henhao
Sign in
PhysiologyMolecular Metabolism
Peer-reviewed journal article

Exercise training remodels inter-organ endocrine networks

Cheehoon AhniD, Andrea Lynn HeveneriD, Laurie J. GoodyeariD, Sue C. BodineiD, Karyn Ann EsseriD, Marcus Michael SeldiniD, Lauren Marie SparksiD

Abstract

Exercise induces organism-wide molecular adaptations, partly mediated by humoral factors released in response to acute and chronic physical activity. However, the extent and specificity of endocrine effects from training-induced secreted factors remain unclear. Here, we applied systems genetics approaches to quantify inter-organ endocrine networks using multi-tissue transcriptomics and proteomics data collected from endurance-trained rats in The Molecular Transducers of Physical Activity Consortium (MoTrPAC). Eight weeks of endurance training significantly altered both the magnitude and specificity of endocrine effects across multiple origin-target tissue pairs. Subcutaneous white adipose tissue emerged as a key endocrine regulator impacted by training, while extracellular matrix-derived factors were identified as globally regulated secretory features in trained vs sedentary animals. Notably, secretory Wnt signaling factors were identified as key mediators of exercise-induced endocrine adaptations in multiple tissues. Our systems genetics framework provides an unprecedented atlas of inter-organ communication significantly remodeled by endurance exercise, serving as a valuable resource for novel exerkine discovery.

PDFDOICodeData
Full-text digest ↓

02 /Specification & Classification

Click any tag to browse matching papers
Citations
10
FWCI
6.43
Field percentile
top 3%
OpenAlex Topics

In-depth summary

✦ Multi-agent verified full-text synthesis

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (4 May 2025) · CC BY 4.0

Correlation-based endocrine network analysis of MoTrPAC rat multi-tissue omics shows 8-week endurance training reshapes inter-organ secretory crosstalk, centering subcutaneous adipose, extracellular matrix factors, and Wnt regulators.

Exerkines carry exercise benefits between organs, but methodological limits have left unclear how strongly and how selectively training-induced secreted factors act on target tissues, and how chronic training rewires the inter-organ endocrine network. Applying cross-tissue correlation frameworks (QENIE, GD-CAT) to MoTrPAC rat multi-omics yields a training-remodeled endocrine atlas for discovering candidate exerkines.

Key results

  1. ShownSecretome scores and correlation-based pathway inference recovered known exercise-modulated endocrine axes (adipose leptin to hypothalamus, muscle IL15 to adipose, adipose TGF-β2), with stronger links after 8 weeks of training than in sedentary rats.

    Lep (scWAT-to-hypothalamus) Ssec rose progressively, with the largest difference TR8W vs CON; gastrocnemius Il15 Ssec toward scWAT was over four-fold higher in TR8W, with scWAT lipid catabolism/oxidation enriched only in TR8W (adjusted p<0.05); scWAT autocrine TGF-β2 protein Ssec increased with training. · Application of QENIE and GD-CAT

  2. ShownEndurance training broadly changed inter-organ crosstalk strength. Subcutaneous white adipose tissue (scWAT) became the dominant origin tissue, and adipose-to-muscle signaling was the connection most altered by training.

    Mean Ssec differed between TR8W and CON in 187 of 256 gene-to-gene and 32 of 49 protein-to-protein origin-target pairs (adjusted p<0.05); Ssec ranks differed in 186 gene pairs and 31 of 49 protein pairs; scWAT-to-vastus lateralis had the top significance and effect size; scWAT replaced vena cava as top origin by median Ssec at TR4W and TR8W. · Endurance training alters Ssec ranks and…

  3. ShownExtracellular matrix remodeling factors were prominent among globally expressed secretory transcripts with the largest training-associated rank shifts, and individual factors showed large rank gains specific to particular tissue pairs.

    Of 524 secretory transcripts expressed in all 16 tissues, the top 20 rank shifters included Adamts3, Adamtsl4, Col12a1, Endou and Loxl2 (each average Ssec p<0.05, TR8W vs CON); Gdf10 (scWAT-to-hypothalamus) rose from 880th to 6th, and Grn protein (gastrocnemius-to-heart) from 274th to 3rd. · Global and tissue-specific Ssec changes by…

  4. ShownSecreted Wnt pathway regulators shifted with training in ways specific to origin tissue, target tissue and direction: positive regulators gained endocrine weight (strongly toward scWAT), while inhibitors such as Nog, Mdk and Wif1 fell in rank.

    From 13 positive, 19 negative and 8 dual secretory Wnt regulators (extracellular score >4), positive-regulator Ssec toward scWAT rose most from BAT, heart and adrenal gland; all top 10 Rspo1 pairs were higher in TR8W; nine of the top 10 Wnt3a connections came from vena cava; Wif1 had the largest negative rank shift in lung-to-cortex proteins. · Global and tissue-specific Ssec changes by…; Secretory Wnt factors as central mediators of…; Curation of secretory Wnt signaling lists

  5. SuggestedEndocrine Wnt links that strengthened with training correlated with target pathways (muscle mitochondrial metabolism, hypothalamic synaptic activity) and differed from Wnt associations within the brain, implying the brain effects are not simply secondary to local Wnt signaling.

    scWAT Sfrp4 fell with training (FDR 9.6e-8) yet its vastus lateralis Ssec rank rose, tracking muscle mitochondrial metabolism in TR8W; scWAT Wnt4 and Ccn4 linked to hypothalamic synaptic and mitochondrial terms (adjusted p<0.05), whereas hypothalamus-derived Ccn4 dropped from 8th to 615th in intra-hypothalamic rank. · Global and tissue-specific Ssec changes by…

Methods, data and reproducibility

Per group, biweight midcorrelations between origin- and target-tissue features, restricted to UniProt-annotated secreted origin features, gave secretome scores (Quantitative Endocrine Network Interaction Estimation, QENIE). GD-CAT ran fgsea on target genes ranked by bicor, and Wilcoxon signed-rank tests compared 8-week trained (TR8W) vs sedentary (CON) rats.

  • MoTrPAC (Molecular Transducers of Physical Activity Consortium) endurance cohort: treadmill training for 1, 2, 4 or 8 weeks plus sedentary controls; transcriptomics of 16 tissues · rat (Fischer 344, six-month-old males and females) · n = n=10 per group; 50 samples per tissue
  • MoTrPAC rat proteomics from 7 of the 16 tissues, used for protein-to-protein secretome scoring · rat (Fischer 344) · n = not stated separately for proteomics (7 tissues)
  • UniProt rat secretome database used to filter origin-tissue features to known secreted proteins · rat · n = 1321 unique features
  • COMPARTMENTS rat subcellular localization scores (extracellular region, space, exosome, vesicle; 0-5 scale) · rat
  • Gene Ontology gene sets for positive and negative regulation of Wnt signaling, filtered to QENIE secretory genes with extracellular score >4 · n = 13 positive, 19 negative, 8 dual regulators · GO:0030177; GO:0030178
Methods & tools
QENIE (Quantitative Endocrine Network Interaction Estimation), GD-CAT (Gene-Derived Correlations Across Tissues), Biweight midcorrelation (bicor), WGCNA (weighted gene co-expression network analysis), fgsea (Fast Gene Set Enrichment Analysis, R), TMsig R package, hypeR (overrepresentation analysis), MotrpacRatTraining6moData R package, UniProt rat secretome database, COMPARTMENTS subcellular localization database, Gene Ontology, Wilcoxon signed-rank test, paired t-test, Benjamini-Hochberg correction
Code
https://github.com/ahnchi/MoTrPAC-PASS1B-QENIE
Data & accessions
https://motrpac-data.org/https://motrpac-data.org/searchhttps://github.com/MoTrPAC/MotrpacRatTraining6moDatahttps://github.com/ahnchi/MoTrPAC-PASS1B-QENIE
Software & environment
R (WGCNA, fgsea, TMsig, hypeR, MotrpacRatTraining6moData); no package versions or containers reported.
Reproducibility gaps
Software versions and compute resources are not reported; per-tissue WGCNA soft-thresholding powers are not listed in the text.
Limitations (6)
  • The analysis is correlational: it cannot prove causality or separate direct endocrine signaling from parallel exercise adaptations arising independently in origin and target tissues, so candidate factors need mechanistic follow-up. · stated by the authors
  • Rats within each group were relatively homogeneous, which may restrict biological variability; for instance, IL-6 was detected only in spleen and scWAT, preventing evaluation of its systemic endocrine role. · stated by the authors
  • Untrained controls exist only at the 8-week time point, so Ssec trends across 1 to 4 weeks of training lack matched sedentary comparators and should be interpreted cautiously. · stated by the authors
  • Whether the extracellular Wnt factors implicated in peripheral-to-brain signaling can cross the blood-brain barrier is unknown; exosomal secretion is offered as a possible route but not tested. · stated by the authors
  • Correlations use n=10 rats per group that include both males and females; no sex adjustment is described for QENIE, so sex-dimorphic expression shared across tissues could inflate cross-tissue correlations and Ssec. · noted in this summary
  • Large rank shifts between TR8W and CON (e.g., Gdf10 from 880th to 6th) rest on single correlation estimates from 10 animals per group, with no permutation or bootstrap assessment of rank stability reported. · noted in this summary

Who should read:Read this if you study exercise physiology or exerkines and want a rat multi-tissue atlas of training-remodeled secretory crosstalk, or plan to apply QENIE/GD-CAT cross-tissue correlation methods to your own data.

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISGroup '22Olsen '25Taha '24Wei '22Vamvini '23Ahn '26
68% semantic matchMolecular Biology
MoTrPAC Study Group, David Amar, Nicole R. Gay+97 more·bioRxiv·Sep 21, 2022· 19 citations