Original Paper
Neuroendocrinology 1992;55:695-707
Department o f Anatomy and Neuroscience Training Program, Tulane University School o f Medicine, New Orleans, La., U S A ; Department o f Anatomy, University o f Ghana Medical School, Accra, G hana; Baker Medical Institute, Prahan, Australia
Key Words Glucocorticoids Receptors Immunocytochemistry Cerebellum Brain Stem
Postnatal Development of Corticosteroid Receptor Immunoreactivity in the Rat Cerebellum and Brain Stem
Abstract The postnatal development o f corticosteroid receptor immunoreactivity in the rat cerebellum and related brainstem nuclei was studied using a type I receptor antiserum, M IN R E C 4 , and a type II receptor monoclonal antibody, B U G R 2 . Type 1 receptor immunoreactive (ir) Purkinje cells were first observed at postnatal day 5 (P5), and increased to adult levels by P20.Type I-ir cells, presumably migrating granule cells, were observed in the developing molecular layer o f the cerebellum at P5. By P30, the density o f type I-ir cells in the definitive molecular and granular layers was still less than adult levels. In contrast, type I I-ir Purkinje cells were first observed at PI 5 and increased to adult levels by P20. No type I I-ir cells were observed in the proliferative and migratory zones o f the molecular layer. By P30, the density o f type I I-ir cells in the molecular and granular layers was far less than adult levels. In the deep cerebellar nuclei and most brain stem nuclei type I-ir was observed at P5 and developed to adult levels by P30. Type I I-ir was observed in the deep cerebel lar nuclei, red and medial vestibular nuclei by P I 5. The pontine and inferior olivary nuclei showed type I I-ir cells by P10. Type I I-ir in these regions de veloped to adult levels by P30. The earlier development o f type I-ir suggests that the type I receptor may mediate the actions o f corticosteroids in the cere bellum and related brain stem nuclei during early postnatal life. Both type I and type II corticosteroid receptors may mediate corticosteroid effects in the cerebellum and brain stem nuclei during late postnatal life and adulthood.
Glucocorticoids have diverse effects on the developing nervous system [I, 2]. Neuronal and glial proliferation and differentiation are regulated by glucocorticoids. While glucocorticoids are required for normal C N S de velopment at certain critical periods, high levels have
deleterious effects. This is especially evident in germinal zones o f the developing brain, e.g. cerebellum, dentate gyrus and olfactory bulb, which undergo considerable de velopment during the postnatal period [1, 3, 4], Excess levels o f glucocorticoids inhibit D N A synthesis, and neu-
Tliis work was supported by N S24I48 (R .E .H ). Special thanks to R Harrison for supplying B U G R 2 antibody and P3 AgX-653 myeloma cell supernatant.
Richard E. Harlan, Ph i) Dept, o f Anatomy, Tulane University Medical School 1430 Tulane Avenue New Orleans. L A 70112 (U SA )
Received: July I I . 1991 Accepted after revision: September 12, 1991
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Aaron Lawson“ h Rexford S. Ahima“ Zygmunt Krozowskic Richard E. Harlan“ d
696
type Il-ir have been localized in the Purkinje and granular layers o f the cerebellar cortex, deep cerebellar nuclei and red, pontine, inferior olivary and medial vestibular nuclei [15, 16]. We report, based on immunocytochemical local ization o f corticosteroid receptors, that there are age-re lated regional differences in the distribution o f these re ceptors during postnatal development.
Materials and Methods Animals and Tissue Preparation Four pregnant Sprague-Dawley rats were used for the study. Each littered 12 pups which were kept with their dams and were not handled until sacrifice. Six pups each were sacrificed at postnatal day zero (P0; day o f birth), P5, P I0, P15, P20, P25 and P30. Three adult male rats (3 months old) were used for comparison. PO and P5 pups were anaesthetized by hypothermia and P I0-P 30 pups by methoxyflurane (Pitman-Moore) inhalation. The animals were per fused transcardially with phosphate buffered saline (PBS), pH 7.2, for 5 min followed by 3% phosphate-buffered paraformaldehyde for 8 min. Brains were dissected out. postfixed in the same fixative for 2 h and cryoprotected until they sank in 30% sucrose. They were then frozen with dry ice and stored wrapped in aluminium foil at -7 0 ° C until sectioning.
Antibodies M IN R F .C 4 antiserum, raised against a fusion protein G T M R 4 comprising a 167 amino acid fragment o f the N-terminal domain of the human type 1 receptor and glutathione-S-transferase (GST), G T M R 4 fusion protein and G S T antiserum were supplied by Krozowski’s group [13, 19, 20]. This antiserum has the same specificity as the previously characterized M IN R E C 2 antiserum [22] and has been used in studies in the rat C N S [13, 19, 20], B U G R 2 mouse antirat liver type 11 receptor monoclonal antibody and nonimmune P3 AgX-653 myeloma cell medium were supplied by Harrison’s group [21],
Immunocytochemistry Coronal sections, 30-pm thick, were cut on a cryostat and thawmounted onto chrom-alum-coated slides. They were processed for immunocytochemistry using Vectastain immunoperoxidase kits (Vector) and a modification o f a protocol by Hancock [23]. M IN R E C 4 antiserum (against type I receptor) was used at dilutions o f 1:500, 1: 1.000, 1:2,000 and 1:4,000. Adjacent control sections were incubated with either M IN R E C 4 antiserum (1:1,000) preabsorbed with 10 \iM G T M R 4 fusion protein or 10 u M G S T or 1:500, 1:1,000 and 1:2,000 dilution o f G S T antiserum. B U G R 2 antibody (against type II receptor) was used at dilutions o f 1:250, 1:500, 1:1,000 and 1:2,000. Adjacent control sections were incubated with either nor mal mouse IgG or nonimmune P3 AgX-653 myeloma cell medium. The chromogen used was 25 mg% 3,3'-diaminobenzidine tetrahydrochloride (D A B ) and 1.25% nickel ammonium sulfate, which produced a dark blue reaction product. Sections were dehydrated in ethanol, defatted with Histoclear (National Diagnostics) and coverslipped with Permount (Fisher Scientific). Sections adjacent to those selected for microscopic analysis were stained with 0.5% cresyl violet to facilitate delineation o f neuronal groups.
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
rona! and glial proliferation leading to reduced brain weight [ 1,4-6], Two different effects o f excess glucocorti coids on rat cerebellar development have been described [ 1, 4], Treatment o f rats with high doses o f cortisol be tween postnatal days 1 and 4 (P1-P4) inhibits cell prolife ration and alters the pattern o f foliation resulting in small and abnormally shaped cerebelli in adulthood. This pro cess is, however, reversible if treatment is terminated early. Late postnatal treatment with cortisol (P7-P18), produces greater inhibition o f neuronal proliferation and differentiation and smaller cerebelli compared to early treatment, and no effect on foliation. Neonatal treatment with cortisol delays the maturation o f motor behaviors, such as swimming [7] and motor coordination [8]. C o n versely, early adrenalectomy improves subsequent maze learning behavior [9], It is possible these effects o f gluco corticoids on motor function are mediated via a direct action on neuronal development and function in motorcoordinating centers, including the cerebellum. The actions o f glucocorticoids in the postnatal rat brain are mediated via two types o f corticosteroid receptors [10-12], A type I receptor, with the same characteristics as the classical mineralocorticoid receptor, binds with high affinity to corticosterone, the major glucocorticoid, and to aldosterone. Ligand-binding studies have indicated that this receptor in the forebrain is restricted to limbic regions [10, 12], A type II receptor, or classical glucocorticoid re ceptor with a lower affinity and higher capacity for cortico sterone, has been localized with binding studies in the lim bic brain o f postnatal rats [10]. Immunocytochemical studies [ I I , 13] have revealed a widespread distribution of type 11 receptors. We [ 13] have recently described the post natal development o f types I and II immunoreactive (ir) cells in the rat hippocampus. The patterns o f corticosteroid receptor binding [10, 12] and immunocytochemical local ization [11, 13] vary with age. This may be related to the roles o f these receptors in mediating the modulatory ef fects o f glucocorticoids on critical aspects o f C N S develop ment [10-12], Corticosteroid receptors have been local ized in the adult rat cerebellum using binding studies [14], in situ hybridization [15, 16] and immunocytochemistry [17-19], Presumably these receptors may mediate some of the effects ofglucocorticoids on postnatal cerebellar devel opment [1-4]. We have used an antiserum against the type I receptor [13, 19, 20] and an anti-type II receptor monoclonal antibody [ 18,21 ] to examine the development o f corticosteroid receptor-like immunoreactivity in the postnatal rat cerebellum and its brainstem connection. In the adult rat brain, moderate to high densities o f cells ex pressing type I and type II receptor m R N A or showing
Microscopy Sections were examined with a Nikon Optiphot microscope equipped with a Nikon F X 3 5A camera and photographs taken with Technical Pan (Kodak) film. Three coronal sections o f the cerebellum stained for type I and Il-ir from three pups per age were selected, matched and coded by one investigator. The levels were (i) a rostral section showing vermal lobules 1-6; (ii) a caudal section showing vermal lobules 7-10, and (in) a section between these two showing all deep cerebellar nuclei. Sections o f some brainstem nuclei with cerebellar connections, i.e. red, pontine, inferior olivary and medial vestibular nuclei were also analyzed. Sections stained for type I and Il-ir through the thickness o f these nuclei and separated by 90 pm (i.e. every third section) from three pups per age were selected, matched and coded by one investigator. The sections were analyzed blindly by two investigators observ ing the following parameters: (i) distribution o f immunoreactive cells; (ii) intracellular location o f receptors; (iii) relative intensity o f immunostaining - this was scored qualitatively as ‘strong’ as in fig ure 4A or ‘weak’ as in figure 8A. The distribution o f Purkinje cells in representative sections o f rostral and caudal cerebelli was plotted using a Bausch and Lomb projection microscope.
Specificity o f Antibodies The specificity o f M IN R E C 4 antiserum and B U G R 2 monoclonal antibody for type I and II receptors have been demonstrated previously [13, 18, 19, 24, 25]. In the present experiment, preabsorption with G T M R 4 fusion protein abolished type I-ir (fig. 4B) while preabsorption with G S T peptide did not affect immunoreactivity. For type Il-ir, adjacent sections incubated with P3 AgX-653 myeloma cell supernatant produced only nonspecific staining (fig. 4E).
Cerebellum No corticosteroid immunoreactive cells were observed in the cerebellum at P0. By P5 (fig. 1) all lobules showed a monolayer o f type I-ir Purkinje cells. The number in creased steadily to adult levels by P20 (fig. 3). In all lob ules examined, very few type I-ir cells were observed in the external granular layer (developing molecular layer) and internal granular layer before P10. By PI 5 (fig. 4A) cells showing strong type I-ir were observed in the molec ular layer. By P30 (fig. 4C) the number o f type I-ir cells in the definitive granular layer was close to adult levels. Fewer type I-ir cells were observed in the definitive mo lecular layer (fig. 4C). The timing o f development of type I-ir in the cerebellar cortex was similar in all lobules. Type Il-ir Purkinje cells were first observed in the cere bellar hemispheres, flocculus, paraflocculus and tuber
(vermal lobule 7) by P I 5 (fig. 2 ,4D ). No differences were observed in anterior and posterior cerebelli. By P20 type Il-ir Purkinje cells were observed in all lobules (fig. 3). Moreover, the number and spatial distribution o f type Il-ir Purkinje cells at P20 was close to adult levels. Very few type Il-ir cells were observed in the molecular and granular layers as late as P30 (fig. 4F). Type I-ir cells were first observed in all deep cerebellar nuclei by P5 (data not shown), and increased subse quently in number (fig. 5 A -C ). In comparison, type Il-ir cells were first observed at P I 5 (data not shown). Few type Il-ir cells were observed at P20 (fig. 5E). By P30 (fig. 5F) the number had increased but was still less than levels (data not shown).
Brain Stem No corticosteroid immunoreactive cells were observed in the red, pontine, inferior olivary or medial vestibular nuclei at P0 (data not shown). Few type I-ir cells were observed in all nuclei by P5 (data not shown) except the red nucleus, where type I-ir cells were first observed at
697
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Results
Fig. 1. Distribution o f corticosteroid receptor immunoreactive (ir) Purkinje cells in the cerebellum at P5 and P10. Bach dot repre sents one labeled Purkinje neuron. Type I-ir cells were observed in all lobules. No type Il-ir cells were observed. Numbers refer to lo bules. I V = Fourth ventricle; Sim = simple lobule; Crus I =crus I ansiform lobule; Crus 2 = crus 2 ansiform lobule; PF1 = parafloccu lus; F L = flocculus; PM = paramedian lobule.
Fig. 2. Distribution o f corticosteroid re ceptor immunoreactive (ir) Purkinje cells in the cerebellum at P I 5. Each dot represents one labeled Purkinje neuron. Note the pres ence o f type l-ir cells in all lobules. Type IIir cells were observed only in few regions o f the cerebellar hemispheres, paraflocculus and flocculus. Numbers and abbreviations as in figure 1.
698
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Fig. 3. Distribution o f corticosteroid re ceptor immunoreactive (ir) Purkinje cells in the cerebellum at P20. Each dot represents one labeled Purkinje neuron. Type I-ir and type I I-ir cells were present in all lobules. The densities were similar to adult levels. C o p = Copula pyramis. Numbers and other abbreviations as in figure 1.
/
m
P /
A
D
P
P
9
9 B
E
m
••
« •
Fig. 4. Photomicrographs o f corticosteroid receptor immunoreactivity in the cerebellar cortex; A PI 5: type I-ir. Note the presence o f immunoreactive Purkinje cells (P) and cells with strong immunoreactivity (arrows) in the molecular layer (m). These may represent migrating type I-ir granule cells (see C). B P I 5: Preabsorption o f M IN R E C 4 antiserum with G T M R 4 fusion protein abolished type I-ir. P = Purkinje cell layer; m = molecular layer; g = granular layer. C P30: type I-ir. More Purkinje cells with strong nuclear and cytoplasmic type I-ir and intensely stained granular cells are
• 9
9
present. P = Purkinje cell layer: m = molecular layer; g = granular layer. D P15: type I I-ir. Note the paucity o f immunoreactive cells. P = Purkinje cell. E P15; Sections incubated with non-immune P3 AgX-653 myeloma medium did not show specific staining. P = Pur kinje cell layer; m = molecular layer; g = granular layer. F P30: type I I-ir. Note the presence o f immunoreactive Purkinje cells (P). The granular layer (g) shows very few immunoreactive cells. Bar = 50 pm.
699
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
►
% • V
'V
t «
*
1 ‘
*
>
f.
?
:
, - .•
i
%
1 *.
”
t A
*
’
>
*
' .*
v
;
i
V
*
-
♦
• *
t
:*■
*
'
1
•
1
f
i K
I
%
i *
.. •
••
^
11 • «
¥
**
9
/
0
• « ’I
f *• i
* *
*
i
«
'
t
v ^
■'
* ^
.
Fig. 5. Photomicrographs o f corticosteroid receptor immunoreactivity in the dentate nucleus. A P10: type I-ir. Note the presence o f type I-ir cells. B P20: type I-ir. There was an increase in the density o f immunoreactive cells. Most o f them showed both nuclear and cytoplasmic distribution o f receptor (arrow). C P30: type I-ir. The
700
7
* '
>
• *’ ,
*
9 \
»
density o f type I-ir cells was similar to adult levels. D PI 0: type I I-ir. Note the absence o f immunoreactive cells. E P20: type Il-ir. Few type I I-ir cells were observed. F P30: type Il-ir. The density of type 1I-ir cells increased. Note the presence o f both nuclear and cy toplasmic immunoreactivity (arrow). Bar = 50 um.
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
.
,1»
)4
\
i '
«
%4
1
i *
*
Neuroendocrinology 1992;55:695-707
Department o f Anatomy and Neuroscience Training Program, Tulane University School o f Medicine, New Orleans, La., U S A ; Department o f Anatomy, University o f Ghana Medical School, Accra, G hana; Baker Medical Institute, Prahan, Australia
Key Words Glucocorticoids Receptors Immunocytochemistry Cerebellum Brain Stem
Postnatal Development of Corticosteroid Receptor Immunoreactivity in the Rat Cerebellum and Brain Stem
Abstract The postnatal development o f corticosteroid receptor immunoreactivity in the rat cerebellum and related brainstem nuclei was studied using a type I receptor antiserum, M IN R E C 4 , and a type II receptor monoclonal antibody, B U G R 2 . Type 1 receptor immunoreactive (ir) Purkinje cells were first observed at postnatal day 5 (P5), and increased to adult levels by P20.Type I-ir cells, presumably migrating granule cells, were observed in the developing molecular layer o f the cerebellum at P5. By P30, the density o f type I-ir cells in the definitive molecular and granular layers was still less than adult levels. In contrast, type I I-ir Purkinje cells were first observed at PI 5 and increased to adult levels by P20. No type I I-ir cells were observed in the proliferative and migratory zones o f the molecular layer. By P30, the density o f type I I-ir cells in the molecular and granular layers was far less than adult levels. In the deep cerebellar nuclei and most brain stem nuclei type I-ir was observed at P5 and developed to adult levels by P30. Type I I-ir was observed in the deep cerebel lar nuclei, red and medial vestibular nuclei by P I 5. The pontine and inferior olivary nuclei showed type I I-ir cells by P10. Type I I-ir in these regions de veloped to adult levels by P30. The earlier development o f type I-ir suggests that the type I receptor may mediate the actions o f corticosteroids in the cere bellum and related brain stem nuclei during early postnatal life. Both type I and type II corticosteroid receptors may mediate corticosteroid effects in the cerebellum and brain stem nuclei during late postnatal life and adulthood.
Glucocorticoids have diverse effects on the developing nervous system [I, 2]. Neuronal and glial proliferation and differentiation are regulated by glucocorticoids. While glucocorticoids are required for normal C N S de velopment at certain critical periods, high levels have
deleterious effects. This is especially evident in germinal zones o f the developing brain, e.g. cerebellum, dentate gyrus and olfactory bulb, which undergo considerable de velopment during the postnatal period [1, 3, 4], Excess levels o f glucocorticoids inhibit D N A synthesis, and neu-
Tliis work was supported by N S24I48 (R .E .H ). Special thanks to R Harrison for supplying B U G R 2 antibody and P3 AgX-653 myeloma cell supernatant.
Richard E. Harlan, Ph i) Dept, o f Anatomy, Tulane University Medical School 1430 Tulane Avenue New Orleans. L A 70112 (U SA )
Received: July I I . 1991 Accepted after revision: September 12, 1991
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Aaron Lawson“ h Rexford S. Ahima“ Zygmunt Krozowskic Richard E. Harlan“ d
696
type Il-ir have been localized in the Purkinje and granular layers o f the cerebellar cortex, deep cerebellar nuclei and red, pontine, inferior olivary and medial vestibular nuclei [15, 16]. We report, based on immunocytochemical local ization o f corticosteroid receptors, that there are age-re lated regional differences in the distribution o f these re ceptors during postnatal development.
Materials and Methods Animals and Tissue Preparation Four pregnant Sprague-Dawley rats were used for the study. Each littered 12 pups which were kept with their dams and were not handled until sacrifice. Six pups each were sacrificed at postnatal day zero (P0; day o f birth), P5, P I0, P15, P20, P25 and P30. Three adult male rats (3 months old) were used for comparison. PO and P5 pups were anaesthetized by hypothermia and P I0-P 30 pups by methoxyflurane (Pitman-Moore) inhalation. The animals were per fused transcardially with phosphate buffered saline (PBS), pH 7.2, for 5 min followed by 3% phosphate-buffered paraformaldehyde for 8 min. Brains were dissected out. postfixed in the same fixative for 2 h and cryoprotected until they sank in 30% sucrose. They were then frozen with dry ice and stored wrapped in aluminium foil at -7 0 ° C until sectioning.
Antibodies M IN R F .C 4 antiserum, raised against a fusion protein G T M R 4 comprising a 167 amino acid fragment o f the N-terminal domain of the human type 1 receptor and glutathione-S-transferase (GST), G T M R 4 fusion protein and G S T antiserum were supplied by Krozowski’s group [13, 19, 20]. This antiserum has the same specificity as the previously characterized M IN R E C 2 antiserum [22] and has been used in studies in the rat C N S [13, 19, 20], B U G R 2 mouse antirat liver type 11 receptor monoclonal antibody and nonimmune P3 AgX-653 myeloma cell medium were supplied by Harrison’s group [21],
Immunocytochemistry Coronal sections, 30-pm thick, were cut on a cryostat and thawmounted onto chrom-alum-coated slides. They were processed for immunocytochemistry using Vectastain immunoperoxidase kits (Vector) and a modification o f a protocol by Hancock [23]. M IN R E C 4 antiserum (against type I receptor) was used at dilutions o f 1:500, 1: 1.000, 1:2,000 and 1:4,000. Adjacent control sections were incubated with either M IN R E C 4 antiserum (1:1,000) preabsorbed with 10 \iM G T M R 4 fusion protein or 10 u M G S T or 1:500, 1:1,000 and 1:2,000 dilution o f G S T antiserum. B U G R 2 antibody (against type II receptor) was used at dilutions o f 1:250, 1:500, 1:1,000 and 1:2,000. Adjacent control sections were incubated with either nor mal mouse IgG or nonimmune P3 AgX-653 myeloma cell medium. The chromogen used was 25 mg% 3,3'-diaminobenzidine tetrahydrochloride (D A B ) and 1.25% nickel ammonium sulfate, which produced a dark blue reaction product. Sections were dehydrated in ethanol, defatted with Histoclear (National Diagnostics) and coverslipped with Permount (Fisher Scientific). Sections adjacent to those selected for microscopic analysis were stained with 0.5% cresyl violet to facilitate delineation o f neuronal groups.
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
rona! and glial proliferation leading to reduced brain weight [ 1,4-6], Two different effects o f excess glucocorti coids on rat cerebellar development have been described [ 1, 4], Treatment o f rats with high doses o f cortisol be tween postnatal days 1 and 4 (P1-P4) inhibits cell prolife ration and alters the pattern o f foliation resulting in small and abnormally shaped cerebelli in adulthood. This pro cess is, however, reversible if treatment is terminated early. Late postnatal treatment with cortisol (P7-P18), produces greater inhibition o f neuronal proliferation and differentiation and smaller cerebelli compared to early treatment, and no effect on foliation. Neonatal treatment with cortisol delays the maturation o f motor behaviors, such as swimming [7] and motor coordination [8]. C o n versely, early adrenalectomy improves subsequent maze learning behavior [9], It is possible these effects o f gluco corticoids on motor function are mediated via a direct action on neuronal development and function in motorcoordinating centers, including the cerebellum. The actions o f glucocorticoids in the postnatal rat brain are mediated via two types o f corticosteroid receptors [10-12], A type I receptor, with the same characteristics as the classical mineralocorticoid receptor, binds with high affinity to corticosterone, the major glucocorticoid, and to aldosterone. Ligand-binding studies have indicated that this receptor in the forebrain is restricted to limbic regions [10, 12], A type II receptor, or classical glucocorticoid re ceptor with a lower affinity and higher capacity for cortico sterone, has been localized with binding studies in the lim bic brain o f postnatal rats [10]. Immunocytochemical studies [ I I , 13] have revealed a widespread distribution of type 11 receptors. We [ 13] have recently described the post natal development o f types I and II immunoreactive (ir) cells in the rat hippocampus. The patterns o f corticosteroid receptor binding [10, 12] and immunocytochemical local ization [11, 13] vary with age. This may be related to the roles o f these receptors in mediating the modulatory ef fects o f glucocorticoids on critical aspects o f C N S develop ment [10-12], Corticosteroid receptors have been local ized in the adult rat cerebellum using binding studies [14], in situ hybridization [15, 16] and immunocytochemistry [17-19], Presumably these receptors may mediate some of the effects ofglucocorticoids on postnatal cerebellar devel opment [1-4]. We have used an antiserum against the type I receptor [13, 19, 20] and an anti-type II receptor monoclonal antibody [ 18,21 ] to examine the development o f corticosteroid receptor-like immunoreactivity in the postnatal rat cerebellum and its brainstem connection. In the adult rat brain, moderate to high densities o f cells ex pressing type I and type II receptor m R N A or showing
Microscopy Sections were examined with a Nikon Optiphot microscope equipped with a Nikon F X 3 5A camera and photographs taken with Technical Pan (Kodak) film. Three coronal sections o f the cerebellum stained for type I and Il-ir from three pups per age were selected, matched and coded by one investigator. The levels were (i) a rostral section showing vermal lobules 1-6; (ii) a caudal section showing vermal lobules 7-10, and (in) a section between these two showing all deep cerebellar nuclei. Sections o f some brainstem nuclei with cerebellar connections, i.e. red, pontine, inferior olivary and medial vestibular nuclei were also analyzed. Sections stained for type I and Il-ir through the thickness o f these nuclei and separated by 90 pm (i.e. every third section) from three pups per age were selected, matched and coded by one investigator. The sections were analyzed blindly by two investigators observ ing the following parameters: (i) distribution o f immunoreactive cells; (ii) intracellular location o f receptors; (iii) relative intensity o f immunostaining - this was scored qualitatively as ‘strong’ as in fig ure 4A or ‘weak’ as in figure 8A. The distribution o f Purkinje cells in representative sections o f rostral and caudal cerebelli was plotted using a Bausch and Lomb projection microscope.
Specificity o f Antibodies The specificity o f M IN R E C 4 antiserum and B U G R 2 monoclonal antibody for type I and II receptors have been demonstrated previously [13, 18, 19, 24, 25]. In the present experiment, preabsorption with G T M R 4 fusion protein abolished type I-ir (fig. 4B) while preabsorption with G S T peptide did not affect immunoreactivity. For type Il-ir, adjacent sections incubated with P3 AgX-653 myeloma cell supernatant produced only nonspecific staining (fig. 4E).
Cerebellum No corticosteroid immunoreactive cells were observed in the cerebellum at P0. By P5 (fig. 1) all lobules showed a monolayer o f type I-ir Purkinje cells. The number in creased steadily to adult levels by P20 (fig. 3). In all lob ules examined, very few type I-ir cells were observed in the external granular layer (developing molecular layer) and internal granular layer before P10. By PI 5 (fig. 4A) cells showing strong type I-ir were observed in the molec ular layer. By P30 (fig. 4C) the number o f type I-ir cells in the definitive granular layer was close to adult levels. Fewer type I-ir cells were observed in the definitive mo lecular layer (fig. 4C). The timing o f development of type I-ir in the cerebellar cortex was similar in all lobules. Type Il-ir Purkinje cells were first observed in the cere bellar hemispheres, flocculus, paraflocculus and tuber
(vermal lobule 7) by P I 5 (fig. 2 ,4D ). No differences were observed in anterior and posterior cerebelli. By P20 type Il-ir Purkinje cells were observed in all lobules (fig. 3). Moreover, the number and spatial distribution o f type Il-ir Purkinje cells at P20 was close to adult levels. Very few type Il-ir cells were observed in the molecular and granular layers as late as P30 (fig. 4F). Type I-ir cells were first observed in all deep cerebellar nuclei by P5 (data not shown), and increased subse quently in number (fig. 5 A -C ). In comparison, type Il-ir cells were first observed at P I 5 (data not shown). Few type Il-ir cells were observed at P20 (fig. 5E). By P30 (fig. 5F) the number had increased but was still less than levels (data not shown).
Brain Stem No corticosteroid immunoreactive cells were observed in the red, pontine, inferior olivary or medial vestibular nuclei at P0 (data not shown). Few type I-ir cells were observed in all nuclei by P5 (data not shown) except the red nucleus, where type I-ir cells were first observed at
697
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Results
Fig. 1. Distribution o f corticosteroid receptor immunoreactive (ir) Purkinje cells in the cerebellum at P5 and P10. Bach dot repre sents one labeled Purkinje neuron. Type I-ir cells were observed in all lobules. No type Il-ir cells were observed. Numbers refer to lo bules. I V = Fourth ventricle; Sim = simple lobule; Crus I =crus I ansiform lobule; Crus 2 = crus 2 ansiform lobule; PF1 = parafloccu lus; F L = flocculus; PM = paramedian lobule.
Fig. 2. Distribution o f corticosteroid re ceptor immunoreactive (ir) Purkinje cells in the cerebellum at P I 5. Each dot represents one labeled Purkinje neuron. Note the pres ence o f type l-ir cells in all lobules. Type IIir cells were observed only in few regions o f the cerebellar hemispheres, paraflocculus and flocculus. Numbers and abbreviations as in figure 1.
698
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
Fig. 3. Distribution o f corticosteroid re ceptor immunoreactive (ir) Purkinje cells in the cerebellum at P20. Each dot represents one labeled Purkinje neuron. Type I-ir and type I I-ir cells were present in all lobules. The densities were similar to adult levels. C o p = Copula pyramis. Numbers and other abbreviations as in figure 1.
/
m
P /
A
D
P
P
9
9 B
E
m
••
« •
Fig. 4. Photomicrographs o f corticosteroid receptor immunoreactivity in the cerebellar cortex; A PI 5: type I-ir. Note the presence o f immunoreactive Purkinje cells (P) and cells with strong immunoreactivity (arrows) in the molecular layer (m). These may represent migrating type I-ir granule cells (see C). B P I 5: Preabsorption o f M IN R E C 4 antiserum with G T M R 4 fusion protein abolished type I-ir. P = Purkinje cell layer; m = molecular layer; g = granular layer. C P30: type I-ir. More Purkinje cells with strong nuclear and cytoplasmic type I-ir and intensely stained granular cells are
• 9
9
present. P = Purkinje cell layer: m = molecular layer; g = granular layer. D P15: type I I-ir. Note the paucity o f immunoreactive cells. P = Purkinje cell. E P15; Sections incubated with non-immune P3 AgX-653 myeloma medium did not show specific staining. P = Pur kinje cell layer; m = molecular layer; g = granular layer. F P30: type I I-ir. Note the presence o f immunoreactive Purkinje cells (P). The granular layer (g) shows very few immunoreactive cells. Bar = 50 pm.
699
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
►
% • V
'V
t «
*
1 ‘
*
>
f.
?
:
, - .•
i
%
1 *.
”
t A
*
’
>
*
' .*
v
;
i
V
*
-
♦
• *
t
:*■
*
'
1
•
1
f
i K
I
%
i *
.. •
••
^
11 • «
¥
**
9
/
0
• « ’I
f *• i
* *
*
i
«
'
t
v ^
■'
* ^
.
Fig. 5. Photomicrographs o f corticosteroid receptor immunoreactivity in the dentate nucleus. A P10: type I-ir. Note the presence o f type I-ir cells. B P20: type I-ir. There was an increase in the density o f immunoreactive cells. Most o f them showed both nuclear and cytoplasmic distribution o f receptor (arrow). C P30: type I-ir. The
700
7
* '
>
• *’ ,
*
9 \
»
density o f type I-ir cells was similar to adult levels. D PI 0: type I I-ir. Note the absence o f immunoreactive cells. E P20: type Il-ir. Few type I I-ir cells were observed. F P30: type Il-ir. The density of type 1I-ir cells increased. Note the presence o f both nuclear and cy toplasmic immunoreactivity (arrow). Bar = 50 um.
Lawson/Ahima/Krozowski/Harlan
Development of Corticosteroid Receptors in Cerebellum
Downloaded by: King's College London 137.73.144.138 - 1/13/2018 8:47:43 PM
.
,1»
)4
\
i '
«
%4
1
i *
*
