Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Activation of the µ-opioid receptor (MOR) system leads to antinociception under most pain conditions, and a major site of opioid action is the dorsal horn of the spinal cord, which receives inputs from primary afferent fibers and integrates the multiple ascending and descending pathways that contribute to pain modulation (Ossipov et al., 1995). However, several studies have suggested that morphine is less effective at treating chronic neuropathic pain, as occurs after nerve injury in humans (Sindrup and Jensen, 1999) and in animal models (Wegert et al., 1997). Furthermore, the neuronal plasticity that underlies the development of neuropathic pain syndrome in some instances leads to a reduced morphine antinociception in drug-naïve conditions (Mao et al., 1995; Mayer et al., 1999)

The development of hyperalgesia (increased sensitivity to a repeated painful stimulus) in neuropathic pain states, and the insensitivity to morphine that has been observed to accompany it, are thought to share common mechanisms and neuronal substrates (Mayer et al., 1999). Although the details of these mechanisms are poorly understood, several adaptive changes in neuronal pathways have been proposed (Dickenson, 1997; Mayer et al., 1999), and changes in G protein-coupled receptor (GPCR) signaling in pain pathways have been established previously (Bohn et al., 1999; Przewlocka et al., 2002).

The coincident regulation of nociception and opioid responsiveness likely involves members of the regulator of G protein signaling (RGS) family of proteins, many of which are expressed in the nervous system (Gold et al., 1997) and serve to promote the attenuation of G protein signaling by stimulating the ability of G subunits to hydrolyze GTP and adopt an inactive GDP-bound state (Watson et al., 1996). The physiological importance of the known RGS isoforms in regulating nociception in rats has been suggested by the effects of specific antisense oligonucleotides injected into the lateral ventricle of the brain, leading to a reduction in the expression of various RGS genes (Garzòn et al., 2001). The inhibition of RGS4, 7, 9, 12, 14, or 16 expression was shown in this study to increase to varying extents the acute antinociceptive effects of morphine and/or delay the loss of sensitivity to opioids.

RGS isoforms can act upon members of the Gi/o classes of heterotrimeric G proteins (Ross and Wilkie, 2000), which mediate the physiological effects of opioid receptors. RGS2 and RGS9 have been shown to attenuate signaling by MOR expressed heterologously in frog melanocytes (Rahman et al., 1999). Recombinant RGS4 has been shown to blunt the inhibitory activity of [Leu]-enkephalin, acting via the -opioid receptor on cAMP accumulation in NG108-15 membranes (Hepler et al., 1997).

Although RGS proteins have the potential to modulate nociceptive signaling pathways, it is unknown whether they are directly involved in physiological adaptations that lead to hyperalgesia and a reduction in morphine antinociception in neuropathic pain conditions. Defining a role for RGS proteins in the modulation of nociception requires, in part, demonstrating that their expression, localization, or function changes in relevant regions of the nervous system under conditions in which chronic neuropathic pain occurs.

In this study, we have determined whether the expression of several RGS genes is regulated in the spinal cord in concert with the development of neuropathic pain induced by partial ligation of the sciatic nerve. We report that RGS4 mRNA levels are specifically up-regulated in spinal cord when hyperalgesia is fully established and insensitivity to morphine is apparent. Moreover, we provide evidence that RGS4 is expressed in the dorsal horn of the spinal cord and show that RGS4 overexpression in vitro can attenuate MOR activity. These results suggest that dynamic regulation of RGS4 expression may contribute to the changes in pain signaling that occur in neuropathic pain conditions.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials. Drugs and chemicals used in this study were obtained from the following sources: morphine-hydrochloride (Salars, Como, Italy); [D-Pen²,D-Pen⁵]-enkephalin, [D-Ala²,N-Me-Phe⁴,Gly⁵-ol]-enkephalin (DAMGO), forskolin, 3-isobutyl-1-methylxantine, and poly-D-lysine hydromide (Sigma-Aldrich, Milan, Italy). S-()-2-(1-Pyrrolidinylmethyl)-1-(4-trifluoromethylphenyl)acetyl piperidine hydrochloride (BRL 52656) was synthesized as described by Brooks et al. (1993). Tissue culture plastics were purchased from NUNC (Milan, Italy); cell culture medium reagents, fetal bovine serum, hygromycin B, nonessential amino acids, FuGENE6 transfection reagent, TRIzol RNA extraction reagent, oligo(dT)_12-18, and SuperscriptII reverse transcriptase were from Invitrogen (San Giuliano Milanese, Italy); and geneticin (G418 sulfate) was from Calbiochem Inalco (Milan, Italy). Culture plates and reagents used in the luciferase gene-reporter assay were from Packard Bioscience (Milan, Italy). All PCR reagents were from Applied Biosystems (Foster City, CA). TaqMan probes and oligonucleotides were synthesized by Chem Progress (Sesto Ulteriano, Italy).

Sciatic Nerve Ligation: Measurement of Thermal Hyperalgesia. Adult male Sprague-Dawley rats (Charles River Italica, Calco, Italy), weighing 275 to 300 g, were maintained at 22°C on a 12-h light/dark cycle and provided with a standard diet (Mucedola, Italy) and water ad libitum. After a week of acclimatization, the animals underwent sciatic nerve ligation following the method described by Seltzer et al. (1990). The left sciatic nerve of anesthetized rats (sodium pentobarbital, 50 mg/kg i.p.) was exposed surgically at high-thigh level. The dorsum of the nerve was carefully freed from surrounding connective tissues. A 7-0 silicon-treated silk suture was inserted into the nerve with a three-eighths-curved, reversed-cutting mini-needle, tightly ligated so that the dorsal one-third to one-half of the nerve thickness was trapped in the ligature, and the wound was closed with clips. After habituation to the equipment, baseline withdrawal latencies to a focused, radiant, thermal stimulus (plantar analgesia instrument; Ugo Basile, Comario, Italy) were measured in freely moving rats (Hargreaves et al., 1988) at different time points after nerve ligation. Thermal hyperalgesia was indicated if rats exhibited an altered paw posture on the side ipsilateral to the nerve injury, and a left (ligated)/right paw latency ratio of 80% or less. All experiments were performed according to the Institutional Italian Guidelines for Animal Experiments and the internal GlaxoSmithKline Animal User Committee Regulations.

Quantitative Reverse Transcription-PCR Analysis of RGS mRNA Expression. RNA was extracted from lumbar spinal cords of three sham-operated rats or three rats with a tight partial ligation of the sciatic nerve at 24 h, 3, 7, 14, 21, or 28 days after surgery. Tissues were homogenized in TRIzol reagent and total RNA was prepared from pools of three independent extractions followed by treatment with RNase-free DNase (Promega, Madison, WI) to remove DNA. First-strand cDNA was synthesized using oligo(dT)_12-18 and Superscript-II reverse transcriptase in triplicate reactions from each RNA sample. Real-time TaqMan PCR (Heid et al., 1996) was performed in MicroAmp optical 96-well reaction plates containing cDNA from 100 ng of total RNA, or with different amounts of a pCR3-mycRGS4 plasmid as a standard reference (Heximer et al., 1999). PCR master mix (20 µl) containing 1× TaqMan buffer A, 5 mM MgCl₂, 0.3 mM dATP/dCTP/dGTP, 0.6 mM dUTP, AmpErase uracil N-glycosilase (0.01 U), AmpliTaq gold DNA polymerase (1.25 U), 0.3 mM of each primer, and 0.2 mM of TaqMan probe was added to each well. Amplification conditions were 2 min at 50°C, a further 10 min at 95°C followed by 45 cycles at 95°C for 15 s, 60°C for 90 s. Reactions were performed using an ABI PRISM 7700 sequence detector (Applied Biosystems) following the manufacturer's instructions. The levels of RGS4 mRNA were normalized to the endogenous reference GAPDH and expressed as the number of copies detected after extrapolation from pCR3-mycRGS4 plasmid DNA standard curves. Each data point shown is the mean value of three determinations (each performed in triplicate) on pooled reverse transcribed RNA from three operated or three sham animals. Primers and probe for rat GAPDH were from PerkinElmer (Monza, Italy). For RGS sequences, sense, antisense, and probe oligonucleotide sequences (labeled on 5' end with 5-carboxyfluorescein and on 3' end with 5-carboxytetramethylrhodamine) were as follows: RGS4 (U27767), 5'-tgtgcaggcaacaaaagagg-3', 5'-tctgggcttcatcaaaacagg-3' and 5'-cctggattcttgcaccagagaggagacaa-3'; RGS6 (U32436), 5'-ggccgtccaagatctcaagaag-3', 5'-ccaggttgattgcacttggg-3', 5'-atgtggccaagagggtggaggaaatct-3'; RGS7 (AB024398), 5'-ggttctggttggcagtggag-3', 5'-tcgtcctggttccttcacattc-3', 5'-gaaaaggcctatccgagaggtcccctc-3'; RGS8 (AB006013), 5'-agccttccgtgccttcctg-3', 5'-aaagatcctgtgggccttgg-3', 5'-gagttctggctggcctgtgaggagttc-3'; RGS9 (AF038006), 5'-gtgcaaggcgatggatcaac-3', 5'-tgtaaatgtgggtctgcgcc-3', 5'-ctgagacacccccaccgctatgtgtt-3'; RGS11 (U32438), 5'-gctgcccgctggatcaac-3', 5'-gagcatgtagatgtgcagttgtgc-3', 5'-agcagaacaatggagtggaccctggag-3'; RGS12 (U92280), 5'-ttcctgtgcagcaaagccac-3', 5'-gctgttccttgaacatgtcgg-3', 5'-gcccagctggcagacgatattctcaat-3'; RGS14 (U92279), 5'-ttccaacagatcccagccag-3', 5'-gcctgtcggtcgatgttcac-3', 5'-tcaggaggcccacaacatctaccatga-3'; RGS17 (AW140991), 5'-tttgacaagatgatgaaggctcc-3', 5'-cttcacaggccagccagaag-3', 5'-ccttttccgagagttcctccgaacaga-3'; and GAIP (AF068136), 5'-actcccgtgttcgagaaggc-3', 5'-taggagtcccggtgcatgag-3', 5'-gcaagaaccatcaccacacacattcga-3'.

Cloning of Rat RGS4 Plasmids and Riboprobes. Rat RGS4 single-stranded cDNA was produced by reverse transcription using a gene-specific primer (5'-gtctgcagaactcttgg-3', nucleotide 572, gene accession number NM_017214). This reaction product and the specific primers 5'-aaaaaaggatccttctggatcagctgtgaggagta-3' and 5'-aaaaaatctagagcagctggaaggat tggtcaggtc-3' (BamHI and XbaI sites used for subsequent clone are underlined) were used to prepare a double-stranded RGS4 cDNA (282 base pairs) by PCR. After the PCR product was purified and cleaved with BamHI and XbaI, it was cloned into pBluescript SKII(+) (Stratagene, La Jolla, CA). The sequence of the RGS4 cDNA was checked by sequencing and this plasmid was used to prepare antisense and sense riboprobes by in vitro transcription in the presence of 50 pCi of [³³P]UTP (PerkinElmer Life Sciences, Boston, MA), according to the manufacturer's instructions (Ambion, Austin, TX).

In Situ Hybridization and Nissl Staining of Spinal Cord Sections. In situ hybridization was performed in spinal cord sections as described by Golden et al. (1999). Sections (20 µm) were cut using a cryostat (Brights, Huntingdon, UK), dried 1 to 2 h, and postfixed with cold 4% (v/v) paraformaldehyde for 15 min. After washing three times with 1× phosphate-buffered saline (pH 7.4 in diethyl pyrocarbonate-treated double-distilled water), sections were acetylated for 10 min with 0.1 M triethanolamine (Sigma-Aldrich, St. Louis, MO) containing 0.025% (v/v) acetic anhydride (Sigma-Aldrich), followed by three washes with 1× phosphate-buffered saline for 5 min. Sections were sequentially dehydrated with 50, 75, 95, and 100% ethanol (v/v), chloroform, 100 and 95% ethanol (v/v), and air-dried for 30 min. Sections were hybridized overnight with labeled probes (1 × 10⁶ cpm/section) at 55°C in 20 mM Tris-HCl (pH 7.5), 1 mM EDTA (pH 8.0), 1× Denhardt's solution, 0.3 M NaCl, 50% (v/v) deionized formamide, 20% (w/v) dextran sulfate, 300 µg/ml salmon sperm RNA, and 150 µg/ml yeast tRNA. After hybridization, samples were washed four times with 4× SSC (1× SSC consists of 30 mM sodium citrate, pH 7.4, containing 0.15 M NaCl) for 15 min at 55°C and then with 2× SSC for 20 min at 55°C, 1× RNase buffer (10 mM Tris-HCl, pH 8.0, containing 0.5 M NaCl and 1 mM EDTA) for 15 min at 37°C. Samples were then treated with RNase (20 µg/ml) for 30 min at 37°C followed by washing with 1× RNase buffer for 15 min at 37°C, 2× SSC for 20 min at 55°C, 1× SSC for 20 min at 55°C, and 0.1× SSC for 20min at 65°C. Samples were sequentially dehydrated with 50, 70, 95, and 100% ethanol (v/v) containing 0.3M sodium acetate, respectively, and air-dried for 3 h. Sections were exposed to NTB2 emulsion (Eastman Kodak, Rochester, NY) for 2 weeks at 4°C. The developed slides were counterstained with hematoxylin and eosin stains. Images were acquired by dark-field microscopy with a digital camera. For subsequent Nissl staining of sections, slides were demounted with xylene for 1 week, dehydrated with ethanol, hydrated with successive washes with ethanol and deionized H₂O, and cryeyl violet stain was applied for 1 min. Slides were then rinsed in 95% ethanol (v/v) for 1 min, 100% ethanol (v/v) for 1 min, ethanol/xylene (50:50, v/v) for 2 min, and 100% xylene for 2 min. Slides were mounted with Permount (Sigma-Aldrich) and imaged by bright-field microscopy and a digital camera.

Opioid Receptor Cell Lines and Signaling Assays. Clonal cell lines stably expressing the human µ- (hMOR) and  (hDOR)-opioid receptor were established in our laboratory to assess opioid receptor signaling activity using a gene-reporter assay. Briefly, a human embryonic kidney (HEK) 293 cell clone (5 × 10⁶ cells) stably expressing a cAMP-responsive MRE/CRE luciferase gene-reporter construct (Fitzgerald et al., 1999) was transfected with hDOR (Simonin et al., 1994) or hMOR (Mestek et al., 1995) cDNA subcloned into pcDNA3.1()/Hygro (5 µg) using the FuGENE6 transfection reagent (Roche Molecular Biochemicals, Monza, Italy), as recommended by the manufacturer. Cloning was done by growth selection using hygromycin B (400 µg/ml). Selected clones expressed a homogenous population of high affinity of receptors (approximately 1.8 pmol/mg protein for hDOR and 0.1 pmol/mg protein for hMOR) as determined by saturation binding studies using specific radioligands. Clonal cell lines stably expressing the human -opioid receptor (hKOR) (approximately 2.8 pmol/mg protein of receptor binding sites) were generated by stable expression in HEK293 cells of a hKOR cDNA after its integration into a pcDNA expression vector (GlaxoSmithKline, Harlow, UK). Gene-reporter assays in hKOR-expressing HEK293 cells were performed after transient transfection of a cAMP-responsive MRE/CRE luciferase gene-reporter construct.

RGS4 Overexpression. RGS4 overexpression in hMOR and hDOR cells was performed by transient transfection using a pCR3-mycRGS4 expression construct that directs expression of rat RGS4 functionally tagged with three copies of the c-Myc epitope at the C terminus (Heximer et al., 1999). For hKOR cells, cells were cotransfected with pCR3-mycRGS4 and an MRE/CRE-luciferase gene-reporter construct (Fitzgerald et al., 1999) to allow measurement of receptor signaling using the gene-reporter assay. Cells (4 × 10⁶ cells in a 100-mm culture dish) were seeded in 10 ml of medium and subsequently incubated for 24 h in the presence of 800 µl of transfection mix containing various amounts of DNA, and DNA/FuGENE (microgram per microliter) ratios as indicated in figure legends of the experiments described. Cells were recovered by gentle trypsinization and used for receptor gene-reporter signaling studies. Overexpression of RGS4-myc was checked by immunoblotting analysis. Briefly, cells (10⁶ cells/100 µl) were recovered in Laemmli lysis buffer. After 3 min of boiling, equal amounts of protein were resolved by electrophoresis using 12% SDS-polyacrylamide gels and subsequently blotted onto Hybond ECL membranes (Amersham Biosciences UK, Ltd. (Little Chalfont, Buckinghamshire, UK). Blots were incubated for 2 h in a 1:500 dilution of anti-c-Myc mouse monoclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA) followed by 1-h incubation with a 1:7000 dilution of horseradish peroxidase conjugated secondary antibody (Sigma-Aldrich). After washing for 2 h in 50 mM Tris-HCl buffer, pH 7.4, containing 0.15 M NaCl and 0.05% (w/v) Tween 20, RGS4-myc was detected on hyperfilm ECL by enhanced chemiluminescence (Amersham Biosciences UK, Ltd.).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Development of Hyperalgesia and Reduced Morphine Antinociception after Partial Sciatic Nerve Ligation. With the aim to establish that hyperalgesia was associated with reduced responsiveness to subsequent opioid treatment, we initially determined the effect of morphine in a rat model of neuropathic pain established in our laboratory. After partial ligation of the rat sciatic nerve (Seltzer et al., 1990), thermal hyperalgesia was fully established 3 days after ligation and was maximal 7 to 14 days after surgery (Fig. 1A). We compared the nociceptive effect of morphine (0.03, 0.1, and 0.3 mg/kg s.c.) on thermal hyperalgesia using the plantar test 1 day after ligation, at which time the animals exhibit a postoperative pain not related to neuropathic hyperalgesia, with the effect of morphine 7 days after surgical ligation of the sciatic nerve, at which time neuropathic pain is established (Fig. 1A). The antinociceptive effects of morphine (0.03 and 0.1 mg/kg s.c.) were significantly reduced by 27 and 57%, respectively, 7 days after surgery when hyperalgesia is maximal relative to 1 day after surgery (Fig. 1B). No changes in the antinociceptive effects of morphine were observed in sham-operated rats (Fig. 1C). This apparent reduction of morphine antinociception was completely overcome when a higher morphine dose (0.3 mg/kg s.c.) was used, demonstrating changes in the effective morphine dosage. These data, showing that hyperalgesia induced a reduction of morphine antinociception in the rat partial sciatic nerve ligation model of neuropathic pain, has been previously reported as well in the chronic nerve constriction injury model (Mao et al., 1995).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Morphine tolerance after hyperalgesia. A, time course of hyperalgesia in the neuropathic rat model. Hyperalgesia was measured using the plantar test in sham-operated animals and in rats with partial ligation of the sciatic nerve (Seltzer model) at different times after surgery. Paw withdrawal latencies are reported and expressed as mean values ± S.E.M. from four animals. Vertical lines show S.E.M. **, P < 0.01 and *, P < 0.05 represent significant differences compared with sham animals. B, effect of morphine on thermal hyperalgesia in neuropathic rats. Antinociceptive effects of morphine on thermal hyperalgesia induced by partial rat sciatic nerve ligation measured 1 and 7 days after surgery. C, effect of morphine on thermal hyperalgesia in sham-operated rats. Antinociceptive effects of morphine on thermal hyperalgesia induced by partial rat sciatic nerve ligation measured 1 and 7 days after surgery. Morphine was administered subcutaneously 30 min before thermal hyperalgesia measurements. Results are means ± S.E.M of 10 animals. *, P < 0.05 compared with 1-day values after injury.

Changes in RGS Expression in Spinal Cord after Partial Nerve Ligation. To investigate whether RGS proteins are up-regulated specifically in the spinal cord in conditions of injury, we assessed the spinal expression levels of nine members of the RGS family, including RGS9, which had been implicated previously in the regulation of acute morphine activity (Garzòn et al., 2001). As shown in Fig. 2, the mRNA levels of GAIP, RGS6, RGS11, and RGS17 were invariant in rats 7 days after surgery, relative to sham-operated control animals. In contrast, the expression level of three RGS genes were significantly decreased by 50% (RGS9 and RGS12) and 75% (RGS7) after nerve ligation. RGS8 and RGS14 were not expressed at detectable levels in the spinal cords of control or ligated rats. Therefore, among the 10 RGS members studied, only RGS4 showed a significant 2-fold increase in neuropathic rats, and several RGS species were shown to exhibit significant decreases in expression under these conditions (RGS9, RGS12, and RGS7). Although the reduced expression of RGS7, 9 and 12 could also contribute to the reduced activity of morphine by regulating pronociceptive receptors, we focused on RGS4, with the view that an increase in RGS expression could result in reduced MOR antinociception activity.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   RGS mRNA expression in rat spinal cord after partial nerve ligation. The mRNA level of 10 RGS genes was measured by means of quantitative TaqMan PCR in the lumbar spinal cord (pool from three animals) of rats 7 days after tight ligation of the sciatic nerve and in sham-operated animals. Data were normalized to the GAPDH reference gene and are expressed as mRNA copy numbers (triplicate determinations) in ligated animals relative to sham-operated animals.

Spinal RGS4 Expression Coincides with the Development of Hyperalgesia. To correlate changes in RGS up-regulation with the development of hyperalgesia, RGS4 mRNA levels were measured at various times after nerve ligation by performing quantitative TaqMan reverse transcription-PCR analysis. As shown in Fig. 3, RGS4 mRNA expression was up-regulated in response to sciatic nerve ligation, whereas spinal levels of RGS4 mRNA in control (sham-operated) animals were invariant 3 and 7 days after surgery. In contrast, spinal RGS4 mRNA levels 3 and 7 days after ligation were about 2-fold higher than in sham-operated animals at the same time points. Fourteen days after ligation, spinal RGS4 mRNA levels in nerve-ligated animals were similar to sham-operated control levels.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   RGS4 mRNA expression in rat spinal cord after partial nerve ligation. RGS4 mRNA levels were measured by means of quantitative TaqMan PCR in the lumbar spinal cord (pool from three animals) of sham-operated animals and of rats with a tight ligation of the sciatic nerve, at different times after surgery. Data are expressed as mRNA copy number detected after normalization to the GAPDH reference gene and are means ± S.E.M. from triplicate determinations. Statistical differences in RGS4 mRNA copy number in ligated versus sham animals are indicated as *, P < 0.05 and **, P < 0.01.

Expression of RGS4 in Dorsal Horn of Spinal Cord. RGS4 and MOR expression in the spinal cord would be expected to overlap if the up-regulation of RGS4 mRNA expression in the spinal cord as a result of nerve injury directly plays a role in the loss of sensitivity to morphine in neuropathic pain conditions. We attempted to determine the location of RGS4 expression via in situ hybridization experiments using RGS4-specific probes (Fig. 4A). Although the RGS4 in situ hybridization signal in spinal cord was weaker than in rat brain (data not shown), RGS4 mRNA could be detected throughout much of the dorsal horn of the spinal cord of control rats. Experiments using a RGS4 sense probe confirmed that the signal observed in the dorsal horn was specific (Fig. 4B). Therefore, it is likely that the up-regulation of RGS4 mRNA expression after sciatic nerve ligation that we observed occurred in the dorsal horn of the spinal cord. However, it should be noted that at the level of resolution provided by this technique, we were unable to detect changes in the level or sites of RGS4 mRNA expression in spinal cords of neuropathic rats (data not shown). Because rat MOR mRNA is known to be expressed most abundantly in laminae I and II of the dorsal horn (Peckys and Landwehrmeyer, 1999), these results point to an overlap in RGS4 and MOR expression. Therefore, an up-regulation of RGS4 mRNA after sciatic nerve ligation could have the potential to regulate MOR signaling, contributing to the attenuation of morphine antinociception.

View larger version (85K): [in this window] [in a new window]   Fig. 4.   RGS4 expression in rat spinal cord by in situ hybridization. RGS4 antisense (A) and sense (B) probes were prepared as described under Materials and Methods. RGS4 mRNA was detected in superficial layers of the dorsal horn of lumbar spinal cord prepared from sham-operated controls (A). Preservation of the lamellar structure of spinal cord sections was demonstrated by Nissl staining (C).

Attenuation of µ-Opioid Receptor Signaling by RGS4. To establish a mechanistic basis for our results associating RGS4 up-regulation with reduced sensitivity to morphine, we further investigated whether RGS4 can affect MOR signaling at the cellular level. Studies were performed to evaluate the effects of RGS4 overexpression on the ability of MOR agonists to inhibit adenylyl cyclase activity. We used a previously characterized transcriptional reporter cell line that derives from HEK293 cells and stably coexpresses hMOR (100 fmol/mg protein) and a cAMP-responsive CRE-luciferase gene-reporter construct (Fitzgerald et al., 1999).

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Inhibition of µ-opioid receptor signaling after RGS4 overexpression. DAMGO-induced inhibition of FSK-stimulated cAMP accumulation by MOR stably expressed in HEK293 cells was determined using a luciferase gene-reporter assay. Data are presented as chemiluminescence intensities relative to FSK levels as a function of DAMGO concentration (nanomolar) in control cells () and in cells transiently transfected (106 cells) with 1 µg () or 3 µg () of pCR3-mycRGS4 plasmid, using 3 µl of FuGENE/µg DNA. Data points are means ± S.E.M. from triplicate determinations from a representative experiment.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   RGS4-induced reduction of opioid agonist signaling activity. DAMGO- and morphine-induced inhibition of FSK-stimulated cAMP accumulation was determined using the luciferase gene-reporter assay at a submaximal (30 mM DAMGO; 100 nM morphine) and at a saturating (300 nM DAMGO; 1000 nM morphine) concentration of agonist. Agonist activity was measured in hMOR-expressing control cells and in cells transiently transfected with pCR3-mycRGS4 plasmid (3 µg/106 cells) using 3 µl of FuGENE/µl DNA. The extent of the reduction of agonist-induced inhibition of luciferase expression was calculated using the formula 100  [100 × IRGS4/Icontrol] where IRGS4 and Icontrol represent the extent of DAMGO-induced inhibition of FSK-stimulated luciferase expression in transfected and in control cells, respectively. Results are expressed as percentage of maximal reduction of drug activity by RGS4 in transfected cells relative to control cells and are means ± S.E.M. from three independent experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study has provided evidence suggesting that RGS4 overexpression is associated with the development of hyperalgesia and may further play a role in the altered response to morphine that can accompany the development of neuropathic pain by regulating G protein-coupled receptor signaling pathways involved in the control of nociception. After partial ligation of the rat sciatic nerve, we showed that both thermal hyperalgesia and reduced sensitivity to morphine develop and are associated with changes in RGS4 gene expression in the lumbar spinal cord. Furthermore, we demonstrated that RGS4 is expressed in the dorsal horn of the rat spinal cord and that RGS4 overexpression in cell culture strongly attenuates signaling by all subtypes of opioid receptor, including the µ-opioid receptor subtype that mediates the actions of morphine. Therefore, the up-regulation of RGS4 in lumbar spinal cord that occurs in response to nerve injury may potentially be part of a feedback mechanism to control nociceptive signaling events, including those modulated by µ-opioid receptors.

We were unable to demonstrate that RGS4 mRNA up-regulation corresponded to an increase in RGS4 protein, because of the low levels detected and lack of specificity of RGS4 antibodies. Previous studies using cell membrane fractions and recombinant RGS4 protein have suggested that RGS4 can affect opioid receptor signaling (Hepler et al., 1997), although these results were not reproduced in a coexpression system using frog melanocytes (Potenza et al., 1999). Our findings indicate that RGS4 overexpression in HEK293 cells can blunt Gi-mediated MOR signaling, a result that is consistent with the previously reported inhibition of dopamine D2 and serotonin 5-hydroxytryptamine_1B receptor activity by overexpressed RGS4 (Yan et al., 1997; Leone et al., 2000). RGS4 could additionally affect G protein-coupled inwardly rectifying potassium channel currents through a G-mediated mechanism, as well as cAMP signaling responses, as shown in several Gi-coupled receptor systems (Inanobe et al., 2001; Keren-Raifman et al., 2001). Because the RGS domain of RGS4 and the Gi proteins are required for RGS modulation of G protein-coupled inwardly rectifying potassium channel and RGS modulation of GTPase activity, it is likely that spinal RGS4 up-regulation could completely block MOR signaling by affecting both G- and G-mediated MOR signaling pathways.

Increases in RGS4 expression relative to that of the MOR, causing a change in the available pool of active (G protein-complexed) receptor, may lead in vivo to the observed losses in morphine antinociception. In support of this view, a study by Sora et al. (2001) has shown that changes in µ-opioid receptor reserve in vivo can effect morphine dose-response relationships. The extent of RGS4-mediated inhibition of Gi-mediated MOR signaling in vitro can be partially overcome by saturating concentrations of agonist. This suggests that higher opioid concentrations would be needed in vivo to counteract RGS4 negative regulation and produce an antihyperalgesic effect. Indeed, reductions in rat neuropathic pain at 7 days postligation required the use of elevated morphine doses. The timing of RGS4 mRNA up-regulation after sciatic nerve ligation suggests its involvement in the initiation rather than the maintenance of hyperalgesia and/or development of insensitivity to morphine. The increased expression of RGS4 mRNA we observed occurs within 7 days after partial nerve ligation, by which time neuropathic pain is fully established and is associated with a reduced activity of morphine.

Studies by Nakagawa et al. (2001), showing that agonist activation of MOR and KOR results in a transient up-regulation of RGS4 mRNA in PC12 cells with a time-course profile that parallels that of opioid receptor desensitization, suggest that RGS4 can also be dynamically regulated by opioid receptors and act as a negative feedback regulator of opioid activity. Although RGS4 mRNA expression could be regulated by a variety of other pharmacological systems in our model, it is possible that RGS4 up-regulation results from the activation of antinociceptive opioid receptor systems in response to nerve injury and contributes to the development of a reduced functional activity.

Two weeks after ligation the relative levels of RGS4 mRNA were similar to that in control sham animals, even though hyperalgesia was still observed, suggesting that other regulatory mechanisms will have likely been invoked to maintain the observed hyperalgesia. RGS4 up-regulation could have broader effects on the regulation of nociception in the dorsal horn of the spinal cord beyond the modulation of opioid responsiveness. In addition to its ability to attenuate signaling by receptors coupled to Gi/Go, including antinociceptive MOR, RGS4 may inhibit signaling by receptors coupled to Gq, such as group I metabotropic glutamate receptors (Saugstad et al., 1998) and 5-hydroxytryptamine receptors (Leone et al., 2000) that have pronociceptive effects.

The regulation of both pro- and antinociceptive pathways involved in the establishment of neuropathic pain and consequent reduced sensitivity to morphine may require orchestrated changes in the activity of several different RGS family members. Studies from Garzòn et al. (2001) using oligonucleotide antisense probes suggest that blocking the expression of RGS4, RGS7, RGS9, or RGS12 increases the duration of acute morphine antinociception. Conversely, it would be expected that a reduction in morphine antinociception would be accompanied with an increase in RGS expression. However, it was our finding that RGS4 was the only family member shown to be up-regulated in vivo among the 10 RGS members studied and that RGS7, RGS9, and RGS12 seemed to be down-regulated in these studies. These RGSs may constitutively regulate antinociceptive receptors whose reduced expression and activity would contribute to increased hyperalgesia and neuropathic pain.

We suggest two models in which RGS4 up-regulation could contribute to the induction of hyperalgesia and reduced morphine antinociception. First, RGS4 up-regulation in nociceptive-specific neurons could sensitize postsynaptic responses to excitatory signals triggered by enhanced activity of C-fibers or excitatory interneurons. Postsynaptic sensitization could be due to direct inhibition of MOR signaling, thereby relieving µ-opioid-mediated inhibition of N-methyl-D-aspartate receptor activation, for example. Second, RGS4 up-regulation could exert its effects presynaptically in C-fibers or excitatory interneurons. In this model, RGS4 would attenuate signaling by GPCRs, including MOR that otherwise inhibit glutamate release; thus, up-regulation of RGS4 would facilitate glutamate release. Either mechanism could contribute to long-lasting adaptive changes, possibly coupling the induction and/or maintenance of hyperalgesia with morphine insensitivity. This is consistent with hypotheses proposing that plastic changes in the sensory and central nervous systems are closely associated with and are of critical importance to the development and maintenance of pathological pain states (Mayer et al., 1999).

It will be important to additionally determine which receptors involved in nociception besides MOR can be targeted by RGS4 inhibition. In this regard, recombinant RGS4 has been shown to preferentially attenuate signaling by muscarinic receptors relative to cholecystokinin receptors (Xu et al., 1999) in a perfused cell system, and we have found that in HEK293 cells RGS4 is able to attenuate signaling by - and -opioid receptors. The receptor targeting of RGS4 overexpression may both directly (as suggested in this study) or indirectly regulate MOR function. For instance, serotonin infusion in spinal cord can accelerate the development of morphine insensitivity by reducing the number of MOR binding sites (Li et al., 2001).

In conclusion, RGS proteins are beginning to emerge as important factors in a host of pathological conditions affecting the nervous system. These include anxiety and male aggression in mice lacking RGS2 (Olivera-Dos-Santos et al., 2000) and overexpression of RGS4 in certain populations of human schizophrenics (Mirnics et al., 2001). Our studies suggest that a family of RGS genes expressed in spinal cord may participate in the regulation of nociceptive GPCR signaling and adaptive changes of the nervous system that occur in association with hyperalgesia and accompanying morphine tolerance. Elucidation of these mechanisms may open new therapeutic directions by leading to the identification of small molecules that modulate the activities of these RGS proteins and enhance the efficacy of morphine in neuropathic pain.