Acne vulgaris is a prevalent multifactorial skin disorder significantly impacting individuals both physically and psychologically. Characterized by the blockage of pilosebaceous follicle outlets and excessive keratin production, it creates an environment conducive to bacterial proliferation. Traditional approaches to acne treatment often involve antimicrobial and anti-inflammatory agents; however, the increasing interest in cosmeceuticals—products bridging cosmetics and pharmaceuticals—has spurred the search for novel therapeutic agents with enhanced properties. Keratinases, enzymes capable of degrading keratin, hold significant promise in cosmeceutical applications by regulating keratin within skin pores, thereby aiding in the removal of excess keratin that can lead to clogged pores and acne outbreaks. While previous research has largely focused on fungal keratinases, bacterial keratinases offer notable advantages such as faster enzymatic activity, broader substrate specificity, increased stability, and cost-effective production, making them efficient and adaptable for industrial applications. This in silico study aims to address the research gap concerning the in-depth evaluation of bacterial keratinases in anti-acne cosmeceutical formulations by investigating their structural and functional properties and their interactions with standard anti-acne drugs.
Methods
This study employed in silico methodologies to evaluate two bacterial keratinases: one from mesophilic Pseudomonas aeruginosa (6FZX) and another from thermophilic Meiothermus taiwanensis WR-220 (5WSL). Structural analyses involved assessing physicochemical properties, primary, secondary, and tertiary structures using tools like ExPASy-ProtParam, GPMAW, SOPMA, ESBRI, SWISS-MODEL, PROCHECK, ProSA-web, and ERRAT. Functional analyses included protein–protein interaction networks (STRING database), subcellular localization (TMHMM), and domain/motif presence (MEME suite, InterProScan). Finally, molecular docking against adapalene, erythromycin, tretinoin, isotretinoin, and benzoyl peroxide was performed using the CB-Dock2 server, and ligand-protein interactions were analyzed using LigPlot.
Section-wise Key Points
•Physicochemical Characterization: The thermophilic keratinase (5WSL) exhibited a lower instability index and a higher aliphatic index compared to the mesophilic enzyme (6FZX), suggesting greater structural and thermal stability. Notably, 5WSL had an alkaline isoelectric point (pI), which could be advantageous for acne-prone skin that tends to have a higher pH.
•Primary Structure Determination: Both keratinases contained a high percentage of alanine and glycine, amino acids crucial for structural stabilization. The mesophilic keratinase (6FZX) had a higher number of both positively and negatively charged amino acids.
•Secondary Structure Prediction: The mesophilic keratinase (6FZX) showed a higher percentage of α-helix forming amino acids, while the thermophilic keratinase (5WSL) had a greater number of random coils, indicating potentially higher flexibility and functionality.
•Tertiary Structure Prediction and Homology Modelling: The thermophilic keratinase (5WSL) possessed a higher number of salt bridges (Arg-Asp), contributing to its stability. Structural validation using Ramachandran plots indicated that both models were reliable, with over 87% of amino acid residues in the most favored regions. The SWISS-MODEL QMEAN analysis suggested better overall model quality for the thermophilic keratinase (5WSL).
•Determination of Protein–Protein Interaction Network: The PPI network analysis revealed functional associations for both keratinases, with the thermophilic enzyme showing similarity to a protein from Meiothermus cerbereus and the mesophilic enzyme exhibiting more complex interactions, suggesting involvement in a broader range of biological functions.
•Functional Analysis: Both keratinases were predicted to be cytoplasmic. Conserved motifs containing glutamine were identified in both enzymes, potentially enhancing substrate interaction. InterProScan analysis revealed that the mesophilic keratinase (6FZX) has metalloendopeptidase activity and belongs to the peptidase M4 family, while the thermophilic keratinase (5WSL) has serine-type endopeptidase activity and belongs to the Peptidase S8 subtilisin-related family.
•Analysis of Non-covalent Interaction: Analysis using the Protein Contact Atlas provided insights into the non-covalent interactions of both keratinases with their respective bound ligands, highlighting the atomic contacts involved in maintaining protein stability.
•Protein–Ligand Interaction Analysis by Molecular Docking Study: Both keratinases exhibited favorable binding affinities for all tested anti-acne drugs, with adapalene showing the strongest binding affinity (8.5$ kcal/mol for 5WSL and 9.6$ kcal/mol for 6FZX) and erythromycin the lowest.
•LigPlot Analysis: LigPlot analysis identified specific amino acid residues involved in hydrogen bond and hydrophobic interactions with the docked anti-acne ligands for both keratinases.
This in silico evaluation provides compelling evidence for the potential of bacterial keratinases, particularly the thermophilic enzyme from Meiothermus taiwanensis, in anti-acne cosmeceutical applications. The novelty of this research lies in its comparative in silico analysis of mesophilic and thermophilic bacterial keratinases, coupled with a detailed investigation of their molecular interactions with a panel of standard anti-acne drugs. The findings indicate that the thermophilic keratinase possesses enhanced structural stability and functionality, especially under conditions relevant to acne-prone skin (high pH and temperature), and demonstrates strong binding affinity for adapalene, a key anti-acne agent. These insights pave the way for future wet lab validation and industrial experiments to explore the incorporation of thermophilic bacterial keratinase into effective and stable anti-acne cosmeceutical formulations, offering a promising alternative for treating acne and other keratin-related skin disorders.
Link to the study: https://link.springer.com/article/10.1007/s43994-025-00220-x
