Calreticulin (crt) is an ubiquitously expressed and multifunctional Ca2+-binding protein that regulates diverse vital cell functions, including Ca2+ storage in the ER and protein folding. Calreticulin deficiency in mice is lethal in utero due to defects in heart development and function. Herein, we used crt−/− embryonic stem (ES) cells differentiated in vitro into cardiac cells to investigate the molecular mechanisms underlying heart failure of knockout embryos. After 8 d of differentiation, beating areas were prominent in ES-derived wild-type (wt) embryoid bodies (EBs), but not in ES-derived crt−/− EBs, despite normal expression levels of cardiac transcription factors. Crt−/− EBs exhibited a severe decrease in expression and a lack of phosphorylation of ventricular myosin light chain 2 (MLC2v), resulting in an impaired organization of myofibrils. Crt−/− phenotype could be recreated in wt cells by chelating extracellular or cytoplasmic Ca2+ with EGTA or BAPTA, or by inhibiting Ca2+/calmodulin-dependent kinases (CaMKs). An imposed ionomycin-triggered cystolic-free Ca2+ concentration ([Ca2+]c) elevation restored the expression, phosphorylation, and insertion of MLC2v into sarcomeric structures and in turn the myofibrillogenesis. The transcription factor myocyte enhancer factor C2 failed to accumulate into nuclei of crt−/− cardiac cells in the absence of ionomycin-triggered [Ca2+]c increase. We conclude that the absence of calreticulin interferes with myofibril formation. Most importantly, calreticulin deficiency revealed the importance of a Ca2+-dependent checkpoint critical for early events during cardiac myofibrillogenesis.
Cardiogenesis is one of the earliest processes during vertebrate embryogenesis (Olson and Srivastava, 1996; Lin et al., 1997b; Black and Olson, 1998; Charron and Nemer, 1999; Harvey, 1999; Harvey and Rosental, 1999; Schwartz and Olson, 1999; Frey et al., 2000; Srivastava and Olson, 2000). Although intracellular Ca2+ is a critical second messenger in the heart, regulating diverse functions from gene expression to cell contraction and relaxation (Pozzan et al., 1994; Clapham, 1995), the potential contribution of Ca2+ signaling during early stages of cardiac development remains largely unexplored.
Within the ER, the Ca2+ reservoir of nonmuscle cells, Ca2+ is mainly bound to calreticulin (Michalak et al., 1999). Calreticulin is a highly conserved, 46-kD protein that modulates many cellular functions, including Ca2+ homeostasis (Liu et al., 1994; Bastianutto et al., 1995; Camacho and Lechleiter, 1995; Mery et al., 1996; Coppolino et al., 1997; Fasolato et al., 1998; John et al., 1998), protein folding (Helenius et al., 1997; Krause and Michalak, 1997; Ellgaard and Helenius, 2001), integrin-dependent cell adherence (Coppolino et al., 1995; Opas et al., 1996; Coppolino et al., 1997), and steroid-dependent gene expression (Burns et al., 1994; Dedhar et al., 1994).
Calreticulin is essential for proper cardiac development because knockout of the calreticulin gene is embryonically lethal due to impaired cardiac development (Mesaeli et al., 1999; Rauch et al., 2000). Calreticulin-null mouse embryos show a marked decrease in ventricular wall thickness and deep intertrabecular recesses in the ventricular walls (Mesaeli et al., 1999). Importantly, at early stages of cardiac development, calreticulin is highly expressed in differentiating cardiomyocytes (Mesaeli et al., 1999). It is subsequently downregulated in the mature heart (Mesaeli et al., 1999) because elevated expression of calreticulin in newborn hearts of transgenic mice leads to complete heart block and sudden death (Nakamura et al., 2001a).
Mechanisms responsible for impaired cardiac development in calreticulin-deficient (crt−/−) mice are not well understood. Here we used calreticulin-deficient crt−/− embryonic stem (ES)* cells to investigate the role of the protein in early events of cardiac differentiation.
We show that calreticulin-deficient cardiomyocytes derived from ES cells had an impaired myofibrillogenesis due to decreased ventricular myosin light chain 2 (MLC2v) expression and phosphorylation. Nuclear translocation of myocyte enhancer factor C2 (MEF2C), a transcription factor responsible for activation of several cardiac specific embryonic genes, including the MLC2v, was inhibited in crt−/− cardiomyocytes. Furthermore, calreticulin-deficient myocytes featured disorganized myofilaments and inhibited contractile activity. Most importantly, this phenotype was reversed by transient elevation of cystolic-free calcium ([Ca2+]c) with a Ca2+ ionophore. This reveals a Ca2+-dependent checkpoint in early events associated with cardiac myofibrillogenesis.
Phenotype of calreticulin-deficient embryoid bodies
Fig. 1 shows that crt−/− embryoid bodies (EBs) cultured in suspension between day 2 (D2) and D6 had a significantly larger size than wild-type (wt) EBs (669 ± 60 μm, vs. 359 ± 26 μm, respectively), as measured at D6 (Fig. 1 A, left). Despite the size difference, both cell types had similar protein content as measured for 20 pooled EBs (4.66 ± 0.55 and 4.82 ± 1.06 μg/μl in wt and crt−/− EBs, respectively; n = 3; Fig. 1 A, right). In addition, no difference in the growing capacity (mitotic rate) of cultured wt versus crt−/− undifferentiated ES cells was observed (unpublished data).
Removal of leukemia inhibitory factor (LIF) from the culture medium induces ES cell differentiation (Maltsev et al., 1994) into EBs containing clusters of contracting cardioblasts after 7 d of culture (Meyer et al., 2000). Therefore, we analyzed for the presence of the contractile activity and the beating frequency of cardiomyocytes derived from wt and crt−/− ES cells at D8 of culture. Approximately 70% of the adherent wt EBs were strongly contracting and contained more than two beating areas, each beating with a frequency of 72 ± 8 contractions/min. In contrast, 90% of crt−/− EBs did not feature any contracting foci (Fig. 1 B). Wild-type EBs loaded with the Ca2+-indicator dye fluo3 exhibited normal [Ca2+]c oscillations (Fig. 1 C, left) (Bony et al., 2001), whereas crt−/− EBs did not show any measurable [Ca2+]c spikes (Fig. 1 C, right).
The myofibrillogenesis was evaluated in 8-d-old EBs by in situ immunostaining of sarcomeric proteins α-actinin and MLC2v. As expected (Meyer et al., 2000), in wt ES-derived cardiomyocytes, α-actinin and MLC2v appeared organized as an interconnected network with the myofibrils stretching throughout the cytoplasm even by running from one cell into another (Fig. 2 A, a and b). These newly formed sarcomeres featured a length of 2 μm, similar to that of adult cardiac sarcomeres, indicating that the contractile apparatus was fully organized at this stage of EBs development. In contrast, crt−/− cardiomyocytes showed an anomalous distribution of both α-actinin and MLC2v (Fig. 2 A, c and d). α-actinin was not incorporated into the Z-disks but remained localized in cytosolic spots. MLC2v also failed to form and localize to the A bands. It was distributed within the cytosol as short disorganized filaments. In addition, a severe reduction in MLC2v mRNA and protein was also observed as revealed by reverse transcription–PCR (unpublished data) and Western blot analysis (Fig. 2 B), respectively. This indicated that calreticulin deficiency interferes with myofibril formation.
Effect of extracellular Ca2+ chelation, intracellular [Ca2+]c buffering, and inhibition of calmodulin-dependent protein kinases on ES cell differentiation
Calreticulin-deficient mouse embryonic fibroblasts (MEFs) have impaired Ca2+ release from ER (Nakamura et al., 2001b). To test whether an altered Ca2+ homeostasis in calreticulin-deficient cardiomyocytes was responsible for an abnormal myofibrillogenesis and impaired contractility, we modulated extracellular or intracellular [Ca2+]c. Starting at D4 of differentiation, we either removed extracellular Ca2+ from the medium with EGTA (2 mM) or chelated [Ca2+]c with BAPTA/AM (25 μM) during differentiation of wt EBs. D4 was chosen because at this time, although cardiac transcription factors are fully detectable by reverse transcription–PCR, no genes encoding constitutive cardiac proteins are yet expressed. Removal of extracellular Ca2+ or chelation of [Ca2+]c led to a dramatic reduction in beating wt EBs observed at D8 (Fig. 3 A), and impaired myofibrillogenesis reminiscent of the calreticulin-deficient phenotype of EBs (Fig 2). Concomitantly, there was a reduction of MLC2v protein (Fig. 3 B).
The activation of Ca2+/calmodulin-dependent protein kinases (CaMKs) is essential for myofibrillogenesis (Zhang et al., 2002) and sarcomere organization (Aoki et al., 2000). To test the involvement of CaMKs in EGTA or BAPTA-dependent inhibition of myofibrillogenesis we used KN93, an inhibitor of CaMKII (Sumi et al., 1991). wt EBs treated with KN93 from D4 to D8 had impaired distribution of both α-actinin and MLC2v, similar to the EGTA- or BAPTA-treated cells. Most importantly, they exhibited a phenotype similar to crt−/− cardiomyocytes. MLC2v was spread within the cytosol as short disorganized filaments (Fig. 3 C, d, f, and h vs. b for control), whereas α-actinin showed a spotted distribution (Fig. 3 C, c, e, and g). We concluded that elevation of [Ca2+] and activation of CaMK are essential for early stages of myofibrillogenesis, and that these processes are impaired in the absence of calreticulin.
Ca2+-dependent rescue of the myofibrillogenesis in calreticulin-deficient cardiomyocytes
The phenotype of wt cells differentiated in the absence of extracellular Ca2+ or under Ca2+ buffering conditions was reminiscent of the phenotype of crt−/− cells. Therefore, we tested whether a transient elevation of [Ca2+]c prior to myofibrillogenesis (i.e., appearance of recognizable cardiac cells) would initiate myofibril formation and restore the beating activity in the calreticulin-deficient EBs.
To elicit a transient increase in [Ca2+]c, we incubated crt−/− EBs at D4 with the Ca2+ ionophore ionomycin (10–1,000 nM) for 2 h, as indicated in Fig. 4 A. Only a 2-h treatment (followed by a washout) was used to exclude any potential toxic effect of ionomycin. Western blot analysis revealed no significant change in the level of GRP94, a heat-shock ER chaperone of the HSP90 family normally upregulated after prolonged exposure to increased intracellular Ca2+ (Mitani et al., 1996) (unpublished data). This confirmed that this exposure of crt−/− cells to Ca2+ ionophore did not induce notable cell stress. Furthermore, we did not detect any cell death after ionomycin treatment, as determined by cell viability using trypan blue exclusion (3.8 ± 0.2% positive cells in control EBs vs. 3.4 ± 0.3% in ionomycin-treated EBs) and by extracellular release of cytosolic lactate dehydrogenase (1.5 ± 0.1% from control EBs vs. 2.0 ± 0.2% from ionomycin-treated EBs; n = 3).
Fig. 4 B shows that addition of 100 nM ionomycin to wt and crt−/− EBs induced a transient elevation in [Ca2+]c in both types of EBs. After washout, [Ca2+] returned to basal level within 5 h.
Crt−/− ES-derived EBs were incubated with 10–100 nM ionomycin, at D4 (Fig. 4, C and D) or D7 (Fig. 4 D), followed by washout and analysis of their beating activity 4 d after addition of the ionophore. Fig. 4 C shows that crt−/− EBs treated with ionomycin had fully restored beating activity as rapidly as 4 d after ionomycin treatment. Partial recovery of EBs (20% beating EBs) was already observed the first day after incubation with ionomycin (unpublished data).
Next, we tested MLC2v expression in crt−/− cardiomyocytes within ionomycin-treated EBs. There was a correct insertion of both α-actinin and MLC2v into sarcomeric structures (Fig. 5 A), indicating that myofibrillogenesis was restored in crt−/− cells as a result of ionomycin treatment. Importantly, crt−/− EBs treated with ionomycin expressed high levels of MLC2v protein as observed for wt (Fig. 5 B).
Phosphorylation of MLC2v by the Ca2+/calmodulin-dependent myosin light chain kinase (MLCK) is a prerequisite for its insertion into the A-band of the sarcomere (Sanbe et al., 1999). The phosphorylated and unphosphorylated forms of MLC2v can be readily identified by two-dimensional gel electrophoresis and followed by Western blot analysis with specific antibodies. As expected, wt EBs contained both phosphorylated and unphosphorylated forms of MLC2v. In contrast, in crt−/− cardiomyocytes, there was no detectable phosphorylated MLC2v (Fig. 5 C, middle). Most importantly, treatment of crt−/− EBs with ionomycin resulted in appearance of phosphorylated MLC2v (Fig. 5 C, bottom). This indicated that crt−/− cardiomyocytes feature impaired Ca2+-dependent phosphorylation of MLC2v.
Interestingly, the content of atrial MLC2 (MLC2a) was not significantly affected by treating wt or crt−/− EBs with ionomycin (Fig. 5 D), and it was correctly inserted into sarcomeres in crt−/− cardiomyocytes (Fig. 5 E). We concluded that crt−/− EBs are deficient in a specific Ca2+ signal (“checkpoint”) required for elevated expression of MLC2v (but not MLC2a), its phosphorylation, and assembly into functional sarcomeres.
Nuclear translocation of MEF2C is impaired in calreticulin-deficient cardiomyocytes
To investigate mechanisms responsible for downregulation of MLC2v in the absence of calreticulin, we measured expression and nuclear targeting of MEF2C, GATA4, and Nkx2.5, a group of cardiac transcription factors known to regulate the MLC2v gene expression. MEF2C, GATA4, and Nkx2.5 are detected as early as D4 to D5 of EB culture (Meyer et al., 2000). Real-time quantitative PCR analysis revealed no differences in the expression of all three transcription factors at D8 in wt and crt−/− EBs (Fig. 6 A). Ionomycin treatment had no significant effect on expression of MEF2C, GATA4, and Nkx2.5 in wt cells (Fig. 6 A); however, Western blot analysis revealed an increase in MEF2C levels in ionomycin-treated crt−/− EBs (Fig. 6 B). In contrast, GATA4 and NKx2.5 mRNA were decreased to 40 and 50%, respectively, in ionomycin-treated crt−/− EBs (Fig. 6 A). The increased level of MEF2C in ionomycin-treated cells was observed at the protein level in both wt and crt−/− EBs, whereas we did not detect any significant difference of MEF2C content in wt versus crt−/− under control conditions (Fig. 6 B).
Next, we investigated the localization of MEF2C, which regulates the activity of MLC2v promoter (Zou and Chien, 1995). Fig. 7 illustrates the intranuclear distribution of MEF2C within EBs. Distribution of nuclei within EBs was identified by Hoechst staining (Fig. 7, inset). In wt EBs, nuclei were strongly stained by the anti-MEF2C antibody, as visualized by the intranuclear three-dimensional distribution of MEF2C using the shadow projection mode (Fig. 7 A, a), as well as by the summed fluorescence across the nuclei (Fig. 7 A, a′). In contrast, MEF2C failed to accumulate within the nuclei of crt−/− EBs and remained perinuclear (Fig. 7 A, b–b′). Importantly, the ionomycin-induced [Ca2+]c elevation resulted in the restoration of MEF2C translocation into nuclei of crt−/− EBs (Fig. 7 A, d–d′), as well as in an increased accumulation of MEF2C into nuclei of wt cells (Fig. 7 A, c–c′). The impaired nuclear transport was specific to MEF2C because the intracellular distribution of another cardiac transcription factor, GATA4, was not affected. Fig. 7 C shows that GATA4 was localized to the nuclei in both wt and crt−/− cells, indicating that the nuclear import of this transcription factor was not affected by the absence of calreticulin. This indicates that Ca2+-dependent processes are involved in the specific translocation of MEF2C to the nucleus of wt and crt−/− cells.
Next, we tested for a role of CaMKs in nuclear translocation of MEF2C. wt EBs were treated with KN93 from D4 to D8 to inhibit CaMK. In KN93-treated EBs, MEF2C was not targeted into the nucleus, resulting in the absence of contractile foci in these 8-d-old EBs (unpublished data; Fig. 3 A). KN92 was used as a negative control, and as expected, it did not affect MEF2C nuclear targeting or EB contractility.
The generation of mice deficient in calreticulin has uncovered a critical role of the ER Ca2+-binding protein (Mesaeli et al., 1999). Calreticulin-deficient mice display defects in heart development and function due to reduced thickness of the ventricular wall and impaired trabeculation (Mesaeli et al., 1999). Here we used calreticulin-deficient ES cells to investigate the molecular mechanisms responsible for the heart failure of crt−/− mice. We demonstrated that during the differentiation process, crt−/− ES-derived cardiomyocytes feature a severe disruption of myofibrillogenesis due to insufficient expression and Ca2+-dependent phosphorylation of MLC2v. This is likely due to insufficient availability and mobilization of Ca2+ from ER in crt−/− cells as manifested by inhibited CaMK-dependent nuclear translocation of MEF2C. The importance of Ca2+ in these processes is documented by rescue of crt−/− phenotype with Ca2+ ionophore. We show that differentiation proceeds normally at the very early stages of crt−/− EBs formation, but that they fail to undergo normal myofibril formation in the absence of calreticulin and a specific ER-dependent Ca2+ signal. We propose that a specific Ca2+-dependent checkpoint is required for myofibrillogenesis. This signal is not generated in the absence of calreticulin.
Early cardiogenesis is regulated by three families of transcription factors (i.e., Nkx2.5, MEF2, GATA) (Srivastava and Olson, 2000). Among these factors, MEF2C is an essential regulator of ventricular development (Black and Olson, 1998). The targeted disruption of the MEF2C gene in the mouse results in embryonic lethality due to cardiac developmental arrest and severe downregulation of a number of cardiac markers (Durocher et al., 1996; Lin et al., 1997a; Charron and Nemer, 1999) including MLC2v (Liu et al., 2001). MEF2C is a major cardiac transcription factor regulating the activity of MLC2v promoter (Chen et al., 1998; Nguyen-Tran et al., 1999). Therefore, it is not surprising that impaired Ca2+-dependent nuclear translocation of MEF2C in crt−/− cells results in low expression of MLC2v, and consequently impaired myofibrillogenesis.
MEF2 is at the crossroad of multiple [Ca2+]c signals, which affect transcriptional processes (Olson and Srivastava, 1996). Here we show that MEF2C function depends on the presence of a calreticulin-dependent signal. MEF2 integrates Ca2+-dependent signals involving calcineurin, CaMKI, CaMKIV, MKK6/p38 MAPK, and extracellular signal-regulated kinase 5 (Yang et al., 1998; Blaeser et al., 2000; Friday et al., 2000; Han and Molkentin, 2000; Passier et al., 2000). Crt−/− ES-derived cardiomyocytes have normal expression of MEF2C, GATA4, or Nkx2.5, but have an impaired nuclear translocation of MEF2C. MEF2C remains in the cytosol and in perinuclear area in crt−/− ES-derived cardiomyocytes, whereas it is inside the nucleus in wt cells. This is specific for MEF2C because nuclear localization of GATA4 is not affected in crt−/− cardiomyocytes. Indeed, GATA4 was correctly targeted to the nucleus of these cells. Therefore, our data do not favor the hypothesis that the absence of calreticulin, by affecting the Ca2+ filling state of the stores and the nuclear cisternae, could indiscriminately impair the nuclear import of transcription factors through a conformational closure of nuclear pores after an upward shift of the transporter (Perez-Terzic et al., 1996). How nuclear import of macromolecules including MEF2C is impaired in crt−/− cardiomyocytes is currently under investigation.
As expected, MLC2a expression is significantly less affected in crt−/− cardiomyocytes, as it is likely activated by GATA4 and Nkx2.5 factors (Gruber et al., 1998; Franco et al., 1999; Nguyen-Tran et al., 1999; Doevendans et al., 2000). The presence of correctly inserted MLC2a is likely to account for some contractility observed in 5–10% of crt−/− EBs. This is in line with calreticulin-knockout gene studies which showed that calreticulin deficiency primarily affects ventricular development and has no significant effect on atrial cells (Mesaeli et al., 1999).
Although inhibited nuclear translocation of MEF2C in crt−/− cardiomyocytes explains the reduced expression of MLC2v, it cannot account for the lack of insertion of the existing light chains into functional sarcomeres. We show that the latter defect is likely attributed to low active Ca2+/calmodulin-dependent MLCK, which is required to phosphorylate MLC2v prior to its assembly into sarcomeric units and to ensure a normal Ca2+ sensitivity of myofilaments (Clement et al., 1992; Sanbe et al., 1999).
The most important finding of this study is that the calreticulin-deficient phenotype can be rescued by increasing [Ca2+]c in crt−/− EBs. Treatment of EBs at D4 or D7 with ionomycin restores the CaMK-dependent intranuclear localization of MEF2C, MLC2v phosphorylation, its sarcomeric insertion and contractility in crt−/− EBs. In line with these observations, inhibition of CaMK with KN93, chelation of intracellular Ca2+ (BAPTA), or preventing Ca2+ influx in EGTA-buffered extracellular medium abolishes nuclear localization of MEF2C in wt EBs. Thus, phosphorylation of MEF2C by a cytosolic CaMK (Colomer and Means, 2000; Olson and Williams, 2000; Corcoran and Means, 2001), is required for its intranuclear targeting.
An elevation of [Ca2+]c by either a Ca2+ influx across the plasma membrane and/or Ca2+ release from the ER is required to activate CaMKs, including MLCK to drive cardiac myofibrillogenesis. In early cardiomyocytes differentiated within EBs, the initiation of spontaneous contractions is triggered by inositol 1,4,5-trisphosphate–dependent [Ca2+]c oscillations from the ER and it does not require Ca2+-induced Ca2+ release from the sarcoplasmic reticulum (Viatchenko-Karpinski et al., 1999). Crt−/− cardiomyocytes do not feature any spontaneous [Ca2+]c spikes, suggesting a defect in inositol 1,4,5-trisphosphate signaling in calreticulin-deficient cardiomyocytes. Although rapid [Ca2+]c signals induce events such as muscle contraction, secretion, adhesion, and and synaptic transmission, more sustained [Ca2+]c signals are known to be involved in cell proliferation and differentiation. We rescued the expression, the phosphorylation, and the assembly of MLC2v into sarcomeric units by triggering a transient but long-lasting [Ca2+]c increase in crt−/− EBs before the appearance of contractions (i.e., D4 or D7) using the Ca2+ ionophore, ionomycin. This treatment led to a recovery of contractility in crt−/− ES-derived cardiomyocytes. Although applied for a brief period (2 h), ionomycin induced a transient but long-lasting Ca2+ influx into the embryoid body as previously shown in other cell types (Timmerman et al., 1996; Durham and Russo, 2000). This ionophore-dependent Ca2+ increase mimics receptor-mediated Ca2+ elevation into the cytosol, an event missing in crt−/− EBs. It further resets the cardiac differentiation program, paused after an early stage of cardiac commitment.
In summary, although crt−/− cells seem to undergo a normal cardiac commitment up to D4 (time of maximal expression of cardiac transcription factors) (Meyer et al., 2000), they have lost the ability to achieve the full differentiation program, waiting for a Ca2+ signal. This is similar to a Ca2+-dependent block involving CaMKs activation as reported during the development of Xenopus eggs (Lorca et al., 1993). Thus, we provide evidence that a Ca2+-dependent checkpoint involving [Ca2+]c elevation in the cytoplasm and activation of CaMKs is required at an early stage of differentiation to allow cardiac precursors to fully differentiate into cardiomyocytes. This Ca2+ checkpoint plays a dual role: it modulates Ca2+-dependent nuclear translocation of specific transcription factor, and it activates Ca2+-dependent kinases (and phosphatases) involved in early cardiac myofibrillogenensis.
Materials And Methods
Embryonic stem cell culture and differentiation
wt or crt−/− ES cells derived from J1 129/Sv mice were propagated on feeder cells (mitomycin-treated mouse embryonic fibroblasts) using BHK21 medium (GIBCO BRL) supplemented with pyruvate, nonessential amino acids, mercaptoethanol, glutamine, 10% fetal calf serum (Biomedia) (Smith, 1992), and LIF-conditioned medium, obtained from preconfluent 740 LIF-D cells, stably transfected with a plasmid encoding LIF (Meyer et al., 2000).
The cells were maintained at <70% confluency to keep an undifferentiated phenotype. ES cells were differentiated in vitro as embryo-like aggregates (EBs) using the hanging drop method (Maltsev et al., 1994). Briefly, EBs were formed for 2 d (D0–D2) in hanging drops (450 cells/20μl) of differentiation medium (BHK21, as described above), containing 20% fetal calf serum and lacking of LIF. The EBs were then incubated in suspension for 3 d (D2–D5) and transferred to gelatin-coated dishes or glass coverslips (at D6). The differentiating EBs were examined daily under phase-contrast microscopy for the presence of contracting cardiomyocytes.
Antibodies and immunocytochemistry
For immunostaining, EBs were fixed with 3% paraformaldehyde, permeabilized with 0.1% Triton X-100, and labeled with mouse monoclonal anti-MEF2C (1/50) (Cell Signaling), anti-α-actinin antibodies (1/500) (Sigma-Aldrich), with rabbit anti-MLC2v (1/50), or anti-MLC2a (1/500) antisera (Meyer et al., 2000). Secondary antibodies included rhodamine-conjugated goat anti–rabbit IgG and FITC-conjugated goat anti–mouse IgG (Sigma-Aldrich) (dilutions 1/100) (Jaconi et al., 2000). Cell nuclei were visualized within the EBs by Hoechst staining.
Imaging of cytosolic calcium levels
EBs adherent on glass coverslips were loaded at D 4 or D8 with 5 μM fluo3 or fluo4/AM (Molecular Probes) for 20 min, and transferred to the stage of an epifluorescence microscope and perfused with a medium containing 20 mM Hepes, 117 mM NaCl, 5.7 mM KCl, 1.2 mM NaH2PO4, 4.4 mM NaHCO3, 1.7 mM MgCl2, and 1.8 mM CaCl2, pH 7.4. The field was illuminated at 485 ± 22 nm with a Xenon lamp. Fluo3 or fluo 4 emission fluorescence was recorded through a dichroic mirror (cutoff 510 nm) and a long pass emission filter (cutoff 520 nm) (Jaconi et al., 2000). The fluorescent images were recorded at 530 nm long pass filter, using a Nikon (diaphot) or Leica (DMRA) microscope coupled to a CCD camera (micromax 1300Y/HS; Princeton) and digitized on line by a computer (Metamorph software; Universal Imaging). To calculate the frequency of [Ca2+]c spikes in embryoid bodies, a region of interest was selected within a beating area and the average pixel intensity was plotted as a function of time. To record long-term changes in [Ca2+]c induced by ionomycin using time-lapse video microscopy (Metamorph software), EBs were loaded with fluo4 prior to ionomycin application and reloaded 2 h later after washout of the ionophore. EBs were also reloaded with fluo4 at the end of the time-lapse experiment to determine whether the low levels of Ca2+ were due to a leakage of the probe. EBs were illuminated for 2 s, and z-stacks of images were acquired every 15 min for up to 10 h.
All experiments were performed at 35 ± 2°C, under 95%O2/5% CO2 atmosphere, and have been repeated on three separate occasions.
Images of the adherent EBs were acquired with a 12-bit, cooled charge-coupled device interlined camera (Visicam; Visitron System) mounted on a Nikon Eclipse epifluorescence microscope, equipped with 40×, 63×, 100×, 1.3 NA oil-immersion objectives. A monochromator coupled to an optic fiber (DeltaRam, Photon Technology International) allowed the excitation for the FITC (488 nm, DM505, BA 520–560 nm) and the rhodamine (545 nm, DM575, BA 590) fluorescence respectively. Optical z-sectioning (0.2-μm step) was achieved using an objective-mounted piezo-electric controller (LVPDZ position servo controller; Physik Instrumente) driven by the Metamorph 4.1 software (Visitron; Universal Imaging). The stack of images, stored as volume files, were processed by the software Huygens to perform digital restoration of the images (Huygens 2.2.1; Scientific Volume Imaging) and visualized after three-dimensional reconstruction using the Imaris software (Bitplane). All calculations were performed using an Octane workstation (Silicon Graphics).
RNA extraction and reverse transcription reaction
Total cellular RNA was isolated from EBs using the RNA Extraction Kit (Perfect RNA Mini Kit; Eppendorf). Residual genomic DNA was removed by incubating the RNA solution with 15 units of RNase-free DNase I (CLONTECH Laboratories, Inc.) in 2 mM MgCl2 for 45 min at 37°C, followed by inactivation at 90°C for 5 min. 1 μg of DNase-treated RNA was reverse transcribed using the Superscript II kit (GIBCO BRL). The cDNA was diluted 10-fold prior to PCR amplification.
Real-Time quantitative PCR by SYBR green detection
The nucleotide sequences of the PCR primers used were: MEF2C forward 5′-AGATACCCACAACACACCACGCGCC and reverse 5′-ATCCTTCAGAGAGTCGCATGCGCTT; GATA4 forward 5′-CGAGATGGGACGGGACACT and reverse 5′-CTCACCCTCG-GCCATTACGA; and Nkx2.5 forward 5′-TGCAGAAGGCAGTGGAGCTGGACAAGCC and reverse 5′-TTGCACTTGTAGCGACGGTTCTGGAACCAG; β-tubulin forward 5′-CCGGACAGTGTGGCAACCAGATCGG and reverse 5′-TGGCCAAAAGGACCTGAGCGAACGG. Real-time quantitative PCR was performed using a LightCycler rapid thermal cycler (Roche). Amplification was carried out as recommended by the manufacturer. 12 μl reaction mixture contained 1 μl of LightCycler-DNA Master SYBR Green I mix (FAST Start KIT, containing Taq DNA polymerase, reaction buffer, deoxynucleoside trisphosphate mix, and SYBR Green I dye), 3 mM MgCl2, and 0.5 μM concentration of appropriate primer and 2 μl of cDNA. The relative cDNA concentrations were established by a standard curve using sequential dilutions of corresponding PCR fragments. The data were normalized by PCR analysis of the β-tubulin. The amplification program included the initial denaturation step at 95°C for 8 min, and 40 cycles of denaturation at 95°C for 3–5 s, annealing at 60–65°C for 6–10 s, and extension at 72°C for 5–10 s. The temperature transition rate was 20°C/s. Fluorescence was measured at the end of each extension step. After amplification, a melting curve was acquired by heating the product at 20°C/s to 95°C, cooling it at 20°C/s to 70°C, keeping it at 70°C for 20 s, and then slowly heating it at 0.1°C/s to 95°C. Fluorescence was measured through the slow heating phase. Melting curves were used to determine the specificity of PCR products, which were further confirmed using conventional gel electrophoresis.
The Student's t test was used to analyze statistical significance. All p values corresponded to two-tailed tests, and a P of <0.05 was considered statistically significant.
Western blotting analyses
The MLC2v and MEF2C proteins were identified by Western blot analysis (Towbin et al., 1979; Clement et al., 1992). In brief, whole-cell lysate from 50 to 60 EBs was obtained using RIPA buffer (20 mM Tris-HCl, pH 7.8, 150 mM NaCl, 1% Triton X-100, 1% deoxycholic acid, 1mM EDTA, 0.05% SDS) supplemented with a protease inhibitors cocktail (CompleteTM Mini; Boehringer Mannheim). Protein concentrations were determined by BioRad protein assay. Proteins (50–70 μg) were fractionated by SDS-PAGE, electroblotted onto PVDF Immobilon-P transfer membrane (Millipore), and probed with anti–MLC2v (dilution 1/500) or anti–MLC2a antisera (1/5,000), or with anti–MEF2C antibody (dilution 1/1,000). Antibodies were visualized by addition of HRP goat anti–rabbit IgG and enhanced chemiluminescence reagents (ECL reagent; Amersham Biosciences).
Two-dimensional gel electrophoresis
wt or crt−/− cell lysates from D8 EBs were prepared in a buffer containing (in mM): NaCl 150, EDTA 1, Tris 20, Na3O4-vanadate 10, and NP-40 1%, pH 8.2. Only the Triton insoluble fraction containing the myofilaments was resuspended in O'Farell's buffer (O'Farrel, 1975) before loading into the electrofocalization capillaries. Unphosphorylated MLC2v was separated from the phosphorylated protein by high-resolution two-dimensional gel electrophoresis (PAGE) as previously described (Hochstrasser et al., 1988; Gravel et al., 1995). Nonlinear immobilized gradient strips (pH 3–10, NL IPG 18 cm; Amersham Biosciences) were used as the first dimension. Hydration of gel strips was performed by sample (100 μg contractile proteins). Focusing started at 500 V for 2 h, and then the voltage was increased to 1,000 V for 2.5 h, to 3,000 V for 1 h, and subsequently it was gradually increased to 5,000 V (at 500 V/15 min) and kept constant overnight (total 24 h). Separation in the second dimension was performed on 9–16% gradient polyacrylamide/piperazine diacrylamide gels (30:0.8 proportion; BioRad Laboratories). The gels (180 × 200 × 1.5 mm) were run in BioRad apparatus at 40 mA per gel, for 3–4 h. Subsequently, proteins were transferred on PVDF membranes (Amersham Biosciences), and Western blot was performed as described above.
We are grateful to Dr. Pierre Travo (the Core Facility of Bitplane European Advanced Imaging Center, CRBM, CNRS UPR1086, member of the European Light Microscopy Initiative), for continuous support and skillful advice on image restoration and analysis, Dr. Frederick Villhardt for his advice on two-dimensional gel electrophoresis, and M. Oliver Plastre for technical help.
This work was supported by grants from the Swiss National Science Foundation (NRP-4046-058712 and 31-55344.98 to M. Jaconi and KH Krause), Fondation de France (2000003470), and Association Française contre les Myopathies (7778 to M. Pucéat), and from Canadian Institutes for Health Research (to M. Michalak). We also gratefully acknowledge the Region Languedoc-Roussillon, INSERM, and Association Française contre le Cancer for their financial support for the acquisition of the light cycler by the Institut Fédératif de Recherche Jean-François Pechère (IFR24) at Montpellier. M. Pucéat is an established investigator of INSERM, and M. Michalak is CIHR Senior Investigator and AHFMR Medical Scientist.
C. Perez-Terzic's present address is Division of Cardiovascular Diseases, Department of Medicine and Department of Physical Medicine and Rehabilitation, Mayo Clinic, Mayo Foundation, Rochester, MN 55905.
K. Nakamura's present address is Department of Pediatrics, Kumamoto University, School of Medicine, Kumamoto 860-8556, Japan.
Abbreviations used in this paper: [Ca2+]c, cystolic-free Ca2+ concentration; CaMK, Ca2+/calmodulin-dependent kinase; crt, calreticulin; EB, embryoid body; ES, embryonic stem; LIF, leukemia inhibitory factor; MEF, mouse embryonic fibroblast; MEF2C, myocyte enhancer factor C2; MLC2a, atrial myosin light chain 2; MLC2v, ventricular myosin light chain 2; MLCK, myosin light chain kinase; wt, wild-type.