Investigating Rapid Nuclear Mechanotransduction by Mapping Genome-wide Changes in Transcription and Chromatin Accessibility in Response to Mechanical Stretch

Introduction: Cells can mediate tissue maintenance and growth through mechanotransduction, i.e., the translation of mechanical forces into biochemical signals. Whereas multiple mechanotransduction pathways have been identified at the plasma membrane and in the cytoplasm, mounting evidence suggests that the nucleus also plays a crucial part in cellular mechanotransduction. Proteins at the nuclear envelope, such as lamins and the LINC complex, play a crucial part in the transmission of force from the cytoskeleton to the nuclear interior, with the potential to influence genome organization and transcriptional regulation. It is not yet known, however, if changes in chromosomal organization and the concurrent gene activation, initiated by increased binding of transcription factors (TFs) or polymerases, following mechanical stimuli are a direct consequence of nuclear deformation, or if they are due to the action of upstream cytoplasmic mechanotransduction pathways. To address this question, we investigated the regulatory mechanisms mediated by nuclear mechanotransduction using a custom-developed in vitro assay that can probe the immediate (< minutes) genome-wide response of both the transcriptome and chromatin accessibility to cyclic mechanical stretch with unprecedented sensitivity and resolution (Fig. 1A).

Materials and

Methods: Human skin fibroblasts were seeded on silicone membranes that were subsequently subjected to cyclic biaxial stretching for 0, 10, or 30 minutes. To interrogate the immediate response of the transcriptome response following cyclical stretch, cells were processed by precision nuclear run-on sequencing (PRO-seq), a novel genome-wide transcriptome sequencing technique that, unlike traditional RNA sequencing, enriches for nascent RNA transcripts and simultaneously provides a base-pair resolution map of active RNA polymerases. To probe the effect of short-term mechanical stress on genome organization, we also performed an improved version of Assay for Transposase-Accessible Chromatin with high-throughput sequencing (Omni-ATAC) on stretched fibroblasts. Comparative analyses between each timepoints allowed for the identification of differentially expressed genes (DEGs), differentially activated transcriptional regulatory elements (dTREs), and associated TF motifs using PRO-seq data, as well as TF footprints and differentially accessible regions (DARs) using Omni-ATAC data.

Results, Conclusions, and Discussions: PRO-seq experiments revealed a rapid increase in RNA polymerase II (PolII) activity within the gene body of several known (EGR1, FOS/B) mechanoresponsive genes as early as 10 minutes after mechanical stretch (Fig. 1B), a much shorter timeframe than previously reported. PRO-seq PolII processing along the gene body suggests that gene activation was initiated < 1min after the stretch, i.e., substantially faster than canonical cytoplasmic mechanotransduction cascades, which take several minutes to reach the nucleus. Our data also identified different modalities of activation within the stretch-activated genes, based on the presence or lack of accumulated paused PolII within the promoter-proximal region of the gene prior to mechanical stimulus (Fig. 1C). This suggests that stretch can lead to gene activation by either inducing the release of paused PolII (e.g., EGR1, FOS), or leading to the immediate recruitment and elongation of PolII (e.g., NFATC2). The increased sensitivity of PRO-seq, compared to traditional RNA-seq, allowed detection of NFATC2 as a novel mechanoresponsive gene. Several of the identified mechanoresponsive genes were also found in mouse myotubes (Figure 1D), suggesting that the underlying molecular mechanism is conserved across cell types. Cross-referencing PRO-seq data with Omni-ATAC revealed that the promoter of identified mechanically-induced genes were already accessible before mechanical stretching was induced (Fig. 1E). Analysis of predicted upstream regulators (Fig. 1F) based on PRO-seq and Omni-ATAC data identified several potential regulators related to important pathways like MAPK, NFkB, and the FOX family, which are involved in stress response and cell proliferation. Our results suggest that these regulators, previously known to respond to mechanotransduction downstream of cytoplasmic signaling pathways, may also already be present in the nucleus, where they can be rapidly activated by mechanical stimulation, thus presenting an alternative and more rapid mechanotransduction pathway.
In conclusion, our findings suggest that genomic regions associated with mechanoresponsive genes are already poised for gene transcription, triggering either polymerase pause release or recruitment and elongation, within seconds of the mechanical stimulus. These findings point to an exciting, previously unrecognized role of the nucleus itself in regulating the activation of the identified mechanoresponsive genes, which we are currently investigating in more detail.

Acknowledgements (Optional):