Spontaneous single-channel activity of neuronal TRP5 channel recombinantly expressed in HEK293 cells

Neuroscience Letters 285 (2000) 111±114

www.elsevier.com/locate/neulet

Spontaneous single-channel activity of neuronal TRP5 channel recombinantly expressed in HEK293 cells Hisanobu Yamada a, b, Minoru Wakamori a,*, Yuji Hara a, Yasuo Takahashi b, Kazuhiko Konishi b, Keiji Imoto a, Yasuo Mori a a

Department of Information Physiology, National Institute for Physiological Sciences, Okazaki, Aichi 444-8585, Japan Laboratory of Neuroinformation Science, Faculty of Liberal Arts, Tohoku Gakuin University, Sendai 981-3193, Japan

b

Received 22 February 2000; received in revised form 21 March 2000; accepted 23 March 2000

Abstract Mammalian homologues of the Drosophila transient receptor potential (trp) protein (TRP) form Ca 21 permeable cation channels activated in response to stimulation of G-protein-coupled receptors. Establishing biophysical characteristics of basal TRP activity is of great importance in understanding modulatory processes, which underlie enhancement of TRP activity via receptor stimulation. We have examined spontaneous activity of the TRP5 channel recombinantly expressed in human embryonic kidney cells, using the conventional whole-cell mode of the patch-clamp technique in a low-Ca 21 external solution. The unitary Na 1 conductance of the TRP5 channel was linear, being 47.6 pS. By contrast, the open probability of the TRP5 channel showed a voltage-dependent decrease below 250 mV. These biophysical properties are important hallmarks in distinguishing the TRP5 channel in native neuronal preparations, whose spontaneous activity may contribute to control of resting membrane potentials and generation of action potentials. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cation channel; Receptor-activated channel; Transient receptor potential protein; Patch clamp; Single-channel recording; HEK293 cell

Receptor-activated cation channels (RACCs) are the group of monovalent and divalent cation-permeable channels activated in response to stimulation of G-proteincoupled or tyrosine kinase-coupled cell membrane receptors [5]. Importance of RACCs has been recognized in a variety of cellular responses, including lymphocyte activation, contraction of smooth muscle, and excitability of neurons [5,16]. RACCs are highly diverse in functional properties. In fact, RACCs open in response to various activation triggers including depletion of the stores via Ca 21 release through inositol 1,4,5-trisphosphate (IP3) receptors, direct action of G-protein, diacylgrycerol, arachidonic acid and second messengers such as Ca 21, IP3, and inositol 1,3,4,5-tetraphosphate (IP4) [3,5,6,16,21]. A cation selectivity also varies among RACC types: many of the RACCs are non-selective or slightly selective to divalent cations over monovalents [5], whereas store-operated, Ca 21 release-activated Ca 21 (CRAC) channels show an extremely high Ca 21 selectivity * Corresponding author. Tel.: 181-564-55-7851; fax: 181-56455-7853. E-mail address: [email protected] (M. Wakamori).

in the presence of extracellular Ca 21 [16] but are highly permeable to monovalent cations in the absence of Ca 21 and Mg 21 [10]. It is not readily feasible to distinguish RACC types and to elucidate physiological roles of each RACC type, since selective blockers for respective RACCs are unavailable. However, recent molecular studies of mammalian homologues of transient receptor potential (trp) and trp-like (trpl) gene, that have been implicated in generating light-induced depolarization in Drosophila phototransduction [11], provided us with tools to solve this problem. Seven mammalian cDNAs encoding trp proteins, TRP1±TRP7, have been so far isolated [14,15,17±19,23±26]. TRP homologues show Ca 21-permeable cation channel activity activated by stimulation of G-protein-coupled receptors, and show spontaneous activity, as well [7,15,22]. TRP5, that distributes predominantly in the brain, forms a cation channel with relatively high permeability to Ca 21 when recombinantly expressed in human embryonic kidney 293 (HEK293) cells [14,19]. The activation mechanism of the TRP5 channel still remains elusive [14,19], and biophysical properties of the TRP5 channel are yet to be elucidated. To address this issue, we have here focused on characterizing the

0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 0) 01 03 3- 8

112

H. Yamada et al. / Neuroscience Letters 285 (2000) 111±114

basal activity of recombinant TRP5 channels at the microscopic level in HEK293 cells with whole-cell patch-clamp recording. The procedure for expression of the TRP5 in HEK293 cells is similar to that described elsewhere [14]. Currents were recorded using the patch-clamp technique of the whole-cell mode with an Axopatch 200B ampli®er (Axon Instruments, Foster City, CA). Pipette resistance ranged from 2 to 4 MV when ®lled with the pipette solution described below. Currents were sampled at 10 kHz after low-pass ®ltered at 1 kHz, and analyzed with pCLAMP 6.02 software (Axon Instruments). NPo was calculated as sum of total open time at individual open-channel current levels divided by the total length of the record. To construct amplitude histograms and current±voltage relationship (Fig. 1), only openings lasting longer than 1 ms were analyzed. The pipette solution contained (in mM): CsOH 105, aspartic acid 105, CsCl 40, MgCl2 2, CaCl2 1.3, EGTA 5, Na2ATP 2, HEPES 5, adjusted to pH 7.2 with CsOH. Calculated free Ca 21 concentration was 50 nM. The low-Ca 21 external solution contained (in mM): NaCl 121.7, MgCl2 1.2, CaCl2 1.2, EGTA 2, glucose 10, HEPES 11.5, mannitol 51, adjusted to pH 7.4 with NaOH (90 nM calculated free Ca 21 ([Ca 21]o)). The osmolarity of the solutions were adjusted to about 320 mOsm. In the previous experiments [14], we found that addition of 10 mM Ca 21 to the external solution induced a shift of reversal potential of TRP5 currents by only 9 mV to the depolarizing direction. This indicates that the TRP5 channel is capable of permeating substantial amount of monovalent cations. To characterize monovalent cation currents carried by the TRP5 channel, whole-cell mode of patch-clamp technique was employed and single-channel activities were studied in the low-Ca 21 external solution. HEK293 cells of relatively small sizes were selected for the measurement. Combination of small cell capacitance (7.1 ^ 0.3 pF, n ˆ 88) and high input resistance (7.8 ^ 1.6 GV, n ˆ 7) allowed us to record single-channel events with high resolution using whole-cell mode. This recording strategy has an advantage that unitary currents are measured with controlled intracellular ionic composition without changing the wholecell structure. Eighty-eight out of 116 TRP5-transfected HEK293 cells showed single-channel activities at a holding potential (Vh) of 250 mV (Fig. 1Ab), whereas none of 40 control HEK293 cells had measurable single-channel activities (Fig. 1Aa) indicating that spontaneous single-channel activities in the transfected HEK293 cells are currents through the TRP5 channel. Since the unitary conductance is an important biophysical criterion in characterizing ion permeation properties of channels, we ®rst determined the single-channel conductance. Elementary currents recorded at Vhs of 250, 270 and 290 mV are shown in Fig. 1B, and an example of amplitude histogram constructed from the currents at a Vh of 250 mV is shown in Fig. 1C. The histogram was well ®tted by the sum of ®ve equidistant Gaussians with a peak distance of 3.2 pA. Plots of unitary current±voltage relationship yield a single-

channel conductance of 47.6 ^ 1.7 pS (n ˆ 7) and the apparent reversal potential of 17.2 ^ 1.9 mV (n ˆ 7). This value excludes the possibility that our recording of the single-channel currents results from the contamination of endogenous Cl 2 channels, since the calculated reversal potential of Cl 2 current is 225.6 mV. Among seven TRP homologues, the single-channel currents of TRP3 [7,9,25] and TRP6 [6] have been examined in the presence of Mg 21. Although we cannot directly compare the single-channel conductance with those reported, because of differences in recording conditions such as the ionic composition of the solutions, TRP5 appears to have a characteristic single-channel conductance. The recombinantly expressed TRP3 has single-channel conductances of 60 pS in Chinese hamster ovary (CHO) cells [25] and in HEK293 cells [7] and 23 pS in bovine pulmonary artery endothelial cells [9], whereas TRP6 has a slope conductance of 37.5 pS in CHO cells [6]. Besides the recombinantly expressed TRP channels, the native CRAC channels activated by store-depletion in T lymphocytes, have the similar range of conductance (36±40 pS) when the concentration of external divalent cations is lowered to micromolar range [10], whereas the unitary cord conductance estimated by ¯uctuation analysis is 24 fS in the solutions containing 100

Fig. 1. Single-channel current±voltage relationship. (A) Wholecell recording from CD8 positive control HEK293 cells without transfection of TRP5 (a) and from CD8 positive HEK293 cells transfected with TRP5 (b) at a Vh of 250 mV. The lines and the dotted lines indicate the closed and the open current levels. (B) Examples of single-channel currents recorded at 250 (a), 270 (b) and 290 mV (c). (C) Amplitude histogram for the whole-cell recording illustrated in (B) at a Vh of 250 mV. Bin width was 0.1 pA. Continuous curve is ®tted with Gaussian distributions. (D) Current±voltage relationship of the unitary TRP5 channel. Data points represent mean ^ SEM of seven cells. Liner regression yields a single-channel conductance of 47.6 pS. The currents in Ab and B were recorded from different HEK cells.

H. Yamada et al. / Neuroscience Letters 285 (2000) 111±114

mM Ca 21 [26]. External Mg 21 at a concentration of 3 mM blocks monovalent CRAC currents, converting long-duration opening into ¯icker opening [10]. The TRP5 channel expressed in HEK293 cells, however, showed long-duration opening even in the presence of 1.1 mM Mg 21 (Fig. 2Da), and this opening pattern persisted after the removal of extracellular Mg 21 (Fig. 2Db). These results suggest that the TRP5 is unique in pore properties which differ from those of the other TRP cation channels and the native CRAC channel. Macroscopic current±voltage (I±V) relationships of TRP channels have demonstrated heterogeneity in voltagedependence. TRPC1A expressed in CHO cells [24], TRP6 [2] and TRP7 [15] in HEK293 cells showed linear I±V relationship, while TRPC3 in CHO cells [25] and TRP4 in HEK293 [18] showed signi®cant inward recti®cation. In Fig. 1B, whereas simultaneous open of four channels was observed at a Vh of 250 mV, only two channels open simultaneously at a Vh of 290 mV from the same patch, suggesting a voltage-dependence of TRP5 channel gating. Since we could not obtain the number of channels (N) in the patch, we estimated the voltage-dependency of the channel activity (NPo; Po is the probability of the channel being open). In Fig. 2B we calculated NPo every 200 ms at Vhs of 250 mV

Fig. 2. Voltage-dependency of single-channel opening. (A) Two consecutive traces of single-channel currents recorded from a cell at 250 (a) and 290 mV (b) between 6 and 7 s in (B). The lines and the dotted lines indicate the closed and the open current levels. (B) Plot of NPo against time. Values of NPo are calculated every 200 ms. The thick and the thin lines indicate the average of the NPo at 250 (O) and 290 mV (W), respectively. (C) Voltage-dependency of NPo. The NPo was normalized to that at a Vh of 250 mV as shown by the asterisk. Data points represent mean ^ SEM of six cells and are ®tted to the Boltzmann's equation with a V0.5 of 262.4 mV and a k of 7.2 mV. (D) Two consecutive traces of single-channel currents recorded from a cell at 260 mV in the external solutions containing 1.1 mM (a) and 10 mM (b) Mg 21. The currents in (A) and (D) were recorded from different HEK cells.

113

and 290 mV and plotted NPo values against time, indicating the tendency that the NPo values at a Vh of 290 mV showed smaller values than those at a Vh of 250 mV. To quantitatively demonstrate the striking kinetic differences, NPo values at Vhs between 2110 mV and 240 mV were normalized to that at a Vh of 250 mV. Normalized channel activity was ®tted with the Boltzmann's equation to yield the midpoint potential (V0.5) of 262.4 mV and the slope factor (k) of 7.2 mV (Fig. 2C). It is interesting to note that this voltage-dependent change of channel opening occurs at around resting membrane potentials in neuronal cells. Extracellular Mg 21 is involved in the negative slope conductance of the NMDA receptor channel at negative membrane potentials [13]. We examined the effects of extracellular Mg 21 on the unitary currents of the TRP5 at a Vh of 260 mV, where the I±V relationship shows the negative slope and the NPo is smaller than that at a Vh of 250 mV. Neither opening behavior (Fig. 2D) nor NPo (1.7 ^ 2.3% increase, n ˆ 3) was affected signi®cantly by the reduction of external free Mg 21 concentration from 1.1 mM (Fig. 2Da) to 10 mM (Fig. 2Db). This strongly suggests that the `J-shaped' I±V relationship of TRP5 channel at whole-cell level (see Fig. 7 of Ref. [14]) is due to voltagedependence of single-channel gating, since the unitary current±voltage relationship is linear between 290 and 240 mV. However, it cannot be excluded that local Ca 21 voltage-dependently inactivated the CRAC channel-like activity generated by the TRP5 [16]. To obtain a clue to understand intracellular mechanisms that regulate spontaneous activity of TRP5 channels, contribution of intracellular ATP and Mg 21 to the spontaneous openings was examined. The average single-channel activity (NPo) was 0.67 ^ 0.34 (n ˆ 5) in the control condition, and the average NPo was 0.77 ^ 0.11 (n ˆ 5, P . 0:05) in the absence of ATP, and was 0.89 ^ 0.33 (n ˆ 6, P . 0:05) in the absence of Mg 21 at 5±6 min after the beginning of the internal perfusion with the pipette solutions (data not shown). These results may suggest that ATP is not essential for the basal activity of the TRP5 channel. It is interesting to test whether modulatory factors such as intracellular Ca 21, calmodulin, GTP and other second messengers are involved in the basal TRP5 channel activity or in the activity obtained after receptor stimulation. The spontaneous cation channel activity has been also reported for Drosophila Trpl expressed in Sf9 cells [22], and TRP3 [7] and TRP7 [15] expressed in HEK293 cells. Physiological roles of spontaneously active or background non-selective cation currents mediated by TRP homologues may include the pace-making activity in the sino-atrial node cells [4], the spontaneous secretion in pace-making endocrine cells [20], regulation of resting membrane potential [1], and the frequency and pattern modulation of action potentials [8] in smooth muscle cells. In this context, constitutive cation channel activity of TRP5 may depolarize resting membrane potential, and consequently may modulate frequency and pattern of action potentials in neurons. In fact, background non-selective

114

H. Yamada et al. / Neuroscience Letters 285 (2000) 111±114

cation current was recorded in rat hippocampal neurons [12], where the TRP5 is abundantly expressed [14,19]. Under the circumstances that selective blockers for respective TRP channels are not available, the biophysical properties, including `J-shaped' I±V relationship, singlechannel conductance of 47.6 pS, and insensitivity to Mg 21 are useful hallmarks in distinguishing native TRP activity. However, we may have to consider a possibility that electrophysiological properties of recombinantly expressed TRP5 channels are not completely identical with those in the native preparations, because the host cells may lack native modulators. This study was supported by research grants from the Ministry of Education, Science, Sports, and Culture of Japan, and by `the Research for the Future Program' of the Japan Society for the Promotion of Science, and by the research grant for cardiovascular diseases from the Ministry of Health and Welfare of Japan. We thank Drs Brian Seed and Gary Yellen for the CD8 expression plasmid. [1] Bae, Y.M., Park, M.K., Lee, S.H., Ho, W.K. and Earm, Y.E., Contribution of Ca 21-activated K 1 channels and non-selective cation channels to membrane potential of pulmonary arterial smooth muscle cells of the rabbit, J. Physiol., 514 (1999) 747±758. [2] Boulay, G., Zhu, X., Peyton, M., Jiang, M., Hurst, R., Stefani, E. and Birnbaumer, L., Cloning and expression of a novel mammalian homolog of Drosophila transient receptor potential (Trp) involved in calcium entry secondary to activation of receptors coupled by the Gq class of G protein, J. Biol. Chem., 272 (1997) 29672±29680. [3] Chyb, S., Raghu, P. and Hardie, R.C., Polyunsaturated fatty acids activate the Drosophila light-sensitive channels TRP and TRPL, Nature, 397 (1999) 255±259. [4] Denyer, J.C. and Brown, H.F., Pacemaking in rabbit isolated sino-atrial node cells during Cs 1 block of the hyperpolarization-activated current if, J. Physiol., 429 (1990) 401±409. [5] Fasolato, C., Innocenti, B. and Pozzan, T., Receptor-activated Ca 21 in¯ux: how many mechanisms for how many channels? Trends Pharmacol. Sci., 15 (1994) 77±83. [6] Hofmann, T., Obukhov, A.G., Schaefer, M., Harteneck, C., Gudermann, T. and Schultz, G., Direct activation of human TRPC6 and TRPC3 channels by diacylglycerol, Nature, 397 (1999) 259±263. [7] Hurst, R.S., Zhu, X., Boulay, G., Birnbaumer, L. and Stefani, E., Ionic currents underlying HTRP3 mediated agonistdependent Ca 21 in¯ux in stably transfected HEK293 cells, FEBS Lett., 422 (1998) 333±338. [8] Inoue, R. and Kuriyama, H., Dual regulation of cation-selective channels by muscarinic and a1-adrenergic receptors in the rabbit portal vein, J. Physiol., 465 (1993) 427±448. [9] Kamouchi, M., Philipp, S., Flockerzi, V., Wissenbach, U., Mamin, A., Raeymaekers, L., Eggermont, J., Droogmans, G. and Nilius, B., Properties of heterologously expressed hTRP3 channels in bovine pulmonary artery endothelial cells, J. Physiol., 518 (1999) 345±358. [10] Kerschbaum, H.H. and Cahalan, M.D., Single-channel recording of a store-operated Ca 21 channel in Jurkat T lymphocytes, Science, 283 (1999) 836±839. [11] Montell, C. and Rubin, G.M., Molecular characterization of the Drosophila trp locus: a putative integral membrane

[12]

[13]

[14]

[15]

[16] [17]

[18]

[19]

[20]

[21]

[22] [23]

[24]

[25]

[26]

protein required for phototransduction, Neuron, 2 (1989) 1313±1323. Nakazawa, K., Inoue, K., Watano, T., Koizumi, S. and Inoue, K., Zinc potentiation of neurotransmission and inhibition of background cationic conductance in rat cultured hippocampal neurones, J. Physiol., 484 (1995) 447±462. Nowak, L., Bregestovski, P., Ascher, P., Herbet, A. and Prochiantz, A., Magnesium gates glutamate-activated channels in mouse central neurones, Nature, 307 (1984) 462± 465. Okada, T., Shimizu, S., Wakamori, M., Maeda, A., Kurosaki, T., Takada, N., Imoto, K. and Mori, Y., Molecular cloning and functional characterization of a novel receptor-activated TRP Ca 21 channel from mouse brain, J. Biol. Chem., 273 (1998) 10279±10287. Okada, T., Inoue, R., Yamazaki, K., Maeda, A., Kurosaki, T., Yamakuni, T., Tanaka, I., Shimizu, S., Ikenaka, K., Imoto, K. and Mori, Y., Molecular and functional characterization of a novel mouse transient receptor potential protein homologue TRP7. Ca 21-permeable cation channel that is constitutively activated and enhanced by stimulation of G proteincoupled receptor, J. Biol. Chem., 274 (1999) 27359±27370. Parekh, A.B. and Penner, R., Store depletion and calcium in¯ux, Physiol. Rev., 77 (1997) 901±930. Petersen, C.C., Berridge, M.J., Borgese, M.F. and Bennett, D.L., Putative capacitative calcium entry channels: expression of Drosophila trp and evidence for the existence of vertebrate homologues, Biochem. J., 311 (1995) 41±44. Philipp, S., CavalieÂ, A., Freichel, M., Wissenbach, U., Zimmer, S., Trost, C., Marquart, A., Murakami, M. and Flockerzi, V., A mammalian capacitative calcium entry channel homologous to Drosophila TRP and TRPL, EMBO J., 15 (1996) 6166±6171. Philipp, S., Hambrecht, J., Braslavski, L., Schroth, G., Freichel, M., Murakami, M., CavalieÂ, A. and Flockerzi, V., A novel capacitative calcium entry channel expressed in excitable cells, EMBO J., 17 (1998) 4274±4282. Sankaranarayanan, S. and Simasko, S.M., A role for a background sodium current in spontaneous action potentials and secretion from rat lactotrophs, Am. J. Physiol., 271 (1996) C1927±C1934. Tsubokawa, H., Oguro, K., Robinson, H.P., Masuzawa, T. and Kawai, N., Intracellular inositol 1,3,4,5-tetrakisphosphate enhances the calcium current in hippocampal CA1 neurones of the gerbil after ischaemia, J. Physiol., 497 (1996) 67±78. Vaca, L., Sinkins, W.G., Hu, Y., Kunze, D.L. and Schilling, W.P., Activation of recombinant trp by thapsigargin in Sf9 insect cells, Am. J. Physiol., 267 (1994) C1501±C1505. Zhu, X., Jiang, M., Peyton, M., Boulay, G., Hurst, R., Stefani, E. and Birnbaumer, L., trp, a novel mammalian gene family essential for agonist-activated capacitative Ca 21 entry, Cell, 85 (1996) 661±671. Zitt, C., Zobel, A., Obukhov, A.G., Harteneck, C., Kalkbrenner, F., LuÈckhoff, A. and Schultz, G., Cloning and functional expression of a human Ca 21-permeable cation channel activated by calcium store depletion, Neuron, 16 (1996) 1189± 1196. Zitt, C., Obukhov, A.G., StruÈbing, C., Zobel, A., Kalkbrenner, F., LuÈckhoff, A. and Schultz, G., Expression of TRPC3 in Chinese hamster ovary cells results in calcium-activated cation currents not related to store depletion, J. Cell Biol., 138 (1997) 1333±1341. Zweifach, A. and Lewis, R.S., Mitogen-regulated Ca 21 current of T lymphocytes is activated by depletion of intracellular Ca 21 stores, Proc. Natl. Acad. Sci. USA, 90 (1993) 6295±6299.