and Zajac AJ (2015). T cell exhaustion during persistent viral infections. Virology 479-480 180–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kedei N, Lundberg DJ, Toth A, Welburn P, Garfield SH, and Blumberg PM (2004). Characterization of the interaction of ingenol 3-angelate with protein kinase C. Cancer Res 64 3243–3255. [DOI] [PubMed] [Google Scholar]
  33. Kolde R (2013). pheatmap: Pretty Heatmaps. .
  34. Kong KF, Yokosuka T, Canonigo-Balancio AJ, Isakov N, Saito T, and Altman A (2011). A motif in the V3 domain of the kinase PKC-theta determines its localization in the immunological synapse and functions in T cells via association with CD28. Nat Immunol 12 1105–1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kurtulus S, Madi A, Escobar G, Klapholz M, Nyman J, Christian E, Pawlak M, Dionne D, Xia J, Rozenblatt-Rosen O, et al. (2019). Checkpoint Blockade Immunotherapy Induces Dynamic Changes in PD-1(−)CD8(+) Tumor-Infiltrating T Cells. Immunity 50 181–194 e186 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Lachmann A, Torre D, Keenan AB, Jagodnik KM, Lee HJ, Wang L, Silverstein MC, and Ma’ayan A (2018). Massive mining of publicly available RNA-seq data from human and mouse. Nature Communications 9, 1366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Le TT, Gardner J, Hoang-Le D, Schmidt CW, MacDonald KP, Lambley E, Schroder WA, Ogbourne SM, and Suhrbier A (2009). Immunostimulatory cancer chemotherapy using local ingenol-3-angelate and synergy with immunotherapies. Vaccine 27 3053–3062. [DOI] [PubMed] [Google Scholar]
  38. Leach DR, Krummel MF, and Allison JP (1996). Enhancement of antitumor immunity by CTLA-4 blockade. Science 271 1734–1736. [DOI] [PubMed] [Google Scholar]
  39. Lebwohl M, Swanson N, Anderson LL, Melgaard A, Xu Z, and Berman B (2012). Ingenol mebutate gel for actinic keratosis. N Engl J Med 366, 1010–1019. [DOI] [PubMed] [Google Scholar]
  40. Lee WY, Hampson P, Coulthard L, Ali F, Salmon M, Lord JM, and Scheel-Toellner D (2010). Novel antileukemic compound ingenol 3-angelate inhibits T cell apoptosis by activating protein kinase Ctheta. J Biol Chem 285 23889–23898. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lewicki H, Tishon A, Borrow P, Evans CF, Gairin JE, Hahn KM, Jewell DA, Wilson IA, and Oldstone MB (1995). CTL escape viral variants. I. Generation and molecular characterization. Virology 210 29–40. [DOI] [PubMed] [Google Scholar]
  42. Liang X, Grue-Sorensen G, Mansson K, Vedso P, Soor A, Stahlhut M, Bertelsen M, Engell KM, and Hogberg T (2013). Syntheses, biological evaluation and SAR of ingenol mebutate analogues for treatment of actinic keratosis and non-melanoma skin cancer. Bioorg Med Chem Lett 23 5624–5629. [DOI] [PubMed] [Google Scholar]
  43. Love MI, Huber W, and Anders S (2014). Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 15 1–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Mackay LK, Minnich M, Kragten NA, Liao Y, Nota B, Seillet C, Zaid A, Man K, Preston S, Freestone D, et al. (2016). Hobit and Blimp1 instruct a universal transcriptional program of tissue residency in lymphocytes. Science 352 459–463. [DOI] [PubMed] [Google Scholar]
  45. Mognol GP, Spreafico R, Wong V, Scott-Browne JP, Togher S, Hoffmann A, Hogan PG, Rao A, and Trifari S (2017). Exhaustion-associated regulatory regions in CD8+ tumor-infiltrating T cells. Proceedings of the National Academy of Sciences 114 E2776–E2785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Mueller SN, and Ahmed R (2009). High antigen levels are the cause of T cell exhaustion during chronic viral infection. Proceedings of the National Academy of Sciences 106 8623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Oldstone MB, Nerenberg M, Southern P, Price J, and Lewicki H (1991). Virus infection triggers insulin-dependent diabetes mellitus in a transgenic model: role of anti-self (virus) immune response. Cell 65 319–331. [DOI] [PubMed] [Google Scholar]
  48. Pircher H, Burki K, Lang R, Hengartner H, and Zinkernagel RM (1989). Tolerance induction in double specific T-cell receptor transgenic mice varies with antigen. Nature 342 559–561. [DOI] [PubMed] [Google Scholar]
  49. Reinhardt RL, Liang HE, Bao K, Price AE, Mohrs M, Kelly BL, and Locksley RM (2015). A novel model for IFN-gamma-mediated autoinflammatory syndromes. J Immunol 194 2358–2368. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Ritchie ME, Smyth GK, Phipson B, Wu D, Hu Y, Shi W, and Law CW (2015). limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Research 43, e47–e47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Rizk AM, Hammouda FM, Elmissiry MM, Radwan HM, and Evans FJ (1985). Constitutents Egyptian Euphorbiaceae .13. Biologically-Active Diterpene Esters from Euphorbia-Peplus. Phytochemistry 24, 1605–1606. [Google Scholar]
  52. Robert C, Thomas L, Bondarenko I, O’Day S, Weber J, Garbe C, Lebbe C, Baurain JF, Testori A, Grob JJ, et al. (2011). Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med 364 2517–2526. [DOI] [PubMed] [Google Scholar]
  53. Rosen RH, Gupta AK, and Tyring SK (2012). Dual mechanism of action of ingenol mebutate gel for topical treatment of actinic keratoses: rapid lesion necrosis followed by lesion-specific immune response. J Am Acad Dermatol 66, 486–493. [DOI] [PubMed] [Google Scholar]
  54. Scott-Browne JP, Lopez-Moyado IF, Trifari S, Wong V, Chavez L, Rao A, and Pereira RM (2016). Dynamic Changes in Chromatin Accessibility Occur in CD8(+) T Cells Responding to Viral Infection. Immunity 45 1327–1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Sharma P, Hu-Lieskovan S, Wargo JA, and Ribas A (2017). Primary, Adaptive, and Acquired Resistance to Cancer Immunotherapy. Cell 168 707–723. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sheppard KA, Fitz LJ, Lee JM, Benander C, George JA, Wooters J, Qiu Y, Jussif JM, Carter LL, Wood CR, et al. (2004). PD-1 inhibits T-cell receptor induced phosphorylation of the ZAP70/CD3zeta signalosome and downstream signaling to PKCtheta. FEBS Lett 574 37–41. [DOI] [PubMed] [Google Scholar]
  57. Siddiqui I, Schaeuble K, Chennupati V, Fuertes Marraco SA, Calderon-Copete S, Pais Ferreira D, Carmona SJ, Scarpellino L, Gfeller D, Pradervand S, et al. (2019). Intratumoral Tcf1(+)PD-1(+)CD8(+) T Cells with Stem-like Properties Promote Tumor Control in Response to Vaccination and Checkpoint Blockade Immunotherapy. Immunity 50 195–211 e110. [DOI] [PubMed] [Google Scholar]
  58. Sullivan BM, Juedes A, Szabo SJ, von Herrath M, and Glimcher LH (2003). Antigen-driven effector CD8 T cell function regulated by T-bet. Proc Natl Acad Sci U S A 100 15818–15823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Team, R.D.C. (2008). R: A language and environment for statistical computing. R Foundation for Statistical Computing; (Vienna, Austria: ). [Google Scholar]
  60. Teixeiro E, Daniels MA, Hamilton SE, Schrum AG, Bragado R, Jameson SC, and Palmer E (2009). Different T cell receptor signals determine CD8+ memory versus effector development. Science 323 502–505. [DOI] [PubMed] [Google Scholar]
  61. Topalian SL, Hodi FS, Brahmer JR, Gettinger SN, Smith DC, McDermott DF, Powderly JD, Carvajal RD, Sosman JA, Atkins MB, et al. (2012). Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med 366 2443–2454. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Varga SM, and Welsh RM (1998). Detection of a high frequency of virus-specific CD4+ T cells during acute infection with lymphocytic choriomeningitis virus. J Immunol 161 3215–3218. [PubMed] [Google Scholar]
  63. West EE, Jin HT, Rasheed AU, Penaloza-Macmaster P, Ha SJ, Tan WG, Youngblood B, Freeman GJ, Smith KA, and Ahmed R (2013). PD-L1 blockade synergizes with IL-2 therapy in reinvigorating exhausted T cells. J Clin Invest 123 2604–2615. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Wherry EJ, Blattman JN, Murali-Krishna K, van der Most R, and Ahmed R (2003). Viral Persistence Alters CD8 T-Cell Immunodominance and Tissue Distribution and Results in Distinct Stages of Functional Impairment. Journal of Virology 77 4911. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Wherry EJ, Ha SJ, Kaech SM, Haining WN, Sarkar S, Kalia V, Subramaniam S, Blattman JN, Barber DL, and Ahmed R (2007). Molecular signature of CD8+ T cell exhaustion during chronic viral infection. Immunity 27 670–684. [DOI] [PubMed] [Google Scholar]
  66. Yokosuka T, Kobayashi W, Sakata-Sogawa K, Takamatsu M, Hashimoto-Tane A, Dustin ML, Tokunaga M, and Saito T (2008). Spatiotemporal regulation of T cell costimulation by TCR-CD28 microclusters and protein kinase C theta translocation. Immunity 29 589–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Zajac AJ, Blattman JN, Murali-Krishna K, Sourdive DJ, Suresh M, Altman JD, and Ahmed R (1998). Viral immune evasion due to persistence of activated T cells without effector function. J Exp Med 188 2205–2213. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1060479-supplement-1.pdf (993.7KB, pdf)

Data Availability Statement

The accession number for the RNA sequence data reported in this paper is “GSE130406.”

RESOURCES