Protein Modeling

Our team expanded on a computational protein design pipeline developed by Washington iGEM 2024 and optimized it to design novel protein minibinders targeting a specifically chosen epitope in Domain I of Exotoxin A. This pipeline consists of the deep learning tools RFdiffusion, ProteinMPNN, and AlphaFold3. Our documentation specifically outlines the iterative process of 1) choosing a binding epitope and specific hotspots on our target protein, 2) running full or partial diffusion, 2) selecting a subset of designed backbones to generate sequences for 3) forward-folding sequences and collecting descriptive statistics, 4) visually analyzing and comparing forward-folded designs with original backbones, and 5) calculating additional statistics for analysis using a custom script.
This specific pipeline is well-suited for student research because it closely resembles higher-throughput computational methods used in labs such as the Institute for Protein Design at the University of Washington but is scaled down to be managed easily with less computing power. With computational power being a limitation for many computational protein design methods, our method maximizes the number of high-quality candidates generated without the need for a high-performance computing cluster. Though our pipeline cannot entirely bridge the gap in throughput between that of a computationally-driven research lab and an iGEM team such as ours, we have optimized our protocols to work as well as possible with a limited amount of resources.
From this work, our protein modeling team designed 96 entirely novel proteins which showed strong in silico binding metrics which were then ordered by our wetlab team for in vitro testing. Beyond our project, we hope that this document provides useful information for other iGEM teams or research groups to design novel proteins for a variety of applications outside of ExoA inhibition.

Our team developed and performed protein inhibitor purification, LRP1 transfection, and mutated and recombinant Exotoxin A purification. View our modeling page to learn more

Protein Inhibitor Purification

For our protein inhibitor purification, we created a protocol to purify our 96 protein inhibitor candidates. We ordered gene fragments from Twist Bioscience and IDT that encode for our mini-proteins. We then cloned the gene encoding for the mini proteins into expression vectors, transfected host cells, lysed the host cells, and then used His-tagging for protein purification.

LRP1 Transfection, Western Blot, and Immunofluorescence

In our LRP1 transfection, we used a purified LRP1 plasmid to introduce the LRP1 protein into human embryonic kidney cells (HEK293T cells). This allows the human cell line to uptake ExoA. In order to validate our transfection results, we performed an immunofluorescence assay and western blot. For the immunofluorescence assay, we used a primary antibody that binds to the LRP1 protein and a secondary antibody conjugated with a fluorophore that binds to the primary antibody to evaluate how well our transfection worked. We then performed a Western blot to confirm the presence of the transfected LRP1 protein in our HEK293T cell line.

Exotoxin A Gene Fragments

In our recombinant ExoA purification, we mutated the toxic ExoA to use as a less toxic form for lab experiments and made ExoA conjugated with fluorospheres for future experiments. We created and purified 5 different recombinant forms of ExoA: 1) a Domain 1a construct. 2) Truncated Domain 1a, without residues 232 - 252 of ExoA. 3) Full-length Exotoxin A with Mutations H440A and E553Δ. 4) Domain 1a fused to mCherry and sfGFP at the C terminus. 5) Domain 1 and 2 fused to mCherry and sfGFP at the C terminus. The process for creating the mutants is the same as our protein inhibitor purification. We ordered gene fragments, cloned the gene encoding for our mutated ExoA into expression vectors, and transfected the host cells. We created these recombinant ExoA proteins to be used in future SPR and ExoA fluorosphere experiments.

For more detailed information on our WetLab experimental design, refer to our experiments page! Beyond our work with Exo A, we hope our investigation can offer a foundation for researchers to better design experiments for testing recombinant protein mini binders.