Cardiac effects of seasonal ambient particulate matter and ozone co-exposure in rats

Abstract

Background: The potential for seasonal differences in the physicochemical characteristics of ambient particulate matter (PM) to modify interactive effects with gaseous pollutants has not been thoroughly examined. The purpose of this study was to compare cardiac responses in conscious hypertensive rats co-exposed to concentrated ambient particulates (CAPs) and ozone (O3) in Durham, NC during the summer and winter, and to analyze responses based on particle mass and chemistry.

Methods: Rats were exposed once for 4 hrs by whole-body inhalation to fine CAPs alone (target concentration: 150 μg/m3), O3 (0.2 ppm) alone, CAPs plus O3, or filtered air during summer 2011 and winter 2012. Telemetered electrocardiographic (ECG) data from implanted biosensors were analyzed for heart rate (HR), ECG parameters, heart rate variability (HRV), and spontaneous arrhythmia. The sensitivity to triggering of arrhythmia was measured in a separate cohort one day after exposure using intravenously administered aconitine. PM elemental composition and organic and elemental carbon fractions were analyzed by high-resolution inductively coupled plasma-mass spectrometry and thermo-optical pyrolytic vaporization, respectively. Particulate sources were inferred from elemental analysis using a chemical mass balance model.

Results: Seasonal differences in CAPs composition were most evident in particle mass concentrations (summer, 171 μg/m3; winter, 85 μg/m3), size (summer, 324 nm; winter, 125 nm), organic:elemental carbon ratios (summer, 16.6; winter, 9.7), and sulfate levels (summer, 49.1 μg/m3; winter, 16.8 μg/m3). Enrichment of metals in winter PM resulted in equivalent summer and winter metal exposure concentrations. Source apportionment analysis showed enrichment for anthropogenic and marine salt sources during winter exposures compared to summer exposures, although only 4% of the total PM mass was attributed to marine salt sources. Single pollutant cardiovascular effects with CAPs and O3 were present during both summer and winter exposures, with evidence for unique effects of co-exposures and associated changes in autonomic tone.

Conclusions: These findings provide evidence for a pronounced effect of season on PM mass, size, composition, and contributing sources, and exposure-induced cardiovascular responses. Although there was inconsistency in biological responses, some cardiovascular responses were evident only in the co-exposure group during both seasons despite variability in PM physicochemical composition. These findings suggest that a single ambient PM metric alone is not sufficient to predict potential for interactive health effects with other air pollutants.

Figure 1

A map of superimposed backward trajectories using the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT) at the Real-Time Environmental Applications and Display sYstem (READY) website developed by the Air Resources Laboratory of the National Oceanic and Atmospheric Administration. Archived meteorological data was utilized to model the direction and location of the test air mass for the previous 24 hr before 10 am local time (2 pm UTC (Universal Time Coordinated)) of each exposure day. Each trajectory represents a different air mass for each winter exposure day (2/28/12, and 2/29/12, and 3/07/12 and blue in color) and each summer exposure day (8/17/11 and 8/18/11 and red in color) and terminates at the exposure facility in A-Building of the US EPA campus in Durham, NC (indicated by a star). Triangle symbols indicate time of day (UTC) in 6 hour increments with larger symbols corresponding to midnight UTC (8 pm local time) in each path.

Figure 2

Pie Charts illustrating sources contributing to the concentrated ambient particulate (CAPs) and CAPs + ozone (O₃) mixtures during the summer and winter exposures. Sources for each season were quantified using the EPA Chemical Mass Balance Model.

Figure 3

Mean change in Heart rate (HR) from pre-exposure values during summer and winter exposures. HR values for each animal at each time point during exposure or after exposure were subtracted from corresponding time-matched pre-exposure baseline data, which was recorded while the animals were either in the chamber (for “during exposure” data) or in their home cages (for “after exposure” data). Values represent mean change in HR in beats per minute ± standard error of the mean (n = 6). a - significantly less than filtered air control (p < 0.05). b - significantly greater than filtered air control (p < 0.05).

Figure 4

Mean change in PR interval from pre-exposure values during summer and winter exposures. PR values for each animal at each time point during exposure or after exposure were subtracted from corresponding time-matched pre-exposure baseline data, which was recorded while the animals were either in the chamber (for “during exposure” data) or in their home cages (for “after exposure” data). Values represent mean change in PR interval in msec ± standard error of the mean (n = 6). a - significantly less than filtered air control (p < 0.05). b - significantly greater than filtered air control (p < 0.05).

Figure 5

Mean change in QT_c interval from pre-exposure values during summer and winter exposures. QT_c values for each animal at each time point during exposure or after exposure were subtracted from corresponding time-matched pre-exposure baseline data, which was recorded while the animals were either in the chamber (for “during exposure” data) or in their home cages (for “after exposure” data). Values represent mean change in QT_c interval in msec ± standard error of the mean (n = 6). b - significantly greater than filtered air control (p < 0.05).

Figure 6

Mean change in SDNN from pre-exposure values during summer and winter exposures. SDNN values for each animal at each time point during exposure or after exposure were subtracted from corresponding time-matched pre-exposure baseline data, which was recorded while the animals were either in the chamber (for “during exposure” data) or in their home cages (for “after exposure” data). Values represent mean change in SDNN in msec ± standard error of the mean (n = 6). a - significantly less than filtered air control (p < 0.05).

Figure 7

Cumulative dose of infused aconitine necessary to trigger ventricular premature beats (VPB), ventricular tachycardia (VT), ventricular fibrillation (VF), and cardiac arrest (CA) in rats one day after a single exposure. Values represent mean dose ± standard error of the mean (n = 5). a - significantly less than filtered air control (p < 0.05).

Figure 8

N-acetyl B-D glucosaminidase, lactate dehydrogenase, CuZn superoxide dismutase, and glutathione S-transferase in lung lining fluid one day after summer or winter exposures to concentrated ambient particulate (CAPs), ozone (O₃), CAPs + O₃, or filtered air. Bars represent means ± SEM for each marker shown (n = 6/group). a - significantly less than filtered air control p < 0.05). b - significantly greater than filtered air control (p < 0.05).

Figure 9

Schematic of concentrated ambient particulates (CAPs) and ozone (O₃) co-exposure system showing concentrator, O₃ generator, exposure chambers, and particulate matter (PM) and O₃ monitoring systems. Receivers were placed within each chamber to monitor electrocardiogram, heart rate and body temperature. Particle concentration and sizing were tracked in real-time using a scanning mobility particle sizer and an aerodynamic particle sizer. Additional aerosol monitors (DustTrak and P-Trak) were used to track PM levels in real-time.