Refers to article:
Single-cell Sequencing Reveals Variants in ARID1A, GPRC5A and MLL2 Driving Self-renewal of Human Bladder Cancer Stem Cells
Accepted 20 June 2016
January 2017 (Vol. 71, Issue 1, pages 8 - 12)
Are we to see the forest or the trees? Cancer researchers are often confronted with this age-old conundrum. Thanks to dramatic technological improvements and cost reductions, it is now feasible to sequence the whole genome/exome of an increasing number of tumor specimens. This has offered unprecedented opportunities to profile the entire genomic landscape of a tumor type without preconceived notion. Bladder cancer is no exception. To date, ∼1300 specimens have been sequenced on a whole-genome/exome level, with more specimens being accrued and sequenced . Not only have these studies validated the key mutations previously identified from candidate gene sequencing, but they have also uncovered new recurrent mutations that may serve as tumor drivers. They showed that bladder cancer is among the most mutated cancers, in line with its being a chemically induced cancer. The studies taught us that the collective alterations for a signaling pathway, such as RTK-RAS-PI3K and p14Arf-MDM2-p53-p21, are more biologically meaningful than the alteration of any component of a pathway. This has clear implications for devising targeted therapies, such as targeting one molecule versus targeting the entire pathway. The studies revealed prominent defects in histone remodeling, DNA repair, and cell division that can be exploited for enhancement of tumor-cell sensitivity to DNA-damaging agents. Last, but not the least, multiplatform analyses indicated the existence of phenotypic subtypes within muscle-invasive bladder cancer . This finding is highly significant, as the subtypes show divergent responses to neoadjuvant therapies and hence divergent prognoses.
Notwithstanding the tremendous progress in profiling bladder cancer globally, an outstanding issue remains. Every tumor tissue comprises the differentiated or “bulk” tumor cells and the relatively minor stem/progenitor cells . Without separating the two at the sampling stage, it is difficult later to determine, by mixed cell sequencing, what specific cell type the genetic alterations belong to. While relatively inconspicuous, genetic alterations in cancer stem/progenitor cells are of critical importance, because they are the key to tumor initiation, recurrence, progression, metastasis, and chemoresistance . Studies of other cancer types suggest that conventional therapies primarily target the bulk tumor cells, leaving stem/progenitor cells largely untouched and leading to “escaped clones” and treatment failure. In this issue of European Urology, Yang and colleagues  report on their use of single-cell sequencing of CD44–differentially sorted normal urothelial stem cells, normal urothelial non-stem cells, bladder cancer stem cells, and bladder cancer non-stem cells . A total of 59 individual cells from three bladder cancer patients (2 with pT1 and 1 with pT2; ∼5 cells/group/patient) were subjected to exome sequencing. Phylogenetic analyses of single nucleotide polymorphisms (SNPs)/single nucleotide variants (SNVs) suggested that two of the bladder cancers originated from normal urothelial stem cells and the third from bladder cancer non-stem cells. In addition, the authors identified 21 mutated genes in bladder cancer stem cells, of which six were novel. Finally, the authors demonstrated that mutation of ARID1A, GPRC5A, and MLL2 all together, but not singly or in various combinations, conferred stem-cell characteristics to bladder cancer non-stem cells .
The study by Yang et al is the first to define genomic alterations in bladder cancer stem cells using single-cell sequencing. It was well conceived and executed, especially with the inclusion of functional assays to establish the validity of normal urothelial and bladder cancer stem cells. The sample size (n = 3) was relatively small, with two cancers showing one “cell of origin” and the third showing another. Little overlapping of somatic exonic mutations existed between the two cancers showing the normal urothelial stem cell origin . The authors did not address how their findings could reconcile with recent data from mouse lineage tracing suggesting divergent progenitor cells for bladder cancer variants  (and see below). From a technical standpoint, despite the best efforts, allele dropout, the false discovery effect, and low sensitivity could not be avoided completely with single-cell sequencing. The authors are, however, not to be faulted for these limitations, given the state of the technology and the expensive nature of this line of study. It is to be hoped that this study will spur considerable future efforts that will allow more definitive conclusions on the issues touched on by Yang et al.
The coming years are likely to see a significant expansion of single-cell exome sequencing in tackling the problems unsuitable for mixed cell sequencing. New platforms are emerging that allow single-cell RNA sequencing or in situ sequencing that preserves the spatial relationship of individual cells . One area that can benefit particularly from these new approaches is the root cause of inter- and intratumor heterogeneity. Bladder cancer comprises three major phenotypic variants (low-grade papillary, high-grade papillary, and high-grade invasive)  and several subtypes within the muscle-invasive category (luminal, p53-like, and basal) . It is not uncommon for a single lesion to contain mixed lineages, for example, transitional as well as squamous. Within a lineage, different cell populations may exhibit varied differentiation. One possibility for such heterogeneity is that the lineage-committed normal urothelial or bladder cancer stem cells remain highly plastic, capable of differentiating into multiple lineages. Tumor heterogeneity may represent clonal evolution along the hierarchy of urothelial/bladder cancer stem cells. Another possibility is that urothelium possesses divergent progenitor cells, giving rise to phenotypic variants, a scenario supported by linage-tracing experiments in mice. Van Batavia and colleagues  used the uroplakin IIIa promoter and cytokeratin 5 promoter to indelibly mark intermediate/superficial cells and basal cells, respectively. The uroplakin-marked cells formed noninvasive papillary lesions, whereas the cytokeratin 5–marked basal cells contributed to carcinoma in situ and muscle-invasive lesions . Two other studies used the sonic hedgehog promoter  and cytokeratin 14 promoter  to mark urothelial basal cells, and showed that muscle-invasive bladder cancer originated from basal cells. It should be noted that all these lineage tracing studies used the N-butyl-N-(4-hydroxybutyl)nitrosamine-based chemical carcinogenesis model, and none addressed the subtypes of muscle-invasive bladder cancer. Phylogenetic analyses of single-cell genomics and transcriptomics, coupled with additional mouse lineage tracing, may solve the unanswered questions surrounding bladder cancer heterogeneity. The results should help improve our understanding of bladder cancer chemoresistance and how to prevent it.
Conflicts of interest
The author has nothing to disclose.
Work in the author's laboratory is supported in part by grants from the United States National Institutes of Health (P01 CA165980) and Veterans Affairs Office of Research and Development (Biomedical Laboratory Research and Development Service; 1I01BX002049) and a grant-in-aid from the Goldstein Fund for Urological Research of the New York University School of Medicine.
-  S.P. Lerner. Update on The Cancer Genome Atlas project on muscle-invasive bladder cancer. Eur Urol Focus. 2015;1:94-95
-  W. Choi, B. Czerniak, A. Ochoa, et al. Intrinsic basal and luminal subtypes of muscle-invasive bladder cancer. Nat Rev Urol. 2014;11:400-410 Crossref
-  N. Takebe, P.J. Harris, R.Q. Warren, S.P. Ivy. Targeting cancer stem cells by inhibiting Wnt, notch, and hedgehog pathways. Nat Rev Clin Oncol. 2011;8:97-106 Crossref
-  D.W. Stuckey, K. Shah. Stem cell-based therapies for cancer treatment: separating hope from hype. Nat Rev Cancer. 2014;14:683-691 Crossref
-  Z. Yang, C. Li, H. Liu, et al. Single-cell sequencing reveals variants in ARID1A, GPRC5A and MLL2 driving self-renewal of human bladder cancer stem cells. Eur Urol. 2017;71:8-12
-  J. Van Batavia, T. Yamany, A. Molotkov, et al. Bladder cancers arise from distinct urothelial sub-populations. Nat Cell Biol. 2014;16:982-991 Crossref
-  Ke R, Mignardi M, Hauling T, Nilsson M. Fourth generation of next-generation sequencing technologies: promise and consequences. Hum Mutat. In press. http://dx.doi.org/10.1002/humu.23051.
-  X.R. Wu. Urothelial tumorigenesis: a tale of divergent pathways. Nat Rev Cancer. 2005;5:713-725 Crossref
-  K. Shin, A. Lim, J.I. Odegaard, et al. Cellular origin of bladder neoplasia and tissue dynamics of its progression to invasive carcinoma. Nat Cell Biol. 2014;16:469-478 Crossref
-  G. Papafotiou, V. Paraskevopoulou, E. Vasilaki, Z. Kanaki, N. Paschalidis, A. Klinakis. KRT14 marks a subpopulation of bladder basal cells with pivotal role in regeneration and tumorigenesis. Nat Commun. 2016;7:11914 10.1038/ncomms11914
a Departments of Urology and Pathology, New York University School of Medicine, New York, NY, USA
b Veterans Affairs, New York Harbor Healthcare System, New York, NY, USA
Corresponding author. Department of Urology, New York University School of Medicine, Veterans Affairs Medical Center, 423 E23 Street, New York, NY 10010, USA. Tel. +1 212 9515429; Fax: +1 212 9515424.
© 2016 European Association of Urology, Published by Elsevier B.V.