Enhancement of central transmission to sympathetic preganglionic neurons by phosphodiesterase inhibitors and its prevention by clonidine

Journal of the Autonomic" Nervous System, 19 (1987) 199-209 Elsevier

199

JAN 00723

Enhancement of central transmission to sympathetic preganglionic neurons by phosphodiesterase inhibitors and its prevention by clonidine Donald N. Franz, Robert G. Peterson, Parley W. Madsen and Chaichan Sangdee * Department of Pharmacoloy2v, University of Utah School of Medicine, Salt Lake City', UT84132 (U.S.A.) (Received 2 May 1986) (Revised version received 7 November 1986) (Accepted 3 February 1987)

Key words: Adenylate cyclase; a2-Adrenergic receptor; Clonidine; Cyclic AMP: Phosphodiesterase inhibitor; Sympathetic preganglionic neuron

Summa~ The effects of 3 phosphodiesterase inhibitors, aminophylline, isobutylmethylxanthine (IBMX), and RO 20-1724, were tested on descending intraspinal and spinal reflex transmission to sympathetic preganglionic neurons in unanesthetized spinal cats. Sympathetic discharges, recorded from upper thoracic preganglionic white rami, were evoked by stimulation (0A Hz) of descending excitatory pathways in the cervical dorsolateral funiculus (imraspinal) or of adjacent intercostal nerves (spinal reflex). Each phosphodiesterase rapidly and markedly enhanced transmission through intraspinal pathways but only slowly and modestly enhanced transmission through spinal reflex pathways. Pretreatment with a methyltyrosine-reserpine combination, chlorpromazine, or prazosin markedly restricted the enhancement of intraspinal transmission by IBMX to levels typically produced on spinal refle~ pathways. Clonidine markedly depressed transmission through both pathways and prevented enhancement by the phosphodiesterase inhibitors. Yohimbine or tolazoline antagonized the depressant effects of clonidine and restored the ability of the phosphodiesterase inhibitors to enhance transmission. Somatic spinal reflexes were not affected by the phosphodiesterase inhibitors. The results suggest that descending norepinephrine pathways to sympathetic preganglionic neurons activate adenylate cyclase to generate cyclic AMI' which increases neuronal excitability, possibly by phosphorylating membrane proteins. Clonidine appears to depress neuronal excitability by inhibiting adenylate cyclase through activation of ~2-adrenergic receptors.

Introduction Sympathetic preganglionic neurons in the intermediolateral columns of the spinal cord receive a

* Present address: Department of Pharmacology, Faculty of Medicine, Chiang Mai University, Chiang Mai 50000, Thailand Correspondence: D.N. Franz, Department of Pharmacology, University of Utah Medical Center, Salt Lake City, UT 84132, U.S.A.

dense innervation of norepinephrine terminals [1,8,23,30,49] and a lesser but exclusive innervation of epinephrine terminals [20,26,40,42] from neurons associated with autonomic centers in the brainstem. The respective functional roles of these catecholamine pathways have not been resolved because conflicting evidence has been advanced for both excitation and inhibition by each pathway [5,10,18,20,31,40]. The apparent innervation by dopamine pathways [48] adds a further complication to interpretations of the available evidence.

0165-183g/87/$03.50 ','~'1987 Elsevier Science Publishers B.V. (Biomedical Division)

20{)

Our own previous findings that descending excitatory transmission to sympathetic preganglionic neurons is rapidly and markedly enhanced by desipramine [43] or dextroamphetamine [45], and is markedly reduced but not abolished by chlorpromazine [38] and by combined reserpine-methyltyrosine pretreatment [14] suggest that norepinephrine pathways are excitatory. However, this conclusion seems to conflict with the markedly inhibitory effects of systemically administered clonidine on spinal sympathetic transmission [2,3,13,15,16,29,47] and of iontophoretically applied clonidine and catecholamines to sympathetic preganglionic neurons [7,11,24,25], all of which are antagonized by a2-receptor antagonists. A possible resolution to this apparent dichotomy was recently proposed from our findings that clonidine prevents the marked enhancement of intraspinal transmission produced by aminophylline and isobutylmethylxanthine (IBMX), two xanthine phosphodiesterase inhibitors which inhibit the catabolism of cyclic AMP (cAMP) [17,18]. These results suggest that the excitability of sympathetic preganglionic neurons is regulated by intraneuronal cAMP through activation of adenylate cyclase by excitatory norepinephrine pathways or suppression of adenylate cyclase by inhibitory epinephrine pathways which, like ctonidine, act on postsynaptic az-receptors. Although aminophylline and IBMX inhibit cAMP phosphodiesterase, they both inhibit cyclic G M P phosphodiesterase as well [9,56] and also antagonize the neuronal effects of adenosine [9,27,56]. Therefore, the present study was designed to compare the effects of the two xanthines with those of a more selective, non-xanthine phosphodiesterase inhibitor, RO 20-1724, which does not inhibit cyclic GMP phosphodiesterase [46] or block adenosine receptors [27,53]. Each of the 3 inhibitors was tested on both intraspinal and spinal reflex pathways to sympathetic preganglionic neurons in order to detect possible differences in their respective effects on transmission. In addition, the typical effects of IBMX on intraspinal transmission were compared with those determined after depletion or antagonism of central catecholamines to assess the role of intraspinal norepinephrine or dopamine pathways in mediat-

ing the enhanced transmission during inhibition of phosphodiesterase. The interactions of clonidine with each of the 3 phosphodiesterase inhibitors were also examined.

Materials and Methods

Surgical procedures Experiments were conducted on adult spinal cats of either sex weighing 2.2-4.0 kg. Under brief ether or methohexitat anesthesia, the trachea was cannulated, the right common carotid artery was ligated and the left common carotid artery was cannulated for continuous monitoring of central blood pressure. The vertebral arteries were occluded at C 2 and the spinal cord was transected at C 1 level. Ventilation was sustained with a respirator adjusted to maintain end-expiratory CO 2 concentration between 3 and 4% (Spinco Medical Gas Analyzer). A cephalic vein was cannulated for administration of drugs. Arterial blood pressure usually stabilized between 70 and 90 mm Hg. Small volumes of 10% dextran-40 were infused as necessary to maintain mean blood pressure above 70 mm Hg. Body temperature was maintained between 36 and 38°C by an automatically regulated heating plate. Gallamine triethiodide (Flaxedil) was administered as needed to provide muscle paralysis throughout experiments. The left thoracic sympathetic preganglionic rami communicantes and intercostal nerves of segments T2, T 3, and T 4 were exposed by removal of overlying muscle and ribs. Care was taken to leave

SPINALREFLEX

A

( ' - ~ - ~ ~ T R A SPINAL

B

Fig. 1. Diagrammatic representations of spinal reflex (A) and intraspinal pathways (B) to sympathetic preganglionic neurons (SPGN) showing positions of stimulating and recording electrodes and locations of bulbospinal norepinephrine (NE) and epinephrine (EPI) pathways. Other descending monoaminergic, neuropeptide, and amino acid pathways are not shown.

201

the parietal pleural membrane intact to serve as the floor of the recording pool. Intercostal nerves were sectioned approximately 2 cm distal to the vertebral column and were dissected free of adjacent tissue to a point just proximal to the sympathetic trunk. Under warm mineral oil, sympathetic preganglionic rami were carefully cleaned of remaining connective tissue between their merger with the sympathetic trunk and their juncture with a corresponding spinal nerve, and were sectioned proximal to the sympathetic chain. An ipsilateral pneumothorax and a rigid blunt retainer positioned against ribs and exposed pleural membrane minimized respiratory movement in the vicinity of the exposed nerves. In experiments depending upon intraspinal stimulation, a dorsolateral laminectomy was performed to expose the spinal cord from C a to C 3. The dura was opened and reflected to expose the dorsal roots which were sectioned distally and displaced medially. In 6 experiments, two phosphodiesterase inhibitors were tested on spinal somatic monosynaptic and polysynaptic reflexes evoked in lumbar ventral roots by stimulation of adjacent dorsal roots exposed by laminectomy. Exposed tissues were covered with warm mineral oil. Spinal sympathetic reflexes were evoked by stimulation of small myelinated afferent fibers in one of the freed intercostal nerves with supramaximal rectangular pulses (0.2 ms; 1-3 V) delivered through bipolar silver-wire electrodes from an isolated stimulator (Devices, type 2533) at a frequency of 0.2 Hz (Fig. 1A). Intraspinal pathways in the ipsilateral dorsolateral funiculus of the cervical cord were stimulated with bipolar tungsten microelectrodes (tip exposure, 25 >V; 3-9 MS2; separation, 1 ram) oriented vertically in the sagittal plane (Fig. 1B). Maximal responses were evoked with biphasic pulses (0.2 ms) at a frequency of 0.1 Hz and at stimulus intensities between 50 and 150 ~A (W-P Bipolar stimulator, Model 601). The use of biphasic pulses at a low frequency of stimulation conferred good stability of responses for many hours. Evoked sympathetic discharges were recorded with bipolar silver-wire electrodes from one or two of the exposed preganglionic rami. Evoked re-

sponses were amplified and displayed on a dualbeam oscilloscope for continuous monitoring. Response sizes were measured on-line by integrating 16 consecutive responses with a signal averaging computer (Nicolet 1072). Representative individual discharges and the corresponding analog computer displays were photographed for permanent records.

Experimental procedures At least 6 h elapsed between spinal cord transection and the beginning of recording. Two additional hours were utilized for a control period during which evoked responses were sampled periodically to insure their stability and to generate data for calculating a mean control value and its variance. In the absence of drug treatment, evoked sympathetic discharges routinely remained within _+2 Standard Deviations ( < + 10%) of mean control values for more than 5 h. Therefore, changes occurring after drug administration were considered significant when consecutive readings fell outside this range. Following drug administration, evoked activity was sampled at 5- or 10-rain intervals. Absolute sizes of evoked discharges generated by the averaging computer were converted to percentages of mean control values to standardize data for comparison among different experiments. Although some quantitative differences among individual experiments were noted, drug effects were qualitatively consistent and variations were minor. Some variability among individual experiments may have been due to relative differences in sensitivity or proportion of preganglionic neurons available for discharge. These procedures were also followed in experiments on somatic spinal reflexes.

Drugs Unless otherwise specified, all drugs were administered intravenously. Drug solutions were prepared with warm 0.9% NaCI or commercially available parenteral forms were used. Phosphodiesterase inhibitors were routinely injected slowly during a 5-min period to minimize vasodilatation and tachycardia. Other drugs were administered within 2 rain. Sources of drugs used: gallamine triethiodide, Davis-Geck; aminophylline,

202

Abbott; theophylline, Sigma; isobutylmethylxanthine (IBMX), Aldrich; RO 20-1724, Roche; chlorpromazine HC1, Smith-Kline; alphamethyltyrosine methylester HCI, Regis; prazosin HC1, Pfizer; clonidine HC1, Bohringer; tolazoline HCI. Ciba; yohimbine HC1, Aldrich.

8

A 200-

1724

g ~00• 50 m~lkq T4r o 25 rag/kg X3 r

i Omg/kg T4f 0 5 mq/kg T2r

0

0

]

I

I

I

2

5

0

2

5

i

C 3

• bO mg/kg TZr o 0 5 mg/kg T3r

2

3

TI ME HOURS

Effects of phosphodiesterase inhibitors on intraspinally evoked sympathetic discharges Intravenous administration of aminophylline (25-50) mg/kg), isobutylmethylxanthine (IBMX; 0.5-1 mg/kg), or RO 20-1724 (0.5-1 mg/kg; 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone) consistently produced marked enhancement of transmission through intraspinal excitatory pathways to sympathetic preganglionic neurons (SPGNs). The onset of enhancement was rapid, becoming prominent within 5 min and reaching a maximum within 15-30 rain (Fig. 2). The larger doses of each drug produced greater enhancement than the smaller doses, after which responses gradually returned to control levels by 2-3 h (Fig. 3). After termination of enhancement by the larger doses of each drug, a second dose produced only modest or no enhancement.

CONTROL

RO 2C

~ 150-

5O C

Results

C IBMX

AMINOPHYLLINE

DRUG rng/kg phylline

TIME-MINUTES ~, " ~ J ~

.i

The average time courses of enhancement produced by 50 m g / k g of aminophylline, 1 m g / k g of IBMX, or 1 m g / k g of RO 20-1724 (Fig: 4) show that these respective doses produced about the same degree of enhancement with similar rates of onset. The duration of enhancement produced by IBMX and RO 20-1724 was somewhat longer than that produced by aminophylline. At the doses tested, none of the 3 phosphodiesterase inhibitors provoked spontaneous sympathetic activity during enhancement or depressed transmission after recovery. Equimolar doses of theophylline (40 mg/kg) enhanced intraspinal transmission to the same extent as produced by 50 m g / k g of aminophylline.

Effects of phosphodiesterase inhibitors on spinal sympathetic reflexes

30

L ,~ 6o

Fig. 3. Time courses of enhancement of intraspinal transmission in individual experimentsby different doses of aminophylline (A), IBMX (B), and RO 20-1724(C). The rami from which recordings were analyzed are indicated next to dosages. Each point represents an average of 16 consecutiveresponses.

/,

,2o

5<__ Fig. 2. Recordings of discharges recorded from T3 preganglionic rami evoked by intraspinal stimulation before and at several times after i.v. administration of aminophylline or IBMX in separate experiments. Upper traces of each series are individual responses; lower traces are averages of 16 consecutive responses evoked at 0.1 Hz. Calibrations: fop traces; 25 ms, 50 #V.

In contrast to the rapid, marked enhancement of intraspinal transmission to sympathetic preganglionic neurons by each of the 3 phosphodiesterase inhibitors, transmission through spinal sympathetic reflex pathways was only slowly and modestly enhanced by the same respective doses of each drug (Fig. 4). The onset of enhancement was delayed and the rates of onset were much more gradual so that maximal enhancement was reached at about 1 h before declining to control levels by about 2 h. Maximal enhancements produced by each drug were about equal (Table I).

203

200

c

B

A AMINOPHYLLINE 50 rng / kg

RO 2 0 - 1 7 2 4 I rng/kg

IBMX I mg/kg

• INTRASPINAL

175-

REFLEX

._J

oO~ 150Z

o cD o~ 1 2 5 o

I00 •N--4 oN:4

75 0

•N=7 oN= 5

I

I

I

I

2

5

7-

0

I

• N: 4 oN:3 I

I

2

5

0

I

[

I

t

2

5

TIME-HOURS

Fig. 4. Time courses of average enhancement of intraspinal and spinal reflex transmission in multiple experiments by aminophyllinc (A), IBMX (B), and RO 20-1724 (C). Vertical bars represent means _+S.E.M. N indicates number of experiments in each series.

TABLE 1

Maximum enhancement of sympathetic discharges by phosphodieslerase inhibitors Values are means + S.E.M. and (ranges) of maximal effects.

Drug (rag/ kg)

lntraspinal

n

Spinal reflex

n

Aminophylline (50) RO 20-1724 (1) IBMX(1) After pretreatment: AMPT (350)/reserpine (5) Chlorpromazine (4.5) Prazosin (0.5)

176 _ 8 (157-196) 185 _+13 (161-209) 190_+ 8(156-216)

4 4 7

127 + 8 (123-132) 128 + 2 (126-131) 123+3(115 128)

4 3 5

136 _+ 4 (126-146) * 128 + 4 (121-138) * 134 _+ 4 (116-150) *

4 4 6

* P < 0.001, compared to IBMX in untreated cats.

Effects of I B M X on intraspinally euoked sympathetic discharges after depletion or blockade of catecholamines In o r d e r to assess the role of d e s c e n d i n g c a t e c h o l a m i n e p a t h w a y s in p r o d u c i n g the large r a p i d e n h a n c e m e n t of i n t r a s p i n a l t r a n s m i s s i o n to S P G N s b y the p h o s p h o d i e s t e r a s e inhibitors, several p r o c e d u r e s were used to i m p a i r such t r a n s m i s s i o n b e f o r e a d m i n i s t e r i n g 1 m g / k g of I B M X . T h e first p r o c e d u r e was designed to d e p l e t e central stores of c a t e c h o l a m i n e s . Cats were pret r e a t e d with a l p h a m e t h y l - p a r a t y r o s i n e ( A M P T ) ,

2 × 100 m g / k g , i.p., on the d a y before experim e n t s a n d 3 × 50 m g / k g , i.v., d u r i n g the 6 h surgical p r e p a r a t i o n . A f t e r baseline intraspinal t r a n s m i s s i o n was stabilized, 5 m g / k g of reserpine was given i,v. As shown in Fig. 5A, reserpine r a p i d l y d e p r e s s e d i n t r a s p i n a l t r a n s m i s s i o n to a b o u t 50%. The second p r o c e d u r e involved b l o c k i n g c a t e c h o l a m i n e t r a n s m i s s i o n with c h l o r p r o m a z i n e which blocks b o t h d o p a m i n e a n d ~ F a d r e n e r g i c r e c e p t o r s [34,39] or with p r a z o s i n which blocks o n l y cq-adrenergic receptors [35]. C h l o r p r o m a z i n e p r o d u c e d p r o m p t b u t only p a r t i a l d e p r e s s i o n of

204 A

A ] ~ ' L O R P R O M L ~ i N~

RESERPINE

x

,2o-] ~ !

~ z ~ ,oo

1 ix

B

IBMX

AMrNOPHYLLrNE

125 2 mg1 g

°°°°°° -~--'--~-:---,......

o

017;

~

,

T2 r

"5,

• MONOSYNAPT 80

~D oz

Img/kg '~

~

AMPT/RESERPiNE 35015mg/kg

/,/" \,

175

ZBMX Irng/kg ~ T : "\ , /

SONTRC: C'ILORPROMAZ ;NE 4 5 mq/kg

% /

*%

,

*

~ :;

C

l 2

ic IoPoLYsY"~P"C ~ O

t

2

TFME-HOURS

N-4

N=4 75

J

.

Fig. 6. Effects of aminophylline (A) and IBMX (B) on transmission through a monosynaptic and a polysynaptic somatic reflex pathway in the exposed, lower lumbar spinal cord (L6 and L7) in separate experiments. Reflexes recorded from a ventral root were evoked by stimulation of a dorsal root.

150

Ill

0

o...,.

~-~-.-~-~-: o+, .......

] 3

TiME

:

2

3

HOdRS

Fig. 5. Depression of intraspinal transmission by reserpine (A) after 36-h pretreatment with divided doses of alphamethylparatyrosine (AMPT) or by chlorpromazine(B) in single experiments. Pretreatment with AMPT/reserpine (C) or chlorpromazine (D) markedly restricted enhancementof intraspinal transmission by IBMX. Dashed curves illustrate average enhancement by IBMX alone.

intraspinal transmission as shown in Fig. 5B (average, 49 _+ 7% of control, n = 4). Prazosin (0.5 m g / k g ) also promptly depressed transmission to 61 + 3% (n = 6) of control. Pretreatment with AMPT-reserpine (Fig, 5C), chlorpromazine (Fig. 5D), or prazosin markedly impaired the ability of IBMX to enhance transmission. Maximal enhancements were restricted to about one-third of those produced in untreated animals (Table I) and were delayed in onset and of shorter duration. These patterns of enhancement were similar to those produced on the spinal reflex pathway (Fig. 4).

Effects of aminophylline and I B M X on spinal somatic reflexes Somatic monosynaptic and polysynaptic reflexes in lumbar spinal pathways were not significantly affected by cumulative doses of aminophylline (25-75 mg/kg, n = 2) or IBMX (1-2 mg/kg, n = 4). Monosynaptic reflexes were slightly depressed and polysynaptic reflexes were only

slightly depressed or enhanced by the higher doses (Fig. 6). No convulsive-type discharges were evoked until the dose of aminophylline was increased to 100 m g / k g or higher and that of IBMX was increased to 4 - 6 mg/kg. Reflexes evoked between convulsive-type discharges were essentially unchanged from control values.

Effects of clonidine on sympathetic transmission and phosphodiesterase inhibition Clonidine HC1 produced a dose-dependent, marked depression of transmission through intraspinal (Fig. 7A, B, D) and spinal reflex (Fig. 7C) pathways to sympathetic preganglionic neu, rons. In addition, depending on dose, clonidine also partially (Fig. 7A) or completely (Fig. 7B, C) prevented enhancement of transmission by 50 m g / k g of aminophylline. Blockade of c~z-adrenergic receptors by tolazoline or yohimbine has been shown previously to antagonize the depressant effects of clonidine on both spinal reflex and intraspinal pathways, restoring transmission only to the original control levels [2,13,15,16,47]. However, since aminophylline was still active in the present experiments, the a2-receptor antagonists not only antagonized the depressant effects of clonidine but also restored the ability of aminophylline to enhance transmission to levels well above the original control levels. Clonidine also prevented enhancement of transmission by IBMX or RO 20'1724 and the enhancement was rapidly restored when a2,receptors were antagonized by yohimbine (Fig. 7D).

205

A

B

CLONIDINE 200lO#glkg

TOLAZOLINE 2mgl kg

AMIN.

C CLONIDINE 25 pg/kg

~

TOLAZOLINE

D

CLONIDINE 25 #g/kg

2.5 mg/kg

YOHIMBINE 0.5 mglkg

t

AMIN.

AMIN.

YOHIMBINE 0.Smqlkg

oRO20-1724 T ~

50mg/kg

5Omglkg

CLONIDINE 25 ,ug/kg

elBeX

/,~'o

+

150_J

ooc: t--

z

o

I00 50SPINAL REFLEX

PATHWAY

0 o

I

N=2 I

I

2

~

N=3

0

I

I

I

2

o

I

1

I

2

N=3 N=3 1

0

2

TIME-HOURS Fig. 7. Depressant effects of clonidine on intraspinal (A, B, D) or spinal (C) reflex transmission and on enhancement by phosphodiesterase inhibitors with subsequent antagonism by tolazoline (A, B) or yohimbine (C, D). A and B show the graded depressant effects of clonidine on intraspinal transmission and on enhancement by aminophylline and antagonism of both by tolazoline. C shows the depressant effects of clonidine on spinal reflex transmission and on enhancement by aminophylline and antagonism of both by yohimbine. D shows that clonidine also prevented enhancement by RO 20-1724 and IBMX and thal yohimbine restored the enhancement.

In contrast to the early termination of maximal intraspinal enhancements produced by the phosphodiesterase inhibitors alone (Fig. 4), the enhancements produced after antagonism of the depressant effects of clonidine did not show a tendency to decline (Fig. 7A, B, D) but continued to increase for up to 5 h [17,18] as would be expected on the basis of their estimated half-lives [21]. This observation suggests that the cAMP-induced increase in excitability may be self-limited by a local inhibitory process involving a2-receptors.

Discussion

The present results demonstrating rapid and marked enhancement of transmission through intraspinal excitatory pathways to sympathetic preganglionic neurons by aminophylline, IBMX, and RO 20-1724 extend our previous findings and add further support to the proposal [17,18] that descending synaptic activation of adenylate cyclase generates cAMP as a second messenger in these neurons. The increased levels of intraneuronal cAMP appear to increase the excitability of pre-

ganglionic neurons, probably by phosphorylating membrane proteins through activation of protein kinase A [36,37]. As an important criterion for demonstrating the mediation of synaptic events by a second messenger, the inhibition of cyclic AMP phosphodiesterase during synaptic activation permits accumulation of cAMP, which is normally rapidly hydrolyzed, thereby revealing its effects on neuronal excitability [4]. Each phosphodiesterase inhibitor also enhanced transmission through spinal reflex pathways, but the enhancement was much less and more gradual than produced on intraspinal pathways. This pattern is compatible with a gradual rise in cAMP levels due to a basal activity of adenylate cyclase as has been demonstrated in spinal cord tissue in vitro [27,28]. The possibility that intraspinal and spinal reflex pathways activate distinctly different populations of preganglionic neurons with differing responses to the phosphodiesterase inhibitors is not supported by interaction studies. Preganglionic discharges evoked by simultaneous activation of both pathways are not additive, and conditioning discharges in one pathway markedly depress test discharges in the other pathway for up to 200 ms (Neumayr

206

and Franz, unpublished observations). The effects of the 3 phosphodiesterase inhibitors on both intraspinal and spinal reflex pathways appear to be due to inhibition of cAMP phosphodiesterase rather than to inhibition of cGMP phosphodiesterase, to blockade of adenosine receptors, or to non-selective increases in neuronal excitability. RO 20-1724 does not inhibit cyclic GMP phosphodiesterase [46] or block adenosine receptors as do the xanthines [28,49], but it enhanced intraspinal transmission as much as and more rapidly than did the two xanthines. Furthermore, the potencies of IBMX and theophylline as adenosine antagonists are about equal [9], whereas their relative potencies for enhancing intraspinal transmission are consistent with those as phosphodiesterase inhibitors [9,56]. The possibility that the phosphodiesterase inhibitors enhance excitability of sympathetic preganglionic neurons by non-specific neuronal stimulation or by a generalized effect on interneurons is not supported by their lack of effects on spinal somatic reflexes. Monosynaptic reflexes and polysynaptic reflexes, which share common interneurons with spinal sympathetic reflexes [121, were not significantly affected by doses of either aminophylline or IBMX that markedly enhance intraspinal transmission. These results also suggest that the excitability of motoneurons and their relevant spinal reflex interneurons is not regulated by cAMP. The rapid depression of intraspinal transmission and the markedly reduced enhancement by IBMX after depletion of catecholamines or blockade of al-adrenergic receptors (Fig. 5, Table I) suggest the descending norepinephrine pathways are largely responsible for activating adenylate cyclase to increase intraneuronal cAMP and neuronal excitability. Activation of spinal cord adenylate cyclase by norepinephrine has also been demonstrated in vitro [27,28]. Although AMPT/reserpine treatment also depletes dopamine, and chlorpromazine blocks dopamine as well as al-receptors [34,39], the equivalent effect of prazosin, a selective al-receptor antagonist [35], suggests that norepinephrine pathways activate al-receptors that are positively coupled to adenylate cyclase. A previous study [33] showed that

propranolol (5 mg/kg, n = 7) does not depress intraspinal transmission at all but rather produces a late, gradual enhancement (20-30%). Therefore, fl-adrenergic receptors do not appear to be involved, and activation of a2-receptors by clonidine produces opposite effects. The possibility that descending dopamine pathways [48] also activate adenylate cyclase cannot be entirely discounted, but their contribution appears to be much less than that of norepinephrine pathways, lmmunohistochemical studies suggest that norepinephrine innervation of sympathetic preganglionic neurons is predominantly axodendritic [1,23]. The failure of catecholamine depletion or a~-receptor antagonism to prevent all of the enhancement by IBMX may be due to preservation of basal production levels of cAMP as discussed above in relation to spinal reflexes. Alternatively, other unidentified descending pathways may account for the modest enhancement if their receptors are also positively coupled to adenylate cyclase. However, since none of the pretreatments reduced intraspinal transmission by more than about 50%, it is clear that the remaining transmission to preganglionic neurons is mediated by other descending excitatory pathways. Excitatory amino acids such as glutamate [22] or neuropeptides such as substance P [32] are likely candidates for other excitatory neurotransmitters which may not directly influence adenylate cyclase. However, their synaptic efficacy could be markedly affected by the concurrent effects of those neurotransmitters that do regulate neuronal excitability through adenylate cyclase and cAMP, as has been demonstrated in the cerebellum [19]. The ability of clonidine to depress transmission through both intraspinal and spinal reflex pathways and to prevent enhancement by phosphodiesterase inhibitors indicates that activation of a2-receptors inhibits adenylate cyclase and prevents synthesis of cAMP. In the absence of cAMP synthesis, inhibition of its phosphodiesterase cannot increase cAMP levels and therefore does not enhance transmission. Although both depression of transmission and inhibition of adenylate cyclase by clonidine are dose-dependent, the results do not eliminate the possibility that other mechanisms might contribute to the total depressant

207

effect of clonidine. A secondary mechanism might account for the greater depressant effects of clonidine than of AMPT/reserpine, chlorpromazine, or prazosin on intraspinal transmission (cf., Fig. 5A, B and Fig. 7). Nevertheless, the results indicate that a large part of the depressant effect of clonidine is mediated by inhibition of adenylate cyclase and reduction of intraneuronal levels of cAMP. Depression of excitability by this mechanism could effectively reduce or enhance the respective synaptic efficacy of excitatory or other inhibitory inputs to these neurons. Several lines of evidence converge to suggest that the a2-receptors activated by clonidine are located postsynaptically on sympathetic preganglionic neurons and are innervated by bulbospinal epinephrine pathways. First, direct iontophoretic application of epinephrine, norepinephrine, or clonidine depresses preganglionic neurons by acting on a2-receptors [7,11,24,25]. Second, bulbospinal epinephrine pathways terminate only in the vicinity of sympathetic preganglionic neurons [26.40,42], and neither intraspinal nor spinal reflex transmission to motoneurons is affected by doses of clonidine that markedly depress transmission to preganglionic neurons [13]. Third, selective inhibition of brain epinephrine synthesis leads to a gradual and marked increase in intraspinal, but not spinal reflex, transmission [44]. Fourth, epinephrine has a higher affinity than norepinephrine for brain a2-receptors [52], and it is also a more effective inhibitor of sympathetic preganglionic neurons than norepinephrine [24]. Finally, there is an intimate relationship between epinephrine terminal fields and az-receptors throughout the brain [50,54], and central a2-receptors appear to be predominantly postsynaptic [51]. The similarity of effects and interactions of clonidine and phosphodiesterase inhibitors on transmission through both spinal reflex and intraspinal pathways in the present study is also consistent with a postsynaptic locus of action. The synaptic isolation necessary for the independent and reciprocal regulation of adenylate cyclase, cAMP levels, and neuronal excitability by the two catecholamines may be accomplished by differences in the respective locations of their synapses and receptors. Activation of negatively cou-

pied a2-receptors by epinephrine pathways and of positively coupled al-receptors by norepinephrine pathways would allow the two pathways to exert a wide range of bulbospinal regulation on the excitability of sympathetic preganglionic neurons and on the intensity of the sympathetic outflow. As shown in other biological systems that are regulated by hormonal or neurotransmitter influences on adenylate cyclase [6,41], it is reasonable to assume that the a 1- and a2-receptors are positively and negatively coupled to adenylate cyclase by respective stimulatory (N~) and inhibitory (N,), GTP-dependent regulatory proteins. Previous studies [55,57] showing a direct correlation between central levels of cAMP and blood pressure and heart rate suggested that cAMP is involved in central regulation of cardiovascular centers. The results of the present study support this proposal and, furthermore, implicate the sympathetic preganglionic neurons as one likely site of this regulatory function.

Acknowledgements This work was supported by U.S. Public Health Service Grants HL-24085 and GM-07579 and by a grant-in-aid from the Montana Heart Association.

References 1 Akeyson, E.W., Lewis, M.S., Molliver, M.E. and Grzanna, R., Ultrastructural demonstration of noradrenergic synapses in the rat spinal cord by dopamine-/~-hydroxylase immunocytochemistry. Soc. Neurosci. Abstr., 9 (1983) 285. 2 Baum, T. and Shropshire, A.T., Susceptibility of spontaneous sympathetic outflow and sympathetic reflexes to depression by clonidine, Eur. J. Pharmacol., 44 (1977) 121-129. 3 Bernthal, P.J. and Koss, M.C., Effects of clonidine and chlorpromazine on a sympathetic-cholinergic reflex, Eur. J. Pharmacol., 60 (1979) 23-29. 4 Bloom, F.E., The role of cyclic nucleotides in central synaptic function, Ret,. Pt~vsiol. Biochem. Pharmacol., 74 (1975) 1 103. 5 Chalmers, J.P., Blessing, W.W.. West, M.J., Howe, P.R.C., Costa, M. and Furness, J.B., Importance of new catecholamine pathways in control of blood pressure, Clin E~p. Hvpertens., 3 (1981) 393-416.

208 6 Codina, J., Hildebrandt, J., Sunyer, T., Sekura, R.D., Manclark, C.R., lyengar, R. and Birnbaumer, L., Mechanisms in the vectorial receptor-adenylate cyclase signal transduction, Adv. Cyclic Nucleotide Prot. Phosphorylation Res., 17 (1984) 111-125. 7 Coote, J.H., Macleod, V.H., Fleetwood-Walker, S. and Gilbey, M.P. The response of individual sympathetic preganglionic neurons to microphoretically applied endogenous monoamines, Brain Res., 215 (1981) 135--145. 8 Dahlstr~Sm, A. and Fuxe, K., Evidence for the existence of monoamine nuerons in the central nervous system. II. Experimentally induced changes in the intraneuronal amine levels of bulbospinal neuron systems, Acta Physiol. Scand., 64, Suppl. 247 (1965) 1-36. 9 Daly, J., Cyclic Nucleotides in the Nervous System. Plenmn, New York, 1977. 10 Dampney, R.A.L., Functional organization of central cardiovascular pathways, Clin. Exp. Pharmacol. Physiol., 8 (1981) 241-259. ] 1 De Groat, W.C. and Ryall, R.W., An excitatory action of 5-hydroxytryptamine on sympathetic preganglionic neurons, Exp. Brain Res., 2 (1967) 299-305. 12 Fields, H.L., Meyer, G.A. and Partridge, Jr., L.D., Convergence of visceral and somatic input onto spinal neurons, Exp. Neurol., 26 (1970) 36-52. 13 Franz, D.N., Spinal sites of action of ctonidine. In P. Milliez and M. Safar (Eds.), Recent Advances in ttypertension, Vol. 3, Boehringer, Ingelheim, 1976, pp. 133-138. 14 Franz, D.N., Role of catecholamines in bulbospinal sympathetic pathways. In E. Usdin, I.J. Kopin and J. Barchas (Eds.), Catecholamines: Basic and Clinical Frontiers, Pergamon, New York, 1979, pp. 649-651. 15 Franz, D.N., Hare, B.D. and McCloskey, K.L., Spinal sympathetic neurons: possible sites of opiate-withdrawal suppression by clonidine, Science, 215 (1982) 1643-1645. 16 Franz, D.N. and Madsen, P.W., Differential sensitivity of four central sympathetic pathways to depression by clonidine, Eur. J. Pharrnacol., 78 (1982) 53 59. 17 Franz, D.N. and Madsen, P.W.. Clonidine prevents enhancement of spinal sympathetic transmission by phosphodiesterase inhibitors, Neurosei. Lett, 28 (1982) 211-2 l 6. 18 Franz, D.N., Madsen, P.W., Peterson, R.G. and Sangdee, C., Functional roles of monoaminergic pathways to sympathetic preganglionic neurons, Clin. Exp. Hypertens., A4 (1982) 543-562. 19 Freedman, R., Hoffer, B.J., Woodward, D.J. and Puro, D., Interaction of norepinephrine with cerebellar activity evoked by mossy and climbing fibers, Exp. Neurol., 55 (1977) 269-288. 20 Fuxe, K., Bolme, P., Agnati, L.F., Jonsson, G., Andersson, K., Kohler, C. and H~Skfelt, T., On the role of central adrenaline neurons in central cardiovascular regulation. In K. Fuxe, M. Goldstein, B. H~Skfelt and T. H~Skfelt (Eds.L Central Adrenaline Neurons: Basic Aspects and Their Role in Cardiovascular Functions, Pergamon, New York, 1980, pp. 161-182. 21 Gal, P., Jusko, W.J., Yurichak, A.M. and Franklin, B.A.,

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36 37

Theophylline disposition in obesity. Clin. Pharmacol. Ther.. 23 (1978) 438-444. Ganong, A.H., Lanthorn, T.H. and (_otman, C.W.. Kynurenic acid inhibits synaptic and acidic amino acidinduced responses in the rat hippocampus and spinal cor& Brain Res., 273 (1983) 170-174. Glazer, E.J. and Ross, L.L.. Localization of noradrenergic terminals in sympathetic preganglionic nuclei of the rat: demonstration by immunocytochemical localization of dopamine-,8-hydroxylase, Brain Res., 185 (1980) 39-49. Guyenet, P.G. and Cabot, J.H., Inhibition of sympathetic preganglionic neurons by catecholamines and clonidine: mediation by an c~-adrenergic receptor, J. ,Veurosct., l (1981) 908-917. Guyenet, P.G. and Stornetta, R.L., Inhibition of sympathetic pregangiionic discharges by epinephrine and c~methylepinephrine, Brain Res., 235 (1982) 271-283. H~Skfelt, T., Fuxe, K., Goldstein, M. and Johansson, O., Immunohistochemical evidence for the existence of adrenaline neurons in the rat brain, Brain Re,~., 66 (1975) 235-251. Jones, D.J., Adenosine regulation of cyclic 3",5'-adenosine monophosphate formation in rat spinal cord. J Pharmacol. Exp. Ther., 219 (1981) 370-376. Jones, D.J. and McKenna, L.F., Alpha adrenergic receptor mediated formation of cyclic AMP in rat spinal cord, J. Cyclic Nucleotide Res., 6 (1980) t33-141. Krier, J., Thor, K.B. and De Groat, W.C., Effects of clonidine on the lumbar sympathetic pathways to the large intestine and urinary bladder of the cat, Eur. J. Pharmacol., 59 (1979) 47-53. loewy, A.D., McKellar, S. and Saper, C.B., Direct projections from the A5 catecholamine cell group to the intermediolateral cell column, Brain Res., 174 (1979) 309-314. Loewy, A.D. and Nell, J.J., The role of descending monoaminergic systems in central control of blood pressure. Fed. Proc. Fed. Am. Soc. Exp. Biol., 40 (1981) 2778-2785. Loewy, A.D. and Sawyer, W.B., Substance P antagonist inhibits vasomotor responses elicited from ventral medulla in rat, Brain Res., 245 (1982) 379-383. Madsen, P.W. and Franz, D.N., Contrasting effects of clonidine and propranolol on three central sympathetic pathways, Fed. Proc. Fed. Am. Soc. Exp. Biol., 39 (1980) 247. Marwaha, J. and Aghajanian, G.K., Typical and atypical neuroleptics are potent antagonists at %-receptors of the dorsal lateral geniculate nucleus, Naunvn-Schmiedeberg's Arch. Pharmacol., 321 (1982) 32-37. Marwaha, J. and Aghajanian, G.K, Relative potencies of alpha-1 and alpha-2 antagonists in the locus ceruleus, dorsal raphe and dorsal lateral geniculate nuclei: an electrophysiological study, J. Pharmacol. Exp. Ther., 222 (1982) 287 293. Nestler, E.J. and Greengard, P.. Protein Phosphorvlation in the Nervous Svstern, Wiley, New York. 1983. Nestler, E.J. and Greengard, P., Neuron-specific phosphoproteins in mammalian brain, Adv. E),cfic Nucl. Protein Phosphorvlation Res., 17 (1984)483-488.

209 38 Neumayr, R.J., Hare, B.D. and Franz, D.N., Evidence for bulbospinal control of sympathetic preganglionic neurons by monoaminergic pathways, Life Sci., 14 (1974) 793-806. 39 Palmer, G.C. and Manian, A.A., Actions of neuroleptic agents on central cyclic nucleotide systems. In G.C. Palmer (Ed.), Neuropharmacologv of Cyclic Nucleotides, Urban and Schwarzenberg, Baltimore, 1979, pp. 53-111. 40 Reis, D.J., Granata, A.R., Joh, T.H., Ross, C.A., Ruggiero, D.A. and Park, D.H., Brain stem catecholamine mechanisms in tonic and reflex control of blood pressure, Hypertension, 6 (1984) 117-115. 41 Rodbell, M., The role of hormone receptors and GTP-regulatory proteins in membrane transduction, Nature (London). 284 (1980) 17-22. 42 Ross, C.A., Armstrong, D.M., Ruggiero, D.A., Pickel, V.M., Joh, l.H. and Reis, D.J., Adrenaline neurons in the rostral ventrolateral medulla innervate thoracic spinal cord: a combined immunocytochemical and retrograde transport demonstration, Neurosci Lett., 25 (1981) 257-262. 43 Sangdee, C. and Franz, D.N., Enhancement of central norepinephrine and 5-hydroxytryptamine transmission by tricyclic antidepressants: a comparison, P~vchopharmacologv, 62 (1979) 9-16. 44 Sangdee, C. and Franz, D.N., Evidence for inhibition of sympathetic preganglionic neurons by bulbospinal epinephrine pathways, Neurosci. Lett., 37 (1983) 167-173. 45 Sangdee, C., Steffensen, S.C. and Franz, D.N., Enhancemere of bulbospinal excitatory transmission to sympathetic preganglionic neurons by desipramine and by dextroamphetamine. Soc. Neurosci. Abstr., 10 (1984) 713. 46 Schwabe, U., Miyake, M., Ohga, Y. and Daly, J.W., 4-( 3-Cyclopenlyloxy-4-methoxyphenyl)-2-pyrrolidone (ZK62711): a potent inhibitor of adenosine cyclic 3',5'monophosphate phosphodiesterase in homogenates and tissue slices from rat brain, Mol. Pharrnacol., 12 (1976) 900 910. 47 Sinha, J.N., Atkinson, J.M. and Schmitt, H., Effects of clonidine and L-dopa on spontaneous and evoked splanchnic nerve discharges, Eur. J. Pharmacol., 24 (1973) 113 119. 48 Skagerberg, G., BjtSrklund, A., Lindvall, O. and Schmidt, R.H., Origin and termination of the diencephalo-spinal

49

50

51

52

53

54

55

56

57

dopamine system in the rat, Brain Res. Bull., 9 (1982) 237-244. Smolem A.J. and Ross, L.L., The bulbospinal monoaminergic system of the chick: degeneration in the sympathetic nucleus following surgical and chemical lesions. Brain Res., 139 (1978) 153 159. Stolk, J.M., Vantini, G., Perry, B.D., Guchhait, R.B. and U'Prichard, D.C., Assessment of the functional role ol brain adrenergic neurons: chronic effects of phenylethanolamine-N-methyltransferase inhibitors and alpha adrenergic receptor antagonists on brain norepinephrinc metabolism, J. Pharmacol. Exp. Ther., 230 (1984) 577-586 U'Prichard, D.C., Bechtel. W.D, Rouot, B.M. and Snyder S.H., Multiple apparent o~-noradrenergic receptor binding sites in rat brain: effect of 6-hydroxydopamine, Mol Pharmacol., 16 (1979) 47-60. U'Prichard, D.C., Mitrius, J., Kahn, D.J. and Daiguji, M. Neural and non-neural adenylate cyclase coupled c~2-receptots: agonist and antagonist radioligand interactions, h H.J. Yamamura, R.W. Olsen and E. Usdin (Eds.), P~vcho pharrnacologv and Biochemistn of Neurotransmitler Recep tors, Elsevier/North Holland, Amsterdam, 1980, pp 247 259. Van Caulker, D., Muller, M. and Hamprecht, B., Receptors regulating the level of cyclic AMP in primary cultures ~v, perinatal mouse brain. In E. Meisami and M.A.B. Brazier (Eds.), Neural Growth and Differentiation, Raven, Ne~ York, 1979, pp. 11-25. Vantini, G., Perry, B.D., Guchhait, R.B., U'Prichard, D.(" and Stolk, J.M., Brain epinephrine systems: detailed comparison of adrenergic and noradrenergic metabolism, receptor number and in vitro regulation, in two inbred rat strains, Brain Res., 296 (1984) 49 65. Walland, A., Further evidence for the involvement of cAMP in central blood pressure regulation, Naunvn-Schmiedeberg'; Arch. Pharrnacol., 296 (1977) 177-181. Wells, J.N. and Kramer, G.L.. Phosphodiesterase inhibitoras tools in cyclic nucleotide research: a precautionary comment. Mol. Cell. Endocrinol., 23 (1981) 1 9. While, H.S., Driver, P.S. and lsom, G.E., Studies on the central pressor activity of dibutyryl cyclic AMP, Eur. J Pharmacol., 57 (1979) 107-113.