The central dogma of molecular biology states, “DNA makes RNA, and RNA makes protein.” Accordingly, protein expression has been believed to be primarily regulated by the production and degradation of mRNAs. Everyone believes it is a matter of course that protein expression is well-correlated thanks to mRNA expression.
However, recent studies have shown that their relationships are not so simple, but rather controversial. A recent review published in Cell  examined their relationships in various conditions and concluded that mRNA levels by themselves are often not sufficient to predict protein levels.
For example, a study cited in the review conducted a genome-wide, time-series survey using yeast that was subjected to osmotic stress and examined the correlation between the protein and mRNA expressions . The correlation was not at all clear in two of the four most regulated pathways, and they only mentioned a possibility it “may be mainly due to a slight delay observed for the protein response compared to the RNA response.”
A potential solution comes from a Ph.D. project that aimed to develop an analytical framework for identifying specific gene groups that have a significant correlation between mRNA and protein from such genome-wide time-series data. The study mined the data of the previous study using yeast  provided by the first author Dr. Nathalie Selevsek at ETH Zürich, and made it possible to account for potential time delays in their relationships and classify genes according to the time delays and concordance of the time course of mRNA and protein expressions.
The framework was indeed effective in finding stronger correlations between mRNA and protein abundance among genes in the two pathways. Moreover, from the genome-wide data, it identified a pair of stress-responsive genes that showed a statistically significant correlation (Figure 1). The genes are distantly located in the same chromosome and encode different proteins both of which were reported to be important for cells to survive after stress. A concerted role of these genes would be vital in cellular stress response in yeast, and further studies are warranted to understand molecular mechanisms that enable such a long-distant interaction between the genes.
In addition, the study applied the analysis framework to another genome-wide time-series data in mammalian cells after stress and identified a group of genes related to cytoskeletons that show similar significant correlation. Another concerted role of these genes would be vital in cellular stress response in mammalian cells.
The study identified the two types of gene groups from the two genome-wide time-series data. On the other hand, such a significant correlation was not identified in the vast majority of genes, indicating relationships between mRNA and protein abundance are not as simple at all as suggested from the central dogma! Further studies are warranted to disentangle the relationship after accounting for various factors in addition to the time delays and concordance of the time course of mRNA and protein expressions examined in the study.
These findings are described in the article entitled Identification of stress responsive genes by studying specific relationships between mRNA and protein abundance, recently published in the journal Heliyon. This study was conducted by Shimpei Morimoto from the Kurume University School of Medicine and Koji Yahara from the National Institute of Infectious Diseases.
- Liu Y, Beyer A, Aebersold R. On the Dependency of Cellular Protein Levels on mRNA Abundance. Cell. 2016;165(3):535-50.
- Selevsek N, Chang CY, Gillet LC, Navarro P, Bernhardt OM, Reiter L, et al. Reproducible and consistent quantification of the Saccharomyces cerevisiae proteome by SWATH-mass spectrometry. Mol Cell Proteomics. 2015;14(3):739-49
- Shimpei Morimoto, Koji Yahara. Identification of stress responsive genes by studying specific relationships between mRNA and protein abundance. Heliyon, Volume 4, Issue 3, March 2018, Article e00558