Contribution of axonal transport to the renewal of myelin phospholipids in peripheral nerves. II. Biochemical study

Brain Research, 219 (1981) 73-84

73

Elsevier/North-HollandBiomedicalPress

CONTRIBUTION OF AXONAL TRANSPORT TO THE RENEWAL OF MYELIN PHOSPHOLIPIDS IN PERIPHERAL NERVES. II. BIOCHEMICAL STUDY

MARINA BRUNETTI, LUIGI DI GIAMBERARDINO, GIUSEPPE PORCELLATIand BERNARD DROZ Istituto di Chimica Biologica, Universitd di Perugia, CP no. 3, 06100 Perugia (Italy) and D~partement de Biologic, Commissariat d l'Energie Atomique, CEN Saclay, 91191 Gif-sur- Yvette Cedex (France)

(Accepted January 1st, 1981) Key words: neuron-glia-- phospholipid-- myelin-- axonal transport

SUMMARY The classes of radioactive phospholipids appearing in the ciliary ganglion (CG) and especially in the myelin sheath of the intraorbital part of the oculomotor nerve (OMN) were determined after the intracerebral injection of [2-all]glycerol and [methyl-laC]choline to chickens. Analysis of the radioactive compounds in water-soluble fractions and chloroform-methanol extracts was performed by thin-layer chromatography (TLC). The water-soluble content of the OMN and CG was much poorer in [2-all]glycerol and metabolites than in [methyl-14C]choline and derivatives. All classes of glycerophospholipids were found to be axonaUy transported along the OMN and into the CG, but choline-phosphoglycerides (CPG) were largely predominant. In myelin fractions from the OMN, the specific radioactivity (SRA) of CPG labeled with [2-3H]glycerol reached a maximum earlier (40 h) than the SRA of CPG labeled with [methyl-14C]choline. A 25-fold enhancement of the [14C]SRA of sphingomyelin (SM) was observed between I2 h and 7 days. These results indicate that: (1) axonally transported phospholipids labeled with [2-all]glycerol consist mainly of CPG; (2) small amounts of CPG are translocated from the axon to myelin; and (3) the progressive enrichment of myelin in [14C]CPG and, to a greater extent, SM draws attention to the importance of the base recycling for local synthesis of myelin phospholipids. Thus the axonal supply of Schwann cells with choline and the transfer of axonal phospholipids to myelin would probably contribute to the metabolic interdependence existing between neuron and glia.

0006-8993/81/0000-0000/$2.50 © Elsevier/North-HollandBiomedicalPress

74 INTRODUCTION In the course of the axonal transport of labeled phospholipids along peripheral nerves, radioautography has shown that phospholipid components are translocated from the axons to their encompassing myelin sheath6-S,lL The aim of the present investigation is to specify the classes of labeled phospholipids which are axonally transported along the nerve fibres to the axon terminals and those which are eventually recovered in myelin fractions of the same nerve. After the intracerebral injection of [2-3H]glycerol and [methyl-t4C]choline to chickens, the different classes of radioactive phospholipids which are axonally transported along the oculomotor nerve (OMN) were determined: (1) in the ciliary ganglion (CG) in which they are accumulated in nerve endings; (2) in the orbital segment of the OMN, in which they are in transit; and (3) in myelin fractions isolated from this OMN portion. MATERIALS AND METHODS Two-week-old chickens (Leghorn, I.N.R.A., Nouzilly), lightly anesthetized with chloroform, were given a simultaneous injection of 800 #Ci of [2-ZH]glycerol (spec. act. 9.57 Ci/mmol, N.E.N.) and 70 /zCi of [methyl-14C]choline (spec. act. 46.0 mCi/mmol, N.E.N.) into the cerebral aqueduct. At various time intervals after the intracerebral injection of both [2-3H]glycerol and [methyl-14C]choline, 2 or 4 series of 5 chickens were sacrificed by decapitation. The CG and orbital segments of the OMN were excised. The CG were stored at ---60 °C until use. The segment of the OMN, which was chosen for myelin analysis, was located in the orbital cavity at the same distance from the cell bodies as the ciliary ganglion. The orbital segments of the O M N were immediately homogenized in 0.32 M sucrose solution. An aliquot was utilized for lipid analysis of the total nerve homogenate while another was used for the preparation of the myelin fraction. Analysis of lipids and water-soluble compounds in the CG. Ten pooled CG for each series were homogenized with 2 ml of chloroform-methanol (2:1, v/v) and centrifuged at 5000 rpm for 15 rain. 1 ml of water was added to the organic extract and the two phases separated by centrifugation. The water-methanol phase was collected and saved, while the chloroform phase was washed twice with 0.6 ml of methanol-water (1:1, v/v). The chloroform, which contained the lipid material, was collected and stored in the dark under Ne and at --20 °C until use. The lipid-free pellet was rehomogenized with 0.5 ml of ethanol-water (1:4, v/v) and centrifuged. The procedure was repeated 3 times. The combined water-ethanol extracts were added to the pooled original upper water-methanol phase from the lipid extraction (see above) for further analyses. The pellet was finally dissolved in 0.1 M N a O H and used for protein determination. The methanol-chloroform extract was dried under N2 and the lipids resuspended in known volumes of chloroform-methanol (2 : 1, v/v); an aliquot was utilized for the direct counting of the total radioactivity and the remainder was divided into two parts and separately chromatographed by using the two-dimensional thin-layer chromatography (TLC) procedure according to Horrocks and Sun 15. One plate was

75 used for the determination of radioactivity of the different separated phospholipids and the other for the assay of the phospholipid P content. The aqueous phases were evaporated to complete dryness to avoid the presence of tritiated water, which might eventually have arisen from the oxidation of glycerol-3phosphate to dihydroxyacetone phosphate, and then dissolved in small volumes of distilled water. An aliquot of the hydrosoluble extract was utilized for the direct counting of the total radioactivity, while the remainder was chromatographed according to Yavin 8°. Isolation o.fmyelin. The Norton and Poduslo procedure 20 for the preparation of myelin fractions was adapted to small samples. Ten nerve segments for each experiment (about 6 mg of fresh weight) were pooled for each preparation of myelin. The nerves were homogenized with 19 vols. of cold 0.32 M sucrose and the crude myelin collected at the interface between 0.32 M and 0.85 M sucrose following centrifugation in a Beckman SW 50.1 rotor at 75,000 g for 30 rain (total volume of each tube: 700 #1). The myelin fraction was diluted with ice-cold water and centrifuged at 75,000 g for 15 min. The pellet was resuspended in distilled water and washed twice by centrifugation at 12,000 g for 10 min. The osmotically shocked myelin was resuspended in 0.32 M sucrose and recycled through the discontinuous gradient step, as above. The purified myelin was collected from the gradient, diluted with water and collected by centrifugation at 75,000 g for 15 min. The myelin membrane were finally resuspended in a small amount of water and completely dissolved in few volumes of chloroform-methanol (1:2, v/v). Myelin purity. The enrichment of the isolated fraction was verified both by gel electrophoresis and electron microscopy. The gel electrophoresis displayed the characteristic protein pattern of purified myelin. Fig. 1 shows electron micrographs of the isolated fraction. Enzyme analysis could not be performed, owing to the low amount recovered in this fraction.

Fig. 1. a and b: Electron microscopic appearance of myelin fractions prepared from the orbital portion of the OMN of chicken.

76

3H

3H

180

180

(a)

140 100

(b)

160 10(]

60

a.

2 0 ~ _

2(~

t 12h

o

ml 4Oh

12h

7(1

4Oh

7d

E

,4C

'4C

40

(a)

(b)

30 T

20

20

10

1

12h

m

4Oh

i

71:1

12h

40h

7d

Fig. 2. Levels of radioactivity (dpm//~g phospholipid P) found in the CG (a) and the OMN (b) at various time intervals (12 h; 40 h; 7 days) after the administration of [2-3H]glycerol [3H] and [methyl14C]choline [lac] into the cerebral aqueduct of young chickens. Values for CG are shown on the left (a) and those for OMN on the right (b). White bars, radioactivity of the lipid phase; dashed bars, radioactivity of the aqueous phase. Mean data from 2-4 experiments, each carried out with 10 pooled CG or OMN, with deviation values.

Lipid analysis. 1.5 ml-of c h l o r o f o r m - m e t h a n o l

(1:2, v/v) were added to the nerve homogenate and myelin, and the extracts were left for 3 h at 4 °C. The tubes were then centrifuged at 50,00 r p m for 15 min. C h l o r o f o r m and water were added to the organic extract to give a final ratio o f c h l o r o f o r m - m e t h a n o l - w a t e r equal 8:4:3 (v/v/v) and the ensuing layers were separated 30. The upper methanol-water layer was dried under N2 directly into counting vials and counted for total radioactivity content. The lower chloroform layers, dried under N2, were resuspended in small volumes of chlorof o r m - m e t h a n o l (2:1, v/v), and aliquots utilized for the determination of the total radioactivity and total phospholipid P content. The remainder was subjected to TLC, and the various lipids separated by two-dimensional T L C 15.

77

Other analyses. Protein was determined according to Lowry et al. 17. Phospholipid P was assayed following the procedure of Ernster et alP. Couting was carried out in Insta-gel (Packard, Ziirich) by liquid scintillation spectrometry with an Intertechnique scintillation spectrometer. Counting efficiency was computed from the external standard ratio index calibrated with increasing quenched vials containing 3H and 14C standards (Laboratoire de Metrologie des Rayonnements Ionisants, C.E.N. de Saclay, Gif-sur-Yvette, France). TLC of lipids and water-soluble compounds was carried out on precoated Polygram Sil.G plates (0.25 mm thickness, Macherey Nagel). RESULTS

Ciliary ganglion The phospholipid composition of the C G showed the presence of all main classes of phospholipids. Ethanolamine-phosphoglycerides (EPC: 35.9 ~o) and choline-phosphoglycerides (CPG: 30.5 ~o) are more abundant than serine-phosphoglycerides (SPG: 14.3 ~), inositol-phosphoglycerides (IPG: 11.4 ~o) or sphingomyelin (SM: 7.8 ~o)- The relative proportion of diacyl-, alkenylacyl- and alkyl-derivatives was not determined. After the intracerebral injection of [2-3H]glycerol (Fig. 2a, upper part), the [SH]label was recovered in all classes of phosphoglycerides extracted from the CG, but with striking differences of labeling (Fig. 3). At 12 h, large amounts of [aH]-label were indeed incorporated into CPG (67 ~ ) whereas only minor parts were taken by E P G (22 ~ ) as well as SPG and IPG (11 ~o). Only traces of [SH]-labeled lysophosphatidylcholine were identified. Seven days after injection, the proportion of [aH]-label in CPG decreased relative to that of EPG and SPC.

UC-Choline

3H-Glycerol

1OO

75 -[= 5C

2~

SM

CPG

CPG

EPG

SPG

r~

IPG

Fig. 3. Percentage distribution of labeled phospholipid classes in the CG at 3 time intervals after the intracerebral injection of [2-all]glycerol (on the right) and [methyl-zaC]choline(on the left). Each of the 3 adjacent columns represents, respectively,the per cent of label at 12 h, 40 h and 7 days after the injection. These values are calculated over total lipid radioactivity after TLC and represent means from 2--4 experiments with corresponding deviation data.

78 In the water-soluble fraction (Fig. 2a, upper part), the [3H]-radioactivity derived from the intracerebrally injected [2-3H]glycerol was extremely low at the 3 time intervals examined and never exceeded 4 ~ of the total [3H]-radioactivity8. Analysis of the [3HI-soluble compounds at early time intervals indicated that they consist of glycerol, glycerophosphorylcholine to a lesser extent, and only traces of glycerol-3phosphate. The hydrosoluble [14C]-radioactivity contributed to about 25 ~ of the total [14C]radioactivity at early time intervals in the CG (fig. 2a, lower part). A similar rise of the water-soluble radioactivity was also found after the intracerebral injection of [methyl3H]cholineS. TLC of the hydrosoluble molecules performed at time intervals close to the peak indicated that 50 ~ of the water-soluble [14C]-radioactivity was associated with free choline and possibly acetylcholine. A minor part of the t4C label was bound to phosphorylcholine and cytidine-5'-diphosphate choline (including possibly betaine) whereas [14C]glycero-phosphorylcholine culminated later. [Methyl-14C]choline-labeled phospholipids were found to consist of CPG for 95 ~ and of SM for only 5 ~ at 12 h (Fig. 3). Seven days later, SM represented more than 20 ~,, of the total [14C]-labeled lipids with a corresponding relative decrease of the part taken by the CPG. At any time, [14C]-labeled lysophosphatidylcholine never exceeded 2 ~. O c u l o m o t o r nerve

The water-soluble and the lipid-specific radioactivity (SRA) counted in the OMN was lower than that counted in the CG (Fig. 2b). The [2-3H]glycerol-labeled lipids displayed an evolution (Fig. 2a and b, upper part) and distribution throughout the different classes of phospholipids which are similar in the O M N (Fig. 4) and in the CG (Fig. 3). The high peak of [aH]SRA recorded at 40 h in CPG and, to a lesser degree, EPG was followed at 7 days by a fall to 20-25 ~ of the maximal value (Fig. 5).

14C-Choline

:Ill-Glycerol

10(2

i

75

50

25

/ SM

7

CPG

ClOG

E PG

SPG

I PG

Fig. 4. Percentage distribution of labeled phospholipid classes in the OMN at 3 time intervals alter the intracerebral injection of [2-3H]glycerol(on the right) and [methyl-14C]choline(on the left). The 3 adjacent columns represent respectively an interval of 12 h, 40 h and 7 days between injection and sacrifice.

79

3H-Glycerol

'4C-Chol.

f

i

~. 20

lo

i

SM

CPG

CPG

EPG

SPG

IPG

Fig. 5. Specific radioactivity (SRA) values, expressed as dpm//~g P, of labeled phospholipid classes in the OMN at 3 time intervals after the intracerebral injection of either [2-aH]glycerol (on the right) or [methylA4C]choline (on the left). Each of the adjacent columns to a time interval of 12 h, 40 h and 7 days. The increase o f [methyl-14C]choline-labeled lipids o b s e r v e d between 12 h a n d 7 d a y s (Fig. 2b, lower part) resulted in p a r t f r o m the d r a m a t i c increase o f [14C]-labeled S M (Fig. 5). C o n t r a r y to w h a t was observed with 8H, the p e a k o f [14C]SRA at 40 h was n o t followed by a clear decline at 7 d a y s (Fig. 5). F u r t h e r m o r e , d u r i n g this period, the [14C]SRA o f S M was m o r e t h a n 10 times enhanced.

Myelin fractions As in the whole nerve, all the classes o f [2-3H]glycerol-labeled p h o s p h o l i p i d s were represented in the myelin fractions (Fig. 6). T h e relative p r o p o r t i o n o f [SH]-

laC'Choline

% lOO~

3H-Glycerol

50

SM

CPG

CPG

EPG

SPG

I PG

Fig. 6. Percentage distribution of labeled phospholipid classes in the myelin fraction isolated from the OMN at various time intervals after the intracerebral injection of either [2-3H]glycerol (on the right) or [methyl-laC]choline (on the left). Each group of 3 adjacent columns represents the percentage of radioactive phospholipid recovered at 12 h, 40 h and 7 days after administration of the precursor.

80

140 "Ch°line !

1000 i

3H-Glycer-ol

I 750[i i o..

i

I f

250~-

F l SM

CPG

CPG

E PG

S PG

I PG

Fig. 7. Specificradioactivity (SRA) values, expressed as dpm//~g P, of labeled phospholipid classes in the myelin fraction isolated from the OMN at 3 time intervals after the intracerebral injection of either [2-3H]glycerol (on the right) or [methyl-14C]choline(on the left). Each of the 3 grouped columns corresponds to 12 h, 40 h and 7 days after injection. abeled CPG (60 %) remained unchanged between 12 h and 7 days whereas the relative content of [3H]-labeled SPG and EPG was respectively increased and decreased. The [3H]SRA of the phosphoglycerides in myelin fraction was lower than in the whole OMN but reached a maximum by 40 h; here again, the highest value was recorded in CPG (Fig. 7). At 7 days, the [3H]SRA of CPG and EPG decreased by 50 % of their maximal value; this fall was therefore less pronounced than that observed in the whole nerve. The [methyl-t4C]choline-labeled lipids displayed roughly a similar distribution in the myelin fraction (Fig. 6) and in the whole OMN (Fig. 4). Although the [t4C]-specific radioactivity was rather low, its evolution in CPG and SM was essentially characterized by a respective 3- and 25-fold increase between 12 h and 7 days. As distinct from the whole O M N (Fig. 5), the myelin fractions exhibited a late but definite enhancement of the [14C]SRA of CPG from 40 h to 7 days (Fig. 7). DISCUSSION

Classes of axonally tran~portedphospholipids CPG labeled with either [2-~H]glycerol or [methyl-t4C]choline is the most represented class of axonally transported phospholipids (Figs. 3 and 4). In spite of the fact that EPG are particularly abundant in chicken peripheral nerves, the low proportion of [3H]-labeled EPG results probably from the loss of [3HI-label during synthesis of ethanolamine-plasmalogens through the dihydroxyacetone-phosphate pathway19, 29. The gradual increase of the [14C]-labeled glycerophosphorylcholine, liberated by breakdown of CPG 25, as well as the following decrease of the labeled glycerophospholipids suggest a local catabolism of the axonally transported phospholipids. In CPG indeed, the faster decline of [3HI- than of [14C]SRA reflects that [2-3H]glycerol is lost whereas [14C]-labeled choline is reineorporated into new CPG. The rise of [14C]-label

81 observed in SM within a 7-day period (Fig. 5) could result from the local reincorporation of radioactive choline or from axonal transport 26. Although slow axonal transport of mitochondria might carry some labeled lipid (Tables I and II in ref. 8), these organelles are particularly poor in SM 22. Thus [14C]-labeled SM should derive mainly, if not exclusively, from local re-incorporation of recycled choline. Thus the re-utilization of labeled choline liberated from axonally transported CPG is consistent with the early arrival of fast transported lipids in the OMN and CG as well as with the rise of free tracer in the water-soluble fraction (Figs. 2a and b). The rapid base-exchange reaction1, a and the extensive re-incorporation of released choline after lipolysis19, 25 could account for the labeling of the Schwann cells and ganglion cell bodies. Under these conditions, radioactive choline reveals not only the axonal movement of labeled phospholipids but also a local re-utilization of the base by adjacent structures. In contrast, [2-all]glycerol, which is incorporated into newly formed phospholipids and is scarcely recycled, must be selected to demonstrate the kinetics of axonally transported phospholipids and their eventual transfer to other cells.

Neuronal contribution of phospholipid constituents to myelin The small but definite amounts of radioactive phospholipids found in myelin fractions of the OMN (Fig. 7) confirm that part of the axonally transported label is transferred to the myelin sheath in peripheral nerves6-8,12. The relatively low SRA of [3H]- and [14C]-labeled phospholipids in myelin fractions (Fig. 7) as compared with that of whole nerve homogenates (Fig. 5) is due to dilution of the label in the high lipid content of myelin. The radioautographic studies indeed reveal that the distribution of the label is initially restricted to certain myelin areas before eventually being redistributed throughout the myelin layers (Tables I and II and Figs 6, 7 of ref. 8). The transcellular passage of label from axon to myelinating Schwann cells might result from a translocation of whole lipid molecules or of a lipid moiety bearing the label. In the case of [2-SH]glycerol-labeled phospholipids, there is only a scarce re-utilization of [2-ZH]glycerol, at least in a manner which preserves the [3H]labellL Hence [2-ZH]glycerol-labeled phospholipids recovered in myelin should have been translocated from axon to myelin as entire molecules of phospholipids such as CPG which account for more than 60 ~. Once translocated into myelin, [2-3H]glycerol-labeled CPG as well as EPG and IPG show a pronounced decrease of their specific radioactivity between 40 h and 7 days (Fig. 7). This loss of label results mainly from the catabolism of labeled phospholipid in myelin 28. In the case of [methyl-14C]choline-labeled phospholipids, one part of the [14C]labeled CPG would also enter myelin from axon by a translocation mechanism of the entire phospholipid molecule, but another part is probably produced in the Schwann cells by local re-incorporation of radioactive choline into CPG. The gradual and net enrichment of CPG and especially SM in [14C]-label observed in myelin fractions (compare Figs. 5 and 7) points to an efficient mechanism of choline re-utilization in myelinated nerve fibres (Figs. 3 and 8-12 of the companion paperS). As proposed in brain 16, CPG could act as donor of radioactive choline for SM, although this pathway could not be demonstrated in vitro in isolated brain microsomes zT. Since the entry of

82 choline into peripheral nerves is limited by the blood-nerve barrier and the perineuri al sheath 11,14, re-utilization of choline by Schwann cells for biosynthesis of new myelin phospholipids would contribute to increase the thickness of the myelin sheath in young and to compensate for myelin catabolism in adult 2s.

Functional significance of the axon-myelin translocation of phospholipid constituents In the OMN system 6-8 as well as in the sciatic nerve 12 or the optic tract 13, axonally transported phospholipid constituents are exported from the axon to myelin sheath. Such an export could supply an exchangeable pool of myelin phospholipids with axonally transported CPG. CPG, which are mainly inserted into the outer leaflet of the axolemma, could indeed be more readily exchanged with inner myelin layers than EPG ~3, which are mainly inserted into the inner leaflet of the axolemma. Other membrane components of the axons such as proteins or glycoproteins s do not seem to contribute to the renewal of myelin in the peripheral 5 and central nervous systems2,13, is,z4, except for occasional recycling of the label 10. The amount o f C P G translocated from axon to myelin cannot be estimated from the present data. However, the translocation of phospholipids appears to be quantitatively a minor event as compared with the large amounts of myelin lipids synthesized by Schwann cells 11,14. Nevertheless this intercellular passage of CPG probably exerts a deep influence on a limited part of the phospholipid pool of myelin. The axon-myelin transfer of CPG could correspond to a series of equilibration processes4A3, 28 and give rise to continuous and reversible exchanges of lipids between the following membrane structures: axonal SER ~ axolemma ~ glial plasmalemma ~ loose myelin of the Schmidt-Lanterman clefts ~ compact myelin. Such an equilibration mechanism could account for a one-for-one exchange as well as for a net transfer of CPG to myelin depleted of phospholipid 4, possibly at the level of the Schmidt-Lanterman incisures. An exchange or a transfer process would permit to rapidly balance the distribution of phospholipids throughout the inner myelin layers within a few minutes rather than a few days after synthesis of new lipids in Schwann cells 11. The supply of Schwann cells with choline of axonal origin may be considered as a definite contribution of neuron to myelin economy in peripheral nerves. Owing to the limited entry of plasmatic choline into the endoneural space 11, axonal choline, mainly generated by axonal flow and turnover of CPG, is efficiently taken up by adjacent Schwann cells. The recycling of the base within a space of diffusion restricted to a minimum by the close association of axon and Schwann cells reveals another fascinating aspect of the metabolic cooperation existing between neurons and myelinating cells 21. ACKNOWLEDGEMENTS The technical assistance of Mrs. J. Boyenval and R. H/issig was extremely appreciated. M.B. received a traineeship from the European Training Programme in Brain and Behaviour Research, then from the Commissariat ~t l'Energie Atomique. L. Di G. is Charg6 de Recherches I.N.S.E.R.M.

83 REFERENCES 1 Arienti, G., Brunetti, M., Gaiti, A., Orlando, P. and Porcellati, G., The contribution of net synthesis and base-exchange reaction in phospholipid biosynthesis. In G. Porcellati, L. Armaducci and C. Galli (Eds.), Function and Metabolism of Phospholipids in the Central and Peripheral Nervous Systems, Raven Press, New York, 1976, pp. 63-78. 2 Autillio-Gambetti, L., Gambetti, P. and Shafer, B., Glial and neuronal contribution to proteins and glycoproteins recovered in myelin fractions, Brain Research, 84 (1975) 336-340. 3 Brunetti, M., Giuditta, A. and Porcellati, G., The synthesis of choline phosphoglycerides in the giant fibre system of the squid, J. Neurochem., 32 (1979) 319-324. 4 Carey, E. M., The role of soluble, cytoplasmic phospholipid exchange proteins in myelin sheath formation and turnover of myelin phospholipids, Proc. Europ. Soc. Neurochem., 1 (1978) 138. 5 Di Giamberardino, L., Bennett, G., Koenig, H. L. and Droz, B., Axonal migration of protein and glycoprotein to nerve endings III. Cell fraction analysis of chicken ciliary ganglion after intracerebral injection of labeled precursors of proteins and glycoproteins, Brain Research, 60 (1973) 147159. 6 Droz, B., Di Giamberardino, L., Koenig, H. L., Boyenval, J. and Hassig, R., Axon-myelin transfer of phospholipid components in the course of their axonal transport as visualized by radioautography, Brain Research, 155 (1978) 347-353. 7 Droz, B., Brunetti, M., Di Giamberardino, L., Koenig, H. L. and Porcellati, G., Transfer of phospholipid constituents to glia during axonal transport, Soc. Neurosci. Symp., 4 (1979) 344360. 8 Droz, B., Di Giamberardino, L. and Koenig, H. L., Contribution of axonal transport to the renewal of myelin phospholipids in peripheral nerves. I. Quantitative radioautographic study, Brain Research, 219 (1981) 57-71. 9 Ernster, L., Zetterstr6m, R. and Lindberg, O., Method for the determination of tracer phosphate in biological material, Acta chem. scand., 4 (1950) 942-950. 10 Giorgi, P. P., Relationship between axonal transport and the Wolfgram proteins in myelin, Proc. Europ. Soc. Neurochem., 1 (1978) 118. 11 Gould, R. M. and Dawson, R. M. C., Incorporation of newly formed lecithin into peripheral nerve myelin, J. Cell Biol., 68 (1976) 480~96. 12 Gould, R. M., Sinatra, R. S., Spivack, W., Berti, M., Wisniewski, H. M., Lindquist, T. and Ingoglia, N., Axonal transport of phospholipids in rat sciatic nerve, Neurosci. Abstr., 5 (1979) 60. 13 Haley, J. E. and Ledeen, R. W., Incorporation of axonally transported substances into myelin lipids, J. Neurochem., 32 (1979) 735-742. 14 Hendelman, W. and Bunge, R. P., Radiautographic studies of choline incorporation into peripheral nerve myelin, J. Cell Biol., 40 (1969) 190-208. 15 Horrocks, L. A. and Sun, G. Y., Ethanolamine plasmalogens. In R. Rodnight and H. Marks (Eds.), Research Methods in Neurochemistry, Plenum Press, New York, 1972, pp. 223-231. 16 Jungalwala, F. B., The turnover of myelin phosphatidylcholine and sphingomyelin in the adult rat brain, Brain Research, 78 (1974) 99-108. 17 Lowry, O. H., Rosebrough, N. J., Farr, L. A. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 18 Matthieu, J. M., Webster, H. de F., De Vries, G. H., Corthay, S. and Koellreuter, B., Glial versus neuronal origin of myelin proteins and glycoproteins studied by combined intraocular and intracranial labelling, J. Neurochem., 31 (1978) 93-102. 19 Miller, S. L., Benjamins, J. A. and Morell, P., Metabolism of glycerophospholipids of myelin and microsomes in rat brain. Re-utilization of precursors, J. biol. Chem., 252 (1977) 4025-4037. 20 Norton, W. T. and Poduslo, S. E., Myelination in rat brain: method of myelin isolation, J. Neurochem., 21 (1973) 749-757. 21 Quarles, R. H., Mclntyre, L. J. and Sternberger, N. H., Glycoproteins and cell surface interactions during myelinogenesis, Soc. Neurosci. Symp., 4 (1979) 322-343. 22 Rouser, G. and Yamamoto, A., Lipids. In A. Lajtha (Ed.), Handbook ofNeurochemistry, Vol. 1, Plenum Press, New York, 1969, pp. 121-169. 23 Sandra, A. and Pagano, R. E., Liposome-cell interactions. Studies of lipid transfer using isotopically asymmetric vesicles, J. biol. Chem., 254 (1978) 2244-2249. 24 Schonbach, J., Schonbach, C. and Cuenod, M., Distribution of transported proteins in the slow phase of axoplasmic flow. An electron microscopical autoradiographic study, J. comp. NeuroL, 152 (1973) 1-16.

84 25 Spanner, S. and Ansell, G. B., Enzymes involved in the metabolism of lipid bases, Bioehem. Soc. Trans. 7 (1979) 338-341. 26 Toews, A. D. Goodrum, J. F. and Morell, P., Axonal transport of phospholipids in rat visual system, J. Neurochem., 32 (1979) 1165-1173. 27 Ullman, M. D. and Radin, N. S., The enzymatic formation of sphingomyelin from ceramide and lecithin in mouse liver, J. biol. Chem., 249 (1974) 1506-1512. 28 Woelk, H. and Porcellati, G., Myelin catabolism, Proc. Europ. soc. Neurochem., 1 (1978) 64-77. 29 Wykle, R. L. and Snyder, F., The glycerol source for the biosynthesis of alkylglycerol ethers, Biochem. Biophys. Res. Commun., 37 (1969) 658-662. 30 Yavin, E., Regulation of phospholipid metabolism in differentiating cells from rat brain cerebral hemispheres in culture, J. biol. Chem., 251 (1976) 1392-1397.