Requirement of cognate CD4+ T-cell recognition for the regulation of allospecific CTL by human CD4+ CD127- CD25+ FOXP3+ cells generated in MLR

Abstract

Although immunoregulation of alloreactive human CTLs has been described, the direct influence of CD4(+) Tregs on CD8(+) cytotoxicity and the interactive mechanisms have not been well clarified. Therefore, human CD4(+)CD127(-)CD25(+)FOXP3(+) Tregs were generated in MLR, immunoselected and their allospecific regulatory functions and associated mechanisms were then tested using modified (51)Chromium release assays (Micro-CML), MLRs and CFSE-based multi-fluorochrome flow cytometry proliferation assays. It was observed that increased numbers of CD4(+)CD127(-)CD25(+)FOXP3(+) cells were generated after a 7 day MLR. After immunoselection for CD4(+)CD127(-)CD25(+) cells, they were designated as MLR-Tregs. When added as third component modulators, MLR-Tregs inhibited the alloreactive proliferation of autologous PBMC in a concentration dependent manner. The inhibition was quasi-antigen specific, in that the inhibition was non-specific at higher MLR-Treg modulator doses, but non-specificity disappeared with lower numbers at which specific inhibition was still significant. When tested in micro-CML assays CTL inhibition occurred with PBMC and purified CD8(+) responders. However, antigen specificity of CTL inhibition was observed only with unpurified PBMC responders and not with purified CD8(+) responders or even with CD8(+) responders plus Non-T "APC". However, allospecificity of CTL regulation was restored when autologous purified CD4(+) T cells were added to the CD8(+) responders. Proliferation of CD8(+) cells was suppressed by MLR-Tregs in the presence or absence of IL-2. Inhibition by MLR-Tregs was mediated through down-regulation of intracellular perforin, granzyme B and membrane-bound CD25 molecules on the responding CD8(+) cells. Therefore, it was concluded that human CD4(+)CD127(-)CD25(+)FOXP3(+) MLR-Tregs down-regulate alloreactive cytotoxic responses. Regulatory allospecificity, however, requires the presence of cognate responding CD4(+) T cells. CD8(+) CTL regulatory mechanisms include impaired proliferation, reduced expression of cytolytic molecules and CD25(+) activation epitopes.

Figure 1. Flow diagram depicting the culture system for the generation of MLR-Tregs (step #1) and their utilization in various MLR, micro- CML and flow cytometric assays (step #2).

Figure 2. Purity and FOXP3 expression of MLR-Tregs.

PBMCs from a healthy volunteer were stimulated with irradiated PBMC from an HLA fully mismatched donor. After 7 days, the CD4⁺CD127⁻CD25⁺ cells were isolated by the Treg isolation kit (methods). These cells were designated as MLR-Tregs. The purity of the isolated cells was assessed for the indicated markers by flow cytometry. The gating strategy is indicated by arrows and title headings on histograms. Thus, the viable cells (top left) were gated based on forward scatter and side scatter and then on CD4⁺ cells (top middle) followed by CD25⁺ or CD25^high expression (top right). The FOXP3 and CD127 levels were then assessed on indicated gates (bottom). These dot plots depict 1 example of >30 experiments performed in this report.

Figure 3. The ability of MLR-Tregs to allospecifically suppress MLR proliferation.

MLR-Tregs were added as modulators in descending concentrations of 1×10⁴, 2×10³ or 0.4×10³ cells per well to 1×10⁵ fresh responding PBMC from the same individual as the one from whom MLR-Tregs were generated (i.e., the responders were autologous to MLR-Tregs). These were stimulated with 1×10⁵ irradiated PBMC and 18-hour ³H-Thymidine incorporation assays were performed (as diagrammatically shown in Figure 1; step# 2, left) and the data are shown as: (A) CPM: ³H-TdR uptake in MLR of the responder to the original stimulator used in generating the MLR-Tregs in the presence of the indicated number of modulator cells. Note that inhibition by MLR-Treg modulators is demonstrated by the differences between modulator Treg points, (right side), vs. fresh Ax (autologous irradiated PBMC) added as modulator controls (left side) (** = p<.01; n = 10). (B) Percentage inhibition: The CPM values (from A) were converted to percent inhibition (Tregs vs. Ax; see Methods for the formula) and allospecific vs. non-specific inhibition is shown. For allospecific inhibition the stimulators were from the original stimulators used for generating MLR-Tregs and for non-specific inhibition the stimulator PBMC were from a different totally HLA mismatched (third party) individual. Note the drastic decrease in the inhibitory effect by MLR-Tregs in the non-specific culture combinations as the modulator cell concentrations decreased (** = p<0.01; n = 5).

Figure 4. Regulatory effects of MLR-Tregs in micro-CML assays of responding whole PBMC.

MLR-Tregs were added as modulators in descending concentrations of 1×10⁴, 2×10³ or 0.4×10³ cells per well to 1×10⁵ fresh responding PBMC from the same individual as the one from whom MLR-Tregs were generated. These were stimulated with 1×10⁵ irradiated allogeneic PBMC and 4-hour ⁵¹Cr release assays against target cells from the stimulator were performed (as diagrammatically shown in Figure 1; step# 2, middle). The data are depicted as: (A) Percent specific lysis: against the specific stimulator used both in MLR-Treg generation and the micro-CML readout. Similar to the data shown in Figure 3, the inhibitory effects by MLR-Tregs are depicted by the points on the right, and the absence of inhibitory effects by modulator controls (Ax) is shown by points on the left side of each graph (** = p<.01; n = 10). (B) Percentage inhibition: The percent specific lysis values were converted to percent inhibition (see Methods for this calculation). To test for specific inhibition, the stimulator/targets were from the original donor used for generating MLR-Tregs and for non-specific inhibition stimulator/target cells were from a third party donor. Note that allospecificity of CTL regulation is demonstrated with more significant inhibition at higher concentrations of MLR-Tregs in responses against stimulating cells from the original donor vs. those of the third party donor (** = p<0.01; * = p<0.05; n = 5).

Figure 5. Regulatory effects of MLR-Tregs in micro-CML assays of responding purified CD8⁺ cells.

Micro-CML inhibition assays were performed as described in Figure 4, except that 5×10⁴ purified CD8⁺ cells rather than whole PBMC (1×10⁵) were used as responders. The data are depicted as: (A) Percent specific lysis: against the specific stimulator used both in MLR-Treg generation and the micro-CML readout. Similar to the data in Figure 4A, the lysis of target cells was decreasingly inhibited by decreasing concentrations of MLR-Tregs (** = p<.01; n = 6). (B) Percentage inhibition: the percent specific lysis values were converted to percent inhibition. In contrast to the findings depicted in Figure 4B, inhibition of purified responding CD8⁺ CTL activity did not appear to be as clearly allospecific, i.e. there was a lack of significant differences between the points on the right side vs the left side of each graph in the lower row (p>0.05; n = 5).

Figure 6. MLR-Tregs non-specifically regulate cytotoxic activity in micro-CML generated by purified CD8⁺ plus autologous non-T “APC” responders.

Micro-CML inhibition assays were performed with purified CD8⁺ responders as described in Fig. 5, but in presence of non-T “APCs” autologous to the responders, and the data are shown as: (A) Percent specific lysis: against the specific stimulator used both in MLR-Treg generation and the micro-CML readout. The lysis of target cells was decreasingly inhibited by decreasing concentrations of modulating MLR-Tregs (points on the right side) as opposed to none seen using control (Ax) modulators (points on the left side) (** = p<.01; n = 10). (B) Percentage inhibition: the percent specific lysis values were converted to percent inhibition. In contrast to the findings depicted in Figure 4B, using the original vs third party stimulating cells, inhibition of CTL activity generated by purified CD8⁺ plus non-T “APC” did not appear to be allospecific. Note the lack of significant differences between the points on the right side vs the left side of each graph in the lower row (n = 5).

Figure 7. MLR-Tregs allospecifically regulate cytotoxic activity in micro-CML generated by purified CD8⁺ plus purified autologous CD4⁺ responders.

Micro-CML inhibition assays were performed with purified CD8⁺ responders as described in Figure 5, but in presence of purified CD4⁺ cells autologous to the responders; the data are depicted as: (A) Percent specific lysis: against the specific stimulator used both in MLR-Treg generation and the micro-CML readout. The lysis of target cells was decreasingly inhibited by decreasing concentrations of modulating MLR-Tregs (points on the right side) as opposed to none seen using control modulators (points on the left side) (** = p<.01; n = 10). (B) Percentage inhibition: the percent specific lysis values were converted to percent inhibition. Noteworthy is that in contrast with the findings depicted in Figure 5B (but similar to those of Figure 4B), using the original vs third party stimulator/target cells, allospecific lytic inhibition was restored by using purified CD8⁺ to which purified autologous CD4⁺ cells were added (* = p<0.05; ** = p<0.01; n = 5).

Figure 8. MLR-Tregs suppress purified CD8 allospecific proliferation.

5×10⁴ CFSE labeled purified CD8 cells were cultured with 1×10⁵ irradiated PBMC from the original stimulator (used in generating MLR-Tregs) together with indicated numbers of autologous MLR-Treg modulators or autologous irradiated controls (Ax), either the presence (top row) or absence (bottom row) of IL-2 (10 U/ml). After 7 days in culture flow cytometric assays were performed and the percentage of CFSE-diluted cells was estimated after gating on viable lymphocytes followed by CD8⁺ cells. It was observed that the irradiated stimulators and Ax died off by day 7 (not shown); even the few that remained were gated out on CFSE vs. CD8 density-plot during the analysis. Note the increasing percentages of (CFSE diluted) proliferating cells with decreasing concentrations of the MLR-Treg modulators. In the left column are depicted the results of negative control cultures (CD8 responders) in the absence of allogeneic stimulators or modulators. The figure is representative of 3 such experiments. (** = p<0.01; n = 3).

Figure 9. MLR-Tregs inhibit the expression of perforin, granzyme B and CD25 on responding CD8⁺ cells.

5×10⁴ PKH26 labeled purified responder CD8 cells were cultured with the original stimulators (1×10⁵) used in generating MLR-Tregs, in the presence of 1×10⁴ autologous modulator MLR-Tregs (right) vs. autologous modulator controls (Ax; middle). After 7 days in culture, the expression of intracellular Perforin-A, Granzyme-B, and membrane CD25 was assessed by flow cytometry. The CD8⁺ responder cells were gated and the PKH26^high non-proliferating and PKH26-diluted proliferating cells were analyzed. It was observed that the irradiated stimulators and Ax died off by day 7 (not shown); even the few that remained were gated out on CFSE vs. CD8 density-plot during the analysis. Note that there was a profound inhibition of both proliferation (PKH26 dilution) and expression of Perforin-A, Granzyme-B and CD25 by MLR-Tregs. This experiment is representative of 4 similar ones. (** = p<0.01; n = 4).