NEW YORK (GenomeWeb) – A form of melanoma marked by amplification of an oncogene called MITF appears to also involve gains of a long, non-coding RNA called SAMMSON that is widely expressed in melanoma, according to a study published online today in Nature.
By focusing on melanoma samples with focal amplifications of a chromosome 3 region containing MITF, researchers from Belgium and France found that SAMMSON is typically gained alongside MITF. Though missing from unaffected melanocyte skin cells, available transcriptome data suggested the same lncRNA is expressed in more than 90 percent of human melanoma tumors, suggesting it may have a widespread role in the disease.
"[T]he long non-coding RNA gene SAMMSON is specifically expressed in human melanomas and duplicated or amplified in about 10 [percent] of the cases," co-senior author Pieter Mestdagh, a medical genetics and cancer researcher at Ghent University, said in a statement. "This unique expression profile of SAMMSON led us to hypothesize that this gene might play an important role in the etiology of melanoma."
Using available sequence data from the Cancer Genome Atlas for tumor and normal samples from more than 300 individuals with melanoma, the team used in silico SNP analyses to search for recurrent alterations in tumors with chromosome 3 amplifications.
Along with the MITF, which has been described previously, the researchers picked up a corresponding gain involving a lncRNA roughly 30,000 bases downstream of that melanoma-specific oncogene.
But SAMMSON expression was not limited to the 10 percent or so of tumors with the characteristic chromosome 3 amplification. When the team scoured the TCGA's RNA sequence data, it picked up SAMMSON expression in more than 90 percent of primary cutaneous melanoma or metastatic tumor samples, albeit at lower levels than those found in tumors with the MITF-SAMMSON co-amplification.
Likewise, the researchers unearthed SAMMSON expression in all but one of the 17 melanoma cell lines they examined, regardless of the broader transcriptional patterns present. Their RT-qPCR-based look at dozens of cancer cell lines indicated that this relationship was specific, since SAMMSON did not turn up in non-melanoma samples.
TCGA data for thousands more tumor samples from two-dozen cancer types supported this melanoma-specificity, though low levels of the lncRNA turned up in precursors of normal human melanocyte cells. Similarly, chromatin immunoprecipitation sequence data pointed to chromatin binding in the SAMMSON promoter region of melanoma cells that was missing in other cancer cell lines.
The team's follow-up experiments hinted that SAMMSON expression is mediated by the transcription factor SOX10, while knock down of the lncRNA led to diminished melanoma growth — an effect that the group ultimately attributed to decreased interactions with a mitochondrial metabolism regulator called p32.
Along with new clues to melanoma biology, the researchers said their findings could open up potential treatment avenues, since results in mouse models of melanoma and in vitro studies of patient-derived xenograft models hinted that a SAMMSON inhibitor called GapmeR3 could suppress melanoma growth.
"SAMMSON addiction is a clear vulnerability that we can combat through targeted therapy, without affecting the normal cells from the host or patient," co-senior author Jean-Christophe Marine, a human genetics researcher affiliated with KU Leuven and VIB, said in a statement.