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
. 2019 Sep 15;25(18):5663-5673.
doi: 10.1158/1078-0432.CCR-19-0366. Epub 2019 Jul 8.

BAP1 Loss Is Associated with DNA Methylomic Repatterning in Highly Aggressive Class 2 Uveal Melanomas

Affiliations

BAP1 Loss Is Associated with DNA Methylomic Repatterning in Highly Aggressive Class 2 Uveal Melanomas

Matthew G Field et al. Clin Cancer Res. .

Abstract

Purpose: The strong association between BAP1 mutations and metastasizing Class 2 uveal melanoma (UM) suggests that epigenetic alterations may play a significant role in tumor progression. Thus, we characterized the impact of BAP1 loss on the DNA methylome in UM.Experimental Design: Global DNA methylation was analyzed in 47 Class 1 and 45 Class 2 primary UMs and in UM cells engineered to inducibly deplete BAP1. RNA-Seq was analyzed in 80 UM samples and engineered UM cells.

Results: Hypermethylation on chromosome 3 correlated with downregulated gene expression at several loci, including 3p21, where BAP1 is located. Gene set analysis of hypermethylated and downregulated genes identified axon guidance and melanogenesis as deregulated pathways, with several of these genes located on chromosome 3. A novel hypermethylated site within the BAP1 locus was found in all Class 2 tumors, suggesting that BAP1 itself is epigenetically regulated. Highly differentially methylated probes were orthogonally validated using bisulfite sequencing, and they successfully distinguished Class 1 and Class 2 tumors in 100% of cases. In functional validation experiments, BAP1 knockdown in UM cells induced methylomic repatterning similar to UM tumors, enriched for genes involved in axon guidance, melanogenesis, and development.

Conclusions: This study, coupled with previous work, suggests that the initial event in the divergence of Class 2 UM from Class 1 UM is loss of one copy of chromosome 3, followed by mutation of BAP1 on the remaining copy of chromosome 3, leading to the methylomic repatterning profile characteristic of Class 2 UMs.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: Dr. Harbour is an inventor of intellectual property discussed in this study. He is a paid consultant for Castle Biosciences, licensee of this intellectual property, and he receives royalties from its commercialization. All remaining authors declare no competing financial interests.

Figures

Figure 1.
Figure 1.
Global methylation analysis in primary uveal melanomas (UMs). (A) Unsupervised principal component analysis (PCA) based on methylation profiling showing differential clustering of Class 2 (red) and Class 1 (blue) UMs from the 80 TCGA cases. (B) Unsupervised PCA analysis on an independent dataset consisting of 12 additional cases that shows similar clustering based on methylation profiling. Red and blue ellipsoids represent 95% confidence intervals for each respective group. (C) Global CpG-island feature analysis of the normalized percent of hypermethylated (red) and hypomethylated (blue) probe sites in Class 2 UMs compared to Class 1 TCGA samples. (D) Chromosomal location analysis of the normalized percent of hypermethylated (red) and hypomethylated (blue) probe sites within the promoter region in Class 2 UMs compared to Class 1 TCGA samples. The promoter region was defined as starting at the transcription start site (TSS) and extending upstream 1500 base pairs. CpG-islands are defined as regions > 500 base pairs with > 55% GC content and an expected/observed CpG ratio of > 0.65. CpG shores are ~2Kb from islands and CpG shelves are ~4Kb from islands. *p < 0.05, **p < 0.01, ***p < 0.001 with binomial testing.
Figure 2.
Figure 2.
Integration of DNA methylation with RNA expression data. (A) Inner quadrant graph demonstrates methylation of individual probe sites (x-axis, Class 2 Beta – Class 1 Beta) plotted against gene expression (y-axis, log2 fold change Class 2/Class 1) in 80 TCGA samples. Red, hypermethylated/downregulated genes; blue, hypomethylated/upregulated genes; black, other genes. Outer Circos plot demonstrates hypermethylated/downregulated genes (red bars) and hypomethylated/upregulated genes (blue bars) that met filtering cutoffs (Delta Beta ± 0.05 and log2 fold change ± 1) with respect to chromosomal location. (B) Global CpG-island feature analysis of the filtered hypermethylated/downregulated genes and hypomethylated/upregulated genes showing the normalized percent of hypermethylated (red) and hypomethylated (blue) probe sites. (C) Gene set enrichment analysis of chromosomal location for hypermethylated/downregulated genes (red lines) and hypomethylated/upregulated genes (blue lines) in Class 2 UMs compared to Class 1 UMs from the TCGA dataset. Copy number gains (blue circles) and losses (red circles) that commonly occur in UM on chromosomes 3, 6, and 8 are indicated next to the enriched methylated chromosomal regions. CpG-islands are defined as regions > 500 base pairs with > 55% GC content and an expected/observed CpG ratio of > 0.65. CpG shores are ~2Kb from islands and CpG shelves are ~4Kb from islands. *p < 0.05, **p < 0.01, ***p < 0.001 with binomial testing.
Figure 3.
Figure 3.
Epigenetic deregulation of axon guidance and melanogenesis pathways. Protein-protein interaction network of axon guidance (top) and melanogenesis (bottom) pathways identified using STRING, Gene Set Enrichment Analysis (GSEA), and the Molecular Signatures Database (MSigDB) for hypermethylated/downregulated (red) and hypomethylated/upregulated (blue) genes in Class 2 UMs compared to Class 1 UMs from the TCGA dataset. The source of evidence for each interaction is indicated by the legend.
Figure 4.
Figure 4.
Epigenetic silencing of key melanocyte-related developmental genes on chromosome 3 in Class 2 uveal melanoma. (Left) Melanogenesis and neural crest genes (SATB1, MITF, RAF1, DVL3) with box plots of normalized RNA read counts for all Class 1 UMs (blue) and Class 2 UMs (red) from the TCGA dataset with gene track plots of hypermethylated (red) and hypomethylated (blue) probe sites. (Middle, Top to Bottom) Ideogram of chromosome 3; Gene region with hypermethylated/downregulated loci (3p26–25, 3p23–21, 3q12–21, 3q27); Genomic location (purple bars) of hypermethylated/downregulated genes involved in axon guidance, neural crest, and melanogenesis; LOESS plot of DNA methylation (Beta Value) on chromosome 3 in Class 2 UMs (red line) and Class 1 UMs (blue line) from the TCGA dataset, including all probe sites associated with significantly hypermethylated/downregulated and hypomethylated/upregulated genes; Dot plot demonstrating significance (–logFDR) of differential methylation in Class 1 versus Class 2 UMs; Dot plot demonstrating fold change (log2 scale) in methylation in Class 1 versus Class 2 UMs. (Right) Axon guidance genes (PLXN1, CHL1, ROBO1, SEMA3B) with box plots of normalized RNA read counts for all Class 1 UMs (blue) and Class 2 UMs (red) from the TCGA dataset with gene track plots of hypermethylated (red) and hypomethylated (blue) probe sites. For box-and-whiskers plots, the central box represents the 25th to 75th percentiles, the middle of the notch represents the median, whiskers extend from the box hinges to 1.5 times the interquartile range (IQR), and outlier values that extend beyond the whiskers are indicated by dots. Notches within the central box extend 1.58 * IQR / sqrt(n), which gives a 95% confidence interval for comparing whether medians of box plots are significantly different.
Figure 5.
Figure 5.
Association between BAP1 methylation status, BAP1 gene expression, molecular prognostic classification, and BSE mutation status. (A) Chromosome 3 ideogram with the BAP1 locus indicated at chr3p21 (red bar), an expanded map of BAP1 locus (chr3:52435020–52444121), with exons indicated by boxes, and direction of transcription indicated by arrows. Significantly methylated (red bars) and non-significantly methylated (grey bars) CpG probe sites within the BAP1 locus are indicated. (B) Normalized BAP1 gene expression plotted against methylation (Beta Value) of cg16871520, the most significantly hypermethylated probe site in Class 2 UMs relative to Class 1 UMs. LOESS linear regression line (black) is plotted along with the 95% confidence interval (grey). The r- and p-values were calculated using Spearman correlation coefficient. The left graph illustrates Class 1 versus Class 2 molecular prognostic classification status. The right graph highlights the BSE (BAP1, SF3B1 and other splicing factors, and EIF1AX) mutation status.
Figure 6.
Figure 6.
Orthogonal validation of highly differentially methylated CpG sites between Class 1 and Class 2 uveal melanomas. Fourteen UM samples that were previously unexamined for methylation were analyzed using the bisulfite conversion followed by Sanger sequencing for methylation of CpG sites within the IL12RB2, SESN1, SATB1, and ENPP2 genes (see legend at lower right). Methylation calls are displayed in pie chart form. Grey, non-methylated; blue, methylated; green, ambiguous methylation. Masked Class 1 and Class 2 predictions based on methylation (blue and red boxes, respectively) corresponded to the UM gene expression profile in all cases. Absent pie pieces indicate technical failure.

References

    1. Ramaiya KJ, Harbour JW. Current management of uveal melanoma. Expert Review of Ophthalmology 2007;2(6):939–46 doi doi: 10.1586/17469899.2.6.939. - DOI
    1. Harbour JW, Chao DL. A molecular revolution in uveal melanoma: implications for patient care and targeted therapy. Ophthalmology 2014;121(6):1281–8 doi 10.1016/j.ophtha.2013.12.014. - DOI - PMC - PubMed
    1. Onken MD, Worley LA, Ehlers JP, Harbour JW. Gene expression profiling in uveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res 2004;64:7205–9. - PMC - PubMed
    1. Tschentscher F, Husing J, Holter T, Kruse E, Dresen IG, Jockel KH, et al. Tumor classification based on gene expression profiling shows that uveal melanomas with and without monosomy 3 represent two distinct entities. Cancer Res 2003;63(10):2578–84. - PubMed
    1. van Gils W, Lodder EM, Mensink HW, Kilic E, Naus NC, Bruggenwirth HT, et al. Gene expression profiling in uveal melanoma: two regions on 3p related to prognosis. Invest Ophthalmol Vis Sci 2008;49(10):4254–62. - PubMed

Publication types

MeSH terms