- Email: [email protected]
Down-regulation of α2- and β-adrenoceptor binding sites in rat cortex caused by amygdalar kindling
EXPERIMENTAL
NEUROLOGY
90, 108- 117 ( 1985)
Down-Regulation of cy2- and ,&Adrenoceptor in Rat Cortex Caused by Amygdalar S.CLARE STANFORDANDJOHNG.
Binding Sites Kindling
R.JEF'FERYS'
The University Laboratory of Physiology, Oxford, and The Sobell Department of Nemophysiology, Institute of Neurology, Queen Square, London WClN 3BG, United Kingdom Received January 15. 1985; revision received May 30, I985 The role of central noradrenergic neurons in kindled seizures was assessed by comparison of az- and &adrenoceptor binding in the cerebral cortex from kindled and control rats. To minimize handling, which may modify kindling-induced changes in binding, the kindling protocol involved stimulation of the amygdala every hour for a maximum of 26 h. Twenty-four hours after kindling, down-regulation of /3adrenoceptors was found in both olfactory cortex and the remaining neocortex, whereas CQ down-regulation was confined to the olfactory cortex. At 21 days after kindling, the only change found was a down-regulation of &adrenoceptors in the neocortex. The results support the view that functional changes in central noradrenergic transmission are associated with the reduction in seizure threshold induced by kindling. 0 1985 Academic P~cs, hc.
INTRODUCTION Repeated electrical stimulation in vivo of the amygdalar nuclei (among other brain regions), can lead to progressive increases in response, culminating in generalized convulsions. This process is termed “kindling” and is widely used as an experimental model of epilepsy (5). Despite numerous studies, the mechanisms underlying the persistent reduction in seizure threshold characteristic of kindling remain poorly understood. The central noradrenergic system is implicated in kindled seizures as both the establishment and Abbreviation: [SH]DHA-[3H]dihydroaIprenolol. ’ The present address of Dr. Stanford is Department of Pharmacology and Therapeutics, Middlesex Hospital Medical School, Cleveland Street, London W 1P 7PN, UK. The authors thank the Medical Research Council (S.C.S.) and the Thorn Trust (J.G.R.J.) for support and Elizabeth A. Paul for statistical advice. 108 0014-4886/85 $3.00 Copyright 0 1985 by Academic Press. Inc. All rights of reproduction in any form reserved.
a*- AND B-BINDING
AFTER
KINDLING
109
expression of kindling are sensitive to pharmacological manipulations of noradrenergic function [( 1, 12, 13) reviewed in (lo)]. However, studies of the effects of kindling on content, release, and turnover of noradrenaline have failed to show any consistent changes [e.g., (10, 17)]. Measurement of adrenoceptor number and affinity, using the radioligand binding technique, provides an alternative index of changes in noradrenergic function. This technique has already revealed a reduction in the number of both (Ye-and ,&adrenoceptors in rat cortex after repeated electroconvulsive shock (22). More recently, McIntyre and Roberts (14) using a once-daily stimulation kindling protocol, showed a reduction in P-adrenoceptor number in the whole cortex after kindling. We report here the effects of kindling on (Ye-and /3-adrenoceptor binding in rat cerebral cortex; the olfactory cortex (i.e., prepiriform, piriform, and entorhinal cortices) was examined separately as it is particularly sensitive to kindling (4, 8). We adopted a once-hourly stimulation protocol [as used in (3, 15)] to minimize handling of the animals, a procedure that itself causes adrenoceptor down-regulation (2 1). This may be especially important because rats become jumpy or hyperreactive during kindling and thus might have neurochemical responses to handling which differ from those of unkindled animals. Preliminary reports of these data have been presented (6, 7). METHODS Kindling. Male Sprague-Dawley rats weighing 280 to 300 g at the start of the experiment (from either Bantin & Kingman or OLAC) were maintained on a 12-: 12-h light-dark schedule, with free access to food and water. At least 7 days after receipt of the rats, the left amygdala of two-thirds of each batch was implanted with bipolar, Teflon-coated, 125~pm, platinumiridium wire electrodes, (tips stripped of insulation for approx 300 pm). The implantation coordinates were 8 mm below cortical surface, 5 mm to the left of, and 0 and 1.0 mm caudal to bregma. The operation was carried out under halothane anesthesia. As a part of this study we confirmed that anesthesia had no direct effect on binding 24 h later (data not shown). Subsequently, all rats were housed individually and received the minimum of handling necessary for maintenance. Between 1600 and 1800 h, 10 to 14 days after implantation, pairs of rats were connected to stimulators through counterbalanced leads and mercury slip rings; a corresponding “unhandled,” unoperated, control rat was housed nearby. One of each pair of implanted rats, the “sham” control, received no electric stimuli, and the other “kindled” rat received trains of 100 biphasic, 400~PA, I-ms pulses at 50/s, once an hour, until three stage 5 (generalized) seizures were seen in response to these stimuli (20). (Two kindled rats not attaining this criterion
110
STANFORD
AND
JEFFERYS
after 24 to 26 stimulations were excluded from the study.) Thus, each kindled rat was matched with a sham and an unhandled control throughout the experiment and subsequent binding assay. Through the use of Student’s paired t test this yoked design excluded uncontrolled environmental and analytical variations from the comparison. Preparation of Tissues and Binding. The implanted rats were disconnected from their stimulator leads and, either 24 h or 21 days later, were stunned and killed by cervical dislocation. The cerebral cortex was dissected on ice and divided at the rhinal fissure to yield “olfactory cortex” and the “remaining” neocortex. The neocortex from each animal and the olfactory cortex pooled from three similarly treated animals were stored at -20°C prior to binding. Membrane suspensions were prepared as described elsewhere (22) and incubated 40 min at 26°C in 50 mMTris buffer, pH 7.8, containing either [3H]clonidine (for binding to high-affinity a,binding sites) or [3H]dihydroalprenolo1 ([3H]DHA; for binding to high-affinity P-binding sites). Nonspecific binding was assessed in the presence of 15 PM phentolamine or 3 PM (f)-propranolol, respectively. Six different concentrations of radioligand, in the range 0.1 to 4.0 n&f (each in duplicate) were used to construct binding curves. Over this range of concentrations phentolamine displaced at least 80% total bound [3H]clonidine, and (+)-propanolol displaced at least 70% of total bound [3H]DHA. Incubation was terminated by filtration over Whatman GF/B filters and washing with 15 ml Tris buffer. Proteins were measured by the method of Lowry et al. (11). Binding Analysis. Each binding curve was analyzed by the iterative, nonlinear regression computer program “Ligand” ( 16) which estimates the binding B,, (density of binding sites) and Kd (which is inversely proportional to the affinity for the radioligand). All binding data were best fitted by a single-site model. RESULTS Kindling. Successive kindling stimuli led to the sequence of behavioral responses described elsewhere in the literature (5, 20). The first stimulation typically caused an arrest of movement, or, occasionally chewing or other mild facial clonus. Subsequently the stimuli elicited myoclonus of the face and forelimbs, and then a more generalized myoclonus, and finally, usually , after 18 to 20 stimulations, the rearing and falling fits characteristic of stage i 5. The stimulation plugs were removed after the third stage 5 seizure had , been observed; in a few rats this required as many as five further stimuli as milder seizures intervened. This suggested that a postepileptic inhibitory process was occurring and that a 1.5- or 2-h interval might have been more suitable (19). After each seizure, the rats usually groomed or explored their cages.
(Y*- AND &BINDING
AFfER
111
KINDLING
Three rats (not used in the binding analysis) were restimulated 21 days after completion of the kindling procedure described above. All three showed generalized convulsions (two stage 5, one stage 4) on the first stimulation, demonstrating the persistence of kindling under our conditions. The kindled rats were hyperreactive in their response to handling for about 3 days after kindling, but subsequently returned to normal in this respect. Adrenoceptor Binding in Olfactory Cortex 24 h after Kindling. No statistically significant differences were detected between left and right olfactory cortices, therefore the data from the two sides were pooled for subsequent analysis. Comparison with unhandled controls showed that the sham operation had no effect on (Ye-or P-adrenoceptor density (B,,) or on the apparent affinity of these binding sites for their respective radioligands ( I/-Kd) (Table 1, Fig. IA and B). In contrast, kindling significantly reduced the B,,, for both receptor subtypes compared with either of the control groups; the differences between individual kindled rats and their respective controls are plotted in Fig. 1A, and B. The Kd for P-adrenoceptor binding was significantly lower in the kindled group than in the sham controls, but did not differ from the unhandled controls (Fig. 1B); no significant changes were found in the Kd for a2 binding (Fig. 1A). Adrenoceptor Binding in Neocortex 24 h after Kindling. Kindling had no effect on Kd for either receptor subtype (Fig. 2A, B). However, there was a statistically significant reduction in fi B,,, for the kindled group compared with either of the control groups (illustrated in Fig. 2B). In contrast, B,,, for a2-adrenoceptors was not affected by kindling 24 h earlier (Table 2, Fig. 2A). Long-Term Efects of Kindling on Adrenoceptor Binding. Both B,, and Kd of p binding in the neocortex were decreased 21 days after kindling compared with unhandled controls (Table 3). However, sham controls had intermediate values which were not significantly different from those of TABLE 1 Olfactory Cortex Adrenoceptor Binding 24 h after Kindling”
Unhandled control Sham control Kindled
1.5 + 0.2 1.7 k 0.3 1.3 f 0.3
162 rfr 17 177 + 18 133 + 17*.**
0.7 If 0.1 0.9 f 0.1 0.5 + 0.1**
143 * 17 152 If: 17 98 + 7*~**
a Values show Kd (nM) and B,, (pmol/g . protein), X + SE, N = 5. * P < 0.05 cf. unhandled control; ** P < 0.05 cf. sham control; matched-paired t test.
112
STANFORD A
SHAM
AND JEFFERYS
KINDLED
SHAM
KINDLED
a
: a
z
-40 -00
I
B
SHAM
KINDLED
SHAM
KINDLED
200
z0
100
0
5h
1
F i s
0
l ?
o
,P :
d a
0
0 0 ma
0
-100
t-200
1
0
FIG. 1. Changes in adrenergic binding parameters (B- and &) in olfactory cortex 24 h after kindling: comparison of SHAM controls with unhandled controls, and of KINDLED with unhandled controls. A-differences for [‘Hlclonidine binding. B-differences for [‘Hldihydroalprenolol binding. O-Right olfactory cortex, e-Left olfactory cortex.
either the kindled or unhandled groups. The cY2-adrenoceptor binding showed no significant differences 21 days after kindling, with the one exception that the Kd for sham controls was greater than for unhandled controls. Olfactory cortex was also collected from this 2 l-day-survival experiment. The amount of material was insufficient for full statistical analysis, but the
q- AND P-BINDING A
SHAM
113
AFTER KINDLING
KINDLED
KINDLED
SHAM 2
1
. .
s 2
0
l
l-
Q
-1 l -2 b ::
B 200
1
SWAM
-3
i
l
SHAM
KINDLED 2
1
1
KINDLED
0 .
b
FIG. 2. Changes in adrenergic binding parameters (B,, and &) in neocortex 24 h atier kindling: comparison of SHAM controls with unhandled controls, and of KINDLED with unhandled controls. A-differences for [3H]clonidine binding. B-differences for [3H]dihydroalprenolol binding.
limited data available suggested that the differences found at 24 h were no longer apparent. Binding in the control animals differed between the 24-h and 21-day experiments. As the two experiments were conducted at different times, many factors could explain the discrepancy, for instance, seasonal variations (9) or housing conditions. The latter may be particularly relevant, as adrenoceptor binding is sensitive to environmental factors, even those as
114
STANFORD
AND JEFFERYS TABLE 2
Neocortex Adrenoceptor Binding 24 h after Kindling”
Unhandled control Sham control Kindled
2.8 f 0.4 2.2 f 0.3 2.3 + 0.5
108k 9 895 15 84+ 16
1.9 f 0.5 1.7 + 0.3 1.6 + 0.5
212 + 59 176 f 25 87 f 15*~**
’ Values show Kd (nM) and B,, (pmol/g . protein), X + SE, N = 5. * P < 0.05 cf. unhandled control; ** P < 0.05 cf. sham control; matched-paired t test.
mild as handling and stroking for 1 min daily (21), and systematic differences in treatment occurred between the 24-h and 21-day experiments (for example only the latter group of rats was returned to the animal house for the period between kindling and killing). Furthermore, genetic differences cannot be excluded because different suppliers of Sprague-Dawley rats were used for the two experiments, (interestingly, we also have preliminary evidence that concentrations of vesicular noradrenaline differ in rats from these two suppliers). Whatever caused the variation in binding in the controls, the importance of using a yoked design and paired t test analysis is emphasized. DISCUSSION The results show that 24 h after kindling, fi-adrenoceptor numbers were reduced in both olfactory cortex and neocortex, whereas q-binding decreased only in the olfactory cortex. By 21 days after kindling only the reduction TABLE 3 Neocortical Adrenoceptor Binding 2 1 Days after Kindling“ a2
P
Group
&
B mu
Unhandled control Sham control Kindled
1.3 -t 0.4 3.4 * 0.2* 2.6 2 0.8
68 2 12 106 + 22 87+ 17
Lx
0.9 k 0.2 0.7 + 0.2 0.5 + 0.6*
0 Values show Kd (nM) and B,, (pmol/g * protein), X + SE, N = 5. * P -C 0.05 cf. unhandled control; matched-paired t test.
141 k 19 128 Z!I26 92 f 7*
a*- AND &BINDING
AFTER KINDLING
115
in neocortical P-binding remained. Changes in the affinities of adrenoceptors showed no consistent pattern and included increased affinity of ,&binding sites of the olfactory cortex 24 h after kindling, and at 2 1 days, a reduction of affinity of az-binding sites of the sham controls, and an increase in the affinity of P-binding sites in neocortex of the kindled rats. Regional variation in receptor changes, as seen here, are not unprecedented; the effects of antidepressant treatments, for instance, have also been shown to vary throughout the brain, although all the regions studied receive an input from the nucleus locus coeruleus (23). However, this is the first time that regional differences have been demonstrated within the cerebral cortex in response to a particular treatment. The reduction in seizure threshold caused by kindling is very persistent; under our conditions the stimulus that was initially subconvulsive still triggered generalized convulsions 2 1 days after kindling had been completed. Any neurochemical change directly relevant to the changes in seizure threshold should have a similar time course. Our limited data for olfactory cortex at 2 1 days suggest that adrenoceptor changes were no longer apparent. In the neocortex, on the other hand, P-binding remained significantly below that of the unhandled controls. However, the sham-operated group had an intermediate level of P-binding, and did not differ significantly either from the kindled group or the unhandled controls. This apparently equivocal result suggests that the prolonged presence of the implant in the sham animals can itself cause a down-regulation of ,&adrenoceptors, which may explain why kindling occurs more rapidly after electrodes have been in place for about 28 days (2). Our data thus further support a relationship between the noradrenergic system and the establishment of kindling. McIntyre and Roberts (14) showed that fl-adrenoceptors were still downregulated 23 days after kindling compared with sham-operated controls, whereas we found changes only at 21 days when kindled and unoperated animals were compared. There were several differences in protocol that could explain this apparent discrepancy in detail between the two studies. First, an end point of six stage 5 seizures was used in (14) compared with three used here. Secondly, a daily kindling protocol was used in (14) which involves more handling than our hourly protocol [see (2 1) and Introduction]. Nevertheless, we are in broad agreement that kindling has a persistent effect on neocortical @-adrenoceptors. Another study of this problem by McNamara and colleagues (3, 15), who used a single concentration of [3H]DHA to assess P-binding, also found a decrease at 3 days, but not at 15 h, after kindling. Because our data for 24 h after the last seizure showed a clear down-regulation, it appears that the kindling-induced reduction in the number of fi-adrenoceptors occurs between 15 and 24 h. These findings at 24 h do not support McIntyre and Roberts’s suggestion ( 14) that seizures cause a transient up-regulation of receptors.
116
STANFORD
AND JEFFERYS
It is likely that the down-regulation of adrenoceptors is a consequence of excessive neuronal activity and release of noradrenaline during kindling. The physiologic significance of these changes remains to be established, but possibilities are first, that the persistent changes in neocortical P-binding have a role in the changes in seizure threshold responsible for the kindling phenomenon; second, that the transient changes in binding observed in olfactory cortex may underlie the short-term behavioral consequences of kindling, such as hyperreactivity; and third, that the changes in a,binding, found only in olfactory cortex at 24 h, may account for the sensitivity of this brain region to kindling and for the possible proconvulsant actions of a2 antagonists ( 18). REFERENCES I. ALTMAN, 1. M. AND M. E. CORCORAN. 1983. Facilitation of neocortical kindling by depletion of forebrain noradrenaline. Bruin Res. 270: 174-177. 2. BLACKWOOD, D. H. R., M. J. MARTIN, AND J. K. MCQUEEN. 1982. Enhanced rate of kindling after prolonged electrode implantation into the amygdala of rats. J. Neurosci. Methods 5: 343-348. 3. BYRNE, M. C., R. GO~LIEB, AND J. 0. MCNAMARA. 1980. Amygdala kindling induces muscarinic cholinergic receptor declines in a highly specific distribution within the limbic system. Exp. Neurol. 69: 85-98. 4. CAIN, D. P., AND M. E. CORCORAN. 1978. Kindling in olfactory-lesioned rats. Behav. Biol. 22: 264-268. 5. GODDARD, G. V., D. C. MCINTYRE, AND C. K. LEECH. 1969. A permanent change in brain function resulting from daily electrical stimulation. Exp. Neural. 25: 294-330. 6. JEFFERYS,J. G. R., AND S. C. STANFORD. 1983. Changes in (Y*- and fl-adrenoceptor binding sites in amygdalar kindled rat brain. Sot. Neurosci. Abstr. 9: 487. 7. JEFFERYS,J. G. R., AND S. C. STANFORD. 1984. The effects of amygdalar kindling on LY*and fl-adrenoceptor binding in rat cortex. J. Physiol. (London) 346: 43P. 8. KAIRESS, E. W., R. J. RACINE, AND G. K. SMITH. 1984. The development of the interictal spike during kindling in the rat. Bruin Res. 322: 101-l IO. 9. KAFKA, M. S., A. WIRZ-JUSTICE, AND D. NABER. 1981. Circadian and seasonal rhythms in (Y- and @-adrenergic receptors in the rat brain. Bruin Res. 207: 409-419. 10. KALICHMAN, M. W. 1982. Neurochemical correlates of the kindling model of epilepsy. Neurosci. Biol. Behav. Rev. 6: 165-181. I I. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. I95 I. Protein measurements with the Folin phenol reagent. J. Biol. Chem. 193: 265-273. 12. MCINTYRE, D. C. 1980. Amygdala kindling in rats: facilitation after local amygdala norepinephrine depletion with 6-hydroxydopamine. Exp. Neurol. 69: 395-407. 13. MCINTYRE, D. C., AND N. EDSON. I98 I. Facilation of amygdala kindling alter nompinephrine depletion with 6-hydroxydopamine in rats. Exp. Neural. 74: 148-757. 14. MCINTYRE, D. C., AND D. C. S. ROBERTS. 1983. Long-term reduction in beta-adrenergic receptor binding after amygdala kindling. Exp. Neurol. 82: 17-24. 15. MCNAMARA, J. 0. 1978. Selective alterations of regional adrenergic receptor binding in the kindling model of epilepsy. Exp. Neural. 61: 582-59 I. 16. MUNSON, P. J., AND D. RODBARD. 1980. Ligand: a versatile computerized approach for characterization of ligand binding systems.Anal. B&hem. 107: 220-239.
a?- AND P-BINDING
AFTER KINDLING
117
17. OKAZAKI, M. M., W. M. BURNHAM, AND J. J. WARSH. 1983. Unchanged noradrenaline levels and turnover two months alter amygdalar kindling in the rat. Sot. Neurosci. Abstr. 9: 487. 18. PAPANICOLAOU, J., R. J. SUMMERS, F. J. E. VAJDA, AND W. J. LOUIS. 1982. The relationship between a*-adrenoceptor selectivity and anticonvulsant effect in a series of clonidine-like drugs. Bruin Res. 241: 393-397. 19. PETERSON, S. L., T. E. ALBERTSON, AND L. G. STARK. 1981. Intertrial intervals and kindled seizures. Exp. Neurol. 71: 144-153. 20. RACINE, R. J. 1972. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroenceph. Clin. Neurophysiol. 32: 28 1-294. 21. STANFORD, S. C., M. FILLENZ, AND E. RYAN. 1984. The effect of repeated mild stress on cerebral cortical adrenoceptors and noradrenaline synthesis in the rat. Neurosci. Left. 45: 163-167.
STANFORD, S. C., AND D. J. NUTT. 1982. Comparison of the effects of repeated electroconvulsive shock on Q- and b-adrenoceptors in different regions of rat brain. Neuroscience 7: 1753-1757. 23. STANFORD C., D. J. Nut-r, AND P. J. COWEN. 1983. Comparison of the effects of chronic desmethylimipramine administration on (Y*-and &adrenoceptors in different regions of rat brain. Neuroscience 8: 16 1- 164. 22.
NEUROLOGY
90, 108- 117 ( 1985)
Down-Regulation of cy2- and ,&Adrenoceptor in Rat Cortex Caused by Amygdalar S.CLARE STANFORDANDJOHNG.
Binding Sites Kindling
R.JEF'FERYS'
The University Laboratory of Physiology, Oxford, and The Sobell Department of Nemophysiology, Institute of Neurology, Queen Square, London WClN 3BG, United Kingdom Received January 15. 1985; revision received May 30, I985 The role of central noradrenergic neurons in kindled seizures was assessed by comparison of az- and &adrenoceptor binding in the cerebral cortex from kindled and control rats. To minimize handling, which may modify kindling-induced changes in binding, the kindling protocol involved stimulation of the amygdala every hour for a maximum of 26 h. Twenty-four hours after kindling, down-regulation of /3adrenoceptors was found in both olfactory cortex and the remaining neocortex, whereas CQ down-regulation was confined to the olfactory cortex. At 21 days after kindling, the only change found was a down-regulation of &adrenoceptors in the neocortex. The results support the view that functional changes in central noradrenergic transmission are associated with the reduction in seizure threshold induced by kindling. 0 1985 Academic P~cs, hc.
INTRODUCTION Repeated electrical stimulation in vivo of the amygdalar nuclei (among other brain regions), can lead to progressive increases in response, culminating in generalized convulsions. This process is termed “kindling” and is widely used as an experimental model of epilepsy (5). Despite numerous studies, the mechanisms underlying the persistent reduction in seizure threshold characteristic of kindling remain poorly understood. The central noradrenergic system is implicated in kindled seizures as both the establishment and Abbreviation: [SH]DHA-[3H]dihydroaIprenolol. ’ The present address of Dr. Stanford is Department of Pharmacology and Therapeutics, Middlesex Hospital Medical School, Cleveland Street, London W 1P 7PN, UK. The authors thank the Medical Research Council (S.C.S.) and the Thorn Trust (J.G.R.J.) for support and Elizabeth A. Paul for statistical advice. 108 0014-4886/85 $3.00 Copyright 0 1985 by Academic Press. Inc. All rights of reproduction in any form reserved.
a*- AND B-BINDING
AFTER
KINDLING
109
expression of kindling are sensitive to pharmacological manipulations of noradrenergic function [( 1, 12, 13) reviewed in (lo)]. However, studies of the effects of kindling on content, release, and turnover of noradrenaline have failed to show any consistent changes [e.g., (10, 17)]. Measurement of adrenoceptor number and affinity, using the radioligand binding technique, provides an alternative index of changes in noradrenergic function. This technique has already revealed a reduction in the number of both (Ye-and ,&adrenoceptors in rat cortex after repeated electroconvulsive shock (22). More recently, McIntyre and Roberts (14) using a once-daily stimulation kindling protocol, showed a reduction in P-adrenoceptor number in the whole cortex after kindling. We report here the effects of kindling on (Ye-and /3-adrenoceptor binding in rat cerebral cortex; the olfactory cortex (i.e., prepiriform, piriform, and entorhinal cortices) was examined separately as it is particularly sensitive to kindling (4, 8). We adopted a once-hourly stimulation protocol [as used in (3, 15)] to minimize handling of the animals, a procedure that itself causes adrenoceptor down-regulation (2 1). This may be especially important because rats become jumpy or hyperreactive during kindling and thus might have neurochemical responses to handling which differ from those of unkindled animals. Preliminary reports of these data have been presented (6, 7). METHODS Kindling. Male Sprague-Dawley rats weighing 280 to 300 g at the start of the experiment (from either Bantin & Kingman or OLAC) were maintained on a 12-: 12-h light-dark schedule, with free access to food and water. At least 7 days after receipt of the rats, the left amygdala of two-thirds of each batch was implanted with bipolar, Teflon-coated, 125~pm, platinumiridium wire electrodes, (tips stripped of insulation for approx 300 pm). The implantation coordinates were 8 mm below cortical surface, 5 mm to the left of, and 0 and 1.0 mm caudal to bregma. The operation was carried out under halothane anesthesia. As a part of this study we confirmed that anesthesia had no direct effect on binding 24 h later (data not shown). Subsequently, all rats were housed individually and received the minimum of handling necessary for maintenance. Between 1600 and 1800 h, 10 to 14 days after implantation, pairs of rats were connected to stimulators through counterbalanced leads and mercury slip rings; a corresponding “unhandled,” unoperated, control rat was housed nearby. One of each pair of implanted rats, the “sham” control, received no electric stimuli, and the other “kindled” rat received trains of 100 biphasic, 400~PA, I-ms pulses at 50/s, once an hour, until three stage 5 (generalized) seizures were seen in response to these stimuli (20). (Two kindled rats not attaining this criterion
110
STANFORD
AND
JEFFERYS
after 24 to 26 stimulations were excluded from the study.) Thus, each kindled rat was matched with a sham and an unhandled control throughout the experiment and subsequent binding assay. Through the use of Student’s paired t test this yoked design excluded uncontrolled environmental and analytical variations from the comparison. Preparation of Tissues and Binding. The implanted rats were disconnected from their stimulator leads and, either 24 h or 21 days later, were stunned and killed by cervical dislocation. The cerebral cortex was dissected on ice and divided at the rhinal fissure to yield “olfactory cortex” and the “remaining” neocortex. The neocortex from each animal and the olfactory cortex pooled from three similarly treated animals were stored at -20°C prior to binding. Membrane suspensions were prepared as described elsewhere (22) and incubated 40 min at 26°C in 50 mMTris buffer, pH 7.8, containing either [3H]clonidine (for binding to high-affinity a,binding sites) or [3H]dihydroalprenolo1 ([3H]DHA; for binding to high-affinity P-binding sites). Nonspecific binding was assessed in the presence of 15 PM phentolamine or 3 PM (f)-propranolol, respectively. Six different concentrations of radioligand, in the range 0.1 to 4.0 n&f (each in duplicate) were used to construct binding curves. Over this range of concentrations phentolamine displaced at least 80% total bound [3H]clonidine, and (+)-propanolol displaced at least 70% of total bound [3H]DHA. Incubation was terminated by filtration over Whatman GF/B filters and washing with 15 ml Tris buffer. Proteins were measured by the method of Lowry et al. (11). Binding Analysis. Each binding curve was analyzed by the iterative, nonlinear regression computer program “Ligand” ( 16) which estimates the binding B,, (density of binding sites) and Kd (which is inversely proportional to the affinity for the radioligand). All binding data were best fitted by a single-site model. RESULTS Kindling. Successive kindling stimuli led to the sequence of behavioral responses described elsewhere in the literature (5, 20). The first stimulation typically caused an arrest of movement, or, occasionally chewing or other mild facial clonus. Subsequently the stimuli elicited myoclonus of the face and forelimbs, and then a more generalized myoclonus, and finally, usually , after 18 to 20 stimulations, the rearing and falling fits characteristic of stage i 5. The stimulation plugs were removed after the third stage 5 seizure had , been observed; in a few rats this required as many as five further stimuli as milder seizures intervened. This suggested that a postepileptic inhibitory process was occurring and that a 1.5- or 2-h interval might have been more suitable (19). After each seizure, the rats usually groomed or explored their cages.
(Y*- AND &BINDING
AFfER
111
KINDLING
Three rats (not used in the binding analysis) were restimulated 21 days after completion of the kindling procedure described above. All three showed generalized convulsions (two stage 5, one stage 4) on the first stimulation, demonstrating the persistence of kindling under our conditions. The kindled rats were hyperreactive in their response to handling for about 3 days after kindling, but subsequently returned to normal in this respect. Adrenoceptor Binding in Olfactory Cortex 24 h after Kindling. No statistically significant differences were detected between left and right olfactory cortices, therefore the data from the two sides were pooled for subsequent analysis. Comparison with unhandled controls showed that the sham operation had no effect on (Ye-or P-adrenoceptor density (B,,) or on the apparent affinity of these binding sites for their respective radioligands ( I/-Kd) (Table 1, Fig. IA and B). In contrast, kindling significantly reduced the B,,, for both receptor subtypes compared with either of the control groups; the differences between individual kindled rats and their respective controls are plotted in Fig. 1A, and B. The Kd for P-adrenoceptor binding was significantly lower in the kindled group than in the sham controls, but did not differ from the unhandled controls (Fig. 1B); no significant changes were found in the Kd for a2 binding (Fig. 1A). Adrenoceptor Binding in Neocortex 24 h after Kindling. Kindling had no effect on Kd for either receptor subtype (Fig. 2A, B). However, there was a statistically significant reduction in fi B,,, for the kindled group compared with either of the control groups (illustrated in Fig. 2B). In contrast, B,,, for a2-adrenoceptors was not affected by kindling 24 h earlier (Table 2, Fig. 2A). Long-Term Efects of Kindling on Adrenoceptor Binding. Both B,, and Kd of p binding in the neocortex were decreased 21 days after kindling compared with unhandled controls (Table 3). However, sham controls had intermediate values which were not significantly different from those of TABLE 1 Olfactory Cortex Adrenoceptor Binding 24 h after Kindling”
Unhandled control Sham control Kindled
1.5 + 0.2 1.7 k 0.3 1.3 f 0.3
162 rfr 17 177 + 18 133 + 17*.**
0.7 If 0.1 0.9 f 0.1 0.5 + 0.1**
143 * 17 152 If: 17 98 + 7*~**
a Values show Kd (nM) and B,, (pmol/g . protein), X + SE, N = 5. * P < 0.05 cf. unhandled control; ** P < 0.05 cf. sham control; matched-paired t test.
112
STANFORD A
SHAM
AND JEFFERYS
KINDLED
SHAM
KINDLED
a
: a
z
-40 -00
I
B
SHAM
KINDLED
SHAM
KINDLED
200
z0
100
0
5h
1
F i s
0
l ?
o
,P :
d a
0
0 0 ma
0
-100
t-200
1
0
FIG. 1. Changes in adrenergic binding parameters (B- and &) in olfactory cortex 24 h after kindling: comparison of SHAM controls with unhandled controls, and of KINDLED with unhandled controls. A-differences for [‘Hlclonidine binding. B-differences for [‘Hldihydroalprenolol binding. O-Right olfactory cortex, e-Left olfactory cortex.
either the kindled or unhandled groups. The cY2-adrenoceptor binding showed no significant differences 21 days after kindling, with the one exception that the Kd for sham controls was greater than for unhandled controls. Olfactory cortex was also collected from this 2 l-day-survival experiment. The amount of material was insufficient for full statistical analysis, but the
q- AND P-BINDING A
SHAM
113
AFTER KINDLING
KINDLED
KINDLED
SHAM 2
1
. .
s 2
0
l
l-
Q
-1 l -2 b ::
B 200
1
SWAM
-3
i
l
SHAM
KINDLED 2
1
1
KINDLED
0 .
b
FIG. 2. Changes in adrenergic binding parameters (B,, and &) in neocortex 24 h atier kindling: comparison of SHAM controls with unhandled controls, and of KINDLED with unhandled controls. A-differences for [3H]clonidine binding. B-differences for [3H]dihydroalprenolol binding.
limited data available suggested that the differences found at 24 h were no longer apparent. Binding in the control animals differed between the 24-h and 21-day experiments. As the two experiments were conducted at different times, many factors could explain the discrepancy, for instance, seasonal variations (9) or housing conditions. The latter may be particularly relevant, as adrenoceptor binding is sensitive to environmental factors, even those as
114
STANFORD
AND JEFFERYS TABLE 2
Neocortex Adrenoceptor Binding 24 h after Kindling”
Unhandled control Sham control Kindled
2.8 f 0.4 2.2 f 0.3 2.3 + 0.5
108k 9 895 15 84+ 16
1.9 f 0.5 1.7 + 0.3 1.6 + 0.5
212 + 59 176 f 25 87 f 15*~**
’ Values show Kd (nM) and B,, (pmol/g . protein), X + SE, N = 5. * P < 0.05 cf. unhandled control; ** P < 0.05 cf. sham control; matched-paired t test.
mild as handling and stroking for 1 min daily (21), and systematic differences in treatment occurred between the 24-h and 21-day experiments (for example only the latter group of rats was returned to the animal house for the period between kindling and killing). Furthermore, genetic differences cannot be excluded because different suppliers of Sprague-Dawley rats were used for the two experiments, (interestingly, we also have preliminary evidence that concentrations of vesicular noradrenaline differ in rats from these two suppliers). Whatever caused the variation in binding in the controls, the importance of using a yoked design and paired t test analysis is emphasized. DISCUSSION The results show that 24 h after kindling, fi-adrenoceptor numbers were reduced in both olfactory cortex and neocortex, whereas q-binding decreased only in the olfactory cortex. By 21 days after kindling only the reduction TABLE 3 Neocortical Adrenoceptor Binding 2 1 Days after Kindling“ a2
P
Group
&
B mu
Unhandled control Sham control Kindled
1.3 -t 0.4 3.4 * 0.2* 2.6 2 0.8
68 2 12 106 + 22 87+ 17
Lx
0.9 k 0.2 0.7 + 0.2 0.5 + 0.6*
0 Values show Kd (nM) and B,, (pmol/g * protein), X + SE, N = 5. * P -C 0.05 cf. unhandled control; matched-paired t test.
141 k 19 128 Z!I26 92 f 7*
a*- AND &BINDING
AFTER KINDLING
115
in neocortical P-binding remained. Changes in the affinities of adrenoceptors showed no consistent pattern and included increased affinity of ,&binding sites of the olfactory cortex 24 h after kindling, and at 2 1 days, a reduction of affinity of az-binding sites of the sham controls, and an increase in the affinity of P-binding sites in neocortex of the kindled rats. Regional variation in receptor changes, as seen here, are not unprecedented; the effects of antidepressant treatments, for instance, have also been shown to vary throughout the brain, although all the regions studied receive an input from the nucleus locus coeruleus (23). However, this is the first time that regional differences have been demonstrated within the cerebral cortex in response to a particular treatment. The reduction in seizure threshold caused by kindling is very persistent; under our conditions the stimulus that was initially subconvulsive still triggered generalized convulsions 2 1 days after kindling had been completed. Any neurochemical change directly relevant to the changes in seizure threshold should have a similar time course. Our limited data for olfactory cortex at 2 1 days suggest that adrenoceptor changes were no longer apparent. In the neocortex, on the other hand, P-binding remained significantly below that of the unhandled controls. However, the sham-operated group had an intermediate level of P-binding, and did not differ significantly either from the kindled group or the unhandled controls. This apparently equivocal result suggests that the prolonged presence of the implant in the sham animals can itself cause a down-regulation of ,&adrenoceptors, which may explain why kindling occurs more rapidly after electrodes have been in place for about 28 days (2). Our data thus further support a relationship between the noradrenergic system and the establishment of kindling. McIntyre and Roberts (14) showed that fl-adrenoceptors were still downregulated 23 days after kindling compared with sham-operated controls, whereas we found changes only at 21 days when kindled and unoperated animals were compared. There were several differences in protocol that could explain this apparent discrepancy in detail between the two studies. First, an end point of six stage 5 seizures was used in (14) compared with three used here. Secondly, a daily kindling protocol was used in (14) which involves more handling than our hourly protocol [see (2 1) and Introduction]. Nevertheless, we are in broad agreement that kindling has a persistent effect on neocortical @-adrenoceptors. Another study of this problem by McNamara and colleagues (3, 15), who used a single concentration of [3H]DHA to assess P-binding, also found a decrease at 3 days, but not at 15 h, after kindling. Because our data for 24 h after the last seizure showed a clear down-regulation, it appears that the kindling-induced reduction in the number of fi-adrenoceptors occurs between 15 and 24 h. These findings at 24 h do not support McIntyre and Roberts’s suggestion ( 14) that seizures cause a transient up-regulation of receptors.
116
STANFORD
AND JEFFERYS
It is likely that the down-regulation of adrenoceptors is a consequence of excessive neuronal activity and release of noradrenaline during kindling. The physiologic significance of these changes remains to be established, but possibilities are first, that the persistent changes in neocortical P-binding have a role in the changes in seizure threshold responsible for the kindling phenomenon; second, that the transient changes in binding observed in olfactory cortex may underlie the short-term behavioral consequences of kindling, such as hyperreactivity; and third, that the changes in a,binding, found only in olfactory cortex at 24 h, may account for the sensitivity of this brain region to kindling and for the possible proconvulsant actions of a2 antagonists ( 18). REFERENCES I. ALTMAN, 1. M. AND M. E. CORCORAN. 1983. Facilitation of neocortical kindling by depletion of forebrain noradrenaline. Bruin Res. 270: 174-177. 2. BLACKWOOD, D. H. R., M. J. MARTIN, AND J. K. MCQUEEN. 1982. Enhanced rate of kindling after prolonged electrode implantation into the amygdala of rats. J. Neurosci. Methods 5: 343-348. 3. BYRNE, M. C., R. GO~LIEB, AND J. 0. MCNAMARA. 1980. Amygdala kindling induces muscarinic cholinergic receptor declines in a highly specific distribution within the limbic system. Exp. Neurol. 69: 85-98. 4. CAIN, D. P., AND M. E. CORCORAN. 1978. Kindling in olfactory-lesioned rats. Behav. Biol. 22: 264-268. 5. GODDARD, G. V., D. C. MCINTYRE, AND C. K. LEECH. 1969. A permanent change in brain function resulting from daily electrical stimulation. Exp. Neural. 25: 294-330. 6. JEFFERYS,J. G. R., AND S. C. STANFORD. 1983. Changes in (Y*- and fl-adrenoceptor binding sites in amygdalar kindled rat brain. Sot. Neurosci. Abstr. 9: 487. 7. JEFFERYS,J. G. R., AND S. C. STANFORD. 1984. The effects of amygdalar kindling on LY*and fl-adrenoceptor binding in rat cortex. J. Physiol. (London) 346: 43P. 8. KAIRESS, E. W., R. J. RACINE, AND G. K. SMITH. 1984. The development of the interictal spike during kindling in the rat. Bruin Res. 322: 101-l IO. 9. KAFKA, M. S., A. WIRZ-JUSTICE, AND D. NABER. 1981. Circadian and seasonal rhythms in (Y- and @-adrenergic receptors in the rat brain. Bruin Res. 207: 409-419. 10. KALICHMAN, M. W. 1982. Neurochemical correlates of the kindling model of epilepsy. Neurosci. Biol. Behav. Rev. 6: 165-181. I I. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, AND R. J. RANDALL. I95 I. Protein measurements with the Folin phenol reagent. J. Biol. Chem. 193: 265-273. 12. MCINTYRE, D. C. 1980. Amygdala kindling in rats: facilitation after local amygdala norepinephrine depletion with 6-hydroxydopamine. Exp. Neurol. 69: 395-407. 13. MCINTYRE, D. C., AND N. EDSON. I98 I. Facilation of amygdala kindling alter nompinephrine depletion with 6-hydroxydopamine in rats. Exp. Neural. 74: 148-757. 14. MCINTYRE, D. C., AND D. C. S. ROBERTS. 1983. Long-term reduction in beta-adrenergic receptor binding after amygdala kindling. Exp. Neurol. 82: 17-24. 15. MCNAMARA, J. 0. 1978. Selective alterations of regional adrenergic receptor binding in the kindling model of epilepsy. Exp. Neural. 61: 582-59 I. 16. MUNSON, P. J., AND D. RODBARD. 1980. Ligand: a versatile computerized approach for characterization of ligand binding systems.Anal. B&hem. 107: 220-239.
a?- AND P-BINDING
AFTER KINDLING
117
17. OKAZAKI, M. M., W. M. BURNHAM, AND J. J. WARSH. 1983. Unchanged noradrenaline levels and turnover two months alter amygdalar kindling in the rat. Sot. Neurosci. Abstr. 9: 487. 18. PAPANICOLAOU, J., R. J. SUMMERS, F. J. E. VAJDA, AND W. J. LOUIS. 1982. The relationship between a*-adrenoceptor selectivity and anticonvulsant effect in a series of clonidine-like drugs. Bruin Res. 241: 393-397. 19. PETERSON, S. L., T. E. ALBERTSON, AND L. G. STARK. 1981. Intertrial intervals and kindled seizures. Exp. Neurol. 71: 144-153. 20. RACINE, R. J. 1972. Modification of seizure activity by electrical stimulation. II. Motor seizure. Electroenceph. Clin. Neurophysiol. 32: 28 1-294. 21. STANFORD, S. C., M. FILLENZ, AND E. RYAN. 1984. The effect of repeated mild stress on cerebral cortical adrenoceptors and noradrenaline synthesis in the rat. Neurosci. Left. 45: 163-167.
STANFORD, S. C., AND D. J. NUTT. 1982. Comparison of the effects of repeated electroconvulsive shock on Q- and b-adrenoceptors in different regions of rat brain. Neuroscience 7: 1753-1757. 23. STANFORD C., D. J. Nut-r, AND P. J. COWEN. 1983. Comparison of the effects of chronic desmethylimipramine administration on (Y*-and &adrenoceptors in different regions of rat brain. Neuroscience 8: 16 1- 164. 22.