Author for correspondence: Lucindo J. Quintans-Júnior, Federal University of Sergipe., Department of Physiology. Av Marechal Rondon, sn, Rosa Elze, São Cristóvão, Sergipe, CEP 49.100-000 Brazil (fax +55 79 3212 6640, e-mail email@example.com; firstname.lastname@example.org).
Citronellol (CT) is a monoterpenoid alcohol present in the essential oil of many medicinal plants, such as Cymbopogon citratus. We evaluated the antinociceptive effects of CT on orofacial nociception in mice and investigated the central pathway involved in the effect. Male Swiss mice were pretreated with CT (25, 50 and 100 mg/kg, i.p.), morphine (5 mg/kg, i.p.) or vehicle (saline + tween 80 0.2%). Thirty minutes after the treatment, we injected formalin (20 μl, 2%), capsaicin (20 μl, 2.5 μg) or glutamate (40 μl, 25 μM) into the right limb. For the action in the CNS, ninety minutes after the treatment, the animals were perfused, the brains collected, crioprotected, cut in a criostate and submitted in an immunofluorescence protocol for Fos protein. CT produced significant (p < 0.01) antinociceptive effect, in all doses, in the formalin, capsaicin and glutamate tests. The immunofluorescence showed that the CT activated significantly (p < 0.05) the olfactory bulb, the piriform cortex, the retrosplenial cortex and the periaqueductal grey of the CNS. Together, our results provide first-time evidence that this monoterpene attenuates orofacial pain at least, in part, through an activation of CNS areas, mainly retrosplenial cortex and periaqueductal grey.
Pain is a complex and multidimensional experience, having physiologically an important protective function and being activated by the spinal reflex withdrawal mechanisms [1-3]. The pain has a particular expression in the orofacial region, having a special biological meaning, making the orofacial pain a condition which is difficult to be handled by health professionals, with a prevalence estimated around 13–26% in the community [4, 5]. In fact, the orofacial region focuses some of the most common acute or chronic pains such as migraine, post-herpetic neuralgia and teeth pain .
The management of pain continues to be a major challenge for medicine. Opioids, anticonvulsants and non-steroidal anti-inflammatory drugs (NSAIDs) are the main agents used to relieve acute and chronic pain . An actual approach is to develop a new biological compound that inhibits pain from natural products, such as medicinal plants or its secondary metabolites, with enhanced efficacy and minimal side effects . Thus, the natural products remain an important source for the discovery of new drugs . Notable progress has been made in the recent years in the development of natural therapies, but there is an urgent need to discover effective and potent analgesic agents .
Therefore, the search for new compounds as therapeutic alternatives for orofacial pain has progressed, including the recent studies conducted by our group that showed the antinociceptive activity of citronellal , p-cymene , Sida cordifolia , Ocimum basilicum, (-)-linalool , atranorin  in mice orofacial nociception.
Citronellol (CT) is a monoterpenoid alcohol present in essential oils of various aromatic plant species such as Cymbopogon citratus and C. winterianus , medicinal plants used in pain and inflammatory disorders . Recently, Guimarães et al.  have demonstrated the importance of monoterpenes as possible new analgesic compounds. Some pharmacological effects such as hypotensive, vasorelaxant [18, 19], anticonvulsant  and analgesic  activities were described for CT. For these effects, probable mechanisms action have been described, such as calcium and sodium channel blockade, activation of GABAA receptors [19, 20], reduction in pro-inflammatory cytokines and activation of opioid system . However, there are no data that have evaluated the effects of CT on orofacial pain. Hence, the purpose of the present study was to evaluate the antinociceptive effects of citronellol (CT) on orofacial nociception in mice and investigate the central pathway involved in the effect.
Materials and Methods
Glutamate, capsaicin, formalin, Tween 80, ((S)-(-)-B-Citronellol, CT, 97% purity), fluoromount G, glycine and bovine serum albumin (BSA) were purchased from Sigma (MO, USA). Morphine (MOR) and Diazepam (DZP) was purchased from Cristália (SP, Brazil). Rabbit anti-Fos and donkey anti-rabbit Alexa Fluor 594 were obtained from Santa Cruz Biotechnology (CA, USA).
Adult (3-month-old) male albino Swiss mice (28–32 g) were randomly housed in appropriate cages at 21 ± 2°C with a 12-hr light: dark cycle (light from 06:00 to 18:00), with free access to food (Purina®, Rio Grande do Sul, Brazil) and tap water. All experiments were carried out between 09:00 am and 14:00 pm in a quiet room. All nociception tests were carried out by the same visual observer. Experimental protocols were approved by the Animal Care and Use Committee at the Federal University of Sergipe (CEPA/UFS # 72/11). All efforts were made to minimise the number of animals used and their discomfort.
Orofacial nociception was induced in mice by injection (s.c.) of 20 μl of 2% formalin into the right upper limb. This volume and the percentage concentration of formalin were selected from pilot studies that showed a nociceptive-related biphasic behavioural response (face-rubbing) of great intensity at periods of 0–5 min. (first phase) and 15–40 min. (second phase). Nociception was quantified at these periods by measuring the time (sec.) that the animal spent face-rubbing in the injected area with its fore or hind paws. To assess the effect of the test drug, groups of mice (n = 6, per group) were pretreated with vehicle (Saline + Tween 80 0.2%), CT (25, 50 or 100 mg/kg, i.p.) or MOR (5 mg/kg; i.p.), 0.5 hr before the local injection of formalin.
Orofacial nociception was induced in mice by injection (s.c.) of 20 μl (2.5 μg) of capsaicin into the right upper limb. Capsaicin was dissolved in ethanol, dymethil sulfoxide and distilled water (1:1:8). In pilot studies, the animals manifested nociceptive-related face-rubbing behaviour with a high intensity for 10–20 min. after capsaicin injection. Therefore, quantification of nociception was performed at this period by measuring the time (sec.) that the animal spent face-rubbing in the injected area with its fore or hind paws. Vehicle (Saline + Tween 80 0.2%), CT (25, 50 or 100 mg/kg, i.p.) or MOR (5 mg/kg; i.p.) were given to animals as described for the formalin test, 0.5 hr before the local injection of formalin.
In an attempt to provide the interaction of the CT with glutamate system, it was investigated whether the CT was able to antagonize the glutamate-induced orofacial nociception in mice. The procedure was similar to that previously described by Beirith et al.  but with some alterations. A volume of 40 μl of glutamate (25 μM) was injected into the right upper limb. Animals were observed individually for 15 min. after the glutamate injection. Quantification of nociception was performed by measuring the time (sec.) that the animal spent face-rubbing in the injected area with its fore or hind paws. Animals (n = 6, per group) were pretreated with vehicle (Saline + Tween 80 0.2%), CT (25, 50 or 100 mg/kg, i.p.) or MOR (5 mg/kg; i.p.), 0.5 hr before the local injection of glutamate.
Spontaneous locomotor activity
Mice divided into five groups (n = 6, per group) were treated with vehicle, CT (25, 50 or 100 mg/kg; i.p.) or DZP (1.5 mg/kg; i.p.). The spontaneous locomotor activity of the animals was assessed on the basis of cage activity (50 × 50 × 50 cm) at 0.5, 1 and 2 hr, for 3 min., after administration .
To evaluate the action of the test drug on the central nervous system, 90 min. after the injection of CT (25, 50 and 100 mg/kg; i.p.) or vehicle (Saline + Tween 80 0.2%), the animals (n = 6, per group) were perfused, and the brains collected and crioprotected for immunofluorescence processing to Fos protein. The time for realisation of immunofluorescence protocol was based on the studies of Barr  and Bai et al. .
Frozen serial transverse sections (20 μm) of all brains were collected on gelatinized glass slides. The tissue sections were stored at −80°C until use. The sections were washed with phosphate buffer (0.01 M) saline isotonic (PBS) five times for 5 min. and incubated with 0.1 M glycine in PBS for 10 min. Non-specific protein binding was blocked by incubation of the sections for 30 min. in a solution containing 2% BSA. Then, the sections were incubated overnight with rabbit anti-Fos as primary antibodies (1:2000). Afterwards, the sections were incubated for 2 hr with donkey anti-rabbit Alexa Fluor 594 as secondary antibodies (1:2000). The cover slip was mounted with Fluoromount G. As an immunofluorescence control for non-specific labelling, sections were incubated without primary antibody. After each stage, slides were washed with PBS five times for 5 min.
Acquisition and analyses of images
Pictures from Fos-positive brain areas were acquired for each animal with an Axioskop 2 plus, Carl Zeiss, Germany. The brain regions were classified according to Paxinus and Watsu Atlas, 1997. Neurons were counted by the free software Image J (National Institute of Health, MD, USA) using a plug-in (written by the authors) that uses the same level of label intensity to select and count the Fos-positive cells.
Values are expressed as mean ± S.E.M. The data obtained were evaluated through one-way analysis of variance (anova) followed by Tukey's test. In all cases, differences were considered significant if p < 0.05. All statistical analyses were performed using the software GraphPad Prism 5.0 (GraphPad Prism Software Inc., San Diego, CA, USA).
Administration of CT produced a reduction in face-rubbing behaviour induced by formalin (fig. 1). All test doses of CT significantly (p < 0.001) increased antinociception in the first and in the second phase when compared with control (vehicle).
CT, in all doses, significantly (p < 0.001) reduced the face-rubbing induced by administration of capsaicin (fig. 2).
Figure 3 represents the results of the orofacial nociception induced by the glutamate. CT, in all doses, decreased significantly (p < 0.01 or p < 0.001) the face-rubbing behaviour compared with the control group (vehicle).
Table 1 shows the spontaneous locomotor activity of mice treated with different doses of CT. In this test, CT, in all doses, was unable to cause a significant decrease in ambulation (number of crossings) at 0.5, 1 and 2 hr after administration, unlike DZP.
Table 1. Effect of CT (25, 50 or 100 mg/kg; i.p.) or DZP (1.5 mg/kg; i.p.) on the locomotor activity of mice
Number of crossings
Values are mean ± S.E.M. (n = 6, per group).
p < 0.001 as compared with control (vehicle) (anova followed by Tukey's test).
In the olfactory bulb, piriform cortex, retrosplenial and in the periaqueductal grey of the animals brains, the average number of neurons showing Fos protein was significantly (p < 0.05) increased by an intraperitoneal injection of CT when compared with control (vehicle). However, the intraperitoneal injections of CT at a dose of 25 mg/kg did not change the average number of neurons showing Fos protein in the piriform cortex when compared with control (figs 4 and 5).
Many studies have shown that monoterpenes have pharmacological activities, such as antinociceptive and anti-inflammatory, including in orofacial pain [12, 16, 17, 26]. Therefore, the aim of this study was to evaluate the antinociceptive effects of citronellol (CT) using orofacial nociception tests in mice and investigate a possible central nervous system (CNS) involvement.
Acute administration of CT caused an antinociceptive effect in the formalin test, which was evidenced by a decrease in face-rubbing behaviour in both phases. The biphasic component of formalin-induced nociception reflects different underlying mechanisms. The first phase is related to the direct chemical stimulation of nociceptive nerve endings, which reflects centrally mediated pain with the release of substance P [27, 28], and the second phase depends on a combination of inputs from nociceptive afferents, due to the release of excitatory amino acids, PGE2, nitric oxide (NO), tachykinin, kinins and other peptides . It has been reported that the development of hyperalgesia due to injection of formalin involves glutamatergic system, mainly NMDA receptors .
Recently, some studies have proposed that inhibition of both phases of formalin test by monoterpenes may be associated with the blockade of the voltage-dependent Na+ channels [26, 30]. As De Sousa et al.  have demonstrated that CT partially blocks voltage-dependent Na+ channels and consequently reduces neuronal excitability, it is possible to suggest that the observed antinociceptive response of CT is related to this effect. Besides, Su et al.  showed that CT inhibited iNOS enzymatic activity and was able to significantly attenuate cyclooxygenase-2 (COX-2) protein and LPS-induced mRNA expression. Brito et al.  demonstrated that CT reduced TNF-α level in pleurisy test; this suggests an involvement of cytokines in antinociceptive response. Thus, these pharmacological properties of CT may involve the inhibition of inflammatory mediators and have also been implicated in antinociceptive effect.
In the capsaicin test, CT also inhibited the nociceptive behaviour in mice. Capsaicin has a selective action on sensory fibres that convey pain sensations and elicit axon reflex vasodilatation that results from the activation of the capsaicin vanilloid receptor, a ligand and heat-gated ion channel present in small-diameter sensory neurons. A capsaicin injection is also able to increase the excitability of the spinal and trigeminal nociceptive neurons . This inhibition observed on formalin and capsaicin tests effect may be a result of an inhibition of substance P or by a blocking action on receptor neurokinin-1 (NK-1) , because studies demonstrated the evidence for the activation of NK-1 receptors, through NK-1 antagonist administration blocked the second-phase formalin test .
Brito et al.  also showed that CT is able to reduce the abdominal constriction induced by acid acetic and the nociceptive behaviour (‘paw licking’) induced by intraplantar injection of formalin. In addition, Brito et al.  described that CT has a central analgesic effect, as evidenced by the prolonged delay in response time when mice were subjected to a nociceptive stimulus during a hot plate test, corroborating the antinociceptive effect of CT demonstrated in this study.
Additionally, the results also showed that intraperitoneal administration of CT produced an inhibition of nociceptive behaviour induced by injection of glutamate into the right upper limb. Glutamate is a candidate to activate primary afferent nociceptors following its release from inflamed or injured tissues, and the N-methyl-d-aspartate (NMDA) receptor antagonist ketamine injection into the temporomandibular joint (TMJ) causes significant attenuation of the glutamate-induced TMJ pain . Thus, the inhibition of glutamate-induced nociception by CT treatment can be associated with its interaction with the glutamatergic system.
Studies have suggested that the CNS depression and the non-specific muscle relaxation effects can reduce the response of motor coordination and might invalidate the formalin and hot plate tests results [27, 28]. However, a previous study with CT, using the same doses as the present study, did not show any interference with motor coordination of the animals in the rota-rod test, hence eliminating a non-specific muscle relaxation effect of CT . To confirm this result, we evaluated the effect of CT on spontaneous locomotor activity, and our results show that CT is not capable of changing the motor coordination of animals in all doses. This observation confirms that the action of CT on orofacial pain, observed in this study, is not entirely due to an inhibitory effect on the CNS.
To demonstrate CNS action of CT, Fos protein labelled by immunofluorescence was performed, showing a significant activation of the olfactory bulb, piriform cortex, retrosplenial cortex and periaqueductal grey.
The olfactory bulb projections to piriform cortex lack any apparent spatial organisation. Mitral and tufted cells from a single glomerulus in the olfactory bulb innervate large and very diverse regions of piriform cortex, which receive direct convergent inputs from multiple mitral cells that innervate broadly distributed glomeruli . Piriform cortex also receives information from the amygdala and hippocampus and project their axons to amygdala and hypothalamus, where they may influence aggressive and mating behaviour .
The retrosplenial (RSP) cortex projects to the ipsilateral anterior pretectal nucleus (APtN), a structure implicated in antinociception, which involves local muscarinic, cholinergic, opioid and serotonergic receptors . Reis et al.  demonstrated that the electric stimulation of the occipital or RSP cortex induces antinociception in the rat tail-flick and formalin tests. This action probably occurs because of the capacity of RSP cortex to activate the APtN and, consequently, stimulate the activation of the descending pathway which runs in the dorsolateral funiculus to inhibit lamina V-type spinothalamic neurons [40, 41].
The periaqueductal grey (PAG) surrounds the midbrain aqueduct and is implicated in a wide variety of functions including opioid-mediated analgesia , so the activation of PAG exerts antinociceptive effect and inhibits the responses of spinal neurons to central and peripheral nociceptive stimulation . The antinociception evoked from the PAG is opioid receptor-mediated and can be attenuated by concurrent administration of an opioid antagonist such as naloxone [44, 45]. The antinociceptive effect of CT, as described by Brito et al. , is blocked by the administration of naloxone on the hot plate test, suggesting the involvement of the opioid central receptors in this response.
So, the activation of the olfactory bulb and piriform cortex indicates that CT has some influence on the aggressive and mating behaviour. Already, the activation of RSP cortex and PAG suggests the involvement of CNS in the antinociceptive effect of CT, corroborating the results shown in the first phase of the formalin test.
Thus, it can be concluded that CT reduced the orofacial nociceptive behaviour in mice, probably by the inhibition of peripheral mediators, as well as the activation, direct or indirect, of CNS regions, especially the RSP and PAG, cerebral areas involved in pain, being the opioid central receptors involved in this action. Further studies will enable us to understand the precise mechanisms of central and peripheral action of CT on nociception.
We thank Mr. Osvaldo Andrade Santos for technical support. This work was supported by grants from the National Council of Technological and Scientific Development (CNPq/Brazil), the Research Supporting Foundation of the State of Sergipe (FAPITEC-SE/Brazil) and Coordinating Development of Senior Staff (Capes/Brazil). We thank teacher Abilio Borghi for the grammar review on the manuscript.