Male FVB/N-TgN (Ho1-luc) Xen mice were obtained from Xenogen Corp. (Alameda, CA). These mice carry a fusion of a 15-kb mouse HO-1 promoter with modified firefly luciferase cDNA (Promega pGL-3) and were used to assess HO-1 activation. Male C57BL/6J and B6.129P2-Apoe^tm1Unc (ApoE^−/−) mice were obtained from Jackson Laboratory (Bar Harbor, ME). The ApoE^−/−mice lack apolipoprotein E (a high-affinity ligand for lipoprotein receptors), consequently have elevated plasma levels of cholesterol and triglycerides, and develop atherosclerotic plaques in a fashion similar to humans (Reddick et al. 1994). Mice were between 2 and 3 months of age when exposures began, and were fed an irradiated commercial mouse chow diet (LM-485; Harlan, Indianapolis, IN). For some of the atherosclerosis studies, ApoE^−/− mice were fed a high fat diet (15.8% fat; TD 88051; Harlan), which exacerbates the atheroma formation. All mice were provided tap water ad libitum and were housed in ventilated cages on autoclaved hardwood chip bedding. Sentinel mice were free of endogenous pathogens. Animal room conditions included HEPA-filtered air, controlled temperature and humidity, and a 12-hr light/dark cycle. Animal care and use procedures, including sacrifice by carbon dioxide asphyxiation, were conducted in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals (Office of Laboratory Animal Welfare 2002) and the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources 1996); these procedures were approved by the National Institute for Occupational Safety and Health institutional animal care and use committee. All mice used in these studies were treated humanely and with regard for alleviation for suffering.

To screen for systemic oxidative effects of SWCNT exposure, Ho1-luc reporter mice or C57BL/6 mice (at least 4 mice per treatment) were exposed to SWCNT in doses of 10–40 μg/mouse by single intrapharyngeal instillation and were sacrificed at time points including 1, 7, 28, and 56 days after exposure. These experimental settings were selected to correspond to previous pulmonary toxicity studies (Shvedova et al. 2005). Pulmonary toxicity has been characterized with early but acute and transient inflammatory response, granuloma formation around SWCNT agglomerates, and a fibrogenic response. To evaluate SWCNT effects on atherosclerosis progression, a chronic process, ApoE^−/− mice (10/treatment) were exposed by pharyngeal aspiration to a medium dose of SWCNT (20 μg/mouse) via multiple exposures (once every other week for 8 weeks). In all studies the control animals were exposed by pharyngeal aspiration to vehicle [sterile phosphate-buffered saline (PBS)].

SWCNT (CNI, Houston, TX) were produced by the high pressure carbon monoxide proportionation process (HiPco) technique (Scott et al. 2003; Scott and Smalley 2003) and purified by acid treatment to remove metal contaminants (Gorelik et al. 2000). Their physicochemical characteristics, including chemical content, mean diameter, and surface area, were reported previously (Shvedova et al. 2005). Ultrafine carbon black (UfCB) particles (Printex 90, Degussa, Germany), used as a control material, were a gift from G. Oberdörster (University of Rochester, Rochester, NY). Suspensions of SWCNTs and UfCBs were prepared by sonification using a Branson Sonifier 450 (Branson Ultrasonics Corp., Danbury, CT) in PBS for 3 min (at 400 W) at room temperature before animal exposure. For particulate administration, we used the pharyngeal aspiration technique (Rao et al. 2003). Briefly, mice were anesthetized with isoflurane (Abbott Laboratories, North Chicago, IL) and placed on a slant board, and the tongue was gently held in full extension while a 60-μL suspension of particles was pipetted onto the base of the tongue. This technique provides for a wide dissemination of particles in a peribronchiolar pattern within the alveolar region (Rao et al. 2003).

Food was removed from the cage for ~ 8 hr before sacrifice. Whole blood samples were drawn from the inferior vena cava into EDTA-containing tubes. The plasma was collected and stored for later analysis. The heart and vascular tree were perfused with ice-cold PBS containing heparin (10 U/mL) for the atherosclerotic lesion analysis. Animals used for evaluation of oxidative stress were perfused with ice-cold PBS containing 20 mM butylated hydroxytoluene (BHT) to prevent further oxidation. For DNA damage studies, the aortas were retrieved and rapidly frozen in liquid nitrogen and stored at −80°C before analysis. For histopathologic studies, the aorta and heart were harvested and stored in buffered formalin (10%), or the brachiocephalic artery (BCA) was dissected and snap-frozen in Tissue-Tek OCT-compound (Sakura Finetek, Torrance, CA) to be used for cryostat sections.

Pieces (10 mg each) of lung, heart, and aorta from Ho1-luc mice were removed, placed in 0.2 mL tissue lysis buffer (Promega, Madison, WI), chopped into fine pieces, and left overnight at 4°C on a shaker. After brief centrifugation, supernatants of the tissue lysates were collected and luciferase activity was measured using a commercial kit (Promega) and a MicroLumat Plus LB 96V luminometor (EG&G Berthold, Bad-Wildbad, Germany).

The QPCR assay, which evaluates the average lesion frequency per strand for the two template strands in the segment of interest, was conducted according to a protocol described previously (Ballinger et al. 1999, 2000) with slight modifications. Briefly, genomic DNA was isolated using the Qiagen genomic tip 20G kit (Qiagen Inc., Valencia, CA). The DNA concentration was measured using the PicoGreen dsDNA quantitation kit (Molecular Probes, Eugene, OR). QPCR of extra long amplification product was carried out using the Expand 20 kb PCR^plus System (Roche, Mannheim, Germany). A mouse 12-kb mtDNA fragment was amplified using the following primers: F5′-AGAGAT-TTCTCTACACCTTCGAATTTGC-3′ and R 5 ′ - A T T T A A T T T G A G G G T G A -CGGGCGG-3′. The PCR reaction system contained 100 ng DNA as a template. Amplification consisted of an initial cycle of denaturation (92°C, 2 min) followed by 25 cycles of denaturation (92°C, 30 sec), annealing (65°C, 30 sec), and extension (68°C, 15 min), with a final extension (68°C, 10 min) cycle. A short (180 bp) fragment from the D-loop part of mtDNA was amplified using the following primers: F5′-TAATCAGCC-CATGACCAACA-3′ and R5′-GATTGGG-TTTTGCGGACTAA-3′. Ten microliters of PCR products were applied to gel electrophoresis (0.8% agarose gel, 75 V, 1 hr). Gels were stained by ethidium bromide, imaged using the Alphaimager System (Alpha Innotech Corp., San Leandro, CA), and quantified by scanning laser densitometry.

Mitochondria were isolated from fresh aortas by differential centrifugation (Young et al. 2005). Briefly, tissues were homogenized in isotonic buffer containing 10 μg/mL protease inhibitors (aprotinin, phenylmethylsulphonylfluoride, leupeptin, sodium orthovanadate) and BHT (5 mM) at 4°C. Tissue homogenates were centrifuged at 1,500 × g and 4°C for 15 min, and the supernatants were collected. The supernatants were centrifuged at 10,000 × g for 30 min at 4°C, and the mitochondria were collected in the pellet. Mitochondrial protein concentration was measured in BSA-free isolation buffer by a Lowry Assay (Bio-Rad Corp., Hercules, CA). Samples were stored at −70°C.

Mitochondrial reduced glutathione/oxidized glutathione (GSH/GSSG) ratios were measured by a Bioxytech GSH/GSSG-412 colorimetric assay kit (OxisResearch, Portland, OR) with minor modifications. Briefly, 10 μL 1-methyl-2-vinylpyridinium trifluoromethane-sulfonate was added to scavenge for GSH during the mitochondrial isolation. The reaction mixture was added to a 96-well microtiter plate at a volume of 50 μL and incubated at room temperature for 5 min. NADPH was added to each well, and the change in absorbance at 412 nm was recorded in a microplate reader (Molecular Device, Sunnyvale, CA) for 3 min. GSH and GSSG levels were normalized to protein concentrations and the GSH/GSSG ratio was calculated. Mitochondrial carbonyl groups were detected by slot blot (Amersham Biosciences, Piscataway, NJ). Briefly, 10 μL mitochondrial protein was derivatized with 2,4-dinitrophenylhydrazine and blotted to polyvinylidene fluoride membrane (Amersham Biosciences, Piscataway, NJ). The dirivatized carbonyl groups, detected by rabbit anti-dinitrophenol (1:500; Molecular Probes) and horseradish peroxide–conjugated anti-rabbit IgG (1:1,000) were visualized by an ECL system (Amersham, Piscataway, NJ). Band densities were quantified by densitometry (ImageQuant, Molecular Dynamics).

We determined the extent of aortic atherosclerotic lesions by the en face method, as described previously (Simeonova et al. 2003). Briefly, the aorta was fixed in 10% neutral buffered formalin, opened longitudinally, and stained with Sudan IV. Aorta images were captured with a Sony DXC-960MD three-chip-coupled device color video camera (Sony, Tokyo, Japan), connected to an Olympus SZX12 dissection microscope (Olympus, Melville, NY). Among the individual segments of the same aorta, the smallest coefficient of variation occurred in the segment that contained the most extensive lesions (Palinski et al. 1994); thus the thoracic part of the aorta was used for quantitative assessment. Quantitative analysis of the lesion size was performed using a Simple PCI C-imaging system (Compix Inc., Cranberry Township, PA).

Plaque formation in brachiocephalic arteries was quantified as described by Teupser et al. (2003). Briefly, BCAs were sectioned from distal to proximal at a 10-μm thickness. Atherosclerotic lesions luminal to the internal elastic lamina were quantified in eight equidistant oil red O–stained sections 200, 300, 400, 500, 600, 700, and 800 μm from the branching point of the BCA into the carotid and subclavian arteries. The images were captured by a Sony DXC-960MD three-chip-coupled device color video camera (Sony) connected to an Olympus AX70 microscope (Olympus) and analyzed by Simple PCI C-imaging system (Compix Inc., Sewickley, PA). For immunostaining analysis, we applied specific rat anti-mouse monoclonal antibodies for Mac-3 (a marker of monocytes/macrophages) and vascular cell adhesion molecule 1 (VCAM-1) (both from BD PharMingen, San Diego, CA) in an immunostaining protocol conducted on acetone-fixed transverse cryosections; the positive staining was evaluated as described previously (Warren et al. 2005).

Analysis of cytokines from plasma samples was conducted using a mouse inflammation cytometric bead array kit and a FACSCalibur flow cytometer (both from BD Biosciences, San Diego, CA) following the manufacturer’s instructions. Standard curves were determined for each cytokine from the range of 20–5,000 pg/mL. The lower detection limit of the assay is 2.5–52.7 pg/mL, depending on types of cytokines. The following cytokines were measured: interleukin-6 (IL-6), monocyte chemoattractant protein-1 (MCP-1), interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), and IL-12.

All experiments were replicated, and representative findings are shown. Differences among treatment groups were detected by analysis of variance. Differences in which the p-value was < 0.05 were considered statistically significant. When significant differences were detected, post hoc comparisons between treatment groups were examined with the Holm-Sidak method.

We conducted initial screening for extrapulmonary effects using Ho1-luc reporter transgenic mice. To evaluate HO-1 gene expression, a surrogate biomarker of oxidative stress (Prawan et al. 2005; Rao et al. 2003), we measured the activation of tissue luciferase in Ho1-luc reporter transgenic mice. HO-1 demonstrated a trend for activation in the lung, aorta, and heart tissue 7 days after SWCNT exposure; the activity declined approximately to control levels by 28 days after exposure (Figure 1).

Furthermore, we conducted specific evaluation of vascular oxidative perturbations in C57BL/6 mice. Because circular mtDNA is highly susceptible to oxidative damage and mtDNA damage has been suggested as an initiating event in atherogenesis (Ballinger et al. 2000, 2002), we assessed whether pulmonary SWCNT exposure induces aortic mtDNA damage. Significant reduction in mtDNA amplification, reflecting an increase in mtDNA damage, was observed in the aortic tissue of C57BL/6 mice exposed to SWCNT (10 and 40 μg/mouse) at days 7, 28, and 60 after a single exposure, compared with age-matched, vehicle-treated control mice (p < 0.01) (Figure 2A). Maximum mtDNA damage occurred 7 days after exposure in both treatment groups and persisted in the high-dose treatment group 60 days after exposure. Amplification of a small fragment of the D-loop of the mitochondrial genome (Figure 2A), as well as nuclear 18S rRNA (data not shown), was not affected by the exposure. Accordingly, the amplification of the large mtDNA fragment was normalized to the amplification of the small fragment. In contrast to SWCNT, UfCB pharyngeal instillation did not result in mtDNA damage in the aortic tissue at comparable doses of exposure (Figure 2B).

Changes in mitochondrial GSH levels are also manifestations of oxidative stress (Suliman et al. 2004). The aortic mitochondrial GSH/GSSG ratios were significantly reduced 7 days after pulmonary exposure to a single high dose of SWCNT (Figure 3A). Oxidative stress also may lead to oxidation of amino acid residues on proteins, forming protein carbonyls (Dalle-Donne et al. 2003). Consistently, significant accumulation of aortic mitochondrial protein carbonyl groups occurred in SWCNT-exposed mice compared with the age-matched vehicle-exposed mice at both exposure levels (p < 0.05) (Figure 3B).

The effects of pulmonary SWCNT exposure on atherosclerosis development were evaluated in ApoE^−/− mice. ApoE^−/− mice were exposed chronically to PBS or SWCNT (four instillations of SWCNT at 20 μg/mouse every other week for 8 weeks) and fed a regular chow diet (regimen 1) or a fat chow diet for 4 weeks, followed by regular chow for 4 weeks (regimen 2). Vehicle-exposed ApoE^−/− mice on regimen 1 had lower plasma cholesterol (p ≤ 0.01) and minimal plaque occurrence in the aorta compared with the mice on regimen 2. The body weights, as well as plasma cholesterol, glucose, and lactate dehydrogenase (LDH) levels were comparable in SWCNT-and vehicle-exposed mice under both diet regimens (Table 1). Plaque formation, although appearing slightly increased after exposure to SWCNT, compared with the vehicle exposure in mice on regimen 1, was not quantified because the size of the plaques was not large enough for meaningful analysis or comparisons to be made (representative images presented in Figure 4A). However, in mice on regimen 2, the aortic morphometric en face analysis demonstrated a significant increase in the plaque areas in SWCNT-exposed mice compared with the vehicle-treated mice (p < 0.05) (Figure 4B, C). Histologic assessment of the lesion size in the BCAs of these mice confirmed that SWCNT exposure accelerated atherosclerosis progression (Figure 5). The BCA was quantified at seven sections: 200, 300, 400, 500, 600, 700, and 800 μm proximal from its branching site into the carotid and subclavian arteries. SWCNT exposure was associated with considerably larger atherosclerotic lesion area at each BCA location examined (representative images shown in Figure 5A). The overall lesion area of the BCA was significantly higher in the SWCNT-exposed mice than in the PBS-treated mice (mean ± SE, 447,239 ± 164,242 μm² and 192,911 ± 90,755 μm², respectively; p < 0.001; Figure 5B).

To assess the influence of SWCNT exposure on inflammatory characteristics of the plaque, serial sections of the BCA were immunostained with antibodies to Mac-3 (Ralph et al. 1983) and VCAM-1 (Matheny et al. 2000). Consistent with the larger size of the plaques, the number of macrophages and VCAM-1–positive cells were proportionally higher in the BCA plaques from SWCNT-exposed mice compared with plaques in PBS-treated mice (Figure 5C, D). In the evaluated sections, we found no evidence that SWCNT exposure resulted in increased immunostaining in the vascular wall, which was not affected by plaque formation.

We also evaluated the involvement of systemic inflammation and aortic mtDNA damage in SWCNT-related atherosclerosis progression. The plasma MCP-1, IL-12, IL-6, TNF-α, and IFN-γ levels were not significantly different in ApoE^−/− mice treated with PBS or SWCNT (Figure 6A). However, the aortas of SWCNT-exposed ApoE^−/− mice demonstrated significantly increased mtDNA damage.