A protein called TPT1 has emerged as the leading candidate in a study seeking molecular signs associated with sarcopenia, a condition linked to muscle wasting. Its expression separated sarcopenia samples from controls better than the other genes tested in two human muscle-transcriptomic cohorts, although the result is not yet evidence that TPT1 causes muscle loss or can diagnose the condition in clinical practice.
The strongest result came from an analysis of public human datasets. Four bulk skeletal-muscle transcriptomic datasets were combined into a training cohort of 103 samples, including 66 controls and 37 sarcopenia samples. A separate external cohort contained 40 samples, split evenly between controls and sarcopenia. Researchers used the second group to check whether the pattern held outside the data used to select candidates.
TPT1’s area under the receiver operating characteristic curve, a measure of how well a marker distinguishes two groups, was 0.819 in the training cohort, with a 95% confidence interval of 0.737 to 0.900. In the external cohort, the AUC was 0.753, with a confidence interval of 0.598 to 0.907. Those figures made TPT1 the strongest single-gene discriminator reported in the analysis, but the wider external interval also signals uncertainty around how well the result would generalize.
How TPT1 was selected
The researchers first looked for genes that differed between the sarcopenia and control samples in the training data, then compared them with a curated set of 781 genes linked to PANoptosis. In this study, PANoptosis-related genes were treated as genes connected with inflammatory and programmed-cell-death pathways. The analysis found 608 differentially expressed genes, of which 47 overlapped the PANoptosis-related set. Forty of those overlapping genes were more active in sarcopenia samples and seven were less active.
The overlapping genes were associated with inflammatory and programmed-cell-death signaling, including NF-κB, TNF, NOD-like receptor, cytokine and interleukin pathways. They were also linked with regulated necrosis and changes in extracellular-matrix remodeling, while oxidative phosphorylation and aerobic respiration showed negative enrichment. These are patterns in gene activity, not direct evidence that PANoptosis was being executed in muscle cells.
Three machine-learning methods were then used to narrow the list. LASSO, random forest and XGBoost all selected the same three consensus candidates: TRAP1, TPT1 and NTRK1. TPT1 had the highest single-gene AUC in both cohorts. NTRK1 and TRAP1 had training AUCs of 0.721 and 0.744, and external AUCs of 0.715 and 0.720, respectively.
Aging context, not proof of sarcopenia
A separate single-nucleus RNA-sequencing analysis helped show where TPT1 was active in muscle tissue. It retained 97,154 nuclei from 17 donors, including six young and 11 older donors, and grouped them into 12 transcriptional clusters and nine cell classes. TPT1 was broadly expressed across several muscle, stromal, vascular and immune compartments and was lower in older samples.
That analysis adds age-related context, but it was not a direct comparison of people with clinically defined sarcopenia against controls. The combined PANoptosis module score tended to be higher in macrophages, T/NK cells, endothelial cells and fibro-adipogenic progenitors than in major myonuclear populations, but the analysis did not demonstrate PANoptosis execution.
The study also used laboratory models to see whether the TPT1 pattern appeared alongside muscle-wasting changes. In male mice exposed to D-galactose, lower TPT1 protein accompanied reductions in the muscle-related proteins MyHC, MyoD and myogenin, while the atrophy-associated proteins Atrogin-1 and MuRF-1 increased. The experiment involved 12 eight-week-old mice, six receiving D-galactose and six receiving vehicle, with daily exposure for 10 weeks.
D-galactose-treated C2C12 muscle cells showed a similar pattern: higher p53 and p16, lower MyHC, MyoD and myogenin, higher Atrogin-1 and MuRF-1, and markedly lower TPT1 protein. The cell findings support consistency across the study’s models, but they are not direct evidence in humans.
The next test is whether the signal holds up
Several other analyses were exploratory. A computational method called CIBERSORT estimated 22 immune-cell fractions and suggested differences in selected estimated cell populations, including lower plasma-cell and higher CD8+ T-cell and M1-macrophage estimates in sarcopenia samples. The study also identified 106 compound or chemical-perturbagen signature terms, while lower TPT1 expression coincided with higher scores for apoptosis, reactive oxygen species, inflammatory, stress-response, hypoxia and glycolysis pathways. These results generate ideas for follow-up work, but they do not show that any compound would treat sarcopenia.
The authors describe TPT1 as a candidate marker associated with the bulk-transcriptomic sarcopenia phenotype. They say prospective clinical validation and targeted functional studies are still needed to establish whether the signal is specific to sarcopenia, useful in practice or biologically causal. In particular, the current findings do not show that changing TPT1 would prevent muscle wasting.
Paper data and sources
Original title: Integrative transcriptomic and experimental analyses prioritize TPT1 as a PANoptosis-associated candidate molecular marker in sarcopenia.
Authors: Shijie Dong, Min Wang, Chen Liang et al.
Journal/Repository: Frontiers in cell and developmental biology
Status: Peer-reviewed
First online: 2026-08-20
DOI: 10.3389/fcell.2026.1912393
Original paper