Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 18 (9), 1394-400

Bat3 Promotes T Cell Responses and Autoimmunity by Repressing Tim-3–mediated Cell Death and Exhaustion

Affiliations

Bat3 Promotes T Cell Responses and Autoimmunity by Repressing Tim-3–mediated Cell Death and Exhaustion

Manu Rangachari et al. Nat Med.

Abstract

T cell immunoglobulin and mucin domain–containing 3 (Tim-3) is an inhibitory receptor that is expressed on exhausted T cells during infection with HIV-1 and hepatitis C virus. By contrast, Tim-3 expression and function are defective in multiple human autoimmune diseases. However, the molecular mechanisms modulating Tim-3 function are not well understood. Here we show that human leukocyte antigen B (HLA-B)-associated transcript 3 (Bat3) binds to, and represses the function of, Tim-3. Bat3 protects T helper type 1 (TH1) cells from galectin-9–mediated cell death and promotes both proliferation and proinflammatory cytokine production. Bat3-deficient T cells have elevated expression of exhaustion-associated molecules such as Tim-3, Lag3, Prdm1 and Pbx3, and Bat3 knockdown in myelin-antigen–specific CD4+ T cells markedly inhibits the development of experimental autoimmune encephalomyelitis while promoting the expansion of a dysfunctional Tim-3hi, interferon-γ (IFN-γ)loCD4+ cell population. Furthermore, expression of Bat3 is reduced in exhausted Tim-3+ T cells from mouse tumors and HIV-1–infected individuals. These data indicate that Bat3 acts as an inhibitor of Tim-3–dependent exhaustion and cell death. Bat3 may thus represent a viable therapeutic target in autoimmune disorders, chronic infections and cancers.

Figures

Figure 1
Figure 1. Bat3 binds to the Tim-3 tail in a galectin-9-dependent manner
a. Domain structure and location of tyrosine residues in the Tim-3 tail, and depiction of Tim-3 deletion and substitution constructs used for yeast two-hybrid screening and co-IP studies. b. Bat3-myc, Tim-3-HA and ICOS-HA fusion constructs were generated to confirm the specificity of the Bat3/Tim-3 interaction in mammalian cells. Bat3+Tim-3 or Bat3+ICOS were co-expressed in 293T cells and cell lysates were prepared for IP with anti-myc and blotting with anti-HA. Representative of eight experiments. c. Bat3-myc was co-expressed in 293T cells with HA-tagged Tim-3 deletion and substitution mutants. Lysates were prepared for IP with anti-myc and blotting with anti-HA. Representative of three experiments. d. Tim-3 and Bat3, or Tim-32YF and Bat3, were co-expressed in 293T with or without a galectin-9 expression plasmid. Cell lysates were prepared for IP with anti-myc and blotting with anti-HA. Representative of three experiments.
Figure 2
Figure 2. Bat3 promotes Th1 cell function and protects from galectin-9 - mediated cell death
a. B6 CD4+ T cells were polarized under Th0, Th1 or Th17 conditions for 4 days. Cell lysates were prepared and probed with Bat3 antisera (top) or anti-β-tubulin (bottom). Representative of three experiments. b,c. 2D2 CD4+ cells specific for myelin oligodendrocyte glycoprotein 35–55 (MOG35–55) were polarized under Th1 conditions and transferred to Rag−/− recipients. At 4 weeks post-transfer, splenocytes were recovered and analyzed for intracellular cytokine (ICC) expression on GFP+ cells by flow cytometry (b), or were restimulated with MOG35–55 peptide and assessed for presence of IFNγ in culture supernatant 48 hours later (c). d. B6 CD4+ cells were Th1 polarized and were infected with Bat3-RV or empty RV. GFP+ cells were sorted 2 days after 3rd round polarization and then cultured with or without galectin-9 for 3 hours. Cell death was assessed on the basis of 7AAD-positivity. Data in the bar graph (right) represent the %7-AAD+ cells post-galectin-9 treatment divided by the %7-AAD+ cells pre-treatment. Representative of three experiments.
Figure 3
Figure 3. Bat3 expression in the hematopoietic compartment promotes CNS autoimmunity
a. Bat3+/−Rag−/− and Bat3−/−Rag−/− fetal liver chimeras were immunized subcutaneously with 100 μg MOG in complete Freund's adjuvant (CFA), and EAE development was monitored. The mean disease score of each group is shown. Representative of two experiments. 11 Bat3+/−Rag−/− (■) and 8 Bat3−/−Rag−/− (▲) mice were analyzed in total. b. Linear regression curves of the acute phase of the disease are shown for the experiment represented in (a). The slopes are significantly different between the two groups (p < 0.0001). The 95% confidence intervals for each curve are depicted with dashed lines. c. CD4+ cells from immunized Bat3+/−Rag−/− (■; n=7) and Bat3−/−Rag−/− (▲; n=5) mice were isolated 38 days after MOG immunization, and intracellular cytokine production was measured. Each data point represents an individual mouse.
Figure 4
Figure 4. Loss of Bat3 function induces an exhausted T cell phenotype and Bat3 regulates T cell receptor signaling
a,b,c,e. CD4+ T cells were isolated from Bat3+/−Rag−/− and Bat3−/− → Rag−/− mice, and were stimulated with anti-CD3+anti-CD28 for 5 days. a. Cells were assessed for intracellular expression of IFNγ, IL-2 and IL-10, and for surface expression of Tim-3. b. mRNA were extracted, and Lag3, Prdm1 and Pbx3 message were assessed by quantitative PCR. c. Cells were gated on level of CD3ε expression and were analyzed for Tim-3 surface levels by flow cytometry. d. 293T cells were transfected with Lck plus either HA-tagged wildtype Tim-3 or HA-tagged Tim-3 lacking the intracellular domain (Tim-3ΔIC). Cell lysates were prepared, immunoprecipitated with anti-HA and blotted with anti-Lck. e. Th1 cells were transduced with a Bat3 RV construct containing a C-terminal myc tag and bicistronic GFP reporter. Cells were sorted for GFP positivity and were incubated for 20 minutes with 10μg/mL anti-Tim-3 or mIgG1. Lysates were prepared for IP with anti-myc and IB with anti-phospho-Src to detect pLck(Y394), pLck(Y505) or anti-Lck(total). Representative of three experiments. f. Bat3+/−Rag−/− and Bat3−/−Rag−/− CD4+ T cells were stimulated for 5 days with anti-CD3+anti-CD28. Cell lysates were prepared and blotted for anti-phospho-Lck(Y505). Representative of three experiments.
Figure 5
Figure 5. Bat3 expression in CD4+ T cells is critical for the induction of Th1-driven EAE
a. GFP+-sorted, control- or Bat3KD1-RV infected, 2D2 Th1 cells were transferred to Rag−/− recipients, which were monitored for EAE development. Control 2D2 → Rag−/− (■), Bat3KD1-2D2 → Rag−/− (▲). Representative of 1 of 3 transfers. 13 control-2D2 and 14 Bat3KD1-2D2 recipients were analyzed in total. b. CNS-infiltrating mononuclear cells were collected at day 38 post-transfer from Rag−/− mice that had received control- or Bat3KD1 RV-infected 2D2 Th1 cells, and were analyzed for CD4 expression by flow cytometry. Representative of 4 control-2D2 and 2 Bat3KD1-2D2 recipients. c,d. Lymph node cells were collected at day 38 post-transfer from Rag−/− mice that had received control- or Bat3KD1 RV-infected 2D2 Th1 cells, and ICC staining was performed for IFNγ (c, d) and IL-10 (d). c. Plots gated on CD4+. Data are representative of 4 (control; ■) or 5 (Bat3KD1; ▲) mice. d. Data first gated on CD4+GFP+ cells. Each data point represents an individual mouse. e. Th1 cells were transduced with empty, Tim-3IC or Tim-3ICΔBat3BD-expressing retrovirus. GFP+ cells were analyzed for intracellular expression of IFNγ 5 days after 2nd round stimulation. Representative of three experiments. f. Tim-3+/+ and Tim-3−/− Th1 cells were infected with either control-, Bat3KD1- or Bat3KD2-expressing RV and cultured for 5 days. Cells were sorted for GFP positivity and were restimulated with 1 μg/mL anti-CD3+anti-CD28 for 48 hours before [3H]-thymidine pulse for an additional 16 hours. Representative of three experiments. g. Tim-3+/+ and Tim-3−/− Th1 cells were infected with either control- or Bat3KD1-expressing RV and cultured for 5 days. Cells were assessed for intracellular expression of IFNγ and IL-2 (gated on GFP+). Representative of three experiments.
Figure 6
Figure 6. Loss of Bat3 results in an exhausted-like phenotype in T cells
a. TILs were isolated from implanted 4T1 mammary adenocarcinomas 26 days post-implantation, and were analyzed for intracellular production of IFNγ and TNFα. Each data point represents an individual mouse. b. Lymphocytes in draining lymph nodes from CT26 tumor bearing mice treated with anti-Tim-3 or anti- PD-L1 antibody, or both, were stimulated in vitro with tumor antigen (AH1) for 48 hours and then measured for IFNγ in the supernatant. * p>0.01, ** p>0.05 c. TILs were isolated from implanted CT26 colorectal carcinomas, 21 days post-injection. Cells were sorted into Tim-3+PD-1+ and Tim-3PD-1+ subpopulations, mRNA were isolated and Bat3 message expression was assessed. Data were normalized to β-actin levels. Data represent pooled tumors from ten recipients. d. CD4+Tim-3-PD-1+ and CD4+Tim-3+PD-1+ populations were sorted from the peripheral blood of three antiretroviral therapy-naïve individuals with chronic progressive HIV-1 infection. mRNA were prepared from each subpopulation, and Bat3 message was assessed by quantitative PCR. Patient 1, p<0.024; patient 2, p<0.004; patient 3, p<0.004. e. 2D2 Th1 cells infected with Bat3KD1 RV were transferred into Rag−/− mice, and were analyzed 38 days later for IFNγ and Tim-3 co-expression. 5 of 11 Bat3KD1-2D2 recipients displayed GFPCD4+Tim-3hiIFNγlo cells in peripheral lymphoid tissues (right panel). GFPCD4+ cells from control-2D2 mice were Tim-3IFNγhi (left panel, representative of 9 mice).

Comment in

Similar articles

See all similar articles

Cited by 87 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback