
Visceral leishmaniasis remains a significant global health burden, with current therapeutic options hampered by high toxicity and the emergence of drug-resistant strains. Thiosemicarbazones, a class of compounds with well-established antimicrobial and antiparasitic properties, represent a promising avenue for novel therapeutics. This study screened 18 thiosemicarbazone derivatives against Leishmania infantum and identified two lead candidates, TBD and TBB, with potent antileishmanial activity. TBD and TBB inhibited promastigotes (IC50 = 5.43 μM and 3.14 μM, respectively) and axenic amastigotes (EC50 = 8.22 μM and 16.11 μM), with favorable selectivity indices in human peripheral blood mononuclear cells (SI = 51.23 for TBD; SI = 9.58 for TBB). Cell death assays revealed that TBD induced dose-dependent apoptosis in promastigotes, with late-apoptotic populations increasing from 5.75% at 1× IC50 to 33.6% at 4× IC50. In axenic amastigotes, TBD at 4× EC50 triggered early apoptosis in 31.0% of cells, significantly surpassing amphotericin B (20.1%, p < 0.01). Furthermore, TBD induced mitochondrial membrane depolarization in L. infantum promastigotes, with Variation Index values of −0.451, −0.444, and −0.805 at 1×, 2×, and 4× IC50, respectively. Importantly, TBD exhibited no detectable cytotoxicity in human PBMCs. These findings position TBD as a compelling candidate for further preclinical development as a novel treatment for visceral leishmaniasis.
Anisakiasis and parasite-associated hypersensitivity remain important food-safety concerns where raw marine fish is consumed. Although wood creosote (WC) has been studied mainly for effects on anisakid larval motility, its influence on host-relevant epithelial responses to larva-derived products is unknown. We examined whether WC pretreatment of thawed anisakid third-stage larvae alters the biological activity of crude larval extracts in Caco-2 cells. Crude extract from WC-pretreated larvae (CEWC) was compared with paired control extract (CE). CE and CEWC had similar protein concentrations and were not cytotoxic at 0.2-2.0 mg/mL. At 1.0 mg/mL, CE increased Lucifer Yellow flux, whereas CEWC produced a lower increase. Two independently performed cellstimulation experiments using LPS or IL-1 beta as positive controls showed concordant reductions in IL-6 and IL-8 secretion and mRNA expression with CEWC compared with CE. At non-cytotoxic concentrations, WC alone caused only small changes from the control and produced cytokine levels clearly below those induced by IL-1 beta. High-performance liquid chromatography (HPLC) spike-comparison analysis detected WC-associated peaks in CEWC spiked with 1 mu g/mL WC but not in unspiked CEWC filtrate under the tested conditions. These findings support an upstream effect of WC pretreatment on epithelial responses to larval material. The concordant protein and transcript results across the two independent experiments support reproducibility of the cytokine findings, although reduced clinical allergenicity was not assessed.
Artificial intelligence (AI) is increasingly being investigated to address challenges in parasite detection, disease surveillance, and therapeutic development. This narrative review examines current and emerging AI applications in parasitology across diagnostics, drug discovery, genomics, epidemiology, education, and public health decision-making, with particular emphasis on diagnostic applications. Machine learning and deep learning methods support automated parasite detection, classification, and quantification in microscopic images, alongside medical image analysis and arthropod identification. Applications are discussed for malaria, leishmaniasis, schistosomiasis, cystic echinococcosis, soil-transmitted helminth infections, and intestinal protozoal infections. Beyond diagnosis, AI-assisted approaches support compound screening, drug repurposing, toxicity prediction, and prioritization of potential therapeutic targets. Genomic and transcriptomic analyses also offer opportunities to investigate parasite biology and predict drug resistance. Surveillance models may inform outbreak forecasting and targeted control, although their operational value requires evaluation in the intended settings. Educational applications remain emerging, and the benefits of digital learning should be distinguished from those attributable specifically to AI. Across these domains, performance on curated datasets does not establish clinical utility or generalizability. Major barriers include limited representative datasets, insufficient external and prospective validation, infrastructure constraints, and concerns regarding privacy, bias, and accountability. Future progress requires reproducible evaluation, experimental validation of therapeutic candidates, and assessment of clinical and public health benefits. AI is therefore best positioned as a complementary tool that supports parasitological expertise, with responsible implementation guided by transparency, equitable access, and human oversight.
Visceral leishmaniasis (VL) remains a major public health concern in Latin America, particularly in Brazil, where Leishmania infantum is primarily transmitted by the sand fly Lutzomyia longipalpis. Understanding parasite development within the vector is essential for elucidating transmission dynamics, yet classical microscopic assessments often lack sensitivity for detecting early or low intensity infections. In this study, Lu. longipalpis females were experimentally infected with either promastigote or amastigote forms of Le. infantum and the infection dynamics were evaluated by both optical microscopy and digital PCR (dPCR). Microscopic examination revealed successful midgut colonization in both groups, with higher parasite counts in amastigote-infected sand fly females during the early phase of infection. Quantification by dPCR confirmed this pattern, detecting significantly higher parasite loads at the initial time points in amastigote-fed sand flies. However, parasite loads converged between groups at later stages, indicating comparable infection dynamics regardless of the parasite infective form. Both methods showed consistent patterns of parasite dynamics, whereas dPCR provided greater sensitivity for detecting and quantifying low parasite burdens, particularly during periods of reduced infection intensity. Overall, our findings provide new insights into parasite-vector interactions and validate dPCR as a precise and sensitive tool for assessing Leishmania infections in sand fly vectors, with potential applications for entomological surveillance and transmission studies.
Background: Toxocara canis infection is a neglected zoonosis associated with multisystemic involvement and persistent larval infection. Murine models are widely used to investigate host-parasite interactions, but methodological heterogeneity limits biomarker comparison and validation. Objective: To characterize experimental murine T. canis infection models and synthesize parasitological, pathological, immunological, molecular, and functional biomarkers and outcomes. Methods: PubMed/MEDLINE, Embase, Scopus, Web of Science Core Collection, and Virtual Health Library were searched from inception to September 1, 2026, without publication-date restrictions. Only English-language articles were eligible. Studies were screened using Rayyan and assessed according to a PECOS framework. Risk of bias and methodological quality were evaluated using SYRCLE and an adapted CAMARADES checklist. Findings were synthesized descriptively. Results: Sixty-seven studies were included. The central nervous system was most investigated (39/67; 58.2%), followed by pulmonary (32/67; 47.8%) and hepatic (24/67; 35.8%) compartments. Larval recovery and tissue PCR characterized parasite localization or persistence, whereas serological, inflammatory, histopathological, molecular, and functional outcomes captured distinct dimensions of infection. Type-2-associated responses, particularly eosinophilia and IL-4, IL-5, and IL-13, were recurrent. Recent studies incorporated transcriptomics, lipidomics, extracellular-vesicle miRNAs, single-nucleus RNA sequencing, and spatial transcriptomics. Methodological limitations were particularly evident in randomization, blinding, and sample-size reporting. Conclusions: Murine T. canis infection is a dynamic, multisystem process in which biomarker interpretation depends on tissue, timing, and biological context. Evidence supports complementary, purpose-driven outcome assessment rather than a single biomarker or universal protocol. Greater methodological standardization, longitudinal assessment, and multimodal designs may improve reproducibility and biomarker validation.