In silico analysis of hypothetical proteins in Pseudomonas aeruginosa PAC1: Structural and functional insights

Abstract

Background: Bacterial genomes contain numerous hypothetical proteins (HPs) with uncharacterized roles. This study used computational methods to identify and predict the functions of such proteins in the Pseudomonas aeruginosa PAC1 strain.

Methods: The PAC1 genome (GenBank: CP053706.1) was analyzed, starting with 828 HPs. Proteins shorter than 50 amino acids (unlikely to form stable structures) were excluded, leaving 807 HPs. Physicochemical properties were assessed to filter unstable proteins, resulting in 272 candidates. Subcellular localization tools predicted cytoplasmic localization for 58 proteins. Functional annotation identified conserved domains, and homology modeling generated 3D structures for proteins with >80 % similarity to known templates. Structural validation and active site prediction were performed to assess biological relevance.

Results: Two HPs, WP_003099663.1 (186 residues) and WP_010793930.1 (455 residues), exhibited structural stability and functional potential. WP_003099663.1 was annotated as a zinc-dependent enzyme involved in carbon dioxide regulation, while WP_010793930.1 was linked to amino acid biosynthesis. Structural models confirmed stable folds, and ligand-binding site predictions highlighted conserved regions, suggesting roles in metabolic pathways.

Conclusion: This study demonstrates a systematic computational approach for characterizing hypothetical proteins in bacterial genomes. WP_003099663.1 and WP_010793930.1 exhibit promising structural and functional features and warrant further experimental investigation.

Keywords: Bioinformatics analysis; Cytoplasmic; Homology modeling; Hypothetical proteins; Pseudomonas aeruginosa.

Fig. 1

Screening and selection of hypothetical proteins (WP_003099663.1 and WP_010793930.1). The figure depicts the sequential process of filtering and selecting HPs from the initial dataset.

Fig. 2

Secondary structure prediction of the hypothetical protein (WP_003099663.1) generated by the PSIPRED server. The figure's yellow, pink, and grey represent strand, helix, and coil structures.

Fig. 3

Secondary structure prediction of the hypothetical protein (WP_010793930.1) generated by the PSIPRED server. The figure's yellow, pink, and gray represent strand, helix, and coil structures.

Fig. 4

Multiple Sequence Alignment of Various carbonate dehydratase proteins with the target WP_003099663.1 protein. The figure was generated using CLC Sequence Viewer version 8.

Fig. 5

Multiple Sequence Alignment of Various 2-isopropyl malate synthase proteins with the target WP_010793930.1 protein. The figure was generated using CLC Sequence Viewer version 8.

Fig. 6

Phylogenetic tree depicting the evolutionary relationship of HPs: A) WP_003099663.1 and B) WP_010793930.1, together with their closely related proteins. The tree was created using CLC Sequence Viewer Version 8. The scale bar shows sequence divergence, while the line segments show genetic variance, with values of 0.004 for WP_003099663.1 and 0.003 for WP_010793930.1.

Fig. 7

Predicted the target protein's three-dimensional structure using SWISS-MODEL. A) WP_003099663.1, refined structure template-based homology modeling structure. B) WP_010793930.1, refined structure template-based homology modeling structure. (Visualized by UCSF Chimera 1.16).

Fig. 8

Evaluation of model quality. A) The Ramachandran plot for the WP_003099663.1 model structure, validated by the PROCHECK server, indicated that 91.7% of amino acid residues were in the most favored regions. B) The Ramachandran plot for the WP_010793930.1 model structure, validated by the PROCHECK server, showed 94.6% of the amino acid residues in the most favored regions.

Fig. 9

Model quality assessment using QMEAN. QMEAN Z-scores for the homology models of (A) WP_003099663.1 (Z-score: 1.71) and (B) WP_010793930.1 (Z-score: 1.03). The Z-scores represent deviations from empirical distributions of high-resolution experimental structures of comparable size. Lower absolute Z-scores indicate closer agreement with native-like geometric properties. WP_010793930.1 (Z-score: 1.03) demonstrates strong global structural reliability, consistent with high-quality models, while WP_003099663.1 (Z-score: 1.71) resides within the “dark grey zone” of acceptable model quality.

Fig. 10

Structural analysis of ligand-binding pockets in WP_003099663.1 and WP_010793930.1. A) The refined structure of WP_003099663.1 exhibits two distinct ligand-binding pockets. Pocket 1 (red) contains the catalytic residues, while pocket 2 (yellow) represents a secondary binding site. B) The WP_010793930.1 structure features seven ligand-binding pockets, differentiated by color: red (pocket 1), yellow (pocket 2), orange (pocket 3), light-blue (pocket 4), green (pocket 5), teal (pocket 6), and blue (pocket 7). Structural visualizations were generated using the PrankWeb server for binding pocket prediction and annotation.