Ganglioside GM1 prevents and reverses toluene-induced increases in membrane fluidity and calcium levels in rat brain synaptosomes.

The effects of exposure to ganglioside GM 1 and to toluene in vitro upon synaptosomal integrity have been examined using fluorescence polarization of two probes: 1-[4(trimethylamino)phenyl]-l,3,5-hexatriene (TMA-DPH) and l,6-diphenyl-1,3,5-hexatriene (DPH) to measure membrane anisotropy, and the fluorescent indicator fura-2 to assay levels of cytosolic calcium ([Ca 2 +l). The anisotropy of both TMA-DPH and DPH was decreased by toluene, implying increased membrane fluidity. The decrease in TMA-DPH but not in DPH anisotropy was prevented by pretreatment with GM 1 in concentrations as low as 10 11M. This is not an additive interaction since 10 µM of GM 1 alone did not significantly modulate TMA-DPH anisotropy. When the GM 1 treatment succeeded the addition of toluene the decrease in anisotropy of both probes was reversed. Toluene treatment increased [Ca 2 +]; in a dose- and time-dependent manner. This increase could partially be both prevented and reversed by treatment with 50 µM of GM 1• These effects may reflect an additive interaction, since this concentration of GM 1 alone reduced [Ca 2 +];. The present results show that toluene increases membrane fluidity and intracellular calcium levels. These effects may be counteracted by the endogenous compound GM 1.


INTRODUCTION
The monosialoganglioside GM 1 is an endogenous constituent of the plasma membrane, and is concentrated in nerve endings 29 . GM 1 , in contrast to polysialogangliosides, has been implicated in synaptic transmission 1.
3 • 2 1. 3 6 .4 4 .4 5 and behavior 23 . Thus, gangliosides have been demonstrated to affect adenylate cyclase 7 · 33 and protein kinase C 7 • 24 . Furthermore, GM 1 can bind calcium in vitro 20 • 35 .4°, and may thus be involved in the control of calcium flux over the nerve cell membrane 46 . Treatment with GM 1 has by some authors been implicated to decrease membrane fluidity 2 .4 2 , while others have found increases of membrane fluidity 31 • GM 1 has frequently been reported to increase neural regenerative capacity 22 · 28 · 32 .4 1 . More specifically, this ganglioside can decrease synaptosomal vulnerability to chemical injury 4 . The organic solvent toluene has been suggested to induce its effects on chemical neurotransmission in the nervous system by altering membrane fluidity 13 · 15 and calcium flux 9 • 10 . In view of the above, GM 1 may thus interfere with the actions of toluene. In line with this hypothesis, treatment with GM 1 antagonizes the effects of toluene on central dopamine 0 2 receptors 11 . Therefore, we have investigated the effects of GM 1 in vitro on toluene-induced changes in membrane fluidity parameters and changes in synaptosomal calcium levels.

Synaptosomal preparation
Brains were rapidly removed and the forebrain was dissected out on ice, weighed and homogenized in 10 vols. of 0.32 M sucrose at 0 °C. The homogenate was centrifuged (1500 g, 10 min) to give a post nuclear supernatant which was layered over 1.2 M sucrose and centrifuged for 25 min (250,000 g). The interphase band was removed and layered over 0.8 M sucrose and centrifuged again for 25 min at 250,000 g 8 • 16 . The purified synaptosomes were resuspended in HEPES buffer (pH 7.4) at 0.15 g-equiv/ml corresponding to a final concentration of 120-140 µg protein/ml 5 . The HEPES buffer was composed of (in mM): NaCl 125, KCl 5, NaH 2 P0 4 1.2, MgCl 2 1.2, NaHC0 3 5, Glucose 6, CaCl 2 1, HEPES 25; and adjusted with NaOH to pH 7.4.

Membrane order
Synaptosomal membrane order was evaluated by fluorescence polarization studies using two probes. 1,6-Dipbenyl-1,3,5-hexatriene (DPH) is a no-polar lipophilic molecule capable of penetrating into and through inner lipid-rich membrane layers 38 . 1-[4(trimethylamino)phenyl}-l,3,5-hexatriene (TMA-DPH) is a related compound with a polar-constituent that causes the molecule to be aligned at the outer surface of limiting membranes with the polar head at the hydrophilic surface, while the non-polar body penetrates the lipid interior 26 • 34 . Synaptosomes were prepared as described above and were incubated at 37 °C in the presence of TMA-DPH (30 min) or DPH  The synaptosomes were reprecipitated (13,000 g, 2 min) using a microfuge (235B, Fisher Scientific), resuspended in HEPES buffer and allowed to equilibrate for 10 min at 37 °C prior to measuring of fluorescence intensity and polarization in a water-jacketed cuvette holder maintained at 37 °C (Aminco SPF-500 spectrofluorometer, American Instrument Co., Urbana, IL, U.S.A.). An excitation wavelength of 360 nm (bandwidth 10 nm) was used with a determination of emission at 430 nm (bandwidth 10 nM). Corrections for light scattering (membrane suspension minus probe) and for fluorescence in the ambient medium (after pelleting membranes) were made.
Fluorescence anisotropy (r) was determined by the formula r = lvv-lvHl(lvv + 2(/vH)]. lvv is the fluorescence intensity with excitation and emitted light polarized vertically and lvH is the intensity obtained with a vertical orientation of the exciting polarizer with the emitted fluorescence passing through a horizontal polarizer. Total fluorescence intensity F = lvv + 2(/vH). Relative microviscosity is proportional to rjr-1, where r 0 is the maximal limiting anisotropy of the probe; 0.362 for DPH 37 and 0.39 for TMA-DPH 34 . A correction factor (G) for instrument asymmetry was also made using G = 1 11 vll 1 rn where /Hv is fluorescence intensity with horizontal excitation light and emitted light read vertically, and /HH is the corresponding value with the entire light path horizontally aligned. This compensates for the sensitivity of the detection system toward vertically and horizontally polarized light. All values of fvu were multiplied by G in the calculation of r. All corrections made amounted to less than 6% of the original unmodified readings.
lntrasynaptosomal calcium levels Synaptosomes were prepared as described above and were incubated at 37 °C in the presence of fura-2 dissolved in DMSO for 10 min 25 . They were then diluted 10 times in HEPES buffer and incubated for another 5 min. The final concentration of fura-2 was 5 µM. The synaptosomes were centrifuged for 8 min at 3000 g and the pellet was resuspended in HEPES buffer.
For each assay, 0.5 ml of synaptosomes was rapidly centrifuged (2 min, 13,000 g) and the resulting pellet was resuspended in 1 ml HEPES buffer at 37 °C. The buffer was as described above without NaHC0 3 and NaH 2 P0 4 to prevent the precipitation of calcium at elevated pH (required during the determination of minimal fluorescence). The tube was rinsed with another 1 ml of HEPES buffer and the total 2 ml sample was placed in a quartz cuvette at 37 °C and left to equilibrate for 10 min. Excitation of fura-2 was at 340 and 380 nm (bandwidth 3 nm) and emission determinations were made at 510 nm (bandwidth 20 nm). Corrections for light scattering (membrane suspension minus probe) and fluorescence in the where Kd is the dissociation constant of the fura-2-Ca 2 + complex, taken to be 224 nM 17 , Rmin is the ratio in the presence of excess amounts of EGTA (10 mM) and Rmax the ratio in excess amounts of calcium (18 mM). S 12 and Sb 2 denote fluorescence of fura-2 at zero calcium concentration and full calcium saturation, respectively, at an excitation wavelength of 380 nm.
Dye leakage into the extracellular compartment was calculated after the addition of MnC1 2 to a final concentration of 4 µM. The resulting depression in emitted fluorescence when excitation was at 340 nm, was expressed as a percentage of the difference between the corresponding value prior to MnC1 2 addition and the value at zero calcium conditions (EGTA and 0.1 o/r sodium dodecyl sulfate being present).

Incubations
After initial determinations, toluene (0.06-6 ,ul) or ganglioside GM 1 (Hl-50 µ1) were added to the synaptosomal suspension and values were measured after a 10-min incubation for calcium determinations. or after a 30-min equilibration for fluorescence polarization assays. Toluene was mixed with the synaptosomes by vortexing (5 s) in 10 ml polypropylene test tubes. Following this, any further additions were made and the fluorescence was read again following a further 10 or 30 min, respectively, of incubation. ..

Statistical analysis
Differences between groups were assessed by Fisher's Least Significant Difference Test after one-way analysis of variance. The acceptance level of significance was P < 0.05 using a two-tailed distribution.
Ganglioside GM 1 (50 µM but not 10 µM) was found to increase anisotropy of both TMA-DPH and DPH (Table I).
Pretreatment with 10 µM of GM 1 completely abolished the reduction of TMA-DPH anisotropy induced by 2 µI toluene, without affecting the toluene-induced reduction in DPH anisotropy (Fig. 2). Following pretreatment with   50 µM GM 1 , the inhibition by toluene on TMA-DPH anisotropy remained, whereas the toluene-induced decrease in DPH anisotropy was reduced by 50%.
The toluene-induced decrease in TMA-DPH and DPH anisotropy was reversed by subsequent treatment with GM 1 . Ten µM of GM 1 counteracted the decrease in TMA-DPH and DPH anisotropy by 50%, and 50 µM of GM 1 reversed the decrease completely (Fig. 3). The anisotropy of treated and non-treated synaptosomes did not change over time (last measurement made after 60 min of incubation, data not shown).

Studies on [Ca2+ ];
Toluene was found to increase the levels of intrasynaptosomal calcium in a concentration-and time-dependent way without affecting membrane leakage (Figs. 4 and 5). At 2 µI of toluene the [Ca 2 +]; was close to the upper limit of the assay capacity of fura-2, and at 6 µl of toluene the synaptosomes clumped together. Absence of vortexing did not allow toluene to act on synaptosomal calcium levels (data not shown). This time-dependent increase in [Ca 2 +]; by toluene contrasted with the   GMi-induced reduction in calcium levels (Fig. 5). Under control conditions, the [Ca 2 +]; increased slowly. Pretreatment with GM 1 was found to prevent the toluene-induced increase in calcium levels (Table II). The effect was most pronounced at 50 µM of GM 1 , at which concentration GM 1 almost completely blocked the toluene effects. Addition of GM 1 (50 µM but not 10 µM) subsequent to toluene treatment was found to reverse the toluene-induced increase of [Ca 2 +]i by 50% (Table II). DISCUSSION It was found that toluene in vitro reduces the anisotropy of both TMA-DPH and DPH in synaptosomal membranes. Our study directly shows that toluene affects and increases membrane fluidity as earlier postulated 15 . These actions may have neurotoxic consequences in vivo, and are paralleled by effects of other organic solvents 19 · 30 • 39 . Although statistically significant, the observed changes in anisotropy are relatively small in absolute terms. However, as is clear from enzyme studies, even minor variations in the physical form of proteins can have major functional effects. Thus changes in biological activity often have a greater magnitude than the structural alterations of the molecules that underlie such changes 43 . However, intraperitoneal injections with toluene in vivo had no effect on membrane fluidity as measured by fluorescence polarization 27 . The present results confirm that GM 1 increases the rigidity of the cell membrane 2 .4 2 rather than decreases it 31 • Already at 10 µM, a concentration at which GM 1 by itself has no effect on membrane fluidity, GM 1 is able to completely antagonize the toluene-induced reduction in TMA-DPH anisotropy. However, GM 1 is unable to prevent the actions of toluene on DPH anisotropy, perhaps due to the exclusive localization of GM 1 to the external surface of the cell membrane 18 . However, GM 1 reversed the toluene-induced reductions of both TMA-DPH and DPH anisotropy.
This may be the first example where treatment with GM 1 after a neurotoxic insult can block potentially deleterious changes in nerve tissue. The ability of GM 1 to reverse the DPH effects may be due to a damage to the membrane integrity by toluene, allowing GM 1 to gain access to the inner, hydrophobic part of the cell membrane.
Toluene increased intracellular Ca 2 + levels, but interestingly, membrane leakage was not affected by toluene. Other studies have shown that toluene increases calcium uptake in synaptosomes 9 • 10 and affects calcium regulated protein phosphorylation 12 • 1 4 . Thus, the increase in calcium levels may be another mechanism underlying toluene neurotoxicity. GM 1 , on the other hand, stabilized intracellular calcium levels, in contrast to the slow increase in [Ca 2 +]; seen under control conditions. Thus, both the preventive and the reversal effects of GM 1 on the toluene-induced increases in calcium levels may represent an additive interaction.
The importance of vortexing for attaining the effects of toluene established in the present study emphasizes the need for standardization of vortexing protocols. Thus, it is difficult to quantitate the relation between the amount of toluene added and the resulting concentration in the membrane bilayers. The same problem is also valid for the discussion on whether the results can be transferred to in vivo situations.
In conclusion, the present study shows protective and reversal effects of ganglioside GM 1 on toluene-induced increases in membrane fluidity and intrasynaptosomal levels of free ionic calcium. It may be that treatment with GM 1 may counteract neurotoxic effects induced by toluene and other organic solvents. This suggests a potential therapeutic use of this ganglioside.