- Email: [email protected]
Stimulation of GABA release from the rat neostriatum and globus pallidus in vivo by corticotropin-releasing factor
Neuroscience Letters, 100 (1989) 203 209
203
Elsevier Scientific Publishers Ireland Ltd.
NSL 06043
Stimulation of GABA release from the rat neostriatum and globus pallidus in vivo by corticotropin-releasing factor D.J.S. Sirinathsinghji and R.P. Heavens Department of Neuroendocrinology, AFRC Institute of Animal Physiology and Genetics Research, Babraharn, Cambridge ( U.K. ) (Received 16 November 1988; Revised version received 6 January 1989; Accepted 12 January 1989)
Key words." Corticotropin-releasing factor; Corticotropin-releasing factor antagonist; or-Helical CRFg~I; 7-Aminobutyric acid; Neostriatum; Globus pallidus; In vivo push-pull release; Rat This study conducted in vivo examined the changes in ?-aminobutyric acid (GABA) release in push-pull perfusates of the caudate nucleus (CN) and the globus pallidus (GP) in response to corticotropin releasing factor (CRF). In the CN, CRF (10- ~2, 10-10, 10 s M) stimulated GABA release in a dose-related manner, the highest dose (10 8 M) also potentiating the 25 mM K ~-evoked response. The release of GABA in response to CRF (10 8 M) was completely blocked by e-helical CRFg4L (10 6 M ) which also attenuated the K+-evoked response to control K+-stimulated levels. In the GP, only the highest dose of CRF (10 s M) significantly stimulated GABA release, this dose also potentiating the K+-evoked release. Both responses were attenuated by the CRF receptor antagonist (10 6 M). These results thus demonstrate that CRF can exert potent effects on GABA release within the rat neostriatum-pallidum by increasing the membrane excitability of GABA neurons/terminals and that such effects are mediated via receptors present on both the cell bodies/terminals of GABA-containing neurones. These effects of CRF suggest that the peptide may be an integral component of the neurochemical circuitry in the basal ganglia with relevance to the regulation of motor behaviour.
Corticotropin releasing factor (CRF) when infused intracerebroventricularly (i.c.v.) produces a dose-dependent increase in locomotor activity in rats [4, 13, 22]. Although the neuroanatomical or neurochemical basis of this action of CRF has not been established, it is feasible that CRF may produce these effects by acting on neuronal systems within the basal ganglia. Indeed, several lines of evidence indicate that CRF may function as a neurotransmitter within the striatum. For example, both immunoreactive CRF [10, 14, 23] and receptors [(~8, 15] for CRF have been detected in the striatum of the rat. Moreover, CRF has been demonstrated to be released in a calcium-dependent manner from rat striatal slices in vitro in response to potassium depolarization [20] and to stimulate adenylate cyclase activity [5]. In addition, CRFlike immunoreactivity has been shown to be markedly decreased in the striatum of Huntington's disease [9], a neurodegenerative disorder characterized neuropathologiCorrespondence." D.J.S. Sirinathsinghji, Department of Neuroendocrinology, AFRC Institute of Animal Physiology and Genetics Research, Babraham, Cambridge CB2 4AT, U.K. 0304-3940/89/$ 03.50 ~3 1989 Elsevier Scientific Publishers Ireland Ltd.
204 cally by profound cell loss in the caudate-putamen and clinically by severe motor dysfunction, together with complex cognitive and emotional disturbances [I]. If CRF affects locomotor activity by acting at the level of the neostriatum, it is possible that CRF produces these effects either by affecting the release of dopamine (DA) from nigrostriatal DA terminals or by modulating the activity of intrinsic neuronal populations in the neostriatum, e.g. ),-aminobutyric acid (GABA)-containing neurones. Much evidence indicates that GABA is present not only in medium spiny neurones that contribute heavily to the innervation of striatal output sites, the globus pallidus (GP) and the substantia nigra (SN) [12, 21] but also in small to medium aspiny neurons [2, 3] in the striatum. As a logical first step to define the neurochemical basis for a possible action of CRF in the striatum, we conducted experiments to evaluate the influence of CRF on the activity of GABA neurones in the rat striatum. We, thus, used the push-pull perfusion technique to monitor the in vivo release profiles of GABA in both the caudate nucleus (CN) and its projection site, the GP under basal conditions and in response to CRF. In addition, we employed a-helical CRF9 41 a specific CRF receptor antagonist [18], to determine if the observed effects of CRF are indeed mediated via receptors specific for CRF. Adult male Wistar rats (280 320g) were used for the experiments. They were housed in groups of 6 per cage in temperature- (20-22°C), humidity- (approx. 50%) and light - (14 h: 10 h light dark cycle; lights on at 00.05 h) controlled environment. Food and water were freely available. Rats were anaesthetized with chloral hydrate (300 mg/kg i.p. supplemented with 90 mg/kg at 1 h intervals for the duration of the experiment) and positioned in a Kopf stereotaxic instrument. Body temperature was monitored and maintained at 37°C. A push-pull cannula was then implanted into the desired brain site, the anteromedial CN or GP at the following coordinates [16]: CN, AP: + 2.0, L: 3.0 V: 4.5; GP, AP: + 0.8, L: 3.3. V: 6.5. Perfusion of the brain site was immediately initiated at a rate of l0/tl/min with artificial cerebrospinal fluid CSF (pH 7.4). A detailed description of the push-pull procedure is given elsewhere [18]. The first 15 min collection of the perfusate was discarded and successive 15 min samples were continuously collected on ice in Eppendorf microtubes containing 1/11 0.1 M HCIO4 and 1 itl 1/tM EDTA. At the end of each collection, the sample was immediately frozen on dry-ice and stored at - 7ffC until the levels of GABA could be measured by HPLC with fluorescence detection (Perkin Elmer LS1, 340 nm excitation and 450 nm emission filters) following precolumn derivatization with o-phthalaldehyde. The method was based essentially on that of Jarret et al. [11] using a Gilson autoinjector for derivatization and a binary methanol gradient with Gilson pumps controlled by an Apple IIe computer. The column was a 5 l;m C!8 Resolve (Waters) protected with a C18 guard column. Using this system the limit of detection of GABA was 100 fmol. The perfusates were analysed within 1 week of collection. To examine the effects of CRF (Peninsula, Merseyside, U.K.), it was dissolved in the perfusing medium and delivered into the brain site via the push-tube of the cannula in doses of 10 12, 10 10 and 10 s M. Four basal samples were collected, 3 more
205
CRF
0---0
Control
=
~. lO'eM CRF
E
-"
-" lO'mM CRF
,( rn
¢
e 10"1=M CRF
o
L9
/
I
;Sr..
I
I
I
I
I
2
4
6
8
10
B
|
12
a-helicalCRF9.41(lO'eM) + CRF l O e M
y,
•
/,
o---o
a-helicalCRFg.4+ (IO'SM)
A
I
~.
\ \
Control
~ lO'SM CRF
~ a-helicalCRF + lO'SM CRF
E 13. rn
(.9
.
.
.
/ 25ram ,
.
~'~l~l~-O--~=ll
KCI
I
I
I
I
I
|
I
0
2
4
6
8
10
12
Collection Intervals (15 rain)
Fig. 1. The effects of CRF (10 12, 10 ~0and 10 8 M) (A) on the in vivo release of GABA from the caudate nucleus of chloral hydrate-anaesthetized rats. CRF was infused for 45 min (intervals 5 7); 25 mM K + was infused for 45 min (intervals 10-12) in each group. GABA release rates at intervals ~ 8 for CRF 10 i_, and 10 ~0M and at intervals 5 9 for CRF 10 -8 M were significantly higher than control values at corresponding intervals. Notice potentiation of the K +-evoked GABA release following CRF 10 8 M pretreatment. The effects of the antagonist, a-helical C R F g ~ (10 -6 M) on the release of GABA induced by CRF (10 8 M) are shown in the lower diagram (B). Perfusion of the antagonist was initiated 15 min before CRF and continued in the presence of 10 -8 M CRF (45 min). The levels of GABA in the combined presence of the antagonist (10 6 M) and CRF (10 8 M) were significantly lower at intervals 5-10 compared to CRF (10 ~ M) alone. Notice the reduction in the K+-evoked response in the presence of the antagonist ( n = 4 - 5 animals/group; values are mean + S.E.M. "P<0.05, *'P<0.01, ""P<0.001 (one-way analysis of variance and non-paired Student's t-test).
206
in the presence of C R F (45 min duration) followed by 2 basal samples and 3 more in the presence of 25 mM K + (45 min duration). In each experiment a 25 m M K + pulse was applied to verify that the site was functionally active. To examine the effects of a-helical CRF9 41 on the CRF-induced responses a similar protocol was followed except that perfusion of the antagonist (10 ++6 M ) began 15 min before C R F ( 10 s M) and was continued in the presence of CRF. Following completion of the experiments, some of the animals were given a lethal dose of chloral hydrate, the brain removed and fixed in phosphate-buffered formalin. Sections (50 /~m) were then cut in the frontal plane in a freezing microtome and stained with Cresyl violet to verify cannula placement. In some animals, Methylene blue was introduced via the push-pull cannula at the end of the experiment, the animal was given a lethal dose of chloral hydrate and the brain removed for gross examination of the perfusion site (typically an area of 0.54).7 m m diameter). The temporal profiles of the release of G A B A in the CN under basal conditions and in response to C R F and to 25 m M K + are shown in Fig. 1. Push-pull perfusion of the striatum resulted in readily detectable and highly reproducible levels of GABA. In the control animals there was a 4-fold increase in the rate of release of G A B A in response to 25 mM K + compared to basal levels indicating that the sites were functionally active. In addition, this K+-evoked G A B A release was Ca2+-dependent, since no release of G A B A was observed when Ca 2+ was removed from the perfusion medium (to which was added 2 m M EGTA, data not shown). It can be seen from Fig. 1A that C R F produced a clear dose-related increase in G A B A levels in the CN, the highest dose (CRF 10 s M) producing a 6-fold increase in G A B A release. This dose (but not l0 m or l0 12 M) also significantly (P<0.005) enhanced the K ~evoked release compared to the responses in the controls (Fig. 1A). Perfusion of the striatal site with :~-helical CRF9 41 ( 1 0 - 6 M ) completely blocked the release of G A B A induced by C R F (10 8 M) and also normalized the K +-evoked release to stimulated levels seen in the controls (Fig. I B). Under basal conditions, the levels of G A B A in push-pull perfusates of the GP were about 3-fold those observed in the CN. In the control animals there was a 2-fold increase in the rate of release of G A B A in response to 25 mM K ~ again confirming
Fig. 2. The effects of CRF (10 I: 10 u, and 10 ~ M (A) on the in vivo release of GABA from the globus pallidus of chloral hydrate-anaesthetized rats. CRF was infused for 45 rain (intervals 5 7); 25 mM K ~ was infused for 45 min (intervals 10 12) in each group. The release of GABA in response to 10 L2--M CRF and 10 ~u M CRF were not significantly different from controls at any time interval. The GABA release rates in response to l0 8 M CRF were significantly higher at interval 7 compared to the corresponding interval in the controls. Notice potentiation of the K+-evoked GABA release following 10 8 M CRF pretreatment. The effects of the antagonist (10 6 M) on the release of GABA induced by 10+ ~ M CRF are shown in the lower diagram (B). Perfusion of the antagonist was initiated 15 min before CRF and continued in the presence of 10 ~ M CRF (45 min). The levels of GABA in the combined presence of the antagonist (10 6 M) and CRF (10 8 M) were significantly lower at interval 7 compared to CRF ( 10 s M) alone. Notice the reduction in the K+-evoked response in the presence of the antagonist (n = 4 5 animals/group; values are mean + S.E.M. "P<0.05+ "'P<0.01 (one-way analysis of variance and nonpaired Student's t-test).
207
0--0
A
Control
=
¢ lO'SM CRF
,L
i
¢
¢ 10"12M CRF
o
2 ~n
lO-mM CRF
(9
L
l
25 mM KCI I
i
I
I
I
I
2
4
6
8
10
12
B a-helical CRFI.,H (lO'eM)
0 - - - 0 Control
+ CRF IO'=M i.
A C
£:
!
Q.
1
~
/ a-helical CRFI. =
i
(lO-eM)
2 t
O0 < (9
I
6
/
I
2 5 m M KCI I,
|
i
j
l
I
0
2
4
6
8
10
Collection Intervals (15 rain)
I
12
=
~- lO-eM CRF
--
=
a-helical CRF9.41 + lO'SM CRF
208
that the pallidal sites were functionally active. The lower doses of C R F (10 12, 10 l0 M) had no significant effects on pallidal G A B A release (Fig. 2A). The highest dose (10 s M), however produced a significant (P<0.001) (I.4 fold) increase in G A B A levels and also significantly ( P < 0.01) enhanced the K +-evoked G A B A release (Fig. 2A). Perfusion of the pallidal site with ~-helical CRF9 41 (10 6 M) completely blocked the G A B A release induced by C R F (10 s M) and also attenuated the K ~evoked release to control stimulated levels (Fig. 2B). The present studies have clearly demonstrated that C R F exerts potent activating effects on G A B A neurones in the rat neostriatum as seen by the enhanced G A B A release in the CN and to some extent in the GP in response to CRF. Release of G A B A in the CN may come from axonal terminals of G A B A interneurones and collaterals of medium spiny projection neurones while release in the G P may either come from terminals of the latter efferent neurones [21] or from collaterals of pallidal G A B A neurones which send outputs to the subthalamic nucleus and SN [I 7]. The results showing the enhancement of the K+-evoked release with C R F (10 s M) suggest that C R F may increase G A B A efflux by increasing the membrane excitability of G A B A neurones/terminals. The ability of the C R F receptor antagonist, ~-helical CRF9.41 to completely block the release of G A B A induced by C R F (10 8 M) in both the C N and G P and to normalize the K +-evoked response to levels obtained in controls may suggest that C R F receptors are present on cell bodies of G A B A interneurones and/or projection neurones in the CN as well as on terminals of the latter neurones in the GP. However, the inability of low doses of C R F to cause an increase in G A B A release in the GP may favour a suggestion that the density of C R F binding sites in the GP may be lower than that in the CN. High densities of C R F receptors have been found in the rat striatum [6 8, 15] although the specific localization of such receptors in the GP has not been reported. In conclusion, the potent stimulatory effects of C R F on G A B A release in the neostriatum indicate that C R F can have a major influence on an important population of striatal output neurones with possible relevance to the control of motor function and to disorders of movement that originate from the basal ganglia. I Bird. E.D.. Chemical pathology of ttuntington's disease. Annu. Rev. Pharmacol. Toxicol., 20 (1980) 533 55 I. 2 Bolam, J.P., Clarke, D.J., Smith, A.D. and Somogyi, P., A type of aspiny neuron in the rat neostriatum accumulates [~H]-gamma-aminobutyric acid: combination of Golgi-staining, auloradiography and electron microscopy, J. Comp. Neurol., 213 (1983) 121 134. 3 Bolam, J.P., Powell, J.F., Wu, J-Y. and Smith, A.D., Glutamate decarboxylase immunoreactive structures in the rat neostriatum: a correlated light and electron microscopic study including a combination of Golgi impregnation with immunocytochemistry, J. Comp. Neurol., 237 (1985) 121 134. 4 Britton, D.R., Koob, G.F., Rivier, J. and Vale, W., lntraventricular corticotropin-releasing factor enhances behavioural effects of novelty, Life Sci., 31 (1982) 363 367. 5 Chen, F.M., Bilezikjian, L.M., Perrin, M.H., Rivier, J. and Vale, W., C R F receptor-mediated stimulation of adenylate cyclase ativity in rat brain, Brain Res., 381 (1986) 49 57. 6 DeSouza, E,B., Corticotropin-releasing factor receptors in the rat central nervous system: characterization and regional distribution, J. Neurosci., 7 (1987) 88 100.
209 7 DeSouza, E.B., Insel, T.R., Perrin, M.H., Rivier, J., Vale, W.W. and Kuhar, M.J., Corticotropinreleasing factor receptors are widely distributed within the central nervous system: an autoradiographic study, J. Neurosci., 5 (1985) 3189-3203. 8 DeSouza, E.B., Perrin, M.H., Insel, T.R., Rivier, J., Vale, W.W. and Kuhar, M.J., Corticotropinreleasing factor receptors in rat forebrain: autoradiographic identification, Science, 224 (1984) 1449 1451. 9 DeSouza, E.B., Whitehouse, P.J., Folstein, S.E., Price, D.L. and Vale, W.W., Corticotropin-releasing hormone (CRH) is decreased in the basal ganglia in Huntington's disease, Brain Res., 437 (1987) 355 359. 10 Fischman, A.J. and Moldow, R.L., Extrahypothalamic distribution of CRF-like immunoreactivity in the rat brain, Peptides, 1 (1982) 149 153. 11 Jarret, H.W., Cooksy, K.D., Ellis, B. and Andersen, J.M., The separation of o-phthalaldehyde derivatives of amino acids by reverse phase chromatography on octylsilica columns. Anal. Biochem., 153 (1986) 189 198. 12 Jessel, T.M., Emson, P.C., Paxinos, C. and Cuello, A.C., Topographic projections of substance P and GABA pathways in the striato- and pallido-nigral system: a biochemical and immunohistochemical study, Brain Res., 152 (1978)487-498. 13 Koob, G.F., Swerdlow, N., Seeligson, M., Eaves, M., Sutton, R., Rivier, J. and Vale, W., Effects of alpha-flupenthixol and naloxone on CRF-induced locomotor activation, Neuroendocrinology, 39 (I 984) 459-464. 14 Merchenthaler, I., Corticotropin releasing factor (CRF)-like immunoreactivity in the rat central nervous system. Extrahypothalamic distribution, Peptides, 5 (1984) 53~59. 15 Millan, M.A., Jacobowitz, D.M., Hauger, R.L., Cart, K.J. and Aguilera, G., Distribution of corticotropin-releasing factor receptors in primate brain, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 1921 1925. 16 Pelligrino, L.J., Pelligrino, A.S. and Cushman, A.J., A Stereotaxic Atlas of the Rat Brain, 2nd end., Plenum, New York, 1979. 17 Pycock, C.J. and Phillipson, O.T., A neuroanatomical and neuropharmacological analysis of basal ganglia output. In L.~ !ve:sen, S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 18, Plenum, New York, 1984, pp. 191 278. 18 Rivier, J., Rivier, C. and Vale, W., Synthetic competitive antagonists of corticotropin releasing factor: effect on ACTH secretion in the rat, Science, 224 (1984) 889 891. 19 Sirinathsinghji, D.J.S., Heavens, R.P. and McBride, C.S., Dopamine-releasing action of 1-methyl-4phenyll,2,3,6-tetrahydropyridine (MPTP) and I-methyl-4-phenylpyridine (MPP ÷) in the neostriatum of the rat as demonstrated in vivo by the push-pull perfusion technique: dependence on sodium but not calcium ions, Brain Res., 443 (1988) 101 116. 20 Smith, M.A., Bissette, G., Slotkin, T.A., Knight, D.L. and Nemeroff, C.B., Release of corticotropinreleasing factor from rat brain regions in vitro, Endocrinology, 118 (1986) 1997-2001. 21 Somogyi, P. and Smith, A.D., Projection of neostriatal spiny neurons to the substantia nigra. Application of a combined Golgi-staining and horseradish peroxidase transport procedure at both light and electron microscope levels, Brain Res., 178 (1979) 3-15. 22 Sutton, R.E., Koob, G.F., LeMoal, M., Rivier, J. and Vale, W., Corticotropin releasing factor produces behavioural activation in rats, Nature (Lond.), 297 (1982) 331-333. 23 Swanson, L.W., Sawchenko, P.E., Rivier, J. and Vale, W., Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study, Neuroendocrinology, 36 (1983) 164-186.
203
Elsevier Scientific Publishers Ireland Ltd.
NSL 06043
Stimulation of GABA release from the rat neostriatum and globus pallidus in vivo by corticotropin-releasing factor D.J.S. Sirinathsinghji and R.P. Heavens Department of Neuroendocrinology, AFRC Institute of Animal Physiology and Genetics Research, Babraharn, Cambridge ( U.K. ) (Received 16 November 1988; Revised version received 6 January 1989; Accepted 12 January 1989)
Key words." Corticotropin-releasing factor; Corticotropin-releasing factor antagonist; or-Helical CRFg~I; 7-Aminobutyric acid; Neostriatum; Globus pallidus; In vivo push-pull release; Rat This study conducted in vivo examined the changes in ?-aminobutyric acid (GABA) release in push-pull perfusates of the caudate nucleus (CN) and the globus pallidus (GP) in response to corticotropin releasing factor (CRF). In the CN, CRF (10- ~2, 10-10, 10 s M) stimulated GABA release in a dose-related manner, the highest dose (10 8 M) also potentiating the 25 mM K ~-evoked response. The release of GABA in response to CRF (10 8 M) was completely blocked by e-helical CRFg4L (10 6 M ) which also attenuated the K+-evoked response to control K+-stimulated levels. In the GP, only the highest dose of CRF (10 s M) significantly stimulated GABA release, this dose also potentiating the K+-evoked release. Both responses were attenuated by the CRF receptor antagonist (10 6 M). These results thus demonstrate that CRF can exert potent effects on GABA release within the rat neostriatum-pallidum by increasing the membrane excitability of GABA neurons/terminals and that such effects are mediated via receptors present on both the cell bodies/terminals of GABA-containing neurones. These effects of CRF suggest that the peptide may be an integral component of the neurochemical circuitry in the basal ganglia with relevance to the regulation of motor behaviour.
Corticotropin releasing factor (CRF) when infused intracerebroventricularly (i.c.v.) produces a dose-dependent increase in locomotor activity in rats [4, 13, 22]. Although the neuroanatomical or neurochemical basis of this action of CRF has not been established, it is feasible that CRF may produce these effects by acting on neuronal systems within the basal ganglia. Indeed, several lines of evidence indicate that CRF may function as a neurotransmitter within the striatum. For example, both immunoreactive CRF [10, 14, 23] and receptors [(~8, 15] for CRF have been detected in the striatum of the rat. Moreover, CRF has been demonstrated to be released in a calcium-dependent manner from rat striatal slices in vitro in response to potassium depolarization [20] and to stimulate adenylate cyclase activity [5]. In addition, CRFlike immunoreactivity has been shown to be markedly decreased in the striatum of Huntington's disease [9], a neurodegenerative disorder characterized neuropathologiCorrespondence." D.J.S. Sirinathsinghji, Department of Neuroendocrinology, AFRC Institute of Animal Physiology and Genetics Research, Babraham, Cambridge CB2 4AT, U.K. 0304-3940/89/$ 03.50 ~3 1989 Elsevier Scientific Publishers Ireland Ltd.
204 cally by profound cell loss in the caudate-putamen and clinically by severe motor dysfunction, together with complex cognitive and emotional disturbances [I]. If CRF affects locomotor activity by acting at the level of the neostriatum, it is possible that CRF produces these effects either by affecting the release of dopamine (DA) from nigrostriatal DA terminals or by modulating the activity of intrinsic neuronal populations in the neostriatum, e.g. ),-aminobutyric acid (GABA)-containing neurones. Much evidence indicates that GABA is present not only in medium spiny neurones that contribute heavily to the innervation of striatal output sites, the globus pallidus (GP) and the substantia nigra (SN) [12, 21] but also in small to medium aspiny neurons [2, 3] in the striatum. As a logical first step to define the neurochemical basis for a possible action of CRF in the striatum, we conducted experiments to evaluate the influence of CRF on the activity of GABA neurones in the rat striatum. We, thus, used the push-pull perfusion technique to monitor the in vivo release profiles of GABA in both the caudate nucleus (CN) and its projection site, the GP under basal conditions and in response to CRF. In addition, we employed a-helical CRF9 41 a specific CRF receptor antagonist [18], to determine if the observed effects of CRF are indeed mediated via receptors specific for CRF. Adult male Wistar rats (280 320g) were used for the experiments. They were housed in groups of 6 per cage in temperature- (20-22°C), humidity- (approx. 50%) and light - (14 h: 10 h light dark cycle; lights on at 00.05 h) controlled environment. Food and water were freely available. Rats were anaesthetized with chloral hydrate (300 mg/kg i.p. supplemented with 90 mg/kg at 1 h intervals for the duration of the experiment) and positioned in a Kopf stereotaxic instrument. Body temperature was monitored and maintained at 37°C. A push-pull cannula was then implanted into the desired brain site, the anteromedial CN or GP at the following coordinates [16]: CN, AP: + 2.0, L: 3.0 V: 4.5; GP, AP: + 0.8, L: 3.3. V: 6.5. Perfusion of the brain site was immediately initiated at a rate of l0/tl/min with artificial cerebrospinal fluid CSF (pH 7.4). A detailed description of the push-pull procedure is given elsewhere [18]. The first 15 min collection of the perfusate was discarded and successive 15 min samples were continuously collected on ice in Eppendorf microtubes containing 1/11 0.1 M HCIO4 and 1 itl 1/tM EDTA. At the end of each collection, the sample was immediately frozen on dry-ice and stored at - 7ffC until the levels of GABA could be measured by HPLC with fluorescence detection (Perkin Elmer LS1, 340 nm excitation and 450 nm emission filters) following precolumn derivatization with o-phthalaldehyde. The method was based essentially on that of Jarret et al. [11] using a Gilson autoinjector for derivatization and a binary methanol gradient with Gilson pumps controlled by an Apple IIe computer. The column was a 5 l;m C!8 Resolve (Waters) protected with a C18 guard column. Using this system the limit of detection of GABA was 100 fmol. The perfusates were analysed within 1 week of collection. To examine the effects of CRF (Peninsula, Merseyside, U.K.), it was dissolved in the perfusing medium and delivered into the brain site via the push-tube of the cannula in doses of 10 12, 10 10 and 10 s M. Four basal samples were collected, 3 more
205
CRF
0---0
Control
=
~. lO'eM CRF
E
-"
-" lO'mM CRF
,( rn
¢
e 10"1=M CRF
o
L9
/
I
;Sr..
I
I
I
I
I
2
4
6
8
10
B
|
12
a-helicalCRF9.41(lO'eM) + CRF l O e M
y,
•
/,
o---o
a-helicalCRFg.4+ (IO'SM)
A
I
~.
\ \
Control
~ lO'SM CRF
~ a-helicalCRF + lO'SM CRF
E 13. rn
(.9
.
.
.
/ 25ram ,
.
~'~l~l~-O--~=ll
KCI
I
I
I
I
I
|
I
0
2
4
6
8
10
12
Collection Intervals (15 rain)
Fig. 1. The effects of CRF (10 12, 10 ~0and 10 8 M) (A) on the in vivo release of GABA from the caudate nucleus of chloral hydrate-anaesthetized rats. CRF was infused for 45 min (intervals 5 7); 25 mM K + was infused for 45 min (intervals 10-12) in each group. GABA release rates at intervals ~ 8 for CRF 10 i_, and 10 ~0M and at intervals 5 9 for CRF 10 -8 M were significantly higher than control values at corresponding intervals. Notice potentiation of the K +-evoked GABA release following CRF 10 8 M pretreatment. The effects of the antagonist, a-helical C R F g ~ (10 -6 M) on the release of GABA induced by CRF (10 8 M) are shown in the lower diagram (B). Perfusion of the antagonist was initiated 15 min before CRF and continued in the presence of 10 -8 M CRF (45 min). The levels of GABA in the combined presence of the antagonist (10 6 M) and CRF (10 8 M) were significantly lower at intervals 5-10 compared to CRF (10 ~ M) alone. Notice the reduction in the K+-evoked response in the presence of the antagonist ( n = 4 - 5 animals/group; values are mean + S.E.M. "P<0.05, *'P<0.01, ""P<0.001 (one-way analysis of variance and non-paired Student's t-test).
206
in the presence of C R F (45 min duration) followed by 2 basal samples and 3 more in the presence of 25 mM K + (45 min duration). In each experiment a 25 m M K + pulse was applied to verify that the site was functionally active. To examine the effects of a-helical CRF9 41 on the CRF-induced responses a similar protocol was followed except that perfusion of the antagonist (10 ++6 M ) began 15 min before C R F ( 10 s M) and was continued in the presence of CRF. Following completion of the experiments, some of the animals were given a lethal dose of chloral hydrate, the brain removed and fixed in phosphate-buffered formalin. Sections (50 /~m) were then cut in the frontal plane in a freezing microtome and stained with Cresyl violet to verify cannula placement. In some animals, Methylene blue was introduced via the push-pull cannula at the end of the experiment, the animal was given a lethal dose of chloral hydrate and the brain removed for gross examination of the perfusion site (typically an area of 0.54).7 m m diameter). The temporal profiles of the release of G A B A in the CN under basal conditions and in response to C R F and to 25 m M K + are shown in Fig. 1. Push-pull perfusion of the striatum resulted in readily detectable and highly reproducible levels of GABA. In the control animals there was a 4-fold increase in the rate of release of G A B A in response to 25 mM K + compared to basal levels indicating that the sites were functionally active. In addition, this K+-evoked G A B A release was Ca2+-dependent, since no release of G A B A was observed when Ca 2+ was removed from the perfusion medium (to which was added 2 m M EGTA, data not shown). It can be seen from Fig. 1A that C R F produced a clear dose-related increase in G A B A levels in the CN, the highest dose (CRF 10 s M) producing a 6-fold increase in G A B A release. This dose (but not l0 m or l0 12 M) also significantly (P<0.005) enhanced the K ~evoked release compared to the responses in the controls (Fig. 1A). Perfusion of the striatal site with :~-helical CRF9 41 ( 1 0 - 6 M ) completely blocked the release of G A B A induced by C R F (10 8 M) and also normalized the K +-evoked release to stimulated levels seen in the controls (Fig. I B). Under basal conditions, the levels of G A B A in push-pull perfusates of the GP were about 3-fold those observed in the CN. In the control animals there was a 2-fold increase in the rate of release of G A B A in response to 25 mM K ~ again confirming
Fig. 2. The effects of CRF (10 I: 10 u, and 10 ~ M (A) on the in vivo release of GABA from the globus pallidus of chloral hydrate-anaesthetized rats. CRF was infused for 45 rain (intervals 5 7); 25 mM K ~ was infused for 45 min (intervals 10 12) in each group. The release of GABA in response to 10 L2--M CRF and 10 ~u M CRF were not significantly different from controls at any time interval. The GABA release rates in response to l0 8 M CRF were significantly higher at interval 7 compared to the corresponding interval in the controls. Notice potentiation of the K+-evoked GABA release following 10 8 M CRF pretreatment. The effects of the antagonist (10 6 M) on the release of GABA induced by 10+ ~ M CRF are shown in the lower diagram (B). Perfusion of the antagonist was initiated 15 min before CRF and continued in the presence of 10 ~ M CRF (45 min). The levels of GABA in the combined presence of the antagonist (10 6 M) and CRF (10 8 M) were significantly lower at interval 7 compared to CRF ( 10 s M) alone. Notice the reduction in the K+-evoked response in the presence of the antagonist (n = 4 5 animals/group; values are mean + S.E.M. "P<0.05+ "'P<0.01 (one-way analysis of variance and nonpaired Student's t-test).
207
0--0
A
Control
=
¢ lO'SM CRF
,L
i
¢
¢ 10"12M CRF
o
2 ~n
lO-mM CRF
(9
L
l
25 mM KCI I
i
I
I
I
I
2
4
6
8
10
12
B a-helical CRFI.,H (lO'eM)
0 - - - 0 Control
+ CRF IO'=M i.
A C
£:
!
Q.
1
~
/ a-helical CRFI. =
i
(lO-eM)
2 t
O0 < (9
I
6
/
I
2 5 m M KCI I,
|
i
j
l
I
0
2
4
6
8
10
Collection Intervals (15 rain)
I
12
=
~- lO-eM CRF
--
=
a-helical CRF9.41 + lO'SM CRF
208
that the pallidal sites were functionally active. The lower doses of C R F (10 12, 10 l0 M) had no significant effects on pallidal G A B A release (Fig. 2A). The highest dose (10 s M), however produced a significant (P<0.001) (I.4 fold) increase in G A B A levels and also significantly ( P < 0.01) enhanced the K +-evoked G A B A release (Fig. 2A). Perfusion of the pallidal site with ~-helical CRF9 41 (10 6 M) completely blocked the G A B A release induced by C R F (10 s M) and also attenuated the K ~evoked release to control stimulated levels (Fig. 2B). The present studies have clearly demonstrated that C R F exerts potent activating effects on G A B A neurones in the rat neostriatum as seen by the enhanced G A B A release in the CN and to some extent in the GP in response to CRF. Release of G A B A in the CN may come from axonal terminals of G A B A interneurones and collaterals of medium spiny projection neurones while release in the G P may either come from terminals of the latter efferent neurones [21] or from collaterals of pallidal G A B A neurones which send outputs to the subthalamic nucleus and SN [I 7]. The results showing the enhancement of the K+-evoked release with C R F (10 s M) suggest that C R F may increase G A B A efflux by increasing the membrane excitability of G A B A neurones/terminals. The ability of the C R F receptor antagonist, ~-helical CRF9.41 to completely block the release of G A B A induced by C R F (10 8 M) in both the C N and G P and to normalize the K +-evoked response to levels obtained in controls may suggest that C R F receptors are present on cell bodies of G A B A interneurones and/or projection neurones in the CN as well as on terminals of the latter neurones in the GP. However, the inability of low doses of C R F to cause an increase in G A B A release in the GP may favour a suggestion that the density of C R F binding sites in the GP may be lower than that in the CN. High densities of C R F receptors have been found in the rat striatum [6 8, 15] although the specific localization of such receptors in the GP has not been reported. In conclusion, the potent stimulatory effects of C R F on G A B A release in the neostriatum indicate that C R F can have a major influence on an important population of striatal output neurones with possible relevance to the control of motor function and to disorders of movement that originate from the basal ganglia. I Bird. E.D.. Chemical pathology of ttuntington's disease. Annu. Rev. Pharmacol. Toxicol., 20 (1980) 533 55 I. 2 Bolam, J.P., Clarke, D.J., Smith, A.D. and Somogyi, P., A type of aspiny neuron in the rat neostriatum accumulates [~H]-gamma-aminobutyric acid: combination of Golgi-staining, auloradiography and electron microscopy, J. Comp. Neurol., 213 (1983) 121 134. 3 Bolam, J.P., Powell, J.F., Wu, J-Y. and Smith, A.D., Glutamate decarboxylase immunoreactive structures in the rat neostriatum: a correlated light and electron microscopic study including a combination of Golgi impregnation with immunocytochemistry, J. Comp. Neurol., 237 (1985) 121 134. 4 Britton, D.R., Koob, G.F., Rivier, J. and Vale, W., lntraventricular corticotropin-releasing factor enhances behavioural effects of novelty, Life Sci., 31 (1982) 363 367. 5 Chen, F.M., Bilezikjian, L.M., Perrin, M.H., Rivier, J. and Vale, W., C R F receptor-mediated stimulation of adenylate cyclase ativity in rat brain, Brain Res., 381 (1986) 49 57. 6 DeSouza, E,B., Corticotropin-releasing factor receptors in the rat central nervous system: characterization and regional distribution, J. Neurosci., 7 (1987) 88 100.
209 7 DeSouza, E.B., Insel, T.R., Perrin, M.H., Rivier, J., Vale, W.W. and Kuhar, M.J., Corticotropinreleasing factor receptors are widely distributed within the central nervous system: an autoradiographic study, J. Neurosci., 5 (1985) 3189-3203. 8 DeSouza, E.B., Perrin, M.H., Insel, T.R., Rivier, J., Vale, W.W. and Kuhar, M.J., Corticotropinreleasing factor receptors in rat forebrain: autoradiographic identification, Science, 224 (1984) 1449 1451. 9 DeSouza, E.B., Whitehouse, P.J., Folstein, S.E., Price, D.L. and Vale, W.W., Corticotropin-releasing hormone (CRH) is decreased in the basal ganglia in Huntington's disease, Brain Res., 437 (1987) 355 359. 10 Fischman, A.J. and Moldow, R.L., Extrahypothalamic distribution of CRF-like immunoreactivity in the rat brain, Peptides, 1 (1982) 149 153. 11 Jarret, H.W., Cooksy, K.D., Ellis, B. and Andersen, J.M., The separation of o-phthalaldehyde derivatives of amino acids by reverse phase chromatography on octylsilica columns. Anal. Biochem., 153 (1986) 189 198. 12 Jessel, T.M., Emson, P.C., Paxinos, C. and Cuello, A.C., Topographic projections of substance P and GABA pathways in the striato- and pallido-nigral system: a biochemical and immunohistochemical study, Brain Res., 152 (1978)487-498. 13 Koob, G.F., Swerdlow, N., Seeligson, M., Eaves, M., Sutton, R., Rivier, J. and Vale, W., Effects of alpha-flupenthixol and naloxone on CRF-induced locomotor activation, Neuroendocrinology, 39 (I 984) 459-464. 14 Merchenthaler, I., Corticotropin releasing factor (CRF)-like immunoreactivity in the rat central nervous system. Extrahypothalamic distribution, Peptides, 5 (1984) 53~59. 15 Millan, M.A., Jacobowitz, D.M., Hauger, R.L., Cart, K.J. and Aguilera, G., Distribution of corticotropin-releasing factor receptors in primate brain, Proc. Natl. Acad. Sci. U.S.A., 83 (1986) 1921 1925. 16 Pelligrino, L.J., Pelligrino, A.S. and Cushman, A.J., A Stereotaxic Atlas of the Rat Brain, 2nd end., Plenum, New York, 1979. 17 Pycock, C.J. and Phillipson, O.T., A neuroanatomical and neuropharmacological analysis of basal ganglia output. In L.~ !ve:sen, S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 18, Plenum, New York, 1984, pp. 191 278. 18 Rivier, J., Rivier, C. and Vale, W., Synthetic competitive antagonists of corticotropin releasing factor: effect on ACTH secretion in the rat, Science, 224 (1984) 889 891. 19 Sirinathsinghji, D.J.S., Heavens, R.P. and McBride, C.S., Dopamine-releasing action of 1-methyl-4phenyll,2,3,6-tetrahydropyridine (MPTP) and I-methyl-4-phenylpyridine (MPP ÷) in the neostriatum of the rat as demonstrated in vivo by the push-pull perfusion technique: dependence on sodium but not calcium ions, Brain Res., 443 (1988) 101 116. 20 Smith, M.A., Bissette, G., Slotkin, T.A., Knight, D.L. and Nemeroff, C.B., Release of corticotropinreleasing factor from rat brain regions in vitro, Endocrinology, 118 (1986) 1997-2001. 21 Somogyi, P. and Smith, A.D., Projection of neostriatal spiny neurons to the substantia nigra. Application of a combined Golgi-staining and horseradish peroxidase transport procedure at both light and electron microscope levels, Brain Res., 178 (1979) 3-15. 22 Sutton, R.E., Koob, G.F., LeMoal, M., Rivier, J. and Vale, W., Corticotropin releasing factor produces behavioural activation in rats, Nature (Lond.), 297 (1982) 331-333. 23 Swanson, L.W., Sawchenko, P.E., Rivier, J. and Vale, W., Organization of ovine corticotropin-releasing factor immunoreactive cells and fibers in the rat brain: an immunohistochemical study, Neuroendocrinology, 36 (1983) 164-186.