A landscape of driver mutations in melanoma

Abstract

Despite recent insights into melanoma genetics, systematic surveys for driver mutations are challenged by an abundance of passenger mutations caused by carcinogenic UV light exposure. We developed a permutation-based framework to address this challenge, employing mutation data from intronic sequences to control for passenger mutational load on a per gene basis. Analysis of large-scale melanoma exome data by this approach discovered six novel melanoma genes (PPP6C, RAC1, SNX31, TACC1, STK19, and ARID2), three of which-RAC1, PPP6C, and STK19-harbored recurrent and potentially targetable mutations. Integration with chromosomal copy number data contextualized the landscape of driver mutations, providing oncogenic insights in BRAF- and NRAS-driven melanoma as well as those without known NRAS/BRAF mutations. The landscape also clarified a mutational basis for RB and p53 pathway deregulation in this malignancy. Finally, the spectrum of driver mutations provided unequivocal genomic evidence for a direct mutagenic role of UV light in melanoma pathogenesis.

Figure 1. Detection of Positive Selection for Non-Silent Mutations

(A) Gene A locus displaying a greater rate of nonsilent mutation compared to silent and intronic mutation rate (left) indicative of positive selection for nonsilent mutations, and Gene B locus displaying approximately equivalent rates of nonsilent mutation and silent/intronic mutation (right) indicative of a nonsilent mutation rate that matches the basal locus mutation rate. (B) Schema of permutation-based framework for identifying genes harboring positively-selected nonsilent mutations. (C) Q-Q plot of functional mutation burden test (λ=1.02) and synonymous mutation burden test across all genes with at least one mutation in the set of 121 sequenced samples. Dashed line indicates q≤0.2 for the functional mutation burden test. Grey-shaded area represents 95% confidence interval for expected p-values. (Please see Figure S1).

Figure 2. Significantly Mutated Genes PPP6C, STK19, SNX31, TACC1

(A–D) Schematic diagram of domains and mutations of PPP6C, STK19, SNX31 and TACC1. (A, bottom panel) Structure of the PPP6C homologous protein, PP2A (PDB: 2IAE), with mapped PPP6C somatic mutations (all mutated residues are conserved between the two proteins except for PPP6C S270, which maps to PP2AC A274). Salt bridge interactions represented by dashed lines in zoom image. PPP motifs: protein phosphatase; Ub-like: ubiquitin-related fold; PTB-like: PTB: phosphotyrosine-binding; PX: Phox homology; FERM-like domain: Band 4.1 (F), Ezrin (E), Radixin (R), and Moesin (M); SPAZ: Ser-Pro Azu-1 motif; TACC: transforming acidic coiled-coil. (Please see Figure S2).

Figure 3. RAC1 hot spot Mutation in Switch I Implicates Rho Family of GTPases in Melanoma

(A) Schematic diagram (left) and image of RAC1 crystallographic model (right) (PDB: 1MH1) with P29 shown. (B) Distribution of P29-homologous or known-activating mutations in Rho family members, RAC1, RAC2, RHOT1, and CDC42.

Figure 4. RAC1 P29S Is Activating

(A) Homology model (based on PDB entry 2G0N) zoom images onto P29 (top panels) and P29S (bottom panels) in the GDP-bound form are shown. (B) Homology model (based on PDB entry 2QME) zoom images onto P29 (top panels) and P29S (bottom panels) in the GTP-bound form are shown. Relevant amino acids are highlighted in both sphere (left panels) and cartoon representation (right panels). (C) GFP-tagged RAC1 GTP-bound status assayed by p21-binding domain (PBD) of p21-activated protein kinase 1 (PAK1) pull downs in HEK293FT cells (T17N: dominant negative; Q61L: constitutively active); (D) In presence of exogenous GDP or GTPγS (NT: No Treatment; GTPγS: non-hydrolysable GTP analog); (E) Following transfection of immortalized melanocytes (pMEL) stably expressing mutant forms of NRAS or (F) BRAF. (Please see Figure S4).

Figure 5. Loss-of-Function Mutations in ARID2

(A) Schematic diagram of domains and mutations in ARID2. (B) Loss-of-function (nonsense, frame-shift indel, splice site) mutations in ARID2, ARID1B, ARID1A, SMARCA4 across sequenced samples.

Figure 6. Landscape of Driver Mutations in Melanoma

(A) (Top Panel) Per-sample mutation rate. (Middle Panel) Color-coded matrix of individual mutations and copy number alterations. In cases where multiple mutations per gene were found in a sample, only one mutation is shown, with preference given to LoF (nonsense/splice/frameshift) mutations and then hot spot/COSMIC-recurrent mutations. Final row indicates primary origin of melanoma. (Bottom Panel) Mutation spectra of all samples. (B) Distribution of selected mutations and copy number amplifications in BRAF, NRAS, NF1, HRAS, RAF1, MAP2K1, KIT, GNA11, CCND1 and CDK4 are shown across all samples. (Please see Figure S6).

Figure 7. Signature of UV Mutagenesis Across Driver Mutations

(A) Total number and (B) % of driver mutations caused by UVB single nucleotide variant (SNV) (C>T), UVA SNV (G>T), UVB in half of dinucleotide variant (DNV) (NC>NT; CN>TN) and UVA in half of DNV (NG>NT; GN>TN) are indicated. Dotted line indicates exome-wide sample median % UVB SNV (C>T).