2020-12-17-tumor=immunity

Anti-tumour immunity induces aberrant peptide presentation in melanoma

Abstract

Extensive tumour inflammation, which is reflected by high levels of infiltrating T cells and interferon-γ (IFNγ) signalling, improves the response of  patients with melanoma to checkpoint immunotherapy1,2. Many tumours, however, escape by activating cellular pathways that lead to immunosuppression. One such mechanism is the production of  tryptophan metabolites along the kynurenine pathway by the enzyme  indoleamine 2,3-dioxygenase 1 (IDO1), which is induced by IFNγ3,4,5. However, clinical trials using inhibition of IDO1 in combination with  blockade of the PD1 pathway in patients with melanoma did not improve  the efficacy of treatment compared to PD1 pathway blockade alone6,7, pointing to an incomplete understanding of the role of IDO1 and the  consequent degradation of tryptophan in mRNA translation and cancer  progression. Here we used ribosome profiling in melanoma cells to  investigate the effects of prolonged IFNγ treatment on mRNA translation. Notably, we observed accumulations of ribosomes downstream of  tryptophan codons, along with their expected stalling at the tryptophan  codon. This suggested that ribosomes bypass tryptophan codons in the  absence of tryptophan. A detailed examination of these  tryptophan-associated accumulations of ribosomes—which we term  ‘W-bumps’—showed that they were characterized by ribosomal frameshifting events. Consistently, reporter assays combined with proteomic and  immunopeptidomic analyses demonstrated the induction of ribosomal  frameshifting, and the generation and presentation of aberrant trans-frame peptides at the cell surface after treatment with IFNγ. Priming of  naive T cells from healthy donors with aberrant peptides induced  peptide-specific T cells. Together, our results suggest that  IDO1-mediated depletion of tryptophan, which is induced by IFNγ, has a  role in the immune recognition of melanoma cells by contributing to  diversification of the peptidome landscape.

Fig. 1: IFNγ induces IDO1-mediated ribosome pausing on tryptophan codons,  and the formation of W-bumps downstream of these codons.

a, Schematic depicting the effect of IFNγ signalling on IDO1+ cells. IFNγ induction leads to increased expression of IDO1, which  catalyses the conversion of tryptophan to kynurenine. This leads to an  increase in uncharged tRNAs, which negatively affects the process of  protein translation. But, on the other hand, the production of  kynurenine inhibits T cell function. b, Metagene density profiles depicting global shifts of RPFs to the start of the coding sequence in  12T cells after treatment with IFNγ (red and yellow) as compared to mock treatment (control (ctrl)) (black and grey). The y axis shows the average intra-gene normalized density of RPFs for each transcript. UTR, untranslated region. c, Quantification of flow cytometry analysis of OPP incorporation assays  as a readout for nascent protein synthesis in mock-treated or  IFNγ-treated 12T cells. Data are mean ±s.d. of three independent  experiments. d, Diricore analysis line plots depicting  differential ribosome occupancy in 12T cells (5′-RPF) at −30 to +60  codons surrounding the initiator methionine (ATGstart) (green), tryptophan (red) and cysteine codons (grey). The y axis shows the ratio between the number of reads in IFNγ versus control conditions. e, f, Diricore analysis bar plots depicting differential codon usage (at position 15 of the RPFs) in IFNγ-treated versus control (e), and IFNγ-treated versus IFNγ + IDOi-treated (f) 12T cells. g, Diricore analysis line plots depicting differential ribosome occupancy  in 12T cells (5′-RPF) at −30 to +60 codons surrounding the initiator  methionine (ATGstart), tryptophan and cysteine codons in  IFNγ-treated cells versus control (grey) or IFNγ + IDOi-treated cells  versus control (red) comparisons.

Fig. 2: IFNγ-induced formation of W-bumps is associated with the presence of multiple tryptophan codons within a region of eight codons, and is  indicative of a reduction in protein synthesis.

a, Density of codons for alanine, serine and tryptophan, 60 codons  upstream and downstream of bumps identified with bump-finder after  treating 12T cells with IFNγ. b, Ratio of reads 30 codons upstream versus 30 codons downstream of bumps identified in control and IFNγ-treated conditions. c–f, RPF density in the 200 nucleotides surrounding the tryptophan codon (W) closest to bumps in control versus IFNγ-treated 12T cells (c), IDOi-treated versus IFNγ +IDOi-treated 12T cells (d), control versus tryptophan-depleted 12T cells (e) or control versus tyrosine (Y)-depleted 12T cells (f). g, Mean tryptophan codon enrichment in the ‘bumps’ group over the ‘no  bumps’ group. Red arrows indicate the enrichment of multiple alternating tryptophan codons within a region of eight codons in the bumps group  over the no bumps group. h, Box plots depicting the fold change  in protein levels (left (red); average of three replicates) and mRNA  levels (right (black); average of two replicates) in IFNγ-treated versus control cells. Proteins were grouped according to the number of  tryptophan residues in the protein sequence. P < 0.005, **P* < 0.0005, NS, not significant. Actual P values are 0.14, 1.2 × 10−3, 1.5 × 10−3 and 4.5 × 10−11 for the protein quantification and 0.8, 0.6, 0.7 and 0.9 for the RNA quantifications (right), in the order shown. i, Box plots depicting the fold change in protein levels (blue; average of three replicates) in IFNγ-treated versus control cells with inclusion  of proteasomal inhibition (MG132). Actual P values are 0.13, 3.2 × 10−4, 9.8 × 10−6 and 8.9 × 10−12, in the order shown. P values by two-tailed Wilcoxon test; boxes depict the first, second and third  quartiles and whiskers show the range excluding the outliers (h, i).