Restoration of Normal L-Type Ca2+ Channel Function During Timothy Syndrome by Ablation of an Anchoring ProteinNovelty and Significance
Rationale: L-type Ca2+ (CaV1.2) channels shape the cardiac action potential waveform and are essential for excitation–contraction coupling in heart. A gain-of-function G406R mutation in a cytoplasmic loop of CaV1.2 channels causes long QT syndrome 8 (LQT8), a disease also known as Timothy syndrome. However, the mechanisms by which this mutation enhances CaV1.2-LQT8 currents and generates lethal arrhythmias are unclear.
Objective: To test the hypothesis that the anchoring protein AKAP150 modulates CaV1.2-LQT8 channel gating in ventricular myocytes.
Methods and Results: Using a combination of molecular, imaging, and electrophysiological approaches, we discovered that CaV1.2-LQT8 channels are abnormally coupled to AKAP150. A pathophysiological consequence of forming this aberrant ion channel-anchoring protein complex is enhanced CaV1.2-LQT8 currents. This occurs through a mechanism whereby the anchoring protein functions like a subunit of CaV1.2-LQT8 channels that stabilizes the open conformation and augments the probability of coordinated openings of these channels. Ablation of AKAP150 restores normal gating in CaV1.2-LQT8 channels and protects the heart from arrhythmias.
Conclusion: We propose that AKAP150-dependent changes in CaV1.2-LQT8 channel gating may constitute a novel general mechanism for CaV1.2-driven arrhythmias.
L-type Ca2+ (CaV1.2) channels are expressed in the sarcolemma of atrial and ventricular myocytes, where they play a critical role in activating Ca2+ release from the sarcoplasmic reticulum (SR) during excitation–contraction (EC) coupling. The magnitude and time course of the CaV1.2 current determine the waveform of the cardiac action potential (AP).1 Thus, changes in CaV1.2 channel function can have profound effects on cardiac EC coupling and excitability. Accordingly, a recent study2 discovered that a single amino acid substitution (G406R) in CaV1.2 is linked to Timothy syndrome. Timothy syndrome is characterized by prolongation of the electrocardiogram (ECG) QT interval and lethal arrhythmias, which is why it is also known as long QT syndrome 8 (LQT8). Interest in the mechanisms of LQT8 has been intense because it is a multisystem disease, with many patients also afflicted by autism. Thus, a single amino acid mutation in CaV1.2 causes clinically significant disorders in the cardiac and central nervous systems.
Electrophysiological studies have revealed 2 distinctive features of LQT8 mutant CaV1.2 channels (CaV1.2-LQT8). First, these channels inactivate at a slower rate than do wild-type (WT) channels.2–4 Second, small clusters of CaV1.2-LQT8 channels have a higher probability of undergoing coordinated openings and closings (“coupled gating”) than do WT channels.5 Although recent reports suggested that the G406R substitution in CaV1.2 creates a new phosphorylation site for the Ca2+/calmodulin–dependent kinase II (CaMKII), which contributes to an increase in the open probability (Po) of CaV1.2-LQT8 channels, others suggested that phosphorylation by CaMKII is not necessary for their slower rate of inactivation.4,6,7 Thus, the mechanism by which the activity of CaV1.2-LQT8 channels is coordinated to generate irregular cardiac rhythm is unclear.
A potential mechanism regulating the activity of CaV1.2-LQT8 channels involves the anchoring protein AKAP150. AKAP150 targets specific protein kinases and phosphatases to regions near CaV1.2 channels in ventricular myocytes and neurons.8,9 Furthermore, AKAP150 binds to the carboxyl tail of CaV1.2 channels via leucine zipper (LZ) motifs in these 2 proteins,10 facilitating physical interactions between CaV1.2 carboxyl tails. AKAP150 increases the probability of long openings and coupled gating events between CaV1.2 channels.5 At present, however, whether the interaction with AKAP150 modulates the abnormal CaV1.2-LQT8 channel activity is unknown.
Here, we used a combination of cellular, molecular, imaging, and electrophysiological approaches to investigate this important issue. We discovered that AKAP150 is required for abnormal gating of CaV1.2-LQT8 channels. Importantly, our data indicate that ablation of AKAP150 corrects arrhythmogenic CaV1.2-LQT8 channel activity in ventricular myocytes.
An expanded Methods section is available in the Online Supplemental Material at http://circres.ahajournals.org.
Mice were euthanized using a lethal dose of sodium pentobarbital as approved by the University of Washington Institutional Animal Care and Use Committee. Details about the generation of our LQT8 mouse are available in the Online Supplemental Material. Ventricular myocytes were isolated as described previously.8 Electrophysiological signals were recorded using HEKA EPC10 or Axopatch 200B amplifiers. Images were obtained using a confocal microscope. Data are presented as mean±SEM. A probability value of less than 0.05 was considered significant. An asterisk was used in the figure to illustrate a significant difference between groups.
Ablation of AKAP150 Protects Against Cardiac Hypertrophy During LQT8
We generated a transgenic mouse that expresses CaV1.2-LQT8 channels fused to the tag-red fluorescent protein (tRFP) solely in cardiac myocytes (LQT8; Figure 1A) and crossed them with AKAP150 null mice (LQT8/AKAP150−/−).11 Online Table I summarizes 21 different anatomic and functional features of these mice. We found that the heart-to-body weight ratio of LQT8 hearts was larger than that of WT, AKAP150−/−, and LQT8/AKAP150−/− mice. Indeed, LQT8 myocytes were longer and wider than WT, AKAP150−/−, and LQT8/AKAP150−/− myocytes. These findings suggest that expression of CaV1.2-LQT8 promotes cardiac hypertrophy and loss of AKAP150 protects LQT8 mice against it.
AKAP150 Is Not Required for the Expression or Spatial Organization of CaV1.2-LQT8 Channels in Adult Ventricular Myocytes
Western blot analysis of biotinylated endogenous WT CaV1.2 (CaV1.2-WT) and CaV1.2-LQT8 indicated that sarcolemmal CaV1.2-WT expression was similar in WT, LQT8, AKAP150−/−, and LQT8/AKAP150−/− myocytes (Figure 1B). CaV1.2-LQT8 channels comprised 41%±5 (n=6 mice) and 43%±4% (n=6 mice) of the total sarcolemmal CaV1.2 population in LQT8 and LQT8/AKAP150−/− myocytes, respectively. As with CaV1.2-WT channels in WT and AKAP150−/− myocytes, CaV1.2-LQT8 channels were prominently expressed along the transverse tubules (T-tubules) of LQT8 and LQT8/AKAP150−/− myocytes. However, unlike CaV1.2-WT channels, CaV1.2-LQT8 channels were also expressed in the intercalated discs and seemed to form multiple clusters in the sarcolemma and near the nuclear envelope of LQT8 and LQT8/AKAP150−/− cells (Figure 1C). The number of CaV1.2-LQT8 clusters were similar in LQT8 (Online Figure II, 154±7 clusters/cell, n=7) and LQT8/AKAP150−/− cells (142±68 clusters/cell, n=5; P>0.05) (see Online Supplemental material for a description of this analysis). Collectively, these data suggest that CaV1.2-LQT8 and CaV1.2-WT channels are differentially expressed in ventricular myocytes, but that AKAP150 does not regulate the expression or distribution of these channels in these myocytes.
Loss of AKAP150 Restores Normal Inactivation of ICa in LQT8 Myocytes
We recorded macroscopic CaV1.2 currents (ICa) from WT, AKAP150−/−, LQT8, and LQT8/AKAP150−/− ventricular myocytes. Although the amplitude of ICa was similar in WT, AKAP150−/−, LQT8, and LQT8/AKAP150−/− ventricular myocytes (P>0.05), there were striking differences in the rate of inactivation of these currents (Figure 2A and 2B and Online Table I). Indeed, the fraction of ICa remaining 50 ms (r50) after the onset of depolarization to +10 mV from LQT8 myocytes was larger (n=8) than in WT (n=9) and AKAP150−/− myocytes (n=5; P<0.05), suggesting expression of functional CaV1.2-LQT8 channels in LQT8 myocytes. Indeed, from these ICa currents, we determined that CaV1.2-LQT8 channels account for ≈32% of the total CaV1.2 channel population in LQT8 myocytes (see Online Supplemental Material). Interestingly, the r50 of ICa in LQT8/AKAP150−/− (n=9) was similar to that of WT and AKAP150−/−. These data suggest that loss of AKAP150 restores normal ICa inactivation in LQT8 myocytes.
Our ICa data raise an important question: is AKAP150 required for the expression of functional CaV1.2-LQT8 channels? To address this question, we expressed these channels in WT and AKAP150−/− mouse embryonic fibroblasts (MEFs). As shown in Figure 2C, we recorded robust ICa (1 to 3 pA/pF) only in cells transfected with CaV1.2-WT or CaV1.2-LQT8. In WT MEFs (Figure 2C and 2D and Online Table I), CaV1.2-LQT8 currents (r50=0.73±0.10, n=5) inactivated at a much slower rate than did CaV1.2-WT currents at +10 mV (r50=0.25±0.02, n=6; P<0.05). However, in AKAP150−/− MEFs, CaV1.2-LQT8 channels (r50=0.28±0.03, n=5) produced currents with a similar time course to that of CaV1.2-WT channels (r50=0.35±0.03 at +10 mV; n=5 cells; P>0.05). Thus, although AKAP150 is not necessary for the expression of functional WT or LQT8 CaV1.2 channels, it is required for defective inactivation of CaV1.2-LQT8 channels.
A potential mechanism by which AKAP150 could promote a slow rate of inactivation of CaV1.2-LQT8 currents is by acting as an anchor for protein kinase A (PKA).8,10 Another possibility is that the effects of AKAP150 on CaV1.2-LQT8 channel inactivation depend on CaMKII activity. Application of ht31 (PKA-AKAP interaction inhibitor, 10 μmol/L), Rp-cAMP (PKA inhibitor, 100 μmol/L), or KN-93 (CaMKII inhibitor, 5 μmol/L) did not change the r50 of ICa in LQT8 myocytes (Online Figure I, P>0.05), which suggests that PKA or CaMKII activity is not responsible for the potentiation of ICa during LQT8. Furthermore, these data support the view that the necessity of AKAP150 for decreased CaV1.2-LQT8 channel inactivation is not dependent on CaMKII activity or its ability to target PKA locally.
AKAP150 Is Required for Increased CaV1.2 Channel Activity and Coupled Gating Seen in LQT8 Myocytes
To test the hypothesis that ablation of AKAP150 decreases the Po, open time, and frequency of coupled gating events by CaV1.2 channels in LQT8 myocytes, we recorded the in situ activity of CaV1.2 channels in WT, LQT8, and LQT8/AKAP150−/− myocytes using the cell-attached configuration of the patch clamp technique (Figure 3A and Online Table I). AKAP150−/− myocytes were not included in these experiments because the amplitude, rate of inactivation, and voltage dependence of ICa in these cells is similar to that of WT cells and LQT8/AKAP150−/− cells. Thus, it is unlikely that single CaV1.2 channel activity in AKAP150 null myocytes would be different from that of WT and LQT8/AKAP150−/− cells.
The amplitudes of elementary Ca2+ currents were similar in WT (0.55±0.10 pA, n=8 cells), LQT8 (0.60±0.11 pA, n=12 cells), and LQT8/AKAP150−/− (0.58±0.12 pA, n=10 cells) myocytes at −30 mV (P>0.05). Consistent with our ICa data, the activity (ie, NPo where N is the number of channels and Po is the open probability) of CaV1.2 channels in LQT8 myocytes (0.11±0.04) was ≈10-fold higher than in WT (0.01±0.01) and LQT8/AKAP150−/− (0.02±0.01) myocytes (P<0.05; Figure 3B). Furthermore, analysis of the open dwell times from CaV1.2 channels revealed that a larger proportion of channel openings are long openings in LQT8 myocytes in comparison with those recorded from LQT8/AKAP150−/− and WT myocytes. The open time histograms from WT and LQT8/AKAP150−/− myocytes could be fit with a single exponential function with a time constant (τshort) of 0.8 ms and 0.6 ms, and the open time histogram of CaV1.2 channels in LQT8 myocytes could be fit with the sum of 2 exponential functions with τshort of 1.3 ms and τlong of 9.4 ms, which accounted for 95% and 5% of the channel openings, respectively (Figure 3B). The time constants from LQT8 myocytes likely represents a mixed population of WT and LQT8 CaV1.2 channels operating in 2 gating modalities in LQT8 myocytes. By contrast, the long CaV1.2 channel openings observed in LQT8 myocytes were completely absent in LQT8/AKAP150−/− cells. Collectively, these data suggest that AKAP150 is required for long openings of CaV1.2 channels in LQT8 myocytes.
To test the hypothesis that CaV1.2-LQT8 channels have a higher probability of coupled gating than do CaV1.2-WT channels in ventricular myocytes, we implemented a coupled Markov chain model to determine the coupling coefficient (κ) among CaV1.2 channels.5,12 The mean coupling coefficient was 0.13±0.03 for Ca2+ channels in LQT8 myocytes and 0.03±0.01 for WT and 0.03±0.01 for LQT8/AKAP−/− cells (Figure 3D). Indeed, the frequency of coupled gating events (κ > 0.1) was higher in LQT8 (43%±10%) myocytes than in WT (8%±4%) and LQT8/AKAP150−/− (10%±6%) myocytes (P<0.05; Figure 3E).
Loss of AKAP150 Restores Normal [Ca2+]i, AP Waveform, and Cardiac Rhythm in LQT8 Mice
We recorded AP-evoked [Ca2+]i transients in WT, LQT8, AKAP150−/−, and LQT8/AKAP150−/− myocytes (Figure 4A and Online Table I). The amplitudes of the AP-evoked [Ca2+]i transient in WT myocytes (n=7), AKAP150−/− (n=7), and LQT8/AKAP150−/− myocytes (n=9) were similar (P>0.05). The [Ca2+]i transient was larger in LQT8 myocytes (n=9) than in these myocytes (P<0.05). Furthermore, although 56% of LQT8 myocytes had spontaneous Ca2+ release (SCR) events under control conditions, none was detected in WT, AKAP150−/−, or LQT8/AKAP150−/− myocytes under similar experimental conditions. Because AKAP150 is required for β-adrenergic–induced increases in the amplitude of the AP-evoked [Ca2+]i transient in ventricular myocytes,8 we examined the effects of the β-adrenergic agonist isoproterenol (ISO, 100 nmol/L) on WT, LQT8, AKAP150−/−, and LQT8/AKAP150−/− myocytes (Figure 4A and Online Table I). We found that ISO increased the amplitude of the AP-evoked [Ca2+]i in WT and LQT8, but not in AKAP150−/− or LQT8/AKAP150−/− myocytes, providing functional confirmation of the loss of AKAP150 in these cells (P<0.05). ISO also increased the number of spontaneous Ca2+ release events in CaV1.2-LQT8 cells from 40% to 85%, but not in WT, AKAP150−/−, and LQT8/AKAP150−/− myocytes.
We investigated whether restoration of normal inactivation of ICa in LQT8/AKAP150−/− myocytes translated to changes in AP waveform in these cells. Consistent with our ICa data, the duration of the AP at 90% repolarization (APD90) was longer in LQT8 (n=10) than in WT (n=5), AKAP150−/− (n=5), and LQT8/AKAP150−/− (n=11) myocytes (P<0.05; Figure 4B and Online Table I). In addition, analysis of records with trains of APs revealed that LQT8 myocytes had a higher frequency of early (EADs) and delayed afterdepolarizations (DADs) than did WT, AKAP150−/−, and LQT8/AKAP150/− myocytes (Figure 4C and Online Table I).
To determine the electrophysiological phenotype of WT, LQT8, AKAP150−/−, and LQT8/AKAP150−/− mice, we implanted telemetric ECG transmitters13 (Figure 4D and Online Table I). Heart rate was similar in WT (n=6), LQT8 (n=5), AKAP150−/− (n=6), and LQT8/AKAP150−/− at rest (n=6) or during mild exercise (P>0.05). However, consistent with our ICa and AP data, the QT interval—corrected for heart rate using Bazet's formula (ie, QTc)—of LQT8 mice (116±1 ms) is longer than that of WT (97±1 ms), AKAP150−/− (98±1 ms), and LQT8/AKAP150−/− mice (108±1 ms; P<0.05). During exercise, although multiple premature ventricular depolarizations (PVDs) and episodes of torsades de pointes (TdPs, a hallmark of LQT) were observed in LQT8 mice, none was recorded from WT, AKAP150−/−, and LQT8/AKAP150−/− mice (Figure 4D and Online Table I). Thus, loss of AKAP150 was protective against arrhythmias in mice expressing CaV1.2-LQT8.
Our findings suggest a new model of CaV1.2-LQT8 channel dysfunction during Timothy syndrome (Figure 4E). In this model, the anchoring protein AKAP150 and CaV1.2-LQT8 form a complex that is necessary for aberrant CaV1.2-LQT8 channel gating and arrhythmias. CaV1.2-LQT8 channels likely interact with AKAP150 via LZ motifs in the carboxyl tails of both proteins.10 We propose that AKAP150 functions like an allosteric modulator of CaV1.2-LQT8 channels, increasing CaV1.2-LQT8 currents by stabilizing the open conformation and increasing the probability of coupled gating between CaV1.2-LQT8 channels. This leads to increased Ca2+ influx, AP prolongation, cardiac hypertrophy, and arrhythmias. Coupled gating of CaV1.2-LQT8 channels presumably occurs because AKAP150 promotes physical interactions of adjacent channels via their carboxyl tails.5,10,14
Our data provide insights into the cellular mechanisms by which CaV1.2-LQT8 channels increase the probability of arrhythmias. We found that expression of CaV1.2-LQT8 channels increased the frequency of arrhythmogenic EADs and DADs. EADs are likely produced by reactivation of CaV1.2 channels during the long APs of LQT8 myocytes. It is intriguing to speculate that the larger Ca2+ influx associated with CaV1.2-LQT8 channels leads to SR Ca2+ overload and thus to SRC events and DADs in LQT8 myocytes. Future experiments should examine in detail the relationship between Ca2+ influx via CaV1.2-LQT8 and EADs and DADs in these cells.
Ablation of AKAP150 corrects pathological CaV1.2-LQT8 channel gating and arrhythmias and prevents hypertrophy of LQT8 hearts presumably by decreasing Ca2+ influx via CaV1.2-LQT8 channels. Because AKAP150 does not bind CaMKII, loss of this scaffolding protein is not expected to affect CaMKII-dependent modulation of CaV1.2-LQT8 channels in ventricular myocytes. However, our data suggest that AKAP150 is required for any potential CaMKII-induced changes in CaV1.2-LQT8 gating. Thus, we propose that disrupting the interaction between AKAP150 and CaV1.2-LQT8 is a potential target for novel therapeutics for treating the broad spectrum of Timothy syndrome's symptoms, including lethal arrhythmias and autism.
Sources of Funding
Supported by the National Institutes of Health and the American Heart Association.
We thank Dr Simon Hinke, Ms Jennifer Cabarrus and Katherine Forbush for technical assistance, Dr Richard D. Palmiter for helpful discussions, and Dr Michael T. Chin for reviewing ECG records.
In May 2011, the average time from submission to first decision for all original research papers submitted to Circulation Research was 14.5 days.
Brief UltraRapid Communications are designed to be a format for manuscripts that are of outstanding interest to the readership, report definitive observations, but have a relatively narrow scope. Less comprehensive than Regular Articles but still scientifically rigorous, BURCs present seminal findings that have the potential to open up new avenues of research. A decision on BURCs is rendered within 7 days of submission.
Non-Standard Abbreviations and Acronyms
- A-kinase anchoring protein 150
- action potential
- intracellular Ca2+ concentration
- Ca2+/calmodulin-dependent kinase II
- CaV1.2 channels with the long QT syndrome mutation
- wild-type CaV1.2 channels
- EC coupling
- excitation–contraction coupling
- long QT syndrome 8 (Timothy syndrome)
- leuzine zipper
- mouse embryonic fibroblast
- Torsades de pointes
- tag red fluorescent protein
- wild type
- Received May 10, 2011.
- Revision received May 31, 2011.
- Accepted June 13, 2011.
- © 2011 American Heart Association, Inc.
- Luo CH,
- Rudy Y
- Splawski I,
- Timothy KW,
- Sharpe LM,
- Decher N,
- Kumar P,
- Bloise R,
- Napolitano C,
- Schwartz PJ,
- Joseph RM,
- Condouris K,
- Tager-Flusberg H,
- Priori SG,
- Sanguinetti MC,
- Keating MT
- Barrett CF,
- Tsien RW
- Thiel WH,
- Chen B,
- Hund TJ,
- Koval OM,
- Purohit A,
- Song LS,
- Mohler PJ,
- Anderson ME
- Navedo MF,
- Cheng EP,
- Yuan C,
- Votaw S,
- Molkentin JD,
- Scott JD,
- Santana LF
- Erxleben C,
- Liao Y,
- Gentile S,
- Chin D,
- Gomez-Alegria C,
- Mori Y,
- Birnbaumer L,
- Armstrong DL
- Nichols CB,
- Rossow CF,
- Navedo MF,
- Westenbroek RE,
- Catterall WA,
- Santana LF,
- McKnight GS
- Coghlan VM,
- Perrino BA,
- Howard M,
- Langeberg LK,
- Hicks JB,
- Gallatin WM,
- Scott JD
- Tunquist BJ,
- Hoshi N,
- Guire ES,
- Zhang F,
- Mullendorff K,
- Langeberg LK,
- Raber J,
- Scott JD
- Gold MG,
- Stengel F,
- Nygren PJ,
- Weisbrod CR,
- Bruce JE,
- Robinson CV,
- Barford D,
- Scott JD
Novelty and Significance
What Is Known?
A single amino acid substitution in CaV1.2 L-type Ca2+ channels causes long QT syndrome 8 (LQT8).
CaV1.2-LQT8 channels are characterized by an abnormally slow rate of inactivation and by exhibiting a high frequency of coordinated openings between nearby channels.
The A-kinase anchoring protein 150 (AKAP150) is a CaV1.2 channel–associated scaffolding protein that regulates CaV1.2 channel function and excitation–contraction (EC) coupling by targeting adenyl cyclase 5, protein kinase A, and calcineurin near these channels.
What New Information Does This Article Contribute?
AKAP150 is required for the expression of the LQT8 phenotype in a mouse model of this disease.
AKAP150 functions like an allosteric modulator of CaV1.2-LQT8 channels that increases the opening time and also facilitates coupled gating between these channels in LQT8 cardiac myocytes.
AKAP150 directly modulates the gating of CaV1.2-LQT8 without the aid of kinases.
The mechanism by which the LQT8 mutation alters the function of CaV1.2-LQT8 and EC coupling is unclear. Here, we establish that AKAP150 is necessary for the expression of the LQT8 phenotype. We find that AKAP150 functions as an accessory protein to the mutant CaV1.2-LQT8 channels, directly modulating the gating of these channels independently of its role in targeting adrenergic signaling. We also find that the coupled gating modality plays an important role in the pathophysiology of LQT8. The increased activity of CaV1.2-LQT8 in complex with AKAP150 increases the frequency of arrhythmogenic voltage fluctuations and arrhythmias. Our findings establish a novel role for AKAP150 as a CaV1.2 accessory protein in LQT8, and suggest that disruption of the interaction between CaV1.2 and AKAP150 could be a potential novel therapeutic target for LQT8 and other arrhythmias.