In March 2018, researchers published a paper that could signal a break in the spiraling battle between antibiotics and superbugs1. It looks at the properties of an infection-fighting synthetic polymer, the most eye-catching element of which is that the polymer doesn’t appear to set off an adaptive response in bacteria at all.
The standard treatment for a serious bacterial infection is a dose of antibiotics, which slow or halt the infection by hindering critical cellular processes within the bacteria. However, some bacteria have evolved devious mechanisms to protect themselves against antibiotics, for instance by producing enzymes that can destroy the antibiotic molecules, or by making themselves less permeable to the antibiotic. The most resilient become better at this every time they’re exposed to the drugs. The class of multi-drug resistant bacteria that have become the most difficult, or, even impossible to treat, are known as superbugs. Last year, a UK government report suggested that, by 2050, drug-resistant infections could kill one person every three seconds.
In the 2018 study by a team working under Yi Yan Yang, a bioengineer from the Institute of Bioengineering and Nanotechnology (IBN), researchers revealed a polymer that is biodegradable, non-toxic and can kill the Acinetobacter baumannii superbug within an hour. Genomic tests conducted by the Genome Institute of Singapore (GIS) then confirmed that the bacteria did not develop resistance to the polymer.
Serendipity: how computer hardware led to infection treatments
Last year, the World Health Organization (WHO) released a list of 12 of the world’s most dangerous superbugs — a move intended to underline them as the highest priorities for antibiotic research. Marie-Paule Kieny, the WHO’s assistant director-general for health systems, was pointed in a WHO communiqué: “If we leave it to market forces alone, the new antibiotics we most urgently need are not going to be developed in time […] The pipeline is practically dry.” However, a treatment that kills bacteria and doesn’t stimulate resistance could represent a new chapter in the fight against superbugs.
Yang was aware of the superbug problem when her team started their polymer work, but superbugs also just pique her curiosity. Bacteria, she says, are not unlike our own mammalian cells in their ability to develop and learn. “They’re very smart.” If bacteria were in a classroom studying survival, superbugs would be top students, she says.
Yang, who is also interested in using nanotechnology for drug delivery, seeks and attracts interesting collaborators. The development of the most recent polymer began with multinational tech company IBM. While at a conference in Australia in 2007, Yang heard IBM’s James Hedrick talk about a unique, positively charged synthetic polymer his team had accidentally made while working on creating silicon wafers for computer semiconductor technology. The polymer was particularly interesting to Yang: its positive charge means it has the potential to become attached to negatively charged bacteria surfaces when introduced to the bloodstream, while leaving human cells alone. Today, Hedrick, the lead scientist for IBM Research’s advanced organic materials group, is one of Yang’s main collaborators.
Since then, Yang and a handful of other researchers have looked at various polymers, but each new iteration encountered problems. Biodegradability is one – without a route for excretion, non-biodegradable polymers could build up in a patient’s body, potentially causing problems. Yang’s new polymer is broken down via a chemical reaction with water, the products of which are excreted within days, meaning it's a much more promising candidate for drug development than pervious examples.
In 2016, a University of Melbourne study published in Nature Microbiology drew attention when it found a similar lack of drug resistance in their polymer2. Newspapers put a positive spin on the fact that it “ripped the bacteria apart”. But Yang says that, actually, this is not really a good thing. “With polymers that break down the bacteria cell membrane, you worry that a lot of endotoxins will be released from the bacteria cells, and those endotoxins may cause sepsis.” In other words, they could be toxic. This was exactly what stopped Yang from proceeding with drug trials for a polymer she published about in 2011 in Nature Chemistry3. In large quantities, endotoxins – toxins present in the membranes of many types of Gram-negative bacteria – can cause severe diarrhea, septic shock, and even death. Smaller amounts may cause fever, chills, and can lower the number of white blood cells in the body, leading to a compromised ability to fight infection.
Yang’s new polymer, by contrast, binds to bacteria cell surfaces, moving through the cells' membranes into the cells. Once inside, the polymer condenses the proteins and DNA inside, killing the cells. The dead bacterial cells are then digested by macrophages, and preclinical tests showed that the products of the polymer are excreted within days.
Getting all these elements right took years of communication between Yang and chemists at IBM, who worked on the polymer design. They eventually settled on a guanidinium-functionalized polycarbonate, which was then synthesized in Yang's laboratory.
In the first stage of testing the researchers compared the polymer with imipenem, an important injected antibiotic used in hospitals to treat pneumonia, sepsis, joint infection and several other bacterial infections. They did this by repeatedly applying sub-lethal concentrations of the two agents on A. baumannii bacteria and looking for changes in the concentration at which the agents were effective. “Any increase in the effective concentration of antimicrobial agents against the resilient bacteria means that the bacteria have developed resistance against the antimicrobial agents,” explains Yang.
With imipenem at sub-lethal concentrations, the group began to see an increase in the effective concentration levels of imipenem within eight to ten applications. By contrast, the effective concentration of their polymer stayed roughly the same even after 30 applications at sub-lethal concentrations.
The group then proved that the polymer is not toxic to mice and is effective against five globally important superbugs: E. coli, methicillin-resistant Staphylococcus aureus (MRSA), P. aeruginosa, K. pneumoniae and A. baumannii.
Yang next looked to the genomes of the treated bacteria for more detail on the mechanisms behind their findings. It was the team of Paola Florez de Sessions at A*STAR’s Genome Institute of Singapore who were tasked with analyzing the antibiotic and polymer-treated bacteria’s genes using RNA sequencing.
In the imipenem-treated bacteria, the team observed an upregulation of genes associated with imipenem resistance, as well as those linked to resistance against other classes of antibiotics – a clear indication that the imipenem-treated bacteria had developed cross-resistance, which eventually leads to multi-drug resistance. By contrast, none of these genes were upregulated in the polymer-treated bacteria. These results, together with the other studies, strongly reinforce the polymer’s promise as an effective, broad-spectrum and long-lasting antibacterial agent.
Looking at outbreak genomes
Genome technology’s dramatic drop in price over the last few years has a lot to offer for the study of infection, says Florez de Sessions. She adds that some of GIS’s current work supports economic arguments for using genomic analysis to isolate and halt superbug outbreaks. A recent GIS study showed that whole-genome sequencing was very effective at revealing the development of a Streptococcus outbreak in a male ward in a 200+ patient psychiatric hospital4.
“Even though the upfront sequencing costs are more, if you have an outbreak in a hospital and say you have to keep screening and swab all patients once a month for six months and they’re approximately swabs, now, suddenly, by sheer numbers and sheer amount, it ends up being cheaper to do whole-genome sequencing.”
While Florez de Sessions leads the GIS Efficient Rapid Microbial Sequencing (GERMS) platform, GERMS consists of three labs: Swaine Chen’s lab focuses on infectious diseases; Niranjan Nagarajan’s lab on computational and systems biology of microbes; and Florez de Sessions’ lab on microbial sequencing.
Chen’s lab was called upon in 2015 when Singapore had an outbreak of Group B Streptococcus. Throughout that year, a severe and invasive strain of this bacteria caused serious infections including meningitis, septic arthritis and spinal infection in over 300 people who were often young and otherwise healthy. However, many of these patients had recently consumed raw freshwater fish. After fishmongers and food handlers were screened and shown not to be carriers of the superbug, Chen's lab led the GERMS genome sequencing effort to identify the bacteria causing the infection5.
By analyzing the genome sequences of bacteria isolated from sick patients and fish dishes, Chen and his team provided genomic evidence that traced the infections to two types of fishes, Song (Asian bighead carp) and Toman (snakehead fish), which are used in yu sheng, a raw fish dish often served with congee. This corroborated traditional epidemiological studies suggesting that the infections were emanating from those consuming contaminated raw fish. To prevent future outbreaks, Singapore’s National Environment Agency has now banned food outlets from serving raw freshwater fish dishes.
- Nanomedicine superbug breakthrough hailed as ‘world changing’
- Triple-threat mixture overcomes drug-resistant lung infections
The brave new world of macromolecular therapeutics
GIS and gene sequencing will also be key to the next stage of research on the polymer. Hedrick has described the March paper as the beginning of a whole new class of treatments, which he calls 'macromolecular therapeutics'. While Hedrick may be correct in describing their findings as seminal, the field’s newness means that to move toward drug development, Yang and GIS will have to continue to rigorously test the polymer, which they are doing in collaboration with the U.S. National Institutes of Health.
However, the hope is that, just as genome technology accelerates new discoveries, it can help to speed up the vetting of drugs as well. Florez de Sessions says GIS is keen to harness IBM’s machine-learning technology to look in the treated bacteria sequence data for any other important patterns that might affect a person’s health.
“The premise of this particular study is that the polymer is relatively inert compared to the standard-of-care antibiotic, imipenem or penicillin … or other antibiotics,” she says. “So, you don’t see a resistance, you don’t see some of the things that you expect from antibiotics … but what about the things that we don’t know?” What GIS can do, says Florez de Sessions, is help to show IBM what genome specialists already know about bacterial resistance regulation. Those are already well-annotated responses, she says. GIS and IBM can then run the treated bacteria’s genomic data through a machine-learning program and highlight any other patterns that might be relevant to a person’s health outcome.
As for Yang, she thinks synthetic polymers have the potential to provide a better and cheaper response than one of the main alternatives to antibiotics — peptides (there are others alternatives, including immune modulation, antibody therapy and phage therapy). Peptides are “not actually that stable in vivo” and can easily be degraded by enzymes, she notes. “So they’re high cost and unstable. That’s why there are very few anti-microbial peptides in the clinic, and those peptides are only used in topical applications.” Until now, polymers had been a relatively untapped direction for drug research. “Compared to peptide people there are still very few polymer people — although they're now increasing,” Yang explains.
Yang and her team are looking for partners to start working on the comprehensive preclinical studies they will need before they can start to look at clinical trials.
Note on the term 'superbugs': We’d like to note that, although commonly used, many microbiologists are trying not to proliferate the use and misuse of the term 'superbug'. The Mayo Clinic defines superbugs as “a term used to describe strains of bacteria that are resistant to the majority of antibiotics commonly used today”. In this story, the term refers to bacteria that are resistant to many drugs.
Collaborate with the Genome Institute of Singapore (GIS):
GIS can rapidly pinpoint the source of many viral and bacterial outbreaks by drawing down on its other platforms, such as scientific and research computing led by Chih Chuan Shih, research pipeline development led by Andreas Wilm and the next-generation sequencing platform led by Wendy Soon. They welcome collaborators.
For more information about GIS’s technology platforms, click here.
- Chin, W., Zhong, G., Pu, Q., Yang, C., Lou, W. et al. A macromolecular approach to eradicate multidrug resistant bacterial infections while mitigating drug resistance onset. Nature Communications 9, 917 (2018). | article
- Lam, S., J., O’Brien-Simpson, N., M., Pantarat, N., Sulistio, A., Wong, E., H., H., et al. Combating multidrug-resistant Gram-negative bacteria with structurally nanoengineered antimicrobial peptide polymers. Nature Microbiology 1, 16162 (2016). | article
- Nederberg, F., Zhang, Y., Tan, J., P., K., Xu, K., Wang, H., et al. Biodegradable nanostructures with selective lysis of microbial membranes. Nature Chemistry 3, 409 (2011). | article
- Bergin, S., M., Periaswamy, B., Barkham, T., Chua, H., C., Mok, M., Y., et al. An Outbreak of Streptococcus pyogenes in a Mental Health Facility: Advantage of Well-Timed Whole-Genome Sequencing Over emm Typing. Infection Control & Hospital Epidemiology published online 09 May 2018, 1-9. | article
- Kalimuddin, S., Chen, S., L., Lim, C., T., K., Koh, T., H., Tan, T., Y., et al. 2015 Epidemic of Severe Streptococcus agalactiae Sequence Type 283 Infections in Singapore Associated With the Consumption of Raw Freshwater Fish: A Detailed Analysis of Clinical, Epidemiological, and Bacterial Sequencing Data. Clinical Infectious Diseases 64,S145-S152 (2017).| article