Multi-omics characterization of the loss of transcriptomic modulators SOX10 and BAP1 in cutaneous and uveal melanomas

Timothy J. Purwin

doi:10.17918/00001773

Identifying alterations to the transcriptomic networks in cancers is crucial for studying their mechanisms and exploiting their weaknesses. In cutaneous melanoma (CM), the deadliest form of skin cancer, and uveal melanoma (UM), the most common type of intraocular cancer, loss of specific transcriptomic modulators significantly impacts the regulation of these networks and is associated with disease progression. In CM, recent research into transcriptomics at the cellular level has revealed that cell state plasticity plays a major role in therapeutic resistance. Loss of SOX10 transcription factor activity has been linked to drug tolerance and an invasive phenotype. Its involvement with immunotherapy resistance is currently unknown. Identifying gene targets and determining its activity in transcriptomic data are important for studying its role in disease progression. In UM, the median overall survival time for a patient with metastatic disease is between six months and a year. Loss-of-function mutations in BAP1, an epigenetic regulator that deubiquitinates monoubiquitinated Histone H2A, occur in 45% of primary-site tumors. Despite this, 80% of metastatic tumors present genomic loss of BAP1. The goals of this thesis are to characterize a transcriptome profile due to SOX10 loss in CM and to identify transcriptomic targets linked to disease progression due to BAP1 loss in UM. By integrating multiple datasets, layers of 'omics data, and gene-dependent results with patient-derived data, we identified genes core to the loss of SOX10 in CM and BAP1 in UM. We developed a robust SOX10 loss regulatory gene signature by integrating SOX10 ChIP-seq with time-course knockdown ATAC-seq and CRISPR knockout RNA-seq data. We validate the gene signature using multiple SOX10 knockdown, cell line panel, and drug-resistant gene expression datasets. We determine if SOX10 activity is lower in immune checkpoint inhibitor-resistant tumors using patient-derived single-cell RNA-seq data. We use similar methods to reveal BAP1 dependent cell migration and invasion pathway genes in UM. Comparisons between BAP1 mutant and wild-type patient tumors, cell line panels, re-expression, and knockdown datasets reveal consistent changes in gene expression. Patient-derived single-cell RNA-seq data show a selective expression of these genes in malignant cells, and reverse-phase protein array data confirm these differences at the protein level. Genes and mechanisms central to the loss of SOX10 and BAP1 activities identified in this study will enable further investigation into disease progression and discovery of targets for therapeutic intervention in CM and UM.

Multi-omics characterization of the loss of transcriptomic modulators SOX10 and BAP1 in cutaneous and uveal melanomas

Files and links (1)

Abstract

Metrics

Details

Multi-omics characterization of the loss of transcriptomic modulators SOX10 and BAP1 in cutaneous and uveal melanomas

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media