During this reporting period the Laboratory of Genetics and Physiology has made progress and elucidated molecular mechanisms that enable cytokines to control the mammary genome during pregnancy through transcription factors. Notably, we identified, for the first time, mammary super-enhancers capable of activating the mammary genome. These super enhancers are activated by cytokine-sensing transcription factors. We have also discovered that epigenetic regulators establish a genomic clock ensuring correct temporal activation of the mammary genome. Progressive establishment of the mammary genome and launching cytokine-sensing super-enhancers Differentiation of mammary epithelium during pregnancy is characterized by the sequential activation of genes that control the production of milk. We have now identified the chronological order by which transcription factors are recruited to the mammary genome during pregnancy and the establishment of epigenetic histone codes (ref. 5, 7, 9, 13). Binding of the key mammary transcription factor STAT5 to genes induced during pregnancy was low in immature mammary tissue but increased with epithelial differentiation. Augmented STAT5 binding was associated with the establishment of H3K4me3 marks and transcriptional activation. Through integration of RNA-seq information with STAT5 and H3K4me4 ChIP-seq data sets, we discovered novel mammary-specific transcriptional units, including noncoding RNAs. Autoregulation of the Stat5 locus controls its exceptional activity STAT5 is likely the key transcription factor conveying the exceptional magnitude of gene activation in the mammary gland. It remains to be understood why its premier, non-redundant functions are restricted to prolactin-induced activation of the mammary genome. We have discovered that the Stat5a/b locus is subject to lineage-specific transcriptional control in mammary epithelium. Genome-wide surveys of epigenetic status and transcription factor occupancy uncovered a putative mammary-specific enhancer within the intergenic sequences separating the two Stat5 genes (ref. 7). Mammary-specific STAT5 binding was observed at two sites and CRISPR/Cas9 gene editing was employed to delete these sites in mice. Mutant animals exhibited a reduction of Stat5 levels in mammary epithelium and STAT5-dependent gene expression was greatly reduced. A class of mammary-restricted genes particularly dependent on high STAT5 levels was identified. Taken all together, the mammary-specific enhancer facilitates a positive feedback circuit that yields a remarkable abundance of STAT5 in the mammary gland and aids the activation of genetic programs underlying lactation and milk production. Mammary super-enhancers Super-enhancers comprise dense transcription factor platforms highly enriched for active chromatin marks. A paucity of functional data led us to investigate the role of super-enhancers in the mammary gland, an organ characterized by exceptional gene regulatory dynamics during pregnancy. ChIP-seq analysis for the master regulator STAT5A, the glucocorticoid receptor, H3K27ac and MED1 identified 440 mammary-specific super-enhancers, half of which were associated with genes activated during pregnancy (ref. 10). We interrogated the Wap super-enhancer, generating mice carrying mutations in STAT5-binding sites within its constituent enhancers. Individually, the most distal site displayed the greatest enhancer activity. However, combinatorial mutation analysis showed that the 1,000-fold induction in gene expression during pregnancy relied on all enhancers. Disabling the binding sites of STAT5, NFIB and ELF5 in the proximal enhancer incapacitated the entire super-enhancer. Altogether, these data suggest a temporal and functional enhancer hierarchy. The identification of mammary-specific super-enhancers and the mechanistic exploration of the Wap locus provide insights into the regulation of cell-type-specific expression of hormone-sensing genes. Histone methylation controls the mammary genomic clock We have shown previously that the temporal differentiation of mammary alveoli during pregnancy is controlled by the pituitary hormone prolactin and this signal is executed by the transcription factor STAT5 activating the mammary genome. It had been debated that the histone code, specifically the trimethylation status of lysine 27 on histone H3 (H3K27me3), controls the maintenance of mammary stem and epithelial cells and possibly their differentiation. We have addressed this question by mutating in mammary epithelium within mice those enzymes that control the establishment of (H3K27me3) marks (ref. 12) as well as their decommission (ref. 13). EZH2 is the enzyme (methyltransferase) that catalyzes the introduction of methyl groups onto H3K27 generating H3K27me3 and KDM6A (UTX) catalyzes the removal of methyl groups. Through deleting the Ezh2 gene specifically in mammary epithelial cells we addressed the question whether the methyltransferase and transcriptional co-activator EZH2 controls the differentiation clock (ref. 12). Ablation of Ezh2 from mammary stem cells resulted in precocious differentiation of alveolar epithelium during pregnancy and the activation of mammary-specific STAT5 target genes. This coincided with enhanced occupancy of these loci by STAT5, EZH1 and RNA Pol II. Limited activation of differentiation-specific genes was observed in mammary epithelium lacking both EZH2 and STAT5, suggesting a modulating but not mandatory role for STAT5. Loss of EZH2 did not result in overt changes in genome-wide and gene-specific H3K27me3 profiles, suggesting compensation through enhanced EZH1 recruitment. Differentiated mammary epithelia did not form in the combined absence of EZH1 and EZH2. Deletion of Kdm6a, the gene encoding the H3K2me3 demethylase, in the mammary luminal cell lineage led to a paucity of luminal cells and the expansion of basal cells (ref. 13). Genetically, these cells did not exhibit their distinctive landscape of transcription factor binding and displayed basal characteristics. The genomic H3K27me3 landscape was unaffected by the loss of KDM6A, supporting the notion for a demethylase-independent mechanisms. This was validated by analyzing mice expressing a catalytically inactive KDM6A. Normal mammary development was obtained in these mice. Importantly, putative mammary enhancers were enriched for KDM6A binding as determined by experiments that couple chromatin immunoprecipitation with deep sequencing (ChIP-seq). Thus, the mammary luminal lineage relies on KDM6A to ensure a transcription program leading to differentiated alveoli. Key collaborations LGP scientists engaged in several collaborations with NIH scientists (ref. 11), scientists within the extramural community within the US (ref. 1, 8), Germany (ref. 3, 6), Austria (ref. 9), Japan (ref. 4) and South Korea (ref. 2, 5).
Showing the most recent 10 out of 64 publications
