Neuroscicm'e Letters, 165 (1994) 71 74 71 :C, 1994 Elsevier Science Ireland Ltd. All rights reserved 0304-3940/94/$ 07.00

NSL 10100

Nitric oxide evokes pain in humans on intracutaneous injection

Holger Hol thusen*, Joach im Ot to Arnd t

Heinrieh-Heine-UniversileTt Dfisseldorf ln,s'litul liir Experimenlelh' .4mw,sthc,~ioh~gi¢', Geh. 23.02.01. PostJach 10100, 40001 Dfi.s,sehlor/i f'RG

(Received 3 .luned 1993; Revised version received 14 October 1993: Accepted 15 October 1993)

Key words: Nociception" Nitric oxide: Human

To test the hypothesis that nitric oxide (NO) acts algetically in humans, we determined pain intensity/dose relations for intracutaneously applied NO solutions. NO, dissolved in isoosmolar phosphate buffer, was injected in the forearm of six volunteers and the subjects rated NO-ew)kcd pain continuously with the help of an electronically controlled visual analogue scale. Pain always occurred at a NO dose of 12 nmol, increased with dose and reached the tolerance maximum at 50 nmol. This shows for the tirst lime the genuine pain evoking properties of NO.

Our study focuses on the effects of nitric oxide (NO) on cutaneous nociceptors. NO, which is released fi~om k-arginine by the action of the NO synthase, is thought to play a role in the processing and perhaps also in the gen- eration of pain. This view rests mainly on observations on animals, in which NO synthesis has been manipulated with NO synthase inhibitors, mostly with N"-nitro-i. - arginine methyl ester (k-NAME) or the NO precursor L-arginine. Consistent with expectations, nitric oxide synthase inhibitors attenuated behavioral pain responses when applied parenterally [9, 14], topically to the spinal cord [9, 12], and also intracerebroventricularly [14] ex- cept in one study [2]. Unexpectedly, however, L-arginine which usually reversed these effects also acted analgeti- cally by itself, on parenteral application in animals [6, 7] and in humans [5], as well as on topical application to the spinal cord of rats [7]. In the latter preparation both L- arginine and I.-NAME reduced spike traffic in nocicep- tive spinal neurons [4].

Regardless of the controversial effects of L-arginine, most observations suggest NO to play a part in the cen- tral processing of pain. Its role at peripheral nociceptors, however, is less clear. For example, carrageenin-induced hyperalgesia was attenuated but not aggravated by the injection of l-arginine into pretreated skin areas of rats [6]. Also, subcutaneous injections of a NO synthase in- hibitor decreased rather than increased the action of an- algesics [3]. Nevertheless, injection of L-NAME into the receptive field of spinal nociceptive neurons decreased

* ('orrespond#lg authm:

the formalin-evoked spike traffic in these neurons [4], suggesting that NO is a chemical link in nociception. Re- gardless of the reasons for the peculiar analgesic action of l.-arginine, available information on the effects of NO rest entirely on inferences about its local concentrations. In our attempt to test NO for its genuine pain evoking properties via cutaneous nociceptors, we therefore deter- mined pain intensity/dose relations for intracutancously applied NO solutions in humans.

The data derive from six experiments oll six men. NO solutions at doses of 5, 12, 25, and 50 nmol of equal volume (always 50 pl) were injected (at random: ever\. 10 15 min) intracutaneously with a gas-tight syringe (Hamilton) via a hypodermic needle (gauge 27) al wtri- ous sites of the volar forearm. The subjects rated pain between threshold (0%) and maximum of tolerance (100%) with an electronically controlled visual analogue scale (VAS) which permitted continuous recording of the time course of pain intensity.

Stock solutions of NO were prepared by saturating isoosmolar, deoxygenated phosphate buffer (pH 7.3) with nitric oxide (Linde): the NO concentration in the stock solution was measured by the oxyhemoglobin method [8]. Appropriate dilutions with deoxygenated phosphate buffer were kept under argon atmosphere in vials sealed with silicone rubber [16].

All experiments were performed between 09.00 tl and 13.00 h with the subjects sitting semi-recumbent at a thermoneutral room temperature of 24 25 ° C.

NO evoked pain at the site of injection in all subjects. This is shown by the individual pain intensity/time curves in Fig. I. These curves exhibited a biphasic pattern with

72

100 ] , NO: 12 nmol

0'1t ' ~ / ',~ , / . / - , , . - - . l : : : l G , " " , ,

100 ,, N6; 55 nmol]

It ~ 5O

? o o q,.

z /.,.,;, : : . ,

z 0 • Y - ' ~ "~ " ~*, -

[ 0 0 . . . . . . . °

• ' / ,m ,

~,, ~ ° °

NO: 50 nmol

0 o 0 30 60 90

T I M E [ s ]

Fig. 1. Time course and intensity of pain on intracutaneous injection of nitric oxide (NO). NO, dissolved in isoosmolar phosphate buffer, was injected intracutaneously at various doses but equal volume (50 #1) in six volunteers who rated pain on a visual analogue scale (VAS); thresh- old = 0%, maximum of tolerance = 100%. Shown are the pain inten- sity/time curves for each subject. Note, that after disappearance of the first pain coincident with injection both the intensity and duration of

pain increase with the dose of NO.

an early peak immediately on injection, which was fol-

lowed in every case by a pronounced second peak. The first pain, described as sharp and pricking, likely results

from mechanical distension of the skin by the injection

process as it is independent of dose both in duration and

intensity. By the same token, the second pain, always

described as burning, reflects in all likelihood the al-

gogenic properties of NO, because both the duration and intensity of pain increased with NO doses. The second pain usually occurred with a delay of 10 to 20 s after the start of injection, reached the maximum within 20 to 50 s, and faded within 90 s at the most.

In each subject, the pain intensity increased in a pre- dictable manner with the NO doses applied, as is shown by the individual pain intensity/dose relations in Fig. 2. NO at a dose of 5 nmol never evoked pain. However, in every subject, pain of intensities between 2-19% VAS occurred at a NO dose of 12 nmol and increased in a

dose-related fashion to reach the tolerance maximum in

four of the six subjects at a dose of 50 nmol. Thus, the

threshold dose for NO is in the range between 5 and 12 nmol and about 5- to 10-fold higher doses are required

for attaining the tolerance maximum.

Reddening of the skin but without swelling and itching

was always observed around the NO injection sites. This

effect occurred with a delay of 10 to 15 s alter injection,

spread centrifugally within approximately 2 min and

faded without sequelae in the following 5 min at the

most. The diameter of the reddened area increased with

the dose of NO and attained a maximum of as much as

6 cm.

In humans, intracutaneously applied NO evokes pain

in a dose-related manner. This pain is described as burn-

ing at the injection site, i.e. it does not spread with the

accompanying flare. These flares outlasted pain by sev- eral minutes but they were never accompanied with an

early white response or with skin edema as is characteris-

tic of the classical triple response. Therefore, these flares

are in all likelihood manifestations of the endothelium-

mediated vasoditator action of NO [15] rather than due

to secondarily released algogens or mediators of inflam-

mation. By activating the NO synthase, these substances

are known to increase NO synthesis but there is hitherto

no evidence that they would be released by NO. Thus, the possibility is remote that pain in our experiments

.,r

[..

Z [.. Z

Z <

100

50.

/

0 20 40 60

N O - D O S E [nmol]

Fig. 2. NO-dose/pain intensities relations. NO, dissolved in isoosmolar phosphate buffer was injected intracutaneously in the volar forearm of six men. Shown are for each subject the pain maxima reached at a given dose. Note, that pain occurred always at a NO dose of 12 nmol and reached the tolerance maximum in most subjects at a dose of 50 nmol.

73

may have resulted indirectly from a NO-triggered release of algogens or mediators of inflammation or blood flow- related alteration of their local concentrations.

Intracutaneously injected bradykinin, a natural al- gogen and NO synthase activator, evokes pain in a dose- related manner in humans [11], and it also increases the spike traffic in polymodal nociceptors of animals [13]. Therefore, our approach of injecting intracutaneously deoxygenated phosphate buffer (pH 7.3:300 mosM) containing NO at various concentrations should have been appropriate for studying the effect of NO on cuta- neous nociceptors, i.e. for determining both the time course and dose dependency of NO-evoked pain.

In fact, the second pain apparently reflects the action of NO, because it depends on dose, unlike the first pain, which in agreement with others [1, 11] results from skin stretching concomitant with the injection process.

The time course of the NO-evoked pain, in particular its delayed onset and the prolonged duration can be ex- plained by pharmacokinetics, i.e. by diffusion of NO from the injected pool of deoxygenated and highly con- centrated NO solution (0.1 l mM) to the site of action. In fact, our intensity/time curves conform with that seen on intracutaneous injection of other algogens, particu- larly bradykinin [I1], which is known to release NO within seconds [10]. Without reliable information on the length of the diffusion path and the inactivation rate of NO, which is several seconds only in oxygenated biologi- cal tissues but is several minutes in oxygenated solutions [16], we are hesitant even to speculate about the pain- evoking concentrations at the nociceptors. Nevertheless. the NO concentrations at the sites of injection and action most likely vary in parallel, so that our observations jus- tify the conclusion that NO evokes pain by exciting cuta- neous nociceptors in humans.

This is consistent with those neurophysiological and behavioral studies on animals in which NO synthase in- hibitors have been utilized for reducing local NO concen- trations at peripheral nociceptors. For instance, L- NAME by its local action attenuated carrageenin-in- duced hyperalgesia [4] and also reduced the formalin- evoked spike traffic in spinal nociceptive neurons in rats [4]. At first glance, it is therefore enigmatic why the NO precursor substance L-arginine, which should increase NO concentrations, also acts analgetically, as on spinal [4] and supraspinal application [2, 7], as well as on paren- teral administration in animals [6, 7]. The difficulty ap- parently resides in the formation of kyotorphin [6, 7] from L-arginine, which as an endogenous analgetic neu- ropeptide probably counteracts the action of NO. By ap- plying NO directly into the skin we could circumvent this problem and show that pain intensity correlates with the locally applied doses of NO.

Our observations together with available information on the antinociceptive effects of locally applied NO syn- thase inhibitors are compatible with the hypothesis that NO plays a role as a chemical link in the generation of pain at peripheral nociceptors. Since subcutaneously in- jected NO synthase inhibitors also reduce spike traffic in electrically activated spinal neurons [4], we infer that NO excites nociceptors directly rather than indirectly. Re- gardless of this inference, we show for the first time that NO evokes pain in humans, probably by exciting poly- modal nociceptors of the skin directly.

We thank Professor S.J. Cleveland (Department of Physiology, Heinrich-Heine Universit/it) for his advice and help with the English.

1 Armstrong, D., Dry, R.M.L., Keele, C.A. and Markham, J.W., Observations on chemical excitants of cutaneous pain in man, J. Physiol., 120 (1953) 326 351.

2 Duarte, I.D.G. and Ferreira, S.H., The molecular mechanism of central analgesia induced by morphine or carbachol and the L- arginine-nitric oxide-cGMP pathway, Eur. J. Pharmacol., 221 (1992) 171 174.

3 Ferreira, S.H., Duarte. I.D.G. and Lorenzetti, B.B., The molecular mechanism of action of peripheral morphine analgesia: stimulation of the cGMP system via nitric oxide release, Eur. J. Pharmacol., 201 (1991) 121 122.

4 Haley, J.E., Dickenson, A.H. and Schachter, M., Flectrophysiologi- cal evidence for a role of nitric oxide in prolonged chemical nocicep- lion in the rat, Neuropharmacology, 31 (1992) 251 258.

5 Harima, A., Shimizu, H. and Takagi, H., Analgesic effect of i- arginine in patients with persistent pain, Eur. Neuropsychopharma- col., 1 (1991)529 533.

6 Kawabata, A., Fukuzumi, Y., Fukushima, Y. and Takagi, H., Antinociceptive effect of i.-arginine on the carrageenin-induced hy- peralgesia of the rat: possible involvement of central opioidergic systems. Eur. J. Pharmacoh, 218 (1992) 153 158.

7 Kawabata, A., Nishimura, Y. and Takagi, T., L-Leucyl-L-argininc, naltrindole and D-arginine block antinociceplion elicited by 1 -argin- ine in mice with carrageenin-induced hyperalgesia, Br. J. Pharma- col., 107(1992) 1096 1101.

8 Kelm, M. and Schrader, J., Control of vascular tone by nitric oxide, Circ. Res.. 66 (1990) 1561 1575.

9 Kitto. K.F.. Haley, J.E. and Wilcox. G.L., Involvement of nitric oxide in spinally mediated hyperalgesia in the mouse, Neurosci. kett., 148 (1992) 1 5.

10 Malinski, T. and Taha. Z., Nitric oxide release from a single cell measured in situ by a porphyrinic-based microsensor, Nature, 358 (1992) 676 677.

11 Manning. D.C., Raja, S.N., Meyer, R.A. and Campbell, J.N., Pain and hyperalgesia after intradermal injection of bradykinin in hu- mans, Clin. Pharmacoh Ther., 50 (1991) 721 729.

12 Meller, S.T. and Gebhart, G.F., Nitric oxide (NO} and nociceptive processing in the spinal cord, Pain, 52 (1993) 127 136.

13 Mizumura, K., SaW, J. and Kumazawa, T., Strong heat stimulation sensitizes the heat response as well as the bradykinin response of visceral polymodal receptors, J. Neurophysiol.. 68 (1992) 1209 1215.

74

14 Moore, RK., Oluyomi, A.O., Babbedge, R,(. , Wallace, P. and Hart, S.L., L-N<LNitro arginine methyl ester exhibits antinocicep- tive activity in the mouse, Br, J. Pharmacol., 102 (1991 } 198 202.

15 Palmer, R.M.J., Ferrige, A.G. and Moncada, S., Nitric oxide re- lease accounts for the biological activity of endothelium-derived relaxing factor, Nature, 11 (1987) 524- 526.

I 6 Taha, Z., Kiechle, F. and Malinski, 'F., Oxidation ol nitric oxide by oxygen in biological systems monitored by porphyrinic sensor. Bio- chem. Biophys. Res, Commun.. 188 (1992) 734 730