1. Supplementary Table A1. List of siRNAs used in this study NameDistributorSequence siCtrl Allstars Negative Control siRNA Qiagen, Venlo, The Netherlands seq not provided Rat siPTPN2 #1 ON-TARGETplus rat PTPN2 SMARTpool® Thermo Scientific, Chicago, IL, USA seq #1 5’-GUAGAGAAAGAAUCGGUUA-3’ seq #2 5’-GAAACAGGAUUCAGCGUGA-3’ seq #3 5’-UGAUCACAGUCGUGUUAAA-3’ seq #4 5’-CUACAUGGCCAUAAUAGAA-3’ Rat siPTPN2 #2 Silencer® Select Pre-designed siRNA rat PTPN2 Applied Biosystems, Austin, TX, USA 5’-GCAUUCGAGAGGAUAGAAA-3’ Human siPTPN2 #1 Silencer® Select Validated siRNA human PTPN2 #1 "5’-GGAGAUUCUAGUAUACAGA-3’ Human siPTPN2 #2 Silencer® Select Validated siRNA human PTPN2 #2 "5’-GUACAGGACUUUCCUCUAA-3’ Rat siSTAT1 BLOCK-iT Stealth™ Select siRNA Invitrogen, Paisley, UK 5’-CCCUAGAAGACUUACAAGAUGAAUA-3’

2. Std real time PCR Sequence (5’- 3’) (bp) Sequence (5’- 3’) (bp) Rat GAPDH F ATGACTCTACCCACGGCAAG 975 F AGTTCAACGGCACAGTCAAG 118 R TGTGAGGGAGATGCTCAGTG R TACTCAGCACCAGCATCACC Human GAPDH F CTGAGAACGGGAAGCTTGTC 555 F CAGCCTCAAGATCATCAGCAA 106 R AGGTCAGGTCCACCACTGAC R TGTGGTCATGAGTCCTTCCA Rat  -actin F ATGGTGGGTATGGGTCAGAA 918 F CTGTGCCCATCTATGAGGGT 133 R CAGTGAGGCCAGGATAGAGC R CTCTCAGCTGTGGTGGTGAA Rat srp-14 F AGACCAGGATGGTGTTGCTC 436 F CGTGTTCATCACCCTGAAGA 109 R TAGACCCTTCCGTGAAAACG R CGTGGCCCTTAACAGACACT Mouse/rat/human PTPN2 F TGTCCGCTGTGGTAGTTCC 328 F ATCGAGCGGGAGTTCGA 110 R AATGGCAGCATGTGTTCG R TCTGGAAACTTGGCCACTC Mouse/rat/human PTPN22 F ATTTGACATGCCCTCCCT 398 F GGCATGGACCAAAGAGAAAT 102 R ATCATCCTCCAGAAGTCCAG R CCTTTTCAGCTTCAGAAATTCA Rat Insulin 1 F CATCAGCAAGCAGGTCATTG 316 F ACCTTTGTGGTCCTCACCTG 118 R TGCAGCACTGATCCACAATG R AGCTCCAGTTGTGGCACTTG Rat Insulin 2 F TGACCAGCTACAGTCGGAAA 390 F TGTGGTTCTCACTTGGTGGA 108 R GTTGCAGTAGTTCTCCAGTTGG R CTCCAGTTGTGCCACTTGTG Supplementary Table A2. List of primers used for real time PCR Laemmli Buffer 25 mM Tris-HCl pH 6.8, 10 % glycerol, 1 % SDS, 350 mM 2-mercaptoethanol, 175 mM dithiothreitol, 0.005 % Bromophenol Blue Phospho Lysis Buffer 1% NP40, 25 mM Hepes pH 7.8, 150 mM NaCl, 1 mM CaCl 2, 1 mM MgCl 2, 50 mM NaF, 1 mM Na 3 VO 4, 1 mM PMSF Supplementary Table A3. Lysis buffers used for Western blot

3. Supplementary Table A4. List of antibodies used for Western blot analysis AntibodyCompanyReferenceDilution PTPN2 R&D Systems, Abingdon, UK clone 252294 1:1000 STAT1 Santa Cruz Biotechnology, Santa Cruz, CA., U.S.A. sc-3461:1000 Insulin R  sc-7111:1000 phospho-STAT1 (Y701) Cell Signaling, Danvers, MA, USA #91711:1000 phospho-STAT3 (Y705)#91311:1000 STAT3#91321:1000 phospho-p44/42 MAPK#43771:1000 p44/42 MAPK#91021:1000 phospho-EGFR#22361:1000 EGFR#22321:1000 phospho-Insulin R (pYpY 1162/1163 ) Biosource, Camarillo, CA, USA 44-804G1:1000  -tubulin Sigma, Bornem, Belgium T90261:5000 HRP-conjugated anti-rabbit IgG Lucron Bioproducts, De Pinte, Belgium 711-036- 152 1:5000 HRP-conjugated anti-mouse IgG 715-036- 150 1:5000

4. * GAPDH  -actin srp-14 Supplementary Figure A1. Comparison between the expression of different housekeeping genes in INS-1E cells. INS-1E cells were left untreated or treated with either IL-1  (100 U/ml) or IL-1  (10 U/ml) + IFN-  (100 U/ml) for 24h. GAPDH,  -actin and srp-14 mRNA expression were assayed by real time PCR. Results are mean ± SEM of 4 independent experiments; * p<0.05 vs untreated cells by Student’s t test. Supplementary Figure A2. PTPN22 is not or poorly expressed in primary FACS-purified rat  -cells, human islets or INS-1E cells. PTPN22 mRNA expression was assayed in the same samples as in Fig. 1 and in rat spleen and lymph nodes, used as positive controls. Results are mean ± SEM of 3-5 independent experiments. N.S., non stimulated controls.

5. ** *** * ** *** a a a b a Supplementary Figure A3. PTPN2 inhibition does not influence insulin mRNA expression in INS-1E cells. INS-1E were transfected as described in Fig. 3 and left untreated or treated with either IL-1  (100 U/ml), IFN-  (100 U/ml) or IL-1  (10 U/ml) + IFN-  (100 U/ml) for 24h. Insulin 1 and insulin 2 mRNA expression were assayed by RT-PCR and normalized for the housekeeping gene GAPDH. Results are mean ± SEM of 4 independent experiments; * p<0.05, ** p<0.01 and *** p<0.001 vs untreated cells by Student’s t test. Supplementary Figure A4. PTPN2 inhibition does not exacerbate palmitate or CPA-induced cell death in INS-1E cells. INS-1E cells were left untransfected (NT), or transfected with 30 nM of either siCtrl, siPTPN2 #1 or siPTPN2 #2. After 2 days of recovery, cells were left untreated, or treated for 24h with 0.5 mM palmitate, 28 mM glucose, IL-1  (10 U/ml) + IFN-  (100 U/ml) or 25 µM CPA as indicated. The control condition for CPA (DMSO) contained a similar dilution of DMSO. Apoptosis was evaluated using HO/PI staining. Results are mean ± SEM of 4 experiments; a: p<0.001 vs untreated NT or untreated transfected with the same siRNA; b: p<0.001 vs IL-1  + INF-  -treated NT & siCtrl; ANOVA followed by Student’s t test with Bonferroni correction.

6.  -TUBULIN ERK phospho-ERK PTPN  A NT siCtrl siPTPN2 #1siPTPN2 #2 -30’ 2h 4h14h 24h IL + IFN NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2  -TUBULIN IR  phospho-IR  PTPN2 C NT siCtrl siPTPN2 #1siPTPN2 #2 -30’ 2h 4h14h 24h IL + IFN NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2  -TUBULIN EGFR phospho-EGFR -1’ 5’ 15’30’ 1h EGF PTPN2 B NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 NT siCtrl siPTPN2 #1siPTPN2 #2 Supplementary Figure A5. siRNA-mediated PTPN2 knockdown does not influence ERK or EGFR phophorylation, but increases IR  phosphorylation. INS-1E cells were left untransfected (NT), or were transfected with 30 nM of either siCtrl, siPTPN2 #1 or siPTPN2 #2. After 2 days of recovery, cells were left untreated, or treated for different time points with IL-1  (10 U/ml) + IFN-  (100 U/ml) or rrEGF (100 ng/ml) as indicated. (A) phospho-ERK and total ERK; (B) phosphor-EGFR and total EGFR; (C) phospho-IR  and total IR  were evaluated by Western blot. PTPN2 and  -tubulin proteins were probed to ascertain accurate inhibition and equal loading respectively. Each result is representative of 4 independent experiments.

7. a,c,d b a g a a,e,f a a,g,h a,e Supplementary Figure A6. Double knockdown of PTPN2 (by two different siRNAs) and STAT1 protects INS-1E cells from cytokine-induced apoptosis. INS-1E cells were left untransfected (NT), or were transfected with 60 nM of a control siRNA (siCtrl) or with 30 nM of either a pool of siRNAs targeting PTPN2 (siPTPN2), or another siRNA targeting PTPN2 (siPTPN2 #2), or a siRNA targeting STAT1 (siSTAT1), or double transfected with 30 nM of both siPTPN2 and siSTAT1, or siPTPN2 #2 and siSTAT1. After 2 days of recovery, cells were left untreated, or treated for 24h with IFN-  (100 U/ml), IL-1  (10 U/ml) + IFN-  (100 U/ml) or TNF-  (1000 U/ml) + IFN-  (100 U/ml). Apoptosis was then evaluated using HO/PI staining. Results are mean ± SEM of 4 independent experiments; a: p<0.001 and b: p<0.01 vs untreated NT or untreated transfected with the same siRNA; c: p<0.001 vs IFN-  -treated NT & siCtrl; d: p<0.001 vs IFN-  - treated siSTAT1, siPTPN2 + siSTAT1 & siPTPN2 #2 + siSTAT1; e: p<0.001 vs IL-1  + INF-  -treated NT & siCtrl; f: p<0.001 vs IL-1  + INF-  -treated siSTAT1, siPTPN2 + siSTAT1 & siPTPN2 #2 + siSTAT1; g: p<0.001 vs TNF-  + INF-  -treated NT & siCtrl; h: p<0.001 vs TNF-  + INF-  -treated siSTAT1, siPTPN2 + siSTAT1 & siPTPN2 #2 + siSTAT1; ANOVA followed by Student’s t test with Bonferroni correction.