Argonaute2 and LaminB modulate gene expression by controlling chromatin topology (neuRNA-seq)

Drosophila Argonaute2 (AGO2) is known to regulate expression of certain loci in an RNA interference-independent manner, but its genome-wide function on chromatin remains unknown. In order to elucidate the mechanism by which AGO2 regulates transcription, we performed mass spec analysis of factors that interact with AGO2 in nuclear extracts and identified RNA Polymerase II (Pol II) and LaminB among the top-associated proteins. We found that AGO2 and LaminB attenuate transcription genome-wide, including borders between both LaminB-associated domains (LADs) and topologically-associated domains (TADs). Furthermore, we identified a somatic target of AGO2 transcriptional repression, nht, which is immersed in a LAD located within a repressive TAD. Null mutation of AGO2 leads to ectopic expression of spermatogenesis genes dependent on nht. Finally, depletion of either AGO2 or LaminB results in reduced looping interactions within the TAD as well as ectopic inter-TAD interactions, as detected by 4C-seq analysis. Overall, our findings reveal coordination of AGO2 and LaminB to dictate overall genome architecture and thereby regulate gene expression. Nascent RNA-seq (neuRNA-seq) in mock, AGO2, LaminB, and Dcr-2 knockdowns. In addition, AGO2 was depleted using an alternative siRNA designed against its 5'UTR (si5UTRAGO2) and rescue with a wild type version of AGO2 (pwahf_ago2) or catalytic mutant AGO2V966M (pwahf_v966m). Three separate experiments are included here, and each experiment has between 2 and 3 biological replicates as indicated. The supplemental counts table contains raw read counts in genes in each sample, and the supplemental_sample_lookup table describes the experimental design. The "experiment" column of the supplemental_sample_lookup table indicates the experiment from which each sample came from. When performing differential expression analysis with these counts, only samples that came from the same experiment should be compared; in particular note that each experiment has its own set of mock control samples.