Thioridazine

Limits on what this shows

• Key negative/null finding: Thioridazine did not inhibit Complex I activity, unlike chlorpromazine and haloperidol (Modica-Napolitano et al. 2003) - evidence against a uniform "typical-antipsychotic mitochondrial toxicity" class effect.
• At-risk subgroups: None identified clinically; the leukemia-cell and some hepatocyte data used concentrations well above standard therapeutic exposure.

Regulatory status

Not addressed in the FDA label.

Population evidence

None human. Isolated rat liver mitochondria exposed to thioridazine showed stimulated respiration consistent with uncoupling of oxidative phosphorylation (Soberón et al. 1976); a later comparative study across typical and atypical antipsychotics, with thioridazine specifically named, confirmed thioridazine acts as a "modest uncoupler" of oxidative phosphorylation but, unlike chlorpromazine and haloperidol, does not inhibit respiratory Complex I (Modica-Napolitano et al. 2003). Separately, 10 µM thioridazine protected isolated rat liver mitochondria against induced permeability transition and cytochrome c release (Rodrigues et al. 2002). In freshly isolated rat hepatocytes (Eftekhari et al. 2018) and in leukemia cell lines at cytotoxic, anticancer-research concentrations of 7.5 to 30 micromolar (Moraes et al. 2022), thioridazine caused calcium-dependent mitochondrial membrane depolarization and outer-membrane permeabilization leading to apoptosis.

Mechanism

Direct interaction with, and uncoupling of, the inner mitochondrial membrane at lower concentrations; calcium-dependent mitochondrial outer-membrane permeabilization at higher, cytotoxic concentrations used in cancer-repurposing research.

Counter-evidence held on record

The protective/antioxidant finding (Rodrigues et al. 2002) directly competes with the uncoupling and permeabilization findings above; net clinical direction is unresolved and appears concentration- and model-dependent. Do not extrapolate preclinical uncoupling/permeabilization data into a clinical mitochondrial-disease claim or a supplement recommendation.

Limits on what this shows

• Key negative/null finding: One isolated-mitochondria study found the opposite direction: 10 µM thioridazine suppressed ROS accumulation and lipid peroxidation and protected against mitochondrial permeability transition in rat liver mitochondria, an antioxidant/protective effect at the inner mitochondrial membrane (Rodrigues et al. 2002). The pro-oxidant and antioxidant findings come from different tissue/model systems (whole melanocytes and hepatocytes versus isolated mitochondrial membrane) and cannot be reconciled from the literature as reviewed.
• At-risk subgroups: None identified; all evidence is preclinical.

Regulatory status

Not addressed in the FDA label; no antioxidant-status or oxidative-stress biomarker monitoring is recommended.

Population evidence

None human. In vitro/ex vivo/animal only: human melanocytes (HEMn-DP) exposed to 0.01 to 2.5 micromolar thioridazine for up to several days showed dose-dependent H₂O₂ elevation (+9.8% to +67%), SOD increase (+12.5% to +52.6%), and CAT/GPx decrease at higher concentrations (Otręba et al. 2015). Isolated rat brain myelin showed increased lipid peroxidation after thioridazine exposure (Dhaunsi et al. 1993; abstract text could not be retrieved during this review, finding drawn from title and confirmed journal metadata). Freshly isolated rat hepatocytes exposed to thioridazine showed ROS formation and lipid peroxidation, attenuated by melatonin and/or coenzyme Q10 (Eftekhari et al. 2018); a companion rat-hepatocyte study using the same metabolic-activation paradigm found the oxidative pattern blunted by cimetidine, N-acetylcysteine, and taurine (title and DOI confirmed; full text blocked by repeated rate-limiting during this review). NIH LiverTox attributes some thioridazine-associated liver injury to hepatic bioactivation toward electrophilic iminoquinone intermediates, a reactive-metabolite mechanism consistent with the above.

Mechanism

Hepatic oxidative metabolism (sulfoxidation) generating reactive intermediates; in melanocytes, D2-receptor antagonism was proposed to dysregulate normal ROS handling, with accumulating H₂O₂ outpacing catalase/GPx capacity.

Counter-evidence held on record

Preclinical and mechanistic only; the direction of effect is model- and concentration-dependent and unresolved in the literature. This is not evidence that antioxidant supplementation is needed, or beneficial, during thioridazine therapy. Shares its underlying sources with the Mitochondrial burden entry below - see Overlap map.

Limits on what this shows

• Key negative/null finding: Haloperidol and clozapine, tested in the same preparation, showed no stimulation of iron uptake - this is not a uniform antipsychotic-class effect.
• At-risk subgroups: None demonstrated in vivo; ex vivo rat brain tissue only.

Regulatory status

Not addressed in the FDA label.

Population evidence

None human. Rat cerebral-cortex synaptosome preparations exposed to several neuroleptics, with thioridazine specifically named and tested, showed stimulated uptake of both ⁵⁵Fe-citrate and ⁵⁵Fe-transferrin by thioridazine and chlorpromazine; fluphenazine increased citrate uptake only; haloperidol and clozapine had no effect (Penatti et al. 1998).

Mechanism

Proposed calmodulin-antagonism-linked stimulation of synaptosomal iron uptake, with additional involvement of protein phosphorylation and membrane-polarization changes.

Counter-evidence held on record

This is an excess/retention finding (Hard Rule 5), not a depletion finding. It does not imply iron intake should change in either direction, and no supplementation or restriction is recommended on this basis. A plausible but unconfirmed mechanistic link exists to the Oxidative stress entry above (iron catalyzes Fenton-type radical chemistry) - no source directly measured both, so this is flagged in the Overlap map as a conceptual link only, not a shared citation.

Limits on what this shows

• Key negative/null finding: Direct searches for "thioridazine copper" and "thioridazine ceruloplasmin" returned no on-topic drug-nutrient-status study beyond the Løvstad mechanism paper; the other results returned (retinopathy case literature, unrelated fluorescence-chemosensor chemistry papers, unrelated cytotoxicity assays) do not address copper status and are not cited as support.
• At-risk subgroups: None identified.

Regulatory status

Not addressed in the FDA label; no copper or ceruloplasmin monitoring is recommended.

Population evidence

None human, and no study of thioridazine's effect on serum copper, ceruloplasmin levels, or copper absorption was located. The only on-topic source is a biochemical mechanism study: ceruloplasmin oxidase activity generates a phenothiazine radical intermediate that oxidizes oxyhemoglobin to methemoglobin, using thioridazine and related phenothiazines as the test compounds (Løvstad 1981).

Mechanism

Ceruloplasmin (copper-dependent oxidase) activity toward thioridazine generates a reactive radical intermediate; this is an enzyme-mechanism finding, not a copper-status finding.

Counter-evidence held on record

Do not interpret this entry as a copper-deficiency or copper-excess claim. It is enzyme-mechanism evidence only, from a single 1981 in vitro study. Shares its ceruloplasmin source with the Oxygen delivery/utilization entry above - see Overlap map.

Limits on what this shows

• Key negative/null finding: This is an antimicrobial-repurposing dosing paradigm targeting the gut lumen directly, not the standard oral antipsychotic dosing paradigm; whether standard psychiatric dosing reaches comparable gut-luminal concentrations to meaningfully shift microbiota composition is not established in any source located. A structurally related, but distinct, phenothiazine derivative (not thioridazine itself, and not confirmed to be one of this project's other reviewed phenothiazines) was also reported active against C. difficile with low cytotoxicity in a separate paper (Ronco et al. 2021) - noted here as class-level corroboration of antibacterial activity among phenothiazines generally, explicitly labeled as class extrapolation per Hard Rule 6, and not counted as thioridazine-specific evidence.
• At-risk subgroups: None identified for standard psychiatric dosing; all direct evidence is from an antimicrobial-dosing mouse model.

Regulatory status

Not addressed in the FDA label; thioridazine is not approved or labeled as an antimicrobial.

Population evidence

None human. Mouse model of C. difficile infection: thioridazine inhibited the C. difficile ABC efflux pump CD2068 (IC₅₀ for ATP hydrolysis approximately 30 micromolar; MIC 64 to 128 micrograms per milliliter against C. difficile) and showed direct antibacterial/cell-wall-disruptive activity; in treated mice, Lactobacillus, Muribaculaceae, and Lachnospiraceae were preserved or restored and Proteus was reduced, relative to vancomycin-treated controls (Janvilisri et al. 2025).

Mechanism

Efflux-pump (CD2068) inhibition and direct antibacterial/cell-wall activity against C. difficile, with a broader, comparatively microbiota-sparing effect than vancomycin in this model.

Counter-evidence held on record

Evidence here points away from, not toward, a dysbiosis risk in this specific model, but the model and dosing context differ substantially from standard antipsychotic use. Neither a depletion/dysbiosis claim nor a probiotic-benefit claim is supported for routine thioridazine therapy.

Limits on what this shows

• Key negative/null finding: No electron-transport (Complex I) inhibition was found for thioridazine specifically, distinguishing it from chlorpromazine and haloperidol in the same comparative study (Modica-Napolitano et al. 2003).
• At-risk subgroups: None identified; isolated-organelle data only.

Regulatory status

Not addressed in the FDA label.

Population evidence

None human. Isolated rat liver mitochondria exposed to thioridazine showed stimulated oxygen consumption characteristic of a mild uncoupler (Soberón et al. 1976), and a later study confirmed thioridazine acts as a "modest uncoupler" of oxidative phosphorylation without inhibiting Complex I (Modica-Napolitano et al. 2003).

Mechanism

Uncoupling of the proton gradient from ATP synthase activity at the inner mitochondrial membrane, increasing oxygen consumption relative to ATP yield, without blocking electron-transport-chain flux.

Counter-evidence held on record

No human ATP or energy-metabolism data exist. Do not interpret patient-reported fatigue or sedation on thioridazine as evidence of this mechanism: sedation on this drug is primarily receptor-mediated (histamine H1, dopamine D2, and muscarinic M1 blockade), the drug's own intended/incidental pharmacology, not a nutrient- or energy-metabolism-mediated effect (Hard Rule 2). Shares its underlying sources with the Mitochondrial burden entry above - see Overlap map.

Limits on what this shows

• Key negative/null finding: A human study using thioridazine as a repurposed CXCR4-antagonist hematopoietic-progenitor-mobilizing agent in healthy volunteers did not report clinically significant drops in hemoglobin or hematocrit (Chigaev et al. 2015) - reassuring, though the study was not designed to detect subclinical methemoglobinemia or eryptosis.
• At-risk subgroups: Concentrations used in the mechanistic studies (6 to 30 micromolar free drug) are well above typical steady-state plasma concentrations reported for standard oral dosing; clinical relevance at therapeutic exposure is unconfirmed.

Regulatory status

The FDA label lists anemia and aplastic anemia only as rare, class-level, post-marketing blood dyscrasias, without a measured mechanism or biomarker; this is an outcome-level adverse-event listing, not a functional oxygen-delivery marker, and is not the basis for this entry's grade (Hard Rule 2).

Population evidence

None human measuring oxygen delivery directly. In vitro: ceruloplasmin (a copper-dependent plasma oxidase) catalyzed oxidation of human oxyhemoglobin to methemoglobin by thioridazine and related phenothiazines, via a free-radical intermediate; the reaction was enhanced by EDTA and blunted by ascorbate/NADH (Løvstad 1981). Separately, ex vivo human erythrocytes exposed to thioridazine at 6 to 30 micromolar (significant effects at ≥12 µM) underwent eryptosis, that is, cell shrinkage, calcium entry, and phosphatidylserine exposure, via calcium entry and p38 kinase activation (Lang et al. 2013).

Mechanism

(a) Ceruloplasmin-catalyzed, copper-dependent free-radical oxidation of oxyhemoglobin to methemoglobin. (b) Calcium-dependent, p38-kinase-mediated eryptosis.

Counter-evidence held on record

Chigaev et al. 2015 (above) found no clinically significant hematologic drop in healthy volunteers given thioridazine for a repurposed indication. Mechanistic/in vitro only, at concentrations exceeding typical therapeutic plasma levels. Not evidence of clinical anemia or hypoxia risk, and not a substitute for the label's own hematologic-dyscrasia monitoring guidance. Shares its ceruloplasmin source with the Copper entry below - see Overlap map.