West Bengal University of Animal and Fishery Sciences

This study investigated the antimicrobial resistance (AMR) profile of Salmonella enterica (n = 446) of poultry origin (n = 1020) collected from West Bengal, India. Salmonella Typhimurium (n = 328) were the most frequently isolated serovar and found resistant to tetracycline (72

Abstract Florfenicol (FFC) has become a significant antibiotic in the global aquaculture industry, particularly in the treatment of bacterial infections in finfish and shellfish. FFC is valued for its broad-spectrum bacteriostatic activity and effectiveness. This comprehensive review consolidates and expands upon existing research, offering in-depth information on FFC application across various aquaculture practices. This review addresses the key pharmacokinetic aspects of FFC, such as its absorption, distribution, metabolism, and excretion, in aquatic species. Significant attention is given to the safety profile of FFC, with discussions on residue accumulation, elimination dynamics, and the established withdrawal periods necessary to ensure consumer safety. Furthermore, this review highlights emerging issues related to FFC in aquaculture, particularly with respect to plasmid-mediated resistance, the dispersal of resistant bacteria and resistance genes, and their impact on the aquatic environment and nontarget organisms, laying a foundation for future research in this critical area. This evaluative assessment offers a novel contribution to the field, serving as an essential reference for ongoing and future studies on the appropriate use and prevention of FFC resistance in aquatic animal medicine.

The increasing prevalence of diseases in aquaculture systems has led to an increased use of veterinary medicinal products, but their pathophysiological effects on fish remain poorly understood. The current study assessed the effects of dietary administration of the antiparasitic drug lufenuron (LN) at the recommended dose (5 mg) and overdose (12.5 mg/kg biomass/day) for seven consecutive days on the safety, residue distribution, and biological responses of Nile tilapia Oreochromis niloticus juveniles. The LN caused a dose-dependent decrease in feed intake, survival, and biomass. Plasma biochemical analyses revealed an elevation in glucose, creatinine, aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase levels, and a decrease in calcium, chloride, and acetylcholinesterase in LN-fed O. niloticus. Haematological deviations, and erythrocyte cellular and nuclear alterations, such as vacuolation, crenation, irregularly shaped, teardrop-like cells, micronucleus, and blebbed and notched nuclei, were observed. Oxidative stress biomarkers, including malondialdehyde, ferric-reducing antioxidant power, and total nitric oxide, increased, while glutathione-S-transferase and catalase levels decreased during administration. Nonetheless, these changes were reversible after stopping the dose. The liver was more susceptible to LN, as indicated by the integrated biomarker response assessment. Residue peaks occurred on day 7 of administration in plasma, muscle + skin, liver, and kidney. Following dose suspension, residues declined in both groups but remained detectable until day 35 post-dosing. The withdrawal period for LN in O. niloticus was estimated to be 2 days, considering a maximum residue limit of 1350 µg/kg. Despite short-term adverse effects, dietary administration of LN appears safe for O. niloticus juveniles at the recommended dose.

Campylobacter jejuni (C. jejuni) is the leading cause of bacterial gastroenteritis in humans. Primarily, contaminated poultry meat and meat products account for nearly 80% of human infections. Reducing C. jejuni colonization of the chicken gastrointestinal tract is a key strategy to limit the risk of foodborne transmission to humans. To this end, outer membrane vesicles (OMVs) derived from C. jejuni were shown to have strong immunogenic potential in both avian and mammalian models. However, a major limitation of OMV-based vaccines against C. jejuni is the presence of lipooligosaccharide (LOS) in the outer leaflet of OMVs, which often confers endotoxicity. Additionally, some strains of C. jejuni also express outer-core structures of LOS that mimic human gangliosides, potentially inducing cross-reactive antibody responses. Although avians are less sensitive to endotoxin exposure, a high number of C. jejuni infections in chickens may cause anorexia and impaired nutrient absorption, leading to reduced feed intake and suppressed growth performance in broilers. Here, we explored a simple approach to address the risk of LOS-mediated endotoxic effects of OMVs using a non-ganglioside-mimicking C. jejuni strain (TGH 9011; Penner serotype O: 3). For this, C. jejuni OMVs were treated with an optimal amount of polymyxin B (PMB), a polycationic peptide known to neutralize LOS activity. We confirmed that PMB treatment of C. jejuni LOS substantially reduces the expression of TNFα and IL-1β genes, hallmarks of LOS-driven endotoxic activity. Furthermore, oral administration of chitosan-coated PMB-treated OMVs (CS-PMB-OMVs) retained their ability to induce functional intestinal secretory IgA (sIgA) while eliciting modest cellular immune responses in immunized chickens. Similar to natural OMVs, CS-PMB-OMVs immunized birds exhibited a marked reduction in cecal C. jejuni colonization following C. jejuni infection compared with unvaccinated controls. These findings demonstrate that PMB treatment does not compromise the inherent immunogenicity of natural OMVs, while effectively mitigating the risk of LOS-mediated endotoxicity.

Precise fish species identification plays a vital role in fisheries management, biodiversity monitoring, and automated aquaculture systems. Vision Transformers (ViTs) have shown impressive capabilities in visual recognition by capturing long-range dependencies; however, their performance in fine-grained fish classification is often constrained by strong inter-class visual similarity and the absence of domain-specific contextual guidance. Although recent instruction-driven vision models attempt to mitigate these issues, they commonly rely on large-scale multimodal pretraining, which limits their applicability to small and specialized datasets.To address these challenges, this paper introduces an Instruction Token Injection Vision Transformer (ITI-ViT) for fine-grained fish species classification. The proposed framework leverages a pretrained BERT model to encode domain-relevant natural language instructions, which are then transformed into a compact set of learnable instruction tokens and injected directly into the ViT token sequence. This design enables instruction-aware self-attention throughout the transformer layers without the need for costly instruction tuning or multimodal pretraining. Using different instruction prompt templates makes the model tougher when dealing with all kinds of images. In tests, ITI-ViT hit an average accuracy of 99.13% and a macro-F1 score of 99.15% with five-fold cross-validation, outperforming an image-only ViT baseline, making it suitable for practical fish classification scenarios.