Proarrhythmic defects in Timothy syndrome require calmodulin kinase II

Abstract

Background: Timothy syndrome (TS) is a disease of excessive cellular Ca(2+) entry and life-threatening arrhythmias caused by a mutation in the primary cardiac L-type Ca(2+) channel (Ca(V)1.2). The TS mutation causes loss of normal voltage-dependent inactivation of Ca(V)1.2 current (I(Ca)). During cellular Ca(2+) overload, the calmodulin-dependent protein kinase II (CaMKII) causes arrhythmias. We hypothesized that CaMKII is a part of the proarrhythmic mechanism in TS.

Methods and results: We developed an adult rat ventricular myocyte model of TS (G406R) by lentivirus-mediated transfer of wild-type and TS Ca(V)1.2. The exogenous Ca(V)1.2 contained a mutation (T1066Y) conferring dihydropyridine resistance, so we could silence endogenous Ca(V)1.2 with nifedipine and maintain peak I(Ca) at control levels in infected cells. TS Ca(V)1.2-infected ventricular myocytes exhibited the signature voltage-dependent inactivation loss under Ca(2+) buffering conditions, not permissive for CaMKII activation. In physiological Ca(2+) solutions, TS Ca(V)1.2-expressing ventricular myocytes exhibited increased CaMKII activity and a proarrhythmic phenotype that included action potential prolongation, increased I(Ca) facilitation, and afterdepolarizations. Intracellular dialysis of a CaMKII inhibitory peptide, but not a control peptide, reversed increases in I(Ca) facilitation, normalized the action potential, and prevented afterdepolarizations. We developed a revised mathematical model that accounts for CaMKII-dependent and CaMKII-independent effects of the TS mutation.

Conclusions: In TS, the loss of voltage-dependent inactivation is an upstream initiating event for arrhythmia phenotypes that are ultimately dependent on CaMKII activation.

Figure 1

Dihydropyridine-resistant Ca_V1.2 α subunit Timothy Syndrome (TS) model. (A) A topology diagram of Ca_V1.2 depicting dihydropyridine resistance mutation (DHP^R, black circle), extracellular hemaglutanin epitope (HA, green circle) and the TS mutation (G406R, gray circle) on the I-II intracellular loop. (B) Immunoblot (HA Ig) of HEK293T cells expressing the modified Ca_V1.2 or empty vector control. (C) FITC immunofluorescence (HA Ig) of HEk293T cells expressing the modified Ca_V1.2 with corresponding nuclear stain by DAPI (Scale bar, 10μm). (D) Ca_V1.2 TS expressing HEK293T cells show a reduction in VDI as compared to HEK293T cells transfected with Ca_V1.2 WT (N=5 cells/point). (E) Exogenous Ca_V1.2 with dihydropyridine mutation is resistant to nifedipine. (F) Preserved I_Ca during exposure to nifedipine. The single arrow indicates the nifedipine concentration (10nM) used to study the cellular consequences of the TS mutation, and the double arrow indicates the nifedipine concentration (1μM) to overcome dihydropyridine resistance and block the majority of I_Ca (N=5-8 cells/point, P<0.05 at each nifedipine concentration).

Figure 2

CaMKII recruitment in the TS adult ventricular myocyte model. (A-C) Non-transduced, (D-F) WT and (G-I) TS adult ventricular myocytes (field stimulated 1Hz for 5 minutes in Tyrodes with 1.8mM CaCl₂). (A,D,G) Exogenous Ca_V1.2 channels are expressed in regularly distributed punctae across ventricular myocytes as shown by HA immunostaining. Both WT and TS Ca_V1.2 show spacing consistent with T-tubule network localization. HA immunofluorescence section of Ca_V1.2 WT, TS mutation, and uninfected negative control. (H) More activated CaMKII (pCaMKII Thr²⁸⁶) immuno-stained with (I) TS ventricular myocytes as compared to WT (E and F) and non-transduced (B and C) ventricular myocytes. (Scale bar, 10μm)

Figure 3

CaMKII inhibition reverses TS ventricular myocyte action potential (AP) prolongation and afterdepolarizations. (A) Action potential recordings from WT and TS ventricular myocytes. The first action potential for each sweep was initiated by injected current (arrow head), but the subsequent action potentials in TS arose from spontaneous afterdepolarizations. (B) Ca_V1.2 TS results in an increased action potential duration (N=5-10 cells/group, *P=0.018) and (C) afterdepolarizations (N=5-10 cells/group, *P=0.033). Numerals indicate the fraction of cells studied with afterdepolarizations. 1μm Nifedipine inhibited a majority of I_Ca and so prevented the Ca_V1.2 TS increase in (B) action potential duration (N=5-7 cells/group, P-0.39) and (C) frequency of afterdepolarizations (N=5-7 cells/group, P=1.0). (D) Action potential recordings from TS ventricular myocytes with either the CaMKII inhibitory peptide, AC3-I, or a control peptide, AC3-C. (E and F) Dialyzing AC3-I restored action potential duration in TS to WT levels and prevented afterdepolarizations (N=5-10 cells/group, TS with AC3-I compared to WT: APD90% P=0.403, afterdepolarizations P=1.0). AC3-I resulted in a non-significant shortening of the WT AP duration (N=5 cells/group, P=0.25) and no significant change in afterdepolarizations (N=5 cells/group, P=1.0) compared to WT with no peptide. (D-F) Dialyzing the control peptide, AC3-C, did not alter the TS mutation affects on action potential duration or afterdepolarizations (N=5-10 cells/group, TS with AC3-C compared to TS with AC3-I: APD90% *P=0.017, afterdepolarizations *P=0.044). AC3-C did not alter the AP duration (N=5 cells/group, P=0.28) or afterdepolarizations (N=5 cells/group, P=1.0) of WT with no peptide.

Figure 4

TS mutation shifts the VDI independent of Ca²⁺ signaling. (A) The TS mutation shifts the Ca_V1.2 I_Ba VDI (N=5 cells/point, *P=0.008), (B) without changing the current-voltage (IV) relationship (N=5 cells/group, P=0.88).

Figure 5

TS mutation enhances I_Ca facilitation. (A) TS ventricular myocytes exhibit increased peak I_Ca (arrows) during the first depolarizing voltage clamp command step (-80mV to 0mV, 300ms, 0.5Hz) and slowing of inactivation during all depolarizing steps. (B) Integrated I_Ca evoked by repetitive depolarizing voltage command steps (as in A above) is greater in TS mutation than WT (N=6-7 cells/point, ANOVA P<0.001, * P<0.05). (C) The time constant of the fast component of I_Ca inactivation (τ_fast) is significantly slower in TS ventricular myocytes than WT (N=6-7 cells/point, ANOVA P<0.001, * P<0.05). (D and E) Integrated I_Ca and τ_fast were restored to WT levels in TS ventricular myocytes dialyzed with the CaMKII inhibitory peptide, AC3-I (N=5-6 cells/point, TS with AC3-I compared to WT: integrated I_Ca ANOVA P=0.522, τ_fast ANOVA P=0.294). Dialyzing the control peptide, AC3-C, did not alter the TS mutation affects on I_Ca facilitation (N=5 cells/group, TS with AC3-C compared to TS with AC3-I: integrated I_Ca ANOVA P<0.001, * P<0.05; τ_fast ANOVA P<0.001, * P<0.05).

Figure 6

The TS mutation augments intracellular Ca²⁺ handling. (A) Confocal Ca²⁺ transient recordings from WT and TS ventricular myocytes. (B) Summary data showing TS mutation causes an increase in the peak Ca²⁺ transient during 1Hz stimulations (N=14-28 cells/group, *P=0.042). (C) Summary data showing the 50% decay time of the whole cell Ca²⁺ transients were faster in TS ventricular myocytes (N=14-28 cells/group, *P=0.047). (D) No difference was observed in SR Ca²⁺ content between TS and WT ventricular myocytes (N=14-28 cells/group, P=0.524). (E) Ca²⁺ sparks recorded from WT and TS ventricular myocytes. (F) Summary data showing TS infected ventricular myocytes exhibited an increased frequency of Ca²⁺ sparks during diastole (N=22-37 cells/group, *P=0.001).

Figure 7

LRd modeling of WT, TS and TS with AC3-I based upon experimental data from ventricular myocytes. (A) Schematic of LRd model. (B) LRd model indicates CaMKII activation in TS causes increased I_Ca and action potential prolongation (CL = 700ms) and (C) afterdepolarizations.