- Email: [email protected]
Effects of anti-white matter serum on myelin and lipid synthesis in brain prisms
Brain Research, 173 (1979) 513-526 © Elsevier/North-Holland Biomedical Press
513
EFFECTS OF ANTI-WHITE MATTER SERUM ON MYELIN AND LIPID SYNTHESIS IN BRAIN PRISMS
EVELYN P. LAPIN, HOWARD S. MAKER, GERARD M. LEHRER, SULAMITH WEISSBARTH, CEDRIC S. RAINE, ANNE B. JOHNSON and MURRAY B. BORNSTEIN
Department of Neurology, Mount Sinai School of Medicine, New York, N. Y. 10029 and Department of Neurology, Albert Einstein College of Medicine, New York, N. Y. 10461 (U.S.A.) (Accepted January 4th, 1979)
SUMMARY
Tissue prisms prepared by chopping whole mouse brain maintained respiratory capacity and ultrastructural integrity for 3 h in vitro. Normal rabbit serum (ca. 25 70) caused no morphological change but inhibited the synthesis of galactolipids by the prisms. Heating the serum abolished the inhibition. Complement containing antiwhite matter rabbit serum destroyed myelin and inhibited galactolipid synthesis to a greater degree than did normal serum. Structures other than myelin were unaffected by the antiserum. Incubation in the presence of heated anti-white matter serum eliminated the myelin destruction but resulted in specific morphological changes characterized by the doubling of the myelin lamellae at the intraperiod line. Immunoperoxidase studies suggest specific binding of immunoglobulin to components of myelin located at the intraperiod line. These changes were similar to those found in organotypic cultures. Heated antiserum did not inhibit galactolipid synthesis but addition of complement (normal guinea pig serum) to the heated antiserum restored only that portion of the inhibition which exceeded that caused by normal serum. Heat labile factors in normal rabbit serum which inhibit myelin lipid synthesis in the prisms must be corrected for in studies in which the heating of serum is used to demonstrate that the effect is complement dependent. The prism system is simpler than that of organotypic cultures and may be useful in the study of myelinotoxic factors. INTRODUCTION
The sera of rabbits sensitized with whole white matter2,3,11,25, myelin25 or galactocerebroside8,18 but not with myelin basic protein alone26 contain complementdependent factors which reversibily demyelinate or inhibit myelination in organotypic cultures of fetal rodent central nervous system. While such cultures may represent an optimal system for detecting and quantifying demyelinating factors in sera of
514 experimental animals or humans suffering from demyelinating disease, they are not always available or convenient. Use of the tadpole optic nerve for detecting such factors may also present problems3L For some studies simpler methods would be helpful particularly if a large patient population is to be followed. SmithZ9, 30 and Smith and Hasinoff 31 have demonstrated that 300-500 mg, 0.5 mm Stadie-Riggs slices of young adult rat brain or spinal cord in 1 ml of medium can synthesize myelin lipids. However, during incubation basic protein was lost and myelin 'dissociation' was inferred from a change in its sedimentation properties 30. Pellkofer and Jatzkewitz 23 showed that the serum of rabbits immunized with myelin basic protein which does not normally demyelinate cultures 27 inhibited the incorporation of tyrosine into basic and proteolipid proteins and of 3sSO4 into cerebroside sulfates when 0.35 mm slices of 15 day of age rabbit spinal cord were incubated for 4 h. No ultrastructural studies of the slices were reported in any of these investigations. The present study used brain cylinders 1.5 mm long × 75 ,urn radius. These dimensions should allow complete oxygenation and substrate access when the medium is equilibrated with 95 % oxygen 36. Tissue respiration and ultrastructural integrity were maintained for up to 3 h. When cylinders were incubated with anti-white matter rabbit serum, complement dependent myelin breakdown as well as inhibition of myelin lipid synthesis were observed. Incubation with heated sera (depleted of complement) produced changes similar to those observed in tissue cultures under similar conditions. MATERIALS AND METHODS
Tissue preparation and respiration studies Tissue cylinders were prepared from the pooled brains of 2 or 3 15-22-day-old S-W albino mice. Animals were decapitated and their brains were rapidly removed and cut into coronal slices 1.5-2.0 mm thick. The slices were chopped with a Mcllwain tissue chopper (Brinkmann Instrument Inc., Westbury, N.Y.) 2o, set at 150 # m with a second cut perpendicular to the first to yield tissue prisms approximately 0.15 × 0.15 × 1.5 mm. Suspended in liquid the prisms became cylindrical. The cylinders were dispersed (2 ~ w/v) in gas equilibrated Krebs-Ringer bicarbonate buffered medium 9 containing 10 or 40 mM glucose, with and without 10 % fetal calf serum (Grand Island Biological Co., Grand Island, N.Y.) by gentle stirring at 4 °C and uniform samples were pipetted into Warburg Flasks. The flasks were gassed with 95 ~ oxygen/5 % CO2 for 10 min before measuring oxygen consumption at 37 °C for periods up to 6 h. Antiserum preparation Antiserum was prepared by inoculating 0.1 ml of whole bovine white matter in complete Freund's adjuvant 4 into the footpads of New Zealand rabbits and bleeding them upon the initial appearance of neurological signs of experimental allergic encephalomyelitis (EAE). Serum was used fresh without prior freezing or after heating at 56 °C for 30 min to destroy complement.
515
Effect of incubation on 2', 3'-cyclic nucleotide 3'-phosphohydrolase (CNP) (EC 3.1.4.16) For studies of the effect of antiserum on the activity of the myelin associated enzyme, CNP, cylinders were incubated at 0 or 37 °C for intervals of 0, 0.5, 1, 1.5, 2, 4, 6 or 8 h in Krebs-Ringer medium containing (final concentration) 16~ nutrient culture medium of Edith R. Peterson (33 ~o Eagles minimal essential medium plus glutamine, 7 ~ balance salt solution, 40 mM glucose, 1.3 #g/ml tetracycline), 5 guinea pig serum as a source of complement and either 12.5 ~ normal rabbit serum or anti-whole white matter serum. Incubation conditions were based on those developed for the study of organotypic cultures z. CNP in both tissue and medium was initially measured by the method of Olafson et al. z2 and later by the method of Sogin 33 adapted as a fluorometric method in our laboratory by S. Weissbarth (in preparation). Also tested was 32~ nutrient medium, 23~ normal or antiserum and 10~ guinea pig serum. Media were used unheated or heated at 56 °C for 30 min.
Galactolipid synthesis Tissue cylinders ( 2 ~ w/v) prepared from 15-19-day-old mouse brain were incubated in Krebs-Ringer bicarbonate medium containing 23 mM glucose and 12 or 23 ~ heated or unheated normal rabbit or anti-whole bovine white matter serum, with or without 10 ~ guinea pig serum as a source of complement. In one experiment D-[U14C]galactose, (New England Nuclear Corp., Boston, Mass.), final concentration 37 #M (2.5 #Ci in 270 #1 incubation medium) or Na2~5SO4, 27.6 #Ci/ml (Amersham Corp., Arlington Heights, Ill., 552 #Ci in 270 #1) were added to matched flasks. In other experiments only Na2SO4 incorporation was determined. Following a 10 min equilibration with 95 ~ oxygen/5 ~ CO2, vessels were incubated at 37 °C for 1 h in a metabolic shaker. The tissue was then washed 4 times with 1.0 ml aliquots of Krebs-Ringer medium containing 37 mM unlabeled galactose or sulfate. The sedimented tissue was resuspended in 250 #l of medium and sonicated (Model W140 Heat Systems-Ultrasonics Inc., Plainview, N.Y.) through 8-9 bursts of 0.3 sec each. Ten #1 aliquots of the sonicates were taken for protein determination17. The lipids were extracted from the remaining portion and partitioned by the method of Folch et al. 10 as modified by SuzukiTM The lower phase lipids were dried under a nitrogen stream, dissolved in 150/A of chloroform/methanol/water (70/30/4 by vol.) and chromatographed on silica gel G chromatography plates (Quantum Industries, Fairfield, N.J.) which had previously been washed with acetone, prerun with the developing solvent and activated. Chromatographically pure lipid standards of phosphatidyl choline (PC), phosphatidyl ethanolamine, cholesterol, cholesterol sulfate, sphingomyelin, cerebrosides and cerebroside sulfates (Applied Science Lab. Inc., State College, Pa.) individually and in combination were chromatographed simultaneously. The plates were developed in chloroform/methanol/water (70/30/4 by vol.). The lipid bands were scraped into scintillation vials. A mixture of toluene/Triton X/liquifluor (New England Nuclear Corp., Boston, Mass.) was added to each vial and the radioactivity assayed in a Packard Model 2450 liquid scintillation counter. All chemicals used were reagent grade.
516
Uhrastructural studies For electron microscopy, tissue cylinders were fixed for 30 min in cold 2.5 glutaraldehyde in pH 7.4 phosphate buffer and post-fixed in similarly buffered 1 osmic acid for 30 rain, dehydrated and embedded in Epon 24. One/zm Epon sections were taken and stained with toluidine blue for light microscopy. Based on examination of these, selected areas were trimmed from the blocks and thin sections cut for electron microscopy. Sections were placed on uncoated grids, double stained with lead and uranyl salts, and carbon coated. The specimens were coded with regard to treatment before fixation and all electron micrographs were prepared and examined prior to breaking the code. Immunohistochemical studies on prisms were performed essentially as in the previous studies with cultured nervous tissuOS, 16. Briefly, the prisms, suspended in Krebs-Ringer buffer, were incubated with 25 ~ (v/v) heated rabbit anti-white matter serum for 3 h at 37 °C, washed in buffer, fixed for 30 min in cold 5 ~ formalin/0.5 glutaraldehyde in Millonig's phosphate buffer, washed in phosphate buffered saline, and incubated with a 30-fold dilution of peroxidase-conjugated sheep anti-rabbit immunoglobulin Fab fragments (Pasteur Institute, 1.25 mg Fab per ml) for 1 h at room temperature. The prisms were then fixed in 2 . 5 ~ glutaraldehyde, reacted with diaminobenzidine medium, osmicated, dehydrated and embedded in Epon. Controls were run by substituting heated normal rabbit serum or buffer alone for the anti-white matter serum. Thin sections were studied without electron counterstains, or after contrasting with uranyl acetate. RESULTS
Viability of tissue cylinders Oxygen consumption by mouse brain tissue cylinders following equilibration at 37 °C is shown in Fig. 1. Initially high rates of activity were most likely due to the accumulation during tissue preparation of metabolic products that stimulate respi~" 6.0" 03 0')
7-- 5.0
__q 4.0 .-~ 3.0 E 2.0 Z
X 0
BO
120
IBO
240
MINUTES OF INCUBATION
Fig. 1. Respiratory rates (ml O2/g tissue/h) of tissue prisms prepared from the brains of 17-day-old mice and incubated (20 mg/ml) at 37 °C in Krebs-Ringer bicarbonate medium containing 10 m M glucose (©), 40 m M glucose (O), or 40 m M glucose in 63 ~ tissue culture nutrient medium (containing 2 ~ fetal calf serum) ( ~ ) . Oxygen consumption values are corrected for tissue and reagent blanks.
517 ration. A uniform rate of oxygen utilization was then maintained for 90 min at a level of 80 ~ of that of adult human brain and equal to that of 0.35 mm thick unstimulated guinea pig cortical sliceslL There was then a steady decline in 02 consumption and the tissue ceased to respire after 4-6 h in vitro. Only minor variations in the rate of deterioration of respiration were noted on repeated determinations. Increasing the glucose concentration from 10 to 40 mM (necessary for myelination of organotypic cultures) or adding fetal calf serum did not prolong the period of respiration. The ultrastructural integrity of the tissue was maintained for at least 3 h. Mitochondria and myelin lamellae showed only minor artefacts in buffer alone or in normal rabbit serum-enriched media (Fig. 3a). Incubation at 37 °C caused a small increase in the activity of tissue CNP (Fig. 2). The increase may have been due to improved accessibility of the enzyme to its substrate even above that resulting from sonification and use of 1.0 ~ of Triton X-100 (ref. 26). There was a lowered or no activation of the enzyme at 0 °C. No CNP activity was detected in the medium or in serum before incubation. Incubation caused the release of CNP into the medium within 30 min (37 mmol/gram tissue protein/h at 37 °C and 26 mmol/gram tissue protein/h at 0 °C). However, no further release occurred during incubations as long as 8 h. Tissue CNP activity increased from 90 -I- 10 mmol/gram protein/h in cylinders prepared from the brains of mice 15 days of age to 375 q- 25 mmol/gram protein/h at 27 days of age.
Normal rabbit serum Incubation for 3 h in medium containing decomplemented normal rabbit serum had no noticeable effect on myelin morphology (Fig. 3a). The slow deterioration of tissue respiration remained unchanged and the activity of CNP was stable for 6 h (Figs. 1 and 2) and as long as 8 h (not shown). However, as compared with medium alone, non-immune rabbit serum reduced the incorporation of z5SO4 into cerebroside sulfates by up to one half (Tables I and II). Incorporation of [U-14C]galactose into bJ t/) <[
0.-I nr",>_
I I NO SERUM-S7* J~ HEATED ANTISERUM-37* ANT SERUM-37*
0 "IQ.
o
200
t~ ".,.; 150 ]
:;...,~ ~ . . . . . . . ....::::,,~,.~
V-2 bJ
~ ~
I°° 1
-~--" ~-°
50 1
0
HOURS OF INCUBATION Fig. 2. 2',3'-cyclic nucleotide phosphohydrolase activities (mmol/g protein/h) of tissue prisms prepared from the brains of 17-day-old mice and incubated in Krebs-Ringer bicarbonate medium containing 26 mM glucose at 37 °C (solid lines) or 0 °C (broken lines) without serum (O), with 23.0 % heated (ll), or unheated (A) anti-white matter rabbit serum.
518
Fig. 3. a: brain tissue cylinders incubated for 3 h with heated (decomplemented) normal rabbit serum. A portion of a myelin sheath displays the normal periodicity of 11 nm. The intraperiod line is artefactually enhanced in density. The minor membrane discontinuities are preparation artefacts. Axon below, x 150,000. b: cylinder incubated for 3 h with unheated serum from a rabbit with EAE sensitized against whole white matter. A segment of myelin displays swelling of the outer layers. This swelling (arrows) occurs in the space of the intraperiod line. x 150,000. c: same preparations as b. A small area of a myelin sheath displays swelling at the intraperiod line (arrows), while the major dense line is intact, x 150,000. d: same preparation as b. A similar area shows advanced swelling and breakdown of myelin. × 150,000.
519 TABLE I
[14CJgalactose and zs SO a uptake into lipids in myelinating mouse brain prisms The significance of differences between all groups were calculated, but only those whose significance might be in doubt are shown. Significance of differences between means and that of medium without serum (a) ; between means and normal rabbit serum (b) ; between means and heated antiserum (c) are shown; single letter P < 0.05, double letters P < 0.01. Assays were performed in quadruplicate. Values in parentheses indicate standard deviation.
Serum added
Cerebroside [14CJgalactose (pmol/mg prot./h)
Cerebroside sulfate a5S04 (pmol/mg prot./h)
(A) None (B) Unheated rabbit
1.58 (0.20) 1.18 (0.30)
(C) Heated antiserum (---complement)
2.90 (0.27) aa; bb 1.15 (0.30) cc
14.0 (0.9) 7.19 (0.40) aa* 19.1 (1.6) aa; bb 8.75 (1.00) aa; cc
(D) Unheated antiserum (+complement)
* Incorporation of [14C-U]galactose into cerebrosides and [35SO4]sulfate into cerebroside sulfate (sulfatide) of tissue prisms incubated for 1 h at 37 °Cin Krebs-Ringer bicarbonate mediawith addition of 25 ~ serum and complement (10 % guinea pig serum) as indicated.
TABLE II
z5SO a uptake into cerebroside sulphate in myelinating mouse brain prisms The significance of differences between all groups were calculated, but only those whose significance might be in doubt are shown, a, significance of difference between means and that of medium A, without added serum; b, difference between means, and that containing heated normal rabbit serum B; c, difference between means and that containing unheated rabbit serum C; d, difference between means and that containing heated antiserum to bovine white matter D ; e, difference between means and that containing unheated antiserum E; f, difference between means and that containing heated normal rabbit serum plus guinea pig complement G. Single letter indicates P < 0.05 ; double letter P < 0.01. A and B were two separate studies with different tissue and serum preparations. Values in parentheses indicate standard deviation. N D = not determined.
Serum added
(A) None (B) Heated rabbit (C) Unheated rabbit (D) Heated antiserum (E) Unheated antiserum
Cerebroside sulfate synthesis (pmol/mg protein~h) A
B
6.18 (0.58) 9.43 (0.44) aa 4.79 (0.74) a; bb 9.74 (0.86) aa 2.94 (0.06)
3.95 (0.16) 4.65 (0.21) aa 2.49 (0.26) aa 5.40 (0.88) cc 1.26 (0.16)
CC
CC
(F) Heated rabbit (+complement)
ND
(G) Heated antiserum (+complement)
ND
(H) Unheated antiserum (+complement)
3.23 (0.45) c; dd
4.19 (0.70) cc; ee 3.51 (0.30) a; bb; cc; dd 1.37 (0.20) cc; dd; gg
520 TABLE III [14C]galactose uptake into (galacto)cerebroside and into phosphatidyl choline in myelinating mouse brain prisms (cpm/mg protein/h)
n - number of samples per assay. All other indications and symbols are explained in Tables I and 1I. Values in parentheses indicate standard deviation. ND = not determined. Experiment C (n = 3)
Experiment D (n == 4)
Cerebroside
PC
Cerebroside
PC
975 (124) ND
245 (68) ND
4779 (742) 6008 (514)
1669 (351) 1663 (254)
(C) Unheated rabbit
728 (184)
145 (88)
(D) Heated antiserum
1786 (169) aa ND ND ND
Serum added
(A) None (B) Heated rabbit
a
(E) Unheated antiserum (F) Heated rabbit (+-complement) (G) Heated antiserum (+ complement) (H) Unheated antiserum (+ complement)
275 (49) aa ND ND
2807 (815) aa 5146 (1429)
568 (151) aa 1522 (143)
989 (123) 4541 (664)
374 (71) 1120 (199)
ND
4780 (1043)
889 (120)
928 (74)
375 (145)
C
705 (188) dd
148 (27) d
cerebroside and cerebroside sulfate was reduced 25 ~ and 40 % respectively in one experiment and 40 ~ and 65 ~ respectively in a second experiment, but these decreases did not always reach statistical significance because of the greater variability of the galactose incorporation data. Heating the rabbit serum under the conditions used to destroy complement abolished the inhibition of galactolipid and PC synthesis (Tables I and lII). Heated rabbit serum actually increased the incorporation of 3~SO4 into sulphatide and of [U-14C]galactose into cerebroside above that found in tissue incubated in buffer alone. The addition of complement to heated normal rabbit serum appeared to compensate for some of the loss of heat-labile factor(s) in normal serum but did not restore either its sulfate or galactose incorporation inhibiting property (Table II). Anti-whole white m a t t e r serum
Anti-whole white matter serum containing complement caused wide separation, swelling and breakdown of myelin sheaths (Figs. 3, 4, 5) similar to that seen in myelinating organotypic cultures2L Heating the antiserum (destroying complement) eliminated the myelin breakdown. The activity of C N P fell slightly more rapidly in tissue cylinders incubated with antiserum than with normal serum (Fig. 2) but the difference was small and not consistent. C N P activity was not a reliable indicator of myelin disruption under the conditions of incubation. Longer exposure to the antiserum may have produced greater changes in CNP activity since this does occur in culture11, 35- Antiserum reduced the incorporation of [U-14C]galactose into galactolipid and of 35SO4 into cerebroside sulfate by more than 50 ~ below that in tissue incubated in heated (complement depleted) serum enriched medium (Tables I, II, III). Lipid synthesis in medium supplemented with heated antiserum was similar to that in medium containing heated normal serum. Part of the inhibition produced by the
521 u n h e a t e d a n t i s e r u m could be attributed to the heat labile factor(s) in n o n - i m m u n i z e d r a b b i t serum a n d was n o t i m m u n e related. Correcting for these n o n - i m m u n e factors reduced by one-half the a p p a r e n t i n h i b i t i o n of the antiserum. A d d i n g c o m p l e m e n t to the heated a n t i s e r u m restored only that p o r t i o n of the i n h i b i t i o n due to the c o m p l e m e n t dependent i m m u n o l o g i c a l factors. Myelin d i s r u p t i o n required both the specific i m m u n e factors a n d c o m p l e m e n t a n d was n o t seen within 3 h with n o r m a l r a b b i t or c o m p l e m e n t depleted antiserum.
Fig. 4. a: cylinder incubated for 3 h with heated (decomplemented) EAE rabbit serum. The myelin period has been increased from 11 nm to about 22 nm due to a swelling occurring in the space of the intraperiod line. Instead of two leaflets in this space, 4 leaflets can now be seen (arrows). x 150,000. b: similar preparation to a, A myelin sheath is incompletely swollen in that outer layers are increased to 22 nm (arrows), with the concomitant duplication of structure, and the innermost layers display the normal 11 nm periodicity, x 150,000.
522
Fig. 5. Immunostained myelin in prisms incubated sequentially with 25 ~ , heated, rabbit EAE serum (3h), peroxidase-labeled anti-rabbit immunoglobulin Fab fragments (1 h), and diaminobenzidine peroxidase medium (0.5 h) and then processed for electron miscroscopy. Reaction product is present at the intraperiod line (arrows) in A and B and on the outer oligodendroglial plasmalemma (double arrow) in A. Arrowheads point to major dense lines. Immunoreactive myelin is of the previously described, wide-spaced configuration (22 nm periodicity with 4 linear components to the intraperied line instead of two), and the alteration of the intraperiod line is evident in A in the two lamellae below the arrow and in B adjacent to the middle two arrowheads. A non-immunoreactive lamella of normal periodicity (11 nm) can be seen in B (double arrow). Uranyl acetate contrasting. A, x 152,000; B, × 142,000.
H e a t e d a n t i s e r u m did n o t cause myelin disruption, but h a d a n effect on the tissue cylinders identical to t h a t recently described in o r g a n o t y p i c cultures o f mouse spinal c o r d by Bornstein a n d R a i n e 4. I n c u b a t i o n with heated a n t i s e r u m (but n o t h e a t e d n o r m a l serum) consistently caused an increase in the spacing between the m a j o r dense lines a c c o m p a n i e d by the a p p e a r a n c e o f an a d d i t i o n a l p a i r o f electron dense leaflets replacing the i n t r a p e r i o d line, p r o d u c i n g 4 instead o f two leaflets between the m a j o r dense lines a n d a periodicity o f 22 n m instead o f the usual t 1 nm (Figs. 3 a n d 4). The overall effect was t h a t o f swelling o f the myelin sheaths. R e a c t i o n p r o d u c t indicating b o u n d i m m u n o g l o b u l i n originating in heated a n t i s e r u m was present a r o u n d myelinated axons a n d sometimes at the a r e a o f the i n t r a p e r i o d line. All i m m u n o p e r o x i d a s e stained myelin was o f the previously described wide-spaced configuration (Fig. 5). In
523 controls, no reaction product was observed around or within myelin lamellae. No oligodendrocytes were identified in the prisms. Sulfate incorporation was not significantly greater in tissue exposed to heated antiserum than to heated normal serum. Organotypic cultures exposed to heated antiserum for periods over 24 h show both morphological4 and biochemical (Lehrer et al., Brain Research, in press) evidence of increased lipid synthesis. DISCUSSION Tissue prisms, which assume a cylindrical shape of 75 #m radii when suspended in liquid, allow a better diffusion of oxygen and metabolites than do slices and possess a larger surface exposed to potential toxins in the medium. In addition, uniform replicate samples are readily prepared, whereas the metabolic properties of slices differ regionally. The cylinders maintain metabolic activity and some degree of ultrastructural integrity for several hours in vitro and are more practical than organotypic cultures for the rapid evaluation of the effect of serum on myelin associated lipid synthesis. The respiration studies showed that prolonged incubation periods used in some previous in vitro studies are not practical. If energy for lipid synthesis were dependent on tissue respiration such synthesis should be limited after 2 h of incubation and precursor incorporation after 3-4 h in vitro would be of doubtful significance. Myelin was not separated from the cylinders for the metabolic studies because a significant incorporation of newly synthesized myelin sulpholipid into membrane with the density of mature myelin would not be expected after only a 1 h incubation1,14. The ultrastructural response of the cylinders to specific immune serum is similar to that found in cultures. We have previously shown that, in organotypic CNS cultures, EAE serum immunoglobulin binds to myelin and to oligodendrocyteslS,16. Ultrastructurally, immunoreactive myelin was of the wide-spaced configuration10 and immunoglobulin binding was noted in the area of the intraperiod line. The present study has extended these observations to show that a similar localization occurs in CNS myelin from young mice, and the findings suggest that the bound immunoglobulin might be directly related (by induced dehiscence)to the widened myelin configuration and the biochemical changes noted in these prisms of mouse brain treated with anti-white matter serum (Fig. 5). Although the washed lipid extract of every tissue sample was examined by TLC and carefully monitored for radioactivity, the only lipid which demonstrated consistent [U-t4C]galactose incorporation in 1 h, with the exception of galactocerebroside itself, was PC. Even sulphatide, which might be expected to utilize newly labeled galactose via cerebroside, exhibited virtually no 14C uptake in our experiments. This most likely is because the sulphatide is being synthesized primarily from a pool of cerebroside formed before incubation. Radiocarbon incorporation into PC (Table III) in all likelihood comes by way of [14C]glucose derived from galactose by the active epimerase in brain. Glucose carbon readily labels phosphatidyl choline in brain 19. Apparently the turnover of phospholipid is greater than that iv the other lipid components which we were able to isolate and determine.
524 For the investigation of the effects of sera containing immunologically induced anti-myelin factors on myelinating cultures most investigators, including ourselves, have used antisera heated to 56 °C for 30 min to destroy complement as control for the non-complement dependent (and thus non-immunological) effects of sera. The possibility that heat-labile factors other than complement might be present in normal serum was not specifically investigated. The morphological data supported the assumption that no such factors were significant since normal rabbit serum caused no ultrastructural change3,5, 7. Indeed, the ultrastructure and the activity of the myelin associated enzyme CNP of tissue cylinders were also unaffected by the normal serum. However, some normal serum does inhibit the synthesis of the myelin associated galactolipids and phosphatidyl choline and the factor or factors involved are heat labile. This introduces a complication into the evaluation of the decrease in galactolipid synthesis found when tissue is exposed to intact antiserum. Part of the greater synthesis found when the antiserum has been heated to destroy complement is due to the concomitant loss of non-specific factors inhibiting galactolipid synthesis. To determine the inhibition specifically due to complement dependent and presumably immunological factors the data must be corrected for the heat-labile inhibition normally present in rabbit serum. Alternatively, complement may be added to heated rabbit antiserum in order to measure the immune specific response after destruction of the non-specific heat-labile inhibitors. However, it is possible that immune specific factors other than complement are lost on heating and that complement addition will not restore the full inhibiting properties of the antiserum. Such failure to restore full inhibiting properties to heated antiserum has been reported for human demyelinating serum acting on organotypic myelinated cultures 18 and for the tadpole optic nerve system for detecting complement dependent antimyelin activity in human spinal fluid 35. Among the heat-labile factors in immune serum not restored by adding guinea pig serum are the proteolytic enzymes produced and released by activated lymphocytes and which have been shown to degrade purified myelin 6. ACKNOWLEDGEMENTS This work was supported by National Institute of Neurological Diseases and Stroke NS 11920 and the Medical Research Service of the Veterans Administration. REFERENCES 1 Benjamins, J. A. and lwata, R., Entry ofgalactolipids and phospholipids into myelin, Trans. Amer. Soc. Neurochem., 9 (1978) 179. 2 Bornstein, M. B. and Appell, S. H., The application of tissue culture to the study of experimental 'allergic' encephalomyelitis. I. Patterns of demyelination, J. Neuropath. exp. NeuroL, 20 (1961) 141 157. 3 Bornstein, M. B. and Raine, C. S., Experimental allergic encephalomyelitis antiserum inhibition of myelination in vitro, Lab. Invest., 23 (1970) 536-542. 4 Bornstein, M. B. and Raine, C. S., The initial structural lesion in serum-induced demyelination in vitro, Lab. Invest., 35 (1976) 391-401. 5 Brosnan, C. F., Stoner, G. L., Bloom, B. R. and Wisniewski, H. M., Studies on demyelination by activated lymphocytes in the rabbit eye: 1I. Antibody-dependent cell-mediated demyelination, J. lmmunol., 118 (1977) 2103-2110.
525 6 Cammer, W., Bloom, B. R., Norton, W. T. and Gordon, S., Degradation of basic protein in myelin by neutral proteases secreted by stimulated macrophages. A possible mechanism of inflammatory demyelination, Proc. nat. Acad. Sci. (Wash.), 75 (1978) 1554-58. 7 Crain, S. H., Bornstein, M. B. and Lennon, V. A., Depression of complex bioelectric discharges in cerebral tissue by thermostable complement-dependent serum factors, Exp. Neurol., 49 (1975) 330-335. 8 Dubois-Dalcq, M., Niedieck, B. and Buyse, M., Action of anticerebroside sera on myelinated tissue cultures, Path. Europ., 5 (1970) 331-347. 9 Duca, H. F. Le and Cohen, P. P., Preparation of tissues and enzymes. In W. W. Umbreit, R. H. Burris and J. 13".Stauffer (Eds.), Manometric Techniques, Burgess, Minneapolis, Minn., 1964, pp. 132-133. 10 Folchi-Pi, J., Lees, M. and Sloane-Stanley, G. H., A simple method for the isolation and purification of total lipids from animal tissues, J. biol. Chem., 226 (1957) 497-509. 11 Fry, J. M., Lehrer, G. M. and Bornstein, M. B., Sulphatide synthesis: inhibition by experimental allergic encephalomyelitis serum, Science, 175 (1972) 192-194. 12 Fry, J. M., Lehrer, G. M. and Bornstein, M. B., Experimental inhibition of myelination in spinal cord tissue cultures: Enzyme assays, J. Neurobiol., 4 (1973) 453-459. 13 Fry, J. M., Weissbarth, S., Lehrer, G. M. and Bornstein, M. B., Cerebroside antibody inhibits sulphatide synthesis and myelination and demyelination in spinal cord tissue cultures, Science, 183 (1974) 540-542. 14 Hayes, L. W. and Jungawala, F. B., Synthesis and turnover of cerebrosides and phosphatidyl serine of myelin and microsomal fractions of adult and developing rat brain, Biochem. J., 160 (1976) 195-204. 15 Johnson, A. B. and Bornstein, M. B., Myelin-binding antibodies in vitro. Immunoperoxidase studies with EAE, antigalactocerebroside and MS sera, Brain Research, 159 (1978) 173-182. 16 Johnson, A. B., Raine, C. S. and Bornstein, M. B., Experimental allergic encephalomyelitis: serum immunoglohulin binds to myelin and oligodendrocytes in cultured tissue. Ultrastructuralimmuno-peroxidase observation, Lab. Invest., (1979) in press. 17 Lowry, O. H., Rosebrough, H. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 18 Lumsden, C. E., The clinical pathology of multiple sclerosis. In D. McAlpine, C. E. Lumsden and E. D. Acheson (Eds.), Multiple Sclerosis: A Reappraisal, 2nd ed., Williams and Wilkins, Baltimore, Md., 1972, pp. 311-621. 19 Maker, H. S., Clarke, D. D., and Lajtha, A. L., Intermediary metabolism of carbohydrates and amino acids. In G. J. Siegel, R. W. Albers, R. Katzman and B. W. Agranoff (Eds.), Basic Neurochemistry, 2nd ed., Little, Brown, Boston, Mass., 1976, pp. 274-307. 20 Mcllwain, H. and Bachelard, H. S., In Biochemistry and the Central Nervous System, 4th ed., Williams and Wilkins, Baltimore, Md., 1971, pp. 15-69. 21 Mcllwain, H. and Buddle, H. L., Techniques in tissue metabolism. 1. A mechanical chopper, Biochem., J., 53 (1953) 412-420. 22 Olafson, R. W., Drummond, G. I. and Lee, J. F., Studies on 2',3'-cyclic nucleotide-3'-phosphohydrolase from brain, Canad. J. Biochem., 47 (1969) 961-966. 23 Pellkofer, R. and Jatzkewitz, H., Alteration of myelin biosynthesis in slices of rabbit spinal cord by antiserum to myelin basis protein and by puromycin, J. Neurochem., 27 (1976) 351-384. 24 Raine, C. S., Ultrastructural applications of cultured nervous system tissues to neuropathology. In H. M. Zimmerman (Ed.) Progress in Neuropathology, Vol. 2, Grune and Stratton, New York, 1973, p. 27. 25 Raine, C. S. and Bornstein, M. B., Experimental allergic encephalomyelitis. An ultrastructural study of experimental demyelination in vitro, J. Neuropath. exp. Neurol., 29 (1970) 177-191. 26 Seil, F. J., Falk, G. A., Kies, M. W. and Alvord, Jr., E. C.,The in vitro demyelinating activity of sera from guinea pigs sensitized with whole CNS and with purified encephalitogen, Exp. NeuroL, 22 (1968) 545-555. 27 Seil, F. J., Smith, M. E. and Lerman, A. L., Myelination inhibition and neuroelectric blocking factors in experimental allergic encephalomyelitis, Science, 187 (1975) 951-953. 28 Sims, N. R. and Carnegie, P. R., Release of 2',3'-cyclic nucleotide 3-phosphohydrolase from myelin, Proc. Austr. biochem. Soc., 8 (1975) 75. 29 Smith, M. E., Myelin metabolism in vitro in experimental allergic encephalomyelitis, J. Neurochem., 16 (1969) 1099-1104.
526 30 Smith, M. E., A regional survey of myelin development: some compositional and metabolic aspects, J. Lipid Res., 14 (1973) 541-551. 31 Smith, M. E. and Hasinoff, C. M., Biosynthesis of myelin in vitro, J. Neurochem., 18 (1971) 739-748. 32 Smith, M. E. and Sedgewick, L. M., Studies of the mechanism of demyelination : regional differences in myelin stability in vitro, J. Neurochem., 24 (1975) 763-770. 33 Sogin, D. C., 2',3'-cyclic N A D P as substrate for 2",3'-cyclic nucleotide 3'-phosphohydrolase, J. Neurochem., 27 (1976) 1333-1338. 34 Suzuki, K., Pattern of mammalian gangliosides. IX. Evaluation of the extraction procedures, post-mortem changes and the effect of formalin preservation, J. Neurochem., 12 (1965) 629-638. 35 Tabira, T., Webster, H. DeF and Wray, S. H., Multiple sclerosis cerebrospinal fluid produces myelin lesions in tadpole optic nerves, New Eng. J. Med., 295 (1976) 644-649. 36 Thews, G., Implications to physiology and pathology of oxygen diffusion at the capillary level. In J. P. Schade and W. H. McMenemy (Eds.), Selective Vulnerability o/'the Brain in Hypoxiemia, Davis, Philadelphia, Pa., 1963, pp. 27-36. 37 Tsukada, Y., Shibuya, M. and Ogawa, Y., Changes of 2'-,3'-cyclic nucelotide 3'-phosphohydrolase on myelinating neural tissue grown in culture, Abstracts, 4th Int. Meeting of the Int. Soc. Neurochem., (1973) Tokyo, Japan, p. 299.
513
EFFECTS OF ANTI-WHITE MATTER SERUM ON MYELIN AND LIPID SYNTHESIS IN BRAIN PRISMS
EVELYN P. LAPIN, HOWARD S. MAKER, GERARD M. LEHRER, SULAMITH WEISSBARTH, CEDRIC S. RAINE, ANNE B. JOHNSON and MURRAY B. BORNSTEIN
Department of Neurology, Mount Sinai School of Medicine, New York, N. Y. 10029 and Department of Neurology, Albert Einstein College of Medicine, New York, N. Y. 10461 (U.S.A.) (Accepted January 4th, 1979)
SUMMARY
Tissue prisms prepared by chopping whole mouse brain maintained respiratory capacity and ultrastructural integrity for 3 h in vitro. Normal rabbit serum (ca. 25 70) caused no morphological change but inhibited the synthesis of galactolipids by the prisms. Heating the serum abolished the inhibition. Complement containing antiwhite matter rabbit serum destroyed myelin and inhibited galactolipid synthesis to a greater degree than did normal serum. Structures other than myelin were unaffected by the antiserum. Incubation in the presence of heated anti-white matter serum eliminated the myelin destruction but resulted in specific morphological changes characterized by the doubling of the myelin lamellae at the intraperiod line. Immunoperoxidase studies suggest specific binding of immunoglobulin to components of myelin located at the intraperiod line. These changes were similar to those found in organotypic cultures. Heated antiserum did not inhibit galactolipid synthesis but addition of complement (normal guinea pig serum) to the heated antiserum restored only that portion of the inhibition which exceeded that caused by normal serum. Heat labile factors in normal rabbit serum which inhibit myelin lipid synthesis in the prisms must be corrected for in studies in which the heating of serum is used to demonstrate that the effect is complement dependent. The prism system is simpler than that of organotypic cultures and may be useful in the study of myelinotoxic factors. INTRODUCTION
The sera of rabbits sensitized with whole white matter2,3,11,25, myelin25 or galactocerebroside8,18 but not with myelin basic protein alone26 contain complementdependent factors which reversibily demyelinate or inhibit myelination in organotypic cultures of fetal rodent central nervous system. While such cultures may represent an optimal system for detecting and quantifying demyelinating factors in sera of
514 experimental animals or humans suffering from demyelinating disease, they are not always available or convenient. Use of the tadpole optic nerve for detecting such factors may also present problems3L For some studies simpler methods would be helpful particularly if a large patient population is to be followed. SmithZ9, 30 and Smith and Hasinoff 31 have demonstrated that 300-500 mg, 0.5 mm Stadie-Riggs slices of young adult rat brain or spinal cord in 1 ml of medium can synthesize myelin lipids. However, during incubation basic protein was lost and myelin 'dissociation' was inferred from a change in its sedimentation properties 30. Pellkofer and Jatzkewitz 23 showed that the serum of rabbits immunized with myelin basic protein which does not normally demyelinate cultures 27 inhibited the incorporation of tyrosine into basic and proteolipid proteins and of 3sSO4 into cerebroside sulfates when 0.35 mm slices of 15 day of age rabbit spinal cord were incubated for 4 h. No ultrastructural studies of the slices were reported in any of these investigations. The present study used brain cylinders 1.5 mm long × 75 ,urn radius. These dimensions should allow complete oxygenation and substrate access when the medium is equilibrated with 95 % oxygen 36. Tissue respiration and ultrastructural integrity were maintained for up to 3 h. When cylinders were incubated with anti-white matter rabbit serum, complement dependent myelin breakdown as well as inhibition of myelin lipid synthesis were observed. Incubation with heated sera (depleted of complement) produced changes similar to those observed in tissue cultures under similar conditions. MATERIALS AND METHODS
Tissue preparation and respiration studies Tissue cylinders were prepared from the pooled brains of 2 or 3 15-22-day-old S-W albino mice. Animals were decapitated and their brains were rapidly removed and cut into coronal slices 1.5-2.0 mm thick. The slices were chopped with a Mcllwain tissue chopper (Brinkmann Instrument Inc., Westbury, N.Y.) 2o, set at 150 # m with a second cut perpendicular to the first to yield tissue prisms approximately 0.15 × 0.15 × 1.5 mm. Suspended in liquid the prisms became cylindrical. The cylinders were dispersed (2 ~ w/v) in gas equilibrated Krebs-Ringer bicarbonate buffered medium 9 containing 10 or 40 mM glucose, with and without 10 % fetal calf serum (Grand Island Biological Co., Grand Island, N.Y.) by gentle stirring at 4 °C and uniform samples were pipetted into Warburg Flasks. The flasks were gassed with 95 ~ oxygen/5 % CO2 for 10 min before measuring oxygen consumption at 37 °C for periods up to 6 h. Antiserum preparation Antiserum was prepared by inoculating 0.1 ml of whole bovine white matter in complete Freund's adjuvant 4 into the footpads of New Zealand rabbits and bleeding them upon the initial appearance of neurological signs of experimental allergic encephalomyelitis (EAE). Serum was used fresh without prior freezing or after heating at 56 °C for 30 min to destroy complement.
515
Effect of incubation on 2', 3'-cyclic nucleotide 3'-phosphohydrolase (CNP) (EC 3.1.4.16) For studies of the effect of antiserum on the activity of the myelin associated enzyme, CNP, cylinders were incubated at 0 or 37 °C for intervals of 0, 0.5, 1, 1.5, 2, 4, 6 or 8 h in Krebs-Ringer medium containing (final concentration) 16~ nutrient culture medium of Edith R. Peterson (33 ~o Eagles minimal essential medium plus glutamine, 7 ~ balance salt solution, 40 mM glucose, 1.3 #g/ml tetracycline), 5 guinea pig serum as a source of complement and either 12.5 ~ normal rabbit serum or anti-whole white matter serum. Incubation conditions were based on those developed for the study of organotypic cultures z. CNP in both tissue and medium was initially measured by the method of Olafson et al. z2 and later by the method of Sogin 33 adapted as a fluorometric method in our laboratory by S. Weissbarth (in preparation). Also tested was 32~ nutrient medium, 23~ normal or antiserum and 10~ guinea pig serum. Media were used unheated or heated at 56 °C for 30 min.
Galactolipid synthesis Tissue cylinders ( 2 ~ w/v) prepared from 15-19-day-old mouse brain were incubated in Krebs-Ringer bicarbonate medium containing 23 mM glucose and 12 or 23 ~ heated or unheated normal rabbit or anti-whole bovine white matter serum, with or without 10 ~ guinea pig serum as a source of complement. In one experiment D-[U14C]galactose, (New England Nuclear Corp., Boston, Mass.), final concentration 37 #M (2.5 #Ci in 270 #1 incubation medium) or Na2~5SO4, 27.6 #Ci/ml (Amersham Corp., Arlington Heights, Ill., 552 #Ci in 270 #1) were added to matched flasks. In other experiments only Na2SO4 incorporation was determined. Following a 10 min equilibration with 95 ~ oxygen/5 ~ CO2, vessels were incubated at 37 °C for 1 h in a metabolic shaker. The tissue was then washed 4 times with 1.0 ml aliquots of Krebs-Ringer medium containing 37 mM unlabeled galactose or sulfate. The sedimented tissue was resuspended in 250 #l of medium and sonicated (Model W140 Heat Systems-Ultrasonics Inc., Plainview, N.Y.) through 8-9 bursts of 0.3 sec each. Ten #1 aliquots of the sonicates were taken for protein determination17. The lipids were extracted from the remaining portion and partitioned by the method of Folch et al. 10 as modified by SuzukiTM The lower phase lipids were dried under a nitrogen stream, dissolved in 150/A of chloroform/methanol/water (70/30/4 by vol.) and chromatographed on silica gel G chromatography plates (Quantum Industries, Fairfield, N.J.) which had previously been washed with acetone, prerun with the developing solvent and activated. Chromatographically pure lipid standards of phosphatidyl choline (PC), phosphatidyl ethanolamine, cholesterol, cholesterol sulfate, sphingomyelin, cerebrosides and cerebroside sulfates (Applied Science Lab. Inc., State College, Pa.) individually and in combination were chromatographed simultaneously. The plates were developed in chloroform/methanol/water (70/30/4 by vol.). The lipid bands were scraped into scintillation vials. A mixture of toluene/Triton X/liquifluor (New England Nuclear Corp., Boston, Mass.) was added to each vial and the radioactivity assayed in a Packard Model 2450 liquid scintillation counter. All chemicals used were reagent grade.
516
Uhrastructural studies For electron microscopy, tissue cylinders were fixed for 30 min in cold 2.5 glutaraldehyde in pH 7.4 phosphate buffer and post-fixed in similarly buffered 1 osmic acid for 30 rain, dehydrated and embedded in Epon 24. One/zm Epon sections were taken and stained with toluidine blue for light microscopy. Based on examination of these, selected areas were trimmed from the blocks and thin sections cut for electron microscopy. Sections were placed on uncoated grids, double stained with lead and uranyl salts, and carbon coated. The specimens were coded with regard to treatment before fixation and all electron micrographs were prepared and examined prior to breaking the code. Immunohistochemical studies on prisms were performed essentially as in the previous studies with cultured nervous tissuOS, 16. Briefly, the prisms, suspended in Krebs-Ringer buffer, were incubated with 25 ~ (v/v) heated rabbit anti-white matter serum for 3 h at 37 °C, washed in buffer, fixed for 30 min in cold 5 ~ formalin/0.5 glutaraldehyde in Millonig's phosphate buffer, washed in phosphate buffered saline, and incubated with a 30-fold dilution of peroxidase-conjugated sheep anti-rabbit immunoglobulin Fab fragments (Pasteur Institute, 1.25 mg Fab per ml) for 1 h at room temperature. The prisms were then fixed in 2 . 5 ~ glutaraldehyde, reacted with diaminobenzidine medium, osmicated, dehydrated and embedded in Epon. Controls were run by substituting heated normal rabbit serum or buffer alone for the anti-white matter serum. Thin sections were studied without electron counterstains, or after contrasting with uranyl acetate. RESULTS
Viability of tissue cylinders Oxygen consumption by mouse brain tissue cylinders following equilibration at 37 °C is shown in Fig. 1. Initially high rates of activity were most likely due to the accumulation during tissue preparation of metabolic products that stimulate respi~" 6.0" 03 0')
7-- 5.0
__q 4.0 .-~ 3.0 E 2.0 Z
X 0
BO
120
IBO
240
MINUTES OF INCUBATION
Fig. 1. Respiratory rates (ml O2/g tissue/h) of tissue prisms prepared from the brains of 17-day-old mice and incubated (20 mg/ml) at 37 °C in Krebs-Ringer bicarbonate medium containing 10 m M glucose (©), 40 m M glucose (O), or 40 m M glucose in 63 ~ tissue culture nutrient medium (containing 2 ~ fetal calf serum) ( ~ ) . Oxygen consumption values are corrected for tissue and reagent blanks.
517 ration. A uniform rate of oxygen utilization was then maintained for 90 min at a level of 80 ~ of that of adult human brain and equal to that of 0.35 mm thick unstimulated guinea pig cortical sliceslL There was then a steady decline in 02 consumption and the tissue ceased to respire after 4-6 h in vitro. Only minor variations in the rate of deterioration of respiration were noted on repeated determinations. Increasing the glucose concentration from 10 to 40 mM (necessary for myelination of organotypic cultures) or adding fetal calf serum did not prolong the period of respiration. The ultrastructural integrity of the tissue was maintained for at least 3 h. Mitochondria and myelin lamellae showed only minor artefacts in buffer alone or in normal rabbit serum-enriched media (Fig. 3a). Incubation at 37 °C caused a small increase in the activity of tissue CNP (Fig. 2). The increase may have been due to improved accessibility of the enzyme to its substrate even above that resulting from sonification and use of 1.0 ~ of Triton X-100 (ref. 26). There was a lowered or no activation of the enzyme at 0 °C. No CNP activity was detected in the medium or in serum before incubation. Incubation caused the release of CNP into the medium within 30 min (37 mmol/gram tissue protein/h at 37 °C and 26 mmol/gram tissue protein/h at 0 °C). However, no further release occurred during incubations as long as 8 h. Tissue CNP activity increased from 90 -I- 10 mmol/gram protein/h in cylinders prepared from the brains of mice 15 days of age to 375 q- 25 mmol/gram protein/h at 27 days of age.
Normal rabbit serum Incubation for 3 h in medium containing decomplemented normal rabbit serum had no noticeable effect on myelin morphology (Fig. 3a). The slow deterioration of tissue respiration remained unchanged and the activity of CNP was stable for 6 h (Figs. 1 and 2) and as long as 8 h (not shown). However, as compared with medium alone, non-immune rabbit serum reduced the incorporation of z5SO4 into cerebroside sulfates by up to one half (Tables I and II). Incorporation of [U-14C]galactose into bJ t/) <[
0.-I nr",>_
I I NO SERUM-S7* J~ HEATED ANTISERUM-37* ANT SERUM-37*
0 "IQ.
o
200
t~ ".,.; 150 ]
:;...,~ ~ . . . . . . . ....::::,,~,.~
V-2 bJ
~ ~
I°° 1
-~--" ~-°
50 1
0
HOURS OF INCUBATION Fig. 2. 2',3'-cyclic nucleotide phosphohydrolase activities (mmol/g protein/h) of tissue prisms prepared from the brains of 17-day-old mice and incubated in Krebs-Ringer bicarbonate medium containing 26 mM glucose at 37 °C (solid lines) or 0 °C (broken lines) without serum (O), with 23.0 % heated (ll), or unheated (A) anti-white matter rabbit serum.
518
Fig. 3. a: brain tissue cylinders incubated for 3 h with heated (decomplemented) normal rabbit serum. A portion of a myelin sheath displays the normal periodicity of 11 nm. The intraperiod line is artefactually enhanced in density. The minor membrane discontinuities are preparation artefacts. Axon below, x 150,000. b: cylinder incubated for 3 h with unheated serum from a rabbit with EAE sensitized against whole white matter. A segment of myelin displays swelling of the outer layers. This swelling (arrows) occurs in the space of the intraperiod line. x 150,000. c: same preparations as b. A small area of a myelin sheath displays swelling at the intraperiod line (arrows), while the major dense line is intact, x 150,000. d: same preparation as b. A similar area shows advanced swelling and breakdown of myelin. × 150,000.
519 TABLE I
[14CJgalactose and zs SO a uptake into lipids in myelinating mouse brain prisms The significance of differences between all groups were calculated, but only those whose significance might be in doubt are shown. Significance of differences between means and that of medium without serum (a) ; between means and normal rabbit serum (b) ; between means and heated antiserum (c) are shown; single letter P < 0.05, double letters P < 0.01. Assays were performed in quadruplicate. Values in parentheses indicate standard deviation.
Serum added
Cerebroside [14CJgalactose (pmol/mg prot./h)
Cerebroside sulfate a5S04 (pmol/mg prot./h)
(A) None (B) Unheated rabbit
1.58 (0.20) 1.18 (0.30)
(C) Heated antiserum (---complement)
2.90 (0.27) aa; bb 1.15 (0.30) cc
14.0 (0.9) 7.19 (0.40) aa* 19.1 (1.6) aa; bb 8.75 (1.00) aa; cc
(D) Unheated antiserum (+complement)
* Incorporation of [14C-U]galactose into cerebrosides and [35SO4]sulfate into cerebroside sulfate (sulfatide) of tissue prisms incubated for 1 h at 37 °Cin Krebs-Ringer bicarbonate mediawith addition of 25 ~ serum and complement (10 % guinea pig serum) as indicated.
TABLE II
z5SO a uptake into cerebroside sulphate in myelinating mouse brain prisms The significance of differences between all groups were calculated, but only those whose significance might be in doubt are shown, a, significance of difference between means and that of medium A, without added serum; b, difference between means, and that containing heated normal rabbit serum B; c, difference between means and that containing unheated rabbit serum C; d, difference between means and that containing heated antiserum to bovine white matter D ; e, difference between means and that containing unheated antiserum E; f, difference between means and that containing heated normal rabbit serum plus guinea pig complement G. Single letter indicates P < 0.05 ; double letter P < 0.01. A and B were two separate studies with different tissue and serum preparations. Values in parentheses indicate standard deviation. N D = not determined.
Serum added
(A) None (B) Heated rabbit (C) Unheated rabbit (D) Heated antiserum (E) Unheated antiserum
Cerebroside sulfate synthesis (pmol/mg protein~h) A
B
6.18 (0.58) 9.43 (0.44) aa 4.79 (0.74) a; bb 9.74 (0.86) aa 2.94 (0.06)
3.95 (0.16) 4.65 (0.21) aa 2.49 (0.26) aa 5.40 (0.88) cc 1.26 (0.16)
CC
CC
(F) Heated rabbit (+complement)
ND
(G) Heated antiserum (+complement)
ND
(H) Unheated antiserum (+complement)
3.23 (0.45) c; dd
4.19 (0.70) cc; ee 3.51 (0.30) a; bb; cc; dd 1.37 (0.20) cc; dd; gg
520 TABLE III [14C]galactose uptake into (galacto)cerebroside and into phosphatidyl choline in myelinating mouse brain prisms (cpm/mg protein/h)
n - number of samples per assay. All other indications and symbols are explained in Tables I and 1I. Values in parentheses indicate standard deviation. ND = not determined. Experiment C (n = 3)
Experiment D (n == 4)
Cerebroside
PC
Cerebroside
PC
975 (124) ND
245 (68) ND
4779 (742) 6008 (514)
1669 (351) 1663 (254)
(C) Unheated rabbit
728 (184)
145 (88)
(D) Heated antiserum
1786 (169) aa ND ND ND
Serum added
(A) None (B) Heated rabbit
a
(E) Unheated antiserum (F) Heated rabbit (+-complement) (G) Heated antiserum (+ complement) (H) Unheated antiserum (+ complement)
275 (49) aa ND ND
2807 (815) aa 5146 (1429)
568 (151) aa 1522 (143)
989 (123) 4541 (664)
374 (71) 1120 (199)
ND
4780 (1043)
889 (120)
928 (74)
375 (145)
C
705 (188) dd
148 (27) d
cerebroside and cerebroside sulfate was reduced 25 ~ and 40 % respectively in one experiment and 40 ~ and 65 ~ respectively in a second experiment, but these decreases did not always reach statistical significance because of the greater variability of the galactose incorporation data. Heating the rabbit serum under the conditions used to destroy complement abolished the inhibition of galactolipid and PC synthesis (Tables I and lII). Heated rabbit serum actually increased the incorporation of 3~SO4 into sulphatide and of [U-14C]galactose into cerebroside above that found in tissue incubated in buffer alone. The addition of complement to heated normal rabbit serum appeared to compensate for some of the loss of heat-labile factor(s) in normal serum but did not restore either its sulfate or galactose incorporation inhibiting property (Table II). Anti-whole white m a t t e r serum
Anti-whole white matter serum containing complement caused wide separation, swelling and breakdown of myelin sheaths (Figs. 3, 4, 5) similar to that seen in myelinating organotypic cultures2L Heating the antiserum (destroying complement) eliminated the myelin breakdown. The activity of C N P fell slightly more rapidly in tissue cylinders incubated with antiserum than with normal serum (Fig. 2) but the difference was small and not consistent. C N P activity was not a reliable indicator of myelin disruption under the conditions of incubation. Longer exposure to the antiserum may have produced greater changes in CNP activity since this does occur in culture11, 35- Antiserum reduced the incorporation of [U-14C]galactose into galactolipid and of 35SO4 into cerebroside sulfate by more than 50 ~ below that in tissue incubated in heated (complement depleted) serum enriched medium (Tables I, II, III). Lipid synthesis in medium supplemented with heated antiserum was similar to that in medium containing heated normal serum. Part of the inhibition produced by the
521 u n h e a t e d a n t i s e r u m could be attributed to the heat labile factor(s) in n o n - i m m u n i z e d r a b b i t serum a n d was n o t i m m u n e related. Correcting for these n o n - i m m u n e factors reduced by one-half the a p p a r e n t i n h i b i t i o n of the antiserum. A d d i n g c o m p l e m e n t to the heated a n t i s e r u m restored only that p o r t i o n of the i n h i b i t i o n due to the c o m p l e m e n t dependent i m m u n o l o g i c a l factors. Myelin d i s r u p t i o n required both the specific i m m u n e factors a n d c o m p l e m e n t a n d was n o t seen within 3 h with n o r m a l r a b b i t or c o m p l e m e n t depleted antiserum.
Fig. 4. a: cylinder incubated for 3 h with heated (decomplemented) EAE rabbit serum. The myelin period has been increased from 11 nm to about 22 nm due to a swelling occurring in the space of the intraperiod line. Instead of two leaflets in this space, 4 leaflets can now be seen (arrows). x 150,000. b: similar preparation to a, A myelin sheath is incompletely swollen in that outer layers are increased to 22 nm (arrows), with the concomitant duplication of structure, and the innermost layers display the normal 11 nm periodicity, x 150,000.
522
Fig. 5. Immunostained myelin in prisms incubated sequentially with 25 ~ , heated, rabbit EAE serum (3h), peroxidase-labeled anti-rabbit immunoglobulin Fab fragments (1 h), and diaminobenzidine peroxidase medium (0.5 h) and then processed for electron miscroscopy. Reaction product is present at the intraperiod line (arrows) in A and B and on the outer oligodendroglial plasmalemma (double arrow) in A. Arrowheads point to major dense lines. Immunoreactive myelin is of the previously described, wide-spaced configuration (22 nm periodicity with 4 linear components to the intraperied line instead of two), and the alteration of the intraperiod line is evident in A in the two lamellae below the arrow and in B adjacent to the middle two arrowheads. A non-immunoreactive lamella of normal periodicity (11 nm) can be seen in B (double arrow). Uranyl acetate contrasting. A, x 152,000; B, × 142,000.
H e a t e d a n t i s e r u m did n o t cause myelin disruption, but h a d a n effect on the tissue cylinders identical to t h a t recently described in o r g a n o t y p i c cultures o f mouse spinal c o r d by Bornstein a n d R a i n e 4. I n c u b a t i o n with heated a n t i s e r u m (but n o t h e a t e d n o r m a l serum) consistently caused an increase in the spacing between the m a j o r dense lines a c c o m p a n i e d by the a p p e a r a n c e o f an a d d i t i o n a l p a i r o f electron dense leaflets replacing the i n t r a p e r i o d line, p r o d u c i n g 4 instead o f two leaflets between the m a j o r dense lines a n d a periodicity o f 22 n m instead o f the usual t 1 nm (Figs. 3 a n d 4). The overall effect was t h a t o f swelling o f the myelin sheaths. R e a c t i o n p r o d u c t indicating b o u n d i m m u n o g l o b u l i n originating in heated a n t i s e r u m was present a r o u n d myelinated axons a n d sometimes at the a r e a o f the i n t r a p e r i o d line. All i m m u n o p e r o x i d a s e stained myelin was o f the previously described wide-spaced configuration (Fig. 5). In
523 controls, no reaction product was observed around or within myelin lamellae. No oligodendrocytes were identified in the prisms. Sulfate incorporation was not significantly greater in tissue exposed to heated antiserum than to heated normal serum. Organotypic cultures exposed to heated antiserum for periods over 24 h show both morphological4 and biochemical (Lehrer et al., Brain Research, in press) evidence of increased lipid synthesis. DISCUSSION Tissue prisms, which assume a cylindrical shape of 75 #m radii when suspended in liquid, allow a better diffusion of oxygen and metabolites than do slices and possess a larger surface exposed to potential toxins in the medium. In addition, uniform replicate samples are readily prepared, whereas the metabolic properties of slices differ regionally. The cylinders maintain metabolic activity and some degree of ultrastructural integrity for several hours in vitro and are more practical than organotypic cultures for the rapid evaluation of the effect of serum on myelin associated lipid synthesis. The respiration studies showed that prolonged incubation periods used in some previous in vitro studies are not practical. If energy for lipid synthesis were dependent on tissue respiration such synthesis should be limited after 2 h of incubation and precursor incorporation after 3-4 h in vitro would be of doubtful significance. Myelin was not separated from the cylinders for the metabolic studies because a significant incorporation of newly synthesized myelin sulpholipid into membrane with the density of mature myelin would not be expected after only a 1 h incubation1,14. The ultrastructural response of the cylinders to specific immune serum is similar to that found in cultures. We have previously shown that, in organotypic CNS cultures, EAE serum immunoglobulin binds to myelin and to oligodendrocyteslS,16. Ultrastructurally, immunoreactive myelin was of the wide-spaced configuration10 and immunoglobulin binding was noted in the area of the intraperiod line. The present study has extended these observations to show that a similar localization occurs in CNS myelin from young mice, and the findings suggest that the bound immunoglobulin might be directly related (by induced dehiscence)to the widened myelin configuration and the biochemical changes noted in these prisms of mouse brain treated with anti-white matter serum (Fig. 5). Although the washed lipid extract of every tissue sample was examined by TLC and carefully monitored for radioactivity, the only lipid which demonstrated consistent [U-t4C]galactose incorporation in 1 h, with the exception of galactocerebroside itself, was PC. Even sulphatide, which might be expected to utilize newly labeled galactose via cerebroside, exhibited virtually no 14C uptake in our experiments. This most likely is because the sulphatide is being synthesized primarily from a pool of cerebroside formed before incubation. Radiocarbon incorporation into PC (Table III) in all likelihood comes by way of [14C]glucose derived from galactose by the active epimerase in brain. Glucose carbon readily labels phosphatidyl choline in brain 19. Apparently the turnover of phospholipid is greater than that iv the other lipid components which we were able to isolate and determine.
524 For the investigation of the effects of sera containing immunologically induced anti-myelin factors on myelinating cultures most investigators, including ourselves, have used antisera heated to 56 °C for 30 min to destroy complement as control for the non-complement dependent (and thus non-immunological) effects of sera. The possibility that heat-labile factors other than complement might be present in normal serum was not specifically investigated. The morphological data supported the assumption that no such factors were significant since normal rabbit serum caused no ultrastructural change3,5, 7. Indeed, the ultrastructure and the activity of the myelin associated enzyme CNP of tissue cylinders were also unaffected by the normal serum. However, some normal serum does inhibit the synthesis of the myelin associated galactolipids and phosphatidyl choline and the factor or factors involved are heat labile. This introduces a complication into the evaluation of the decrease in galactolipid synthesis found when tissue is exposed to intact antiserum. Part of the greater synthesis found when the antiserum has been heated to destroy complement is due to the concomitant loss of non-specific factors inhibiting galactolipid synthesis. To determine the inhibition specifically due to complement dependent and presumably immunological factors the data must be corrected for the heat-labile inhibition normally present in rabbit serum. Alternatively, complement may be added to heated rabbit antiserum in order to measure the immune specific response after destruction of the non-specific heat-labile inhibitors. However, it is possible that immune specific factors other than complement are lost on heating and that complement addition will not restore the full inhibiting properties of the antiserum. Such failure to restore full inhibiting properties to heated antiserum has been reported for human demyelinating serum acting on organotypic myelinated cultures 18 and for the tadpole optic nerve system for detecting complement dependent antimyelin activity in human spinal fluid 35. Among the heat-labile factors in immune serum not restored by adding guinea pig serum are the proteolytic enzymes produced and released by activated lymphocytes and which have been shown to degrade purified myelin 6. ACKNOWLEDGEMENTS This work was supported by National Institute of Neurological Diseases and Stroke NS 11920 and the Medical Research Service of the Veterans Administration. REFERENCES 1 Benjamins, J. A. and lwata, R., Entry ofgalactolipids and phospholipids into myelin, Trans. Amer. Soc. Neurochem., 9 (1978) 179. 2 Bornstein, M. B. and Appell, S. H., The application of tissue culture to the study of experimental 'allergic' encephalomyelitis. I. Patterns of demyelination, J. Neuropath. exp. NeuroL, 20 (1961) 141 157. 3 Bornstein, M. B. and Raine, C. S., Experimental allergic encephalomyelitis antiserum inhibition of myelination in vitro, Lab. Invest., 23 (1970) 536-542. 4 Bornstein, M. B. and Raine, C. S., The initial structural lesion in serum-induced demyelination in vitro, Lab. Invest., 35 (1976) 391-401. 5 Brosnan, C. F., Stoner, G. L., Bloom, B. R. and Wisniewski, H. M., Studies on demyelination by activated lymphocytes in the rabbit eye: 1I. Antibody-dependent cell-mediated demyelination, J. lmmunol., 118 (1977) 2103-2110.
525 6 Cammer, W., Bloom, B. R., Norton, W. T. and Gordon, S., Degradation of basic protein in myelin by neutral proteases secreted by stimulated macrophages. A possible mechanism of inflammatory demyelination, Proc. nat. Acad. Sci. (Wash.), 75 (1978) 1554-58. 7 Crain, S. H., Bornstein, M. B. and Lennon, V. A., Depression of complex bioelectric discharges in cerebral tissue by thermostable complement-dependent serum factors, Exp. Neurol., 49 (1975) 330-335. 8 Dubois-Dalcq, M., Niedieck, B. and Buyse, M., Action of anticerebroside sera on myelinated tissue cultures, Path. Europ., 5 (1970) 331-347. 9 Duca, H. F. Le and Cohen, P. P., Preparation of tissues and enzymes. In W. W. Umbreit, R. H. Burris and J. 13".Stauffer (Eds.), Manometric Techniques, Burgess, Minneapolis, Minn., 1964, pp. 132-133. 10 Folchi-Pi, J., Lees, M. and Sloane-Stanley, G. H., A simple method for the isolation and purification of total lipids from animal tissues, J. biol. Chem., 226 (1957) 497-509. 11 Fry, J. M., Lehrer, G. M. and Bornstein, M. B., Sulphatide synthesis: inhibition by experimental allergic encephalomyelitis serum, Science, 175 (1972) 192-194. 12 Fry, J. M., Lehrer, G. M. and Bornstein, M. B., Experimental inhibition of myelination in spinal cord tissue cultures: Enzyme assays, J. Neurobiol., 4 (1973) 453-459. 13 Fry, J. M., Weissbarth, S., Lehrer, G. M. and Bornstein, M. B., Cerebroside antibody inhibits sulphatide synthesis and myelination and demyelination in spinal cord tissue cultures, Science, 183 (1974) 540-542. 14 Hayes, L. W. and Jungawala, F. B., Synthesis and turnover of cerebrosides and phosphatidyl serine of myelin and microsomal fractions of adult and developing rat brain, Biochem. J., 160 (1976) 195-204. 15 Johnson, A. B. and Bornstein, M. B., Myelin-binding antibodies in vitro. Immunoperoxidase studies with EAE, antigalactocerebroside and MS sera, Brain Research, 159 (1978) 173-182. 16 Johnson, A. B., Raine, C. S. and Bornstein, M. B., Experimental allergic encephalomyelitis: serum immunoglohulin binds to myelin and oligodendrocytes in cultured tissue. Ultrastructuralimmuno-peroxidase observation, Lab. Invest., (1979) in press. 17 Lowry, O. H., Rosebrough, H. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 18 Lumsden, C. E., The clinical pathology of multiple sclerosis. In D. McAlpine, C. E. Lumsden and E. D. Acheson (Eds.), Multiple Sclerosis: A Reappraisal, 2nd ed., Williams and Wilkins, Baltimore, Md., 1972, pp. 311-621. 19 Maker, H. S., Clarke, D. D., and Lajtha, A. L., Intermediary metabolism of carbohydrates and amino acids. In G. J. Siegel, R. W. Albers, R. Katzman and B. W. Agranoff (Eds.), Basic Neurochemistry, 2nd ed., Little, Brown, Boston, Mass., 1976, pp. 274-307. 20 Mcllwain, H. and Bachelard, H. S., In Biochemistry and the Central Nervous System, 4th ed., Williams and Wilkins, Baltimore, Md., 1971, pp. 15-69. 21 Mcllwain, H. and Buddle, H. L., Techniques in tissue metabolism. 1. A mechanical chopper, Biochem., J., 53 (1953) 412-420. 22 Olafson, R. W., Drummond, G. I. and Lee, J. F., Studies on 2',3'-cyclic nucleotide-3'-phosphohydrolase from brain, Canad. J. Biochem., 47 (1969) 961-966. 23 Pellkofer, R. and Jatzkewitz, H., Alteration of myelin biosynthesis in slices of rabbit spinal cord by antiserum to myelin basis protein and by puromycin, J. Neurochem., 27 (1976) 351-384. 24 Raine, C. S., Ultrastructural applications of cultured nervous system tissues to neuropathology. In H. M. Zimmerman (Ed.) Progress in Neuropathology, Vol. 2, Grune and Stratton, New York, 1973, p. 27. 25 Raine, C. S. and Bornstein, M. B., Experimental allergic encephalomyelitis. An ultrastructural study of experimental demyelination in vitro, J. Neuropath. exp. Neurol., 29 (1970) 177-191. 26 Seil, F. J., Falk, G. A., Kies, M. W. and Alvord, Jr., E. C.,The in vitro demyelinating activity of sera from guinea pigs sensitized with whole CNS and with purified encephalitogen, Exp. NeuroL, 22 (1968) 545-555. 27 Seil, F. J., Smith, M. E. and Lerman, A. L., Myelination inhibition and neuroelectric blocking factors in experimental allergic encephalomyelitis, Science, 187 (1975) 951-953. 28 Sims, N. R. and Carnegie, P. R., Release of 2',3'-cyclic nucleotide 3-phosphohydrolase from myelin, Proc. Austr. biochem. Soc., 8 (1975) 75. 29 Smith, M. E., Myelin metabolism in vitro in experimental allergic encephalomyelitis, J. Neurochem., 16 (1969) 1099-1104.
526 30 Smith, M. E., A regional survey of myelin development: some compositional and metabolic aspects, J. Lipid Res., 14 (1973) 541-551. 31 Smith, M. E. and Hasinoff, C. M., Biosynthesis of myelin in vitro, J. Neurochem., 18 (1971) 739-748. 32 Smith, M. E. and Sedgewick, L. M., Studies of the mechanism of demyelination : regional differences in myelin stability in vitro, J. Neurochem., 24 (1975) 763-770. 33 Sogin, D. C., 2',3'-cyclic N A D P as substrate for 2",3'-cyclic nucleotide 3'-phosphohydrolase, J. Neurochem., 27 (1976) 1333-1338. 34 Suzuki, K., Pattern of mammalian gangliosides. IX. Evaluation of the extraction procedures, post-mortem changes and the effect of formalin preservation, J. Neurochem., 12 (1965) 629-638. 35 Tabira, T., Webster, H. DeF and Wray, S. H., Multiple sclerosis cerebrospinal fluid produces myelin lesions in tadpole optic nerves, New Eng. J. Med., 295 (1976) 644-649. 36 Thews, G., Implications to physiology and pathology of oxygen diffusion at the capillary level. In J. P. Schade and W. H. McMenemy (Eds.), Selective Vulnerability o/'the Brain in Hypoxiemia, Davis, Philadelphia, Pa., 1963, pp. 27-36. 37 Tsukada, Y., Shibuya, M. and Ogawa, Y., Changes of 2'-,3'-cyclic nucelotide 3'-phosphohydrolase on myelinating neural tissue grown in culture, Abstracts, 4th Int. Meeting of the Int. Soc. Neurochem., (1973) Tokyo, Japan, p. 299.