Grantee Research Project Results

Final Report: Hypersensitivity to Ozone-Induced Inflammation During Pregnancy and Lactation

EPA Grant Number: R826195
Title: Hypersensitivity to Ozone-Induced Inflammation During Pregnancy and Lactation
Investigators: Gunnison, Albert F. , Chan, L. C.
Institution: New York University Medical Center
EPA Project Officer: Hahn, Intaek
Project Period: November 10, 1997 through November 9, 2000
Project Amount: $412,124
RFA: Exploratory Research - Human Health (1997) RFA Text |  Recipients Lists
Research Category: Human Health

Objective:

The lungs of pregnant and lactating rats have been shown in previous studies to be hyperresponsive to inhaled ozone compared to the lungs of prepregnant or postlactating rats. The major objective of this research project was to elucidate the mechanism responsible for this enhanced sensitivity of pregnant and lactating rats. A published report has shown that acute exposure of rats to a high concentration of ozone caused a several fold enhancement of the transcription factor, NF-κB (nuclear factor κB). Other evidence indicates that oxidant activity causes activation of NF-κB, resulting in an upregulation of the transcription of various genes involved in inflammation. The primary hypothesis examined in this research is that pro-oxidant conditions in pregnant and lactating rats produce higher levels of reactive oxygen species (ROS) during ozone exposure that triggers greater activation of NF-κB in pregnant and lactating rats compared to virgin female rats. The specific objective of this research project was to investigate the mechanism responsible for the relative ineffectiveness of exogenous glucocorticoid in the inhibition of ozone-induced pulmonary neutrophilic inflammation in lactating rats compared to virgin female rats.

Summary/Accomplishments (Outputs/Outcomes):

Exposure of rats to 3 ppm ozone for 6 hours has been shown in another laboratory to significantly upregulate nuclear NF-κB. There was concern in our laboratory that exposure to ozone at this high concentration could cause severe tissue damage, which might stimulate alternative or additional signaling pathways compared to stimulation at lower, more realistic ozone exposures. Therefore, initially, virgin female Sprague Dawley rats were exposed to 2 ppm for 4 and 6 hours, and measurements of nuclear NF-κB were made by electrophoretic mobility shift assay using whole-lung tissue collected 1-2 hours following exposure. No increase in nuclear NF-κB was observed in exposed animals compared to sham-exposed animals (n = 3-4/group). Elevation of the ozone concentration to 3.0 ppm for a duration of 4 or 6 hours also did not produce an increase in nuclear NF-κB compared to air controls (n = 4/group). In spite of the lack of evidence for ozone-induced NF-κB signaling in virgin female rats, the possibility of activation of this signaling in lactating rats was investigated. An exposure regimen of 1 ppm ozone for 6 hours was used because previous experiments in our laboratory had shown that lactating rats exposed to higher concentrations for a 6-hour period experienced severe lung tissue damage and necrosis. Nuclear NF-κB activity of lactating rats following exposure was no greater than that of similarly exposed virgin females (n = 4/group).

The seminal experiment from another laboratory that resulted in a significant NF-κB upregulation following 3.0 ppm ozone for a duration of 6 hours had used Brown Norway (BN) rats, a strain that is sensitive to inhaled allergens and often exhibits an apparent idiopathic pulmonary eosinophilic inflammation. Two experiments were performed in our laboratory with BN rats in which groups of three to four rats were exposed to either 3.0 ppm ozone or air for 6 hours. In one experiment, ozone caused a statistically significant, 2.2-fold upregulation of nuclear NF-κB. However, in the other experiment, the air controls had a robust nuclear NF-κB signal, and this signal did not increase in the ozone-exposed rats. The rats used in the latter experiment¾but not the first experiment¾had a high background level of pulmonary eosinophils, indicating the rats probably had been exposed to some unknown agent (e.g., infectious agent) that had stimulated NF-κB activity and a mild inflammation. A third group of rats received from the vendor also had substantial eosinophilic pulmonary inflammation and was not exposed to ozone. Thus, although BN rats appear to produce nuclear NF-κB in response to ozone exposure at a high concentration, the difficulty in obtaining animals without idiopathic pulmonary eosinophilic inflammation makes it a difficult model to use for investigating ozone toxicitity.

The results obtained from these experiments seeking to relate NF-κB activity to acute ozone exposure suggested that pulmonary inflammation caused by ozone is not mediated by NF-κB, at least in the Sprague Dawley rat. To further explore this relationship between NF-κB activation and pulmonary neutrophilic inflammation, ozone was compared with lipopolysaccharide (LPS), a known potent NF-κB activator and inducer of neutrophilic inflammation. Virgin female Sprague Dawley rats were exposed for 4 hours to LPS aerosol generated by nebulization of solutions containing either 2.2 or 7 mg LPS/L, resulting in chamber concentrations of approximately 2 and 6.5 μg LPS/m³, respectively. Controls were exposed to saline aerosol. Other groups of rats were exposed to either 3 ppm ozone or air for 4 hours. Neutrophilic lung inflammation was determined from the numbers of neutrophils recovered from the lungs by lavage approximately 20 hours following exposure. Nuclear NF-κB was assessed from lung tissue collected 1-2 hours following exposure. Neutrophils recovered multiplied by 10⁶ (mean ± SD) were as follows: 1.06 ± 0.50 for the 2 μg/m³ LPS group (n = 7), 3.08 ± 1.99 for the 3 ppm ozone group (n = 6), and 4.09 ± 2.91 for the 6.5 μg/m³ LPS group (n = 3). The greater numbers of neutrophils recovered from ozone-exposed rats compared to rats exposed to 2 μg/m³ LPS was statistically significant (p < 0.05). The enhancement of NF-κB activity in exposed groups compared to their respective controls (mean ± SD) were as follows: 3.7 ± 1.0 for the 2 μg/m³ LPS group, 1.25 ± 0.4 for the 3 ppm ozone group, and 7.1 ± 2.0 for the 6.5 μg/m³ LPS group (n = 3 exposed and n = 3 controls for each comparison). Comparison of neutrophilic inflammation and NF-κB data shows that although ozone caused greatly enhanced neutrophilic inflammation that was significantly greater than that caused by exposure to 2 μg/m³ LPS, no significant enhancement of NF-κB activity was observed. In contrast, the 2 μg/m³ LPS exposure resulted in significant NF-κB activation. These data suggest that neutrophilic inflammation caused by LPS is mediated by NF-κB, but inflammation caused by ozone is not. In addition, TNFa, which is known to induce NF-κB, was elaborated by cells lavaged from the lungs of rats exposed to 2 μg/m³ LPS, whereas TNFa was not elaborated by cells lavaged from the lungs of rats following exposure to 3 ppm ozone.

Additional experiments were conducted to investigate the possible involvement of the transcription factor AP-1 in ozone-induced inflammation. Sprague Dawley virgin female rats were exposed to 3 ppm ozone (n = 4) or air (n = 4) for 6 hours, and their lung tissue was assayed for nuclear AP-1 activity. Ozone induced a large AP-1 response that was approximately 13-fold greater than the relatively weak baseline signal in the sham-exposed controls. However, exposure of other virgin female Sprague Dawley rats to 1 ppm ozone or air for 6 hours failed to show a clear upregulation by ozone. This suggested that the dramatic AP-1 response observed following 3 ppm ozone exposure occurred¾at least partly¾in response to severe tissue damage, and was not relevant to oxidative signaling at more realistic oxidant exposure levels. Because exposure to ozone stimulated AP-1 activity in virgin females, it was desired to determine if lactating rats exhibited an enhanced AP-1 response. For this comparison, the animals were exposed to 1 ppm for 6 hours to preclude the potential complication of signaling triggered by extensive tissue damage and necrosis. There was no enhancement of nuclear AP-1 signal in the lactating rats relative to the virgin females.

Redirection of Experimental Objectives

The absence of any demonstrative role for NF-κB in ozone-induced inflammation in Sprague Dawley rats precluded conducting some of the originally planned experiments, and necessitated re-evaluation of our experimental approach. Failure to demonstrate a role for NF-κB in ozone-induced inflammation was not interpreted as ruling out a role for ROS or other pro-oxidants in signaling pathways leading to pulmonary inflammation. Because there is considerable evidence in the literature of pro-oxidant conditions in the lungs of pregnant and lactating animals and humans, it is possible that the presence of oxidants in the lungs could initiate or participate in a signaling mechanism for pulmonary inflammation, both in ozone-exposed and unexposed lungs, perhaps mediated by AP-1. The first step in addressing this possibility is to determine whether there is evidence of oxidative stress in the lungs of pregnant and/or lactating rats. The lactating rat appears to be a good model for pursuing this line of investigation, because naïve lactating rats show evidence of idiopathic inflammation, suggesting that a signaling pathway has been activated in these animals. Two previous studies in this laboratory showed that the lungs of naïve lactating Sprague Dawley rats exhibit very mild neutrophilic inflammation. Lactating rats had approximately 3.5 ± 0.7 percent (mean ± SD) neutrophils, and virgin females or postlactating rats had approximately 0.3 ± 0.1 percent neutrophils in the lung cells collected by bronchoalveolar lavage (BAL). The focus of this project was redirected to search for evidence of oxidants and molecular species in the lungs of lactating rats that might function in a signaling pathway resulting in pulmonary inflammation.

8-Isoprostane-Indicator of Oxidative Stress

Arachidonic acid in the extracellular fluid lining the respiratory tract is readily oxidized nonenzymatically to 8-epi-prostaglandin F_2a (8-isoprostane), a stabile eicosanoid that is used as an indicator of oxidative stress. Extracellular lining fluid of naïve lactating and naïve virgin female Sprague Dawley rats (n = 5) was collected by BAL. The lavage fluid was concentrated and analyzed for 8-isoprostane by enzyme immunoassay (EIA). There was no statistically significant difference in 8-isoprostane concentration between lactating and virgin female rats, and, therefore, no support for the hypothesis of increased ROS or oxidative stress in the extracellular environment of cells lining the airways of naïve lactating rats relative to virgin females. A comparison of pregnant and virgin female rats exposed to 1 ppm ozone also gave no difference in 8-isoprostane concentration in lung extracellular lining fluid.

Mitochondrial Damage as an Indicator of Oxidant Exposure

A 50-percent increase in the basal ventilation of lactating rats compared to virgin females is required to satisfy the increased requirement for oxygen that stems from the higher rate of metabolism of lactating rats. This should result in increased electron flow along the electron transport chain in lactating rats. Consequently, a greater leakage of electrons from the mitochondria and a higher concentration of extramitochondrial ROS might initiate a signaling pathway leading to neutrophilic inflammation.

For the purpose of estimating the relative concentration of ROS in lactating and virgin female rats, damage to mitochondrial DNA (mt DNA) was assessed using the extra-long polymerase chain reaction (XL-PCR) amplification technique. The rationale of this assay is that lesions in mt-DNA caused by ROS will prevent or inhibit DNA amplification by PCR. Mt-DNA is much more sensitive to oxidative damage than nuclear DNA, probably because of its location in the cell cytoplasm and the absence of protective histone proteins. Thus, ROS are more likely to cause damaging lesions in mt-DNA than in nuclear DNA. Other laboratories have used XL-PCR to demonstrate the differential sensitivity of mitochondrial and nuclear DNA to the damage inflicted by exogenous hydrogen peroxide. Our laboratory verified the effectiveness of this assay using cultured CCL-149 cells exposed to exogenous hydrogen peroxide.

The possibility of greater mt-DNA damage in the lungs and liver of lactating rats compared to virgin rats was investigated. Groups of Sprague Dawley lactating and virgin female rats were exposed to 1 ppm ozone for 6 hours. DNA was extracted from both lung and liver tissue and mt-DNA, and nuclear DNA was amplified by XL-PCR. Inhibition of mt-DNA amplification in liver DNA was assumed to reflect differences between lactating and virgin rats in the concentration of endogenous ROS, whereas the lung DNA would reflect endogenous ROS plus that generated by ozone exposure. No statistically significant differences were found between lactating and virgin rats in XL-PCR amplification of mt-DNA of either tissue. Thus, no evidence was obtained for greater ROS activity in lactating rats either from endogenous metabolism or as a result of ozone exposure.

AP-1

Experiments showed that exposure to ozone at 3 ppm upregulates nuclear AP-1, although this activity was not enhanced in lactating rats exposed to 1 ppm. The nuclear activity of AP-1 in naïve lactating rats and virgin females was compared to investigate the possibility that this transcription factor is involved in the mild idiopathic neutrophilic inflammation of lactating rats. The mean nuclear AP-1 signal was 76 percent higher in a group of lactating rats compared to virgin rats (n = 5/group). The difference between groups was not statistically significant, but the direction of the change did correlate with the difference in neutrophils recovered by BAL (12.5 percent ± 4.3 percent for the lactating group and 0.6 percent ± 0.6 percent for the virgin rats). Another comparison between lactating and virgin rat groups of the same size resulted in a nonsignificant 38-percent increase of nuclear AP-1 signal in the lactating rat group. The difference in lavaged neutrophils between the lactating and virgin female groups also was smaller in this experiment compared to the first (4.6 percent ± 1.2 for the lactating group and 2.0 ± 0.8 for the virgin females). These data are consistent with a role for AP-1 in mediating the mild pulmonary inflammation observed in lactating rats; however, the sensitivity and reproducibility of the gel shift assay for AP-1 is not good enough to establish a clear correlation of these two variables.

Conclusions:

Ozone exposure did not cause activation of the transcription factor, NF-κB, in Sprague Dawley rats. Thus, NF-κB does not mediate ozone-induced pulmonary inflammation, at least in this rat strain, as had been assumed. There was some evidence that the transcription factor AP-1 may be upregulated by ozone, indicating a possible role in signaling; however, upregulation was convincing only at very high ozone concentrations. There also was no difference in activation of either NF-κB or AP-1 between lactating and virgin female rats exposed to ozone as had been hypothesized. Furthermore, evidence of elevated levels of ROS in the lungs of lactating rats compared to virgin female rats was not detected using either 8-isoprostane or inhibition of XL-PCR amplification of mitochondrial DNA as indicators of ROS.

Supplemental Keywords:

ozone, reactive oxygen species, ROS, transcription factor, lactation, pregnancy, sensitive populations, dose response, pulmonary inflammation, neutrophil, mammalian, susceptibility, NF-κB, neutrophilic inflammation, glucocorticoids, antioxidandts, health effects, mechanism, lung inflammation, pregnant, lactating, Sprague Dawley,, Health, RFA, Scientific Discipline, Susceptibility/Sensitive Population/Genetic Susceptibility, Health Risk Assessment, Risk Assessments, genetic susceptability, Atmospheric Sciences, Biochemistry, Toxicology, environmental mutagens, neutrophilic inflammation, oxidant generation, air quality, health risks, sensitive populations, ozone induced inflammation, lactation, pregnancy, mothers, animal model, hypersensitivity, pregnant women, ozone, environmentally caused disease, toxics, biomarker, human health risk, isoprostane, highrisk groups, hypersensitive people, reactive oxygen species, metabolic activation, pulmonary toxicity

Progress and Final Reports:

Original Abstract

1998

1999