Multiple myeloma (MM) is a plasma cell (PC) malignancy characterized by a unique ability to evade immunosurveillance through the induction of antigen-presenting cell dysfunction, the release of immunoregulatory cytokines and the expansion of regulatory T cells (Treg). Hepatocyte growth factor (HGF), a mesenchyme-derived cytokine, is greatly elevated in MM and confers an unfavorable prognosis. We have previously shown that HGF induces the tryptophan-degrading enzyme indoleamine 2,3-dioxygenase (IDO) in human monocyte-derived dendritic cells. This study aimed to determine whether HGF may also promote IDO expression and/or function in MM plasma cells. We first measured IDO-1 expression by a panel of human MM cell lines using real-time PCR, before and after their in vitro treatment with 10 ng/ml IFN-γ. Exposure to IFN-γ translated into a 5-fold induction of IDO-1 mRNA in LP-1 and OPM-2 cells, in parallel with an increased ability to release kynurenines in culture supernatants. Conversely, mRNA signals for IDO-1 were undetectable in MOLP-8 and HUNS-1 cells. Next, the IDO-expressing MM cell lines were evaluated for their ability to release HGF. Whereas OPM-2 cells were incapable of producing HGF, LP-1 cells secreted copious amounts of HGF (8,393 pg/ml on average after 96 hours of culture), and promoted the in vitro conversion of naïve allogeneic CD4+CD25 T cells into bona fide CD4+CD25+FoxP3+ Treg cells, and this was reverted by the IDO inhibitor 1-methyl-tryptophan. In order to provide convincing evidence in favor of HGF effects on IDO expression, we challenged IDO-negative HUNS-1 cells with 100 ng/ml HGF. Under these conditions, an average 6-fold induction of IDO mRNA could be detected starting from 24 hours of culture. Accordingly, the addition of blocking antibodies against HGF to IDO-positive LP-1 cells was associated with an average 6-fold reduction of IDO mRNA. Interestingly, the co-culture of IDO-negative HUNS-1 cells with normal bone marrow (BM)-derived stromal cells promoted the induction of IDO mRNA, suggesting that microenvironmental interactions might also be implicated in IDO expression by the malignant PC. In a final set of ex vivo experiments, we purified CD56+CD138+ malignant PC from 9 patients with either newly diagnosed MM or with MM in partial remission. IDO-1 mRNA could be readily detected in 6 out of 9 cases (median IDO-1 mRNA content equal to 5 [range 2.1–14.1] relative to the HUNS-1 MM cells set arbitrarily at 1). IDO-1 mRNA levels in the malignant PC positively correlated with the frequency of circulating CD4+CD25+FoxP3+ Treg cells (p = 0.04). Furthermore, serum HGF was increased in patients’ peripheral blood (median 1,407 pg/ml, range 370–4940) and BM fluid (median 2,735 pg/ml, range 870–8,620) compared with healthy controls (median 1,000 pg/ml, range 710–1,300 and 1,875 pg/ml, range 1,310–2,430 in peripheral blood and BM, respectively). HGF levels positively correlated with the amount of M-component (p = 0.002), a measure of disease burden, and with the percentage of circulating Treg cells (p = 0.02). Serum kynurenines were increased in 6 patients with MM (median level equal to 3.8 μM, range 2.9–4.43, compared with 2.5 μM in healthy controls; p = 0.03). In line with this, serum tryptophan was significantly decreased in patients with MM (median value equal to 42.34 μM) compared with healthy controls (median value equal to 83.9 μM; p = 0.0012). Collectively, these data suggest that IDO-1 may be expressed by MM PC and that the IDO-mediated generation of immunosuppressive tryptophan metabolites may play a role in the in vivo expansion of the Treg compartment. Whether this will impact on the ability of the immune system to control disease progression remains to be prospectively investigated in larger cohorts of patients.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author