Regulation of Phospholipid Metabolism in Differentiating Cells from Rat Brain Cerebral Hemispheres in Culture

Cultured dissociated cells from rat embryo cerebral hemis-phere iqcorporate [gH]- and [U-14C]ethanolamine into cellular lipids. Nearly all radioactivity in the lipid fractions is incorporated into 1,2diacylethanolamine phosphoglycerides and I-alkenyl,P-acylethanolamine phosphoglycerides (plasmalo-gen). Kinetic data suggest that the rate of labeling of both ethanolamine phospholipids from the phosphorylethanolamine is similar.

# Present address, the Department of Biochemistry, The Weizmann Institute of Science, Rehovot, Israel. ethanolamine lipids in brain tissue could be derived from the corresponding alkyl-ether lipids. The latter are believed to be formed via the cytidine nucleotide pathway (13). Studies by Joffe (14) suggested that the alkyl, alkenyl, and acyl derivatives are synthesized independently providing there is an adequate supply of the appropriate diglyceride.
Little is known about the relationship between free ethanolamine and phosphorylethanolamine utilization and the biosynthesis of plasmalogens and diacyl ethanolamine phosphoglycerides in growing brain cells in culture. In the preceding paper (15) it was demonstrated that ethanolamine is converted to phosphorylethanolamine which accumulates within the cells. It was therefore of interest to examine the role of phosphorylethanolamine for the formation of ethanolamine-containing phospholipids. This report provides evidence that phosphorylethanolamine is the major precursor for the ethanolamine moiety of these phospholipids. It also suggests that the pattern of labeling of plasmalogens and diacyl ethanolamine phosphoglycerides are different. A short communication of this work has appeared (16).

MATERIALS AND METHODS
The general methodology for the growth of brain cells, the incubation conditions, and the extraction of lipids were described (15).

Purification
and Specific Activity Determination of Ethanolamine Phosphoglycerides The total cell lipid extract was applied on Silica Gel G plates (Analtech) and the various lipids were separated by two-dimensional thin layer chromatography (17). The area corresponding to the ethanolamine phosphoglyEerides (6 to 10 pg of Pi), containing both alk-1-envl-EPG ' and EPG. was scraned from the nlate and eluted twice with 2 ml of chloroiorm-methanol-water (6532534 by volume). More than 95y0 of the radioactivity was recovered by this procedure. The pooled eluate was subjected to alkaline hydrolysis in order to separate the EPG from alk-1-enyl-EPG by a slight modification of the procedure described by Horrocks  suggested that a small fraction of the label could correspond either to alkyl ether derivatives or to cyclic acetal artifacts (22). In all subsequent studies, the specific radioactivity referred to as alk-1-enyl-EPG may also contain between 10 to 15% of these saturated ether bond containing compounds. A minor, unidentified radioactive compound migrated behind the lyso derivative of alk-1-enyl-EPG. The remaining radioactivity was equally distributed between sphingomyelin and choline phosphoglyceride.
Alk-1-enyl-EPG and EPG Levels in Growing Brain Cultures-Most of the label present in the lipid fraction from the [U-lC]ethanolamine was incorporated in the alk-1-enyl-EPG and EPG. Since the developmental patterns of rat brain lipids indicated a considerable increase in the content of alk-1-enyl-EPG (3)(4)(5), the levels of these compounds were determined in cultured rat brain cells. At 2 days in culture, when the cells are at their initial stage of morphological differentiation as judged microscopically by the presence of neuritic extensions, the levels of alk-l-enyl-EPG and EPG are 15.8 and 32.0 nmol/mg of protein, respectively ( Table  I). The EPG approaches maximum levels by the second week in culture, while alk-1-enyl-EPG levels are still increasing. Microscopic observations of the brain cells at this time revealed the presence of long and thick neuritic processes which may indicate a state of differentiation (23).

Time Course for [ U-"C]Ethanolamine
Incorporation in Alk-lenyl-EPG and EPG-When 7-day-old cultures were incubated with [ U-14C]ethanolamine for various periods of time, the specific radioactivities of both alk-1-enyl-EPG and EPG were constantly increasing (Table II). The specific radioactivities were low at the initial times examined; however, that of EPG was greater than that of alk-1-enyl-EPG.
The specific activity of the alk-l-enyl-EPG was greater than that of EPG after 6 hours and at all subsequent times examined. The ratio of specific radioactivity of alk-1-enyl-EPG to EPG at 24 hours was about 4-fold higher than that observed after 5 min.
Labeling Patterns of Ethanolamine-containing Phospholipids jrom Etharwlamine and Phosphorylethunolamine-The incorporation of [ U-lF]ethanolamine into lipids was greatly impaired when the incubation temperature was reduced from 37 to 15 or 22" as previously demonstrated (15). After 6 hours of incubation at 15", the incorporation of labeled ethanolamine into alk-l-enyl-EPG and EPG was reduced more than 50-fold and 20-fold, respectively (Table III). The ratio of specific activity of alk-1-enyl-EPG to EPG was greater at 37 than at 15". At 22", the incorporation of  increase in the alk-1-enyl-EPG and EPG specific radioactivities. However, the ratio of both compounds remained unchanged. After 6 hours of incubation at 22", about 26% of the radioactivity was present intracellularly as phosphorylethanolamine. The finding that the ratio of alk-1-enyl-EPG to EPG remained unchanged after cells were transferred from 22 to 37" (Table III) did not parallel the expected patterns as depicted in Table II. The experimental conditions between these two sets of experiments were not identical. The major difference was the absence of free [ U-i4C]ethanolamine (Table III). This observation prompted us to examine the metabolic turnover of phosphorylethanolamine and free ethanolamine in relation to alk-l-enyl-EPG and EPG. Brain cultures were incubated with [U-WJethanolamine for a period of 3 hours. This procedure enabled us to label the endogenous pool of phosphorylethanolamine since preliminary studies indicated that phosphorylethanol- A half-life of 8.5 hours was calculated for the ["Clphosphorylethanolamine.
About 7% of the r4C label was present in the medium after 5 hours of incubation.
The changes in the specific radioactivities of the three major labeled compounds is presented in Fig. 2 were the same as described for Fig. 1. The specific activities expressed as counts per min/nmol of lipid phosphorus represent the material obtained from pooled extracts of two to three cultures. PEa, phosphorylethanolamine; PEPG, alk-1-dnyl-EPG. bation in the presence of [*H]ethanolamine, the specific radioactivities of alk-1-enyl-EPG and EPG were 146 and 215, respectively (Fig. 2). Between the 1st and the 5th hour, the incorporation of *H into EPG and alk-1-enyl-EPG increased 8-to 11-fold, while the incorporation from the 14C label increased by a factor of 1.6 to 1.7. The relative aH :"C ratio in alk-1-enyl-EPG and EPG compounds is constantly increasing, reaching values of 0.8 and 0.62, respectively, after 5 hours incubation. The ratio of alk-1-enyl-EPG:EPG specific radioactivity with respect to the "C-isotope was constant (0.7) for all time points examined. This would corroborate the data presented in Table III, suggesting that the endogenous ["C]phosphorylethanolamine is incorporated into both phospholipids at the same rate. The specific radioactivity ratio with respect to the tritium label showed a significant increase from 0.68 at 1 hour to 0.93 by 5 hours. This could indicate that the rate of labeling of alk-1-enyl-EPG was higher than that of EPG from the tritium labeled precursors, in an analogy to the experimental conditions presented in Table II. E$ect of Various Bases on Ethanolamine Phosphoglyceride Metabolism-Pulse-chase studies in the presence of various bases were conducted to determine the changes in the specific radioactivities of the alk-1-enyl-EPG and EPG compounds. Cells had been prelabeled for 6 hours with [ U-14C]ethanolamine and were further incubated at 37" in the presence of large concentrations of either serine, choline, or ethanolamine (Table IV). After 18 hours of incubation the release of label from the cells into the medium in the presence of ethanolamine was 3-to 4-fold higher than in any of the other experimental conditions employed. No significant differences were observed either in the release of radioactivity into the medium or in quantity of radioactivity present in the lipids when serine or choline were present as compared to the control. Thin layer chromatography of the medium indicated that about 60% of the radioactivity was present in the form of phosphorylethanolamine and about 40% as free ethanolamine in all cases. The specific radioactivities of the alk-1-enyl-EPG and EPG compounds in controls continued to increase up to 30 to 50% of the initial values. Except for the study in which ethanolamine was present, a constant ratio of the alk-1-enyl-EPG:EPG specific radioactivities (1.08-1.11) was observed in all cases.

DISCUSSION
This report provides further evidence for the role of phosphorylethanolamine as the major precursor for the ethanolamine moiety of phospholipids in cultured brain cells. During a 24-hour exposure to [ U-"Clethanolamine there is a greater incorporation into alk-1-enyl-EPG than EPG (Table II). This is concluded from results of studies in which cells prelabeled with [ U-"Clethanolamine continued to produce labeled alk-1-enyl-EPG and EPG even after the removal of labeled ethanolamine (Tables III and  IV, Fig. 2). In the absence of free ethanolamine, there was no subsequent increase of the alk-1-enyl-EPG specific radioactivity over that of EPG. When a small dose of [8H]ethanolamine is added to the medium the specific activity of [BH]alk-l-enyl-EPG was greater than that of [8H]EPG (from 0.68 to 0.92). The incorporation of label from the ["Clphosphorylethanolamine precursor into both ethanolamine phospholipids under those conditions proceeded at the same rate (0.7) (Fig. 2). By employing this double label approach it seems possible to differentiate between the metabolic roles of phosphorylethanolamine and ethanolamine as precursors of phospholipids.
In order to explain the relative increase in the alk-1-enyl-EPG labeling, an additional sequence of reactions which is dependent on the availability of free ethanolamine is suggested. This is summarized in the following scheme where PEA is phosphorylethanolamine and PEPG is alk-lenyl-EPG.
The direct conversion of [aH]phosphorylethanolamine to [*HIalk-1-enyl-EPG (Reaction Sequence I) seems to lag in its capacity of labeling this compound. This is seen from the precursor product-time relationship (Table II and Fig. 2). The lag period could be explained by a compartmentalization of Reaction Sequences I and II. It has already been pointed out that the disso- Ii3 HIEPG ciated brain cell cultures contain a mixed population of neurons, neuroglia, and ependymal-like cells (23). Thus, it is possible that certain types of cells will rapidly take up ethanolamine and accumulate it as a phosphorylated compound. The resulting phosphorylethanolamine would then be incorporated into alk-l-enyl-EPG and EPG at the same rate, as suggested by the data presented in Tables III and IV, and Fig. 2. Other cell types which may exhibit a greater lipid synthesizing capacity might incorporate ethanolamine more efficiently in alk-I-enyl-EPG than in EPG. In view of the recent in vitro study by Binaglia et al. (24) on the phospholipid metabolism of neuronal and glial cells preparations, such a possibility has to be considered. The slow rate of alk-1-enyl-EPG labeling at the initial time points studied may reflect a different site of synthesis of certain species of alk-l-enyl-EPG within the cell. The presence of long neuritic extensions in the brain cultures, in some cases up to 2 mm, has been documented (23). Studies employing labeled precursors indicated an active transport of a variety of metabolites along the axons (25). The possibility that phosphorylethanolamine is such a metabolite cannot be excluded. The physiological significance of the plasmalogen accumulation in the growing brain cells is as yet not clear (Table I). Histological stainings employing 1~x01 dye techniques (26) of 3-to 4-week-old brain cultures did not reveal the presence of myelin such as those observed in cerebellar explants (27). Biochemical parameters such as activities of synthetic enzymes and accumulation of "characteristic myelin lipids" predates the histological appearance of myelin in the brain (1, 2, 28-31). It is possible that in this culture system, which is cerebral in origin, alk-1-enyl-EPG is synthesized and accumulated in certain cell types, and initiation (microscopically undetectable) of myelin formation may indeed occur.
Further analysis of the various cell types and their metabolic capacity to utilize the free bases for phospholipid biosynthesis is in order.