Sodium nitrate

nPM collection and transfer into aqueous suspension. We collected nPM with a high-volume ultrafine particle (HVUP) sampler (Misra et al. 2002 ) at 400 L/min flow in Los Angeles City near the CA-110 Freeway. These aerosols represent a mix of fresh ambient PM mostly from vehicular traffic nearby this freeway (Ning et al. 2007 (link)). The HVUP sampler consists of an ultrafine particle slit impactor, followed by an after-filter holder. The nPM (diameter < 200 nm) was collected on pretreated Teflon filters (20 × 25.4 cm, polytetrafluoroethylene, 2 μm pore; Pall Life Sciences, Covina, CA). We transferred the collected nPM into aqueous suspension by 30 min soaking of nPM-loaded filters in Milli-Q deionized water (resistivity, 18.2 MW; total organic compounds < 10 ppb; particle free; bacteria levels < 1 endotoxin units/mL; endotoxin-free glass vials), followed by vortexing (5 min) and sonication (30 min). As a control for in vitro experiments with resuspended nPM, fresh sterile filters were sham extracted. Aqueous nPM suspensions were pooled and frozen as a stock at –20°C, which retains chemical stability for ≥ 3 months (Li N et al. 2003; Li R et al. 2009). For in vitro experiments, nPM suspensions were diluted in culture medium, vortexed, and added directly to cultures.
Animals and exposure conditions. The nPM suspensions were reaerosolized by a VORTRAN nebulizer (Vortran Medical Technology 1 Inc., Sacramento, CA) using compressed particle-free filtered air [see Supplemental Material, Figure S1 (doi:10.​1289/ehp.1002973)]. Particles were diffusion dried by passing through silica gel; static charges were removed by passing over polonium-210 neutralizers. Particle sizes and concentrations were continuously monitored during exposure at 0.3 L/min by a scanning mobility particle sizer (SMPS model 3080; TSI Inc., Shoreview, MN). The nPM mass concentration was determined by pre- and postweighing the filters under controlled temperature and relative humidity. Inorganic ions [ammonium (NH₄⁺), nitrate (NO₃^–), sulfate (SO₄^2–)] were analyzed by ion chromatography. PM-bound metals and trace elements were assayed by magnetic-sector inductively coupled plasma mass spectroscopy. Water-soluble organic carbon was assayed by a GE-Sievers liquid analyzer (GE-Sievers, Boulder, CO). Analytic details for nPM-bound species are given by Li R et al. (2009). Samples of the reaerosolized nPM were collected on parallel Teflon filters for electron paramagnetic resonance (EPR) analysis.
Mice (C57BL/6J males, 3 months of age) were maintained under standard conditions with ad libitum Purina Lab Chow (Newco Purina, Rancho Cucamonga, CA) and sterile water. Just before nPM exposure, mice were transferred from home cages to exposure chambers that allowed free movement. Temperature and airflow were controlled for adequate ventilation and to minimize buildup of animal-generated contaminants [skin dander, carbon dioxide (CO₂), ammonia]. Reaerosolized nPM or ambient air (control) was delivered to the sealed exposure chambers for 5 hr/day, 3 days/week, for 10 weeks. Mice did not lose weight or show signs of respiratory distress. Mice were euthanized after isoflurane anesthesia, and tissue was collected and stored at –80°C. All rodents were treated humanely and with regard for alleviation of suffering; all procedures were approved by the University of Southern California Institutional Animal Care and Use Committee.
EPR spectroscopy of nPM. The reaerosolized nPM was collected on filters (described above), which were inserted directly in the EPR quartz tube (Bruker EPR spectrometer; Bruker, Rheinstetten, Germany); spectra were measured at 22°C. The g-value was determined following calibration of the EPR instrument using DPPH (2,2-diphenyl-1-picrylhydrazyl) as a standard. The EPR signal for DPPH was measured and the corresponding g-value was calculated. The difference from the known g-value of 2.0036 for DPPH was then used to adjust the observed g-value for the sample.
Cell culture and nPM exposure. Hippocampal slices from postnatal day 10–12 rats were cultured 2 weeks in a humidified incubator (35°C/5% CO₂) (Jourdi et al. 2005 (link)) with nPM suspensions added for 24–72 hr of exposure. Primary neurons from embryonic day 18 rat cerebral cortex were plated at 20,000 neurons/cm² on cover slips coated with poly-d-lysine/laminin and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with B27, at 37°C in 5% CO₂ atmosphere (Rozovsky et al. 2005 (link)). Primary glial cultures from cerebral cortex of neonatal day 3 rats (F344) were plated at 200,000 cells/cm² in DMEM/F12 medium supplemented with 10% fetal bovine serum and 1% l-glutamine and incubated as described above (Rozovsky et al. 1998 (link)). For conditioned medium experiments, glial cultures were treated with 10 mg nPM/mL; after 24 hr, media were transferred by pipette to neuron cultures.
Neurite outgrowth and toxicity assays. After treatments, neurons were fixed in 4% paraformaldehyde and immunostained with anti–β-III-tubulin (1:1,000, rabbit; Sigma Chemical Co., St. Louis, MO); F-actin was stained by rhodamine phalloidin (1:40; Molecular Probes, Carlsbad, CA). A neurite was defined as a process extending from the cell soma of the neuron that was immunopositive for both β-III-tubulin (green) and F-actin (red). The length of neurites was measured using NeuronJ software (Meijering et al. 2004 (link)). Growth cones were defined by the presence of actin-rich filopodia and lamellipodia (Kapfhammer et al. 2007 ). Collapsed growth cones were defined as actin-rich neuritic endings in which filopodia and lamellipodia were indistinguishable. In neurite outgrowth and growth cone collapse assays, individual neurons were selected from two cover slips per condition; n is the total number of neurons analyzed per treatment. Cytotoxicity in slice cultures was assayed by lactate dehydrogenase (LDH) release to media and by cellular uptake of propidium iodide (PI) (Jourdi et al. 2005 (link)). Neuronal viability was assayed by Live/Dead Cytotoxicity Kit (Invitrogen, Carlsbad, CA) by computer-assisted image analysis of fluorescent images. Mitochondrial reductase was assayed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) at 585 nm in undifferentiated PC12 cells (Mosmann 1983 (link)). For viability assays, n is the total number of hippocampal slices analyzed (LDH release and PI uptake) or the total number of cell culture wells analyzed per condition.
Immunoblotting. Mouse hippocampi were homogenized using a glass homogenizer in cold lysis buffer as described by Jourdi et al. (2005) (link). After sample preparation, 20 μg protein was electrophoresed on 10% sodium dodecyl sulfate polyacrylamide gels, followed by transfer to polyvinylidene fluoride (PVDF) membranes. The PVDF membranes were blocked with 5% bovine serum albumin for 1 hr and probed with primary antibodies overnight at 4°C: anti-GluA1 (glutamate receptor subunit 1; 1:3,000, rabbit; Abcam, Cambridge, MA), anti-GluA2 (1:2,000, rabbit; Millipore, Billerica, MA), anti-PSD95 (1:1,000, mouse; Abcam), anti-synaptophysin (1:5,000, mouse; Stressgene; Enzo, Plymouth Meeting, PA), and anti-β-III tubulin (loading control; 1:15,000, rabbit; Sigma), followed by incubation with secondary antibodies (1:10,000) conjugated with IRDye 680 (rabbit, LI-COR Biosciences, Lincoln, NE) and IRDye 800 (mouse, LI-COR). Immunofluorescence was detected by infrared imaging (Odyssey, LI-COR).
Quantitative polymerase chain reaction (qPCR). Total cellular RNA was extracted from cerebral cortex of nPM-exposed mice and rat primary glia (Tri Reagent; Sigma), and cDNA (2 μg RNA; Superscript III kit; Invitrogen) was analyzed by qPCR, with primers appropriate for mouse (in vivo) or rat (in vitro). Genes examined by qPCR were CD14, CD68, CD11b, CD11c, GFAP (glial fibrillary acidic protein), IFN-γ (interferon-γ), IL-1α, IL-1, IL-6, and TNFα. Data were normalized to β-actin.
Statistical analysis. Data are expressed as mean ± SE. The numbers of individual measurements (n) are described above and listed in the figure legends. Single and multiple comparisons used Student’s t-test (unpaired) and one-way analysis of variance (ANOVA)/Tukey’s honestly significant difference, with statistical significance defined as p < 0.05.

We obtained daily counts of hospital admissions for the period 2000–2006 from billing claims of enrollees in the U.S. Medicare system. Because the Medicare data analyzed for this study did not include individual identifiers, we did not obtain consent from individuals. This study was reviewed and exempted by the Institutional Review Board at the Johns Hopkins Bloomberg School of Public Health.
Each billing claim contains the date of service, disease classification using International Classification of Diseases, 9th Revision (ICD-9) codes (Centers for Disease Control and Prevention 2008 ), age, and county of residence. We considered two broad classes of outcomes based on ICD-9 codes: urgent or emergency cardiovascular admissions and urgent or emergency respiratory admissions. The classification of “urgent” and “emergency” is designated directly on each Medicare hospital admissions record. We excluded other classifications, such as “elective.” A recent study (Dominici et al. 2006 (link)) considered a number of different cardiovascular and respiratory outcomes. Because of the sparser sampling of the PM_2.5 component data compared with the PM_2.5 total mass data, to obtain sufficient statistical power we collapsed the data into two broad categories of hospital admissions: a) CVD, which includes heart failure (ICD-9 code 428), heart rhythm disturbances (426–427), cerebrovascular events (430–438), ischemic heart disease (410–414, 429), and peripheral vascular disease (440–448); and b) respiratory diseases, which include chronic obstructive pulmonary disease (490–492) and respiratory infection (464–466, 480–487). We excluded admissions for injuries and for external causes (800–849). By collapsing these health outcomes, we increased statistical power and obtained more stable estimates of risk at the cost of some specificity of the outcome.
We analyzed each outcome (respiratory or cardiovascular admissions) separately. We calculated the daily counts of hospitalizations by summing the hospital admissions for each disease of interest recorded as a primary diagnosis. To calculate daily hospitalization rates, we constructed a parallel time series of the numbers of individuals enrolled in Medicare that were at risk in each county on each day. We based the location of each hospital admission on the county of residence of the enrollee.
The U.S. EPA established the PM Speciation Trends Network (STN) to measure more than 50 PM_2.5 chemical components, in addition to total mass. The STN includes > 50 national air monitoring stations (NAMS) and > 200 state and local air monitoring stations (SLAMS) (U.S. EPA 1999 ). Air pollution concentrations were typically measured on a 1-in-3–day schedule in the NAMS and on a 1-in-6–day schedule in the SLAMS. We removed suspect data and extreme values from the original monitor records; monitors with very little data were omitted altogether. Full details of the construction of the database can be found else-where (Bell et al. 2007 (link)). We also used PM_2.5 total mass measurements from the U.S. EPA’s Air Quality System as in our previous analyses (Dominici et al. 2006 (link)). Of the 187 counties described in the Bell et al. (2007) (link) analysis, we restricted the present analysis to counties with general populations larger than 150,000 and with at least 100 observations on components of PM_2.5. These requirements ensured that we would have enough data in a particular location to estimate an association between PM_2.5 components and hospital admissions. The study population consisted of 12 million Medicare enrollees living in 119 urban counties in the United States (Figure 1).
We limited our analysis to the components making up a large fraction of the total PM_2.5 mass or covarying with total mass (Bell et al. 2007 (link)): sulfate, nitrate, silicon, elemental carbon (EC), organic carbon matter (OCM), sodium ion, and ammonium ion. These seven components, in aggregate, constituted 83% of the total PM_2.5 mass, whereas all other components individually contributed < 1%. We computed countywide averages for each of these components and for PM_2.5 total mass by averaging the daily values from all monitors in a county. We adjusted organic carbon measurements for field blanks to estimate OCM. We used a standard approach such that OCM = k(OC_m − OC_b), where OCM represents organic carbon matter, OC_m represents measured organic carbon, OC_b represents organic carbon for blank filters, and k is the adjustment factor to account for non-carbon organic matter. We applied a k value of 1.4, as in a previous analysis (Bell et al. 2007 (link)). We obtained temperature and dew-point temperature data from the National Climatic Data Center on the Earth-Info CD database (EarthInfo 2006 ).
As a check on the consistency of the chemical component data, we first assessed whether three different PM_2.5 indicators (four scenarios total) provided comparable estimates of the short-term associations of PM_2.5 with cardiovascular and respiratory admissions: PM_2.5 (1), PM_2.5 measured by the national PM_2.5 monitoring network for the period 1999–2006; PM_2.5 (1a), PM_2.5 (1) for the period 2000–2006 and including only days with available measurements for all the seven PM_2.5 components from the STN; PM_2.5 (2), PM_2.5 measured by the STN for the period 2000–2006 and including only days with available measurements for all the seven PM_2.5 components from the STN; and PM_2.5 (3), PM_2.5 estimated as the sum of the seven largest components of PM_2.5 mass for the period 2000–2006. Significant differences between these estimates would raise uncertainty as to the recorded values of PM_2.5 total mass and its components. The estimates obtained under the scenarios 1a, 2, and 3 use data on the same subset of days. Each of these measures of PM_2.5 was available in all 119 counties.
We estimated the within-county monitor-to-monitor correlation for each of the seven PM_2.5 components to obtain a measure of the spatial homogeneity of each component. For this calculation we used a subset of 12 counties that had more than one monitor (27 monitors total): Jefferson, Alabama; Washington, DC; Cook, Illinois; Jefferson, Kentucky; Wayne, Michigan; Bronx, New York; Cuyahoga, Ohio; Allegheny, Pennsylvania; Philadelphia, Pennsylvania; Providence, Rhode Island; King, Washington; and Kanawha, West Virginia. We computed correlations only if at least 90 paired observations were available between two monitors. We also estimated the median within-county correlations between the seven PM_2.5 components, and the three measures of PM_2.5 total mass by a) estimating the correlations between time series data for each pair of air pollutants within each county and b) taking the median of the estimated correlations across the 119 counties. As a separate measure of spatial homogeneity, we calculated, for each of the seven components and using all monitors, the distance at which the correlation between pairs of monitors was 0.5 on average.
We applied Bayesian hierarchical statistical models to estimate county-specific and national average associations between daily variation in the seven PM_2.5 chemical components and daily variation in hospital admissions rates. This approach was originally developed for the National Morbidity, Mortality, and Air Pollution Study (Bell et al. 2004 (link); Samet et al. 2000 (link)) and subsequently extended (Dominici et al. 2006 (link)) to provide a consistent and unified methodology for analyzing data from multiple locations. We fit log-linear Poisson regression models with overdispersion to county-specific time-series data on hospital admissions and chemical components, accounting for potential confounders such as weather, day of the week, unobserved seasonal factors, and long-term trends. In each county-specific regression model, we included an indicator for the day of the week, a smooth function of time with 8 degrees of freedom (df) per calendar year to control for seasonality and long-term trends, a smooth function of current-day temperature (6 df), a smooth function of the 3-day running mean temperature (6 df), a smooth function of current-day dew-point temperature (3 df), and a smooth function of the 3-day running mean dew-point temperature (3 df). For all of the smooth functions we used a natural spline basis. We conducted a sensitivity analysis with respect to the smooth function of time to determine the degree to which risk estimates changed with varying levels of adjustment for smooth unmeasured confounders. Although other information about Medicare enrollees is available, such as sex and race, we excluded these factors from all models because they do not vary over time and should not play a role in our time series analysis.
For the exposure concentrations, we examined 0-, 1-, and 2-day lag concentrations because our previous work with PM_2.5 total mass and hospital admissions showed little evidence of a strong association with admissions at a lag of ≥ 3 days (Dominici et al. 2006 (link)). We examined each lag separately because the 1-in-6–day sampling of the chemical component data from the STN prohibited the use of distributed lag models where all lags can be examined simultaneously.
For estimating the health effects of the PM_2.5 components, we employed single-pollutant and multipollutant models. In single-pollutant models, we included each PM_2.5 component in the regression model individually, without adjusting for any other chemical component (the model does adjust for other time-varying factors). In multipollutant models, we included PM_2.5 components simultaneously to obtain estimates of the regression coefficients for each component adjusted for the other components. Ammonium was excluded from models that included sulfate and nitrate because of the high correlation among these three components. We included ammonium in a separate multipollutant model that did not include sulfate or nitrate but included the remaining four components.
We combined the county-specific risk estimates to form a national average using a Bayesian hierarchical model. In the single-pollutant models, we combined the log-relative risks separately for each pollutant using TLNise two-level normal independent sampling estimation software (Everson and Morris 2000 ). For the multiple-pollutant models, the risks were treated as a vector for each county and combined using a multivariate normal hierarchical model. We used Markov chain Monte Carlo methods to obtain the posterior distribution of the national average component effects. We assessed statistical significance by the posterior probability that the national average relative risk for a component was greater than zero. Values of the posterior probability > 0.95 were considered statistically significant (Dominici et al. 2006 (link); Peng et al. 2008 (link)).
We evaluated whether the relative risks of each PM_2.5 component in a multipollutant model were equal. In this analysis, the risks represent the percent increase in admissions associated with a 1-μg/m³ increase in each PM_2.5 component in a multipollutant model. We assessed the evidence against equal component risks using a chi-square statistic applied to the national average estimates. We also estimated the posterior probability that the coefficient for a particular component was greater than the mean of the coefficients for the other components.
For statistical calculations we used R statistical software, version 2.7.0 (R Foundation for Statistical Computing, Vienna, Austria).

Standard chemiluminescence assays for measuring nitrite and nitrate contents were performed according to previously published protocols [38 (link),39 (link)]. Blood was drawn from the femoral artery into vacutainer tubes containing sodium citrate (Becton Dickinson, Franklin Lakes, NJ, USA) and was immediately centrifuged to obtain platelet-free plasma using a PDG_TM platelet centrifuge (Bio/Data, Horsham, PA, USA). Tissue samples were collected and proteins from all samples were precipitated by adding methanol (dilution 1:1), followed by subsequent centrifugation at 11,000× g for 15 min at 4 °C. Supernatants were used to determine nitrite and nitrate contents using chemiluminescence (Sievers 280i Nitric Oxide Analyzer, GE Analytical Instruments, Boulder, CO, USA).

Redistilled solvents and Milli-Q water (>18
MΩ cm) were used for substrate cleaning and preparation of solutions.
An FEP (30 μm, DAIKIN) film was used for triboelectrification
with liquid droplets. Sodium chloride (99.5%), sodium iodide (99.5%),
sodium nitrate (99%), sodium bicarbonate (99.5%), sodium carbonate
(99.5%), sodium sulfite (98%), sodium sulfate (99%), magnesium sulfate
(99%), potassium chloride (99.5%), potassium ferricyanide (99.5%),
calcium chloride (97%), chromic nitrate (99%), manganese(II) chloride
(99%), manganese(II) sulfate (99%), ferric nitrate (98.5%), nickel(II)
chloride (99%), copper sulfate (99%), copper(II) nitrate (99%), zinc
nitrate (99%), zinc acetate (98%), hydrochloric acid (37%), sulfuric
acid (98%), nitric acid (68%), potassium hydroxide (99.9%), and sodium
hydroxide (97%) were purchased from Macklin. Ethanol (99.7%) and acetone
(99.5%) were obtained from Yong Da Chemical.