permeability, cytotoxicity, and genotoxicity of cr(iii) complexes and some cr(v) analogues in v79...

7
Permeability, Cytotoxicity, and Genotoxicity of Cr(III) Complexes and Some Cr(V) Analogues in V79 Chinese Hamster Lung Cells Carolyn T. Dillon, ² Peter A. Lay,* Antonio M. Bonin, Marian Cholewa, §,| and George J. F. Legge § Centre for Heavy Metals Research, School of Chemistry, The University of Sydney, NSW 2006, Australia, National Occupational Health and Safety Commission (NOHSC), P.O. Box 58, Sydney, NSW 2001, Australia, Micro Analytical Research Centre, School of Physics, The University of Melbourne, Parkville 3052, Australia, and Institute of Nuclear Physics, Cracow, Poland Received January 27, 2000 The permeabilities and genotoxicities of the Cr(III) complexes [Cr(en) 3 ] 3+ , mer-[Cr(glygly) 2 ] - , cis-[Cr(phen) 2 (OH 2 ) 2 ] 3+ , and trans-[Cr(salen)(OH 2 ) 2 ] + and the Cr(V) analogues of cis-[Cr(phen) 2 - (OH 2 ) 2 ] 3+ and trans-[Cr(salen)(OH 2 ) 2 ] + [en being 1,2-ethanediamine, glygly being glycylglycine, phen being 1,10-phenanthroline, and salen being N,N-ethylenebis(salicylideneiminato)] have been studied in V79 Chinese hamster lung cells. Following exposure of 10 6 cells to 0.4 mM Cr(III) for 4 h, the Cr uptake by single cells was less than 10 -14 g/cell (as determined by GFAAS analysis and as confirmed by PIXE analysis where the Cr concentration was below the limit of detection). Importantly, the Cr(V) analogue of cis-[Cr(phen) 2 (OH 2 ) 2 ] was significantly more permeable than the Cr(III) complex. The cytotoxicity of the Cr(III) complexes increased in the following order: mer-[Cr(glygly) 2 ] - < [Cr(en) 3 ] 3+ cis-[Cr(phen) 2 (OH 2 ) 2 ] 3+ < trans-[Cr(salen)- (OH 2 ) 2 ] + . No genotoxic effects were observed following exposure to mer-[Cr(glygly) 2 ] - or [Cr- (en) 3 ] 3+ at concentrations up to 6 mM. The Cr(III) imine complexes trans-[Cr(salen)(OH 2 ) 2 ] + and cis-[Cr(phen) 2 (OH 2 ) 2 ] 3+ , which could be oxidized to Cr(V) complexes, induced MN in vitro at rates of 13.6 and 3.3 MN/1000 BN cells/μmol of Cr, respectively. The comparative permeabilities and genotoxicities of trans-[Cr(salen)(OH 2 ) 2 ] + and [CrO(salen)] + were similar due to the instability of the Cr(V) complex at physiological pH values (7.4). There was a substantial increase in the permeability of [Cr(O) 2 (phen) 2 ] + , compared to that of the Cr(III) analogue, which was accompanied by a highly genotoxic response. Consequently, any Cr(III) complex that is absorbed by cells and can be oxidized to Cr(V) must be considered as a potential carcinogen. This has potential implications for the increased use of Cr(III) complexes as dietary supplements and highlights the need to consider the genotoxicities of a variety of Cr(III) complexes when determining the carcinogenic potential of Cr(III) particularly when “high” deliberately administered doses are concerned. Introduction The recommended dietary intake of Cr ranges from 50 to 200 μg of Cr/day (1). The endorsement of Cr stems from findings that it plays a beneficial role in glucose tolerance and diabetes (2, 3). Recently, a naturally occurring oligopeptide, low-molecular weight chromium-binding substance (LMWCr), which binds four Cr(III) ions has been identified. The protein is capable of activating insulin receptor kinase by up to 7-fold (3). Cr is also involved in lipid and carbohydrate metabolism in mam- mals with a Cr deficiency (1, 4, 5). Consequently, Cr, in the form of tris(picolinato)chromium(III) [Cr(pic) 3 ](6), is promoted as a muscle-building agent, although there is some controversy surrounding its use (7-12). While Cr(VI) has been established as a carcinogen on the basis of epidemiological evidence and animal experiments, no such evidence has been accumulated for Cr(III), which has been important in its acceptance as a dietary supple- ment (13, 14). The main reason for this is that there is little evidence to show that Cr(III) compounds induce tumors in mice or rats (13). Furthermore, while there is overwhelming evidence to show that Cr(VI) complexes are mutagenic in bacterial and mammalian cells, most Cr(III) complexes are not mutagenic (13, 15). However, Warren et al. (15) found that Cr(III) complexes containing aromatic imine ligands, cis-[Cr- (phen) 2 Cl 2 ] + , cis-[Cr(bipy) 2 Cl 2 ] + , and [Cr(bipy) 2 (ox)]I4H 2 O, are mutagenic in Salmonella typhimurium (TA92, TA98, and TA100). cis-[Cr(phen) 2 Cl 2 ] + also induces mu- tations at the HGPRT locus in V79 cells (16). The bacterial mutagenicity of the phen-containing complexes cannot be attributed directly to the ligand [i.e., phen is not mutagenic in S. typhimurium (17, 18)], although it is believed that the lipophilic nature of the phen and bipy ligands increases the permeabilities of the complexes, leading to the observed genotoxicities (16). It has been shown recently that trans-[Cr(salen)(OH 2 ) 2 ] + is also mu- tagenic in S. typhimurium TA97a, TA98, TA100, and TA102, while the salen ligand is not mutagenic (19). In * To whom correspondence should be addressed. ² The University of Sydney. National Occupational Health and Safety Commission. § The University of Melbourne. | Institute of Nuclear Physics. 742 Chem. Res. Toxicol. 2000, 13, 742-748 10.1021/tx0000116 CCC: $19.00 © 2000 American Chemical Society Published on Web 07/26/2000

Upload: george-j-f

Post on 21-Feb-2017

212 views

Category:

Documents


0 download

TRANSCRIPT

Permeability, Cytotoxicity, and Genotoxicity of Cr(III)Complexes and Some Cr(V) Analogues in V79 Chinese

Hamster Lung Cells

Carolyn T. Dillon,† Peter A. Lay,*,† Antonio M. Bonin,‡ Marian Cholewa,§,| andGeorge J. F. Legge§

Centre for Heavy Metals Research, School of Chemistry, The University of Sydney, NSW 2006,Australia, National Occupational Health and Safety Commission (NOHSC), P.O. Box 58, Sydney,

NSW 2001, Australia, Micro Analytical Research Centre, School of Physics, The University ofMelbourne, Parkville 3052, Australia, and Institute of Nuclear Physics, Cracow, Poland

Received January 27, 2000

The permeabilities and genotoxicities of the Cr(III) complexes [Cr(en)3]3+, mer-[Cr(glygly)2]-,cis-[Cr(phen)2(OH2)2]3+, and trans-[Cr(salen)(OH2)2]+ and the Cr(V) analogues of cis-[Cr(phen)2-(OH2)2]3+ and trans-[Cr(salen)(OH2)2]+ [en being 1,2-ethanediamine, glygly being glycylglycine,phen being 1,10-phenanthroline, and salen being N,N′-ethylenebis(salicylideneiminato)] havebeen studied in V79 Chinese hamster lung cells. Following exposure of ∼106 cells to 0.4 mMCr(III) for 4 h, the Cr uptake by single cells was less than 10-14 g/cell (as determined by GFAASanalysis and as confirmed by PIXE analysis where the Cr concentration was below the limitof detection). Importantly, the Cr(V) analogue of cis-[Cr(phen)2(OH2)2] was significantly morepermeable than the Cr(III) complex. The cytotoxicity of the Cr(III) complexes increased in thefollowing order: mer-[Cr(glygly)2]- < [Cr(en)3]3+ ∼ cis-[Cr(phen)2(OH2)2]3+ < trans-[Cr(salen)-(OH2)2]+. No genotoxic effects were observed following exposure to mer-[Cr(glygly)2]- or [Cr-(en)3]3+ at concentrations up to 6 mM. The Cr(III) imine complexes trans-[Cr(salen)(OH2)2]+

and cis-[Cr(phen)2(OH2)2]3+, which could be oxidized to Cr(V) complexes, induced MN in vitroat rates of 13.6 and 3.3 MN/1000 BN cells/µmol of Cr, respectively. The comparativepermeabilities and genotoxicities of trans-[Cr(salen)(OH2)2]+ and [CrO(salen)]+ were similardue to the instability of the Cr(V) complex at physiological pH values (7.4). There was asubstantial increase in the permeability of [Cr(O)2(phen)2]+, compared to that of the Cr(III)analogue, which was accompanied by a highly genotoxic response. Consequently, any Cr(III)complex that is absorbed by cells and can be oxidized to Cr(V) must be considered as a potentialcarcinogen. This has potential implications for the increased use of Cr(III) complexes as dietarysupplements and highlights the need to consider the genotoxicities of a variety of Cr(III)complexes when determining the carcinogenic potential of Cr(III) particularly when “high”deliberately administered doses are concerned.

Introduction

The recommended dietary intake of Cr ranges from 50to 200 µg of Cr/day (1). The endorsement of Cr stems fromfindings that it plays a beneficial role in glucose toleranceand diabetes (2, 3). Recently, a naturally occurringoligopeptide, low-molecular weight chromium-bindingsubstance (LMWCr), which binds four Cr(III) ions hasbeen identified. The protein is capable of activatinginsulin receptor kinase by up to 7-fold (3). Cr is alsoinvolved in lipid and carbohydrate metabolism in mam-mals with a Cr deficiency (1, 4, 5). Consequently, Cr, inthe form of tris(picolinato)chromium(III) [Cr(pic)3] (6), ispromoted as a muscle-building agent, although there issome controversy surrounding its use (7-12). WhileCr(VI) has been established as a carcinogen on the basisof epidemiological evidence and animal experiments, nosuch evidence has been accumulated for Cr(III), which

has been important in its acceptance as a dietary supple-ment (13, 14). The main reason for this is that there islittle evidence to show that Cr(III) compounds inducetumors in mice or rats (13). Furthermore, while there isoverwhelming evidence to show that Cr(VI) complexesare mutagenic in bacterial and mammalian cells, mostCr(III) complexes are not mutagenic (13, 15).

However, Warren et al. (15) found that Cr(III)complexes containing aromatic imine ligands, cis-[Cr-(phen)2Cl2]+, cis-[Cr(bipy)2Cl2]+, and [Cr(bipy)2(ox)]I‚4H2O, are mutagenic in Salmonella typhimurium (TA92,TA98, and TA100). cis-[Cr(phen)2Cl2]+ also induces mu-tations at the HGPRT locus in V79 cells (16). Thebacterial mutagenicity of the phen-containing complexescannot be attributed directly to the ligand [i.e., phen isnot mutagenic in S. typhimurium (17, 18)], although itis believed that the lipophilic nature of the phen and bipyligands increases the permeabilities of the complexes,leading to the observed genotoxicities (16). It has beenshown recently that trans-[Cr(salen)(OH2)2]+ is also mu-tagenic in S. typhimurium TA97a, TA98, TA100, andTA102, while the salen ligand is not mutagenic (19). In

* To whom correspondence should be addressed.† The University of Sydney.‡ National Occupational Health and Safety Commission.§ The University of Melbourne.| Institute of Nuclear Physics.

742 Chem. Res. Toxicol. 2000, 13, 742-748

10.1021/tx0000116 CCC: $19.00 © 2000 American Chemical SocietyPublished on Web 07/26/2000

vitro DNA studies of these complexes and their Cr(V)analogues revealed that the Cr(V) complexes interactedstrongly with DNA at 37 °C and pH values of 3.3 and7.4. Furthermore, the controversial dietary supplement(20, 21) [Cr(pic)3] is clastogenic (11), such that Crconcentrations of 0.05-1 mM induce chromosomal dam-age 3-18-fold above the control levels in Chinese hamsterovary cells (12). The level of DNA fragmentation is alsoincreased by 1.2-1.6-fold when cells are incubated with[Cr(pic)3] (30-50 µg/mL) (10).

Previously (22), we studied the permeabilities, cyto-toxicities, and genotoxicities of Cr(V) and Cr(VI) com-plexes in V79 cells. PIXE (particle-induced X-ray emis-sion) analysis using a scanning proton microprobe (1 µmbeam diameter) enabled the determination of intracel-lular Cr levels in single cells following exposure of theV79 cells to Cr(V) and Cr(VI) complexes (0.5 µmol of Cr/dish, 4 h). These results allowed the quantification of Crcytotoxicity and genotoxicity at a cellular level, andrevealed that the potency of Cr(V) species was similar to(if not greater than) that of Cr(VI) complexes.

In this paper, the permeability, cytotoxicity, and geno-toxicity of several Cr(III) complexes are reported and,where possible, those of their Cr(V) analogues. Thecomplexes studied included [Cr(en)3]3+, a trivalent com-plex containing nonaromatic ligands; mer-[Cr(glygly)2]-,a biologically relevant anionic peptide-containing com-plex; cis-[Cr(phen)2(OH2)2]3+, a mutagenic complex andits Cr(V) analogue, [Cr(O)2(phen)2]+; and trans-[Cr(salen)-(OH2)2]+, a monovalent complex containing a tetradentateimine ligand and its Cr(V) analogue, [CrO(salen)]+. Thecomplexes were chosen because they either were amongthe few Cr(III) complexes that are mutagenic (15, 16, 18,19) or represent typical classes of Cr(III) complexes thatare not mutagenic. By studying this series of complexes,we anticipated that insights would be gained into thereasons why some Cr(III) complexes are genotoxic whileothers are not. This is important not only for understand-ing potential hazards of Cr(III) dietary supplements andCr(III) complexes encountered in industry but also in thecontext of undergraduate and research laboratories,where such complexes are encountered frequently. Thebiochemistry and chemistry of the dietary supplementsthemselves are somewhat more complex than the systemsreported here and will require further study to rationalizetheir bioactivity. The work is in progress and will bereported at a later date.

Experimental Procedures

Caution: cis-[Cr(phen)2(OH2)2]3+, trans-[Cr(salen)(OH2)2]+,and their Cr(V) analogues are genotoxic and possible carcinogens(15, 16, 18, 19). Due care should be taken to avoid inhalation ofthe compounds and to avoid contact with skin.

Syntheses. [Cr(en)3]Cl3 was prepared according to themethod of Gillard et al. (23). Na{mer-[Cr(glygly)2]} was syn-thesized using the method described by Murdoch (24), and thepurity was determined by electronic absorption spectroscopy inH2O: λmax (εmax) 550 (161 M-1 cm-1), 418 nm (37.4 M-1 cm-1)[lit., 552 (185 M-1 cm-1), 420 nm (40 M-1 cm-1) (24)]. cis-[Cr-(phen)2(OH2)2](NO3)3‚5/2H2O was prepared according to themethod of Inskeep et al. (25, 26). Anal. Calcd for CrC24N7-H25O13.5: C, 42.42; H, 3.71; N, 14.43. Found: C, 42.60; H, 3.68;N, 14.38. trans-[Cr(salen)(OH2)2]Cl was prepared as describedby Yamada and Iwasaki (27, 28). Mass spectroscopy (electronimpact, Kratos MS9 upgraded to MS50 configuration) yieldeda parent peak at m/z 353 and a base peak at m/z 318corresponding to the weights of [Cr(salen)Cl]+ and [Cr(salen)]+,

respectively. Electronic absorption spectroscopy in H2O resultedin a visible peak at 382 nm: εmax ) 4.4 × 103 M-1 cm-1 [lit.,λmax ) 381 nm, εmax ) 5 × 103 M-1 cm-1 (29)].

The Cr(V) complexes were prepared by oxidizing the Cr(III)complexes in acetate buffer (pH 3.3) at 37 °C using a 4-fold molarexcess of PbO2 (Merck). A 60 min oxidation of cis-[Cr(phen)2-(OH2)2]3+ produced [Cr(O)2(phen)2]+, which was identified by anEPR signal at giso ) 1.9386 [cf. giso ) 1.937 (19, 30, 31)]. A 30min oxidation of trans-[Cr(salen)(OH2)2]+ yielded [CrO(salen)]+,which was identified by a five-line EPR signal at giso ) 1.9755,AN ) 2.05 × 10-4 cm-1 [cf. giso ) 1.978, AN ) 2.00 × 10-4 cm-1,in acetonitrile (32)]. The features of the absorption spectrumincluded high absorption below 430 nm, and a peak at 590 nm,with an intensity that indicated >99% conversion to the Cr(V)analogue (32).

Reactivity of Cr Complexes in Tissue Culture Medium.[Cr(en)3]Cl3 (5 µmol, 100 µL) was added to MEM tissue culturemedium (2.9 mL, Eagle’s, 9.78 g L-1) to produce a final Cr(III)concentration of 1.67 mM, and the reactivity was monitored byUV/vis spectroscopy using a Hewlett-Packard 8452A diode arrayspectrophotometer. A Hewlett-Packard 8909A Peltier temper-ature control unit maintained the solution temperature at 37°C, and the spectra were collected at 5 min intervals for 2 h.The phenol red component of the MEM interfered with thevisible spectrum at approximately 550 nm (despite backgroundsubtraction of MEM), causing some difficulties with the analysisof the reactivities of mer-[Cr(glygly)2]-, cis-[Cr(phen)2(OH2)2]3+,and trans-[Cr(salen)(OH2)2]+. Consequently, the reactivities ofthese complexes were monitored in MEM-P, a minimal essentialsalt medium, which did not contain phenol red, (0.97 g/100 mL,Flow Laboratories). Solutions were prepared as 1.67 mM Cr,and the reactivities were monitored at 37 °C using the procedurementioned above.

Tissue Culture Procedures Used for V79 Cells. V79 cellswere employed for the permeability, cytotoxicity, and genotox-icity studies (33). The cells, supplied initially by R. Newbold(Institute of Cancer Research, Sutton, Surrey, U.K.), werecultured from frozen stocks (maintained at -196 °C) and grownto confluency in a 5% CO2/95% air-humidified incubator (model3029, Forma Scientific) at 37 °C. The cells were cultured ingrowth medium (GM) containing MEM (9.76 g L-1, Eagle’s) andsodium hydrogen carbonate (2.2 g L-1, Sigma) supplementedwith 10% fetal calf serum (Multi Ser) and 50 IU mL-1 penicillinand 50 µg mL-1 streptomycin (Multi Ser) (18, 22, 34).

Cytotoxicity and Genotoxicity Assays. The cytotoxicitiesof the Cr(III) complexes (in 5 mL of GM) following a 4 h exposurewere determined using the clonal assay, and the maximumCr(III) concentrations were established by the solubility limitsof the complexes. The assays were performed in a manneridentical to that described previously (18, 22), whereby thesurvival percentage was determined 7 days after seeding 180treated V79 cells.

The genotoxicities of the Cr complexes were determined bythe in vitro micronucleus assay employing the cytokinesis blockmethod for micronucleus detection (22, 35, 36). V79 cells (5 ×105) were seeded in GM (5 mL) contained in 60 mm tissueculture dishes. After 24 h, the cells were washed with PBS, andnew GM (5 mL) was added followed by the freshly prepared Crsolutions (0.1 mL, duplicate dishes). The Cr(V) solutions wereprepared by oxidation of the stock Cr(III) solution in acetatebuffer (pH 3.6) at 37 °C using a 4-fold molar excess of PbO2.The solutions were filtered through a 0.2 µm filter, and thefiltrates were stored on ice until dilution and cell treatment.After the 4 h treatment period, the cells were washed with PBS,and new GM (5 mL) was added. The cells were reincubated for24 h, after which time cytochalasin B (15 µg/dish, Sigma) wasadded to block cytokinesis. After a further 24 h incubation, thecells were harvested and centrifuged onto slides (Shandon,cytospin 2). The cells were stained using Diff-quik stain, andthe slides were scored blind at 600× magnification. Theincidence of micronuclei was determined from a minimum of1000 BN cells per duplicate dish (2000 BN cells per dose) using

Permeability and Genotoxicity of Cr(III) and Cr(V) Chem. Res. Toxicol., Vol. 13, No. 8, 2000 743

the following criteria: the micronucleus must be present in aBN cell with two discrete main nuclei, the micronucleus mustbe one-twentieth to one-fifth the size of the main nuclei, themicronucleus must have staining characteristics identical tothose of the main nuclei, and the micronucleus must not be incontact with the main nuclei (37-39).

Permeability Studies. Cells were seeded at a density of 5×105 cells/dish and incubated at 37 °C and 5% CO2 for 24 h.The cells were washed twice with PBS prior to treatment, andnew GM (5 mL) was added, followed immediately by the freshlyprepared Cr solution (2 µmol/dish, 100 µL). Control cells wereprepared in an identical manner, except that 0.1 mL of sterilewater (filtered through a 0.2 µm filter) was substituted for theCr solution. Triplicate dishes were prepared for each treatment.Following the treatment period (4 h), the GM was removed andthe cells were washed thoroughly using PBS solution (pH 7.4,Oxoid). Trypsin (2.5% in PBS) was added, and the cells wereharvested after 5 min. The dishes were washed thoroughly withPBS to ensure that all of the cells were transferred to thepolyethylene tube. The cells were centrifuged for 7 min at 1100rpm, and the trypsin was removed. Saline solution (2 mL, 0.9%NaCl, 99.999% pure, Aldrich) was added; the suspensions werecentrifuged for 7 min, and the supernatants were discarded.This procedure was repeated, and the final saline solutions weredrained from the pellets. The number of cells per dish wasdetermined by scoring cells that were grown under identicalconditions but were untreated. Cr analyses of the treated cellswere performed following digestion of the cells in HNO3 (Tra-cePur, Merck, 0.9 ppb Cr batch analysis) using presoakedglassware (RBS, 3 days; 1.5 M HNO3, 3 days; distilled water, 3days; rinsed five times in milliQ water×). The digestions wereperformed for approximately 6 h, and the resultant ash wasdissolved in Milli-Q water in presoaked 1 mL volumetric flasks.The solutions were analyzed for Cr using a Varian GTA96Graphite Tube Atomizer SpectrAA.

The permeabilities of the Cr complexes were also investigatedin individual cells by PIXE analyses by employing the scanningproton microprobe at MARC, School of Physics, The Universityof Melbourne, as described in detail elsewhere (22, 34). The cellswere treated as described above, harvested, and freeze-driedfrom 200 mM ammonium acetate solution (22, 34). The PIXEdata were collected as a function of the X-ray energy and theassociated count frequencies at each spatial (X,Y) coordinate ofthe beam. The PIXE spectra of the cells were normalized to thecharge, and the absolute amount of Cr was determined withthe aid of a Mn standard (22).

Statistical Analyses. MN data were analyzed by simplelinear regressions, and a ø2 test was used to compare thefrequencies of MN obtained for various treatments and thecontrols. A ranking of the cytotoxicities of the Cr(III) complexeswas determined by Tukey-Kramer analysis of variance for thesurvival data obtained at 3 µmol of Cr/dish. The results of thepermeability assays were assessed using ANOVA and the two-tailed t test. In all cases, a P value of <0.05 was consideredsignificant (36, 40).

Results

Reactivity of Cr Complexes in the Cell GrowthMedium. Figure 1 shows the electronic absorptionspectra of a MEM solution of [Cr(en)3]Cl3 (1.67 mM)maintained at 37 °C. The absorbance of the peaks at 352and 456 nm decreased by less than 10% over 2 h, and nonoticeable changes in peak positions were apparent.

There was a substantial decrease in the absorbance ofmer-[Cr(glygly)2]- (1.67 mM) at λmax ) 420 and 550 nmfollowing reactions in MEM-P solution (Figure 2). Themagnitude of the peak at 550 nm decreased by ap-proximately 30% within 2 h, indicating a degree ofinstability of the peptide complex in this medium.Similarly, a loss in the absorbance of the peaks was also

observed following the reaction of mer-[Cr(glygly)2]- (1.67mM) in phosphate buffer (pH 7.4) maintained at 37 °Cfor 2 h, although it was less pronounced (10% decreasein the magnitude of the peak at λmax ) 550 nm) (18).

The addition of cis-[Cr(phen)2(OH2)2]3+ (1.67 mM) toMEM-P solution resulted in immediate deprotonation ofthe two aqua ligands, as noted by the shift in the λmax

from 497 nm (0.1 M HNO3) to 519 nm (pH 7.4) (26, 41).No further changes in λmax of εmax were observed duringthe 2 h reaction period. The reaction of trans-[Cr(salen)-(OH2)2]+ in MEM-P for 2 h was negligible, producing onlya small decrease in the absorbance intensity at allwavelengths.

Cytotoxicity of Cr(III) Complexes. The cytotoxici-ties of cells treated for 4 h with Cr(III) complexes weredetermined by clonal assays (Figure 3), and none of thetested complexes induced total cell death at the highestconcentrations that were tested. The most toxic substancewas trans-[Cr(salen)(OH2)2]+, which induced 95% celldeath at 10 µmol of Cr/dish. There were only 66%survivors following treatment with cis-[Cr(phen)2(OH2)2]3+

(10 µmol of Cr/dish, 2 mM Cr), while exposure to 30 µmolof [Cr(en)3]3+ and mer-[Cr(glygly)2]-/dish resulted in 71%and 77% survivors, respectively. The cytotoxicities cor-rected to a Cr(III) concentration of 10 µmol of Cr(III)/dish showed the following rank order: mer-[Cr(glygly)2]-

< [Cr(en)3]3+ ∼ cis-[Cr(phen)2(OH2)2]3+ < trans-[Cr(salen)-(OH2)2]+.

Figure 1. Electronic absorption spectra of [Cr(en)3]3+ (1.67mM) in MEM-P solution obtained at 37 °C over 2 h. The spectrawere collected 2, 60, and 120 min after mixing. The arrowsindicate the decrease in the absorption with time.

Figure 2. Electronic absorption spectra of [Cr(glygly)2]- (1.67mM) in MEM-P solution obtained at 37 °C over 2 h. The spectrawere collected 1, 30, 60, 90, and 120 min after mixing. The arrowindicates the decay of the absorption with time.

744 Chem. Res. Toxicol., Vol. 13, No. 8, 2000 Dillon et al.

Genotoxicity. The genotoxic effects of the Cr com-plexes, as determined by the in vitro micronucleus assay,are shown in Figure 4. Quantitative and statisticalanalysis data are provided in Table 1. The complex mer-[Cr(glygly)2]- was not genotoxic under the reportedconditions (ø2, P > 0.05) as was the case for [Cr(en)3]3+

(P ) 0.13) where the linear response was poor (R2 ) 0.45)and was not dose-dependent. Furthermore, the responsesof [Cr(en)3]3+ and mer-[Cr(glygly)2]- were not significantlydifferent from each other (ANOVA). The Cr(III) iminecomplexes trans-[Cr(salen)(OH2)2]+ and cis-[Cr(phen)2-(OH2)2]3+ were genotoxic and induced 14 MN/1000 BN/µmol of Cr and 3 MN/1000 BN/µmol of Cr, respectively.The Cr(V) analogue of cis-[Cr(phen)2(OH2)2]3+ was themost genotoxic of all the complexes and induced 125 MN/1000 BN/µmol of Cr, which was significantly greater than

the effect observed for cis-[Cr(phen)2(OH2)2]3+. While thegenotoxic response of [CrO(salen)]+ appears to be moresubstantial than that of trans-[Cr(salen)(OH2)2]+, theslopes of the curves were not significantly different.

Permeability. The GFAAS permeability results areshown in Figure 5 for cells that were treated with 2 µmolof Cr(III) or Cr(V) per dish (0.4 mM Cr) for 4 h. Alsoincluded in the graph are the results of the 4 h treatmentof the cells with the positive control, [Cr2O7]2- (0.5 µmolof Cr/dish). The Cr permeability determined by GFAASwas e10-14 g/cell for all Cr(III) complexes that weretested, as was also confirmed by PIXE analyses ofindividual cells. Furthermore, the Cr uptake by the cellswas not significantly greater than the controls for all thetested Cr(III) complexes (P > 0.5, ANOVA, Tukey-Kramer and Bonferroni). The Cr uptake observed follow-ing treatment of the cells with [CrO(salen)]+ was notsignificantly greater than the control, nor was it signifi-cantly different from that of trans-[Cr(salen)(OH2)2]+. Asignificant increase in the cellular Cr level was observedfollowing exposure to [Cr(O)2(phen)2]+ (P < 0.01, Tukey-Kramer). The Cr uptake was also significantly greaterthan that observed for cis-[Cr(phen)2(OH2)2]3+ exposure.As expected, the Cr(VI) permeability following treatmentwith 0.5 µmol/dish was highest and was significantlygreater than those of all of the tested Cr complexes.

PIXE analysis of single cells treated with Cr(III)complexes confirmed that the cellular Cr concentrationswere less than 10-14 g/cell. Due to the sensitivity limita-tions of the instrument, no distinction in the Cr contentcould be made for the various Cr(III) treatments abovethat of the control. Importantly, however, there wasevidence of an increased cellular Cr level when the cellswere treated with [Cr(O)2(phen)2]+ as CrKR peak mag-nitudes up to 8 times greater than the background wereobserved.

Discussion

The Cr(III) complexes [Cr(en)3]3+, mer-[Cr(glygly)2]-,cis-[Cr(phen)2(OH2)2]3+, and trans-[Cr(salen)(OH2)2]+ werechosen to determine whether Cr(III) permeability, cyto-toxicity, and genotoxicity were influenced by ligandspecificity and/or the sign and size of the charge of thecomplex. [Cr(en)3]3+ and cis-[Cr(phen)2(OH2)2]3+ wereadded to the culture medium as triply charged cations.It is known that [Cr(en)3]3+ undergoes ligand-exchangereactions in H2O (followed by subsequent deprotonation

Figure 3. Survival percentages of V79 cells obtained followingtreatment with Cr(III) complexes. Data points are means ( SD(n ) 3).

Figure 4. Incidence of micronucleated cells obtained from 1000binucleated cells following the treatment of V79 cells with thespecified Cr(III) or Cr(V) complexes. Data points are means (SD (n ) 2).

Table 1. Summary of the Micronucleus Assays of CrComplexes in V79 Cells

Cr complex MN/1000 BN cells/µmol of Cra R2, P valueb

[Cr(glygly)2]- 0.01(0.08) 0.004, 0.9[Cr(en)3]3+ 0.2(0.1) 0.47, 0.13[Cr(salen)(OH2)2]+ 13.6(1.7) 0.94, 0.001[CrO(salen)]+ 18.2(3.4) 0.88, 0.012[Cr(phen)2(OH2)2]3+ 3.3(0.4) 0.93, 0.0005[Cr(O)2(phen)2]+ 125(19) 0.93, 0.007

a The values of MN/1000 BN cells/µmol of Cr were obtained bylinear regression analysis (Figure 4). b Two-sided ø2 test, compar-ing the responses of the highest Cr(III) concentration with that ofthe control for each respective assay (Figure 4).

Figure 5. Permeability results obtained from GFAAS analysisof cell pellets following treatment with 2 µmol of Cr(III) or Cr-(V) per dish or 0.5 µmol of Cr(VI) per dish for 4 h. Data pointsare means ( SD (n ) 3).

Permeability and Genotoxicity of Cr(III) and Cr(V) Chem. Res. Toxicol., Vol. 13, No. 8, 2000 745

of the aqua ligands at pH 7.4) to produce [Cr(en)2(OH)2]+,which is characterized by peaks at 392 and 528 nm inthe UV/vis spectrum (23). This ligand-exchange reactionwas quite slow in MEM, however, as determined by the2 h study that showed that the peaks at 352 and 456nm, indicative of [Cr(en)3]3+, did not shift (Figure 1).Inskeep et al. (25) showed that the aqua ligands of cis-[Cr(phen)2(OH2)2]3+ are deprotonated at pH 7.4 (pKa1 )3.4 and pKa2 ) 6.0, reactions 1 and 2, respectively). Thiscoincides with a shift of λmax from 497 nm for the diaquaspecies to 510 nm for the mixed aqua/hydroxo species to519 nm for the dihydroxo species (25, 26, 41).

The observation of a peak at 519 nm (pH 7.4) shows thatcis-[Cr(phen)2(OH)2]+ was produced immediately after thedissolution of the corresponding diaqua complex inMEM-P solution. The peptide complex mer-[Cr(glygly)2]-

represented a negatively charged structure that maymimic those found in some of the Cr(III) products of theintracellular reduction of Cr(VI). Its electronic absorptionspectrum in MEM-P solution revealed a degree of insta-bility, as shown by the 30% decay of the visible peak atλmax ) 550 nm. This was consistent with ligand-exchangereactions of glygly with the ligands present in the MEM-Psolution, as the decay of the peak corresponds to thegeneration of Cr(III) species with lower εmax values (42).Finally, the addition of trans-[Cr(salen)(OH2)2]+ to theMEM results in the partial deprotonation of one of theaqua ligands (pKa1 ) 7.54) in producing a mixture of thecation and the neutral species, trans-[Cr(salen)(OH2)-(OH)] (43). The small decrease in the absorption of theUV/vis peaks associated with trans-[Cr(salen)(OH2)2]+

may be due to the production of further species resultingfrom ligand-exchange reactions. For instance, the com-plex undergoes ligand-exchange reactions with azide,thiocyanate, pyridine, imidazole, and nicotinic acid toform monosubstituted products and with oxalate to formthe disubstituted complex (43-45). With the exceptionof the glygly complex, the fact that there were only minorchanges in the UV/vis spectra of all Cr(III) complexes inthe medium indicated that they retained their integritiesduring the assay period. Therefore, some possible factorscontributing to the differences in the genotoxicitiesbetween the different Cr(III) complexes may be thecharge of the complexes, and the physical and/or chemicalproperties imparted by the ligands.

The lack of a detectable Cr accumulation (GFAAS andPIXE analysis) in V79 cells exposed to 0.4 mM [Cr(en)3]3+,mer-[Cr(glygly)2]-, cis-[Cr(phen)2(OH)2]+, or trans-[Cr-(salen)(OH2)2]+ for 4 h contrasts with the results obtainedfollowing exposure to Cr(VI) and Cr(V) (0.1 mM Cr for 4h) (22). The low permeabilities of the Cr(III) complexesare consistent with those reported by a number ofinvestigators (46-49). For example, Kortenkamp et al.(48) showed that the intracellular Cr detected in humanerythrocytes following exposure to Cr(III) complexes (1mM for 1 h) was approximately 0.5-2.0 × 10-18 mol/cell.It was not possible to detect any differences in the uptakeof the complexes using GFAAS or PIXE analysis of cells

in this study due to the low permeability of V79 cells toCr(III). Electrothermal atomic absorption studies per-formed by Kortenkamp et al. (48) also showed thatintracellular Cr concentrations were similar when humanerythrocytes were exposed to the following Cr(III) com-plexes: [Cr(phen)2Cl2]+, [Cr(bipy)2Cl2]+, [Cr(2,4-pen-tanedionato)3], [Cr(glycinato)3], [Cr(glutathionato)2]-, and[Cr(cysteinato)2]-. The study concluded that there waslittle discrimination among the permeability propertiesof the complexes based on charge and ligand lipophilicity(48).

Consistent with previous studies, the cytotoxicities ofthe Cr(III) complexes were much lower than those of Cr-(V) and Cr(VI) complexes (46, 50, 51). Following identi-cally performed studies with V79 cells (44), negativelycharged Cr(V) and Cr(VI) complexes exhibited LD50

values in the range of 0.05-0.1 µmol of Cr/dish, whilethe LD50 values for Cr(III) complexes were in the rangeof 5-30 µmol/dish. There was no evidence that cytotox-icity was enhanced by more anionic species. For instance,trans-[Cr(salen)(OH2)2]+ (existing in equilibrium withtrans-[Cr(salen)(OH2)(OH)]) was the most toxic, while thenegatively charged mer-[Cr(glygly)2]- was the least toxicof the complexes.

There are distinct differences between the Cr(III)complexes containing aliphatic groups and those contain-ing aromatic imine ligands. While the difference was notas significant in the cytotoxicity assays, i.e., trans-[Cr-(salen)(OH2)2]+ was the most toxic, followed by cis-[Cr-(phen)2(OH2)2]3+ and [Cr(en)3]3+ (both with similar tox-icities), there was a marked difference in the genotoxicityassays. There was an increased incidence of MN followingexposure to cis-[Cr(phen)2(OH2)2]3+ or trans-[Cr(salen)-(OH2)2]+, although no well-defined dose response wasobserved following exposure to [Cr(en)3]3+ or mer-[Cr-(glygly)2]-, even at extremely high Cr(III) concentrations(30 µmol/dish). The significantly different genotoxicresponses induced by trans-[Cr(salen)(OH2)2]+ and mer-[Cr(glygly)2]- suggest that a negative charge of thecomplex does not play a major role in producing thegenotoxic effects. In previous studies, we have shown thatCr(V) mutagenicities and cytotoxicities were consistentwith the chemical reactivities of the Cr(V) complexes (19,22, 52). While there is no consistency between ligand-exchange reactions of the Cr(III) complexes and theirrespective genotoxicity, there is one important chemicalreaction that both cis-[Cr(phen)2(OH2)2]3+ and trans-[Cr-(salen)(OH2)2]+ undergo. Unlike the other complexes, bothof these imine complexes are readily oxidized to Cr(V)as reported previously (18, 19, 31, 32).1,2

The permeability of [Cr(O)2(phen)2]+ was significantlygreater than that of cis-[Cr(phen)2(OH2)2]3+, suggestingthat the Cr(V) species assisted uptake. The high geno-toxicity of [Cr(O)2(phen)2]+ was consistent with the

1 Abbreviations: AAS, atomic absorption spectroscopy; ANOVA,analysis of variance; bipy, 2,2′-bipyridine; BN, binucleated; CHO,Chinese hamster ovary; en, 1,2-ethanediamine; GFAAS, graphitefurnace atomic absorption spectroscopy; glygly, glycylglycine; GM,growth medium; HGPRT, hypoxanthine-guanine phosphoribosyl trans-ferase; LMWCr, low-molecular weight chromium-binding substance;MEM, minimal essential medium; MEM-P, minimal essential mediumwithout phenol red; MN, micronuclei; ox, oxalato(2-); phen, 1,10-phenanthroline; PIXE, particle-induced X-ray emission; salen, N,N′-ethylenebis(salicylideneiminato).

2 Sulfab, Y., and Nasreldin, M. Synthesis and characterization ofdioxochromium(V) complexes with 1,10-phenanthroline and 2,2′-bipy-ridine. Electron transfer between dioxochromium(V) with an imine-oxine copper(II) complex (to be submitted for publication).

cis-[Cr(phen)2(OH2)2]3+ + OH- h

cis-[Cr(phen)2(OH2)(OH)]2+ + H2O (1)

cis-[Cr(phen)2(OH2)(OH)]2+ + OH- h

cis-[Cr(phen)2(OH)2]+ + H2O (2)

746 Chem. Res. Toxicol., Vol. 13, No. 8, 2000 Dillon et al.

increased level of Cr detected in the cells. If the geno-toxicity of [Cr(O)2(phen)2]+ were solely attributed to theincreased Cr uptake, however, then the genotoxicity ofcellular Cr would be expected to be similar to that of cis-[Cr(phen)2(OH2)2]3+, but reference to Table 2 shows thatthis is not the case. Furthermore, it has been shown thatthe Cr(V) species cleaves DNA at physiological pH values(19). Although it has been shown that trans-[Cr(salen)-(OH2)2]+ is readily oxidized to [CrO(salen)]+, as evidencedby the rapid production of the dark blue coloration, thisCr(V) species is also unstable in cell medium at pH 7.4(19, 32). Consequently, it is not surprising that thegenotoxicity of [CrO(salen)]+ closely resembled that oftrans-[Cr(salen)(OH2)2]+. Importantly, the permeabilitiesand consequently the cellular genotoxicities of the twocomplexes were also similar. This raises the question asto whether the uptake and genotoxicity are assisted bythe ligand lipophilicity or whether they are being oxidizedon the cell wall to assist in permeability and consequentlygenotoxicity. Previously, in vitro DNA studies haveshown that [CrO(salen)]+ can interact strongly with DNAat physiological pH values in the presence of the oxidant(19). Importantly, the genotoxicity of the Cr(salen) com-plexes is much greater than those of [Cr(glygly)2]- and[Cr(en)3]3+ and is similar to that of [Cr(O)2(phen)2]+. Thepermeability of the complex is not significantly greaterthan those of the other Cr(III) complexes, although thegenotoxicity and, consequently, cellular genotoxicity aremarkedly greater than those of the other Cr(III) com-plexes.

In conclusion, if these Cr(V) species are producedintracellularly via oxidative enzymes (e.g., molybdoen-zymes, oxidases), they would contribute to the observedgenotoxic effects (18, 22). Furthermore, Cr(III) complexescan be oxidized to Cr(V) complexes by organic hydro-peroxides, as occurs with the macrocyclic tetraamido-Nligands (53). Such organic peroxides are produced asnormal metabolic byproducts of inter- and intracellularprocesses (54-57). The mechanism implicating the in-volvement of a Cr(V) species is consistent with the resultsreported by Sugden et al. (58), where it was shown that[Cr(bipy)2Cl2]+, which exhibits reactivities similar to thatof the Cr phen analogue, was mutagenic only whenbacterial cells were incubated under aerobic conditions.This difference was previously attributed to the involve-ment of OH• radicals, but as outlined in a recent paper,the results are more consistent with the involvement ofCr(V) (19). It was also shown by Sugden et al. (59) thatboth Cr(III) (cis-[Cr(phen)2Cl2]+ and cis-[Cr(bipy)2Cl2]+)and Cr(VI) complexes reverted S. typhimurium strainsthat were sensitive to oxidative mutations (TA102 andTA2638). This implies that both Cr(III) and Cr(VI)complexes may exert their mutagenicities by similar

mechanisms, and it is feasible that Cr(V) intermediatesare responsible.

Despite the low potencies of the Cr(III) complexes, highconcentrations of certain complexes, particularly thosecontaining imine ligands, are genotoxic in mammaliancells. In conclusion, evidence shown here indicates thatCr(III) complexes that are readily oxidized to Cr(V)should be considered as potential carcinogens. Thisfinding has potential adverse health implications withrespect to the increased use of Cr(III) complexes asdietary supplements (20). It is also important in under-standing the potential hazards of Cr(III) complexes thatare often encountered in teaching and research labora-tories.

Acknowledgment. We thank Mr. Tony Romeo fortechnical support. P.A.L. is grateful for support from theAustralian Research Council (ARC) and the NationalHealth and Medical Research Council. G.J.F.L. is gratefulfor support from the ARC. The views expressed in thisarticle are those of the authors and do not necessarilyreflect those of NOHSC.

References

(1) Langard, S. (1982) Biological and environmental aspects ofchromium, Elsevier Biomedical, Amsterdam.

(2) Anderson, R. A. (1998) Chromium, glucose intolerance anddiabetes. J. Am. Coll. Nutr. 17, 548-555.

(3) Vincent, J. B. (1999) Mechanisms of chromium action: low-molecular-weight chromium-binding substance. J. Am. Coll. Nutr.18, 6-12.

(4) Anderson, R. A. (1997) Nutritional factors influencing the glucose/insulin systemschromium. J. Am. Coll. Nutr. 16, 404-410.

(5) Anderson, R. A., Cheng, N. Z., Bryden, N. A., Polansky, M. M.,Cheng, N. P., and Chi, J. M. (1997) Elevated intakes of supple-mental chromium improve glucose and insulin variables inindividuals with type 2 diabetes. Diabetes 46, 1786-1791.

(6) Mooney, K. W., and Cromwell, G. L. (1997) Efficacy of chromiumpicolinate and chromium chloride as potential carcass modifiersin swine. J. Anim. Sci. 75, 2661-2671.

(7) Lukaski, H. C., Bolonchuk, W. W., Siders, W. A., and Milne, D.B. (1996) Chromium supplementation and resistance training:effects on body composition, strength, and trace element statusof men. Am. J. Clin. Nutr. 63, 954-965.

(8) Trent, L. K., and Thiedingcancel, D. (1995) Effects of chromiumpicolinate on body composition. J. Sport Med. 35, 273-280.

(9) Speetjens, J. K., Collins, R. A., Vincent, J. B., and Woski, S. A.(1999) The nutritional supplement chromium(III) tris(picolinate)cleaves DNA. Chem. Res. Toxicol. 12, 483-487.

(10) Bagchi, D., Bagchi, M., Balmoori, J., Ye, X., and Stohs, S. J. (1997)Comparative induction of oxidative stress in cultured J774A.1macrophage cells by chromium picolinate and chromium nicoti-nate. Mol. Pathol. Pharmacol. 97, 335-346.

(11) Stearns, D. M., and Armstrong, W. H. (1992) Mononuclear andbinuclear chromium(III) picolinate complexes. Inorg. Chem. 31,5178-5184.

(12) Stearns, D. M., Wise, J. P., Patierno, S. R., and Wetterhahn, K.E. (1995) Chromium(III) picolinate produces chromosome damagein Chinese hamster ovary cells. FASEB J. 9, 1643-1469.

(13) IARC (1990) IARC Monographs on the Evaluation of CarcinogenicRisk to Humans: Chromium, Nickel and Welding, Vol. 49, pp 49-508, International Agency for Research on Cancer, Lyon, France.

(14) Yassi, A., and Nieboer, E. (1988) Carcinogenicity of chromiumcompounds. In Chromium in the Natural and Human Environ-ments (Nriagu, J. O., and Nieboer, E., Eds.) pp 443-495, Wiley,New York.

(15) Warren, G., Schultz, P., Bancroft, D., Bennett, K., Abbott, E. H.,and Rogers, S. (1981) Mutagenicity of a series of hexacoordinatechromium(III) compounds. Mutat. Res. 90, 111-118.

(16) Beyersmann, D., and Koster, A. (1987) On the role of trivalentchromium in chromium genotoxicity. Toxicol. Environ. Chem. 14,11-22.

(17) Hashimoto, T., Negishi, T., Namba, T., Hayakawa, S., andHayatsu, H. (1979) Mutagenicity of quinoline derivatives andanalogs: quinoxaline 1,4-dioxide is a potent mutagen. Chem.Pharm. Bull. 27, 1954-1956.

Table 2. Summary of the Cellular Genotoxicity of the CrComplexes

complex

cellularCr level

(×10-15 g)

genotoxicity(MN/1000

BN/µmol of Cr)

genotoxicity/cellular Cr

level (MN/1000BN) (×1012)

[Cr(glygly)2]- 8.2 0.01 1.2[Cr(en)3]3+ 17.3 0.1971 11.4[Cr(salen)(OH2)2]+ 13.3 13.56 1020[CrO(salen)]+ 10.3 18.18 1760[Cr(phen)2(OH2)2]3+ 7.2 3.29 458[Cr(O)2(phen)2]+ 65.7 125 1902

Permeability and Genotoxicity of Cr(III) and Cr(V) Chem. Res. Toxicol., Vol. 13, No. 8, 2000 747

(18) Dillon, C. T. (1995) The role of Cr(V) complexes in chromium-induced genotoxicities. Ph.D. Thesis, The University of Sydney,NSW, Australia.

(19) Dillon, C. T., Lay, P. A., Bonin, A. M., Dixon, N. E., and Sulfab,Y. (2000) DNA interactions and bacterial mutagenicities of somechromium(III) imine complexes and their chromium(V) analogues.Evidence for chromium(V) intermediates in the genotoxicity ofchromium(III). Aust. J. Chem. (in press).

(20) Lefavi, R. C., Anderson, R. A., Keith, R. E., Wilson, C. D.,McMillan, J. L., and Stone, M. H. (1992) Efficacy of chromiumsupplementation in athletes: emphasis on anabolism. Int. J. SportNutr. 2, 111-122.

(21) Stearns, D. M., Belbruno, J. J., and Wetterhahn, K. E. (1995) Aprediction of chromium(III) accumulation in humans from chro-mium dietary supplements. FASEB J. 9, 1650-1657.

(22) Dillon, C. T., Lay, P. A., Bonin, A. M., Cholewa, M., Legge, G. J.F., Collins, T. J., and Kostka, K. L. (1998) Permeability, cytotox-icity, and genotoxicity of chromium(V) and chromium(VI) com-plexes in V79 Chinese hamster lung cells. Chem. Res. Toxicol.11, 119-129.

(23) Gillard, R. D., and Mitchell, P. R. (1972) Tris(diamine)chromium-(III) salts. Inorg. Synth. 13, 184-186.

(24) Murdoch, C. M. (1989) Chromium(III) and the glucose tolerancefactor. Ph.D. Thesis, The University of Sydney, NSW, Australia.

(25) Inskeep, R. G., and Bjerrum, J. (1961) Preliminary investigationsof the reactions of the hexaquo chromium(III) ion with R,R′-dipyridyl and with o-phenanthroline. Acta Chem. Scand. 15, 62-68.

(26) Inskeep, R. G., and Benson, M. (1961) Further studies of thehexaquo chromium(III) ion o-phenanthroline reaction. J. Inorg.Nucl. Chem. 20, 290-294.

(27) Yamada, S., and Iwasaki, K. (1969) Chromium(III) complexescontaining bis(salicylaldehyde)ethylenediiminate as a ligand.Bull. Chem. Soc. Jpn. 42, 1463.

(28) Yamada, S., and Iwasaki, K. (1971) Chromium(III) complexescontaining N,N′-ethylene-bis(salicylideneiminate) as a ligand.Inorg. Chim. Acta 5, 3-5.

(29) Coggon, P., McPhail, A. T., Mabbs, F. E., Richards, A., andThornley, A. S. (1970) Preparation, magnetic and electronicspectral properties of some chromium(III)-N,N′-ethylenebis(sali-cylideneiminato) complexes: crystal and molecular structure ofN,N′-ethylenebis(salicylideneiminato)diaquo chromium(III) chlo-ride. J. Chem. Soc., Sect. A, 3296-3303.

(30) Saha, H. K., Ghosh, S. K., and Ray Chaudhuri, T. K. (1983)Studies on the hydrolytic reactions of oxochloro complexes ofquinquivalent chromium and synthesis of oxalato derivatives.Indian J. Chem., Sect. A 22A, 346-347.

(31) Sulfab, Y., Al-Shatti, N. I., and Hussein, M. A. (1984) Synthesisand characterization of a novel periodato-bis(1,10-phenanthroli-ne)chromium(V) complex. Inorg. Chim. Acta 86, L59-L60.

(32) Srinivasan, K., and Kochi, J. K. (1985) Synthesis and molecularstructure of oxochromium(V) cations. Coordination with donorligands. Inorg. Chem. 24, 4671-4679.

(33) Ford, D. K., and Yerganian, G. (1958) Observations on thechromosomes of Chinese hamster cells in tissue culture. J. Natl.Cancer Inst. 21, 393-425.

(34) Cholewa, M., Turnbull, I. F., Legge, G. J. F., Weigold, H.,Marcuccio, S. M., Holan, G., Tomlinson, E., Wright, P. J., Dillon,C. T., Lay, P. A., and Bonin, A. M. (1995) Investigation of theuptake of drugs, carcinogens and mutagens by individual mam-malian cells using a scanning proton microprobe. Nucl. Instrum.Methods Phys. Res., Sect. B 104, 317-323.

(35) Fenech, M., and Morley, A. A. (1985) Measurement of micronucleiin lymphocytes. Mutat. Res. 147, 29-36.

(36) Fenech, M., and Morley, A. (1985) Solutions to the kinetic problemin the micronucleus assay. Cytobios 43, 233-246.

(37) Lasne, C., Gu, Z. W., Venegas, W., and Chouroulinkov, I. (1984)The in vitro micronucleus assay for detection of cytogenetic effectsinduced by mutagen-carcinogens: comparison with the in vitrosister-chromatid exchange assay. Mutat. Res. 130, 273-282.

(38) Howard, W. R., Hoffman, S. A., and Kochhar, T. S. (1992)Enhanced production of micronuclei by hexavalent chromium incultured CHO cells. Bull. Environ. Contam. Toxicol. 49, 535-540.

(39) Raj, A. S., and Heddle, J. A. (1980) Simultaneous detection ofchromosomal aberrations and sister-chromatid exchanges. Expe-rience with DNA intercalating agents. Mutat. Res. 78, 253-260.

(40) Pagano, M., and Gauvreau, K. (1993) Principles of Biostatistics,Duxbury, Belmont.

(41) Hancock, M. P., Josephson, J., and Schaffer, C. E. (1976) Bis-(2,2′-bipyridine) and bis(1,10-phenanthroline) complexes of chro-mium(III) and cobalt(III). Acta Chem. Scand., Sect. A 30, 79-97.

(42) Larkworthy, L. F., Nolan, K. B., and O’Brien, P. (1987) Chromium.In Comprehensive Coordination Chemistry (Wilkinson, G., Ed.)Vol. 3, pp 716-718, Pergamon, Oxford, U.K.

(43) Lloret, F., Julve, M., Mollar, M., Castro, I., LaTorre, J., Faus, J.,Solans, X., and Morgenstern-Badarau, I. (1989) Formation insolution, synthesis, and electrochemical study of oxalato com-plexes of N,N′-ethylenebis(salicylideneiminato)chromium(III) andiron(III): Crystal structures of piperidinium [N,N′-ethylenebis-(salicylideneiminato)](oxalato-O1,O2)-chromate(III) and ferrate-(III). J. Chem. Soc., Dalton Trans., 729-738.

(44) Prasad, D. R., Ramasami, T., Ramaswamy, D., and Santappa, M.(1980) Substitution lability of chromium(III) complexes withground-state distortion. Inorg. Chem. 19, 3181-3182.

(45) Prasad, D. R., Ramasami, T., Ramaswamy, D., and Santappa, M.(1982) Kinetics and mechanisms of the equilibration reactions ofdiaquochromium(III)-Schiff base derivatives, Cr(Schiff base)-(H2O)2

+, and their conjugate bases with thiocyanate, azide,imidazole, pyridine and nicotinic acid as ligands. Inorg. Chem.21, 850-854.

(46) Popper, H. H., Grygar, E., Ingolic, E., and Wawschinek, O. (1993)Cytotoxicity of chromium-III and -VI compounds. I. In vitrostudies using different cell culture systems. Inhalation Toxicol.5, 345-369.

(47) Gray, S. J., and Sterling, K. (1950) The tagging of red cells andplasma proteins with radioactive chromium. J. Clin. Invest. 29,1604-1613.

(48) Kortenkamp, A., Beyersmann, D., and O’Brien, P. (1987) Uptakeof chromium(III) complexes by erythrocytes. Toxicol. Environ.Chem. 14, 23-32.

(49) Barr-David, G., Hambley, T. W., Irwin, J. A., Judd, R. J., Lay, P.A., Martin, B. D., Bramley, R., Dixon, N. E., Hendry, P., Ji, J.-Y.,Baker, R. S. U., and Bonin, A. M. (1992) Suppression byvanadium(IV) of chromium(V)-mediated DNA cleavage and chro-mium(VI/V)-induced mutagenesis. Synthesis and crystal structureof the vanadium(IV) complex. (NH4)[V(O){HOC(Et)2COO}{OC-(Et)2COO}]. Inorg. Chem. 31, 4906-4908.

(50) Biedermann, K. A., and Landolph, J. R. (1990) Role of valencestate and solubility of chromium compounds on induction ofcytotoxicity, mutagenesis and anchorage independence in diploidhuman fibroblasts. Cancer Res. 50, 7835-7842.

(51) Levis, A. G., Bianchi, V., Tamino, G., and Pegoraro, B. (1978)Cytotoxic effects of hexavalent and trivalent chromium on mam-malian cells in vitro. Br. J. Cancer 37, 386-396.

(52) Dillon, C. T., Lay, P. A., Bonin, A. M., Dixon, N. E., Collins, T. J.,and Kostka, K. L. (1993) In vitro DNA damage and mutationsinduced by a macrocyclic tetraamide chromium(V) complex:implications for the role of Cr(V) peptide complexes in chromium-induced cancers. Carcinogenesis 14, 1875-1888.

(53) Collins, T. J., Slebodnick, C., and Uffelman, E. S. (1990) Chro-mium(V)-oxo complexes of macrocyclic tetraamido-N ligandstailored for highly oxidized middle transition metal complexes:a new 18O-labeling reagent and a structure with four nonplanaramides. Inorg. Chem. 29, 3433-3436.

(54) Fu, S. L., Hick, L. A., Sheil, M. M., and Dean, R. T. (1995)Structural identification of valine hydroperoxides and hydroxideson radical-damaged amino acid, peptide and protein molecules.Free Radical Biol. Med. 19, 281-292.

(55) Davies, M. J., Fu, S. L., and Dean, R. T. (1995) Protein hydro-peroxides can give rise to reactive free radicals. Biochem. J. 305,643-649.

(56) Fu, S. L., Gebicki, S., Jessup, W., Gebicki, J. M., and Dean, R. T.(1995) Biological fate of amino acid, peptide and protein hydro-peroxides. Biochem. J. 311, 821-827.

(57) Soszynski, M., Filipiak, A., Bartosz, G., and Gebicki, J. M. (1996)Effect of amino acid peroxides on the erythrocyte. Free RadicalBiol. Med. 20, 45-51.

(58) Sugden, K. D., Burris, R. B., and Rogers, S. J. (1990) An oxygendependence in chromium mutagenesis. Mutat. Res. 244, 239-244.

(59) Sugden, K. D., Geer, R. D., and Rogers, S. J. (1992) Oxygenradical-mediated DNA damage by redox active Cr(III) complexes.Biochemistry 31, 11626-11631.TX0000116

748 Chem. Res. Toxicol., Vol. 13, No. 8, 2000 Dillon et al.