Indoleamine 2,3 dioxygenase and human leucocyte antigen-G inhibit the T-cell alloproliferative response through two independent pathways

Abstract

Both human leucocyte antigen (HLA)-G and indoleamine 2,3 dioxygenase (IDO) are key molecules involved in immune tolerance. HLA-G is a non-classical HLA class I molecule that can be expressed in both membrane-bound (HLA-G1) and soluble (HLA-G5) forms, both of which exhibit tolerogenic properties via interaction with inhibitory receptors present on natural killer (NK) cells, T cells and antigen-presenting cells (APC). IDO is an enzyme that acts by depleting the surrounding microenvironment of the essential amino acid, tryptophan, thereby inhibiting T-cell proliferation. Our present study was aimed at analysing the potential link that may exist between IDO and HLA-G. Our results showed that during allogeneic reactions, soluble HLA-G expression was not regulated by the addition of IDO substrate (i.e. tryptophan), metabolite (i.e. kynurenine) or inhibitor (i.e. 1-methyl-tryptophan), that IDO activity was not altered by HLA-G5 treatment, and that HLA-G5-mediated inhibition of the T-cell alloproliferative response was neither affected by the presence of tryptophan and kynurenine nor reversed after IDO activity blockage, demonstrating that HLA-G5 can exert its function in the absence of functional IDO. Similarly, inhibition of the T-cell alloresponse, induced by HLA-G1-expressing antigen-presenting cells, was not altered by IDO metabolites or inhibitor. Taken together, these findings show that the function and expression of IDO and HLA-G5 are not mutually influenced, but rather inhibit the T-cell alloproliferative response through two independent pathways. IDO and HLA-G are thus complementary for inducing and maintaining immune tolerance in physiological (pregnancy) and pathological (tumour and allograft) situations.

Figure 1

Peripheral blood mononuclear cells (PBMC) sensitized for 18 hr with human leucocyte antigen (HLA)-G5 are unresponsive to subsequent allogeneic stimulus. PBMC from healthy donors (1 and 2) were pretreated for 18 hr with adherent and confluent γ-irradiated (125 Gy) M8-HLA-G5 cells (HLA-G5-sensitized PBMC) or M8-pcDNA cells (control PBMC). These pretreated PBMC were then used as responder cells towards irradiated stimulator PBMC (1*, 2*, 3*, 4*), in a classical mixed lymphocyte reaction (MLR), at a ratio of 1 : 1. Tritiated thymidine incorporation was measured after 6 days of the MLR. Five representative allogeneic combinations are shown. Results are expressed as the mean ± standard deviation (SD) thymidine incorporation [in counts per minute (c.p.m.)] of triplicate wells, corrected for background values (Δc.p.m.). Stimulator cells are indicated by an asterisk (*).

Figure 2

Inhibition of the alloproliferative response by the main indoleamine 2,3 dioxygenase (IDO) metabolite, kynurenine. A total of 0·5 mm tryptophan, kynurenine or 1-methyl-tryptophan was added on day 0 of the classical mixed lymphocyte reaction (MLR) between responder PBMC from two healthy donors (5 and 7) and irradiated stimulator peripheral blood mononuclear cells (PBMC) (5*, 6*, 7*, 8*). Tritiated thymidine incorporation was measured after 6 days of the MLR. Two out of nine allogeneic combinations are shown. Similar results were obtained by using 1 mm tryptophan, kynurenine or 1-methyl-tryptophan (data not shown). Results are expressed as the mean ± standard deviation (SD) thymidine incorporation [in counts per minute (c.p.m.)] of triplicate wells, corrected for background values (Δc.p.m.). Stimulator cells are indicated by an asterisk (*).

Figure 3

Indoleamine 2,3 dioxygenase (IDO) and secreted human leucocyte antigen (HLA)-G5 inhibit the T-cell alloproliferative response through two independent pathways. Peripheral blood mononuclear cells (PBMC) from one healthy donor (donor 9) were pretreated for 18 hr with the culture supernatant of M8-HLA-G5 cells (HLA-G5-sensitized PBMC) or M8-pcDNA cells (control PBMC). These pretreated PBMC were then used as responder cells in a classical mixed lymphocyte reaction (MLR) towards stimulator PBMC from one donor (donor 10*). On the first day of the MLR, 0·5 mm tryptophan, kynurenine, or 1 methyl-tryptophan was added. Tritiated thymidine incorporation was measured after 6 days of the MLR. One representative allogeneic combination out of five is shown. Similar results were obtained when experiments were carried out with adherent and confluent irradiated (125 Gy) M8-HLA-G5 or M8-pcDNA cells instead of the supernatants, as well as by using IDO metabolites at a concentration of 1 mm (data not shown). Results are expressed as the mean ± standard deviation (SD) thymidine incorporation [in counts per minute (c.p.m.)] of triplicate wells, corrected for background values (Δc.p.m.).

Figure 4

A mild human leucocyte antigen (HLA)-G5-mediated inhibition is not altered under various concentrations of indoleamine 2,3 dioxygenase (IDO) metabolite or inhibitor. (a) Peripheral blood mononuclear cells (PBMC) from one healthy donor (donor 11) were pretreated for 18 hr with the culture supernatant of M8-HLA-G5 cells (HLA-G5-sensitized PBMC) or were not pretreated (medium). These PBMC were then used as responder cells in a classical mixed lymphocyte reaction (MLR) towards stimulator PBMC from one donor (donor 12*). On the first day of the MLR, various doses of kynurenine (0·01 to 20 mm) were added. (b) PBMC from one healthy donor (donor 13) were pretreated for 18 hr with the culture supernatant of M8-HLA-G5 cells (HLA-G5-sensitized PBMC) or not pretreated (medium). These PBMC were then used as responder cells in a classical MLR towards stimulator PBMC from one donor (donor 14*). On the first day of the MLR, various doses (0·01 to 20 mm) of 1-methyl-tryptophan were added. Tritiated thymidine incorporation was measured after 6 days of the MLR. One representative allogeneic combination out of three is shown. Results are expressed as the mean ± standard deviation (SD) thymidine incorporation [in counts per minute (c.p.m.)] of triplicate wells, corrected for background values (Δc.p.m.).

Figure 5

Indoleamine 2,3 dioxygenase (IDO) and membrane-bound human leucocyte antigen (HLA)-G1 inhibit the T-cell alloproliferative response through two independent pathways. LCL (HLA-G1-negative stimulator cells) or LCL-HLA-G1 cells (HLA-G1-positive stimulator cells) were γ-irradiated (75 Gy) and used as allogeneic stimulator cells against responder PBMC from one healthy donor (donor 15). On the first day of the mixed lymphocyte reaction (MLR), various concentrations of kynurenine (a) or 1-methyl-tryptophan (b) were added. Tritiated thymidine incorporation was measured after 6 days of the MLR. One allogeneic combination out of three is shown as a representative result. Results are expressed as the mean ± standard deviation (SD) thymidine incorporation [in counts per minute (c.p.m.)] of triplicate wells, corrected for background values (Δc.p.m.).