Monocyte-mediated T-cell suppression and augmented monocyte tryptophan catabolism after human hematopoietic stem-cell transplantation

Abstract

T-cell dysfunction after human hematopoietic stem-cell transplantation (HSCT) is generally attributed to intrinsic T-cell defects. Here we show that the characteristic impaired proliferative responses to polyclonal stimulation of post-HSCT peripheral blood mononuclear cells (PB-MCs) were markedly (4-fold) improved by T-cell enrichment. Conversely, addback of post-HSCT monocytes to these enriched T cells dampened their proliferative responses, suggesting that post-HSCT monocytes effectively mediate T-cell suppression. As a mechanism possibly contributing to monocyte-mediated T-cell suppression, we investigated monocyte tryptophan catabolism by indoleamine 2,3-dioxygenase into kynurenine, which has been implicated in regulating T-cell responses. Compared with controls, all post-HSCT monocyte-containing cell cultures (total PBMCs, monocytes, and monocyte/T-cell cocultures), but not monocyte-depleted populations, secreted elevated amounts of kynurenine. Blockade of tryptophan catabolism improved the proliferative responses. The slightly increased kynurenine release and substantial release of neopterin by unstimulated post-HSCT monocytes suggests that they were in a state of continuous activation. Superimposed on this state, stimulation of these cells caused a striking, additional increase (10-fold) in kynurenine release, and they triggered marked apoptosis of autologous post-HSCT T cells. We conclude that the amplified kynurenine release by post-HSCT monocytes, particularly induced upon stimulation, may underlie their suppressor activity, which in turn may contribute to the depressed T-cell immune responses after HSCT.

Figure 1.

Monocytes mediate inhibition of T-cell proliferative responses after HSCT. (A) Total PBMCs or T cells highly enriched from total PBMCs obtained from healthy controls (□) or HSCT recipients (▪) were stimulated by immobilized anti-CD3 (250 ng/mL) for 72 hours and the proliferative response was quantified by net [³H]thymidine incorporation. Each bar depicts the median proliferative response of 15 experiments. Proliferation of post-HSCT PBMCs was significantly lower than that of control PBMCs (P < .001). The asterisk indicates that in HSCT recipients proliferation of enriched T cells was significantly increased over that of total PBMCs (P < .001). (B) T cells were cocultured with graded proportions of autologous monocytes as indicated and stimulated as in panel A. The bars depict net [³H]thymidine incorporation of control cultures (□) and cultures obtained from HSCT recipients (▪) as percent of the response of the respective purified T cells alone (horizontal line = 100%). ^*Significant (P < .05) differences of post-HSCT and control cultures. Error bars indicate standard error of the mean (SEM).

Figure 2.

Monocyte-containing cell cultures after HSCT contain elevated amounts of kynurenine. (A) The culture supernatants of cell populations assessed for their proliferative response to immobilized anti-CD3 in Figure 1 were examined for kynurenine content by HPLC. Each point represents the μM kynurenine released per single experiment in cultures of total PBMCs or of T cells, maintained in resting conditions (top row) or stimulated with anti-CD3 (bottom row). ♦ indicates controls; ▪, HSCT. The median of 15 experiments is shown by the horizontal line. (B) T cells were cocultured with increasing proportions of autologous monocytes (as in Figure 1B). The bars indicate the median fold increase of kynurenine release over that of T cells alone. □ indicates controls; ▪, HSCT. - indicates not done. ^*Significant differences of post-HSCT and control cultures (P < .05). Error bars indicate SEM.

Figure 3.

IDO expression and activity by CD14⁺ monocyte population. (A) Total PBMCs obtained from normal controls or from HSCT recipients were cultured in medium supplemented with M-CSF (20 ng/mL) for 4 days, and IFN-γ (100 U/mL) that was added for the last 18 hours of culture and examined for IDO expression by flow cytometry as described in “Patients, materials, and methods.” Gray shaded histograms represent the CD14⁺ cell population; dark lines, the CD14^- cell population; and light gray lines, negative control (second antibody only). (B) Total PBMCs were separated into a CD14⁺ (diamonds) and a CD14^- (squares) cell population and were cultured in medium supplemented with M-CSF (20 ng/mL) for 4 days. During the last 18 hours of culture, graded amounts of IFN-γ were added as indicated. At the end of culture, concentrations of neopterin (top row) and of kynurenine (bottom row) were determined in cell culture supernatant by ELISA and HPLC, respectively. Results of one experiment, representative of 3 experiments, are shown.

Figure 4.

Inhibition of post-HSCT PBMC proliferation is improved by blockade of IDO activity with MDLT. The effect of the addition of 0.5 M MDLT on the proliferative responses of control PBMCs (□) and of post-HSCT PBMCs (▪) to PHA was assessed. Cells were stimulated with PHA and, in parallel, exposed to MDLT, either immediately after separation from whole blood (fresh) or after pre-exposure to M-CSF (20 ng/mL) for 4 days, and IFN-γ (100 U/mL) that was added for the last 18 hours of culture. Error bars indicate SEM.

Figure 5.

Post-HSCT monocyte suppressor activity affects post-HSCT T cells. (A) Total PBMCs obtained from a control (□) or an HSCT recipient (▪) were stimulated by highly enriched allogeneic control monocytes and proliferative responses were determined by [³H]thymidine incorporation after 5 days of culture. (B) Highly enriched T cells (Tc) or T cells cocultured with their autologous monocytes (Tc/m) of controls (□) or of HSCT recipients (▪) were stimulated by allogeneic cells and their proliferative response determined as in Figure 5A. (C) In a crossover experiment, control monocytes were stimulators of allogeneic post-HSCT T cells (□) and, vice versa, post-HSCT monocytes were stimulators of allogeneic control T cells (▪), and the respective proliferative responses were determined as in Figure 5A. Error bars indicate SEM.

Figure 6.

Post-HSCT monocytes mediate T-cell apoptosis. Total PBMCs or highly enriched T cells alone or highly enriched T cells cocultured with highly enriched autologous monocytes (40%), were stimulated with PHA for 3 days. T cells were identified as CD3⁺ and gated. Histogram plots depict apoptotic T cells identified as an annexin V-positive and propidium iodide-negative population. Assessment of the mean peak channel (mpc) in histogram analyses was used to detect differences in density of cell-surface binding of annexin V. Open histograms represent control cells; filled histograms, post-HSCT cells. MPC values are shown at the top right of each panel.