Identification of molecular determinants of channel gating in the ...

The FASEB Journal article fj.08-107425. Published online June 16, 2008.

The FASEB Journal • Research Communication

Identification of molecular determinants of channel gating in the transient receptor potential box of vanilloid receptor I Pierluigi Valente,*,1 Nuria Garcı´a-Sanz,*,1 Ana Gomis,† Asia Ferna´ndez-Carvajal,* Gregorio Ferna´ndez-Ballester,* Fe´lix Viana,† Carlos Belmonte,† and Antonio Ferrer-Montiel*,2 *Instituto de Biologı´a Molecular y Celular, Universidad Miguel Herna´ndez, Elche, Alicante, Spain; and †Instituto de Neurociencias de Alicante, Universidad Miguel Herna´ndez-Consejo Superior de Investigaciones Cientı´ficas, San Juan de Alicante, Alicante, Spain ABSTRACT Transient receptor potential vanilloid receptor subtype I (TRPV1) is an ion channel gated by physical and chemical stimuli that belongs to the TRPV protein family. TRPV receptors contain a highly conserved, 6-mer segment near the channel gate, known as the TRP box, whose function remains unknown. Here, we performed an alanine scanning mutagenesis of the TRP box of TRPV1 (IWKLQR) and found that mutation of this motif affected channel gating by raising the free energy of channel activation. Functional characterization of TRPV1 mutants showed that substitution of I696, W697, and R701 by alanine severely affected voltage- and heat-dependent activation and notably reduced the capsaicin responsiveness and tachyphylaxia, while mutation of K698, L699, and Q700 had minor effects. In addition, mutation of I696 to alanine promoted a strong outward rectification at negative membrane potentials, and slowed the kinetics of channel activation. Taken together, our findings suggest that modification of I696, W697, and R701 to alanine altered channel function by affecting events downstream of the initial stimuli-sensing step and imply that intersubunit interactions within the TRP box play an important role in TRPV1 gating.—Valente, P., Garcı´a-Sanz, N., Gomis, A., Ferna´ndez-Carvajal, A., Ferna´ndez-Ballester, G., Viana, F., Belmonte, C., and Ferrer-Montiel, A. Identification of molecular determinants of channel gating in the transient receptor potential box of vanilloid receptor I. FASEB J. 22, 000 – 000 (2008)

Key Words: structure-function 䡠 thermoreceptor 䡠 functional coupling 䡠 nociceptor 䡠 sensory transduction 䡠 oligomerization

The capsaicin receptor is a member of the transient receptor potential (TRP) ion channel superfamily that gave name to the vanilloid subfamily (TRPV) of receptors (1, 2). This membrane protein plays an important role in noxious sensing and pain perception. TRPV1 is activated by harmful temperatures (ⱖ42°C) and exhibits remarkable heat sensitivity, with a Q10 ⱖ20 (3, 4). A seminal study concluded that the channel thermosen0892-6638/08/0022-0001 © FASEB

sitivity of TRPV1 arises from the difference in activation energies associated with voltage-dependent gating (5, 6). In addition, this channel protein is gated by vanilloid molecules such as capsaicin and resiniferatoxin, extracellular pH, and proinflammatory substances (1, 2, 7, 8). It has been proposed that chemical agonists of this thermoreceptor function as gating modifiers that mimic and potentiate the thermal responses (5). A functional TRPV1 channel is a homotetramer of subunits assembled around a central aqueous pore (9, 10). The overall topological organization of TRPV1 subunits is akin to that displayed by the hexahelical Shaker-like K⫹ channels, having 6 transmembrane segments (S1–S6) and cytosolic N and C termini (2). Whereas the N terminus contains ankyrin domains that contribute to channel assembly and function (11–16), the C terminus appears to be important for subunit tetramerization, temperature sensing, and modulation of channel gating (17–21). Near the channel gate, there is a 6-mer segment, referred to as the TRP box, which is highly conserved among the TRP channel family (2). In TRPM channels, this motif is implicated in the modulation of channel activity by phosphoinositides (22–24), and it seems to mediate the coupling of menthol binding to channel opening (25). These observations strongly suggest that the TRP box may also serve as a molecular determinant of TRPV1 channel function. In support of this hypothesis, we have recently reported that the TRP domain of TRPV1, a 25-mer segment at the C terminus that contains the TRP box motif, plays a role in modulating channel gating (21). Here, by using alanine scanning mutagenesis of the entire TRP box, we found that mutation of this motif alters the channel response to voltage, capsaicin, and heat, and we identified residues I696, W697 and R701 as molecular determinants of channel gating. We re1

These authors contributed equally to this work. Correspondence: Instituto de Biologı´a Molecular y Celular, Universidad Miguel Herna´ndez, Av. de la Universidad s/n, 03202 Elche, Alicante, Spain. E-mail [email protected] doi: 10.1096/fj.08-107425 2

1

port that replacement of these residues by alanine notably impaired capsaicin-, voltage-, and heat-induced channel activity by a mechanism that is consistent with the alteration of downstream gating events rather than affecting the sensitivity to the activating stimuli. These findings suggest that inter- and intrasubunit interactions at the level of the TRP box in TRPV1 are important for efficient channel gating and suggest that this motif may be involved in the functional coupling of stimulus sensing and pore opening.

MATERIALS AND METHODS TRPV1 receptor mutagenesis Site-directed mutagenesis of the residues encompassing the TRP box to alanine was carried out by PCR as described (26). Mutant receptors were confirmed by DNA sequencing. For mutants, the number indicates the position of the residue in the protein sequence; the first letter is the natural amino acid in the wild-type protein, and the second is the residue that substitutes it. Cell culture and transfection Human embryonic kidney (HEK) 293 cells were cultured in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal calf serum, 2 mM l-glutamine, and 1% penicillin-streptomycin solution at 37°C in 5% CO2. Cells were transfected with 2 ␮g of DNA encoding the TRPV1 and mutant channels with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), following the manufacturer’s recommendations. Cells were used 48 h after transfection. Intracellular Ca2ⴙ imaging Cells were plated at a density of 106 cells/cm2 onto poly-dlysine-coated 25-mm coverslips. Cells were transfected with 2 ␮g of DNA encoding the TRPV1 and mutant channels with Lipofectamine 2000 (Invitrogen), following the manufacturer’s recommendations. Ca2⫹ image measurements were carried out 48 h after transfection. Cells were incubated with 5 ␮M Fluo-4 AM (Molecular Devices, Sunnyvale, CA, USA) in the presence of 0.02% pluronic F-127 (Biotium, Hayward, CA, USA) in isotonic standard solution [in mM: 140 NaCl, 4 KCl, 2 MgCl2, 2 CaCl2, 5 glucose, and 10 HEPES, pH 7.4; the osmolarity was adjusted to ⬇315 mosmol with manitol (Gonotec Osmometer; Gonotec GmbH, Berlin, Germany)] for 1 h at 37°C. For Ca2⫹ imaging, cells were continuously perfused (1 ml/min) with isotonic standard solution at ⬇22°C. TRPV1 activity was evoked with 15-s pulses of capsaicin (1 and 100 ␮M) using a multibarreled, gravity-driven perfusion system. Fluorescence measurements were carried out with a Zeiss Axiovert 200 inverted microscope (Carl Zeiss, Oberkochen, Germany) fitted with an ORCA-ER CCD camera (Hamamatsu, Shizuoka, Japan) through an ⫻20 objective. Fluo-4 was excited at 500 nm using computer-controlled Lambda 10 –2 filter wheel (Sutter Instruments, Novato, CA, USA), and emitted fluorescence filtered with a 510-nm long-pass filter. Images were acquired and processed with AquaCosmos package software (Hamamatsu). 2

Vol. 22

September 2008

Patch-clamp measurements in HEK293 cells HEK293 cells were cotransfected with TRPV1 species and the EYFP protein (pEYFP, Clontech, Palo Alto, CA, USA). Membrane currents were recorded with the whole-cell configuration using patch clamp as described (27). For whole-cell recordings, pipette solution contained (in mM): 150 NaCl, 3 MgCl2, 5 EGTA, and 10 HEPES, pH 7.2, adjusted with CsOH; and bath solution contained (in mM): 150 NaCl, 6 CsCl, 1 MgCl2, 1.5 CaCl2, 10 glucose, and 10 HEPES, pH 7.4, adjusted with NaOH. For capsaicin-induced tachyphylaxia experiments, isotonic standard solution was used as bath solution, and the pipette solution contained (in mM): 144 KCl, 2 MgCl2, 5 EGTA, and 10 HEPES, pH 7.2, adjusted with KOH. The different saline solutions were applied with a gravitydriven local microperfusion system with a rate flow of ⬃200 ␮l/min positioned at ⬃100 ␮m of the recorded cells. All measurements were performed at 20 –22°C. Voltage step protocols consisting of 200-ms depolarizing pulses from ⫺120 to 160 mV, or 50-ms depolarizing pulses from ⫺120 to 300 mV, in steps of 20 mV, were used. The holding potential was 0 mV, and the time interval between each pulse was 5 s. I-V relations were studied using a ramp protocol consisting of a voltage step of 300 ms from the holding potential of 0 mV to ⫺120 mV, followed by 350-ms linear ramp-up to 160 mV. Time interval between each pulse was 10 s. For capsaicin-induced tachyphylaxia experiments, we used a ramp protocol consisting of a voltage step from a holding potential of 0 mV to ⫺80 mV, followed by 1.5 s linear ramp-up to 80 mV, and ionic current values were collected every 3 s (28). Data were sampled at 10 kHz (EPC10 with pulse software; HEKA Elektronik, Lambrecht, Germany) and low-pass filtered at 3 kHz for analysis (PulseFit 8.54; HEKA Elektronik). The series resistance was usually ⬍10 M⍀ and to minimize voltage errors was compensated to 70 –90%. The G-V curves were obtained by converting the maximal current values, evoked with the voltage step protocols, to conductance using the relation G ⫽ I/(V⫺VR), where G is the conductance, I is the peak current, V is the command pulse potential, and VR is the reversal potential of the ionic current obtained from the I-V curves. G-V curves for each cell were fitted to the Boltzmann equation: G ⫽ Gmax/{1⫹exp[(V⫺V0.5)/ an]}, where Gmax is the maximal conductance, V0.5 is the voltage required to activate the half-maximal conductance, and an is the slope of the G-V curve. Thereafter, estimated Gmax values were used for obtaining the normalized G/Gmax-V curves. For voltage-dependent gating, the free energy difference between the closed and the open states at 0 mV and 25°C for a two-state model (⌬GO) was calculated using ⌬GO(V) ⫽ zgFV0.5 (29), where F is the Faraday constant (0.023 kcal/mol mV) and zg is the apparent gating valence obtained using zg ⫽ 25.69 mV/an. For channels that display voltage-dependent and -independent components in the G-V curve in the presence of capsaicin, the free energy of the activation process was obtained using ⌬GO ⫽ ⌬GO(V) ⫹ ⌬GO(I), where ⌬GO(V) denotes the free energy of the voltage-dependent component, and ⌬GO(I) denotes the free energy of the voltageindependent part. This energy was calculated using ⌬GO(I) ⫽ ⫺RT ln (PO/1⫺PO) (30), where PO denotes the probability of channel opening at hyperpolarized potentials. PO was calculated as the y intercept of the fitted straight line of the G/Gmax values from ⫺120 to ⫺20 mV. Temperature stimulation Coverslips with transfected HEK293 cells were placed in a microchamber and continuously perfused (⬃1 ml/min) with

The FASEB Journal

VALENTE ET AL.

solutions warmed to 25°C. The temperature of the solutions was controlled using a CL-100 bipolar temperature controller and an SC-20 dual inline heater/cooler and was measured by a TA-29 thermistor (Warner Instruments, Hamden, CT, USA). The temperature probe was placed near the solution outflow and ⬍500 ␮m of the patch-clamped cell. The time course of the temperature change was ⬃0.3°C/s. Temperature ramps were first obtained at ⫺60 mV and thereafter at ⫹50 mV in the same cell. Macroscopic ion currents were recorded with a Multiclamp amplifier using pCLAMP software and a Digidata 1322A digitizer (Axon Instruments; Molecular Devices). Data analysis Analysis was performed with either the PulseFit 8.11 (HEKA Elektronik) or with pClamp9, WinASCD software (G. Droogmans, Katholieke Universiteit Leuven, Belgium; ftp://ftp. cc.kuleuven.ac.be/pub/droogmans/winascd.zip) and Origin 7.0 (MicroCal, Northhampton, CA, USA). Data were expressed as means ⫾ sd, with n ⫽ number of cells. Statistical analysis was performed using the 1-way ANOVA, and P ⬍ 0.05 was taken as the level of significance.

RESULTS Mutation of the TRP box altered the channel activity of TRPV1 The TRP box is a highly conserved protein motif located at the core of the TRP domain in TRPV channels (18), as well as in other TRP proteins (2). To investigate its functional role, we performed an alanine scanning mutagenesis of all residues encompassing this region in TRPV1 (Fig. 1A) and characterized the phenotype of mutant channels in HEK293 cells. Mutation of TRP box residues did not alter significantly subunit multimerization and channel trafficking to the plasma membrane, as implied by the colocalization of mutant proteins with the cell surface marker wheat agglutinin and a surface biotinylation assay (see Supplemental Fig. S1). All mutant subunits gave rise to functional channels that exhibited capsaicin-evoked Ca2⫹ influx (Fig. 1B) but displayed distinct sensitivity to the vanilloid. Whereas mutation of I696, W697, and R701 to alanine produced proteins that did not respond to 1 ␮M capsaicin, mutants K698A, L699A, and Q700A were activated by this low agonist concentration (Fig. 1C, top panel; D). All mutants, however, responded to 100 ␮M capsaicin (Fig. 1C, bottom panel; D). Note that mutant I696A exhibited the lower Ca2⫹ entry at the high vanilloid concentration (Fig. 1C, D). These observations indicate that mutants I696A, W697A, and R701A display lower vanilloid sensitivity than K698A, L699A, and Q700A mutants. Results obtained from Xenopus oocytes heterologously expressing K698A, L699A, and Q700A mutants corroborated that they produce capsaicin responses akin to those of TRPV1 and showed that I696A, W697A, and R701A mutants were not activated by 10 ␮M capsaicin or by pH 6.0 at ⫺60 mV (Supplemental Fig. S2). Thus, residues I696, W697, and R701 DETERMINANTS OF TRPV1 CHANNEL GATING

Figure 1. Mutation of the TRP box produced functional channels. A) Amino acid sequence of the TRP domain of TRPV1. Red type highlights the TRP box sequence, whose amino acids have been individually mutated to alanine. B) Images show the intracellular rise in Ca2⫹ evoked by 1 ␮M capsaicin (cap) from HEK293 cells expressing the TRPV1 and mutant species. C) Change in Ca2⫹-dependent fluorescence of TRPV1-transfected and mock cells as a function of time, before and after the exposure to 1 ␮M (top) and 100 ␮M capsaicin (bottom) for 15 s. Images and traces are representative of 150 cells that showed capsaicin-induced Ca2⫹ flux. Cells were loaded with Fluo4-AM to record intracellular calcium signals and activated as indicated. Extracellular concentration of Ca2⫹ was 2.0 mM. D) Quantification of the relative capsaicin-evoked fluorescence change for TRPV1 and mutants. Data are given as mean ⫾ sd; n ⫽ 150 cells; 3 independent experiments. *No fluorescence change detected. **P ⬍ 0.05 vs. TRPV1.

appear to be more sensitive to replacement by alanine than K698, L699, and Q700. Mutation of the TRP box affected TRPV1 voltagedependent gating To further characterize the effects of mutating the TRP box on channel gating, we studied the voltage-dependent activation of mutant channels with whole-cell patch clamp (5). Application of 200-ms depolarizing pulses from ⫺120 to 160 mV in steps of 20 mV evoked noninactivating ionic currents from TRPV1, K698A, L699A, and Q700A mutants, but not from I696A, W697A, and R701A (Fig. 2). The I-V relations for voltage-activated mutants were similar to that of TRPV1 (Fig. 2). For the voltage-inactive mutants, we reasoned that they may require stronger depolarizing potentials. Because we could not achieve depolarizing potentials higher than 160 mV using 200-ms pulses, we applied shorter voltage pulses to reach more depolarized potentials without perturbing the high resistance seal. Stimulation with 50-ms depolarizing voltage steps from 3

Figure 2. Mutation of the TRP box residues to alanine produces channels that are not gated by voltage. Representative families of ionic currents elicited with a voltage step protocol consisting of 200-ms depolarizing pulses from ⫺120 to 160 mV in steps of 20 mV. Holding potential ⫽ 0 mV; time between pulses ⫽ 5 s. I-V relations for each TRPV1 species are shown underneath the ionic currents. I-V curves were studied using a ramp protocol consisting of a voltage step of 300 ms from the holding potential of 0 mV to ⫺120 mV, followed by 350-ms linear ramp-up to 160 mV.

⫺120 to 300 mV elicited noninactivating macroscopic currents from TRPV1 and I696A, K698A, L699A, and Q700A mutants, but not from cells expressing the W697A (n⫽20) and R701A (n⫽22) channels or mocktransfected cells (n⫽6) (Supplemental Fig. S3A). Note that all active mutants displayed current densities lower than those of wild-type channels (Supplemental Fig. S3B). The lowest current density is exhibited by the I696A mutant. This small current density was not due to a lower plasma membrane expression of mutants, as evidenced by the similar levels of surface expression revealed by a biotinylation assay (Supplemental Fig. S1). Functional characterization of active channels shows that mutation of K698, L699, and Q700 to alanine did not affect the kinetics of voltage-dependent channel activation from 120 to 160 mV (Supplemental Fig. S4A) nor the reversal potential of the channel (Supplemental Fig. S4B). In addition, these voltagegated ionic currents did not show a significant desensitization on repeated voltage stimulation (see below). Thus, mutation of I696, W697, and R701 to alanine significantly affected the voltage sensitivity of TRPV1.

We next obtained the G-V relations for TRPV1 and active mutants using the steady-state ionic currents evoked with the long (200 ms) depolarizing pulses (Fig. 2). Normalized conductance-to-voltage (G/Gmax-V) curves yielded a depolarized V0.5 for TRPV1 of 86 ⫾ 2 mV (n⫽8) and a gating valence (zg) of 0.70 ⫾ 0.10 (Fig. 3A–C), in accordance with reported values (5, 20). The K698A mutant exhibited a ⬇10 mV (95⫾3 mV, n⫽7) rightward change in V0.5, whereas mutation of L699 and Q700 to alanine shifted the voltage-dependent activation to 126 ⫾ 2 mV (n⫽8) and 135 ⫾ 2.4 mV (n⫽7), respectively (Fig. 3A, B). None of these mutations appear to alter the apparent gating valence moved by the change in voltage (Fig. 3C). For I696A, we could not determine these values because we could not elicit ionic currents with the 200-ms pulse protocol, and, with the shorter voltage paradigm, we could not get saturation of the G-V curve, indicating that V0.5 is ⱖ300 mV (Supplemental Fig. S3). Therefore, these results suggest that mutation of the TRP box altered the voltagedependent gating, presumably by raising the free energy of voltage activation. This notion was substantiated

Figure 3. Mutation of TRP box residues alters the energetics of voltage-dependent channel opening. A) G-V relations for TRPV1 wild-type and mutant channels. Conductance changes were obtained from the ionic currents shown in Fig. 2, using G ⫽ I/(V⫺VR), where V is the stimulation potential value and VR is the reversal obtained from the I-V curves. Conductance values for each cell were fitted to the Boltzmann equation: G ⫽ Gmax/{1⫹exp[(V⫺V0.5)/an]}, where Gmax is the maximal conductance, V0.5 is the voltage required to activate the half-maximal conductance, and an is the slope of the G-V curve. Gmax was thereafter used to obtain the G/Gmax-V curves shown. Solid lines depict the best fit to a Boltzmann distribution. B) V0.5 values for the different TRPV1 species obtained from the Boltzmann distribution of G-V relations. C) Apparent gating valence of the activation process: zg ⫽ 25.69 mV/an. D) Bar graph of the free energy at 0 mV and 25°C (⌬GO) for TRPV1 and mutants assuming a two-state model. ⌬GO ⫽ zgFV0.5, where F is the Faraday constant (0.023 kcal/mol mV). All values are means ⫾ sd; n ⱖ 5. 4

Vol. 22

September 2008

The FASEB Journal

VALENTE ET AL.

on evaluation of the free energy difference between the closed and the open states at 0 mV and 25°C (⌬GO) calculated from the values of V0.5 and zg, considering a two-state model (29). Although TRPV1 gating may be more complex than a two-state model (5, 31, 32), this approach is a simple and valuable means to compare the effect of mutations on the energetics of channel gating. In addition, it has been shown that a two-state model appropriately describes the voltage-dependent gating of the structurally related TRPM8 channel (32). Figure 3D depicts the values of ⌬GO obtained for TRPV1 and mutant channels. For TRPV1, the ⌬GO was 1.4 ⫾ 0.3 kcal/mol. Inspection of the plot reveals that K698A displayed a ⌬GO similar to wild type (1.6⫾0.2 kcal/mol), whereas L699A and Q700A augmented ⌬GO by ⬇0.9 kcal/mol and 1.1 kcal/mol, respectively. Therefore, these results imply that mutation of residues encompassing the TRP box has an effect on the activation energy of voltage-dependent gating. Furthermore, our findings signal that positions I696, W697, and R701 are important determinants of channel gating. Mutant channels I696A, W697A, and R701A exhibit voltage-dependent gating in the presence of capsaicin The impaired voltage-dependent activation of I696A, W697A, and R701A could be due to abrogation of stimuli sensing or an alteration in channel gating that sets the activation energy at very high values. Because voltage appears to be a partial activator of TRPV1, and capsaicin potentiates voltage responses and displaces the G-V relation of TRPV1 toward physiological potentials (5, 33), we next evaluated the effect of the vanilloid on the voltage activation of these three mutants. As illustrated in Fig. 4A–C, the three mutants showed voltage-dependent gating in the presence of capsaicin (see also Supplemental Fig. S5). Mutants W697A and R701A displayed ionic currents at negative potentials. In contrast, I696A exhibited a strong outward rectifying I-V relation with a barely detectable ionic current at hyperpolarized membrane voltages. Interestingly, voltage-activated ionic currents from these mutants did not exhibit significant vanilloid-induced desensitization, as evidenced by the virtually identical currents obtained with two consecutive depolarizing step protocols evoked in the continuous presence of 100 ␮M capsaicin (Supplemental Fig. S5). The G-V curves were obtained from ionic currents evoked with 100-ms depolarizing pulses from ⫺120 to 160 mV in the presence of increasing concentrations of capsaicin (Supplemental Fig. S5). The normalized G-V relations showed that 1) the three mutants exhibited voltage-dependent gating (Fig. 4D–F); 2) the vanilloid modulated differently the voltage-activated response of the three mutants (Fig. 4D–H); 3) a capsaicin-dependent, voltage-independent ionic conductance at negative membrane potentials was characteristic of W697A and R701A (Fig. 4E, F), similar to that reported for TRPV1 wild type (33); 4) the agonist shifted the V0.5 of I696A and W697A toward lower values in a doseDETERMINANTS OF TRPV1 CHANNEL GATING

dependent manner, but it did not alter that of R701A (Fig. 4G); and, 5) the apparent gating valence was unaffected by the presence of the vanilloid and was not significantly changed by the mutations (Fig. 4H). Analysis of the free energy of gating showed that capsaicin dose dependently reduced the free-energy difference between the closed and the open states at 0 mV and 25°C of I696A (Fig. 4I). For the W697A mutant, a decrement in free energy was also detected at 100 ␮M capsaicin; whereas for R701A, a small increment in ⌬GO as a function of the capsaicin concentration was observed (Fig. 4I). Note that for these two mutants, ⌬GO reflects the sum of the voltage-dependent and -independent components at capsaicin concentrations ⱖ10 ␮M. The voltage-independent component arises from the conductance displayed by both mutants at negative membrane potentials in the presence of the vanilloid. Taken together, these data imply a contribution of residues I696, W697, and R701 to the activation energy of channel gating. Mutant channels I696A, W697A, and R701A exhibit reduced sensitivity to capsaicin Because these mutants show capsaicin-dependent activity at depolarized potentials, we next investigated their vanilloid sensitivity by obtaining their respective capsaicin dose-response relations at ⫹60 mV using patch clamp. Higher potential wild-type channels gave rise to large ionic currents that could not be accurately clamped. Conversely, at negative potentials, the I696A mutant exhibited barely detectable ionic currents (Fig. 4A). As illustrated in Fig. 5A, the three mutants responded to capsaicin in a dose-dependent manner. Analysis of these curves revealed half-maximal effective concentration (EC50) values of 4.6 ⫾ 0.8 ␮M (n⫽4) for I696A, 8.1 ⫾ 1.3 ␮M (n⫽4) for W697A, and 45 ⫾ 5 ␮M (n⫽4) for R701A, which are significantly higher than that of TRPV1 (0.3⫾0.2 ␮M, n⫽4). For the three mutants, the Hill coefficient was significantly reduced (Fig. 5B), suggesting that subunit cooperativity during the gating step was also affected in these channels. Furthermore, their low capsaicin responsiveness was accompanied by a significant decrease in the maximal response evoked by the vanilloid (Fig. 5C). When compared with wild-type channels, mutants exhibited higher activation energies in the presence of a concentration of capsaicin that equaled their respective EC50 for the vanilloid receptor, in agreement with their higher energetics of channel gating (Fig. 6A). Taken together, these results demonstrate that mutation of I696, W697, and R701 to alanine reduced capsaicin sensitivity. Moreover, our data suggest that the change in the vanilloid EC50 of the mutants is attributed to a significant degree to the reduction in the maximal response evoked by capsaicin. This result is compatible with the notion that mutation of these residues in the TRP box affects events downstream of the initial binding step, such as channel gating, akin to the effect of mutations in the TRP box of TRPM8 (25). A similar 5

Figure 4. I696A, W697A, and R701A mutant channels exhibit voltage-dependent gating in the presence of capsaicin. Ionic currents evoked by increasing concentrations of capsaicin from the I696A (A), W697A (B) and R701A (C) mutants. I-V relations were recorded using a ramp protocol consisting of a voltage step of 300 ms from 0 to ⫺120 mV, followed by a 350-ms linear ramp-up to 160 mV. Cells were sequentially exposed to the increasing capsaicin concentrations. Inset (A): I-V curve for TRPV1 wild type in the absence and presence of 1 ␮M capsaicin. Representative current traces from n ⫽ 4 cells measured are shown. Whole-cell currents were obtained in symmetrical 150 mM NaCl, from a holding potential of 0 mV. D–F) G-V relations for I696A (D), W697A (E), and R701A (F) mutants were obtained at increasing capsaicin concentrations. Conductance values were obtained from the ionic currents evoked with 100-ms depolarizing pulses from ⫺120 to 160 mV in steps of 20 mV as in Fig. 3 (Supplemental Fig. S5), and fitted to a Boltzmann distribution (solid lines). G, H) Values of V0.5 (G) and zg (H) as a function of the vanilloid concentration for the three mutants, obtained from the G-V curves. I) Free energy of the activation process was obtained as described in Materials and Methods, using ⌬GO ⫽ ⌬GO(V) ⫹ ⌬GO(I), where ⌬GO(V) and ⌬GO(I) denote the free energy of the voltage-dependent and -independent components, respectively. ⌬GO(I) ⫽ 0 for I696A at all capsaicin concentrations, and for W697A and R701A at 1 ␮M capsaicin, since no conductance could be measured at hyperpolarized potentials. For W697A, ⌬GO(I) ⫽ 0.64 ⫾ 0.10 and 0.093 ⫾ 0.008 kcal/mol at 10 and 100 ␮M capsaicin, respectively; for R701A, 0.73 ⫾ 0.12 and 0.70 ⫾ 0.16 kcal/mol at 10 and 100 ␮M capsaicin, respectively. All values are means ⫾ sd; n ⱖ 5.

result is likely for pH-induced activation; however, because of the presence of pH-dependent ionic currents in HEK293 cells, we could not study this stimulus in our mutants. Mutation of I696 appears to stabilize the closed state of the channel To learn more of the I696A functional phenotype, we first examined the I-V relation in the presence of increasing concentrations of the vanilloid (Fig. 4A). As depicted, the I696A mutant exhibited a distinct I-V characterized by a strong outward rectification to the entire range of the vanilloid concentrations tested. At variance with W697A and R701A (Fig. 4B, C), the I696A mutant showed modest ionic current at negative poten6

Vol. 22

September 2008

tials even in the presence of 100 ␮M of capsaicin. As a result, the estimated rectification index for this mutant significantly increased as a function of the vanilloid concentration (Fig. 6B). This outward rectification was not due to a Ca2⫹ or Mg2⫹ blockade of the mutant channel, because the rectification index was unaffected by removal of both divalent cations from the external medium (data not shown). It also did not appear to be due to a change in permeation properties, as it displayed a reversal potential within the range of the other channels (Supplemental Fig. S4B). Thus, the strong rectification suggests an impaired channel activity at negative potentials that could be due, at least in part, to stabilization of the channel closed state under hyperpolarization, which would result in a low probability of pore opening. In support of this notion, this mutant

The FASEB Journal

VALENTE ET AL.

Figure 5. I696A, W697A, and R701A display altered capsaicin sensitivity. A) Dose-response curves for capsaicin activation of mutants. Ionic currents were evoked at 60 mV with a 100-ms voltage pulse. Cells were depolarized in the presence of the corresponding vanilloid concentration. The dose response was obtained at 60 mV, since ionic currents evoked with ⱖ1 ␮M capsaicin at higher potentials gave rise to sizable currents from TRPV1 wild type that could not be properly clamped. Normalized ionic currents were plotted as a function of the capsaicin concentration and fitted to a Michaelis-Menten binding isotherm to obtain the EC50 and Hill coefficient of the activation process (26). TRPV1: EC50 ⫽ 0.3 ⫾ 0.2 ␮M, nH ⫽1.48 ⫾ 0.15; I696A: EC50 ⫽ 4.6 ⫾ 0.8 ␮M, nH ⫽ 0.97 ⫾ 0.13; W697A: EC50 ⫽ 8.1 ⫾ 1.3 ␮M, nH ⫽ 0.8 ⫾ 0.2; R701A: EC50 ⫽ 45 ⫾ 5 ␮M, nH ⫽ 0.72 ⫾ 0.3. B) Hill coefficient for TRPV1 and mutant channels. C) Maximal ionic current at 60 mV for TRPV1and the mutant channels, measured with 10 and 100 ␮M capsaicin, respectively. Data are means ⫾ sd; n ⱖ 4.

exhibited slower kinetics of its voltage-dependent activation in the presence of capsaicin than TRPV1 and the W697A and R701A mutants (Fig. 6C and Supplemental Fig. S6). The time constant for channel activation of I696A was accelerated by the vanilloid in a dosedependent manner, although, even at 100 ␮M capsaicin, it was 10-fold slower than that of TRPV1 (Fig. 6C). Collectively, these findings imply that incorporation of a smaller hydrophobic amino acid at position I696

Figure 6. Mutation of I696 notably affects TRPV1 channel properties. A) Variation of the free energy of channel activation of mutants I696A, W697A, and R701A with respect to TRPV1 in the presence of a concentration of capsaicin that matches their respective EC50 values. ⌬GO values were obtained as in Fig. 3. B) Rectification index (RI) of mutants as a function of the capsaicin concentration, calculated as the ratio of conductance at 150 and ⫺100 mV. RI ⫽ [(I150/ (150⫺VR)]/[I–100/(⫺100⫺VR], where I is ionic current at the indicated potential, and VR is the reversal potential. C) Kinetics of the activation process at 160 mV for mutants as a function of vanilloid concentration. The activation process was fitted to one exponential function of the form I(t) ⫽ Ae⫺t/␶, where A is the amplitude of the ionic current I, and ␶ is the time constant of the activation process (see Supplemental Fig. S6). Ionic currents were fitted with the Pulse/PulseFit software. All values are means ⫾ sd; n ⱖ 5. DETERMINANTS OF TRPV1 CHANNEL GATING

increases the stability of the closed state of the channel, which requires higher energetic input for gating. Mutants I696A, W697A, and R701A display reduced capsaicin-induced tachyphylaxia To further characterize the functional properties of TRP box mutants, we next investigated the channel desensitization and tachyphylaxia evoked by 100 ␮M capsaicin at ⫺80 and ⫹ 80 mV in physiological solutions. We used this high concentration of capsaicin to ensure complete and fast desensitization and tachyphylaxia of TRPV1 species. The experimental paradigm consisted in the application of 1-min capsaicin pulses, interspersed with 1-min washing periods. The activity of the channel during agonist application was measured using a ramp protocol from ⫺80 to ⫹80 mV, and ionic current values were collected every 3 s (28). As illustrated in Fig. 7A, for wild type channels, the application of 100 ␮M capsaicin evoked a rapid desensitization (ⱖ80% initial current) of the channels at ⫺80 mV. This effect was followed by a strong tachyphylaxia, as evidenced by the lack of capsaicin-evoked responses on additional agonist pulses. In contrast, at the depolarized potential, the vanilloid produced a slower desensitization and a partial tachyphylaxia, similar to results reported by other groups (28). Note the presence of a relatively constant ionic current at ⫹80 mV with all the capsaicin pulses applied. These data suggest that depolarizing potentials could rescue, at least partially, the channel from the desensitized stated promoted by the agonist. Quantification of the normalized current as a function of the pulse number for the first 5 agonist applications shows a marked tachyphylaxia for TRPV1 at both voltages, although for depolarizing potentials, the effect reached a steady-state value of ⬇25% of the initial ionic current (Fig. 7H). Inspection of Fig. 7D–F, H reveals that mutants K698A, L699A, and Q700A 7

potentials (Fig. 7B). This channel mutant displayed a slow activating response that was not desensitized during capsaicin application. The agonist-induced tachyphylaxia was also markedly attenuated, compared to the other mutants (Fig. 7B, H). At variance with W697A and R701A, the agonist could be readily washed out from this mutant. Taken together, these results further support a critical role of I696, W697, and R701 in agonist-induced responses, and mutation of these positions has a significant effect in the desensitization and tachyphylaxia evoked by capsaicin. Because of the lower agonist EC50 displayed by these mutants, the reduced capsaicin-induced tachyphylaxia does not seem to arise from an increment in ligand affinity but rather from alteration of the gating steps near the channel gate. Mutation of the TRP box altered the temperature sensitivity of TRPV1

Figure 7. Mutation of the TRP box alters the capsaicininduced desensitization and tachyphylaxia. A–G) Capsaicin ionic currents evoked from TRPV1 wild-type and mutant channels at ⫺80 and 80 mV. The experimental paradigm consisted of pulses of 1 min of 100 ␮M capsaicin, interspersed by 1-min washout periods with standard extracellular buffer. Ionic currents were monitored using a ramp protocol from ⫺80 to 80 mV in 1.5 s, from a holding of 0 mV. Ionic currents were measured every 3 s. Data are representative of ⱖ5 cells for each channel species. H) Variation of the normalized peak current measured for the first 5 capsaicin pulses at ⫺80 and 80 mV as a function of pulse number. Data were normalized with respect to the maximal ionic current activated with the first pulse of capsaicin. Data are means ⫾ sd; n ⫽ 5.

displayed a desensitizing phenotype akin to that of TRPV1. In contrast, W697A and R701A mutants exhibited a noticeably different behavior (Fig. 7C, G). As seen in Fig. 7C, G, the desensitization was slower, such that the channels could only be partially desensitized during the first capsaicin application. Furthermore, the agonist could not be completely washed out, and additional vanilloid pulses resulted in small evoked currents. Because of the persistent ionic current seen for these channel mutants, the tachyphylaxia kinetics appear to be lower than those of wild-type channels, as clearly discerned at ⫺80 mV (Fig. 7H). The I696A mutant did not display measurable ionic currents at ⫺80 mV and was only investigated at depolarizing 8

Vol. 22

September 2008

Because the C terminus of TRPV1 is a key molecular determinant of heat sensing (20, 24), we next investigated the heat sensitivity of our mutant channels to determine the contribution of the TRP box. For this purpose, we obtained temperature-evoked responses at ⫺60 and ⫹50 mV (Supplemental Fig. S7). Heat ramps activated ionic currents from K698A, L699A, and Q700A mutants but not from I696A and W697A (Fig. 8A and Supplemental Fig. S7). For the R701A mutants, a hardly detectable ionic current could be observed at high depolarized potentials (Fig. 8A and Supplemental Fig. S7). Heat-evoked ionic currents from active mutants were similar to TRPV1, although their magnitude was smaller, especially at hyperpolarized membrane potentials (Fig. 8A). We next evaluated the voltage dependency of heat-activated currents by obtaining the I-V relations for all channels at 45°C (Fig. 8B). Mutants K698A, L699A, and Q700A displayed I-V curves similar to TRPV1. Notably, the R701A mutant, that was insensitive to voltage changes at 25°C, was voltage-gated at 45°C (Fig. 8B). In marked contrast, the I696A and W697A mutants did not produce measurable heatactivated currents within the voltage range applied (up to ⫹160 mV) (Fig. 8B). Thus, mutation of the TRP box affected the heat responsiveness of TRPV1. These findings point to I696 and W697 as two important determinants of heat sensing in TRPV1. The lower temperature sensitivity of the mutants may arise from an alteration of the heat sensor or to a defect in the temperature-induced gating. To address whether mutations affected heat sensing, we determined the threshold temperature of channel activation from the I-T relations of TRPV1 and active mutants at ⫹50 mV. As illustrated in Fig. 8C, wild-type and mutant channels displayed a similar threshold temperature of ⬇42°C (40 – 44°C) for channel activation. Even for mutant R701A, which exhibits lower heat-activated currents than the other mutants, the threshold temperature was also ⬇43°C. In contrast, mutant channels depicted an altered temperature-dependent gating, as determined from

The FASEB Journal

VALENTE ET AL.

Figure 8. Heat-dependent activation of TRP box mutants. A) Mean amplitude of the ionic currents evoked by a heat ramp from 25 to 46°C, recorded at 50 mV and 45°C for TRPV1 and chimeric channels. Ionic currents were obtained from the recordings shown in Supplemental Fig. S7. B) I-V relations for TRPV1 and mutant channels obtained at 45°C. Cells were depolarized from ⫺120 to 160 mV in 350 ms. Traces are representative of n ⱖ 3 cells. C, D) Activating threshold temperature T (C) and heat sensitivity Q10 (D) of channel gating, determined at 50 mV for TRPV1 and mutants. T was determined as the crossover of the lines approximating the stable baseline current and that reflecting the temperature-dependent response (38, 39). Q10 was obtained for the inverse slope of log (I) vs. 1/T (Kelvin) between 35 and 46°C as described by Xu et al. (39). Data are means ⫾ sd; n ⱖ 3 cells. *P ⬍ 0.05; 1-way ANOVA.

the Q10 value obtained at 35– 46°C (6, 30, 31). Whereas TRPV1 showed a mean Q10 of 23 ⫾ 5 (n⫽8), mutant channels K698A, L699A, and Q700A displayed a ⬇50% reduction of this parameter (Fig. 8D). For R701A, the Q10 value could not be properly calculated because of its small heat-activated currents. These data support the tenet that the lower heat responsiveness of TRP box mutants is probably due to an alteration of the temperature-induced gating events and support the tenet that this domain does not act as the temperature sensor.

domain, a coiled-coil structure that has a dual functionality, serving as a molecular determinant of subunit multimerization and as a transduction domain important for channel gating (18, 21). Furthermore, the TRP domain of TRPM channels appeared to be required for phosphoinositide-mediated channel gating (23, 34), as well as for activation with menthol (25). Similarly, the corresponding domain in TRPC channels is required

Mutants I696A and W697 have markedly affected temperature sensitivity The important effect on heat sensitivity of mutating I696, W697, and R701 to alanine suggests a critical role of these residues in heat sensing. To additionally address this issue, we investigated the effect of raising the temperature to 45°C on the capsaicin response of these mutants. To investigate the heat potentiation, we used a capsaicin concentration (1 ␮M) that barely activated the mutants at 25°C (Fig. 9). At variance with the I696A mutant (Fig. 9A), the heat- and vanilloid-evoked ionic currents of W697A and R701A were significantly potentiated (⬇50% and ⬇4-fold, respectively) by the simultaneous presence of both activating stimuli (Figs. 9B, C). Note, however, that W697A displayed a modest heat potentiation compared with R701A. In contrast, for the I696A mutant, the capsaicin responsiveness was unchanged by the rise in temperature (Fig. 9A). These data imply that R701A preserves heat sensitivity and further substantiates that I696A and W697A have a noticeably affected temperature-dependent gating that makes these channel mutants rather heat insensitive, especially the I696A mutant.

DISCUSSION The TRP box is a highly conserved motif in the cytosolic C terminus domain of TRP channels (Fig. 10A). In TRPV1, this segment is at the core of the TRP DETERMINANTS OF TRPV1 CHANNEL GATING

Figure 9. Effect of heat on the capsaicin responses of TRPV1 mutants. Mean amplitude of the ionic currents evoked by heat (45°C) and 1 ␮M capsaicin at 25 and 45°C for I696A, W697A, and R701A. Ionic currents recorded at 50 mV. Data are means ⫾ sd; n ⱖ 3 cells. *P ⬍ 0.05 obtained with the 1-way ANOVA. 9

Figure 10. A putative model for the proposed coiled-coil structure of the TRP box in TRPV1. A) Molecular conservation of the TRP box (red box). Multiple-sequence alignments of TRPV, TRPM, and TRPC families were done with Clustalw (http://www.expasy.org/tools/), showing the high degree of conservation of the TRP box region. B) Structural arrangement of the TRP box in TRPV1 according to our proposed model for the C terminus of TRPV1 (18). The TRP box could participate in the formation of a tetrameric coiled-coil structure, with I696 in the bundle core, W697 interacting with adjacent subunits, and R701 exposed to the solvent and interacting either with the lipid bilayer or other receptor regions.

for efficient transient light response in flies (35). Therefore, these observations suggest that the TRP box of TRPV1 may be functionally relevant. The salient contribution of our study is the demonstration that this protein motif in TRPV1 participates in channel gating by contributing to set the activation energy of pore opening. By using an alanine-scanning mutagenesis 10

Vol. 22

September 2008

approach, we report that structural changes in the TRP box notably affected the response of the channel to the activating stimuli. This effect was not due to loss of stimulus recognition, since we found that mutation of TRP box residues did not change the ability of the channel to sense the gating stimuli, but it rather affected the gating reaction, conceivably by altering steps downstream of the initial sensing event, akin to the findings reported for TRPM8 (25). Our findings are compatible with a model in which residues of the TRP box might contribute to couple the activating stimuli to pore opening in TRPV1. The alteration of specific inter- and intrasubunit interactions at the level of this motif appears to have a significant effect on channel gating. This result substantiates our previous results showing that the TRP domain plays a role in TRPV1 function (21) and identifies molecular determinants of channel gating within the TRP box of the receptor. We found a differential contribution of TRP box residues to channel activation. Substitutions of K698, L699, and Q700 modestly affected TRPV1 function, moderately augmented the free energy of pore opening, and did not significantly alter the ligand-induced activation and desensitization responses. In marked contrast, mutation of I696, W697, and R701 drastically altered the sensitivity to the activating stimuli and significantly raised the energetic barrier between the closed and open states of the channel. Replacement of I696, W697, and R701 with alanine decreased the capsaicin responsiveness of the mutant receptors and, concomitantly, attenuated the agonist-induced desensitization and the tachyphylaxia provoked by its repetitive exposure. The effect on the vanilloid efficacy could arise from a change of the ligand affinity or to a modification of the activation step (36). The important decrease in the maximal response elicited by capsaicin in these mutants, along with their lower Hill coefficient, an indicator of the subunit cooperativity during channel gating (36), suggests that the lower vanilloid responsiveness may arise primarily from an alteration of the gating step rather than the initial agonist binding event (25). This observation implies that the unliganded state of these mutants is similar to that of TRPV1 in its ability to undergo the allosteric, agonist-induced conformational change required for channel opening. Nonetheless, a change in agonist affinity produced by the mutations cannot be completely ruled out, especially taking into account an allosteric protein, such as TRPV1. Indeed, it should be considered that the ligand affinity depends on both the occupancy of the binding sites and the conformational state of the protein (36). Further experimental support is needed to address this issue. Mutation of I696, W697, and R701 markedly affected the ability of voltage to activate ionic currents in resting conditions, although they displayed voltage-dependent gating in the presence of capsaicin. The vanilloid receptor shifted leftward the voltage-dependent channel activation toward lower potentials in a dose-depen-

The FASEB Journal

VALENTE ET AL.

dent fashion for I696A and W697A, but not as much for the less sensitive mutant R701A. This effect on the voltage dependency of channel activation was not accompanied by a change in the apparent gating valence. These observations suggest that the altered gating mechanism of I696A, W697A, and R701A, which arose primarily from attenuation of stimulus efficacy, was possibly due to a change of events downstream of voltage sensing that lead to pore opening. In addition, mutation of TRP box residues affected heat sensitivity by decreasing the temperature-dependent gating, as indicated by the lower Q10 value. Taken together, these findings further substantiate that the TRP box plays a function in channel gating, and identify I696, W697, and R701 as molecular determinants of channel function. In addition, these data hint to the exciting hypothesis that mutation of these residues might affect the coupling between stimulus sensing and channel gating. The TRP box of TRPV1 is located a few amino acids after the putative channel gate, at the core of the predicted coiled-coil structure of the TRP domain (18). Although caution must be exercised when inferring protein structure from functional assays using sitedirected mutagenesis, our data support a model where the TRP domain folds in a 4-strand, parallel coiled-coil in the homomeric channel (18, 21). However, the folding of this putative structure may show an organization that differs from a typical left-handed coiled-coil, as suggested by the occupancy of the e and g positions by noncharged residues. Taking into consideration these sequence variations and the mutational findings, we have refined our initial structural prediction of this region (18) and provide an improved model consistent with the functional data (Fig. 10B). Inspection of the TRP box sequence in the refined model shows that amino acids I696 and W697 are located at the a and b positions of the putative coiled-coil motif, where they could mediate intersubunit interactions of the fourhelix bundle assembled by TRPV1 subunits (Fig. 10B). Substitution of I696 and/or W697 with the smaller alanine could draw the TRP domains of the channel subunits closer, strengthening the subunit-subunit contacts near the channel gate. This stronger interaction would raise the free energy for stimulus-mediated channel activation. Because of the central position of I696, the model suggests a stronger effect of mutating this residue than altering W697. Note that the higher activation energy of I696A would arise at least in part from stabilization of the closed state of this channel. The slower activation time constants of the voltagegated currents in the presence of capsaicin lend support to such hypothesis. Furthermore, the strong outward rectifying of I696A is compatible with an increase in the stability of the closed state at negative potentials that significantly reduces the open probability. Therefore, the preservation of compatible intersubunit interactions at the level of I696 and W697 appears to be involved, critical for efficient channel gating. The effect of mutating R701 to alanine is more intriguing since this amino acid does not seem to be DETERMINANTS OF TRPV1 CHANNEL GATING

involved in protein-protein interactions within the C end of TRPV1, as are residues I696 or W697 (Fig. 10B). Mutation of the corresponding residue (R998) in TRPM8 produced a similar phenotype, characterized by lower menthol responsiveness and decreased sensitivity to PIP2 modulation (25). Furthermore, a chimeric TRPM8 channel incorporating the TRP domain of TRPV1 identified R701 as a molecular determinant of PIP2 activation but not of menthol gating (24). Although some structural differences between TRPV1 and TRPM8 plausibly exist, collectively, these findings lend support to a model in which R701 would be oriented externally (Fig. 10B), perhaps interacting with the S4 –S5 and/or PIP2, which would be also consistent with a role of this residue in channel gating. A recent study found that R701 could be involved in Ca2⫹dependent desensitization of TRPV1 (37), further substantiating its implication in channel gating. In conclusion, our findings suggest that the TRP box of TRPV1 is a domain that may act as a coiled-coil zipper that could contribute to hold the gate in the closed state, and imply that this motif may operate as a modulator of channel opening. The energy released by the activating stimuli might break intersubunit interactions at the level of the TRP box, favoring pore opening. Mutations that strengthen subunit-subunit interactions in this stretch could increase the energetics of pore opening. Although our results showed that the TRP box plays a role in channel gating, the precise mechanism remains unknown. For instance, our results suggest that this protein region is important for channel gating, although we cannot distinguish which step of the gating process is primarily altered by the mutations. Thus, additional experimental support is needed to uncover the mechanistic contribution of these residues to the functional coupling between stimulus sensing and channel gating. We thank Drs. Rosa Planells and Marco Caprini for critically reading the manuscript; and Reme Torres for cDNA and cRNA preparation, oocyte injection, and manipulation. We are indebted to Dr. Davis Julius (University of California, San Francisco, CA, USA) for the cDNA encoding rat TRPV1. This work was supported by grants from the Spanish Ministry of Education and Science (MEC) (SAF2006 –2580 to A.F.-M. and BFU2005– 03986 to A.G.), the Fundacio´n Ramo´n Areces (to A.F.-M), and the Generalitat Valenciana (GV-ACOMP06/202 to A.F-M. and GV05/076 to A.G.).

REFERENCES 1.

Caterina, M. J., and Julius, D. (2001) The vanilloid receptor: a molecular gateway to the pain pathway. Annu. Rev. Neurosci. 24, 487–517 2. Ramsey, I. S., Delling, M., and Clapham, D. E. (2006) An introduction to TRP channels. Annu. Rev. Physiol. 68, 619 – 647 3. Caterina, M. J., Schumacher, M. A., Tominaga, M., Rosen, T. A., Levine, J. D., and Julius, D. (1997) The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 389, 816 – 824 4. Caterina, M. J., Rosen, T. A., Tominaga, M., Brake, A. J., and Julius, D. (1999) A capsaicin-receptor homologue with a high threshold for noxious heat. Nature 398, 436 – 441

11

5.

6. 7.

8. 9.

10.

11.

12.

13.

14. 15.

16.

17. 18.

19.

20.

21.

22.

12

Voets, T., Droogmans, G., Wissenbach, U., Janssens, A., Flockerzi, V., and Nilius, B. (2004) The principle of temperaturedependent gating in cold- and heat-sensitive TRP channels. Nature 430, 748 –754 Nilius, B., Talavera, K., Owsianik, G., Prenen, J., Droogmans, G., and Voets, T. (2005) Gating of TRP channels: a voltage connection? J. Physiol. 567, 35– 44 Planells-Cases, R., Garcia-Sanz, N., Morenilla-Palao, C., and Ferrer-Montiel, A. (2005) Functional aspects and mechanisms of TRPV1 involvement in neurogenic inflammation that leads to thermal hyperalgesia. Pflu¨gers Arch. 451, 151–159 Scholz, J., and Woolf, C. J. (2002) Can we conquer pain? Nat. Neurosci. 5(Suppl.), 1062–1067 Hoenderop, J. G., Voets, T., Hoefs, S., Weidema, F., Prenen, J., Nilius, B., and Bindels, R. J. (2003) Homo- and heterotetrameric architecture of the epithelial Ca2⫹ channels TRPV5 and TRPV6. EMBO J. 22, 776 –785 Kedei, N., Szabo, T., Lile, J. D., Treanor, J. J., Olah, Z., Iadarola, M. J., and Blumberg, P. M. (2001) Analysis of the native quaternary structure of vanilloid receptor 1. J. Biol. Chem. 276, 28613–28619 Jung, J., Lee, S. Y., Hwang, S. W., Cho, H., Shin, J., Kang, Y. S., Kim, S., and Oh, U. (2002) Agonist recognition sites in the cytosolic tails of vanilloid receptor 1. J. Biol. Chem. 277, 44448 – 44454 Erler, I., Hirnet, D., Wissenbach, U., Flockerzi, V., and Niemeyer, B. A. (2004) Ca2⫹-selective transient receptor potential V channel architecture and function require a specific ankyrin repeat. J. Biol. Chem. 279, 34456 –34463 Chang, Q., Gyftogianni, E., van de Graaf, S. F., Hoefs, S., Weidema, F. A., Bindels, R. J., and Hoenderop, J. G. (2004) Molecular determinants in TRPV5 channel assembly. J. Biol. Chem. 279, 54304 –54311 Hellwig, N., Albrecht, N., Harteneck, C., Schultz, G., Schaefer, M. (2005) Homo- and heteromeric assembly of TRPV channel subunits. J. Cell Sci. 118, 917–928 Arniges, M., Fernandez-Fernandez, J. M., Albrecht, N., Schaefer, M., and Valverde, M. A. (2006) Human TRPV4 channel splice variants revealed a key role of ankyrin domains in multimerization and trafficking. J. Biol. Chem. 281, 1580 –1586 Morenilla-Palao, C., Planells-Cases, R., Garcia-Sanz, N., and Ferrer-Montiel, A. (2004) Regulated exocytosis contributes to protein kinase C potentiation of vanilloid receptor activity. J. Biol. Chem. 279, 25665–25672 Clapham, D. E. (2003) TRP channels as cellular sensors. Nature 426, 517–524 Garcia-Sanz, N., Fernandez-Carvajal, A., Morenilla-Palao, C., Planells-Cases, R., Fajardo-Sanchez, E., Fernandez-Ballester, G., and Ferrer-Montiel, A. (2004) Identification of a tetramerization domain in the C terminus of the vanilloid receptor. J. Neurosci. 24, 5307–5314 Vlachova, V., Teisinger, J., Susankova, K., Lyfenko, A., Ettrich, R., and Vyklicky, L. (2003) Functional role of C-terminal cytoplasmic tail of rat vanilloid receptor 1. J. Neurosci. 23, 1340 –1350 Brauchi, S., Orta, G., Salazar, M., Rosenmann, E., and Latorre, R. (2006) A hot-sensing cold receptor: C-terminal domain determines thermosensation in transient receptor potential channels. J. Neurosci. 26, 4835– 4840 Garcia-Sanz, N., Valente, P., Gomis, A., Fernandez-Carvajal, A., Fernandez-Ballester, G., Viana, F., Belmonte, C., and FerrerMontiel, A. (2007) A role of the transient receptor potential domain of vanilloid receptor I in channel gating. J. Neurosci. 27, 11641–11650 Rohacs, T., and Nilius, B. (2007) Regulation of transient receptor potential (TRP) channels by phosphoinositides. Pflu¨gers Arch. 455, 157–168

Vol. 22

September 2008

23. 24.

25.

26.

27.

28. 29.

30. 31.

32.

33. 34.

35. 36.

37.

38.

39.

Voets, T., and Nilius, B. (2007) Modulation of TRPs by PIPs. J. Physiol. 582, 939 –944 Brauchi, S., Orta, G., Mascayano, C., Salazar, M., Raddatz, N., Urbina, H., Rosenmann, E., Gonzalez-Nilo, F., and Latorre, R. (2007) Dissection of the components for PIP2 activation and thermosensation in TRP channels. Proc. Natl. Acad. Sci. U. S. A. 104, 10246 –10251 Bandell, M., Dubin, A. E., Petrus, M. J., Orth, A., Mathur, J., Hwang, S. W., and Patapoutian, A. (2006) High-throughput random mutagenesis screen reveals TRPM8 residues specifically required for activation by menthol. Nat. Neurosci. 9, 493–500 Ferrer-Montiel, A. V., Sun, W., and Montal, M. (1995) Molecular design of the N-methyl-d-aspartate receptor binding site for phencyclidine and dizolcipine. Proc. Natl. Acad. Sci. U. S. A. 92, 8021– 8025 Caprini, M., Ferroni, S., Planells-Cases, R., Rueda, J., Rapisarda, C., Ferrer-Montiel, A., and Montal, M. (2001) Structural compatibility between the putative voltage sensor of voltage-gated K⫹ channels and the prokaryotic KcsA channel. J. Biol. Chem. 276, 21070 –21076 Lishko, P. V., Procko, E., Jin, X., Phelps, C. B., and Gaudet, R. (2007) The ankyrin repeats of TRPV1 bind multiple ligands and modulate channel sensitivity. Neuron 54, 905–918 Caprini, M., Fava, M., Valente, P., Fernandez-Ballester, G., Rapisarda, C., Ferroni, S., and Ferrer-Montiel, A. (2005) Molecular compatibility of the channel gate and the N terminus of S5 segment for voltage-gated channel activity. J. Biol. Chem. 280, 18253–18264 Liu, B., Hui, K., and Qin, F. (2003) Thermodynamics of heat activation of single capsaicin ion channels VR1. Biophys. J. 85, 2988 –3006 Brauchi, S., Orio, P., and Latorre, R. (2004) Clues to understanding cold sensation: thermodynamics and electrophysiological analysis of the cold receptor TRPM8. Proc. Natl. Acad. Sci. U. S. A. 101, 15494 –15499 Voets, T., Owsianik, G., Janssens, A., Talavera, K., and Nilius, B. (2007) TRPM8 voltage sensor mutants reveal a mechanism for integrating thermal and chemical stimuli. Nat. Chem. Biol. 3, 174 –182 Matta, J. A., and Ahern, G. P. (2007) Voltage is a partial activator of rat thermosensitive TRP channels. J. Physiol. 585, 469 – 482 Rohacs, T., Lopes, C. M., Michailidis, I., and Logothetis, D. E. (2005) PI(4,5)P2 regulates the activation and desensitization of TRPM8 channels through the TRP domain. Nat. Neurosci. 8, 626 – 634 Wang, T., Jiao, Y., and Montell, C. (2005) Dissecting independent channel and scaffolding roles of the Drosophila transient receptor potential channel. J. Cell Biol. 171, 685– 694 Colquhoun, D. (1998) Binding, gating, affinity and efficacy: the interpretation of structure-activity relationships for agonists and of the effects of mutating receptors. Br. J. Pharmacol. 125, 924 –947 Novakova-Tousova, K., Vyklicky, L., Susankova, K., Benedikt, J., Samad, A., Teisinger, J., and Vlachova, V. (2007) Functional changes in the vanilloid receptor subtype 1 channel during and after acute desensitization. Neuroscience 149, 144 –154 Watanabe, H., Vriens, J., Suh, S. H., Benham, C. D., Droogmans, G., and Nilius, B. (2002) Heat-evoked activation of TRPV4 channels in a HEK293 cell expression system and in native mouse aorta endothelial cells. J. Biol. Chem. 277, 47044 – 47051 Xu, H., Ramsey, I. S., Kotecha, S. A., Moran, M. M., Chong, J. A., Lawson, D., Ge, P., Lilly, J., Silos-Santiago, I., Xie, Y., DiStefano, P. S., Curtis, R., and Clapham, D. E. (2002) TRPV3 is a calcium-permeable temperature-sensitive cation channel. Nature 418, 181–186

The FASEB Journal

Received for publication February 5, 2008. Accepted for publication May 15, 2008.

VALENTE ET AL.