Effects of exposure to Freon 11, 1,1,1-trichloroethane or perchloroethylene on the lipid and fatty-acid composition of rat cerebral cortex.

Organic solvents are often present as mixtures in various industrial and house-hold products. The adverse effects arising from exposure to these solvents have often been generalized to concern the whole group of solvents. In an examination of the possibility that organic solvents have general effects on experimental animals, rats were continuously exposed to vapors of the halogenated solvents Freon 11, perchloroethylene, and 1,1,1-trichloroethane. The lipid composition and fatty-acid pattern of ethanolamine phosphoglyceride from the cerebral cortex were analyzed. It was observed that only perchloroethylene had effects on the brain lipid composition. Cholesterol and total phospholipids were slightly reduced. Among the fatty acids the proportion of stearic acid was reduced and those of docosapentanoic, 22:5 (N = 6), and of docosahexanoic, 22:6 (N = 3), acids were increased. The changes in the fatty-acid pattern indicate that an alteration occurs in the desaturation of fatty acids. It seems probable that the chloroethylenes have specific effects on the fatty-acid pattern of brain phospholipids not shared by other solvents.

The biological effects of or ganic so lvent s and related compounds on th e nervou s system ar e of great interest due to their widespread use in industrial and household products. Long-term exposures have been suspected to result in a solvent -relate d "psychoorganic syndr ome. " In a series of pap ers (10,11,12,14) we have reponed the effects o f chlorinated ethylenes, tr ichloroeth ylene, and per chloroethylene on the lipid compositio n o f th e brain and, in particular, on the fatty-acid pattern of eth anolamine phosphoglyceride (EPG) . We ob served th at the fatty-acid pattern of EPG was changed among the mo st pol yun saturated fatty acids (P UFA), particularly in the region s rich in gray matter , such as the cerebral co rtex and the hippocampus. Th ese changes wer e characterized by a relative increase in the linoleic ac id-de rived (N-6) PUFA , and a decrease in the linolenic acid-derived (N-3) PUFA. Both gerbils and rats wer e stud ied at various du ration s of expos ur e to tr ichloroeth ylene. Significant alteration s appeared a fter 30 d of continuous exposure to 320 ppm. Since continuous expo sure to organic solvents for 30 d at moderate con centrations seems to be a sufficient tim e to det ect earl y changes in br ain lipid s, we cho se thi s expo sure for the screening o f three commonly used , poorly metabolized halogenated solvents, ie, perchlor o-I Institute of Neurobiology, University of Goreborg, Goreborg, Sweden. z Gambro AB, Lund, Sweden.
ethylene (P C E), I , I, I-trichloroethane (l , 1,1-TCE), and Freon II , in the pre sent study . We had studied P CE previously (10,14), but onl y in the gerbil and for much lon ger exposure times. The other two solvents have never pr eviou sly been studied with respect to their effe cts on brain membrane lipids. The gross lipid compo sition and the fatt y-acid pattern of EPG were studied in the cerebral cort ex of animals expo sed continuously for 30 d to P CE, I,I,I-TCE, or Freon II.

Materials and methods
Sepa rate experiments were performed, on e for each of th e sol vents. Male Sprague-Dawley rats were used (Anticimex AB, Sweden). Animal-weight means at the beginning of the experiments were as follows: P CE co ntrols 196 g, PCE-exposed 197 g; I ,I,I-TCE co ntrols 210 g, I , I, I-TCE-expo sed 208; Freon II controls 269 g, Freon l l -exposed 270 g . The animals were housed in pla stic cag es placed inside th e inha lation chambers during the exposure period. Commercial laboratory rat chow (Ewos, Sweden) and water were freely availabl e.
The expo sure equipment has been de scribed in detail previously (8). PCE or I ,I,I-TCE (Uddeholm , Sweden) was mixed with ai r (2 200 II h) to the des ired concentration of 320 ppm ; for Freon II (Imperial Ch emical Industries, Great Britain) the con centration used was 580 ppm . Animals were continuously expo sed for 30 d and each exposed gro up was matched by a control group that was housed in an ident ical chamber and exposed to air throughout the experimental period. At the end of the experiment the animals were immediately decapitated as they were brought from the exposure chambers.
After decapitation the brains were rapidly removed from the skulls and the cerebral cortex was dissected out. The dissected specimens were weighed, folded in aluminum foil, and then individually frozen on dry ice to be used later for the lipid determination.
Only the right cerebral cortex was analyzed. The tissue was homogenized in a glass homogenizer in 2 ml of methanol at O°C, and I ml of chloroform was added during the homogenization. The homogenate was centrifuged at 1 000 g for 10 min in the homogenizers and reextracted by the aforementioned procedure.
The crude lipid extract was purified from nonlipid contaminants by a reversed-phase column method (9). To 6 ml of the crude extract 8 ml of methanol/water (I: I) was added and mixed to homogeneity. The entire mixture was poured onto a 3-ml C-18 "Bond Elut" column (Analytichem International, California, United States) equipped with a 50-ml reservoir. Column elution was carried out under a constant gentle vacuum with an elution speed of 1-2 ml/min. Another 8 ml of methanol/water (I: I) was added to the eluate, mixed, and again passed through the column. The resulting eluate was once again mixed with 8 ml of methanol/water and passed through the column, as before. This last eluate was discarded before elution of the bound lipids. A 5-ml container was mounted on top of the columns, and the lipids were eluted at a moderate rate with two portions of 4.0-ml chloroform/methanol (I : 2). This final eluate was used directly for quantitative analyses of phospholipids and cholesterol.
Cholesterol was analyzed with a colorimetric method (4), and lipid phosphorous was measured according to Bartlett (3). The EPG fraction was isolated by thinlayer chromatography (TLC). The bands corresponding to 1.5 urnol of phospholipids were applied to 0.25-mm TLC plates (Merck, Federal Republic of Ger-many) that were developed unidirectionally in chloroform/methanol/water (65: 25 :4). The lipid bands were visualized by a spray application of a bromphenol blue reagent to the TLC plate (15). The lipid bands corresponding to the EPG fraction were scraped into tubes with a teflon-lined screw cap. Fatty-acid methyl esters were prepared by transmethylation with sodium methylate in methanol at 37°C for 1 h (IS). The methyl esters were extracted in petroleum ether and analyzed by gas chromatography on a 2-m glass column packed with 10 070 OEGS·PS on Supelcoport" (Supelco Inc, Pennsylvania, United States).
A statistical evaluation with Student's t-test was used to compare the exposed groups with their appropriate control animals.

Discussion
Of the three solvents investigated (PCE, I ,I,I-TCE,  and Freon II), only PCE affected the tissue weight s and the brain lipid and fatty-acid composition. In our previous report on PCE we found the whole-brain weights to be reduced in gerbils that where continuously exposed for 90 d at 320 ppm (14). Thi s observation was not reproduced in the present experiment exposing rats for 30 d to PCE. The small decrease in the brain-to-body weight ratio observed in this report is probably of little importance. The slight decrease in the contents of cholesterol and phospholipids of the cerebral cortex after exposure to PCE might indicate a loss of lipids. Of the other solvent s we ha ve investigated, only toluene has resulted in a decrease in phospholipids (13). The importance of changes in the brain lipid class composition after exposure to PCE has to be further investigated.
After exposure to PCE, an increase in the proportion of long-chain PUFA 22: 5 (N-6) and 22: 6 (N-3) at the expense of stearic acid was observed. These two fatty acids are the end products of the desaturation and elongation sequences of the two essential fatt y-acid families, the linoleic acid series (N-6) and the linolenic acid series (N-3). The general increase in the 22-carbon fatty acids of both the N-6 and N-3 series would suggest an increased desaturation and elongation of the last steps in the sequence of the ir syntheses. Such a conclusion is in agreement with that of our previous study on gerbils (14) . However in the present study onl y the metabolic end products , PUFA, were significantly changed. This finding contrasts with those of our previous study (14), in which some of the intermediate PUFA were also affected . The other two solvents in this study, 1,1,I-TCE and Freon II, had no effects on the animals. An explanation fo r this different behavior of the so lvents might be the very low uptake of freons in blood (2) and the relatively low uptake and solubility of 1,1, l-trichloroethane in blood (I, 6). Another explanation could be that the chloroethylenes specifically interact with the enzyme systems which regulate desaturation and elongation.
Changes in the membrane lipid and fatty-acid composition aft er the expo sure of animals to anesthetics and related compounds, such as organic solvents, have often been attributed a compensatory role for the increased fluidity or disturbed packing of membrane lipids which results from the introduction of foreign lipophilic molecules into the membrane (7). The changes in the fatty-acid pattern observed after exposure to PCE in this study did not fit directly into an y 3 of the common hypotheses for membrane adaptation. However, in view of the complex behavior of anesthetics on membranes (5), the po ssibility that a compensatory mechanism was responsible for the observed changes in the fatty-acid pattern cannot be completely excluded.
From this and from previous studies regarding xylene and toluene (13), it appears that 30 d of continuous exposure of rats to organic solvents at a moderate concentration (320 ppm) is suitable for the screening of lipid change s in the brain during such exposures. From these experiments it can be concluded that some of the solvents investigated (xylene, Freon II , and 1, I , I-trichloroethane) had little or no effect on the brain lipid or fatty-acid composition. Other solvents like toluene and the chloroethylenes had specific effects related to the compound. No general sol vent effects could be observed. These findings stress the importance of knowledge about the specific toxic effects of various solvents which are often present together as mixtures during commercial and household use.