De novo design of potent inhibitors of Clostridioides difficile toxin B

Robert J Ragotte,John Tam,Sean Miletic,Roger Palou,Connor Weidle,Zhijie Li,Matthias Gloegl,Greg L Beilhartz,Huazhu Liang,Kenneth D Carr,Andrew J Borst,Brian Coventry,Xinru Wang,John L Rubinstein,Mike Tyers,Roman A Melnyk,David Baker

DOI: https://doi.org/10.1101/2024.08.26.609740

2024-08-26

Abstract:is a major cause of secondary disease in hospitals. During infection, toxin B drives disease pathology. Here we use deep learning and Rosetta-based approaches to de novo design small proteins that block the entry of TcdB into cells. These molecules have binding affinities and neutralization IC50s in the pM range and are compelling candidates for further clinical development. By directly targeting the toxin rather than the pathogen, these molecules have the advantage of immediate cessation of disease and lower selective pressure for escape compared to conventional antibiotics. As infects the colon, the protease and pH resistance of the designed proteins opens the door to oral delivery of engineered biologics.

Microbiology

What problem does this paper attempt to address?

This paper attempts to solve the major public health problems caused by _Clostridioides difficile_ infection (CDI). Specifically, the goal of the paper is to design small - molecule proteins that can effectively inhibit _Clostridioides difficile_ toxin B (TcdB). These proteins can prevent TcdB from entering cells, thereby reducing the occurrence and development of the disease.

### Main research questions
1. **Design of highly efficient small - molecule protein inhibitors**:
- Use deep learning and the Rosetta method to design small - molecule proteins from scratch. These proteins can specifically bind to two key receptor - binding sites of TcdB: the Frizzled/TFPI - binding site and the CSPG4 - binding site.
- The designed proteins have high affinity (binding affinity in the pM range) and strong neutralizing ability (IC50 values in the pM range).

2. **Improve the stability of proteins and the possibility of oral administration**:
- Introduce disulfide bonds to enhance the protease stability and pH stability of the designed proteins, enabling them to remain stable in a simulated intestinal environment.
- Study the degradation of these proteins in simulated intestinal fluid to evaluate their feasibility in practical applications.

3. **Verify the neutralizing activity of the designed proteins**:
- Verify the binding ability and neutralizing activity of the designed proteins to TcdB through in vitro experiments (such as SPR, BLI, neutralization experiments, etc.).
- Compare the neutralizing effects of different designed proteins, especially when proteins targeting different receptor - binding sites are used in combination.

4. **Explore the application of biosynthesis platforms**:
- Use probiotic yeasts (such as _Saccharomyces boulardii_) as production platforms to achieve efficient secretion of the designed proteins, providing the possibility for oral administration.
- Evaluate the neutralizing activity of fusion proteins (such as fusion proteins targeting Frizzled and CSPG4 simultaneously), and verify their effects in a dual - receptor - dependent system.

### Research background
- **_Clostridioides difficile_ infection (CDI)**: It causes more than 500,000 cases in the United States every year, among which 30,000 cases are fatal. Existing treatment methods (such as vancomycin and fidaxomicin) have the problem of high recurrence rates.
- **Antitoxin monoclonal antibodies**: For example, Bezlotoxumab, although it can reduce the recurrence rate, it is costly and requires intravenous injection, which limits its wide application.
- **The need for new therapies**: There is a need for new anti - TcdB therapies that do not require intravenous injection, can act directly on the infection site, are reasonably priced, and can be used preventively during an epidemic outbreak without significantly increasing the risk of antibiotic resistance.

### Research significance
- **Directly target toxins**: Unlike traditional antibiotics, these designed proteins directly target TcdB, reducing the evolutionary pressure on pathogens and the development of drug resistance.
- **High - efficiency neutralizing ability**: The designed proteins have very high affinity and neutralizing ability and are expected to become new candidate drugs for the treatment of CDI.
- **Potential for oral administration**: By improving the stability of proteins and using probiotic yeasts for production, the possibility of oral administration is provided, which improves patient compliance and the convenience of treatment.

In general, this paper aims to provide a new and more effective treatment strategy for CDI by designing highly efficient small - molecule protein inhibitors from scratch.