Trypanosomatid flagellates such as Trypanosoma brucei and T. cruzi have long fascinated cell biologists because their endocytic traffic is confined to a single invagination of the plasma membrane called the flagellar pocket. This unusual geometry makes them powerful models for testing how universal the principles of membrane trafficking really are. Previous work had catalogued many components of clathrin-mediated endocytosis in these parasites, but the composition and evolutionary standing of their ESCRT machinery remained incompletely resolved. In animals and fungi, a subcomplex called ESCRT-0, built around the Vps27/Hrs and STAM proteins, recognises ubiquitylated substrates and hands them to the downstream complexes. Whether trypanosomes possess anything comparable has been an open question.

The team approached the problem from two directions: comparative genomics across the eukaryotic tree and experimental biochemistry in the parasite itself. Using hidden Markov model searches and reciprocal best BLAST analyses, they surveyed hundreds of genomes from every major eukaryotic supergroup for orthologs of candidate sorting proteins. In parallel, they engineered procyclic-form T. brucei cells to express an endogenous GFP-tagged version of Vps23, the Tsg101 orthologue that anchors the ESCRT-I complex. Affinity purification followed by quantitative mass spectrometry then pulled down the proteins that travel with TbVps23 in living cells, allowing the researchers to confirm predicted partners and uncover unexpected ones.

The most striking biochemical discovery was that trypanosome ESCRT-I is not the canonical heterotetramer familiar from yeast and animals. Instead, the purification identified two novel Vps23-interacting proteins, encoded by adjacent open reading frames Tb927.11.2020 and Tb927.11.2030, that travel stoichiometrically with the core subunits. These proteins are specific to trypanosomes, with no obvious homologues outside the kinetoplastid lineage, and their association with Vps23 indicates that the trypanosome ESCRT-I subcomplex is a heteropentamer. The finding adds a new tier of compositional divergence to a complex long considered structurally conserved, and it underscores how evolutionary flexibility can be woven into even the most fundamental cellular machines.

Equally consequential is the team’s answer to the question of what serves as the trypanosome ESCRT-0 analogue. Rather than a Vps27/STAM pair, which the analyses confirmed is absent from most non-opisthokont lineages, the parasites employ Tom1, a protein carrying the VHS and GAT domains that is broadly distributed across eukaryotes. The researchers demonstrated that trypanosome Tom1, termed TbTom1, colocalises with TbVps23 at endosomal membranes and interacts with a kinetoplastid-specific partner protein the authors nicknamed Jerry, a multi-domain protein studded with three FYVE domains, lipid-binding modules that recognise endosomal phosphoinositides. Together, Tom1 and Jerry appear to fulfil the substrate-recognition role that ESCRT-0 plays in fungi and animals, but through entirely different molecular parts.

Structural predictions reinforced the ancient pedigree of this arrangement. AlphaFold3 modelling of TbTom1 revealed a compact, well-ordered VHS helical bundle linked to a GAT domain, closely superimposable on the VHS domain of human TOM-1, while the remaining regions of the protein are predicted to be intrinsically disordered. This architecture supports the authors’ proposal that Tom1 is the ancestral VHS-GAT-domain-containing cargo adaptor, present in the last common ancestor of eukaryotes and retained across supergroups, whereas the Vps27/Hrs-STAM system of ESCRT-0 is a later innovation restricted to animals and fungi. By mapping the occurrence of Tom1 and Jerry orthologues across the Euglenozoa, the researchers showed that Tom1 is universally retained within the phylum while Jerry is absent from diplonemids, tracing a lineage-specific layer of specialisation on top of a conserved core.

The experimental evidence connecting these proteins to endocytic function came from cellular phenotyping. Cells expressing tagged versions of TbTom1 or TbJerry showed significantly enlarged flagellar pockets, with morphometric analysis of DAPI-stained cells revealing a marked increase in pocket area compared with the parental line, and tagged TbJerry cells additionally displaying a subtle nozzle-like distortion of the pocket. Because the flagellar pocket is the sole site of endocytosis in trypanosomes, such expansion implies a direct perturbation of membrane uptake and trafficking throughput. The phenotypes were measured across individual cloned lines and validated by immunoblotting of endogenous tags, giving the morphological data a firm molecular footing.

Curiously, the machinery appears dispensable for parasite viability in laboratory culture. RNA interference targeting TbTom1 in bloodstream-form cells reduced transcript levels by roughly 65 percent yet left proliferation unimpaired over an eight-day course, and mass spectrometric profiling of the silenced proteome showed that overall protein composition shifted only modestly. Concanavalin A uptake assays, which track soluble cargo from the flagellar pocket through endosomes to the lysosome, proceeded indistinguishably in TbTom1-depleted cells and controls. The authors interpret this resilience as evidence of redundancy or adaptability within the trypanosome sorting network rather than irrelevance of Tom1, a pattern consistent with the plasticity that their comparative analyses document throughout the system.

Broadly, the study reframes how biologists should think about the early ESCRT machinery. Evidence that Tom1 functions within the ESCRT pathway of trypanosomes supplies the fourth eukaryotic supergroup in which such a role has been demonstrated, and the authors argue that Tom1, not Vps27/STAM, is likely part of the pan-eukaryotic early ESCRT apparatus. At the same time, the heteropentameric ESCRT-I complex and the kinetoplastid-restricted Jerry protein illustrate how individual lineages continue to elaborate and customise the machinery, echoing earlier discoveries such as the plant-specific FREE1 adaptor, whose orthologues the team also surveyed across embryophytes. Evolution, in this view, has conserved the skeleton of the sorting system while repeatedly rebuilding its sensory apparatus.

For parasitologists, the implications extend toward intervention. Trypanosomes depend on relentless endocytic turnover of their variant surface glycoproteins and invariant surface receptors to evade the immune system, and any component that is parasite-specific, such as the novel ESCRT-I subunits or Jerry, represents a potential Achilles heel that host proteins do not share. Conversely, the conserved Tom1-dependent pathway offers a benchmark against which the animal and fungal systems can be reinterpreted, potentially correcting an animal-centric bias that has coloured membrane trafficking research for decades. What this study makes clear is that the deepest lessons about how eukaryotic cells move their membranes may come not from yeast or human tissue culture, but from an ancient parasite whose cellular geography forces every molecular handoff to happen in one remarkable pocket of membrane.

Subject of Research: The composition and evolution of the ESCRT endosomal sorting system in trypanosome parasites

Article Title: The trypanosome ESCRT system possesses both novel components and under-recognised pan-eukaryotic features

Article References: Padilla-Mejia, N. E., Durante, I. M., Záhonová, K., Stříbrná, E., Heller, J., Dacks, J. B., Lukeš, J., & Field, M. C. (2026). The trypanosome ESCRT system possesses both novel components and under-recognised pan-eukaryotic features. BMC Biology. https://doi.org/10.1186/s12915-026-02730-4

Image Credits: AI Generated

DOI: 10.1186/s12915-026-02730-4

Keywords: ESCRT, endocytosis, trypanosomes, Trypanosoma brucei, Tom1, Vps23, flagellar pocket, protein trafficking, endosomes, eukaryotic evolution, VHS domain, FYVE domain

The burden of sleeping sickness is considerable. The disease progresses in stages, beginning with fever, headaches and lymphadenopathy as parasites multiply in the blood and lymph, and advancing to neurological involvement once the parasites cross the blood-brain barrier, producing sleep disturbances, cognitive decline and, if untreated, death. The parasite’s most formidable weapon is antigenic variation: a dense coat of variant surface glycoproteins, or VSGs, that is continually reshuffled through gene conversion, allowing the parasite to stay one step ahead of the host antibody response. This immune evasion strategy, well documented in the literature, has long frustrated conventional vaccine development, which is one reason the researchers turned to conserved, functionally essential proteins as alternative targets.

Specifically, the team selected three T. brucei proteins as the basis for their construct: the variant surface glycoprotein itself, heat shock protein 70, and the vacuolar transporter chaperone complex. Heat shock protein 70 is a highly conserved molecular chaperone central to protein folding and stress responses, while the vacuolar transporter chaperone complex is involved in polyphosphate synthesis and acidocalcisome function, processes essential to parasite survival. By mining these proteins for immunogenic peptides, the researchers aimed to build a construct that combines surface exposure with conservation across strains, increasing the likelihood that an immune response raised against the vaccine would recognize the parasite before it can establish infection.

The design workflow followed the now-standard logic of reverse vaccinology. Protein sequences were retrieved from the UniProt database and aligned with tools such as Clustal Omega to assess conservation. Cytotoxic T lymphocyte epitopes were predicted for MHC class I presentation, and helper T lymphocyte epitopes for MHC class II, using the Immune Epitope Database analysis resource and related servers. B-cell epitopes were predicted with linear and discontinuous methods, including ElliPro for structure-based antibody epitope mapping. Each candidate epitope was then filtered for allergenicity with AllerTOP, for toxicity with dedicated peptide toxicity predictors, and for antigenicity with VaxiJen, ensuring that only immunogenic, non-allergenic and non-toxic peptides entered the final construct.

Population coverage analysis, which estimates how many people worldwide carry HLA alleles capable of presenting the chosen epitopes, returned a striking result: the vaccine candidate achieved 100 percent global population coverage. This metric matters because a vaccine that only fits a narrow slice of HLA diversity would leave large populations unprotected. Biophysical characterization of the final multi-epitope protein showed an aliphatic index of 71.23, indicating good thermal stability, and a GRAVY score of minus 0.719, indicating a hydrophilic, soluble protein likely to fold and express well. Solubility and instability assessments supported the view that the construct should behave as a stable, expressible protein in a cellular context.

Structural modeling came next. The tertiary structure of the vaccine construct was predicted and evaluated with a TM-score of 0.65 plus or minus 0.13 and a C-score of minus 0.50, values consistent with a reliable fold. The model was then refined, and validation metrics confirmed its quality: a Ramachandran score of 86.8 percent, meaning the vast majority of residues occupy favored or allowed backbone conformations, and a ProSA Z-score of minus 5.26, within the range expected for proteins of comparable size. Disulfide engineering was considered to further stabilize the fold, and secondary structure predictions from PSIPRED and SOPMA were used to cross-check the modeled architecture.

To test whether the vaccine could actually engage the innate immune sensors that trigger adaptive responses, the team docked the construct against Toll-like receptors 2 and 4, key pattern-recognition receptors on antigen-presenting cells. Molecular docking predicted strong binding, with energy scores of minus 1013.5 kJ/mol for TLR-2 and minus 1002.8 kJ/mol for TLR-4. These docked complexes were then subjected to molecular dynamics simulation, principal component analysis, dynamic cross-correlation matrix analysis and MM-GBSA binding free energy calculations, all of which supported the stability and favorable energetics of the receptor-vaccine interactions. In practical terms, the simulations suggest the vaccine construct should bind robustly to the very receptors that initiate the innate immune cascade.

Immune simulation provided the most direct readout of the construct’s potential immunogenicity. Using computational immune system modeling, the researchers predicted robust humoral and cell-mediated responses, including elevated B lymphocyte and T lymphocyte populations and rising titers of IgM and IgG antibodies over the simulated immunization course. Cytokine profiles indicated activation of both Th1-type and Th2-type pathways, the dual signature generally desired in a prophylactic vaccine. While such simulations are approximations of a vastly more complex biological reality, they serve as a critical screening step, allowing weak candidates to be discarded before any laboratory resource is spent.

Because the platform is mRNA, the team also optimized the nucleic acid sequence itself. Codon optimization for expression in Escherichia coli strain K12, conducted for in-silico cloning into the pET-28a(+) vector, yielded a codon adaptation index of 0.9688 and a GC content of 44.70 percent, both indicative of high expression potential. In-silico cloning confirmed that the construct could be inserted into the vector without disrupting restriction sites. Finally, minimum free energy analysis of the mRNA sequence was used to evaluate the structural integrity and stability of the transcript, an important consideration since mRNA secondary structure influences translation efficiency and vaccine performance.

The authors are careful to frame the work as a computational proof of concept rather than a finished vaccine. As they conclude, the in-silico designed candidate demonstrated strong structural stability, favorable receptor interactions and promising immunogenic potential against T. brucei, but experimental validation and in-vivo studies are required to verify its safety and efficacy. That caveat applies to the entire field of computational vaccinology: docking scores and immune simulations can prioritize candidates and dramatically shorten development timelines, but only animal studies and clinical trials can establish whether a designed construct protects against real infection. Still, for a disease with no licensed vaccine, in which drug therapy is costly, logistically difficult and increasingly challenged by resistance, a rationally designed mRNA candidate that clears every computational hurdle represents a meaningful step forward, and a template for applying the same pipeline to other neglected tropical parasites.

Subject of Research: Computational immunoinformatics design of an mRNA vaccine candidate against the sleeping sickness parasite Trypanosoma brucei

Article Title: Immunoinformatics Approach for the Designing of a Novel mRNA Vaccine Candidate Against Trypanosoma brucei

Article References: Nil, M. S., Khandker, S., Sadaf, S., Ahmed, N., Reja, S., Sayfullah, M., Ferdous, J., Saha, S., Alamri, A., Khan, M. S., Ahmed, S., Wajed, S., Mahdeen, A. A., & Siddiquee, N. H. (2026). Immunoinformatics Approach for the Designing of a Novel mRNA Vaccine Candidate Against Trypanosoma brucei. Acta Parasitologica, 71(5), Article 205. https://doi.org/10.1007/s11686-026-01388-w

Image Credits: AI Generated

DOI: 10.1007/s11686-026-01388-w

Keywords: Trypanosoma brucei, human African trypanosomiasis, mRNA vaccine, immunoinformatics, reverse vaccinology, epitope prediction, molecular docking, molecular dynamics simulation, TLR-2, TLR-4, codon optimization, neglected tropical disease