ELSEVIER

Neuroscience Letters 200 (1995) 109-112

HEUROSCIEHC[
I[ITfRS

Apoptotic cell death in the cerebellum of mutant w e a v e r and l u r c h e r mice

U. Wtillner*, P.-A. L6schmann, M. Weller, T. Klockgether
Department of Neurology, Eberhard-Karls-University, Hoppe-Seyler-Strasse 3, D-72076 Tiibingen, Germany

Received 21 August 1995; revised version received 9 October 1995; accepted 9 October 1995

Abstract

Apoptosis in the central nervous system of mutant weaver (wv) and lurcher (lc) mice was studied using in situ DNA end-labeling.
The number of apoptotic cells in the extemal granule cell layer of weaver (wv/+) mice at postnatal day 9 was increased six- to eightfold compared to (+/+) littermates. Purkinje cells and Bergmann glia cells were not affected. No labeled cells were found in the substantia nigra. In lurcher (lc/+) mice, labeled nuclei of large cells were found in the Purkinje cell layer, suggesting that Purkinje cells of
mutant lurcher mice undergo apoptosis. Apoptotic granule cell nuclei were found in the internal granule cell layer. In contrast to
weaver (wv/+) mice, the number of apoptotic cells in the external granule cell layer was not increased. Apoptosis appears to be the
common pathway of cell death in the degenerative processes induced by the weaver and the lurcher gene, respectively. These mutant
mice offer the opportunity to study the mechanisms that underlie inappropriate apoptotic cell death in vivo.
Keywords: Apoptosis; Mutant; Mouse; Cerebellum; Neurodegenerative disease

Programmed cell death (PCD) plays a major role in the

differentiation and development of the nervous system
[11]. Most types of PCD in the central nervous system
exhibit morphological and biochemical features of apoptosis, i.e. chromatin condensation and DNA fragmentation in oligonucleosomal multiples of 180-200 bp [18].
PCD of neuronal cells appears to be an active process that
requires de novo R N A and protein synthesis for the
completion of cell death [19]. While apoptosis traditionally has been regarded as a developmental event, recent
experimental evidence suggests the involvement of apoptotic cell death also in neurodegenerative processes [13].
In human neurodegenerative disease the presence of cells
undergoing apoptosis is controversial; recently, apoptosis
has been demonstrated in Huntington's and Alzheimer's
disease [4,14]. However, previous work failed to identify
apoptotic cell death in Alzheimer's disease, Amyotrophic
lateral sclerosis and in human prion disease [9]. At present, it is unknown whether a p o p t o f c cell death might also
play a role in the pathogenesis of Parkinson's disease and
the cerebellar ataxias. We have examined two genetic
animal models of neurodegenerative disease, the mutant
* Corresponding author. Tel.: +49 7071 297616; fax: +49 7071
296507.

mouse strains w e a v e r (wv) and lurcher (lc). Both mutations cause early postnatal loss of neurons in the cerebellar cortex [20]. W e a v e r is an autosomal recessive mutation; homozygous animals develop a characteristic behavioral syndrome consisting of gait instability, tremor and
ataxia [20]. Histological analysis of homozygous and
heterozygous animals reveals loss of cerebellar granule
cells during the first postnatal weeks, probably related to
an intrinsic defect of progenitor cells [5,6,15,16]. Homozygous w e a v e r mice exhibit additional pathology of the
dopaminergic nigrostriatal system [17]. L u r c h e r is an
autosomal dominant mutation located on chromosome 6
that resembles human adult dominant ataxia in its histological and clinical features [12,20]. While homozygous
animals die shortly after birth, heterozygous animals develop gait instability and ataxia during the second postnatal week [1]. Histologically, Purkinje cells and, to a lesser
extent, granule and inferior olive neurons are affected [2].
Mutant mice were obtained from heterozygous parents
(B6CBACa-A(w-J)/A-wv and B6CBACa-A(w-J)/A-lc;
Jackson Laboratories, Bar Harbor, ME). C57BL/6J mice
were purchased from Interfauna, Germany. Gait instability was the first symptom in both w e a v e r and lurcher
mice. Although the clinical phenotype had not yet fully
evolved, litters could be separated based on gait instabii-

0304-3940/95/$09.50 1995 Elsevier Science Ireland Ltd. All rights reserved

SSDI 0304-3940(95)12090-N

U. Wiillner et al. / Neuroscience Letters 200 (1995) 109-112

I10

ext

ext
i i~

gcl
gcl

~',

,a

,>

gcl

i:

9ii

:if!

:ii(~

~!~i:: (i:

g
Fig. 1. (A) Cerebellum of C57BL mice at P9; occasional apoptotic cells are found in the internal (gcl) and external (ext) granule cell layer. (B) Cerebellum of (wv/+) weaver mice at P9; apoptotic progenitor and granule cells are found in the external granule cell layer. (C) Cerebellum of (lc/+)
lurcher mice at P9; large labeled nuclei are found in the Purkinje cell layer. This section is double-stained with Cresyl violet; the arrow points at an
intact Purkinje cell, note the dens nucleolus. (D) Cerebellum of (lc/+) lurcher mice at P60; Purkinje cells are continuously lost through apoptosis, no
apoptotic granule cells are detectable; the arrow points at the former external granule cell layer. Scale bar, 50/zm.

ity and cerebellar size into 'probably affected' and

'probably not affected' mice at postnatal days 9-10 (P9).
At least three 'probably affected' animals from each
strain were studied and compared to their control littermates. In addition, lurcher mice (lc/+) were studied at
P60. The brains were fixed in 4% paraformaldehyde in
phosphate buffered saline (PBS, pH 7.4), paraffin embedded, and sagittal 8/zm sections were processed according
to the TUNEL technique [7]. Briefly, sections were defatted and treated with proteinase K (20/zg/ml, 15 min).
Endogenous peroxidase was inactivated by 2% H202 . The
sections were rinsed in PBS and incubated in a humid
chamber at 37C with terminal deoxynucleotidyl transferase (TdT, 200 U/ml; Boehringer, Mannheim) and biotinylated dUTP (10/tmol/ml; Boehringer, Mannheim) in
TdT buffer (25 mM Tris-HC1, 200 mM sodium cacodylate, 5 mM cobalt chloride) for 15 min. The reaction was
terminated with citrate buffer (300 mM sodium chloride,
30 mM sodium citrate), the sections covered with bovine
serum albumin (2%, 30 min), rinsed in PBS, incubated
with streptavidin-biotin peroxidase complex (Elite, Vectastain ABC kit; 30 min) and developed with diaminobenzidine (Sigma). In each experiment, control sections

were processed in the absence of cobalt chloride. No nuclei were labeled in these sections. Labeled cells were
considered apoptotic if they (1) were isolated and (2)
showed signs of chromatin condensation or fragmentation, according to Migheli et al. [9]. Apoptotic cells were
counted on at least eight regularly spaced sections per
animal at the midsagittal level of the cerebellum. To
compare the number of apoptotic granule cells, the length
of the external and the internal granule cell layer on each
section was measured using an image processing system
(MCID M4, Imaging Research, St. Catharines, Ontario)
and the number of apoptotic cells per mm determined.
Statistical analysis was performed using one-way
ANOVA and post hoc Scheff6 F-test. Adjacent sections
were processed for Cresyl violet stain, tyrosine hydroxylase-(TH)-, and glial fibrillary acidic protein-(GFAP)immunohistochemistry according to standard protocols.
In C57BL mice at P9 few cells in the external and the
internal granule cell layer were labeled (Fig. la). In addition, very small nuclei throughout the cerebellar cortex
were labeled that could represent late stage pyknotic
granule cells or possibly astrocytes. Outside the cerebellum a few labeled cells were found at the former border

U. Wiillner et al. / Neuroscience Letters 200 (1995) 109-112

a
6O
[]

1+); a ffe c t e d at P 9

[]

( ); n o t a f f e c t e d a t P 9

5O

40

3o
-6
20

1o

I!
C57BL

12-

[]

(+); af f ected at P9

[]

(-); n o t af f ec t e d at P9

Wv

10"

8"

6-

E
2

C57BL

Wv

Fig. 2. (a) Number of apoptotic cells per 2 mm external granule cell

layer. (b) Number of apoptotic cells per 2 mm internal granule cell
layer; this includes possible Golgi cells which at this stage cannot be
clearly differentiated. Cells were counted on at least eight regularly
spaced sections (at the midsagittal level of the cerebellum) per animal,
on each section, the length of the external (a)/internal (b) granule cell
layer was measured and the number of cells per unit determined. Bars
represent the average +__SEM of at least three animals per group.
of the subependymal plate in the parietal cortex and in the
ventral striatum, close to the globus pallidus.
In affected w e a v e r mice, the number of apoptotic cells
found in the external granule cell layer was increased sixto eight-fold compared to C57BL and unaffected w e a v e r
littermates (Figs. lb, 2a,b). The majority of labeled nuclei
were located close to the molecular layer. Purkinje cell
nuclei were not labeled; similarly, Bergmann glia cells
appeared to be unaffected (Fig. lb). With respect to areas
other than the cerebellum, scattered apoptotic cells were
found in the striatum, somewhat concentrated at the border to the globus pallidus, and again at the former border
of the subependymal plate in the parietal cortex. No
apoptotic cells were present in hippocampus, olfactory
bulb or in the midbrain, particularly in the substantia nigra. None of the affected w e a v e r mice examined displayed the cytoarchitectural features of homozygous
(wv/wv) animals, i.e. a thickened Purkinje cell layer of
several cells with almost no granule cells in the usual
zone below them but vertical 'stacks' of postmitotic
granule cells extending from the external granule cell
layer [ 15,20]. Instead, the layering and order of the exter-

111

nal granule cell layer, molecular layer and internal granule cell layer appeared grossly intact. It is therefore most
likely that only heterozygous (wv/+) w e a v e r mice are
included in this group. This is supported by the results of
TH immunohistochemistry. No overt difference in the
number of TH positive cells in the midbrain and no difference in TH fiber density in the substantia nigra pars
reticulata was detectable (data not shown) [21]. In addition, all not affected w e a v e r mice displayed a normal
cerebellar histology without increase in apoptotic granule
cells, suggesting that these animals were true (+/+) wild
type littermates.
In all affected l u r c h e r mice, large labeled nuclei were
found in the Purkinje cell layer, suggesting that Purkinje
cells undergo apoptotic cell death (Fig. lc). In contrast to
affected w e a v e r mice the number of apoptotic cells in the
external granule cell layer did not differ significantly
from unaffected l u r c h e r littermates or C57BL controls
(Fig. 2a). However, an increased number of apoptotic
granule cell nuclei was found in the internal granule cell
layer (Fig. 2b). Outside the cerebellum, the number of
labeled cells at the former border of the subependymal
plate in the parietal cortex and in the ventral striatum appeared to be increased; however, these areas were not
quantified. In affected l u r c h e r mutants at P60 most of the
very few Purkinje cells that were still present were labeled, indicating that Purkinje cells were continuously
lost through apoptosis (Fig. ld). At this age, no apoptotic
granule cells were detectable in either affected l u r c h e r ,
healthy iittermates or C57BL mice.
Our data demonstrate that the loss of granule cells in
w e a v e r and of Purkinje cells in l u r c h e r mutant mice is
brought about by apoptosis. In w e a v e r mice, it is not entirely clear which cell type in the cerebellum is primarily
affected. Although early in vivo studies suggested that
defective outgrowth and maintenance of Bergmann glia
processes leads to a migration failure and subsequent
death of granule cells, cultured w e a v e r granule cells show
gene-dosage dependent loss of viability and defective
neurite elongation [8,15,22]. Implantation of w e a v e r
granule cells in wild type cerebellar cortex rescues neuronal differentiation, suggesting an intrinsic defect of
neuronal progenitor cells [5,6]. In our experiments, the
majority of apoptotic cells in the external granule cell
layer was found adjacent to the molecular layer, i.e. in an
early stage of the migration process. No apoptotic
Bergmann glia cells were identified.
In l u r c h e r mice, convincing evidence points to the
Purkinje cells as the primary target of the l u r c h e r gene
action [1,2]. This is consistent with our findings that
Purkinje cells undergo apoptotic cell death while granule
cells are probably secondarily affected. The hypothesis
that granule cell death in l u r c h e r mice is a target related
phenomenon secondary to Purkinje cell loss is supported
by the finding that the number of apoptotic granule cells
in the external granule cell layer is not significantly in-

112

u. Wallner et al. / Neuroscience Letters 200 (1995) 109-112

creased, w h i l e m o r e apoptotic granule cells are f o u n d in

the internal g r a n u l e cell layer.
W e h a v e s h o w n that apoptosis is the c o m m o n final
p a t h w a y o f cell death in the d e g e n e r a t i v e process induced
by the w e a v e r and the l u r c h e r genes. W h i l e this w o r k was
in preparation, H e i n t z et al., c o n f i r m i n g and e x t e n d i n g the
data reported here, also o b s e r v e d that the loss o f Purkinje
cells in l u r c h e r mutant m i c e e v o l v e s via apoptosis [10].
S i m i l a r results h a v e been p r e v i o u s l y reported in retinal
d e g e n e r a t i o n (rd) mice, w h e r e D N A f r a g m e n t a t i o n occurred during the p e r i o d o f p h o t o r e c e p t o r d e g e n e r a t i o n
[3]. A t least two o f these three mutations (lc and rd) are
g e n e t i c a l l y distinct in that they h a v e b e e n m a p p e d to different loci. A l t h o u g h different cell types c o u l d require
specific g e n e s to i n d u c e p r o g r a m m e d cell death, various
e n d o g e n o u s d e g e n e r a t i v e p r o c e s s e s m a y ultimately result
in a c o m m o n p a t h w a y o f apoptosis. The mutant m i c e
l u r c h e r and w e a v e r offer a u n i q u e opportunity to study in
v i v o the m e c h a n i s m s leading to inappropriate apoptotic
cell death.

[1] Caddy, K.W.T. and Biscoe, T.J., Preliminary observations on the

cerebellum in the mutant mouse Lurcher, Brain Res., 91 (1975)
276-280.
[2] Caddy, K.W.T. and Biscoe, T.J., Structural and quantitative studies on normal C3H and Lurcher mutant mouse, Phil. Trans. R.
Soc. London, 278 (1979) 167-201.
[3] Chang, G.-Q., Hao, Y. and Wong, F., Apoptosis: final common
pathway of photoreceptor death in rd, rds, and rhodopsin mutant
mice, Neuron, 11 (1993) 595--605.
[4] Dragunow, M., Faull, R.L.M., Lawlor, P., Beilharz, E.J., Singleton, K., Walker, E.B. and Mee, E., In situ evidence for DNA
fragmentation in Huntington's disease striatum and Alzheimer's
disease temporal lobes, NeuroReport, 6 (1995) 1053-1057.
[5] Gao, W.-Q and Hatten, M.E., Neuronal differentiation rescued by
implantation of weaver granule cell precursors into wild-type
cerebellar cortex, Science, 260 (1993) 367-369.
[6] Gao, W.-Q., Liu, X.L. and Hatten, M.E., The weaver gene encodes a nonautomatous signal for CNS neuronal differentiation,
Cell, 68 (1992) 841-851.
[7] Gavrieli, Y., Sherman, Y. and Ben-Sasson, S.A., Identification of
programmed cell death in situ via specific labeling of nuclear
DNA fragmentation, J. Cell Biol., 119 (1992) 493-501.

[8] Goldowitz, D., The weaver granuloprival phenotype is due to

intrinsic action of the mutant locus in granule cells: evidence from
homozygous weaver chimeras, Neuron, 2 (1989) 1565-1575.
[9] Migheli, A., Cavalla, P., Marino, S. and Schiffer, D., A study of
apoptosis in normal and pathologic nervous tissue after in situ
end-labeling of DNA strand breaks, J. Neuropathol. Exp. Neurol.,
53 (1994) 606-616.
[10] Norman, D.J., Feng, L., Cheng, S.S., Gubbay, J., Chan, E. and
Heintz, N., The lurcher gene induces apoptotic death in cerebellar
Purkinje cells, Development, 121 (1995) 1183-1193.
[11] Oppenheim, R.W., Schwartz, L.M. and Shatz, C.J., Neuronal
death, J. Neurobiol., 23 (1992) 1111-1351.
[12] Phillips, R.J.S., 'Lurcher', a new gene in linkage group XI of the
house mouse, J. Genet., 17 (1958) 35-42.
[13] Pittman, R.N., Mills, J.C., DiBenedetto, A.J., Hynicka, W.P. and
Wang, S., Neuronal cell death: searching for the smoking gun,
Curr. Opin. Neurobiol., 4 (1994) 87-94.
[14] Portera-Cailliau, C., Hedreen, J.C., Price, D.L. and Koliatsos,
V.E., Evidence for apoptotic cell death in Huntington disease and
exitotoxic animal models, J. Neurosci., 15 (1995) 3775-3787.
[15] Rakic, P. and Sidman, R.L., Weaver mutant mouse cerebellum:
defective neuronal migration secondary to abnormality of
Bergmann glia, Proc. Nat. Acad. Sci. USA, 70 (1973) 240-245.
[16] Rezai, Z. and Yoon, C.H., Abnormal rate of granule cell migration in the cerebellum of 'weaver' mutant mice, Dev. Biol., 29
(1972) 17-26.
[17] Roffier-Tarlov, S. and Graybiel, A.M., The postnatal development
of the dopamine-containing innervation of dorsal and ventral
striatum: effects of the weaver gene, J. Neurosci., 7 (1987) 23642372.
[18] Rubin, L.L., Gatchalian, C.L., Rimon, G. and Brooks, S.F., The
molecular mechanism of neuronal apoptosis, Curr. Opin. Neurobiol., 4 (1994) 696-702.
[19] Schwartzman, R.A. and Cildowski, J.A., Apoptosis: the biochemistry and molecular biology of programmed cell death, Endocr.
Rev., 14 (1993) 133-151.
[20] Sidman, R.L., Green, M.C. and Appel, S.H., Catalog of the Neurological Mutants of the Mouse, Harvard University Press, Cambridge, MA, 1965, pp. 66~7.
[21] Triarhou, L.C. and Ghetti, B., The dendritic dopamine projection
of the substantia nigra: phenotypic denominator of weaver gene
action in hetero- and homozygosity, Brain Res., 501 (1989) 373381.
[22] Willinger, M., Margolis, D.M. and Sidman, R.L., Neuronal differentiation in cultures of weaver mutant mouse cerebellum, J. Supramol. Struct. Cell Biochem., 17 (1981) 79-86.