Neurochemical Research, Vol. 20, No. It), 1995, pp. 123~1248

Biosynthesis and Compartmentalization of Po, Apolipoprotein A-I, and Lipids in the Myelinating Chick Sciatic Nerve M. J o a n n e L e m i e u x , 1 Catherine Mezei, 1 and W. Carl Breckenridge 1,2

(Accepted August 24, 1995)

Myelin deposition in developing chick sciatic nerve is associated with rapid synthesis of lipids, the major myelin protein Po and apo A-I, a major constituent of plasma lipoproteins. In order to understand possible roles of apo A-I in myelin assembly the synthesis and appearance of Po, apo A-I and lipids was studied in an intracellular fraction, an intralamellar fraction thought to be related to, or derived from, myelin and compact myelin from rapidly myelinating sciatic nerve of 1 day chicks. Incorporation with methionine or pulse-chase experiments indicated that initial synthesis of Po occurs in the intracellular fraction followed by movement to the intralamellar fraction and myelin. Incorporation of labelled oleate into phospholipids suggested that initial synthesis occurs in the intracellular and intralamellar fractions with slow movement to myelin. Incorporation of labelled galactose into cerebrosides suggested that initial synthesis occurs partially in myelin with slow loss from this fraction to the intralamellar fraction. However, incorporation of methionine into apo A-I indicated that initial synthesis occurred in the intracellular fraction with some transfer to the intralamellar fraction and secretion of a major portion into the incubation medium. It is concluded that the subcellular distribution of nascent apo A-I is not well coordinated with the distribution of other nascent constituents of the myelin membrane. The accumulation of nascent Po, phospholipids and cerebrosides in the intralamellar fraction compared to compact myelin suggests that this fraction may play a role as a precursor membrane or as a storage site for assembly of myelin constituents into compact myelin.

KEY WORDS: Apolipoprotein A-I; Po; avian; myelination; PNS.

INTRODUCTION

mechanism responsible for the selective deposition o f specific lipids and proteins in myelin remains poorly understood but vesicular transport is thought to be important in the selective deposition of proteins and lipids into pre-myelin membrane which becomes compact myelin (2-6). During membrane biogenesis Po tends to diffuse slowly from the site of initial deposition during myelination while labeled lipids are deposited in the same manner but equilibrate more rapidly throughout the m y -

Myelin biosynthesis by Schwann cell in the developing peripheral nerve (PN) represents one of the most rapid processes of membrane biogenesis. The lipid composition of myelin is specific with high concentrations of cholesterol, phospholipids and cerebrosides and Po, a myelin specific glycoprotein, which constitutes 50% of the proteins in compact myelin (for review ref. 1). The J Department of Biochemistry, Dalhousie University, Halifax, Nova Scotia B3H 4H7. 2 To whom to address reprint requests.

Abbreviations: PAGE: polyacrylamide gel electa'ophoresis; SDS: sodimn dodecyl sulphate; apo: apolipoprotein, PC: phosphafidylcholine, PE: phosphatidylethanolamine

1239 0364-3190/95/1000q239507.50/0 9 1995PtemtmPublishingCorporation

1240 elin sheath (7). Several investigators have isolated a membrane fraction associated with myelin from CNS or PNS that can be released by osmotic shock from compact myelin (8-15). The function of this fraction, refen'ed to as the myelin-like, SN4, SO, and intralametlar fraction, in relation to assembly of myelin has not been extensively investigated in PNS. The rate of biosynthesis and secretion of apolipoprotein (apo) A-I is greatly induced in the avian peripheral nerve during rapid myelination (11,15). Because rapid myelination requires the synthesis, transport and incorporation of large amounts of lipids into the newly formed myelin membrane it is possible that apo A-I is involved either in the transport or in selective incorporation of lipids into the various membranes of the sciatic nerve. Phospholipids, particularly phosphatidylcholine (PC) and phosphatidylethanolamine (PE) are one major component of PNS myelin (1). Amongst the glycosphingolipids, cerebrosides are increasingly enriched in compact myelin during development (1). The present investigation of avian sciatic nerve in short term organ culture combined with immunoprecipitation of specific proteins in subcellular fractions was undertaken to assess the subcellular sites and kinetics of synthesis and organization of PC, PE, cerebrosides and Po into myelin and potential roles for apo A-I in this process.

EXPERIMENTAL PROCEDURE Materials. [3sS]methionine (1186 Ci/mmol, translation ~ade), [9,10-3H]oleate (10 Ci/mmol), [6;-3H]galactose (35 Ci/mmol) were purchased from NEN-DuPont Canada Ltd., Mississauga, ON. The experimental animals were 1 day old chickens (Golden Comet or Brown Highliner strain, Berwick, NS). Plastic incubation wells (Falcon, 3037), all glass Dounce homogenizers, protein A Sepharose, low molecular weight protein standards, REDI plate silica gel G, 20 • 20 cm, lipid standards were obtained from Becton Dickinson Labware, Lincoln Park, NJ, Kontes Glass Co., Pharmacia, Bate Durfe, PQ, Fisher Scientific Co. Montreal, PQ and Matrea Inc., Pittsburgh, Penn., respectively. All other reagents use d were of the highest quality available. The rabbit anti-chick apo A-I immune serum was kindly donated by D. L. Williams, Dept. of Pharmaceutical Science, School of Medicine, State University of New York of Stony Brook, Stony Brook, NY. Labeling of Avian Sciatic Nerves in Organ Culture. Sciatic nerves of 1 day old chicks were quickly dissected free of connective and other adherent tissue and the endoneurium was removed from the perineurium and epineurium at 4~ under the dissecting microscope. The tissues were then placed immediately into cold sterile incubation medium [NaC1 (101 raM), CaCI 2 (1.8 raM), KCI (4.7 raM), MgClz (1.4 mM), Na~H PO4 (0.85 raM), NaHCOs (24.7 raM), glucose (10 raM)], which was pl,eviously saturated with 95% 02/5% COz at 2~ as described by Hu and Mezei (16). Nerves were sliced into 2 to 3 mm segments and placed into sterile plastic incubation wells with 0.5 ml of the above incubation medium-bicarbonate solution and containing

Lemieux, Mezei, and Breckenridge either 200 gCi of [3H]oleate or 250 gCi of [3sS]methionine with 250 gCi of [3H]galactose. For pulse-chase studies radioactive incubation medium was removed from the tissue by centrifugation (600 g • 5 min). The tissue was washed twice with incubation medium and returned to the incubation wells for varying times up to 4 b. The incubation medium contained the following unlabelled substrates as appropriate: for [35]S-methionine--a 69 fold excess, for pH]oleate--a 75 molar excess, for [3H]galactose a 714 molar excess. The tissues in the wells were incubated at 37~ in a moist atmosphere of 95% 02/5% CO2 with constant gentle shaking for the times indicated in the figure legends.

Subcellular Fractionation of Tissues for Solubilization and Analysis of Proteins. A modified version o{"the subcellular ffactionation of tissues for solubilization and analysis of proteins (11) is shown in Fig. 1. After incubation, the medium (secreted fraction) was carefully removed by centrifugation and dialyzed in 3 changes of 500 ml of solution containing sodium phosphate (0.02 M, pH 7.4), sodium chloride (0.15 M), PMSF (100 gg/ml), NaN 3 (0.05 mg/ml), mercaptoethanol (1% v/v), methionine (1 mM) and galactose (10 raM), when double labeling experiments were carried out, or (fatty acid free) albumin (0.7% w/v), when [3H]oleate was the precursor. The remaining endoneurial tissue was washed three times with incubation medium at 4~ followed by centrifugation at 600 g for 5 rain. The tissue was then homogenized (by 25 strokes at 2,800 rpm at 4~ in 2.5 ml sucrose (0.85 M) containing EDTA (1 raM) in an all glass homogenizer (Dotmce, Kontes Glass Co.). The homogenate was layered onto 0.8 ml sucrose (1.2 M). Then 0.4 ml of sucrose (0.32 M) was layered on top and the sample was ultracentrifuged at 100,000 g for 120 rain at 4~ in a Ti75 rotor (Beckman Ins.). Two major fractions (not osmotically shocked) were recovered from this Nadient and processed further as follows, The crude myelin layer at the 0.32 M/0.85 M sucrose interface and a supernatant in 0.85 M sucrose below the crude myeIin band and between the very thin lower band at the 0.85 M/t.2 M sucrose interface were carefully collected by a Pasteur pipette. Previously the supernatant in 0.85 M sucrose was termed as supernatant 2 (S-2, ref. 11). This supematant was then dialyzed as described above for the secreted fraction. The resulting solution is termed and referred to as the "intracellular" fraction in this paper_ To purify the crude myelin fraction, the crude myelin layer was collected from the 0.32 M/0.85 M sucrose interface and was subjected to "osmotic shock" treatment by diluting it with 10 volumes of ice-cold distilled water, followed by hand homogenization with 10 up and down strokes in an all glass Dounce homogenizer at 0~ The suspension was allowed to stand on ice for 15 min and then centrifuged at 32,800 g for 30 rain. The resulting supernatant solution (previously termed the SO fraction (1 l)) was collected and designated as the "intralamellar" fraction. This fraction represents cytoplasmic inclusions and loosely associated myelin fragments released by osmotic shock from the lamellae of isolated myelin (17). The myelin pellet obtained after the osmotic shock treatment was washed three times with 20 volumes of ice-cold distilled water containing PMSF (100 gg/ml) followed by centrifugation at 32,800 g for 30 rain at 4~ to obtain the compact myelin fraction. [mmunopreeipitation of Labeled Proteins. Immunoprecipitations were carried out as described by Blue et al. (18) and LeBlanc et al. (11,19), with the following modifications. Labeled endoneurial proteins (50-200 I-tg in 100 gl) from the various subcellular fractions were reacted with excess rabbit anti-chick apo A-I immune serum (7.5 Ixl) or anti-chick Po IgG (5 gl) or preimmune serum in an ammunoprecipitation buffer (12 gl) to attain a final solution of sodium phosphate buffer pH 7.4, (0.02 M), NaC1 (0.15 M), EDTA (0.005 M), containing PMSF (100 gg/ml). This mixture was incubated at room temperature for 2.5 h and then overnight at 4~ To each 120 gl of the antibody-

Coordination of Po, Apo A-I, and Lipid Biosynthesis in Myelinating Chick Sciatic Nerve antigen complex, 70 gl of immunoprecipitate washing buffer consisting of sodium phosphate buffer, pH 7.4 (0.02 M), NaC1 (0.15 M), EDTA (0.005 M), and PMSF (100 ~tg/ml), was added followed by the addition of 100 gl of Protein A Sepharose which had been equilibrated in the washing buffer mixture (as described above) at a ratio of 1:1 [v/v]. The immunoprecipitation mixture was further incubated at room temperature for 2.5 h followed by 1.5 h at 4~ The amount of Protein A-Sepharose used in the immunoprecipitation was determined by a dose study with the appropriate antibody. No radioactive proteins were immunoprecipitated when preimmune serum was used instead of the anti Po or anti apo A-I antibodies (see refs. 11 and 19 for results). For immunoprecipitation of Po protein from the myelin fraction, myelin proteins were extracted by suspending the myelin pellet in 0.2 ml of SDS (2%) and boiling the suspension for 2 rain. The solubilized proteins were collected by centrifugation at 16,000 g for 5 min and were diluted with sterile water to obtain a final concentration of SDS of 0.4% [v/v]. An aliquot of this solution was then reacted with an excess of anti-chick Po antibody. After incubation with the appropriate antiserum and Protein A-Sepharose was complete, the incubation mixture was centrifuged at 16,000 g for 5 min. The supernatant of the immunoprecipitation was removed and the Sepharose pellet complexed with the protein antibody-antigen complex, was washed three times with 1 ml of immunoprecipitate washing buffer. SDS-PAGE and Fluorography of Labeled Proteins. Samples were prepared for polyacrylamide gel electrophoresis as described by LeBlanc et al. (t 1). Briefly, washed immunoprecipitates and original tissue extracts were diluted with a deoxycholate solution and protein solubilizer [mercaptoethanol (4%, v/v), Tris-HC1 (0.25 M, pH 8.6), glycine (192 raM), and SDS (1%, v/v) to obtain a final solution containing deoxycholate (0.05% v/v) and protein sotubilizer (75%, v/v). The samples were boiled for 2 min, centrifuged at 16,000 g for 5 min and then cooled. An aqueous solution containing bromophenol blue (l.2%, vet/v) and glycerol (75%, v/v) was added to the protein solution at 10% [v/v] before applying the mixture to an electrophoresis gel. Labelled ilnmunoprecipitated protein and total tissue extracts were fractionated by SDS-10% polyacrylamide slab gels as described previously (20). Radioactive proteins were visualized by fluorography (21). Low molecular weight protein standards were routinely run on gels for molecular weight calibration. Samples from the gels were excised and counted immediately after the gel was dried. The samples were also counted 6 months later to determine the level of [3H]galaetose which may be associated with Po, or apo A-I. Radioactive incorporation of [3sS]methioninein Po and apo A-I proteins was determined by subtracting the [3H]galactose counts from the pS]methionine plus the [3H]galactose counts. Extraction of Labeled Lipids. Lipids in the subcellular fractions were extracted according to a modified procedure of Oulton and Mezei (22). Aliquots of aqueous subcellular fractions were extracted 3 times with 1 ml of chloroform-methanol (2:1, v/v), followed by the addition of cold Krebs chicken-Ringer solution (101 mM NaC1, 1.8 mM CaC12, 4.7 mM KC1, 1.4 mM MgC12, 0.85 mM Na~HPO4, 24.7 mM NaHCO3, 10 mM glucose) to reach a final proportion of 8:4:3 [v/v/v] of chloroform-methanol-Krebs chicken-Ringer solution. All tubes were shaken for approximately 10 min and centrifuged at 1,000 rpm for 5 min. To the lower chloroform-methanol phase, 0.2 volumes of cold Krebs chicken-Ringer-methanol (1:1, v/v) was added, shaken for 10 min and centrifuged at 1,000 rpm for 5 mfu. The lower phase was then washed three times with 2 ml of cold chloroform-methanol-water (3:48:47, v/v/v), shaken for 10 min and centrifuged at 1,000 rpm for 5 min. The lipid extracts were then dried under Nz, and made up to a known volume of chloroform or chloroform-methanol (2:1, v/v) and stored at 4~

1241

Thin Layer Chromatography and Quantitation of Labeled Lipids, Lipid extracts were spotted on REDI plate Silica gel G, under N2. For phospholipid analysis samples were separated in a solvent system consisting of chloroform-methanol-28% ammonia (65:25:5, v/v/v). Galactose radiolabelled lipids were separated in a solvent system consisting of chloroform-methanol-water (60:17:2, v/v/v). Dry plates were scanned on a Bioscan Model 200 radioimaging scanner for 20 rain. Lipid standards containing PC, PE, phosphatidyl serine, sphingomyelin, and eerebrosides, as well as lipids from purified myelin of chick PNS containing cholesterol, PC, PE, phosphatidylserine, sphingomyelin, cerebrosides and cerebroside sulfates (22) were run with the thin layer plates. Lipids were stained with flnorescein and visualized with a UV lamp.

RESULTS In order to compare the biosynthesis o f Po, apo AI and cerebrosides and their appearance in subcellular fractions o f m y e l i n a t i n g sciatic nerve u n d e r the same conditions, double labeling experiments were carried out (Figs. 2 to 4). In this protocol endoneurial slices o f one day old chicks were incubated with p s S ] m e t h i o n i n e and [3H]galactose to assess overall incorporation with time a n d u n d e r pulse-chase conditions. The tissues were fractionated into intracellular, intralamellar and compact m y elin fractions (Fig. 1) a n d the fractions were assayed for labeled Po (Fig. 2 A and B), apo A - I (Fig. 3 A and B), and cerebrosides (Fig. 4A, B). Figure 2 A shows the time course o f [3sS]methionine incorporation into Po protein. After 1 - 2 h o f radiolabelling the majority o f [35S]methionine labelled Po is located in the intracellular fraction. After 4 h o f incub a t i o n the level o f radiolabeled Po increases approxim a t e l y 4 fold with the majority detected in the intracellular and intralamellar compartments. The level o f intralamellar Po approaches that o f the intracellular fraction after 24 h o f i n c u b a t i o n time. As expected, the incorporation o f Po into compact m y e l i n was very low during the 1st h o f incubation, but thereafter it increased at a slow, b u t steady rate throughout tile 24 h i n c u b a t i o n period. Pulse-chase experiments were carried out to assess the metabolic stability o f Po in the various fractions. Figure 2B shows that after a 1 h pulse followed b y a 1 or 3 h chase period with u n l a b e l e d methionine, n a s c e n t Po could be partially chased from the intracellular fraction, but not from the intralamellar fraction. After a 3 h chase period, n e w l y synthesized Po accumulated pred o m i n a n t l y in compact m y e l i n . These results indicate that the initial site o f synthesis o f Po is the intracellular fraction, whereas the intralamellar fraction m a y serve as a temporary storage c o m p a r t m e n t for the n e w l y f o r m e d glycoprotein. Po is a stable c o m p o n e n t o f compact m y elin, where it accumulates during the chase periods.

1242

Lemieux, Mezei, and Breckenridge

Nerve and Radioactive Incubation Medium

A 1000

SECRETED FRACTION: SECRETED PROTEINS AND LIPIDS

800 > Homogenize Nerve in 0.85 M Sucrose + 1 mM EDTA

z N 9

oo/U 40~M

Ultracentrifugation

0.32 M Sucrose ~ 0.85 M Sucrose ~ ~ lrnM EDTA ~

o l d ,

CRUDE IVI~ELIN_

1.2 M Sucrose {

0

INTRACELLULAR FRACTION: CYTOSOLIC PROTEINS AND LIPIDS

5

10

......... 15

20

25

INCUBATION TIME (h)

B

35~I

Osmotic Shock 10 vol dH20

300I

25~

Centrifuge at 32,000 x g, 30 min

20O Supematant of smotic Shock Collect Pellet : Osmotically Shocked Myelin ~ t B ~

INTRALAMELLAR FRACTION: CYTOSOLIC PROTEINS AND LIPIDS BETWEEN MYELIN LAMELLAE PURIFIED COMPACT MYELIN

Fig. 1. Subcellular fractionation protocol of endoneurial sciatic nerve slices.

150 9

100 50

0

INTRACELLULAR INTRALAMELLAR

MYELIN

FRACTION

In order to compare the biosynthesis of apo A-I with that of complex lipids and Po, we determined the levels of labeled apo A-I in the intracellular, intralamellar and secreted fractions after various incubation periods of endoneurial slices with pS]methione. Figure 3A shows that during the first 4 h of incubation time the highest amounts of nascent apo A-I occurred in the intracellular fraction. The amount of newly synthesized apo A-I in the intralamellar fraction lagged behind that in the intracellular fraction during the first 2 h of incubation time, but rose steadily between 4 and 24 h. A substantial portion of the labeled apo A-I is also secreted into the incubation medium of the organ culture. While the amount of secreted apo A-I was low after 2 h of incubation it surpassed the amount in the other two fractions after 24 h. The distribution of newly synthesized

Fig. 2. Time course and pulse-chase of [35S]methionine incorporation into Po protein of the rapidly myelinating sciatic nerve of the chick. Endoneurial slices of 10 sciatic nerves from 1 day post-hatch chicks were incubated at 37~ with [SSlmethionine for 1, 2, 4, and 24 h, panel A; and a 1 h pulse (blank bars), plus a 1 h pulse followed by 1 h (hatched bars) and 3 h chases (solid bars), panel B. The tissue was fractionated and extracted for proteins as described in the materials and methods. Po was immunoprecipitated from the intracellular (0), intralamellar (11), and myelin (Q)) fractions of the sciatic nerve containing - 5 0 - 2 0 0 gg protein. The immunoprecipitated Po was separated by PAGE, excised and counted as described in Experiment Procedure. The results are data from a single typical experiment using dual labeling experiment with pS]methionine and [SH]galactose.

apo A-I after a 24 h incubation with labelled methionine was: secreted (51%), intracellular (27%) and intralamellar (22%). Pulse-chase experiments with unlabelled methionine showed that nascent apo A-I levels increased slightly in the secreted, and more substantially in the

Coordination of Po, Apo A-I, and Lipid Biosynthesis in Myelinating Chick Sciatic Nerve

1243

A

A

12

1000 m

z

800

6001 0

<

400C

2

9

<

m

200G

i

I

5

t0

20

15

0

25

,

- . . . . .

5

l0

15

20

"O I

25

/ N C U B A T I O N T I M E (h)

INCUBATION TIME (h)

B

2

B ;>

z ta~

L)


c~

~3

0 INTRA C E L L U L A ~ I N T RA L A M E L L A R INTRACELLULAR

SECRETED

MYELIN

INTKA L A M E L L A R

FRACTION Fig. 3. Time caurse and pulse-chase of [~sSJmethionine incoqJoration into apo A-I protein of the rapidly myelinating sciatic nerve of the chick. Endoneurial slices of I0 ~ciatic nerves from 1 day post-hatch chicks were incubated at 37~ with [358]methionine for 1, 2, 4, and 24 h, panel A; and a 1 h pulse (blank bars), plus a 1 h pulse followed by 1 h (hatched bars) and 3 h cha~es (solid bars), panel B. The tissue was fractionated into secreted (~), intracellular (0), and intralamellar (m) fractions and extracted for proteins as described in the materials and methods. Ape, A-I was immunoprecipitated from ~be intraaellular, secreted, and intralamellar fractions of the sciatic nerve containing ~50200 I-tg protein, The immunoprecipitated apo A-I was separated by PAGE, excised and counted as described in Experimental Procedm'e. The results of are from a single experiment using dual labeling with [35S]methionine and [3H]galactose.

intralamellar fractions after the chase (Fig. 3B). H o w ever, there was a slow, but definite chase o f labelled apo

FRACTION Fig, 4. Time course and pulse-chase of pHJgalactose incorporation into cerebrosides of the rapidly myelinating sciatic nerve of the chick. Endoneurial slices of 10 sciatic nerves from 1 day post-hatch chicks were incubated at 37~ with ~3H]galactose for 1, 2, 4, mad 24 h, panel A; and a [ h pulse (blank bars), plus a 1 h pulse followed by 1 h (hatched bars) and 3 h chases (solid bars), panel B. The tissue was fractionated into intracellular (0), intralamellar ( I ) , and myelin (O) fractions and extracted for tipids as described in the materials and methods. Lipids were separated by thin layer chromatography. Cerebrosides were identified through co-migration with standards on the thin/ayer plates, and cour~ted as described in Experimental Procedure. The results are from a typical experiment with dual labeling with [3sS]methionine and [~H]galactose.

A - I f r o m the intracellular fraction, These results indicate a s l o w t u r n o v e r o f apo A - I in the intracellular fraction and an a c c u m u l a t i o n o f this protein in the secreted and intralamellar compartments.

1244

Lemieux, Mezei, and Breckenridge

A 12

6 i:x.,

4

r.) ...-----O

0 0

~

,

[

i

I

,

i

5

i

10

,i

'

9 '

'

I

. . . .

15

I

. . . .

20

I

25

I N C U B A T I O N T I M E (h)

B

,-?, O

r) r,2

\\

\'<

INTRACELLULAR INTRALAMELLAR

MYELIN

FRACTION Fig. 5. Time course and pulse-chase of pH]oleate incorporationinto phosphatidylcholine of the rapidly myelinating sciatic nerve of the chick. Endoneurial slices of 10 sciatic nerves from 1 day post-hatch chicks were incubated at 37~ with [3H]oleatefor 2, 4, 12, and 24 h, panel A; and a 2 h pulse (hatched bars), or a 2 h pulse followedby a 4 h chase (solid bars), panel B. The tissue was fractionatedinto intracellular (0), intralamellar (B), and myelin (Q)) fractions and extracted for lipids as described in the materialsand methods. Lipids were separated by thin layer chromatography.Phosphatidylcholinewas identified through co-migrationwith standards on the thin layerplates, and counted as described in ExperimentalProcedure. The results are from a single typical experimentwith [3H]oleateand simultaneous analysis of PC and PE for each time point.

The subcellular fate of [3H]galactose labeled cerebrosides in the various fractions was quite distinct from that observed for [35S]methionine incorporation into Po

and apo A-I proteins. Incorporation of galactose label into cerebrosides was highest in myelin initially and the amount in this fraction remained fairly constant between the 4th and 24th h of incubation time (Fig. 4A), while the incorporation of label into the intracellular and intralamellar fraction was very low initially. A definite lag in the incorporation of label into the intracellular fraction occurred between 0 and 2 h. However, the incorporation of label continued to increase with time such that the intracellular and intralamellar fractions contain mo~t of the labeled cerebroside after 24 h. The pulse-chase experiments with labeled and unlabeled galactose indicated that labeled intracellular cerebrosides could be chased from the intracellular fraction (Fig. 4B). None of the labeled cerebrosides could be chased from the intralamellar fraction, but rather they accumulated in that compartment. However, nascent cerebrosides could be slowly chased from the compact myelin fraction after a 1 or 3 h chase with unlabelled galactose, indicating that, in contrast to Po protein, the newly synthesized cerebrosides may not be stable constituents of the myelin membrane (compare Fig. 4B with 2B). [3H]oleate was used as a precursor for incorporation into PC and PE either through de novo synthesis or by acylation of lysophospholipids, Incorporation of [3H]oleate into PC (Fig. 5A) and PE (Fig. 6A) occurred at a high rate in the intralamellar fractions during the first 4 h of incubation. After that time period, the radiolabeled PC in the intralamellar fraction remained high, whereas the level of newly labelled PE rose at a slow, but constant rate. The rate of incorporation of oleic acid into intracellular PC was rapid during the first 4 h and decreased after this period while that of PE remained slow and constant throughout the incubation. During the incubation, the accumulation of PC and PE in myelin, which was about 20% of the accumulation in intralamellar fraction, peaked at 12 h and remained constant up to 24 h. Pulse-chase experiments with labeled and unlabeled oleate demonstrated that radiolabeled PC (Fig. 5B) and PE (Fig. 6B) could be depleted from the intracellular fraction, but not from the intralamellar compartment. PE appeared to be chased into compact myelin over the 3 h chase. These results indicate that the turnover of phospholipids is fast in the intracellular compartment, but slow in the intralamellar and compact myelin fractions. The kinetics of synthesis and distribution of phospholipids and cerebrosides in cellular fractions are quite distinct from that of apo A-I. Although a substantial amount of apo A-I appears in the intracellular and intralamellar fraction a major proportion of apo A-I is secreted into the medium by 24 h. More apo A-I is found

Coordination of Po, Apo A-I, and Lipid Biosynthesis in Myelinating Chick Sciatic Nerve A

f

;o Z

Y

o~

A

5

t0

I

i

i

15

20

25

INCUBATION TIME (h)

B

Z g-.

0

INTRACELLULAR INTRALAMELLAR

MYELIN

FRACTION Fig. 6. Time course and pulse-chaseof [3H]oleate incorporation into phosphatidyl ethanolanfineof the rapidly myelinating sciaticnerve of the chick. Endoneurial slices of i0 sciatic nerves from 1 day posthatch chicks were incubatedat 37~ with [3H]oleate for 2, 4, 12, and 24 h, panel A; and a 2 b pulse (hatchedbars), or a 2 h pulse followed by a 4 h chase (solidbars), panel B. The tissue was fractionatedinto intracellular (O), intralamellar ( I ) , and myelin (Q)) fractions and extracted for lipids as described in the materials and methods. Lipids were separated by thin layer chromatography. Phosphatidylethanolamine was identified through co-migration with standards on the thin layer plates, and counted as described in Experimental Procedure. The results are from a single typical experiment with pH]oleate and simultaneous analysis of PC and PE for each time point.

in the intraceUular fraction than the intralamellar fraction while incorporation of oleate into PE and PC is highest in the intralamellar fraction. Cerebrosides are highest initially in the myelin fraction. Very little labeled lipid was secreted into the medium (less than 2% of label at 24 h, data not shown) indicating that much less lipid was associated with apo A-I than with membranes.

1245

DISCUSSION The results of our investigation indicate that incubation of avian sciatic nerves in short term organ cultures with radioactive protein and lipid precursors provides an assessment of the incorporation of specific proteins and complex lipids into membranes and myelin in the rapidly myelinating peripheral nerve. The incorporation of [3sS]methionine, [3H]galactose and [3H]oleate into proteins and lipids continued over 24 h of incubation (Figs. 2A-6A). These results indicate that the conditions of the incubation resulted in synthesis of proteins and lipids for up to 24 h although the incorporation was not linear over the entire period. The dissection of the sciatic nerve disrupts the Schwann cell-axonal contact and Wallerian degeneration eventually occurs (23). Wallerian degeneration has been demonstrated in transected peripheral nerves after 2 days (24). However, in the present studies radiolabeled Po is incorporated into myelin for up to at least 24 h of incubation indicating the stability of the myelin sheath during this incubation period. To examine the subcellular route of biosynthesis and compartmentalization of Po, apo A-I, phospholipids and cerebrosides, the nerves were subjected to iso-osmotic homogenization (11). The homogenate was applied directly on a discontinuous sucrose density gradient and ultracentrifuged. This protocol produced fractions in which the integrity of the myelin and other subcellular compartments is maintained. The cytosolic compartment was found in the gradient below the myelin band and was designated as the intracellular fraction. Loosely bound membrane proteins and myelin cytoplasmic inclusions were trapped in the crude myelin fraction and released by osmotically shocking the crude myelin, followed by centrifugation. This osmotically shocked supernatant of the crude myelin was designated as the intralamellar fraction and consisted of proteins found between or loosely associated with the myelin lamellae prior to formation of compact myelin. The intralamellar fraction contains small amounts of microsomes and Golgi apparatus which are potentially engaged in protein and lipid biosynthesis (17). These three fractions, intracellular, intralamellar and myelin, represent the three major compartments in the myelinating sciatic nerve. After a 24 h incubation the content of protein in the fractions was: secreted, 1115 + 211 gg (38.2%); intracellular, 229 + 52 pg (7.8%); intralamellar, 795 + 115 gg (27.2%); myelin, 783 + 182 gg (26.8%), indicating that a substantial proportion of the total endoneurial protein was reproducibly recovered in the intralamellar fraction.

1246 Our experiments indicated that in the one day post hatch sciatic nerve of chicks nascent Po appeared first in the intracellular, followed by the intralamellar fractions and, after a definite lag period, in compact myelin (Fig. 2A). Other investigators have examined the subcellular location of Po in the peripheral nerve. In the developing rat, Po was detected immunochemically in the Golgi-enriched cytoplasm of the Schwann cell and in some areas of myelin sheath showing only three lamellae. In the adult rat, very little Po was detected in the cytoplasm but the myelin sheath was intensely stained (25,26). Our results indicate that Po was synthesized in the intraceltular compartment and transferred to myelin. Po accumulated in the intralamellar fraction which may serve as a pool for this protein destined for myelin. The intralamellar fraction is thought to be a premyelin fraction which consists of membrane fragments and cytoplasmic inclusions of the myelin sheath. Linington and Waehneldt (27) have also examined a similar, but not identical fraction in the rabbit sciatic nerve and found a higher specific activity for labeled proteins in the so called heavy myelin fraction as opposed to the compact myelin fraction. Rapaport and Benjamins (28) also detected a lag in the incorporation of radiolabeled Po in a peripheral nerve total homogenate and in its appearance in myelin. From our results and those of others, we can infer that Po is synthesized in the intracellular compartment, transferred and possibly stored in the intralamellar fraction from where it is deposited in the myelin sheath. This concept is supported from data of the pulse-chase study (Fig. 2B), which shows that nascent Po appears early in the intracellular fraction, but after a 1 and 3 h chase period a portion of it accumulates in the intralamellar fraction. In compact myelin radiolabeled Po increases with increasing chase time. Studies with various radioactive tracers for complex lipids also indicated that the intralamellar compartment from the avian sciatic nerve may be functioning as a precursor for myelin phospholipids. [3H]Oleate was incorporated effectively into PC and PE and probably represents synthesis via acylation of lysophospholipids or via the nucleotide pathway for choline and ethanolamine (29). In the intracellular and intralamellar fractions, the [3H]oleate label was found predominantly in PC and PE with small amounts detected in phosphatidylserine and sphingomyelin (data not shown). These data are consistent with changes in the phospholipid composition of myelin during development (22,30). The pulse-chase experiments indicated that initial incorporation of label into PC and PE occurred in the intracellular and intralamellar compartments (Figs. 5B and 6B). The pattern of incorporation of [3H]oleate into phospholipids indicated that

Lemieux, Mezei, and Breckenridge labelled PC and PE also appeared rapidly in the intralamellar compartment. The pulse-chase data indicated that [3H]oleate labeled PE is chased into the myelin fraction. These results provide evidence for the concept that phospholipids are synthesized in the intracellular compartment and transported to the intralamellar and myelin fractions. [3H]Galactose is expected to label a variety of galactolipids. Incubation of avian sciatic nerves with [3H]galactose resulted in the majority of the label appearing in cerebrosides indicating little conversion of galactose to other lipids in this time period. Cerebrosides are a major component of myelin (30). Studies have been reported describing the changes in cerebroside composition of the developing avian sciatic nerve (22) and cerebroside biosynthesis in the rat sciatic nerve during development (31). Lipid analysis of the [3H]galactose labeled subcellular fractions of the sciatic nerve indicated that considerable synthesis of cerebrosides occurred in compact myelin (Fig. 4A). After a 24 h incubation, nascent cerebrosides accumulated in the intralamellar fraction. These results suggest that cerebrosides are synthesized in the myelin, and also may be synthesized in the intracellular and intralamellar compartments or transferred to these fractions after synthesis in the myelin. Other studies performed on brain slices in which the Golgi apparatus was disrupted with colchicine and monensin showed a decline in the appearance of cerebroside sulfates in myelin as compared to the control (32). However, cerebroside deposition in myelin was not affected by colchicine and monensin. On the basis of these results it was proposed that there are two pools of cerebrosides in oligodendroglia. It is possible that in the sciatic nerve the intralamellar fraction may represent one of the pools. The early detection of nascent cerebrosides in higher proportions in the myelin fraction compared to intralamellar fraction is in accordance with previous publications that demonstrate synthesis of galactocerehrosides within the myelin sheath (33). The final step in the synthetic pathway of galactocerebrosides is the transfer of galactose to ceramide by the UDP-galactose transferase reaction (34). The activity of this enzyme is increased in myelin during development (35-37). The production of another cerebroside, glucocerebroside, occurred in the microsomal fraction. Enzymes for the transferase reaction for both galactocerebroside and glucocerebroside synthesis were also located in the microsomal compartments of the PNS (38). The production of glucoeerebrosides occurred in the Golgi apparatus prior to myelination (5) and during degeneration in the PNS (39). During degeneration, the synthesis of galactocere-

Coordination of Po, Apo A-I, and Lipid Biosynthesis in Myelinating Chick Sciatic Nerve broside switched to glucocerebrosides as a result o f loss o f axonal contact between the Schwann cell and the axon (40) and proliferation o f the myelinating and nonmyelinating Schwann cell (41). A n increase in glucosylceramide synthesis in the avian endoneurial slices in organ culture m a y result in a decrease in levels o f [3H]galactose labeled galactosylceramides detected in the compact myelin after 1 and 3 h chase. The precise role o f apo A-I in the rapidly myelihating avian nerve is unclear. Although the ontogeny o f apo A - I parallels myelination and preliminary experiments indicated that some lipids, particularly cholesterol, cholesteryl esters and phospholipids were specifically associated with apo A - I containing lipoprotein particles (11-13), the synthesis and distribution o f the major portion o f nascent PC, PE and cerebrosides were not coordinated with those o f apo A - I in the current studies. The findings presented in this paper, however, do not exclude a role o f apo A - I in equilibrating or delivering lipids among subcellular or cellular compartments o f the endoneurium which m a y provide a major sink for lipids. Pulse-chase experiments also suggest a role o f apo A-1 in the intralamellar compartment. Nascent apo A-I is partially depleted from the intracellular fraction, indicating that this fraction represents the initial site o f synthesis o f the protein, Upon chase, nascent apo A - I accumulates in the intralamellar ftaction (Fig. 3B). In peripheral tissues, apo A - I participates in reverse cholesterol transport and possibly acquires lipids b y aqueous diffusion between plasma membranes and the lipoprotein particles (42). Other investigators have demonstrated that an apo A-1/HDL receptor is responsible for the efflux o f lipid involved in reverse cholesterol transport (43). Thus, apo A-I m a y be involved in the accumulation o f lipids by these two mechanisms in the intracellular and intralamellar fractions. However, there are a variety o f cell types in the endoneurium (44). It is also possible tbat apo A - I is not synthesized b y Schwaml cells, but b y resident macrophages as demonstrated for apo E in m a m m a l i a n (45), or b y endothelial cells o f the microvasculature (46). Immunocytochemical studies, in situ hybridization with anti-sense R N A or biochemical experiments with isolated and specific cell types o f the avian sciatic nerve would further contribute to the knowledge o f apo A - I in the process o f myelination.

ACKNOWLEDGMENTS This investigation was supported by the Medical Research Council of Canada. The clerical assistance of Mrs. B. Bigelow is greatly appreciated.

1247

REFERENCES 1. Momll, P., Quarles, R. H., and Norton, W. T. 1994. Myelin formation, structure, and biochemistry. Pages 117-143, in Siegel O. J., Agranoff B. W., Albers R. W., Molinoff P. B. (eds): "Basic neurochemistry: Molecular, Cellular, and Medical Aspects." 5th Ed. New York: Raven Press. 2. Herschkowitz, N., McKharm, G. M., Saxena, S., and Shooter, E. M. 1968. Characterization of sulphatide-containinglipoproteins in rat brain. J. Neurochem. 15:1181-1188. 3. Benjamins, J. A. and McKhann, G. M. 1973. Properties and metabolism of soluble lipoproteins containing choline and ethanolamine phospholipids in rat brain. J. Neurochem. 20:1121-1129. 4. Brammer, M. J. 1978. The protein-mediated transfer of lecithin to sub-fractions of mature and developing rat myelin. 1. Neurochem. 31:1435-1440. 5. Koul, O., Singh, I., and Jungalwala, F. B. 1988. Synthesis and transport of cerebrosides and sulfatides in rat brain during development. J. Ne~ochem. 50:580-588. 6. Brown, M. C., Moreno, M. B., Bongarzone, E. R., Cohen, P. D., Soto, E. F. and Pasqumi, J. M. 1993. Vesicular transport of myelin proteolipid and cerebroside sulfates to the myelin membrane. J. Neurosci. Res. 35:402-408. 7. Gould, R. M. 1977. Incorporation of glycoproteins into peripheral nerve myelin. J. Ceil Biol. 75:326-328. 8. Agrawal, H. C., Banik, N. L., Boge, A. H., Davison, A. N., Mitchell, R. F., and Spohn, M. 1970. The identity of a myelin-like fraction isolated from developing brain. Biochem. J. 120:635-642. 9. Fischer, C. A., and Morell, P. 1974. Turnover of proteins in myelin and myelin-like material of mouse brain. Brain Res. 74:5165. 10. Waehneldt, T. V. 1978. Protein heterogeneity in rat CNS myelin subfractions. Adv. Exp. Med. Biol. 100:117-133. 11. LeBlanc, A. C., Foldvari, M., Spencer, D. F., Breckenridge, W. C., Fenwick, R. G., Williams, D. L., and Mezei, C. 1989. The apolipoprotein AI gene is actively expressed in the rapidly myelinating avian peripheral nerve. J. Cell Biol. 109:1245-1256. 12. Lemiemx,M. J., Breckenridge, W. C., and Mezei, C. 1993. Apolipoprotein A-I containing lipoprotein particles may serve as lipid transporters in the avian peripheral nerve. J. Neurochem. 61 Supp. S194B. 13. Mezei, C., Lemieux, M. J., and Breckenridge, W. C. 1994. Myelinogenesis and apo A-I associated lipids in the avian PNS. J. Neurochem. 62 Supp. Abst. $54C. I4. Lemieux, J., Gimmoulis, S., Breckenridge, W. C., and Mezei, C. 1995a. Post-I-ranslationalmodifications of apolipoprotein Ad and Po proteins in the avian peripheral nerve. Neurochem. Res. 20: 25-34. 15. Lemieux, M. J., Mezei, C., and Breckev.ridge,W. C. t995b. Biosynthesis of apolipoprotein A-I in the avian sciatic nerve during development. Submitted. 16. Hu, Y. W. and Mezei, C. 1971. RNA synthesis in the peripheral nerve in the chick during development. Can. J. Biochem. 49:320327. 17. Ntmn, D. J., LeBlanc, A. C., and Mezei, C. 1987. A 42K protein of chick sciatic nerve is inmmnologicallyrelated to PO protein of periperal nerve myelin. Neurochem. Res. 12:377-384. 18. Blue, M.-L., Ostapchuk, P., Gordon, J. S., and Williams, D. L. 1982. Synthesis of apolipoprotein AI by peripheral tissues of the rooster. J. Biol. Chem. 257:11151-11159. 19. LeBlanc, A. C., Poduslo, J., and Mezei, C. 1987. Gene expression in the presence and absence of myelin assembly. Mol. Brain Res. 2:5%67. 20. Nuun, D. J. and Mezei, C. 1984. Solid-phase immunoassay of Po glycoprotein of peripheral nerve myelin. J. Neurochem. 42:158165.

1248 21. Bormer, W. M., and Laskey, R. A. 1974. A film detection method for tritium-labelled proteins and nucleic acids in polyacrylamide gels. Eur. J. Biochem. 46:83-88. 22. Oulton, M. R. and Mezei, C~ 1976. Characterization of myelin of chick sciatic nerve during development. J. Lipid Res. 17:167-175. 23. Bignarni, A. M., Dahl, D., Nguyen, B. T., and Crosby, C. J. 1981. The fate of axonal debris in Wallerian degeneration of rat optic and sciatic nerves. J. Neuropathol. Exp. Neurol. 40:53%550. 24. Poduslo, J. F. 1993. Regulation of myelin gone expression in the peripheral nervous system. Pages 282-289, in Dyck, P. J., Thomas, P. K., Griffin, J. W., Low, P. A. and Poduslo, J. F., (eds): Peripheral Neuropathies, 3rd ed. 25. Trapp, B. D., Itoyama, Y., Stemberger, N. H., Quarles, R. H., and deF. Webster, H. 1981. Immunocytochemical localization of Po protein in:Golgi complex membranes and myelin of developing rat Schwann cells. J. Cell Biol. 90:1-6. 26. Baron, P., Shy, M., Honda, I-l., Sessa, M., Kamholz, J., and Pleasure, D. 1994. Developmental expression of PO mRNA and PO protein in the sciatic nerve and the spinal nerve roots of the rat. J. Neurocytol. 23:249-257. 27. Linington, C., and Waehneldt, T. V. 1981. The in vivo synthesis of myelin proteins in rabbit sciatic nerve. Neurochem. Internat. 3: 385-395. 28. Rapaport, R. N., and Benjamins, J. A. 1981. Kinetics of entry of Po protein into peripheral nerve myelin. J. Neurochem. 37:164171. 29. Vance, D. 1985. Phospholipid metabolism in eukaryotes. Pages 243-270, in Vance, D. and Vance, J. E. (eds.): Biochemistry of lipids and membranes. Benjamin/Cummings Pub. Co. 30. Mezei, C. 1993. Myelination in the peripheral nerve during development. Pages 267-281, in Dyck P. J., Thomas P. K., Griffin J. W., Low PA, Poduslo J. F., (eds.): Peripheral Neuropathies, 3rd ed. 31. Yao, J. K. 1989. Metabolism of peripheral nerve monogalactosylceramides. Lipids 24:837-841. 32. Townsend, L. E., Benjamins, J. A. and Skoff, R. P. 1984. Effects of nonensin and colchicine on myelin galactolipids. J. Neurochem. 43:139-145. 33. Cuzner, L. M. and Davison, A. N. 1968. The lipid composition of rat brain myelin and subceUular fraction during development. Biochem. J. 106:29-34. 34. Agranoff, B. W. and Hajra, A. K, 1994. Lipids. Pages 97-116, in Siegel G. J., Agranoff B. W., Albers, R. W., MolinoffP. B. (eds):

Lemieux, Mezei, and Breckenridge Basic Neurochemistry: Molecular Cellular, and Medical Aspects, 5th. ed. New York: Raven Press. 35. Costantino-Ceccarini, E. and Suzuki, K. 1975. Evidence for the presence of UDP-galactose:ceramide galactosyltransferase in rat myelin. Brain Res. 93:358-362. 36. Koul, O., Chou, K. H. and Jmlgalwala, F. B. 1980. UDP-galactose-eeramide galactosyltransferase in rat brain myelin subfractions during development. Biochem. J. 186:959-969. 37. Ledeen, R. W. 1984. Lipid-metabolizing enzymes of myelin and their relation to axon. J. Lipid Res. 25:1548-1554. 38. Yahara, S., Sing,h, I. and Kishimoto, Y. 1980. Cerebroside and cerebroside III-sulfate in brain cytosol. Evidence for their involvement in myelin assembly. Biochim. Biophys. Acta. 619:177-185. 39. Yao, J. K. and Poduslo, J. F. 1988. Biosynthesis of neutral glucocerebroside homologues in the absence of myelin assembly after nerve transection. J. Neurochem. 50:630-638. 40. Yao, J. K., Windebank, A. J., Poduslo, J. F., and Yoshino, J. E. 1990. Axonal regulation of Schwann cell glycolipid biosynthesis. Neurochem. Res. 15:279-282. 41. Yao, J. K. and Yoshino, J'. E. 1994. Association of glucocerebroside homolog biosynthesis with Schwarm cell proliferation. Neurochem. Res. 19:31-35. 42. Johnson, W. J., Malhberg, F. H., Rothblat, G. H., and Philips, M. 1991. Cholesterol transport between cell and high-density lipoproteins. Biochim. Biophys. Acta. 1085:273-298. 43. Mendez, A., Oram, J. F. and Bierman, E. L. 1991. Protein kinase C as a mediator of high density lipoprotein receptor dependent efflux of intracellular cholesterol. J. Biol. Chem. 266:1010410111. 44. Thomas, P. K., Berthold, C. H. and Ochoa, J. 1993. Microscopic anatomy of the peripheral nervous system. Pages 28-91, in Dyck, P. J., Thomas, P. K., Griffin, J. W., Low, P. A. and Poduslo, J. F., (eds.): Peripheral Neuropathies, 3rd ed. 45. Boyles, J. K., Zoellner, C. D., Anderson, L. J., Kosik, L. M., Pitas, R. E., Weisgraber, K. H., I-Iui, D. Y., Mahley, R. W., GebickeHaerter, P. J., Ignatius, M. J., and Shooter, E. M. 1989. A role for apolipoprotein E, apolipoprotein AI and low density receptors in cholesterol transport during regeneration and remyefination of the rat sciatic nerve. J. Clin. Invest. 83:1015-1031. 46. Mockel, B., Zinke, H., Flaeh, R., Weib, B., Weiler-Butter, H., and Gassen, H. G. 1994. Expression of apolipoprotein A-I in porcine endothelium in vitro. J. Neurochem. 62:788-798.