- Email: [email protected]
Enhanced nociceptive behaviour following conditioning injection of formalin in the perioral area of the rat
BRAIN RESEARCH ELSEVIER
Brain Research 676 (1995) 189-195
Research report
Enhanced nociceptive behaviour following conditioning injection of formalin in the perioral area of the rat R. Cadet, L. Aigouy, A. Woda * Laboratoire de Physiologie Orofaciale, Facult~ Chirurgie Dentaire, 11 Boulevard Charles de Gaulle, 63000 Clermont-Ferrand, France Accepted 27 December 1994
Abstract
The possible existence of long-term modifications in response to a transient nociceptive conditioning stimulation was investigated in the rat in three experiments. (1) A nociceptive conditioning stimulus was delivered in the form of a s.c. formalin injection (conditioning injection) in the left upper lip. Evaluation of the nociceptive behaviour triggered by another formalin injection (testing injection) made in the controlateral right upper lip was carried out in distinct groups of rats 7, 14 or 28 days after the conditioning. An enhanced nociceptive response at days 7 and 14 and a return to the baseline at day 28 were observed. (2) A similar protocol was developed with formalin used for both conditioning and testing but an anaesthetic blockade of the infraorbital nerve was performed just before the conditioning injection to suppress the initial barrage. The change observed at day 7 was suppressed by the nerve block. (3) A conditioning nociceptive stimulus was applied either ipsilaterally to the right lower lip or to the tail. An increased nociceptive response was observed when the conditioning stimulus was applied to the same side as the test stimulus but no increase in the formalin test response was detected when the conditioning stimulus was applied to the tail. These results indicated that, after a single formalin injection in the left upper lip, a hyperexcitability developed that depended on the initial barrage, lasted for at least 2 weeks, was no longer present at 4 weeks and might rely on a segmental mechanism. The hypothesis of a central sensitization triggered by an initial barrage and maintained by an ongoing input induced from the periphery is discussed.
Keywords: Formalin; Hyperalgesia; Orofacial; Plasticity; Rat; Trigeminal
1. Introduction
Plasticity has become a major topic in the study of pain mechanisms. It has been suggested that peripheral events can produce sustained changes in the nervous system and consequently can influence subsequent nociceptive processing. Sustained increased nociceptive behaviour depending at least in part on central changes have been observed after various events occurring at the periphery, such as deafferentation, tissue injury and inflammation. The sensitization of trigeminal and spinal second- order neurones is frequently invoked to explain peripherally induced hyperalgesia (see [9] for a review). However, the exact conditions required to induce these changes are not clearly understood. Central hyperexcitable states can be triggered by many
* Corresponding author. Fax: (33) 73.43.64.09. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 0 5 5 - 0
kinds of peripheral stimulation, from a transient reversible stimulus to severe persistent inflammation. For example, electrical stimulation applied to nociceptive muscle fibre afferents increased the flexor reflex excitability for at least 90 rain [29]. It seemed to depend on C-fibre inputs without requiring an ongoing activity for its maintenance [30,35]. Sustained central facilitation can also be inferred from observation of modifications in the properties of dorsal horn neurones after conditioning by electrical [4] or non-invasive mechanical stimulations [5]. These situations all involved transient stimulation and the changes were reported to last for < 2 h. However, plastic changes of longer duration have been recorded in totally different conditions involving deafferentation procedures [18], partial constriction of the sciatic nerve [11,36] or sustained peripheral inflammatory states, such as those induced by noxious thermal stimulation, mustard oil, carrageenin or Freund's adjuvant [15,16,34,35]. Importantly, in these
190
R. Cadet et al. / Brain Research 676 (1995) 189-195
conditions, the nerve impairment or inflammatory state was still present when the test stimulus was applied and the duration of the central changes after the peripheral consequences of the stimulus were over were not studied. In short, plasticity in pain has mostly been shown in one of the following two conditions: (1) a brief stimulus which induces a short-lasting hyperexcitability; and (2) a long-duration stimulus inducing an increase in nociceptive reactions which are concomitant with their causes. Accordingly, the duration of changes in nociceptive behaviour after a reversible but high- intensity barrage of nociceptive inputs required investigation. Injection of formalin induces an intense barrage of powerful nociceptive inputs with effects which last for 45-60 min as inferred from behavioural observations [6,12] and electrophysiological recordings [10,22]. The following questions were addressed. (1) Can enhanced nociceptive behaviour be observed with this stimulus and, if so, what is its time course? (2) Do the induced changes depend on the initial barrage? (3) Can a similar modification be observed when the conditioned locus is remote from the tested area; in other words, are the phenomena underlying the change segmental or not? Each of these questions was addressed in one of the three separate experiments described later in the paper. All of them were carried out in freely moving rats and used two s.c. formalin injections, one as conditioning nociceptive stimulation and another several days later to trigger measurable nociceptive behaviour. At the beginning of these experiments, it was assumed that, if they existed at all, any effects of conditioning nociceptive stimulation would be limited. Consequently, a high concentration of formalin was used for the conditioning nociceptive stimulation and the trigeminal area, known for its high density of nociceptors, was chosen (see [25] for a review). An adaptation for the orofacial area [6] of the formalin test [12] was selected to evaluate any changes in nociceptive behaviour. Some results have been published in a short note [3].
2. Materials and methods
In the first experiment in this study, a nociceptive conditioning stimulus was delivered in the form of a s.c. formalin injection in the left upper lip. The evaluation of the nociceptive behaviour was carried out 7, 14 and 28 days later with a testing stimulus consisting of another formalin injection in the controlateral right upper lip (Table 1). In the second experiment, anaesthetic blockade of the left infraorbital nerve was performed just before the conditioning injection to the left upper lip to suppress the initial barrage due to the formalin injection. Again, the evaluation of the noci-
ceptive behaviour was carried out with another formalin injection in the controlateral right upper lip (Table 2). Finally, in the third experiment, the conditioning nociceptive stimulus was not applied to the left upper lip controlaterally to the test injection site as in the two first experiments but ipsilaterally to the right lower lip and in another group to the tail (Table 3).
2.1. Animals Experiments were carried out on male SpragueDawley rats weighing 180-220 g, a weight at which rats exhibit a consistent late-phase response during the formalin test [31]. The animals were housed in climateand light-controlled rooms (21 + F C; 12-2-h darklight cycle, with lights on at 07.00 h) for at least 1 week before the experiments. Testing took place during the light phase. 2 h before a testing session, the rat was placed in a dedicated noise-proof observation room, maintained at 26 _+ 1° C [23,27]. Each animal was used only once and was sacrificed at the end of each session, using a lethal dose of pentobarbital, to avoid unnecessary suffering.
2.2. Conditioning injections The nociceptive conditioning stimulations were delivered in the form of 100/zl of either saline or 10% formalin prepared from 37% formaldehyde solution. The site of the injections was the left upper lip in the first and second experiments. In the third experiment, it was the right lower lip or the tail. The rats were anaesthetized before the conditioning injections with i.p. 2% sodium methohexital solution (Brietal, Lilly France) which covered the 50-60 min of the formalininduced nociceptive behaviour.
2.3. Testing injections and measurement of pain behaviour The evaluation of the nociceptive behaviour was carried out with the orofacial formalin test [6-8]. Test sessions were performed between 14.00 and 18.00 h because of a suspected nyctohemeral rhythm of nociception in the rat [1]. Formalin tests were initiated by a s.c. injection of 50/zl of a 5% (first and second experiments) or 1% (third experiment) formalin solution made in the right upper lip, lateral to the nose of the rat, using a 1-ml syringe with a 27-gauge needle. A change in the concentration of formalin solution from 5 to 1% was decided since the dose-response curve observed after injection of different concentrations of formalin might not be linear and instead might display a plateau or a bell-shape curve [7,32]. In this case, an increase in the painful sensation would not be reflected
R. Cadet et al. / Brain Research 676 (1995) 189-195
191
Table 1 Experimental design of first experiment Treatments applied to six groups (n = 10) Conditioning injection (day 0) Testing injection (day x)
Left upper lip
Formalin 10%
Saline
Formalin 10%
Saline
Formalin 10%
Saline
Right upper lip
Formalin 5% at day 7
Formalin 5% at day 7
Formalin 5% at day 14
Formalin 5% at day 14
Formalin 5% at day 28
Formalin 5% at day 28
After nociceptive or control conditioning injection, testing injection of 5% formalin was made controlaterally at day 7 for two groups, at day 14 for two other groups and at day 2:~ for the last two groups
by the observed nociceptiw,~ behaviour. Any enhanced nociception could be partly masked by 5% formalin and better revealed by a lower concentration (1%). To ensure calm conditions, the testing injections were made under brief anaesthesia with halothane applied through a nasal mask (Halothane, Laboratoire Belamont) [21]. Successive steps of halothane anaesthesia were estimated by observation of the struggling and falling reactions of the animal. Anaesthesia was discontinued when the first eyeblink was observed. Under these conditions, the usual duration of awakening was ~15s. Just after the testing injection, the experimenter left the observation room and the rat behaviour was recorded with a video camera placed in front of a 30-cm side cubic glass observation box with three mirrored sides. The videotapes were analysed, the observer being unaware of the rat's conditioning treatment. The observed response exhibited two phases of rubbing. The first phase started 15-30 s after the testing injection of formalin and lasted for 3-6 rain. After a time of relative inactivity, a second phase of intense rubbing activity lasted 12-45 min after the injection. Pain behaviour was measured by monitoring the duration of rubbing for each 3-min block. In the first experiment, both first and second phases were considered. Since no significant effect was noted on the first phase, only the second phase of the response was observed in the subsequent experiments. The most reproducible and objective behavioural response related to pain intensity was the rubbing of the injected area with the ipsilateral forepaw often accompanied in its movements by the controlateral forepaw. Sometimes, the rat used the hindpaw to rub its face. The total time the rat spent rubbing its face was counted irrespective of the paw used.
2.4. Expt. 1: time course of effect of formalin conditioning nociceptive stimulation (Table 1) Six groups of 10 rats were used. Three groups received formalin and three received saline in the left upper lip, either as a nociceptive or a control conditioning injection. One saline- and one formalin-conditioned group were tested 7, 14 and 28 days after the conditioning injection. For the six groups, the testing injection was performed controlaterally to the conditioning injection site, i.e. in the right upper lip, with 5% formalin solution. Values of each test group were compared only with those of their respective control group. Data were analysed using a non-parametric Mann--Whitney U test because the distribution of five out of the six groups was not Gaussian. P < 0.05 was regarded as significant. The three control groups were compared using a Kruskal-Wallis H test followed by a Newman-Keuls test (P < 0.05 regarded as significant).
2.5. Expt. 2: effect of nerve block at time of conditioning (Table 2) 45 animals were used. Nerve block was performed or simulated by injection in the left infraorbital fissure of 300 /xl bupivacaine (Marca'ine 0.5%, Laboratoire Roger Bellon) or saline. A preliminary experiment with bupivacaine injection in the infraorbital fissure and formalin injection in the ipsilateral upper lip indicated that the anaesthetic block was complete after 30 min and lasted for 90 min, i.e. it began to fade 120 min after the bupivacaine injection. 30 min after the nerve block, the nociceptive conditioning injection was made in the left upper lip. The testing injection was performed controlaterally in the right upper lip with 5%
Table 2 Experimental design of second experiment showing treatments applied to four groups Treatments applied to four groups Conditioning injection (day 0) Testing injection (day 7)
Left infra orbital fissure Left upper lip Right upper lip
Saline Saline Formalin 5%
Saline Formalin 10% Formalin 5%
Bupivacaine Saline Formalin 5%
Bupivacaine Formalin 10% Formalin 5%
Testing injection of 5% formalin was made controlaterally 7 days after nociceptive or control conditioning injection. Nerve block was performed or simulated by injection of bupivacaine or saline in left infraorbital fissure 30 min before conditioning injection
192
R. Cadet et al. / Brain Research 676 (1995) 189-195
formalin solution, 7 days after the nociceptive conditioning injection. Comparison between the four groups were performed using a Kruskal-Wallis H test followed by a Newman-Keuls test. P < 0.05 was regarded as significant. A non-parametric test was used because the distribution of none of the four groups was Gaussian.
2.6. Expt. 3: effect of ipsilateral and heterosegmental conditioning (Table 3) 30 rats were split into three groups characterized by their conditioning stimulation. Two groups received nociceptive conditioning stimulation applied to either the tail or the right lower lip, i.e. to the same side as the testing injection. The third group received saline. Testing injection of 1% formalin solution was performed 7 days after the conditioning stimulation. Comparison between the three groups were performed using a Kruskal-Wallis H test followed by a Newman-Keuls test. P < 0.05 was regarded as significant. A non-parametric test was used because the distribution of none of the three groups was Gaussian.
3. Results
3.1. Expt. 1: time course of hyperalgesia after formalin conditioning nociceptive stimulation Duration of rubbing behaviour in the three control groups (n = 10) did not differ significantly ( P > 0.05): 223 + 38 s at 7 days, 247 + 50 s at 14 days and 230 + 45 s at 28 days. For each test group (n = 10), duration of rubbing was calculated with reference to their respective control groups and expressed as % (mean + S.E.M.), followed by absolute values in second. The % were 170 + 41 (397 + 75 s) at 7 days, 162 + 22 (401 + 55 s) at 14 days and 99 + 15 (228 + 35 s) at 28 days. No significant difference ( P > 0.05) was found between the two groups tested 7 and 14 days after the conditioning although a significant difference was observed between these two groups and the 28-day group. Compared with their respective control group, a significant increase ( P < 0.05) in the intensity of the nociceptive response was observed at 7 and 14 days but not 28 days after the nociceptive conditioning stimulation (Fig. 1).
250 ,1
1
Day 7 ,
Day 14
Day 28
200
/ 150 ]
.r, 00
e50
0 Control
Formalin
Control
Formalin
/ 0/
Control
Formalin
Fig. 1. Duration of rubbing in saline- and formalin-conditioned groups, after a testing injection performed 7, 14 or 28 days after conditioning stimulation. Data are given, for each conditioning-test interval as % of value of corresponding control group (mean + S.E.M.) (see absolute values in text). Conditioned groups differed significantly from their respective control groups (*) at days 7 and 14 (P < 0.05) (Mann-Whitney test).
3.2. Expt. 2: effect of nerve block at time of conditioning The experimental groups were named according to: (1) the infraorbital injection of saline or bupivacaine; and (2) the conditioning injection of saline or formalin in the upper lip. Thus, the four groups were called Saline-Saline, Saline-Formalin, Bupivacaine-Saline and Bupivacaine-Formalin. The data are given as % of the value of the Saline-Saline control group (145 + 15 s, n = 12). Again, pretreatment with a chemical nociceptive stimulation resulted 7 days later in a significant enhancement ( P < 0.05) of the amplitude of the response in the Saline-Formalin group (178 + 27%, 250 + 39 s, n = 9) (Fig. 2). This effect did not occur if the first injection was made during a 2-h anaesthesia of the lip; the mean value of the Bupivacaine-Formalin group (112 + 14%, 158 + 20 s, n = 12) was significantly lower than in the Saline-Formalin group (178 + 27%, n = 9). Also, there was no difference in the mean values for duration of rubbing between the Bupivacaine-Formalin (112 + 14%, 158 + 20 s, n = 12) and the Bupivacaine-Saline (110 + 19%, 154 + 13s, n = 12) groups nor between the Bupivacaine-Formalin and Saline-Saline groups. These results were not due to an artefactual sustained effect of the bupivacaine since the Bupivacaine-Saline and Saline-Saline mean values did not differ either.
Table 3 Experimental design of third experiment Treatments applied to three groups Conditioning injection (day 0) Testing injection (day 7)
Right lower lip saline . Right upper lip formalin 1%
Right lower lip formalin 10% Right upper lip formalin 1%
Tail formalin 10% Right upper lip formalin 1%
For first and second groups, injection of 1% formalin was made ipsilateraly to conditioning injection performed 7 days before. Third group received 1% formalin in right upper lip 7 days after nociceptive conditioning injection into tail
R. Cadetet al. /Brain Research 676 (1995) 189-195 250 . m
tioning stimulation was applied to the tail, the delayed response to the testing injection was not modified compared with the control group (101 + 24, 96 + 22 s, n = 10) (Fig. 3).
*
200
193
j
150
.o e ioo o
•
_
1 4. D i s c u s s i o n
50
e~ o
Saline-Saline
Saline-Formalin
Bupivacaine-Saline Bnpivacaine-Formalin
Fig. 2. Duration of rubbing during formalin test performed 7 days after conditioning stimuli applied with or without nerve block. For each experimental group, values are expressed as % (mean+ S.E.M.) of value of control group injected with saline at infraorbital fissure and in left upper lip (Saline-Saline) (absolute values in text). Bupivacaine-Formalin (bupivacaine injected in left infraorbital fissure and formalin in left upper lip as a conditioning, n = 12), BupivacaineSaline (n = 12) and Saline-Saline (n = 12) groups, compared with Saline-Formalin (n = 9) group, displayed a significantly different rubbing duration (*). Values of rubbing duration in BupivacaineFormalin, Bupivacaine-Saline and Saline-Saline groups did not differ significantly(P > 0.05) (IO'uskal-Wallis test).
3.3. Expt. 3: effect of ipsilateral and heterosegmental conditioning The data are given as % of the value of the ipsilateral control group (96 ___20 s, n = 10). After nociceprive conditioning stimulation applied to the lower lip, duration of rubbing during a testing injection performed ipsilateraly 7 days later was significantly longer (186 ___25%, 179 + 24 s, n = 10) than for the ipsilateral control group ( P < 0.05). 'When the nociceptive condi-
250
200
150
._= 100
,=
50
0
[psi-Saline
Ip~i-Formalin
1
Tail-Formalin
Fig. 3. Duration of rubbing during formalin test performed 7 days after an ipsilateral or remote conditioningstimulus. For each experimental group, values are expressed as % (mean+ S.E.M.) of value of control group injected with saline in right upper lip as a conditioning injection (Ipsi-Saline) (n = 10) (absolute values in text). Rubbing duration of group injected with formalin in tail (Tail-Formalin, n = 10) did not differ significantlyfrom Ipsi-Saline (P > 0.05). Compared with group injected with fi)rmalin in right upper lip (Ipsi-Formalin, n = 10), both Tail-Formalin and Ipsi-Saline displayed a significantly different rubbing time (P < 0.05) (*) (Kruskal-Wailis test).
These experiments show that a conditioning formalin injection applied to the left upper lip of the rat can induce, 1 and 2 weeks later, an increase in the nociceptive response after a test injection of formalin in the right controlateral upper lip. The basic data, i.e. the presence of an increased reponse at day 7, is robust since it was observed in all three experiments. The 1or 2-week duration of the change is particularly striking. Long-duration increase in nociception has already been described with other inflammatory models, such as carrageenin or Freund's adjuvant [15,16]; however, in these models the enhanced response can be considered as the direct consequence of the concomitant peripheral inflammation whereas the effects seen in this study after formalin injection appeared to result from an intense but transient initial barrage. The modifications seen at day 7 can be suppressed by a nerve block performed just before the application of the nociceptive conditioning stimulus. Of critical importance for asserting the central origin of the observed changes is the fact that, in the first and second experiments, conditioning and testing sites were remote from each other. Also, the anaesthetic block was made at the infraorbital nerve rather than directly into the lip so that the anaesthesia was remote from the conditioning injection site. Thus, inhibition by the bupivacaine of the release in the perioral tissues of substances critical for the early stages of the inflammatory process was avoided. Therefore, it can be concluded that an initial barrage reaching the CNS and triggered by the conditioning injection of formalin is necessary to initiate the change. Does this conclusion suggest that, a contrario, the maintenance of enhanced nociception is independent of ongoing inputs arising from the peripheral tissue injured by the conditioning stimulation? Though this seems to be fully or partially the case for short-duration changes [19,28], it appears that, when the central sensitization develops over a long period, the ongoing input from the injured tissue is necessary. In the clinical situations recently described by Gracely et al. [13] concerning four patients with reflex sympathetic dystrophy, a long-duration sensitization was observed. Local anaesthesia of a small area involving a surgical scar suppressed, for the time of the block, a chronic pain referred to surrounding and much larger cutaneous areas. The authors proposed that, after being initiated by a peripheral trauma, impaired central processing resulting in chronic pain was maintained by
194
R. Cadet et al. / Brain Research 676 (1995) 189-195
ongoing peripheral nociceptive inputs originating from the scar. The same conclusion was reached after observing decreased thermal hyperalgesia when a local block or intrathecal injection of an N-methyl-D-aspartare antagonist had been performed several days after the initial barrage caused by a loose nerve ligation of the sciatic nerve [20]. Therefore, ongoing inputs could be, together with the initial nociceptive barrage, a necessary condition for the development of chronic pain at least when a nerve injury is implicated in the aetiology. This, transposed to the experimental inflammatory conditions described in the present study, would mean that, although necessary as demonstrated by the anaesthetic block, the initial nociceptive discharge triggered by the conditioning stimulus is not sufficient to maintain an enhanced central excitability and that ongoing inputs from the formalin-induced inflammation are also required. If this is so, then: (1) the time course of the cutaneous inflammatory reactions induced by the formalin must be consistent with the duration of the observed increased responsiveness; and (2) the cutaneous inflammation must be able to produce ongoing inputs. (1) Cutaneous reactions produced by the formalin injection match the time course of the increased nociception. This is indicated in the rat, in a study by Clavelou et al. (in press) [7] which shows that, after a 10% formalin injection in the orofacial area, a residual oedema is present at day 5 and cicatrization is still not complete at day 10. Histological observation showing cutaneous reactions with similar time courses were described in the mouse by Rosland et al. [24] after formalin injection. These histological data match our behavioural results, suggesting that the nociceptive response increase dies away along with oedema and cicatrization. (2) Definite knowledge concerning the presence of ongoing inputs arising from formalin-induced scars in the hours and days after the injection is not available since the only information on first-order neurone activity after formalin injection was acquired during the 1st hour after the needle prick [17]. However, ongoing inputs certainly exist because primary hyperalgesia develops from the local inflammation and is accompanied by mechanical allodynia which should cause peripheral input triggering during the common animal behaviour. Overall, these results are consistent with the hypothesis of a central sensitization triggered by an initial barrage and maintained by an ongoing input induced from the periphery. The fact that in this study the changes were over after 4 weeks would indicate that peripheral cutaneous reactions and the corresponding ongoing inputs had also faded out as strongly suggested by Clavelou et al. [7]. In the third experiment, an increase in the formalin test response was also observed when the conditioning stimulus was applied to the same side as the test stimulus but no effect could be detected when the
conditioning stimulus was applied to the tail. The tail was chosen rather than a paw as a remote conditioning site to avoid any interference with the behavioural test. An injection of formalin in a hindpaw, e.g., would have modified the posture of the rat and impeded the use of the forepaw to rub its lip during the test. The fact that an increased nociceptive response from the trigeminal field was obtained by conditioning a closely related ipsi- or controlateral area but not by conditioning the tail suggests segmental rather than suprasegmental mechanisms underlying the central changes. However, this negative result must be treated with caution since it is hazardous to compare formalin conditioning applied in two areas as different as the face and tail. However, other arguments suggest that these central changes take place at the segmental level. For example, Wolpaw and Lee [33] have reported conditioning of stretch reflex lasting for at least 3 days after a spinal cord transection made above the level of the recorded reflex. Evidence has also been presented suggesting that the hyperalgesia surrounding the site of a capsaicin injection is not due to the sensitization of peripheral nociceptors but on the contrary to the sensitization of dorsal horn neurones [2,26]. Demonstration by Heinricher et al. [14] of an operant conditioning of evoked potentials in the rat trigeminal complex is also strong evidence of plastic changes occurring at the segmental level. After a single formalin injection in the left upper lip, a change in the response to nociceptive stimulation develops that lasts for at least 2 weeks and is no longer present at 4 weeks. This change depends on an initial barrage and possibly also on ongoing inputs. It may rely on a segmental mechanism.
Acknowledgements We wish to thank A.M. Gaydier and S. Millien for their technical assistance and P. Raboisson for his help in writing the manuscript. The work was supported by a grant from Fondation Dentaire de France.
References [1] Bar-Or, A. and Brown, G.M., Pineal involvement in the diurnal rhythm of nociception in the rat, Life Sci., 44 (1989) 1067-1075. [2] Baumann, T.K., Simone, D.A., Shain, C.N. and Lamotte, R.H., Neurogenic hyperalgesia: the search for the primary cutaneous afferent fibers that contribute to capsaicine-induced pain and hyperalgesia, J. Neurophysiol., 66 (1991) 212-227. [3] Cadet, R., Aigouy, L. and Woda, A., Sustained hyperalgesia can be induced in the rat by a single formalin injection and depends on the initial nociceptive inputs, Neurosci. Lett., 156 (1993) 43-46.
IL Cadet et al. / Brain Research 676 (1995) 189-195 [4] Cervero, F., Schouenborg, J., Sj61und, B.H. and Waddell, P.J., Cutaneous inputs to dorsal horn neurones in adult rats treated at birth with capsaicin, Brain Res., 301 (1984) 47-57. [5] Cervero, F., Laird, J.M.A. and Pozo, M.A., Selective changes of receptive field properties of spinal nociceptive neurones induced by noxious visceral stimulation in the cat, Pain, 51 (1992) 335-342. [6] Clavelou, P., Pajot, J., Dallel, R. and Raboisson, P., Application of the formalin test to the study of orofacial pain in the rat, Neurosci. Lett., 103 (1989) 2,49-353. [7] Clavelou, P., Dallel, R., Orliaguet, T., Woda, A. and Raboisson, P., The orofacial formalin test in rats: effects of different formalin concentrations, Pain, in press. [8] Dallel, R., Raboisson, P., C]avelou, P., Saade, M. and Woda, A., Evidence for a peripheral origin of the tonic nociceptive response to subcutaneous formalin, Pain, in press. [9] Dickenson, A.H., Recent advances in the physiology and pharmacology of pain: plasticity and its implications for clinical analgesia, J. Psychopharmacol., 5 (1991) 342-351. [10] Dickenson, A.H. and Sullivan, A.F., Subcutaneous formalin-induced activity of dorsal horn neurones in the rat: differential response to an intratheca]! opiate administered pre or post formalin, Pain, 30 (1987) 349-360. [11] Dougherty, P.M., Garrison, C.J. and Carlton, S.M., Differential influence of local anaesthetic upon two models of experimentally induced peripheral mononeuropathy in the rat, Brain Res., 570 (1992) 109-115. [12] Dubuisson, D. and Dennis, S.G., The formalin test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats, Pain, 4 (1977) 164-174. [13] Gracely, R.H., Lynch, S.A. and Bennett, G.J., Painful neuropathy: altered central processing maintained dynamically by peripheral input, Pain, 51 (1992) 175-194. [14] Heinricher, M.M., Rosenfeld, J.P. and Dowman, R., Operant conditioning of trigeminal nuclear evoked potentials, Brain Res., 7 (1981) 353-358. [15] Kayser, V. and Guilbaud, G., Further evidence for changes in the responsiveness of somatosensory neurons in arthritic rats: a study of the posterior intralaminar region of the thalamus, Brain Res., 323 (1984) 144-147. [16] Kayser, V. and Guilbaud, G , Local and remote modifications of nociceptive sensitivity durinb, carrageenin-induced inflammation in the rat, Pain, 28 (1987) 9'9-107. [17] Klemm, F., Carli, G. and Reeh, P.W., Peripheral neural correlates of the formalin test in the rat, Pfliigers Arch., (1989) 42. [18] Kwan, C.W., Hu, J.W. and Sessle, B.J., Effects of tooth pulp deafferentation on brainstem neurons of the rat trigeminal subnucleus oralis, Somatosens. Motor Res., 10 (1993) 115-131. [19] LaMotte, R.H., Shain, C.N., Simone, D.A. and Tsai, E.F.P., Neurogenic hyperaigesia: psychophysical studies of underlying mechanisms, J. Neurophysioi., 66 (1991) 190-211. [20] Map, J., Price, D.D., Mayer, D.J., Lu, J. and Hayes, R.L., Intrathecal MK-801 and local nerve anaesthesia synergistically reduce nociceptive behaviours in rats with experimental peripheral mononeuropathy, Brain Res., 576 (1992) 254-262.
195
[21] Murray, C.W., Porreca, F. and Cowan, A., Methodological refinements to the mouse paw formalin test: an animal model of tonic pain, J. Pharmacol. Methods, 20 (1988) 175-186. [22] Raboisson, P., Bourdiol, P., Dallel, R., Clavelou, P. and Woda, A., Responses of trigeminal subnucleus oralis nociceptive neurones to subcutaneous formalin in the rat, Neurosci. Lett., 125 (1991) 179-182. [23] Rosland, J.H., The formalin test in mice: the influence of ambient temperature, Pain, 45 (1991)211-216. [24] Rosland, J.H., Tjolsen, A., Maehle, B. and Hole, K., The formalin test in mice: effect offormalin concentration, Pain, 42 (1990) 235-242. [25] Shults, R.C., Trigeminal primary afferents receptors. In A.R. Light (Ed.), The Initial Processing of Pain and Its Descending Control: Spinal and Trigeminal Systems, Karger, Basel, Switzerland, 1992, pp. 50-65. [26] Simone, D.A., Sorkin, L.S., Oh, U., Chung, J.M., Owens, C., Lamotte, R.H. and Willis, W.D., Neurogenic hyperalgesia: central neural correlates in responses of spinothalamic tract neurons, J. Neurophysiol., 66 (1991) 228-246. [27] Tjolsen, A., Berge, O.G., Hunskaar, S., Rosland, J.H., Hole, K., The formalin test: an evaluation of the method, Pain, 51 (1992) 5-17. [28] Torebj6rk, H.E., Lundberg, L. and LaMotte, R.H., Central changes in processing of mechanoreceptive input in capsaicininduced secondary hyperalgesia in humans, J. Physiol., 448 (1992) 765-780. [29] Wall, P.D. and Woolf, C.J., Muscle but not cutaneous C-afferent input produces prolonged increases in the excitability of the flexion reflex in the rat, J. PhysioL, 356 (1984) 443-458. [30] Wall, P.D. and Woolf, C.J., The brief and the prolonged facilitatory effects of unmyelinated afferent input on the rat spinal cord are independently influenced by peripheral nerve injury, Neuroscience, 17 (1986) 1199-1206. [31] Wheeler-Aceto, H. and Cowan, A., Standardisation of the rat paw formalin test for the evaluation of analgesics, Psychopharmacology, 104 (1991) 35-44. [32] Wheeler-Aceto, H. and Cowan, A., Naloxone causes apparent antinociception and pronociception simultaneously in the rat paw formalin test, Eur. Z Pharmacol., 236 (1993) 193-199. [33] Wolpaw, J.R. and Lee, C.L., Memory traces in primate spinal cord produced by operant conditioning of H-reflex, J. Neurophysiol., 61 (1989) 563-572. [34] Woolf, C.J., Long-term alterations in the excitability of the flexion reflex produced by peripheral tissue injury in the chronic decerebrate rat, Pain, 18 (1984) 325-343. [35] Woolf, C.J. and Wall, P.D., Relative effectiveness of C afferent primary fibres of different origins in evoking a prolonged facilitation of the flexor reflex in the rat, J. Neurosci., 6 (1986) 1433-1442. [36] Yamamoto, T., Shimoyama, N. and Mizuguchi, T., Role of the injury discharge in thedevelopment of thermal hyperesthesia after sciatic nerve constriction injury in the rat, Anesthesiology, 79 (1993) 993-1002.
Brain Research 676 (1995) 189-195
Research report
Enhanced nociceptive behaviour following conditioning injection of formalin in the perioral area of the rat R. Cadet, L. Aigouy, A. Woda * Laboratoire de Physiologie Orofaciale, Facult~ Chirurgie Dentaire, 11 Boulevard Charles de Gaulle, 63000 Clermont-Ferrand, France Accepted 27 December 1994
Abstract
The possible existence of long-term modifications in response to a transient nociceptive conditioning stimulation was investigated in the rat in three experiments. (1) A nociceptive conditioning stimulus was delivered in the form of a s.c. formalin injection (conditioning injection) in the left upper lip. Evaluation of the nociceptive behaviour triggered by another formalin injection (testing injection) made in the controlateral right upper lip was carried out in distinct groups of rats 7, 14 or 28 days after the conditioning. An enhanced nociceptive response at days 7 and 14 and a return to the baseline at day 28 were observed. (2) A similar protocol was developed with formalin used for both conditioning and testing but an anaesthetic blockade of the infraorbital nerve was performed just before the conditioning injection to suppress the initial barrage. The change observed at day 7 was suppressed by the nerve block. (3) A conditioning nociceptive stimulus was applied either ipsilaterally to the right lower lip or to the tail. An increased nociceptive response was observed when the conditioning stimulus was applied to the same side as the test stimulus but no increase in the formalin test response was detected when the conditioning stimulus was applied to the tail. These results indicated that, after a single formalin injection in the left upper lip, a hyperexcitability developed that depended on the initial barrage, lasted for at least 2 weeks, was no longer present at 4 weeks and might rely on a segmental mechanism. The hypothesis of a central sensitization triggered by an initial barrage and maintained by an ongoing input induced from the periphery is discussed.
Keywords: Formalin; Hyperalgesia; Orofacial; Plasticity; Rat; Trigeminal
1. Introduction
Plasticity has become a major topic in the study of pain mechanisms. It has been suggested that peripheral events can produce sustained changes in the nervous system and consequently can influence subsequent nociceptive processing. Sustained increased nociceptive behaviour depending at least in part on central changes have been observed after various events occurring at the periphery, such as deafferentation, tissue injury and inflammation. The sensitization of trigeminal and spinal second- order neurones is frequently invoked to explain peripherally induced hyperalgesia (see [9] for a review). However, the exact conditions required to induce these changes are not clearly understood. Central hyperexcitable states can be triggered by many
* Corresponding author. Fax: (33) 73.43.64.09. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 0 5 5 - 0
kinds of peripheral stimulation, from a transient reversible stimulus to severe persistent inflammation. For example, electrical stimulation applied to nociceptive muscle fibre afferents increased the flexor reflex excitability for at least 90 rain [29]. It seemed to depend on C-fibre inputs without requiring an ongoing activity for its maintenance [30,35]. Sustained central facilitation can also be inferred from observation of modifications in the properties of dorsal horn neurones after conditioning by electrical [4] or non-invasive mechanical stimulations [5]. These situations all involved transient stimulation and the changes were reported to last for < 2 h. However, plastic changes of longer duration have been recorded in totally different conditions involving deafferentation procedures [18], partial constriction of the sciatic nerve [11,36] or sustained peripheral inflammatory states, such as those induced by noxious thermal stimulation, mustard oil, carrageenin or Freund's adjuvant [15,16,34,35]. Importantly, in these
190
R. Cadet et al. / Brain Research 676 (1995) 189-195
conditions, the nerve impairment or inflammatory state was still present when the test stimulus was applied and the duration of the central changes after the peripheral consequences of the stimulus were over were not studied. In short, plasticity in pain has mostly been shown in one of the following two conditions: (1) a brief stimulus which induces a short-lasting hyperexcitability; and (2) a long-duration stimulus inducing an increase in nociceptive reactions which are concomitant with their causes. Accordingly, the duration of changes in nociceptive behaviour after a reversible but high- intensity barrage of nociceptive inputs required investigation. Injection of formalin induces an intense barrage of powerful nociceptive inputs with effects which last for 45-60 min as inferred from behavioural observations [6,12] and electrophysiological recordings [10,22]. The following questions were addressed. (1) Can enhanced nociceptive behaviour be observed with this stimulus and, if so, what is its time course? (2) Do the induced changes depend on the initial barrage? (3) Can a similar modification be observed when the conditioned locus is remote from the tested area; in other words, are the phenomena underlying the change segmental or not? Each of these questions was addressed in one of the three separate experiments described later in the paper. All of them were carried out in freely moving rats and used two s.c. formalin injections, one as conditioning nociceptive stimulation and another several days later to trigger measurable nociceptive behaviour. At the beginning of these experiments, it was assumed that, if they existed at all, any effects of conditioning nociceptive stimulation would be limited. Consequently, a high concentration of formalin was used for the conditioning nociceptive stimulation and the trigeminal area, known for its high density of nociceptors, was chosen (see [25] for a review). An adaptation for the orofacial area [6] of the formalin test [12] was selected to evaluate any changes in nociceptive behaviour. Some results have been published in a short note [3].
2. Materials and methods
In the first experiment in this study, a nociceptive conditioning stimulus was delivered in the form of a s.c. formalin injection in the left upper lip. The evaluation of the nociceptive behaviour was carried out 7, 14 and 28 days later with a testing stimulus consisting of another formalin injection in the controlateral right upper lip (Table 1). In the second experiment, anaesthetic blockade of the left infraorbital nerve was performed just before the conditioning injection to the left upper lip to suppress the initial barrage due to the formalin injection. Again, the evaluation of the noci-
ceptive behaviour was carried out with another formalin injection in the controlateral right upper lip (Table 2). Finally, in the third experiment, the conditioning nociceptive stimulus was not applied to the left upper lip controlaterally to the test injection site as in the two first experiments but ipsilaterally to the right lower lip and in another group to the tail (Table 3).
2.1. Animals Experiments were carried out on male SpragueDawley rats weighing 180-220 g, a weight at which rats exhibit a consistent late-phase response during the formalin test [31]. The animals were housed in climateand light-controlled rooms (21 + F C; 12-2-h darklight cycle, with lights on at 07.00 h) for at least 1 week before the experiments. Testing took place during the light phase. 2 h before a testing session, the rat was placed in a dedicated noise-proof observation room, maintained at 26 _+ 1° C [23,27]. Each animal was used only once and was sacrificed at the end of each session, using a lethal dose of pentobarbital, to avoid unnecessary suffering.
2.2. Conditioning injections The nociceptive conditioning stimulations were delivered in the form of 100/zl of either saline or 10% formalin prepared from 37% formaldehyde solution. The site of the injections was the left upper lip in the first and second experiments. In the third experiment, it was the right lower lip or the tail. The rats were anaesthetized before the conditioning injections with i.p. 2% sodium methohexital solution (Brietal, Lilly France) which covered the 50-60 min of the formalininduced nociceptive behaviour.
2.3. Testing injections and measurement of pain behaviour The evaluation of the nociceptive behaviour was carried out with the orofacial formalin test [6-8]. Test sessions were performed between 14.00 and 18.00 h because of a suspected nyctohemeral rhythm of nociception in the rat [1]. Formalin tests were initiated by a s.c. injection of 50/zl of a 5% (first and second experiments) or 1% (third experiment) formalin solution made in the right upper lip, lateral to the nose of the rat, using a 1-ml syringe with a 27-gauge needle. A change in the concentration of formalin solution from 5 to 1% was decided since the dose-response curve observed after injection of different concentrations of formalin might not be linear and instead might display a plateau or a bell-shape curve [7,32]. In this case, an increase in the painful sensation would not be reflected
R. Cadet et al. / Brain Research 676 (1995) 189-195
191
Table 1 Experimental design of first experiment Treatments applied to six groups (n = 10) Conditioning injection (day 0) Testing injection (day x)
Left upper lip
Formalin 10%
Saline
Formalin 10%
Saline
Formalin 10%
Saline
Right upper lip
Formalin 5% at day 7
Formalin 5% at day 7
Formalin 5% at day 14
Formalin 5% at day 14
Formalin 5% at day 28
Formalin 5% at day 28
After nociceptive or control conditioning injection, testing injection of 5% formalin was made controlaterally at day 7 for two groups, at day 14 for two other groups and at day 2:~ for the last two groups
by the observed nociceptiw,~ behaviour. Any enhanced nociception could be partly masked by 5% formalin and better revealed by a lower concentration (1%). To ensure calm conditions, the testing injections were made under brief anaesthesia with halothane applied through a nasal mask (Halothane, Laboratoire Belamont) [21]. Successive steps of halothane anaesthesia were estimated by observation of the struggling and falling reactions of the animal. Anaesthesia was discontinued when the first eyeblink was observed. Under these conditions, the usual duration of awakening was ~15s. Just after the testing injection, the experimenter left the observation room and the rat behaviour was recorded with a video camera placed in front of a 30-cm side cubic glass observation box with three mirrored sides. The videotapes were analysed, the observer being unaware of the rat's conditioning treatment. The observed response exhibited two phases of rubbing. The first phase started 15-30 s after the testing injection of formalin and lasted for 3-6 rain. After a time of relative inactivity, a second phase of intense rubbing activity lasted 12-45 min after the injection. Pain behaviour was measured by monitoring the duration of rubbing for each 3-min block. In the first experiment, both first and second phases were considered. Since no significant effect was noted on the first phase, only the second phase of the response was observed in the subsequent experiments. The most reproducible and objective behavioural response related to pain intensity was the rubbing of the injected area with the ipsilateral forepaw often accompanied in its movements by the controlateral forepaw. Sometimes, the rat used the hindpaw to rub its face. The total time the rat spent rubbing its face was counted irrespective of the paw used.
2.4. Expt. 1: time course of effect of formalin conditioning nociceptive stimulation (Table 1) Six groups of 10 rats were used. Three groups received formalin and three received saline in the left upper lip, either as a nociceptive or a control conditioning injection. One saline- and one formalin-conditioned group were tested 7, 14 and 28 days after the conditioning injection. For the six groups, the testing injection was performed controlaterally to the conditioning injection site, i.e. in the right upper lip, with 5% formalin solution. Values of each test group were compared only with those of their respective control group. Data were analysed using a non-parametric Mann--Whitney U test because the distribution of five out of the six groups was not Gaussian. P < 0.05 was regarded as significant. The three control groups were compared using a Kruskal-Wallis H test followed by a Newman-Keuls test (P < 0.05 regarded as significant).
2.5. Expt. 2: effect of nerve block at time of conditioning (Table 2) 45 animals were used. Nerve block was performed or simulated by injection in the left infraorbital fissure of 300 /xl bupivacaine (Marca'ine 0.5%, Laboratoire Roger Bellon) or saline. A preliminary experiment with bupivacaine injection in the infraorbital fissure and formalin injection in the ipsilateral upper lip indicated that the anaesthetic block was complete after 30 min and lasted for 90 min, i.e. it began to fade 120 min after the bupivacaine injection. 30 min after the nerve block, the nociceptive conditioning injection was made in the left upper lip. The testing injection was performed controlaterally in the right upper lip with 5%
Table 2 Experimental design of second experiment showing treatments applied to four groups Treatments applied to four groups Conditioning injection (day 0) Testing injection (day 7)
Left infra orbital fissure Left upper lip Right upper lip
Saline Saline Formalin 5%
Saline Formalin 10% Formalin 5%
Bupivacaine Saline Formalin 5%
Bupivacaine Formalin 10% Formalin 5%
Testing injection of 5% formalin was made controlaterally 7 days after nociceptive or control conditioning injection. Nerve block was performed or simulated by injection of bupivacaine or saline in left infraorbital fissure 30 min before conditioning injection
192
R. Cadet et al. / Brain Research 676 (1995) 189-195
formalin solution, 7 days after the nociceptive conditioning injection. Comparison between the four groups were performed using a Kruskal-Wallis H test followed by a Newman-Keuls test. P < 0.05 was regarded as significant. A non-parametric test was used because the distribution of none of the four groups was Gaussian.
2.6. Expt. 3: effect of ipsilateral and heterosegmental conditioning (Table 3) 30 rats were split into three groups characterized by their conditioning stimulation. Two groups received nociceptive conditioning stimulation applied to either the tail or the right lower lip, i.e. to the same side as the testing injection. The third group received saline. Testing injection of 1% formalin solution was performed 7 days after the conditioning stimulation. Comparison between the three groups were performed using a Kruskal-Wallis H test followed by a Newman-Keuls test. P < 0.05 was regarded as significant. A non-parametric test was used because the distribution of none of the three groups was Gaussian.
3. Results
3.1. Expt. 1: time course of hyperalgesia after formalin conditioning nociceptive stimulation Duration of rubbing behaviour in the three control groups (n = 10) did not differ significantly ( P > 0.05): 223 + 38 s at 7 days, 247 + 50 s at 14 days and 230 + 45 s at 28 days. For each test group (n = 10), duration of rubbing was calculated with reference to their respective control groups and expressed as % (mean + S.E.M.), followed by absolute values in second. The % were 170 + 41 (397 + 75 s) at 7 days, 162 + 22 (401 + 55 s) at 14 days and 99 + 15 (228 + 35 s) at 28 days. No significant difference ( P > 0.05) was found between the two groups tested 7 and 14 days after the conditioning although a significant difference was observed between these two groups and the 28-day group. Compared with their respective control group, a significant increase ( P < 0.05) in the intensity of the nociceptive response was observed at 7 and 14 days but not 28 days after the nociceptive conditioning stimulation (Fig. 1).
250 ,1
1
Day 7 ,
Day 14
Day 28
200
/ 150 ]
.r, 00
e50
0 Control
Formalin
Control
Formalin
/ 0/
Control
Formalin
Fig. 1. Duration of rubbing in saline- and formalin-conditioned groups, after a testing injection performed 7, 14 or 28 days after conditioning stimulation. Data are given, for each conditioning-test interval as % of value of corresponding control group (mean + S.E.M.) (see absolute values in text). Conditioned groups differed significantly from their respective control groups (*) at days 7 and 14 (P < 0.05) (Mann-Whitney test).
3.2. Expt. 2: effect of nerve block at time of conditioning The experimental groups were named according to: (1) the infraorbital injection of saline or bupivacaine; and (2) the conditioning injection of saline or formalin in the upper lip. Thus, the four groups were called Saline-Saline, Saline-Formalin, Bupivacaine-Saline and Bupivacaine-Formalin. The data are given as % of the value of the Saline-Saline control group (145 + 15 s, n = 12). Again, pretreatment with a chemical nociceptive stimulation resulted 7 days later in a significant enhancement ( P < 0.05) of the amplitude of the response in the Saline-Formalin group (178 + 27%, 250 + 39 s, n = 9) (Fig. 2). This effect did not occur if the first injection was made during a 2-h anaesthesia of the lip; the mean value of the Bupivacaine-Formalin group (112 + 14%, 158 + 20 s, n = 12) was significantly lower than in the Saline-Formalin group (178 + 27%, n = 9). Also, there was no difference in the mean values for duration of rubbing between the Bupivacaine-Formalin (112 + 14%, 158 + 20 s, n = 12) and the Bupivacaine-Saline (110 + 19%, 154 + 13s, n = 12) groups nor between the Bupivacaine-Formalin and Saline-Saline groups. These results were not due to an artefactual sustained effect of the bupivacaine since the Bupivacaine-Saline and Saline-Saline mean values did not differ either.
Table 3 Experimental design of third experiment Treatments applied to three groups Conditioning injection (day 0) Testing injection (day 7)
Right lower lip saline . Right upper lip formalin 1%
Right lower lip formalin 10% Right upper lip formalin 1%
Tail formalin 10% Right upper lip formalin 1%
For first and second groups, injection of 1% formalin was made ipsilateraly to conditioning injection performed 7 days before. Third group received 1% formalin in right upper lip 7 days after nociceptive conditioning injection into tail
R. Cadetet al. /Brain Research 676 (1995) 189-195 250 . m
tioning stimulation was applied to the tail, the delayed response to the testing injection was not modified compared with the control group (101 + 24, 96 + 22 s, n = 10) (Fig. 3).
*
200
193
j
150
.o e ioo o
•
_
1 4. D i s c u s s i o n
50
e~ o
Saline-Saline
Saline-Formalin
Bupivacaine-Saline Bnpivacaine-Formalin
Fig. 2. Duration of rubbing during formalin test performed 7 days after conditioning stimuli applied with or without nerve block. For each experimental group, values are expressed as % (mean+ S.E.M.) of value of control group injected with saline at infraorbital fissure and in left upper lip (Saline-Saline) (absolute values in text). Bupivacaine-Formalin (bupivacaine injected in left infraorbital fissure and formalin in left upper lip as a conditioning, n = 12), BupivacaineSaline (n = 12) and Saline-Saline (n = 12) groups, compared with Saline-Formalin (n = 9) group, displayed a significantly different rubbing duration (*). Values of rubbing duration in BupivacaineFormalin, Bupivacaine-Saline and Saline-Saline groups did not differ significantly(P > 0.05) (IO'uskal-Wallis test).
3.3. Expt. 3: effect of ipsilateral and heterosegmental conditioning The data are given as % of the value of the ipsilateral control group (96 ___20 s, n = 10). After nociceprive conditioning stimulation applied to the lower lip, duration of rubbing during a testing injection performed ipsilateraly 7 days later was significantly longer (186 ___25%, 179 + 24 s, n = 10) than for the ipsilateral control group ( P < 0.05). 'When the nociceptive condi-
250
200
150
._= 100
,=
50
0
[psi-Saline
Ip~i-Formalin
1
Tail-Formalin
Fig. 3. Duration of rubbing during formalin test performed 7 days after an ipsilateral or remote conditioningstimulus. For each experimental group, values are expressed as % (mean+ S.E.M.) of value of control group injected with saline in right upper lip as a conditioning injection (Ipsi-Saline) (n = 10) (absolute values in text). Rubbing duration of group injected with formalin in tail (Tail-Formalin, n = 10) did not differ significantlyfrom Ipsi-Saline (P > 0.05). Compared with group injected with fi)rmalin in right upper lip (Ipsi-Formalin, n = 10), both Tail-Formalin and Ipsi-Saline displayed a significantly different rubbing time (P < 0.05) (*) (Kruskal-Wailis test).
These experiments show that a conditioning formalin injection applied to the left upper lip of the rat can induce, 1 and 2 weeks later, an increase in the nociceptive response after a test injection of formalin in the right controlateral upper lip. The basic data, i.e. the presence of an increased reponse at day 7, is robust since it was observed in all three experiments. The 1or 2-week duration of the change is particularly striking. Long-duration increase in nociception has already been described with other inflammatory models, such as carrageenin or Freund's adjuvant [15,16]; however, in these models the enhanced response can be considered as the direct consequence of the concomitant peripheral inflammation whereas the effects seen in this study after formalin injection appeared to result from an intense but transient initial barrage. The modifications seen at day 7 can be suppressed by a nerve block performed just before the application of the nociceptive conditioning stimulus. Of critical importance for asserting the central origin of the observed changes is the fact that, in the first and second experiments, conditioning and testing sites were remote from each other. Also, the anaesthetic block was made at the infraorbital nerve rather than directly into the lip so that the anaesthesia was remote from the conditioning injection site. Thus, inhibition by the bupivacaine of the release in the perioral tissues of substances critical for the early stages of the inflammatory process was avoided. Therefore, it can be concluded that an initial barrage reaching the CNS and triggered by the conditioning injection of formalin is necessary to initiate the change. Does this conclusion suggest that, a contrario, the maintenance of enhanced nociception is independent of ongoing inputs arising from the peripheral tissue injured by the conditioning stimulation? Though this seems to be fully or partially the case for short-duration changes [19,28], it appears that, when the central sensitization develops over a long period, the ongoing input from the injured tissue is necessary. In the clinical situations recently described by Gracely et al. [13] concerning four patients with reflex sympathetic dystrophy, a long-duration sensitization was observed. Local anaesthesia of a small area involving a surgical scar suppressed, for the time of the block, a chronic pain referred to surrounding and much larger cutaneous areas. The authors proposed that, after being initiated by a peripheral trauma, impaired central processing resulting in chronic pain was maintained by
194
R. Cadet et al. / Brain Research 676 (1995) 189-195
ongoing peripheral nociceptive inputs originating from the scar. The same conclusion was reached after observing decreased thermal hyperalgesia when a local block or intrathecal injection of an N-methyl-D-aspartare antagonist had been performed several days after the initial barrage caused by a loose nerve ligation of the sciatic nerve [20]. Therefore, ongoing inputs could be, together with the initial nociceptive barrage, a necessary condition for the development of chronic pain at least when a nerve injury is implicated in the aetiology. This, transposed to the experimental inflammatory conditions described in the present study, would mean that, although necessary as demonstrated by the anaesthetic block, the initial nociceptive discharge triggered by the conditioning stimulus is not sufficient to maintain an enhanced central excitability and that ongoing inputs from the formalin-induced inflammation are also required. If this is so, then: (1) the time course of the cutaneous inflammatory reactions induced by the formalin must be consistent with the duration of the observed increased responsiveness; and (2) the cutaneous inflammation must be able to produce ongoing inputs. (1) Cutaneous reactions produced by the formalin injection match the time course of the increased nociception. This is indicated in the rat, in a study by Clavelou et al. (in press) [7] which shows that, after a 10% formalin injection in the orofacial area, a residual oedema is present at day 5 and cicatrization is still not complete at day 10. Histological observation showing cutaneous reactions with similar time courses were described in the mouse by Rosland et al. [24] after formalin injection. These histological data match our behavioural results, suggesting that the nociceptive response increase dies away along with oedema and cicatrization. (2) Definite knowledge concerning the presence of ongoing inputs arising from formalin-induced scars in the hours and days after the injection is not available since the only information on first-order neurone activity after formalin injection was acquired during the 1st hour after the needle prick [17]. However, ongoing inputs certainly exist because primary hyperalgesia develops from the local inflammation and is accompanied by mechanical allodynia which should cause peripheral input triggering during the common animal behaviour. Overall, these results are consistent with the hypothesis of a central sensitization triggered by an initial barrage and maintained by an ongoing input induced from the periphery. The fact that in this study the changes were over after 4 weeks would indicate that peripheral cutaneous reactions and the corresponding ongoing inputs had also faded out as strongly suggested by Clavelou et al. [7]. In the third experiment, an increase in the formalin test response was also observed when the conditioning stimulus was applied to the same side as the test stimulus but no effect could be detected when the
conditioning stimulus was applied to the tail. The tail was chosen rather than a paw as a remote conditioning site to avoid any interference with the behavioural test. An injection of formalin in a hindpaw, e.g., would have modified the posture of the rat and impeded the use of the forepaw to rub its lip during the test. The fact that an increased nociceptive response from the trigeminal field was obtained by conditioning a closely related ipsi- or controlateral area but not by conditioning the tail suggests segmental rather than suprasegmental mechanisms underlying the central changes. However, this negative result must be treated with caution since it is hazardous to compare formalin conditioning applied in two areas as different as the face and tail. However, other arguments suggest that these central changes take place at the segmental level. For example, Wolpaw and Lee [33] have reported conditioning of stretch reflex lasting for at least 3 days after a spinal cord transection made above the level of the recorded reflex. Evidence has also been presented suggesting that the hyperalgesia surrounding the site of a capsaicin injection is not due to the sensitization of peripheral nociceptors but on the contrary to the sensitization of dorsal horn neurones [2,26]. Demonstration by Heinricher et al. [14] of an operant conditioning of evoked potentials in the rat trigeminal complex is also strong evidence of plastic changes occurring at the segmental level. After a single formalin injection in the left upper lip, a change in the response to nociceptive stimulation develops that lasts for at least 2 weeks and is no longer present at 4 weeks. This change depends on an initial barrage and possibly also on ongoing inputs. It may rely on a segmental mechanism.
Acknowledgements We wish to thank A.M. Gaydier and S. Millien for their technical assistance and P. Raboisson for his help in writing the manuscript. The work was supported by a grant from Fondation Dentaire de France.
References [1] Bar-Or, A. and Brown, G.M., Pineal involvement in the diurnal rhythm of nociception in the rat, Life Sci., 44 (1989) 1067-1075. [2] Baumann, T.K., Simone, D.A., Shain, C.N. and Lamotte, R.H., Neurogenic hyperalgesia: the search for the primary cutaneous afferent fibers that contribute to capsaicine-induced pain and hyperalgesia, J. Neurophysiol., 66 (1991) 212-227. [3] Cadet, R., Aigouy, L. and Woda, A., Sustained hyperalgesia can be induced in the rat by a single formalin injection and depends on the initial nociceptive inputs, Neurosci. Lett., 156 (1993) 43-46.
IL Cadet et al. / Brain Research 676 (1995) 189-195 [4] Cervero, F., Schouenborg, J., Sj61und, B.H. and Waddell, P.J., Cutaneous inputs to dorsal horn neurones in adult rats treated at birth with capsaicin, Brain Res., 301 (1984) 47-57. [5] Cervero, F., Laird, J.M.A. and Pozo, M.A., Selective changes of receptive field properties of spinal nociceptive neurones induced by noxious visceral stimulation in the cat, Pain, 51 (1992) 335-342. [6] Clavelou, P., Pajot, J., Dallel, R. and Raboisson, P., Application of the formalin test to the study of orofacial pain in the rat, Neurosci. Lett., 103 (1989) 2,49-353. [7] Clavelou, P., Dallel, R., Orliaguet, T., Woda, A. and Raboisson, P., The orofacial formalin test in rats: effects of different formalin concentrations, Pain, in press. [8] Dallel, R., Raboisson, P., C]avelou, P., Saade, M. and Woda, A., Evidence for a peripheral origin of the tonic nociceptive response to subcutaneous formalin, Pain, in press. [9] Dickenson, A.H., Recent advances in the physiology and pharmacology of pain: plasticity and its implications for clinical analgesia, J. Psychopharmacol., 5 (1991) 342-351. [10] Dickenson, A.H. and Sullivan, A.F., Subcutaneous formalin-induced activity of dorsal horn neurones in the rat: differential response to an intratheca]! opiate administered pre or post formalin, Pain, 30 (1987) 349-360. [11] Dougherty, P.M., Garrison, C.J. and Carlton, S.M., Differential influence of local anaesthetic upon two models of experimentally induced peripheral mononeuropathy in the rat, Brain Res., 570 (1992) 109-115. [12] Dubuisson, D. and Dennis, S.G., The formalin test: a quantitative study of the analgesic effects of morphine, meperidine, and brain stem stimulation in rats and cats, Pain, 4 (1977) 164-174. [13] Gracely, R.H., Lynch, S.A. and Bennett, G.J., Painful neuropathy: altered central processing maintained dynamically by peripheral input, Pain, 51 (1992) 175-194. [14] Heinricher, M.M., Rosenfeld, J.P. and Dowman, R., Operant conditioning of trigeminal nuclear evoked potentials, Brain Res., 7 (1981) 353-358. [15] Kayser, V. and Guilbaud, G., Further evidence for changes in the responsiveness of somatosensory neurons in arthritic rats: a study of the posterior intralaminar region of the thalamus, Brain Res., 323 (1984) 144-147. [16] Kayser, V. and Guilbaud, G , Local and remote modifications of nociceptive sensitivity durinb, carrageenin-induced inflammation in the rat, Pain, 28 (1987) 9'9-107. [17] Klemm, F., Carli, G. and Reeh, P.W., Peripheral neural correlates of the formalin test in the rat, Pfliigers Arch., (1989) 42. [18] Kwan, C.W., Hu, J.W. and Sessle, B.J., Effects of tooth pulp deafferentation on brainstem neurons of the rat trigeminal subnucleus oralis, Somatosens. Motor Res., 10 (1993) 115-131. [19] LaMotte, R.H., Shain, C.N., Simone, D.A. and Tsai, E.F.P., Neurogenic hyperaigesia: psychophysical studies of underlying mechanisms, J. Neurophysioi., 66 (1991) 190-211. [20] Map, J., Price, D.D., Mayer, D.J., Lu, J. and Hayes, R.L., Intrathecal MK-801 and local nerve anaesthesia synergistically reduce nociceptive behaviours in rats with experimental peripheral mononeuropathy, Brain Res., 576 (1992) 254-262.
195
[21] Murray, C.W., Porreca, F. and Cowan, A., Methodological refinements to the mouse paw formalin test: an animal model of tonic pain, J. Pharmacol. Methods, 20 (1988) 175-186. [22] Raboisson, P., Bourdiol, P., Dallel, R., Clavelou, P. and Woda, A., Responses of trigeminal subnucleus oralis nociceptive neurones to subcutaneous formalin in the rat, Neurosci. Lett., 125 (1991) 179-182. [23] Rosland, J.H., The formalin test in mice: the influence of ambient temperature, Pain, 45 (1991)211-216. [24] Rosland, J.H., Tjolsen, A., Maehle, B. and Hole, K., The formalin test in mice: effect offormalin concentration, Pain, 42 (1990) 235-242. [25] Shults, R.C., Trigeminal primary afferents receptors. In A.R. Light (Ed.), The Initial Processing of Pain and Its Descending Control: Spinal and Trigeminal Systems, Karger, Basel, Switzerland, 1992, pp. 50-65. [26] Simone, D.A., Sorkin, L.S., Oh, U., Chung, J.M., Owens, C., Lamotte, R.H. and Willis, W.D., Neurogenic hyperalgesia: central neural correlates in responses of spinothalamic tract neurons, J. Neurophysiol., 66 (1991) 228-246. [27] Tjolsen, A., Berge, O.G., Hunskaar, S., Rosland, J.H., Hole, K., The formalin test: an evaluation of the method, Pain, 51 (1992) 5-17. [28] Torebj6rk, H.E., Lundberg, L. and LaMotte, R.H., Central changes in processing of mechanoreceptive input in capsaicininduced secondary hyperalgesia in humans, J. Physiol., 448 (1992) 765-780. [29] Wall, P.D. and Woolf, C.J., Muscle but not cutaneous C-afferent input produces prolonged increases in the excitability of the flexion reflex in the rat, J. PhysioL, 356 (1984) 443-458. [30] Wall, P.D. and Woolf, C.J., The brief and the prolonged facilitatory effects of unmyelinated afferent input on the rat spinal cord are independently influenced by peripheral nerve injury, Neuroscience, 17 (1986) 1199-1206. [31] Wheeler-Aceto, H. and Cowan, A., Standardisation of the rat paw formalin test for the evaluation of analgesics, Psychopharmacology, 104 (1991) 35-44. [32] Wheeler-Aceto, H. and Cowan, A., Naloxone causes apparent antinociception and pronociception simultaneously in the rat paw formalin test, Eur. Z Pharmacol., 236 (1993) 193-199. [33] Wolpaw, J.R. and Lee, C.L., Memory traces in primate spinal cord produced by operant conditioning of H-reflex, J. Neurophysiol., 61 (1989) 563-572. [34] Woolf, C.J., Long-term alterations in the excitability of the flexion reflex produced by peripheral tissue injury in the chronic decerebrate rat, Pain, 18 (1984) 325-343. [35] Woolf, C.J. and Wall, P.D., Relative effectiveness of C afferent primary fibres of different origins in evoking a prolonged facilitation of the flexor reflex in the rat, J. Neurosci., 6 (1986) 1433-1442. [36] Yamamoto, T., Shimoyama, N. and Mizuguchi, T., Role of the injury discharge in thedevelopment of thermal hyperesthesia after sciatic nerve constriction injury in the rat, Anesthesiology, 79 (1993) 993-1002.