Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Case Reports
. 2019 Apr;25(4):597-602.
doi: 10.1038/s41591-019-0373-y. Epub 2019 Mar 4.

Clinical genome sequencing uncovers potentially targetable truncations and fusions of MAP3K8 in spitzoid and other melanomas

Affiliations
Case Reports

Clinical genome sequencing uncovers potentially targetable truncations and fusions of MAP3K8 in spitzoid and other melanomas

Scott Newman et al. Nat Med. 2019 Apr.

Abstract

Spitzoid melanoma is a specific morphologic variant of melanoma that most commonly affects children and adolescents, and ranges on the spectrum of malignancy from low grade to overtly malignant. These tumors are generally driven by fusions of ALK, RET, NTRK1/3, MET, ROS1 and BRAF1,2. However, in approximately 50% of cases no genetic driver has been established2. Clinical whole-genome and transcriptome sequencing (RNA-Seq) of a spitzoid tumor from an adolescent revealed a novel gene fusion of MAP3K8, encoding a serine-threonine kinase that activates MEK3,4. The patient, who had exhausted all other therapeutic options, was treated with a MEK inhibitor and underwent a transient clinical response. We subsequently analyzed spitzoid tumors from 49 patients by RNA-Seq and found in-frame fusions or C-terminal truncations of MAP3K8 in 33% of cases. The fusion transcripts and truncated genes all contained MAP3K8 exons 1-8 but lacked the autoinhibitory final exon. Data mining of RNA-Seq from the Cancer Genome Atlas (TCGA) uncovered analogous MAP3K8 rearrangements in 1.5% of adult melanomas. Thus, MAP3K8 rearrangements-uncovered by comprehensive clinical sequencing of a single case-are the most common genetic event in spitzoid melanoma, are present in adult melanomas and could be amenable to MEK inhibition.

PubMed Disclaimer

Conflict of interest statement

COMPETING INTERESTS

The authors declare no competing interests.

Figures

Extended Data Fig. 1
Extended Data Fig. 1. Clinical Course
Timeline of the clinical course imaging, and treatment regimens are shown as colored rectangles along the horizontal timeline. For treatments, HLP is hyperthermic limb perfusion, TVEC is talimogene laherparaepvec and LTT462 is an ERK inhibitor. Subsequent figures within this manuscript showing specimens or images from this timeline are indicated.
Extended Data Fig. 2
Extended Data Fig. 2. TERT rearrangement in the initial tumor
A) Copy number segments from CONSERTING ; red shows a gain in copy number and blue, a loss. B) Scatterplot of normalized sequencing coverage from whole genome sequencing confirms a 75 kb region of neutral copy number encompassing the TERT locus flanked by deletions. C) Structural variant junctions are shown in blue for junctions within the same chromosome and red for junctions linking different chromosomes; the genome position of the partner locus is shown inline. D) RefSeq gene model showing the position of TERT. E) RNA-Seq coverage shows that TERT is expressed. We inspected RNA-Seq coverage at this locus for our 49 additional patient samples and found TERT expression to absent. F) Splice junctions detected in RNA-Seq and quantified according to the y-axis. Canonical splices are shown in blue and non-canonical splices are in red. G) An image of break-apart FISH for TERT performed on the primary tumor from the index case shows split red and green signals, consistent with TERT rearrangement. We performed FISH on X additional tumors and found this locus not to be rearranged. All FISH images are at 100X magnification.
Extended Data Fig. 3
Extended Data Fig. 3. MAP3K8 and GNG2 copy number and expression in initial and relapsed tumors
Copy number profile from CONSERTING for A) chromosomes 10 and B) chromosome 14. Grey scatterplot represents normalized whole genome sequencing read depth arranged by genome position on the x-axis. Changes in copy number are evident as segmental shifts on the y-axis. Although a deletion is evident at the start of 14q, no copy number changes are present at the MAP3K8 (chr10:30,720,950–30,752,762) or GNG2 (chr14:52,325,022–52,438,518) loci in the initial tumor sample (black arrows represent the locations of MAP3K8 and GNG2) on chromosome 10 and 14 respectively. In the relapsed sample, the copy number of both loci are increased copy number. C) RNA-Seq coverage with the initial tumor sample’s MAP3K8 and GNG2 loci in the upper panels and the relapsed tumor below. MAP3K8 and GNG2 expression are increased approximately four-fold in the relapsed sample.
Extended Data Fig. 4
Extended Data Fig. 4. MAP3K8 mutations across cBioPortal studies
ProteinPaint representation of MAP3K8 mutations found over 226 studies housed within cBioPortal . Specific amino acid changes are shown as lollypops with a size proportional to their frequency in the cohort. The most frequently mutated tumor types were Uterine Endometrioid Carcinoma (n=25), Colorectal Adenocarcinoma (n=21) and Cutaneous Melanoma (n=20). A hotspot at R442 is indicated by the black arrow. According to ELM (http://elm.eu.org/), Arg442 it is part of the sequence KRQRSLYIDL described previously as a MAPK docking site in MAP kinase substrates . Interestingly, the MAP3K8 truncation in SJMEL054992_D1 is at codon R440, thus is just R442 excluded from the truncated MAP3K8.
Extended Data Fig. 5
Extended Data Fig. 5. MAP3K8 Fused and Truncated samples from TCGA
RNA-Seq coverage overlapping the MAP3K8 locus is shown by the blue histograms. Each histogram is scaled according to the corresponding Y-axis. Black arrows indicate the positions of fusions or truncations as outlined in Supplementary Table 6. The MAP3K8 locus of all seven rearranged samples is shown.
Figure 1.
Figure 1.. Diagnosis and treatment history
A) H and E stained sections of the initial biopsy of the primary tumor and B) resection of an in-transit metastatic nodule showing morphologic features consistent with spitzoid melanoma at the primary site and progression into an overtly malignant phenotype at the metastatic site. C) Lesions on the right lower extremity throughout trametinib therapy. D) Left PET scan at time of enrollment on a clinical trial for ERK inhibitor, LTT462, showing hypermetabolic lesions in the lower extremities as well as avid osseous metastases. Right PET scan after 2 cycles showing overall decrease in the avidity of the subcutaneous nodules. Regions of interest are indicated by black arrows.
Figure 2.
Figure 2.. Clinical genomics sequencing
A) Somatic findings from the patient’s tumor. Chromosomes are plotted as ideograms around the outside of the circle plot. Moving inwards, copy number abnormalities are shown as histograms with gains in red and losses in blue relative to a normal diploid genome (dotted line) copy number alterations included gains at chromosome 1q, 8q and 20 and losses at 8p, 9p, 13q, 14q and 22; Structural variants are shown as inner grey links with notable variants highlighted: CDKN2A deletion in yellow, TERT deregulating translocation in dark blue and a complex translocation t(10;14) fusing MAP3K8 (RefSeq NM_005204) and GNG2 (NM_ 053064) in red. B) Schematic representation of the MAP3K8-GNG2 fusion protein from ProteinPaint . The MAP3K8 Serine Threonine kinase domain is shown in green, and the autoinhibitory region , (amino acids 425–467) is highlighted in orange; the GNG2 G protein gamma subunit-like motifs are in purple. The fusion point is represented by the dotted line. The fusion protein incorporates a short region of the GNG2 5’-UTR derived from the non-coding portion of GNG2 exon 2, however, the protein reading frame is preserved. C) Quantification of RNA-Seq splice junctions and sequence coverage at the MAP3K8 locus. The upper two panels show splice junctions and coverage from the MAP3K8 fused patient sample, the lower two panels show the wildtype MAP3K8 locus from an unrearranged sample (SJMEL055003_D1). The exon 8–9 region of interest is highlighted in blue and expression level is indicated on the y-axis. The fused patient sample has relatively lower MAP3K8 exon 9 expression (grey histogram) and relatively fewer exon 8–9 splices (blue links) than the unfused sample. The black arrow shows the position of the 19 MAP3K9-GNG2 spliced reads (red lollypop). D) Immunohistochemical staining for MAP3K8 and Phospho-MEK1/2 show higher expression levels in the tumor sample (upper panels) compared to a normal skin sample (lower panels); all images at 40X magnification. E) Western blot shows weak/absent MAP3K8 expression at 53 kDa in melanocytes, expression of both wildtype and fused MAP3K8 at 53 and 57 kDa in the tumor. Also shown is a truncation mutant transfected into melanocytes at 49 kDa and the MAP3K8-GNG2 fusion protein transfected into melanocytes. The black arrow indicates the position of the fusion protein at 57 kDa. The molecular weight ladder is shown in lanes 1 and 5.
Figure 3.
Figure 3.. Recurrent MAP3K8 fusions/truncations in a spitzoid melanoma cohort.
A) Tile plot showing fusions, truncations and hotspot mutations in 49 spitzoid melanomas/atypical Spitz tumors. Each patient is a single column of the grid. Fusions are shown in purple, truncations in black, hotspot mutations in green, and no mutation detected in grey. MAP3K8 fusions and truncations were the most common event in the cohort, found in 33% of cases. B) Schematic representation of MAP3K8 fusions and truncations detected by RNA-Seq. The RNA-Seq fusion point and partner genes are shown as a black/white lollypop plot. All fusions joined exon 8 of MAP3K8 to a partner locus and removed autoinhibitory exon 9. C) Representative FISH and IHC images from MAP3K8 wildtype, fused and truncated samples. Break apart FISH (column 1) confirms the translocation in two samples with the 5’ of MAP3K8 shown in green and the 3’ shown in red. All FISH images are at 100X magnification; at least 100 nuclei were analyzed across multiple fields in each tumor section; MAP3K8 rearrangements predicted by RNA-Seq were confirmed in 13/17 samples; no MAP3K8 rearrangements were observed in 11 control samples. Immunohistochemistry shows MAP3K8 and Phospho-MEK1/2 expression in the tumors. IHC staining for phospho MEK1/2 was done in 48 tumor specimens: the majority of tumor cells expressed pMEK1/2. The intensity of expression was weak in 13 tumors and moderate in 35 tumors. IHC staining for MAP3K8 was done in 45 tumor specimens. In the majority of tissue specimens (42 of 45), the antibody stained the cytoplasm of tumor cells with moderate to strong intensity compared to the adjacent squamous cells that were stained weakly or negatively. All IHC images at 40X magnification. D) Results from colony forming assays in soft agar with NIH 3T3 cells transfected, with empty vector, truncated MAP3K8 and mutant NRAS. Bar height shows the number of colonies normalized to the empty vector control for two replicates (blue and orange). Underlying colony counts for replicate one (blue) are vector=46, truncated MAP3K8=66, NRAS=126 and replicate two (orange) are 146, 226 and 424, respectively.
Figure 4.
Figure 4.. MAP3K8 fusions and truncations in TCGA melanoma samples.
A) Scatterplot showing differential expression of exon 9 relative to exon 8 in 472 TCGA “SKMEL” melanoma samples. Raw read counts are plotted on a log2 scale. The seven samples with highly differential exon 8/9 expression are plotted in red corresponding to TCGA-ER-A196–01A, TCGA-EE-A20I-06A, TCGA-FR-A2OS-01A, TCGA-EB-A4IQ-01A, TCGA-D3-A2J6–06A, TCGA-FS-A1Z4–06A and TCGA-Z2-AA3S-06A. B) IGV visualization of RNA-Seq data from TCGA-FS-A1Z4. The upper grey histogram shows relatively higher expression of MAP3K8 exon 8 compared to exon 9, similar to rearranged spitzoid samples. TCGA-FS-A1Z4 is shown here as a representative rearranged sample. All seven rearranged samples from TCGA were inspected in a similar manner and showed the same pattern of low exon 9 expression and clipped reads at the exon 8 boundary whereas a random selection of ten un-rearranged samples did not display this pattern. Aligned reads are shown below the histogram as small grey rectangles. A cluster of clipped reads indicating the presence of a MAP3K8-LZYL2 fusion is indicated with a black arrow; clipped reads on the LZYL2 side of the fusion junction corroborated this, as did BLAT alignment of the clipped reads. C) Schematic representation of MAP3K8-LZYL2 fusion found in TCGA-FS-A1Z4.

Similar articles

Cited by

References

    1. Lu C et al. The genomic landscape of childhood and adolescent melanoma. J. Invest. Dermatol. 135, 816–823 (2015). - PMC - PubMed
    1. Wiesner T et al. Kinase fusions are frequent in Spitz tumours and spitzoid melanomas. Nat Commun 5, 3116 (2014). - PMC - PubMed
    1. Salmeron A et al. Activation of MEK-1 and SEK-1 by Tpl-2 proto-oncoprotein, a novel MAP kinase kinase kinase. EMBO J. 15, 817–826 (1996). - PMC - PubMed
    1. Hagemann D, Troppmair J & Rapp UR Cot protooncoprotein activates the dual specificity kinases MEK-1 and SEK-1 and induces differentiation of PC12 cells. Oncogene 18, 1391–1400 (1999). - PubMed
    1. Alexandrov LB et al. Signatures of mutational processes in human cancer. Nature 500, 415–421 (2013). - PMC - PubMed

METHODS ONLY REFERENCES

    1. Lee S et al. TERT Promoter Mutations Are Predictive of Aggressive Clinical Behavior in Patients with Spitzoid Melanocytic Neoplasms. Sci Rep 5, 11200 (2015). - PMC - PubMed
    1. Rusch M et al. Clinical cancer genomic profiling by three-platform sequencing of whole genome, whole exome and transcriptome. Nature Communications 9, 3962 (2018). - PMC - PubMed
    1. Wu G et al. The genomic landscape of diffuse intrinsic pontine glioma and pediatric non-brainstem high-grade glioma. Nat. Genet. 46, 444–450 (2014). - PMC - PubMed
    1. Roberts KG et al. Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia. N. Engl. J. Med. 371, 1005–1015 (2014). - PMC - PubMed
    1. Anders S, Pyl PT & Huber W HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics 31, 166–169 (2015). - PMC - PubMed

Publication types

-