α2-adrenoceptor-mediated enhancement of glycine response in rat sacral dorsal commissural neurons

Pergamon

Neuroscience Vol. 89, No. 1, pp. 29~1, 1999 Copyright @ 1998IBRO. Publishedby ElsevierScienceLtd Printed in Great Britain. All rights reserved PII: S0306-4522(98)00303-0 0306-4522/99 $19.00+0.00

0t2-ADRENOCEPTOR-MEDIATED ENHANCEMENT OF GLYCINE RESPONSE IN RAT SACRAL DORSAL COMMISSURAL NEURONS J. NABEKURA,*:~§ T.-L. XU,*t:~ J.-S. RHEE,* J.-S. L i t and N. A K A I K E * *Department of Physiology, Faculty of Medicine, Kyushu University, Fukuoka 812-82, Japan tDepartment of Anatomy, K. K. Leung Brain Research Centre, The Fourth Military Medical University, Xi'an, Shaanxi 710032, People's Republic of China Abstract--The effect of noradrenaline on the glycine response was investigated in neurons acutely dissociated from the rat sacral dorsal commissural nucleus using nystatin perforated patch recording configuration under voltage-clamp conditions. Noradrenaline reversibly potentiated the 10 5M glycineinduced CI- current in a concentration-dependent manner. Single channel recordings in a cell-attached mode revealed that noradrenaline decreased the closing time of the glycine-activated channel activity. Noradrenaline neither changed the reversal potential of the glycine response nor affected the affinity of glycine to its receptor. Clonidine mimicked and yohimbine blocked the noradrenaline action on glycine response. N-[2(methylamino)ethyl]-5-isoquinolinesulfonamide dihydrochloride, protein kinase A inhibitor, mimicked the effect of noradrenaline on glycine response. Noradrenaline failed to affect the glycine response in the presence of these intracellular cyclic AMP and protein kinase A modulators. However, noradrenaline further enhanced the glycine response even in the presence of phorbol-12-myristate-13acetate and chelerythrine, a protein kinase C inhibitor. Pertussis toxin treatment for 6-8 h blocked the noradrenaline facilitatory effect on the glycine response. In addition, noradrenaline potentiated the strychnine-sensitive postsynaptic currents evoked in a slice preparation of sacral dorsal commissural nucleus. These results suggest that the activation of c~2-adrenoceptor by noradrenaline coupled with pertussis toxin-sensitive G-proteins reduces intracellular cyclic AMP formation through the inhibition of adenyl cyclase. The reduction of cyclic AMP decreases the protein kinase A activity, thus resulting in the potentiation of the glycinergic inputs to the sacral dorsal commissural neurons. It is thus feasible that the noradrenergic input to the sacral dorsal commissural nucleus modulates such nociceptive signals as pain by intracellular enhancing the glycine response. ~ 1998 IBRO. Published by Elsevier Science Ltd. Key words: a2 adrenoceptor, glycine response, rat sacral dorsal commissural neuron, protein kinase A,

glycinergic input, single channel.

Considerable evidence supports the idea that glycine acts as a major inhibitory transmitter in both the brainstem and spinal cord where glycine is likely to play an important role in modulating spinal nociceptive transmission. Glycine-containing neurons and nerve terminals are a b u n d a n t in the spinal dorsal horn. 3v Autoradiographic studies have shown that most labelling of all glycine receptors is in laminae II-III as well as the deeper laminae. 4 The :~Both authors contributed equally to this work. §To whom correspondence should be addressed. Abbreviations: APV, DL-2-amino-5-phosphonovalerate; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; GRKs, G protein-coupled receptor kinases; H-89, N-[2 (methylamino)ethyl]-5-isoquinolinesulfonamide dibydrochloride; HEPES, N-2-hydroxyethylpiperazine-N'-2ethanesulphonic acid; IAP, pertussis toxin; IBMX, isobutyl-methylxanthine; Ic~y, glycine-induced current; IPSC, inhibitory postsynaptic current; NA, noradrenaline; NMDA, N-methyl-D-aspartate; PKA, protein kinase A; PKC, protein kinase C; PLC, phospholipase C; PMA, phorbol-12-myristate-13-acetate; SDCN, sacral dorsal commissural nucleus. 29

presence of primary afferent-evoked glycinemediated inhibitory postsynaptic potentials (IPSCs) in the dorsal horn neurons has been reported. 43 Physiological studies have also shown that glycine depressed the activities of dorsal horn neurons and inhibited the glutamate-evoked depolarization. 4~ In the last few years, the strychnine-sensitive glycine receptors have been extensively investigated at both the functional and the molecular levels. 6 In cultured spinal trigeminal neurons, the activation of cAMP-dependent protein kinase A (PKA) potentiated the glycine-induced C1- current (IGly) by increasing the probability of channel opening. 36 Conversely, an increase in the P K A activity reduced the IGly in both the substantia nigra neurons 20 and the ventromedial hypothalamic neurons.1 The activation of protein kinase C (PKC) inhibited the function of glycine receptors in Xenopus oocytes injected with brain m R N A , 38 and potentiated the IGly in cultured hippocampal neurons 34 and acutely dissociated substantia nigra neurons of rat. 28 Thus, the mechanisms underlying the intracellular

30

J. Nabekura et al.

modulation of glycine receptors and the functional significance are valid topics for further investigation. Recent evidence has suggested that a variety of neurotransmitters and/or neuropeptides that regulate intracellular second messengers may affect the efficacy of synaptic transmission by modulating the phosphorylation of ion channels. 33 In the spinal nociception-transmission neurons, the possibility of this regulation is especially appealing. It has been established that the stimulation of the midbrain periaqueductal gray and the rostral ventral medulla inhibits the response of dorsal horn neurons to noxious stimuli, however, the inhibitory mechanism for this remains unclear. Noradrenaline (NA) and 5-hydroxytryptamine, which are present in the descending pathways emanating from the locus coeruleus and the nucleus raphe magnus, respectively, suppress the glutamate- and pinch-induced excitatory activities in the primate spinothalamic tract neurons. 4° Since N A is known to affect the second messenger levels in the target cells, 17 the inhibitory effect of N A thus may result from an enhancement of the inhibitory glycinergic transmission by glycine receptor phosphorylation mediated through the intracellular second messenger pathways. Therefore, in this study the modulation of glycine response by N A was investigated in acutely dissociated neurons from the rat sacral dorsal commissural nucleus (SDCN). One of the major termination sites of the descending noradrenergic pathway is the spinal intermediate zone and lamina X, especially in the lumbosacral segments. 21 The S D C N region represents the area just dorsal to the central canal in the sacral spinal cord and is also known to be implicated in nociceptive transmission.11'19"25 It is therefore an ideal place to study the modulation of the nociceptive signals. 19,39,42

EXPERIMENTAL

PROCEDURES

Cell preparation

The SDCN neurons were acutely dissociated as described elsewhere, 4a with some minor modifications. Briefly, two-week-old Wistar rats, supplied by the department animal center, were decapitated under pentobarbital-Na anesthesia. A segment of the lumbosacral (L5-$3) spinal cord was then dissected out and sectioned with a Vibratome tissue slicer (Dosaka, DTK-1000, Kyoto, Japan) to yield several transverse slices (400-gm-thick) containing the SDCN region. The slices were preincubated in welloxygenated incubation solution (see below for the composition) for 50rain at room temperature (22 25°C). Thereafter, the slices were treated enzymatically in an incubation solution containing ling/G8 ml pronase for 20rain at 31°C followed by exposure to l mg/(~8ml thermolysin for another 15 min in the same conditions. Then a portion of SDCN region was punched out with an electrolytically polished injection needle and transferred into a culture dish filled with standard external solution (see below). The neurons were mechanically dissociated with fire-polished Pasteur pipettes under visual guidance under a phase-contrast microscope (Nikon, TMS-I, Tokyo, Japan).

To elucidate the effect of enzyme treatment on the glycine response, the neurons were dissociated mechanically in some experiments. In brief, the slices were immediately transferred into the dish and a fire-polished glass pipette was touched lightly on to the surface of the SDCN. The pipette was vibrated horizontally at about 60 Hz for 5 s, and then slices were removed. The mechanically and/or enzymatically dissociated neurons adhered to the bottom of the dish within 20 min, thus enabling the electrophysiological studies to be conducted. The neurons which retained their original morphological features such as the dendritic processes were used for the experiments. Recordings Electrical measurements were carried out by using nystatin perforated patch recording configuration under voltageclamp conditions at room temperature (22-25°C). Patch pipettes were pulled from glass capillaries with an outer diameter of 1.5 mm on a two-stage puller (Narishige, PB-7, Tokyo, Japan). The resistance between the recording electrode filled with pipette solution and the reference electrode was 4-6 MF2. Series resistance checked every 10 min was 10-30 Mf2. The change of series resistance through recordings was less than 10%. The current and voltage were both measured with a patch-clamp amplifier (Nihon Koden, CEZ-2300, Tokyo, Japan), filtered at 1 kHz (NF Electronic Instruments, FV-665, Tokyo, Japan), and monitored on both a storage oscilloscope (Iwatsu Electronic, 5100A, Tokyo, Japan) and a pen recorder (Nippondenki San-ei, Recti-Horiz-8K21, Tokyo, Japan) and then stored on video tape after digitalization at a rate of 44 kHz (Nihon Koden, PCM 501 ESN). The membrane potential was held at - 40mV throughout the experiment, except when examining the current-voltage relationships. All measurements were started after stabilization of the glycine response (15 25 min after cell attachment). To analyse the glycine-operated single channel activity obtained in a cell attached mode, single channel currents were stored on videotape, and thereafter loaded into a computer at a sampling frequency of 2.5 kHz and analysed using pCLAMP (Axon Instruments, Version 6.0, U.S.A.). To evaluate the NA action on the glycinergic inputs to the SDCN neurons, slice patch recordings was employed. A concentric stimulating electrode was placed in the SDCN. To elicit the postsynaptic potentials, isolated square pulses with a duration of 100 ms were given at a rate of 0.05 Hz. To discriminate glycinergic inputs from GABAergic and glutamatergic inputs, 10 - 6 M bicuculline, 3 x 10 5 M 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 3x 10 5 M DL-2-amino-5-phosphonovalerate (APV) were added to the standard extracellular solution. The synaptic currents were stored in the DAT data recorder (TEAC. RD-120, Tokyo, Japan). Five successive synaptic responses with and without NA were averaged using an addscope (Nihon Koden, ATAC-150, Tokyo, Japan). Chemicals The ionic composition of the incubation solution was (mM): NaCI 124, NaHCO 3 24, KCl 5, KHzPO 4 1.2, CaC12 2.4, MgSO 4 1.3 and glucose 10, aerated with 95% O215% CO 2 to a final pH of 7.4. The normal external standard solution was (mM): NaC1 150, KCl 5, CaC1z 2, MgCI2, HEPES 10 and glucose 10. The pH was adjusted to 7.4 with Tris-hydroxymethyl aminomethane (Tris-base). The patchpipette solution for nystatin perforated patch recording was (mM): CsCI 150, HEPES 10 and 200 gg/ml nystatin. The pipette solution for the slice patch experiment was (mM) 150 KC1, 6 MgC12, 2 ATP, 0.5 GTP and 10 HEPES. The pH was adjusted to 7.2 with Tris-base. Drugs used in the present experiments were: thermolysin, nystatin, NA, phenylephrine, clonidine, isoprenaline, prazosin, yohimbine, propranolol, dibutyl cyclic AMP, CNQX,

Noradrenergic modulation of glycine response

A

Gly (M) 3x10-6

10.4

10"5

B 10.3 m

31 STR (M) 10-8 I I ml

Gly 10-SM mm

"~~" ~l~Ss300pA ~1~~

I

10-7 I mm

J100 15s

pA

1.0 3,0 i- 0.8

2.o

= o.0 ~ 0.4

1.0

I~ 0.2

0

0 I

10-6

I

r I|||||I

|

||||mi

10"S 10-4 Gly (M)

|

I

||||.J

10"3

10"9

I

i llllnl

i

l iiiiii|

!

10-e 10.7 STR (M)

l llllld

10"6

Fig. I. Glycine responses in the SDCN neurons. (A) The concentration response relationship for glycine (Gly) in SDCN neurons. All currents were normalized to the peak current amplitude induced by 3 x 10-SM glycine (*). (B) The concentration-inhibition curve for strychnine (STR) on 10-SM glycine-induced currents. The currents were expressed as the relative value to the control response induced by 10 5M glycine. STR was perfused from 30 s before and up until the end of the 10-SM glycine application. The holding potential (VH) was - 4 0 m V . Each point and vertical bar represents the mean 4- S.E.M. of six neurons.

APV, bicuculline, isobutyl-methylxanthine (IBMX) and forskolin from Sigma (St Louis, U.S.A); pronase from Calbiochem (La Jolla, U.S.A.); N-[2(methylamino)ethyl]-5isoquinoline sulfonamide dihydrochloride (H-89) from Seikagaku Corporation (Tokyo, Japan); phorbol-12myristate-13-acetate (PMA), chelerythrine and 4c~-phorbol12-myristate-13-acetate (4c~-PMA) from Funakoshi (Tokyo, Japan), pertussis toxin (IAP) from Kaken Seiyaku (Tokyo, Japan), and tetrodotoxin from Sankyo (Tokyo, Japan). The stock solutions of clonidine, dibutyl cyclic AMP, IBMX, forskolin, PMA, and 4c~-PMA were all prepared in dimethylsulfoxide and diluted to final concentrations in the external standard solution. The final concentration of dimethylsulfoxide was always less than 0.1%, which did not induce any ionic current and had no effect on glycine response. All other drugs were dissolved in the external solution just before use. The drug solutions were all administered by the "Y-tube" method. 26 This system allows the complete exchange of external solution surrounding a neuron wilhin 20 ms.

Statistical analysis When the relationships between the peak current amplitude and the glycine concentration were examined, continuous lines were fitted according to the following equation: 1= Ima×-cn/(Cn+ECnso) where I is the normalized value of the current, I,,ax the maximal response, C the glycine concentration, ECso the concentration which induced the half-maximal response, and n the apparent Hill coefficient. The experimental values are shown as the mean ± S.E.M. Student's t-test was used when two groups were compared.

RESULTS

Strychnine-sensitive glycine response In keeping with previous histological observ a t i o n s ] 9'25 the dissociated S D C N neurons were morphologically heterogeneous. The most c o m m o n neurons found in S D C N were medium-sized (1015 gm in diameter), fusiform-like cells that had oval or triangular soma and two to three apical stem dendrites. Larger neurons (15-25 ~tm in diameter) were also found. Some neurons were bipolar in shape and had round or oval soma. In the present study, medium-sized neurons were chosen for the further experiment. The glycine-evoked inward currents in all the S D C N neurons tested at a holding potential (VH) of - 4 0 m V . The currents became detectable at a concentration of about 3 x 1 0 - 6 M and thereafter increased in a sigmoidal fashion with increasing glycine concentration (Fig. 1A). The average threshold concentration, ECso value, the apparent Hill coefficient were 3 x 10 6 M, 4.0 x 10 5 M and 1.42, respectively (n=24). In addition, the reversal potential (EGly) of the glycine response in all tested neurons was - 2 . 0 + 0 . 4 4 m V (n=8). This value was close to the C I - equilibrium potential (Ecl) calculated by the Nernst equation from the given extra- and intracellular C1- concentrations, thus suggesting that the current is passing through the CI channels. Moreover, the glycine response of all neurons was

J. Nabekura et al.

32

A

Gly IO-SM

NA (M) 10-8 mmm ~

/

10-7 ~

/

10-6

/

mm

10-5 ~1

/

10-4 mmm

J 3(~pA '15s

B

~

2.0

O

~

1.5

1.11 I 10..9

I 10-8

I 10-7

I 10-6

I 10-.5

I 10-4

NA (M)

Fig. 2. Potentiation of Ic~yby NA. (A) A typical example of the IGly potentiated by NA of various concentrations. All recordings were obtained from the same neuron. NA was applied from 30 s prior to glycine (Gly) application. (B) The concentration-response relationship for the NA-mediated augmentation of IGly. The ordinate indicates the enhancement ratio of the glycine response by NA. The dashed line indicates the control induced by 10 5 M glycine alone. VH was -40mV. Each point and bar indicates the mean • S.E.M. of six measurements. inhibited by strychnine in a concentration-dependent manner (Fig. 1B). The ICso value was 3.2 x 10 s M.

Potentiation of glycine responses by noradrenaline In the present experimental conditions for recording glycine-induced C1- current (lGly), the application of N A itself did not induce any current even at high concentrations beyond 10 - 6 M. However, simultaneous treatment with N A and glycine increased the 10 -5 M glycine response in 61% (n=51) of the 83 SDCN neurons tested. The effects of N A at various concentrations on the I6~y were quantitatively studied. Figure 2 shows that N A potentiated Ic~y in a concentration-dependent manner at the concentration ranges from 10 9 to 10 4M, though the largest potentiation was induced by adding 1 0 - 6 M N A (190+33% increase, n=6; P<0.001). At N A concentrations higher than 10 6 M , the enhancement ratio decreased markedly, thus showing a bellshaped potentiation curve (Fig. 2B). The inactivation of the IGjy also clearly became faster as the N A concentrations increased (Fig. 2A). To elucidate the effect of enzyme treatment on the glycine response and on N A effect, we employed the mechanical dissociation method as described in the Experimental Procedures. The mechanically

dissociated neurons also responded to glycine and showed the potentiation of the 1 0 - S M glycine response by 10 6 M N A (182+26% increase, n=8). The sustained pretreatment with NA for more than 1 min did not produce any further potentiation of the IGly but accelerated the inactivation (desensitization) of the IGly (Fig. 3Aa). The augmentatory effect of NA was reversible although the recovery time of the NA effect depended on the N A concentrations used. The potentiation effect of NA at concentrations of 1 0 - 7 M or lower on IGly disappeared immediately after washing out NA (Fig. 3Ab), while 10 6 M NA caused a long-lasting modulation of the IGly in both the amplitude and the desensitization even after the washout of N A (Fig. 3B). Figure 3Bb shows the slow recovery time-course of 1 0 - S M glycine response facilitated by 10 6 M NA. Therefore, to evaluate the NA action on the IGly, the pretreatment of 1 0 - 7 M NA for 1 rain was employed in all successive experiments. To explore the facilitating mechanism of NA on IGly, we investigated the current-voltage (I-V) relationships and the concentration-response curves of 10 5 M glycine in the presence or absence of NA. In the present experimental conditions, the I - V relationship of glycine alone was almost linear at the potential range between - 8 0 and +60 inV. After adding 10 7 M NA, the current amplitude at all potentials increased without any change in the reversal potential (Fig. 4A). The reversal potential (Ec~y) for the IGly in the presence of NA ( - 2 . 4 1 +0.18 mV, n=5) was close to that of the control before adding NA ( - 2 . 8 2 ± 0 . 1 0 m V , n--5). NA also increased the maximum value of the concentration-response relationship without affecting the threshold concentration, thus indicating that N A did not change the apparent affinity of glycine to its receptor (Fig. 4B). The ECso values for IGJy were 3.2 x 10 5 M (n=6) in the control and 3.5 x 10 5 M (n=6) in the presence of 10 7M NA. No significant difference was observed between the two values. The Hill coefficients were 1.38 and 1.48 without and with NA, respectively.

Pharmacological analysis" of adrenoceptor subtypes The subtypes of the adrenoceptor, termed cd, a2, 131 and 132, have been defined both biochemically and pharmacologically. 17 To clarify the subtypes of adrenoceptor involved in potentiating the IGly, the effects of N A agonists and antagonists on the Ic~y were investigated. Clonidine is a selective agonist of the ct2 adrenoceptor. Clonidine, 10 7 M , effectively facilitated the IGly and mimicked the action of NA. Both phenylephrine of an a l agonist and isoproterenol of a 13 agonist failed to mimic the NA action (Fig. 5A). The addition of an c~2 antagonist yohimbine completely blocked the NA potentiation of the IG]y. In the presence of ctl antagonist prazosin, the facilitatory effect of NA on IGly was decreased

Noradrenergic modulation of glycine response

Aa

Gly 10-SM i

J

NAI O'SM 4

1 m

i

B a Gly 7mln 10.SM ~:.M B

f

eel

33

remov81 3 6

9

m

mmm

15mln

m

m

12 -Wli vT" J

2.5

b

b 104M

m

m

0

10-7M

IBIB

B

3OOpA

15s

m IIIIIB

_. 2.0 .iu ,-

i

1.5 i

@

rr 1.0

A

I

I

0

300pA

I

I

I

I

5 10 Recovery (mln)

I

15

15s Fig. 3. Facilitatory effect of NA on IGly. (Aa) The perfusion of 10 5 M glycine at an interval of 3 min during a continuous application of 10-6 M NA. Similar results were obtained from the other five neurons. (Ab) Rapid recovery of 10 8 M and 10 7 M NA-induced potentiation of IGly. (Ba) Slow recovery of the 10 5 M glycine-induced current once enhanced by 10 - 6 M NA. (Bb) The recovery time-course of the 1 0 - 6 M NA effect on IGly. All responses were normalized to the peak amplitude induced by 10-SM glycine alone (dashed line). The arrow shows the start of the recovery after washing out NA. Each point and bar indicates the mean± S.E.M. of five neurons.

A

B Relativecurrent ~i ~ ~

3.0

Ecl -60

[

.._

~,-~'/10

Contro~

Control

~ a.o

. -90

IO

30

60

-2

q) 1.o nO ]

T/i

A10-7M

-4

10-6

!

I IIInl]

i

10-6

I I|1111]

10-4

Gly (M)

I

I lUlII|

10-3

Fig. 4. The effects of NA on the I - V relationships and concentration-response curves for glycine. (A) The

I-V relationships of the glycine response with or without 10 7 M NA. The arrow indicates the C1 equilibrium potential (Ect). All currents were normalized to the peak current induced by 10-5 M glycine alone at a V H of - 4 0 m V (*). Each point and bar indicates the average and + S.E.M. of five neurons. (B) The concentratio~response curves of glycine in the presence or absence of 10 7 M NA. Data were obtained from the same neuron. All responses were normalized to the peak current induced by 3 x 10 5 M glycine alone (*). Similar results were obtained from the other five neurons.

slightly. However, the change in the p o t e n t i a t i o n ratio was n o t statistically significant (from 173 4- 16% to 1584-22%, P > 0 . 0 5 , n = 5 ) . A 13 antagonist, propranolol, h a d n o effect o n the N A facilitating effect (Fig. 5B). These results thus suggest t h a t the N A p o t e n t i a t i o n on the IGly is m a i n l y m e d i a t e d by the c~2 adrenoceptor.

Intracellular mechanism of noradrenaline action The IGly is k n o w n to be m o d u l a t e d by intracellular P K A 27 a n d P K C . 28'34'36 T o evaluate the c o n t r i b u t i o n o f P K C p a t h w a y to the facilitating action of N A on the IGly, the effects o f P K C activator a n d inhibitor on IGly were thus investigated. T h e application of P M A ,

34

J. Nabekura et al.

B

A a

Gly 10-SM

10"7M NA

NA Phe

CIo

Iso

mm

J

a

1

10"SM Pra

i

Yoh

Pra,Yoh

i

d o2

300pA

15S

15S

b

IGly 10"SM I

2,0

Pro

i

b

~0~

I~'y l°~MI

2.0-

tO @ > .~1.5

±

o

>o 1.5"

Ill

i

Q n.

1,0

NA

Phe

CIo

Iso

1.0

NA

+ Pra

÷

÷

÷

Yoh

Pra,

Pro

Yoh

Fig. 5. Pharmacological estimation of the adrenoceptor subtypes. (A) The effects of NA agonists. (Aa) IG~y was potentiated by 10 -7 M NA and 10 -7 M clonidine (Clo) but neither by 10 -7 M phenylephrine (Phe) nor by 10 7M isoproterenol (Iso). All recordings were obtained from the same neuron. (Ab) The statistical data obtained from five neurons. **P<0.001. (B) The effects of NA antagonists. (Ba) A typical example of the effects of NA antagonists on NA-mediated potentiation of I~ly. All recordings were obtained from the same neuron. The neuron was pretreated with each antagonist for 15 s before the simultaneous application with 10-7M NA and 10-SM glycine. (Bb) The statistical data showing the effects of NA antagonists. Each column and bar represents the averageq-S.E.M, of five neurons. **P<0.001. Pra, prazosin; Pro, propranolol; Yoh, yohimbine. a PKC activator, significantly enhanced the I~]y. The effect of P M A did not show any recovery after 30 min of removal of P M A (Fig. 6A). The inactive phorbol ester, 4c~-PMA, had no effect (Fig. 6Ab). Even in the presence of 10 7 M PMA, 10 7 M N A could further potentiate the I~ly by 39 + 16% (n--4, P<0.001). On the other hand, the addition of 3 x 1 0 - 6 M chelerythrine, an inhibitor of PKC, slightly decreased the peak amplitude of the IG~y by 12+5%, (n=7). During the suppression of IGly by chelerythrine, the application of 10 7 M NA induced a potentiation of the Ic~y (139 + 14%, n = 8) (Fig. 6Bb) This value was somewhat less than that before chelerythrine application (162 4- 12%, n = 8). This finding thus suggests that the PKC cascade is not a principal pathway in N A potentiation although the involvement of PKC can not be completely ruled out. In the present experiment, the N A potentiation on the Ia~y was mediated by c~2-adrenoceptor. Since a2-adrenoceptor regulates the cAMP production, 17 a possible involvement of cAMP in the N A facilitating action on IG~y was examined. The extracellular application for 5 min of 3 x 10 S M forskolin, an activator of adenylate cyclase, significantly inhibited the IGly (Fig. 7Aa). IBMX (10 5M), which inhibits phosphodiesterase, together with dibutyl cyclic A M P (3 x b l 0 - 5 M), a membrane-permeable cAMP

analog, also inhibited the |Gly (Fig. 7Ab). Their inhibitory effects lasted for more than 30 min (Fig. 7B). To clarify the suppression of the IGly by intracellular cAMP, 10 6 M cAMP was directly applied intracellularly in the conventional whole-cell patch recording mode. The peak amplitude of lGly decreased by 36.3:k6.8% (n=5) at 15min after a rupture of the membrane (Fig. 7C). In contrast, IGly decreased only by - 3.2 + 3.6% (n = 4) in the control experiment with patch-pipette solution without cAMP, thus indicating that intracellular cAMP potentially suppressed the peak amplitude of IGly. Forskolin, IBMX and dibutyl cyclic A M P themselves did not elicit any current at the concentrations used. During the suppression of I ~ y by 3 x 1 0 - S M forskolin or the mixture of 1 0 - S M IBMX and 3 x 10 5M dibutyl cyclic AMP, the application of l0 7 M N A only slightly potentiated the peak of IGly by 2.4±0.9% (n--5, P>0.05; Fig. 8A) and 1.7±0.4% (n=5, P>0.05; Fig. 8B), respectively, of that before N A application. The treatment of 1,9-dideoxyforskolin (10-5 M) did not affect the peak amplitude of IGly. Since we previously observed that the suppression of glycine receptor-mediated response by the intracellular application of cAMP could be reversibly relieved by a P K A antagonist in substantia nigra neurons, TM the effect of H-89, a potent inhibitor of

Noradrenergic modulation of glycine response

Aa Gly 3xl0-SM m

ili

:V----

PMA 10"7M

removal

r///.-//1////J

3

Ba

5

~

10

:V---• oval

r

3x10"SM

~

5

10

15

20

.

.

.

.

.

.

.

""

.

b

~

.

.

.

.

.

.

.

.

f

/ / H H / / I

mm

r

I"

~

I

I

I

25

30

35

Time (rain)

Chelerythrlne 3x 10-6M ~ H / / H / / / / / IIBI ~

mm

"

0

.

10.7M mm

~ pA 15s

I • PMA 10 .8 M [ 0 PMA 10 -7 M ] I E] 4a-PMA 10 -7 M I

1.o

NA

PMA 10"7M

Gly 20min

35

- -

|

3OO pA

,.,a

15 s

Fig. 6. Lack ofa PKC cascade in the NA potentiating effect. (A) Potentiation of I~Lyby PMA. The results were also quantitatively analysed in Ab. All responses were normalized to the peak current induced by 3 x 10 5M glycine alone. Each point and bar represents the mean± S.E.M. of four to seven measurements. The arrow indicates the start of recovery. (Ba) Even in the presence of 10 7 M PMA, 10 7 M NA further potentiated the Icily. (Bb) Chelerythrine (3 x 10 6 M) did not block the NA facilitatory effect on the I~ly.

PKA, was examined on the IGly in SDCN neurons. Pretreatment with 10 6 M H-89 for 5 rain enhanced the [Gly (Fig. 7Ac, B). In the presence of H-89, 10 7 M NA did not produce further increase in the peak amplitude of IGly (--2.3 ±4.3% of that before NA application, n--5, P>0.05; Fig. 8C). These results therefore indicate that the NA potentiation on IGly in the SDCN neurons is mediated by the PKA pathway.

14.1 +2.45 ms, and 0.007±0.001 in the control, and 1.74-0.05 ms, 2.5+0.65 ms, and 0.028+0.003 in the presence of 1 0 - 7 M NA, respectively. NA decreased the closing time of glycine-gated single channel activity, thus resulting in an increase of open probability of the C1 channel. These findings were confirmed in another three SDCN neurons (Fig. 10).

Involvement o f pertussis toxin-sensitive G proteins

Noradrenaline action on the strychnine-sensitive postsynaptie currents

To clarify the type of G-proteins involved in the NA action on the Ic]y, the effect of pertussis toxin (IAP) was examined. The dissociated SDCN neurons were incubated in standard external solution either with or without IAP (300ng/ml) for 6 to 8 h at room temperature. In the control neurons without IAP treatment, 1 0 - 7 M NA enhanced the [Gly (172± 13%, n=7; P<0.001, Fig. 9Aa, B). However, NA had no effect on the Ic,y in the neurons treated with IAP (2± 1%, n--7; P>0.05, Fig. 9Ab, B). IAPsensitive G-proteins are thus considered to be related to the facilitatory action of NA on the IGly-

Glycine-gated single channel activities Single channel recordings in a cell-attached mode revealed that the respective values for the open time, closed time and open probability of 3 x 10 7M glycine-gated channel activities are 1.7+0.04ms,

Glycine is an inhibitory transmitter in the spinal cord 43 and the NA descending input to this region is also present. 31 To evaluate the potentiating action of NA on the glycinergic inputs, conventional wholecell recording ("blind patch recording") was performed on the SDCN neurons in slice preparations of 400gm thickness. In the presence of 10 6M bicuculline, 3 x l 0 5 M C N Q X a n d 3 x I 0 5MAPV, six out of 10 SDCN neurons responded to the electrical stimulation applied through the concentric stimulating electrode located within the SDCN. The current responses were completely antagonized by 3 x 1 0 ¢'M strychnine (Fig. llA), thus suggesting the existence of functional glycinergic synaptic inputs to the SDCN neurons. NA (10 7 M ) applied to the perfusate potentiated the amplitude of the evoked strychnine-sensitive postsynaptic currents (135 =k 10%; n=8).

36

J. Nabekura et al.

Forskolln 3xlO-*

GUY

3xlosm

m

b

H-89lO-Q

C

7-f i-f- ’ i

1-f--

B

Im IBMX+db-cAMP I I

0.5 t C 1.5 -

E t % 0

l.O-

ii

B 0.5 0

10

20 Time (min)

Fig. I.

30

Noradrenergic modulation of glycine response

A

NA

Gly 3xl0-SM

Forskolin 3xlO'SM lmOO'7M I¢/'////////f/////////A

5

B

7 /

J

IBMX 10"SM + db-cAMP 3xl0"SM

"

J

;iT, I-I-89 IO'SM } r fl l l l l l l ,

I

~

~¢[[llllllllllllllll/]

I

C

8 mln

B

J

facilitated strychnine-sensitive glycine-induced C1 response in the isolated SDCN neurons. The c~l antagonist prazosin, which has also been nominated as an antagonist for c~2B- and ct2C-adrenoceptors, 3° partially blocked the NA potentiation on Ioly. Yohimbine completely blocked the effect of NA on the Io~y. On the other hand, clonidine mimicked the NA potentiation of IG]y. These results thus suggest that the c~2A- and probably c~2D-subtypes are mainly involved in NA action, although the additional involvement of a2B- and ct2C-adrenoceptors could not be ruled out. The present results also correlate with the findings of a recent in situ hybridization study in which the ct2A-subtype is the primary ct2adrenoceptor found in the substantia gelatinosa of both the spinal cord and the trigeminal nucleus.~9

Protein kinase A mediates the noradrenaline-induced potentiation of glycine-induced current

m

f ~J'lllllll~l m

37

m

155

Fig. 8. Involvement of PKA on IGly. NA (10 7 M) failed to restore the IGly suppressed by 3 x 10-5 M forskolin (A) or the mixture of 10 s M IBMX and 3 x 10 5 M dibutyl cyclic AMP (B). (C) During the potentiation of Ioly by 10 -6 M H-89, NA did not produce any further increase of IGly. DISCUSSION

The present study has demonstrated that NA facilitates IGly in SDCN neurons. An a2-adrenoceptor and lAP-sensitive G-proteins mediate the effect, and an inhibitor of PKA mimicks and occludes the effect, whereas activators of adenylate cyclase or PKA prevent the effect, presumably by bypassing the receptor-mediated mechanism. It is concluded that NA reduces intracellular cAMP levels by inhibiting adenylate cyclase, which in turn reduces PKA activity, resulting in the enhancement of the glycine receptor-mediated response in SDCN neurons.

cL2-adrenoceptor-mediated enhancement of the glycineinduced current Noradrenergic receptors are classically divided into c~l, a2, [31 and [32 subtypes. Furthermore, a2adrenoceptor has been subclassified into ct2A, a2B, c~2C and probably a2D subtypes.~7 The NA potently

The potentiation effect of NA on IGly was mimicked with the PKA inhibitor H-89 and prevented by activators of adenylate cyclase or PKA, indicating that the NA-induced IGly increase was mediated by the inhibition of cAMP PKA pathway. On the other hand, an increase of PKC activity also potentiates IGly in the SDCN neurons42 (Fig. 6A). The IGly potentiated by a PKC activator, PMA, could be further enhanced by adding NA (Fig. 6Ba). In addition, the Io]y suppressed by the PKC inhibitor was potentiated by NA (Fig. 6Bb). Therefore, the PKC cascade does not seem to be the principle pathway in the NA action on the glycine response. However, intensive observations revealed that the potentiating rate by NA in the presence of PKC modulators is less than that in their absence (Figs 2 and 6B, also see Results). Liu and Simon24 also demonstrated that PKA phosphorylates some type of phospholipase C (PLC) and thus inhibits its activity. Therefore, besides the intracellular pathway without PKC, another pathway might also exist in which a decrease in the PKA activity by NA may lead to the increase in PLC activity and PKC, thus resulting in the potentiation of the glycine response. 12

Possible mechanisms of protein kinase A-mediated modulation of glycine-induced current by noradrenaline Over the past several years, it has been well established that the phosphorylation of the

Fig. 7. Intracellular cAMP-dependent PKA-mediated potentiation of IGly. (A) The effects of PKA activator and inhibitor on IGly. The long-lasting inhibition of Iojy by 3 x 10 5 M forskolin (Aa) or the mixture of 10 5M IBMX and 3 x 10-SM dibutyl cyclic AMP (db-cAMP, Ab). (Ac) H-89 (10 -6 M) long-lasting potentiated lGly. The current measurements were made from three different neurons. (B) The effects of PKA activator and inhibitor on IGly. All responses were normalized to the peak response induced by 3 x 10-5 M glycinealone. The inactive analog, 1,9-dideoxyforskolin,did not affect the amplitude of Ioly. Each point and bar represents the mean + S.E.M. of four to seven measurements. The arrow indicates the start of washing out the PKA modulators. (C) The intracellular application of 1 mM cAMP in conventional whole-cell recording mode gradually decreased the peak amplitude of the 3 x 10- 5 M glycineresponse.

38

J. Nabekura et al.

Aa

B

Control Gly 10"KM BB

NA 10.7M m

Potentiating rate of IGly

mml

NA without lAP

J lAP-treated m

NA with lAP

m m

72 ± 13% (n=7)

2 _+1%

(n=7)

15s

Fig. 9. Effect of lAP treatment on the potentiation of IGly by NA. (A) Facilitatory effect of 10 7 M NA on IGly (Aa) disappeared in the neurons treated with lAP (300 ng/ml) for 6 to 8 h (Ab). (B) Potentiating rate of IGly by 10 -7 M NA with and without IAP treatment. There is a statistically significant difference between the two groups (P<0.01).

+ N A 10"7M

control

50ms Fig. 10. Single CI channel currents in a cell-attached mode before (left traces) and during 10 7 M NA applied by Y-tube (right traces). The pipette solution contained the standard extracellular solution with 3 x 10 7 M glycine. The pipette potential was - 6 0 inV. The dashed lines indicate the open state current levels.

i•

strychnine 3x10"6M

b/./ ~

100ms

I

Fig. 11. NA effects on the strychnine-sensitive postsynaptic currents (PSCs). The strychnine-sensitive postsynaptic current could be elicited by an electrical stimulation applied within the SDCN. NA (10 .7 M) applied in the perfusate potentiated the peak amplitude of evoked PSCs. Each current trace of PSC was obtained after averaging five successive PSCs in each condition. The extracellular solution contained 10 6 M bicuculline, 3 x 10 5 M CNQX, and 3 x 1 0 5M APV. VH= 40mV.

n e u r o t r a n s m i t t e r - a c t i v a t e d receptor i o n o p h o r e complexes is a c o m m o n m e c h a n i s m for the regulation o f receptor function. 23"33 The G A B A A receptor, N-methyl-D-aspartate ( N M D A ) a n d n o n - N M D A receptors as well as the glycine receptor have been d e m o n s t r a t e d to be p h o s p h o r y l a t e d by the intracellular P K A a n d / o r PKC. 33 The glycine receptor is c o m p o s e d of a c o m b i n a t i o n with two kinds of h o m o l o g o u s polypeptide, a a n d [3, as integral m e m b r a n e - s p a n n i n g subunits a n d these subunits have been further classified into several subclasses. A l t h o u g h n o n e o f the four c~ a n d two [3 subunits identified up to n o w contains a strong consensus site for P K A p h o s p h o r y l a t i o n , the strychnine-sensitive structure within the glycine receptor has been rep o r t e d to be p h o s p h o r y l a t e d by P K A as well as P K C . 38 Therefore, the m o s t s t r a i g h t f o r w a r d explanation for o u r results is that P K A p h o s p h o r y l a t e s the glycine receptor directly, decreasing the function of the glycine receptor in the S D C N neurons. Nevertheless, P K A m a y p h o s p h o r y l a t e a regulatory protein which in turn indirectly alters IGly.

Noradrenergic modulation of glycine response Another possibility is that P K A phosphorylates a cytoskeletal protein that influences the glycine receptor channel activity. Indeed, the polymerization state of microtubes has functional consequences on the glycine-evoked currents in cultured spinal cord neurons, m At the subsynaptic level, the glycine receptor is associated with an intracellular protein, gephyrin, which acts as a linker tethering the receptor to the microtube network. A recent study suggests that the intracellular regulatory interaction between the glycine receptor-gephyrin complex and the cytoskeleton could thus be an important mechanism for glycine receptor modulation.~°

Desensitization o f noradrenaline-induced induced current increase

glycine-

In the present experimental conditions for recording IGJy, 10 5M glycine evoked a current that showed little desensitization. However, the sustained pretreatment with clonidine as well as N A gradually accelerated the inactivation (desensitization) of IGly following the acute facilitation of IGly in the SDCN neurons (Fig. 3Aa), thus suggesting that both the enhancement and the acceleration of desensitization by N A shared the c~2 receptor subtype. The desensitization increased in a concentration-dependent manner, thus probably giving a bell-shaped dose-response curve x7 (Fig. 2A). Many G protein-coupled receptor systems also display desensitization to agonist action after repeated or persistent stimulation. ~2'15 In cardiac muscle, for example, [31-adrenergic receptors desensitize as a result of their phosphorylation by P K A (heterogeneous desensitization) or G protein-coupled receptor kinases (GRKs) (homologous desensitization), j4 The former can proceed in the absence of agonist-receptor binding, but homologous desensitization occurs only in the presence of high agonist concentrations. 16 It is thus likely that similar mechanisms control the efficacy of NA-induced Ic~y increase. Homologous desensitization is of particular interest, since NA-induced desensitization became faster as the N A concentrations increased and since G R K s are expressed widely and abundantly in the nervous system 3 Although the c~2-adrenoceptor is known to be the substrate of GRKs, 5"~2 the underlying biochemical explanation for NA-induced desensitization is unknown, and its exploration is complicated by several properties inherent in the signaling pathways.

Comparison with previous results The most closely function-related region in which the modulation of IGly by intracellular P K A has been examined in the trigeminal nucleus which is analogous to the dorsal horn in the spinal cord. Song and Huang 36 found that the activation of the P K A

39

potentiated the IGly in cultured spinal trigeminal neurons. This finding is in apparent contradiction to the present study. However, our study differed in several aspects. Firstly, although the subunit compositions of glycine receptor of the cells studied here are unknown, we used the SDCN neurons which are cytochemically and biologically different from the neurons of the superficial dorsal horn. 4'7 Thus, t h e apparent discrepancy is probably due to such biologically relevant factors as the heterogeneity of the subunit compositions of glycine receptor in different brain areas. 6 Secondly, in the substantia nigra neurons 2°'27 and ventromedial hypothalamic neurons, l the activation of PKA has also been reported to reduce the 1Gly. The control of channel properties by phosphorylation is complex and depends on many factors. The different mechanisms after PKA activation to modulate such a glycine response as PKA itself, 3~ phospholipase C 24 and phosphatase 9 in different preparations may also have contributed to the discrepancy between the present and previous results. Thirdly, in other preparations, 36'3~ cultured neurons were generally used. Although we employed enzyme treatment to dissociate the neurons, this treatment does not seem to interfere with the neurons regarding the NA effect on the glycine response, because the mechanically dissociated neurons exhibited quite a similar potentiation of the glycine response by NA. The time-span of primary cultures is considerably prolonged in comparison to the rapid turn-over rate of glycine receptor channel proteins, which is of the order of one to two days. ~8 The characteristics of the receptors expressed in the cultured neurons are also susceptible to environmental changes, so that there is a possibility that cultured neurons might not exactly reflect the properties under in vivo conditions. Functional considerations The descending pathways from supraspinal sites to the dorsal horn of the spinal cord are important in endogenous antinociception. Two brain areas implicated in endogenous antinociception are the nucleus raphe magnus and the locus coeruleusY The excitation of these two areas, either by electrical stimulation 2 or iontophoresis of glutamate, 22 produces an antinociceptive effect. The N A from the locus coeruleus reduces the response of the dorsal horn neurons to noxious stimuli. In line with the role of N A in antinociception, the intrathecal administration of N A causes antinociception and NA antagonist reduces antinociception produced by supraspinal stimulation (reviewed in Ref. 22). Glycine is an inhibitory transmitter in the spinal cord. 43 Indeed, six out of 10 SDCN neurons responded with strychninesensitive synaptic currents in the present experiment (Fig. 11). Recently, there have been reports that the

40

J. Nabekura et al.

adenylate cyclase cascade regulates spinal synaptic transmission, s'32 The i n n e r v a t i o n o f the sensory afferent to the S D C N region has been d e m o n s t r a t e d b o t h m o r p h o l o g i c a l l y 35 a n d physiologically, j3 a n d the noradrenergic descending i n p u t to this region is also present. 31 In addition, N A p o t e n t i a t e d the evoked glycinergic postsynaptic current, IPSCs, in a m p l i t u d e (Fig. 11). As a result, the facilitation of the glycine response by N A t h r o u g h a c A M P - d e p e n d e n t

m e c h a n i s m m i g h t thus be one o f the antinociceptive m e c h a n i s m s in the S D C N in vivo.

Acknowledgements We would like to give our thanks to Dr B. Quinn for the critical reading of the manuscript. This study was funded by Grants-in-Aids for Scientific Research (Nos 07276101 and 07407002) to N. Akaike and (Nos 10156231 and 09670046) to J. Nabekura from The Ministry of Education, Science and Culture, Japan.

REFERENCES

1. Agopyan N., Tokutomi N. and Akaike N. (1993) Protein kinase A-mediated phosphorylation reduces only the fast desensitizing glycine current in acutely dissociated ventromedial hypothalamic neurons. Neuroscience 56, 605-615. 2. Aimone L. D. and Gebhart G. F. (1986) Stimulation-produced spinal inhibition from the midbrain in the rat is mediated by an excitatory amino acid neurotransmitter in the medial medulla. J. Neurosci. 6, 1803-1813. 3. Arriza J. L., Dawson T. M., Simerly R. B., Marin L. J., Caron M. G., Snyder S. H. and Lefkowitz R. J. (1992) The G protein-coupled receptor kinases [3ARK1 and [3ARK2 are widely distributed at synapses in rat brain. J. Neurosci. 12, 4045-4055. 4. Basbaum A. I. (1988) Distribution of glycine receptor immunoreactivity in the spinal cord of the rat: cytochemical evidence for a differential glycinergic control of lamina I and V nociceptive neurons. J. comp. Neurol. 278, 330-336. 5. Benovic J. L., Regan J. W., Matsui H., Mayor F., Cotecchia S., Leeb-Lundberg L. M. F., Caron M. G. and Lefkowitz R. J. (1987) Agonist-dependent phosphorylation of the c~2-adrenergic receptor by [3-adrenergic receptor kinase. J. biol. Chem. 266, 17,251 17,253. 6. Betz H. (1992) Structures and function of inhibitory glycine receptors. Q. Rev. Biophysics 25, 381 394. 7. Bohlhalter S., Weinmann O., Mohler H. and Fritschy J. M. (1996) Laminar compartmentation of GABAA-receptor subunits in the spinal cord: an immunohistochemical study. J. Neurosci. 16, 283-297. 8. Cerne R., Jiang M. and Randic M. (1992) Cyclic adenosine 3',5'-monophosphate potentiates excitatory amino acid and synaptic responses of rat spinal dorsal horn neurons. Brain Res. 596, 111-123. 9. Colwell C. S. and Levine M. S. (1995) Excitatory synaptic transmission in neostriatal neurons: regulation by cyclic AMP-dependent mechanisms. J. Neurosci. 15, 1704-1713. 10. Delon J. and Legendre P. (1995) Effects of nocodazole and taxol on glycine evoked currents on rat spinal cord neurones in culture. NeuroReport 6, 1932 1936. 11. Ding Y. Q., Qin B. Z., Li J. S. and Mizuno N. (1994) Induction of c-fos-tike protein in spinoparabrachial tract-neurons locating within the sacral parasympathetic nucleus in the rat. Brain Res. 659, 283-286. 12. Diverse-Pierluissi M., lnglese J., Stoffel R. H., Lefkowitz R. J. and Dunlap K. (1996) G protein-coupled receptor kinase mediates desensitization of norepinephrine-induced Ca 2+ channel inhibition. Neuron 16, 579 585. 13. Fitzgerald M. (1985) The post-natal development of cutaneous afferent fibre input and receptive organization in the rat dorsal horn. J. Physiol., Lond. 364, 1 18. 14. Freedman N. J., Liggett S. B., Drachman D. E., Pei G., Caron M. G. and Lefkowitz R. J. (1995) Phosphorylation and desensitization of the human [31-adrenergic receptor: involvement of G protein-coupled receptor kinases and cAMP-dependent protein kinase. J. biol. Chem. 270, 17,953-17,961. 15. Haga T., Haga K. and Kameyana K. (1994) G protein-coupled receptor kinases. J. Neurochem. 63, 400-411. 16. Hausdorf W. P., Caron M. G. and Lefkowitz R. J. (1990) Turning off the signal: desensitization of the 13-adrenergic receptor. Fedn Am. Socs exp. Biol. J. 4, 2881 2889. 17. Hein L. and Kobilka B. K. (1995) Adrenergic receptor signal transduction and regulation. Neuropharmacology 34, 357-366. 18. Hoch W., Betz H. and Becker C. M. (1989) Primary cultures of mouse spinal cord express the neonatal isoform of the inhibitory glycine receptor. Neuron 3, 339-348. 19. Honda C. N. (1985) Visceral and somatic afferent convergence onto neurons near the central canal in the sacral spinal cord of the cat. J. Neurophysiol. 53, 1059 1078. 20. Inomata H., Nabekura J. and Akaike N. (1993) Suppression of taurine response in acutely dissociated substantia nigra neurons by intracellular cyclic AMP. Brain Res. 615, 347 350. 21. Jones B. E. and Yang T. Z. (1985) The efferent projections from the reticular formation and the locus coeruleus studied by anterograde and retrograde axonal transport in the rat. J. comp. Neurol. 242, 56-92. 22. Jones S. L. (1992) Descending control of nociception. In The Initial Processing o f Pain and its Deseending Control: Spinal and Trigeminal Systems (ed. Light A. R.), pp 203-295. Karger, Basel. 23. Lev S., Moreno H., Martinez R., Canoll P., Peles E., Musacchio J. M., Plowman G. D., Rudy B. and Schlessinger J. (1995) Protein tyrosine kinase PYK2 involved in Ca2+-induced regulation of ion channel and MAP kinase functions. Nature 376, 737 745. 24. Liu M. and Simon M. I. (1996) Regulation of cAMP-dependent protein kinase of a G-protein-mediated phospholipase C. Nature 382, 83-87. 25. Lu Y., Jin S. X., Xu T. L., Qin B. Z., Li J. S., Ding Y. Q., Shigemoto R. and Mizuno N. (1995) Expression of c-Jbs protein in substance P receptor-like immunoreactive neurons in response to noxious stimuli on the urinary bladder: an observation in the lumbosacral cord segments of the rat. Neurosei. Lett. 198, 139 142. 26. Murase K., Randic M., Shirasaki T., Nakagawa T. and Akaike N. (1990) Serotonin suppresses N-methyl-D-aspartate responses in acutely isolated spinal dorsal horn neurons of the rat. Brain Res. 525, 84-91. 27. Nabekura J., Omura T. and Akaike N. (1996) Alpha 2 adrenoceptor potentiates glycine receptor-mediated taurine response through protein kinase A in substantia nigra neurons. J. Neurophysiol. 76, 2445 2447.

Noradrenergic modulation of glycine response 28.

41

Nabekura J., Omura T., Horimoto N., Ogawa T. and Akaike N. (1996) Alpha 1 adrenoceptor activation potentiates taurine response mediated by protein kinase C in substantia nigra neurons. J. Neurophysiol. 76, 2446-2460. 29. Nicholas A. P., Pieribone V. and Hokfett T. (1993) Distributions of mRNAs for alpha-2 adrenergic receptor subtypes in rat brain: in situ hybridization study. J. comp. Neurol. 328, 575-594. 30. Ordway G. A., Jaconetta S. M. and Halaris A. E. (1993) Characterization of subtypes of alpha-2 adrenoceptor in the human brain. J. Pharmac. exp. Ther. 264, 967-976. 31. Rajaofetra N., Poulat P., Marlier L., Geffard M. and Privat A. (1992) Pre- and postnatal development of noradrenergic projections to the rat spinal cord: an immunocytochemical study. Devl Brain Res. 67, 237--246. 32. Randic M., Cheng G. and Kojic L. (1995) k-Opioid receptor agonists modulate exitatory transmission in substantia gelatinosa neurons of the rat spinal cord. J. Neurosci. 15, 6809-6826. 33. Raymond L. A., Blackstone C. D. and Huganir R. L. (1993) Phosphorylation of amino acid neurotransmitter receptors in synaptic plasticity. Trends Neurosci. 16, 147-153. 34. Schonrock B. and Bormann J. (1995) Modulation of hippocampal glycine receptor channels by protein kinase C. NeuroReport 6, 301-304. 35. Smith C. L. (1983) The development and postnatal organization of primary afferent projections to the rat thoracic spinal cord. J. comp. Neurol. 220, 2943. 36. Song Y. and Huang L. Y. M. (1990) Modulation of glycine receptor chloride channels by cAMP-dependent protein kinase in spinal trigeminal neurons. Nature 348, 242-245. 37. Todd A. J. (1990) An electron microscope study of glycine-like immunoreactivity in laminae I-III of the spinal dorsal horn cf the rat. Neuroscience 39, 387-394. 38. Vaello M. L., Ruiz-Gomez A., Lerma J. and Mayor F. Jr (1994) Modulation of inhibitory glycine receptors by phosphorylation by protein kinase C and cAMP-dependent protein kinase. Ji biol. Chem. 269, 2002 2008. 39. Vizzard M. A., Erdman S. L. and de Groat W. C. (1995) Increased expression of neuronal nitric oxide synthase (NOS) in visceral neurons after nerve injury. J. Neurosci. 15, 40334045. 40. Willcockson W. S., Chung J. M., Hori Y., Lee K. H. and Willis W. D. (1984) Effects of iontophoretically released amino acids and amines on primate spinothalamic tract cells. J. Neurosci. 4, 732-740. 41. Willis W. D. Jr and Coggeshall R. E. (1991) Functional organization of dorsal horn interneurons. In Sensory Mechanisms o f the Spinal Cord, pp. 153 215. Plenum, New York. 42. Xu T. L., Nabekura J. and Akaike N. (1996) Protein kinase C-mediated enhancement of giycine response in rat sacral dorsal commissural neurones by serotonin. J. Physiol., Lond. 496, 491 501. 43. Yoshimura M. and Nishi S. (1995) Primary afferent-evoked glycine- and GABA-mediated IPSPs in substantia gelatinosa neurones in the rat spinal cord in vitro. J. Physiol., Lond 482, 29 38. (Accepted 19 May 1998)