- Email: [email protected]
Effects of undernutrition on glial maturation
Brain Research, 149 (1978) 379-397 © Elsevier/North-Holland Biomedical Press
379
EFFECTS OF U N D E R N U T R I T I O N ON G L I A L M A T U R A T I O N *
O.
ROBAIN and G. PONSOT
I N S E R M U 154 Hopital Saint Vincent de Paul, 75674 Paris, Cedex 14 (France)
(Accepted October 27th, 1977)
SUMMARY Rats of Wistar strain were conceived and breast-fed until the 25th day by mothers maintained on a low protein (5 ~ ) and low caloric (21 calories/day) diet, producing a severe deficiency in weight body growth (more than 50 ~ at the 10th day) and of the weight of the central nervous system (40 ~ on the 15th day) both in the cerebral hemispheres and the spinal cord. Histological and biochemical analysis of the central nervous system shows: (1) Glial proliferation is insufficient and delayed, the number of glial cells is reduced by 50 ~ on the 10th day in the cuneatus and gracilis tracts and the density of the glial cells is reduced by 5 0 ~ in the corpus callosum at the 19th day. (2) Maturation of the glial cells is greatly retarded, especially in the corpus callosum a structure which matures late. On the 19th day, the majority of the cells in this structure still have a glioblastic appearence, whereas in the normal rat, at this age the majority of the glial cells are oligodendrocytes. (3) These abnormalities of glial maturation agree well with the delay of the increase of DNA, RNA and protein measured in the spinal cord and cerebral hemispheres. (4) There is a defect in myelination assessed by estimation of the density of the myelin fibres, and a definitive reduction in the caliber of the spinal tracts.
INTRODUCTION The effect of undernutrition on the developing brain in the rat has been extensively investigated. It is well documented that undernutrition is particularly harmful during the period of rapid brain development 9. The effect on myelination, which occurs in the rat entirely after birth, and mainly during the first month of life, * This study was supported by ATP N ° 27 INSERM.
380 has been demonstrated biochemically2,S, 2s. Histologically, this abnormal myelination is shown by a reduction in the number and thickness of the myelin sheaths, both in the central and peripheral nervous system4,t1,le,2~, 2~. The effect of undernutrition on gliat maturation has been investigated less. The only definite finding so far observed is a reduction in the total number of cells of the central nervous system, assessed by D N A estimations. Histologically, Bass et al."- described a defect in migration and differentiation of glial cells; Cragg 6, Krigman and Hogan t~ mentioned a scarcity of glia in th e gray and white matter. A reduction in the area occupied by gtia in the molecular layer of the cerebellum has been describedL Effects of undernutrition on the mouse anterior commissure have been examined quantitatively by Sturrock et al. 2~ but only at 19 weeks afterbirth and by Siassi and Siassi 2° but only in the somatosensory cortex. The purpose of our work was to provide more information on the effects of undernutrition on glial maturation, and on the myelination of some tracts maturing at different periods. MATERIALS AND METHODS Adult female rats of Wistar strain, weighing at least 250 g, were ted with a restricted calorie and protein diet having the following composition: Casein 5 °/E, tournesol oil 1%, carbohydrates 88 % (half starch, half sugar), cellulose 2 o~, mineral salts 5 % (calcium carbonate 24 %, bicalcium phosphate 20 %, potassium phosphate 15%, calcium lactate 10%, magnesium sulphate 10~o, iron lactate 10~i, sodium chloride 6 %, sodium phosphate 4 %, copper sulphate 0.6 %, cobalt nitrate 0. 1%, zinc acetate 0.2 %), methionine 0.3 %, vitamin B complex 0.5 %, vitamin D 3000 I.U./Kg, vitamin E 30 I.U./Kg. The rats used for this experiment received 6 g per day of this diel (i.e. 21 calories) for 8 days. They were then mated 5 successive nights and the time of fertilisation determined by daily vaginal smears. The fertilised rats were maintained on the same diet throughout pregnancy and lactation, up to the 25th day. The control animals had free access to a diet of pellets containing 22 % protein. 5 o/lipids, 52 o/~ carbohydrate. They consumed an average of 15 g daily, i.e. 51 calories. Litters of the control and undernourished rats were adjusted to 5-7 animals. The pups were examined on the 5th, 10th, 15th, 19th and 25th days after birth.
Biochemical analysis After ether anesthesia, the pups were decapitated and the brains were rapidly removed from the skull. The cerebral hemispheres were separated from the cerebellum and brain stem, by a section just above the superior colliculus, weighed and frozen at 30 °C. Measurements were carried out on 8 pups from at least three different litters. D N A and R N A were measured by Wannemacker's method z5 as modified by Shibko is, the total protein by Lowry's method 15. These estimations were carried out on 8 undernourished rats and on 4 controls from at least 3 different litters.
381
Histological stud), The pups anaesthesized with ether were fixed by intra-aortic perfusion in vivo, according to a previously described method iv. The whole head and spinal cord were kept overnight in Karnovsky solution modified (1 ~ glutaraldehyde, 1 ~ formaldehyde, in 0.12 M phosphate buffer with 0.02 m M CaClz). Corpus callosum was removed from coronal slices of brain at the level of the optic chiasm. Transverse sections (1 ram) of spinal cord were obtained at the dorsal level, half way between lumbar and cervical enlargements. Blocks were postfixed in 2 ~ osmium tetroxide in 0.12 M phosphate buffer at pH 7.4 for 4 h, stained with uranyl acetate and embedded in araldite. Thick sections were stained with 1 ~ toluidine blue, and thin sections were stained with uranyl acetate saturated in alcohol 50° and after by Reynolds' method.
Quantitative stud), The total number of glial nuclei present in each fasciculus was counted in each slice examined, the counts being made directly on the photomicrographs at a magnification of 188 ×. The surface area of the fasciculi were measured with a polar planimeter. The surface of the corpus callosum could not be traced with precision therefore glial cells were counted not in the whole corpus callosum but in 10 squares, measuring 36 × 36 #m, with an ocular net micrometer, using paraffin sections 7.5 #m thick. The density of myelinated fibers was measured on photographs of thin horizontal sections enlarged to 7500 × in the 3 fasciculi of the posterior column of the spinal cord and on sagittal sections of the corpus callosum. Electronmicrographs were taken of the anterior region of the corticospinal fasciculus in order to have the minimum number of aberrant fibers. They were also taken of the posterior region of the gracilis and cuneatus fasciculi, since the delineation of these fasciculi is more clear at this level than further from the surface. In the corpus callosum, the myelinated fibers were counted on sagittal sections, on photographs also magnified to 7500 x . All measurements were carried out on 5 normal rats and 5 undernourished rats at 5, 10, 15, 19 and 25 days, obtained from at least 3 different litters. RESULTS Delivery occurred only in 50 ~ of fertilised rat dams in the starved group. The mortality of the offsprings was 50 ~. Deaths occurred mostly during the first 2 or 3 days. Lactation in undernourished rats was not regularly induced. The undernourished pups were smaller than the controls: by day 10 they weighed 5 g and 12.5 g respectivily (Fig. 1A). Opening of the eyes was delayed by three days and occurred around the 18th day in the undernourished group. HISTOLOGICAL RESULTS
(1) Glial cells (A) Quantitation study Normal rats. Table I indicates the number of glial cells in the fasciculi cuneatus,
382 Body weight
mg
A
Hemispheres weight , ~ _ . _ . . ~ B
i
900
30
700
25
500
20
300
15
106 0
/ i,iJ F
1
I
I
l
I
-t I1
5
10
15
19
25
days
10
mg 200 -Spinal cord weight
0
l
I
i
l
5
10
15
19
C
150 25
days
Well Nourished .....
Poorly Nourished
0
5
10
15
19
25
days Fig. 1. The difference in the weight of the body (A), cerebral hemispheres (B), and the spinal cord (C) of well nourished and poorly nourished animals are shown. The values are the means, and the bars indicate standard deviation. Eight rats from at least three different litters are used,
gracilis and corticospinal. As already noted by Matthews and Duncan 16, there is a definitive chronological difference in the proliferation of cells in these different tracts. In the cuncatus, glial cells were already numerous by the 5th day and proliferation was almost completed by the 10th day. In the fasciculus gracilis, glial multiplication occurred a little later and continued until the 15th day. In the corticospinal tract, cell multiplication continued beyond the 15th day and at least till the 25th. The reduction in the number of cells observed between the 19th and 25th days in the cuneate and graeilis tracts need not be interpreted as a true rarefaction but may derive from a dispersion depending on the increase in length of the spinal cord la. In the undernourished rats, there was a reduction of the total number of glial cells from the 10th day onwards. It is conspicuous in the posterior column of the spinal cord. The difference is significant in each of the three fasciculi studied. It reaches 50 % by the 10th day in the gracilis and cuneatus tracts and by the t5th dayin the corticospinal tract.
383 TABLE I
Number of glia cells in posterior column of the spinal cord of control and undernourished rats Values are m e a n s ± S t a n d a r d Deviation. T h e results are f r o m 5 rats with no m o r e t h a n 2 from a litter. Difference between control a n d u n d e r n o u r i s h e d are evaluated with M a n n - W h i t n e y test (one-tailed test)
Days
5 10 15 19 25
Fasciculus cuneatus
Fasciculus gracilis
Fasciculus corticospinal
Normal
Undernourished Normal
Undernourished Normal
Undernourished
59.2±8.64 100.0±5.38 121.8±7.08 111.8±7.19 105.2±8.25
53.6±6.84NS 53.2±6.69"* 85.8±7.53** 87.2±6.53** 76.6±7.64**
22.4±2.61NS 28.8±3.35"* 60.4±5.32"* 68.4±4.16'* 43.8±3.38**
10.4±2.6"* 16.8±2.23"* 23.0±1.58"* 27.2t-l.92"* 33.4:~3.85"*
21.6±4.16 58.6±6.27 100.4~6.39 103.6±7.50 92.2±-8.58
17 ± 1 . 5 8 23.8±4.49 42 ± 4 . 4 7 41.5±5.47 55.6±3.71
NS, P > 0.05. * 0.05 > P > 0.01. ** 0.01 ~ - P .
Table II indicates the average number of glial cells counted in the corpus callosum: In the normal rats, cell density increases rapidly between the 10th and 19th days, and this is in agreement with Schonbach18; in the undernourished rats, the cell multiplication is diminished, the difference with control rats is significant from the 15th day onward and very evident on the 19th day (Fig. 2).
( B) Maturation of glial cells Normal rats: corpus callosum. This is specially well seen in longitudinal section (Fig. 3). In the normal rats free subependymal cells 14 or glioblasts 2~ constitute the large majority of cells up to the 10th day (Fig. 4). The length of the cells is sometimes 4 times greater than the width. They have a nucleus which contains one or two large T A B L E II
Number of glia cells in corpus callosum In l0 square m e a s u r i n g 36 × 36/~m. Values are m e a n s ± Standard Deviation. T h e results are f r o m 5 rats with n o m o r e t h a n 2 f r o m a litter. Difference between control a n d u n d e r n o u r i s h e d are evaluated with M a n n - W h i t n e y test (one-tailed test).
Days'
Normal
5 10 15 19 25
24.4 28.4 39.6 49.33 50.33
± ± ± ± ±
NS, P > 0.05. * 0.05 > P > 0.01. ** 0.01 ~> P.
Undernourished 3.05 4.16 2.41 6.35 3.88
20 22 23 23.83 34
± ± ± ± ±
2.34* 2.92* 6.48** 3.43** 4.73**
Fig. 2. R a t c o r p u s c a l l o s u m 19th day after birth, t00. A : control rat : note the a r r a n g e m e n t of glial cells in rows. B: u n d e r n o u r i s h e d rat: note the decrease a n d the dispersion o f the glial ceH~.
4 P
Fig. 3. Rat corpus callosum 19 days after birth. ~: 650. A : control rat : most of the glial cells present in the picture are dark (arrow) or light oligodendrocytes (arrowhead). B: undernourished rat: note the scarcity of the cells, which are dark and light glioblast. Note on the right top two light ligodendrocytes (arrowhead).
386 nucleoli. Their cytoplasm is reduced and is concentrated at each pole of the nucleus: it contains a great number of ribosomes grouped in rosettes. Two kinds of glioblasts can be seen (Fig. 4A and B). They differ mainly by their nuclei: some cells have dark nucleus, with aggregated chromatin and have an ovoid and sometimes irregular shape (Fig. 4A); the other gliobtasts have a clear nucleus with dispersed chromatin and, often, have a very regular shape (Fig. 4B). The nucleus of both cell types may exceed 15 microns in length. The proportion of these cells decreases rapidly after the 10th day. At the 19th day they constitute 10-20 ~o of the glial cell population (Fig. 3A) and at the 25th day only about 10 °/ /0 ' Oligodendrocytes cannot be recognized at the 5th day; young oligodendroglia cells appear on the 10th day: they have a large nucleus, abundant cytoplasm and ~1 narrow electron lucent endoplasmic reticulum. The intermediate and dark forms of adult type can be recognized on the 15th day. At the 19th day, the oligodendrocytes have mainly an immature appearence and constitute 75'i~o of the cells (Fig. 3A). They form rows of 6-8 cells at the 25th day. Astrocytes may be identified easily at the 5th day : most of them are youn~ cells and their transition from the glioblasts with light nuclei can be presumed, because these cells have already a very elongated nucleus. The chromatin, however, tends to aggregate near the nuclear membrane. The cytoplasm of these cells, contains large cisternae with an electron dense content ~:~. in these young cells there are neither glycogen granules nor gliofilaments. Microglia were identified on the 5th day. They have a nucleus with chromatm aggregated beneath the nuclear membrane and in the cytoplasm rough endoplasmic reticulum is proeminent with very elongated cisternae. Some dense bodies ale visible, Spinal cord. Maturation occurred earlier than in the corpus callosum: by the 15th day (oligodendrocytes appeared by the 5th day) oligodendrocytes constitute more than half of cell population, dark adult oligodendrocytes appear by the 5th day in the cuneatus, by the 10th day in gracilis and by the 19th day in the corticospinal tract. Undernourished rats: corpus callosum. Free subependymal cells persist for a much longer time in great number and at 19th day they constitute half the cell population of the corpus callosum (Fig. 3B). The delay in appearence of oligodendrocytes is evident: these cells cannot be identified before the 15th day. at the t9th day they represent less than 20~);i of the cell population (Fig. 3B). Rows of oligodendrocytes are short and scarce and at the 25th day only doublets or triplets can be seen. We did not see any difference in the maturation and proliferation of astrocytes except for a higher density of gtiofilaments from the 15th day. Microglia cells were identified at the 5th day as in the normal rats. Some microglia cells have numerous intracytoplasmic inclusions resembling lipids, Sometimes. intracytoplasmic inclusions are dense, large with irregular shape. The cells containing these inclusions have sometimes disposed fine cisternae. It is difficult to be sure that all these cells are microglia or dark oligodendrocytes containing inclusions (Fig. 5B).
Fig. 4. Undernourished rat corpus callosum 10 days after birth. A : glioblast with dark nucleus, x 7000. The cytoplasm contains mitochondria, free ribosomes and numerous scattered polysomes; Note the absence of endoplasmic reticulum (inset). B: glioblast with light and very elongated nucleus. > 8000. As in the top cell, the cytoplasm contains only mitochondria, free ribosomes and polysomes (inset).
388
389
(2) Myelin sheath~s Table 1II indicates progress in myelination in the posterior column of the spinal cord and in the corpus callosum from 5 to 25 days. In the normal rat, chronological difference in myelination between the cuneate, gracilis, corticospinal tract, and corpus callosum is quite clear, the multiplication of the myelin sheats is conspicuous between the 5th and the 15th day in the fasciculus cuneatus and gracilis, between the 10th and the 25th day in the fasciculus corticospinal and the corpus callosum in the undernourished rat, the myelination is very delayed. The difference with the normal rat can be seen on the 5th day in the cuneatus, from the 5th to the 15th in the gracilis, from the 15th to the 25th day in the corticospinal tract and in the corpus callosum. The density of the myelin sheaths became identical from the 10th onwards in the cuneatus and from the 19th day in the gracilis tract. This phenomenon does not mean that the absolute number of myelin sheaths is identical, because the striking increase of the caliber of the bundles has to be taken in account. Table IV showed a definite defect in the growth of the bundles of the posterior column in the undernutrition. This defect indicates a deficiency in the total number of myelin sheaths. BIOCHEMICAL RESULTS
(1) DNA Cerebralhemispheres (Fig. 7A). From the 5th day onwards, the amount of DNA in the cerebral compared to the 35 ~. In control days, whereas in and 25th days.
hemispheres of undernourished animals is reduced to 25~,,, as controls. D N A reduction is maximal at the 15th day when it reaches animals the increase in D N A is maximal between the 10th and 15th undernourished rats, the maximum increase occurs between the 15th
Spinal cord (Fig. 7B). Cell multiplication, as expressed by DNA, is more affected in the spinal cord than in the cerebral hemispheres. The evolution in the level of DNA during maturation is the same as in the cerebral hemispheres. (2) RNA Cerebral hemispheres (Fig. 7C). In control animals the maximal increase in R N A level occurs between the 10th and 15th day. After the 19th day, there is a
Fig. 5. A : undernourished rat 5 days after birth, corpus callosum, x 7000.Young astrocyte with elongated nucleus containing two nucleoli and peripheral clumps of chromatin. The cytoplasm exhibits large cisternae of rough endoplasmicreticulum, which are filled with gray material, some mitochondria and polysomes. Note the absence of gliofilaments and glycogengranules. (Inset) Arrow indicates one lipid body. B : undernourished rat 10 days after birth, fasciculus gracilis. × 7000. Glial cell with nucleus containing clumps of chromatin beneath the nuclear membrane, and a striking nucleolus. This dark cytoplasm contains numerous dark bodies and lipid droplets, short fine strands of rough endoplasmic reticulum and important Golgi apparatus. This cell is probably dark mature oligodendrocyte.Note in the corner below the dark cytoplasm of an oligodendrocyte.
390
0
e-
.
.
.
.
[
i
**zz ..~
e-
qb
•
.
ZZZ~ O
¢
~ESE
_= Av/v/ ~ck~ <
~**
391
e-
8 e.
o t..,,I
Z~*** o o ¢-
o
o
¢..
F
.I
<
,,4 c~qo o~ AV/?/
!
r~
~i ~¸¸¸,ii!
393 reduction in RNA as already reported by Fish and Winick 1°. In undernourished rats the maximal increase occurs between the 15th and the 19th days and continues until the 25th day. Therefore, the curves for controls and undernourished rats cross one another. Spinal cord (Fig. 7D). The RNA content of spinal cord of undernourished rats is more diminished than that of the cerebral hemispheres. The difference reaches 50 ~ at the 15th day. In normal rats there is a maximal increase in RNA between the 10th and 15th days and a reduction after the 19th day, whereas in undernourished pups, RNA continues to increase after this date.
(3) Changes in protein content (Fig. 7E and F) Cerebral hemispheres protein synthesis is highly disturbed by undernutrition: there is a reduction of almost 5 0 ~ in the quantity of protein at the 15th day. The difference is still 25 ~ at the 25th day. In the spinal cord (Fig. 7F), the reduction is even more marked. By the 15th day, protein levels are reduced by almost 6 0 ~ in undernourished animals and by the 25th day the difference is still 4 0 ~ .
(4) Protein/DNA ratio In undernourished pups the more severe involvement in protein synthesis compared with that of DNA produces a reduction in the protein/DNA ratio, both in the cerebral hemispheres and in the spinal cord. This reduction is significant from the 15th day in the cerebral hemispheres (controls = 54.14 ~ 2.90, undernourished = 41.38 ± 4.05, P < 0.001) and from the 19th day in the spinal cord (controls = 40.05 ~0.93, undernourished -- 33.05 ± 1.98, P < 0.02). DISCUSSION
(1) Our quantitative study of glia shows a reduction in the absolute number of these cells in the posterior column of the spinal cord and of their density in the corpus callosum. As shown by Matthews and Duncan ~5, glial proliferation closely follows the volumetric increase of the tracts undergoing myelination. Counting of the total number of glial cells present on the whole section of the posterior column eliminates the variation due to different rates of growth of the caliber of the tracts. However, the variation due to the different rates of increase in length of the spinal cord is not suppressed. Thus, the true difference in the number of glial cells between normal and undernourished rats is more marked than that which we measured. In the corpus callosum, cell density gives a very imperfect indication of the cell population. Schonbach et al. 17, showed that, in the normal rat, cell density ceases to increase after the 20th day, whereas cell proliferation continues. Our findings in control were identical. This may be explained by the dispersion of the cells: the myelination fibres, increase in diameter and separate the glial cells from one another. The defect of myelination seen in the undernourished rats implies that the increase in volume of the corpus callosum is lower than that for the controls. The true
- - .
Well Nour,shed
.....
Poorly Nourished
DNA
Fg
Hemispheres
A
1000
;_I_--i .... {- '{
500
0
I
0
I
I'5
10
5
~
--
19
25
o
s
10
days
pg
R N A Hemispheres
Hg
~5
spinal card
RNA
oot_
D
//
.-
/
,/
300 b
l/z/
1000
)I
/
200F
/
-"
~"
looI_ {"
500
0
____J-
I
0
5
_
~
10
t
~
.
15
.
.
.
.
.
0 . . . .
.
25
19
0
5
10
days
t9
15
Proteins spinal cord
Fj,~I
16
Proteins Hemispheres
E
60
~2
/
/
40 ¸
/~/
2O
ok_ 0
i/ •
25
oovs
mg mg
2'~
C
2000
1500
1~
days
/
~
..{.-"
/
~1/
i
8
/q
.,"""
_j/"
i
I
J
5
10
15 days
19
25
0
o
~
~o
K is
~
i 2s
days
Fig. 7. The graphs indicate the total DNA-RNA and proteins in the cerebral hemispheres and spinal cord from birth to the 25th day. The values are means (8 rats from at least 3 different litters are used for the undernourished group, and 4 rats for the control group) and the bars indicate standard deviation.
395 reduction in glia is thus, here also more important than that which we measured. Whatever the lack of precision of our histological counts they clearly show the essential role of deficiency of the glia in reduction in cell proliferation as shown by DNA estimation. With our model of ante and postnatal undernutrition, this deficiency is greater than that reported by Swaimann et al. 24, Culley and Linerberger 7, in postnatal undernutrition, and is comparable with that observed by Winnick 2s, with ante and postnatal undernutrition and that of Krigman and Hogan 12, in undernutrition restricted to the postnatal period. It is generally accepted that the disturbance in cell proliferation in undernutrition concerns the glial population; the number of neurones is approximately stable after birth a. This view has received very little accurate histological confirmation. Krigman and Hogan 12 simply found a reduction of glial cells in the white and gray matter; Bass et al. 2, also mentioned, a rarefaction of glia in the subcortical white matter but these authors have not done a quantitative evaluation. The more important work is this of Sturrock et al. 25: these authors described reduction of the oligodendrocytes in the mouse anterior commissure at 19th weeks after birth. Our data quantitatively demonstrate important glial deficiency in malnutrition during the first 25 days. Can the low level of DNA in undernourished rats be explained only by the scarcity of glial cells? Histological quantification is not sufficient to state this categorically. Glia/neurone ratio increases 200~ from 0 to 22 days in the spinal cord of normal rat la. This shows the importance which can be attributed to the impairment of glial proliferation at these ages. (2) A deficiency in maturation of glial cells is demonstrated by our study especially in the corpus callosum, where immature cells still persist in large numbers on the 25th day after birth. We did not count separately the different cell types. The cells are very different in size and shape and therefore Abercrombie's correction, taking into consideration the diameter of the nucleus, is difficult to apply to these cells. The length of their nuclei may be 4 times greater than the width and it is sometimes difficult to say along which axis they have been sectioned. Our findings concerning the delay in the maturation of glia is different from that reported by Krigman and Hogan 12, (rarefaction without morphological changes) and confirms the findings briefly mentioned by Bass et al. 2 in samples from the subcortex. This delay in cell maturation is shown also by our biochemical findings: in undernourished rats a fall in protein/DNA ratio is noted from the 15th day onwards in the cerebral hemispheres and from the 10th day onwards in the spinal cord. This ratio gives an approximate indication of the nucleo-cytoplasmic ratio. Probably the delay in the peak of the RNA curve in undernutrition (Fig. 7C and D) is related to the delay in maturation. This defect in maturation of glia may explain the delay in myelination and the persistence at 30 days of a significant percentage of promyelin fibres noted by Krigman and Hogan 12, in the pyramidal tract. (3) The effect on myelination is the best known histological feature in undernutrition. The morphological evaluation of myelin deficiency in growing structures is difficult. Study of the rate of increase in myelinated fibres and the thickness of myelin
396 s h e a t h s in u n d e r n u t r i t i o n d i d n o t s h o w t h e t r u e myelin deficiency. T h e deficiency tn g r o w t h o f t h e calibre o f the t r a c t s s h o w n in o u r s t u d y gives a g o o d i n d i c a t i o n o f t h e deficiency o f m y e l i n a t i n g fibres. It is in t h e r e g i o n s w h i c h m a t u r e late, such as t h e c o r p u s c a l l o s u m , t h a t gtial a b n o r m a l i t i e s are t h e m o s t p r o m i n e n t a n d it is in this s t r u c t u r e t h a t the deficiency in m y e l i n a t i o n is also t h e m o s t m a r k e d (Fig. 6). L i k e w i s e , t h e c o r t i c o s p i n a l t r a c t , w h e r e m a t u r a t i o n o c c u r s later, is m o s t g r e a t l y a f f e c t e d t h a n gracilis a n d c u n e a t u s .
REFERENCES 1 Albercrombie, M., Estimation of nuclear population from mlcrotome section. Attar. Rec.. 94 (1946) 239-247. 2 Bass, N. H., Netsky, M. G. and Young, E.. Effect of neonatal malnutrition on developing cerebrum. I. Microchemical and histologic study of cellular differentiation in the rat. 11. Microchemical and histologic study of myelin formation in the rat. Arch. Neurol. (Chic.~. 23 (1970) 289-302 and 303-313. 3 Brizzee, K. R., Vigt, J. and Kharetchko, X., Postnatal changes in glial neuron index with a comparison of methods of cell enumeration in the white rat, Prog. Brain Res., 4 (1964) 136--149. 4 Clos, J. and Legrand, J., Influence de la d6ficience thyroidienne et de la sous-atimentation sur la croissance et la myetinisation des fibres nerveuses du nerf sciatique chez le jeune rat blanc: 6tude au microscope electronique, Brain Research., 22 (1970) 285-297. 5 Clos, J., Rebiere, A. and Legrand, J., Differential effects of hypothyroidism and undernutrition on the development of glia in the rat cerebellum, Brain Research, 63 (1973) 445-449. 6 Cragg, B. G.. The development of cortical synapses during starvation in the rat. Brain. 95 (1972l 143-150.
7 Culley, W. J. and Lineberger, R. O., Effect of undernutrition on the size and composition of the rat brain, J. Nutr., 96 (1968) 375-386. 8 Davison. A. N. and Dobbing, J., Myelination as a vulnerable period in brain development, Brain reed. Bull., 22 (1966) 404-444. 9 Dobbing, J., Vulnerable periods in developing brain. In A. N. Davison and J. Dobbing (Eds.), Applied Neurochemistry, Davis, Philadelphia, 1968, pp. 9-29. 10 Fish, I. and Winick, M., Cellular growth in various regions of the developing rat brain, Ped. Res., 3 (1969) 407412. 1l Hedley-Whyte, E. T., Myelination of rat sciatic nerve: comparison of undernutrition and cholesterol biosynthesis inhibition, J. Neuropath. exp. Nearol., 32 (1973) 284-302. 12 Krigman, M. R. and Hogan, E. L., Undernutrition in the developing rat: effect upon myelination. Brain Research. 107 (1976) 239-255. 13 Ling, E. A.. Study in the changes of the proportions and numbers of the various glial celt types in the spinal cord of neonatal and young adult rats, Acta anat. (Basel), 96 (1976) 188-195. 14 Ling, E. A., Paterson, J. A., Privat, A., Mori, S. and Leblond, C. P.. Investigation of glint cells in Semithin sections. I. Identification of glial cells in the brain of young rats, 3". comp. Neurol,, 149 (1973) 43-72. 15 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 16 Matthews, M. A. and Duncan. D.. A quantitative study of morphological changes accompartymg the initiation and progress of myelin production in the dorsal funiculus of the rat spinal cord. J. comp. Neurol. 142 (1971) 1-22. 17 Privat, A., Robain, O. et Mandel, P., Aspects ultrastructuraux du corps calleux chez la souns Jimpy, Acta neuropath. (BerL), 21 (1972) 282-295. 18 Schonbach, J., Hu, K. H. and Friede, R. L., Cellular and chemical changes during myelination" histologic, autoradiographic, histochemical, and biochemical data on myelination in the pyramidal tract and corpus callosum of rat, J. comp. NeuroL, 413 (1968) 21-38. 19 Shibko, S., Koivistoinen, P., Tratnyck, C. A., Newhall, A. R. and Friedman, L., A method for sequential quantitative separation and determination of protein, RNA, lipid and glycogen from a single rat liver homogenate or from a subcellular fraction, Analyt. Bioehern., t9 (1967~ 514-528.
397 20 Siassi, F. and Siassi, B., Differential effects of protein-calorie restriction and subsequent rejection in neuronal and non-neuronal components of cerebral cortex in newborn rat, J. Nutr., 103 (1973) 1625-1633. 21 Sima, A., Studies on fibre size in developing sciatic nerve and spinal roots in normal, undernourished and rehabilitated, Acta physiol, scand., Suppl. 406 (1974) 5-55. 22 Sima, A. and Sourander, P., The effect of early undernutrition on the calibre spectrum of the rat optic nerve, Acta neuropath. (Berl.), 28 (1974) 1-10. 23 Skoff, R. P., Price, D. L. and Stocks, A., Electron microscopic autoradiographic studies of gliogenesis in rat optic nerve. I. Cell proliferation, lI. Time of origine, J. comp. Neurol., 169 (1976) 291-311 and 313 - 333. 24 Sturrock, R. R., Light microscopic identification of immature glial cells in semithin sections of the developing mouse corpus callosum. J. Anat. (Lond.), 122 (1976) 521 537. 25 Sturrock, R. R., Smart, J. C. and Dobbing J., Effect on denutrition during the suckling period on growth of the anterior and posterior limbs of the mouse anterior commissure, Neuropath. appl. Neurobiol., 2 (1976) 239-247. 26 Swaiman, K. F., Daleiden, J. M. and Wolfe, R. N., The effect of food deprivation on enzyme activity in developing brain, J. Neurochem., 17 (1970) 1387-1391. 27 Wannemaker, R. W., Banks, W. L. and Wunner, W. H., Use of a single tissue extract to determine cellular protein and nuclein acid concentration and rate of amino-acid in corporation, Analyt. Biochem., I I (1967) 320-326. 28 Wiggins, R. C., Miller, S. L., Benjamin, J. A., Krigman, M. R. and Morell, P., Myelin synthesis during postnatal nutrition deprivation and subsequences rehabilitation, Brah~ Research., 107 (1976) 257-273. 29 Winick, M., Nutrition and nerve cell growth, Fed. Prod., (1970) 1510-1515.
379
EFFECTS OF U N D E R N U T R I T I O N ON G L I A L M A T U R A T I O N *
O.
ROBAIN and G. PONSOT
I N S E R M U 154 Hopital Saint Vincent de Paul, 75674 Paris, Cedex 14 (France)
(Accepted October 27th, 1977)
SUMMARY Rats of Wistar strain were conceived and breast-fed until the 25th day by mothers maintained on a low protein (5 ~ ) and low caloric (21 calories/day) diet, producing a severe deficiency in weight body growth (more than 50 ~ at the 10th day) and of the weight of the central nervous system (40 ~ on the 15th day) both in the cerebral hemispheres and the spinal cord. Histological and biochemical analysis of the central nervous system shows: (1) Glial proliferation is insufficient and delayed, the number of glial cells is reduced by 50 ~ on the 10th day in the cuneatus and gracilis tracts and the density of the glial cells is reduced by 5 0 ~ in the corpus callosum at the 19th day. (2) Maturation of the glial cells is greatly retarded, especially in the corpus callosum a structure which matures late. On the 19th day, the majority of the cells in this structure still have a glioblastic appearence, whereas in the normal rat, at this age the majority of the glial cells are oligodendrocytes. (3) These abnormalities of glial maturation agree well with the delay of the increase of DNA, RNA and protein measured in the spinal cord and cerebral hemispheres. (4) There is a defect in myelination assessed by estimation of the density of the myelin fibres, and a definitive reduction in the caliber of the spinal tracts.
INTRODUCTION The effect of undernutrition on the developing brain in the rat has been extensively investigated. It is well documented that undernutrition is particularly harmful during the period of rapid brain development 9. The effect on myelination, which occurs in the rat entirely after birth, and mainly during the first month of life, * This study was supported by ATP N ° 27 INSERM.
380 has been demonstrated biochemically2,S, 2s. Histologically, this abnormal myelination is shown by a reduction in the number and thickness of the myelin sheaths, both in the central and peripheral nervous system4,t1,le,2~, 2~. The effect of undernutrition on gliat maturation has been investigated less. The only definite finding so far observed is a reduction in the total number of cells of the central nervous system, assessed by D N A estimations. Histologically, Bass et al."- described a defect in migration and differentiation of glial cells; Cragg 6, Krigman and Hogan t~ mentioned a scarcity of glia in th e gray and white matter. A reduction in the area occupied by gtia in the molecular layer of the cerebellum has been describedL Effects of undernutrition on the mouse anterior commissure have been examined quantitatively by Sturrock et al. 2~ but only at 19 weeks afterbirth and by Siassi and Siassi 2° but only in the somatosensory cortex. The purpose of our work was to provide more information on the effects of undernutrition on glial maturation, and on the myelination of some tracts maturing at different periods. MATERIALS AND METHODS Adult female rats of Wistar strain, weighing at least 250 g, were ted with a restricted calorie and protein diet having the following composition: Casein 5 °/E, tournesol oil 1%, carbohydrates 88 % (half starch, half sugar), cellulose 2 o~, mineral salts 5 % (calcium carbonate 24 %, bicalcium phosphate 20 %, potassium phosphate 15%, calcium lactate 10%, magnesium sulphate 10~o, iron lactate 10~i, sodium chloride 6 %, sodium phosphate 4 %, copper sulphate 0.6 %, cobalt nitrate 0. 1%, zinc acetate 0.2 %), methionine 0.3 %, vitamin B complex 0.5 %, vitamin D 3000 I.U./Kg, vitamin E 30 I.U./Kg. The rats used for this experiment received 6 g per day of this diel (i.e. 21 calories) for 8 days. They were then mated 5 successive nights and the time of fertilisation determined by daily vaginal smears. The fertilised rats were maintained on the same diet throughout pregnancy and lactation, up to the 25th day. The control animals had free access to a diet of pellets containing 22 % protein. 5 o/lipids, 52 o/~ carbohydrate. They consumed an average of 15 g daily, i.e. 51 calories. Litters of the control and undernourished rats were adjusted to 5-7 animals. The pups were examined on the 5th, 10th, 15th, 19th and 25th days after birth.
Biochemical analysis After ether anesthesia, the pups were decapitated and the brains were rapidly removed from the skull. The cerebral hemispheres were separated from the cerebellum and brain stem, by a section just above the superior colliculus, weighed and frozen at 30 °C. Measurements were carried out on 8 pups from at least three different litters. D N A and R N A were measured by Wannemacker's method z5 as modified by Shibko is, the total protein by Lowry's method 15. These estimations were carried out on 8 undernourished rats and on 4 controls from at least 3 different litters.
381
Histological stud), The pups anaesthesized with ether were fixed by intra-aortic perfusion in vivo, according to a previously described method iv. The whole head and spinal cord were kept overnight in Karnovsky solution modified (1 ~ glutaraldehyde, 1 ~ formaldehyde, in 0.12 M phosphate buffer with 0.02 m M CaClz). Corpus callosum was removed from coronal slices of brain at the level of the optic chiasm. Transverse sections (1 ram) of spinal cord were obtained at the dorsal level, half way between lumbar and cervical enlargements. Blocks were postfixed in 2 ~ osmium tetroxide in 0.12 M phosphate buffer at pH 7.4 for 4 h, stained with uranyl acetate and embedded in araldite. Thick sections were stained with 1 ~ toluidine blue, and thin sections were stained with uranyl acetate saturated in alcohol 50° and after by Reynolds' method.
Quantitative stud), The total number of glial nuclei present in each fasciculus was counted in each slice examined, the counts being made directly on the photomicrographs at a magnification of 188 ×. The surface area of the fasciculi were measured with a polar planimeter. The surface of the corpus callosum could not be traced with precision therefore glial cells were counted not in the whole corpus callosum but in 10 squares, measuring 36 × 36 #m, with an ocular net micrometer, using paraffin sections 7.5 #m thick. The density of myelinated fibers was measured on photographs of thin horizontal sections enlarged to 7500 × in the 3 fasciculi of the posterior column of the spinal cord and on sagittal sections of the corpus callosum. Electronmicrographs were taken of the anterior region of the corticospinal fasciculus in order to have the minimum number of aberrant fibers. They were also taken of the posterior region of the gracilis and cuneatus fasciculi, since the delineation of these fasciculi is more clear at this level than further from the surface. In the corpus callosum, the myelinated fibers were counted on sagittal sections, on photographs also magnified to 7500 x . All measurements were carried out on 5 normal rats and 5 undernourished rats at 5, 10, 15, 19 and 25 days, obtained from at least 3 different litters. RESULTS Delivery occurred only in 50 ~ of fertilised rat dams in the starved group. The mortality of the offsprings was 50 ~. Deaths occurred mostly during the first 2 or 3 days. Lactation in undernourished rats was not regularly induced. The undernourished pups were smaller than the controls: by day 10 they weighed 5 g and 12.5 g respectivily (Fig. 1A). Opening of the eyes was delayed by three days and occurred around the 18th day in the undernourished group. HISTOLOGICAL RESULTS
(1) Glial cells (A) Quantitation study Normal rats. Table I indicates the number of glial cells in the fasciculi cuneatus,
382 Body weight
mg
A
Hemispheres weight , ~ _ . _ . . ~ B
i
900
30
700
25
500
20
300
15
106 0
/ i,iJ F
1
I
I
l
I
-t I1
5
10
15
19
25
days
10
mg 200 -Spinal cord weight
0
l
I
i
l
5
10
15
19
C
150 25
days
Well Nourished .....
Poorly Nourished
0
5
10
15
19
25
days Fig. 1. The difference in the weight of the body (A), cerebral hemispheres (B), and the spinal cord (C) of well nourished and poorly nourished animals are shown. The values are the means, and the bars indicate standard deviation. Eight rats from at least three different litters are used,
gracilis and corticospinal. As already noted by Matthews and Duncan 16, there is a definitive chronological difference in the proliferation of cells in these different tracts. In the cuncatus, glial cells were already numerous by the 5th day and proliferation was almost completed by the 10th day. In the fasciculus gracilis, glial multiplication occurred a little later and continued until the 15th day. In the corticospinal tract, cell multiplication continued beyond the 15th day and at least till the 25th. The reduction in the number of cells observed between the 19th and 25th days in the cuneate and graeilis tracts need not be interpreted as a true rarefaction but may derive from a dispersion depending on the increase in length of the spinal cord la. In the undernourished rats, there was a reduction of the total number of glial cells from the 10th day onwards. It is conspicuous in the posterior column of the spinal cord. The difference is significant in each of the three fasciculi studied. It reaches 50 % by the 10th day in the gracilis and cuneatus tracts and by the t5th dayin the corticospinal tract.
383 TABLE I
Number of glia cells in posterior column of the spinal cord of control and undernourished rats Values are m e a n s ± S t a n d a r d Deviation. T h e results are f r o m 5 rats with no m o r e t h a n 2 from a litter. Difference between control a n d u n d e r n o u r i s h e d are evaluated with M a n n - W h i t n e y test (one-tailed test)
Days
5 10 15 19 25
Fasciculus cuneatus
Fasciculus gracilis
Fasciculus corticospinal
Normal
Undernourished Normal
Undernourished Normal
Undernourished
59.2±8.64 100.0±5.38 121.8±7.08 111.8±7.19 105.2±8.25
53.6±6.84NS 53.2±6.69"* 85.8±7.53** 87.2±6.53** 76.6±7.64**
22.4±2.61NS 28.8±3.35"* 60.4±5.32"* 68.4±4.16'* 43.8±3.38**
10.4±2.6"* 16.8±2.23"* 23.0±1.58"* 27.2t-l.92"* 33.4:~3.85"*
21.6±4.16 58.6±6.27 100.4~6.39 103.6±7.50 92.2±-8.58
17 ± 1 . 5 8 23.8±4.49 42 ± 4 . 4 7 41.5±5.47 55.6±3.71
NS, P > 0.05. * 0.05 > P > 0.01. ** 0.01 ~ - P .
Table II indicates the average number of glial cells counted in the corpus callosum: In the normal rats, cell density increases rapidly between the 10th and 19th days, and this is in agreement with Schonbach18; in the undernourished rats, the cell multiplication is diminished, the difference with control rats is significant from the 15th day onward and very evident on the 19th day (Fig. 2).
( B) Maturation of glial cells Normal rats: corpus callosum. This is specially well seen in longitudinal section (Fig. 3). In the normal rats free subependymal cells 14 or glioblasts 2~ constitute the large majority of cells up to the 10th day (Fig. 4). The length of the cells is sometimes 4 times greater than the width. They have a nucleus which contains one or two large T A B L E II
Number of glia cells in corpus callosum In l0 square m e a s u r i n g 36 × 36/~m. Values are m e a n s ± Standard Deviation. T h e results are f r o m 5 rats with n o m o r e t h a n 2 f r o m a litter. Difference between control a n d u n d e r n o u r i s h e d are evaluated with M a n n - W h i t n e y test (one-tailed test).
Days'
Normal
5 10 15 19 25
24.4 28.4 39.6 49.33 50.33
± ± ± ± ±
NS, P > 0.05. * 0.05 > P > 0.01. ** 0.01 ~> P.
Undernourished 3.05 4.16 2.41 6.35 3.88
20 22 23 23.83 34
± ± ± ± ±
2.34* 2.92* 6.48** 3.43** 4.73**
Fig. 2. R a t c o r p u s c a l l o s u m 19th day after birth, t00. A : control rat : note the a r r a n g e m e n t of glial cells in rows. B: u n d e r n o u r i s h e d rat: note the decrease a n d the dispersion o f the glial ceH~.
4 P
Fig. 3. Rat corpus callosum 19 days after birth. ~: 650. A : control rat : most of the glial cells present in the picture are dark (arrow) or light oligodendrocytes (arrowhead). B: undernourished rat: note the scarcity of the cells, which are dark and light glioblast. Note on the right top two light ligodendrocytes (arrowhead).
386 nucleoli. Their cytoplasm is reduced and is concentrated at each pole of the nucleus: it contains a great number of ribosomes grouped in rosettes. Two kinds of glioblasts can be seen (Fig. 4A and B). They differ mainly by their nuclei: some cells have dark nucleus, with aggregated chromatin and have an ovoid and sometimes irregular shape (Fig. 4A); the other gliobtasts have a clear nucleus with dispersed chromatin and, often, have a very regular shape (Fig. 4B). The nucleus of both cell types may exceed 15 microns in length. The proportion of these cells decreases rapidly after the 10th day. At the 19th day they constitute 10-20 ~o of the glial cell population (Fig. 3A) and at the 25th day only about 10 °/ /0 ' Oligodendrocytes cannot be recognized at the 5th day; young oligodendroglia cells appear on the 10th day: they have a large nucleus, abundant cytoplasm and ~1 narrow electron lucent endoplasmic reticulum. The intermediate and dark forms of adult type can be recognized on the 15th day. At the 19th day, the oligodendrocytes have mainly an immature appearence and constitute 75'i~o of the cells (Fig. 3A). They form rows of 6-8 cells at the 25th day. Astrocytes may be identified easily at the 5th day : most of them are youn~ cells and their transition from the glioblasts with light nuclei can be presumed, because these cells have already a very elongated nucleus. The chromatin, however, tends to aggregate near the nuclear membrane. The cytoplasm of these cells, contains large cisternae with an electron dense content ~:~. in these young cells there are neither glycogen granules nor gliofilaments. Microglia were identified on the 5th day. They have a nucleus with chromatm aggregated beneath the nuclear membrane and in the cytoplasm rough endoplasmic reticulum is proeminent with very elongated cisternae. Some dense bodies ale visible, Spinal cord. Maturation occurred earlier than in the corpus callosum: by the 15th day (oligodendrocytes appeared by the 5th day) oligodendrocytes constitute more than half of cell population, dark adult oligodendrocytes appear by the 5th day in the cuneatus, by the 10th day in gracilis and by the 19th day in the corticospinal tract. Undernourished rats: corpus callosum. Free subependymal cells persist for a much longer time in great number and at 19th day they constitute half the cell population of the corpus callosum (Fig. 3B). The delay in appearence of oligodendrocytes is evident: these cells cannot be identified before the 15th day. at the t9th day they represent less than 20~);i of the cell population (Fig. 3B). Rows of oligodendrocytes are short and scarce and at the 25th day only doublets or triplets can be seen. We did not see any difference in the maturation and proliferation of astrocytes except for a higher density of gtiofilaments from the 15th day. Microglia cells were identified at the 5th day as in the normal rats. Some microglia cells have numerous intracytoplasmic inclusions resembling lipids, Sometimes. intracytoplasmic inclusions are dense, large with irregular shape. The cells containing these inclusions have sometimes disposed fine cisternae. It is difficult to be sure that all these cells are microglia or dark oligodendrocytes containing inclusions (Fig. 5B).
Fig. 4. Undernourished rat corpus callosum 10 days after birth. A : glioblast with dark nucleus, x 7000. The cytoplasm contains mitochondria, free ribosomes and numerous scattered polysomes; Note the absence of endoplasmic reticulum (inset). B: glioblast with light and very elongated nucleus. > 8000. As in the top cell, the cytoplasm contains only mitochondria, free ribosomes and polysomes (inset).
388
389
(2) Myelin sheath~s Table 1II indicates progress in myelination in the posterior column of the spinal cord and in the corpus callosum from 5 to 25 days. In the normal rat, chronological difference in myelination between the cuneate, gracilis, corticospinal tract, and corpus callosum is quite clear, the multiplication of the myelin sheats is conspicuous between the 5th and the 15th day in the fasciculus cuneatus and gracilis, between the 10th and the 25th day in the fasciculus corticospinal and the corpus callosum in the undernourished rat, the myelination is very delayed. The difference with the normal rat can be seen on the 5th day in the cuneatus, from the 5th to the 15th in the gracilis, from the 15th to the 25th day in the corticospinal tract and in the corpus callosum. The density of the myelin sheaths became identical from the 10th onwards in the cuneatus and from the 19th day in the gracilis tract. This phenomenon does not mean that the absolute number of myelin sheaths is identical, because the striking increase of the caliber of the bundles has to be taken in account. Table IV showed a definite defect in the growth of the bundles of the posterior column in the undernutrition. This defect indicates a deficiency in the total number of myelin sheaths. BIOCHEMICAL RESULTS
(1) DNA Cerebralhemispheres (Fig. 7A). From the 5th day onwards, the amount of DNA in the cerebral compared to the 35 ~. In control days, whereas in and 25th days.
hemispheres of undernourished animals is reduced to 25~,,, as controls. D N A reduction is maximal at the 15th day when it reaches animals the increase in D N A is maximal between the 10th and 15th undernourished rats, the maximum increase occurs between the 15th
Spinal cord (Fig. 7B). Cell multiplication, as expressed by DNA, is more affected in the spinal cord than in the cerebral hemispheres. The evolution in the level of DNA during maturation is the same as in the cerebral hemispheres. (2) RNA Cerebral hemispheres (Fig. 7C). In control animals the maximal increase in R N A level occurs between the 10th and 15th day. After the 19th day, there is a
Fig. 5. A : undernourished rat 5 days after birth, corpus callosum, x 7000.Young astrocyte with elongated nucleus containing two nucleoli and peripheral clumps of chromatin. The cytoplasm exhibits large cisternae of rough endoplasmicreticulum, which are filled with gray material, some mitochondria and polysomes. Note the absence of gliofilaments and glycogengranules. (Inset) Arrow indicates one lipid body. B : undernourished rat 10 days after birth, fasciculus gracilis. × 7000. Glial cell with nucleus containing clumps of chromatin beneath the nuclear membrane, and a striking nucleolus. This dark cytoplasm contains numerous dark bodies and lipid droplets, short fine strands of rough endoplasmic reticulum and important Golgi apparatus. This cell is probably dark mature oligodendrocyte.Note in the corner below the dark cytoplasm of an oligodendrocyte.
390
0
e-
.
.
.
.
[
i
**zz ..~
e-
qb
•
.
ZZZ~ O
¢
~ESE
_= Av/v/ ~ck~ <
~**
391
e-
8 e.
o t..,,I
Z~*** o o ¢-
o
o
¢..
F
.I
<
,,4 c~qo o~ AV/?/
!
r~
~i ~¸¸¸,ii!
393 reduction in RNA as already reported by Fish and Winick 1°. In undernourished rats the maximal increase occurs between the 15th and the 19th days and continues until the 25th day. Therefore, the curves for controls and undernourished rats cross one another. Spinal cord (Fig. 7D). The RNA content of spinal cord of undernourished rats is more diminished than that of the cerebral hemispheres. The difference reaches 50 ~ at the 15th day. In normal rats there is a maximal increase in RNA between the 10th and 15th days and a reduction after the 19th day, whereas in undernourished pups, RNA continues to increase after this date.
(3) Changes in protein content (Fig. 7E and F) Cerebral hemispheres protein synthesis is highly disturbed by undernutrition: there is a reduction of almost 5 0 ~ in the quantity of protein at the 15th day. The difference is still 25 ~ at the 25th day. In the spinal cord (Fig. 7F), the reduction is even more marked. By the 15th day, protein levels are reduced by almost 6 0 ~ in undernourished animals and by the 25th day the difference is still 4 0 ~ .
(4) Protein/DNA ratio In undernourished pups the more severe involvement in protein synthesis compared with that of DNA produces a reduction in the protein/DNA ratio, both in the cerebral hemispheres and in the spinal cord. This reduction is significant from the 15th day in the cerebral hemispheres (controls = 54.14 ~ 2.90, undernourished = 41.38 ± 4.05, P < 0.001) and from the 19th day in the spinal cord (controls = 40.05 ~0.93, undernourished -- 33.05 ± 1.98, P < 0.02). DISCUSSION
(1) Our quantitative study of glia shows a reduction in the absolute number of these cells in the posterior column of the spinal cord and of their density in the corpus callosum. As shown by Matthews and Duncan ~5, glial proliferation closely follows the volumetric increase of the tracts undergoing myelination. Counting of the total number of glial cells present on the whole section of the posterior column eliminates the variation due to different rates of growth of the caliber of the tracts. However, the variation due to the different rates of increase in length of the spinal cord is not suppressed. Thus, the true difference in the number of glial cells between normal and undernourished rats is more marked than that which we measured. In the corpus callosum, cell density gives a very imperfect indication of the cell population. Schonbach et al. 17, showed that, in the normal rat, cell density ceases to increase after the 20th day, whereas cell proliferation continues. Our findings in control were identical. This may be explained by the dispersion of the cells: the myelination fibres, increase in diameter and separate the glial cells from one another. The defect of myelination seen in the undernourished rats implies that the increase in volume of the corpus callosum is lower than that for the controls. The true
- - .
Well Nour,shed
.....
Poorly Nourished
DNA
Fg
Hemispheres
A
1000
;_I_--i .... {- '{
500
0
I
0
I
I'5
10
5
~
--
19
25
o
s
10
days
pg
R N A Hemispheres
Hg
~5
spinal card
RNA
oot_
D
//
.-
/
,/
300 b
l/z/
1000
)I
/
200F
/
-"
~"
looI_ {"
500
0
____J-
I
0
5
_
~
10
t
~
.
15
.
.
.
.
.
0 . . . .
.
25
19
0
5
10
days
t9
15
Proteins spinal cord
Fj,~I
16
Proteins Hemispheres
E
60
~2
/
/
40 ¸
/~/
2O
ok_ 0
i/ •
25
oovs
mg mg
2'~
C
2000
1500
1~
days
/
~
..{.-"
/
~1/
i
8
/q
.,"""
_j/"
i
I
J
5
10
15 days
19
25
0
o
~
~o
K is
~
i 2s
days
Fig. 7. The graphs indicate the total DNA-RNA and proteins in the cerebral hemispheres and spinal cord from birth to the 25th day. The values are means (8 rats from at least 3 different litters are used for the undernourished group, and 4 rats for the control group) and the bars indicate standard deviation.
395 reduction in glia is thus, here also more important than that which we measured. Whatever the lack of precision of our histological counts they clearly show the essential role of deficiency of the glia in reduction in cell proliferation as shown by DNA estimation. With our model of ante and postnatal undernutrition, this deficiency is greater than that reported by Swaimann et al. 24, Culley and Linerberger 7, in postnatal undernutrition, and is comparable with that observed by Winnick 2s, with ante and postnatal undernutrition and that of Krigman and Hogan 12, in undernutrition restricted to the postnatal period. It is generally accepted that the disturbance in cell proliferation in undernutrition concerns the glial population; the number of neurones is approximately stable after birth a. This view has received very little accurate histological confirmation. Krigman and Hogan 12 simply found a reduction of glial cells in the white and gray matter; Bass et al. 2, also mentioned, a rarefaction of glia in the subcortical white matter but these authors have not done a quantitative evaluation. The more important work is this of Sturrock et al. 25: these authors described reduction of the oligodendrocytes in the mouse anterior commissure at 19th weeks after birth. Our data quantitatively demonstrate important glial deficiency in malnutrition during the first 25 days. Can the low level of DNA in undernourished rats be explained only by the scarcity of glial cells? Histological quantification is not sufficient to state this categorically. Glia/neurone ratio increases 200~ from 0 to 22 days in the spinal cord of normal rat la. This shows the importance which can be attributed to the impairment of glial proliferation at these ages. (2) A deficiency in maturation of glial cells is demonstrated by our study especially in the corpus callosum, where immature cells still persist in large numbers on the 25th day after birth. We did not count separately the different cell types. The cells are very different in size and shape and therefore Abercrombie's correction, taking into consideration the diameter of the nucleus, is difficult to apply to these cells. The length of their nuclei may be 4 times greater than the width and it is sometimes difficult to say along which axis they have been sectioned. Our findings concerning the delay in the maturation of glia is different from that reported by Krigman and Hogan 12, (rarefaction without morphological changes) and confirms the findings briefly mentioned by Bass et al. 2 in samples from the subcortex. This delay in cell maturation is shown also by our biochemical findings: in undernourished rats a fall in protein/DNA ratio is noted from the 15th day onwards in the cerebral hemispheres and from the 10th day onwards in the spinal cord. This ratio gives an approximate indication of the nucleo-cytoplasmic ratio. Probably the delay in the peak of the RNA curve in undernutrition (Fig. 7C and D) is related to the delay in maturation. This defect in maturation of glia may explain the delay in myelination and the persistence at 30 days of a significant percentage of promyelin fibres noted by Krigman and Hogan 12, in the pyramidal tract. (3) The effect on myelination is the best known histological feature in undernutrition. The morphological evaluation of myelin deficiency in growing structures is difficult. Study of the rate of increase in myelinated fibres and the thickness of myelin
396 s h e a t h s in u n d e r n u t r i t i o n d i d n o t s h o w t h e t r u e myelin deficiency. T h e deficiency tn g r o w t h o f t h e calibre o f the t r a c t s s h o w n in o u r s t u d y gives a g o o d i n d i c a t i o n o f t h e deficiency o f m y e l i n a t i n g fibres. It is in t h e r e g i o n s w h i c h m a t u r e late, such as t h e c o r p u s c a l l o s u m , t h a t gtial a b n o r m a l i t i e s are t h e m o s t p r o m i n e n t a n d it is in this s t r u c t u r e t h a t the deficiency in m y e l i n a t i o n is also t h e m o s t m a r k e d (Fig. 6). L i k e w i s e , t h e c o r t i c o s p i n a l t r a c t , w h e r e m a t u r a t i o n o c c u r s later, is m o s t g r e a t l y a f f e c t e d t h a n gracilis a n d c u n e a t u s .
REFERENCES 1 Albercrombie, M., Estimation of nuclear population from mlcrotome section. Attar. Rec.. 94 (1946) 239-247. 2 Bass, N. H., Netsky, M. G. and Young, E.. Effect of neonatal malnutrition on developing cerebrum. I. Microchemical and histologic study of cellular differentiation in the rat. 11. Microchemical and histologic study of myelin formation in the rat. Arch. Neurol. (Chic.~. 23 (1970) 289-302 and 303-313. 3 Brizzee, K. R., Vigt, J. and Kharetchko, X., Postnatal changes in glial neuron index with a comparison of methods of cell enumeration in the white rat, Prog. Brain Res., 4 (1964) 136--149. 4 Clos, J. and Legrand, J., Influence de la d6ficience thyroidienne et de la sous-atimentation sur la croissance et la myetinisation des fibres nerveuses du nerf sciatique chez le jeune rat blanc: 6tude au microscope electronique, Brain Research., 22 (1970) 285-297. 5 Clos, J., Rebiere, A. and Legrand, J., Differential effects of hypothyroidism and undernutrition on the development of glia in the rat cerebellum, Brain Research, 63 (1973) 445-449. 6 Cragg, B. G.. The development of cortical synapses during starvation in the rat. Brain. 95 (1972l 143-150.
7 Culley, W. J. and Lineberger, R. O., Effect of undernutrition on the size and composition of the rat brain, J. Nutr., 96 (1968) 375-386. 8 Davison. A. N. and Dobbing, J., Myelination as a vulnerable period in brain development, Brain reed. Bull., 22 (1966) 404-444. 9 Dobbing, J., Vulnerable periods in developing brain. In A. N. Davison and J. Dobbing (Eds.), Applied Neurochemistry, Davis, Philadelphia, 1968, pp. 9-29. 10 Fish, I. and Winick, M., Cellular growth in various regions of the developing rat brain, Ped. Res., 3 (1969) 407412. 1l Hedley-Whyte, E. T., Myelination of rat sciatic nerve: comparison of undernutrition and cholesterol biosynthesis inhibition, J. Neuropath. exp. Nearol., 32 (1973) 284-302. 12 Krigman, M. R. and Hogan, E. L., Undernutrition in the developing rat: effect upon myelination. Brain Research. 107 (1976) 239-255. 13 Ling, E. A.. Study in the changes of the proportions and numbers of the various glial celt types in the spinal cord of neonatal and young adult rats, Acta anat. (Basel), 96 (1976) 188-195. 14 Ling, E. A., Paterson, J. A., Privat, A., Mori, S. and Leblond, C. P.. Investigation of glint cells in Semithin sections. I. Identification of glial cells in the brain of young rats, 3". comp. Neurol,, 149 (1973) 43-72. 15 Lowry, O. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J., Protein measurement with the Folin phenol reagent, J. biol. Chem., 193 (1951) 265-275. 16 Matthews, M. A. and Duncan. D.. A quantitative study of morphological changes accompartymg the initiation and progress of myelin production in the dorsal funiculus of the rat spinal cord. J. comp. Neurol. 142 (1971) 1-22. 17 Privat, A., Robain, O. et Mandel, P., Aspects ultrastructuraux du corps calleux chez la souns Jimpy, Acta neuropath. (BerL), 21 (1972) 282-295. 18 Schonbach, J., Hu, K. H. and Friede, R. L., Cellular and chemical changes during myelination" histologic, autoradiographic, histochemical, and biochemical data on myelination in the pyramidal tract and corpus callosum of rat, J. comp. NeuroL, 413 (1968) 21-38. 19 Shibko, S., Koivistoinen, P., Tratnyck, C. A., Newhall, A. R. and Friedman, L., A method for sequential quantitative separation and determination of protein, RNA, lipid and glycogen from a single rat liver homogenate or from a subcellular fraction, Analyt. Bioehern., t9 (1967~ 514-528.
397 20 Siassi, F. and Siassi, B., Differential effects of protein-calorie restriction and subsequent rejection in neuronal and non-neuronal components of cerebral cortex in newborn rat, J. Nutr., 103 (1973) 1625-1633. 21 Sima, A., Studies on fibre size in developing sciatic nerve and spinal roots in normal, undernourished and rehabilitated, Acta physiol, scand., Suppl. 406 (1974) 5-55. 22 Sima, A. and Sourander, P., The effect of early undernutrition on the calibre spectrum of the rat optic nerve, Acta neuropath. (Berl.), 28 (1974) 1-10. 23 Skoff, R. P., Price, D. L. and Stocks, A., Electron microscopic autoradiographic studies of gliogenesis in rat optic nerve. I. Cell proliferation, lI. Time of origine, J. comp. Neurol., 169 (1976) 291-311 and 313 - 333. 24 Sturrock, R. R., Light microscopic identification of immature glial cells in semithin sections of the developing mouse corpus callosum. J. Anat. (Lond.), 122 (1976) 521 537. 25 Sturrock, R. R., Smart, J. C. and Dobbing J., Effect on denutrition during the suckling period on growth of the anterior and posterior limbs of the mouse anterior commissure, Neuropath. appl. Neurobiol., 2 (1976) 239-247. 26 Swaiman, K. F., Daleiden, J. M. and Wolfe, R. N., The effect of food deprivation on enzyme activity in developing brain, J. Neurochem., 17 (1970) 1387-1391. 27 Wannemaker, R. W., Banks, W. L. and Wunner, W. H., Use of a single tissue extract to determine cellular protein and nuclein acid concentration and rate of amino-acid in corporation, Analyt. Biochem., I I (1967) 320-326. 28 Wiggins, R. C., Miller, S. L., Benjamin, J. A., Krigman, M. R. and Morell, P., Myelin synthesis during postnatal nutrition deprivation and subsequences rehabilitation, Brah~ Research., 107 (1976) 257-273. 29 Winick, M., Nutrition and nerve cell growth, Fed. Prod., (1970) 1510-1515.