Gating of TRP channels: a voltage connection
1 Department of Physiology, Campus Gasthuisberg, KU Leuven, Leuven, Belgium
Abstract
TRP channels represent the main pathways for cation influx in non-excitable cells. Although TRP channels were for a long time considered to be voltage independent, several TRP channels now appear to be weakly voltage dependent with an activation curve extending mainly into the non-physiological positive voltage range. In connection with this voltage dependence, there is now abundant evidence that physical stimuli, such as temperature (TRPV1, TRPM8, TRPV3), or the binding of various ligands (TRPV1, TRPV3, TRPM8, TRPM4), shift this voltage dependence towards physiologically relevant potentials, a mechanism that may represent the main functional hallmark of these TRP channels. This review discusses some features of voltage-dependent gating of TRPV1, TRPM4 and TRPM8. A thermodynamic principle is elaborated, which predicts that the small gating charge of TRP channels is a crucial factor for the large voltage shifts induced by various stimuli. Some structural considerations will be given indicating that, although the voltage sensor is not yet known, the C-terminus may substantially change the voltage dependence of these channels.
, http://www.100md.com
Introduction
The superfamily of TRP channels (‘transient receptor potential’) currently comprises nearly 30 mammalian members, which, based on amino acid homology, can be subdivided into six subfamilies, i.e. TRPC, TRPM, TRPV, TRPA, TRPP and TRPML (Montell et al. 2002a,b; Clapham, 2003; Corey, 2003; Delmas, 2004; Moran et al. 2004). All TRPs form cation-selective channels, and are assembled as homo- or heterotetramers from subunits that contain six putative transmembrane domains (TM) and a pore region between TM5 and TM6. The TRPC (‘Canonical’) and TRPM (‘Melastatin’) subfamilies consist of seven and eight different channels, respectively (i.e. TRPC1–TRPC7 and TRPM1–TRPM8). The TRPV (‘Vanilloid’) subfamily at present comprises six members (TRPV1–TRPV6) whereas the most recently proposed subfamily, TRPA (‘Ankyrin’), has only a single mammalian member (TRPA1). The TRPP (‘Polycystin’) and TRPML (‘Mucolipin’) families are not well characterized, but gain increasing interest because of their involvement in several human diseases. Another TRP subfamily, TRPN (NOMP, NO-Mechano-Potential, assisting sensory neurones in hearing in Drosophila and zebra fish), has so far only been detected in worm, fly and zebra fish, and is proposed to be a mechano-sensing channel (Walker et al. 2000; Sidi et al. 2003).
, 百拇医药
One of the most manifest hallmarks of TRP channels is the huge versatility of activating stimuli, ranging from physical (force, temperature ...) and chemical (ligands, Ca2+ ...) stimuli to protein–protein interaction (e.g. TRPP1–TRPP2).
Recent findings indicate that the gating of several TRP channels shows weak voltage dependence, contradicting the longstanding dogma that the gating of TRP channels is voltage independent. Here we will discuss the relevance of this voltage dependence to the functioning of TRP channels. In particular, we will illustrate how small changes of the free energy of activation of these channels can results in large shifts of their voltage-dependent activation curves, and concomitant gating of these channels.
, 百拇医药
TRP channels: abundant examples of a voltage connection
Close inspection of the current traces in the exciting early reports on the activation of TRPV1 by heat or capsaicin reveals clear signs of time-dependent activation at positive and deactivation at negative potentials, indicative of voltage-dependent gating of this channel (see, e.g. Caterina et al. 1997; Premkumar & Ahern, 2000; Ahern & Premkumar, 2002). It was also obvious that the time- and voltage-dependent activation and deactivation depended on temperature, the type and concentration of agonist, and the PKC-dependent phosphorylation state of the channel (Premkumar & Ahern, 2000; Ahern & Premkumar, 2002; Vlachova et al. 2003). This voltage dependence was intrinsic and not due to voltage-dependent block or unblock by divalent cations (Piper et al. 1999), as has been described in detail for TRPV5 and TRPV6 (Nilius et al. 2000; Voets et al. 2003). Similar activation and deactivation kinetics have also been observed for TRPV3 (Xu et al. 2002; Chung et al. 2004; Chung et al. 2005) and TRPM3 (Grimm et al. 2005). However, the voltage dependence of TRP channels has often been overlooked, because currents were either recorded at a constant holding potential or during a voltage ramp, rather than using voltage step protocols, which are better suited for analysing time- and voltage-dependent features of ion channels.
, 百拇医药
Various compounds that relieve the phosphatidylinositol-4,5-biphosphate (PtdIns(4,5)P2)-mediated inhibition of TRPV1 appear to modulate gating by shifting the voltage dependence of the channels towards more negative potentials (Chuang et al. 2001; Prescott & Julius, 2003). Application of Ca2+ and calmodulin, but not Ca2+ alone, to excised patches clearly shifts the open probability of TRPV1 channels towards more positive potentials (i.e. causing inhibition) (Rosenbaum et al. 2004). The sensitizing effect of ethanol on the activation of TRPV1 by capsaicin (Trevisani et al. 2002) is another illustration of a shift of its activation curve. Similarly, activation for TRPV3 by 2-APB is obviously induced by a shift of activation (Chung et al. 2004; Chung et al. 2005). Very similar results were obtained for TRPM4 (Nilius et al. 2003, 2004, 2005) and TRPM5 (Hofmann et al. 2003; Ullrich et al. 2005). Because these shifts in voltage dependence were often ignored and never analysed, we started a systematic analysis of the voltage-dependent properties of TRPM4 channels by measuring activation and deactivation kinetics, constructing Boltzmann-type activation curves and measuring shifts of activation induced by several tools (Nilius et al. 2003). We have also found that the TRPV1 ligand capsaicin and the TRPM8 ligand menthol induce dramatic shifts in the voltage dependence of these channels (Voets et al. 2004). In this review we summarize the main features of the voltage-dependent gating of TRP channels, and discuss the voltage-dependent shift of activation by various manoeuvres, such as physical stimuli, agonists, endogenous ligands, phosphorylation and PtdIns(4,5)P2, as an important modulator of TRP channel gating.
, 百拇医药
TRP channels: a prototype of voltage dependence
Voltage-dependent activation of ion channels involves the movement of a charged voltage sensor, characterized by an effective gating charge of valence z moving from the inner membrane surface to the outer. If this charge is large, such as for the Shaker K+ channel (approximately 13e, Aggarwal & MacKinnon, 1996), activation occurs in a relatively narrow voltage range (Fig. 1A). The gating charge of TRP channels appears to be much smaller, i.e. about –0.7e for TRPM4, resulting in a relatively shallow activation curve. Under ‘resting’ conditions, the midpoint of the activation curve is mostly at very positive potentials, such that channel openings are very rare at physiological membrane potentials. However, a leftward shift of the activation curve can result in channel activation (Fig. 1B), as will be shown below.
, 百拇医药
A, activation curve of a Shaker K+ channel. Because of the large gating charge, activation occurs in a very narrow voltage range. B, activation curve of a voltage-dependent TRP channel. This curve is shallow, which reflects a small gating charge. The open probability changes over a broad voltage range, which for most TRP channels extends into a range of non-physiological positive potentials. Activation of TRP channels in response to various stimuli results from a shift of their voltage dependence towards negative potentials and opening at physiological potentials. C, activation of TRPM4 channels (inside out patch, 300 μM Ca2+ at the inner side of the membrane) by depolarizing voltage steps. Note deactivation at negative potentials and activation at positive potentials. Patches were held at 0 mV, steps from –100 mV to +180 mV (increment 20 mV) for 400 ms, and then back to –100 mV for 100 ms. D, current–voltage relationships (I–V curves) from the experiment shown in C: steady state I–V curves () show outward rectification, I–V curves from the tail currents saturate at positive prepotentials (). E, voltage dependence of the time constants of current activation obtained from mono-exponential fits. F, voltage dependence of the open probability, Po, of the TRPM4 channel obtained from tail current measurements as shown in C and D. Continuous line is fit with the Boltzmann equation, V= 92 mV a s = 32 mV, i.e. z = 0.75. G, calculation of the rate coefficients and for the two state model from E and F. Data were fitted to y = yo + Aexp(t/) (continuous lines).
, http://www.100md.com
Figure 1C–H illustrates the fingerprints of the voltage-dependent TRPM4 channel activated under conditions of elevated [Ca2+]i. Voltage steps to positive potentials induce activation, whereas steps to negative potentials cause deactivation (Fig. 1C). Some other TRP channels, e.g. TRPM8 or TRPV1, show a similar activation pattern at very positive potentials in the absence of any additional stimulus. The voltage-dependent gating of TRPM4 apparently requires the stimulus of elevated [Ca2+]i, but it cannot fully be ruled out that the activation curve of these channels at low [Ca2+]i is shifted to extreme positive potentials that cannot be reached under patch-clamp conditions. Stepping back to negative potentials induces deactivating tail currents due to channel closure. The typical outwardly rectifying steady-state current–voltage relationship (Fig. 1D, circles) can be accounted for by the voltage-dependent open probability, as assessed from the amplitude of these tail currents at each test potential (Fig. 1D, triangles). In most cases, current kinetics can be fitted adequately by a mono-exponential time course (Fig. 1E) and approximated by a simple closed–open kinetic scheme:
, http://www.100md.com
where and are the rate coefficients for channel opening and closing, respectively. The open probability (Po) can be calculated by fitting the tail current amplitudes (Fig. 1D, triangles) with a sigmoid Boltzmann relationship:
(V is the potential of half-maximal activation, s = RT/ zF, the slope factor with z the valence of the gating charge) and normalizing the measured tail current amplitudes to the maximal current amplitude Imax (Fig. 1F):
, 百拇医药
The instantaneous I–V curve, derived from the amplitude of tail currents during steps to different potentials following maximal activation by a depolarizing prepulse, is approximately linear, indicative of an Ohmic conductance of the open channel. Rate coefficients for opening and closing can readily be calculated from Po and ( = Po/; = (1 – Po)/) (Fig. 1G). Very similar voltage-dependent characteristics have been observed for at least TRPV1, TRPV3, TRPM3, TRPM4, TRPM5 and TRPM8.
, http://www.100md.com
Shift of the activation curve: a frequently used trick of TRP channels
There is convincing evidence that activation of TRP channels is often accompanied by a shift of their voltage-dependent gating, as illustrated with a few examples in Fig. 2. The first example concerns activation of TRP channels by changes in temperature. Various TRP channels (so-called ‘thermoTRPs’) are involved in temperature sensing, not only in sensory nerve fibres originating from the dorsal root or trigeminal ganglia but also in keratinocytes (Peier et al. 2002; Patapoutian et al. 2003; Moqrich et al. 2005). Four TRPV channels are activated upon heating, and are responsible for detection of temperatures that we experience as either warmth, heat or noxious heat (Patapoutian et al. 2003). Two other TRP channels, TRPM8 and TRPA1, are cold-activated, sensing cool and noxious cold, respectively. We have recently shown that the temperature sensitivity of TRPV1 and TRPM8 is modulated by the transmembrane voltage, and that changes in ambient temperature result in graded shifts of the voltage dependence of activation (Voets et al. 2004). Figure 2A shows that the activation curve of TRPV1 is shifted leftward by nearly 200 mV if the temperature is increased from 17 to 42°C (Voets et al. 2004). On the other hand, cooling from 37°C to 15°C induces a leftward shift of 170 mV for TRPM8 (Fig. 2B, Voets et al. 2004). The TRPV1-specific ligand capsaicin shifts the TRPV1 activation curve at constant temperature (the maximal capsaicin induced shift at 24°C is 200 mV with an EC50 of 30 nM, Fig. 2C). Similarly, the TRPM8 activator menthol induces a leftward shift of the activation curve (Fig. 2D, the maximal menthol induced shift at 34°C is 200 mV with an EC50 of 30 μM, Voets et al. 2004). Figure 2E illustrates how increasing the intracellular Ca2+ concentration shifts the activation curve of TRPM4 towards more negative potentials (Fig. 2E, Nilius et al. 2004). This activation is followed by a fast desensitization, which is accompanied by (or is due to) a dramatic rightward shift of the activation curve. Importantly, interventions that sensitize TRPM4 to Ca2+ shift the activation curve towards negative potentials, e.g. phosphorylation via protein kinase C activation or addition of calmodulin to excised patches (Nilius et al. 2005). Vice versa, desensitization of TRPM4 to Ca2+, which can be achieved by deletion of putative Ca2+/calmodulin binding sites in the channel C-terminus or by coexpression of the dominant negative calmodulin mutant CaM1,2,3,4 (with the four EF-hand Ca2+ bindings sites deleted), shifts channel activation towards positive potentials (Nilius et al. 2005). Interestingly, neutralization of positive charges in the fourth transmembrane spanning helix (S4), which has some homology with the voltage sensing S4 helix of Shaker, shifts the activation curve towards positive potentials (Fig. 2F, K. Talavera & B. Nilius, unpublished).
, 百拇医药
A, voltage dependence of the open probability of TRPV1 channels at 17°C () and 42°C (). The inset shows the respective current families for a voltage step programme (holding potential 0 mV, test steps from –120 to +160 mV, increment 40 mV), back step to +60 mV. Note the dramatic shift towards negative potentials from +198 to –33 mV. B, same protocol as in A, but for TRPM8 at 37°C () and 15°C (). Note the leftward shift of the activation curve. V changed from +203 to 53 mV. C, same voltage protocol as in A. TRPV1 is activated by 100 nM capsaicin () causing a leftward voltage shift of from +95 to –22 mV. Temperature was 24°C. D, 30 μM menthol () induced a leftward shift from 203 to 74 mV of the TRPM8 activation curve at 34°C (circles are controls, same protocol as in A). E, TRPM4 was activated in inside-out patches by different [Ca2+]i. Voltage protocol consists of steps from –100 to +160 mV (increment 40 mV) from a holding potential of 0 mV. A change from 10 μM [Ca2+]i () to 1000 μM [Ca2+]i () induced a leftward shift of the activation curve from 162 to 68 mV. F, TRPM4 activation in the presence of 300 μM [Ca2+]i. Neutralizing the positive charge K919 caused a shift of V from +60 (wild type TRPM4) to +192 mV (K919A mutant). Holding potential was 0 mV, steps from –25 to 250 (mV). (Calibration bars in the insets are identical for examples shown in each panel). (A–D are adapted from Voets et al. (2004) with permission from Nature, http://www.nature.com/; E is adapted from Nilius et al. (2004).)
, 百拇医药
A thermodynamic approach of the voltage-dependent shift of channel activation
We have recently applied a simple thermodynamic formalism to describe the shifts in voltage dependence due to changes in temperature (Voets et al. 2004). For the simple gating scheme (1), the open probability of the channel is given by:
where z is the gating charge, F is Faraday's constant, R is the gas constant, T is the absolute temperature and G is the free energy difference between open and closed states of the channel, i.e.
, 百拇医药
where H and S are the differences in enthalpy and entropy between open (o) and closed (c) states. The potential V for half maximal activation (see eqn (2)) is determined by the condition: G – zFV = 0. Hence:
The gating charge of TRP channels is much smaller than that of classical voltage-gated channels. This implies that any factor inducing changes in H or S will provoke larger larger shifts of the activation curve of TRP channels.
, 百拇医药
The rate coefficients and for the simple closed–open model can be calculated from Po and the time constants of current activation using the equations Po = /( + ) and = 1/( + ), and are related to the differences in free energy between an intermediate activated channel state and the closed or open state of the channel Gc and Go by the equations (derived from Eyring's rate theory):
where is the fraction of the electric field that the gating charge z has to cross in moving from the closed to the intermediate activated state and f is a frequency factor, representing the number of potential transitions per unit time. This prefactor, which is not known, is assumed to be independent of temperature at least in the investigated temperature range. The exponential terms give the probabilities that the thermal energy exceeds the energy required to induce the conformational changes of the channel. Gc and Go are given by:
, 百拇医药
It follows from these equations that the slopes of ln() and ln() as a function of 1/T (Arrhenius plots) are given by (– Hc + z FV)/R and (– Hc – z(1 – )FV)/R, respectively.
This formalism has been used to describe the temperature-induced voltage shifts of the activation of TRPV1 and TRPM8 (Fig. 2A and B). The Arrhenius plots for TRPM8 reveal a weak temperature dependence of the opening rates, but a steep temperature dependence of the closing rate (Fig. 3A). The plots at different voltages are almost parallel in the investigated temperature range, indicating that the contribution of the voltage terms to the slopes is small compared to that of the enthalpy terms Hc and Ho in the physiological voltage range. The linear trend of these plots further implies that Hc and Ho and therefore also H = Hc – Ho can be considered to be independent of temperature. The values for Ho and H are quite large, suggesting that the conformational changes during channel closure are accompanied by a large energetic increase. On the other hand, the reverse has been found for TRPV1 channels, i.e. temperature strongly affects the opening rate but has only minor effects on the closing rate. The trends of the corresponding Arrhenius plots are also linear, with slopes that are virtually independent of the applied voltage (Fig. 3B).
, 百拇医药
A, Arrhenius plots for the rate constants and of TRPM8 for the two state model at –80 (), +80 () and +160 mV (). The slope corresponds to the enthalpy. The enthalpy for channel opening is small, corresponding to a Q10 values of 1.2, but is large for channel closing (corresponding Q10 is 9.4). From the shifts in V, the following data were obtained: Ho = 13 kJ mol–1, Hc = 170 kJ mol–1, So = –166 J mol–1 K–1, Sc = 392 J mol–1 K–1. B, same analysis as in A but for TRPV1. Note that the enthalpy for channel opening is much higher (corresponding Q10 is 14.8) than for channel closing (Q10 is 1.35). Considering the shifts in V, the following data were obtained: Ho = 205 kJ mol–1, Hc = 21 kJ mol–1, So = 468 J mol–1 K–1, Sc = –130 J mol–1 K–1. C, shift of the potential for half-maximal activation of TRPM8 as a function of temperature. With V = (H – ST)/zF, the positive slope corresponds to the negative change in entropy. Menthol induced a parallel shift of this curve toward more negative potentials, likely indicating a change in enthalpy. (A and B are adapted from Voets et al. (2004) with permission from Nature, http://www.nature.com/.)
, 百拇医药
Figure 3C shows that the plot of V as a function of T is linear over a wide temperature range for TRPM8. Its slope, i.e. –S/zF according to eqn (5), and therefore also S apparently does not depend on temperature. The value of S is –558 J mol–1 K–1 for TRPM8 and 598 J mol–1 K–1 for TRPV1. Note that the changes in enthalpy and entropy for the warm-sensing TRPV1 channel have the same sign, implying that the difference in free energy G = H – TS between closed and open state, which determines open probability, can be small despite the large values of H and S. In other words, the large change in enthalpy can be compensated by a large change in entropy. A similar argument applies for TRPM8, for which both H and S are negative. This makes it possible that the relatively small energy provided by the membrane potential is able to significantly modulate the open probability in the physiological range of potentials. From the slope of the V versus T plot, we can calculate a shift –7 mV °C–1 (see also legend Fig. 3A) for TRPM8 and of 9 mV °C–1 for TRPV1. More general, the sign of S = So – Sc determines the direction in which a change in temperature shifts the voltage-dependent activation of the channel.
, 百拇医药
To resolve the question why cooling activates a channel like TRPM8 and warming activates a channel like TRPV1, we have calculated the change of Po with temperature T from eqn (3). It can be readily calculated that:
from which it is evident that the sign of Po/ T, which determines whether a channel is warm or cold activated, is the same as that of (H – zFV). The voltage term zFV amounts to about 10 kJ mol–1 at 100 mV, which is much smaller (in absolute value) than the value of H (in the order of 100–200 kJ mol–1 for TRPV1 and TRPM8). Since H is positive for TRPV1 and negative for TRPM8, these channels are warm- and cold-activated, respectively. An intriguing inference is that the term (H – zFV) changes sign at a potential V = H/zF, thereby switching from a warm-activated TRPV1 channel to a cold-activated channel at extreme positive potentials (V > 2.5 V), or from a cold-activated TRPM8 channel to a warm-activated channel for V < –2.1 V.
, 百拇医药
Voltage shift of channel activation by ‘gating’ modifiers
It has also been observed that some ligands cause prominent shifts of the voltage-dependent channel activation at constant temperature (see Fig. 2C and D). Figure 3E gives as an example of the voltage shift of TRPM8 by menthol at various temperatures. This shift of the V versus T curve is roughly parallel, indicating that, according to eqn (5), menthol mainly affects the enthalpy rather than the entropy. Figure 4 illustrates some examples of the shift of the voltage-dependent activation induced by agents that modify gating of TRPM4, the first TRP channel whose intrinsic voltage dependence was analysed in detail (Nilius et al. 2003). From comparison with the KscA channel, one might expect that the intrinsic voltage sensor is localized in the fourth membrane spanning -helix (S4) of TRPM4, TRPM5 and TRPM8.
, 百拇医药
A, a structural model of the C-terminus of TRPM4 from S1044 to D1214 (see text for more details), viewed using DS ViewerPro 5.0 software (Accelrys Inc., San Diego, CA, USA) (parental structure score 1.29, serine acetyltransferase-apoenzyme, chain A (1smmA) score). Indicated are the two putative Ca2+–calmodulin binding sites in blue (for details see Nilius et al. 2005). -Helices are shown in red, turns in grey. The putative binding site for decavanadate, R1136ARDKR1141) is shown in cyan (for details see Nilius et al. 2004). In the modelled coiled-coil region (1134–1154 and 1158–1185, centre of the figure) are two PKC phosphorylation sites indicated in yellow (Nilius et al. 2005). B, deletion of the calmodulin binding site induced a rightward shift of the activation curve of TRPM4. The inset shows current traces from voltage steps for controls () and the deletion mutant of the first C-terminal Ca2+–calmodulin binding site A1076–S1098 (). [Ca2+]i was 100 μM, voltage steps from –100 to +200 mV, 20 mV increment for the control and from –100 to +260 mV, increment +40 mV for the deletion mutant (inside out patches). Note the shift of V from +112 to +235 mV. C, effects of decavanadate (DV) on the activation curve of TRPM4. The inset shows current traces from inside patches with [Ca2+]i of 300 μM (controls, and after application of 10 μM DV, ). Activation curves show a shift of more than 200 mV toward negative potentials in the presence of DV. Note the very shallow voltage dependence, and the loss of activation and deactivation kinetics (Nilius et al. 2004). D, mutation of the C-terminal putative PKC phosphorylation sites S1145 and S1152 to aspartates induces a leftward shift of the TRPM4 activation curves. The inset shows current traces from inside out patches with wild-type TRPM4 () and with the mutation S1145D (). Shift is from V of +124 to +12 mV (100 μM [Ca2+]i, steps from –75 to +200, +150 mV, respectively, holding potential 0 mV). This shift is in agreement with a sensitizing effect of PKC phosphorylation of TRPM4 (Nilius et al. 2005). (C, modified from Nilius et al. (2004).).
, http://www.100md.com
We have shown that modifications to the C-terminus of TRPM4 can induce alterations of the voltage dependence of this channel (Nilius et al. 2005). We obtained a putative 3D structure of the C-terminus of TRPM4 using the automated comparative protein modelling with prediction of structure of the ROBETTA server (‘ab initio’ and comparative modelling). We used as template the serine acetyltransferase-apoenzyme, chain A, a protein with a coiled coil domain as described for the C-terminus of TRPM4 (Perraud et al. 2003). In addition, we used previously obtained information of the C-terminal localization of Ca2+–calmodulin binding sites in the positions A1076–S1098 and R1095–E1130. Figure 4A shows the results of this structural approach. Indicated are the two putative Ca2+–calmodulin binding sites as helical structures, two coiled coil regions (1134–1154 and 1158–1185), which contain the two functionally important PKC phosphorylation sites (Nilius et al. 2005) and the putative decavanadate (DV) binding site R1136ARDKR1141 (Nilius et al. 2004). We propose that these indicated regions provide sensitive structural elements that may modify the voltage sensing of TRPM4.
, http://www.100md.com
Mutation or deletion of putative Ca2+–calmodulin binding sites in the C-terminus induce a large shift of the voltage-dependent activation towards positive potentials (Fig. 4B, Nilius et al. 2005). Vice versa, application of calmodulin to the inner side of cell free patches diminished channel desensitization and induces a shift towards negative potentials (B. Nilius, unpublished and Nilius et al. 2005). Figure 4C illustrates that DV induces a dramatic shift in the voltage dependence and decrease the voltage sensitivity of TRPM4 when applied to the inner side of excised patches (see Fig. 1C–H). Tail currents deactivate rapidly and almost completely in the absence of DV, but more slowly and incomplete in the presence of DV. DV, a compound with six negative charges, is widely used as a tool to interact with ATP binding in transport ATPases (Toyoshima et al. 2000; Pezza et al. 2002; Clausen et al. 2003; Tiago et al. 2004). We have therefore searched for a motif in TRPM4 similar to the FSRDRK motif in SERCA (Toyoshima et al. 2000). Such a motif was identified in the C-terminal coiled-coil region as a cluster of positive charges, R1136ARDKR1141 (R/K motif, see also Fig. 4A). Deletion of this cluster or replacement of the C-terminus with that of TRPM5, which does not contain this motif, eliminate the effect of DV on the voltage-dependent shift of TRPM4 activation (Nilius et al. 2004), strongly indicating that the site of DV action resides in the C-terminus of TRPM4. Interestingly, this site shows some similarities with the pleckstrin domain in phospholipases C, which mediates interaction with second messenger lipids such as PIP2 (Harlan et al. 1994, 1995). Interestingly, phosphorylation of the two putative PKC sites (S1145 and S1152) increased the apparent Ca2+ sensitivity of TRPM4 (Nilius et al. 2005). Mutants that mimic the phosphorylated state (S1145D and S1152D) displayed a delayed desensitization and a remarkable shift of the voltage dependence of activation towards negative potentials. (V = 56 ± 11 mV for the S1145D mutant, V = 65 ± 19 mV for the S1152D mutant, B Nilius & J. Prenen, unpublished).
, http://www.100md.com
All these examples clearly show that TRP channels perform a variety of important physiological functions such as temperature sensing, Ca2+ sensing, and detecting natural or endogenous compounds by shifting their activation curve towards less positive potentials. This shift is supported by the relatively small gating charge of TRP channels (see eqn (5)). Thus, the here-described unique voltage dependence of TRP channels induces equally unique gating properties, which are of physiological importance rather than being an epiphenomenon based on structural similarities with voltage-gated channels.
, 百拇医药
Conclusions
Without doubt, temperature sensing of several thermoTRPs is connected to a shift of their voltage dependence. This has been shown in detail for TRPV1 and TRPM8 (Brauchi et al. 2004; Voets et al. 2004), and more recently for TRPV3 (Chung et al. 2005). The small gating charge of TRP channels compared to that of classical voltage-gated channels could lie at the basis of the large shifts of their voltage-dependent activation curves, and may be essential for their gating versatility. We postulate that many physiological stimuli may exploit this feature and activate TRPs by shifting their voltage dependence from a physiologically uninteresting voltage range into a functionally relevant voltage window by inducing small changes in the Gibbs free energy of the channel.
, http://www.100md.com
Footnotes
This report was presented at The Journal of Physiology Symposium on TRP channels: physiological genomics and proteomics, San Diego, CA, USA, 5 April 2005. It was commissioned by the Editorial Board and reflects the views of the authors.
References
Aggarwal SK & MacKinnon R (1996). Contribution of the S4 segment to gating charge in the Shaker K+ channel. Neuron 16, 1169–1177.
, 百拇医药
Ahern GP & Premkumar LS (2002). Voltage-dependent priming of rat vanilloid receptor: effects of agonist and protein kinase C activation. J Physiol 545, 441–451.
Brauchi S, Orio P & 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.
Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD & Julius D (1997). The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 389, 816–824.
, 百拇医药
Chuang HH, Prescott ED, Kong H, Shields S, Jordt SE, Basbaum AI, Chao MV & Julius D (2001). Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns (4,5),P2-mediated inhibition. Nature 411, 957–962.
Chung MK, Guler AD & Caterina MJ (2005). Biphasic currents evoked by chemical or thermal activation of the heat-gated ion channel, TRPV3. J Biol Chem 280, 15928–15941.
Chung MK, Lee H, Mizuno A, Suzuki M & Caterina MJ (2004). 2-Aminoethoxydiphenyl borate activates and sensitizes the heat-gated ion channel TRPV3. J Neurosci 24, 5177–5182.
, http://www.100md.com
Clapham DE (2003). TRP channels as cellular sensors. Nature 426, 517–524.
Clausen JD, McIntosh DB, Vilsen B, Woolley DG & Andersen JP (2003). Importance of conserved N-domain residues Thr441, Glu442, Lys515, Arg560, and Leu562 of sarcoplasmic reticulum Ca2+-ATPase for MgATP binding and subsequent catalytic steps. Plasticity of the nucleotide-binding site. J Biol Chem 278, 20245–20258.
Corey DP (2003). New TRP channels in hearing and mechanosensation. Neuron 39, 585–588.
, http://www.100md.com
Delmas P (2004). Polycystins: from mechanosensation to gene regulation. Cell 118, 145–148.
Grimm C, Kraft R, Schultz G & Harteneck C (2005). Activation of the melastatin-related cation channel TRMP3 by D-erythro-sphingosine. Mol Pharm 67, 798–805.
Harlan JE, Hajduk PJ, Yoon HS & Fesik SW (1994). Pleckstrin homology domains bind to phosphatidylinositol-4,5-bisphosphate. Nature 371, 168–170.
, 百拇医药
Harlan JE, Yoon HS, Hajduk PJ & Fesik SW (1995). Structural characterization of the interaction between a pleckstrin homology domain and phosphatidylinositol 4,5-bisphosphate. Biochemistry 34, 9859–9864.
Hofmann T, Chubanov V, Gudermann T & Montell C (2003). TRPM5 is a voltage-modulated and Ca2+-activated monovalent selective cation channel. Current Biol 13, 1153–1158.
Montell C, Birnbaumer L & Flockerzi V (2002a). The TRP channels, a remarkable functional family. Cell 108, 595–598.
, http://www.100md.com
Montell C, Birnbaumer L, Flockerzi V, Bindels RJ, Bruford EA, Caterina MJ, Clapham D, Harteneck C, Heller S, Julius D, Kojima I, Mori Y, Penner R, Prawitt D, Scharenberg AM, Schultz G, Shimizu S & Zhu MX (2002b). A unified nomenclature for the superfamily of TRP cation channels. Mol Cell 9, 229–231.
Moqrich A, Hwang SW, Earley TJ, Petrus MJ, Murray AN, Spencer KS, Andahazy M, Story GM & Patapoutian A (2005). Impaired thermosensation in mice lacking TRPV3, a heat and camphor sensor in the skin. Science 307, 1468–1472.
, 百拇医药
Moran MM, Xu H & Clapham DE (2004). TRP ion channels in the nervous system. Current Opin Neurobiol 14, 362–369.
Nilius B, Prenen J, Droogmans G, Voets T, Vennekens R, Freichel M, Wissenbach U & Flockerzi V (2003). Voltage dependence of the Ca2+-activated cation channel TRPM4. J Biol Chem 278, 30813–30820.
Nilius B, Prenen J, Janssens A, Voets T & Droogmans G (2004). Decavanadate modulates gating of TRPM4 cation channels. J Physiol 560, 753–765.
, 百拇医药
Nilius B, Prenen J, Tang J, Wang C, Owsianik G, Janssens A, Voets T & Zhu MX (2005). Regulation of the Ca2+ sensitivity of the nonselective cation channel TRPM4. J Biol Chem 280, 6423–6433.
Nilius B, Vennekens R, Prenen J, Hoenderop JG, Bindels RJ & Droogmans G (2000). Whole-cell and single channel monovalent cation currents through the novel rabbit epithelial Ca2+ channel ECaC. J Physiol 527, 239–248.
Patapoutian A, Peier AM, Story GM & Viswanath V (2003). ThermoTRP channels and beyond: mechanisms of temperature sensation. Nature Rev Neurosci 4, 529–539.
, http://www.100md.com
Peier AM, Reeve AJ, Andersson DA, Moqrich A, Earley TJ, Hergarden AC, Story GM, Colley S, Hogenesch JB, McIntyre P, Bevan S & Patapoutian A (2002). A heat-sensitive TRP channel expressed in keratinocytes. Science 296, 2046–2049.
Perraud AL, Schmitz C & Scharenberg AM (2003). TRPM2 Ca2+ permeable cation channels: from gene to biological function. Cell Calcium 33, 519–531.
Pezza RJ, Villarreal MA, Montich GG & Argarana CE (2002). Vanadate inhibits the ATPase activity and DNA binding capability of bacterial MutS. A structural model for the vanadate–MutS interaction at the Walker A motif. Nucl Acids Res 30, 4700–4708.
, http://www.100md.com
Piper AS, Yeats JC, Bevan S & Docherty RJ (1999). A study of the voltage dependence of capsaicin-activated membrane currents in rat sensory neurones before and after acute desensitization. J Physiol 518, 721–733.
Premkumar LS & Ahern GP (2000). Induction of vanilloid receptor channel activity by protein kinase C. Nature 408, 985–990.
Prescott ED & Julius D (2003). A modular PIP2 binding site as a determinant of capsaicin receptor sensitivity. Science 300, 1284–1288.
, http://www.100md.com
Rosenbaum T, Gordon-Shaag A, Munari M & Gordon SE (2004). Ca2+/calmodulin modulates TRPV1 activation by capsaicin. J General Physiol 123, 53–62.
Sidi S, Friedrich RW & Nicolson T (2003). NompC TRP channel required for vertebrate sensory hair cell mechanotransduction. Science 301, 96–99.
Tiago T, Aureliano M & Gutierrez-Merino C (2004). Decavanadate binding to a high affinity site near the myosin catalytic centre inhibits F-actin-stimulated myosin ATPase activity. Biochemistry 43, 5551–5561.
, 百拇医药
Toyoshima C, Nakasako M, Nomura H & Ogawa H (2000). Crystal structure of the calcium pump of sarcoplasmic reticulum at 2.6 resolution. Nature 405, 647–655.
Trevisani M, Smart D, Gunthorpe MJ, Tognetto M, Barbieri M, Campi B, Amadesi S, Gray J, Jerman JC, Brough SJ, Owen D, Smith GD, Randall AD, Harrison S, Bianchi A, Davis JB & Geppetti P (2002). Ethanol elicits and potentiates nociceptor responses via the vanilloid receptor-1. Nature Neursci 5, 546–551.
, http://www.100md.com
Ullrich ND, Voets T, Prenen J, Vennekens R, Talavera K, Droogmans G & Nilius B (2005). Comparison of functional properties of the Ca2+-activated cation channels TRPM4 and TRPM5 from mice. Cell Calcium 37, 267–278.
Vlachova V, Teisinger J, Suankova K, Lyfenko A, Ettrich R & Vyklicky L (2003). Functional role of C-terminal cytoplasmic tail of rat vanilloid receptor 1. J Neurosci 23, 1340–1350.
Voets T, Droogmans G, Wissenbach U, Janssens A, Flockerzi V & Nilius B (2004). The principle of temperature-dependent gating in cold- and heat-sensitive TRP channels. Nature 430, 748–754.
, 百拇医药
Voets T, Janssens A, Prenen J, Droogmans D & Nilius G (2003). Mg2+-dependent gating and strong inward rectification of the cation channel TRPV6. J General Physiol 121, 245–260.
Walker RG, Willingham AT & Zuker CS (2000). A Drosophila mechanosensory transduction channel. Science 287, 2229–2234.
Xu HX, Ramsey IS, Kotecha SA, Moran MM, Chong JHA, Lawson D, Ge P, Lilly J, Silos Santiago I, Xie Y, DiStefano PS, Curtis R & Clapham DE (2002). TRPV3 is a calcium-permeable temperature-sensitive cation channel. Nature 418, 181–186., http://www.100md.com(Bernd Nilius, Karel Talav)
Abstract
TRP channels represent the main pathways for cation influx in non-excitable cells. Although TRP channels were for a long time considered to be voltage independent, several TRP channels now appear to be weakly voltage dependent with an activation curve extending mainly into the non-physiological positive voltage range. In connection with this voltage dependence, there is now abundant evidence that physical stimuli, such as temperature (TRPV1, TRPM8, TRPV3), or the binding of various ligands (TRPV1, TRPV3, TRPM8, TRPM4), shift this voltage dependence towards physiologically relevant potentials, a mechanism that may represent the main functional hallmark of these TRP channels. This review discusses some features of voltage-dependent gating of TRPV1, TRPM4 and TRPM8. A thermodynamic principle is elaborated, which predicts that the small gating charge of TRP channels is a crucial factor for the large voltage shifts induced by various stimuli. Some structural considerations will be given indicating that, although the voltage sensor is not yet known, the C-terminus may substantially change the voltage dependence of these channels.
, http://www.100md.com
Introduction
The superfamily of TRP channels (‘transient receptor potential’) currently comprises nearly 30 mammalian members, which, based on amino acid homology, can be subdivided into six subfamilies, i.e. TRPC, TRPM, TRPV, TRPA, TRPP and TRPML (Montell et al. 2002a,b; Clapham, 2003; Corey, 2003; Delmas, 2004; Moran et al. 2004). All TRPs form cation-selective channels, and are assembled as homo- or heterotetramers from subunits that contain six putative transmembrane domains (TM) and a pore region between TM5 and TM6. The TRPC (‘Canonical’) and TRPM (‘Melastatin’) subfamilies consist of seven and eight different channels, respectively (i.e. TRPC1–TRPC7 and TRPM1–TRPM8). The TRPV (‘Vanilloid’) subfamily at present comprises six members (TRPV1–TRPV6) whereas the most recently proposed subfamily, TRPA (‘Ankyrin’), has only a single mammalian member (TRPA1). The TRPP (‘Polycystin’) and TRPML (‘Mucolipin’) families are not well characterized, but gain increasing interest because of their involvement in several human diseases. Another TRP subfamily, TRPN (NOMP, NO-Mechano-Potential, assisting sensory neurones in hearing in Drosophila and zebra fish), has so far only been detected in worm, fly and zebra fish, and is proposed to be a mechano-sensing channel (Walker et al. 2000; Sidi et al. 2003).
, 百拇医药
One of the most manifest hallmarks of TRP channels is the huge versatility of activating stimuli, ranging from physical (force, temperature ...) and chemical (ligands, Ca2+ ...) stimuli to protein–protein interaction (e.g. TRPP1–TRPP2).
Recent findings indicate that the gating of several TRP channels shows weak voltage dependence, contradicting the longstanding dogma that the gating of TRP channels is voltage independent. Here we will discuss the relevance of this voltage dependence to the functioning of TRP channels. In particular, we will illustrate how small changes of the free energy of activation of these channels can results in large shifts of their voltage-dependent activation curves, and concomitant gating of these channels.
, 百拇医药
TRP channels: abundant examples of a voltage connection
Close inspection of the current traces in the exciting early reports on the activation of TRPV1 by heat or capsaicin reveals clear signs of time-dependent activation at positive and deactivation at negative potentials, indicative of voltage-dependent gating of this channel (see, e.g. Caterina et al. 1997; Premkumar & Ahern, 2000; Ahern & Premkumar, 2002). It was also obvious that the time- and voltage-dependent activation and deactivation depended on temperature, the type and concentration of agonist, and the PKC-dependent phosphorylation state of the channel (Premkumar & Ahern, 2000; Ahern & Premkumar, 2002; Vlachova et al. 2003). This voltage dependence was intrinsic and not due to voltage-dependent block or unblock by divalent cations (Piper et al. 1999), as has been described in detail for TRPV5 and TRPV6 (Nilius et al. 2000; Voets et al. 2003). Similar activation and deactivation kinetics have also been observed for TRPV3 (Xu et al. 2002; Chung et al. 2004; Chung et al. 2005) and TRPM3 (Grimm et al. 2005). However, the voltage dependence of TRP channels has often been overlooked, because currents were either recorded at a constant holding potential or during a voltage ramp, rather than using voltage step protocols, which are better suited for analysing time- and voltage-dependent features of ion channels.
, 百拇医药
Various compounds that relieve the phosphatidylinositol-4,5-biphosphate (PtdIns(4,5)P2)-mediated inhibition of TRPV1 appear to modulate gating by shifting the voltage dependence of the channels towards more negative potentials (Chuang et al. 2001; Prescott & Julius, 2003). Application of Ca2+ and calmodulin, but not Ca2+ alone, to excised patches clearly shifts the open probability of TRPV1 channels towards more positive potentials (i.e. causing inhibition) (Rosenbaum et al. 2004). The sensitizing effect of ethanol on the activation of TRPV1 by capsaicin (Trevisani et al. 2002) is another illustration of a shift of its activation curve. Similarly, activation for TRPV3 by 2-APB is obviously induced by a shift of activation (Chung et al. 2004; Chung et al. 2005). Very similar results were obtained for TRPM4 (Nilius et al. 2003, 2004, 2005) and TRPM5 (Hofmann et al. 2003; Ullrich et al. 2005). Because these shifts in voltage dependence were often ignored and never analysed, we started a systematic analysis of the voltage-dependent properties of TRPM4 channels by measuring activation and deactivation kinetics, constructing Boltzmann-type activation curves and measuring shifts of activation induced by several tools (Nilius et al. 2003). We have also found that the TRPV1 ligand capsaicin and the TRPM8 ligand menthol induce dramatic shifts in the voltage dependence of these channels (Voets et al. 2004). In this review we summarize the main features of the voltage-dependent gating of TRP channels, and discuss the voltage-dependent shift of activation by various manoeuvres, such as physical stimuli, agonists, endogenous ligands, phosphorylation and PtdIns(4,5)P2, as an important modulator of TRP channel gating.
, 百拇医药
TRP channels: a prototype of voltage dependence
Voltage-dependent activation of ion channels involves the movement of a charged voltage sensor, characterized by an effective gating charge of valence z moving from the inner membrane surface to the outer. If this charge is large, such as for the Shaker K+ channel (approximately 13e, Aggarwal & MacKinnon, 1996), activation occurs in a relatively narrow voltage range (Fig. 1A). The gating charge of TRP channels appears to be much smaller, i.e. about –0.7e for TRPM4, resulting in a relatively shallow activation curve. Under ‘resting’ conditions, the midpoint of the activation curve is mostly at very positive potentials, such that channel openings are very rare at physiological membrane potentials. However, a leftward shift of the activation curve can result in channel activation (Fig. 1B), as will be shown below.
, 百拇医药
A, activation curve of a Shaker K+ channel. Because of the large gating charge, activation occurs in a very narrow voltage range. B, activation curve of a voltage-dependent TRP channel. This curve is shallow, which reflects a small gating charge. The open probability changes over a broad voltage range, which for most TRP channels extends into a range of non-physiological positive potentials. Activation of TRP channels in response to various stimuli results from a shift of their voltage dependence towards negative potentials and opening at physiological potentials. C, activation of TRPM4 channels (inside out patch, 300 μM Ca2+ at the inner side of the membrane) by depolarizing voltage steps. Note deactivation at negative potentials and activation at positive potentials. Patches were held at 0 mV, steps from –100 mV to +180 mV (increment 20 mV) for 400 ms, and then back to –100 mV for 100 ms. D, current–voltage relationships (I–V curves) from the experiment shown in C: steady state I–V curves () show outward rectification, I–V curves from the tail currents saturate at positive prepotentials (). E, voltage dependence of the time constants of current activation obtained from mono-exponential fits. F, voltage dependence of the open probability, Po, of the TRPM4 channel obtained from tail current measurements as shown in C and D. Continuous line is fit with the Boltzmann equation, V= 92 mV a s = 32 mV, i.e. z = 0.75. G, calculation of the rate coefficients and for the two state model from E and F. Data were fitted to y = yo + Aexp(t/) (continuous lines).
, http://www.100md.com
Figure 1C–H illustrates the fingerprints of the voltage-dependent TRPM4 channel activated under conditions of elevated [Ca2+]i. Voltage steps to positive potentials induce activation, whereas steps to negative potentials cause deactivation (Fig. 1C). Some other TRP channels, e.g. TRPM8 or TRPV1, show a similar activation pattern at very positive potentials in the absence of any additional stimulus. The voltage-dependent gating of TRPM4 apparently requires the stimulus of elevated [Ca2+]i, but it cannot fully be ruled out that the activation curve of these channels at low [Ca2+]i is shifted to extreme positive potentials that cannot be reached under patch-clamp conditions. Stepping back to negative potentials induces deactivating tail currents due to channel closure. The typical outwardly rectifying steady-state current–voltage relationship (Fig. 1D, circles) can be accounted for by the voltage-dependent open probability, as assessed from the amplitude of these tail currents at each test potential (Fig. 1D, triangles). In most cases, current kinetics can be fitted adequately by a mono-exponential time course (Fig. 1E) and approximated by a simple closed–open kinetic scheme:
, http://www.100md.com
where and are the rate coefficients for channel opening and closing, respectively. The open probability (Po) can be calculated by fitting the tail current amplitudes (Fig. 1D, triangles) with a sigmoid Boltzmann relationship:
(V is the potential of half-maximal activation, s = RT/ zF, the slope factor with z the valence of the gating charge) and normalizing the measured tail current amplitudes to the maximal current amplitude Imax (Fig. 1F):
, 百拇医药
The instantaneous I–V curve, derived from the amplitude of tail currents during steps to different potentials following maximal activation by a depolarizing prepulse, is approximately linear, indicative of an Ohmic conductance of the open channel. Rate coefficients for opening and closing can readily be calculated from Po and ( = Po/; = (1 – Po)/) (Fig. 1G). Very similar voltage-dependent characteristics have been observed for at least TRPV1, TRPV3, TRPM3, TRPM4, TRPM5 and TRPM8.
, http://www.100md.com
Shift of the activation curve: a frequently used trick of TRP channels
There is convincing evidence that activation of TRP channels is often accompanied by a shift of their voltage-dependent gating, as illustrated with a few examples in Fig. 2. The first example concerns activation of TRP channels by changes in temperature. Various TRP channels (so-called ‘thermoTRPs’) are involved in temperature sensing, not only in sensory nerve fibres originating from the dorsal root or trigeminal ganglia but also in keratinocytes (Peier et al. 2002; Patapoutian et al. 2003; Moqrich et al. 2005). Four TRPV channels are activated upon heating, and are responsible for detection of temperatures that we experience as either warmth, heat or noxious heat (Patapoutian et al. 2003). Two other TRP channels, TRPM8 and TRPA1, are cold-activated, sensing cool and noxious cold, respectively. We have recently shown that the temperature sensitivity of TRPV1 and TRPM8 is modulated by the transmembrane voltage, and that changes in ambient temperature result in graded shifts of the voltage dependence of activation (Voets et al. 2004). Figure 2A shows that the activation curve of TRPV1 is shifted leftward by nearly 200 mV if the temperature is increased from 17 to 42°C (Voets et al. 2004). On the other hand, cooling from 37°C to 15°C induces a leftward shift of 170 mV for TRPM8 (Fig. 2B, Voets et al. 2004). The TRPV1-specific ligand capsaicin shifts the TRPV1 activation curve at constant temperature (the maximal capsaicin induced shift at 24°C is 200 mV with an EC50 of 30 nM, Fig. 2C). Similarly, the TRPM8 activator menthol induces a leftward shift of the activation curve (Fig. 2D, the maximal menthol induced shift at 34°C is 200 mV with an EC50 of 30 μM, Voets et al. 2004). Figure 2E illustrates how increasing the intracellular Ca2+ concentration shifts the activation curve of TRPM4 towards more negative potentials (Fig. 2E, Nilius et al. 2004). This activation is followed by a fast desensitization, which is accompanied by (or is due to) a dramatic rightward shift of the activation curve. Importantly, interventions that sensitize TRPM4 to Ca2+ shift the activation curve towards negative potentials, e.g. phosphorylation via protein kinase C activation or addition of calmodulin to excised patches (Nilius et al. 2005). Vice versa, desensitization of TRPM4 to Ca2+, which can be achieved by deletion of putative Ca2+/calmodulin binding sites in the channel C-terminus or by coexpression of the dominant negative calmodulin mutant CaM1,2,3,4 (with the four EF-hand Ca2+ bindings sites deleted), shifts channel activation towards positive potentials (Nilius et al. 2005). Interestingly, neutralization of positive charges in the fourth transmembrane spanning helix (S4), which has some homology with the voltage sensing S4 helix of Shaker, shifts the activation curve towards positive potentials (Fig. 2F, K. Talavera & B. Nilius, unpublished).
, 百拇医药
A, voltage dependence of the open probability of TRPV1 channels at 17°C () and 42°C (). The inset shows the respective current families for a voltage step programme (holding potential 0 mV, test steps from –120 to +160 mV, increment 40 mV), back step to +60 mV. Note the dramatic shift towards negative potentials from +198 to –33 mV. B, same protocol as in A, but for TRPM8 at 37°C () and 15°C (). Note the leftward shift of the activation curve. V changed from +203 to 53 mV. C, same voltage protocol as in A. TRPV1 is activated by 100 nM capsaicin () causing a leftward voltage shift of from +95 to –22 mV. Temperature was 24°C. D, 30 μM menthol () induced a leftward shift from 203 to 74 mV of the TRPM8 activation curve at 34°C (circles are controls, same protocol as in A). E, TRPM4 was activated in inside-out patches by different [Ca2+]i. Voltage protocol consists of steps from –100 to +160 mV (increment 40 mV) from a holding potential of 0 mV. A change from 10 μM [Ca2+]i () to 1000 μM [Ca2+]i () induced a leftward shift of the activation curve from 162 to 68 mV. F, TRPM4 activation in the presence of 300 μM [Ca2+]i. Neutralizing the positive charge K919 caused a shift of V from +60 (wild type TRPM4) to +192 mV (K919A mutant). Holding potential was 0 mV, steps from –25 to 250 (mV). (Calibration bars in the insets are identical for examples shown in each panel). (A–D are adapted from Voets et al. (2004) with permission from Nature, http://www.nature.com/; E is adapted from Nilius et al. (2004).)
, 百拇医药
A thermodynamic approach of the voltage-dependent shift of channel activation
We have recently applied a simple thermodynamic formalism to describe the shifts in voltage dependence due to changes in temperature (Voets et al. 2004). For the simple gating scheme (1), the open probability of the channel is given by:
where z is the gating charge, F is Faraday's constant, R is the gas constant, T is the absolute temperature and G is the free energy difference between open and closed states of the channel, i.e.
, 百拇医药
where H and S are the differences in enthalpy and entropy between open (o) and closed (c) states. The potential V for half maximal activation (see eqn (2)) is determined by the condition: G – zFV = 0. Hence:
The gating charge of TRP channels is much smaller than that of classical voltage-gated channels. This implies that any factor inducing changes in H or S will provoke larger larger shifts of the activation curve of TRP channels.
, 百拇医药
The rate coefficients and for the simple closed–open model can be calculated from Po and the time constants of current activation using the equations Po = /( + ) and = 1/( + ), and are related to the differences in free energy between an intermediate activated channel state and the closed or open state of the channel Gc and Go by the equations (derived from Eyring's rate theory):
where is the fraction of the electric field that the gating charge z has to cross in moving from the closed to the intermediate activated state and f is a frequency factor, representing the number of potential transitions per unit time. This prefactor, which is not known, is assumed to be independent of temperature at least in the investigated temperature range. The exponential terms give the probabilities that the thermal energy exceeds the energy required to induce the conformational changes of the channel. Gc and Go are given by:
, 百拇医药
It follows from these equations that the slopes of ln() and ln() as a function of 1/T (Arrhenius plots) are given by (– Hc + z FV)/R and (– Hc – z(1 – )FV)/R, respectively.
This formalism has been used to describe the temperature-induced voltage shifts of the activation of TRPV1 and TRPM8 (Fig. 2A and B). The Arrhenius plots for TRPM8 reveal a weak temperature dependence of the opening rates, but a steep temperature dependence of the closing rate (Fig. 3A). The plots at different voltages are almost parallel in the investigated temperature range, indicating that the contribution of the voltage terms to the slopes is small compared to that of the enthalpy terms Hc and Ho in the physiological voltage range. The linear trend of these plots further implies that Hc and Ho and therefore also H = Hc – Ho can be considered to be independent of temperature. The values for Ho and H are quite large, suggesting that the conformational changes during channel closure are accompanied by a large energetic increase. On the other hand, the reverse has been found for TRPV1 channels, i.e. temperature strongly affects the opening rate but has only minor effects on the closing rate. The trends of the corresponding Arrhenius plots are also linear, with slopes that are virtually independent of the applied voltage (Fig. 3B).
, 百拇医药
A, Arrhenius plots for the rate constants and of TRPM8 for the two state model at –80 (), +80 () and +160 mV (). The slope corresponds to the enthalpy. The enthalpy for channel opening is small, corresponding to a Q10 values of 1.2, but is large for channel closing (corresponding Q10 is 9.4). From the shifts in V, the following data were obtained: Ho = 13 kJ mol–1, Hc = 170 kJ mol–1, So = –166 J mol–1 K–1, Sc = 392 J mol–1 K–1. B, same analysis as in A but for TRPV1. Note that the enthalpy for channel opening is much higher (corresponding Q10 is 14.8) than for channel closing (Q10 is 1.35). Considering the shifts in V, the following data were obtained: Ho = 205 kJ mol–1, Hc = 21 kJ mol–1, So = 468 J mol–1 K–1, Sc = –130 J mol–1 K–1. C, shift of the potential for half-maximal activation of TRPM8 as a function of temperature. With V = (H – ST)/zF, the positive slope corresponds to the negative change in entropy. Menthol induced a parallel shift of this curve toward more negative potentials, likely indicating a change in enthalpy. (A and B are adapted from Voets et al. (2004) with permission from Nature, http://www.nature.com/.)
, 百拇医药
Figure 3C shows that the plot of V as a function of T is linear over a wide temperature range for TRPM8. Its slope, i.e. –S/zF according to eqn (5), and therefore also S apparently does not depend on temperature. The value of S is –558 J mol–1 K–1 for TRPM8 and 598 J mol–1 K–1 for TRPV1. Note that the changes in enthalpy and entropy for the warm-sensing TRPV1 channel have the same sign, implying that the difference in free energy G = H – TS between closed and open state, which determines open probability, can be small despite the large values of H and S. In other words, the large change in enthalpy can be compensated by a large change in entropy. A similar argument applies for TRPM8, for which both H and S are negative. This makes it possible that the relatively small energy provided by the membrane potential is able to significantly modulate the open probability in the physiological range of potentials. From the slope of the V versus T plot, we can calculate a shift –7 mV °C–1 (see also legend Fig. 3A) for TRPM8 and of 9 mV °C–1 for TRPV1. More general, the sign of S = So – Sc determines the direction in which a change in temperature shifts the voltage-dependent activation of the channel.
, 百拇医药
To resolve the question why cooling activates a channel like TRPM8 and warming activates a channel like TRPV1, we have calculated the change of Po with temperature T from eqn (3). It can be readily calculated that:
from which it is evident that the sign of Po/ T, which determines whether a channel is warm or cold activated, is the same as that of (H – zFV). The voltage term zFV amounts to about 10 kJ mol–1 at 100 mV, which is much smaller (in absolute value) than the value of H (in the order of 100–200 kJ mol–1 for TRPV1 and TRPM8). Since H is positive for TRPV1 and negative for TRPM8, these channels are warm- and cold-activated, respectively. An intriguing inference is that the term (H – zFV) changes sign at a potential V = H/zF, thereby switching from a warm-activated TRPV1 channel to a cold-activated channel at extreme positive potentials (V > 2.5 V), or from a cold-activated TRPM8 channel to a warm-activated channel for V < –2.1 V.
, 百拇医药
Voltage shift of channel activation by ‘gating’ modifiers
It has also been observed that some ligands cause prominent shifts of the voltage-dependent channel activation at constant temperature (see Fig. 2C and D). Figure 3E gives as an example of the voltage shift of TRPM8 by menthol at various temperatures. This shift of the V versus T curve is roughly parallel, indicating that, according to eqn (5), menthol mainly affects the enthalpy rather than the entropy. Figure 4 illustrates some examples of the shift of the voltage-dependent activation induced by agents that modify gating of TRPM4, the first TRP channel whose intrinsic voltage dependence was analysed in detail (Nilius et al. 2003). From comparison with the KscA channel, one might expect that the intrinsic voltage sensor is localized in the fourth membrane spanning -helix (S4) of TRPM4, TRPM5 and TRPM8.
, 百拇医药
A, a structural model of the C-terminus of TRPM4 from S1044 to D1214 (see text for more details), viewed using DS ViewerPro 5.0 software (Accelrys Inc., San Diego, CA, USA) (parental structure score 1.29, serine acetyltransferase-apoenzyme, chain A (1smmA) score). Indicated are the two putative Ca2+–calmodulin binding sites in blue (for details see Nilius et al. 2005). -Helices are shown in red, turns in grey. The putative binding site for decavanadate, R1136ARDKR1141) is shown in cyan (for details see Nilius et al. 2004). In the modelled coiled-coil region (1134–1154 and 1158–1185, centre of the figure) are two PKC phosphorylation sites indicated in yellow (Nilius et al. 2005). B, deletion of the calmodulin binding site induced a rightward shift of the activation curve of TRPM4. The inset shows current traces from voltage steps for controls () and the deletion mutant of the first C-terminal Ca2+–calmodulin binding site A1076–S1098 (). [Ca2+]i was 100 μM, voltage steps from –100 to +200 mV, 20 mV increment for the control and from –100 to +260 mV, increment +40 mV for the deletion mutant (inside out patches). Note the shift of V from +112 to +235 mV. C, effects of decavanadate (DV) on the activation curve of TRPM4. The inset shows current traces from inside patches with [Ca2+]i of 300 μM (controls, and after application of 10 μM DV, ). Activation curves show a shift of more than 200 mV toward negative potentials in the presence of DV. Note the very shallow voltage dependence, and the loss of activation and deactivation kinetics (Nilius et al. 2004). D, mutation of the C-terminal putative PKC phosphorylation sites S1145 and S1152 to aspartates induces a leftward shift of the TRPM4 activation curves. The inset shows current traces from inside out patches with wild-type TRPM4 () and with the mutation S1145D (). Shift is from V of +124 to +12 mV (100 μM [Ca2+]i, steps from –75 to +200, +150 mV, respectively, holding potential 0 mV). This shift is in agreement with a sensitizing effect of PKC phosphorylation of TRPM4 (Nilius et al. 2005). (C, modified from Nilius et al. (2004).).
, http://www.100md.com
We have shown that modifications to the C-terminus of TRPM4 can induce alterations of the voltage dependence of this channel (Nilius et al. 2005). We obtained a putative 3D structure of the C-terminus of TRPM4 using the automated comparative protein modelling with prediction of structure of the ROBETTA server (‘ab initio’ and comparative modelling). We used as template the serine acetyltransferase-apoenzyme, chain A, a protein with a coiled coil domain as described for the C-terminus of TRPM4 (Perraud et al. 2003). In addition, we used previously obtained information of the C-terminal localization of Ca2+–calmodulin binding sites in the positions A1076–S1098 and R1095–E1130. Figure 4A shows the results of this structural approach. Indicated are the two putative Ca2+–calmodulin binding sites as helical structures, two coiled coil regions (1134–1154 and 1158–1185), which contain the two functionally important PKC phosphorylation sites (Nilius et al. 2005) and the putative decavanadate (DV) binding site R1136ARDKR1141 (Nilius et al. 2004). We propose that these indicated regions provide sensitive structural elements that may modify the voltage sensing of TRPM4.
, http://www.100md.com
Mutation or deletion of putative Ca2+–calmodulin binding sites in the C-terminus induce a large shift of the voltage-dependent activation towards positive potentials (Fig. 4B, Nilius et al. 2005). Vice versa, application of calmodulin to the inner side of cell free patches diminished channel desensitization and induces a shift towards negative potentials (B. Nilius, unpublished and Nilius et al. 2005). Figure 4C illustrates that DV induces a dramatic shift in the voltage dependence and decrease the voltage sensitivity of TRPM4 when applied to the inner side of excised patches (see Fig. 1C–H). Tail currents deactivate rapidly and almost completely in the absence of DV, but more slowly and incomplete in the presence of DV. DV, a compound with six negative charges, is widely used as a tool to interact with ATP binding in transport ATPases (Toyoshima et al. 2000; Pezza et al. 2002; Clausen et al. 2003; Tiago et al. 2004). We have therefore searched for a motif in TRPM4 similar to the FSRDRK motif in SERCA (Toyoshima et al. 2000). Such a motif was identified in the C-terminal coiled-coil region as a cluster of positive charges, R1136ARDKR1141 (R/K motif, see also Fig. 4A). Deletion of this cluster or replacement of the C-terminus with that of TRPM5, which does not contain this motif, eliminate the effect of DV on the voltage-dependent shift of TRPM4 activation (Nilius et al. 2004), strongly indicating that the site of DV action resides in the C-terminus of TRPM4. Interestingly, this site shows some similarities with the pleckstrin domain in phospholipases C, which mediates interaction with second messenger lipids such as PIP2 (Harlan et al. 1994, 1995). Interestingly, phosphorylation of the two putative PKC sites (S1145 and S1152) increased the apparent Ca2+ sensitivity of TRPM4 (Nilius et al. 2005). Mutants that mimic the phosphorylated state (S1145D and S1152D) displayed a delayed desensitization and a remarkable shift of the voltage dependence of activation towards negative potentials. (V = 56 ± 11 mV for the S1145D mutant, V = 65 ± 19 mV for the S1152D mutant, B Nilius & J. Prenen, unpublished).
, http://www.100md.com
All these examples clearly show that TRP channels perform a variety of important physiological functions such as temperature sensing, Ca2+ sensing, and detecting natural or endogenous compounds by shifting their activation curve towards less positive potentials. This shift is supported by the relatively small gating charge of TRP channels (see eqn (5)). Thus, the here-described unique voltage dependence of TRP channels induces equally unique gating properties, which are of physiological importance rather than being an epiphenomenon based on structural similarities with voltage-gated channels.
, 百拇医药
Conclusions
Without doubt, temperature sensing of several thermoTRPs is connected to a shift of their voltage dependence. This has been shown in detail for TRPV1 and TRPM8 (Brauchi et al. 2004; Voets et al. 2004), and more recently for TRPV3 (Chung et al. 2005). The small gating charge of TRP channels compared to that of classical voltage-gated channels could lie at the basis of the large shifts of their voltage-dependent activation curves, and may be essential for their gating versatility. We postulate that many physiological stimuli may exploit this feature and activate TRPs by shifting their voltage dependence from a physiologically uninteresting voltage range into a functionally relevant voltage window by inducing small changes in the Gibbs free energy of the channel.
, http://www.100md.com
Footnotes
This report was presented at The Journal of Physiology Symposium on TRP channels: physiological genomics and proteomics, San Diego, CA, USA, 5 April 2005. It was commissioned by the Editorial Board and reflects the views of the authors.
References
Aggarwal SK & MacKinnon R (1996). Contribution of the S4 segment to gating charge in the Shaker K+ channel. Neuron 16, 1169–1177.
, 百拇医药
Ahern GP & Premkumar LS (2002). Voltage-dependent priming of rat vanilloid receptor: effects of agonist and protein kinase C activation. J Physiol 545, 441–451.
Brauchi S, Orio P & 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.
Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD & Julius D (1997). The capsaicin receptor: a heat-activated ion channel in the pain pathway. Nature 389, 816–824.
, 百拇医药
Chuang HH, Prescott ED, Kong H, Shields S, Jordt SE, Basbaum AI, Chao MV & Julius D (2001). Bradykinin and nerve growth factor release the capsaicin receptor from PtdIns (4,5),P2-mediated inhibition. Nature 411, 957–962.
Chung MK, Guler AD & Caterina MJ (2005). Biphasic currents evoked by chemical or thermal activation of the heat-gated ion channel, TRPV3. J Biol Chem 280, 15928–15941.
Chung MK, Lee H, Mizuno A, Suzuki M & Caterina MJ (2004). 2-Aminoethoxydiphenyl borate activates and sensitizes the heat-gated ion channel TRPV3. J Neurosci 24, 5177–5182.
, http://www.100md.com
Clapham DE (2003). TRP channels as cellular sensors. Nature 426, 517–524.
Clausen JD, McIntosh DB, Vilsen B, Woolley DG & Andersen JP (2003). Importance of conserved N-domain residues Thr441, Glu442, Lys515, Arg560, and Leu562 of sarcoplasmic reticulum Ca2+-ATPase for MgATP binding and subsequent catalytic steps. Plasticity of the nucleotide-binding site. J Biol Chem 278, 20245–20258.
Corey DP (2003). New TRP channels in hearing and mechanosensation. Neuron 39, 585–588.
, http://www.100md.com
Delmas P (2004). Polycystins: from mechanosensation to gene regulation. Cell 118, 145–148.
Grimm C, Kraft R, Schultz G & Harteneck C (2005). Activation of the melastatin-related cation channel TRMP3 by D-erythro-sphingosine. Mol Pharm 67, 798–805.
Harlan JE, Hajduk PJ, Yoon HS & Fesik SW (1994). Pleckstrin homology domains bind to phosphatidylinositol-4,5-bisphosphate. Nature 371, 168–170.
, 百拇医药
Harlan JE, Yoon HS, Hajduk PJ & Fesik SW (1995). Structural characterization of the interaction between a pleckstrin homology domain and phosphatidylinositol 4,5-bisphosphate. Biochemistry 34, 9859–9864.
Hofmann T, Chubanov V, Gudermann T & Montell C (2003). TRPM5 is a voltage-modulated and Ca2+-activated monovalent selective cation channel. Current Biol 13, 1153–1158.
Montell C, Birnbaumer L & Flockerzi V (2002a). The TRP channels, a remarkable functional family. Cell 108, 595–598.
, http://www.100md.com
Montell C, Birnbaumer L, Flockerzi V, Bindels RJ, Bruford EA, Caterina MJ, Clapham D, Harteneck C, Heller S, Julius D, Kojima I, Mori Y, Penner R, Prawitt D, Scharenberg AM, Schultz G, Shimizu S & Zhu MX (2002b). A unified nomenclature for the superfamily of TRP cation channels. Mol Cell 9, 229–231.
Moqrich A, Hwang SW, Earley TJ, Petrus MJ, Murray AN, Spencer KS, Andahazy M, Story GM & Patapoutian A (2005). Impaired thermosensation in mice lacking TRPV3, a heat and camphor sensor in the skin. Science 307, 1468–1472.
, 百拇医药
Moran MM, Xu H & Clapham DE (2004). TRP ion channels in the nervous system. Current Opin Neurobiol 14, 362–369.
Nilius B, Prenen J, Droogmans G, Voets T, Vennekens R, Freichel M, Wissenbach U & Flockerzi V (2003). Voltage dependence of the Ca2+-activated cation channel TRPM4. J Biol Chem 278, 30813–30820.
Nilius B, Prenen J, Janssens A, Voets T & Droogmans G (2004). Decavanadate modulates gating of TRPM4 cation channels. J Physiol 560, 753–765.
, 百拇医药
Nilius B, Prenen J, Tang J, Wang C, Owsianik G, Janssens A, Voets T & Zhu MX (2005). Regulation of the Ca2+ sensitivity of the nonselective cation channel TRPM4. J Biol Chem 280, 6423–6433.
Nilius B, Vennekens R, Prenen J, Hoenderop JG, Bindels RJ & Droogmans G (2000). Whole-cell and single channel monovalent cation currents through the novel rabbit epithelial Ca2+ channel ECaC. J Physiol 527, 239–248.
Patapoutian A, Peier AM, Story GM & Viswanath V (2003). ThermoTRP channels and beyond: mechanisms of temperature sensation. Nature Rev Neurosci 4, 529–539.
, http://www.100md.com
Peier AM, Reeve AJ, Andersson DA, Moqrich A, Earley TJ, Hergarden AC, Story GM, Colley S, Hogenesch JB, McIntyre P, Bevan S & Patapoutian A (2002). A heat-sensitive TRP channel expressed in keratinocytes. Science 296, 2046–2049.
Perraud AL, Schmitz C & Scharenberg AM (2003). TRPM2 Ca2+ permeable cation channels: from gene to biological function. Cell Calcium 33, 519–531.
Pezza RJ, Villarreal MA, Montich GG & Argarana CE (2002). Vanadate inhibits the ATPase activity and DNA binding capability of bacterial MutS. A structural model for the vanadate–MutS interaction at the Walker A motif. Nucl Acids Res 30, 4700–4708.
, http://www.100md.com
Piper AS, Yeats JC, Bevan S & Docherty RJ (1999). A study of the voltage dependence of capsaicin-activated membrane currents in rat sensory neurones before and after acute desensitization. J Physiol 518, 721–733.
Premkumar LS & Ahern GP (2000). Induction of vanilloid receptor channel activity by protein kinase C. Nature 408, 985–990.
Prescott ED & Julius D (2003). A modular PIP2 binding site as a determinant of capsaicin receptor sensitivity. Science 300, 1284–1288.
, http://www.100md.com
Rosenbaum T, Gordon-Shaag A, Munari M & Gordon SE (2004). Ca2+/calmodulin modulates TRPV1 activation by capsaicin. J General Physiol 123, 53–62.
Sidi S, Friedrich RW & Nicolson T (2003). NompC TRP channel required for vertebrate sensory hair cell mechanotransduction. Science 301, 96–99.
Tiago T, Aureliano M & Gutierrez-Merino C (2004). Decavanadate binding to a high affinity site near the myosin catalytic centre inhibits F-actin-stimulated myosin ATPase activity. Biochemistry 43, 5551–5561.
, 百拇医药
Toyoshima C, Nakasako M, Nomura H & Ogawa H (2000). Crystal structure of the calcium pump of sarcoplasmic reticulum at 2.6 resolution. Nature 405, 647–655.
Trevisani M, Smart D, Gunthorpe MJ, Tognetto M, Barbieri M, Campi B, Amadesi S, Gray J, Jerman JC, Brough SJ, Owen D, Smith GD, Randall AD, Harrison S, Bianchi A, Davis JB & Geppetti P (2002). Ethanol elicits and potentiates nociceptor responses via the vanilloid receptor-1. Nature Neursci 5, 546–551.
, http://www.100md.com
Ullrich ND, Voets T, Prenen J, Vennekens R, Talavera K, Droogmans G & Nilius B (2005). Comparison of functional properties of the Ca2+-activated cation channels TRPM4 and TRPM5 from mice. Cell Calcium 37, 267–278.
Vlachova V, Teisinger J, Suankova K, Lyfenko A, Ettrich R & Vyklicky L (2003). Functional role of C-terminal cytoplasmic tail of rat vanilloid receptor 1. J Neurosci 23, 1340–1350.
Voets T, Droogmans G, Wissenbach U, Janssens A, Flockerzi V & Nilius B (2004). The principle of temperature-dependent gating in cold- and heat-sensitive TRP channels. Nature 430, 748–754.
, 百拇医药
Voets T, Janssens A, Prenen J, Droogmans D & Nilius G (2003). Mg2+-dependent gating and strong inward rectification of the cation channel TRPV6. J General Physiol 121, 245–260.
Walker RG, Willingham AT & Zuker CS (2000). A Drosophila mechanosensory transduction channel. Science 287, 2229–2234.
Xu HX, Ramsey IS, Kotecha SA, Moran MM, Chong JHA, Lawson D, Ge P, Lilly J, Silos Santiago I, Xie Y, DiStefano PS, Curtis R & Clapham DE (2002). TRPV3 is a calcium-permeable temperature-sensitive cation channel. Nature 418, 181–186., http://www.100md.com(Bernd Nilius, Karel Talav)