Ex Vivo Culture with Albumin Greatly Increases the Immunosuppressive Subset of T Lymphocytes

Emerging studies suggest that the expression of co-inhibitory receptors such as PD1, TIM3, TIGIT, and LAG3 on the T cell surface is important in regulating T cell response. However, the low number of cells expressing such co-inhibitory receptors in peripheral blood severely limits their clinical potential in treating GVHD. In this study, we introduce a novel method to culture and obtain a high number of CD3⁺PD1⁺TIM3⁺ lymphocytes that also exert strong immunosuppression. We demonstrate that culturing CD3⁺ cells isolated from G-CSF mobilized peripheral blood stem cells (G-PBSCs) with human serum albumin (HSA) greatly increases the number of immunosuppressive CD3⁺ cells. Also, we confirm that the CD3⁺PD1⁺TIM3⁺ lymphocytes are the most potent subset of the CD3⁺ cells that induce immunosuppression and inhibit T cell proliferation.

Methods. The donors were subcutaneously injected with G-CSF (10μg/kg) for five days. G-PBSCs were collected from the donors using a COBE spectra cell separator, and CD3⁺ cells were isolated by positive selection by magnetic-activated cell sorting (MACS) using CD3 Dynabeads™. Isolated CD3⁺ cells were cultured with 5% HSA for 4 days (1x10⁵ cells/mL). PD1, TIM3, LAG3, and TIGIT expressions were analyzed using flow cytometry (FACS). CD3⁺, CD3⁺PD1⁺, and CD3⁺PD1⁺TIM3⁺ lymphocytes were sorted and cultured with irradiated allo-MNCs for three days (Mixed Lymphocyte Reaction; MLR). 5,6-carboxyfluorescein diacetate succinimidyl ester (CFSE) was used as an intracellular fluorescent dye in MLR (CFSE-MLR) to measure T cell proliferation.

Results. We discovered that treating G-PBSC cells with 5% HSA resulted in a dramatic increase in the number of immunosuppressive lymphocytes after the HSA treatment. In untreated G-PBSCs, there were very low numbers of CD3⁺PD1⁺TIM3⁺, CD3⁺PD1⁺LAG3⁺, and CD3⁺PD1⁺TIGIT⁺ lymphocytes (0.9±0.3x10³, 0.1±0.1x10³, and 0.3±0.2x10³, respectively). In G-PBSCs, the HSA treatment markedly increased the cell numbers (47.6±1.9x10³, 7.8±3.0x10³, and 0.7±0.3x10³, respectively) while resulting in a mild increase in the total number of CD3⁺ cells (1.2-fold increase, 1x10⁵ to 1.2±0.4x10⁵). Our data also demonstrated significant percent changes in subpopulations of CD3⁺ lymphocytes. In untreated G-PBSCs, very low percentages of cells were CD3⁺PD1⁺TIM3⁺, CD3⁺PD1⁺LAG3⁺, and CD3⁺PD1⁺TIGIT⁺ (9.0±3.1%, 1.2±0.1% and 3.2±2.8% in G-PBSC, respectively). After the HSA treatment, the percentages of the immunosuppressive lymphocytes in G-PBSCs markedly increased to 39.7±9.1%, 6.5±1.3%, and 0.6±0.4%, respectively. In particular, CD3⁺PD1⁺TIM3⁺ lymphocytes demonstrated the greatest increase in the cell number and the percentage after the HSA treatment, increasing by ~52 fold.

Next, we tested whether the HSA-treatment induced proliferation of immunosuppressive T cells translated into increased immunosuppression. To do this, we used CSFE-MLR, co-cultured the CD3⁺ G-PBSCs with allo-MNCs, and analyzed the number of unstimulated T cells. The HSA-treated CD3⁺ G-PBSCs significantly increased the levels of unstimulated T cells compared to the untreated G-PBSCs (% of unstimulated T cells in untreated cells vs. in HSA-treated cells; 0.5±0.1 vs. 15.4±11.7), indicating that the HSA treatment increased CD3⁺ cell-mediated immunosuppression.

Lastly, we tested the immunosuppressive effects of CD3⁺PD1⁺ and CD3⁺PD1^-lymphocytes in HSA-treated CD3⁺ G-PBSCs. Both CD3⁺PD1⁺ and CD3⁺PD1⁺TIM3⁺ lymphocytes demonstrated significantly enhanced levels of immunosuppression compared to CD3⁺PD1^-lymphocytes and CD3⁺PD1⁺TIM3^- lymphocytes, respectively (% of unstimulated T cells in CD3⁺PD1⁺ vs. in CD3⁺PD1^-lymphocytes, 27.9±5.5 vs. 4.3±7.5, and % of unstimulated T cells in CD3⁺PD1⁺TIM3⁺ vs. in CD3⁺PD1⁺TIM3^- lymphocytes, 46.1±5.5 vs. 17.4±7.4). Collectively, these data suggest that PD1 and TIM3 are key markers in the HSA-dependent increase in immunosuppression.

Conclusion. The ex vivo culture of CD3⁺ G-PBSCs with HSA markedly increases both the number and the effect of immunosuppressive CD3⁺ lymphocytes. Our analysis also indicates that the increased immunosuppression is likely due to the proliferation of CD3⁺PD1⁺TIM3⁺ lymphocytes. Taken together, our data suggest that the ex vivo HSA treatment of CD3⁺ G-PBSCs shows promising therapeutic potential for treating patients with GVHD.

No relevant conflicts of interest to declare.