Yang et al. 2024 — Metallothionein: a comprehensive review of classification, structure, biological functions, and applications
Yang and colleagues at the Key Laboratory of Precision Nutrition and Food Quality, College of Food Science and Nutritional Engineering, China Agricultural University, with co-authors from the College of Engineering, review the published literature on metallothionein (MT) — a cysteine-rich, low-molecular-weight metal-binding protein first isolated from equine kidneys by Margoshes and Vallee in 1957. The review covers MT classification (15 evolutionary families across vertebrates, invertebrates, plants, and microbes), structure (α- and β-domains forming a dumbbell-like tertiary fold), and five categories of biological function (heavy metal detoxification, antioxidant activity, neuroprotection, anticancer effects, anti-inflammatory effects). Sections 4.1–4.3 then catalogue applications in environmental monitoring and removal of Pb, Cd, Cr⁶⁺, and Ni from contaminated water; in disease prediction and diagnosis (Wilson disease, schizophrenia, hepatocellular carcinoma); and in skincare functional ingredients. The review is a secondary narrative synthesis with no primary measurements; its value to the wiki is conceptual — it organises the MT isoform vocabulary (MT1–MT4, MT1A/B/E/F/G/H/M/X, MT2A) and the chelation chemistry that downstream pages on peptide-based detoxification and metal-binding peptide mechanisms will need.
Why this matters
- It catalogues the metallothionein superfamily classification system of Binz and Kägi, dividing MT into 15 evolutionary families plus a Family 99 for phytochelatins and other non-proteinaceous MTs; this is the closed taxonomy the MT-mechanism literature uses.
- It documents the human MT family of four classes (MT1–MT4) encoded by 11 active genes (MT1A, MT1B, MT1E, MT1F, MT1G, MT1H, MT1M, MT1X, MT2A, MT3, MT4) clustered on chromosome 16, with tissue-expression patterns (MT1/MT2 ubiquitous, MT3 brain-dominant, MT4 stratified squamous epithelium).
- It compiles five tabulated studies (Tables 1–5) summarising heavy metal detoxification experiments (Cd-dominant), antioxidant assays, neuroprotection assays, anticancer assays, and anti-inflammatory assays, with MT source, target system, metal/disease focus, and reported effect — a convenient closed list of mechanism citations for downstream synthesis.
- It positions MT as a candidate for three application classes the wiki’s downstream programmes touch on: environmental remediation (wastewater Pb/Cd/Cr⁶⁺/Ni removal by engineered MT-expressing bacteria), disease biomarker (MT1B/MT1F/MT1H/MT1L in colorectal cancer; MT-1 in Wilson disease; MT-3 in hepatocellular carcinoma prognosis), and skincare functional ingredient (anti-UV repair via MT-1A upregulation, cadmium-toxicity protection via curcumin-induced MT2A).
Key concepts and structure
The review is organised into a brief introduction (Section 1), the classification and structure of MT (Section 2 with subsections 2.1 Classification and 2.2 Structure), five biological functions (Section 3 with subsections 3.1 Heavy metal detoxification, 3.2 Antioxidant effect, 3.3 Neuroprotective effect, 3.4 Anticancer effect, 3.5 Anti-inflammatory effect), three application areas (Section 4 with subsections 4.1 Detection and removal of heavy metal ions from the environment, 4.2 Disease prediction and diagnosis in medicine, 4.3 Development of products with skincare functions), and a conclusions section (Section 5).
Classification (source Section 2.1, p. 2)
The Binz–Kägi metallothionein superfamily comprises 15 families plus Family 99 for non-proteinaceous MTs:
| Family | Description |
|---|---|
| Family 1 | Vertebrate MTs |
| Family 2 | Mollusc MTs |
| Family 3 | Crustacean MTs |
| Family 4 | Echinodermata MTs |
| Family 5 | Diptera MTs |
| Family 6 | Nematoda MTs |
| Family 7 | Ciliata MTs |
| Family 8 | Fungi-I MTs |
| Family 9 | Fungi-II MTs |
| Family 10 | Fungi-III MTs |
| Family 11 | Fungi-IV MTs |
| Family 12 | Fungi-V MTs |
| Family 13 | Fungi-VI MTs |
| Family 14 | Prokaryota MTs |
| Family 15 | Plant MTs |
| Family 99 | Phytochelatins and other non-proteinaceous MTs |
The human MT family is divided into four classes (MT1–MT4) encoded by 11 active genes (MT1A, MT1B, MT1E, MT1F, MT1G, MT1H, MT1M, MT1X, MT2A, MT3, MT4) located in the chromosome 16 gene cluster. MT1 and MT2 are expressed in most organs and tissues; MT3 is mainly expressed in the brain; MT4 is expressed in stratified squamous epithelial cells.
Structure (source Section 2.2, p. 2)
The MT apoprotein has no typical secondary structure; tertiary structure forms only after metal binding, with the fold depending on the nature and quantity of bound metal ions. Metal binding occurs principally through the thiol groups of cysteine residues, forming MT–metal clusters with strong affinity for Cu²⁺ and Zn²⁺. Mammalian MT has a unique three-dimensional structure containing two independent structural domains — N-terminal β-domain (three metal-binding sites) and C-terminal α-domain (four metal-binding sites) — that combine into a dumbbell-like fold. The β- and α-domain NMR structures are shown using rat MT-2 (PDB 1MRT and 2MRT) as the model.
Heavy metal detoxification (source Section 3.1, p. 3–5; Table 1)
The review states the reaction rate constant of MT with hydroxyl radicals is approximately 300 times that of glutathione, attributing this to the high sulfhydryl content. Cadmium poisoning is described as causing organ damage (liver, kidney), osteoporosis, increased cancer risk, immune-system effects, and metabolic disorders; metallothionein was first discovered in cadmium-rich environments bound to cadmium.
Studies on the role of MT in heavy metal detoxification (source Table 1; reproduced verbatim from the source’s own tabulation):
| MT source | Target of action | Metal ion | Effect (per source) | Source ref |
|---|---|---|---|---|
| Anabaena PCC 7120 NmtA | E. coli cells | Cd²⁺ | NmtA-expressing E. coli exhibits better growth at certain cadmium concentrations | [27] |
| MT expressed by Alishewanella sp. WH16-1-MT | Rice | Cd²⁺ | Increased plant height, spike length, and thousand-grain weight; significant reduction in Cd²⁺ content in brown rice, rice husk, roots, and shoots | [28] |
| S. cerevisiae expressing PtMT2b (from Populus trichocarpa) | S. cerevisiae | Cd²⁺ | Enhanced Cd²⁺ tolerance in S. cerevisiae | [29] |
| Ipomoea aquatica IaMT expressed in E. coli | E. coli | Cd²⁺ | Increased tolerance to and accumulation of Cd²⁺ in E. coli | [30] |
| Rabbit liver MT-2 | Grass carp | Cd²⁺ | Reduced cadmium levels in kidneys and blood; attenuated organ damage | [31] |
| Commercial MT | PC12 cells | As³⁺ | Reduced As³⁺-induced metabolic disturbances; inhibited ROS accumulation | [32] |
| ShMT (freshwater crab Sinopotamon henanense) expressed in E. coli | Metal ions in solution | Zn²⁺, Cu²⁺, Cd²⁺ | Strong binding ability; affinity order Cu > Cd > Zn | [33] |
| Human MT HsMT1L expressed in tobacco | Tobacco | Zn²⁺, Cd²⁺ | Increased accumulation of Zn²⁺ and Cd²⁺ in tobacco; enhanced tolerance | [34] |
| MT expressed by engineered bacterium EcN-MT (E. coli Nissle, per ref 35) | Mice | Cd²⁺ | Significant reduction in liver Cd content; accelerated Cd metabolism | [35] |
| MT expressed by Rhizobium leguminosarum (pea MT gene) | CdCl₂ solution | Cd²⁺ | Enhanced Cd tolerance in peas | [36] |
The review also describes specific experimental observations from the underlying primary literature without tabulation: PtMT2b expression in S. cerevisiae enabled Cd²⁺ tolerance up to 50 μM (growth completely inhibited above that concentration; ref 29); MT chelation of As³⁺ scavenges ROS and alleviates As³⁺ cytotoxicity (ref 32); ShMT has a Cu > Cd > Zn affinity order (ref 33).
Antioxidant effect (source Section 3.2, p. 5–6; Table 2)
MT exhibits approximately 50-fold higher antioxidant activity against oxidative DNA damage and approximately 10-fold higher antioxidant activity against lipid peroxidation than glutathione (source’s narrative citing ref 44; original primary reference not specified in the body text). MTF-1 and Nrf2 regulate MT expression via the antioxidant-response element (ARE) in the promoter region; Nrf2 and its downstream antioxidant genes may also be regulated by MT. Table 2 tabulates eight MT-source/target/effect tuples spanning E. coli, mouse lungs, Caenorhabditis elegans, yeast and Arabidopsis thaliana, HT1376 bladder carcinoma cells, 3T3-L1 adipocytes, mouse cardiomyocytes, and transgenic Arabidopsis (refs 46, 45, 47, 48, 42, 49, 50, 51, 52, 53, 54).
Neuroprotective effect (source Section 3.3, p. 6–7; Table 3)
The review describes Alzheimer’s disease (Aβ deposition rich in Zn and Cu metal ions) and Parkinson’s disease (α-synuclein/Lewy-body formation in the presence of metal ions) as the two principal neurodegenerative settings for MT. Zn₇MT-3 removes Cu(II) from α-syn-Cu(II) complexes and inhibits the toxic effects of α-syn-Cu(II) (ref 60). Treatment with human MT-2 peptide (hMT2) mitigated paraquat-induced brain damage in zebrafish (ref 61). Table 3 tabulates eight MT-source/target/effect tuples on neuroprotection across aged transgenic C. elegans, dentate granule cells, α-syn-Cu(II) complexes, zebrafish brain, astrocytes, spinal motor neurons, central nervous system, and dentate granular cell layer (refs 58, 59, 60, 61, 62, 8, 63, 64).
Anticancer effect (source Section 3.4, p. 7–9; Table 4)
The review identifies a negative correlation between MT expression and tumour grade/inflammation level in feline injection site fibrosarcoma; MT down-regulation may raise DNA damage and cancer risk (ref 9). MT1M overexpression in esophageal squamous cell carcinoma induces apoptosis, reduces cell viability, and inhibits epithelial-mesenchymal transition. Overexpression of MT1G and MT2A synergistically enhances the anticancer effect of cannabidiol in colorectal cancer (ref 67). MT1E knockdown decreases apoptosis in hepatocellular carcinoma (ref 68). MT1M inhibits proliferation, migration, and invasion of gastric cancer cells and increases 5-fluorouracil sensitivity (ref 69). MT1G inhibits proliferation/migration/invasion of hepatocellular carcinoma cells in vivo and in vitro, with synergistic inhibition with sorafenib (ref 70). MT2A overexpression inhibits proliferation/migration of colorectal cancer cells (ref 72). MT1G limits activin A secretion in pancreatic ductal adenocarcinoma, overcoming gemcitabine resistance (ref 73). MT2A knockdown increases cisplatin sensitivity in malignant pleural mesothelioma (ref 75) and osteosarcoma (ref 76). Carbon monoxide reduces nuclear MT levels in ovarian cancer cells, enhancing cisplatin effect (ref 77). Table 4 tabulates 11 MT-isoform/cancer-type/effect tuples.
Anti-inflammatory effect (source Section 3.5, p. 9–11; Table 5)
The primary mechanism is ROS scavenging; MT also acts as a zinc chaperone, triggering matrix metalloproteinases (MMPs) that facilitate tissue repair during inflammation (ref 10). Tabulated studies (Table 5) span MT isoforms across osteoarthritis (MT-1; ref 80), non-alcoholic steatohepatitis (MT1; ref 81), LPS-induced macrophage inflammation (MT1G; ref 78), colitis (MT1+MT2; ref 82; MT-2; ref 88), As³⁺-induced inflammation (MT; ref 83), neuroinflammation (MT1; ref 79), non-canonical inflammasome activation (MT3; ref 84), inflammatory liver injury (MT; ref 85), ankylosing spondylitis (MT-1; ref 86), pre-eclampsia (MT; ref 87), alcoholic hepatitis (MT1+MT2; ref 89), rheumatoid arthritis (MT-1+MT-2; ref 90).
Environmental detection and removal (source Section 4.1, p. 12)
MT is positioned as a biomarker for aquatic-organism exposure to heavy metal ions; fish-MT-specific polyclonal antibodies enable early-warning detection of pollution (ref 93). Plankton MT reflects organism heavy-metal contamination and supports environmental quality assessment (ref 94). Fish MT levels increase as a result of metal pollution in waters (ref 95).
Specific engineered-MT bioremediation results cited:
- Cellulose–MT biosorbent removes Pb(II) and Zn(II) from polluted water (ref 97).
- SmtA-modified selenium nanoparticles adsorb Cd²⁺ and Pb²⁺ from wastewater to meet Chinese national wastewater discharge standards (ref 96).
- E. coli genetically modified with the MT gene and co-assembled with magnetic nanoparticles removed Pb²⁺ and Cd²⁺ with >80 % efficiency (ref 98).
- E. coli expressing MT2A and MT3 effectively removed Cr⁶⁺ from aqueous solution via hydroxyl, phosphoryl, and carbonyl functional groups (ref 99).
- MT3-expressing E. coli encapsulated in calcium-alginate bio-beads removed Cu, Zn, and Cd from water with the most significant effect on Cu (ref 100).
- E. coli overexpressing MTA achieved a seven-fold increase in nickel bioaccumulation from mine wastewater compared with the control (ref 101).
Disease prediction and diagnosis (source Section 4.2, p. 12–13)
MT1E and MT1F as biomarkers for hepatocellular carcinoma recurrence (down-regulation correlates with cancer risk; ref 103). MT-1 reduction associated with elevated schizophrenia risk via oxidative damage (ref 102). MT immunostaining as a useful diagnostic tool for Wilson disease (a copper-metabolism inherited disorder; ref 104). MT-3 included in a seven-oxidative-stress-related gene panel for hepatocellular carcinoma prognosis (ref 105). MT1B, MT1F, and MT1G down-regulation in colorectal cancer; MT1B, MT1H, and MT1L expression levels predict colorectal cancer prognosis (ref 106). MT1X expression predicts tumorigenesis and clear-cell renal-cell carcinoma prognosis (ref 107). MT mRNA correlates positively with overall survival, first-progression survival, and post-progression survival in gastric cancer (ref 108). MT1JP expression provides reference for glioma diagnosis (ref 109).
Skincare applications (source Section 4.3, p. 13)
MT1X expression reduces allergen-induced inflammatory response in dermatitis patients; MTF1 expression prevents allergen-induced oxidative stress (ref 110). MT binds cytoplasmic zinc and controls zinc homeostasis, delaying the onset of hand-foot skin reaction (ref 111). 1,25-dihydroxyvitamin D3 decreases UV-induced sun-damaged cells and improves skin resistance via MT synthesis stimulation (ref 112). Isoflavonoid-containing skincare products elevate epidermal MT expression after UV-simulated radiation (ref 113). MT-null mice show increased UVB-induced sun damage and skin thickening (ref 114). MT-1A elevation after sun exposure is linked to skin repair (ref 115). Curcumin protects keratin-forming cells from cadmium-induced apoptosis by upregulating MT2A (ref 116). Skin-wound newly-growing epidermis expresses more MT than surrounding normal skin, contributing to anti-inflammatory zinc activity and MMP-1-driven keratinocyte migration (ref 117).
Methods (brief)
The paper is a narrative review; it reports no original experimental work, no primary chemical or biological measurements, and no analytical methods of its own. The reference list contains 117 entries spanning 1957 (the founding Margoshes and Vallee paper) to 2023, drawn from journals including J. Am. Chem. Soc., Cell Mol. Life Sci., Antioxidants, Aquat. Toxicol., Environ. Res., Ecotoxicol. Environ. Saf., J. Agric. Food Chem., Inflamm. Res., Sci. Rep., Front. Microbiol., Front. Endocrinol., Cancer Cell Int., Chemosphere, ACS Appl. Mater. Interfaces, Biomedicines, Histopathology, and others. The review does not declare a formal search strategy, inclusion/exclusion criteria, PRISMA flow, or risk-of-bias assessment — it is a narrative review in the older tradition rather than a systematic review.
The journal (Antioxidants, MDPI) is open-access; the article is published under CC BY 4.0. Article processing charges are paid by the authors. Funding declared: National Key R&D Program of China (grant 2020YFD1000300) and Beijing Fishery Innovation Team (grant BAIC07-2023-13). Conflicts of interest: the authors declare none. Academic editor: Simone Ciofi-Baffoni. Received 27 May 2024; revised 24 June 2024; accepted 26 June 2024; published 9 July 2024.
Implications
- Certification: The review contributes no occurrence data and no exposure data, so it does not move any HMTc threshold-setting work. Its value for HMTc is indirect — it provides background on MT-based biological detoxification mechanisms relevant to any future mitigation chapter and underpins the MT-versus-phytochelatin distinction in the peptide-binding-mechanism literature.
- App: No routing to ingredient or product pages. This source contributes background reading for the metals/cadmium, metals/lead, metals/arsenic, metals/chromium-hexavalent, metals/nickel, and metals/mercury pages on the topic of MT-mediated binding, sequestration, and detoxification mechanisms, not on contamination occurrence.
- Courses: Useful as a single-source orientation to the metallothionein vocabulary (MT1–MT4 isoforms, the 15-family Binz–Kägi classification, the α-/β-domain dumbbell structure) for an educator-audience module on heavy-metal detoxification mechanisms. Should not be cited as the authority for any specific quantitative claim; trace claims to their primary references first.
- Microbiome: Marginal. The review touches microbial detoxification (engineered MT-expressing E. coli, S. cerevisiae, Rhizobium leguminosarum, Alishewanella) but does not engage the gut microbiome or the heavy-metal-microbiome axis. WikiBiome federation is unlikely to draw on this source.
Limitations
This is a narrative review with no declared search strategy, no inclusion or exclusion criteria, no PRISMA flow, and no risk-of-bias assessment. Quantitative claims (the 300-fold hydroxyl-radical reaction rate constant versus glutathione, the 50-fold antioxidant activity against oxidative DNA damage, the 10-fold antioxidant activity against lipid peroxidation, the >80 % Pb²⁺/Cd²⁺ removal efficiency by genetically modified E. coli, the seven-fold nickel bioaccumulation increase) are reported with attribution to single primary references but without consistent context (assay conditions, dose-response framing, replicate counts) and should be traced to the underlying primary papers before propagation to downstream synthesis pages. The five tabulated effect summaries (Tables 1–5) collapse heterogeneous experimental systems (cell culture, transgenic plant, fish injection, mouse model) into a single “effect” column and the directionality language is the review authors’ interpretation rather than verbatim primary-source claims. The review’s coverage of mercury and arsenic is comparatively thin: mercury appears only in passing context (Wilson disease being copper-specific is the only metal-disorder pairing developed at length), and arsenic appears principally in the As³⁺-cytotoxicity primary reference 32 and the As³⁺-inflammation primary reference 83. Speciation (iAs vs tAs; MeHg vs tHg) is not addressed by the review authors.
Wiki pages this source may touch
Verification notes
Existing-page check. DOI grep (10.3390/antiox13070825), raw_handle grep (MFK_05-metallothionein-a-comprehensive-review-of-its-c), and cite-key glob (yang2024-metal*) over wiki/sources/ on 2026-06-08 returned no hits. This is a NEW source page — no prior version to merge-enhance. (The folder also contains a duplicate-titled PDF, 21_Metallothionein_A_Comprehensive_Review.pdf, at the identical filesize 1,044,450 bytes; that file is almost certainly the same DOI and should be skipped as a downstream duplicate when its turn comes in the manual-fetch queue.)
Evidence tier. B (secondary narrative review). The paper reports no primary measurements and declares no systematic search strategy. A-tier is reserved for primary peer-reviewed studies and authoritative agency monographs; this is neither.
Metals frontmatter. The review explicitly discusses Cd (dominant; Sections 3.1, 3.5, 4.1, 4.3), Pb (Section 4.1 wastewater removal), As³⁺ (Sections 3.1 and 3.5 — primary references 32 and 83), Cr⁶⁺ (Section 4.1 wastewater removal by MT2A/MT3-expressing E. coli), Ni (Section 4.1 nickel bioaccumulation by MTA-overexpressing E. coli), and Cu/Zn extensively but as physiological-essential context rather than contamination context. Mercury appears only in the introductory chelation-affinity sentence (“strong binding capacity for metal ions such as Cu²⁺ and Zn²⁺ … Cd …”) and in the Cu > Cd > Zn affinity order of ShMT; mercury speciation (MeHg, iHg, tHg) is not addressed. From the HMTc 10-analyte priority list, Cd, Pb, iAs, Cr-VI, Ni, and tHg are recorded in the metals: frontmatter. Arsenic is recorded as iAs (not tAs) because the source consistently uses As³⁺ (trivalent inorganic arsenic) wherever it specifies a species. Mercury is recorded as tHg (not MeHg) because the source uses unspeciated “mercury”/“Hg” only and never specifies methylmercury. Cu, Zn, Fe, Al are out-of-scope for the HMTc-priority detoxification framing here and are not added to frontmatter.
Ingredients, products, matrices, jurisdictions frontmatter. All empty. The source measures nothing in any food, beverage, personal-care, or environmental sampling matrix; it reviews MT biology and applications conceptually. Skincare-functional-ingredient discussion in Section 4.3 names MT as the active rather than measuring metals in skincare products, so no products: slug applies. The remediation experiments in Section 4.1 use synthetic wastewater or “polluted water” generically rather than a sampled matrix from any specific jurisdiction. The corresponding author affiliations are Beijing (China), but the review’s literature scope is international and no national regulatory or occurrence framework is applied.
Sample size. Null. This is a review with no sampling frame.
Brand firewall (Part 12). No brand names appear in the source. Engineered bacterial strain names (Anabaena PCC 7120 NmtA, Alishewanella sp. WH16-1-MT, EcN-MT) and reference materials (rat MT-2 PDB 1MRT/2MRT) are scientific-method vendor/material names per the Exception 2 carve-out and are preserved. No firewall action required.
HMTc firewall (Part 2). The review contains no HMTc-threshold language, no “consistent with the literature consensus” framing in either direction, and no consumer-audience risk advisories. Skincare-product application discussion in Section 4.3 is framed as MT being an active ingredient with antioxidant/anti-UV/anti-cadmium properties — it does not propose any HMTc threshold for metals in skincare products and is therefore firewall-clean. No firewall action required.
Scope fit. This paper is in scope per the 2026-06-02 commit 3f47f95 — scope: mitigation/remediation is in-scope, not a skip. The “Black Market Peptide Metal Survey” folder context suggests Karen is collecting peptide-metal literature to inform either the peptide-therapeutic-contamination programme (research peptides sold as supplements may carry heavy-metal residues from synthesis catalysts) or the mitigation/remediation-evidence page. This source supports the latter but not the former: nothing in the paper addresses contamination of MT preparations themselves.
Date arithmetic. Received 27 May 2024, revised 24 June 2024, accepted 26 June 2024, published 9 July 2024 — all consistent with the year frontmatter (2024) and the citation. Article DOI 10.3390/antiox13070825 resolves to Antioxidants 2024, Vol 13, Article 825.
Audit subagent (2026-06-08) verdict: REVISE. Five checks (numerical fidelity, slug vocabulary, speciation/methods, brand firewall, HMTc firewall) returned ✅✅✅✅✅ at the body level with two ⚠️ findings, both minor; no ❌. (1) Table 1 row 9 initially read “MT expressed by engineered Rhodococcus (EcN-MT)”; verified against the PDF Table 1 Cont. on p. 5 and reference [35] (Zou et al. 2022, Front. Pharmacol. 13:857869, “Engineered Bacteria EcN-MT Alleviate Liver Injury in Cadmium-Exposed Mice via its Probiotics Characteristics and Expressing of Metallothionein”) — the PDF says “engineered bacterium EcN-MT” and EcN denotes E. coli Nissle, not Rhodococcus; corrected to “engineered bacterium EcN-MT (E. coli Nissle, per ref 35)“. (2) The auditor flagged [[mitigation/remediation-evidence]] in ## Wiki pages this source may touch as outside the taxonomy snapshot’s vocabulary; verified against wiki/mitigation/remediation-evidence.md (which exists) and against the 2026-06-02 scope commit 3f47f95 — scope: mitigation/remediation is in-scope, not a skip — finding is a false positive because the taxonomy snapshot scopes ingredients/products/metals/regulations as GPT’s drafting vocabulary, not the full wiki section index; the mitigation/ subtree is a legitimate routing destination (and the same wikilink form is used in the sibling luo2024-peptides-heavy-metal-remediation page). No change applied on finding (2).
Page history
The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.
| Commit | Date | Description |
|---|---|---|
| f4c7a4e | 2026-06-08 | ingest: jarin2025-plant-responses-heavy-metal-stresses fresh from MFK/June 8 Kimi_Agent_Black Market Peptide Metal Survey/heavy_metals_peptides |