Introduction

It is well known that enzymatic antioxidants like superoxide dismutase (SOD), catalase (CAT), and enzymes involved in glutathione metabolism, like glutathione peroxidase (GSH-Px) and glutathione reductase (GR; Matés 2000; Urso and Clarkson 2003), as well as non-enzymatic antioxidants, such as reduced glutathione (GSH), vitamin E, β-carotene, and flavonoids, play an important role in the defense of the organism against reactive oxygen species (ROS) and free radical attacks (Meister 1992; Lindsay and Astley 2002; Bliska et al. 2007). They act as a coordinated system, where deficiencies in one component may affect the efficiency of the other one (Evans and Halliwell 2001).

Vanadium (V) is an element with complex chemistry and it can exist in many oxidation states (from −1 to +5), which enables the element to function as an electron transfer catalyst in a wide variety of reactions (Badmaev et al. 1999). Since it is a transitional metal having redox properties, it may enter redox processes and produce free radicals (Byczkowski and Kulkarni 1998), in this way inducing lipid peroxidation (LPO) in both in vitro (Donaldson and LaBella 1983; Younes et al. 1991; Younes and Strubelt 1991) and in vivo conditions (Elfant and Keen 1987; Sheriff 1991; Thompson and McNeill 1993; Russanov et al. 1994) and disturbing, thereby, the pro-antioxidative balance in the organism. V may also decrease the activities of some antioxidative enzymes in erythrocytes (RBCs; Soussi et al. 2006), in the liver (Oster et al. 1993; Saxena et al. 1993; Russanov et al. 1994), kidneys (Russanov et al. 1994), and testes (Chandra et al. 2007) of rats. It may diminish the concentrations of GSH and/or l-ascorbic acid in blood and/or internal organs (especially in liver and/or kidneys) of these animals (Chakraborty et al. 1977; Elfant and Keen 1987; Saxena et al. 1993; Zaporowska 1994; Zaporowska et al. 1997) and act as a mitogen, tumor promoter and co-carcinogen. On the other hand, V compounds may prevent chemical carcinogenesis and act as potential antimetastatic agents, which suggests their use in treatment of cancer in humans (Papaioannou et al. 2004; Laizé et al. 2010). An osteogenic role of V and, thereby, its possible therapeutic application in osteoporosis has been described, too (Facchini et al. 2006; Laizé et al. 2010). Due to its antioxidant action, vanadate may also act as a scavenger of superoxide radicals (O •−2 ; Matsubara et al. 1995). Its antioxidant properties have been mentioned, too, with respect to 1,2-dimethylhydrazine (DMH)-induced colon cancer in male rats (Kanna et al. 2004). The chemopreventive effects of V, as the authors reported, were accompanied by changes in the activities of some antioxidant enzymes, normalization of which might be caused by the antioxidant properties of the metal interacting with free radicals. Thus, V may behave as an antioxidant or as a pro-oxidant, depending on experimental conditions, the dose and form of its species as well as cell type (Matsubara et al. 1995; Badmaev et al. 1999; Aureliano and Crans 2009). Its “biphasic” effect—essentiality at low concentrations and toxicity at high concentrations is well known and has been presented in the literature (Chatterjee and Bishayee 1998).

As far as the antioxidant properties of trivalent chromium (Cr+3) are concerned, it has been reported that Cr3+ as chromium polynicotinate led to reduction in the hepatic and renal LPO in rats with hypertension (Preuss et al. 1998); as chromium picolinate (CrPic) it produced a decrease in LPO in the serum in cold-stressed Japanese quails (Sahin et al. 2003b), in laying hens and in broiler chickens reared at a low (Sahin et al. 2002; Onderci et al. 2003) and a high ambient temperature (Sahin et al. 2003a), respectively, and in hepatic LPO in spontaneously hypertensive rats (Preuss et al. 1997). Supplementation with chromium (Cr), as chromium pidolate or chromium yeast has also been shown to (a) result in a decrease in the plasma thiobarbituric acid reactive substances (TBARS) level in Tunisian adult subjects with type 2 diabetes mellitus (Anderson et al. 2001) and (b) be an effective treatment strategy to minimize increased oxidative stress in type 2 diabetes mellitus patients whose glycated hemoglobin (HbA1C) level was >8.5% (Cheng et al. 2004), respectively. Chromium chloride (CrCl3) has been reported, too, to be able to reduce V toxicity in rats (Curran 1954), chicks (Hill and Matrone 1970; Hafez and Kratzer 1976; Cupo and Donaldson 1987), and in laying hens (Jensen and Maurice 1980). Some of the antagonistic effects between those metals included alterations in growth, mortality, oxidative phosphorylation, lipogenesis and cholesterol, and fatty acid synthesis.

On the other hand, Cr+3 is well known to be capable of producing free radicals from hydrogen peroxide (H2O2) and lipid hydroperoxides (LOOH; Shi et al. 1993). Its reduction to divalent chromium (Cr+2) by some biological reductants such as l-cysteine and reduced nicotinamide adenine dinucleotide (NADH) and its reaction with H2O2 to yield hydroxyl radicals (HO•) have been reported, too (Ozawa and Hanaki 1990). This element, likewise V (Keller et al. 1988), may be involved in redox cycling with ROS production, which clearly shows that Cr+3 may be one of the biologically most active oxidation states. It may maintain redox activity in vivo through Cr3+/pentavalent chromium (Cr5+) or Cr3+/Cr2+ redox pairs, thus re-activating the radical cascade (Bal and Kasprzak 2002). So, Cr3+ may enter redox processes and induce oxidative stress due to its redox properties. Moreover, Cr3+ may accumulate in animals and humans (Stearns et al. 1995) and be genotoxic (Błasiak and Kowalik 2000; Codd et al. 2001; O’Brien et al. 2003), likewise V, (Owusu-Yaw et al. 1990; Bonnefont-Rousselot 2004; Leopardi et al. 2005). Its interactions with deoxyribonucleic acid (DNA) have been presented in the literature (Codd et al. 2001; Blankert et al. 2003; Levina and Lay 2005). The study of Kirpnick-Sobol et al. (2006) revealed that Cr3+ (as CrCl3) is a more potent inducer of DNA deletions than hexavalent chromium (Cr6+) and that nutritional supplement Cr3+ increases DNA damage in vitro and in vivo, when ingested with drinking water. CrCl3 has also been demonstrated in the study of Lushchak et al. (2009) to induce stronger oxidative stress than Cr6+ at the same concentration. Besides CrCl3, CrPic and chromium nicotinate have also been shown to increase oxidative stress, the former more than the latter, and to cause DNA damage (Bagchi et al. 1995, 2002; Hepburn et al. 2003; Vincent 2003; Whittaker et al. 2005).

As it is well known, several Cr3+ salts such as chloride, picolinate and nicotinate, likewise V salts, are used as common and available dietary supplements, especially popular among athletes, who take them for a long time (Gerrard et al. 1993; Anderson 1998). On the one hand, V and Cr dietary supplementation has been postulated to facilitate weight loss, a decrease in serum cholesterol and an increase in muscle mass (Lefavi et al. 1993; Anderson 1998; Barceloux 1999; Lukaski 1999; Lamson and Plaza 2002; Vincent 2003); on the other hand, prolonged consumption of high-doses of Cr3+ and/or V supplements has been reported to give rise to deleterious effects, too (Vincent 2003; Talbott and Hughes 2007). Kirpnick-Sobol et al. (2006) paid attention to the fact that even small amounts of absorbed Cr3+ are potentially dangerous. This observation, as the authors stressed, has important implications for public health due to the widespread and unregulated consumption of Cr3+-containing dietary supplements that are designed for efficient absorption. Stearns et al. (1995) have also indicated that Cr3+ supplementation might have possible adverse biological effects on the organism due to its accumulation in various tissues and that long-term use of Cr dietary supplements, especially in excessive doses and for extended periods, should be approached with caution. The safety of Cr3+, as Bagchi and co-investigators (2002) emphasize, is largely dependent on the ligand; therefore, adequate clinical studies should be warranted to demonstrate the safety and efficacy of Cr3+ for human consumption. CrPic, for example, the most popular and widely marketed trivalent chromium supplement, has been observed to produce an increase in LPO in liver and kidney as well as a decrease in the activity and concentration of some antioxidants in these two tissues in rats (Bagchi et al. 2002). Some of V nutritional supplements are also known to be a very important source of its exposure because they may provide a high-dose of this element (Barceloux 1999; Chemical Information Review Document 2008). Leopardi et al. (2005) stress that in consideration of V genotoxic properties and its ubiquitous occurrence in water and diet items, assessment of the risk posed by the intake of low doses of this metal is an important public health issue.

Therefore, the study of Cr3+ and V toxicity is still necessary and further examinations are required because the description of their possible side effects may be helpful in establishing the optimal dose of those elements for pharmacological use. V and Cr3+ are well known to display an insulin-mimetic action (Anderson 2000; Crans 2000; Anderson et al. 2001; Thompson and Orvig 2004), and there are trials to use them as therapeutic agents in the therapy of diabetes in humans. V, besides its insulin-mimetic properties, has been reported to prevent and/or treat pancreatic B-cell lesions induced by the streptozotocin treatment (Mongold et al. 1990). Thus, studies of their pro-oxidant potential and other toxicological aspects examined under different experimental conditions are reasonable and still needed. It is essential to realize that although a lot has already been written about V and Cr3+ with respect to their toxicity and physiological activities, not all aspects of their action in the organism have been sufficiently recognized and explained until now; especially, there is lack of information regarding changes in the antioxidant defense system caused by precise V and/or Cr3+ doses, ingested orally during a specific period of time. It is clear that the effects of those elements on the antioxidant system, and more generally on living organisms, depend not only on the form in which they are administered, but also on the dose ingested and time of exposure.

Therefore, taking into account the above data and the fact that there has been increased interest in V and Cr3+ supplements, we wanted to answer some questions: (1) about the extent to which both metals administered separately and in combination in drinking water at the concentrations used will influence the diet and fluid intake, body weight gain, antioxidative enzyme activities, malondialdehyde (MDA) level and l-ascorbic acid concentration, (2) whether they will reveal their pro-oxidant action on RBCs at the doses consumed, (3) whether they will interact to affect the above-mentioned parameters, and (4) whether the changes, if they occur, will result only from independent action of V and/or Cr. It is important to notice that until now the mutual interactions between V and Cr3+ have not been fully studied; therefore, we decided to explore this subject. Examination of interactions between V and trivalent Cr is important because it may have medical significance. The knowledge about their effects, in the future, might be helpful in the therapy in people who are intoxicated with V+5 and Cr3+ environmentally and/or in individuals who consume pharmaceutical preparations containing the two elements separately or together for a long time. To the best of our knowledge, this is the first study of the antioxidant defense system in erythrocytes under co-administration of V5+ and Cr3+. We chose erythrocytes because they are not only well equipped with antioxidative enzymes (SOD, CAT, cGSH-Px, GST) which scavenge free radicals and remove peroxides (Siems et al. 2000; Tsantes et al. 2006) as well as with GSH and l-ascorbic acid which are involved in detoxification of vanadium and in formation of Cr3+ and VO2+ (vanadyl) complexes (Hojo and Satomi 1991; Yamamoto 1995; Baran 1998; Karan et al. 2005), but also because they provide antioxidant protection to other tissues and organs (Siems et al. 2000). The measurement of the above-mentioned antioxidants in those cells has been considered essential and useful in evaluating possible oxidative stress caused by V5+ and/or Cr3+ administration in drinking water.

Materials and methods

Reagents

Kits for GSH, cGSH-Px (EC 1.11.1.9), GR (EC 1.6.4.2), GST (EC 2.5.1.18), CAT (EC 1.11.1.6), and Cu,Zn-SOD (EC 1.15.1.1) were obtained from OXIS International Inc., Portland, USA. Kit for Total Antioxidant Status (TAS) was obtained from Calbiochem, San Diego, CA, USA. l-ascorbic acid and nitric acid (65%, Suprapure) were obtained from Sigma Chemicals, St. Louis, USA and from Merck, Darmstadt, Germany, respectively. Buffered solution of physiological salt (PBS) was obtained from Serum and Vaccine Factory (BIOMED, Lublin, Poland). Ultra-pure water assigned for atomic absorption spectrometry (AAS) was received from water purification system Elix 10 (Millipore Corporation, Billerica, MA, USA). All the chemicals were of the highest quality available. Deionized water was used throughout.

Animals and diet

Outbred 2-month old, albino male Wistar rats obtained from Brwinów/Warsaw (Poland) and weighing approximately 241 g (mean) at the beginning of the experiment were used. All the rats were individually housed in stainless steel cages (one rat per cage) and kept in standard laboratory conditions at constant temperature (20–21°C), relative air humidity (55 ± 10%) and standard light–dark cycles (12 h/12 h). The experimental protocol was approved by the First Local Ethical Committee for Animal Studies in Lublin (registration number 48/2007). All the animals were randomly divided into four groups (14 animals per each group). Group I (control) received daily deionized water to drink from ARIES deionizer (Resin Tech., Inc., USA) over a 12-week period. The other three treatment groups received deionized water, to which V or/and Cr was added; Group II—water solution of sodium metavanadate—NaVO3 (SMV; Sigma, St. Louis, MO, USA) at a concentration of 0.100 mg V/mL (1.96 mM; pH 6.48), Group III—water solution of chromium chloride hexahydrate—CrCl3·6H2O (CC; Sigma, St. Louis, MO, USA) at a concentration of 0.004 mg Cr/mL (0.077 mM; pH 4.94), Group IV—water solution of SMV-CC at a concentration of 0.100 mg V and 0.004 mg Cr/mL (1.96 and 0.077 mM; pH 6.66) also over the period of 12 weeks. The mean start weight of rats in Groups I, II, III, and IV was about 240, 239, 238, and 245 g, respectively. V and Cr concentrations in drinking water were chosen on the basis of our previous studies (Zaporowska and Ścibior 1998; Terpiłowska et al. 2004; Ścibior 2005) and studies of other authors (Russanov et al. 1994; Bataineh et al. 1997). The V and Cr intake was calculated on the basis of the amount of SMV, CC, and SMV-CC solutions consumed by the rats. All the animals were fed with the same standard granulated rodent laboratory chow (Labofeed B; Fodder and Concentrate Factory, Kcynia, Poland), having the shape of pellets 12 mm in diameter, with the following ingredients per kilogram of Labofeed B (Table 1).

Table 1 Diet composition for animals used in the experiment
Full size table

Diet, fluids and deionized water were offered ad libitum and their consumption was monitored daily over the experimental period. Daily intake of fluids and water was measured with a measuring cylinder and the remaining diet plus additional spillage were weighed.

Biochemical analyses

Blood samples were taken from the jugular vein following anesthesia with pentobarbital (30 mg/kg body weight; intraperitonealy) into plastic tubes with bipotassium ethylene-diaminetetraacetic acid (K2EDTA) as an anticoagulant. The plasma was separated by centrifugation of whole blood (5 min, 1,500×g, 4°C) and used for determination of l-ascorbic acid concentration and TAS. The l-ascorbic acid content in the plasma was determined by Kyaw’s method (Kyaw 1978) and the results were expressed as micromoles per liter (μmol/L). As far as TAS is concerned, in this method 2,2′-azino-bis-3-ethyl-benzothiazoline-6-sulfonic acid (ABTS) is incubated with metmyoglobin (a peroxidase) and H2O2 to produce the radical cation (ABTS•+), which has a relatively stable blue-green color and can be monitored by reading the absorbance at 600 nm. A 20-µl sample was used in the analysis and the results were expressed as millimoles per liter (mmol/L).

The l-ascorbic acid concentration in RBCs, which were washed three times in ice-cold 0.9% sodium chloride (NaCl), was also determined by Kyaw’s method (Kyaw 1978) and the results were expressed as micromoles per liter (μmol/L). The MDA level in these cells was determined as TBARS using Stocks and Dormandy’s method (Stocks and Dormandy 1971), modified by Gilbert et al. (1984) and it was calculated using the molar extinction coefficient 1.56 × 105 M−1 cm−1. The results were expressed as nanomoles per milligram of hemoglobin (nmol/mg Hb). The GSH concentration in RBCs was determined according to the Bioxytech® GSH-420 assay procedure and the results were expressed as micromoles per liter (µmol/L). For determination of Cu,Zn-SOD, CAT, cGSH-Px, GR, and GST activities, washed RBCs were resuspended in four packed-cell volumes of ice-cold deionized water. For determination of Cu,Zn-SOD activity which was measured in the aqueous phase, erythrocyte lysates were treated with a cold extraction reagent (absolute ethanol/chloroform 62.5/37.5 v/v) in order to remove Hb; while measuring CAT activity, the erythrocyte lysates were diluted with a sample diluent, and the results for both enzymes were expressed in units per milligram of hemoglobin (U/mg Hb). To measure the activity of cGSH-Px, GR and GST (determined using 1-chloro-2,4-dinitrobenzene (CDNB) as substrate), the erythrocyte lysates were diluted with buffer—pH 7.6 or with a diluent—pH 7.5 or with a sample diluent, respectively, and the results were expressed as milliunits per milligram of hemoglobin (mU/mg Hb).

Hb concentration

The Hb concentration in the erythrocyte lysates was measured by the cyanmethemoglobin method using Alpha Diagnostics reagents (Warsaw, Poland) in order to determine the specific activity of Cu,Zn-SOD, CAT, cGSH-Px, GR, and GST.

Determination of V and Cr in the blood

The V and Cr concentrations in the chow and in the plasma were determined by inductively coupled plasma-atomic emission spectrometry (ICP-AES) using the Liberty II AX plasma emission spectrometer (VARIAN) with the Ultrasonic Nebulizer (U-5000 AT+, CETAC), whereas V and Cr concentrations in RBCs were determined by Graphite Furnace Atomic Absorption Spectrometry (GF-AAS) using the Z-8200 HITACHI atomic absorption spectrophotometer after microwave digestion. Before measurement, all the samples were diluted with ultra-pure water. An internal quality control was employed to ensure that the measurement process was reliable. The V and Cr concentrations in the chow were expressed as milligrams per kilogram (mg/kg); whereas in the plasma and in RBCs they were expressed as micrograms per milliliter (μg/mL).

Statistical analysis

The data were processed with Statistica version 12.0 PL for Windows. Normal distribution was tested by the Shapiro–Wilk’s test. Homogeneity of variances was analyzed employing Levene’s test. Since there were no significant deviations from the assumptions of the ANOVA, the two-way analysis of variance was used (two-way ANOVA; factors of vanadium and chromium, test F). If the two-way ANOVA indicated a significant V-treatment effect, Cr-treatment effect or V × Cr interaction, subsequent post-hoc comparisons between the individual groups were performed using Tukey’s multiple range test when the variances were homogeneous or the Dunnett’s T3 test when the variances were not homogeneous. Student’s ‘t’ test was applied for detection of significant differences in the concentrations of V in the plasma and RBCs only between Group II and IV (the two-way ANOVA was not performed in this case). Student’s ‘t’ test was also applied for comparison of the consumed V doses between Group II and IV and Cr doses between Group III and IV. The differences between the Groups were considered to be significant at P < 0.05. All the results were expressed as means ± SEM.

Results

During the 12 weeks of the experimental period, no distinctive differences in physical appearance and in motor behavior were observed in most of the rats treated with SMV and/or CC, compared with Group I. Four rats intoxicated with SMV alone, nine treated with CC alone and five exposed to SMV and CC in combination had transient diarrhea, whose main cause was probably the intake of V and/or Cr solutions at the concentrations used. More details of the diarrhea are presented in Table 2.

Table 2 The animals in which diarrhea occurred during a 12-week experimental period
Full size table

The two-way ANOVA revealed a significant main effect of V on the diet and fluid intake, body weight gain and food efficiency ratio (FER); and, as the results of those ANOVA analysis indicated, the changes in the above-mentioned parameters in the rats co-treated with SMV and CC only resulted from independent action of V (Table 3). Post-hoc comparisons demonstrated that the administration of SMV alone (Group II) or in combination with CC (Group IV) produced a decrease in the diet and fluid intake as well as in body weight gain and FER, compared with Group I (diet intake was lower by: 14.6%, P < 0.001 and 11.5%, P < 0.01; fluid intake by: 31%, P < 0.001 and 24.7%, P < 0.001; body weight gain by: 25%, P < 0.01 and 23%, P < 0.05; FER by: 16%, P < 0.05 and 17%, P < 0.05, respectively) and compared with Group III (diet intake was lower by: 12.5%, P < 0.01 and 9.5%, P < 0.05; fluid intake by: 29%, P < 0.001 and 22.7%, P < 0.001; body weight gain by: 29.8%, P < 0.001 and 28%, P < 0.01; FER by: 22.2%, P < 0.01 and 23%, P < 0.001, respectively; Table 3).

Table 3 Intake of diet and water/fluids as well as body weight gain and food efficiency ratio (FER) in the tested animal groups
Full size table

The rats from the particular experimental groups consumed with drinking water about 9 mg V/kg b.w./24 h (0.18 mmol V/kg b.w./24 h; Group II), 0.47 mg Cr/kg b.w./24 h (0.010 mmol Cr/kg b.w./24 h; Group III), 9.8 mg V (0.19 mmol) and 0.39 mg (0.007 mmol) Cr/kg b.w./24 h (Group IV; Table 4). The rats in Group IV ingested about 5% of V more, compared with the animals in Group II, but this difference was not statistically significant. However, the dose of Cr ingested by the rats in Group IV was lower (by 17%, P < 0.001), compared with that ingested by the animals in Group III (Table 4).

Table 4 Vanadium and chromium doses consumed by the rats with drinking water during 12 weeks
Full size table

V concentrations in RBCs and in the plasma in Group I and III were below the detection limit (Table 5). The exposure to SMV alone or in combination with CC resulted in a significant increase in V concentration in the plasma and RBCs, compared with Group I and III; but no significant changes in its content between the rats in Group II and IV were demonstrated (Table 5). Despite the fact that the two-way ANOVA of Cr concentration in RBCs yielded a significant main effect of Cr, post-hoc comparisons did not demonstrate any marked changes in this parameter in all the treated Groups of rats; although a visible tendency to an increase in Cr content in RBCs was observed. The concentration of Cr in the plasma also slightly increased but no significant changes were demonstrated, too (Table 5).

Table 5 Concentrations of vanadium and chromium in the erythrocytes (RBCs) and plasma in the tested animal groups
Full size table

A significant main effect of Cr on l-ascorbic acid concentration in the plasma was revealed by the two-way ANOVA and according to those ANOVA analysis the changes in the content of this antioxidant in the plasma of rats co-treated with SMV and CC were a consequence of independent action of Cr (Table 6). Post-hoc comparisons demonstrated that in the rats treated with CC alone, the plasma l-ascorbic acid concentration was higher (by 54%, P < 0.05), as compared to the animals in Group II. However, no significant changes were found in the values of MDA and l-ascorbic acid in RBCs and in the plasma TAS level in all the Groups of the rats studied (Table 6).

Table 6 The level of malondialdehyde (MDA) in the erythrocytes (RBCs) and the concentration of l-ascorbic acid in the plasma and erythrocytes as well as the total antioxidant status (TAS) in the plasma in the tested animal groups
Full size table

The two-way ANOVA did not reveal any significant main effect of V and/or Cr on the activities of Cu-Zn-SOD, CAT, cGSH-Px, and GR in RBCs but it indicated a significant effect of Cr on the GSH concentration in RBCs as well as a significant effect of Cr and V × Cr interaction on the GST activity in RBCs (Tables 7 and 8). Post-hoc comparisons demonstrated that the administration of CC alone or in combination with SMV resulted in an increase in GSH content (by 59%, P < 0.05 and by 53.6%, P < 0.01, respectively) and led to a decrease in the activity of GST (by 56.7%, P < 0.001 and by 40%, P < 0.01), compared with Group I. The administration of SMV alone also caused an increase in GSH level (by 28%, P > 0.05) and produced a decrease in GST activity (by 30%, P < 0.05), in comparison with Group I (Table 8). As the two-way ANOVA showed, the changes in GSH concentration in the rats following SMV and CC co-treatment resulted from independent action of Cr; however, the changes in GST activity in the same Group of rats were a consequence of independent effect of Cr and they also resulted from V–Cr interaction (Table 8).

Table 7 The activity of Cu, Zn-superoxide dismutase (Cu, Zn-SOD) and catalase (CAT) in the erythrocytes (RBCs) in the tested animal groups
Full size table
Table 8 The concentration of reduced glutathione (GSH) and activity of cellular glutathione peroxidase (cGSH-Px), glutathione reductase (GR) and glutathione transferase (GST) in the erythrocytes (RBCs) in the tested animal groups
Full size table

Discussion

We studied the changes induced in certain blood biochemical parameters upon administration to rats SMV and CC separately and in combination in drinking water. Although it was difficult to compare our results with the studies of other researchers, because of different V and Cr compounds used, modes of their administration, concentrations and length of exposure, we decided to present some of them taking into account the above-mentioned differences to compare toxicological effects of the two different metals, especially in the sense of their some toxic effects, which will partially be presented in the further part of the discussion.

As it was here demonstrated, V at the dose of 9 mg V/kg b.w./24 h produced a significant fall in the diet and fluid intake, body weight gain and FER (Table 3). The decreased diet and/or fluid intake and/or body weight gain in rats after administration of vanadium compound solutions as vanadate or vanadyl was also observed in our previous studies (Zaporowska et al. 1997; Ścibior and Zaporowska 1998; Zaporowska and Ścibior 1999; Zaporowska et al. 1999; Ścibior and Zaporowska 2002; Ścibior 2005) and in the studies of other authors (Meyerovitch et al. 1989, Thompson and McNeill 1993, Thompson et al. 1993, Russanov et al. 1994, Dai et al. 1995, Poggioli et al. 2001, Thompson et al. 2002, Mohamad et al. 2004, Kordowiak et al. 2005, Dąbroś et al. 2006). However, other investigators did not demonstrate any significant changes in the diet and fluid intake (Bishayee and Chatterjee 1995; De Tata et al. 2000; Poggioli et al. 2001) or in body weight gain (Dai et al. 1995; Thompson et al. 2002) after treating rats with vanadate or vanadyl in drinking water. A reduction in diet intake and/or body weight gain after administration of vanadium was also shown in other animals: in laying hens (Ousterhout and Berg 1981), sheep (Hansard et al. 1982), chicks (Cupo and Donaldson 1987), and in male mice (Llobet et al. 1993).

Reduced fluid intake after SMV treatment, as it was shown (Table 3), might be caused by an altered taste of vanadate solution which might lead to development of rats’ aversion to drink, and in consequence, cause its decreased consumption. Heyliger et al. (1985) suggested that vanadate ingestion may induce water retention, which is subsequently responsible for the diminished water intake. Mongold et al. (1990) paid attention to a hypodipsic effect of the vanadium salts. Thus, this phenomenon still remains to be explained. The addition of CrCl3 (CC) into SMV solution (Group IV) did not improve the SMV-CC taste and did not elevate its consumption. The animals drank significantly less SMV-CC solution compared with the control rats and with those drinking CC alone (Group III). It is important to mention that this decrease, as the two-way ANOVA indicated, was only a consequence of independent action of V. On the other hand, the administration of SMV and CC in combination did not result in further reduction in fluid intake, which allows us to claim that the rats simultaneously treated with SMV and CC did not develop stronger aversion to drink (Table 3). It is also worth noting that the animals in Group IV ingested a significantly lower Cr dose in comparison with that consumed by the rats treated with CC alone (Table 4) and, as a result, Cr concentration in the plasma and RBCs in these rats was lower (but not significantly) than in the animals of Group III (Table 5).

As far as the inhibitory effect of orally (via drinking water) administered NaVO3 to male Wistar rats on diet intake is concerned, Meyerovitch et al. (1989) suggest that this effect may result from direct action of the ion in the central nervous system (CNS) involving stimulation of local glucose uptake. Thus, as the authors stress, vanadate is capable of blocking diet intake by a specific effect in the CNS mentioned above. This local effect, which leads to a rise in intracellular glucose, transduces a signal that can suppress diet intake and, consequently, body weight gain. Other authors reported that decreased appetite produced by vanadium administered in drinking water may be linked to decreased neuropeptide Y (NPY) levels in the hypothalamus (Wang et al. 2001), which was observed in the Zucker fatty rats, characterized by hyperinsulinemia, high NPY levels, and hyperphagia. However, this mechanism cannot be directly related to our experiment because, among others, a different V compound—bis(maltolato)oxovanadium(IV) (BMOV) and a different animal model were used by the mentioned authors. In contrast, the studies of Malabu et al. (1994) demonstrated an increase (higher than in the control) in the hypothalamic NPY concentration in the non-diabetic and diabetic rats treated by gavage NaVO3. The difference in the results obtained by Wang et al. (2001) and by Malabu et al. (1994), first of all, was due to the different animal model used and to the application of different experimental conditions.

Cr, at the dose of 0.47 mg Cr/kg b.w./24 h, ingested by rats via drinking water during 12 weeks, did not produce any significant changes in the diet and fluid intake and in body weight gain (Table 3), which is in agreement with our previous studies, in which the level of the consumed Cr dose during only 6 weeks was very similar to that presented above (Ścibior 2003; 2005). Unchanged diet intake was also shown in pigs (Mooney and Cromwell 1997), in broiler chicks (Kalaycioglu et al. 1999), and in laying hens (Piva et al. 2003) when CrCl3 was added into the diet. More data about the changes in body weight gain in rats and also in other animals after Cr+3 administration were presented and discussed in our previous paper (Ścibior 2005).

The decreased body weight gain in the rats co-treated with SMV and CC (Table 3), which (as the two-way ANOVA indicated) was only a consequence of action of V alone, was also shown by other authors in chicks fed with a diet supplemented with ammonium metavanadate (NH4VO3) and chromium acetate (Wright 1968; Nielsen et al. 1980), or with NH4VO3 and CrCl3 (Hill and Matrone 1970; Hafez and Kratzer 1976) or with NaVO3 and CrCl3 (Cupo and Donaldson 1987).

The unchanged l-ascorbic acid concentration in RBCs between the control and SMV-only treated rats shown in the present experiment (Table 6) was also observed in our previous studies after ammonium metavanadate (AMV) intoxication (Zaporowska et al. 1993, 1997). The plasma l-ascorbic acid content did not change either; however, in our earlier studies, its significant decrease in the plasma following vanadate treatment in drinking water was found (Zaporowska et al. 1997; Ścibior and Zaporowska 1998; Ścibior 1999). This is not surprising, because previously we used a different V compound and the period of its administration was shorter (4 weeks). The dose of V ingested was also higher (10–11 mg V/kg b.w./24 h) than that consumed by the rats in this study.

l-ascorbic acid, a potent antioxidant (Lukaski 2004), is an endogenous compound for rats, which can synthesize it in the liver (Meister 1992). By causing a decrease in its concentration, which could be due to its inhibited biosynthesis, enhanced catabolism and/or increased utilization under some experimental conditions, as it was demonstrated in our previous experiment and in the studies of other researchers (Chakraborty et al. 1977), V influences all the metabolic processes, in which this non-enzymatic antioxidant performs regulatory functions. This is very important because l-ascorbic acid in high concentrations acts as an antioxidant (Niki 1991; Halliwell 1996), but in low concentrations acts as a pro-oxidant (Song et al. 1999). V and l-ascorbic acid have been reported to act synergistically in some conditions, too (Sakurai et al. 2004). l-ascorbic acid has also been demonstrated to form complexes with many metals (Kleszczewska 2001), and it has often been considered one of the possible natural reducing agents of vanadate to vanadyl (VO2+), which is then complexed with the acid (Baran 1998). Attention needs to be paid to the fact that during vanadate reduction mechanism in the presence of ascorbic acid generation of free radicals with strong oxidation potential has been postulated (Ding et al. 1994) and that some complexes of VO2+ with this acid may cause DNA nicks, LPO, cytotoxicity as well as depletion of the level of GSH and changed activity of GSH-Px, GR and GST. Cr3+ complexes with ascorbic acid may also cause DNA nicks, cytotoxicity and produce an increase in LPO, but they do not show any effect on the GSH level and on the activity of the above-mentioned enzymes (Sreedhara et al. 1997).

The administration of CC alone (Group III) or in combination with SMV (Group IV) resulted in a significant increase in the GSH concentration in RBCs, which was about 59% and 53.6% higher, respectively, than in Group I (Control), whereas in Group II (SMV-only intoxicated rats) it was elevated by 28% (Table 8). GSH is known to play an important role in detoxification processes and in protecting cells against oxidative stress (Cnubben et al. 2001). Its concentration in RBCs may reflect GSH status in other tissues. The increased GSH concentration might be, at least in part, a consequence of elevated GR activity (by 19% in Group II, by 21% in Group III, and by 46% in Group IV); on the other hand, it might be associated with an increase in the erythrocyte gamma-glutamylcysteine synthetase (γ-GCS) activity, an enzyme which together with glutathione synthetase participates in the synthesis of GSH de novo from three amino acids such as glycine, cysteine, and glutamic acid (Wang and Ballatori 1998). Furthermore, the significantly decreased GST activity in Groups II, III, and IV (by 30, 56 and 40%, respectively) might also be, at least partially, responsible for the enhancement of GSH content, which in this case might indicate its decreased use. We cannot exclude a reduction in its export, either. Thus, the above-mentioned effects could take place together, and although the mechanism responsible for the observed increase in the erythrocyte GSH level is not fully clear at present and further studies are necessary to confirm our assumptions, we may claim that the increase probably resulted from an additive and/or an adaptive response to SMV and/or CC exposure. It is important to stress that in our previous studies, in which we observed slightly decreased (Zaporowska et al. 1993) or significantly lowered (Ścibior 1999) GSH concentration after vanadate treatment in rats, different experimental conditions were used and this antioxidant was assessed in whole blood, not in RBCs.

The unchanged level of MDA, which is considered to be one of decomposing product of LOOH (Niki 2009), observed in RBCs after SMV treatment (Group II, Table 6) is at odds with our previous study, in which a significant increase in its content was demonstrated (Ścibior 1999). Like in the case of the plasma l-ascorbic acid concentration, these discrepancies might be due to large differences between V doses consumed, V compounds used, and the time of their administration.

Vanadate is well known to be transported into RBCs through anion channels and reduced to vanadyl (Baran 1998) which is potentially very damaging because it can autoxidize a thus produced superoxide anion radical and vanadate, which may subsequently react with this radical to yield reactive intermediates responsible for initiation of LPO (Byczkowski and Kulkarni 1998). Moreover, vanadyl may react with H2O2 and produce reactive HO• in a Fenton-type reaction (Byczkowski and Kulkarni 1998). However, the results of our study showed that V (as SMV), at the dose of 9 mg V/kg b.w./24 h, ingested by rats in drinking water during 12 weeks, did not have a pro-oxidant effect on RBCs, which was reflected in the unchanged l-ascorbic acid concentration and MDA level as well as Cu,Zn-SOD, CAT, cGSH-Px, and GR activities in these cells (Tables 6, 7, and 8). The unaltered SOD activity in RBCs after treating the rats vanadate in drinking water was also demonstrated in our previous study (Ścibior 1999) and in the studies of Sekar et al. (1990), Oster et al. (1993) and Russanov et al. (1994); whereas the unchanged erythrocyte CAT activity after vanadate intoxication was shown by Sekar et al. (1990), Zaporowska et al. (1993) and by Russanov et al. (1994). The unaltered erythrocyte cGSH-Px activity found in the present experiment was also demonstrated in the studies of Russanov et al. (1994) and in our earlier study (Ścibior 1999), but contradictory results showing increased activity (Tas et al. 2006) have been reported, too. These conflicting results might be due to the different experimental conditions used by the authors (Tas et al. 2006). The lack of significant changes in the plasma TAS value after SMV treatment (Table 6) remains consistent with the results from our previous experiment (Ścibior 2005).

The treatment of rats with CC alone (Group III) did not change the level of MDA (Table 6) and the activities of Cu,Zn-SOD, CAT and cGSH-Px in RBCs (Tables 7 and 8), which allows us to conclude that CrCl3, (0.47 mg Cr/kg b.w./24 h) did not act as a pro-oxidant at the dose ingested. However, its administration separately (Group III) or together with SMV (Group IV) caused a significant decrease in GST activity (by 56% and 40%, respectively, Table 8). Since in the available literature we did not find any information about the decrease in GST activity in RBCs of rats after V+5 and/or Cr+3 administration in drinking water, it is difficult to discuss our results. Our in vivo studies are the first, to the best of our knowledge, which show this effect. However, the decreased GST activity in blood was described in lead-exposed workers by Hunaiti et al. (1995), who postulated direct or indirect inhibition of GST activity and suggested that this reduction might be caused by depletion of GSH or binding of lead to sulfhydryl (SH) groups. A decrease in GST activity was also demonstrated by Hunaiti and Sound (2000) during incubation of human whole blood containing some lead compounds. Reddy et al. (1981) also observed inhibition of GST by lead and by other metals, too, such as cadmium and mercury in in vitro studies on calf liver homogenates, and the authors suggested that these metals inhibit GST activity by direct binding with an essential group(s) of the protein which promotes the catalytic function.

It is difficult to explain why in all the treated Groups of rats the GST activity, an enzyme which catalyzes the conjugation of GSH to a wide variety of endogenous toxic electrophilic compounds (Townsend et al. 2003), significantly decreased in RBCs, instead of increasing, when, paradoxically, GSH concentration in these Groups of animals increased. Cr+3 is well known to have affinity for the histidyl, methionyl, and cysteinyl groups of proteins as well as for such sulfur-type ligands as GSH (El-Shahawi 1995; Baran 1998; O’Brien et al. 2003). Thus, Cr+3 may bind tightly to many types of proteins and GSH to form stable complexes, some of which may damage the structure and function of proteins (Yamamoto 1995). The reaction of V+5 with SH-containing molecules is also well known to result in the reduction of vanadium and oxidation of thiol-containing molecules (Ballistreri et al. 2000; Çakir and Biçer 2004). Therefore, we may suppose that inhibition of GST activity in our experimental model might result from a direct interaction of V and/or Cr with the SH group of cysteine, which led to its modification and consequently diminished its activity. On the other hand, we did not exclude the possibility that V and/or Cr might produce some other factors leading to site-specific inactivation of GST. Therefore, our speculations need to be elucidated in further investigations to find the correct mechanism by which both elements administered separately or in combination decrease the erythrocyte GST activity, because the present results cannot explain this issue at this stage of studies. Recently, vanadium-binding GST has been identified from the vanadium-rich ascidian (Ascidia sydneiensis samea; Yoshinaga et al. 2006), which seems to have the ability to bind vanadium ions and seems to be able to act as a dual function protein as vanadium-specific metallochaperone and as a GST enzyme; this may suggest a new role of this enzyme in vanadium homeostasis and detoxification (Yoshinaga et al. 2007). However, as the authors indicated, further studies are needed to confirm this hypothesis.

To summarize, the results of this study showed that V (as NaVO3, SMV) and Cr (as CrCl3, CC) administered to rats separately (at the doses ingested, 9 mg V and 0.47 mg Cr/kg b.w./24 h) or in combination (9.8 mg V and 0.39 mg Cr/kg b.w./24 h) in drinking water for 12 weeks did not demonstrate their pro-oxidant potential on RBCs, which was confirmed by the unaltered MDA level and l-ascorbic acid concentration as well as Cu,Zn-SOD, CAT, cGSH-Px, and GR activities in these cells; however, a significantly decreased diet and fluid intake as well as body weight gain and FER were observed in the rats after the treatment with SMV alone or together with CC. The present experiment also indicated that both the elements applied separately or in conjunction elevated the GSH concentration, whose rise could probably be an adaptive response to exposure to the two metals, and decreased GST activity in RBCs. An important finding in the study is that interactions between V and Cr were involved in the changes in GST activity at co-treatment with both elements at the doses mentioned. However, more studies are needed to interpret correctly the V and Cr interactive effect on the activity of that enzyme and to elucidate the mechanism of their action at the molecular level, which cannot be explained thoroughly at the moment. Finally, we may conclude that in spite of the visible changes in some of the parameters examined, V and Cr at the doses ingested were not able to induce oxidative stress in our experimental conditions.