Expression-dependent but strand-independent synonymous single-nucleotide polymorphism in the Escherichia coli chromosome

This study analyzes synonymous single-nucleotide polymorphisms across 157 *Escherichia coli* strains to demonstrate that specific mutation patterns are expression-dependent but strand-independent, providing evidence for the role of transcription-induced mutagenesis in shaping genomic variation.

The Gist

Imagine the E. coli bacterium's DNA as a massive, double-stranded instruction manual for building a living machine. Usually, we think of mistakes (mutations) in this manual happening mostly when the book is being copied (replication), like a tired scribe making typos while photocopying pages. However, scientists also know that reading the book (transcription) can sometimes cause damage, like a reader accidentally smudging ink while turning pages.

The big question this paper asks is: Can we actually see the difference between mistakes made while copying the book versus mistakes made while just reading it?

To find out, the researchers acted like detectives, examining the "typos" (specifically, harmless changes called synonymous SNPs) in the instruction manuals of 2091 different genes across 157 different strains of E. coli. They looked at two specific lanes of traffic on the DNA highway: the Leading Strand (the smooth, fast lane) and the Lagging Strand (the bumpy, stop-and-go lane).

Here is what they discovered, broken down into simple analogies:

1. The "Smudged Ink" Effect (Transcription)

The researchers found that certain types of typos happened more often when the genes were being read loudly and frequently (high expression).

• The Metaphor: Imagine a popular book in a library. The more people read it (high expression), the more likely the pages are to get worn out or smudged, regardless of which side of the page you are looking at.
• The Finding: Specifically, the transition from T to C and A to G (a specific type of letter swap) was heavily influenced by how much the gene was being used. Crucially, these mistakes happened equally on both the smooth lane and the bumpy lane. This proves that the act of reading the gene causes these specific errors, not the act of copying it.

2. The "Copy-ist's Bias" (Replication)

Other types of typos showed a different pattern. They happened more often on one specific lane (the Leading or Lagging strand) than the other.

• The Metaphor: This is like a scribe who has a bad habit of making a specific mistake only when they are moving their hand in one direction (say, left-to-right) but not the other.
• The Finding: Changes like C to T and G to A were heavily influenced by which "lane" the gene was on. This suggests these errors are tied to the mechanics of copying the DNA, where the two strands are treated differently.

3. The "Universal Preference"

The study also noticed that some typos just happened more often than their opposite versions, no matter what.

• The Metaphor: It's like a coin that is slightly weighted; it lands on "Heads" more often than "Tails" every single time, regardless of who flips it or where they are standing.
• The Finding: Certain letter swaps (like A turning into T) were universally more common than their reverse (T turning into A).

The Bottom Line

The main takeaway is that the researchers successfully separated the "noise" of copying from the "noise" of reading. They proved that transcription (reading the gene) causes specific mutations that happen equally on both strands, depending on how busy the gene is. This supports the idea that the simple act of a cell reading its own DNA instructions is a significant source of genetic change, distinct from the errors made during the copying process.

Technical Summary

Technical Summary: Expression-Dependent but Strand-Independent Synonymous SNP in E. coli

Problem Statement
While mutation is traditionally understood to arise primarily during DNA replication, transcription is also recognized as a mutagenic process. Despite recent reports suggesting genome-wide transcription-induced mutagenesis, a distinct demonstration establishing specific mutations as either replication-dependent or transcription-dependent within genomes remains to be fully established. This study addresses the need to differentiate the contributions of replication and transcription to mutational patterns, specifically examining whether mutation frequencies vary based on the DNA strand (leading vs. lagging) and gene expression levels.

Methodology
The authors conducted a comparative analysis of synonymous single-nucleotide polymorphisms (SNPs) across 157 strains of Escherichia coli. The study focused on 2,091 individual coding sequences (CDS) distributed across both the leading strand (LeS) and the lagging strand (LaS) of the chromosome. To assess parity violation and mutational biases, the researchers compared the frequencies of complementary transitions (ti) and complementary transversions (tv) within each CDS. The analysis specifically evaluated how these frequencies correlated with the strand location and the expression levels of the genes.

Key Results
The analysis yielded distinct patterns regarding the drivers of mutational inequality:

• Transitions (ti): The C→T and G→A transitions exhibited the highest frequencies and the most prominent strand inequality. These mutations were found to be influenced by both the specific DNA strand (LeS vs. LaS) and the level of gene expression.
• Expression-Dependent, Strand-Independent Inequality: A notable finding was that the inequality between T→C and A→G transitions was dependent on gene expression but independent of the DNA strand.
• Transversions (tv): The A→T and G→T transversions were universally more frequent than their complementary counterparts (T→A and C→A, respectively).

Significance and Claims
The paper claims to demonstrate the existence of strand-independent but expression-dependent synonymous SNP inequality within coding sequences. By identifying mutation patterns that correlate with expression levels rather than solely with strand-specific replication mechanics, the study supports the role of transcription-induced mutagenesis as a contributing factor to strand inequality in the E. coli chromosome. The authors conclude that these findings provide evidence for a mechanism where transcription, distinct from replication, drives specific mutational biases in the bacterial genome.