- Email: [email protected]
Early adrenalectomy increases myelin content of the rat brain
DevelopmentalBrain Research, 17 (1985) 1-9 Elsevier
I
BRD 50119
Research Reports
Early Adrenalectomy Increases Myelin Content of the Rat Brain JERROLD S. MEYER and KENNETH R. FAIRMAN
Division of Neuroscienceand Behavior, Departmentof Psychology, Universityof Massachusetts, Amherst, MA 01003 (U.S.A.) (Accepted June 19th, 1984)
Key words: myelin - - brain weight - - brain development - - adrenalectomy
Rats were adrenalectomized (ADX) or sham-operated (SHAM) on the llth day of life and killed on days 35-36, 63, or 151-153 for the isolation of cerebral myelin from each animal. Despite having lower overall body weights, ADX rats had heavier cerebra than SHAM control rats at all ages. Mean cerebral weight increases were 10.0% at day 35-36, 15.3% at day 63, and 16.7% at day 151-153. Recovered myelin dry weights were even more elevated in the ADX rats, but only at day 63 (41.7% increase) and 151-153 (42.1% increase). At both of these ages, there was a clear linear relationship between cerebral wet weights and the amount of myelin recovered from the cerebra. Analysis of the day-63 myelin samples showed no group differences in total cholesterol or protein concentration or in the specific activity of the myelin marker enzyme 2':3'-cyclic-nucleotide3'-phosphodiesterase (CNP). However, myelin isolated from the ADX rats appeared to be deficient in both galactolipid and phospholipid. Optic nerve myelination was assessed in all animals by measuring CNP activity in homogenates prepared from this tissue. No difference between ADX and SHAM rats was observed at any age. These results indicate that early adrenalectomy stimulates myelin deposition in the rat brain as part of a more general, long-lasting enhancement of brain growth. Myelin from the brains of ADX animals may be slightly abnormal in its lipid composition. Finally, the optic nerve data may mean that myelination is not affected equally in all areas of the CNS by the loss of adrenal glands. INTRODUCTION Considerable evidence suggests that adrenal hormones, particularly the glucocorticosteroids, can influence CNS myelinogenesis in developing rats. Most investigators have studied this p r o b l e m by administering pharmacological doses of glucocorticoids to young subjects and then examining biochemical or morphological indices of myelination at a later time. A n u m b e r of studies 5,9,11A3,24,29 have suggested that early glucocorticoid t r e a t m e n t retards the rate of myelin deposition, at least temporarily. Some of these studies m a d e use of biochemical m a r k e r s of myelination in their assessment of glucocorticoid effects. F o r e x a m p l e , cerebral sulfatide concentrations were significantly lower in rats t r e a t e d neonatally with corticosterone 13, while cerebral 2':3'-cyclic-nucleotide 3 ' - p h o s p h o d i e s t e r a s e (CNP) activity was
likewise reduced in animals given cortisol early in life 24. It is interesting to note that the latter effect, as well as various other deleterious consequences of glucocorticoid administration, was substantially reversed by daily growth h o r m o n e injections until the time of weaning 24. Ultrastructurai studies of myelination following early exposure to glucocorticoids have e x p a n d e d upon the above findings. G u m b i n a s et al. u e x a m i n e d axonal growth and myelination in the p y r a m i d a l tract of rats treated on the 6th day of life with a single injection of methylprednisolone. A reduction in the percentage of myelinated fibers (in relation to the total n u m b e r of fibers per unit area) was noted at both 3 and 6 weeks of age, although there was some evidence that the e x p e r i m e n t a l and control groups were converging at the later time point. The results also showed that the n u m b e r of lamellae as a function of
Correspondence and reprint requests: J. S. Meyer, Department of Psychology, University of Massachusetts, Amherst, MA 01003, U.S.A. 0165-3806/85/$03.30 © 1985 Elsevier Science Publishers B.V.
axon circumference was decreased in the glucocorticold-treated animals, indicating a breakdown in the normal relationship between axon size and myelin sheath development. Bohn and Friedrich 5 studied optic nerve myelination and oligodendrocyte proliferation in rats given cortisol from days 7-18 postnatal. As in the case of the pyramidal tract, the percentage of optic nerve axons that were myelinated was significantly lower in the glucocorticoid-treated subjects at day 21. In this instance, however, examination of older animals (60 days of age) revealed a complete normalization of all myelin-related parameters. Parallel studies of optic nerve cell proliferation using [3H]thymidine autoradiography revealed a stimulation of oligodendrocyte formation following the end of hormone administration. This effect may have been a compensatory response to the well-known suppressive action of high-dose glucocorticoid treatment on. neural cell division 3 and furthermore may have been responsible for the ultimate recovery of myelination in the experimental rats. Although the above-cited experiments show that glucocorticoid administration can inhibit CNS myelinogenesis in young rats, at least one study has suggested the opposite relationship. Casper et al. 6 reported that cortisol treatment from the 6th to the 10th day of life increased the cerebroside concentration in the spinal cord when measured on day 12. Furthermore, glucocorticoids stimulate the activity of cytoplasmic glycerol-3-phosphate dehydrogenase (GPDH) in rat brain 7, where the enzyme is localized primarily in oligodendrocytes 2°. Because G P D H is thought to be important in phosphoglyceride biosynthesis 18 and oligodendrocytes must synthesize large amounts of phospholipid for incorporation into myelin, Leveille et al.20 have proposed that glucocorticoids might have a positive influence on the maintenance and/or development of myelin sheaths in the rat brain. A question thus arises as to the exact role of the adrenal glands in the normal process of myelinogenesis, partially because of the inconsistencies noted above but also because the effects of high-dose glucocorticoid treatment may not reflect the actions of these hormones at their usual circulating concentrations. We have recently begun to reexamine the role of the adrenal glands in a number of aspects of brain development by studying rats that have been bilater-
ally adrenalectomized (ADX) on the l l t h day of life. Despite reductions in overall body weight, A D X rats showed significant increases in wet and dry brain weight (both cerebrum and cerebellum) and in cerebral D N A content when compared to sham-operated (SHAM) controls at days 65-67 of life 23. We also found a substantial increase in the amount of purified myelin that could be recovered from the cerebra of A D X animals killed at that age. Hormone replacement studies are still in progress, hence we have not yet determined whether these changes are due to the loss of glucocorticoids or other adrenal secretions. The present paper further characterizes the relationship between the adrenal glands and myelination, focusing on the time course of the adrenalectomy effect and also providing a detailed lipid analysis of the myelin isolated from A D X vs SHAM subjects at one of the time points. The results show that early adrenalectomy produces long-term increases in brain myelin content and that the myelin formed under these conditions may not be completely normal in its composition. MATERIALS AND METHODS Animals and surgery The animals used were male and female SpragueDawley rats (Charles River CD strain) bred in our laboratory under previously described conditions 23. Litters were culled to 8 pups (sex-balanced where possible) on the day after birth. On the l l t h day of life (counting the first day that pups were observed as day 1), rats were subjected to bilateral adrenalectomy or sham operation (laparotomy only) under ether anesthesia. Prior to weaning, A D X animals were maintained on the mineralocorticoid deoxycorticosterone pivalate (Percorten Pivalate, Ciba Pharmaceutical Co.) as previously described 23. Such treatment promotes sodium retention and is necessary for the survival of the A D X rats. SHAM controls received saline vehicle instead, inasmuch as Percorten was previously found to have no effect on brain growth in such animals (J. S. Meyer, K. R. Fairman, and M. Friedman, unpublished observations). The subjects were weaned on day 30 of life, transferred to standard wire-bottom group cages, and given ad libiturn access to Purina Rodent Laboratory Chow, tap water, and 0.85% saline (SHAM as well as A D X
subjects). All animals were maintained under a 14:10 light-dark cycle (lights on at 06.00 h) at a temperature of approximately 22 °C.
Sample preparation On days 35-36, 63, or 151-153 of life, rats were weighed, decapitated, and their brains (minus the olfactory bulbs) rapidly removed and placed on a chilled ground-glass plate. A single vertical cut was made between the cerebral hemispheres and the cerebellum and the resulting tissue sample (cerebrum) used for myelin isolation. Each cerebrum was weighed, rinsed with ice-cold 0.32M sucrose, minced finely, and homogenized in 10 vols. of 0.32 M sucrose using 4 slow strokes at 300 rpm with a 0.20-0.23 mm clearance Teflon-glass Duall homogenizer (Kontes). This and all subsequent steps were carried out at 4 °C. Myelin was isolated from the crude homogenates by the procedure of Jungawala and Dawson 15, omitting the final continuous density gradient step. The myelin was then washed 3 times in deionized water to remove residual sucrose. The purity of the myelin isolated by this method has been confirmed biochemically by the absence of the microsomal marker enzyme NADPH-cytochrome c reductase and the mitochondrial marker cytochrome c oxidase 15. We have also examined the myelin by standard transmission electron microscopy and found it to consist exclusively of membranous profiles with many multilamellar structures. In the case of the 35-36- and 151-153-day-old animals, all of the recovered myelin was lyophilized in tared tubes to obtain the dry weight of each sample. For the 63-day-old group, the myelin was suspended in deionized water to a total volume of 10.0 ml. One ml of this suspension was frozen at -40 °C for subsequent CNP and protein determination (see below), while the remainder was lyophilized. The lyophilized samples were weighed (weight data presented below have been corrected to reflect total recovered myelin) and then later processed for lipid determinations. The intracranial portion of the optic nerves from each animal in all age groups was dissected at the same time as the brain samples. Optic nerves were homogenized in 1.0 ml of ice-cold deionized water and frozen at -40 °C for measurement of CNP and protein.
Lipid extraction and analysis Lyophilized myelin from the 63-day-old animals was dissolved in chloroform-methanol (CM) 2:1 (vol./vol.) and the resulting mixture filtered through glass wool to remove the insoluble protein. The lipids were partitioned by the addition of 0.2 vols. of 0.88% aqueous KC110, followed by two washes of the chloroform-rich lower phase with KCl-containing Folch 'theoretical upper phase'. The lower phase was subsequently evaporated to dryness at 40 °C in vacuo and the lipid residue redissolved in CM 2:1. A portion of each sample was subjected to silicic acid column chromatography in which cholesterol, glycolipids, and phospholipids were eluted with chloroform, acetone, and methanol respectively28. All eluent fractions were stored at 4 °C for subsequent analysis. Total cholesterol (free plus esterified) in the chloroform eluates was determined by the enzymaticspectrophotometric method of Allain et al. 1. The acetone eluates were assayed for lipid-bound galactose by the orcinol-sulfuric acid method of Balazs et al. 2. Galactose values were multiplied by 4.7 to estimate myelin galactolipid weight 26. A modification of Kean's ~6azure A dye-binding method was used to determine myelin sulfatides in the same acetone fractions. Total phospholipid phosphorus was assayed in each methanol fra~ction by first hydrolyzing the lipids using sulfuric acid and hydrogen peroxide as previously described23 and then measuring the liberated phosphorus according to Bartlett 4. Finally, additional aliquots of the methanol eluates were subjected to quantitative TLC to determine the concentrations of the major myelin phospholipid constituents: ethanolamine phosphatides (EP), phosphatidylcholine (PC), sphingomyelin (SP), and phosphatidylserine (PS). These aliquots were evaporated to dryness under nitrogen gas, the lipids redissolved in small amounts of CM 2:1, and the solutions then applied to pre-activated (110 °C overnight) TLC plates coated with a 250/~m thick pre-channeled layer of silica gel G (Uniplates, Analtech). The plates were developed in a solvent system of chloroform-methanolconcentrated ammonium hydroxide 130:50:8 (by vol) 12. Spots were visualized by brief exposure to iodine vapor and their identities determined by comparison with standard phospholipids run on each plate. Despite the fact that most of the EP in myelin is of the plasmalogen form 25, phosphatidylethanolamine
(PE) was used as the standard because this chromatographic system does not separate PE from ethanolamine plasmalogen. Following sublimation of the iodine, the spots were scraped directly into assay tubes and their phospholipid content determined without prior elution. Samples were centrifuged briefly before the absorbance reading to spin down the silica gel. Because the silica gel itself gave a small but measurable reaction in the assay, non-lipid containing areas comparable in size to the spots were also scraped and the readings obtained from these silica gel blanks subtracted from the appropriate sample values.
CNP The specific activity of CNP (2':3'-cyclic-nucleotide 3'-phosphodiesterase; EC 3.1.4.37), a myelinrelated enzyme (see Discussion), was determined in cerebral myelin suspensions and in optic nerve homogenates by the method of Weissbarth et al. 34. This procedure uses 2':3'-cyclic-NADP as the substrate 3] and is based on the conversion of the N A D P ÷ product to N A D P H using glucose-6-phosphate and glucose-6-phosphate dehydrogenase present in excess. Myelin and optic nerve samples were prepared as indicated above and small portions then diluted 1:400 and 1:150 respectively in ice-cold 1% Triton X-100. Aliquots of 10/~1 of each diluted sample were added to 0.2 ml of the enzyme reagent described by Weissbarth et al. 34 and incubated at 25 °C for 1 h with gentle shaking. The reaction was stopped by addition of 2.0 ml of 0.05 M sodium bicarbonate, pH 10.5, and the N A D P H fluorescence measured in a Perkin-Elmer No. 1000 spectrofluorometer using excitation and emission wavelengths of 340 and 460 nm respectively. Blanks and N A D P + standards (5-40 nmol) were run along with the samples. Proteins Myelin suspensions were solubilized in 5% sodium dodecyl sulfate in 0.5 M NaOH at room temperature for 1 h and then assayed for protein according to the Lees and Paxman 19modification of the Lowry et al. 21 procedure. Optic nerve homogenates were incubated in 0.5 M NaOH at 40 °C overnight and then assayed by the method of Lowry et al. 21. In both cases, bovine serum albumin (BSA) served as the standard.
Chemicals The following items were purchased from the Sigma Chemical Co.: cholesterol, galactose, orcinol (recrystallized twice from benzene before use), 2' :3'cyclic-NADP, and BSA (fraction V). Cholesterol esterase (26 U/mg), cholesterol oxidase (25 U/mg), and N A D P (monopotassium salt) were obtained from Boehringer Mannheim Biochemicals. Silicic acid (Unisil, 100-200 mesh) came from Clarkson Chemical Co., azure A (71% dye content) from Aldrich Chemical Co., bovine sulfatides and the phospholipid standards (PC, PE, PS, and SP) from Supelco Co., and pre-mixed 'Fiske-Subbarow' reagent (in dry powder form) for phosphate analysis from Fisher Scientific Co. Other reagents were purchased from various suppliers and were of reagent grade wherever possible. Statistics Body weight, cerebral weight, myelin weight, and optic nerve CNP data were all subjected to analyses of variance followed by Fisher's Least Significant Difference post-hoe tests 17 where appropriate. Myelin biochemical data were subjected to Student's ttests. Results occurring with a chance probability of less than 0.05 were considered statistically significant.
5O0
40(I
~
I
ADX ~
•/'
2o0
20
40
60
"- .....
80
100
120
-
ADX 9
140
160
Age Idaysl
Fig. 1. Body weights (mean + S.E.M.) of male (M) and female (F) rats adrenalectomized (ADX) or sham-operated (SHAM) on the llth day of life and then killed on day 35-36, 63, or 151-153. Group sizes were as follows: day 35-36 ADX = 10 (4 M, 6 F), SHAM = 8 (4 M, 4 F); day 63 ADX = 10 (6 M, 4 F), SHAM = 8 (5 M, 3 F); day 151-153 ADX = 6 (4 M, 2 F), SHAM = 6 (4 M, 2 F). ADX males and females were significantly different from the respective SHAM rats at all time points (all differences at least P < 0.05).
RESULTS 50
Rats adrenalectomized on day 11 of life displayed significantly lower body weights than SHAM controls at all time points (Fig. 1), the difference being greatest at 151-153 days of age. This result is in accord with previous findings 23 and is thought to reflect the influence of adrenalectomy on peripheral energy metabolism and the deposition of fat stores. Despite the body weight reduction, the cerebral wet weights of the A D X animals were considerably greater than those of the SHAMs (10.0% increase at day 35-36, 15.3% at day 63, and 16.7% at day 151-153; see Fig. 2). The groups were already different at day 35-36, and diverged further with age. Thus, the adrenalectomy-induced increase in brain weight is not simply due to accelerated maturation (in which case the SHAMs might be expected to be 'catching up' by day 151-153), but rather appears to be a large, permanent increment in tissue mass. It should also be noted that although the data for male and female rats are not shown separately in Fig. 2, male rats had significantly heavier cerebra than female rats in all cases (P < 0.001). The results of the myelin isolation procedure (Fig. 3) partially paralleled the brain weight results in that the amount of recovered cerebral myelin (dry weight following lyophilization) was greater in the A D X animals than in the SHAMs and the two groups
1.8
ADX 1.6 .c
1.4 .....
~'---
SHAM
1.2
1.0
t
i
J
20
40
60
_ _ /
80
~
i
i
t
100
120
140
160
Age (days)
Fig. 2. Cerebral wet-weights (mean _+S.E.M.) of the same rats shown in Fig. 1. Although the data analysis included sex as a factor, the values were pooled for presentation in order to emphasize the treatment effect. Statistical analysis revealed a significant main effect of treatment (P < 0.001) and a treatment × age interaction (P < 0.0l).
J
J
J
F-
SHAM
g~
20
40
60
80
100
120
140
160
Age (days]
Fig. 3. Dry weights (mean + S.E.M.) of myelin recovered from the cerebra shown in Fig. 2. Data analysis and presentation as in Fig. 2. The analysis yielded a significant main effect of treatment (P<0.001) and a treatment × age interaction (e < 0.001).
again diverged with increasing age. In this case, however, the differences were significant only at the day63 (41.7% increase in recovered myelin) and 151-153 (42.1% increase) time points, suggesting that enhanced myelination contributed measurably to the brain weight effect at those ages but not at day 35-36. Likewise, cerebra from male rats yielded more myelin than cerebra from females (P < 0.001; data not shown), but only at the two later points. Fig. 4 shows a scatterplot of recovered myelin dry weight vs cerebral wet weight for the individual 63day-old A D X and SHAM animals. The results have been presented in this way not only to demonstrate the almost complete lack of overlap between the two treatment groups with regard to these variables, but also to illustrate the relationship between the variables. A least-squares regression analysis performed on these data revealed that the amount of myelin recovered from the cerebra was a linear function of the cerebral weights (r = +0.93, P < 0.01). Given that the analysis included both male and female rats from each treatment condition, a single function was therefore capable of closely predicting the amount of myelin recovered from animals with differing cerebral weights, regardless of whether the weight differences were treatment- or sex-related (see above). Similar regression analyses were performed on the day 35-36 and 151-153 results. The correlation was only + 0.40 (not significant) on day 35-36, which is not surprising considering that adrenalectomy pro-
6 duced a clear brain weight increase at that age but had much less effect on myelination (Fig. 3). On the other hand, the correlation between cerebral weight
40
30
y = 38.27x - 32.60 E
/@
r = + .93
.E
o/
20
?
O
//~
ee-
O
•
SHAM ADX
10
///./
1
L
i
i
i
1.0
1.2
1.4
1.6
1.8
Cerebral
Wet
Weight
Ig I
Fig. 4. Scatterplot of recovered myelin dry weight vs cerebral wet weight for individual A D X and S H A M rats comprising the day-63 group. The best-fit straight line and correlation coefficient for all the points taken together are also shown.
and the amount of recovered myelin was + 0.92 (P < 0.01) for the data from day 151-153. These data yielded a steeper line (y = 56.45x-48.61) than that shown in Fig. 4, reflecting the greater quantity of myelin obtained per unit brain weight. Portions of the myelin isolated from the 63-day-old rats were analyzed for various lipid constituents, total protein content, and CNP activity. As indicated in Table I, there were no differences between ADX and SHAM animals in terms of myelin cholesterol, protein, or CNP specific activity. However, myelin from the ADX animals was significantly deficient in total galactolipid (not including gangliosides) and phospholipid. Sulfatides were only slightly reduced, suggesting that the principal deficit occurred in the remaining galactolipid, most of which is cerebroside 25. The decrease in phospholipid was observed to some extent across all of the major myelin phospholipid classes (Table I), although the effect was statistically significant only in the case of PC. Finally, it is curious to note that of the myelin components measured in this experiment, none was increased in the ADX animals (see Discussion). In addition to characterizing the effect of day-ll adrenalectomy on the cerebrum, a region where myelination of many tracts begins relatively late and occurs at a slow rate 14, we were also interested in examining a system that develops earlier and more rapidly. For this purpose we chose the optic nerve, a tract that begins to myelinate around day 6 postna-
TABLE I
Composition of myelin isolated from 63-day-old adrenalectomized (ADX) and sham-operated (SHAM) rats Data represent the m e a n + S.E.M. for 8 S H A M and 10 A D X rats except for the individual TLC-separated phospholipids, which are based on 6 animals from each group.
SHAM Cholesterol (~g/mg myelin) Total galactolipid (,ug/mg) Sulfatides Total phospholipid (nmol/mg) Ethanolamine phosphatides Phosphatidylcholine Phosphatidylserine Sphingomyelin % of total phospholipid represented by the above c o m p o u n d s Total protein ~ g / m g ) CNP activity (~mol/mg prot./h)
171.3 203.6 44.6 373.6 166.1 94.1 61.7 25.7
+ + + + + + + +
ADX 3.2 4.9 1.1 5.5 2.4 1.7 0.9 0.6
92.6% 255.7 +_ 2.5 1441 + 40
* P < 0.05, ** P < 0.01, *** P < 0.001 compared to S H A M group.
172.2 176.5 41.9 358.2 158.9 86.2 57.3 23.7
___3.3 + 3.7*** + 1.0 + 3.2" + 2.8 + 1.5"* + 2.3 + 1.0
91.9% 255.1 + 6.4 1326 + 46
taP e, 5 days prior to the surgery. Rather than attempt to isolate and quantify purified myelin from such a small tissue, we measured the specific activity of rCNP directly in optic nerve homogenates, using the enzyme as an index of myelin formation30. It is clear from Table II that adrenalectomy had no effect on CNP specific activity at any age. This could mean that myelination in the optic nerve was unaffected by the treatment or alternatively that the amount of myelin was increased but to the same extent as total tissue protein. The latter possibility could have been tested by comparing total CNP activity in the entire optic nerves, but this was precluded by incomplete removal of the tissue from some animals in the course of rapid dissection. DISCUSSION Our results demonstrate first that the increased brain weight produced by early adrenalectomy persists well into adulthood, with no sign of a late convergence toward the brain weight of control animals, This stimulation of brain growth occurs despite a concomitant reduction in body weight and therefore may represent a unique response of the brain to the loss of adrenal secretions. Apart from their increased size, the brains of A D X rats appear normal upon gross observation. We assume that the cranium of these animals is likewise enlarged to provide the needed space, although no actual measurements of cranial volume have yet been performed. The amount of purified myelin that could be recovered from the cerebrum was greatly increased in A D X vs SHAM subjects at 63 and 151-153 days of age. Although a small trend in the same direction was
TABLE II 2':3'-Cyclic-nucleotide 3'-phosphodiesterase (CNP) activity in optic nerve homogenates from adrenalectomized (ADX) and sham-operated (SHAM) rats
Data represent the mean ± S.E.M. for the number of animals shown in parentheses. Age
35-36 days 63 days 151-153 days
CNP (izol/mg protein~h) SHAM
ADX
1084 ± 32 (8) 819±47 (8) 855±46 (6)
1084 ± 14 (10) 847±52(10) 898 ± 32 (6)
apparent on day 35-36, the effect was not statistically significant. This was somewhat surprising because measurable brain weight differences are present at that age and because myelination is proceeding rapidly from the time of the surgery up to about 30 days of age26. We cannot be certain that the differences between the A D X and SHAM animals in recovered myelin weight represent the exact differences that existed in vivo because we did not have a measure of percentage myelin recovery. On the other hand, our isolation procedure did yield quantitatively reproducible amounts of myelin within each age and treatment group (note the small S.E.M. bars in Fig. 3 ) , and the yield in the older subjects was closely related to the size of each brain sample. Furthermore, our earlier work showed that whole cerebral CNP activity (i.e., without prior myelin isolation) in 65-day-old A D X rats was increased by 29% over that of SHAM animals 23. Therefore, it seems reasonable to conclude in the present study that at least by adulthood, substantially more myelin was present in the brains of the A D X rats than in the controls. As we previously demonstrated an elevated cerebral D N A content in A D X animals at 65 days of age 23, perhaps our treatment results in an increased number of oligodendrocytes, thereby leading to increased myelination. Myelination could be altered by changes in the percentage of axons myelinated, total number of axons (of which the same percentage is myelinated), and/or sheath thickness. Morphometric studies currently in progress should provide further information concerning the effects of adrenalectomy on axon-myelin relationships. Myelin isolated from the A D X animals was found to be measurably deficient in total galactolipid and phospholipid concentrations. It should be noted that such differences were not detected in a previous study 23, however, in that study fewer animals were examined and the animals were of a different strain (Wistar). Furthermore, at least in the case of the galactolipids, the assay method used here was more sensitive and precise. On the other hand, if an average phospholipid molecular weight of 775 is used to substitute for the molar values reported in Table I, it is possible to calculate total recovery of myelin constituents, including protein. The results indicate recoveries of only 92% and 88% for the SHAM and A D X groups respectively. Although it is the case that
several lipid classes (notably the gangliosides and polyphosphoinositides) were not measured in our protocol, these compounds represent only a small fraction of the dry weight of normal rat myelin 25. Because we found no myelin component that was increased in the A D X animals, the possibility must be considered that the difference in myelin composition between the groups was due to differential recovery of lipids (e.g. from the silicic acid columns) in the two sets of samples. This hypothesis will need to be addressed in subsequent studies. Myelination does not commence simultaneously in all CNS areas because of regional variation in the timetables of oligodendrocyte differentiation and axonal growth. Early myelinating tracts such as the optic nerve might be relatively less vulnerable than late myelinating tracts to the effects of adrenalectomy on day 11 postnatal. Indeed, we failed to find any difference in optic nerve CNP specific activity between ADX and SHAM animals. CNP is often used as a marker for myelination 30, although its activity is actually highest in so-called 'myelin-like' fractions22. 23 which are thought to consist primarily of uncompacted myelin and oligodendroglial plasma membranes undergoing transition to myelin. For this reason and also because of the point raised earlier concerning interpretation of specific activity values, the failure to observe CNP changes should not be taken as unequivocal evidence that optic nerve myelination is unaffected by removal of the adrenal glands. Although no hormone replacement condition was run in the present series of experiments, it is obviously important to determine which hormones are involved in the increased brain weight and myelination effects. Adrenalectomy results in a complex set of endocrine changes, including the loss of cortical steroids and medullary catecholamines, and an increase in circulating A C T H concentrations due to the absence of corticosteroid negative feedback. Studies cited in the Introduction have shown that exogenous glucocorticoid administration early in life can reduce subsequent myelination, hence it is conceivable that the removal of endogenous glucocorticoids might have the opposite effect. Furthermore, Devenport and Devenport 8 reported that corticosterone replacement could reverse the increased brain weight found after adrenalectomy on day 25 of life, although these investigators did not study myelination in their
experiments. These considerations suggest that the effects reported here are most likely to have resulted from the loss of glucocorticoids, but this conclusion must remain tentative until the appropriate replacement studies have been performed. Following submission of this manuscript for publication, a paper appeared by Preston and McMorris27 in which rats were adrenalectomized on the 14th day of life, weaned at day 18 and given drinking solutions of 5% glucose and 0.9% saline (used instead of the present mineralocorticoid replacement), and then sacrificed on the 21st or 22nd day of life for myelin determinations. In contrast to our present results, Preston and McMorris observed a 25% reduction in the amount of myelin isolated from the A D X animals which they attributed to adrenalectomy-related decreases in G P D H activity (see Introduction). We attempted to replicate these findings using the Preston and McMorris protocol but none of the rats survived more than 1-2 days following weaning. We then performed additional experiments with all animals (SHAM as well as ADX) given Percorten at the time of surgery and left with their mothers until sacrifice. Under these conditions, virtually all of the rats survived and we observed a small (non-significant) increase in myelination despite a significant reduction in G P D H activity. Given the fragile nature of early A D X rats, we would argue that the hypomyelination reported by Preston and McMorris was a secondary consequence of the premature weaning and the attendant anorexia produced by such treatment. These issues are discussed in greater detail elsewhere (J. S. Meyer and K. R. Fairman, in preparation). In conclusion, we have begun to describe some of the cellular and biochemical changes underlying the increase in brain weight seen following early adrenalectomy. Although the present paper focuses on adrenalectomy-induced changes in myelination, it should be noted that if myelin comprises an estimated 20% of the dry weight of a 60-day-old rat brain26, then even a 40% increase in myelin deposition could not nearly account for a 15% increase in overall brain weight (as shown in Meyer 23, the remaining difference is not due simply to elevated brain water content). Early adrenalectomy undoubtedly influences a variety of neural elements and thus provides an important new model for studying plasticity of the developing brain.
ACKNOWLEDGEMENTS
typing the final m a n u s c r i p t . This r e s e a r c h was supp o r t e d by G r a n t BNS-8118073 f r o m the N a t i o n a l Sci-
W e w o u l d like to t h a n k Ms. M e l a n i e B e l l e n o i t for
REFERENCES 1 Allain, C. C., Poon, L. S., Chart, C. S. G., Richmond, W. and Fu, P. C., Enzymatic determination of total serum cholesterol, Clin. Chem., 20 (1974) 470-475. 2 Balazs, R., Brooksbank, B. W. L., Patel, A. J., Johnson, A. L. and Wilson, D. A., Incorporation of [35S]sulfate into brain constituents during development and the effects of thyroid hormone on myelination, Brain Res., 30 (1971) 273-293. 3 Balazs, R., Patel, A. J. and Hajos, F., Factors affecting the biochemical maturation of the brain: effects of hormones during early life, Psychoneuroendocrinology, 1 (1975) 25-36. 4 Bartlett, G. R., Phosphorus assay in column chromatography, J. biol. Chem., 234 (1959) 466-468. 5 Bohn, M. C. and Friedrich, V. L. Jr., Recovery of myelination in rat optic nerve after developmental retardation by cortisol, J. Neurosci., 2 (1982) 1292-1298. 6 Casper, R., Vernadakis, A. and Timiras, P. S., Influence of estradiol and cortisol on lipids and cerebrosides in the developing brain and spinal cord of the rat, Brain Res., 5 (1967) 524-525. 7 De Vellis, J. and Inglish, D., Hormonal control of glycerolphosphate dehydrogenase in the rat brain, J. Neurochem., 15 (1968) 1061-1070. 8 Devenport, L. D. and Devenport, J. A., The effects of adrenal hormones on brain and body size, Physiol. Psychol., 10 (1982) 399-404. 9 Field, E. J., Effect of cortisone on the neonatal rat, Nature (Lond.), 174 (1954) 182. 10 Folch, J., Lees, M. and Sloane Stanley, G. H., A simple method for the isolation and purification of total lipides from animal tissues, J. biol. Chem., 226 (1957) 497-509. 11 Gumbinas, M., Oda, M. and Huttenlocher, P., The effects of corticosteroids on myelination of the developing rat brain, Biol. Neonate, 22 (1973) 355-366. 12 Horrocks, L. A., Thin-layer chromatography of brain phospholipids, J. Amer. Oil Chem. Soc., 40 (1963) 235-236. 13 Howard, E. and Benjamins, J. A., DNA, ganglioside and sulfatide in brains of rats given corticosterone in infancy, with an estimate of cell loss during development, Brain Res., 92 (1975) 73-87. 14 Jacobson, S., Sequence of myelination in the brain of the albino rat. A. Cerebral cortex, thalamus and related structures, J. comp. Neurol., 121 (1963) 5-29. 15 Jungalwala, F. B. and Dawson, R. M. C., The turnover of myelin phospholipids in the adult and developing rat brain, Biochem. J., 123 (1971) 683-693. 16 Kean, E. L., Rapid, sensitive spectrophotometric method for quantitative determination of sulfatides, J. Lipid Res., 9 (1968) 319-327. 17 Kirk, R. E., Experimental Design: Procedures for the Behavioral Sciences, Brooks/Cole, Belmont, CA, 1968, 577 PP. 18 Kornberg, A. and Pricer, W. E., Enzymatic esterification of a-glycerophosphate by long chain fatty acids, J. biol. Chem., 204 (1953) 345-347.
ence Foundation.
19 Lees, M. B. and Paxman, S., Modification of the Lowry procedure for the analysis of proteolipid protein, Anal. Biochem., 47 (1972) 184-192. 20 Leveille, P. J., McGinnis, J. F., Maxwell, D. S. and de Vellis, J., lmmunocytochemical localization of glycerol-3phosphate dehydrogenase in rat oligodendrocytes, Brain Res., 196 (1980) 287-305. 21 Lowry, O. J., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 22 Matthieu, J. M., Quarles, R. H., Brady, R. O. and Webster, H. deF., Variation of proteins, enzyme markers and gangliosides in myelin subfractions, Biochim. biophys. Acta, 329 (1973) 305-317. 23 Meyer, J. S., Early adrenalectomy stimulates subsequent growth and development of the rat brain, Exp. Neurol., 82 (1983) 432-446. 24 Noguchi, T., Sugisaki, T., Watanabe, M., Kohsaka, S. and Tsukada, Y., Effects of bovine growth hormone on the retarded cerebral development induced by neonatal hydrocortisone intoxication, J. Neurochem., 38 (1982) 246-256. 25 Norton, W. T., Biochemistry of myelin. In S. G. Waxman and J. M. Ritchie (Eds.), Demyelinating Disease. Basic and Clinical Electrophysiology, Raven Press, New York, 1981, pp. 93-121. 26 Norton, W. T. and Poduslo, S. E., Myelination in rat brain: changes in myelin composition during brain maturation, J. Neurochem., 21 (1973) 759-773. 27 Preston, S. L. and McMorris, F. A., Adrenalectomy of rats results in hypomyelination of the central nervous system, J. Neurochem., 42 (1984) 262-267. 28 Rouser, G., Kritchevsky, G. and Yamamoto, A., Column chromatography of phosphatides and glycolipids. In G. V. Marinetti (Ed.), Lipid Chromatographic Analysis, 1Iol. 1, Marcel Dekker, New York, 1967, pp. 99-162. 29 Schapiro, S., Some physiological, biochemical, and behavioral consequences of neonatal hormone administration: cortisol and thyroxine, Gen. Comp. End., 10 (1968) 214-228. 30 Sims, N. R. and Carnegie, P. R., 2':3'-Cyclic-nucleotide 3'phosphodiesterase. In B. W. Agranoff and M. H. Aprison (Eds.), Advances in Neurochemistry, Vol. 3, Plenum, New York, 1978, pp. 1-41. 31 Sogin, D. C., 2':3'-Cyclic-NADP as a substrate for 2':3'cyclic-nucleotide 3'-phosphohydrolase, J. Neurochem., 27 (1976) 1333-1337. 32 Tennekoon, G. I., Cohen, S. R., Price, D. L. and McKhann, G. M., Myelinogenesis in optic nerve. A morphological, autoradiographic, and biochemical analysis, J. Cell Biol., 72 (1977) 604-616. 33 Waehneldt, T. V., Ontogenetic study of a myelin-derived fraction with 2':3'-cyclic-nucleotide 3'-phosphohydrolase activity higher than that of myelin, Biochem. J., 151 (1975) 435-437. 34 Weissbarth, S., Maker, H. S., Raes, I., Brannan, T. S., Lapin, E. P. and Lehrer, G. M., The activity of 2':3'-cyclicnucleotide 3'-phosphodiesterase in rat tissues, J. Neurochem., 37 (1981) 677-680.
I
BRD 50119
Research Reports
Early Adrenalectomy Increases Myelin Content of the Rat Brain JERROLD S. MEYER and KENNETH R. FAIRMAN
Division of Neuroscienceand Behavior, Departmentof Psychology, Universityof Massachusetts, Amherst, MA 01003 (U.S.A.) (Accepted June 19th, 1984)
Key words: myelin - - brain weight - - brain development - - adrenalectomy
Rats were adrenalectomized (ADX) or sham-operated (SHAM) on the llth day of life and killed on days 35-36, 63, or 151-153 for the isolation of cerebral myelin from each animal. Despite having lower overall body weights, ADX rats had heavier cerebra than SHAM control rats at all ages. Mean cerebral weight increases were 10.0% at day 35-36, 15.3% at day 63, and 16.7% at day 151-153. Recovered myelin dry weights were even more elevated in the ADX rats, but only at day 63 (41.7% increase) and 151-153 (42.1% increase). At both of these ages, there was a clear linear relationship between cerebral wet weights and the amount of myelin recovered from the cerebra. Analysis of the day-63 myelin samples showed no group differences in total cholesterol or protein concentration or in the specific activity of the myelin marker enzyme 2':3'-cyclic-nucleotide3'-phosphodiesterase (CNP). However, myelin isolated from the ADX rats appeared to be deficient in both galactolipid and phospholipid. Optic nerve myelination was assessed in all animals by measuring CNP activity in homogenates prepared from this tissue. No difference between ADX and SHAM rats was observed at any age. These results indicate that early adrenalectomy stimulates myelin deposition in the rat brain as part of a more general, long-lasting enhancement of brain growth. Myelin from the brains of ADX animals may be slightly abnormal in its lipid composition. Finally, the optic nerve data may mean that myelination is not affected equally in all areas of the CNS by the loss of adrenal glands. INTRODUCTION Considerable evidence suggests that adrenal hormones, particularly the glucocorticosteroids, can influence CNS myelinogenesis in developing rats. Most investigators have studied this p r o b l e m by administering pharmacological doses of glucocorticoids to young subjects and then examining biochemical or morphological indices of myelination at a later time. A n u m b e r of studies 5,9,11A3,24,29 have suggested that early glucocorticoid t r e a t m e n t retards the rate of myelin deposition, at least temporarily. Some of these studies m a d e use of biochemical m a r k e r s of myelination in their assessment of glucocorticoid effects. F o r e x a m p l e , cerebral sulfatide concentrations were significantly lower in rats t r e a t e d neonatally with corticosterone 13, while cerebral 2':3'-cyclic-nucleotide 3 ' - p h o s p h o d i e s t e r a s e (CNP) activity was
likewise reduced in animals given cortisol early in life 24. It is interesting to note that the latter effect, as well as various other deleterious consequences of glucocorticoid administration, was substantially reversed by daily growth h o r m o n e injections until the time of weaning 24. Ultrastructurai studies of myelination following early exposure to glucocorticoids have e x p a n d e d upon the above findings. G u m b i n a s et al. u e x a m i n e d axonal growth and myelination in the p y r a m i d a l tract of rats treated on the 6th day of life with a single injection of methylprednisolone. A reduction in the percentage of myelinated fibers (in relation to the total n u m b e r of fibers per unit area) was noted at both 3 and 6 weeks of age, although there was some evidence that the e x p e r i m e n t a l and control groups were converging at the later time point. The results also showed that the n u m b e r of lamellae as a function of
Correspondence and reprint requests: J. S. Meyer, Department of Psychology, University of Massachusetts, Amherst, MA 01003, U.S.A. 0165-3806/85/$03.30 © 1985 Elsevier Science Publishers B.V.
axon circumference was decreased in the glucocorticold-treated animals, indicating a breakdown in the normal relationship between axon size and myelin sheath development. Bohn and Friedrich 5 studied optic nerve myelination and oligodendrocyte proliferation in rats given cortisol from days 7-18 postnatal. As in the case of the pyramidal tract, the percentage of optic nerve axons that were myelinated was significantly lower in the glucocorticoid-treated subjects at day 21. In this instance, however, examination of older animals (60 days of age) revealed a complete normalization of all myelin-related parameters. Parallel studies of optic nerve cell proliferation using [3H]thymidine autoradiography revealed a stimulation of oligodendrocyte formation following the end of hormone administration. This effect may have been a compensatory response to the well-known suppressive action of high-dose glucocorticoid treatment on. neural cell division 3 and furthermore may have been responsible for the ultimate recovery of myelination in the experimental rats. Although the above-cited experiments show that glucocorticoid administration can inhibit CNS myelinogenesis in young rats, at least one study has suggested the opposite relationship. Casper et al. 6 reported that cortisol treatment from the 6th to the 10th day of life increased the cerebroside concentration in the spinal cord when measured on day 12. Furthermore, glucocorticoids stimulate the activity of cytoplasmic glycerol-3-phosphate dehydrogenase (GPDH) in rat brain 7, where the enzyme is localized primarily in oligodendrocytes 2°. Because G P D H is thought to be important in phosphoglyceride biosynthesis 18 and oligodendrocytes must synthesize large amounts of phospholipid for incorporation into myelin, Leveille et al.20 have proposed that glucocorticoids might have a positive influence on the maintenance and/or development of myelin sheaths in the rat brain. A question thus arises as to the exact role of the adrenal glands in the normal process of myelinogenesis, partially because of the inconsistencies noted above but also because the effects of high-dose glucocorticoid treatment may not reflect the actions of these hormones at their usual circulating concentrations. We have recently begun to reexamine the role of the adrenal glands in a number of aspects of brain development by studying rats that have been bilater-
ally adrenalectomized (ADX) on the l l t h day of life. Despite reductions in overall body weight, A D X rats showed significant increases in wet and dry brain weight (both cerebrum and cerebellum) and in cerebral D N A content when compared to sham-operated (SHAM) controls at days 65-67 of life 23. We also found a substantial increase in the amount of purified myelin that could be recovered from the cerebra of A D X animals killed at that age. Hormone replacement studies are still in progress, hence we have not yet determined whether these changes are due to the loss of glucocorticoids or other adrenal secretions. The present paper further characterizes the relationship between the adrenal glands and myelination, focusing on the time course of the adrenalectomy effect and also providing a detailed lipid analysis of the myelin isolated from A D X vs SHAM subjects at one of the time points. The results show that early adrenalectomy produces long-term increases in brain myelin content and that the myelin formed under these conditions may not be completely normal in its composition. MATERIALS AND METHODS Animals and surgery The animals used were male and female SpragueDawley rats (Charles River CD strain) bred in our laboratory under previously described conditions 23. Litters were culled to 8 pups (sex-balanced where possible) on the day after birth. On the l l t h day of life (counting the first day that pups were observed as day 1), rats were subjected to bilateral adrenalectomy or sham operation (laparotomy only) under ether anesthesia. Prior to weaning, A D X animals were maintained on the mineralocorticoid deoxycorticosterone pivalate (Percorten Pivalate, Ciba Pharmaceutical Co.) as previously described 23. Such treatment promotes sodium retention and is necessary for the survival of the A D X rats. SHAM controls received saline vehicle instead, inasmuch as Percorten was previously found to have no effect on brain growth in such animals (J. S. Meyer, K. R. Fairman, and M. Friedman, unpublished observations). The subjects were weaned on day 30 of life, transferred to standard wire-bottom group cages, and given ad libiturn access to Purina Rodent Laboratory Chow, tap water, and 0.85% saline (SHAM as well as A D X
subjects). All animals were maintained under a 14:10 light-dark cycle (lights on at 06.00 h) at a temperature of approximately 22 °C.
Sample preparation On days 35-36, 63, or 151-153 of life, rats were weighed, decapitated, and their brains (minus the olfactory bulbs) rapidly removed and placed on a chilled ground-glass plate. A single vertical cut was made between the cerebral hemispheres and the cerebellum and the resulting tissue sample (cerebrum) used for myelin isolation. Each cerebrum was weighed, rinsed with ice-cold 0.32M sucrose, minced finely, and homogenized in 10 vols. of 0.32 M sucrose using 4 slow strokes at 300 rpm with a 0.20-0.23 mm clearance Teflon-glass Duall homogenizer (Kontes). This and all subsequent steps were carried out at 4 °C. Myelin was isolated from the crude homogenates by the procedure of Jungawala and Dawson 15, omitting the final continuous density gradient step. The myelin was then washed 3 times in deionized water to remove residual sucrose. The purity of the myelin isolated by this method has been confirmed biochemically by the absence of the microsomal marker enzyme NADPH-cytochrome c reductase and the mitochondrial marker cytochrome c oxidase 15. We have also examined the myelin by standard transmission electron microscopy and found it to consist exclusively of membranous profiles with many multilamellar structures. In the case of the 35-36- and 151-153-day-old animals, all of the recovered myelin was lyophilized in tared tubes to obtain the dry weight of each sample. For the 63-day-old group, the myelin was suspended in deionized water to a total volume of 10.0 ml. One ml of this suspension was frozen at -40 °C for subsequent CNP and protein determination (see below), while the remainder was lyophilized. The lyophilized samples were weighed (weight data presented below have been corrected to reflect total recovered myelin) and then later processed for lipid determinations. The intracranial portion of the optic nerves from each animal in all age groups was dissected at the same time as the brain samples. Optic nerves were homogenized in 1.0 ml of ice-cold deionized water and frozen at -40 °C for measurement of CNP and protein.
Lipid extraction and analysis Lyophilized myelin from the 63-day-old animals was dissolved in chloroform-methanol (CM) 2:1 (vol./vol.) and the resulting mixture filtered through glass wool to remove the insoluble protein. The lipids were partitioned by the addition of 0.2 vols. of 0.88% aqueous KC110, followed by two washes of the chloroform-rich lower phase with KCl-containing Folch 'theoretical upper phase'. The lower phase was subsequently evaporated to dryness at 40 °C in vacuo and the lipid residue redissolved in CM 2:1. A portion of each sample was subjected to silicic acid column chromatography in which cholesterol, glycolipids, and phospholipids were eluted with chloroform, acetone, and methanol respectively28. All eluent fractions were stored at 4 °C for subsequent analysis. Total cholesterol (free plus esterified) in the chloroform eluates was determined by the enzymaticspectrophotometric method of Allain et al. 1. The acetone eluates were assayed for lipid-bound galactose by the orcinol-sulfuric acid method of Balazs et al. 2. Galactose values were multiplied by 4.7 to estimate myelin galactolipid weight 26. A modification of Kean's ~6azure A dye-binding method was used to determine myelin sulfatides in the same acetone fractions. Total phospholipid phosphorus was assayed in each methanol fra~ction by first hydrolyzing the lipids using sulfuric acid and hydrogen peroxide as previously described23 and then measuring the liberated phosphorus according to Bartlett 4. Finally, additional aliquots of the methanol eluates were subjected to quantitative TLC to determine the concentrations of the major myelin phospholipid constituents: ethanolamine phosphatides (EP), phosphatidylcholine (PC), sphingomyelin (SP), and phosphatidylserine (PS). These aliquots were evaporated to dryness under nitrogen gas, the lipids redissolved in small amounts of CM 2:1, and the solutions then applied to pre-activated (110 °C overnight) TLC plates coated with a 250/~m thick pre-channeled layer of silica gel G (Uniplates, Analtech). The plates were developed in a solvent system of chloroform-methanolconcentrated ammonium hydroxide 130:50:8 (by vol) 12. Spots were visualized by brief exposure to iodine vapor and their identities determined by comparison with standard phospholipids run on each plate. Despite the fact that most of the EP in myelin is of the plasmalogen form 25, phosphatidylethanolamine
(PE) was used as the standard because this chromatographic system does not separate PE from ethanolamine plasmalogen. Following sublimation of the iodine, the spots were scraped directly into assay tubes and their phospholipid content determined without prior elution. Samples were centrifuged briefly before the absorbance reading to spin down the silica gel. Because the silica gel itself gave a small but measurable reaction in the assay, non-lipid containing areas comparable in size to the spots were also scraped and the readings obtained from these silica gel blanks subtracted from the appropriate sample values.
CNP The specific activity of CNP (2':3'-cyclic-nucleotide 3'-phosphodiesterase; EC 3.1.4.37), a myelinrelated enzyme (see Discussion), was determined in cerebral myelin suspensions and in optic nerve homogenates by the method of Weissbarth et al. 34. This procedure uses 2':3'-cyclic-NADP as the substrate 3] and is based on the conversion of the N A D P ÷ product to N A D P H using glucose-6-phosphate and glucose-6-phosphate dehydrogenase present in excess. Myelin and optic nerve samples were prepared as indicated above and small portions then diluted 1:400 and 1:150 respectively in ice-cold 1% Triton X-100. Aliquots of 10/~1 of each diluted sample were added to 0.2 ml of the enzyme reagent described by Weissbarth et al. 34 and incubated at 25 °C for 1 h with gentle shaking. The reaction was stopped by addition of 2.0 ml of 0.05 M sodium bicarbonate, pH 10.5, and the N A D P H fluorescence measured in a Perkin-Elmer No. 1000 spectrofluorometer using excitation and emission wavelengths of 340 and 460 nm respectively. Blanks and N A D P + standards (5-40 nmol) were run along with the samples. Proteins Myelin suspensions were solubilized in 5% sodium dodecyl sulfate in 0.5 M NaOH at room temperature for 1 h and then assayed for protein according to the Lees and Paxman 19modification of the Lowry et al. 21 procedure. Optic nerve homogenates were incubated in 0.5 M NaOH at 40 °C overnight and then assayed by the method of Lowry et al. 21. In both cases, bovine serum albumin (BSA) served as the standard.
Chemicals The following items were purchased from the Sigma Chemical Co.: cholesterol, galactose, orcinol (recrystallized twice from benzene before use), 2' :3'cyclic-NADP, and BSA (fraction V). Cholesterol esterase (26 U/mg), cholesterol oxidase (25 U/mg), and N A D P (monopotassium salt) were obtained from Boehringer Mannheim Biochemicals. Silicic acid (Unisil, 100-200 mesh) came from Clarkson Chemical Co., azure A (71% dye content) from Aldrich Chemical Co., bovine sulfatides and the phospholipid standards (PC, PE, PS, and SP) from Supelco Co., and pre-mixed 'Fiske-Subbarow' reagent (in dry powder form) for phosphate analysis from Fisher Scientific Co. Other reagents were purchased from various suppliers and were of reagent grade wherever possible. Statistics Body weight, cerebral weight, myelin weight, and optic nerve CNP data were all subjected to analyses of variance followed by Fisher's Least Significant Difference post-hoe tests 17 where appropriate. Myelin biochemical data were subjected to Student's ttests. Results occurring with a chance probability of less than 0.05 were considered statistically significant.
5O0
40(I
~
I
ADX ~
•/'
2o0
20
40
60
"- .....
80
100
120
-
ADX 9
140
160
Age Idaysl
Fig. 1. Body weights (mean + S.E.M.) of male (M) and female (F) rats adrenalectomized (ADX) or sham-operated (SHAM) on the llth day of life and then killed on day 35-36, 63, or 151-153. Group sizes were as follows: day 35-36 ADX = 10 (4 M, 6 F), SHAM = 8 (4 M, 4 F); day 63 ADX = 10 (6 M, 4 F), SHAM = 8 (5 M, 3 F); day 151-153 ADX = 6 (4 M, 2 F), SHAM = 6 (4 M, 2 F). ADX males and females were significantly different from the respective SHAM rats at all time points (all differences at least P < 0.05).
RESULTS 50
Rats adrenalectomized on day 11 of life displayed significantly lower body weights than SHAM controls at all time points (Fig. 1), the difference being greatest at 151-153 days of age. This result is in accord with previous findings 23 and is thought to reflect the influence of adrenalectomy on peripheral energy metabolism and the deposition of fat stores. Despite the body weight reduction, the cerebral wet weights of the A D X animals were considerably greater than those of the SHAMs (10.0% increase at day 35-36, 15.3% at day 63, and 16.7% at day 151-153; see Fig. 2). The groups were already different at day 35-36, and diverged further with age. Thus, the adrenalectomy-induced increase in brain weight is not simply due to accelerated maturation (in which case the SHAMs might be expected to be 'catching up' by day 151-153), but rather appears to be a large, permanent increment in tissue mass. It should also be noted that although the data for male and female rats are not shown separately in Fig. 2, male rats had significantly heavier cerebra than female rats in all cases (P < 0.001). The results of the myelin isolation procedure (Fig. 3) partially paralleled the brain weight results in that the amount of recovered cerebral myelin (dry weight following lyophilization) was greater in the A D X animals than in the SHAMs and the two groups
1.8
ADX 1.6 .c
1.4 .....
~'---
SHAM
1.2
1.0
t
i
J
20
40
60
_ _ /
80
~
i
i
t
100
120
140
160
Age (days)
Fig. 2. Cerebral wet-weights (mean _+S.E.M.) of the same rats shown in Fig. 1. Although the data analysis included sex as a factor, the values were pooled for presentation in order to emphasize the treatment effect. Statistical analysis revealed a significant main effect of treatment (P < 0.001) and a treatment × age interaction (P < 0.0l).
J
J
J
F-
SHAM
g~
20
40
60
80
100
120
140
160
Age (days]
Fig. 3. Dry weights (mean + S.E.M.) of myelin recovered from the cerebra shown in Fig. 2. Data analysis and presentation as in Fig. 2. The analysis yielded a significant main effect of treatment (P<0.001) and a treatment × age interaction (e < 0.001).
again diverged with increasing age. In this case, however, the differences were significant only at the day63 (41.7% increase in recovered myelin) and 151-153 (42.1% increase) time points, suggesting that enhanced myelination contributed measurably to the brain weight effect at those ages but not at day 35-36. Likewise, cerebra from male rats yielded more myelin than cerebra from females (P < 0.001; data not shown), but only at the two later points. Fig. 4 shows a scatterplot of recovered myelin dry weight vs cerebral wet weight for the individual 63day-old A D X and SHAM animals. The results have been presented in this way not only to demonstrate the almost complete lack of overlap between the two treatment groups with regard to these variables, but also to illustrate the relationship between the variables. A least-squares regression analysis performed on these data revealed that the amount of myelin recovered from the cerebra was a linear function of the cerebral weights (r = +0.93, P < 0.01). Given that the analysis included both male and female rats from each treatment condition, a single function was therefore capable of closely predicting the amount of myelin recovered from animals with differing cerebral weights, regardless of whether the weight differences were treatment- or sex-related (see above). Similar regression analyses were performed on the day 35-36 and 151-153 results. The correlation was only + 0.40 (not significant) on day 35-36, which is not surprising considering that adrenalectomy pro-
6 duced a clear brain weight increase at that age but had much less effect on myelination (Fig. 3). On the other hand, the correlation between cerebral weight
40
30
y = 38.27x - 32.60 E
/@
r = + .93
.E
o/
20
?
O
//~
ee-
O
•
SHAM ADX
10
///./
1
L
i
i
i
1.0
1.2
1.4
1.6
1.8
Cerebral
Wet
Weight
Ig I
Fig. 4. Scatterplot of recovered myelin dry weight vs cerebral wet weight for individual A D X and S H A M rats comprising the day-63 group. The best-fit straight line and correlation coefficient for all the points taken together are also shown.
and the amount of recovered myelin was + 0.92 (P < 0.01) for the data from day 151-153. These data yielded a steeper line (y = 56.45x-48.61) than that shown in Fig. 4, reflecting the greater quantity of myelin obtained per unit brain weight. Portions of the myelin isolated from the 63-day-old rats were analyzed for various lipid constituents, total protein content, and CNP activity. As indicated in Table I, there were no differences between ADX and SHAM animals in terms of myelin cholesterol, protein, or CNP specific activity. However, myelin from the ADX animals was significantly deficient in total galactolipid (not including gangliosides) and phospholipid. Sulfatides were only slightly reduced, suggesting that the principal deficit occurred in the remaining galactolipid, most of which is cerebroside 25. The decrease in phospholipid was observed to some extent across all of the major myelin phospholipid classes (Table I), although the effect was statistically significant only in the case of PC. Finally, it is curious to note that of the myelin components measured in this experiment, none was increased in the ADX animals (see Discussion). In addition to characterizing the effect of day-ll adrenalectomy on the cerebrum, a region where myelination of many tracts begins relatively late and occurs at a slow rate 14, we were also interested in examining a system that develops earlier and more rapidly. For this purpose we chose the optic nerve, a tract that begins to myelinate around day 6 postna-
TABLE I
Composition of myelin isolated from 63-day-old adrenalectomized (ADX) and sham-operated (SHAM) rats Data represent the m e a n + S.E.M. for 8 S H A M and 10 A D X rats except for the individual TLC-separated phospholipids, which are based on 6 animals from each group.
SHAM Cholesterol (~g/mg myelin) Total galactolipid (,ug/mg) Sulfatides Total phospholipid (nmol/mg) Ethanolamine phosphatides Phosphatidylcholine Phosphatidylserine Sphingomyelin % of total phospholipid represented by the above c o m p o u n d s Total protein ~ g / m g ) CNP activity (~mol/mg prot./h)
171.3 203.6 44.6 373.6 166.1 94.1 61.7 25.7
+ + + + + + + +
ADX 3.2 4.9 1.1 5.5 2.4 1.7 0.9 0.6
92.6% 255.7 +_ 2.5 1441 + 40
* P < 0.05, ** P < 0.01, *** P < 0.001 compared to S H A M group.
172.2 176.5 41.9 358.2 158.9 86.2 57.3 23.7
___3.3 + 3.7*** + 1.0 + 3.2" + 2.8 + 1.5"* + 2.3 + 1.0
91.9% 255.1 + 6.4 1326 + 46
taP e, 5 days prior to the surgery. Rather than attempt to isolate and quantify purified myelin from such a small tissue, we measured the specific activity of rCNP directly in optic nerve homogenates, using the enzyme as an index of myelin formation30. It is clear from Table II that adrenalectomy had no effect on CNP specific activity at any age. This could mean that myelination in the optic nerve was unaffected by the treatment or alternatively that the amount of myelin was increased but to the same extent as total tissue protein. The latter possibility could have been tested by comparing total CNP activity in the entire optic nerves, but this was precluded by incomplete removal of the tissue from some animals in the course of rapid dissection. DISCUSSION Our results demonstrate first that the increased brain weight produced by early adrenalectomy persists well into adulthood, with no sign of a late convergence toward the brain weight of control animals, This stimulation of brain growth occurs despite a concomitant reduction in body weight and therefore may represent a unique response of the brain to the loss of adrenal secretions. Apart from their increased size, the brains of A D X rats appear normal upon gross observation. We assume that the cranium of these animals is likewise enlarged to provide the needed space, although no actual measurements of cranial volume have yet been performed. The amount of purified myelin that could be recovered from the cerebrum was greatly increased in A D X vs SHAM subjects at 63 and 151-153 days of age. Although a small trend in the same direction was
TABLE II 2':3'-Cyclic-nucleotide 3'-phosphodiesterase (CNP) activity in optic nerve homogenates from adrenalectomized (ADX) and sham-operated (SHAM) rats
Data represent the mean ± S.E.M. for the number of animals shown in parentheses. Age
35-36 days 63 days 151-153 days
CNP (izol/mg protein~h) SHAM
ADX
1084 ± 32 (8) 819±47 (8) 855±46 (6)
1084 ± 14 (10) 847±52(10) 898 ± 32 (6)
apparent on day 35-36, the effect was not statistically significant. This was somewhat surprising because measurable brain weight differences are present at that age and because myelination is proceeding rapidly from the time of the surgery up to about 30 days of age26. We cannot be certain that the differences between the A D X and SHAM animals in recovered myelin weight represent the exact differences that existed in vivo because we did not have a measure of percentage myelin recovery. On the other hand, our isolation procedure did yield quantitatively reproducible amounts of myelin within each age and treatment group (note the small S.E.M. bars in Fig. 3 ) , and the yield in the older subjects was closely related to the size of each brain sample. Furthermore, our earlier work showed that whole cerebral CNP activity (i.e., without prior myelin isolation) in 65-day-old A D X rats was increased by 29% over that of SHAM animals 23. Therefore, it seems reasonable to conclude in the present study that at least by adulthood, substantially more myelin was present in the brains of the A D X rats than in the controls. As we previously demonstrated an elevated cerebral D N A content in A D X animals at 65 days of age 23, perhaps our treatment results in an increased number of oligodendrocytes, thereby leading to increased myelination. Myelination could be altered by changes in the percentage of axons myelinated, total number of axons (of which the same percentage is myelinated), and/or sheath thickness. Morphometric studies currently in progress should provide further information concerning the effects of adrenalectomy on axon-myelin relationships. Myelin isolated from the A D X animals was found to be measurably deficient in total galactolipid and phospholipid concentrations. It should be noted that such differences were not detected in a previous study 23, however, in that study fewer animals were examined and the animals were of a different strain (Wistar). Furthermore, at least in the case of the galactolipids, the assay method used here was more sensitive and precise. On the other hand, if an average phospholipid molecular weight of 775 is used to substitute for the molar values reported in Table I, it is possible to calculate total recovery of myelin constituents, including protein. The results indicate recoveries of only 92% and 88% for the SHAM and A D X groups respectively. Although it is the case that
several lipid classes (notably the gangliosides and polyphosphoinositides) were not measured in our protocol, these compounds represent only a small fraction of the dry weight of normal rat myelin 25. Because we found no myelin component that was increased in the A D X animals, the possibility must be considered that the difference in myelin composition between the groups was due to differential recovery of lipids (e.g. from the silicic acid columns) in the two sets of samples. This hypothesis will need to be addressed in subsequent studies. Myelination does not commence simultaneously in all CNS areas because of regional variation in the timetables of oligodendrocyte differentiation and axonal growth. Early myelinating tracts such as the optic nerve might be relatively less vulnerable than late myelinating tracts to the effects of adrenalectomy on day 11 postnatal. Indeed, we failed to find any difference in optic nerve CNP specific activity between ADX and SHAM animals. CNP is often used as a marker for myelination 30, although its activity is actually highest in so-called 'myelin-like' fractions22. 23 which are thought to consist primarily of uncompacted myelin and oligodendroglial plasma membranes undergoing transition to myelin. For this reason and also because of the point raised earlier concerning interpretation of specific activity values, the failure to observe CNP changes should not be taken as unequivocal evidence that optic nerve myelination is unaffected by removal of the adrenal glands. Although no hormone replacement condition was run in the present series of experiments, it is obviously important to determine which hormones are involved in the increased brain weight and myelination effects. Adrenalectomy results in a complex set of endocrine changes, including the loss of cortical steroids and medullary catecholamines, and an increase in circulating A C T H concentrations due to the absence of corticosteroid negative feedback. Studies cited in the Introduction have shown that exogenous glucocorticoid administration early in life can reduce subsequent myelination, hence it is conceivable that the removal of endogenous glucocorticoids might have the opposite effect. Furthermore, Devenport and Devenport 8 reported that corticosterone replacement could reverse the increased brain weight found after adrenalectomy on day 25 of life, although these investigators did not study myelination in their
experiments. These considerations suggest that the effects reported here are most likely to have resulted from the loss of glucocorticoids, but this conclusion must remain tentative until the appropriate replacement studies have been performed. Following submission of this manuscript for publication, a paper appeared by Preston and McMorris27 in which rats were adrenalectomized on the 14th day of life, weaned at day 18 and given drinking solutions of 5% glucose and 0.9% saline (used instead of the present mineralocorticoid replacement), and then sacrificed on the 21st or 22nd day of life for myelin determinations. In contrast to our present results, Preston and McMorris observed a 25% reduction in the amount of myelin isolated from the A D X animals which they attributed to adrenalectomy-related decreases in G P D H activity (see Introduction). We attempted to replicate these findings using the Preston and McMorris protocol but none of the rats survived more than 1-2 days following weaning. We then performed additional experiments with all animals (SHAM as well as ADX) given Percorten at the time of surgery and left with their mothers until sacrifice. Under these conditions, virtually all of the rats survived and we observed a small (non-significant) increase in myelination despite a significant reduction in G P D H activity. Given the fragile nature of early A D X rats, we would argue that the hypomyelination reported by Preston and McMorris was a secondary consequence of the premature weaning and the attendant anorexia produced by such treatment. These issues are discussed in greater detail elsewhere (J. S. Meyer and K. R. Fairman, in preparation). In conclusion, we have begun to describe some of the cellular and biochemical changes underlying the increase in brain weight seen following early adrenalectomy. Although the present paper focuses on adrenalectomy-induced changes in myelination, it should be noted that if myelin comprises an estimated 20% of the dry weight of a 60-day-old rat brain26, then even a 40% increase in myelin deposition could not nearly account for a 15% increase in overall brain weight (as shown in Meyer 23, the remaining difference is not due simply to elevated brain water content). Early adrenalectomy undoubtedly influences a variety of neural elements and thus provides an important new model for studying plasticity of the developing brain.
ACKNOWLEDGEMENTS
typing the final m a n u s c r i p t . This r e s e a r c h was supp o r t e d by G r a n t BNS-8118073 f r o m the N a t i o n a l Sci-
W e w o u l d like to t h a n k Ms. M e l a n i e B e l l e n o i t for
REFERENCES 1 Allain, C. C., Poon, L. S., Chart, C. S. G., Richmond, W. and Fu, P. C., Enzymatic determination of total serum cholesterol, Clin. Chem., 20 (1974) 470-475. 2 Balazs, R., Brooksbank, B. W. L., Patel, A. J., Johnson, A. L. and Wilson, D. A., Incorporation of [35S]sulfate into brain constituents during development and the effects of thyroid hormone on myelination, Brain Res., 30 (1971) 273-293. 3 Balazs, R., Patel, A. J. and Hajos, F., Factors affecting the biochemical maturation of the brain: effects of hormones during early life, Psychoneuroendocrinology, 1 (1975) 25-36. 4 Bartlett, G. R., Phosphorus assay in column chromatography, J. biol. Chem., 234 (1959) 466-468. 5 Bohn, M. C. and Friedrich, V. L. Jr., Recovery of myelination in rat optic nerve after developmental retardation by cortisol, J. Neurosci., 2 (1982) 1292-1298. 6 Casper, R., Vernadakis, A. and Timiras, P. S., Influence of estradiol and cortisol on lipids and cerebrosides in the developing brain and spinal cord of the rat, Brain Res., 5 (1967) 524-525. 7 De Vellis, J. and Inglish, D., Hormonal control of glycerolphosphate dehydrogenase in the rat brain, J. Neurochem., 15 (1968) 1061-1070. 8 Devenport, L. D. and Devenport, J. A., The effects of adrenal hormones on brain and body size, Physiol. Psychol., 10 (1982) 399-404. 9 Field, E. J., Effect of cortisone on the neonatal rat, Nature (Lond.), 174 (1954) 182. 10 Folch, J., Lees, M. and Sloane Stanley, G. H., A simple method for the isolation and purification of total lipides from animal tissues, J. biol. Chem., 226 (1957) 497-509. 11 Gumbinas, M., Oda, M. and Huttenlocher, P., The effects of corticosteroids on myelination of the developing rat brain, Biol. Neonate, 22 (1973) 355-366. 12 Horrocks, L. A., Thin-layer chromatography of brain phospholipids, J. Amer. Oil Chem. Soc., 40 (1963) 235-236. 13 Howard, E. and Benjamins, J. A., DNA, ganglioside and sulfatide in brains of rats given corticosterone in infancy, with an estimate of cell loss during development, Brain Res., 92 (1975) 73-87. 14 Jacobson, S., Sequence of myelination in the brain of the albino rat. A. Cerebral cortex, thalamus and related structures, J. comp. Neurol., 121 (1963) 5-29. 15 Jungalwala, F. B. and Dawson, R. M. C., The turnover of myelin phospholipids in the adult and developing rat brain, Biochem. J., 123 (1971) 683-693. 16 Kean, E. L., Rapid, sensitive spectrophotometric method for quantitative determination of sulfatides, J. Lipid Res., 9 (1968) 319-327. 17 Kirk, R. E., Experimental Design: Procedures for the Behavioral Sciences, Brooks/Cole, Belmont, CA, 1968, 577 PP. 18 Kornberg, A. and Pricer, W. E., Enzymatic esterification of a-glycerophosphate by long chain fatty acids, J. biol. Chem., 204 (1953) 345-347.
ence Foundation.
19 Lees, M. B. and Paxman, S., Modification of the Lowry procedure for the analysis of proteolipid protein, Anal. Biochem., 47 (1972) 184-192. 20 Leveille, P. J., McGinnis, J. F., Maxwell, D. S. and de Vellis, J., lmmunocytochemical localization of glycerol-3phosphate dehydrogenase in rat oligodendrocytes, Brain Res., 196 (1980) 287-305. 21 Lowry, O. J., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 22 Matthieu, J. M., Quarles, R. H., Brady, R. O. and Webster, H. deF., Variation of proteins, enzyme markers and gangliosides in myelin subfractions, Biochim. biophys. Acta, 329 (1973) 305-317. 23 Meyer, J. S., Early adrenalectomy stimulates subsequent growth and development of the rat brain, Exp. Neurol., 82 (1983) 432-446. 24 Noguchi, T., Sugisaki, T., Watanabe, M., Kohsaka, S. and Tsukada, Y., Effects of bovine growth hormone on the retarded cerebral development induced by neonatal hydrocortisone intoxication, J. Neurochem., 38 (1982) 246-256. 25 Norton, W. T., Biochemistry of myelin. In S. G. Waxman and J. M. Ritchie (Eds.), Demyelinating Disease. Basic and Clinical Electrophysiology, Raven Press, New York, 1981, pp. 93-121. 26 Norton, W. T. and Poduslo, S. E., Myelination in rat brain: changes in myelin composition during brain maturation, J. Neurochem., 21 (1973) 759-773. 27 Preston, S. L. and McMorris, F. A., Adrenalectomy of rats results in hypomyelination of the central nervous system, J. Neurochem., 42 (1984) 262-267. 28 Rouser, G., Kritchevsky, G. and Yamamoto, A., Column chromatography of phosphatides and glycolipids. In G. V. Marinetti (Ed.), Lipid Chromatographic Analysis, 1Iol. 1, Marcel Dekker, New York, 1967, pp. 99-162. 29 Schapiro, S., Some physiological, biochemical, and behavioral consequences of neonatal hormone administration: cortisol and thyroxine, Gen. Comp. End., 10 (1968) 214-228. 30 Sims, N. R. and Carnegie, P. R., 2':3'-Cyclic-nucleotide 3'phosphodiesterase. In B. W. Agranoff and M. H. Aprison (Eds.), Advances in Neurochemistry, Vol. 3, Plenum, New York, 1978, pp. 1-41. 31 Sogin, D. C., 2':3'-Cyclic-NADP as a substrate for 2':3'cyclic-nucleotide 3'-phosphohydrolase, J. Neurochem., 27 (1976) 1333-1337. 32 Tennekoon, G. I., Cohen, S. R., Price, D. L. and McKhann, G. M., Myelinogenesis in optic nerve. A morphological, autoradiographic, and biochemical analysis, J. Cell Biol., 72 (1977) 604-616. 33 Waehneldt, T. V., Ontogenetic study of a myelin-derived fraction with 2':3'-cyclic-nucleotide 3'-phosphohydrolase activity higher than that of myelin, Biochem. J., 151 (1975) 435-437. 34 Weissbarth, S., Maker, H. S., Raes, I., Brannan, T. S., Lapin, E. P. and Lehrer, G. M., The activity of 2':3'-cyclicnucleotide 3'-phosphodiesterase in rat tissues, J. Neurochem., 37 (1981) 677-680.