SIMULTANEOUS ISOLATION OF PURIFIED MICROSOMAL and MYELIN FRACTIONS FROM RAT SPINAL CORD

Abstract— The fraction that sediments between 2 × 10⁵ g -min and 6 × 10⁶ g -min from dilute dispersions of rat brain in 0.32 m -sucrose is a microsomal fraction with very little contamination by myelin. A crude microsomal fraction prepared in the same way from rat spinal cord contains more myelin than microsomes. Centrifugation of the crude microsomal fraction in 0.85 m -sucrose gave a floating fraction, an infranatant fraction (purified microsomes) and a small pellet. The purified microsomes contained very little myelin as judged by electron microscopy and polyacrylamide gel electrophoresis. The lipid composition resembled that of spinal cord myelin except that the purified microsomes contained relatively less cholesterol and ethanolamine plasmalogens. The content of galactolipids was much greater in spinal cord microsomes than in brain microsomes. The spinal cord CDP-ethanol-amine:diglyceride ethanolaminephosphotransferase activity (EC 2.7.8.1) was concentrated in the purified microsomes.
A spinal cord myelin fraction isolated from the 2 × 10⁵ g -min pellet was quite pure as judged by electron microscopy, enzyme activities and polyacrylamide gel electrophoresis. No NADPH-cyto-chrome c reductase activity (EC 1.6.2.3) could be detected in the purified myelin. The ethanolaminephosphotransferase specific activity was about 5% of that found in the purified microsomal fraction. The protein content was 25% by weight for spinal cord myelin and 31% for brain myelin. Of the total spinal cord 2',3'-cyclic nucleotide-3'-phosphohydrolase activity, 16% was lost from the crude myelin during purification, 21% was recovered in the purified myelin, and 11% was found in the floating fraction from the crude microsomes. The purified myelin and microsomal fractions from spinal cord were relatively pure. Additional myelin was recovered in the floating fraction from the crude microsomes.     

UDP-galactose-ceramide galactosyltransferase in rat brain myelin subfractions during development.

The localization and activity of the enzyme UDP-galactose-hydroxy fatty acid-containing ceramide galactosyltransferase is described in rat brain myelin subfractions during development. Other lipid-synthesizing enzymes, such as cerebroside sulphotransferase, UDP-glucose-ceramide glucosyltransferase and CDP-choline-1,2-diacylglycerol cholinephosphotransferase, were also studied for comparison in myelin subfractions and microsomal membranes. The purified myelin was subfractionated by isopycnic sucrose-density-gradient centrifugation. Four myelin subfractions, three floating respectively on 0.55 M- (light-myelin fraction), 0.75 M- (heavy-myelin fraction) and 0.85 M-sucrose (membrane fraction), and a pellet, were isolated and purified. At all ages, 70--75% of the total myelin proteins was found in the heavy-myelin fraction, whereas 2--5% of the protein was recovered in the light-myelin fraction, and about 7--12% in the membrane fraction. Most of the galactosyltransferase was associated with the heavy-myelin and membrane fractions. Other lipid-synthesizing enzymes studied appeared not to associate with purified myelin or myelin subfractions, but were enriched in the microsomal-membrane fraction. During development, the specific activity of the microsomal galactosyltransferase reached a maximum when the animals were about 20 days old and then declined. By contrast the specific activity of the galactosyltransferase in the heavy-myelin and membrane fractions was 3--4 times higher than that of the microsomal membranes in 16-day-old animals. The specific activity of the enzyme in the heavy-myelin fraction sharply declined with age. Chemical and enzymic analyses of the heavy-myelin and membrane myelin subfractions at various ages showed that the membrane fraction contained more proteins in relation to lipids than the heavy-myelin fraction. The membrane fraction was also enriched in phospholipids compared with cholesterol and contrined equivalent amounts of 2':3'-cyclic nucleotide 3'-phosphohydrolase compared with heavy- and light-myelin fractions. The membrane fraction was deficient in myelin basic protein and proteolipid protein and enriched in high-molecular-weight proteins. The specific localization of galactosyltransferase in heavy-myelin and membrane fractions at an early age when myelination is just beginning suggests that it may have some role in the myelination process.     

UDP-Galactose: Ceramide galactosyltransferase of rat central nervous system myelin during development

The activity of UDP-galactose: hydroxy fatty acid containing ceramide galactosyltransferase was studied in the myelin and microsomal fractions of rat cerebral hemispheres, cerebellum and spinal cord during development. In all three regions, the specific activity of the enzyme reached a maximum in myelin prior to that in the microsomal membranes. This temporal relationship between myelin and microsomal fraction was similar in all the three regions, although the overall timing was shifted corresponding to known differential timing of myelin deposition in these regions. The activity of the enzyme from both the membranes, during development, increased in parallel with temperature up to 45°C. Specific localization of galactosyltransferase in early myelin may suggest specific role of the enzyme in the myelination process.     

SUBCELLULAR DISTRIBUTION AND SOME PROPERTIES OF ACETYL-COENZYME A HYDROLASE IN THE BRAIN

—The detailed subcellular distribution and some properties of acetyl-CoA hydrolase were studied in the rat brain. The brain homogenate (S₁) hydrolysed acetyl-CoA at a rate of approx 2·3 nmol/min/mg of protein at 37°C. The total activity of acetyl-CoA hydrolase was distributed in the following order: soluble > mitochondrial > microsomal, synaptosomal > myelin fraction. The order of the specific activity of the enzyme was: soluble, microsomal > mitochondrial > synaptosomal > myelin fraction. The synaptic vesicle fraction (D) had relatively high specific activity among the intraterminal particulate fractions, having two or three times higher specific activity than that of the synaptic cytoplasmic membrane fraction (F or G). Attempts to de-occlude acetyl-CoA hydrolase in the particulate fraction showed that only the enzyme activity in the myelin fraction was increased markedly by the treatment with ether or Triton X-100. Lineweaver-Burk plots gave straight lines for each subcellular fraction and apparent K_m values for acetyl-CoA were between 0·1 and 0·2 mM. Neither diisopropyl fluorophosphate nor physostigmine at the concentration of 0·1 mm inhibited the enzyme activity.     

THE DISTRIBUTION AND BIOSYNTHESIS OF THE MYELIN -GALACTOLIPIDS IN THE SUBCELLULAR FRACTIONS OF BRAINS OF QUAKING AND NORMAL MICE DURING DEVELOPMENT

Abstract— The biosynthesis and accumulation of monogalactosyl diglyceride, galacto-cerebrosides and sulfatides were studied in the brain of quaking mouse during myelination. The specific activity of monogalactosyl diglyceride synthesis of the mutant mouse was reduced to 50% of the control of the same age, comparable to the reduction in the biosynthesis of galactosylcerebrosides and sulfatides. The three galactolipids were largely associated with the myelin and microsomal fractions in the normal and quaking mice at the ages studied. Although the concentrations of microsomal galactolipids (expressed as nmol/g wet wt of brain) were lower in quaking mice than in the controls at all ages, the percentage of total brain monogalactosyl diglyceride recovered in the microsomes of the mutant mouse was always larger than in the microsomes of the controls. Between 16 and 41 days, the monogalactosyl diglyceride content of the control myelin increased 10-fold, whereas the concentrations in the mutant increased only 2-fold. In normal animals, the percentage of total myelin galactolipids in the 'small myelin' decreased over the age of 1841 days with concomitant increase in the 'large myelin'. In contrast, in the mutant, large percentages of these compounds remained associated with the small myelin even at late periods of myelin development. These findings indicate that the slow rate of deposition of myelin in the brain of quaking mouse may be due to a defective transport mechanism of the galactolipids from the site of synthesis (microsomes) to the site of deposition (myelin), or to a defect in the mechanism of final myelin assembly, rather than to a lipid-specific genetic error.     

Cholesterol ester hydrolase(s) in mammalian brain: Is there a myelin-specific cholesterol ester hydrolase?

The present study compared the properties of cholesterol ester hydrolase(s) in myelin and microsomes from rat, mouse and human brain. The results indicated that the enzyme activity in both myelin and microsomes from rat, mouse and human brain was optimal at pH 6.5 and required Triton X-100 for optimal activity. The enzyme activity in myelin was 3- to 4-fold higher in the presence of Trition X-100 than taurocholate. Addition of phosphatidyl serine enhanced (2 to 4 fold) the hydrolase activity in both myelin and microsomes. The properties of the enzyme in solubilized preparation of myelin were also similar to the properties of the enzyme in partially delipidated and solubilized preparations of microsomes. The activity was again optimal at pH 6.5, required Triton X-100 for optimal activity and was stimulated by phosphatidyl serine. These results indicate that the properties of cholesterol ester hydrolase in myelin are similar to those of the microsomal enzyme and that this is true for the fractions from both human and rodent brain. The data thus lead us to believe that the hydrolase activity in mammalian brain myelin and microsomes may reflect the distribution of a single enzyme in the two fractions rather than two distinct enzymes, one being specific to each fraction.     

UDP-glucose: ceramide glucosyltransferase of rat brain: an association with smooth microsomes and requirement of an intact membrane for enzyme activity

Subcellular distribution of rat brain UDP-glucose:ceramide glucosyltransferase, the enzyme which catalyses the first step during the sequential addition of carbohydrate moieties for ganglioside biosynthesis, was studied. The activity of the enzyme was highest in the fraction rich in microsomes. Subfractionation of crude microsomal fractions resulted in a further enrichment of the enzyme activity in the fraction which contained smooth microsomes, thus suggesting that the enzyme is associated with microsomal membranes. The enzyme does not appear to be associated with synaptosomes or myelin. Treatment of the microsomal fraction with phospholipase A and C or detergents resulted in the loss of enzyme activity. Preincubation of the microsomal fraction at 37 °C also resulted in a loss of enzyme activity. These results suggest the requirement of specific membrane structure for the activity of the enzyme UDP-glucose:ceramide glucosyltransferase of rat brain. The amount of the enzyme activity lost during preincubation was dependent on the composition of the incubation medium and the age of the rats from which microsomal fractions were obtained.     

Magnesium-dependent sphingomyelinase.

An enzyme which requires divalent metals and hydrolyses sphingomyelin to ceramide and phosphorylcholine is present in rat and human brain and practically absent from other organs. The greatest activity is associated with the microsomal fraction. It had an optimal pH at about 7.4, required magnesium or manganese ions and was completely inhibited by EDTA. Triton X-100 was required for optimal activity and this detergent could also be used to partly solubilize the enzyme from rat brain microsomes. Lecithin was hydrolyzed at only 2% of the corresponding rate of hydrolysis of sphingomyelin.     

Evidence for the Presence of CDP-Ethanolamine: 1,2-diacyl-sn-glycerol Ethanolaminephosphotransferase in Rat Central Nervous System Myelin

Highly purified rat brain myelin isolated by two different procedures showed appreciable activity for CDP-ethanolamine: 1,2-diacyl-sn-glycerol ethanolaminephosphotransferase (EC 2.7.8.1). Specific activity was close to that of total homogenate and approximately 12-16% that of brain microsomes. Three other lipid-synthesizing enzymes, cerebroside sulfotransferase, lactosylceramide sialyltransferase, and serine phospholipid exchange enzyme, were found to have less than 0.5% the specific activity in myelin compared with microsomes. Washing the myelin with buffered salt or taurocholate did not remove the phosphotransferase, but activity was lost from both myelin and microsomes by treatment with Triton X-100. It resembled the microsomal enzyme in having a pH optimum of 8.5 and a requirement for Mn2+ and detergent, but differed in showing no enhancement with EGTA. The diolein Km was similar for the two membranes (2.5-4 x 10(-4) M), but the CDP-ethanolamine Km was lower for myelin (3-4 x 10(-5) M) than for microsomes (11 - 13 x 10(-5 M). Evidence is reviewed that this enzyme is able to utilize substrate from the axon in situ.     

ISOLATION AND PROPERTIES OF THE PLASMA MEMBRANE OF KB CELLS

Plasma membranes from KB cells were isolated by the method of latex bead ingestion and were compared with those obtained by the ZnCl₂ method. Optimal conditions for bead uptake and the isolation procedure employing discontinuous sucrose gradient centrifugation are described. All steps of preparative procedure were monitored by electron microscopy and specific enzyme activities. The plasma membrane fraction obtained by both methods is characterized by the presence of the Na⁺ + K⁺-activated ATPase and 5'-nucleotidase, and contains NADPH-cytochrome c reductase and cytochrome b₅. The latter two enzymes are also present in lower concentrations in the microsomal fraction. Unlike microsomes which are devoid of the Na⁺ + K⁺-activated ATPase and which contain only traces of 5'-nucleotidase activity, the plasma membrane fraction contains only trace amounts of the rotenone-insensitive NADH-cytochrome c reductase but no cytochrome P-450, both of which are mainly microsomal components. Morphologically the plasma membrane fraction isolated by the latex bead method is composed of vesicles of 0.1–0.3 µm in diameter. On the basis of the biochemical and morphological criteria presented, it is concluded that the plasma membrane fraction isolated by the above methods are of high degree of purity.     

BRAIN CARBONIC ANHYDRASE: ACTIVITY IN ISOLATED MYELIN AND THE EFFECT OF HEXACHLOROPHENE

Abstract— Animals receiving hexachlorophene (HCP) in their diet develop cerebral edema with vacuolation of the myelin sheath. When carbonic anhydrase activities were measured in homogenates of brains from HCP-fed and control rats, the HCP-fed rats showed small decreases in the enzyme activity, but these changes were not statistically significant. HCP did inhibit the enzyme in vitro but at higher concentrations (10⁻⁵-10⁻⁴ m ) than have been reported for HCP levels in brains of experimental animals. Carbonic anhydrase activity was present in myelin preparations obtained by gradient centrifugation and osmotic shock or by subcellular fractionation. When the latter procedure was used, myelin carbonic anhydrase had a specific activity which was higher than that of the mitochondrial fraction. The myelin enzyme was inhibited by 10⁻⁹10⁻⁸ m -acetazolamide and, like the homogenates and the commercial enzyme, was inhibited by HCP. The mechanism for HCP toxicity remains unknown, but this study does suggest that carbonic anhydrase is an intrinsic component of the myelin sheath.     

PROPERTIES OF THIAMINE DI- AND TRIPHOSPHATASES IN RAT BRAIN MICROSOMES: EFFECTS OF CHLORPROMAZINE

Abstract— The mechanism of the action of chlorpromazine on rat brain thiamine phosphatases were studied to clarify the properties of these enzymes in the CNS. Chlorpromazine at concentrations of 0.25-1.0 m m caused marked decrease of microsomal and soluble thiamine triphosphatase (TTPase) activities and marked increase of microsomal thiamine diphosphatase (TDPase) activity. Imipramine and desipramine also inhibited TTPase but did not cause any marked change in TDPase activities. Addition of chlorpromazine (0.5 m m ) decreased the V_max of microsomal TTPase by about one-half, increased that of TDPase about 3-fold, and lowered the K _m value for TDP but not for TTP.
Acetone treatment of the microsomal fraction lowered the TTPase activity and markedly enhanced the TDPase activity. In acetone-treated microsomes, chlorpromazine also inhibited TTPase activity but did not activate TDPase. Deoxycholate had similar effects to chlorpromazine on these enzyme activities.     

Distribution of lipid synthesizing enzymes, 2′,3′-cyclic nucleotide 3′-phosphodiesterase,and myelin proteins in rat forebrain subfractions during development

The distribution of UDP-galactose: ceramide galactosyltransferase (CGalT) was studied in subcellular fractions of rat forebrain during development using zonal centrifugation on linear gradients. Specialized subfractions: SN 1, a microsomal fraction, SN 4, a myelin-related fraction, and purified myelin were also used for this study. For comparison, two microsomal lipid synthesizing enzymes, a myelin-specific enzyme, 2,3-cyclic nucleotide 3-phosphodiesterase and myelin proteins were measured in the same subfractions. UDP-glucose: ceramide glucosyltransferase and cerebroside sulfotransferase were confined to microsomes. CGalT was ferase and cerebroside sulfotransferase were confined to microsomes. CGalT was localized in microsomes, but also in myelin and myelin-related fractions. The developmental change in distribution of CGalT in adult animals toward myelin containing fractions could indicate that the replacement of galactosylceramide in compact myelin could be carried out in close proximity to compact myelin (mesaxon, paranodal loops) rather than in the distant oligodendrocyte perikaryon.     

Development of cholinephosphotransferase in guinea pig lung mitochondria and microsomes

Development of mitochondrial and microsomal choline phosphotransferase in the fetal guinea pig lung was investigated. The activity in fetal mitochondria was more than twice of that in fetal microsomes. However, in adult lung, the enzyme was distributed mostly in microsomes. In fetal lung, both the mitochondrial and microsomal enzyme activity was greatest at approx. 81% of the total gestation period (55 days). The specific activity in the microsomal fraction then declined until term, but increased again in the 24-h newborn from 1.0 to 2.3 nmol/min per mg protein. The activity in the mitochondrial fraction declined after 61 days (2.8 nmol/min per mg) to a minimal level at term (0.6 nmol/min per mg). Although the enzyme activity decreased from day 55 (1.2 nmol/min per mg), the amount of phosphatidylcholine gradually increased between day 55 and term.     

Localization and solubilization of a rat liver microsomal carnitine acetyltransferase.

Carnitine acetyltransferase activity had been previously shown to occur in peroxisomes, mitochondria, and a membranous fraction of rat and pig hepatocytes. When components of this third subcellular fraction (plasma membranes, components of the Golgi apparatus, and microsomes) were further separated, carnitine acetyltransferase fractionated with the microsomes. Microsomes isolated by three different methods (isopycnic sucrose density zonal centrifugation, high-speed differential centrifugation, and aggregation with Ca²⁺ followed by low-speed differential centrifugation) all contained carnitine acetyltransferase activity. The lability of carnitine acetyltransferase in microsomes isolated by different methods and in different isolation media is reported.When total microsomes were subfractionated into rough and smooth components, carnitine acetyltransferase activity was found to the same extent in both and was tightly associated with the microsomal membrane. The microsomal enzyme was rapidly inactivated in 0.25 m sucrose or 0.1 m phosphate, but was stable for at least 2 weeks in 0.4 m KCl. Extensive treatment with high ionic strength salt solutions, 1% Triton X-100, or a combination of the two was used to solubilize microsomal carnitine acetyltransferase activity.Carnitine octanoyltransferase activity was also found in the microsomal fractions isolated by three different methods, but no carnitine palmitoyltransferase was detected in the microsomal fractions. It is proposed that microsomal carnitine acetyl- and octanoyltransferases could be involved in the transfer of acyl groups across the microsomal membrane, thereby providing a source of acetyl and other acyl CoA's at sites of acetylation reactions and synthesis.     

Further Characterization of a Myelin-Associated Neuraminidase: Properties and Substrate Specificity

A neuraminidase activity in myelin isolated from adult rat brains was examined. The enzyme activity in myelin was first compared with that in microsomes using N-acetylneuramin(alpha 2----3)lactitol (NL) as a substrate. In contrast to the microsomal neuraminidase which exhibited a sharp pH dependency for its activity, the myelin enzyme gave a very shallow pH activity curve over a range between 3.6 and 5.9. The myelin enzyme was more stable to heat denaturation (65 degrees C) than the microsomal enzyme. Inhibition studies with a competitive inhibitor, 2,3-dehydro-2-deoxy-N-acetylneuraminic acid, showed the Ki value for the myelin neuraminidase to be about one-fifth of that for the microsomal enzyme (1.3 X 10(-6) M versus 6.3 X 10(-6) M). The apparent Km values for the myelin and the microsomal enzyme were 1.3 X 10(-4) M and 4.3 X 10(-4) M, respectively. An enzyme preparation that was practically devoid of myelin lipids was then prepared and its substrate specificity examined. The "delipidated enzyme" could hydrolyze fetuin, NL, and ganglioside substrates, including GM1 and GM2. When the delipidated enzyme was exposed to high temperature (55 degrees C) or low pH (pH 2.54), the neuraminidase activities toward NL and GM3 decreased at nearly the same rate. Both fetuin and 2,3-dehydro-2-deoxy-N-acetylneuraminic acid inhibited NL and GM3 hydrolysis. With 2,3-dehydro-2-deoxy-N-acetylneuraminic acid, inhibition of NL was greater than that of GM3; however, the Ki values for each substrate were almost identical. GM3 and GM1 also competitively inhibited the hydrolysis of NL and NL similarly inhibited GM3 hydrolysis by the enzyme. These results indicate that rat brain myelin has intrinsic neuraminidase activities toward nonganglioside as well as ganglioside substrates, and that these two enzyme activities are likely catalyzed by a single enzyme entity.     

TURNOVER OF PHOSPHATIDYLCHOLINE IN MICROSOMES AND MYELIN IN BRAINS OF YOUNG AND ADULT RATS

We have tested the hypothesis that the turnover of phosphatidylcholine in subcellular fractions of rat brain is a function of the age at which this lipid is deposited. Rats, 60 days of age, were injected intracranially with [2-³H]glycerol and either [methyl-¹⁴C]choline (to label the base moiety) or [U-¹⁴C]glucose (to label acyl moieties). Littermates were killed up to 90 days after injection and brain microsomes and myelin isolated. Lipids were extracted and the phosphatidylcholine was isolated by 2-dimensional TLC and hydrolyzed to its constituent moieties. The ³H in the glycerol backbone and ¹⁴C in the choline or acyl residues was quantitated. The microsomal and myelin ³H/¹⁴C ratios decreased with time with either set of precursors, indicating that labeled choline and acyl moieties were reutilized more efficiently than the glycerol backbone. The various precursors exhibited first order decay curves with half-lives for the glycerol backbone of 6 and 11 days for the microsomal and myelin fractions respectively. These results contrast with those previously obtained with identical experimental procedures when 17-day-old animals were injected. In that study, although much of the phosphatidylcholine turned over rapidly as for the older animals, by 2 weeks after injection most of the remaining phosphatidylcholine was turning over more slowly with a half-life of 13 and 25 days for microsomes and myelin respectively (Miller et al., 1977). The base and acyl moieties also had a corresponding shorter half-life in older animals relative to the slow turnover phase in younger rats.     

SUBCELLULAR AND SUBMICROSOMAL DISTRIBUTION OF GLYCOLIPID-SYNTHESIZING TRANSFERASES IN YOUNG RAT BRAIN

The subcellular and submicrosomal distributions of four glycolipid-synthesizing transferases were studied in young rat brains. (1) Two galactosyl transferases involved in the synthesis of cerebrosides, the cerebroside sulphotransferase which catalyses the synthesis of sulphatides, and the glucosyl transferase which plays an important role in the ganglioside biosynthesis were localized essentially in the microsomal fraction. Only low activities were detected in the crude mitochondrial and synaptosome-enriched fractions. (2) A comparison of the activities of these enzymes in the crude myelin and two myelin subfractions showed that the galactosyl transferases and the cerebroside sulphotransferase had similar activities in the crude myelin and myelin-like fractions. A considerable galactosyl transferase activity was found in purified myelin. In this respect these two enzymes were different from cerebroside sulphotransferase, whose activity was much lower in purified myelin. On the other hand, glucosyl transferase had a relatively low specific activity in all three myelin fractions. Analysis of different markers showed that the activities were considerably higher than those expected from the maximum microsomal contamination calculated. (3) Subfractionation of the microsomes demonstrated that the galactosyl transferases were more concentrated in the lower parts of the gradient, containing vesicles with attached ribosomes. Cerebroside sulphotransferase and glucosyl transferase were found predominantly in the upper and intermediate parts of the gradient, which were composed essentially of smooth-surfaced vesicles and membrane fragments. Chemical analysis of submicrosomal fractions confirmed the morphological observations.     

Studies on the latency of UDP-galactose-asialo-mucin galactosyltransferase activity in microsomal and Golgi subfractions from rat liver

The activity of UDPgalactose-asialo-mucin galactosyltransferase (EC 2.4.1.74) in microsomal and Golig subfractions was stimulated 2.4-fold after disruption of the membrane permeability barrier by hypotonic incubation. In the presence of Triton X-100, galactose transfer to asialo-mucin was increased 12-fold in rough microsomes and 5-fold in smooth microsomes both with and without hypotonic incubation; while in the Golgi subfractions no stimulation by detergent was observed. These experiments indicate differences in enzyme-lipid or enzyme-protein interactions in microsomes and Golgi membranes. Furthermore, these results strongly support the conclusion that the UDP-galactose-asialo-mucin galactosyltransferase activity in microsomal fractions is not due to contamination by Golgi vesicles but represents an enzyme activity endogenous to the endoplasmic reticulum.     

ISOLATION AND CHARACTERIZATION OF MYELIN-RELATED MEMBRANES

Abstract— Myelin related membrane fractions from rat brain and spinal cord were isolated from material normally discarded during standard myelin isolation procedures. A fraction which floated on 0.32 M-sucrose (F) and the material released after subjecting the myelin fraction to osmotic shock at two stages in the purification (W₁ and W₂) were characterized. These fractions were subjected to subfractionation on three step discontinuous sucrose gradients. Morphologically, the heavier subfrac-tions of W₁ and W₂ were shown to consist mainly of single membranes and vesicles. Sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis showed that, relative to myelin, proteolipid and basic protein were reduced in all subfractions, while the high molecular weight proteins were increased. The specific activity of 2′,3′-cyclic nucleotide 3′-phosphohydrolase (CNP) was up to 2-fold higher than that of myelin in the heavier subfractions of W₁ and W₂. The major myelin-associated glycoprotein was also increased in these subfractions as determined by periodic acid-Schiff staining. Differential centrifugation of the initial tissue homogenate to remove microsomes prior to myelin isolation gave rise to W₁ and W₂ subfractions with a CNP specific activity 3–4 times that of myelin. The high molecular weight proteins and glycoproteins were enriched in these microsome-depleted subfractions, but were qualitatively similar to those of myelin. Some of the membranes in these fractions may be derived from the continuum between the plasma membrane of the oligodendrocyte and compact myelin. Fraction F consisted of small membrane fragments and many vesicles, and was particularly deficient in proteolipid. The specific activity of CNP in fraction F was about the same as myelin, while the major myelin associated glycoprotein could not be detected. Fraction F from normal CNS tissue appears to be similar to the floating fractions previously isolated in larger amounts from pathological brain undergoing edematous demyelination.