- Email: [email protected]
Histamine Produces Opposing Effects to Serotonin in the Knee Joint of Rats
The Journal of Pain, Vol 14, No 8 (August), 2013: pp 808-817 Available online at www.jpain.org and www.sciencedirect.com
Histamine Produces Opposing Effects to Serotonin in the Knee Joint of Rats Eduardo Souza-Silva, Daniel Teixeira de Oliveira, Carolina Eto, Taciane Stein, and Carlos Rog erio Tonussi polis, SC, Brazil. Department of Pharmacology, Federal University of Santa Catarina, Floriano
Abstract: Formalin injected in the knee joint of rats produces concentration-dependent nociception, edema, and plasma leakage (PL). Herein, we investigated the effect of histamine H1 receptor (H1R) antagonists in this model. Articular nociception was inferred from the paw elevation time (PET; seconds) during 1-minute periods of stimulated walking, determined every 5 minutes, throughout a 60-minute experimental session. Edema was evaluated by the increase in articular diameter (AD; mm), and PL was measured by the amount of Evans blue dye in the synovial fluid (PL; mg/mL). Loratadine and cetirizine, given systemically, both increased the PET. None of the treatments changed the AD and PL. Loratadine given locally with formalin increased the PET but was without effect when given in the contralateral knee. Systemic loratadine was also without effect when formalin was coinjected with sodium cromoglycate. Histamine and the selective H1R agonist 2-pyridylethylamine decreased the PET and potentiated morphine spinal analgesia, but did not affect the AD and PL. Cetirizine prevented the antinociceptive effect of the H1R agonist. The N-methyl-D-aspartate/ histamine-site agonist tele-methylhistamine coinjected with formalin only increased PET. Serotonin alone had no effect on the PET and increased the AD, and the highest dose increased the PL. When coinjected with formalin, serotonin only caused hypernociception, and the highest dose also increased AD. NAN 190, cyproheptadine, and ondansetron (respectively, 5-HT1, 5-HT2, and 5-HT3 receptor antagonists) decreased the PET without changing the AD or PL. Collectively, these results suggest that in rats, the H1R plays an antinociceptive role within the knee joint, while serotonin receptors play a pronociceptive role. Perspective: The present study revealed an antinociceptive mechanism that has previously not been detected by traditional nociceptive tests. Our observations may help to improve the development of new pharmacological strategies for the treatment of clinically relevant pains that generally originate in deep structures. ª 2013 by the American Pain Society Key words: H1R antagonists, mast cells, articular pain, formalin test, cetirizine, loratadine.
T
he nociceptive role of histamine has generally been studied in cutaneous tissue and it is accepted to be a hyperalgesic mediator in inflammation, but its role in deep tissues still remains poorly understood. Evidence that supports the cutaneous nociceptive role
Received October 2, 2012; Revised February 7, 2013; Accepted February 13, 2013. This work received financial support from the following Brazilian government agencies: CNPq, CAPES, and FAPESC (pronex). E.S.S., D.T.O., and T.S. were recipients of graduate, and C.E. an undergraduate, fellowships from CNPq. C.R.T. was recipient of a research grant from CNPq. The authors state that there is no conflict of interest. rio Tonussi, Department of Address reprint requests to Carlos Roge Pharmacology, Universidade Federal de Santa Catarina, 8804-0900 Florianopolis, SC, Brazil. E-mail: [email protected] 1526-5900/$36.00 ª 2013 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2013.02.006
808
of histamine may have been, at least in some cases, overestimated. Thus, while locally applied firstgeneration histamine H1 receptor (H1R) antagonists can inhibit formalin-induced nociception,25,32 this effect could well be due to a local anesthetic property of the antihistamine employed.26 In other studies, the lack of selectivity of the antagonist used may be another problem.7,24 Thus, in an acute inflammatory model, the peripheral H1R antagonist thiazinamium prevented the paw hyperalgesia induced by carrageenan, but this agent also possesses an antiserotoninergic property.20 All of these studies were conducted in rodent models in which the possible contribution of serotonin derived from tissue mast cells cannot be omitted. The local injection of histamine has also been used to support a role for endogenous histamine in the
Souza-Silva et al
The Journal of Pain 25
formalin-induced nociception in skin and articular tissue.32 However, the concentrations of histamine used to produce nociceptive sensitization were far above that which is likely to exist per gram of rat skin.10,31 Furthermore, the data available concerning the effect of histamine in articular afferents are scarce,11 as are behavioral studies. In these studies, another source of ambiguity may be pruritus. Pruritus induced by histamine is not adequately discriminated from nociception by behavioral models, and therefore the summation of similar behaviors may be misinterpreted as sensitization.28 In addition, the nociceptive behavior produced by formalin is only one component of a local reaction which also includes edema and plasma leakage (PL).12 These vascular events can be influenced by the action of histamine in structures other than sensory fibers, such as blood vessels and immune cells, which may render their role even more complex. For example, the arthritis response induced by Freund’s adjuvant was shown to be reduced by an H1R agonist34 as well as by an antagonist.3 In a previous report, we observed that the peripheral H1R antagonist loratadine, or low doses of the brain-permeant meclizine, produced a hypernociceptive effect in the new model of formalin-induced articular incapacitation,17 suggesting that peripheral H1R activation could unexpectedly produce antinociception in the knee joint of rats. In view of the above observations, the effect of antihistamines on deep nociception may not be easily predicted. Formalin presents an interesting tool to study the mechanisms underlying persistent pain in an early inflammatory condition with the involvement of amines released by mast cells,25,19 and the model of knee incapacitation induced by formalin17 may be an interesting tool to study the contribution of these amines to articular nociception. The present study was designed to assess the role of the H1R in the deep tissue nociception induced by formalin in the knee joint, in parallel with the edematous and articular PL effects, aiming to provide consistent information that may explain how antihistamines can produce hypernociception in this model.
Methods Animals The experiments were performed on male Wistar rats (250–300 g) housed in a temperature-controlled room (21 6 2 C), under a 12/12-hour light/dark cycle, with free access to water and food. This study followed the ethical guidelines of the International Association for the Study of Pain14 and was previously approved by the local ethics committee for animal use (CEUA-UFSC: 23080.042991/2008-16).
Substances and Vehicles Cetirizine (MW 388.9), loratadine (MW 382.9), and sodium cromoglycate were obtained from Galena ^utica LTDA (Campinas, SP, Brazil). Quımica Farmace Ondansetron and morphine were acquired from
809
rios Crista lia LTDA (Itapira, SP, Brazil). Laborato Histamine, serotonin (5-HT), and tele-methylhistamine were acquired from Sigma-Aldrich (St. Louis, MO), and 2-pyridylethylamine, NAN 190 and cyproheptadine from Tocris (Minneapolis, MN). Loratadine was dissolved in polyoxyethylenesorbitan monoleate (Tween 80), not exceeding 5% of the total volume in physiological saline. Cetirizine, sodium cromoglycate, ondansetron, NAN 190, cyproheptadine, histamine, serotonine, tele-methylhistamine, 2-pyridylethylamine, and morphine were dissolved in physiological saline. Systemic treatments (loratadine and cetirizine) were applied by intraperitoneal (i.p.) route, 60 minutes before formalin. Local treatments were applied intrathecally (morphine) 20 minutes before or coinjected (loratadine, histamine, 2-pyridylethylamine, sodium cromoglycate serotonin, NAN 190, cyproheptadine, and ondansetron) with intra-articular formalin. Formalin (formaldehyde 37%; Merck KGaA, Darmstadt, Germany) was dissolved in saline at 1.5%, considering the initial concentration as 100%. Control group treatments were carried out with the vehicle of the respective test drugs.
Knee Joint Incapacitation Induced by Formalin The rat knee joint incapacitation test has been described in detail elsewhere.33 Briefly, in this test, rats are placed on a revolving cylinder (30-cm diameter; 3 rpm) for 1-minute periods and a computer-assisted device measured the total time that a specific hind paw was not in contact with the cylinder surface (paw elevation time [PET]). Normally, control animals display a PET of approximately 10 seconds, whereas algogenic substances injected into the knee joint increased this value only in the affected limb. In order to induce incapacitation, 50 mL of formalin solution, diluted in sterile saline, was injected into the right knee joints of the rats. The injection site was first shaved and treated with an iodine alcohol antiseptic solution. The animals were gently restrained in a supine position by hand, and the intra-articular injection was quickly performed with a 30-gauge needle. The PET was measured every 5 minutes after formalin injection.17 In addition to the nocifensive behavior being scored automatically, the experimenter was blind to the treatment protocols.
Evaluation of Articular Edema The animals were immobilized in the same way as for the intra-articular injection. Articular diameter (AD) was measured through the knee joint mediolateral axis applying the micrometer at 3 arbitrary levels along the knee joint proximodistal axis, and taking the highest value. The AD increase in millimeters was simply calculated subtracting the AD taken immediately before from that taken 1 hour after formalin injection.
Evaluation of the Synovial PL Under isoflurane anesthesia, the animals received an intravenous injection of Evans blue (25 mg/kg;
810
Histamine Inhibits Formalin-Induced Response
The Journal of Pain 8
.1 mL/100 g) through the gingival vein, 30 minutes before the formalin injection. One hour after formalin, they were euthanized by anesthetic deepening with 15% chloral hydrate until the respiratory arrest. A mediolateral incision was made on the knee joint overlying skin just above the patella, and the synovial cavity was washed out with 100 mL of .2% EDTA solution in physiological saline. This synovial lavage was centrifuged (3,000 rpm; 15 minutes) for debris sedimentation, and the supernatant (60–70 mL) was used for absorbance photometry (630 nm). Data were presented as mg of protein/mL of synovial fluid.
Motor Impairment Assessment When the cylinder started revolving, the animals promptly and spontaneously walked to keep themselves on top, without assistance from the experimenter. Each experiment comprised 13 periods of gait registering for each animal. Any repeated delay in the reaction to the cylinder movement or an animal fall was considered as motor impairment. A treatment was identified as causing no motor impairment if the abovementioned events were not observed.
Statistical Analysis All statistical analyses were carried out using the software Statistica 7.0 (StatSoft, Tulsa, OK). Data were expressed as the mean 6 SEM as shown in the figures. The PET time-course curves of the second phase of formalin response were compared by 2-way analysis of variance (ANOVA) followed by the Duncan post hoc test when a P level of less than .05 was detected. In this test, the P level indicated refers to the whole time-course curve (treatment). The AD or PL was analyzed by 1-way ANOVA followed by Dunnett’s post hoc test when a P level of less than .05 was detected. The Student’s t test was used when only 2 means were compared.
Results Concentration-Response Effect of Intra-Articularly Injected Formalin on Nociception, Edema, PL Formalin injected into the knee joint induced a concentration-dependent (.5, 1.5, 2, and 3%) increase in the PET (Fig 1, top), AD (Fig 1, bottom left), and synovial PL (Fig 1, bottom right), compared with saline-treated knee joint values. The time course of the PET behavior presented an initial, short-lasting phase (0–5-minute interval, P1) immediately after the injection, followed by a longlasting phase (5–60-minute interval, P2). The concentrations of 1.5, 2, and 3% of formalin produced a significantly different PET when compared to the control (P < .05). Similarly, only the .5% formalin did not produce a significant AD increase (.25 6 .09 mm), while all other higher concentrations produced AD values ranging between .38 6 .08 mm and .9 6 .12 mm, which were significantly higher than those presented by the saline-treated group (.1 6 .04 mm; P < .05). Regarding
Figure 1. Concentration-response relationship of the formalininduced articular effects. (Top) Paw elevation time (PET). Formalin (.5, 1.5, 2, and 3%) was injected into the knee joint in a volume of 50 mL, and PET was measured immediately after. C represents PET before formalin injection (time zero). P1 and P2 represent first and second phases, respectively, of formalininduced response. (Bottom left) Increase of AD in mm taken 1 hour after formalin injection. (Bottom right) PL to synovial fluid sampled 1 hour after formalin injection and presented in mg of protein per mL. Each plot symbol and texture box indicates the respective formalin percent concentrations tested. Saline represent the control group, which received only physiological saline in the knee joint. #P < .05 (1-way ANOVA followed by Dunnett’s test); *P < .05 (2-way ANOVA, followed by Duncan test between treatments). the PL, when all treatments were compared together, only the 2 and 3% formalin concentrations were found to be statistically significant according to Dunnett’s test; however, 1.5% formalin was also found to be significantly different from the saline group in the Student’s t test (P < .05). For subsequent experiments, the intermediate 1.5% concentration was used as a standard stimulus for inducing all 3 parameters, ie, incapacitation, AD increase, and PL. Furthermore, treatment effects on PET were analyzed in the early second phase only (5–35 minutes) since there were no pharmacological effects in the first phase (0–5 minutes), and effects were generally over by 35 minutes.
Effect of Systemic and Local Administration of the Peripherally Acting H1R Antagonists Intraperitoneal pretreatment with loratadine (10 mg/kg) and cetirizine (1 and 10 mg/kg) 1 hour before increased
Souza-Silva et al
The Journal of Pain
811
4,21
to histamine release. Formalin coinjected with sodium cromoglycate (1.6 mg/joint) prevented the increased incapacitation produced by loratadine (10 mg/kg) (Fig 2, bottom). Since both peripheral H1R antagonists only produced a PET increase, we extended this study by testing the effect of locally applied loratadine. Loratadine (20 pmol, knee) coinjected with formalin reproduced the PET increase in the formalin response (P < .05) (Fig 3, top). Supporting the local effect, the same dose injected into the left knee joint did not modify the incapacitation induced by formalin in the right knee joint (Fig 3, bottom). Also, the coinjection of loratadine (20 pmol) with formalin, but not that injected contralaterally, caused a significant decrease in the PL (P < .05) (Table 1). However, this treatment did not change the AD.
Effect of Histamine, Tele-Methylhistamine, and 2-Pyridylethylamine Into the Joint
Figure 2. Enhancing effect of H1R antagonists on the articular incapacitation and its blockade by sodium cromoglycate. (Top) LOR 2.5 and 10; and (Middle) CET .01, .1, 1, and 10 indicate, respectively, loratadine and cetirizine treatments at doses in mg/kg, given i.p. (Bottom) CRO 1.6, represents the dose of sodium cromoglycate coinjected with formalin. CRO 1 LOR indicates pretreatment with loratadine (10 mg/kg, i.p.) before cromoglycate-formalin coinjection; all loratadine and cetirizine treatments were given 1 hour before formalin injection. Open square represents control groups, which received 1.5% formalin alone in the knee joint and vehicle i.p., ie, 5% Tween 80 in physiological saline for loratadine, or physiological saline alone for cetirizine. The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
only the PET (P < .05). The lower doses of both antihistamines were without effect (Fig 2, top and middle). None of these treatments changed the AD and PL values (Table 1), even though doses chosen were previously reported as being able to inhibit vascular reactions due
A wide range of histamine doses (.2–20,000 nmol) was injected alone in na€ıve knee joints. Only the higher doses (200 and 20,000 nmol) produced slight incapacitation (P < .05) (Fig 4, top), PL, and AD increase (P < .05) (Table 1). However, the lowest doses of histamine (.2 and 20 nmol, knee) coinjected with formalin caused a potent inhibition of the formalin-induced incapacitation (P < .05) (Fig 5, top). From the 200-nmol dose and above, the inhibitory effect on formalininduced incapacitation was lost, although no increase was observed. The histamine metabolite tele-methylhistamine (1 and 10 pmol, knee), which acts selectively at the histaminergic site of the N-methyl-D-aspartate (NMDA) receptor,5 only increased the incapacitating effect of the formalin (P < .05) (Fig 5, middle). The selective H1R agonist 2-pyridylethylamine (.05, .5, and 5 nmol, knee) decreased the formalininduced PET (P < .05). On the other hand, a higher dose of the H1R agonist (500 nmol, knee) increased the formalin-induced PET (P < .05) (Fig 6, top). Cetirizine (.1 mg/kg, i.p.), given 1 hour before the articular coinjection of 2-pyridylethylamine with formalin, prevented the antinociceptive effect (Fig 6, middle). With the exception of the 20,000-nmol histamine dose combined with formalin, which enhanced the AD (P < .05), no other treatment affected the PL and AD induced by formalin (Table 1).
Potentiation of Spinal Morphine Analgesia by an Intra-Articular H1R Agonist Morphine (9 nmol) administered in the lumbar spinal channel 20 minutes before intra-articular formalin decreased the PET (P < .05). The lowest dose (.009 nmol) of morphine that had no effect on PET alone was potentiated by an also subeffective dose of 2-pyridylethylamine (.005 nmol) coinjected intraarticularly with formalin (P < .05) (Fig 6, bottom). None of these treatments changed the AD (data not shown).
812
Histamine Inhibits Formalin-Induced Response
The Journal of Pain
Table 1.
Effect of Histamine, H1R Agonist, and Antagonists on AD and Synovial PL in Formalin Test TREATMENTS
Histamine alone (nmol, knee)
Formalin with
Histamine (nmol, knee)
LOR (mg/kg, i.p.)
CET (mg/kg, i.p.)
LOR (pmol, knee)
LOR (pmol, contralateral knee) 2-PEA (nmol, knee)
CRO (mg, knee) LOR (i.p.) 1 CRO CET (i.p.) 1 2-PEA
DOSE
AD (MM)
PL (mG/ML)
0 0.2 20 200 20,000 0 200 20,000 0 2.5 10 0 1 10 0 2.1 20 0 20 0 5 500 1.6 2.5 1 1.6 .1 1 5
.1 6 .03 .2 6 .09 .4 6 .05 .9 6 .1* 1.2 6 .1* .4 6 .1 .7 6 .1 1.9 6 .05* .57 6 .10 .35 6 .09 .68 6 .13 .2 6 .02 .3 6 .04 .3 6 .06 .62 6 .15 .55 6 .15 .57 6 .11 .53 6 .1 .8 6 .10 .4 6 .04 .3 6 .06 .3 6 .05 .5 6 .1 .5 6 .13 .3 6 .07
7.5 6 .92 6.5 6 1.54 9.33 6 1.42 11.0 6 1.26* 12.0 6 1.06* 14.66 6 2.7 13.83 6 1.24 10.0 6 2.04 18.8 6 1.18 14.8 6 2.02 11.6 6 3.33 13.5 6 2.53 11.4 6 2.42 10.0 6 .68 23.6 6 2.86 20.6 6 2.4 14.1 6 .6* 15.5 6 3.04 13.5 6 .76 19.27 6 1.88 17.5 6 2.56 21.4 6 2.15 16.57 6 2.86 13.0 6 1.18 16.66 6 2.41
Abbreviations: LOR, loratadine; CET, cetirizine; 2-PEA, 2-pyridyl-ethylamine; CRO, sodium cromoglycate. NOTE. Values are presented as the mean 6 SEM (n = 6). Dose = 0 means the respective vehicle-treated group of each experiment. I.p. treatments were given 60 minutes before formalin. Excepting for histamine alone, all other knee treatments were coinjected with formalin. Some doses were omitted to make the table clearer, since they do not differ from controls. *Indicates P < .05 compared to the vehicle-treated group (1-way ANOVA followed by Dunnett’s post hoc test).
Effect of Local Administration of Serotonin and Serotonin Receptor Antagonists We further evaluated the involvement of serotonin (5-HT) in the present model of the knee joint nociception. Firstly, serotonin doses (1, 10, and 100 nmol) given alone in the knee joint did not increase the PET (Fig 4, bottom, only highest dose was shown), but they did increase the AD, and the highest dose also produced PL (P < .05) (Table 2). However, the same doses coinjected with formalin increased the PET (P < .05) (Fig 5, bottom) when compared with formalin alone. Only the highest dose (100 nmol) (P < .05) caused the AD to increase, and none increased the PL (Table 2). Due to the multiplicity of serotonin receptors known to be expressed in small afferent fibers, we evaluated the possible involvement of 3 that have been implicated in nociception. Antagonists of 5-HT1 (NAN-190; .1 and .6 pmol, knee), 5-HT2 (cyproheptadine; .16 and .5 pmol, knee), and 5-HT3 (ondansetron; .26 pmol, knee) receptors coinjected with formalin decreased the PET (P < .05) (Fig 7), but not the AD or PL (Table 2).
Discussion In the present study we confirmed and extended the previous observation17 that peripheral histamine H1R antagonists only increase the nocifensive response
induced by formalin in the knee joint. Also, it was shown that low doses of histamine, and the H1R agonist 2-pyridylethylamine (2-PEA), coinjected with formalin decreased the PET induced by formalin. None of these effects appeared to be due to a modification in the vascular permeability. This observation suggests that peripheral H1R activation is responsible for eliciting some kind of antinociceptive mechanism and that this mechanism is activated during formalin reaction. This conclusion was further supported when a nonanalgesic dose of morphine in the spinal cord was rendered effective by the concomitant intra-articular administration of an equally noneffective dose of an H1R agonist. The above observations contradict the general notion that the peripheral H1R activation by tissuereleased histamine plays a role in promoting pain.18,36,25 However, to date there is no clear evidence for the involvement of histamine in peripheral inflammatory pain, which has been verified for human pruritus.2,13,28 Furthermore, the role of histamine in many experimental models may be misinterpreted since the animal models employed to evaluate the histamine contribution to nociception do not discriminate nociceptionrelated behavior from pruritus-related behavior,16,29 essentially because the application of either histamine or algogens to the foot will elicit similar
Souza-Silva et al
Figure 3. Enhancing effect of locally applied loratadine on the
The Journal of Pain
813
articular incapacitation. (Top) LOR 2.1 and 20 indicate loratadine doses in pmol, coinjected with formalin. (Bottom) LOR 20 indicates loratadine dose in pmol, given in the left knee joint immediately before formalin injection in the right knee joint. Control groups (open square) received only 1.5% formalin diluted in a 5% Tween 80 physiological saline solution. The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
Figure 4. Effect of histamine and serotonin alone on the PET. (Top) HIST 20, 200, and 20,000 indicate histamine doses in nmol, knee. (Bottom) 5-HT 100 indicates serotonin dose in nmol, knee. The PET was taken before (C), immediately after the injection of both amines (0), and every 5 minutes after. Control groups received only physiological saline in the right knee joint (open square). *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
behaviors.25 Typically, the histamine doses employed to elicit nociceptive-related behaviors are far above25,32 those that may be expected to be released after a stimulus such as formalin in the tissues.10,31 Our results with the higher histamine doses are in agreement with this since we also observed an incapacitation response, as well as PL and edema with such doses. Another problem related to the correct pharmacological interpretation when using high histamine concentrations is the possibility of a direct facilitatory action at the NMDA receptor. Tele-methylhistamine is a major metabolite of histamine and a selective agonist at the histaminergic site of the NMDA receptor, but not in the histamine receptors.5 The injection of huge amounts of histamine in the tissue would likely also result in large amounts of this metabolite. It is conceivable that this facilitatory effect at NMDA receptors also facilitates formalin nociception,6 and in fact, this was observed by coinjecting tele-methylhistamine with formalin into the knee joint. However, the histamine concentration found in the synovial fluid of humans with or without arthritis is 1,000-fold lower, and in this dose range, histamine did not change the inflammatory response induced in the mouse knee joint,1 suggesting
that histamine does not participate as a physiological proinflammatory mediator. Second-generation H1R antagonists such as loratadine and cetirizine do not easily cross the blood-brain barrier, in contrast to first-generation antihistamines. Indeed, the higher loratadine and cetirizine doses used here have not been reported to produce changes in behavioral and cognitive tests.22 Furthermore, the hypernociceptive effect produced by loratadine given directly into the knee joint that received formalin, but not when given in the contralateral joint, excludes the possibility that this effect was due to an action outside the knee joint. In a previous report,17 we speculated that a blockade of a histamine vasodilating effect by the H1R antagonists could enhance the formalininduced nociception by simply inhibiting its scavenging from the synovial space. In an attempt to provide evidence for this hypothesis, we collected data on the AD and PL in the synovia. However, no changes were observed in the AD or PL after systemic treatments with H1R antagonists, and even the highest concentration of locally given loratadine only slightly diminished PL, which indicates that the vascular effect induced by formalin in the knee joint may be, at most, marginally mediated by local histamine release. Thus,
814
The Journal of Pain
Figure 5. Effect of histamine, tele-methylhistamine, and serotonin coinjected with formalin on the articular incapacitation. (Top) HIST .2, 20, 200, and 20,000 indicate histamine doses in nmol. (Middle) TELE 1 and 10 indicate tele-methylhistamine doses in pmol. (Bottom) 5-HT 1, 10, and 100 indicate serotonin doses in nmol. Control groups received 1.5% formalin alone (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
the hypernociceptive effect of the antihistamines may not be attributed to a modification of synovial permeability. The hypernociceptive effect of the H1R antagonists presupposes the release of histamine in the knee joints after formalin injection. This unexpected role of histamine was further confirmed by the antinociceptive effect produced by the local injection of either histamine or the selective H1 agonist 2-PEA. It is important to note that this effect was observed with doses lower than those usually reported as causing pronociceptive effects. In addition, the antinociceptive
Histamine Inhibits Formalin-Induced Response
Figure 6. Antinociceptive effect of H1R agonist, its reversal by a H1R antagonist, and potentiation of spinal morfine analgesia. (Top) 2-PEA .05, .5, 5, and 500 indicate the H1R agonist 2-pyridylethylamine doses in nmol coinjected with formalin in the knee joints. (Middle) CET .1 indicates the cetirizine dose in mg/kg given i.p. 60 minutes before formalin alone. In CET 1 2-PEA, the same dose of cetirizine was given before 2-PEA coinjected with formalin. (Bottom) MOR .009 and 9 indicate the morphine dose, in nmol, injected in the lumbar spinal channel 20 minutes before formalin in the knee joint. In MOR 1 2-PEA, the lowest dose of morphine was given intrathecally before 2-PEA coinjected with formalin; control groups received 1.5% formalin only (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
effect of 2-PEA was inhibited by a nonhypernociceptive dose of cetirizine, which reinforces the hypothesis that this mechanism is mediated specifically through H1 receptors. Previous reports have implicated local mast cells in the response elicited by formalin paw injection,25,19 thus raising the possibility that antihistamines may
Souza-Silva et al
The Journal of Pain
815
Effect of Serotonin and Antagonists on AD and Synovial PL
Table 2.
TREATMENTS Serotonin alone (nmol)
DOSE
0 1 10 100 Formalin with Serotonin (nmol) 0 1 10 100 NAN (pmol) .05 .1 .6 CYPRO (pmol) .16 .5 ONDA (pmol) .13 .26
AD (MM)
PL (mG/ML)
.3 6 .04 .6 6 .01* 1.0 6 .01* 1.2 6 .07* .48 6 .2 .6 6 .03 .7 6 .05 1.3 6 .2* .1 6 .08 .3 6 .1 .2 6 .05 .2 6 .06 .2 6 .05 .3 6 .1 .2 6 .1
7.5 6 1.75 8.7 6 1.21 9.06 6 2.38 14.5 6 1.85* 14.70 6 2.74 16.33 6 2.65 14.37 6 4.16 16.03 6 3.85 17.55 6 3.84 13.16 6 2.30 18.66 6 5.89 16.22 6 3.56 17.01 6 1.98 17.64 6 3.24 18.20 6 2.97
Abbreviations: NAN, NAN-190; CYPRO, cyproheptadine; ONDA, ondansetron. NOTE. Values are presented as the mean 6 S.E.M (n = 6). Dose = 0 means the respective vehicle treated group of each experiment. Excepting for serotonin alone, all other treatments were coinjected with formalin. *Indicates P < .05 compared to the vehicle-treated group (1-way ANOVA followed by Dunnett’s post hoc test).
antagonize an effect produced by histamine released from articular mast cells. The prevention of the hypernociceptive effect of loratadine with sodium cromoglycate is consistent with a role of mast cells in this model of formalin-induced nociception.23 Rodent mast cells differ from their human counterparts by way of their high content of serotonin.16 Since serotonin is more consistently reported to play a role in nociception than histamine, an investigation of the role of this amine was also carried out. Indeed, serotonin only enhanced the formalin response, even in the presence of significant PL. Furthermore, selective serotonin receptor antagonists applied with formalin only inhibited the incapacitation response, in a dosedependent fashion, and is in agreement with predictions based on previously reported data.9 The importance of the results obtained with serotonin and its antagonists is 2-fold: first, they reinforce the reliability of the model concerning to the findings to histamine and its antagonists; and second, they suggest that mast cell amines have a mutually opposing role in the articular nociception. Finally, the observation that a nonanalgesic dose of morphine in the spinal cord was rendered highly effective by the intra-articular injection of an H1R agonist strongly suggests a specific antinociceptive pathway that seems to be activated by H1 receptors in the knee joint of rats. We do not know how the neural substrate could subserve this connection from knee joint to spinal cord; however, there is evidence for a subset of small diameter primary afferent neurons expressing H1 receptors and neuropeptide Y (NPY) in dorsal root ganglion neurons of guinea pigs.15 Such fibers undergoing H1R activation in the peripheral end could promote the release of NPY in the spinal cord, where this peptide has been
Figure 7. Antinociceptive effect produced by 5-HT receptor antagonists coinjected with formalin on the articular incapacitation. (Top) CYP.16 and .5 indicate the 5-HT2 receptor antagonist cyproheptadine doses. (Middle) NAN .05, .1, and .5 indicate the 5-HT1 receptor antagonist NAN-190 doses. (Bottom) ONDA .13 and .26 indicate the 5-HT3 receptor antagonist ondansetron doses. All doses indicated are in pmol. Control groups received 1.5% formalin only (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
demonstrated to inhibit nociceptive transmission.30,35 A similar action still remains to be demonstrated in rats, although NPY has been detected in the sciatic nerves of rats,27 also suggesting the existence of sensory fibers that could be an appropriate candidate for this species. In conclusion, these results represent an important contribution to understanding the role of mast cell amines in the modulation of nociception in an early inflammatory condition, suggesting that, at least for knee joint sensory fibers, rat mast cell amines may
816
The Journal of Pain
Histamine Inhibits Formalin-Induced Response
exert opposite actions. We do not know whether histamine can play a similar role in humans. However, the further exploration of the mechanisms underlying
this antinociceptive effect of histamine in the present model may reveal a new approach to obtaining clinical analgesia.
References
17. Martins MA, Bastos LC, Tonussi CR: Formalin injection into knee joint of rats: Pharmacologic characterization of a deep somatic nociceptive model. J Pain 7:100-107, 2006
1. Adlesic M, Verdrengh M, Bokarewa M, Dahlberg L, Foster SJ, Tarkowski A: Histamine in rheumatoid arthritis. Scand J Immunol 65:530-537, 2007 €nder S, Steinhoff M, Miyashi Y, 2. Akihiko I, Roman R, Sta Schmelz M: Neurophysiology of pruritus: Interaction of itch and pain. Arch Dermatol 139:1475-1478, 2003 3. Al-Haboubi HA, Zeitlin IJ: The actions of cimetidine hydrochloride and mepyramine maleate in rat adjuvant arthritis. Eur J Pharmacol 78:175-185, 1982 4. Andersson M, Persson CG, Svensson C, Cervin-Hoberg C, Greiff L: Effects of loratadine on red wine-induced symptoms and signs of rhinitis. Acta Otolaryngol 123: 1087-1093, 2003 5. Burban A, Faucard R, Armand V, Bayard C, Vorobjev V, Arrang JM: Histamine potentiates N-methyl-D-aspartate receptors by interacting with an allosteric site distinct from the polyamine binding site. J Pharmacol Exp Ther 332: 912-921, 2010 6. Carlton SM: Peripheral NMDA receptors revisited - Hope floats. Pain 146:1-2, 2009 7. Chia YY, Lo Y, Liu K, Tan PH, Chung NC, Ko NH: The effect of promethazine on postoperative pain: A comparison of preoperative, postoperative, and placebo administration in patients following total abdominal hysterectomy. Acta Anaesthesiol Scand 48:625-630, 2004 8. De Oliveira DT, Souza-Silva E, Tonussi CR: Gingival vein punction: A new simple technique for drug administration or blood sampling in rats and mice. Scand J Lab Anim Sci 36:109-111, 2009 9. Doak GJ, Sawynok J: Formalin-induced nociceptive behavior and edema: Involvement of multiple peripheral 5-hydroxytryptamine receptor subtypes. Neuroscience 80: 939-949, 1997 10. Hardwick DC: Age changes in the histamine content of rat skin. J Physiol 28:157-165, 1954 11. Herbert MK, Schmidt RF: Histamine excites groups III and IV afferents from the cat knee joint depending on their resting activity. Neurosc Letters 305:95-98, 2001 12. Hong Y, Abbott FV: Peripheral opioid modulation of pain and inflammation in the formalin test. Eur J Pharmacol 277:21-29, 1995 €nder S, Yosipovitch G: The 13. Ikoma A, Steinhoff M, Sta neurobiology of itch. Nat Rev Neurosci 7:535-547, 2006 14. International Association For The Study Of Pain (IASP): Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16:109-110, 1983 15. Kashiba H, Fukui H, Senba E: Histamine H1 receptor mRNA is expressed in capsaicin-insensitive sensory neurons with neuropeptide Y-immunoreactivity in guinea pigs. Brain Research 901:85-93, 2001 16. Kuraishi Y, Nagasawa T, Hayashi K, Satoh M: Scratching behaviour induced by pruritogenic but not algesiogenic agents in mice. Eur J Pharmacol 275:229-233, 1995
18. Mobarakeh JI, Sakurada S, Hayashi T, Orito T, Okuyama K, Sakurada T, Kuramasu A, Watanabe T, Watanabe T, Yanai K: Enhanced antinociception by intrathecally-administered morphine in histamine H1 receptor gene knockout mice. Neuropharmacology 42:1079-1088, 2002 19. Nakajima K, Obata H, Ito N, Goto F, Saito S: The nociceptive mechanism of 5-hydroxytryptamine released into the peripheral tissue in acute inflammatory pain in rats. Eur J Pain 13:441-447, 2009 20. Neil A, Benoist JM, Kayser V, Guilbaud G: Initial nociceptive sensitization in carrageenin-induced rat paw inflammation is dependent on amine autacoid mechanisms: Electrophysiological and behavioural evidence obtained with a quaternary antihistamine, thiazinamium. Exp Brain Res 65:343-351, 1987 21. Nielsen PN, Skov PS, Poulsen LK, Schmelz M, Petersen LJ: Cetirizine inhibits skin reactions but not mediator release in immediate and developing late-phase allergic cutaneous reactions. A double-blind, placebo-controlled study. Clin Exp Allergy 31:1378-1384, 2001 22. Nishiga M, Fujii Y, Konishi M, Hossen MA, Chiaki K: Effects of second-generation histamine H1 receptor antagonists on the active avoidance response in rats. Clin Exp Pharmacol Physiol 30:60-63, 2003 23. Okayama Y, Church MK: Drugs modifying the responses of mast cell and basophils, in Foreman JC (ed): Immunopharmacology of Mast Cell and Basophils. New York, NY, Academic, 1993, pp 139-152 24. Paalzow GH, Paalzow LK: Promethazine both facilitates and inhibits nociception in rats: Effect of the testing procedure. Psychopharmacology 85:31-36, 1985 25. Parada CA, Tambeli CH, Cunha FQ, Ferreira SH: The major role of peripheral release of histamine and 5-hydroxytryptamine in formalin-induced nociception. Neuroscience 102:937-944, 2001 26. Raffa RB: Antihistamines as analgesics. J Clin Pharm Ther 26:81-85, 2001 27. Roddy DR, Yaksh TL, Aimone LD, Go VL: Distribution of neuropeptide Y in the spinal cords of cat, dog, rat, man and pig. Regul Pept 29:81-92, 1990 28. Schmelz M: Specific C–receptors for itch in human skin. J Neurosci 17:8003-8008, 1997 29. Shimada GS, LaMotte RH: Behavioural differentiation between itch and pain in mouse. Pain 139:681-687, 2008 30. Smith PA, Moran TD, Abdulla F, Tumber KK, Taylor BK: Spinal mechanisms of NPY analgesia. Peptides 28:464-474, 2007 31. Telford JM, West GB: Histamine and 5-hydroxytryptamine content of tissues after prolonged treatment with polymyxin. J Pharm Pharmacol 12:254-255, 1960 32. Ting E, Roveroni RC, Ferrari LF, Lotufo CM, Veiga MC, Parada CA, Tambeli CH: Indirect mechanism of
Souza-Silva et al histamine-induced nociception in temporo-mandibular joint of rats. Life Sci 81:765-771, 2007 33. Tonussi CR, Ferreira SH: Rat knee-joint carrageenan incapacitation test: An objective screen for central and peripheral analgesics. Pain 48:421-427, 1992 34. Vickers MR, Sykes KJ: The effect of histamine on the development of adjuvant arthritis in the rat. Agents Actions 12:691-698, 1982
The Journal of Pain
817
35. Wiley RG, Lemons LL, Kline RH IV: Neuropeptide Y receptor-expressing dorsal horn neurons: Role in nocifensive reflex response to heat and formalin. Neuroscience 161:139-147, 2009 36. Yoshida A, Mobarakeh JI, Sakurai E, Sakurada S, Orito T, Kuramasu A, Kato M, Yanai K: Intrathecally-administered histamine facilitates nociception through tachykinin NK-1 and histamine H1 receptors: A study in histidine decarboxylase gene knockout mice. Eur J Pharmacol 17:55-62, 2005
Histamine Produces Opposing Effects to Serotonin in the Knee Joint of Rats Eduardo Souza-Silva, Daniel Teixeira de Oliveira, Carolina Eto, Taciane Stein, and Carlos Rog erio Tonussi polis, SC, Brazil. Department of Pharmacology, Federal University of Santa Catarina, Floriano
Abstract: Formalin injected in the knee joint of rats produces concentration-dependent nociception, edema, and plasma leakage (PL). Herein, we investigated the effect of histamine H1 receptor (H1R) antagonists in this model. Articular nociception was inferred from the paw elevation time (PET; seconds) during 1-minute periods of stimulated walking, determined every 5 minutes, throughout a 60-minute experimental session. Edema was evaluated by the increase in articular diameter (AD; mm), and PL was measured by the amount of Evans blue dye in the synovial fluid (PL; mg/mL). Loratadine and cetirizine, given systemically, both increased the PET. None of the treatments changed the AD and PL. Loratadine given locally with formalin increased the PET but was without effect when given in the contralateral knee. Systemic loratadine was also without effect when formalin was coinjected with sodium cromoglycate. Histamine and the selective H1R agonist 2-pyridylethylamine decreased the PET and potentiated morphine spinal analgesia, but did not affect the AD and PL. Cetirizine prevented the antinociceptive effect of the H1R agonist. The N-methyl-D-aspartate/ histamine-site agonist tele-methylhistamine coinjected with formalin only increased PET. Serotonin alone had no effect on the PET and increased the AD, and the highest dose increased the PL. When coinjected with formalin, serotonin only caused hypernociception, and the highest dose also increased AD. NAN 190, cyproheptadine, and ondansetron (respectively, 5-HT1, 5-HT2, and 5-HT3 receptor antagonists) decreased the PET without changing the AD or PL. Collectively, these results suggest that in rats, the H1R plays an antinociceptive role within the knee joint, while serotonin receptors play a pronociceptive role. Perspective: The present study revealed an antinociceptive mechanism that has previously not been detected by traditional nociceptive tests. Our observations may help to improve the development of new pharmacological strategies for the treatment of clinically relevant pains that generally originate in deep structures. ª 2013 by the American Pain Society Key words: H1R antagonists, mast cells, articular pain, formalin test, cetirizine, loratadine.
T
he nociceptive role of histamine has generally been studied in cutaneous tissue and it is accepted to be a hyperalgesic mediator in inflammation, but its role in deep tissues still remains poorly understood. Evidence that supports the cutaneous nociceptive role
Received October 2, 2012; Revised February 7, 2013; Accepted February 13, 2013. This work received financial support from the following Brazilian government agencies: CNPq, CAPES, and FAPESC (pronex). E.S.S., D.T.O., and T.S. were recipients of graduate, and C.E. an undergraduate, fellowships from CNPq. C.R.T. was recipient of a research grant from CNPq. The authors state that there is no conflict of interest. rio Tonussi, Department of Address reprint requests to Carlos Roge Pharmacology, Universidade Federal de Santa Catarina, 8804-0900 Florianopolis, SC, Brazil. E-mail: [email protected] 1526-5900/$36.00 ª 2013 by the American Pain Society http://dx.doi.org/10.1016/j.jpain.2013.02.006
808
of histamine may have been, at least in some cases, overestimated. Thus, while locally applied firstgeneration histamine H1 receptor (H1R) antagonists can inhibit formalin-induced nociception,25,32 this effect could well be due to a local anesthetic property of the antihistamine employed.26 In other studies, the lack of selectivity of the antagonist used may be another problem.7,24 Thus, in an acute inflammatory model, the peripheral H1R antagonist thiazinamium prevented the paw hyperalgesia induced by carrageenan, but this agent also possesses an antiserotoninergic property.20 All of these studies were conducted in rodent models in which the possible contribution of serotonin derived from tissue mast cells cannot be omitted. The local injection of histamine has also been used to support a role for endogenous histamine in the
Souza-Silva et al
The Journal of Pain 25
formalin-induced nociception in skin and articular tissue.32 However, the concentrations of histamine used to produce nociceptive sensitization were far above that which is likely to exist per gram of rat skin.10,31 Furthermore, the data available concerning the effect of histamine in articular afferents are scarce,11 as are behavioral studies. In these studies, another source of ambiguity may be pruritus. Pruritus induced by histamine is not adequately discriminated from nociception by behavioral models, and therefore the summation of similar behaviors may be misinterpreted as sensitization.28 In addition, the nociceptive behavior produced by formalin is only one component of a local reaction which also includes edema and plasma leakage (PL).12 These vascular events can be influenced by the action of histamine in structures other than sensory fibers, such as blood vessels and immune cells, which may render their role even more complex. For example, the arthritis response induced by Freund’s adjuvant was shown to be reduced by an H1R agonist34 as well as by an antagonist.3 In a previous report, we observed that the peripheral H1R antagonist loratadine, or low doses of the brain-permeant meclizine, produced a hypernociceptive effect in the new model of formalin-induced articular incapacitation,17 suggesting that peripheral H1R activation could unexpectedly produce antinociception in the knee joint of rats. In view of the above observations, the effect of antihistamines on deep nociception may not be easily predicted. Formalin presents an interesting tool to study the mechanisms underlying persistent pain in an early inflammatory condition with the involvement of amines released by mast cells,25,19 and the model of knee incapacitation induced by formalin17 may be an interesting tool to study the contribution of these amines to articular nociception. The present study was designed to assess the role of the H1R in the deep tissue nociception induced by formalin in the knee joint, in parallel with the edematous and articular PL effects, aiming to provide consistent information that may explain how antihistamines can produce hypernociception in this model.
Methods Animals The experiments were performed on male Wistar rats (250–300 g) housed in a temperature-controlled room (21 6 2 C), under a 12/12-hour light/dark cycle, with free access to water and food. This study followed the ethical guidelines of the International Association for the Study of Pain14 and was previously approved by the local ethics committee for animal use (CEUA-UFSC: 23080.042991/2008-16).
Substances and Vehicles Cetirizine (MW 388.9), loratadine (MW 382.9), and sodium cromoglycate were obtained from Galena ^utica LTDA (Campinas, SP, Brazil). Quımica Farmace Ondansetron and morphine were acquired from
809
rios Crista lia LTDA (Itapira, SP, Brazil). Laborato Histamine, serotonin (5-HT), and tele-methylhistamine were acquired from Sigma-Aldrich (St. Louis, MO), and 2-pyridylethylamine, NAN 190 and cyproheptadine from Tocris (Minneapolis, MN). Loratadine was dissolved in polyoxyethylenesorbitan monoleate (Tween 80), not exceeding 5% of the total volume in physiological saline. Cetirizine, sodium cromoglycate, ondansetron, NAN 190, cyproheptadine, histamine, serotonine, tele-methylhistamine, 2-pyridylethylamine, and morphine were dissolved in physiological saline. Systemic treatments (loratadine and cetirizine) were applied by intraperitoneal (i.p.) route, 60 minutes before formalin. Local treatments were applied intrathecally (morphine) 20 minutes before or coinjected (loratadine, histamine, 2-pyridylethylamine, sodium cromoglycate serotonin, NAN 190, cyproheptadine, and ondansetron) with intra-articular formalin. Formalin (formaldehyde 37%; Merck KGaA, Darmstadt, Germany) was dissolved in saline at 1.5%, considering the initial concentration as 100%. Control group treatments were carried out with the vehicle of the respective test drugs.
Knee Joint Incapacitation Induced by Formalin The rat knee joint incapacitation test has been described in detail elsewhere.33 Briefly, in this test, rats are placed on a revolving cylinder (30-cm diameter; 3 rpm) for 1-minute periods and a computer-assisted device measured the total time that a specific hind paw was not in contact with the cylinder surface (paw elevation time [PET]). Normally, control animals display a PET of approximately 10 seconds, whereas algogenic substances injected into the knee joint increased this value only in the affected limb. In order to induce incapacitation, 50 mL of formalin solution, diluted in sterile saline, was injected into the right knee joints of the rats. The injection site was first shaved and treated with an iodine alcohol antiseptic solution. The animals were gently restrained in a supine position by hand, and the intra-articular injection was quickly performed with a 30-gauge needle. The PET was measured every 5 minutes after formalin injection.17 In addition to the nocifensive behavior being scored automatically, the experimenter was blind to the treatment protocols.
Evaluation of Articular Edema The animals were immobilized in the same way as for the intra-articular injection. Articular diameter (AD) was measured through the knee joint mediolateral axis applying the micrometer at 3 arbitrary levels along the knee joint proximodistal axis, and taking the highest value. The AD increase in millimeters was simply calculated subtracting the AD taken immediately before from that taken 1 hour after formalin injection.
Evaluation of the Synovial PL Under isoflurane anesthesia, the animals received an intravenous injection of Evans blue (25 mg/kg;
810
Histamine Inhibits Formalin-Induced Response
The Journal of Pain 8
.1 mL/100 g) through the gingival vein, 30 minutes before the formalin injection. One hour after formalin, they were euthanized by anesthetic deepening with 15% chloral hydrate until the respiratory arrest. A mediolateral incision was made on the knee joint overlying skin just above the patella, and the synovial cavity was washed out with 100 mL of .2% EDTA solution in physiological saline. This synovial lavage was centrifuged (3,000 rpm; 15 minutes) for debris sedimentation, and the supernatant (60–70 mL) was used for absorbance photometry (630 nm). Data were presented as mg of protein/mL of synovial fluid.
Motor Impairment Assessment When the cylinder started revolving, the animals promptly and spontaneously walked to keep themselves on top, without assistance from the experimenter. Each experiment comprised 13 periods of gait registering for each animal. Any repeated delay in the reaction to the cylinder movement or an animal fall was considered as motor impairment. A treatment was identified as causing no motor impairment if the abovementioned events were not observed.
Statistical Analysis All statistical analyses were carried out using the software Statistica 7.0 (StatSoft, Tulsa, OK). Data were expressed as the mean 6 SEM as shown in the figures. The PET time-course curves of the second phase of formalin response were compared by 2-way analysis of variance (ANOVA) followed by the Duncan post hoc test when a P level of less than .05 was detected. In this test, the P level indicated refers to the whole time-course curve (treatment). The AD or PL was analyzed by 1-way ANOVA followed by Dunnett’s post hoc test when a P level of less than .05 was detected. The Student’s t test was used when only 2 means were compared.
Results Concentration-Response Effect of Intra-Articularly Injected Formalin on Nociception, Edema, PL Formalin injected into the knee joint induced a concentration-dependent (.5, 1.5, 2, and 3%) increase in the PET (Fig 1, top), AD (Fig 1, bottom left), and synovial PL (Fig 1, bottom right), compared with saline-treated knee joint values. The time course of the PET behavior presented an initial, short-lasting phase (0–5-minute interval, P1) immediately after the injection, followed by a longlasting phase (5–60-minute interval, P2). The concentrations of 1.5, 2, and 3% of formalin produced a significantly different PET when compared to the control (P < .05). Similarly, only the .5% formalin did not produce a significant AD increase (.25 6 .09 mm), while all other higher concentrations produced AD values ranging between .38 6 .08 mm and .9 6 .12 mm, which were significantly higher than those presented by the saline-treated group (.1 6 .04 mm; P < .05). Regarding
Figure 1. Concentration-response relationship of the formalininduced articular effects. (Top) Paw elevation time (PET). Formalin (.5, 1.5, 2, and 3%) was injected into the knee joint in a volume of 50 mL, and PET was measured immediately after. C represents PET before formalin injection (time zero). P1 and P2 represent first and second phases, respectively, of formalininduced response. (Bottom left) Increase of AD in mm taken 1 hour after formalin injection. (Bottom right) PL to synovial fluid sampled 1 hour after formalin injection and presented in mg of protein per mL. Each plot symbol and texture box indicates the respective formalin percent concentrations tested. Saline represent the control group, which received only physiological saline in the knee joint. #P < .05 (1-way ANOVA followed by Dunnett’s test); *P < .05 (2-way ANOVA, followed by Duncan test between treatments). the PL, when all treatments were compared together, only the 2 and 3% formalin concentrations were found to be statistically significant according to Dunnett’s test; however, 1.5% formalin was also found to be significantly different from the saline group in the Student’s t test (P < .05). For subsequent experiments, the intermediate 1.5% concentration was used as a standard stimulus for inducing all 3 parameters, ie, incapacitation, AD increase, and PL. Furthermore, treatment effects on PET were analyzed in the early second phase only (5–35 minutes) since there were no pharmacological effects in the first phase (0–5 minutes), and effects were generally over by 35 minutes.
Effect of Systemic and Local Administration of the Peripherally Acting H1R Antagonists Intraperitoneal pretreatment with loratadine (10 mg/kg) and cetirizine (1 and 10 mg/kg) 1 hour before increased
Souza-Silva et al
The Journal of Pain
811
4,21
to histamine release. Formalin coinjected with sodium cromoglycate (1.6 mg/joint) prevented the increased incapacitation produced by loratadine (10 mg/kg) (Fig 2, bottom). Since both peripheral H1R antagonists only produced a PET increase, we extended this study by testing the effect of locally applied loratadine. Loratadine (20 pmol, knee) coinjected with formalin reproduced the PET increase in the formalin response (P < .05) (Fig 3, top). Supporting the local effect, the same dose injected into the left knee joint did not modify the incapacitation induced by formalin in the right knee joint (Fig 3, bottom). Also, the coinjection of loratadine (20 pmol) with formalin, but not that injected contralaterally, caused a significant decrease in the PL (P < .05) (Table 1). However, this treatment did not change the AD.
Effect of Histamine, Tele-Methylhistamine, and 2-Pyridylethylamine Into the Joint
Figure 2. Enhancing effect of H1R antagonists on the articular incapacitation and its blockade by sodium cromoglycate. (Top) LOR 2.5 and 10; and (Middle) CET .01, .1, 1, and 10 indicate, respectively, loratadine and cetirizine treatments at doses in mg/kg, given i.p. (Bottom) CRO 1.6, represents the dose of sodium cromoglycate coinjected with formalin. CRO 1 LOR indicates pretreatment with loratadine (10 mg/kg, i.p.) before cromoglycate-formalin coinjection; all loratadine and cetirizine treatments were given 1 hour before formalin injection. Open square represents control groups, which received 1.5% formalin alone in the knee joint and vehicle i.p., ie, 5% Tween 80 in physiological saline for loratadine, or physiological saline alone for cetirizine. The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
only the PET (P < .05). The lower doses of both antihistamines were without effect (Fig 2, top and middle). None of these treatments changed the AD and PL values (Table 1), even though doses chosen were previously reported as being able to inhibit vascular reactions due
A wide range of histamine doses (.2–20,000 nmol) was injected alone in na€ıve knee joints. Only the higher doses (200 and 20,000 nmol) produced slight incapacitation (P < .05) (Fig 4, top), PL, and AD increase (P < .05) (Table 1). However, the lowest doses of histamine (.2 and 20 nmol, knee) coinjected with formalin caused a potent inhibition of the formalin-induced incapacitation (P < .05) (Fig 5, top). From the 200-nmol dose and above, the inhibitory effect on formalininduced incapacitation was lost, although no increase was observed. The histamine metabolite tele-methylhistamine (1 and 10 pmol, knee), which acts selectively at the histaminergic site of the N-methyl-D-aspartate (NMDA) receptor,5 only increased the incapacitating effect of the formalin (P < .05) (Fig 5, middle). The selective H1R agonist 2-pyridylethylamine (.05, .5, and 5 nmol, knee) decreased the formalininduced PET (P < .05). On the other hand, a higher dose of the H1R agonist (500 nmol, knee) increased the formalin-induced PET (P < .05) (Fig 6, top). Cetirizine (.1 mg/kg, i.p.), given 1 hour before the articular coinjection of 2-pyridylethylamine with formalin, prevented the antinociceptive effect (Fig 6, middle). With the exception of the 20,000-nmol histamine dose combined with formalin, which enhanced the AD (P < .05), no other treatment affected the PL and AD induced by formalin (Table 1).
Potentiation of Spinal Morphine Analgesia by an Intra-Articular H1R Agonist Morphine (9 nmol) administered in the lumbar spinal channel 20 minutes before intra-articular formalin decreased the PET (P < .05). The lowest dose (.009 nmol) of morphine that had no effect on PET alone was potentiated by an also subeffective dose of 2-pyridylethylamine (.005 nmol) coinjected intraarticularly with formalin (P < .05) (Fig 6, bottom). None of these treatments changed the AD (data not shown).
812
Histamine Inhibits Formalin-Induced Response
The Journal of Pain
Table 1.
Effect of Histamine, H1R Agonist, and Antagonists on AD and Synovial PL in Formalin Test TREATMENTS
Histamine alone (nmol, knee)
Formalin with
Histamine (nmol, knee)
LOR (mg/kg, i.p.)
CET (mg/kg, i.p.)
LOR (pmol, knee)
LOR (pmol, contralateral knee) 2-PEA (nmol, knee)
CRO (mg, knee) LOR (i.p.) 1 CRO CET (i.p.) 1 2-PEA
DOSE
AD (MM)
PL (mG/ML)
0 0.2 20 200 20,000 0 200 20,000 0 2.5 10 0 1 10 0 2.1 20 0 20 0 5 500 1.6 2.5 1 1.6 .1 1 5
.1 6 .03 .2 6 .09 .4 6 .05 .9 6 .1* 1.2 6 .1* .4 6 .1 .7 6 .1 1.9 6 .05* .57 6 .10 .35 6 .09 .68 6 .13 .2 6 .02 .3 6 .04 .3 6 .06 .62 6 .15 .55 6 .15 .57 6 .11 .53 6 .1 .8 6 .10 .4 6 .04 .3 6 .06 .3 6 .05 .5 6 .1 .5 6 .13 .3 6 .07
7.5 6 .92 6.5 6 1.54 9.33 6 1.42 11.0 6 1.26* 12.0 6 1.06* 14.66 6 2.7 13.83 6 1.24 10.0 6 2.04 18.8 6 1.18 14.8 6 2.02 11.6 6 3.33 13.5 6 2.53 11.4 6 2.42 10.0 6 .68 23.6 6 2.86 20.6 6 2.4 14.1 6 .6* 15.5 6 3.04 13.5 6 .76 19.27 6 1.88 17.5 6 2.56 21.4 6 2.15 16.57 6 2.86 13.0 6 1.18 16.66 6 2.41
Abbreviations: LOR, loratadine; CET, cetirizine; 2-PEA, 2-pyridyl-ethylamine; CRO, sodium cromoglycate. NOTE. Values are presented as the mean 6 SEM (n = 6). Dose = 0 means the respective vehicle-treated group of each experiment. I.p. treatments were given 60 minutes before formalin. Excepting for histamine alone, all other knee treatments were coinjected with formalin. Some doses were omitted to make the table clearer, since they do not differ from controls. *Indicates P < .05 compared to the vehicle-treated group (1-way ANOVA followed by Dunnett’s post hoc test).
Effect of Local Administration of Serotonin and Serotonin Receptor Antagonists We further evaluated the involvement of serotonin (5-HT) in the present model of the knee joint nociception. Firstly, serotonin doses (1, 10, and 100 nmol) given alone in the knee joint did not increase the PET (Fig 4, bottom, only highest dose was shown), but they did increase the AD, and the highest dose also produced PL (P < .05) (Table 2). However, the same doses coinjected with formalin increased the PET (P < .05) (Fig 5, bottom) when compared with formalin alone. Only the highest dose (100 nmol) (P < .05) caused the AD to increase, and none increased the PL (Table 2). Due to the multiplicity of serotonin receptors known to be expressed in small afferent fibers, we evaluated the possible involvement of 3 that have been implicated in nociception. Antagonists of 5-HT1 (NAN-190; .1 and .6 pmol, knee), 5-HT2 (cyproheptadine; .16 and .5 pmol, knee), and 5-HT3 (ondansetron; .26 pmol, knee) receptors coinjected with formalin decreased the PET (P < .05) (Fig 7), but not the AD or PL (Table 2).
Discussion In the present study we confirmed and extended the previous observation17 that peripheral histamine H1R antagonists only increase the nocifensive response
induced by formalin in the knee joint. Also, it was shown that low doses of histamine, and the H1R agonist 2-pyridylethylamine (2-PEA), coinjected with formalin decreased the PET induced by formalin. None of these effects appeared to be due to a modification in the vascular permeability. This observation suggests that peripheral H1R activation is responsible for eliciting some kind of antinociceptive mechanism and that this mechanism is activated during formalin reaction. This conclusion was further supported when a nonanalgesic dose of morphine in the spinal cord was rendered effective by the concomitant intra-articular administration of an equally noneffective dose of an H1R agonist. The above observations contradict the general notion that the peripheral H1R activation by tissuereleased histamine plays a role in promoting pain.18,36,25 However, to date there is no clear evidence for the involvement of histamine in peripheral inflammatory pain, which has been verified for human pruritus.2,13,28 Furthermore, the role of histamine in many experimental models may be misinterpreted since the animal models employed to evaluate the histamine contribution to nociception do not discriminate nociceptionrelated behavior from pruritus-related behavior,16,29 essentially because the application of either histamine or algogens to the foot will elicit similar
Souza-Silva et al
Figure 3. Enhancing effect of locally applied loratadine on the
The Journal of Pain
813
articular incapacitation. (Top) LOR 2.1 and 20 indicate loratadine doses in pmol, coinjected with formalin. (Bottom) LOR 20 indicates loratadine dose in pmol, given in the left knee joint immediately before formalin injection in the right knee joint. Control groups (open square) received only 1.5% formalin diluted in a 5% Tween 80 physiological saline solution. The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
Figure 4. Effect of histamine and serotonin alone on the PET. (Top) HIST 20, 200, and 20,000 indicate histamine doses in nmol, knee. (Bottom) 5-HT 100 indicates serotonin dose in nmol, knee. The PET was taken before (C), immediately after the injection of both amines (0), and every 5 minutes after. Control groups received only physiological saline in the right knee joint (open square). *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
behaviors.25 Typically, the histamine doses employed to elicit nociceptive-related behaviors are far above25,32 those that may be expected to be released after a stimulus such as formalin in the tissues.10,31 Our results with the higher histamine doses are in agreement with this since we also observed an incapacitation response, as well as PL and edema with such doses. Another problem related to the correct pharmacological interpretation when using high histamine concentrations is the possibility of a direct facilitatory action at the NMDA receptor. Tele-methylhistamine is a major metabolite of histamine and a selective agonist at the histaminergic site of the NMDA receptor, but not in the histamine receptors.5 The injection of huge amounts of histamine in the tissue would likely also result in large amounts of this metabolite. It is conceivable that this facilitatory effect at NMDA receptors also facilitates formalin nociception,6 and in fact, this was observed by coinjecting tele-methylhistamine with formalin into the knee joint. However, the histamine concentration found in the synovial fluid of humans with or without arthritis is 1,000-fold lower, and in this dose range, histamine did not change the inflammatory response induced in the mouse knee joint,1 suggesting
that histamine does not participate as a physiological proinflammatory mediator. Second-generation H1R antagonists such as loratadine and cetirizine do not easily cross the blood-brain barrier, in contrast to first-generation antihistamines. Indeed, the higher loratadine and cetirizine doses used here have not been reported to produce changes in behavioral and cognitive tests.22 Furthermore, the hypernociceptive effect produced by loratadine given directly into the knee joint that received formalin, but not when given in the contralateral joint, excludes the possibility that this effect was due to an action outside the knee joint. In a previous report,17 we speculated that a blockade of a histamine vasodilating effect by the H1R antagonists could enhance the formalininduced nociception by simply inhibiting its scavenging from the synovial space. In an attempt to provide evidence for this hypothesis, we collected data on the AD and PL in the synovia. However, no changes were observed in the AD or PL after systemic treatments with H1R antagonists, and even the highest concentration of locally given loratadine only slightly diminished PL, which indicates that the vascular effect induced by formalin in the knee joint may be, at most, marginally mediated by local histamine release. Thus,
814
The Journal of Pain
Figure 5. Effect of histamine, tele-methylhistamine, and serotonin coinjected with formalin on the articular incapacitation. (Top) HIST .2, 20, 200, and 20,000 indicate histamine doses in nmol. (Middle) TELE 1 and 10 indicate tele-methylhistamine doses in pmol. (Bottom) 5-HT 1, 10, and 100 indicate serotonin doses in nmol. Control groups received 1.5% formalin alone (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
the hypernociceptive effect of the antihistamines may not be attributed to a modification of synovial permeability. The hypernociceptive effect of the H1R antagonists presupposes the release of histamine in the knee joints after formalin injection. This unexpected role of histamine was further confirmed by the antinociceptive effect produced by the local injection of either histamine or the selective H1 agonist 2-PEA. It is important to note that this effect was observed with doses lower than those usually reported as causing pronociceptive effects. In addition, the antinociceptive
Histamine Inhibits Formalin-Induced Response
Figure 6. Antinociceptive effect of H1R agonist, its reversal by a H1R antagonist, and potentiation of spinal morfine analgesia. (Top) 2-PEA .05, .5, 5, and 500 indicate the H1R agonist 2-pyridylethylamine doses in nmol coinjected with formalin in the knee joints. (Middle) CET .1 indicates the cetirizine dose in mg/kg given i.p. 60 minutes before formalin alone. In CET 1 2-PEA, the same dose of cetirizine was given before 2-PEA coinjected with formalin. (Bottom) MOR .009 and 9 indicate the morphine dose, in nmol, injected in the lumbar spinal channel 20 minutes before formalin in the knee joint. In MOR 1 2-PEA, the lowest dose of morphine was given intrathecally before 2-PEA coinjected with formalin; control groups received 1.5% formalin only (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
effect of 2-PEA was inhibited by a nonhypernociceptive dose of cetirizine, which reinforces the hypothesis that this mechanism is mediated specifically through H1 receptors. Previous reports have implicated local mast cells in the response elicited by formalin paw injection,25,19 thus raising the possibility that antihistamines may
Souza-Silva et al
The Journal of Pain
815
Effect of Serotonin and Antagonists on AD and Synovial PL
Table 2.
TREATMENTS Serotonin alone (nmol)
DOSE
0 1 10 100 Formalin with Serotonin (nmol) 0 1 10 100 NAN (pmol) .05 .1 .6 CYPRO (pmol) .16 .5 ONDA (pmol) .13 .26
AD (MM)
PL (mG/ML)
.3 6 .04 .6 6 .01* 1.0 6 .01* 1.2 6 .07* .48 6 .2 .6 6 .03 .7 6 .05 1.3 6 .2* .1 6 .08 .3 6 .1 .2 6 .05 .2 6 .06 .2 6 .05 .3 6 .1 .2 6 .1
7.5 6 1.75 8.7 6 1.21 9.06 6 2.38 14.5 6 1.85* 14.70 6 2.74 16.33 6 2.65 14.37 6 4.16 16.03 6 3.85 17.55 6 3.84 13.16 6 2.30 18.66 6 5.89 16.22 6 3.56 17.01 6 1.98 17.64 6 3.24 18.20 6 2.97
Abbreviations: NAN, NAN-190; CYPRO, cyproheptadine; ONDA, ondansetron. NOTE. Values are presented as the mean 6 S.E.M (n = 6). Dose = 0 means the respective vehicle treated group of each experiment. Excepting for serotonin alone, all other treatments were coinjected with formalin. *Indicates P < .05 compared to the vehicle-treated group (1-way ANOVA followed by Dunnett’s post hoc test).
antagonize an effect produced by histamine released from articular mast cells. The prevention of the hypernociceptive effect of loratadine with sodium cromoglycate is consistent with a role of mast cells in this model of formalin-induced nociception.23 Rodent mast cells differ from their human counterparts by way of their high content of serotonin.16 Since serotonin is more consistently reported to play a role in nociception than histamine, an investigation of the role of this amine was also carried out. Indeed, serotonin only enhanced the formalin response, even in the presence of significant PL. Furthermore, selective serotonin receptor antagonists applied with formalin only inhibited the incapacitation response, in a dosedependent fashion, and is in agreement with predictions based on previously reported data.9 The importance of the results obtained with serotonin and its antagonists is 2-fold: first, they reinforce the reliability of the model concerning to the findings to histamine and its antagonists; and second, they suggest that mast cell amines have a mutually opposing role in the articular nociception. Finally, the observation that a nonanalgesic dose of morphine in the spinal cord was rendered highly effective by the intra-articular injection of an H1R agonist strongly suggests a specific antinociceptive pathway that seems to be activated by H1 receptors in the knee joint of rats. We do not know how the neural substrate could subserve this connection from knee joint to spinal cord; however, there is evidence for a subset of small diameter primary afferent neurons expressing H1 receptors and neuropeptide Y (NPY) in dorsal root ganglion neurons of guinea pigs.15 Such fibers undergoing H1R activation in the peripheral end could promote the release of NPY in the spinal cord, where this peptide has been
Figure 7. Antinociceptive effect produced by 5-HT receptor antagonists coinjected with formalin on the articular incapacitation. (Top) CYP.16 and .5 indicate the 5-HT2 receptor antagonist cyproheptadine doses. (Middle) NAN .05, .1, and .5 indicate the 5-HT1 receptor antagonist NAN-190 doses. (Bottom) ONDA .13 and .26 indicate the 5-HT3 receptor antagonist ondansetron doses. All doses indicated are in pmol. Control groups received 1.5% formalin only (open square). The PET was taken before (C) and every 5 minutes after formalin injection. *P < .05 (2-way ANOVA, followed by Duncan test between treatments).
demonstrated to inhibit nociceptive transmission.30,35 A similar action still remains to be demonstrated in rats, although NPY has been detected in the sciatic nerves of rats,27 also suggesting the existence of sensory fibers that could be an appropriate candidate for this species. In conclusion, these results represent an important contribution to understanding the role of mast cell amines in the modulation of nociception in an early inflammatory condition, suggesting that, at least for knee joint sensory fibers, rat mast cell amines may
816
The Journal of Pain
Histamine Inhibits Formalin-Induced Response
exert opposite actions. We do not know whether histamine can play a similar role in humans. However, the further exploration of the mechanisms underlying
this antinociceptive effect of histamine in the present model may reveal a new approach to obtaining clinical analgesia.
References
17. Martins MA, Bastos LC, Tonussi CR: Formalin injection into knee joint of rats: Pharmacologic characterization of a deep somatic nociceptive model. J Pain 7:100-107, 2006
1. Adlesic M, Verdrengh M, Bokarewa M, Dahlberg L, Foster SJ, Tarkowski A: Histamine in rheumatoid arthritis. Scand J Immunol 65:530-537, 2007 €nder S, Steinhoff M, Miyashi Y, 2. Akihiko I, Roman R, Sta Schmelz M: Neurophysiology of pruritus: Interaction of itch and pain. Arch Dermatol 139:1475-1478, 2003 3. Al-Haboubi HA, Zeitlin IJ: The actions of cimetidine hydrochloride and mepyramine maleate in rat adjuvant arthritis. Eur J Pharmacol 78:175-185, 1982 4. Andersson M, Persson CG, Svensson C, Cervin-Hoberg C, Greiff L: Effects of loratadine on red wine-induced symptoms and signs of rhinitis. Acta Otolaryngol 123: 1087-1093, 2003 5. Burban A, Faucard R, Armand V, Bayard C, Vorobjev V, Arrang JM: Histamine potentiates N-methyl-D-aspartate receptors by interacting with an allosteric site distinct from the polyamine binding site. J Pharmacol Exp Ther 332: 912-921, 2010 6. Carlton SM: Peripheral NMDA receptors revisited - Hope floats. Pain 146:1-2, 2009 7. Chia YY, Lo Y, Liu K, Tan PH, Chung NC, Ko NH: The effect of promethazine on postoperative pain: A comparison of preoperative, postoperative, and placebo administration in patients following total abdominal hysterectomy. Acta Anaesthesiol Scand 48:625-630, 2004 8. De Oliveira DT, Souza-Silva E, Tonussi CR: Gingival vein punction: A new simple technique for drug administration or blood sampling in rats and mice. Scand J Lab Anim Sci 36:109-111, 2009 9. Doak GJ, Sawynok J: Formalin-induced nociceptive behavior and edema: Involvement of multiple peripheral 5-hydroxytryptamine receptor subtypes. Neuroscience 80: 939-949, 1997 10. Hardwick DC: Age changes in the histamine content of rat skin. J Physiol 28:157-165, 1954 11. Herbert MK, Schmidt RF: Histamine excites groups III and IV afferents from the cat knee joint depending on their resting activity. Neurosc Letters 305:95-98, 2001 12. Hong Y, Abbott FV: Peripheral opioid modulation of pain and inflammation in the formalin test. Eur J Pharmacol 277:21-29, 1995 €nder S, Yosipovitch G: The 13. Ikoma A, Steinhoff M, Sta neurobiology of itch. Nat Rev Neurosci 7:535-547, 2006 14. International Association For The Study Of Pain (IASP): Ethical guidelines for investigations of experimental pain in conscious animals. Pain 16:109-110, 1983 15. Kashiba H, Fukui H, Senba E: Histamine H1 receptor mRNA is expressed in capsaicin-insensitive sensory neurons with neuropeptide Y-immunoreactivity in guinea pigs. Brain Research 901:85-93, 2001 16. Kuraishi Y, Nagasawa T, Hayashi K, Satoh M: Scratching behaviour induced by pruritogenic but not algesiogenic agents in mice. Eur J Pharmacol 275:229-233, 1995
18. Mobarakeh JI, Sakurada S, Hayashi T, Orito T, Okuyama K, Sakurada T, Kuramasu A, Watanabe T, Watanabe T, Yanai K: Enhanced antinociception by intrathecally-administered morphine in histamine H1 receptor gene knockout mice. Neuropharmacology 42:1079-1088, 2002 19. Nakajima K, Obata H, Ito N, Goto F, Saito S: The nociceptive mechanism of 5-hydroxytryptamine released into the peripheral tissue in acute inflammatory pain in rats. Eur J Pain 13:441-447, 2009 20. Neil A, Benoist JM, Kayser V, Guilbaud G: Initial nociceptive sensitization in carrageenin-induced rat paw inflammation is dependent on amine autacoid mechanisms: Electrophysiological and behavioural evidence obtained with a quaternary antihistamine, thiazinamium. Exp Brain Res 65:343-351, 1987 21. Nielsen PN, Skov PS, Poulsen LK, Schmelz M, Petersen LJ: Cetirizine inhibits skin reactions but not mediator release in immediate and developing late-phase allergic cutaneous reactions. A double-blind, placebo-controlled study. Clin Exp Allergy 31:1378-1384, 2001 22. Nishiga M, Fujii Y, Konishi M, Hossen MA, Chiaki K: Effects of second-generation histamine H1 receptor antagonists on the active avoidance response in rats. Clin Exp Pharmacol Physiol 30:60-63, 2003 23. Okayama Y, Church MK: Drugs modifying the responses of mast cell and basophils, in Foreman JC (ed): Immunopharmacology of Mast Cell and Basophils. New York, NY, Academic, 1993, pp 139-152 24. Paalzow GH, Paalzow LK: Promethazine both facilitates and inhibits nociception in rats: Effect of the testing procedure. Psychopharmacology 85:31-36, 1985 25. Parada CA, Tambeli CH, Cunha FQ, Ferreira SH: The major role of peripheral release of histamine and 5-hydroxytryptamine in formalin-induced nociception. Neuroscience 102:937-944, 2001 26. Raffa RB: Antihistamines as analgesics. J Clin Pharm Ther 26:81-85, 2001 27. Roddy DR, Yaksh TL, Aimone LD, Go VL: Distribution of neuropeptide Y in the spinal cords of cat, dog, rat, man and pig. Regul Pept 29:81-92, 1990 28. Schmelz M: Specific C–receptors for itch in human skin. J Neurosci 17:8003-8008, 1997 29. Shimada GS, LaMotte RH: Behavioural differentiation between itch and pain in mouse. Pain 139:681-687, 2008 30. Smith PA, Moran TD, Abdulla F, Tumber KK, Taylor BK: Spinal mechanisms of NPY analgesia. Peptides 28:464-474, 2007 31. Telford JM, West GB: Histamine and 5-hydroxytryptamine content of tissues after prolonged treatment with polymyxin. J Pharm Pharmacol 12:254-255, 1960 32. Ting E, Roveroni RC, Ferrari LF, Lotufo CM, Veiga MC, Parada CA, Tambeli CH: Indirect mechanism of
Souza-Silva et al histamine-induced nociception in temporo-mandibular joint of rats. Life Sci 81:765-771, 2007 33. Tonussi CR, Ferreira SH: Rat knee-joint carrageenan incapacitation test: An objective screen for central and peripheral analgesics. Pain 48:421-427, 1992 34. Vickers MR, Sykes KJ: The effect of histamine on the development of adjuvant arthritis in the rat. Agents Actions 12:691-698, 1982
The Journal of Pain
817
35. Wiley RG, Lemons LL, Kline RH IV: Neuropeptide Y receptor-expressing dorsal horn neurons: Role in nocifensive reflex response to heat and formalin. Neuroscience 161:139-147, 2009 36. Yoshida A, Mobarakeh JI, Sakurai E, Sakurada S, Orito T, Kuramasu A, Kato M, Yanai K: Intrathecally-administered histamine facilitates nociception through tachykinin NK-1 and histamine H1 receptors: A study in histidine decarboxylase gene knockout mice. Eur J Pharmacol 17:55-62, 2005