Supplemental Information. Oncogenic RAS Signaling Promotes Tumor. Immunoresistance by Stabilizing PD-L1 mrna

Transcription

1 Immunity, Volume 7 al Information Oncogenic RAS Signaling Promotes Tumor Immunoresistance by Stabilizing PD-L mrna Matthew A. Coelho, Sophie de Carné Trécesson, Sareena Rana, Davide Zecchin, Christopher Moore, Miriam Molina-Arcas, Philip East, Bradley Spencer- Dene, Emma Nye, Karin Barnouin, Ambrosius P. Snijders, Wi S. Lai, Perry J. Blackshear, and Julian Downward

2 al Figure Legends Figure S, related to Figure. Cell-intrinsic Upregulation of PD-L through Oncogenic RAS Signalling (A) qpcr analysis of PD-L mrna expression at 6 h and h after addition of -OHT or EtOH vehicle in ER-HRAS GV MCFA cells. Mean ± SEM of two independent experiments. P<.; unpaired, two-tailed Student s t- tests. (B) Flow cytometry analysis of PD-L surface protein expression at h and four days after addition of -OHT or EtOH vehicle in ER-HRAS GV MCFA cells and four days in ER-HRAS GV HKE- cells. (C) Flow cytometry analysis of PD-L surface protein four days after addition of -OHT or EtOH vehicle to parental MCFA and HKE- cells. (D) Western blotting analysis of ER-KRAS GV type II pneumocytes treated with -OHT, MEK inhibitor or PIK inhibitor in starvation medium for h. (E) qpcr analysis of PD-L expression in H and H79 cells h after addition of MEK inhibitor or PIK inhibitor or the combination. Mean ± SEM of three (H) or two (H79) independent experiments. P<.5, *P<.5; unpaired, two-tailed Student s t-tests. (F) qpcr analysis of PD-L expression in H58 and H cells h after addition of the ERK/ inhibitor SCH7798 (5 nm). Mean ± SD of biological duplicates. P<.; unpaired, two-tailed Student s t-tests. (G) qpcr analysis of PD-L expression in H79 cells treated with PMA for h following a min pre-treatment with or MEK inhibitor. Data represent the mean ± SEM of two independent experiments. P<.; unpaired, two-tailed Student s t-test. (H) Surface expression of PD-L was measured by flow cytometry. MFI values are adjusted for the isotype control in each condition. Mean ± SEM of biological duplicates. (I) qpcr analysis of transcripts encoding antigen processing and presentation machinery in ER-KRAS GV type II pneumocytes simulated with -OHT for

3 h in starvation medium. Data represent the mean ± SEM of four independent experiments. P<.; paired, two-tailed Student s t-test. MFI, Mean Fluorescence Intensity. -OHT, nm. IFN-γ, ng/ml. MEK inhibitor GSK, 5 nm. PIK inhibitor GDC-9, 5 nm. PMA, nm.

4 Figure S, relating to Figure. RAS Signalling Increases PD-L mrna Stability through AU-rich Elements in the UTR (A) Normalised luciferase signal from the indicated human CD7 promoter region reporter constructs in H58 cells treated for 6.5 h with medium only, PMA ( nm) or IFN-γ ( ng/ml). Numbering corresponds to the GRCh8 assembly. Data are representative of two independent experiments. P<.5, two-way ANOVA. (B) Stability of murine PD-L mrna measured by qpcr after the addition of actinomycin D (5 μg/ml) and or PIK inhibitor GDC-9 (5 nm). KPB6 cells were pre-treated with or PIK inhibitor for min before actinomycin D addition. Data represent the mean ± SEM and are normalised to the h time point when actinomycin D was added, and are representative of two independent experiments. (C) Stability of murine Tusc mrna (left panel) and Ptgs mrna (right panel) measured by qpcr after the addition of actinomycin D (5 μg/ml) and or MEK inhibitor GSK (5 nm). KPB6 cells were pre-treated with or MEK inhibitor for min before actinomycin D addition. Data represent the mean ± SEM and are normalised to the h time point when actinomycin D was added.

5 Figure S, relating to Figure. AU-rich element Binding Proteins TTP and KSRP are Negative Regulators of PD-L Expression (A-C) qpcr analysis of knockdown efficiency 8 h after sirna transfections, relative to rambled control. Data represent the mean ± SD of triplicates. (D) qpcr analysis of PD-L expression h after transfection with the indicated constructs. Data represent the mean ± SEM of two independent experiments. P<.; P<.; unpaired, two-tailed Student s t-test. (E) qpcr analysis of PD-L expression in H cells 8 h after transfection with sirnas targeting AU-rich binding proteins (AU-BPs), relative to rambled () control. Data represent the mean ± SEM of two independent experiments. (F) qpcr analysis of knock-down efficiency in H cells 8 h after sirna transfections, relative to rambled control. Data represent the mean ± SEM of two independent experiments. (G) Western blotting analysis of TTP expression in H, A7 and H58 cells 8 h after sirna transfection with sirna pools against TTP relative to rambled. Overexpression of Myc-TTP serves as a positive control for immunodetection. (H) qpcr analysis of PD-L and TTP expression in H and H58 cells 8 h after sirna transfection with sirna pools or single sirnas against TTP relative to rambled. Data represent the mean ± SEM of biological duplicates.

6 Figure S, relating to Figure. RAS Regulates PD-L Expression through TTP (A) qpcr analysis of PD-L and KSRP expression in H58 cells following sirna mediated knock-down of KSRP ( h) followed by MEK inhibition ( h) with GSK (5 nm). Data represent the mean ± SEM of two independent experiments. P<.; *P<.; P<.5; *P<.5; n.s; not significant. (B) qpcr analysis of PD-L mrna from RNA immunoprecipitates using IgG control or anti-ttp antibody, or anti-ksrp antibody, using KPB6 cells pretreated with or MEK inhibitor GSK for 5.5 h (5 nm). Data represent the mean ± SD of biological triplicate IPs. (C) qpcr analysis of Gapdh mrna (control, lacking AU-rich elements) from RNA immunoprecipitates using IgG control or anti-ttp antibody, or anti- KSRP antibody, using KPB6 cells pre-treated with or MEK inhibitor GSK for 5.5 h (5 nm). Data represent the mean ± SD of biological triplicate IPs. (D) qpcr analysis of PD-L expression in H58 cells after transfection with empty, wild-type KSRP or phospho-mutant KSRP S9A constructs. Data represent the mean ± SEM of two independent experiments. P<.; *P<.; NS, not significant. Unless otherwise stated, data were compared using unpaired, two-tailed Student s t-tests. 5

7 Figure S5, relating to Figure 5. RAS-ROS-p8 Signalling Controls TTP Activity (A) MS/MS spectra for phosphopeptides STphSLVEGR (S5) and QSIphSFSGLPSGR (S78). -98 indicates the loss of H PO. (B) Flow cytometry analysis of PD-L surface protein, and intracellular ROS measured by staining with H DCFDA, in MCFA ER-HRASGV cells treated with -OHT or vehicle ± NAC ( mm) for h. The same dataset is represented as a dot-plot and a histogram and data are representative of two independent experiments. (C) Western blotting analysis of MCFA cells harbouring an inducible version of the kinase domain of MEKK (ER- MEKK), h after the addition of - OHT ( nm) or vehicle. (D) qpcr analysis of PD-L mrna expression h after treatment with NAC ( mm), reduced glutathione ( mm) or MK inhibitor III ( μm). Data represent the mean ± SEM of two independent experiments. P<., *P<.5, P<.; two-tailed Student s t-tests comparing to control condition. (E) Western blotting analysis of CT6 TTP KO cells harbouring doxycyclineinducible WT or phospho-mutant Myc-TTP constructs, treated with doxycycline or vehicle for h. Arrow indicates Myc-TTP. 6

8 Figure S6, relating to Figure 6. RAS Pathway Activation is Associated with PD-L Upregulation in Human Cancers (A) Heat maps showing LUAD and COAD samples from the TCGA dataset clustered into RAS high or low pathway activity groups using RNA sequencing expression data and published RAS activity gene expression signatures (Loboda et al., ; Sweet-Cordero et al., 5). KRAS mutation status (codons, and 6) is shown for each sample. (B) qpcr analysis of PD-L and IFNGR expression in ER-KRAS GV type II pneumocytes h after treatment with vehicle, -OHT ( nm), or -OHT with IFN-γ blocking antibody ( μg/ml) or with ruxolitinib (5 nm). Mean ± SEM of two independent experiments. The panel on the right shows a qpcr analysis of PD-L expression in ER-KRAS GV type II pneumocytes following a h treatment with IFN-γ ( ng/ml) and IFN-γ with IFN-γ blocking antibody ( μg/ml) or with ruxolitinib (5 nm) to verify blocking of IFN-γ IFGR signalling. (C) TTP mrna expression in human patient lung and colon normal tissue versus adenocarcinoma, from publically available datasets (Selamat et al., ; Skrzypczak et al., ). Wilcoxon signed-rank test. (D) qpcr analysis of PD-L and TTP expression in FACS purified CD5 - CD - DAPI - EpCAM + cells derived from lung tumours or matched normal adjacent lung tissue from Kras LSL-GD/+ ; Trp5 F/F mice. Each point represents data from an individual mouse and is normalised to the matched normal lung tissue. *P<.5; unpaired, two-tailed Student s t-tests. (E) Flow cytometry analysis of PD-L expression on CD5-CD-DAPI- cells derived from macroscopically dissected lung tumours or normal adjacent lung tissue from Kras LSL-GD/+ ; Trp5 F/F mice. Each point represents data from an individual mouse and is normalised to the matched normal lung tissue. Data are pooled from two independent experiments. MFI; Mean Fluorescence Intensity. *P<.5; unpaired, two-tailed Student s t-test. 7

9 Figure S7, relating to Figure 7. Restoration of Tumour Cell TTP Expression Enhances Anti-tumour Immunity (A) Stability of murine PD-L mrna measured by qpcr analysis. CT6 TTP (teton) cells were pretreated with doxycycline (Dox.; μg/ml) or vehicle for 6 h and then MEK inhibitor (GSK, trametinib; 5 nm) for an additional min before actinomycin D (ActD; μg/ml) was added. Data are normalised to time h when ActD was added and represent the mean ± SEM of two independent experiments. (B) Western blotting analysis of stable MC8 cell lines expressing Myctagged, mouse TTP under a tetracycline-inducible promoter (TTP tet-on). Cells were treated with doxycycline (Dox.,. μg/ml or μg/ml) or vehicle in starvation medium for h before analysis. Arrow indicates Myc-TTP. (C) qpcr analysis of PD-L expression from stable MC8 (TTP tet-on) cell lines treated with doxycycline (Dox., μg/ml) or vehicle in starvation medium for h before analysis. Data represent the mean ± SEM of biological duplicates. (D) Confluency was measured using IncuCyte for CT6 stable derivative cell lines treated with the indicated concentrations of doxycycline or vehicle at t = h. Data represent the mean ± SD of biological triplicates and are representative of two independent experiments. (E) Confluency was measured using IncuCyte for MC8 stable derivative cell lines treated with the indicated concentrations of doxycycline or vehicle at t = h. Data represent the mean ± SD of biological triplicates. (F) Representative histograms from flow cytometry analysis of PD-L surface expression in TTP (teton) CT6 stable cells lines expressing endogenous PD-L, PD-L + wild-type UTR cdna, or PD-L UTR cdna, after treatment with doxycycline (Dox., μg/ml) or vehicle for 7 h. Data are representative of two independent experiments and also form part of Figure 7B. (G) Tumour growth curves for the indicated CT6-derived cell lines shown in Figure S7C, subcutaneously transplanted into BALB/c mice (n = 5 WT UTR + Dox and UTR + H O; n = WT UTR + H O and UTR + Dox; n = 8

10 empty + H O and empty + Dox). Vehicle or doxycycline (Dox., 5 mg/kg) was administered by oral gavage and commenced from day three after tumour cell injection. Mean ± SEM. P<., *P<., *P<.5; two-way ANOVA. (H) Representative histological analysis of CD+ cells in subcutaneous tumours at the end-point from the experiment described in Figure 7C, and quantified in Figure 7G. Scale bar is 5 µm. (I) Proposed molecular model. Oncogenic RAS activity leads to hyperphosphorylation, whereas PPA activity promotes hypophosphorylation of TTP (Bourcier et al., ; Deleault et al., 8; Essafi-Benkhadir et al., 7; Hardle et al., 5; Rahman et al., 5; Sun et al., 7), constituting a rapid switch controlling TTP activity. Low TTP expression and activity in tumour cells represents a permissive context for PD-L expression and immune evasion. 9

11 Table S, relating to Figure 5 and Figure S5. TTP phosphopeptides. Identified mouse TTP phosphopeptides from MS analyses pooled from two independent biological experiments. Identifications are % FDR controlled. PEP indicates the probability that the identification is incorrect. Phosphosite assignment probabilities are indicated in parenthesis. PEP Score Position Modified sequence Phospho (STY) Probabilities Number of Phospho (STY).67 5 _STS(ph)LVEGR_ STS()LVEGR.E- 7 8 _PGPELS(ph)PSPT(ph)SPTATPTTSSR_ PGPELS(.998)PS(.6)PT(.77)S(.)PT(.68)AT(.)P T(.)T(.)S(.)S(.)R.E _PGPELSPS(ph)PTSPTATPTTSSR_ PGPELS(.6)PS(.9)PT(.)S(.9)PT(.)AT(.)P TTSSR.7E _PGPELSPSPTS(ph)PTATPTTSSR_ PGPELSPSPT(.8)S(.857)PT(.)ATPTTSSR. 7 5 _TYS(ph)ESGRCR_ TYS()ESGRCR 6.9E-8 78 _QSIS(ph)FSGLPSGR_ QSIS()FSGLPSGR RS(.7)S(.767)PPPPGFS(.95)GPS(.6)LS(.)S(.).5E _RSS(ph)PPPPGFS(ph)GPSLSSCSFSPSSSPPPP CS(.)FS(.)PS(.9)S(.85)S(.8)PPPPGDLPLSPSAFSA GDLPLSPSAFSAAPGTPVTR_ APGTPVTR.5E _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_.5E _RSS(ph)PPPPGFS(ph)GPS(ph)LSSCSFSPSSSP PPPGDLPLSPSAFSAAPGTPVTR_ 9.68E-5 69 _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E-5 69 _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E-5 69 _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ 9.68E-5 69 _RSSPPPPGFS(ph)GPS(ph)LSSCSFSPSSSPPPP GDLPLSPSAFSAAPGTPVTR_ RS(.5)S(.5)PPPPGFS(.67)GPS(.7)LS(.69)S(.67) CS(.66)FS(.65)PS(.65)S(.6)S(.6)PPPPGDLPLSPSA FSAAPGTPVTR RS(.7)S(.68)PPPPGFS(.79)GPS(.)LS(.)S(.97)C S(.9)FS(.8)PS(.7)S(.7)S(.65)PPPPGDLPLSPSAFSA APGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR RS(.)S(.77)PPPPGFS(.)GPS(.)LS(.)S(. )CS(.)FS(.)PS(.)S(.)S(.)PPPPGDLPLSPSA FSAAPGTPVTR _S(ph)TTPSTIWGPLGGLAR_ S(.)T(.)T(.)PSTIWGPLGGLAR _S(ph)TTPSTIWGPLGGLAR_ S(.)T(.)T(.)PSTIWGPLGGLAR.9E _STT(ph)PSTIWGPLGGLAR_ ST(.)T(.997)PSTIWGPLGGLAR 6.7E _S(ph)PSAHSLGSDPDDYASSGSSLGGSDSPVFE 6.7E _S(ph)PSAHSLGSDPDDYASSGSSLGGSDSPVFE 5.7E _SPSAHS(ph)LGSDPDDYASSGSSLGGSDSPVFE.5E _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 79 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 8 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 8 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 8 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 87 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE.5 89 _SPSAHSLGS(ph)DPDDYASSGSSLGGSDSPVFE S(.6)PS(.)AHS(.87)LGS(.)DPDDY(.)AS(. 5)S(.5)GS(.5)S(.5)LGGS(.5)DS(.5)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.65)LGS(.6)DPDDY(.)AS(.) S(.)GS(.)S(.)LGGS(.)DS(.)PVFEAGVFGP PQTPAPPR S(.7)PS(.7)AHS(.7)LGS(.)DPDDY(.)AS(. )S(.)GS(.)S(.)LGGS(.)DS(.)PVFEAGVF GPPQTPAPPR S(.9)PS(.9)AHS(.78)LGS(.)DPDDY(.7)AS(.7 8)S(.78)GS(.78)S(.78)LGGS(.78)DS(.78)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR S(.)PS(.)AHS(.)LGS(.9)DPDDY(.)AS(. 9)S(.9)GS(.9)S(.9)LGGS(.9)DS(.9)PVFEAGVF GPPQTPAPPR

12 Table S, relating to STAR Methods, Method Details. Cell lines and growth conditions. Cell line Normal medium Starvation medium H58 RPMI + % FCS N/A A7 RPMI + % FCS N/A H79 RPMI + % FCS N/A KPB6 DMEM + % FCS N/A Type II pneumocytes DCCM- + % FCS +.5 % FCS for qpcr and FACS + % FCS for mrna half-life analysis SW87 DMEM + % FCS N/A H RPMI + % FCS N/A 9FT DMEM + % FCS N/A TTP KO and TTP DMEM + % FCS +.5 % FCS WT MEFs MCFA F:DMEM mix (:) + 5 % horse serum and 5 % horse serum, ng/ml EGF, µg/ml insulin, ng/ml cholera toxin,.5 µg/ml hydrocortisone CT6 RPMI + % FCS N/A A59 DMEM + % FCS N/A MC8 RPMI + % FCS +.5 % FCS N/A, not applicable.

13 Figure S A B PD-L expression realtive MCFA-ER HRAS GV EtOH -OHT 6 h h % of maximum 8 6 MCFA-ER HRAS GV 5 PD-L - PE Unstained Isotype EtOH h -OHT h EtOH d -OHT d 8 6 HKE--ER HRAS GV 5 PD-L - PE Unstained Isotype EtOH -OHT C % of maximum 8 6 MCFA - parental 8 6 HKE- - parental Unstained Isotype EtOH -OHT D -OHT EtOH MEK i PIK i p-erk ERK p-akt (S7) AKT Vinculin 5 5 PD-L - PE E PD-L expression relative..5. MEK i H * PIK i MEK i + PIK i PD-L expression relative..5. MEK i H79 * PIK i MEK i + PIK i F PD-L expression relative..5. H58 ERKi PD-L expression relative..5. H ERKi G PD-L expression relative H79 PMA PMA + MEKi H PD-L (relative MFI) 5 Human Mouse A7 H58 H79 68T MEK i PIK i MEK i + PIK i I Expression relative Type II pneumocytes - ER-KRAS GV ns BM HLA-A HLA-B HLA-C TAP TAP TAPBP Et-OH -OHT

14 Figure S A pgl Firefly luciferase reporters: SV FF luciferase SV enh. SV (Control) GRCh8 Homo sapiens Chromosome 9 FF luciferase CD7 pro. FF luciferase Basic (Empty) kb pro. 587 CD7 pro FF luciferase kb pro CD7 pro. FF luciferase CD7 pro. FF luciferase CD7 enh. kb pro. kb pro. + enh. ø PMA IFN- 6 8 Firefly luciferase normalised to Renilla B KPB6 PD-L expression relative to Gapdh Time (h) + Act D (5 ug/ml) PIK i + Act D (5 ug/ml) C KPB6 KPB6 Tusc expression relative to Gapdh Time (h) Ptgs expression relative to Gapdh Time (h) + Act D (5 ug/ml) MEK i + Act D (5 ug/ml)

15 Figure S A H58 sirna knock-down efficiency D H58 - h Expression relative..5. * * siauf siksrp sihur sittp PD-L expression relative Empty + Emtpy Empty + TTP KSRP + Empty KSRP + TTP B Expression relative..5. A7 sirna knock-down efficiency * siauf siksrp sihur sittp E PD-L expression relative 6 sibrf- H sibrf- * sittp C F H sirna knock-down efficiency H sirna knock-down efficiency Expression relative..5. siauf siksrp sihur sittp Expression relative..5. sibrf- sibrf- sittp G H A7 H58 H58 kda sittp sittp sittp Empty Myc-TTP 5 TTP 8 Vinculin H H58 H Expression relative 6 5 sittp # sittp # sittp # sittp # sittp pool Expression relative sittp # sittp # sittp # sittp # sittp pool PD-L TTP

16 Figure S A PD-L expression relative H58 - PD-L * MEKi n.s MEKi KSRP expression relative H58 - KSRP * n.s MEKi MEKi siksrp siksrp B KPB6 KPB6 6 RNA IP - PD-L n.s *.9.5 MEKi 5 RNA IP - PD-L n.s.78 * 7.9 MEKi % Input.. IgG TTP IgG TTP % Input IgG KSRP IgG KSRP C RNA IP - Gapdh (5.5 h) RNA IP - Gapdh (5.5 h) % Input 6 n.s n.s.9... IgG TTP IgG TTP MEKi % Input n.s * * IgG KSRP IgG KSRP MEKi D. H58 - h * n.s PD-L expression relative egfp-empty egfp-ksrp egfp-ksrp S9A

17 Figure S5 A MS/MS spectrum for STphSLVEGR MS/MS spectrum for QSIphSFSGLPSGR B 5 MCFA-ER HRAS GV Unstained ROS - DCF % of maximum 8 6 % of maximum 8 6 Isotype EtOH -OHT -OHT + NAC 5 PD-L - PE 5 ROS - DCF 5 PD-L - PE C MCFA-ER MEKK D A7 H kda 5 8 EtOH -OHT p-p8 (T8, Y8) Vinculin PD-L expression relative..5. NAC * Glut. MK i III PD-L expression relative..5. NAC * Glut. MK i III E CT6 TTP KO H79 H58 kda 5 8 TTP WT HO Dox. S5A S78A TTP HO Dox. Myc Vinculin PD-L expression relative NAC Glut. MK i III PD-L expression relative NAC Glut. MK i III

18 Figure S6 A Loboda-Watters signature Lung Loboda-Watters signature Colon CD7 FRL IL8R IFNGR SLCA LCP CDC SOCS PSEN ILRB PTPRC TP5 TIGIT CDD IL8RAP PSEN CDE RPS6 EOMES Expression IFNGR CD7 SOCS SLCA IL8RAP FRL RPS6 TIGIT PTPRC ILRB LCP CDE IL8R PSEN CDC CDD EOMES PSEN TP5 Expression High KRAS mutation NO KRAS mutation NO RAS pathway activity YES YES Low p<.5 p-value p<.5 p-value B Expression relative Type II pneumocytes - ER-KRAS GV n.s * n.s EtOH -OHT -OHT + IFN-g block -OHT + Rux. n.s EtOH -OHT -OHT + IFN-g block -OHT + Rux. PD-L IFNGR PD-L expression relative Type II pneumocytes - ER-KRAS GV ø IFN- IFN- + Rux. IFN- + blocking ab. C TTP expression (log) Selamat - Lung P =.5 x Skrzypczak - Colon P = 9. x - D E mrna expression relative to control KrasLSL-GD/+; Trp5F/F mouse lung: CD5 - CD - DAPI - EpCAM + PD-L TTP KrasLSL/+; Trp5 F/F mouse lung: CD5 - CD - DAPI - Normal adjacent Tumour * * Lung (n = 58) Lung adenocarcinoma (n = 58) Colon (n = ) Colorectal adenocarcinoma (n = 5) PD-L (relative MFI) Normal adjacent Tumours

19 Figure S7 A CT6 TTP (teton) B MC8 TTP (teton) C MC8 TTP (teton) PD-L expression relative to Gapdh..5. Time (h) ActD + H O MEKi Dox. Dox. + MEKi kda 5 8 HO Dox. Dox. Myc Vinculin PD-L expression relative to Gapdh..5. HO Dox. D CT6 TTP (teton) E MC8 TTP (teton) Confluency (%) Time (h) Empty: TTP (teton) Empty: TTP (teton) + H O Empty: TTP (teton) + Dox.. g/ml Empty: TTP (teton) + Dox. g/ml PD-L 'UTR: TTP (teton) PD-L 'UTR: TTP (teton) + H O PD-L 'UTR: TTP (teton) + Dox.. g/ml PD-L 'UTR: TTP (teton) + Dox. g/ml % Confluency Time (h) TTP WT ø TTP WT H O TTP WT Dox.. µg/ml TTP WT Dox. µg/ml % of maximum F 8 6 CT6 Unstained Isotype TTP (teton): Empty + HO TTP (teton): Empty + Dox. TTP (teton): PD-L WT UTR + HO TTP (teton): PD-L WT UTR + Dox. TTP (teton): PD-L UTR + HO Tumour volume (mm ) G 8 6 CT6 - BALB/c * n.s * TTP (teton): Empty + HO TTP (teton): Empty + Dox. TTP (teton): PD-L WT UTR + HO TTP (teton): PD-L WT UTR + Dox. TTP (teton): PD-L UTR + HO TTP (teton): PD-L UTR + Dox. 5 PD-L - PE TTP (teton): PD-L UTR + Dox. 5 5 Time (days) H TTP (teton): Empty TTP (teton): Empty + H O + Dox. TTP (teton): PD-L 'UTR TTP (teton): PD-L 'UTR + H O + Dox. I ACTIVE MK OA ERK INACTIVE H&E TTP PPA Trametinib TTP P P P PD-L LOW PD-L HIGH CD