Exome sequencing identifies recurrent somatic RAC1 mutations in melanoma

PMID: 22842228 · DOI: 10.1038/ng.2359 · Journal: Nature Genetics (2012)

TL;DR

This study performed whole-exome sequencing of 147 melanomas and identified RAC1 P29S as a recurrent activating UV-signature mutation present in 9.2% of sun-exposed melanomas, making it the third most frequent driver after BRAF and NRAS. The study also discovered PPP6C as a new melanoma gene (mutated in 12% of sun-exposed tumors, exclusively co-occurring with BRAF/NRAS mutations) and defined three molecular classes of melanoma based on mutation burden and copy-number alterations.

Cohort & data

• 147 melanomas (primary and metastatic) with matched germline DNA subjected to whole-exome sequencing; expanded to 364 melanomas for RAC1 P29S validation via Sanger sequencing.
• Cancer types: cutaneous melanoma (sun-exposed), acral melanoma, mucosal melanoma, uveal melanoma.
• Dataset: skcm_yale.
• Exome capture: Roche/NimbleGen SeqCap EZ; sequencing on Illumina GAIIx and HiSeq 2000; 22.4 Mb capture region covering 15,714 genes.
• Independent validation cohort of 76 Australian melanoma cell lines.
• Reference genome: GRCh37/hg19.

Key findings

• Sun-exposed melanomas had a median of 171 somatic mutations per tumor vs. 9 per sun-shielded tumor (P = 1.6 x 10^-5), with excess C>T transitions at dipyrimidine sequences characteristic of UV damage.
• RAC1 P29S (c.85C>T) was identified as a recurrent UV-signature mutation in 9.2% (20/217) of sun-exposed melanomas; absent in sun-shielded melanomas.
• RAC1 P29S showed 4.5- to 9-fold increased binding to PAK1-PBD compared to wild-type RAC1 and enhanced ERK phosphorylation, cell proliferation, and migration in melanocytes.
• RAC1 P29S was significantly more prevalent in males (12.8%) vs. females (2.4%, P = 0.01), consistent with greater UV exposure in men.
• PPP6C harbored mutations in 12.4% of sun-exposed tumors, all co-occurring with BRAF or NRAS mutations (P = 0.007); mutations clustered in the catalytic active site, suggesting loss-of-function.
• NF1 was mutated in 30% of BRAF/NRAS-wild-type sun-exposed melanomas.
• Three major melanoma classes identified: (1) sun-shielded with high copy gains and low mutation burden; (2) sun-exposed BRAF/NRAS-wild-type with high mutation load; (3) sun-exposed BRAF/NRAS-mutant with frequent PTEN/CDKN2A losses and PPP6C mutations.
• Focal copy-number gains detected at RICTOR (5p13), CCND1 (11q13), and CDK4 (12q14), predominantly in acral/mucosal melanomas.
• DCC and ARID2 identified as putative tumor suppressors with high rates of inactivating mutations.

Genes & alterations

Gene	Alteration	Finding
RAC1	P29S (gain-of-function)	Recurrent in 9.2% of sun-exposed melanomas; activating mutation with increased effector binding
BRAF	V600E and others	Most frequent driver in cutaneous melanoma (~50%)
NRAS	Q61 and others	Second most frequent driver (15-20%)
PPP6C	H92Y, R301C (loss-of-function)	12.4% of sun-exposed tumors; exclusively with BRAF/NRAS mutations
NF1	Inactivating mutations	30% of BRAF/NRAS-WT sun-exposed melanomas
TP53	Nonsense/truncating	Enriched in BRAF/NRAS-WT class
ARID2	Nonsense/truncating	Putative tumor suppressor; part of SWI/SNF complex
PTEN	Copy loss (10q23)	Frequent in BRAF/NRAS-mutant class
CDKN2A	Copy loss (9p21)	Frequent in BRAF/NRAS-mutant class
DCC	Inactivating mutations	Positively associated with RAC1 P29S
PTPRK	Missense (19.7% of sun-exposed)	Protein tyrosine phosphatase; TGF-beta target
PTPRD	Multiple (27 mutations in 17 tumors)	Protein tyrosine phosphatase
BAP1	Frameshift (homozygous)	Identified in uveal melanoma
PAK1	Downstream effector of RAC1	Potential therapeutic target

Clinical implications

• RAC1 P29S identifies a new molecular subgroup of sun-exposed melanomas (~9%) potentially amenable to targeted therapy via pharmacological inhibition of RAC1 downstream effectors (e.g., PAK family kinases).
• PPP6C inactivation co-occurring exclusively with BRAF/NRAS mutations suggests a cooperative pathway for overcoming oncogene-induced senescence, distinct from PTEN loss; Aurora A kinase inhibition may be relevant in PPP6C-mutant tumors.
• The molecular classification into three melanoma classes (sun-shielded/copy-gain-high, sun-exposed/mutation-high/BRAF-NRAS-WT, sun-exposed/BRAF-NRAS-mutant) has implications for treatment stratification.
• Vemurafenib and dabrafenib are referenced as BRAF-targeted therapies with known clinical benefit in BRAF-mutant melanoma.

Limitations & open questions

• The study is cross-sectional; longitudinal data on clinical outcomes and treatment response are not provided.
• Functional validation of RAC1 P29S was performed in cell lines and mouse melanocytes, not in patient-derived models or in vivo melanoma models.
• PPP6C loss-of-function is inferred from structural clustering of mutations and LOH but lacks direct phosphatase activity assays.
• The clinical actionability of RAC1 pathway inhibition (e.g., PAK inhibitors) remains to be tested in clinical trials.
• PREX2 (previously reported) did not reach significance in this study due to its large size and lack of expression in melanocytes, highlighting challenges in distinguishing passengers from drivers in highly mutated tumors.

Citations from this paper used in the wiki

• “RAC1P29S was present in 20 melanomas, all of which originated in the head, neck, limbs or upper trunk, comprising 9.2% of this type of lesion.” (Results, RAC1P29S mutations section)
• “PPP6C (protein phosphatase 6, catalytic subunit), mutations of which affected 12.4% of sun-exposed tumors […] all of which also had BRAF or RAS mutations (P = 0.007)” (Results)
• “The binding activity of GTP-RAC1P29S with PAK1-PBD was similar to that of GTP-RAC1F28L, and that GTP-RAC1P29S was about 4.5-fold to ninefold more active than GTP-RAC1WT” (Results, Effector binding)
• “30% of the melanomas in this class (10 out of 33 samples) harbored deleterious mutations in NF1” (Results, Melanoma classification)

This page was processed by paper-compiler on 2026-05-06.