Artificial Activation of Toxin-Antitoxin Systems

Due to the role of plasmids in the spread of multi-drug resistance, I propose a novel strategy to activate internal cell death mechanisms as a novel antimicrobial strategy.

Antibiotic resistance is one of the greatest threats to public health worldwide, and this is often facilitated by the transfer of resistance plasmids through horizontal gene transfer. However, one internal cell death mechanism carried by plasmids are toxin-antitoxin systems, which act as a mechanism carried by plasmids to ensure the survival of the plasmid. This is through post-segregational killing, where all daughter cells that have failed to inherit the plasmid are killed due to the cleavage of the antitoxin.

My solution proposes to artificially activate this system by disrupting the interaction between the protein toxin and protein antitoxin, which will free the toxin from the complex and allow it to act lethally on the cellular target. This can be achieved with a small-molecule inhibitor of the protein-protein interaction.

Through computational drug design, I developed a novel small-molecule inhibitor of the Kid-Kis toxin-antitoxin system by targeting the binding residues responsible for the majority of the binding energy. This novel drug molecule was evaluated for molecular properties and the biological activity was predicted with the construction of a QSAR model.

According to the World Health Organization (WHO), antibiotic resistance is one of the greatest threats to global health today, as a growing number of infections, such as pneumonia and tuberculosis are becoming harder to treat because antibiotic resistance is arising. Every year, there are more than 2.8 million cases of antibiotic resistant infections in the United States alone, and more than 35,000 deaths due to antibiotic resistance. Today, more than 70% of bacteria that cause infections have acquired resistance to at least one antibiotic commonly used to treat the infection, and many bacteria are beginning to develop resistance to last resort antibiotics, such as colistin and vancomycin. However, it has been observed that bacteria often have pre-existing resistance to the main antibiotic targets, which shows that more novel approaches to antibiotic development must be considered in order to combat the issue of antibiotic resistance (Hughes and Karlén 2014).

In the pharmaceutical development industry, plasmids have largely been overlooked as an antimicrobial target due to the lack of suitable antibiotic targets. Because plasmids serve as an accessory to the bacterial genome, plasmids do not contain any components vital to the survival of the cell. Antibiotics are largely targeted towards vital cellular processes to ensure cell death. However, because of this, bacteria have a significant selective pressure to evolve resistance against the antibiotic in order to survive, which has led to the rapid rise in antibiotic resistance towards all novel antimicrobials (Bennett 2008). Yet, plasmids are a promising target in antimicrobial development due to their role in facilitating antibiotic resistance. Plasmids have been found to be major carriers of antibiotic resistance genes and responsible for the rise of many multi-drug resistant infections (Clewell 2014). However, because of the role of plasmids as accessories, there have been no antibiotic targets that have been observed directly linked to the antibiotic resistance genes. But in the 1980s, toxin-antitoxin systems were identified in plasmids, and hypothesized to aid in plasmid competition and ensure the retention of the plasmid in the population. In type II toxin-antitoxin systems, the plasmid encodes a protein toxin and protein antitoxin, which forms a stable complex in cells containing the plasmid. However, if the plasmid was lost in daughter cells during cell division, all plasmid-free daughter cells would be killed through post-segregational killing due to the cleavage of the antitoxin, activating the toxin. Toxin-antitoxin systems have also been identified to be highly prevalent in resistance plasmids, as it was found by Williams et al. that toxin-antitoxin systems were found in all clinical isolates identified. Additionally, toxin-antitoxin systems are suitable targets for plasmids as these systems have been found to be highly prevalent in plasmids conferring antibiotic resistance, including in vancomycin-resistant enterococcus (VRE) and carbapenem-resistant enterobacteriaceae (CRE). In my research, I propose a novel application of these internal cell death mechanisms as a novel antimicrobial strategy, which has broad implications in healthcare and in science.

There are two spectrums of antibiotic activity - broad-spectrum and narrow-spectrum antibiotics. Broad-spectrum antibiotics can kill a large spectrum of bacterial species, regardless of pathogenicity. The importance of broad-spectrum antibiotics lies in instances of prescribing where the specific strain of infection-causing bacteria cannot be identified. In this case, all bacteria targeted by the broad-spectrum antibiotic are killed, including commensal bacteria. However, this can cause devastating consequences on the gut microbiome, as the non-selective killing of bacteria can severely disrupt the microbes present in the human body essential for human health and function. While narrow spectrum antibiotics can target more specific strains of bacteria, today, there have been no antibiotics developed capable of specifically targeting bacteria conferring virulence or resistance. Because plasmids can confer antibiotics resistance or virulence to the bacterial host, plasmids provide a promising target for the specific killing of bacteria conferring resistance or virulence. This antibiotic drug lead provides a new hope for circumventing the issues faced by standard antibiotics, as only bacteria containing resistance plasmids will be killed by the antibiotic, without harming non-resistant commensal bacteria.

As a strategy directly targeting a mechanism facilitating the evolution of antibiotic resistance, this innovative solution to antimicrobial development is unlikely to face resistance and may also serve to limit the spread of antibiotic resistance through horizontal gene transfer. In standard antibiotics, antibiotic resistance can arise when enzymes are produced to degrade the antibiotic, the target of the antibiotic is mutated, or removing the antibiotic from the cell. Because the small-molecule inhibitor does not directly facilitate the death of the cell, resistance mechanisms removing the antibiotic from the cell through efflux pumps or changes in membrane permeability are unlikely to occur. The novel antibiotic itself is not causing toxicity to the cell, but the antibiotic only activates once within the cell to cause cell death through internal mechanisms of the bacteria itself. Therefore, any resistance that develops to prevent cell death would be adaptations against the internal cell death mechanism of plasmids, toxin-antitoxin systems, rather than the antibiotic itself. This is significant because a major issue facing the pharmaceutical industry in antibiotic development is the rapid evolution of resistance towards any novel antibiotic developed even in the preclinical stages of development. With this advancement indirectly causing cell death through the bacteria’s own mechanisms, this could have wide implications as the only antimicrobial to have no risk of detrimental antibiotic resistance arising to the drug. If the bacteria does acquire resistance to the mechanism of cell death, the primary purpose of toxin-antitoxin systems is to ensure the plasmid in the population. If bacteria modify the targets of toxin-antitoxin systems or otherwise evolve resistance to the toxin, the toxin-antitoxin system will not be effective in ensuring plasmid retention. Therefore, the plasmid will be more likely to be lost in the population, re-establishing antibiotic susceptibility to the bacteria. In any case, the risk of resistance arising to this novel antibiotic is much lower than resistance to standard antibiotics targeting vital cellular processes.

Antibiotic resistance has an important personal impact on me, as I began researching antibiotic resistance when my grandmother died of a bacterial infection in 2019. Because of her encouragement of my big dreams, I began researching about the lack of antibiotic development in the field. Eventually, I conducted a literature review and meta-analysis on toxin-antitoxin systems, and proposed the application in antimicrobial development. The meta-analysis and proposal has currently been accepted for peer-review publication in a journal and will be published soon after revision. Since this project involves the pharmaceutical industry, I discussed and worked with experts in the field on my idea and received feedback on how I could improve the drug lead. My project has evolved significantly from peptide nucleic acids to a small-molecule inhibitor since I've worked on this idea for over two years now, and I believe that toxin-antitoxin systems have a great potential in future antimicrobial development due to its role in the horizontal gene transfer of resistance genes. Today, I've worked with professors at my local university and other scientists to develop my idea, and I hope to take it to the next step with MIT SOLV[ED]!

Currently, I've developed a drug lead with promising molecular properties and biological predictions, as it's been predicted by machine learning models and QSAR models to show no toxicity and good biological activity. All molecular properties also fall under suitable Lipinski Rule of Five values.

However, I hope to begin the lead optimization process of the drug lead using computational drug design software such as the Schrodinger software to improve the structure of the drug lead for in-vitro and in-vivo testing. This will be a great advancement for pharmaceutical development, as in-silico testing strategies are capable of saving thousands of dollars in laboratory testing funds by predicting biological activity beforehand. By conducting lead optimization using computational software, this will be able to develop a promising drug lead for future clinical development.

This research incorporates a multidisciplinary approach using computational drug design to develop a drug lead for a promising target identified by the creative application of a discovery in microbiology. This involved computational software including LEA3D, OCHEM (eADMET), PyMOL, Protein Data Bank, SWISS-DOCK, and HADDOCK molecular docking. Computational structure-based drug design has significant promise in reducing the cost of drug development and improving efficiency by reducing the need to test large libraries of compounds. As a novel target identified in plasmids, the in-silico development of a small-molecular inhibitor provides a starting point for further advancement and optimization for further drugs targeting toxin-antitoxin systems. The results of in-silico molecular properties and prediction evaluations shows that the drug lead has values within a suitable range for a starting point of an effective drug, but further modifications can be made to improve the activity and structure of the drug. In-silico predictions are an efficient method of evaluating the in-vitro activity of drug leads before undergoing costly laboratory testing, so because I didn’t have access to a laboratory and could not perform costly laboratory tests, I used a multidisciplinary approach by combining computational biology and microbiology, which provides quantitative evaluations for determining opportunity for advancement when laboratory resources are available for further development. For me, I’m currently working on using in-silico strategies to improve the structure of the drug lead with computational drug design software, but I hope that one day, I can synthesize the drug lead in the laboratory and conduct an in-vitro bioassay!

Antibiotic resistance is a significant issue for global health worldwide, as it's been estimated that by 2050, antibiotic resistance will kill more people than cancer. In addition, antibiotic development has significantly declined since the peak of antibiotic development in the 1950s, as the lack of market incentive in developing new antibiotics have led to a decrease in options for antibiotics as the number of antibiotic resistant bacteria increases. This means that if my research were to reach the market after pharmaceutical development, it could have the potential to help thousands of people who suffer from bacterial infections, particularly for those suffering from multi-drug resistant infections. Currently, the CDC estimates the number of people who get antibiotic resistant infections a year at 2.8 million, and this number is only expected to increase. However, it's important to note that my idea will need time in pharmaceutical development to reach the market, but with the SOLV[ED] program, I would like to use more accessible computational tools to reach that goal.

The goal of my project is to undergo lead optimization for the drug lead that I designed in my previous research targeting the Kid-Kis toxin-antitoxin system in the R1 plasmid using computational Schrodinger software. This project involves free energy perturbation calculations of the ligand to the drug target to determine the binding affinity. Based on the binding affinity, structural changes for the molecule will be made based on electronic and steric features, with further modifications being made to improve the interaction between the binding site and the drug lead. This process will be undertaken with computational drug development, which helps save time and money in pharmaceutical drug development as drugs with poor molecular properties have already been screened out. The reason for implementing this plan is to help  contribute to the field of targeting evolution mechanisms against antibiotic resistance, as in the future, I hope to be able to test the optimized drug lead in the laboratory with an AlphaScreen bioassay. While it's unlikely that I'll be able to develop the drug lead into an actual marketable drug in the near future, the development of a drug lead itself will contribute greatly to research in toxin-antitoxin systems, and further research on these systems will greatly benefit people worldwide as it could serve as a promising strategy to limit multi-drug resistance.

The UN Sustainable Development Goal that I hope to contribute to with my project are Good Health and Well-Being and Industry, Innovation, and Infrastructure. I understand that pharmaceutical drug development is a long process. However, I believe that the strategy of targeting evolutionary mechanisms of antibiotic resistance has great potential in being explored, so I would like to contribute to innovation in the field of antibiotic development by designing a drug lead that has activity in-vitro, later on conducting a bioassay to show proof-of-concept for the artificial activation of toxin-antitoxin systems. My idea has significant promise for innovation, so because of the lack of antibiotic development in the field, in the future, I hope to start a non-profit research organization to develop novel drugs against diseases that are overlooked by the pharmaceutical industry. This research that I'm currently conducting can be taken to the next step when I receive more resources to conduct clinical trials and undergo in-vitro and in-vivo testing, but as of now, using computational tools is a feasible and an economically-viable way of contributing to the UN Sustainable Development goals. I will measure my progress based on computational calculations on molecular properties, including binding affinity and drug-like properties, calculated by Schrodinger software. This is a significant measure of progress because it shows quantitative evaluations for the promise and proof-of-concept for the artifical activation of toxin-antitoxin systems.

The actual drug development process for pharmaceuticals is very long. However, in participating significantly in the citizen science movement throughout my life, this has inspired me to help contribute to pharmaceutical development as a student researcher, even if the chance of reaching clinical trials is very slim at this time. As a student researcher, I hope to provide creative ideas to the field, and contribute as much as I can using computational tools and the guidance from my mentors in the field. However, in actually reaching the market as a drug, this can require thousands or millions of dollars in pre-clinical testing and clinical trials. Therefore, the goals of my project is to contribute to the field and inspire further research in toxin-antitoxin systems, rather than develop the drug as a marketable antibiotic. This will be a significantly more feasible goal, although some barriers I may come across may be with publishing my research. It can be difficult to publish research and receive resources for computational tools as a high schooler, which is why I am applying to the MIT SOLV[ED] program to gain the resources for my project!

The opportunity presented by MIT SOLV[ED] will be incredibly valuable to me since I am very passionate about my research in antibiotic resistance. I began researching toxin-antitoxin systems in bacteria because I was mind-blown by the unique abilities of bacteria to transfer their genes through horizontal transfer. As a critical vector for resistance, I wondered if it was possible to target these vectors of resistance as an antimicrobial strategy. Since the search for antibiotics has exhausted the targets for life processes in bacteria, the idea of targeting toxin-antitoxin systems is a novel and promising strategy. I never imagined that I would have the opportunity to test my idea, so I am incredibly hopeful to potentially be able to see if the concept of targeting toxin-antitoxin system will work.

I’ve read over 200 pieces of scientific literature on the structure and function of Kid-Kis and related toxin-antitoxin systems, and since then, my research has progressed into the development of a small molecule to disrupt the protein-protein interactions between the toxin and antitoxin, which can artificially activate the system to cause cell death. With the MIT SOLV[ED] program, I hope to be able to optimize the drug lead for in-vitro and in-vivo activity and potentially test the concept. I’ve learned many skills in computational biology, including the use of bioinformatics and web servers, the process of de-novo design, data processing, and the construction of QSAR models based on experimental data in PubChem. I have learned skills in computational biology to ensure that I have sufficient data and knowledge to conduct this project as proposed.

I have worked with the University of Washington and the University of Washington, Bothell, as well as other independent scientists to ensure the success of my project. I've also submitted my project proposals to the MIT THINK program and the Davidson Fellows Scholarship, and I hope to gain the resources to continue my project since I'm very passionate about the research I've done!

As a novel approach to antibiotic development, this project qualifies for the HP Girls Save the World Prize because it aims specifically to help antibiotic resistance that is most commonly faced in developing countries. Ramirez et. al. showed that broad-spectrum antibiotics can have severe consequences on the gut microbiome, including a link between early childhood exposure to antibiotics and gastrointestinal, neurological, and immunological damage that results from disruption of the gut microbiota due to antibiotics (Ramirez et. al. 2020). These findings demonstrate the need for novel antibiotic targets to prevent pre-existing resistance and disruption of the beneficial microbiome that populates the human gut, and with my project, I aim to research this issue which will benefit from sustainability funding from the prize.

This project qualifies for The Pozen Social Innovation Prize because it aims to solve the antibiotic resistance crisis which is frequently faced by women and girls, particularly in developing nations. The developpment of a novel antibiotic targeting antibiotic resistance can have a great potential in improving the quality of life for women and girls worldwide, as bacterial infections are a major cause of death. In fact, in Bangladesh, a pandemic of multi-drug resistant has killed thousands of childrens, so the research conducted by my project will have the potential to cure antibiotic resistance with a novel strategy. As I began poring over journals and articles in novel antimicrobial development, I was devastated by the lack of funding and lack of research into solving antibiotic resistance. Our rising antibiotic crisis meant that families were being torn apart, all because the pharmaceutical industry was tied to the confines of capitalism. With my love for investigation, the personal impact antibiotic resistance had on my life inspired my search for overlooked novel drug targets for antibiotics.