Despite the importance of intratumor phenotypic heterogeneit
Despite the importance of intratumor phenotypic heterogeneity for tumor progression and therapy resistance (Marusyk et al., 2012, Marusyk and Polyak, 2010), our understanding of regulators of this process and our ability to modulate them are very limited. Recent advances in genomic sequencing and single-cell technologies have enabled the detailed characterization of tumors at the single-cell level (Macaulay et al., 2017). Although most of the single-cell studies thus far have focused on defining individual cell types (Tirosh et al., 2016), scRNA-seq has also been used to characterize cell-to-cell variability in immune c-kit inhibitor in aging (Martinez-Jimenez et al., 2017). Epigenetic regulators such as histone modifying enzymes are critical for the establishment of cell-type-specific gene expression patterns, and, thus, they are also likely to play a role in modulating cell-to-cell variability in transcription, but this has been mostly investigated in lower-level organisms during aging (Booth and Brunet, 2016). We have previously shown that neoplastic and stem cell-like mammary epithelial cells have higher transcriptomic diversity than normal and more differentiated cells based on the analysis of bulk gene expression data (Wu et al., 2010). Here we describe that KDM5 histone demethylase is a regulator of cellular transcriptomic heterogeneity in ER+ luminal breast cancer, and its higher expression in ER+ breast tumors is associated with higher transcriptomic, but not genetic, heterogeneity and shorter overall survival. Higher cell-to-cell variability increases the probability of therapeutic resistance (Chisholm et al., 2016). Most studies analyzing intratumor heterogeneity have focused on genetic alterations and in many cases therapeutic resistance is due to mutations in genes and pathways targeted by the treatment (McGranahan and Swanton, 2017). However, non-genetic variability such as epigenetic heterogeneity also contributes to therapeutic resistance by multiple different mechanisms (Brock et al., 2009). One possibility is that the distinct epigenetic state of the cells could determine cellular response to treatment (Shibue and Weinberg, 2017). Another option is that subpopulations of phenotypically different cells (e.g., persisters) provide a temporary pool for selection during treatment and facilitate the outgrowth of drug-resistant mutants as demonstrated by the emergence of EGFR(T790M)-positive clones from drug-tolerant subpopulations of lung cancer cells (Hata et al., 2016). Because KDM5 activity regulates both differentiated luminal epithelial epigenetic states and cellular transcriptomic diversity, KDM5i could decrease the probability of therapeutic resistance in different ways in multiple different cancer types including ER+ luminal breast cancers.
Acknowledgments We thank members of our laboratories for their critical reading of this manuscript and useful discussions. We thank members of Allon Klein's laboratory and the Single Cell Core at Harvard Medical School, particularly Allon Klein, Rapolas Zilionis, Sarah Boswell, and Alex Ratner, for providing instructions and guidance for setting up our single-cell RNA sequencing system. We thank Bob Yauch (Genentech, San Francisco) for providing us the KDM5 inhibitor 48 and the Lurie Family Imaging Center for performing the in vivo xenograft experiments. This research was supported by the National Cancer Institute PSOC U54 CA193461 (to F.M. and K.P.), R35 CA197623 (to K.P.), P01 CA080111 (to K.P., M.B., and P.S.), R01 CA202634 (to P.S.), the Ludwig Center at Harvard (to K.P., F.M., and M.B.), and the Division of Preclinical Innovation of the National Center for Advancing Translational Sciences (NCATS), NIH (to S.C.K., A.S., D.J.M., G.R., A.J., and M.L.-N.).
Introduction The histone tails are extensively posttranslationally modified, including by phosphorylation, ubiquitination, acetylation, and methylation, and these modifications collectively serve as regulators for chromatin template-based events including gene transcription and changes in chromatin architecture (Jenuwein and Allis, 2001, Strahl and Allis, 2000). Histone methylation was considered an irreversible PTM since its discovery in the 1960s, but this dogma was overturned by the discovery of the first histone demethylase, LSD1 (Lysine-Specific Demethylase 1; also known as KDM1A), in 2004 (Shi et al., 2004). Since then, the repertoire of histone demethylases has been greatly expanded, and today, it consists of 33 histone demethylases (Allis et al., 2007, Pedersen and Helin, 2010).