Stochastic activation of a DNA damage response causes cell-to-cell mutation rate variation

Abstract

Cells rely on the precise action of proteins that detect and repair DNA damage. However, gene expression noise causes fluctuations in protein abundances that may compromise repair. For the Ada protein in Escherichia coli, which induces its own expression upon repairing DNA alkylation damage, we found that undamaged cells on average produce one Ada molecule per generation. Because production is stochastic, many cells have no Ada molecules and cannot induce the damage response until the first expression event occurs, sometimes delaying the response for generations. This creates a subpopulation of cells with increased mutation rates. Non-genetic variation in protein abundances thus leads to genetic heterogeneity in the population. Our results further suggest that cells balance reliable repair against toxic side-effects of abundant DNA repair proteins.

Single-cell microscopy of suspension cultures using a microfluidics-assisted cell screening platform

“…Regardless of the underlying mechanism we have found that the slowing down of diffusion allows for visualization and even accurate counting of fluorescent proteins at exceptionally low abundances, using conventional TIRF microscopy setups 10,11 . As opposed to chemical fixation protocols, this mechanical fixation can be achieved with no loss of fluorescence, and extensive controls with photobleaching steps and tandem dimer FPs 11 suggest that we do detect close to all mature fluorescent molecules.…”

“…For instance, in bacteria, a large fraction of the proteome is present in just a handful copies per cell 8 . These proteins have been very difficult to analyze 9 , and include many key regulatory proteins that contribute greatly to the phenotypic heterogeneity of cells 10 . We recently developed a microfluidic platform, termed MACS for M icrofluidics- A ssisted C ell S creening, to address the limitations of flow cytometry and traditional microscopy 11 .…”

“…As opposed to chemical fixation, this procedure does not quench the fluorescence signal, and makes it possible to accurately identify single FPs in living cells. We previously quantified this effect by applying our method to single FPs as well as FP-FP fusions, by counting photobleaching steps 11 , and by comparing the results to other methods 10 , showing that the approach makes it possible to accurately count the number of fluorescent proteins per cell even when cells contain just one or two protein copies 10 . The flattening of cells in the z-dimension and widening in the x-y plane also helps with counting in three ways: (1) by reducing the projected autofluorescence per area unit and thereby increasing the signal-to-noise ratio for FP detection, (2) by further separating the proteins from each other to reduce the risk of spatial overlap, and (3) by making it possible to keep the whole cell in focus.…”

“…The MOPS-based rich defined medium EZRDM (Teknova) works even better for this purpose. By paying close attention to cleaning procedures at different stages, background fluorescence drops to as few as 0.3 spots/cell for a control strain that does not express any fluorescent proteins 10 .…”

The Design and Transformation of Biofundamentals: A Nonsurvey Introductory Evolutionary and Molecular Biology Course

“…As part of this focus, we have included materials based on the work of Vilar et al . (2003) and, more recently, Uphoff et al . (2016) on the roles of regulatory noise, including the stochastic processes that underlie DNA–protein interactions (Elowitz et al ., 2002; Maheshri and O’Shea, 2007).…”