Interaction of British Anti-Lewisite (BAL) with Copper(I) and Copper(II) compounds in conjunction with Wilson’s disease

Abstract

The chemistry of British Anti-Lewisite (BAL) and copper in relation to the therapeutic action for the treatment of Wilson’s disease was studied by reacting selected Cu(I) and Cu(II) compounds with a range of thiol compounds. Optimum conditions for the preparation of cuprous mercaptides, Cu^I–SR, from thiophenol, 3-mercapto-1,2-propanediol (MPDH), cysteine (CysH), and glutathione (GSH), were established. BAL displaced Cu(I) from Cu^I–MPD giving Cu^I₂–BAL irrespective of stoichiometry^, while from Cu^I–cys and Cu^I–SG, Cu–BAL were obtained. BAL passivated Cu^I–SPh. The direct reaction of BAL with either Cu(II) or Cu(I) gave insoluble amorphous solids assumed to be Cu^I₂-(oxidized dimeric BAL) or Cu^I₂–BAL, respectively, that could not be completely analyzed or solubilized under various conditions. Thus, the detection of elevated copper in the urine of Wilson’s disease patients treated with BAL may not be directly due to the solubilization of endogenous copper by BAL. Cuprous mercaptides are easily autoxidized by attack of O₂ to Cu^I to give Cu^II probably as peroxo complexes.

Keywords

Wilson’s disease Copper(II)Copper(I)cuprous mercaptides British Anti-Lewisite

1 Introduction

Wilson’s disease [45] (hepatolenticular degeneration) is a rare genetic disorder of copper metabolism leading to copper accumulation in the liver and extrahepatic organs such as the brain and cornea [2]. Patients may present with combinations of hepatic, neurological and psychiatric symptoms. Copper overload in patients with Wilson’s disease is caused by impairment to the biliary route for excretion of dietary copper. For the majority of patients, Wilson’s disease is a treatable condition but requires life-long use of drugs.

Five drugs have commonly been used to treat Wilson’s disease [30]: BAL; D-penicillamine; zinc sulfate (or acetate); trientine dihydrochloride; and ammonium tetrathiomolybdate, Fig. 1. The BAL analogues, meso-2,3-dimercaptosuccinic acid (DMSA) or its disodium salt and sodium 2,3-dimercapto-1-propanesulfonate (DMPS), Fig. 1, have also been investigated, particularly in China [46]. Each drug can promote excretion of copper from the body and/or transform copper into a metabolically unavailable form in the patient. Four of these drugs contain thiol groups: the monothiol D-penicillamine and the dithiols BAL, DMSA and DMPS.

Fig.1

Compounds used to treat Wilson’s disease in chronological order of their introduction.

$2 {Cu}^{2 +} + 2 RSH \to 2 {Cu}^{+} + RSSR + 2 H^{+}$ (1) $2 {Cu}^{+} + 2 RSH \to 2 Cu - SR + 2 H^{+}$ (2)

Thiols, RSH, containing non-functional groups are quickly oxidized by Cu(II) to the corresponding disulfide with concomitant reduction of Cu(II) to Cu(I) [12], Equation (1). Copper(I) then forms the cuprous mercaptide, Cu–SR, Equation (2). If steric effects hinder the formation of the disulfide, Cu(II) does not form a cuprous mercaptide [12]. The mechanism of this redox process is not well understood, but it is believed to involve an electron transfer within [Cu^II-RS]⁺ from sulfur to Cu(II) to give Cu(I) and a thiyl radical, RS·, which dimerizes to the disulfide [12].

The end product can be more complicated in examples where the thiol contains functional groups, the stoichiometry [Cu(II)]:[thiol] is not 1:2, and air (which oxidizes cuprous mercaptides) is present or absent. For example, the interaction of Cu(II) with cysteine can produce a complex [Cu^I(cys)₂]^– [6]. With penicillamine, mixed valence compounds can be obtained [4, 36], and ternary complexes such as albumin/copper (I or II)/penicillamine [37] are possible. A water-soluble neutral mixed valence complex, Cu^IICu^I(DL-pen)₂, is obtained from excess CuCl₂ and DL-penicillamine at pH 6.2 [36], while [Cu₆^IICu₈^I(D-pen)₁₂Cl]^5– has been characterised in the intensely purple solution obtained by mixing D-penicillamine and CuCl₂·2H₂O in sodium acetate buffer at pH 6.2 [4]. However, there are doubts whether this copper–penicillamine solution chemistry operates as a route for the decoppering properties of penicillamine in vivo [41], and other mechanisms have been suggested, for example the induction of hepatic metallothionein [35].

British Anti-Lewisite (BAL, dimercaprol, 2,3-dimercapto-1-propanol), Fig. 1, has an interesting story [42] as an antidote to Lewisite (ClCH = CHAsCl₂); its interaction with many heavy metal cations in vivo [39] and in vitro [17] has been reviewed. BAL is a viscous oil, moderately soluble in water (6–8 g/100 mL). Racemic BAL is marketed for clinical use as a solution in peanut oil with the addition of benzyl benzoate as a stabilizer [40].

An observation that BAL promoted the urinary excretion of copper led J.N. Cumings to suggest in 1948 that BAL might be a rational therapy for the treatment of Wilson’s disease [30, 31]. This followed Cumings’ (and others’) finding of an excessive accumulation of copper in the brain and the liver of Wilson’s disease patients [31]. In 1951, Cumings in the UK and independently Denny-Brown and Porter in the USA reported the use of BAL for treating Wilson’s disease patients, thus identifying copper chelation as a rational therapy for this metabolic disorder [31]. Although BAL produced relief of symptoms in some Wilson’s disease patients, it also gave rise to toxic reactions, and resistance, both biochemical and clinical, developed in many patients [43].

Three pathways have been suggested by which BAL can detoxify heavy metals and metalloids:

The formation of a five-membered chelate ring (e.g. with As(III)). This was postulated [39, 42], and verified in vitro [18], and the term “chelation therapy” was established in the literature [8]. The water insoluble Pb–BAL·H₂O most likely forms such a chelate [17].

If the formation of a five-membered ring is prevented, (e.g. with Hg(II) [7]) an oligomer or polymer may be formed, which is insoluble in aqueous solutions at physiological pH.

A redox reaction between BAL and a reducible cation, such as Au(III) [17] or Cu(II), and further reactions of the formed products.

There are only two reports on the interaction of Cu(II) with BAL in vitro. In the first [44], mixing a very dilute aqueous solution of excess CuSO₄ and a known amount of BAL afforded a dark blue solid that on drying turned green. Titration of unreacted Cu(II) revealed that the solid had a copper/BAL 1:1 molar ratio. In the second report [26], an aqueous mixture of Cu^II/BAL/HCl 1:32:320 molar ratio formed a butanol-soluble complex with a 1:1 molar ratio of copper/BAL. ${({Ph}_{3} P)}_{3} Cu - Cl + RSH \to {({Ph}_{3} P)}_{2} {Cu - SR + Ph}_{3} P + HCl$ (3)

In this paper, we report the reactions of four monothiols [thiophenol, 3-mercapto-1,2-propanediol (MPDH), cysteine and glutathione] and a dithiol (BAL) with selected Cu(II) {CuCl₂·2H₂O, CuSO₄·5H₂O, Cu(AcO)₂·H₂O and cis-Cu(gly)₂·H₂O [28]} and Cu(I) {CuCl, Cu₂O, (Ph₃P)₃Cu–Cl [19, 32] and (Ph₃P)₂Cu (NO₃) [14]} compounds. The monothiols, a priori, were expected to react with Cu(II) compounds according to Equations (1) and (2) using a 2:1 molar ratio of monothiol to Cu(II). The chemistry of Cu(II) and BAL was anticipated to be more complicated. Copper(I) compounds such as CuCl and Cu₂O should react with monothiols according to Equation (2), while (Ph₃P)₃Cu–Cl should react with monothiols according to Equation (3) [32]. For the reaction of BAL with Cu(I), the products will depend on the ratio Cu^I:BAL. We checked if BAL can sequester Cu(I) from isolated cuprous-organic acids, cuprous-amino acids and cuprous mercaptides. The oxidized (by H₂O₂) BAL was studied in the presence of Cu(II) and Cu(I) compounds for comparative reasons. We examined whether the copper–BAL insoluble compounds can be solubilized under various conditions, especially by selected urinary constituents. Attempts were also made towards elucidating the process of autoxidation of cuprous mercaptides by ¹H NMR.

From the pharmacological point of view our results agree with the suggestion that chelating agents act therapeutically by mobilizing relatively small amounts of toxic metals [8]. However, the observation that BAL glucoside (BAL-Intrav) (2–4 g/man) increased the excretion of copper in urine of six subjects about 20 times (3.6–14.0μg/h before to 65.8–235.0μg/h after treatment) [27] raises questions about the mechanism of this enhanced copper excretion. The finding [25, 41] that the water insoluble Cu(I) complexes with cysteine, penicillamine and glutathione forms water soluble ternary complexes in the presence of a second thioamino acid, e.g. Cu^I–cysteine-penicillamine, may offer a new explanation for the detoxification of heavy metal ions, including copper, by chelatingagents.

2 Experimental section

2.1 Materials and methods

Silica gel 60 H for thin layer chromatography (TLC) was obtained from Merck. Deionized water pH 6.5, aqueous NaCl solutions pH 5.6–6.27, aqueous urea solutions pH 7.8–8.07 and phosphate buffer pH 7.00 were used. Water and methanol solvents were deaerated by boiling and bubbling nitrogen just before use and are regarded as not strictly deoxygenated. TLCs were run on microslides and the spots were made visible by iodine vapors. IR spectra were taken in KBr discs on a Perkin-Elmer, model 16PC, FT-IR spectrometer, ¹H NMR (at 400 MHz) spectra were obtained on a Bruker, model DPX Avance, spectrometer and X-ray powder diffractograms were recorded with a Siemens powder X-ray diffractometer, model D5000. Electrospray ionization (ESI positive ion detection) mass spectra of the products of oxidation of BAL with hydrogen peroxide in methanol were recorded on a Micromass-Platform LC spectrometer. Elemental analyses were obtained through the Center of Instrumental Analyses, University of Patras, Patras, Greece.

Copper(II) chloride dihydrate (Alfa Aesar), copper(II) sulfate pentahydrate (Riedel – de Haën), copper(II) acetate monohydrate (Ferak), 3-mercapto-1,2-propanediol [95%; (TLC (Et₂O) was used to exclude glycerol or 3-chloro-1,2-propanediol as impurities) (Aldrich), and BAL (Merck; TLC (Et₂O) determined traces of oxidized BAL at R_f 0.60) were used as received. The following copper compounds were prepared by literature methods and had the following properties:

cis-B is(glycinato)copper(II) monohydrate 28] {85%; insoluble in CHCl₃, MeOH, 95% EtOH, slightly to moderately soluble in H₂O and moderately soluble in DMSO; IR (KBr): as in ref. [28].

C opper(I) chloride [21] {IR (KBr): ṽ = 3446 s, broad, 1000 vs, very broad}.

C opper(I) oxide from CuSO₄·5H₂O (80 mmol), glucose (40 mmol), NaOH (200 mmol) in warm water (150 mL) [33] as a brick-colored solid (100%) after filtration and washing with water, methanol and ether.

Tris(triphenylphosphine)copper(I) chloride [19] {96%; recrystallized from methanol (1.00 g/100 mL) with 75% recovery, insoluble in Et₂O, Me₂CO, sparingly-to-moderately soluble in DMSO and soluble in CHCl₃; m.p. at 163°C shrinks to a globule, at 168°C gives an opalescent globule that clears at approx. 210°C, lit. [19] 7–170°C, lit. [10] 166°C that clears at 202°C; IR (KBr): ṽ = 3046 m, 1480 s, 1434 s, 1090 s, 744 vs, 734 vs, 694 vs, 618 s, 518 vs, 508 vs, 482 vs, see Costa et al. [10]; ¹H NMR (CDCl₃): δ= 7.23 (m, 2 H, meta Ph-H), 7.33 (m, 3 H, ortho and para Ph-H).

B i s(triphenylphosphine)copper(I) nitrate [14] {98%; recrystallized from methanol (1.00 g/100 mL) with 50% recovery; soluble in CHCl₃, CH₂Cl₂, MeCN, DMSO, DMF, sparingly to moderately soluble in MeOH, EtOH, insoluble in Et₂O, Me₂CO; m.p. 6–247°C dec., lit. [14] 248°C dec.; IR (KBr): has a medium ionic NO₃^- at 1384 cm^–1 [16] and a very strong chelated NO₃ at 1464 and 1274 cm^–1 [11, 20]. ¹H NMR (CDCl₃, TMS): δ= 7.29 (s, 2 H, meta Ph-H), 7.37, 7.48, 7.54 and 7.67 (m, 3 H, ortho and para Ph-H). The NMR solution in 30 min gave a bluish fluffy solid.

The yellow C u–S Ph [1] is insoluble in MeOH and CHCl₃ and sparingly soluble in DMSO. M.p.: at 245°C darkens to brownish and at 260°C shrinks and turns black. IR (KBr): ṽ = 3048 m, 1576 s, 1540 w, 1474 vs, 1436 s, 1296 w, 1175 w, 1078 s, 1066 m, 1022 s, 998 m, 956 w, 898 w, 730 vs, 684 vs, 476 s.

The very pale yellow (Ph₃P)₂Cu–S Ph [32] {calculated for C₄₂H₃₅SP₂Cu (M_r 697.20): C 72.35, H 5.06, S 4.59%; found C 71.80, H 4.22, S 5.35% } is soluble in CHCl₃, sparingly soluble in Et₂O, Me₂CO and MeOH, and moderately soluble in DMSO. M.p. 0–162°C; lit. [32] 0–162°C. IR (KBr): ṽ = 3048 m, 1570 m, 1476 m, 1466 m, 1432 s, 1308 w, 1182 w, 1154 w, 1090 s, 1082 s, 1020 m, 996 m, 850 w, 744 vs, 732 s, 692 vs, 618 m, 520 s, 512 vs, 500 s, 422 m. ¹H NMR (CDCl₃, TMS): δ= 7.25 (m, all Ph-H protons).

2.2 Preparation of cuprous 2,3-dihydroxypropyl-1-mercaptide, Cu^I-MPD

To a centrifuge tube 3-mercapto-1,2-propanediol (MPDH) (450 mg, 4.16 mmol) was dissolved in non deaerated water (1 mL) and copper(I) chloride (375 mg, 3.78 mmol) was added. At once a yellow gel was formed that on stirring at room temperature (RT) gave a yellow solution. After 30 min stirring at RT, methanol (9 mL) was added drop-wise to give a yellowish suspension that was stirred at RT for 2 h. TLC (Et₂O) showed the excess MPDH that had not reacted at R_f 0.36 and the impurity in the 3-mercapto-1,2-propanediol at R_f 0.16. Centrifugation, washing with methanol (5 mL) and drying in vacuo gave Cu^I-MPD as a yellow powder containing approx. 5% methanol (628 mg, 97%) insoluble in methanol, sparingly soluble in water and soluble in DMSO giving a yellow solution that in approx. 15 min turned brown. M.p.: at approx. 60°C starts shrinking slowly, at 85°C swells and bubbles, at 125°C shrinks to a glassy oily mass, at approx. 190°C starts darkening, at approx. 200°C turns light orange and at 210°C turns black. Calculated for C₃H₇O₂SCu (M_r 170.69): C 21.11, H 4.14, S 18.78%; found C 21.63, H 4.53, S 18.22%. IR (KBr, yellow disc): ṽ = 3332 vs, broad, 2920 s, 2860 s, 1643 m, 1408 s, 1326 m, 1288 m, 1218 m, 1064 vs, 1012 vs, 924 w, 878 s, 814 w, 730 m, 628 m, 544 m. ¹H NMR (DMSO-d₆, TMS; yellow solution): δ= 2.70 (m, broad, 1 H, CHHSCu), 2.89 (m, broad, 1 H, CHHSCu), 3.3–3.6 (m, broad, 2 H, CH₂OH), 3.62 (m, broad, 1 H, CHOH), 4.70 (very broad, 1 H, CH₂OH), 5.50 (very broad, 1 H, CHOH), Fig. 2A. At the end of the run a brown tint was visible at the top of the solution. ¹H NMR (DMSO-d₆, TMS; brown solution: run 3 h after running the yellow sample): δ= 2.72 (m, broad, 1 H, CHHSCu), 2.90 (m, broad, 1 H, CHHSCu), 3.43 (broad,1 H, CHHOH), 3.48 (broad, 1 H, CHHOH), 3.64 (m, broad, 1 H, CHOH), 4.09 (broad, 0.03 H, MeOH), 4.63 (t, J 5.6 Hz, 0.23 H, OH), 4.72 (broad, 0.72 H, OH), 4.87 (d, J 5.2 Hz, 0.14 H, OH), 5.50 (broad, 0.62 H, OH). The methanol singlet was broadened, Fig. 2B. After 12 h the OH signals were broad and their integrations became 0.28, 0.23, 0.89, 0.20 and 0.78, respectively.

Fig.2

The yellow cuprous 2,3-dihydroxypropyl-1-mercaptide (Cu^I-MPD) in DMSO-d₆ gave a yellow solution that was run just after dissolution, A. After 3 h at room temperature the yellow solution became brown and had the spectrum B. The solid yellow mercaptide on standing became brown and a brown sample in DMSO-d₆ had the spectrum C. 2.50 ppm: DMSO, 3.17 ppm: MeOH impurity, 3.33 ppm: water in DMSO.

The product stored at –20°C became darker yellow and remained so for 10 days while it developed a greenish tint after 2 months. After 1 month storing at +4°C the solid Cu^I-MPD turned brown and had the following thermal behavior: at 68°C starts shrinking slowly, at approx. 80°C shrinks, at 108°C brown bubbles formed, at approx. 130°C a brown film forms, at 185°C orange bubbles appeared and at 210°C turns black. ¹H NMR of the brown solid (DMSO-d₆, TMS; brown solution): nearly the same as the above brown solution, Fig. 2C. At room temperature the solid in an open vial turned brown in 7 days while when stored in a closed vial at room temperature darkened in 1 day, turned khaki in 3 days, light brown in 6 days and brown in 10 days.

2.3 Reactions of cysteine with copper(II) and copper(I) compounds. Preparation of cuprous cysteinate, Cu^I–cys

A variety of starting conditions (e.g. adding a mixture of the solid reagents to a deaerated solvent or vice versa, adding a solution of a copper compound to a solution of cysteine in the absence and in the presence of a base to scavenge the acid produced etc.) were tried. The best preparation, as regards the stability of the Cu^I–cys during its preparation, was from Cu₂O: In a 50 mL round-bottomed flask containing a nitrogen inlet, water (10 mL) was deaerated at 100°C while bubbling nitrogen. A mixture of Cu₂O (143 mg, 1 mmol) and cysteine (242 mg, 2 mmol) was then added and the system stirred at 100°C (oil bath temperature) for 3 h. Cu₂O had reacted in approx. 45 min. The cooled to RT suspension was transferred to a centrifuge tube and the solid was washed with methanol (2 mL) and ether (3 mL) and dried in vacuo to give the product, Cu^I–cys, (315 mg, 86%) as an off-white powder insoluble in H₂O and DMSO (no signals in D₂O or DMSO-d₆ in the ¹H NMR spectra). M.p.: from 50°C starts darkening slowly, at 140°C turns brownish and at approx. 170°C turns brown. Calculated for C₃H₆NSO₂Cu (M_r 183.69): C 19.61, H 3.29, N 7.63, S 17.45%; found C 19.52, H 3.00, N 8.13, S 17.01%. IR (KBr): ṽ = 3416 m, broad, 3000 s, very broad, 1620 vs, 1598 vs, 1560 s, shoulder, 1502 s, 1426 s, 1396 vs, 1336 s, 1298 s, 1180 m, 1146 w, 1104 w, 1038 m, 782 w, 656 w, 606 w, 544 m. The powder X-ray spectrum of Cu^I–cys showed some crystallinity.

2.4 Reactions of glutathione with copper(II) and copper(I) compounds. Preparation of cuprous glutathionate, Cu^I–SG

These reactions were generally carried out by adding the solid reactants to a deaerated solvent in centrifuge tubes. Pure Cu^I–SG was prepared from Cu(AcO)₂·H₂O, cis-Cu(gly)₂·H₂O or Cu₂O. To a centrifuge tube containing deaerated water (5 mL) was added together Cu(AcO)₂·H₂O (40 mg, 0.2 mmol) and GSH (123 mg, 0.4 mmol) and stirred at RT for 2 h. The white suspension formed was centrifuged, washed with deaerated water (2 mL) and dried in vacuo over P₂O₅ to give the product, Cu^I–SG, as a white, somewhat hard solid (71 mg, 95%), insoluble in H₂O, MeOH and warm DMSO. M.p.: at approx. 180°C darkens to brown and at 200°C swells to a brown foam. Calculated for C₁₀H₁₆N₃O₆SCu (M_r 369.85): C 32.47, H 4.36, N 11.36, S 8.67; found C 32.52, H 4.65, N 11.54, S 9.02%. IR (KBr): ṽ = all bands broad: 3400 s, 3280 s, 3078 s, 2900 m, 1726 m, 1640 vs, 1538 vs, 1412 m, 1222 m, 668 w. The powder X-ray spectrum of Cu^I–SG revealed a total absence of crystallinity.

2.5 Reactions of BAL with some Cu(II) compounds

In a centrifuge tube containing a solution of BAL (0.5 mmol) in deaerated methanol (1 mL) was added a solution of CuCl₂·2H₂O (0.5 mmol) in deaerated water (3 mL) and the dark blue to grey solvated suspension stirred at RT for 3 h. TLC (Et₂O) of the supernatant did not show BAL at R_f 0.82 but traces of oxidized BAL at R_f 0.61. Centrifugation and washing (by stirring for 5 min each time) with H₂O (1×3 mL), MeOH (3×3 mL) and Et₂O (1×4 mL) gave a dark blue to grey solid (92.8 mg, 100%) probably as a mixture of isomers A (see Section 3.2, Fig. 4) insoluble in 1 M HCl and MeCN, sparingly soluble in warm DMSO giving a yellow-green light suspension that on cooling transformed to a yellow-brown suspension. M.p.: at approx. 150°C shrinks, at approx. 195°C swells a little and does not change up to 300°C. Calculated for {(oxidized dimeric BAL)·2Cu⁺} C₆H₁₂O₂S₄Cu₂ (M_r 371.50): C 19.40, H 3.26, S 34.53%; found C 19.95, H 3.75, S 34.12%. IR (KBr): ṽ = 3334 vs, broad, 2912 s, 2860 s, 1636 m, 1456 m, 1404 s, 1258 s, 1220 s, 1042 vs, 1006 vs, 970 s, 884 s, 654 s, 570 s, 520 s. Its powder X-ray spectrum showed a total absence of crystallinity, Fig. 3. Under the same conditions CuSO₄·5H₂O and BAL (0.5 mmol each) gave 94.6 mg (102%) of a dark blue - grey solid. Cu(AcO)₂·H₂O and cis-Cu(gly)₂·H₂O formed dark blue solvates from which no solid could be isolated by centrifugation.

Fig.3

Powder X-ray spectrum of the dark blue to grey colored solid assumed to be mixture A (Fig. 4) obtained from the reaction of BAL with Cu(II) compounds. The same spectra were obtained from the dark green solids from the reaction of BAL with Cu(I) compounds assumed to be Cu^I₂–BAL and Cu^I–BAL.

Fig.4

Mixture A expected from the initial reaction of BAL with Cu(II) compounds (1:1 molar ratio). Polymeric or polynuclear complexes can then be formed because Cu(I) has free coordination sites. The chelate i or a polymer thereof (BAL:copper 1:1) can also be trapped in the mixture A to give mixed valence products.

2.6 Reactions of BAL with some Cu(I) compounds

In a centrifuge tube containing a solution of BAL (248 mg, 2 mmol) in deaerated methanol (7 mL) was added solid CuCl (396 mg, 4 mmol) and the dark green suspension was stirred at RT for 30 h. TLC (Et₂O) of the supernatant showed traces of oxidized BAL. Centrifugation, washing with methanol (4×3 mL) by stirring for 5 min each time (the last washing gave an opalescent supernatant) and drying in vacuo over P₂O₅ gave a dark green solid assumed to be Cu^I₂–BAL (454 mg, 91%) insoluble in boiling MeCN and warm DMSO. M.p. 198°C shrinks and turns black. Calculated for C₃H₆OS₂Cu₂ (M_r 249.28): C 14.45, H 2.43, S 25.72%; found C 14.42, H 2.45, S 25.06%. IR (KBr): Similar to that described in Section 2.5. Its powder X-ray spectrum showed absence of crystallinity (see Fig. 3). CuCl and BAL under 1:1 and 1:2 molar ratios after 18 h gave Cu^I₂–BAL and BAL that had not reacted was detected by TLC. The product Cu^I₂–BAL keeps tenaciously methanol and if not dried over P₂O₅ it was analyzed as the green methanolate: C₃H₆OS₂Cu₂·2CH₃OH (M_r 313.36): C 19.16, H 4.50, S 20.46%; found C 20.20, H 3.19, S 21.02% and its IR (KBr) spectrum now had a stronger –OH band.

Cu₂O and BAL (1:1 molar ratio) did not react in methanol (105% Cu₂O recovered), while (Ph₃P)₃Cu-Cl and BAL (2:1 molar ratio) in toluene at 110°C for 6 h gave a greenish light suspension from which on cooling to room temperature 33% (Ph₃P)₃Cu-Cl was recovered.

2.7 Reaction of BAL with cuprous mercaptides

2.7.1 Reaction of BAL with Cu–SPh

To a suspension of Cu–SPh (30 mg, 0.17 mmol) in deaerated MeOH (2 mL), BAL (17μL, 0.17 mmol) was syringed in and the greenish suspension stirred at RT for 2 days. TLC (Et₂O) showed the presence of BAL, traces of oxidized BAL and PhSH/PhSSPh while TLC (petroleum ether) showed the presence of PhSH and/or PhSSPh at R_f 0.43. Centrifugation and washing with MeOH gave a green powder (30 mg) that by IR (KBr) contained sharp bands due to Cu–SPh. The same IR (KBr) spectra were obtained from Cu–SPh and BAL under 2:1 and 1:2 molar ratios.

2.7.2 Reaction of BAL with Cu^I-MPD

To a suspension of the mercaptide (176 mg, 1.032 mmol) in deaerated water (1 mL) was added a solution of BAL (64 mg, 0.516 mmol) in deaerated methanol (2 mL) and the greenish suspension stirred at RT for 5 h. TLC showed only MPDH at R_f 0.42. Centrifugation and washing of the green precipitate by methanol (3×3 mL) by stirring for 5 min each time removed all MPDH. After drying in vacuo, Cu^I₂–BAL·2MeOH as a green powder (156 mg, 97%) was obtained, insoluble in DMSO. M.p.: at 185°C darkens and at 0–202°C shrinks and turns black. IR (KBr): same as described in Section 2.5.

When the reaction was run in deaerated methanol it took 24 h for BAL to react, while when a solution of MPDH and solid CuCl in deaerated water was treated with BAL (2:2:1 molar ratio) no BAL was detected by TLC after 30 min. Centrifugation and washing with methanol (2×3 mL) to extract MPDH and HCl gave the product Cu^I₂–BAL·2MeOH (64 mg, 100%), insoluble in DMSO and having the same IR and thermal behavior as the one obtained above.

Running the above reactions under 1:1 molar ratios, BAL was detected by TLC and Cu^I₂–BAL·2MeOH was obtained in nearly 100% yields. The in situ preparation and reaction of Cu^I-MPD with BAL under 1:1:1 and 1:1:2 molar ratios again gave Cu^I₂–BAL·2MeOH in nearly 100% yields.

2.7.3 Reaction of BAL with Cu^I–cysteinate

To a suspension of Cu^I–cys (0.5 mmol) in deaerated water (3 mL), BAL (0.5 mmol) was syringed in and the greenish suspension was stirred at RT for 2 days. TLC (Et₂O) showed traces of BAL and an intense spot at R_f 0.0 due to cysteine. Because the product did not settle on centrifugation, it was filtered (sintered glass, porosity 4, very slow filtration rate with a water aspirator), washed with deaerated water, and dried in vacuum to give the product assumed to be isomers of Cu^I–BAL (78 mg, 84%) as a greenish powder insoluble in warm DMSO. M.p.: at approx. 165°C starts darkening, at approx. 185°C turns brown-black and at 1–203°C shrinks and turns black. Calculated for C₃H₇OS₂Cu (M_r 186.75): C 19.29, H 3.78, S 34.34%; found C 19.04, H 3.56, S 32.85%. IR (KBr): ṽ = 3382 vs, broad, 2918 s, 2850 s, 1630 m, 1408 m, 1258 m, 1220 w, 1050 s, 1004 s, 884 w, 616 w. The powder X-ray spectrum of the solid showed a total absence of crystallinity.

2.7.4 Reaction of BAL with Cu^I-glutathionate

To a suspension of Cu^I–SG (0.38 mmol) in deaerated water (2 mL), BAL (0.38 mmol) was syringed in and the greenish suspension that was formed at once was stirred at RT for 2 days and worked up as above. The product as isomers Cu^I–BAL (89%) was a greenish powder having the same solubility and thermal characteristics, IR (KBr) spectrum and analyzed as C₃H₇OS₂Cu indicating the same product as that obtained from Cu^I–cys of 2.7.3 but different from that obtained from Cu^I-MPD.

Cu^I–SG formed in situ from the reaction of GSH and CuCl₂·2H₂O, Cu(AcO)₂·H₂O, cis-Cu(gly)₂·H₂O or Cu₂O in the presence of added NaCl in water, reacted with BAL (1:1 molar ratio) giving greenish suspensions. The isolated greenish solids were by weight isomers Cu^I–BAL, had the same IR (KBr) spectra as the solid obtained from solid Cu^I–SG and their powder X-ray spectra revealed no crystallinity.

2.8 Oxidation of BAL

The oxidation of BAL with 0.88 M H₂O₂ (2:1 molar ratio) in methanol was complete after approx. 30 h when TLC (Et₂O) showed no BAL at R_f 0.79 but only spots at R_f 0.65 (main) and R_f 0.0 (minor). Evaporation and drying in vacuum gave 100% of a clear, colorless, very viscous oil. Its IR (neat) spectrum revealed the presence of –OH (3354 cm^–1, very strong, broad), –SH (2550 cm^–1) and –CH₂–OH (1038 cm^–1, very strong), while the ¹H NMR spectra in CD₃OD and DMSO-d₆ showed six signals that could not positively assigned and the ¹³C NMR spectra in the same solvents revealed the presence of many signals grouped as 3–5 bundles of singlets that obviously indicate many closely related compounds. The ESI mass spectrum (positive mode) in methanol did not show the presence of a {dimer·Na⁺} at 269 but the presence of {trimer·Na⁺} at 391, {tetramer·Na⁺} at 513 and {pentamer·Na⁺} at 647.5.

The oxidation of BAL in CDCl₃ with I₂ till the color persisted showed that it was incomplete by ¹H NMR, while the oxidation of BAL with I₂ (1:1 molar ratio) either in Et₂O or CHCl₃ gave colored solids (insoluble in ether, chloroform and acetone) and supernatants whose TLC (Et₂O) showed a spot at R_f 0.95 and not at 0.65. Chromatographic isolation of the product(s) at R_f 0.95 revealed very complex ¹H and ¹³C NMR spectra that could not be analyzed. The oxidation of BAL by O₂ was very slow and incomplete the product running on TLC (Et₂O) at R_f 0.56 and a spot at R_f 0.0 was present as well.

The reaction of the colorless oil with various copper compounds will be summarized in the Section 3.4.

3 Results and discussion

3.1 Selection and preparation of some starting compounds

Of the four monothiols used for this study, thiophenol and 3-mercapto-1,2-propanediol (MPDH) are useful control compounds while cysteine and glutathione are cellular constituents. These monothiols, on reaction with Cu(I) and Cu(II) compounds should afford the mercaptides Cu^I-thiophenolate (Cu–SPh), Cu^I-1,2-propanediol-3-mercaptide (Cu^I-MPD), Cu^I–cysteinate (Cu^I–cys) and Cu^I-glutathionate (Cu^I–SG), respectively. The water-soluble Cu(II) compounds were chosen in order to study their reactivities and the effect of the anion on the solubility of the mercaptides. Of the Cu(I) compounds used in this study, CuCl and Cu₂O can reveal reactivities and, especially, the effect of Cl^– on the solubility of cuprous mercaptides, while (Ph₃P)₃Cu–Cl might afford a derivative, (Ph₃P)₂Cu–SR, suitable for X-ray determination. The complex (Ph₃P)₂Cu(NO₃) was not studied in detail because the HNO₃ produced may partly oxidize the thiol and/or the cuprous mercaptide as has been shown in a previous study with Bi(III) thiolates [16].

Thiophenol reacted with the Cu(II) and Cu(I) compounds to form the yellow [1] polymeric [32] Cu–SPh. The optimum preparation of Cu–SPh in terms of its stability on storage [13] was from Cu₂O. CuCl₂·2H₂O and CuCl, in methanol, produced yellow gels, while (Ph₃P)₃Cu–Cl and (Ph₃P)₂Cu(NO₃) in CHCl₃/MeOH reacted only in the presence of an equivalent amount of NaOH, which scavenged the HCl or HNO₃, to give moderate yields of Cu–SPh.

3-Mercapto-1,2-propanediol reacted with CuCl in deaerated water to give a yellow solution from which the yellow mercaptide was obtained by adding methanol. The isolated Cu^I-MPD was virtually insoluble in water suggesting that the HCl solubilized it during its preparation by binding to Cu(I), because Cu(I) has free coordination site(s). The product as a solid and in solution was not stable (see Section 2.2). The probable product(s) of the autoxidation of Cu^I-MPD will be discussed in Section 3.4.

Cysteine reacted with the four Cu(II) compounds under 2:1 molar ratios of cysteine to Cu(II) according to Equations (1) and (2) [24], and formed an inseparable mixture of Cu^I–cys and cystine, both insoluble in water. The tint of the solid mixture indicated autoxidation of Cu^I–cys. Pure, off-white to yellowish Cu^I–cys was obtained from water with CuCl plus NaOH at room temperature or Cu₂O at 100°C. CuCl gave low yields of Cu^I–cys·HCl, (Ph₃P)₃Cu-Cl in methanol under reflux did not react, and (Ph₃P)₂Cu(NO₃) in methanol gave colored suspensions indicating oxidation by O₂ and/or HNO₃ [16]. It is of interest that the titration of cysteine by ${CuCl}_{2}^{-} / {CuCl}_{3}^{2 -}$ gives Cu^I(cys)_1.2 which is a polymer (with bridging cysteine sulfur), that was not reactive towards CuCl₂ [34].

Glutathione and CuCl₂·2H₂O (2:1 molar ratio) in water reacted at once to give a clear colorless solution, indicating that the product, Cu^I–SG, and HCl interacted and solubilized the water insoluble Cu^I–SG. On neutralizing the HCl with NaOH, a gel was formed from which no product could be isolated. However, CuSO₄·5H₂O gave a white suspension from which low yield of Cu^I–SG was obtained, again indicating some solubilization in the presence of H₂SO₄. CuCl and GSH (1:1 molar ratio) in water gave a gel for Cu^I–SG·HCl that did not settle on centrifugation. However, under a 1:2 to 1:8 of CuCl/GSH stoichiometries a soluble polymeric complex, Cu^I(SG)_1.2, has been obtained where each Cu(I) is coordinated to three sulfur atoms [9]. The water soluble Cu^I(cys)_1.2 and Cu^I(SG)_1.2 seems to be equivalent with the water soluble ternary complexes [25, 41] mentioned in the Introduction. A complex Cu^I(GS)₂ has also been reported [3]. As with cysteine, GSH did not react with (Ph₃P)₃Cu-Cl in refluxing MeOH and from the clear solution on cooling 60% (Ph₃P)₃Cu-Cl was recovered. Pure and not autoxidized Cu^I–SG was obtained using Cu(AcO)₂·H₂O, cis-Cu(gly)₂·H₂O or Cu₂O. The product (white solid) was insoluble in water and warm DMSO, had no sharp m.p., its IR (KBr) was not informative and the only criterion of purity was its elemental analyses. Glutathione, by powder X-ray analysis, was crystalline, but the product Cu^I–SG was non-crystalline.

3.2 The reaction of BAL with various Cu(II) and Cu(I) compounds

The two reports [26, 44] on the interaction of Cu(II) and BAL under different conditions were mentioned in the Introduction. In both cases the oxidation state of copper [Cu(I) or Cu(II)] and of BAL (reduced or oxidized) was not reported. Based on Equations (1) and (2) the solid obtained [44] should contain Cu(I) and oxidized BAL.

The reaction of BAL with CuCl₂·2H₂O and with CuSO₄·5H₂O (1:1 molar ratios) in H₂O/MeOH gave dark blue to grey solvated suspensions that settled on centrifugation. In the supernatant liquid, BAL was not detected by TLC. The yields were 100% (based on the mixture A of Fig. 4). Running the same reactions in phosphate buffer pH 7.0 under nitrogen, dark blue to grey solvates were also been formed and the solids isolated in 81% yields (based on mixture A of Fig. 4) due to opalescent washing supernatants. Their identical IR (KBr) spectra did not contain phosphate groups. The dark blue to grey colors of the dried solids in all preparations described contrasted with the green color previously reported for this reaction [44]. Elemental analyses indicated the presence of two Cu⁺ and one oxidized (dimeric) BAL. The solids had relatively simple IR (KBr) spectra featuring a very strong, broad band at 3334 cm^–1 due to O–H stretching and two very strong characteristic bands at 1042 and 1006 cm^–1 also seen in BAL as a film due to C–O stretching. The powder X-ray spectra, Fig. 3, of the products as mixture of isomers (mixture A; Fig. 4) showed total absence of crystallinity. BAL plus Cu(AcO)₂·H₂O and BAL plus cis-Cu(gly)₂·H₂O (1:1 stoichiometry), respectively, gave dark blue solvates that did not settle on centrifugation indicating that simple carboxylic or amino acids coordinated with Cu(II) in vivo can be displaced by BAL, with concomitant reduction of Cu(II) to Cu(I).

The solids obtained from BAL and Cu(II) compounds are thought to be mixtures based on the following considerations. Addition of CuL₂ to BAL should give the monothiolate which is not reduced but, because of the proximity of a second –SH group, will give a chelated Cu(II) dithiolate, i. This compound cannot disproportionate to Cu(0) and a four-membered ring C₂S₂, because the ring decomposes spontaneously by forming S₂ [38]. Nor can the five-membered ring i be stabilized in solution, by analogy with the unstable adduct of BAL with non-reducible Hg(II), which produces a polymer [7]. Intramolecular electron transfer from sulfur to Cu(II) in i could give primary and/or secondary thiyl radicals, which on recombination conceivably could initially afford three products (mixture A) containing oxidized dimeric BAL and two Cu⁺ i.e. 1:1 stoichiometry of Cu/BAL (Fig. 4). Because Cu(I) has free coordination sites there is a chance for the formation of polynuclear complexes, the nature of which could not be revealed due to (total) insolubilities of the solids. The chelate Cu^II–BAL i, or a polymer thereof, could conceivably be trapped in the insoluble mixture A to give mixed valence products all having an 1:1 ratio of copper to BAL. Mixed valence insoluble solids have been isolated in the case of reducible Au(III) compounds reacting with BAL [17].

Dihydro-α-lipoic acid (6,8-dimercaptooctanoic acid) and its amide have α,γ-SH groups. On treatment with CuCl in DMF, both compounds gave bis-copper(I) salts as yellow solids, which were presumably insoluble in organic solvents, and were not analyzed further [5]. BAL, which has α,β-SH groups, on reaction with CuCl [1:2 molar ratio of BAL to Cu(I)] in methanol gave the bis-copper(I) salt Cu^I₂–BAL as a dark green solid insoluble in organic solvents, which has the same IR (KBr) spectrum as the product obtained from CuCl₂·2H₂O or CuSO₄·5H₂O and BAL (1:1 molar ratio). With excess BAL again the salt Cu^I₂–BAL was produced. The powder X-ray spectra of the solids of various runs showed total absence of crystallinity. Cu^I₂–BAL retains tenaciously methanol forming the green methanolate Cu^I₂–BAL·2MeOH. No reaction between Cu₂O and stoichiometric or excess BAL in methanol was observed at room temperature, while BAL and (Ph₃P)₃Cu-Cl in toluene at 110°C gave a small amount of a green solid and recovered (Ph₃P)₃Cu-Cl. Reichle [32] obtained the Cu(II) salt of a dithiol by refluxing (Ph₃P)₃Cu-Cl with the disodium salt of a phenyl-1,2-dithiol.

3.3 The reaction of BAL with cuprous mercaptides

Although BAL and (Ph₃P)₂Cu–SPh (1:1 molar ratio) in CHCl₃ reacted (by TLC), no pure product could be isolated corresponding to (Ph₃P)₂Cu–SCH₂CH(SH)CH₂OH. BAL reacted incompletely (by TLC) with Cu–SPh (2:1, 1:1 or 1:2 molar ratios) in methanol because the green solids obtained were mainly Cu–SPh (by IR). It seems that BAL reacted at the surface of the granules and the green product passivated the core Cu–SPh.

In contrast to Cu–SPh, BAL did extract Cu(I) from Cu^I-MPD under 1:1 and 2:1 molar ratios of mercaptide to BAL giving the green methanolate Cu^I₂–BAL·2MeOH. The reaction was slow in MeOH, faster in H₂O/MeOH 1:2 v/v and even faster in H₂O. In these cases, too, excess BAL gave the bis-copper(I) and not the mono-copper(I) salt.

If Cu^I–cys and Cu^I–SG in any form are cellular constituents, then the possibility arises that their copper(I) ion may be sequestered by BAL. Experiments using Cu^I–cys and Cu^I–SG as suspensions in deaerated water showed that they both reacted at once with added BAL under 1:1 stoichiometry giving greenish suspensions that did not settle on centrifugation. The greenish products, isolated by filtration, were analyzed as the adduct Cu^I–BAL, and were most likely mixtures of isomers, with Cu^I binding to either α or β –SH groups. Their IR (KBr) spectra did not show bands due to cysteine and glutathione and we did not detect a band for free –SH at approx. 2550 cm^–1 the spectrum of the latter being identical to that of Cu^I₂–BAL·2MeOH. The same results were obtained when the white suspension of Cu^I–SG formed i n situ from GSH and Cu(AcO)₂·H₂O, cis-Cu(gly)₂·H₂O or Cu₂O in water containing added NaCl was treated with BAL under 1:1 stoichiometry.

The observation that BAL and Cu^I-MPD (1:1) gave Cu^I₂–BAL, while BAL and Cu^I–cys (1:1) or BAL and Cu^I–SG (1:1) gave Cu^I–BAL could not be explained.

3.4 Oxidation of BAL by O₂, I₂ and H₂O₂ and reaction of the products with Cu(I) and Cu(II) compounds

In order to verify the production of the disulfides of Fig. 4, BAL was treated with O₂, I₂ and H₂O₂, respectively. Complete oxidation only occurred with H₂O₂.

The oxidation of BAL by H₂O₂ (1:1 molar ratio) in methanol for 17 h gave a glassy mass insoluble in methanol and chloroform. A 2:1 molar ratio of BAL to H₂O₂ stirred for 30 h at room temperature afforded a clear, colorless oil that had quite complex ¹H and ¹³C NMR spectra in CD₃OD and DMSO-d₆ solvents. Its mass spectrum did not show the disulfide dimers (precursors of mixture A of Fig. 4) but trimer(s), tetramer(s) and pentamer(s) arising by reaction of the dimer(s) with BAL under these oxidizing conditions. In the presence of Cu(I), however, the dimers of Fig. 4 can be captured by Cu(I), probably not allowing the formation of oligomers to a significantextent.

The reaction of this oily product with the water soluble CuCl₂·2H₂O gave a grey colored solid, while the products from the water insoluble CuCl, Cu^I–cys and Cu^I–SG, were greenish. The IR (KBr) spectra of the solids from CuCl₂ and CuCl were quite similar but those from Cu^I–cys and Cu^I–SG were dominated by the cysteine and glutathione bands. In all cases the solids were water (and DMSO) insoluble that can not be excreted in urine.

3.5 Attempted solubilization of Cu^I–BAL compounds

3.5.1 With urea/NaCl

Because patients with Wilson’s disease treated with BAL showed elevated copper levels in their urine [27, 30], and because the main constituents of urine are urea (9.3 g/L), Cl^– (1.87 g/L) and Na⁺ (1.17 g/L) with K⁺ (0.75 g/L), we investigated the possibility of solubilization of BAL/Cu(I) solids by urea and NaCl.

The reaction of BAL with CuSO₄·5H₂O (0.5 mmol each) in deaerated water (5 mL) in the presence of 1.55×0.5 or 15.5×0.5 mmol urea and then adding 0.55×0.5 or 5.5×0.5 mmol NaCl gave dark blue to grey solvated suspensions from which 2–99% blue hard solids were isolated, which when powdered gave green powders. Their IR (KBr) spectra were identical featuring absence of –SH and presence of –OH and –CH₂–OH bands. The same results were obtained reversing the mode of addition: first NaCl and then urea.

Attempted solubilization of various solids assumed to be Cu^I₂–BAL·2MeOH and mixtures of Fig. 4 by urea followed by NaCl and stirring at room temperature for 24 h all produced green suspensions that settled easily giving, after drying, dark green solids in nearly quantitative recoveries the IR (KBr) of which were dissimilar and resembled those of the starting solids.

3.5.2 With excess BAL

With excess BAL to Cu(II) compounds, a more complicated mixture of products can be obtained. Thus, performing the reactions of CuCl₂·2H₂O and CuSO₄·5H₂O with excess BAL (1:2 molar ratios) in deaerated water green colored solvated suspensions were formed. The green solids isolated had identical IR (KBr) spectra with the products obtained from BAL and CuCl₂·2H₂O or CuSO₄·5H₂O under 1:1 stoichiometries.

3.5.3 With excess Cu(II) compounds

Attempted solubilization of mixtures of Fig. 4 by adding CuSO₄·5H₂O in water resulted in nearly quantitative recovery of the dark blue to grey starting solid (by weight and IR). However, with CuCl₂·2H₂O the recovery of the dark green to grey solid was >100% and its IR (KBr) spectrum was different from that of the starting indicating a probable involvement of chloride ions.

Summarizing, qualitative experiments on the solubilization of the insoluble solids: Cu^I₂–BAL and mixtures of Fig. 4 gave negative results and, therefore, the increase of copper excretion in the urine in the microgram/hour range [27] of patients with Wilson’s disease is not likely to be due to these binary Cu^I/BAL compounds and the enhanced excretion must follow another mechanism that presently remains unknown. A similar mechanism can lead to the detoxification [39] of the water insoluble [17] Pb–BAL·H₂O.

3.6 The structure and stability of cuprous mercaptides and of the copper(I)/BAL compounds

The simple cuprous thiolates, such as Cu–SPh, are yellow [1,13, 1,13] and are polymeric due to secondary Cu…S bonds [9 and references therein]. In yellow cuprous–thiolate multimetallic clusters, e.g. [Cu₄(SPh)₆]^2–, the Cu…Cu distances are approx. 270 pm [29 and references therein]. Functionalized monothiols, such as 3-mercapto-1,2-propanediol, cysteine and glutathione gave yellow, yellowish and nearly colorless cuprous mercaptides, respectively. The α,γ-dithiols: dihydrolipoic acid and its amide also gave yellow insoluble bis-copper(I) mercaptides [5] but with the α,β-dithiol (BAL) the insoluble copper(I) compounds, such as Cu^I₂–BAL, Cu^I₂–BAL·2MeOH, isomers of Cu^I–BAL or mixtures of Fig. 4, were not yellow but dark green or dark blue to grey colored solids probably indicating Cu…Cu interactions. However, mixed valence complexes cannot a priori be excluded if they contain Cu^II–BAL, i, of Fig. 4 or its polymers. Therefore, the real structure of the these highly insoluble solids is not known.

It was reported that the solid obtained from CuSO₄ and BAL at high dilution, apparently remained (visually) unchanged after several months [44], but as early as 1931 [13], it was found that simple cuprous mercaptides were not stable to storage, obviously being autoxidized. The nature of the products has not been reported. An attempt was, therefore, made to identify the site of attack of O₂ (Cu⁺ and/or the organic part of a mercaptide) by ¹H NMR using the easily autoxidized Cu^I-MPD in DMSO-d₆.

The ¹H NMR spectrum of MPDH in DMSO-d₆ [15] showed sharp carbon-hydrogen and sulfur-hydrogen but broad oxygen-hydrogens signals due to hydrogen bonding to DMSO. A yellow solution of Cu^I-MPD in DMSO-d₆, run just after dissolution, Fig. 2A, did not show the –SH proton (at 2.06 ppm). The carbon-hydrogen signals became a little broad and resonated approx. 0.1 ppm downfield compared to those of MPD except those of diastereomeric –CH₂SCu which resonated approx. 0.25 ppm downfield indicating bonding of Cu^I to sulfur. However, the –CH₂OH and >CHOH protons were very broad (reaching the base line) and resonated at 4.70 and 5.50 ppm, respectively, probably indicating a rearranging process the nature of which could not be established because of a quick development of a brown coloration due to production of Cu²⁺. The brown solutions obtained in DMSO-d₆ under various conditions, Fig. 2B and 2C, differed from the yellow one in that the carbon-hydrogen signals were broader but without shifting indicative of the presence of Cu²⁺ which causes long relaxation times [23]. Thus, the brown coloration indicates preferential attack of O₂ at the Cu⁺ and not at –S^– or –CH₂OH to give a superoxo (RSCu^II-O₂·) or more likely [22] a peroxo (RSCu^II-O-O-Cu^IISR) product. The paramagnetic Cu²⁺ can then affect the ¹H NMR spectra of both the peroxo product and the starting Cu^I-MPD.

4 Conclusions

In an attempt to understand the pharmacology of BAL in the treatment of Wilson’s disease, BAL was treated with a variety of Cu(II) and Cu(I) compounds. BAL with CuCl₂·2H₂O and CuSO₄·5H₂O, respectively, gave an insoluble mixture of compounds assumed to be A or polynuclear complexes thereof and probably of mixed valence complexes containing i or polymers thereof with a ratio 2Cu⁺/oxidized dimeric BAL or Cu(II)/BAL 1:1 (Fig. 4), while with Cu(AcO)₂·H₂O and cis-Cu(gly)₂·H₂O solvates were obtained that did not settle on centrifugation. With CuCl, but not with the non-reactive Cu₂O, the product Cu^I₂–BAL and its methanolate were obtained. BAL passivated Cu–SPh but completely extracted Cu(I) from Cu^I-MPD, Cu^I–cys and Cu^I–SG giving a water-insoluble Cu^I₂–BAL·2MeOH and Cu^I–BAL. Thus, BAL formed water-insoluble compounds with Cu(II) and Cu(I) as well as with the mercaptides Cu^I-MPD, Cu^I–cys and Cu^I–SG that, logically, could not be excreted in urine as binary copper/BAL compounds. Also, the Cu^I/BAL compounds were not solubilized by the main constituents of urine (urea/NaCl), excess BAL or excess Cu(II) compounds. Overall, the increase of urinary copper levels when Wilson’s disease patients are treated with BAL may not be due directly to the solubilization of copper by BAL and another, unknown, mechanism must be involved. The same mechanism will probably operate for the removal of lead from the body [39]. The autoxidation product of a copper(I) mercaptide is probably a peroxo compound containing Cu^II as deduced from the ¹H NMR spectra of the easily autoxidized Cu^I-MPD. More work on this subject is obviously required.

Footnotes

Acknowledgments

We thank Professor P.G. Koutsoukos of the Department of Chemical Engineering, University of Patras, for running the powder X-ray spectra.

References

Adams

, Reifschneider

and Nair

M.D.

, The preparation of aryl thioethers by reaction of aromatic halogen compounds with cuprous thiophenolate or cuprous thiobutylate, Croatica Chem Acta 29 (1957), 277–285.

Ala

, Walker

A.P.

, Ashkan

, Dooley

J.S.

and Schilsky

M.L.

, Wilson’s disease, Lancet 369 (2007), 397–408.

Aliaga

M.E.

, López-Alacrón

, Bridi

and Speisky

, Redox-implications associated with the formation of complexes between copper ions and reduced or oxidized glutathione, J Inorg Biochem 154 (2015), 77–88.

Birker

P.J.M.W.L.

and Freeman

H.C.

, Structure, properties, and function of a copper(I)-copper(II) complex of D-penicillamine: Pentathallium(I) μ₈-chlorododeca(D-penicillaminato)-octacuprate(I)hexacuprate(II)-hydrate, J Am Chem Soc 99 (1997), 6890–6899.

Bonomi

, Pagani

, Cariati

, Pozzi

, Grisponi

, Cristiani

, Diaz

and Zanoni

, Synthesis and characterization of metal derivatives of dihydrolipoic acid and dihydrolipoamide, Inorg Chim Acta 192 (1992), 237–242.

Buzuk

, Brinić

, Vladislavić

, Bralić

, Buljac

and Škugor Rončević

, Real-time monitoring the “self-oxidation” of cysteine in presence of Cu²⁺: Novel findings in the oxidation mechanism, Monatsh Chem 147 (2016), 359–367.

Canty

A.J.

and Kishimoto

, Mercury(II) and organomercury(II) complexes of thiols and dithiols, including British Anti-Lewisite, Inorg Chim Acta 24 (1977), 109–122.

Catsch

and Harmuth-Hoene

A.-E.

, New developments in metal antidotal properties of chelating agents, Biochem Pharmacol 24 (1975), 1557–1562.

Corazza

, Harvey

and Sadler

P.J.

, ¹H, ¹³C-NMR and X-ray absorption studies on copper(I) glutathione complexes, Eur J Biochem 236 (1996), 697–705.

10.

Costa

, Reisenhofer

and Stefani

, Complexes of copper(I) with triphenylphosphine, J Inorg Nucl Chem 27 (1965), 2581–2584.

11.

Cotton

F.A.

and Goodgame

D.M.L.

, Tetrakis(triphenylphosphine-silver(I) and copper(I) complexes, J Chem Soc (1960), 5267–5269.

12.

Downes

J.M.

, Whelan

and Bosnich

, Biological analogues. Spectroscopic characteristics of mercapto- and disulfide-copper(II) coordination in relation to type I proteins, Inorg Chem 20 (1981), 1081–1086.

13.

Duncan

W.E.

, Ott

and Reid

E.E.

, Some copper mercaptides and their reaction with carbon disulfide, Ind Eng Chem 23 (1931), 381–384.

14.

Gysling

H.J.

, Coordination complexes of copper(I) nitrate, I norg Synthesis 19 (1979), 92–97.

15.

http://sdbs.db.aist.go.jp; Accessed 15.8.2017.

16.

Ioannou

P.V.

, Dioxygen activation by Bi(III) compounds: Air oxidation of thiols to their disulfides, Main Group Chem 12 (2013), 125–138.

17.

Ioannou

P.V.

, Vachliotis

D.G.

and Chrissanthopoulos

, Chelation therapy: The interaction of British Anti-Lewisite (BAL) with some heavy metal cations of p and d blocks, Main Group Chem 16 (2017), 125–139.

18.

Ioannou

P.V.

, Vachliotis

D.G.

and Kouli

, The reaction of British Anti-Lewisite (dimercaprol) with As(III), Sb(III), and Bi(III) oxides and salts, Main Group Chem 12 (2013), 235–249.

19.

Jardine

F.H.

, Rule

and Vohra

A.G.

, The chemistry of copper(I) complexes. Part I. Halogeno-complexes, J Chem Soc (A) (1970), 238–240.

20.

Jardine

F.H.

, Vohra

A.G.

and Young

F.J.

, Copper(I) nitrato and nitrate complexes, J Inorg Nucl Chem 33 (1971), 2941–2945.

21.

Keller

R.N.

and Wycoff

H.D.

, Copper(I) chloride, Inorg Synthesis 2 (1946), 1–4.

22.

Kitajima

and Moro-oka

, Copper dioxygen complexes. Inorganic and bioinorganic perspectives, Chem Rev 94 (1994), 737–757.

23.

Knowles

P.F.

, March

and Rattle

H.W.E.

, Magnetic Resonance of Biomolecules, Wiley, London, 1976, 68–71.

24.

Kolthoff

I.M.

and Stricks

, Polarographic investigations of reactions in aqueous solutions containing copper and cysteine (cystine). II. Reactions in ammoniacal medium in the presence and in the absence of sulfite, J Am Chem Soc 73 (1951), 1728–1733.

25.

Königsberger

L.-C.

, Königsberger

, Hefter

and May

, Formation constants of copper(I) complexes with cysteine, pencillamine and glutathione: Implications for copper speciation in the human eye, Dalton Trans 44 (2015), 20413–20425.

26.

Majumdar

J.C.

and Haldar

B.C.

, Extraction of Cu(II) with 2,3-dimercaptopropanol (BAL) in butanol, Radiochem Radioanal Lett 48 (1981), 247–252.

27.

McCance

R.A.

and Widdowson

E.M.

, Observations of the administration of BAL-Intrav to man, Nature 157 (1946), 837.

28.

O’Brien

, The preparation and characterization of the geometric isomers of a coordination complex- and-bis glycinato copper(II) monohydrates, J Chem Educ 59 (1982), 1052–1053.

29.

Pickering

I.G.

, George

G.N.

, Dameron

C.T.

, Kurz

, Winge

D.R.

and Dance

I.G.

, X-ray absorption spectroscopy of cuprous-thiolato clusters in proteins and model systems, J Am Chem Soc 115 (1993), 9498–9505.

30.

Purchase

, The treatment of Wilson’s disease, a rare genetic disorder of copper metabolism, Sci Prog 96 (2013), 19–32.

31.

Purchase

, The link between copper and Wilson’s disease, Sci Prog 96 (2013), 213–223.

32.

Reichle

W.T.

, Preparation, physical properties and reactions of copper(I)-triphenyl-M complexes (M=P, As, Sb), Inorg Chim Acta 5 (1971), 325–332.

33.

Remy

, (translated by Anderson

J.C.

and Kleinberg

, Treatise on Inorganic Chemistry Amsterdam, Elsevier.II (1956), 379.

34.

Rigo

, Corazza

and di Paolo

M.L.

, Rossetto

, Ugolini

and Scarpa

, Interaction of copper with cysteine: Stability of cuprous complexes and catalytic role of cupric ions in anaerobic thiol oxidation, J Inorg Biochem 98 (2004), 1495–1501.

35.

Scheinberg

I.H.

, Sternlieb

, Schilsky

and Stockert

R.J.

, Penicillamine may detoxify copper in Wilson’s disease, Lancet 330 (1987), 95.

36.

Sigiura

and Tanaka

, Studies on the sulfur-containing chelating agents. XXV. Chelate formation of penicillamine and its related compounds with copper(II), Chem Pharm Bull 18 (1970), 368–373.

37.

Sigiura

and Tanaka

, Evidence for a ternary complex containing albumin, copper and penicillamine, Mol Pharmacol 8 (1972), 249–255.

38.

Steliou

, Salama

, Brodeur

and Gareau

, Diatomic sulfur, J Am Chem Soc 109 (1987), 926–927.

39.

Stocken

L.A.

and Thompson

R.H.S.

, Reactions of British Anti-Lewisite with arsenic and other metals in living systems, Physiol Rev 29 (1949), 168–192.

40.

The Merck Index, 15th edn., Royal Society of Chemistry, Cambridge, 2013 p. 584, entry 3231.

41.

Tran-Ho

L.-C.

, May

P.M.

and Hefter

G.T.

, Complexation of copper(I) by thioamino acids. Implication for copper speciation in blood plasma, J Inorg Biochem 68 (1997), 225–231.

42.

Vilensky

J.A.

and Redman

, British Anti-Lewisite (dimercaprol): An amazing history, Ann Emerg Med 41 (2003), 378–383.

43.

Walshe

J.M.

, Endogenous copper clearance in Wilson’s disease: A study of the mode of action of penicillamine, Clin Sci 26 (1964), 461–469.

44.

Webb

E.C.

and van

, Heyningen, The action of British Anti-Lewisite (BAL) on enzyme systems, Biochem J 41 (1947), 74–78.

45.

Wilson

S.A.K.

, Progressive renticular degeneration: A familial nervous disease associated with cirrhosis of the liver, Brain 34 (1912), 295–509.

46.

Xie

J.-J.

and Wu

Z.-Y.

, Wilson’s disease in China, Neurosci Bull 33 (2017), 323–330.

Interaction of British Anti-Lewisite (BAL) with Copper(I) and Copper(II) compounds in conjunction with Wilson’s disease

Abstract

Keywords

1 Introduction

2.1 Materials and methods

2.2 Preparation of cuprous 2,3-dihydroxypropyl-1-mercaptide, Cu I -MPD

2.4 Reactions of glutathione with copper(II) and copper(I) compounds. Preparation of cuprous glutathionate, Cu I –SG

2.5 Reactions of BAL with some Cu(II) compounds

2.7 Reaction of BAL with cuprous mercaptides

2.7.1 Reaction of BAL with Cu–SPh

2.7.2 Reaction of BAL with Cu I -MPD

2.7.3 Reaction of BAL with Cu I –cysteinate

2.7.4 Reaction of BAL with Cu I -glutathionate

2.8 Oxidation of BAL

3 Results and discussion

3.1 Selection and preparation of some starting compounds

3.2 The reaction of BAL with various Cu(II) and Cu(I) compounds

3.3 The reaction of BAL with cuprous mercaptides

3.4 Oxidation of BAL by O2, I2 and H2O2 and reaction of the products with Cu(I) and Cu(II) compounds

3.5 Attempted solubilization of Cu I –BAL compounds

3.5.1 With urea/NaCl

3.5.2 With excess BAL

3.5.3 With excess Cu(II) compounds

3.6 The structure and stability of cuprous mercaptides and of the copper(I)/BAL compounds

4 Conclusions

Footnotes

Acknowledgments

References

2.2 Preparation of cuprous 2,3-dihydroxypropyl-1-mercaptide, Cu^I-MPD

2.4 Reactions of glutathione with copper(II) and copper(I) compounds. Preparation of cuprous glutathionate, Cu^I–SG

2.7.2 Reaction of BAL with Cu^I-MPD

2.7.3 Reaction of BAL with Cu^I–cysteinate

2.7.4 Reaction of BAL with Cu^I-glutathionate

3.4 Oxidation of BAL by O₂, I₂ and H₂O₂ and reaction of the products with Cu(I) and Cu(II) compounds

3.5 Attempted solubilization of Cu^I–BAL compounds