Whole-genome and whole-exome sequencing of bladder cancer identifies frequent alterations in genes involved in sister chromatid cohesion and segregation

Authors

Guangwu Guo

Xiaojuan Sun

Chao Chen

Song Wu

Peide Huang

Zesong Li

Michael Dean

Jun Wang

Yaoting Gui

Zhiming Cai

Doi

10.1038/ng.2798

PMID: 24121792 · DOI: 10.1038/ng.2798 · Journal: Nature Genetics (2013)

TL;DR

Guo et al. performed whole-genome, whole-exome, and transcriptome sequencing on 99 transitional cell carcinoma (TCC) bladder tumors with matched peripheral blood from a Chinese cohort recruited through the Urinogenital Cancer Genomics Consortium (UCGC). Beyond confirming recurrent mutations in known TCC drivers (TP53, FGFR3, PIK3CA, RB1, HRAS, KRAS, TSC1) and chromatin-remodeling genes, they identified frequent loss-of-function alterations in STAG2 (encoding a cohesin complex subunit) and ESPL1, plus a recurrent FGFR3-TACC3 in-frame gene fusion. Overall, 32/99 (32%) tumors harbored genetic alterations in the sister chromatid cohesion and segregation (SCCS) process, and STAG2 alterations associated with significantly worse survival.

Cohort & data

  • 99 individuals with transitional cell carcinoma (TCC) of the bladder, recruited through member institutions of the Urinogenital Cancer Genomics Consortium (UCGC) in China; tumors with malignant cell purity >85% selected.
  • Cancer type: BLCA — bladder urothelial carcinoma (TCC).
  • Dataset: blca_bgi — cBioPortal study ID for this cohort.
  • Assays: tumor/normal whole-exome-seq on Illumina HiSeq 2000 with Agilent SureSelect Human All Exon 50Mb capture; whole-genome-seq (fourfold mean haploid coverage); rna-seq on a subset of 42 DNA-sequenced tumors (with 16 matched normal bladder tissues); validation by sanger-sequencing. CNAs called using SegSeq with a reimplemented gistic algorithm. DHFR amplifications validated by fish.
  • Reads aligned to hg18 (NCBI human reference).

Key findings

  • Validation of 11,240 candidate somatic mutations by Sanger sequencing: 1,023 of 1,119 (91%) predicted substitutions and 67 of 91 (74%) indels confirmed.
  • 37 significantly mutated genes identified (P < 0.01 vs background; non-silent mutations in ≥5/99 tumors), including 7 previously known bladder cancer genes (TP53, HRAS, FGFR3, PIK3CA, RB1, KRAS, TSC1) and 13 newly identified TCC mutated genes.
  • STAG2 ranked first in mutational significance among the 13 newly identified genes (nonsynonymous mutation enrichment P = 8 × 10⁻¹¹; non-syn:syn ratio enrichment P = 0.02). 11/99 (11%) tumors carried STAG2 point mutations — 9 of 11 truncating (3 frameshift indels, 4 nonsense, 2 splice-site). An additional 5/99 tumors harbored STAG2 genomic deletions (16% combined alteration frequency).
  • Screening STAG2 exons 3–35 in 50 additional TCC tumor/normal pairs by Sanger sequencing identified 5 somatic STAG2 mutations in 4 tumors.
  • STAG2 promoter hypermethylation observed in 7/30 (23%) TCCs by bisulfite sequencing (≥10% methylation, log2 ratio ≥1, P ≤ 0.001).
  • Kaplan-Meier analysis: individuals with STAG2 somatic alterations (n = 16) had significantly worse survival than wild-type STAG2 (n = 83; log-rank P < 0.001). Significant survival difference persisted within both superficial and invasive TCC subtypes (P < 0.001).
  • Tumors with STAG2 alterations had higher aneuploidy than tumors with wild-type SCCS genes (P = 0.01).
  • 58 of 99 cases (58%) harbored somatic mutations in chromatin-remodeling genes. Histone demethylase KDM6A/UTY: 30%. Chromatin remodelers ARID1A/ARID4A: 17%. Histone lysine methyltransferases KMT2A/KMT2C/KMT2E: 16%. Histone acetyltransferases EP300/EP400: 15%. SWI/SNF SMARCA4/SMARCC1: 7%. Histone demethylases KDM5A/KDM5B: 6%.
  • Recurrent in-frame FGFR3-TACC3 gene fusion detected by RNA-seq in 2/42 (5%) tumors (B59-3 and B100): intron 17 of FGFR3 fused with intron 10 of TACC3, joining exon 17 of FGFR3 5’ to exon 11 of TACC3. Both tumors showed outlier high TACC3 expression by RPKM. Junction-spanning and mate-pair reads from WGS confirmed the fusion.
  • CNA findings (broad arms): gain of 5p, 8q, 13p, 20p, 20q; loss of 8p, 9p, 9q, 11p, 14p, 15p, 17p, 21p.
  • GISTIC-style analysis yielded 84 focal amplification regions and 80 focal deletion regions. Frequent amplifications: TRIO, MDM2, MYC, E2F3, CCND1, ERBB2, and (first reported in bladder) CCNE1, CEBPA, E2F1, MUC1. DHFR amplification at 5q in 14/99 (14%) tumors. Common focal deletion at 9p21 containing CDKN2A/CDKN2B in 50/99 (50%) tumors. Focal deletions also affected RB1 and CREBBP.
  • SCCS pathway alteration summary (32/99 = 32% of cohort): STAG2 16%, ESPL1 6%, TACC3 (fusion) 5%, NIPBL 4%, SMC1A 3%, SMC3 2%. Tumors with SCCS alterations showed significantly higher aneuploidy (P = 0.01).
  • Cell cycle pathway altered in 86/99 (86%) of tumors, with G1/S (RB1, MYC, CCND1, CCNE1, E2F1, E2F3, CDKN2A, CDKN2B) and G2/M (ATM, ATR, CREBBP, EP300, BRCA1, BRCA2, TP53, MDM2) checkpoints affected.

Genes & alterations

  • STAG2 — predominantly truncating somatic mutations (frameshift indels, nonsense, splice-site) in 11/99 tumors; additional 5/99 with genomic deletions; promoter hypermethylation in 23% (7/30); alterations associated with worse overall survival and increased aneuploidy. Newly identified as a high-frequency bladder cancer driver.
  • ESPL1 — significantly mutated in 6% of tumors; encodes separase that cleaves cohesin during chromosome segregation.
  • FGFR3 — recurrent activating point mutations (known TCC driver) plus participation in FGFR3-TACC3 in-frame fusion (2/42 = 5% by RNA-seq).
  • TACC3 — partner in FGFR3-TACC3 fusion; encodes microtubule-associated SCCS protein; outlier high expression driven by FGFR3 promoter rather than amplification.
  • NIPBL, SMC1A, SMC3 — additional cohesin loaders/subunits altered in 4%, 3%, 2% of tumors.
  • TP53, HRAS, KRAS, PIK3CA, RB1, TSC1 — confirmed recurrent driver mutations in TCC.
  • Chromatin-remodeling genes: KDM6A/UTY, ARID1A/ARID4A, KMT2A/KMT2C/KMT2E, EP300/EP400, CREBBP, NCOR1, CHD6, SMARCA4, KDM5A/KDM5B — mutations totaling 58% of cases.
  • ERCC2, TRRAP, FAT4 — additional newly identified TCC mutated genes (frequencies in Supplementary Note).
  • MDM2, MYC, CCND1, CCNE1, E2F1, E2F3, ERBB2, TRIO, CEBPA, MUC1 — focal amplifications.
  • DHFR — focal amplification at 5q in 14/99 (14%) tumors; flagged as therapeutically relevant given it is the target of antifolate anticancer agents.
  • CDKN2A/CDKN2B — homozygous focal deletion at 9p21 in 50/99 (50%).
  • BRCA1, BRCA2, ATM, ATR — inactivating mutations contributing to G2/M checkpoint pathway alteration.
  • BUB1 and other mitotic-checkpoint genes (BUB3, MAD1L1, MAD2L1, CENPE) — no genomic alterations observed in this cohort despite being reported in other cancers.

Clinical implications

  • STAG2 may serve as an independent biomarker of poor prognosis in TCC: STAG2-altered individuals had significantly worse survival than wild-type both overall (log-rank P < 0.001) and within superficial and invasive TCC subtypes.
  • The 32% prevalence of SCCS-pathway genetic lesions positions bladder cancer as the first solid tumor type with a predominant SCCS alteration burden, suggesting potential therapeutic vulnerabilities targeting cohesion/segregation machinery.
  • DHFR amplification in 14% of TCCs is noted by the authors as a therapeutic possibility given DHFR is the target of many antifolate anticancer agents.
  • The recurrent FGFR3-TACC3 fusion, together with frequent FGFR3 mutations and amplification, reinforces FGFR3 as a high-value therapeutic target in bladder cancer.

Limitations & open questions

  • Neither whole-genome (fourfold mean haploid coverage) nor whole-exome sequencing achieved high coverage; the authors note additional sequencing is needed to identify rare mutations.
  • Cohort is single-country (China) recruited through UCGC; generalizability across ancestries not assessed.
  • Mechanism by which STAG2 loss contributes to TCC pathogenesis remains undefined.
  • Discrepancy with leukemia data: STAG2-mutated primary leukemias have normal karyotypes, in contrast to the aneuploidy observed here and in other solid tumors — the authors call out the need for further investigation of STAG2’s role in aneuploidy and tumorigenesis.
  • Functional characterization of newly identified mutated genes (ERCC2, TRRAP, FAT4) in TCC is not performed here and is explicitly recommended.
  • Survival association for STAG2 is based on Kaplan-Meier with n = 16 altered vs n = 83 wild-type; multivariable models adjusting for stage, grade, and treatment are not presented in the main text.

Citations from this paper used in the wiki

  • “we performed whole-exome sequencing of tumor and matched peripheral blood samples from 99 individuals with TCC” — cohort definition.
  • “32 of the 99 tumors (32%) harbored genetic alterations in the SCCS process” — abstract finding.
  • “11 cases (11%) harbored mutations in STAG2, and 9 of the 11 mutations were truncating mutations” — STAG2 mutation spectrum.
  • “STAG2 genomic deletions were observed in 5 of the 99 tumors” — STAG2 CNAs.
  • “STAG2 promoter hypermethylation in 7 tumors (23%) relative to matched normal samples” — methylation finding (n = 30 tested).
  • “individuals with STAG2 somatic alterations had a much worse prognosis compared to individuals with wild-type STAG2” — survival claim (log-rank P < 0.001).
  • “STAG2, NIPBL, SMC1A and SMC3, four genes with important roles in sister chromatid cohesion, were altered in 16%, 4%, 3% and 2% of the tumors, respectively” — SCCS pathway frequencies.
  • “frequent mutations in ESPL1 (6% of tumors)” — ESPL1 finding.
  • “The only recurrent fusion event involved FGFR3 fusion in frame with TACC3 in cases B59-3 and B100 (2/42, 5%)” — fusion finding.
  • “An interesting finding was frequent amplification of DHFR … at 5q … in 14 tumors (14%)” — DHFR amplification.
  • “One of the most common focal deletions, detected in 50 tumors (50%), was a deleted region at 9p21 containing CDKN2A and CDKN2B” — 9p21 deletion.
  • “at least 57 of the 99 cases (58%) harbored somatic mutations in chromatin-remodeling genes” — chromatin-remodeling burden.

This page was processed by crosslinker on 2026-05-09.