Accelerating phage evolution and tools via synthetic biology and machine learning

NIH RePORTER · NIH · R01 · $643,178 · view on reporter.nih.gov ↗

Abstract

Summary Phages, which are the naturally evolved predators of bacteria, may hold the key to combating bacterial pathogens, including the looming threat of multidrug resistant bacteria. Phages are viruses which while harmless to humans and have been successfully engineered as tools to separate, concentrate, and detect their bacterial hosts. Additionally, phages have been used as therapeutic agents to treat patients infected with pathogens resistant to known antibiotics. While the potential benefits of phages are numerous, certain limitations must be addressed in order to fully employ them. The central hypothesis of this proposal is that both top-down and bottom-up approaches can be utilized to design and synthesize novel phages, through a combination of synthetic biology and machine learning. This will result in phage-based tools with increased functionality and customizable host ranges. The rationale for the proposed research is that as the threat of bacterial infections including those with multi-drug resistance continues to grow, phages, which have evolved to efficiently recognize and kill bacteria, will become indispensable tools. Therefore, the ability to rapidly design and engineer new phages for biosensing and therapeutics will be a critical advantage to human health. The proposal contains three specific aims which are supported by preliminary data and cited literature. Aim 1: Site-directed conjugation for advanced phage-based biosensors and therapeutics. Under this aim, phages will be modified with alkyne-containing unnatural amino acids allowing their direct conjugation to 1) azide decorated magnetic nanoparticles, and 2) azide terminated polyethylene glycol. The modifications will allow the development of magnetic phages for bacteria separation and detection, and phages that are more effective therapeutics due to their ability to avoid a patient’s innate immune response, respectively. Aim 2: Decoding phage biorecognition elements using machine learning. In this aim, machine learning will be used to model the binding of phages and their bacterial hosts. The model will enable the prediction of host interactions as well as allow the design and synthesis of novel phage tail fibers which can target specific bacterial isolates. Aim 3: Repurposing phage biorecognition for a broader host ranges. Under the final aim, phage-binding proteins will be replaced with those known to recognize conserved regions of the bacterial LPS, resulting in a phage with a much broader host range. This approach is innovative because it uses top-down characterizations for bottom-up design and synthesis of novel phages. Traditional phage screening methods will be replaced with the rapid synthesis of phages, which are optimized for a particular bacterial isolate. Following the successful completion of the specific aims, the expected outcome is the design and synthesis of phages that can be used to target a selected group of bacteria within Enterobacteriaceae for advanced...

Key facts

NIH application ID
10443537
Project number
5R01EB027895-03
Recipient
CORNELL UNIVERSITY
Principal Investigator
Sam R Nugen
Activity code
R01
Funding institute
NIH
Fiscal year
2022
Award amount
$643,178
Award type
5
Project period
2019-09-16 → 2024-05-31