Journal of Neurophysiology

Bradykinin Lowers the Threshold Temperature for Heat Activation of Vanilloid Receptor 1

Takeshi Sugiura, Makoto Tominaga, Hirotada Katsuya, Kazue Mizumura


Bradykinin (BK) is an inflammatory mediator that plays a pivotal role in pain and hyperalgesia to heat in inflamed tissues by exciting nociceptors and sensitizing them to heat through activation of protein kinase C (PKC). It has been suggested that the capsaicin receptor (VR1), a nociceptor-specific cation channel sensitive to noxious heat, protons, and capsaicin, is a channel that is modified by BK in these effects. In this study, we examined how BK modulates the activity of VR1. We measured VR1 currents using the patch-clamp technique in human embryonic kidney-derived (HEK293) cells expressing VR1 and B2 BK receptor. We found that BK lowered the threshold temperature for activation of VR1 currents in HEK cells down to well below the physiological body temperature in a concentration-dependent manner through PKC activation. We also demonstrated that in capsaicin-sensitive dorsal root ganglion (DRG) neurons the activation threshold of heat-induced current, which is considered to be VR-1 mediated, was lowered by BK and that this effect was also mediated by PKC. These data further support the supposition that modulation of VR1 is a mechanism for the BK-induced excitation of nociceptors and their sensitization to heat.


Bradykinin (BK), a nonapeptide released into inflamed tissues, induces pain and hyperalgesia to heat (Manning et al. 1991) when applied to human skin, and its antagonists alleviate inflammatory hyperalgesia (Steranka et al. 1988). These observations suggest a pivotal role of BK in inflammatory pain and hyperalgesia. BK is also known to induce excitation of nociceptors in many tissues (seeMizumura 1998 for review), as well as augment the magnitude of their response to heat and lower the threshold temperature for the response to heat (Koltzenburg et al. 1992;Kumazawa et al. 1991). The B2 BK receptor (B2R) has been implicated in BK-induced nociceptor activities and nociceptive behaviors (Dray and Perkins 1993; Mizumura 1998), and its cDNA has already been cloned (Ma et al. 1994). Evidence for the contribution of the protein kinase C (PKC) pathway to the B2 receptor-mediated algesic action of BK has accumulated (Cesare et al. 1999; Mizumura 1998).

The capsaicin receptor (VR1) is a nonselective cation channel sensitive to capsaicin as well as to heat and low pH (see Caterina and Julius 2001 for review). Localization of this channel in small dorsal root ganglion (DRG) neurons, likely to be soma of nociceptors, suggests its role as a heat transducer of nociceptors. Recently, a novel mechanism for VR1 modulation by ATP and BK through a PKC-dependent pathway was demonstrated (Premkumar and Ahern 2000; Tominaga et al. 2001; Vellani et al. 2001). Thus it is highly possible that VR1 is a channel that is modified by BK to sensitize and excite nociceptors. However, whether and to what extent BK lowers the threshold temperature of VR1 activation, and whether this modulation can be induced in the concentration range where nociceptors are sensitized and excited in tissues, is not yet clarified. In the present study, we investigated these issues using HEK293 cells expressing VR1 and B2R as well as in capsaicin-sensitive DRG neurons. We found that BK lowered the threshold temperature of heat-induced currents to below skin temperature in a concentration-dependent manner through PKC activation.

Preliminary results have appeared in abstract form (Sugiura et al. 2001).


Human embryonic kidney-derived (HEK293) cells were grown in Dulbecco's modified Eagle medium (DMEM, Gibco/BRL, Rockville, MD; supplemented with 10% fetal bovine serum, penicillin, streptomycin, and l-glutamine) under 95% O2 and 5% CO2 at 34°C. The rat VR1 cDNA in the pcDNA3 vector (Invitrogen, Carlsbad, CA) was a generous gift from Dr. David Julius (UCSF), and the mouse B2R cDNA in the pRc/CMV vector (Invitrogen) was a generous gift from Dr. Haruhiro Higashida (Kanazawa University, Japan). HEK293 cells were transfected with both VR1 and B2R plasmid DNAs (0.7 and 0.5 μg each) using LipofectAMINE PLUS Reagent (Gibco/BRL). Their expression was confirmed by fluorescence from simultaneously transfected GFP.

DRG neurons were obtained from male Wistar rats (80–150 g) according to the methods described in a previous paper (Kasai et al. 1998). Briefly, L4–L6 DRGs were dissected from a rat quickly killed by stunning the neck and treated with 0.2% collagenase (Wako, Osaka, Japan). Neurons were then mechanically dissociated on poly-L lysine-coated glass coverslips. They were incubated in DMEM with NGF (100 ng/ml, mouse 7S-NGF, pH 7.4; Sigma, St. Louis, MO) at 37°C. All the experimental procedures were approved by the Committee for Animal Experiments, Research Institute of Environmental Medicine, Nagoya University.

Whole-cell patch-clamp recording from HEK293 cells or DRG neurons was carried out with a holding potential of −60 mV at 2 days after transfection or cultivation using an Axopatch 200B amplifier (Axon Instruments, Foster City, CA). The standard bath solution contained the following (mM): 140 NaCl, 5 KCl, 2 MgCl2, 2 CaCl2, 10 HEPES, and 10 glucose (pH 7.4). The Ca-free solution was made by replacing CaCl2 with 5 mM EGTA in the standard solution. Electrodes were filled with pipette solution containing the following (mM): 140 CsCl, 5 EGTA, and 10 HEPES (pH 7.4), and had resistances of 3–5 MΩ. One hundred nanometers of TTX was added to the extracellular solutions, and 2 mM ATP and 0.2 mM guanosine 5′-triphosphate (GTP) were added to the pipette solution when recordings were made from DRG neurons. The chamber volume was about 0.3 ml, and extracellular solution was dropped by gravity at a flow rate of about 3 ml/min and was changed to other test solution by using a valve controller (VC-6; Warner Instrument, Hamden, CT). Thermal stimulation was applied by increasing the bath temperature at a rate of about 1.0°C/s with a preheated extracellular solution. When the heat-activated current started to inactivate, the heat solution was changed to the solution of the room temperature. The temperature of the solution was monitored with a thermocouple placed within approximately 100 μm of the patch-clamped cell (accuracy ±0.1°C).

BK was purchased from Peptide Institute (Osaka, Japan), phorbol 12-myristate 13-acetate (PMA) was purchased from Wako, Calphostin C and PKC ɛ translocation inhibitor peptide were purchased from Calbiochem (San Diego, CA), and TTX and 4-α-phorbol 12,13-didecanoate (4αPDD) were purchased from Sigma.


Lowering of the heat activation threshold of VR-1 in transfected HEK293 cells by BK

Heat-evoked inward currents (I heat) were compared between different cells in this experiment becauseI heat show significant desensitization on repetitive heat application even in the absence of extracellular Ca2+ (Tominaga et al. 1998) and because the thermal sensitivity of VR1 changes with repeated heat application (Caterina et al. 1999). When a heat ramp was applied to an HEK293 cell,I heat developed at about 42°C with an extremely steep temperature dependency (Fig.1 A). The heat threshold (Tt ) of VR1 activation was defined as the intersection where two lines approximating the stable baseline current and the clearly increasing temperature-dependent current cross (Fig.1 A, bottom). Tt thus measured was 41.9 ± 0.7°C (n = 11) in the absence of any treatment (Fig. 1, A and C).

Fig. 1.

Bradykinin (BK) lowered the threshold temperature (T t) of VR1 activation in HEK293 cells co-expressing VR1 and B2R. A and B(top): a sample recording of heat-evoked current (I heat); (middle): temperature change; (bottom): temperature-response profile up to the deactivation point.T t (arrow point) was defined as described in the text. T t of VR1 activation was lowered by BK (10 nM, gray bar). C: BK lowered T t of VR1 activation in a concentration-dependent manner. EC50 = 34.0 pM. Mean values ± SE. *P < 0.001 versus control (CTR); one-way analysis of variance (ANOVA) with Dunnet's multiple comparison test. n, number of neurons tested.

The effects of BK on I heat in VR1-expressing cells were examined by applying BK for 1 min before the heat stimulus. BK had no effects onI heat in five cells transfected with VR1 cDNA and the vector of B2R cDNA (Tt 41.4 ± 1.3°C). BK treatment, however, lowered theTt for VR1 activation in cells expressing both VR1 and B2R but it seldom modified the basal current in these cells (Fig. 1 B). The Tt lowering effect of BK was concentration-dependent and was observed at a concentration as low as 10 pM (Tt 38.5 ± 1.1°C, range 32.4–41.8°C, n = 9). It reached a plateau at 100 nM (30.6 ± 0.8°C, range 28.4–33.2°C, n = 4) (Fig. 1 C).

In some cases, especially when a higher concentration of BK was applied (in 2 of 11 cases with 100 nM and 11 of 20 cases with 1000 nM), a small current was observed at room temperature (23°C) (Fig.2 A, left). This current was partially blocked by capsazepine (Fig. 2 B) and was never observed when the BK solution was cooled down to 10°C (a sample in the middle panel of Fig. 2 A, n = 4 with 100 nM and n = 5 with 1000 nM). The incidence of this current by BK 1000 nM was significantly lower at 10°C (P < 0.01, Fisher's exact probability test). We confirmed that 1 μM capsaicin still induced inward currents in three cells at this temperature (Fig. 2 A, right), indicating that VR1 is functioning. We examined the current-voltage relationship of BK-induced current in three cases, and it turned out to have the reversal potential at about 0 mV (Fig.2 C) and outward rectification similar to the capsaicin-induced current, revealing a change in gating of the same channel. These results would suggest that this current was due to the lowered Tt of VR1 activation.

Fig. 2.

BK occasionally elicited VR1 current at room temperature in transfected HEK293 cells. A: sample recording of BK-induced current (left) at room temperature and its absence at lower temperature (10°C, middle). BK, 100 nM (gray bar). Capsaicin (1 μM) still elicited inward current at 10°C (right), demonstrating that VR1 is functional at this temperature. B: capsazepine (10 μM, black bar) partially inhibited BK (1 μM, gray bar)-induced current at 23°C.C: current-voltage (I-V) relationship during the responses induced by BK (1 μM) and capsaicin (1 μM) at room temperature. Voltage-ramp is indicated at the left. Note thatI-V relationships of responses to capsaicin and BK application reveal a change in gating of the same channel.

A role of PKC in the BK-induced sensitization of nociceptors to heat has been suggested, and we therefore examined the effects of PKC-activating PMA. A recent report has suggested that PMA may directly activate VR1 as well as PKC (Chuang et al. 2001); therefore, we used PMA at 0.1 μM, a concentration that is lower than that of PMA has this effect. The inward current was seldom increased during application of PMA alone (Fig.3 A), yet PMA significantly reduced the Tt for VR1 activation (32.8 ± 0.7°C, n = 11, P < 0.0001 compared with the control response without any treatment) (Fig.3 A, right). Inactive phorbol ester, 4-αPDD, at 0.1 μM had no effect (Tt 40.4 ± 0.8°C, n = 4, Fig. 3 A).

Fig. 3.

PKC activator lowered the T t of VR1 activation (A), and PKC inhibitors completely abolished the sensitization of VR1 induced by BK (B) in transfected cells. Left: sample recording.Right: summary graph (mean ± SE). Phorbol 12-myristate 13-acetate (PMA): 100 nM, gray bar; 4-α-phorbol 12,13-didecanoate (4αPDD): 100 nM; BK: 10 nM, gray bar; calpC: calphostin C, 1 μM, black bar; PKCɛInh: PKC-ɛ translocation inhibitor peptide 200 μM (sample recording is not shown). Recording was started ≥3 min after obtaining a whole-cell configuration. n.s., not significantly different vs. CTR. *P < 0.001 vs. CTR. #P < 0.001 vs. BK; one-way ANOVA with Bonferroni's multiple comparison test. n, number of neurons tested.

We further examined whether BK-induced lowering ofTt is mediated through PKC, by applying a selective PKC inhibitor, calphostin C. Using the pipette solution with calphostin C (1 μM), the effect of 10 nM BK was completely blocked (Tt was 41.5 ± 0.8°C,n = 7, n.s. compared with the control) (Fig.3 B). A PKC ɛ isoform has been implicated in the effect of BK in DRG neurons (Cesare et al. 1999). To confirm this, a PKC ɛ translocation inhibitor peptide (200 μM) (Johnson et al. 1996) was added to the pipette solution. Again, the effect of 10 nM BK was completely abolished, withTt here (41.8 ± 1.2°C,n = 5) not different from the controlTt (n.s., Fig. 3 B). These results demonstrate that BK lowers the heat-activation threshold of VR1 in a PKC-dependent manner.

Activation threshold of Iheat in capsaicin-sensitive DRG neurons was also lowered by BK

I heat were measured in DRG neurons with diameter <30 μm. Capsaicin sensitivity was examined either before heat stimulation in the absence of extracellular calcium or after heat application to avoid desensitization of VR1 (Tominaga et al. 1998). The latter procedure might have lowered the ratio of capsaicin-sensitive DRG neurons in the control group, because the neurons in this group must have been heated >50°C (Fig. 4 C).

Fig. 4.

BK reduced the threshold temperature ofI heat in capsaicin-sensitive dorsal root ganglion (DRG) neurons. A: sample recordings ofI heat obtained from a DRG neuron (left). The sensitivity to capsaicin was examined after a heat experiment (right). B: temperature-response profile. T t(arrow point) was reduced by BK (100 nM) pretreatment. Data were obtained from two different neurons. C: summary of the effects of BK and BK in the presence of calphostin C onT t in DRG neurons. Triangle: neuron in which capsaicin (1 μM) was not examined; solid circle: neuron that was capsaicin-sensitive; open circle: neuron that did not respond to capsaicin. BK10/clapC: BK 10 nM in the presence of calphostin C (1 μM).

Similar to HEK 293 cells expressing VR1 and B2R,I heat of capsaicin-sensitive DRG neurons steeply increased above around 40°C in the absence of BK (a sample in Fig. 4 B, left). The distribution ofTt is summarized in Fig. 4 C. The averageTt was 41.2 ± 0.5°C (n = 5) in capsaicin-sensitive neurons without any treatment [control (CTR) Fig. 4 C]. BK at 10 and 100 nM lowered Tt ofI heat (Fig. 4, A andC) in most cases, with considerable variation at 10 nM.Tt of one capsaicin-sensitive neuron with BK of 10 nM was in the same range as that of the control. BK might have only slightly lowered Tt of this neuron, or the neuron might not have expressed B2R. There were several neurons with Tt at around 50°C in the control as well as in the presence of BK. The averageTt of capsaicin-sensitive neurons was 32.2 ± 2.4°C (n = 6) and 24.0 ± 2.3°C (n = 5) with 10 and 100 nM BK, respectively, significantly different values from the control [P < 0.001, one-way analysis of variance (ANOVA) with Dunnet's multiple comparison test]. Thus BK lowered Tt of capsaicin-sensitive DRG neurons in a concentration-dependent manner. BK (10 nM)-induced lowering of Tt of capsaicin-sensitive DRG neurons was completely blocked by calphostin C (Tt 43.7 ± 0.5°C,n = 5, n.s. compared with theTt of the control).

BK at a high concentration (1 μM) sometimes induced inward currents in DRG neurons at room temperature (data not shown). These currents were partly suppressed by capsazepin, suggesting them to be VR1 mediated similar to the phenomenon observed in the HEK293 cells expressing VR1.


One of the major findings of the present study is thatTt ofI heat in HEK cells expressing VR1 and B2R was lowered by activation of B2R and that this effect was dependent on the concentration of BK. The minimum effective concentration of BK was as low as 10 pM. This concentration is a little lower than what is reported for BK-induced sensitization to heat in visceral nociceptors (Kumazawa et al. 1991) and close to the concentration reported for the inflammatory exudates (see Mizumura 1998).

Similar to this observation in HEK 293 cells, we observed lowering ofTt ofI heat by BK at concentrations of 10 and 100 nM in capsaicin-sensitive DRG neurons. These currents appeared to be mediated by VR1 because they were observed in capsaicin–sensitive DRG neurons with diameters <30 μm (Tominaga et al. 1998). The magnitude of the decrease inTt (down to 24°C with 100 nM BK) was greater than in HEK 293 cells. This might be related to ATP (Kwak et al. 2000) and/or GTP in the pipette solution used for recording from DRG neurons. The present observations strongly suggest that VR1 can be activated by BK in pathophysiological conditions. Methodological limitation in this experiment did not allow us to examine whether the maximal heat-current was also increased, butI heat has been reported to be increased by BK in a certain temperature range with a short heat pulse (Vellani et al. 2001). Presently observed lowering ofTt together with augmentation ofI heat contribute to the BK-induced sensitization of nociceptors to heat, and in turn to hyperalgesia, to heat in inflamed tissues.

Neurons having I heat with higherTt (around 50°C) existed in both the presence and the absence of BK. Nagy and Rang (1999)reported two groups of heat-sensitive neurons with different activation threshold: one group has threshold at around 45°C, and the other group has threshold at around 51°C. The former is supposed to have VR1 and the latter VRL-1 (Caterina et al. 1999). So far no DRG neurons with higher heat activation threshold have been reported. Thus it is highly possible that presently observed neurons with Tt at around 50°C have VRL-1. The small number of observations do not allow us to make any solid conclusions; however, the present observation might raise a possibility that VRL-1 does not undergo modulation by BK.

A second finding is that the Tt lowering effect of BK is mediated by a PKC, specifically by PKC ɛ. Similar PKC-dependent activation of VR1 through P2Y receptor activation has been reported (Tominaga et al. 2001), although no isoform was implicated in that report. The present observation supports a previous report on DRG neurons (Cesare et al. 1999).

The third interesting finding of the present study is that theTt of VR1 activation is lowered by BK down to 30.6°C in HEK293 cells and 24.0°C in capsaicin-sensitive DRG neurons (with 100 nM BK), which is well below the physiological body temperature and even below the range of the normal skin temperature. The present observation suggests that the BK-induced currents reported by Premkumar and Ahern (2000) and by Chuang et al. (2001) at room temperature were also the result of a decrease in the threshold of VR1. The presently observed lowering of the Tt of VR1 activation below the body temperature implies that BK-induced excitation of nociceptors may be the result of sensitization of VR1. Whether VR1 is the sole ion channel that BK modulates to induce excitation in nociceptors is yet to be determined.

PLC-mediated hydrolysis of phosphatidylinositol-4,5-bisphosphate (PIP2) is the essential step after binding of BK with B2R before PKC activation. Recently, Chuang et al. (2001) reported that PLC activation releases VR1 from PIP2-mediated inhibition. The release from PIP2blockade and PKC-mediated activation of VR1 by BK might work together to bring about a greater change in VR1 activities in nociceptor neuron.


This work was partly supported by a grant from the Ministry of Education, Culture, Sports, Science, and Technology in Japan (to K. Mizumura).


  • Address for reprint requests: K. Mizumura, Dept. of Neural Regulation, Research Institute of Environmental Medicine, Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8601, Japan (E-mail:mizu{at}


View Abstract