Antimicrobial Resistance
Antimicrobial resistance poses a major global health risk. Infections that were once simple to cure can become life-threatening as bacteria develop resistance to previously effective drugs. This challenge impacts not only common illnesses but also critical medical procedures such as surgeries and cancer therapies.
Rising resistance rates have forced reliance on older antibiotics, such as polymyxins, as last-resort treatments despite their limited effectiveness. This dependence is especially evident in low-income settings where alternatives are scarce.
How do polymyxins kill bacteria?
Discovered over 80 years ago, polymyxins are used to treat infections caused by Gram-negative bacteria such as Escherichia coli. These kinds of bacteria possess an outer membrane that helps block many antibiotics from entering the cell. The main components of this outer membrane are lipopolysaccharides (LPS) and outer membrane porins (OMPs). While polymyxins are known to attack this protective layer, the precise way in which they disrupt it and successfully kill the bacteria remains unclear.
One common misconception in the field is that membrane-targeting antimicrobials are effective against dormant bacteria. Dormant bacteria are bacteria that enter a temporary inactive state where they stop growing and dividing but are still alive. This makes them able to survive harsh conditions and “wake up” when the environment is more favourable. Because the membrane is essentially the same whether the bacteria are active or dormant, the hypothesis is that polymyxins can target it regardless of their activity.
But is that really the case for polymyxins? Do they, in fact, kill dormant bacteria?
Our methods
By employing a mix of biochemical and biophysical methods, including Atomic Force Microscopy (AFM), we investigated how polymyxin B (PmB) acts at the molecular level in dormant and active bacteria. AFM makes use of an incredibly sharp tip that gently raster scans the surface of a sample (i.e. bacteria) and creates a high-resolution image of the sample topography. This is a relatively novel tool in microbiology, and our group is among the first to apply it directly to live bacterial cells. To fully harness this approach, our collaboration brings together expertise across three leading universities—Imperial College London, UCL, and the University of Nottingham.
Polymyxins highly disrupt the membrane of E. coli
The first thing we observed when attempting to answer this question was that PmB highly disrupts the outer membrane of active E. coli when in favourable conditions i.e., the presence of nutrients. This effect was visible in AFM images of bacteria undergoing treatment in real-time: the otherwise smooth outer membrane quickly became packed with protrusions and overall disorganisation. With AFM, we were able to track the antibiotic's effect in real time over 90 minutes. We found it fascinating to see just how quickly PmB caused disruption to this outer membrane, and how much damage the bacteria can endure before ultimately dying. Not only that, but through a series of microbiological and biochemical assays, we proved that a lot of the LPS in the outer membrane was released from the bacteria. It seems as though the cell is sensing the PmB binding to it and is compelled to generate more outer membrane as a result of the treatment, so rapidly that it becomes destabilised, creating openings that allow the antibiotic to enter.
Polymyxins fail to kill metabolically inactive bacteria
However, when treating dormant bacteria with PmB in the absence of nutrients like glucose, this outer membrane remained intact, and there was no shedding of LPS from it, leading us to believe that this class of antibiotics is ineffective against dormant bacteria. Through microbiological assays, we observed that PmB was still able to bind the outer membrane but was unable to penetrate it and kill these dormant cells.
Our collaboration
Throughout the research, the collaboration between the institutions is what we believe really made this work so impactful and allowed the rapid progress achieved in this study. The interdisciplinary nature of this collaboration enabled us to examine the same question from multiple angles, each bringing something unique and valuable to the answer. For instance, several treatment conditions were initially evaluated using microbiological assays (e.g., time-kill experiments). These conditions were subsequently reproduced in the AFM or Scanning Electron Microscope (SEM) experiments to allow direct comparison. In microscopy, we usually look at one or two cells at a time, especially when tracking the effects of a treatment, so we were able to correlate what was happening at the single-cell level to the population level.
This work proposes a new model for how polymyxins kill bacteria. They first bind to the outer membrane of E. coli, regardless of whether cells are metabolically active. Then, in active cells, this results in LPS release, which is dependent on the synthesis and transport of new LPS. This, in turn, causes outer membrane destabilisation and protrusions, which result in membrane gaps that allow polymyxins to access the inner membrane and ultimately kill the bacteria. In metabolically inactive cells, the binding is not enough to allow access to the inner membrane.