Simultaneous detection of circulating autoreactive CD8+ T-cells specific for different islet cell-associated epitopes using combinatorial MHC multimers

Abstract

Objective: Type 1 diabetes results from selective T-cell-mediated destruction of the insulin-producing beta-cells in the pancreas. In this process, islet epitope-specific CD8(+) T-cells play a pivotal role. Thus, monitoring of multiple islet-specific CD8(+) T-cells may prove to be valuable for measuring disease activity, progression, and intervention. Yet, conventional detection techniques (ELISPOT and HLA tetramers) require many cells and are relatively insensitive.

Research design and methods: Here, we used a combinatorial quantum dot major histocompatibility complex multimer technique to simultaneously monitor the presence of HLA-A2 restricted insulin B(10-18), prepro-insulin (PPI)(15-24), islet antigen (IA)-2(797-805), GAD65(114-123), islet-specific glucose-6-phosphatase catalytic subunit-related protein (IGRP)(265-273), and prepro islet amyloid polypeptide (ppIAPP)(5-13)-specific CD8(+) T-cells in recent-onset diabetic patients, their siblings, healthy control subjects, and islet cell transplantation recipients.

Results: Using this kit, islet autoreactive CD8(+) T-cells recognizing insulin B(10-18), IA-2(797-805), and IGRP(265-273) were shown to be frequently detectable in recent-onset diabetic patients but rarely in healthy control subjects; PPI(15-24) proved to be the most sensitive epitope. Applying the "Diab-Q-kit" to samples of islet cell transplantation recipients allowed detection of changes of autoreactive T-cell frequencies against multiple islet cell-derived epitopes that were associated with disease activity and correlated with clinical outcome.

Conclusions: A kit was developed that allows simultaneous detection of CD8(+) T-cells reactive to multiple HLA-A2-restricted beta-cell epitopes requiring limited amounts of blood, without a need for in vitro culture, that is applicable on stored blood samples.

FIG. 1.

Flow cytometric analysis of epitope-specific CD8⁺ T-cells using the combinatorial Qdot approach. A: Gating strategy: viable CD8⁺ single T-cells were analyzed by gating lymphocytes on the basis of FSC-A and SSC-A. Subsequently, single cells were gated (FSC-W and FSC-H) and CD8-APC–positive but dump-channel fluorescein isothiocyanate (CD4 + CD14 + CD16 + CD20 + CD40)–negative cells were gated, of which the 7-AAD–positive cells were gated out. B: Qdot staining: within the viable CD8⁺ single T-cells, the cells recognizing the epitopes in the viral mix (Qdot 585 + 655) and insulin B_10–18 (Qdot 605 + 655) are shown as a typical example for a healthy control, a recent-onset type 1 diabetic patient (T1D), and a pretransplantation islet cell recipient.

FIG. 2.

Frequencies of epitope-specific CD8⁺ T-cells in recent-onset diabetic patients, their siblings, and healthy control subjects. The frequencies of CD8⁺ T-cells recognizing the epitopes HLA-A2_140–149, the viral mix, insulin B_10–18, PPI_15–24, GAD65_114–123, IA-2_797–805, IGRP_265–273, and ppIAPP_5–13 in HLA-A2 as determined by flow cytometry are depicted. First, the frequency detected in recent-onset diabetic patient material (RO, n = 3) and that of their siblings (Sibs, n = 5) was compared (left panels). Statistical analysis was performed using the Wilcoxon matched-pairs test. The frequencies detected in all RO (n = 20) and control subjects (Con, n = 15) were compared (right panels). Statistical analysis was performed using the unpaired t test with Welch correction (for HLA-A2 peptide and PPI_15–24) or the Mann-Whitney test (all other epitopes).

FIG. 3.

Frequencies of epitope-specific CD8⁺ T-cells in islet cell recipients over time. The frequencies of CD8⁺ T-cells recognizing the epitopes HLA-A2_140–149, the viral mix, insulinB_10–18, PPI_15–24, GAD65_114–123, IA-2_797–805, IGRP_265–273, and ppIAPP_5–13 in HLA-A2 as determined by flow cytometry are depicted. The frequencies of CD8⁺ T-cells were measured at four different time points: before transplantation, 6 weeks thereafter (reconstitution of the T-cell compartment after ATG induction treatment), 26 weeks after transplantation, and 52 weeks after transplantation. Changes in epitope reactivity of islet cell transplant recipients over time were tested using the Friedman test followed by a Dunn multiple comparisons test with *P < 0.05 values considered statistically significant. Data points of recent-onset diabetic patients are depicted to allow easy comparison of “recent-onset reactivity” and “islet cell transplant reactivity.”

FIG. 4.

Cumulative frequencies of epitope-specific CD8⁺ T-cells in islet cell recipients at different time points. The cumulative frequencies of CD8⁺ T-cells recognizing the epitopes insulin B_10–18, PPI_15–24, GAD65_114–123, IA-2_797–805, IGRP_265–273, and ppIAPP_5–13 in HLA-A2 as determined by flow cytometry are depicted for each islet cell recipient individually. Note the different axis for patients βct #6 and βct #7.