Join the Merrikh Lab!

Postdoc positions are available.



Lab Photo 2017

Research Overview

What are replication-transcription conflicts?

Bacteria (and most other organisms) have a common requirement for accurate, timely, and faithful DNA replication. Yet a wide variety natural impediments are known to physically slow or stall DNA replication forks, reducing fitness and increasing mutation rates. Impediments include chemical lesions, broken DNA strands, and tightly bound proteins. Much of our interest lies in identifying such obstacles, and determining what happens when DNA replication forks encounter them. In particular, we have found that actively transcribing RNA polymerases represent the most significant impediment to DNA replication. Using single molecule analysis of DNA repliction fork dynamics, we have shown that transcription collapses the replication fork and the replisome multiple times per cell cycle, increasing mutation rates, causing DNA strand breakage, and cutting the average rate of DNA replication in half. Hence replication-transcription conflicts are a major and fundamental problem for bacteria.

Our lab uses a variety of cutting edge tools to study replication-transcription conflicts. These include in vivo single molecule microscopy, deep sequencing, bioinformatics, and lab-based experimental evolution. We pair these new technologies with classical genetics and molecular biology techniques, typically using the model bacterium Bacillus subtilis. This naturally competent bacteria grows quickly, is genetically tractable, and a deletion collection is available. In recent years we have also branched out into pathogenic species including Mycobacterium tuberculosis, Pseudomonas aeruginosa, and Salmonella enterica.

How are replication-transcription conflicts resolved?

Replication and transcription are not temporally separated. This leads to collisions between the two machines as they move along the DNA. While the directionality of DNA replication is fixed, genes can be encoded on either the Watson or the Crick strand. This leads to two types of conflicts: head-on and co-directional. Though head-on collisions are far more severe than co-directional collisions, both can promote strand breakage and replisome collapse, requiring the action of replisome restart proteins. To maximize fitness and protect the genome cells have developed mechanisms to resolve conflicts. Thus far, we have identified several critical conflict resolution mechanisms. These include the accessory helicase PcrA, which appears to bolster the strength of the replisome and allow it to advance through even the most heavily transcribed head-on genes, and RNase HIII, an enzyme that degrades the RNA portion of stable DNA:RNA hybrids. We found that R-loops form specifically at head-on (but not at co-directional) conflict regions, permanently blocking the replication fork and killing the cell if left unresolved. Interestingly, our discovery of the orientation-specific nature of R-loop formation and the need for RNase HIII coincided with a similar discovery in human cells. The consistency between our results in bacteria and other labs' findings in higher organisms demonstrate that replication-transcription conflicts and their downstream consequences are similar throughout all life.

How do replication-transcription conflicts influence genome organization?

Replication-transcription conflicts are especially harmful when the two machineries meet head-on, i.e. when lagging strand genes are expressed. Presumably to avoid these encounters, bacteria have co-oriented the majority of their genes with replication by encoding them on the leading strand. While this greatly reduces the frequency of head-on conflicts, many lagging strand genes remain. Others have speculated that the continued existence of head-on genes could be explained if they were simply not expressed, or if their protein products are irrelevant for cell survival. Yet we find that many head-on genes are highly conserved and have essential functions. Furthermore, many are expressed (some constitutively) during growth, causing conflicts and blocking replication. Most recently, we investigated of a broad array of bacterial genomes and found that surprisingly, new genes are continually being added to the lagging strand over evolutionary time! This incredible finding indicates that the number of head-on conflicts are increasing in bacterial species across all major phyla! We are actively investigating these and other exciting patterns at the intersection between repliction-transcription conflicts and the evolution of genomic architectures.

Blocking evolution to prevent antibiotic resistance development

We recently discovered that head-on conflicts drive up mutation rates. This finding could potentially explain a second observation - that in the wild, head-on genes evolve at a faster rate (on average) than their co-directional counterparts. Our lab has made major progress towards identifying and characterizing some of the underlying mechanisms that actively increase mutation rates after replication-transcription collisions. In particular, we have identified a new class of proteins that we are terming "evolvability factors". Our new data show that these proteins promote rapid adaptation to new environmental stresses including exposure to antibiotics. Using new state of the art high-throughput lab-based evolution experiments and bioinformatics analyses, we are currently investigating the molecular mechanisms underlying these adaptive changes.