Abstract

Avoidance of immunosurveillance is presently considered as a hallmark of cancer. Indoleamine 2,3-dioxygenase (IDO) is a cytosolic enzyme which metabolizes L-tryptophan to kynurenines and induces T cell suppression either directly on T cells, or by altering antigen presenting cell function and generating regulatory T cells (Treg). Recently, IDO expression by cancer cells and/or cancer-associated stromal cells has been correlated with the induction of immune dysfunction by human solid tumors, such as ovarian and colon cancer. It has been previously shown that IDO may be expressed by human acute myeloid leukemia, where it promotes the differentiation of Treg cells (

Curti A et al.
Blood
2007
;
109
:
2871
–7
). Cyclooxygenase (COX) is the rate-limiting enzyme in the synthesis of prostaglandins and exists in two isoforms, COX-1 and COX-2. COX-2 is over-expressed by several tumor types and reportedly affects multiple pathways involved in tumorigenesis, including angiogenesis, invasion, and tumor-induced immune suppression. The present study aimed to determine whether COX-2 inhibitors may interfere with the IFN-γ-induced expression of IDO-1 in leukemia cells. The IDO-negative HL-60 acute myeloid leukemia cells, characterized by a CD34 CD33+CD119+ phenotype, were challenged with 100 ng/ml IFN-γ, a well-known inducer of IDO. Cytokine-treated HL-60 cells markedly up-regulated IDO-1 mRNA (~ 66-fold after 96 hours of culture) and protein and released a significant amount of kynurenines (12.58±2.66 μM after 72 hours and 19.21±9.8 μM after 96 hours of culture compared with 1.35±0.36 μM and 1.31±0.42 μM from untreated HL-60 cells; p =0.0004). This was associated with an average 35% depletion of tryptophan from culture supernatants. Not unexpectedly, IFN-γ-treated HL-60 cells promoted the in vitro conversion of naïve allogeneic CD4+CD25− T cells into bona fide CD4+CD25+FoxP3+ Treg cells. Phosphorylated signal transducer and activator of transcription (STAT)-1 could be detected in HL-60 cells starting from 48 hours of IFN-γ treatment, but not in untreated HL-60 cells, and its expression kinetics and relative amount closely paralleled those of IDO-1. IFN-γ strongly induced COX-2 mRNA (average fold-induction equal to 17.7±1.37 compared with untreated HL-60 cells) and protein. Conversely, IFN-γ exerted no appreciable effects on the expression levels of the tryptophan transporter molecule (LAT-1/CD98 receptor complex) on HL-60 leukemia cells. The addition of 100 μM nimesulide, a preferential COX-2 inhibitor, to HL-60 cells that were pre-treated with IFN-γ as above detailed translated into an average 54% reduction of kynurenines in culture supernatants (18.78±4.02 μM in the absence of nimesulide compared with 8.75±3.26 μM in cultures established with the drug; p = 0.0082). Superimposable results in terms of kynurenine production and tryptophan depletion were recorded in cultures of HL-60 cells that were simultaneously established with IFN-γ and nimesulide. Exposure to nimesulide attenuated mRNA signals for IDO-1, suggesting that the overall inhibition of IDO activity leading to suppressed kynurenine synthesis might also be attributed to the reduction of IDO-1 gene transcription. Interestingly, nimesulide diminished the expression of phosphorylated STAT-1 in HL-60 cells, indicating the involvement of this signaling pathway in the regulation of IDO-1 expression. In a further set of experiments, aspirin was added for 24 hours at 1 mM (final concentration) to IFN-γ-treated HL-60 cells in order to investigate its effects on IDO expression and function. In good agreement with the above results with nimesulide, aspirin promoted an average 65% inhibition of kynurenine release from IFN-γ-treated HL-60 cells and abrogated tryptophan consumption from culture supernatants. Collectively, these data suggest that COX-2 inhibition might be exploited as a valuable strategy to interfere with IDO-mediated tryptophan catabolism and hopefully to interfere with the tumor-induced immune dysfunction.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author