Irisin ameliorates obesity and insulin resistance via adipose tissue IL-33 and regulatory T cells.
The main result did not reproduce in this reproduction attempt. Where our recomputation produced values that differ from the published ones, those discrepancies are listed below. This is a single automated attempt — not peer review and not a finding of error or misconduct — and differences can also arise from data access, undocumented parameters or the computing environment. The verdict can be contested via “report an error”.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓Same input data as the authors
- 🟡Reported values were only indirectly comparable
- 🟡A deviation arose in the data or preprocessing
- 🟡A deviation was attributed to the published material
- 🟡Reported values were not (fully) derivable from the shared data
- 🟡The deviation was non-trivial in magnitude
- 🟡The central claim did not (fully) hold under reproduction
- 🟡Overall, the reproduction showed a material discrepancy
A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.
▸Reproduction agent’s raw note
Reproduced the two pipeline-derived RNA-seq analyses underlying PMID 41933175's irisin/IL-33 findings from raw/deposited data on the account holder's own SLURM account («our HPC»/Uni Hamburg, governed portal lane). For the mSC dataset (GSE282942), an independently-run DESeq2 analysis on the GEO-deposited raw count table reproduced the paper's Fig. 3g claims directionally and statistically for all 10 named marker genes (Il33 ~12.4x up vs. paper's stated ~10x; Il6/Fgf23/Ccl2/Ccl7/Cxcl5 up; Fabp4/Adipoq/Cd36 down past both significance thresholds; Pparg down and significant but below the |log2FC|>=1 magnitude threshold) -- graded within-tol/partial. The 75%-of-195-gene adipogenesis-signature statistic could not be tested because that reference gene list is not available anywhere in the retrievable paper text or source data; this was left as explicitly untestable rather than fabricated. For the whole-tissue dataset (GSE283234), all 24 samples were downloaded from SRA and independently realigned/quantified from FASTQ (Hisat2 -> featureCounts -> DESeq2, matching the paper's exact stated padj<=0.05 & |log2FC|>=1 threshold); the eWAT_HFD KEGG enrichment correctly shows no thermogenesis pathway (consistent with the paper), but the iWAT_HFD contrast -- the tissue where the paper claims thermogenesis IS enriched -- yielded essentially no significantly upregulated genes in our pipeline, so that specific positive claim could not be reproduced (graded mismatch, with the caveat that the authors' exact alignment/quantification pipeline is not fully specified in the retrievable methods). Also flagged: the paper's Data Availability statement describes GSE283234 as HFD-only, but the deposited data actually includes a full CHW arm as well.
These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.
✎ I am an author of this paper
Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.
Provenance — full disclosure
When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.
- Reproduced
- 2026-07-31
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-07-31no human curator yet
- Last updated
- 2026-07-31
Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.
Deep full-text extraction
Model: opusDoes chronic elevation of the exercise-induced myokine irisin ameliorate diet-induced obesity and insulin resistance through immunomodulation of adipose tissue, and if so, what cellular and molecular pathways mediate these effects? The authors test whether irisin acts via an IL-33–ST2 axis involving mesenchymal stromal cells and ST2+ regulatory T cells in white adipose tissue.
- ★ Irisin modulates adipose tissue inflammation by increasing IL-33 production and preserving ST2+ regulatory T cells in white adipose tissues, thereby improving obesity and glucose intolerance. finding
- ★ Chronic AAV-mediated irisin elevation reduces body weight and fat mass and improves glucose tolerance and insulin sensitivity in HFD-fed male mice without reduced food intake, loss of lean mass, or changes in physical activity. finding
- ★ Irisin preserves VAT (eWAT) ST2+ Treg cells that normally contract after long-term HFD feeding, and reduces CD4+ and CD8+ effector T cell accumulation. finding
- ★ Irisin's effect on eWAT Treg cells is indirect: it acts primarily by increasing IL-33-producing mesenchymal stromal cells and IL-33 levels in VAT, as irisin directly induces IL-33 expression in cultured VAT mesenchymal stromal cells. mechanism
- ★ Irisin-driven changes in VAT IL-33 dynamics are accompanied by IL-33-dependent upregulation of thermogenic gene expression in subcutaneous adipose tissue (iWAT), including UCP1 and futile creatine cycle genes. mechanism
- Irisin's immunomodulatory effects are depot-specific, occurring in eWAT but not in iWAT or BAT at either 8 or 18 weeks of HFD feeding, and are not seen in normal chow-fed mice. finding
- Irisin reduces crown-like structures, a marker of inflammation and adipocyte death, in eWAT but not iWAT of HFD-fed mice. finding
- AAV8-mediated hepatic expression of FLAG-tagged irisin provides a model of stable, chronic (~20-fold) elevation of plasma irisin for studying long-term irisin action. method
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| AAV8-mediated gene delivery with longitudinal metabolic phenotyping (body weight, food intake) | 4-week-old male C57BL/6J mice, liver-directed expression; HFD or normal chow feeding | AAV8-FLAG-irisin overexpression versus AAV8-GFP control; high-fat diet versus normal chow | Plasma irisin levels, body weight over time, food consumption | AAV8 vector, tail-vein injection |
| Body composition analysis by magnetic resonance imaging | HFD-fed male C57BL/6J mice | AAV-irisin versus AAV-GFP | Fat mass and lean mass | MRI |
| Intraperitoneal glucose tolerance test and insulin tolerance test | HFD-fed male C57BL/6J mice (8 and 18 weeks of HFD) | AAV-irisin versus AAV-GFP | Blood glucose excursion over time | — |
| Indirect calorimetry / metabolic cage analysis | HFD-fed male C57BL/6J mice | AAV-irisin versus AAV-GFP | Energy expenditure and physical activity | CLAMS (comprehensive lab animal monitoring system) |
| Whole-tissue bulk RNA sequencing with pathway analysis | BAT, inguinal white adipose tissue (iWAT) and epididymal white adipose tissue (eWAT) from mice fed HFD for 18 weeks | AAV-irisin versus AAV-GFP | Differential gene expression and enriched pathways (fold change > 2, adjusted P < 0.05) | — |
| RT–qPCR of thermogenic and lipolysis gene programmes | iWAT, eWAT and BAT from HFD-fed and normal chow-fed mice | AAV-irisin versus AAV-GFP | Expression of Ucp1, Ckb, Alpl (TNAP) and lipolysis genes | — |
| Multiparameter flow cytometry (13-parameter T cell panel; t-SNE clustering) of adaptive and innate immunocytes | eWAT, iWAT and BAT from male C57BL/6J mice fed normal chow or HFD for 8 or 18 weeks | AAV-irisin versus AAV-GFP; HFD versus normal chow | Frequencies/numbers of CD45+ cells, CD64+ and CD11c-hi macrophages, total/CD8+/Foxp3−CD4+ T cells, total and ST2+ Foxp3+CD4+ Treg cells, IL-33+ mesenchymal stromal cells | — |
| Ex vivo cell culture with recombinant irisin plus HSP90α, including conditioned-medium transfer with IL-33 blockade, plus ELISA for IL-33 and histological crown-like structure quantification | Cytofluorimetrically sorted eWAT Treg cells from 12-week-old male Foxp3.Thy1.1 mice; IL-33(EGFP)+ and IL-33(EGFP)− eWAT mesenchymal stromal cells from 12-week-old male Il33.Egfp mice; eWAT and iWAT tissue lysates and plasma from AAV-treated mice | Irisin/HSP90α (doses up to 10 nM) versus PBS for 3 days; anti-IL-33 antibody (2 μg ml−1) versus control IgG on conditioned medium | Total Treg numbers and ST2+ Treg fraction; number and mean fluorescence intensity of IL-33(EGFP)+ mSCs; IL-33 concentration in eWAT lysates and plasma; crown-like structure counts | Flow cytometry; ELISA |
- ▼ AAV-irisin mice had lower body weight than AAV-GFP controls after 8 weeks of HFD feeding, with no significant change in food intake
- ▼ Lower body weight in the irisin group was largely due to reduced fat mass with no loss of lean mass; subcutaneous depots accounted for most of the reduction at this stage
- ▲ Irisin improved glucose tolerance (GTT) and insulin sensitivity (ITT) in HFD-fed mice at the late time point, with no differences at 8 weeks of HFD or in normal chow-fed mice
- ▲ Irisin elevation increased Ucp1, Ckb and Alpl expression specifically in iWAT, with no changes in eWAT, BAT or chow-fed mice; RNA-seq pathway analysis showed thermogenesis upregulated exclusively in iWAT Ucp1 ~1.9-fold; Ckb 3.3-fold; Alpl 2.4-fold
- ▲ Long-term HFD strongly reduced total eWAT Treg cells and particularly the ST2+ subtype; irisin strongly attenuated these reductions, almost totally for ST2+ Treg cells
- ▼ Irisin prevented HFD-induced increases in CD4+ and CD8+ effector T cells and decreased the CD8+ T cell fraction under both feeding regimes, with enrichment of ST2+ Treg cells and a trending increase in naive CD8+ T cells by t-SNE analysis
- ▼ Irisin reduced crown-like structure counts in eWAT at both 8 and 18 weeks of HFD, but not in iWAT at either time point 40% reduction at 18 weeks
- – Culturing sorted eWAT Treg cells directly with irisin/HSP90α did not change total Treg numbers or the ST2+ Treg fraction, whereas irisin/HSP90α increased the number and mean fluorescence intensity of IL-33(EGFP)+ mSCs, and irisin increased IL-33+ mSC numbers and IL-33 protein in eWAT lysates and plasma in vivo
- fold_change approximately 20-fold above basal levels (Plasma irisin concentration after AAV injection, stable from 4 weeks to at least 6 months (prior study))
- fold_change ~1.9-fold (Ucp1), 3.3-fold (Ckb), 2.4-fold (Alpl/TNAP) (Irisin-induced increases in thermogenic gene expression in iWAT after 18 weeks HFD (RT–qPCR))
- fold_change twofold increase (Total CD45+ immunocytes in eWAT of control-AAV mice on HFD versus normal chow at 8 weeks; increase absent in AAV-irisin mice)
- fold_change 40% reduction (Irisin-mediated reduction of eWAT crown-like structure counts at 18 weeks of HFD)
- pvalue **P = 0.0047 (MRI body composition difference between AAV-irisin and AAV-GFP (n = 8, 8))
- pvalue **P = 0.0011, ****P < 0.0001, ****P < 0.0001 (GTT); **P = 0.0018, ****P < 0.0001, ***P = 0.0003 (ITT) (Two-way ANOVA for intraperitoneal GTT (n = 8, 5) and ITT (n = 5, 5))
- pvalue *P = 0.0299, *P = 0.0335, *P = 0.0207, ***P = 0.0002 (One-way ANOVA for total and ST2+ eWAT-Treg percentages across NC/HFD and GFP/irisin groups (n = 3, 3, 8, 8))
- pvalue *P = 0.0127 (IL-33+ mSC numbers); **P = 0.0019, ****P < 0.0001, ****P < 0.0001 (eWAT IL-33 ELISA); *P = 0.0342 (plasma IL-33 ELISA); **P = 0.0019, ***P = 0.0006, ****P < 0.0001 (IL-33(EGFP)+ mSC culture) (IL-33 measurements in vivo (n = 3, 3, 6, 7 and n = 3, 3, 6, 6; plasma n = 6, 9) and in cultured mSCs (n = 4))
Statistical methods review
Model: sonnetA neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.
The study used a diet-induced obesity mouse model (AAV-irisin vs AAV-GFP control, fed high-fat diet or normal chow) with group comparisons analyzed by unpaired two-sided Student's t-tests for two-group endpoints, one-way ANOVA for comparisons across three or more groups (e.g., feeding/treatment combinations), and two-way ANOVA for glucose and insulin tolerance time-course data. Whole-tissue RNA-seq pathway analysis used a fold-change and adjusted P-value threshold to identify differentially regulated pathways. Results are reported per figure as mean ± s.d. with exact P values given in figure legends, and the specific test used is stated for each panel; explicit discussion of test assumptions (e.g., normality, variance homogeneity) is not present in the provided text.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Unpaired two-sided Student's t-test | Fig. 1b body weight over time (AAV-irisin vs AAV-GFP) | n=15,15 (n=10,10 after week 17) | not stated |
| Unpaired two-sided Student's t-test | Fig. 1c food consumption | n=10,10 | not stated |
| Unpaired two-sided Student's t-test | Fig. 1d body composition by MRI | n=8,8 | not stated |
| Two-way ANOVA | Fig. 1e intraperitoneal glucose tolerance test (GTT) | n=8,5 | not stated |
| Two-way ANOVA | Fig. 1f intraperitoneal insulin tolerance test (ITT) | n=5,5 | not stated |
| One-way ANOVA | Fig. 2b–d flow cytometry of macrophage/T cell/Treg populations in eWAT across 4 conditions | n=3,3,8,8 | not stated |
| One-way ANOVA | Fig. 3b,c,e IL-33+ mSC counts and IL-33 ELISA measurements | n=3,3,6,7 (Fig. 3b); n=3,3,6,6 (Fig. 3c); n=4 (Fig. 3e) | not stated |
| Unpaired two-sided Student's t-test | Fig. 3d plasma IL-33 ELISA | n=6,9 | not stated |
| Fold-change and adjusted P-value threshold (differential/pathway analysis) | RNA-seq of BAT, iWAT, eWAT (fold change > 2; adjusted P value < 0.05) | — | not stated |
-
Many pairwise comparisons (t-tests) and multi-group comparisons (ANOVAs) are reported across numerous figure panels without a stated correction for the overall number of tests performed in the study.↳ Could also: A multiple-comparison correction (e.g., Holm-Šídák, Benjamini-Hochberg) applied across the full family of tests within a figure or dataset — This would jointly control the family-wise error rate or false discovery rate across the many statistical comparisons made within the study, which is a standard consideration when numerous tests are reported together.
-
Glucose and insulin tolerance time courses (Fig. 1e,f) were analyzed with two-way ANOVA treating time and treatment as factors.↳ Could also: A repeated-measures or linear mixed-effects model accounting for within-animal correlation across timepoints — This approach explicitly models the fact that the same animals are measured repeatedly over time, and can also accommodate any missing timepoints without requiring complete data at every measurement.
-
Multi-group comparisons (e.g., Fig. 2b–d, 3b,c,e) were analyzed by one-way ANOVA.↳ Could also: A named post-hoc multiple comparison test (e.g., Tukey HSD or Šídák) following the omnibus ANOVA — This would identify which specific pairs of groups differ, in addition to the overall significance of group differences indicated by the ANOVA.
-
Results are summarized as mean ± s.d. throughout the figures.↳ Could also: Reporting mean with 95% confidence intervals alongside or instead of s.d. — A confidence interval directly conveys the precision of the estimated group difference, which some readers find more informative for judging the reliability of an effect, particularly with modest group sizes.
-
RNA-seq differential expression/pathway analysis used a fold-change > 2 and adjusted P value < 0.05 cutoff, without naming the specific statistical software or model.↳ Could also: A named differential expression framework such as DESeq2 or edgeR with an explicitly stated FDR (Benjamini-Hochberg) correction — Naming the specific tool, version and correction method increases transparency and reproducibility of the multiple-testing adjustment applied across thousands of simultaneously tested genes.
-
Several in vitro experiments (e.g., Fig. 3a,e) used relatively small group sizes (n=3–4) analyzed with parametric tests (t-test/ANOVA).↳ Could also: A nonparametric test such as Mann-Whitney U or Kruskal-Wallis as a robustness check — With small sample sizes, normality is harder to verify, and a nonparametric alternative can serve as a complementary check that does not rely on distributional assumptions.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
Fig. 3g reproduces well from identical data. DESeq2 on the GEO-deposited raw count table (GSE282942) confirms all six named upregulated genes (Il33 log2FC=3.638, ~12.4x vs the paper's 'tenfold', padj=5.17e-72; Il6, Fgf23, Ccl2, Ccl7, Cxcl5) and three of four downregulated ones past the paper's own dual threshold; only Pparg (-0.857, padj=5.51e-09) misses the |log2FC|>=1 half. Two gaps are on the authors' side of reporting, not fabrication. The '147 of 195 adipogenesis-up genes (75%)' statistic is untestable because that reference gene list exists nowhere in the paper or source data, and the Data Availability statement misdescribes GSE283234's scope (HFD-only claimed, CHW+HFD deposited, plus undocumented G1-G8 FPKM columns). The one hard contradiction is confounded. Our independent FASTQ->Hisat2->featureCounts->DESeq2->enrichKEGG pipeline found 0 upregulated genes in iWAT_HFD irisin-vs-gfp at the paper's stated FC>2/FDR<0.05, so the Extended Data Fig. 1k thermogenesis enrichment did not reproduce — but GEO deposits only FPKM and the paper names no aligner or quantifier, so a preprocessing-chain difference cannot be separated from genuine non-replication. Net: core mSC/IL-33 mechanism confirmed, iWAT thermogenic arm unconfirmed, severity moderate.
Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.
Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.
🚩 Report an error in this record
Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.
Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.