Main

It is now well established that the cardiac neural crest makes major contributions to normal cardiovascular development. This is supported by extensive work in an embryonic chick model of neural crest-related heart development and, more recently, by work in genetically defined mouse models(for reviews, seeRefs. 1–3). These studies have led to the classification of a whole series of human congenital heart anomalies as neural crest-related. Complete bilateral ablation of the cardiac neural crest in the chick embryo (neural crest extending from mid-otic placode to the caudal region of somite 3) before migration, leads to failure of septation of the cardiac outflow tract and is referred to as PTA. This is the most severe of the neural crest-related heart defects with a mortality of 38% even with extensive clinical intervention (from 1992 figures of the Pediatric Cardiac Care Quality Assurance Consortium). Chick embryos with PTA are edematous and have low body weight as well as a large heart with respect to body weight(4). Survivability is poor and presumably due to heart failure. These observations in the chick embryo parallel the phenomenon of “fetal wastage” often observed during prenatal life in humans with outflow tract defects(5). Such findings are suggestive of poor myocardial function. That there is impaired myocardial function is supported by recent ultrasound studies of embryos with PTA that demonstrate a reduction in the fraction of the diastolic volume of blood ejected during systole(6). In fact, in ovo microcinephotographic studies demonstrate a reduction in the ejection fraction in very young embryos with neural crest ablation even before the period when septation of the outflow tract would normally occur(7, 8). These whole-animal observations are further supported by a reduction in contractile force seen in ventricular trabeculae(9) and reduced L-type Ca2+ current in isolated myocytes from hearts with PTA(10, 11).

The reduced L-type Ca2+ current suggests that impaired EC coupling is contributing to poor myocardial function in hearts with PTA. In the present study, Ca2+ transients in fura-2-loaded myocytes from normal hearts and hearts with PTA were measured. The results show that Ca2+ transients are reduced in hearts with PTA, which corresponded to a reduction in the L-type Ca2+ current. The results further indicate an impairment of SR function. These results strongly support our hypothesis that the poor viability of embryos with PTA is due to impaired cardiac EC coupling.

METHODS

Animal surgery. Fertilized Arbor Acre chicken eggs(Seaboard Hatchery, Athens, GA) were incubated in forced draft incubators maintained at 90% humidity and 38°C. The eggs for sham-operated and experimental animals were windowed at Hamburger-Hamilton stages 8-10(approximately 30 h of development). The embryos were stained with neutral red, and the cardiac neural crest cells were ablated bilaterally with a pulsed nitrogen/dye laser (model VSL-337/DLM-110. Laser Science, Inc.) as previously described in detail by Kirby et al.(12). This lesion produces PTA in >90% of embryos that survive to d 15 as the cardiac neural crest normally provides ectomesenchymal cells essential for septation of the cardiac outflow tract(13). The microsurgery was performed by a surgery Core Unit, under the supervision of Dr. Margaret L. Kirby, using procedures which are standardized to produce PTA. For sham-operated embryos, the embryos were stained with neutral red but the laser ablation was not performed. All eggs were sealed with cellophane tape, reincubated and harvested at embryonic d 15. The hearts from embryos with neural crest ablations were examined macroscopically at 10 × magnification with a dissecting microscope. Only those diagnosed with PTA were used. The hearts from sham-operated embryos were considered“normal” if no outflow tract anomalies were detected at the time of macroscopic inspection. By d 15, the mortality of embryos with neural crest ablation was >70%. PTA was never observed in sham-operated embryos.

Myocyte preparation. Ventricular myocytes from sham-operated and experimental embryos were dissociated and cultured overnight either on coverslips or plastic culture dishes as previously described in detail elsewhere(10, 11, 14). All cultures were used within 18-24 h of dissociation. All experiments were carried out at room temperature.

Ca2+ transient measurements. The coverslips with the attached cells were transferred to a perfusion chamber (Medical Systems) and gently washed several times with a Ringer solution containing (in mM): 142.5 NaCl, 4.0 KCl, 1.8 MgCl2, 1.8 CaCl2, 5.0 glucose, and 5.0 HEPES (pH 7.4, NaOH). After washing the cells with the Ringer solution they were loaded with fura-2/AM as previously described(14). We previously determined that under our experimental conditions cytoplasmic fura-2 concentrations ranging from 10 to 45 μM did not affect the magnitude and the shape of the Ca2+ transients(14). In the present study we have estimated a fura-2 cytoplasmic concentration of approximately 20 μM using the methods described by Moore et al.(15) and as modified by Brotto and Creazzo(14). A Photon Technology International microspectrofluorometer with dual monochromators was used to collect the fura-2 calcium transients (340/380 nm ratios) as previously described in detail elsewhere(14). The myocytes were electrically stimulated at 0.2 Hz by applying 10 V for 7 ms via a pulse stimulator (Hewlet Packard, model II) through two platinum electrodes placed on either side of a single myocyte. Myocytes were continuously stimulated for several minutes before measurements were made. This was sufficient to achieve stable transients and diastolic Ca2+ levels in these myocytes(14).

Patch clamp procedure. The voltage clamp protocols and solutions used in this report were the same as previously described(10, 11). Briefly the extracellular solution was comprised of (in mM): 1.8 CaCl2, 20 CsCl, 120 tetraethylammonium chloride, 1.8 MgCl2, 10 HEPES, 3 μM tetrodotoxin (approximately 500 times KI for the chick cardiac sodium channel(16); 5 dextrose, adjusted to pH 7.4 with CsOH. The perforated patch configuration was achieved using nystatin essentially as described by Horn and Marty(17) and as routinely used in this laboratory(14, 18). The perforated patch solution was comprised of (in mM): 55 KCl, 70 K2SO4, 7 MgCl2, 10 HEPES, and 5 dextrose (pH 7.3, KOH). In some experiments Cs2SO4 was used instead of K2SO4 with no noticeable difference in the calcium current. Seal resistances were typically>50 Gohms. Currents were measured with a Dagon 3900 amplifier. Data acquisition and voltage pulses were controlled with pClamp software and the TL-125 Labmaster system from Axon Instruments. Cell capacitances ranged from 4 to 8 pF.

The current records were analyzed using pClamp software. Cursor measurements were used to determine leak current and the peak magnitude as previously described(10, 11). After correction for leak current, which increases linearly with membrane potential, the peak magnitude is measured as the difference between the peak inward current and the current at the end of the 400-ms voltage pulse when the Ca2+ current has completely decayed(10, 11). Automated computer analysis was not possible as the Ca2+ currents were often to small to be accurately measured by the pClamp software, especially in hearts with PTA.

RyR binding. [3H]RyR binding was carried out by modification of the technique of Kim et al.(19). The ventricles were dissected and homogenized in 10 volumes of assay buffer, containing 10 μM leupeptin and 200 μM phenylmethylsulfonyl fluoride, using a Dounce tissue homogenizer and a Cafamo stirrer type RZR 50 (Ontario, Canada). Four to six hearts were pooled for each assay. The assay buffer consisted of 1 M KCl, 20 mM 4-morpholinepropanesulfonic acid (pH 7.4), and 3 μM added CaCl2. Each assay tube contained 200 μL of homogenate and 50 nM[3H]ryanodine (68.4 Ci/mmol, NEN) such that the final volume was 250μL in a 1.5-mL Eppendorf tube. To determine nonspecific binding, some tubes contained 100 μM unlabeled ryanodine. Incubation was for 2 h at 37°C in a rotating water bath (80 rpm). The reaction was terminated by adding 100μL of 40% polyethylene glycol solution and incubating at room temperature for 5 min. The tubes were then centrifuged at 14 000 rpm in an Eppendorf microcentrifuge for 10 min, and the pellets were washed twice with 400 μL of assay buffer. The pellets were dissolved by sonication in 250 μL of ScintiGest tissue solubilizer (Fisher Scientific Co.) and subsequent incubation for 2 h at 70°C. The samples were counted at 50% efficiency in 10 mL of Scinti Verse scintillation cocktail (Fisher Scientific Co.). Samples contained from 30 000-40 000 cpm with about 40-50% background. Protein content was determined using the bicinchoninic acid protein assay kit (BCA; Pierce, Rockford, Il) after dissolving the protein in 4% SDS. Each assay tube contained 1.5-1.8 mg of protein.

Statistics. Sham-operated and PTA groups were compared using the t test in the SigmaStat (Jandel Corp.) computer software package.

RESULTS

Ca2+ transients in normal hearts.Figure 1, A and B, shows examples of Ca2+ transients in a d 15 embryonic chick ventricular myocyte from a normal heart of a sham-operated embryo, in the absence or presence of 100μM ISO. The myocyte was stimulated at 0.2 Hz. The extracellular Ringer solution contained 1.8 mM CaCl2. Under these conditions, ISO nearly doubled the magnitude of the Ca2+ transients and accelerated both the time to peak and the exponential decay. We have previously shown that this dose of ISO maximally stimulated the magnitude of Ca2+ transients in chick myocytes(14). The results are summarized inTable 1. These measured parameters are in the range of those reported for avian and mammalian myocyte preparations by other investigators(20–22).

Figure 1
figure 1

Ca2+ transients in ventricular myocytes from a normal heart from a sham-operated embryo and hearts with PTA at embryonic age d 15. (A and B) A myocyte from a sham-operated embryo(Sham) in the absence and in the presence of 100 μM ISO.(C) A myocyte from a heart with PTA. The inset shows transients from the same myocyte after signal averaging. (D) A transient from a myocyte from a heart with PTA in the presence of 100 μM ISO. Note that transients from hearts with PTA were small and poorly distinguishable from background noise unless signal-averaged or treated with ISO. The transients were elicited by field stimulation at 0.2 Hz.

Full size image
Table 1 Parameters for calcium transients in myocytes from normal hearts (sham) and hearts with PTA
Full size table

Reduced Ca2+ transients in hearts with PTA. In contrast with normal hearts, field stimulation-induced Ca2+ transients were very small in myocytes from hearts with PTA and difficult to observe above background noise (Fig. 1C). Measurements of these tiny transients were facilitated with the use of signal averaging(Fig. 1C inset and Table 1). The peak systolic Ca2+ was 43% less than that of the sham. These data are consistent with previous observations which show that L-type Ca2+ current is similarly reduced in hearts with PTA at d 11 of incubation(10, 11). The kinetics of the transients were also significantly slowed (Table 1). The time to the peak of the transient and the time constant of decay were slowed by 40 and 42%, respectively. Ca2+ transients were more clearly observed when the myocytes were treated with 100 μM ISO, and signal averaging was not necessary (Fig. 1D). However, even in the presence of ISO, the Ca2+ transients were still significantly smaller than those observed in preparations from sham-operated embryos in the absence of ISO(Table 1). Second, the time-to-peak and decay kinetics in ISO stimulated myocytes from normal hearts were nearly twice as fast compared with the myocytes from embryos with PTA in the presence of ISO(Table 1). The kinetic parameters for the Ca2+ transients in hearts with PTA treated with ISO appeared similar to the kinetics in myocytes from sham-operated embryos in the absence of ISO.

Reduced L-type Ca2+ current in hearts with PTA. In previous studies it was found that the peak magnitude of the L-type Ca2+ current was reduced by approximately 50% in ventricular myocytes from chick hearts with PTA at embryonic age d 11(10, 11). In the present report, we find a similar reduction of L-type current at embryonic age d 15 (Fig. 2). The decreased L-type Ca2+ current could account for the reduced Ca2+ transients, because there would have been less extracellular Ca2+ entering the myocyte and, in addition, less Ca2+ to trigger CICR from the SR. Thus, it should be possible to restore the Ca2+ transients by increasing L-type current comparable to the level found in normal hearts from the sham-operated embryos. To test this possibility, we examined whether the dose of ISO used to stimulate the Ca2+ transients in hearts with PTA (Fig. 2) could elevate the L-type Ca2+ currents to a level equal to or greater than that observed in myocytes from normal hearts. The results from these experiments indicated that 100 μM ISO elevated the L-type Ca2+ current to a level comparable with that seen in myocytes from normal hearts. However, as already described above, exposure to ISO did not completely restore the magnitude of the Ca2+ transients in myocytes from hearts with PTA(Table 1). These results suggested the possibility that there is decreased Ca2+ release from the SR in addition to reduced L-type Ca2+ current in hearts with PTA.

Figure 2
figure 2

L-type calcium current in ventricular myocytes from normal hearts from sham-operated embryos (Sham) and hearts with PTA.(A) Examples of calcium current records in the absence and presence of 100 μM ISO. The currents were elicited by a 500-ms prepulse to -40 mV from a holding potential of -80 mV followed by a 400-ms test pulse to +10 mV(10, 11). Horizontal lines indicate zero current and leak correction was not applied. (B) Mean current-voltage relationships illustrating the effects of ISO on myocytes from normal hearts and hearts with PTA. n = 5-7 myocytes per point.

Full size image

Reduced caffeine-stimulated Ca2+ transients in hearts with PTA. If the Ca2+ current is decreased then there should be less loading of the SR leading to diminished SR Ca2+ release. To test this hypothesis, fura-2-loaded myocytes from embryonic d 15 normal hearts and hearts with PTA were exposed to 50 mM caffeine to release all of the available Ca2+. Caffeine was applied to myocytes that were stimulated for at least 5 min to achieve steady-state loading of the SR. The examples inFigure 3 and the results given in Table 2 showed that, although caffeine was able to induce calcium release from the SR of myocytes from hearts with PTA, the time to peak, the exponential decay (τ), and the magnitude of the calcium transients were all significantly reduced. These data indicate that there is less loading of the SR, which would lead to less SR Ca2+ release in hearts with PTA. The slowed kinetics further suggest that there may be impaired CICR by a mechanism other than a reduction in number of Ca2+ release channels.

Figure 3
figure 3

Examples of caffeine (50 mM)-stimulated Ca2+ transients in quiescent myocytes. (A) A myocyte from a sham-operated embryo (Sham). (B) A myocyte from a heart with PTA. See parameters in Table 1. Field stimulation was at 0.2 Hz.

Full size image
Table 2 Parameters for caffeine stimulated calcium transients
Full size table

[3H]Ryanodine binding is not reduced in hearts with PTA. One possible explanation for decreased SR Ca2+ release in hearts with PTA would be a reduction in the number of Ca2+ release channels or RyRs. Therefore, we compared the number of[3H]ryanodine binding sites in normal hearts from sham-operated embryos with hearts with PTA. For these experiments, RyRs were assayed in two separate whole-ventricle homogenates each from sham-operated and neural crestablated embryos. Each homogenate contained tissue from 4-6 ventricles and a total of 9 hearts with PTA and 11 normal hearts from sham-operated embryos were used. Contrary to expectation, the results showed that there was about a 25% elevation of RyRs in hearts with PTA (sham, 730 ± 70, and PTA, 970± 210 fmol/mg of protein; ± SD). The respective number of RyRs was 6.6 and 8.0 times the number of DHPRs, previously reported by Aiba and Creazzo(11) in hearts from sham-operated embryos and in hearts with PTA. Although the previous report compared DHPRs binding at d 11 of incubation, DHPR binding in ventricles was relatively constant during this period of development in the chick embryo (T. L. Creazzo, unpublished results). Furthermore, the results reported here are consistent with a study indicating that the ratio of DHPRs to RyRs ranges from 4 to 10 in the heart, depending on the species(23). These results indicate that the reduced Ca2+ transients in hearts with PTA are not due to a reduction in RyRs.

DISCUSSION

A number of experimental observations of hearts with PTA, a lethal cardiac neural crest-related heart defect, indicate poor myocardial function, which is suggestive of impaired EC coupling (see Introduction). Previous reports from this laboratory indicate a reduction of L-type Ca2+ current in embryonic chick hearts with PTA at d 11 of incubation(10, 11). This report confirms these findings at d 15 of incubation in the chick (Fig. 2). In agreement with these findings, the present report shows that the magnitude of the field stimulation-induced Ca2+ transients is reduced in hearts with PTA. Thus, these hearts do not compensate for the loss of Ca2+ entering via sarcolemmal Ca2+ channels by other means such as enhanced CICR and reverse Na+-Ca2+ exchange. In fact, in addition to a reduced L-type current, there appears to be less Ca2+ release from the SR as well as less Ca2+ available for release by the SR.

SR function and structure is well developed in the embryonic chick heart by embryonic d 15. Imunofluorescence labeling indicates the presence of SR Ca2+-ATPase in the developing chick myocardium at stages 9-10 (≅36 h of incubation) when the primitive heart tube begins contracting(24). Ca2+ sequestration by SR has been detected at low levels at d 2.5, and the level increases in a near linear fashion through hatching(25). RyR binding studies indicate the presence of Ca2+ release channels by d 4 of incubation(26). Second, we have found that 100 μM ryanodine inhibits Ca2+ transients by 30% at d 11(14) and that the inhibition increases to about 60% by d 15 (T. L. Creazzo and A. P. Brotto, unpublished results). Electron microscopic observations and imunofluorescent co-labeling of DHPR and RyR indicate that the size and density of SR junctions with the surface membrane is similar to adult chicken by d 15(27). Phosphorylation of the SR protein phospholamban has also been detected as early as d 4 in the chick embryo(28). ISO would be expected to stimulate Ca2+ uptake and loading of the SR via phosphorylation of phospholamban, thereby increasing the amount of Ca2+ available for release(29–31). Despite the possibility that loading of the SR may be enhanced in addition to elevation of the Ca2+ current, ISO was not able to completely restore the magnitude of the Ca2+ transients to normal levels.

Caffeine-stimulated transients were substantially reduced in myocytes from hearts with PTA. The most likely explanation is that there is less Ca2+ loading of the SR as a consequence of less Ca2+ entering the cytoplasm due to a reduction in L-type Ca2+ current. A second possibility is that there is less uptake due to an impairment of the SR Ca2+-ATPase mechanism. Reduced SR Ca2+-ATPase activity and expression of this enzyme has been demonstrated in experimental models of adult cardiac hypertrophy and in humans with heart failure(32). The slowed decay kinetics of field-stimulated transients (Table 1) in hearts with PTA support this hypothesis. The relatively slow kinetics of the caffeine transients suggests that Ca2+ release is also impaired in hearts with PTA (Table 2). The results indicate a need for more detailed studies of SR function in hearts with PTA.

Another possible explanation for the slowed decay kinetics of both field-stimulated and caffeine-stimulated Ca2+ transients is that the level of Na+-Ca2+ exchange activity may be decreased in hearts with PTA. The Na+-Ca2+ exchanger is quite prominent in the avian heart and plays the predominate role in extruding cytosolic Ca2+ in early heart development when the SR is sparse(25). It is also possible that a decrease in the rate of Ca2+ extrusion, by decreased Na+-Ca2+ exchange and/or SR Ca2+-ATPase activity, could account for some of the decrease in L-type Ca2+ current. The prolonged elevation of [Ca2+]i may result in greater Ca2+-dependent inactivation of L-type Ca2+ channels. However, the effect on inactivation is probably not significant as we have found that the decay kinetics for L-type Ca2+ current are not different from normal(11). It remains to be determined whether the Na+-Ca2+ exchanger is affected during development in hearts with a neural crest-associated heart defect.

The data discussed above are suggestive of a reduction in the gain for CICR in hearts with PTA. This implies that, for each Ca2+ entering the cytosol via Ca2+ channels, fewer Ca2+ ions are released from the SR when compared with normal hearts from sham-operated embryos. Although it is not possible to accurately determine gain from the experiments in this study, we can make a rough comparison between normal hearts and hearts with PTA from the ratio of the average peak change in Ca2+ during a transient(Δ[Ca2+]i = S[Ca2+]i - D[Ca2+]i; see Table 1) divided by the average peak of the Ca2+ current at +10 mV. These calculations provide arbitrary numbers for relative comparisons and do not represent the actual gains of CICR. The calculated ratios are 164 and 76 and 233 and 56 in the presence of ISO for sham-operated and PTA, respectively. These calculations suggest that the gain for CICR in considerably reduced in hearts with PTA both in the absence or presence of ISO. The gain appears to increase in normal hearts in the presence of ISO, which is not unexpected because ISO-stimulated phosphorylation of phospholamban should enhance Ca2+ loading of the SR. Part of the reason for the relatively low gain with PTA, with or without ISO, may be related to the fact that the SR is poorly loaded as indicated by the caffeine experiments. Bassani et al.(33) have shown that the gain of CICR is nonlinearly related to the degree of SR Ca2+ loading such that the gain is quite low in poorly loaded SR.

Reduced Ca2+ transients are observed in hearts with PTA despite the fact that the number of RyRs is not decreased. This is consistent with the possibility of impaired SR Ca2+ release in these hearts as there are a number of endogenous factors known to modulate RyR activity in addition to Ca2+ which may have been affected by PTA. In heart, these could include ATP, Mg2+, calmodulin, sphingosine, pH, cyclic ADP ribose, and protein phosphorylation [reviewed by Meissner(34)]. In a similar fashion, the reduction in L-type calcium current in hearts with PTA occurs without a concomitant reduction in the number of DHPRs(11). It remains to be determined whether regulation of RyRs by endogenous factors differs in normal hearts and hearts with PTA.

A final important consideration is that the neural crestablated embryos used in this study probably represent a subpopulation of“survivors” as the majority of embryos with this lesion die before d 15. This subpopulation apparently has a less severe spectrum of neural crest-associated anomalies than the embryos that die earlier. At d 15, chick embryos are well passed the period of cardiac morphogenesis and are equivalent to the fetal stage in humans. Thus the survivors examined in this study, may be more representative of the clinical spectrum of liveborn children with heart defects.

In summary, this is the first report to demonstrate a reduction in the cardiac systolic Ca2+ transient in an experimental model of congenital heart disease. The results from this study demonstrate that there is marked impairment of EC coupling in embryos with PTA. Second, the data show that decreased L-type Ca current alone cannot account for the diminished Ca2+ transients in myocytes from these hearts. Finally, the results are a strong indication that impaired EC coupling is the major cause of poor cardiac function leading to edema and poor viability of embryos with PTA.