Final Report: Airborne Particulate Matter-Induced Lung Inflammation

EPA Grant Number: R826782
Title: Airborne Particulate Matter-Induced Lung Inflammation
Investigators: Morandi, Maria T. , Holian, Andrij , Parsley, Edwin
Institution: The University of Texas at Houston , University of Montana
EPA Project Officer: Chung, Serena
Project Period: October 1, 1998 through September 30, 2001 (Extended to September 30, 2002)
Project Amount: $674,288
RFA: Health Effects of Particulate Matter and Associated Air Pollutants (1998) RFA Text |  Recipients Lists
Research Category: Air Quality and Air Toxics , Air , Health Effects , Particulate Matter


The overall objective of this research project was to test the hypothesis that fine particulate matter (PM) induces apoptosis of the suppresser population of alveolar macrophage (AM), which contributes to activation of T helper cell cytokine cascades and development of lung inflammation. We further proposed that these effects would be more pronounced in individuals with chronic lung disease. The specific objectives of this research project were to: (1) characterize PM-induced apoptosis and phenotype shifts in human AM in vitro and AM apoptosis, and T helper cell activation in murine models in vivo; (2) characterize the influence of age in murine models on the bioactivity of PM; (3) characterize the effects of PM on AM apoptosis and phenotype shifts in human AM isolated from patients with chronic lung disease; and (4) characterize the bioactive chemical components of PM that affect apoptosis and phenotype shifts of human AM and T helper cell activation in murine models. Recent epidemiological studies have reported statistically significant associations between short-term increases in airborne respirable PM concentrations and increased mortality and morbidity from respiratory and cardiovascular disease. Although toxic effects of PM have been demonstrated with a variety of animal models and, to a more limited extent, with human subjects, the mechanism(s) that would explain the reported associations between exposure to PM and adverse health effects remains to be elucidated. We propose that one of the important targets of PM-induced inflammation is the AM.

Summary/Accomplishments (Outputs/Outcomes):

Studies With Animal Models

Surface Components of Airborne PM Induce Apoptosis Through Scavenger Receptors (SR). Epidemiology studies have linked mortality, increased asthma morbidity, and other respiratory disorders in urban areas to increases in fine airborne PM concentrations. However, neither the bioactive components of PM, nor the biological mechanisms of the reported health effects, have been elucidated. A number of studies have implicated soluble metals, the strong acid fraction, and/or other components of PM as possible bioactive mediators. AM apoptosis, mediated through SR, may be important in the response to inflammatory particles. Therefore, this study explored the hypothesis that organic and metallic components of PM induce apoptosis by interacting with SR. Freshly isolated AM from Balb/c mice were incubated with PM 1648 samples untreated or extracted with Milli-Q water, acetone or cyclohexane, acid digested, or heated at 100°C or 500°C. Cell viability was assessed by trypan blue exclusion, and apoptosis was demonstrated by examination of cell morphology and Cell Death ELISA. Untreated PM induced necrosis and apoptosis in AM. Treatment of PM by organic solvent extraction, acid digestion, or high heat modified the particle surface composition and decreased apoptosis. Apoptosis induced by untreated, acetone extracted, and high temperature-treated PM, was blocked by polyinosinic acid or 2F8 antibody. These results demonstrated that PM-induced apoptosis is mediated by Class A Type I/II SR. Altering the surface characteristics of PM interferes with recognition by SR, resulting in decreased apoptosis of AM. Therefore, altering the surface chemistry by removal of one or more PM components, such as the various treatments conducted in this study, are sufficient to alter PM bioactivity. These results also may help explain why PM from many different sources and airsheds, with known differences in chemical composition, are all bioactive because it is the overall PM matrix structure that is important, not just one specific chemical component.

Asthma Mouse Model (2 manuscripts in preparation). Epidemiological evidence suggests that the increased prevalence of asthma and other respiratory diseases may be linked to elevated levels of urban PM. To test this hypothesis in an animal model, we have developed a new model using DO11.10 mice (on a Balb/c background) that are transgenic for an ovalbumin (OVA) specific alpha-beta T cell receptor, eliminating the need for prior immunization to the antigen. Bronchoconstriction was determined using whole body plethysmography (Buxco) 72 hours after intranasal administration (IA) of PM, and 4 hours after IA of OVA. Methacholine (0-50 mg/ml) was aerosolized as a standard method of further enhancing bronchoconstriction, which was quantified by penh (enhanced pause) measurements taken at regular time intervals.

Using this model, we demonstrated that PM 1648 caused significant bronchoconstriction in a dose-dependent manner in the range from 125 µg to 750 µg. Acid digested PM 1648, described above, was essentially non-bronchoconstrictive. These results suggested that the bronchoconstriction was because of the character of the PM and because not all PM components are biologically active. The acid digested PM 1648 served as a particle control for treatment with untreated PM 1648. To determine whether endotoxin contamination contributed to the bronchoconstriction, PM 1648 was treated with polymyxin B prior to instillation. Removal of endotoxin had no effect on the bronchoconstrictive activity of PM 1648, indicating that endotoxin was not essential to the biological activity of PM 1648.

Because previous studies suggested that PM acts on AM through the scavenger receptor (SR), we tested whether the bronchoconstrictive effect of PM 1648 could be blocked in vivo using the 2F8 antibody. 2F8 caused a significant reduction in the exacerbation of bronchoconstriction induced by PM 1648, indicating a possible role for the macrophage scavenger receptor A in this process. These results are being investigated further.

Taken together, these studies demonstrated that PM 1648 caused reproducible bronchoconstriction in a murine model. Furthermore, this effect was dependent on particle characteristics and was independent of possible endotoxin contamination. Additional studies were conducted using fresh PM2.5 collected in Houston, TX. The fresh PM2.5 appeared to be much more bioactive than PM 1648 with regard to bronchoconstriction. PM2.5 at 250 µg approximately doubled the penh values produced by PM 1648 (500 µg). Additional studies are being conducted with PM2.5.

Alterations of AM Antigen Presentation Cell (APC) Activity. Previously, we had demonstrated that PM 1648 could increase the APC activity of human AM. However, additional studies were needed to determine whether APC activity was increased in vivo and whether any changes were consistent with the time course of the changes in bronchoconstriction. In these experiments, Balb/c mice were given differing amounts of PM 1648 (250, 500, and 750 µg) by intranasal instillation. In addition, acid-extracted PM 1648 was used as a control particle. The lung AMs were harvested at 4 hours, 3 days, and 7 days post-instillation, and cultured for 2 hours with OVA (1 mg/ml). These cells then were co-cultured with DO11.10 spleen-derived lymphocytes at a 1:4 ratio for 2 days. After 2 days, the culture media was retrieved and two cytokine levels were determined (interferon gamma [IFN,TH1 pathway] and interleukin-13 - [IL-13, TH2 pathway]). Interleukin-4 (IL-4) levels were not measured as they were determined to be below detection in this model. All PM 1648 particle loads produced similar effects with regard to APC activity, possibly because they were all in excess. Pre-exposure to PM 1648 caused significant increases in IFN levels short-term (4 hours group) and resolved in the 3 day and 7 day exposure groups. In contrast, IL-13 was enhanced significantly in the 3-day and 7-day groups, but not in the 4-hours group. The control particles produced no significant enhancements of APC activity at any time point. Taken together, these results suggest that AM exposure to PM 1648 is a potent inducer of TH2 response long-term, with the time of enhancement consistent with the time of the measured bronchoconstriction. Furthermore, the APC activity in these results clearly was TH2, consistent with the known TH2 pathway of human asthma and the previously published work with human AM described next.

Human Models

Silica and PM 1648 Modify Human Alveolar Macrophage (HAM) Antigen Presenting Cell Activity In Vitro. Some respired particles are known to lead to inflammation and lung pathology, while others do not appear to have long-term effects. Potential mechanisms to account for these differences are only now beginning to be understood. In this study, we examined whether silica and PM 1648 caused selective deletion of the suppressor human (HAM) phenotype (RFD 1+/7+), and whether this affected cytokine production in an APC assay with autologous T lymphocytes. HAM were exposed to the bioactive particulates, silica, and PM 1648 for 24 hours, then isolated free of particulates and non-viable cells; HAM then were cultured with autologous lymphocytes in an 11-day APC assay. Silica exposure up-regulated a TH1 lymphocyte-derived cytokine, IFN, and a TH2 lymphocyte-derived cytokine, IL-4. PM 1648 exposure primarily up-regulated IL-4. Neither particle exposure had a significant effect on interleukin-10 (IL-10) production. Control particulate exposures with titanium dioxide (TiO2), and wollastonite (Woll), produced no altered APC activity. Silica and PM 1648 demonstrated selective toxicity to suppressor macrophages (RFD 1+/7+), probably because of the involvement of surface proteins such as the class-A scavenger receptor (SR-A). We proposed that with the suppressor macrophage phenotype disabled, the activator macrophage (RFD1+/7-) operates free of the suppressor macrophage's influence, enhancing APC activity with increased lymphocyte-derived pro-inflammatory cytokine production.

A Comparison of Murine and HAM Responses to Urban Particle Matter. Rodents frequently are used as an animal model to help elucidate the mechanisms of toxicity that may provide clues for the understanding of PM toxicity in humans; however, the relationships between murine and human PM toxicity have not been established. PM is known to target the pulmonary epithelium and resident AM. PM can initiate cytotoxic effects on the AM, including apoptosis and necrosis depending on the particle concentration. This study examined AM apoptosis and necrosis initiated by PM in a human and murine (Balb/c) in vitro model. Consequently, a direct comparison of the effects of PM on human and murine AM cytotoxicity performed under similar conditions was examined. Freshly isolated AM from human volunteers was incubated with seven different residual fractions of PM 1648 derived from organic solvent extractions, high-temperature heating, and acid digestions to change the surface characteristics of the PM (described earlier). These results were compared to the murine results found in Obot, et al. The results obtained suggest that at the same concentration of PM, the pattern of toxicity and the effects in ex-vivo experiments with Balb/c and human AM have a similar pattern. Furthermore, altering the surface chemistry by removal of one or more PM components, such as the various treatments conducted in this study, was sufficient to alter PM bioactivity in both human and murine AM in a similar pattern. In addition, the human and murine models were compared with regard to in vitro cytotoxicity using freshly collected PM2.5. The cytotoxic PM2.5 effects were identical in both human and mouse models, with PM2.5 being significantly more toxic than PM 1648 based on mass. Taken together, the results demonstrate that the Balb/c mouse can be used as a suitable model for PM cytotoxicity in AM as it parallels the human AM responses very closely.

PM2.5 Alters Human Antigen Presentation (APC) Activity In Vitro. Additional studies were done to examine the effect of PM2.5 on human APC activity using the in vitro model described above in Hamilton, et al. Briefly, human AM were exposed to PM2.5 (50 or 100 µg/ml, collected in Houston, TX) for 24 hours in suspension culture. After 24 hours, viable cells were separated from the loose particles and dead cells by differential centrifugation. The viable AM then were co-cultured with autologous lymphocytes at a 1:4 ratio in an 11-day APC assay as described above. At the end of this culture, the medium was retrieved and assayed for T lymphocyte products IFN, IL-4, and IL-13. Pre-exposure to PM2.5 significantly altered APC activity producing significant increases in all cytokines measured. However, the largest increases occurred with IL-13 and IL-4, again indicating a PM-stimulated TH2 response. The higher dose of PM2.5 resulted in greater cytokine production. In this study, there clearly was a subset of non-responding subjects' cells (>50 percent). This may suggest that there is a differential susceptibility in the general population to the effect of PM on APC activity based on our hypothesis; a differential sensitivity to the effects of PM on airways. Lastly, a limited sample of PM2.5 was collected in rural setting (Missoula, MT) and tested using the same assay. The results indicated that the Missoula PM2.5 was not as bioactive as the PM2.5 collected in a much more industrial urban setting (Houston, TX). These results suggest that different sources are important in the magnitude of the resulting biological responses.

Characterization of Residual Oil Fly Ash (ROFA)-Induced Cytotoxicity in HAM. In this work, three forms of ROFA were characterized by examining the toxicity with human AM in vitro. All three forms produced significant decreases in AM viability and significant increases in apoptosis in a concentration-dependent manner. The water-extracted portion of ROFA, which contained toxic metals, was determined not to be the source of the cytotoxicity. In addition, the major toxic metal components of ROFA (vanadium [IV], vanadium [V], copper, and nickel) were examined individually for toxicity with human AM. Although all of the metals were determined to be toxic to AM, much higher concentrations were required than those possible based on the metal content from the ROFA particles. It was concluded that the particle-cell interaction (such as that for PM through SR) must be important for cytotoxicity to occur. To test this hypothesis, polyinosinic acid (100 µg/ml) was used to block the AM SR prior to ROFA exposure. Polyinosinic acid significantly attenuated the cytotoxicity of ROFA particles (both necrosis and apoptosis). It was concluded that particle surface characteristics, and not soluble metal content, were responsible for ROFA toxicity in AM.


Objective 1: To Characterize PM-Induced Apoptosis and Phenotype Shifts in Human AM In Vitro and AM Apoptosis and T Helper Cell Activation in Murine Models In Vivo. Significant work has been done in this area using PM 1648 and PM2.5. The studies conducted thus far in human and mouse models have demonstrated that both PMs are potent inducers of TH2 responses in APC stimulated lymphocytes. This is consistent with the observation that PM is involved in asthma exacerbation. PM exposure also may play a role in other TH2 mediated diseases. It is clear from the human study (Hamilton, et al., 2001) that suppressive AM phenotypes are selectively susceptible to PM 1648 toxicity. This may partially explain the APC results. Activator AM phenotypes may dominate the response to antigens and the interactions with TH lymphocytes, in the absence of sufficient numbers of suppressor AM, producing an overreaction to the APC activity. It remains to be determined, however, what regulates a TH1 verses a TH2 dominated response. Some particles such as silica and asbestos produce an augmented TH1 response in similar conditions. It is not clear how PM selectively activates or primes a TH2 lymphocytic response. More work is needed in this area.

Combined with an earlier published study (Holian, et al., 1998), these results provide insight into the relative human AM cytotoxicity by PM. Comparing PM on the basis of mass, the potential human AM cytotoxicity is ranked as follows: ROFA > PM2.5 > PM 1648 = PM 1649. The only PMs tested in a mouse model are PM2.5 and PM 1648, and it does appear that PM2.5 is more toxic than PM 1648 at equal mass in vitro.

Objective 2: To Characterize the Influence of Age in Murine Models on the Bioactivity of PM. Some work has been done on this specific objective but the availability of "old" mice is very limited. Therefore, the majority of this work was focused on "young" mice. Initial observations in the murine asthma model using 3-month old mice showed no differences compared to normal 7-week old mice in PM exacerbation of bronchoconstriction. In another experiment, 2-week old DO11.10 mice were instilled with 250 µg 1648 or saline. Again the penh values after 72 hours and OVA were not significantly different between the two groups. The cell differentials were 50 percent macrophages and 50 percent neutrophils. Two-week old DO11.10 mice were instilled as above, but not challenged until they were 4 weeks old. Bronchoconstriction at 4 weeks was significantly suppressed when instilled at 2 weeks compared to instillation at 4 weeks old. The cell differentials were closer to "normal" (90 percent macrophages and 10 percent neutrophils). More work needs to be done in this area to clarify the role of age-dependent susceptibility to PM exposure.

Objective 3: To Characterize the Effects of PM on AM Apoptosis and Phenotype Shifts in Human AM Isolated From Patients With Chronic Lung Disease. Due to the lack of samples from patients with chronic lung disease no work was done on this objective.

Objective 4: To Characterize the Bioactive Chemical Components of PM That Affect Apoptosis and Phenotype Shifts of Human AM and T Helper Cell Activation in Murine Models. Significant work has been done on this specific objective. PM 1648 was used instead of PM2.5 because of the large amounts of particle needed for the extraction processes. The amount of PM2.5 available is extremely limited at any given time. Obot, et al., 2002, described the cytotoxicity of PM 1648 and its altered-extracted forms in a murine in vitro model. Acid-extraction of PM removing labile chemical constituents appears to make the particle biologically inert, providing a good control particle for all PM studies. No extraction process of PM made the particle more toxic. It is clear that the surface components of PM mediate the AM toxicity. Changing these surface characteristics alters the AM's response to the particle. In addition, the SR-A appears to play an important role in AM toxicity with regard to PM and other toxic particles. The results of analogous experiments with human AM are consistent with those in the mouse model, suggesting that the PM toxicity in the murine model is indicative of that occurring in the human system, making the murine model appropriate for PM studies.

Journal Articles on this Report : 5 Displayed | Download in RIS Format

Other project views: All 8 publications 5 publications in selected types All 5 journal articles
Type Citation Project Document Sources
Journal Article Archer AJ, Cramton JLH, Pfau JC, Colasurdo G, Holian A. Airway responsiveness after acute exposure to urban particulate matter 1648 in a DO11.10 murine model. AJP Lung Cellular and Molecular Physiology 2004;286(2):L337-L343. R826782 (Final)
  • Abstract from PubMed
  • Full-text: AJP-Full Text HTML
  • Abstract: AJP-Abstract
  • Other: AJP-Full Text PDF
  • Journal Article Hamilton RF, Marshall GD, Holian A. Silica and PM1648 modify human alveolar macrophage antigen-presenting cell activity in vitro. Journal of Environmental Pathology, Toxicology and Oncology 2001;20(Suppl 1):75-84. R826782 (Final)
  • Abstract from PubMed
  • Abstract: Begell House-Abstract
  • Journal Article Hamilton Jr. RF, Holian A, Morandi MT. A comparison of asbestos and urban particulate matter in the in vitro modification of human alveolar macrophage antigen-presenting cell function. Experimental Lung Research 2004;30(2):147-162. R826782 (Final)
  • Abstract from PubMed
  • Abstract: Taylor&Francis-Abstract
  • Journal Article Obot CJ, Morandi MT, Beebe TP, Hamilton RF, Holian A. Surface components of airborne particulate matter induce macrophage apoptosis through scavenger receptors. Toxicology and Applied Pharmacology 2002;184(2):98-106. R826782 (Final)
  • Abstract from PubMed
  • Full-text: ScienceDirect-Full Text PDF
  • Abstract: ScienceDirect-Abstract
  • Journal Article Obot CJ, Morandi MT, Hamilton RF, Holian A. A comparison of murine and human alveolar macrophage responses to urban particulate matter. Inhalation Toxicology 2004;16(2):69-76. R826782 (Final)
  • Abstract from PubMed
  • Abstract: Taylor&Francis-Abstract
  • Supplemental Keywords:

    asthma, reactive airways, animal models., RFA, Health, Scientific Discipline, Air, Geographic Area, particulate matter, Toxicology, Health Risk Assessment, State, Risk Assessments, Susceptibility/Sensitive Population/Genetic Susceptibility, indoor air, genetic susceptability, Atmospheric Sciences, Biology, Immunology, EPA Region, sensitive populations, particulates, health risks, human health effects, inhalability, morbidity, Texas, airway disease, cardiovascular vulnerability, exposure, pulmonary disease, air pollution, pre-existing conditions, human exposure, cardiopulmonary response, mortality studies, lung inflammation, Region 6, coronary artery disease, Acute health effects, indoor air quality, mortality, environmental hazard exposures, toxics

    Progress and Final Reports:

    Original Abstract
  • 1999
  • 2000
  • 2001