T cell exhaustion is associated with antigen abundance and promotes transplant acceptance

Abstract

Exhaustion of T cells limits their ability to clear chronic infections or eradicate tumors. Here, in the context of transplant, we investigated whether T cell exhaustion occurs and has a role in determining transplant outcome. A peptide/MHC tetramer-based approach was used to track exhausted CD8⁺ T cells in a male-to-female skin transplant model. Transplant of large whole-tail skins, but not small tail skins (0.8 cm × 0.8 cm), led to exhaustion of anti-male tetramer⁺ CD8⁺ T cells and subsequently the acceptance of skin grafts. To study CD4⁺ T cell exhaustion, we used the TCR-transgenic B6 TEa cells that recognize a major transplant antigen I-Eα from Balb/c mice. TEa cells were adoptively transferred either into B6 recipients that received Balb/c donor skins or into CB6F1 mice that contained an excessive amount of I-Eα antigen. Adoptively transferred TEa cells in skin-graft recipients were not exhausted. By contrast, virtually all adoptively transferred TEa cells were exhausted in CB6F1 mice. Those exhausted TEa cells lost ability to reject Balb/c skins upon further transfer into lymphopenic B6.Rag1^−/− mice. Hence, T cell exhaustion develops in the presence of abundant antigen and promotes transplant acceptance. These findings are essential for better understanding the nature of transplant tolerance.

1 |. INTRODUCTION

The term “T cell exhaustion” refers to a dysfunctional state of T cells resulting from chronic antigen stimulation.^1,2 During chronic infections or cancer, exhausted T (Tex) cells progressively lose cytokine production (eg, interferon [IFN]-γ, tumor necrosis factor [TNF]-α), express high levels of inhibitory receptors (eg, programmed cell death protein 1 [PD-1], T cell immunoglobulin and mucin-domain containing-3 [TIM-3], lymphocyte-activation gene 3 [LAG-3], T cell immunoreceptor with Ig and ITIM domains [TIGIT]), become less proliferative, and, most importantly, have a very limited ability to control infections or inhibit tumor growth.² Several transcription factors (eg, thymocyte selection-associated high mobility group box protein [TOX], nuclear receptor subfamily 4 group A [NR4A]) are highly expressed in Tex cells, promote the expression of multiple inhibitory receptors, and may be key drivers of T cell exhaustion.^3–6

Although sharing a core molecular signature, Tex cells are differently developed during chronic infection and cancer due to differences in immunogenicity and microenvironmental factors.¹ In a well-documented lymphocytic choriomeningitis virus (LCMV) clone-13 infection model, most virus-specific CD8⁺ T cells in the lymphoid organs are robustly activated and subsequently exhausted after infection.⁷ By contrast, in murine and human cancers, CD8⁺ Tex cells are often identified in tumors. Of note, tumor-infiltrating T cells are highly heterogeneous in terms of phenotypic and functional states. Exhaustion represents only 1 of those cell states.^8,9 Given the wide varieties of immune responses, T cell exhaustion is profound and complex and should be studied in a context-dependent manner.

It remains unclear whether T cell exhaustion naturally occurs in response to transplant.¹⁰ We found that IRF4 deletion leads to transplant tolerance by inducing an exhaustion-like phenotype of T cells, evident by their high PD-1 expression and impaired cytokine production.^11,12 Miller et al¹³ demonstrated that CD154 blockade plus donor-specific transfusion, an experimental approach to transplant tolerance, also induces a PD-1^hi phenotype of alloreactive T cells. Because IRF4 deletion or CD154 blockade impairs T cell effector function, T cell state under those conditions was called “T cell dysfunction” rather than exhaustion.

Herein, we studied T cell exhaustion in response to transplant without immunosuppression or gene deletion in T cells. To study CD8⁺ T cell exhaustion, female B6 recipients were transplanted with either large whole-tail skins or small tail skins from male B6 donors. Transplant of large but not small skins resulted in the exhaustion of anti-male tetramer⁺ CD8⁺ cells and, subsequently, the acceptance of skin grafts. To study CD4⁺ T cell exhaustion, TCR-transgenic B6 TEa cells (recognize Balb/c I-Eα antigen) were adoptively transferred either into B6 recipients that received Balb/c skins or into F1 offspring of B6 and Balb/c mice (CB6F1) that contain abundant I-Eα antigen. Adoptively transferred TEa cells in B6 recipients responded to Balb/c skin transplant and were not exhausted. Nevertheless, almost all adoptively transferred TEa cells were exhausted in CB6F1 mice. The exhaustion of TEa cells was revealed by their transcriptional landscape and failure to reject Balb/c skins on further transfer into lymphopenic mice.

2 |. MATERIALS AND METHODS

2.1 |. Mice

C57BL/6 (B6), B6.SJL CD45.1 congenic, TEa TCR transgenic, Balb/c, B6.Rag1^−/−, and CB6F1 (Balb/c × C57BL/6)F1 mice were purchased from Jackson Laboratory (Bar Harbor, MA). CD45.1 congenic mice were crossed to TEa mice to create CD45.1⁺ TEa mice. All animal experiments in this study were approved by the Houston Methodist Animal Care Committee in accordance with institutional animal care and use guidelines.

2.2 |. Murine skin transplant and histology staining

Skin transplant was performed using a previously described method.¹² Large whole-tail skins, small (0.8 cm × 0.8 cm) tail skins, or ear skins from male B6 donors were transplanted onto female B6 recipients. Balb/c tail skins (1 cm × 1 cm) were transplanted onto wild-type (WT) B6 or B6.Rag1^−/− mice. More than 90% necrosis of the donor skin tissue was considered as rejection. Hematoxylin & eosin (H&E) staining was performed on paraffin sections of skin grafts.

2.3 |. Flow cytometric analysis and cell sorting

T cells were analyzed on an LSR II or Fortessa flow cytometer (BD Biosciences) or sorted from splenocytes by an FACSAria cell sorter (BD Biosciences). PE-conjugated H-2D^b HY Uty tetramer was prepared at the National Institutes of Health Tetramer Core Facility. Fluorochrome-conjugated antibodies were purchased from BioLegend, Thermo Fisher Scientific, Santa Cruz Biotechnology, or Miltenyi Biotec. Dead cells were excluded from analysis by using a Zombie Aqua Fixable Viability Kit (BioLegend). Expression of transcription factors was determined by using the Foxp3/Transcription Factor Staining Buffer Set (Thermo Fisher Scientific). For intracellular staining of cytokines, cells were stimulated for 5 hours with 50 ng/mL phorbol 12-myristate 13-acetate and 500 ng/mL ionomycin (Sigma-Aldrich) in the presence of GolgiStop (BD Biosciences), fixed and permeabilized with Cytofix/Cytoperm solution (BD Biosciences), and stained with fluorochrome-labeled antibodies against cytokines.¹¹ The data were processed by using the FlowJo v10 software (Tree Star, Inc).

2.4 |. Tracking of anti-male tetramer⁺ CD8⁺ T cells

Following male-to-female tail skin transplant, T cells in peripheral blood were analyzed on days 7, 14, 21, 28, and 42, and splenocytes were analyzed on day 28. Single-cell suspensions were incubated with PE-conjugated H-2D^b HY Uty tetramer for 15 minutes at 37°C. Without washing, the cells were further stained with Zombie Aqua and fluorochrome-conjugated antibodies. The expression of cell surface and intracellular molecules was analyzed as previously described.¹¹

2.5 |. Analysis of adoptively transferred TEa cells

On day 0, CB6F1 and WT B6 mice were adoptively transferred intravenously with 1 × 10⁶ CD45.1⁺ TEa cells, which were isolated from spleens of CD45.1⁺ TEa mice using the Dynabeads untouched mouse CD4 cells kit (Thermo Fisher Scientific). WT B6 mice were also transplanted with Balb/c tail skins on day 0, whereas CB6F1 mice were intraperitoneally injected with 500 μg anti-NK1.1 (clone PK136, Bio X Cell) on days −3 and 15 to deplete NK cells. The following experiments were then performed: (1) on different days after TEa cell transfer, TEa cell states in peripheral blood and spleens were determined by flow cytometric analysis; (2) on day 6 after TEa-cell transfer, TCR(Vα2⁺ Vβ6⁺)CD45.1⁺CD4⁺ TEa cells were sorted from splenocytes, followed by RNA-seq analysis as described later; and (3) on day 30 after TEa cell transfer, TCR(Vα2⁺ Vβ6⁺)CD45.1⁺CD4⁺ TEa cells were sorted from splenocytes and further adoptively transferred into B6.Rag1^−/− mice. One day later, B6.Rag1^−/− mice were transplanted with Balb/c tail skins to determine the antigraft function of TEa cells.

2.6 |. RNA-seq analysis of adoptively transferred TEa cells

Adoptively transferred TEa cells were sorted from splenocytes and stored in TRIzol (Thermo Fisher Scientific). Samples were then sent to Novogene Inc for low-input total RNA library construction using the SMRTER kit lib (Takara Inc Mountain View). The genes with RNA Integrity number (≥5.8) were further sequenced on the Illumina HiSeq platform. Paired-end libraries with an average insert size of 150 bp were obtained. The RNA sequence qualities, including GC content and ErrorRate, were evaluated by the in-house pipeline from Novogene. Raw reads were aligned to mouse reference (mm10) using STAR aligner (version 2.6).¹⁴ Analysis of differentially expressed genes was performed by edgeR, which calculates the log expression ratio (trimmed mean of M values [TMM]) for each gene in effector (from skin-graft recipients) and exhaustion (from CB6F1 mice) TEa cells.¹⁵

2.7 |. Statistical analysis

Data were represented as mean ± SD and analyzed with Prism version 8 (GraphPad Software). The P values of skin-graft survival were determined with the Mann-Whitney test. Other measurements were performed using unpaired Student t test. Differences were considered significant when P < .05.

3 |. RESULTS

3.1 |. Large but not small male skin grafts are accepted by female recipients

To determine whether antigen abundance affects transplant survival, female B6 recipients were transplanted with either large whole-tail skins or small (0.8 cm × 0.8 cm) tail skins from male B6 donors (Figure 1A). All large tail skin grafts were accepted by recipients (mean survival time [MST] of > 100 days; n = 15). In contrast, all small tail skin grafts were rejected (MST = 48.4 ± 13.8 days; n = 15) (Figure 1B). Figure 1C shows the representative images of accepted skin grafts on recipients at 140 days after transplant.

FIGURE 1.

Large but not small male skin grafts are accepted by female recipients. A-C, Female B6 mice were transplanted with either large whole-tail skins or small tail skins from male B6 donors. A, Schematic of the experimental design. B, The percentage of skin-graft survival after transplant (n = 15). ****P < .0001; Mann-Whitney test. C, Representative images of the accepted male whole-tail skins on female recipients at 140 days postgrafting. D-F, More than 100 days after accepting the primary large whole-tail skins, female recipients were transplanted again with male ear skins as the secondary (2nd) grafts. D, Schematic of the experimental design for secondary grafting. E, The percentage of secondary skin-graft survival (n = 6). F, Representative images of the accepted secondary (ear) skin grafts, indicated by white arrows. G, Representative H&E staining images (×200) of small tail-skin graft (day 28; left), large whole-tail skin graft (day 420; middle), and secondary (ear) skin graft (day 280, right). Tx, transplantation; 2nd, secondary

Next, at >100 days after accepting the large male tail skins, female recipients were transplanted again with male ear skins as secondary grafts (Figure 1D). Four of 6 ear skin grafts have survived long term (>100 days) on those recipients (Figure 1E). On the contrary, all primary male ear skin grafts were rejected after transplanting onto naive female recipients (MST = 29.2 ± 5.01 days; n = 6) (Figure S1). Figure 1F shows the representative images of the accepted primary tail- and secondary ear-skin grafts at 280 days after ear-skin transplant. Figure 1G shows the representative H&E images. Excessive infiltrating cells were found in the small tail-skin graft but not in the accepted primary tail- and secondary ear-skin grafts. Hence, the abundance of transplant antigens (indicated by large tail skin) induces graft acceptance in the male-to-female skin transplant model.

3.2 |. Anti-male CD8⁺ T cells display an exhaustive phenotype in female recipients that received large but not small male skin grafts

To study the T cell states correlated with transplant outcome, female B6 recipients were transplanted with either large or small male tail skins, followed by flow cytometric analysis of anti-male H-2D^b HY Uty tetramer⁺ CD8⁺ T cells in peripheral blood or spleens. Figure 2A shows the representative plots for detecting tetramer⁺ CD8⁺ cells. In blood, tetramer⁺ CD8⁺ cell frequencies in both skin-graft groups were gradually increased and then declined. Tetramer⁺ CD8⁺ cell frequencies in the large skin-graft group were significantly lower than those of the small skin-graft group on days 28 and 42 and were barely detectable on day 42 (Figure 2A,B).

FIGURE 2.

Transplant of large but not small male skins induces exhaustion of anti-male CD8⁺ T cells in female recipients. Female B6 recipients were transplanted with either small tail skins or large whole-tail skins from male B6 donors, followed by flow cytometric analysis. A, Representative plots show the detection of tetramer⁺ CD8⁺ T cells. B, % tetramer⁺ cells among CD8⁺ T cells in peripheral blood at indicated days postgrafting. Data are mean ± SD (n = 4) from 1 experiment, representative of 2 independent experiments. C-E, Splenocytes were analyzed on day 28 postgrafting. Representative contour plots were gated on living tetramer⁺ CD8⁺ T cells and display the expression of CD62L and CD44 (C), PD-1 and CD38 (D), and IFN-γ and TNF-α (E). The left bar graph in C shows % tetramer⁺ in CD8⁺ splenocytes. Other bar graphs show % CD44⁺ cells (C), % PD-1⁺ and % PD-1⁺CD38⁺ cells (D), and % IFN-γ⁺TNF-α⁺ cells (E) among living tetramer⁺ CD8⁺ splenocytes. Data are mean ± SD (n = 4) from 1 experiment, representative of 3 independent experiments. *P < .05; **P < .01, ***P < .001; ****P < .0001 (unpaired studen’s t test)

Splenocytes were analyzed at 28 days postgrafting. Tetramer⁺ CD8⁺ splenocytes in both skin-graft groups display an activated CD62L⁻CD44⁺ T cell phenotype (Figure 2C, contour plots and right bar graph), but their frequencies were significantly lower in the large skin-graft group (Figure 2C, left bar graph). Importantly, compared with the small skin-graft group, tetramer⁺ CD8⁺ splenocytes in the large skin-graft group expressed significantly higher levels of PD-1, CD38, TIM-3, 2B4, and LAG-3 (Figure 2D, Figure S2A) and produced significantly less IFN-γ, TNF-α, and Perforin (Figure 2E, Figure S2B). Expressions of BTLA, CD160, TIGIT, IL-2 and Granzyme B by tetramer⁺ CD8⁺ splenocytes were not significantly different between 2 groups (Figure S2). Therefore, anti-male CD8⁺ T cells exhibit an exhaustive phenotype in female recipients that were transplanted with large but not small male skins.

3.3 |. Adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients exhibit an exhaustion gene signature

To investigate CD4⁺ T cell exhaustion, CD45.1⁺ TEa cells were adoptively transferred either into B6 mice that received Balb/c skin grafts (skins rejected on days 11-13; data not shown), or into CB6F1 mice that contained abundant I-Eα antigen (Figure 3A). Figure S3A shows the gating strategy for detecting the adoptively transferred TEa cells. In blood, TEa cell frequencies in CB6F1 mice were significantly higher on day 7 and significantly lower on days 14, 21, and 28 compared with those of skin-graft recipients (Figure 3B). Thus, TEa cells in CB6F1 mice expanded and declined rapidly.

FIGURE 3.

Adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients display an exhaustion gene signature. 1 × 10⁶ CD45.1⁺ TEa cells were adoptively transferred either into B6 recipients that received Balb/c skins, or into CB6F1 mice that received the anti-NK1.1 (αNK1.1) treatment on days −3 and 15. A, Schematic of the experimental design. B, % TEa cells among CD4⁺ cells in peripheral blood at days 7, 14, 21, and 28 post cell transfer. Data are mean ± SD (n = 5) from 1 experiment, representative of 2 independent experiments. *P < .05; **P < .01 (unpaired Student t test). C, D, On day 6 post cell transfer, TEa cells were isolated from splenocytes for RNA-seq analysis. One RNA sample was obtained from sorted TEa cells pooled from 5 skin-graft recipients, whereas the other RNA sample was obtained from sorted TEa cells pooled from 3 CB6F1 mice. C, The volcano plot shows up- and down-regulated genes in TEa cells from the CB6F1 mice (vs skin-graft recipients), represented by the red and green dots, respectively. The STAR program was used to align the reads to the mouse reference file (GENCODE mouse reference GRCm38).¹⁴ Differential expression was analyzed using R package (edgeR)²⁶ with the raw read count of each gene adjusted by trimmed mean of M values (TMM).¹⁵ The significant differentially expressed genes are both q value < 0.005 and |log2(FoldChange)|> 1. Highlighted genes are exhaustion- and effector-related genes. D, Heatmap of selected differentially expressed genes. Read count of each gene were normalized to fragments per kilobase of transcript sequence per millions base pairs sequenced (FPKM), which takes into consideration the effects of both sequencing depth and gene length on counting of fragments. Z-score transformation $[\text{Z}=(\text{x}-\mu )/(\sigma /\sqrt{\text{n}})]$ was performed to obtain the relative expression of each gene cross each sample. Color is proportional to the relative abundance of gene expression across each sample (Z-score). Exhaustion- and effector-related genes were highlighted in a larger font size

We isolated TEa cells from spleens of skin-graft recipients and CB6F1 mice at day 6 post cell transfer and determined their gene expression profiles. The expression of 268 genes were increased and 316 genes were downregulated in TEa cells from CB6F1 mice compared with those from skin-graft recipients (Figure 3C). In particular, compared with TEa cells from skin-graft recipients, TEa cells from CB6F1 mice expressed upregulated levels of key exhaustion genes such as Pdcd1 (encoding PD-1), Tox, Lag3, Cd200, and Cd38, as well as downregulated levels of effector genes Ifng, Gzmb, Prf1, Tbx21, Il2ra, Fasl, and Klrg1. Of note, the expression of chemokine receptor genes Cxcr3, Cxcr5, Ccr2, Ccr5, and Ccr9 were also downregulated in TEa cells from CB6F1 mice (Figure 3C,D). Therefore, adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients exhibit a core gene signature of T cell exhaustion.

3.4 |. Adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients are exhausted

To further investigate CD4⁺ T cell exhaustion, CD45.1⁺ TEa cells were adoptively transferred into either CB6F1 mice or skin-graft B6 recipients, followed by cytometric analysis of TEa cells in spleens. At 6 days post cell transfer, TEa cell frequencies were significantly higher in spleens of CB6F1 mice than those of skin-graft recipients (Figure S3B). Nevertheless, in contrast to TEa cells from skin-graft recipients, TEa cells from CB6F1 mice barely produced effector cytokines IFN-γ and TNF-α (Figure 4A), and expressed extremely high levels of PD-1 and BTLA (Figure 4B,C). In addition, compared with TEa cells from skin-graft recipients, TEa cells from CB6F1 mice expressed significantly higher levels of TOX, NFATc1, and MAF, and significantly lower levels of Ki67, T-bet, CXCR3, Fas ligand (FasL), KLRG1, Granzyme A, and Perforin (Figure 4D, S3C). Expression of Granzyme B, RORγt, IL-4, and IL-17A in TEa cells were not significantly different between 2 groups (Figure S3C).

FIGURE 4.

Adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients display an exhaustion-like phenotype on day 6 post cell transfer. CD45.1⁺ TEa cells were adoptively transferred either into B6 recipients that received Balb/c skins or into CB6F1 mice that received the anti-NK1.1 treatment on day −3. TEa cell states in spleens were determined by flow cytometric analysis on day 6 post cell transfer. All contour plots and overlay histograms were gated on living CD45.1⁺Vβ6⁺ TEa cells. A-C, Representative plots and bar graphs display % IFN-γ⁺ TNF-α⁺ cells (A), % PD-1⁺ cells (B), and % BTLA⁺ cells (C) among TEa cells. D, Representative overlay histograms and bar graphs show the expression levels of TOX, NFATc1, Ki67, T-bet, CXCR3, and FasL in TEa cells. Data are mean ± SD (n = 5) from 1 experiment, representative of 3 independent experiments. *P < .05; **P < .01, ***P < .001; ****P < .0001 (unpaired Student t test)

At 30 days post cell transfer, the majority of TEa cells in spleens of CB6F1 mice were CD62L⁻CD44⁺, whereas TEa cells in skin-graft recipients were heterogeneous in the expression of CD62L and CD44 (Figure S4A). Compared with TEa cells from skin-graft recipients, TEa cells from CB6F1 mice expressed extremely high levels of PD-1 and CD38 (Figure 5A) and barely produce effector cytokines IFN-γ and TNF-α (Figure 5B). Moreover, compared with TEa cells from skin-graft recipients, TEa cells from CB6F1 mice expressed significantly higher levels of TOX, NR4A1, NR4A2, LAG-3, TIM-3, TIGIT, CD39, and 2B4 (Figure 5C, Figure S4B).

FIGURE 5.

Adoptively transferred TEa cells in CB6F1 mice but not skin-graft recipients are exhausted on day 30 post cell transfer. CD45.1⁺ TEa cells were adoptively transferred either into B6 recipients that received Balb/c skins or into CB6F1 mice that received the anti-NK1.1 treatment on days −3 and 15. TEa cell states in spleens were determined by flow cytometric analysis on day 30 post cell transfer. A, Representative plots display % PD-1⁺CD38⁺ cells among TEa cells. Bar graphs show the PD-1 and CD38 mean fluorescent intensity (MFI) of TEa cells. B, Representative plots and bar graph show the % IFN-γ⁺ TNF-α⁺ cells among TEa cells. C, Representative overlay histograms and bar graphs show the expression levels of TOX, NR4A1, and NR4A2 in TEa cells. Data are mean ± SD (n = 3) from 1 experiment, representative of 3 independent experiments. *P < .05; **P < .01, ***P < .001; ****P < .0001 (unpaired Student t test). D, E, TEa cells were sorted from CB6F1 mice and skin-graft B6 recipients at 30 days post initial TEa cell transfer. Five hundred of those sorted TEa cells were further adoptively transferred into Rag1^−/− mice. One day later, Rag1^−/− mice were transplanted with Balb/c tail skins. D, Schematic of the experimental design. E, The percentage of skin-graft survival after Balb/c skin transplant (n = 3); *P < .05; Mann-Whitney test

TEa cells were isolated from spleens of CB6F1 mice and skin-graft B6 recipients at 30 days post initial TEa cell transfer and then were further adoptively transferred into B6.Rag1^−/− mice. One day later, B6.Rag1^−/− mice were transplanted with Balb/c tail skins (Figure 5D). TEa cells isolated from skin-graft B6 recipients but not from CB6F1 mice were potent to mediate skin-graft rejection in B6.Rag1^−/− mice (Figure 5E). Therefore, an abundance of I-Eα antigen in CB6F1 mice induces exhaustion of adoptively transferred TEa cells.

4 |. DISCUSSION

Deletion, anergy, and regulatory T cell suppression of anti-graft T cells have been reported as mechanisms underlying transplant tolerance or acceptance.^16–19 Here, we demonstrated that exhaustion of anti-graft T cells is also a mechanism underlying transplant acceptance. Indeed, anti-male specific CD8⁺ T cells were exhausted in female recipients that had accepted the large male skin grafts. Adoptively transferred CD4⁺ TEa cells in CB6F1 mice were also exhausted and lost the potency to reject Balb/c skin grafts after further transfer into lymphopenic mice.

The transferred TEa cells were exhausted in CB6F1 mice but not in B6 recipients that received Balb/c skins. In these 2 TEa-cell transfer models, differences in the amount of I-Eα antigen and other factors (eg, surgical trauma, influence from other immune cells) may affect the phenotype/expression profile of the transferred TEa cells. A relevant question is whether increasing the Balb/c graft size can induce T cell exhaustion in B6 recipients. Balb/c-to-B6 transplant is a fully MHC-mismatched model and involves many endogenous alloreactive T cells. We demonstrated previously that transplant of Balb/c hearts (larger than the skin grafts) into B6 mice effectively activated both endogenous alloreactive T cells and the transferred TEa cells and did not induce T cell exhaustion/dysfunction. Indeed, we had to delete a transcription factor IRF4 to induce T cell dysfunction and transplant tolerance.^11,12 Interestingly, dramatically increasing graft size (eg, transplant of 2 kidneys plus 2 hearts in 1 recipient) or reducing alloreactive T cell frequency promotes the acceptance of MHC-mismatched grafts.^20,21 It would be informative to know whether these conditions can induce T cell exhaustion in fully MHC-mismatched transplant.

It has been shown that as early as on days 5 (a cancer model) and 8 (the LCMV infection model) post antigen exposure, T cells with the exhausted fate have already expressed high levels of inhibitory receptors and lost cytokine production.^22,23 In this study, at 6 days post cell transfer, TEa cells also displayed exhaustive changes in CB6F1 mice. Exhaustive differentiation is a continuous process. Although exhaustive changes can be detected within 5-8 days post antigen exposure in these models, chronic antigen exposure may further change the gene expression profile and the epigenetic state of exhausted T cells.^22–24

In terms of molecular mechanism, recent reports indicated that the TCR-NFAT signaling, a key T cell activation pathway, also governs the T cell exhaustion program.²⁵ Enhanced NFAT activation up-regulates the expression of exhaustion drivers TOX and NR4As.^3,6 Here we found that compared with the transferred TEa cells in Balb/c skin-transplanted B6 mice, TEa cells in CB6F1 mice have increased expression of NFATc1, TOX, NR4A1, and NR4A2. Hence, the abundant I-Eα antigen in CB6F1 mice may promote the TCR-NFAT-TOX/NR4A exhaustion program in TEa cells. By all means, the molecular mechanism of T cell exhaustion remains unclear and requires further investigation.

It is a real challenge to achieve clinical transplant tolerance. Transplant patients require lifelong use of FK506 or other immunosuppressants. However, recent reports indicated that FK506 inhibits the exhaustive/dysfunctional differentiation of T cells by inhibiting NFAT activation,^22,25 and in turn repressing the expression of exhaustion drivers TOX and NR4A.^3,6 For developing an optimal strategy for transplant tolerance, it is necessary to identify the therapeutic approaches that not only inhibit the effector programs in alloreactive T cells but also promote the exhaustive/dysfunctional differentiation of alloreactive T cells.

Abbreviations:

B6

C57BL/6

CB6F1

(Balb/c × C57BL/6)F1

FasL

Fas ligand

H&E

hematoxylin and eosin

IFN

interferon

LAG-3

lymphocyte-activation gene 3

LCMV

lymphocytic choriomeningitis virus

MFI

mean fluorescent intensity

MST

mean survival time

NR4A

nuclear receptor subfamily 4 group A

PD-1

programmed cell death protein 1

Tex

exhausted T

TIGIT

T cell immunoreceptor with Ig and ITIM domains

TIM-3

T cell immunoglobulin and mucin-domain containing-3

TNF

tumor necrosis factor

TOX

thymocyte selection-associated high mobility group box protein

Tx

transplant

WT

wild-type