Cytomegalovirus-Specific CD8+ T-Cells With Different T-Cell Receptor Affinities Segregate T-Cell Phenotypes and Correlate With Chronic Graft-Versus-Host Disease in Patients Post-Hematopoietic Stem Cell Transplantation

Abstract

Virus-specific T-cell responses are crucial to control cytomegalovirus (CMV) infections/reactivation in immunocompromised individuals. Adoptive cellular therapy with CMV-specific T-cells has become a viable treatment option. High-affinity anti-viral cellular immune responses are associated with improved long-term immune protection against CMV infection. To date, the characterization of high-affinity T-cell responses against CMV has not been achieved in blood from patients after allogeneic hematopoietic stem cell transplantation (HSCT). Therefore, the purpose of this study was to describe and analyze the phenotype and clinical impact of different CMV-specific CD8+ cytotoxic T-lymphocytes (CMV-CTL) classes based on their T-cell receptor (TCR) affinity. T-cells isolated from 23 patients during the first year following HSCT were tested for the expression of memory markers, programmed cell death 1 (PD-1), as well as TCR affinity, using three different HLA-A*02:01 CMV_NLVPMVATV-Pp65 tetramers (wild-type, a245v and q226a mutants). High-affinity CMV-CTL defined by q226a tetramer binding, exhibited a higher frequency in CD8+ T-cells in the first month post-HSCT and exhibited an effector memory phenotype associated with strong PD-1 expression as compared to the medium- and low-affinity CMV-CTLs. High-affinity CMV-CTL was found at higher proportion in patients with chronic graft-versus-host disease (p < 0.001). This study provides a first insight into the detailed TCR affinities of CMV-CTL. This may be useful in order to improve current immunotherapy protocols using isolation of viral-specific T-cell populations based on their TCR affinity.

Keywords: T-cell; T-cell receptor affinity; cytomegalovirus; cytomegalovirus-specific CD8+ cytotoxic T-lymphocytes; hematopoietic stem cell transplantation; programmed cell death-1; tetramers.

Figure 1

HLA-A2 cytomegalovirus (CMV) tetramer reactive CD8+ T-cells and CMV-specific T-cells post-hematopoietic stem cell transplantation (HSCT). (A) Analysis of the CMV-specific CD8+ cytotoxic T-lymphocytes (CMV-CTL) in 23 patients followed individually over time post-HSCT. (B) IFN-γ production evaluated after overnight incubation of CMV-CTL isolated by q226a mutant tetramer (square) and total fraction of CMV-CTL (circle) with different concentration of CMV peptide and CD8 blocking antibody (in red, DK25) or isotype control (green). (C) HLA-A2 CMV tetramer reactive CD8+ T-cells frequency post-HSCT. Dark circles represent wild-type tetramer reactive T-cells, blue squares represent a245v mutant tetramer reactive T-cells and red triangles represent q226a mutant tetramer reactive T-cells. (D) HLA-A2 CMV tetramer reactive CD8+ T-cells allow the analysis of the different CMV-specific CD8+ T-cells depending on the affinity of the T-cell receptor (TCR). (E) Frequency of CMV-CTL with different TCR affinity in peripheral blood of patient’s post-HSCT. Mean values are represented. (F) Analyses of the difference between the CMV-CTL subpopulations frequency in CD8+ T-cells. Mean and SD are represented. (G) Proportion analysis of the different subpopulations (low, medium, high-affinity) in the total of CMV-CTL. Mean and SD are represented. Black circles: low-affinity CMV-CTL, blue squares: medium-affinity CMV-CTL, red triangles: high-affinity CMV-CTL. Differences were analyzed using Wilcoxon matched-pairs signed rank test or Friedman test to detect differences between two groups of paired samples (A,E) and two-way ANOVA was used to detect differences between two groups over time (F,G). *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 2

Memory phenotype and PD-1 expression of the cytomegalovirus(CMV)-specific CD8+ cytotoxic T-lymphocytes (CMV-CTL) with different T-cell receptor affinities over time post-hematopoietic stem cell transplantation (HSCT). (A) Memory phenotype of the different CMV-CTL subpopulations. (B) PD-1 expression in the different CMV-CTL subpopulations (C) Proportion of the high/medium/low-affinity CMV-CTL in the CD8+PD-1+ T-cell population. (D) Comparison of the high-affinity CMV-CTL frequency between the total CD8+ T-cell population and the CD8+PD-1+ T-cell population. Mean and SD are represented. Two-way ANOVA was used to detect differences between two groups over time. *p < 0.05, **p < 0.01, ***p < 0.001. (E) Pie charts with the proportion of high/medium/low-affinity CMV-CTL the total CD8+ T-cell population and the CD8+PD-1+ T-cell population at month 1 and 2 post-HSCT.

Figure 3

Correlation with the donor serostatus and cytomegalovirus (CMV) reactivation. (A) Comparison of CMV-CTL frequency, low-, medium-, and high-affinity proportion between patients with donor presenting different CMV-serological status (D+: blue square and D−: black circle) (B) Analysis of CMV-CTL in between patients diagnosed with CMV reactivation (red circle) and patients who did not present sign of reactivation (blue square). Mean and SD are represented. Two-way ANOVA was used to detect differences between two groups over time. *p < 0.05.

Figure 4

Correlation between cytomegalovirus(CMV)-specific CD8+ cytotoxic T-lymphocytes(CTL) affinity and cgraft-versus-host disease(GVHD). (A) CMV-CTL frequency, low and high-affinity proportion correlated with cGVHD. Green circle: no cGVHD, orange square: cGVHD (I–III) (B) Proportion of high/medium/low-affinity CMV-CTL in patients presenting clinical symptoms of cGVHD (I–III). Black circles: low-affinity CMV-CTL, blue squares: medium-affinity CMV-CTL, red triangles: high-affinity CMV-CTL. Mean and SD are represented. Two-way ANOVA was used to detect differences between two groups over time.