Increased Free lntrasynaptosomal ca 2 + by Neurotoxic Organometals: Distinctive Mechanisms

Increased Free Intrasynaptosomal CaH by Neurotoxic Organometals: Distinetive Mecha nisms. KOMULAINEN, H., AND BONDY. S. c. (1987). Toxico/. Appl. Pharmacol. 88, 77-86. Effects of several alkylmetals on free intrasynaptosomal Ca 2 + concentration, [Ca2+],, were stud ied in vitro using the fluorescent Ca 2 + indicator fura-2. Neurotoxic alkylmetals methylmercury (Met-Hg). triethyllead (TEL), triethyltin (TET), and trimethyltin (TMT) (at 2.5-30 µM) in creased [Ca 2 +], to different degrees. Met-Hg was the most potent, elevating [Ca 2 +], 100-800 nM, dose dependently and significantly more than high K+ (150 nM) or veratridine (350 nM). The effect of Met-Hg could not be inhibited with a Cai+ channel blocker, verapamil, nor with a Na+ channel blocker, tetrodotoxin. Inhibition of the mitochondrial Ca 2 + uptake in situ with rotenone + oligomycin decreased the potency of Met-Hg to elevate [Ca 2 +], but did not change the resting [Ca 2 +],. Met-Hg also slightly decreased synaptosomal ATP. TEL and TET elevated [Ca 1 +]; by 100-200 nM. The effect of TEL. but not that ofTET. could be blocked with veraparnil (36%) and veratridine (67%). TEL was less efficient in the presence of ouabain. Neither TEL nor TET had significant mitochondrial effects in situ contributing to [Ca 2 +],. TMT increased [Ca 2 +], less than TET while dimethyltin and methyl tin were inactive. These results indicate that neurotoxic derivatives of alkylmetals studied increase [Caz+];. This occurs mainly either by nonspecific increase (Met-Hg, TET) ofCa 2 + leakage through the plasma membrane and/or specific interfer ence with the mechanisms regulating Ca 2 + fluxes through the plasma membrane (TEL). '9 1987

The plasma membrane is always traversed before intracellular targets are reached and it is possible that some intracellular effects of alkylmetals are indirect due to altered characteristics of the plasma membrane. The plasma membrane maintains vital ionic gradients and is the main barrier to separate 1 mM extracellular Ca 2 + from 0.1-0.3 µM intracellular Ca 2 + (>3000-fold concentration gradient;Tsien et al., 1982;Hesketh et al., 1983;Berthon et al., 1984;Sistare et al., 1985, Komulainen andBondy, 1987). Ca 2 + is an important messenger in the cell (Rasmussen et al., 1984) and by increasing free intracellular Ca 2 +, organometals might initiate chains of events exciting and finally killing the cell. In nerve endings, Ca 2 + triggers release of neurotransmitters and there is already evidence that Met-Hg stimulates release in vitro (Komulainen and Tuomisto, 1981;Atchison et al., 1984). We are reporting here effects of several alkylmetals on free intrasynaptosomal Ca 2 + ([Ca 2 +]i) in vitro. Synaptosomes were used because they are readily available metabolically intact fragments of isolated neural tissue. Various pharmacological, neuroacti ve agents were used for comparison and to assess the specificity of the effect of alkylmetals and to elucidate their sites of action.
Assay of [Ca 2 +],. Free intrasynaptosomal Ca 2 ' was measured using the fluorescent Ca 2 + indicator dye fura-2. Details of the method and characteristics of the dyeloaded synaptosomes are in Komulainen and Bondy ( 1987). Briefly, in order to load synaptosomes with fura-2 ( 150-200 µM concentration). I-ml aliquots of synaptosome suspension were incubated with 5 µM acetoxymethyl ester of fura-2 (fura-2/ AM) for 20 min at 37'C. Then 9.0 ml Hepes buffer (A) at 37'C was added and incubation continued for the next 25 min (total incubation time 45 min). Synaptosomes were immediately spun down (9000g, 5 min, Sorvall RC-5B) and resuspended in 5 ml ice-cold Hepes buffer (A). An identical tissue sample incubated with 5 µI DMSO was prepared in parallel to assay autofluorescence of synaptosomes.
The fluorescence was measured with Aminco SPF-500 spectrofluorometer (Silver Spring, MA) equipped with a thermostated sample holder (37'C) and a magnetic stirrer (lnstech Laboratories, Inc., Horsham, PA). The excitation wavelengths were 340 and 380 nm (bandpass I nm) and the emission wavelength 510 nm (bandpass 8 nm). For calculation of [Ca 2 +]i, R, the ratio of the fluorescence at 340 nm/380 nm, was determined for each synaptosomal sample as follows: 0.5 ml offura-2-loaded synaptosome suspension (containing 0.14-0.16 mg protein) was rapidly spun down with a microcentrifuge ( 13.000g, 30 sec) and synaptosomes resuspended in 2 ml Hepes buffer (at 37'C) omitting NaH2P0 4 and NaHCO, (Hepes buffer B). These salts were excluded because otherwise Ca 2 + precipitated out at elevated pH during calibration of the fura-2 fluorescence. Synaptosomes were allowed to equilibrate for 10 min in a cuvette before addition of any chemical under assay. Autofluorescence of synaptosomes was separately measured and subtracted from the fluorescence of the sample. Extrasynaptosomal fura-2 was quenched by Mn 2 + (20-40 µM). For calculations of [Ca 2 +],. two other ratios were determined for each fura-2-loaded homogenate (Komulainen and Bondy, 1987): Rmin• the ratio of the fluorescence at zero free Ca 2 +. and Rmax. the ratio of the fluorescence at full Ca 2 ' saturation. This calibration of fura-2 fluorescence was performed in the presence of JO µM diethylenetriamine pentaacetic acid (DTPA), a heavy metal chelator. to ensure chelation of all added Mn 2 + in the sample. The final [Ca 2 +j, was calculated as described by Grynkiewicz et al. ( 1985): Sf 2 is fluorescence of fura-2 at zero Ca 2 + and Sb 2 at Ca 2 ' saturation at the excitation wavelength 380 nm. Kd, the dissociation constant of the fura-2-Ca2+ complex, was taken to be 224 nM (Grynkiewicz et al.. 1985).
Assay ofsynaptosomal ATP. Synaptosomal ATP was measured as described by Cheung and Verity ( 1981) after slight modifications of the method. Synaptosomes were made and resuspended in Hepes buffer (A) at the protein concentration 1.4-1.6 mg/ml. Aliquots (0.5 ml) of this suspension were incubated in 1.5 ml Hepes buffer (B) with indicated concentrations of organometals or other agents for 10 min at 37•c. Immediately after incubation. a 0.5-ml sample of the suspension was mixed with 0.5 ml 0.6 Mice-cold perchloric acid. Then samples were rapidly microcentrifuged and 62.5 µl 2 M K2 C03 was added. The precipitate formed was spun down and a 100-µ.l aliquot of the supernatant diluted with 0.9 ml Tris buffer. 40 mM, pH 7.4. ATP was assayed with luciferase-luciferin mixture (Sigma No. L 1761, 40 mg/ml) which was aged overnight in a refrigerator. For each assay. 0.5 ml of Tris buffer and 30-40 µ.l ofluciferase-luciferin mixture were pipetted into a glass scintillation vial; 30-60 µI of the sample was added and the mixture vortexed. The peak luminescence was immediately recorded with a liquid scintillation counter (Beckmann LS-3100) operated manually. For quantitation of ATP, standards containing known amounts of ATP (disodium salt) were measured at the same concentration range as in samples. The linal ATP concentration was read graphically on a loglog paper.
45 Ca J+ accumulation. In experiments intended to study effects of organometals on Caz+ accumulation in synaptosomes, synaptosomes (on an average 0.16 mg/ml protein. a final concentration) were incubated in Hepes buffer(A) containing0.5 µ.Ci/ml 45 Ca 2 + (55.9 Ci/mo!) for 10 min at 37°C Synaptosomes were filtered on glass-liber filters (Type A/E, Gelman Sciences, Inc.), washed thrice with 5 ml cold Hepes buffer (A) omitting CaC1 2 and containing I mM EGT A, and the radioactivity remaining in the tissue was measured (Kornulainen and Bondy, 1987).
Protein assay. The protein content of synaptosomal suspensions was determined according to Lowry et al.
( i 951) using bovine albumin as a reference.

Statistics. Statistical analysis of the results was done
with Fisher's least significant difference test after one-way analysis of variance. p <0.05, with two-tailed I-distribution, was considered to be significant.

Effect of alkyl metals on the resting [ ca2+];.
Of the organometals studied only those which are known to be potentially neurotoxic elevated (Ca 2 +]i in vitro. Met-Hg increased [Ca 2 +]i from its basal level of 350 nM in a dose-dependent manner (Fig. I). Its initial effect was rapid and the Ca 2 + level showed a continuing increase with time (Fig. 2).
TEL and TET were much less potent at the same concentration range, elevating [Ca 2 +]i by 50-170 nM (Fig. l ). This effect was saturable and the Ca 2 + level changed initially more slowly than with Met-Hg (data not shown).
When Ca 2 + was omitted from the incubation medium (nominally Ca2+-free buffer), Time course of the increase of free intrasynaptosomal Ca 2 + caused by methylmercury (10 µM) . Synaptosomes were incubated at 37°C with methylmercury either for I min (n = 7) or for 10 min (n = 4). The increase (mean ± SE) is that over the resting [Ca 2 +);. **Significantly different from the value at I min (p < 0.00 I).
the [Ca2+]i dropped to 230 ± 15 nM (n = 6). Under such circumstances neither TEL nor TET increased [Ca 2 +]i at all and Met-Hg only marginally (Fig. 3). 45 Ca 2 + accumulation. Met-Hg stimulated accumulation of 45 Ca 2 + into synaptosomes in a dose-dependent manner from the incubation medium containing 1 mM Ca 2 + (Table  I). TEL also enhanced Ca 2 + uptake at the highest concentration studied (Table 1).
Effect of verapamil or tetrodotoxin on the increase of [Ca 2 +];. In order to elucidate Note. Synaptosomes (about 0.16 mg protein/ml) were incubated with alkylmetals in Hepes buffer (A) containing traces of 45 Ca 2 + (total Ca 2 + J mM) at 37°C for l 0 min (see details under Methods). Ca 2 + uptake in control samples was 12.6 ± 1.6 nmol/mg protein/IO min (n = 5). SE of the mean are presented.
"Number of samples in triplicate. hp< 0.05, significantly higher compared to control. c p < 0.0 l, significantly higher compared to control.
whether Ca 2 + fluxed in through the voltagesensitive Ca 2 + channels after synaptosomal exposure to organometals, synaptosomes were preincubated for I 0 min with verapamil (30 µM), a blocker of one type of Ca 2 + channels, before the addition of alkylmetals. Verapamil inhibited 36% of the TEL-induced elevation of[Ca2+]i but could not modify the increased Ca 2 + level effected by Met-Hg or TET (  (6) 112 ±WC (7) 97 ± 7 (7) Note. The mean ± SE resting [Ca 2 +]; was 352 ± 6 nM (n = 26) before any addition of chemicals. Verapamil (in 60 µI H 2 0) was added IO min before addition of 50 mM K +(with which synaptosomes were incubated for 1 min), veratridine, or organometals. Synaptosomes were incubated with veratridine or organometals for I 0 min. " Number of determinations. 'p < 0.00 I, significantly lower compared to corresponding value in the absence ofverapamil.
' p < 0.0 I, significantly lower compared to corresponding value in the absence ofverapamil.
(85 %) the effect of veratridine upon [Ca 2 +]; (Fig. 4) and inhibited 67% of the increase of free Ca 2 + caused by TEL (Fig. 5). In contrast, it could not significantly block the effect of TET or Met-Hg (Fig. 5). Hence, TEL might specifically increase Na+ influx through Na+ channels. Whereas, if there was any increase in intrasynaptosomal Na+ concentration by Met-Hg or TET, it was not through Na+ channels.
Modulation of the effect of ouabain. Another mechanism to achieve an increase in intrasynaptosomal Na+ concentration is an inhibition of the plasma membrane "Na+ pump," Na+ -K +-A TPase. Inhibition of this enzyme with ouabain (I mM) increased gradually [Ca2+];, about 100 nM in IO min (Table  3). Alkylmetals, added after a preincubation of synaptosomes with ouabain, had different effects on [Ca 2 +];. TET was additive to ouabain, Met-Hg even more potent with ouabain, but TEL lost about 50% of its potency in the presence of ouabain (Table 3)   (370 ± 6, n = 25); ouabain was added in 10 µI DMSO 2 min before organometals or H 2 0 (20 µI) and incubation continued for 10 min.
a Number of measurements. b p < 0.01, significantly higher compared to corresponding value without ouabain. c p < 0.00 I, significantly higher compared to corresponding value without ouabain. µM; Rasmussen et al., 1984), and they provide ATP for ion pumps maintaining the ionic gradients across the plasma membrane.

Alkylmetals could increase [Ca 2 +l by inhibit-
ing Ca 2 + uptake in mitochondria and/or decreasing intrasynaptosomal ATP. The combination of rotenone (inhibition of respiration) + oligomycin (inhibition of mitochondrial Mg 2 + -A TPase and subsequently both ATP synthesis and hydrolysis by this enzyme) + iodoacetic acid (inhibition of glycolysis ), both inhibits mitochondrial Ca 2 + uptake and severely depletes synaptosomal ATP (Akerman and Nicholls, 198 la,b, Table 6). These agents in combination caused [Ca 2 +]; to rise linearly with time by 156 ± 16 nM (n = 5) in 10 min (Table 4) and by 280 ± 22 nM (n = 6) in 20 min. In these ATP-depleted synaptosomes, TEL increased [Ca 2 +]i in an additive manner with the increase due to ATP depletion. The effect of TET upon such ATP-depleted preparations was enhanced (p < 0.05) but the rise in [Ca 2 +]i caused by Met-Hg was considerably below ( 42%) that effected in normal synaptosomes (Table 4). Inhibition of mitochondrial Ca 2 + uptake by rotenone + oligomycin did not alter resting [Ca 2 +]i (Table 5). After such treatment of syn-aptosomes with rotenone and oligomycin the response of [Ca 2 +]i to TET or TEL was unaltered [Ca 2 +]i while the effect of Met-Hg was again diminished 42% (Table 5).

DISCUSSION
"Number of determinations. "p < 0.01, significantly lower compared to corresponding value in the absence ofrotenone + oligomycin.
free intrasynaptosomal Ca2+ in vitro. Each alkylmetal appeared to have distinctive mechanisms of action. Although more work is needed to elucidate the details of the mechanisms, some sites of action can already be pinpointed. Regulation of [Ca2+]i is complex with many simultaneous reactions that balance the net free Ca 2 + concentration (Rasmussen et al., 1984). The most important barrier to Ca 2 + appears to be the plasma membrane, across which Ca2+ cycles through the influx (leakage, voltage-dependent Ca 2 + channels) and efflux (Ca 2 + pump, Na+ /Ca 2 + exchange) mechanisms. Inside Ca 2 + is bound to cytoplasmic Ca2+ binding and membrane proteins and is sequestered in the endoplasmic reticulum. The role of mitochondria in the regulation of synaptosomal resting [Ca 2 +]i is at present controversial (Akerman and Nicholls, 1981 b;Nachshen, 1985; present results) but they may take up Ca 2 + at elevated [Ca 2 +]j. Hence, there are many sites in Ca 2 + metabolism as possible targets for alkylmet-als. The main effect of alkylmetals was in the plasma membrane. Perturbation of the mechanisms maintaining the huge Ca 2 + gradient and/or the integrity of the plasma membrane were crucially affected.
Of the alkylmetals studied, Met-Hg was the most potent in increasing [Ca 2 +]i. It affected the plasma membrane nonspecifically, rendering it leaky to Ca 2 +. Because Met-Hg did not cause excess release of fura-2 from synaptosomes, the plasma membrane was not lysed. The initial effect on Ca 2 + level was so immediate that a direct binding to the plasma membrane (SH-groups), and a subsequent opening of extra "gates" to Ca2+ (and possibly also to Na+), is suggested. That extrasynaptosomal Ca 2 + was the main source ofCa 2 + is supported by two observations. Met-Hg could not increase [Ca 2 +]i much in nominally Ca2+ -free buffer, and Met-Hg stimulated accumulation of 45 Ca2+ into synaptosomes from the extrasynaptosomal medium.
However, the effect of Met-Hg differed clearly from the specific depolarization of the plasma membrane such as that induced by high K+ or veratridine (Krueger et al., 1980). During such depolarization, different subclasses of Ca 2 + channels open to Ca 2 + (Nowycky et al.. 1985), and verapamil blocks probably one type of these channels. Consequently, it inhibits elevation of[Ca 2 +]i by 30-50% (present results; Komulainen and Bondy, 1987). However, verapamil could not inhibit to any extent the increase of [Ca 2 +]i by Met-Hg, suggesting that the elicited Ca 2 + influx was not through verapamil-sensitive Ca 2 + channels. The pattern of the increase of [Ca 2 +]i by Met-Hg resembled somewhat that of veratridine. After an initial rapid increase, there was a slower linear phase. In the case of veratridine the latter slower phase is likely due to increased release of Ca 2 + from mitochondria by Na+;ca 2 + exchange (Akerman and Nicholls, 1981b).
Met-Hg increases synaptosomal Na+ (Cheung and Verity, 1981) and if the Na+ influx is through Na+ channels, tetrodotoxin should block it. The effect of Met-Hg on [Ca 2 +]i could not be blocked by tetrodotoxin, suggesting that Met-Hg did not have any specific veratridine-like effect on Na+ channels. Met-Hg also inhibits Na+-K+-ATPase (Henderson et al., 1979) and could increase synaptosomal Na+ also by inhibiting the extrusion of Na+. A specific inhibition of Na+-K+-A TPase with ouabain increased [Ca 2 +]i slowly, but by only about 100 nM in the time range when Met-Hg had already doubled the resting level. Because the Met-Hg effect on [Ca 2 +]i was additive with the corresponding effects of both ouabain and veratridine (unpublished result), an increase in intrasynaptosomal Na+ could contribute only in part to the observed elevation of [Ca 2 +]h by Met-Hg.
Met-Hg might also elevate [Ca 2 +]i slightly from intrasynaptosomal sources, since it marginally increased [Ca 2 +]i in nominally Ca 2 +-free medium. Met-Hg is known to affect mitochondrial functions in situ (Verity et al., 1975;Fox et al., 1975;Cheung and Verity, 1981 ), suggesting that it can reach mitochondria. Met-Hg was less effective in elevating [Ca 2 +]i in synaptosomes, where Ca 2 + uptake into mitochondria was inhibited (rotenone + oligomycin) or where in addition ATP was depleted (rotenone + oligomycin + iodoacetic acid) beforehand. This clearly suggests that mitochondria are a target of Met-Hg which possibly inhibits mitochondrial Ca 2 + uptake. Although Met-Hg decreased synaptosomal ATP, a similar change effected by rotenone and oligomycin in combination (about 40%) did not yet effect [Ca 2 +]j. Only a nearly complete depletion of ATP increased [Ca 2 +]i by 250-300 nM as observed also by Nachshen ( 1985).
TEL and TET differed clearly from Met-Hg in that their effects were saturable and both induced only a 100-200 nM increase in [Ca 2 ili. TEL had a veratridine-like effect whereas TET was completely nonspecific. The data suggest that TEL increases intrasynaptosomal Na+ concentration by stimulating Na+ influx through Na+ channels and by inhibiting extrusion of Na+ out by Na+ -K+ -A TPase. Such a change in intrasynaptosomal Na+ could gradually depolarize the plasma membrane and allow the specific Ca 2 + influx through the voltage-sensitive Ca2+ channels. It remains uncertain by which mechanism TET increased [Ca 2 +];, but it did not render the plasma membrane similarly leaky to Ca 2 t as did Met-Hg.
Neither TEL nor TET significantly decreased synaptosomal ATP content and their effects on [Ca 2 +]i were independent of the simultaneous modulation of the function of mitochondria with other agents. Hence, the effect on the plasma membrane was far more important with respect to [Ca 2 +]i than any possible mitochondrial effect.
The observation that neurotoxic derivatives of alkylmetals increased free intrasynaptosomal Ca 2 ' has important implications. Such an increase in free Ca 2 + is expected to stimulate intracellular Ca 2 + -sensitive reactions providing alkylmetals do not directly inhibit them. The spontaneous release of monoamines from synaptosomes (Komulainen and Tuomisto, 1981) and that of acetylcholine from motor nerve endings (Miya-