Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Granulosa cells (GC) surround the oocyte within the ovarian follicle and play an essential role in creating the conditions required for follicular development, ovulation, fertilization, and implantation (Salustri et al., 1993). During folliculogenesis, GC undergo a series of mitotic divisions (proliferation) then acquire gonadotropin receptors and enhanced steroidogenic activity (differentiation). Autocrine-paracrine and endocrine regulation of the granulosa cell maturation have been extensively investigated and much is known about the specific roles of various growth factors, hormones, transmembrane receptors, and second messengers (Steele and Leung, 1993). In contrast, the functional significance of voltage-dependent ion channels in GC is far from understood.

It has been reported that GC can generate action potentials (Mealing et al., 1994), and indirect evidence suggests that modulation of granulosa cell electrical activity may provide a means to regulate cell function (Mattioli et al., 1990, 1991, 1993; Kusaka et al., 1993). For example, granulosa cell depolarization has been described as a consistent feature of oocyte maturation in different experimental systems (Mattioli et al., 1990). Voltage-gated potassium currents with distinct electrophysiological and pharmacological properties have been described in both acutely isolated and cultured GC, and have been shown to regulate granulosa cell resting membrane potential (Mattioli et al., 1991, 1993; Kusaka et al., 1993). Delayed rectifier currents with both slow and rapid activation and inactivation kinetics have been described in porcine GC (Mattioli et al., 1991, 1993; Kusaka et al., 1993). However, the molecular correlates of these currents have not been identified.

The major objective of this study was to elucidate the molecular bases for K⁺ current diversity in porcine GC. Specific antibodies were used to demonstrate directly the existence in GC of Kv1 (KCNA) and KCNQ1 (KvLQT1) channel proteins, as well as Kv (KCNAB) and KCNE1 (minK or IsK) auxiliary subunits. Coimmunoprecipitation was used to establish the hetero-oligomeric nature of granulosa cell Kv channels. Whole-cell patch-clamp techniques were used to record granulosa cell K⁺ currents and to document the specific effects of a variety of K⁺ channel antagonists.

We report not only that a variety of voltage-gated K⁺ channel - (pore-forming) and - (accessory) subunits are present in freshly isolated GC but also that their expression is temporally regulated as GC spontaneously differentiate (luteinize) in culture. Our data suggest that the diverse electrophysiological and pharmacological properties of native granulosa cell K⁺ currents, described here and elsewhere (Mattioli et al., 1991, 1993; Kusaka et al., 1993), may reflect variation in the composition and stoichiometry of the hetero-oligomeric channel complexes, as well as differences in post-translational modification of contributing channel subunits. The documented potential for dynamic interchange of various associated K⁺ channel subunits as a function of the cells' metabolic status suggests that these ion channels may participate in control of granulosa cell proliferation or differentiation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents. Cell culture media, supplements, and sera were obtained from Invitrogen (Carlsbad, CA), unless stated otherwise. Chemicals were obtained from Sigma Chemical (St. Louis, MO), unless stated otherwise. Regular pork insulin and LY-97241 were obtained from Eli Lilly (Indianapolis, IN). Margatoxin and charybdotoxin were obtained from Alomone Laboratories (Jerusalem, Israel). Anti-phosphotyrosine cocktail (PY-Plus-HRP) was purchased from Zymed Laboratories (South San Francisco, CA). Polyclonal rabbit antibodies directed against Kv1.1, Kv1.2, Kv1.3, Kv1.5, and Kv1.6 -subunits were purchased from Alomone Laboratories. Mouse monoclonal antibodies directed against Kv1.1, -1.2, -1.4, and -1.5 were obtained from Upstate Biotechnology (Lake Placid, NY). Primary antibodies directed against HERG were a gift from J. Nerbonne (Pond et al., 2000). Primary antibody directed against KvLQT1 was a gift from D. Roden and S. Kupershmidt (Vanderbilt University, Nashville, TN). Primary antibody directed against minK was a gift from R. Dumaine (Masonic Medical Research Laboratory, Utica, NY). Primary antibody directed against Kv1.3 was a gift from M. Tamkun (Colorado State University, Ft. Collins, CO). Nitrocellulose membranes (Hybond ECL), secondary antibodies, chemiluminescence reagent (ECL), and film (Hyperfilm ECL) were obtained from Amersham Biosciences, Inc. (Piscataway, NJ). Porcine FSH was a gift from the National Hormone and Pituitary Program. Compound 293B was a gift from Aventis (Frankfurt, Germany). ()-(3R,4S)-293B was obtained from Procter & Gamble (Cincinnati, OH). Compounds MK-499, L-735,821, and L-768,673 were obtained from Merck Research Laboratories (Westpoint, PA). HEK-293 cells stably transfected with HERG were provided by Z. Zhou and C. T. January (University of Wisconsin, Madison, WI). HERG cDNA was obtained from Robert Kass (Columbia University, New York, NY) Primers for non-nested PCR of KCNE1 (~170 bp) were supplied by M. Scofield (Creighton University, Omaha, NE).

Granulosa Cell Isolation and Culture. GC were isolated from the ovaries of mature swine by using techniques described in detail previously (Barano and Hammond, 1985). Briefly, small (1-3 mm in diameter) to medium (4-6 mm in diameter) follicles were aspirated by hand with a 19-gauge needle attached to a 10-ml syringe. GC were separated from follicular fluid by centrifugation at 500g for 5 min. Cells were washed twice with a 1:1 mixture of Ham's F-10 nutrient medium and Dulbecco's modified Eagle's medium (DMEM) containing HEPES (25 mM), penicillin (50 U/ml), and streptomycin (50 µg/ml). For RNA isolation or membrane protein preparation, the granulosa cell pellet was washed again with phosphate-buffered saline, flash frozen in liquid nitrogen, and kept at 80°C.

Electrophysiological Recordings. Membrane currents were recorded using standard patch-clamp procedures in the whole-cell configuration. Voltage-clamp protocols and solutions for measuring slow (I_Ks) and ultrarapid (I_Kur) cardiac delayed rectifier currents under whole-cell recording conditions have been described in detail previously (Arena and Kass, 1988; Nattel et al., 1999); similar protocols were used to elicit granulosa cell currents. Briefly, recording pipettes were pulled to resistances of 2.5 to 6 M when filled with intracellular (pipette) solution containing 110 mM potassium aspartate, 1 mM MgCl₂, 11 mM EGTA, 1 mM CaCl₂, 10 mM HEPES, 10 mM K₂ATP, pH 7.3, attained by addition of 1 N KOH to bring the final potassium concentration to 140 mM. Extracellular recording (bath) solution consisted of 132 mM NaCl; 1.2 mM MgCl₂; 1 mM CaCl₂; 5 mM glucose; 0.0, 0.5, or 4.8 mM KCl; 10 mM HEPES, pH adjusted to 7.4 with NaOH. The reference electrode was an Ag/AgCl half-cell immersed in the pipette solution and connected to the bath via a 3 M KCl-agar salt bridge. Tip potentials were zeroed before seal formation. All recordings were performed at room temperature (22-24°C) from a Plexiglas chamber mounted on an inverted microscope (Nikon Diaphot 300; Nikon, Tokyo, Japan). Data acquisition and analysis were accomplished using an IBM compatible computer interfaced to an Axopatch 200-A amplifier driven by pClamp software (Axon Instruments, Union City, CA). The standard voltage-clamp protocol for activation of I_Ks consisted of a series of 1- to 4-s depolarizing test pulses from a holding potential of 40 mV to test potentials (V_t) ranging from 40 to +60 mV at an interpulse interval of 14 s. I_Kur was elicited from holding potentials of either 80 to 40 mV by 80- to 1000-ms depolarizing test pulses to V_t between 40 and +60 mV.

Reverse Transcription-Polymerase Chain Reaction. mRNA was isolated from fresh GC by using a Micro-Fast Track kit (Invitrogen). Reverse transcription was performed using random hexamers and enhanced avian reverse transcriptase (Sigma Chemical) under recommended conditions. PCR detection of KCNE1 was performed using gene-specific oligonucleotide primers designed to amplify a 170-bp fragment of sense: 5'-ACCCTGGGCATCATGCTGAGT-3'; antisense: 5'-TGCCGCCTGGTTTTCAATGAC-3'. Four additional primers were used for nested PCR designed to amplify an ~180-bp KCNE1 product for the outer reaction [5'-AAGCTGGAAGCACTCTAC-3'; 5'-CTCCAGAACCCGGGCCTG-3'], and an ~110-bp KCNE1 product for the inner reaction [5'-GGTTTCTTCACTCTGGGC-3'; 5'-CTTCTCCTGCCAGGCGTC-3']. The PCR reaction mixtures (25 µl) contained 0.4 mM each 5' and 3' primer, 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl₂, 0.2 mM each dNTP, and 1 unit of REDTaq DNA polymerase (Sigma). These PCR reactions were amplified for 30 cycles consisting of denaturation for 30 s at 94°C, annealing at 60°C for 1 min, and extension at 72°C for 1 min. The PCR product was cloned into INVaF' cells by using an Original TA Cloning kit (Invitrogen) and sequenced using T7 and M13 primers.

Preparation of Granulosa Cell Lysates and Membrane Proteins. Whole-cell lysates were made from GC monolayers by standard techniques with a lysis buffer consisting of phosphate-buffered saline with 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, and protease inhibitor cocktail (1:500, P8340; Sigma Chemical). Lysis buffer was added to the culture dish after washing with cold phosphate-buffered saline three times. The culture dishes were scraped and the lysate was aspirated into a syringe with a 21-gauge needle to shear DNA. The lysates were rocked in the cold for 1 h and centrifuged for 10 min at 10,000g.

Immunoblotting. Solubilized proteins were separated by SDS-PAGE under reducing conditions and transferred to nitrocellulose membranes by the semidry transfer method. The membranes were blocked for 1 h at room temperature with 5% nonfat milk in Tris-buffered saline (TBS; 100 mM Tris, 0.9% NaCl, pH 7.5) containing 0.1% Tween 20 then incubated overnight at 4°C with primary antibody diluted in the blocking solution. Primary antibody dilutions were -Kv1.1 (1:1000), -Kv1.2 (1:1000), -Kv1.3 (1:500), -Kv1.4 (1:500), -Kv1.5 (1:500), -Kv1.6 (1:500), -KvLQT1 (1:500), -minK (1:500), and -HERG (1:500). Unless otherwise stated, after three washes with 0.1% Tween 20/TBS, membranes were incubated for 1 h at room temperature with the appropriate HRP-conjugated secondary antibody diluted 1:1500 in 0.1% Tween 20/TBS. After four additional washes with 0.1% Tween 20/TBS, bound primary antibodies were visualized using an ECL detection system (Amersham Biosciences, Inc.) and recorded on radiographic film. Densitometric analysis was performed using Scion Image beta version 4.02 (Scion Corporation, Frederick, MD). Blots that were reprobed with antiphosphotyrosine were stripped with IgG elution buffer (Pierce Chemical) for 30 min then washed 3 × 10 min with TBS. Blots were then treated with ECL reagent as described for Western blots to confirm that antibodies were stripped completely. The blots were then reprobed with HRP-linked antiphosphotyrosine (1:500) and developed as described above.

Deglycosylation. The product of a Kv1.3 immunoprecipitation reaction was subjected to enzymatic deglycosylation by using peptide-N-glycosidase F (PNGase-F) according to the manufacturer's instructions (Glyko Deglycosylation Plus kit; Glyko, Inc., Novato, CA). Briefly, the glycoprotein was denatured before digestion by using SDS and -mercaptoethanol. Samples were heated to 100°C for 5 min then cooled on ice. Excess Nonidet P-40 was added to the cooled reaction to complex the SDS then PNGase-F was added (0.2 U/ml) and the sample was incubated overnight at 37°C. A control reaction containing buffer instead of enzyme was incubated in parallel. After incubation, sample buffer was added, and the reactions were analyzed by SDS-PAGE.

Immunoprecipitation. To preclear nonspecific binding, protein G-agarose beads (Immunopure; Pierce Chemical) were incubated with samples for 2 h at 4°C with continuous gentle agitation then pelleted by centrifugation (700g). Antibodies directed against channel proteins were subsequently added to the supernatant and mixed overnight at 4°C. Immune complexes were then immobilized onto protein G-agarose beads, washed three times with cold radioimmunoprecipitation assay buffer, and eluted from the beads with SDS sample buffer containing 5% -mercaptoethanol. Immunoprecipitated proteins and supernatant fractions were analyzed by immunoblotting as described above.

Statistical Analysis. Data are expressed as mean ± S.E.M. Significant differences between treatment groups were identified by analysis of variance with appropriate general linear models. Multiple comparisons were made using least significant difference procedure (Statistix; Analytical Software, Tallahassee, FL). Differences were considered significant when p < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Electrophysiological Differentiation of Delayed Rectifier K⁺ Currents in Porcine Granulosa Cells. Two different outward potassium currents with distinct activation and inactivation kinetics could be elicited by a series of depolarizing test pulses from a 40-mV holding potential: a slowly activating, noninactivating current (Fig. 1) and an ultrarapidly activating, slowly inactivating current (Fig. 2). A discrete rapidly activating and inactivating (A-type) current was rarely seen when a similar voltage protocol was combined with a 80-mV holding potential (2/50 cells); this current was not further characterized.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   A, left, slowly inactivating, nonactivating delayed rectifier K+ current (IKs) elicited by a series of 4-s depolarizing test pulses (0 to +60 mV, step 20), followed by return to the 40-mV holding potential [K+]out = 0 mM, 22°C. Right, normalized isochronal (4-s) activation curve for IKs determined from tail currents (n = 9). Continuous curve is a Boltzmann relationship fit to the mean at each test potential. The half-activation voltage (V1/2) and slope factor (k) are 12.8 and 16.9 mV, respectively. B, left, granulosa IKs recorded at +40 mV by using the protocol described above, in the presence of increasing concentrations of the 3R,4S-enantiomer of chromanol 293B. Right, isochronal activation curves recorded in the absence and presence of 10 µM racemic 293B (n = 3). Data were normalized to the maximal control (drug-free) conductance. Continuous curves are Boltzmann relationships fit to the mean data. C, left, granulosa IKs recorded at +40 mV by using the protocol described above, before and after addition of 10 nM L-735,821. Right, granulosa IKs recorded at +40 mV by using the protocol described above, before and after addition of 20 nM L-768,673.

View larger version (21K): [in this window] [in a new window]   Fig. 2.   A, two examples of IKur recorded from granulosa cells maintained in monolayer culture for 48 h before patch-clamp recording. The current traces in the left and center were elicited by an identical series of 1-s depolarizing test pulses (0 to +40 mV, step 20), followed by return to the 40-mV holding potential. Scale bars, 250 ms, 25 pA. Right, current density plotted as a function of test potential for the traces on the left (, solid line) and center (, dash-dot line). B, frequency distributions for IKur half-activation voltage (left, n = 22), inactivation time constant at +40 mV (center, n = 18), and ratio of peak to sustained current at +40 mV (n = 23).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Cumulative (use-dependent) inactivation of IKur was variable. Top, currents elicited from a granulosa cell maintained in monlayer culture for 96 h by a train of seven depolarizing voltage steps to +40 mV once every second from a holding potential of 80 mV. Cumulative inactivation can be visualized as a reduction in current amplitude with each successive pulse. Bottom, currents elicited from another granulosa cell maintained in monolayer culture for 96 h by using the voltage protocol described above. Cumulative inactivation is not apparent.

View larger version (19K): [in this window] [in a new window]   Fig. 4.   A, IKur recorded from granulosa cell maintained in monolayer culture for 72 h before patch-clamp recording has a moderately TEA-sensitive component. Left, control current recorded in the absence of TEA. Middle, TEA-resistant current recorded 5 min after application of 10 mM TEA (extracellular solution with 10 mM sodium replaced by 10 mM TEA). Right, TEA-sensitive current obtained by subtracting TEA-resistant current from control. Scale bars, 250 ms, 50 pA. B, peptide toxins inhibit components of IKur. Left, inhibition by CTX (10 nM) of IKur recorded from a granulosa cell after 48 h in culture. Middle, inhibition by DTX (20 nM) of IKur recorded from a granulosa cell after 48 h in culture. Right, inhibition by KTX (50 nM) of IKur recorded from a granulosa cell after 72 h in culture. Scale bars, 125 ms, 50 pA. C, effects of nonspecific K+ channel antagonists on IKur recorded from granulosa cells maintained in monolayer culture for 48 h before patch-clamp recording. Effects of clofilium, 4-AP, and verapamil are shown in the left, middle, and right, respectively. Similar drug effects were observed in granulosa cells cultured for 72 or 96 h. Scale bars, 125 ms, 50 pA.

Potassium Channel Transcripts and Proteins Expressed in Porcine Granulosa Cells. Our goal was not only to increase knowledge about the electrophysiological and pharmacological properties of GC delayed rectifier K⁺ currents but also to determine which potassium channel subunits might contribute to GC K⁺ channels. To this end, we used qualitative RT-PCR and Western analysis to examine K⁺ channel protein expression at the mRNA and protein levels, respectively. To address directly the issue of molecular diversity in GC Kv channel assembly, we used sequential coimmunoprecipitation and immunoblotting to document not only the presence of but also the potential for heteromultimer formation by channel subunits from the ERG (KCNH), KCNQ, KCNE, Kv1, and Kv families.

View larger version (38K): [in this window] [in a new window]   Fig. 5.   ERG1 (KCNH2) is not expressed in GC. Lysates (15 µg) of CHO cells transiently transfected with HERG cDNA and porcine granulosa cells were fractionated by SDS-PAGE, transferred to nitrocellulose, and immunoblotted (IB) with C-terminal anti-HERG antibody. As expected, a single HERG1 protein (~145 kDa) is detected in the transfected cells (Pond et al., 2000). ERG1 is undetectable in GC. Arrows indicate molecular mass. Results representative of four experiments.

View larger version (50K): [in this window] [in a new window]   Fig. 6.   Top, KCNE1 mRNA expression in porcine granulosa cells. RT-PCR was performed using gene-specific primers. Amplified products were separated by gel electrophoresis and visualized by ethidium bromide staining. Left, predicted product of 170 bp. Arrow indicates 200 bp. Lanes: 1) 50-bp ladder; 2) porcine granulosa cell poly(A)+ mRNA. Right, nested PCR with predicted products of 183 bp (outer reaction) and 110 bp (inner reaction). Arrow indicates 150 bp. Lanes: 1) 100-bp ladder; 2) 50-bp ladder; 3) porcine granulosa cell poly(A)+ mRNA, outer reaction; 4) porcine granulosa cell poly(A)+ mRNA, outer reaction (primers removed with spin column); 5) lane 4 product, inner reaction; 6) guinea pig heart poly(A)+ mRNA, outer reaction; 7) lane 6 product, inner reaction; and lane 8, no DNA control, inner reaction. Bottom, partial deduced amino acid sequence of pig granulosa KCNE1 (GenBank accession number AF233358) compared with other KCNE clones.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   KCNQ1 and KCNE1 expression in freshly isolated porcine GC. Top, KCNQ1 immunoreactivity in membranes (10 µg) prepared from GC, guinea pig heart, and untransfected CHO cells. KCNQ1 antibody (-KCNQ1) was applied to the membrane directly () or after preincubation (+) with 10 µg/ml of peptide against which the antibody was generated. Bottom, left, KCNE1 immunoreactivity in membranes (20 µg) from GC, CHO cells transfected with KCNQ1 (CHO-Q1), and CHO cells cotransfected with KCNQ1 and KCNE1 (CHO-Q1/E1). In this instance, bound KCNE1 antibody (-KCNE1) was visualized using amplified alkaline phosphatase (Bio-Rad, Hercules, CA). Three bands corresponding to differentially glycosylated forms of KCNE1 are visible. Bottom, middle, Immunoprecipitate (IP) of GC membranes by -KCNE1, resolved by SDS-PAGE on a 4 to 20% gel, and immunoblotted (IB) with -KCNQ1. Bottom, right, IP by -Kv1.6 and the associated supernatant (SP) were probed with -KCNQ1 as a negative control. Arrows indicate molecular mass.

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Changes in expression of Kv channels during culture. Left, immunoblot (IB) of GC membrane proteins (30 µg) from fresh isolates (F) and 24 h primary cultures (C), separated by SDS-PAGE on a 4 to 20% gel, and probed with antibody to KCNE1 (top) or KCNQ1 (bottom). Right, immunoblots of membrane proteins (30 µg) isolated from freshly harvested PGC cells (F) or PGC maintained in culture (C) for 72 h. Membrane proteins were separated by SDS-PAGE on a 4 to 20% gradient gel, transferred to nitrocellulose, and detected with anti-Kv1.2 (top), anti-Kv1.5 (middle), or anti-Kv1.6 (bottom). Arrows indicate molecular mass.

View this table: [in this window] [in a new window]   TABLE 2 Molecular basis for IKur Interactions between different Kv -subunits as demonstrated by co-immunoprecipitation. Where low- and high-molecular-mass forms of Kv protein were detected, the relative amounts that coassociate are indicated. Minus sign signifies undetectable by immunoblotting. Plus sign indicates apparent relative abundance (+ < ++ < +++). Each cell represents data obtained by immunoprecipitation (IP) and immunoblotting (IB) of multiple membrane preparations from fresh granulosa cells (n = 2-5). Typically, membranes were prepared using granulosa cells harvested from the small and medium follicles on 150 to 300 pig ovaries.

View larger version (42K): [in this window] [in a new window]   Fig. 9.   Heteromultimeric coassembly of Kv channel subunits in porcine granulosa cells. Immunoprecipitation (IP) from the membrane protein fraction was performed then the resulting IP and supernatant (SP) were resolved by SDS-PAGE (4 to 20% gradient gels) and analyzed by immunoblotting (IB). A, IP of 35S-labeled proteins from Chinese hamster ovary cells stably transfected with Kv1.5 (Kv1.5) and porcine GC. B, proteins from GC and Chinese hamster ovary cells stably transfected with Kv1.5 (CHO-Kv1.5) were IP with antibody to Kv1.1. The IP and SP were then IB with anti-Kv1.5, in the absence and presence (bottom) of Kv1.5 fusion protein (FP). C, membrane proteins from porcine granulosa cells were IP with antibody to Kv1.5. The IP and SP were probed with antibody to Kv1.1. D, proteins IP with antibodies (rabbit polyclonal) to Kv1.1-Kv1.6 were IB with antibody (mouse monoclonal) to Kv1.4. Arrows indicate molecular mass.

View larger version (69K): [in this window] [in a new window]   Fig. 10.   Expression and phosphorylation of Kv1.3 in fresh and cultured porcine granulosa cells. Western blot of membrane proteins isolated from freshly harvested PGC cells (F) or PGC maintained in culture for 72 h (C). Membrane proteins were separated by SDS-PAGE on a 4 to 20% gradient gel, transferred to nitrocellulose, and detected with anti-Kv1.3 (top). The blot was then stripped and reprobed with HRP-linked anti-phosphotyrosine (bottom).

View larger version (32K): [in this window] [in a new window]   Fig. 11.   Expression of Kv -subunits in porcine GC. Top, left, membrane proteins from porcine GC were immunoprecipitated with antibody to Kv1.3, Kv1.1, Kv1.2, or Kv1.5 and resolved by SDS-PAGE (4 to 20% gradient gels). The Western blot was probed with antibody to Kv1.3. The wide band at 50 kDa in the lane sequentially immunoprecipitated then immunoblotted with rabbit anti-Kv1.3 is IgG detected by the anti-rabbit secondary antibody. This band is lacking in the other lanes, because Kv -subunits were immunoprecipitated with mouse monoclonal antibodies. Left arrows indicate molecular mass standards. Right arrow (~40 kDa) indicates Kv1.3 protein. Top, right, mRNA for Kv2, but not Kv1 nor Kv3, is detected in porcine GC. RT-PCR was performed using gene-specific primers and poly(A)+ mRNA. Amplified products were separated by gel electrophoresis and visualized by ethidium bromide staining. Lanes: 1) 50-bp ladder; 2) primers for 1, 3) primers for 2, and 4) primers for 3. Right arrow (~140 bp) indicates Kv2 product. Bottom, partial deduced amino acid sequence of pig granulosa Kv2 (GenBank accession number AF348084), compared with previous clones. Dashes indicate homology to Pig GC Kv2.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Kv channels not only influence the electrical properties of excitable and nonexcitable cells but also regulate cell proliferation and differentiation (Attali et al., 1997; Sobko et al., 1998; Kotecha and Schlichter, 1999; Rane, 1999). K⁺-selective channel pores are formed by tetrameric complexes of integral membrane proteins with six transmembrane-spanning domains. Formation of homo- and heterotetramers of Kv -subunits has been demonstrated not only in heterologous expression systems but also native cells (Scott et al., 1994; Attali et al., 1997; Koch et al., 1997; Shamotienko et al., 1997; Sobko et al., 1998; Yuan et al., 1998; Schmidt et al., 1999). The subunit composition of Kv channels influences the expression, and electrophysiological and pharmacological properties of the associated currents (Hopkins, 1998). Kv channel expression and function may be modulated not only by coassociation of the various pore-forming -subunits with accessory ()-subunits and other regulatory proteins but also by post-translational modifications of either the Kv - or Kv -subunits (Martens et al., 1999; Petersen and Nerbonne, 1999). Several K⁺ channel subunits are typically expressed in a single cell. The extensive diversity in Kv channels has made it difficult to determine definitively the molecular identify of native K⁺ currents in many cases (Attali et al., 1997; Shamotienko et al., 1997; Sobko et al., 1998; Yuan et al., 1998; Schmidt et al., 1999).

A variety of K⁺ currents with different activation and inactivation kinetics have been described in porcine GC subjected to whole-cell patch clamp. In GC cultured for 2 days in serum-free media containing FSH (10 ng/ml) then an additional 3 to 5 days in FSH-free media, Kusaka et al. (1993) described a rapidly inactivating transient outward (A-type) current activated from a 70-mV holding potential, and a rapidly activating noninactivating delayed rectifier K⁺ current activated from a holding potential of 30 mV. Both currents were sensitive to 4-AP but only the delayed rectifier was sensitive to TEA (10 mM). In freshly isolated GC and GC maintained cultured for up to 3 days in serum-containing media (10% fetal bovine serum), Mattioli et al. (1991, 1993) described two currents: a slowly activating, noninactivating, TEA- and 4-AP-insensitive K⁺ current that disappeared after 24 h in monolayer culture, and a rapidly activating K⁺ current with an inactivation time constant that increased from 10 to 300 ms over 72 h in culture. The latter could be activated from a holding potential of 70 but not 40 mV and was sensitive to 4-AP. We recorded routinely a slowly activating, noninactivating delayed rectifier K⁺ current and an ultrarapidly activating, slowly inactivating K⁺ current. A rapidly inactivating A-type current was elicited from a 80-mV holding potential in less than 5% of screened GC.

The slowly activating, noninactivating current in Fig. 1 is identical to that described previously (Mattioli et al., 1991, 1993). Its electrophysiological properties and drug sensitivity are consistent with those of I_Ks currents and channels formed by coassociation of KCNQ1 and KCNE1 proteins (Barhanin et al., 1996; Busch et al., 1996; Sanguinetti et al., 1996; Selnick et al., 1997). In fact, both of these channel proteins are expressed in GC. The disappearance of I_Ks as GC differentiate in culture, along with the observation that the current is inhibited by luteinizing hormone, cAMP, and protein kinase C, has led to the speculation that I_Ks may play a specific role in granulosa cell maturation (Mattioli et al., 1991, 1993; L. C. Freeman, unpublished observations). Our identification not only of KCNQ1 and KCNE1 as the molecular correlates of GC I_Ks but also of 293B, L-735,821, and L-768,673 as specific antagonists of the current, will facilitate further investigation of its functional role. Additional experiments will be required to determine the basis for tissue-specific differences in I_Ks modulation, particularly to address the discrepant effects of cAMP on GC I_KS compared with recombinant KCNQ1/KCNE1 currents (Blumenthal and Kaczmarek, 1992) and native cardiac currents in guinea pig (Walsh et al., 1989) and pig (L. C. Freeman, unpublished observations).

The ultrarapidly activating, slowly inactivating K⁺ current recorded from GC by using whole-cell patch clamp seems to reflect overlapping components carried by homomeric and heteromeric Kv channels. The variability in the electrophysiological and pharmacological properties of granulosa cell delayed rectifier currents described here and elsewhere (Kusaka et al., 1993; Mattioli et al., 1993) is not surprising given the results shown in Figs. 7 to 10. GC can express at least six Kv -subunits, and their expression varies with time in culture. The diversity of granulosa cell K⁺ currents is further increased by the potential for not only coassembly of nonidentical Kv -subunits but also modulation by accessory subunits. Our data indicate that the potential for K⁺ channel diversity in GC is comparable with that seen in the central nervous system. Further experiments are required to determine whether GC also expresses non-Shaker-related Kv channel proteins (Kv2.x, Kv3.x, Kv4.x, etc.).

Heterotetrameric complexes of -subunits have been demonstrated previously among the following: Kv1.1/Kv1.2, Kv1.2/Kv1.4, Kv1.2/Kv1.3, Kv1.4/Kv1.6, Kv1.5/Kv1.2, Kv1.5/Kv1.4, Kv1.1/Kv1.2/Kv1.4, Kv1.1/Kv1.2/Kv1.6, Kv1.1/Kv1.2/Kv1.3, Kv1.2/Kv1.3/Kv1.4, and Kv1.2/Kv1.4/Kv1.6 (Koch et al., 1997; Shamotienko et al., 1997; Koschak et al., 1998; Sobko et al., 1998). In addition, the potential for heteromultimer formation has been demonstrated in oligodendrocyte progenitors for Kv1.4, Kv1.5, and Kv1.6 (Attali et al., 1997; Schmidt et al., 1999). In GC that express Kv1.1 to Kv1.6 proteins, we were able to demonstrate many of these Kv -subunit coassociations (Table 2).

The strong coassociation observed between the glycosylated forms of Kv1.4 and Kv1.6 is noteworthy and may explain our failure to detect rapidly inactivating A-type currents in a significant number of GC. Heteromeric expression of Kv1.4 and Kv1.6 has been shown to result in a slowly inactivating current (Roeper et al., 1998), because the N terminus of Kv1.6 possesses an N-type inactivation prevention domain, which prevents the fast, N-type inactivation typically associated with homomeric Kv1.4 channels. A potentially significant pattern of coassociation was also evident between the 60-kDa form of Kv1.1 and the low-molecular-mass forms of Kv1.2, Kv1.3, Kv1.4, and Kv1.5. These relationships may reflect a physiological role for Kv1.1 in limiting surface expression of the other Kv subunits. Kv1.1 has been reported to have a dominant negative effect on the surface expression of Kv1.2 and Kv1.4 (Manganas and Trimmer, 2000).

Kv -subunits can also affect dramatically both the gating and cell surface expression of coassociated Kv -subunits (Martens et al., 1999). Our data demonstrate that GC express at least two -subunits encoded by distinct genes. Kv1 and Kv2 family members not only interact with -subunits via distinct functional stoichiometries (_4n for Kv1 and ₄₄ for Kv2) but also affect Kv currents in distinct manners (Xu et al., 1998; Martens et al., 1999). For example, Kv1.3 converts Kv1.5 from a delayed rectifier channel to one with rapid, but partial inactivation, whereas Kv2.1 shifts the state-state activation and inactivation of Kv1.5 without inducing rapid inactivation. Kv2 subunits have also been shown to enhance N-glycosylation and/or surface expression of associated Kv -subunits, including Kv1.1, Kv1.2, Kv1.3, and Kv1.6. Clearly, the presence of Kv1 and Kv2 accessory subunits in GC could contribute substantially to Kv current diversity, not only by influencing the electrophysiological properties of expressed delayed rectifier currents but also by controlling the number of -subunits available for tetramer assembly.

Phosphorylation has been shown also to modulate current amplitudes and kinetics by influencing interactions between not only - and -subunits but also Kv channel complexes and other intracellular proteins (Bowlby et al., 1997; Martens et al., 1999; Rane, 1999). Basal tyrosine phosphorylation of Kv1.3 suggests that the activity of GC Kv channels may be influenced by signaling pathways associated with intraovarian growth factors (Steele and Leung, 1993). Tyrosine phosphorylation has been identified previously as an important influence on the activity of Kv1.3 (Bowlby et al., 1997; Rane, 1999), an ion channel with a well established role in modulating proliferation and differentiation of other nonexcitable cells (Kotecha and Schlichter, 1999; Rane, 1999).

As GC spontaneously luteinized in culture, expression of Kv1.3 increased. The mature, fully glycosylated form of Kv1.3 was present only in cultured GC. Whole-cell K⁺ currents recorded from cultured GC contained a TEA-sensitive component that exhibited cumulative inactivation, consistent with expression of Kv1.3 homotetramers. It will be interesting to determine what role, if any, Kv1.3 plays in granulosa cell maturation. Increased expression of Kv1.3 has been positively correlated with days in culture and acquisition of a proliferative phenotype in microglia (Kotecha and Schlichter, 1999).

Associating specific K⁺ channels with cellular processes in GC will be technically challenging as a result of the molecular diversity. However, it will be critical to define the roles of specific K⁺ channels in granulosa cell proliferation, differentiation, and apoptosis, and the temporal pattern of K⁺ channel expression during the estrous cycle, because these proteins may represent either novel targets for assisted reproduction or potential sites of toxicity for drugs designed to act on channels in other tissues, including heart, brain, and lymphocytes. We document here distinct differences in the pharmacological sensitivities and temporal expression patterns of voltage-gated K⁺ currents and channel proteins in fresh isolates and primary cultures of porcine GC. Our findings provide an essential background for experimental definition of granulosa K⁺ channel function(s).