Nortriptyline

Limits on what this shows

• Key negative/null findings: No study identified shows nortriptyline impairing or depleting mitochondrial function in normal, non-cancerous tissue at therapeutic concentrations. The one opposite direction finding (Zhu 2024) is specific to cancer cell lines being deliberately targeted for cell death at cytotoxic, drug repurposing concentrations, not a patient safety signal.
• At-risk subgroups: Not established as a safety concern, since the direction found in normal tissue models is protective; the cancer cell finding describes a research target, not a risk to a treated patient's own tissue.
• Counter-evidence on record: The Zhu 2024 cancer cell finding is recorded here explicitly and must not be conflated with normal tissue risk. This finding does not replicate, and is not the same evidence as, the CoQ10 depletion signal identified for amitriptyline in the companion dossier; that Tier A human finding does not carry over to nortriptyline based on the evidence located in this independent review.

Regulatory status

Not addressed in the FDA label.

Population evidence

Zhang WH, Wang H, Wang X, et al, "Nortriptyline protects mitochondria and reduces cerebral ischemia/hypoxia injury," Stroke 2008, DOI 10.1161/strokeaha.107.496810, PMID 18174477 (Tier B, cultured neurons plus mouse middle cerebral artery occlusion model) - nortriptyline was identified via screening of FDA approved drugs using isolated mitochondria as a potent mPTP inhibitor; it reduced oxygen glucose deprivation neuronal death, blocked caspase-3 activation, and decreased infarct volume with improved neurological outcome in mice. Wang H, Guan Y, Wang X, et al, "Nortriptyline delays disease onset in models of chronic neurodegeneration," Eur J Neurosci 2007, DOI 10.1111/j.1460-9568.2007.05663.x, PMID 17686041 (Tier B, mouse ALS and Huntington's disease models, same laboratory group) - nortriptyline significantly delayed disease onset and extended lifespan in ALS mice and showed comparable benefit in Huntington's disease models, via inhibition of cytochrome c release and caspase-3 activation. Jeong JH, Kim DK, Lee NS, et al, "Neuroprotective Effect of Nortriptyline in Overt Hepatic Encephalopathy Through Attenuation of Mitochondrial Dysfunction," ASN Neuro 2018, DOI 10.1177/1759091418810583, PMID 30428281 (Tier B, independent laboratory, bile duct ligated rat model plus PC-12 cell glutamate excitotoxicity model) - nortriptyline reduced motor deficits, neuronal cell death, synaptic loss, and dopaminergic neuron apoptosis in rats, and in PC-12 cells reduced mitochondrial membrane potential dissipation and blocked cytochrome c release, the same mPTP blocking mechanism, independently confirmed in a third laboratory and a third disease model. Yazdani I, Ghazi-Khansari M, Saeedi Saravi SS, et al, "Nortriptyline protects testes against germ cell apoptosis and oxidative stress induced by testicular ischaemia/reperfusion," Andrologia 2016, DOI 10.1111/and.12605, PMID 27147301 (Tier B, rat, N=96) - nortriptyline pretreatment reduced testicular oxidative and apoptotic injury in a dose dependent manner; the authors attribute the protective effect to mPTP inhibition, consistent with the three sources above. A companion 2022 paper (Yazdani et al, PMID 35932053) extended this finding in combination with cyclosporine A. Contrasting context: Zhu C, Lu Y, Wang S, et al, "Nortriptyline hydrochloride, a potential candidate for drug repurposing, inhibits gastric cancer by inducing oxidative stress by triggering the Keap1-Nrf2 pathway," Sci Rep 2024, DOI 10.1038/s41598-024-56431-5, PMID 38480798 (Tier B, gastric cancer cell lines AGS and HGC27, plus nude mouse xenografts) found nortriptyline decreased mitochondrial membrane potential and increased oxidative stress markers specifically in cancer cells, as part of oncology drug repurposing research, an opposite direction finding confined to a cytotoxic cancer cell line context, not normal tissue.

Mechanism

Direct inhibition of the mitochondrial permeability transition pore, blocking cytochrome c release and downstream caspase-3 mediated apoptosis, named consistently across the four protective studies above.

Counter-evidence held on record

This is a real, consistent, drug specific, multi-laboratory preclinical finding, but it is entirely animal and in vitro data; no human data exists, so it must never be presented as a confirmed human neuroprotective benefit, and never used to suggest nortriptyline should be taken for mitochondrial protection. This entry shares its evidentiary core with Oxidative stress and ATP production below; see the Overlap map.

Limits on what this shows

• Key negative/null findings: No human oxidative stress marker study was located for nortriptyline in either direction; all evidence identified is preclinical.
• At-risk subgroups: None established in normal tissue. The pro-oxidant finding applies to gastric cancer cell lines being studied as a treatment target, not to a patient's own tissue during standard therapeutic use.
• Counter-evidence on record: The two directions found, protective in ischemia/reperfusion models and pro-oxidant in cancer cells, are themselves the recorded counter-evidence to each other and must not be merged into one directional claim.

Regulatory status

Not addressed in the FDA label.

Population evidence

Yazdani 2016 (PMID 27147301, Tier B, rat testicular ischemia/reperfusion, N=96, detailed under Mitochondrial burden) found malondialdehyde and caspase-3 elevated and antioxidant enzyme activity decreased by injury, all normalized in a dose dependent manner by nortriptyline pretreatment. Zhang 2008 and Jeong 2018 (detailed under Mitochondrial burden) both used cellular oxidative and apoptotic stress endpoints as part of their neuroprotection readouts, consistent protective direction. Zhu 2024 (PMID 38480798, Tier B, gastric cancer cell lines and xenograft, detailed under Mitochondrial burden) found nortriptyline increased intracellular reactive oxygen species and malondialdehyde and depleted glutathione specifically in cancer cells via Keap1-Nrf2 pathway activation, an effect partially rescued by the antioxidant N-acetylcysteine, opposite direction, cancer cell specific.

Mechanism

Mitochondrial permeability transition pore inhibition (protective, normal tissue models) versus Keap1-Nrf2 mediated pro-oxidant induction (cancer cell specific, drug repurposing context) - two distinct, context dependent mechanisms rather than one general effect.

Counter-evidence held on record

Do not present nortriptyline as either a general antioxidant or a general pro-oxidant for a treated patient; the direction found depends entirely on tissue and disease context, and none of it is human data. This entry shares its evidentiary core with Mitochondrial burden above; see the Overlap map.

Limits on what this shows

• Why no grade: Both conditions for ungraded are met. Three original PubMed search-listing queries were run: "nortriptyline copper" returned a confirmed clean zero-result count. "nortriptyline superoxide dismutase" returned four results, none measuring copper-dependent enzyme activity in relation to nortriptyline directly (the Yazdani testicular protective study, already counted under Oxidative stress, measured general antioxidant-enzyme activity without a copper-specific marker; the remaining three were an unrelated CYP2C8 inactivation paper, an unrelated Parkinson's disease combination-therapy paper, and an unrelated irritable bowel syndrome rat oxidative-stress paper). "nortriptyline trace elements OR nortriptyline zinc OR nortriptyline ceruloplasmin" returned six results, all irrelevant to copper biology. A dedicated repair pass re-ran this category with a class-level phrasing, "tricyclic antidepressant copper metabolism," which returned 21 results, four relevant to copper biology and antidepressants in general: Schlegel-Zawadzka M, Nowak G, "Alterations in serum and brain trace element levels after antidepressant treatment. Part II. Copper," Biol Trace Elem Res 2000, PMID 10949967 (Tier C, rat, chronic imipramine and citalopram, plus electroconvulsive shock, nortriptyline not among the drugs tested) found chronic treatment did not alter brain copper levels, a null finding for two different antidepressant classes, not evidence of a copper effect for any drug including nortriptyline; a case report of acute focal dystonia from an unnamed tricyclic in a patient with pre-existing Wilson disease (a copper-storage disease, not evidence the drug altered copper status); and two rat studies of antidepressant effects on Cu/Zn-superoxide dismutase (an antioxidant enzyme, not a copper-status marker) and on serum zinc/copper ratio, neither of which could be confirmed to include nortriptyline in the tested cohort before PubMed's abstract pages and OpenAlex's search endpoint both became rate-limited or bot-challenged partway through this repair pass (see tooling disclosure). None of this newly located material meets the bar for even a yellow class-extrapolation grade, since the one drug-specific-enough finding (Schlegel-Zawadzka 2000) is a null result for other TCA/SSRI members, not supporting data, and per the governing protocol a null class-level finding argues against, not for, a claim. No absorption-interference, chelation, or copper-enzyme mechanism connecting nortriptyline to copper status is documented in the FDA label or in any source located in either research pass.

Tooling disclosure

The original three PubMed queries and this repair pass's class-level query all returned readable, non-error result sets, so the core zero-result finding for nortriptyline specifically is assessed as a genuine literature gap rather than a tooling failure. However, two follow-up queries in this repair pass, intended to confirm the exact drug panels in the Cu/Zn-SOD and zinc/copper-ratio rat studies, hit a PubMed bot-verification challenge page and a 429 rate limit on OpenAlex's search endpoint before either could be resolved, so those two studies' cohorts remain unconfirmed rather than ruled in or out. The Europe PMC REST API and Crossref API were rejected with HTTP 403 at the outbound proxy gateway itself during this repair pass (confirmed via the proxy's own status log, a persistent organization-level policy denial on these hosts, not a per-request rate limit), and OpenAlex's per-work lookup endpoint (as opposed to its search endpoint) remained the more reliable channel, consistent with the original review. A future revision with working access to the blocked hosts, or a successful PubMed abstract fetch, should re-confirm the two unresolved cohorts.

Limits on what this shows

• Why no grade: Both conditions for ungraded are met. A PubMed search for "nortriptyline iron OR nortriptyline ferritin OR nortriptyline anemia" returned six results, all irrelevant: a ferroptosis lung-cancer mechanism paper unrelated to nortriptyline's own effect on iron, a 1974 pharmacogenetics editorial, an unrelated antibacterial wound-healing hydrogel paper, an analytical spectrophotometric assay method paper using nortriptyline only as the analyte being measured, not iron as a biological variable, an unrelated infant sleep case report, and a second unrelated spectrophotometric assay method paper. A follow-up search for "nortriptyline erythrocyte OR nortriptyline hemolysis OR nortriptyline hemoglobin" returned 27 results; the first page of ten was reviewed in full and consisted entirely of tricyclic-antidepressant blood-level pharmacokinetic and forensic-toxicology papers, or diabetes-comorbidity treatment-response papers, none measuring iron status or red-cell iron content. A dedicated repair pass re-ran this category with a fresh class-level phrasing, "tricyclic antidepressant iron status deficiency," and confirmed a clean zero-result count. The FDA label's hematologic adverse-reaction list (bone marrow depression including agranulocytosis, eosinophilia, purpura, thrombocytopenia) does not include anemia, a genuine null finding recorded here rather than silently omitted.

Tooling disclosure

All three PubMed queries across the original review and this repair pass returned readable, non-error result sets. Europe PMC and Crossref were rejected with HTTP 403 at the outbound proxy gateway itself during this repair pass (a confirmed organization-level policy denial on those hosts, per the proxy's own status log, not a per-request rate limit), and a follow-up attempt to cross-check via "nortriptyline hepcidin OR nortriptyline transferrin OR nortriptyline iron absorption" was itself rate-limited (HTTP 429) before returning a result. This is assessed as a genuine literature gap for nortriptyline specifically, since the core zero-result and low-relevance findings above came from clean, readable PubMed result sets, not a blocked query; the blocked channels should still be re-checked in a future revision in case they hold something PubMed's own index does not.

Limits on what this shows

• Why no grade: Both conditions for ungraded are met, with a genuine, now more thoroughly documented access limitation disclosed below. A PubMed search for "nortriptyline gut microbiome OR nortriptyline microbiota OR nortriptyline gut bacteria" returned two results, neither a drug-specific microbiome study (an insulin-resistance and depression comorbidity review, and an unrelated irritable bowel syndrome practical-approach review). The one large-scale study that might have resolved this either way, Maier L, Pruteanu M, Kuhn M, et al, "Extensive impact of non-antibiotic drugs on human gut bacteria," Nature 2018, DOI 10.1038/nature25979, PMID 29555994, screened more than 1,000 marketed drugs against 40 gut bacterial strains and found antipsychotics, a chemically diverse class, overrepresented among growth-inhibiting drugs. A dedicated repair pass tried nine separate channels to reach the full compound list or supplementary tables and could not confirm nortriptyline's inclusion or hit status either way; the openly accessible abstract and results text on nature.com names only a handful of example drugs (metformin, posaconazole, moricizine, olanzapine, methotrexate, and others) and does not include or exclude nortriptyline, and the complete compound list lives only in Supplementary Table 1, which could not be reached (see tooling disclosure). This remains disclosed as an unresolved access gap, not as support or rebuttal for nortriptyline specifically.

Tooling disclosure

Channels tried in this repair pass, all unsuccessful for the full compound list specifically: (1) PMC full text (pmc.ncbi.nlm.nih.gov/articles/PMC6108420) served a reCAPTCHA browser-verification page instead of content. (2) europepmc.org, both the /article/MED/29555994 and /articles/PMC6108420 paths, were disallowed by that site's own robots.txt. (3) The Europe PMC REST API (www.ebi.ac.uk/europepmc) and the Crossref API were rejected with HTTP 403 at the outbound proxy gateway itself; a direct curl to these hosts, run outside the WebFetch tool as a cross-check, confirmed this is a persistent organization-level policy denial recorded in the proxy's own status log, not a transient rate limit. (4) The Semantic Scholar API returned HTTP 429 on repeated attempts spaced roughly a minute apart. (5) OpenAlex's per-DOI and per-PMID lookup endpoints returned full bibliographic metadata (title, journal, authors, citation count, open-access location) but no abstract text and no compound list; OpenAlex's search endpoint was itself rate-limited (429) when tried as a fallback. (6) Unpaywall's API confirmed the paper is green open access with its only full-text location being the same PMC6108420 record blocked in (1). (7) Nature's own supplementary-file host, static-content.springer.com, was disallowed by robots.txt. (8) The Nature.com article page itself was reachable and readable for its visible abstract and results text, but did not surface a working link to Supplementary Table 1 in the fetched content. (9) A second attempt at PMC's PDF path was rate-limited before a response could be evaluated. Because PubMed's own search index returned a genuine, readable, non-overlapping zero-relevance result for nortriptyline specifically, this is recorded as a real, disclosed evidence gap in what this review could access, not as reassurance that no effect exists. A future revision with working access to PMC full text, the Europe PMC REST API, or the Maier et al. supplementary compound list should re-check this specific question directly.

Limits on what this shows

• Key negative/null findings: A PubMed search combining nortriptyline with ATP synthase or mitochondrial respiration returned only three results: the Jeong 2018 paper already counted above, a mouse fibroblast housekeeping-gene expression paper (Sugden 2010, PMID 19074533) that does not measure ATP, and a primary astrocyte oxygen-glucose deprivation paper (Gabryel 2010, PMID 21098865) that measures cell survival via cytosolic phospholipase A2 inhibition, not ATP content directly.
• At-risk subgroups: Cannot be specified; no direct human or animal ATP measurement exists for this drug.
• Counter-evidence on record: None identified; the entry is explicitly labeled as extrapolation rather than direct evidence, and the one study that might have added direct enzyme-activity data (Scaini 2010) could not be read for its direction in this review.

Regulatory status

Not addressed in the FDA label.

Population evidence

Scaini G, Santos PM, Benedet J, et al, "Evaluation of Krebs cycle enzymes in the brain of rats after chronic administration of antidepressants," Brain Res Bull 2010, DOI 10.1016/j.brainresbull.2010.03.006, PMID 20347017 (Tier B, rat) - confirmed via OpenAlex indexed metadata to have tested nortriptyline specifically, alongside paroxetine and venlafaxine, measuring citrate synthase and succinate dehydrogenase, two Krebs cycle enzymes that feed the electron transport chain and ultimately ATP synthase, in rat brain. Neither the abstract nor full text was accessible during this review (PubMed's individual abstract page and the Crossref API for this DOI were both blocked, HTTP 429 or bot verification, across repeated attempts), so the direction of the enzyme activity change with nortriptyline specifically cannot be stated here. Separately, the mitochondrial permeability transition pore inhibition findings detailed under Mitochondrial burden (Zhang 2008, Wang 2007, Jeong 2018) imply preserved rather than impaired electron transport chain integrity, since an intact mPTP supports the proton gradient that drives ATP synthase, but none of those studies measured ATP directly either.

Mechanism

Indirect inference only, from mitochondrial membrane integrity to ATP synthesis capacity; not directly observed in any source located.

Counter-evidence held on record

This entry shares its entire evidentiary basis with Mitochondrial burden above and adds no independently confirmed data of its own; see the Overlap map. The Scaini 2010 Krebs-cycle-enzyme direction must not be presented as either increased or decreased; a future revision with working PubMed abstract or Crossref access should re-check this specific study.

Limits on what this shows

• Why no grade: Both conditions for ungraded are met, though this repair pass located a genuine drug-specific arterial-oxygenation finding that a careful read of Hard Rule 2 excludes from this category rather than upgrading it. The PubMed search for "nortriptyline erythrocyte OR nortriptyline hemolysis OR nortriptyline hemoglobin" described under Iron above (27 results, first page reviewed in full) returned no study measuring red-cell oxygen-carrying parameters. The FDA label's hematologic adverse-reaction list does not include anemia (see PK basis and the Iron entry above). A fresh repair-pass query, "nortriptyline hematocrit OR nortriptyline oxygen saturation OR nortriptyline tissue oxygenation," returned five results, four irrelevant (drug-distribution and therapeutic-drug-monitoring assay papers, and an unrelated 1985 gastric-cell secretion paper) and one directly on point: Hosseini R, Borghei P, Amali A, Heidari R, Erfanian R, "Short-term Effects of Nortriptyline on Sleep Parameters for Residual OSA after UPPP," Iran J Otorhinolaryngol 2025, trial registration IRCT20230614058482N1 (Tier A, human, N=24, prospective, one month of open-label nortriptyline in patients with residual obstructive sleep apnea after uvulopalatopharyngoplasty surgery). Mean overnight oxygen saturation rose from 94.1 to 95.4 percent alongside a drop in apnea-hypopnea index (16.8 to 11.4 events per hour) and improved Epworth Sleepiness Scale and snoring scores. This is a real, drug-specific, human, arterial-oxygenation measurement, but it was not counted as a category finding here: nortriptyline was given specifically as an intended treatment for the underlying sleep-disordered breathing, working through its known pharmacology of reducing apnea and hypopnea events (tricyclics are studied in OSA for noradrenergic effects on upper-airway dilator muscle tone and REM suppression), not through a hemoglobin, red-cell, or nutrient-style oxygen-delivery mechanism. Per Hard Rule 2 and this protocol's own specific warning for this category, a drug's intended pharmacologic effect on a clinical outcome is a mechanism clarification, not a nutrient-style oxygen-delivery finding, and grading it here would misfile a sleep-apnea treatment effect as a deficiency-adjacent claim. It is disclosed in full here, by name, rather than silently left out, consistent with how the Gabryel 2010 oxygen-glucose-deprivation finding above was handled. A second related finding, Gabryel B, Bielecka A, Stolecka A, Bernacki J, Langfort J, "Cytosolic phospholipase A2 inhibition is involved in the protective effect of nortriptyline in primary astrocyte cultures exposed to combined oxygen-glucose deprivation," 2010, PMID 21098865 (Tier B, rat primary astrocyte culture), found nortriptyline protective against a combined oxygen-glucose deprivation insult via cytosolic phospholipase A2 inhibition, again cellular hypoxia tolerance rather than hemoglobin, red-cell, or arterial-oxygenation status, and remains recorded as context alongside Mitochondrial burden above.

Tooling disclosure

Same channels as Iron above; both the original and repair-pass PubMed queries returned genuine, readable result sets, so this is assessed as a real category-definition boundary rather than a tooling failure. Europe PMC and Crossref were rejected with HTTP 403 at the outbound proxy gateway during this repair pass, a confirmed organization-level policy denial rather than a rate limit, and could not independently cross-check this category; OpenAlex's per-PMID lookup confirmed the Hosseini 2025 study's bibliographic details and abstract content.