Non-proline cis peptide bonds in proteins

Area of Science:

• Structural Biology
• Protein Chemistry
• Computational Biology

Background:

• Peptide bonds in proteins typically exist in a trans conformation.
• Non-proline cis peptide bonds are less common but structurally significant.
• Understanding cis peptide bond geometry is crucial for accurate protein structure refinement.

Purpose of the Study:

• To identify and characterize non-proline cis peptide bonds in a protein dataset.
• To analyze the geometrical parameters and conformational preferences associated with these bonds.
• To develop a computational method for detecting overlooked cis peptide bonds.

Main Methods:

• Analysis of a non-redundant protein set (571 proteins) from the Brookhaven Protein Data Base.
• Examination of geometrical parameters (bond angles, dihedral angles) of identified cis peptide bonds.
• Comparison with small molecule cis amide bonds from the Cambridge Structural Data Base.
• Development of an algorithm based on deviations in geometrical parameters.

Main Results:

• 43 non-proline cis peptide bonds were identified.
• Specific deviations in bond angles and striking preferences in main-chain dihedral angles (beta-region, alpha-helix start) were observed.
• Intimate side-chain interactions and occurrence in functionally important regions (e.g., active sites) were noted.
• Higher frequency in high-resolution structures suggests underestimation of their prevalence.

Conclusions:

• Non-proline cis peptide bonds exhibit distinct structural features and conformational preferences.
• A new set of parameters is proposed for refining protein structures with cis peptide bonds.
• The developed algorithm can help identify potentially missed cis peptide bonds in protein databases.
• These bonds may be more prevalent than previously recognized, particularly in carbohydrate-related proteins.