Exercise modulation of the alternative splicing landscape in human tissues

This study reveals that acute endurance and resistance exercise induce distinct, time-dependent patterns of alternative splicing across human skeletal muscle, adipose tissue, and blood, largely independent of RNA expression changes, thereby establishing alternative splicing regulation as a major molecular mediator of the multi-organ responses to physical activity.

The Gist

The Big Picture: Exercise is a "Software Update" for Your Body

Imagine your body is a massive, high-tech factory. For a long time, scientists knew that exercise (like running or lifting weights) told the factory to produce more of certain machines (proteins) and produce less of others. This is like telling the factory to run the "Assembly Line A" faster and slow down "Assembly Line B."

But this new study discovered something even more fascinating: Exercise doesn't just change how much the factory produces; it changes how the machines are built.

The study looked at a process called Alternative Splicing. Think of your DNA as a giant instruction manual for building proteins. Usually, the factory reads the manual from start to finish. But "splicing" is like having a master editor who can cut out certain paragraphs and rearrange others before the machine is built. This means one single instruction manual can be edited to build three slightly different versions of the same machine.

The main finding: When you exercise, your body acts like a frantic editor, rewriting the instructions for thousands of proteins to create new, specialized versions of them. This happens even if the total amount of the protein doesn't change.

---

The Experiment: The "MoTrPAC" Factory Tour

The researchers (part of a huge team called MoTrPAC) recruited sedentary adults and put them through two different types of workouts:

1. Endurance Exercise (EE): Like a long, steady bike ride (aerobic).
2. Resistance Exercise (RE): Like lifting heavy weights (anaerobic).

They took samples from three key "departments" in the body factory:

• Skeletal Muscle: The engines that move you.
• Adipose Tissue: The storage tanks (fat).
• Blood: The delivery trucks carrying messages.

They checked these samples immediately before the workout, and then at various times after (15 minutes, 3.5 hours, and 24 hours later) to see how the "editing" changed over time.

Key Discoveries

1. The Muscle Factory is the Busiest Editor

The most dramatic changes happened in the muscles. While fat and blood saw some editing, the muscles were rewriting thousands of instructions.

• The Analogy: If your body is a library, the muscle section is where the librarians are frantically rewriting the books while you run. The fat and blood sections are just making a few notes in the margins.

2. Two Different Workouts, Two Different Editors

Endurance and Resistance exercises didn't just do the same thing twice as hard; they triggered different editing styles.

• Endurance (Running): The editing happened fast (peaking at 15 minutes) and then settled down. It focused on things like heat shock proteins (repairing heat damage) and energy production.
• Resistance (Lifting): The editing started later but kept going for much longer (peaking at 3.5 hours and still active at 24 hours). It focused heavily on the structural parts of the muscle (the "scaffolding" and "beams" that hold the muscle together).
• The Analogy: Running is like a quick, intense storm that fixes the roof immediately. Lifting weights is like a slow, steady renovation crew that is still hammering away the next day, reinforcing the foundation.

3. The "Quantity vs. Quality" Surprise

Scientists used to think that if you wanted a new protein, you had to make more of the gene (the instruction manual).

• The Finding: In this study, 67% of the changes happened without changing the amount of the gene at all.
• The Analogy: Imagine a bakery. Usually, to get more chocolate cake, you bake more cakes. But here, the baker realized they could just change the recipe for the existing batter. They took the same amount of batter but swapped out the chocolate chips for nuts, creating a completely different product without baking a single extra cake. This is "splicing": changing the recipe, not the volume.

4. The "Foremen" (Regulators)

Who is telling the editors what to cut and paste? The study found two types of "foremen":

• RNA-Binding Proteins (RBPs): These are like the line managers holding the scissors. The study found that exercise changes how these managers are "phosphorylated" (a chemical switch that turns them on or off).
• Transcription Factors (TFs): These are the architects who decide which parts of the DNA are accessible. The study found that exercise changes the activity of these architects, opening up new areas of the instruction manual for editing.

Why Does This Matter?

This study explains how exercise makes us healthier at a molecular level.

• It's not just about building bigger muscles or burning more calories.
• It's about customizing the proteins in your body to handle stress, repair damage, and function more efficiently.
• Because the editing process is so specific to the type of exercise, this helps explain why running and weightlifting give you different health benefits.

The Takeaway

Exercise is a powerful signal that tells your body to re-engineer its own software. It doesn't just turn the volume up on your genes; it rewrites the code to create specialized tools that help you survive and thrive. Whether you are a runner or a weightlifter, your body is frantically editing its instruction manual to build the perfect version of you for that specific activity.

Technical Summary

1. Problem Statement

While the health benefits of exercise (cardiovascular, metabolic, cognitive) are well-established, the precise molecular mechanisms driving acute physiological adaptations remain incompletely understood. Previous research has largely focused on changes in gene expression (transcript abundance) and protein levels. However, Alternative Splicing (AS) is a critical co-transcriptional mechanism that generates transcriptome and proteome diversity, with an estimated 95% of human genes producing multiple isoforms.

• Gap: There is a lack of comprehensive, genome-wide data on how acute exercise (both endurance and resistance) modulates the AS landscape across multiple human tissues.
• Unknowns: It is unclear whether exercise-induced AS changes are independent of gene expression changes, which biological pathways are affected, and what regulatory mechanisms (RNA-binding proteins, transcription factors, chromatin accessibility) drive these splicing alterations.

2. Methodology

The study utilized data from the MoTrPAC (Molecular Transducers of Physical Activity Consortium) pre-pandemic cohort, employing a multi-omic, time-series approach.

• Study Design:
  • Cohort: 175 healthy, sedentary adults randomized into three groups: Endurance Exercise (EE, n=65), Resistance Exercise (RE, n=73), and Non-Exercise Controls (n=37).
  • Intervention: Acute bouts of cycling (EE) or weightlifting (RE).
  • Tissues: Skeletal Muscle (SKM), Adipose Tissue, and Blood.
  • Timepoints: Samples collected pre-exercise and at multiple intervals post-exercise (e.g., 15 min, 3.5 hr, 24 hr) to capture dynamic temporal responses.
• Multi-Omic Data Generation:
  • Transcriptomics: Deep paired-end RNA-seq (avg. 60M reads) to quantify Percent Spliced In (PSI) values.
  • Epigenomics: ATAC-seq for chromatin accessibility.
  • Proteomics/Phosphoproteomics: LC-MS/MS (TMT 16-plex) for protein abundance and phosphorylation states.
• Computational Analysis:
  • AS Detection: Combined two complementary algorithms, rMATS-turbo (junction-based) and SUPPA2 (isoform-based), to identify Differential Alternative Splicing (DAS) events (Skipped Exon, Alternative 5'/3' Sites, Retained Intron).
  • Statistical Modeling: Linear mixed-effects models with participant ID as a random effect to account for within-subject correlations. DAS was defined as significant changes in PSI independent of gene expression changes in controls.
  • Integrative Analysis:
    • WGCNA: To identify co-expression modules of spliceosome genes.
    • RBP Analysis: Enrichment of RNA-binding protein (RBP) motifs (using ENCODE eCLIP data) at DAS sites; correlation of RBP phosphorylation with target PSI changes.
    • TF Analysis: Used the MAGICAL framework to infer Splice Regulatory Circuits (SRCs) linking chromatin accessibility, Transcription Factor (TF) binding motifs, and DAS events.

3. Key Contributions

• First Multi-Tissue AS Landscape: Provides the first comprehensive map of acute exercise-induced alternative splicing across skeletal muscle, adipose, and blood in humans.
• Independence of Regulation: Demonstrates that the majority of exercise-induced AS events occur independently of changes in total gene expression (transcript abundance).
• Modality Specificity: Distinguishes the distinct temporal and functional signatures of Endurance vs. Resistance exercise on the splicing landscape.
• Mechanistic Inference: Integrates phosphoproteomics and chromatin accessibility to infer the regulatory drivers (RBPs and TFs) of splicing changes, moving beyond correlation to mechanistic hypothesis generation.

4. Key Results

A. Global DAS Landscape

• Volume: Identified 5,102 distinct DAS events.
• Tissue Specificity: Skeletal muscle showed the highest abundance of DAS events, followed by adipose tissue and blood.
• Temporal Dynamics:
  • EE: Peak DAS events occurred early (15 min post-exercise) and declined rapidly.
  • RE: DAS events increased over time, peaking at 3.5 hours and remaining elevated at 24 hours.
• Independence from Expression: 67% of DAS events were independent of differential gene expression (DEG). In SKM, ~82% of early-phase DAS events occurred in non-DEGs.
• Protein Impact: 89% of DAS events resulted in changes to the protein-coding sequence (CDS), suggesting profound impacts on protein function and structure.

B. Biological Pathways

• Shared Pathways: Both EE and RE regulated genes involved in RNA splicing, focal adhesion, and cell-substrate junctions.
• Modality-Specific Pathways:
  • EE: Enriched for heat shock protein binding (early phase) and mitochondrial processes.
  • RE: Enriched for sarcomere organization, actin cytoskeleton, and muscle system processes (especially at later timepoints).

C. Regulatory Mechanisms

• Spliceosome Regulation: The spliceosome machinery itself is a target of exercise. 66% of spliceosome-related genes showed regulation at the RNA expression, AS, or protein phosphorylation levels.
• RNA-Binding Proteins (RBPs):
  • Identified 49 RBPs with enriched binding sites at DAS loci.
  • Phosphorylation: Key RBPs (e.g., SERBP1, ABCF1, PCBP1) showed differential phosphorylation post-exercise.
  • Correlation: Strong correlations were found between the phosphorylation status of specific RBPs (e.g., SERBP1 at Ser-330) and the PSI values of their target AS events (e.g., SRSF5).
• Transcription Factors (TFs) & Chromatin:
  • Inferred 2,764 Splice Regulatory Circuits (SRCs).
  • Top TFs included immediate early genes (FOS, JUN, JUNB, NR4A1) and others like MEF2C, STAT1, RXRA.
  • Mechanism: Changes in chromatin accessibility at TF binding sites (e.g., ~75kb upstream of PPM1A) correlated with DAS events and TF phosphorylation (e.g., NR4A1), suggesting co-transcriptional regulation.

5. Significance

• Mechanistic Insight: This study establishes that AS is a primary, rapid, and independent mechanism by which exercise remodels the proteome, distinct from simple gene up/down-regulation.
• Personalized Medicine: The identification of modality-specific splicing signatures (EE vs. RE) and inter-individual variation in spliceosome "set points" (via WGCNA) provides a basis for tailoring exercise prescriptions for specific therapeutic goals (e.g., metabolic health vs. muscle hypertrophy).
• Therapeutic Targets: The identification of specific RBPs (like PCBP1) and TFs (like MEF2C, NR4A1) as regulators of exercise-induced splicing offers new molecular targets for treating chronic diseases (diabetes, cardiovascular disease, sarcopenia) where exercise is a primary intervention.
• Resource: The study provides a rich, multi-omic dataset and analytical framework (MAGICAL adaptation for splicing) that serves as a foundation for future research into post-transcriptional regulation in human physiology.