Main

Tetralogy of Fallot is a common congenital cardiac defect with a wide spectrum of clinical manifestations depending on the degree of obstruction to pulmonary blood flow(1, 2). One of the most challenging aspects in the management of these patients, before surgical palliation or repair and frequently during repair, is the treatment of“tet spells” in a subgroup of patients with this diagnosis(1). Tet spells are characterized by paroxysmal episodes of hyperpnea, hypoxemia resulting in severe cyanosis and occasionally leading to dysrhythmias, and loss of consciousness(1). Neurologic sequelae and cardiac dysfunctions are the possible serious consequences of these tet spells(1). Tet spells or paroxysmal hypoxic spells are precipitated by agitation, crying, and situations associated with increased sympathetic activity(1). The mechanism(s) underlying the development of paroxysmal hypoxic spells have not been clearly defined, but are probably multifactorial(1). One proposed mechanism is a dynamic process involving hypercontractility of the right ventricular outflow tract (infundibulum). Because propranolol, a βAR antagonist, has often been effective in reducing the frequency and severity of paroxysmal hypoxic spells(3, 4), excessive βAR stimulation of the right ventricular infundibulum leading to hypercontractility has been specifically implicated. Obstruction from the contraction of the hypertrophied infundibulum would severely reduce pulmonary arterial perfusion and increase right to left shunting, which could result in ever-worsening hypoxemia or the clinical manifestations of paroxysmal hypoxic spells(1).

No study to date has specifically examined the relationship between infundibular myocardial βAR function and the clinical history of paroxysmal hypoxic spells. The purpose of the present study is to characterize the right ventricular infundibular βAR-adenylyl cyclase complex, to determine its function in tetralogy patients, and relate the results to their clinical presentation. Our results indicate that symptomatic tetralogy of Fallot patients have an increase in βAR density compared with asymptomatic patients, suggesting that altered infundibular βAR function may contribute to the development of tet spells.

METHODS

The study was approved by our Institutional Review Board for human studies.

Patient groups. Right ventricular tissues from a total of 29 patients with tetralogy of Fallot were studied. Patients were divided into two groups based on the presence or absence of cyanotic spells. Symptomatic patients (n = 15, age 16.5 ± 26.1 mo, range 1-97 mo; 8 male and 7 female subjects; O2 saturation at rest was 87.7 ± 13.1%, means ± SD) were those who had a prior history of clinical paroxysmal hypoxic spells. Asymptomatic patients (n = 14, age 13 ± 11.9 mo, range 0.5-47 mo; 10 male and 4 female subjects; O2 saturation at rest was 90.4 ± 14.3%, means ± SD) were those who had no prior history of spells preoperatively and who did not show arterial oxygen desaturation or became hypotensive (more than 25% decrease in blood pressure) at the time of aortic cannulation. Patients who had systemic-pulmonary artery anastomoses (shunts) in place or who had received βAR blocker therapy any time before surgery were excluded from the study. None of the patients was receiving any medication preoperatively or had any clinical evidence of cardiac failure. All patients received a standardized anesthetic regimen, as per our usual routine. General anesthesia was induced with halothane and oxygen, supplemented with 75 μg/kg Fentanyl i.v. before initiation of CPB. Muscle relaxation was maintained with vecuronium i.v. During bypass, nitroprusside at 4 μg/kg/min was administered i.v. to facilitate cooling.

Tissue handling. Right ventricular infundibular tissues were obtained at surgery from patients undergoing surgical repair of tetralogy of Fallot. Tissues were then rinsed clean of any obvious blood, blotted dry, quickly frozen in liquid nitrogen, and stored at -70 °C.

Preparation of sarcolemmal membranes. Membranes used for adenylyl cyclase assay were prepared from each patient individually. Whenever possible, membranes were also prepared individually for binding assays, but if available membrane protein was insufficient, then membranes from several patients within the same group were pooled and used for binding assays. Tissues were minced in 6 volumes of a solution containing: 50 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 250 mM sucrose, and 30 mM histidine (pH 7.4), on ice, then homogenized with a Polytron with three 15-s bursts. The homogenate was first centrifuged at 15 000 × g for 15 min at 0°C, and the resultant supernatant was recentrifuged twice at 105 000× g for 45 min. The final pellet was resuspended in 4 volumes of non-EDTA-containing solution and used as membrane preparations for enzyme assays immediately. The final pellet was stored at -70 °C, resuspended as above, and used within 24 h for binding assays. Protein determination was performed immediately before each experiment according to the method of Bradford(5), using BSA as the standard.

βAR radioligand binding experiments. Saturation experiments were performed in duplicates using the radiolabeled antagonist(-)-[125I]ICYP to determine total βAR density. Membranes (25-40μg of protein) were incubated with 8-11 concentrations of [125I]ICYP ranging from 2 to 160 pM in the absence (total binding) and presence of 10μM propranolol (nonspecific binding) in the saturation binding studies. To determine the relative proportion of subpopulations of βAR, we performed competition binding experiments in duplicates using 120 pM [125I]ICYP and with increasing concentrations of CGP 20712 A (10-10-5 × 10-3 M) as the unlabeled β1 antagonist. In both saturation and competition experiments, incubation was for 1 h at 37 °C. Reaction was then terminated by rapid vacuum filtration over glass fiber filters that had been presoaked in 0.5% polyethylenimine, followed by three washing with 5 mL of ice-cold 50 mM Tris buffer (pH 7.5). Filters were counted in a gamma counter at 70% efficiency.

Adenylyl cyclase assay. For adenylyl cyclase assays, membranes(10 μg of protein) were incubated in triplicate at 37 °C for 15 min in a final volume of 100 μL, containing the following reaction mixture (final concentration): 1 mM 3-isobutyl-methylxanthine, 3 mM MgSO4, 50 mM Tris(pH 7.5), 0.2 mM ATP, an ATP regenerating system (creatine phosphate 10 mM and 50 U/mL creatine phosphokinase). Basal adenylyl cyclase activity was measured in the absence of GTP. Adenylyl cyclase assay was performed with 200 μM GTP+ 10 μM isoproterenol to determine response to βAR stimulation. Adenylyl cyclase assays were also performed to examine the response to direct stimulation of G protein by either 200 μM GTP or 10 mM NaF and response to direct activation of the adenylyl cyclase enzyme using 50 μM forskolin. For assessment of β2 receptor-mediated activation, the reaction was performed with the addition of 200 μM GTP and 10 μM isoproterenol in the presence of 5 μM CGP 20712 A. The reaction was stopped by the addition of 1% SDS. After centrifuge at 1000 × g for 15 min, cAMP generated was then determined using RIA (Biomedical Technologies).

Data analysis. When multiple assays have been performed from a single patient, data from only the first set of assay results were included in the analysis. Scatchard analysis of the saturation binding experiments was performed to determine the apparent affinity (Kd) of the ligand for the βAR and the apparent maximum number of binding sites(Bmax)(6) (bound versus bound/free ligand and using least square linear regression). The negative reciprocal of the slope of the best fit yielded the Kd, and the x-intercept the Bmax(6).Bmax was used as an index of βAR density. Data from competition experiments were analyzed using the LIGAND program (Biosoft, Milltown, NJ)(7). To detect differences in adenylyl cyclase activities within each patient group and between patient groups, analysis was by repeated measure analysis of variance, followed by Scheffe's test, if needed. Differences in Kd and Bmax between patient groups were assessed by unpaired t test(8). Data are expressed as means ± SE; p< 0.05 was deemed significant.

RESULTS

βAR density in the group with a history of hypoxic spells(symptomatic) was significantly higher than in the group with no such history(asymptomatic) (Fig. 1). Antagonist affinity was comparable for both groups (Kd = 25.2 ± 7.7 and 23.3± 4.7 pM for asymptomatic and symptomatic groups, respectively).

Figure 1
figure 1

Total βAR density in symptomatic (n = 5) and asymptomatic (n = 6) patients. The antagonist affinity(Kd) was comparable in both groups (asymptomatic, 25.2± 7.7 pM, and symptomatic, 23.3 ± 4.7 pM). *p< 0.05 vs asymptomatic patients. Data are expressed as means± SEM.

Full size image

Basal adenylyl cyclase activity was significantly higher in symptomatic patients. GTP and isoproterenol significantly stimulated adenylyl cyclase activity in both symptomatic and asymptomatic patients, indicating that functional βAR-Gs-adenylyl cyclase signaling pathway was present in the right ventricular infundibular myocardium in all Tetralogy patients (Fig. 2A). In addition, GTP-, NaF-, and isoproterenol-stimulated adenylyl cyclase activities were greater in symptomatic compared with asymptomatic patients (Fig. 2, A and B). Adenylyl cyclase activity in response to forskolin stimulation (Fig. 2B), however, was comparable in both groups. Therefore, although unstimulated adenylyl cyclase activity was higher in the symptomatic patient, there was no difference between the two groups in the stimulated catalytic activity of the cyclase enzyme.

Figure 2
figure 2

(A) Basal and GTP- and isoproterenol-stimulated adenylyl cyclase activities in symptomatic(n = 11) and asymptomatic (n = 9) patients. (B) Basal and NaF- and forskolin-stimulated adenylyl cyclase activities in symptomatic (n = 3) and asymptomatic (n = 4) patients. GTP, 200 μM; isoproterenol (ISO), 10 μM + GTP 200 μM; NaF, 10 mM; forskolin, 50 μM. *p < 0.05 vs same condition in asymptomatic patients. ^p < 0.05 vs basal.§p < 0.05 vs GTP-stimulated. Data are expressed as means ± SEM.

Full size image

Because subpopulations of βARs are differentially regulated, we examined whether there might be a selective upregulation of a βAR subtype in the symptomatic patient. The up-regulation of βAR was not subtype-specific. The distribution of β1 and β2 receptors was comparable in symptomatic and asymptomatic patients (Fig. 3). In the presence of the β1 selective antagonist, CGP 20712 A, isoproterenol failed to stimulate adenylyl cyclase activity in both patient groups (Fig. 4). These results indicate that β1 receptors were the predominant, if not the exclusive, functionally coupled receptor subtype in the right ventricular infundibular myocardium in all tetralogy patients.

Figure 3
figure 3

The relative proportion of β1 andβ2 βAR in symptomatic and asymptomatic patients (n= 5 for each group). Error bars for asymptomatic group are not visible. Data are expressed as means ± SEM.

Full size image
Figure 4
figure 4

Effect of the β1 AR antagonist, CGP 20712 A(CGP), on isoproterenol (ISO)-stimulated adenylyl cyclase activity (n = 6). ^p < 0.05 vs basal. §< 0.05 vs GTP-stimulated. Data are expressed as means ± SEM.

Full size image

DISCUSSION

Our results indicate that total βAR density was increased in the right ventricular infundibular myocardium of symptomatic tetralogy of Fallot patients, or those who had paroxysmal hypoxic spells. There was a proportional increase in both β1 and β2 receptor subtypes so that subtype distribution remained unchanged. Although the βAR signaling pathway was functional in all Tetralogy patients, the up-regulation ofβAR density in symptomatic patients was associated with a corresponding increase in βAR-mediated adenylyl cyclase activity.

βAR is the most potent of the excitatory receptor pathways that mediate positive inotropy in humans(9). Stimulation ofβARs regulates myocardial contractile function mostly through modulation of intracellular cAMP content. βARs couple to a stimulatory GTP binding protein, Gs, to stimulate adenylyl cyclase activity and increase cAMP. cAMP acts at several different sites to increase inotropy(10). cAMP activates protein kinase A, which in turn promotes trans-sarcolemmal Ca2+ entry through L-type Ca2+ channels, increase sarcoplasmic reticulum Ca2+ release and decreases myofilament Ca2+ sensitivity(11). βARs have also been shown to cause direct increase in current through voltage-gated calcium channels via coupling to the stimulatory G protein(12). Infundibular hypercontractility may result from a direct increase in βAR-mediated contractile function due to up-regulation of βAR and/or in βAR-stimulated adenylyl cyclase activity and/orβAR-mediated non-cAMP dependent function.

Several studies have examined the infundibular βAR complex in patients with tetralogy of Fallot. McGrath et al.(13) used the radiolabeled ligand 3H-dihydroalprenolol to determine βAR density in patients with tetralogy and ventricular septal defects and found that the βAR density was not significantly different in these two groups of patients(13). Moreover, βAR density and antagonist affinity could not be correlated to the resting arterial desaturation in their patients(13). However, resting arterial desaturation has not been shown to correlate with the risk for paroxysmal hypoxic spells(14). Therefore, their study provided little information regarding the function of the βAR pathway and its relation to the clinical event of paroxysmal hypoxic spells. Brodde et al.(14) found βAR density andβAR-stimulated adenylyl cyclase activity in tetralogy patients to be less than values reported in right ventricular tissues from adult patients(15). Their patients spanned a wide age range from under 1 y to 23 y of age. No age-matched controls were used in their study, and no attempt was made to distinguish clinical subgroups of patients. Kaumann et al.(16) reported that infundibular contractile function in response to catecholamines stimulation was related toβAR-mediated adenylyl cyclase activity in tetralogy patients. However, only a total of three tetralogy patients were studied. Their clinical history was not reported, so there was no information regarding the occurrence of paroxysmal hypoxic spells in these patients.

Our study was the first that attempted to examine the infundibular βAR complex in tetralogy of Fallot patients on the basis of clinical presentation. Clinical symptoms were defined based on documented episodes of paroxysmal hypoxic spells. We chose to exclude patients who had systemic-pulmonary arterial shunts in place, because we were uncertain whether the presence of shunts might alter infundibular βAR function. Systemic-pulmonary arterial shunts usually alter volume loading and the degree of cyanosis, both of which could secondarily modify βAR density and function.

All of the data in the present study were obtained from infundibular tissues excised more than 30 min after the onset of hypothermic cardioplegic CPB. A possible limitation of the present study was the effect of CPB onβAR function(17, 18). βAR desensitization was reported in right atrial tissues from children between 21 and 95 min after initiation of hypothermic CPB. Although human myocardium contains β2 ARs which have been shown to be functionally coupled to the βAR-Gs-adenylyl cyclase signaling pathway(16), we failed to observe any β2-mediated adenylyl cyclase activity in any patient. Thus, β2 AR desensitization might have occurred in the infundibular tissues obtained at surgery. Studies of effect of CPB on βAR function have documented that human myocardial β2 ARs were preferentially desensitized during CPB(17). Alterations of βAR function, however, should be similar in symptomatic and asymptomatic patient because bypass time, degree of hypothermia, cardioplegic solutions were comparable in both groups.

We found βAR density and βAR-stimulated adenylyl cyclase activity to be higher in symptomatic patients. Furthermore, basal cyclase activity and response to stimulation of the βAR-Gs-adenylyl cyclase signaling pathway at the level of the G protein, using either GTP or NaF, were also enhanced. However, there was no difference in the stimulated catalytic activity of the cyclase enzyme itself in the two groups, because response to forskolin was comparable. These findings suggest that not only were there more infundibular βAR, but the signaling pathway in symptomatic patients might also be more “sensitized.”

Whereas our study has focused on the βAR-Gs-adenylyl cyclase signaling pathway, which has been considered as the signaling pathway, activation of non-cAMP-dependent signaling pathways by βAR has been well documented(12, 19). It is possible that infundibular hyperactivity in the symptomatic tetralogy patient might also be related to βAR activation of additional pathways such as direct coupling of βAR to the voltage-gated calcium channels. Finally, our study did not specifically examine the cause and effect relationship between infundibularβAR up-regulation and clinical manifestation of paroxysmal hypoxic spells. Our results, however, were more consistent with the idea that enhanced infundibular βAR function contributed to the development episodes of paroxysmal hypoxic spells. Although pressure overload hypertrophy might be expected to cause receptor up-regulation, this has not been documented when there was concomitant hypoxemia(20). In experimental right ventricular outflow tract obstruction with right to left shunting, right ventricular hypertrophy and hypoxemia did not induce any change in right ventricular βAR number or function(20). It was therefore more likely that infundibular βAR up-regulation contributed to its hyperreactivity in the symptomatic patient.

In summary, we examined the infundibular myocardium tissues obtained at surgery from tetralogy patients. Our results indicate that patients who have a history of paroxysmal hypoxic spells have an increase in total βAR number and enhanced βAR-mediated adenylyl cyclase activity in the infundibular tissues. Findings from the present study are consistent with the idea that infundibular βARs may play a role in the development of paroxysmal hypoxic spells. Future studies are needed to examine non-cAMP-dependentβAR-mediated functions, infundibular response to βAR agonist in vivo, and modulators of βAR function in the symptomatic and asymptomatic tetralogy patient.