Disruption of the potential across the synaptosomal plasma membrane and mitochondria by neurotoxic agents

SUMMARY The effect of a neurotoxic organometal, methyl mercuric iodide, and an aromatic solvent, toluene, upon the transmembrane potential (l/f), across both the limiting membrane of isolated nerve terminals and their mitochondria, has been studied. Exposure of nerve endings to either of these toxicants in vitro resulted in a dose-dependent diminution of 'I' that was especially pronounced in the case of mitochondria. This was not prevented by a concurrent exposure to an antioxidant (ll-tocopherol), or an iron chelator (deferoxamine), or ganglioside GM 1• No significant changes were detected in synaptosomal potentials derived from cortices of rats exposed to methyl mercury or toluene at levels known to increase the rate of formation of reactive oxygen species within this region. The especial vulnerability of mitochondrial 'I' to these agents may be due to the disruption of oxidative phosphorylation and may be related to the increase in intrasynaptosomal free ionic calcium that both of these chemicals can induce.


INTRODUCTION
Biological membranes present a distinct target to chemicals that are injurious to the nervous system. This may in part be due to the critical geometry of the hydrophilic and lipophilic regions within such membranes. The maintenance of this bilayered structure is essential for membrane function. Both mitochondria and the plasma membrane are characterized by a transmembrane potential generated by ion pumps. In the case of the plasma membrane, this electrical discontinuity represents an energy reserve which can be drawn upon to allow the passage of nerve impulses. Synaptosomal membrane potential may also be directly involved in effecting neurotransmitter release [1]. The mitochondrial membrane potential is the result of uniporter-effected translocation of protons created at the inner surface of the mitochondrial membrane by operation of the respiratory chain.
A wide range of pharmacological and neurotoxic agents can cause hyperexcitability. One of the means of effecting this is by partial depolarization of the axonal limiting membrane. Since maintenance of this gradient is an energy-requiring process, any means of depletion of energy reserves (largely represented by the ATP/ADP ratio) may diminish the magnitude of the membrane potential. Xenobiotic agents having an adverse effect on mitochondrial energy generation may also impair the transmitochondrial membrane potential generated by pumping of protons to the cytosol. This proton gradient is critical for the formation of ATP from ADP.
We have studied the effects of two neurotoxic agents upon the potential across the synaptosomal limiting membrane and the mitochondrial potential. One of these is an aromatic solvent (toluene), while the other is an organometal (methyl mercuric iodide). Both of these chemicals have been found to induce oxidative stress both within brain regions of exposed animals and within isolated synaptosomes [2,3]. Methyl mercury has previously been found to increase lipid peroxidation levels [4] and vitamin E can protect against the toxicity of alkyl mercurials [5,6]. The ability of various compounds with antioxidant potential, to mitigate induced changes, has also been investigated in the present work.
Reactive oxygen species have previously been reported to modulate plasma membrane potential [7]. The relevance of membrane depolarization to neurotoxic events is further illustrated by evidence that persistent depolarization may underlie longterm synaptic failure and may be an early critical step toward neuronal degeneration [8].

Animals
Adult male CR 1 CD rats (250-300 g) were obtained from Charles River Breeding Laboratories (Wilmington, MA), and were maintained in the animal facility in clear polypropylene cages with water and food provided ad libitum.

Preparation of synaptosomes
Adult male CR 1 CD rats (Charles River Breeding Laboratories, Inc., Wilmington, MA) 2-3 months old, weighing 175-280 g were used. Rats were decapitated, the brains excised and the whole brain except the cerebellum and pons-medulla dissected out. Synaptosomes were made by the modification of Dodd et al. [9] of the method of Gray and Whittaker [l O]. After homogenization in 10 volumes of cold 0.32 M sucrose, the tissue homogenate was centrifuged (1900 x g, 10 min) and the supernatant Sl laid over 10 ml 1.2 M sucrose. After high-speed centrifugation (250 000 x g, 25 min), the layer at the interface was collected, diluted 2.5-fold with 0.32 M sucrose and 15 laid over 8 ml 0.8 M sucrose. After further centrifugation at high speed, the synaptosomal pellet was suspended in HEPES buffer (pH 7.4) to give a tissue concentration of ~.l~ g-equivalent/ml (about 1.~ mwml protein). For ~tudie~ of oxygen radical formation, crude synaptosomes were prepared from the Sl fraction by centrifugation at 31 500 x g for 10 min to yield the P2 fraction. This was suspended in HEP ES buffer to a concentration of 0.0185 g-equivalent/ml.

Determination of electrical potential
Synaptosomes were incubated 5 min at 37°C in the presence of 1 µM rhodamine 6G [11]. This concentration of rhodamine has a negligible effect on synaptosomal metabolism [12]. The final synaptosomal concentration was 0.015 g-equivalent tissue/ ml containing 0.27-0.32 mg protein/ml, in HEPES buffer.
In order to test the effects of agents upon the potential (I/I) of mitochondria within synaptosomes, KCl was added to a final concentration of 100 mM in order to almost totally depolarize the plasma membrane while not modulating mitochondrial l/f [13,14). The fluorescence of rhodamine was then determined at 550 nm, using an excitation wavelength of 520 nm. When toxic agents were to be studied in vitro, they were added after this determination, and incubation continued for a further 10 min. Any ensuing loss of mitochondrial l/f was evaluated by an increased fluorescence of rhodamine. In order to completely depolarize mitochondria, carbonyl cyanide pfluoromethoxyphenylhydrazone (FCCP) was then added to a final concentration of 1 µM [15), after which fluorescence was again determined.
Plasma membrane l/f was estimated in a parallel manner. Rhodamine fluorescence was measured 5 min after the depolarization of mitochondria by 1 µM FCCP [16). Plasma membrane l/f has been shown to remain unaltered for some time after such a suppression of mitochondrial l/f [16). Any toxic agent under study was then added and, after a further 10 min incubation, fluorescence determined once more. Finally, fluorescence was determined after depolarization of the plasma membrane with 100 mM KCI. This permitted the extent of reduction of the transmembrane l/f by the toxicant to be calculated.
Values of l/f of fully polarized membranes were derived from the equation:  calculated potentials is minimal [17,19]. This may be due to the deleterious effect of low potassium in the preparative medium on free mitochondria.

Protein determination
Protein content of synaptosomes was assayed by the method of Bradford [22] using bovine serum albumin as a reference.

Statistical analysis
Differences between groups were assessed by Fisher's Least Significant Difference Test after one-way analysis of variance. The acceptable level of significance was P<0.05 using a two-tailed distribution.

RESULTS
The effect of various concentrations of methyl mercury upon the potentials of synaptosomes exposed in vitro was studied. The membrane potentials of both mitochondria and plasma membranes were depressed by methyl mercury in a dose-related manner (Fig. 1). However, mitochondrial 'I' was considerably more susceptible to inhibition by this neurotoxic agent than was 'I' of the plasma membrane. A variety of chemicals known to be protective against oxidative stress were tested for their ability to protect against changes in mitochondrial 'I' caused by methyl mercury. These compounds were 5 µM a-tocopherol succinate, 1 µM ganglioside GM 1 , and 100 µM deferoxamine. In no instance was any mitigation effected by such chemicals (data not shown).
Methyl mercury was next administered to rats by a single intraperitoneal injection at a dose of 10 mg/kg body wt., synaptosomes being prepared after 2 days of exposure. These conditions are known to significantly elevate the rate of generation of reactive oxygen species within the cerebral cortex of the rat [2]. Following such an in-vivo exposure, however, there was no significant effect on either plasma membrane or mitochondrial potential (Table I).
In a parallel series of experiments, the effect of toluene upon synaptosomal potentials was studied. This aromatic solvent also depolarized both mitochondrial and plasma membranes, and this effect was dose-related (Fig. 2). In view of the insolubility of toluene in water and uncertainty of toluene content within the tissue lipid compartment, toluene additions are given in µ112 ml incubation medium, rather than in concentrations. Mitochondrial 1f1 was again especially sensitive to inhibition by this solvent exposure in an isolated system (Fig. 2). Once more, the presence of each of several potentially protective agents (cx-tocopherol succinate, ganglioside GM1' deferoxamine), was ineffective in preventing the depolarization of mitochondria by toluene (data not shown). Isolated synaptosomes were also exposed to a related organic Values are derived from 5 animals/group and are expressed as -mV. Rats were sacrificed 2 days after methyl mercury (10 mg/kg) or I h after toluene (I g/kg) treatment. solvent, benzene. This chemical has not been found to induce the formation of reactive oxygen species in synaptosomes [3]. Benzene also inhibited mitochondrial 1f1 but to a lesser extent than toluene (Fig. 3). One hour after treatment of rats with toluene (I g/kg by i.p. injection using 50% corn oil, v/v, as a vehicle), animals were sacrificed and synaptosomes prepared. The mitochondrial 1f1 was depressed, but this did not reach significance (Table I). DISCUSSION Assay of synaptosomal 1f1 can only be carried out on fresh tissue. This is in contrast to several other procedures for evaluating synaptosomal status which can be performed on carefully frozen and subsequently re-thawed synaptosomes using the procedure of Hardy et al. [23]. These latter include assay oflevels of cytosolic calcium and pH, rates of free radical generation, and high-affinity uptake of neurotransmitters. This suggests that depolarization may be an especially sensitive index of synaptosomal damage. However, adverse effects upon nerve endings incurred in vivo may be harder to detect. This may be related to the powerful ability of isolated nerve ending to maintain homeostasis. Synaptosomes isolated from toluene-treated rats, and subsequently incubated in toxicant-free HEPES buffer, have normal levels of [Ca 2 +]i although the permeability of the limiting membrane is elevated [Von Euler et al., in preparation]. This differs from results obtained using synaptosomes exposed to toluene in vitro, where these parameters are both elevated [24]. Another possible reason for the apparent resistance of nerve endings to detectable changes following their exposure to neurotoxic agents in vivo is their much higher density per unit of toxicant encountered in the living animal than is the case with isolated tissue fractions. Thus, isolated synaptosomes exposed to levels of an agent, corresponding to that concentration found in the CNS of an exposed species, are present in the incuba-tion medium at a protein concentration that is several orders of magnitude below that present in the intact brain. A critical relation between synaptosomal vulnerability to the tissue/ toxicant ratio, rather than to the absolute concentration of the toxic agent, has been previously described [25].
The concurrent presence of inhibitors of free radical formation had no effect on mitochondrial depolarization caused by either methyl mercury or toluene. Thus the susceptibility of this 'I' to toluene or methyl mercury cannot be due to the known ability of these neurotoxicants to stimulate the generation of oxygen free radicals. The overall neurotoxicity of the two agents under study can perhaps be ascribed to a combination of both oxidative and non-oxidative events.
The especial sensitivity of the mitochondrial membrane to depolarization by the agents under study here may be a reflection of the deleterious effects of these agents upon the electron transport processes underlying oxidative phosphorylation. One would expect the synaptosomal membrane potential to be less sensitive to disruptions in energy metabolism than the mitochondrial potential, because resting potentials across nerve terminals are largely dependent on the K+ gradient which, in tum, is largely determined by the Donnan equilibrium, a non-energy-consuming process. Exposure to methyl mercury has been found to effect reduction of ATP-generating capacity [26,27]. The elevation of cytosolic calcium, [Ca 2 +];, reported to be effected by both methyl mercury and toluene [19,24,27], may in part be due to the collapse of the mitochondrial potential, leading to release of calcium sequestered within mitochondrial binding sites.
The results obtained with methyl mercury parallel those reported in a previous study on synaptosomes in that the mitochondrial was much more susceptible to depolarization than was the plasma membrane [27]. However, these latter authors found inhibition of 'I' to require considerably higher concentrations of methyl mercury. A similar reduction of mitochondrial 'If and elevation of [Ca 2 +l has also been reported for rat hepatocytes exposed to inorganic mercuric chloride [28]. In many of these studies, the omission of calcium from the external medium greatly diminishes the effect of toxicants on cytosolic calcium levels. Thus it is not clear whether collapse of the mitochondrial potential is a cause or a consequence of elevations of ionic intracellular calcium. Methyl mercury has been reported to inhibit the mitochondrial Ca/H exchanger [29], and there is a close link between [Ca 2 +l and mitochondrial potential. Calcium uptake by mitochondria is driven by mitochondrial 'If [18], and calcium overload can effect the collapse of the mitochondrial 'I' [20]. The results ofKaupinnen et al. [27] suggest that lower concentrations of methyl mercury cause elevation of free cytosolic calcium by disruption of mitochondrial energy production, while higher levels of this toxicant disrupt the ability of the plasma membrane to exclude extracellular calcium.
The locus of action of a neurotoxicant may be either predominantly at the plasma membrane or at the mitochondrion. Damage at either of these sites has the potential of initiating a sequence of deleterious events which may include elevation of free calcium or rates of free radical generation within cells. Determination of the state of