Genome-wide chromatin and gene expression profiling during memory formation and maintenance in adult mice

SEAMoD: A fully interpretable neural network for cis-regulatory analysis of differentially expressed genes

A common way to investigate gene regulatory mechanisms is to identify differentially expressed genes using transcriptomics, find their candidate enhancers using epigenomics, and search for over-represented transcription factor (TF) motifs in these enhancers using bioinformatics tools. A related follow-up task is to model gene expression as a function of enhancer sequences and rank TF motifs by their contribution to such models, thus prioritizing among regulators. We present a new computational tool called SEAMoD that performs the above tasks of motif finding and sequence-to-expression modeling simultaneously. It trains a convolutional neural network model to relate enhancer sequences to differential expression in one or more biological conditions. The model uses TF motifs to interpret the sequences, learning these motifs and their relative importance to each biological condition from data. It also utilizes epigenomic information in the form of activity scores of putative enhancers and automatically searches for the most promising enhancer for each gene. Compared to existing neural network models of non-coding sequences, SEAMoD uses far fewer parameters, requires far less training data, and emphasizes biological interpretability. We used SEAMoD to understand regulatory mechanisms underlying the differentiation of neural stem cell (NSC) derived from mouse forebrain. We profiled gene expression and histone modifications in NSC and three differentiated cell types and used SEAMoD to model differential expression of nearly 12,000 genes with an accuracy of 81%, in the process identifying the Olig2, E2f family TFs, Foxo3, and Tcf4 as key transcriptional regulators of the differentiation process.

A brain-specific pgc1α fusion transcript affects gene expression and behavioural outcomes in mice

This study shows that loss of a brain-specific fusion isoform of PGC1a leads to up-regulation of genes and motor impairments in mice, suggesting functional differences between PGC1 isoforms in the brain.

PGC1α is a transcriptional coactivator in peripheral tissues, but its function in the brain remains poorly understood. Various brain-specific Pgc1α isoforms have been reported in mice and humans, including two fusion transcripts (FTs) with non-coding repetitive sequences, but their function is unknown. The FTs initiate at a simple sequence repeat locus ∼570 Kb upstream from the reference promoter; one also includes a portion of a short interspersed nuclear element (SINE). Using publicly available genomics data, here we show that the SINE FT is the predominant form of Pgc1α in neurons. Furthermore, mutation of the SINE in mice leads to altered behavioural phenotypes and significant up-regulation of genes in the female, but not male, cerebellum. Surprisingly, these genes are largely involved in neurotransmission, having poor association with the classical mitochondrial or antioxidant programs. These data expand our knowledge on the role of Pgc1α in neuronal physiology and suggest that different isoforms may have distinct functions. They also highlight the need for further studies before modulating levels of Pgc1α in the brain for therapeutic purposes.

A hybrid de novo assembly of the sea pansy (Renilla muelleri) genome

BACKGROUND

More than 3,000 species of octocorals (Cnidaria, Anthozoa) inhabit an expansive range of environments, from shallow tropical seas to the deep-ocean floor. They are important foundation species that create coral "forests," which provide unique niches and 3-dimensional living space for other organisms. The octocoral genus Renilla inhabits sandy, continental shelves in the subtropical and tropical Atlantic and eastern Pacific Oceans. Renilla is especially interesting because it produces secondary metabolites for defense, exhibits bioluminescence, and produces a luciferase that is widely used in dual-reporter assays in molecular biology. Although several anthozoan genomes are currently available, the majority of these are hexacorals. Here, we present a de novo assembly of an azooxanthellate shallow-water octocoral, Renilla muelleri.

FINDINGS

We generated a hybrid de novo assembly using MaSuRCA v.3.2.6. The final assembly included 4,825 scaffolds and a haploid genome size of 172 megabases (Mb). A BUSCO assessment found 88% of metazoan orthologs present in the genome. An Augustus ab initio gene prediction found 23,660 genes, of which 66% (15,635) had detectable similarity to annotated genes from the starlet sea anemone, Nematostella vectensis, or to the Uniprot database. Although the R. muelleri genome may be smaller (172 Mb minimum size) than other publicly available coral genomes (256-448 Mb), the R. muelleri genome is similar to other coral genomes in terms of the number of complete metazoan BUSCOs and predicted gene models.

CONCLUSIONS

The R. muelleri hybrid genome provides a novel resource for researchers to investigate the evolution of genes and gene families within Octocorallia and more widely across Anthozoa. It will be a key resource for future comparative genomics with other corals and for understanding the genomic basis of coral diversity.

A novel method for culturing stellate astrocytes reveals spatially distinct Ca2+ signaling and vesicle recycling in astrocytic processes

Communication between astrocytes and neurons has been difficult to study because cultured astrocytes do not resemble those in vivo. Wolfes et al. develop a stellate astrocyte monoculture with physiological characteristics and find that VAMP2 and SYT7 mark distinct vesicle populations in astrocytes.

Interactions between astrocytes and neurons rely on the release and uptake of glial and neuronal molecules. But whether astrocytic vesicles exist and exocytose in a regulated or constitutive fashion is under debate. The majority of studies have relied on indirect methods or on astrocyte cultures that do not resemble stellate astrocytes found in vivo. Here, to investigate vesicle-associated proteins and exocytosis in stellate astrocytes specifically, we developed a simple, fast, and economical method for growing stellate astrocyte monocultures. This method is superior to other monocultures in terms of astrocyte morphology, mRNA expression profile, protein expression of cell maturity markers, and Ca²⁺ fluctuations: In astrocytes transduced with GFAP promoter–driven Lck-GCaMP3, spontaneous Ca²⁺ events in distinct domains (somata, branchlets, and microdomains) are similar to those in astrocytes co-cultured with other glia and neurons but unlike Ca²⁺ events in astrocytes prepared using the McCarthy and de Vellis (MD) method and immunopanned (IP) astrocytes. We identify two distinct populations of constitutively recycling vesicles (harboring either VAMP2 or SYT7) specifically in branchlets of cultured stellate astrocytes. SYT7 is developmentally regulated in these astrocytes, and we observe significantly fewer synapses in wild-type mouse neurons grown on Syt7^−/− astrocytes. SYT7 may thus be involved in trafficking or releasing synaptogenic factors. In summary, our novel method yields stellate astrocyte monocultures that can be used to study Ca²⁺ signaling and vesicle recycling and dynamics in astrocytic processes.

Genome-wide chromatin and gene expression profiling during memory formation and maintenance in adult mice

Recent evidence suggests that the formation and maintenance of memory requires epigenetic changes. In an effort to understand the spatio-temporal extent of learning and memory-related epigenetic changes we have charted genome-wide histone and DNA methylation profiles, in two different brain regions, two cell types, and three time-points, before and after learning. In this data descriptor we provide detailed information on data generation, give insights into the rationale of experiments, highlight necessary steps to assess data quality, offer guidelines for future use of the data and supply ready-to-use code to replicate the analysis results. The data provides a blueprint of the gene regulatory network underlying short- and long-term memory formation and maintenance. This 'healthy' gene regulatory network of learning can now be compared to changes in neurological or psychiatric diseases, providing mechanistic insights into brain disorders and highlighting potential therapeutic avenues.