A Hyperlipidemic Rabbit Model Provides New Insightsinto Pulmonary Zinc Exposure Effects on Cardiovascular Health

Adriana J. LaGier Æ Nick D. Manzo Æ Alex P. Carll ÆRichard H. Jaskot Æ Ralph Slade Æ Judy H. Richards ÆDarrell W. Winsett Æ Aimen K. Farraj Æ Janice A. Dye

Published online: 25 October 2008

� Humana Press 2008

Abstract This study ascertains the effects of zinc, a

major component of particulate matter, on pulmonary and

systemic endpoints using hyperlipidemic rabbits to model

diet-induced human atherosclerosis. New Zealand White

rabbits were fed a normal or cholesterol-enriched diet and

then were intratracheally instilled 19/week for 4 weeks with

saline or 16 lg/kg of zinc, equal parts sulfate and oxide.

Physiologic responses, blood after each exposure, and ter-

minal bronchoalveolar lavage (BAL) were assessed. Rabbits

fed a cholesterol-rich diet developed hyperlipidemia and

had consistently higher circulating leukocyte counts than

rabbits fed normal chow. Within minutes after zinc instil-

lation, saturation of peripheral oxygen was decreased in

hyperlipidemic rabbits and heart rate was increased in

hyperlipidemic rabbits with total serum cholesterol levels

greater than 200 mg/dl. Total circulating leukocytes levels

were increased 24 h after the first zinc instillation, but upon

repeated exposures this effect was attenuated. After repeated

zinc exposures, BAL fluid (BALF) N-acetylglucosaminidase

activity was increased regardless of hyperlipidemic state.

Hyperlipidemic rabbits had an increase in BALF-oxidized

glutathione and a decrease in serum nitrite. The study elu-

cidates mechanisms by which the zinc metal component of

PM drives cardiovascular health effects, as well as the pos-

sible susceptibility induced by hyperlipidemia. Furthermore,

the study exemplifies the benefits of monitoring circulatory

physiology during exposure as well as after exposure.

Keywords Zinc � Rabbit � Cardiovascular �Particulate matter � Hyperlipidemia

Introduction

Increases in particulate air pollution are associated with

elevated cardiovascular hospital admissions and mortality

(reviewed in [33]). Patients with existing cardiovascular

disease (CVD) are more strongly affected than other

patients [41]. These data indicate that individuals with pre-

existing cardiovascular disease are at increased risk from

particulate matter (PM) exposure. In addition, the pul-

monary inflammation initiated by repeated PM insults may

sustain sub-clinical systemic inflammation, which has a

pivotal role in the development of atherosclerosis. Hence,

chronic PM exposure could contribute to the development

of cardiovascular disease [26].

Animal models of atherosclerosis have been used to

study pulmonary PM exposure effects on this disease [4, 6,

This article was reviewed by the National Health and Environmental

Effects Research Laboratory, U.S. Environmental Protection Agency,

and approved for publication. Approval does not signify that the

contents necessarily reflect the views and policies of the agency, nor

does mention of trade names or commercial products constitute

endorsement or recommendation for use.

A. P. Carll and N. D. Manzo supported by EPA-CT82947101 and

EPA-CT826512010, respectively.

A. J. LaGier (&) � R. H. Jaskot � R. Slade �J. H. Richards � D. W. Winsett � A. K. Farraj � J. A. Dye

Experimental Toxicology Division, National Health and

Environmental Effects Research Laboratory, Office of Research

and Development, U.S. Environmental Protection Agency,

109 TW Alexander Drive, MD: B143-01, Research Triangle

Park, NC 27711, USA

e-mail: lagier.adriana@epa.gov

N. D. Manzo

Department of Molecular Biomedical Sciences, North Carolina

State University, Raleigh, NC 27606, USA

A. P. Carll

School of Public Health, ENVR, University of North Carolina,

Chapel Hill, NC 27599, USA

Cardiovasc Toxicol (2008) 8:195–206

DOI 10.1007/s12012-008-9028-9

29]. The models used to date have some limitations in

terms of mimicking human disease because they are largely

based on genetic defects, i.e. the apolipoprotein E knockout

(ApoE-/-) mouse [44] or the Watanabe heritable hyper-

lipidemic (WHHL) rabbit that has a natural defect in low-

density lipoprotein (LDL) receptors [42]. These genetic

defects lead to alterations in cholesterol clearance and

spontaneous increases in total serum cholesterol levels that

are unlike humans with diet-mediated atherosclerosis.

To better represent the at-risk hypercholesterolemia-

induced atherosclerotic human population as a whole, the

study presented here used a diet-induced rabbit model of

hyperlipidemia and atherosclerosis. In this model, total

serum cholesterol levels and the nature of the target lesions

are controlled, at least in part, by adjusting dietary cho-

lesterol intake [25, 38]. Additionally, these animals

develop transitional atherosclerotic plaques comprised

predominantly of foamy macrophages [25], which are

important for PM-induced progression of atherosclerosis

[36, 43].

One of the common PM-associated elements found in

ambient air, especially PM samples collected near indus-

trial sources [10], is zinc, a transition metal. Past reports

have suggested that the zinc content of PM samples is

responsible for the responses to inhaled particulates [2, 23].

In fact, zinc oxide fume inhalation causes metal fume

fever, an occupational health hazard. In addition, zinc is

one of the largest metal fractions of EHC-93 [2], the

ambient PM sample that reportedly increased atheroscle-

rotic plaque volume in WHHL rabbits [36].

Using zinc as a plausible driver of PM toxicity, we

hypothesize that PM exposure in the hyperlipidemic rabbits

will elicit a greater adverse cardiovascular health outcome

than the non-lipemic rabbits. In addition, we propose that

these responses will be influenced by the degree of

hyperlipidemia. Finally, by monitoring immediate physio-

logic responses, changes in systemic biomarkers and

terminal pulmonary endpoints, we evaluate the potential

utility of hyperlipidemic rabbits as a model to assess PM

exacerbation of existing cardiovascular disease.

Materials and Methods

Animals

Male New Zealand White rabbits were used in all experi-

ments (Charles River Laboratory, Wilmington, MA).

Animals were housed in an isolated animal room in an

AAALAC-approved facility (21 ± 1�C, 50 ± 5% relative

humidity, 12 h light–dark cycle) and allowed free access to

water. Rabbits weighed between 1.8 and 2.1 kg prior to

start of diet-induction phase and were 36 weeks old with

weights between 3.6 and 4.5 kg at the start of the intra-

tracheal instillation regimen. All protocols were approved

by the Institutional Animal Care and Use Committee of the

U.S. EPA.

Diet

Rabbits were randomized by body weight into two groups

(N = 6 per group). One group (normal) was fed certified

high fiber rabbit diet 2031C and the other group (hyper-

lipidemic) was fed this rabbit diet supplemented with 0.2%

cholesterol (Harlan Teklad, Madison, WI). Based on the

diet’s mineral content, rabbits ingested approximately

7 mg zinc per day. A 0.2% cholesterol-supplemented diet

for at least 16 weeks increases total plasma cholesterol

levels and induces aortic lesions similar to atherosclerotic

plaques [38]. The diet-induction phase was administered

beyond this time point as prolonged ingestion of a cho-

lesterol-rich diet leads to more fibrous and less cellular

plaques [1] suggestive of human lesions prone to fissure,

hematoma, and/or thrombus.

Body Weight, Temperature, and Identification

Animals were weighed weekly during the diet-induction

phase and prior to, 24 and 48 h after instillations on an

ES50L standard electronic balance (Ohaus, Pine Brook,

NJ). Identification and temperature were monitored using

Implantable Programmable Temperature Transponder-

200TM (Biomedic Data Systems, Seaford, DE) that was

inserted subcutaneously as per manufacturer’s instructions.

Core temperatures were determined with a rectal

thermometer.

Blood Collection and Analysis

Blood was collected biweekly during the diet-induction

phase, 24 h after each instillation, and at necropsy. Because

of the known diurnal variation in total white blood counts,

all samples were obtained in the morning after an overnight

fast. Rabbits were sedated with 0.5 mg/kg acepromazine

IM (Henry Schein, Melville, NY). Blood was drawn from

the central ear artery and aliquots were collected into

additive-free or EDTA-treated BD vacutainers (Becton

Dickinson (BD), Franklin Lakes, NJ).

Aliquots of additive-free blood were centrifuged at

3000 9 g at room temperature and the serum was collected

into separate tubes. Using commercially available kits

modified and adapted for use on the KONELAB 30 clinical

chemistry analyzer (Thermo Clinical Labsystems, Espoo,

Finland), serum was analyzed for total cholesterol, low-

density lipoprotein (LDL), high-density lipoprotein (HDL),

and triglyceride content (ThermoElectron Corp., Melbourne,

196 Cardiovasc Toxicol (2008) 8:195–206

Australia). In addition, serum C-reactive protein (CRP) and

nitrite was assessed as per manufacturer’s instructions by a

rabbit-specific ELISA (American Laboratory Products Co.,

Windham, NH) and by a Griess reagent kit (Molecular

Probes, Invitrogen, Carlsbad, CA), respectively. Colori-

metric assays were read and analyzed on a Thermomax�

plate reader outfitted with Softmax Pro� v2.6.1 software

(Molecular Devices Corp., Menlo Park, CA).

Aliquots of EDTA-collected blood were analyzed for

complete blood counts (CBC), which included white blood

cell (WBC) counts and hematocrit, with an AcT blood

analyzer (Beckman-Coulter Inc., Fullerton, CA) and for

differential cell determination ([200 cells counted per

slide) from air-dried blood smear preparations stained with

LeukoStat (Fisher Scientific Co., Pittsburgh, PA). A post-

instillation/baseline ratio was calculated by dividing the

total WBC count 24 h post-instillation by the average pre-

instillation WBC count from week 14–36 of the diet-

induction phase.

Intratracheal Zinc Instillation

We designed the present study to experimentally resemble

a study by Suwa et al. [36] that used Watanabe rabbits to

conclude that exposure to ambient PM led to progression of

atherosclerosis. In this regard, we chose intratracheal

instillation as an exposure method. Based on total choles-

terol levels at the end of the diet-induction phase, we chose

to diverge from our original 2 9 2 experimental matrix

(±diet, ±zinc) in order to assess whether the degree of

hyperlipidemia was correlated to the risk of adverse health

outcome following the same zinc exposure. Rabbits were

assigned to groups for instillation of control saline (n = 3)

or zinc (n = 3 each for rabbits with normal (N), moderate

(M) or high (H) lipid levels). Zinc oxide (ZnO) and zinc

sulfate heptahydrate (ZnSO4) (Sigma Chemical Co., St.

Louis, MO) were added to pyrogen-free sterile saline,

probe sonicated for 1 min (Microson Ultrasonic cell dis-

ruptor, Misonix, Farmingdale, NY), vortexed, and instilled

intratracheally (IT) at a total dose of 16 lg/kg of zinc,

containing equal parts of ZnO and ZnSO4, in a volume of

2 ml (0.5 ml/kg). Control rabbits were instilled with

0.5 ml/kg sterile saline.

The zinc instillate was prepared fresh for each instilla-

tion and contained 50% soluble zinc (i.e. sulfate) and 50%

insoluble zinc (i.e. oxide), similar to the zinc content of

EHC-93 [2]. The 16 lg/kg dose of zinc given per instil-

lation approximates the amount of zinc contained in the

1.56 mg/kg of EHC-93 used in comparable rabbit studies

[29, 36] (approximately 10.4 lg zinc per mg EHC-93 [2]

times 1.56 mg/kg equals 16 lg zinc/kg body weight). An

initial dose–response pilot experiment in rabbits, which

received a single instillation of saline or 8, 16, or 32 lg/kg

of the zinc sulfate–oxide (50:50) instillate confirmed

that the 16 lg/kg dose resulted in mild pulmonary

inflammation.

For the instillations, rabbits were pre-medicated with

1 mg/kg midazolam IM, given 0.05 mg/kg atropine IM

and anesthetized by inhalation of vaporized isoflurane

via face mask to allow transoral intubation. Two appli-

cations, at least 3 min apart, of a topical anesthetic,

cetacaine (14.0% benzocaine), were applied to the larynx

with a cotton-tipped applicator. All pharmaceuticals were

purchased from Henry Schein (Melville, NY). Anesthe-

tized rabbits were placed in sternal recumbency on a

circulating warm water blanket and intubated with a

2.5 mm ID, 10FR, 14 cm long, uncuffed, Murphy oral/

nasal ET tube (Hudson RCI, Temecula, CA). Placement

was confirmed by visualizing a breathing pattern on a

SurgiVet� V9400 capnograph (Smiths Medical PM Inc.,

Waukesha, WI), which was attached to the ET tube

during placement. A 5FR feeding tube (Henry Schein) cut

to a length of 34 cm was passed through the endotracheal

tube to the level of the carina. The saline or zinc sus-

pension was instilled via the feeding tube, as two separate

1 ml boluses, followed by 20 ml of air to extrude any

instillate remaining in the tubing.

Vital signs were monitored for at least 10 min before,

during, and for at least 10 min after instillations. In addi-

tion to monitoring lung sounds, rabbits were instrumented

with a capnograph (Smiths Medical PM Inc) to monitor

end-tidal CO2 and respiratory rate (RR). Saturation of

peripheral oxygen (SpO2) was determined at 2 min inter-

vals with an Ohmeda Biox340 pulse oximeter (DRE Inc.,

Louisville, KY) attached between the footpads. Heart rate

was determined by way of auscultation and during the third

and fourth instillation via electrocardiogram (ECG). ECG

were recorded continuously (beginning 5 min prior to and

ending 10 min after instillation) with a radiotelemetry

transmitter (Model TA11CTA-F40; Data Sciences Inter-

national Inc., St. Paul, MN) attached externally with

alligator clips to shaved skin moistened with conductive

gel. Signals were transmitted to a receiver placed beneath

the animal, electronically recorded at a rate of 1,000 Hz on

a 54 mV scale and analyzed with ECG-Auto v1.5.12.48

(EMKA Technologies, Falls Church, VA). A minimum of

50 waveforms were analyzed per 1 min segment. Absolute

area of T represents the absolute value of the area under the

curve between the ECG trace and the isoelectric line.

BAL Fluid and Analyses

Two days after the fourth instillation, a time point used in

comparable rabbit studies [29], rabbits were weighed,

anesthetized as described above and bronchoalveolar

lavage (BAL) was performed, followed by euthanasia with

Cardiovasc Toxicol (2008) 8:195–206 197

Euthanosol (250 ll/kg, IV). BAL was performed similar

to procedures recommended for clinical use in small

companion animals [18] and optimized for use in the

laboratory rabbit [11]. Briefly, lungs were lavaged with a

volume of sterile 0.9% NaCl (Braun Medical, Inc., Irvine,

CA) equal to 10 ml/kg body weight. The entire volume

was instilled through the pre-placed endotracheal tube and

was aspirated back a single time. The BAL fluid was

quantified, examined grossly for evidence of surfactant,

hemorrhage, mucus strands, or cloudiness and kept on ice

for further analysis.

Total white cell counts were determined from analyses

of aliquots of whole BALF using a Coulter counter

(Coulter Inc., Miami, FL). Differential cell determinations

([200 cells counted per slide) were made from air-dried

cytospin preparations (Shandon, Pittsburgh, PA) stained

with LeukoStat (Fisher Scientific Co., Pittsburgh, PA). The

remaining BAL fluid was centrifuged at 1500 9 g to

remove cells. Using commercially available kits modified

and adapted for use on the KONELAB 30 clinical chem-

istry analyzer (Thermo Clinical Labsystems, Espoo,

Finland), cell-free BAL fluid was analyzed for albumin

(Diasorin, Saluggia, Italy), total protein content (Coo-

massie Plus Protein assay kit, Pierce, Rockford, IL with

BSA standards from Sigma, St. Louis, MO), lactate

dehydrogenase (LDH) activity (ThermoElectron, Corp.),

and N-acetyl-b-D-glucosaminidase (NAG) activity (Roche

Diagnostics, Indianapolis, IN). In addition, reduced (GSH)

and oxidized (GSSG) glutathione was labeled with dansyl

chloride [21] and analyzed by HPCL using a modification

of a previously described method [13].

Statistics

The data were analyzed by analysis of variance (ANOVA),

or where appropriate, a Student’s t-test, and by Bonfer-

roni’s corrections for multiple comparisons, using Prism 4

(GraphPad Software Inc., San Diego, CA). Statistical sig-

nificance was stated when P value of B0.05 was reached.

Correlations were determined using the Excel 2003 data

analysis tool (Microsoft Corp., Redmond, WA).

Results

Diet-Induction Phase

During the 36 weeks of the diet-induction phase (prior to

zinc instillation), all rabbits were serially monitored

(Fig. 1). Serum was assessed for lipid profile changes (total

cholesterol, LDL, HDL, triglyceride). Physical parameters,

i.e. temperature and weight, were measured and CBCs

were determined.

Lipid Profile

Rabbits fed normal rabbit chow (N, circle) had total serum

cholesterol levels below 50 mg/dl, while rabbits fed the

cholesterol-rich diet for 2 weeks and onward through the

end of the study fell into two distinct groups: moderate (M,

gray squares) and high (H, black squares) total serum cho-

lesterol levels, which were 50–200 mg/dl and[200 mg/dl,

respectively (Fig. 1a). This pattern of increase in total serum

Fig. 1 Responses to

cholesterol-supplemented diet.

Dotted gridlines on (a) delineate

normal (N \ 50 mg/dl),

moderate (M, 50–200 mg/dl),

and high (H [ 200 mg/dl) total

serum cholesterol. Relative to

rabbits fed normal chow (N,

open circle, n = 6), rabbits fed

a diet supplemented with 0.2%

cholesterol developed either

moderate (M, gray square,

n = 3) or high (H, black square,

n = 3) elevations in a total

serum cholesterol and had

altered, b low-density

lipoprotein (LDL), c body

weight gain, and d number of

circulating leukocytes. Data

shown are mean ± SE.

* P \ 0.05 versus rabbits fed

normal chow. # P \ 0.05 versus

rabbits with M lipid profiles

198 Cardiovasc Toxicol (2008) 8:195–206

cholesterol was correlated linearly with the increase in LDL

(r2 [ 0.95 at 2–36 weeks on diet) or hyperlipidemia

(Fig. 1b). HDL and triglyceride levels were not different

between groups, except at 14 weeks on diet, when the rab-

bits in group H had higher triglyceride levels (188 ± 32 mg/

dl) than rabbits in group N (54 ± 9 mg/dl) or M

(19 ± 1 mg/dl) (P \ 0.05, ANOVA).

Physical Parameters

No significant differences were noted in body temperatures

by way of subcutaneous transponders. After 8 weeks on

diet, the overall body weight gain was 0.23 ± 0.12 kg and

0.17 ± 0.03 kg higher in M and H hyperlipidemic rabbits,

respectively, when compared to N normolipidemic rabbits.

However, the weight difference between groups was always

\0.5 kg. Weight gain reached a plateau at approximately

16 weeks. Therefore, the diet-induction phase did not

cause the animals to become overtly obese or cachectic

(Fig. 1c). In general, all the rabbits appeared healthy and

did not show outward signs of poor health, e.g. appetite

loss or behavioral changes.

CBC

As previously reported [12], the rabbits with moderate (M)

and high (H) hyperlipidemia had elevated circulating leu-

kocytes when compared to rabbits with normal (N) lipid

levels. The leukocytosis in the hyperlipidemic rabbits was

noted consistently throughout the diet-induction phase

(Fig. 1d). Hematocrit levels were not different amongst

groups. Differential cell counts revealed that leukocytosis

was not driven by a single cell type, but rather reflected

increased lymphocytes, monocytes, and heterophils. (Het-

erophils in the rabbit represent neutrophils found in other

species.) However, total monocyte numbers correlated

linearly with total heterophil numbers (r2 = 0.98, 0.99, and

0.96, at 8, 10, and 14 weeks on diet, respectively), indic-

ative of the chronic inflammatory processes that

purportedly beget atherosclerosis.

Zinc Instillation Studies

Pilot Zinc Dose–Response Study in Normolipidemic

Rabbits

Past reports suggested that PM-associated zinc content is

an important causal component involved in acute lung and

sub-acute cardiac responses to inhaled particulates [2, 10,

24]. To ensure that these previous findings, which utilized

rodent models to implicate zinc, could be extended to the

rabbit species, we conducted an exploratory study to

examine the acute pulmonary effects of instilled zinc in

rabbits. All rabbits used in this dose–response study had

normal total serum cholesterol levels, \50 mg/dl.

Acute lung injury was assessed 48 h after instillation of

a single bolus of saline; or 8, 16, or 32 lg/kg zinc, com-

posed of half zinc sulfate and half zinc oxide (see

‘‘Materials and Methods’’). In a dose-dependent manner,

the total number of heterophils in BAL increased in zinc-

exposed rabbits compared with rabbits exposed to saline

(Fig. 2a). The percentage of lavageable macrophages and

lymphocytes decreased in zinc-exposed rabbits compared

with rabbits exposed to saline (Fig. 2b). Based on dose,

BAL fluid levels of LDH and total protein were elevated in

zinc-exposed rabbits in comparison to rabbits exposed to

saline (Fig. 2c). When compared to saline control, a slight

increase in NAG activity was noted after zinc instillation,

but this was not dose-dependent (data not shown). In

addition, serum levels of CRP, a purportedly sensitive

biomarker of systemic inflammation and vascular disease,

were increased as zinc dose increased (Fig. 2d). These data

indicate that zinc induced an acute, dose-dependent lung,

and systemic inflammatory response in rabbits that was

similar to previous reports in zinc-exposed rodents.

We also observed that a dose of 32 lg/kg zinc was not

well tolerated in that immediately after instillation the

subject developed tachypnea associated with depressed

SpO2. Hence, for the formal repeated zinc instillation

study, we utilized a dose of 16 lg/kg zinc (sulfate–oxide)

because it (1) was well-tolerated acutely, (2) was compa-

rable to that used in previous hyperlipidemic rabbit studies

[36], and (3) caused a detectable lung response character-

ized by moderate neutrophilic inflammation and only

minor lung injury.

Repeated Zinc Instillation Study in Rabbits with Normal,

Moderate or High Lipid Levels

Based on our pilot dose–response study and on the previous

study in WHHL rabbits [36], in the formal zinc instillation

study, we instilled normal (N), moderate (M), and high (H)

hyperlipidemic rabbits with saline or 16 lg/kg zinc sul-

fate–oxide (50:50) once a week for 4 weeks. We assumed

the weekly dosing regimen allowed the rabbits time for

clearance as most soluble and insoluble zinc clears the lung

2 days after instillation [20, 39]. We assessed immediate

physiologic responses at the time of each instillation, blood

biomarkers 24 h after each exposure, and terminal end-

points, e.g. BAL fluid indices.

Immediate Physiological Response to Zinc Expo-

sure Because zinc oxide inhalation has been associated

with metal fume fever, we monitored the rabbits for body

temperature before and after instillation. No significant

alteration in temperature was evident between groups,

Cardiovasc Toxicol (2008) 8:195–206 199

likely as a result of the route of exposure [27] or because of

anesthesia, which causes hypothermia.

Previous reports have shown that pulmonary exposure to

PM-associated transition metals effect physiological

parameters [5]. To this end, we monitored several vital

signs prior to, during, and after instillation. We noted that

instillation led to an altered breathing pattern. However,

because rates exceeded the detection limits of the capno-

graph, no significant disparities in the respiratory rate or

end-tidal CO2 could be ascertained.

SpO2 was decreased in zinc-instilled rabbits compared

to saline-instilled animals for up to 8 min after exposure

(Fig. 3a). This response was noted after each of the four

zinc instillations (data not shown). In rabbits receiving

zinc, the SpO2 decrease was more pronounced in moderate-

and high-hyperlipidemic rabbits when compared to normal

lipidemic rabbits at 2 min after exposure. SpO2 differences

between rabbits were not present 1 h post-instillation,

indicating that the effect was transient. Red blood cell

counts were not significantly different between groups

indicating that the decrease in SpO2, a measure of oxygen-

bound hemoglobin, was likely not due to a decrease in

hemoglobin levels.

Zinc instillation stimulated a rise in heart rate (*30–

80 bpm) in high-hyperlipidemic animals when compared to

all other groups (Fig. 3b). The increase in heart rate in

high-hyperlipidemic rabbits occurred during the entire

assessed period, from instillation to 8 min after instillation.

Heart rates in all rabbits returned to established baseline

1 h after instillation.

The alterations in heart rate were accompanied by chan-

ges in the morphology of the ECG waveforms of all the high-

hyperlipidemic rabbits, which after zinc instillation revealed

a flattening of the T-wave that was not present in the mod-

erate hyperlipidemic, normal lipidemic, or saline groups

(Fig. 3c). Analysis of the ECG waveforms demonstrated

that zinc instillation led to a decrease in absolute area of the

T-wave (Fig. 3d). Combined, these data indicate that the

immediate effect of instilling a bolus of zinc metal on cir-

culatory physiology occurs only in rabbits with

hyperlipidemia, an indicator of poor cardiovascular health.

Circulatory Responses to Repeated Zinc Instilla-

tions Systemic inflammation has been recognized as a

factor that predicts cardiovascular mortality. In this regard,

we assessed blood samples for inflammatory responses in

the vasculature, i.e. circulating leukocyte, serum CRP, and

serum nitrite levels.

Twenty-four hours after the first instillation, the total

number of circulating leukocytes relative to the total number

of circulating leukocytes before instillations were increased

in all zinc-exposed rabbits, but not in saline controls. This

response declined in magnitude after the second instillation,

and was negligible by the third and fourth instillation, when

compared to saline controls (Fig. 4a). Although there was a

trend towards an accelerated attenuation of lymphocyte

numbers and more persistent heterophil numbers, differen-

tial counts did not show a consistent change in any particular

cell type examined. Eosinophil involvement was not

evident. By the middle of the 4 week zinc exposure

Fig. 2 Pilot zinc dose–response

study using normolipidemic

rabbits. Pulmonary response

determined in bronchoalveolar

lavage fluid (BALF) recovered

from rabbits with normal lipid

levels 48 h after a single

instillation of saline (0), or 8,

16, or 32 lg/kg of zinc sulfate–

oxide mix (n = 1 per

treatment). a Total numbers of

heterophils (striped bar). bPercentage of total BAL

alveolar macrophages (graybar), lymphocytes (black bar),

and heterophils (striped bar). cTotal protein concentration

(dark gray bar) and lactate

dehydrogenase activity (LDH,

dotted bar). d C-reactive protein

(CRP)

200 Cardiovasc Toxicol (2008) 8:195–206

experimental period, i.e. before the third instillation, there

was no significant change in total number of circulating

leukocytes relative to the levels before instillations indi-

cating that leukocyte counts after the first instillation did not

remain elevated through subsequent instillations (data not

shown).

By the end of the 4 week zinc exposure experimental

period (post-treatment), there was no significant change,

relative to the levels before instillations (pre-treatment), in

total number of circulating leukocytes, total serum choles-

terol, LDL, HDL, triglycerides, and body weight (Table 1).

However and likely because diet was continued throughout

the instillation period (week 36–39), the significant leuko-

cytosis, hypercholesterolemia, and hyperlipidemia in the

hyperlipidemic rabbits evident at the end of the diet-induc-

tion phase (week 36) was also noted 48 h after the fourth

instillation procedure (week 39). We noted that the total

cholesterol levels were greater than the sum of the HDL,

LDL, and triglycerides, which was likely due to the contri-

bution of lipoproteins not quantified in this study, e.g. VLDL

or IDL [38].

In addition, we assessed serum for determinants that

predict cardiovascular events, i.e. CRP. The serum CRP

levels in all rabbit groups before zinc instillations were

similar to the 90th percentile of humans (\3 mg/l) [35].

Serum CRP levels relative to the levels before instillations

were elevated, but not significantly, in all zinc-exposed

rabbits when compared to saline controls at 24 h after the

fourth zinc instillation (Fig. 4b).

Because nitric oxide levels are relevant to atheroscle-

rosis and reduced nitric oxide activity has been associated

with increased lipid levels (reviewed in [19]), we also

assayed serum for nitrite, a nitric oxide metabolite. At 24 h

after the fourth zinc instillation, serum nitrite levels relative

to the levels before instillations were depressed in high-

hyperlipidemic rabbits when compared to saline controls

and moderate hyperlipidemic and normolipidemic rabbits

(Fig. 4c). The depression in nitrite relative to levels before

instillations that was demonstrated in high-hyperlipidemic

rabbits was also evident after the first, second, and third

instillation when compared to saline controls (data not

shown).

BAL Findings After Repeated Zinc Instillations Results

from the current pilot dose–response (Fig. 2) and from

others [16] have shown that a single zinc exposure sig-

nificantly increases the number of granulocytes obtained

from BAL collected 48 h post-exposure. We show here

(Fig. 5a) that after four consecutive weekly zinc exposures,

there is no longer an appreciable increase in the number of

Fig. 3 Immediate physiological response to zinc exposure. Satura-

tion of peripheral oxygen (SpO2) (a), heart rate (b), electrocardiogram

(ECG) traces (c), and T-wave absolute area (T-area) (d) recorded

prior to, during, and 8 min after instillation in rabbits with normal

lipid levels instilled with saline (open circle or N-Sal) or rabbits

instilled with 16 lg/kg zinc sulfate–oxide mix with normal lipid

levels (open triangle or N-Zn), moderate hyperlipidemia (gray square

or M-Zn), or high hyperlipidemia (black square or H-Zn). SpO2 (a),

heart rate (b), and T-area (d) values represented as change from

baseline that was calculated per animal immediately prior to

instillation and are mean ± SE from three animals. ECG waveforms

depicted (c) represent averaged waveforms compiled by ECG-Auto

software from a 1-min segment of ECG activity as was recorded prior

to (gray line) and 4 min after (bold line) instillation in one animal

from each group (N-Sal, N-Zn, M-Zn, or H-Zn). The morphology of

the ECG waveforms depicted (c) was consistent for all three animals

per group and the changes to T-wave area for all three animals are

quantified in (d). All figures shown are from the fourth instillation.

* P \ 0.05 versus N-Sal. # P \ 0.05 versus M-Zn

b

Cardiovasc Toxicol (2008) 8:195–206 201

heterophils lavaged 48 h after zinc instillation (N-Zn),

compared to saline controls (N-Sal) and relative to the

increase that occurs acutely following zinc instillation in

a naı̈ve animal (0.11 9 103 vs. 3.05 9 103 heterophils/ll

BAL from rabbits receiving a single saline or zinc instilla-

tion, respectively, compared to 0.38 9 103 vs. 0.42 9 103

heterophils/ll BAL from rabbits receiving repeated saline or

zinc instillations, respectively). In addition, after repeated

zinc exposures, there was no significant increase in lavage-

able lung heterophils in hyperlipidemic rabbits (M-Zn and

H-Zn) when compared to N-Sal and N-Zn (Fig. 5a).

At the dosing regimen used, repeated intratracheal zinc

instillation did not alter protein, albumin, or LDH levels in

BALF. However, NAG activity was significantly increased

in rabbits instilled with zinc when compared to those

receiving saline, consistent with increased lysosomal

enzyme release from alveolar macrophages (Fig. 5b). No

significant changes were noted in the glutathione levels of

lung tissue collected from any group. The lung responses

were independent of the rabbit’s lipidemia. In contrast,

oxidized glutathione (GSSG) levels were increased in

hyperlipidemic rabbits receiving zinc when compared to

rabbits with normal lipid levels receiving either zinc or

saline, consistent with development of increased oxidative

stress in animals with excess circulating cholesterol and

LDL. Reduced glutathione (GSH) levels were not signifi-

cantly different among the groups (Fig. 5c). Together these

data suggest that repeated zinc exposures lead to an

inflammatory response whose character differs from the

inflammation elicited by a single instillation.

Discussion

This study provides new insights into whether exposure of

an at-risk population to PM elicits a greater adverse car-

diovascular health outcome than in a population not at-risk.

The at-risk human population was represented by a rabbit

model of hyperlipidemia, and an intratracheally instilled

mixture of soluble and insoluble zinc forms was used as a

Fig. 4 Systemic responses to repeated zinc instillation. Vascular

response determined in blood recovered from rabbits with normal

lipid levels instilled with saline (N-Sal, open bar) or rabbits instilled

with 16 lg/kg zinc sulfate–oxide mix with normal lipid levels (N-Zn,

gray bar), moderate hyperlipidemia (M-Zn, striped bar), or high

hyperlipidemia (H-Zn, black bar). a Number of circulating leukocytes

recovered 24 h after the first (1), second (2), third (3), and fourth (4)

instillations, which were performed weekly for a month. b C-reactive

protein (CRP) and c nitrite levels were assessed 24 h after the fourth

instillation. Relative levels were calculated by dividing the yield in

each treated animal by the mean yield calculated per animal prior to

any instillations. Data shown are mean ± SE from three animals.

* P \ 0.05 versus matched N-Sal. # P \ 0.05 versus same treatment

group after three and four instillations

202 Cardiovasc Toxicol (2008) 8:195–206

plausible driver of PM toxicity. Herein, we demonstrated

that immediately upon instillation of zinc, severely

hyperlipidemic rabbits had altered cardiovascular physiol-

ogy, characterized by a decrease in SpO2, an increase in

heart rate, and an alteration in the morphology of the ECG

waveforms, i.e. T-wave flattening. In addition, after repe-

ated zinc instillations, serum nitrite levels were depressed

and BALF-oxidized glutathione levels were increased in

high-hyperlipidemic rabbits when compared to all other

groups. These findings are consistent with our hypothesis

that hyperlipidemic subjects (similar to humans with ath-

erosclerosis) are at an increased risk for developing

untoward CVD health indices after repeated PM, i.e. zinc,

exposures and that these responses are influenced by the

degree of hyperlipidemia.

We also evaluated the potential utility of hyperlipidemic

rabbits as a model to assess air pollutant health effects on

populations susceptible to cardiovascular disease. By using

the rabbit model, we could non-invasively detect several

physiologic parameters and serially monitor a variety of

serum determinants. In this regard, we were able to detect

immediate changes in circulatory physiology and to deter-

mine that repeated intratracheal exposures to a mixture

of soluble and insoluble zinc forms resulted in attenuated

leukocyte increase. In this regard, this diet-induced hyper-

lipidemic rabbit model appears to be an ideal model with

Fig. 5 BAL findings after repeated zinc instillations. Pulmonary

response determined in bronchoalveolar lavage (BAL) recovered 48 h

after the fourth instillation in rabbits with normal lipid levels instilled

with saline (N-Sal) or rabbits instilled with 16 lg/kg zinc sulfate–

oxide mix with normal lipid levels (N-Zn), moderate hyperlipidemia

(M-Zn), or high hyperlipidemia (H-Zn). a Numbers of BAL

heterophils and b N-acetyl-b-D-glucosaminidase (NAG), [N-Sal (openbar), N-Zn (gray bar), M-Zn, (striped bar), H-Zn (black bar)]. cReduced (GSH, open bar) and oxidized (GSSG, gray bar) glutathi-

one. Data shown are mean ± SE from three animals. * P \ 0.05

versus N-Sal. # P \ 0.05 versus GSH

Table 1 Systemic response per treatment group before the first zinc

instillation and terminally, 48 h after the fourth zinc instillation

Before

instillations

After 4th

instillationa,b

Leukocytes (106/ml) N-Sal 5.2 ± 0.4 5.3 ± 0.8

N-Zn 3.6 ± 0.3 4.9 ± 0.7

M-Zn 7.6 ± 0.3 6.3 ± 0.7

H-Zn 9.3 ± 1.4*^ 11.4 ± 1.8*^

Cholesterol (mg/dl) N-Sal 16 ± 2 9 ± 6

N-Zn 19 ± 1 12 ± 3

M-Zn 70 ± 6*^ 97 ± 2*^

H-Zn 439 ± 101*^# 746 ± 254*^#

LDL (mg/dl) N-Sal 7 ± 0 13 ± 2

N-Zn 7 ± 1 13 ± 1

M-Zn 18 ± 1*^ 30 ± 2*^

H-Zn 150 ± 29*^# 249 ± 70*^#

HDL (mg/dl) N-Sal 9 ± 1 14 ± 3

N-Zn 7 ± 2 13 ± 1

M-Zn 12 ± 2 16 ± 2

H-Zn 19 ± 4 14 ± 1

Triglycericides (mg/dl) N-Sal 44 ± 5 39 ± 1

N-Zn 82 ± 35 59 ± 14

M-Zn 32 ± 5 21 ± 4

H-Zn 103 ± 51 111 ± 62

Weight (kg) N-Sal 4.0 ± 0.1 4.0 ± 0.1

N-Zn 3.8 ± 0.1 3.8 ± 0.1

M-Zn 4.2 ± 0.1 4.2 ± 0.0

H-Zn 3.9 ± 0.1 3.8 ± 0.1

Values are mean ± SE (n = 3 per group)a No significant differences versus before instillationsb All values from 48 h after fourth instillation, terminus of study

* P \ 0.05 versus N-Sal at matched time

^ P \ 0.05 versus N-Zn at matched time# P \ 0.05 versus M-Zn at matched time

Cardiovasc Toxicol (2008) 8:195–206 203

which to assess PM health effects in susceptible populations,

particularly when pathophysiology is involved.

Our findings are consistent with previous findings that

pulmonary exposure to the metal component of PM elicits

a cardiopulmonary inflammatory response. In this regard,

inhalation of heavy metal oxide, in particular zinc oxide,

leads acutely to metal fume fever, an occupational health

disease that in humans is accompanied by increased heart

rate [17] and leukocytosis [9]. Both of these metal fume-

induced effects were noted in our zinc-exposed hyperlipi-

demic rabbits.

Previous reports indicate that repeated exposure to zinc

fume leads to ‘pulmonary tolerance’, which is defined as the

lung’s ability to withstand the detrimental effects of a toxic

compound following multiple exposures [40]. We report

here that repeated instillation of zinc likely leads to more

global ‘tolerance’ as assessed by the blunting of zinc-

induced circulating leukocyte increases. Diminished leu-

kocyte increases could occur as a consequence of reduced

cellular production. However, reduction in production is not

indicated by a recent study demonstrated that increased

marrow production occurs in PM-exposed hyperlipidemic

rabbits [43].

In this study, we demonstrated that when leukocyte

infiltrate was reduced, i.e. after the fourth instillation,

serum CRP was elevated, serum nitrite was reduced, and

oxidized glutathione in airways was increased. Taking this

data into consideration, leukocyte numbers were likely

affected by cell influx because leukocyte migration is

modulated by CRP [22], nitric oxide [3], and oxidative

stress [28]. Furthermore, the anti-inflammatory milieu

established to terminate the first zinc-induced inflammatory

response would likely inhibit the influx of leukocytes

responding to successive zinc exposures. In addition to

production and cell influx, other causes of leukocytosis,

such as delayed apoptosis, altered clearance, or a shift from

marginating pools to circulating compartments, should be

considered for further study.

Systemic inflammation in combination with T-wave

changes on ECG traces has been previously shown to be

predictive of cardiovascular disease mortality [30]. This

study demonstrated that PM, i.e. zinc, initiated systemic

inflammation and altered the circulatory physiology of

hyperlipidemic rabbits, predicting deleterious cardiovas-

cular disease health effects in subjects with atherosclerosis.

In particular, we determined that zinc instillation led to an

immediate decrease in SpO2 and increase in heart rate in

the high-hyperlipidemic rabbits. Our data are consistent

with reports in humans with cardiovascular disease that

pulmonary PM exposure alters oxygen saturation [8], heart

rate [32], and ECG waveforms [31].

The altered SpO2 and heart rate that was noted in

hyperlipidemic subjects immediately upon pulmonary zinc

exposure could occur as a consequence of modified vascular

tone [34, 37]. We demonstrated that severely hyperlipidemic

rabbits had depressed levels of serum nitrite, tentatively

suggesting a reduction in the vasodilator activity of nitric

oxide and an alteration in vascular tone. Decreased nitric

oxide activity, as suggested by depressed levels of serum

nitrite, has also been associated with oxidative stress [14],

which is along with PM exposure involved with athero-

genesis [26]. In this regard, we showed as assessed

terminally with an increase in BALF-oxidized glutathione

that the airways in zinc-exposed hyperlipidemic rabbits had

an increase in oxidative stress. Combined these data suggest

that after repeated exposure to instilled zinc, hyperlipidemic

rabbits may have diminished antioxidant capacity and that

susceptibility to PM, i.e. zinc, may relate to being less able to

respond to the repeated oxidative stress associated with

continued zinc exposure.

Physiological effects elicited in already susceptible

cardiovasculature may lead to late-onset adverse effects

possibly modulating by decreased nitric oxide activity and

increased oxidative stress. The immediacy of the changes

in circulatory physiology suggests that the zinc-induced

pathophysiology is likely mediated by a neural reflex. In

this regard, atherosclerosis in combination with reduced

nitric oxide levels and increased oxidative stress have been

associated with reduced sensitivity of the baroreflex, which

detects and corrects changes in vascular tone [15]. Recent

evidence indicates that repeated activation of the barore-

ceptor reflex results in risk for late-onset cardiovascular

mortality [7].

In conclusion, using a diet-induced rabbit model of

hyperlipidemia, we present data that pulmonary instillation

of zinc induces both pulmonary and systemic inflammatory

responses whose character changes upon repeated instilla-

tions. Furthermore, the intratracheal instillation of zinc to

an already susceptible cardiovascular system likely leads to

immediate pathophysiology, decreased systemic nitric

oxide activity, and increased pulmonary oxidative stress.

These data may explain why sensitive groups exposed to

particle pollution experience heart-related health effects

after breathing polluted air.

Acknowledgments The authors thank Henry Daes, Allen Ledbetter,

and the animal staff for excellent technical assistance. We also thank

Dr. David Kurtz, Dr. Urmila Kodavanti, Dr. Gary Hatch, and Dr.

MaryJane Selgrade for critical review of the manuscript.

References

1. Adams, C. W., Miller, N. E., Morgan, R. S., & Rao, S. N. (1982).

Lipoprotein levels and tissue lipids in fatty-fibrous atherosclerosis

induced in rabbits by two years’ cholesterol feeding at a low

level. Atherosclerosis, 44, 1–8. doi:10.1016/0021-9150(82)90

047-8.

204 Cardiovasc Toxicol (2008) 8:195–206

2. Adamson, I. Y., Prieditis, H., Hedgecock, C., & Vincent, R. (2000).

Zinc is the toxic factor in the lung response to an atmospheric

particulate sample. Toxicology and Applied Pharmacology, 166,

111–119. doi:10.1006/taap.2000.8955.

3. Belenky, S. N., Robbins, R. A., & Rubinstein, I. (1993). Nitric

oxide synthase inhibitors attenuate human monocyte chemotaxis

in vitro. Journal of Leukocyte Biology, 53, 498–503.

4. Campen, M. J., McDonald, J. D., Reed, M. D., & Seagrave, J.

(2006). Fresh gasoline emissions, not paved road dust, alter

cardiac repolarization in ApoE-/- mice. Cardiovascular Toxi-cology, 6, 199–210. doi:10.1385/CT:6:3:199.

5. Campen, M. J., Nolan, J. P., Schladweiler, M. C., Kodavanti, U.

P., Evansky, P. A., Costa, D. L., et al. (2001). Cardiovascular and

thermoregulatory effects of inhaled PM-associated transition

metals: A potential interaction between nickel and vanadium

sulfate. Toxicological Sciences, 64, 243–252. doi:10.1093/toxsci/

64.2.243.

6. Chen, L. C., & Nadziejko, C. (2005). Effects of subchronic

exposures to concentrated ambient particles (CAPs) in mice. V.

CAPs exacerbate aortic plaque development in hyperlipidemic

mice. Inhalation Toxicology, 17, 217–224. doi:10.1080/0895837

0590912815.

7. De Ferrari, G. M., Sanzo, A., Bertoletti, A., Specchia, G., Vanoli,

E., & Schwartz, P. J. (2007). Baroreflex sensitivity predicts long-

term cardiovascular mortality after myocardial infarction even in

patients with preserved left ventricular function. Journal of theAmerican College of Cardiology, 50, 2285–2290. doi:10.1016/j.

jacc.2007.08.043.

8. DeMeo, D. L., Zanobetti, A., Litonjua, A. A., Coull, B. A.,

Schwartz, J., & Gold, D. R. (2004). Ambient air pollution and

oxygen saturation. American Journal of Respiratory and CriticalCare Medicine, 170, 383–387. doi:10.1164/rccm.200402-244OC.

9. Drinker, P., Thomson, R. M., & Finn, J. L. (1927). Metal fume

fever: II. Resistance acquired by inhalation of zinc oxide on two

successive days. Journal of Industrial Hygiene, 9, 98–105.

10. Dye, J. A., Lehmann, J. R., McGee, J. K., Winsett, D. W., Led-

better, A. D., Everitt, J. I., et al. (2001). Acute pulmonary toxicity

of particulate matter filter extracts in rats: Coherence with epi-

demiologic studies in Utah Valley residents. EnvironmentalHealth Perspectives, 109(Suppl 3), 395–403. doi:10.2307/343

4787.

11. Dye, J. A., Slade, R., Jaskot, R. H., Richards, J. H., Manzo, N.,

Taylor, G. M., & LaGier, A. J. (2008). Use of repeated bron-

choalveolar lavage in rabbits to assess pollutant-induced lung

changes in an animal model of cardiovascular (CV) disease.

American Thoracic Society International Conference, Toronto,

A426.

12. Feldman, D. L., Mogelesky, T. C., Liptak, B. F., & Gerrity, R. G.

(1991). Leukocytosis in rabbits with diet-induced atherosclerosis.

Arteriosclerosis and Thrombosis, 11, 985–994.

13. Gan, J., Harper, T. W., Hsueh, M. M., Qu, Q., & Humphreys, W.

G. (2005). Dansyl glutathione as a trapping agent for the quan-

titative estimation and identification of reactive metabolites.

Chemical Research in Toxicology, 18, 896–903. doi:10.1021/

tx0496791.

14. Ganafa, A. A., Socci, R. R., Eatman, D., Silvestrova, N., Abu-

khalaf, I. K., & Bayorh, M. A. (2002). Acute inhibition of

glutathione biosynthesis alters endothelial function and blood

pressure in rats. European Journal of Pharmacology, 454, 217–

223. doi:10.1016/S0014-2999(02)02500-1.

15. Gianaros, P. J., Jennings, J. R., Olafsson, G. B., Steptoe, A.,

Sutton-Tyrrell, K., Muldoon, M. F., et al. (2002). Greater intima-

media thickness in the carotid bulb is associated with reduced

baroreflex sensitivity. American Journal of Hypertension, 15,

486–491. doi:10.1016/S0895-7061(02)02923-0.

16. Gilmour, P. S., Nyska, A., Schladweiler, M. C., McGee, J. K.,

Wallenborn, J. G., Richards, J. H., et al. (2006). Cardiovascular

and blood coagulative effects of pulmonary zinc exposure. Tox-icology and Applied Pharmacology, 211, 41–52. doi:10.1016/j.

taap.2005.06.002.

17. Hassaballa, H. A., Lateef, O. B., Bell, J., Kim, E., & Casey, L.

(2005). Metal fume fever presenting as aseptic meningitis with

pericarditis, pleuritis and pneumonitis. Occupational Medicine(Oxford, England), 55, 638–641. doi:10.1093/occmed/kqi141.

18. Hawkins, E. C., & Berry, C. R. (1999). Use of a modified

stomach tube for bronchoalveolar lavage in dogs. Journal of theAmerican Veterinary Medical Association, 215, 1635–1639.

1620.

19. Henry, P. D., Cabello, O. A., & Chen, C. H. (1995). Hypercho-

lesterolemia and endothelial dysfunction. Current Opinion inLipidology, 6, 190–195.

20. Hirano, S., Higo, S., Tsukamoto, N., Kobayashi, E., & Suzuki, K.

T. (1989). Pulmonary clearance and toxicity of zinc oxide

instilled into the rat lung. Archives of Toxicology, 63, 336–342.

doi:10.1007/BF00278649.

21. Jones, D. P., Carlson, J. L., Samiec, P. S., Sternberg, P., Jr.,

Mody, V. C., Jr., Reed, R. L., et al. (1998). Glutathione mea-

surement in human plasma. Evaluation of sample collection,

storage and derivatization conditions for analysis of dansyl

derivatives by HPLC. Clinica Chimica Acta, 275, 175–184. doi:

10.1016/S0009-8981(98)00089-8.

22. Kew, R. R., Hyers, T. M., & Webster, R. O. (1990). Human

C-reactive protein inhibits neutrophil chemotaxis in vitro: Pos-

sible implications for the adult respiratory distress syndrome. TheJournal of Laboratory and Clinical Medicine, 115, 339–345.

23. Kodavanti, U. P., Schladweiler, M. C., Gilmour, P. S., Wallen-

born, J. G., Mandavilli, B. S., Ledbetter, A. D., et al. (2008). The

role of particulate matter-associated zinc in cardiac injury in rats.

Environmental Health Perspectives, 116, 13–20.

24. Kodavanti, U. P., Schladweiler, M. C., Ledbetter, A. D., Hauser,

R., Christiani, D. C., Samet, J. M., et al. (2002). Pulmonary and

systemic effects of zinc-containing emission particles in three rat

strains: Multiple exposure scenarios. Toxicological Sciences, 70,

73–85. doi:10.1093/toxsci/70.1.73.

25. Kolodgie, F. D., Katocs, A. S., Jr., Largis, E. E., Wrenn, S. M.,

Cornhill, J. F., Herderick, E. E., et al. (1996). Hypercholesterol-

emia in the rabbit induced by feeding graded amounts of low-

level cholesterol. Methodological considerations regarding indi-

vidual variability in response to dietary cholesterol and

development of lesion type. Arteriosclerosis, Thrombosis, andVascular Biology, 16, 1454–1464.

26. Kunzli, N., & Tager, I. B. (2005). Air pollution: From lung to

heart. Swiss Medical Weekly, 135, 697–702.

27. Mori, T., Akashi, S., & Nukada, A. (1975). Effects of the inha-

lation of catalytically active metallic oxide fumes on rabbits.

International Archives of Occupational and EnvironmentalHealth, 36, 29–39. doi:10.1007/BF01267849.

28. Moutet, M., d’Alessio, P., Malette, P., Devaux, V., & Chaudiere,

J. (1998). Glutathione peroxidase mimics prevent TNFalpha- and

neutrophil-induced endothelial alterations. Free Radical Biologyand Medicine, 25, 270–281. doi:10.1016/S0891-5849(98)000

38-0.

29. Mukae, H., Hogg, J. C., English, D., Vincent, R., & van Eeden, S.

F. (2000). Phagocytosis of particulate air pollutants by human

alveolar macrophages stimulates the bone marrow. The AmericanJournal of Physiology, 279, L924–L931.

30. Okin, P. M., Roman, M. J., Best, L. G., Lee, E. T., Galloway, J.

M., Howard, B. V., et al. (2005). C-reactive protein and elec-

trocardiographic ST-segment depression additively predict

mortality: The Strong Heart Study. Journal of the American

Cardiovasc Toxicol (2008) 8:195–206 205

College of Cardiology, 45, 1787–1793. doi:10.1016/j.jacc.

2005.02.072.

31. Pekkanen, J., Peters, A., Hoek, G., Tiittanen, P., Brunekreef, B.,

de Hartog, J., et al. (2002). Particulate air pollution and risk of

ST-segment depression during repeated submaximal exercise

tests among subjects with coronary heart disease: The Exposure

and Risk Assessment for Fine and Ultrafine Particles in Ambient

Air (ULTRA) study. Circulation, 106, 933–938. doi:

10.1161/01.CIR.0000027561.41736.3C.

32. Peters, A., Perz, S., Doring, A., Stieber, J., Koenig, W., &

Wichmann, H. E. (1999). Increases in heart rate during an air

pollution episode. American Journal of Epidemiology, 150,

1094–1098.

33. Pope, C. A., III, & Dockery, D. W. (2006). Health effects of fine

particulate air pollution: Lines that connect. Journal of the Air &Waste Management Association (1995), 56, 709–742.

34. Sartori, C., Lepori, M., & Scherrer, U. (2005). Interaction

between nitric oxide and the cholinergic and sympathetic nervous

system in cardiovascular control in humans. Pharmacology andTherapeutics, 106, 209–220. doi:10.1016/j.pharmthera.2004.11.

009.

35. Shine, B., de Beer, F. C., & Pepys, M. B. (1981). Solid phase

radioimmunoassays for human C-reactive protein. Clinica Chi-mica Acta, 117, 13–23. doi:10.1016/0009-8981(81)90005-X.

36. Suwa, T., Hogg, J. C., Quinlan, K. B., Ohgami, A., Vincent, R., &

van Eeden, S. F. (2002). Particulate air pollution induces pro-

gression of atherosclerosis. Journal of the American College ofCardiology, 39, 935–942. doi:10.1016/S0735-1097(02)01715-1.

37. Talke, P., & Stapelfeldt, C. (2006). Effect of peripheral vaso-

constriction on pulse oximetry. Journal of Clinical Monitoringand Computing, 20, 305–309. doi:10.1007/s10877-006-9022-3.

38. Thomas, L. J., Picard, M. D., Miller, D. P., Adari, H., Beattie, D.

T., Emmett, C. D., et al. (2003). Modifications of the cholesterol-

fed rabbit model of atherosclerosis using diets with different

levels of cholesterol supplementation. Preclinica, 1, 60–69.

39. Wallenborn, J. G., McGee, J. K., Schladweiler, M. C., Ledbetter,

A. D., & Kodavanti, U. P. (2007). Systemic translocation of

particulate matter-associated metals following a single intratra-

cheal instillation in rats. Toxicological Sciences, 98, 231–239.

doi:10.1093/toxsci/kfm088.

40. Wesselkamper, S. C., Chen, L. C., & Gordon, T. (2005). Quan-

titative trait analysis of the development of pulmonary tolerance

to inhaled zinc oxide in mice. Respiratory Research, 6, 73. doi:

10.1186/1465-9921-6-73.

41. Wichmann, H. E., Mueller, W., Allhoff, P., Beckmann, M.,

Bocter, N., Csicsaky, M. J., et al. (1989). Health effects during a

smog episode in West Germany in 1985. Environmental HealthPerspectives, 79, 89–99. doi:10.2307/3430534.

42. Yamamoto, T., Bishop, R. W., Brown, M. S., Goldstein, J. L., &

Russell, D. W. (1986). Deletion in cysteine-rich region of LDL

receptor impedes transport to cell surface in WHHL rabbit. Sci-ence, 232, 1230–1237. doi:10.1126/science.3010466.

43. Yatera, K., Hsieh, J., Hogg, J. C., Tranfield, E., Suzuki, H., Shih,

C. H., et al. (2008). Particulate matter air pollution exposure

promotes the recruitment of monocytes into atherosclerotic pla-

ques. American Journal of Physiology. Heart and CirculatoryPhysiology, 294, H944–H953.

44. Zhang, S. H., Reddick, R. L., Piedrahita, J. A., & Maeda, N.

(1992). Spontaneous hypercholesterolemia and arterial lesions in

mice lacking apolipoprotein E. Science, 258, 468–471. doi:

10.1126/science.1411543.

206 Cardiovasc Toxicol (2008) 8:195–206