henhaobio
Sign in

bioRxivCell BiologyPreprint

Loss of TAFAZZIN leads to perturbation of amino acid metabolism and reduction of collagen synthesis

Abu Mohammed RamimiD, Tyler Ralph-EppsiD, Linh VoiD, Hyejeong JangiD, Janaka S. S. LiyanageiD, Kaitlin LowraniD, Miriam L. GreenbergiD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

Barth syndrome is a life-threatening genetic disorder caused by mutations in the TAFAZZIN (TAZ) gene, which disrupt remodeling of cardiolipin in mitochondria. The disease is associated with cardiac and skeletal myopathy, neutropenia, fatigue, and metabolic dysfunction. Previous studies showed that loss of TAZ decreases pyruvate dehydrogenase activity, reduces glucose flux into the TCA cycle, and impairs fatty acid metabolism. To test the hypothesis that amino acid (AA) metabolism may be altered to compensate for these deficiencies, we characterized AA metabolism in TAZ-deficient mouse myoblasts (TAZ-KO). Levels of branched-chain amino acids (BCAAs) were reduced, while proline levels were increased in TAZ-KO cells. Levels of proline dehydrogenase and glutamate dehydrogenase, which convert proline to TCA cycle intermediates, were increased. 13C5-proline isotope tracing demonstrated elevated conversion of proline into glutamate and TCA cycle intermediates. SILAC analysis using [U-13C6, 15N2]-Lys and [U-13C6]-Arg revealed decreased synthesis of collagen and proteins associated with extracellular matrix (ECM). Gene expression and protein analyses revealed reduced collagen expression, lower total collagen content, decreased collagen crosslinking enzymes, decreased proline hydroxylation and reduced synthesis of new collagen and cell-adhesion proteins. SILAC analysis using [U-13C6, 15N2]-proline also showed diminished incorporation of proline into newly synthesized ECM proteins. Together, our findings reveal that loss of TAZ leads to increased proline catabolism to the TCA cycle, decreased incorporation of proline into collagen, and impaired collagen synthesis and ECM remodeling.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (20 Jul 2026) · CC BY 4.0

TAFAZZIN-knockout C2C12 myoblasts show elevated proline catabolism into the TCA cycle alongside reduced collagen content, proline hydroxylation, crosslinking enzyme LOX, and proline incorporation into ECM proteins.

Barth syndrome arises from TAZ mutations that disrupt cardiolipin remodeling; prior work showed TAZ loss impairs pyruvate dehydrogenase, glucose flux into the TCA cycle, and fatty acid oxidation, leaving open how cells sustain energy metabolism. This work shows amino acid metabolism is rewired in TAZ-deficient myoblasts, with proline diverted toward TCA anaplerosis and away from collagen and extracellular matrix (ECM) synthesis, a candidate link to skeletal myopathy.

Key results

  1. ShownProline is catabolized into the TCA cycle more in TAZ-KO myoblasts. Protein levels of PRODH and GDH (glutamate dehydrogenase) rise, and labeled proline enriches glutamate and TCA intermediates more strongly, consistent with compensatory anaplerosis.

    Western blot (n = 3, p < 0.05): PRODH and GDH elevated in TAZ-KO; 8-h [U-13C5]proline tracing (n = 4): M5 glutamate, M5 alpha-ketoglutarate and M4 succinate, fumarate, malate and citrate fractional enrichment all higher vs WT (p < 0.01 to p < 0.0001). · 3.5. TAZ-KO cells exhibit increased flux of…; Figure legends

  2. ShownCollagen production is reduced in TAZ-KO cells. Total collagen is lower, Col1a1 protein is undetectable, Type I and VI collagen transcripts are down, the labeled hydroxyproline pool is smaller even though P4H is unchanged, and the LOX crosslinking enzyme is decreased.

    Hydroxyproline assay (N = 4, 3 technical replicates each) showed lower collagen content; 13C5-hydroxyproline pool reduced after 8-h proline labeling (n = 4, p < 0.001); LOX protein reduced by immunoblot (N = 3); collagen I/VI downregulation from previously reported RNA-seq FPKM values. · 2.4. Collagen assay; 3.6. Collagen synthesis is impaired in TAZ-KO…; Figure legends

  3. SuggestedStable isotope labeling (SILAC) indicates less new synthesis of collagen and ECM-associated proteins in TAZ-KO cells, including Type IV collagen and integrin alpha 6, and less proline incorporated into ECM remodeling proteins.

    Lys/Arg SILAC (24 h): newly made Type IV collagen and Itga6 lower. 13C5-proline SILAC (about 80 h, n = 4 per group): 6,767 proteins identified, 1,380 analyzed; 72 up, 126 down (|FC| > 1.5, unadjusted p < 0.05), including 11 ECM proteins such as FN1, SPARC, HSPG2, SERPINH1. · 2.6.2. Growth condition; 2.8. Statistical analysis; 3.6. Collagen synthesis is impaired in TAZ-KO…; 3.7. Impaired proline incorporation into…

  4. ShownTAZ-KO myoblasts have lower intracellular branched-chain amino acids (BCAAs) and higher proline, which altered transporter levels do not explain. Glucose-derived carbon also enters most nonessential amino acids less efficiently.

    LC-MS (N = 5): leucine 1.77-fold and isoleucine/valine 2-fold lower, proline 1.8-fold higher vs WT; LAT1, SLC25A44, SNAT2, SLC6A20 protein unchanged; [U-13C]glucose 4-h tracing (n = 4): M3 serine and M2/M4 glutamate, proline, aspartate reduced, M3 alanine increased. · 3.1. TAZ-KO cells exhibit altered levels of…; 3.2. BCAA transporter expression or uptake is…; 3.4. Levels of proline transporters are…; Figure legends

  5. ShownUnder high glucose, labeled BCAAs reach branched-chain keto acids (BCKAs) but not TCA intermediates in either genotype. TAZ-KO cells show higher inhibitory BCKDH E1alpha phosphorylation and export fewer labeled BCKAs, with no increase in branched-chain fatty acid labeling.

    BCKDHE1alpha mRNA and protein and phospho-BCKDHE1alpha elevated, BCAT2 protein unchanged (western n = 3; p < 0.05 or p < 0.01); 4-h [U-13C]BCAA tracing in 25 mM glucose (n = 4): KIV and 3-OMV labeling unchanged, no TCA labeling, lower BCKA enrichment in TAZ-KO media; BCFA incorporation unchanged. · 2.7. Tracing of BCAA-Derived Branched-Chain…; 3.3. BCAA catabolic enzymes are upregulated in…; 4. Discussion; Figure legends

Methods, data and reproducibility

Wild-type and TAZ-knockout C2C12 myoblasts were compared with several methods: LC-MS amino acid profiling; stable isotope tracing of 13C-glucose, 13C-BCAAs and 13C5-proline; qPCR and western blots of metabolic enzymes and transporters; a hydroxyproline collagen assay; and SILAC proteomics with heavy Lys/Arg or proline.

  • Wild-type and TAZ-KO C2C12 myoblasts (previously generated knockout); steady-state intracellular amino acid profiling by LC-MS · mouse · n = N = 5
  • [U-13C]glucose and [U-13C]BCAA (leucine, isoleucine, valine) tracing in cells and media, 4-h labeling · mouse · n = N = 4 biological replicates
  • [U-13C5]proline tracing into glutamate, TCA intermediates and hydroxyproline, 8-h labeling · mouse · n = n = 4
  • SILAC proteomics: heavy Lys/Arg (about 24 h) for global synthesis and 13C5-proline (about 80 h) for collagen/ECM synthesis · mouse · n = N = 4 per group; 6,767 proteins identified, 1,380 retained (proline SILAC) · PRIDE PXD078912 (10.6019/PXD078912)
  • Hydroxyproline-based total collagen assay · mouse · n = N = 4 per group, 3 technical replicates each
  • Western blots and qPCR of BCAA, proline and collagen pathway enzymes and transporters · mouse · n = at least three or four biological replicates
  • Previously reported RNA-seq of undifferentiated WT vs TAZ-KO C2C12 myoblasts (collagen type I and VI FPKM) · mouse
  • BCAA-to-branched-chain fatty acid tracing (isomyristic, isopalmitic, isostearic acids) at Wayne State Lipidomics Core · mouse
Methods & tools
LC-MS: Vanquish LC with Orbitrap Exploris 240 (ion-paired, negative mode), Easy-nLC 1200 with Orbitrap Fusion (data-dependent LC-MS/MS), Proteome Discoverer 2.4 with Sequest HT, Mouse protein database UP000000589, SILAC ([U-13C6,15N2]-Lys, [U-13C6]-Arg, 13C5-proline), Hydroxyproline Assay Kit (MAK463, Sigma-Aldrich), qPCR on QuantStudio 3 (2^-ΔΔCT method), Western blot with iBright densitometry, Random forest imputation, iPathwayGuide (gene ontology analysis), Two-sample t-tests
Reagents & environment
Proteome Discoverer 2.4 with Sequest HT; mouse protein database UP000000589 (downloaded March 30, 2021); iBright and iPathwayGuide software used without stated versions.
Reproducibility gaps
No analysis code is shared. Only the proteomics data are deposited (PRIDE PXD078912); no repository is stated for the LC-MS metabolomics and isotope tracing data or the reused RNA-seq dataset.
Limitations (4)
  • It remains unresolved where the BCKA pool goes when TAZ-KO cells do not export it. The authors cannot exclude subtle differences in BCAA export, or more incorporation into new protein, as causes of the low intracellular BCAA levels. · stated by the authors
  • The source that keeps serine and glycine pools steady despite lower glucose labeling was not determined (uptake is proposed). A direct link from PDH loss to alanine synthesis, and greater alanine use or export, still need validation. · stated by the authors
  • All work uses one knockout line of undifferentiated C2C12 myoblasts, with no TAZ re-expression rescue and no differentiated myotubes, tissue or patient cells. This leaves the link from the collagen/ECM phenotype to myopathy untested. · noted in this summary
  • Proteomic hits rely on unadjusted p < 0.05 with n = 4 per group. Reporting is also inconsistent: the Fig. 4 legend cites adjusted p, and asparagine glucose labeling is called reduced in Results but unchanged in Discussion. · noted in this summary

Read this if you study Barth syndrome or cardiolipin-linked mitochondrial disease and want isotope tracing and SILAC evidence connecting TAZ loss to proline anaplerosis and impaired collagen/ECM synthesis in myoblasts.

Citations
1

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
THISCarney '24Carlein '25Liang '24Matias '25Senoo '26Cole '22
67% semantic matchCell Biology
Olivia Sniezek Carney, Kodi William Harris, Yvonne Wohlfarter+7 more·Human Molecular Genetics·Oct 29, 2024· 9 citations