Bonet C, Luciani F, Ottavi JF, Leclerc J, Jouenne FM, Boncompagni M, Bille K, Hofman V, Bossis G, Marco de Donatis G, Strub T, Cheli Y, Ohanna M, Luciano F, Marchetti S, Rocchi S, Birling MC, Avril MF, Poulalhon N, Luc T, Bertolotto C
J. Natl. Cancer Inst. 2017 08;109(8)
Background: MITF encodes an oncogenic lineage-specific transcription factor in which a germline mutation ( MITFE318K ) was identified in human patients predisposed to both nevus formation and, among other tumor types, melanoma. The molecular mechanisms underlying the oncogenic activity of MITF E318K remained uncharacterized.
Methods: Here, we compared the SUMOylation status of endogenous MITF by proximity ligation assay in melanocytes isolated from wild-type (n = 3) or E318K (n = 4) MITF donors. We also used a newly generated Mitf E318K knock-in (KI) mouse model to assess the role of Mitf E318K (n = 7 to 13 mice per group) in tumor development in vivo and performed transcriptomic analysis of the tumors to identify the molecular mechanisms. Finally, using immortalized or normal melanocytes (wild-type or E318K MITF, n = 2 per group), we assessed the role of MITF E318K on the induction of senescence mediated by BRAF V600E . All statistical tests were two-sided.
Results: We demonstrated a decrease in endogenous MITF SUMOylation in melanocytes from MITF E318K patients (mean of cells with hypoSUMOylated MITF, MITF E318K vs MITF WT , 94% vs 44%, difference = 50%, 95% CI = 21.8% to 67.2%, P = .004). The Mitf E318K mice were slightly hypopigmented (mean melanin content Mitf WT vs Mitf E318K/+ , 0.54 arbitrary units [AU] vs 0.36 AU, difference = -0.18, 95% CI = -0.36 to -0.007, P = .04). We provided genetic evidence that Mitf E318K enhances BRaf V600E -induced nevus formation in vivo (mean nevus number for Mitf E318K , BRaf V600E vs Mitf WT , BRaf V600E , 68 vs 44, difference = 24, 95% CI = 9.1 to 38.9, P = .006). Importantly, although Mitf E318K was not sufficient to cooperate with BRaf V600E alone in promoting metastatic melanoma, it accelerated tumor formation on a BRaf V600E , Pten-deficient background (median survival, Mitf E318K/+ = 42 days, 95% CI = 31 to 46 vs Mitf WT = 51 days, 95% CI = 50 to 55, P < .001). Transcriptome analysis suggested a decrease in senescence in tumors from Mitf E318K mice. We confirmed this hypothesis by in vitro experiments, demonstrating that Mitf E318K impaired the ability of human melanocytes to undergo BRAF V600E -induced senescence.
Conclusions: We characterized the functions of melanoma-associated MITF E318K mutations. Our results demonstrate that MITF E318K reduces the program of senescence to potentially favor melanoma progression in vivo.