Interaction between Cu and Thiols of Biological and Environmental Importance: Case Study Using Combined Spectrophotometric/Bathocuproine Sulfonate Disodium Salt Hydrate (BCS) Assay

Considering the biological and ecological importance of Cu–thiol interactions and the discrepancies in previous research, this study focuses on Cu interactions with biologically and ecologically relevant thiols: glutathione (GSH), L-cysteine (L-cys), 3-mercaptopropionic acid (MPA), and thioacetic acid (TAA) in aqueous solution. The addition of Cu(II) to a thiol-containing solution led to a rapid reduction of Cu(II) and the formation of a Cu(I)–thiol complex. The mechanism of Cu(II) reduction and Cu(I) complex formation as well as the kinetics of Cu(I) oxidation strongly depend on the structural properties of the individual thiols investigated. The reducing power of the investigated thiols can be summarized as follows: L-cys ≅ GSH > MPA > TAA. The reaction order, with respect to Cu(I) oxidation, also changes over the time of the reaction course. The deviation of the reaction kinetics from the first order with respect to Cu(I) in the later stages of the reaction course can be attributed to a Fenton-like reaction occurring under low thiol concentration conditions. At high Cu:thiol ratios, in the case of GSH, L-cys, and MPA, the early stage of the reaction course is characterized by high Cu(I) stability, most likely as a result of Cu(I) complexation by the thiols present in excess in the reaction mixture.


Introduction
As one of the most abundant trace elements in living organisms, copper (Cu) is an essential metal in the biochemistry and physiology of all organisms [1]. Because of its ability to undergo redox changes between the oxidation states of Cu(I) and Cu(II), Cu is an important catalytic cofactor in redox reactions involved in electron transfer [2,3]. Thiol compounds are known to interact strongly with Cu and can inhibit Cu-containing lysyl oxidase and tyrosinase enzymes [4][5][6][7]. Furthermore, biological thiols containing Cu in the active sites are an example of substrates used by copper-zinc superoxide dismutase (Cu/Zn-SOD), which catalyzes the dismutation of superoxide radicals to oxygen and hydrogen peroxide [8,9]. However, Cu(II) reduced to Cu(I) by hydrogen peroxide, superoxide, or thiol compounds can catalyze the formation of reactive oxygen species (ROS) that can lead to cell damage, with Cu acting as a pro-oxidant in this scenario [10,11]. Winterbourn et al. reported that thiols, especially aminothiols, such as cysteine, enhance the reduction of the Cu active site of CuZn-SOD and lead to the formation of hydrogen peroxide and disulfide, though CuZn-SOD is less reactive than free Cu(II) [12]. Many authors attribute Cu toxicity to the oxidative damage caused by free radicals, which leads to DNA damage, increased lipid peroxidation, or decreased enzymatic activity [13][14][15][16]. Murakami et al. showed that Cu ions inhibit GSH reductase activity, with Cu(I) ions acting more strongly than Cu(II) ions, which is consistent with previous studies by Xiao et al. [17,18]. Considering Cu toxicity, it is but the exact form of GSH in water and the mechanisms of Cu-GSH complex formation are still under investigation [39]. It has been suggested that GSH measured in water may be in its oxidized form, i.e., glutathione disulfide (GSSG), or in the form of complexes with trace metals [39]. In addition to GSH, other LMW thiols, such as cysteine, have also been detected in lakes, estuaries, and marine waters and have been proposed as Cu(I)-binding ligands, though the exact mechanisms of Cu(I) complexation with thiols are still under investigation [41,42]. L-cysteine is a non-essential amino acid that plays an important role in protein synthesis and stability through the formation of disulfide bonds [43]. The reactivity of L-cys is due to the sulfhydryl group (-SH), and L-cys serves as a ligand for the binding of trace metals, including Cu, and for active sites in enzymes. Deprotonation of the sulfhydryl group produces a negatively charged and reactive thiolate anion that can participate in oxidation or alkylation by electrophiles [44]. The thiolate anion is known to form mononuclear complexes with Cu(I), including mononuclear mixed thiolate-nitrogenbonded species as well as Cu(I)-thiolate clusters [45]. In addition to glutathione and L-cysteine, 3-mercaptopropionic acid (MPA) is widely distributed in freshwater and marine environments, where it may be of biological and abiotic origins [38]. Biologically produced MPA has been addressed as a metabolite product of other thiol species, but it is also found in the methanogens M. Jannaschii, where its presence is associated with the coenzyme M rather than the metabolism of other thiols [46,47]. Abiotically generated MPA may originate from the abiotic transformation of sulfur species, such as hydrogen sulfide (H 2 S) [46,47]. Although interaction with Cu ions has been reported in marine and freshwater systems, it has not been thoroughly investigated because of the rapid oxidation of MPA in natural environments and the presence of other possible MPA-binding trace metals, such as iron, Fe [38]. Finally, thioacetic acid (TAA) is a thiocarboxylic acid that has been implicated in the origin of life as a potential acetyl-CoA analogue in prebiotic chemistry [48][49][50]. The oxidation of TAA and ethanethiol (ET) by Fe(III) has already been studied, and the mechanism by which the Fe(III) is reduced to Fe(II) while oxidizing thioacetate to diacetyl disulfide, leading to the formation of thioester and iron sulfide (FeS), has been proposed [50]. However, similar interactions of Cu with TAA are not known so far.
In this manuscript we present a kinetic study of Cu(II) reduction with thiols of biological and ecological importance (GSH, L-cys, MPA, and TAA). The kinetics of Cu(II) reduction and the Cu(I) oxidation formed are evaluated under different experimental conditions, i.e., different Cu and thiol concentration ratios. Thiols such as GSH are present in excess in many prokaryotic and mammalian cells, as well as in the aquatic environment; therefore, we decided to investigate Cu(II) reduction and Cu(I) oxidation with excess thiol in relation to Cu [32][33][34][38][39][40]. However, previous research on the interactions of Cu with thiols, including GSH, lacks data on Cu(II) reduction when Cu(II) and GSH are present in equimolar ratios. Therefore, we decided to additionally investigate Cu(II) reduction in the presence of thiols when Cu:thiol = 1:1.

Interaction of Cu and GSH
The interactions of Cu(II) with GSH were studied at Cu:GSH ratios of 1:1 and 1:10 over the course of 100 min at pH = 8.4 in 0.1 M borate buffer under oxic conditions (Figures 1  and 2). For both Cu:GSH ratios studied, Cu(II) reduction occurred immediately after the addition of Cu(II) to the buffered GSH solution, and Cu(I) concentrations were monitored using a BCS assay as described in Section 3.2. The kinetics of Cu(II) reduction and Cu(I) oxidation are shown in Figure 1. At a ratio of Cu:GSH = 1:10, all the Cu present in the solution was in its reduced form at the beginning of the reaction (the added Cu(II) was equal to Cu(I) detected using the BCS assay), implying the rapid reduction of Cu(II) in the presence of excess GSH. Oxidation of Cu(I) under the conditions of a tenfold excess of GSH with respect to the initial Cu(II) started after 60 min, and after 100 min, Cu(I) was no longer present in the solution. The UV-Vis spectra of Cu:GSH = 1:10 show peaks indicative of the ligand charge transfer band of a Cu(I)-GSH complex, with A max values observed at 255 nm and 300 nm [51,52]. After 60 min, the oxidation of Cu(I) began and a decrease in A 255 and A 300 was also observed, while a peak characteristic of the Cu(II) complex with oxidized glutathione, Cu(II)-GSSG, appeared with an A max of 625 nm [52,53]. A peak with an A max of 250 nm was also observed in the solution with Cu:GSH = 1:1, but a peak at 300 nm was not present. A peak with an A max of 625 nm, which can be attributed to the Cu(II)-GSSG complex, was present from the beginning of the reaction, which is consistent with Cu(I) accounting for a maximum of 52% of the Cu present in the solution. An examination of reaction kinetics (ln Cu(I) and 1/Cu(I) concentration-time) shows significant differences between the reaction kinetics of mixtures with Cu:GSH = 1:1 and those of mixtures with Cu:GSH = 1:10 ( Figure [54,55]. The early stage of the reaction, in the case of GSH excess, can be explained by strong interactions between the Cu(I) and GSH present in excess, which prevent fast Cu(I) oxidation with molecular O 2 . Ngamchuea et al. proposed a two-phase mechanism in which a Cu(II)-GSH complex with a stoichiometry of 1:2 is formed in the first phase, and the second phase comprises the Cu-catalyzed oxidation of GSH to GSSG [56]. In general, our results are in agreement with the two-phase mechanism of GSH oxidation proposed by Ngamchuea et al. [56]. However, while Ngamchuea et al. proposed the formation of a Cu(II) complex with GSH as an intermediate species in the first reaction phase, our results support the formation of a Cu(II) complex with oxidized GSH in the early as well as later stages of the reaction, depending on the Cu(II)-to-GSH ratio [57,58]. In the solution containing Cu:GSH = 1:1, a peak with an A max of 625 nm was present from the very beginning of the reaction, indicating the presence of a Cu(II)-GSSG complex even at the early reaction stage. In contrast, for Cu:GSH = 1:10, a peak with an A max of 625 nm was not present in the first reaction stage, since 100% of the Cu present in the solution was in its reduced form in the first reaction stage. For Cu:GSH = 1:10, a peak with an A max of 625 was formed when Cu(I) oxidation began.

Interaction of Cu and L-cys
The kinetics of Cu(II) reduction and Cu(I) oxidation over time at Cu:L-cys ratios of 1:1, 1:2.5, 1:5, and 1:10 are shown in Figures 3 and 4. Immediately after the addition of Cu(II) to the L-cys solution, the reduction of Cu(II) to Cu(I) was observed, as in the study of the interaction of Cu and GSH presented in this manuscript (Section 2.1.). The presence of Cu(I) was detected in all Cu:L-cys ratios studied during the 180 min, indicating the formation of a Cu(I) complex with L-cys that was resistant to oxidation with molecular O 2 . The amount of reduced Cu(II) and the stability of Cu(I) to oxidation increased with increasing Cu:L-cys ratios, as shown in Figure 3. The UV-Vis spectrum of L-cys alone in 0.1 M borate buffer is characterized by a peak with an A max around 235 nm, which is consistent with the literature data [59]. Furthermore, Battin et al. reported A max values at the same wavelength for all sulfur-mediated compounds, which prevented Cu-mediated DNA damage [51]. Immediately after the addition of Cu(II) to a solution containing L-cys and borate buffer, the previously observed peak at 230 nm disappeared, indicating the loss of free L-cys. At the same time, two new peaks appeared with A max values of 260 nm and 336 nm. Similar peaks were also observed in studies by Pecci et al. and Rigo et al. and were attributed to Cu(I)-cys complexes, with a difference in the position of the second peak (300 nm), which could be due to the different reaction conditions (different buffers and pH values, and different oxic and anoxic reaction conditions) [60,61]. Cavalini et al. reported a compound with an A max of 330 nm at a pH of >8 as an intermediate Cu(II)-cysteine complex with a stoichiometry of 1:2, which disappeared after cysteine oxidation during the reaction [59]. Also in our case, the peaks at 260 and 330 nm disappeared during the course of the reaction when L-cys was oxidized and cystine was formed, which was confirmed by an increase in A in the 200-400 nm region where cystine absorbs [61].   Figure S2). In the model solutions containing equimolar Lcys and Cu(II), linearity was observed in the ln Cu(I) concentration-time plots during the first minutes of the reaction course (up to 20 min), which is characteristic of first-order reaction kinetics. As the L-cys concentration increased, reaching a tenfold excess over the initial Cu(II) concentration, the linearity disappeared even in the early stages of the reaction course, and the reaction course showed a pattern that was independent of the Cu(I) concentration. A comparison of the calculated rate constants (k, Table S1 in the Supplementary Materials) for ln Cu(I) concentration-time indicates faster Cu(I) oxidation in the presence of L-cys compared to GSH (under the conditions of an equimolar Cu:thiol ratio and under the conditions of linearity in ln Cu(I) concentration-time). The changes in the reaction kinetics over time in the solution with equimolar L-cys and Cu can be attributed to the Fenton-like process. The same reaction pattern was previously observed by Moffet and Zika for Cu(I) oxidation in solutions containing NaCl [55]. The pattern is explained by the existence of a reverse reaction in which formed Cu(II) is reduced with ROS. It is known that the absence of free thiol in the solution produces H 2 O 2 , and Fenton-like reactions between Cu(I)-thiol and ROS occur [54]. During the reverse reaction, Cu(I) was formed and additionally influenced the reaction kinetics, as shown by a change in the reaction slope (k) as well as the reaction order. When L-cys was present in excess relative to the added Cu, the reaction kinetics were independent of the Cu(I) concentration alone, indicating a complex reaction scheme involving oxidation of the free L-cys and the Cu(I)-cys complex, followed by a competition between cystine and L-cys for Cu(I) [60,62,63].  Similar to what has already been described in this manuscript, we investigated the reaction kinetics in more detail (Supplementary Materials, Figure S3). The reaction course allowed us to analyze reaction kinetics with respect to the Cu(I) concentration at all four investigated Cu:MPA ratios. An analysis of k for the plots of ln Cu(I)-time (in the parts where linearity was observed) shows the dependence of k on MPA concentration, as shown in Table S1 (Supplementary Materials). The first-order reaction rate (with respect to the Cu(I) concentration) for the Cu(I) oxidation reaction decreased with an increasing MPA concentration. This observed pattern again points to the importance of the experimental conditions used, i.e., the initial ratio of Cu(II) to thiol, as already observed in the case of Cu and GSH and L-cys. When thiol (in this case, MPA) is present in a tenfold excess with respect to the initial concentration of Cu(II), the formation of Cu(I)-MPA is favored and the oxidation of Cu(I) by O 2 is strongly inhibited as a result of the Cu(I) complexation ( Figure 6a). As was already observed (in Section 2.2, where the interaction of Cu with L-cys is investigated), a shift in the reaction kinetics from first to second order occurred over the course of the reaction, which in turn can be explained by the fact that more than just Cu(I) was involved in the reaction mechanism, namely ROS and Cu(II).

Interaction of Cu with TAA
The interaction of Cu(II) with TAA was studied by adding an aliquot of Cu(II) to solutions with increasing TAA in 0.1 M borate buffer at pH 8.4, as shown in Figures 7 and 8. Four different ratios of Cu to TAA were studied over 180 min: Cu:TAA = 1:1, Cu:TAA = 1:2.5, Cu:TAA = 1:5, and Cu:TAA = 1:10. Immediately after the addition of Cu(II) to TAA, a reduction of Cu(II) took place, but at a much slower rate than when Cu(II) was mixed with GSH, L-Cys, or MPA. For Cu:TAA = 1:1, the Cu(I) maximum was reached after about 100 min, while for Cu:TAA = 1:2.5, Cu:TAA = 1:5, and Cu:TAA = 1:10, the maximum of Cu(I) was reached after 80, 40, and 7 min, respectively. Both Cu:TAA = 1:5 and Cu:TAA = 1:10 resulted in the 100% reduction of Cu(II), but the maximum was reached faster with a higher Cu:TAA ratio. In addition to monitoring Cu(I) with the BCS assay, UV-Vis spectra of the Cu:TAA = 1:5 solution were recorded because the 1:5 ratio resulted in the 100% reduction of Cu(II), and the reduction was slow enough to monitor with UV-Vis spectrophotometry. The UV-Vis spectrum of TAA in 0.1 M borate buffer without Cu addition resulted in a peak with an A max of 246 nm and a lower peak with an A max of 300 nm. After the addition of Cu(II) to the TAA solution, the peak with an A max of 246 decreased while the Cu(I) concentration increased, indicating a loss of free thiol due to its complexation with Cu(I) and its oxidation after reaction with Cu(II). After 40 min, the concentration of Cu(I) began to decrease (Figure 7), followed by an increase in the peak with an A max of 246 nm (Figure 8), indicating the release of the thiol group bound to Cu. These results are consistent with our recent study on the interaction of vanadates with thiols, which also showed the lower reducing ability of TAA compared to L-cysteine and MPA [64]. It seems very likely that, because of the different structural properties of the thiols studied (different chelating properties, i.e., monodentate vs. bidentate), TAA is not able to stabilize the formed Cu(I) against oxidation and thus exhibits a lower reduction of Cu(II) at lower Cu:TAA ratios. Moreover, the SH group in the TAA molecule is located next to the C=O group, resulting in a resonant structure, which is not the case for any other thiol studied (GSH, L-cys, or MPA), affecting the electron-donating properties of the sulfur atom of TAA [65].  After closely inspecting the reaction kinetics (Supplementary Materials, Figure S4) for Cu:TAA = 1:1, we were able to determine the k value of the reduction reaction of Cu(II) with TAA (Table S1, Supplementary Materials). The figure in the Supplementary Materials ( Figure S4a-h) also indicates rapid changes in the reaction sequence as well as in the course of the reaction (reduction vs. oxidation). Unfortunately, we were able to determine k for the Cu(I) oxidation reaction only in the cases where TAA was in five-and tenfold excess with respect to the added Cu (Table S1, Supplementary Materials). However, as noted in the case of L-cys and MPA, the initial reduction of Cu(II) was followed by the Cu(I) oxidation reaction, which, in turn, shows changes in the reaction order from first to second that, again, can be explained as earlier.

Materials
All solutions were prepared with deionized water from the Milli-Q (MQ) system (18.2 MΩ, Millipore, Burlington, MA, USA), and all chemicals were of analytical grade. The Cu(I) standard solution used for the calibration curve of the Cu(I)-bathocuproine complex (see Section 3.2) was prepared by dissolving copper(I) chloride (CuCl; Thermo Fisher Scientific, Waltham, MA, USA) in a solution containing 1 M sodium chloride (NaCl; Grammol, North Salt Lake, UT, USA) and 0.1 M hydrochloric acid (HCl; Roth, Newport Beach, CA, USA), which was previously purged with high purity nitrogen to remove oxygen [66]. The bathocuproine sulfonate disodium salt hydrate (BCS; Thermo Fisher Scientific), was prepared by dissolving BCS in MQ water to a concentration of 1000 µM.
The Cu(II) standard solution was prepared by dissolving copper(II) sulfate (CuSO 4 ; VWR BDH Prolabo Chemicals, Radnor, PA, USA) in MQ water to a final concentration of 0.01 M Cu(II). Thiol solutions of L-cysteine (L-cys), reduced glutathione (GSH), thioacetic acid (TAA), and 3-mercaptopropionic acid (MPA) were purchased from Thermo Fisher Scientific and prepared fresh daily by dissolving the thiols in MQ water to a final concentration of 0.01 M. A constant ionic strength and a pH = 8.4, relevant to the environmental and physiological conditions, in the model solutions containing Cu and thiol were achieved using 0.1 M borate buffer. The borate buffer was prepared from ortho-boric acid (VWR BDH Prolabo Chemicals), and its pH was adjusted to pH = 8.4 with sodium hydroxide (NaOH; Lach-ner Chemicals, Neratovice, Czech Republic).
Copper reduction by individual thiol species was investigated by measuring the Cu(I) concentration with a UV-Vis spectrophotometer (Analytik Jena, Jena, Germany) in a 1 cm quartz cuvette.

Procedure
The interaction of Cu with thiols was studied by adding aliquots of a Cu(II) stock solution to a solution containing thiol (L-cysteine, glutathione, 3-mercaptopropionic acid, or thioacetic acid) buffered to pH = 8.4 with 0.1 M borate buffer. The Cu concentration of 100 µM was the same in all experiments, while the thiol concentrations ranged from 100 to 1000 µM. The kinetics of the reduction of Cu(II) and the oxidation of Cu(I) were studied by adding aliquots of the Cu-thiol solution to the mixture of BCS and EDTA (BCS assay). Previous studies have shown that BCS effectively binds Cu(I) in an orange complex with an absorption maximum (A max ) of 484 nm, while a masking ligand is necessary to avoid Cu(II) interference [66,67]. In this study, EDTA was used as the masking ligand for Cu(II) with a fivefold excess of EDTA over BCS, which has been found to be optimal for ensuring Cu(II) complexation while avoiding Cu(I) oxidation [66,67]. The Cu(I) standard solution was freshly prepared before the experiment, following the procedure described in Section 3.1. To better understand the reaction mechanism, in addition to monitoring the Cu(I) concentration, UV-Vis spectra of solutions containing Cu and L-cys were recorded at the same reaction time as was the Cu(I) determination.

Conclusions
The study of the interactions of Cu with GSH (glutathione), L-cys (L-cysteine), MPA (3-mercaptopripionic acid), and TAA (thioacetic acid) revealed a complex reaction scheme with a rapid Cu(II) reduction in the case of GSH, L-cys, and MPA. The reduction reaction was rapid (within seconds), and we could instead follow the kinetics of Cu(I) oxidation under conditions where the thiol compound being studied was varied in stoichiometry with respect to the Cu(II) added. An examination of the ln Cu(I) and 1/Cu(I) concentrations over time shows that the reaction kinetics changed over the course of the reaction. Only the solution with Cu:GSH = 1:1 showed first-order reaction kinetics with respect to Cu(I) oxidation, and the kinetics did not change during the 180 min of the reaction course. During the reaction courses of the Cu:L-cys = 1:1 and Cu:MPA = 1:1 mixtures, changes in the reaction kinetics for the Cu(I) oxidation process were observed in the later stages of the reaction course. In the later stages of the reaction, the H 2 O 2 formed during Cu(I) oxidation with O 2 accumulated and triggered a Fenton-like reaction. The addition of Cu(I) to the already-ongoing reaction and the involvement of further reactants (Cu(II) and ROS) changed the reaction kinetics, as can be observed. Increasing the thiol concentration also increased the stability of Cu(I) against oxidation, which is probably due to the formation of stable Cu(I) complexes with the investigated thiols. Only in the case of TAA were we able to follow the reduction kinetics of added Cu(II), and the reducing power of the thiols studied can be summarized as follows: L-cys ∼ = GSH > MPA > TAA. The lower reducing ability of TAA towards Cu(II) can be explained by its monodentate nature, as well as by the SH group being located next to the C=O group, with respect to the GSH, L-cys, and MPA. Data Availability Statement: Data available on request due to restrictions. The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy.