Transcriptional errors and the drift barrier

Abstract

Population genetics predicts that the balance between natural selection and genetic drift is determined by the population size. Species with large population sizes are predicted to have properties governed mainly by selective forces; whereas species with small population sizes should exhibit features governed by mutational processes alone. This "drift-barrier hypothesis" has been successful in explaining extensive variation in genome size, mutation rate, transposable element abundance, and other molecular features across diverse taxa (1-3). However, in PNAS Traverse and Ochman (4) report a striking exception to this theory by showing that transcriptional error rates are nearly equal across several bacterial species with very different population sizes.Although the term "drift barrier" was coined in the context of mutation rates (5), the drift-barrier hypothesis applies to any heritable trait (e.g., ref. 6), and so it can provide a simple explanation for patterns of many molecular traits across the diversity of life. The explanation relies on the fact that mutations whose effects on fitness are smaller than the inverse of the population size behave like neutral mutations. As a result, natural selection will tend to optimize a trait until the selective benefits of further optimization are smaller than the inverse of the population size (7): that is, until the population hits the drift barrier. Because the location of the drift barrier depends on the population size, so too should the values of these phenotypic traits. Thus, differences in population size among species can explain a concerted syndrome of traits at the cellular and molecular levels (1-3, 5, 6).In the context of broad empirical support for the drift-barrier hypothesis, the results of Traverse and Ochman (4) are particularly surprising. Using a CircSeq (8) strategy (based on the sequencing of short, circularized fragments of mRNA that are copied multiple times by rolling-circle amplification before sequencing) to reduce errors in high-throughput sequencing, Traverse and Ochman measured transcriptional error rates in Escherichia coli along with two endosymbiotic prokaryotes, Buchnera aphidicola and Carsonella ruddii. These endosymbionts feature dramatically reduced genome sizes, increased mutation rates, and other features typical of their small population sizes, including the loss of several transcriptional fidelity factors (4). Nevertheless, Traverse and Ochman report less than twofold variation in the rates of transcriptional errors among these three species. This result is in sharp contrast to mutation rates, which differ by orders of magnitude between species (5). Possible Explanations for Conserved Error RatesWhat could explain the apparent constancy of transcriptional error rates across species with such different population sizes? One hypothesis we must consider is errors in the sequencing procedure itself. If the rate of sequencing error exceeds the true transcriptional error rates, then it would give the impression of constant error rates. Ho...

Drift Barriers to Quality Control When Genes Are Expressed at Different Levels

“…In agreement with drift barrier theory, large-N e E. coli exhibits a local solution-a tendency for transcription errors to have synonymous effects-while small-N e B. aphidicola does not (Traverse and Ochman 2016a). While, as predicted, the global solution of low transcriptional error rates does not obey the naïve drift barrier expectation of being higher in B. aphidicola than in E. coli (Traverse and Ochman 2016a), nor are transcription error rates drastically lower in B. aphidicola as predicted by previous theory on the interplay between global and local solutions (Rajon and Masel 2011;McCandlish and Plotkin 2016). This significantly lower rate relative to E. coli is, however, found in intermediate-N e C. elegans.…”

“…When we also account for mutation bias that tends to increase rather than decrease the error rate r, our model can explain the previously puzzling observation that the rate of transcriptional errors in small-N e endosymbiont bacteria Buchnera is so much higher than that of C. elegans, and almost as high as that of large-N e E. coli (McCandlish and Plotkin 2016;Traverse and Ochman 2016b). In extremely small populations, even the global solution is subject to a drift barrier, making r higher than its optimal value.…”

Drift barriers to quality control when genes are expressed at different levels

“…While, as predicted, the global solution of low transcriptional error rates does not obey the naïve drift barrier expectation of being higher in B. aphidicola than in E. coli (Traverse and Ochman 2016a), nor are transcription error rates drastically lower in B. aphidicola as predicted by previous theory on the interplay between global and local solutions (Rajon and Masel 2011;McCandlish and Plotkin 2016). This significantly lower rate relative to E. coli is, however, found in intermediate-N e C. elegans.…”

“…In any case, assuming mutation bias toward deleterious options is biologically reasonable, and Figure 3 shows that results are not sensitive to the quantitative strength of our assumption on this count. When we also account for mutation bias that tends to increase rather than decrease the error rate r, our model can explain the previously puzzling observation that the rate of transcriptional errors in small-N e endosymbiont bacteria Buchnera is so much higher than that of C. elegans, and almost as high as that of large-N e E. coli (McCandlish and Plotkin 2016;Traverse and Ochman 2016b). In extremely small populations, even the global solution is subject to a drift barrier, making r higher than its optimal value.…”

A Narrow Range of Transcript-error Rates Across the Tree of Life

“…Taken together, these observations strongly support the idea that the strength and/or efficiency of selection for error-rate reduction is substantially reduced at the level of transcription relative to replication. The lack of patterning in Figure 1a may lead to the impression that there are no explanatory biological factors associated with interspecies variation in the transcript-error rate, and that there is a uniform optimum error rate across the Tree of Life, contrary to the situation with genomic mutation rates (Lynch et al 2016;McCandlish and Plotkin 2016). However, the error-rate estimates displayed in Figure 1a have standard errors that are typically < 10% of the estimates, implying the presence of true parametric differences between species.…”