- Email: [email protected]
Changes in the cerebellar cortex of hairless Rhino-J mice (hr-rh-j)
Neuroscience Letters 256 (1998) 13–16
Changes in the cerebellar cortex of hairless Rhino-J mice (hr-rh-j) N. Garcı´a-Atares a,1, I. San Jose a , b,1, R. Cabo b, J.A. Vega c, J. Represa a , b ,*
b
a Departamento de Anatomı´a Humana, Universidad de Valladolid, Valladolid, Spain Instituto de Biologı´a y Gene´tica Molecular, Universidad de Valladolid-CSIC, C/ Ramo´n y Cajal 7, 47007 Valladolid, Spain c Departamento de Morfologı´a y Biologı´a Celular, Universidad de Oviedo, Oviedo, Spain
Accepted 1 September 1998
Abstract A mutation in the hr gene is responsible for typical epithelium phenotype in hairless mice. As this gene is expressed at high levels not only in the skin but also in the brain, the aim of the study was to clarify its role in the central nervous system. We have analyzed by morphological and immunocytochemical methods (calbindin D-28k, phosphorylated and 200 kDa neurofilament protein) the cerebellum of a mutated mouse strain, the hairless (hr-rh-j) type carrying the homozygous hr gene rhino mutation. The cerebellar cortex was studied in young (3 months) and adult (9 months) wild type and mutated mice. No major structural change was found in any of the groups and neuronal density or neuronal arrangement were similar in mutated animals to their age-matched controls. Nevertheless there were changes in shape and size of the Purkinje neurons in the old mutated animals respect to their normal littermates, while the molecular and the granule cell layers were apparently invariable. Calbindin (CB) immunohistochemistry revealed a significant decrease in the expression of this protein in the Purkinje cells of the aged mutated mice. Immunohistochemistry for a neurofilament protein (NFP) showed a reduction of staining in all the cerebellar cortex layers in the older animals, which was much more evident in the (hr-rh-j) mutated mice. These results suggest that hr gene is involved in the structural maintenance of the mature cerebellar cortex, rather than in the development. Our findings may also be consistent with an accelerated aging of the central nervous system in rh-rh-j mice. 1998 Elsevier Science Ireland Ltd. All rights reserved
Keywords: Hairless hr-rh-j mice; Cerebellum; Structure
Hairless mice strains show defects in the keratinization of hair follicles that results in typical total hairless phenotype. In addition to alopecia, they develop age-related immunodeficiency [13,14], and increased incidence of skin neoplasias [7,14]. The hairless (hr) mutation in mice, responsible for the phenotype of these animals, has been recently characterized [5]. Consistently with the phenotype of the hairless mice the hr gene is expressed at high levels in the skin, but also in the brain, however the role of hr gene in the central nervous system remained unclear [5]. The homozygous hr/hr animals did not show apparent neurological diseases, suggesting absence of abnormalities in pathways and/or nuclei of the central nervous system. On the other hand, it has been * Corresponding author. Tel.: +34 83 423057; e-mail: [email protected] 1 These authors contributed equally to this paper.
demonstrated that hr gene in the brain showed a specific spatial-temporal pattern of expression that is regulated by thyroid hormone, all this is representative of genes that are important to nervous system development and/or maintenance. Nevertheless, the structure and/or histochemistry of central nervous in hr/hr mice have not been accurately studied yet. The present work was aimed to analyze the structure of the brain, focused on the cerebellar cortex, of young (3 months) and adult (9 months) homozygous hr-rh-j mice. We used histological techniques, and immunohistochemistry for calbindin D-28k (CB, [2]), and phosphorylated 200 kDa neurofilament proteins (NFP, [15]). Because of its heterogeneous morphology and characteristic arrangement, the cerebellar cortex represents a good model to address possible changes in the central nervous system due to the expression of the hr [4]. This has allowed us to study structural parameters such as neuronal density or
0304-3940/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00757- 5
14
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
CB and NFP immunoreactive nerve fibers. The parameters we analyzed here are sensitive to aging [2,15] and therefore results in young and adult hr-rh-j animals were compared with their normal littermates. The brain of homozygous male hr-rh-j mice aged 3 (n = 3) and 9 (n = 4) months, and of age-matched wildtype mice (n = 3 and n = 4, respectively) were used in this study. Mice of the mutated strain and normal littermates were obtained from the colony of the National Institute for Medical Research, Mill Hill, London and their genotype was determined by RT-PCR [5]. The animals were sacrificed by decapitation after deep ether anesthesia, the brain quickly removed and the cerebellum fixed in Bouins’s solution for 24 h, then dehydrated and embedded in paraffin. Serial frontal sections, 10 mm thick, were obtained from wild-type and mutant mice, mounted on gelatin-coated microscope slides and processed for histology or immunohistochemistry. The morphological structure of cerebellar cortex in hr-rh-j mice and their littermates was studied following a modified Giemsa’s method [9]; and 10 sections per animal, 50 mm apart, were analyzed. For immunohistochemistry, 10 sections per animal and antibody, 30 mm apart, were used as follows. In rehydrated sections the endogenous peroxidase activity and non-specific binding were blocked by standard procedures, then sections were incubated overnight (4ºC) with mouse monoclonal antibodies against: (1) calbindin D-28k (clone CL-300; Sigma, St. Louis, MO) used diluted 1:200 [2]; and (2) phosphorylated 200 kDa neurofilament proteins (clone RT-97, BoehringerMannheim, Germany) used diluted 1 mg/ml [15]. Sections were then washed and incubated with peroxidase-labeled sheep anti-mouse IgG (Amersham, UK) diluted 1:100 (1 h, room temperature) and the immunoreaction visualized using 3-3′ diaminobenzidine (Sigma, St. Louis, MO) as a chromogen. Control representative sections were processed as above, using a non-immune mouse serum instead of the primary antibodies, or omitting primary antibodies. Under these control conditions no immunostaining was observed.
Since we compared littermates, all these experiments were carried out in parallel and using the same solutions to eliminate variations on staining sensitivity. Quantitative image analysis was carried out using a MIP image analysis system (Centro de Ana´lisis de Ima´genes, University of Oviedo, Spain), to assess different parameters as follows: (1) The density of cerebellar neurons was evaluated in 10 sections per animal, 50 mm apart, stained according to the modified Giemsa’s method. The number of Purkinje neurons was counted in ten randomly selected 0.5 mm long portions per section, likewise the density of granule neurons was determined in ten randomly selected fields 0.5 mm2 each. These counts were performed by two researchers independently in a blind manner. (2) The number of neurons displaying CB immunoreactivity and the intensity of CB immunostaining were studied in ten sections per animal, 30 mm apart. The number of Purkinje neurons showing CB immunoreactivity was counted as above, then on the same neurons density of immunostaining was evaluated by microdensitometry [2] using arbitrary units of gray level ranging between 1 (black) and 256 (white). (3) The area occupied by NFP immunoreactive nerve fibers in hr-rhj and wild type mice was estimated in 10 sections per animal, 30 mm apart. Ten randomly selected fields of Purkinje’s and molecular neuron layers and the white matter were evaluated. The area occupied by NFP immunoreactive was expressed as a percentage (for details see [15]). Values of individual animals within each group were measurement means of the mentioned parameters. The average of each group were then determined from these individual means. The number of neurons, the number of CB immunoreactive Purkinje cells, the intensity of CB immunostaining and the area occupied by NFP were studied. Statistical differences between the above mentioned parameters were evaluated by analysis of variance (ANOVA) and covariance (ANCOVA), using the values of immunostaining in control sections as covariate. Data are the mean ± SEM. A probability at the 0.05 level was considered significant.
Table 1 Calbindin D-28k (CB) and 200 kDa neurofilament protein (NFP) immunoreactivity in the cerebellar cortex
(a) Number of nerve cell profiles Purkinje neurons (mm) Granule neurons (mm2) (b) % of CB D28k IR Purkinje neurons (mm) Intensity of CB D28k immunoreaction (c) % of area occupied by NFP immunoreactive nerve fibres (mm2) Layer of Punkinje cells Stratum moleculare White matter
Young wild-type (n = 3)
Young hr-rh-j (n = 3)
Old wild-type (n = 4)
Old hr-rh-j (n = 4)
84 ± 3 4435 ± 104 81 ± 4 93 ± 1.8
86 ± 5 4501 ± 98 85 ± 5 106 ± 1.1
89 ± 7 4287 ± 148 78 ± 7 99 ± 0.9
82 ± 6 4359 ± 119 80 ± 5 67 ± 2.3*
17.3 ± 0.5 16.8 ± 0.9 8.1 ± 0.4
16.7 ± 0.5 15.9 ± 1.1 5.3 ± 0.6**
12.9 ± 0.6 11.8 ± 0.9 4.9 ± 0.4**
7.1 ± 0.3 6.1 ± 0.5 3.1 ± 0.3*
(a) The number of different types of neurons (neuronal density); (b) the percentage of Purkinje neurons displaying CB immunoreactivity and intensity of CB immunostaining; (c) the density of the NFP immunoreactive nerve fibers. Numbers for wild type and hr-rh-j mutant mice of different age are shown. *P , 0.05 vs. all other groups. **P , 0.05 vs. young wild-type.
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
Morphological analysis showed a normal structure of the cerebellar cortex in the rh-rh-j mice of both young and old animals, without decrease in the number of Purkinje and granule neurons (Table 1) nor in the extent of these layers. Although, no major structural changes were found in the cerebellar cortex in any of the groups, there were changes in the shape and size of the Purkinje neurons in old mutated animals with respect to their age-matched controls (Fig. 1). Measurements of 200 Purkinje cellular profiles per animal were carried out in three mutated and three wild type mice, as it is explained elsewhere. There was a significant reduction of 27% (3.4 in the size of Purkinje neurons in old mutated mice.
15
Immunohistochemical study revealed specific changes in the expression Calbindin (CB) and neurofilament protein (NFP) in the cerebellar cortex of some of the studied groups. In the cerebellar cortex, CB immunoreactivity labeled the cytoplasm and dendritic arbor of the Purkinje neurons entering the molecular layer (Fig. 1). In young animals no differences in CB labelling were observed between rh-rh-j and controls (Fig. 1A,B; Table 1). With aging, there was a decrease in the intensity of both CB cytoplasmic immunostaining and CB density of positive dendritic profiles in the old mice, which was much more intense in old mutated mice than in their littermates (Fig. 1C–D and Table 1). Immunoreactivity for NFP was localized in the axon of basket neu-
Fig. 1. Immunohistochemical localization of calbindin D-28k in the cerebellum. Normal cerebellar cortex of wild type mice are shown in (A) the young animal and (C,E) the old animal, while the cerebellar cortex of hr-rh-j mutant mice are illustrated in (B) young animal and (D,F) old animal. Calbindin immunoreactivity appears highly concentrated in the Purkinje’s cells of both young wild type (A) and young mutant mice (B). The expression of calbindin is decreased in the Purkinje’s cells of aged mutant mice (arrows in D) respect to the Purkinje’s cells of wild type (arrow heads in C). (E) and (F) pictures show the differences in size and shape of Purkinje’s cells between old wild type (E) and old mutant mice (F). g, granule neurons layer; m, molecular layer; P, Purkinje’s neurons layer. Magnification: (A,B) 280×; (C,E) 420×; (D,F) 1000×. Fig. 2. Immunohistochemical localization of phosphorylated 200 kDa neurofilament protein (NFP) in the cerebellum. The microphotographs show the cerebellar cortex of young (A) and old (B,C) wild type mice, as well as young (D) and old (E,F) hr-rh-j mutant mice. NFP immunoreactivity appeared in basket fibers surrounding the soma of Purkinje’s neurons, in fibers entering the molecular layer and in axons of Purkinje’s neurons crossing the granule neuron layer. It can be observed an age-dependent decreased density of NFP, with the greatest reduction in old rh-rh-j mice. (C,F) Illustrate in detail the differences in NFP immunoreactivity in basket fibers (arrows). g, granule neurons layer; m, molecular layer; P, Purkinje’s neurons layer. Magnification: (A,B,D,E) 280×; (E,F) 720×.
16
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
rons (which encircles the soma of Purkinje cells) and also in a lesser extent within the axon of Purkinje neurons (Fig. 2). In the young rh-rh-j mice a reduction in the density of NFP immunoreactive nerve fibers was observed (Fig. 2A,D and Table 1), particularly in those involving the Purkinje’s cell axons. With aging there was a decrease in the density of NFP immunoreactive structures, that was much more evident in old hr-rh-j mice with respect to their age-matched controls (Fig. 2B,C,E,F and Table 1). Although the gene responsible for the hairless phenotype is expressed at high levels in the brain [5], we observed no major changes in neuronal density and/or neuronal arrangements. These structural parameters are considered to be related to the ability of the brain to analyze and store information [6]. No large structural changes in the cerebellar cortex have been reported during the first year of normal life [6]. Our results revealed that no evident changes in the neuronal density of the cerebellar cortex were encountered between hr-rh-j mice and the corresponding controls at any age. The absence of gross abnormalities in the cerebellar cortex of young and old animals carrying the hr gene mutation, suggest that this gene is not essential in the embryonic development nor is required for neuronal maintenance. In old rh-rh-j mice there was a reduction in the size as well as modifications in the morphology of the Purkinje’s neurons respect to their age-matched controls, this could be regarded as an evidence of accelerated aging in rh-rh-j mice [2]). Our findings confirm that CB is highly concentrated in the Purkinje’s cells of normal cerebellum [3]. Moreover, the density of neurons expressing this calcium-binding protein remained unchanged in all groups of animals. We have found a non-significant decrease in CB immunostaining in old wild type animals respect to young ones. However, the expression of CB appeared significantly decreased in the Purkinje’s cells of aged mutant mice with respect to their age-matched controls. Expression of neuronal CB has been found modified in some physiological and pathological conditions of the nervous system [1]. For instance, both the intensity of immunostaining and the number of immunoreactive neurons remained elevated postnatally [8] and decrease with aging [2,8]. In addition, reduced CB has been reported in spino-cerebellar degeneration [10] and in genetically epilepsy-prone rats [12]. Thus, the remarkable decrease of CB observed in old hr-rh-j mice respect to their littermates may be suggesting an accelerate aging of the cerebellum in these mutant animals. The nerve fibers detected by silver-gold methods or immunohistochemistry for NFP reflects the size and complexity of neuronal interconnections in the cerebellum. In this study we observed age-dependent decreased density of NFP with the greatest reduction in old rh-rh-j mice, suggesting that age-dependent physiological impairment in NFP expression is accelerated in mutated mice. This effect could be explained by a more reduced synthesis or a specific phosphorylation defect of NFP in old rh-rh-j mice [11].
Special thanks are due to B. Cachon-Gonza´lez for kindly providing us with homozygous hr-rh-j and wild type mice. We wish to thank C. Valero for technical assistance and C. Layburn for useful discussion and criticisms. This work was supported in part by grants from Fondo de Investigaciones Sanitarias and Junta de Castilla y Leo´n to J.A Vega and J. Represa. [1] Airaksinen, M.S., Eilers, J., Garaschuk, O., Thoenen, H., Nonnerth, A. and Meyer, M., Ataxia and altered dendritic calcium signaling in mice carrying a targeted null mutation of the calbindin D28k gene, Proc. Natl. Acad. Sci. USA, 94 (1997) 1488–1493. [2] Amenta, F., Cavalotti, D., Del Valle, M.E., Mancini, M., Sabbatini, M., Torres, J.M. and Vega, J.A., Calbindin D-28k immunoreactivity in the rat cerebellar cortex: age-related changes, Neurosci. Lett., 178 (1994) 131–134. [3] Baimbridge, K.G., Celio, M.R. and Rogers, J., Calcium-binding proteins in the nervous system, Trends Neurosci., 15 (1992) 303–308. [4] Bakalian, A., Delhaye-Bouchaud, N. and Mariani, J., Quantitatve analysis of Purkinje cells and the granule cells populations in the cerebellum of nude mice, J. Neurogenet., 9 (1995) 207– 218. [5] Cacho´n-Gonza´lez, M.B., Fenner, S., Coffin, J.M., Moran, C., Best, S. and Stoye, J.P., Structure and expression of the hairless gene of mice, Proc. Natl. Acad. Sci. USA, 91 (1994) 7717– 7721. [6] Flood, D.G. and Coleman, P.D., Neuron number and sizes in ageing brain: comparisons of human, monkey and rodent data, Neurobiol. Aging, 9 (1988) 453–463. [7] Gruijil, F.R. and Forbes, P.D., UV-induced skin cancer in a hairless mouse model, Bioassays, 17 (1995) 651–660. [8] Iacopino, A.M., Rhoten, W.B. and Christakos, S., Calcium-binding protein (calbindin-D-28k) gene expression in the developing and aging mouse cerebellum, Mol. Brain Res., 8 (1990) 283– 290. [9] In˜iguez, C., Gayoso, M.J. and Carreres, J., A versatile and simple method for staining nervous tissue using Giemsa dye, J. Neurosci. Meth., 13 (1985) 77–86. [10] Ishikawa, K., Mizusawas, H., Fujita, T., Ohkoshi, N., Doi, M., Komatsuzaki, Y., Iwamoto, H., Ogata, T. and Shoji, S., Calbindin-D 28k immunoreactivity in the cerebellum of spinocerebellar degeneration, J. Neurol. Sci., 129 (1995) 179–185. [11] Marc, C., Clavel, M.-C. and Rabie, A., Non-phosphorylated and phosphorylated neurofilaments in the cerebellum of the rat: an immunohistochemical study using monoclonal antibodies: development in normal and thyroid-deficient animals, Dev. Brain Res., 26 (1986) 249–260. [12] Montpied, P., Winsky, L., Dailey, J.W., Jobe, P.C. and Jacobowitz, D.M., Alterations in levels of expression of brain calbindin D-28k and calretinin mRNA in genetically epilepsyprone rats, Epilepsia, 36 (1995) 911–921. [13] Morrisey, P.J., Parkinson, D.R., Schwarz, R.S. and Waksal, S.D., Immunologic abnormalities in HRS/J mice. I. Specific deficit in T lymphocyte helper function in a mutant mouse, J. Immunol., 125 (1980) 1558–1562. [14] Poland, A., Palen, D. and Glover, E., Tumor promotion by TCDD in skin of HRS/J hairless mice, Nature, 300 (1982) 271–273. [15] Vega, J.A., Del Valle, M. and Amenta, F., Expression of neurofilament proteins in the rat cerebellar cortex as a function of age: an immunohistochemical study, Mech. Ageing Dev., 73 (1994) 9–16.
Changes in the cerebellar cortex of hairless Rhino-J mice (hr-rh-j) N. Garcı´a-Atares a,1, I. San Jose a , b,1, R. Cabo b, J.A. Vega c, J. Represa a , b ,*
b
a Departamento de Anatomı´a Humana, Universidad de Valladolid, Valladolid, Spain Instituto de Biologı´a y Gene´tica Molecular, Universidad de Valladolid-CSIC, C/ Ramo´n y Cajal 7, 47007 Valladolid, Spain c Departamento de Morfologı´a y Biologı´a Celular, Universidad de Oviedo, Oviedo, Spain
Accepted 1 September 1998
Abstract A mutation in the hr gene is responsible for typical epithelium phenotype in hairless mice. As this gene is expressed at high levels not only in the skin but also in the brain, the aim of the study was to clarify its role in the central nervous system. We have analyzed by morphological and immunocytochemical methods (calbindin D-28k, phosphorylated and 200 kDa neurofilament protein) the cerebellum of a mutated mouse strain, the hairless (hr-rh-j) type carrying the homozygous hr gene rhino mutation. The cerebellar cortex was studied in young (3 months) and adult (9 months) wild type and mutated mice. No major structural change was found in any of the groups and neuronal density or neuronal arrangement were similar in mutated animals to their age-matched controls. Nevertheless there were changes in shape and size of the Purkinje neurons in the old mutated animals respect to their normal littermates, while the molecular and the granule cell layers were apparently invariable. Calbindin (CB) immunohistochemistry revealed a significant decrease in the expression of this protein in the Purkinje cells of the aged mutated mice. Immunohistochemistry for a neurofilament protein (NFP) showed a reduction of staining in all the cerebellar cortex layers in the older animals, which was much more evident in the (hr-rh-j) mutated mice. These results suggest that hr gene is involved in the structural maintenance of the mature cerebellar cortex, rather than in the development. Our findings may also be consistent with an accelerated aging of the central nervous system in rh-rh-j mice. 1998 Elsevier Science Ireland Ltd. All rights reserved
Keywords: Hairless hr-rh-j mice; Cerebellum; Structure
Hairless mice strains show defects in the keratinization of hair follicles that results in typical total hairless phenotype. In addition to alopecia, they develop age-related immunodeficiency [13,14], and increased incidence of skin neoplasias [7,14]. The hairless (hr) mutation in mice, responsible for the phenotype of these animals, has been recently characterized [5]. Consistently with the phenotype of the hairless mice the hr gene is expressed at high levels in the skin, but also in the brain, however the role of hr gene in the central nervous system remained unclear [5]. The homozygous hr/hr animals did not show apparent neurological diseases, suggesting absence of abnormalities in pathways and/or nuclei of the central nervous system. On the other hand, it has been * Corresponding author. Tel.: +34 83 423057; e-mail: [email protected] 1 These authors contributed equally to this paper.
demonstrated that hr gene in the brain showed a specific spatial-temporal pattern of expression that is regulated by thyroid hormone, all this is representative of genes that are important to nervous system development and/or maintenance. Nevertheless, the structure and/or histochemistry of central nervous in hr/hr mice have not been accurately studied yet. The present work was aimed to analyze the structure of the brain, focused on the cerebellar cortex, of young (3 months) and adult (9 months) homozygous hr-rh-j mice. We used histological techniques, and immunohistochemistry for calbindin D-28k (CB, [2]), and phosphorylated 200 kDa neurofilament proteins (NFP, [15]). Because of its heterogeneous morphology and characteristic arrangement, the cerebellar cortex represents a good model to address possible changes in the central nervous system due to the expression of the hr [4]. This has allowed us to study structural parameters such as neuronal density or
0304-3940/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00757- 5
14
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
CB and NFP immunoreactive nerve fibers. The parameters we analyzed here are sensitive to aging [2,15] and therefore results in young and adult hr-rh-j animals were compared with their normal littermates. The brain of homozygous male hr-rh-j mice aged 3 (n = 3) and 9 (n = 4) months, and of age-matched wildtype mice (n = 3 and n = 4, respectively) were used in this study. Mice of the mutated strain and normal littermates were obtained from the colony of the National Institute for Medical Research, Mill Hill, London and their genotype was determined by RT-PCR [5]. The animals were sacrificed by decapitation after deep ether anesthesia, the brain quickly removed and the cerebellum fixed in Bouins’s solution for 24 h, then dehydrated and embedded in paraffin. Serial frontal sections, 10 mm thick, were obtained from wild-type and mutant mice, mounted on gelatin-coated microscope slides and processed for histology or immunohistochemistry. The morphological structure of cerebellar cortex in hr-rh-j mice and their littermates was studied following a modified Giemsa’s method [9]; and 10 sections per animal, 50 mm apart, were analyzed. For immunohistochemistry, 10 sections per animal and antibody, 30 mm apart, were used as follows. In rehydrated sections the endogenous peroxidase activity and non-specific binding were blocked by standard procedures, then sections were incubated overnight (4ºC) with mouse monoclonal antibodies against: (1) calbindin D-28k (clone CL-300; Sigma, St. Louis, MO) used diluted 1:200 [2]; and (2) phosphorylated 200 kDa neurofilament proteins (clone RT-97, BoehringerMannheim, Germany) used diluted 1 mg/ml [15]. Sections were then washed and incubated with peroxidase-labeled sheep anti-mouse IgG (Amersham, UK) diluted 1:100 (1 h, room temperature) and the immunoreaction visualized using 3-3′ diaminobenzidine (Sigma, St. Louis, MO) as a chromogen. Control representative sections were processed as above, using a non-immune mouse serum instead of the primary antibodies, or omitting primary antibodies. Under these control conditions no immunostaining was observed.
Since we compared littermates, all these experiments were carried out in parallel and using the same solutions to eliminate variations on staining sensitivity. Quantitative image analysis was carried out using a MIP image analysis system (Centro de Ana´lisis de Ima´genes, University of Oviedo, Spain), to assess different parameters as follows: (1) The density of cerebellar neurons was evaluated in 10 sections per animal, 50 mm apart, stained according to the modified Giemsa’s method. The number of Purkinje neurons was counted in ten randomly selected 0.5 mm long portions per section, likewise the density of granule neurons was determined in ten randomly selected fields 0.5 mm2 each. These counts were performed by two researchers independently in a blind manner. (2) The number of neurons displaying CB immunoreactivity and the intensity of CB immunostaining were studied in ten sections per animal, 30 mm apart. The number of Purkinje neurons showing CB immunoreactivity was counted as above, then on the same neurons density of immunostaining was evaluated by microdensitometry [2] using arbitrary units of gray level ranging between 1 (black) and 256 (white). (3) The area occupied by NFP immunoreactive nerve fibers in hr-rhj and wild type mice was estimated in 10 sections per animal, 30 mm apart. Ten randomly selected fields of Purkinje’s and molecular neuron layers and the white matter were evaluated. The area occupied by NFP immunoreactive was expressed as a percentage (for details see [15]). Values of individual animals within each group were measurement means of the mentioned parameters. The average of each group were then determined from these individual means. The number of neurons, the number of CB immunoreactive Purkinje cells, the intensity of CB immunostaining and the area occupied by NFP were studied. Statistical differences between the above mentioned parameters were evaluated by analysis of variance (ANOVA) and covariance (ANCOVA), using the values of immunostaining in control sections as covariate. Data are the mean ± SEM. A probability at the 0.05 level was considered significant.
Table 1 Calbindin D-28k (CB) and 200 kDa neurofilament protein (NFP) immunoreactivity in the cerebellar cortex
(a) Number of nerve cell profiles Purkinje neurons (mm) Granule neurons (mm2) (b) % of CB D28k IR Purkinje neurons (mm) Intensity of CB D28k immunoreaction (c) % of area occupied by NFP immunoreactive nerve fibres (mm2) Layer of Punkinje cells Stratum moleculare White matter
Young wild-type (n = 3)
Young hr-rh-j (n = 3)
Old wild-type (n = 4)
Old hr-rh-j (n = 4)
84 ± 3 4435 ± 104 81 ± 4 93 ± 1.8
86 ± 5 4501 ± 98 85 ± 5 106 ± 1.1
89 ± 7 4287 ± 148 78 ± 7 99 ± 0.9
82 ± 6 4359 ± 119 80 ± 5 67 ± 2.3*
17.3 ± 0.5 16.8 ± 0.9 8.1 ± 0.4
16.7 ± 0.5 15.9 ± 1.1 5.3 ± 0.6**
12.9 ± 0.6 11.8 ± 0.9 4.9 ± 0.4**
7.1 ± 0.3 6.1 ± 0.5 3.1 ± 0.3*
(a) The number of different types of neurons (neuronal density); (b) the percentage of Purkinje neurons displaying CB immunoreactivity and intensity of CB immunostaining; (c) the density of the NFP immunoreactive nerve fibers. Numbers for wild type and hr-rh-j mutant mice of different age are shown. *P , 0.05 vs. all other groups. **P , 0.05 vs. young wild-type.
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
Morphological analysis showed a normal structure of the cerebellar cortex in the rh-rh-j mice of both young and old animals, without decrease in the number of Purkinje and granule neurons (Table 1) nor in the extent of these layers. Although, no major structural changes were found in the cerebellar cortex in any of the groups, there were changes in the shape and size of the Purkinje neurons in old mutated animals with respect to their age-matched controls (Fig. 1). Measurements of 200 Purkinje cellular profiles per animal were carried out in three mutated and three wild type mice, as it is explained elsewhere. There was a significant reduction of 27% (3.4 in the size of Purkinje neurons in old mutated mice.
15
Immunohistochemical study revealed specific changes in the expression Calbindin (CB) and neurofilament protein (NFP) in the cerebellar cortex of some of the studied groups. In the cerebellar cortex, CB immunoreactivity labeled the cytoplasm and dendritic arbor of the Purkinje neurons entering the molecular layer (Fig. 1). In young animals no differences in CB labelling were observed between rh-rh-j and controls (Fig. 1A,B; Table 1). With aging, there was a decrease in the intensity of both CB cytoplasmic immunostaining and CB density of positive dendritic profiles in the old mice, which was much more intense in old mutated mice than in their littermates (Fig. 1C–D and Table 1). Immunoreactivity for NFP was localized in the axon of basket neu-
Fig. 1. Immunohistochemical localization of calbindin D-28k in the cerebellum. Normal cerebellar cortex of wild type mice are shown in (A) the young animal and (C,E) the old animal, while the cerebellar cortex of hr-rh-j mutant mice are illustrated in (B) young animal and (D,F) old animal. Calbindin immunoreactivity appears highly concentrated in the Purkinje’s cells of both young wild type (A) and young mutant mice (B). The expression of calbindin is decreased in the Purkinje’s cells of aged mutant mice (arrows in D) respect to the Purkinje’s cells of wild type (arrow heads in C). (E) and (F) pictures show the differences in size and shape of Purkinje’s cells between old wild type (E) and old mutant mice (F). g, granule neurons layer; m, molecular layer; P, Purkinje’s neurons layer. Magnification: (A,B) 280×; (C,E) 420×; (D,F) 1000×. Fig. 2. Immunohistochemical localization of phosphorylated 200 kDa neurofilament protein (NFP) in the cerebellum. The microphotographs show the cerebellar cortex of young (A) and old (B,C) wild type mice, as well as young (D) and old (E,F) hr-rh-j mutant mice. NFP immunoreactivity appeared in basket fibers surrounding the soma of Purkinje’s neurons, in fibers entering the molecular layer and in axons of Purkinje’s neurons crossing the granule neuron layer. It can be observed an age-dependent decreased density of NFP, with the greatest reduction in old rh-rh-j mice. (C,F) Illustrate in detail the differences in NFP immunoreactivity in basket fibers (arrows). g, granule neurons layer; m, molecular layer; P, Purkinje’s neurons layer. Magnification: (A,B,D,E) 280×; (E,F) 720×.
16
N. Garcı´a-Atares et al. / Neuroscience Letters 256 (1998) 13–16
rons (which encircles the soma of Purkinje cells) and also in a lesser extent within the axon of Purkinje neurons (Fig. 2). In the young rh-rh-j mice a reduction in the density of NFP immunoreactive nerve fibers was observed (Fig. 2A,D and Table 1), particularly in those involving the Purkinje’s cell axons. With aging there was a decrease in the density of NFP immunoreactive structures, that was much more evident in old hr-rh-j mice with respect to their age-matched controls (Fig. 2B,C,E,F and Table 1). Although the gene responsible for the hairless phenotype is expressed at high levels in the brain [5], we observed no major changes in neuronal density and/or neuronal arrangements. These structural parameters are considered to be related to the ability of the brain to analyze and store information [6]. No large structural changes in the cerebellar cortex have been reported during the first year of normal life [6]. Our results revealed that no evident changes in the neuronal density of the cerebellar cortex were encountered between hr-rh-j mice and the corresponding controls at any age. The absence of gross abnormalities in the cerebellar cortex of young and old animals carrying the hr gene mutation, suggest that this gene is not essential in the embryonic development nor is required for neuronal maintenance. In old rh-rh-j mice there was a reduction in the size as well as modifications in the morphology of the Purkinje’s neurons respect to their age-matched controls, this could be regarded as an evidence of accelerated aging in rh-rh-j mice [2]). Our findings confirm that CB is highly concentrated in the Purkinje’s cells of normal cerebellum [3]. Moreover, the density of neurons expressing this calcium-binding protein remained unchanged in all groups of animals. We have found a non-significant decrease in CB immunostaining in old wild type animals respect to young ones. However, the expression of CB appeared significantly decreased in the Purkinje’s cells of aged mutant mice with respect to their age-matched controls. Expression of neuronal CB has been found modified in some physiological and pathological conditions of the nervous system [1]. For instance, both the intensity of immunostaining and the number of immunoreactive neurons remained elevated postnatally [8] and decrease with aging [2,8]. In addition, reduced CB has been reported in spino-cerebellar degeneration [10] and in genetically epilepsy-prone rats [12]. Thus, the remarkable decrease of CB observed in old hr-rh-j mice respect to their littermates may be suggesting an accelerate aging of the cerebellum in these mutant animals. The nerve fibers detected by silver-gold methods or immunohistochemistry for NFP reflects the size and complexity of neuronal interconnections in the cerebellum. In this study we observed age-dependent decreased density of NFP with the greatest reduction in old rh-rh-j mice, suggesting that age-dependent physiological impairment in NFP expression is accelerated in mutated mice. This effect could be explained by a more reduced synthesis or a specific phosphorylation defect of NFP in old rh-rh-j mice [11].
Special thanks are due to B. Cachon-Gonza´lez for kindly providing us with homozygous hr-rh-j and wild type mice. We wish to thank C. Valero for technical assistance and C. Layburn for useful discussion and criticisms. This work was supported in part by grants from Fondo de Investigaciones Sanitarias and Junta de Castilla y Leo´n to J.A Vega and J. Represa. [1] Airaksinen, M.S., Eilers, J., Garaschuk, O., Thoenen, H., Nonnerth, A. and Meyer, M., Ataxia and altered dendritic calcium signaling in mice carrying a targeted null mutation of the calbindin D28k gene, Proc. Natl. Acad. Sci. USA, 94 (1997) 1488–1493. [2] Amenta, F., Cavalotti, D., Del Valle, M.E., Mancini, M., Sabbatini, M., Torres, J.M. and Vega, J.A., Calbindin D-28k immunoreactivity in the rat cerebellar cortex: age-related changes, Neurosci. Lett., 178 (1994) 131–134. [3] Baimbridge, K.G., Celio, M.R. and Rogers, J., Calcium-binding proteins in the nervous system, Trends Neurosci., 15 (1992) 303–308. [4] Bakalian, A., Delhaye-Bouchaud, N. and Mariani, J., Quantitatve analysis of Purkinje cells and the granule cells populations in the cerebellum of nude mice, J. Neurogenet., 9 (1995) 207– 218. [5] Cacho´n-Gonza´lez, M.B., Fenner, S., Coffin, J.M., Moran, C., Best, S. and Stoye, J.P., Structure and expression of the hairless gene of mice, Proc. Natl. Acad. Sci. USA, 91 (1994) 7717– 7721. [6] Flood, D.G. and Coleman, P.D., Neuron number and sizes in ageing brain: comparisons of human, monkey and rodent data, Neurobiol. Aging, 9 (1988) 453–463. [7] Gruijil, F.R. and Forbes, P.D., UV-induced skin cancer in a hairless mouse model, Bioassays, 17 (1995) 651–660. [8] Iacopino, A.M., Rhoten, W.B. and Christakos, S., Calcium-binding protein (calbindin-D-28k) gene expression in the developing and aging mouse cerebellum, Mol. Brain Res., 8 (1990) 283– 290. [9] In˜iguez, C., Gayoso, M.J. and Carreres, J., A versatile and simple method for staining nervous tissue using Giemsa dye, J. Neurosci. Meth., 13 (1985) 77–86. [10] Ishikawa, K., Mizusawas, H., Fujita, T., Ohkoshi, N., Doi, M., Komatsuzaki, Y., Iwamoto, H., Ogata, T. and Shoji, S., Calbindin-D 28k immunoreactivity in the cerebellum of spinocerebellar degeneration, J. Neurol. Sci., 129 (1995) 179–185. [11] Marc, C., Clavel, M.-C. and Rabie, A., Non-phosphorylated and phosphorylated neurofilaments in the cerebellum of the rat: an immunohistochemical study using monoclonal antibodies: development in normal and thyroid-deficient animals, Dev. Brain Res., 26 (1986) 249–260. [12] Montpied, P., Winsky, L., Dailey, J.W., Jobe, P.C. and Jacobowitz, D.M., Alterations in levels of expression of brain calbindin D-28k and calretinin mRNA in genetically epilepsyprone rats, Epilepsia, 36 (1995) 911–921. [13] Morrisey, P.J., Parkinson, D.R., Schwarz, R.S. and Waksal, S.D., Immunologic abnormalities in HRS/J mice. I. Specific deficit in T lymphocyte helper function in a mutant mouse, J. Immunol., 125 (1980) 1558–1562. [14] Poland, A., Palen, D. and Glover, E., Tumor promotion by TCDD in skin of HRS/J hairless mice, Nature, 300 (1982) 271–273. [15] Vega, J.A., Del Valle, M. and Amenta, F., Expression of neurofilament proteins in the rat cerebellar cortex as a function of age: an immunohistochemical study, Mech. Ageing Dev., 73 (1994) 9–16.