Potassium Channels Regulated by Inositol 1,3,4,5-Tetrakisphosphate and Internal Calcium in DDTl MF-2 Smooth Muscle Cells*

This study was carried out to determine the intra- cellular components responsible for the transmembrane current evoked by stimulation of H1-histamin- ergic receptors in DDT, MF-2 smooth muscle cells. Histamine elicited an outward current that was reversed below the K+ equilibrium potential and passed voltage-independent K+ channels. A histamine concen-tration-dependent rise in outward current and in cy- toplasmic-free Ca2+ with similar time courses was observed. The histamine-induced current was not found after depletion of internal Ca2+ stores, suggesting a coupling between internal Ca2+ and K+ current. The time course of the initial increase in inositol (1,4,5)-trisphosphate (Ins (1,4,5)P3) caused by histamine dif- fers from that of the internal Ca2+ response. However, a significant concentration-dependent increase in ino- sitol (1,3,4,5)-tetrakisphosphate (Ins (1,3,4,5)P4) was seen during the whole stimulating period. The role of internal Ca2+, Ins (1,4,5)P3, and Ins (1,3,4,5)P4 on the outward current was also examined by the addition of these substances directly to the cytoplasm. Internal application of Ca2+ increased the amplitude and duration FRG). Inositol 1,3,4,5-tetrakisphos- phate sodium salt and inositol 1,4,5-trisphosphate sodium salt were obtained from Boehringer Mannheim. D-[2-%]inositol 1,4,5-tris- phosphate and ~-[2-'HH]inositol 1,3,4,5-tetrakisphosphate were from Du Pont-New England Nuclear. The other salts were of analytical grade (Merck).

This study was carried out to determine the intracellular components responsible for the transmembrane current evoked by stimulation of H1-histaminergic receptors in DDT, MF-2 smooth muscle cells.
Histamine elicited an outward current that was reversed below the K+ equilibrium potential and passed voltage-independent K+ channels. A histamine concentration-dependent rise in outward current and in cytoplasmic-free Ca2+ with similar time courses was observed. The histamine-induced current was not found after depletion of internal Ca2+ stores, suggesting a coupling between internal Ca2+ and K+ current. The time course of the initial increase in inositol (1,4,5)trisphosphate (Ins (1,4,5)P3) caused by histamine differs from that of the internal Ca2+ response. However, a significant concentration-dependent increase in inositol (1,3,4,5)-tetrakisphosphate (Ins (1,3,4,5)P4) was seen during the whole stimulating period. The role of internal Ca2+, Ins (1,4,5)P3, and Ins (1,3,4,5)P4 on the outward current was also examined by the addition of these substances directly to the cytoplasm. Internal application of Ca2+ increased the amplitude and duration of the histamine-induced current whereas internal EGTA suppressed the outward current. Internal Ins (1,4,5)P3 did not affect the histamine-induced K+ current, Ins ( 1,3,4,5)P4 inhibited the outward current, and the combination of Ins (1,3,4,5)P4 and Ca2+ abolished this response. The noradrenaline response evoked under normal conditions is not reflected by a change in transmembrane current or a change in Ins (1,3,4,5)P4 but is associated with an increase in Ins (1,4,5)P3 and internal Ca2+. Stimulation of a,-adrenoceptors, however, also evoked an outward current after the addition of Ins (1,3,4,5)P4 intracellularly. It is concluded that K+ channels, carrying the histamine outward current, are activated from the combined action of internal Ca2+ and Ins (1,3,4,5)P4.
Activation of transmembrane ion fluxes through stimulation of external receptor sites by transmitters or hormones is not yet fully understood. Agonist-receptor interaction may facilitate membrane currents by a direct action on ionic channels (Benham and Tsien, 1987) or indirectly via second messengers activating ionic channels from the inside (Hescheler et al., 1987). DDT, MF-2 smooth muscle cells, derived from hamster vas deferens, possess histaminergic receptors of the HI subtype, as demonstrated by receptor binding experi-* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ments (Mitsuhashi and Payan, 1988). An increase in intracellular Ca2+ upon HI histaminergic receptor stimulation was observed in swine carotid artery (Rembold and Murphy, 1989), in rat aortic smooth muscle (Matsumoto et al., 1989), and in DDTl MF-2 cells (Mitsuhashi and Payan, 1989). Stimulation of al-adrenoceptors also increased internal Ca2+ in a variety of smooth muscle cell types (Chiu et al., 1987;Nelemans and Den Hertog, 1987a;Minneman, 1988) including DDT, MF-2 cells (Nelemans et al., 1990). Intracellular Ca2+ originates from the extracellular space or from internal stores and functions to trigger the contractile apparatus in smooth muscle cells (Hartshorne, 1987). Ca2+ originating from the extracellular space may enter the cytoplasm, passing channel-like structures , or via channels coupled with the receptor (Benham and Tsien, 1987).
There is substantial evidence that phosphatidylinositol metabolism plays a physiological role in the release of calcium from intracellular structures. In particular, the metabolites inositol trisphosphate (Ins (1,4,5)P3)' and inositol tetrakisphosphate (Ins (1,3,4,5)P4) are considered to function as second messengers, contributing to the release of calcium from internal stores (Berridge and Irvine, 1989). It is noticed that GTP may function similar to Ins (1,3,4,5)P4 in DDT, MF-2 cells (Mullaney et al., 1988;Ghosh et al., 1989). The localization of the Ca2+ stores involved in the cellular response to receptor stimulation and their sensitivity to inositol phosphates is still under discussion (Volpe et al., 1988;Berridge and Irvine, 1989). Besides their action on Ca2+ stores, inositol phosphates may also play a role in the activation of ionic channels in the plasma membrane. Certain K+ channels are regulated by inositol phosphates as observed in mouse lacrimal acinar cells (Morris et al., 1987;Changya et al., 1989). In fact, phosphatidylinositol metabolism was activated in several smooth muscle cells in the presence of histamine (Villalobos-Molina and Garcia-Sainz, 1983;Donaldson and Hill, 1985;Bielkiewicz-Vollrath et al., 1987;Hall and Hill, 1988) and by stimulation of a,-adrenoceptors (Nelemans and Den Hertog, 198713;Minneman, 1988;Nelemans et al., 1990).

EXPERIMENTAL PROCEDURES
Cell Culture-The DDT, MF-2 cells, derived from a Syrian hamster vas deferens (Norris et al., 1974) were cultured in Dulbecco's modified Eagle's medium supplemented with 7 mM NaHCO:,, 10 mM HEPES at pH 7.2, and 10% fetal calf serum a t 37 "C in 95% 0 2 , 5% CO, (Molleman et al., 1989). In electrophysiological measurements the cells were plated on glass coverslips, and in Ca" measurements and inositol phosphate determination the cells were brought into suspension.
Membrane Currents-Microelectrode and whole-cell patch-clamp measurements were performed a t 20 "C as described earlier (Molleman et al., 1989). Microelectrodes (Clark GC150F-15 glass, Reading, England) were filled with 1 M KC1 and had a typical resistance of 50-80 megohms. Cells superfused at a constant flow rate (1 ml/min) by means of a multichannel peristaltic pump (IPS Ismatec, Zurich, Switzerland) were penetrated with a microelectrode using a nanostepper (model B, World Precision Instruments, New Haven, CT), and membrane potentials were recorded by a high resistance amplifier (model M4, World Precision Instruments, New Haven, CT) on a paper writer (Kipp BD-8, Delft, The Netherlands). Patch pipettes (Clark GC150TF-15 glass) were heat polished and filled with intracellular solution and had a typical resistance of 2-10 megohms. Establishment of the whole-cell patch-clamp configuration was monitored via a CCD camera (Sony AVD-D5CE) mounted on the microscope (Olympus CK) and detected electrically by measuring the resistance and capacity. Membrane currents of the cells were recorded under voltage-clamp (List EPC-7, Federal Republic of Germany).
Data are presented as mean f S.E. and considered to be significantly different ( p < 0.05) from control values using Student's t test.
Intracellular Calcium-Cytoplasmic free Ca" levels were determined by indo-1 fluorescence, as reported previously (Hoiting et al., 1990;Nelemans et al., 1990). Cell suspensions a t a density of 2 x lo5 cells/ml were loaded with indo-1 ester (2 p~) for 45 min a t 37 "C. Recordings were made at an excitation wavelength of 325 nm and emission wavelengths of 400 and 480 nm at 20 "C using a fluorescence spectrophotometer (Hitachi). Cytoplasmic free Ca' ' concentrations were calculated (Hesketh et al., 1983) using 0.015% Triton X-100 as a permeabilizing agent.

RESULTS AND DISCUSSION
The membrane current, the intracellular free Ca2+ concentration, and the formation of inositol phosphates were measured upon stimulation of DDTl MF-2 cells by histamine. After establishment of the whole-cell patch, the cells showed an outward current on exposure to histamine (Fig. 1A). The transmembrane current evoked by histamine was transient in nature and reached a maximum (0.5 f 0.1 nA; n = 19) after about 30 s at a holding potential (-50 mV) near the resting potential (-48 f 6 mV, n = 8) of the cells. The histamineinduced current increased linearly at positive holding potentials and was reversed at values below the potassium equilibrium potential (-76 mV; Fig. 1C). These data indicate that the outward current is carried by K+ ions since this current could not be evoked by histamine after replacing cytoplasmic K+ with Cs+ via the patch pipette (not shown). Hyperpolarization of the cells in the presence of histamine, measured by using microelectrodes, reached a maximum (-32 f 2.4 mV; n = 6) after about the same time as the outward current (  2. The relationship between the histamine concentration and the evoked cellular response. A , the maximum current measured by using the whole-cell patch-clamp method (holding potential, -10 mV). The response to M histamine was taken as 100% ( n = 4). B, the internal Ca'+ concentration determined by using indo-1 fluorescence. C, the Ins ( 1,3,4,5)P4 formation expressed as dps/ lo6 cells and analyzed using HPLC. All data represent increases above basal values as given in Table I. potassium current can readily explain hyperpolarization of the cells in the presence of histamine. The similarity of both responses also indicates that dialysis of the cells with ICs in the whole-cell patch-clamp configuration does not interfere with the cellular response to histamine. Therefore, the EGTA concentration (7.7 x M) used in this study to avoid spontaneously developing current fluctuation in the absence of this chelator' is considered acceptable. It is noticed that the histamine-induced current was abolished by dialyzing the cell with a 4-fold higher concentration of EGTA ( 3 X

M;
see Fig. 6B). In addition to the voltage insensitivity of the histamine-evoked K+ current, it was not affected by apamin ( 3 X M), known to block certain receptor-activated K+ channels (Den Hertog, 1982). It was also unaffected by 3,4diaminopyridine ( M) inactivating slow rectifying K' conductances (Okabe et al., 1987) or by glipizide ( M) blocking ATP-mediated K+ channels (Hescheler et al., 1987; not shown). Thus, the outward current evoked by histamine in DDTl MF-2 cells presumably passed novel K+ channels, which are voltage independent (Fig. 1C) and insensitive to established channel blocking agents.
The amplitude of the outward current increased at higher agonist concentrations to reach a maximum at about M histamine (Fig. 2 A ) . The exclusive expression of H1 histaminergic receptors on these cells was demonstrated by binding experiments (Mitsuhashi and Payan, 1989). In agreement, blocking Hz histaminergic receptors by cimetidine

M)
did not affect the outward current elicited by histamine M), but mepyramine M), acting as a HI histaminergic receptor antagonist, blocked this response completely (not shown). The experiments described below are performed at a holding potential of -10 mV in case of the whole-cell patchclamp and are showing cellular responses to a maximal effective histamine concentration ( M), if not stated otherwise. It was found in several cell types that changes in membrane current are linked to the internal Ca2+ concentration (Morris et al., 1987;Molleman et al., 1989;Hoiting et al., 1990). To examine whether such a coupling also exists with respect to the histamine-induced K+ current in DDTl MF-2 cells, internal Ca'+ was measured. Stimulation of the HI histaminergic receptors caused a concentration-dependent rise in internal Ca'+ ( Fig. 2B) followed by a low sustained phase (Fig. 3B). The sustained component (12 f 4 nM; n = 8) amounted to less than 10% of the maximum value (Table I) in agreement with previous reports (Mitsuhashi and Payan, 1988). This increase in internal Ca'+ with histamine is not affected by blocking voltage-dependent Ca2+ channels with diltiazem M; n = 4; not shown). The sustained phase, however, A. Den Hertog, personal observation.

Ins (1,3,4,5)P4 and [Ca2+Ji
was not observed in the absence of external Ca'+ (15 min; Fig.  3B) and accordingly is most likely due to Ca'+ entry from the extracellular space via a voltage-insensitive pathway. The remaining transient increase in Ca2+ caused by histamine under Ca2+-free conditions could be evoked only once (Fig.  3 B ) . The transient rise in cytosolic Ca'+ is obviously due to release from internal stores, a process observed in many cell types (Den Hertog, 1981;Takemura et al., 1989;Irvine, 1990).
Refilling of the stores with Ca2+ from the extracellular space is required to evoke a second Ca'+ response with histamine under Ca'+-free conditions (not shown), indicating that reuptake from the cytoplasm, as in parotid acinar cells, does not play an important role .

TABLE I The effect of histamine and noradrenaline stimulation on the outward current (I,,,), internal Ca'+ ([Ca'+]J and the formation of Ins
Maximal changes in the amplitude of the outward current at a holding potential of -10 mV and in [Ca'+], are presented. The inositol phosphate values were obtained 30 s after stimulation and represent changes from the control level. Values obtained under extracellular Ca2+-free conditions are marked with asterix (*). Data related to noradrenaline were obtained previously (Nelemans et al., 1990). ND, not detectable (background 0.5 dps/106 cells; 10 PA). The number of exDeriments is in uarentheses.

Control
Histamine Noradrenaline caeci smooth muscle cells (Den Hertog, 1981;Okabe et al., 1987) and DDTl MF-2 cells (Hoiting et al., 1990;Nelemans et al., 1990) on stimulation of Pz purinoceptors or al-adrenoceptors. The changes in internal Ca2+ under different experimental conditions described here are accompanied by an outward current following similar time-related characteristics ( Fig. 3A; Table I). It is noticed that the inability to evoke a second histamine-induced response under Ca2+-free conditions is also reflected by the absence of a change in transmembrane current (Fig. 3A).
The results obtained so far are consistent with the concept that internal Ca2+ is involved in activation of K+ channels.
Several models are presented to describe the inositol phosphate-regulated Ca2+ release from internal stores Takemura et al., 1989;Berridge and Irvine, 1989).
Stimulation of Ins (1,4,5)P3-sensitive receptors on the endoplasmic reticulum or related structures (Volpe et al., 1989) is thought to cause Ca2+ release. In view of the inositol phosphates formed throughout stimulation of the cells with histamine, receptor desensitization or lack of inositol phosphates cannot account for the transient characteristics of the rise in internal Ca2+. The transient nature of the Ca2+ response is explained in terms of a quantal release process (Muallem et al., 1989) or a feedback mechanism (Irvine, 1990). In the quantal release process the Ca2+ stores are supposed to display different sensitivities to Ins (1,4,5)P3, and in the feedback process regulation of the Ins (1,4,5)P3 receptor affinity is considered to occur by stored Ca2+ itself. The shortlasting transient increase in Ins (1,4,5)P3 upon HI histaminergic    (1,3,4,5)P, promoted Ca2+ release from internal stores and opened potassium channels, resulting in a transmembrane outward current, which events were observed by stimulation of HI histaminergic receptors.
Ca2+ release process is attributed to translocation of Ca2+ toward the Ins (1,4,5)P3-sensitive store (Berridge and Irvine, 1989;Irvine, 1990). Thus, Ins (1,4,5)P3 and Ins (1,3,4,5)P4 are apparently both involved in the generation of the Ca2+ release, in particular in the onset and in the development of the Ca2+ response. Based on the histamine concentration-related action and the time-related characteristics it is concluded that the inositol phosphate formation, enhancement of internal Ca2+, and the potassium current are related phenomena in DDTl MF-2 cells.

,5)P4
and Ca2+ applied to the cytoplasm caused a sustained outward current. Thus, preactivation of K' channels under these conditions prevented the noradrenaline-or histamine-induced current to become prominent (Fig. 7B).
A simplified model, depicting the action of internal Ins (1,4,5)P3, Ins (1,3,4,5)P4, and Ca2+ on the transmembrane current, achieved via receptor stimulation or by direct addition, is presented (Fig. 8). These results show for the first time that the combined action of internal Ca2+ and Ins ( 1,3,4,5)P4 is required to activate K+ channels responsible for the histamine-evoked transmembrane current in DDTl MF-2 cells.