study of myelin purity in relation to axonal contaminants

Cellular and Molecular Neurobiology, Vol. 1, No. 2, 1981

Study of Myelin Purity in Relation to Axonal Contaminants

James E. Haley , ~'2 Fred G. Samuels , 1 and Robert W. Ledeen 1'3

Received October 20, 1980; revised December 18, 1980; accepted December 22, 1980

Axonal remnants are considered a probable source o f contamination o f isolated myelin in view of the relatively tight axon-glial intercellular junction. Using the rabbit optic system to label specifically axonal components, we have found the levels o f such contaminants to depend on the myelin isolation procedure, the tissue source, and the nature o f the contaminant. A procedure employing repetitive treatments with EGTA was found to be highly effective in removing proline-labeled axonal proteins, the estimated upper limit o f such contamination being approximately 0.6-1.2% of the myelin protein. The standard isolation procedure o f Norton and Poduslo, supplemented with an additional discontinuous gradient step, proved equally effective in removing rapidly transported proteins from myelin isolated from the superior colliculus or lateral geniculate body. When the optic tract was the source, however, the EGTA procedure proved more effective in removing both rapidly and slowly transported proteins. Axonal gangliosides labeled with N-[SH]aeetylmannosamine were efficiently removed by both procedures, adding support to the proposition that gangliosides detected in isolated myelin are intrinsic to that membrane.

KEY WORDS: myelin; myelin purity; myelin proteins; myelin gangliosides.

INTRODUCTION

Elaboration of myelin isolation procedures many years ago (August et al., 1961; Laatsch et al., 1962; Autilio et al., 1964; Gerstl et al., 1966) led to identification of the major protein and lipid components in this membrane and, eventually, of numerous minor components as well (for review, see Norton, 1977). A common question pertaining to the latter is whether they are true myelin constituents or components of contaminating membranes. This problem has special pertinence, in the case of proteins, to the growing list of enzymes found to be present in myelin whose purity has been difficult to define. Examples of minor lipids are the gangliosides, which comprise

Departments of Neurology and Biochemistry, Albert Einstein College of Medicine, The Bronx, New York 10461.

2 Present address: Department of Ophthalmological Research, College of Physicians and Surgeons of Columbia University New York, New York 10032.

3 To whom all correspondence should be addressed.

175

0272-4340/81/0600~0175503.00/0 © 1981 Plenum Publishing Corporation

!76 Haley, Samuels, and Ledeen

0.5% or less of the total lipid in mammalian myelin (Suzuki et al . , 1967; Ledeen et al., 1973). Their distinctive tlc pattern along with other lines of evidence (Suzuki et al., 1967, 1968; Suzuki, 1970) previously suggested that they are intrinsic to myelin. However, the recent demonstration that an axolemma-enriched fraction of brain has an appreciable ganglioside concentration (DeVries and Zmachinski, 1980) requires reconsideration of the possibility that a portion of gangliosides previously attributed to myelin may instead belong to tightly adhering axolemmal fragments.

Myelin purity is central to a i~esolution of such questions, and we have therefore undertaken a study of myelin isolation procedures from the standpoint of efficiency in removing axonal contamination. We have used the established procedure of Norton and Poduslo (1973) as our starting point and assessed the benefit of adding a third discontinuous density gradient step to the procedure. We have also examined the composition of myelin isolated by a technique employing EGTA 4 as an agent for stripping away axonal membranes. This is based on the method developed by DeVries (1976) to obtain an axolemma-enriched fraction from mammalian white matter, following reports by others (Schlaepfer and Bunge, 1973; Blank et al., 1974; Yu and Bunge, 1975) that a low-calcium medium effectively loosens the intracellular junction between axon and myelin. Previous work in this laboratory (Haley and Ledeen, 1979; Samuels et al., 1979; Ledeen et al., 1980) provided evidence that myelin subjected to repetitive treatment with EGTA showed progressively lower levels of axonal components, as judged by residual radioactivity and an enzyme marker. A further rationale for the use of chelating agents was the observation of Elam (1978) that EDTA treatment of myelin from goldfish optic tecta removed a portion of radioactivity associated with axonal proteins.

Our approach in this study has been to selectively label axonal elements of the optic system through axonal transport of substances synthesized in the ganglion cell bodies of the retina and to use these as impurity markers. We have found both methods of myelin isolation to be effective in removing all but a very small fraction of axonal protein contaminants. Our results also provide evidence that the gangliosides present in isolated myelin do not derive from the axon.

MATERIALS AND METHODS

Materials

Forty-day-old Sprague-Dawley rats and 6-week-old New Zealand white rabbits were used as sources of myelin. Radioisotopes, obtained from New England Nuclear (Boston, Mass.), included N-[3H(G)]acetyl-D-mannosamine (2.8 Ci/mmol), L- [14C(U)]proline (270 mCi/mmol), and L-[4-3H(N)]proline (15.6 Ci/mmol). EGTA and N-tris[hydroxymethyl]methyl-2-aminoethanesulfonic acid (TES) were purchased from Sigma Chemical Co. (St. Louis, Mo.).

4 Abbreviations used: CNPase, 2',Y-cyclic nucleotidase-Y-phosphodiesterase; EGTA, ethylene glycol bis(f3- aminoethyl ether)-N,N'-tetraacetic acid; TES, N-tris [hydroxymethyl]methyl-2-aminoethanesulfonic acid; SC, superior colliculus; LGB, lateral geniculate body; OT, optic tract; tic, thin-layer chromatography; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel eleetrophoresis.

Study of Myelin Purity 177

Myelin Isolation Procedure

Rats were sacrificed by decapitation and rabbits by an overdose of Nembutal. Brains were rapidly removed and cooled, and the brain stems dissected for myelin isolation. All operations were carried out at 4°C. For the EGTA method, all centrifugations were performed with an SW-27 rotor at 82,500g (average). Solutions were buffered at pH 7.5 with 10 mM TES. The first objective was the isolation of myelinated axons in a salt-containing medium (DeVries et al., 1978). Approximately 9 g of brain-stem tissue was minced with a razor, mixed with 37 ml of buffered 1.2 M sucrose containing 150 mM NaC1, and dispersed in a Dounce homogenizer with a loose-fitting pestle. After centrifugation for 20 min the floating fraction, consisting of myelinated axons (DeVries et al., 1972), was removed and successively processed in a similar fashion in buffered salt solutions containing 1.0 M sucrose and, finally, 0.8 M sucrose. The final floating fraction was dispersed in 37 ml of 10 m M EGTA (previously adjusted to pH 7.5) using a Dounce homogenizer with a tight-fitting pestle, and the mixture was stirred in the cold for approximately 1 hr. Centrifugation for 30 min produced a pellet which was homogenized in 37 ml of buffered 0.8 M sucrose containing 1 mM EGTA and centrifuged again for 1 hr. The floating myelin fraction was removed with a spatula and dispersed as above in buffered 10 mM EGTA, and after centrifugation the recovered pellet was again treated with buffered 0.8 M sucrose/EGTA to obtain the second floating myelin fraction. A total of four such cycles of EGTA treatment and flotation was performed (including the initial osmotic shock), aliquots being taken at each stage for analysis. The final myelin fraction was washed four times with water and lyophilized to dryness before storing in the freezer over a desiccant. The yield was approximately 11 mg of myelin per brain stem.

Myelin was also isolated by the standard method of Norton and Poduslo (1973). For some experiments further purification was achieved by the use of an additional discontinuous sucrose gradient following the low-speed centrifugations (three gradients en toto); in this instance the pellet was dispersed in 0.85 M sucrose and overlayered with 0.32 M sucrose prior to centrifugation. The band forming at the 0 . 3 2 M / 0 . 8 5 M interface was washed four times with water and lyophilized to dryness.

Radiolabeling of Axons via Axonal Transport

To label axonal gangliosides, two groups of rabbits were injected in the vitreous bodies of both eyes with 75 uCi of N-[3H]acetylmannosamine in 25 #1 of saline. Five days later the animals were sacrificed, the optic tracts were dissected out, and myelin was immediately isolated by the two methods described above. Gangliosides were isolated from the myelin, quantified, and counted (see below). To label axonal proteins, two groups of six rabbits were injected in both eyes with 20 uCi of [IgC]proline in 25 ~1 of saline. Seven days later the same rabbits were again injected in both eyes with 80 uCi of [3H]proline in 25 ~xl of saline. The animals were sacrificed 2 days later and the labeled tissues removed by dissection: these included superior colliculi (SC), lateral geniculate bodies (LGB), and optic tracts (OT). Myelin was immediately isolated from pooled OT and SC, while LGB were stored overnight on ice

178 Haley, Samuels, and Ledeen

prior to myelin isolation. One set of rabbits was used to prepare EGTA-myelin, and the other standard (Norton-Poduslo) myelin as described above.

Ganglioside Methodology

Gangliosides were isolated from myelin by the previously described procedure (Ledeen et al., 1973). When smaller amounts of myelin (ca. 40 mg) were extracted, the DEAE-Sephadex procedure and subsequent operations were scaled down to one-third the original (Ledeen and Yu, 1978). In those runs dialysis was sometimes replaced with a Sephadex (3-50 column procedure, as described by Ueno et al. (1978). Sialie acid content was determined by gas-liquid chromatography (Yu and Ledeen, 1970). For thin-layer chromatography, ganglioside mixtures containing 10 #g sialic acid were applied to Merck HPTLC precoated plants with 200 #m silica gel 60 (EM Laboratories, Inc., Elmsford, N.Y.) which were developed with chloroform-metha- nol-0.2% aqueous CaC12 (50:40:10). Bands were visualized with resorcinol spray (Svennerholm, 1957) and scanned densitometrically by the procedure of Ando et al.

(1978).

Assay Procedures

Myelin proteins were quantified by the method of Lowry et al. (1951) after delipidation with ether-ethanol (3:2) and solubilization of the residue in 1% SDS. Lipid phosphorus was measured by the procedure of Marinetti et al. (1959), as modified by Norton and Autilio (1966). CNPase activity was assayed according to Prohaska et al. (1973); after isobutanol-benzene extraction of the phosphomolybdic acid, liberated phosphorus was determined by the colorimetric procedure of Lindberg and Ernster (1956). Radioactivities of myelin samples were measured after dissolving aliquots in 1.1 ml of Protosol containing 10% water, followed by 10 ml of cocktail containing 0.7% Permablend III in toluene. Ganglioside samples were dissolved in 0.8 ml water and then mixed with 10 ml Aquasol.

SDS-Polyacrylamide Gel Electrophoresis (PAGE)

Discontinuous SDS-PAGE was performed by the method of Maizel (1966) as modified by Greenfield et al. (1971) for myelin proteins. Myelin samples were delipidated by extracting several times with ether-ethanol (3:2) and the dried protein residues were solubilized in 1% SDS at a concentration of 1.5-2 mg protein/ml. Approximately 150 #g protein was subjected to SDS-PAGE with 15% polyacrylamide. The gels were stained in 1% fast green, destained, and scanned at 580 nm with a Gilford spectrophotometer.

RESULTS

The first set of experiments was designed to establish the general properties of EGTA-treated myelin, in comparison to standard myelin. As indicated in Table I for rat brain-stem myelin, the properties were very similar with respect to protein content,


Table I. Comparison of Myelin Properties a

EGTA Protein Weight ratio, protein treatments (% dry wt) to phospholipid b CNPasC

1 27 0.87 10.8 2 26 0.87 12.2 3 25 0.86 8.0 4 29 0.85 10.2

Standard myelin 27 0.86 11.5

aMyelin was prepared from rat brain stem by the two procedures outlined in Materials and Methods. bPhospholipid weights were calculated as lipid P x 25. c2',Y-Cyclic nucleotide-3'-phosphodiesterase in units of/~mol/min/mg protein.

protein to phospholipid ratio, and CNPase activity. Furthermore, those parameters did not change significantly with each subsequent E G T A treatment.

Despite the relative constancy of total protein, the gel patterns (Fig. 1) showed a progressive loss of some of the minor high molecular weight proteins with each E G T A treatment. Since such proteins comprised a small proportion of the total, their removal had little effect on total protein content. Quantification of the protein distributions through scanning of the gels is shown in Table II. The patterns are generally similar to those reported by Morell et al. (1973). Myelin subjected to four E G T A treatments had, compared to s tandard myelin, roughly 50% less of the minor high molecular weight proteins that migrated more rapidly ( H M W P f ) and more slowly (HMWP~) than the Wolfgram complex. This applied to both rat and rabbit brain-stem prepara- tions. The various major proteins, expressed as a percentage of the total, were unchanged or slightly higher in EGTA-myel in .

To study protein contaminants quantitatively, [~4C] proline and [3H]proline were injected 7 days apart and the animals sacrificed 2 days after the latter (see Materials and Methods). This schedule permitted comparison of contaminat ing axonal proteins that arrived by either slow or fast transport. Since background labeling with these precursors is negligible (Elam and Agranoff, 1971), both eyes of each rabbit could be employed. Myelin isolated immediately from freshly dissected SC showed no difference between the two procedures with respect to ~4C and only a moderate (24%)

Table II. Comparison of Myelin Protein Patterns =

Rat Rabbit

Standard EGTA Standard EGTA

Basic proteins 15 18 Basic proteinL 11 12 18 21 Intermediate 10 13 16 19 Proteolipid protein 29 33 30 35 HMWPf 9 5 9 5 Wolfgram 16 15 16 14 HMWPs 11 4 11 6

aProteins from myelin isolated by the standard and EGTA procedures were separated by SDS-PAGE and quantified by scanning. These correspond to gels D, E, F, and G in Fig. 1. Results are expressed as percentages of the total protein detected on the gel. Each value is the average of two scans of a single gel. Areas were measured by triangulation.

180 Haley, Samuels, and Leedeen

A B C D E F G Fig. 1. SDS-polyacrylamide gel electrophoresis of myelin proteins. Each gel contains 150 ~g of myelin protein, stained with fast green. Gels A-D are protein samples from rat brain-stem myelin representing successive EGTA treatments; E is protein from rat brain-stern myelin isolated by the standard (Norton- Poduslo) procedure; F and G are proteins from rabbit brain-stem myelin isolated by the EGTA (4x) and standard procedures, respectively. Arrow indicates the Wolfgram complex; the different migratory rates result from the samples not having been run in parallel.

Study of Myelin Purity

Table III.

Method of isolation

Myelin Contamination by Axonal Proteins ~ i

OT SC LGB

3 H 14 C 3 H 14 C 3 H 14 C

181

EGTA 7,380 7,110 3,280 1,280 3,430 1,890 Standard 11,750 15,100 4,060 1,160 4,350 1,900 Standard plus one gradient 11,030 15,000 3,760 1,170 3,770 1,700

i

aProteins in the rabbit optic system were labeled via axonal transport following intraocular injection of [3H]proline for fast transport and [~4C]proline for slow transport. Myelin was isolated from fresh SC and from LGB following storage at 0°C overnight. Results are expressed as dpm/mg myelin protein, converted from cpm on the basis of 46 and 9.7% efficiencies for ~4C and 3H, respectively. 3H radioactivities were corrected for ~4C spillover of 10.6%. Values are normalized to take account of variation in homogenate radioactivity. Results are from a single experiment.

difference with respect to 3H (Table III). The results were generally similar for LGB, whose myelin was isolated after allowing the tissues to stand in the cold overnight. Standard myelin subjected to an additional discontinuous gradient lost part of its "excess" 3H radioactivity. The greatest difference between the two procedures was seen for OT myelin, that prepared by the standard procedure having 59% (3H) and 112% (14C) more radioactivity than that prepared with EGTA. These figures represent normalized values which take account of variation in homogenate radioactivities. Most of this additional radioactivity was not removed from the standard myelin after purification with a third discontinuous gradient.

Table IV shows that appreciable radioactivity was removed with each EGTA treatment. Such myelin isolated from the SC and LGB had less than 3% of the original homogenate radioactivity for either isotope after the fourth EGTA treatment. Virtually the same results were obtained with standard myelin (data not shown). EGTA-myelin isolated from the OT contained 5 and 2.8% of the homogenate 3H and 14C radioactivities, respectively, while the corresponding figures for standard myelin were 9.2 and 7.3%. These values represent recoveries of radioactivity in the isolated myelin and are not corrected for myelin yield.

Similar experiments were carried out by labeling gangliosides through intraocular injections of N-[3H] acetylmannosamine. Such gangliosides were previously shown

Table IV. Progressive Removal of Axonal Contaminants from Myelin with Successive EGTA Treatments a

OT SC LGB

3 H 14 C 3 H 14 C 3 H 14 C

Homogenate 38,040 64,800 30,500 6,890 22,060 9,670 1st EGTA 62,300 60,200 8,210 2,590 9,050 5,350 2nd EGTA 15,600 13,300 6,410 2,130 5,940 3,460 3rd EGTA 10,300 8,740 5,040 1,650 4,310 2,350 4th EGTA 7,380 7,110 3,280 1,280 3,430 1,89'0 Recovered radioactivity (%) 5.0 2.8 1.3 2.2 1.9 2.4

aSamples from Table III, showing progressive purification with EGTA treatments. Footnote a, Table III, pertains here. Results are expressed as dpm/mg total protein in the case of homogenate, and dpm/mg myelin protein for the remainder. "Recovered radioactivity" compares myelin after the fourth EGTA treatment to the original homogenate.


(Ledeen et al., 1976) to undergo axonal transport in the rabbit optic system and would thus constitute another axonal marker. Myelin was isolated from the OT 5 days after injection and the gangliosides were found to have 40% more counts in standard myelin as compared to that isolated with EGTA, after normalizing for the difference in radioactivity of the two homogenates. Purification of the standard myelin with an additional discontinuous sucrose gradient reduced its radioactivity to approximately the same level as that of EGTA-myelin. Based on this amount of radioactivity, myelin was calculated to contain 5% of the radioactivity originally present in the OT homogenate. Comparison of ganglioside concentrations and patterns showed no difference between standard myelin and EGTA-myelin (Table V).

DISCUSSION

The widely used density gradient centrifugation procedure of Norton and Poduslo (1973) was originally shown to yield myelin of high purity, as judged by the low concentrations of such negative markers as nucleic acids and Na +, K+-ATPase. Because of the growing interest in minor constituents, we have sought to obtain additional information on myelin purity and have focused on axonal components which are thought to be a likely source of contamination due to the specialized junctions between the myelin sheath and its investing axon (for review, see Peters et al., 1976). Consistent with this view was the observation (DeVries et al., 1972) that axons isolated from myelinated fibers are devoid of axolemma, suggesting that this membrane is retained by the myelin. Our approach was to use the rabbit optic system to specifically label axonal proteins and gangliosides, in the same manner employed to study axonal transport of these substances (Karlsson and Sjostrand, 1971; McEwen and Grafstein, 1968; Ledeen et al., 1976). By measuring the residual radioactivities

Table V. Myelin Ganglioside Patterns ~

Rat Rabbit

Standard EGTA Standard EGTA Content b 43 ± 5 44 _+ 9 51 47

% Distribution

GM4 5.2 5.9 1.2 1.2 GM3 GMz 1.4 0.9 GMI 48.0 45.0 52.0 49.0 Go3 - - - - 5.8 5.3 Gola 6.9 6.6 7.4 8.3 Gvla 5.7 6.8 3.8 3.1 Go2 Golb 12.0 16.0 14.0 15.0 Grlb 13.0 12.0 9.7 11.0 GQlb 7.7 6.6 5.9 7.1

aGangliosides were obtained from myelin isolated by the two procedures. Distributional values were the average of at least two scans. GM3 and GM4 were unresolved. Symbols are those of Svennerholm (1963).

bMicrograms sialic acid per 100 mg myelin. Rat values are the average of three determinations ± SD.


representing these axonal constituents, we have found that the levels contaminating myelin depend on (1) the nature of the contaminant, (2) the tissue from which the myelin was isolated, and (3) the procedure employed for myelin isolation. The latter aspect was examined by comparing standard myelin (Norton and Poduslo, 1973) with myelin isolated by a procedure employing multiple treatments with EGTA, following the observation (DeVries, 1976)that this reagent facilitates stripping away of the axolemma.

The experimental protocol was designed to differentiate between protein contaminants arriving via fast and slow axonal flow. Thus, [3H]proline injected 2 days prior to sacrifice should appear only in rapidly transported proteins of the OT, SC, and LGB, whereas [14C]proline, injected 9 days prior to sacrifice, should appear in both slowly and rapidly transported proteins of the OT but only in the fast transport wave reaching the SC and LGB (Karlsson and Sjostrand, 1971). Hence virtually all 3H and 14C radioactivity detected in myelin from the SC and LGB represented axonal contaminants that arrived by fast transport. 5 The fact that such radioactivities represented only 1-3% of the homogenate levels indicated efficient removal of these substances by both isolation procedures. Myelin isolated by the standard procedure lost 10-15% of this residual radioactivity on being subjected to another density gradient purification, the final value being close to that of EGTA-myelin. Myelin isolated from LGB was not significantly more contaminated than myelin isolated from SC (as percentage homogenate), indicating that storage of dissected tissues overnight at 0°C before isolation did not have a deleterious effect.

When myelin was isolated from the optic tract, however, the EGTA procedure resulted in substantially less protein contamination than the standard procedure (Table III). Moreover, the difference was not significantly narrowed by an additional gradient purification of the standard myelin. It is not known why the OT should differ in this manner from the SC and LGB, although we can speculate that it might relate in part to the "tougher" quality of this tissue and the greater difficulty encountered in grinding and dispersing it. An additional factor could be the presence of slowly transported proteins, which are thought to account for the greater contamination by 14C than 3H (see below).

The amounts of 14C and 3H radioactivities remaining in OT myelin after the fourth EGTA treatment were 2.8 and 5%, respectively, of the homogenate. From these values alone it is difficult to estimate the absolute levels of contamination since it is not known what fraction of the nonmyelin protein corresponded to labeled (e.g., axonal) and nonlabeled (e.g., glial) material. However, a rough estimate of contamination can be made by comparison of radioactivities in the myelin fraction after the first and fourth EGTA treatments (Table IV). Although a considerable amount of radioactivity was removed by the first EGTA stage, the s p e c i f i c radioactivity of that preparation was comparable to or greater than that of the homogenate. It seems likely that those labeled proteins were components of the more tightly adhering axonal membranes (e.g., axolemma) whose removal was effected only gradually during subsequent EGTA treatments. Our assumption is that such contaminants comprised at most

s Using Willard's designation (Willard et al., 1974), this would correspond to group I and possibly group II proteins.


5-10% of the total protein after the first EGTA treatment. On that basis it may be calculated that the amount of radioactivity remaining after the fourth EGTA treatment corresponded to 5.9-11.8 Izg of axonal protein or 0.6-1.2% of the myelin fraction protein. Essentially the same estimate is obtained for both isotopes. Calcula- tions of this kind are not valid for SC or LGB owing to the dilution effect stemming from the large amount of myelin associated with unlabeled axons.

It may be noted that despite the high specific radioactivity of myelin at the first EGTA stage, its analytical properties (Table I) and overall gel pattern (Fig. 1) were similar to those of purified myelin. This indicated that the radioactivity was concentrated in a relatively small quantity of protein and that the assumption of 5-10% contamination at that stage is probably an upper limit.

The fact that the calculated level of contamination in OT myelin was the same for each isotope suggests that rapidly transported proteins were the principal contribu- tors in the final EGTA myelin. It may be noted that the 14C/3H ratio of OT homogenate decreased with the first EGTA treatment (0.86 ~ 0.51) and remained relatively constant throughout subsequent stages; such behavior is consistent with the removal of a large proportion of loosely attached ~4C-labeled (e.g.,slowly transported) protein at the first stage of purification. The SC and LGB, on the other hand, which would not have received slowly transported proteins, showed increases in the 14C/3H ratio between the homogenate and the first EGTA treatment. Since rapidly transported proteins are thought to belong primarily to membrane components (McEwen and Grafstein, 1968), these results are consistent with the hypothesis that tightly adhering axonal membrane fragments are the principal contaminants of myelin. Removal of such fragments would thus be reflected in the progressive loss of radioactivity and high molecular weight proteins observed here and also in the gradual reduction of Na+,K÷-ATPase demonstrated in an earlier study (Haley and Ledeen, 1979). Although the EGTA procedure appeared more effective than the standard procedure in stripping away this membrane for OT myelin isolation, the two methods gave equivalent results when the SC or LGB were the source. As rfientioned, the latter tissues did not contain radioactivity from slowly transported proteins and there is some indication that such proteins, while effectively removed from EGTA-myelin, did contribute contamination to standard myelin (see below).

Association of axonally transported protein radioactivity with isolated myelin was previously described by Elam (1974, 1975), Autilio-Gambetti et al. (1975), and Prensky et al. (1975). The fact that radioactivity was confined for the most part to high molecular weight proteins may be related to our observation that such proteins were reduced with EGTA treatment. Our results are similar in some respects to those of Elam (1974), who found relatively high specific radioactivity in myelin from goldfish optic tecta following intraocular labeling with [3H]proline. This represented rapidly transported proteins which, since the myelin-associated radioactivity was confined to the higher molecular weight proteins concentrated in a myelin subfraction of relatively high density, were concluded to be retained axonal components rather than proteins which became intrinsic to myelin itself. In another study by the same author (Elam, 1975), slowly transported proteins were found to impart even more radioactivity to tectal myelin, which had been isolated by the standard procedure. The present study similarly indicates that myelin obtained by that procedure can, in some


circumstances, show contamination by slowly transported proteins. This is inferred in the observation (Table III) that whereas OT myelin isolated by the standard procedure had 59% more 3H radioactivity than EGTA myelin, the relative 14C contamination was nearly twice as great (112%). Our results differ from those of Elam in showing much lower specific radioactivities in myelin relative to homogenate for both fast and slow transport, possibly due to the different animal models employed as well as variation in other experimental parameters.

Myelin gangliosides labeled with N-[3H]acetylmannosamine also showed a progressive loss of radioactivity with each EGTA treatment, the final level corresponding to approximately 5% of that present in OT homogenate. Optic tract myelin isolated by the standard procedure had gangliosides with 40% more radioactivity than those in EGTA myelin, but the difference disappeared when the former was subjected to a third density gradient. In this respect axonal gangliosides differed from axonal protein contaminants, suggesting that gangliosides may have a somewhat different distribution than the proteins that were labeled in the axonal membrane. This calls to mind recent evidence for heterogeneity of the axon membrane in myelinated fibers (Waxman and Foster, 1980). The fact that one myelin ganglioside preparation, despite 40% more radioactivity, showed virtually the same concentration and tlc pattern as the less-labeled preparation is taken to indicate that the radioactive portion comprised a minor part of the total gangliosides present in the myelin fraction. From this and the low recovery of original homogenate radioactivity, we infer that the portion of gangliosides in isolated myelin associated with axonal contaminants is small. This provides further evidence that the gangliosides detected in isolated myelin are intrinsic to that membrane. 6

Additional studies (not shown) have demonstrated that axonal lipids in general, labeled with [3H]glycerol or [14C]serine, behave similarly to the gangliosides and appear as contaminants in isolated myelin. Their levels were also reduced by EGTA treatment or, in the case of the standard procedure, purification with an additional sucrose gradient. For these substances, however, it becomes imperative to distinguish between lipids that bear label through association with axonal membranes and those that are genuine myelin components. In the latter case the lipids could enter myelin through direct transfer from the axon or by a reutilization mechanism involving resynthesis with labeled precursor taken up from the axon (Droz et a l . , 1978; Haley and Ledeen, 1979). Since neither gangliosides nor proteins appeared to participate in those processes, we infer that radioactivity in these components represents axonal material only.

In summary, this study has illustrated the beneficial effect of repeated EGTA treatment in reducing axonal contaminants to a negligible level and has shown that, in most instances, a similar result can be achieved with the standard Norton and Poduslo isolation procedure supplemented with one additional gradient purification step. The results also indicate that the axon contributes insignificantly to the gangliosides

6 The ganglioside concentrations for rat and rabbit myelin shown in Table V are lower than values reported for fully mature animals (e.g., Ledeen et al., 1980). This may be attributed to the developmentally related increases observed in some species (Yu and Yen, 1975).


de t ec t ed in i sola ted myel in , thus s t r eng then ing the a r g u m e n t t ha t these mino r l ipids

a re in t r ins ic to tha t m e m b r a n e .

ACKNOWLEDGMENT

This s tudy was suppor ted by P H S G r a n t s N S 03356, N S 04834, and N S 16181.

R E F E R E N C E S

Ando, S., Chang, N. -C. and Yu, R. K. (1978). High-performance thin-layer chromatography and densitometric determination of brain ganglioside compositions of several species. Anal. Biochem. 89:437-450.

August, C., Davison, A. N., and Williams, F. M. (1961). Phospholipid metabolism in nervous tissue. Biochem. J. 81:8 12.

Autilio, L. A., Norton, W. T., and Terry, R. D. (1964). The preparation and some properties of purified myelin from the central nervous system. J. Neurochem. 11:17-27.

Autilio-Gambetti, L., Gambetti, P., and Shafer, B. (1975). Glial and neuronal contribution to proteins and glycoproteins recovered in myelin fractions. Brain Res. 84:336-340.

Blank, W. F., Jr., Bartlett-Bunge, M., and Bunge, R. P. (1974). The sensitivity of the myelin sheath, particularly the Schwann cell-axolemmal junction, to lowered calcium levels in cultured sensory ganglia. Brain Res. 67:503-518.

DeVries, G. H. (1976). Isolation and axolemma-enriched fractions from bovine central nervous system. Neurosci. Lett. 3:117 122.

DeVries, G. H., and Zmachinski, C. J. (1980). The lipid composition of rat CNS axolemma-enriched fractions. J. Neurochem. 34:424-430.

DeVries, G. H., Norton, W. T., and Raine, C. S. (1972). Axons: Isolation from mammalian central nervous system. Science 172:1370 ~1372

DeVries, G. H., Matthieu, J. -M., Beny, M., Chicheportiche, R., Lazdunski, M., and Dolivo, M. (1978). Isolation and partial characterization of rat CNS axolemma enriched fractions. Brain Res. 147:339- 352.

Droz, D., DiGiamberardino, L., Koenig, H. L., Boyenval, J., and Hassig, R. (1978). Axon-myelin transfer of phospholipid components in the course of their axonal transport as visualized by radioautography. Brain Res. 155:347-353.

Elam, J. S. (1974), Association of axonally transported proteins with goldfish brain myelin fractions. J. Neurochem. 23:345-354.

Elam, J. S. (1975). Association of proteins undergoing slow axonal transport with goldfish visual system myelin. Brain Res. 97:303-315.

Elam, J. S. (1978). Dissociation of axonally transported proteins from myelin by ethylenediamine tetraacetate (EDTA). J. Neurochem. 31:351-353.

Elam, J. S., and Agranoff, B. W. (1971). Rapid transport of protein in the optic system of the goldfish. J. Neurochem. 18:375-387.

Gerstl, B., Rubinstein, L. -J., Eng, L. F., and Tavaststjerna, M. (1966). A neurochemical study of a case of sudanophilic leukodystrophy. Arch. Neurol. 15:603 614.

Greenfield, S., Norton, W. T., and Morell, P. (1971). Quaking mouse: Isolation and characteristics of myelin protein. J. Neurochem. 18:2119-2128.

Haley, J. E., and Ledeen, R. W. (1979). Incorporation of axonally transported substances into myelin lipids. J. Neurochem. 32:735 742.

Karlsson, J. -O., and Sjostrand, J. (1971). Synthesis, migration and turnover of protein in retinal ganglion cells. J. Neurochem. 18:749-767.

Laatsch, R. H., Kies, M. M., Gordon, S., and Alvord, E. C., Jr. (1962). The encephalomyelitic activity of myelin isolated by ultracentrifugation. J. Exp. Med. 115:777-788.

Ledeen, R. W., and Yu, R. K. (1978). Methods for isolation and analysis of gangliosides. In Research Methods in Neurochemistry (Marks, N., and Rodnight, R., Ed.), Plenum Press, New York, pp. 371 410.

Ledeen, R. W., Yu, R. K., and Eng, L. F. (1973). Gangliosides of human myelin: Sialosylgalactosylceram- ide (GT) as a major component. J. Neurochem. 21:829-839.


Ledeen, R. W., Skrivanek, J. A., Tirri, L. J., Margolis, R. K., and Margolis, R. U. (1976). Gangliosides of the neuron: localization and origin. In Ganglioside Function: Biochemical and Pharmacological Implications (Porcellati, G., Ceccarelli, B., and Tettamanti, G., Ed.), Plenum Press, New York, pp. 83-103.

Ledeen, R. W., Cochran, F. B., Yu, R. K., Samuels, F. G., and Haley, J. E. (1980). Gangliosides of the CNS myelin membrane. In Structure and Function of the Gangliosides (Svennerholm, L., Mandel, P., Dreyfus, H., and Urban, P. -F., Eds), Plenum Press, New York, pp. 167-176.

Lindberg, O., and Ernster, L. (1956). Determination of organic phosphorus compounds by phosphate analysis. In Methods of Biochemical Analysis (Gluck, D., Ed.), Interscience, New York, pp. 7-9.

Lowry, O. H., Rosebrougb, N. J., Farr, A. L., and Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193:265-275.

Maizel, J. V., Jr. (1966). Acrylamide-gel electrophorograms by mechanical fractionation: Radioactive adenovirus proteins. Science 151:988 990.

Marinetti, G. V., Erbland, J., and Stotz, E. (1959). The quantitative analysis of plasmalogens by paper chromatography. Biochim. Biophys. Acta 31:251-252.

McEwen, B. S., and Grafstein, B. (1968). Fast and slow components in axonal transport of protein. J. Cell Biol. 38:494-508.

Morell, P., Lipkind, R., and Greenfield, S. (1973). Protein composition of myelin from brain and spinal cord of several species. Brain Res. 58:510-514.

Norton, W. T. (1977). Isolation and characterization of myelin. In Myelin (Morell, P., Ed.), Plenum Press, New York, pp. 161-199.

Norton, W. T., and Autilio, L. A. (1966). The lipid composition of purified bovine brain myelin. J. Neurochem. 13:213-222.

Norton, W. T., and Poduslo, S. (1973). Myelination in rat brain: Method of myelin isolation. J. Neurochem. 21:749-757.

Peters, A., Palay, S. L., and Webster, H. DeF. (1976). The Fine Structure of the Nervous System: The Neurons and Supporting Cells', W. B. Saunders, Philadelphia, pp. 191 226.

Prensky, A. L., Fujimoto, K., and Agrawal, H. C. (1975). Are myelin proteins synthesized in retinal ganglion cells'?. J. Neurochem. 25:883-887.

Prohaska, J. R., Clark, D. A., and Wells, W. W. (1973). Improved rapidity and precision in the determination of brain 2', 3'-cyclic nucleotide 3'-phosphohydrolase. Anal. Biochem. 56:275-282.

Samuels, F. G., Haley, J. E., and Ledeen, R. W. (1979). Purity of myelin isolated in presence of EGTA. Abstracts, American Society for Neurochemistry, Tenth Annual Meeting, 89.

Schlaepfer, W. W., and Bunge, R. P. (1973). Effects of calcium ion concentration on the degeneration of amputated axons in tissue culture. J. Cell Biol. 59:456-470.

Suzuki, K. (1970). Formation and turnover of myelin ganglioside. J. Neurochem. 17:209-213. Suzuki, K., Poduslo, S. E., and Norton, W. T. (1967). Gangliosides in the myelin fraction of developing rats.

Biochim. Biophys. Acta 144:375-381. Suzuki, K., Poduslo, J. F., and Poduslo, S. E. (1968). Further evidence for a specific ganglioside fraction

closely associated with myelin. Biochim. Biophys. Acta 152:576-586. Svennerholm, L. (1957). Quantitative estimation of sialic acids. II. Colorimetric resorcinol-hydrochloric

acid method. Biochim. Biophys. Acta 24:604-611. Svennerholm, L. (1963). Chromatographic separation of human brain gangliosides. J. Neurochem.

10:613-623. Ueno, K., Ando, S., and Yu, R. K. (1978). Gangliosides of human, cat, and rabbit spinal cords and cord

myelin. J. Lipid Res. 19:863-871. Waxman, S. G., and Foster, R. E. (1980). Ionic channel distribution and heterogeneity of the axon

membrane in myelinated fibers. Brain Res. Rev. 2:205-234. Willard, M., Cowan, W. M., and Vagelos, P. R. (1974). The polypeptide composition of intra-axonally

transported proteins: Evidence for four transport velocities. Proc. Nat. Acad. Sci. USA 71:2183-2187. Yu, R. C. -P., and Bunge, R. P. (1975). Alterations in the peripheral myelin sheath and node of Ranvier

produced by treatment with trypsin. J. Cell Biol. 64:1-14. Yu, R. K., and Ledeen, R. W. (1970). Gas-liquid chromatographic assay of lipid-bound sialic acids:

Measurement of gangliosides in brain of several species. J. Lipid Res. 11:506-516. Yu, R. K., and Yen, S. I. (1975). Gangliosides in developing mouse brain myelin. J. Neurochem.

25:229-232.

study of myelin purity in relation to axonal contaminants

Documents