Characterization and identification of extrachromosomal circular DNA in cholangiocarcinoma

Abstract

Extrachromosomal circular DNAs (eccDNAs) have gained attention as key players in cancer heterogeneity, potentially associated with elevated oncogene copy numbers in many cancers. While the presence of eccDNA in both normal and cancer cells is confirmed, its influence on gene-level alterations in cancer cells remains largely unexplored. This study delves into the genomic profiles of eccDNA in cholangiocarcinoma (CCA), an aggressive biliary tract cancer with extensive heterogeneity and diverse molecular alterations, using a modified long-read CircleSeq method. We reveal distinct eccDNA characteristics in CCA compared to non-tumor cells, focusing on genic components and chromosomal origins. Analysing read depth differences in oncogene-containing eccDNA; we identified potential eccDNA candidates that may be relevant for CCA biology. Subsequent bioinformatics analysis was performed using the established CReSIL tool, revealing distinct patterns of these oncogenes, particularly genes in the RAS/BRAF pathway, suggesting a potential functional role. These findings highlight the remarkable heterogeneity and diverse origins of eccDNA in CCA. This study establishes the first profiling of eccDNA in cholangiocarcinoma and paves the way for further investigation of its potential contribution to oncogene amplification and disease progression.

Fig 1. Genome mapping comparison.

(A) Normalized eccDNA count per million of all samples. (B and C) Chromosomal distribution of eccDNAs origins in (B) Cas9-treated samples and (C) MssI-treated sample. (D) Principal component analysis of genomic coverage of all samples based on their mapped reads to the human reference genome. (E) Read depth distribution of all samples across the reference genome. (F) Region of lncRNA gene XR_934236.3 on chromosome 17. (G) Region of proto-oncogene PIM1 on chromosome 6.

Fig 2. eccDNA characteristics in CCA.

(A) The occurrence of eccDNAs under 3 kb in size was higher in tumor samples than in non-tumor samples. (B) Density plot of length distribution of eccDNAs across the samples, variability and peaks of eccDNA lengths were shown in the figure. (C) Genomic composition of identified eccDNAs in MssI-approach (D) Genomic composition analysis in Cas-9 approach. (E) The number of eccDNAs containing GTTTAAAC sequence, the recognition site for MssI enzyme.

Fig 3. Recurring motifs in CCA-derived eccDNAs.

Matches of the common eccDNA motif of KKU213A cell line found in HOCOMOCO v12 CORE database. P-value refers to the TomTom estimate value using a null model consisting of sampling motif columns from all the columns in the set of target motifs. (R = Reverse complement, N = Normal).

Fig 4. Pathway network analysis based on genes annotated in eccDNAs.

Annotated genes in eccDNAs were used for pathway enrichment analysis. Each enriched pathway was represented as a node. Connected nodes indicated that they shared at least one gene from enrichment analysis. (Yellow = contained only CCA-derived genes, Blue = contained only MMNK1-derived genes, Green = contained both CCA and MMNK1 derived genes).

Fig 5. Analysis of CCA-associated genes and pathways in CCA and non-tumor cell lines.

(A) Read depth of oncogene; the read depth values shown are summation of all 3 replicates for CCA and non-tumor cholangiocyte cell lines. Multiple genes, particularly those in the Ras/BRAF pathway showed elevated read depth in CCA cell line. (B) Illustration of RAS/BRAF Pathway: This schematic depicts the RAS-BRAF- MAP2K1-MAPK1 signaling pathway, highlighting key components and their interactions. (C-K) Coverage of eccDNA carrying genes in RAF/BRAF pathway between CCA and non-tumor cholangiocyte cell lines: (C), (D) show high-coverage of BRAF-eccDNA in two samples of KKU213A (CCA cell line). (E), (F) Low-Coverage of BRAF-eccDNA in two samples of MMNK1 (non-tumor cholangiocyte cell line). (G) High coverage of eccDNA carrying MAP2K1 (MEK1) gene in the KKU213A cell line. (H) Low coverage of eccDNA carrying MAP2K1 (MEK2) gene in the MMNK1 cell line. Additional genes found in eccDNAs linked to the RAF/BRAF pathway (MAP2K1 (MEK2), MAPK1 (ERK2) and KRAS) and these eccDNA-derived genes were only found in CCA samples (I) MAP2K1 (MEK2), (J) MAPK1 (ERK2) (K) KRAS. (C: Cas9-treated. M: MssI-treated. R: Replicate number. Blue region: BRAF gene. Black regions: repetitive elements. Green region: CpG-rich area).