Extensive experimental and epidemiological evidence points to the cardiac toxicity of bisphenol-type environmental chemicals. However, the impact of bisphenols and other phenols on the cardiac electrical properties in humans is entirely unknown. This is the first study to examine the associate of phenol exposure and ECG alterations in a human cohort, and provides insights into phenol associated, sex-specific changes in cardiac electrical properties. The key findings of this study relate to the changes observed in ECG parameters, PR interval and QTc in particular, as it pertains to BPA and other phenol levels when analyzed by sex. These alterations of cardiac electrical properties have clear physiological and pathological implications.
PR interval represents the electrical conduction between the atria and the ventricle, and prolongation of the interval is associated with increasing age as the AV node function decreases [37, 38]. This is relevant as advanced AV blocks are a common indication for placement of artificial cardiac pacemakers [39] and are also relevant in an aging population that is prone to develop atrial arrhythmias (such as atrial fibrillation) that depend on normal AV conduction to maintain cardiac output [40–42]. In this study we found the PR interval increases with age in both males and females as expected, and increases in females in association with BPA and BPF exposure. Further, the combined effect of BPA exposure with elevated BMI was associated with an approximate 10% increase in PR interval in females. This association between PR prolongation and bisphenol exposure is generally consistent with published experimental data showing that exposure to BPA delayed AV conduction and increased the PR interval in excised female rat hearts [43]. Conduction velocity in cardiac tissue is determined by the size of the action potential-generating inward current [44]. It is possible that BPA reduces AV node conduction through inhibition of the L-type Ca2+ current, which is the dominant inward current in AV node myocytes and a key determinant of AV node conduction velocity [44, 45]. Recently, we have shown that 1 nM BPA acutely inhibited the L-type Ca2+ current in hiPSC-CMs, including in nodal-like myocytes [16], and that long term exposure to 1 nM BPA inhibited the L-type Ca current in hiPSC-CMs through reduction of the L-type Ca2+ channel protein expression level [19]. Similarly, low-doses of BPA inhibited the L-type Ca2+ current in female rat cardiomyocytes in a dose-dependent manner [14]. At higher, µM doses, BPA has also been shown to suppress the L-type current/channel in hiPSC-CMs and heterologously expressed in HEK cells [29, 30]. Such BPA-mediated inhibition of the L-type Ca2+ channel may provide a cellular mechanism for the observed prolongation of the PR interval in the Fernald Cohort.
It has been well-established that prolongation of the QT interval is a key arrhythmogenic mechanism and are associated with adverse cardiac outcomes, specifically sudden cardiac death [25–27]. QT prolongation is an indication of delay in ventricular repolarization, which can lead to ectopic ventricular excitation and malignant ventricular arrhythmias [25, 26, 46]. QT prolongation is often caused by genetic mutations in cardiac ion channels, predominantly K+ and Na+ channels, and the resulting prolongation of ventricular action potential [47, 48]. QT prolongation can also be produced by adverse blockade of cardiac ion channels by drugs or environmental chemicals, and is one of the central issues in cardiac safety and cardiotoxicity assessment [49–53]. Examples of QT prolongation-related cardiac toxicity of environmental chemicals include trivalent arsenic As3+ and organophosphate poisoning [54–57]. Here, we show that higher urine level of TCC was associated with longer QTc in males but not females. TCC, an antibacterial chemical once common in personal care products, is now banned for usage in the US. Information on the association between TCC exposure and QT prolongation toxicity may help retrospective understanding of the health outcomes of the Fernald Cohort. In addition, even though there is no association between BPA exposure and QT change at the cohort level, we found that in older males, higher BPA exposure may have an association with QTc prolongation. While QTc difference between the lowest to the highest groups was modest (about 20 msec), from a population perspective even smaller increases in QTc have been shown to be associated with increased risk of adverse outcomes. For instance, Zhang and colleagues found that any increase in QT interval was associated with a small but measurable change in total and cardiovascular mortality and specifically sudden cardiac death [58]. Our findings have important implications for the QT prolongation cardiotoxicity of BPA, and likely other related bisphenol-type chemicals, in older males.
From a cellular physiologic perspective, BPA has been shown to inhibit the IKr K current, which plays a key role in cardiac repolarization. We showed in our previous studies that acute exposure to environmentally relevant low dose BPA (1 nM) inhibited the IKr current in hiPSC-CMs and canine ventricular myocytes. Such IKr inhibition resulted in delay in action potential repolarization in hiPSC-CMs and canine ventricular myocytes, and prolongation of QT interval in canine ventricular tissue [15, 16]. The pro-arrhythmic toxicity of such BPA-induced repolarization delay was illustrated by the marked increase in arrhythmic events under BPA exposure combined with pathophysiological conditions [16]. In addition, acute exposure to high doses (mid to high µM) BPA has also been shown to inhibit IKr in hiPSC-CMs and hERG (the molecular correlate of human IKr current) expressed in HEK 293 [30]. While these acute experimental exposure findings cannot be directly extrapolated to the exposure scenario in humans, they nevertheless offer a possible cellular mechanism of the observed QT prolongation associated with bisphenol exposure in the cohort.
No change in QRS duration or heart rate was found to be associated with exposure to any single tested phenol, while higher urine BPA plus BPF level was associated with slight QTS prolongation in females. QRS duration represents depolarization of the ventricles and is prolonged under disease conditions including heart failure, prior infarction, and/or development of a bundle branch block. The etiology of the QRS prolongation associated with BPA plus BPF levels is unknown, and further experimental and human population studies are needed to fully clarify this. That being noted, Hnatkova and colleagues found that, in healthy individuals, QRS duration is slightly longer in males in comparison with females (103.2 msec vs 98.7 msec) [59], which is similar to the small differences we observed in this study.
The lack of effect of phenols on heart rate in the Fernald participants contrasts with previously reported acute stimulatory effect of bisphenols on ex vivo rodent heart rate and pacing rate of nodal hiPSC-CMs [16, 30, 31], and likely reflects the difference between chronic and acute phenol exposures. Further, the human heart rate is controlled by cardiac factors (such as sinus node automaticity) as well as extracardiac factors such as vagal tone, adrenaline secretion and sympathetic activity. The autonomic control of the heart rate in vivo may be another reason for the contrasting cohort vs experimental findings on this parameter.
We show here that higher exposure to some phenols was associated with alternations of ECG properties in the Fernald Cohort. Individual persons may experience more or less of an effect of BPA exposure on the ECG properties due to individual differences in absorption and metabolism of BPA. Regardless, it should be noted that these ECG changes at the population level are moderate and generally within the normal ECG parameter ranges. While these changes alone are unlikely to result in clinically significant cardiac electrical disease in healthy individuals, we believe that higher phenol exposure is a risk factor that can contribute to cardiac electrical abnormalities, particularly in subpopulations with existing pathophysiological conditions and predisposition. One such scenario is the combined effect of bisphenol-associated PR interval prolongation and BMI in females. As shown in Fig. 2, high exposure combined with high BMI can lead to sizable PR alteration in females. It is also possible that bisphenol-associated PR prolongation, particularly in high exposure individuals, can exacerbate existing AV block in patients with heart conduction disorders. Also, as noted above, subsetting the cohort into tertiles based on age reveals clue on the possible QT prolongation toxicity of BPA in the older subpopulation. Such a subpopulation-based toxicity assessment approach can better identify cardiac toxicities of environmental chemicals that would otherwise be difficult to identify in the general population.
The cross-sectional nature of the study allows us to determine association but not causality. Further, the temporal variability of BPA, and likely other phenols, is well known [60–63], and a one-time measurement of phenol levels does not represent lifetime exposure. However, even with these limitations, the FCC is highly valuable because it is unusual to have a large population-based cohort with simultaneous urine specimen collection and ECG recordings, which allowed us to determine the association of real-time (not lifetime) phenol exposure with ECG markers.
In conclusion, our population-based study shows, for the first time, that higher urinary levels of environmental phenols, including BPA, BPF, and TCC, were associated with sex-specific alterations in cardiac electrical-conduction properties in the Fernald Cohort. Our findings correlate with clinically relevant parameters that are associated with known pathologic conditions in humans. Our findings may have implication for the cardiac toxicity of these chemicals, particularly in predisposed subpopulations.