Carcinoma cells residing in an intermediate phenotypic state along the epithelial–mesenchymal (E–M) spectrum are associated with malignant phenotypes, such as invasiveness, tumor-initiating ability, and metastatic dissemination. Using the recently described CD104+/CD44hi antigen marker combination, we isolated highly tumorigenic breast cancer cells residing stably—both in vitro and in vivo—in an intermediate phenotypic state and coexpressing both epithelial (E) and mesenchymal (M) markers. We demonstrate that tumorigenicity depends on individual cells residing in this E/M hybrid state and cannot be phenocopied by mixing two cell populations that reside stably at the two ends of the spectrum, i.e., in the E and in the M state. Hence, residence in a specific intermediate state along the E–M spectrum rather than phenotypic plasticity appears critical to the expression of tumor-initiating capacity. Acquisition of this E/M hybrid state is facilitated by the differential expression of EMT-inducing transcription factors (EMT-TFs) and is accompanied by the expression of adult stem cell programs, notably, active canonical Wnt signaling. Furthermore, transition from the highly tumorigenic E/M state to a fully mesenchymal phenotype, achieved by constitutive ectopic expression of Zeb1, is sufficient to drive cells out of the E/M hybrid state into a highly mesenchymal state, which is accompanied by a substantial loss of tumorigenicity and a switch from canonical to noncanonical Wnt signaling. Identifying the gatekeepers of the various phenotypic states arrayed along the E–M spectrum is likely to prove useful in developing therapeutic approaches that operate by shifting cancer cells between distinct states along this spectrum.

To determine the prevalence of cotranscriptional splicing in Drosophila, we sequenced nascent RNA transcripts from Drosophila S2 cells as well as from Drosophila heads. Eighty-seven percent of the introns assayed manifest >50% cotranscriptional splicing. The remaining 13% are cotranscriptionally spliced poorly or slowly, with~3% being almost completely retained in nascent pre-mRNA. Although individual introns showed slight but statistically significant differences in splicing efficiency, similar global levels of splicing were seen from both sources. Importantly, introns with low cotranscriptional splicing efficiencies are present in the same primary transcript with efficiently spliced introns, indicating that splicing is intron-specific. The analysis also indicates that cotranscriptional splicing is less efficient for first introns, longer introns, and introns annotated as alternative. Finally, S2 cells expressing the slow RpII215 C4 mutant show substantially less intron retention than wild-type S2 cells.

Spliceosome assembly and/or splicing of a nascent transcript may be crucial for proper isoform expression and gene regulation in higher eukaryotes. We recently showed that cotranscriptional splicing occurs efficiently in Drosophila, but there are not comparable genome-wide nascent splicing data from mammals. To provide this comparison, we analyze a recently generated, high-throughput sequencing data set of mouse liver nascent RNA, originally studied for circadian transcriptional regulation. Cotranscriptional splicing is approximately twofold less efficient in mouse liver than in Drosophila, i.e., nascent intron levels relative to exon levels are~0.55 in mouse versus 0.25 in the fly. An additional difference between species is that only mouse cotranscriptional splicing is optimal when 59-exon length is between 50 and 500 bp, and intron length does not correlate with splicing efficiency, consistent with exon definition. A similar analysis of intron and exon length dependence in the fly is more consistent with intron definition. Contrasted with these differences are many similarities between the two systems: Alternatively annotated introns are less efficiently spliced cotranscriptionally than constitutive introns, and introns of single-intron genes are less efficiently spliced than introns from multi-intron genes. The most striking common feature is intron position: Cotranscriptional splicing is much more efficient when introns are far from the 39 ends of their genes. Additionally, absolute gene length correlates positively with cotranscriptional splicing efficiency independently of intron location and position, in flies as well as in mice. The gene length and distance effects indicate that more ''nascent time'' gives rise to greater cotranscriptional splicing efficiency in both systems.

Drosha is a type III RNase, which plays a critical role in miRNA biogenesis. Drosha and its double-stranded RNA-binding partner protein Pasha/DGCR8 likely recognize and cleave miRNA precursor RNAs or pri-miRNA hairpins cotranscriptionally. To identify RNAs processed by Drosha, we used tiling microarrays to examine transcripts after depletion of drosha mRNA with dsRNA in Drosophila Schneider S2 cells. This strategy identified 137 Drosha-regulated RNAs, including 11 putative pri-miRNAs comprising 15 annotated miRNAs. Most of the identified pri-miRNAs seem extremely large, >10 kb as revealed by both the Drosha knockdown strategy and by RNA PolII chromatin IP followed by Drosophila tiling microarrays. Surprisingly, more than a hundred additional RNAs not annotated as miRNAs are under Drosha control and are likely to be direct targets of Drosha action. This is because many of them encode annotated genes, and unlike bona fide pri-miRNAs, they are not affected by depletion of the miRNA processing factor, dicer-1. Moreover, application of the evofold analysis software indicates that at least 25 of the Drosha-regulated RNAs contain evolutionarily conserved hairpins similar to those recognized by the Drosha-Pasha/DGCR8 complex in pri-miRNAs. One of these hairpins is located in the 59 UTR of both pasha and mammalian DGCR8. These observations suggest that a negative feedback loop acting on pasha mRNA may regulate the miRNA-biogenesis pathway: i.e., excess Drosha cleaves pasha/DGCR8 primary transcripts and leads to a reduction in pasha/DGCR8 mRNA levels and Pasha/DGCR8 synthesis.