Molecular-Modeling Design of Cadmium-Mobilizing Agents: A Novel

In Vivo Cadmium Mobilization by Three Novel Bis(carbodithioates). Pramod K. Singh and Mark M. Jones, Krista Kostial, Maja Blanuša, and Martina Piasek...
0 downloads 0 Views 2MB Size
Download PDF
Chem. Res. Toxicol. 1994, 7, 614-620

614

Molecular-ModelingDesign of Cadmium-Mobilizing Agents: A Novel Biscarbodithioate Pramod K. Singh, Cunyong Xu, and Mark M. Jones* Department of Chemistry and Center i n Molecular Toxicology, Vanderbilt University, Nashville, Tennessee 37235

Krista Kostial and Maja Blanusa Department of Mineral Metabolism, Znstitute for Medical Research and Occupational Health, University of Zagreb, Ksaverska cesta 2, 41001 Zagreb, Croatia Received March 15, 1994@ This report describes the design, synthesis, characterization, and in vivo cadmium-mobilizing properties of a novel biscarbodithioate chelating agent, namely, disodium N,N'-diglucosyl-1,Snonanediamine-N,N'-biscarbodithioate (CSGBDTC), HOCH~(CHOH)~CH~(CS~N~)N(CH~)SN(CS~Na)CH2(CHOH)4CH20H, which can coordinate to a single cadmium ion with both of its carbodithioate groups (CS2Na) in its folded configuration. When valuated for its cadmium efficacy at 1.0 mmoVkg x 5 ip in lo9Cd-pretreatedrats against sodium N-benzyl-D-glucamineN-carbodithioate (BGDTC) a s standard, the biscarbodithioate was found to reduce the wholebody levels of cadmium more rapidly in the rat t h a n BGDTC which contains only one CSzNa group. The whole-body Cd depletions after the first ip injection of the new and the standard compound were 52% and 23%, respectively. The CSG2DTC was found to be more effective in removing cadmium from the liver (% Cd reductions compared to controls: CSGBDTC, 94, and BGDTC, 851, but slightly less effective in reducing renal cadmium levels (% Cd-reductions: CSGBDTC, 44 , and BGDTC, 60). The ip LD50 of the bis-DTC was estimated to be slightly in excess of ca. 4.0 mmoVkg in the rat. A molecular model of this chelating agent indicates that, because of the flexibility of the nonane chain, both carbodithioate groups can approach closely enough to each other to permit complexation with the same cadmium ion to give a resulting structure without significant strain. A mechanism for the removal of Cd from CdMT by CSGBDTC is also proposed.

Introduction

The chelating agents developed for the other metals are designed to occupy all of the available coordination positions of the toxic metal ion (12-14). The monocarbodithioates occupy only two of the coordination positions on the Cd2+ion (15). One may well ask if it is possible to design a chelating agent of either of these types which could occupy four of the coordination positions around the Cd2+ ion. The present study was conducted to determine whether a chelating agent possessing two carbodithioate groups, both of which can coordinate to the same cadmium ion, would in fact enhance the in vivo Cd mobilization in comparison with a DTC having only one such group. The compound selected for examination, disodium N,N'-di-D-glucosyl-l,9-nonanediamine-N,N'-biscarbodithioate, has two such groups (Figure 1A). Because of the flexibility of the nonane chain, (CH& these carbodithioate groups can position themselves (Figure 1B) so that they can both bind to the same Cd2+ion.

It is well established that, shortly after ingestion or administration, the cadmium ion (Cd2+)which is retained by a n animal achieves intracellular sites from which its removal becomes progressively more difficult because of the subsequent formation of cadmium-metallothionein (CdMT)l complex (1). While there are a large number of chelating agents which can accelerate the excretion of cadmium when given simultaneously with or very shortly after the cadmium (2, 31, there are far fewer which are effective if the chelation therapy is delayed for 24 h or more (4). Vicinal dithiols such as 2,3-dimercapto-1propanol (BAL) and the mono- and diesters of meso-2,3dimercaptosuccinic acid (DMSA) have the ability to enhance the excretion of cadmium in animals very efficiently when given days, weeks, and even months after the cadmium intoxication (5-8). The compounds having one carbodithioate function have also been demonstrated to be effective cadmium-mobilizing agents in animal studies even when administered long after cadmium exposure (9-11). A comparison of these chelating agents with those used for several other toxic metals, e.g., Fe(II1) and Pu(IV), shows some significant differences.

Materials and Methods

* Address correspondence: to this author at Box 1583,Station B, Department of Chemistry,Vanderbilt University, Nashville, TN 37235. @Abstractpublished in Advance ACS Abstracts, August 15,1994. * Abbreviations: CSG2DTC, disodiumN~-di-~glucosyl-l,9-nonarAediamine-NJV-biscarbodithioate;BGDTC,sodium N-benzyl-D-glucamineN-carbodithioate;MT, metallothionein; CdMT, cadmium-metallothionein complex. 0893-228x/94/2707-0614$04.50/0

Preparation of the Chelating Agent. Anhydrous a-D-(+)glucose, approximate purity 96% (l),was purchased from Sigma Chemical Co. (St. Louis, MO), and l,g-diaminononane, 98% (21, was obtained from Aldrich Chemical Co. (Milwaukee, WI). Platinum(IV) oxide type D catalyst (assay 74.82%) was purchased from Aesar/Johnson-Matthey (Ward Hills, MA). Solvents were degassed with nitrogen to avoid oxidation of t h e amine. Melting points were determined on a Thomas-Hoover stirred liquid apparatus, and proton NMR spectra were recorded on a 300-MHz FT-NMR spectrometer in deuterium oxide using the HOD peak at 6 4.77 (a singlet) as standard. The chemical

0 1994 American Chemical Society

Molecular-Modeling Design of Cd Mobilizing Agents

Chem. Res. Toxicol., Vol. 7, No. 5, 1994 615

brought to 22.5 "C and left overnight to ensure complete reaction. The product C9G2DTC (4) separated out as a white solid. Acetone (100 mL) was added and shaken well, and 4 was collected on a Buchner funnel and washed with acetone (3 x 25 mL) and finally with ether (50 mL). The biscarbodithioate 4 was dried under vacuum for 24 h and stored refrigerated under nitrogen. Yield 12.20 g (87.0%) as a fairly nonhygroscopic white solid; mp 155-160 "C dec; lH NMR (D20) 6 4.45-4.23 (m, 6 H), 3.94-3.59 (m, 12 H), 1.72 (br m, 4 H), 1.31 (br s, 10 H). a Anal. Calcd for C23H44N2S4010Na2: C, 40.46; H, 6.50; N, 4.10. Found: C, 41.16; H, 6.97; N, 3.72. 1A. (Unfolded Configuration) Cd-CSGZDTC 1:l Complex (5). The preparation of the 1:l complex was based on a reported Cd-DTC complexation procedure (18).Addition of an aqueous solution (20 mL) of CdC12-2.5H20 (0.334 g, 1.464 mmol) to the stirred aqueous solution (50 mL) of the bis-DTC 4 (1.00 g, 1.464 mmol) at 22 "C over 1 h followed by an overnight stirring gave the complex 6 a s a white precipitate. Yield 0.84 g (76.6%); mp 185-190 "C dec. Anal. Calcd for C23 H44N2S4010Cd: C, 36.87; H, 5.92; N, 3.74. Found: C, 36.71; H, 5.95; N, 3.72. Simultaneous additions of dilute aqueous solutions (10 mL) each of 4 and CdCl2 in 50 mL of water while stirring at room temperature over a period of 30 min afforded the same product. 1B. (Folded Configuration) Animal Studies. (A) Toxicity Study. The in vivo toxicity of C9G2DTC was estimated using a limited number of rats. Figure 1. Molecular model of C9G2DTC showing unfolded (A) and folded (B) configurations (Na atoms are not shown; empty Male Wistar rats (Sasco, Omaha, NB), average weight 275 f circles = H, dark circles = C, circles with dots = 0, "brickwork" 25 g, were divided into 3 groups. Group 1(2 animals) received circles = N, lined circles = S). 3.5 mmol of C9G2DTCkg ip in 2 mL of deionized water; group 2 (4 animals) received 4.0 of mmol C9G2DTCkg ip in 2 mL of shifts are reported in parts per million (ppm, 6). Elemental deionized water; and group 3 (2 animals) received 5.0 mmol of analyses were performed by Atlantic Microlabs, Inc. (Norcross, C9G2DTCkg ip in 3 mL of deionized water. Lethality was monohyGA). Sodium N-benzyl-D-glucamine-N-carbodithioate observed over a period of 7 days. There were no mortalities in drate (BGDTC), which served as the standard chelator, was group 1, and only 1animal out of 4 died after 6.5 h in group 2, prepared by a reported procedure (16)and was fully characterwhereas 100% mortality occurred in group 3 after 2.5 h. The ized before animal experimentation. surviving animals appeared normal and did not show any sign Disodium N,iV'-Di-D-glucosyl-l,S-nonanediamhe-N,iV'- of prolonged apparent sickness. The ip LD50 of C9G2DTC was, biscarbodithioate (4; CSGZDTC). The synthesis was based therefore, estimated to be ca. 4.0 mmolkg or slightly higher in on the reported procedure used for the preparation of substithe rat based on the lethality and observation of the animals. tuted glucamine monocarbodithioates (11,17),as follows (Figure (B) In Vivo Cadmium Excretion. The experiment was 2). performed on 3-week-old female albino rats (average body A mixture of 20.00 g (111.00 mmol) of anhydrous a-D-(+)weight 38 g) from the breeding farm of the Institute of Medical glucose (1) and 8.78 g (55.50 mmol) of 1,9-diaminononane (2) Research (University of Zagreb, Croatia). The experiment lasted was allowed to react for 15-20 min in 6 mL of water at 50-60 13 days. On the first day each rat was injected intraperitoneally "C under N2 with mechanical stirring to give a transparent with 0.03 mg of CdCly2.5H20 in 1 mL of 0.9% NaCl solution viscous oil. This crude imine was dissolved in 150 mL of MeOH containing 2.0 pCi (74 kBq) of lo9CdC12. and then subjected to hydrogenation at 60-65 psi in the presence of PtO2 catalyst (1.00 g) for 72 h. White solid started Seven days later the rats were randomly divided into 3 groups to appear after ca. 7 h of reduction. At the the end of the of 6 animals in each for further treatment. Two groups received hydrogenation reaction the massive white solid which formed chelation therapy with BGDTC or CSGBDTC by ip injection on was dissolved in aqueous MeOH and passed through a cellulose 5 successive days (i.e., from days 7 to 11 of the experiment). filter aid to get rid of the black catalyst (Caution:the recovered Both chelators were administered at a daily dose of 1.0 mmol/ catalyst is highly flammable, and copious nitrogen flow on the kg in a 1-mL volume (cumulative dose 5.0 mmolkg). Controls top of funnel is required; the used catalyst must be stored wet received 5 ip injections of 0.9% saline (1mL) at the same time with water). The clear colorless filtrate was treated twice with intervals. decolorizing charcoal and let stand to crystallize. After 8 h at Whole-body (WB) retention of lo9Cdwas determined 5 times, 22 "C the secondary amine N,N-di-D-glucosyl- 1,g-nonanedii.e., 24 h after the first 4 treatments, and 48 h after the fifth amine (3)separated as white amorphous solid: yield 10.7 g treatment, i.e., at the end of the experiment when the rats were (37%) I crop; mp 145-147 "C; lH NMR (D2O) 6 3.70-3.58 (m, sacrificed (under ether anaesthesia) by cardiac exsanguination. 12 H, CHOH and CH20H), 2.74-2.56 (m, 8 H, CHzNCHz), 1.60After dissecting the organs, the retention of lo9Cd was deter1.70 (m, 4 H, alkyl H2 at C-2 and C-81, 1.82 [br s, 10 H, alkyl mined in the liver, both kidneys, and the brain. Whole-body (CH-&J. Anal. Calcd for C21H46N2010: C, 51.83; H, 9.50; N, 5.76. radioactivity measurements were performed in a double-crystal Found: C, 51.65; H, 9.51; N, 5.68. Raney-Ni catalyst was used scintillation counter (Tobor Nuclear, Chicago). The results for economical reasons instead of PtO2 for larger batches of 3 corrected for radioactive decay and geometry of the samples (ca. 1g of wet Ni catalysvg of glucose), and hydrogenation was were expressed as percentage of the lo9Cd dose and are carried out at 700 psi H2 pressure and at 40 "C for ca. 60-70 h. presented as the arithmetic mean and standard error of the The bis-secondary amine 3 (10.00 g, 20.55 mmol) was dismean. The rats were placed in special containers to restrain solved in 10% aqueous MeOH (150 mL) by warming u p slightly their movements during counting. At the end of the experiment, followed by dropwise addition of an aqueous solution (5 mL) of the in vivo radioactivity measurements were compared with the NaOH (1.64 g, 41.00 mmol) under N2 over 15 min while stirring radioactivity of the same rats after sacrifice, and no differences a t 22.5 "C to give a clear colorless mixture. After 2 h of were observed. The statistical significance of differences beadditional stirring, a solution of CS2 (10 mL) in methanol (25 tween groups was determined by Duncan's multiple-range test mL) was added slowly (20 min) at 10 "C. After CS2 addition was complete, the transparent yellowish reaction mixture was using the SAS/STAT program (SAS Institute, Cary, NC).

QQ

Singh et al.

616 Chem. Res. Toxicol., Vol. 7, No. 5, 1994 2 HOCH2(HOCH)&HO

+ HzN(CH2)9NHz 2

1

1

i. HzO, 60 'C, N2 ii. H2Pt02,60 psi, 72 h. CH,OH

HOCH2(CHOH)4CH2-NH-(CH2)9-NH-CH2(CHOH)4CH2OH 3 ZNaOH, CS2, CH30H

t

Table 1. ComparativeWhole-Body Cadmium Retention after Treatments with a Mono- and a Biscarbodithioate (WJ3 % "Cd ip Dose)a

treatment day

controls

treatment

C9G2DTC treatment

1 2 3 4 5

72.4 f 2.5 71.1 f 2.2 75.4 f 2.6 68.8 f 2.4 67.0 f 1.5

55.5 f 1.3' 40.5 f 1.9' 33.1 f 1.2' 29.5 f 1.1' 25.1 f 1.0'

34.5 f 1.7',' 28.0 f 1.4'8' 25.8 f 1.2bc 24.8 f 1.0'3' 22.6 k 1.0'

BGDTC

a The results are given as the arithmetic mean f SEM. There were 6 rats in each group. BGDTC or C9G2DTC given 7 days after lo9Cdadministration 5 times (daily ip dose 1.0 mmovkg). WB measurements performed 24 h after each treatment and 48 h after the last treatment on the 13th day after lo9Cd application. The statistical differences were calculated using Duncan's multiplerange test. Significantly different from control group; p < 0.05. Significantly different from BGDTC-treated group on the same day;p < 0.05.

'

Table 2. Comparative Organ Cadmium Retention after Treatments with a Mono- and a Biscarbodithioate (% "Cd ip Dose)"

s;c;s

organ

controls

treatment

C9G2DTC treatment

s,/?,s

liver kidney

59.6 f 2.0 4.3 & 0.2 0.10 f 0.01

9.0 f 0.6' 1.7 f 0.1' 0.10 k 0.01

3.4 f O.lb8' 2.4 f 0.2',c 0.09 f 0.01

BGDTC

I

"i"'

c

brain

Results and Discussion

The results are given as the arithmetic mean & SEM. There were 6 rats in each group. BGDTC or C9G2DTC was given 7 days after lo9Cd administration 5 times (daily ip dose 1.0 mmolflcg). Organ measurements were performed 48 h after the last treatment, i.e., the 13th day after lo9Cd application. The statistical differences were calculated using Duncan's multiple-range test. Significantly different from control group;p < 0.05. Significantly different from BGDTC-treated group on the same day; p < 0.05.

Synthesis and Properties of Biscarbodithioate CSGBDTC. The synthesis of CSGBDTC (4) was achieved by use of the reaction sequence illustrated in Figure 2. The imine formation was effected in the presence of a small amount of water because of the limited solubility of glucose (1) in 1,9-diaminononane (2) as with earlier substituted glucose imines, but surprisingly the yield of secondary amine 3 after hydrogenation was relatively lower (37%) even though the duration of reduction was prolonged from 24 to 72 h. Our attempts to improve the yield of 3 did not succeed using R O Zcatalyst. Conversion of amine 3 to the corresponding biscarbodithioate was quite straightforward and afforded 4 in 87% yield. The design of the biscarbodithioate was proposed with a view to complex one cadmium, with both carbodithioate groups of one molecule having two carbodithioate functions tethered together with a long and flexible C-9 alkyl chain. Earlier studies have demonstrated that cadmium forms Cd(DTC)Z type complexes with monocarbodithioates (15). In the solid state, four of the coordination positions of cadmium are occupied by the sulfur atoms of the two dithiocarbamate groups and a fifth coordination position is occupied by a S atom of a dithiocarbamate which is bonded to another Cd2+ion (15). The cadmium complex 5 was prepared in high yield (76.6%)as a white insoluble solid with excellent elemental analysis (Figure 2). Modeling and Design of Effective and Less Toxic Chelating Agents. The structure of CSGBDTC, folded to allow both carbodithioate groups to coordinate to a single cadmium ion, is shown in Figure 1B. While the current study does not prove that the cadmium i n vivo is coordinated to both of the carbodithioate groups, it does prove that such chelating agents can be prepared and

are effective in the mobilization of cadmium from the liver. The more rapid rate of this process, in comparision with BGDTC, is noteworthy. This is consistent with our earlier reports (9, 10, 19) showing an increase in liver cadmium mobilization with increasing molecular weight. The molecular weight of the CSGBDTC anion is 636.86, well above the 300 level at which biliary excretion becomes important in the rat (20). Cadmium stored in the liver as CdMT slowly enters the blood circulation and accumulates in the kidney, causing nephrotoxicity (2123). A compound which is very effective in rapidly reducing hepatic cadmium levels is thus of potential interest as a n agent capable of preventing the continued damage to renal function which is a characteristic feature of high-level chronic cadmium intoxication. The fact that there is essentially no net reduction in the whole-body levels of cadmium after the second day suggests that the remaining cadmium is present in sites to which C9D2DTC cannot gain access. The notable difference in the rates a t which cadmium is removed by BGDTC and CSG2DTC combined with the similar long-term results is consistent with the hypothesis that these two compounds remove cadmium from essentially the same in vivo sites but that the CSGBDTC does this more rapidly because it has two chelating groups rather than the one of BGDTC. More detailed rate information on these processes could be obtained from studies using isolated perfused livers where shorter time intervals could be examined. The molecular modeling of biscarbodithioates for use as cadmium-mobilizing agents must be subject to restrictions in addition to the coordination geometry of the Cd2+ ion. The final molecule must also incorporate appropriate polar groups to reduce the toxicity of the compound

b

N\CH2(CHOH)4CH20H

Cd-CPGZDTC complex (5)

Figure 2. Scheme showing the preparation of C9G2DTC (4) and its 1:l Cd complex (5).

Chem. Res. Toxicol., Vol. 7, No. 5, 1994 617

Molecular-Modeling Design of Cd Mobilizing Agents

G

G

+ s\ 'S

/cCW2DTC G t CH2(CHOH)4CH20H

CdMT

G

G

I Fw%AO-pL?sS'

N

CdMT-C%2DTC Complex G

I

G-N

Cd-C9G2DTC Complex

Figure 3. Proposed mechanism for the removal of Cd from CdMT by CSGBDTC (4). The 4-Cd cluster of CdMT is omitted in the last two steps a t the bottom of the diagram in the interest of clarity.

and to prevent the carbodithioate from transferring cadmium into the brain. The design of a chelating agent to remove cadmium from an intracellular site is thus subject to a considerable number of restraints. Such a design must incorporate structural features which ensure the following: (1)The chelating agent can remove cadmium from the complexes in which it is found inside the cell, including CdMT. (2) The chelating agent can gain access to these intracellular sites by passing through the cellular membranes of those cells in which cadmium is concentrated as CdMT. (3) The cadmium complex which is formed with the chelator is not capable of passing through the blood-brain barrier. (4) The cadmium-chelator complex which is formed does not cause any toxic reaction as it is excreted through the liver or the kidneys. (5) The cadmium complex is rapidly

excreted and does not undergo processes such as enterohepatic circulation or reabsorption in the proximal tubules. (6)The chelating agent is a compound of modest toxicity. The compound should be capable of effecting a significant reduction in cadmium levels when administered a t a dosage which is appreciably less than its LD50 value. The structure of CSG2DTC incorporates features to allow it to meet these desiderata as follows: (1)The carbodithioate function has been demonstrated to remove cadmium from CdMT in vitro (24) and from the in vivo deposits in which cadmium was present as CdMT. (2) High molecular weight (683.83)combined with the amphipathic structure of CSG2DTC favors uptake by hepatocytes and excretion into the bile. Since the great

Singh et al.

618 Chem. Res. Toxicol., Vol. 7, No. 5, 1994 Table 3. LDSO Values of Some Cadmium-MobilizingAgents

LD50 (mmolkg)

chelating agent MGDTC BGDTC MeOBGDTC HBGDTC CBGDTC BLDTC MeBLDTC CSGBDTC

'26.6 11.2 10.0 26.5 >30.0 >6.0 '6.0 >4.0

dl-DMPA BAL Mi-ADMS a

0.82 1.48 >2.0

structure

Dithiocarbamates: R(R')NCSaa R = CH3; R = G" R = C6HsCHz; R' = G" R = p-CH30Cs&CHz; R' = G" R = p-HOCHzCaH4CHz; R' = G" R = p-NaOOCCsHaCHz;R = G" R = CsH&Hz; R = Lb R =p-CH3CsH4CHz; R = Lb (Figure 1) Vicinial Dithiols: RCH(SH)CH(SH)R' R =H R = CH&HCOC~H~COOH-O; R = CHzOH; R = H R = COOCHzCHzCH(CH3)z;R' = COOH

ref 31 32 33 34 34 10 19 this study 35 35 36

Denotes residue derived from a-D-glucose, Le., G = CH~(CHOH)~CHZOH. Denotes residue derived from a-lactose, Le., L =

CH2(CHOH)2CH(CHOH)CH20H I

O

1

O

1

CH(CHOH)3CHCH20H

-

majority of the cadmium in these animals is present in the liver, these features help the chelating agent in gaining access to the cadmium sequestered within hepatocytes. (3) The two highly hydrophilic residues, derived from glucose attached one on each nitrogen, present in CSGBDTC, provide a sufficient degree of polarity to the cadmium-chelator complex to ensure its inability to pass through the blood-brain barrier. This feature can also be seen in a large number of carbodithioates previously reported which contain moieties derived from sugars. ( 4 ) The residues derived from glucose, in a manner analogous to what is found with products of glucuronidation, ensure that cadmium complexes which form are less toxic than the cadmium complexes present within the cells (25). ( 5 ) The combination of the amphipathic structure and a high molecular weight in the chelating agent ensures that the cadmium complex formed will be rapidly cleared in the bile. (6)The toxicity of the present compound is reduced by the incorporation of the two residues derived from glucose. The advantage of the design factors described above are clearly demonstrated in a comparison of CSGBDTC with the biscarbodithioates that have been examined previously [i.e., R(CSZNa)NCHzCHZN(CSzNa)R,where R = CH3, C&5, HOCHzCHz, and CHzC&I (26)a~ ad"mobilizing agents. The lack of any significant degree of activity among these previously studied bis-DTCs was attributed to two major facts: first, they did not possess a tether long enough to allow the two carbodithioate functions to come to closure on a single Cd2+ ion, and second, the absence of sufficient nonionic hydrophilicity rendered them more toxic than desirable. The most important point to note is that CSGPDTC is a more effective cadmium-mobilizing agent than any of these for several reasons. Its significantly greater molecular weight makes it,much more effective in the removal of cadmium, especially from the liver. The presence of nonpolar and polar groups to give a n amphipathic structure overall prevents any transport of cadmium to the brain and modulates its toxicity. Finally, the presence of two carbodithioate groups which can both, at least in principle, coordinate to the same cadmium ion provides CSGPDTC with a n additional advantage over those biscarbodithioates whose carbodithioate groups can nei-

ther coordinate to a single cadmium ion nor function in a n independent fashion. Cadmium-Mobilizing Efficacy of CSGBDTC. The in vivo results presented in Table 1 show that both the mono- and the biscarbodithioate caused a significant reduction in whole-body losCd retention as compared to controls. This reduction was more rapid in animals treated with CSGBDTC than with BGDTC, especially a t the begining of the treatment. In decreasing organ cadmium retention (Table 21, CSGBDTC was found to be much more effective in removing cadmium from the liver but slightly less effective in reducing the renal cadmium levels than the BGDTC. Neither of the chelators had any effect on brain cadmium retentions. After 2 days of treatment, CSGBDTC removes 60% of the whole-body cadmium burden as compared with only 43% by BGDTC. After 5 days of treatment the bis-DTC removes 95%of the hepatic cadmium. I t is of interest to note that treatment with the biscarbodithioate for 2 days produces whole-body cadmium levels comparable to those resulting from 4 days of treatment with the monocarbodithiate BGDTC (Table 1). This is consistent with the clearly demonstrated superiority of the biscarbodithioate in reducing hepatic cadmium levels. In this animal model the great majority of the administered cadmium is present in the liver (see data in Table 21, a site from which cadmium is mobilized preferentially by compounds of higher molecular weight (19). A significant advantage of removing cadmium from the liver into the bile is the potential benefit of such a process to the kidney. The transfer of CdMT from the liver to the kidney is a probable cause of the nephrotoxicity which is found in chronic cadmium intoxication (21). The data given in Table 2 also prove that the biscarbodithioate does not cause any detectable adverse effect on the brain cadmium levels. This inability to carry cadmium across the bloodbrain barrier is a common feature of carbodithioates which contain residues derived from glucose or other sugars. The higher efficiency of CSGPDTC a t the begining of the treatment than a t later intervals remains to be explained but indicates that better results might be obtained by a different treatment schedule. There was undoubtedly some loss of zinc, as we have previously shown that dithiocarbamates enhance the excretion of zinc (27,28). This loss of zinc which is expected to occur concurrently with the mobilization of cadmium may be

Molecular-Modeling Design of Cd Mobilizing Agents

estimated from previous studies on related dithiocarbamates. Mechanism of Removal of Cd from CdMT by COGPDTC. The ability of CSGPDTC to rapidly remove Cd from intracellular sites in the liver strongly suggests that this compound can remove Cd from CdMT complex. A partial mechanism for this process is illustrated in Figure 3, where one of the two carbodithioate groups of CSGBDTC attacks one of the Cd2+ions bound in MT and begins the process of removing it from its bonds to the SH groups of MT. The structure shown for CdMT is patterned after that of ZnMT given by Vallee and Auld (29). The initial attack involves the formation of 5-coordinated cadmium, subsequently followed by the stepwise replacement of S-bonds from thionein to Cd. After one end of the CSG2DTC is bonded to Cd, the proximity of the free carbodithioate group at the other end would greatly facilitate the reaction of this second carbodithioate group and the removal of the Cd from its bonds to MT. The cooperativity of binding of at least part of the cadmium in CdMT suggests that the initial disruption of a cadmium thiolate cluster may facilitate the removal of other cadmium ions from this cluster (30). The mechanism presented in Figure 3 is for CdMT which contains only Cd2+. In tissue one generally expects to find both Zn2+ and Cd2+present in metallothionein induced by the administration of cadmium salts, with Cd2+present in the 4-metal ion cluster and Zn2+present in the 3-metal ion cluster (30). In such a case one would expect that the administration of the dithiocarbamate would simultaneously increase the excretion of zinc and cadmium, and this is what is found (27, 28). It is probable that a reverse cooperativity operates to facilitate the removal of cadmium and zinc from CdZnMT once one of the metal ions has been removed. The approximate ip LD50 value of the new compound which is >4.0 mmolkg is compared in Table 3 with those of the most effective sugar-based dithiocarbamates (MGDTC, BGDTC, MeOBGDTC, HMBGDTC, CBGDTC, BLDTC, and MeBLDTC) reported to date which contain only one CS2Na group, and with 3 vicinal dithiols (dlDMPA, BAL, and Mi-ADMS). The cause of lethality among rats given higher dosages (5.0 mmoYkg or more) of CSGBDTC was ascribed to an acute respiratory failure. Two hours aRer the administration of this compound, the animals exhibited severe respiratory distress which resulted in rapid, shallow, and ineffectual breathing, a purple skin, and lethargy, indicating the onset of progressive respiratory failure due to C02 accumulation, causing death after a n additional 30 min. The demonstration that CSG2DTC is an effective cadmium-mobilizing agent suggests that other biscarbodithioates may be even more effective. The exact structural features which are needed to produce a n optimum cadmium-mobilizing agent of this type are presently under investigation.

Acknowledgment. We wish to acknowledge with thanks the support received for this study by Grant ES02638 from the National Institute of Environmental Health Sciences and the Center in Molecular Toxicology via grant ES-00268 from this same institute (M.M.J., P.K.S., and C.X.);and by the Ministry of Science, Republic of Croatia, and the International Atomic Energy Agency, Vienna, Austria (K.K. and M.B.).

Chem. Res. Toxicol., Vol. 7, No. 5, 1994 619

References Waalkes, M. P., Watkins, J. B., and Klaassen, C. D. (1983) Minimal role of metallothionein in decreased chelator efficacy for cadmium. Toxicol. Appl. Pharmacol. 88,392-398. Cantilena, L. R.,and Klaassen, C. D. (1982)Decreased effectiveness of chelation therapy with time after acute cadmium poisoning. Toxicol.Appl. Pharmacol. 83, 173-180. Eybl, V.,Sykora, J., and Mertl, F. (1966)Effect of CaEDTA and CaDTPA on cadmium poisoning. Acta. Biol. Med. German. 17, 178-185. Jones, M. M., and Cherian, M. G. (1990)The search for chelate antagonists for chronic cadmium intoxication. Toxicology 82, 1-25. Cherian, M. G. (1980)Chelation of cadmium with BAL and DTPA in rats. Nature (London) 281,871-872. Jones, M. M., Singh, P. K., Gale, G. R., Atkins, L. M., and Smith, A. B. (1988)Esters of meso-2,3-dimercaptosuccinicacid as cadmium mobilizing agents. Toxicol.Appl. Pharmacol. 96,507-514. Jones, M. M., Singh, P. K., Gale, G. R., Smith, A. B., and Atkins, L. M. (1992)Cadmium mobilization in vivo by intraperitoneal or oral administration of monoalkvl esters of meso-2,3-dimercautosuccinc acid in the mouse. Phaimacol. Toxicol. 70,336-343 (8)Shaikh, Z.A., and Lucis, 0. J. (1972)Biological differences in cadmium and zinc turnover. Arch. Environ. Health 24,410-418. (9)Jones, M. M., Singh, P. K., Jones, S. G., Gale, G. R., Atkins, L. M., and Smith, A. B. (1991)Structure-activity relationships for the in vivo cadmium mobilization by dithiocarbamates derived from lactose and maltotriose. Chem. Res. Toxicol. 4,692-698. (10)Jones, M. M., Singh, P. K., and Jones, S. G. (1990)Optimization of chelating agent structure for the mobilization of aged renal and hepatic deposits: Sodium N-benzyl-4-O-@-~-galactopyranosy1)-D-glucamine-N-carbodithioate. Chem. Res. Toxicol. 3, 248253. (11)Jones, S. G., Singh, P. K., and Jones, M. M. (1988)Use of the Topliss scheme for the design of more effective chelating agents for cadmium decorporation. Chem. Res. Toxicol. 3, 248-253. (12)Ma, R.,and Martell, A. E. (1993)Protonation constanta and metal ion binding constants of N,N'-bis(2-hydroxyphenyl)-N,N'-ethylenediaminediacetic acid. Inorg. Chim. Acta 209,71-78. (13)Hancock, R. D., Bhavan, R., Wade, P. W., Boeyens, J. C. A., and Dobson, S. M. (1989)Ligand design for complexation in aqueous solution. 1. Neutral oxygen donor bearing groups as a means of controlling size-based selectivity for metal ions. Inorg. Chem. 28, 187-194. (14)White, D. L., Durbin, P. W., Jeung, N., and Raymond, K. N. (1988) Specific sequestering agents for the actinides. 16.Synthesis and initial biological testing of polydentate oxohydroxypyridinecarboxylate ligands. J. Med. Chem. 31,11-18. (15)Casas, J. S.,Sbnchez, A., Bravo, J., Garcia-Fontan, S., Castellano, E. E., and Jones, M. M. (1989)Cadmium coordination chemistry related to chelate therapy. Znorg. Chim. Acta 158,119-126. (16)Kojima, S.,Kaminaka, K., Kiyozumi, M., and Honda, T. (1986) Comparative effects of three chelating agents on distribution and excretion of cadmium in rats. Toxicol.Appl. Pharmacol. 83,516524. (17)Jones, M. M., Singh, P. K., Mukundan, C. R., Banton, J. A., Gale, G. R., Atkins, L. M., and Smith, A. B. (1991)Chirality and chain branching effects on the dithiocarbamate-induced mobilization of cadmium from intracellular deposits in mice. Chem. Res. Toxicol. 4,27-34. (18)Jones, M. M., Cherian, M. G., Singh, P. K., Basinger, M. A,, and Jones, S. G. (1991)Comparative study of the influence of vicinal dithiols and a dithiocarbamate on the biliary excretion of cadmium in rat. Toxicol. Appl. Pharmacol. 110,241-250. (19)Jones, M. M., Singh, P. IC,Jones, S. G., and Holscher, M. A. (1991) Dithiocarbamates of improved efficacy for the mobilization of retained cadmium from renal and hepatic deposits. Pharmacol. Toxicol. 88, 115-120. (20)Klaassen, C. D., and Watkins, J. B., I11 (1984)Mechanism of bile formation, hepatic uptake, and biliary excretion. Pharmacol. Rev. 36,1-66. (21)Wang, X.P., Chan, H. M., Goeyer,R. A., and Cherian, M. G. (1993) Nephrotoxicity of repeated injection of cadmium-metallothionein in rats. Toxicol. Appl. Toxicol. 119,11-16. (22)Kido, T., Shaikh, Z. A., Kito, H., Honda, R., and Nogawa, K. (1991) Dose-response relationship between dietary cadmium intake and metallothioneinuria in a popuation from cadmium-polluted area of Japan. Toxicology 68, 271-278. (23)Nogawa, K., Kido, T., and Shaikh, Z. A. (1993)Dose-response for renal dysfunction in a population enviromentally exposed to cadmium. ZARC Sci. Publ. 118,311-318. (24)Planas-Bohne, F.,Jones, M. M., and Singh, P. K. (1991)Effect of chelating agents on the cadmium burden of cells in culture. Cell Biol. Toxicol. 7,167-181.

620 Chem. Res. Toxicol., Vol. 7, No. 5, 1994 (25) Jones, S. G., Holscher, M. A., Basinger, M. A,, and Jones, M. M. (1988) Dependence on chelating agent properties of nephrotoxicity and testicular damage in male mice during cadmium decorporation. Toxicology 63,135-146. (26) Singh, P. K., Jones, M. M., Jones, M. M, Gale, G. R., Atkina, L. M., Smith, A. B., Bulman, R. A. (1989) Effect of chelating agent structure on the mobilization of cadmium from intracellular deposits. J. Toxicol. Environ. Health 28, 501-518. (27) Kargacin, B., Blanusa, M., Arezina, R., Kostial, K., Singh, P. K., and Jones, M. M. (1991) Effect of cadmium mobilizing dithiocarbamates on esential trace metal metabolism in rats. Bull. Environ. Contam. Toxicol. 47, 160-164. (28) Jones, M. M., Basinger, M. A., Holscher, M. A., and Singh, P. K. (1992) Effect of dithiocarbamates and dithiocarbamate-induced cadmium mobilization on essential, trace metal metabolism in female rat. Fundam. Appl. Toxicol. 19, 432-437. (29) Vallee, B. L., and Auld, D. 5. (1990) Zinc coordination, function, and structure of zinc enzvmes and other Droteins. Biochemistrv 29,5647-5659. (30) Waalkes. M. P.. and Goerine. P. L. (1990) Metallothionein and other ca&nium:binding proteins: Recent developments. Chem Res. Toxicol. 3, 281-288.

Singh et al. (31) Shinobu, L. A., Jones, S. G., and Jones, M. M. (1984) Sodium N-methyl-D-glucamine dithiocarbamate and cadmium intoxication. Acta Pharmacol. Toxicol. 54, 189-194. (32) Kojima, S., Kiyozumi, M., Honda, T., Kaminaka, K., Oda, Y., and Senba, Y. (1987) Effect of N-benzyl-D-glucamine dithiocarbamte on distribution and excretion of cadmium in rats. Toxicology 45, 93-102. (33) Gale, G. R., Atkins, L. M., Smith, A. B., Singh, P. K., and Jones, M. M. (1989) N,N-Disubstituted dithiocarbamates as cadmium antagonists: N-(4-Methoxybenzyl)-N-dithiocaboxy-~-glucamine, Toxicol. Lett. 48, 105-155. (34) Shimada, H., Kamenosono, T., Kawagoe, M., Funakoshi, T., Kiyozumi, M., Takadate, A., and Kojima, S. (1991) Mobilization of renal and hepatic cadmium by dithiocarbamates in rats J. Pharmacobio-Dyn. 14,555-560. (35) Aposhian, H. V., Carter, D. E., Hoover, T. D., Hsu,C.-A., Maiorino, R. M., and Stine, E. (1984) DMSA, DMPS, and DMPA as arsenic antidotes. Fundam. Appl. Toxicol. 4, 558-570. (36) Jones, M. M., Singh, P. K., Glen, G. R., Smith, A. B., and Atkins, L. M. (1992) Cadmium mobilization in vivo by intraperitoneal or oral administration of monoalkyl esters meso-2,3-dimercaptosuccinic acid in mouse. Pharmacol. Toxicol. 70, 336-343.