Evaluation of Novel Inhibitors of Tryptophan Dioxygenases for Enzyme and Species Selectivity Using Engineered Tumour Cell Lines Expressing Either Murine or Human IDO1 or TDO2

Abstract

Indoleamine 2, 3-dioxygenase 1 (IDO1) is commonly expressed by cancers as a mechanism for evading the immune system. Preclinical and clinical studies have indicated the potential of combining IDO1 inhibitors with immune therapies for the treatment of cancer, strengthening an interest in the discovery of novel dioxygenase inhibitors for reversing tumour-mediated immune suppression. To facilitate the discovery, development and investigation of novel small molecule inhibitors of IDO1 and its hepatic isozyme tryptophan dioxygenase (TDO2), murine tumour cells were engineered to selectively express either murine or human IDO1 and TDO2 for use as tools to dissect both the species specificity and isoenzyme selectivity of newly discovered inhibitors. Lewis lung carcinoma (LLTC) lines were engineered to express either murine or human IDO1 for use to test species selectivity of the novel inhibitors; in addition, GL261 glioma lines were engineered to express either human IDO1 or human TDO2 and used to test the isoenzyme selectivity of individual inhibitors in cell-based assays. The 20 most potent inhibitors against recombinant human IDO1 enzyme, discovered from a commissioned screening of 40,000 compounds in the Australian WEHI compound library, returned comparable IC₅₀ values against murine or human IDO1 in cell-based assays using the LLTC-mIDO1 and LLTC-hIDO1 line, respectively. To test the in vivo activity of the hits, transfected lines were inoculated into syngeneic C57Bl/6 mice. Individual LLTC-hIDO1 tumours showed variable expression of human IDO1 in contrast to GL261-hIDO1 tumours which were homogenous in their IDO1 expression and were subsequently used for in vivo studies. W-0019482, the most potent IDO1 inhibitor identified from cell-based assays, reduced plasma and intratumoural ratios of kynurenine to tryptophan (K:T) and delayed the growth of subcutaneous GL261-hIDO1 tumours in mice. Synthetic modification of W-0019482 generated analogues with dual IDO1/TDO2 inhibitory activity, as well as inhibitors that were selective for either TDO2 or IDO1. These results demonstrate the versatility of W-0019482 as a lead in generating all three subclasses of tryptophan dioxygenase inhibitors which can be applied for investigating the individual roles and interactions between IDO1 and TDO2 in driving cancer-mediated immune suppression.

Keywords: cancer models; immune modulation; tryptophan dioxygenase inhibitors.

Figure 1

(A) Expression of human IDO1 in NZM lines without and after incubation with hIFNγ. (B). Expression of murine or human IDO1 in LLTC−WT, LLTC-mIDO1 and LLTC-hIDO1 lines without or after 72 h incubation with mIFNγ. (C) LLTC-WT, LLTC-mIDO1 and LLTC-hIDO1 cells immunostained with anti-human IDO1 or anti-mouse IDO1antibodies. (D) IC₅₀ values measured in cell-based assays against LLTC-hIDO1 plotted against IC₅₀ values measured against LLTC-mIDO1 for WEHI compounds (red); reference controls INCB14943 and 4-PI (blue). Spearman’s ρ and the p-value in brackets denote Spearman’s rank correlation coefficient and the probability value of that correlation, respectively.

Figure 2

(A) Growth of LLTC-WT (black circle) and LLTC-hIDO1 (red square) subcutaneous tumours in C57Bl/6 mice (mean ± s.e.m. of n = 6 per group). p-values determined using two-way Anova and * denotes significance p ≤ 0.05. (B) Western blot of hIDO1 or mIDO1 protein in four individual 14-day LLTC-hIDO1 tumours (T1–T4) and in LLTC-hIDO1, LLTC-WT and LLTC-mIDO1 cells. (C) Western blots of hIDO1 expression in individual 22-day tumours from sequential generations of implants from a high-IDO1-expressing LLTC-hIDO1 tumour (T1).

Figure 3

(A) Growth of GL261-WT (black circles) and GL261-hIDO1 (red squares) tumours implanted subcutaneously in syngeneic C57Bl/6 mice. p-values determined using two-way Anova and * denotes significance p ≤ 0.05 (B) Western blot of hIDO1 or mIDO1 expression in GL261-WT or GL261-hIDO1 cells or 14-day GL261-WT and GL261-hIDO1 tumours. (C) Plasma K:T ratios measured in mice with GL261-WT (a) or GL261-hIDO1 (b) tumours on day 14 (black), day 17 (red) or day 21 (blue) after implantation. (D) Plasma K:T ratio plotted against tumour weight and analysed by linear regression and Pearson correlation. (E) Percentage of CD45⁺ cells from GL261-WT (blue) and GL261-hIDO1 (red) tumours that were B220⁺ B-lymphocytes or CD3⁺ T-lymphocytes. (F) Percent of CD3⁺ T-lymphocytes in GL261-WT or GL261-hIDO1 tumours that were CD4⁺ T-helpers (white), Foxp3⁺ suppressors (hatched), CD8⁺ cytotoxic T-cells (green) and CD4/CD8 null cells (orange). Data are from tumours (n = 3), enzyme digested and the single cells in suspension immunostained with antibodies for the various subtypes and then quantitated by flow cytometry.

Figure 4

Cell viability (black triangles) and inhibition of hIDO1 activity measured using the LLTC-hIDO1 cell-based assay (blue squares) compared to that measured using an enzymatic assay (red circles) with increasing concentrations of (a) W-0019482, (b) NLG919 and (c) INCB14943.

Figure 5

(A) K:T ratios at indicated times following treatment with W-0019482 (150 mg/kg) collected from mice with 16-day GL261-hIDO1 tumours (15–20 mm in size), in (a) plasma (red) and (b) tumours (blue). Bars represent mean ± s.e.m. of 3 mice per time point. DMSO vehicle control (black), bar represents mean ± s.e.m. from 21 samples pooled from 3 mice per timepoint following treatment with DMSO. * and ** denote significance (p < 0.05, p <0.01, respectively) by one-way ANOVA and Sidak’s multiple comparisons compared to DMSO controls. (B) Growth of subcutaneous GL261-hIDO1 tumours following daily ip injection of W-0019482 at 75 mg/kg (red squares), 100 mg/kg (blue triangles), and 150 mg/kg (green) compared to DMSO vehicle controls (black). * denotes significant difference (p < 0.05) between treated and vehicle controls by two-way ANOVA and Tukey’s multiple comparisons.

Figure 6

Percent inhibition of hIDO1 (red square) and hTDO2 (blue circle) activity and cell viability (% of control) of LLTC-hIDO1 (red triangle) and GL261-hTDO2 (blue triangle) in cell-based assays versus concentration of inhibitor for (A) Dual IDO1/TDO2 inhibitors; (B) hTDO2 selective; (C) hIDO1 selective; and (D) Reference controls W-0019482 and INCB14934.