Cytomegalovirus-responsive γδ T cells: novel effector cells in antibody-mediated kidney allograft microcirculation lesions

Abstract

Human cytomegalovirus infection in transplant recipients has been associated with adverse renal allograft outcome and with a large γδ T-cell response, but whether both mechanisms are connected is unknown. We previously showed that most expanded circulating cytomegalovirus-responsive γδ T cells express the Fcγ-receptor CD16, suggesting that γδ T cells may participate in allograft lesions mediated by donor-specific antibodies through antibody-dependent cellular cytotoxicity. Here, we show that cytomegalovirus-specific CD16(pos) γδ T cells can perform antibody-dependent cellular cytotoxicity against stromal cells coated with donor-specific antibodies in vitro. In vivo, graft-infiltrating γδ T cells localized in close contact with endothelial cells only in patients who experienced cytomegalovirus infection and were more frequent within peritubular capillaries and glomeruli from antibody-mediated acute rejections than within those from T cell-mediated acute rejections. Finally, a persistently increased percentage of circulating cytomegalovirus-induced γδ T cells correlated inversely with the 1-year eGFR only in kidney recipients with donor-specific antibodies. Collectively, these data support the conclusion that cytomegalovirus-induced γδ T cells are involved in, and may serve as a clinical biomarker of, antibody-mediated lesions of kidney transplants. Moreover, these findings offer a new physiopathologic link between cytomegalovirus infection and allograft dysfunction in recipients with donor-specific antibodies.

Figure 1.

Anti-class I HLA antibodies from sensitized kidney transplant recipients sera bind on fibroblastic and endothelial cells. (A) Representative binding on FSF of IgG from different sera using a goat anti-human IgG antibody coupled to FITC and analyzed by flow cytometry (numbers indicate the MFI). Pool S+, pool of sensitized sera; S1, nonsensitized serum; S3 and S4, sera containing CLSA; S9, sensitized serum without CLSA. (B) MFI obtained with each serum on all three cell lines. Data are representative of three different experiments. (C) Linear regression analysis of CLSA MFI analyzed by SAB assay versus the MFI of the IgG binding analyzed by flow cytometry on the three cell lines.

Figure 2.

CD16^pos γδ T cells can perform CLSA dependent cytotoxicity (ADCC) against endothelial and fibroblastic cells. (A) FSF, MRC5, and IVEC cell lines were labeled with ⁵¹Cr and pre incubated or not pre incubated with sera S3, S4, S7, and S8, which contained different levels of CLSA (indicated as + to +++) and sera that did not (indicated as −). Ability of the CD16^pos γδ T cell line to induce lysis of preincubated cell lines was evaluated by ⁵¹Cr release in the supernatant. Results are the mean specific lysis of culture triplicates from three independent experiments. SD was always <15% of the mean value (not shown). (B) Linear regression analysis between the MFI of IgG binding analyzed by flow cytometry on the three cell lines and the percentage of specific lysis (FSF: r²=0.83 [P=0.01]; IVEC: r²=0.93 [P=0.04]; MRC5: r²=0.74 [P=0.14]).

Figure 3.

CLSA-dependent ADCC by γδ T cells depends on CD16 and perforin. (A) CD16^pos (left panel) or CD16^neg (right panel) γδ T cell–mediated lysis of the FSF primary cell line preincubated with the serum S8 (+++) or S1 (−) in the presence or absence of a blocking anti-CD16 mAb or control mAb. (B) Expression of intracellular granzyme B and perforin in CD16^pos γδ T cells. Numbers indicate percentages of positive cells. (C) As in part A with the addition or no addition of indicated concentrations of concanamycin A (CMA). Viability of γδ T cells was >80% even after incubation of 7 hours with the highest concanamycin A concentration (data not shown). Data are representative of three different experiments. Ctrl, control.

Figure 4.

γδ T cells are present in antibody-mediated peritubular capillaritis and glomerulitis of CMV-infected patients. Triple-immunofluorescence staining for CD31 (green), γδ T cells (red), and nuclei (blue) on graft biopsy specimens from patients with acute AMR (n=10) and acute TCMR (n=13). (A) Representative staining of γδ T cells in microcirculation lesions associated with acute AMR in peritubular capillaritis (upper panel) and in glomerulitis (lower panel). Arrows indicate γδ T cells and arrowheads indicate endothelial cells of the microcirculation. Original magnification ×600. (B) Quantification of γδ T cells in peritubular capillaries and glomeruli of acute AMR or TCMR graft biopsy specimens in CMV-experienced (CMV⁺, upper panels) and CMV-naive (CMV⁻, lower panels) patients.

Figure 5.

γδ T-cell levels correlate with eGFR at 12 months after transplant in patients with DSA. (A) Median, 25th–75th percentiles and 10th–90th percentiles of γδ T cell (right panel) or Vδ2^neg γδ T cell (left panel) percentages in patients with acute TCMR, with acute AMR, or without rejection. (B) Linear regression analysis between circulating γδ T cell percentages (right panel) or Vδ2^neg γδ T-cell percentages (left panel) and the 12 month post-transplant eGFR in KTRs with DSA the day of the graft. Plain circles are CMV-experienced patients and open circles are CMV-naive patients. (C) Same analysis as in part B in anti-HLA nonsensitized KTRs.