2010 Progress Report: Endothelial Cell Responses to PM—In Vitro and In Vivo

EPA Grant Number: R832414C002
Subproject: this is subproject number 002 , established and managed by the Center Director under grant R832414
(EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).

Center: San Joaquin Valley Aerosol Health Effects Research Center (SAHERC)
Center Director: Wexler, Anthony S.
Title: Endothelial Cell Responses to PM—In Vitro and In Vivo
Investigators: Wilson, Dennis , Anastasio, Cort , Barakat, Abdul , Rutledge, John , Tablin, Fern
Institution: University of California - Davis
EPA Project Officer: Hunt, Sherri
Project Period: October 1, 2005 through September 30, 2010 (Extended to September 30, 2011)
Project Period Covered by this Report: July 1, 2009 through June 30,2010
RFA: Particulate Matter Research Centers (2004) RFA Text |  Recipients Lists
Research Category: Health Effects , Air


The overall goal of this project is to determine the relationship between vascular disease and systemic effects of particulate matter.

Progress Summary:

In vitro responses to collected ambient PM

The goals of this work are to characterize cellular responses that are specific to ambient particulate matter and determine whether differences in either the extent or nature of these responses occurs in PM collected from different sites or seasons. We compared the response of PM collected from urban and rural sites in winter and summer with responses from similarly sized laboratory-generated PM of various compositions.

We also have measured the ability of the particles to generate reactive oxygen species (ROS), specifically HOOH in a PBS solution containing 50 "μM ascorbate. Our results show that: (1) Fresno PM is much more reactive than Westside particles, (2) PM2.5 produces much more HOOH than coarse particles, and (3) that soluble copper can explain most of the PM-mediated generation of HOOH.

Pulmonary Microvascular Endothelium

Because of our ongoing interest in the interface between the pulmonary microvasculature and systemic inflammatory responses, we have focused on cultured human pulmonary microvascular endothelial cells (HPMVEC) as in vitro targets for toxicity. Some of this work has been reported in previous progress reports. In summary, our previous work compared gene responses of HPMVEC to PM collected during summer and winter from our urban Fresno sampling site using microarray-based screening. We found summer responses to be significantly more pro-inflammatory than winter source samples. In contrast, we found consistent upregulation of the PAH-metabolizing enzyme CYP 1A1 in both seasons with a more robust response in winter than in summer. CYP 1A1 induction is regulated by the AhR receptor pathway. We have verified that this pathway is stimulated by collected PM and demonstrated that AhR driven promoter activity is unique to PM containing PAH compounds.

In the current year's work, we have extended these initial findings to compare them with rural source PM. We also used laboratory generated carbon particles containing PAHs (soot) for comparison. We find the trends for summer pro-inflammatory responses and winter PAH responses to be consistent between sites (figure 1 A+B). Soot treatment also upregulated AhR responses but was less potent in induction of inflammatory responses. In addition to PAH related metabolism, ambient PM upregulated prostagladin S2 synthase (COX-2) and the antioxidant response enzyme aldehyde dehydrogenase.

Figure 1A: Induction of CYP1A1, ALDH1A3, PTGS2 and TIPARP gene expression by urban and rural APM from different seasons and soot. Urban and rural APM were compared to H2O (control) and Soot was compared to DMSO (control). N = 4, # = P ≤0.0005, * = P ≤0.005, $ = P ≤0.05.

Figure 1B: Induction of E-Selectin, ATF3, ICAM-1, CCL2, CXCL2 (MIP-2α) and IL-6 by urban and rural APM for different season and Soot. Urban and rural APM were compared to H2O (control) and Soot was compared to DMSO (control). N = 4, # = P ≤0.0005, * = P ≤0.005, $ = P ≤0.05.

To confirm AhR selectivity and promotional activation, we used a luciferase-based assay in a hepatocyte tumor line construct (Figure 2A). Relative to the inactivity we previously reported for similar concentrations of silica, soot was more active in stimulating AhR responses than ambient PM from all sources. This finding is somewhat tempered by the compositional differences in that 100% of the soot was carbon particles containing PAHs, while ambient source material contains a large proportion of salts that likely are soluble under in vitro conditions. To determine whether APM or soot preparations were likely to induce apoptosis, we evaluated caspase 3 activation as a marker of apoptosis (Figure 2B). APM was significantly more inductive of apoptosis than soot.

Figure 2A: Induction of luciferase in H1L1.1c2 cells containing a stably integrated AhR responsive DRE-driven firefly luciferase reporter gene plasmid. Response to a 3 hour incubation of 10 µg/mL APM collected in Summer 2006 or Winter 2007 or Summer 2008 or Winter 2008 or Soot was compared to TCDD (100% as positive control). N=3.

Figure 2B: APM-induced caspase3/7 activity was increased significantly by APM collected in Summer 2008. Urban and rural APM were compared to H2O (control) and Soot was compared to DMSO (control). N = 5, * = P ≤0.05


Investigation of endothelial responses in the current year also evaluated whether PAH, metal, ROS generation or endotoxin components of PM drives the pattern of gene responses we have described. Based on experiments with monocytes, described below, we used inhibitors of AhR activation (a-napthoflavone) endotoxin (polymyxin B) and ROS generation (n-acetyl cysteine and resveratrol) to determine whether gene responses in endothelial cells were selectively attenuated by any of these treatments. As expected aNF treatment inhibited AhR mediated CYP 1A1 upregulation. This upregulation was also inhibited by antioxidant treatment with resveratrol but not nAC. In general antioxidant treatment effectively reduced all inflammatory responses. Binding endotoxin with polymyxin B prevented cytokine synthesis but not transcription of adhesion molecules (Figure 3 A+B).

Figure 3 Effect of AhR inhibition (a-napthoflavone aNF) endotoxin binding (polymyxin B PolB) or antioxidant treatment (n-acetyl cysteine and resveratrol) on gene responses (RT-PCR) to collected urban summer source ambient particulate matter.

Bronchial Epithelium: Compared with endothelial cell responses, cultured human bronchial epithelium has somewhat lesser responses to collected ambient PM. Our studies reported in previous progress reports suggest a consistent PAH response as demonstrated by upregulation of CYP 1A1 and 1B1. As in endothelial cell responses, this was more robust in winter source PM than in summer. Also similar is the pro-inflammatory response, in this case represented by the dendritic cell chemotractant OCL-20.

Our ongoing work evaluates the contribution of bronchial epithelium to pro-inflammatory responses using ELIZA and multiplex based assays for protein. We examined both IL1b and CCL-20, important components of innate immune system responses. We find little treatement associated differences in protein expression in the supernatants of HBE cells treated with urban summer source PM (Figure 4).

Figure 4: Protein levels of IL1b and CCL 20 in supernatants of cultured human bronchial epithelial cells treated with urban summer source ambient PM.


We have used freshly isolated human blood monocytes as well as monocyte cell lines (THP-1 cells) to evaluate the potential that PM crossing the alveolar barrier might interact either directly with circulating inflammatory cells or alternatively through the interface with alveolar endothelium activated endothelium. We previously reported that PM treated monocytes upregulate a variety of pro-inflammatory genes. We also demonstrated that PM treated monocytes increase their adhesion to cultured aortic endothelium. In more recent experiments, we asked whether pulmonary microvascular endothelium secretes activities that in turn stimulate monocyte responses. We find that monocytes treated with supernatants of PM treated HPMVEC’s have a similar pro-inflammatory gene response to that resulting from direct PM treatment (Figure 5)

Figure 5: Pro-inflammatory gene activation in monocytes incubated with supernatants from HPMVEC treated with rural winter source ambient PM.

We further asked whether specific pro-inflammatory mediators could be demonstrated in EC supernates using a multiplex protein assay system. Of 32 activities evaluated only IL6, IL8 and MCP-1 were present in greater than background expression levels (Figure 6)

Figure 6: Positive results of multiplex (Bioplex) assays for cytokines in HPMVEC treated with rural winter source ambient PM.

Results of our monocyte experiments suggested that these results could be used to compare location and season source responses. We asked whether differences in monocyte responses occurred between our rural and urban source PM. We found that rural source PM had a greater inflammatory response than urban but that there was no difference in PAH related responses (Figure 7).

Figure 7: Comparative gene responses to urban (F) or rural (W) winter source ambient PM treatment of monocytes in vitro.

We next asked whether differing components of PM selectively stimulated specific responses using inhibitors of transition metals (DFM chelation) AhR activation (a-napthoflavone), endotoxin (polymixin B) and intracellular and extracellular ROS generation (n-acetyl cysteine and catalase respectively). As might be predicted, metal chelation and intercellular ROS quenching inhibited inflammatory gene responses but did not alter PAH related responses. Endotoxin binding by polymixin B significantly inhibited pro-inflammatory responses. Using these inhibitors, we compared winter rural and urban source PM. We found that endotoxin binding was more effective in blunting responses to rural source PM while metal chelation uniquely inhibited urban source PM pro-inflammatory genes (Figure 8 A+B).

Figure 8: Percent inhibition of pro-inflammatory gene upregulation by A: endotoxin binding (poymixin B treatment) or B: metal chelation (DFM). Monocytes treated with urban source (F) or rural source (W) PM were compared in the presence or absence of indicated inhibitors.

We extended these studies to show that polymyxin B treatment abrogates the PM induced adhesion of monocytes to endothelium. This suggests that LPS is the key component driving pro-inflammatory responses. Conversely, polymyxin B treatment has no effect on CYP 1A1 transcription suggesting that the PAH response pathway is separate from the pro-inflammatory responses (Figure 9).

Figure 9: Comparison of polymyxin B inhibition of endotoxin on monocyte adhesion to endothelium or induction of CYP 1A1 gene transcription in response to rural winter source PM.

Finally, we asked whether ROS generation occurs in PM treated monocytes and whether this is an extracellular or intracellular phenomenon. We demonstrated that PM treated monocytes have increased ROS in a flow cytometric assay (Figure 10) Furthermore, we find that inhibition of extracellular ROS by pretreatment with catalase does not influence PM induced pro-inflammatory responses but that inhibition of intracellular ROS by NAC greatly blunts inflammatory gene responses (Figure 11).

Figure 10: Flow cytometric analysis of intracellular ROS in monocytes treated with winter rural source PM.

Monocyte ROS.tif

Figure 11: Comparison of inflammatory gene responses in monocytes treated with ambient PM in the presence of catalase or n-acetyl cysteine (NAC).

In vivo responses to CAPs and collected ambient PM Intra-tracheal instillations:

With additional support from the California Air Resources Board, we have been translating our in vitro findings to in vivo experiments with intra-tracheal instillations of collected ambient PM. Our previous CAPs studies demonstrated upregulation of systemic cytokines and evidence of platelet priming in mice exposed for 2 weeks. We have repeated these studies with 3 more exposures from differing seasons and locations. We find inconsistent changes in circulating cytokine levels in response to CAPs but consistent evidence of platelet priming in all CAPs exposed animals. We also find subjective evidence that the platelet responses in vivo correlates with the diversity and extent of changes in our in vitro assays.

A key new approach in our most recent CAPs studies has been the use of Laser Capture Microdissection to compare gene responses in differing regions of lung. While our experiments with whole lung homogenates generally do not demonstrate any CAPs associated gene activation in microarray studies, we do find gene responses in specific lung locations that correlate with our in vitro gene screening assays. Specifically we find upregulation of CYP 1A1 as a signature response in several locations including bronchial epithelium (Figure 12). Analysis of alveolar parenchyma did not demonstrate specific gene responses. Dissected pulmonary vessels, however, had upregulation of CYP 1A1 as well as expression of NOX-2, an endogenous ROS generating enzyme, and the adhesion molecule ICAM-1. The findings in pulmonary vessels add credence to our evolving hypothesis that the pulmonary vasculature may be the interface between pulmonary and systemic inflammatory responses.

Figure 12: Gene responses in airways or vessels of mice exposed to CAPs for 2 weeks: C: Gene responses in dissected bronchial epithelium D: Gene responses in dissected pulmonary vessels

To more directly control for dose and time course of changes, we have performed intra-tracheal instillations of collected ambient PM. While results of these experiments are only partially complete, several key findings have emerged. Firstly, intra-tracheal (IT) instillation reliably replicates activation of circulating platelets as we have previously described in CAPs studies. Secondly, IT experiments elicit regional pulmonary inflammatory responses that center on small pulmonary arterioles.

ROS generation:

We have also quantified the generation of ROS by particles collected at the Fresno and Westside sites during the Project 3 animal exposures from 2006 to 2009. We have finished with our HOOH measurements and are now working on OH measurements. We have found that Fresno PM generally generates much more HOOH than Westside PM and that, normalized by air volume, PM2.5 generally makes more HOOH than the corresponding coarse PM (PMcf, i.e., 2.5 to10 µm) (Figure 13). PM2.5 dominates primarily because the mass concentration of PM2.5 is much higher than that of PMcf in our aerosol samples. In nearly all of the samples the addition of a physiologically relevant concentration of ascorbate greatly enhances HOOH formation, by a factor of approximately 100. The amount of HOOH produced by SJV PM is reduced on average by (78 ± 15) % when the transition metal chelator desferoxamine (DSF) is added to the extraction solution, indicating that transition metals play a dominant role in HOOH generation. More specifically, we found through other experiments that PBS-soluble copper is primarily responsible for HOOH production by the Fresno PM. Extrapolating our results to expected concentrations of PM-derived HOOH in human lungs suggests that typical daily PM exposures in the San Joaquin Valley are unlikely to cause HOOH-mediated acute health effects, but that very high PM events might lead to cytotoxic levels of pulmonary HOOH.

Figure 13. Air-volume-normalized initial rates of HOOH generation in the presence of 50 µM ascorbate in PBS. Key to sample names: SU = summer, WI = winter, 0x = year (200x). Values are means ± SD, n = 3. Letters above bars indicate statistically different rates: a > b > c for fine PM, while a' > b' > c' > d' for coarse PM. The asterisk for the Westside Winter 2008 sample indicates that HOOH formation was not statistically different from zero.

Future Activities:

Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other subproject views: All 15 publications 9 publications in selected types All 9 journal articles
Other center views: All 111 publications 54 publications in selected types All 48 journal articles
Type Citation Sub Project Document Sources
Journal Article den Hartigh LJ, Lame MW, Ham W, Kleeman MJ, Tablin F, Wilson DW. Endotoxin and polycyclic aromatic hydrocarbons in ambient fine particulate matter from Fresno, California initiate human monocyte inflammatory responses mediated by reactive oxygen species.Toxicology In Vitro 2010;24(7):1993-2002. R832414 (2010)
R832414 (Final)
R832414C002 (2010)
R832414C002 (Final)
  • Abstract from PubMed
  • Full-text: ScienceDirect - Full Text HTML
  • Abstract: ScienceDirect - Abstract
  • Other: ScienceDirect - Full Text PDF
  • Journal Article Nakayama Wong LS, Lame MW, Jones AD, Wilson DW. Differential cellular responses to protein adducts of naphthoquinone and monocrotaline pyrrole. Chemical Research in Toxicology 2010;23(9):1504-1513. R832414 (2010)
    R832414 (Final)
    R832414C002 (2010)
    R832414C002 (Final)
  • Abstract from PubMed
  • Abstract: ACS - Abstract
  • Journal Article Shen H, Barakat AI, Anastasio C. Generation of hydrogen peroxide from San Joaquin Valley particles in a cell-free solution. Atmospheric Chemistry and Physics 2011;11(2):753-765. R832414 (2010)
    R832414 (Final)
    R832414C002 (2010)
    R832414C002 (Final)
  • Full-text: ACP-Full Text PDF
  • Abstract: ACP-Abstract
  • Supplemental Keywords:

    , RFA, Health, Scientific Discipline, Air, particulate matter, Health Risk Assessment, Environmental Chemistry, Epidemiology, Risk Assessments, ambient aerosol, lung injury, acute cardiovascular effects, long term exposure, toxicology, lung disease, air toxics, airway disease, airborne particulate matter, ambient particle health effects, human exposure, inhalation toxicology, epidemiological studies, concentrated air particles, vascular dysfunction, microarray studies, PM, cardiovascular disease

    Relevant Websites:

    Progress and Final Reports:

    Original Abstract
  • 2006 Progress Report
  • 2007 Progress Report
  • 2008 Progress Report
  • 2009 Progress Report
  • Final Report

  • Main Center Abstract and Reports:

    R832414    San Joaquin Valley Aerosol Health Effects Research Center (SAHERC)

    Subprojects under this Center: (EPA does not fund or establish subprojects; EPA awards and manages the overall grant for this center).
    R832414C001 Project 1 -- Pulmonary Metabolic Response
    R832414C002 Endothelial Cell Responses to PM—In Vitro and In Vivo
    R832414C003 Project 3 -- Inhalation Exposure Assessment of San Joaquin Valley Aerosol
    R832414C004 Project 4 -- Transport and Fate Particles
    R832414C005 Project 5 -- Architecture Development and Particle Deposition