Plastics and cardiovascular health: phthalates may disrupt heart rate variability and cardiovascular reactivity

Phthalates are widely used in the manufacturing of consumer and medical products. In the present study, di-2-ethylhexyl-phthalate exposure was associated with alterations in heart rate variability and cardiovascular reactivity. This highlights the importance of investigating the impact of phthalates on health and identifying suitable alternatives for medical device manufacturing.

Abstract

Plastics have revolutionized medical device technology, transformed hematological care, and facilitated modern cardiology procedures. Despite these advances, studies have shown that phthalate chemicals migrate out of plastic products and that these chemicals are bioactive. Recent epidemiological and research studies have suggested that phthalate exposure adversely affects cardiovascular function. Our objective was to assess the safety and biocompatibility of phthalate chemicals and resolve the impact on cardiovascular and autonomic physiology. Adult mice were implanted with radiofrequency transmitters to monitor heart rate variability, blood pressure, and autonomic regulation in response to di-2-ethylhexyl-phthalate (DEHP) exposure. DEHP-treated animals displayed a decrease in heart rate variability (−17% SD of normal beat-to-beat intervals and −36% high-frequency power) and an exaggerated mean arterial pressure response to ganglionic blockade (31.5% via chlorisondamine). In response to a conditioned stressor, DEHP-treated animals displayed enhanced cardiovascular reactivity (−56% SD major axis Poincarè plot) and prolonged blood pressure recovery. Alterations in cardiac gene expression of endothelin-1, angiotensin-converting enzyme, and nitric oxide synthase may partly explain these cardiovascular alterations. This is the first study to show an association between phthalate chemicals that are used in medical devices with alterations in autonomic regulation, heart rate variability, and cardiovascular reactivity. Because changes in autonomic balance often precede clinical manifestations of hypertension, atherosclerosis, and conduction abnormalities, future studies are warranted to assess the downstream impact of plastic chemical exposure on end-organ function in sensitive patient populations. This study also highlights the importance of adopting safer biomaterials, chemicals, and/or surface coatings for use in medical devices.

NEW & NOTEWORTHY Phthalates are widely used in the manufacturing of consumer and medical products. In the present study, di-2-ethylhexyl-phthalate exposure was associated with alterations in heart rate variability and cardiovascular reactivity. This highlights the importance of investigating the impact of phthalates on health and identifying suitable alternatives for medical device manufacturing.

the development of plasticized polyvinyl chloride (PVC) materials has promoted versatility in medical device manufacturing, which has transformed hematological care, transfusion medicine, and cardiac surgery. Di-2-ethylhexyl-phthalate (DEHP) is a commonly used plasticizer that imparts flexibility to PVC medical devices (24). Plasticized PVC products allow for the use of individual sterile storage containers and tubing, which has reduced the risk of infection, improved stability, and increased shelf life. DEHP-containing blood storage bags are also associated with a reduced incidence of red blood cell hemolysis. The latter has been attributed to DEHP chemical migration out of the plastic material and into the stored blood and stabilization of red blood cell membranes (2).

Since plasticizers are not covalently bound to the PVC polymer, they are highly susceptible to leaching when in contact with lipophilic solutions (i.e., blood, nutrition solutions, and medication formulation aids) (47, 68). Consequently, patients can be exposed to high phthalate doses, particularly when undergoing multiple medical interventions that use plasticized PVC products (i.e., storage bags containing blood, plasma, intravenous fluids, total parenteral nutrition, tubing associated with their administration, nasogastric tubes, enteral feeding tubes, catheters, extracorporeal membrane oxygenation circuits, hemodialysis tubing, respiratory masks, and endotracheal tubes) (34, 35, 39, 48, 62, 67). Moreover, pediatric patients <1 yr old can be subjected to prolonged phthalate exposure due to an underdeveloped glucuronidation pathway and slowed excretion. As a result, patient exposure can exceed safe levels by 4,000–160,000 times (13, 17, 49).

Emerging epidemiological studies have shown correlations between phthalate exposure and human health concerns, including neurological disorders and cardiovascular disease (14, 22, 32, 58, 76, 78). The latter includes an association between phthalate plasticizer exposure and increased systolic blood pressure (SBP) in adolescents (83, 84) and an increased coronary risk in elderly populations (46, 60). Additionally, our laboratory has previously shown that phthalate exposure adversely affects the electrical activity and contractility of cardiac muscle cells (30, 64–66). To further address the potential clinical impact of phthalates on health, we used an in vivo radiotelemetry system in mice to monitor the direct effect of phthalates on cardiovascular [blood pressure (BP) and heart rate (HR)] and autonomic physiology [HR variability (HRV)]. Modifications in HRV indexes serve as an early and sensitive indicator of cardiovascular dysfunction, which is manifested by a decrease in parasympathetic (16, 31) and/or increase in sympathetic tone (50). A reduction in HRV often precedes the development of hypertension (73), atherosclerosis (72), and conduction abnormalities. HRV markers have also been associated with the Framingham risk score (92), coronary events (20), and dysrhythmia (86). In the present study, animals were monitored under both resting conditions and during and after Pavlovian fear conditioning, which provokes a sympathetically mediated cardiovascular response (19, 44, 77, 85). We hypothesized that phthalate exposure alters autonomic regulation of the cardiovascular system, which is evidenced by a decrease in basal HRV and exaggerated cardiovascular reactivity to a conditioned stressor.

MATERIALS AND METHODS

Animal protocol and experimental design.

All animal procedures were reviewed and approved by the Animal Care and Use Committee at the George Washington University and Children’s National Health System. Male 12-wk-old C57BL/6 mice were purchased from Hilltop Laboratory Animals. Mice were housed individually in ventilated acrylic cages with nesting material. Animals were fed a normal laboratory diet ad libitum. Room lights were cycled every 12 h: on from 0700 to 1900 hours and off from 1900 to 0700 hours. Baseline data were recorded for 2–3 days, and animals were then separated into two groups differing in drinking water supplementation: 1) Captisol (4 mg/ml) for vehicle control or 2) DEHP-Captisol (1 µg-4 mg/ml). The DEHP dose was chosen to fall within the range of clinical exposure for a pediatric patient undergoing medical treatment over an extended period of time (13, 32, 49). Open-field testing was performed using the Tru Scan system to assess locomotion.

Telemetry transmitters (n = 12, HDX-11, Data Sciences) were implanted surgically under aseptic conditions with the blood pressure transducer placed in the left carotid artery and fed inferiorly. Electrodes were placed on the right atrium and apex to produce a lead II ECG. Animals were anesthetized with ketamine-xylazine via intraperitoneal injection; depth of anesthesia was assessed frequently via toe pinch. Body temperature was regulated using a feedback-controlled heating pad. Mice recovered from surgery for 7–10 days before baseline data collection was started. Cardiovascular reactivity testing to fear conditioning was performed after 4-wk exposure. HR and BP recordings were collected for continuous segments for 3–4 days/wk to conserve transmitter battery power for the total 6-wk experimental protocol. At the end of the study, animals were anesthetized with ketamine-xylazine and then euthanized by exsanguination after heart excision.

Telemetry data acquisition and analysis of HRV.

Telemetry data, including ECG lead II, arterial BP, activity, and temperature, were acquired using DataQuest ART (Data Sciences) and analyzed using ecgAUTO (Emka Technologies). From the BP waveform, the following parameters were determined: diastolic, systolic, mean, and dP/dt. R waves were detected from the ECG to determine the mean RR interval and HR. Each hour, HR and BP signals were sampled and analyzed [120-s continuous segment (80)], resulting in 12 data points per daytime (12 h) and nighttime (12 h). Daytime and nighttime data were averaged across the 12-h timeframe and displayed with two representative data points per week of the 6-wk study, similar to other analysis methods for long-term telemetry recordings (58, 80).

For HRV analysis, an interbeat-interval waveform was calculated for 500 continuous normal beats (~30–60 s of recording), as previously described (8, 10). The interbeat-interval waveform was further processed by cubic spline interpolation and then through fast Fourier transform to determine high-frequency (HF) power between 2.5 and 5 Hz, a suggested bandwidth that is predominantly parasympathetic activity for mouse HRV analysis (6). HF power was normalized by dividing the squared mean RR interval (in s) at that input epoch, a common transform to compensate for the nonlinear relationship between HR and RR (8, 69). Low-power and low-frequency-to-HF ratio were not investigated due to concerns of accuracy (9, 51, 57), particularly for long-term in vivo studies. The SD of normal beat-to-beat intervals (SDNN) was calculated based on guidelines for the measurement of overall HRV (33). Similarly to HF power, SDNN was normalized by dividing by the mean RR interval (in s) of the input epoch (8, 69).

Conditioned cardiovascular and autonomic reactivity to Pavlovian fear conditioning.

Enhanced sympathetic nerve activity is involved in the development and progression of hypertension (37, 43). Indeed, spontaneously hypertensive animals display exaggerated cardiovascular reactivity and prolonged BP elevation (44) in response to fear conditioning due to heightened sympathetic neural activity (37). Therefore, we used a similar Pavlovian fear conditioning paradigm (53) to provoke a conditioned cardiovascular response that is sympathetically mediated (23, 37, 41, 71). Cue-dependent fear conditioning consists of pairing a conditioned stimulus (CS; audible tone) and an unconditioned stimulus (US; mild foot shock) over five consecutive trials. After repeated CS-US pairings, an association is formed such that the CS alone elicits the same behavioral and autonomic responses (conditioned responses) formerly produced by the US (44). After fear conditioning (>22 h), cardiovascular reactivity to the conditioned cue was evaluated in a novel test chamber. For this study, we focused on the cardiovascular components (HR, HRV, and BP), which are mediated by the autonomic nervous system (14, 19, 44, 59, 77).

After 4 wk of exposure, mice underwent auditory fear conditioning (30 min, Coulbourn Instruments). This included five pairings of the CS (audible tone: 30 s, 6 kHz, 75 db) coterminating with an US (mild foot shock: 500 ms, 0.6 mA) (53). Poincaré plots of RR intervals were produced for each animal after the first tone shock pair as a measure of HRV during a provoked sympathetic response. Poincaré plots were averaged by treatment group (control vs. DEHP treated) to produce ensemble means, from which RR variability was quantitated by the SD from the major (SD1) and minor axis (SD2) (38). After the fear conditioning protocol, continuous BP and ECG measurements were collected for ~22 h in the homecage environment (recovery). The next day, cardiovascular reactivity to the CS (tone) was performed in the absence of the US (shock). BP and ECG were continuously monitored. After cardiovascular reactivity testing, animals were returned to their homecage to continuously monitor BP and ECG over 22 h (recovery).

Ganglionic blockade via chlorisondamine administration.

To measure the contribution of the sympathetic nervous system on the cardiovascular system, animals were treated with the ganglionic blocker chlorisondamine (12 mg/kg ip) (7, 29, 88, 91). Blockage of nicotinic acetylcholine receptors in the autonomic ganglia attenuates sympathetic input to arteries, resulting in vasodilation and a decrease in BP (18). After animals underwent treatment with control or DEHP-supplemented water (6 wk, end of study), chlorisondamine was administered and the magnitude of sympathetic input was determined by the fall in mean arterial pressure (MAP) after ganglionic blockade.

Mass spectrometry to quantitate plasma phthalate concentration.

At the end of the treatment study (6 wk), blood samples were collected via cardiac puncture and anticoagulated. Plasma samples were isolated and then analyzed at the Bioanalytical Laboratory at the School of Pharmacy, Virginia Commonwealth University. Samples were subjected to a high-throughput solid phase extraction and analyzed using liquid chromatography-tandem mass spectrometry/mass spectrometry, as previously described (79). Briefly, samples were subjected to reversed phase chromatography using isocratic conditions (85% acetonitrile and 15% of 0.05% acetic acid). An Xterra 2.1 × 100-mm, 3.5 μM HPLC column (Waters, Milford, MA) was used. Concentrations of DEHP and its primary metabolite mono-2-ethylhexylphthalate (MEHP) were quantitated. Replicate analyses of spiked samples containing multiple concentrations of the target molecules were used to generate a calibration curve (DEHP: r² = 0.992644 and MEHP: r² = 0.998872).

Gene expression analysis of cardiac tissue.

At the end of the treatment study (6 wk), total RNA was isolated from heart tissue using an RNeasy fibrous tissue kit (Qiagen, Germantown MD) per the manufacturer’s instructions. RNA concentration and quality were assessed via 260:280-nm absorbance readings, and total RNA was reverse transcribed using a SuperScript VILO cDNA synthesis kit (ThermoFisher Scientific, Waltham, MA). The resulting cDNA was used as a template for real-time quantitative PCR (qPCR) using Taqman Gene Expression assays (ThermoFisher Scientific, Waltham, MA) for GAPDH, angiotensin-converting enzyme (ACE), ACE2, angiotensin II type 1 receptor, α_1A-adrenergic receptor, β₂-adrenergic receptor, endothelin-1, endothelin receptor-α, endothelin receptor-β, and nitric oxide synthase 3 (NOS3). qPCR was performed using the CFX96 detection system (Bio-Rad, Hercules, CA). Quantification and normalization of relative mRNA gene expression was accomplished using the comparative threshold cycle (C_T) method (ΔΔC_T); ΔΔC_T values were converted to ratios by $2^{{}^{–\Delta \Delta {\text{C}}_{\text{T}}}}$ averaged across replicates.

Data presentation and statistical analysis.

Data normality was confirmed by Shapiro-Wilk test, and comparisons between sample means were performed using two-sample Student’s t-test or Student’s t-test with multiple comparisons as needed (R statistical package). Within-group analyses were performed as the comparison between baseline and week 6 by repeated-measures one-way or two-way ANOVA. Sidak-Bonferroni methodology was used to correct for multiple comparisons (Prism, Graphpad). The level of statistical significance was defined as P < 0.05. All results are reported as means ± SE.

RESULTS

Animal characteristics and phthalate exposure.

Mice exposed to DEHP-supplemented drinking water achieved a circulating MEHP concentration of 403 ± 28.8 ng/ml compared with 0.922 ± 0.14 ng/ml in control animals (Table 1). Phthalate exposure did not alter average water intake, body weight, or heart weight (Table 1). Analysis of animal movement within the cage was also assessed to ensure that heart mass, HR, or BP measurements under basal conditions were not confounded by hyperactivity or lethargy. There were no differences in the number of entries to center, time spent in cage center, or total travel distance from the 30-min Tru Scan session (Table 1).

Table 1.

Animal characteristics

	Control	DEHP	P Value
MEHP serum concentration, ng/ml	0.922 ± 0.14	403 ± 28.8	<0.001*
Heart mass, mg	170 ± 8.1	166 ± 7.1	0.709
Heart mass/body mass, mg/g	5.41 ± 0.3	5.34 ± 0.2	0.853
Body mass, g	31.6 ± 0.8	31.2 ± 1.1	0.525
Water consumption, ml/day	5.05 ± 1.0	5.34 ± 0.6	0.651
Tru Scan
Number of center entries	84.3 ± 27	93.2 ± 30	0.561
Time in center, min	11.8 ± 4.1	12.3 ± 5.9	0.712
Distance traveled, m	29.3 ± 4.9	29.8 ± 8.6	0.653

All values reported as means ± SE. DEHP, di-2-ethylhexyl-phthalate. Mono-2-ethylhexyl-phthalate (MEHP) serum concentration was measured by mass spectrometry. All values were taken at the end of the study except for water consumption, which was averaged throughout the study. Tru Scan activity measurements were from a 30-min session.

^*P < 0.05.

Exaggerated cardiovascular reactivity and poststress recovery in DEHP-exposed animals.

Cardiovascular reactivity testing was assessed after 4 wk of DEHP exposure to model prolonged clinical exposure resulting from multiple medical interventions. To measure the impact of DEHP exposure on autonomic function, we monitored cardiovascular response (HR, BP, and HRV) during and after (recovery) a fear-conditioning protocol (Fig. 1). As shown by others (19, 44, 85), this paradigm elicits a strong cardiovascular and autonomic response to a conditioned cue (tone). Cardiovascular reactivity to stress is an early and sensitive predictor of future pathologies (56, 76); therefore, we hypothesized that an imbalance in autonomic regulation after DEHP exposure would initially present as exaggerated cardiovascular reactivity and prolonged poststress recovery of BP (19, 37, 44).

Fig. 1.

Cardiovascular (CV) reactivity to Pavlovian fear conditioning and poststress recovery (4 wk). A: Pavlovian fear conditioning (CV reactivity) protocol timeline. B: mean systolic blood pressure (SBP) monitored throughout the course of the CV reactivity protocol (30 min). A significant elevation in SBP was observed in the di-2-ethylhexyl-phthalate (DEHP)-treated group during both recovery periods after CV reactivity testing. All values are reported as means ± SE. *P < 0.05 (two-way ANOVA). C and D: Poincaré plots were computed by the first 500 beats after the first tone/shock pair during fear conditioning (the first CV reactivity session) and ensemble averaged for the control and DEHP-treated group. DEHP-treated animals have reduced variability along the major axis (SD2). CS, conditioned stimulus; US, unconditioned stimulus.

At rest and before cardiovascular reactivity testing (4-wk treatment), no difference in SBP or HR was observed between treatment groups (114 ± 9 mmHg and 435 ± 25 beats/min for control animals and 115 ± 9.5 mmHg and 418 ± 24 beats/min for DEHP-treated animals), although HRV was significantly lower in DEHP-treated animals (HF power; Table 2). HRs were comparable during cardiovascular reactivity testing (control: 742 ± 25 beats/min vs. DEHP: 730 ± 38 beats/min). Compared with basal levels in the homecage, HRV parameters diminished significantly after initiation of the fear-conditioning protocol (tone + shock pair) in both control and DEHP-treated animals, thus highlighting the link between stress and sympathetic activity (Table 2). However, DEHP-treated animals displayed significantly lower SD2 (−56%, P < 0.05) compared with control animals, as computed by Poincaré (Fig. 1C), indicating a more intense reduction in HRV (40).

Table 2.

HR variability during cardiovascular reactivity testing (4-wk DEHP exposure)

	Homecage Baseline (4-wk exposure)			Cardiovascular Reactivity Test 1 (4-wk exposure)
	Control	DEHP	P value	Control	DEHP	P value
HR, beats/min	431 ± 35	413 ± 31	0.629	780 ± 13	794 ± 15	0.279
RR, ms	139 ± 6.8	145 ± 7.5	0.642	76.9 ± 0.8	75.5 ± 0.5	0.279
SD1, ms	5.4 ± 0.9	3.7 ± 0.5	0.270	0.63 ± 0.08	0.63 ± 0.1	0.971
SD2, ms	17.2 ± 2.2	18.3 ± 3.3	0.832	3.19 ± 0.2	1.79 ± 0.2*	0.020
HF power, normalized units	353 ± 48	194 ± 23*	0.0001	47.7 ± 17	51.3 ± 28.8	0.935

All values reported as means ± SE. RR and heart rate (HR) variability parameters were computed from a 500-beat segment during a homecage basal period and again after the initiation of the first conditioned stimulus (tone)-unconditioned stimulus (shock) pair (cardiovascular test 1). High-frequency (HF) power was significantly lower in di-2-ethylhexyl-phthalate (DEHP)-treated animals before reactivity testing. SD from the major (SD1) and minor axis (SD2) were computed from Poincare plots computed in the same timeframe. SD2 was significantly lower (P < 0.05) in the DEHP-treated group than in the control group during the cardiovascular test, indicating heightened reactivity.

^*P < 0.05.

After the 30-min reactivity test, animals were returned to their home cage, and SBP and HR were continuously monitored during the 22-h recovery period. As shown in Fig. 1, DEHP-treated animals exhibited an exaggerated and delayed poststress SBP recovery (+13% over 22 h; Fig. 1B), which may be attributed to enhanced sympathetic activity. After recovery, cardiovascular reactivity was measured in response to a conditioned stimulus (audible tone). In DEHP-treated animals, cardiovascular reactivity remained significantly elevated during the day/night recovery period (+20% over 22 h; Fig. 1B). SBP returned to baseline in both groups 72 h after completion of the cardiovascular reactivity protocol. Taken together, these data suggest that DEHP-treated animals display an exaggerated response to stress that is mediated by sympathetic dominance (19, 70).

DEHP exposure increases sympathetic dominance and decreases HRV.

Alterations in cardiovascular reactivity to stress suggest an imbalance in autonomic regulation in DEHP-treated animals. To further address the impact of DEHP exposure on autonomic regulation, we analyzed HRV indexes under basal resting conditions. Time-domain measurements by SDNN showed a decrease in HRV in DEHP-treated animals (78 ± 4.3) compared with control (95 ± 4.4) at the end of the study (6 wk, P < 0.05; Fig. 2A). As a marker for parasympathetic innervation on the heart, HF power was computed and corrected for mean RR interval. HF power was also lower in the DEHP-treated group compared with the control group at rest (6 wk, P < 0.05; Fig. 2B). Raw HF power in the DEHP-treated group was 6.31 ms² compared with 9.53 ms² in the control group. After RR correction, normalized HF power was computed as 585 ± 117 in the control group and 374 ± 61 in the DEHP-treated group (normalized units).

Fig. 2.

Measurement of heart rate variability and sympathetic tone under basal resting conditions (6 wk, end of study). A: SD of normal beats (SDNN) was corrected for mean RR interval. A significant decrease in SDNN was found at the end of the study in the di-2-ethylhexyl-phthalate (DEHP)-treated group compared with the control (Ctrl) group. B: high-frequency (HF) power was computed by power in the band 2.5–5 Hz corrected by squared mean RR interval. A significant decrease in HF power was found in the DEHP-treated group at 6 wk compared with the Ctrl group. C: after treatment with chlorisondamine (12 mg/kg), mean arterial pressure (MAP) of the DEHP-treated group decreased significantly compared with the Ctrl group, indicating an elevated state of sympathetic tone (means ± SE). All values reported are as means ± SE. *P < 0.05.

DEHP exposure increases sympathetic tone on the cardiovascular system.

To directly investigate the contribution of the sympathetic nervous system on the cardiovascular system, animals were treated with the ganglionic blocker chlorisondamine at the end of the study (6 wk). Blockage of nicotinic acetylcholine receptors in the autonomic ganglia attenuates sympathetic input to arteries, resulting in vasodilation and a decrease in BP (7, 18, 29, 91). After administration of chlorisondamine (30 min), the magnitude of sympathetic input was measured by the resulting change in MAP. Before drug administration, control and DEHP-treated groups had initial HRs of 350 and 360 beats/min, respectively (P = 0.31), and an average initial MAP of 101 ± 14 mmHg for the control group and 117 ± 3 mmHg for the DEHP-treated group. After administration of chlorisondamine, DEHP-treated animals displayed greater sympathetic tone under basal conditions, as measured by an exaggerated MAP response to ganglionic blockade (−50.9 mmHg ± 6.0 in the DEHP-treated group vs. −31.6 mmHg ± 0.6 in the control group, P < 0.05; Fig. 2C). After normalization of the initial MAP, the DEHP-treated group displayed a −69.75 ± 4.13% drop in MAP compared with −38.24 ± 7.85% in the control group.

Effect of DEHP exposure on resting HR and BP.

Heightened cardiovascular reactivity and reduced HRV are early markers of cardiovascular dysfunction, which can be used to predict future pathologies (56, 76). After 6 wk of treatment, basal SBP was slightly increased in DEHP-treated animals (125 ± 4.9 mmHg) compared with control animals (116 ± 1.4 mmHg), although this parameter did not reach significance (P = 0.09; Fig. 3F). No significant difference in basal resting HR was detected between DEHP-treated (490 ± 5 beats/min, nighttime) and control animals (508 ± 3.1 beats/min; Fig. 3E). These results were not unexpected, as the timespan required to develop hypertension may exceed the time limitation of a continuous radiotelemetry monitoring protocol.

Fig. 3.

Heart rate (HR) and systolic blood pressure (SBP) under basal resting conditions. SBP (A–C) and HR measurements (D–F) were separated by time of day. SBP decreased marginally in the control (Ctrl) group to 96–97% of the baseline value and increased marginally in the di-2-ethylhexyl-phthalate (DEHP)-treated group to 105–109% of the baseline value (F). No significant difference in HR was observed between groups (F). All values are reported as means ± SE. BPM, beats/min.

DEHP exposure alters cardiac tissue gene expression.

To identify potential mechanisms contributing to HRV alterations and poststress BP recovery, we investigated genes associated with the renin-angiotensin system, BP regulation, and autonomic balance. We detected elevated levels of NOS3 mRNA in the cardiac tissue of DEHP-treated animals (6-wk exposure; Fig. 4), which may serve as a compensatory effect to offset sympathetically mediated stimulation of β-adrenergic signaling (3, 54, 89).

Fig. 4.

Gene expression profiling of cardiac tissue. Shown are gene expression levels (ΔΔC_T, where C_T is threshold cycle) of cardiovascular genes involved in blood pressure regulation [angiotensin-converting enzyme (ACE), ACE2, and angiotensin II type 1 receptor (AGTR1)], autonomic receptors [α_1A-adrenoceptor (ADRA1A) and β₂-adrenoceptor (ADRB2)], cardiac contractility [nitric oxide synthase 3 NOS3)], and vasoconstriction [endothelin-1 (EDN1), endothelin receptor type A receptor (EDNRA), and endothelin receptor type B receptor (EDRNB)]. Ctrl, control; DEHP, di-2-ethylhexyl-phthalate. All values are reported as means ± SE. *P < 0.05.

We also observed an upregulation in genes associated with hypertension, including ACE and endothelin-1, in isolated cardiac tissue from DEHP-treated animals (Fig. 4). In the renin-angiotensin system, ACE converts angiotensin I to the active vasoconstrictor angiotensin II, which exerts its actions through interactions with the angiotensin II type 1 receptor. Increased ACE expression could predispose DEHP-treated animals to elevated serum angiotensin II, provided that ACE enzymatic activity has been positively correlated with the angiotensin II serum level (27). No significant alterations in ACE2 or angiotensin II type 1 receptors were observed. Endothelin-1 is also a potent vasoconstrictor, which exerts its hypertensive effects through interaction with endothelin type A receptor. Upregulation of both endothelin-1 and endothelin type A receptors was found in DEHP-treated samples.

We did not detect a difference in the expression of genes associated with adrenergic receptors, including the α_1A-adrenergic receptor, β₁-adrenergic receptor, or β₂-adrenergic receptor, although we cannot rule out alterations in adrenergic receptor protein expression, activity, and/or sensitivity, which were beyond the scope of this study.

DISCUSSION

For the first time, we have shown that phthalate exposure alters autonomic regulation, resulting in decreased HRV, increased sympathetic tone, and exaggerated cardiovascular reactivity and recovery to a stress stimulus. After DEHP treatment, time- and frequency-domain measurements of HRV were reduced, indicating impaired parasympathetic activity. DEHP-treated animals also exhibited an increase in total sympathetic tone, as determined by the exaggerated MAP response to ganglionic blockade. In response to conditioned stress cue, DEHP-treated animals displayed a more significant reduction in HRV and exhibited sustained SBP elevation over a 22-h recovery period. Phthalate-treated animals also had increased endothelin-1 and ACE cardiac expression, which can predispose animals to vasoconstriction, increased vascular resistance and autonomic dysfunction. Overall, our study supports our previous findings that phthalate plasticizer exposure has a profound effect on cardiovascular function (30, 64, 65).

For the described experiments, we focused our attention on early and sensitive indicators of cardiovascular and autonomic dysfunction, including HRV, sympathetic tone on the vasculature, and the magnitude of the cardiovascular response to stress. Alterations in autonomic balance can initially present as a change in HRV, even in normotensive subjects who later develop hypertension (36). Short-term HRV is dominated by respiratory sinus arrhythmia, which manifests as parasympathetic input on the sinus node. In the present study, we found a reduction in parasympathetic input on the heart, as evidenced by a decrease in HF power (2.5–5 Hz) and SDNN of the interbeat intervals, a more significant reduction in HRV during cardiovascular reactivity testing, and prolonged BP elevation. Cardiovascular reactivity to a conditioned stimulus is mediated by mass sympathetic discharge, a resulting release of catecholamines, and an increase in HR and total peripheral resistance through stimulation of adrenergic receptors in the heart and blood vessels (81). DEHP-treated animals displayed an exaggerated cardiovascular response to a conditioned stimulus, with SBP remaining 10 mmHg higher in treated animals after cardiovascular reactivity testing. We speculate that delayed BP recovery is attributed to sympathetic dominance, reduced β-adrenergic sensitivity, and/or norepinephrine spillover (19, 44, 70). Similar to HRV indexes, an exaggerated cardiovascular response to a stressor has been shown to predict future hypertension risk (55, 74, 76). Finally, abnormal autonomic nervous system cardiac control is also associated with anxiety, and reduced HRV has been consistently observed in patients with anxiety disorders (11, 28). This highlights the extensive influence of autonomic regulation on multiple disease pathologies, including cardiovascular function, behavior, and anxiety, which warrant future study in relation to phthalate exposure.

In the present study, we observed a slight elevation in SBP in DEHP-treated animals under resting basal conditions, although this trend was not significant over the 6-wk timeframe (P = 0.09). Phthalate-treated animals also displayed an upregulation of ACE and endothelin-1 in cardiac tissue, genes that encode ACE and endothelin-1, respectively. ACE is responsible for converting angiotensin I to angiotensin II, resulting in elevated vasoconstrictor activity and enhanced sympathetic drive. Importantly, we did not see a change in the expression of ACE2, the gene responsible for encoding the protein for converting angiotensin II to inactive angiotensin 1–7, which would counter the hypertensive effects of ACE1 (21). Endothelin-1 is also known for its vasoconstrictive effects and has been shown to decrease cardiac output due to increased systemic vascular resistance (1, 87). Recent studies have also highlighted the role of endothelin-1 in the interaction between sympathetic neurons and cardiac myocytes (45). Elevated levels of endothelin-1 have been observed in various pathological conditions, including heart failure, myocardial infarction, kidney disease, and hypertension (45). Increased endothelin-1 and ACE expression in phthalate-treated animals may predispose these animals to vasoconstriction and increase vascular resistance.

Epidemiological studies have established an association between DEHP exposure and a mild increase in SBP in adolescents (83, 84), estimating an approximate 1-mmHg increase with a threefold increase of DEHP exposure (84). Studies have also shown a link between elevated SBP and exposure to DEHP alternatives, including diisononyl phthalate (DINP) and diisodecyl phthalate (DIDP) (83). Moreover, ex vivo and in vitro studies have identified links between phthalate exposure and alterations in cardiac contractility and vasculature function. Labow et al. (42) showed that MEHP has a hypertensive effect on the pulmonary vasculature and a negative inotropic effect on isolated human atrial trabeculae (4). Similarly, our group has previously shown that phthalate exposure can alter Ca²⁺ handling and contractility in rodent and human cardiomyocytes (64, 65). In the present study, we detected an increase in NOS3 gene expression in cardiac tissue, which may attenuate β-adrenergic effects via cGMP-dependent mechanisms (3, 12, 54, 89). The latter may serve as a compensatory effect to offset stimulation of β-adrenergic signaling.

Followup studies are needed to pinpoint the underlying mechanisms responsible for the observed effects on autonomic and cardiovascular physiology. Phthalate esters, including DEHP, are endocrine disruptors that interfere with hormone homeostasis (15). Previous studies have shown that DEHP exposure reduces circulating levels of aldosterone and testosterone (52), increases adrenocorticotropin and corticosterone levels (78), and results in widespread disruption of steroidogenesis. Such alterations can contribute to autonomic imbalance and diminished HRV (61, 90). In addition, phthalates can cause a wide range of nonendocrine-related toxicities (49), including inflammation and oxidative stress (25, 26, 68). The underlying effects of DEHP on autonomic and cardiovascular function are likely to be multifactorial.

Limitations.

The present study was limited by a small sample size in a mouse model. Cardiotoxicity studies performed in mice may not yield the same results in humans, which is due in part to species differences in HR, size, and ion channel expression. To remedy high HRs and artifacts that are common in mouse radiotelemetry protocols, HR and BP data analysis was limited to 2-min continuous recordings, as previously described (80). In addition, our method of administration (oral via water supplementation) resulted in a circulating MEHP plasma level (0.403 µg/ml) that is 40 times less than maximum levels measured in pediatric patients (15.6 µg/ml) (5, 75). Additional dose-response studies are necessary to accurately predict patient safety after high clinical exposures.

Conclusion.

For the first time, we have identified a relationship between phthalate exposure, altered autonomic regulation, and cardiovascular function under basal and stress conditions (conditioned cardiovascular reactivity). HRV indexes are often used as a sensitive early indicator of cardiovascular dysfunction, serving as a precursor to clinical manifestations of hypertension, atherosclerosis, and conduction abnormalities. Sufficient HRV is considered part of the healthy interplay between cardiovascular and nervous systems, providing adaptation and regulatory capacity. HRV indexes are sensitive indicators of cardiovascular and autonomic health and are often used to predict patient recovery from intensive care, extracorporeal membrane oxygenation, and cardiopulmonary bypass. Future studies are needed to examine the risk of phthalate exposure to individuals, particularly in high-risk, high-exposure scenarios.

GRANTS

This work was supported by National Institutes of Health Grants R00-ES-023477, R00-HL-10767, UL1-TR-000075, and KL2-TR-000076, the Children’s Research Institute (Divisions of Cardiology and Cardiac Surgery, to N. G. Posnack), and American Heart Association Grant 15CSA2430001.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.J., M.S., and N.G.P. analyzed data; R.J., A.S., P.M., and N.G.P. interpreted results of experiments; R.J., M.S., and N.G.P. prepared figures; R.J., P.M., and N.G.P. drafted manuscript; R.J., A.S., P.M., and N.G.P. edited and revised manuscript; R.J., A.S., M.S., N.M., P.M., and N.G.P. approved final version of manuscript; A.S., M.S., N.M., and N.G.P. performed experiments; P.M. and N.G.P. conceived and designed research.