Quantitative mechanism of heavy-metal-ion inhibition of protein folding: thermodynamics, structural determinants, reversibility, and in vivo relevance of multidentate Cd$^{2+}$/Hg$^{2+}$/Pb$^{2+}$ complexes with non-native polypeptides

Environmental heavy metal ions Cd$^{2+}$, Hg$^{2+}$ and Pb$^{2+}$ inhibit the refolding of chemically denatured proteins at nanomolar concentrations, apparently by forming high-affinity multidentate (multi-ligand) complexes with the accessible functional groups of non-native polypeptide chains, and they likewise defeat ATP-dependent chaperone-assisted refolding and disaggregation (Sharma, Goloubinoff and Christen, 2008). This inhibition has only been characterized by apparent half-inhibitory concentrations ($IC_{50}$) measured at a single, higher substrate concentration, so the underlying binding thermodynamics, stoichiometries and structural rules remain unknown. The problem is to establish the quantitative mechanism of this folding inhibition: direct dissociation constants and stoichiometries (including non-thiol ligands for cadmium and mercury, since even a cysteine-free enzyme is inhibited by these ions), the sequence and structural determinants of susceptibility across proteins, the reversibility and chelator-mediated extractability of the complexes, and whether the mechanism operates in living cells at physiologically relevant exposures (notably for Hg$^{2+}$ and Pb$^{2+}$, where in vivo folding effects are untested). How metal-arrested folding could link to protein-misfolding disease endpoints is a downstream open question that motivates the program.

Background

Proteins reach their working three-dimensional structures through folding, and while folding they transiently expose backbone and side-chain groups (thiols of cysteine, imidazoles of histidine, carboxylates) that are buried or geometrically constrained in the native state. Molecular chaperones — ATP-driven machines such as the bacterial Hsp70 system (DnaK with its co-chaperones DnaJ and GrpE) — assist the folding of nascent and stress-denatured proteins and can solubilize and refold aggregates. 'Proteostasis' (protein homeostasis) denotes the cell-wide balance of folding, disaggregation and degradation that maintains a functional proteome.

Classical heavy-metal toxicology attributed the toxicity of cadmium, mercury and lead mainly to attacks on specific native proteins: binding of free thiols or displacement of essential metal cofactors, for example in zinc-finger domains. Sharma, Goloubinoff and Christen (BBRC 2008) demonstrated an additional, more general mode: Cd$^{2+}$, Hg$^{2+}$ and Pb$^{2+}$ potently inhibit the spontaneous refolding of chemically denatured model enzymes (firefly luciferase, lactate dehydrogenase, malate dehydrogenase, glucose-6-phosphate dehydrogenase) with $IC_{50}$ values of roughly 40–520 nM — the concentration halving recovered enzymatic activity after refolding. Inhibition was immediate, far stronger than toward the same proteins in the native state (a >100-fold shift in dose response for Cd$^{2+}$ and Pb$^{2+}$), and extended to chaperone-assisted refolding and chaperone-mediated disaggregation of heat-aggregated substrate. Because these apparent $IC_{50}$ values sit two to three orders of magnitude below the dissociation constants of monodentate Cd$^{2+}$/Pb$^{2+}$ complexes with single thiol, imidazole or carboxyl groups, the authors inferred formation of multidentate, chelate-like complexes — one ion coordinated by several neighboring groups of the same flexible chain — that lock the unfolded polypeptide and block folding. Three observations anchor that inference and its puzzles: even a cysteine-free enzyme (glucose-6-phosphate dehydrogenase) is inhibited by Cd$^{2+}$ and Hg$^{2+}$, whereas Pb$^{2+}$ shows no inhibition of that substrate up to 600 nM (Tamás et al. 2014), so non-thiol ligands participate for cadmium and mercury; the chaperone-side defect traces to the zinc-finger of DnaJ (substitution by CbpA, which lacks the zinc-binding domain, restores refolding); and neither reduced glutathione nor the chelator EDTA rescues protein that has misfolded in the presence of Cd$^{2+}$, while the ATP-dependent Hsp70 chaperone system significantly attenuates the folding-inhibitory effect of the metal ions (Tamás et al. 2014, restating the 2008 work) and the metal-bound chains form aggregates with amyloid-like dye responses.

Crucially, the quantitative footing is incomplete: $IC_{50}$ values had to be measured at a fixed substrate concentration of 350 nM (above some of the values themselves), so they are apparent numbers that underestimate the true potency and yield neither dissociation constants nor stoichiometries. Follow-up work has built around, not filled, this gap. For the metalloid arsenic(III) — chemically adjacent but thiol-directed — the corresponding questions were answered: As(III) binds unfolded reduced proteins at defined stoichiometries (one arsenic per two or three thiols), and synthetic multivalent arsenicals achieve ~50 nM affinity for unfolded, cysteine-rich substrates even against millimolar glutathione (Ramadan et al. 2009; Sapra et al. 2015). In vivo, arsenite and cadmium were shown to misfold and aggregate nascent proteins in yeast and Escherichia coli (Jacobson et al. 2012, 2017; Cornejo et al. 2023), cadmium impairs folding in the endoplasmic reticulum and triggers the unfolded protein response — the ER stress signaling pathway activated when folding demand exceeds capacity (Le et al. 2016) — and genome-wide screens have mapped cellular determinants of arsenite aggregation (Andersson et al. 2021). A 2023 authoritative review (Wysocki et al.) records that the in vivo effects of Hg$^{2+}$ and Pb$^{2+}$ on protein folding 'is not known'; the anticipation it states that susceptibility depends on the target's amino acid sequence, thiol arrangement and folding pathway is formulated for the metalloid arsenite, and no corresponding residue-level susceptibility rule exists for the heavy metals. Binding studies on individual native or intrinsically disordered targets (Hg$^{2+}$–apolipoprotein E, Ag$^{+}$–amyloid-β) find only micromolar affinities, far weaker than the folding-inhibition potency, leaving the high-affinity complex on non-native chains uncharacterized to date.

Problem Statement

For the heavy metal ions Cd$^{2+}$, Hg$^{2+}$ and Pb$^{2+}$, whose inhibition of the refolding of non-native proteins was discovered by Sharma, Goloubinoff and Christen (2008): establish the quantitative mechanism of folding inhibition, comprising (i) direct measurement of the thermodynamics and stoichiometry of the multidentate metal–polypeptide complexes that arrest refolding — true dissociation constants and metal-to-chain stoichiometries on defined non-native substrates (the published $IC_{50}$ values, the metal concentrations halving recovered activity, are apparent, measured at a single 350 nM substrate concentration and underestimating potency), including identification of the coordinating ligands, which for cadmium and mercury must include non-thiol groups since even a cysteine-free enzyme is inhibited by these ions; (ii) identification of the sequence and structural determinants — that is, all of: ligand identity (cysteine and histidine as the most important potential ligands, plus non-thiol groups for cadmium and mercury), ligand arrangement and spacing, side-chain accessibility, and backbone flexibility/motility — that permit high-affinity multidentate complex formation on non-native chains, and thereby explain why some non-native proteins are more susceptible than others; (iii) determination of the kinetics and reversibility of complex formation, and whether cellular or therapeutic chelating agents (glutathione, EDTA-like chelators, metallothionein) can extract metal from already-formed complexes and restore folding competence — motivated by the observation that neither reduced glutathione nor EDTA rescues protein that has misfolded in the presence of Cd$^{2+}$, whereas the ATP-dependent Hsp70 chaperone system (DnaK/DnaJ/GrpE) significantly attenuates the folding-inhibitory effect of the metal ions; whether the complexes behave as effectively irreversible on the experimental timescale, and for which metals, is itself open; and (iv) demonstration (or exclusion) of the mechanism's operation in vivo at physiologically relevant exposures, that is, at metal concentrations, speciation and durations that cells or organisms may actually encounter, with free-ion availability under cellular metal buffering rather than total metal as the relevant measure, including nascent chains and co-translational folding and the modulating role of cellular metal buffering — established for cadmium and arsenite in yeast and E. coli but explicitly unknown for Hg$^{2+}$ and Pb$^{2+}$. Establishing the quantitative mechanism requires all four parts (i)–(iv) together; a result on any single part is a partial resolution only. At minimum, parts (i) and (ii) must cover all three metal ions and the model substrates named in Background (firefly luciferase, lactate dehydrogenase, malate dehydrogenase, glucose-6-phosphate dehydrogenase); part (iii) must test both a cellular thiol and an EDTA-like non-protein chelator; and part (iv) must demonstrate or exclude the mechanism in at least one cell or organism system, including for Hg$^{2+}$ and Pb$^{2+}$. How metal-arrested protein folding links to the pathologies of protein-misfolding (neurodegenerative and age-related) disorders associated with chronic metal exposure remains a downstream open question and motivation of this program (Tamás et al. 2014), rather than a required component of the mechanistic account.

References

1. Sharma SK, Goloubinoff P, Christen P (2008) Heavy metal ions are potent inhibitors of protein folding. Biochemical and Biophysical Research Communications 372(2):341–345. DOI 10.1016/j.bbrc.2008.05.052. https://doi.org/10.1016/j.bbrc.2008.05.052
2. Sharma SK, Goloubinoff P, Christen P (2011) Non-native Proteins as Newly-Identified Targets of Heavy Metals and Metalloids. In: Cellular Effects of Heavy Metals (Bánfalvi G, ed.), Springer, pp. 263–274. DOI 10.1007/978-94-007-0428-2_12. https://doi.org/10.1007/978-94-007-0428-2_12
3. Tamás MJ, Sharma SK, Ibstedt S, Jacobson T, Christen P (2014) Heavy metals and metalloids as a cause for protein misfolding and aggregation. Biomolecules 4(1):252–267. DOI 10.3390/biom4010252. https://doi.org/10.3390/biom4010252
4. Ramadan D, Rancy PC, Nagarkar RP, Schneider JP, Thorpe C (2009) Arsenic(III) species inhibit oxidative protein folding in vitro. Biochemistry 48(2):424–432. DOI 10.1021/bi801988x. https://doi.org/10.1021/bi801988x
5. Sapra A, Ramadan D, Thorpe C (2015) Multivalency in the inhibition of oxidative protein folding by arsenic(III) species. Biochemistry 54(2):612–621. DOI 10.1021/bi501360e. https://doi.org/10.1021/bi501360e
6. Jacobson T, Navarrete C, Sharma SK, Sideri TC, Ibstedt S, Priya S, Grant CM, Christen P, Goloubinoff P, Tamás MJ (2012) Arsenite interferes with protein folding and triggers formation of protein aggregates in yeast. Journal of Cell Science 125(Pt 21):5073–5083. DOI 10.1242/jcs.107029. https://doi.org/10.1242/jcs.107029
7. Jacobson T, Priya S, Sharma SK, Andersson S, Jakobsson S, Tanghe R, Ashouri A, Rauch S, Goloubinoff P, Christen P, Tamás MJ (2017) Cadmium causes misfolding and aggregation of cytosolic proteins in yeast. Molecular and Cellular Biology 37(17):e00490-16. DOI 10.1128/MCB.00490-16. https://doi.org/10.1128/MCB.00490-16
8. Tamás MJ, Fauvet B, Christen P, Goloubinoff P (2018) Misfolding and aggregation of nascent proteins: a novel mode of toxic cadmium action in vivo. Current Genetics 64(1):177–181. DOI 10.1007/s00294-017-0748-x. https://doi.org/10.1007/s00294-017-0748-x
9. Le QG, Ishiwata-Kimata Y, Kohno K, Kimata Y (2016) Cadmium impairs protein folding in the endoplasmic reticulum and induces the unfolded protein response. FEMS Yeast Research 16(5):fow049. DOI 10.1093/femsyr/fow049. https://doi.org/10.1093/femsyr/fow049
10. Cornejo FA, Muñoz-Villagrán C, Luraschi RA, Sandoval-Díaz MP, Cancino CA, Pugin B, Morales EH, Piotrowski JS, Sandoval JM, Vásquez CC, Arenas FA (2023) Soft-metal(loid)s induce protein aggregation in Escherichia coli. Frontiers in Microbiology 14:1281058. DOI 10.3389/fmicb.2023.1281058. https://doi.org/10.3389/fmicb.2023.1281058
11. Andersson S, Romero A, Rodrigues JI, Hua S, Hao X, Jacobson T, Karl V, Becker N, Ashouri A, Rauch S, Nyström T, Liu B, Tamás MJ (2021) Genome-wide imaging screen uncovers molecular determinants of arsenite-induced protein aggregation and toxicity. Journal of Cell Science 134(11):jcs258338. DOI 10.1242/jcs.258338. https://doi.org/10.1242/jcs.258338
12. Wysocki R, Rodrigues JI, Litwin I, Tamás MJ (2023) Mechanisms of genotoxicity and proteotoxicity induced by the metalloids arsenic and antimony. Cellular and Molecular Life Sciences 80(11):342. DOI 10.1007/s00018-023-04992-5. https://doi.org/10.1007/s00018-023-04992-5
13. Berntsson E, Sardis M, Noormägi A, Jarvet J, Roos PM, Tõugu V, Gräslund A, Palumaa P, Wärmländer SKTS (2022) Mercury ion binding to apolipoprotein E variants ApoE2, ApoE3, and ApoE4: similar binding affinities but different structure induction effects. ACS Omega 7(33):28924–28931. DOI 10.1021/acsomega.2c02254. https://doi.org/10.1021/acsomega.2c02254
14. Lakela AL, Berntsson E, Vosough F, Jarvet J, Paul S, Barth A, Gräslund A, Roos PM, Wärmländer SKTS (2025) Molecular interactions, structural effects, and binding affinities between silver ions (Ag+) and amyloid beta (Aβ) peptides. Scientific Reports 15:5439. DOI 10.1038/s41598-024-59826-6. https://doi.org/10.1038/s41598-024-59826-6
15. Lorentzon E, Horvath I, Kumar R, Rodrigues JI, Tamás MJ, Wittung-Stafshede P (2021) Effects of the toxic metals arsenite and cadmium on α-synuclein aggregation in vitro and in cells. International Journal of Molecular Sciences 22(21):11455. DOI 10.3390/ijms222111455. https://doi.org/10.3390/ijms222111455
16. Alazoumi KKM, Sharma P, Islam A, Farooqi H (2024) Mitigation of the deleterious effect of heavy metals on the conformational stability of ubiquitin through osmoprotectants. Cell Biochemistry and Biophysics 82(1):193–202. DOI 10.1007/s12013-023-01188-3. https://doi.org/10.1007/s12013-023-01188-3
17. Blanc FEC, Hologne M, Girod M, Demontrond M, Chermette H, Walker O (2026) Mechanism of Ag+-induced folding of a bacterial peptide from replica-exchange molecular simulations. Journal of Physical Chemistry Letters 17(8):2455–2463. DOI 10.1021/acs.jpclett.6c00044. https://doi.org/10.1021/acs.jpclett.6c00044
18. Devi S, Chaturvedi M, Fatima S, Priya S (2022) Environmental factors modulating protein conformations and their role in protein aggregation diseases. Toxicology 465:153049. DOI 10.1016/j.tox.2021.153049. https://doi.org/10.1016/j.tox.2021.153049
19. Ghosh P, Murumalla L, Das S, Alavala S, Challa S (2026) Heavy metal toxicity as a driver of endoplasmic reticulum stress and dysfunction. The FEBS Journal 293(15):4415–4453. DOI 10.1111/febs.70527. https://doi.org/10.1111/febs.70527