Direct phosphorylation of capsaicin receptor VR1 by protein kinase Cepsilon and identification of two target serine residues.

The capsaicin receptor, VR1, is a sensory neuron-specific ion channel that serves as a polymodal detector of pain-producing chemical and physical stimuli. It has been reported that ATP, one of the inflammatory mediators, potentiates the VR1 currents evoked by capsaicin or protons and reduces the temperature threshold for activation of VR1 through metabotropic P2Y(1) receptors in a protein Kinase C (PKC)-dependent pathway, suggesting the phosphorylation of VR1 by PKC. In this study, direct phosphorylation of VR1 upon application of phorbol 12-myristate 13-acetate (PMA) was proven biochemically in cells expressing VR1. An in vitro kinase assay using glutathione S-transferase fusion proteins with cytoplasmic segments of VR1 showed that both the first intracellular loop and carboxyl terminus of VR1 were phosphorylated by PKCepsilon. Patch clamp analysis of the point mutants where Ser or Thr residues were replaced with Ala in the total 16 putative phosphorylation sites showed that two Ser residues, Ser(502) and Ser(800) were involved in the potentiation of the capsaicin-evoked currents by either PMA or ATP. In the cells expressing S502A/S800A double mutant, the temperature threshold for activation was not reduced upon PMA treatment. The two sites would be promising targets for the development of substance modulating VR1 function, thereby reducing pain.

The sensation of pain allows us to recognize injury and triggers appropriate protective responses. A specific population of primary afferent neurons called nociceptors are known to be involved in the detection of noxious thermal, mechanical, or chemical stimuli and can be distinguished by their sensitivity to capsaicin, the pungent ingredient in hot chili peppers (1)(2)(3)(4). The capsaicin receptor VR1 1 is a nonspecific cation channel with six transmembrane domains expressed predominantly in unmyelinated C fibers and activated not only by capsaicin but also by noxious heat (with a thermal threshold Ͼ 43°C) or protons (acidification), both of which cause pain in vivo (5)(6)(7)(8)(9). This sensitivity of VR1 to multiple noxious stimuli might explain certain properties of so called polymodal nociceptors. Furthermore, analyses of mice lacking VR1 have shown that VR1 is essential for selective modalities of pain sensation and for tissue injury-induced thermal hyperalgesia, further suggesting a critical role for VR1 in the detection or modulation of pain (10,11).
Tissue damage associated with infection, inflammation, or ischemia produces an array of chemical mediators that activate or sensitize nociceptor terminals to elicit or exacerbate pain at the site of injury in addition to the release of the mediators from the niciceptor terminals themselves known as neurogenic inflammation. An important component of this pro-algesic response, adenosine 5Ј-triphosphate (ATP), has recently been found to potentiate the VR1 currents evoked by capsaicin or protons through metabotropic P2Y 1 receptor activation in a protein kinase C (PKC)-dependent pathway (12). In the presence of extracellular ATP, the temperature threshold for VR1 activation was reduced from 42°C to 35°C, such that normally nonpainful thermal stimuli (i.e. normal body temperature) were capable of activating VR1 and thereby causing pain. Bradykinin has also been reported to enhance VR1 activity through a PKC-dependent pathway (13,14). These data suggest that direct phosphorylation of VR1 or a closely associated protein by PKC changes the agonist sensitivity of this ion channel.
It has to be addressed whether VR1 is directly phosphorylated by PKC and if so which amino acid residues are involved in the phosphorylation. For the purpose, we tried to examine the phosphorylation of VR1 expressed in human embryonic kidney-derived HEK293 cells biochemically and identified two Ser residues in the cytoplasmic domains of VR1, Ser 502 and Ser 800 .

EXPERIMENTAL PROCEDURES
In Vivo Phosphorylation-HEK293 cells were maintained in Dulbecco's modified Eagle's medium (supplemented with 10% fetal bovine serum, penicillin, streptmycin, and L-glutamine) and plated at 60 -70% confluence in 100-mm dishes, then transfected with 1 g of rat VR1 plasmid DNA using LipofectAMINE plus reagent (Invitrogen) as described previously (6). In vivo phosphorylation was confirmed as described previously (15). In brief, after transfection the cells were serumdeprived for 36 h in serum-free medium and then labeled with [ 32 P]orthophosphate (300 Ci/ml) for 3 h at 37°C. Following PMA (Sigma) stimulation (50 ng/ml) for 10 min at 37°C, the cells were washed with ice-cold phosphate-buffered saline and resuspended in TNE buffer (10 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, complete EDTA-free protease inhibitor mixture (Roche Molecular Biochemicals), * This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology in Japan and by The Japan Health Sciences Foundation, the Mochida Memorial Foundation, the Suzuken Memorial Foundation, the Mishima Kaiun Memorial Foundation, the Ichiro Kanehara Foundation, and the Naito Foundation (to M. T.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
ʈ To whom correspondence should be addressed: Dept. of Physiology, Mie University School of Medicine, Edobashi 2-174, Tsu, Mie 514-8507, Japan. Tel.:/Fax: 81-59-231-5004; E-mail: tominaga@doc.medic.mie-u. ac.jp. phosphatase inhibitor mixture (Sigma)). Samples were centrifuged for 15 min at 100,000 ϫ g. The pellets were resuspended in TNE buffer with 1% Nonidet P-40 and sonicated for 30 s. Following centrifugation at 100,000 ϫ g for 30 min, the supernatants were pre-cleared with protein A and then incubated at 4°C for 3 h with 1 g of rabbit anti-rat VR1 antibodies. Anti-rabbit IgG was added and incubated at 4°C for 1 h. Immunoprecipitated proteins were boiled in SDS sample buffer and separated by SDS-PAGE (8% polyacrylamide). The gel was exposed for autoradiography.
Anti-rat VR1 antibody was made as follows. A peptide encoding the predicted carboxyl terminus of VR1 (EDAEVFKDSMVPGEK) was coupled to keyhole limpet hemocyanin via an amino-terminal cystein and used to immunize rabbits.
Bacterial Expression of Glutathione S-Transferase (GST)-VR1 Fusion Proteins-Fusion proteins comprising GST at the amino terminus in-frame with amino (NH 2 )-terminal, the first intracellular loop and carboxyl (COOH)-terminal were generated by PCR and standard cloning techniques as described previously (16). The PCR products were subcloned into the pGEX vector (Amersham Biosciences). The final constracts were verified by sequencing. GST-VR1 protein was purified according to the manufacturer's manuscript (Amersham Biosciences).
In Vitro Kinase Assays-For the in vitro kinase assays, each purified fusion protein was incubated with the following reagents: 20 mM Tris-HCl, 0.01% Triton-X, 100 mM MgCl 2, 200 g/ml phosphatidylserine, 10 M ATP, 0.5 mM CaCl 2 , 0.25% bovine serum albumin, 0.5 mM dithioerythritol, and 1 Ci of [␥-32 P]ATP. The reaction, in a 50-l final volume, was started by adding 0.04 unit of PKC⑀ (Panvera) and incubated at 30°C for 15 min. Following the addition on a 10% SDSpolyacrylamide gel, the gel was exposed for autoradiography. The amount of used GST-VR1 proteins was analyzed by Coomassie Brilliant Blue (CBB) staining.
Mutagenesis-Point mutations were introduced by using oligonucleotide-directed mutagenesis. All constracts were verified by DNA sequencing. cDNAs were subcloned into pcDNA3 vector (Invitrogen).
Electrophysiology-Whole-cell patch clamp recordings were carried out at one or 2 days after transfection of VR1 cDNA to HEK293 cells as described previously (6). Standard bath solution contained 140 mM NaCl, 5 mM KCl, 2 mM MgCl 2 , 5 mM EGTA, 10 mM HEPES, 10 mM glucose, pH 7.4 (adjusted with NaOH). Pipette solution contained 140 mM KCl, 5 mM EGTA, 10 mM HEPES, pH 7.4 (adjusted with KOH). All patch clamp experiments were performed at room temperature (22°C) unless otherwise noted. When examining the heat-evoked current responses, bath temperature was increased using a preheated solution with the rate of 1-1.5°C/s. When the heat-activated currents started to inactivate, the heat solution was changed to a 22°C one. Chamber temperature was monitored (accuracy Ϯ0.1°C) with a thermocouple placed within 4 mm of the patch-clamped cell. The solutions containing drugs were applied to the chamber (180 l) by a gravity at a flow rate of 5 ml/min.

RESULTS AND DISCUSSION
To confirm the in vivo phosphorylation of VR1 by PKC, we used HEK293 cells expressing VR1 heterologously. Activation of PKC was achieved by incubating the transfected cells for 10 min with 50 ng/ml PMA, a potent and specific PKC activator. Following the treatment with [␥-32 P]ATP, the cells were stimulated with PMA. VR1 protein immunoprecipitated with antirat VR1 antibody showed more 32 P incorporation into VR1 upon PMA stimulation compared with the VR1 without PMA stimulation (Fig. 1A), indicating the direct phosphorylation of VR1 by PKC. There are 16 putative Ser or Thr residues that are candidate substrates for PKC-dependent phosphorylation in the VR1 NH 2 terminus, first intracellular loop, and COOH terminus (Fig. 1B). To distinguish among these possibilities, recombinant proteins carrying GST fused to the three segments of the cytoplasmic domains of VR1 were generated for use in an in vitro kinase assay. This assay demonstrated that the first intracellular loop and the COOH-terminal contained the substrates for PKC⑀ (Fig. 1C, panel b).
To identify the specific VR1 amino acids involved, eight Ser or Thr residues in the first intracellular loop and the COOHterminal were individually replaced with Ala, and the resulting mutant proteins were subjected to functional analysis using a whole-cell patch clamp technique. In voltage clamp experi-ments, a low dose of capsaicin (20 nM) evoked small inward currents in the HEK293 cells expressing VR1. In the absence of extracellular calcium, no change was observed in the magnitude of responses evoked by repetitive capsaicin applications. In contrast, after a 1-min pretreatment with 100 nM PMA, the same dose of capsaicin produced a much larger current responses (7.95 Ϯ 2.72 (means Ϯ S.E.)-fold, n ϭ 8) (Fig. 2A). Other electrophysiological properties of these capsaicin-evoked responses were unchanged by the presence of PMA (data not shown). Among the mutants tested, S502A and S800A showed significantly smaller potentiation of capsaicin-evoked current responses by PMA, although normalized currents after treatment of PMA varied in the eight mutants (2.13 Ϯ 0.41-fold, n ϭ 9 for S502A; 2.76 Ϯ 0.52-fold, n ϭ 11 for S800A) (p Ͻ 0.05) (Fig.  2, B, C, and E). In wild type VR1, PKC activation works by increasing the potency of capsaicin but not its efficacy (12). Therefore, we ruled out the possibility that the two mutants, S502A and S800A, already have a high affinity for capsaicin by examining that higher doses of capsaicin produced bigger current responses in the two mutants like wild type (data not shown). Patch clamp recordings in mutants bearing Ala substitutions of eight other Ser and Thr residues in the NH 2 terminus of VR1 provided no evidence for the involvement of these residues in PMA-mediated potentiation (9.18 Ϯ 4.05-fold, n ϭ 7; 14.7 Ϯ 4.58-fold, n ϭ 9; 10.4 Ϯ 4.26-fold, n ϭ 5; 7.20 Ϯ 1.50-fold, n ϭ 6; 9.85 Ϯ 2.37-fold, n ϭ 5; 7.32 Ϯ 3.77-fold, n ϭ 5; 14.6 Ϯ 5.39-fold, n ϭ 6 and 20.4 Ϯ 7.46-fold, n ϭ 9 for T42A, S93A, S139A, S153A, S185A, T322A, T329A, and S366A, respectively). Furthermore, double mutant S502A/S800A exhibited almost no PMA potentiation effect (0.95 Ϯ 0.04-fold, n ϭ 7) (p Ͻ 0.05) (Fig. 2, D and E), suggesting that these two Ser residues were the major substrates for PKC-dependent phosphorylation. Because ATP is a more physiological stimulus   (17), ATP was applied to cells expressing wild type, single (S502A and S800A) or double (S502A/S800A) VR1 mutants to confirm the involvement of these two Ser residues in VR1 potentiation. All of the S502A, S800A, and S502A/S800A mutants again showed no potentiation of VR1 currents evoked by 20 nM capsaicin upon pretreatment of 100 M ATP, whereas capsaicin-evoked currents were significantly potentiated by ATP in wild type VR1 (2.61 Ϯ 0.19-fold, n ϭ 7 for wild type; 1.05 Ϯ 0.24-fold, n ϭ 7 for S502A; 0.63 Ϯ 0.10-fold, n ϭ 6 for S800A; 0.79 Ϯ 0.12-fold, n ϭ 7 for S502A/S800A) (p Ͻ 0.05) (Fig. 3), indicating that these two Ser residues were phosphorylated in the physiological condition.
Of great physiological relevance is whether these mutants affect the response of VR1 to heat. Therefore, potentiating effects of PMA were examined on heat-evoked responses in HEK293 cells expressing wild type VR1 or S502A/S800A mutant. For this analysis, heat-evoked current responses were compared between different cells, rather than within the same cell, because repetitive heat-evoked currents show significant desensitization even in the absence of extracellular Ca 2ϩ (7) and because the thermal sensitivity of VR1 increases with repeated heat application (18). When temperature ramps were applied to HEK293 cells expressing wild type VR1, heat-evoked currents developed at about 42°C with an extremely steep temperature dependence (Fig. 4A). PMA (100 nM) treatment lowered the temperature threshold for wild type VR1 activation significantly (41.9 Ϯ 0.9°C, n ϭ 3 and 31.8 Ϯ 1.6°C, n ϭ 4, without and with PMA treatment, respectively, p Ͻ 0.01) (Fig.  4, A and C). On the other hand, HEK293 cells expressing the S502A/S800A mutant showed a little lower temperature threshold for activation without PMA treatment, although there was no significant difference between wild type and S502A/S800A mutant. However, no reduction of the threshold was observed in the mutant upon PMA treatment (38.0 Ϯ 1.4°C, n ϭ 7 and 37.4 Ϯ 0.7°C, n ϭ 9 without and with PMA treatment, respectively, p ϭ 0.7). These data further indicate the involvement of these two Ser residues in VR1 sensitization.
To further confirm that Ser 502 and Ser 800 function as substrates for PKC⑀-dependent phosphorylation, an in vitro kinase assay was carried out in those mutants. Phosphorylation was significantly reduced in both S502A and S800A mutants upon PKC⑀ application when the same amount of proteins were loaded (Fig. 1C, panels a and b). Some residual signals in both S502A and S800A mutants might suggest phosphorylation of other amino acids in the fusion proteins by PKC⑀. However, it is not likely that the phosphorylation in the mutants has significant meaning in terms of potentiation of VR1 currents by PKC, since the double mutant, S502A/S800A, showed no potentiation of VR1 currents evoked by capsaicin upon both PMA and ATP stimuli (Figs. 2 and 3) and no reduction of temperature threshold for activation upon PMA stimulus (Fig. 4).
One of the mechanisms underlying inflammatory pain is sensitization of ion channels expressed in nociceptor terminals such as VR1 (1-3, 5). Sensitization is triggered by extracellular inflammatory mediators, including ATP and bradykinin, re- . Cells were perfused for 1 min with solution containing ATP before exposure to capsaicin. Holding potential was Ϫ60 mV. C, effects of ATP on the capsaicin-activated currents in HEK293 cells expressing wild type and point mutants of VR1. Currents were normalized to the currents evoked initially by capsaicin (20 nM) before application of ATP, and the normalized values represent the means Ϯ S.E. Normalized currents were 2.61 Ϯ 0.19 (n ϭ 7), 1.05 Ϯ 0.24 (n ϭ 7), 0.63 Ϯ 0.10 (n ϭ 6) and 0.79 Ϯ 0.12 (n ϭ 7) for wild type, S502A, S800A and S502/S800A, respectively. *, p Ͻ 0.05 versus wild type; two-tailed unpaired t test.
Direct Phosphorylation of Capsaicin Receptor VR1 by PKC⑀ 13377 leased from surrounding damaged, inflamed, or ischemic tissues and from nociceptors themselves. Our data and those of others suggest that a system consisting of VR1 and certain metabotropic receptors exists that causes nociceptor sensitization by increasing VR1 sensitivity to noxious stimuli (12,13,17,19). In addition, a series of observations indicate that PKC plays an important role in this system (12)(13)(14)20). PKC⑀, among many PKC isoforms, has been reported to be predominantly and specifically involved in nociceptor sensitization (21)(22)(23). In the present study, direct in vivo phosphorylation of VR1 by PKC was proven for the first time, and PKC⑀ was found to phosphorylate two Ser residues. The replacement of these residues with Ala results in blunting of PKC⑀-mediated VR1 phosphorylation and a complete loss of VR1 potentiation or sensitization by ATP or PMA. Another isoform of PKC, PKC␥, has been shown to be pivotal for enhancing the sensation of pain in the spinal cord dorsal horn neurons (24,25). Therefore, two different isoforms of PKC, PKC⑀ and PKC␥, exhibit distinct roles at two levels of the pain pathway: the primary afferent neuron and the spinal cord dorsal horn, respectively. Our findings suggest that inhibitors of PKC (especially PKC⑀) as well as compounds acting at Ser 502 or Ser 800 of VR1 could prove useful in the treatment of pain by interfering with phosphorylation-mediated sensitization events.