Mitochondrial Complex I Emerges as a Resistance-Proof Weak Point in Parasite-Infected Cells

A team of researchers in India has found that shutting down a single component of the mitochondrial electron transport chain can kill cells infected by the tick-borne parasite Theileria annulata, even when those cells have become resistant to the only drug approved against the disease. The study, published in Frontiers in Cellular and Infection Microbiology, positions mitochondrial bioenergetics as a resistance-independent therapeutic node that may extend across evolutionarily distant protozoan pathogens, from the apicomplexan Theileria to the kinetoplastid Leishmania donovani, the amoebozoan Entamoeba histolytica, and the free-living amoeba Naegleria fowleri.

The work began with an unbiased phenotypic screen of the Medicines for Malaria Venture Global Health Priority Box, a curated collection of 240 chemically diverse compounds organized into sections covering drug-resistant malaria, zoonotic and neglected diseases, and vector control. Using a resazurin-based viability assay at 10 micromolar, the team identified 15 compounds that reduced the viability of T. annulata-infected bovine leukocytes by at least 40 percent. Secondary validation at 2 micromolar narrowed the field to three leads: rotenone, a classical inhibitor of mitochondrial complex I, designated VEC C04, along with MMV692115 and MMV1828058, which showed half-maximal inhibitory concentrations of 1.30 plus or minus 0.27 micromolar, 2.40 plus or minus 0.03 micromolar, and 1.72 plus or minus 0.43 micromolar respectively.

Selectivity proved to be a defining feature of all three hits. Uninfected bovine peripheral blood mononuclear cells retained more than 90 percent viability at 10 micromolar, and rotenone showed no detectable toxicity in those cells even at 100 micromolar. Structural similarity analysis based on Tanimoto coefficients confirmed low chemical overlap between the lead compounds and the reference drugs buparvaquone and artesunate, hinting at mechanisms of action distinct from existing therapies. Time-course experiments revealed that rotenone acted fastest, producing observable cell death within 24 hours and preceding the other two leads as well as buparvaquone. Western blotting showed a dose-dependent reduction of the parasite surface protein TaSP, confirming that the compounds suppressed parasite viability within the infected host cells.

The resistance data carry the greatest clinical weight. Buparvaquone, the sole approved drug for bovine theileriosis, targets the parasite’s mitochondrial cytochrome bc1 complex, and its exclusive use has driven the emergence and spread of resistant T. annulata strains in endemic regions. When the researchers tested a buparvaquone-resistant cell line they generated through stepwise drug exposure, the two other lead compounds showed minimal activity, suggesting shared resistance mechanisms with buparvaquone. Rotenone, by contrast, retained potent efficacy, inducing roughly 50 percent cell death in the resistant line. Because rotenone acts on complex I of the host mitochondrial electron transport chain, an entirely different node from the parasite-encoded complex III targeted by buparvaquone, host-directed mitochondrial targeting appears able to bypass parasite-intrinsic resistance mechanisms altogether.

Mechanistically, the study delineates a temporally ordered, three-tiered cascade of cell death. Complex I inhibition first collapses the mitochondrial membrane potential, impairing oxidative phosphorylation and depleting cellular energy. Measurements showed that ATP content fell from 32.33 plus or minus 3.19 nanomoles per million cells in controls to 7.21 plus or minus 0.69 nanomoles after 24 hours of rotenone treatment, a dramatic bioenergetic collapse. JC-1 staining confirmed depolarization, with the red-to-green fluorescence ratio dropping from about 2.5 in untreated cells to roughly 0.88 after 48 hours, while MitoTracker Green staining corroborated reduced mitochondrial mass and activity.

The bioenergetic insult then drives electron leak and superoxide generation. MitoSOX staining revealed progressive accumulation of mitochondrial reactive oxygen species, and total intracellular ROS measured by DCFDA fluorescence rose approximately 1.94-fold after 24 hours of treatment. The ROS scavenger N-acetylcysteine reduced that fluorescence by about 37.1 percent, supporting a functional contribution of oxidative stress to the rotenone-induced phenotype. The oxidative burden in turn inflicted genomic damage: both immunofluorescence microscopy and Western blotting demonstrated increased gamma-H2AX, a marker of DNA double-strand breaks, after 48 hours. Phosphorylation of the tumor suppressor p53 at serine 15 confirmed engagement of the DNA damage response, and cells accumulated in the G2/M phase of the cell cycle before dying.

The final tier of the cascade is apoptosis through the intrinsic, mitochondrial pathway. TUNEL assays revealed significant DNA fragmentation within 24 hours, and Western blotting showed elevated cleavage of caspase-9 and caspase-3, while caspase-8 remained unchanged, ruling out the death receptor pathway. At its half-maximal inhibitory concentration, rotenone induced apoptosis in approximately 70 percent of infected cells after 48 hours, substantially exceeding the roughly 45 and 43 percent responses elicited by the other two leads. DAPI staining revealed characteristic nuclear condensation, providing additional morphological confirmation of programmed cell death.

Transcriptomic profiling by RNA sequencing added a systems-level view of how the infected cells respond. Differential expression analysis identified 26 upregulated and 36 downregulated genes, revealing a coordinated stress response coupled with suppression of biosynthetic and proliferative programs. The heat shock gene HSPA6 showed the largest positive fold change at log2 11.28, alongside upregulation of the p53-regulated ubiquitin ligase G2E3, the lysosomal glycoprotein GPNMB, the respiratory attenuator NDUFA4L2, and the co-chaperone DNAJB1, all consistent with proteotoxic, oxidative, and mitochondrial adaptive stress. Conversely, genes required for translational fidelity and proliferation, including GARS1, RB1, EIF1, and EEF1E1, were downregulated. Because the RNA-seq analysis used only two biological replicates per condition, the researchers validated the principal signatures by RT-qPCR with three independent replicates, confirming the direction of change for nine of the key genes.

Drug combination studies extended the therapeutic logic. Checkerboard assays quantified by fractional inhibitory concentration indices revealed the strongest synergy between rotenone and buparvaquone, at 0.2389 plus or minus 0.12, followed by rotenone with atovaquone at 0.433 plus or minus 0.084, indicating that co-targeting complexes I and III produces cooperative effects exceeding either agent alone. Independent analysis with the SynergyFinder platform, using the Highest Single Agent and Loewe models, yielded positive synergy scores that reinforced the value of hitting distinct electron transport chain nodes, while also underscoring the importance of evaluating drug interactions with complementary analytical models. Testing of additional complex I inhibitors showed that piericidin A had measurable but poorly selective activity, whereas 1-methyl-2-pentyl-4(1H)-quinolinone showed no inhibition, demonstrating that complex I blockade is not a uniform class effect and that structural specificity is required for productive target engagement.

Perhaps the most intriguing finding is the cross-phylum activity of rotenone. The compound inhibited L. donovani promastigotes with an IC50 of 10.12 plus or minus 0.98 micromolar, showed moderate activity against E. histolytica at 32.94 plus or minus 0.27 micromolar, and killed 44.27 plus or minus 4.02 percent of N. fowleri at 20 micromolar. This poses a puzzle, since protozoan parasites are widely described as lacking the rotenone-sensitive mammalian-type complex I and instead relying on the alternative type II NADH dehydrogenase, NDH2. Comparative sequence, motif, and phylogenetic analyses revealed a shared PTZ00318 superfamily domain architecture across Plasmodium, Theileria, Babesia, Toxoplasma, Leishmania, and Naegleria, while the NDH2 ortholog of Entamoeba, which lacks canonical mitochondria and retains only non-respiring mitosomes, diverged from this pattern, correlating with its reduced susceptibility. The authors caution that these in silico observations do not establish a direct rotenone-NDH2 interaction and require functional validation. They also emphasize that rotenone itself, with its systemic toxicity, is unsuitable for clinical use and serves instead as a chemical probe. The broader principle, that mitochondria-dependent vulnerabilities such as membrane potential disruption, ROS induction, and metabolic dependency are shared across mitochondrion-bearing protozoa, now provides a foundation for designing safer, broad-spectrum antiparasitic agents that selectively exploit mitochondrial pathways, independent of whatever resistance mechanisms parasites may evolve against existing drugs.

Subject of Research: Mitochondrial complex I inhibition as a resistance-independent therapeutic strategy against protozoan parasites

Article Title: Complex I inhibition collapses host bioenergetics and activates p53‐dependent apoptosis in Theileria annulata infected cells

Article References: Subudhi, M., Singothu, S., Bhandari, V., Brahma, U., Verma, K., Navyaka, P., Suresh, A., Singh, S., & Sharma, P. (2026). Complex I inhibition collapses host bioenergetics and activates p53‐dependent apoptosis in Theileria annulata infected cells. Frontiers in Cellular and Infection Microbiology, 16, Article 1908136. https://doi.org/10.3389/fcimb.2026.1908136

Image Credits: AI Generated

DOI: 10.3389/fcimb.2026.1908136

Keywords: Theileria annulata, mitochondrial complex I, rotenone, buparvaquone resistance, p53, apoptosis, reactive oxygen species, electron transport chain, Leishmania donovani, Entamoeba histolytica, Naegleria fowleri, drug repurposing