Atom-level backbone engineering preserves peptide function while enhancing stability

Mingzhu He, Kai Fan Cheng, Tuan A. Vu, Marcelo D. T. Torres, Stavros Zanos, Ibrahim T. Mughrabi, César de la Fuente‐Núñez, Yousef Al‐AbedbioRxiv 2025 · DOI

Abstract

Peptide therapeutics offer unmatched potency and selectivity but are limited by rapid proteolysis and poor pharmacokinetics. Backbone engineering provides a rational approach to enhance stability while preserving function, yet direct comparisons across strategies remain scarce. Using bradykinin as a model, we systematically evaluated four backbone modifications—D-amino acid substitution, N-methylation, α-methylation, and azapeptide incorporation. Each modification produced distinct outcomes in synthesis, conformation, proteolytic stability, and receptor pharmacology. While D- and N-methyl substitutions yielded high stability, they compromised receptor binding and in vivo function. In contrast, the azapeptide analogue maintained native-like affinity and physiological activity while achieving an enhanced stability profile. These findings highlight the need to balance stability and function in peptide design and position azapeptides as an underexplored class with strong therapeutic potential. More broadly, this study establishes a framework for systematic, data-driven peptide design and optimization.

Introduction

Peptides occupy a unique therapeutic niche, combining the high specificity of proteins with the potential for chemical optimization. However, their clinical utility is often hampered by an "Achilles' heel": rapid enzymatic degradation. The human body is replete with proteases that recognize the standard peptide backbone, often leading to half-lives measured in minutes. To overcome this, researchers employ backbone engineering—modifying the peptide's "spine" at the atomic level to evade enzymatic recognition.

Despite the variety of available modifications, head-to-head comparisons of different strategies on a single peptide scaffold are rare. This study addresses this gap by systematically evaluating how different backbone changes affect synthesis efficiency, conformational stability, receptor binding, and in vivo efficacy.

Methods: Systematic Backbone Engineering

The researchers utilized Bradykinin (BK), a nonapeptide hormone critical for blood pressure regulation, as a model system. They identified a proteolytic hotspot at the 8th position ($Phe^8$), which is susceptible to cleavage by kininases. Four distinct backbone modification strategies were applied at this site:

  1. D-amino acid substitution: Inverting the stereochemistry of the $Phe^8$ residue.
  2. N-methylation: Adding a methyl group to the backbone nitrogen to reduce hydrogen bonding and increase steric hindrance.
  3. $\alpha$-methylation: Adding a methyl group to the central $\alpha$-carbon to increase steric bulk and restrict backbone mobility.
  4. Azapeptide incorporation: Replacing the $\alpha$-carbon with a nitrogen atom, creating a non-natural backbone that increases rigidity.
Overview of systematic backbone engineering strategies applied to Bradykinin.

The analogues were synthesized using microwave-assisted Solid-Phase Peptide Synthesis (SPPS). The team then conducted a comprehensive evaluation, including ex vivo stability in mouse whole blood, Circular Dichroism (CD) for structural analysis, B2 receptor (B2R) binding assays, and in vivo hypotensive effect measurements in C57BL/6 mice.

Key Results and Discussion

The comparative analysis revealed a stark trade-off between proteolytic stability and biological function. While all modifications improved the peptide's half-life compared to native Bradykinin ($t_{1/2} = 3.9$ min), the most stabilizing modifications often proved the least effective biologically.

Summary of synthesis, stability, binding affinity, and in vivo activity of Bradykinin analogues.

The Stability-Function Trade-off

D-amino acid substitution and N-methylation provided the greatest stability gains ($t_{1/2}$ of 51.4 and 46.3 minutes, respectively). However, these changes significantly compromised receptor interaction. The D-Phe analogue showed a nearly 60-fold decrease in binding affinity ($IC_{50} = 400$ nM vs. 6.4 nM for native BK), and the N-methylated version showed no inhibition at 1 $\mu$M. CD spectra indicated that these modifications significantly altered the peptide's conformational ensemble, explaining the loss of receptor recognition.

Azapeptides: The "Goldilocks" Modification

The azapeptide analogue emerged as the superior candidate. By replacing a single carbon with a nitrogen atom, this modification achieved an 8-fold increase in stability ($t_{1/2} = 28.6$ min) while maintaining a binding affinity ($IC_{50} = 4.3$ nM) nearly identical to the native peptide. Crucially, in live mouse models, only the azapeptide preserved the significant blood-pressure-lowering effect (hypotension) characteristic of Bradykinin.

Synthesis Considerations

While azapeptides offer a balanced profile, they present greater synthetic challenges. The synthesis time for the aza-analogue was 4 hours, significantly longer than the 1 hour required for the other modifications. This highlights a practical hurdle in the broader adoption of azapeptide technology.

Conclusion and Future Directions

This study demonstrates that subtle, atom-level changes to the peptide backbone can have profound effects on therapeutic potential. Azapeptides, though synthetically more demanding, provide a promising route to stabilize peptides without sacrificing potency. Future research will focus on automating azapeptide synthesis to improve throughput and exploring multi-site modifications to further extend the therapeutic window of peptide drugs. This systematic framework provides a roadmap for the rational design of the next generation of peptide therapeutics.

Summary updated

Abstract

Peptide therapeutics offer unmatched potency and selectivity but are limited by rapid proteolysis and poor pharmacokinetics. Backbone engineering provides a rational approach to enhance stability while preserving function, yet direct comparisons across strategies remain scarce. Using bradykinin as a model, we systematically evaluated four backbone modifications-D-amino acid substitution, N-methylation, α-methylation, and azapeptide incorporation. Each modification produced distinct outcomes in synthesis, conformation, proteolytic stability, and receptor pharmacology. While D- and N-methyl substitutions yielded high stability, they compromised receptor binding and <i>in vivo</i> function. In contrast, the azapeptide analogue maintained native-like affinity and physiological activity while achieving an enhanced stability profile. These findings highlight the need to balance stability and function in peptide design and position azapeptides as an underexplored class with strong therapeutic potential. More broadly, this study establishes a framework for systematic, data-driven peptide design and optimization.
BibTeX
@misc{he2025atom,
  author = {Mingzhu He and Kai Fan Cheng and Tuan A. Vu and Marcelo D. T. Torres and Stavros Zanos and Ibrahim T. Mughrabi and César de la Fuente‐Núñez and Yousef Al‐Abed},
  title = {{Atom-level backbone engineering preserves peptide function while enhancing stability}},
  year = {2025},
  doi = {10.1101/2025.10.06.678421},
  howpublished = {bioRxiv},
}
Figure