Primary mucosal melanomas of the head and neck are characterised by overexpression of the DNA mutating enzyme APOBEC3B

Abstract

Aims: Primary head/neck mucosal melanomas (MMs) are rare and exhibit aggressive biologic behaviour and elevated mutational loads. The molecular mechanisms responsible for high genomic instability observed in head/neck MMs remain elusive. The DNA cytosine deaminase APOBEC3B (A3B) constitutes a major endogenous source of mutation in human cancer. A3B-related mutations are identified through C-to-T/-G base substitutions in 5'-TCA/T motifs. Herein, we present immunohistochemical and genomic data supportive of a role for A3B in head/neck MMs.

Methods and results: A3B protein levels were assessed in oral (n = 13) and sinonasal (n = 13) melanomas, and oral melanocytic nevi (n = 13) by immunohistochemistry using a custom rabbit α-A3B mAb (5210-87-13). Heterogeneous, selective-to-diffuse, nuclear only, A3B immunopositivity was observed in 12 of 13 (92.3%) oral melanomas (H-score range = 9-72, median = 40) and 8 of 13 (62%) sinonasal melanomas (H-score range = 1-110, median = 24). Two cases negative for A3B showed prominent cytoplasmic staining consistent with A3G. A3B protein levels were significantly higher in oral and sinonasal MMs than intraoral melanocytic nevi (P < 0.0001 and P = 0.0022, respectively), which were A3B-negative (H-score range = 1-8, median = 4). A3B levels, however, did not differ significantly between oral and sinonasal tumours (P > 0.99). NGS performed in 10 sinonasal MMs revealed missense NRAS mutations in 50% of the studied cases and one each KIT and HRAS mutations. Publicly available whole-genome sequencing (WGS) data disclosed that the number of C-to-T mutations and APOBEC3 enrichment score were markedly elevated in head/neck MMs (n = 2).

Conclusion: The above data strongly indicate a possible role for the mutagenic enzyme A3B in head/neck melanomagenesis, but not benign melanocytic neoplasms.

Keywords: HRAS; KIT; NRAS; APOBEC3B; APOBEC3G; DNA cytosine deamination; oral mucosal melanoma; sinonasal mucosal melanoma.

Figure 1

Endogenous A3B protein expression in primary oral mucosal melanomas (MMs) and oral melanocytic nevi. (A) A3B staining and corresponding H&E‐stained photomicrographs of representative primary oral MMs. Scale bars are 60 μm and inset images are magnified 8‐fold. (B) A3B immunophenotype and corresponding H&E‐stained photomicrographs of representative benign oral melanocytic nevi. Scale bars are 60 μm and inset images are magnified 8‐fold. (C) Collective presentation of quantified A3B IHC score (H‐score) of oral MMs (n = 13) and oral nevi (n = 13, control). The black horizontal dotted line indicates the A3B H‐score median (40) of the oral MMs group. [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2

Endogenous A3B protein expression in primary sinonasal MMs. (A) A3B staining and corresponding H&E‐stained photomicrographs of representative primary sinonasal MMs. Scale bars are 60 μm and inset images are magnified 8‐fold. (B) Collective presentation of quantified A3B H‐scores of sinonasal MMs (n = 13). The black horizontal dotted line indicates the A3B H‐score median (24) of this group. (C) Collective presentation and comparison of quantified A3B H‐scores of oral melanocytic nevi (n = 13), oral MMs (n = 13) and sinonasal MMs (n = 13). The H‐score median and interquartile range for each group are shown and statistical significance for key comparisons is indicated (Kruskal–Wallis test). [Color figure can be viewed at wileyonlinelibrary.com]

Figure 3

Endogenous A3G expression in primary sinonasal MMs. (A) H&E‐stained and corresponding 5210‐87‐13 immunohistochemical photomicrographs of the two cases with pronounced cytoplasmic positivity. Scale bars are 60 μm and inset images are magnified 8‐fold. (B) Staining with a commercial A3G specific mAb corroborates the presence of cytoplasmic A3G in these two sinonasal MMs. Scale bars are 60 and 20 μm, respectively. (C) Quantitative RT‐PCR investigating mRNA levels of A3A, A3B, and A3G expression in these two cases. Experiments were performed in triplicate. The mean and SEM of at least three independent experiments is shown. [Color figure can be viewed at wileyonlinelibrary.com] [Color figure can be viewed at wileyonlinelibrary.com]

Figure 4

APOBEC3 does not drive NRAS, HRAS, or KIT genetic alterations in head/neck MMs. (A) NGS findings in primary sinonasal MMs (n = 10). (B) Type and nucleotide context of the most common pathogenic mutations discovered by NGS in primary sinonasal MMs. None of the NRAS, HRAS, or KIT base substitutions occurred in a 5′‐TCA‐3′ or 5′‐TCT‐3′ trinucleotide context, which represent the preferred DNA motifs for APOBEC3. [Color figure can be viewed at wileyonlinelibrary.com] [Color figure can be viewed at wileyonlinelibrary.com]

Figure 5

APOBEC3 enrichment scores are high in head/neck MMs. (A) Total number of mutations, (B) Number of C‐to‐T mutations, (C) APOBEC3 enrichment scores, and (D) A3B‐specific enrichment scores of eight cases of human MMs, including two nasal (red bars), available through public WGS databases. Enrichment score is a metric to represent the volume of APOBEC3‐specific single base substitutions T[C>T/G]W over background mutation at cytosine (at nonspecific contexts). A3B‐specific enrichment score exclusively considers C‐T/G mutations in an (A/G)TCW context, which is the context preferred by A3B. Red bars represent the two head/neck MMs that are significantly enriched for APOBEC3 and A3B mutations (q < 0.05). The dotted line at enrichment score = 1 is shown to visualize samples enriched (above 1) or depleted (below 1) for APOBEC3 or A3B mutations. These analyses remove C‐to‐A mutations. [Color figure can be viewed at wileyonlinelibrary.com] [Color figure can be viewed at wileyonlinelibrary.com]