Introduction

Trigeminal neuropathic pain is a frequently occurring pathological condition in humans that differs in numerous aspects from neuropathic pain affecting limbs, is mostly resistant to morphine, and represents a real challenge to therapy ^1,2,3. Although antidepressants and anticonvulsants offer some therapeutic benefit in trigeminal neuralgia, an important proportion of patients become refractory to these drugs ⁴. A recently developed rat model of trigeminal neuropathic pain obtained by chronic constrictive injury of the infraorbital nerve (CCI-ION) shares many characteristics with clinical disorders in humans suffering from trigeminal neuralgia or trigeminal neuropathic pain ^5,6,7. In particular, pain-related behavioral abnormalities observed in rats with CCI-ION are difficult to treat with tricyclic antidepressants and single or repeated administrations of morphine are ineffective on the mechanical hyperresponsiveness of animals ⁷. Chronic administration of a high dose of morphine (5 mg day^-1) starting 1 month after the ION ligation was shown to reduce mechanical hypersensitivity in CCI-ION. However, this effect rapidly vanished as tolerance to morphine developed and was finally completely absent when chronic infusion of morphine started during the first 2 weeks following ION ligation ⁸. Interestingly, by contrast to neuropathic pain-related behavior involving the limbs, mechanical allodynia associated with CCI-ION has been recently shown to be reduced by triptans, the potent antimigraine molecules with 5-HT_1B/1D receptor agonist properties ⁹. These later data thus further suggest that neuropathic pain in the trigeminal area presents important differences, compared with peripheral pain at the spinal level, which probably reflect anatomophysiological particularities of the trigeminal sensory system ^10,11.

We and others have recently demonstrated that herpes simplex virus type 1 (HSV-1)-mediated transfer and overexpression of proenkephalin A (PA) in primary sensory neurons at the lumbar level has a potent antihyperalgesic effect in rat models of chronic pain ^12,13,14. In addition, in a rat model of neuropathic pain consecutive to spinal nerve ligation, Hao and co-workers ¹⁵ also reported that overexpression of PA in lumbar dorsal root ganglia ameliorated mechanical allodynia.

Here, we examined in this particular model of severe trigeminal neuropathic pain the possible effects of targeted overproduction of PA-derived opioid peptides in trigeminal ganglion sensory neurons, with special attention to the period (2 weeks after infraorbital nerve ligation) when mechanical hypersensitivity was fully established but chronic morphine treatment was devoid of any antiallodynic potency.

Results

Peripheral inoculation of HSVLatEnk on rat vibrissal pad results in robust met-enkephalin production in trigeminal sensory ganglia neurons

Three weeks following unilateral application of an HSVLatEnk suspension on the slightly scarified left rat face, in situ hybridization histochemistry with cRNA probe of HSV latency-associated transcripts (LATs) showed accumulation of LATs only in the left trigeminal ganglion, ipsilateral to the infection. Indeed, in the right ganglion, as in the trigeminal ganglia from uninfected rats (not shown), we detected no LAT hybridization (Fig. 1A). The presence of LATs in numerous nerve soma of the left ganglion demonstrated the ability of the HSV vector to penetrate into this structure and to establish a latent infection of sensory nerves after peripheral infection.

Figure 1.

Presence of latency-associated transcripts and increases in PA mRNA levels in left trigeminal ganglion 3 weeks after HSVLatEnk inoculation on left rat vibrissal pad. (A) Ten-micrometer sections of both left and right trigeminal ganglia from HSVLatEnk-infected or uninfected rats were incubated with digoxigenin-labeled LAT cRNA. LAT expression was detected in numerous nerve cell bodies in the left (ipsilateral to the infection) trigeminal ganglion of HSVLatEnk-infected rats. In contrast, comparable to both ganglia from uninfected rats (not shown), no LAT-containing neurons were detected in the right ganglion contralateral to the HSVLatEnk-inoculated vibrissal pad. The photomicrographs are representative of three animals per group. Scale bar, 100 m. (B) Total RNA was extracted from left or right trigeminal ganglion of HSVLatEnk-infected rats (n = 3) and 2 g of total RNA was reverse-transcribed in the presence of six different dilutions of synthetic fragment and amplified for 30 cycles. Optical density of PCR amplification products of PA mRNA (402 bp) or of the synthetic fragment (241 bp) was used for the plot drawing ¹³. Representative gel analyses of PCR products are shown.

Full figure and legend (108K)

The levels of pEnkA mRNA measured by quantitative RT-PCR in the left trigeminal sensory ganglion were about ninefold higher (P < 0.001; n = 3) than those found in the right ganglion (contralateral to the infection) (Fig. 1B). Proenkephalin A mRNA concentrations in the right ganglion of HSVLatEnk-infected rats were comparable to those determined in both trigeminal ganglia of control uninfected animals (not shown). Finally, infection of rats with control HSVLat-gal vector did not affect the amount of pEnkA mRNA that remained, comparable to control values (not shown).

Immunofluorescence investigations showed that no or only a few neurons containing met-enkephalin (ME)-like immunoreactive material (MELM) were present in trigeminal sensory ganglia of uninfected or sham-infected (HSVLat-gal) rats. Infraorbital nerve, which includes peripheral branches of trigeminal sensory neurons, contained rare and discrete nerve fibers stained for MELM. In contrast, in HSVLatEnk-infected rats, numerous neuronal cell bodies moderately or heavily stained for MELM were present in the left trigeminal ganglion (Fig. 2A). Numerous nerve processes containing dense MELM labeling were visualized in the left infraorbital nerve (Fig. 2B). On the other hand, as in control rats, almost no MELM-containing neurons or nerve processes were present in the contralateral (right) ganglion or infraorbital nerve in HSVLatEnk-infected rats (Figs. 2C and 2D).

Figure 2.

Immunofluorescent detection of met-enkephalin-like material in left and right trigeminal ganglia and infraorbital nerves in HSVLatEnk-infected rats. Fifteen-micrometer sections were stained with anti-ME monoclonal antibody (Valbiotech). (A) Numerous neuronal cell bodies in the left (ipsilateral to the infection) trigeminal ganglion were heavily (arrows) or moderately (arrowheads) stained for ME. (B) The presence of immunoreactive ME-like material was also visualized in numerous nerve fibers in the left infraorbital nerve. In contrast, in (C) trigeminal ganglion and (D) infraorbital nerve dissected out from the right, uninfected side of the rat head, no, or only scarce, ME labeling was detected. The photomicrographs are representative of three to five animals per group. Scale bar, (A, C) 100 m, (B, D) 50 m.

Full figure and legend (121K)

As a major part of the central branches of the trigeminal sensory neurons project to the trigeminal spinal nucleus, we examined the presence of MELM in particular in the caudal part of this nucleus (subnucleus caudalis), where thermal and pain messages from the face territory are conveyed ^16,17. In HSVLatEnk-infected rats, MELM labeling present in superficial layers of the left part (ipsilateral to the infection) of nucleus was unchanged compared with the density and area of MELM labeling in the contralateral part of this structure (Fig. 3). Immunohistochemical staining for MELM in the trigeminal nucleus of HSVLatEnk-infected rats was consistently comparable to that found in control rats (not shown).

Figure 3.

Photomicrograph reconstitution of MELM immunolabeling in left and right trigeminal nucleus in HSVLatEnk-infected rats. Fifteen-micrometer sections prepared from the brain stem including the subnucleus caudalis of the trigeminal nucleus were processed for ME labeling. Analysis of immunolabeling through the whole trigeminal nucleus revealed no differences in density or distribution of ME staining between the left, infected side and the right side of the nucleus. Every fifth section is represented. The photomicrographs are representative of five animals per group. Scale bar, 300 m.

Full figure and legend (98K)

Chronic constriction injury of the infraorbital nerve induces severe mechanical allodynia that is attenuated by local HSVLatEnk-mediated ME overproduction

Two weeks after CCI of the left infraorbital nerve, uninfected or HSVLat-gal-infected rats presented considerably decreased response threshold to mechanical stimulation of the left face (0.20 0.07 g, mean SEM, n = 15; 0.20 0.04 g, mean SEM, n = 6, respectively) compared with values measured in sham-operated rats (12.5 g, i.e., the experimental cut-off value) or with preligation thresholds of each animal (12.5 g) (Fig. 4A). Hypersensitivity to mechanical stimulation remained constant for at least 4 weeks after the infraorbital nerve ligation (not shown). Most of the lesioned animals (16 of 21) presented a brisk withdrawal to the mechanical stimulus followed by an attack reaction. As we measured no difference in mechanical stimulation hypersensitivity after lesion of the infraorbital nerve in uninfected versus HSVLat-gal-infected rats, we will hereafter refer to animals of both groups as lesioned control rats. Mechanical stimulation of the right face territory innervated by the uninjured infraorbital nerve also showed a markedly decreased response threshold (0.26 0.07 g vs. 12.5 g in sham-operated rats) that was comparable to the values measured in the territory ipsilateral to the lesioned left infraorbital nerve.

Figure 4.

Mechanical response thresholds in control rats (preligation) and in rats with ligation of left infraorbital nerve treated or not with HSVLatEnk. Sensitivity of rats to mechanical stimulation of the vibrissal pad territory was evaluated with a graded series of von Frey filaments. (A) Unilateral loose ligation of left infraorbital nerve (ION) induced 15 days later a strong mechanical allodynia at both left face (L) and contralateral right (R) face of animals (n = 22). No difference in mechanical stimulation hypersensitivity was measured in uninfected versus HSVLat-gal-infected rats and consequently both groups were termed lesioned control rats. In rats inoculated on their left vibrissal pad with HSVLatEnk 1 week before their left ION was constricted, ligation-induced mechanical allodynia was attenuated 3 weeks (3 w) later (i.e., 2 weeks after ION ligation) only on the left face, ipsilateral to the HSVLatEnk application (n = 21). Decreased mechanical hypersensitivity in HSVLatEnk-treated rats persisted 5 weeks (5 w) after left face infection (n = 7). Data (means SEM) are expressed in grams. ^*P < 0.005 for ION ligation control versus HSVLatEnk-infected rats; ^**P < 0.001 for preligation versus ION ligation rats; two-tailed unpaired t test. (B) Animals used for the evaluation of the mechanical response threshold 3 weeks after they were inoculated with HSVLatEnk were implanted subcutaneously for 3 days with an Alzet osmotic minipump delivering 3 mg kg-1 day-1 of either naloxone (n = 6) or naloxone methiodide (n = 5) and mechanical sensitivity in their left and right (not shown) vibrissal pad territory was assessed once again. ^*P < 0.05 and ^**P < 0.001 for untreated versus naloxone and naloxone methiodide-treated HSVLatEnk-infected animals, respectively; two-tailed paired t test.

Full figure and legend (96K)

By contrast, in rats infected with HSVLatEnk on the left vibrissal pad 1 week before their left infraorbital nerve was lesioned, hypersensitivity to mechanical stimulation of the left face, developing 2 weeks after infraorbital nerve CCI, was significantly weaker than that determined in lesioned control rats (3.25 0.87 g, P < 0.005, n = 21). We never observed the abrupt attack reaction after mechanical stimulation of the left face in HSVLatEnk-infected rats. The reduced mechanical response threshold, determined in a group of HSVLatEnk-infected rats 3 weeks after infection (i.e., 4 weeks after infraorbital nerve ligation), persisted during this observation period. On the other hand, mechanical stimulation of the right face (contralateral to the infection and lesion side) of HSVLatEnk-infected rats was unaffected at any time after the treatment and the mechanical response threshold remained comparable to the values observed in lesioned control rats (Fig. 4A).

Antiallodynic potency of targeted ME overproduction is reversed by a peripherally acting opioid receptor antagonist

Mechanical hypersensitivity successive to the infraorbital nerve CCI was not modified in control or HSVLatEnk-infected rats implanted with saline-containing minipumps (not shown). Three days administration of naloxone or naloxone methiodide at the dose of 3 mg kg^-1 day^-1 was without any significant effect on bilateral mechanical response threshold in lesioned control rats (Fig. 4B), whereas in HSVLatEnk-infected rats either naloxone or naloxone methiodide reversed the antiallodynic effect (P < 0.05, n = 6 and P < 0.001, n = 5, respectively (Fig. 4B).

Discussion

Trigeminal sensory nerves relay sensory and proprioceptive information from orofacial regions and meninges. Despite evident analogies with the spinal sensory system, accumulating data are in favor of fundamental anatomical, functional, as well trophic factor requirement particularities of the trigeminal sensory system ^10,18. Together with the functional complexity of trigeminal sensory systems, this specificity might account, at least in part, for the real therapeutic challenge that represents trigeminal neuropathic pain. Interestingly, the particularity of the trigeminal sensory system is, at least in some aspects, also found in the rat model of trigeminal pain. Indeed, unlike neuropathic pain, which develops in the limb only on the side of the chronically constricted sciatic nerve ¹⁹, a pronounced bilateral mechanoallodynia results from the unilateral loose ligation of the ION ⁵, probably reflecting the anatomofunctional organization of the trigeminal complex ^11,20,21. Here we report that in this rat model of severe trigeminal neuropathic pain, resistant to morphine treatment, which thus shares numerous behavioral abnormalities with human trigeminal neuralgia and neuropathic pain, targeted genomic HSV vector-mediated overproduction of pEnkA-derived peptides in trigeminal ganglion sensory neurons evoked a potent antiallodynic effect.

The antinociceptive efficacy of HSV-driven overexpression of pEnkA was recently reported at the spinal level ^12,14 and, in particular, we demonstrated both antihyperalgesic and antiinflammatory potency of overproduced enkephalin-derived peptides in lumbar dorsal root ganglia sensory neurons of rats suffering from chronic inflammatory pain ¹³. Using a similar approach, Hao and co-workers ¹⁵ extended these data by showing that transgene-mediated enkephalin production in lumbar dorsal root ganglia may also attenuate neuropathic pain in the rat induced by lumbar fifth spinal nerve ligation and, contrary to the morphine antiallodynic effect, without signs of tolerance.

The HSV-derived vectors bearing the rat pEnkA coding sequence under the control of the LAT-long terminal repeat (LTR) promoter that we generated were deleted for the HSV thymidine kinase gene and were thus severely impaired for acute replication in neurons. After peripheral (foot pad) inoculation, these vectors have been shown to drive robust expression of pEnkA in primary sensory neurons of the lumbar dorsal root ganglia. Indeed, the production of transgene-derived authentic met-enkephalin rose progressively during the first 3 weeks after the infection ²² and then lasted for at least 2 months ¹³. In this study, peripheral application of HSVLatEnk on depilated and slightly scarified rat vibrissal pad led 3 weeks later to an important increase in transgene mRNA levels and the appearance of numerous cell bodies synthesizing and accumulating MELM in the trigeminal ganglion ipsilateral to the infection. Similar to our observation in dorsal root ganglia ¹³, HSVLatEnk infection, revealed by both LAT and transgene-derived pEnkA RNA accumulation, was restricted to the sensory trigeminal ganglion without any sign of cell damage. Numerous nerve fibers containing MELM were visualized in the left infraorbital nerve (innervating the infected face) of HSVLatEnk-infected rats, suggesting effective transport of MELM from nerve cell bodies to the peripheral terminals of infected sensory neurons. In addition, in infected animals no difference in the intensity or area of MELM staining was apparent through the subnucleus caudalis of the trigeminal spinal nucleus in the infected side where most of the central projections of sensory neurons innervating the rat face terminate ^16,17. Consistent with our previous observations at the spinal level, showing mainly peripherally oriented transport and release of overproduced MELM ^13,23, these data suggest that the major part of viral vector-derived MELM was transported to the peripheral processes of trigeminal primary sensory neurons. However, the existence of a limited contingent of sensory neurons transporting overproduced MELM also to their central projections could not be excluded. Indeed, at the spinal level we showed that few MELM-containing nerve processes projected also to the spinal cord and, using a similar approach, Goss et al. ¹⁴ reported that the antinociceptive effect of enkephalin overproduced in the dorsal root ganglia could be reversed after intrathecal administration of an opioid receptor antagonist.

Peripheral inoculation of HSVLatEnk and subsequent overproduction of pEnkA-derived peptides in trigeminal sensory ganglia attenuated the marked mechanical hypersensitivity of rats developing after the nerve injury. Although constriction of ION induces bilateral mechanical allodynia, an antiallodynic effect was observed only at the side infected with HSVLatEnk, and the mechanical stimulation of the contralateral (right) face resulted in hypersensitivity comparable to that measured in control rats with CCI-ION. In addition to the significantly increased mechanical response threshold, the HSVLatEnk-infected animals did not present the attack reaction after the mechanical stimulation of the left face, frequently observed in lesioned control rats, suggesting that the paroxysmal stabbing pain (experienced in human as an explosive-like sensation) was absent. The decrease in mechanical hypersensitivity in CCI-ION rats was comparable at 3 and 5 weeks after peripheral inoculation of HSVLatEnk, suggesting a prolonged antiallodynic effect of overproduced enkephalins. If applicable in human, the chronic character of trigeminal pain should clearly need the long-lasting efficacy of viral vector-driven enkephalin production. Long-term activity of promoters based on LAT-derived elements combined with the Moloney murine leukemia virus LTR, such as that used in our recombinant vector to drive pEnkA synthesis, has been clearly documented ^24,25. Apart from the efficacy of the vector promoter and the active production of transgene-derived pEnkA, progressive development of tolerance to continuously delivered enkephalins might represent a possible limitation of such a treatment. Experiments of longer duration than those reported here should thus be conducted in this model of trigeminal neuropathic pain to ensure the persistence of the antiallodynic efficacy of this treatment. Nevertheless, several elements are in favor of a sustained antinociceptive activity of HSV-vector-mediated enkephalin production. Indeed, in a rat model of polyarthritis we have previously shown that the antihyperalgesic effect of HSVLatEnk persisted with the same magnitude at least for 8 weeks without any signs of tolerance ¹³. The absence of tolerance development in response to continuous local production of pEnkA-derived peptides is further supported by the data of Hao et al. ¹⁵. In a model of neuropathic pain at the spinal level, these authors demonstrated that, in contrast to rapidly developing tolerance to repeated administration of morphine, enkephalins produced locally from very similar vectors maintained a stable antiallodynic effect for 4 weeks. The apparent lack of tolerance development might also reflect the conditions of the release of overproduced enkephalins from nerve terminals, apparently similar to the inducible, physiological release of endogenous opioid peptides. We, and others, have previously demonstrated that viral vector-mediated, forced production of enkephalins in control healthy rats does not induce any basal analgesic effect ^12,13. The antinociceptive potency of this treatment is "uncovered" under conditions of prolonged or chronic pain, associated with continuous stimulation and sensitization of primary sensory neurons, presumably leading to the release of overproduced enkephalins from vector-infected neurons. Finally, our observation that, in primary sensory neurons at the lumbar level, overproduced enkephalins are stored and transported in vesicular form and released after electrical stimulation of these neurons further supports the idea of an inducible rather than constitutive release of viral vector-overproduced pEnkA-derived peptides ²³.

The finding that the antiallodynic effect that followed the enkephalin overproduction in trigeminal ganglion may be reversed by the opioid receptor antagonist naloxone suggests that PA-derived opioid peptides evoked local antinociceptive effects through the stimulation of opioid receptors. The spinal trigeminal nucleus caudalis, which represents the major input of sensory neurons from the orofacial region, contains a high density of opiate receptors and both in vitro and in vivo experiments demonstrated direct inhibitory activity of morphine and of - or -opioid receptor agonists to inhibit sensory neurons in this structure ^26,27,28. Although activation of these receptors might account for the reduced mechanical hypersensitivity in HSVLatEnk-treated CCI-ION rats, our data rather support the idea that the major part of the antiallodynic potency of this treatment was mediated through the activation of peripherally located opioid receptors. Indeed, immunohistochemical experiments showed that, in line with our previous observations at the lumbar level, the major part of the transgene-derived MELM was present in the peripheral part of sensory nerves. Moreover, the finding that naloxone methiodide, an opioid receptor antagonist acting only at the peripheral level, fully reversed the antiallodynic effect observed in HSVLatEnk-treated CCI-ION rats strongly supports the hypothesis that peripherally located opioid receptors are involved in the attenuation of trigeminal neuropathic pain in this model.

Abundant literature supports the reduced responsiveness of neuropathic pain to systemic morphine (for review, see Dellemijn ²⁹). Although some efficacy of morphine is occasionally found in clinical management of neuropathic pain of other origins ³⁰, facial pain ³¹ and trigeminal neuralgia ³² appear particularly resistant to opioid therapy. In the rat model of trigeminal neuropathic pain induced by CCI-ION, variable results were reported regarding morphine efficacy. While Deseure et al. ³³ reported that a high (and, actually, sedative) dose of morphine (10 mg kg^-1 ip) could reverse mechanical allodynia in this model, single or repeated iv injections of morphine at a dose 10-fold higher (1 mg kg^-1) than that sufficient to reduce mechanical allodynia successive to CCI of the sciatic nerve ³⁴ were without any effect on the mechanical response threshold in CCI-ION rats ^7,35.

By contrast, the present data demonstrated that targeted HSV-mediated overproduction of pEnkA-derived opioid peptides in the primary sensory neurons of the trigeminal area evoked an antiallodynic effect in this model. Although sustained attenuation of hypersensitivity of animals against mechanical stimulation was without apparent signs of tolerance or undesirable side effects during the observation period, long-term experiments should nevertheless be performed to demonstrate a real potential for such approaches in the management of some extreme forms of chronic pain.

Materials and methods

HSV-derived vector construction

The recombinant HSV-derived vectors impaired for replication were constructed and produced as described previously ¹³. Briefly, a rat PA-encoding sequence under the control of the LAT-LTR promoter ²⁴ was inserted into the gC locus of HSV-1 genomic DNA deleted for the thymidine kinase gene (HSVLatEnk). The modified HSV LAT promoter allowed long-term in vivo production of transgene in rat primary sensory neurons ¹³. Control vectors were the same, except that the transcriptional unit inserted into the gC locus contained the Escherichia coli -galactosidase (LacZ) coding sequence (HSVLat-gal). Recombinant vectors were isolated by PCR analysis and purified by successive limiting dilutions. Single-plaque-isolated recombinants were then amplified on Vero cells and purified at about 10⁶ plaque-forming units (pfu) l^-1. Viral stock was saved at -80°C in saline containing 10% sucrose.

Animals and Treatments

All experiments were performed in conformity with the institutional guidelines, which are in compliance with national and international laws and policies for use of animals in neuroscience research (European Communities Council Directive No. 87848, October 1987, Ministère de l'Agriculture et de la Forêt, Service Vétérinaire de la Santé et de la Protection Animale; Permissions No. 6186 to M.P.). Male Sprague–Dawley rats (Centre d'Élevage R. Janvier, Le Genest-St Isle, France), weighing 175–200 g on arrival, were used. All animals were maintained under the same conditions (22 1°C, 60 10% relative humidity, 12 h/12 h light/dark cycle, food and water ad libitum). The animals were accustomed to the housing facilities for at least 1 week before any treatment.

Infection with recombinant vectors

Rats were profoundly anesthetized (Nembutal, 50 mg kg^-1 ip) and their left side vibrissal pad territory was depilated using a razor blade and slightly scarified. Five microliters of viral suspension (5 10⁶ pfu) of either HSVLatEnk or HSVLat-gal was applied and spread using glass tips. Treatment of sham-infected rats consisted of vehicle (5 l of 10% sucrose in 0.9% NaCl) application onto slightly scarified vibrissal pad. Animals were treated 1 week before induction of trigeminal neuropathic pain. This experimental protocol was chosen because previous studies showed that the increase in the concentrations of MELM in rat sensory ganglia following peripheral application of HSVLatEnk was maximum 3 weeks after infection ²², thus coinciding with the hyperresponsiveness to mechanical stimulation, which is fully developed 2 weeks after the nerve lesion ^5,6. In addition, 3 weeks after the infection, the rats' vibrissae were grown again and the animals' faces were completely healed, allowing the evaluation of mechanical allodynia with von Frey filaments. A group of control rats, unilaterally infected with either HSVLatEnk or HSVLat-gal but in which the infraorbital nerve was not constricted, was also prepared.

Surgery

Rats were anesthetized with an intraperitoneal injection of sodium pentobarbital (Nembutal, 50 mg kg^-1) and the head of the rat was fixed in a Horsley–Clark stereotaxic frame. The unilateral chronic constriction injury to the left infraorbital nerve was performed under direct visual control using a Zeiss operation microscope (10–25). A midline scalp incision was made, exposing skull and nasal bone. The infraorbital part of the left infraorbital nerve was exposed using a surgical procedure adapted from Gregg ³⁶ and Jacquin and Zeigler ³⁷. The edge of the orbit, formed by the maxillary, frontal, lacrimal, and zygomatic bones, was dissected free. To give access to the infraorbital nerve, the orbital contents were gently deflected with a cotton-tipped wooden rod. The infraorbital nerve was dissected free at its most rostral extent on the orbital cavity, just caudal to the infraorbital foramen. Two chromic catgut (5-O) ligations were tied loosely (with about 2 mm spacing) around the nerve. To obtain the desired degree of constriction, a criterion formulated by Bennett and Xie ¹⁹ was used: the ligations reduced the diameter of the nerve by a just noticeable amount and retarded, but did not interrupt the epineural circulation. Blood circulation through epineural vessels was checked under direct visual control using the Zeiss operation microscope. The scalp incision was closed using silk sutures (5-O).

Nociceptive test procedures

Animals were placed individually in plastic cages and allowed to adapt to the testing environment as described previously ⁹. Mechanical sensitivity was determined with a graded series of 10 von Frey filaments (Semmes–Weinstein monofilaments; Stoelting, Wood Dale, IL, USA), producing a bending force of 0.217, 0.445, 0.745, 0.976, 2.35, 4.19, 4.64, 6.00, 7.37, and 12.5 g. The 12.5-g filament, the bending force of which already turned the head of the rat, was chosen as the cutoff. Stimuli were applied within the infraorbital nerve territory, near the center of the vibrissal pad, on the hairy skin surrounding the mystacial vibrissae. These areas were stimulated on both sides of the face, ipsilateral and contralateral to the nerve constriction and/or infection. Each stimulation consisted of three consecutive applications (1 s apart) of the filament, beginning with the filament producing the lowest force. A complete series of von Frey filaments was applied following an increasing force order, until a well-defined behavioral response was triggered. As previously described ⁹, this response consisted of a brisk withdrawal of the head and/or an attack/escape reaction. The minimal force applied through von Frey filaments to trigger at least one of these behaviors was considered the mechanical response threshold. Thresholds to stimulation of the face by von Frey filaments were determined 2 days before and 3 and 5 weeks after infection (i.e., 2 and 4 weeks after surgery).

Pharmacological treatments

Mechanical hypersensitivity of the different groups of CCI-ION or infected control rats was first assessed 3 weeks after infection. The next day, animals were lightly anesthetized (Nembutal, 30 mg kg^-1 ip). The skin was incised at the level of the scapula, an Alzet osmotic minipump (Model 1007D; delivery rate 0.5 l h^-1) was implanted subcutaneously, and the incision was then sutured. Following the manufacturer's instructions osmotic minipumps were filled with either naloxone or naloxone methiodide to administer each of these drugs at the dose of 3 mg kg^-1 day^-1. "Sham"-treated animals were implanted with saline-delivering minipumps. Behavioral studies were performed on the third day after minipump implantation. Observers were blinded to the groups of CCI-ION rats (sham-, HSVLat-gal-, or HSVLatEnk-infected) and to the drug delivered. Behavioral experiments did not reveal any differences between sham-infected and HSVLat-gal-infected controls on the one hand and CCI-ION rats on the other hand.

Quantitative reverse transcription-PCR

Animals used for RT-PCR, immunohistochemical, or in situ hybridization procedures were killed by decapitation 3 weeks after infection, i.e., 2 weeks after infraorbital nerve constriction. Trigeminal ganglia were dissected at 0–4°C. Tissue pieces for RNA analyses were frozen in liquid nitrogen and stored at -80°C. Total RNA, extracted using the NucleoSpin RNA II extraction kit (Macherey-Nagel, Hoerdt, France), was quantified using as reference a scale of total RNA prepared on a cesium chloride gradient and estimated from the optical density at 260 nm. RT-PCR was performed, as previously described ²², with 2 g of each RNA sample in the presence of various amounts (0.1–80 fg) of internal synthetic standard prepared according to the PCR MIMIC construction kit (Clontech). This 241-base standard fragment, flanked with PA sequences (21 bases), was amplified with the same set of rat PA-specific primers as the cDNA. Reverse-transcribed RNA was amplified with 30 cycles (96, 58, and 72°C; 1 min each) according to the Access RT-PCR system instructions (Promega, Madison, WI, USA) using 40 pmol of primers in a mixture containing 10 mM each dNTP, 25 mM MgSO₄, 2.5 u of AMV reverse transcriptase, 2.5 u of Tfl DNA polymerase, 6.5 u of RNase inhibitor (RNasin) in 1 reaction buffer. The RT-PCR products were electrophoresed on 1.2% ethidium bromide-stained agarose gel and quantified with the gel analyzer GDS 5000 (UVP, Cambridge, UK).

Immunohistochemistry

Deeply anesthetized animals (Nembutal, 50 mg kg^-1 ip) were perfused transcardially with 100 ml of saline (0.9% NaCl) supplemented with 0.1% sodium nitrite, followed by 600 ml of 4% paraformaldehyde in PBS, at room temperature. Segments of infraorbital nerves and trigeminal ganglia were dissected out and cryoprotected in 10% sucrose (24 h, 4°C). Fifteen-micrometer cryostat sections were preincubated (30 min, room temperature) in PBS containing 0.3% Triton X-100 and 3% normal donkey serum (Jackson ImmunoResearch, USA) and then incubated overnight at 4°C in the same buffer supplemented with a monoclonal anti-ME antibody (1:1000; Valbiotech, France). After being washed in PBS, sections were incubated for 1 h with rhodamine (Cy3)-conjugated anti-mouse immunoglobulin (1:800; Interchim, France), rinsed in PBS, mounted in Fluoromount-G (Clinisciences, France), and examined using a Leica confocal microscope.

In situ hybridization

Animals were anesthetized and perfused as described in the protocol for immunohistochemistry. Trigeminal sensory ganglia were dissected, postfixed for 2 h in the same solution at 4°C, cryoprotected in 10% sucrose–PBS, frozen, and stored at -80°C until used. Ten-micrometer cryostat sections were mounted on slides, dehydrated through a graded series of ethanol concentrations (30–100%), and incubated in the presence of a cRNA probe for LATs labeled with digoxigenin-11–UTP according to the instructions of the manufacturer (Promega). Hybridization was performed overnight at 65°C in 1 SSC, 50% formamide, 10% dextran sulfate, 1 mg/ml rRNA, and 1 Denhardt's solution. On the following day, sections were washed twice in 1 SSC, 50% formamide, and 0.1% Tween 20, at 65°C, and twice with 100 mM maleic acid, 150 mM NaCl, and 1% Tween 20, at room temperature. The digoxigenin-labeled hybrids were detected with alkaline phosphatase-conjugated anti-digoxigenin antibody following the instructions of the manufacturer (Roche Products, Hertfordshire, UK).

Statistical analyses

Data presented as means SEM were subjected to the unpaired Student t test. Mechanical response thresholds after treatment versus before treatment with opioid receptor antagonists were compared using the paired Student t test. When P > 0.05, the corresponding difference was considered to be not significant.

References

1. Fromm, G. H., Terrence, C. F. and Maroon, J. C. (1984). Trigeminal neuralgia, current concepts regarding etiology and pathogenesis. Arch. Neurol. 41: 1204–1207. | PubMed | ChemPort |
2. Sweet, W. H. (1984). Deafferentation pain after posterior rhizotomy, trauma to a limb and herpes zoster. Neurosurgery 15: 928–932. | PubMed | ChemPort |
3. Gregg, J. M., Walter, J. M. and Driscoll, R. (1979). Neurosensory studies of trigeminal dysesthesia following trigeminal nerve injury. Advanced Pain Research Therapy. In: Bonica J.J., Liebeskind J.C., and Albe-Fessard D.G. (Eds). Raven Press: New York.311–315.
4. Swerdlow, M. (1984). Anticonvulsivant drugs and chronic pain. Clin. Neuropharmacol. 7: 51–82. | PubMed | ChemPort |
5. Kryzhanovski, G. N., Reshnyak, V. K., Dolgikh, V. G., Gorizontova, P. P. and Speranskaya, T. V. (1992). Trigeminal neuralgia of neuropathic origin. Bull Exp. Biol. Med. 112: 1059–1062.
6. Vos, B. P., Strassman, A. M. and Maciewicz, R. J. (1994). Behavioral evidence of trigeminal neuropathic pain following chronic constriction injury to the rat's infraorbital nerve. J. Neurosci. 14: 2708–2723. | PubMed | ChemPort |
7. Idänpään-Heikkilä, J. J. and Guilbaud, G. (1999). Pharmacological studies on a rat model of trigeminal neuropathic pain: baclofen, but not carbamazepine, morphine or tricyclic antidepressants, attenuates the allodynia-like behaviour. Pain 79: 281–290. | Article | PubMed |
8. Deseure, K., Koek, W., Adriaensen, H. and Colpaert, F. C. (2003). Continuous administration of the 5-hydroxytryptamine_1A agonist (3-chloro-4-fluoro-phenyl)-[4-fluoro-4{[(5-methyl-pyridin-2-ylmethyl)-amino]-methyl}piperidin-1-yl]-methadone (F13640) attenuates allodynia-like behavior in a rat model of trigeminal neuropathic pain. J. Pharmacol. Exp. Ther. 306: 505–514. | Article | PubMed | ChemPort |
9. Kayser, V., Aubel, B., Hamon, M. and Bourgoin, S. (2002). The antimigraine 5-HT_1B/1D receptor agonists, sumatriptan, zolmitriptan and dihydroergotamine, attenuate pain-related behaviour in a rat model of trigeminal neuropathic pain. Br. J. Pharmacol. 137: 1287–1297. | Article | PubMed | ChemPort |
10. Bereiter, D. A. and Bereiter, D. F. (2000). Morphine and NMDA receptor antagonism reduces c-fos expression in spinal trigeminal nucleus produced by acute injury to the TMJ region. Pain 85: 65–77. | Article | PubMed | ChemPort |
11. Benoist, J. M., Gautron, M. and Guilbaud, G. (1999). Experimental model of trigeminal pain in the rat by constriction of one infraorbital nerve: changes in neuronal activities in the somatosensory cortices corresponding to the infraorbital nerve. Exp. Brain Res. 126: 383–398. | Article | PubMed | ChemPort |
12. Wilson, S. P., Yeomans, D. C., Bender, M. A., Lu, Y., Goins, W. and Glorioso, J. C. (1999). Antihyperalgesic effects of infection with a preproenkephalin-encoding herpes virus. Proc. Natl. Acad. Sci. U. S. A. 96: 3211–3216. | Article | PubMed | ChemPort |
13. Braz, J., Beaufour, C., Coutaux, A., Epstein, A. L., Cesselin, F., Hamon, M. and Pohl, M. (2001). Therapeutic efficacy in experimental polyarthritis of viral-driven enkephalin overproduction in sensory neurons. J. Neurosci. 21: 7881–7888. | PubMed | ISI | ChemPort |
14. Goss, J. R., Mata, M., Goins, W. F., Wu, H. H., Glorioso, J. C. and Fink, D. J. (2001). Antinociceptive effect of a genomic herpes simplex virus-based vector expressing human proenkephalin in rat dorsal root ganglion. Gene Ther. 8: 551–556. | Article | PubMed | ISI | ChemPort |
15. Hao, S., Mata, M., Goins, W., Glorioso, J. C. and Fink, D. J. (2003). Transgene-mediated enkephalin release enhances the effect of morphine and evades tolerance to produce a sustained antiallodynic effect in neuropathic pain. Pain 102: 135–142. | Article | PubMed | ChemPort |
16. Dallel, R., Villanueva, L., Woda, A. and Voisin, D. (2003). Neurobiologie de la douleur trigéminale. Méd./Sci. 19: 567–574.
17. Dodd, J. and Kelly, J. P. (1991). Trigeminal system. Principles of Neural Science. In: Kandel E.R., Schwarz J.H., and Jessel T.M. (Eds). Elsevier: New York.701–710.
18. Kvinnsland, I. H. et al. (2004). Glial cell line-derived neurotrophic factor (GDNF) from adult rat tooth serves a distinct population of large-sized trigeminal neurons. Eur. J. Neurosci. 19: 2089–2098. | Article | PubMed |
19. Bennet, G. J. and Xie, Y. K. A. (1988). Peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man. Pain 33: 87–107. | Article | PubMed | ISI | ChemPort |
20. Vos, B. P., Benoist, J. M., Gautron, M. and Guilbaud, G. (2000). Changes in neuronal activities in the two ventral posterior medial thalamic nuclei in an experimental model of trigeminal pain in the rat by constriction of one infraorbital nerve. Somatosens. Mot. Res. 17: 109–122. | Article | PubMed | ChemPort |
21. Samsam, M. M. et al. (2001). Depletion of substance P, neurokinin A and calcitonin gene-related peptide from the contralateral and ipsilateral caudal trigeminal nucleus following unilateral electrical stimulation of the trigeminal ganglion: a possible neurophysiological and neuroanatomical link to generalized head pain. J. Chem. Neuroanat. 21: 161–169. | Article | PubMed | ChemPort |
22. Antunes-Bras, J. M., Epstein, A. L., Bourgoin, S., Hamon, M., Cesselin, F. and Pohl, M. (1998). Herpes simplex virus 1-mediated transfer of preproenkephalin A in rat dorsal root ganglia. J. Neurochem. 70: 1299–1303. | PubMed | ChemPort |
23. Antunes-Bras, J. M. et al. (2001). Met-enkephalin is preferentially transported to the peripheral processes of primary afferent fibres in both normal and proenkephalin A overexpressing rats. Neuroscience 103: 1073–1083. | Article | PubMed | ChemPort |
24. Lokensgard, J. R., Bloom, D. C., Dobson, A. T. and Feldman, L. T. (1994). Long-term promoter activity during herpes simplex virus latency. J. Virol. 68: 7148–7158. | PubMed | ISI | ChemPort |
25. Palmer, J. M. et al. (2000). Development and optimization of herpes simplex virus vectors for multiple long-term gene delivery to the peripheral nervous system. J. Virol. 74: 5604–5618. | Article | PubMed | ISI | ChemPort |
26. Suarez-Roca, H. and Maixner, W. (1992). Delta-opioid-receptor activation by [D-Pen2, D-Pen5]enkephalin and morphine inhibits substance P release from trigeminal nucleus slices. Eur. J. Pharmacol. 229: 1–7. | Article | PubMed | ChemPort |
27. Dallel, R., Dualé, C. and Molat, J. L. (1998). Morphine administered in the substantia gelatinosa of the spinal trigeminal nucleus caudalis inhibits nociceptive activities in the spinal trigeminal nucleus oralis. J. Neurosci. 18: 3529–3536. | PubMed | ChemPort |
28. Storer, R. J., Akerman, S. and Goadsby, P. J. (2003). Characterization of opioid receptors that modulate nociceptive neurotransmission in the trigeminocervical complex. Br. J. Pharmacol. 138: 317–324. | Article | PubMed | ChemPort |
29. Dellemijn, P. (1999). Are opioids effective in relieving neuropathic pain?. Pain 80: 453–462. | Article | PubMed | ISI | ChemPort |
30. Rowbotham, M. C., Reisner-Keller, L. A. and Fields, H. L. (1991). Both intravenous lidocaine and morphine reduce the pain of postherpetic neuralgia. Neurology 41: 1024–1028. | PubMed | ISI | ChemPort |
31. Zenz, M., Strumpf, M. and Tryba, M. (1992). Long-term oral opioid therapy in patients with chronic non-malignant pain. J. Pain Symptom Manage. 7: 69–77. | Article | PubMed | ChemPort |
32. Mauskop, A. (1993). Trigeminal neuralgia (tic douloureux). J. Pain Symptom Manage. 8: 148–154. | Article | PubMed | ChemPort |
33. Deseure, K., Koek, W., Colpaert, F. C. and Adriaensen, H. (2002). The 5-HT_1A receptor agonist F 13640 attenuates mechanical allodynia in a rat model of trigeminal neuropathic pain. Eur. J. Pharmacol. 456: 51–57. | Article | PubMed | ChemPort |
34. Attal, N., Chen, Y. L., Kayser, V. and Guilbaud, G. (1991). Behavioural evidence that systemic morphine may modulate a phasic pain-related behaviour in a rat model of peripheral mononeuropathy. Pain 47: 65–70. | Article | PubMed | ChemPort |
35. Christensen, D., Gautron, M., Guilbaud, G. and Kayser, V. (1999). Combined systemic administration of the glycine/NMDA receptor antagonist, (+)-HA966 and morphine attenuates pain-related behaviour in a rat model of trigeminal neuropathic pain. Pain 83: 433–440. | Article | PubMed | ChemPort |
36. Gregg, J. M. (1973). A surgical approach to the ophthalmic–maxillary nerve trunks in the rat. J. Dent. Res. 52: 392. | PubMed | ChemPort |
37. Jacquin, F. M. and Ziegler, H. P. (1983). Trigeminal orosensation and ingestive behavior in the rat. Behav. Neurosci. 97: 62–97. | Article | PubMed |

Acknowledgements

We are grateful to Andre Bogdan (INSERM U 713) for critical reading of the manuscript and helpful discussions. This work was supported by grants from INSERM, Bristol–Myers Squibb Foundation (Unrestricted Biomedical Research Grant), and Institut UPSA de la Douleur.