Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 May 2;132(9):e152864.
doi: 10.1172/JCI152864.

Tim-3 mediates T cell trogocytosis to limit antitumor immunity

Affiliations

Tim-3 mediates T cell trogocytosis to limit antitumor immunity

Ornella Pagliano et al. J Clin Invest. .

Abstract

T cell immunoglobulin mucin domain-containing protein 3 (Tim-3) negatively regulates innate and adaptive immunity in cancer. To identify the mechanisms of Tim-3 in cancer immunity, we evaluated the effects of Tim-3 blockade in human and mouse melanoma. Here, we show that human programmed cell death 1-positive (PD-1+) Tim-3+CD8+ tumor-infiltrating lymphocytes (TILs) upregulate phosphatidylserine (PS), a receptor for Tim-3, and acquire cell surface myeloid markers from antigen-presenting cells (APCs) through transfer of membrane fragments called trogocytosis. Tim-3 blockade acted on Tim-3+ APCs in a PS-dependent fashion to disrupt the trogocytosis of activated tumor antigen-specific CD8+ T cells and PD-1+Tim-3+ CD8+ TILs isolated from patients with melanoma. Tim-3 and PD-1 blockades cooperated to disrupt trogocytosis of CD8+ TILs in 2 melanoma mouse models, decreasing tumor burden and prolonging survival. Deleting Tim-3 in dendritic cells but not in CD8+ T cells impeded the trogocytosis of CD8+ TILs in vivo. Trogocytosed CD8+ T cells presented tumor peptide-major histocompatibility complexes and became the target of fratricide T cell killing, which was reversed by Tim-3 blockade. Our findings have uncovered a mechanism Tim-3 uses to limit antitumor immunity.

Keywords: Cancer immunotherapy; Immunology; Melanoma; Oncology; T cells.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest: HMZ reports Bristol Myers Squibb, Checkmate Pharmaceuticals, GlaxoSmithKline (research support) and Bristol Myers Squibb, Checkmate Pharmaceuticals, GlaxoSmithKline, Vedanta Biosciences (consulting). DD reports Merck, Bristol Myers Squibb, Checkmate Pharmaceuticals, CellSight Technologies, MedPacto, GlaxoSmithKline (research support); Array Biopharma, Checkmate Pharmaceuticals, Incyte, Immunocore, Merck, Shionogi (consulting); and Vedanta Biosciences (scientific advisory board). JMK reports Amgen, Bristol Myers Squibb, Castle Biosciences, Checkmate Pharmaceuticals, Immunocore, Iovance, Novartis (research support) and Amgen, Bristol Myers Squibb, Checkmate Pharmaceuticals, Novartis (consulting). ACA reports Tizona Therapeutics, Compass Therapeutics, Zumutor Biologics, and ImmuneOncia (scientific advisory board), which have interests in cancer immunotherapy. VKK reports Compass Therapeutics, Tizona Therapeutics, and ImmuneOncia (scientific advisory board), which have interests in cancer immunotherapy; iTeos Therapeutics (consulting); Tizona Therapeutics (ownership interest and scientific advisory board); and Celsius Therapeutics and Bicara therapeutics (cofounder and ownership interest). ACA’s and VKK’s interests were reviewed and managed by the Brigham and Women’s Hospital and Mass General Brigham in accordance with their conflict-of-interest policies. AJK, AR, SMW, MH, CB, XD, XMS, and PS are employees of Bristol Myers Squibb.

Figures

Figure 1
Figure 1. PD-1+Tim-3hi CD8+ TILs express myeloid markers acquired from APCs via trogocytosis in metastatic melanoma.
(A) Summary data (n = 20) showing expression of PD-1 and Tim-3 on CD8+ TILs in metastatic melanoma (MM). (B) Summary data showing frequencies of perforin+ (n = 16), CD107a+ (n = 24), and PS+ (n = 13) CD8+ TILs in MM according to PD-1 and Tim-3 expression. (C) Pooled data (n = 11) showing the correlation between CD107a and PS expression in PD-1+Tim-3hi CD8+ TILs in MM. (D) Representative dot plots (left) and summary data (right, n = 10) showing frequencies of CD11c+, CD14+, and PD-L1+ CD8+ TILs according to PD-1 and Tim-3 expression. (E) Confocal micrographs of PKH67+ PD-1+Tim-3 and PD-1+Tim-3hi CD8+ TILs coincubated for 30 minutes with autologous PKH26+CD45+CD3 cells. (F) Representative ImageStream analysis (upper) and summary data of cell frequencies with flow cytometry (lower, n = 8) showing cell surface coexpression of PKH26 and CD11c, CD14, or PD-L1 on PKH67+ PD-1+Tim-3 and PD-1+Tim-3hi CD8+ TILs after 30 minutes’ incubation with autologous PKH26+CD45+CD3 cells. P values were obtained by 1-way ANOVA followed by Holm-Šídák multiple comparisons test and Friedman’s test followed by Dunn’s multiple-comparison test (B), simple linear regression test (C), and paired t test and Wilcoxon’s matched pairs signed-rank test (D and F). *P < 0.05; **P < 0.01; ****P < 0.0001. Data indicate mean ± SD.
Figure 2
Figure 2. Transcriptomic features of Tim-3+ tumor-infiltrating myeloid cells using single-cell RNA sequencing.
(A) Summary data showing frequencies of Tim-3+CD11c+ and Tim-3+CD14+ CD45+CD3 cells in MM (n = 10). (B) Single-cell RNA sequencing (scRNA-Seq) of CD45+ cells sorted from 18 PD-1–naive MM cases. Uniform manifold approximation and projection of 3872 myeloid cells shown in dots. Monocyte (Mono) and conventional dendritic cell (cDC) clusters were identified. (C) Heatmap of normalized HAVCR2 (Tim-3) expression. (D) Summary of pathway enrichment analysis using Metascape. Myeloid cells were separated into HAVCR2+ and HAVCR2 groups.
Figure 3
Figure 3. Two anti–Tim-3 antibodies bind with high affinity to Tim-3 but target 2 nonoverlapping epitopes.
(A) Yeast display epitope mapping of aTim-3 mAbs to human Tim-3 (hTim-3): aTim-3.18 mAb binds to hTim-3 (A, shown in cyan) on the PS binding site (orange sphere), while aTim-3.22 binds the opposite face of hTim-3 (B, red) away from the PS binding site. (B) X-ray crystal structure of aTim-3.18:hTim-3 complex reveals heavy chain CDR2 (blue) binds the PS binding loops (orange), with residues F56 and Y58 inserted into the PS binding pocket. CDR2, complementarity-determining region 2; h, human; IgV, Ig variable. (C) Cell-free PS blocking assay showing relative binding of PS liposomes to hTim-3 saturated with aTim-3.18 or aTim-3.22. (D) Kinetic affinity measurements of aTim-3.18 and aTim-3.22 mAbs. Surface plasmon resonance sensorgrams of hTim-3 binding in increasing concentrations to aTim-3.18 (left) or aTim-3.22 (right) mAbs (data in color, fit in black). Each mAb was captured in duplicate on different flow cells.
Figure 4
Figure 4. Tim-3 blockade acts in APCs in a PS-mediated fashion and cooperates with PD-1 blockade to decrease trogocytosis of TA-specific CD8+ T cells.
(A and B) Representative flow cytometry dot plots (A) and summary data (B) evaluating PKH26 and PS expression by CD8+ T cells. PKH26+HLA-A2+ DCs were incubated with indicated antibodies, then added to wells with NY-ESO-1 157-165–specific clone 95/3 and cognate peptide for 5 days before flow cytometry (A, upper); CD8+ T cell clone 95/3 was incubated with indicated antibodies, then added to wells with PKH26+HLA-A2+ DCs pulsed with cognate peptide for 5 days before flow cytometry (A, lower). Summary data showing fold changes of PKH26+ and PS+ CD8+ T cells upon DC blockade (B, upper) or CD8+ T cell blockade (B, lower) as compared with IgG mAbs. Data shown are representative of 3 independent experiments. (C) Representative ImageStream images (left) and flow cytometry analysis of PKH26+ CD8+ T cells (middle and right panels). PKH67+ PD-1+Tim-3 and PD-1+Tim-3+ CD8+ TILs were incubated for 30 minutes with PKH26+CD45+CD3 cells isolated from MM in the presence of indicated antibodies before analysis. Representative summary data showing fold change of PKH26+ PD-1+Tim-3+ CD8+ TILs in wells with aTim-3.18 or aPD-1 as compared with IgG control (n = 5). P values shown (B and C) were obtained from paired t tests in Supplemental Figure 3, A and B. Data indicate mean ± SD.
Figure 5
Figure 5. Frequencies of PS+, CD11c+, and CD14+ PD-1+Tim-3+ CD8+ TILs as well as Tim-3+ APCs increase in mouse melanoma upon tumor progression.
(AD) Flow cytometry analysis of YUMMER 1.7 melanoma collected at days 7, 13, and 18 after subcutaneous implantations. (A) Representative dot plots (left) and summary data (right) showing frequencies of PD-1+Tim-3+ CD8+ TILs. (B) Summary data showing frequencies of PS+ (left) and CD11c+ (right) PD-1+Tim-3+ CD8+ TILs. (C) Summary data showing frequencies of CD11c+ (left) and CD14+ (right) CD45+CD3 cells. (D) Representative dot plots (left) and summary data (right) showing frequencies of Tim-3+CD11c+ and Tim-3+CD14+ CD45+CD3 cells. Results shown are from 1 experiment (n = 5), representative of 2 independent experiments. P values were obtained by 1-way ANOVA followed by Tukey’s multiple-comparison test (AD). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data indicate mean ± SD.
Figure 6
Figure 6. Tim-3 and PD-1 blockades impede the trogocytosis of CD8+ TILs in vivo.
(A and B) Representative dot plots (left) and summary data (middle and right) showing frequencies of PD-1+Tim-3+ and PS+PD-1+Tim-3+ CD8+ TILs (day 21) in mice implanted with YUMMER 1.7 (A) or B16-OVA (B) and treated with indicated mAbs. (C) Representative dot plots (left) and summary data (n = 5–8, right) showing frequencies of CD11c+, CD209+, and PD-L1+ PD-1+Tim-3 and PD-1+Tim-3+ CD8+ TILs (day 21) in mice implanted with YUMMER 1.7 and treated with indicated mAbs. Results shown are from 1 experiment, representative of 3 independent experiments (IgG, n = 8; aTim-3, n = 8; aPD-1, n = 6; aPD-1 + aTim-3, n = 5). P values were obtained by 1-way ANOVA followed by Tukey’s multiple-comparison test (AC) or Kruskal-Wallis test followed by Dunn’s multiple-comparison test (C). **P < 0.01; ***P < 0.001; ****P < 0.0001. Data indicate mean ± SD.
Figure 7
Figure 7. Deleting Tim-3 on DCs but not on CD8+ T cells impedes the trogocytosis of CD8+ TILs in vivo.
Representative flow cytometry dot plots and summary data (n = 5, left) showing frequencies of CD11c+, CD209+, and CD14+ PD-1+Tim-3+ CD8+ TILs in Havcr2fl/fl, Havcr2fl/fl Cd11ccre, and Havcr2fl/fl E8icre mice implanted with YUMMER 1.7 (day 21). Results shown are from 1 experiment (n = 5), representative of 2 independent experiments. P values were obtained from unpaired t tests. **P < 0.01; ***P < 0.001; ****P < 0.0001. Data indicate mean ± SD.
Figure 8
Figure 8. Trogocytosed CD8+ T cells acquire peptide-MHC complexes in a Tim-3–mediated fashion.
(AC) Representative dot plots (left) and summary data (right) showing frequencies of OVA257-264 tet+ and PD-1+Tim-3+tet+ CD8+ TILs (A), peptide-MHC complex (pMHC)/OVA257-264 H-2Kb+ Tim-3+CD11c+ and Tim-3+CD14+ cells (B), or pMHC/OVA257-264-H-2Kb+ PD-1+Tim-3+ CD8+ TILs (C) in B16-OVA (day 21) treated with indicated mAbs. Results shown are from 1 experiment, representative of 2 independent experiments (n = 5). Data are representative of 3 independent experiments. P values were obtained by 1-way ANOVA followed by Tukey’s multiple-comparison test (A and C). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. Data indicate mean ± SD.
Figure 9
Figure 9. Trogocytosed CD8+ T cells become the target of fratricide killing.
(A) Flow cytometry data showing fratricide killing of PD-1+Tim-3+ CD8+ TILs isolated from human or mouse melanoma. PD-1+Tim-3 TILs sorted from human MM or YUMMER 1.7 (day 21 postimplantation) were labeled with PKH67 and cocultured with autologous PKH26-labeled PD-1Tim-3 (group “A” in the figure) or PD-1+Tim-3+ (group “B” in the figure) CD8+ TILs before flow cytometry (left). Representative dot plots (middle) and summary data (right, n = 6) showing percentages of PKH67+CD107a+ CD8+ TILs. (B) Flow cytometry data showing that Tim-3 blockade impedes fratricide killing of NY-ESO-1–specific CD8+ T cells in vitro. DCs obtained from HLA-A2+ HDs were pulsed with either tumor lysate (T. Lysate) from an HLA2/NY-ESO-1+ melanoma cell line, MEL 285, or NY-ESO-1 157-165 peptide (NY-ESO-1) with aTim-3 or IgG control mAbs, then washed and cocultured with PKH26-labeled clone 95/3. PKH26+CD8+ T cells were sorted and coincubated with PKH67+ clone 95/3 for 6 hours before flow cytometry (left). Representative dot plots (middle) showing percentages of PKH67+PKH26CD11c+ and PKH67+PKH26CD107a+ CD8+ T cell clones and summary data of frequency fold changes upon aTim-3 as compared with IgG control mAbs (right, n = 6). Data are representative of 3 independent experiments. P values were obtained by paired t test (A). P values shown in B were obtained from paired t tests in Supplemental Figure 5B. ***P < 0.001; ****P < 0.0001. Data indicate mean ± SD.

References

    1. Blackburn SD, et al. Coregulation of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral infection. Nat Immunol. 2009;10(1):29–37. doi: 10.1038/ni.1679. - DOI - PMC - PubMed
    1. Barber DL, et al. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature. 2006;439(7077):682–687. doi: 10.1038/nature04444. - DOI - PubMed
    1. Topalian SL, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366(26):2443–2454. doi: 10.1056/NEJMoa1200690. - DOI - PMC - PubMed
    1. Hamid O, et al. Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369(2):134–144. doi: 10.1056/NEJMoa1305133. - DOI - PMC - PubMed
    1. Larkin J, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373(1):23–34. doi: 10.1056/NEJMoa1504030. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources