The present study was designed to determine if naturally-occurring oils with antioxidant properties could be utilized to provide protection against proinflammatory challenges to the upper respiratory tract. The mechanisms involved with such protection would likely include their direct ability to scavenge ROS that arise as a consequence of toxicant exposure in the nasal epithelial mucosa. In addition, the presence of molecular constituents within these oils could presumably have an effect on antioxidant gene expression within mucosal cells, contributing further to their innate defense against agents that induce inflammation through oxidant-related pathways.
Exposure to ozone has long been known to lead to an inflammatory response in the upper and lower respiratory tracts characterized by the influx of PMNs [4, 41]. Under conditions of controlled exposure of subjects to 0.25 ppm ozone for 2 hours in the present study, this response was observed in the upper respiratory tract. Pretreatment of the nasal passages by aerosol spray with a natural oil preparation inhibited the inflammatory response. Because of the reactive target that the antioxidant oil mixture and the soy oil carrier (diluent) might present to inhaled ozone or its reactive products, some degree of protection could have been provided by a simple "barrier" effect . This would likely be greatest at times early in the exposure period prior to a reduction in surface oil levels that might result from nasal mucociliary clearance . If this were the sole mechanism involved, it might be expected that administration of the oil would reduce or, at best, eliminate the ozone-induced influx of inflammatory cells. However, the data demonstrated that PMN levels in the nasal lumen at the 18 hour post-exposure time point were significantly reduced below those observed at baseline in the pre-exposure samples. This observation suggests that proinflammatory signaling was abrogated in the nasal mucosa by the treatment, possibly through mechanisms involving increased intracellular antioxidant activity leading to reduced inflammatory drive. In addition, it indicates that this activity persisted in the tissue, at least to the sampling point 18 hours after the ozone exposure period.
Ozone has been shown to stimulate the influx of PMNs into the airway lumen as early as 1 hr following exposure  suggesting very rapid initiation of pro-inflammatory signaling. In the present study, the effectiveness of the oil in reducing inflammation following exposure to ozone could have had several temporal components. In addition to the likely ROS scavenger effect of the mixed oil preparation previously described, pre-treatment of the nasal mucosa prior to exposure could have stimulated early activation of intracellular antioxidant pathways to increase baseline cellular protection. Involvement of antioxidant genes with varying kinetic profiles of activity could provide both immediate and more prolonged antioxidant capacity. The activity of such a mechanism would be consistent with the observed early induction of the rapidly-responding enzyme, HO-1. HO-1 gene expression was increased by 3-fold in cultured cells within 3 hours of oil treatment and was also found to be elevated in the nasal epithelium of naïve subjects 8 hr following administration of the mixed oil preparation. Furthermore, the demonstration of related antioxidant genes with delayed expression kinetics in the cell culture studies provides the mechanistic basis for a sustained antioxidant effect.
The four antioxidant genes investigated in this study, HO-1, NQO1, GCLc, and GCLm, are known to have a common antioxidant response element (ARE) in their promoters and are expressed in an Nrf2-dependent manner. The basic leucine zipper (bZip) transcription factor, Nrf2, acting via an antioxidant/electrophile response element, regulates the global expression of a family of antioxidant enzymes and functions to maintain cellular redox homeostasis . The present investigation of the mixed oil preparation demonstrated that the Nrf2-associated gene and protein expression observed was predominantly associated with the orange oil component. Nrf2 activation by that oil was confirmed in a battery of cell culture studies that demonstrated activation kinetics, dose-dependency, translocation of Nrf2 protein to the nucleus, and the gene and protein expression of Nrf2-activated antioxidants. Although the focus of the present study was directed toward the Nrf2 system, the possibility remains that other components of the mixed oil preparation activated additional antioxidant or anti-inflammatory-associated transcription factors that contributed to the observed reduction of nasal mucosal inflammation.
Among the antioxidant genes studied, HO-1 induction in response to oil treatment was the most rapid and dramatic. Heme oxygenase catalyzes the rate limiting steps of heme oxidation to biliverdin, carbon monoxide and iron. Biliverdin is rapidly converted to bilirubin, a potent endogenous antioxidant. Three isoforms of heme oxygenase have been reported: the inducible HO-1 and the constitutively expressed HO-2 and HO-3. An increasing number of studies implicate HO-1 in the regulation of inflammation. The induction of HO-1 has been demonstrated in many models of lung injury including hyperoxia, endotoxemia, bleomycin, asthma, acute complement-dependent lung inflammation, and heavy metals [49, 50].
In an in vitro model of oxidative stress using pulmonary epithelial cells stably transfected to over-express HO-1, Lee et al.  demonstrated that these cells exhibited increased resistance to hyperoxic cell injury. In studies by Petrache et al.  and Soares et al. , HO-1 also prevented TNF-α-mediated apoptosis in fibroblasts and endothelial cells, respectively. Such findings further underscore the importance of HO-1 in cytoprotection and the potential prophylactic benefits of its up-regulation.
Otterbein and colleagues [54–56] have demonstrated that HO-1 induction correlated with cytoprotection against oxidative stress in vivo. Using hyperoxia as a model of acute respiratory distress syndrome in rats, they demonstrated that the exogenous administration of HO-1 by gene transfer could confer protection against oxidant-induced tissue injury. Adenoviral gene transfer of HO-1 (Ad5-HO-1) into the lungs of rats resulted in increased expression of HO-1 and, importantly, induced a marked resistance to hyperoxic lung injury [56, 57]. Rats treated with Ad5-HO-1 showed reduced levels of hyperoxia-induced pleural effusion, neutrophil alveolitis, and bronchoalveolar lavage protein leakage. Furthermore, rats over-expressing HO-1 showed increased survivability in the presence of hyperoxic stress versus those treated with the vector control virus [56, 57].
Another of the antioxidants observed to undergo up-regulation was NQO1, an enzyme primarily expressed in tissues requiring a high level of antioxidant protection, such as the epithelial cells of the lung, breast, colon, and vascular endothelium. Expression of NQO1, suggests that this molecule may play a key role in establishing the antioxidant capacity in these cells . Oxidant pollutants, including diesel exhaust particles, induce NQO1 expression which plays a role in mitigating pollutant-enhanced IgE responses [58, 59]. Furthermore, over-expression of phase II enzymes, including NQO1, inhibited IgE production and supports the concept that chemical up-regulation of these enzymes may represent a chemopreventative strategy in airway allergic diseases [58, 59]. Thus, the oil-induced inductions of NQO1 and related antioxidants may have broader implications for protection of the respiratory mucosa, extending to pollutant-related pro-allergenic effects in susceptible individuals.
Cellular antioxidant defenses can counter inflammation by limiting the levels of ROS generated. Expression of genes involved in glutathione biosynthesis (i.e. GCLc, and GCLm) was significantly up-regulated in response to oil pretreatment. Further, pretreatment of BEAS-2B cells with the mixed oil preparation or orange oil alone was observed to greatly suppressed TNFα activation in response to LPS treatment. Lower levels of glutathione have been reported to augment activation of the proinflammatory transcription factor, NF-κB , consistent with our previous report of a protective role of glutathione peroxidase in LPS-induced septic inflammation . In addition, induction of HO-1 may exert anti-inflammatory functions through the generation of carbon monoxide and has been shown to inhibit the LPS-induced expression of proinflammatory cytokines .
The results of the present study indicate that the mixture of natural oils was capable of reversing the nasal inflammatory response to ozone exposure in healthy human subjects in a manner that persisted for up to 18 hours. In human airway epithelial cells in culture, short duration (15 min) treatment with the mixture resulted in activation of Nrf2 and increases in the expression of several representative antioxidant genes with both rapid response and late activation profiles. The effectiveness of the short treatment duration was surprising, given the significantly longer period identified in our preliminary studies (12 hr) and utilized by others  (24 hr) to elicit robust activation of Nrf2-dependent enzyme systems using the water-soluble Nrf2 activator, sulforaphane. It may be that the oil-based preparation allows for rapid integration of the active component(s) into the cell membrane which then acts as a repository for sustained release into the cell. It will be of interest to determine if the presence of a lipid-based carrier plays a role in increasing the duration of antioxidant pathway activation. Consistent with the observed up-regulation of Nrf2-activated antioxidant genes in cultured airway epithelial cells, treatment of test subjects with the oil mixture by nasal spray induced detectable increases in nasal mucosal HO-1 gene expression. Although the easiest of the genes to assess in vivo due to its rapid- and high-responding profile, the increased expression of this gene in vivo provides support for the notion that the mechanism of action identified for the oil in the cell culture studies is associated with the protection observed in the nasal ozone intervention study.
Although seemingly unrelated in terms of their sources and the nature of their interactions with the respiratory system, many environmental challenges share the development of cellular oxidative stress as a common pathway for cell activation, inflammation and, in some cases, cytotoxicity. Such challenges include virus and bacterial infection [10, 63], allergen challenge , and exposure to common gaseous and particulate air pollutants [65, 66], tobacco smoke , and bacterial endotoxin . It is important to note that many of these diverse exposures occur in combination and may synergize to produce greatly amplified responses within epithelial cells of the respiratory tract. For example, in a study of human bronchial epithelial cells in culture, it was observed that the consequences of oxidant stress induced by the oxidant pollutants ozone or nitrogen dioxide in combination with rhinovirus infection resulted in release of the proinflammatory mediator, IL-8, at levels as much as 2.5-fold greater than those predicted by the individual exposures . Thus, targeting oxidant-driven pathways leading to inflammatory responses in the upper respiratory tract may offer a means to provide cytoprotection against a range of environmental challenges to those tissues. Such antioxidant strategies may be especially beneficial in individuals who have reduced or absent phase II enzyme activity, as may result from certain genetic polymorphisms. Furthermore, the unique mechanisms of action afforded by natural oil-derived preparations may offer opportunities to broaden therapeutic approaches for those individuals who are poorly responsive to current treatments.