Antibiotics emerged as miracle drugs and “silver bullets” in the early 20th century, revolutionizing medicine and our ability to combat infectious disease while positively impacting health and lifespans on a large scale. This remarkable triumph held steady for many years, and consequently antibiotic research and development diminished as a priority due to the seeming defeat of bacterial infections. However, the selective pressure that came with antibiotic exposure led to the development of bacterial resistance to these compounds, motivating renewed interest in what is now an extremely important public health issue. Mechanisms of resistance are many and ever-evolving, and we know now that it is not a matter of IF bacteria will become resistant to a class of antibiotics, but when. The search for new and potentially exploitable bacterial vulnerabilities, then, becomes a constant enterprise in order for us to keep pace with the bacteria in the antibiotics/resistance arms race.
A new study this week in Nature Genetics describes how manipulating the bacterial DNA methylome affects susceptibility to multiple classes of antibiotics. The authors observed that deleting the dam gene, encoding a DNA methyltransferase, from E. coli causes increased susceptibility to sub-lethal doses of the β-lactam antibiotic ampicillin. Dam specifically methylates GATC sites, and deletion of any of the other three DNA methyltransferases found in E. coli had no effect on the level of antibiotic susceptibility. Using SMRT sequencing, the authors saw that genome-wide GATC methylation patterns did not change after exposure to ampicillin, so they sought alternative explanations for the observed phenotype.
As newly replicated DNA is not immediately methylated, GATC methylation enables strand discrimination by marking template DNA. This hemimethlyated state coordinates methyl-dependent mismatch repair (MMR), insuring the correction of mismatches exclusively on the nascent strand. Repair is mediated by endonucleolytic cleavage so in the absence of the GATC methylation guide, the MMR pathway can introduce double-stranded breaks during the repair process. This could be exacerbated under antibiotic stress, which induces the error-prone polymerase PolIV, resulting in a higher frequency of mismatches. The authors hypothesized that disruption of methylation led to unregulated MMR in the context of increased mutagenesis under stress, resulting in the potentiation of antibiotics.
Through deletion experiments, the authors were able to show that the enhanced susceptibility of dam-deleted E. coli to ampicillin was indeed mediated by components of the MMR pathway and the error-prone polymerase. They also tested different quinolones and found that the dam-deleted E. coli, including a uropathogenic (UPEC) strain, were also hypersensitive to these compounds. Finally, they demonstrated that deleting dam in a ciprofloxacin-resistant clinical UPEC isolate partially restored sensitivity, highlighting the ability to potentiate antibiotics through methylation manipulation in resistant strains. This has possible implications for the enhancement of antibiotic action through the targeting of bacterial DNA methylation, representing a potential therapeutic strategy for the treatment of both antibiotic-sensitive as well as antibiotic-resistant bacteria.
We spoke with lead investigator James Collins to get some background on this research.
Were you surprised that you didn’t see perturbations of the methylation patterns upon treatment with ampicillin?
Absolutely. Given the pervasive and multifaceted bacterial responses documented to occur following drug-induced stress, the quasi-complete stability of the E. coli methylome over the course of ampicillin exposure was a totally unexpected finding that left us scratching our heads for quite some time.
Do you think that targeting bacterial methylation would be an effective strategy to potentiate antibiotics? What might some of the advantages and drawbacks be?
Dam is a very attractive drug target as it has no human homolog but is conserved in several clinically important organisms. Because GATC methylation can regulate pathogenicity factors, Dam inhibition has also been proposed as an anti-virulence strategy. Going forward, much remains to be learned about the effects of Dam inhibition in the context of an infection. One concern is that elevated rates of mutagenesis could complicate a Dam inhibitor-based therapy.
While we expect this effect would be mitigated in the context of combination treatment, it will be important to verify that this is the case. Although further investigation is clearly needed, we feel that targeting bacterial methylation to potentiate antibiotics is an exciting strategy that is certainly worth exploring further.
One of the most interesting findings is the sensitization of a dam-deleted strain of antibiotic-resistant pathogenic E. coli to ciprofloxacin. What clinical implications do you think this might have?
We agree! This was another completely unexpected finding that challenges the long-held notion that highly resistant organisms (for instance organisms in which the drug target is mutated such that the drug can not longer bind) are completely impervious to treatment. The finding that antibiotics can be potentiated in even the most resistant bacteria suggests that stress responses are occurring in these organisms, and that these responses might be therapeutically exploited.
Bacteria have evolved many ways to combat antibiotics. Obviously DNA methylation plays various important roles in bacterial physiology. Do you think antibiotic exposure could have had any influence on the evolution of Dam-dependent methylation?
This is an interesting question. Given the fundamental nature of the processes that are regulated by GATC methylation (cell division, DNA repair etc.), and the ubiquitous presence of antimicrobial compounds in nature, it is difficult to know what role antibiotic exposure may have played in the evolution of genomic Dam methylation.
Here you show how these antibiotics essentially exploit the cell’s intrinsic mechanisms of strand discrimination (methylation) and DNA repair (MMR and error-prone polymerase) to act. How universal do you think this phenomenon is?
The GATC methylation system in E. coli is by far the best-characterized strand discrimination system, and while Dam is conserved in several pathogenic organisms, many organisms also lack Dam (Gram-positive bacteria, for example). In these cells, strand discrimination mechanisms remain poorly understood. However, because the MMR machinery is more broadly conserved, it is possible that interfering with strand discrimination in such organisms could potentiate antibiotics via a similar mechanism. More generally, we find that exploring the possibility of corrupting bacterial endonuclease function and exploiting their potential genotoxicity is an interesting avenue.