Abstract
Bradykinin (BK) is well known as a potent mediator of pain and hyperalgesia. Using a highly sensitive nociception test, we found that intraplantar (i.pl.) injection of BK produced nociceptive hyper-responses in partial sciatic nerve-injured mice, compared with the control sham-operated animals. By use of selective agonists and antagonists, we revealed that BK nociception in sham-operated mice was mediated through B2 receptor, whereas that in injured mice was mediated through B1 receptor. When we examined the activation of extracellular signal-regulated protein kinase (ERK) in dorsal root ganglion (DRG) neurons upon i.pl. injection of BK, phosphorylated ERK was mainly observed in unmyelinated neurons in sham-operated mice, and in case of nerve-injured mice, ERK was mainly activated in myelinated neurons and satellite cells. The B1 receptor agonist, [Lys-des-Arg9]-BK also produced nociceptive response and activated ERK only in nerve-injured mice. BK or B1 agonist-induced activation of ERK in DRG neurons of nerve-injured mice was completely blocked by pretreatment with antisense oligodeoxynucleotide (AS-ODN) for B1 receptor. We found that in sham-operated mice mainly B2 receptors were expressed in unmyelinated DRG neurons with a very little presence of B1 receptor. After nerve injury, B2 receptor expression drastically decreased, whereas B1 receptors were newly expressed mainly in myelinated DRG neurons and satellite cells. Finally, BK nociception in sham-operated mice was blocked by AS-ODN for B2 receptors and that in injured mice by AS-ODN for B1 receptors. Altogether, these findings confirm a switching of receptor and fiber subtype for BK nociception after peripheral nerve injury, which might contribute to the pathobiology of neuropathic pain.
Traumatic nerve injury is often associated with local inflammation of the injured nerve and release of proinflammatory mediators including bradykinin, prostaglandins, tumor necrosis factor-α, interleukin-1β, etc. It has been proposed that the local release of such proinflammatory mediators might play a critical role in the development and maintenance of neuropathic pain (Bennett 1999; Cui et al., 2000; Schafers et al., 2003). The nonapeptide bradykinin is one of the most potent among the proinflammatory mediators and is produced from plasma globulin kininogens by action of specific enzyme kallikreins (Bhoola et al., 1992). It participates in processes such as tissue permeability, vascular dilation, smooth muscle contraction, and sensory nerve-ending stimulation (Regoli and Barabe 1980; Bhoola et al., 1992). Since the kallikreins are normally present in their inactive forms, only a small amount of bradykinin is present in the plasma and tissues under normal physiological conditions (Bhoola et al., 1992). After tissue injury, kallikreins are activated, and bradykinin is released from damaged tissue and mediates pain and hyperalgesia (Bhoola et al., 1992; Calixto et al., 2000).
Upon release, bradykinin exerts its physiological effects through activation of two pharmacologically distinct receptors, designated B2 and B1. BK is also rapidly inactivated by proteolytic enzymes to yield potent B1 agonist, des-[Arg9]bradykinin (Bhoola et al., 1992). Proteolytic degradation of bradykinin to des-[Arg9]-bradykinin is reported to occur in human tumor ascites (Matsumura et al., 1990). Release of bradykinin and its subsequent conversion to B1 agonist, des-[Arg9]-BK, has also been reported in experimental nasal inflammation in humans (Proud et al., 1987). Although increase in plasma concentration of des-[Arg9]-bradykinin following nerve injury is unknown, rapid rise in plasma concentration of des-[Arg9]-bradykinin by bacterial endotoxins has been reported in rabbits (Raymond et al., 1995). Of the two types of bradykinin receptors, B2 receptor is constitutively expressed in normal tissue, whereas the B1 receptor expression is induced after tissue injury (Prado et al., 2002). Previous binding studies confirmed that in sensory neurons bradykinin acts through the B2 receptor under normal physiological conditions (Steranka et al., 1988). Recently, we reported that bradykinin can directly activate the peripheral nociceptor endings to produce nociceptive flexion responses in naive mice (Inoue and Ueda 2000). Moreover, neonatal capsaicin treatment, which destroys most unmyelinated C-fibers, caused complete loss of bradykinin nociception, suggesting that BK acts on unmyelinated nociceptor endings under naive condition (Ueda et al., 2000). On the other hand, although the plasticity in the expression of bradykinin B1 and B2 receptors in DRG neurons following peripheral nerve injury has been reported by agonist binding studies and polymerase chain reactions (Petersen et al., 1998; Levy and Zochodne 2000), their expression pattern and exact localization in the DRG neurons in naive and nerve-injured states have been largely unknown. Moreover, it remains unclear about the type of nociceptive fibers and subtype of bradykinin receptors that mediate bradykinin-induced nociceptive responses after nerve injury.
The extracellular signal-regulated protein kinases (ERK), a member of the mitogen-activated protein kinase family, are activated by membrane depolarization and calcium influx during generation of action potentials in neuronal cells (Baraban et al., 1993). Very recently, activity-dependent phosphorylation of ERK in DRG neurons by application of peripheral noxious stimulation has been reported (Dai et al., 2002). In the present study, using this principle of activity-dependent phosphorylation of ERK in DRG neurons by peripheral bradykinin stimulation, we identified the type of primary afferent neurons and subtype bradykinin receptors that mediate bradykinin nociception in the nerve-injured state. Moreover, using both behavioral and immunohistochemical techniques, we confirmed the switching of bradykinin nociception from B2 to B1 receptors after peripheral nerve injury.
Materials and Methods
Animals. Male ddY mice weighing 20 to 25 g were used throughout the experiments. The mice were housed in a room maintained at 21 ± 2°C, 55 ± 5% relative humidity, and an automatic 12-h light/dark cycle with free access to standard laboratory diet and tap water. The animals were adapted to the testing environment (maintained at 21 ± 2°C, 55 ± 5% relative humidity, and 12-h light/dark cycle) by keeping them in the testing room 24 h before the experiments. Experiments were performed during the light phase of the cycle (10:00 AM-5:00 PM). All procedures were approved by Nagasaki University Animal Care Committee and complied with the recommendations of International Association for the Study of Pain (Zimmermann, 1983). Neonatal capsaicin treatment, which degenerates small diameter sensory neurons, was done as described previously (Rashid et al., 2003) by injecting 50 mg/kg capsaicin (Nacalai Tesque, Kyoto, Japan) into the back of newborn (P4) pup.
Drugs. The following drugs were purchased: bradykinin, Hoe140, [des-Arg10]-Hoe140, [Lys-des-Arg9]-BK (Sigma-Aldrich, St. Louis, MO), morphine hydrochloride (Takeda Chemical Industries, Osaka, Japan), and [Tyr8]-BK (Peptide Institute, Inc., Osaka, Japan). FR-173657 (Inamura et al., 1997) was kindly provided by Fujisawa Pharmaceutical Co. (Osaka, Japan). All drugs were dissolved in physiological saline.
Oligonucleotides. The antisense oligodeoxynucleotide (ASODN) for B2 receptor (5′-AGAATTCTGTTCACTGTTTCTTCCCTG-3′) and its missense oligodeoxynucleotide (MS-ODN; 5′-AGGATCCGAAATGTTCAACGTCACCACA-3′), the AS-ODN for B1 receptor (5′-AGGTTCCTGTGGATGGCGTCCC-3′), and its MS-ODN (5′-AGTGTCTCGGTAGTGCGGCTCC-3′) were synthesized, freshly dissolved in physiological saline, and used for intrathecal (i.t.) injections. Ten micrograms of AS-ODN or MS-ODN were injected i.t. 1 day before surgery and then on days 1, 3, and 5 after surgery. On day 6, the treated mice were assessed for BK nociception, and the DRGs were removed for immunohistochemical experiment. The dosage and time schedule for antisense oligodeoxynucleotide treatment was selected according to the findings from our previous study where significant down-regulation of μ-opioid receptor expression was observed in DRG after i.t. injections of its AS-ODN (Inoue and Ueda 2000).
Partial Ligation of Sciatic Nerve. Partial ligation of sciatic nerve of mice was performed under pentobarbital anesthesia (50 mg/kg, i.p.), following the method of Malmberg and Basbaum (1998). Briefly, common sciatic nerve of the right hind limb was exposed at high thigh level through a small incision, and dorsal one-third to one-half of the nerve thickness was tightly ligated with a silk suture. Sham operation was performed similarly, except without touching the sciatic nerve. Immediately following surgery, the animals were kept in a soft bed cage with some food inside so they could feed themselves without difficulty when standing. The wound healed within 1 to 2 days, and the animals behaved normally. In our experiments, all of the nerve-ligated mice showed hyper-responses to BK in the algogenic-induced nociceptive flexion (ANF) test. Moreover, similar extent of hyper-responses to i.pl. BK was observed in nerve-injured mice at days 7 and 14 after nerve ligation (data not shown). However, experiments in the present study were carried out 7 days after nerve ligation or sham operation to lessen animal sufferings. No hyper-responses to i.pl. BK was observed in the contralateral paw of the nerve-injured mice (data not shown).
ANF Test. Experiments were performed as described previously (Inoue and Ueda 2000; Ueda et al., 2000). Briefly, mice were lightly anesthetized with ether and held in a cloth sling suspended on a metal bar with four limbs hanging free through holes. Three limbs were fixed to the floor, whereas the right hind limb was connected to an isotonic transducer and recorder. All experiments were started after complete recovery from the light ether anesthesia. A polyethylene cannula (0.61 mm in outer diameter) filled with algogenic substance bradykinin or B1 and B2 agonists was connected to a 50-μl Hamilton microsyringe and then carefully inserted into the under-surface of the right hind paw via a 30-gauge hypodermic needle at the tip. The pain due to insertion of the needle caused some immediate nonspecific flexion responses recorded as the peak height in centimeters over the recording sheet. The biggest response among these nonspecific flexor responses was considered as the maximal reflex. The nociceptive flexion responses induced by BK or B1 and B2 agonists infused in a volume of 2 μl were then measured. The flexion responses induced by BK, or the B1 and B2 agonists, were represented as the percentage of maximal reflex in each mouse. In antagonist experiments, B1 or B2 antagonists were injected i.pl. through another cannula 10 min before BK injection. In some experiments, the results were represented as the percentage of control BK response. There was no change in the paw volume following i.pl. infusion of the kinin agonists in the ANF test.
Immunohistochemistry. For B1 and B2 immunohistochemistry, DRG sections were processed as described previously (Rashid et al., 2003). L4-L5 DRGs from naive mice and, the ipsi- and contralateral L4-L5 DRGs from sham-operated or nerve-injured mice were cut at 10 μm with a cryostat, thaw-mounted on a silane-coated glass slide and air-dried overnight at room temperature (RT). The DRG sections were washed and blocked with 2% bovine serum albumin in PBST (0.1% Triton X-100 in 0.1 M potassium-free phosphate-buffered saline). The sections were reacted with the primary antibody (goat polyclonal antibodies for B2 and B1 receptors; 1:100; Santa Cruz Biotechnology, Santa Cruz, CA) for 48 h at 4°C, and the sections were then placed in Texas Red-conjugated anti-goat IgG secondary antibody (1:200; Rockland, Gilbertsville, PA) for 60 min at RT. For double immunolabeling with N52 or GFAP, these sections were rinsed and first incubated with anti-mouse IgG (1:50; Cappel Laboratories, Aurora, Ohio) for 60 min and then reacted with mouse monoclonal antibody against the N52 clone of Neurofilament 200, a marker of myelinated fibers (anti-N52; 1:30,000; Sigma-Aldrich) or mouse monoclonal antibody against glial fibrillary acid protein (anti-GFAP; DAKO, Glostrup, Denmark) at 4°C overnight. The sections were then placed in fluorescein isothiocyanate-conjugated anti-mouse IgG (1:200; Cappel Laboratories) for 60 min at RT. The sections were rinsed and coverslipped with Perma Fluor (Thermo Shandon, Pittsburgh, PA) and examined under fluorescence microscope (Olympus, Tokyo, Japan). Specificity of the antibodies for B1 and B2 receptors were confirmed by no staining of the DRG sections in experiments excluding the primary antibodies or by blocking the antibody binding with receptor sequence-specific blocking peptide (Santa Cruz Biotechnology; sc-15045P for B1 and sc-15050P for B2). Experiments for blocking peptides were carried out according to the manufacturer's instructions. Immunohistochemistry for phosphorylated-ERK (pERK) was performed as described previously (Dai et al., 2002). Sham-operated or nerve-injured mice were deeply anesthetized with sodium pentobarbital (50 mg/kg, i.v.) and then injected i.pl. (10 nmol/20 μl) with BK or B1 agonist, [Lys-des-Arg9]-BK (Schanstra et al., 1998) ipsilateral to the operated paw. Two minutes after the i.pl. injections, mice were perfused transcardially and DRGs were processed as described above. For immunolabeling, DRG sections were washed with PBST, incubated with 3% bovine serum albumin in PBST for 60 min at RT, and then reacted with rabbit polyclonal antibody for phosphorylated-ERK (phospho-p44/42 MAP kinase, 1:100; Cell Signaling Technology, Beverly, MA) overnight at 4°C. The next day, the sections were washed and then placed in fluorescein isothiocyanate-conjugated anti-rabbit IgG (1:100; Santa Cruz Biotechnology) for 60 min at RT. Double-labeling with N52 or GFAP was then performed as described above.
Statistical Analysis. Statistical evaluations were performed using Student's t test following comparison with repeated measures analysis of variance. In experiments with different doses of antagonists, statistical evaluations were performed using one-way analysis of variance followed by Dunnett's test for post hoc analysis. Data were expressed as mean ± S.E.M. Significance was established at p < 0.05.
Results
Switching of BK Nociception after Peripheral Nerve Injury in Mice. As shown in Fig. 1A, i.pl. injection of BK produced increased nociceptive responses in partial sciatic nerve-injured mice compared with the control sham-operated animals. BK nociception in sham-operated mice was dose-dependently blocked by i.pl. pretreatment with specific B2 receptor antagonists Hoe140 and FR173657 (Fig. 1B). B1 receptor-specific antagonist, [des-Arg10]-Hoe140 did not block BK responses in sham-operated mice. On the other hand, BK responses in nerve-injured mice were not significantly affected by B2 receptor antagonists, Hoe140 or FR173657 (Fig. 1C). Instead, B1 antagonist, [des-Arg10]Hoe140 dose-dependently inhibited the BK nociception in nerve-injured mice (Fig. 1C). Pretreatment with saline instead of the antagonists did not have significant effects (data not shown). Next we examined the effects of B1 and B2 receptor-specific agonists in sham-operated and nerve-injured mice. As shown in Fig. 1D, the B2 receptor-specific agonist, [Tyr8]-BK (Tousignant et al., 1991) produced dose-dependent nociceptive flexion in sham-operated mice. However, the dose-response curve of [Tyr8]-BK was shifted rightward and flattened in nerve-injured mice, suggesting a drastic decrease in B2-mediated nociception in the injured state (Fig. 1D). On the other hand, the B1 receptor-specific agonist, [Lys-des-Arg9]-BK (Schanstra et al., 1998) did not induce significant nociceptive flexion responses in sham-operated mice whereas it produced dose-dependent flexion responses in nerve-injured animals (Fig. 1E). In neonatal capsaicin-treated sham mice, where most of the small diameter sensory neurons degenerated (Hiura and Ishizuka, 1989), BK nociception was almost completely abolished (Fig. 1F). However, partial sciatic nerve injury in neonatal capsaicin-treated mice caused reappearance of BK nociception in a dose-response pattern almost similar to that in nerve-injured mice (Fig. 1E). All these results suggest a switching of receptor subtype and nociceptive fiber type for BK nociception in nerve-injured mice.
BK-Induced Phosphorylation of ERK in Myelinated DRG Neurons and Non-Neuronal Satellite Cells in Nerve-Injured Mice. Using the principle of activity-dependent phosphorylation of ERK in DRG neurons by peripheral noxious stimulation (Dai et al., 2002), we identified the type of primary afferent neurons that mediate BK nociception in the naive- and nerve-injured state. As shown in Fig. 2A, i.pl. injection of BK in sham-operated mice caused the appearance of many pERK immunoreactive cells (green) in the DRG, which were not colocalized with the myelinated fiber marker N52 (red), suggesting that BK acts on small diameter unmyelinated primary afferent neurons under naive condition. However, in nerve-injured mice, i.pl. injection of BK ipsilateral to the injured paw induced phosphorylation of ERK in many myelinated DRG neurons (Fig. 2B) where no activation of ERK was observed in sham mice. These results suggest the appearance of bradykinin responsiveness in a previously unresponsive fiber type in nerve-injured mice. BK also activated ERK in some unmyelinated DRG neurons (Fig. 2B) as well as in non-neuronal satellite cells surrounding the large diameter DRG neurons as revealed by colocalization with GFAP (red) (Fig. 2C). The B1 receptor agonist, [Lys-des-Arg9]-BK, did not induce phosphorylation of ERK in the DRG of sham-operated mice (Fig. 2D) indicating the absence of functional B1 receptors in the naive state. On the other hand, similar to BK, [Lys-des-Arg9]-BK induced activation of ERK in large diameter DRG neurons and surrounding non-neuronal satellite cells in nerve-injured mice (Fig. 2, E and F). Moreover, i.t. pretreatment with AS-ODN for B1 receptor, but not with MS-ODN, in nerve-injured mice prevented BK- or [Lys-des-Arg9]-BK-induced activation of ERK in the DRG, confirming that B1 receptors were involved in the activation of ERK in nerve-injured mice (Fig. 2, G and H, and BK data are not shown). Since MS-ODN was designed by random variation of nucleotide positions in AS-ODN sequence, MSODN did not mask the AS-ODN, and no loss of AS-ODN′s effect was observed when MS-ODN and AS-ODN were coadministered (data not shown). In naive, sham-operated, or nerve-injured mice without the i.pl. injection of BK or [Lysdes-Arg9]-BK, no phosphorylation of ERK was observed (data not shown).
Novel Expression of B1 Receptors on Myelinated DRG Neurons and Non-Neuronal Satellite Cells after Peripheral Nerve Injury. To confirm the behavioral data of BK action through B1 receptors in nerve-injured mice, we performed further immunohistochemical studies to see the expression of B1 and B2 receptors in DRG of sham-operated and nerve-injured mice. First of all, we confirmed the specificity of the antibodies by neutralizing or blocking peptides. As shown in Fig. 3, A through D, blocking peptides for B2 and B1 receptor antibodies completely prevented the immunostaining of the DRG sections in naive and nerve-injured mice, respectively. Next, we performed double immunolabeling of B1 and B2 antibodies with an N52 clone of the Neurofilament 200, a myelinated fiber marker. We found that in control sham-operated mice mainly B2 receptors were expressed (41.4 ± 2.8% of total cells) on small diameter DRG neurons with a very little expression of B1 receptors also in unmyelinated DRG neurons (3.0 ± 0.2% of total cells) (Fig. 3, E, F, and L). Similar results were also observed in DRGs from untreated naive mice as well as in DRGs from the contralateral side of the nerve-injured and sham-operated mice (data not shown). In DRG sections of partial sciatic nerve-injured mice, B2 receptor expression drastically decreased (9.2 ± 2.2% of total cells) whereas B1 receptors were newly expressed (31.4 ± 2.3% of total cells) (Fig. 3, G, H, and L). The newly induced B1 receptors in nerve-injured mice were mainly located on myelinated large diameter DRG neurons (24.5 ± 2.0% of total cells were N52 positive). As indicated by arrows in Fig. 3G, B1 receptors were also expressed in non-neuronal satellite cells surrounding the large diameter DRG neurons in nerve-injured mice. Further immunohistochemical double labeling between glial cell marker GFAP and B1 receptor antibodies confirmed the novel expression of B1 receptors in non-neuronal satellite cells in the DRG of injured mice (Fig. 3, I-K). Figure 3L showed the percentage of cells that were immunoreactive for B1 or B2 receptor in sham-operated and nerve-injured mice. B1 receptor expression was significantly increased, whereas B2 receptor expression was significantly decreased after nerve injury.
Blockade of BK Nociception in Nerve-Injured Mice by AS-ODN for B1 Receptor. We also examined the effects of the AS-ODN for B2 and B1 receptors on BK nociception in sham-operated and nerve-injured mice. Consistent with the results in antagonist experiments (Fig. 1B), i.t. pretreatment with AS-ODN for B2 receptor, but not B1 receptor, almost completely blocked the BK nociception in sham-operated mice (Fig. 4A). In nerve-injured mice, AS-ODN for B1 receptors, but not B2 receptors, completely blocked the BK nociception (Fig. 4B). Immunohistochemical experiments confirmed significant decrease in B2 and B1 receptor expression in the DRG neurons of mice treated with AS-ODN for B2 and B1 receptors, respectively (Fig. 4, C-G). Since MS-ODN was designed by random variation of nucleotide positions in ASODN sequence, MS-ODN did not mask the AS-ODN, and no loss of AS-ODN′s effect was observed when MS-ODN and AS-ODN were coadministered (data not shown).
Discussion
Although it is unknown whether endogenous bradykinin is involved in the development and/or maintenance of neuropathic pain, its release from injured tissue site as well as the inhibition of neuropathic hyperalgesia by its receptor blockade (Levy and Zochodne, 2000; Yamaguchi-Sase et al., 2003) indicate its role in neuropathic pain following nerve injury. In this report, we showed that exogenous application of very low doses of BK at the peripheral nerve endings in mice produced potent nociceptive flexion response, and these responses were sensitized after peripheral nerve injury. We also found that, although BK nociception in control sham mice was mediated through the B2 receptors located in small diameter unmyelinated DRG neurons, in nerve-injured mice it was mediated through newly expressed B1 receptors in the myelinated DRG neurons. BK is well known to sensitize nociceptive sensory neurons during inflammatory pain (Calixto et al., 2000). BK also lowers the activation threshold of heat-activated current in nociceptive sensory neurons (Cesare and McNaughton, 1996). Such sensitization was mediated through transactivation of the vanilloid receptor 1 (VR1) by bradykinin (Chuang et al., 2001). Although we do not know whether VR1 is involved in the sensitization of BK nociception after peripheral nerve injury, our previous study (Rashid et al., 2003) showed novel expression of VR1 in large diameter myelinated DRG neurons after nerve injury where the B1 receptors were also newly expressed.
Previous studies demonstrated the presence of B1 and B2 receptors in DRG neurons mainly by reverse transcription-polymerase chain reaction or agonist binding studies (Petersen et al., 1998; Levy and Zochodne, 2000). In this report, we showed the expression of B1 and B2 receptor protein by immunohistochemical experiments. We found that in naive state, mainly B2 receptors, were expressed in the small diameter unmyelinated DRG neurons. Very scarce expression of B1 receptor was also observed in naive DRG. There have been contradictory reports on the expression of B1 receptor in naive DRG. Although some reports showed the presence of B1 receptor mRNA in naive DRG (Levy and Zochodne, 2000; Ma et al., 2000), others reported no expression of B1 receptor in normal DRG (Petersen et al., 1998; Brand et al., 2001). Our results showed very little expression of B1 receptors in naive DRG, which is consistent with our behavioral data of very little nociceptive response produced by B1 agonist [Lysdes-Arg9]-BK in sham-operated mice. On the other hand, expression of B2 receptor drastically decreased in nerve-injured mice. The drastic decrease in B2 receptor expression in DRG of partial sciatic nerve-injured mice might be due to their presence on unmyelinated C fibers where similar down-regulation of various peptides and receptors are reported (Hokfelt et al., 1994; Malmberg and Basbaum, 1998; Zhang et al., 1998). Loss of unmyelinated fibers due to nerve injury might underlie such reduced gene expression. Further studies could identify the reasons behind such reduction in B2 expression in DRG following nerve injury. These results, however, are in contrast with previous reports of increase in B2 mRNA expression (Levy and Zochodne, 2000; Lee et al., 2002). Such discrepancy might be due to differences in the neuropathy models and species of animals used, measurement of different biochemical markers like mRNA and protein, as well as different time points of measurement of these markers. Nonetheless, in nerve-injured mice B1 receptors were newly expressed in large diameter myelinated DRG neurons. Potent nociceptive response induced by B1 agonist [Lys-des-Arg9]-BK in nerve-injured mice suggests that these newly expressed B1 receptors were also functionally active. Similar induction of B1 receptor expression and their activation by B1 agonist, des-[Arg9]-BK has been reported in a rat model of inflammatory hyperalgesia (Fox et al., 2003).
The novel induction of B1 receptor expression in DRG neurons after peripheral nerve injury should be particularly interesting because, unlike B2 receptor, B1 receptor is not readily desensitized after agonist binding (Prado et al., 2002). Such a property could be useful for the sustained signaling of neuropathic pain. Although exact mechanisms of the novel expression of B1 receptors following peripheral nerve injury is unknown, cytokines such as interleukins, leukemia inhibitory factor, and tumor necrosis factor-α that are released following nerve injury are thought to up-regulate B1 receptor expression (Marceau et al., 1998). For novel expression of B1 receptor, activation of p38 MAP kinase pathway is reported to be necessary (Larrivee et al., 1998). Indeed, very recently activation of p38 MAP kinase in DRG neurons following peripheral nerve injury has been reported (Schafers et al., 2003). On the other hand, the expression of B1 receptor in non-neuronal satellite cells surrounding large diameter DRG neurons in nerve-injured mice might indicate a cross-link between neuron and glial cells in neuropathic pain, whereby non-neuronal satellite cells might somehow modulate the nociceptive signaling in neuronal cells after nerve injury. Recent findings indicate the immense role of glial cells in pathological pain (Watkins et al., 2001). Indeed, bradykinin is reported to indirectly activate neonatal rat DRG neurons in culture through activation of non-neuronal satellite cells (Heblich et al., 2001). Very recently, expression of functional bradykinin receptors has also been reported in microglial primary culture cells from postnatal rat cortex (Noda et al., 2003). Further studies can identify the exact role of the newly expressed B1 receptors in peripheral non-neuronal satellite cells in neuropathic pain.
The switching of BK nociception following partial sciatic nerve injury was further confirmed by using the principle of activity-dependent phosphorylation of ERK (Dai et al., 2002) upon i.pl. injection of BK or B1 receptor agonist. Consistent with the findings of B2 receptor expression in small diameter unmyelinated DRG neurons in sham-operated mice, i.pl. BK induced the activation of ERK only in the small-sized unmyelinated DRG neurons in sham mice. In nerve-injured mice, BK activated ERK mainly in large diameter myelinated DRG neurons where B1 receptors were newly expressed. Activation of ERK in some unmyelinated DRG neurons might be due to its effect on the residual B2 receptors in nerve-injured mice. However, pretreatment of nerve-injured mice with ASODN for the B1 receptor almost completely prevented the activation of ERK by i.pl. BK, confirming that BK acted in nerve-injured mice through the newly expressed B1 receptors. Although in vitro studies showed that BK has higher affinity toward B2 receptors than the B1 receptors (Marceau et al., 1998), our in vivo and biochemical results suggest that BK can also activate the B1 receptor with considerable potency. Activation of ERK only in nerve-injured mice by high affinity B1 receptor agonist, [Lys-des-Arg9]-BK and its blockade by pretreatment with AS-ODN for B1 receptor, further confirmed that the newly induced B1 receptors were functionally active. Another interesting finding was the activation of ERK on non-neuronal satellite cells surrounding the myelinated DRG neurons by i.pl. BK or [Lys-des-Arg9]-BK in nerve-injured mice. Although we do not know whether it is correlated to the de novo expression of B1 receptors in non-neuronal satellite cells surrounding the myelinated DRG neurons in nerve-injured mice, one possibility could be that the activation of ERK in the cell body of DRG neurons may in turn immediately activate nearby satellite cells via non-genomic pathways such as by activation of cytoskeletal proteins or modulation of membrane ion channels (Lewis et al., 1998). Activation of ERK by BK or B1 agonist in both neurons and non-neuronal satellite cells suggests that peripheral nerve injury-induced release of kinin agonists may induce long-term changes to produce the chronic pain. Dina et al. (2003) recently reported the induction of an ERK-dependent secondary hyperalgesia through involvement of cytoskeletal proteins that occurs long after the initial inflammatory reactions.
In conclusion, the present study revealed a switching of receptor subtype and nociceptive fiber type for BK-mediated nociception after partial sciatic nerve injury in mice. Such phenotypic changes may contribute to the development and/or maintenance of neuropathic pain following peripheral nerve injury. Moreover, novel expression of functionally active B1 receptors in myelinated DRG neurons after nerve injury indicates the potential of B1 antagonists to treat neuropathic pain.
Acknowledgments
We thank Fumiko Fujiwara, Saori Kondo, Toshiko Kawashima, and Megumi Kawakami for technical assistance.
Footnotes
-
Parts of this study were supported by grants-in-aid from the Ministry of Education, Science, Culture, and Sports of Japan and a grant for the Human Frontier Science Program.
-
DOI: 10.1124/jpet.103.060335.
-
ABBREVIATIONS: BK, bradykinin; ANF, algogenic-induced nociceptive flexion; DRG, dorsal root ganglion; i.pl., intraplantar; ERK, extracellular signal-regulated protein kinase; pERK, phosphorylated-ERK; AS-ODN, antisense oligodeoxynucleotide; MS-ODN, missense oligodeoxynucleotide; RT, room temperature; MAP kinase, mitogen-activated protein kinase; N52, the N52 clone of the neurofilament 200.
- Received September 19, 2003.
- Accepted November 20, 2003.
- The American Society for Pharmacology and Experimental Therapeutics