Initial Steps in the Drug-Discovery Pipeline: In-Silico Screening and Biochemical Validation of β-Clamp Peptide Inhibitors

Context & Rationale

This pre-registration is a direct extension of our recently published ResearchHub Journal article, “Structural Basis for Differential Affinity and Competitive Binding of DNA Polymerase Peptides to the Mycobacterial β-Clamp” (https://doi.org/10.55277/ResearchHub.ac502975.1). Author’s note: I realize that it doesn’t appear like this paper is published on the ResearchHub Journal, but I’ve been told that will be fixed fairly soon. I was considering waiting until the the paper had higher visability on the platform, but I spring semester just ended and I wanted to get going with this.

That paper described the first crystal structure and biochemical characterization of the ImuB-derived β-clamp binding hexapeptide QLPLWG, revealing a polar glutamine “anchor” and a conformationally constrained hydrophobic patch that together drive sub-micromolar affinity. The data suggest that the β-clamp–ImuB protein–protein interaction (PPI) is a privileged chemical space for anti-TB lead discovery, yet they leave two fundamental questions open:

Is ImuB’s QLPLWG motif already operating at the upper limit of β-clamp affinity, or can an unbiased search uncover linear sequences that bind even more tightly?
If such higher-affinity linear motifs exist, would they constitute a better starting point for downstream optimization¹ (e.g., cyclization, stapling, N-methylation, or any other peptide-engineering strategy) than the native ImuB-derived peptide sequence?

The nano-grant pre-registration proposed here addresses both of those questions by docking a curated set of 76,223 PPI interface-derived peptides² against the β-clamp, synthesizing the three top scorers, and measuring whether any of them can out-compete ImuB in direct biophysical assays. All experiments will be documented in near-real time in a public ResearchHub Lab Notebook ³, with the raw data permanently archived in Iowa State University’s public DataShare ⁴ repository, ensuring full transparency and community participation.

Why a ResearchHub Nano-Grant?

I could easily cover a $644 experiment from my laboratory’s research budget, but doing so would miss an opportunity to highlight an interesting use case of ResearchHub pre-registration funding mechanism: funding the extension of a study published in the ResearchHub Journal, and carrying it out under a fully open-science workflow. I intend for this small project to be a live demonstration of how open, community-supported science can unfold from publication to follow-on data to subsequent publication.

Also, the timing is ideal for me. With spring semester just finishing and no teaching or committee obligations during the summer, I have a relatively uninterrupted stretch of time to work on this small project.

A. Overall Objective

To identify a tight-binding linear peptide that can engage the mycobacterial β-clamp groove and out-compete the native ImuB peptide sequence. We will (i) dock a curated set of 76,233 protein-interface peptides, (ii) synthesize the three highest scorers, and (iii) measure their affinity and ability to out-compete the native ImuB peptide by ITC and fluorescence polarization.

A linear sequence that outperforms ImuB would become an immediate candidate for any commonly used peptide hit-to-lead strategy^1,5,6: head-to-tail macrocyclization, side-chain stapling, N-methyl or d-amino scans, or peptidomimetic grafting. Conversely, if none surpasses ImuB, we will have established that evolution is already near the affinity ceiling, directing future chemistry toward conformational locking rather than sequence design. Either outcome will be achieved within eight weeks and documented in real time in a public ResearchHub Lab Notebook.

B. Significance

Clinical need. Multidrug-resistant tuberculosis still claims ~1.3 million lives each year⁷.
Target rationale. The β-clamp is essential for chromosomal replication⁸ and for the mutasome that drives de novo drug resistance⁹, making it a double-value target.
Preliminary data. Our recent ResearchHub article¹⁰ revealed the clamp bound to the ImuB hexapeptide QLPLWG with ~0.9 µM affinity, anchored by Gln¹ and a pre-organized hydrophobic cluster.
Key unknown. Has evolution already “maxed out” clamp affinity with QLPLWG, or can a broader search uncover linear motifs that bind even tighter? The answer will dictate whether future medicinal chemistry should focus on sequence design (if a stronger linear epitope exists) or on conformational optimization of the ImuB motif (if it is already at the ceiling).
Translational impact. Peptides have emerged as a versatile therapeutic class capable of engaging targets that small molecules often miss. Their value is reflected in market trends: over the last ten years roughly one in twelve new FDA approvals (~8%) has been a peptide-based drug⁶. Any linear peptide that out-competes ImuB would provide an immediate springboard for further engineering (e.g., macrocyclization, stapling, N-methyl scans, or peptidomimetic grafting) ultimately yielding β-clamp inhibitors with better potency and cellular uptake than the bulky depsipeptide griselimycin¹¹.

C. Research Plan

**Phase 1 – In-silico triage (Weeks 0–2)**

Instead of separate modeling and docking, we will use MDockPeP2_VS², which uses pre-existing peptide structures and docks them in a single workflow.

Input. The 76,223 linear peptides we will screen are not randomly generated but a curated, structure-derived collection generated in the original MDockPeP2_VS study². Briefly, every monomeric protein chain in the Protein Data Bank was fragmented into overlapping 7- to 15-residue windows; fragments whose flanking Cα atoms were ≤ 6 Å apart, which indicates a compact, surface-exposed conformation, were retained. For each fragment the atomic contacts it makes inside its parent protein were stored as a physicochemical “interaction fingerprint.” Redundant sequences were collapsed at 90 % identity and discarded, yielding 76,233 unique peptides that (i) adopt well-defined backbones and (ii) already satisfy a realistic interaction pattern in their native context.
Run. MDockPeP2_VS will be launched on the ISU High Performance Computing cluster. Each peptide typically yields 5–10 poses scored by a knowledge-based energy function. The complete config file and the workflow will be provided in the ResearchHub notebook.
Ranking. The ~1000 best-scoring docked peptides are first clustered with CD-HIT¹² at 70 % sequence identity. Within each cluster we will calculate ESOL logS¹³ (Estimated SOLubility, based on molecular weight and cLogP). Peptides with logS > -4.0 (≈ 50 µM at 25 °C, which is adequate for ITC and FP assays) will be retained. From the surviving set we take the best scoring representative of three diverse clusters for synthesis, ensuring both sequence diversity and workable solubility.

Preliminary data for phase 1. A pilot run on ~1 % of the library (765 peptides) produced several sequences whose docking score exceeded that of the ImuB-derived peptide when both were processed with the same software, validating the pipeline and strongly suggesting we will discover peptides that out-compete ImuB (Table I).

Table I - Vina_pep Docking Results
Model (file)	Calculated affinity¹⁴ (kcal · mol⁻¹) †
ImuB_peptide.pdbqt	–7.2
5awvA_205-214_out.pdbqt	–8.0
3ukmA_195-208_out.pdbqt	–7.8
3vkgA_4242-4254_out.pdbqt	–7.6
5vkwA_201-210_out.pdbqt	–7.6
3lovA_284-294_out.pdbqt	–7.5
† Values are for model 1 of each docking run (more-negative = stronger predicted binding). Affinity is calculated using the same scoring function as Autodock 4.0 (adapted for vina_pep).

Figure 1. A single subunit of the β-clamp dimer with the ImuB, awvA, and 3ukmA peptides bound (red, blue, and green, respectively). These peptides correspond to the top three lines of Table 1. The peptide awvA shows a high degree of binding site overlap with the ImuB peptide, whereas 3ukmA only overlaps at a few key locations. Both peptides are predicted to compete for native mycobacterial protein binding.

Phase 2 – Biochemical validation (Weeks 2–8)

Peptide synthesis. Order the three top ranked peptides from Genscript (≥ 85 % purity, 3 mg each) plus a 6FAM-QLPLWG tracer.

Prepare β-clamp for biochemical assays. Expression and purification will be performed as described in our ResearchHub Journal paper.¹⁰
Isothermal titration calorimetry (ITC). Measure binding to purified M. tuberculosis β-clamp (20 µM, 25 °C).
Fluorescence-polarization (FP) competition. Titrate peptides (10 µM→10 nM) against 400 nM 6FAM-QLPLWG + 500 nM clamp; derive Ki.

A control for non-specific binding in the ITC and FP assay will use a scrambled version of the highest-scoring peptide (same composition, permuted order). A peptide achieving Kd ≤ 0.9 µM will be considered a superior lead; failure to surpass ImuB will suggest that the native motif is already near the affinity ceiling (an unlikely outcome based on preliminary data).

All raw ITC heats, FP traces, processed data and analysis scripts will be uploaded to DataShare, providing a public record from initial FASTA file to quantitative affinity data.

D | Budget (USD $644)

Table 2 - Budget
Item	Cost
Three custom 7–15-mer peptides (3 mg ea)	$120 x 3 = $360
6FAM-QLPLWG tracer (1 mg)	$284
Total requested	$644

E. Expected Outcomes & Impact

Should even one of the three candidate linear peptides bind the β-clamp with low-micromolar affinity and displace ImuB, we will have revealed a previously unknown sequence family that can immediately be advanced through medicinal-chemistry routes such as cyclization, stapling, or N-methyl scanning. A negative outcome would be equally valuable, demonstrating that the native ImuB motif is already near the clamp’s affinity ceiling and that future efforts should focus on conformational locking rather than sequence replacement. In either case, this nano-grant provides decisive, openly shared data at minimal cost, clarifying the most productive next steps for β-clamp inhibitor development.

F. Team & Environment

Role	Investigator	Position	Expertise
PI	Scott Nelson, PhD	Associate Professor, Iowa State University, Ames, IA.	Biochemistry and biophysics of DNA replication enzymes.

Relevant facilities: ISU Protein Facility, and ISU High Performance Computing cluster.

G. Intellectual Property & Patent Strategy

All patentable discoveries generated by this project will be governed by Iowa State University’s IP policy and managed through the Office of Innovation & Commercialization / ISU Research Foundation (see https://www.techtransfer.iastate.edu/). Iowa State will retain ownership of the intellectual property and will decide, at its discretion, to file U.S. and foreign patent applications covering any high-affinity peptide sequences that are discovered. The Principal Investigator will submit invention disclosure forms to the Office before public release of sequence information in the ResearchHub Lab Notebook or Journal, ensuring that provisional patents can be filed in a timely manner. Any future licensing or option agreements will be negotiated by the Office, with revenue sharing following standard ISU policy.

References

1. Otvos, L. The latest trends in peptide drug discovery and future challenges. Expert Opinion on Drug Discovery 19, 869–872 (2024).

2. Xu, X. et al. In silico screening of protein-binding peptides with an application to developing peptide inhibitors against antibiotic resistance. PNAS Nexus 3, pgae541 (2024).

3. Notebook | ResearchHub Docs. https://docs.researchhub.com/researchhub/product-features/notebook (2024).

4. DataShare: ISU’s Data Repository - University Library. https://www.lib.iastate.edu/collections/datashare-isu-data-repository.

5. Otvos, L. & Wade, J. D. Big peptide drugs in a small molecule world. Front. Chem. 11, (2023).

6. Gare, C. L., White, A. M. & Malins, L. R. From lead to market: chemical approaches to transform peptides into therapeutics. Trends in Biochemical Sciences (2025) doi:10.1016/j.tibs.2025.01.009.

7. Global tuberculosis report 2024. https://www.who.int/publications/i/item/9789240101531.

8. Altieri, A. S. & Kelman, Z. DNA sliding clamps as therapeutic targets. Frontiers in Molecular Biosciences 5, 87 (2018).

9. Boshoff, H. I. M., Reed, M. B., Barry, C. E. & Mizrahi, V. DnaE2 polymerase contributes to in vivo survival and the emergence of drug resistance in Mycobacterium tuberculosis. Cell 113, 183–193 (2003).

10. Structural Basis for Differential Affinity and Competitive Binding of DNA Polymerase Peptides to the Mycobacterial β-Clamp. https://www.researchhub.com/paper/9385004/structural-basis-for-differential-affinity-and-competitive-binding-of-dna-polymerase-peptides-to-the-mycobacterial-clamp.

11. Alghamdi, S. et al. Promising Lead Compounds in the Development of Potential Clinical Drug Candidate for Drug-Resistant Tuberculosis. Molecules 25, 5685 (2020).

12. Fu, L., Niu, B., Zhu, Z., Wu, S. & Li, W. CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28, 3150–3152 (2012).

13. Delaney, J. S. ESOL: Estimating Aqueous Solubility Directly from Molecular Structure. J. Chem. Inf. Comput. Sci. 44, 1000–1005 (2004).

14. Trott, O. & Olson, A. J. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J Comput Chem 31, 455–461 (2010).

Initial Steps in the Drug-Discovery Pipeline: In-Silico Screening and Biochemical Validation of β-Clamp Peptide Inhibitors

Initial Steps in the Drug-Discovery Pipeline: In-Silico Screening and Biochemical Validation of β-Clamp Peptide Inhibitors

Funding Progress

Funders

Nonprofit Recipient

Topics

DOI