Regulatory and functional architecture of 'Nested-Sensitivity' microexon programs

Project Information

NESTEDMICS

Grant agreement ID: 101002275

DOI

10.3030/101002275

Project closed

EC signature date 7 May 2021

Start date 1 July 2021

End date 30 June 2026

Funded under

• EXCELLENT SCIENCE - European Research Council (ERC)

Total cost

€ 1 974 541,00

EU contribution

€ 1 974 541,00

Coordinated by

UNIVERSIDAD POMPEU FABRA
Spain

Periodic Reporting for period 3 - NESTEDMICS (Regulatory and functional architecture of 'Nested-Sensitivity' microexon programs)

Reporting period: 2023-10-01 to 2025-03-31

Summary of the context and overall objectives of the project

Neural microexons are a paradigmatic example of a cell type-specific transcriptomic program. Microexons are tiny exons that we revealed to have striking neuronal specificity established by their master splicing regulator Srrm4, which activates them during neuronal differentiation. However, our data challenge this on/off regulatory and functional paradigm. We found that a related paralog, Srrm3, is lowly but significantly expressed also in endocrine pancreas and, together with Srrm4, configure a 3-step switch of Srrm3/4 activity in pancreas (low), brain (mid) and retina (high). These different levels of expression activate increasingly larger subsets of microexons in the three tissues, configuring a triple-nested microexon program. Remarkably, initial results support a model in which microexon subclass inclusion is dictated largely by their sensitivity to Srrm3/4, and each subclass is differentially enriched for distinct functional categories including vesicle-mediated transport, neuronal differentiation and cilium biogenesis.

This project assesses the regulatory and functional architecture of this new paradigm by answering:
(1) How are the different levels of the master regulators controlled in each cell type?
(2) How are the distinct sensitivities of microexons to Srrm3/4 genomically encoded?
(3) What are the functional implications of the 'nestedness' of the microexon programs?
(4) How does misregulation of the nested programs contribute to disease?

We are pursuing these goals by a combination of high-throughput methods and focused experiments using in vitro and in vivo systems, using mouse and zebrafish as model organisms. The expected results will provide a transformative multi-level portrait of microexons, from quantitative regulatory logic to organismal functions. Moreover, this novel paradigm is likely to apply to many other master regulators, expanding the impact of the project and shedding new light into how cell type-specific transcriptomes are established. Therefore, the work from this project will provide novel conceptual tools for researchers working on genes involved in gene regulation in different fields.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

In general terms, the project is advancing correctly for most Aims, and some important articles have been published. The project consists of four Aims, which can be summarized by four questions on the functional and regulatory architecture of the nested microexon programs:

(1) How are the different levels of the master regulators controlled in each cell type? Through the analysis of RNA-seq, ATAC-seq and H3K27ac ChIP-seq data we identified (i) a glucose-responsive, pancreas-specific superenhancer regulating SRRM3, and (ii) a weak enhancer in the first intron of SRRM3, also relevant for its expression (Juan-Mateu et al, Nat Metab 2023). We found that both regulatory elements contain human variants associated with pathophysiological conditions. At the experimental level, sorting fluorescent cells from our zebrafish reporter lines for pancreas is proven difficult, but we hope to obtain them for ATAC-seq from collaborators.

(2) How are the distinct sensitivities of microexons to Srrm3/4 genomically encoded? We have designed five minigene libraries, covering length, cis-regulatory rules and evolution, expanding what we have originally planned. These libraries were tested with different levels of Srrm3/4 expression. The results led to an unexpected conclusion: the sensitivities are mainly due to differences in the core splicing architecture of microexons and not to differences in binding and effect of SRRM4. The manuscript with these results has been posted in bioRxiv (Bonnal et al 2024) and it is in advance stages of peer review.

(3) What are the functional implications of the 'nestedness' of the microexon programs? This is the largest Aim and can be divided into pancreas, neural and retina sub-aims. For pancreas, we have already published a first study with the functional characterization of pancreas microexons (IsletMICs) using cell cultures and a constitutive mouse KO for Srrm3 (Juan-Mateu et al, Nat Metab 2023). We have also investigated the role of Srrm3 and Srrm4 in glucose homeostasis in zebrafish, with non-conclusive results for now. For the neural part, we have succeeded in completing and publishing our work on the functional characterization of the impact of a neural-specific microexon in DAAM1. It has been published in Nature Communications (Polinski et al 2025). In addition, we are preparing a small follow up to be submitted. The work on another neural-specific microexon candidate (Unc13b) is ongoing.. For the retina, we have already published a first manuscript corresponding to the first tasks (Ciampi et al, PNAS 2022) and have generated five retina-specific microexon (RetMIC) KOs in zebrafish, and have assayed them for vision defects.

(4) How does misregulation of the nested programs contribute to disease? We have performed multiple computational analyses with available and de novo data for the three nested programs. The key results are reported in the manuscripts mentioned above. We are now assessing the potential of antisense oligonucleotides to manipulate IsletMIC inclusion with the hope of using them in diabetogenic contexts.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

The published and submitted manuscripts resulting from the work done so far, particularly for Aims 3 and 4, report for the first time the impact of microexons in endocrine pancreas and photoreceptor biology and suggest their importance for certain human disease. Specifically, we have found human genomic variants associated with glycemic control and type 2 diabetes risk in SRRM3, and discovered that both IsletMICs and RetMICs are in genes enriched for loci genetically linked to type 2 diabetes and retina degenerative disorders, respectively. Moreover, using high-throughput splicing assays, we have made solid advances towards elucidating the cis-regulatory code for the sensitivity of microexon splicing to a specific splicing factor, which had never been investigated before, and much less so at this level of resolution.

We expect to further elucidate the impact of microexons for endocrine pancreas and glucose homeostasis, neuronal differentiation and function, and photoreceptor development and vision. From these insights altogether, we will better understand how microexons, as a special type of genetic element, contribute to module tissue-specific proteomes. As recent progress, we have now generated a new model system (conditional Srrm3 KO mouse) that is providing insights into the nested microexon programs in a more precise manner. We are also delving into the metabolism of Srrm3 depleted cells.

We also expect to get exciting insights with respect to Srrm3 and microexon's relevance for type 2 diabetes risk and, ideally, whether they can be used as therapeutic targets. We are beginning to explore the modulation of IsletMICs using antisense oligonucleotides in vitro but also in mouse models of type 2 diabetes. In parallel, we are investigating whether patients with unexplained genetic retinal dystrophies have mutations in RetMICs or NeuralMICs that might have been previously overlooked.

Finally, we are excited about applying the power of artificial intelligence to our parallel splicing assay data, and we have initiated two collaborations with dedicated experts. We hope this can help us to obtain a reliable predictor of sensitivity to then apply to any custom sequence.

Summary of Aims and rationale