In order for populations of cells to respond coherently to extracellular signals, individual cells must react accurately despite inevitable differences in gene expression and metabolic state. This principle is critical during processes such as embryonic development and maintenance of tissue homeostasis in adults, in which uncoordinated behavior can be deleterious. Some mechanisms that suppress cell-to-cell variability, such as negative feedback control of gene expression, have been known for half a century, but how robustness in signal transmission is maintained at the whole-cell level remains largely unknown.
Studying variability in signal transduction can be difficult due to the necessity of measuring quantitative, often subtle, phenotypes at the single cell level. The Brent Lab (Basic Sciences Division) uses the pheromone response system (PRS) in the budding yeast Saccharomyces cerevisiae to study cell-to-cell variability in signaling. In this process, a yeast cell detects pheromone excreted by a compatible mating partner using a G-protein coupled receptor on its surface, initiating a series of signaling events in which G-proteins, scaffolding proteins and MAP kinase cascade proteins congregate at the cell membrane. Subsequent signaling via a phosphorylation cascade leads to induction of genes that facilitate mating with the partner cell.
In work initiated over a decade ago at The Molecular Sciences Institute (MSI) in Berkeley, California and recently published in Molecular Systems Biology, Dr. Roger Brent set out to identify genes that affect cell-to-cell variability in the PRS. Dr. Gustavo Pesce, then a research fellow at MSI, constructed a library of yeast strains carrying necessary reporter genes and single deletions in non-essential genes. Using flow cytometry, Dr. Pesce and his colleagues screened 1000 members of the library for mutants that exhibited differences in variability and/or strength of the PRS output signal.
The screen identified 50 genes of interest. While some genes influenced both variation and overall signal strength, many only affected one of these properties, thus demonstrating that signal strength and variability can be genetically separated. This observation indicates that variability is not simply a function of signal strength, suggesting that evolutionary forces may shape the degree of variability in different contexts.
Six of the 50 genes implicated in the screen were already known to be related to the PRS; many of the others are involved in cell cycle regulation, gene expression or metabolism. The researchers honed in on three genes--BIM1, GIM2 and GIM4--that caused increased variability in signaling when disrupted and are known to be involved in the function of microtubules. During the pheromone response, microtubules in the cytoplasm form a bridge connecting the nucleus to the signaling site at the cell membrane and are instrumental in bringing together the nuclei of the mating cells.
To explore how microtubule function suppresses variability in PRS output, the researchers used genetic and chemical means to perturb microtubule function in specific ways. For example, time-lapse microscopy of cells harboring mutations in BIM1, GIM2 and GIM4 revealed that these cells were less able to form signaling sites at the plasma membrane, and the amount of time required for site formation was both longer and more variable. In addition, a strain expressing a mutant allele of the motor protein Kar3 that freezes microtubule extension and contraction was found to exhibit high levels of signal variability, indicating that microtubules must be attached to the cell membrane and be able to exert pushing and pulling forces on the nucleus in order to suppress signal variation.
The researchers next sought to identify the step of PRS signaling at which microtubules affect signal variability. They found that the static arises at or upstream of the Ste5 scaffold recruitment step of the PRS and depends on Fus3, one of the two PRS MAP kinases, but it is unclear how microtubule perturbations cause erratic signaling by Fus3. “We speculate that, once initiated, small instabilities in signaling can become amplified into larger instabilities in recruitment of proteins to the signaling site, leading to larger instabilities in signaling," says Dr. Brent.
The authors propose that suppression of signal variability by microtubules likely occurs in multicellular organisms as well. Interestingly, the vertebrate orthologs of BIM1 interact with the tumor suppressor APC and other genes affecting cytoplasmic microtubule function have been associated with cancer. Given that one in a thousand people carries a polymorphism in one of the three Bim1 orthologs, it seems possible these and other microtubule-related polymorphisms could contribute to cancer initiation in humans. Asks Dr. Brent: "How much morbidity might be due to small, quantitative, functional differences in allelic forms present in the population, and could never be detected by blunt tools such as genome-wide association studies?"
Pesce CG, Zdraljevic S, Peria WJ, Bush A, Repetto MV, Rockwell D, Yu RC, Colman-Lerner A and Brent R. 2018. Single-cell profiling screen identifies microtubule-dependent reduction of variability in signaling. Mol Sys Biol. 14: e7390. doi:10.15252/msb.20167390
This work was supported by the National Institutes of Health and the Argentine Agency of Research and Technology.