Arun Kumar Jaiswal

A bioinformática, hoje, é uma disciplina central na interseção entre a biologia, a computação e a ciência de dados, com impacto crescente à medida que grandes bancos de dados de informação biológica se expandem. O campo já impacta as áreas da saúde, agricultura, meio ambiente e indústria, sendo motor de inovação e descoberta científica. No entanto, seu crescimento, impulsionado pelas eras do Big Data e Inteligência Artificial, traz desafios estruturais e éticos. Buscamos, por meio deste artigo, discutir como preparar uma nova geração de bioinformatas treinados para navegar a próxima década da bioinformática de forma crítica. Destacamos a necessidade de repensar a formação acadêmica no campo, enfatizando as bases biológicas e computacionais do conhecimento, e de apoiar programas de pós-graduação na área, marcados por precarização e escassez de recursos. Argumentamos que a IA, por mais que seja promissora como ferramenta de apoio a análises complexas, não substitui o pensamento crítico e a compreensão fundamental de dados biológicos. Propomos estratégias pedagógicas interdisciplinares, parcerias com o setor privado, e fortalecimento da bioinformática no âmbito da política pública para garantir a formação de novos bioinformatas com perspectivas de futuro. Este artigo, portanto, busca chamar a atenção da comunidade científica e das agências financiadoras de pesquisa para a urgência de formar e valorizar profissionais de bioinformática capazes de impulsionar o Brasil até 2035.

Background: Cytochrome P450 (CYP450) enzymes play a central role in the metabolism of xenobiotics, including plant-derived compounds such as terpenoids. Objectives: This study aimed to predict the molecular interactions of linalool (LIN) and linalyl acetate (LINAct), major constituents of lavender essential oil, with the canine CYP2B11, CYP2C21, and CYP2D15 isoforms, using in silico approaches. Methods: Three-dimensional (3D) models of the target enzymes were generated through homology modeling using SWISS-MODEL and validated based on global model quality estimate (GMQE) and QMEAN Z-score metrics. Ligand structures were optimized in the Molecular Operating Environment (MOE), and pharmacophoric features were analyzed. Molecular docking simulations were performed using AutoDock Vina, followed by visualization of interactions in MOE. Results: LIN and LINAct exhibit favorable binding affinities with all three isoforms, suggesting their potential as substrates or modulators. Hydrogen bonding and hydrophobic interactions were the predominant forces stabilizing the ligand–enzyme complexes. Conclusions: These findings provide a computational basis for understanding the hepatic metabolism of LIN and LINAct in dogs, offering preliminary insights into the role of specific CYP isoforms in their biotransformation.

Syphilis, a global healthcare burden, is a sexually transmitted infection caused by the spirochete Treponema pallidum, a spiral-shaped, Gram-negative obligate human pathogen. Despite its easy identification and treatability, the disease affects over 50 million people worldwide, with 8 million new cases in the 15–49 age group annually, as per the WHO 2024 report. If left untreated, syphilis progresses through its primary, secondary, latent, and tertiary stages, causing severe complications like neurosyphilis, congenital syphilis, and organ damage. The first-line treatment, penicillin, faces challenges, including logistical issues, shortages, allergic reactions, and patient non-compliance. Secondary treatment options are sparse, and there are reported cases of T. pallidum strains resistant to those antibiotics. The absence of an effective vaccine for syphilis has led to efforts to control its spread through sexual education, condom usage, and post-exposure prophylaxis with doxycycline, which raises concerns about antimicrobial resistance (AMR). The continued reliance on penicillin and the increasing rates of doxycycline post-exposure prophylaxis (DoxyPEP) use have both contributed to concerns about AMR development. Recent works pointing to emerging antibiotic resistance and treatment failures highlight the urgent need for new antibiotics to manage syphilis effectively and reduce dependency on penicillin. This review has focused on the shortcomings and limitations of penicillin treatment, recently identified antimicrobial-resistant strains of T. pallidum, and case studies where its application failed to treat the disease adequately.

Gardnerella vaginalis is a Gram-variable bacillus capable of causing bacterial vaginosis, a condition prevalent in reproductive-age women, this bacterium is present in almost 100% of cases and is also considered a gateway to various sexually transmitted infections. This organism exhibits high pathogenicity linked to virulence and resistance genes acquired throughout evolution, showcasing elevated resistance to a broad spectrum of drug classes. This study conducted comparative genomic analyses to identify these genes and correlate their presence with positive Darwinian selection. Additionally, new drug targets were selected through docking and molecular modeling, guided by the heightened antimicrobial resistance exhibited by this microbial species. The available genomes of G. vaginalis were analyzed, and the orthologous genes were delineated and positively selected, whereby 29 groups were found. Of these genes, one of great importance was predicted, Mef(A), which is related to resistance to the macrolide group of antibiotics, which are one of the main choices for the treatment of sexually transmitted infections. Additionally, two potential protein candidates were selected as drug targets. These proteins were linked with a natural compound each and are considered good potential drug targets. The analyses in this study contribute to analyzing the evolution of the species and how resistance genes are related to their permanence as a potential pathogen.

Equine strangles is a prevalent disease that affects the upper respiratory in horses and is caused by the Gram-positive bacterium Streptococcus equi. In addition to strangles, other clinical conditions are caused by the two S. equi subspecies, equi and zooepidemicus, which present relevant zoonotic potential. Treatment of infections caused by S. equi has become challenging due to the worldwide spreading of infected horses and the unavailability of effective therapeutics and vaccines. Penicillin treatment is often recommended, but multidrug resistance issues arised. We explored the whole genome sequence of 18 S. equi isolates to identify candidate proteins to be targeted by natural drug-like compounds or explored as immunogens. We considered only proteins shared among the sequenced strains of subspecies equi and zooepidemicus, absent in the equine host and predicted to be essential and involved in virulence. Of these, 4 proteins with cytoplasmic subcellular location were selected for molecular docking with a library of 5008 compounds, while 6 proteins were proposed as prominent immunogens against S. equi due to their probabilities of behaving as adhesins. The molecular docking analyses revealed the best ten ligands for each of the 4 drug target candidates, and they were ranked according to their binding affinities and the number of hydrogen bonds for complex stability. Finally, the natural 5-ring compound C₂₅H₂₀F₃N₅O₃ excelled in molecular dynamics simulations for the increased stability in the interaction with UDP-N-acetylenolpyruvoylglucosamine reductase (MurB). This research paves the way to developing new therapeutics to minimize the impacts caused by S. equi infections.Communicated by Ramaswamy H. Sarma.

Histoplasmosis is a widespread systemic disease caused by Histoplasma capsulatum, prevalent in the Americas. Despite its significant morbidity and mortality rates, no vaccines are currently available. Previously, five vaccine targets and specific epitopes for H. capsulatum were identified. Immunoinformatics has emerged as a novel approach for determining the main immunogenic components of antigens through in silico methods. Therefore, we predicted the main helper and cytotoxic T lymphocytes and B-cell epitopes for these targets to create a potential multi-epitope vaccine known as HistoVAC-TSFM. A total of 38 epitopes were found: 23 common to CTL and B-cell responses, 11 linked to HTL and B cells, and 4 previously validated epitopes associated with the B subunit of cholera toxin, a potent adjuvant. In silico evaluations confirmed the stability, non-toxicity, non-allergenicity, and non-homology of these vaccines with the host. Notably, the vaccine exhibited the potential to trigger both innate and adaptive immune responses, likely involving the TLR4 pathway, as supported by 3D modeling and molecular docking. The designed HistoVAC-TSFM appears promising against Histoplasma, with the ability to induce important cytokines, such as IFN-γ, TNF-α, IL17, and IL6. Future studies could be carried out to test the vaccine’s efficacy in in vivo models.

Bacteria of the Leuconostoc genus are Gram-positive bacteria that are commonly found in raw milk and persist in fermented dairy products and plant food. Studies have already explored the probiotic potential of L. mesenteroides, but not from a probiogenomic perspective, which aims to explore the molecular features responsible for their phenotypes. In the present work, probiogenomic approaches were applied in strains F-21 and F-22 of L. mesenteroides isolated from human milk to assess their biosafety at the molecular level and to correlate molecular features with their potential probiotic characteristics. The complete genome of strain F-22 is 1.99 Mb and presents one plasmid, while the draft genome of strain F-21 is 1.89 Mb and presents four plasmids. A high percentage of average nucleotide identity among other genomes of L. mesenteroides (≥ 96%) corroborated the previous taxonomic classification of these isolates. Genomic regions that influence the probiotic properties were identified and annotated. Both strains exhibited wide genome plasticity, cell adhesion ability, proteolytic activity, proinflammatory and immunomodulation capacity through interaction with TLR-NF-κB and TLR-MAPK pathway components, and no antimicrobial resistance, denoting their potential to be candidate probiotics. Further, the strains showed bacteriocin production potential and the presence of acid, thermal, osmotic, and bile salt resistance genes, indicating their ability to survive under gastrointestinal stress. Taken together, our results suggest that L. mesenteroides F-21 and F-22 are promising candidates for probiotics in the food and pharmaceutical industries.

Listeria monocytogenes is an important human and animal pathogen able to cause an infection named listeriosis and is mainly transmitted through contaminated food. Among its virulence traits, the ability to form biofilms and to survive in harsh environments stand out and lead to the persistence of L. monocytogenes for long periods in food processing environments. Virulence and biofilm formation are phenotypes regulated by quorum sensing (QS) and, therefore, the control of L. monocytogenes through an anti-QS strategy is promising. This study aimed to identify, by in silico approaches, proteins secreted by lactic acid bacteria (LAB) potentially able to interfere with the agr QS system of L. monocytogenes. The genome mining of Lacticaseibacillus rhamnosus GG and Lactobacillus acidophilus NCFM revealed 151 predicted secreted proteins. Concomitantly, the three-dimensional (3D) structures of AgrB and AgrC proteins of L. monocytogenes were modeled and validated, and their active sites were predicted. Through protein–protein docking and molecular dynamic, Serine-type D-Ala-D-Ala carboxypeptidase and L,D-transpeptidase, potentially secreted by L. rhamnosus GG and L. acidophilus NCFM, respectively, were identified with high affinity to AgrB and AgrC proteins, respectively. By inhibiting the translocation of the cyclic autoinducer peptide (cyclic AIP) via AgrB, and its recognition in the active site of AgrC, these LAB proteins could disrupt L. monocytogenes communication by impairing the agr QS system. The application of the QS inhibitors predicted in this study can emerge as a promising strategy in controlling L. monocytogenes in food processing environment and as an adjunct to antibiotic therapy for the treatment of listeriosis. Graphical Abstract

Biofilms are complex microecosystems with valuable ecological roles that can shelter a variety of microorganisms. Spirochetes from the genus Leptospira have been observed to form biofilms in vitro, in rural environments, and in the kidneys of reservoir rats. The genus Leptospira is composed of pathogenic and non-pathogenic species, and the description of new species is ongoing due to the advent of whole genome sequencing. Leptospires have increasingly been isolated from water and soil samples. To investigate the presence of Leptospira in environmental biofilms, we collected three distinct samples of biofilms formed in an urban setting with poor sanitation: Pau da Lima, in Salvador, Bahia, Brazil. All biofilm samples were negative for the presence of pathogenic leptospires via conventional PCR, but cultures containing saprophytic Leptospira were identified. Whole genomes were generated and analyzed for twenty isolates obtained from these biofilms. For species identification, we used digital DNA-DNA hybridization (dDDH) and average nucleotide identity (ANI) analysis. The obtained isolates were classified into seven presumptive species from the saprophytic S1 clade. ANI and dDDH analysis suggest that three of those seven species were new. Classical phenotypic tests confirmed the novel isolated bacteria as saprophytic Leptospira. The isolates presented typical morphology and ultrastructure according to scanning electron microscopy and formed biofilms under in vitro conditions. Our data indicate that a diversity of saprophytic Leptospira species survive in the Brazilian poorly sanitized urban environment, in a biofilm lifestyle. We believe our results contribute to a better understanding of Leptospira biology and ecology, considering biofilms as natural environmental reservoirs for leptospires.