No abstract

A complex landscape of genomic regulatory elements underpins patterns of metazoan gene expression, yet it has been technically difficult to disentangle composite regulatory elements within their endogenous genomic context. Expression of the Sox2 transcription factor (TF) in mouse embryonic stem cells (mESCs) depends on a distal regulatory cluster of DNase I hypersensitive sites (DHSs), but the contributions of individual DHSs and their degree of independence remain a mystery. Here, we comprehensively analyze the regulatory architecture of the Sox2 locus in mESCs using Big-IN to scarlessly deliver payloads ranging up to 143 kb, permitting deletions, rearrangements and inversions of single or multiple DHSs, and surgical alterations to individual TF recognition sequences. Multiple independent mESC clones were derived for each payload, extensively sequence-verified, and profiled for expression of Sox2 specifically from the engineered allele. We find that a single core DHS comprising a handful of key TF recognition sequences is sufficient to sustain significant expression in mESCs, though its contribution is modulated by additional DHSs. Moreover, their overall activity is influenced by specific DHS order and/or orientation effects. We built a highly predictive model for locus regulation which includes nonlinear components indicating both synergy and redundancy among. Our results suggest that composite regulatory elements and their influence on gene expression can be resolved to a tractable set of sequence features using synthetic regulatory genomics.

Prostate cancer (PCa) is a very complex disease that is a major cause of death in men worldwide. Currently, PCa dependence on the androgen receptor (AR) has resulted in use of AR antagonists and anti-androgen therapies that reduce endogenous steroid hormone production. However, within two to three years of receiving first line androgen deprivation therapy, the majority of patients diagnosed with PCa progress to castration resistant prostate cancer (CRPC). There is an urgent need for therapies that are more durable than antagonism of the AR axis. Studies of Runt-related transcription factors (RUNX) and their heterodimerization partner, Core-Binding Factor Subunit b (CBFβ), are revealing that the RUNX family are drivers of CRPC. In this review, we describe what is presently understood about RUNX members in PCa, including what regulates and is regulated by RUNX proteins, and the role of RUNX proteins in the tumor microenvironment and AR signaling. We discuss the implications for therapeutically targeting RUNX, the potential for RUNX as PCa biomarkers, and the current pressing questions in the field.

Genetically Engineered Mouse Models (GEMMs) aid in understanding human pathologies and developing new therapeutics, yet recapitulating human diseases authentically in mice is challenging to design and execute. Advances in genomics have highlighted the importance of non-coding regulatory genome sequences controlling spatiotemporal gene expression patterns and splicing to human diseases. It is thus apparent that including regulatory genomic regions in GEMMs is highly preferable for disease modeling, with the prerequisite of large-scale genome engineering ability. Existing genome engineering methods have limits on the size and efficiency of DNA delivery, hampering routine creation of highly informative GEMMs. Here, we describe mSwAP-In (mammalian Switching Antibiotic resistance markers Progressively for Integration), a method for efficient genome rewriting in mouse embryonic stem cells. We first demonstrated the use of mSwAP-In for iterative genome rewriting of up to 115 kb of the Trp53 locus, as well as for genomic humanization of up to 180 kb ACE2 locus in response to the COVID-19 pandemic. Second, we showed the hACE2 GEMM authentically recapitulated human ACE2 expression patterns and splicing, and importantly, presented milder symptoms without mortality when challenged with SARS-CoV-2 compared to the K18-ACE2 model, thus representing a more authentic model of infection.

No abstract