Cobbett & Goldsbrough 2002 — Phytochelatins and metallothioneins: roles in heavy metal detoxification and homeostasis

Cobbett (Department of Genetics, University of Melbourne) and Goldsbrough (Department of Horticulture and Landscape Architecture, Purdue University) review the two best-characterised classes of cysteine-rich heavy-metal-binding ligands in plant cells: the phytochelatins (PCs), enzymatically synthesised peptides derived from glutathione, and the metallothioneins (MTs), gene-encoded polypeptides. The review covers PC structure and biosynthesis, the identification of PC synthase genes across plant, animal, and fungal taxa, regulation of PCS at the enzyme and gene-expression levels, vacuolar sequestration of PC-metal complexes, the role of sulfide in HMW complex formation, MT structural classification into four plant types (Type 1–4) based on cysteine arrangement, MT gene organisation in the Arabidopsis genome, MT protein characterisation, MT gene expression patterns, and the still-unresolved question of MT function in plants. The review is a comprehensive narrative synthesis at the time of completion of the Arabidopsis genome sequence; it is one of the canonical entry-points to the PC/MT plant biology literature. Its value to the wiki is biological background for the cadmium, arsenic, mercury, lead, copper, zinc, and nickel pages, and for the remediation-evidence section when it discusses phytoremediation as a research goal. The review reports no contamination occurrence data in any food matrix.

Why this matters

• This is the foundational 2002 review establishing the dichotomy and the eventual convergence between phytochelatins (originally considered plant-specific) and metallothioneins (originally considered animal-specific) as the two best-characterised heavy-metal-binding ligand classes in plant cells. The review documents that the cloning of PC synthase genes from plants in 1999 and the subsequent identification of functional PCS homologues in Caenorhabditis elegans, Caenorhabditis briggsae, the slime mould Dictyostelium discoideum, and partial sequences in Chironomus (aquatic midge) and earthworms discarded the artificial PC-versus-MT taxonomic split.
• It establishes the canonical PC biosynthesis pathway diagram (Figure 1, p. 163) covering GCS/GS-mediated GSH biosynthesis, PCS-catalysed PC formation from GSH in the presence of Cd²⁺, LMW PC-Cd complex formation, hmt1-mediated vacuolar transport in S. pombe (with ycf1 as the S. cerevisiae analogue), and HMW PC-CdS complex formation in the vacuole with sulfide derived from cysteine sulfinate via ade2/6/7/8 and hmt2.
• It catalogues the PCS gene family across taxa (Table 2, p. 164): AtPCS1 and AtPCS2 in Arabidopsis, BjPCS1 in Brassica juncea, TaPCS1 in wheat, OsPCS1 in rice, SpPCS1 in Schizosaccharomyces pombe, CePCS1 in Caenorhabditis elegans, DdPCS1 in Dictyostelium discoideum — with predicted molecular weights ranging from 42 kD to 70 kD and a conserved N-terminal catalytic domain plus variable C-terminal cysteine-rich domain.
• It establishes the plant MT classification system (Figure 2, p. 171) into four types based on cysteine arrangement, replacing earlier ad hoc classifications: Type 1 MTs (Cys-Xaa-Cys motifs in two domains separated by ~40 amino acids), Type 2 MTs (first pair of cysteines as Cys-Cys in positions 3–4, with conserved N-terminal MSCCGGNCGCS motif), Type 3 MTs (only four Cys in the N-terminal domain with the conserved Gln-Xaa-Lys-Lys-Gly motif; expressed in ripening fruits), Type 4 MTs (three cysteine-rich domains, exemplified by the wheat Ec protein; expressed in developing seeds with ABA-response promoter elements). The Arabidopsis MT gene family (MT1a, MT1c, MT2a, MT2b, MT3, MT4a, MT4b plus the MT1b pseudogene) is laid out on chromosomes 1, 2, 3, and 5.
• It documents the asymmetric state of evidence between PCs (whose detoxification role is supported by PC-deficient mutants like cad1 in Arabidopsis and corresponding S. pombe mutants) and MTs (where, at the time, no MT-deficient mutants in Arabidopsis had been characterised and the function in plants remained an “enigma”). The review explicitly notes that PCs and MTs may not even overlap in their spheres of function in plant cells.

Key concepts and structure

The review is organised into three main sections following the introduction: Phytochelatins (covering structure and biosynthetic pathway, PC synthase gene identification, animals that express PC synthase, PCS regulation, vacuolar sequestration, sulfide ions, metals other than Cd, and the roles of PCs); Metallothioneins (covering structure, gene structure, MT proteins, gene expression, and function); and Future Prospects.

Phytochelatin structure and biosynthesis

PCs form a family with the general structure (γ-GluCys)_n-Gly, where n is generally 2 to 5. Structural variants identified in some plant species include (γ-GluCys)_n-β-Ala (homophytochelatins, in legumes), (γ-GluCys)_n-Ser (hydroxymethyl-PCs), and (γ-GluCys)_n-Glu. PCs are structurally related to glutathione (GSH; γ-GluCysGly), and GSH is the substrate for PC biosynthesis. GSH-deficient mutants of S. pombe and Arabidopsis are PC-deficient and hypersensitive to Cd.

The biosynthetic pathway and its mutants (Table 1, p. 162):

• GSH biosynthesis — Gsh1 (γ-glutamylcysteine synthetase, in S. pombe) and CAD2/RML1 (in Arabidopsis) encode γ-glutamylcysteine synthetase. Gsh2 (in S. pombe) encodes glutathione synthetase.
• PC biosynthesis — CAD1 (Arabidopsis), Pcs1 (S. pombe), Pcs1 (C. elegans) encode PC synthase. The cad1 Arabidopsis mutants are Cd-sensitive, PC-deficient, but have wild-type GSH levels.
• PC function — Hmt1 (S. pombe) encodes a PC-Cd vacuolar membrane ABC-type transporter. ade2, 6, 7, 8 (S. pombe) catalyse cysteine sulfinate metabolism feeding sulfide biosynthesis. Hmt2 (S. pombe) encodes a mitochondrial sulfide:quinone oxidoreductase for sulfide detoxification. Hem2 (Candida glabrata) encodes porphobilinogen synthase, a cofactor for sulfite reductase in siroheme biosynthesis.

Identification of PC synthase genes

PC synthase was first characterised enzymatically by Grill et al. in 1989, but the genes were not cloned until 1999. Three independent approaches converged on the same gene: two groups used Arabidopsis and wheat cDNA libraries in S. cerevisiae to identify genes (AtPCS1 and TaPCS1) conferring increased Cd resistance; the third group identified AtPCS1 by positional cloning of the CAD1 gene. A S. pombe homologue (SpPCS1) and a C. elegans homologue (CePCS1) were subsequently characterised. Heterologous expression and tagged-protein purification demonstrated that these gene products are both necessary and sufficient for GSH-dependent PC biosynthesis in vitro.

Surprisingly, PC synthase homologues are absent from S. cerevisiae, Drosophila melanogaster, mouse, and human genomes. One initial hypothesis — that PC synthase is restricted to organisms with aquatic or soil habitats — is undermined by the report of a partial PC synthase–homologous EST in the mosquito-borne parasitic nematode Brugia malayi. Suppression of CePCS1 in C. elegans by RNA interference produces Cd sensitivity, confirming an essential role for PCs in heavy metal detoxification in animals.

PCS enzymes and regulation

The predicted molecular weights of PC synthase enzymes range from 42 kD to 70 kD (Table 2, p. 164). N-terminal regions across plant, yeast, and animal PC synthases share 40–50% identity; C-terminal regions show little apparent conservation at the amino acid level but share the feature of multiple cysteine residues, often as adjacent pairs (CC) or near pairs. AtPCS1 has 10 Cys residues (4 as adjacent pairs) in its C-terminal region; SpPCS1 has 7 Cys (6 as adjacent pairs).

The first-characterised PC synthase activity (from cultured Silene cucubalis cells) was described as a γ-GluCys dipeptidyl transpeptidase (EC 2.3.2.15) catalysing transpeptidation of γ-GluCys from GSH onto a second GSH (forming PC2) or onto a PC_n molecule (forming PC_n+1). The enzyme was reported as a tetramer of Mr 95,000 with K_m for GSH of 6.7 mM — but the deduced MWs of cloned plant PCS genes (42–70 kD) and the absence of evidence for multimerisation in cloned-enzyme studies suggest the S. cucubalis MW determination may have reflected a protein mixture. PC synthase activity has also been detected in pea, tomato, and Arabidopsis.

Activation pattern. PC biosynthesis can be induced by a range of metal ions in S. pombe and in intact plants and plant cell cultures. In vitro studies of partially purified PC synthase from S. cucubalis and recombinant enzyme in E. coli or S. cerevisiae found activation by Cd, Cu, Ag, Hg, Zn, and Pb ions, with Cd being the best activator. The activation list for the original S. cucubalis enzyme is Cd, followed by Ag, Pb, Bi, Zn, Cu, Hg cations (PDF p. 166).

Mechanism of activation. Early models assumed direct metal–enzyme interaction. The Vatamaniuk et al. 2000 study (PDF p. 166, reference [79]) demonstrated that metal binding to the enzyme per se is not responsible for catalytic activation. AtPCS1 binds Cd at high affinity (K_d = 0.54 ± 0.20 µM) and high capacity (stoichiometric ratio = 7.09 ± 0.94) but has much lower affinity for Cu, even though Cu activates equally well. The current model is that heavy metal glutathione thiolates (e.g., Cd·GS₂) and free GSH act as γ-Glu-Cys acceptor and donor — the metal ion forms part of the substrate (a metal thiolate), not a direct enzyme effector. S-alkylglutathiones (where the SH is blocked) can also participate in PC biosynthesis in the absence of heavy metals, consistent with this substrate-blocking model. Termination of biosynthesis occurs simply through exhaustion of metal-thiolate substrate.

Expression and regulation. PC synthase is expressed constitutively; mRNA levels of AtPCS1/CAD1 are not influenced by exposure to Cd or other metals (i.e., no transcriptional regulation in Arabidopsis). However, TaPCS1 mRNA levels in wheat roots increase on Cd exposure, suggesting transcriptional regulation exists in some species. PC biosynthesis occurs within minutes of Cd exposure and is independent of de novo protein synthesis, indicating regulation is primarily through activation of pre-existing enzyme. Tissue specificity is incompletely characterised: in tomato, activity was detected in roots and stems but not leaves or fruits.

Vacuolar sequestration

Both plant and yeast PC-Cd complexes are sequestered to the vacuole. In S. pombe, two PC-Cd complex classes — HMW and LMW — are resolved by gel-filtration chromatography; the hmt1 mutant is unable to form HMW complexes. Hmt1 is an ABC-family transporter located in the vacuolar membrane; both Hmt1 and ATP are required for LMW PC-Cd transport into vacuolar membrane vesicles. In S. cerevisiae, which apparently does not express a PC synthase, YCF1 (also ABC-family) carries (GSH)_2-Cd complexes to the vacuole. In C. elegans, mutations affecting ABC transporters also confer heavy metal sensitivity.

In plants, sequestration of PCs to the vacuole has been observed in tobacco mesophyll protoplasts (where almost all Cd and PCs accumulated were confined to the vacuole). An ATP-dependent, proton-gradient-independent transport activity for PCs and PC-Cd complexes has been identified in oat roots, but the plant gene encoding this function has not been identified. An inventory of ABC transporter genes in the Arabidopsis genome has not revealed a clearly identifiable HMT1 homologue.

Sulfide ions and HMW complex formation

In S. pombe, Candida glabrata, and some plants, sulfide ions are important for Cd detoxification. HMW PC-Cd complexes contain both Cd and acid-labile sulfide; sulfide incorporation increases the amount of Cd per molecule and the stability of the complex. Some complexes contain a CdS crystallite core (20-Å-diameter particles) coated with PCs, with a comparatively high S²⁻:Cd ratio. Cd-sensitive mutants of S. pombe (deficient in PC-Cd complexes) include ade-pathway mutants where cysteine sulfinate metabolism is disrupted; in C. glabrata, the hem2 mutant (porphobilinogen synthase) shows Cd sensitivity through a siroheme/sulfite-reductase deficiency. The hmt2 mutant of S. pombe (mitochondrial sulfide:quinone oxidoreductase) hyperaccumulates sulfide and shows Cd sensitivity; the role of HMT2 in Cd tolerance may be to detoxify excess sulfide generated during HMW PC-Cd complex formation.

Metals other than Cd

Although PC synthase activation in vitro and PC induction in vivo occurs with a range of metal ions, evidence supporting a role for PCs in detoxification of metals other than Cd is limited. Maitani et al. used ICP-AES with HPLC separation in Rubia tinctorum roots and found that the most effective PC inducers were Ag, arsenate, Cd, Cu, Hg, and Pb; the only PC complexes identified in vivo were with Cd, Ag, and Cu ions. PC complexes formed in response to Pb and arsenate contained copper, not the inducing metal — suggesting metal exchange between PC complexes. Schmöger et al. clearly demonstrated formation of PC-As complexes in vivo and in vitro.

PC-synthase-deficient mutants of Arabidopsis and S. pombe show similar but not identical sensitivity patterns: both are highly sensitive to Cd and arsenate; both show little or no sensitivity to Cu, Hg, Ag, Zn, Ni, and selenite. Suppression of PC synthase in C. elegans gives Cd sensitivity, but other-metal responses have not been reported. Thus, despite Cu being a strong activator of PC biosynthesis in vivo and in vitro, PC-deficient mutants show relatively little sensitivity to Cu — suggesting PC-Cu complexes may be poorly sequestered to the vacuole, comparatively transient, or that an alternative more effective mechanism exists for Cu detoxification.

The roles of PCs in unpolluted environments

Most experimental PC studies use Cd concentrations above 1 µM, while nonpolluted soils may have solutions containing up to 0.3 µM Cd. Wagner has argued that PCs play a role only at high (non-natural) Cd exposure. Counter to this: a PC-deficient Arabidopsis mutant is highly sensitive to Cd as low as 0.6 µM in agar medium, and even at Cd concentrations where the mutant is not obviously sensitive, the wild type may have a selective advantage. The review’s stance is that PCs likely have a role in heavy metal detoxification even in unpolluted environments.

Metallothionein classification

MT proteins are low-molecular-weight, cysteine-rich, metal-binding proteins. They typically contain two metal-binding, cysteine-rich domains giving a dumbbell conformation. The pre-2002 classification distinguished:

• Class I MTs — 20 highly conserved Cys residues arranged as in mammalian MTs; widespread in vertebrates.
• Class II MTs — without this strict cysteine arrangement; includes plant, fungal, and nonvertebrate animal MTs.
• Class III MTs — confusingly, this was the term applied to PCs.

The review proposes a four-type plant MT classification (Figure 2, p. 171) based on cysteine arrangement, building on the Robinson et al. system:

• Type 1 MTs — six Cys-Xaa-Cys motifs distributed equally between two domains, separated by ~40 amino acid spacer containing aromatic residues. Brassicaceae Type 1 MTs (Arabidopsis, Brassica) have a much shorter spacer and an additional Cys residue.
• Type 2 MTs — two cysteine-rich domains separated by ~40 amino acid spacer; first pair of Cys present as Cys-Cys motif at positions 3–4; Cys-Gly-Gly-Cys motif at end of N-terminal cysteine-rich domain. N-terminal domain conserved as MSCCGGNCGCS. C-terminal domain has three Cys-Xaa-Cys motifs. Spacer region more variable between species than in Type 1.
• Type 3 MTs — only four Cys residues in N-terminal domain. Consensus sequence for the first three is Cys-Gly-Asn-Cys-Asp-Cys; the fourth Cys is part of the conserved Gln-Xaa-Lys-Lys-Gly motif. C-terminal domain has six Cys arranged as Cys-Xaa-Cys motifs. Two domains separated by ~40 amino acids.
• Type 4 MTs — three cysteine-rich domains each with 5 or 6 conserved Cys, separated by 10 to 15 residues. Most cysteines present as Cys-Xaa-Cys motifs. Type 4 MTs from dicots contain an additional 8 to 10 amino acids in the N-terminal domain before the first cysteine residue. Exemplified by the wheat Ec protein, the first characterised plant MT.

The vast majority of plant MT genes are found in angiosperms. Some species (Arabidopsis, rice, sugarcane) contain genes encoding all four types, indicating the four MT types evolved before the monocot-dicot split. The MT-encoding gene from Fucus vesiculosus (brown alga) does not fit readily into any of the four plant types but is equally similar to Arabidopsis MT1a and to an oyster MT.

MT gene structure and Arabidopsis MT family

Almost all plant MT genes contain an intron located close to the end of the N-terminal cysteine-rich domain, with position varying by MT type. The Arabidopsis MT family is organised across multiple chromosomes:

• Chromosome 1 — MT1a and MT1c lie within 4 kb as an inverted repeat.
• Chromosome 2 — MT4a and MT4b (not closely linked).
• Chromosome 3 — MT2a and MT3 at distinct positions.
• Chromosome 5 — MT2b; MT1b pseudogene also identified on chromosome 5.

Mapping has demonstrated MT gene distribution across different chromosomes in the tomato and rice genomes. Evidence of MT gene clustering was found in cotton (three MT genes within a 10-kb fragment).

MT proteins

The wheat Ec Type 4 MT protein was purified from embryos as a zinc-binding protein and provided the first evidence that plants contained MTs as cysteine-based metal ligands in addition to PCs. Difficulties in identifying MTs in plants have led to the trend of describing the genes as “metallothionein-like” — but the review argues this is over-cautious given the high expression levels of many MT genes and their similarity to functional animal and fungal MTs.

Evidence for the occurrence of MT proteins in Arabidopsis was obtained by Murphy et al. (PDF p. 173, reference [50]): peptide fragments from MT1a, MT2a, MT2b, and MT3 were identified after purification of the proteins under anaerobic conditions using Zn affinity chromatography. Immunoblot analysis showed MT1 and MT2 proteins corresponded to RNA levels in tissue specificity and copper-treatment induction. MT instability in the presence of oxygen complicates plant MT purification.

Plant MT proteins expressed in microbial hosts (E. coli, MT-deficient yeast, Synechococcus) restore tolerance to copper and zinc, providing evidence that plant MTs can provide a biological metal-tolerance function in nonplant systems. The pea Type 1 PsMTa bound Cu, Cd, and Zn with highest affinity for Cu. A recombinant Fucus MT fusion protein showed greater affinity for Cu than Cd, with a pH for Cd dissociation approximately 2 pH units higher than that of a recombinant human MT.

MT gene expression

Plant MT genes are expressed at very high levels. SAGE profiling in rice (two-week-old seedlings) found transcripts from four MT genes comprised almost 3% of all transcripts; a single Type 2 MT gene contributed 1.26%, and transcripts of two Type 3 MT genes accounted for an additional 1.25% of the mRNAs. ESTs for MT genes are among the most prevalent in randomly sequenced cDNA libraries; a Type 2 MT gene accounted for 0.4% of tomato ESTs and 0.5% of maize ESTs were from a Type 1 MT gene.

Expression patterns by type and tissue:

• Type 4 MTs — restricted to developing seeds; promoter contains ABA-response elements. Proposed to provide a mechanism for storing zinc required during germination (Kawashima et al., PDF p. 174, reference [34]).
• Type 1 MTs — generally higher in roots than shoots; tend to be highly expressed in trichomes (where toxic metals such as Cd accumulate). RNA expression in leaves localised primarily to trichomes (in situ hybridisation in Arabidopsis and Vicia faba) and to phloem.
• Type 2 MTs — reverse pattern: generally higher in shoots than roots. GUS reporter studies in Arabidopsis showed MT1a and MT2a promoters drive expression preferentially in trichomes under some conditions.
• Type 3 MTs — highly expressed in fleshy fruits as they ripen (banana, apple, kiwi); also at high levels in leaves of plants that do not produce fleshy fruits (Arabidopsis).

Senescence and copper homeostasis. Type 1 MT RNA levels increase dramatically in senescing leaves (first reported in Brassica napus, confirmed in Arabidopsis and rice). At least two other genes specifically involved in copper homeostasis — the metal chaperone AtCCH and the copper transporter AtRAN1 — are co-expressed in senescing leaves. The proposed model: MTs chelate copper released from metalloproteins catabolised in senescing leaves; without MTs or another ligand, free copper would precipitate a cascade of oxidative damage. AtRAN1 (Menkes copper transporter homologue) may participate in copper efflux from senescing leaves to sinks such as developing seeds, possibly requiring AtCCH as a partner copper chaperone protein.

Environmental and stress regulation. Environmental modulation of MT expression shows little consistent cross-species pattern. Copper induces a Type 1 MT gene in Arabidopsis, rice, wheat, and tobacco, and MT genes in Fucus and Posidonia oceanica. Type 1 MTs are also induced by aluminium, cadmium, nutrient deprivation, and heat shock in wheat and rice, suggesting MTs may be expressed as part of a general stress response. There is currently no information about the mechanisms regulating MT transcription in plant vegetative tissues — in contrast to the detailed knowledge of metal-regulated MT expression in yeast (Ace1) and mammals (MTF1).

MT function in plants

The function of MTs in plants is still largely unresolved at the time of the review. In animals, MTs protect against cadmium toxicity, but this function in plants is provided by PCs. Evidence supports MT involvement in copper tolerance and homeostasis in plants: some plant MTs are functional copper-binding proteins; some MT genes are induced by copper; MT expression in senescing leaves coordinates with copper-homeostasis genes; Type 2 MT expression correlates with copper tolerance in Arabidopsis ecotypes; a Type 2 MT gene is elevated in a copper-sensitive mutant that accumulates copper; copper-tolerant Silene vulgaris populations have higher RNA expression and gene copy number of a Type 2 MT gene; PCs do not provide copper tolerance in Arabidopsis (implying another mechanism, likely MTs).

The most direct approach to defining MT function — identifying plants with defined MT-null genotypes — is technically difficult because MT genes present very small targets for T-DNA insertional mutagenesis (less than 1 kb), and multiple-MT-knockouts may be required to observe a phenotype. The function of mammalian MTs remains “somewhat of an enigma” despite extensive characterisation; the same can be expected for plants until null mutants are characterised.

Methods (brief)

This is a narrative literature review with no PRISMA flow, no formal inclusion/exclusion criteria, no risk-of-bias assessment, and no primary measurements. The review was completed shortly after the publication of the Arabidopsis genome sequence (2000), which the authors note “has allowed the identification of the entire suite of MT genes in a higher plant.” The review draws on ~85 cited references (per the Literature Cited section beginning at p. 178). Funding: Australian Research Council (C. Cobbett) and USDA-NRI (P. Goldsbrough). Acknowledgements name Metha Meetam and Woei-Jiun Guo for helpful comments. The review was downloaded by the University of Connecticut on 04/25/13 (per page-edge timestamps), suggesting the access copy is the canonical 2002 publication.

Tables:

• Table 1 (p. 162) — Mutants affected in phytochelatin biosynthesis and function (PC biosynthesis mutants in S. pombe and Arabidopsis; PC function mutants in S. pombe and C. glabrata).
• Table 2 (p. 164) — PC synthase enzymes predicted from DNA sequences across plants (Arabidopsis, Brassica juncea, wheat, rice) and others (S. pombe, C. elegans, D. discoideum), with N + conserved domain + variable domain amino acid counts, predicted MW in kD, cysteine arrangement in the variable domain, and GenBank accession numbers.

Figures:

• Figure 1 (p. 163) — Genes and functions contributing to Cd detoxification in plants and fungi, as a composite from Cd-sensitive mutant studies. Spans cytoplasm (Glu + Cys → γGluCys → GSH → PC via GCS/GS/PCS encoded by CAD2/RML1, CAD1; Cd²⁺ activation; LMW PC-Cd complex formation) and vacuole (hmt1-mediated transport to vacuole; HMW PC-CdS complex formation; sulfide pathway via ade6/7/8, ade2, hem2, hmt2).
• Figure 2 (p. 171) — Alignment of plant MT amino acid sequences across the four types, with conserved cysteines marked by stars. Sequences shown for Arabidopsis (At), rice (Os), pea (Ps), alfalfa (Ms), Brassica oleracea (Bo), petunia (Ph), Silene vulgaris (Sv), banana (Ma), kiwifruit (Ad), cotton (Gh), Picea glauca (Pg), maize (Zm), and wheat (Ta).

Implications

• Certification: The review contributes no occurrence data and no exposure data; it does not move any HMTc threshold-setting work. Its value for HMTc is indirect — it is biological background for whether plant-side PC and MT biochemistry is a credible Cd/Cu/Zn-reduction lever for crops supplying HMTc-certifying categories (rice, leafy vegetables, cereals, root vegetables). The review’s own framing emphasises that the PC pathway’s role in detoxification of metals other than Cd is limited and that MT function in plants is still unresolved — caveats that should be preserved when framing plant-side engineering as a potential remediation pathway.
• App: No routing to ingredient or product pages. The review measures nothing in any food matrix; food-relevant plant species (rice, wheat, maize, tomato, pea, Brassica, banana, apple, kiwi) appear as model or experimental host organisms in PC/MT biology studies, not as sampled food commodities with reported metal concentrations.
• Courses: Useful as a single-source orientation to the 2002 state of the PC and MT plant biology literature, the canonical PC biosynthesis pathway (Figure 1), the PCS catalytic mechanism, the plant MT four-type classification (Figure 2), and the cad1/CAD1 allelic basis for genetic evidence of PC function in plants. Should not be cited as the authority for any specific quantitative crop-level metal-reduction claim; trace claims to the cited primary studies. The 2002 vintage means that more recent reviews (Seregin & Kozhevnikova 2023, Marques 2025) are preferred for current PCS-paralogue inventories and engineering-outcome syntheses; Cobbett & Goldsbrough 2002 remains valuable for the foundational pathway and classification framework.
• Microbiome: Marginally relevant. The review notes PCS gene homologues in plants, fungi (S. pombe; SpPCS1 is the sole PCS-expressing fungus catalogued in Table 2), nematodes (C. elegans, C. briggsae, B. malayi), and slime moulds (D. discoideum); the absence of PCS in S. cerevisiae, Drosophila, mouse, and human is part of the phylogenetic puzzle of PC distribution. C. glabrata appears in the review for the hem2 mutant / sulfide-pathway / Cd-tolerance work (PDF p. 168) rather than as a PCS-expressing organism. WikiBiome federation is unlikely to draw on this source directly, but the cross-kingdom PCS distribution discussion is the foundational treatment of the question.

Limitations

• This is a narrative review with no declared inclusion or exclusion criteria, no systematic search protocol, no PRISMA flow, no language restrictions reported, and no risk-of-bias assessment. The cited literature skews toward Arabidopsis, S. pombe, S. cerevisiae, wheat, and other well-characterised model systems; the broader hyperaccumulator literature and the tropical/subtropical crop literature are underrepresented.
• The review is dated 2002. The PCS-gene catalogue (Table 2) is small compared to the contemporary 26+ species PCS-orthologue catalogue in Seregin & Kozhevnikova 2023; the ABCC1/2/3 vacuolar transporter system in plants had not yet been identified at the time of writing (HMT1-like plant function was acknowledged as not identified in the Arabidopsis genome). Subsequent decades have substantially expanded the engineering-outcomes literature for PCS overexpression and the As(III)-PC and Hg-PC complex characterisation. The wiki should prefer the more recent reviews for current state-of-evidence claims; Cobbett & Goldsbrough 2002 remains canonical for the foundational pathway and classification.
• The function of MTs in plants is explicitly framed as unresolved: “no MT-deficient mutants in Arabidopsis have been characterized” at the time of writing. The review’s discussion of MT function in copper tolerance is supported by indirect evidence (expression correlation, recombinant-protein binding studies, ecotype variation) rather than null-mutant phenotypes, and this caveat should be preserved.
• Metals other than Cd receive less attention. PC-As complex formation was being characterised at the time (Schmöger et al.), but the As(III)-versus-As(V) distinction is not consistently maintained in the review’s prose. Pb, Hg, Cu, Zn, and Ni discussion is brief, framed around the in vitro PCS activation pattern and the in vivo PC-deficient-mutant cross-metal sensitivity profile rather than detailed quantitative comparisons.
• No engagement with gut-microbiome, human-exposure, or epidemiological literature on PC or MT intake from plant-derived foods. The food-safety relevance of plant-side PC/MT engineering is implicit, not developed.
• Geographic and institutional scope of the cited primary literature skews toward North America, Europe, Australia, and Japan; the cited Chinese, South American, and African primary literature is sparse.

Wiki pages this source may touch

• cadmium
• arsenic
• arsenic-inorganic
• mercury
• mercury-total
• lead
• copper
• zinc
• nickel
• remediation-evidence

Verification notes

Existing-page check. DOI grep (10.1146/annurev.arplant.53.100301.135154), raw_handle grep (MFK_40-phytochelatins-and-metallothioneins-roles-in-he), and cite-key glob (cobbett2002-*, goldsbrough2002-*) over wiki/sources/ on 2026-06-08 returned no matches. Sibling PC/MT review pages seregin2023-phytochelatins-sulfur-metal-chelating (2023, IJMS), marques2025-phytochelatins-cadmium-mitigation (2025, IJMS), ruttkay-nedecky2013-metallothionein-oxidative-stress (2013), thirumoorthy2007-metallothionein-overview (2007), yang2024-metallothionein-comprehensive-review (2024), and grill1989-phytochelatins-heavy-metal-binding-peptides-plants (1989, the original Grill characterisation cited as reference [24] in this review) are conceptually adjacent but distinct papers with different cite-keys and DOIs. This is a NEW source page — no prior version to merge-enhance.

Evidence tier. B (secondary narrative review with no PRISMA and no primary measurements). The Annual Reviews series is a high-quality invited-review venue, and Cobbett (Melbourne) and Goldsbrough (Purdue) are both prominent researchers in the PC/MT field; however, the review is narrative rather than systematic, declares no inclusion/exclusion criteria, and contains no primary data. A-tier is reserved for primary peer-reviewed studies and authoritative agency monographs; this is neither.

Metals frontmatter. Cd is the primary subject across all sections. As is the secondary subject — discussed as a substantial PC-deficient-mutant sensitivity and as a substrate for in vivo PC complex formation (Schmöger et al.); the review treats arsenate and As(III) somewhat interchangeably but the substantive biological context is inorganic As broadly, so the frontmatter uses iAs. Hg is discussed in PCS activation and the cad1 mutant hypersensitivity context; the review uses Hg without consistently distinguishing methyl from inorganic, so the frontmatter uses tHg as the broader designation per CLAUDE.md Part 14. Pb is discussed as a PCS-activating metal and in the in vivo PC-Pb complex context (where the complex actually contains copper not Pb). Cu and Zn are discussed extensively in the MT section (copper tolerance, zinc binding in Type 4 MTs, copper-homeostasis genes in senescence); both fall outside the HMTc 10-analyte list but are in the wiki taxonomy. Ni receives the lightest coverage but is named in the in vivo PC-deficient-mutant cross-metal sensitivity list. Ag, Bi, Sn, and selenite are mentioned but are not in the wiki metal taxonomy. The metal slugs used (Cd, iAs, tHg, Pb, Cu, Zn, Ni) follow the CLAUDE.md Part 14 abbreviation vocabulary.

Ingredients, products, matrices, jurisdictions frontmatter. All empty. The review measures nothing in any food matrix. Food-relevant plant species (rice Oryza sativa, wheat Triticum aestivum, maize Zea mays, tomato Solanum lycopersicum, pea Pisum sativum, Brassica juncea, B. napus, B. oleracea, banana, apple, kiwi, cotton, sugarcane) appear as model or experimental host organisms in PC/MT biology studies, not as sampled food commodities with reported metal concentrations. Author affiliations are Australia (Cobbett, University of Melbourne) and USA (Goldsbrough, Purdue University); the review is conceptually international and no national regulatory or occurrence frame applies; jurisdictions: remains empty.

Sample size. Null. The review has no sampling frame; it cites ~85 references covering plant, fungal, animal, and bacterial PC/MT studies but reports no human/biological sample numbers of its own.

Brand firewall (Part 12). No commercial brand names appear in the source body for contamination values. No food, supplement, or personal-care brand is named. The only vendor-type mentions are scientific methods context (HPLC, ICP-AES, polymerase chain reaction, gel-filtration chromatography, T-DNA insertional mutagenesis, GUS reporter, SAGE protocol, Zn affinity chromatography) and standard plant-research organisms — none of which are commercial brands subject to the firewall. No firewall action required.

HMTc firewall (Part 2). The review contains no HMTc-threshold language, no claims about HMI certification levels, and no consumer-audience risk advisories. The Future Prospects section discusses “the potential for the use of plants for the detoxification or ‘phytoremediation’ of polluted environments is being increasingly examined” — biological-research framing, not a wiki-side synthesis or threshold proposal. Preserved in Implications without escalation. No firewall action required.

Date arithmetic. Annual Review of Plant Biology, vol. 53 (2002), pages 159–82. DOI 10.1146/annurev.arplant.53.100301.135154 resolves to the 2002 volume. The PDF page-edge metadata reads “5 Apr 2002 10:20” (LaTeX compilation date) and “Downloaded from www.annualreviews.org by University of Connecticut on 04/25/13” (access timestamp). All consistent with year: 2002 frontmatter.

Reviewer’s note on scope fit. This paper is in the “Black Market Peptide Metal Survey / heavy_metals_peptides” Manual Fetch Kimi folder alongside luo2024-peptides-heavy-metal-remediation, marques2025-phytochelatins-cadmium-mitigation, seregin2023-phytochelatins-sulfur-metal-chelating, grill1989-phytochelatins-heavy-metal-binding-peptides-plants, and others. Per the 2026-06-02 commit 3f47f95 — scope: mitigation/remediation is in-scope, not a skip, peptide-mediated mitigation/remediation papers are in scope as background for the mitigation-evidence chapter. Cobbett & Goldsbrough 2002 is the canonical foundational text of the PC/MT plant biology field — earlier than Seregin & Kozhevnikova 2023 and Marques 2025, broader than Grill 1989 (which characterised PCs originally), and the first comprehensive review to synthesise the PC and MT literatures together after the PCS gene cloning of 1999. The four (Grill 1989, Cobbett & Goldsbrough 2002, Seregin & Kozhevnikova 2023, Marques 2025) provide the chronological backbone for the PC/MT physiology section of a future mitigation chapter.

Slug-vocabulary note. [[mitigation/remediation-evidence]] is in the live wiki/mitigation/ directory; the taxonomy snapshot (docs/gpt-collaboration/taxonomy-snapshot.md, generated 2026-05-18) covers ingredients, products, metals, and regulations but not the mitigation/ subdirectory; this is a snapshot-coverage gap, not a missing-slug defect.

As speciation note. The review uses “Cd and arsenate” and “PC-As complexes” interchangeably; arsenate (As(V)) is named explicitly in the Rubia tinctorum induction list (Maitani et al.); As(III)-PC complex formation is mentioned (Schmöger et al.) but not consistently distinguished from As(V)-PC in the prose. The frontmatter uses iAs (inorganic arsenic) per CLAUDE.md Part 14, which captures both arsenite and arsenate. The MeHg/tHg distinction is not maintained in the review (Hg is used throughout without consistent speciation); the frontmatter uses tHg as the broader designation.

License note. Annual Reviews publishes under publisher copyright (the PDF page footer reads “Copyright © 2002 by Annual Reviews. All rights reserved”). The frontmatter license: "publisher-copyright" reflects this; the access_url points to the publisher DOI rather than an open-access repository.

Frontmatter near_duplicates: [] note. Sibling PC/MT reviews are conceptually adjacent but cover distinct ground (different DOIs, different paper bodies, different conclusions). They are not near-duplicates in the schema sense (which is reserved for the same paper published in multiple venues or the same study reported in multiple outputs). The near_duplicates field stays empty.

Audit subagent (2026-06-08) verdict: PROMOTE → two ⚠️ concerns applied. Five checks returned ✅ on slug vocabulary, ✅ on brand firewall, ✅ on HMTc firewall, ⚠️ on numerical fidelity (one omitted SAGE figure), and ⚠️ on speciation/methods (one C. glabrata PCS overreach). Both concerns independently verified against the PDF and applied:

• Finding 1 (SAGE Type 3 MT 1.25% omission): verified PDF p. 174 reports “Transcripts of two Type 3 MT genes accounted for an additional 1.25% of the mRNAs” — added to the MT Gene Expression section for completeness.
• Finding 2 (C. glabrata not a PCS-expressing organism): verified PDF Table 2 (p. 164) lists only SpPCS1 as a fungal PCS; PDF p. 168 names C. glabrata only for the hem2/sulfide-pathway/Cd-tolerance work. Implications/Microbiome paragraph corrected to specify S. pombe as the catalogued fungal PCS-expressing organism, with C. glabrata called out as a hem2/sulfide-context organism rather than a PCS-bearing fungus.

2 findings, 2 applied, 0 rejected. Audit subagent ID afe9fdc3247ec4cda.

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.