Introduction

Mining waste accounts for approximately 60% of total industrial waste produced in Mexico (1). Populations living near mining sites are exposed to soil and dust containing metal mixtures as well as to surface water contaminated with mineral debris. In Villa de la Paz, Mexico, mining waste is used as plaster material, and it is therefore sold for finishing houses and buildings.

Several neurotoxic metals are found among mining waste from Villa de la Paz--arsenic, manganese, lead, and cadmium. Although research dealing with the adverse effects of a single exposure to any of these metals is essential to elucidate mechanisms of neurotoxicity, few results from those studies can be extrapolated because exposure is to a mixture, not to the metals alone.

A number of recent studies demonstrate that striatal dopaminergic markers are vulnerable to exposure to several metals. Occupational exposure to manganese results in symptoms resembling Parkinson's disease (2,3). Animal studies have confirmed that striatal dopaminergic neurons are one of the main targets for manganese neurotoxicity (4-6). Arsenic-induced increments (7) or decrements (8) of striatal dopamine have been reported for rodents. Lead also influences dopaminergic systems, as has been shown by animal studies of D1 and D2 receptors (9,10), dopamine turnover rates (11), behavioral tests (12-14), enzymatic assays (15,16), and in vitro and in vivo release rate studies (17). Similarly, cadmium has been reported to interact with the striatal dopaminergic system (18,19).

In this study, we used an animal model to examine possible neurochemical changes resulting from exposure to mining waste consisting of a mixture of metals. If several single components of the mining waste induce alterations of striatal dopaminergic parameters, it is plausible that a central nervous system alteration provoked by a mixture of metals would be reflected in changes in striatal dopamine. We conducted a microdialysis study of dopamine release in adult rats subchronically exposed to the mining waste by oral route. We chose dopamine release measurements as an index of brain alterations because neurotransmitter release rates are tightly regulated. Therefore, if the synapse is not able to compensate for externally induced modifications to keep transmitter release at the required levels, this is a reliable sign of synaptic alterations.

Materials and Methods

Animals. Sixty male Wistar rats bred in-house weighing 300-350 g were assigned to three experimental groups of 20 animals each. Animals were housed in groups of 6 or 7 during 14 days under controlled conditions of light and temperature with 15 g of food per rat and water ad libitum.

Mining waste administration. Given that arsenic is the metal with the highest concentration in the mining waste (Table 1), it was considered the guide pollutant. Considering the concentration of arsenic and its approximate bioavailability in mining wastes [around 15% (20)], pellets of normal lab diet (Lab Diet, St. Louis, MO) mixed with the mining waste were prepared to administer 0.92 g of the mining waste per day, which corresponds approximately to an intake of 5 mg/kg/day arsenic. One group was exposed for 2 weeks to the mining waste, a second one received an equivalent amount (25 mg/kg/day) of arsenic as sodium arsenite during 2 weeks, and the control group received normal lab chow. As a quality control, the concentrations of arsenic, lead, and manganese in food pellets was verified, with the following results: arsenic 496 ± 184 µg/g, lead 1.04 ± 0.1 µg/g, and manganese 4.44 ± 0.2 µg/g [mean ± standard error (SE)]. Arsenic levels in food pellets prepared with sodium arsenite were not significantly different from those of mining-waste pellets.

Mining waste collection and analysis of metals. A sufficient amount of mining waste was obtained from a mining area in Villa de la Paz, San Luis Potosí, Mexico. The concentration of 10 metals was determined using an inductively coupled plasma (ICP) spectrophotometer (Model 38S, Yobin Ivon, Lonjumeau, CEDEX France). Separately, arsenic concentration in the mining waste and food pellets was determined using a Perkin-Elmer atomic absorption spectrophotometer (Model 2380, Norwalk, CT) by the hydride evolution-atomic absorption technique (21-23). Lead and manganese were determined in food pellets and tissue using the graphite furnace absorption technique (24). For all the analyses the mining waste was solubilized with a nitric-perchloric acid mixture.

Surgical and microdialysis procedures. Control and exposed animals were anesthetized with pentobarbital (Anestesal, Pfizer, Mexico, 25 mg/kg, ip), acepromazine (Calmivet, Vetoquinol, Mexico, 0.68 mg/kg, ip), and ketamine (Ketavet, Revetmex, Mexico, 30 mg/kg, ip). Once anesthetized, the animals were placed in a stereotaxic apparatus (Stoelting, Wood Dale, IL), the skull was exposed, and a hole was drilled for placement of a guide cannula over the right striatum [stereotaxic coordinates, anteroposterior: -0.3 mm; lateral: 3 mm; ventral: 4.0, with reference to bregma, according to the atlas of Paxinos and Watson (25)]. The cannula was fixed to the skull with anchor screws and acrylic cement. After the surgery, rats were individually housed with free access to water and food during a 48-hr recovery period.

For the microdialysis experiments, a probe of concentric design (CMA/10; Carnegie Medicine AB, Stockholm, Sweden), outer diameter 0.5 mm, and 3-mm dialyzing membrane was inserted into the guide cannula. The dialysis probe was continuously perfused at a flow rate of 2 µl/min through a liquid swivel from a microinfusion pump (74900 series; Cole Parmer, Niles, IL) with a solution containing 147 mM Na Cl, 4.0 mM K Cl, and 1.2 mM CaCl₂, pH 6.0-6.5. Sample collection was performed every 20 min and started 1 hr after the beginning of the perfusion. After three baseline samples, a solution with high potassium content (91 mM NaCl, 60 mM KCl, 1.2 mM CaCl₂, pH 6.0-6.5) was infused during 1 hr (three samples), and standard Ringer solution was restored for the last four samples. Monoamine content was immediately determined in the dialysates or stored at -20°C until quantification. For verification of the probe's efficiency, in vitro recoveries were performed before and after each dialysis experiment by placing the probe in a solution containing 100 pg/µl of each of the monoamines and obtaining three consecutive samples of 20 min each (flow rate 2 µl/min). Recovery rates for the monoamines were as follows: dihydroxyphenylacetic acid (DOPAC) 29.8 ± 2.1, homovanillic acid (HVA) 26.1 ± 1.3, and dopamine 22.32 ± 1.2 (mean ± SE; n = 56)]. Data were not corrected for recovery, and probes were discarded at recovery ratios lower than 20%.

Analysis of dopamine and its metabolites in dialysates. Dopamine and metabolites were quantified in the dialysates by HPLC with electrochemical detection as described in Mejía et al. (7). A Perkin Elmer (series 200) pump was used in conjunction with an electrochemical detector (Bioanalitical system LC-4C). A chromatographic column from Alltech Associates Inc. (Deerfield, IL) packed with adsorbosphere (3-mm particle size, 100 4.8 mm) was used. The isocratic mobile phase was a phosphate buffer (pH 3.2) containing 0.2 mM sodium octyl sulfate, 0.1 mM EDTA, and 15% v/v methanol (Mallinckrodt, Mexico), filtered (0.45 mm pores) and degassed before use. The flow was set at 1.1 ml/min and the determinations were made at room temperature. The electrochemical detector was used at a sensitivity of 1 nA, full scale. The monoamines (DOPAC, dopamine, and HVA) were oxidized with a glassy carbon electrode at a potential of 850 mV relative to the Ag/AgCl reference electrode. The peaks generated by the compounds were analyzed with the software Turbochrom 4 (Perkin Elmer, San Jose, CA). External standards (Sigma, St. Louis, MO) were used to construct a calibration curve for each of the monoamines. Results are expressed in picograms per microliter. At the end of each experiment rats were sacrificed with an overdose of sodium pentobarbital and were transcardially perfused with 10% buffered formaldehyde. Slices of 10 µm were stained with hematoxylin-eosin to determine the exact location of the dialysis probe. Data from animals with the probe out of the striatum or showing extensive tissue damage were discarded.

Statistics. Data analysis was performed by means of the SPSS/PC program (SPSS Inc., Chicago, IL). After exploratory analysis of normal distribution of the data, we used one-way analysis of variance to detect treatment effects on body weight, brain arsenic concentration, and basal and stimulated release of dopamine, DOPAC, and HVA. Groups of data showing a significant effect of the treatments were further analyzed by means of a multiple comparison procedure (Tukey's HSD test). Differences among groups were considered significant at p-values <0.05.

Results

Analysis of the mining waste by ICP mass spectrophotometry showed that its main component is arsenic, followed by manganese and other metals as shown in Table 1. Components of the mining waste readily enter the brain. Figure 1 shows the levels of arsenic and manganese in whole brain of rats exposed to the mining waste during 2 weeks, compared to control and arsenite groups. Arsenic is present in the brain of control animals (38 ± 4.2, mean ± SE), while manganese content builds up after 2 weeks of exposure (147 ± 14.6 ng/g). Lead concentration in the brain was below our detection limit (3 ppb), even after the end of the treatment. The whole-brain arsenic concentration was almost eight times higher in the arsenite group (2,092 ± 186.11 ng/g) than in the mining waste group (269 ± 30.5 ng/g).

Figure 1. Content of arsenic and manganese in rat brains. Rats were treated with normal lab diet (control group), sodium arsenite (25 mg/kg of arsenic), and mining waste (0.92 g/day). Each bar represents the mean ± standard error of 6-10 rats.

Neither the arsenite nor the mining waste group showed significant changes in body weight compared to controls (control group 402.8 ± 8.9 g versus arsenite group 385 ± 19.3 g and mining waste group 383.7 ± 12.4 g; mean ± SE; n = 7-12 animals). During the microdialysis experiments all animals displayed similar behavior. Because experiments were performed during the day, animals were quietly awake, drowsy, or asleep.

Figure 2 shows the temporal courses of dopamine, DOPAC, and HVA release in basal conditions (samples 1-3) under depolarization by perfusion with 60 mM K⁺ Ringer solution (samples 4-7) and the recovery phase after restoring 4 mM K⁺ (samples 8-10). The mean basal and stimulated release values of dopamine, DOPAC, and HVA for all three groups are shown in Figure 3. Analysis of variance indicated that there was no significant difference in basal dopamine release rates among groups. In contrast, stimulated dopamine release rates showed an effect of the treatment. Control animals displayed a sixfold increase of dopamine release under 60 mM K⁺ perfusion. Although the arsenite group responded to high potassium with a similar increase in dopamine release, the mining waste group showed only a weak increase in dopamine release in response to a depolarizing stimulus (57% increase).

Figure 2. Temporal course of striatal release of (A) dopamine, (B) dihyroxyphenylacetic acid (DOPAC), and (C) homovanillic acid (HVA). Dialysates were collected every 20 min; K+-infusion (60 mM) lasted from samples 4 to 7; normal Ringer solution was restored after sample number 7. Values shown represent means ± standard errors for 6-8 rats.

Figure 3. Effect of arsenite and mining waste in basal and stimulated values of (A) dopamine, (B) dihyroxyphenylacetic acid (DOPAC), and (C) homovanillic acid (HVA). Rats were treated with normal lab diet (control group), sodium arsenite (25 mg/kg of arsenic), and mining waste (0.92 g/day). Each bar represents the mean ± standard error of 6-8 rats per group. Basal release values are the mean of six samples per animal, three prior to depolarization and three after depolarization. Stimulated release represent the mean of four samples during high potassium infusion. *p<0.01 versus control group.

Concerning DOPAC, basal as well as stimulated levels were significantly lower in the mining waste group as compared to controls (basal: mining waste group 66.7 ± 7.5 pg/µl versus control group 113.7 ±14.3 pg/µl; stimulated: mining waste group 57.3 ± 6.7 pg/µl versus control group 114.1 ± 15.2 pg/µl). HVA levels, although lower than in control animals, only reached statistical significance for the mining waste group under depolarization (mining waste group 41.9 ± 5.2 pg/µl versus control group 84.1 ± 13.3 pg/µl).

Discussion

The present study represents the first application of in vivo microdialysis for assessing the effects of a chemical mixture over striatal neurochemical parameters. We found that at least two of the metals contained in the mining waste do accumulate in rat brain after oral subchronic exposure. Under steady-state conditions, exposed animals had dopamine release rates comparable to controls, although DOPAC levels were significantly decreased. When the system was challenged with a long-standing depolarization through high-potassium perfusion, it became clear that exposure to the mining waste provoked an alteration because exposed animals were not able to sustain an increased dopamine release in response to depolarization. Also, DOPAC and HVA levels were significantly lower in exposed animals than in controls during stimulation with high potassium. Although rats treated with arsenite alone revealed a similar tendency to mining waste-exposed animals, they did not show significant differences in any parameter with respect to controls. This indicates that arsenic alone is not responsible for the effects of the mixture and points to an additive effect of the elements present in the mining waste.

This study sought to establish an animal model for predicting neurological effects by exposure to a chemical mixture. Data from Table 1 indicate that individual analysis of the effects of the mining waste components upon neurotransmitter systems would require a long series of experiments. Furthermore, results from single-exposure testing of a toxic substance can be hardly extrapolated to the real effects of that toxicant in presence of other substances. In this case, arsenite and mining- waste groups were exposed to the same dose of total arsenic; however, brain arsenic content was eight times higher in the arsenite group. This is in agreement with previous estimations of bioavailability of arsenic present in the mining wastes (20,27,28). In contrast, even though the manganese concentration in the mining waste and food pellets was about 100 times lower than that of arsenic, its concentrations in the brain were the same order of magnitude as those of arsenic. Without a previous study of bioavailability of manganese in this mineral mixture, we do not know whether manganese was indeed more bioavailable or if its entrance to the brain was facilitated though other mechanisms, as has been reported for another mixture (7). Similar cases might have been encountered if each metal was analyzed separately; therefore, the direct study of the mining-waste effects appears to be a better alternative for our objective.

In view of the high arsenic content from the mining waste (Table 1), we have recently reported the results of a parallel analysis of arsenic in soil and dust from Villa de la Paz, as well as in urine from children in this population (26). We found 2,904 ± 2,261 ppm in soil and 2,045 ± 1,117 ppm in household dust (mean ± SD). Urinary arsenic concentrations in children were as high as 161 µg/g creatinine, with 69% of the children bearing more that 100 µg As/g creatinine (26). These findings warrant concern about the neurotoxic potential of the mining waste because we found neurobehavioral alterations in children with arsenic levels of 62 µg/g creatinine (29). Given these facts, it was important to include arsenic as a secondary control in this animal model to obtain data about brain availability of arsenic from the mining waste as well as its possible role in dopamine release.

Data presented in Figures 2 and 3 demonstrate that our microdialysis setup was reliable. First, we obtained basal and stimulated dopamine, DOPAC, and HVA levels as well as recovery rates similar to those reported in the literature (30,31). Second, depolarization responses of control preparations were also similar to those reported by others using similar procedures (32-35).

In mining waste-exposed animals, the combined action of metals may compromise the synthesis of dopamine, but efficient modifications of re-uptake and catabolism rates could maintain normal basal release rates. It is plausible that, under basal conditions, a decrease of dopamine synthesis in exposed animals could be counterbalanced by modifications of the re-uptake rates. In this way, less dopamine would be available for metabolism through monoamine oxidase, which would result in decreased basal DOPAC levels (Figs. 2 and 3). Under a situation of higher demand, like prolonged exposure of the striatal cells to high potassium, compensatory mechanisms became insufficient, resulting in significantly lower dopamine, DOPAC, and HVA levels in dialysates.

Another possibility is that exposure to the mining waste is not affecting specific enzymes of dopamine synthesis and metabolism, but that nonspecific membrane alterations caused by the mixture could disturb the normal response to a depolarization through increased release rates, leading to the observed decrease of dopamine, DOPAC, and HVA release under stimulation. Membrane alterations may also modify the activity of membrane-associated enzymes like monoamine oxidase, which is responsible for dopamine conversion to DOPAC, resulting in lower DOPAC basal and stimulated levels.

Processes like behavioral activation, motor activity, attention, and learning are all associated with increased release rates of different neurotransmitters in the areas involved in those tasks. For instance, increases of striatal DA release accompanied eating (43%), drinking (22%), and tactile stimulation (15%) (36,37). More important, increases have been observed in acetylcholine (ACh), which is involved in sleep/wake cycles [52% increase during the night (38)]. Also, sensory stimulation has been reported to provoke hippocampal and cortical (ACh) increases ranging from 40% to 220% (39,40). Our observed 600% increases suggest that a 1-hr sustained depolarization with 60 mM K⁺ is not a common physiological situation. Although mining waste-exposed animals showed a marked reduced capability of the dopaminergic terminals around the probe to sustain a prolonged increase of dopamine release, it is still possible that the system can cope with physiological requirements.

In conclusion, we have demonstrated that in vivo microdialysis is suitable for use with an animal model to study neurotoxic effects of a chemical mixture. Our results point to the need for further research that directly characterizes the effects of chemical mixtures to which our population is exposed (41). Also, these results indicate that the mining waste from Villa de la Paz represents a potential health risk for the residents of that area and that measures should be implemented to avoid health problems.