Farnesol Blocks the L-Type Ca²⁺ Channel by Targeting the α_1C Subunit

Cell Culture

Rat aortic A7r5 cells (passages 10 to 25) were purchased from the American Type Culture Collection (Manassas, Va). The cells were grown in Dulbecco’s modified Eagle’s medium under 95% O₂ and 5% CO₂ in the presence of 10% dialyzed BSA (Sigma-Aldrich), 1.7 mmol/L l-glutamine, streptomycin (30 mg/mL), penicillin (30 U/mL), and nonessential amino acids.⁹ The CHOC9 cells, which do not contain endogenous Ca²⁺ channel β, α2, and γ subunits, were grown as previously described.^{13 14}

Patch Clamp

Ca²⁺ channel currents were recorded in the whole-cell (WC) or perforated-patch (PP) configuration with nystatin^{15 16} and with the use of a List patch-clamp amplifier EPC7. Data acquisition and command potentials were controlled with commercial software and a CED1401 interface (Cambridge Electronic Design Ltd). Currents were recorded from holding potentials of −80 mV (−100 mV) during linear voltage ramps at 0.67 V/s from −100 mV to +100 mV or 300-ms step pulses to different potentials; pulse frequency was 0.2 Hz. Analysis of WC currents was performed using CED patch and voltage clamp software version 6.08 (Cambridge Electronic Design Ltd). Ba²⁺ was used as the charge carrier; K⁺ currents were blocked by Cs⁺. In WC experiments, the extracellular solution contained (in mmol/L) NaCl 125, BaCl₂ 10.8, CsCl 5.4, glucose 10, and Na-HEPES 10 (pH 7.4 at 37°C). The patch pipette was filled with a solution containing (in mmol/L) CsCl 120, MgCl₂ 1, Mg-ATP 3, EGTA 10, and Cs-HEPES 10 (pH 7.4 at 22°C). For PP experiments, the bath solution contained (in mmol/L) NaCl 125, BaCl₂ 10.8, CsCl 5.4, glucose 10, and Na-HEPES 10 (pH 7.4 at 22°C); the pipette was filled with a solution containing (in mmol/L) CsCl 75, MgCl₂ 5, HEPES 5, and aspartate 75 (pH 7.4 at 22°C). The resistance of the pipettes was 4 to 6 MΩ. Experiments were performed at room temperature, ≈22°C.

To test a possible involvement of G proteins in the effects of farnesol, the GDP analogue GDPβS was intracellularly infused via the patch pipette at a concentration of 100 mmol/L by using the WC patch clamp configuration. In control experiments, 100 μmol/L GDPβS (5-minute infusion) prevented G protein–activated cation currents induced by vasopressin (1 μmol/L) in A7r5 cells, which was consistent with the observations of Krautwurst et al.¹⁷ Furthermore, our previous studies demonstrated that 100 μmol/L GDPβS abolished G protein–dependent inhibition of voltage-gated, L-type Ca²⁺ channels by several receptor agonists.^{18 19 20 21}

Reagents

Bay K 8644 and nimodipine were obtained from RBI. EGTA, HEPES, and all other reagents were purchased from Sigma-Aldrich. Farnesol was obtained from Aldrich. All salts were obtained from Merck.

Animal Model

We relied on the Sabra desoxycorticosterone acetate (DOCA) –salt genetic model. The rats used in the current study were subbred from the original colony and are referred to as SBH/yr or SBN/yr.²² Experiments were performed in male, salt-resistant SBN/yr rats weighing 286±7 g and male, salt-sensitive SBH/yr rats weighing 309±7 g. The animals were bred in the animal facility of the Barzilai Medical Center, Ashkelon, Israel. The rats were allowed free access to standard chow (0.3% NaCl, SSNIFF Specialitäten GmbH) and drinking water. The experimental protocol was approved by the local council on animal care, which corresponds to the standards of the American Physiological Society. A 25-mg DOCA tablet (Innovative Research of America) was implanted into SBN/yr and SBH/yr rats below the skin at the nape of the neck. The animals were provided with 1% NaCl solution to drink. After 3 weeks (20±1 days) of DOCA salt treatment, the rats were surgically prepared and outfitted with a radiotelemetry device (DSI) to continuously record mean arterial blood pressure and heart rate. To test the oral effects of farnesol on blood pressure, farnesol (500 mg/kg) was mixed with olive oil and instilled by gavage feeding at a rate of 2 mL/min g body weight. Control rats received only vehicle. Blood pressure and heart rate were recorded for 48 hours thereafter.

Statistics

All values are given as mean±SEM. The term n represents the number of cells tested. For group comparisons, paired and unpaired Student’ s t tests or nonparametric tests were used as appropriate. A value of P<0.05 was considered statistically significant.

Inhibition of Multisubunit, Complex L-Type Ca²⁺ Channels by Farnesol

Calcium currents in A7r5 cells were characterized by using 10.8 mmol/L Ba²⁺ as a divalent charge carrier with the use of the WC or PP configuration of the patch clamp technique. The inward Ba²⁺ currents were reversibly blocked by 1 mmol/L nimodipine or 100 mmol/L Cd²⁺; they were reversibly increased by 1 mmol/L (±)Bay K 8644 (not shown). A fast inactivating (T-type) current was observed in only 10% of all cells tested (N=54). To estimate the current-voltage relationship (IV curves), the test potential was linearly varied from −100 to +100 mV (ramp), or 300-ms step pulses to test potentials between −70 and +100 mV were applied from a holding potential of −80 mV (Figure 1A⇓ through 1C). The current recorded during voltage-ramp pulses (Figure 1C⇓) was U-shaped and exhibited a maximum at 2±8 mV (N=54). The currents corresponded to those measured during step pulses to the respective potentials (Figure 1B⇓). The apparent threshold occurred at −33±7 mV and the reversal potential at 55±9 mV (N=54).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Effects of farnesol on inward Ba²⁺ currents (I_Ba) in A7r5 cells.A, Superimposed current records at different test potentials of a representative cell before (open circles) and after (closed circles) application of 5 mmol/L farnesol. Ba²⁺ currents were evoked by 300-ms depolarizing voltage-step pulses to test potentials of −15, −10, −5, 0, 5, 10, 15, and 20 mV (holding potential, −80 mV; pulse frequency, 0.33 Hz). The dotted lines represent the zero-current levels. B, IV relation of peak Ba²⁺ currents recorded during step pulses (A) before (open circles) and after (closed circles) application of 5 mmol/L farnesol (n=5). Solid lines were drawn according to Equation 1 for the data points (see text). C, Superimposed current records of a representative cell before (control) and after application of 5 mmol/L farnesol. Ba²⁺ currents were evoked by linear voltage ramps at 0.67 V/s from −100 to +100 mV (pulse frequency, 0.2 Hz). Currents either are expressed in pA (I_Ba, top) or have been normalized (I_norm, bottom).

Figure 1A⇑ shows the effect of farnesol (5 mmol/L) on Ca²⁺ channel currents evoked by voltage steps to test potentials between −15 and 20 mV from a holding potential of −80 mV in a representative A7r5 cell. Farnesol significantly reduced the Ba²⁺ current at all potentials, and the currents at relatively positive potentials (≥0 mV) showed a faster decay in the presence of farnesol. Figure 1B⇑ shows the corresponding IV curves determined from peak Ba²⁺ channel currents recorded before and after farnesol application. The IV curves were fitted using the following equation and assuming a Boltzmann type of channel activation: where I_Ba (pA) is the peak Ba²⁺ current, V_0.5 (mV) is the potential of half-maximal activation, V (mV) is the test potential, g_max (pS) is the maximal conductance, V_rev (mV) is the extrapolated reversal potential, and h is the slope factor of Ca²⁺ channel activation. Statistical analysis of the fits revealed that only g_max was significantly decreased by farnesol, 8.08±1.1 and 2.6±0.48 pS for control and farnesol-treated cells, respectively (P<0.05, n_PP=5). The other variables were not significantly altered (V_0.5, −13.8±2.2 and −9.3±2.3 mV; h, 4.2±0.5 and 5.1±0.41 mV; and V_rev, 38.2±1.7 and 35.3±1.9 mV for control and farnesol-treated cells, respectively; n=5).

To further substantiate this point, we recorded Ca²⁺ channel currents by using linear voltage-ramp pulses at 0.67 V/s from −100 to +100 mV; the holding potential was −80 mV. As shown for the normalized data in Figure 1C⇑, farnesol reduced the current without affecting the potential at which peak Ba²⁺ currents occurred. The current was proportionally inhibited at all potentials where Ba²⁺ currents could be evoked; the apparent threshold and the reversal potential were not affected by farnesol. Thus, the data demonstrate that farnesol induces a voltage-independent inhibition of multisubunit, complex Ca²⁺ channels in VSMCs.

The effect of farnesol on steady-state inactivation was also tested. As shown in Figure 2⇓, farnesol affected the voltage dependence of Ca²⁺ channel inactivation in A7r5 cells. The Ca²⁺ channel availability was determined by measuring Ba²⁺ currents during test pulses to 0 mV after 3-second conditioning voltage steps. Representative pulse recordings in control and farnesol-treated cells are shown in Figure 2A⇓. The data were fitted with a Boltzmann equation of the form: where I (pA) is the amplitude of the inward Ba²⁺ current elicited by a test pulse after the conditioning prepulse V (mV), I_max (pA) is the amplitude of the inward Ca²⁺ current elicited by a test pulse from a conditioning prepulse of −80 mV, V_h (mV) is the value of the conditioning potential causing 50% inactivation of the Ca²⁺ current, and k (mV) is the steepness factor characterizing the voltage sensitivity of the channels. Farnesol (5 mmol/L) significantly shifted the midpoint (V_h) of the normalized steady-state inactivation curve by ≈10 mV to more negative membrane potentials in comparison with control conditions; ie, from −29.9±2.0 to −41.4±1.7 mV (P<0.001, n_PP=5; Figure 2B⇓). The slope factor k was not significantly altered; k values were 10.2±0.2 and 9.7±0.63 mV for control and farnesol-treated cells, respectively.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Effects of farnesol on steady-state inactivation of Ca²⁺ channels in A7r5 cells. A, Superimposed records of Ba²⁺ currents of a representative cell in the absence (left) and presence (right) of 5 mmol/L farnesol. The currents were recorded during a 300-ms test pulse to 0 mV after a 3-second conditioning prepulse to voltages between −80 and 0 mV. Holding potential was −80 mV. B, Corresponding voltage dependence of steady-state inactivation of Ca²⁺ channels averaged from 5 cells in the absence (open circles) and presence (closed circles) of 5 mmol/L farnesol. Normalized peak Ca²⁺ channel currents (I/I_max) were plotted vs the voltage of the conditioning prepulses. Pulse protocol was the same as that used in A. Solid lines were drawn according to Equation 2 for the data points.

G Protein–Independent Ca²⁺ Channel Inhibition

In these experiments, we tested the hypothesis that the effect of farnesol was secondary to the activation of a putative G protein. In a first set of experiments, we applied the GDP analogue GDPβS intracellularly, which stabilizes G proteins in the inactive GDP-bound form^{23 24} and thus, prevents receptor-activated, G protein–dependent regulation of ion channels.^{18 19 20 21} GDPβS was intracellularly infused via the patch pipette at a concentration of 100 mmol/L by using the WC patch-clamp configuration. Estimation of the rate of GDPβS diffusion²⁵ and patch-clamp experiments¹⁸ showed that the cytoplasmic concentration of GDPβS should have reached 80% of the concentration in the pipette solution within 5 minutes after membrane disruption. With the use of this protocol, the farnesol block was compared with that obtained in the absence of GDPβS in the recording pipette. As illustrated in Figure 3A⇓ and 3B⇓, GDPβS did not modify farnesol-induced inhibition of Ca²⁺ channels. After 5 minutes of intracellular infusion of GDPβS, farnesol (10 mmol/L) inhibited the peak Ba²⁺ current by 66.1±3.7% (n=6). This result was not significantly different from the value obtained in control experiments with standard pipette solution (77.0±11.2%; P=NS, n_WC=6).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

G protein–independent inhibition of L-type Ca²⁺ channels by farnesol in A7r5 cells. A and B, Effect of GDPβS on the inhibition of Ba²⁺ currents by farnesol. Ba²⁺ currents (I_Ba) were evoked by linear voltage ramps at 0.67 V/s from −100 to +100 mV (pulse frequency, 0.2 Hz). A, The pipette solution was filled with standard pipette solution; B, the pipette solution contained 100 mmol/L GDPβS. Currents were recorded after 5 minutes of intracellular infusion, under control conditions, with 10 mmol/L farnesol and after washout (w.o.). C, Superimposed Ba²⁺ current traces evoked in the presence or absence of 10 mmol/L farnesol in a representative cell by using the double-pulse voltage protocol shown above the traces. In this cell, application of farnesol decreased the current to 22% and 17% of control values before and after the conditioning prepulse to +80 mV, respectively.

Voltage-gated Ca²⁺ channels can be directly inhibited by receptor-activated G proteins via a membrane-delimited pathway independent of soluble messengers. This inhibition results from the interaction of G protein β/γ subunits with the Ca²⁺ channel protein and is typically relieved by strong membrane depolarizations.^{25 26 27} We compared inhibitions of Ca²⁺ currents by farnesol before and after strong membrane depolarization. As shown in Figure 3C⇑, a large conditioning depolarization (+80 mV) inserted between 2 identical test pulses to 0 mV (“double-pulse” protocol) did not affect the amplitude of the control Ba²⁺ currents. Similarly, it did not affect the amplitude of the Ba²⁺ currents in the presence of farnesol (10 mmol/L): inhibition was 71±3% and 75±4% before and after the conditioning prepulse, respectively (n_PP=5, P=NS). These results suggest that farnesol blocks L-type channels by a G protein–independent mechanism.

Effect of Farnesol on α_1C Subunit Currents

As previously described,¹³ the CHOC9 cells exhibited a typical, slowly inactivating Ca²⁺ current (L-type) and no fast inactivating (T-type) current (n=49). The recorded IV curves (Figure 4A⇓) were U-shaped, with a maximum current at a potential of 4±5 mV, an apparent threshold potential at −35±6 mV, and a reversal potential at 52±6 mV (n=49). Ba²⁺ currents were reversibly blocked by 1 mmol/L nimodipine or 100 mmol/L Cd²⁺ and were reversibly increased by 1 mmol/L (±)-Bay K 8644 (not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Effects of farnesol on Ca²⁺ currents in CHOC9 cells. A, Original current records of a representative cell under control conditions, with 5 mmol/L farnesol, and after washout (w.o.). Ba²⁺ currents were evoked by linear voltage ramps at 0.67 V/s from −100 to +100 mV with a 0.2-Hz pulse frequency. The insert shows normalized Ba²⁺ currents (I_norm) in the presence of farnesol and under control conditions. B, Dose-response curves of Ba²⁺ currents and farnesol concentration in CHOC9 (open circles) and A7r5 (closed circles) cells. Inward Ba²⁺ currents were recorded as peak currents at test potentials of 0 mV in A7r5 cells (n=5) and CHOC9 cells (n= 4). I_f represents the percent current in the presence of farnesol as a fraction of its control value at 0 mV. Solid lines are drawn according to Equation 3 for the data points (see text). Symbols represent mean±SEM. Holding potential, −80 mV. C, Corresponding voltage dependence of steady-state inactivation of the channels averaged from 5 cells in the absence (open circles) and presence (closed circles) of 5 mmol/L farnesol. Normalized peak Ba²⁺ currents (I/I_max) were plotted vs voltage of the conditioning prepulses. Pulse protocols are as described in the legend to Figure 2A⇑. Solid lines were drawn according to Equation 2 for the data points.

The same figure illustrates the effect of 5 mmol/L farnesol on CHOC9 cell Ca²⁺ channel currents evoked by linear voltage-ramp pulses at 0.67 V/s from −100 to +100 mV. The holding potential was −80 mV. As shown for the normalized data (Figure 4A⇑, insert), farnesol reduced the current without affecting the potential at which peak currents occurred. Ba²⁺ currents were proportionally inhibited at all potentials where they could be evoked; however, neither the threshold potential nor the reversal potential was affected by farnesol. The effect of farnesol was reversible on washout. The onset and offset of the inhibitory effect of farnesol were similar to those seen in A7r5 cells, ie, ≈1 minute. Thus, the inhibition of α₁ subunits by farnesol occurred in a similar voltage-independent manner as the inhibition of complex, multisubunit, native L-type channels.

Figure 4B⇑ shows the average dose-response curve for inhibition of Ca²⁺ channel currents by farnesol in CHOC9 cells. The curve was fitted to the following equation: where I_f is the percent Ba²⁺ current in the presence of farnesol as a fraction of its control value at 0 mV, [farnesol] is the concentration of farnesol (mmol/L), IC₅₀ is the concentration of farnesol yielding half-maximal inhibition, and n is the Hill coefficient. The results indicate that IC₅₀ was 4.3±1.2 mmol/L and n was 1.18±0.18 (n_PP=4). The dose-response curve to farnesol in A7r5 cells is given for comparison on the same figure. Fitting the data points with Equation 3 revealed that IC₅₀ was 2.6±1.1 mmol/L (n_PP=5) and n was 0.99±0.13 in A7r5 cells (P=NS versus CHOC9 cells).

Farnesol also affected the voltage dependence of the steady-state inactivation of Ba²⁺ currents through α₁ subunits in CHOC9 cells (Figure 4C⇑). The kinetic protocol used with CHOC9 cells was the same as that used with A7r5 cells; the data were fit with Equation 2. Farnesol (5 mmol/L) significantly shifted the midpoint (V_h) of the normalized steady-state inactivation curve by ≈7 mV to more negative membrane potentials: V_h was −30.5±1.4 and −37.2±2.5 mV for control and farnesol, respectively (P<0.05, n_PP=5). Although there was a trend for the inactivation shift in CHOC9 cells to be less than that in A7r5 cells, the difference was not statistically significant. The slope factor k was not significantly altered by farnesol: 12.6±0.6 and 13.1±1.1 mV in control and farnesol-treated cells, respectively.

Initial Block of Multisubunit, Complex Channels and α₁ Subunits

These experiments were designed to determine whether the farnesol-induced inhibition of Ca²⁺ channel currents depended on the opening of L-type channels; ie, whether farnesol could induce an “initial block.” A7r5 cells were patched and held at −80 mV. Control Ba²⁺ currents were recorded during 300-ms pulses to 0 mV (frequency of 0.2 Hz). After ≈10 pulses, the cells were left unstimulated at resting potential (−80 mV). Farnesol (10 mmol/L) was then applied extracellularly; after 2 minutes, pulsing to 0 mV (frequency of 0.2 Hz) was resumed. The first depolarization evoked a current whose amplitude was reduced to 37±8% (n_PP=5) of control values before farnesol application. Thus, the initial block of the current by farnesol was 63%. Subsequent pulsing did not further reduce Ba²⁺ currents. This effect of farnesol on initial block is illustrated in Figure 5A⇓ for a representative A7r5 cell. These experiments were repeated in CHOC9 cells, and similar results were obtained (Figure 5B⇓). For CHOC9 cells, the initial block with 10 mmol/L farnesol was 50±9% (n_PP=7). As observed in A7r5 cells, subsequent pulsing did not yield a significant further decrease in current magnitude (Figure 5B⇓).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Farnesol-induced initial block of L-type Ca²⁺ channels in A7r5 and CHOC9 cells. A, Shown are Ba²⁺ current traces of a representative A7r5 cell. The initial-block protocol is represented below the trace: currents were elicited by applying 6 test pulses to 0 mV (300-ms duration; 0.2-Hz frequency) from a holding potential of −80 mV. Farnesol (10 mmol/L) was then applied to the unstimulated cell, ie, clamped at a holding potential of −80 mV. After 2 minutes, pulsing was resumed. In this cell, the first postfarnesol depolarization evoked a Ca²⁺ current whose amplitude was reduced to ≈25% of the control current before farnesol (initial block of ≈75%). Further pulsing (pulse frequency, 0.2 Hz) did not significantly decrease the initial block. B, Shown are Ba²⁺ currents evoked by the same initial-block protocol as shown in A for a representative CHOC9 cell. In this cell, farnesol induced an initial block of ≈45%. Subsequent pulsing did not significantly decrease Ba²⁺ currents.

Effect of Farnesol on Blood Pressure in Hypertensive Rats

Farnesol administration (500 mg/kg by gavage to SBH/yr rats) reduced blood pressure at 12 hours (Figure 6A⇓) to levels lower than in vehicle-treated rats. This difference in blood pressure persisted and was still detectable at 48 hours. The cumulative 48-hour decrease in blood pressure (Figure 6B⇓) was 12 mm Hg, whereas no decrease was observed in vehicle-treated SBH/yr rats (P<0.05). We also tested the effect of farnesol 500 mg/kg on salt-resistant SBN/yr rats. Farnesol had no effect on the blood pressure of the normotensive strain. Farnesol did not influence the heart rate of either strain (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

Figure 6.

A, Farnesol given by gavage reduced arterial blood pressure to lower levels in SBH/yr rats than in vehicle-treated SBH/yr rats by 12 hours. This decrease persisted for 48 hours. B, The cumulative difference in blood pressure in farnesol-treated, SBH/yr rats compared with vehicle-treated SBH/yr rats was 16 mm Hg (*P<0.05).