Hypercholesterolemia Induces Up-regulation of KACh Cardiac Currents via a Mechanism Independent of Phosphatidylinositol 4,5-Bisphosphate and Gβγ*

Background: KACh channels play a key role in controlling the heart rate. Results: KACh currents are enhanced by cholesterol enrichment and high cholesterol diet. Conclusion: Cholesterol plays a critical role in modulating IK,ACh in atrial cardiomyocytes. Significance: The increase in IK,ACh following cholesterol enrichment is likely to play a critical role in hypercholesterolemia-induced dysfunction of the heart. Hypercholesterolemia is a well-known risk factor for cardiovascular disease. In the heart, activation of KACh mediates the vagal (parasympathetic) negative chronotropic effect on heart rate. Yet, the effect of cholesterol on KACh is unknown. Here we show that cholesterol plays a critical role in modulating KACh currents (IK,ACh) in atrial cardiomyocytes. Specifically, cholesterol enrichment of rabbit atrial cardiomyocytes led to enhanced channel activity while cholesterol depletion suppressed IK,ACh. Moreover, a high-cholesterol diet resulted in up to 3-fold increase in IK,ACh in rodents. In accordance, elevated currents were observed in Xenopus oocytes expressing the Kir3.1/Kir3.4 heteromer that underlies IK,ACh. Furthermore, our data suggest that cholesterol affects IK,ACh via a mechanism which is independent of both PI(4,5)P2 and Gβγ. Interestingly, the effect of cholesterol on IK,ACh is opposite to its effect on IK1 in atrial myocytes. The latter are suppressed by cholesterol enrichment and by high-cholesterol diet, and facilitated following cholesterol depletion. These findings establish that cholesterol plays a critical role in modulating IK,ACh in atrial cardiomyocytes via a mechanism independent of the channel's major modulators.

Hypercholesterolemia is a well-known risk factor for cardiovascular disease. In the heart, activation of K ACh mediates the vagal (parasympathetic) negative chronotropic effect on heart rate. Yet, the effect of cholesterol on K ACh is unknown. Here we show that cholesterol plays a critical role in modulating K ACh currents (I K,ACh ) in atrial cardiomyocytes. Specifically, cholesterol enrichment of rabbit atrial cardiomyocytes led to enhanced channel activity while cholesterol depletion suppressed I K,ACh . Moreover, a high-cholesterol diet resulted in up to 3-fold increase in I K,ACh in rodents. In accordance, elevated currents were observed in Xenopus oocytes expressing the Kir3.1/Kir3.4 heteromer that underlies I K,ACh . Furthermore, our data suggest that cholesterol affects I K,ACh via a mechanism which is independent of both PI(4,5)P 2 and G␤␥. Interestingly, the effect of cholesterol on I K,ACh is opposite to its effect on I K1 in atrial myocytes. The latter are suppressed by cholesterol enrichment and by high-cholesterol diet, and facilitated following cholesterol depletion. These findings establish that cholesterol plays a critical role in modulating I K,ACh in atrial cardiomyocytes via a mechanism independent of the channel's major modulators.
We recently reported cholesterol sensitivity of representative homomeric Kir channels using the Xenopus oocyte heterologous expression system (18). Kir3.4*, the highly active Kir3.4 pore mutant S143T that has similar characteristics to those of the Kir3.1/Kir3.4 complex (19), was stimulated by cholesterol. Cholesterol is one of the major lipid components of the plasma membrane in mammalian cells. While cholesterol is essential for cell function and growth, its excess is associated with multiple pathological conditions (20 -22) including development of cardiovascular disease (23)(24)(25). Thus, the present study was designed to examine the impact of cholesterol on cardiac K ACh currents (I K,ACh ).

Expression of Recombinant Channels in Xenopus Oocytes-
Point mutations on the background of control Kir3.4_S143T (Kir3.4*) (19) were generated using the QuickChange site-di-rected mutagenesis kit (Stratagene). cRNAs were transcribed in vitro using the MessageMachine kit (Ambion). Oocytes were isolated and microinjected as previously described (26). Expression of channel proteins in Xenopus oocytes was accomplished by injection of the desired amount of cRNA, 2 ng per oocyte for Kir3.4* or Kir3.4*D223N, and for coexpression of Kir3.1 with Kir3.4, 1 ng per oocyte per channel. All oocytes were maintained at 17°C. Two-electrode voltage clamp recordings were performed 2 days after injection.
Two-electrode Voltage-clamp Recording and Analysis (Xenopus Oocytes)-Whole-cell currents were measured by conventional two-microelectrode voltage clamp with a GeneClamp 500 amplifier (Axon Instruments), as previously reported (26). A high-potassium (HK) solution was used to superfuse oocytes (in mM: 96 KCl, 1 NaCl, 1 MgCl 2 , 5 HEPES KOH buffer, pH 7.4). Basal currents represent the difference of inward currents obtained (at Ϫ80 mV) in the presence and absence of 3 mM BaCl 2 in high-potassium solution. Each experiment was performed on a minimum of 3-5 oocytes from the same batch, and a minimum of two batches of oocytes were tested for each normalized recording shown. Recordings from different batches of oocytes were normalized by the mean of whole-cell basal currents from untreated oocytes. Statistics (mean Ϯ S.E.) for each construct were calculated from all of the normalized data from different batches of oocytes. Statistical analysis of the data was performed using a two-tailed Student's t test with p Ͻ 0.05 taken to be significant. The fitting of the curves for the time courses was performed using Origin software (Originlab).
Macropatch Recording (Xenopus Oocytes)-Currents were recorded from the inside-out patch configuration using an EPC9 amplifier (HEKA) and the Pulse fit program (HEKA). The bath and pipette solutions of ND96KϩEGTA were composed of 96 mM KCl, 1 mM MgCl 2 , 5 mM EGTA, and 10 mM HEPES, pH 7.4. Currents were recorded at a holding membrane potential of Ϫ80 mV.
Ci-VSP Rundown Experiments in Xenopus Oocytes and Their Analysis-Ci-VSP (28) has a voltage sensor coupled to a phosphoinositide phosphatase, leading to dephosphorylation of PI(4,5)P 2 . The phosphatase activity is activated by membrane depolarization (28). The advantage of this method of depleting PI(4,5)P 2 is that it allows to examine the effect of dephosphorylation of PI(4,5)P 2 on whole-cell currents using two-electrode voltage clamp rather than by inside-out recording from macropatches. As with other methods that remove PI(4,5)P 2 , activation of Ci-VSP results in current rundown, and rundown is slower and less complete with stronger interaction of the channel with PI(4,5)P 2 . Activation of Ci-VSP is achieved by cycling between Ϫ80 mV and ϩ80 mV. The rundown curves for each experiment (oocyte) were fitted to an exponential decay function of a second order: Inhibition ϭ Inhibition (tϭ∞) ϩ A 1 exp(Ϫt/ 1 ) ϩ A 2 exp(Ϫt/ 2 ). The limiting value, Inhibition (tϭ∞) , is the maximal inhibition that would be obtained at infinity. The average of the values obtained in all the experiments is given in the summary figure.
Cholesterol Enrichment and Depletion of Atrial Cardiomyocytes-As previously described (25), cardiomyocytes were depleted of cholesterol by exposure to the cholesterol acceptor methyl-␤-cyclodextrin (M␤CD). Alternatively, cells are enriched with cholesterol by treatment with M␤CD saturated with cholesterol, a well-known cholesterol donor. Depletion was carried out with freshly prepared 5 mM M␤CD in a modified Kraft-Brühe solution (pH 7.2) for 30 min or 1 h. For enrichment, an M␤CD-cholesterol solution was prepared as described previously (29). In brief, a small volume of cholesterol solvated in chloroform was added to a glass tube, the solvent was evaporated, and then 5 mM M␤CD in a modified Kraft-Brühe solution (pH 7.2) was added to the dried cholesterol. This preparation was sonicated and incubated overnight in a rotating bath at 37°C. The cells were incubated with the M␤CDcholesterol solution for 1 h. During enrichment and depletion, myocytes were maintained in a humidified CO 2 incubator at 37°C. After treatment, cells were washed three times with serum-free medium and returned to the incubator.
Rat High Cholesterol Diet-A group of 25-day-old male Sprague-Dawley rats was placed on a high-cholesterol diet (2% cholesterol in standard rodent food) (Harland-Teklad). Another group of the same age was fed an isocaloric, cholesterol-free diet from the same supplier. Rats were sacrificed as atrial tissue donors after 20 -24 weeks on control or high cholesterol diet.
Atrial Myocyte Isolation-Atrial myocytes were freshly isolated from adult New Zealand White rabbits (2.8 -3.1 kg) of either gender or from male rats by a Pronase-collagenase enzymatic dissociation method. Hearts were excised, immediately tied via the aorta to a Langendorff column, and retrogradely perfused for 5 min with Tyrode solutions that were oxygenated and maintained at 37°C, followed by 5 min with a Ca 2ϩ -free Tyrode solution. The heart was then perfused with enzyme solution for ϳ20 min. Atria were excised, minced, and placed in fresh enzyme solution. The tissues were bubbled with an O 2 /CO 2 (95%:5%) gas mixture and gently shaken for two 15-min cycles in a 37°C shaker bath. At the end of each cycle the supernatant was collected and replaced with fresh enzyme solution. Supernatants were filtered through 200 m nylon, and isolated cells were pelleted by gentle centrifugation. Isolated myocytes were stored in a modified Kraft-Brühe solution (pH 7.2) before using. Rod-shaped quiescent cells with clear striations and no membrane blebs or other morphological irregularities were studied within 10 h of isolation. Tyrode solution for myocyte isolation contained (mM): 130 NaCl, 5 KCl, 1.8 mM CaCl 2 , 0.4 KH 2 PO 4 , 3 MgCl 2 , 5 HEPES, 15 taurine, 5 creatine, 10 glucose, pH 7.25 (adjusted with NaOH). For Ca 2ϩ -free Tyrode solution, CaCl 2 was replaced with 0.1 mM Na 2 EGTA. For making enzyme solution, the Ca 2ϩ -free Tyrode solution was supplemented with 0.45 mg/ml collagenase (Cls 4; Worthington Biochemical) and 0.015 mg/ml Pronase (Type XIV; Sigma-Aldrich). The modified Kraft-Brühe myocyte storage solution contained (mM): 120 K-glutamate, 10 KCl, 10 KH 2 PO 4 , 0.5 K 2 EGTA, 10 taurine, 1.8 MgSO 4 , 10 HEPES, 20 glucose, 10 mannitol, pH 7.2 (adjust with KOH).
Whole Cell Patch Clamp and Electrophysiological Recordings (Atrial Myocytes)-Pipettes were pulled using a P-97 micropipette puller (Sutter Instrument) and then fire polished. The final pipette tip diameter was 2-3 m, and the corresponding pipette resistance in bath solution was 2-4 M⍀. Junction potentials were corrected, and a 3 M KCl-agar bridge served as the ground electrode. Freshly isolated atrial myocytes were dispersed over a glass-bottomed cell chamber (ϳ0.3 ml) and extracellular solution was superfused at a rate of 2-3 ml/min. Typical seal resistances were 5-10 G⍀. Myocytes were dialyzed for at least 5 min before data were collected. After obtaining the whole-cell configuration, successive 250 ms long steps were applied from Ϫ50 mV or Ϫ80 mV to test potentials between Ϫ100 and ϩ40 mV in ϩ10 mV increments, and current-voltage (I-V) relationships were plotted from quasi steady-state currents. Currents were recorded with an Axoclamp 200B and Digidata 1322A under pClamp 9 (MDS Analytical Technology) and digitized (5 kHz) after low-pass filtering (Bessel, 2 kHz). , and 100 nM tertiapin-Q (30) (Tocris) was used to block the channels. Tertiapin-sensitive currents were obtained by subtracting the currents recorded after perfusion of tertiapin from the currents of the control and cholesterol-treated ACh-induced currents, and from the basal currents, respectively. To examine the effect of GTP␥S, 100 M GTP␥S were added to the pipette solution, and the myocytes were dialyzed for at least 10 min before data were collected. After basal currents were obtained, the cells were superfused with ACh containing solution followed by a solution containing both ACh and tertiapin-Q. Currents recorded were normalized to cell size by dividing the amplitude of the current (pA) by the capacitance of the cell (pF) to obtain the current density. Statistics (mean Ϯ S.E.) for each construct were calculated from all of the normalized data from different batches of myocytes. Statistical analysis of the data were performed using a two-tailed Student's t test or a one-way ANOVA, and comparisons to the control group were made by the Holm-Sidak method with p Ͻ 0.05 taken to be significant.

ACh-induced Currents in Atrial Cardiomyocytes
Are Up-regulated by Cholesterol-To investigate the role of cholesterol in modulating cardiac K ACh currents (I K,ACh ), we first examined the effect of both cholesterol enrichment and depletion on I K,ACh in atrial cardiomyocytes. As Fig. 1, A and B, shows, cho- lesterol enrichment by exposing myocytes for 1 h to methyl-␤cyclodextrin (M␤CD) saturated with cholesterol, a well-known cholesterol donor (29), resulted in an ϳ2-fold increase in K ACh currents that were sensitive to the selective I K,ACh -blocker tertiapin (30). Conversely, cholesterol depletion following a 30-min exposure to untreated M␤CD, which acts as a cholesterol acceptor (29), decreased cardiomyocyte I K,ACh (Fig. 1, A  and C). Further exposure to M␤CD for one hour suppressed I K,ACh currents almost completely. Summary data in Fig. 1, D and E show that cholesterol enrichment and depletion modulated both inward and outward currents in a similar manner. Thus, while the effect was more visible for the larger inward currents, the physiologically relevant smaller outward currents were also significantly affected by cholesterol.
High Cholesterol Diet Enhances K ACh Currents in Atrial Myocytes-Following our finding that acute cholesterol enrichment enhances K ACh currents in atrial cardiomyocytes, we next evaluated the tertiapin-sensitive K ACh currents in atrial myocytes from rats subject to a hypercholesterolemic diet. As can be seen in Fig. 2, A-D, rats that were on high cholesterol diet for 20 -24 weeks exhibited up to 2-3-fold increase in both inward and outward K ACh currents.
Basal Currents in Atrial Cardiomyocytes Are Down-regulated by Cholesterol-In addition to Kir3.1/Kir3.4 channels that underlie I K,ACh , atrial cardiomyocytes also express inwardly rectifying potassium channels, which are members of the Kir2 family and underlie I K1 (31). We previously showed that cholesterol enrichment suppresses whereas cholesterol depletion enhances Kir2 currents in endothelial cells and heterologously expressed channels in CHO cells or Xenopus oocytes (18,(32)(33)(34). Interestingly, the inwardly rectifying non-I K,ACh -induced basal currents in atrial cardiomyocytes, which are not sensitive to tertiapin, were regulated in an opposite fashion compared with I K,ACh -induced currents (Fig. 3). Specifically, cholesterol enrichment resulted in approximately a 30% decrease in basal currents, whereas cholesterol depletion resulted in about a 70% increase. Furthermore, similarly to our observations in cholesterol-enriched myocytes, high-cholesterol diet resulted in approximately a 60% decrease in basal currents as can be seen in Fig. 3, D-F. Thus, the effect of cho-lesterol on basal currents was similar to its effect on members of the Kir2 subfamily.
Cholesterol Enrichment of Xenopus Oocytes Leads to Increased Kir3.1/Kir3.4 Currents-To gain mechanistic insight into the effect of cholesterol on K ACh currents, we turned to the Xenopus oocyte heterologous expression system. This system is well suited for expression of proteins following cRNA injection, since it contains accumulated stores of enzymes, organelles and proteins that can be recruited for heterologous proteins (35), allowing basal activation of Kir3 currents (36). As Fig. 4 shows, the heterotetrameric Kir3.1/Kir3.4 channel that underlies I K,ACh rendered an ionic current that was facilitated by cholesterol enrichment, albeit to a lesser extent than Kir3.4*. In contrast, the homomer Kir3.1* was suppressed by elevated levels of cholesterol. Because Kir3.4* exhibits enhanced basal currents following cholesterol enrichment similarly to Kir3.1/Kir3.4 in atrial myocytes, we used it as a model system, and confirmed the results obtained using Kir3.1/Kir3.4.
Cholesterol Has a Subtle Effect on Kir3 Channel-PI(4,5)P 2 Interactions-The activity of Kir3 channels depends on PI(4,5)P 2 , and requires the presence of an additional gating molecule, such as the ␤␥ subunits of G proteins that may enhance channel-PI(4,5)P 2 interactions (27,37). Thus, to gain mechanistic insight into the effect of cholesterol on I K,ACh , we first examined whether cholesterol acts through PI(4,5)P 2 using Xenopus oocytes. It was previously shown that channel run-down following macropatch excision is associated with PI(4,5)P 2 dephosphorylation (27). Therefore, when the channel binds more strongly to PI(4,5)P 2 , the rundown is slower. We thus examined the effect of cholesterol enrichment on current run-down. As depicted in supplemental Fig. S1A, the ionic current run-down of cholesterol-enriched oocytes was slightly slower than that of Kir3.4* in absence of cholesterol (control).
To further explore whether PI(4,5)P 2 could account for the cholesterol effect, we utilized Ci-VSP, a protein whose voltage sensor is coupled to a phosphoinositide phosphatase that dephosphorylates PI(4,5)P 2 to PI(4)P following membrane depolarization (28). The advantage of this method is that it allows us to examine the effect of dephosphorylation of PI(4,5)P 2 on whole-cell currents from intact cells using two- electrode voltage clamp (TEVC) rather than by inside-out recording from membrane macropatches. As with other methods that deplete PI(4,5)P 2 , activation of Ci-VSP results in current run-down, and run-down is slower and less complete when the interaction of the channel with PI(4,5)P 2 is stronger. Before examining the effect of cholesterol enrichment on Kir3.4* current rundown, we tested the effect of coexpressing Ci-VSP with Kir3.4* on the consequences of cholesterol enrichment. As can be seen in supplemental Fig. S1B when compared with Fig. 4B, mere coexpression of Ci-VSP with Kir3.4* did not alter the effect of cholesterol enrichment on Kir3.4* currents. We then proceeded to activate Ci-VSP by cycling between Ϫ80 mV and ϩ80 mV to deplete PI(4,5)P 2 . As can be seen in Fig. 5, A and B, activation of Ci-VSP by depolarization caused a rapid decrease of Kir3.4* whole-cell currents to 12% of their initial value exhibited prior to Ci-VSP activation. In cholesterol-enriched oocytes, however, the current decreased to 18% of its initial value. Similarly, a small difference was also observed between control and cholesterol-enriched oocytes in the inhibition of Kir3.1/Kir3.4 currents following Ci-VSP activation (Fig. 5, C  and D).
Could this small difference reflect that PI(4,5)P 2 interaction with the channel was slightly strengthened by cholesterol enrichment? The strength of channel-PI(4,5)P 2 interactions has been found to correlate strongly with current amplitude (38). We proceeded to examine the effect of Ci-VSP on the D223N mutant of Kir3.4*, whose interaction with PI(4,5)P 2 is significantly stronger than that of Kir3.4* (39), and whose  FEBRUARY 10, 2012 • VOLUME 287 • NUMBER 7 whole-cell currents are comparable with the currents of Kir3.4* following cholesterol enrichment (Fig. 5E). As expected, D223N currents decreased only to 56% of their initial value following the activation of Ci-VSP (Fig. 5, A and B). Furthermore, since D223N not only exhibited stronger channel-PI(4,5)P 2 interactions than Kir3.4* but could also be activated by PI(4,5)P 2 alone (39), we examined the effect of this mutation on cholesterol sensitivity. As can be seen in Fig. 5F, cholesterol failed to stimulate D223N currents. This result is similar to our earlier observation regarding the effect on cholesterol sensitivity of mutating position 229, which is located in the cytosolic CD loop across from D223 (see Fig. 5G). Specifically, the I229L mutation also abrogates the cholesterol sensitivity of Kir3.4* (18). Notably, this mutation also strengthens channel-PI(4,5)P 2 interactions (27). In summary, cholesterol stimulated Kir3.4* currents to levels comparable to Kir3.4*D223N but did not affect the currents of this mutant. Whereas dephosphorylation of PI(4,5)P 2 by Ci-VSP greatly reduced Kir3.4* or cholesterolstimulated Kir3.4* currents, it had a significantly smaller effect on Kir3.4*D223N currents. These results suggest that the mechanism by which cholesterol affects Kir3 currents cannot be accounted for by a strengthening of channel-PI(4,5)P 2 interactions, at least in a manner similar to that achieved with the D223N mutation.

Hypercholesterolemia Up-regulates K ACh Currents
Cholesterol Affects Kir3 Channels via a G␤␥-independent Mechanism-Next, we proceeded to examine the relationship of cholesterol sensitivity to gating mechanisms other than PI(4,5)P 2 , such as that by the ␤␥ subunits of G proteins. As can be seen in Fig. 6, A and B, the sensitivity of both Kir3.4* and Kir3.1/Kir3.4 to cholesterol was abrogated by coexpression of G␤ 1 ␥ 2 , suggesting that a G␤␥-related mechanism could play a direct role in cholesterol modulation of Kir3 channels.
Notably, however, since both cholesterol enrichment and increased levels of G␤␥ enhance Kir3 currents (see supplemental Fig. S2, A and B), it is possible that the cholesterol sensitivity of Kir3 channels was abrogated following coexpression of G␤ 1 ␥ 2 due to saturation of the currents by the ␤␥ subunits of the G protein. Thus, to further investigate this possibility, we examined the effect of GTP␥S on cholesterol sensitivity of I K,ACh in atrial myocytes following cholesterol depletion. All heterotrimeric G proteins employ a signaling strategy in which the metastable GTP-bound state of the G␣ subunit is used as a molecular clock (40,41). When GTP is hydrolyzed to GDP, G␣ is inactivated, and its affinity for free G␤␥ is increased. The potential contributions of G protein turnover to the kinetics of G protein-mediated signaling can be eliminated with the hydrolysis-resistant GTP analog, GTP␥S. Binding of GTP␥S to G␣ persistently activates G proteins, constraining both G␣ and G␤␥ to their active states, reducing the signaling pathway to diffusion-limited interactions between G␤␥ and the channel (40,41). As can be seen in Fig. 6, C and D, in the presence of GTP␥S in atrial myocytes, cholesterol depletion did not affect K ACh currents. As expected and in contrast, GTP␥S did not affect cholesterol modulation of Kir2 basal currents whose activation does not depend on G␤␥ (Fig. 6E).
A simple explanation of our findings is that cholesterol may modulate Kir3 channels via a G␤␥-related mechanism. It is also possible, however, that in the presence of elevated levels of G␤␥, stabilization of the channel in its open state becomes dominant. Thus, utilizing the versatility of the Xenopus oocyte heterologous expression system, we carried out the following additional experiments: (1) examination of the effect of coexpression of Kir3.4* and of Kir3.1/Kir3.4 with the pH domain of the ␤ adrenergic receptor kinase, ␤ARK-PH, a G␤␥ scavenger (42-44) (supplemental Fig. S2, C and D), on cholesterol sensitivity; (2) examination of the effect of the E339Q mutation of Kir3.4*, which impairs agonist-induced sensitivity and decreases binding to G␤ 1 ␥ 2 (43) (supplemental Fig. S2E). As can be seen in Fig. 7, A-C, in all cases, the sensitivity of the  channel to cholesterol was not affected compared with the control channel, suggesting that G␤␥ is not necessary for cholesterol modulation of Kir3 channels, and that cholesterol acts independently of G␤␥ (Fig. 7D).

DISCUSSION
The major finding of this study is that cholesterol has a substantial facilitating effect on K ACh channels function in atrial cardiomyocytes. This was observed both following cholesterol cell enrichment (Fig. 1) and in myocytes from rats subject to high-cholesterol diet (Fig. 2). Consistent with these observations, cholesterol depletion resulted in decreased I K,ACh (Fig. 1).
Interestingly, in myocytes that express both Kir2 and Kir3 channels, cholesterol has an opposite effect on basal and AChinduced currents. Thus, whereas under basal stimulation conditions cholesterol enrichment inhibits Kir2 channels, upon stimulation, the dominant effect becomes facilitation of I K,ACh . Furthermore, a facilitatory effect of cholesterol on K ACh channels is unique because the majority of Kir channels are suppressed by cholesterol (18). Differential effects of cholesterol on different types of Kir channels expressed in the same cells may play a major role in fine-tuning the impact of hypercholesterolemia on heart function.
Moreover, our data demonstrate that even within the same Kir subfamily, the impact of cholesterol may vary. Our data suggest that cholesterol-induced facilitation of K ACh currents should be attributed to a mechanism independent of PI(4,5)P 2 . Specifically, Fig. 5, A-D, argues that the effect of cholesterol on channel gating is not predominantly by stabilizing the channel-PI(4,5)P 2 activated state. Moreover, Fig. 5F argues that the channel-PI(4,5)P 2 -stabilized activated state occludes any further cholesterol effect. Thus, since Fig. 5, A-D suggest that PI(4,5)P 2 stabilization of the cholesterol-activated state is a small component of the effect, cholesterol may be acting by a distinct mechanism to stabilize the open state of the channel. Once the open state is stabilized by another mechanism (e.g. strengthening of channel-PI(4,5)P 2 interactions by G␤␥), cholesterol does not produce a further effect as evident in Fig. 6, A-D. In accord, when the open state is not stabilized by another mechanism (e.g. following the removal of G␤␥), cholesterol plays a major role in channel gating as Fig. 7, A-C demonstrate. This interpretation is consistent with our recent work (45) showing that the cholesterol sensitivity of Kir2.1 depends on a belt of residues, which are correlated with the major cytosolic gate of the channel, the G-loop (46). Thus, the PI(4,5)P 2 and the cholesterol-active states are likely to reflect a common open conducting channel state of Kir3.1/Kir3.4.
In contrast and similarly to our previous studies of the effect of cholesterol on Kir2 channels in other cells (18,32,33), cho-  [11][12], alone and coexpressed with G␤ 1 ␥ 2 . For each pair, the currents are normalized relative to control untreated oocytes expressing the same constructs. C, tertiapin-sensitive I-V relationships of I K,ACh in the presence of GTP␥S for control and cholesterol-depleted atrial myocytes (n ϭ 8 -9). D, summary data for I K,ACh at Ϫ80 mV in the presence of GTP␥S. E, summary data for tertiapin-insensitive whole-cell inward current densities at Ϫ80 mV in the presence of GTP␥S. Significant difference is indicated by an asterisk (*, p Ͻ 0.05); (n ϭ 8 -9). lesterol increase inhibits I K1 in atrial myocytes, an effect that has been previously attributed to direct Kir2-cholesterol interactions (34) and to stabilization of the channel in its closed state (33). Furthermore and similarly to the effect of high-cholesterol diet on endothelial cells isolated from hypercholesterolemic animals (47), also in atrial myocytes high-cholesterol diet resulted in significantly lower Kir2 currents compared with cells isolated from control animals. Thus, cholesterol may regulate different Kir channels by opposite mechanisms. Accordingly, whereas cholesterol stabilizes the closed state of Kir2 channels, it stabilizes the open state of Kir3.1/Kir3.4.
Since hypercholesterolemia is a risk factor for cardiovascular disease, the increase in I K,ACh following cholesterol enrichment and in hypercholesterolemic animals is likely to play a key role in the function of the heart. It was previously shown that chronic atrial fibrillation patients exhibit agonist-independent constitutive K ACh activity that contributes to the enhanced basal inward rectifier current (48). Corresponding constitutively active I K,ACh -like currents are up-regulated by canine atrial tachypacing (49,50). In addition, selective blocking of I K,ACh by tertiapin results in action potential prolongation and suppression of inducible atrial-fibrillation (50). Thus, cholesterol-induced up-regulation of K ACh activity may also contribute significantly to the development of atrial fibrillation. Moreover, statins, which are cholesterol lowering drugs, have been suggested as potential treatment for preventing AF (51). Yet, the diverse mechanisms by which statins could reduce AF onset are uncertain (52). Thus, our findings may explain the association of statins to decreased risk of AF, and may lead to a novel therapeutic approach to prevent and reverse AF.