The recent discovery that camphor activates and strongly desensitizes the capsaicin-sensitive and noxious heat-sensitive channel transient receptor potential vanilloid subfamily member 1 (TRPV1) has provided new insights and opened up new research paths toward understanding why this naturally occurring monoterpene is widely used in human medicine for its local counter-irritant, antipruritic, and anesthetic properties. However, the molecular basis for camphor sensitivity remains mostly unknown. The authors attempt to explore the nature of the activation pathways evoked by camphor and narrow down a putative interaction site at TRPV1.
The authors transiently expressed wild-type or specifically mutated recombinant TRPV1 channels in human embryonic kidney cells HEK293T and recorded cation currents with the whole cell, patch clamp technique. To monitor changes in the spatial distribution of phosphatidylinositol 4,5-bisphosphate, they used fluorescence resonance energy transfer measurements from cells transfected with the fluorescent protein-tagged pleckstrin homology domains of phospholipase C.
The results revealed that camphor modulates TRPV1 channel through the outer pore helix domain by affecting its overall gating equilibrium. In addition, camphor, which generally is known to decrease the fluidity of cell plasma membranes, may also regulate the activity of TRPV1 by inducing changes in the spatial distribution of phosphatidylinositol-4,5-bisphosphate on the inner leaflet of the plasma membrane.
The findings of this study provide novel insights into the structural basis for the modulation of TRPV1 channel by camphor and may provide an explanation for the mechanism by which camphor modulates thermal sensation in vivo.
What We Already Know about This Topic
Transient receptor potential vanilloid (TRPV) channels are key participants in thermal and inflammatory pain
The molecular mechanism by which the analgesic agent camphor activates the TRPV1 channel is unknown
What This Article Tells Us That Is New
Camphor modulates TRPV1 by altering the short helical segment within the permeation pore as well as the spatial distribution of lipids on the inner membrane leaflet
These mechanisms are relevant to modulation by camphor of TRPV1 and other TRP family members involved in pain and thermal sensation
CAMPHOR is an organic waxy substance used topically for its counterirritant and analgesic properties.1,2This compound is one of a group of naturally occurring monoterpenes that are known to modulate the activity of various ion channels3,4and, particularly, the thermosensitive ion channels from the transient receptor potential (TRP) channel family expressed in skin and neural tissues.5,–,8The recently proposed mechanism by which camphor can modulate pain is through the activation and strong subsequent desensitization of the TRP vanilloid subfamily member 1 (TRPV1).7,8This channel is most abundantly expressed in small-diameter primary sensory neurons, where it functions as a receptor for capsaicin and as a transducer of noxious thermal stimuli (more than 43°C).9,–,11In addition to a range of vanilloids that bind to the receptor at the same site as capsaicin, such as olvanil, piperine, arvanil, resiniferatoxin, and various endogenous substances produced during inflammatory processes (reviewed by Starowicz et al. 12), TRPV1 can be modulated by protons (pH less than 6.8), which directly activate and potentiate the channel from the extracellular side13,–,15; temperatures greater than 43°C9; and strongly depolarizing voltages (V 50∼+150 mV).16,17Camphor acts as a partial agonist of TRPV1 but has a much lower potency (EC50>3 mM) than capsaicin (EC50∼ 0.3 μM). However, these millimolar concentrations are physiologically relevant because camphor is used efficiently in various ointments at a proportion of approximately 10% by weight. In addition, its strong synergy with other TRPV1 modalities could make this compound a powerful ally for the management of various pain and itch states. Camphor also activates two other mammalian thermosensitive TRP channels, both expressed in primary afferent neurons or keratinocytes: heat-activated TRPV318and cold-activated transient receptor potential melastatin 8 (TRPM8) channel.6,7It also inhibits transient receptor potential ankyrin 1 (TRPA1) channel, another cold-sensitive member of the TRP channel family.8Thus, given that camphor recently has been shown to activate an insect isoform of TRPA1,19it is likely that the molecular determinants by which thermosensitive TRP channels sense camphor are evolutionarily conserved.
Camphor potentiates heat-evoked responses and shifts the voltage-dependence of TRPV1 activation to more negative voltages.8Although strong evidence points toward camphor binding directly to the TRPV1 receptor,8the activation mechanism has not been identified. Camphor activates TRPV1 independently of the vanilloid-binding site (the S2-S4 binding module) and desensitizes the channel more rapidly and completely than does capsaicin.8In a striking contrast to capsaicin, the camphor-mediated desensitization appears to be independent of the presence of external Ca2+.8,20Another important aspect of TRPV1 activation is the apparent weak ability of camphor to alter dynamically TRPV1 permeability to large cations.20Upon capsaicin stimulation, the channel undergoes pore dilation, in which its selectivity for large cations over sodium ions is increased. It has been proposed that, upon activation, camphor may preclude or mask conformational events that change the selectivity filter of the channel. This is in contrast to capsaicin, which induces a concentration-, time-, and Ca2+-dependent functional increase in TRPV1 pore diameter.
Because understanding the principal differences in the properties of capsaicin- and camphor-dependent activation and desensitization could aid in the understanding of the underlying mechanisms, we set out to compare the effects of camphor and capsaicin on the activity of TRPV1. We identify structural elements in the TRPV1 protein that are critical for camphor agonism and demonstrate important aspects in the way camphor might interact with TRPV1.
Materials and Methods
Expression and Constructs
HEK293T cells (American Tissue Cell Collection, Rockville, MD) were cultured in OPTI-MEM I medium (Life Technologies; Darmstadt, Germany) supplemented with 5% fetal calf serum as described previously.21,22Cells were cotransfected transiently with 300–400 ng complementary DNA plasmid encoding wild-type or mutant rat TRPV1 and with 200 ng green fluorescent protein plasmid (TaKaRa; Tokyo, Japan) per 1.6-mm dish using the magnet-assisted transfection (IBA; Göttingen, Germany) method. Standard molecular biologic techniques were as described.23The chimeric (Δ15-TRPV1: Y627-C634) and the deletion (Δ15-TRPV1) constructs of TRPV1 were provided by Feng Qin (PhD., Associate Professor, Department of Physiology and Biophysics, State University of New York at Buffalo, Buffalo, New York). For fluorescence resonance energy transfer (FRET) measurements, HEK293T cells were transfected with cyan (CFP)- and yellow fluorescent protein (YFP)-tagged pleckstrin homology (PH) domains of PLCδ1 (provided by Tamas Balla, D.Sc., Senior Investigator, Section Molecular & Signal Transduction, National Institute of Child Health & Human Development, National Institutes of Health, Bethesda, Maryland). Cells were used 24–48 h after transfection. At least four independent transfections were used for each experimental group. The wild-type channel was regularly tested in the same batch as the mutants.
Standard whole cell electrophysiology techniques were used as described elsewhere.23Experiments were performed at room temperature (23°–25°C). Only one recording was performed on any one coverslip of cells to ensure that recordings were made from cells not previously exposed to chemical stimuli. A system for rapid superfusion of the cultured cells was used for drug application.24The extracellular control solution contained: 160 mM NaCl, 2.5 mM KCl, 1 mM CaCl2, 2 mM MgCl2, 10 mM HEPES, 10 mM glucose; adjusted to pH 7.3 and 320 mOsm. In whole cell, patch clamp experiments, the pipette/intracellular solution contained: 125 mM Cs-gluconate, 15 mM CsCl, 5 mM EGTA, 10 mM HEPES, 0.5 mM CaCl2, 2 mM Mg-adenosine triphosphate; pH 7.3, 286 mOsm. In experiments with dioctanoyl-phosphatidylinositol-4,5-bisphosphate (DiC8-PIP2), adenosine triphosphate- and Mg2+-free intracellular solution was used. Camphor ((1R)-(+)-Camphor, (+)-Camphor, (1R)-1, 7,7-Trimethylbicyclo[2.2.1]heptan-2-one) solution was prepared from a 1-M stock solution in dimethyl sulfoxide stored at −20°C. All chemicals were purchased from Sigma–Aldrich (Prague, Czech Republic).
For FRET measurements, we used the Cell^R imaging system based on an Olympus IX-81 inverted microscope (Olympus, Tokyo, Japan) equipped with a dual-emission setup (Dual-View Optical Insights, LLC; Santa Fe, NM). The excitation wavelength was 436 nm, and emission was detected in parallel at 470 and 535 nm. The excitation light was generated with a Polychrome V polychromator (Till Photonics, Gräfelfing, Germany), and the fluorescence emission was detected with a Hamamatsu Orca-ER camera (Hamamatsu Photonics, Hamamatsu City, Japan). Data were collected using Cell^R software (Olympus), and the fluorescence analysis was done using the program ImageJ (National Institutes of Health, Bethesda, MD).
Automated Ligand Docking
The homology models of the rat25and human26TRPV1 channels were used for automated ligand docking. To perform the molecular docking calculations, we used AutoDock Vina software 1.0.2,27and the input files for both protein and ligand were prepared using AutoDock Tools version 4.228(The Scripps Research Institute, La Jolla, CA). Docking was performed by using a grid box 70 × 70 × 70 Å on each protein subunit or by using a grid box 110 × 110 × 80 Å centered on entire tetramer. Polar hydrogen atoms were added to the protein, and Gasteiger partial charges were calculated using AutoDock Tools. Each docking consisted of 30 runs using the default AutoDock Vina parameters, except the options for maximum number of binding modes20and an exhaustiveness (accuracy) option of 16. In AutoDock Vina, two variants of root-mean-square deviation (RMSD) metrics are provided: RMSD lower bound (matches each atom in one conformation [c1] with the closest atom of the same element type in the other [c2] conformation: max [RMSD(c1, c2), RMSD(c2, c1)]) and RMSD upper bound (matches each atom in one conformation with itself in the other conformation, ignoring any symmetry). Based on the affinities, ranging from less than 2 Å for the lower bound RMSD and less than 3 Å for the upper bound RSMD, two distinct low-energy binding modes were identified.
All data were analyzed using pCLAMP 10 (Molecular Devices GmbH, Ismaning, Germany), and curve fitting and statistical analyses were done in SigmaPlot 10 and SigmaStat 3.5 (Systat Software Inc., Chicago, IL). Significance levels were determined by a two-tailed Student independent t test. Data comparisons of three or more groups were performed by one-way ANOVA followed by Dunnett's post hoc comparison. Correlations were determined with the Spearman rank order correlation test. Differences were considered significant at P < 0.05, unless stated otherwise. Conductance-voltage (G–V ) relationships were obtained from steady-state whole cell currents measured at the end of voltage steps from −80 to +200 mV in increments of +20 mV. Exponential current decays were fitted using the Chebyshev algorithm in pCLAMP 10 software (Molecular Devices, Sunnyvale, CA). All results are presented as mean ± SD.
Camphor-induced Currents Exhibit Fast Kinetics, and Its Potentiating Effects Are Readily Reversible
The activation properties of TRPV1 channels in response to camphor were first examined by recording whole cell currents evoked by camphor at concentrations from 1 mM to 10 mM. As was reported in a previous study,8camphor activated TRPV1 in a dose-dependent manner, with apparent EC50values of approximately 5 mM at +70 mV (fig. 1), reaching a peak of 23 ± 22% and 71 ± 22% of the maximal response to 10 μM capsaicin when measured at −70 mV and +70 mV, respectively (n = 4). We found that the inward currents elicited by 10 mM camphor exhibited acute irreversible desensitization within 15 s of exposure, with a half decay time (T 50) of 3.7 ± 1.6 s at −70 mV (n = 27) and 2.5 ± 1.0 s at +40 mV (n = 3; P = 0.198). The extent of acute desensitization was independent of the peak amplitude (145 ± 66 pA/pF) for inward currents evoked by 10 mM camphor (P = 0.76; r = −0.08; n = 17; Spearman rank order correlation). This is in contrast to capsaicin-induced acute desensitization, which depends on Ca2+influx and is dependent on current size, with large currents exhibiting more pronounced desensitization.29,30The onset rates of camphor activation were characterized by a 10–90% rise-time of 1.7 ± 1.0 s, and the deactivation currents after the removal of 10 mM camphor exhibited a time constant of 230 ± 92 ms (n = 26). Initial responses to 10 mM camphor frequently exhibited a rapid onset followed by a subsequent hump (see fig. 1A), presumably reflecting a kinetic component of the activation process faster than the solution exchange of the system (∼130 ms).
The efficacy of camphor dramatically increases if other modalities, such as heat or proalgesic agents, act on the TRPV1 channel synergistically.8To further explore to what extent distinct activation pathways synergize to promote TRPV1 channel opening, camphor was applied while TRPV1 was activated by capsaicin, heat, or voltage. In the presence of 1–10 μM capsaicin, camphor potentiated TRPV1 in an activation-dependent manner, being more effective on the opened than the resting state of the channel. Figure 2illustrates that camphor, in combination with capsaicin, repeatedly potentiated TRPV1 channels, and this potentiation reached the maximum activation capacity of the capsaicin-occupied form but not of the closed (fig. 2A, i ) or camphor-occupied (fig. 2D) form of the channels. At lower camphor concentrations (5 mM), such responses underwent a reversible acute desensitization within 10 s of exposure (fig. 2A, e , f ), which might indicate that the camphor-dependent desensitization mechanism is functionally preserved. In addition, we found that the channels that previously had been almost fully desensitized by camphor and/or capsaicin could be repetitively potentiated to their close-to-maximum activation capacity by a combination of capsaicin with 5 or 10 mM camphor (fig. 2, A–D). Again, the onset rates of the camphor-mediated potentiation were fast, with a 10–90% rise-time of 0.8 ± 0.8 s (n = 7), and a deactivation time constant of 0.35 ± 0.19 s, apparently being limited by the solution exchange time. Camphor dramatically intensifies the tachyphylaxis of TRPV1 responses to capsaicin and low pH.8However, what we found was that 10 mM camphor applied for 10–15 s did not have any impact on the subsequent response induced by 1 μM capsaicin 30 s later because this response was of maximal amplitude, as can be inferred from that induced by 10 μM capsaicin an additional 30 s later (fig. 2E). Thermal stimuli and depolarizing voltages also strongly potentiated camphor-evoked currents (fig. 3), as has been reported previously.8In addition, we observed that camphor shifted the threshold for heat activation (from 42.1 ± 0.9°C to 31.5 ± 5.4°C; n = 5; fig. 3A), and increasing the temperature above 42°C speeded up camphor-mediated acute desensitization (fig. 3B).
The results from these experiments suggest that long-lasting camphor-induced conformational changes are not propagated to the capsaicin activation pathway. The fast kinetics of the camphor-dependent activation and deactivation, and almost complete repeatability of the camphor-mediated potentiation of capsaicin responses support the hypothesis that camphor activates TRPV1 independently of the capsaicin binding/gating motif most likely by directly acting on the channel protein.
The Outer Pore Region Is Involved in Camphor-mediated Effects
The putative camphor-interacting site has been shown to lie outside the transmembrane 2–4 region (capsaicin binding module), so the outer pore region between transmembrane domains 5 and 6 appears to be one of the likely candidates.8This region plays a general role in TRPV1 channel gating31and is also a key regulatory domain important for activation and potentiation by protons.15,32We reasoned that if camphor acts on the channel by inducing perturbations in the outer pore region, then it might occlude the modulatory effects of protons. Thus, we compared responses to camphor at pH 6.8 with those measured at pH 7.3. Figures 3D and 3E illustrate that the whole cell currents elicited by camphor at pH 6.8 did not exceed the amplitude of the initial control response to 10 mM camphor obtained at pH 7.3 and did not alter the time course of tachyphylaxis. This is in contrast to capsaicin and heat, whose efficacies are strongly increased at acidic pH, and also in contrast to the potentiating effects of camphor on other activating stimuli.
We also examined the possibility that camphor acts at the TRPV1 channel by inducing conformational changes in the hydrophobic pore helix domain (L630-F640; fig. 4). This domain critically affects the balance between the open and closed states of the channel and is essential for low-pH activation and potentiation but not for capsaicin activation.15,31Within the pore helix, a single residue, T633, is crucial for direct activation by acidic pH.15We found that camphor-evoked currents through the T633A mutant were statistically significantly smaller than those induced in wild-type channels (fig. 4, A–C), reaching a peak of only 6.1 ± 3.2% of the maximal response to 10 μM capsaicin when measured at −70 mV (n = 7). The camphor responses were transient, exhibiting a more rapid and pronounced acute desensitization than the wild-type channel (T 50of 2.2 ± 1.3 s, n = 11; P = 0.008).
The results from these experiments identified T633 as a residue involved in camphor activation and suggest that the activation pathways are partly shared between camphor and protons. In an attempt to narrow the critical region for camphor modulation, we tested camphor sensitivity in a TRPV1 chimera (Δ15-TRPV1: Y627-C634) that lacked the stretch of 15 nonconserved residues between the turret and selectivity filter (T612-S626) and in which the pore helix (Y627-C634) was replaced with the counterpart from TRPV2, a camphor-insensitive homolog. As described previously,15this construct exhibited a markedly slower onset and offset of capsaicin responses. We found that, in contrast to T633A, the pore-helix chimera was completely insensitive to depolarizing voltages to +200 mV (fig. 4D). Most importantly, in this chimera, 10 mM camphor neither induced measurable currents nor potentiated capsaicin-induced currents (fig. 4E). The control deletion mutant with a minimal pore lacking the 15-residue stretch had the same sensitivity to camphor as the wild type (fig. 4F). These results strongly indicate that the pore helix mediates the camphor sensitivity of TRPV1.
The proton activation of TRPV1 involves several other sites, including the S3-S4 linker outside the pore domain, and this region is critical for additional stimulation of fully liganded human but not rat TRPV1 channel by protons.32In addition, specific mutation V538L in rat TRPV1 has been shown previously to abrogate the proton-evoked currents while preserving capsaicin and heat responses and their potentiation by mildly acidic pH.15To explore the extent of the overlapping of the activation pathways between camphor and low pH, we also tested the sensitivity to camphor in the human TRPV1 ortholog and in the mutant of rat TRPV1 V538L. We found that human TRPV1 responds poorly to 10 mM camphor (fig. 4G), inducing only 1.3 ± 1.5% of the peak currents induced by 10 μM capsaicin (n = 9). The capsaicin-induced desensitized currents were still robustly potentiated by camphor, which is in apparent contrast with observations made on the chimera Δ15-TRPV1: Y627-C634. The V538L mutant exhibited wild-type sensitivity to camphor, inducing 27 ± 16% of the peak current amplitudes induced by 10 μM capsaicin (n = 6; average data shown in fig. 4C). Together, these results indicate that the camphor and proton activation pathways do not overlap completely and suggest the existence of significant species difference with respect to camphor sensitivity.
Recovery Mutations Are Not Sufficient to Improve Camphor Sensitivity
Camphor acts as a full agonist of the TRPV3 channel, whose pore helix differs in only four residues from TRPV2 and in six residues from TRPV1 (fig. 4A). Thus, we were curious to know whether the TRPV1-to-TRPV3 mutations at the structurally most dissimilar residues of this region, N628G and L630F, might enhance the sensitivity to camphor in TRPV1. The N628G mutant could be activated normally by 10 mM camphor, just like the wild type; however, it exhibited statistically significantly smaller heat-induced responses, which was expected based on the proposed specific role of this residue in the heat-dependent gating of TRPV1.33–34The currents evoked by heat were only mildly potentiated by camphor (fig. 4H). The second mutant, L630F, was normal in all aspects of TRPV1 activation. Next, we measured responses to camphor for two recovery mutations of the chimeric channel, D632S and Y631L. Both recovery mutants resembled the chimeric phenotype (data not shown). Thus, the data from these experiments indicate that the pore helix plays a key role in transducing the camphor signal to gate opening and that mutations in this specific region selectively uncouple camphor activation. T633 seems to play an important part in this process, similar to that previously reported for the de novo proton activation mechanism.15
Camphor Increases FRET between the CFP- and YFP-tagged PH Domains
The selective loss of camphor sensitivity in the pore-helix chimera and the reversible and fast kinetics of camphor modulation appear to be consistent with a direct effect of camphor on TRPV1, rather than a secondary effect through a membrane-delimited signaling process. However, given the substantial lipophilicity of camphor and its ability to partition into membranes, we cannot exclude the possibility that camphor activation pathway is initiated through a “nonspecific” effect on the plasma membrane. Among these effects, a decrease in the phosphatidylinositol-4,5-bisphosphate (PIP2) concentration on the inner leaflet of the plasma membrane might underlie the long-lasting desensitization of the TRPV1 channel due to agonist-induced calcium influx.35,36On the other hand, increases in local PIP2might potentiate or directly activate the channel. Thus, as a next step, we attempted to explore the general effects of camphor on membrane-associated events that might be related to the activation of TRPV1.
Calcium influx through TRPV1 channels activates Ca2+-dependent isoforms of phospholipase C (PLC) that catalyzes the hydrolysis of membrane-bound PIP2to two second messengers: inositol triphosphate and diacylglycerol.37These changes can be detected in real time by FRET between the fluorescent protein-tagged PLC PH domains (PLCδ1PH-CFP and PLCδ1PH-YFP), whose translocation into the cytosol reflects increased PLC activation and PIP2hydrolysis.38,–,42To examine the effects of camphor on the PLC-dependent cascades, HEK293T cells were first transiently transfected with PLCδ1PH-CFP and PLCδ1PH-YFP at a molar ratio of 1:1, and assayed for FRET by simultaneously monitoring the emission of CFP (475 ± 15 nm) and YFP (530 ± 20 nm) while exciting CFP at 425 ± 5 nm. Upon adding camphor, the donor (CFP) emission intensity statistically significantly decreased, whereas the acceptor (YFP) emission increased (fig. 5). This effect was concentration-dependent, rapidly reversible on return to extracellular control solution, and was not observed upon stimulation with either dimethyl sulfoxide (1%), capsaicin (1 μM or 10 μM; fig. 5C), or hypertonic solution (to 410 mOsm, adjusted with sucrose; fig. 5D).
Because several effects may contribute to changes in the fluorescent properties of PLCδ1PH chimeric proteins (e.g. , concentration changes or hydrophobicity), we also explored the effects of capsaicin and camphor on cells that were transfected with either PLCδ1PH-CFP or PLCδ1PH-YFP alone (fig. 5, E and F). Upon capsaicin or camphor stimulation, only a small decrease in fluorescence (less than ∼ 0.8%) was observed in the PLCδ1PH-YFP–expressing cells. This effect resembles the changes induced by bradykinin in N1E-115 neuroblastoma cells reported by van der Wal et al. 40and could be explained analogously by the alterations in the local microenvironment influencing the spectral properties of the fluorescent proteins.
Activation of TRPV1 Increases Camphor-induced Effects on FRET between Tagged PH Domains
As an additional step to analyze the camphor-induced changes in PIP2-related processes, HEK293T cells were cotransfected to express rat TRPV1 together with PLCδ1PH-CFP and PLCδ1PH-YFP (fig. 6). In these cells, we again observed a statistically significant increase in the FRET signal upon 15–20 s of exposure to 10 mM camphor and a slower and delayed decrease in FRET in response to 1 μM or 10 μM capsaicin (fig. 6A, c ). This latter, slower process exhibited a time course corresponding to the TRPV1-mediated, Ca2+-dependent hydrolysis of PIP2and was followed by a slow and incomplete recovery upon washing, most likely corresponding to the replenishment of PIP2levels through the action of phosphatidylinositol phosphate kinases.35,36,43In contrast, the camphor-induced increases in FRET, which were found to be independent of TRPV1, were rapidly reversible upon washout (τoff= 0.5 ± 0.4 s; n = 16), indicating that the PLC-mediated hydrolysis of PIP2is unlikely to be involved. Moreover, the camphor-induced increase in FRET was more pronounced after capsaicin-induced PIP2depletion, as is demonstrated in the representative recording (fig. 6B, a and b ), in which we concurrently measured whole cell inward currents and FRET signals from a HEK293T cell expressing tagged PH domains together with TRPV1.
Inhibition of PLC Is Not Likely to Be the Mechanism Underlying the Modulation of TRPV1 by Camphor
Our results described here could be explained if the effective distances between the tagged fluorophores markedly decrease upon the exposure of HEK293T cells to camphor, apparently independently of the presence of TRPV1. The camphor-induced increase in the FRET signal might reflect changes in the molecular proximity of the membrane-bound PH domains. Another explanation, although less probable and not mutually exclusive, is that camphor reversibly inhibits the basal activity of PLC, which temporarily increases the concentration of PIP2and decreases the cytosolic translocation of PH domains. To examine this latter (less likely) possibility, we tested the effects of the PLC inhibitor U73122 (1-[6-[((17β)-3-Methoxyestra-1,3,5-trien-17-yl)amino]hexyl]-1H-pyrrole-2,5-dione) on camphor-induced changes in the FRET signal in HEK293T cells transiently transfected with PLCδ1PH-CFP and PLCδ1PH-YFP. U73122 is a widely used inhibitor of PLC that, upon prolonged exposure, exhibits several side effects attributable to alkylation of various proteins.35,44For experiments in which the aim is to prevent the depletion of PIP2, these side effects can be minimized by a brief pretreatment with relatively high concentration of U73122, followed by a removal of this compound.35,44We applied 10 mM camphor and treated the cells with a 3-μM concentration of U73122 for 1–2 min, followed by a wash of 1–2 min. Camphor was then applied again. We observed a substantial decrease in the FRET signal in response to U73122, which could not be attributed to the inhibition of PLC (fig. 7A) but could be explained as a direct side effect of this compound. More importantly, after pretreatment with U73122 (but not with the structurally similar inactive analog U73343; 1-[6-[((17β)-3-Methoxyestra-1,3,5-trien-17-yl)amino]hexyl]-2,5-pyrrolidinedione;fig. 7, B and C), camphor increased the FRET ratio signal by a factor of 1.8 ± 0.7 (n = 11; P < 0.005) compared with the initial camphor response, indicating that the inhibition of PLC intensifies the camphor-induced changes in the effective distances between CFP and YFP. However, this effect did not appear to be related to TRPV1 responsiveness because U73122 (at 3 μM applied using an identical protocol) did not influence the inward currents induced by camphor in TRPV1-HEK293T cells, but it did reduce the capsaicin-induced acute desensitization in these cells (data not shown). Although the relevance of these findings to TRPV1 activity remains to be determined, our experiments clearly show that camphor is able to dynamically affect vital processes at the inner leaflet of the plasma membrane.
It recently has been shown that PIP2potentiates TRPV1, and its depletion leads to channel inactivation,35,45,46so we investigated whether camphor-mediated effects on TRPV1 might be linked to changes in the local concentration of the plasma membrane PIP2. We examined whether replenishing the membrane with PIP2would affect the responsiveness of TRPV1 to camphor. We included 25 μM DiC8-PIP2in the whole cell patch pipette and measured the inward currents induced either by 10 mM camphor (n = 4) or by a combination of 10 mM camphor with 10 μM capsaicin (n = 3) before and after 3 min of intracellular dialysis with diC8-PIP2. Representative recordings shown in figure 7D–F demonstrate that there was no difference between dialyzed and nondialyzed cells (n = 6) in the time course of responses. There were also no statistically significant differences in camphor sensitivity in the truncated construct of TRPV1 that lacked the last 42 amino acid residues at the distal C terminal tail (data not shown). This sensitizing mutation has been proposed previously to be attributable to a disruption to the direct interaction of TRPV1 with PIP2.47However, later studies cast doubt on this initial hypothesis and attributed the potentiating effects of the C-terminal truncation to a disrupted interaction of TRPV1 with the scaffolding protein AKAP79/150 (A-kinase anchoring protein 79 is the human ortholog; AKAP150 is the rodent ortholog). This anchoring protein mediates protein-kinase-C-dependent phosphorylation of TRPV1 at the two key phosphorylation sites S502 and S800.48,49It also has been shown recently that PIP2degradation increases AKAP79/150 association with TRPV1, positively modulating receptor/channel activity.50We reasoned that if phosphorylation sites required for TRPV1 modulation were mutated, changes in the responsiveness of TRPV1 to camphor would be indicative of the changes in local concentration of the plasma membrane PIP2. Thus, another attempt to correlate our FRET results with TRPV1 functionality was to assess the camphor sensitivity in the nonphosphorylatable TRPV1 double mutant S502A/S800A (fig. 8). For this purpose, we measured currents induced by 1, 3, and 10 mM camphor, and compared the amplitudes measured at −70 mV and at +70 mV with the currents elicited by camphor in the wild-type. Strikingly, the mutant channels exhibited markedly increased camphor sensitivity at both negative and positive membrane potentials compared with the wild-type, suggesting that the interaction of TRPV1 with camphor depends on the phosphorylation status of the channel.
The data presented here indicate that the camphor-activation pathway in TRPV1 involves the outer pore domain, particularly T633, a specific residue located in the middle of the pore helix that is also critical for direct activation of TRPV1 by protons.15Replacing T633 with alanine reduced camphor activation of the channel, while retaining the capsaicin responses. The TRPV1 chimera in which the N-terminal portion of the pore helix (Y627-C634) was replaced with its counterpart from TRPV2 was completely insensitive to camphor. This construct also lacked the stretch of nonconserved 15 amino acid residues between the turret and selectivity filter (T612-S626), and we show here that this region is not important for camphor sensitivity (fig. 4F), as has also been shown for other stimuli (capsaicin, protons,15and heat51). Our findings, together with the previous observations that the chimeric channel cannot be activated by protons and exhibits a markedly slower onset and offset of capsaicin responses,15extend the proposal that the pore-helix domain represents a universal gating element common to various distinct activation modalities.31The idea that the pore helix forms a movable barrier to ion flux that, depending on the protonation state of the channel, ultimately regulates channel opening is also supported by our observation that the chimeric channel was completely insensitive to depolarizing voltages (fig. 4D).
There are several possible mechanisms by which camphor may activate TRPV1:
(1) Classic ligand-receptor interaction. Given the ability of camphor to partition into the plasma membrane, hydrophobic/intramembrane interaction sites are likely. However, our data showing that camphor occludes the potentiating effect of protons are consistent with the idea that it interacts with the proton-activation pathway directly (i.e. , the channel pore domain from the extracellular side).15In addition, the experiments reported here demonstrate that the camphor activation/deactivation kinetics are remarkably fast and the potentiating effects are readily reversible, indicating a direct and extracellular site of action. To assess the potential mechanism of camphor interaction at TRPV1 and search for possible binding sites at TRPV1, we carried out a series of ligand docking experiments using two previously published homology models of the rat25and human26TRPV1 channel (fig. 9). Generally, the docking results proposed several possible docking sites for camphor: four in rat and three in human TRPV1. When the binding modes with affinities ranging from less than 2 Å for the lower bound RMSD and less than 3 Å for the upper bound RMSD were taken into further consideration for each predicted binding site, both orthologs exhibited the best two docking positions: the first was predicted to be close to the intracellular part of the lipid-exposed face of transmembrane segments S1 and S2 (fig. 9A). The main interactions stabilizing the putative complex between camphor and TRPV1 were identified at W427, K432 in S1 and F496, F490, R491, Y487 in S2. The second docking pose was predicted to be in the pore loop region, close to the selectivity filter of TRPV1 (fig. 9B). The main interactions at TRPV1 were: L638 and F639 in the pore helix and L648 in the external linker between the pore helix and S6. These results increase the possibility that camphor directly binds TRPV1. In this regard, the existence of an extracellular interaction site is also supported by our observations that camphor exhibited a direct blocking effect, as could be seen from the partial current recovery upon washout (fig. 2A, b and 2F). More importantly, it has been shown previously that the selectivity filter of TRPV1 does not undergo dynamic conformational changes during camphor activation,20which further supports a direct “stabilizing” or “rigidifying” effect on the outer pore domain, which serves as an allosteric regulatory site of the channel.31
(2) Another possible way in which camphor could activate TRPV1 is by membrane receptor-mediated effects. Activation might be a consequence of incorporating camphor into the membrane structure, resulting in a change in its physical properties (e.g. , changes in anisotropy or dipolar organization52) and a subsequent conformational change of TRPV1. The partial agonism of camphor at TRPV1 indicates that the energetic coupling from the putative camphor interaction site to the gate is not strong enough to induce a full response in a resting channel. Our results indicate that the potentiating effects of camphor may be related to the changes in the molecular proximity of PIP2at the inner leaflet of the plasma membrane or to the changes in the molecular distance between PIP2and the putative PIP2-binding domain in TRPV1.46Strikingly, despite the relatively high concentrations required for activation and potentiation, the effects of camphor were found to be rapidly and readily reversible, indicating a mechanical, rather than biochemical, linkage to the pore gating machinery. Monoterpenes have been clearly demonstrated as compounds that affect general membrane properties such as fluidity, membrane polarity, and the dipolar organization of the receptor's environment (see Turina et al. 52and Sánchez et al. 53). However, compared with other TRPV1 agonists, camphor has a relatively low lipophilicity (logP value of 2.2) and thus is expected to exhibit a slow diffusion rate into the cell membrane.54Therefore, the effects of camphor at TRPV1 do not appear to be related solely to the physicochemical properties of this compound. It is conceivable that camphor may exert a condensation effect on lipids, and whether increasing temperature can counteract the camphor-induced FRET changes is an interesting subject for future study.
(3) Finally, camphor could act by affecting the interaction of TRPV1 with a specific regulatory molecule, such as membrane PIP2,25,47AKAP79/150,55or Pirt.56We found that the inclusion of PIP2in the whole cell patch pipette does not affect the camphor-mediated potentiation of responses induced by 10 μM capsaicin. This result may indicate either that the redistribution of PIP2is not the primary signal for camphor-mediated effects on TRPV1 or that the affinity of the receptor for PIP2markedly increases in the presence of camphor: If camphor tightened the receptor-phospholipid interaction, increasing PIP2levels should not cause any additional effects. Camphor has been reported to inhibit the tonic activity of PLC in HEK293T cells in a TRP-independent manner.57We explored the possibility that if camphor inhibits PLC, it might increase the accessible concentration of PIP2in the inner leaflet vicinity of the channel and reduce its activation energy for opening. In our experiments, camphor increased the FRET signal between CFP- and YFP-tagged PH domains in cells treated with the PLC inhibitor U73122 as well as in TRPV1-expressing cells in which PLC was activated by capsaicin stimulation. Therefore, the PLC signaling pathway is not likely to be involved in this camphor-induced effect. Interestingly, we observed that the activation of TRPV1 by camphor elicited substantially smaller Ca2+transients compared with that of 1 μM capsaicin (∼50% of the maximum 340/380 ratio measured with FURA-2, data not shown), but this Ca2+influx may still be sufficient to induce changes in the PLC-dependent cascades.
Previous studies have shown that the TRPV1-mediated currents induced by a near-threshold concentration of camphor (1.5 mM) can be markedly potentiated by Gq/11protein activation.8Stimulation of the PLC-coupled receptors (muscarinic M1 or bradykinin B2) results in the hydrolysis of PIP2and the activation of protein kinase C. Both these pathways contribute to the potentiation of responses to low concentrations of capsaicin (∼20 nM) and converge at the key phosphorylation site Ser502 on TRPV1. These potentiating effects critically depend on the presence of the scaffolding protein AKAP79/15050that forms a signaling complex with protein kinase Cε, protein kinase A, and protein phosphatase 3 (calcineurin), and binds to the C-terminal domain of TRPV1 (K735-N748).55We found that the protein kinase C (and partially protein kinase A) nonphosphorylatable double mutant of TRPV1, S502A/S800A, exhibited statistically significantly higher sensitivity to camphor (fig. 8). This result indicates that the apparent affinity of TRPV1 to camphor might depend on the degree of basal sensitization by protein kinases.49Our observation that extracellularly applied camphor induced reversible PIP2redistribution on the inner leaflet of the plasma membrane provides a possible mechanistic explanation; the extent to which camphor modulates the channel could depend on the degree of its association with the scaffolding protein that controls the protein kinase A- and C-mediated phosphorylation of TRPV1.50In this regard, it is interesting that the two other camphor-activated TRP channels, TRPV358and TRPM8,42are critically regulated by membrane PIP2.
In conclusion, our results provide functional support for the role of the putative outer pore region in controlling the camphor-dependent gating of the TRPV1 channel. Furthermore, these results extend the recent proposal that the pore-helix domain of TRPV1 represents a universal gating element common to many activation modalities and raise the hypothesis that the mechanisms identified here are indicative of domains in which camphor may interact to gate the channel. Obviously, the overlapping characteristics of thermosensitive TRP channels may help us along the way to understanding the mechanisms by which camphor modulates the sensation of warmth in humans and reveal the critical determinants for the indisputable beneficial effects of this natural compound.
The authors thank Tamas Balla, D.Sc. (Senior Investigator, Section Molecular & Signal Transduction, National Institute of Child Health & Human Development, National Institutes of Health, Bethesda, Maryland), for providing the YFP- and CFP-tagged pleckstrin homology domain vectors; David Julius, Ph.D. (Professor, Department of Physiology, University of California, San Francisco, San Francisco, California), for the rat TRPV1 vector; Feng Qin, Ph.D. (Associate Professor, Department of Physiology and Biophysics, State University of New York at Buffalo, Buffalo, New York), for providing the chimeric (Δ15-TRPV1: Y627-C634) and the deletion (Δ15-TRPV1) constructs of TRPV1; and Jan Teisinger, Ph.D. (Senior Scientist, Department of Protein Structure, Institute of Physiology Academy of Sciences of the Czech Republic, Prague, Czech Republic), for the T633A mutant. The authors also thank Gregorio Fernandez-Ballester, Ph.D. (Associate Professor, Molecular and Cellular Design Unit, Molecular and Cell Biology Institute, University Miguel Hernandez, Alicante, Spain), for providing the homology model of human TRPV1.