henhao
Sign in
Cell BiologybioRxiv
Preprint

A proteomic and phosphoproteomic comparison of mouse spermatogonial stem cells and progenitor spermatogonia

David Anthony Skerrett-ByrneiD, Brett NixoniD, Adrián Sanz‐MorenoiD, Filip DámekiD, Patricia da Silva‐ButtkusiD, Raffaele TeperinoiD, Gailus-Durner, Helmut FuchsiD, Martin Hrabě de AngelisiD, Jon M. OatleyiD, Connor Cason, Ilana R. Bernstein, Tessa LordiD

Abstract

In this manuscript, we used an Id4-eGfp mouse line to produce the first proteomic and phosphoproteomic database for mouse spermatogonial stem cells (SSCs) and progenitor spermatogonia. Our proteomic analyses identified 8,464 proteins in spermatogonia, superseding the depth of previously published spermatogonia datasets by >2000 proteins. While the comparison of SSCs and progenitors revealed few unique proteins (18 and 3, respectively), 532 proteins exhibited significantly different abundance (FC {+/-} 1.5, p-value [≤] 0.05) between these sub-populations. Interestingly, in overlaying the proteome with transcriptomic data, correlation was poor (R2 = 0.236), re-iterating discordance between transcript and protein abundance in the testis. In our phosphoproteomic analyses, phosphosites were identified in 19.5% of proteins (3,604 total phosphosites). Unique protein phosphorylation was substantially more common than unique protein expression when comparing SSCs and progenitors, with 38 and 191 unique phosphosites identified, respectively, in addition to significant differences in abundance at 60 and 257 phosphosites. In identifying a wave of phosphorylation that accompanies the progenitor transition, we performed predictive analyses to identify three potential master kinases for follow up validation studies (PAK1, BUB1, ABL2). The inhibition of these kinases resulted in a significant reduction in the capacity for progenitor spermatogonia to differentiate, and caused elevated apoptosis and DNA damage. Finally, we explored the testis phenotype of 42 knockout mouse lines for proteins that were either differentially expressed (26) or differentially phosphorylated (15) in our dataset. Of these lines, 21 exhibited a testis phenotype in at least one of two biological replicates observed (severity score [≥] 1, compared to 0.15 for controls), with increased numbers of Sertoli-only tubules being evident in four of these lines. This manuscript provides a comprehensive roadmap to understand the multifaceted layers of regulation over fate decisions in undifferentiated spermatogonia. These data have been provided in an accessible platform via our ShinySpermatogoniaCells app to encourage future progress in the field.

PDFDOICodeData
Full-text digest ↓

02 /Specification & Classification

Click any tag to browse matching papers
Citations
1
OpenAlex Topics

In-depth summary

✦ Multi-agent verified full-text synthesis

Claude Sonnet 4.6 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (16 Jul 2026) · CC BY 4.0

Deep proteome and phosphoproteome of Id4-eGFP-sorted mouse spermatogonial stem and progenitor cells reveal 8,464 proteins, poor RNA-protein concordance, and phosphorylation-driven kinase candidates tested functionally.

Transcript levels predict protein abundance poorly in testis, and to the authors' knowledge no phosphoproteome existed for stem spermatogonia, a rare population that is hard to purify. This work supplies protein and phosphosite maps for the stem-to-progenitor transition from sorted Id4-eGFP populations.

Key results

  1. ShownThe proteome of sorted stem cells and progenitors is very deep and nearly identical in composition, yet hundreds of proteins differ in abundance between the two populations.

    8,461 proteins in SSCs and 8,446 in progenitors (8,464 total); 99.8% overlap; 18 SSC-unique and 3 progenitor-unique proteins; 532 proteins differ at FC ±1.5, p≤0.05 (274 up, 258 down in progenitors); n=4 biological replicates. · Abstract; Generating a proteomic database for SSC and…; Integrative Proteomic and Transcriptomic…; Statistical Analysis

  2. ShownProtein and transcript changes between SSCs and progenitors correlate weakly, consistent with marked discordance between RNA and protein abundance in spermatogonia.

    Pearson R^2 = 0.236 across 7,660 overlapping proteins vs a previously published RNA-seq dataset of the same populations; 198 proteins significant in both, 174 with the same direction of change. · Integrative Proteomic and Transcriptomic…

  3. ShownPhosphorylation differs more sharply than protein presence between the populations, and a pipeline built on these data nominated many candidate kinases.

    3,633 phosphosites in SSCs and 3,790 in progenitors; phosphosites on 19.5% of proteins; 38 vs 191 unique phosphopeptides (SSC vs progenitor); 60 vs 257 over-represented at FC ±1.5, p≤0.05; 328 kinases predicted. · Abstract; Characterization of the phosphoproteomic changes…

  4. ShownIn cultured undifferentiated spermatogonia, inhibiting PAK1, BUB1 and ABL2 lowered differentiation capacity and cell number without altering the SSC:progenitor ratio, while apoptosis and DNA damage markers rose with some inhibitors and the combined Trio.

    Treatments for 24h: cKIT+ cells reduced in all (p<0.05); cell number reduced (p<0.01); TUNEL+ up with Trio (p<0.05); Caspase 3 up with BUB1, ABL2, Trio (p<0.001); γH2AX+ up with BUB1 and Trio (p<0.0001). · Master kinases PAK1, BUB1 and ABL2 protect…; Figure legends

  5. SuggestedKnockouts of genes encoding differentially abundant or phosphorylated proteins frequently showed testis pathology, with more Sertoli-only tubules in four lines, but without statistical testing.

    Of 27 abundance-selected lines 10, and of 16 phosphorylation-selected lines 11, scored ≥1 in at least one replicate (controls 0.15, 82 animals). Sertoli-only tubules: Nucks1 12.9 ± 5.3, Gstt1 7.8 ± 4.3, Cnot6l 5.1 ± 1.5, Rtn4 3.7 ± 1.5 vs wild type 0.3 ± 0.1. · Abstract; Testis phenotypes following knockout of genes…

Methods, data and reproducibility

FACS-sorted ID4-eGFP Bright (SSC) and Dim (progenitor) spermatogonia from P6-8 mice were processed with EasyPhos for label-free LC-MS/MS proteomics and phosphoproteomics, compared at FC ±1.5 and p≤0.05, then integrated with RNA-seq, IPA, kinase prediction, inhibitor assays and knockout histopathology.

  • FACS-sorted ID4-eGFP Bright (SSC) and Dim (progenitor) spermatogonia, proteome and phosphoproteome, label-free LC-MS/MS · mouse (Mus musculus), P6-8 · n = 4 biological replicates per population (litter-pooled) · PXD066114
  • Previously published RNA-seq of ID4-eGFP Bright and Dim spermatogonia, used for proteome-transcriptome comparison · mouse · n = 7,660 overlapping proteins compared
  • UniProt Mus musculus search database for SEQUEST HT · mouse · n = 25,444 sequences
  • Primary cultures of THY1+ undifferentiated spermatogonia treated with PAK1, BUB1, ABL2 or combined (Trio) inhibitors · mouse · n = n=3 to 10 biological replicates depending on assay
  • German Mouse Clinic / IMPC knockout lines, testis and epididymis histopathology at 16 weeks · mouse · n = 42 lines, 2 knockout mice per line; 82 control animals
Methods & tools
Id4-eGfp reporter mouse, BD FACS AriaIIu, EasyPhos, Orbitrap Eclipse nLC-MS/MS, Proteome Discoverer 2.5 (SEQUEST HT, Percolator), Perseus 2.0.7.0, Ingenuity Pathway Analysis, PhosphoSitePlus, KinMap, NVS-PAK1-1, BAY-1816032, Asciminib, R Shiny 1.9.1
Code
https://github.com/DavidSBEire/ShinySpermagoniaCells
Data & accessions
http://proteomecentral.proteomexchange.org/https://reproproteomics.shinyapps.io/ShinySpermatogoniaCells/
Software & environment
Proteome Discoverer 2.5; Perseus 2.0.7.0; Xcalibur 4.2.47; Shiny app built with shiny 1.9.1, R 4.3.3, RStudio 2024.04.1+748; GraphPad Prism 11.0.0.
Reproducibility gaps
Raw MS data are deposited under PRIDE with reviewer-account access in this preprint; additional reanalysis information only from the Lead Contact upon request; the cited RNA-seq dataset has no accession in the text.
Limitations (5)
  • Only postnatal day 6-8 mouse spermatogonia were profiled; adult spermatogonia may differ in expression and metabolome, so findings may not transfer to adult testis. · stated by the authors
  • Kinase roles were inferred from pharmacological inhibitors in primary cultures; conditional knockouts and phosphosite-directed CRISPR mutations would be needed to establish causality in vivo. · stated by the authors
  • Only two biological replicates per knockout line were available, so no statistical comparisons could be made and phenotype trends remain provisional. · stated by the authors
  • Inhibitor specificity and off-target effects are not reported, and retinoic acid in vitro drives differentiation in all spermatogonia, which limits interpretation of the differentiation readout. · noted in this summary
  • Differential thresholds (FC ±1.5, p≤0.05) with n=4 replicates are described without multiple-testing correction, so some reported protein and phosphopeptide changes may be false positives. · noted in this summary

Who should read:Read this if you study spermatogonial stem cell fate, need a proteome and phosphoproteome resource for SSCs vs progenitors, or want evidence of RNA-protein discordance in testis.

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
THISSkerrett-By '25Cheng '20Bush '24Fang '20Liao '21Liao '18
65% semantic matchCell Biology
David Anthony Skerrett-Byrne, Amanda L. Anderson, Raffaele Teperino+9 more·bioRxiv·Sep 18, 2025· 1 citations