Factoring-in agglomeration of carbon nanotubes and nanofibers for better prediction of their toxicity versus asbestos

Abstract

Background: Carbon nanotubes (CNT) and carbon nanofibers (CNF) are allotropes of carbon featuring fibrous morphology. The dimensions and high aspect ratio of CNT and CNF have prompted the comparison with naturally occurring asbestos fibers which are known to be extremely pathogenic. While the toxicity and hazardous outcomes elicited by airborne exposure to single-walled CNT or asbestos have been widely reported, very limited data are currently available describing adverse effects of respirable CNF.

Results: Here, we assessed pulmonary inflammation, fibrosis, oxidative stress markers and systemic immune responses to respirable CNF in comparison to single-walled CNT (SWCNT) and asbestos. Pulmonary inflammatory and fibrogenic responses to CNF, SWCNT and asbestos varied depending upon the agglomeration state of the particles/fibers. Foci of granulomatous lesions and collagen deposition were associated with dense particle-like SWCNT agglomerates, while no granuloma formation was found following exposure to fiber-like CNF or asbestos. The average thickness of the alveolar connective tissue--a marker of interstitial fibrosis--was increased 28 days post SWCNT, CNF or asbestos exposure. Exposure to SWCNT, CNF or asbestos resulted in oxidative stress evidenced by accumulations of 4-HNE and carbonylated proteins in the lung tissues. Additionally, local inflammatory and fibrogenic responses were accompanied by modified systemic immunity, as documented by decreased proliferation of splenic T cells ex vivo on day 28 post exposure. The accuracies of assessments of effective surface area for asbestos, SWCNT and CNF (based on geometrical analysis of their agglomeration) versus estimates of mass dose and number of particles were compared as predictors of toxicological outcomes.

Conclusions: We provide evidence that effective surface area along with mass dose rather than specific surface area or particle number are significantly correlated with toxicological responses to carbonaceous fibrous nanoparticles. Therefore, they could be useful dose metrics for risk assessment and management.

Figure 1

Scanning Electron Microscopy images of CNF (A), asbestos (B), SWCNT (C, Inset: TEM showing individual roped SWCNT), structure numbers per dose (D) and particles size (length) distribution (E) presented as % of total particles.

Figure 2

Schematic representation of the SWCNT bundles and CNF. (A) A hexagonal arrangement model of a bundle containing two layers of SWCNT. The labels d_SWCNT and d_bundle correspond to the diameters of SWCNT and a bundle of SWCNT, respectively. (B) A model of CNF showing five carbon layers. The labels d_OD, and d_HC correspond to the outer and hollow core diameters of CNF, respectively.

Figure 3

Cell profile in BAL fluids of C57BL/6 mice after pharyngeal aspiration with CNF, Crocidolite Asbestos, or SWCNT. A: polymorphonuclear leukocytes (PMNs); B: alveolar macrophages (AMs). Open columns - exposure with SWCNT (40 μg/mouse); gray columns - exposure with CNF (120 μg/mouse); black columns - exposure with crocidolite asbestos (120 μg/mouse). Mice were exposed via pharyngeal aspiration to doses indicated. Animals were sacrificed 1, 7, and 28 days post exposure. Average control (PBS-treated mice) values for PMNs (cells, x10³) on day 1, 7 or 28 post exposure were 1.58 ± 0.69, 1.39 ± 0.45 or 0.71 ± 0.24, respectively. Average control values for AMs (cells, x10³) on day 1, 7 or 28 post exposure were 389.06 ± 22.80, 345.01 ± 30.01 or 229.79 ± 27.81, respectively. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control mice, ^αp < 0.05 vs. mice exposed to CNF, ^βp < 0.05 vs. mice exposed to asbestos.

Figure 4

Pulmonary cell damage as evaluated by changes in LDH activity and air/blood barrier damage was evaluated by protein in the bronchoalveloar lavage fluid of C57BL/6 mice in response aspiration of CNF, Crocidolite Asbestos or SWCNT. A: protein; B: LDH. Open columns - exposure with SWCNT (40 μg/mouse); gray columns - exposure with CNF (120 μg/mouse); black columns - exposure with crocidolite asbestos (120 μg/mouse). Mice were exposed via pharyngeal aspiration to the doses indicated. Animals were sacrificed 1, 7, and 28 days post exposure. Average control (PBS-treated mice) values for protein (mg/ml) on day 1, 7 or 28 post exposure were 0.31 ± 0.05, 0.33 ± 0.05 or 0.38 ± 0.05, respectively. Average control values for LDH (U/ml) on day 1, 7 or 28 post exposure were 25.12 ± 0.70, 26.81 ± 1.03 or 25.47 ± 1.06, respectively. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control mice, ^αp < 0.05 vs. mice exposed to CNF, ^βp < 0.05 vs. mice exposed to asbestos.

Figure 5

Biomarkers of oxidative stress in the lung of C57BL/6 Mice following aspiration with CNF, Crocidolite Asbestos, or SWCNT. A: 4-Hydroxynonenal; B: Protein Carbonyl. Mice were exposed via pharyngeal aspiration to the doses indicated. Open columns - exposure with SWCNT (40 μg/mouse); gray columns - exposure with CNF (120 μg/mouse); black columns - exposure with crocidolite asbestos (120 μg/mouse). Animals were sacrificed 1, 7, and 28 days post exposure. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control (PBS) exposed mice; ^αp < 0.05 vs. CNF exposed mice; ^βp < 0.05 vs. asbestos exposed mice.

Figure 6

Cytokine accumulation in bronchoaleolar lavage fluid of C57BL/6 mice following aspiration of CNF, Crocidolite Asbestos, or SWCNT. A: TNF-α; B: MCP-1; C: IL-6; D: IL-12p70; E: IFN-γ; F: IL-10. Open columns - exposure with SWCNT (40 μg/mouse); gray columns - exposure with CNF (120 μg/mouse); black columns - exposure with crocidolite asbestos (120 μg/mouse). Mice were exposed via pharyngeal aspiration to the indicated doses. Animals were sacrificed 1, 7, and 28 days post exposure. Means ± SE (n = 6 mice per group).*p < 0.05 vs. control (PBS) exposed mice, ^βp < 0.05 vs. mice exposed to asbestos.

Figure 7

Fibrogenic response as assessed by cytokine release in BAL fluid and collagen accumulation in the lung of C57BL/6 mice following CNF, Crocidolite Asbestos or SWCNT exposure. A: TGF-β; B: Total collagen measured on day 28 post exposure. Open columns - exposure with SWCNT (40 μg/mouse); gray columns - exposure with CNF (120 μg/mouse); black columns - exposure with crocidolite asbestos (120 μg/mouse). Mice were exposed via pharyngeal aspiration to the indicated doses and animals were sacrificed 1, 7, and 28 days post exposure. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control (PBS) exposed mice; ^αp < 0.05 vs. mice exposed to CNF, ^βp < 0.05 vs. mice exposed to asbestos.

Figure 8

Morphometric changes (A) and Sirius red-stained lung sections (B-E) from C57BL/6 mice 28 days following exposure to CNF, Crocidolite Asbestos or SWCNT. A: Collagen fiber content determined as average thickness of connective tissue; Inset: Granulomatous cellular tissue; B: Control; C: CNF; D: Asbestos; E: SWCNT (Arrow indicate collagen accumulation). Mice were exposed via pharyngeal aspiration to 120 μg/mouse CNF or Asbestos, or 40 μg/mouse SWCNT. Animals were sacrificed 28 days post exposure. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control (PBS) exposed mice, ^αp < 0.05 vs. mice exposed to CNF, ^βp < 0.05 vs. mice exposed to asbestos.

Figure 9

Light micrographs of H&E-stained sections from lung of C57BL/6 mice 28 days following exposure to PBS (A), CNF (B), Crocidolite Asbestos (C) or SWCNT (D). Mice were exposed via pharyngeal aspiration to 120 μg/mouse CNF or Asbestos, or 40 μg/mouse SWCNT. Animals were sacrificed 28 days post exposure.

Figure 10

Splenic T Cell Proliferation 7 and 28 days following pulmonary exposure to CNF, Crocidolite Asbestos or SWCNT. Mice were exposed via pharyngeal aspiration to 120 μg/mouse CNF or Asbestos, or 40 μg/mouse SWCNT. Animals were sacrificed 7 and 28 days post exposure (white and black bars, respectively) and splenic T cell proliferation was evaluated. Means ± SE (n = 6 mice per group). *p < 0.05 vs. control (PBS) exposed mice.

Figure 11

Correlation between specific surface area or effective surface area of nanomaterials administered and pulmonary outcomes in the lung. (A) Correlation between specific surface area of NM administered (as measured by BET) and pulmonary outcomes in the lung following exposure to Crocidolite Asbestos, CNF, or SWCNT on day 1 post exposure. Pearson's correlation coefficients were calculated for pairs of variables including NM dose (expressed as specific surface area of NM per mouse) and the relative values of the respective pulmonary outcomes: ♦ - NM dose, m²/mouse; ■ - alveolar wall thickness, day 28 post exposure (r = 0.995, p > 0.05); ▲ - 4-HNE, day1 post exposure (r = 0.837, p > 0.05; × - PMN counts, day1 post exposure (r = 0.908, p > 0.05); ✱ - IL-6, day1 post exposure (r = 0.733, p > 0.05); • - BAL protein, day1 post exposure (r = 0.979, p > 0.05). (B) Correlation between effective surface area of NM administered and pulmonary outcomes. Pearson's correlation coefficients were calculated for pairs of variables including NM dose (expressed as effective surface area of NM per mouse) and the relative values of the respective pulmonary outcomes: ♦ - NM dose, m²/mouse; ■ - alveolar wall thickness, day 28 post exposure (r = 0.963, p > 0.05); ▲ - 4-HNE, day1 post exposure (r = 0.974, p > 0.05; × - PMN counts, day1 post exposure (r = 0.997, p < 0.05); ✱ - IL-6, day1 post exposure (r = 0.905, p > 0.05); • - BAL protein, day1 post exposure (r = 0.997, p < 0.05).