ENZYMIC STUDIES ON GLIAL AND NEURONAL CELLS DURING MYELINATION
Helmut Woelk and Rosemarie Jahrreiss Einheit fUr Neurobiochemie der UniversitatsNervenklinik Erlangen, 852 Erlangen, and Universitats-Nervenklinik der Universitat des Saarlandes 665 Homburg/Saar, G.F.R. SUMMARY The formation of ethanolamine plasmalogen from labelled 1-alkyl-2-acyl-sn-glycero-3-phosphorylethanolamine was st udi ed in neurons and glial cells of the developing rat brain. It was found that the conversion of the ether to the enol-ether bond of the 1-alkyl moiety by the neuronal and glial desaturase system requires unsaturated fatty acids at the 2 position of the substrate. There is almost no difference between the activity of the neuronal and glial desaturase during the period of active myelination, whereas the neuronal cell fraction of the adult rats displays a threefold higher enzyme activity as compared to the glial cells. Evidence for the involvement of a microsomal electron transport system in the enzymic conversion of alkylacyl-glycero-3phosphorylethanolamine to ethanolamine plasmalogen was obtained by using specific antibodies against NADH-cytochrome b s reductase. Cytochrome b 5 stimulated the biosynthesis of ethanolamine plasmalogen.
Abbreviations used: GPC, sn-glycero-3-phosphorylcholine; GPE, sn-glycero-3-phosphorylethanolamine; GPS, sn-glycero-3-phosphorylserine.
43
J. Palo (ed.), Myelination and Demyelination © Plenum Press, New York 1978
H. WOELK AND R. JAHRREISS
INTRODUCTION Experimental evidence has been forwarded during the last years for marked differences in phospholipid metabolism between glialand neuronal cell-enriched fractions. The neuronal cell bodies were found to possess a much higher rate of base-exchange for both serine and ethanolamine than the glial cell-enriched fraction (6) and Freysz et al. (5) showed in their extensive investigation on the kinetics of the biosynthesis of phospholipids in neurons and glial cells, isolated from rat brain cortex, that neuronal phospholipids had a faster turnover than glial phospholipids. Recently, we obtained indirect evidence for a faster turnover of glycerophosphatides in neurons than in glial cells, since neurons contained a considerably higher phospholipase A1 and A2 activity when compared to the glial cell-enriched fraction (11). There is only little information on the mechanism. of the biosynthesis of ethanolamine plasmalogen (1-alk-1'-enyl-2-acylsn-glycero-3-phosphorylethanolamine) by brain cells. In the brain tissue ethanolamine plasmalogen reaches particularly high values accounting for approximately 80% of total ethanolamine phosphatides in the myelin sheath. Investigation of plasmalogen biosynthesis in intact cells has led to conflicting opinions regarding the aliphatic precursor of the alkenyl moiety. The postmitochondrial fraction of Ehrlich ascites cells and preputial gland tumors can synthesize ethanolamine plasmalogens from the long chain fatty alcohols and dihydroxyacetone phosphate (10) or l-alkyl-2-acylsn-glycero-3-phosphate (1). It appears to be well established that one mechanism by which l-alk-1 '-enyl-2-acyl-sn-glycero-3phosphorylethanolamine can be formed involves dehydrogenation of the 1-alkyl moiety and that cytochrome b 5 participates into the reaction (8). EXPERIMENTAL Preparation of glia and neurons from rat and rabbit brain. The neuronal and glial fractions were prepared from young and adult Wistar rats and from white rabbits weighing about 1,5 kg. The animals were anaesthetized with sodium pentobarbitone and killed by intracardiac perfusion of Ringer solution. The whole brain was removed quickly, weighed and sliced. The procedure employed for the preparation and identification of neurons and glia was essentially by the methods of Blomstrand and Hamberger (2, 3) and Goracci et al. (6) as indicated elsewhere (11). The~onal fraction contained more than 90% of neuronal cell bodies, freed for the greater part of their axonal processes and with very little contamination by glial cells. The glial cellenriched fraction contained about 90% glial cells. Endothelial cells and free nuclei were the main contaminants of both cell
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
45
fractions. The purity of the neuronal and glial cell suspensions was further assessed by using the base-exchange enzymic system as a marker for the neuronal cell bodies (6). Using serine and ethanolamine as nitrogeneous bases, neuronal/glial ratios of about 3 to 4 were found for the base-exchange activity. Oligodendroglia were obtained by the method of Poduslo and Norton (9). Subcellular fractionation of the cell-enriched preparations. Neuronal and glial microsomes were obtained as described in detail by Goracci et al. (6). The microsomes displayed 88% of the glucose6-phosphatase activity and 91% of the NADPH:cytochrome c oxidoreductase activity of t~~ whole cell homogenate. Substrates. 1-[ C]alkyl-2-acyl-GPE (substrate A) was prepared fro~4the brain phospholipids after intracerebral injection of 1-[ C}hexadecanol (52 ~Ci/~ole/animal) into 14 day-old-rats. Twenty-four hours after the administration of the labelled alcohol, the animals were sacrificed and the brain ethanolamine phosphoglycerides purified. In order to remove the 1-alk-1'-enyl portion, the ethanolamine phosphoglycerides were subjected to mild acid hydrolysis according to a modified Dawson~s procedure (4). 1,2Diacyl-GPE was removed with the use of purified lipase from porcine pancreas. The method is based on the selective deacylation by lipase action at the 1 position of the 1,2-diacyl compounds of naturally occurring phosphatide mixtures, containing 1,2-diacyl-, alkenyl-acyl-,1l:nd alkylacyl-glycerophosphatides. The final preparation of 1[ Clalkyl-2-acyl-GPE (specific activity of 4,2 ~Ci/ ~ole) was obtained by silicic acid or Florisil column chromatograph~14
L C}-labelled 1-alkyl-2-acyl-GPE (substrate B) was ~~epared from 14-day-old rats after intracerebral injection of 1-[ C ] acetate. Twenty-four hours after the administration of the labelled precursor the rats were sacrificed, the brain lipids extracted and the ethanolamine phosphoglycerides purified. Th r4 1-alk-1'-enyl-2acyl portion (ethanolamine plasmalogen) of the [ C] -labelled ethanolamine phosphoglycerides was transformed lnto the corresponding acylalkyl-compound by catalytic hydrogenation. In order to remove the the remaining plasmalogens, the ethanolamine phosphoglycerides were subjected to mild acid hydrolysis according to a modified Dawson~s procedure (4). The phosphoglyceride mixture, containing 1,2 diacyland 1-alkyl-2-acyl-GPE, was purified by means of column chromatography o~4Florisil. 1- L C]alkyl-GPE was obtained by silicic acid column chroma togra~uy after removing the fatty acids from the 2 position of 1- [ c] alkyl-2-acyl-GPE with phospholipase A2 from Naja naja venom. The labelled 1,2-diacyl-, 2-acyl-1-alk-1'-enyl- and 2-acyl-1-alkyl glycerophosphatides, used for the measurement of the phosphollpase A activity, were prepared and checked as described previously (13). 2 . . . . . . HydrolYS1S of the phosphoglycerldes, WhlCh dlffered ln the radlcal at the 1 position and were labelled at the 2 position with different fatty acids, by phospholipase A2 from naja naja venom, showed that the radioactivity was almost exclusively recovered in the fatty
46
H. WOELK AND R. JAHRREISS
acids freed from the substrates, indicating that specific incorporation of the labelled fatty acids into the 2 position of the phosphoglycerides had occurred. Hydrolysis of the 1,2-diacyl-glycerophosphatides, labelled at the 1 position, by phospholipase A2 from crotalus atrox venom, showed that 94-96% of the radioactivity was recovered in the lyso-compounds, indicating that specific incorporation of the labelled fatty acids into the 1 position of the lyso derivatives had occurred.
RESULTS The formation of ethanolamine plasmalogen from 1-l14c] alkyl(prepared ~,~m brain tissue after injection of 1- L Clhexadecanol) and L cj-labelled 1-al~~1-2-acYl-GPE (prepared trom brain tissue after injection of [ Cl acetate and catalytic hydrogenation of the plasmalogen portion) has ieen studied in neurons and glial cells of the d1~eloping rat brain (Table 1). As can be seen from Table 1, 1- [ c l alkyl-2-acyl-GPE is a much better substrate for the formation of ethanolamine plasmalogen by the neuronal and glial desaturation system than the corresponding hydrogenated glycerophospatide. It appears that the conversion of the ether to the enol-ether bond of the 1-alkyl moiety by the neuronal and glial desaturation system requires unsaturated fatty acids at the 2 position of the substrate. Furthermore, there is almost no difference between the activity of the neuronal and glial desaturase during the period of active myelination, whereas the neuronal cell fraction of the adult rats displays a threefold higher enzyme activity as compared to the glial cells. With increasing age of the animals a decrease in the desaturase activity can be observed. The decrease is much more pronounced in glia than in neurons (Table 1). The dehydrogenation of the 1-alkyl moiety requires a reduced pyridine nucleotide (NADH or NADPH) and is strongly inhibited by KCN, but not by CO/0 2 . Further evidence for the involvement of a mi~psomal electron transport system in the enzymic conversion of 1-L14 C]alkyl-2-acyl-GPE to ethanolamine plasmalogen was obtained by using specific antibodies against NADH-cytochrome b5 reductase. Table 2 clearly demonstrates the inhibition of plasmalogen formation by adding increasing amounts of antibody against NADH-cytochrome b 5 reductase to the incubation system. On the other hand, addition of cytochrome b5 stimulated the biosynthesis of ethanolamine plasmalogen from the corresponding alkyl-ether by oligodendroglial microsomes (Table 3). In order to investigate whether Piracetam (2-oxo-pyrrolidine-1-acetamide) has some effect on the biosynthesis of ethanolamine plasmalogen, the animals were injected i.p. for 10 days with increasing amounts of the nootropic substance. Table 4 shows that the effect of Piracetam on the formation of ethanolamine plasmalogen resembles the action of cytochrome b5 on the dehydrogenation of the 1-alkyl moiety as shown in Table 3. 2-aA~1-sn-GPE
- - - -
2.11 1.23 0.91
Neurons
------
2.05 0.84 0.36
Glia
----
1.03 1.46 2.53
Neuronal/ glial ratio 0.31 0.28 0.12
Neurons
0.33 0.22 0.05
Glia
B
0.94 1.27 2.40
Neuronal/ glial ratio
Each incubation mixture c~ntained in a total volume of 0.25 ml 0.1 ~ Tris/HCl buffer, pH 7.1: 15 nmoles of 1-r1 Clalkyl-2-acyl-sn-GPE (substrate A) or ~ C1-labelled 1-alkyl-2acyl-sn-GPE (hydrogenate~ s~strate, B), NaF (12 mM), ATP (10 mM), NArlP+ (2 mM). glucose-6-P (6 mM), MgC12 (4 mM). glucose-6-P-dehydrogenase (0.41 U), GSH (2 mM) and 325 to 480~g of glial or neuronal protein. The substrates were added as a sonicated dispersion in 0.15% Tween 80. Incubations were carried out for 60 min at 37°C_~n a shaking water bath. Values are expressed as nmol ethanolamine-plasmalogen formed x mg prot. x h- 1 •
14 days 21 days Adult
Age
A
Table 1. Formation of ethanolamine-plasmalogen from the corresponding 1-alky.1-2-acyl-snglycero-3-phosphorylethanolamine in neurons and glial cells of the rat brain during myelination.
~
oz
~
Z
r
-< m
s::
G)
Z
C :D
o
en
r
r
m
(')
» r
o z
C :D
m
Z
z o
»
r
5>
G)
r
o"T1
en
-< s:: m
N
z
m
2.0 4.0 8.0 10.0
-
(mg)
Control globulin
0.80 0.37 0.22
1.42
o .9A 31.3 43.8 74.5 82.2
1.26 1.13 1.10 1.14
10.8 20.4 22.5 19.7
I
4.0 8.0 16.0 20.0
Ethanolamine-Plasmalogen formed in the presence of Antibody Inhibition Control Inhibition Ratio globulin globulin antibody-micro somal protein (nmol) (nmol) (% ) (%)
Each incubation mixture 9Rntained in a total volume of 1.0 ml 0.1 M Tris/HCl buffer, pH 7.1: 60 nmoles of 1-[ Clalk.yl-2-acyl-GPE, NaF (12 mM), ATP (10 mM), NADH (2 mM), glucose-6-P (6 mM), MgCl2 (4 mM), glucose-6-P-dehydrogenase (1.64 U), GSH (2 mM), 0., ml of the 100,000 x g supernatant, 498 ~g microsomal protein and antibody or control globulin as indicated. The substrates were added as a sonicated dispersion in 0.15% Tween 80. Incubations were carried out for 60 min at 37°C in a shaking water bath.
2.0 4.0 8.0 10.0
-
(mg)
Antibody against NADH-cytochrome b5 reductase
Table 2. Influence~n fOti NADH-cytochrome b5 reductase on the formation of ethanolamineplasmalogen from l-L' Cjalkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by oligodendroglial microsomes of 20-d~-old rats.
~
m
::0 ::0
'-
» J:
::0
c
»z
~
r-
m
;t:
:e o
t
49
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
Both substances enhance the formation of ethanolamine plasmalogen from 1-[1 C] alkyl-2-acyl-GPE (Tables 3 and 4). In another series of experiments exogenously added glycerophosphatides, specifically labelled either in the 1 or in the 2 position, were used to measure the activity of phospholipase A2 from the neuronal-enriched cell fraction of the rabbit. As can be seen from Table 5 the enzyme hydrolysed the 1,2-diacylglycerophosphatides more rapidly than the acyl alkyl- and acylalkenyl-compounds. Choline plasmalogen and the corresponding alkyl-derivative were cleaved at almost similar rates by the phospholipase A2 . Among the various 1,2-diacyl-glycerophosphatides, the neuronal phospholipase A2 preferred phosphatidylcholine as a substrate, whereas phosphatidylserine was hydrolysed less actively. Table 5 shows, furthermore, that norepinephrine, injected into the lateral ventricle of the rabbit brain, stimulated the hydrolysis of the various glycerophosphatide substrates.
Table 3. Effect ofc~tfchrome b 5 on the formation of ethanolamineplasmalogen from 1-[1 Cjalkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by oligodendroglial microsomes of 20-day-old rats. Addition
Control IG Cytochrome b5
Concentration (mg/mg of micro protein)
Ethanolami nePlasmalogen formed (nmol/mg protein/h)
1.8 0.5 1.2
3.6
4.5
Incubation conditions as in Table 2.
2.94
2.78 3.85
4.49
5.67 6.23
Difference
C%)
6
+. 31 + 53 + 93 + 112
50
H. WOELK AND R. JAHRREISS
Table 4. Influence Of Piracetam on the formation of ethanolamineplasmalogen from 1-[14C] alkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by neuronal microsomes of 20-d~-old rats. Piracetam (mg/kg/day)
60
80 100 120 140
Ethanolamine-plasmalogen formed (nmol/mg protein/h)
2.96
Increase
(%)
3.87
30.7 54.0
6.02 6.12
103.3 108.7
4.56 5.76
94.6
The incubation conditions were as described ln Table 2 with the exception that instead of oligondendroglial microsomes neuronal microsomal protein was used. The animals were pretreated for 10 days with the dose of Piracetam.
DISCUSSION The present results have shown that the microsomal 1~alkyl-2acyl-GPE desaturation reaction is catalyzed by a membrane-bound multi component desaturase system requiring the supply of reducing equivalents from pyridine nucleotides. It may be supposed that in the desaturase reaction, which transforms the alkyl-ether to the enol-ether moiety, a cyanide sensitive factor (CSF) is operating as the terminal oxidase as it has been proposed for the oxidative conversion of stearoyl-CoA to oleoyl-CoA (7). The use of various specific inhibitors including antisera against NADH-cytochrome b5 reductase indicate the role of microsomal flavoproteins in the supply of reducing equivalents from NADPH or NADH to the cyanide sensitive factor. Cytochrome b5 seems to be functional as an electron carrier between the flavoproteins and the CSF. 1-alkyl-2-acyl-GPE with unsaturated fatty acids at the 2 position served as a good substrate for plasmalogen synthesis, whereas 1-alkyl-GPE was transformed to the corresponding plasmalogen only after the addition of CoA to the incubation system, indicating that the dehydrogenation of the 1.l..alkyl moiety only occurs on the intact phospholipid molecule. The observation that norepinephrine stimulates the hydrolysis of glycerophosphatides by phospholipase A2 (Table 5) could suggest that phospholipase A2 might be concerned in the molecular changes taking place in membranes during synaptic transmission. In an investigation on the incorporation of 3 2p into the phospholipids
33.2 27.8 27.2
20.3 18.5
17.8
73.9 61.3
52.8
50.2
63.5
54.7
51.7
(% )
Dlfference
.
(b)
~g of L-norepinephrine (for details,
(b)Per cent of increase in activity of norepinephrine experiments, as compared to controls.
(a)The animals were injected intraventricularly with 100 see Experimental.
Each incubation mixture contained in a total volume of 1.0 ml: 0.1 M sodium acetate buffer, pH 5.4; 1.0 ~mol of the glycerophosphatide substrate; 3 mg of sodium taurocholate and varying amounts of neuronal protein (ranging from 248 to 524 ~g). Incubation at 37°C for 1 h. For details, see Experimental. Each figure represents the average of five experiments; SEM was less than 8%.
1-[14cJstearOYl-2-aCY1-~-GPS 1-Alk-1'-enYl-2-[14cJlinOleOY1-~-GPC 1-AlkYl-2-[14c]linOleOY1-~-GPC
39.6
1-[14C]stearOYl-2-aCY1-Sn-GPE
[nmol/(mg x h)] 48.7
[nmol/(mg x h)]
Phospholipase A2 phospholipase A2
. . (a) Noreplnephrlne
Hydrolysis of different glycerophosphatides by neuronal phospholipase A2 .
1-[14c]stearOYl-2-aCY1-~-GPC
Substrates
Table 5.
m Z
~
z r (') m r r en o
o
C :0
m
Z
o
Z
l>
» r
r
Cl
."
o
en
-< s: m
N
52
H. WOELK AND R. JAHRREISS
of neuronal- and glial cell-enriched fractions, Woelk et al. (12) presented experimental evidence that norepinephrine, injected into the ~ateral ventricle of rabbit brain, increased the incorporation of 3 Pinto phosphatidylinositol of both glial and neuronal cell bodies, but had no effect on the specific radioactivity of glial and neuronal ethanolamine plasmalogen and sphingo~elin. The labelling of phosphatidylcholine was slightly inhibited by norepinephrine in the glial cell-enriched fraction, whereas with neurons no inhibition could be observed (12). On the basis of these observations it was concluded that increased incorporation of the label into phosphatidic acid and phosphatidylinositol may occur in neurons which respond to the neurotransmitter with excitation, whereas inhibition of 32p incorporation into phosphatidylcholine may be connected with inhibitory transmission (12). REFERENCES 1.
2.
3.
4. 5.
6.
8. 9. 10.
Blank, M.L., Wykle, R.L. and Snyder, F., Enzymic synthesis of ethanolamine plasmalogens from an O-alkyl glycerolipid, FEBS Lett. 18 (1971) 92-94. Blomstrand, C. and Hamberger, A., Protein turnover in cellenriched fractions from rabbit brain, J. Neurochem. 16 (1969) 1401-1407. Blomstrand, C. and Hamberger, A., Amino acid incorporation in vitro into protein of neuronal and glial cell-enriched -fractions, J. Neurochem. 17 (1970) 1187-1195. Dawson, R.M.C., A hydrolytic procedure for the identification and estimation of individual phospholipids in biological samples, Biochem. J. 75 (1960) 45-53. Freysz, L., Bieth, R. and Mandel, P., Kinetics of the biosynthesis of phospholipids in neurons and glial-cells isolated from rat brain cortex, J. Neurochem. 16 (1969) 1417-1424. Goracci, G., Blomstrand, C., Arienti, G., Hamberger, A. and Porcellati, G., Base-exchange enzymic system for the synthesis of phospholipids in neuronal and glial cells and their subfractions: a possible marker for neuronal membranes, J. Neurochem. 20 (1973) 1167-1180. Oshino, N. and Omura, T., Immunochemical evidence for the participation of cytochrome b5 in microsomal stearoyl-CoA desaturation reaction, Arch. Biochem. Biophys.157(1973) 395-404. Paltauf, F., Prough, R.A., Masters, B.S.S. and Johnston, J.M., Evidence for the participation of cytochrome b S in plasmalogen biosynthesis, J. Biol. Chem. 249 (1974) 2661-2662. Poduslo, S.E. and Norton, W.T., Isolation and some chemical properties of oligodendroglia from calf brain, J. Neurochem. 19 (1972) 727-736. Snyder, F., Blank, M.L. and Wykle, R.L., The enzymic synthesis of ethanolamine plasmalogens, J. Biol. Chem. 246 (1971) 36393645.
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
11.
12.
13.
53
Woelk, H., Goracci, G., Gaiti, A., Porcellati, G., Phospholipase A1 and A activities of neuronal and glial cells of the rabbit brain, ~oppe Seyler's Z. Physiol. Chern. 354 (1973) 729-736. Woelk, H., Kanig, K., Peiler-Ichikawa, K., Phospholipid metabolism in experimental allergic encephalomyelitis: Activity of mitochondrial phospholipase A2 of rat brain towards specifically labelled 1,2-diacyl-, 1-alk-1'-enyl-2-acyl- and 1-alkyl2-acyl-sn-glycero-)-phosphorylcholine, J. Neurochem. 23 (1974) 745-750. Woelk, H., Porcellati, G., Subcellular distribution and kinetic properties of rat brain phospholipases A1 and A2 . ~ Seyler's Z. Physiol. Chern. 354 (1973) 90-100.
Helmut Woelk and Rosemarie Jahrreiss Einheit fUr Neurobiochemie der UniversitatsNervenklinik Erlangen, 852 Erlangen, and Universitats-Nervenklinik der Universitat des Saarlandes 665 Homburg/Saar, G.F.R. SUMMARY The formation of ethanolamine plasmalogen from labelled 1-alkyl-2-acyl-sn-glycero-3-phosphorylethanolamine was st udi ed in neurons and glial cells of the developing rat brain. It was found that the conversion of the ether to the enol-ether bond of the 1-alkyl moiety by the neuronal and glial desaturase system requires unsaturated fatty acids at the 2 position of the substrate. There is almost no difference between the activity of the neuronal and glial desaturase during the period of active myelination, whereas the neuronal cell fraction of the adult rats displays a threefold higher enzyme activity as compared to the glial cells. Evidence for the involvement of a microsomal electron transport system in the enzymic conversion of alkylacyl-glycero-3phosphorylethanolamine to ethanolamine plasmalogen was obtained by using specific antibodies against NADH-cytochrome b s reductase. Cytochrome b 5 stimulated the biosynthesis of ethanolamine plasmalogen.
Abbreviations used: GPC, sn-glycero-3-phosphorylcholine; GPE, sn-glycero-3-phosphorylethanolamine; GPS, sn-glycero-3-phosphorylserine.
43
J. Palo (ed.), Myelination and Demyelination © Plenum Press, New York 1978
H. WOELK AND R. JAHRREISS
INTRODUCTION Experimental evidence has been forwarded during the last years for marked differences in phospholipid metabolism between glialand neuronal cell-enriched fractions. The neuronal cell bodies were found to possess a much higher rate of base-exchange for both serine and ethanolamine than the glial cell-enriched fraction (6) and Freysz et al. (5) showed in their extensive investigation on the kinetics of the biosynthesis of phospholipids in neurons and glial cells, isolated from rat brain cortex, that neuronal phospholipids had a faster turnover than glial phospholipids. Recently, we obtained indirect evidence for a faster turnover of glycerophosphatides in neurons than in glial cells, since neurons contained a considerably higher phospholipase A1 and A2 activity when compared to the glial cell-enriched fraction (11). There is only little information on the mechanism. of the biosynthesis of ethanolamine plasmalogen (1-alk-1'-enyl-2-acylsn-glycero-3-phosphorylethanolamine) by brain cells. In the brain tissue ethanolamine plasmalogen reaches particularly high values accounting for approximately 80% of total ethanolamine phosphatides in the myelin sheath. Investigation of plasmalogen biosynthesis in intact cells has led to conflicting opinions regarding the aliphatic precursor of the alkenyl moiety. The postmitochondrial fraction of Ehrlich ascites cells and preputial gland tumors can synthesize ethanolamine plasmalogens from the long chain fatty alcohols and dihydroxyacetone phosphate (10) or l-alkyl-2-acylsn-glycero-3-phosphate (1). It appears to be well established that one mechanism by which l-alk-1 '-enyl-2-acyl-sn-glycero-3phosphorylethanolamine can be formed involves dehydrogenation of the 1-alkyl moiety and that cytochrome b 5 participates into the reaction (8). EXPERIMENTAL Preparation of glia and neurons from rat and rabbit brain. The neuronal and glial fractions were prepared from young and adult Wistar rats and from white rabbits weighing about 1,5 kg. The animals were anaesthetized with sodium pentobarbitone and killed by intracardiac perfusion of Ringer solution. The whole brain was removed quickly, weighed and sliced. The procedure employed for the preparation and identification of neurons and glia was essentially by the methods of Blomstrand and Hamberger (2, 3) and Goracci et al. (6) as indicated elsewhere (11). The~onal fraction contained more than 90% of neuronal cell bodies, freed for the greater part of their axonal processes and with very little contamination by glial cells. The glial cellenriched fraction contained about 90% glial cells. Endothelial cells and free nuclei were the main contaminants of both cell
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
45
fractions. The purity of the neuronal and glial cell suspensions was further assessed by using the base-exchange enzymic system as a marker for the neuronal cell bodies (6). Using serine and ethanolamine as nitrogeneous bases, neuronal/glial ratios of about 3 to 4 were found for the base-exchange activity. Oligodendroglia were obtained by the method of Poduslo and Norton (9). Subcellular fractionation of the cell-enriched preparations. Neuronal and glial microsomes were obtained as described in detail by Goracci et al. (6). The microsomes displayed 88% of the glucose6-phosphatase activity and 91% of the NADPH:cytochrome c oxidoreductase activity of t~~ whole cell homogenate. Substrates. 1-[ C]alkyl-2-acyl-GPE (substrate A) was prepared fro~4the brain phospholipids after intracerebral injection of 1-[ C}hexadecanol (52 ~Ci/~ole/animal) into 14 day-old-rats. Twenty-four hours after the administration of the labelled alcohol, the animals were sacrificed and the brain ethanolamine phosphoglycerides purified. In order to remove the 1-alk-1'-enyl portion, the ethanolamine phosphoglycerides were subjected to mild acid hydrolysis according to a modified Dawson~s procedure (4). 1,2Diacyl-GPE was removed with the use of purified lipase from porcine pancreas. The method is based on the selective deacylation by lipase action at the 1 position of the 1,2-diacyl compounds of naturally occurring phosphatide mixtures, containing 1,2-diacyl-, alkenyl-acyl-,1l:nd alkylacyl-glycerophosphatides. The final preparation of 1[ Clalkyl-2-acyl-GPE (specific activity of 4,2 ~Ci/ ~ole) was obtained by silicic acid or Florisil column chromatograph~14
L C}-labelled 1-alkyl-2-acyl-GPE (substrate B) was ~~epared from 14-day-old rats after intracerebral injection of 1-[ C ] acetate. Twenty-four hours after the administration of the labelled precursor the rats were sacrificed, the brain lipids extracted and the ethanolamine phosphoglycerides purified. Th r4 1-alk-1'-enyl-2acyl portion (ethanolamine plasmalogen) of the [ C] -labelled ethanolamine phosphoglycerides was transformed lnto the corresponding acylalkyl-compound by catalytic hydrogenation. In order to remove the the remaining plasmalogens, the ethanolamine phosphoglycerides were subjected to mild acid hydrolysis according to a modified Dawson~s procedure (4). The phosphoglyceride mixture, containing 1,2 diacyland 1-alkyl-2-acyl-GPE, was purified by means of column chromatography o~4Florisil. 1- L C]alkyl-GPE was obtained by silicic acid column chroma togra~uy after removing the fatty acids from the 2 position of 1- [ c] alkyl-2-acyl-GPE with phospholipase A2 from Naja naja venom. The labelled 1,2-diacyl-, 2-acyl-1-alk-1'-enyl- and 2-acyl-1-alkyl glycerophosphatides, used for the measurement of the phosphollpase A activity, were prepared and checked as described previously (13). 2 . . . . . . HydrolYS1S of the phosphoglycerldes, WhlCh dlffered ln the radlcal at the 1 position and were labelled at the 2 position with different fatty acids, by phospholipase A2 from naja naja venom, showed that the radioactivity was almost exclusively recovered in the fatty
46
H. WOELK AND R. JAHRREISS
acids freed from the substrates, indicating that specific incorporation of the labelled fatty acids into the 2 position of the phosphoglycerides had occurred. Hydrolysis of the 1,2-diacyl-glycerophosphatides, labelled at the 1 position, by phospholipase A2 from crotalus atrox venom, showed that 94-96% of the radioactivity was recovered in the lyso-compounds, indicating that specific incorporation of the labelled fatty acids into the 1 position of the lyso derivatives had occurred.
RESULTS The formation of ethanolamine plasmalogen from 1-l14c] alkyl(prepared ~,~m brain tissue after injection of 1- L Clhexadecanol) and L cj-labelled 1-al~~1-2-acYl-GPE (prepared trom brain tissue after injection of [ Cl acetate and catalytic hydrogenation of the plasmalogen portion) has ieen studied in neurons and glial cells of the d1~eloping rat brain (Table 1). As can be seen from Table 1, 1- [ c l alkyl-2-acyl-GPE is a much better substrate for the formation of ethanolamine plasmalogen by the neuronal and glial desaturation system than the corresponding hydrogenated glycerophospatide. It appears that the conversion of the ether to the enol-ether bond of the 1-alkyl moiety by the neuronal and glial desaturation system requires unsaturated fatty acids at the 2 position of the substrate. Furthermore, there is almost no difference between the activity of the neuronal and glial desaturase during the period of active myelination, whereas the neuronal cell fraction of the adult rats displays a threefold higher enzyme activity as compared to the glial cells. With increasing age of the animals a decrease in the desaturase activity can be observed. The decrease is much more pronounced in glia than in neurons (Table 1). The dehydrogenation of the 1-alkyl moiety requires a reduced pyridine nucleotide (NADH or NADPH) and is strongly inhibited by KCN, but not by CO/0 2 . Further evidence for the involvement of a mi~psomal electron transport system in the enzymic conversion of 1-L14 C]alkyl-2-acyl-GPE to ethanolamine plasmalogen was obtained by using specific antibodies against NADH-cytochrome b5 reductase. Table 2 clearly demonstrates the inhibition of plasmalogen formation by adding increasing amounts of antibody against NADH-cytochrome b 5 reductase to the incubation system. On the other hand, addition of cytochrome b5 stimulated the biosynthesis of ethanolamine plasmalogen from the corresponding alkyl-ether by oligodendroglial microsomes (Table 3). In order to investigate whether Piracetam (2-oxo-pyrrolidine-1-acetamide) has some effect on the biosynthesis of ethanolamine plasmalogen, the animals were injected i.p. for 10 days with increasing amounts of the nootropic substance. Table 4 shows that the effect of Piracetam on the formation of ethanolamine plasmalogen resembles the action of cytochrome b5 on the dehydrogenation of the 1-alkyl moiety as shown in Table 3. 2-aA~1-sn-GPE
- - - -
2.11 1.23 0.91
Neurons
------
2.05 0.84 0.36
Glia
----
1.03 1.46 2.53
Neuronal/ glial ratio 0.31 0.28 0.12
Neurons
0.33 0.22 0.05
Glia
B
0.94 1.27 2.40
Neuronal/ glial ratio
Each incubation mixture c~ntained in a total volume of 0.25 ml 0.1 ~ Tris/HCl buffer, pH 7.1: 15 nmoles of 1-r1 Clalkyl-2-acyl-sn-GPE (substrate A) or ~ C1-labelled 1-alkyl-2acyl-sn-GPE (hydrogenate~ s~strate, B), NaF (12 mM), ATP (10 mM), NArlP+ (2 mM). glucose-6-P (6 mM), MgC12 (4 mM). glucose-6-P-dehydrogenase (0.41 U), GSH (2 mM) and 325 to 480~g of glial or neuronal protein. The substrates were added as a sonicated dispersion in 0.15% Tween 80. Incubations were carried out for 60 min at 37°C_~n a shaking water bath. Values are expressed as nmol ethanolamine-plasmalogen formed x mg prot. x h- 1 •
14 days 21 days Adult
Age
A
Table 1. Formation of ethanolamine-plasmalogen from the corresponding 1-alky.1-2-acyl-snglycero-3-phosphorylethanolamine in neurons and glial cells of the rat brain during myelination.
~
oz
~
Z
r
-< m
s::
G)
Z
C :D
o
en
r
r
m
(')
» r
o z
C :D
m
Z
z o
»
r
5>
G)
r
o"T1
en
-< s:: m
N
z
m
2.0 4.0 8.0 10.0
-
(mg)
Control globulin
0.80 0.37 0.22
1.42
o .9A 31.3 43.8 74.5 82.2
1.26 1.13 1.10 1.14
10.8 20.4 22.5 19.7
I
4.0 8.0 16.0 20.0
Ethanolamine-Plasmalogen formed in the presence of Antibody Inhibition Control Inhibition Ratio globulin globulin antibody-micro somal protein (nmol) (nmol) (% ) (%)
Each incubation mixture 9Rntained in a total volume of 1.0 ml 0.1 M Tris/HCl buffer, pH 7.1: 60 nmoles of 1-[ Clalk.yl-2-acyl-GPE, NaF (12 mM), ATP (10 mM), NADH (2 mM), glucose-6-P (6 mM), MgCl2 (4 mM), glucose-6-P-dehydrogenase (1.64 U), GSH (2 mM), 0., ml of the 100,000 x g supernatant, 498 ~g microsomal protein and antibody or control globulin as indicated. The substrates were added as a sonicated dispersion in 0.15% Tween 80. Incubations were carried out for 60 min at 37°C in a shaking water bath.
2.0 4.0 8.0 10.0
-
(mg)
Antibody against NADH-cytochrome b5 reductase
Table 2. Influence~n fOti NADH-cytochrome b5 reductase on the formation of ethanolamineplasmalogen from l-L' Cjalkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by oligodendroglial microsomes of 20-d~-old rats.
~
m
::0 ::0
'-
» J:
::0
c
»z
~
r-
m
;t:
:e o
t
49
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
Both substances enhance the formation of ethanolamine plasmalogen from 1-[1 C] alkyl-2-acyl-GPE (Tables 3 and 4). In another series of experiments exogenously added glycerophosphatides, specifically labelled either in the 1 or in the 2 position, were used to measure the activity of phospholipase A2 from the neuronal-enriched cell fraction of the rabbit. As can be seen from Table 5 the enzyme hydrolysed the 1,2-diacylglycerophosphatides more rapidly than the acyl alkyl- and acylalkenyl-compounds. Choline plasmalogen and the corresponding alkyl-derivative were cleaved at almost similar rates by the phospholipase A2 . Among the various 1,2-diacyl-glycerophosphatides, the neuronal phospholipase A2 preferred phosphatidylcholine as a substrate, whereas phosphatidylserine was hydrolysed less actively. Table 5 shows, furthermore, that norepinephrine, injected into the lateral ventricle of the rabbit brain, stimulated the hydrolysis of the various glycerophosphatide substrates.
Table 3. Effect ofc~tfchrome b 5 on the formation of ethanolamineplasmalogen from 1-[1 Cjalkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by oligodendroglial microsomes of 20-day-old rats. Addition
Control IG Cytochrome b5
Concentration (mg/mg of micro protein)
Ethanolami nePlasmalogen formed (nmol/mg protein/h)
1.8 0.5 1.2
3.6
4.5
Incubation conditions as in Table 2.
2.94
2.78 3.85
4.49
5.67 6.23
Difference
C%)
6
+. 31 + 53 + 93 + 112
50
H. WOELK AND R. JAHRREISS
Table 4. Influence Of Piracetam on the formation of ethanolamineplasmalogen from 1-[14C] alkyl-2-acyl-sn-glycero-3-phosphorylethanolamine by neuronal microsomes of 20-d~-old rats. Piracetam (mg/kg/day)
60
80 100 120 140
Ethanolamine-plasmalogen formed (nmol/mg protein/h)
2.96
Increase
(%)
3.87
30.7 54.0
6.02 6.12
103.3 108.7
4.56 5.76
94.6
The incubation conditions were as described ln Table 2 with the exception that instead of oligondendroglial microsomes neuronal microsomal protein was used. The animals were pretreated for 10 days with the dose of Piracetam.
DISCUSSION The present results have shown that the microsomal 1~alkyl-2acyl-GPE desaturation reaction is catalyzed by a membrane-bound multi component desaturase system requiring the supply of reducing equivalents from pyridine nucleotides. It may be supposed that in the desaturase reaction, which transforms the alkyl-ether to the enol-ether moiety, a cyanide sensitive factor (CSF) is operating as the terminal oxidase as it has been proposed for the oxidative conversion of stearoyl-CoA to oleoyl-CoA (7). The use of various specific inhibitors including antisera against NADH-cytochrome b5 reductase indicate the role of microsomal flavoproteins in the supply of reducing equivalents from NADPH or NADH to the cyanide sensitive factor. Cytochrome b5 seems to be functional as an electron carrier between the flavoproteins and the CSF. 1-alkyl-2-acyl-GPE with unsaturated fatty acids at the 2 position served as a good substrate for plasmalogen synthesis, whereas 1-alkyl-GPE was transformed to the corresponding plasmalogen only after the addition of CoA to the incubation system, indicating that the dehydrogenation of the 1.l..alkyl moiety only occurs on the intact phospholipid molecule. The observation that norepinephrine stimulates the hydrolysis of glycerophosphatides by phospholipase A2 (Table 5) could suggest that phospholipase A2 might be concerned in the molecular changes taking place in membranes during synaptic transmission. In an investigation on the incorporation of 3 2p into the phospholipids
33.2 27.8 27.2
20.3 18.5
17.8
73.9 61.3
52.8
50.2
63.5
54.7
51.7
(% )
Dlfference
.
(b)
~g of L-norepinephrine (for details,
(b)Per cent of increase in activity of norepinephrine experiments, as compared to controls.
(a)The animals were injected intraventricularly with 100 see Experimental.
Each incubation mixture contained in a total volume of 1.0 ml: 0.1 M sodium acetate buffer, pH 5.4; 1.0 ~mol of the glycerophosphatide substrate; 3 mg of sodium taurocholate and varying amounts of neuronal protein (ranging from 248 to 524 ~g). Incubation at 37°C for 1 h. For details, see Experimental. Each figure represents the average of five experiments; SEM was less than 8%.
1-[14cJstearOYl-2-aCY1-~-GPS 1-Alk-1'-enYl-2-[14cJlinOleOY1-~-GPC 1-AlkYl-2-[14c]linOleOY1-~-GPC
39.6
1-[14C]stearOYl-2-aCY1-Sn-GPE
[nmol/(mg x h)] 48.7
[nmol/(mg x h)]
Phospholipase A2 phospholipase A2
. . (a) Noreplnephrlne
Hydrolysis of different glycerophosphatides by neuronal phospholipase A2 .
1-[14c]stearOYl-2-aCY1-~-GPC
Substrates
Table 5.
m Z
~
z r (') m r r en o
o
C :0
m
Z
o
Z
l>
» r
r
Cl
."
o
en
-< s: m
N
52
H. WOELK AND R. JAHRREISS
of neuronal- and glial cell-enriched fractions, Woelk et al. (12) presented experimental evidence that norepinephrine, injected into the ~ateral ventricle of rabbit brain, increased the incorporation of 3 Pinto phosphatidylinositol of both glial and neuronal cell bodies, but had no effect on the specific radioactivity of glial and neuronal ethanolamine plasmalogen and sphingo~elin. The labelling of phosphatidylcholine was slightly inhibited by norepinephrine in the glial cell-enriched fraction, whereas with neurons no inhibition could be observed (12). On the basis of these observations it was concluded that increased incorporation of the label into phosphatidic acid and phosphatidylinositol may occur in neurons which respond to the neurotransmitter with excitation, whereas inhibition of 32p incorporation into phosphatidylcholine may be connected with inhibitory transmission (12). REFERENCES 1.
2.
3.
4. 5.
6.
8. 9. 10.
Blank, M.L., Wykle, R.L. and Snyder, F., Enzymic synthesis of ethanolamine plasmalogens from an O-alkyl glycerolipid, FEBS Lett. 18 (1971) 92-94. Blomstrand, C. and Hamberger, A., Protein turnover in cellenriched fractions from rabbit brain, J. Neurochem. 16 (1969) 1401-1407. Blomstrand, C. and Hamberger, A., Amino acid incorporation in vitro into protein of neuronal and glial cell-enriched -fractions, J. Neurochem. 17 (1970) 1187-1195. Dawson, R.M.C., A hydrolytic procedure for the identification and estimation of individual phospholipids in biological samples, Biochem. J. 75 (1960) 45-53. Freysz, L., Bieth, R. and Mandel, P., Kinetics of the biosynthesis of phospholipids in neurons and glial-cells isolated from rat brain cortex, J. Neurochem. 16 (1969) 1417-1424. Goracci, G., Blomstrand, C., Arienti, G., Hamberger, A. and Porcellati, G., Base-exchange enzymic system for the synthesis of phospholipids in neuronal and glial cells and their subfractions: a possible marker for neuronal membranes, J. Neurochem. 20 (1973) 1167-1180. Oshino, N. and Omura, T., Immunochemical evidence for the participation of cytochrome b5 in microsomal stearoyl-CoA desaturation reaction, Arch. Biochem. Biophys.157(1973) 395-404. Paltauf, F., Prough, R.A., Masters, B.S.S. and Johnston, J.M., Evidence for the participation of cytochrome b S in plasmalogen biosynthesis, J. Biol. Chem. 249 (1974) 2661-2662. Poduslo, S.E. and Norton, W.T., Isolation and some chemical properties of oligodendroglia from calf brain, J. Neurochem. 19 (1972) 727-736. Snyder, F., Blank, M.L. and Wykle, R.L., The enzymic synthesis of ethanolamine plasmalogens, J. Biol. Chem. 246 (1971) 36393645.
ENZYMES OF GLIAL AND NEURONAL CELLS DURING MYELINATION
11.
12.
13.
53
Woelk, H., Goracci, G., Gaiti, A., Porcellati, G., Phospholipase A1 and A activities of neuronal and glial cells of the rabbit brain, ~oppe Seyler's Z. Physiol. Chern. 354 (1973) 729-736. Woelk, H., Kanig, K., Peiler-Ichikawa, K., Phospholipid metabolism in experimental allergic encephalomyelitis: Activity of mitochondrial phospholipase A2 of rat brain towards specifically labelled 1,2-diacyl-, 1-alk-1'-enyl-2-acyl- and 1-alkyl2-acyl-sn-glycero-)-phosphorylcholine, J. Neurochem. 23 (1974) 745-750. Woelk, H., Porcellati, G., Subcellular distribution and kinetic properties of rat brain phospholipases A1 and A2 . ~ Seyler's Z. Physiol. Chern. 354 (1973) 90-100.
