Fitness effects of altering gene expression noise in Saccharomyces cerevisiae

• Published in: eLife Vol. 7
• Main Authors: Duveau, Fabien, Hodgins-Davis, Andrea, Metzger, Brian PH, Yang, Bing, Tryban, Stephen, Walker, Elizabeth A, Lybrook, Tricia, Wittkopp, Patricia J
• Format: Journal Article
• Language: English
• Published: England eLife Science Publications, Ltd 20.08.2018; eLife Sciences Publications Ltd; eLife Sciences Publications, Ltd
• Subjects: Analysis; Base sequence; Biochemistry; Cell division; competitive growth; DNA sequencing; evolution; Evolution (Biology); Evolutionary Biology; Gene expression; Gene Expression Regulation, Fungal - genetics; Genes; Genetic aspects; Genetic Fitness - genetics; Genetic research; Genotype; Genotype & phenotype; Genotypes; Glyceraldehyde-3-Phosphate Dehydrogenase (Phosphorylating) - genetics; Natural selection; Noise; Population growth; promoter; Proteins; Reproductive fitness; RNA; Saccharomyces cerevisiae; Saccharomyces cerevisiae - genetics; Saccharomyces cerevisiae Proteins - genetics; selection; Selection, Genetic; Single-Cell Analysis; Standard deviation; TDH3; Transcription factors; Ann Arbor Michigan; United States > US; Israel; evolutionary biology; selection; competitive growth; promoter; evolution; TDH3; gene expression; S. cerevisiae
• ISSN: 2050-084X; 2050-084X
• DOI: 10.7554/eLife.37272

Gene expression noise is an evolvable property of biological systems that describes differences in expression among genetically identical cells in the same environment. Prior work has shown that expression noise is heritable and can be shaped by selection, but the impact of variation in expression noise on organismal fitness has proven difficult to measure. Here, we quantify the fitness effects of altering expression noise for the TDH3 gene in Saccharomyces cerevisiae. We show that increases in expression noise can be deleterious or beneficial depending on the difference between the average expression level of a genotype and the expression level maximizing fitness. We also show that a simple model relating single-cell expression levels to population growth produces patterns consistent with our empirical data. We use this model to explore a broad range of average expression levels and expression noise, providing additional insight into the fitness effects of variation in expression noise. Single-celled organisms that reproduce by dividing, like yeast, can create whole populations of genetically identical cells. However, some differences will exist among such cells, even when they have all experienced the same environment. These differences are known as “noise”. By definition, noise is not caused by differences in DNA sequence, but some DNA sequences are noisier than others (i.e. they cause more differences among cells). Because the amount of noise can be under genetic control, noise could evolve due to natural selection. Scientists often study noise at the level of gene expression – in other words, how many RNA or protein molecules are produced from each gene within each cell. Prior work has suggested that this type of noise can affect how often individual cells divide in a population, which is a component of that population’s fitness. Yet directly measuring these effects has proven challenging. Different studies have in the past reached opposite conclusions about whether a change in gene expression noise would increase or decrease fitness. One major reason for the lack of clear results is that most mutations that alter gene expression noise also alter the average level of expression of that gene. To find DNA sequences that produced the same average amount of protein but different levels of expression noise, Duveau et al. compared the effects of hundreds of mutations in the DNA sequence regulating the expression of a gene in baker’s yeast. Experiments focused on 43 DNA sequences then showed that increased expression noise could either speed up or slow down the growth of the population by affecting how long it took each cell to divide. More specifically, the effects of increasing expression noise depended on the average amount of protein produced among the cells in the population. If the average expression level was close to the optimum amount at which cells divided as fast as possible, increasing expression noise reduced the growth of the whole population. If, however, the average protein level caused cells to divide slower than their maximum rate, increasing expression noise resulted in faster growth of the population as a whole. Duveau et al. explain their results as follows: more expression noise in a population that is already making the optimal amount of protein can reduce fitness because it increases the fraction of that population making a suboptimal amount of the protein. However, when the average expression level is not optimal, more expression noise would mean more cells producing an amount of protein that is closer to the optimum and thus having higher fitness. These findings provide conceptual tools needed to understand how genetic variation affecting expression noise evolves. They could also help understand how expression noise might contribute to biological processes that depend upon cell division, such as diseases like cancer.