Blockade of nitric oxide synthase differentially influences background activity and electrical excitability in rat dorsal horn neurones

ELSEVIER

Neuroscience Letters 188 (1995) 143-146

NEURUSCI[NC[ LETTERS

Blockade of nitric oxide synthase differentially influences background activity and electrical excitability in rat dorsal horn neurones U. Hoheisel, B. Sander, S. M e n s e * Institut fiir Anatomie und Zellbiologie, Im Neuenheimer Feld 307, D-69120 Heidelberg, Germany Received 30 December 1994; revised version received 15 February 1995; accepted 17 February 1995

Abstract

A previous study has shown that inflammation of the gastrocnemius-soleus muscle in rats leads to an increase in excitability of dorsal horn neurones particularly in the spinal segment L3. Here, we have blocked the nitric oxide synthase (NOS) in L3 by spinal cord superfusion with NG-monomethyl-L-arginine (L-NMMA) to find out if this effect is due to a release of nitric oxide (NO). L-NMMA had no influence on the excitability of L3 neurones but caused a marked increase in background activity. The L-NMMA effect on background activity was also present in rats with intact muscle. The data show that the myositis-induced increase in spinal excitability is not mediated by NO. The background activity, however, appears to be strongly dependent on NO production. Keywords: Excitability; Background activity; Nitric oxide; Nitric oxide synthase; Dorsal horn neurones; Inflammation; NG-Mono methyl-L-arginine

A previous study [5] has shown that within 6-8 h following the induction of an experimental myositis of the gastrocnemius-soleus muscle (GSM), distinct changes in electrical excitability and background activity occur in rat lumbar dorsal horn neurones. The changes in excitability (measured as the proportion of cells responding to input from the gastrocnemius-soleus nerves (GSN)) were most marked in the segment L3, i.e. just rostral to the main input region of the GSN. In this segment, the alteration in excitability appeared to be independent of changes in background activity, since background activity in L3 was not affected by the myositis. One possible explanation for the spinal changes is that the increased impulse activity in afferent units from the inflamed GSM releases - or induces production o f - substances in the spinal cord which alter the discharge behaviour of dorsal horn neurones. One of the candidates for such an action is nitric oxide (NO). NO can be synthesized by the neuronal NO synthase (NOS) and is assumed to be involved in neuroplastic processes such as long-term potentiation (LTP) and induction of oncogene expression (for recent reviews, see Refs. [ 16,19]). NO also appears to * Corresponding author, Tel: +49 6221 564193; Fax: +49 6221 563071.

play an important (albeit controversial) role in the spinal processing of nociceptive information [9,10,17]. In the present investigation, the NOS inhibitor Ammonomethyl-L-arginine (L-NMMA) was administered topically to the segment L3 to find out whether the myositis-induced spinal changes are dependent on the synthesis of NO. The experiments were performed on 26 adult SpragueDawley rats. For methods of anaesthesia and surgical preparation, see Ref. [6]. In short, the animals were anaesthetized with thiopental-sodium (Trapanal R, BYK, 5% in distilled water) 100 mg/kg i.p. initially, followed by repeated doses i.v. (approximately 10 mg/kg every hour) to maintain a deep and relatively constant level of anaesthesia. Muscular relaxation was induced by repeated injections of pancuronium bromide (Pancuronium R, Organon, 0.6 mg/kg per h). The absence of flexor reflexes was checked repeatedly after the effect of the muscle relaxant had worn off. The criterion for the end of action of the relaxant was that the hindlimb muscles twitched with normal thresholds upon electrical stimulation of their nerves. The animals were ventilated with a gas mixture of 47.5% 02, 2.5% CO2, and 50% N2. With the slight hyperventilation used, the arterial pO2 was above 100 mmHg, the pCO2 around 40 mmHg, and the pH close to 7.4. The

0304-3940/95/$09.50 © 1995 Elsevier Science Ireland Ltd. All rights reserved SSDI 0304-3940(95)11411-L

144

U Hoheisel et al I Neuroscience Letters 188 (1995) 143-146

Table 1 Proportion of dorsal horn neurones activated by electrical stimulation of peripheral nerves at A-fibre intensity (muscle inflamed)

Control (CSF) (77) a L-NMMA (70) D-NMMA (36) L-NMMA + L-arginine (61 )

GSN/TIB

PER

SU

(%)

(%)

(%)

45.45 38.57 30.55 36.07

54.55 58.57 52.78 47.54

33.77 3143 27.78 26.23

aNumber of cells responding to the search stimulus in parentheses. Of these, the indicated proportion was activated by a given nerve. In animals with intact GS muscle, L-NMMA likewise did not cause significant changes in the proportion of neurones responding to electrical stimulation of the various nerves.

mean blood pressure was continuously measured via a cannula in the common carotid artery; the experiments were discontinued when the pressure dropped below 80 mmHg. The thoracolumbar spinal cord was exposed and around the segment L3, a small plastic trough was built. The segment was continuously superfused with artificial cerebrospinal fluid (CSF; composition according to Ref. [141) or CSF containing a test substance [61. The following substances were administered at a concentration of 100klM in the superfusate: (1) L-NMMA (Sigma), a specific, competitive inhibitor of NOS [ 12,15]; (2) D-NMMA (Sigma), the enantiomer of L-NMMA; and (3) a mixture of L-NMMA ( 1 0 0 ~ M ) and L-arginine (L-arg, 100~M, Sigma), the endogenous substrate of NOS [11]. According to the available dose-response data of L-NMMA as an NOS inhibitor [2,15] the L-NMMA concentration used should produce an almost complete block of the NOS. LN M M A effects on blood pressure and heart rate were not detected; visual inspection of the exposed spinal vessels gave no evidence for local vasoconstriction (see Ref. [7]). The antagonists were administered for the entire duration of the experiments (6-8 h), starting 30 min before the induction of the inflammation in the GSM by infiltrating it with 0.3 ml of carrageenan 2% (Sigma) in Tyrode. The electrical activity of single dorsal horn neurones in L3 was recorded extracellularly at depths of 2 0 0 1300/~m using micropipettes filled with 0 . 8 6 M NaC1 (DC resistance 7 - 1 5 Mff2). The recordings started 2 h after the induction of the myositis. As a search stimulus, electrical stimulation of the dorsal roots L 3 - L 6 at an intensity supramaximal for A fibres was used (5 V amplitude, 0.3 ms duration, repetition rate 0.33 Hz). The excitability of the neurones was studied with electrical stimulation of the GSN, common peroneal n. (PER), and sural n. (SU). To keep the effect of surgery on NO productton small, the hindlimb skin was anaesthetized before incision. The nerves were carefully exposed (largely by blunt dissection) and put over pairs of plati-

num wire electrodes The proportion of cells activated from a nerve was determined relative to the neurones responding to the search stimulus To avoid stimulus spread to neighbouring nerves the maximal stimulus intensity for testing for A fibre input was 2 5 V (pulse width 0 3 ms)• With GSN stimuli exceeding 1 2 V, also A fibres in the tibial n (TIB) were excited (GSNFFIB input)• For studying GSN effects in isolation, only dorsal horn neurones with a GSN threshold of less than 1 V were accepted For supramaximal stimulation of C fibres, pulses of up to 50 V amplitude were used (for details, see R e f [5]) A neurone was considered to have background activity if at least three action potentials occurred during a period of 3 min before application of external stimuli Statistical comparisons were made using the Z 2 test, Fisher's exact test, and the U t e s t of Mann-Whitney A probability level of less than 5 % (two-tailed) was regarded as significant. The compounds were tested in groups of five animals The tests with D-NMMA were discontinued after having recorded from 36 cells in 3 rats A

Proportion of neurones with GS A-fibre input

% I

35

I

30 25

,,..,.., ..,.~,. ;,t.,- . .'., _ . ° .' .,,..,.. ,.-..-, .~.,., ,.-,.-, : ' t "" • .... ,.-,.-.

20 15 10

[1D

5 0

CSF

L-NMMA

G S intact

B

1

-.-j.. • ..., ..... •" -.-.' .-.- t

I I I

CSF

L-NMMA

I I i l D-NMMA

L-arg

+

L-NMMA

GS inflamed

Proportion of neurones with C-fibre input (all nerves)

% 35

I

I .-.....,

30

• .... .....

25

-- ,z, o.t, . .. . , .,,-., °.....° t. .---. -z2, ."....-. ..' .. .-..-t , --.' ;.'.. ., .;-t .° ' . ' •.". . , ,_t , .". ..-.z-.. . . -. ;...-., .-.. ..-... .'., . t , .'

20 15 10 5 0 CSF

L-NMMA G S intact

CSF

L-NMMA

D-NMMA

@ L-arg + L - N M M A

GS inflamed

Fig. 1. Effects of NOS blockade on the excitability of dorsal horn neurones by electrical stimulation of peripheral nerves. (A) Proportion of cells responding to A-fibre input from the gastrocnemius-soleus nerves. GS intact, results from animals with non-inflamed gastrocnemiussoleus muscle. GS inflamed, results from animals with inflamed muscle. Note that the myositis-induced increase in the proportion of cells responding to GS A-fibre input is not influenced by L-NMMA. (B) Proportion of cells responding to C-fibre input from at least one hindlimb nerve. *P < 0,05. Each calculated percentage is based on recordings from at least 33 neurones

U. Hoheisel et al. / Neuroscience Letters 188 (1995) 143-146

A

Proportion of neurones with background activity

% 6O i

I

I

~

1

5O

..-..-o -..:-.:.-., "-:--3:

4O

.?-.-~-. L , "-.~" • ..~,o ...-..,. ,_-...

3O 2O

10 0

CSF

L-NMMA I

II

CSF

L-NMMA

I I

GS intact

D-NMMA

L-arg + L-NMMA

GSinflamed

I

Frequency of background discharge 300

•k'A" I

"/c'k I

I

I

250

N 2oo +

~ 150

_E 50 i

CSF

L-NMMA ] CSF i I GS intact I I

L-NMMA

D-NMMA

L-arg + L-NMMA

GS inflamed

Fig. 2. Effects of NOS blockade on the background activity of dorsal horn neurones. (A) Proportion of ceils exhibiting background activity. (B) Frequency of background discharge. **P < 0.01; ***P < 0.001. Note that L-NMMA caused a highly significant increase in background activity in animals with both intact and inflamed muscle. The other differences between control (CSF) and the compounds tested were statistically not significant. The data of each bar are based on recordings from at least 33 neurones.

because the data were practically identical to those obtained with CSF. As can be seen from Fig. tA, spinal superfusion with L-NMMA had no influence on the proportion of neurones responding to GSN A-fibre input (which exhibited a significant increase in myositis animals). The A-fibre excitability from the other peripheral nerves was likewise not affected (Table 1). The overall proportion of neurones with C-fibre drive increased (not significantly) in myositis animals (Fig. 1B); this proportion was significantly reduced by the combination of L-NMMA and L-arginine, but not by L-NMMA alone. The other response parameters to electrical stimulation tested (threshold, latency, following frequency, input convergence) were not influenced by L-NMMA. In contrast to the lack of action of L-NMMA on the excitability, the NOS inhibitor caused a highly significant

145

increase in background activity (Fig. 2). This action was stereospecific; superfusion with D-NMMA had no effect. By adding L-arginine to the antagonist, the action of LNMMA could be partially reversed. A peripheral inflammation was not a prerequisite for the L-NMMA effect, since it was present also in animals with intact GSM. The marked increase in background activity observed in the present study following administration of an NOS inhibitor could be explained by assuming that normally there is a continuous release of NO in the spinal cord which tonically depresses the background activity of the dorsal horn neurones. A tonic inhibitory action of NO on spinal nociception has been described [20]. However, there is also evidence for NO to promote the inflammation-induced tonic activity of dorsal horn neurones in a different animal model [3] (for review, see Ref. [10]). In a previous study [5], the most marked effect of an experimental GSM inflammation on L3 neurones was an increase in the proportion of cells responding to GSN Afibre input. Blockade of NOS in the present study had no influence on this effect. Preliminary results (U. Hoheisel, B. Sander and S. Mense, in preparation) show that the increase in excitability can be prevented by superfusion of the spinal cord with antagonists to tachykinin and NMDA receptors. In several studies NO has been described as a pronociceptive agent [13] which is also involved in mechanonociceptive reflexes [17] and hyperalgesia [1,4]. The studies suggest that NO promotes hyperalgesia by increasing the excitability of dorsal horn neurones. In the present investigation, the myositis-induced increase in GSN Afibre input to L3 neurones was found to be independent of NO production. Possible reasons for this difference may be that we used a different animal model and studied not only nociceptive neurones. NO is known to have not only excitability-increasing but also excitability-depressing actions. An example of the latter effect is the long-term depression in the hippocampus [21] and cerebellum [18] (for review, see Ref. [8]). Therefore, the lack of action of L-NMMA on the excitability in the present study could also be due to the fact that excitability-increasing and -decreasing actions of NO cancel each other. The data of this study demonstrate that the myositisinduced increase in neuronal excitability is not due to an action of NO. The enormous increase in background activity following blockade of NOS in animals with intact and inflamed GSM suggests that background activity in the dorsal horn is tonically depressed by NO under physiological and pathophysiological circumstances.

[1] Bj6rkman, R., Halhnan, K.M., Hedner, J., Hedner, T. and Henning, M., Acetaminophen blocks spinal hyperalgesia induced by NMDA and substance P, Pain, 57 (1994) 259-264. [2] Garthwaite, J., Garthwaite, G., Pahner, R.M.J. and Moncada, S., NMDA receptor activation induces nitric oxide synthesis from

146

[3]

[4]

[5]

[6]

[7]

[8] [9]

[10] [11]

[12]

U. Hoheisel et al. / Neuroscience Letters 188 (1995) 143-146

arginine in rat brain slices, Eur. J. Pharmacol., 172 (1989) 413416. Haley, J.E., Dickenson, A.H. and Schachter, M., Electrophysiological evidence for a role of nitric oxide in prolonged chemical nociception in the rat, Neuropharmacology, 31 (1992) 251-258. Hao, J.-X., Xu, X.-J. and Wiesenfeld-Hallin, Z., Systemic N-nitroL-arginine-ester (L-NAME), inhibitor of nitric oxide synthase, relieves chronic allodynia-like symptom in rats with spinal cord lesion, Acta Physiol. Scand., 150 (1994) 457-458. Hoheisel, U., Koch, K. and Mense, S., Functional reorganization in the rat dorsal horn during an experimental myositis, Pain, 59 (1994) 110-118. Hoheisel, U., Mense, S. and Ratkai, M., Effects of spinal cord supeffusion with substance P on the excitability of rat dorsal horn neurons processing input from deep tissues, J. Musculoskeletal Pain (1995) in press. Li, P., Tong, C., Eisenaeh, J.C. and Figueroa, J.P., NMDA causes release of nitric oxide from rat spinal cord in vitro, Brain Res., 637 (1994) 287-291. Linden, D.J., Long-term synaptic depression in the mammalian brain, Neuron, 12 (1994) 457-472. Meller, S.T., Pechmann, P.S., Gebhart, G.F. and Maves, T.J., Nitric oxide mediates the thermal hyperalgesia produced in a model of neuropathic pain in the rat, Neuroscience, 50 (1992) 710. Meller, S.T. and Gebhart, G.F., Nitric oxide (NO) and nociceptive processing in the spinal cord, Pain, 52 (1993) 127-136. Moncada, S., Palmer, R.M.J. and Higgs, E.A., Biosynthesis of nitric oxide from L-arginine. A pathway for the regulation of cell function and communication, Biochem. Pharmacol., 38 (1989) 1709-1715. Moncada, S., The L-arginine: nitric oxide pathway, Acta Physiol. Scand., 145 (1992) 201-227.

[13] Moore, P.K., Oluyomi, A.O., Babbcdge, R.C., Wallace, P. and Hart, S.L., L-NG-nitro arginine methyl ester exhibits antinociceptive activity in the mouse, Br. J. Pharmacol., 102 (1991) 198-202. [14] Philippu, A., Use of push-pull cannulae to determine the release of endogenous neurotransmitters in distruct brain areas of anesthetized and freely moving animals. In C.A. Marsden (Ed.), Measurement of Neurotransmitter Release In Vivo, IBRO Handbook Series: Methods in the Neuroscience, Vol. 6, Wiley, New York, 1984, pp. 3 37. [15] Rees, DD., Palmer, R.M.J., Hodson, H.F. and Moncada, S., A specific inhibitor of nitric oxide formation from L-arginine attenuates endothelium-dependent relaxation, Br. J. Pharmucol., 96 (1989) 418-424. [16] Schuman, E.M. and Madison, D.V., Nitric oxide and synaptic function, Annu. Rev. Neurosci., 17 (1994) 153-183. [17] Semos, M.L. and Headley, P.M., The role of nitric oxide in spinal nociceptive reflexes in rats with neurogenic and non-neurogenic peripheral inflammation, Neuropharmacology, 33 (1994) 14871497. [18] Shibuki, K. and Okada, D., Endogenous nitric oxide release required for long-term synaptic depression in the cerebellum, Nature, 349 (1991) 326-328. [19] Vincent, S.R., Nitric oxide: a radical neurotransmitter in the central nervous system, Prog. Neurobiol., 42 (1994) 129-160. [20] Zhuo, M., Meller, S.T. and Gebhart, G.F., Endogenous nitric oxide is required for tonic cholinergic inhibition of spinal mechanical transmission, Pain, 54 (1993) 71-78. [21] Zhuo, M., Kandel, E.R. and Hawkins, R.D., Nitric oxide and cGMP can produce either synaptic depression or potentiation depending on the frequency of presynaptic stimulation in the hippocampus, NeuroReport, 5 (1994) 1033-1036.