Pulmonary toxicity of MWCNTs has been reported in both mouse and rat models [15–17]. Instillation of MWCNT into the lungs of mice and rats has been shown to induce fibrosis ; however, extensive evaluation of pulmonary function changes in animals instilled with MWCNTs has not been reported. In this study, we demonstrated that 30 days following MWCNT instillation, C57BL/6 mice exhibited changes in pulmonary function that were consistent with pulmonary inflammation, increased collagen deposition and granuloma formation. Additionally, increased levels of Ccl3, Ccl11, and Mmp13 were observed in C57BL/6 mice instilled with different doses of MWCNTs. Taken together, these results suggest that MWCNT exposure could lead to impaired pulmonary function due to inflammatory and fibrotic remodeling of lung tissue.
Due to the implication that MWCNTs may adversely affect human health and safety, appropriate dosing of animals was essential to evaluate the pertinence of these findings in regard to human exposure levels. Studies conducted in industrial plants indicated nanoparticle exposure levels up to 0.5 mg/m3 for an 8-hour work day and 40-hour work week . Additional evaluations of carbon nanotubes in manufacturing and research facilities, found airborne levels during handling to be as low as 53 μg/m3
 and as high as 400 μg/m3
[4, 15]. Shvedova et al., report that human occupational exposure levels of 5 mg/m3 over the course of an 8-hour day and 40-hour work week equate to approximately 20 μg of MWCNT aspiration in a mouse model . Current proposed guidelines by the National Institute for Occupational Safety and Health (NIOSH) limits exposure to 7 μg/m3, the lowest detectable level of airborne CNT using the latest analytical methods . The doses used in this study, 1, 2 and 4 mg/kg MWCNTs (equivalent to 27, 54, and 108 μg of MWCNT in the average 27 g mouse used in this study), were selected based on these limited exposure studies and doses previously reported in the literature for rodent studies . Furthermore, the MWCNTs were dispersed in 10% surfactant in saline, a physiologically relevant medium, shown to effectively reduce aggregation of nanotubes in solution . Data from this study demonstrated a persistent inflammatory response and the development of granulomatous and collagen-rich fibrotic tissue in C57BL/6 mice, similar to previously reported findings [4, 7]. Post-exposure to MWCNTs at 90 days exhibited granulomatous foci similar to those found in chronic human granulomatous disorders . In fact, due to their ability to produce a robust granulomatous response, the use of MWCNTs has been proposed as a novel murine model to study chronic granulomatous disease . In the present study, we demonstrated that focal aggregates of MWCNTs were present at the core of the granulomas and surrounding fibrotic tissue (Figure 4D). In addition, our data indicated that intact MWCNTs do not congregate to form airway granulomas, but are dispersed throughout the lung tissue (Figure 4B) contrary to reports from previous investigators . Furthermore, studies have shown that MWCNTs are not only found in the periphery of the lungs post-exposure, but have also been found to translocate to the pleural space . Data from this study are consistent with previous reports that MWCNTs induce pulmonary inflammation and fibrosis, and are widely distributed through the lung following exposure. Further, we have now shown that these severe adverse pulmonary responses have a negative impact on pulmonary function.
In this study, pulmonary functional variables were evaluated in C57BL/6 mice exposed to different doses of MWCNTs using the FlexiVent system. The Snapshot perturbation was imposed to measure resistance (R), dynamic compliance (C), and elastance (E) of the whole respiratory system (including airways, lung, and chest wall). Our data showed that R was significantly increased at the highest dose of MWCNT with a concomitant declined in dynamic compliance (C). The decrease in C is consistent with findings in a bleomycin-induced model of pulmonary fibrosis in mice . However, that study did not demonstrate an increase in R. This discrepancy is likely due to the fact that inflammation was attenuated by use of cyclophosphamide. Thus, in our study, the consistent elevation in numbers of inflammatory cells may have contributed to the increased R observed. Because R reflects the combined resistance contributed by both the airways and lung parenchyma, we employed the constant-phase model to distinguish between central and peripheral respiratory mechanics and to provide information about the heterogeneity of the respiratory response .
Tissue damping (G), which reflects parenchymal distortion, showed a dose-dependent increase in MWCNT instilled mice which was likely due to inflammatory infiltrates, as well as granuloma formation in the peribronchiolar and alveolar regions of the lung. Since these structures account for a major portion of the cross-sectional area of the lung, any obstruction in the distal airways and/or alveolar spaces could contribute to an increase in R. Furthermore, the decrease in C may suggest that the alveoli are not fully expanding as a result of alveolar volume reduction which is likely due to inflammation and/or granuloma formation. This would reduce the traction normally exerted on the conducting airways at high lung volume and thereby also contribute to changes in resistance (R). Similar to Rn, H, a measure of tissue elastance, also displayed a non-significant trend towards increased elastance with increasing doses of MWCNT (data not shown). Tissue elastance is typically elevated with fibrotic lung disease; but the patchy nature of the granulomatous/fibrotic lesions seen in this model may account for the lack of significance. However, eta (G/H) was dose-dependently increased with MWCNT instillation and was significantly elevated at the highest dose, indicating heterogeneity in the combined inflammatory and fibrotic responses to MWCNTs. On the other hand, Rn showed a trend toward dose-dependent increases that did not reach significance, suggesting that MWCNTs may have a minor effect on central airways. In support of this, we observed only minor inflammation within the central airways. However, this effect, while small with MWCNTs alone, may explain enhancement of airway responses to methacholine in models of allergic airway disease by exposure to air pollution  observed in other studies. Along the same line, our study showed a MWCNT-induced increase in eosinophil numbers and eotaxin levels (Ccl11) which may also contribute to MWCNT enhancement of allergic airway disease.
Static compliance (Cst) and the upper portion of the deflation PV curve (K), parameters derived from PVr-P maneuvers, declined significantly at the highest dose of MWCNTs which is typical of fibrotic lung disease. In contrast to our findings, Kamata et al.  reported that carbon black nanoparticles had no effect on lung compliance. This might be due to the physicochemical differences between carbon black nanoparticles and MWCNTs. A decrease in Cst was also observed in the vehicle treated mice suggesting that the administration of high doses of pulmonary surfactant reduces lung compliance.
Taken together, our pulmonary function findings suggest that MWCNT exposure results predominantly in peripheral respiratory disease as a result of combined inflammatory infiltrates and granulomatous/fibrotic parenchymal responses, reflected by a decline in pulmonary function. Consistent with other reports, we observed significant granuloma formation with fibrotic content which likely contributed to restrictive changes in pulmonary function . Translocation and penetration of MWCNTs into pleural space, as previously shown, may also contribute to peripheral respiratory injury and subsequent structural changes [7, 26].
To begin examining potential molecular mechanisms by which MWCNT instillation mediates impaired pulmonary function, gene and protein expression of cytokines and chemokines were examined to identify fibrotic and inflammatory responses. PCR arrays were utilized to identify changes in gene expression related to the development of fibrosis. We identified Mmp13, Ccl3, and Ccl11 as three highly upregulated gene products. Mmps can be divided by structure and substrate specificity into several subgroups including collagenases, gelatinases, stromelysins, and membrane-type (MT) Mmps . Imbalanced expression of MMPs has been associated with fibrosis . Mmp13, an interstitial collagenase, is considered a key activator in the cascade of proinflammatory reactions leading to pulmonary fibrosis . Furthermore, activation of Mmp13 enhances the process of macrophage chemoattraction and infiltration of other inflammatory cells following tissue injury. MWCNT instilled mice exhibited a dose dependent increase in Mmp13 production and activity in BAL fluid. Additionally, Ccl3 (also named MIP-1α) is a critical macrophage chemoattractant in murine wound repair . In our study, increased Ccl3 expression in the lung and Mmp13 activity in BAL fluid following MWCNT exposure are likely associated with the collagen deposition and granuloma formation in mouse lung exposed to MWCNT. Lastly, Ccl11, involved in eosinophil recruitment, was significantly increased in the lung following MWCNT instillation. This likely contributed to the increasing trend in eosinophils in the BALF seen with increasing doses of MWCNT. Consistent with previous studies , an influx of eosinophils into the lungs by a variety of eosinophil chemoattractants, such as Ccl11, was observed. This classic Th2 driven inflammatory response may contribute to the subsequent fibrotic outcome and change in resistance that was observed in response to MWCNT exposure.
Finally, IL-33 was examined for its potential role in inflammatory cell recruitment and Th2 immune responses following MWCNT exposure. IL-33 is a member of the IL-1 cytokine family and the only known ligand for the ST2 receptor . The ST2 receptor is most highly expressed on mast cells and Th2 lymphocytes, and is known to exist in at least two isoforms; a transmembrane form and a soluble form which is cleaved after activation by IL-33 . Haraldsan et al. describe IL-33, as a potential alarmin, or immune stimulating danger signal during trauma or infection [34, 35]. Studies have confirmed IL-33 as a chemoattractant for human and murine Th2 cells . Furthermore, IL-33 has recently been shown to polarize macrophages to a M2 phenotype (or alternatively activated macrophages), resulting in enhanced production of pro-inflammatory and pro-fibrotic cytokines and chemokines in Th2 driven pathologies . The IL-33/ST2 axis, known to influence these Th2 cell types including mast cells, macrophages and eosinophils, has been recognized in modulating disease pathologies such as anaphylaxis and allergic asthma . A study by Mangan et al. demonstrates that expression of transmembrane ST2 on Th2 cells negatively impacts both pulmonary physiologic and pathologic responses in an OVA-induced pulmonary inflammation mouse model. Changes in airways resistance measures, such as PenH, along with pathologic alterations of pulmonary cell infiltrates and cytokine profiles in ST2 knock-out mice indicate a role for IL-33 and ST2 receptor in regulating allergic inflammatory responses . Further, airway hyper-reactivity following allergen challenge can be attenuated by blockade of the transmembrane ST2 receptor in BALB/c mice . Thus, during injury or insult to the lung, IL-33 upregulation and activation of ST2 receptors may significantly impact Th2 driven cell types resulting inflammatory, fibrotic and physiological changes.
Our data show significant increases in IL-33 expression with MWCNT exposure compared to vehicle control, demonstrating a potential role in inflammation and fibrosis. Mechanistically, the release of chemokines (CCL3 and CCL11), cytokines (IL-33) and the activation of MMP13, help to amplify pro-inflammatory and pro-fibrotic mediators, as well as recruit macrophages, neutrophils, and eosinophils into the lung, thereby contributing not only to the pathologic inflammatory and fibrotic responses, but also the physiologic and functional changes induced by MWCNTs.