Latent Antibiotics
Bacteria play an undeniably important role in human health, as crucial symbionts, as harmless bystanders, and also as pathogenic opportunists or invaders. Despite the premature declaration of victory over pathogenic bacteria, untreatable infections have remained, and the growing problem of antibiotic resistance is threatening to turn many previously treatable infections into serious medical conditions with potentially deadly outcomes – just as they were before the age of antibiotics.
The development of novel antibiotics has proven extraordinarily challenging and the majority of our antibacterial arsenal are derived from natural products. These natural product antibiotics were likely evolved as part of a natural arms race between microorganisms over eons of time. From a practical perspective, broad-spectrum activity is essential for antibiotic development (based on current economics, no drug company will invest the resources necessary to develop a narrow spectrum antibiotic), but unfortunately, it appears that most classes of natural product antibiotics with broad-spectrum activity have already been discovered. The reduced spectrum of the relatively more common narrow-spectrum agents is typically considered a result of intrinsic limitations, such as poor penetration or insufficient target conservation. However, if the natural, age-old arms race between microbes is anything like the modern day arms race between medicinal chemists and pathogenic bacteria, previously evolved specific resistance mechanisms might also contribute to the narrow-spectrum of some natural product antibiotics. In contrast to intrinsic limitations, chemists have had much success in optimizing scaffolds to overcome such specific mechanisms of resistance, as evidenced by the many ‘next generation’ antibiotics that have been developed. This suggests that if such ‘latent antibiotics’ could be identified, then they might be optimized into new classes of broad-spectrum agents.
Our first efforts to identify latent antibiotics have been directed towards the arylomycin class of natural products. These compounds selectively bind type I signal peptidases (SPases), which function within protein export and secretion pathways. Despite in vivo activity, and the fact that SPase is essential, conserved, and relatively accessible, the arylomycins appeared to have only a very narrow spectrum of activity, and they were dismissed as drug candidates. Towards understanding what limits the spectrum of arylomycin activity, we synthesized the first member of this class of natural products, arylomycin A2. Using a microbiological approach, we found that S. epidermidis is extremely sensitive (it is actually more sensitive to the arylomycins than it is to antibiotics that are currently prescribed for its treatment), and that it evolves resistance via a specific mechanism based on the introduction of target mutations that reduce arylomycin affinity. Via a phylogenetic analysis, we showed that inherently resistant pathogens already have the target mutations, and moreover, we also identified many bacteria that did not – most of which were then found to be highly sensitive to the arylomycins. Moreover, when the ‘resistance-conferring’ mutations are removed, E. coli, P. aeruginosa, and S. aureus are potently killed. The conclusion of our work to date is that the arylomycins are actually very broad-spectrum antibiotics, and that if optimized to bind their targets with slightly more affinity (to overcome the decrease in affinity caused by the target mutations), the arylomycins would have a spectrum of activity that supports their progression as broad-spectrum therapeutics. Towards this goal we have initiated synthetic chemistry efforts to re-optimize the arylomycins for broad-spectrum activity.