Antibiotics: Halicin is a New Class of Antibiotics discovered by Artificial Intelligence
A large proportion of what we read today is about the current health and economic impact of the Coronavirus COVID-19. So, it is easy to miss other important newsworthy scientific discoveries.
Today I am writing a short piece on antibiotics! A large proportion of what we read today is about the current health and economic impact of the Coronavirus COVID-19. So, it is easy to miss other important newsworthy scientific discoveries.
On February 20th (in the scientific journal ‘Cell’) a new research article was published about the discovery of a completely new class of antibiotic called ‘halicin’. Halicin is named after Hal, the astronaut-bothering AI in the film 2001: A Space Odyssey. This breakthrough did receive some exposure by the press but was well hidden as the impending tide of COVID-19 drew closer to our shores. What was even more remarkable about this new antibiotic is that it was discovered using artificial intelligence. Halicin as a discovery also represents a milestone as it is the first completely new class of antibiotics to be discovered for decades.
The Lack of innovation in antibiotics has been of Huge Concern
Clearly this comes at a time when novel antibiotics are becoming increasingly difficult to find and when drug-resistant bacteria are a growing global threat. What is quite startling is that the WHO estimates that antibiotic treatments add an average of 20 years to all our lives. But in the 80 years since the discovery of penicillin, our overuse of antibiotics has forced bacteria to evolve resistance, leading to the emergence of untreatable superbugs that threaten the basis of modern medicine.
The Interagency Coordination Group (IACG) on Antimicrobial Resistance convened by United Nations released a report in 2019 estimating that drug-resistant diseases could result in 10 million deaths per year by 2050. Despite the urgency in the search for new antibiotics, a lack of financial incentives has caused pharmaceutical companies to scale back their research.
At the beginning of the last century infectious diseases in general remained the leading cause of death, accounting for 25% of England’s mortality. However, by the 1950’s, this figure has shrunk to just under 1% after the commercialisation of antibiotics. The golden age of antibiotic discovery lasted from the 1930-1970’s. Selman Waksman received the Nobel prize in 1952 for his discovery of streptomycin, the first antibiotic effective against tuberculosis. Thereafter pharmaceutical companies went into overdrive until the 1970’s. This hay day has since been followed by a 30-year slump with a sharp drop in the number of antibiotics being discovered. So, currently the pipeline for new antibiotics is looking lacklustre.
It can cost billions of pounds and take many years to produce a new antibiotic. Even then shelf life is short as resistance mechanisms in bacteria develop as soon as clinical introduction. This means many pharmaceutical companies would rather focus on developing other treatments, which have better profit margins and more likely to be sustained. No new classes of antibiotics have been invented for decades. The lack of innovation has meant that some scientists are fearful of the return to the pre-antibiotic era because of the increasing prevalence of multi-drug resistant superbugs.
Antibiotics are the cornerstone of Modern Medicine
Antibiotics are a prerequisite for today’s high-technological healthcare. We typically associate antibiotics with successful treatment of chest infections, urinary tract infections (UTIs) or gastrointestinal infections but they are of course much more than that. Without them it would no longer be possible to perform organ transplants, cancer chemotherapy, intensive care and other surgical procedures, such as the hip and knee replacements we commented on two weeks ago.
The use of antibiotics even filters down to the treatment of COVID-19. One of the several serious complications that COVID-19 can cause in a low percentage of serious cases is sepsis. Sepsis is caused by the body’s own response (our immune system) to an infection that leads to organ damage and failure. Sepsis is already one of the most common causes of death in this country. Sepsis, even without COVID-19, is responsible for roughly 44,000 deaths each year and many of these are now directly related to the rising threat of antibiotic resistant infections.
The Artificial Intelligence behind the New Discovery of Halicin
The newly designed artificial intelligence (AI) tool used in this new research study has identified a molecule capable of wiping out several antibiotic-resistant strains of bacteria. What is interesting is the method by which this molecule was discovered.
The use of AI, it appears, can significantly speed up the process of research and development thereby rationalising development costs. Using a sophisticated algorithm, the AI was unleashed on vast digital libraries of pharmaceutical compounds and delivered results in a timeframe that was previously unimagined. The algorithm learned to predict molecular function without any assumptions about how drugs work and without chemical groups being labelled. As a result, the model can learn new patterns unknown to human experts. James Collins, one of the co-authors of the study from MIT concluded; “I do think this platform will very directly reduce the cost involved in the discovery phase of antibiotic development. With these models, one can now get after novel chemistries in a shorter period involving less investment.” Anything that can speed up early-stage drug discovery has the potential to make a huge impact.
The AI that was used for the project learned specific atomic and molecular features of 2,335 molecules for which the antibacterial activity was known and related this to their efficacy against a common pathogen. These molecules included 300 approved antibiotics and 800 natural products from plant, animal and microbial sources. Once the model was trained, the researchers used it to monitor a library called the Drug Repurposing Hub, which contains around 6,000 molecules under investigation for human diseases. The researchers asked the AI to predict which would be effective against E. coli, and to show them only molecules that look different from conventional antibiotics.
One of the 100 candidates that were returned turned out to be halicin.
As confidence in the AI algorithm increased the team then screened more than 107 million molecular structures in a database called ZINC15. From a shortlist of 23, physical tests identified 8 with antibacterial activity. Two of these had potent activity against a broad range of pathogens and could overcome even antibiotic-resistant strains of E. coli. This approach highlights the potential power of computer-aided drug discovery. It would be impossible to physically test over 100m compounds for antibiotic activity using conventional methods.
Antibiotics work in various ways, such as blocking the enzymes involved in cell-wall synthesis, DNA repair or protein synthesis. But halicin’s mechanism is unconventional: it disrupts the flow of protons across a cell membrane. In initial animal tests, it also seemed to have low toxicity and be robust against resistance. In experiments, resistance to other antibiotic compounds typically arises within one or two days. Remarkably with halicin even after 30 days of testing the researchers could see no evidence of resistance.
Halicin is Effective against high priority Pathogens
Halicin had previously been researched as a potential treatment for diabetes. Amazingly, it also demonstrates an antimicrobial capacity against Mycobacterium tuberculosis (the causative pathogen of tuberculosis), and several other resistant pathogens including Acinetobacter baumannii and Enterobacteriaceae (common causes of post-operative infection and sepsis).
These are considered high-priority pathogens that the World Health Organization ranks as critical for new antibiotics to target. James Collins one of the co-authors of the study comments; “I think this is one of the more powerful antibiotics that has been discovered to date. It has remarkable activity against a broad range of antibiotic-resistant pathogens.”
You may not have heard of Acinetobacter baumannii. However, it is one of the most common pathogens associated with hospital-acquired infections worldwide. Most infections occur in critically ill patients in intensive care unit (ICU) settings. It accounts for up to 20% of infections in ICUs worldwide. The WHO has declared that it is one of the most serious pathogens labelled as ESKAPE organisms. This is an acronym for the six pathogens that are the leading cause of healthcare-acquired infections worldwide. Antibiotic resistance in ESKAPE organisms is usually associated with significant higher morbidity, mortality, as well as economic burden. It just so happens that another one of these six is also Enterobacteriaceae. Some strains of Enterobacteriaceae for instance are resistant to carbapenems, a group of antibiotics that are considered the last resort for infections.