Design of 8-mer peptides that block Clostridioides difficile toxin A in intestinal cells

Infections by Clostridioides difficile, a bacterium that targets the large intestine (colon), impact a large number of people worldwide. Bacterial colonization is mediated by two exotoxins: toxins A and B. Short peptides that can be delivered to the gut and inhibit the biocatalytic activity of these toxins represent a promising therapeutic strategy to prevent and treat C. diff. infection. We describe an approach that combines a Peptide Binding Design (PepBD) algorithm, molecular-level simulations, a rapid screening assay to evaluate peptide:toxin binding, a primary human cell-based assay, and surface plasmon resonance (SPR) measurements to develop peptide inhibitors that block Toxin A in colon epithelial cells. One peptide, SA1, is found to block TcdA toxicity in primary-derived human colon (large intestinal) epithelial cells. SA1 binds TcdA with a KD of 56.1 ± 29.8 nM as measured by surface plasmon resonance (SPR).

The lowest scoring peptides obtained from Cases 2 and 3 are SA3 and SA6. Figure 1A and 1B show examples of the plot of the Score (Γscore) and the RMSD profile v/s the number of sequence and conformation change moves performed in PepBD for Cases 2 and 3.The SA3:TcdA GTD catalytic domain structure obtained from the PepBD algorithm is shown in Figure 1C.Peptide SA3 has a Γscore = -43.21,obtained at the 9401 th evolution step.Likewise, SA6 has a Γscore = -44.37 which is obtained at the 7415 th step of the evolution process.The structure of SA6 complexed with the TcA GTD catalytic domain is shown in Figure 1D.Supplementary Table 3.Comprehensive list of peptide sequences obtained from PepBD, with their corresponding Case, random seed, Γscore, as well as if they were evaluated in MD simulations.Since conformational changes can take place over long time scales, we selected SA1:TcdA GTD, SA2:TcdA GTD, SA4:TcdA GTD and SA5:TcdA GTD complexes for performing 500 ns MD simulation each.We calculated the RMSD (root mean squared deviation) with the starting input structure of the peptide:TcdA GTD as the reference structure, over the course of the simulation (Figure 2).The RMSD was recorded every 1 ns and was mass-weighted on the backbone atoms.To determine if there are any significant conformational changes over the course of the simulation, we performed an unpaired t-test on the 96-100 ns time frame and the 101-500 ns time frame (Note that the 96-100 ns timeframe was selected to evaluate the ∆  ).The unpaired t-test confirmed that the difference in the RMSD of the peptide:protein complex over the 95-100 ns time frame and 101-500 ns timeframe is not statistically significant (with 95% confidence), hence, confirming there were no siginificant conformational changes on the peptide:protein complexes over long time scales.The mean and the standard deviation of the RMSD for the 96-100 ns and 101-500 ns timeframes are reported in Table 4.We also measured the % secondary structure assignment of the peptides at the 100 th , 200 th , 300 th , 400 th and 500 th ns over the course of the 500 ns simulation of the four peptides bound to TcdA GTD.Investigation of the biorecognition mechanism of specific residues on the Toxin A neutralizing peptides will help us understand what the key amino acid interactions are on the peptide:TcdA GTD complexes and also support future efforts to design peptides targeting TcdA.When evaluated experimentally, SA1 neutralized toxin A in both primary-derived human jejunum small intestinal and colon epithelial cells and exhibited a KD of 56.1  29.8 nM measured by surface plasmon resonance (SPR).Hence, to draw further insights as to which SA1 amino acids play key roles in recognizing specific residues on TcdA GTD, we examine the residue-wise decomposition of the interaction energy between SA1 and the catalytic site of TcdA GTD.In Figure 3C of the Main Text we plot the residuewise decomposition of the interaction energy between SA1 and the catalytic site of TcdA GTD.In Figure 3A we construct an energy panel detailing the pair wise interactions of SA1:TcdA GTD complex.

Peptide
Figure 3B shows the number of contacts that each residue on peptide SA1 forms with the TcdA GTD at the binding site.A contact is defined to occur when the distance between a residue on the peptide and a residue on the receptor is ≤ 4.5 Å.

MD Simulation parameter Value Simulation box dimesion
12 Å Total number of residues (6 mer peptide:protein complex) 545 Total number of residues (7 mer peptide:protein complex) 546 Total of residues (8 mer peptide:protein complex) 547 Total water molecules ~36,000 Total number of atoms ~117,000

Supplementary Figure 1 .
The score/RMSD vs the number of sequence and conformation steps for (A) Case 1 and (B) Case 2 with distinct initial random seeds results and their corresponding top peptides (C) SA3: QEWMGRHW and (D) SA6: EGWQHRHR obtained from PepBD.

Supplementary Figure 2 .
RMSD (root mean squared deviation) caclulation of SA1:TcdA GTD, SA2:TcdA GTD, SA4:TcdA GTD and SA5:TcdA GTD complexes over the course of 500 ns MD simulation.The reference structure for the RMSD calculation is the starting input structure for each MD run.Snapshots were generated every 1 ns in the simulation.
Figures 3C of the Main Text, 3A and 3B reveal that the critical residues on SA1 involved in TcdA GTD binding are Trp3, Trp4, Arg5, Arg6, His7 and Asn8.Thus, the four amino acids: tryptophan, arginine, histidine, and asparagine on SA1 play a key role in binding to the TcdA GTD.Phe2 also contributes significantly to the interaction energy but has a slightly lower contribution than the residues mentioned above.It is also worth noting that peptide SA1 contains three aromatic residues (Phe2, Trp3, Trp4), which reflects the importance of aromatic side chains for peptide:TcdA GTD binding.Trp2 and Trp3 both form strong polar-π interaction with Asn513 while cationic Arg5 forms an ionic bond with anionic Glu509.Arg6 forms a cation-π interaction with Trp516 and interacts via coulombic forces with Asn513.His7 and Asn8 form strong polar bonds with Glu509 and Asn330 respectively.The "Number of Contacts" plot also reveals that Trp3, Arg5 and Arg6 form the most contacts with the TcdA GTD at the catalytic domain, further highlighting the importance of tryptophan and arginine in TcdA GTD binding.Glu1 on SA1 does not contribute significantly to the interaction energy but might be necessary to maintain the conformational stability of the peptide, as it can form strong anion-π interactions with neighboring Phe2, Trp3 and Trp4 on the peptide chain.It appears that the 3 three aromatic residues (Phe2, Trp3 and Trp4) also interact with each other on the peptide via π-π interactions and maintain the conformational stability of the peptide.

Table 4 .
The mean and the standard deviation of the RMSD for the 96-100 ns and 101-500 ns timeframes for peptides SA1, SA2, SA4 ans SA5.

Table 5
below summarizes the secondary structure assignment of the peptides.As is expected from short 8-mer peptides, the peptides mostly remain in "coil" or "turn" configuration and at certain points as "3-10 helix".Supplementary

Table 5 .
Secondary structure assignment of peptides SA1, SA2, SA4 and SA5 over the course of