SOX10 ablation arrests cell cycle, induces senescence, and suppresses melanomagenesis

Abstract

The transcription factor SOX10 is essential for survival and proper differentiation of neural crest cell lineages, where it plays an important role in the generation and maintenance of melanocytes. SOX10 is also highly expressed in melanoma tumors, but a role in disease progression has not been established. Here, we report that melanoma tumor cell lines require wild-type SOX10 expression for proliferation and SOX10 haploinsufficiency reduces melanoma initiation in the metabotropic glutamate receptor 1 (Grm1(Tg)) transgenic mouse model. Stable SOX10 knockdown in human melanoma cells arrested cell growth, altered cellular morphology, and induced senescence. Melanoma cells with stable loss of SOX10 were arrested in the G1 phase of the cell cycle, with reduced expression of the melanocyte determining factor microphthalmia-associated transcription factor, elevated expression of p21WAF1 and p27KIP2, hypophosphorylated RB, and reduced levels of its binding partner E2F1. As cell-cycle dysregulation is a core event in neoplastic transformation, the role for SOX10 in maintaining cell-cycle control in melanocytes suggests a rational new direction for targeted treatment or prevention of melanoma.

Figure 1

SOX10 depletion triggers arrested growth and melanoma cell senescence A. Western blot analysis of SOX10 expression in 0380-MMU cell lysates stably transduced with a non-silencing control hairpin (NS) or SOX10-specific lentiviral RNAi hairpins (sh1 and sh2). B. Phase-contrast images of 0380-MMU stable cell lines, 1 week post-infection with NS or SOX10sh1. C. Proliferation assays for 0380-MMU, UACC 1022, and 0002-ARM cell lines transduced with NS, sh1, or sh2. Cells were counted in triplicate 1 to 14 days after plating, and proliferation assays were performed at least twice with subsequent infections. D. Representative images of 0380-MMU cells stained for SA-β-Gal activity one week post-infection with NS, SOX10sh1 and SOX10sh2. E. Significant increases in the percentage of SA-β-Gal positive cells were seen in SOX10sh lines relative to NS lines, as determined by t-test (for 0380-MMU, sh1 p<7.32×10⁻⁵, sh2 p<7.95×10⁻³; for UACC 1022, sh1 p<5.49×10⁻¹¹, sh2 p<2.17×10⁻⁸; for 0002-ARM, sh1 p<2.90×10⁻⁵, sh2 p<8.80×10⁻⁵). F. Stably transduced UACC 1022 cell lines stained with anti-SOX10, anti-HP1β antibodies and DAPI nuclei counterstain. HP1β staining indicated the presence of senescence-associated heterochromatin foci in SOX10sh1 cells. Scale bar = 10 μm.

Figure 2

Cell cycle arrest and expression changes in cell cycle regulators are observed upon stable loss of SOX10 A. Significant increases in G1 cell populations were observed in 0380-MMU, UACC 1022, and 0002-ARM lines stably transduced with SOX10-specific hairpins (sh1 and sh2) relative to lines stably transduced with a non-silencing control hairpin (NS) (t-test p values were: 0380-MMU, sh1 = p<0.007, sh2 = p<0.026; UACC 1022 sh1 = p<0.026, sh2 = 0.17; 0002-ARM sh1 = p<0.023, sh2 = p<0.03). FACS analysis on each line was repeated twice with subsequent infections. B. Western blot analysis of the effects of NS, sh1 and sh2 stable transduction in each line on important cell cycle regulatory proteins. C. Quantitative RT-PCR of RNA isolated from 0380-MMU NS and sh1 stable lines showed significant changes in SOX10 (p<0.006), MITF (p<0.0005), E2F1 (p<0.005) and p21 (p<0.016) expression as determined by t-test. Data are fold change normalized to control, and the average of three replicate assays.

Figure 3

Altered MITF, p21 and E2F1 expression precedes G1 cell cycle arrest A. Quantification of the percentages of SA-β-Gal-positive cells in 0380-MMU and UACC 1022 cells transiently transfected with non-silencing (NS) or SOX10-specific (si1 and si2) siRNA showed no significant differences. B. No significant differences were seen in G1 cell populations in 0380-MMU and UACC 1022 melanoma cell lines transiently transfected with NS, si1, or si2. FACS analysis on independent cell lines was repeated twice with subsequent transfections. C–D. Quantitative RT-PCR analysis of (C) UACC 1022 and (D) 0380-MMU cell lines transiently transfected with NS or siSOX10. Data are fold change normalized to NS control. siSOX10-transfected UACC 1022 cells showed significant changes in expression of SOX10 (p<0.05), MITF (p<0.0004), and E2F1 (p<0.001). siSOX10-transfected 0380-MMU cells showed significant changes in the expression of SOX10 (p<0.01), MITF (p<0.007), E2F1 (p<0.001); p21 and p27 showed increased expression that did not achieve statistical significance. Data analyzed by Student’s t-test.

Figure 4

SOX10 expression is required for melanoma progression in vivo A. Grm1^Tg/+;Sox10^+/+ mice show moderate ear hyperpigmentation (top left panel). This phenotype is exaggerated in Grm1^Tg/Tg;Sox10^+/+ mice, where the pigmented lesions are raised (bottom left panel). Haploinsufficiency for SOX10 (Sox10^LacZ/+) in conjunction with one or two Grm1^Tg alleles reduces pigmented ear lesions (right two panels). B,C. Grm1^Tg phenotypic severity either alone or in the context of Sox10 haploinsufficiency was quantitatively measured by scoring each individual animal on a scale of 1–4 for (B) the extent of pigmentation and (C) burden of raised tumors. Sox10 haploinsufficiency significantly reduced pigmented nevi (p<0.005) in both Grm1^Tg/+ and Grm1^Tg/Tg mice, and also the extent to which they were raised (p<0.05) in Grm1^Tg/Tg mice (Mann-Whitney test with Bonferonni correction). D. No ventral spotting was observed in Grm1^Tg/+ or Grm1^Tg/Tg mice that were also Sox10^+/+ (left two panels), and the presence of one or two Grm1^Tg/+ alleles did not affect ventral spotting caused by the Sox10^LacZ allele. (right two panels; mice shown received a spotting severity score of two). E. Quantitative scoring of belly spots in Sox10^LacZ; Grm1^+/+, Sox10^LacZ; Grm1^Tg/+, and Sox10^LacZ; Grm1^Tg/Tg mice showed no significant differences (p=0.3342; Kruskal-Wallis nonparametric one-way ANOVA).