Role of Lipoproteins in the Delivery of Lipids to Axons during Axonal Regeneration
1997; Elsevier BV; Volume: 272; Issue: 49 Linguagem: Inglês
10.1074/jbc.272.49.30766
ISSN1083-351X
AutoresElena I. Posse de Chaves, Antonio E. Rusiñol, Dennis E. Vance, Robert B. Campenot, Jean E. Vance,
Tópico(s)Peroxisome Proliferator-Activated Receptors
ResumoNerve fiber elongation involves the input of lipids to the growing axons. Since cell bodies are often a great distance from the regenerating tips, alternative sources of lipids have been proposed. We previously demonstrated that axonal synthesis of phosphatidylcholine is required for axonal growth (Posse de Chaves, E., Vance, D. E., Campenot, R. B. and Vance, J. E. (1995) J. Cell Biol. 128, 913–918; Posse de Chaves, E., Vance, D. E., Campenot, R. B. and Vance, J. E. (1995) Biochem. J. 312, 411–417). In contrast, cholesterol is not made in axons. We now show that when compartmented cultures of rat sympathetic neurons are incubated with pravastatin, in the absence of exogenously supplied lipids, cholesterol synthesis is inhibited and axonal growth is impaired. The addition of cholesterol to the axons or cell bodies of neurons treated with this inhibitor restores normal axonal elongation. Similarly, a supply of cholesterol via lipoproteins restores normal axonal growth. In contrast, lipoproteins do not provide axons with sufficient phosphatidylcholine for normal elongation when axonal phosphatidylcholine synthesis is inhibited. Thus, our studies support the idea that during axonal regeneration lipoproteins can be taken up by axons from the microenvironment and supply sufficient cholesterol, but not phosphatidylcholine, for growth. We also show that neither apoE nor apoA-I within the lipoproteins is essential for axonal growth. Nerve fiber elongation involves the input of lipids to the growing axons. Since cell bodies are often a great distance from the regenerating tips, alternative sources of lipids have been proposed. We previously demonstrated that axonal synthesis of phosphatidylcholine is required for axonal growth (Posse de Chaves, E., Vance, D. E., Campenot, R. B. and Vance, J. E. (1995) J. Cell Biol. 128, 913–918; Posse de Chaves, E., Vance, D. E., Campenot, R. B. and Vance, J. E. (1995) Biochem. J. 312, 411–417). In contrast, cholesterol is not made in axons. We now show that when compartmented cultures of rat sympathetic neurons are incubated with pravastatin, in the absence of exogenously supplied lipids, cholesterol synthesis is inhibited and axonal growth is impaired. The addition of cholesterol to the axons or cell bodies of neurons treated with this inhibitor restores normal axonal elongation. Similarly, a supply of cholesterol via lipoproteins restores normal axonal growth. In contrast, lipoproteins do not provide axons with sufficient phosphatidylcholine for normal elongation when axonal phosphatidylcholine synthesis is inhibited. Thus, our studies support the idea that during axonal regeneration lipoproteins can be taken up by axons from the microenvironment and supply sufficient cholesterol, but not phosphatidylcholine, for growth. We also show that neither apoE nor apoA-I within the lipoproteins is essential for axonal growth. During nerve regeneration, large amounts of lipids are required for remyelination and expansion of axonal membranes. Synthesis of new myelin by Schwann cells in the regenerating peripheral nerve has been extensively studied (1Gould R.M. Holshek J. Silverman W. Spivack W. J. Neurochem. 1987; 48: 1121-1131Crossref PubMed Scopus (26) Google Scholar, 2Ledeen R.W. Golly F. Haley J. Mol. Neurobiol. 1992; 6: 179-190Crossref PubMed Scopus (18) Google Scholar, 3Boiron F. Spivack W.D. Deshmukh D.S. Gould R.M. J. Neurochem. 1993; 60: 320-329Crossref PubMed Scopus (7) Google Scholar, 4Goodrum J.F. Earnhardt T. Goines S. Bouldin T.W. J. Neurosci. 1994; 14: 357-367Crossref PubMed Google Scholar). An interesting model for regeneration of injured peripheral nerve, involving apolipoprotein E (apoE) 1The abbreviations used are: apo, apolipoprotein; HDL, high density lipoprotein; LDL, low density lipoprotein; PtdCho, phosphatidylcholine; VLDL, very low density lipoprotein. and the coordinated storage and redistribution of cholesterol, has been proposed (5Mahley R.W. Science. 1988; 240: 622-630Crossref PubMed Scopus (3395) Google Scholar) and supported by experimental evidence (6Boyles J.K. Zoellner C.D. Anderson L.J. Kosic L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Harter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Crossref PubMed Scopus (446) Google Scholar). After peripheral nerve injury, axon degeneration and myelin destruction proceed rapidly. The degenerating nerve is infiltrated by blood-derived macrophages that are responsible for clearing axonal and myelin debris (4Goodrum J.F. Earnhardt T. Goines S. Bouldin T.W. J. Neurosci. 1994; 14: 357-367Crossref PubMed Google Scholar, 7Beuche W. Friede R.L. J. Neurocytol. 1984; 13: 767-796Crossref PubMed Scopus (318) Google Scholar, 8Perry V.H. Brown M.C. Gordon S. J. Exp. Med. 1987; 165: 1218-1223Crossref PubMed Scopus (526) Google Scholar). It has been proposed that most of the cholesterol, and possibly other key lipids released by degenerating axons and myelin, accumulate within Schwann cells and macrophages that remain in the area of degeneration. The salvaged cholesterol appears to be reutilized by the regenerating myelin membranes (4Goodrum J.F. Earnhardt T. Goines S. Bouldin T.W. J. Neurosci. 1994; 14: 357-367Crossref PubMed Google Scholar, 9Rawlins F.A. Hedley-White E.T. Villegas G. Uzman B.G. Lab. Invest. 1970; 22: 237-240PubMed Google Scholar, 10Rawlins F.A. Villegas G. Hedley-White E.T. Uzman B.G. J. Cell Biol. 1972; 52: 615-625Crossref PubMed Scopus (40) Google Scholar) and by neurons for axonal membrane regeneration. However, direct evidence for the reutilization of cholesterol by axons has not been provided. After nerve injury, synthesis of several proteins is induced in the distal, but not the proximal, segment of the injured nerve (11Müller H.W. Ignatius M.J. Hangen D.H. Shooter E.M. J. Cell Biol. 1986; 102: 393-402Crossref PubMed Scopus (56) Google Scholar). One protein in particular, apoE, is produced by infiltrating macrophages and accumulates to levels 100–200-fold greater than in uninjured nerve (12Skene J.H.P. Shooter E.M. Proc. Natl. Acad. Sci. U. S. A. 1983; 80: 4169-4173Crossref PubMed Scopus (137) Google Scholar, 13Ignatius M.J. Gebicke-Härter P.J. Skene J.H.P. Schilling J.W. Weisgraber K.H. Mahley R.W. Shooter E.M. Proc. Natl. Acad. Sci. U. S. A. 1986; 83: 1125-1129Crossref PubMed Scopus (496) Google Scholar, 14Snipes G.J. Skene J.H.P. Freeman J.A. Soc. Neurosci. Abstr. 1983; 9: 52Google Scholar, 15Müller H.W. Gebicke-Härter P.J. Hangen D.H. Shooter E.M. Science. 1985; 228: 499-501Crossref PubMed Scopus (112) Google Scholar). Synthesis of apoE in neurons per se has not been detected. 2M. Bussière, and E. Posse de Chaves, unpublished results. It has been proposed that apoE, together with apoA-I, which enters the nerve from the circulation, play a central role in the reutilization of cholesterol (6Boyles J.K. Zoellner C.D. Anderson L.J. Kosic L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Harter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Crossref PubMed Scopus (446) Google Scholar, 16Ignatius M.J. Gebicke-Härter P.J. Pitas R.E. Shooter E.M. Prog. Brain Res. 1987; 71: 177-184Crossref PubMed Scopus (39) Google Scholar, 17Goodrum J.F. J. Neurochem. 1991; 56: 2082-2086Crossref PubMed Scopus (57) Google Scholar). Current evidence suggests that cholesterol from the cellular and myelin debris is first stored in endoneurial macrophages and is subsequently secreted by the macrophages to form cholesterol-rich, apoE/A-I-containing lipoproteins. When regeneration begins, the proximal stump of the nerve sends out numerous neurites. The tips of the neurites, as well as Schwann cells, express high levels of LDL receptors, which have been postulated to participate in uptake of these lipoproteins (6Boyles J.K. Zoellner C.D. Anderson L.J. Kosic L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Harter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Crossref PubMed Scopus (446) Google Scholar, 17Goodrum J.F. J. Neurochem. 1991; 56: 2082-2086Crossref PubMed Scopus (57) Google Scholar, 18Goodrum J.F. J. Neurochem. 1993; 60: 1564-1566Crossref PubMed Scopus (16) Google Scholar, 19Rothe T. Müller H.W. J. Neurochem. 1991; 57: 2016-2025Crossref PubMed Scopus (28) Google Scholar). More recently, studies on mice lacking functional apoE, apoA-I, or both apoE and apoA-I genes revealed that neither apoE nor apoA-I is required for nerve regeneration in the peripheral nervous system (20Popko B. Goodrum J.F. Bouldin T.W. Zhang S.H. Maeda N. J. Neurochem. 1993; 60: 1155-1158Crossref PubMed Scopus (65) Google Scholar, 21Goodrum J.F. Bouldin T.W. Zhang S.H. Maeda N. Popko B. J. Neurochem. 1995; 64: 408-416Crossref PubMed Scopus (42) Google Scholar). These data suggest that cholesterol reutilization by Schwann cells can occur in the absence of these apolipoproteins, favoring the idea that considerable redundancy is built into the lipoprotein-assisted process of cholesterol reutilization (20Popko B. Goodrum J.F. Bouldin T.W. Zhang S.H. Maeda N. J. Neurochem. 1993; 60: 1155-1158Crossref PubMed Scopus (65) Google Scholar, 21Goodrum J.F. Bouldin T.W. Zhang S.H. Maeda N. Popko B. J. Neurochem. 1995; 64: 408-416Crossref PubMed Scopus (42) Google Scholar). Although the assumption has been that lipoprotein-derived cholesterol is used for nerve regeneration, the use of cholesterol from this source for axonal growth has not been directly demonstrated. Moreover, the fate of other lipoprotein-derived lipids, such as phospholipids, in relation to axonal regeneration is unknown. The model proposed above for peripheral nerve regeneration by reutilization of exogenously supplied cholesterol departs from the generally accepted idea that nearly all membrane materials (i.e. proteins and lipids) required for axonal growth are synthesized in the cell bodies and transported by anterograde transport mechanisms into the axons, where these molecules are utilized for the assembly of new membranes (22Ledeen R.W. Eichberg J. Phospholipids in Nervous Tissues. John Wiley & Sons, Inc., New York1985: 135-172Google Scholar, 23Alberts B. Bray D. Lewis J. Roberts Watson J.D. Molecular Biology of the Cell. 2nd Ed. Garland Press, New York1989: 1059-1136Google Scholar). It is now clear that axons of rat sympathetic neurons synthesize significant amounts of the major phospholipids (PtdCho, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, and sphingomyelin) as well as fatty acids (24Vance J.E. Pan D. Vance D.E. Campenot R.B. J. Cell Biol. 1991; 115: 1061-1068Crossref PubMed Scopus (68) Google Scholar, 25Vance J.E. Pan D. Campenot R.B. Bussière M. Vance D.E. J. Neurochem. 1994; 62: 329-337Crossref PubMed Scopus (94) Google Scholar). Of the total amount of PtdCho present in axons, at least 50% is made locally in the axons. Indeed, axonal synthesis of PtdCho is essential for normal axonal growth (26Posse de Chaves E. Vance D.E. Campenot R.B. Vance J.E. J. Cell Biol. 1995; 128: 913-918Crossref PubMed Scopus (61) Google Scholar). In contrast, cholesterol synthesis does not occur in axons (25Vance J.E. Pan D. Campenot R.B. Bussière M. Vance D.E. J. Neurochem. 1994; 62: 329-337Crossref PubMed Scopus (94) Google Scholar). In the present study, we demonstrate that lipoproteins can be used by axons of rat sympathetic neurons as a source of cholesterol, but not PtdCho, for axonal elongation. We also show that the presence of apoE and/or apoA-I within the lipoproteins is not essential for axonal growth. [1-14C]Acetic acid, sodium salt (specific activity 58 mCi/mmol) was purchased from Amersham Canada (Oakville, Ontario, Canada). Water-soluble cholesterol (methyl-β-cyclodextrin-cholesterol) was provided by Sigma. Pravastatin was a gift from Dr. Shinji Yokoyama (University of Alberta). Thin layer chromatography plates (silica gel G) were obtained from BDH Chemicals (Edmonton, Alberta, Canada). Standard phospholipids and cholesterol were isolated from rat liver or purchased from Avanti Polar Lipids (Birmingham, AL). L15 medium without antibiotics was purchased from Life Technologies, Inc. Mouse 2.5S nerve growth factor was obtained from Cedarlane Laboratories Ltd. (Hornby, Ontario, Canada). Rat serum was provided by the University of Alberta Laboratory Animal Services. Polyclonal antibodies directed against human apoB were purchased from Boehringer Mannheim (Germany). Polyclonal antibodies directed against rat apoE and apoA-I were generated in rabbits and characterized in our laboratories by standard procedures (27Rusiñol A. Verkade H. Vance J.E. J. Biol. Chem. 1993; 268: 3555-3562Abstract Full Text PDF PubMed Google Scholar). The reagents used for electrophoresis were supplied by Bio-Rad. Polyvinylidene difluoride membranes were from Millipore Corp. All other reagents were obtained from Sigma or Fisher. Procedures for growth of rat sympathetic neurons in compartmented cultures have been previously reported (28Campenot R.B. Walji A.H. Draker D.D. J. Neurosci. 1991; 11: 1126-1139Crossref PubMed Google Scholar). Medium supplied to the compartment containing cell bodies (Fig. 1 a) was supplemented with 2.5% rat serum, 1 mg/ml ascorbic acid, 10 mm cytosine arabinoside, and 10 ng/ml nerve growth factor. Cytosine arabinoside was added to prevent the growth of nonneuronal cells, resulting in cultures that were purely neuronal in gross appearance. In all growth experiments, delipidated rat serum (29Yao Z. Vance D.E. J. Biol. Chem. 1988; 263: 2998-3004Abstract Full Text PDF PubMed Google Scholar) was used. Medium supplied to the axon-containing compartments contained 100 ng/ml nerve growth factor. After 6 days, cytosine arabinoside treatment was discontinued, and nerve growth factor was confined to the axon-containing compartments (30Campenot R.B. Draker D.D. Neuron. 1989; 3: 733-743Abstract Full Text PDF PubMed Scopus (40) Google Scholar). Culture medium was changed every 3–6 days. Cells were cultured for 14 days prior to the start of experiments. For center-plated cultures, the center compartment contained cell bodies with some proximal axons; the left and right compartments each contained growing distal axons without cell bodies (Fig. 1 b). Typically, the axons increased in length by ∼ 1 mm/day. In most experiments, as noted, cell bodies were plated in the left compartment, and the center and right compartments contained distal axons (Fig.1 c). When the neurons are plated in the left compartment, the center compartment is occupied by numerous distal axons that can be treated separately from cell bodies, and at the same time neurite elongation can be evaluated in the right compartment. One advantage of the three-compartment model for neuron culture is that metabolic events occurring in axons alone can be studied independently of those in cell bodies. This model is also applicable for studying axonal regeneration, since neurites can be mechanically removed from the side compartments and axonal extension can be accurately measured as neurites regenerate. Neurons were plated in the center compartment of compartmented dishes and cultured for 14 days. The radioactive cholesterol precursor [1-14C]acetic acid (10 μCi/ml) was added to the cell body-containing compartment. After incubation periods of 1 day and 2 days, the radioactive medium was removed, the cells were washed twice with cold phosphate-buffered saline, and cellular material was harvested by the addition of methanol/water (1:1, v/v) to the center and side compartments separately. The lipids were extracted by the addition of chloroform to a final chloroform/methanol/water ratio of 2:1:1 (v/v) (31Folch J. Lees M. Sloane-Stanley G.H. J. Biol. Chem. 1959; 226: 497-509Abstract Full Text PDF Google Scholar). The lipid samples were applied to thin layer chromatography plates, which were developed in the solvent system diisopropyl ether/acetic acid (96:4, v/v) using unlabeled cholesterol as carrier. The band corresponding to authentic cholesterol was scraped from the plate and radioactivity was measured. Radioactivity was normalized to total phospholipid mass measured in the same extract (32Chalvardjian A. Rudnicki E. Anal. Biochem. 1970; 36: 225-226Crossref PubMed Scopus (303) Google Scholar). To evaluate the effect of pravastatin on the incorporation of [1-14C]acetic acid into cholesterol, neurons (14 days old) cultured in 24-well dishes were incubated with medium containing [1-14C]acetic acid in the presence or absence of 50 μm pravastatin. Pravastatin was dissolved in water to make a 10 mm stock solution, which was added to culture medium to give the desired final concentration of inhibitor. Extraction of lipids and isolation of cholesterol was performed as indicated for compartmented cultures. VLDL, LDL, and HDL were isolated from human plasma by sequential ultracentrifugation on a benchtop Beckman TL-100 ultracentrifuge as described by Brousseau et al. (33Brousseau T. Clavey V. Bard J.M. Fruchart J.C. Clin. Chem. 1993; 39: 960-964Crossref PubMed Scopus (79) Google Scholar), which allows the separation to be completed in 1 day. HDL3 and HDL2 subfractions were further separated by heparin-Sepharose affinity chromatography (34Weisgraber K.H. Mahley R.W. J. Lipid Res. 1980; 21: 316-325Abstract Full Text PDF PubMed Google Scholar). The apolipoprotein content of the lipoprotein fractions was determined by sensitive Coomassie Blue staining (Sigma) and immunoblotting, using polyclonal antibodies directed against human apoB, rat apoE, and rat apoA-I. Proteins were separated by electrophoresis on 3–15% gradient polyacrylamide gels that contained 0.1% SDS and then transferred to polyvinylidene difluoride membranes for 12 h at 50 V, and immunoblotting was performed as described previously (27Rusiñol A. Verkade H. Vance J.E. J. Biol. Chem. 1993; 268: 3555-3562Abstract Full Text PDF PubMed Google Scholar). For measurement of axonal extension, distal axons were mechanically removed from left and right compartments in the case of center-plated neurons, or from the right compartment for left-plated neurons, with a jet of sterile distilled water delivered with a syringe through a 22-gauge needle. The water was aspirated, and the wash was repeated twice followed by the addition of fresh culture medium. This procedure, termed axotomy, effectively removes all visible traces of axons from the side compartments. Neurite growth was measured as described previously (35Campenot R.B. Stevenson R.B. Gallin W.J. Paul D.L. Cell-Cell Interactions: A Practical Approach. IRL Press, Oxford1992: 275-298Google Scholar). In previous studies using the compartment model for culture of neurons (25Vance J.E. Pan D. Campenot R.B. Bussière M. Vance D.E. J. Neurochem. 1994; 62: 329-337Crossref PubMed Scopus (94) Google Scholar), we detected no cholesterol biosynthesis, as measured by incorporation of [1-14C]acetic acid into cholesterol, in axons of rat sympathetic neurons. We have now shown that under normal culture conditions cholesterol is synthesized in the cell body-containing compartment and efficiently transported into the distal axons. Rat sympathetic neurons (14 days old), cultured in three-compartment dishes, were incubated with [14C]acetate in the cell body-containing compartment alone. Radiolabeled cholesterol was isolated from the cell body-containing compartment and also from the left and right distal axon-containing compartments after time intervals of 1–3.5 days, and radioactivity was measured. Since the amount of neuronal material was small, we were unable to measure the mass of cholesterol. Instead, in this and other experiments, we related the radioactivity in cholesterol to the mass of cellular phospholipids. Fig.2 shows that [1-14C]acetate had been efficiently incorporated into cholesterol in the cell body-containing compartment by 24 h, and some radiolabeled cholesterol had been transported into the distal axons. After 3.5 days, the amount of radiolabeled cholesterol in the distal axons corresponded to approximately 50% of total neuronal [14C]cholesterol present, and newly synthesized cholesterol had equilibrated throughout all compartments. This experiment implies that under these culture conditions cholesterol utilized for axonal membrane biosynthesis is derived from synthesis in the cell bodies, since no external source of cholesterol was provided to the axons. We next determined if inhibition of cholesterol synthesis would inhibit axonal growth. Since in several cell types cholesterol synthesis is reduced by inhibitors of the rate-limiting enzyme of cholesterol biosynthesis, 3-hydroxy-3-methylglutaryl-CoA reductase (36Tsujita Y. Kuroda M. Shimada Y. Tanzawa K. Arai M. Kaneko I. Tamaka M. Masuda H. Tamuri C. Watanabe Y. Fujii S. Biochim. Biophys. Acta. 1986; 877: 50-60Crossref PubMed Scopus (341) Google Scholar, 37Nakaya N. Homma Y. Tamachi H. Goto Y. Atherosclerosis. 1986; 61: 125-128Abstract Full Text PDF PubMed Scopus (43) Google Scholar), we examined the effect of one of these inhibitors, pravastatin, on cholesterol synthesis in rat sympathetic neurons. Neurons cultured in 24-well dishes for 14 days were incubated with [1-14C]acetate in the presence or absence of 50 μm pravastatin. Fig. 3shows that pravastatin had reduced the incorporation of [14C]acetate into cholesterol by approximately 83% at 24 h, demonstrating that cholesterol synthesis is inhibited by pravastatin in rat sympathetic neurons. The response of neurite growth to the presence of pravastatin and the corresponding reduction of cholesterol biosynthesis was investigated. Neurons were plated in the center compartment of compartmented culture dishes for 14 days. Distal axons were removed from the side compartments by axotomy, and the neurons were subsequently incubated with 50 μm pravastatin in either the center, cell body-containing compartment or the distal, axon-containing compartments. Axon extension was measured every day for the next 3.7 days (Fig. 4). Cultures that had been given pravastatin in the cell body-containing compartment alone had essentially stopped growing after 2.7 days. At all time points, axonal extension of pravastatin-treated neurons was statistically different from that of untreated neurons (p ≤ 0.008). As expected, since cholesterol synthesis is restricted to cell bodies, axons of cultures to which pravastatin had been added to the distal axon-containing compartments alone continued to elongate at the same rate as cells grown in the absence of pravastatin during the 3.7-day treatment period (Fig. 4). The effect of pravastatin on axonal growth was specifically related to the inhibition of cholesterol synthesis at the level of 3-hydroxy-3-methylglutaryl-CoA reductase, since the addition of mevalonic acid, the product of the reaction catalyzed by 3-hydroxy-3-methylglutaryl-CoA reductase, reversed the effect of pravastatin on axonal growth in a dose-dependent manner (Fig. 5). When mevalonic acid was added in the absence of pravastatin, axon growth was the same as in untreated cells (data not shown).Figure 5Mevalonic acid reverses the effect of pravastatin on axonal growth. Neurons (14 days old) were cultured as indicated in the legend to Fig. 4. Axons in the side compartments were axotomized, and medium containing either pravastatin (50 μm) alone or the indicated concentrations of mevalonic acid (100–500 μm) with pravastatin was given to the cell body-containing compartment (CB). Control cultures were given medium lacking both pravastatin and mevalonic acid. Axonal extension was measured 3 days after axotomy. Results are means ± S.E. of measurements from 60–64 tracks for each treatment. The experiment was repeated twice with similar results. Theasterisk signifies the statistical difference between cultures treated with pravastatin in the presence or absence of mevalonic acid (p < 0.01).View Large Image Figure ViewerDownload Hi-res image Download (PPT) We conclude from these experiments that active cholesterol synthesis is required for normal axonal elongation of cultured rat sympathetic neurons and that synthesis of cholesterol in cell bodies supplies sufficient cholesterol for normal axonal extension. Since in most cases in animals the tips of regenerating axons are localized at distances greatly removed from the cell bodies, cholesterol transport from the cell bodies to the regenerating sprout may not be the sole source of axonal cholesterol. An alternative scenario is that axons would "recycle" cholesterol salvaged from degeneration of both myelin and axons. We therefore investigated the ability of axons to utilize exogenous sources of cholesterol for elongation. Sympathetic neurons were plated in the left compartment of the three-compartment dishes (Fig. 1 c) and allowed to grow for 14 days. Distal axons were removed from the right compartment, and the axons were allowed to regenerate under the following conditions. Incubation of neurons with pravastatin in the left, cell body-containing compartment impaired axonal extension, in agreement with the data shown in Figs. 4 and 5. Incubation of neurons with pravastatin and water-soluble cholesterol (cholesterol incorporated into β-cyclodextrin) in the cell body-containing compartment, or with pravastatin in the cell body-containing compartment and cholesterol in the axon-containing compartments (center and right compartments), allowed normal axonal growth (Fig.6). (All values were not statistically different from controls at p < 0.1). Cultures incubated without pravastatin in all three compartments were used as controls (Fig. 6). This experiment demonstrates that sympathetic neurons can use exogenous cholesterol added to either cell bodies or axons for axonal membrane biogenesis. Cholesterol is normally delivered from the circulation to cells in the form of lipoproteins. The presence of lipoproteins has been demonstrated in the vicinity of regenerating and remyelinating nerves; however, no lipoprotein particles have been identified in noninjured nerves (6Boyles J.K. Zoellner C.D. Anderson L.J. Kosic L.M. Pitas R.E. Weisgraber K.H. Hui D.Y. Mahley R.W. Gebicke-Harter P.J. Ignatius M.J. Shooter E.M. J. Clin. Invest. 1989; 83: 1015-1031Crossref PubMed Scopus (446) Google Scholar). These lipoproteins contain apoE and apoA-I as the major or exclusive apolipoproteins. We investigated the ability of lipoproteins to provide cholesterol for axonal growth. Three different preparations of human plasma lipoproteins were used as sources of cholesterol for axonal regeneration of cultured sympathetic neurons: LDL containing apoB; HDL2 containing apoE, apoA-I, and apoA-II; and HDL3 with apoA-I and apoA-II but with no apoE. The apoprotein content of the lipoproteins was verified by immunoblotting with anti-apoB, anti-apoA-I, and anti-apoE antibodies (data not shown). Sympathetic neurons were plated in the left compartment and allowed to grow for 14 days. At that time, the cells were axotomized, and cholesterol synthesis was inhibited by the addition of pravastatin to the cell body-containing compartment. To some cultures, lipoproteins were supplied to the distal axon-containing compartments alone, whereas to other cultures lipoproteins were added to only the cell body-containing compartment. The concentration of lipoproteins was normalized in terms of total cholesterol (100 μg/ml of medium). A few cultures were given cholesterol (100 μg/ml medium) for comparison. Axonal extension was measured after 4.2 days. In pravastatin-treated cultures, axonal extension was 54% less than in untreated cultures (Fig. 7). Cholesterol and all of the lipoproteins, when added to the distal axon-containing compartments, were able to overcome pravastatin inhibition of axonal elongation (Fig. 7), even LDL that lacked apoE and apoA-I. However, when given to the cell body-containing compartment alone (Fig. 7), only LDL, but not HDL2 or HDL3, was as effective as cholesterol in restoring neurite growth. This experiment indicates that cholesterol supplied to the regenerating axons via exogenous delivery of lipoproteins satisfies the cholesterol requirement for axonal extension. Since HDL3 contains no apoE and LDL contains neither apoA-I nor apoE, we conclude that neither apoE nor apoA-I is required for delivery of lipoprotein cholesterol for growth of distal axons. Since lipoproteins contain not only cholesterol but also phospholipids, we tested the hypothesis that lipoproteins can provide PtdCho for axonal growth. We have previously shown that when axonal PtdCho synthesis is inhibited by incubation of distal axons with choline-deficient medium (26Posse de Chaves E. Vance D.E. Campenot R.B. Vance J.E. J. Cell Biol. 1995; 128: 913-918Crossref PubMed Scopus (61) Google Scholar) or with alkylphosphocholines (38Posse de Chaves E. Vance D.E. Campenot R.B. Vance J.E. Biochem. J. 1995; 312: 411-417Crossref PubMed Scopus (37) Google Scholar), neurite growth is strongly impaired. Axonal growth is also inhibited during global choline deficiency. However, decreased PtdCho synthesis in the cell bodies does not inhibit axon extension (26Posse de Chaves E. Vance D.E. Campenot R.B. Vance J.E. J. Cell Biol. 1995; 128: 913-918Crossref PubMed Scopus (61) Google Scholar). From these experiments, we concluded that PtdCho synthesis in distal axons, but not cell bodies, is required for normal axonal growth. Neurons plated in the left compartment of three-compartment dishes were incubated with medium lacking choline in all three compartments for 3 days (preincubation). The axons in the right compartment were then axotomized and allowed to regenerate in the presence or absence of choline or PtdCho-containing lipoproteins. Axonal extension was measured every day for 5 days (Fig. 8). In cultures depr
Referência(s)