TRPC4 can be activated by G-protein-coupled receptors and provides sufficient Ca(2+) to trigger exocytosis in neuroendocrine cells.

Transient receptor potential (TRP) channels form a large family of plasma membrane cation channels. Mammalian members of the "short" TRP family (TRP channel (TRPC) 1-7 are Ca(2+)-permeant, non-selective cation channels that are widely expressed in various cell types, including neurons. TRPC activity is linked through unknown mechanisms to G-protein-coupled receptors or receptor tyrosine kinases that activate phospholipase C. To investigate the properties and function of TRPC4 in neuronally derived cells, we transiently expressed mouse TRPC4 and histamine H(1) receptor in mouse adrenal chromaffin cells and PC12 cells. Histamine, but not thapsigargin, stimulated Mn(2+) influx in transfected cells. In the whole-cell patch clamp mode, histamine triggered a transient current in TRPC4-expressing cells. No current was evoked by perfusion with inositol 1,4,5-trisphosphate. When exocytosis was monitored with the capacitance detection technique, the magnitude of the membrane capacitance increase (Delta C(m)) on application of histamine in H(1) receptor/TRPC4-expressing chromaffin cells was comparable with that triggered by a train of depolarizing pulses. Our results indicate that TRPC4 channels behave as receptor, but not store-operated, channels in neuronally derived cells. TRPC4 channels can provide sufficient Ca(2+) influx to trigger a robust secretory response in voltage-clamped neurosecretory cells. Similar mechanisms may modulate exocytosis in other neuronal systems.

Transient receptor potential (TRP) channels form a large family of plasma membrane cation channels. Mammalian members of the "short" TRP family (TRP channel (TRPC) 1-7 are Ca 2؉ -permeant, non-selective cation channels that are widely expressed in various cell types, including neurons. TRPC activity is linked through unknown mechanisms to G-protein-coupled receptors or receptor tyrosine kinases that activate phospholipase C. To investigate the properties and function of TRPC4 in neuronally derived cells, we transiently expressed mouse TRPC4 and histamine H 1 receptor in mouse adrenal chromaffin cells and PC12 cells. Histamine, but not thapsigargin, stimulated Mn 2؉ influx in transfected cells. In the whole-cell patch clamp mode, histamine triggered a transient current in TRPC4-expressing cells. No current was evoked by perfusion with inositol 1,4,5trisphosphate. When exocytosis was monitored with the capacitance detection technique, the magnitude of the membrane capacitance increase (⌬C m ) on application of histamine in H 1 receptor/TRPC4-expressing chromaffin cells was comparable with that triggered by a train of depolarizing pulses. Our results indicate that TRPC4 channels behave as receptor, but not store-operated, channels in neuronally derived cells. TRPC4 channels can provide sufficient Ca 2؉ influx to trigger a robust secretory response in voltage-clamped neurosecretory cells. Similar mechanisms may modulate exocytosis in other neuronal systems.
The transient receptor potential (trp) Drosophila mutants were discovered in 1969 on the basis of their defective visual response to prolonged illumination (1). Hardie and Minke (2) demonstrated that the product of the trp gene functions as a Ca 2ϩ -permeable channel required for inositide-mediated Ca 2ϩ entry and that flies deficient in trp product lack sustained Ca 2ϩ entry. Since the cloning of the original trp gene (3), the "TRP" family of proteins has expanded rapidly. Recently, TRPs were subdivided into three groups, short, long, and Osm (4). Mammalian "short" TRP channels (TRPCs) 1 form a seven-member group of second messenger-operated, non-selective Ca 2ϩ -permeable cation channels (TRPC1-7) that can be activated by G-protein-coupled receptors or tyrosine receptor kinases. Phospholipase C plays a key role in stimulation of TRPC activity, although the specific mechanism of channel activation is unclear (5). Several activation mechanisms have been proposed, including stimulation by store depletion, direct coupling with IP 3 receptor/ryanodine receptors, and activation by second messengers such as diacylglycerol (6 -10).
TRPCs are widely expressed in mammalian tissues, and the mRNAs of some isoforms such as TRPC4 and TRPC5 are expressed predominantly in brain (11)(12)(13). To date, almost all studies of TRPC isoforms have been performed in non-excitable expression systems such as HEK293 cells, Chinese hamster ovary cells, and oocytes (5). There is no universal agreement on the ion selectivity and kinetic properties of individual TRPCs. At least part of the controversy arises because TRPC isoforms can exhibit different characteristics in different cellular environments (14).
Neurotransmitter release in neurons and neuroendocrine cells is triggered by a rise in [Ca 2ϩ ]. Voltage-gated Ca 2ϩ channels (VGCC) provide highly localized, rapid [Ca 2ϩ ] elevations. Additional Ca 2ϩ entry pathways, however, may trigger or modulate exocytosis. For example, numerous metabotropic presynaptic receptors can modulate synaptic release via second messenger systems. G-protein-coupled receptors that activate phospholipase C␤ cause release of Ca 2ϩ from intracellular stores and may activate additional cationic conductances such as store-operated or other cation channels. To determine whether TRPCs can respond to G-protein-coupled receptor signaling and mediate exocytosis in neuronal cells, we transiently expressed TRPC4 in mouse adrenal chromaffin cells, which are developmentally related to sympathetic neurons.
We found that stimulation of TRPC4 via the G q/11 -linked histamine H 1 receptor induced exocytosis in chromaffin cells co-transfected with TRPC4 and H 1 -R. Thapsigargin and IP 3 did not stimulate TRPC4 in the cells, suggesting that activation is not store-operated. Because TRPC4 is abundant in hippocampal pyramidal neurons and cortical neurons (11), we propose that activation of by agonists of metabotropic receptors may regulate secretory responses in neurons.

EXPERIMENTAL PROCEDURES
Molecular Biology-TRIzol Reagent (Invitrogen) was used to isolate total RNA from the mouse brain. The primers used for the polymerase chain reaction were 5Ј-GTCGACGCCACCATGGCTCAGTTCTAT-TACAAAAG (sense) and 5Ј-GGATCCGTTCACAATCTTGTGGTCA-CATAATC (antisense). The ␤ (accession number AAD10168) isoform of mouse TRPC4 was amplified from total RNA by employing SuperScript one-step reverse transcription-PCR with Platinum Taq System. The PCR product was subcloned into the pCR2.1 cloning vector using the Original TA cloning kit (Invitrogen) and then subcloned into the SalI/ BamHI sites of the pEYFP-N1 vector or the pEYFP-C1 vector (CLONTECH, Palo Alto, CA). The mouse H 1 histamine receptor was amplified using primers 5Ј-TCTCTATGGATTATGTGG (sense) and 5Ј-CGAGCAGAGAATGAATGC (antisense). The product was subcloned into the BamHI/XhoI sites of the pEYFP-N1 vector. The identity of the inserts were confirmed by sequencing.
PC12 cells (a kind gift of Dr. Lisa Elferink, Wayne State University) were maintained in F12-K medium supplemented with 15% heat-inactivated horse serum and 2.5% fetal bovine serum and were seeded onto poly-L-lysine-coated coverslips. Only undifferentiated PC12 cells were used in experiments.
Chromaffin cells as well as PC12 cells were transfected using Lipo-fectAMINE 2000 (Invitrogen). Mouse histamine H 1 receptor fused to EYFP (H 1 -R) was expressed alone or co-expressed with TRPC4 (ratio 1:4). Only about 25% of cells were found expressing EYFP. Cells that were co-transfected with both constructs (H 1 -R/TRPC4) exhibited enhanced cell death, possibly because of increased Ca 2ϩ influx and/or exocytosis. All experiments were performed 1-3 days after transfection.
Fluorescence Imaging-A monochromator-equipped imaging system (TILL-Photonics, Martinsreid, Germany) was used to monitor fluorescence changes in fura-2 (Molecular Probes, Inc., Eugene, OR) loaded cells. Cells were loaded with fura-2/AM (2-5 M) for 30 min in a fura-loading buffer at room temperature. After loading, cells were washed 3 times with loading buffer without fura-2/AM and then incubated for an additional 30 min at room temperature. Fura-2 was excited at 340 and 380 nm for determinations of [Ca 2ϩ ] i and at the isosbestic wavelength of 360 nm for determining Mn 2ϩ quench. Depending on loading, exposure times were 10 -100-ms intervals, indicated in the text. Emitted light was collected with a 510-nm long pass filter. An H Plan Fluotar 40ϫ oil immersion objective was used. Data were analyzed using TILLvisION software (version 3.02). Background fluorescence was subtracted. The intracellular concentration of Ca 2ϩ was determined as described by Grynkiewicz et al. (16). To study the cellular localization of mTRPC4 fusion proteins, a Nikon PCM2000 confocal microscope with a Plan 63ϫ oil-immersion objective was used. EYFP was excited at 488 nm, and emitted light was collected with a 515-nm long pass filter. The pin-hole diameter was set to ϳ1 Airy disc.
Electrophysiological Techniques-Whole-cell patch-clamp experiments as well as capacitance measurements were performed using a List EPC-7 patch-clamp amplifier. To monitor current through mTRPC4 channels, cells were voltage-clamped at a potential of Ϫ60 mV. Current amplitudes were obtained from voltage ramps. The ramps (1 mV/ms) from Ϫ100 to 100 mV were applied at 5-s intervals. The acquisition rate was set to 1 kHz, and currents were filtered at 3 kHz. Data were expressed as means Ϯ S.E.
A computer-based phase-tracking capacitance detection technique in the whole-cell patch clamp mode was used to study exocytosis in the cells (for details see Ref. 17). During capacitance measurements, a 15-mV root mean square, 1.2-kHz sine wave was added to the holding potential of Ϫ90 mV. Membrane capacitance (C m ) and conductance (G) phase angles were calculated and reset at the beginning of each capacitance trace (about 11 s) by switching in a 500-kiloohm resistor in series with ground. Capacitance changes were calibrated by adding 100-fF capacitor in the capacitance compensation circuitry of the patch-clamp amplifier. All experiments were performed at room temperature.

RESULTS
To study the role of TRPCs in mediating exocytosis we isolated the cDNAs of mouse brain ␤-TRPC4 isoform using a reverse transcription-PCR based approach and subcloned the PCR product into the multiple cloning site of the pEYFP-C1 vector upstream of the stop codon of EYFP. Expression of this construct resulted in formation of the EYFP-mTRPC4 fusion protein. We found that in confocal images of PC12 cells the fusion construct of ␤-TRPC4 was localized to the plasma membrane ( Fig. 1, 4 independent experiments). Because fusion of YFP to TRPC4 may affect the channel properties, we subcloned ␤-TRPC4 into pEYFP-N1. YFP is not expressed in the product of this construct because a stop codon follows the coding sequence for TRPC4. In all subsequent experiments, only the construct without fusion to YFP was used.
Expression of TRPC4 in Chromaffin Cells-In a non-excitable expression model, TRPC4 can be activated via phospholipase C␤ by G q/11 -protein-coupled receptors such as the histamine H 1 receptors (HEK-293 cells; Ref. 18). We tested whether mouse chromaffin cells expressed sufficient endogenous histamine H 1 receptors to stimulate phospholipase C␤. In fura-2loaded, non-transfected mouse chromaffin cells, histamine (20 M) evoked only small, infrequent intracellular Ca 2ϩ tran- sients. The same cells responded to high potassium-induced depolarization with a pronounced Ca 2ϩ transient due to Ca 2ϩ influx via voltage-gated calcium channels (averages of data from five cells; Fig. 2A). Similar results were observed in three independent experiments. After transient expression of mouse histamine H 1 receptors, histamine (10 M) consistently evoked a large Ca 2ϩ transient (Fig. 2B). Transient expression of H 1 -R did not impair the depolarization-activated Ca 2ϩ transient (Fig. 2B, note different time scale). In this and subsequent experiments we used mouse H 1 receptors fused to EYFP to serve as a marker of transfected cells.
The histamine-induced intracellular Ca 2ϩ transient could be due to Ca 2ϩ release from intracellular stores or extracellular Ca 2ϩ influx or both. To determine the source of the Ca 2ϩ , we measured Mn 2ϩ quench of fura-2 at the Ca 2ϩ isosbestic point (360 nm), which is a reliable indicator of divalent cation influx (19). In H 1 -R-expressing cells, the slow basal Mn 2ϩ leak was increased only slightly but consistently by histamine application (Fig. 3A, upper panel, n ϭ 5 independent experiments (coverslips)). In the same cells, histamine evoked a large increase in the fura-2 340/380 nm fluorescence ratio, indicating that the Ca 2ϩ transient evoked by histamine was largely caused by release of Ca 2ϩ from intracellular stores (Fig. 3A,  lower panel).
Most chromaffin cells co-expressing TRPC4 with H 1 -R, in contrast, displayed robust Mn 2ϩ quench after histamine application (Fig. 3B, upper panel, n ϭ 7 coverslips). In these cells, rapid Ca 2ϩ release from intracellular stores was followed by a second rise in [Ca 2ϩ ] i that occurred with a delay of 5-60 s (Fig.   3, A and B, lower panel, compare insets), due to Ca 2ϩ entry via TRPC4.
To test whether TRPC4 is gated by a store-depletion mechanism, we used thapsigargin (1 M), a sarco(endo)plasmic reticulum calcium ATPase pump inhibitor, which depletes intracellular Ca 2ϩ stores by preventing the re-uptake of passively leaked Ca 2ϩ . Application of thapsigargin resulted in slow depletion of intracellular Ca 2ϩ stores, which in some cells was manifested by a slow rise in [Ca 2ϩ ] i before clearance by extrusion or alternative uptake mechanisms (Fig. 3, C and D, lower panels). In H 1 -R-expressing cells, thapsigargin stimulated a detectable, but small Mn 2ϩ influx (n ϭ 10 coverslips, Fig. 3C). Histamine application after thapsigargin had little effect on Ca 2ϩ transients, indicating that stores were depleted. Thapsigargin application also had only a small effect on Mn 2ϩ influx in TRPC4/H 1 -R-expressing cells. Subsequent application of histamine, however, induced a marked Mn 2ϩ influx (n ϭ 4 coverslips, Fig. 3D). The fura-2 340/380 ratio showed a robust rise with a delay of 10 -60 s, similar to that observed in the absence of thapsigargin (Fig. 3B), indicating a Ca 2ϩ rise due to influx through TRPC4.
The Mn 2ϩ quench experiments indicate that ␤-TRPC4 can support divalent cation influx when expressed in mouse chromaffin cells. To examine the kinetics of the TRPC4 current, we performed whole-cell, patch-clamp recordings. TRPC4 currents were identified by applying voltage ramps from Ϫ100 mV to ϩ100 mV every 5 s. The extracellular recording solution contained NMDG ϩ as the sole monovalent cation to avoid contamination by voltage-gated Na ϩ channels. Before histamine ap- plication, the ramp evoked an inward current representing the activity of VGCC (Fig. 4A, inset, trace 1, the cell was clamped at Ϫ70 mV to reduce contribution of VGCC activation to the currents at a holding potential). Application of histamine (10 M) activated a large outward current via the TRPC4 channel (202.2 Ϯ 123.7 pA at ϩ100 mV, n ϭ 5, Fig. 4A). Under these conditions, TRPC4 currents exhibit a characteristic outward rectification due to Cs ϩ efflux at positive potentials (Fig. 4A,  inset, trace 2). The dip of the current-voltage relation between 0 and 60 mV results from voltage-dependent, Mg 2ϩ -induced inhibition of TRPC4 (18). The TRPC4 current inactivated spontaneously despite the continued presence of histamine ( decay ϭ 37.9 Ϯ 10.1 s, n ϭ 5). The rapid decay of TRPC4 currents was not due to internalization of H 1 -R because the intensity of EYFP fluorescence at the plasma membrane did not change after stimulation with histamine (data not shown). VGCC currents were effectively inhibited after activation of the histamine H 1 -receptor (Fig. 4A, trace 3), in agreement with a previous report (20).
Some TRPC channels may be activated by a direct interaction with IP 3 receptors (9). To test whether IP 3 receptors are involved in the activation of TRPC4 in adrenal chromaffin cells expressing H 1 -R/TRPC4, we dialyzed the cells with IP 3 (10 M).
In these experiments, the VGCC currents were markedly reduced or absent, probably due to Ca 2ϩ -dependent inactivation of VGCC after Ca 2ϩ release from intracellular stores by IP 3 . IP 3 did not stimulate TRPC4 outward current (Fig. 4B, trace 1). Subsequent application of histamine induced a current nearly identical to that evoked with no IP 3 in the pipette (164 Ϯ 86.5 pA at ϩ100 mV, n ϭ 6, Fig. 4B, trace 2). Currents were induced with a short delay after histamine application and reached a maximum within 10 -25 s. TRPC4 currents inactivated relatively quickly ( decay ϭ 32.8 Ϯ 9.3 s, n ϭ 6). No currents were recorded in control chromaffin cells expressing H 1 -R alone, with (n ϭ 5) or without IP 3 (n ϭ 6, Fig. 4C) in the pipette.
Chromaffin cells are a widely used model system for studying Ca 2ϩ -evoked exocytosis. Typically, a train of depolarizing pulses stimulated a robust rise in the cell capacitance, as illustrated in Fig. 5A (n ϭ 11), and the capacitance response was a function of the integral of the Ca 2ϩ current (Fig. 5B). We tested whether TRPC4 could supply sufficient Ca 2ϩ to support exocytosis in these cells. Capacitance changes were measured in chromaffin cells expressing either H 1 -R alone or a mixture of H 1 -R and TRPC4. Because inward currents through TRPC4 interfered with capacitance measurements (21), we substituted monovalent cations with NMDG ϩ . TRPC4 currents were monitored at positive voltages by applying depolarizing pulses at regular intervals of ϳ11 s. After application of histamine (10   n ϭ 6, Fig. 6). The increase of cell capacitance paralleled the development of the TRPC4 currents (n ϭ 6, Fig. 6A). A comparison of the amount of exocytosis triggered by depolarizing pulses or the activation of TRPC4 by histamine showed that both stimuli were almost equipotent in stimulating secretory responses in chromaffin cells: trains of 10 depolarizing pulses induced a capacitance increase of 369 Ϯ 94.5 fF (n ϭ 9, Fig. 6C), which is not significantly different from the value obtained on stimulation with histamine in chromaffin cells expressing TRPC4/ H 1 -R. Assuming the capacitance contribution of individual catecholaminergic vesicles is about 1.3 fF in mouse adrenal chromaffin cells (22), we estimated that, in chromaffin cells expressing TRPC4, the secretory response may correspond to the fusion of as many as about 300 vesicles at physiological extracellular [Ca 2ϩ ].
Expression of TRPC4 in PC12 Cells-PC12 cells, derived from pheochromocytomas of rat adrenal medulla, are a common model system for studying exocytosis. To test whether the histamine-evoked response is also observed in this neuroendocrine cell line, we expressed TRPC4 in PC12 cells. As in chromaffin cells, fura-2AM-loaded PC12 cells co-expressing TRPC4 and H 1 -R-EYFP displayed marked Mn 2ϩ influx upon H 1 -receptor stimulation (n ϭ 9, data not shown), which was not observed in control PC12 cells expressing only H 1 -R-EYFP (n ϭ 7). Thapsigargin did not stimulate mTRPC4 activity in TRPC4/ H 1 -R-EYFP co-expressing PC12 cells (n ϭ 7, data not shown). Histamine (10 M) stimulated TRPC4 currents in PC12 cells co-expressing TRPC4 and H 1 -R (Fig. 7A). The TRPC4 specific current exhibited the same current-voltage relationship during voltage ramps from Ϫ100 to ϩ100mV as in mouse chromaffin cells (see the inset in Fig. 7A; compare with Fig. 4). The outward current measured at ϩ100 mV was activated with a delay of 10 -30 s after histamine application and averaged 376.6 Ϯ 160.5 pA (n ϭ 4). As in chromaffin cells, TRPC4 currents decayed relatively quickly ( ϭ 23.9 Ϯ 3.9 s, n ϭ 4) in maintained histamine.
PC12 cells possess fewer secretory vesicles compared with primary chromaffin cells (15). Capacitance changes evoked by depolarization-stimulated exocytosis were smaller in these cells, probably as a result of both fewer vesicles and smaller voltage-gated Ca 2ϩ currents ( Fig. 7B; Ref. 23). In addition, Ca 2ϩ entry was less effective in triggering exocytosis because more Ca 2ϩ ions were required to stimulate a capacitance response in PC12 cells comparable with that in a chromaffin cell (compare Fig. 5B and 7B; note the different scale). Histaminestimulated capacitance responses were significantly larger in PC12 cells expressing TRPC4 and H 1 -R than in control H 1 -Rexpressing cells (130 Ϯ 18.1 versus 35.1 Ϯ 7.8, n ϭ 9 and 13, respectively; Fig. 7, C-E). In summary, even in cells that have less robust secretory responses, TRPC4 can provide sufficient Ca 2ϩ to support exocytosis. DISCUSSION Here we demonstrate that the receptor-operated cation channel, TRPC4, can provide sufficient Ca 2ϩ influx to support agonist-evoked secretion in both mouse primary chromaffin cells and rat-derived PC12 cells, two widely used neuronally related models for studying Ca 2ϩ -stimulated exocytosis. We used chromaffin cells and PC12 cells as a functional expression system in which both the G-protein coupled receptor, H 1 , and TRPC4 are transiently expressed. Our results demonstrate that all tested aspects of the G-protein-coupled receptor-signaling pathway and the secretory pathway are intact and functional in our neuroendocrine expression system. Both the H 1 -YFP receptor construct and ␤-TRPC4-YFP were appropriately targeted to the plasma membrane. Sustained histamine application inhibited VGCC for the duration of the recording in chromaffin cells expressing H 1 -R or H 1 -R/␤-TRPC4. Similar effects were previously observed by Currie and Fox (20) in non-transfected chromaffin cells. Histamine activation of TRPC4, on the other hand, was transient. A second application of histamine did not reactivate any TRPC4-like currents in the same cell.
Mammalian TRPCs were cloned by homology to the Drosophila trp gene, with the goal of identifying store-operated channels, such as I CRAC , that are critical for the function of certain cells of the blood lineage (24 -25). Currently, despite numerous reports postulating store-dependent mechanisms for TRPC stimulation, there is little experimental evidence for this hypothesis (5). The exceptions are reports on TRPC1 (26) and TRPC4 (27), although the last report has been disputed (5). Several other mechanisms of activation have been also proposed for each individual TRPCs. TRPC3, -6, and -7 have been shown to be activated by direct coupling with IP 3 receptor/ ryanodine receptors or direct stimulation by diacylglycerol (7)(8)(9)(10). ␤-TRPC4 and TRPC5, two closely related isoforms, are not stimulated by diacylglycerol; however, involvement of some other unidentified messenger downstream of phospholipase C was suggested (18). We corroborate that histamine acting on the histamine H 1 -receptor, a G q/11 -protein-coupled receptor, can stimulate ␤-TRPC4 in mouse chromaffin cells and rat PC12 cells. Neither thapsigargin nor dialysis with IP 3 stimulated ␤-TRPC4 in our experiments. In addition, ␤-TRPC4 was not sensitive to La 3ϩ , a classical inhibitor of store-operated channels. Therefore, our results do not support the hypothesis that the ␤-TRPC4 can form a store-operated channel in neuroendocrine cells, at least under our experimental conditions.
The few physiological functions attributed to TRPCs so far are diverse, reflecting the broad tissue distribution of the channels. Most studies have been carried out in non-excitable cells. TRPC1 is thought to be an important avenue for the agoniststimulated and store-operated Ca 2ϩ influx that regulates saliva flow in salivary glands (26). TRPC2 is prominent in the vomeronasal organ, where it may participate in pheromone transduction (27). In addition, TRPC2 appears to be activated during the mouse sperm acrosome reaction, a form of exocytosis (28). Mice deficient in TRPC4 have markedly reduced vasorelaxation and agonist-induced Ca 2ϩ entry in endothelial cells (29). The ␣ 1 -adrenoreceptor-activated cation channel that is involved in regulating the vascular tone in portal vein resembles TRPC6 (30). In the only report thus far on excitable cells, Li et al. (13) find that TRPC3 may contribute to neurotrophinstimulated Ca 2ϩ and Na ϩ influx in pontine neurons. This phenomenon is transient, occurring only during a specific stage of development. Thus, there is little information from which to predict the possible function of the numerous neuronal TRPCs.
Adrenal chromaffin cells are developmentally related to sympathetic neurons and widely used as a neuron-like model system for studying Ca 2ϩ -stimulated exocytosis. Capacitance changes in the cells directly correlate with the number of secreted vesicles. In chromaffin cells, as in neurons, rapid exocy-tosis is mediated by VGCC, which can provide high [Ca 2ϩ ] beneath the plasma membrane at the sites of exocytosis. At neuronal synapses, it is postulated that certain VGCC directly bind to synaptic vesicle proteins so as to provide Ca 2ϩ influx at precisely the locations that require it (31).
There are numerous reports that Ca 2ϩ pathways other than VGCC can regulate exocytosis in chromaffin cells. Cheek et al. (32) report that agonists of G q/11 -coupled receptors, such as histamine and angiotensin II, stimulated exocytosis in bovine adrenal chromaffin cells by a combination of Ca 2ϩ release from internal stores and additional subsequent Ca 2ϩ entry, which they postulated was "store-operated." Teschemacher and Seward (33) demonstrate that an angiotensin II-induced voltage-independent capacitance increase in bovine chromaffin cells was associated with a small leak current; however, the authors did not discuss the properties of the current. Fomina and Nowycky (34) demonstrate that exocytosis can be triggered by Ca 2ϩ entry via a small current activated on depletion of stores with thapsigargin in bovine chromaffin cells. Zerbes et al. (35,36) report that histamine can trigger exocytosis in bovine chromaffin cells via both store-operated and store-independent mechanisms.
It is not known whether any of these effects may be mediated by any of TRPC family members. Philipp et al. (37) report that TRPC4 is present in the adrenal gland but only in the cortex and not the medulla. A recent report, however, states that PC12 cells express mRNA for TRPC1-6 (38). In our experimental conditions, histamine evoked a small, rapid, and brief but significant increase in Mn 2ϩ influx in cells transfected with H 1 -R alone. reflecting the presence of some type of agonistand/or store-operated pathway ( Fig. 2A). In chromaffin cells transfected with TRPC4, however, there is a much larger his- FIG. 7. Transient expression of TRPC4/H 1 -R in PC12 cells. Histamine was applied during the time indicated by the horizontal bars. A, histamine-stimulated current in a PC12 cell expressing TRPC4/H 1 -R. The holding potential was Ϫ60 mV. Diamonds indicate current amplitudes measured at Ϫ70 mV (lower trace) and ϩ100 mV (upper trace) during the course of the experiment. The inset shows current-voltage relations recorded during voltage ramps from Ϫ100 mV to ϩ100 mV before (1) and after (2) the addition of histamine in the same cell. The [Ca 2ϩ ] out was 1.2 mM. B, depolarization-stimulated exocytosis in PC12 cells. A representative capacitance trace in a PC12 cell is shown. Exocytosis was elicited by a train of 10 depolarizing pulses, 40-ms duration to ϩ10 mV. The arrows indicate depolarizing pulses. The holding potential was set to Ϫ90 mV. The [Ca 2ϩ ] out was 10 mM. In the lower panel, a plot of cumulative ⌬C m versus cumulative Ca 2ϩ influx is shown. The Ca 2ϩ ion influx per pulse was determined from the time integral and is expressed as total charge. The inset shows the first (1) and last (2)  tamine-evoked divalent cation influx. Further studies will be needed to correlate the various inward currents activated by different experimental protocols.
In 1.2 mM external Ca 2ϩ the inward current at negative potentials had a small amplitude compared with that of VGCC. However, ␤-TRPC4 was as effective in mediating the secretory responses in chromaffin cells as VGCC stimulated by a train of 10 depolarizing pulses, although the secretory rate was very slow. There are a number of possible reasons for the similar efficacy. TRPC4 channels are open for many seconds, and Ca 2ϩ may accumulate in the cytoplasm or bind to Ca 2ϩ sensors for exocytosis. In addition, activation of phospholipase C␤ generates diacylglycerol, an activator of protein kinase C that potently facilitates Ca 2ϩ -evoked exocytosis in chromaffin cells (39 -41). Additional second messenger pathways may also be activated by sustained elevation of [Ca 2ϩ ] i . Because we do not have data to compare the absolute permeability of VGCC and TRPC4, a quantitative comparison between the efficacy of Ca 2ϩ influx through the two pathways to evoke exocytosis is not possible.
There is no evidence that TRPCs are closely associated with and can bind to synaptic vesicle proteins, as has been demonstrated for N-and P/Q-type VGCCs (42). In a recent study, TRPC4 and -5 were found to be associated with phospholipase C␤ and NHERF, a regulator of the Na ϩ /H ϩ exchanger (43). NHERF contains two PDZ domains and, through the second PDZ domain, associates with the actin cytoskeleton via interactions with members of the ezrin/radixin/moesin family. The authors termed the TRPC4/5 complex as a "signalplex" by analogy with the INAD scaffold localizing phospholipase C and TRP at the proper site in the rhabdomere of Drosophila photoreceptors (44 -45). Consistent with this finding in HEK293 cells, transiently expressed ␤-TRPC4 was distributed in a punctate manner in chromaffin and PC12 cells. We do not know whether such a signalplex may play a role in exocytosis in chromaffin cells.
In conclusion, we found that H 1 -R-dependent activation of transiently expressed TRPC4 can provide sufficient Ca 2ϩ influx to trigger a robust secretory response comparable with that activated by a train of depolarizing pulses in neurosecretory cells. The secretory response corresponded to the fusion of about 300 vesicles at physiological extracellular [Ca 2ϩ ] in chromaffin cells expressing TRPC4. Thus, together with VGCCs and store-operated channels (34,36), receptor-operated nonspecific cation channels appear to form an additional pathway for Ca 2ϩ entry to support secretory and possibly other Ca 2ϩdependent processes in neurosecretory cells. A similar mechanism may be involved in stimulating or modulating exocytosis in other neuronal systems and non-excitable cells under some physiological or pathophysiological conditions. Characterization of a functional expression system in a neuroendocrine cell will allow further study of both TRPC channels and the mechanisms by which they are activated.