Transcriptome features of striated muscle aging and predictability of protein level changes

Abstract

RNA and protein levels correlate only partially and some transcripts are better correlated with their protein counterparts than others. This suggests that in aging and disease studies, some transcriptomics markers may carry more information in predicting protein-level changes. Here we applied a computational data analysis workflow to predict which transcriptomic changes are more likely relevant to protein-level regulation in striated muscle aging. The protein predictability of each transcript is estimated from existing large proteogenomics data sets, then transferred to new total RNA sequencing data comparing skeletal muscle and cardiac muscle in young adult (~4 months) mice vs. early aging (~20 months) mice. Aging cardiac and skeletal muscles both invoke transcriptomic changes in innate immune system and mitochondria pathways but diverge in extracellular matrix processes. On an individual gene level, we identified 611 age-associated signatures in skeletal and cardiac muscles at 10% FDR, including a number of myokine and cardiokine encoding genes. We estimate that about 48% of the aging-associated transcripts may predict protein levels well (r ≥ 0.5). In parallel, a comparison of the identified aging-regulated genes with public human transcriptomics data showed that only 35-45% of the identified genes show an age-dependent expression in corresponding human tissues. Finally, integrating both protein predictability and human translatability from multiple data sources, we nominate 134 prioritized aging signatures that are predicted to correlate strongly with protein levels and that show age-dependent expression in humans. These prioritized gene signatures may hold promise to understanding heart and skeletal muscle physiology in human and mouse aging.

Protein prediction models support widespread post-transcriptional regulation of protein abundance by interacting partners

“…The predictive models here are trained using publicly available CPTAC data from 8 cancer types, which contain transcriptome and proteome data from both tumor and normal adjacent tissue samples. In prior work, we found general concordance between protein and mRNA correlation in CPTAC samples vs. GTEx tissue proteomics data 15 . However, other studies that compared protein and mRNA correlation in tumors and normal adjacent tissues have found higher inter-sample correlation in tumor samples 20,21 , which may be attributable to the increased translation rates in cancer.…”

“…Transcriptomics experiments often operate on the implicit assumption that identified differential regulation exert their biological effects via their cognate proteins, hence it is important to better understand the relationships between protein and mRNA levels to aid in data interpretation and determining potential protein level changes given a set of transcriptomics data. Alternatively, it would be useful to know the genewise difference in how well a gene’s transcript can predict its protein counterpart as a means to filter or prioritize biologically relevant transcript signatures 15 .…”

Proteogenomic characterization of age and sex interactions in cardiac gene expression

“…Database search results were post-processed using Percolator v.3 (Crux v.4.0 distribution) (The et al, 2016) requiring 1% FDR for identification. TMT intensities were extracted from MS2 spectra using a custom Python script (Dostal et al, 2020) built on the pymzml library (Kösters et al, 2018) then corrected for isotope contamination and normalized across blocks as previously described (Han et al, 2021).…”

DataSheet1.xlsx

“…To estimate the potential number of cardiokine candidates, we queried large-scale human transcriptome data from Genotype-Tissue Expression Project (GTEx) v8. We then sequentially processed, normalized, and batch-corrected the data as previously described (Han et al, 2021). Next, we integrated this data with the circulating protein annotations from the Human Proteome Atlas (HPA) secretome data set (Uhlén et al, 2019), which lists a total of 784 proteins that are annotated as part of the human secretome and moreover was designated to be "secreted to blood" as their location.…”