The RAD18 (E3) ubiquitin ligase, a key DNA damage tolerance regulator that also functions in DNA double-strand break repair, is overexpressed in the brain cancer glioblastoma. Here, we show that RAD18 promotes glioblastoma cell proliferation in the absence of exogenous damage, independently of its catalytic activity. RAD18 downregulation arrests glioblastoma cells in the G1 phase of the cell cycle, leading to senescence onset, with no apparent increase in DNA damage. We also show that RAD18 sustains glioblastoma stem cells' self-renewal and the growth of tumor orthotropic xenografts in mice. Further, we show that increased RAD18 expression enhances the growth of non-transformed cells and induces the features of oncogenic transformation. Mechanistically, RAD18 interacts and negatively regulates the NF2 tumor suppressor through its SAP domain, thereby facilitating stabilization and nuclear retention of the YAP1 transcription factor. Altogether, these data propose RAD18 as a key target to sensitize glioblastoma to therapy.
Abstract How broad phenotypic variation is maintained in natural populations in the face of selection is a central question in evolutionary biology. We address this question in the water strider Microvelia longipes, where males exhibit striking variation in rear leg length used in male-male contests for dominance. Using reaction norm experiments on inbred lines, we demonstrate that phenotypic plasticity contributes to expanding phenotypic variation, but requires high genetic variation to generate the broad range of trait expression observed in natural populations. Experimental evolution favouring trait exaggeration revealed that directional sexual selection not only fails to erode variation of male rear leg length, but rather amplifies it beyond the natural distribution. Additionally, male-limited selection in favour of dominance generated substantial fecundity costs in females, underscoring the role of sexual conflict driven by females in constraining exaggerated secondary sexual traits in males. Our findings show that sexually antagonistic selection and directional sexual selection jointly generate high genetic variation, which phenotypic plasticity inflates into broad phenotypic distribution of male weapon size. This provides an empirical explanation for the high variability of male exaggerated weapons in nature.
Identifying non-coding regulatory elements in the genome poses a challenge in most organisms. Classical methods rely on trial and error to test the regulatory activities of DNA fragments using reporter constructs. In large eukaryotic genomes, where cis-regulatory elements can spread over long distances, separated by large stretches of non-functional DNA, this trial and error approach is particularly challenging. Here, we generate two types of resources that can be used to narrow the search for such cis-regulatory elements in the 3.6 Gbp genome of Parhyale hawaiensis (comparable in size to the human genome). First, we use bulk ATAC-seq to uncover genome-wide patterns of chromatin accessibility in embryonic and adult tissues of Parhyale (whole embryos and legs), and single-nucleus ATAC-seq to identify regions of open chromatin in diverse cell types recovered from adult legs, including epidermal, neuronal, muscle and blood cells. Second, by sequencing the genomes of three congeneric species of Parhyale hawaiensis – P. darvishi, P. aquilina and P. plumicornis – we identify islands of sequence conservation across the genome, corresponding to DNA elements that are functionally constrained during evolution. We present an approach by which low-coverage (10-15x) short-read genome sequencing, without genome assembly, is sufficient to provide reliable maps of sequence conservation. This approach cuts the cost and labour required to generate these maps, making the identification of cis-regulatory elements more widely accessible. We demonstrate the utility of these resources by identifying cis-regulatory elements that drive robust expression of fluorescent reporters ubiquitously and in specific cell types.
The circadian system aligns behavior and physiology with the 24-hour environmental cycle through a distributed network of clocks including the master pacemaker in the suprachiasmatic nucleus (SCN) and an autonomous retinal clock critical for local retinal physiology and function. Although both clocks are entrained by light, they differ in their photoreceptor inputs and light sensitivity. The specific contributions and mechanisms by which distinct photoreceptor pathways drive their photoentrainment, however, remain incompletely understood. In this study, we conducted a comprehensive transcriptomic and integrative comparative analysis of retinal and SCN circadian responses to 530 nm monochromatic light using mouse models lacking specific photoreceptors or key components of signaling pathways. Under photopic conditions, we found that each tissue displays distinct light-responsive transcriptional signatures across genotypes, yet both shared a conserved cluster of rod-driven immediate early-genes. Strikingly, the light-evoked transcriptional response was not sufficient to shift the phase of the SCN clock, in contrast to its robust phase-shifting effect on the retinal clock. Furthermore, by genetically disrupting rod/cone electrical coupling and pharmacologically isolating rod pathways, we identified the OFF-cone bipolar cell circuit as both necessary and sufficient to mediate light-induced phase resetting of the retinal clock. Together, these findings delineate the specialized retinal circuitry that supports circadian entrainment and highlight a fundamental divergence between retinal and SCN mechanisms of photic timekeeping.
Transgenic markers and tools have revolutionised how we study cells and developing organisms. Some of the elements needed to construct those tools are universally applicable (e.g. fluorescent proteins), while others are species-specific (e.g. cis-regulatory elements driving transcription). Membrane-localising signals that target proteins to the plasma membrane have been identified in several model organisms. Unfortunately, the efficacy of these signals varies greatly across species. To address this problem, we generated a toolkit of 11 membrane-localising tags that can be screened rapidly in diverse organisms. The toolkit includes tags that target the plasma membrane through different mechanisms, including signal peptides, the attachment of lipids, and fusion with lipid-binding domains. Each tag has been fused to the red fluorescent protein mScarlet3 and placed downstream of a T7 promoter, allowing the in vitro production of mRNA that can be readily delivered in a wide range of embryos and cells of interest. Through a collaborative effort, we tested this toolkit in ten species of animals spanning diverse phyla, including chordates, echinoderms, arthropods, nematodes, annelids, flatworms and cnidarians. We identify robust membrane-localising tags in each of these animals, and in one of animals' closest relatives, the choanoflagellates. Three tags (KRas, GAP43 and Src64B) work in all of the species tested.
