Effect of Surface Modification on the Pulmonary and Systemic Toxicity of Cellulose Nanofibrils

Abstract

Cellulose nanofibrils (CNFs) have emerged as sustainable options for a wide range of applications. However, the high aspect ratio and biopersistence of CNFs raise concerns about potential health effects. Here, we evaluated the in vivo pulmonary and systemic toxicity of unmodified (U-CNF), carboxymethylated (C-CNF), and TEMPO (2,2,6,6-tetramethyl-piperidin-1-oxyl)-oxidized (T-CNF) CNFs, fibrillated in the same way and administered to mice by repeated (3×) pharyngeal aspiration (14, 28, and 56 μg/mouse/aspiration). Toxic effects were assessed up to 90 days after the last administration. Some mice were treated with T-CNF samples spiked with lipopolysaccharide (LPS; 0.02-50 ng/mouse/aspiration) to assess the role of endotoxin contamination. The CNFs induced an acute inflammatory reaction that subsided within 90 days, except for T-CNF. At 90 days post-administration, an increased DNA damage was observed in bronchoalveolar lavage and hepatic cells after exposure to T-CNF and C-CNF, respectively. Besides, LPS contamination dose-dependently increased the hepatic genotoxic effects of T-CNF.

Figure 1

Atomic force microscopy (AFM; top row) and scanning electron microscopy (SEM; middle row) images, and fibril lateral size distributions (fitted log–normal models; bottom row) of U-CNF (a,d,g), T-CNF (b,e,h), and C-CNF (c,f,i).

Figure 2

Total number of neutrophils in BAL fluid at 1 (A), 28 (B), and 90 (C) days of the vehicle, CNF samples, positive control, and LPS-spiked T-CNF sample exposure. Data are expressed as mean ± SEM. Asterisks designate statistically significant differences compared with the vehicle group at *p < 0.05 and **p < 0.01. The LPS-spiked T-CNF samples were compared with the corresponding T-CNF treatment (14 μg/mouse/aspiration), and no significant differences were found.

Figure 3

Representative images of hematoxylin- and eosin-stained lung sections of mice exposed to the vehicle (A), CNF [(B–E) 28 μg/mouse/aspiration] or the positive control (F). CNF aggregates are marked with black arrows. (A) Vehicle sample on day 90. (B and inset) Neutrophil influx accompanied by eosinophils in a T-CNF sample 1-day post-exposure. (C) Material in lung tissue in a T-CNF sample 28 days post-exposure. (D and inset) Eosinophilic crystals in a C-CNF sample 90 days post-exposure. (E) Material in lung tissue in a U-CNF sample 90 days post-exposure. (F and inset) Granuloma formation around material aggregates in a positive control sample 90 days post-exposure.

Figure 4

Transmission electron microscope (TEM) micrographs of mouse lung tissue 90 days after repeated pharyngeal aspiration with 14 μg/mouse/aspiration of U-CNF (A,C) or C-CNF (B). Presence of free CNF (black arrows) in the lung parenchyma (A and inset). Bronchoalveolar macrophage containing CNF (B,C, black arrows) and eosinophil crystals (C, red arrow).

Figure 5

DNA damage (percentage of DNA in the comet tail; mean ± SEM) in the bronchoalveolar lavage (BAL), lung, and liver cells of mice at 28 and 90 days after repeated (3×) pharyngeal aspiration with the vehicle and the CNF samples. Asterisks designate statistically significant differences compared with the vehicle group at *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. A statistically significant increase in the percentage of DNA in the tail over the negative control values was induced by the positive control (H₂O₂, 20 mM) in all the experiments performed (1.94 ± 0.4-fold increase; p < 0.001), confirming the validity of the assay (data not shown).

Figure 6

DNA damage (percentage of DNA in comet tail; mean ± SEM) in the bronchoalveolar lavage (BAL), lung, and liver cells of mice 28 and 90 days after repeated (3×) pharyngeal aspiration with T-CNF (14 μg/mouse/aspiration) spiked with 0, 0.02, 1, and 50 ng/mouse/aspiration of lipopolysaccharide (LPS). Asterisks designate statistically significant differences compared with the non-spiked sample at *p < 0.05, **p < 0.01, and ***p < 0.001. A statistically significant increase in the percentage of DNA in tail over the negative control values was induced by the positive control (H₂O₂, 20 mM) in all the experiments performed (1.94 ± 0.4-fold increase; p < 0.001), confirming the validity of the assay (data not shown).