gene and protein responses of human lung tissue explants exposed to ambient particulate matter of...

966

Introduction

Robust epidemiological evidence indicates that exposure to ambient particulate air pollution is associated with increased cardiovascular and respiratory morbidity and mortality (Gauderman et al., 2004; Laden et al., 2006; Miller et al., 2007). Associations have been observed

between mortality rates and traffic related air pollution (Beelen et al., 2008; HEI, 2010) and positive associations have also been reported for living near a major road and acute myocardial infarction (Tonne et al., 2007), coro-nary atherosclerosis and higher prevalence of coronary heart disease (Hoffmann et al., 2006).

RESEARCH ARTICLE

Gene and protein responses of human lung tissue explants exposed to ambient particulate matter of different sizes

Bastiaan Hoogendoorn1, Kelly Berube2, Clive Gregory1, Tim Jones3, Keith Sexton2, Paul Brennan4, Ian A. Brewis5, Alexander Murison6, Robert Arthur1, Heather Price3, Huw Morgan7, and Ian P. Matthews1

1Department of Primary Care and Public Health, Neuadd Meirionnydd, School of Medicine, Heath Park, Cardiff, UK, 2Lung and Particle Research Group, School of Biosciences, Cardiff University, Museum Avenue, Cardiff, UK, 3School of Earth and Ocean Sciences, Cardiff University, Cardiff, UK, 4Institute of Cancer and Genetics, School of Medicine, Heath Park, Cardiff University, Cardiff, UK, 5Central Biotechnology Services, School of Medicine, Heath Park, Cardiff University, Cardiff, UK, 6School of Biosciences, Cardiff University, Museum Avenue, Cardiff, UK, and 7Pollution Control Division, Environment Department, City and County of Swansea, The Guildhall, Swansea, UK

AbstractContext: Exposure to ambient particulate air pollution is associated with increased cardiovascular and respiratory morbidity and mortality. It is necessary to understand causal pathways driving the observed health effects, particularly if they are differentially associated with particle size. Objectives: To investigate the effect of different size ranges of ambient particulate matter (PM) on gene and protein expression in an in vitro model. Materials and methods: Normal human tracheobronchial epithelium (NHTBE) three-dimensional cell constructs were exposed for 24 h to washed ambient PM of different sizes (size 1: 7–615 nm; size 2: 616 nm–2.39 µm; size 3: 2.4–10 µm) collected from a residential street. A human stress and toxicity PCR array was used to investigate gene expression and iTRAQ was used to perform quantitative proteomics. Results: Eighteen different genes of the 84 on the PCR array were significantly dysregulated. Treatment with size 2 PM resulted in the greatest number of genes with altered expression, followed by size 1 and lastly size 3. ITRAQ identified 317 proteins, revealing 20 that were differentially expressed. Enrichment for gene ontology classification revealed potential changes to various pathways. Discussion and conclusions: Different size fractions of ambient PM are associated with dysregulatory effects on the cellular proteome and on stress and toxicity genes of NHTBE cells. This approach not only provides an investigative tool to identify possible causal pathways but also permits the relationship between particle size and responses to be explored.Keywords: Particulate matter, ultrafine particles, gene expression, iTRAQ proteomics, gene pathway

Address for Correspondence: Dr Bastiaan Hoogendoorn, Centre for Health and Environmental Research, Department of Primary Care and Public Health, Cardiff University, 4th Floor, Neuadd Meirionnydd, Heath Park, Cardiff CF14 4YS, UK. Tel.: +44 (0)29 2068 7151; Fax: +44 (0)29 20687236. E-mail: [email protected]

(Received 18 May 2012; revised 16 October 2012; accepted 18 October 2012)

Inhalation Toxicology, 2012; 24(14): 966–975© 2012 Informa Healthcare USA, Inc.ISSN 0895-8378 print/ISSN 1091-7691 onlineDOI: 10.3109/08958378.2012.742600

Inhalation Toxicology

24

14

966

975

18May2012

16October2012

18October2012

0895-8378

1091-7691

© 2012 Informa Healthcare USA, Inc.

10.3109/08958378.2012.742600

2012

Gene and protein responses of human lung tissue

B. Hoogendoorn et al.

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.

Gene and protein responses of human lung tissue 967


The majority of epidemiological studies use PM10

, which is a good predictor of the mass dose to the lung but a poor predictor of surface area and particle number (Harrison et al., 2010) as there is no general correlation between PM mass and particle number (Putaud et al., 2010). Estimates of motor-vehicle contribution to PM

2.5 (aerodynamic diameter <2.5 µm) range from 6%

in Beijing, 5% in Pittsburgh to 49% in Phoenix, 53% in Barcelona, and 55% in Los Angles. Particles with an aerodynamic diameter of <100 nm (Ultra-Fine Particles (UFP)) constitute up to 90% of total particle number but only 10% of total mass of PM

2.5 (EPAQS), 1998). In urban

areas, ambient UFP are mainly combustion derived, from transport and industrial sources. Toxicological response varies by size and composition of particles (Araujo & Nel, 2009; Kreyling et al., 2006). However, if UFP cause adverse health effects independently from coarser fractions then Health Impact Assessment undertaken in support of policy options to reduce the disease burden of PM would need to consider these.

An expert group (Knol et al., 2009) reported that, for cardiovascular effects of UFP, the pathway given the highest likelihood rating was that of airway inflammation leading to plaque rupture by one or more intermediate steps, among them endothelial dysfunction, leukocyte and platelet activation, acute phase response and vaso-dilatation/vasoconstriction. Pro-oxidative and/or proin-flammatory mediators released from the lungs into the systemic circulation subsequent to inhalation of PM are likely to mediate cardiovascular responses (Brook, 2008; Simkhovich et al., 2008).

Combustion-derived ambient UFPs have a large sur-face area and usually contain organic chemicals such as polycyclic aromatic hydrocarbons (PAHs) and quinones as well as transition metals (Donaldson et al., 2005). These constituents have been shown to have a role in proinflammatory effects in animal and in vitro models and a common mechanism linking these factors is their ability to generate oxidative stress in lung cells. The mea-surement of oxidative stress potential has been suggested as a promising way forward to evaluate toxicity of UFP (Ayres et al., 2008; Borm et al., 2007). Cells under oxida-tive stress will have tiered (i.e. dose related) responses including cell defense, proinflammation, and mitochon-dria-mediated apoptosis and necrosis (Li et al., 2008). Lower tiered protective responses may be non-harmful but it is possible to explore the level at which responses become potentially harmful by measuring protein dys-regulation consequent to exposure (Xiao et al., 2003). Evidence relating pathophysiological mechanisms to the effect of particle size alone may help to explain adverse health outcomes and provide clues as to which particle sizes in the ambient aerosol are more important.

We have performed an exploratory study into the putative dysregulation of genes and proteins in normal human-derived tracheobronchial epithelial (NHTBE) three dimensional cell constructs exposed to washed and dialyzed PM of different sizes collected from a residential

urban street. In order to address the dynamic and com-plex changes that occur during particulate insult of the tissue culture in a more open and non-hypothesis driven fashion, we employed a quantitative proteomic approach to profile protein changes after 24 h. This approach has become the most popular method of choice in quantita-tive proteomics in recent years (Lau & Chiu, 2009; Zhang et al., 2009).

Methods

Environmental samples and preparation of dosing suspensionsEnvironmental samples were collected over a noncon-secutive 10-week period on a residential street in Swansea using a sampling station run by the City Council. Average daily traffic was 18,000 vehicles/day. The particle col-lection and characterisation has been described in our previous publication (Price et al., 2010). To summarize, for collection we used a DekatiTM ELPI. This equipment divides particles into 12 size fractions, from 7 nm to 10 µm. ELPI cutoff diameters (Keskinen et al., 1992) and particle concentration profiles (Zervas & Dorlhène, 2006) have been confirmed in previous studies. The environmental samples collected at the 12 different size levels were com-bined to yield samples for analysis in the three size ranges (size 1: 7–615 nm; size 2: 616 nm–2.39 µm; size 3: 2.4–10 µm). The samples were then washed by overnight dialysis using a 6–8 KDa Spectrapor membrane in distilled water and the samples were subsequently freeze dried.

ICP-MS analysisSamples were digested for ICP-MS analysis using a CEM MDS-200 microwave system. Particle samples were washed into Teflon-coated composite vessels using 5 ml of 70% nitric acid. The samples were digested using an existing program developed for refractory carbon-based PM (Jones et al., 2006). The microwave program consists of a stepped increase in pressure to 80 psi for a period of 20 min, with a corresponding temperature rise to 180°C. The program lasts for ~2.5 h, including warm-up and cool-down periods. Samples were then diluted to a level of 10 μg/ml (dependent upon their original weight) using deionized (>18 ΩM) H2O. Raw data was corrected for blanks and controls accordingly.

Cells, culture, and exposure conditionsNHTBE cell constructs (EpiAirwayTM tissues, AIR-100), media and sterile phosphate-buffered saline (PBS) (with magnesium and calcium) were supplied by the MatTek Company (Ashland, MA, USA). ViaLight™ Cell Proliferation and Cytotoxicity Bio Assay Kit (Lonza Group Ltd., Basel, Switzerland) were used for study-ing ATP levels. All other chemicals were obtained from Sigma Chemical Co. (St Louis, MO, USA), unless other-wise stated.

The EpiAirway-100 system is a culture of NHTBE cells that forms a pseudo-stratified, highly differentiated

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.

968 B. Hoogendoorn et al.


model closely resembling that of the human tracheo-bronchial region of the respiratory tract (Balharry et al., 2008; Sexton et al., 2008). Cell culture and initial trans epithelial electric resistance (TEER) measurements were performed as described in Balharry et al. (2008).

Dosing solutions of the particles were prepared in ster-ile PBS (+Mg and Ca) supplied by MatTek and warmed to 37°C before dosing. Each insert was apically dosed with 100 µl of warm environmental particle (EP) solution at a concentration of 500 µg/ml. Each dose was replicated (n = 5). TEER and ATP assays were subsequently carried out 24 h after initial dosing. After 24 h the EP dose was removed from the apical surface and the tissue was stored in RNA later (Qiagen, Crawley, UK) at –20°C for genomic studies and simply at –80°C for proteomic analysis.

TEER measurementsTEER measurements were performed as described in Balharry et al. (2008).

ATP measurementsPost incubation, the cell inserts were lysed by addition of 200 µl of Cell Lysis Reagent (Lonza) for 10 min at room temperature. 100 µl of ATP monitoring solution was added to each insert sample and incubated at room temperature for a further 2 min before reading at 590 nm using a BMJ Labtech microplate reader.

Genomic analysesFor genomic analyses the samples (n = 3) were thawed, cells removed from the insert and extracted using QIAZOL (Qiagen). To assess RNA quality and concen-tration, each sample was analyzed using a NANODROP ND-1000 spectrophotometer, samples (S) were used with a 260/280 nm ratio of 1.6 < S < 2.0.

Reverse-transcription PCR and real-time PCR were carried out according to manufacturer’s instructions using SABiosciences first-strand cDNA synthesis fol-lowed by the Human Stress and Toxicity PathwayFinder™ RT² Profiler™ PCR Array (PAHS-003). This array profiles the expression of 84 genes related to stress and toxicity, including genes representative of pathways activated by prolonged stress, ranging from apoptosis/necrosis to growth arrest and senescence to proliferation and car-cinogenesis. Real-time PCR was carried out using an MJ Research Opticon (Biorad, Hercules, CA, USA).

Genomic data analysisRelative gene expression was determined using the integrated RT2 ProfilerTM PCR Array Data Analysis web-based software package according to the comparative Ct method (http://www.sabiosciences.com/rt_pcr_prod-uct/HTML/PAHS-003A.html). Gene transcription levels were considered statistically significant if they had a fold-change >1.5 or <0.66 with a p value < 0.05.

In addition, we tested for significant pairwise differ-ences in fold changes between the particle sizes using a permutation test. Specifically, 10,000 times we randomly

permuted the pair of fold changes associated with each of the 84 genes of interest and the two particle sizes being compared. For each permuted set of data, we calculated the absolute difference between particle size groups in the number of genes whose fold-change was >1.5 or <1/1.5. Under the null hypothesis of no difference, such permutations are inconsequential, so the p value may be computed by determining the fraction of the permuted sets of data whose test statistic (absolute difference in the number of fold change exceedances) equaled or exceeded the value seen in the observed data.

iTRAQ proteomic analysisProtein digestion, peptide labeling, LC separation of iTRAQ reagent-labeled peptides, mass spectrometry, and protein identification were performed using meth-ods previously described (Brennan et al., 2009).

Proteomic data analysisProteins were identified and relative quantification performed using ProteinPilotTM Software (version 2.0.1; Applied Biosystems, MDS Sciex) with the Paragon AlgorithmTM iTRAQ 8plex sample type (Peptide labeled) (Shilov et al., 2007). Data from duplicate LC-MALDI runs were combined for the analysis. The Swiss Prot Database (version Sprot_55.6_20080705.fasta) was searched using Homo sapiens as a species filter. Analysis parameters allowed for cysteine alkylation by methyl methane-thiosulfate. The detected protein threshold (Unused ProtScore (Conf)) was set at 2.00 to attain 99% confidence, bias correction was engaged and the identi-fied proteins were further processed by the ProGroupTM Algorithm to determine the minimal set of justifiable identified proteins. Following this, again using the ProGroup Algorithm within ProteinPilot, protein-based ratios of relative abundance were reported. Average iTRAQ ratios were calculated for each identified protein and p value and error factor (EF) estimates were pro-duced. Ordinarily, within the ProteinPilot analysis of a single 8-plex experiment, those identified proteins with relative quantitation p values <0.05 and EF <2 are con-sidered statistically significant in their fold change. For this research, four biological replicates with a control and three treatments were used. Two separate lots of eight samples were used (see Table 1) for iTRAQ labeling and LC-MALDI analysis. Protein Pilot software was used for the protein identification and quantitation. The output from Protein Pilot was exported into Excel and the two separately labeled lots were combined to generate an average of the four biological replicates. Calculating the standardized departure from the target (SD of all fold changes in treatment) in units of SD allowed the calcula-tion of lower and upper 95% confidence intervals for each fold change in each treatment. This allowed for any varia-tion between the independent experiments, giving a final fold change for each protein accompanied by 95% con-fidence intervals. Proteins were considered significantly altered if the 95% confidence interval did not include 1.

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



Identification of potential interacting proteinsWe used bioinformatics approaches to gather and collate data on the various proteins in the data set. A script was written in Perl to collect data from the Universal Protein Resource (UniProt) and these were stored in a MySQL relational database, allowing data linking and query generation output to an Excel compatible format. The approach we used was based around the methodologies used to predict functional categories from sets of genes derived from microarray data (Richards et al., 2008). Ontological analysis was performed (Khatri & Draghici, 2005) to see if a given set of genes is significantly enriched for a given gene ontology classification (Ashburner et al., 2000), which describes gene products in terms of their associated biological processes, cellular compo-nents and molecular functions. Genes encoding sets of proteins which either interact with a given dysregulated protein or are up to three interaction steps away from proteins already in this set were extracted from Uniprot using our Perl program and examined for enrichment for a gene ontology classification using DAVID. This list of proteins is defined as the interaction set. The UniProt manual curation process comprises manual review of results from a range of sequence analysis programs and literature curation of experimental data as well as attri-bution of all information to its original source. The full text of each paper identified is read and information is extracted and added to the entry using the UniProt cura-tion editor. Curators also assign GO terms to all manually curated entries (www.uniprot.org/program) (Magrane & Consortium, 2011). EASE/DAVID (Huang da et al., 2009) will systematically map the genes in this list to their asso-ciated biological annotation (e.g., gene ontology terms), and then statistically highlight the most overrepresented (enriched) biological annotation out of thousands of linked terms and contents in KEGG pathway to identify biological processes most applicable to the biological occurrence under study. It uses Fisher’s exact test to

calculate p values for over-representation and also uses a “relative enrichment” statistic: the ratio of enriched to nonenriched genes in the sample set, divided by the ratio of enriched to nonenriched genes in the reference set. Multiple testing is addressed by the use of the Benjamini correction (Benjamini & Hochberg, 1995) (also known as false discovery rate).

Results

Tissue culture TEER and ATP measurementsApical treatment with the different size fractions of col-lected EPs caused a non-significant decrease in TEER and ATP values (see Figure 1A and B, respectively).

Genomic analysesAn overview of the Human Stress and Toxicity PCR Array results may be found in Table 2. Genes in this array are arranged according to functional gene group with sta-tistically significant dysregulated genes shown in bold italic. Using the integrated RT2 ProfilerTM PCR Array Data Analysis web-based software package according to the comparative Ct method, for each functional gene group, treatment with size 2 sample resulted in the great-est number of genes being altered (n = 18) when com-pared with the control, followed by size 1 (n = 10) and

Table 1. Overview of experiment repetition for iTRAQ 8plex sample analysis. iTRAQ 8plex experiment Repetition Sample1 A Control

Size 1 (7 nm–615 nm)Size 2 (616 nm–2.39 µm)Size 3 (2.4 µm–10 µm)

B ControlSize 1 (7 nm–615 nm)Size 2 (616 nm–2.39 µm)Size 3 (2.4 µm–10 µm)

2 C ControlSize 1 (7 nm–615 nm)Size 2 (616 nm–2.39 µm)Size 3 (2.4 µm–10 µm)

D ControlSize 1 (7 nm–615 nm)Size 2 (616 nm–2.39 µm)Size 3 (2.4 µm–10 µm)

Figure 1. Measurements for tissue integrity were taken 24 h after exposure using (A) trans epithelial electric resistance (TEER)—units in resistance values (ohms) and (B) ATP.

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



Table 2. Stress toxicity real-time PCR data analyzed with RT2 ProfilerTM PCR array data analysis web-based software package.

Refseq (Genbank) GeneSize 1 (7–615 nm) Size 2 (0.61–2.39 µm) Size 3 (2.4–10 µm)

Fold change p value Fold change p value Fold change p valueOxidative or metabolic stressNM_001752 CAT 0.87 0.544 0.57 0.410 2.65 0.424NM_001885 CRYAB 0.20 0.354 0.28 0.375 0.96 0.541NM_000499 CYP1A1 1.46 0.202 2.42 0.021 0.83 0.920NM_000773 CYP2E1 0.07 0.239 0.04 0.234 0.25 0.761NM_000780 CYP7A1 0.11 0.266 0.07 0.247 0.45 0.578NM_001979 EPHX2 0.20 0.325 0.23 0.327 1.14 0.699NM_002021 FMO1 0.03 0.067 0.02 0.067 0.10 0.753NM_001461 FMO5 0.07 0.197 0.05 0.192 0.31 0.680NM_000581 GPX1 1.01 0.925 1.72 0.325 1.69 0.369NM_000637 GSR 0.20 0.336 0.15 0.323 0.88 0.757NM_000849 GSTM3 0.06 0.298 0.03 0.281 0.22 0.787NM_002133 HMOX1 0.21 0.255 0.18 0.242 0.81 0.828NM_005953 MT2A 0.32 0.083 0.36 0.026 0.34 0.031NM_000941 POR 0.10 0.188 0.09 0.189 1.52 0.520NM_002574 PRDX1 0.41 0.001 0.28 0.000 0.50 0.584NM_005809 PRDX2 0.48 0.394 0.65 0.435 2.29 0.702NM_000962 PTGS1 0.19 0.277 0.20 0.281 0.89 0.659NM_000454 SOD1 0.56 0.311 0.45 0.052 0.43 0.106NM_000636 SOD2 0.34 0.009 0.24 0.000 0.54 0.157Heat shockNM_001539 DNAJA1 0.78 0.505 0.87 0.540 1.75 0.566NM_007034 DNAJB4 0.58 0.295 0.39 0.191 1.33 0.528NM_005526 HSF1 0.35 0.175 0.49 0.230 1.18 0.611NM_005345 HSPA1A 0.29 0.054 0.22 0.041 0.42 0.157NM_005527 HSPA1L 0.14 0.346 0.14 0.337 0.60 0.717NM_021979 HSPA2 0.16 0.143 0.16 0.120 0.60 0.951NM_002154 HSPA4 0.25 0.336 0.22 0.338 1.20 0.645NM_005347 HSPA5 0.19 0.087 0.17 0.076 0.37 0.356NM_002155 HSPA6 0.10 0.220 0.08 0.214 0.43 0.719NM_006597 HSPA8 0.79 0.507 0.51 0.378 1.80 0.554NM_001540 HSPB1 0.72 0.111 1.01 0.919 0.11 0.887NM_001040141 HSP90AA2 0.47 0.046 0.30 0.010 0.54 0.081NM_007355 HSP90AB1 1.34 0.754 1.18 0.622 2.63 0.556NM_002156 HSPD1 0.57 0.413 0.56 0.421 1.57 0.660NM_002157 HSPE1 0.28 0.082 0.29 0.036 0.32 0.063NM_006644 HSPH1 0.20 0.017 0.19 0.012 0.45 0.666Proliferation and carcinogenesisNM_005190 CCNC 0.00 0.374 0.00 0.374 0.01 0.374NM_053056 CCND1 0.29 0.221 0.24 0.206 0.76 0.673NM_004060 CCNG1 0.37 0.047 0.38 0.039 0.63 0.738NM_005225 E2F1 0.03 0.005 0.02 0.004 0.14 0.556NM_001964 EGR1 0.05 0.008 0.03 0.007 0.22 0.257NM_182649 PCNA 0.33 0.255 0.27 0.236 0.69 0.838Growth arrest and senescenceNM_000389 CDKN1A 1.07 0.899 0.84 0.420 1.00 0.827NM_004083 DDIT3 0.00 0.374 0.01 0.374 0.00 0.374NM_001924 GADD45A 0.04 0.073 0.03 0.069 0.14 0.129NM_004864 GDF15 1.82 0.564 2.39 0.273 1.50 0.728NM_002178 IGFBP6 0.69 0.380 0.76 0.410 2.90 0.209NM_002392 MDM2 0.79 0.721 0.38 0.010 1.06 0.727NM_000546 TP53 0.08 0.214 0.07 0.210 1.51 0.500InflammationNM_002989 CCL21 0.11 0.313 0.14 0.311 0.65 0.609

(Continued)

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



lastly size 3, where very little significant gene dysregula-tion was seen (only two genes downregulated out of an array of 84).

Using the permutation methodology, the observed absolute change between size groups 1 and 2 was 5 genes (non-significant, p = 0.0625), whereas the observed dif-ferences between size groups 1 and 3 (14 genes) and size groups 2 and 3 (19 genes) were both statistically signifi-cant (p = 0.0195 and p = 0.0011, respectively).

iTRAQ and LC-MALDI analysisThe three treatments (sizes 1, 2, and 3) were performed in duplicate with a control within an 8-plex iTRAQ experiment and the iTRAQ experiment was repeated for a total of n = 4 for each treatment, with separate extracts analyzed in each repetition. Over the two iTRAQ

experiments, 231 and 209 proteins were identified respectively with two or more peptides, with 124 proteins found in both experiments. Twelve proteins were signifi-cantly altered in both experiments for treatment size 1 (Table 3), seven were significantly altered for treatment size 2 (Table 4) and for treatment size 3 (Table 5), 7 were significantly altered.

Potential protein interaction network identificationAs genetic “modules” are thought to be highly intercon-nected (Hintze & Adami, 2008), three interaction steps from a given dysregulated protein are thought to be sufficient to include the entire neighborhood of a gene while two steps would be too exclusive (Yip & Horvath, 2007). A set of interacting proteins was thus generated for each treatment consisting of all genes in Uniprot up to

NM_002983 CCL3 0.21 0.374 0.20 0.360 0.76 0.628NM_002984 CCL4 0.08 0.300 0.10 0.297 0.29 0.866NM_000758 CSF2 0.10 0.072 0.36 0.420 0.22 0.880NM_001565 CXCL10 0.10 0.187 0.11 0.179 0.54 0.802NM_001562 IL18 0.88 0.571 0.85 0.501 1.55 0.679NM_000575 IL1A 1.26 0.561 1.22 0.538 2.90 0.625NM_000576 IL1B 0.44 0.372 0.43 0.361 1.60 0.696NM_000600 IL6 0.04 0.109 0.03 0.110 0.17 0.919NM_000595 LTA 0.08 0.274 0.06 0.264 0.56 0.827NM_002415 MIF 0.88 0.437 0.81 0.396 0.96 0.830NM_003998 NFKB1 0.23 0.035 0.19 0.027 0.36 0.274NM_000625 NOS2A 0.08 0.324 0.12 0.332 0.29 0.703NM_000602 SERPINE1 0.29 0.234 0.27 0.220 0.50 0.803Necrosis or apoptosis: DNA damage and repairNM_000051 ATM 0.07 0.224 0.08 0.225 0.45 0.692NM_007194 CHEK2 0.25 0.352 0.22 0.342 0.93 0.829NM_001923 DDB1 0.39 0.286 0.25 0.227 0.93 0.669NM_001983 ERCC1 0.13 0.302 0.15 0.307 0.94 0.700NM_000122 ERCC3 0.09 0.220 0.12 0.236 0.35 0.917NM_005053 RAD23A 0.20 0.070 0.15 0.057 0.75 0.910NM_005732 RAD50 0.04 0.005 0.02 0.004 1.10 0.334NM_007120 UGT1A4 0.12 0.305 0.03 0.274 0.50 0.602NM_003362 UNG 0.20 0.063 0.11 0.045 0.50 0.916NM_006297 XRCC1 0.27 0.339 0.14 0.307 0.79 0.869NM_005431 XRCC2 0.04 0.305 0.03 0.298 0.18 0.710Necrosis or apoptosis: apoptosis signalingNM_001154 ANXA5 0.36 0.067 0.25 0.024 0.71 0.898NM_004324 BAX 0.32 0.137 0.22 0.099 1.06 0.676NM_138578 BCL2L1 0.25 0.082 0.10 0.009 0.56 0.886NM_033292 CASP1 0.23 0.306 0.18 0.290 0.72 0.663NM_001230 CASP10 0.14 0.306 0.06 0.285 0.70 0.716NM_001228 CASP8 0.13 0.232 0.10 0.226 0.21 0.717NM_000639 FASLG 0.14 0.194 0.12 0.174 0.34 0.765NM_020529 NFKBIA 0.37 0.018 0.27 0.005 0.39 0.012NM_000594 TNF 0.17 0.193 0.16 0.183 0.69 0.935NM_001065 TNFRSF1A 0.04 0.054 0.03 0.052 0.14 0.141NM_003810 TNFSF10 0.48 0.165 0.35 0.092 1.03 0.671Significant (p < 0.05) fold changes in italic.

Table 2. (Continued).

Refseq (Genbank) GeneSize 1 (7–615 nm) Size 2 (0.61–2.39 µm) Size 3 (2.4–10 µm)

Fold change p value Fold change p value Fold change p value

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



three-interaction steps away. Using DAVID, we examined each set for enrichment for a gene ontology classifica-tion. This approach revealed potential changes to path-ways involved in various cancers, apoptosis, cell cycle regulation, MAPK signaling and p53 signaling. Given the small number of initial proteins, the GO enrichments found in this case were entirely driven by neighboring proteins. Since these enrichments are not influenced by the data measured in this experiment, the results must be interpreted with caution. Future work would involve confirmation of these inferred protein dysregulations.

Discussion

The level of dosing applied was based on earlier experi-ments (data not shown) that showed no significant changes in cell viability at such a concentration. The

results from the TEER and ATP assays (Figure 1A and B) confirmed that, although the viability and cell integrity did decrease, this was not significantly less than the con-trol samples. The cells did not demonstrate a response consistent with toxic overdosing and were still healthy at the utilized dose. The nature of the study provides information on molecular events that would be difficult to monitor in vivo, but there are also clear limitations. In particular, the isolation of epithelial cells removes effects on other resident cell types such as macrophages that could alter the epithelial cell signaling.

Importantly, size fraction 3, representing the larger particle sizes, showed the lowest gene expression change when compared with the control. Risk estimates for health effects associated with PM are based on epide-miological studies using PM

10 or PM

2.5 as the metric of

exposure. Since the greatest mass in PM10

is associated

Table 3. Proteins altered in HTBE cells after dosing with PM in the size range 7–615 nm (size 1). Size 1 (7 nm–615 nm)Swiss-Prot accession Name Fold change Lower 95% CI Upper 95% CIP45880|VDAC2 Voltage-dependent anion-selective channel protein 2 1.659 1.230 2.236P04083|ANXA1 Annexin A1 1.466 1.088 1.976P15941|MUC1 Mucin-1 precursor 0.731 0.542 0.985P00450|CERU Ceruloplasmin precursor 0.710 0.527 0.958P01833|PIGR Polymeric immunoglobulin receptor precursor 0.692 0.513 0.932P01024|CO3 Complement C3 precursor 0.683 0.507 0.921P00751|CFAB Complement factor B precursor 0.681 0.505 0.918Q9Y6R7|FCGBP IgGFc-binding protein precursor 0.680 0.505 0.917P98088|MUC5A Mucin-5AC precursor 0.677 0.502 0.913P11684|UTER Uteroglobin precursor 0.657 0.487 0.885P80188|NGAL Neutrophil gelatinase-associated lipocalin precursor 0.651 0.483 0.877Q9HC84|MUC5B Mucin-5B precursor 0.647 0.480 0.873

Table 4. Proteins altered in HTBE cells after dosing with PM in the size range 616 nm–2.39 µm (size 2). Size 2 (616 nm–2.39 µm)Swiss-Prot accession Name Fold change Lower 95% CI Upper 95% CIP45880|VDAC2 Voltage-dependent anion-selective channel protein 2 1.660 1.321 2.087P04083|ANXA1 Annexin A1 1.494 1.189 1.878P07355|ANXA2 Annexin A2 1.372 1.091 1.724Q71DI3|H32 Histone H3.2 1.341 1.067 1.685P62805|H4 Histone H4 1.331 1.059 1.673P13073|COX41 Cytochrome c oxidase subunit 4 isoform 1, mitochondrial

precursor1.267 1.008 1.593

Q9HC84|MUC5B Mucin-5B precursor 0.700 0.557 0.880

Table 5. Proteins altered in HTBE cells after dosing with PM in the size range 2.4 µm–10 µm (size 3). Size 3 (2.4 µm–10 µm)Swiss-Prot accession Name Fold change Lower 95% CI Upper 95% CIP80188|NGAL Neutrophil gelatinase-associated lipocalin precursor 1.658 1.364 2.015P45880|VDAC2 Voltage-dependent anion-selective channel protein 2 1.510 1.243 1.836P18669|PGAM1 Phosphoglycerate mutase 1 1.291 1.062 1.569P08118|MSMB β-Microseminoprotein precursor 1.268 1.044 1.542

P62805|H4 Histone H4 1.253 1.030 1.523P25705|ATPA ATP synthase subunit α, mitochondrial precursor 1.219 1.003 1.482

P06731|CEAM5 Carcinoembryonic antigen-related cell adhesion molecule 5 precursor

1.217 1.001 1.479

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



with the larger sized particles, it is therefore important to note the relative gene response by size fraction. Results with size 1 and size 2 show a general trend for down-regulated fold change, whereas the largest size fraction showed fold changes in line with the control or greater, although many of these are not significant. The lack of significance can be attributed in part to the small sample sizes investigated in this pilot study.

A total of four genes within the oxidative and meta-bolic stress functional group were dysregulated. This includes the only upregulated gene, CYP1A1, which was found in size 2 only. This gene encodes for one of the cytochrome P450 family of xenobiotic-metabolizing enzymes, and is known to be involved in the conver-sion of PAH’s to cancerous by-products and has been associated with increased lung cancer risk. MT2A was found to be downregulated in size 2 and size 3 but not in size 1. This gene encodes metallothionein 2A, which is a cysteine-rich protein important in maintaining heavy metal tolerance. MT2A is an inhibitor of cell death and these proteins have been linked to enhanced cell prolif-eration in squamous cell carcinoma of the oesophagus (Cui et al., 2003). The generation of PM-mediated toxicity via the generation of ROS is acknowledged and ROS can lead to activation of transcription factors such as NFKB. NFKB1 was the only gene within this functional group to be affected, being downregulated in both size 1 and size 2 but not in size 3. The observation of a differential in gene regulation by particle size for this inflammatory mecha-nism may be of importance.

It has been shown that expression of heat shock pro-teins is linked with exposure to diesel particulates (Jung et al., 2007). Central to the functions of heat shock pro-teins is their ability to protect the airways from damage resulting from the production of reactive oxygen species, hence inhibition of production of these proteins leads to increased cell damage. Four heat shock proteins were downregulated within this group: HSP1A1A, HSP90AA2, HSPE1, and HSPH1. All these were significantly down-regulated in size 2 while only HSP90AA2 and HSPH1 were likewise affected in size 1, which may be explained by the reduced expression seen in NFKB, which, in turn, regu-lates the promoter of inducible HSP90AA2 (Ammirante et al., 2008).

Of the DNA damage and repair and apoptosis signal-ing groups, the “protective genes” RAD50, UNG, ANXA5, and BCL2L1 were all found to be downregulated. RAD50 is a protein involved in DNA double-stranded break repair, cell cycle and checkpoint activation. Studies using knockout mice have indicated that this gene is essential for cell growth and viability (Lamarche et al., 2010) and the gene UNG is associated with the prevention of muta-genesis. Downregulation of these genes would have a detrimental effect on the repair process. ANXA5 has been associated with an antithrombotic function (Cederholm & Frostegård, 2007) and downregulation of this gene could have consequences in relation to the link of athero-sclerosis and air pollution. BCL2L1 is associated with the

protection of respiratory epithelial cells against oxygen-induced toxicity such as hyperoxia; downregulation of this gene could indicate a loss of protection against oxi-dative stress (Staversky et al., 2010).

Of the 124 proteins identified using iTRAQ, 12, 7, and 7 proteins were significantly altered in the treatments with size 1, size 2, and size 3, respectively (Tables 3–5). VDAC2 was upregulated in all three size groups when compared with the controls, demonstrating a non-size-specific involvement of anion transport regulation in the mitochondrial apoptotic pathway (Blachly-Dyson et al., 1993; Ren et al., 2009).

The airway mucosa is secreted by a variety of epithe-lial cells and functions to protect the alveoli and airways from infection and injury. Dysregulated mucosal-related proteins (MUC1, FCGBP, MUC5A, MUC5B) were mostly found in the size 1 group. Mucins are the main com-ponent of mucus, which has an important function in airway clearance, hydration and particulate trapping. Expression of mucin genes has been shown to be regu-lated by proinflammatory and TLR signalling (Parker & Prince, 2011; Voynow & Rubin, 2009). MUC1 performs an anti-inflammatory role by suppression of TLR2, -3, -4, -5, -7, and -9 signaling (Ueno et al., 2008). These mucosal-related proteins and those proteins found to be dysregu-lated that are listed in Entrez or Uniprot as involved in inflammatory response (ANXA1, CP, C3) or heat-stress response (ANXA2) were found mostly in size 1 or size 2 but not in size 3. The larger size fractions would there-fore appear to elicit a greater immune response than the smallest size fraction. The proteins found to be dysregu-lated by size 3 are mostly involved in cell signaling and energy metabolism.

Conclusions

We have performed an exploratory study into the putative dysregulation of genes and proteins in normal human-derived tracheo-bronchial epithelial (NHTBE) three dimensional cell constructs exposed to PM of different sizes collected from a residential urban street. Genomic analysis profiled expression of 84 genes related to stress and toxicity. For each functional gene group, treatment with size 2 sample resulted in the greatest number of genes with altered expression when compared with controls, followed by size 1 and lastly size 3. The iTRAQ quantitative proteomic approach identified 317 proteins over two experiments (124 in common) and revealed 20 different proteins that were differentially expressed in dosed cells compared with controls. Enrichment for gene ontology classification using an in-house script and DAVID revealed putative links to various pathways. These results provide context for the potential function of the identified proteins and present a focus for further investigation.

In the genomic array, the treatment with size 1 as well as size 2 PM demonstrated a greater dysregulatory effect when compared with controls than size 3 PM. The

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



interpretation of the comparative dysregulatory effect between size 1 PM and size 2 PM is less clear. Integrated RT2 ProfilerTM PCR Array Data Analysis reports 8 more statistically significantly dysregulated genes in size 2 compared to size 1 but analysis using the permutation method reports that the difference in dysregulation between size 1 and size 2 just fails to reach statistical significance. An equal mass dose was used for each size fraction. It is known that the total surface area presented by ultrafine particles is much larger than for larger particles so surface area alone would not appear to be responsible for the observed differential in effect by size. Transition metals, such as cobalt (Co), copper (Cu), iron (Fe), manganese (Mn), nickel (Ni), vana-dium (V), and titanium (Ti), contribute to the oxida-tive capacity of PM (Prahalad et al., 1999). However, we have reported (Price et al., 2010) that elemental concentrations of these metals were not greater in size 2 compared with the other sizes (except for Ni) and for most elements were higher for size 3. There is evidence that diesel exhaust particulates and ultrafine particle effects are mediated by adsorbed chemicals rather that the particles themselves (Xia et al., 2004). Significantly higher levels of PAH levels have been measured in ultra-fine particles (<0.1 µm) compared to fine (<2.5 µm) and coarse particles (2.5–10 µm) (Li et al., 2003). One can speculate that the effects we have observed relate to a difference in chemical concentration by size but we are not aware of data in the literature which could answer this question.

In order to develop appropriate policies to control ambient particles in urban areas it is necessary that epidemiological investigation makes use of the metric of PM which best relates to health outcomes. The appro-priate metric can be explored by examining in detail the causal pathways which are likely to contribute to health endpoints and whether these differ markedly by size and constitution of ambient particles. The main objective of this study was to establish a methodology to investigate the effects of ambient PM size fractions on tissue culture with respect to gene and protein expression and we have shown that it is possible to do this.

Acknowledgements

We would like to acknowledge the Pollution Control Division, Environment Department, City and County of Swansea for their assistance with sample collec-tion and traffic and pollution data. The proteomics analyses were conducted in collaboration with Cardiff University CBS Proteomics Facility (http://www.car-diff.ac.uk/cbs).

Declaration of interest

This work was supported by the Natural Environment Research Council (NERC); Department for Environment Food and Rural Affairs (Defra); Environment Agency

(EA); Ministry of Defence (MOD), and Medical Research Council (MRC) for their funding as part of the Joint Environment and Human Health Programme (Award: NE/E00833X/1). The authors report no declaration of interest.

References(EPAQS) EPoAQS. 1998. Final report of the Expert Panel on Air Quality

Standards on Particles. In: Department for Environment FaRA (ed.). The stationery office. National Publishing: London, UK.

Ammirante M, Rosati A, Gentilella A, Festa M, Petrella A, Marzullo L, Pascale M, Belisario MA, Leone A, Turco MC. 2008. The activity of hsp90 alpha promoter is regulated by NF-kappa B transcription factors. Oncogene 27:1175–1178.

Araujo JA, Nel AE. 2009. Particulate matter and atherosclerosis: role of particle size, composition and oxidative stress. Part Fibre Toxicol 6:24.

Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, Lewis S, Matese JC, Richardson JE, Ringwald M, Rubin GM, Sherlock G. 2000. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet 25:25–29.

Ayres JG, Borm P, Cassee FR, Castranova V, Donaldson K, Ghio A, Harrison RM, Hider R, Kelly F, Kooter IM, Marano F, Maynard RL, Mudway I, Nel A, Sioutas C, Smith S, Baeza-Squiban A, Cho A, Duggan S, Froines J. 2008. Evaluating the toxicity of airborne particulate matter and nanoparticles by measuring oxidative stress potential–a workshop report and consensus statement. Inhal Toxicol 20:75–99.

Balharry D, Sexton K, BéruBé KA. 2008. An in vitro approach to assess the toxicity of inhaled tobacco smoke components: nicotine, cadmium, formaldehyde and urethane. Toxicology 244:66–76.

Beelen R, Hoek G, van den Brandt PA, Goldbohm RA, Fischer P, Schouten LJ, Jerrett M, Hughes E, Armstrong B, Brunekreef B. 2008. Long-term effects of traffic-related air pollution on mortality in a Dutch cohort (NLCS-AIR study). Environ Health Perspect 116:196–202.

Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J Royal Statistical Society Series B (Methodological) 57:289–300.

Blachly-Dyson E, Zambronicz EB, Yu WH, Adams V, McCabe ER, Adelman J, Colombini M, Forte M. 1993. Cloning and functional expression in yeast of two human isoforms of the outer mitochondrial membrane channel, the voltage-dependent anion channel. J Biol Chem 268:1835–1841.

Borm PJ, Kelly F, Künzli N, Schins RP, Donaldson K. 2007. Oxidant generation by particulate matter: from biologically effective dose to a promising, novel metric. Occup Environ Med 64:73–74.

Brennan P, Shore AM, Clement M, Hewamana S, Jones CM, Giles P, Fegan C, Pepper C, Brewis IA. 2009. Quantitative nuclear proteomics reveals new phenotypes altered in lymphoblastoid cells. PROTEOMICS – Clinical Applications 3:359–369.

Brook RD. 2008. Cardiovascular effects of air pollution. Clin Sci 115:175–187.

Cederholm A, Frostegård J. 2007. Annexin A5 as a novel player in prevention of atherothrombosis in SLE and in the general population. Ann N Y Acad Sci 1108:96–103.

Cui Y, Wang J, Zhang X, Lang R, Bi M, Guo L, Lu SH. 2003. ECRG2, a novel candidate of tumor suppressor gene in the esophageal carcinoma, interacts directly with metallothionein 2A and links to apoptosis. Biochem Biophys Res Commun 302:904–915.

Donaldson K, Tran L, Jimenez LA, Duffin R, Newby DE, Mills N, MacNee W, Stone V. 2005. Combustion-derived nanoparticles: a review of their toxicology following inhalation exposure. Part Fibre Toxicol 2:10.

Gauderman WJ, Avol E, Gilliland F, Vora H, Thomas D, Berhane K, McConnell R, Kuenzli N, Lurmann F, Rappaport E, Margolis H, Bates

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.



D, Peters J. 2004. The effect of air pollution on lung development from 10 to 18 years of age. N Engl J Med 351:1057–1067.

Harrison RM, Giorio C, Beddows DC, Dall’Osto M. 2010. Size distribution of airborne particles controls outcome of epidemiological studies. Sci Total Environ 409:289–293.

HEI. 2010. HEI Panel on the Health Effects of Traffic-Related Air Pollution. Traffic-related air pollution: a critical review of the literature on emissions, exposure and health effects. HEI Special Report 17. Health Effects Institute: Boston, MA.

Hintze A, Adami C. 2008. Evolution of complex modular biological networks. PLoS Comput Biol 4:e23.

Hoffmann B, Moebus S, Stang A, Beck EM, Dragano N, Möhlenkamp S, Schmermund A, Memmesheimer M, Mann K, Erbel R, Jöckel KH; Heinz Nixdorf RECALL Study Investigative Group. 2006. Residence close to high traffic and prevalence of coronary heart disease. Eur Heart J 27:2696–2702.

Huang da W, Sherman BT, Lempicki RA. 2009. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc 4:44–57.

Jones T, Moreno T, BéruBé K, Richards R. 2006. The physicochemical characterisation of microscopic airborne particles in south Wales: a review of the locations and methodologies. Sci Total Environ 360:43–59.

Jung EJ, Avliyakulov NK, Boontheung P, Loo JA, Nel AE. 2007. Pro-oxidative DEP chemicals induce heat shock proteins and an unfolding protein response in a bronchial epithelial cell line as determined by DIGE analysis. Proteomics 7:3906–3918.

Keskinen J, Pietarinen K, Lehtimäki M. 1992. Electrical low pressure impactor. J Aerosol Sci 23:353–360.

Khatri P, Draghici S. 2005. Ontological analysis of gene expression data: current tools, limitations, and open problems. Bioinformatics 21:3587–3595.

Knol AB, de Hartog JJ, Boogaard H, Slottje P, van der Sluijs JP, Lebret E, Cassee FR, Wardekker JA, Ayres JG, Borm PJ, Brunekreef B, Donaldson K, Forastiere F, Holgate ST, Kreyling WG, Nemery B, Pekkanen J, Stone V, Wichmann HE, Hoek G. 2009. Expert elicitation on ultrafine particles: likelihood of health effects and causal pathways. Part Fibre Toxicol 6:19.

Kreyling WG, Semmler-Behnke M, Möller W. 2006. Ultrafine particle-lung interactions: does size matter? J Aerosol Med 19:74–83.

Laden F, Schwartz J, Speizer FE, Dockery DW. 2006. Reduction in fine particulate air pollution and mortality: Extended follow-up of the Harvard Six Cities study. Am J Respir Crit Care Med 173:667–672.

Lamarche BJ, Orazio NI, Weitzman MD. 2010. The MRN complex in double-strand break repair and telomere maintenance. FEBS Lett 584:3682–3695.

Lau AT, Chiu JF. 2009. Biomarkers of lung-related diseases: current knowledge by proteomic approaches. J Cell Physiol 221:535–543.

Li N, Sioutas C, Cho A, Schmitz D, Misra C, Sempf J, Wang M, Oberley T, Froines J, Nel A. 2003. Ultrafine particulate pollutants induce oxidative stress and mitochondrial damage. Environ Health Perspect 111:455–460.

Li N, Xia T, Nel AE. 2008. The role of oxidative stress in ambient particulate matter-induced lung diseases and its implications in the toxicity of engineered nanoparticles. Free Radic Biol Med 44:1689–1699.

Magrane M, Consortium U. 2011. UniProt Knowledgebase: a hub of integrated protein data. Database (Oxford) 2011:bar009.

Miller KA, Siscovick DS, Sheppard L, Shepherd K, Sullivan JH, Anderson GL, Kaufman JD. 2007. Long-term exposure to air pollution and incidence of cardiovascular events in women. N Engl J Med 356:447–458.

Parker D, Prince A. 2011. Innate immunity in the respiratory epithelium. Am J Respir Cell Mol Biol 45:189–201.

Prahalad AK, Soukup JM, Inmon J, Willis R, Ghio AJ, Becker S, Gallagher JE. 1999. Ambient air particles: effects on cellular oxidant radical generation in relation to particulate elemental chemistry. Toxicol Appl Pharmacol 158:81–91.

Price H, Arthur R, Sexton K, Gregory C, Hoogendoorn B, Matthews I, Jones T, BéruBé K. 2010. Airborne particles in Swansea, UK: their collection and characterization. J Toxicol Environ Health Part A 73:355–367.

Putaud JP, Van Dingenen R, Alastuey A, Bauer H, Birmili W, Cyrys J, Flentje H, Fuzzi S, Gehrig R, Hansson HC, Harrison RM, Herrmann H, Hitzenberger R, Hüglin C, Jones AM, Kasper-Giebl A, Kiss G, Kousa A, Kuhlbusch TAJ, Löschau G, Maenhaut W, Molnar A, Moreno T, Pekkanen J, Perrino C, Pitz M, Puxbaum H, Querol X, Rodriguez S, Salma I, Schwarz J, Smolik J, Schneider J, Spindler G, ten Brink H, Tursic J, Viana M, Wiedensohler A, Raes F. 2010. A European aerosol phenomenology - 3: Physical and chemical characteristics of particulate matter from 60 rural, urban, and kerbside sites across Europe. Atmospheric Environ 44:1308–1320.

Ren D, Kim H, Tu HC, Westergard TD, Fisher JK, Rubens JA, Korsmeyer SJ, Hsieh JJ, Cheng EH. 2009. The VDAC2-BAK rheostat controls thymocyte survival. Sci Signal 2:ra48.

Richards AL, Holmans P, O’Donovan MC, Owen MJ, Jones L. 2008. A comparison of four clustering methods for brain expression microarray data. BMC Bioinformatics 9:490.

Sexton K, Balharry D, BéruBé KA. 2008. Genomic biomarkers of pulmonary exposure to tobacco smoke components. Pharmacogenet Genomics 18:853–860.

Shilov IV, Seymour SL, Patel AA, Loboda A, Tang WH, Keating SP, Hunter CL, Nuwaysir LM, Schaeffer DA. 2007. The Paragon Algorithm, a next generation search engine that uses sequence temperature values and feature probabilities to identify peptides from tandem mass spectra. Mol Cell Proteomics 6:1638–1655.

Simkhovich BZ, Kleinman MT, Kloner RA. 2008. Air pollution and cardiovascular injury epidemiology, toxicology, and mechanisms. J Am Coll Cardiol 52:719–726.

Staversky RJ, Vitiello PF, Yee M, Callahan LM, Dean DA, O’Reilly MA. 2010. Epithelial ablation of Bcl-XL increases sensitivity to oxygen without disrupting lung development. Am J Respir Cell Mol Biol 43:376–385.

Tonne C, Melly S, Mittleman M, Coull B, Goldberg R, Schwartz J. 2007. A case-control analysis of exposure to traffic and acute myocardial infarction. Environ Health Perspect 115:53–57.

Ueno K, Koga T, Kato K, Golenbock DT, Gendler SJ, Kai H, Kim KC. 2008. MUC1 mucin is a negative regulator of toll-like receptor signaling. Am J Respir Cell Mol Biol 38:263–268.

Voynow JA, Rubin BK. 2009. Mucins, mucus, and sputum. Chest 135:505–512.

Xia T, Korge P, Weiss JN, Li N, Venkatesen MI, Sioutas C, Nel A. 2004. Quinones and aromatic chemical compounds in particulate matter induce mitochondrial dysfunction: implications for ultrafine particle toxicity. Environ Health Perspect 112:1347–1358.

Xiao GG, Wang M, Li N, Loo JA, Nel AE. 2003. Use of proteomics to demonstrate a hierarchical oxidative stress response to diesel exhaust particle chemicals in a macrophage cell line. J Biol Chem 278:50781–50790.

Yip AM, Horvath S. 2007. Gene network interconnectedness and the generalized topological overlap measure. BMC Bioinformatics 8:22.

Zervas E, Dorlhène P. 2006. Comparison of exhaust particle number measured by EEPS, CPC, and ELPI. Aerosol Science and Technology 40:977–984.

Zhang L, Wang M, Kang X, Boontheung P, Li N, Nel AE, Loo JA. 2009. Oxidative stress and asthma: proteome analysis of chitinase-like proteins and FIZZ1 in lung tissue and bronchoalveolar lavage fluid. J Proteome Res 8:1631–1638.

Inha

latio

n T

oxic

olog

y D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Fr

anci

s A

Cou

ntw

ay L

ibra

ry o

f M

edic

ine

on 0

5/16

/13

For

pers

onal

use

onl

y.

gene and protein responses of human lung tissue explants exposed to ambient particulate matter of...

Documents