The COMER web server for protein analysis by homology

Gardnerella fibrinogen-binding protein as a candidate adherence factor

Bacterial vaginosis (BV), a form of vaginal dysbiosis, is associated with numerous adverse reproductive and obstetric outcomes. Gardnerella spp. are among the key bacteria identified in most BV cases. The formation of a polymicrobial Gardnerella-dominated biofilm on the vaginal epithelium is a characteristic diagnostic marker of BV. Gardnerella colonization and biofilm formation indicate a significant adhesion potential, the determinants of which remain unexplored. In this initial approach to identify Gardnerella adhesins, we analyzed the Cna protein located on the G. vaginalis ATCC 14018 cell surface as determined previously. Structure modeling of Cna (designated Grd Cna) revealed that the protein contains N2 and N3 domains with an immunoglobulin (IgG)-like fold, which shows structural homology to the corresponding domains in SdrD and UafA proteins of the microbial surface component recognizing adhesive matrix molecules (MSCRAMMs) family. A single B domain shares structural similarity with the corresponding domain of Sdr proteins. The R region is rich in PKD repeats, while the C-terminal contains a non-canonical LVNTG cell wall sorting motif. The cna gene was predominantly detected in G. vaginalis isolates but was absent in other commonly identified Gardnerella species isolates. The recombinant Grd Cna protein binds dose-dependently to human fibrinogen but does not interact with fibronectin or collagen types I, III, or IV. Cna-positive G. vaginalis cells adhered to immobilized fibrinogen; however, recombinant Cna did not inhibit this binding, suggesting that Cna may not be a major adhesin mediating G. vaginalis adherence to this ECM component.

Web-based GTalign: bridging speed and accuracy in protein structure analysis

Accurate protein structure alignment is essential for understanding structural and functional relationships. Here, we introduce GTalign-web, a web-based implementation of GTalign, a spatial index-driven protein structure alignment tool, designed for accessibility and high-performance structural searches. Benchmarked against the DALI and Foldseek servers, GTalign-web demonstrates superior accuracy while maintaining rapid search times. Its utility is further highlighted in annotating uncharacterized proteins through searches against UniRef30. GTalign-web provides a useful resource for protein structure analysis and functional annotation and is available at https://bioinformatics.lt/comer/gtalign. This website is free and open to all users, and there is no login requirement.

A sophisticated virulence repertoire and colistin resistance of Citrobacter freundii ST150 from a patient with sepsis admitted to ICU in a tertiary care hospital in Uganda, East Africa: Insight from genomic and molecular docking analyses

Sepsis and multidrug resistance comprise a complex of factors attributable to mortality among intensive care unit (ICU) patients globally. Pathogens implicated in sepsis are diverse, and their virulence and drug resistance remain elusive. From a tertiary care hospital ICU in Uganda, we isolated a Citrobacter freundii strain RSM030 from a patient with sepsis and phenotypically tested it against a panel of 16 antibiotics including imipenem levofloxacin, cotrimoxazole and colistin, among others. We sequenced the organism's genome and integrated multilocus sequencing (MLST), PathogenFinder with Virulence Factor analyzer (VFanalyzer) to establish its pathogenic relevance. Thereafter, we combined antiSMASH and PRISM genome mining with molecular docking to predict biosynthetic gene clusters (BGCs), pathways, toxin structures and their potential targets in-silico. Finally, we coupled ResFinder with comprehensive antibiotic resistance database (CARD) to scrutinize the genomic antimicrobial resistance profile of the isolate. From PathogenFinder and MLST, this organism was confirmed to be a human pathogen (p = 0.843), sequence type (ST)150, whose virulence is determined by chromosomal type III secretion system (T3SS) (the injectosome) and plasmid-encoded type IV secretion system (T4SS), the enterobactin biosynthetic gene cluster and biofilm formation through the pgaABCD operon. Pathway and molecular docking analyses revealed that the shikimate pathway can generate a toxin targeting multiple host proteins including spectrin, detector of cytokinesis protein 2 (Dock2) and plasmalemma vesicle-associated protein (PLVAP), potentially distorting the host cell integrity. From phenotypic antibiotic testing, we found indeterminate results for amoxicillin/clavulanate and levofloxacin, with resistance to cotrimoxazole and colistin. Detailed genome analysis revealed chromosomal beta lactam resistance genes, i.e. blaCMY-79, blaCMY-116 and blaTEM-1B, along with multiple mutations of the lipopolysaccharide modifying operon genes PmrA/PmrB, pmrD, mgrA/mgrB and PhoP/PhoQ, conferring colistin resistance. From these findings, we infer that Citrobacter freundii strain RSM030 is implicated in sepsis and resistance to standard antibiotics, including colistin, the last resort.

Prediction of protein assemblies by structure sampling followed by interface-focused scoring

Proteins often function as part of permanent or transient multimeric complexes, and understanding function of these assemblies requires knowledge of their three-dimensional structures. While the ability of AlphaFold to predict structures of individual proteins with unprecedented accuracy has revolutionized structural biology, modeling structures of protein assemblies remains challenging. To address this challenge, we developed a protocol for predicting structures of protein complexes involving model sampling followed by scoring focused on the subunit-subunit interaction interface. In this protocol, we diversified AlphaFold models by varying construction and pairing of multiple sequence alignments as well as increasing the number of recycles. In cases when AlphaFold failed to assemble a full protein complex or produced unreliable results, additional diverse models were constructed by docking of monomers or subcomplexes. All the models were then scored using a newly developed method, VoroIF-jury, which relies only on structural information. Notably, VoroIF-jury is independent of AlphaFold self-assessment scores and therefore can be used to rank models originating from different structure prediction methods. We tested our protocol in CASP15 and obtained top results, significantly outperforming the standard AlphaFold-Multimer pipeline. Analysis of our results showed that the accuracy of our assembly models was capped mainly by structure sampling rather than model scoring. This observation suggests that better sampling, especially for the antibody-antigen complexes, may lead to further improvement. Our protocol is expected to be useful for modeling and/or scoring protein assemblies.