A Diagnostic for Fungal Pathogens Invading Amphibian Populations

Purpose: The purpose of my project is to design a low-cost, rapid diagnostic to detect the presence of the fungal pathogen Batrachochytrium dendrobatidis. This emerging fungal amphibian parasitic pathogen is threatening to wipe out many species of endemic amphibians around the world. The current method involves isolating the amphibian, transporting test samples to an available facility, and performing a DNA amplification step and gel electrophoresis analysis before taking action. While this is going on, the test amphibian is quarantined and can only be released after the diagnosis is made. This method requires a highly specialized laboratory, takes several hours to diagnose, and requires a scientist with technical training to diagnose and analyze each sample. 

Design Criteria: The design criteria is to bio-ethically develop a portable diagnostic that will provide a rapid, visual, and accurate indication of the presence of the fungal pathogen. The diagnostic is cheaper than the best-in-class diagnostic (Polymerase Chain Reaction) and does not require any specialized skills to operate it. This test is non-invasive, does not harm the amphibian, and only requires a skin swab.

​

Methodology: My approach was to engineer a protein that has a high binding affinity to the secreted, pathogenic proteins from Batrachochytrium dendrobatidis via the Ribosome Display method. The pathogenic proteins are secreted from the fungus to mediate infection. The secreted proteins are present initially and during the disease state, thus making them ideal candidates for early and active infection.

​

As the fungal proteins all proved to be bactericidal, I was unable to produce any usable colonies even after using the secretion tag Maltose Binding Protein which should have suppressed the toxicity of the protein. So, I used Green Fluorescent Protein (GFP) as a target protein to validate the Ribosome Display concept for engineering a binding protein (from a four-helix bundle source protein) that is specifically bound only to GFP. Other proteins such as Bovine Serum Albumin (BSA) did not bind to our engineered protein during the pull assay showing that the engineered protein specifically binds to GFP. Additionally, when I directly added the engineered protein to a nitrocellulose strip (Dot Blot) mimicking the lateral flow strip, I identified that only GFP bound here as well.

​

Conclusion: My design is successful because it proved the concept that a protein can be engineered and mounted on a portable lateral flow assay to visually and accurately detect the presence of the fungal pathogen. It meets all the other design criteria. Cost: The PCR method costs around ~ $200 including the labor cost of a researcher, versus the cost of my diagnostic is less than $0.25. Speed: It provides rapid results (a couple of minutes) vs days. Ease of Use: It does not need any specialized lab skills. The diagnostic requires an in-field skin swab that is placed on the test strip. The results visually appear like a home pregnancy strip. Bioethics: The diagnostic is non-invasive and does not require removing the amphibian from the environment. Additionally, the method to produce these test strips does not involve the injuring of any animals as in the production of antibodies. Also, one unexpected but significant finding was that the fungal proteins were bactericidal. The in vitro translated products of the fungal proteins were ~1000 times more lethal than ethanol on a per molecular basis. 

Future Improvements: The planned future improvements are to design and manufacture the physical lateral flow strips using Ribosome Display. My experiments so far proved that such lateral flow strips can be designed for this use case scenario. Developing these test strips would allow researchers the unique ability to test in-field and be able to better protect the amphibian population from collapsing, preserving the rich biodiversity of such locations.

Acknowledgments: I would like to thank my parents for helping me procure the material, my teachers, and friends for providing several suggestions to improve my project, and lastly, Eric Espinosa and Johan Sosa, my mentors, without whom this project wouldn't have been successful.

​

If you would like to read the complete research plan and report, please contact me at adarsh.ambati1@gmail.com, and I will send it to you.