Metal Chelation Therapy and Parkinson's Disease: A Critical Review on the Thermodynamics of Complex Formation between Relevant Metal Ions and Promising or Established Drugs.

The present review reports a list of approximately 800 compounds which have been used, tested or proposed for Parkinson's disease (PD) therapy in the year range 2014-2019 (April): name(s), chemical structure and references are given. Among these compounds, approximately 250 have possible or established metal-chelating properties towards Cu(II), Cu(I), Fe(III), Fe(II), Mn(II), and Zn(II), which are considered to be involved in metal dyshomeostasis during PD. Speciation information regarding the complexes formed by these ions and the 250 compounds has been collected or, if not experimentally available, has been estimated from similar molecules. Stoichiometries and stability constants of the complexes have been reported; values of the cologarithm of the concentration of free metal ion at equilibrium (pM), and of the dissociation constant Kd (both computed at pH = 7.4 and at total metal and ligand concentrations of 10-6 and 10-5 mol/L, respectively), charge and stoichiometry of the most abundant metal-ligand complexes existing at physiological conditions, have been obtained. A rigorous definition of the reported amounts is given, the possible usefulness of this data is described, and the need to characterize the metal-ligand speciation of PD drugs is underlined.

1. Introduction

Parkinson’s disease (PD) is a common neurodegenerative disorder (ND) [1]. It is characterized by neuronal cell loss in the substantia nigra (SN), which leads to a progressive central nervous system dysfunction. Symptoms include motor abnormalities like tremors, movement and balance issues, and non-motor problems like difficulty in swallowing and speaking, depression, cognitive impairment, and dementia. Although PD by itself is not a fatal disease, people may die from causes related to it.

Age is the most relevant risk factor: approximately 2% of people over the age of 60 years, and 3% of those at age over 80 years, suffer from PD [2]. Due to the average population aging, the occurrence of PD and of other NDs like Alzheimer’s disease and Amyotrophic Lateral Sclerosis is continuously increasing. The number of persons suffering from PD is expected to reach a value of approximately 9 million in 2030 [3] and up to more than 17 million in 2040 [4]. NDs are considered the leading source of disability around the world, and the fastest growing of these disorders is PD [5]. Although non-infectious, PD exhibits many of the characteristics of a pandemic, and it is experiencing exponential growth worldwide [4,6,7]. Together with the average population age increasing, other factors will likely contribute to increase the incidence of PD over current forecasts. In particular, reducing smoking rates in some countries may lead to a higher incidence of PD, because many studies have found that the risk of this disease is decreased among smokers by approximately 40% [8]. Also, pollutants by-produced in industrialized countries may contribute to the rising rates of PD, because specific pesticides, solvents, and heavy metals have been linked to this disease [9]. Actually, countries that underwent the most rapid industrialization have seen the greatest increase in the rates of PD [5].

PD can be due to genetic factors, and it has been found that race/ethnicity can affect the incidence of PD in the order Hispanics > non-Hispanic Whites > Asians > Blacks [10]. However, genetics appears to justify only a small amount (approximately 5–10%) of all PD cases: it is therefore possible to suggest an important role of external factors, like behavioural and environmental [2,9,11].

Existing drugs for PD provide only the relief of some symptoms, and there are no disease-modifying therapies demonstrated to slow or to stop the ongoing neurodegenerative process. In the search for such therapies, however, the primary issue lies in the multifactorial nature of PD. The main neuropathological hallmark of PD is proteopathy, as the formation and deposition of protein aggregates is generally observed in PD brains. The most typical deposits, called Lewy bodies, are due to α-synuclein (α-syn), an abundant neurological protein with yet unclear physiological functions. Other features characterizing PD brains are mitochondrial dysfunction, oxidative stress, defects in energy metabolism, aberrant axonal transport, and metal ion dyshomeostasis [12]. All the pathways operating in PD appear to be strictly inter-related, so that both the study of the molecular causes of the disease, and the search for an efficient therapy, cannot be limited to a specific pathway: a multi-targeted approach is suggested [1,12]. Much scientific effort in recent years is devoted to the comprehension of each pathological mechanism operating in PD, with the aim to understand and rationalize the biochemical processes occurring during the pathology. This review focuses on metal dyshomeostasis and on the therapy that is aimed to target this pathological process, i.e., metal chelation therapy.

2. Parkinson’s Disease and Metal Ions

Almost one hundred years ago, Lhermitte et al. [13] discovered that the brains of people who died from Parkinsonism, a form of dementia with similar symptoms as those of PD [14], contained a significantly larger amount of iron (Fe) than the corresponding brains of controls. Since that work, several studies have confirmed the abnormally high Fe content also in the brains of PD patients [15,16,17,18,19]. Conversely, other studies could not detect an overload of Fe in PD brains [20,21]. Recently, several studies have attempted to determine Fe brain levels in living patients with PD ([22,23] and references therein). Most but not all studies indicated larger Fe levels in the SN of patients with PD compared to control subjects, whereas no Fe excess was observed in many other brain parts, thus suggesting that a Fe dyshomeostasis occurs in PD brain, especially in the SN [23,24]. Bush et al. found that the reported Fe accumulation is contributed to by a disturbance in Fe export. This was explained by a significant reduction of the specific activity (but not of the levels) of ceruloplasmin in the SN of PD patients [25]. Ceruloplasmin is a multicopper ferroxidase protein facilitating cellular Fe export [26]. The same authors [27] observed a decrease in the PD SN of the soluble levels of another protein, tau, which can lower neuronal Fe levels by promoting the presentation of the amyloid protein precursor to the neuronal surface, where it favours the efflux of Fe [28]. It has also been suggested that PD-induced Fe accumulation is due to a dark pigment contained in SN, neuromelanin, that is able to bind Fe and that may act as a protection against Fe by binding and storing its excessive labile content [29].

A number of studies have also shown alterations in the copper (Cu) concentrations in the brain of post-mortem PD patients compared to non-PD controls [11,30], suggesting that metal dyshomeostasis in PD brains also regards Cu [31]. However, while Fe appeared to be systematically overloaded, Cu was significantly reduced in the degenerating regions of PD brains [30,32,33,34]. Other metal ions have been monitored in PD brains, but less definite results were reported. Conflicting data were reported for zinc (Zn) [35,36] and Genoud et al. recently evidenced no differences in Zn levels between experimental groups [30]. Parkinsonism is reported to rapidly develop in patients subjected to the exposure of high levels of manganese (Mn) [11,37], and a role in Parkinsonism onset has also been suggested for Cu [38]. However, no changes have been detected in the Mn level of post-mortem PD brains with respect to non-PD samples [30], and, as seen, even reduced levels were detected for Cu. Other elements which have been occasionally linked with PD have been aluminium (Al), arsenic (As), bismuth (Bi), cadmium (Cd), mercury (Hg), lead (Pb), thallium (Tl), and titanium (Ti) [2,11]. Bjørklund et al. [11] reviewed the works in which the exposure to metal ions was shown to correlate with the onset of PD and/or of Parkinsonism.

These results have prompted researchers to clarify the role of each metal ion in PD. In literature, the most studied metal ions have been Fe, Cu, Mn and Zn, whereas papers regarding other elements were much fewer. The huge work performed on this matter has been reviewed in detail [11,31,32,39,40,41], and it regarded the molecular mechanisms and biological aspects of these elements in the brains of controls and of PD patients.

Essential metal ions like Cu, Fe, Mn, and Zn are known to be involved in a large number of biochemical processes in the human brain [41,42], where they exert a structural (e.g., stabilizing configurations of macromolecules) or a functional role (e.g., being the active site of metalloenzymes). Both Cu and Fe can exist in vivo under two oxidation states, Cu(II) and Cu(I), Fe(III) and Fe(II), to allow biological systems activating and using O2 for energy purposes. Reactions activating O2, if not tightly regulated, can cause oxidative stress, so that healthy biological systems contain suitable antioxidants and very little exchangeable Cu and Fe ions. This metal ion fraction is also called “labile” or “free” ion [43], and it is thought to be the main contributor of metal-induced oxidative stress [24,44,45]. Metal ions in the labile pool can be loosely bound to peptides, carboxylates and phosphates as compounds with low mass, while some might exist as hydrated free ions. In healthy mammalian cells, the labile Fe concentration is less than 1 µmol/L, and less than 5% of total Fe [46]. The labile Cu and Fe fractions exert their toxicity by generating reactive oxygen species (ROS) via the Fenton and the Haber–Weiss reactions, both related to the presence of the Fe(III)/Fe(II) or of the Cu(II)/Cu(I) redox couple. The Fenton reaction for Fe is:Fe(II) + H

In PD brains, a dysregulation occurs between the production of OH• and OOH• (and of other ROS) and their removal, thus resulting in cellular damage through the oxidation of lipids, proteins, and DNA. The levels of glutathione, one of the most important antioxidants in human brain, were reported to be significantly decreased in the SN of PD patients compared with those of healthy subjects [47]. A dyshomeostasis of Fe or Cu can therefore have a significant impact on ROS regulation. Also, high labile Mn levels have been reported to increase oxidative stress [42]. Labile Zn2+ is more abundant in healthy brain cells, as it is released by neural activity at many central excitatory synapses [48], but still this metal ion was related to oxidative stress [49].

Labile metal ion pools can also undergo a pathogenic relation with α-syn. Being an unfolded protein, α-syn can easily switch in a number of conformational states in response to changes in environmental conditions [50,51]. Temperature changes, presence of pro-oxidative conditions [52,53] and of several metal ions can promote the formation of dimers and other polymeric forms of this protein [54,55]. The misfolding of α-syn is thought to be the most important factor driving the formation of Lewy bodies in PD, and, in turn, toxic forms of aggregated α-syn are released from neurons, and then spread between cells in a prion-like manner [32]. It was shown that part of RNA structure posttranscriptionally regulates α-syn production in response to cellular Fe and redox events [56,57], so that the overexpression of α-syn promotes the neuronal accumulation of Fe. Fe can promote the aggregation of this protein, and post-translational modifications of α-syn have also been found to regulate Fe transport [58]. It was found that α-syn can inhibit the lysosome-mediated degradation of ferritin (a Fe storing protein), resulting in the intracellular build-up of ferritin and consequently of Fe [59]. Also, the direct interaction between metal ions and α-syn in neurons, with the formation of metal–protein complexes, can be of primary importance to justify the protein unfolding and eventually its aggregation. Furthermore, the complexes themselves may be cytotoxic, as, e.g., it was reported to be Cu2+/α-syn [60]. The properties of the complex formation between metal ions and α-syn have been reviewed by several authors (e.g., [12,32,60,61]). As regards the binding moiety, it is known that metal ions can bind to high-affinity N-terminal (containing residues 1–60) and to lower affinity C-terminal sites of α-syn (from 96 to 140 amino acid residues). Studies have been performed to evaluate the stability of the metal–α-syn interactions: information available in the literature, given as dissociation constants (Kd, see below) of the complexes formed between metal ions and α-syn, is resumed in Table 1.

Table 1

Dissociation constants (Kd) obtained at physiological pH for the complexes formed between metal ions and α-synuclein (values computed for the highest affinity binding site), and references (reviews).

Metal Ion	Kd (nmol/L)	References
Cu(II)	102	[12,61]
Cu(I)	104–103	[61]
Fe(III)	10−4	[12,60]
Fe(II)	106–5 × 104	[12,60]
Mn(II)	106	[60]
Zn(II)	>106	[61]

A recently discovered cell death pathway, called ferroptosis or Fe-dependent cell death, has provided further impetus to the “Fe hypothesis” of PD, Alzheimer’s disease, and Amyotrophic lateral sclerosis [39,62,63]. Fe still has an unclear role in ferroptosis, but it has been shown that Fe chelation is beneficial in preventing this cellular damage, which is also characterized by increased levels of lipid hydroperoxides and by a depletion of the important antioxidant glutathione. PD has been linked to ferroptosis because literature studies generally indicate that ferroptosis inhibitors may be effective in PD, too [39]. For example, chelation therapy can be beneficial in PD (see below), and prodrugs such as N-acetylcysteine, which enhances glutathione levels in brain, exert partial protection against PD neurodegeneration.

Many mechanisms have been considered and reviewed for the damage induced by Cu [44], Fe [49,64,65,66], Mn [67] and Zn [49] under PD conditions; in several cases, oxidative damage, metal dyshomeostasis and α-syn aggregation have been demonstrated to be strictly related to each other.

Despite the availability of many useful results, the molecular pathways describing the association between metals and PD onset are still ill defined. It remains controversial whether the dyshomeostasis of Cu, Fe, Mn, Zn, and possibly of other elements, is the primary cause or secondary consequence of PD as well as of other neurological diseases such as Alzheimer’s disease, multiple system atrophy, dementia with Lewy bodies, amyotrophic lateral sclerosis, Huntington’s disease, frontotemporal dementia, corticobasal degeneration, and progressive supranuclear palsy. The possibility that metal dyshomeostasis is just a secondary consequence of other independent molecular paths is supported by considering that the timing for PD onset is much slower (many years) than for Parkinsonism, which in turn is very rapidly induced if, e.g., Mn or other external toxins are administered to animal models. This suggests that endogenous neurotoxins, rather than exogenous ones, are responsible for the extremely slow neurodegeneration observed in PD. In their very recent review, Ndayisaba et al. tried to answer whether neurodegeneration is caused (or co-caused) by Fe dyshomeostasis, whether the latter contributes accelerating the pathological effects due to nerve cell death and to release of intracellular components, or whether neurodegeneration is simply not related to Fe accumulation [68]. The authors were not able to give a definite answer, but they observed that Fe dyshomeostasis occurs already at early PD onset, and that Fe should at least contribute to many aspects of neurodegeneration, in a way such that Fe might be proposed as a biomarker to detect for preclinical stages of NDs. Similarly, prudent conclusions have been drawn in another very recent review by Chen et al. [42], where the authors found that it was unclear whether Fe and Mn are primary or secondary causes of neurodegeneration, as they found that neurodegeneration cannot be reversed if metal overload is removed. The question about the primary or secondary role of Cu and Zn was raised by Barnham and Bush [41] in their review. They concluded that PD, Alzheimer’s disease, Huntington’s disease and amyotrophic lateral sclerosis are not caused by a simple overload of these metals, although the possibility that exposure may alter disease risk was not excluded. The authors found a number of possible molecular pathways induced by altered Cu and Zn levels in PD brains, which can at least contribute to the disease progression. The authors concluded that “the targets of metalloneurobiology are rich with pharmacological opportunities” [41].

3. Metal Chelation Therapy in Parkinson’s Disease

Metal chelation therapy (MCT) was proposed more than 50 years ago for the therapy of pathologies produced in the body by an overload of a metal. Metal chelation therapy involves the use of a chelating agent (CA), i.e., a molecule which forms stable coordination complexes with the target metal ion. Once administered to the patient, the CA acts as a scavenger removing the metal from its stores and favouring its decorporation from the body [69]. An efficient CA should be orally active and have a low cost, and both the ligand and the complexes formed in vivo should possess suitable hydrophilicity/hydrophobicity, and no redox activity [70]. In particular, the CA affinity towards the overloaded metal ion should be as high as possible. Last but not least, CAs and their metal complexes should display no toxicity and no or negligible side effects, but these properties are still only partially verified for established CAs. For example, the common Fe and Al chelator Desferrioxamine (also known as Desferal, DFO or Deferoxamine) is reported to cause a number of severe side effects [71], among which heart diseases [72] and retinopathy [73] appear to be the most important ones. Adverse effects were also reported for the other two established Fe chelators, Deferiprone and Deferasirox. According to Fisher et al. [74], side effects increased in patients treated with Deferiprone compared with Desferrioxamine. Kontoghiorghes et al. [71,75] reported a number of fatal renal, liver, and bone marrow failures due to Deferasirox. Another common CA used for Cu overload, D-Penicillamine, causes neuropsychiatric or hepatic complications in up to one-third of patients [76]. Toxic effects have also been observed with other CAs (2,3-dimercaptopropanol, meso-2,3-dimercaptosuccinic acid, 2,3-dimercaptopropane-1-sulfonic acid, EDTA calcium or sodium salts) used in the therapy for the overload of As, Au, Hg, Pb [77].

Despite these toxicity reports, the occurrence of a metal ion dyshomeostasis in PD has suggested to also employ MCT for the therapy of this disease and of other NDs such as corticobasal degeneration, the Westfal variant of Huntington’s disease, Alzheimer’s disease, Friedreich’s ataxia, pantothenate kinase-associated neurodegeneration, and other neuropathologies associated with brain metal overload [12,24,78,79]. In these cases, MCT was also referred to as “metal targeting”, “metal attenuating” or “metal protein attenuating” [41,80,81], in order to underline the differences occurring when MCT is employed in NDs instead of in metal overloads. Poujois et al. [82] have further improved this chelation strategy, and they called it “conservative chelation”. For the design of metal-based therapeutic strategies in NDs, the complete removal of metals from affected tissues is not the desired mechanism of action of these drugs. The terms “targeting”, “attenuating” and “conservative” evidence that the CA should remove only labile essential metal ions, which are considered to be not functionally required. Essential metal ions are aimed to be removed from the biological targets where they might be harmful, in particular, to avoid the α-syn complex formation and ROS-generating redox reactions, but they should also be allowed acting their normal physiological functions, as for example in metallo-enzymes, thus preventing severe side effects. Another important feature of a conservative CA is that the labile metal pool should be redeployed to cell acceptors or transport proteins (e.g., transferrin for Fe) [82]. This mode of action is expected to correct aberrant metal distribution, minimising systemic loss of chelated metal, thus avoiding the CA to cause metal-deficiency anemia and interfere with metal-dependent mechanisms essential for normal physiological functions. Conservative chelation, instead of a more aggressive metal removal, appeas to be particularly suitable in PD if Fe is the target, as patients suffering from this ND are mainly elderly people who are often on the border or already with Fe deficiency. Nevertheless, the unspecific removal of any essential trace metals may lead to harmful adverse effects to people of all ages, and metal deficiency can be regarding not only targeted but also untargeted metal ions. Excessive removal of Cu and Zn has been often reported for β-thalassemic patients undergoing MCT with Fe-chelators such as Deferiprone, and especially Desferrioxamine and Deferasirox [83,84]. Cu and Zn anemia, in turn, can lead to delay in growth and development, immunodeficiency, and abnormal hematopoiesis [84]. Metal redistribution, rather than metal removal, is therefore the goal in PD [24,85]. To allow a conservative chelation, a CA to be employed in PD should form moderate but not too strong complexes with the target metal ion. As an additional property, the drug should be able to pass the blood–brain barrier [78]. The ability of a CA to pass this barrier can be improved by the prochelator strategy, which for PD and also for other pathologies has been extensively reviewed by Oliveri and Vecchio [86].

In PD, MCT was proposed to target dysregulated essential metal ions [85], mainly the labile pool of Fe and Cu [24,78] but also of Zn [81] and, more rarely, of Mn [40], rather than for the decorporation of total toxic ions such as Al, Hg, Pb, etc. MCT has been tested in a number of translational studies on cell lines or on animal PD models, where Parkinsonism was induced by the administration of OHDA (6-hydroxydopamine) or MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine). The Fe chelator Desferrioxamine was reported to reduce iron-induced oxidative stress in SK-N-SH cell line and dopaminergic cells aggregation [87], and its intranasal administration significantly improved PD symptoms in MPTP-treated mice [88]. The other Fe chelator Deferiprone, differently than Desferrioxamine, is orally active and is better able to cross the blood–brain barrier [89]. Deferiprone demonstrated to be efficacious in MPTP and OHDA-induced animal models of PD [90]. Other CAs tested in cell lines or in animal models for the PD therapy have been Clioquinol [91], VK-28 [92,93], M30 [94], PBT2 [95], Q1, Q4 [96] and several other compounds, as reviewed recently by Singh et al. [89].

The first clinical evidence about the efficacy of a conservative Fe chelation regimen for human PD was given by Devos et al. [90], who orally administered Deferiprone to PD patients for 12 months. The Fe deposits in the SN were significantly reduced, and the Unified Parkinson’s Disease Rating Scale motor indicators of disease progression were significantly improved. However, when the treatment was suspended, Fe started to accumulate again, suggesting a reversal to the pathological state. Deferiprone, differently than other well-known Fe chelators such as Desferrioxamine and Desferal, has the important feature to rescue transfusional hemosiderosis in the hearts of β-thalassemia patients without inducing significant anemia, largely attributable to the redeployment of captured Fe to extracellular iron-free transferrin and subsequent distribution [97], thus allowing this CA to be employed for a conservative chelation strategy. Devos et al. [90] reported that none of the Deferiprone-treated PD patients developed new neurological signs, and no level changes (of Fe and of other transition metals) were detected in brain parts not involved in PD. The conservative chelation strategy used by the authors prevented side effects typically due to Fe-deficiency anemia in the brain such as the restless legs syndrome [98]. Other Fe chelators forming stronger complexes with the targeted metal ion could have caused such unwanted side effects as they would likely also remove the non-labile part of Fe or of other metal ions [90]. The authors concluded that Deferiprone can represent a paradigm for conservative Fe chelation. The encouraging results obtained for this CA prompted the development of other clinical trials with Deferiprone. A search in https://clinicaltrials.gov indicated four ongoing or finished tests of this molecule for the treatment of PD, as also recently reported by Nuñez and Chana-Cuevas [79].

To the best of our knowledge, no other CA is still being subjected to clinical trials. However, other than Deferiprone, there is a number of molecules which have been or are being considered good therapeutic candidates for PD therapy [24,79,85,89]. Nowadays, due to the multi-faced nature of PD, the proposed strategy for MCT requires the use of multifunctional molecules able not only to bind metal ions thus controlling metal dyshomeostasis, but also to counterbalance other toxic pathways in PD. Multifunctional molecules for PD have recently been reviewed by Savelieff et al. [12].

All previously cited reviews list the names and sometimes the chemical structures of CAs used as or tested for PD therapy. However, these reviews do not report which complexes can form between relevant metal ions and CAs (metal–ligand stoichiometries) and how stable they are (metal–ligand stability constants), i.e., they lack speciation information. Some lists of metal–ligand stoichiometries and of stability constants (given as logβ, see below) for promising PD drugs have been reported, for example by Gumienna-Kontecka et al. [24], Kasprzak et al. [99], and Prachayasittikul et al. [100]. The Gumienna-Kontecka logβ list was however limited to very few compounds and to the complexes formed with Fe(III), the Kasprzak and Prachayasittikul lists report only the complex formation of flavonoids and 8-hydroxyquinolines, respectively. Many of the reported lists also lack the ligand acidity constants, which (see below) are necessary to achieve a complete speciation picture.

The knowledge of metal–ligand speciation can allow modeling the CA activity in vitro or in vivo, i.e., to perform calculations (e.g., [101]) describing the distribution of the metal ion of interest at any conditions. Some examples of information which can be gathered from speciation calculations in the frame of MCT will be briefly given below. We think, and it has been stated (see, e.g., [101,102]), that knowledge about metal–ligand speciation should be easily available to researchers interested in the study of pathologies involving metal ions, such as PD. As regards CAs, the aforementioned knowledge should at first regard all essential metal ions (in all their possible oxidation states) undergoing dyshomeostasis in PD. The relevant ions are therefore Cu(II), Cu(I), Fe(III), Fe(II), Mn(II), and Zn(II) (Mn(III) should also be considered in principle, however we decided to exclude it because its content in the body is generally considered to be minimal [103]).

It is also necessary to note that there are several other established or promising PD drugs, not specifically designed for MCT, which can also act as CAs. A typical example is L-Dopa (levodopa), the gold standard in the therapy against PD: its pharmacological activity is aimed to increase the dopamine deficit in vivo, but it can also form stable complexes with several metal ions including the PD relevant ones [104]. These compounds are by themselves multifunctional drugs, and their potential influence on the pathological mechanisms involving metal ions cannot be neglected. In the present review, all drugs displaying metal chelation properties, not only those specifically designed for MCT, will be considered.

A bibliographic search has been performed with the aim to collect all promising and established drugs for the PD therapy. Keywords and boolean logics employed for this bibliographic search are given in the Supplementary Materials. Several reviews appeared in 2014 or later, reporting lists of anti-PD compounds [1,12,105,106,107,108,109,110]. These reviews list mainly molecules in use, or which underwent in vivo, or clinical phase tests against PD; here, we also decided to consider compounds which have just been tested in vitro, or even only proposed, e.g., after an in silico approach, because it is likely that some of these will undergo further tests in the following years. Clearly, if a compound was proposed and/or tested in vitro several years ago, and after not more considered as anti-PD drug, likely it was not suitable for this aim and has been abandoned. Due to the latter consideration, and given the availability of the above listed reviews, we decided to limit our bibliographic search to papers published starting from 2014.

For each compound, name(s), chemical structure and the reference(s) were collected. The whole list of these substances is reported in Table S1 of the Supplementary Materials. This table lists approximately 800 compounds and, to the best of our knowledge, it is the most complete table available to date which reports established or potential anti-PD drugs. Table S1 also contains the compounds listed in all above mentioned reviews of PD drugs [1,12,24,79,85,89,104,105,106,107,108,109,110]. Table S1 does not report natural extracts, e.g., drugs obtained from plants or animals, unless the active components have been identified. This therapeutic approach is extensively considered in the literature: for a recent review see, e.g., [111].

Table 2 represents a subset of Table S1 and it lists all compounds (nearly 250) displaying metal-chelating properties which have been used, tested or only proposed for the therapy against PD.

Table 2

Compounds displaying metal-chelation properties which have been used, tested or proposed for the therapy against PD, as obtained from a literature survey in the year range 2014–2019 (April). Substances are listed in the first column according to their alphabetical order. Only the latest and/or the most important references (e.g., reviews) are given in the last column. This Table is a subset of Table S1, which also includes non-chelating compounds and compounds with non-predictable metal–ligand speciation. The chemical structure of each substance is reported in Table S1.

Compound Name(s)	References
7DH	[112]
7MH	[112]
8A	[89]
8B	[89]
8C	[89]
8E	[89]
8F	[89]
N-Acetylcysteine	[113,114]
ACPT-I	[115]
ADX88178	[116]
Alaternin	[117]
Alvespimycin	[118]
AM-251	[119,120]
Ambroxol	[121,122]
3-(7-Amino-5-(cyclohexyl-amino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide	[123]
Aminothiazoles derivatives as SUMOylation activators	[124]
AMN082	[115]
Amodiaquine	[125,126]
Antagonist of the A(2A) adenosine receptor-derivative 49	[127]
Apigenin	[128,129,130,131]
Apomorphine	[132,133]
l-Arginine	[134]
Aromadendrin	[128]
Ascorbic acid	[135,136]
ASI-1	[12]
ASI-5	[12]
Astilbin	[137]
Azilsartan	[138]
Baicalein	[139,140,141]
Benserazide	[142,143]
7H-Benzo, perimidin-7-one derivatives (R6 = OH)	[144]
4H-1-Benzopyran-4-one	[145]
8-Benzyl-tetrahydropyrazino, purinedione derivatives (derivative n.57)	[146]
Bikaverin	[147]
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydro-benzo, thiazole-2,6-diamine derivatives	[148]
2.2’-Bipyridyl (2,2’-bipyridine)	[112]
4-((5-Bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	[149,150]
C-3 (α carboxyfullerene)	[151]
Caffeic acid amide analogues	[152,153,154,155]
Carbazole-derived compounds	[156]
Carbidopa	[135,157]
Carnosic acid	[154,158]
Catechin	[24,128]
Ceftriaxone	[12,159,160,161]
Celastrol	[162,163]
CEP-1347	[164,165]
Chebulagic acid	[166]
Chlorogenic acid	[167]
3′-O-(3-Chloropivaloyl)quercetin	[168]
Chlorpromazine	[108]
Chrysin	[128,169,170]
Clioquinol	[89,91,171,172]
Clioquinol-selegiline hybrid	[79]
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	[173]
“Compound 1”	[174]
“Compound (−)-8a”	[175]
“Compound 8”	[176]
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	[177]
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydro-benzo[d]thiazole-2,6-diamine	[178]
Creatine	[179,180]
Cudraflavone B	[181]
Curcumin	[89,117,182,183,184]
Cyanidin	[185,186]
D-512	[187]
D-607 (bipyridyl-D2R/D3R agonist hybrid)	[12,188,189]
DA-2 (8D)	[12,89]
DA-3	[12]
DA-4	[12]
Dabigatran etexilate	[190]
Dabrafenib	[191]
(S)-3,4-DCPG	[115]
Deferasirox	[24]
Deferricoprogen	[192]
Delphinidin	[160,185,193,194]
Demethoxycurcumin	[195]
Dendropanax morbifera active compound	[196]
Desferrioxamine (Desferoxamine, Desferal, DFO)	[112]
(S)-N-(3-(3,6-Dibromo-9H-carbazol-9-yl)-2-fluoropropyl)-6-methoxypyridin-2-amine	[197]
4,5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid derivatives (R1, R2, R3, R4, or R5 = caffeoyl)	[198]
Dihydromyricetin	[199]
5-(3,4-Dihydroxybenzylidene) -2,2-dimethyl-1,3-dioxane-4,6-dione	[200]
7,8-Dihydroxycoumarin derivative DHC12	[79]
3′,4′-Dihydroxyflavone	[201]
7,8-Dihydroxyflavone	[202,203]
5,7-Dihydroxy-4′-methoxyflavone	[204]
(E)-3,4-Dihydroxystyryl aralkyl sulfones	[205]
(E)-3,4-Dihydroxystyryl aralkyl sulfoxides	[205]
5,3’-Dihydroxy-3,7,4’-trimethoxyflavone	[206]
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	[207]
DKP	[85]
L-DOPA (levodopa, CVT-301)	[132,135,208]
DOPA-derived peptido-mimetics (deprotected)	[209]
DOPA-derived peptido-mimetics (protected)	[209]
L-DOPA deuterated (D3-L-DOPA)	[210]
Doxycycline	[211,212]
Droxidopa	[110]
Echinacoside	[213]
Ellagic acid	[214]
Entacapone (Comtan, ASI-6)	[12,215,216]
Enzastaurin	[164]
Epicatechin	[128,160,193,194]
Epigallocatechin-3-gallate	[117,217,218]
Etidronate (HEDPA)	[219]
Exifone	[12]
F13714	[220]
F15599	[220,221]
Farrerol	[222]
Fisetin (3,3′,4′,7-Tetrahydroxyflavone)	[223,224,225]
Fraxetin	[117]
Galangin	[226]
Gallic acid and derivatives	[214,227,228]
Gallocatechin	[128]
Garcinol	[229]
Genistein	[117,128,230,231]
Glutamine	[232]
Glutathione derivatives	[63]
Glutathione-hydroxyquinoline compound	[233]
Glutathione-l-DOPA compound	[234]
Gly-N-C-DOPA	[209]
GSK2795039	[108]
Guanabenz	[235]
Hesperidin	[128,236]
Hinokitiol	[237]
8-HQ-MC-5 (VK-28)	[12,24,89,92,93]
4-Hydroxyisophthalic acid	[238]
1-Hydroxy-2-pyridinone derivatives	[89,239]
3-Hydroxy-4(1H)-pyridinone (Deferiprone)	[112,239,240]
8-Hydroxyquinoline	[241]
8-Hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	[242]
Hydroxyquinoline-propargyl hybrid (HLA 20)	[12,79,89]
Hydroxytyrosol butyrate	[243]
Hyperoside	[117]
IC87201	[150]
Icariin	[244]
Icariside II	[245]
l-(7-imino-3-propyl-2,3-dihydrothiazolo [4,5-d]pyrimidin-6(7H)-yl)urea	[246]
Imipramine	[247]
Isobavachalcone	[248]
Isochlorogenic acid	[167]
Isoquercetin (Isoquercitrin)	[249]
Kaempferol	[128,160,193,194]
Kaempferol, 3-O-a-l arabinofuranoside-7-O-a-l-rhamnopyranoside	[214]
KR33493	[250]
Kukoamine	[251]
Lestaurtinib	[164]
Lipoic acid	[252,253,254]
Luteolin	[128]
LY354740	[115]
M10	[24]
M30 (VAR10303)	[112,255]
M99	[24]
Macranthoin G	[256]
Magnesium lithospermate B	[257]
α-Mangostin	[258]
γ-Mangostin	[259]
MAOI-1	[12]
MAOI-2	[12]
MAOI-4	[12]
MAOI-8	[12]
Meclofenamic acid	[260]
Metformin (Met)	[261,262]
Methoxy-6-acetyl-7-methylijuglone	[117]
N’-(4-Methylbenzylidene)-5-phenylisoxazole-3-carbohydrazide	[263]
Mildronate	[264]
Minocycline	[12,160,265]
Mitomycin C	[266]
MitoQ	[108]
Morin (3,5,7,29,49-pentahydroxyflavone)	[160,193,194,267]
[18F]MPPF	[107]
MSX-3	[268]
Myricetin	[128,269,270]
Myricitrin	[271]
Naringenin	[128,272]
Naringin	[117,273,274]
Nicotinamide adenine dinucleotide phosphate (NADPH)	[275,276]
Nicotinamide mononucleotide	[277]
Nitecapone	[278]
Nordihydroguaiaretic	[160,193,194]
Oleuropein	[279]
Opicapone	[278]
P7C3	[280,281]
PBF-509	[282]
PBT2	[89,283]
PBT434	[284]
Petunidin	[185]
Phenothiazine 2Bc (n = 0 and n = 1)	[285]
Phenylhydroxamates	[286]
Piceatannol	[160,193,194]
Pinostrobin	[287]
Piperazine-8-OH-quinolone hybrid	[79]
Preladenant	[282,288]
Promethazine	[289]
Protocatechuic acid	[170]
Protosappanin A	[290]
Punicalangin	[270,291]
Pyrazolobenzothiazine-based carbothioamides	[292]
Pyridoxal isonicotinoyl hydrazone (PIH) and related compounds	[24,89]
Pyrimidinone 8	[293]
Q1	[89]
Q4	[89]
Quercetin	[117,294,295]
Quinolines Derivatives as SUMOylation activators	[124]
Radotinib	[296]
Riboflavin	[297,298]
Rifampicin (ASI-3)	[12,160]
Rimonabant	[119,120,282]
Rosmarinic acid	[154,299]
Rotigotine	[105,133,278,300]
Rutin	[128,249]
Salicylate, sodium salt	[301]
Salvianolic Acid B	[117]
SCH-58261	[105,302]
SCH412348	[105]
Silibinin (silybin) A, B	[89]
Silydianin	[24]
ST1535	[282]
ST4206	[282]
Staurosporine	[164]
Stemazole	[303]
Sulfuretin	[304]
Tannic acid	[160,193,194]
Tanshinol	[117]
Taurine	[305]
Taxifolin	[128]
Tectorigenin	[306]
Tetracycline	[307]
Tolcapone (ASI-7)	[12,105]
Tozadenant	[282,302]
Transilitin	[270]
O-Trensox	[24]
2′,3′,4′-Trihydroxyflavone	[270]
2,3,3-Trisphosphonate	[219]
V81444	[282]
VAS3947	[108]
VAS2870	[108]
Verbascoside	[160,193,194]
WIN 55,212-2	[119,120]
WR-1065	[308]
Zonisamide	[309,310,311]

It is rather easy to predict whether a given compound can form stable complexes with metal ions, and thus whether it in principle can affect metal homeostasis in the brain: the presence of at least two functional groups with metal-binding ability is suggested, and the formed chelation ring should have five or six members. Truly, monodentate ligands and those forming chelating rings of a different size to 5 or 6 can also coordinate metal ions, but the resulting complexes are generally too weak to allow these ligands to affect metal speciation in vivo. Coordinating functional groups can be negatively charged (or partially charged) oxygens such as carboxylates, phenolates, N-oxides, as well as nitrogen and sulphur atoms with non-delocalized lone pairs such as amines and thio-derivatives. For example, l-Dopa (chemical structure drawn in Table S1) has two possible binding sites, one given by the two phenolic oxygens (catechol-like), and the other by the amino-acidic tail (glycine-like). Both binding sites, upon chelation, form a five membered ring with the metal ion. Some simple metal-chelating moieties often contained in CAs are depicted in Figure 1

Figure 1

Simple metal-chelating moieties (listed in alphabetical order) which are often encountered in molecules used or proposed for the metal chelation therapy in Parkinson’s disease.

The formation of stable metal–ligand complexes is more difficult to predict for peptides, because their metal chelation ability is strongly influenced by their spatial configuration. Also, it is not possible to assess the metal–ligand speciation of molecules bearing many chelating functional groups. Molecules of this kind, cited in Table S1, were not reported in Table 2. For the relevant metal ions (Cu(II), Cu(I), Fe(III), Fe(II), Mn(II), and Zn(II)) and for the compounds listed in Table 2, the metal–ligand speciation will be given.

4. The Measurement of the Stability of Metal–Ligand Complexes

While no ambiguities exist for reporting the stoichiometries of the complexes, it is worth describing all possible amounts used in literature to measure the stability of metal–ligand complexes: cumulative stability constants, stepwise stability constants, conditional stability constants, cologarithm of the concentration of free metal ion at equilibrium (pM), association constants, and dissociation constants.

Cumulative (or overall) stability constants are generally indicated with the Greek letter β. Each complex forming in solution is characterized by a β value. If M is the metal ion, H the proton, and L the ligand, and MHL is the complex formed, β is defined as: where square brackets denote concentrations at equilibrium. The use of concentration amounts instead of activities (i.e., concentrations multiplied by activity coefficients) is generally allowed by maintaining a constant ionic strength during the experimental measurements. The (at least formal) presence of activities in equation (2) justifies why β values are commonly indicated without a measuring unit. Cumulative constants are usually given as logarithm (logβ), and their knowledge is required for performing metal–ligand speciation calculations. Their experimental determination is however complicated as many experimental details have to be considered in order to obtain accurate results [312,313]. In addition, β values do not allow to state the effective complex stability, which is also affected by the acid—base properties of metal ion and ligand, by the total metal (cM) and ligand (cL) concentrations, and by the pH. In other words, logβ values by themselves are not informative and cannot be used to compare the stability of complexes formed by different metal ions and ligands.

Stepwise stability constants are generally indicated with the letter K and they are written as logK. Stepwise constants are more commonly employed if the complexes existing in solution contain one metal ion and one or more ligands (ML, with l ≥ 1). For example, for the complex ML, K can be defined as:

Stepwise constants are related with cumulative ones, so that the former can be computed from the latter and vice versa— e.g., for the complexes ML and ML2, logβML = logKML and logβML = logKML + logKML, respectively.

Conditional (or effective) stability constants may be cumulative or stepwise and are indicated with the apostrophe (β ‘ or K ‘). For example, the conditional cumulative constant of the complex ML2 is defined as: where Σ[M′] and Σ[L′] represent the sum of the concentrations of uncomplexed metal ion and uncomplexed ligand at equilibrium, i.e., [M] + [M(OH)] + [M(OH)2] + … and [L] + [HL] + [H2L] + …, respectively. As Σ[M′] and Σ[L′] depend on pH, also logβ′ and logK′ values depend on pH. These constants represent the effective stability of the given complexes in the presence of acid—base equilibria, so that they can be used to compare the stability of complexes formed by different metal ions and ligands. However, the comparison is possible only if the complexes have the same stoichiometries.

The other three amounts used to measure metal–ligand affinities, i.e., pM, association constants, and dissociation constants, differ from logβ, logK, and from their conditional values, because only one number is given to characterize a solution containing the given metal and ligand. This is particularly useful when many complexes are formed and, in overall, one value resumes their strength.

pM represents the cologarithm of the concentration of free metal ion at equilibrium (pM = –log[M]), and it can be calculated if log β or logK values are known. The larger is pM, i.e., the lower is [M], the stronger are the complexes; pM can be used to compare the relative strength of the complexes, irrespective of their number and their stoichiometry. As pM depends on pH and also on cM and cL, calculation must be performed under the same conditions: usually the pM value for MCT relevant conditions is computed at pH = 7.4, cM = 10−6 mol/L and cL = 10−5 mol/L [24,70,314]. The only (but important) disadvantage of pM is that it can be computed if logβ or logK values are known, i.e., the experimental procedure required to gain a pM value remains complicated.

Association constants, indicated as Ka, are defined like β or K if only one complex ML forms in solution and no acid–base equilibria coexist:

If M and/or L undergo acid–base equilibria, Ka for the complex ML is defined like K′ or β′. When more complexes of general stoichiometry MHL coexist in solution, more K′ or β′ are needed, whereas still only one Ka suffices and is defined as: where Σ[MHL] represents the sum of the concentrations of all complexes existing in solution. As Ka values are not thermodynamically defined, i.e., concentration values are employed instead of activities, they bear a measuring unit, which according to equation (6) is L/mol (or, more commonly, a multiple).

The dissociation constant, indicated as Kd, is the inverse of Ka (measuring unit of Kd: mol/L or a multiple). For example, if complexes of general stoichiometry MHL coexist in solution, Kd can be defined as:

Kiss et al. [102] reported a similar definition of Kd, where [M] was used instead of Σ[M’]. However, as the proton content is experimentally not controlled when Kd values are measured, not only for the ligand but also for the metal ion, we think that Equation (7) allows a more rigorous calculation of Kd values. Literature appears to have preference for measuring and reporting Kd more than Ka [313,315]. This is probably due to the chemical usefulness and significance of Kd, as it represents the concentration of free metal ion at which the concentrations of free ligand and of the complexes are equivalent [313]. In the following, Kd values will be considered instead of Ka ones.

Values of Kd can be determined with a much simpler experimental design than that used to obtain logβ, logK and pM [313,315]. This is a crucial advantage when complicated ligands are studied, such as proteins, for which the determination of logβ or logK is practically impossible. Still, Kd can be computed from Equation (7) if logβ or logK values are available, so that the concentrations of all species existing in solution can be calculated. Therefore, Kd represents a simple tool and practically the only way to compare metal–ligand and metal–protein complex stabilities each other.

The main disadvantage of Kd is that it depends not only on pH, cM and cL, but also on the copresence of other ligands or other metal ions [316], as these affect Σ[M′] and Σ[L′] in Equation (7) (the same applies for Ka in equation 6, too). This explains at least in part why reported Kd values are scarcely reproducible (see Table 1) and depend on the experimental conditions [12]. Equations to correct Kd values, by taking into account the effect of a competing ligand (e.g., the buffer) and of the different pH, have been proposed [316]. Standardized conditions to measure the Kd of metal–protein complexes are also being proposed [317,318,319], and this should lead to more reproducible results, thus eventually allowing a more reliable comparison among Kd numbers.

5. The Metal–Ligand Speciation of Anti-Parkinson Drugs

Table S2 (Supplementary Materials) reports the metal–ligand speciation available in the literature for the ligands listed in Table 2 (rows) and the relevant metal ions, i.e., Cu(II), Cu(I), Fe(III), Fe(II), Mn(II), and Zn(II) (columns). If not differently specified in the notes of Table S2, speciation information (stoichiometries of the complexes, and stability constants given as logβ) has been obtained from the IUPAC stability constant database [320].

The ionic product of water, the stability constants of the metal ion hydrolysis products, and the acidity constants of each ligand, have to be considered to complete the speciation picture and allow speciation calculations. The ionic product of water and the stability constants for hydrolysis products of the considered metal ions are resumed in Table S3 in the Supplementary Materials (it is worth noting that these values are only sometimes reported in papers dealing with metal speciation). The acidity constants of the ligands listed in Table 2 have also been taken from the IUPAC stability constant database [320] or from the papers reported in the notes, and they are given as logβ values in Table S2 (column marked ‘H+’).

For some metal–ligand complexes, and for many ligand acidity constants, more than one speciation set has been reported in literature, and/or different logβ values were proposed. For example, 27 different speciation models have been obtained for the Cu(II)/L-Dopa complexes [320] In Table S2 only one speciation set has been reported, obtained at ionic strength and temperature as close as possible to 0.1 mol/L and 25 °C, respectively. This ionic strength represents a reasonable physiological environment; as regards temperature, 37 °C would better resemble physiological conditions, but speciation data at this temperature are few. For comparison purposes, we preferentially reported data at the most studied temperature of 25 °C. Notes were added in Table S2 if the studied temperature and ionic strength were different than 25 °C and 0.1 mol/L, respectively.

For many other ligands listed in Table 2, no metal speciation set, and sometimes also no acidity constants, were available. This can be ascribed to several reasons, which for some CAs include their very recent development or proposal. For example, 3-(7-Amino-5-(cyclohexylamino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide, Aromadendrin, Astilbin, and many other CAs listed in Table 2 have been proposed for MCT in PD only approximately (or less than) one year ago. Possibly, the absence of equilibrium constants can also be justified by the above-mentioned experimental difficulties associated to accurate speciation measurements, which for complicated and often poorly water-soluble molecules may become formidable. Nevertheless, it is possible to tentatively predict the metal speciation of such ligands, by individuating the chelating moiety which is responsible for the complex formation (see also Figure 1), and by considering a simpler ligand having a known metal speciation and bearing the same moiety: ligands having the same chelating functional groups are expected to have a similar metal speciation. For example, complicated molecules bearing a 1,2-diaminoethane chelating group have been considered to have the same metal speciation as 1,2-diaminoethane itself. This predicted speciation should be employed with caution, because inductive and steric effects (and especially resonance ones, if existing) of the remaining part of the molecule might significantly modify the speciation picture. However, these estimations should represent the most reliable values available, until dedicated experimental measurements will be performed. Whenever this kind of assignation has been done, the reported metal-speciation has been marked as “tentative” in the notes of Table S2. A speciation prediction has also been attempted for molecules bearing two or a maximum of three chelating moieties, by considering their speciation to be similar to that of a simpler ligand bearing the moiety forming the most stable complexes. In these cases, however, the inaccuracy of the predicted speciation might be relatively large.

Table 3 reports selected speciation information regarding Cu(II) which has been computed from the data of Tables S2 and S3. Only the ligands for which a Cu(II) speciation was available or has been tentatively estimated were reported in Table 3. Calculations have been performed by the software PITMAT (see [70] and references therein). The first reported value is that of pM (pCu(II) = –log[Cu(II)], which has been computed at pH = 7.4, at cM = 10−6 mol/L and at cL = 10−5 mol/L, as recommended in MCT modeling [70,314]. Besides pM, Kd also was computed according to Equation (7). To the best of our knowledge, no reference pH, cM and cL values have been hitherto adopted for the calculation of Kd. We propose here that this calculation should be performed at the same conditions as for pM: pH = 7.4, cM = 10−6 mol/L, and cL = 10−5 mol/L. The last column of Table 3 reports the most abundant metal complex existing for each ligand at these same physiologically relevant conditions; if available, the charge of this complex is reported as well. Table 4, Table 5, Table 6, Table 7, Table 8 report the same information computed for Cu(I) (Table 4), Fe(III) (Table 5), Fe(II) (Table 6), Mn(II) (Table 7), and Zn(II) (Table 8).

Table 3

pCu(II) and Kd values, and the most abundant Cu(II) complex, obtained at physiologically relevant conditions: pH = 7.4, cCu = 10–6 mol/L, and cL = 10–5 mol/L. Computations have been performed from data of Tables S2 and S3 (Supplementary Materials). Refer to Table S2 to identify ligands for which only tentative speciations were proposed, and ligands whose complexes have unknown charges.

Compound Name(s)	pCu(II)	Kd (nmol/L)	Most Abundant Complex
7DH 7MH	14.2	5.91 × 10−5	CuL2
8A8B8C	14.2	5.91 × 10−5	CuL2
8E8F	10.6	2.35 × 10−1	CuL2
N-Acetyl cysteine	6.2	3.82 × 104	CuL
ACPT-I	7.3	5.93 × 102	CuL
ADX88178	7.4	5.15 × 102	CuL
Alaternin	16.5	3.39 × 10−7	CuL2
Alvespimycin	10.3	5.35 × 10−1	CuL22+
AM-251	6.3	1.47 × 104	CuL
Ambroxol	9.2	7.26	CuL+
3-(7-Amino-5-(cyclohexylamino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide	9.1	7.58	CuL2
Aminothiazoles derivatives as SUMOylation activators	9.7	1.86	CuL2
AMN082	10.3	5.35 × 10−1	CuL22+
Amodiaquine	6.1	1.90 × 107	CuHL
Antagonist of the A(2A) adenosine receptor-derivative 49	6.4	9.65 × 103	CuL
Apigenin	6.8	2.39 × 103	CuH2L+
Apomorphine	7.4	5.06 × 102	CuL
l-Arginine	7.1	9.57 × 102	CuL2+
Aromadendrin	11.1	8.29 × 10−2	CuL
Ascorbic acid	6.1	2.43 × 105	Cu2H–2L2
ASI-1	7.9	1.21 × 102	CuL2
ASI-5	6.1	2.23 × 106	CuL
Astilbin	7.4	5.06 × 102	CuL
Azilsartan	6.1	4.53 × 105	CuL
Baicalein	9.3	5.41	CuL22−
Benserazide	9.3	5.30	CuL2
7H-benzo[e] perimidin-7-one derivatives	14	1.05 × 10−4	CuL2
8-Benzyl-tetrahydropyrazino[2,1-f]purinedione (derivative n. 57)	6.1	3.36 × 106	CuL
Bikaverin	14	1.05 × 10−4	CuL2
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine derivatives	10.3	5.35 × 10−1	CuL22+
2,2′-bipyridyl	10.6	2.35 × 10−1	CuL22+
4-((5-bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	6.2	6.51 × 104	CuL
C-3 (α-carboxyfullerene)	6.9	1.96 × 103	CuL
Caffeic acid amide analogues	7.3	6.02 × 102	CuH−1L
Carbazole-derived compounds	10.3	5.35 × 10−1	CuL22+
Carbidopa	15.1	8.11 × 10−6	CuH−2L
Carnosic acid	7.4	5.06 × 102	CuL
Catechin	7.9	1.50 × 102	CuH2L
Ceftriaxone	6.1	4.04 × 106	CuL
Celastrol	6.5	5.27 × 103	CuL
Chebulagic acid	6.3	1.51 × 104	CuHL
Chlorogenic acid	8.3	5.04 × 101	CuL−
3′-O-(3-chloropivaloyl) quercetin	11.1	8.29 × 10−2	CuL
Chlorpromazine	6.3	1.47 × 104	CuL2+
Chrysin	10.6	2.97 × 10−1	CuHL+
Clioquinol	14.2	5.91 × 10−5	CuL2
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	7.4	5.06 × 102	CuL
“Compound 1”	10.3	5.35 × 10−1	CuL22+
“Compound 8”	6.1	4.37 × 105	CuL2
“Compound 21”, derivative of3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	7.3	5.93 × 102	CuL+
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine	10.3	5.35 × 10−1	CuL23+
Creatine	6.8	2.26 × 103	CuH−1L
Cudraflavone B	11.1	8.29 × 10−2	CuL
Curcumin	7.9	1.21 × 102	CuL2
Cyanidin	7.4	5.06 × 102	CuL
D512	10.3	5.35 × 10−1	CuL22+
D607 (bipyridyl-D2R/D3R agonist hybrid)	10.6	2.35 × 10−1	CuL2
DA-2 (8D)	14.2	5.91 × 10−5	CuL2
DA-3	10.3	5.35 × 10−1	CuL2
DA-4	10.3	5.35 × 10−1	CuL2
Dabigatran etexilate	10.3	5.35 × 10−1	CuL2
Dabrafenib	6.1	3.36 × 106	CuL
(S)-3-4-DCPG	6.1	4.54 × 105	CuL
Deferricoprogen	12.6	3.17 × 10−3	CuHL
Delphinidin	9.3	5.30	CuL2
Demethoxycurcumin	7.9	1.21 × 102	CuL2
Dendropanax morbifera active compound	7.4	5.06 × 102	CuL
Desferrioxamine (Deferoxamine, Desferal, DFO)	11.4	4.98 × 10−2	CuH2L+
(S)-N-(3-(3-6-dibromo-9H-carbazol-9-yl)-2-fluoropropyl)-6-methoxypyridin-2-amine	6.3	1.47 × 104	CuL
4, 5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid(R1, R2, R3, R4, or R5 = caffeoyl)	7.4	5.06 × 102	CuL
Dihydromyricetin	9.3	5.30	CuL2
5-(3,4-Dihydroxybenzylidene)-2,2-dimethyl-1,3-dioxane-4,6-dione	7.4	5.06 × 102	CuL
7,8-Dihydroxycoumarin derivative DHC12	7.4	5.06 × 102	CuL
3′,4′-Dihydroxyflavone	7.4	5.06 × 102	CuL
7,8-Dihydroxyflavone	7.4	5.06 × 102	CuL
5,7-Dihydroxy-4′-methoxyflavone	6.6	4.58 × 103	CuL
(E)-3, 4-Dihydroxystyryl aralkyl sulfones	7.4	5.06 × 102	CuL
(E)-3, 4-Dihydroxystyryl aralkyl sulfoxides	7.4	5.06 × 102	CuL
5,3′-Dihydroxy-3,7,4′-trimethoxyflavone	6.6	4.58 × 103	CuL
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	9.6	2.98	CuL
DKP	8.1	8.43 × 101	CuL
l-DOPA (levodopa, CVT-301)	15.2	7.95 × 10−6	CuH−2L3−
DOPA-derived peptido-mimetics (deprotected)	15.2	7.95 × 10−6	CuH−2L2
DOPA-derived peptido-mimetics (protected)	7.4	5.06 × 102	CuL
l-DOPA deuterated	15.2	7.95 × 10−6	CuH−2L3−
Doxycycline	8.9	1.28 × 101	CuL
Droxidopa	15.2	7.95 × 10−6	CuH−2L3−
Echinacoside	7.4	5.06 × 102	CuL
Ellagic acid	7.4	5.06 × 102	CuL
Entacapone (comtan, ASI-6)	10.1	8.00 × 10−1	CuL22−
Enzastaurin	10.3	5.35 × 10−1	CuL2
Epigallocatechin-3-gallate	6.1	1.33 × 105	CuH2L2
Etidronate (HEDPA)	9	1.11 × 101	CuL2−
Exifone	6.3	1.51 × 104	CuHL
F13714, F15599	9.7	1.86	CuL2
Farrerol	11.1	8.29 × 10−2	CuL
Fisetin (3,3′,4′,7-tetra-hydroxy-flavone)	7.4	5.06 × 102	CuL
Fraxetin	7.4	5.06 × 102	CuL
Galangin	9.1	8.20	CuL
Gallic acid derivatives	6.3	1.51 × 104	CuHL−
Gallocatechin	9.3	5.30	CuL2
Garcinol	7.4	5.06 × 102	CuL
Genistein	11.1	8.29 × 10−2	CuL
Glutamine	7.3	5.38 × 102	CuL+
Glutathione derivatives	6.2	7.23 × 104	CuL−
Glutathione-hydroxy-quinoline compound	9.4	5.00	CuH−1L+
Glutathione-l-DOPA compound	13.5	3.98 × 10−4	CuH−1L
Gly-N-C-DOPA	15.2	7.95 × 10−6	CuH−2L3−
GSK2795039	12	1.02 × 10−2	CuL2
Guanabenz	10.3	5.35 × 10−1	CuL2
Hesperidin	11.1	8.29 × 10−2	CuL
Hinokitiol	7.3	6.69 × 102	CuL+
8-HQ-MC-5 (VK-28)	14.2	5.91 × 10−5	CuL2
4-Hydroxyisophthalic acid	6.3	2.06 × 104	CuL
1-Hydroxy-2-pyridinone derivatives	8.4	4.26 × 101	CuL2
3-Hydroxy-4(1H)pyridinone (Deferiprone)	10.2	6.60 × 10−1	CuL2
3-Hydroxy-4(1H)pyridinone derivatives (R = H)	10.2	6.60 × 10−1	CuL2
8-Hydroxyquinoline	14.2	5.91 × 10−5	CuL2
8-Hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	14.2	5.91 × 10−5	CuL2
Hydroxy-quinoline-propargyl hybrids (HLA20)	14.2	5.91 × 10−5	CuL2
Hydroxytyrosol butyrate	7.4	5.06 × 102	CuL
Hyperoside	9.1	8.20	CuL
IC87201	6.1	1.90 × 107	CuHL
Icariin	6.6	4.58 × 103	CuL
Icariside II	6.6	4.58 × 103	CuL
l-(7-Imino-3-propyl-2,3-dihydrothiazolo[4,5-d]pyrimidin-6(7H)-yl)urea	6.1	4.37 × 105	CuL2
Imipramine	6.3	1.47 × 104	CuL2+
Isobavachalcone	6.2	8.13 × 104	CuL+
Isochlorogenic acid	8.3	5.04 × 101	CuL−
Isoquercetin (isoquercitrin)	9.1	8.20	CuL
Kaempferol	9.1	8.20	CuL
KR33493	7.3	5.93 × 102	CuL
Kukoamine	7.4	5.06 × 102	CuL
Lestaurtinib	10.3	5.35 × 10−1	CuL2
Lipoic acid	6.1	4.41 × 105	CuL+
Luteolin	7.4	5.06 × 102	CuL
M10M30 (VAR10303)M99	14.2	5.91 × 10−5	CuL2
Macranthoin G	7.4	5.06 × 102	CuL
Magnesium lithospermate B	7.4	5.06 × 102	CuL
α-mangostin	7.4	5.06 × 102	CuL
γ-Mangostin	7.4	5.06 × 102	CuL
MAOI-1	9.3	5.30	CuL2
MAOI-2	10.3	5.35 × 10−1	CuL2
MAOI-4	6.3	1.47 × 104	CuL
MAOI-8	6.1	1.90 × 107	CuHL
Metformin (met)	6.3	1.61 × 104	CuL+
Methoxy-6-acetyl-7-methylijuglone	14	1.05 × 10−4	CuL2
N′-(4-methylbenzylidene)-5-phenylisoxazole-3-carbohydrazide	6.3	1.38 × 104	CuL
Minocycline	11.6	2.83 × 10−2	CuL
Mitomycin C	7.1	9.14 × 102	CuL
Morin	6.1	2.27 × 1015	CuH3L
[18F]MPPF	10.3	5.35 × 10−1	CuL2
MSX-3	6.1	2.01 × 106	CuL
MyricetinMyricitrin	9.1	8.20	CuL
Naringin	6.9	1.58 × 103	CuHL
Naringenin	6.9	1.58 × 103	CuHL
Nicotinamide adenine dinucleotide phosphate (NADPH)	6.4	8.91 × 103	CuL
Nicotinamide mononucleotide	6.1	2.01 × 106	CuL
Nitecapone	10.1	8.00 × 10−1	CuL22−
Nordihydroguaiaretic acid	7.4	5.06 × 102	CuL
Oleuropein	7.4	5.06 × 102	CuL
Opicapone	10.1	8.00 × 10−1	CuL2
P7C3	7.8	3.76 × 102	Cu2H−2L2+
PBT2	12.4	4.04 × 10−3	CuL+
PBT434	11.2	7.21 × 10−2	CuL+
Petunidin	7.4	5.06 × 102	CuL
Phenothiazine 2Bc (n=0)	10.3	5.35 × 10−1	CuL22+
Phenothiazine 2Bc (n=1)	6.3	1.47 × 104	CuL2+
Phenylhydroxamates	7	1.46 × 103	CuL
Piceatannol	7.4	5.06 × 102	CuL
Pinostrobin (5-hydroxy-7-methoxy-flavone)	6.6	4.58 × 103	CuL+
Piperazine-8-OH-quinolone hybrid	14.2	5.91 × 10−5	CuL2
Preladenant	10.3	5.35 × 10−1	CuL2
Promethazine	8.2	7.99 × 101	CuL2+
Protocatechuic acid	8.1	8.13 × 101	CuL−
Protosappanin A	7.4	5.06 × 102	CuL
Punicalangin	8.2	6.89 × 101	CuL
Pyrazolobenzothiazine-based carbothioamides	6.1	4.66 × 105	CuL
Pyrimidinone 8	10.3	5.35 × 10−1	CuL2
Q1 Q4	14.2	5.91 × 10−5	CuL2
Quercetin	9.1	8.20	CuL3−
Quinoline derivatives SUMOylation activators	7.2	7.26 × 102	CuL2+
Radotinib	10.6	2.35 × 10−1	CuL2
Riboflavin	6.1	1.65 × 105	CuHL3+
Rifampicin (ASI-3)	6.5	6.97 × 103	CuL
Rimonabant	8.1	8.43 × 101	CuL
Rosmarinic acid	7.4	5.06 × 102	CuL
Rotigotine	6.1	1.45 × 108	CuL2
Rutin	9.1	8.20	CuL
Salicylate, sodium salt	6.3	2.06 × 104	CuL
Salvianolic acid B	7.4	5.06 × 102	CuL
SCH58261SCH412348	9.1	7.58	CuL2
ST1535 ST4206	9.1	7.58	CuL2
Staurosporine	10.3	5.35 × 10−1	CuL2
Stemazole	6.1	4.66 × 105	CuL
Sulfuretin	7.4	5.06 × 102	CuL
Tannic acid	6.1	1.84 × 105	CuL
Tanshinol	7.4	5.06 × 102	CuL
Taurine	6.1	1.23 × 107	CuL+
Taxifolin	10.4	4.55 × 10−1	CuL2−
Tectorigenin	11.1	8.29 × 10−2	CuL
Tetracycline	6.4	1.08 × 104	CuL
Tolcapone (ASI-7)	10.1	8.00 × 10−1	CuL2
Transilitin	7.4	5.06 × 102	CuL
o-Trensox	22.9	1.51 × 10−13	CuL4−
2′, 3′, 4′-Trihydroxyflavone	9.3	5.30	CuL22−
2,3,3-Trisphosphonate	14	9.98 × 10−5	CuL2
V81444	9.7	1.86	CuL2
VAS3947 VAS2870	6.1	4.37 × 105	CuL2
Verbascoside	7.3	6.02 × 102	CuH−1L
WIN 55, 212-2	10.3	5.35 × 10−1	CuL22+
WR-1065	6.6	3.67 × 103	CuL2+
Zonisamide	7.4	4.62 × 102	CuL

Table 4

pCu(I) and Kd values, and the most abundant Cu(I) complex, obtained at physiologically relevant conditions: pH = 7.4, cCu = 10−6 mol/L, and cL = 10−5 mol/L. See caption of Table 3 for other notes.

Compound Name(s)	pCu(I)	Kd (nmol/L)	Most Abundant Complex
7DH7MH	6.3	8.38 × 103	CuL2
8A8B8C	6.3	8.38 × 103	CuL2
8E8F	6.2	1.87 × 104	CuL
ACPT-I	6	2.54 × 108	CuL2
ADX88178	7.7	1.77 × 102	CuL
Alvespimycin	6	7.98 × 107	CuL2+
3-(7-Amino-5-(cyclohexylamino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide	6	7.98 × 107	CuL2
Aminothiazoles derivatives as SUMOylation activators	6	2.66 × 106	CuL2
AMN082	6	7.98 × 107	CuL2
Antagonist of the A(2A) adenosine receptor (derivative 49)	6	3.06 × 107	CuL2
8-Benzyl-tetrahydropyrazino[2,1-f]purinedione (derivative 57)	7.6	2.17 × 102	CuL
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine derivatives	6	7.98 × 107	CuL2+
2,2′-bipyridyl	6.2	1.87 × 104	CuL+
Carbazole-derived compounds	6	7.98 × 107	CuL2+
Ceftriaxone	6	2.54 × 108	CuL2
Clioquinol	7.2	5.79 × 102	CuL2−
“Compound 1”	6	7.98 × 107	CuL2+
“Compound 8”	7.6	2.17 × 102	CuL
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	6	2.54 × 108	CuL2−
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine	6	7.98 × 107	CuL2+
Creatine	6	2.54 × 108	CuL2−
D512	6	7.98 × 107	CuL2+
D607(bipyridyl-D2R/D3R agonist hybrid)	6.2	1.87 × 104	CuL
DA-2 (8D)	6.3	8.38 × 103	CuL2
DA-3	6	7.98 × 107	CuL2
DA-4	6	7.98 × 107	CuL2
Dabigatran etexilate	6	7.98 × 107	CuL2
Dabrafenib	7.7	1.77 × 102	CuL
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	6.9	1.07 × 103	CuH2L2
DKP	7.4	3.64 × 102	CuL
Doxycycline	9.2	1.26 × 101	Cu2L
Enzastaurin	6	7.98 × 107	CuL2
F13714F15599	6	7.83 × 105	Cu2L
Glutathione-hydroxy-quinoline compound	6.3	8.38 × 103	CuL2−
Glutathione derivatives	15.2	6.21 × 10−6	CuHL−
Guanabenz	6	7.98 × 107	CuL2
8-HQ-MC-5 (VK-28)	6.3	8.38 × 103	CuL2
8-hydroxyquinoline	6.3	8.38 × 103	CuL2−
8-hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	6.3	8.38 × 103	CuL2
Hydroxy-quinoline-propargyl hybrids (HLA20)	6.3	8.38 × 103	CuL2
l-(7-Imino-3-propyl-2,3-dihydrothiazolo [4,5-d]pyrimidin-6(7H)-yl)urea	7.6	2.17 × 102	CuL
KR33493	6	2.54 × 108	CuL2
Lestaurtinib	6	7.98 × 107	CuL2
M10M30 (VAR10303)M99	6.3	8.38 × 103	CuL2
MAOI-2	6	7.98 × 107	CuL2
[18F]MPPF	6	7.98 × 107	CuL2
PBF-509	6	7.98 × 107	CuL2
PBT2	6.3	8.38 × 103	CuL2
Phenothiazine 2Bc (n=0)	6	7.98 × 107	CuL2+
Piperazine-8-OH-quinolone hybrid	6.3	8.38 × 103	CuL2
Preladenant	6	7.98 × 107	CuL2
Promethazine	6	7.98 × 107	CuL2+
Pyrimidinone 8	6	7.98 × 107	CuL2
Q1Q4	6.3	8.38 × 103	CuL2
Radotinib	6.2	1.87 × 104	CuL
Rifampicin (ASI-3)	9.2	1.26 × 101	Cu2L
Rimonabant	7.4	3.64 × 102	CuL
Rotigotine	7.7	1.77 × 102	CuL
SCH58261 SCH412348	6	7.98 × 107	CuL2
ST1535 ST4206	6	7.98 × 107	CuL2
Staurosporine	6	7.98 × 107	CuL2
V81444	6	2.66 × 106	CuL2
VAS3947 VAS2870	7.6	2.17 × 102	CuL
WIN 55, 212-2	6	7.98 × 107	CuL2+
WR-1065	7.6	2.17 × 102	CuL

Table 5

pFe(III) and Kd values, and the most abundant Fe(III) complex, obtained at physiologically relevant conditions: pH = 7.4, cFe = 10–6 mol/L, and cL = 10–5 mol/L. See caption of Table 3 for other notes.

Compound Name(s)	pFe(III)	Kd (nmol/L)	Most Abundant Complex
7DH7MH	20.6	2.15 × 10−1	FeL3
8A8B8C	20.6	2.15 × 10−1	FeL3
8E8F	21.5	3.01 × 10−2	FeH–2L2
N-Acetyl cysteine	16.1	4.59 × 109	FeL2−
ACPT-I	16.1	9.59 × 107	FeL2
Ambroxol	16.3	1.70 × 104	FeL2+
Apigenin	16.1	7.55 × 108	FeL2+
Apomorphine	16.3	1.35 × 104	FeL2
l-Arginine	16.1	1.18 × 1012	FeL3+
Aromadendrin	16.1	7.55 × 108	FeL
Ascorbic acid	16.1	4.99 × 1017	FeL2+
ASI-1	16.8	2.28 × 103	FeL
ASI-5	18	1.06 × 102	FeL
Astilbin	16.3	1.35 × 104	FeL2
Baicalein	16.1	7.55 × 108	FeL2+
4H-1-benzopyran-4-one	18.1	8.14 × 101	FeL2
2,2′-bipyridyl	21.5	3.01 × 10−2	FeH−2L2+
4-((5-bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	16.1	4.46 × 107	FeL2
C-3 (α carboxyfullerene)	16.1	2.54 × 1010	FeL2
Caffeic acid amide analogues	16.3	1.35 × 104	FeL2
Carbidopa	16.2	2.96 × 104	FeL
Carnosic acid	16.3	1.35 × 104	FeL2
Catechin	16.1	4.49 × 1017	FeHL
Ceftriaxone	16.1	9.59 × 107	FeL2
Celastrol	19.2	6.33	FeL2
Chebulagic acid	16.1	7.45 × 105	FeHL
Chlorogenic acid	16.1	1.07 × 107	FeL
3′-O-(3-Chloropivaloyl) quercetin	16.1	7.55 × 108	FeL
Chrysin	16.1	7.55 × 108	FeL+
Clioquinol	20.6	2.15 × 10−1	FeL3
Clioquinol-selegiline hybrid	22.9	1.07 × 10−3	FeL2
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	16.3	1.35 × 104	FeL2
“Compound (−)-8a”	16.3	1.35 × 104	FeL2
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	16.1	9.59 × 107	FeL2+
Creatine	16.1	9.59 × 107	FeL2+
Cudraflavone B	16.1	7.55 × 108	FeL
Curcumin	16.6	4.09 × 103	FeL
Cyanidin	16.3	1.35 × 104	FeL2
D607 (bipyridyl-D2R/D3R agonist hybrid)	21.5	3.01 × 10−2	FeH−2L2
DA-2 (8D)	20.6	2.15 × 10−1	FeL3
DA-3	17.2	5.61 × 102	FeL2
DA-4	17.2	5.61 × 102	FeL2
Deferasirox	23.5	3.22 × 10−4	FeL23−
Delphinidin	16.3	1.35 × 104	FeL2
Demethoxycurcumin	16.8	2.28 × 103	FeL
Dendropanax morbifera active compound	16.3	1.35 × 104	FeL2
Desferrioxamine (Deferoxamine, Desferal, DFO)	26.8	1.81 × 10−7	FeHL+
4,5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid derivatives (R1, R2, R3, R4, or R5 = caffeoyl)	16.3	1.35 × 104	FeL2
Dihydromyricetin	16.3	1.35 × 104	FeL2
5-(3,4-dihydroxybenzylidene)-2,2-dimethyl-1,3-dioxane-4,6-dione	16.3	1.35 × 104	FeL2−
7,8-dihydroxycoumarin derivative DHC12	16.3	1.35 × 104	FeL2
3′,4′-dihydroxyflavone	16.3	1.35 × 104	FeL2−
7,8-dihydroxyflavone	16.3	1.35 × 104	FeL2−
5,7-dihydroxy-4′-methoxyflavone	18.1	8.14 × 101	FeL2
(E)-3,4-dihydroxystyryl aralkyl sulfones	16.3	1.35 × 104	FeL2−
(E)-3,4-dihydroxystyryl aralkyl sulfoxides	16.3	1.35 × 104	FeL2−
5,3′-dihydroxy-3,7,4′-trimethoxyflavone	18.1	8.14 × 101	FeL2
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	16.1	1.86 × 1010	FeL
DKP	16.1	7.43 × 1021	FeL2
l-DOPA (levodopa, CVT-301)	16.2	2.96 × 104	FeL
DOPA-derived peptido-mimetics (deprotected)	16.2	2.96 × 104	FeL
DOPA-derived peptido-mimetics (protected)	16.3	1.35 × 104	FeL2
l-dopa deuterated	16.2	2.96 × 104	FeL
Doxycycline	18.1	7.22 × 101	FeL2
Droxidopa	16.2	2.96 × 104	FeL
Echinacoside	16.3	1.35 × 104	FeL2
Ellagic acid	16.3	1.35 × 104	FeL2
Entacapone (comtan, ASI-6)	19.3	3.99	FeL33−
Epicatechin	16.3	1.35 × 104	FeL2
Epigallocatechin-3-gallate	16.1	7.45 × 105	FeHL
Etidronate (HEDPA)	23.5	3.22 × 10−4	FeH−1L
Exifone	16.1	7.45 × 105	FeHL
Farrerol	16.1	7.55 × 108	FeL
Fiset (3,3′,4′,7-tetra-hydroxy-flavone)	16.3	1.35 × 104	FeL2
Fraxetin	16.3	1.35 × 104	FeL2−
Galangin	27.0	9.36 × 10−8	FeH−1L
Gallic acid derivatives	16.1	7.45 × 105	FeHL
Garcinol	16.3	1.35 × 104	FeL2
Genistein	16.1	7.55 × 108	FeL
Glutathione-hydroxy-quinoline compound	18	1.04 × 102	FeH−2L+
Glutathione-l-DOPA compound	16.3	1.35 × 104	FeL2
Gly-N-C-DOPA	16.2	2.96 × 104	FeL
Hesperidin	16.1	7.55 × 108	FeL
Hinokitiol	16.1	4.30 × 108	FeL2+
8-HQ-MC-5 (VK-28)	20.6	2.15 × 10−1	FeL3
4-Hydroxyisophthalic acid	16.1	1.64 × 106	FeL3
1-Hydroxy-2-pyridinone derivatives	17.7	1.50 × 102	FeL3
3-Hydroxy-4(1H)pyridinone (Deferiprone)	19.3	3.92	FeL3
3-Hydroxy-4(1H)pyridinone derivatives (R = H)	19.3	3.92	FeL3
8-Hydroxyquinoline	20.6	2.15 × 10−1	FeL3
8-Hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	20.6	2.15 × 10−1	FeL3
Hydroxy-quinoline-propargyl hybrids (HLA20)	20.6	2.15 × 10−1	FeL3
Hydroxytyrosol butyrate	16.3	1.35 × 104	FeL2−
Hyperoside	27.0	9.36 × 10−8	FeH−1L
Icariin	18.1	8.14 × 101	FeL2
Icariside II	18.1	8.14 × 101	FeL2
Isobavachalcone	16.1	3.23 × 108	FeL2+
Isochlorogenic acid	16.1	1.07 × 107	FeL
Isoquercetin (isoquercitrin)	27.0	9.36 × 10−8	FeH−1L
Kaempferol	27.0	9.36 × 10−8	FeH−1L
Kaempferol, 3-O-a-L arabino-furanoside-7-O-a-L-rhamno-pyranoside	18.1	8.14 × 101	FeL2
KR33493	16.3	1.06 × 104	FeL2
Kukoamine	16.3	1.35 × 104	FeL2
Luteolin	16.3	1.35 × 104	FeL2
M10M30 (VAR10303)M99	20.6	2.15 × 10−1	FeL3
Macranthoin G	16.3	1.35 × 104	FeL2
Magnesium lithospermate B	16.3	1.35 × 104	FeL2
α-mangostin	16.3	1.35 × 104	FeL2
γ-mangostin	16.3	1.35 × 104	FeL2
Metformin (Met)	16.1	1.36 × 109	FeL2+
MitoQ	16.1	5.72 × 1010	FeL2
Morin	18.1	8.14 × 101	FeL2
MyricetinMyricitrin	27.0	9.36 × 10−8	FeH−1L
Naringenin	16.1	7.55 × 108	FeL
Naringin	16.1	7.55 × 108	FeL
Nicotinamide adenine dinucleotide phosphate (NADPH)	16.1	6.01 × 1010	FeL2
Nitecapone	16.8	2.04 × 103	FeL2−
Nordihydroguaiaretic acid	16.3	1.35 × 104	FeL2
Oleuropein	16.3	1.35 × 104	FeL2−
Opicapone	16.8	2.04 × 103	FeL2
PBT2	20.6	2.15 × 10−1	FeL3
PBT434	16.1	3.22 × 109	FeL2+
Petunidin	16.3	1.35 × 104	FeL2
Phenylhydroxamates	16.1	9.46 × 104	FeL2
Piceatannol	16.3	1.35 × 104	FeL2
Pinostrobin (5-hydroxy-7-methoxy-flavone)	18.1	8.14 × 101	FeL2
Piperazine-8-OH-quinolone hybrid	20.6	2.15 × 10−1	FeL3
Protocatechuic acid	22.2	6.16 × 10−3	FeL23−
Protosappanin A	16.3	1.35 × 104	FeL2
Punicalangin	16.1	1.14 × 1012	FeL
Pyridoxal isonicotinoyl hydrazone (PIH)	22.9	1.07 × 10−3	FeL2
Pyridoxal isonicotinoyl hydrazone derivatives: PCIH PCTH H2NPH H2PPH	22.9	1.07 × 10−3	FeL2
Q1 Q4	20.6	2.15 × 10−1	FeL3
Quercetin	27.0	9.36 × 10−8	FeH–1L3−
Quinoline derivatives as SUMOylation activators	16.1	1.14 × 1016	FeL2+
Radotinib	21.5	3.01 × 10−2	FeH−2L2
Rimonabant	16.1	7.43 × 1021	FeL2
Rosmarinic acid	16.3	1.35 × 104	FeL2
Rutin	27.0	9.36 × 10−8	FeH−1L
Salicylate, sodium salt	16.1	1.64 × 106	FeL33−
Salvianolic acid B	16.3	1.35 × 104	FeL2
Silibinin (silybin) A, B	16.1	1.99 × 1019	FeH3L3+
Silydianin	16.1	1.99 × 1019	FeH3L3+
Sulfuretin	16.3	1.35 × 104	FeL2
Tanshinol	16.3	1.35 × 104	FeL2
Tannic acid	16.1	6.05 × 1049	Fe4L
Taxifolin	16.1	7.55 × 108	FeL
Tectorigenin	16.1	7.55 × 108	FeL
Tetracycline	16.1	8.35 × 1010	FeL22−
Tolcapone (ASI-7)	16.8	2.04 × 103	FeL2−
Transilitin	16.3	1.35 × 104	FeL2
o-Trensox	29.5	3.36 × 10−10	FeL3−
2,3,3-Trisphosphonate	18	1.06 × 102	FeL
Verbascoside	16.3	1.35 × 104	FeL2
Zonisamide	16.1	1.86 × 1011	FeL2

Table 6

pFe(II) and Kd values, and the most abundant Fe(II) complex, obtained at physiologically relevant conditions: pH = 7.4, cFe = 10−6 mol/L, and cL = 10−5 mol/L. See caption of Table 3 for other notes.

Compound Name(s)	pFe(II)	Kd (nmol/L)	Most Abundant Complex
7DH7MH	6.9	1.35 × 103	FeL
8A8B8C	6.9	1.35 × 103	FeL
8E8F	6.1	4.62 × 104	FeL2
ACPT-I	6	1.20 × 107	FeL
Alvespimycin	6	2.07 × 107	FeL2+
Aminothiazoles derivatives as SUMOylation activators	6	2.77 × 106	FeL
AMN082	6	2.07 × 107	FeL2+
Apomorphine	6	1.53 × 107	FeHL
l-Arginine	6	3.30 × 107	FeL2+
Ascorbic acid	6	1.89 × 109	FeL+
ASI-1	6	1.78 × 105	FeL+
ASI-5	10.7	1.95 × 10−1	FeL
Astilbin	6	1.53 × 107	FeHL
Azilsartan	6	2.53 × 1015	FeL2
Baicalein	9.9	1.09	FeL22−
Benserazide	9.9	1.09	FeL2
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine derivatives	6	2.07 × 107	FeL2+
2,2′-Bipyridyl	6.1	4.62 × 104	FeL22+
4-((5-bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	6	3.13 × 108	FeL
C-3 (α carboxyfullerene)	6	6.90 × 106	FeL
Caffeic acid amide analogues	6.7	2.02 × 103	FeH−1L
Carbazole-derived compounds	6	2.07 × 107	FeL2+
Carbidopa	6.3	1.07 × 104	FeL
Carnosic acid	6	1.53 × 107	FeHL
Catechin	6	1.53 × 107	FeHL
Ceftriaxone	6	1.20 × 107	FeL
Celastrol	6	1.53 × 107	FeHL
CEP-1347	6	1.53 × 107	FeHL
Chebulagic acid	6	5.66 × 105	FeL
Chlorogenic acid	6	2.93 × 106	FeHL
Clioquinol	7.9	1.01 × 102	FeL2
Clioquinol-selegiline hybrid	7.1	6.37 × 102	FeH2L2
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	6	1.53 × 107	FeHL
“Compound 1”	6	2.07 × 107	FeL2+
“Compound (−)-8a”	6	1.53 × 107	FeHL
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	6	1.20 × 107	FeL+
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine	6	2.07 × 107	FeL2+
Creatine	6	1.20 × 107	FeL+
Curcumin	6	4.16 × 109	FeH2L+
Cyanidin	6	1.53 × 107	FeHL
D512	6	2.07 × 107	FeL2+
D607 (bipyridyl-D2R/D3R agonist hybrid)	6.1	4.63 × 104	FeL2
DA-2 (8D)	6.9	1.35 × 103	FeL
DA-3	6	2.07 × 107	FeL
DA-4	6	2.07 × 107	FeL
Dabigatran etexilate	6	2.07 × 107	FeL
Delphinidin	9.9	1.09	FeL2
Demethoxycurcumin	6	1.78 × 105	FeL
Dendropanax morbifera	6	1.53 × 107	FeHL
Desferrioxamine (Deferoxamine, Desferal, DFO)	6.2	2.24 × 104	FeH2L+
4,5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid derivatives (R1, R2, R3, R4, or R5 = caffeoyl)	6	1.53 × 107	FeHL
Dihydromyricetin	9.9	1.09	FeL2
5-(3,4-Dihydroxybenzylidene)-2,2-dimethyl-1,3-dioxane-4,6-dione	6	1.53 × 107	FeHL+
7, 8-Dihydroxycoumarin derivative DHC12	6	1.53 × 107	FeHL
3′,4′-Dihydroxyflavone	14.8	1.58 × 10−5	FeL+
7,8-Dihydroxyflavone	6	1.53 × 107	FeHL+
(E)-3,4-Dihydroxystyryl aralkyl sulfones	6	1.53 × 107	FeHL+
(E)-3,4-Dihydroxystyryl aralkyl sulfoxides	6	1.53 × 107	FeHL+
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	8.2	5.40 × 101	FeL2
DKP	6	4.33 × 105	FeL
l-DOPA (levodopa, CVT-301)	6.3	1.07 × 104	FeL−
DOPA-derived peptido-mimetics (deprotected)	10.5	2.82 × 10−1	FeHL
DOPA-derived peptido-mimetics (protected)	6	1.53 × 107	FeHL
l-DOPA deuterated	6.3	1.07 × 104	FeL−
Doxycycline	6	4.07 × 108	FeL
Droxidopa	6.3	1.07 × 104	FeL−
Echinacoside	6	1.53 × 107	FeHL
Ellagic acid	6	1.53 × 107	FeHL
Entacapone (comtan, ASI-6)	12.7	1.65 × 10−3	FeL22−
Enzastaurin	6	2.07 × 107	FeL
Epicatechin	6	1.53 × 107	FeHL
Etidronate (HEDPA)	9.9	1.14	FeL2−
F13714F15599	6	2.77 × 106	FeL
FIsetin (3,3′,4′,7-tetra-hydroxy-flavone)	6	1.53 × 107	FeHL
Fraxetin	6	1.53 × 107	FeHL+
Gallocatechin	9.9	1.09	FeL2
Garcinol	6	1.53 × 107	FeHL
Glutamine	6	7.54 × 105	FeL+
Glutathione-hydroxy-quinoline compound	6.9	1.35 × 103	FeL+
Glutathione-l-DOPA compound	6	1.53 × 107	FeHL
Gly-N-C-DOPA	6.3	1.07 × 104	FeL−
Guanabenz	6	2.07 × 107	FeL
8-HQ-MC-5 (VK28)	6.9	1.35 × 103	FeL
4-Hydroxyisophthalic acid	6	3.13 × 108	FeL
8-hydroxyquinoline	6.9	1.35 × 103	FeL+
8-hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	6.9	1.35 × 103	FeL
Hydroxy-quinoline-propargyl hybrids (HLA20)	6.9	1.35 × 103	FeL
Hydroxytyrosol butyrate	6	1.53 × 107	FeHL+
Isochlorogenic acid	6	2.93 × 106	FeHL
KR33493	6	1.20 × 107	FeL
Kukoamine	6	1.53 × 107	FeHL
Lestaurtinib	6	2.07 × 107	FeL
M10M30 (VAR10303)M99	6.9	1.35 × 103	FeL
Macranthoin G	6	1.53 × 107	FeHL
Magnesium lithospermate B	6	1.53 × 107	FeHL
α-mangostin	6	1.53 × 107	FeHL
γ-mangostin	6	1.53 × 107	FeHL
MAOI-1	9.9	1.09	FeL2
MAOI-2	6	2.07 × 107	FeL
Meclofenamic acid	6	2.53 × 1015	FeL2
Mildronate	6	2.53 × 1015	FeL2
Mitomycin C	6	6.57 × 107	FeL
[18F]MPPF	6	2.07 × 107	FeL
Nitecapone	12.7	1.65 × 10−3	FeL22−
Nordihydroguaiaretic acid	6	1.53 × 107	FeHL
Oleuropein	6	1.53 × 107	FeHL+
Opicapone	12.7	1.65 × 10−3	FeL2
PBF-509	6	2.07 × 107	FeL
PBT2	6.9	1.35 × 103	FeL
PBT434	6.2	1.80 × 104	FeL+
Petunidin	6	1.53 × 107	FeHL
Phenothiazine 2Bc (n=0)	6	2.07 × 107	FeL2+
Phenylhydroxamates	6	5.96 × 107	FeL2
Piceatannol	6	1.53 × 107	FeHL
Piperazine-8-OH-quinolone hybrid	6.9	1.35 × 103	FeL
Preladenant	6	2.07 × 107	FeL
Promethazine	6	2.07 × 107	FeL2+
Protosappanin A	6	1.53 × 107	FeHL
Pyridoxal isonicotinoyl hydrazone (PIH)	7.1	6.37 × 102	FeH2L2
Pyridoxal isonicotinoyl hydrazone derivatives: PCIHPCTHH2NPHH2PPH	7.1	6.37 × 102	FeH2L22+
Pyrimidinone 8	6	2.07 × 107	FeL
Q1Q4	6.9	1.35 × 103	FeL
Radotinib	6.1	4.62 × 104	FeL2
Riboflavin	6	1.22 × 105	FeL2+
Rifampicin (ASI-3)	6	4.07 × 108	FeL
Rimonabant	6	4.85 × 1011	FeL
Rosmarinic acid	6	1.53 × 107	FeLH
Salicylate, sodium salt	6	3.13 × 108	FeL
Salvianolic acid B	6	1.53 × 107	FeHL
SCH58261SCH412348	6	2.07 × 107	FeL
ST1535 ST4206	6	2.07 × 107	FeL
Staurosporine	6	2.07 × 107	FeL
Sulfuretin	6	1.53 × 107	FeHL
Tanshinol	6	1.53 × 107	FeHL
Tetracycline	6	2.73 × 106	FeL
Tolcapone (ASI-7)	12.7	1.65 × 10−3	FeL22−
Transilitin	6	1.53 × 107	FeHL
2′,3′,4′-trihydroxyflavone	9.9	1.09	FeL22−
2,3,3-trisphosphonate	10.7	1.95 × 10−1	FeL
V81444	6	2.77 × 106	FeL
Verbascoside	6	2.02 × 106	FeL
WIN 55, 212-2	6	2.07 × 107	FeL2+

Table 7

pMn(II) and Kd values, and the most abundant Mn(II) complex, obtained at physiologically relevant conditions: pH = 7.4, cMn = 10−6 mol/L, and cL = 10−5 mol/L. See caption of Table 3 for other notes.

Compound Name(s)	pMn(II)	Kd (nmol/L)	Most Abundant Complex
7DH7MH	6.7	2.35 × 103	MnL+
8A8B8C	6.7	2.35 × 103	MnL
8E8F	6	2.40 × 106	MnL
N-Acetyl cysteine	6	8.97 × 105	MnHL+
ACPT-I	6	2.25 × 108	MnL
Alvespimycin	6	7.61 × 108	MnL2+
Ambroxol	6.7	2.47 × 103	MnL+
3-(7-amino-5-(cyclohexylamino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide	6	7.61 × 108	MnL
Aminothiazoles derivatives as SUMOylation activators	6	7.64 × 107	MnL
AMN082	6	7.61 × 108	MnL2+
Apomorphine	6	7.16 × 108	MnL
l-Arginine	6	1.45 × 108	MnL2+
ASI-1	6	3.44 × 106	MnL+
ASI-5	6	7.81 × 106	MnL
Astilbin	6	7.16 × 108	MnL
Azilsartan	6	3.71 × 109	MnL
Baicalein	6	2.62 × 107	MnL
Benserazide	6	2.62 × 107	MnL
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine derivatives	6	7.61 × 108	MnL2+
2,2′-bipyridyl	6	2.40 × 106	MnL2+
4-((5-bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	6	8.72 × 108	MnL
C-3 (α carboxyfullerene)	6	5.06 × 106	MnL
Caffeic acid amide analogues	6	6.62 × 107	MnH−1L
Carbazole-derived compounds	6	7.61 × 108	MnL2+
Carbidopa	7.6	2.33 × 102	MnHL
Carnosic acid	6	7.16 × 108	MnL
Catechin	6	7.16 × 108	MnL
Ceftriaxone	6	2.25 × 108	MnL
Celastrol	6	7.16 × 108	MnL
CEP1347	6	3.71 × 109	MnL
Chebulagic acid	6	2.62 × 107	MnL
Chlorogenic acid	6	3.91 × 107	MnL−
Clioquinol	6.7	2.35 × 103	MnL+
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	6	7.16 × 108	MnL
“Compound 1”	6	7.61 × 108	MnL++
“Compound (−)-8a”	6	7.16 × 108	MnL
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	6	2.25 × 108	MnL+
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine	6	7.61 × 108	MnL++
Creatine	6	2.25 × 108	MnL+
Curcumin	6	3.44 × 106	MnL
Cyanidin	6	7.16 × 108	MnL
D512	6	7.61 × 108	MnL2+
D607 (bipyridyl-D2R/D3R agonist hybrid)	6	2.40 × 106	MnL
DA-2 (8D)	6.7	2.35 × 103	MnL
DA-3	6	7.61 × 108	MnL
DA-4	6	7.61 × 108	MnL
Dabigatran etexilate	6	7.61 × 108	MnL
(S)-3,4-DCPG	6	5.91 × 106	MnL
Delphinidin	6	2.62 × 107	MnL
Demethoxycurcumin	6	3.44 × 106	MnL
Dendropanax morbifera active compound	6	7.16 × 108	MnL
4,5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid derivatives (R1, R2, R3, R4, or R5 = caffeoyl)	6	7.16 × 108	MnL
Dihydromyricetin	6	2.62 × 107	MnL
5-(3,4-Dihydroxybenzylidene) -2,2-dimethyl-1,3-dioxane-4,6-dione	6	7.16 × 108	MnL
7,8-Dihydroxycoumarin derivative DHC12	6	7.16 × 108	MnL
3′,4′-Dihydroxyflavone	6	7.16 × 108	MnL
7,8-dihydroxyflavone	6	7.16 × 108	MnL
(E)-3,4-Dihydroxystyryl aralkyl sulfones	6	7.16 × 108	MnL
(E)-3,4-Dihydroxystyryl aralkyl sulfoxides	6	7.16 × 108	MnL
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	6	2.58 × 106	MnL
DKP	6	1.00 × 106	MnHL
l-DOPA (levodopa, CVT-301)	7.6	2.26 × 102	MnHL
DOPA-derived peptido-mimetics (deprotected)	7.6	2.33 × 102	MnHL
DOPA-derived peptido-mimetics (protected)	6	7.16 × 108	MnL
l-dopa deuterated	7.6	2.26 × 102	MnHL
Droxidopa	7.6	2.33 × 102	MnHL
Echinacoside	6	7.16 × 108	MnL
Ellagic acid	6	7.16 × 108	MnL
Entacapone (comtan, ASI-6)	6	6.74 × 105	MnL
Enzastaurin	6	7.61 × 108	MnL
Epicatechin	6	7.16 × 108	MnL
Etidronate (HEDPA)	6	1.03 × 106	MnL2−
F13714F15599	6	7.64 × 107	MnL
Fisetin (3,3′,4′,7-tetra-hydroxy-flavone)	6	7.16 × 108	MnL
Fraxetin	6	7.16 × 108	MnL
Gallocatechin	6	2.62 × 107	MnL
Garcinol	6	7.16 × 108	MnL
Glutamine	6	2.30 × 107	MnL+
Glutathione derivatives	6	4.07 × 105	MnL−
Glutathione-hydroxy-quinoline compound	6.7	2.35 × 103	MnL+
Glutathione-l-DOPA compound	6	7.16 × 108	MnL
Gly-N-C-DOPA	7.6	2.33 × 102	MnHL
Guanabenz	6	7.61 × 108	MnL
Hinokitiol	6.1	2.99 × 104	MnL+
8-HQ-MC-5 (VK-28)	6.7	2.35 × 103	MnL
4-Hydroxyisophthalic acid	6	8.72 × 108	MnL
8-hydroxyquinoline	6.7	2.35 × 103	MnL+
8-Hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	6.7	2.35 × 103	MnL
Hydroxy-quinoline-propargyl hybrids (HLA20)	6.7	2.35 × 103	MnL
Hydroxytyrosol butyrate	6	7.16 × 108	MnL
Isobavachalcone	6.2	1.55 × 104	MnL+
Isochlorogenic acid	6	3.91 × 107	MnL−
KR33493	6	2.25 × 108	MnL
Kukoamine	6	7.16 × 108	MnL
Lestaurtinib	6	7.61 × 108	MnL
Lipoic acid	6	9.53 × 106	MnL+
Luteolin	6	7.16 × 108	MnL
M10M30 (VAR10303)M99	6.7	2.35 × 103	MnL
Macranthoin G	6	7.16 × 108	MnL
Magnesium lithospermate B	6	7.16 × 108	MnL
α-mangostin	6	7.16 × 108	MnL
γ-mangostin	6	7.16 × 108	MnL
MAOI-1	6	2.62 × 107	MnL
MAOI-2	6	7.61 × 108	MnL
Meclofenamic acid	6	3.71 × 109	MnL
Mildronate	6	3.71 × 109	MnL
Mitomycin C	6	7.46 × 107	MnL
MitoQ	6	1.29 × 105	MnL
[18F]MPPF	6	7.61 × 108	MnL
Nicotinamide adenine dinucleotide phosphate (NADPH)	6	2.01 × 107	MnL
Nicotinamide mononucleotide	6	6.59 × 106	MnL
Nitecapone	6	6.74 × 105	MnL
Nordihydroguaiaretic acid	6	7.16 × 108	MnL
Oleuropein	6	7.16 × 108	MnL
Opicapone	6	6.74 × 105	MnL
PBF-509	6	2.41 × 109	MnL
PBT2	6	5.22 × 105	MnL+
Petunidin	6	7.16 × 108	MnL
Phenothiazine 2Bc (n=0)	6	7.61 × 108	MnL2+
Phenylhydroxamates	6	5.81 × 106	MnL
Piceatannol	6	7.16 × 108	MnL
Piperazine-8-OH-quinolone hybrid	6.7	2.35 × 103	MnL
Preladenant	6	7.61 × 108	MnL
Promethazine	6	7.61 × 108	MnL2+
Protocatechuic acid	6	3.66 × 108	MnL−
Protosappanin A	6	7.16 × 108	MnL
Pyrimidinone 8	6	7.61 × 108	MnL
Q1Q4	6.7	2.35 × 103	MnL
Radotinib	6	2.40 × 106	MnL
Riboflavin	6	5.75 × 105	MnHL3+
Rifampicin (ASI-3)	6	1.03 × 106	MnL
Rimonabant	6	1.00 × 106	MnHL
Rosmarinic acid	6	7.16 × 108	MnL
Salicylate, sodium salt	6	8.72 × 108	MnL
Salvianolic acid B	6	7.16 × 108	MnL
SCH58261SCH412348	6	7.61 × 108	MnL
ST1535ST4206	6	7.61 × 108	MnL
Staurosporine	6	7.61 × 108	MnL
Sulfuretin	6	7.16 × 108	MnL
Tanshinol	6	7.16 × 108	MnL
Taurine	6	6.39 × 1011	MnL2
Tetracycline	6	2.14 × 107	MnL
Tolcapone (ASI-7)	6	6.74 × 105	MnL
Transilitin	6	7.16 × 108	MnL
2′,3′,4′-Trihydroxyflavone	6	2.62 × 107	MnL
V81444	6	7.64 × 107	MnL
Verbascoside	6	6.62 × 107	MnH−1L
WIN 55,212-2	6	7.61 × 108	MnL2+

Table 8

pZn(II) and Kd values, and the most abundant Zn(II) complex, obtained at physiologically relevant conditions: pH = 7.4, cZn = 10−6 mol/L, and cL = 10−5 mol/L. See caption of Table 3 for other notes.

Compound Name(s)	pZn(II)	Kd (nmol/L)	Most Abundant Complex
7DH7MH	7.5	3.16 × 102	ZnL
8A8B8C	7.5	3.16 × 102	ZnL
8E8F	6.4	6.31 × 103	ZnL
N-acetyl cystein	6	1.66 × 106	ZnL
ACPT-I	6	1.46 × 107	ZnL
Alaternin	6.8	1.83 × 103	ZnL
Alvespimycin	6	7.85 × 105	ZnL2+
AM-251	6	1.95 × 106	ZnHL
Ambroxol	7.7	1.90 × 102	ZnL2
3-(7-amino-5-(cyclohexylamino)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-2-yl)-2-cyanoacrylamide	6	7.85 × 105	ZnL
Aminothiazoles derivatives as SUMOylation activators	6	2.28 × 105	ZnL
AMN082	6	7.85 × 105	ZnL2+
Antagonist of the A(2A) adenosine receptor - derivative 49	6	1.95 × 106	ZnHL
Apigenin	6	1.23 × 1029	ZnH3L
Apomorphine	6	7.76 × 106	ZnL
l-Arginine	6	4.14 × 106	ZnL2+
Ascorbic acid	6.1	4.67 × 104	ZnL+
ASI-1	6	5.76 × 105	ZnL+
ASI-5	6	9.06 × 106	ZnL
Astilbin	6	7.76 × 106	ZnL
Azilsartan	6	5.22 × 105	ZnL
Baicalein	6	2.71 × 105	ZnL
Benserazide	6	2.71 × 105	ZnL
7H-Benzo[e] perimidin-7-one derivatives (R6 = OH)	6.1	4.82 × 104	ZnL2
4H-1-benzopyran-4-one	6	3.56 × 1020	ZnH3L
8-Benzyl-tetrahydropyrazino[2,1-f]purinedione (derivative n. 57)	6	9.67 × 1011	ZnL
Bikaverin	6.3	1.07 × 104	ZnH−2L2−
(−)-N6-(2-(4-(Biphenyl-4-yl)piperazin-1-yl)-ethyl)-N6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine derivatives	6	7.85 × 105	ZnL2+
2,2′-bipyridyl	6.4	6.31 × 103	ZnL2+
4-((5-Bromo-3-chloro-2-hydroxybenzyl) amino)-2-hydroxybenzoic acid (LX007, ZL006)	6	8.04 × 108	ZnH−1L
C-3 (α carboxyfullerene)	6	2.08 × 106	ZnL
Caffeic acid amide analogues	6	9.29 × 105	ZnH−1L
Carbazole-derived compounds	6	7.85 × 105	ZnL2+
Carbidopa	6	2.96 × 106	ZnHL
Carnosic acid	6	7.76 × 106	ZnL
Cathechin	6	7.76 × 106	ZnL
Ceftriaxone	6.1	3.68 × 104	ZnL
Celastrol	6	7.76 × 106	ZnL
Chebulagic acid	6	1.81 × 1013	Zn2L
Chlorogenic acid	6	6.86 × 105	ZnL−
3′-O-(3-Chloropivaloyl) quercetin	6	5.69 × 1016	ZnH3L
Chlorpromazine	6	1.95 × 106	ZnHL3+
Chrysin	6	5.69 × 1016	ZnH3L
Clioquinol	7.5	3.16 × 102	ZnL+
Clovamide analogues (R1 and R2 = OH, and/or R3 and R4 = OH)	6	7.76 × 106	ZnL
“Compound 1”	6	7.85 × 105	ZnL2+
“Compound (−)-8a”	6	7.76 × 106	ZnL
“Compound 8”	6.4	6.47 × 103	ZnH−1L
“Compound 21”, derivative of 3-methyl-1-(2,4,6-trihydroxyphenyl) butan-1-one	6	1.54 × 106	ZnL+
“Compound (−)-21a”, derivative of N-6-(2-(4-(1H-indol-5-yl)piperazin-1-yl)ethyl)-N-6-propyl-4,5,6,7-tetrahydrobenzo[d]thiazole-2,6-diamine	6	7.85 × 105	ZnL2+
Creatine	6	1.54 × 106	ZnL+
Cudraflavone B	6	5.69 × 1016	ZnH3L
Curcumin	6	5.76 × 105	ZnL
Cyanidin	6	7.76 × 106	ZnL
D512	6	7.85 × 105	ZnL2+
D607 (bipyridyl-D2R/D3R agonist hybrid)	6.4	6.31 × 103	ZnL
DA-2 (8D)	7.5	2.80 × 102	ZnL
DA-3	6	7.85 × 105	ZnL
DA-4	6	7.85 × 105	ZnL
Dabigatran etexilate	6	7.85 × 105	ZnL
Dabrafenib	6	8.23 × 107	ZnL
(S)-3,4-DCPG	6	6.84 × 106	ZnL
Deferricoprogen	8.3	4.35 × 101	ZnHL
Delphinidin	6	2.71 × 105	ZnL
Demethoxycurcumin	6	5.76 × 105	ZnL
Dendropanax morbifera active compound	6	7.76 × 106	ZnL
Desferrioxamine (Deferoxamine, Desferal, DFO)	7.4	3.97 × 102	ZnH2L+
(S)-N-(3-(3,6-Dibromo-9H-carbazol-9-yl)-2-fluoropropyl)-6-methoxypyridin-2-amine	6	1.95 × 106	ZnHL
4,5-O-Dicaffeoyl-1-O-(malic acid methyl ester)-quinic acid derivatives (R1, R2, R3, R4, or R5 = caffeoyl)	6	7.76 × 106	ZnL
Dihydromyricetin	6	2.71 × 105	ZnL
5-(3,4-Dihydroxybenzylidene)-2,2-dimethyl-1,3-dioxane-4,6-dione	6	7.76 × 106	ZnL
7,8-Dihydroxycoumarin derivative DHC12	6	7.76 × 106	ZnL
3′,4′-Dihydroxyflavone	6	7.76 × 106	ZnL
7,8-Dihydroxyflavone	6	7.76 × 106	ZnL
(E)-3,4-Dihydroxystyryl aralkyl sulfones	6	7.76 × 106	ZnL
(E)-3,4-Dihydroxystyryl aralkyl sulfoxides	6	7.76 × 106	ZnL
2-[[(1,1-Dimethylethyl) oxidoimino]-methyl]-3,5,6-trimethylpyrazine	6	2.20 × 106	ZnL
DKP	6	1.52 × 106	ZnL
l-DOPA (levodopa, CVT-301)	6	2.96 × 106	ZnHL
DOPA-derived peptido-mimetics (deprotected)	6	2.96 × 106	ZnHL
DOPA-derived peptido-mimetics (protected)	6	7.76 × 106	ZnL
l-dopa deuterated	6	2.96 × 106	ZnHL
Doxycycline	6.1	3.88 × 104	ZnL
Droxidopa	10.9	1.18 × 10−1	ZnHL
Echinacoside	6	7.76 × 106	ZnL
Ellagic acid	6	7.76 × 106	ZnL
Entacapone (comtan, ASI-6)	6.2	2.62 × 104	ZnL
Enzastaurin	6	7.85 × 105	ZnL
Epicatechin	6	7.76 × 106	ZnL
Epigallocatechin-3-gallate	6	1.81 × 1013	Zn2L
Etidronate (HEDPA)	7.4	4.32 × 102	ZnL2−
Exifone	6	1.81 × 1013	Zn2L
F13714,F15599	6	2.28 × 105	ZnL
Fisetin (3,3′,4′,7-tetra-hydroxy-flavone)	6	7.76 × 106	ZnL
Fraxetin	6	7.76 × 106	ZnL
Gallic acid derivatives	6	1.81 × 1013	Zn2L
Gallocatechin	6	2.71 × 105	ZnL
Garcinol	6	7.76 × 106	ZnL
Glutamine	6	8.61 × 105	ZnL+
Glutathione derivatives	14.8	1.20 × 10−5	ZnH−2L22−
Glutathione-hydroxy-quinoline compound	7.8	1.37 × 102	ZnH−1L+
Glutathione-l-DOPA compound	6.3	1.20 × 104	ZnH−1L
Gly-N-C-DOPA	6	2.96 × 106	ZnHL
GSK2795039	9.7	1.64	ZnL2
Guanabenz	6	7.85 × 105	ZnL
Hinokitiol	6.2	2.06 × 104	ZnL+
8-HQ-MC-5 (VK-28)	7.5	3.16 × 102	ZnL
4-Hydroxyisophthalic acid	6	9.02 × 107	ZnL
1-Hydroxy-2-pyridinone derivatives	6.3	1.01 × 104	ZnL
3-Hydroxy-4(1H)pyridinone (Deferiprone)	6.2	1.45 × 104	ZnL+
3-Hydroxy-4(1H)pyridinone derivatives (R = H)	6.2	1.45 × 104	ZnL
8-hydroxyquinoline-2-carboxaldehyde isonicotinoyl hydrazone	7.5	3.16 × 102	ZnL
Hydroxy-quinoline-propargyl hybrids (HLA20)	7.5	3.16 × 102	ZnL
Hydroxytyrosol butyrate	6	7.76 × 106	ZnL
l-(7-Imino-3-propyl-2,3-dihydrothiazolo [4,5-d]pyrimidin-6(7H)-yl)urea	6.5	5.10 × 103	ZnH−1L
Imipramine	6	1.95 × 106	ZnHL3+
Isobavachalcone	6	2.12 × 105	ZnL+
Isochlorogenic acid	6	6.86 × 105	ZnL−
Isoquercetin (isoquercitrin)	6	3.94 × 1024	ZnH4L
Kaempferol	6	3.94 × 1024	ZnH4L+
Kaempferol, 3-O-a-L arabino-furanoside-7-O-a-L-rhamno-pyranoside	6	3.56 × 1020	ZnH3L
KR33493	6	1.54 × 106	ZnL
Kukoamine	6	7.76 × 106	ZnL
Lestaurtinib	6	7.85 × 105	ZnL
Lipoic acid	6	3.84 × 106	ZnL+
LY354740	6	1.46 × 107	ZnL
M10M30 (VAR10303)M99	7.5	3.16 × 102	ZnL
Macranthoin G	6	7.76 × 106	ZnL
Magnesium lithospermate B	6	7.76 × 106	ZnL
α-mangostin	6	7.76 × 106	ZnL
γ- mangostin	6	7.76 × 106	ZnL
MAOI-1	6	2.71 × 105	ZnL
MAOI-2	6	7.85 × 105	ZnL
MAOI-4	6	5.88 × 105	ZnHL
Metformin (Met)	6	1.87 × 106	ZnL+
Methoxy-6-acetyl-7-methylijuglone	6.1	4.82 × 104	ZnL2
N′-(4-methylbenzylidene)-5-phenylisoxazole-3-carbohydrazide	6	1.76 × 106	ZnL
Minocycline	6.4	7.52 × 103	ZnHL
Mitomycin C	6	3.49 × 106	ZnL
MitoQ	7.1	8.74 × 102	ZnL
Morin	6	5.69 × 1016	ZnH3L
[18F]MPPF	6	7.85 × 105	ZnL
MSX-3	6	8.19 × 106	ZnL
Nicotinamide adenine dinucleotide phosphate (NADPH)	6.1	3.72 × 104	ZnL
Nicotinamide mononucleotide	6	8.19 × 106	ZnL
Nitecapone	6.2	2.62 × 104	ZnL
Nordihydroguaiaretic acid	6	7.76 × 106	ZnL
Oleuropein	6	7.76 × 106	ZnL
Opicapone	6.2	2.62 × 104	ZnL
P7C3	6	1.96 × 107	ZnL2+
PBF-509	6	7.85 × 105	ZnL
PBT2	7.5	3.16 × 102	ZnL
PBT434	7.9	1.19 × 102	ZnL2
Petunidin	6	7.76 × 106	ZnL
Phenothiazine 2Bc (n=0)	6	7.85 × 105	ZnL2+
Phenothiazine 2Bc (n=1)	6	1.95 × 106	ZnHL3+
Phenylhydroxamates	6	2.07 × 105	ZnL
Piceatannol	6	7.76 × 106	ZnL
Piperazine-8-OH-quinolone hybrid	7.5	3.16 × 102	ZnL
Preladenant	6	7.85 × 105	ZnL
Promethazine	6	7.85 × 105	ZnL2+
Protocatechuic acid	6	1.25 × 107	ZnL−
Protosappanin A	6	7.76 × 106	ZnL
Pyrazolobenzothiazine-based carbothioamides	6	4.73 × 107	ZnL
Pyrimidinone 8	6	7.85 × 105	ZnL
Q1Q4	7.5	3.16 × 102	ZnL
Quercetin	6	3.94 × 1024	ZnH4L+
Quinoline derivatives as SUMOylation activators	6	3.94 × 106	ZnL2+
Radotinib	6.4	6.31 × 103	ZnL
Riboflavin	6	2.16 × 105	ZnHL3+
Rifampicin (ASI-3)	6.1	7.04 × 104	ZnL
Rimonabant	6	1.52 × 106	ZnL
Rosmarinic acid	6	7.76 × 106	ZnL
Rutin	6	3.94 × 1024	ZnH4L+
Salicylate, sodium salt	6	9.02 × 107	ZnL
Salvianolic acid B	6	7.76 × 106	ZnL
SCH58261 SCH412348	6	7.85 × 105	ZnL
ST1535 ST4206	6	7.85 × 105	ZnL
Staurosporine	6	7.85 × 105	ZnL
Stemazole	6	4.73 × 107	ZnL
Sulfuretin	6	7.76 × 106	ZnL
Tannic acid	6	1.22 × 105	ZnL
Tanshinol	6	7.76 × 106	ZnL
Taurine	6	1.26 × 1012	ZnL2
Tetracycline	6	7.01 × 106	ZnL
Tolcapone (ASI-7)	6.2	2.62 × 104	ZnL
Tozadenant	8.7	1.62 × 101	ZnL2
Transilitin	6.1	8.21 × 104	ZnL
o-Trensox	21.7	1.95 × 10−12	ZnL4−
2′, 3′, 4′-Trihydroxyflavone	6	2.71 × 105	ZnL
2,3,3-Trisphosphonate	12.1	7.93 × 10−3	ZnL
V81444	6	2.28 × 105	ZnL
VAS3947 VAS2870	6.5	5.10 × 103	ZnH−1L
Verbascoside	6	9.29 × 105	ZnH−1L
WIN 55,212-2	6	7.85 × 105	ZnL2+
WR-1065	6	1.95 × 106	ZnHL3+
Zonisamide	8	8.81 × 101	ZnL

6. Possible Usages of Speciation Data for Metal Chelation Therapy against Parkinson’s Disease

Speciation calculations allow to predict which metal and ligand species exist in solution at a given pH and metal and ligand total concentrations. A speciation model can therefore be obtained: this was done for the calculations performed in Table 3, Table 4, Table 5, Table 6, Table 7 and Table 8, where only some information has been given.

As anticipated, values of pM and Kd are useful to compare the relative strength of the complexes formed by different ligands with the same metal ion, or, conversely, by different metal ions with the same ligand. For example, the complexes formed by Fe(III) with 3-hydroxy-4(1H)-pyridinone (Deferiprone) have larger pM and lower Kd values than those formed with Luteolin, thus Fe(III)-Deferiprone complexes are stronger than Fe(III)-Luteolin ones. The values of pFe(III) for Deferiprone (19.3) or for Desferrioxamine (26.8) are often considered as milestones when new CAs are proposed for the chelation therapy of Fe overload [314]: new compounds are considered to be effective enough if their pFe(III) is larger than that of Deferiprone or of Desferrioxamine.

More importantly, Kd values can be compared with those reported in Table 1, allowing the assessment of whether a given ligand is able to remove a metal ion from α-syn: removal can occur if the Kd value is lower than that of α-syn. This approach has been proposed for MCT in Alzheimer’s disease, where Kd values of the CA + metal ion complexes were recommended to be 10–100 times lower than Kd values of the amyloid β protein + metal ion ones [316]. If the same approach is adopted for PD, it can be, e.g., deduced that Desferrioxamine (Kd = 1.81 × 10−7) is able to remove Fe(III) from α-syn (Kd = 10−4), and L-Dopa (Kd = 7.95 × 10−6) is able to remove Cu(II) from α-syn (Kd = 102). Many other ligands, including L-Dopa (Kd = 2.96 × 104), cannot remove Fe(III) from α-syn. However, it is necessary to underline that the Kd values reported in Table 1 are not completely reliable (the same also applies for Kd values of amyloid β protein + metal ion [12]), for the reasons stated above, so this approach does not (still) allow drawing definite conclusions about the removal of the relevant metal ion from proteins. On the other hand, the control of metal ion dyshomeostasis in PD requires that the CAs do not form too strong metal complexes to avoid metal anemia and allow metal redeployment to other compartments, according to the conservative chelation strategy. Too high or too low Kd values are thus not suitable. Unfortunately, the limiting Kd values which an ideal CA should possess to be employed for the PD therapy are not known.

The information regarding the most abundant complex existing at physiological conditions can be useful for two reasons. The identity of the existing complexes (and in particular of the most abundant one) and their charge are crucial in determining their redistribution once the target metal ion has been complexed by the CA. For example, a charged complex is expected to be hydrophilic, thus being unable to pass cellular barriers and preferring to be solubilised in aqueous solutions (e.g., in the blood), whereas neutral species should behave in an opposite way. The structure of the complexes, which might be deduced from the stoichiometry, also has a main role in determining their properties and toxicity [321].

As regards Fe and Cu, it is necessary to consider that they can undergo redox reactions even when complexed by a ligand. These reactions might be as harmful as or even more dangerous than those caused by the target metal ion at pathological in vivo conditions. The redox-induced toxicity of Fe and Cu complexes formed by several ligands is well known, so it has been used to develop new anti-cancer drugs [322], but it appears to have been generally overlooked when MCT is employed for PD. The redox activity of Fe and Cu complexes depends on the relative stability of the complexes formed by the ions at the two oxidation states. As regards Fe, if a ligand L forms the complexes FeIIIL and FeIIL with Fe(III) and Fe(II), respectively, a redox half-reaction can occur:Fe

By means of simple substitutions in the Nernst equation of Fe(III)/Fe(II), the standard reduction potential of (8) can be computed: where E0Fe(III)/Fe(II) is the standard reduction potential for free Fe (0.771 V), and βFeL and βFeL are the cumulative stability constants of FeIIIL and FeIIL, respectively. Values of E0 of any Fe or Cu complexes can be derived in a similar way if metal–ligand speciation at both oxidation states is known. Alternatively, electrochemical values can experimentally be obtained from voltammetric measurements (see, e.g., [323]). The standard reduction potentials of Fe complexes might have an important role in determining whether they can undergo harmful redox cycling in vivo, as extensively descrived by Merkofer et al. [324] and recently reviewed by Koppenol and Hider [325]: in general, it appears that negative E0 values can guarantee the absence of such toxic phenomena. However, further work is necessary to evaluate limiting Fe and Cu E0 values under which no redox damage occurs in PD brains.

Other possible information which can be gathered from speciation calculations, even if focused on the bloodstream, has recently been reviewed by Kiss et al. [326,327].

7. Concluding Remarks

The development of drugs able to target several pathological pathways appears to be the best approach for PD therapy and of other important NDs such as Alzheimer’s disease and Amyotrophic Lateral Sclerosis. Compounds which form complexes with the PD relevant metal ions, i.e., Cu(II), Cu(I), Fe(III), Fe(II), Mn(II) and Zn(II), are aimed to target metal dyshomeostasis. For these CAs and for these metal ions, the knowledge of metal–ligand speciation is of primary importance to predict the efficacy of the CA, its ability to remove dysregulated metal ions from toxic storages such as the α-syn complex and redeploy metal ions to safe stores (conservative chelation), the possible toxic effects induced by the metal complexes formed in PD brain, and in general to be able to model the distribution of the metal–ligand species in vivo. Still much work has to be performed to define the upper and the lower limiting metal Kd values required by a CA to disrupt the α-syn complex without causing excessive metal removal, as well as the suitable standard reduction potentials required by the complexes to avoid harmful redox cycling in the brain. Also, available speciation information is in part lacking, especially as regards Cu(I), for which very few stability constants values have been hitherto determined (possible strategies for effective studies of Cu(I) speciations have been proposed [328]), but also for other metal ions and for several complicated or recently proposed CAs. If a complete metal–ligand speciation study (aimed to determine stability constants) cannot be performed, a Kd value should at least be determined. This amount represents a key number which can be used to compare simple metal–ligand complexes, for which the full speciation picture is available, with complicated ones like those involving α-syn, for which this information cannot be obtained. However, standardized experimental procedures are recommended to allow Kd values to be more rigorously and reliably compared with each other.