Unexpected Structural and Functional Consequences of the R33Q Homozygous Mutation in Cardiac Calsequestrin
A Complex Arrhythmogenic Cascade in a Knock In Mouse Model
Catecholaminergic polymorphic ventricular tachycardia (CPVT) is an inherited arrhythmogenic disorder characterized by life threatening arrhythmias elicited by physical and emotional stress in young individuals. The recessive form of CPVT is associated with mutation in the cardiac calsequestrin gene (CASQ2). We engineered and characterized a homozygous CASQ2R33Q/R33Q mouse model that closely mimics the clinical phenotype of CPVT patients. CASQ2R33Q/R33Q mice develop bidirectional VT on exposure to environmental stress whereas CASQ2R33Q/R33Q myocytes show reduction of the sarcoplasmic reticulum (SR) calcium content, adrenergically mediated delayed (DADs) and early (EADs) afterdepolarizations leading to triggered activity. Furthermore triadin, junctin, and CASQ2-R33Q proteins are significantly decreased in knock-in mice despite normal levels of mRNA, whereas the ryanodine receptor (RyR2), calreticulin, phospholamban, and SERCA2a-ATPase are not changed. Trypsin digestion studies show increased susceptibility to proteolysis of mutant CASQ2. Despite normal histology, CASQ2R33Q/R33Q hearts display ultrastructural changes such as disarray of junctional electron-dense material, referable to CASQ2 polymers, dilatation of junctional SR, yet normal total SR volume. Based on the foregoings, we propose that the phenotype of the CASQ2R33Q/R33Q CPVT mouse model is portrayed by an unexpected set of abnormalities including (1) reduced CASQ2 content, possibly attributable to increased degradation of CASQ2-R33Q, (2) reduction of SR calcium content, (3) dilatation of junctional SR, and (4) impaired clustering of mutant CASQ2.
Catecholaminergic polymorphic ventricular tachycardia (CPVT) is a genetically transmitted disease characterized by stress- or emotion-induced life-threatening arrhythmias occurring in the structurally intact heart. In 2001,1 we demonstrated that mutations in the cardiac ryanodine receptor cause the autosomal dominant form of CPVT and subsequently reported that approximately 50% of CPVT patients carry mutations in RyR2.2 In 2001, Lahat et al3 showed that mutations in the cardiac calsequestrin gene (CASQ2) are responsible for the rare autosomal recessive form of the disease that accounts for approximately 3% of CPVT.4 The identification of the genes underlying CPVT has had implications that extend beyond those impacting clinical management of patients inasmuch as it stimulated fundamental research targeted to understand the links between intracellular calcium regulation and arrhythmogenesis. We recently developed5 a knock-in mouse model carrying the R4496C RyR2 mutation identified in the first genotyped CPVT family and demonstrated that the RyR2R4496C mice develop bidirectional and polymorphic VT similar to those observed in patients. In this model we also demonstrated6 the occurrence of delayed after depolarizations (DADs) induced by adrenergic stimulation in isolated myocytes from the heart of heterozygous mice, suggesting that arrhythmias are elicited by triggered activity. Recently 2 mutants CASQ2 knock-in mice models were developed by Song et al7: the first strain carries the homozygous point mutation discovered by Lahat et al3 in the first recessive CPVT family (D307H), and the second strain carries a homozygous deletion ΔE9/ΔE9; in analogy with RyR2 mice,5 both models develop bidirectional-polymorphic VTs on sympathetic activation. Interestingly, the finding of CASQ2 reduction and calreticulin increase in both strains led to the hypothesis that reduction in CASQ2 activates a compensatory increase in calreticulin.7 Here we describe a novel knock-in mouse carrier of a CASQ2 point mutation at position 33 (R33Q); the characterization of the CASQ2R33Q/R33Q mouse model provides information that shed new light on the complex pathogenesis of recessive CPVT.
Materials and Methods
Detailed methods for mouse generation, electrophysiological measurements, immunofluorescence, real-time PCR, microarray, protein and electron microscopy analysis are reported in the online supplement (available online at http://circres.ahajournals.org).
Generation of Knock-In of R33Q CASQ2 in Mouse Model
The knock-in strain was generated by homologous recombination of the targeting vector with 129Sv/J embryonic stem cells genome. A 450-bp DNA segment encompassing 5′ UTR and exon 1 of the CASQ2 was used to screen RPCI-21 PAC mouse genomic library. Positive clones were sequenced to define the structure of an 8.5-Kb region encompassing promoter, exon 1, and part of intron 1 of mouse CASQ2. This region was cloned in 3 parts and assembled into the targeting vector pFlrt (supplemental Figure I). The linearized targeting vector was electroporated into 129Sv/J embryonic stem cells. The clone selected with G418 and gancyclovir was injected into C57BL/6NCrL blastocyts and transferred to pseudopregnant CD-1 females. Genotype was determined by sequencing of DNA extracted from tail biopsy specimens (DNasy Tissue Kit, Qiagen).
In Vivo Phenotype
In vivo ECG recording was performed using intraperitoneal devices (Data Sciences International). Surgery was performed under general anesthesia (Avertin 0.025 mg/kg) using a heating pad to keep body temperature at 37°C. Arrhythmias were defined according to Cerrone et al,5 transthoracic 2-dimensional, M-mode, and Doppler echocardiography (CASQ2WT/WT n=6; CASQ2R33Q/R33Q n=6) was performed under isoflurane anesthesia with a 15 MHz linear transducer (HP Sonos 5500).
Hearts of 8-week-old mice (CASQ2WT/WT n=2; CASQ2R33Q/WT n=2; CASQ2R33Q/R33Q n=2) were excised, stored in 10% formalin, and serially sectioned. Presence of macroscopic alterations of the heart was assessed by gross inspection; the heart was weighed to determine the heart/body weight ratio. Sections were stained with hematoxylin-eosin and Masson stain and evaluated by light microscopy.
Isolated ventricular myocytes were processed using an established protocol (see online data supplements). Coverslips were incubated with polyclonal anti-CASQ2 (Affinity Bioreagents, PA1-913) and monoclonal anti-α-actinin (Sigma A7811) antibodies. After washing, cells were incubated with secondary antibodies. Confocal microscopy was performed with a TCS-SP digital scanning confocal microscope (Leica).
The hearts of CASQ2WT/WT (n=2) and CASQ2R33Q/R33Q mice (n=2) were fixed by retrograde aortic perfusion (3.5% glutaraldehyde, 0.1 mol/L Na-Cacodylate buffer, pH 7.2). Specimens were processed as described in the online data supplements. Sections were examined with a Philips 410 Microscope (Fei Co.), equipped with a Hamamatsu C4742-95 digital imaging system (Advanced Microscopy Techniques).
The width of the lumen of junctional sarcoplasmic reticulum (jSR) was measured in electron micrographs taken at 135 000× magnification using Adobe Photoshop and a grid of perpendicular lines randomly drawn perpendicularly across the jSR.
The total SR volume was calculated by the well-established stereology point counting technique8 in electron micrographs taken at 24 000× magnification from cross-sections of papillary cardiomyocytes. Multiple pictures were taken for each myocyte (36 and 46 cells in CSQ2WT/WT and CSQ2R33Q/R33Q mice, respectively). The images were covered with an orthogonal arrays of dots at a spacing of 0.17 μm. The ratio of the dots’ number falling over the SR to the total number of dots covering the whole micrograph was used to calculate the percentage of volume occupied by SR.
Patch Clamp Experiments in Isolated Ventricular Myocytes
Isolated cardiac myocytes were isolated using an established enzymatic digestion protocol.9 Transmembrane action potentials and currents were recorded in whole cell configuration as previously described6 using a MultiClamp 700B amplifier (Axon Instruments). Only quiescent, Ca-tolerant, rod-shaped cells with resting potential ≤−80 mV were used. Myocytes were electrically stimulated with 3ms/1.5 to 2.5 nA depolarizing pulses. Data were corrected for liquid junction potential before analysis. Action potential duration (APD) was measured at 90% and 50% of repolarization (APD90 and APD50). Triggered activity was defined as an unstimulated action potential arising from a delayed afterdepolarization (DAD) or a early afterdepolarization (EAD). L-type calcium current (ICa-L) was recorded with 200 ms depolarizing pulses from a holding potential of −40 mV, with 10 mV steps from −40 mV to +60 mV. Current/voltage (I/V) curves were obtained 10 mV voltage steps (−40 mV to +60 mV) from a holding potential of −40 mV. Protocols to assess releasable SR Ca2+ content by integration of the sodium-calcium exchange current (NCX), steady-state inactivation (SSI), and steady state activation (SSA) curves are reported in the online supplements.
Total RNA from the hearts of 8-week-old CASQ2R33Q/R33Q (n=5) and CASQ2WT/WT mice (n=5) was extracted (RNA Easy, Qiagen) and retro-transcribed with random examers (ThermoScript RT-PCR system-Invitrogen). Real Time quantification of targets genes (CASQ2, Triadin, Junctin, RyR2, SERCA2, Phospholamban and Calreticulin—list of primers provided in online supplements) and housekeeping reference transcript (GAPDH) was performed with CYBR, using the ABI PRISM 7000 detection system (Applied Biosystems). Relative gene expression was quantified as follows (from User Bulletin #2 for the ABI PRISM 7000): fold change= 2−Δ(ΔCt) where ΔCt=Cttarget−Ctreference and Δ(ΔCt)=ΔCtsample−ΔCtcontrol Ct is the fractional cycle number at which the fluorescence passes the fixed threshold. Data were analyzed with the comparative threshold cycle (Ct) relative-quantification method. Variance in fold change was calculated from genes mRNAs values (target gene) compared with control mRNAs (WT counterpart) and reference gene.
Experiments were performed on total RNA from hearts of 8-week-old CASQ2R33Q/R33Q (n=5) and CASQ2WT/WT mice (n=5). Labeled cRNA probes were generated from total RNA using the GeneChip IVT Labeling Kit (Affymetrix). The cRNA was fragmented, biotinylated, and hybridized to Mouse Genome 430 2.0 Array chips, containing 14 000 mouse genes. Quality control and boxplot of raw intensities indicated the absence of outliers. Probe level was converted to expression values using both the Robust Multi-array Average (RMA) procedure and the MAS5.0 algorithms. Differentially expressed genes were identified using SAM by computing single-gene statistics and repeated permutations to determine correlation of expression level with the analyzed phenotypes. Supervised analyses was carried out to identify expression signatures of CASQ2R33Q/R33Q samples as compared to CASQ2WT/WT.
Total homogenates and microsomal fractions were obtained using standard methods (online data supplements) from 8-week-old CASQ2WT/WT, CASQ2R33Q/WT, and CASQ2R33Q/R33Q mice. Western blot relied on the following primary antibodies: anti-CASQ (ABR, PA1-913), anti-Triadin (ABR, MA3-927), anti-RyR2 (ABR, MA3-916), anti-Phospholamban (ABR, MA3-922), anti-Calreticulin (Upstate), anti-GAPDH (Chemicon), anti-SERCA2 (Santa Cruz, SC-8094), and antijunctin (kind gift of Dr D.H. Kim, Gwangju Institute of Science and Technology, Korea). Chemioluminescent detection was performed with substrate reagents by Pierce Biotechnology. Densitometric analysis was performed with Image for Windows software (V. Beta 4.0.2; Scion).
Recombinant CASQ2-WT and CASQ2-R33Q proteins, generated and purified as described,10 were digested by trypsin (CASQ2:trypsin=50:1 w/w; Sigma-Aldrich) in 20 mmol/L MOPS pH 7.2, 500 mmol/L KCl at 25°C for 30 minutes. Soyben trypsin inhibitor (inhibitor:trypsin=1:1 w/w; SERVA) halted the reaction and digested samples were electrophoresed on 10 to 17.5% polyacrylamide gradient gels. Slab gels were stained with Coomassie brilliant blue. Densitometric analysis was performed as described above.
Statistical analysis was performed using SPSS v14 package. One-way ANOVA with Bonferroni posthoc analysis or Student t test were used. Western blot data were analyzed by densitometric analysis which retuned arbitrary units of band intensities. Chi square test (Fisher exact test) was used to compare difference of incidence of DADs/triggered activity. Data are expressed as mean±SD unless specified. Values of P<0.05 were considered statistically significant.
Phenotype of CASQ2R33Q/R33Q Mice
No differences were observed in the duration of the pregnancy, delivery, size, and survival of litters among CASQ2WT/WT, CASQ2R33Q/WT, and CASQ2R33Q/R33Q mice. In the 3 groups of mice, (n=10 for each group) no differences were present in weight (CASQ2WT/WT 20.1±1.6 g; CASQ2R33Q/WT 20.5±1.9 g P=0.63 versus CASQ2WT/WT; CASQ2R33Q/R33Q 21.8±2.1 g P=0.06 versus CASQ2WT/WT) and in heart/body weight ratio (CASQ2WT/WT 5.5±0.6 mg/g; CASQ2R33Q/WT 5.6±0.8 mg/g P=0.9 versus CASQ2WT/WT; CASQ2R33Q/R33Q 6.0±1.2 mg/g; P=0.27 versus CASQ2WT/WT).
Echocardiographic evaluation was unremarkable in CASQ2R33Q/R33Q versus CASQ2WT/WT (see supplemental Table I). Likewise, all the ECG parameters were unremarkable.
Continuous ECG recording (n=7 per group, ages 8 weeks) documented the presence of spontaneous ventricular premature beats (VPBs) in 6/7 CASQ2R33Q/R33Q and in none of the CASQ2WT/WT or CASQ2R33Q/WT mice. Exposure to environmental stressors such as noise and physical contact induced both nSVT as sVT episodes in 6/7 CASQ2R33Q/R33Q mice but not in CASQ2WT/WT or CASQ2WT/R33Q mice (Figure 1). Duration of sustained VT ranged from 5 to 232 seconds.
Light Microscopy of Ventricular Tissue and CASQ2 Localization by Immunohistochemistry
Histological examination of cardiac specimens failed to demonstrate gross structural abnormalities (Figure 2A). CASQ2-R33Q and α-actinin colocalized at the Z-line both in homozygous CASQ2R33Q/R33Q and CASQ2WT/WT mice, as judged by immunofluorescence confocal microscopy (Figure 2B). The reduced fluorescence intensity of CASQ2R33Q/R33Q samples is consistent with the decreased amount of CASQ2-R33Q detected by Western blotting (see below).
Electron microscopy of both CASQ2WT/WT and CASQ2R33Q/R33Q hearts showed that the SR forms junctions with T-tubules (dyads) in proximity of the Z line, as expected (Figure 3). The percentage of cell volume occupied by SR was identical: 2.9±0.9 for CSQ2WT/WT and 2.9±1.0 for CSQ2R33Q/R33Q mice (P=0.4; see supplemental Table II). The jSR cisternae of WT samples were narrow and flat (Figure 3A through 3C), whereas they were much wider (41±10 nm versus 26±4 nm; P<0.0001, see supplemental Table II) and had a more variable size in R33Q samples (Figure 3D through 3F).
The jSR of CASQ2WT/WT myocytes contained an electron-dense chain-like line, referable to condensed CASQ2 molecules, that runs parallel to the SR membrane (Figure 3A through 3C, single arrows). and is confined to those junctional areas bearing feet, ie, the cytoplasmic domains of RyR2 (Figure 3A and 3D, multiple arrows)11. On the other hand, the chain-like line of condensed CASQ2 was missing in CASQ2R33Q/R33Q myocytes, and the SR lumen was either empty or contains some electron-dense material that is not clustered as in SR/T-tubule junctions of WT myocytes (Figure 3D through 3F).
Electrophysiological Characterization of Cardiac Myocytes
We compared the action potential duration (APD) of CASQ2WT/WT, CASQ2R33Q/WT, and CASQ2R33Q/R33Q myocytes at 1 or 5 Hz at 35°C. No significant differences were detected in the APD90, APD50, in the amplitude of AP, and in the resting potential among the 3 groups (Table). At 5Hz, 56% (18/32) CASQ2R33Q/R33Q myocytes, but none of the CASQ2WT/WT and CASQ2R33Q/WT, developed DADs (P<0.001) and (9/32) 28% CASQ2R33Q/R33Q myocytes and none of the CASQ2WT/WT and CASQ2R33Q/WT showed triggered activity (TA) (P=0.041). In the presence of epinephrine 200 nmol/L, no triggered activity was induced in CASQ2WT/WT myocytes (n=15) versus 8% CASQ2R33Q/WT myocytes (1/12) (P=0.44, versus CASQ2WT/WT); finally we perfused with 200 nmol/L epinephrine CASQ2R33Q/R33Q myocytes (n=17) that failed to develop triggered activity in unstimulated settings and observed that 47% of such CASQ2R33Q/R33Q myocytes (8/17) developed triggered activity at 5 Hz (P=0.003, versus CASQ2WT/WT; P=0.043, versus CASQ2R33Q/WT). Interestingly, when paced at 1 Hz in the presence of epinephrine, 23.5% of CASQ2R33Q/R33Q myocytes (4/17) developed EADs at a take off potential between −50 mV and −60 mV (Figure 4).
No significant differences were observed in the ICa-L current between CASQ2WT/WT and CASQ2R33Q/R33Q myocytes either in absence or in the presence of epinephrine. The time course of ICa-L decay was similar in the 2 groups. In the presence of 200 nmol/L epinephrine, SSA curves significantly shifted to left in both WT and R33Q cells with no difference between the 2 groups [V1/2: from −16.6 mV to −21.2 mV in CASQ2WT/WT myocytes (P=0.044) and from −15.3 mV to −21.6 mV in CASQ2R33Q/R33Q myocytes (P<0.001). SSI curves did not shift significantly in the 2 groups leading to the conclusion that there was no significant difference of ICa-L window current between CASQ2WT/WT and CASQ2R33Q/R33Q myocytes in the presence of epinephrine. We assessed the total SR Ca2+ content of CASQ2WT/WT and CASQ2R33Q/R33Q myocytes from the integral of INCX evoked by the application of caffeine and demonstrated that the SR Ca2+ content is decreased by 35% in CASQ2R33Q/R33Q myocytes as compared with CASQ2WT/WT suggesting that the SR ability to store Ca2+ was significantly suppressed in CASQ2R33Q/R33Q myocytes (Figure 5). There was no significant difference of the τ of INCX between CASQ2WT/WT and CASQ2R33Q/R33Q (806±40ms versus 760±66ms, P=0.57), suggesting no increase of NCX function in CASQ2R33Q/R33Q myocytes.
mRNA and Protein Levels in CASQ2R33Q/R33Q, CASQ2R33Q/WT, and CASQ2WT/WT Mice
Western blot of total heart homogenates and of microsomal fractions extracted from the heart of 8-week-old CASQ2WT/WT, CASQ2R33Q/WT, and CASQ2R33Q/R33Q mice revealed a dramatic decrease in the levels of CASQ2 protein (< 50% of normal) in CASQ2R33Q/R33Q mice (Figure 6A; ANOVA P<0.001). Real-time PCR failed to demonstrate changes in the level of mRNA CASQ2R33Q/R33Q as compared to CASQ2WT/WT. CASQ2-R33Q reduction could be accounted for by increased proteolysis: in vitro trypsin digestion of WT-CASQ2 and R33Q-CASQ2 was performed to test this hypothesis (Figure 6B). Densitometric scans showed that about 75% of CASQ2-R33Q was digested versus 23% of CASQ2-WT, ie, mutant CASQ2 is more prone to trypsin digestion than CASQ2-WT (compare lane D with lane B).
Protein levels of RyR2, junctin, triadin, calreticulin, SERCA, and phospholamban were compared in CASQ2R33Q/R33Q versus CASQ2R33Q/WT and CASQ2WT/WT hearts. Densitometric scans showed a 25% reduction of the calsequestrin-binding protein triadin in microsomal fractions from CASQ2R33Q/R33Q hearts as compared to CASQ2WT/WT hearts and a reduction of 70% of junctin. No changes were observed in the protein levels of RyR2, SERCA, calreticulin, and phospholamban in CASQ2R33Q/R33Q hearts (Figure 6C). mRNA levels for RyR2, triadin, junctin, calreticulin, SERCA, and phospholamban extracted from CASQ2R33Q/R33Q mice (n=5) were identical to those observed in CASQ2WT/WT mice (n=5), as judged by real-time PCR. Similarly, we failed to detect differences in mRNA levels in 14 000 well characterized mouse genes in CASQ2R33Q/R33Q hearts (n=5) versus CASQ2WT/WT (n=5), using gene expression profiling microarrays (data not shown).
CPVT is an inherited arrhythmogenic disorder characterized by syncopal events and sudden cardiac death, elicited by physical and emotional stress, that manifests in the pediatric age.2 Mutations in the cardiac ryanodine receptor (RyR2) and in the calsequestrin (CASQ2) genes are responsible for the autosomal dominant and recessive variants of CPVT, respectively. We recently reported the characterization of a knock-in mouse model carrier of the RyR2R4496C mutation identified in the Italian family that led to the discovery of the genetic basis of autosomal dominant CPVT.1 Remarkably the knock-in mice have a phenotype that closely resembles clinical manifestations of CPVT patients.5 The extensive phenotypic characterization of the model showed that delayed after-depolarization and triggered activity represent the pivotal arrhythmogenic mechanism of CPVT.6,12
In the present study, we report the characterization of a knock-in mouse homozygous carrier of the CASQ2R33Q mutation that we identified in highly symptomatic CPVT patients13: the main objective of the present study is to derive information germane to the understanding of the pathophysiology of the autosomal recessive form of CPVT that remains controversial.7,14
Morphological Abnormalities and Contractile Function
CPVT is classified as an inherited arrhythmogenic disorder that occurs in individuals with a structurally intact heart: so far, in fact, no functional or morphological cardiac abnormalities have been described in patients with autosomal recessive CPVT. The murine models of recessive CPVT present modest structural abnormalities: similarly to our CASQ2R33Q/R33Q mice, the CASQ2 knock out mice present normal cardiac contractility. However, ultrastructural assessment of CASQ2R33Q/R33Q mice showed that the total SR volume is unchanged, whereas the width of the jSR is doubled. It remains to be clarified whether such features might represent the initial stages of a cardiomiopathy or are mere adaptive phenomena. In addition, CASQ2-R33Q appears not be properly clustered and this might also be related to the inability of R33Q to form Ca-dependent polymers15 at physiological intra-SR [Ca2+].
In Vivo and In Vitro Electrophysiological Profile
In analogy with what observed in the RyR2 knock-in mouse model,5,6 the CASQ2R33Q/R33Q mice develop polymorphic and bidirectional VT remarkably similar to arrhythmias developed by patients. At variance with the animal model of autosomal dominant CPVT,5 however, CASQ2R33Q/R33Q mice develop sustained VTs on exposure to environmental stress, ie, in the absence of any pharmacological challenge. In analogy with RyR2R4496C/WT mice, cardiac myocytes isolated from the heart of CASQ2R33Q/R33Q mice develop DADs and triggered activity when paced at faster rates in the presence of adrenergic stimulation. Interestingly at slow heart rate (1Hz), R33Q myocytes develop EADs that were never observed in RyR2R4496C/WT myocytes (N Liu. unpublished data). It has been suggested that the occurrence of EADs involves reactivation of the L-type calcium current in the setting of action potential prolongation16. Because we did not observe APD prolongation in CASQ2R33Q/R33Q myocytes (Table), a reactivation of inward calcium current seems an unlikely mechanism for EADs. Interestingly, EADs elicited in R33Q myocytes showed a take off potential between −50 mV and −60 mV that corresponds to the potential associated with the peak amplitude of the transient inward current in mouse myocytes (data not shown).6 We therefore speculate that, in analogy with what we previously observed17 in canine myocytes exposed to adrenergic stimulation, EADs in R33Q cardiac myocytes may be generated by the transient inward current elicited by abnormal SR calcium leak. However, because EADs were observed only at low pacing rates, it is unclear whether they have any physiological role in triggering ventricular tachyarrhythmias during adrenergic activation in vivo.
SR Calcium Content and Calsequestrin Depletion
To further investigate the mechanisms leading to arrhythmias in isolated R33Q cardiac myocytes, we measured the integral of INCX evoked by the application of caffeine18,19 and showed a substantial decrease as compared to what observed in WT cells. There is an apparent discrepancy between these data and those reported by Terentyev et al13 showing that in rat isolated myocytes overexpressing CASQ2-R33Q on top of endogenous CASQ2-WT, INCX is unchanged despite the reduction of SR free calcium. Terentyev et al13 linked the observed decrease in free calcium to the impairment of CASQ2 to act as the intraluminal calcium sensor of the RyR2 macromolecular complex20 and to the impairment of the physiological deactivation of the RyR2 at low intraluminal calcium levels. Furthermore, based on single channel analysis,21 we have shown that R33Q failed to inactivate RyR2 as intra-SR [Ca2+] decreased. Thus, despite the leaky RyR2, normal calcium transients were recorded because of the compensatory effect of the high calcium binding provided by overexpressed CASQ2-R33Q and endogenous CASQ2-WT.
Because CASQ2 levels are markedly reduced in CASQ2R33Q/R33Q mice (Figure 6), the limitations of the model used by Terentyev et al,13 ie, CASQ2-R33Q overexpression in the presence of endogenous CASQ2-WT, have now become clear. Therefore, based on the present data, we propose that 2 consequences prompted by the mutation concur to the pathogenesis of the homozygous CASQ2R33Q/R33Q phenotype: (1) reduction of CASQ2 levels (likely attributable to post/translational regulation leading to increased protein degradation, see below) and (2) impairment of RyR2 deactivation, as shown by Terentyev et al13 and Qin et al.21
Despite a substantial reduction of CASQ2 seems to be central in recessive CPVT, it is still a matter of debate whether a threshold decrease of CASQ2 is required for eliciting the CPVT phenotype or whether even a minor loss of CASQ2 increases propensity to arrhythmias. At variance with data reported in the heterozygous knock out mouse model,22 most human heterozygous individuals carriers of premature truncations of CASQ2 are asymptomatic. We reported that all 9 carriers of the G112+5X mutation leading to a premature stop codon had no clinical signs of CPVT. Postma et al23 reported that 1 of 3 carriers of the heterozygous R33X mutation had arrhythmias, yet the clinical description of this individual is insufficient to support a causal link between the heterozygous mutation and the arrhythmias. It is possible that compensatory mechanisms to haploinsufficency of CASQ2 are different in humans and mice, and human beings may have the ability to upregulate the expression of the wild-type allele thus attenuating the consequences of CASQ2 deficiency in the heterozygous state. In the presence of heterozygous missense mutations, the residual function of the 50% mutant CASQ2 may be adequate to prevent functional impairment, as seen in our CASQ2R33Q/WT mice that do not display reduction of CASQ2 levels and in analogy to the asymptomatic parents of the R33Q-linked CPVT patient.
Mechanisms Leading to Reduced CASQ2 in Recessive CPVT and Compensatory Responses
Recently Song et al7 showed that the CASQ2 reduction observed in CASQ2D307H/D307H mice is not paralleled by a reduction of CASQ2 mRNA, yet they did not provide any explanation to account for the protein reduction. In the CASQ2R33Q/R33Q model, we observed a CASQ2 reduction in the presence of normal levels of mRNA and reasoned that CASQ2-R33Q might be more susceptible to proteolysis: a trypsin digestion test supported the view that accelerated CASQ2 catabolism is likely to underlie the observed CASQ2 reduction. Exposure of new proteolytic sites nearby the N or C terminus of the CASQ2-R33Q protein might be taken as indication of misfolding and protein instability, as compared to CASQ2-WT. CASQ2-R33Q because of its higher conformational flexibility, as recently shown,24 might undergo accelerated in vivo degradation via proteasome, thus reducing effective CASQ2 content within the SR lumen. The relationships linking missense mutation, protein instability, and activation of posttranslational regulatory mechanisms are currently under investigation. A complementary mechanism for CASQ2-R33Q reduction might be attributable to inappropriate CASQ2 polymerization leading to defective SR retention, as suggested by Cala and coworkers.25
In the CASQ2D307H/D307H and CASQ2ΔΕ9/ΔΕ9 models, Song et al7 found an increase in calreticulin and RyR2, and thus considered this process as a compensatory response to reduced CASQ2. In our model, we failed to observe such an adaptive response, and thus we argue that the increase in calreticulin and RyR2 might be a mutation-specific finding and should not be regarded as the univocal response to CASQ2 decrease. In our model, we found, instead, a decrease in the levels of triadin and junctin, in the absence of mRNA reduction, as assessed by real-time PCR: interestingly, also the knock out CASQ2 model14 showed a similar reduction of triadin and junctin, yet the mechanism underlining this response remains so far undefined.
To provide a comprehensive assessment of the heart adaptive response to CASQ2-R33Q, we performed microarray studies to define the gene expression profiling of the heart of CASQ2R33Q/R33Q mice versus that of their CASQ2WT/WT littermates. Unexpectedly, no changes in mRNA levels were identified in any of the 14 000 genes probed suggesting that the heart does not activate adaptive transcriptional responses to mutant R33Q CASQ2, at least by 8 weeks of age.
In light of the unexpected13 finding of the central role of CASQ2 decrease in the R33Q knock in mouse model, we now propose that CASQ2 reduction is one of the common pathogenetic mechanisms of autosomal recessive CPVT, as supported by the evidence that the CASQ2 knock out mice is a CPVT phenocopy.14 We also suggest that the different missense mutations present in CPVT patients may be characterized by additional and specific functional abnormalities of CASQ2 (such as, CASQ2 polymerization impairment, altered calcium affinity, altered interaction with RyR2) leading to variable compensatory responses (increase in RyR2, increase in calreticulin, decrease in triadin and junctin … ). The distinguishing features of each mutation may exert an additive effect on the common background of CASQ2 reduction that is likely to modulate the clinical expressivity of the disease.
We acknowledge the support of the San Raffaele-Telethon Core Facility for Conditional Mutagenesis for embryonic stem cell electroporation and blastocyst injection.
Sources of Funding
This work was supported by Telethon grants No. GGP04066 and GGP06007 and by funds from the Ministero dell'Università e della Ricerca Scientifica e Tecnologica: FIRB RBNE01XMP4_006, RBLA035A4X_002, PRIN 2006055828_002.
↵*These authors contributed equally to this study.
Original received February 1, 2008; revision received May 31, 2008; accepted June 16, 2008.
Priori SG, Napolitano C, Tiso N, Memmi M, Vignati G, Bloise R, Sorrentino V, Danieli GA. Mutations in the cardiac ryanodine receptor gene (hryr2) underlie catecholaminergic polymorphic ventricular tachycardia. Circulation. 2001; 103: 196–200.
Priori SG, Napolitano C, Memmi M, Colombi B, Drago F, Gasparini M, DeSimone L, Coltorti F, Bloise R, Keegan R, Cruz Filho FE, Vignati G, Benatar A, DeLogu A. Clinical and molecular characterization of patients with catecholaminergic polymorphic ventricular tachycardia. Circulation. 2002; 106: 69–74.
Lahat H, Pras E, Olender T, Avidan N, Ben Asher E, Man O, Levy-Nissenbaum E, Khoury A, Lorber A, Goldman B, Lancet D, Eldar M. A missense mutation in a highly conserved region of CASQ2 is associated with autosomal recessive catecholamine-induced polymorphic ventricular tachycardia in Bedouin families from Israel. Am J Hum Genet. 2001; 69: 1378–1384.
Rossenbacker T, Bloise R, De Giuli L, Raytcheva-Buono EV, Theilade J, Keegan R, Drago F, Gasparini M, Napolitano C, Priori SG. Catecholaminergic polymorphic ventricular tachycardia: genetics, natural history and response to therapy. Circulation. 2007; 116: –179. Abstract.
Cerrone M, Colombi B, Santoro M, Raffale di Barletta M, Scelsi M, Villani L, Napolitano C, Priori SG. Bidirectional ventricular tachycardia and fibrillation elicited in a knock-in mouse model carrier of a mutation in the cardiac ryanodine receptor (RyR2). Circ Res. 2005; 96: e77–e82.
Liu N, Colombi B, Memmi M, Zissimopoulos S, Rizzi N, Negri S, Imbriani M, Napolitano C, Lai FA, Priori SG. Arrhythmogenesis in catecholaminergic polymorphic ventricular tachycardia. insights from a RyR2 R4496C knock-in mouse model. Circ Res. 2006; 99: 292–298.
Song L, Alcalai R, Arad M, Wolf CM, Toka O, Conner DA, Berul CI, Eldar M, Seidman CE, Seidman JG. Calsequestrin 2 (CASQ2) mutations increase expression of calreticulin and ryanodine receptors, causing catecholaminergic polymorphic ventricular tachycardia. J Clin Invest. 2007; 117: 1814–1823.
Mobley BA, Eisenberg BR. Sizes of components in frog skeletal muscle measured by methods of stereology. J Gen Physiol. 1975; 66: 31–45.
Raffaele di Barletta M, Viatchenko-Karpinski S, Nori A, Memmi M, Terentyev D, Turcato F, Valle G, Rizzi N, Napolitano C, Gyorke S, Volpe P, Priori SG. Clinical phenotype and functional characterization of CASQ2 mutations associated with Catecholaminergic Polymorphic Ventricular Tachycardia. Circulation. 2006; 114: 1012–1019.
Cerrone M, Noujaim SF, Tolkacheva EG, Talkachou A, O'connell R, Berenfeld O, Anumonwo J, Pandit SV, Vikstrom K, Napolitano C, Priori SG, Jalife J. Arrhythmogenic mechanisms in a mouse model of catecholaminergic polymorphic ventricular tachycardia. Circ Res. 2007; 11: 1039–1048.
Terentyev D, Nori A, Santoro M, Viatchenko-Karpinski S, Kubalova Z, Gyorke I, Terentyeva R, Vedamoorthyrao S, Blom NA, Valle G, Napolitano C, Williams SC, Volpe P, Priori SG, Gyorke S. Abnormal interactions of calsequestrin with the Ryanodine receptor calcium release channel complex linked to exercise-induced sudden cardiac death. Circ Res. 2006; 98: 1151–1158.
Knollmann BC, Chopra N, Hlaing T, Akin B, Yang T, Ettensohn K, Knollmann BE, Horton KD, Weissman NJ, Holinstat I, Zhang W, Roden DM, Jones LR, Franzini-Armstrong C, Pfeifer K. Casq2 deletion causes sarcoplasmic reticulum volume increase, premature Ca2+ release, and catecholaminergic polymorphic ventricular tachycardia. J Clin Invest. 2006; 116: 2510–2520.
January CT, Riddle JM. Early afterdepolarizations: mechanism of induction and block. A role for L-type Ca2+ current. Circ Res. 1989; 64: 977–990.
Qin J, Valle G, Nani A, Nori A, Rizzi N, Priori SG, Volpe P, Fill M. Luminal Ca2+ regulation of single cardiac ryanodine receptors: insights provided by calsequestrin and its mutants. J Gen Physiol. 2008; 131: 325–334.
Chopra N, Kannankeril PJ, Yang T, Hlaing T, Holinstat I, Ettensohn K, Pfeifer K, Akin B, Jones LR, Franzini-Armstrong C, Knollmann BC. Modest reductions of cardiac calsequestrin increase sarcoplasmic reticulum Ca2+ leak independent of luminal Ca2+ and trigger ventricular arrhythmias in mice. Circ Res. 2007; 101: 617–626.
Valle, G, Galla, D, Nori, A, Priori, SG, Gyorke, S, De Filippis, V and Volpe, P. Catecholaminergic polymorphic ventricular tachycardia-related mutations R33Q and L167H alter calcium sensitivity of human cardiac calsequestrin. Biochem J. 2008; 413: 291–303.