Extended Data Fig. 4: Reporter assays for the detection of tryptophan-associated out-of-frame events.

a, A hypothetical model suggesting a possible mechanism causative for W-bumps. In the normal scenario, ribosomes do not encounter problems when translating a tryptophan and progress translation at regular speed (top). Tryptophan shortage, on the other hand, can lead to stalling on the tryptophan codon (bottom left), or could in theory induce frameshifting events, leading to aberrant peptide production (bottom right). As the secondary structure of growing polypeptide chains is attained in the lower tunnel of the ribosome, the loss of an α-helical secondary structure in this tunnel could hamper ribosomal progression. b, Box plot depicting bump-score (from two replicates) in MD55A3 cells from the group of selected peptides with ordered out-of-frame peptides (‘Selected’ from Fig. 3b) and every tryptophan in the proteome (‘All’ from Fig. 3b). Boxes depict first, second and third quartiles; whiskers depict the range excluding the outliers. Test: two-sample t-test; *P = 0.056. c, Schematic representation of the final protein sequences that would form due to frameshifting events. The in-frame construct (top) contains a His tag and would end up in the pull-down (PD) fraction. Whenever a frameshift occurs at the position of the tryptophan in ATF4, this protein would lose its His tag, and consequently would end up in the supernatant fraction (S). The +1 and +2 out-of-frame constructs (bottom) do not contain a His tag, whereby the resulting proteins always end up in the supernatant fraction (S) in a His-tag pull-down assay. When frameshifting events take place, however, the His tag is incorporated into the peptide, whereby the resulting protein ends up in the pull-down fraction (PD). d, Western blot analysis of a sequential V5-tag immunoprecipitation on the supernatant samples of the Frame reporter expressing cells from Fig. 3e (n = 1). The image indicates the presence of V5-tag-containing peptides that do not contain a His tag generated from the in-frame reporter. e, Tryptophan levels in MD55A3 cells expressing the V5-ATF4-His reporter constructs that were used in Fig. 3e and in d. Tryptophan levels were analysed by mass spectrometry after 48 h of treatment. Bars represent the average of three independent replicates ± s.d. f, Tryptophan levels in MD55A3 cells expressing the V5-ATF4-His reporter constructs that were used in h. Tryptophan levels were analysed by mass spectrometry after 24 and 48 h of treatment in triplicate. g, Western blot analysis showing V5-tagged peptides in pull-down samples of MD55A3 reporter cells that were mock-, IFNγ- or IFNγ + IDOi-treated (n = 2). h, Western blot analysis of V5-tagged proteins in pull-down samples of MD55A3 reporter cells that were either mock-treated or cultured in tryptophan-depleted medium (n = 1). i, Western blot analyses showing pull-down assays followed by V5 staining on western blot of MD55A3 cells expressing the original reporter constructs as depicted in Fig. 3d (Wt), or MD55A3 cells expressing the same reporters where the tryptophan codon was mutated to a codon for tyrosine (Y mut, n = 1). j, Amino acid levels as determined by mass spectrometry from lysates of Tyr-depleted cells (48 h) versus control cells. Bars represent the average of three independent replicates ± s.d. k, Western blot analysis showing V5-tagged peptides in pull-down samples of MD55A3 reporter cells that were mock-treated or depleted of tyrosine for 48 h (n = 1). l, Flow cytometry analyses showing the histograms obtained of MD55A3 cells expressing the V5-ATF4-tGFP reporters in all 3 frames in mock- and IFNγ-treated conditions. Plots are a representative graph out of a triplicate biological experiment. m, Flow cytometry analyses showing the quantification of histograms obtained in l. Bars represent the average of three independent replicates ± s.d., *P values in order from left to right: 4.0 × 10⁻⁶ and 9.5 × 10⁻⁴ as determined by a two-sided t-test. n, Anti-V5-tag and anti-tGFP western blot analysis of whole cell lysates from MD55A3 cells expressing the indicated reporters, which were subjected to mock or IFNγ treatments. In each blot the position of the full-length in-frame protein and the shorter out-of-frame protein are marked by the arrowheads (n = 2).o, Western blot analyses with anti-V5 antibody of His-tag pull-down samples of 888-Mel and D10 cells expressing the in-frame and +1 reporters (n = 2). The cells were either grown in isolation (−), or co-cultured with MART-1 specific T cells for 16 h (+) before the pull-down was performed. p, Western blot analysis showing IDO1 upregulation in 888-Mel and D10 cells in control and T cell co-culture conditions (n = 2). The same cells were used for pull-down experiments in Fig. 3m. q, His-tag pull-down was executed on the lysates of mock and IFNγ-treated 888-Mel and D10 cells expressing the in-frame and +1 reporters (n = 2). Both supernatant (S) and His-tag pull-down samples (PD) of these cells were stained with V5 antibodies. r, The lysates of cells used in q were used for a western blot analysis to show the level of IDO1 induction (n = 1). s, Tryptophan levels as determined by mass spectrometry analysis of 888-Mel and D10 lysates in mock, IFNγ-treated and tryptophan-depleted conditions. Bars represent the average of three independent replicates ± s.d., except for the first bar, which is an average of two independent replicates. t, Western blot analyses with anti-V5 antibody of His-tag pull-down samples of 888-Mel and D10 that were mock-treated (+), or grown in tryptophan-less medium for 48 h (−) before the pull-down was performed (n = 2).