- Email: [email protected]
Resting and reactive astrocytes express adrenergic receptors in the adult rat brain
Brain Research Bullerin, Vol. 29, pp. 211-284,
1992
0361-9230/92
Copyright0 I992 Pergamon
Printed in the USA. All rights reserved.
$5.00 + .OO Press Ltd.
Resting and Reactive Astrocytes Express Adrenergic Receptors in the Adult Rat Brain JEROME
SUTIN’
AND
YANPING
SHAO*
Department of Anatomy and Cell Biology, Emory University School of Medicine, Atlanta, GA 30322 Received
27 November
199 1
SUTIN, J. AND Y. SHAO. Resting and reactive astrocytes express adrenergic receptors in the adult rat brain. BRAIN RES BULL 29(3/4) 277-284, 1992.-Adrenergic receptor subtypes were localized in situ and in cells isolated from the trigeminal motor nucleus and several other brain regions. To study receptor expression in reactive astrocytes, motor neuron degeneration and a glial reaction were induced in the trigeminal motor nucleus by the injection ofthe toxic lectin Ricin communis into the trigeminal motor root. Autoradiography following incubation of tissue sections in the a,-ligand ‘r51BE 2254 (HEAT) or the P-l&and ‘2510docyanopindolol (ICYP) showed a decrease in a,- and an increase in P-adrenergic receptor binding in the region of neuronal degeneration and gliosis. Glial hypertrophy, rather than hyperplasia, appears to be mainly responsible for the increased @-binding, since inhibition of mitosis with cytosine arabinofuranoside only partially blocked elevations of /3-adrenergic receptor binding and GFAP immunolabelling in reactive astrocytes. More direct evidence for the expression of adrenergic receptors in normal and reactive astrocytes was obtained by combined autoradiography and immunohistochemistry of cells dissociated from the cerebral cortex, striatum, cerebellum, and trigeminal motor nucleus of adult rats. More than 88% of GFAP-positive astrocytes showed varying densities of fl-adrenergic receptor binding. In each region, the &-subtype was proportionally greater than the &-subtype. Astrocytes also expressed a significant density of cY,-receptors. Trigeminal motor neurons did not show P-receptor binding, but had a density of cu,-receptors tenfold greater than astrocytes. A model for the role of astrocytes in adrenergic receptor-mediated modulation of trigeminal motor neuron excitability is discussed. Glia
Astrocytes
Synapse
Motor neuron
Beta adrenergic receptor
the brain (9,19). It contains few, if any, interneurons ( 13) and the motor neurons receive a monosynaptic glutamate or aspartate input from muscle spindles in the jaw closing muscles (1). Following neonatal 6-OHDA treatment, NE in the cerebral cortex is permanently depleted, while in the trigeminal motor nucleus there is vigorous sprouting resulting in a two- to fourfold increase in NE content. Stimulation of the noradrenergic axons alone does not discharge motor neurons, but does facilitate motor neuron firing evoked by concomitant stimulation of muscle spindle afferent axons. This facilitation is further enhanced in the NE hyperinnervated trigeminal motor nucleus. Intracellular recordings from trigeminal motor neurons indicate that the facilitation involves a presynaptic mechanism (20). The administration of the P-antagonist propranolol or the Nantagonist phentolamine partially blocks the facilitation. When both drugs are given together, the facilitation is virtually abolished. To determine the cellular localization of the LY-and @adrenergic receptors that mediate NE facilitation of trigeminal motor neurons, we employed receptor autoradiography oftissue sections and of astrocytes and neurons isolated from the adult rat brain.
THE work of many investigators has established that astrocytes grown in primary culture can express fl-adrenergic receptors, actively take up norepinephrine (NE), glutamate or other transmitters, and increase CAMP production and intermediate filament phosphorylation in response to P-agonist stimulation. These studies are described in detail by several other speakers at this symposium (McCarthy et al., Stone et al., Hansson et al., Shain et al.). The extent to which the properties of astrocytes is observed in cultures also occur in vivo, and their functional consequences is less well established. Our interest in glial cells as targets of NE grew out of studies of the physiological consequences of axonal sprouting in neurons. Studies with cytotoxic analogs of catecholamines (2,4,10) indicate that only axons arising from the locus coeruleus have the high affinity NE reuptake transporter, while the lateral tegmental NE neurons which innervate cranial and spinal motor neurons appear to have only a low affinity reuptake mechanism (Fig. I). One of the targets of the lateral tegmental system is the trigeminal motor nucleus. This cluster of about 2500 motor neurons innervating the jaw closing and opening muscles (Fig. 2) has the highest concentration of NE in
’ Requests for reprints should be addressed to Dr. Jerome Win. ‘Current address: Department of Pharmacology, University of North Carolina School of Medicine, Chapel Hill, NC 27599-7365.
277
218
MEDIAL
SUTIN
TEGMENTAL
NORADRENERCIC
LATERAL CELLS
AND
SHAO
TECMENTAL
NORADRENERGIC
CELLS
n.............................. Cerebral
cortex
Hypothalamus LOCUS COERULEUS
LATERAL LEMNISCUS
Cerebellum
Motor Sensory
SUB COERULEUS
v
KtiLLlKER FUSE
Spinal
.
= Sprouting
) = Low .
terminals
= No 101s of NE
-
NUCLEUS
cord
of new NE
of NE
REGION
v
termindr after
terminals
after
6-OHDA
6-OHDA
after
6-OHDA
FIG. I, Schematic diagram of the principal projection patterns of medial and lateral tegmental groups of noradrenergic neurons in the mammalian brain. Motor neurons in the brainstem and spinal cord ventral horn receive their innervation from the lateral tegmental group. The medial group projects more widely. including cranial nerve sensory nuclei and the dorsal horn of the spinal cord. Only cells of the locus coeruleus are destroyed by neonatal 6-OHDA treatment. presumably because other terminals lack a high affinity NE reuptake transporter.
METHOD
Adult male and female Sprague-Dawley rats were used in these studies. All surgical procedures were carried out under ketamine (50 mg/kg, diluted I: 1 with xylazine) anesthesia. For full details of the experimental methods see Shao and Sutin (14,15). RESULTS
AND
Digastric and mylohyoid
DISCUSSION
In earlier studies we reported a fivefold greater NE concentration in tissue punches from the trigeminal motor nucleus compared to the cerebral cortex ( 19). The trigeminal motor nucleus is prominently labelled in autoradiographs of tissue sections incubated in the nonselective /3-ligand ‘251CYP (30 PM) or the (Y,ligand “?BE2254 (HEAT, 80 PM) (Figs. 3 and 4). P-Receptor subtypes were determined by coincubation of separate slides with “‘ICYP and the p2 antagonist IPS 339 or the PI antagonist betaxolol. Nonspecific binding was determined by coincubation of slides in ‘251CYP with either I PM dl-propranolol or 200 FM isoproterinol. For the assessment of 01, nonspecific binding, incubations were carried out with ‘*‘HEAT in the presence of either I PM prazosin or 50 PM phentolamine. The difficulty in the cellular localization of silver grains produced by lzsI in tissue sections led to two approaches to identify the cell types with adrenergic receptors; elimination of discrete populations of neurons and dissociation of individual cells. The cytotoxic ligand Ricin communis was injected into the masseter nerve on one side and the animals allowed to survive for 7-28 days. when the tissue was processed for receptor autoradiography. Adjacent slides were immunolabelled with anti-GFAP to identify
FIG. 2. Drawing illustrating features of the innervation of jaw opening and jaw closing groups of muscles. The perikarya of primary afferent tibers associated with muscle spindles in the jaw closing muscles are located in the trigeminal mesencephalic nucleus (Mes V). and their central processes enter the motor nucleus to establish excitatory synapses directly on motor neurons. Interneurons involved in trigeminal reflexes are located outside the motor nucleus in the supratrigeminal nucleus (unlabelled) and reticular formation.
astrocytes or processed for the thiamine pyrophosphate reaction to identify vascular endothelium. The retrogradely transported Ricin destroys motor neurons ( 18.2 I ) and produces an associated microglial proliferation. astroglial hypertrophy, and small vessel angiogenesis. The use of Ricin rather than simple axotomy reduces the variability in motor neuron degeneration due to gender (22). Since motor neurons are the only neural cells in the trigeminal motor nucleus. any receptor binding must be in glial cells, afferent axon terminals, vascular smooth muscle. or endothelium. Changes in receptor density and the intensity of GFAP immunolabeling were assessed by silver grain counts and optical density measurements with the aid of BioQuant IV (R and M Biometrics) image analysis software. After Ricin treatment there is a 42-46s increase in the number of nonneural cells. GFAP and @-adrenergic receptor density (Fig. 2) in the region of motor neuron degeneration compared to the normal contralateral motor nucleus. There is also a substantial reduction of cu,-receptor binding (Fig. 4). To determine the relative contributions of cell division and hypertrophy to the increases in b-adrenergic receptor binding and GFAP. parallel groups of animals were treated with Ricin alone and Rich plus the mitotic inhibitor cytosine arabinofuranoside (ARAc). Compared to the normal contralateral motor nucleus. there was no increase in cell number, and a 28% increase in &receptor labelling in the animals receiving Ricin + ARAc. The decrease in al-binding was similar in both groups of animals. In addition to astrocytes. microglia and vascular endothelium (Fig. 6) also undergo hyperplasia and/or hypertrophy but do not account for the major part of the @-receptor density increase observed. Astrocytes in primary culture and protoplasmic astrocytes isolated from the adult rat cerebral cortex express IS-adrenergic
FIG. 3. (A) autoradiogram of a brain tissue section from the pons incubated in the nonselective fi-adrenergic ligand ‘*‘ICYP. The normal trigeminal motor nucleus is indicated by a single arrow. The left motor nucleus, in which motor neurons have been largely eliminated by Rich injections into the masseter nerve. is marked by two arrows. The increased accumulation of silver grains corresponds to the region of reactive gliosis. (B) photograph of a video image of the autoradiogram showing the 400 X 400 Frn2 area in each trigeminal motor nucleus in which optical density and silver grains were measured. (C) a pixel density histogram measured along the horizontal line through the trigeminal motor nuclei shown in the middle panel. Each pixel has a gray scale value ranging from 0 (0% light transmission) through 255 (loo?+ transmission). The peak densities of silver grains over the trigeminal motor nuclei are shown as white areas near the baseline.
ADRENERGIC
RECEPTORS
IN ASTROCYTES
RICIN SIDE
279
CONTROL SIDE
SUTIN AND SHAO
280
FIG. 4. Autoradiograph of a tissue section mcubated with the cr,-adrenergic receptor ligand “‘IBt,2254 (HEAI ). The trigcminal motet nucleus on the intact side is marked with a single arrow head. and dual arrowheads identify the nucleus on the side in which motor neuron degeneration and a glial reaction was induced by the injection of Ricrn into the masseter nerve.
1
0.1
Normal fzm
ICYP
FIG. 5. Relative optical density measurements and of GFAP labeling from the trigeminal
and standard deviations oftotal &adrenergic receptor binding motor nucleus on the normal and Ricin-treated sides of six
animals treated with cytosine arabinofuranoside to inhibit the mitotic division of glia and endothelial cells.
ADRENERGIC
RECEPTORS
IN ASTROCYTES
281
wa s estimated by counting silver grains in a 350 + 30 pm* circular wi ndow centered over the cell body and proximal processes. Th ie window also included some cell free areas and, therefore,
FIG. 6. Photomicrographs of the trigeminal motor nucleus treated with the tyramine pyrophosphatase reaction to label capillary endothelium in an animal treated with cytosine arabinofuranoside to block cell division. The upper panel shows the intact side, and the lower panel the region of motor neuron degeneration and glial reaction in the Ricin treated side. Hypertrophy of small vessels is typical of the region with reactive gliosis.
receptors (7,12). Our data from tissue sections suggests that normal and reactive astrocytes in the trigeminal motor nucleus possess fl-adrenergic receptors and that motor neurons are the main source of oc,-receptors. For direct verification of the cellular localization of adrenergic receptors, astrocytes and motor neurons were isolated from the brains of adult animals using the method of Farooq and Norton (3) and processed for combined autoradiography and GFAP immunochemistry (8,ll). The number of GFAP positive cells obtained from the cerebral cortex, striatum, cerebellum, and trigeminal motor nucleus ranged from 1 to 3% of the total cells isolated. In the Ricin-treated trigeminal motor nucleus the yield was nearly 9%. Most GFAP labelled cells isolated from each region had morphological features typical of protoplasmic astrocytes (Fig. 7) but were not characterized further for surface antigen properties. Receptor density in astrocytes
FIG. 7. Fluorescence images of anti-GFAP labelled astrocytes isolated from the adult rat cerebral cortex. Cells with the morphological characteristics of protoplasmic astrocytes were most common. An example isolated from the trigeminal motor nucleus is shown in the upper phetograph. Protoplasmic (smaller cells) and fibrous astrocyte (larger cell) from the cerebellum are shown in the lower panel.
SUTIN AND
282
FIG. 8. Combined 1251CYP autoradiographv I for $-adrenergic receptors and immunofluorescence for GFAP illustrating binding density in a reactive astrocytc from the R/tin-treated trigeminal motor nucleus. In the upper photo the focal plane is on the astrocyte. and in the bottom photo on the emulsion.
SHAO
ADRENERGIC
RECEPTORS
283
IN ASTROCYTES
Cb
MOV
RMOV
Brain Region
Beta
1
Beta
2
FIG. 9. The relative density of pi- and &-adrenergic receptor subtypes in identified astrocytes isolated from the striatum (CP), cerebral cortex (CTX), cerebellar cortex (Cb). normal trigeminal motor nucleus (MoV), and Ricin-treated trigeminal motor nucleus (RMoV).
the actual density values would be underestimated. Eighty-eight percent of the GFAP-positive cells showed some degree of “51CYP binding in the cell body and processes (Fig. 8). Although the density of binding of the P-adrenergic receptor subtypes varied from region to region (Table l), the &subtype was always proportionally greater (Fig 9). Reactive astrocytes isolated from the Ricin-treated trigeminal motor nucleus displayed a larger increase in /3,-receptors relative to p2, although the latter was still the dominant subtype (Fig. 9) Since P-adrenergic subtypes were determined in separate groups of single cells, it is not known if the mean differential expression of @,-receptors in the population of reactive astrocytes sampled is a typical feature of each cell.
TABLE
The number of silver grains over isolated trigeminal motor neurons incubated in ‘251CYP was never greater than the background, and these cells were considered devoid of P-receptors. a,-Adrenergic receptors were present in astrocytes isolated from all regions (the striatum was not sampled), and increased by about 50% in reactive astrocytes. Motor neurons, however,
Ia
1
SPECIFIC BINDING
OF ADRENERGIC RECEPTOR SUBTYPES ON ASTROCYTES AND NEURONS ISOLATED FROM ADULT RAT BRAIN
Cell Type
PI-AR binding
@,-AR binding
Cortical astrocytes Striatal astrocytes Cerebellar astrocytes Trigeminal astrocytes Trigeminal react. astrocytes Trigeminal motor neurons
13 Z!Z1.9 [55]* 15 * 2.0 [60] 46 t 3.3 [SO] 15 * 1.9 [45]
3 f 1.7 [49] 6+- 1.6[50] 12 f 1.3 [50] 5 + 15 [40]
26 i- 2.5 [55]
8k21
[55]
al-AR
binding
13 f 3.3 18 + 2.7 12 f 2.9 19 f 3.5 115 * 7.4
/32 binding (“‘ICYP in the presence of betaxolol) and P-binding (“‘ICYP in the presence of ICI-1 18,55 1). All values expressed as silver grain density (mean f SEM/l,OOO pm’) minus nonspecific binding. * Number of astrocytes sampled.
““4-o-.-& FIG. 10. A model of the hypothesized role of astrocytes (A) in the uptake of excitatory amino acid transmitter released from muscle spindle afferent axons (la) forming monosynaptic junctions (diagonal hatching) with trigeminal motor neurons (M). It is speculated that fl-adrenergic receptors on astrocytes stimulated by norepinephrine (NE) released from varicosities (cross hatching) reduce the active transport of excitatory amino acid into the glia, resulting in larger EPSPs in the motor neuron and facilitation of motor neuron discharge.
284
SUTIN
had a density of specific @i-binding nearly tenfold greater than that in astrocytes (Table I). The evidence for adrenergic receptors in astrocytes in the adult brain described here supports the earlier observations of Salm and McCarthy ( 12). and leads to questions about functional significance. Many investigators have developed evidence that glial cells in vitro bind and respond to NE [see (5,I7) for reviews]. Much of this data is related to second messenger and cytoskeletal protein changes. and there is relatively little information about glial cell involvement in adrenergic modulation of synaptic efficacy. The facilitation of the reflex activation of jaw-closing trigeminal motor neurons by norepinephrine appears to involve both rr- and j3-receptors in the rat (20) and cu,-receptors in the cat (16). The dense tu,-receptors we found on isolated motor neurons could mediate a direct action of norepinephrine. However. /3-adrenergic receptor-linked mechanisms must be presynaptic. since no &receptors were found in motor neurons. The presence of/3-receptors on astrocytes from the mature brain and the suppression by NE, at IO ’ M, of the uptake of glutamate into astrocytes in primary culture (6). support a role for these
AND
SHAO
cells in regulating the duration of transmitter action at nonadrenergic synapses. A model based on the in vitro uptake studies is proposed to account for fl-adrenergic receptor-mediated facilitation of trigeminal motor neurons (Fig. IO). Glutamate released by the la muscle spindle afferent axon terminal (cross hatching) on the motor neuron (M), is presumed to be terminated by active reuptake into the terminal and, we speculate. into adjacent astrocytes (A). The rate of uptake is enhanced by high levels of NE in the extracellular space through cu,-receptors, and inhibited by lower concentrations of NE via /3-adrenergic receptors. NE released from the varicosities of adrenergic axons innervating the trigeminal motor nucleus can act indirectly upon (ri- and /j-receptors on astrocytes and directly on n,-receptors on motor neurons. ACKNOWLEDGEMENTS
Supported by National Institutes of Health grant # NSl4778. WC are indebted to Mr. Ronald Griffith for excellent technical assistance and to Dr. Kenneth Minneman for providing radiolabeled ligands.
REFERENCES I. Chandler. S. H. Evidence for excitatory amino acid transmission between mesencephalic nucleus of V afferents and jaw-closer motoneurons in the guinea pig. Brain Res. 447:252-264: 1985. 2. Gulp. S.: Grzanna, R.; De Sousa, E. B.; Fritschy, J.-M.: Zaczek. R. Cortical and hypothalamic norepinephrine (NE) transport processes differ in their kinetic and pharmacological properties. Sot. Neurosci. Abstr. 15:1009; 1989. _7 Farooq, M.; Norton. W. T. A modified procedure for isolation of astrocytes- and neuron-enriched fraction from rat brain. J. Neurothem. 31:887-894; 1978. analysis of the 4. Fritcshy, J.-M.: Grzanna, R. Immunohistochemical neurotoxic effects of DSP-4 identifies two populations of noradrenergic axon terminals. Neuroscience 30: I8 I - 197: 1989. 5. Hansson, E. Co-existence between receptors, carriers, and second messengers on astrocytes grown in primary cultures. Neurochem. Res. l4:8l l-819; 1989. 6. Hansson. E.: Riinnblck, L. Regulation of glutamate and GABA transport by adrenoceptors in primary astroglial ceil cultures. Life Sci. 44:27-34: 1989. 7. Hosli. E.: Hosli, L. Evidence for the existence of N- and P-adrenoceptors on neurons and glial ceils of cultured rat central nervous system: An autoradiographic study. Neuroscience 7:2873-288 I: 1982. of 8. Katz. D. M.: Kimelberg, H. K. Kinetics and autoradiography high affinity uptake of serotonin by primary astrocyte cultures. J. Neurosci. 5:1901-1908: 1985. 9. Levitt. P.: Moore, R. Y. Organization of brainstem noradrenaline hyperinnervation following neonatal 6-OHDA treatment in rat. Anat. Embryol. 158: 133- 150; 1980. IO. McBride. R. L.; Ozment, R. V.: Sutin, J. Neonatal 6-hydroxydopamine dopamine destroys spinal cord noradrenergic axons from the locus coeruleus, but not those from lateral tegmental cell groups. J. Comp. Neurol. 235:375-383: 1985.
I I. McCarthy. K. D. An autoradiographic analysis of beta adrencrgic receptors on immunocytochemically defined astroglia. J. Pharmcol. Exp. Ther. X6282-290: 1983. 12. Salm. A. K.: McCarthy. K. D. Expression of beta-adrenergic receptors by astrocytes isolated from adult rat cortex. Glia 2:346-352: 1989. of alpha and gamma trigeminal moto13. Sessle. B. J. Identification neurons and effects of stimulation of amygdala. cerebellum. and cerebral cortex. Exp. Neurol. 54:303-332: -1977. in individual 14. Shao. Y.: Sutin. J. Exuression of adrenergic_ receotors . astrocytes and motor neurons isolated from the adult rat brain. Glia (in press). 15. Shao. Y.: Sutin. J. Noradrenergic facilitation of motor neurons: Localization of adrenergic receptors in neurons and non-neuronal cells in the trigeminal motor nucleus. Exp. Neurol. 114:216-227; 199 I, 16. Stafford, I. L.: Jacobs. B. L. Noradrenergic modulation of the masseteric reflex in behaving cats. I. Pharmacological studies. J. Neurosci. l&91-98: 1990. 17. Stone. E. A.: Ariano. M. A. Are glial cells targets of the central noradrenergic system: A review of the evidence. Brain Res. Rev. 14: 297-309: 1989. 18. Streit. W. J.: Kreutzberg. W. Response of endogenous glial cells to motor neuron degeneration induced by toxic Ricin. J. Comp. Neurol. 268:248-263; 1988. IY. Sutin. J.: Minneman. K. P. o,- and ij-adrenergic receptors are coregulated during both noradrenergic denervation and hyperinnervation. Neuroscience 14:973-980: 1985. of the motor 20. Vornov, J. J.: Sutin. J. Noradrenergic hyperinnervation trigeminal nucleus: Alternations in membrane properties and responses to synaptic input. J. Neurosci. 6:30-37: 1986. ‘I. Wiley. R. G.: Blessing. W. W.: Reis. D. J. Suicide transport: Destruction of neurons by retrograde transport of ricin, abrin, and modeccin. Science 216:889-X90: 1982. 22. Yu. W.-H. A. Sex difference in neuronal loss induced by axotomy in the rat brain stem motor nuclei. Exp. Neurol. 102:230-235: 1988.
1992
0361-9230/92
Copyright0 I992 Pergamon
Printed in the USA. All rights reserved.
$5.00 + .OO Press Ltd.
Resting and Reactive Astrocytes Express Adrenergic Receptors in the Adult Rat Brain JEROME
SUTIN’
AND
YANPING
SHAO*
Department of Anatomy and Cell Biology, Emory University School of Medicine, Atlanta, GA 30322 Received
27 November
199 1
SUTIN, J. AND Y. SHAO. Resting and reactive astrocytes express adrenergic receptors in the adult rat brain. BRAIN RES BULL 29(3/4) 277-284, 1992.-Adrenergic receptor subtypes were localized in situ and in cells isolated from the trigeminal motor nucleus and several other brain regions. To study receptor expression in reactive astrocytes, motor neuron degeneration and a glial reaction were induced in the trigeminal motor nucleus by the injection ofthe toxic lectin Ricin communis into the trigeminal motor root. Autoradiography following incubation of tissue sections in the a,-ligand ‘r51BE 2254 (HEAT) or the P-l&and ‘2510docyanopindolol (ICYP) showed a decrease in a,- and an increase in P-adrenergic receptor binding in the region of neuronal degeneration and gliosis. Glial hypertrophy, rather than hyperplasia, appears to be mainly responsible for the increased @-binding, since inhibition of mitosis with cytosine arabinofuranoside only partially blocked elevations of /3-adrenergic receptor binding and GFAP immunolabelling in reactive astrocytes. More direct evidence for the expression of adrenergic receptors in normal and reactive astrocytes was obtained by combined autoradiography and immunohistochemistry of cells dissociated from the cerebral cortex, striatum, cerebellum, and trigeminal motor nucleus of adult rats. More than 88% of GFAP-positive astrocytes showed varying densities of fl-adrenergic receptor binding. In each region, the &-subtype was proportionally greater than the &-subtype. Astrocytes also expressed a significant density of cY,-receptors. Trigeminal motor neurons did not show P-receptor binding, but had a density of cu,-receptors tenfold greater than astrocytes. A model for the role of astrocytes in adrenergic receptor-mediated modulation of trigeminal motor neuron excitability is discussed. Glia
Astrocytes
Synapse
Motor neuron
Beta adrenergic receptor
the brain (9,19). It contains few, if any, interneurons ( 13) and the motor neurons receive a monosynaptic glutamate or aspartate input from muscle spindles in the jaw closing muscles (1). Following neonatal 6-OHDA treatment, NE in the cerebral cortex is permanently depleted, while in the trigeminal motor nucleus there is vigorous sprouting resulting in a two- to fourfold increase in NE content. Stimulation of the noradrenergic axons alone does not discharge motor neurons, but does facilitate motor neuron firing evoked by concomitant stimulation of muscle spindle afferent axons. This facilitation is further enhanced in the NE hyperinnervated trigeminal motor nucleus. Intracellular recordings from trigeminal motor neurons indicate that the facilitation involves a presynaptic mechanism (20). The administration of the P-antagonist propranolol or the Nantagonist phentolamine partially blocks the facilitation. When both drugs are given together, the facilitation is virtually abolished. To determine the cellular localization of the LY-and @adrenergic receptors that mediate NE facilitation of trigeminal motor neurons, we employed receptor autoradiography oftissue sections and of astrocytes and neurons isolated from the adult rat brain.
THE work of many investigators has established that astrocytes grown in primary culture can express fl-adrenergic receptors, actively take up norepinephrine (NE), glutamate or other transmitters, and increase CAMP production and intermediate filament phosphorylation in response to P-agonist stimulation. These studies are described in detail by several other speakers at this symposium (McCarthy et al., Stone et al., Hansson et al., Shain et al.). The extent to which the properties of astrocytes is observed in cultures also occur in vivo, and their functional consequences is less well established. Our interest in glial cells as targets of NE grew out of studies of the physiological consequences of axonal sprouting in neurons. Studies with cytotoxic analogs of catecholamines (2,4,10) indicate that only axons arising from the locus coeruleus have the high affinity NE reuptake transporter, while the lateral tegmental NE neurons which innervate cranial and spinal motor neurons appear to have only a low affinity reuptake mechanism (Fig. I). One of the targets of the lateral tegmental system is the trigeminal motor nucleus. This cluster of about 2500 motor neurons innervating the jaw closing and opening muscles (Fig. 2) has the highest concentration of NE in
’ Requests for reprints should be addressed to Dr. Jerome Win. ‘Current address: Department of Pharmacology, University of North Carolina School of Medicine, Chapel Hill, NC 27599-7365.
277
218
MEDIAL
SUTIN
TEGMENTAL
NORADRENERCIC
LATERAL CELLS
AND
SHAO
TECMENTAL
NORADRENERGIC
CELLS
n.............................. Cerebral
cortex
Hypothalamus LOCUS COERULEUS
LATERAL LEMNISCUS
Cerebellum
Motor Sensory
SUB COERULEUS
v
KtiLLlKER FUSE
Spinal
.
= Sprouting
) = Low .
terminals
= No 101s of NE
-
NUCLEUS
cord
of new NE
of NE
REGION
v
termindr after
terminals
after
6-OHDA
6-OHDA
after
6-OHDA
FIG. I, Schematic diagram of the principal projection patterns of medial and lateral tegmental groups of noradrenergic neurons in the mammalian brain. Motor neurons in the brainstem and spinal cord ventral horn receive their innervation from the lateral tegmental group. The medial group projects more widely. including cranial nerve sensory nuclei and the dorsal horn of the spinal cord. Only cells of the locus coeruleus are destroyed by neonatal 6-OHDA treatment. presumably because other terminals lack a high affinity NE reuptake transporter.
METHOD
Adult male and female Sprague-Dawley rats were used in these studies. All surgical procedures were carried out under ketamine (50 mg/kg, diluted I: 1 with xylazine) anesthesia. For full details of the experimental methods see Shao and Sutin (14,15). RESULTS
AND
Digastric and mylohyoid
DISCUSSION
In earlier studies we reported a fivefold greater NE concentration in tissue punches from the trigeminal motor nucleus compared to the cerebral cortex ( 19). The trigeminal motor nucleus is prominently labelled in autoradiographs of tissue sections incubated in the nonselective /3-ligand ‘251CYP (30 PM) or the (Y,ligand “?BE2254 (HEAT, 80 PM) (Figs. 3 and 4). P-Receptor subtypes were determined by coincubation of separate slides with “‘ICYP and the p2 antagonist IPS 339 or the PI antagonist betaxolol. Nonspecific binding was determined by coincubation of slides in ‘251CYP with either I PM dl-propranolol or 200 FM isoproterinol. For the assessment of 01, nonspecific binding, incubations were carried out with ‘*‘HEAT in the presence of either I PM prazosin or 50 PM phentolamine. The difficulty in the cellular localization of silver grains produced by lzsI in tissue sections led to two approaches to identify the cell types with adrenergic receptors; elimination of discrete populations of neurons and dissociation of individual cells. The cytotoxic ligand Ricin communis was injected into the masseter nerve on one side and the animals allowed to survive for 7-28 days. when the tissue was processed for receptor autoradiography. Adjacent slides were immunolabelled with anti-GFAP to identify
FIG. 2. Drawing illustrating features of the innervation of jaw opening and jaw closing groups of muscles. The perikarya of primary afferent tibers associated with muscle spindles in the jaw closing muscles are located in the trigeminal mesencephalic nucleus (Mes V). and their central processes enter the motor nucleus to establish excitatory synapses directly on motor neurons. Interneurons involved in trigeminal reflexes are located outside the motor nucleus in the supratrigeminal nucleus (unlabelled) and reticular formation.
astrocytes or processed for the thiamine pyrophosphate reaction to identify vascular endothelium. The retrogradely transported Ricin destroys motor neurons ( 18.2 I ) and produces an associated microglial proliferation. astroglial hypertrophy, and small vessel angiogenesis. The use of Ricin rather than simple axotomy reduces the variability in motor neuron degeneration due to gender (22). Since motor neurons are the only neural cells in the trigeminal motor nucleus. any receptor binding must be in glial cells, afferent axon terminals, vascular smooth muscle. or endothelium. Changes in receptor density and the intensity of GFAP immunolabeling were assessed by silver grain counts and optical density measurements with the aid of BioQuant IV (R and M Biometrics) image analysis software. After Ricin treatment there is a 42-46s increase in the number of nonneural cells. GFAP and @-adrenergic receptor density (Fig. 2) in the region of motor neuron degeneration compared to the normal contralateral motor nucleus. There is also a substantial reduction of cu,-receptor binding (Fig. 4). To determine the relative contributions of cell division and hypertrophy to the increases in b-adrenergic receptor binding and GFAP. parallel groups of animals were treated with Ricin alone and Rich plus the mitotic inhibitor cytosine arabinofuranoside (ARAc). Compared to the normal contralateral motor nucleus. there was no increase in cell number, and a 28% increase in &receptor labelling in the animals receiving Ricin + ARAc. The decrease in al-binding was similar in both groups of animals. In addition to astrocytes. microglia and vascular endothelium (Fig. 6) also undergo hyperplasia and/or hypertrophy but do not account for the major part of the @-receptor density increase observed. Astrocytes in primary culture and protoplasmic astrocytes isolated from the adult rat cerebral cortex express IS-adrenergic
FIG. 3. (A) autoradiogram of a brain tissue section from the pons incubated in the nonselective fi-adrenergic ligand ‘*‘ICYP. The normal trigeminal motor nucleus is indicated by a single arrow. The left motor nucleus, in which motor neurons have been largely eliminated by Rich injections into the masseter nerve. is marked by two arrows. The increased accumulation of silver grains corresponds to the region of reactive gliosis. (B) photograph of a video image of the autoradiogram showing the 400 X 400 Frn2 area in each trigeminal motor nucleus in which optical density and silver grains were measured. (C) a pixel density histogram measured along the horizontal line through the trigeminal motor nuclei shown in the middle panel. Each pixel has a gray scale value ranging from 0 (0% light transmission) through 255 (loo?+ transmission). The peak densities of silver grains over the trigeminal motor nuclei are shown as white areas near the baseline.
ADRENERGIC
RECEPTORS
IN ASTROCYTES
RICIN SIDE
279
CONTROL SIDE
SUTIN AND SHAO
280
FIG. 4. Autoradiograph of a tissue section mcubated with the cr,-adrenergic receptor ligand “‘IBt,2254 (HEAI ). The trigcminal motet nucleus on the intact side is marked with a single arrow head. and dual arrowheads identify the nucleus on the side in which motor neuron degeneration and a glial reaction was induced by the injection of Ricrn into the masseter nerve.
1
0.1
Normal fzm
ICYP
FIG. 5. Relative optical density measurements and of GFAP labeling from the trigeminal
and standard deviations oftotal &adrenergic receptor binding motor nucleus on the normal and Ricin-treated sides of six
animals treated with cytosine arabinofuranoside to inhibit the mitotic division of glia and endothelial cells.
ADRENERGIC
RECEPTORS
IN ASTROCYTES
281
wa s estimated by counting silver grains in a 350 + 30 pm* circular wi ndow centered over the cell body and proximal processes. Th ie window also included some cell free areas and, therefore,
FIG. 6. Photomicrographs of the trigeminal motor nucleus treated with the tyramine pyrophosphatase reaction to label capillary endothelium in an animal treated with cytosine arabinofuranoside to block cell division. The upper panel shows the intact side, and the lower panel the region of motor neuron degeneration and glial reaction in the Ricin treated side. Hypertrophy of small vessels is typical of the region with reactive gliosis.
receptors (7,12). Our data from tissue sections suggests that normal and reactive astrocytes in the trigeminal motor nucleus possess fl-adrenergic receptors and that motor neurons are the main source of oc,-receptors. For direct verification of the cellular localization of adrenergic receptors, astrocytes and motor neurons were isolated from the brains of adult animals using the method of Farooq and Norton (3) and processed for combined autoradiography and GFAP immunochemistry (8,ll). The number of GFAP positive cells obtained from the cerebral cortex, striatum, cerebellum, and trigeminal motor nucleus ranged from 1 to 3% of the total cells isolated. In the Ricin-treated trigeminal motor nucleus the yield was nearly 9%. Most GFAP labelled cells isolated from each region had morphological features typical of protoplasmic astrocytes (Fig. 7) but were not characterized further for surface antigen properties. Receptor density in astrocytes
FIG. 7. Fluorescence images of anti-GFAP labelled astrocytes isolated from the adult rat cerebral cortex. Cells with the morphological characteristics of protoplasmic astrocytes were most common. An example isolated from the trigeminal motor nucleus is shown in the upper phetograph. Protoplasmic (smaller cells) and fibrous astrocyte (larger cell) from the cerebellum are shown in the lower panel.
SUTIN AND
282
FIG. 8. Combined 1251CYP autoradiographv I for $-adrenergic receptors and immunofluorescence for GFAP illustrating binding density in a reactive astrocytc from the R/tin-treated trigeminal motor nucleus. In the upper photo the focal plane is on the astrocyte. and in the bottom photo on the emulsion.
SHAO
ADRENERGIC
RECEPTORS
283
IN ASTROCYTES
Cb
MOV
RMOV
Brain Region
Beta
1
Beta
2
FIG. 9. The relative density of pi- and &-adrenergic receptor subtypes in identified astrocytes isolated from the striatum (CP), cerebral cortex (CTX), cerebellar cortex (Cb). normal trigeminal motor nucleus (MoV), and Ricin-treated trigeminal motor nucleus (RMoV).
the actual density values would be underestimated. Eighty-eight percent of the GFAP-positive cells showed some degree of “51CYP binding in the cell body and processes (Fig. 8). Although the density of binding of the P-adrenergic receptor subtypes varied from region to region (Table l), the &subtype was always proportionally greater (Fig 9). Reactive astrocytes isolated from the Ricin-treated trigeminal motor nucleus displayed a larger increase in /3,-receptors relative to p2, although the latter was still the dominant subtype (Fig. 9) Since P-adrenergic subtypes were determined in separate groups of single cells, it is not known if the mean differential expression of @,-receptors in the population of reactive astrocytes sampled is a typical feature of each cell.
TABLE
The number of silver grains over isolated trigeminal motor neurons incubated in ‘251CYP was never greater than the background, and these cells were considered devoid of P-receptors. a,-Adrenergic receptors were present in astrocytes isolated from all regions (the striatum was not sampled), and increased by about 50% in reactive astrocytes. Motor neurons, however,
Ia
1
SPECIFIC BINDING
OF ADRENERGIC RECEPTOR SUBTYPES ON ASTROCYTES AND NEURONS ISOLATED FROM ADULT RAT BRAIN
Cell Type
PI-AR binding
@,-AR binding
Cortical astrocytes Striatal astrocytes Cerebellar astrocytes Trigeminal astrocytes Trigeminal react. astrocytes Trigeminal motor neurons
13 Z!Z1.9 [55]* 15 * 2.0 [60] 46 t 3.3 [SO] 15 * 1.9 [45]
3 f 1.7 [49] 6+- 1.6[50] 12 f 1.3 [50] 5 + 15 [40]
26 i- 2.5 [55]
8k21
[55]
al-AR
binding
13 f 3.3 18 + 2.7 12 f 2.9 19 f 3.5 115 * 7.4
/32 binding (“‘ICYP in the presence of betaxolol) and P-binding (“‘ICYP in the presence of ICI-1 18,55 1). All values expressed as silver grain density (mean f SEM/l,OOO pm’) minus nonspecific binding. * Number of astrocytes sampled.
““4-o-.-& FIG. 10. A model of the hypothesized role of astrocytes (A) in the uptake of excitatory amino acid transmitter released from muscle spindle afferent axons (la) forming monosynaptic junctions (diagonal hatching) with trigeminal motor neurons (M). It is speculated that fl-adrenergic receptors on astrocytes stimulated by norepinephrine (NE) released from varicosities (cross hatching) reduce the active transport of excitatory amino acid into the glia, resulting in larger EPSPs in the motor neuron and facilitation of motor neuron discharge.
284
SUTIN
had a density of specific @i-binding nearly tenfold greater than that in astrocytes (Table I). The evidence for adrenergic receptors in astrocytes in the adult brain described here supports the earlier observations of Salm and McCarthy ( 12). and leads to questions about functional significance. Many investigators have developed evidence that glial cells in vitro bind and respond to NE [see (5,I7) for reviews]. Much of this data is related to second messenger and cytoskeletal protein changes. and there is relatively little information about glial cell involvement in adrenergic modulation of synaptic efficacy. The facilitation of the reflex activation of jaw-closing trigeminal motor neurons by norepinephrine appears to involve both rr- and j3-receptors in the rat (20) and cu,-receptors in the cat (16). The dense tu,-receptors we found on isolated motor neurons could mediate a direct action of norepinephrine. However. /3-adrenergic receptor-linked mechanisms must be presynaptic. since no &receptors were found in motor neurons. The presence of/3-receptors on astrocytes from the mature brain and the suppression by NE, at IO ’ M, of the uptake of glutamate into astrocytes in primary culture (6). support a role for these
AND
SHAO
cells in regulating the duration of transmitter action at nonadrenergic synapses. A model based on the in vitro uptake studies is proposed to account for fl-adrenergic receptor-mediated facilitation of trigeminal motor neurons (Fig. IO). Glutamate released by the la muscle spindle afferent axon terminal (cross hatching) on the motor neuron (M), is presumed to be terminated by active reuptake into the terminal and, we speculate. into adjacent astrocytes (A). The rate of uptake is enhanced by high levels of NE in the extracellular space through cu,-receptors, and inhibited by lower concentrations of NE via /3-adrenergic receptors. NE released from the varicosities of adrenergic axons innervating the trigeminal motor nucleus can act indirectly upon (ri- and /j-receptors on astrocytes and directly on n,-receptors on motor neurons. ACKNOWLEDGEMENTS
Supported by National Institutes of Health grant # NSl4778. WC are indebted to Mr. Ronald Griffith for excellent technical assistance and to Dr. Kenneth Minneman for providing radiolabeled ligands.
REFERENCES I. Chandler. S. H. Evidence for excitatory amino acid transmission between mesencephalic nucleus of V afferents and jaw-closer motoneurons in the guinea pig. Brain Res. 447:252-264: 1985. 2. Gulp. S.: Grzanna, R.; De Sousa, E. B.; Fritschy, J.-M.: Zaczek. R. Cortical and hypothalamic norepinephrine (NE) transport processes differ in their kinetic and pharmacological properties. Sot. Neurosci. Abstr. 15:1009; 1989. _7 Farooq, M.; Norton. W. T. A modified procedure for isolation of astrocytes- and neuron-enriched fraction from rat brain. J. Neurothem. 31:887-894; 1978. analysis of the 4. Fritcshy, J.-M.: Grzanna, R. Immunohistochemical neurotoxic effects of DSP-4 identifies two populations of noradrenergic axon terminals. Neuroscience 30: I8 I - 197: 1989. 5. Hansson, E. Co-existence between receptors, carriers, and second messengers on astrocytes grown in primary cultures. Neurochem. Res. l4:8l l-819; 1989. 6. Hansson. E.: Riinnblck, L. Regulation of glutamate and GABA transport by adrenoceptors in primary astroglial ceil cultures. Life Sci. 44:27-34: 1989. 7. Hosli. E.: Hosli, L. Evidence for the existence of N- and P-adrenoceptors on neurons and glial ceils of cultured rat central nervous system: An autoradiographic study. Neuroscience 7:2873-288 I: 1982. of 8. Katz. D. M.: Kimelberg, H. K. Kinetics and autoradiography high affinity uptake of serotonin by primary astrocyte cultures. J. Neurosci. 5:1901-1908: 1985. 9. Levitt. P.: Moore, R. Y. Organization of brainstem noradrenaline hyperinnervation following neonatal 6-OHDA treatment in rat. Anat. Embryol. 158: 133- 150; 1980. IO. McBride. R. L.; Ozment, R. V.: Sutin, J. Neonatal 6-hydroxydopamine dopamine destroys spinal cord noradrenergic axons from the locus coeruleus, but not those from lateral tegmental cell groups. J. Comp. Neurol. 235:375-383: 1985.
I I. McCarthy. K. D. An autoradiographic analysis of beta adrencrgic receptors on immunocytochemically defined astroglia. J. Pharmcol. Exp. Ther. X6282-290: 1983. 12. Salm. A. K.: McCarthy. K. D. Expression of beta-adrenergic receptors by astrocytes isolated from adult rat cortex. Glia 2:346-352: 1989. of alpha and gamma trigeminal moto13. Sessle. B. J. Identification neurons and effects of stimulation of amygdala. cerebellum. and cerebral cortex. Exp. Neurol. 54:303-332: -1977. in individual 14. Shao. Y.: Sutin. J. Exuression of adrenergic_ receotors . astrocytes and motor neurons isolated from the adult rat brain. Glia (in press). 15. Shao. Y.: Sutin. J. Noradrenergic facilitation of motor neurons: Localization of adrenergic receptors in neurons and non-neuronal cells in the trigeminal motor nucleus. Exp. Neurol. 114:216-227; 199 I, 16. Stafford, I. L.: Jacobs. B. L. Noradrenergic modulation of the masseteric reflex in behaving cats. I. Pharmacological studies. J. Neurosci. l&91-98: 1990. 17. Stone. E. A.: Ariano. M. A. Are glial cells targets of the central noradrenergic system: A review of the evidence. Brain Res. Rev. 14: 297-309: 1989. 18. Streit. W. J.: Kreutzberg. W. Response of endogenous glial cells to motor neuron degeneration induced by toxic Ricin. J. Comp. Neurol. 268:248-263; 1988. IY. Sutin. J.: Minneman. K. P. o,- and ij-adrenergic receptors are coregulated during both noradrenergic denervation and hyperinnervation. Neuroscience 14:973-980: 1985. of the motor 20. Vornov, J. J.: Sutin. J. Noradrenergic hyperinnervation trigeminal nucleus: Alternations in membrane properties and responses to synaptic input. J. Neurosci. 6:30-37: 1986. ‘I. Wiley. R. G.: Blessing. W. W.: Reis. D. J. Suicide transport: Destruction of neurons by retrograde transport of ricin, abrin, and modeccin. Science 216:889-X90: 1982. 22. Yu. W.-H. A. Sex difference in neuronal loss induced by axotomy in the rat brain stem motor nuclei. Exp. Neurol. 102:230-235: 1988.