- Email: [email protected]
The content of amino acids in the developing cerebellar cortex and deep cerebellar nuclei of granule cell deficient mutant mice
Brain Research, 247 (1982) 65-73 Elsevier Biomedical Press
65
The Content of Amino Acids in the Developing Cerebellar Cortex and Deep Cerebellar Nuclei of Granule Cell Deficient Mutant Mice SUZANNE ROFFLER-TARLOV* and MAUREEN TUREY
Department of Neurology, Tufts University, School of Medicine, 136 Harrison Avenue, Boston, MA 02111 (U.S.A.) (Accepted February llth, 1982)
Key words: amino acid transmitter - - glutamic acid - - GABA - - cerebellar mutant - - deep cerebellar nucleus - - staggerer - weaver
Glutamic acid is the only free amino acid to be clearly reduced in mature granule cell deficient cerebellum. The correlation between concentration of glutamic acid and extent of granule cell loss suggests that it may serve as a neurotransmitter. Curiously, in two neurological mouse mutants, glutamic acid is also decreased in the deep cerebellar nuclei where there are no granule cells. We have now examined the amino acid content of cerebellar cortex and deep cerebellar nuclei of the granule cell deficient mutants, weaver and staggerer, during the postnatal period in which granule cell development takes place. We have found: (1) an early and transient deficit in taurine in weaver cerebellar cortex during the period of granule cell migration, (2) deficits during the second postnatal week in taurine, aspartic and glutamic acids in both weaver and staggerer cerebellar cortex, (3) that aspartic and glutamic acid deficits result from failure to increase concentrations at the normal rate after birth rather than from a fall from normal levels, (4) decreased concentrations of glutamic acid but not of taurine and aspartic acids apparent in the deep nuclei of both weaver and staggerer at about the same time as in cerebellar cortex, (5) amino acid changes in weaver heterozygote cerebellum which result in values intermediate in magnitude between normal and homozygous weaver animals and (6) an early and persistent reduction in staggerer deep nuclei of gamma-aminobutyricacid (GABA), the Purkinje cell transmitter, indicating early denervation or lack of full innervation of deep nuclei by Purkinje cells. INTRODUCTION A parallel decrease of cerebellar glutamic acid a n d granule cells occurs in m a t u r e animals affected as neonates by m u t a t i o n s , virus a n d X-irradiationT, 11, 12,19,23,32,37, whereas glutamic acid is n o t affected by destruction of inhibitory n e u r o n s in cerebellar cortex 21,23,31. The c o n s t a n t association of concentrat i o n of this excitatory a m i n o acid a n d the n u m b e r s of excitatory granule cells has led to the hypothesis that glutamic acid m a y be the n e u r o t r a n s m i t t e r of the granule cells; however, in the mouse n e u r o logical m u t a n t s , glutamic acid is decreased even in deep cerebellar nuclei (dcn) where there are n o granule cells 23. The purpose of the present report is to examine more closely the association of glutamic acid c o n c e n t r a t i o n a n d granule cell n u m b e r . G l u t a -
mic a n d other free a m i n o acids extracted from n o r m a l a n d granuloprival mouse cerebellar cortex were measured at several time points d u r i n g the first 3 weeks after birth. It is d u r i n g this period that granule cells are b o r n , develop parallel fiber processes a n d migrate from the external surface of the cerebellum t h r o u g h the molecular layer to form the internal g r a n u l a r layerlL D u r i n g the same period, a b e r r a n t events in the two m u t a n t s staggerer a n d weaver lead to granule cell deficiency. Both weaver (gene symbol, wv) a n d staggerer (gene symbol, sg) are a u t o s o m a l recessive m u t a t i o n s which result in a n almost complete loss of cerebellar granule cells 26,z7. The wv a n d sg genes are located on different chrom o s o m e s a n d the two diseases follow different pathological courses. I n n o r m a l development, proliferation of granule cells in the external granule cell
* This work was initiated with Maureen Turey at the Department of Neuroscience, Children's Hospital Medical Center, Boston, MA 02115, and continued at Dr. Roffter-Tarlov's present address. Maureen Turey is now a student of Medicine at Columbia University. 0006-8993/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press
66 layer of mouse cerebellar cortex begins on postnatal day 3 or 413. On about day 4, the first of the postmitotic granule cells lying deep in the external granular cell layer form horizontal cytoplasmic processes in the longitudinal axis of the cerebellar folium. These become the parallel fiber axons. The granule cell body proceeds to migrate deeper into the cerebellum past the Purkinje cell layer to form the internal granular layer. Most wv/wv granule cells die in the proliferative zone, the external granule layer, without developing axons and failing to migrate to the internal granular layerS,16,17,1s,28,29,34. Some granule cells remain in lateral regions. Heterozygote animals suffer a modest loss of granule cells 5. The cause of cell death may be failure of migration due to abnormal Bergman glial 'guide' fibers 16-18 or to a direct effect of the gene on the granule cell 2,29 or both. Purkinje cells remain in this granule cell- free cerebellum though some are lost at the midline 2°. In sg/sg cerebellum the proliferation of granule cells is reduced in the external granule layerg,s4 in addition to death of granule cells which have migrated to the internal granular layer 27. The death of the internal granule cells in this mutant is thought to be secondary to failure of synapse formation with Purkinje cells which are malformedg, 30 and severely reduced in number due to the action of the mutant geneS, 4. The weaver mutant provides an opportunity to chart the time course of glutamic acid depletion during postnatal development of a cerebellum in which the internal granular layer has never formed. The staggerer mutant enables us to follow the time course of glutamate reduction in a cerebellum in which the granule cells have died after their migration into an internal granular layer. The mutants have a number of characteristics in common. These include severe ataxia, and in the mature cerebellum an almost complete loss of cerebellar granule cells as well as reduced glutamic acid concentrations. This report follows the development of the mature free amino acid pattern in the cerebellar cortex of wv/wv, q-/wv, sg/sg and normal littermates from postnatal day 4 when proliferation and migration of granule cells from the external granule cell layer has just begun until day 21 when the mature stratified cerebellum has been formed. The concentrations of amino acids have also been monitored in the deep
cerebellar nuclei which are targets for Purkinje cell axons and which contain none of the other neurons present in cerebellar cortex. MATERIALS AND METHODS
Animals The wv/wv, +/wv and littermate + / + animals used in these studies were of the strain in which the mutation originally occurred, C57B1/6J 26. Affected animals were obtained by breeding heterozygotes. The mature affected animals can be recognized behaviorally by severe ataxia, hypotonia and fine tremor25, 26. Seven-day-old mutant animals were identified by poor righting responses. The 4-day-old animals were distinguishable only histologically. To obtain data on 4-day-old animals, entire litters were taken from known heterozygous parents and the animals were classified in retrospect as wv/wv, q-/wv or q-/q- by histological criteria. For this purpose the 4-day-old cerebella were divided in half sagitally. One half was extracted for amino acid analyses and the other was fixed by immersion in quarter strength Karnovsky's fixatives (glutaraledehyde and paraformaldehyde, each 1 70 in 0.1 M phosphate buffer, pH 7.3) at room temperature. After postfixation in OsO4, dehydration and embedding in Epon, sections 1 #m thick were cut in the sagittal plane and stained with alkaline toluidine blue. The q-/q-, +/wv and wv/wv animals were distinguished by examination of the stained histological sections. At postnatal day 4, + / + animals have a small part of the granule cell population present in the internal granular layer, the wv/wv animals had no granule cells in the internal granular layer and the + / w v animals were intermediate. The cerebella from 7-day and older mice were identified as being q-/q- or +/wv by their appearance when viewed under the dissecting microscope. Compared to a ÷ / ÷ cerebellum, that of the +/wv looks misshapen and slightly deflated. The wv/wv cerebellum appears frankly atrophic.
Staggerer The staggerer (sg) mutation 27, was obtained from the Jackson Laboratory in a C57B1 stock with the closely linked marker genes dilute (d) and short ear (se). The mutation is maintained by brother × sister
67 mating of heterozygotes. Homozygous staggerer mice are recognized by their black fur and abnormal righting responses after about postnatal day 7 (P7). Littermates that are q-/+ at the sg locus are recognized by their diluted coat color and short ears, whereas the heterozygotes have black fur and normal behavior. The cerebellum of an sg/sg animal can be identified easily on the fourth postnatal day by its abnormal appearance as a narrow smooth bar of tissue. Data from heterozygous and homozygous normal littermate controls were pooled since at all ages studied their cerebellar weights and amino acid concentrations were identical.
Separation of cerebellar deep nuclei and cerebellar cortex Animals were killed by decapitation. The brains were quickly removed and the cerebella put into a beaker of ice-cold phosphate-buffered saline to chill for several minutes. Each cerebellum was then placed on the glass stage of a dissecting microscope. Tabs of choroid plexus, cerebellar peduncle and vestibular gray matter were trimmed away from the ventral surface with the aid of the microscope and the parafloccular lobes were discarded. The cerebellum was cut into 10 sagittal sections using very cold, sharp blades. A sagittal face of each section was viewed under the microscope using bright scattered illumination from below. The cortical gray matter was clearly separated visually from white matter which separates the folia and the white matter core containing the deep nuclei. Using cold, sharp blades, the white core containing the deep nuclei was cut from the sections in which it appeared and was analyzed separately from the cortical layers. The dissection of the cerebellum of 4- and 7-day-old mice required that the animals be anesthetized (Avertin) and perfused through the heart briefly with ice-cold phosphate-buffered saline as. The brain after cold perfusion is very hard and can be sliced. Sagittal slices of the young animals when viewed with intense illumination from below showed differential translucency in the deep nuclei. The darker piece (dcn) was separated from the cerebellar cortex on each sagittal section using the small, cold, sharp blades. For all animals, the time between decapitation and homogenization of the tissues was kept constant to control for postmortem chemical
changes.
Amino acid analysis The tissue was homogenized in from 250/~1 to 1 ml of ice-cold 5 ~ perchloric acid 24. The samples were centrifuged at 24,000 g for 30 min. The pellet was dissolved by heating in 1 N NaOH and aliquots of the solution were used for protein determinations 10. The supernatant was brought to pH 4 with 5 ~ KOH, and centrifuged to remove the KC104. The supernatant was dried and reconstituted in 0.2 N sodium citrate buffer, pH 2.2, for analysis. Samples were analyzed by a Beckman model 121M amino acid analyzer using the single column method and a sodium citrate buffer systemaS. Peaks were compared to standards for retention times and absorption ratio between 440 and 570 mm. The wellseparated peaks corresponded to authentic taurine, aspartic acid, glutamic acid, glycine, alanine and GABA. While glutamine and serine appeared in one peak, acid hydrolysis showed that this peak was almost entirely glutamine. Recoveries were estimated by the recovery of norleucine added at the homogenization step and averaged 88 ~. Statistical significance of the differences between values obtained from control and mutant tissue were assessed with the two-tailed Student's t-test or when multiple comparisons were made (between q-/+, q-/wv and wv/wv animals), by a one-way analysis of variance followed by Bonferroni's procedure 14. RESULTS
Weights of cerebella during postnatal development Weaver. Four days after birth, the cerebella from q-/q-, q-/wv and wv/wv animals could not be distinguished from one another except by the histological criteria described in the Methods. Weights of the 4-day-old wv/wv and q-/wv cerebella were not reduced compared to -q-/÷ cerebella. Three days later it is obvious by their appearance under the dissecting microscope that both the wv/wv and the q-/wv cerebella have failed to grow normally. At postnatal day 7, -t-/wv cerebella weigh 80~ and wv/wv cerebella weigh only 56~ of values from -t-/q- littermates (Fig. 1). At 10 days of age the heterozygote cerebellum is still intermediate in size, weighing 69 of -t-/+ values; the wv/wv cerebellum weighs 55
68
50 40
I]]] Control
[ ] +I+ I~ +/wv
[ ] Sg/Sg
[ ] wv/wv
I"1(..9 LIJ
5O Z
F-" LIJ (_9
Z Z % %
4
7
I0
21
11iI 4
7
I0
2i
AGE IN DAYS
Fig. 1. The mean weights :k S.E.M. of cerebella from ÷ / + , +/wv and wv/wvanimals and from control ( + / + and -~/sg) and sg/sg animals at 4 time points (4, 7, 10 and 21 days) during the postnatal development of the cerebellum. The numbers of samples for each graphed value range from 3 to 5.
of + / + values. The + / + and +/wv cerebella achieve further growth between postnatal day 10 and 20 but the wv/wv cerebellum does not. After 3 weeks of age, wv/wv cerebella weigh 39 ~ of + / + controls and +/wv cerebella weigh 9 0 ~ of -k-/+ controls (Fig. 1). Staggerer. The effect of the staggerer mutation is evident even at 4 days when viewed by gross inspection. At this young age it weighs only 62 ~ of normal ( ÷ / + or +/sg) values (Fig. 1). The staggerer cerebellum fails to grow during the following 3 days whereas the size of the normal cerebellum doubles. Therefore at postnatal day 7, the staggerer cerebellum weighs 36 ~ of normal values. Staggerer cerebellum weighs 34 ~ and 28 ~ of normal values at l0 and 21 days respectively. Like wv/wv, the sg/sg cerebellum does not grow after postnatal day 10 (Fig. 1).
The free amino acid content of cerebellar cortex Weaver. Among the free amino acids measured (taurine, aspartic, glutamic, glutamine/serine, glycine, alanine and GABA) only taurine was reduced in concentration (nmol/mg wet weight) in wv/wv cerebellar cortex compared to heterozygote and homozygote controls (Fig. 2). Taurine concentration in wv/wvwas 80 ~o of normal values. Reductions of taurine (64 ~ of + / + value) and glutamate (78 of ÷ / + value)were seen in the 7-day-old weaver animals. A reduction in the concentration in the same amino acids plus aspartate and glutamine/se-
rine was evident at 10 days. Reductions in glutamate, and taurine concentrations occurred also in the heterozygote animals so that values were intermediate between those of the homozygote normal and the mutants. Concentrations of taurine, glutamine/serine and aspartic acid 'recover'. In the case of aspartic acid, concentration increases between 10 and 21 days in all 3 types of animals. A sharp reduction in concentration of taurine occurs between 10 and 21 days in normal cerebellum. The extent of the reduction in taurine is much greater for the homozygote normal and heterozygote animals compared to the weavers to reach the same final concentration of 10 nmol/mg wet weight. Glutamic acid is the only amino acid to be lower in concentration in wv/wv and +/wv animals compared to +/q- when the animals are 21 days old. The concentration of glutamic acid in the 21-day-old heretozygote cerebella was between +/q- and wv/wv values. The concentration of glutamic acid fails to increase normally as the mutant cerebellum matures; glutamic acid concentrations increase by 68 between postnatal days 4 and 10 in normal cerebellum whereas glutamic acid levels do not change at all during this period in wv/wv cerebellum. Glutamic acid concentration in wv/wv does increase between 10 and 21 days but does not reach normal concentrations. GABA and glutamine/serine levels in wv/wv cerebellar cortex were greater than normal at 21 days. Staggerer. None of the free amino acids was reduced in concentration in the 4- or in the 7-dayold staggerer cerebellar cortex compared to controls (Fig. 3) although the staggerer cerebellum was much compromised at these ages in terms of size (Fig. 1). Reductions in the concentrations of 3 amino acids occurred between 7 and 10 days of age. These were glutamic acid, taurine and aspartic acid, which on postnatal day 10 were 75, 84 and 67 ~ of control values respectively. The decreased concentrations of these amino acids persist in staggerer. Glutamate continued to decline relative to control values so that at 21 days its concentration in staggerer cerebellum was 52 ~ of control values. Taurine and aspartate were 71 and 63 ~o of control values at 21 days. The concentrations of GABA and of glutamine/serine in sg/sg cerebellar cortex were not different from control at any age studied.
69
~0 GLUTAMATE
I-T
25
8
20
6
15
4
10
2
5
TAURINE
ASPARTATE
Ld
I.-
LLI
0 r'r" LLI {3_ IC
4
7
10
21
GLUTAMINE / SERINE
0
4
7
10
21
GLYCINE
4
5
d 0
8
4
z
6
3
4
[]+/+ []+/wv
7
10
21
4
7
10
21
7
10
7
10
21
GAi~
4
21
AGE IN DAYS
[] wv/~
Fig. 2. Concentrations of 6 free amino acids extracted from weaver, heterozygousweaver and normal cerebellar cortex expressed as nmol/mg wet weight. Comparison between values for + / ÷ , ÷/wv and wv/wvanimals were made at postnatal days 4, 7, 10 and 21. Values are mean ! S.E.M. for from 3 to 6 animals except for values for 4-day-old + / ÷ animals which are the means of two observations. The values for alanine which are not shown were the same for + / + , +/wv and wv/wvat all ages and were 0.8-1.1 nmol/mg wet weight. *P < 0.05; ** P < 0.01.
The free amino acid content of deep cerebellar nuclei
DISCUSSION
The concentration of glutamic acid fails to increase normally in mutant deep nuclei. The disparity between mutant and control is seen at day 6 in weaver and day 10 in staggerer (Fig. 4). The concentration of GABA is reduced in the deep nuclei but not in the staggerer cerebellar cortex at all postnatal ages studied (Figs. 3 and 4). On postnatal day 4, G A B A concentration in dcn was 72 % of that of normal littermates, whereas none of the other amino acids was affected. The content of G A B A in normal dcn doubles between postnatal days 4 and 7. Such an increase does not occur in staggerer where, on postnatal day 7, G A B A content in dcn is 52% of normal values. The disparity increases; at 3 weeks of age, the staggerer deep nuclei contain only 30% of the normal G A B A content (Fig. 4).
This study points to several developmental features of amino acid content of the granuloprival cerebellar cortex which were not evident in our previous study in which glutamic acid was the only free amino acid whose concentration was reduced in mature and mutant cerebellaZL When the content of free amino acids is followed during the postnatal formation of the cerebellum in weaver, staggerer and normal littermates, it is clear that glutamic acid, aspartic acid and taurine are all abnormally low in both mutant cerebella during the peak migratory period for granule cells. The amino acid deficits are displayed during the same developmental period in both mutants even though the pathologies of the two diseases are very different with death of granule cells occurring after migration in staggerer and prior
70 I0
GLUTAMATE
25
8
4
IO
LU
5
HLU
4
{D
5
ASPARTATE
20 15
HI (._9
TAURINE
IZ: to III £1_ 03 8 LIJ _J 0 6 Z
7
I0
21
1 4
5
GLUTAMINE/SERINE
IO
7
21
4
7
IO
21
I0
21
GLYCINE
GABA
4 3
4
2
I
I
4
7
IO
4
21
[ ] Control
7
I0
21
4
7
AGE IN DAYS
[ ] Sg/Sg
Fig. 3. Concentrations of 6 free amino acids extracted from staggerer and normal cerebellar cortex expressed as nmol/mg wet weight. Comparison between values for control and staggerer were made at postnatal days 4, 7, 10 and 21. Values are mean 3: S.E.M., for from 3 to 5 animals. Values for alanine which are not shown were the same for mutant and control at all ages and were 0.6-0.8 nmol/mg wet weight. *P < 0.01; **P < 0.001.
14oI
WE,4 V E R
STAGGERER
•
120
":'
I..- t 0 0 z
8 I..0
o
. 08
.
•
o
:
oo
§
;o
e • 8O
o
o o
60
o o
•
o
8 o
o o o
•
t
o o o
o
eee •
•
~
4O
2O
o
~
~, ,'2 ,,
16
20
24
A G E I N DAYS
Fig. 4. Concentrations of glutamic acid and GABA in weaver and staggerer deep cerebellar nuclei expressed as percent of control ( + / + values). Points are individual values at several ages from postnatal days 4 to 21. to migration in weaver. The deficits in glutamic and aspartic acids appear between postnatal day 7 and 10 in both weaver and staggerer. Although the
staggerer cerebellum is already diminished to 62 % of normal values for wet weight and protein content on P4-P7 during the proliferative phase of granule cell development, the content of each amino acid is normal when expressed as nmol/mg wet weight whereas low taurine concentration is evident in weaver cerebellar cortex on postnatal days 4--10. The weaver cerebellum 'recovers' with respect to taurine and aspartic acid concentrations so that at 21 days of age, glutamic acid is the only amino acid present in abnormally low concentration. The 3 amino acids are reduced in the 21-day-old staggerer cerebellar cortex but the most prominent deficit is that of glutamic acid. Every change found in the weaver cerebellum is intermediate in magnitude in the heterozygote animals at each age tested. The +/wv animal loses about 20% of its granule cells. We predict that quantitative histology will show the granule cell loss
71 in +/wv to have a regional focus being most prominent in the midline region of the cerebellum because when a midsagittal slice was removed for histology from ÷/wv, no decrease in glutamic acid was detected in the rest of the cerebellum 2a. As shown here in normal mouse cerebellum as well as in prior studies of rat cerebellum 3z, glutamic and aspartic acids increase in concentration during development. Failure to increase concentrations of these acids at the normal rate is characteristic of both weaver and staggerer. In the wv/wv cerebellar cortex, glutamic acid concentration remains the same during at least the first ten days of life, whereas normal levels double. Between 10 and 21 days, the concentration of glutamic acid in the weaver doubles, but is still lower than normal values since glutamic acid increases in the normal cerebellum during this period also. In neither mutant are synapses made between granule cell axons, the parallel fibers, and Purkinje cell dendrites. In the normal animal these synapses are made during this early postnatal period and thus it is tempting to say that the normal increase in glutamic and aspartic acids may be a mark of this event. However, the concentration of glutamic acid in the deep nuclei also fails to increase at the normal rate, suggesting that at least part of the cause is common to both deep nuclei and cerebellar cortex. One possible cause could be retrograde 'transsynaptic' cell death of mossy fibers which normally form prominent synaptic connections with granule cell dendrites and which also innervate the deep nuclei. During the development of the nervous system, the fate of one neuron depends upon that of another to such an extent that neurons may die if unable to form synapses. An example of this extreme is the staggerer cerebellum in which the granule cells, unable to make synapses with their normal targets, die 9,2s. Also relevant is the fact that climbing fibers degenerate following the postnatal degeneration of their primary targets, the Purkinje neurons, in the Lurcher mutant mouse 1. The elimination of Lurcher climbing fibers appears to be caused by the disappearance of their target field since they are not a primary target of the mutation 33. Similarly, mossy fibers may also be eliminated when granule cells disappear. It is not known if all or part of the mossy fiber population contains high concentrations of
glutamate, nor has mossy fiber degeneration been examined quantitatively in these mutant cerebella. However, Sotelo and Changeaux have described degenerative changes in the mossy fibers of staggerer mice 30. It seems likely that a range of reactions occurs in cerebellar neurons and glia which all are induced by death of granule cells or by failure of granule cell migration to take place., The deficits in glutamate as well as changes in other amino acids, such as aspartate and taurine some of which are transient effects, may reflect more subtle reactions such as alterations in the normal pace of development of cerebellar neurons in the mutant mice. One amino acid deficit is specific for the staggerer mutant and it occurs only in the deep nuclei. The concentration of GABA in sg/sg deep nuclei was reduced at P4, the earliest age tested. At this time there were no other amino acid differences in the mutants' deep nuclei and none at all in the cerebellar cortex. This evidence suggests that Purkinje cell terminals which contain GABA and which normally contact the neurons of the deep nuclei as their principal target are not doing so in staggerer. We know that neurons in adult sg/sg dcn are present in normal numbers but their cell bodies are shrunken in size 22. The early decrease in GABA concentration observed here suggests that the explanation for histological changes in the deep nuclei z2 is either early denervation by Purkinje cells or lack of innervation during development. It is possible that most Purkinje cells have never been generated in sg/sg for although 60-90 ~ of Purkinje cells are missing 8, no postnatal degeneration of these cells has ever been detectedZ,9,a0, a4-~6. The present finding is pertinent to the timing of staggerer gene action since early reduction of GABA in the deep nuclei of sg/sg adds to the evidence that staggerer exerts its effects on Purkinje cells early. ACKNOWLEDGEMENTS This study was supported by N I H grants NS 14937 at Children's Hospital and NS 17322 at Tufts University to S.R.T. It is a pleasure to thank Diane Rich and Khalidah Bilal tbr their technical help and to thank Dr. Richard Sidman for teaching us to identify weaver, heterozygote and normal animals in 4-day-old litters.
72 REFERENCES 1 Caddy, K. W. T. and Biscoe, T. J., Structural and quantitative studies on the normal Call and Lurcher mutant mouse, Phil. Trans. B., 287 (1979) 167-201. 2 Goldowitz, D. and Mullen, R. J., Weaver mutant granule cell defect expressed in chimeric mice, Neurosci. Abstr., 6 (1980) 743. 3 Herrup, K. and Mullen, R. J., Regional variation and absence of large neurons in the cerebellum of the staggerer mouse, Brain Research, 172 (1979) 1-12. 4 Herrup, K. and Mullen, R. J., Staggerer chimeras: intrinsic nature of Purkinje cell defects and implications for normal cerebellar development, Brain Research, 178 (1979) 443-458. 5 Hirano, A. and Dembitzer, H. M., Observations on the development of the weaver mouse cerebellum, J. Neuropath. exp. Neurol., 33 (1974) 354-364. 6 Hirano, A. and Dembitzer, H. M., The fine structure of staggerer cerebellum, J. Neuropath. exp. NeuroL, 34 (1975) 1-11. 7 Hudson, D. B., Valcana, T., Bean, G. and Timiras, P. S., Glutamic acid: a strong candidate as the neurotransmitter of the cerebeUar granule cell, Neurochem. Res., 1 (1976) 73-81. 8 Karnovsky, M. J., A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy, J. Cell Biol., 27, (1965) 137A. 9 Landis, D. M. D. and Sidman, R. L., Electron microscopic analysis of postnatal histogenesis in the cerebellar cortex of staggerer mutant mice, J. comp. NeuroL, 179 (1978) 831-864. 10 Lowry, O. H., Rosenbrough, N. J., Farr, A. L. and Randall, R. J., Protein measurements with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. 11 McBride, W. J., Aprison, M. H. and Kusano, K., Contents of several amino acids in the cerebellum, brain stem and cerebrum of the 'staggerer', 'weaver' and 'nervous' neurologically mutant mice, J. Neurochem., 26 (1976) 867-870. 12 McBride, W. J., Nadi, N. S., Airman, J. and Aprison, M. H., Effects of selective doses of X-irradiation on the levels of several amino acids in the cerebellum of the rat, Neurochem. Res., 1 (1976) 141-152. 13 Miale, 1. L. and Sidman, R. L., An autoradiographic analysis of histogenesis in the mouse cerebellum, Exp. Neurol., 4 (1961) 277-296. 14 Miller, R. G., Simultaneous Statistical Inference, McGraw Hill, New York, 1966. 15 Moore, S. and Stein, W. H., Procedures for the chromatographic determination of amino acids on four per cent cross-linked sulfonated polystyrene resins, J. biol. Chem., 211 (1954) 893-906. 16 Rakic, P. and Sidman, R. L., Sequence of developmental abnormalities leading to granule cell deficit in cerebellar cortex of weaver mutant mice, J. comp. Neurol., 152 (1973) 103-132. 17 Rakic, P. and Sidman, R. L., Organization of cerebellar cortex secondary to deficit of granule cells in weaver mutant mice, J. comp. Neurol., 152 (1973) 133-161. 18 Rakic, P. and Sidman, R. L., Weaver mutant mouse cerebellum: defective neuronal migration secondary to abnormality of Bergmann glia, Proc. nat. Acad. Sci. U.S.A., 70 (1973) 240-244.
19 Rea, M. A. and McBride, W. J., Effects of X-irradiation on the levels of glutamate, aspartate and GABA in different regions of the cerebellum of the rat, Life Sci., 23 (1978) 2355-2360. 20 Rezai, Z. and Yoon, C. H., Abnormal rate of granule cell migration in the cerebellum of 'weaver' mutant mice, Develop. Biol., 29 (1972) 17-26. 21 Roffler-Tarlov, S., Beart, P. M., O'Gorman, S. and Sidman, R. L., Neurochemical and morphological consequences of axon terminal degeneration in cerebellar deep nuclei of mice with inherited Purkinje cell degeneration, Brain Research, 168 (1979) 75-95. 22 Roffler-Tarlov, S. and Herrup, K., Quantitative examination of the deep cerebellar nuclei in the staggerer mutant mouse, Brain Research, 215 (1981) 49-59. 23 Roffler-Tarlov, S. and Sidman, R. L., Concentrations of glutamic acid in cerebellar cortex and deep nuclei of normal mice and weaver, staggerer and nervous mutants, Brain Research, 142 (1978) 269-283. 24 Saifet, A., Comparative study of various extraction methods for the quantitative determination of free amino acids from brain tissue, Analyt. Biochem., 40 (1971) 412-423. 25 Sidman, R. L., Development of interneuronal connections in the brain of mutant mice. In F. D. Carlson (Ed.), Physiological and Biochemical Aspects of Nervous Itttegration, Prentice-Hall, Englewood-Cliffs, N.J., 1968, p. 163. 26 Sidman, R. L., Green, M. C. and Appel, S. H., Catalog of the Neurological Mutants of the Mouse, Harvard University Press, Cambridge, MA, 1965. 27 Sidman, R. L., Lane, P. W. and Dickie, M. M., Staggerer, a new mutation in mouse affecting the cerebellum, Science, 137 (1962) 610-612. 28 Sotelo, C., Dendritic abnormalities of Purkinje cells in the cerebellum of neurologic mutant mice (weaver and staggerer). In G. W. Frentzberg (Ed.), Advances in Neurology, I/ol. XII, Raven Press, New York, 1975, pp. 335-351. 29 Sotelo, C. and Changeaux, J.-P., Bergmann fibers and granule cell migration in the cerebellum of homozygous weaver mutant mouse, Brain Research, 77 (1974) 484-491. 30 Sotelo, C. and Changeux, J.-P., Transsynaptic degeneration 'en cascade' in cerebellar cortex of staggerer mutant mice, Brain Research, 67 (1974) 519-526. 31 Tran Vinh Tan and Snyder, S. H., Amino acid neurotransmitter candidates in rat cerebellum: selective effects of kainic acid lesions, Brain Research, 167 (1979) 345-353. 32 Valcana, T., Hudson, D. and Timiras, P. S., Effects of X-irradiation on the content of amino acids in the developing rat cerebellum, J. Neurochem., 19 (1972) 2229-2232. 33 Wetts, R. and Herrup, K., Interaction of granule, Purkinje and inferior olivary neurons in Lurcber chimeric mice, I. Qualitative Studies, in press. 34 Yoon, C. H., Developmental mechanism for changes in the cerebellum of 'staggerer' mouse, a neurological mutant of genetic origin. Neurology, 22 (1972) 743-754. 35 Yoon, C. H., Fine structure of the cerebellum of 'staggerer: reeler', a double mutant of mice affected by staggerer and reeler condition. II. Purkinje cell anomalies, J. Neuropath. exp. Neurol., 36 (1977) 427-439. 36 Yoon, C. H., Pleiotropic effect of the staggerer gene, Brain Research, 109 (1976) 206-215. 37 Young, A. B., Oster-Granite, M. L., Herndon, R. M. and Snyder, S. H., Glutamic acid: selective depletion by viral
73 induced granule cell loss in hamster cerebellum, Brain Research, 73 (1974) 1-13. 38 Zigmond, R. and Ben-Ari, Y., A simple method for the
serial sectioning of fresh brain and the removal of identifiable nuclei from stained sections for biochemical analysis, J. Neurochem., 26 (1976) 1285-1287.
65
The Content of Amino Acids in the Developing Cerebellar Cortex and Deep Cerebellar Nuclei of Granule Cell Deficient Mutant Mice SUZANNE ROFFLER-TARLOV* and MAUREEN TUREY
Department of Neurology, Tufts University, School of Medicine, 136 Harrison Avenue, Boston, MA 02111 (U.S.A.) (Accepted February llth, 1982)
Key words: amino acid transmitter - - glutamic acid - - GABA - - cerebellar mutant - - deep cerebellar nucleus - - staggerer - weaver
Glutamic acid is the only free amino acid to be clearly reduced in mature granule cell deficient cerebellum. The correlation between concentration of glutamic acid and extent of granule cell loss suggests that it may serve as a neurotransmitter. Curiously, in two neurological mouse mutants, glutamic acid is also decreased in the deep cerebellar nuclei where there are no granule cells. We have now examined the amino acid content of cerebellar cortex and deep cerebellar nuclei of the granule cell deficient mutants, weaver and staggerer, during the postnatal period in which granule cell development takes place. We have found: (1) an early and transient deficit in taurine in weaver cerebellar cortex during the period of granule cell migration, (2) deficits during the second postnatal week in taurine, aspartic and glutamic acids in both weaver and staggerer cerebellar cortex, (3) that aspartic and glutamic acid deficits result from failure to increase concentrations at the normal rate after birth rather than from a fall from normal levels, (4) decreased concentrations of glutamic acid but not of taurine and aspartic acids apparent in the deep nuclei of both weaver and staggerer at about the same time as in cerebellar cortex, (5) amino acid changes in weaver heterozygote cerebellum which result in values intermediate in magnitude between normal and homozygous weaver animals and (6) an early and persistent reduction in staggerer deep nuclei of gamma-aminobutyricacid (GABA), the Purkinje cell transmitter, indicating early denervation or lack of full innervation of deep nuclei by Purkinje cells. INTRODUCTION A parallel decrease of cerebellar glutamic acid a n d granule cells occurs in m a t u r e animals affected as neonates by m u t a t i o n s , virus a n d X-irradiationT, 11, 12,19,23,32,37, whereas glutamic acid is n o t affected by destruction of inhibitory n e u r o n s in cerebellar cortex 21,23,31. The c o n s t a n t association of concentrat i o n of this excitatory a m i n o acid a n d the n u m b e r s of excitatory granule cells has led to the hypothesis that glutamic acid m a y be the n e u r o t r a n s m i t t e r of the granule cells; however, in the mouse n e u r o logical m u t a n t s , glutamic acid is decreased even in deep cerebellar nuclei (dcn) where there are n o granule cells 23. The purpose of the present report is to examine more closely the association of glutamic acid c o n c e n t r a t i o n a n d granule cell n u m b e r . G l u t a -
mic a n d other free a m i n o acids extracted from n o r m a l a n d granuloprival mouse cerebellar cortex were measured at several time points d u r i n g the first 3 weeks after birth. It is d u r i n g this period that granule cells are b o r n , develop parallel fiber processes a n d migrate from the external surface of the cerebellum t h r o u g h the molecular layer to form the internal g r a n u l a r layerlL D u r i n g the same period, a b e r r a n t events in the two m u t a n t s staggerer a n d weaver lead to granule cell deficiency. Both weaver (gene symbol, wv) a n d staggerer (gene symbol, sg) are a u t o s o m a l recessive m u t a t i o n s which result in a n almost complete loss of cerebellar granule cells 26,z7. The wv a n d sg genes are located on different chrom o s o m e s a n d the two diseases follow different pathological courses. I n n o r m a l development, proliferation of granule cells in the external granule cell
* This work was initiated with Maureen Turey at the Department of Neuroscience, Children's Hospital Medical Center, Boston, MA 02115, and continued at Dr. Roffter-Tarlov's present address. Maureen Turey is now a student of Medicine at Columbia University. 0006-8993/82/0000-0000/$02.75 © 1982 Elsevier Biomedical Press
66 layer of mouse cerebellar cortex begins on postnatal day 3 or 413. On about day 4, the first of the postmitotic granule cells lying deep in the external granular cell layer form horizontal cytoplasmic processes in the longitudinal axis of the cerebellar folium. These become the parallel fiber axons. The granule cell body proceeds to migrate deeper into the cerebellum past the Purkinje cell layer to form the internal granular layer. Most wv/wv granule cells die in the proliferative zone, the external granule layer, without developing axons and failing to migrate to the internal granular layerS,16,17,1s,28,29,34. Some granule cells remain in lateral regions. Heterozygote animals suffer a modest loss of granule cells 5. The cause of cell death may be failure of migration due to abnormal Bergman glial 'guide' fibers 16-18 or to a direct effect of the gene on the granule cell 2,29 or both. Purkinje cells remain in this granule cell- free cerebellum though some are lost at the midline 2°. In sg/sg cerebellum the proliferation of granule cells is reduced in the external granule layerg,s4 in addition to death of granule cells which have migrated to the internal granular layer 27. The death of the internal granule cells in this mutant is thought to be secondary to failure of synapse formation with Purkinje cells which are malformedg, 30 and severely reduced in number due to the action of the mutant geneS, 4. The weaver mutant provides an opportunity to chart the time course of glutamic acid depletion during postnatal development of a cerebellum in which the internal granular layer has never formed. The staggerer mutant enables us to follow the time course of glutamate reduction in a cerebellum in which the granule cells have died after their migration into an internal granular layer. The mutants have a number of characteristics in common. These include severe ataxia, and in the mature cerebellum an almost complete loss of cerebellar granule cells as well as reduced glutamic acid concentrations. This report follows the development of the mature free amino acid pattern in the cerebellar cortex of wv/wv, q-/wv, sg/sg and normal littermates from postnatal day 4 when proliferation and migration of granule cells from the external granule cell layer has just begun until day 21 when the mature stratified cerebellum has been formed. The concentrations of amino acids have also been monitored in the deep
cerebellar nuclei which are targets for Purkinje cell axons and which contain none of the other neurons present in cerebellar cortex. MATERIALS AND METHODS
Animals The wv/wv, +/wv and littermate + / + animals used in these studies were of the strain in which the mutation originally occurred, C57B1/6J 26. Affected animals were obtained by breeding heterozygotes. The mature affected animals can be recognized behaviorally by severe ataxia, hypotonia and fine tremor25, 26. Seven-day-old mutant animals were identified by poor righting responses. The 4-day-old animals were distinguishable only histologically. To obtain data on 4-day-old animals, entire litters were taken from known heterozygous parents and the animals were classified in retrospect as wv/wv, q-/wv or q-/q- by histological criteria. For this purpose the 4-day-old cerebella were divided in half sagitally. One half was extracted for amino acid analyses and the other was fixed by immersion in quarter strength Karnovsky's fixatives (glutaraledehyde and paraformaldehyde, each 1 70 in 0.1 M phosphate buffer, pH 7.3) at room temperature. After postfixation in OsO4, dehydration and embedding in Epon, sections 1 #m thick were cut in the sagittal plane and stained with alkaline toluidine blue. The q-/q-, +/wv and wv/wv animals were distinguished by examination of the stained histological sections. At postnatal day 4, + / + animals have a small part of the granule cell population present in the internal granular layer, the wv/wv animals had no granule cells in the internal granular layer and the + / w v animals were intermediate. The cerebella from 7-day and older mice were identified as being q-/q- or +/wv by their appearance when viewed under the dissecting microscope. Compared to a ÷ / ÷ cerebellum, that of the +/wv looks misshapen and slightly deflated. The wv/wv cerebellum appears frankly atrophic.
Staggerer The staggerer (sg) mutation 27, was obtained from the Jackson Laboratory in a C57B1 stock with the closely linked marker genes dilute (d) and short ear (se). The mutation is maintained by brother × sister
67 mating of heterozygotes. Homozygous staggerer mice are recognized by their black fur and abnormal righting responses after about postnatal day 7 (P7). Littermates that are q-/+ at the sg locus are recognized by their diluted coat color and short ears, whereas the heterozygotes have black fur and normal behavior. The cerebellum of an sg/sg animal can be identified easily on the fourth postnatal day by its abnormal appearance as a narrow smooth bar of tissue. Data from heterozygous and homozygous normal littermate controls were pooled since at all ages studied their cerebellar weights and amino acid concentrations were identical.
Separation of cerebellar deep nuclei and cerebellar cortex Animals were killed by decapitation. The brains were quickly removed and the cerebella put into a beaker of ice-cold phosphate-buffered saline to chill for several minutes. Each cerebellum was then placed on the glass stage of a dissecting microscope. Tabs of choroid plexus, cerebellar peduncle and vestibular gray matter were trimmed away from the ventral surface with the aid of the microscope and the parafloccular lobes were discarded. The cerebellum was cut into 10 sagittal sections using very cold, sharp blades. A sagittal face of each section was viewed under the microscope using bright scattered illumination from below. The cortical gray matter was clearly separated visually from white matter which separates the folia and the white matter core containing the deep nuclei. Using cold, sharp blades, the white core containing the deep nuclei was cut from the sections in which it appeared and was analyzed separately from the cortical layers. The dissection of the cerebellum of 4- and 7-day-old mice required that the animals be anesthetized (Avertin) and perfused through the heart briefly with ice-cold phosphate-buffered saline as. The brain after cold perfusion is very hard and can be sliced. Sagittal slices of the young animals when viewed with intense illumination from below showed differential translucency in the deep nuclei. The darker piece (dcn) was separated from the cerebellar cortex on each sagittal section using the small, cold, sharp blades. For all animals, the time between decapitation and homogenization of the tissues was kept constant to control for postmortem chemical
changes.
Amino acid analysis The tissue was homogenized in from 250/~1 to 1 ml of ice-cold 5 ~ perchloric acid 24. The samples were centrifuged at 24,000 g for 30 min. The pellet was dissolved by heating in 1 N NaOH and aliquots of the solution were used for protein determinations 10. The supernatant was brought to pH 4 with 5 ~ KOH, and centrifuged to remove the KC104. The supernatant was dried and reconstituted in 0.2 N sodium citrate buffer, pH 2.2, for analysis. Samples were analyzed by a Beckman model 121M amino acid analyzer using the single column method and a sodium citrate buffer systemaS. Peaks were compared to standards for retention times and absorption ratio between 440 and 570 mm. The wellseparated peaks corresponded to authentic taurine, aspartic acid, glutamic acid, glycine, alanine and GABA. While glutamine and serine appeared in one peak, acid hydrolysis showed that this peak was almost entirely glutamine. Recoveries were estimated by the recovery of norleucine added at the homogenization step and averaged 88 ~. Statistical significance of the differences between values obtained from control and mutant tissue were assessed with the two-tailed Student's t-test or when multiple comparisons were made (between q-/+, q-/wv and wv/wv animals), by a one-way analysis of variance followed by Bonferroni's procedure 14. RESULTS
Weights of cerebella during postnatal development Weaver. Four days after birth, the cerebella from q-/q-, q-/wv and wv/wv animals could not be distinguished from one another except by the histological criteria described in the Methods. Weights of the 4-day-old wv/wv and q-/wv cerebella were not reduced compared to -q-/÷ cerebella. Three days later it is obvious by their appearance under the dissecting microscope that both the wv/wv and the q-/wv cerebella have failed to grow normally. At postnatal day 7, -t-/wv cerebella weigh 80~ and wv/wv cerebella weigh only 56~ of values from -t-/q- littermates (Fig. 1). At 10 days of age the heterozygote cerebellum is still intermediate in size, weighing 69 of -t-/+ values; the wv/wv cerebellum weighs 55
68
50 40
I]]] Control
[ ] +I+ I~ +/wv
[ ] Sg/Sg
[ ] wv/wv
I"1(..9 LIJ
5O Z
F-" LIJ (_9
Z Z % %
4
7
I0
21
11iI 4
7
I0
2i
AGE IN DAYS
Fig. 1. The mean weights :k S.E.M. of cerebella from ÷ / + , +/wv and wv/wvanimals and from control ( + / + and -~/sg) and sg/sg animals at 4 time points (4, 7, 10 and 21 days) during the postnatal development of the cerebellum. The numbers of samples for each graphed value range from 3 to 5.
of + / + values. The + / + and +/wv cerebella achieve further growth between postnatal day 10 and 20 but the wv/wv cerebellum does not. After 3 weeks of age, wv/wv cerebella weigh 39 ~ of + / + controls and +/wv cerebella weigh 9 0 ~ of -k-/+ controls (Fig. 1). Staggerer. The effect of the staggerer mutation is evident even at 4 days when viewed by gross inspection. At this young age it weighs only 62 ~ of normal ( ÷ / + or +/sg) values (Fig. 1). The staggerer cerebellum fails to grow during the following 3 days whereas the size of the normal cerebellum doubles. Therefore at postnatal day 7, the staggerer cerebellum weighs 36 ~ of normal values. Staggerer cerebellum weighs 34 ~ and 28 ~ of normal values at l0 and 21 days respectively. Like wv/wv, the sg/sg cerebellum does not grow after postnatal day 10 (Fig. 1).
The free amino acid content of cerebellar cortex Weaver. Among the free amino acids measured (taurine, aspartic, glutamic, glutamine/serine, glycine, alanine and GABA) only taurine was reduced in concentration (nmol/mg wet weight) in wv/wv cerebellar cortex compared to heterozygote and homozygote controls (Fig. 2). Taurine concentration in wv/wvwas 80 ~o of normal values. Reductions of taurine (64 ~ of + / + value) and glutamate (78 of ÷ / + value)were seen in the 7-day-old weaver animals. A reduction in the concentration in the same amino acids plus aspartate and glutamine/se-
rine was evident at 10 days. Reductions in glutamate, and taurine concentrations occurred also in the heterozygote animals so that values were intermediate between those of the homozygote normal and the mutants. Concentrations of taurine, glutamine/serine and aspartic acid 'recover'. In the case of aspartic acid, concentration increases between 10 and 21 days in all 3 types of animals. A sharp reduction in concentration of taurine occurs between 10 and 21 days in normal cerebellum. The extent of the reduction in taurine is much greater for the homozygote normal and heterozygote animals compared to the weavers to reach the same final concentration of 10 nmol/mg wet weight. Glutamic acid is the only amino acid to be lower in concentration in wv/wv and +/wv animals compared to +/q- when the animals are 21 days old. The concentration of glutamic acid in the 21-day-old heretozygote cerebella was between +/q- and wv/wv values. The concentration of glutamic acid fails to increase normally as the mutant cerebellum matures; glutamic acid concentrations increase by 68 between postnatal days 4 and 10 in normal cerebellum whereas glutamic acid levels do not change at all during this period in wv/wv cerebellum. Glutamic acid concentration in wv/wv does increase between 10 and 21 days but does not reach normal concentrations. GABA and glutamine/serine levels in wv/wv cerebellar cortex were greater than normal at 21 days. Staggerer. None of the free amino acids was reduced in concentration in the 4- or in the 7-dayold staggerer cerebellar cortex compared to controls (Fig. 3) although the staggerer cerebellum was much compromised at these ages in terms of size (Fig. 1). Reductions in the concentrations of 3 amino acids occurred between 7 and 10 days of age. These were glutamic acid, taurine and aspartic acid, which on postnatal day 10 were 75, 84 and 67 ~ of control values respectively. The decreased concentrations of these amino acids persist in staggerer. Glutamate continued to decline relative to control values so that at 21 days its concentration in staggerer cerebellum was 52 ~ of control values. Taurine and aspartate were 71 and 63 ~o of control values at 21 days. The concentrations of GABA and of glutamine/serine in sg/sg cerebellar cortex were not different from control at any age studied.
69
~0 GLUTAMATE
I-T
25
8
20
6
15
4
10
2
5
TAURINE
ASPARTATE
Ld
I.-
LLI
0 r'r" LLI {3_ IC
4
7
10
21
GLUTAMINE / SERINE
0
4
7
10
21
GLYCINE
4
5
d 0
8
4
z
6
3
4
[]+/+ []+/wv
7
10
21
4
7
10
21
7
10
7
10
21
GAi~
4
21
AGE IN DAYS
[] wv/~
Fig. 2. Concentrations of 6 free amino acids extracted from weaver, heterozygousweaver and normal cerebellar cortex expressed as nmol/mg wet weight. Comparison between values for + / ÷ , ÷/wv and wv/wvanimals were made at postnatal days 4, 7, 10 and 21. Values are mean ! S.E.M. for from 3 to 6 animals except for values for 4-day-old + / ÷ animals which are the means of two observations. The values for alanine which are not shown were the same for + / + , +/wv and wv/wvat all ages and were 0.8-1.1 nmol/mg wet weight. *P < 0.05; ** P < 0.01.
The free amino acid content of deep cerebellar nuclei
DISCUSSION
The concentration of glutamic acid fails to increase normally in mutant deep nuclei. The disparity between mutant and control is seen at day 6 in weaver and day 10 in staggerer (Fig. 4). The concentration of GABA is reduced in the deep nuclei but not in the staggerer cerebellar cortex at all postnatal ages studied (Figs. 3 and 4). On postnatal day 4, G A B A concentration in dcn was 72 % of that of normal littermates, whereas none of the other amino acids was affected. The content of G A B A in normal dcn doubles between postnatal days 4 and 7. Such an increase does not occur in staggerer where, on postnatal day 7, G A B A content in dcn is 52% of normal values. The disparity increases; at 3 weeks of age, the staggerer deep nuclei contain only 30% of the normal G A B A content (Fig. 4).
This study points to several developmental features of amino acid content of the granuloprival cerebellar cortex which were not evident in our previous study in which glutamic acid was the only free amino acid whose concentration was reduced in mature and mutant cerebellaZL When the content of free amino acids is followed during the postnatal formation of the cerebellum in weaver, staggerer and normal littermates, it is clear that glutamic acid, aspartic acid and taurine are all abnormally low in both mutant cerebella during the peak migratory period for granule cells. The amino acid deficits are displayed during the same developmental period in both mutants even though the pathologies of the two diseases are very different with death of granule cells occurring after migration in staggerer and prior
70 I0
GLUTAMATE
25
8
4
IO
LU
5
HLU
4
{D
5
ASPARTATE
20 15
HI (._9
TAURINE
IZ: to III £1_ 03 8 LIJ _J 0 6 Z
7
I0
21
1 4
5
GLUTAMINE/SERINE
IO
7
21
4
7
IO
21
I0
21
GLYCINE
GABA
4 3
4
2
I
I
4
7
IO
4
21
[ ] Control
7
I0
21
4
7
AGE IN DAYS
[ ] Sg/Sg
Fig. 3. Concentrations of 6 free amino acids extracted from staggerer and normal cerebellar cortex expressed as nmol/mg wet weight. Comparison between values for control and staggerer were made at postnatal days 4, 7, 10 and 21. Values are mean 3: S.E.M., for from 3 to 5 animals. Values for alanine which are not shown were the same for mutant and control at all ages and were 0.6-0.8 nmol/mg wet weight. *P < 0.01; **P < 0.001.
14oI
WE,4 V E R
STAGGERER
•
120
":'
I..- t 0 0 z
8 I..0
o
. 08
.
•
o
:
oo
§
;o
e • 8O
o
o o
60
o o
•
o
8 o
o o o
•
t
o o o
o
eee •
•
~
4O
2O
o
~
~, ,'2 ,,
16
20
24
A G E I N DAYS
Fig. 4. Concentrations of glutamic acid and GABA in weaver and staggerer deep cerebellar nuclei expressed as percent of control ( + / + values). Points are individual values at several ages from postnatal days 4 to 21. to migration in weaver. The deficits in glutamic and aspartic acids appear between postnatal day 7 and 10 in both weaver and staggerer. Although the
staggerer cerebellum is already diminished to 62 % of normal values for wet weight and protein content on P4-P7 during the proliferative phase of granule cell development, the content of each amino acid is normal when expressed as nmol/mg wet weight whereas low taurine concentration is evident in weaver cerebellar cortex on postnatal days 4--10. The weaver cerebellum 'recovers' with respect to taurine and aspartic acid concentrations so that at 21 days of age, glutamic acid is the only amino acid present in abnormally low concentration. The 3 amino acids are reduced in the 21-day-old staggerer cerebellar cortex but the most prominent deficit is that of glutamic acid. Every change found in the weaver cerebellum is intermediate in magnitude in the heterozygote animals at each age tested. The +/wv animal loses about 20% of its granule cells. We predict that quantitative histology will show the granule cell loss
71 in +/wv to have a regional focus being most prominent in the midline region of the cerebellum because when a midsagittal slice was removed for histology from ÷/wv, no decrease in glutamic acid was detected in the rest of the cerebellum 2a. As shown here in normal mouse cerebellum as well as in prior studies of rat cerebellum 3z, glutamic and aspartic acids increase in concentration during development. Failure to increase concentrations of these acids at the normal rate is characteristic of both weaver and staggerer. In the wv/wv cerebellar cortex, glutamic acid concentration remains the same during at least the first ten days of life, whereas normal levels double. Between 10 and 21 days, the concentration of glutamic acid in the weaver doubles, but is still lower than normal values since glutamic acid increases in the normal cerebellum during this period also. In neither mutant are synapses made between granule cell axons, the parallel fibers, and Purkinje cell dendrites. In the normal animal these synapses are made during this early postnatal period and thus it is tempting to say that the normal increase in glutamic and aspartic acids may be a mark of this event. However, the concentration of glutamic acid in the deep nuclei also fails to increase at the normal rate, suggesting that at least part of the cause is common to both deep nuclei and cerebellar cortex. One possible cause could be retrograde 'transsynaptic' cell death of mossy fibers which normally form prominent synaptic connections with granule cell dendrites and which also innervate the deep nuclei. During the development of the nervous system, the fate of one neuron depends upon that of another to such an extent that neurons may die if unable to form synapses. An example of this extreme is the staggerer cerebellum in which the granule cells, unable to make synapses with their normal targets, die 9,2s. Also relevant is the fact that climbing fibers degenerate following the postnatal degeneration of their primary targets, the Purkinje neurons, in the Lurcher mutant mouse 1. The elimination of Lurcher climbing fibers appears to be caused by the disappearance of their target field since they are not a primary target of the mutation 33. Similarly, mossy fibers may also be eliminated when granule cells disappear. It is not known if all or part of the mossy fiber population contains high concentrations of
glutamate, nor has mossy fiber degeneration been examined quantitatively in these mutant cerebella. However, Sotelo and Changeaux have described degenerative changes in the mossy fibers of staggerer mice 30. It seems likely that a range of reactions occurs in cerebellar neurons and glia which all are induced by death of granule cells or by failure of granule cell migration to take place., The deficits in glutamate as well as changes in other amino acids, such as aspartate and taurine some of which are transient effects, may reflect more subtle reactions such as alterations in the normal pace of development of cerebellar neurons in the mutant mice. One amino acid deficit is specific for the staggerer mutant and it occurs only in the deep nuclei. The concentration of GABA in sg/sg deep nuclei was reduced at P4, the earliest age tested. At this time there were no other amino acid differences in the mutants' deep nuclei and none at all in the cerebellar cortex. This evidence suggests that Purkinje cell terminals which contain GABA and which normally contact the neurons of the deep nuclei as their principal target are not doing so in staggerer. We know that neurons in adult sg/sg dcn are present in normal numbers but their cell bodies are shrunken in size 22. The early decrease in GABA concentration observed here suggests that the explanation for histological changes in the deep nuclei z2 is either early denervation by Purkinje cells or lack of innervation during development. It is possible that most Purkinje cells have never been generated in sg/sg for although 60-90 ~ of Purkinje cells are missing 8, no postnatal degeneration of these cells has ever been detectedZ,9,a0, a4-~6. The present finding is pertinent to the timing of staggerer gene action since early reduction of GABA in the deep nuclei of sg/sg adds to the evidence that staggerer exerts its effects on Purkinje cells early. ACKNOWLEDGEMENTS This study was supported by N I H grants NS 14937 at Children's Hospital and NS 17322 at Tufts University to S.R.T. It is a pleasure to thank Diane Rich and Khalidah Bilal tbr their technical help and to thank Dr. Richard Sidman for teaching us to identify weaver, heterozygote and normal animals in 4-day-old litters.
72 REFERENCES 1 Caddy, K. W. T. and Biscoe, T. J., Structural and quantitative studies on the normal Call and Lurcher mutant mouse, Phil. Trans. B., 287 (1979) 167-201. 2 Goldowitz, D. and Mullen, R. J., Weaver mutant granule cell defect expressed in chimeric mice, Neurosci. Abstr., 6 (1980) 743. 3 Herrup, K. and Mullen, R. J., Regional variation and absence of large neurons in the cerebellum of the staggerer mouse, Brain Research, 172 (1979) 1-12. 4 Herrup, K. and Mullen, R. J., Staggerer chimeras: intrinsic nature of Purkinje cell defects and implications for normal cerebellar development, Brain Research, 178 (1979) 443-458. 5 Hirano, A. and Dembitzer, H. M., Observations on the development of the weaver mouse cerebellum, J. Neuropath. exp. Neurol., 33 (1974) 354-364. 6 Hirano, A. and Dembitzer, H. M., The fine structure of staggerer cerebellum, J. Neuropath. exp. NeuroL, 34 (1975) 1-11. 7 Hudson, D. B., Valcana, T., Bean, G. and Timiras, P. S., Glutamic acid: a strong candidate as the neurotransmitter of the cerebeUar granule cell, Neurochem. Res., 1 (1976) 73-81. 8 Karnovsky, M. J., A formaldehyde-glutaraldehyde fixative of high osmolarity for use in electron microscopy, J. Cell Biol., 27, (1965) 137A. 9 Landis, D. M. D. and Sidman, R. L., Electron microscopic analysis of postnatal histogenesis in the cerebellar cortex of staggerer mutant mice, J. comp. NeuroL, 179 (1978) 831-864. 10 Lowry, O. H., Rosenbrough, N. J., Farr, A. L. and Randall, R. J., Protein measurements with the Folin phenol reagent, J. Biol. Chem., 193 (1951) 265-275. 11 McBride, W. J., Aprison, M. H. and Kusano, K., Contents of several amino acids in the cerebellum, brain stem and cerebrum of the 'staggerer', 'weaver' and 'nervous' neurologically mutant mice, J. Neurochem., 26 (1976) 867-870. 12 McBride, W. J., Nadi, N. S., Airman, J. and Aprison, M. H., Effects of selective doses of X-irradiation on the levels of several amino acids in the cerebellum of the rat, Neurochem. Res., 1 (1976) 141-152. 13 Miale, 1. L. and Sidman, R. L., An autoradiographic analysis of histogenesis in the mouse cerebellum, Exp. Neurol., 4 (1961) 277-296. 14 Miller, R. G., Simultaneous Statistical Inference, McGraw Hill, New York, 1966. 15 Moore, S. and Stein, W. H., Procedures for the chromatographic determination of amino acids on four per cent cross-linked sulfonated polystyrene resins, J. biol. Chem., 211 (1954) 893-906. 16 Rakic, P. and Sidman, R. L., Sequence of developmental abnormalities leading to granule cell deficit in cerebellar cortex of weaver mutant mice, J. comp. Neurol., 152 (1973) 103-132. 17 Rakic, P. and Sidman, R. L., Organization of cerebellar cortex secondary to deficit of granule cells in weaver mutant mice, J. comp. Neurol., 152 (1973) 133-161. 18 Rakic, P. and Sidman, R. L., Weaver mutant mouse cerebellum: defective neuronal migration secondary to abnormality of Bergmann glia, Proc. nat. Acad. Sci. U.S.A., 70 (1973) 240-244.
19 Rea, M. A. and McBride, W. J., Effects of X-irradiation on the levels of glutamate, aspartate and GABA in different regions of the cerebellum of the rat, Life Sci., 23 (1978) 2355-2360. 20 Rezai, Z. and Yoon, C. H., Abnormal rate of granule cell migration in the cerebellum of 'weaver' mutant mice, Develop. Biol., 29 (1972) 17-26. 21 Roffler-Tarlov, S., Beart, P. M., O'Gorman, S. and Sidman, R. L., Neurochemical and morphological consequences of axon terminal degeneration in cerebellar deep nuclei of mice with inherited Purkinje cell degeneration, Brain Research, 168 (1979) 75-95. 22 Roffler-Tarlov, S. and Herrup, K., Quantitative examination of the deep cerebellar nuclei in the staggerer mutant mouse, Brain Research, 215 (1981) 49-59. 23 Roffler-Tarlov, S. and Sidman, R. L., Concentrations of glutamic acid in cerebellar cortex and deep nuclei of normal mice and weaver, staggerer and nervous mutants, Brain Research, 142 (1978) 269-283. 24 Saifet, A., Comparative study of various extraction methods for the quantitative determination of free amino acids from brain tissue, Analyt. Biochem., 40 (1971) 412-423. 25 Sidman, R. L., Development of interneuronal connections in the brain of mutant mice. In F. D. Carlson (Ed.), Physiological and Biochemical Aspects of Nervous Itttegration, Prentice-Hall, Englewood-Cliffs, N.J., 1968, p. 163. 26 Sidman, R. L., Green, M. C. and Appel, S. H., Catalog of the Neurological Mutants of the Mouse, Harvard University Press, Cambridge, MA, 1965. 27 Sidman, R. L., Lane, P. W. and Dickie, M. M., Staggerer, a new mutation in mouse affecting the cerebellum, Science, 137 (1962) 610-612. 28 Sotelo, C., Dendritic abnormalities of Purkinje cells in the cerebellum of neurologic mutant mice (weaver and staggerer). In G. W. Frentzberg (Ed.), Advances in Neurology, I/ol. XII, Raven Press, New York, 1975, pp. 335-351. 29 Sotelo, C. and Changeaux, J.-P., Bergmann fibers and granule cell migration in the cerebellum of homozygous weaver mutant mouse, Brain Research, 77 (1974) 484-491. 30 Sotelo, C. and Changeux, J.-P., Transsynaptic degeneration 'en cascade' in cerebellar cortex of staggerer mutant mice, Brain Research, 67 (1974) 519-526. 31 Tran Vinh Tan and Snyder, S. H., Amino acid neurotransmitter candidates in rat cerebellum: selective effects of kainic acid lesions, Brain Research, 167 (1979) 345-353. 32 Valcana, T., Hudson, D. and Timiras, P. S., Effects of X-irradiation on the content of amino acids in the developing rat cerebellum, J. Neurochem., 19 (1972) 2229-2232. 33 Wetts, R. and Herrup, K., Interaction of granule, Purkinje and inferior olivary neurons in Lurcber chimeric mice, I. Qualitative Studies, in press. 34 Yoon, C. H., Developmental mechanism for changes in the cerebellum of 'staggerer' mouse, a neurological mutant of genetic origin. Neurology, 22 (1972) 743-754. 35 Yoon, C. H., Fine structure of the cerebellum of 'staggerer: reeler', a double mutant of mice affected by staggerer and reeler condition. II. Purkinje cell anomalies, J. Neuropath. exp. Neurol., 36 (1977) 427-439. 36 Yoon, C. H., Pleiotropic effect of the staggerer gene, Brain Research, 109 (1976) 206-215. 37 Young, A. B., Oster-Granite, M. L., Herndon, R. M. and Snyder, S. H., Glutamic acid: selective depletion by viral
73 induced granule cell loss in hamster cerebellum, Brain Research, 73 (1974) 1-13. 38 Zigmond, R. and Ben-Ari, Y., A simple method for the
serial sectioning of fresh brain and the removal of identifiable nuclei from stained sections for biochemical analysis, J. Neurochem., 26 (1976) 1285-1287.