SEARCH

SEARCH BY CITATION

Keywords:

  • air pollution;
  • epigenetics;
  • genomics;
  • lung disease;
  • single-nucleotide polymorphism

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Adverse health effects from air pollutants remain important, despite improvement in air quality in the past few decades. The exact mechanisms of lung injury from exposure to air pollutants are not yet fully understood. Studying the genome (e.g. single-nucleotide polymorphisms (SNP) ), epigenome (e.g. methylation of genes), transcriptome (mRNA expression) and microRNAome (microRNA expression) has the potential to improve our understanding of the adverse effects of air pollutants. Genome-wide association studies of SNP have detected SNP associated with respiratory phenotypes; however, to date, only candidate gene studies of air pollution exposure have been performed. Changes in epigenetic processes, such DNA methylation that leads to gene silencing without altering the DNA sequence, occur with air pollutant exposure, especially global and gene-specific methylation changes. Respiratory cell line and animal models demonstrate distinct gene expression signatures in the transcriptome, arising from exposure to particulate matter or ozone. Particulate matter and other environmental toxins alter expression of microRNA, which are short non-coding RNA that regulate gene expression. While it is clearly important to contain rising levels of air pollution, strategies also need to be developed to minimize the damaging effects of air pollutant exposure on the lung, especially for patients with chronic lung disease and for people at risk of future lung disease. Careful study of genomic responses will improve our understanding of mechanisms of lung injury from air pollution and enable future clinical testing of interventions against the toxic effects of air pollutants.


INTRODUCTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Adverse health effects from air pollutants—particulate matter, ozone and nitrogen dioxide—remain important, despite improvement in air quality in the past few decades.1 It has been estimated that exposure to air pollution contributes to 6% of total mortality.2 Epidemiological and clinical studies have shown that exposure to air pollutants is associated with increased mortality, higher admission rates for respiratory diseases and decreased lung function in patients with lung disease. Understanding mechanisms of injury from air pollutants has implications for reducing susceptibility to asthma, chronic obstructive pulmonary disease (COPD), lung cancer and other lung diseases, and for reducing disease severity from asthma and COPD, including airway remodelling and acute exacerbations. However, the exact mechanisms of lung injury from exposure to air pollutants are not yet fully understood. This review will focus on advances in genomics and on how approaches at the genomic level would potentially improve our understanding of the adverse effects of air pollutant exposure in the lung, to better prevent and treat lung disease.

GENOME AND GENE–ENVIRONMENT INTERACTION

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Inter-individual variability in the respiratory effects of air pollutants

Much research has been conducted into the biological response to exposure to air pollutants at the molecular, cellular and whole organism level. This has clearly established that there are significant differences in biological responses induced by differing types of air pollution, containing varying constituents (e.g. particulates, ozone and nitrogen dioxide), derived from a variety of sources. However, equally important in determining the health effects of air pollution exposure on the person is the inter-individual variation in responses to exposure. For example, even in healthy individuals who have undergone controlled exposure to a pollutant, it is clear that there is a wide variability in physiological responses.3–7 This variability in responses to pollutant exposure has been attributed to both extrinsic (e.g. diet, pre- or concurrent exposure, climactic conditions, and pre-existing disease) and intrinsic (e.g. age and gender) factors.8 However, it is clear that key intrinsic factors in determining individual response to air pollution exposure are genetic, and epigenetic differences between subjects, especially given that processes involved in the response to air pollutants—oxidative stress and inflammation—are known to be under genetic regulation.8,9 The contribution that genetic factors can make to responses is highlighted by the differences in responses to controlled exposure to ozone, particulate pollutants and nitrogen dioxide between inbred strains of laboratory animals.10–12 Identifying genetic determinants of variability in response in humans is clearly important for understanding mechanisms of lung disease, and recognizing at-risk groups who would benefit the most from preventive strategies. Furthermore, identification of at-risk groups, the degree of their sensitivity to exposure and their frequency in the population will aid in the cost–benefit analysis of ‘safe’ exposure levels in the public health setting.

Genetic approaches to the study of the respiratory effects of air pollution exposure

There are two main approaches to the study of the genetics of disease, the candidate gene approach and the genome-wide or hypothesis-independent approach. In the candidate gene approach, genetic variation in individual genes is directly assessed for association with the disease phenotype of interest. In general, candidate genes are selected for analysis because of a known or postulated role for the encoded product of the gene in the disease process or an expression pattern associated with the disease. Polymorphisms within the gene that are believed to be functional (i.e. affecting gene expression or encoded protein function), or that are selected for maximal information on the basis of linkage disequilibrium patterns surrounding the gene (often termed tagging single-nucleotide polymorphisms (SNP) ), are then tested for association with the disease or phenotype in question. However, by definition, the candidate-gene approach is not capable of identifying novel genes, contributing to the phenotype being analysed. Thus limiting the insight that such genetic studies can provide into the biological basis of the disease pathogenesis or physiological phenotype (i.e. inflammatory responses to air pollution exposure) being studied.

In the past, the use of hypothesis-independent approaches to identify novel genetic loci involved the genotyping of microsatellite-based polymorphisms and use of linkage analysis in family-based cohorts. In recent years, the study of the genetic basis of complex disease has been revolutionized by technological advances in array-based SNP genotyping technologies and the characterization of millions of SNP variants in the human genome.13 This has made possible the simultaneous determination of the genotype of several million SNP throughout the genome of an individual through the use of array-based platforms. This has allowed the use of genome-wide hypothesis-independent association studies in case-control or population-based samples. Such genome-wide association studies (GWAS) have now transformed the study of genetic factors in complex common disease.14,15 For hundreds of phenotypes, from common diseases to physiological measurements, such as height and body mass index, and biological measurements, such as circulating lipid levels and blood eosinophil levels, GWAS have provided compelling statistical associations for hundreds of different loci in the human genome, giving new insight into the biological processes that underlie these phenotypes and diseases.16

Candidate association studies of the respiratory effects of air pollution

While we still await the results of studies using a genome-wide approach to identify genetic variants that can explain the variability in response to controlled exposure, there have been a number of candidate gene studies that have identified some important genetic determinants of response to exposures such as ozone and particulate matter and sulphur dioxide, and these have been reviewed elsewhere.17 For example, in a study of ozone exposure, 24 healthy non-smokers performed bicycle rides for 2 h outdoors and genotyped for functional polymorphisms in the oxidant/antioxidant genes NAD(P)H dehydrogenase quinone 1 (NQO1) and glutathione-S-transferase MI (GSTM1).18 When the ozone concentration was high (>80 parts per billion), subjects with the high oxidant-producing genotypes (NQO1 Pro/Pro187 and GSTM1-null, n = 8) had a greater fall in forced expiratory volume in 1 s (FEV1), and increased airway epithelial damage, as measured by serum CC16, compared with subjects without this genotype. Subjects with susceptible genotypes also sustained excessive DNA damage by reactive oxygen species, as shown by an increase in the biomarker, 8-OHdG.18 These results were confirmed in a second study, in which ozone exposure during exercise resulted in higher levels of oxidative stress (lipid peroxidation products) in subjects with susceptible genotypes.19 Similar modulatory effects of antioxidant gene polymorphisms have been seen in studies of the effects of controlled exposure to particulate pollution on allergic responses to ragweed allergen with low antioxidant genotypes (GSTM1-null and GSTP1 Ile/Ile) exhibiting enhanced nasal responses, as indicated by increased nasal immunoglobulin E (IgE) and histamine.20 These genotype-specific responses to ragweed were also aggravated in the presence of second-hand smoke.21

As well as genetic variation in metabolizing enzymes modulating the extent of oxidative stress, other studies have examined polymorphisms that may determine the level of inflammatory response induced. For example, we performed a genetic association study of tumour necrosis factor-α (TNF) polymorphisms and ozone exposure in 51 participants who inhaled ozone during intermittent exercise.22 With ozone challenge, there was a statistically significantly greater fall in FEV1 (–9% of baseline) in individuals with the TNF−308G/G genotype, compared with subjects with the −308G/A or A/A genotypes (–3% of baseline). A similar association was found when combinations of polymorphisms (haplotypes) in the TNF gene were analysed. Specifically, the lymphotoxin-α (LTA) +252G/TNF−1031T/TNF−308A/TNF−238G haplotype conferred the smallest change in FEV1 with ozone exposure.

GWAS of the respiratory effects of air pollution

One approach to identification of genetic factors that modulate the biological responses to exposure is to measure the results of controlled exposures in healthy individuals. The advantages of such an approach are that the level, composition and timing of the exposure can be carefully controlled and standardized. This allows the measurement of intermediate phenotypes such as lung function parameters, measurements of inflammation in lavage or biopsies and assessment of changes in gene expression. While to date no study has combined the controlled exposure approach with a GWAS design, it is clear that in the future, this approach may provide important insights into the biological mechanisms underlying the response to exposure.

For example, a recent GWAS has used this controlled exposure approach to identify determinants of response to radiation therapy that varies widely among patients. Niu et al. performed a GWAS using 277 ethnically defined human lymphoblastoid cell lines (LCL) in which data for 1.3 million SNP markers were available.23 Genotypes were tested for association with data from a non-radioactive cell proliferation assay (MTS assay) to obtain area under the curve (AUC) as a radiation response phenotype. The results of the analysis, coupled with genome-wide gene expression analysis, identified 50 SNP in 14 of the 27 loci that were associated with both AUC and the expression of 39 genes, which were also associated with radiation AUC (P < 10−3). Functional validation using siRNA knockdown cell lines showed that C13orf34, MAD2L1, PLK4, TPD52 and DEPDC1B each significantly altered radiation sensitivity in at least two cancer cell lines.23 Similar studies utilizing LCL exposed to pollutants could help to identify novel biomarkers that might contribute to variation in response to air pollution exposure and enhance our understanding of mechanisms underlying that variation.

While there have been no GWAS to date examining the respiratory effects of controlled pollution exposure, recently, there has been several large GWAS of lung function per se. The SPIROMETA study tested genome-wide association with FEV1 and the FEV1/forced vital capacity (FVC) ratio in 20 288 individuals of European ancestry and then replicated the top signals in 32 184 additional individuals together with and in silico summary association data from the CHARGE Consortium (n = 21 209) and the Health 2000 survey (n ≤ 883). A number of loci were identified associated with FEV1 (Hedgehog interacting protein (HHIP) (previously associated with COPD24 and height25), TNS1, HTR4 and GSTCD) and FEV1/FVC (DAAM2, AEGR and THSD4).26 Simultaneously, the CHARGE Consortium presented the results of a similar analysis in 20 890 participants of European ancestry with the identification of eight loci associated with FEV1/FVC (HHIP, GPR126, ADAM19, AGER-PPT2, FAM13A, PTCH1, PID1 and HTR4) and one locus associated with FEV1 (INTS12-GSTCD-NPNT) at or near genome-wide significance (P < 5 × 10–8). A combined analysis of both these studies including a number of addition cohorts (n = 48 201 with genome-wide genotyping and replication in an additional 46 411 individuals), and this has identified a further 16 loci associated with lung function,27 identifying a number of new proteins whose role in determining lung function remain to determined.

It is important to note that in these GWAS, lung function was measured cross-sectionally. The level of FEV1 at a given time point in an individual depends on several potentially independent processes: the maximum lung function obtained during development as well as the rate of lung function decline in later life, disease state, and environmental exposure. It is possible that some of the loci identified as associated with lung function will modulate the respiratory effects of air pollution exposure on any of these processes. For example, the GSTCD gene, while encoding a protein of unknown function, does have partial homology with the glutathione-S-transferases that have shown to modulate the respiratory effects of exposure to oxidants such as ozone,28 and interact with air pollution exposure to determine lung function growth in children,29 and rate of decline in lung function in adults.30 However, further analysis of these variants in controlled exposure studies is now required to validate this.

Gene–environment interaction: air pollution exposure, genetic variation and respiratory disease

The development and severity of lung disease such as asthma and COPD depends on the interaction between inherited susceptibility (genetic and epigenetic) and response to exposures (environment), including exposure to air pollution.31–33 Effects of genetic polymorphisms may be only seen in exposed populations, or vice versa, and in some circumstances, ‘flip-flop’ effects may be present where the effect of alleles on disease outcome is reversed depending on environmental exposure. Thus, the rationale for study the interaction between genes and environment includes:34

  • • 
    Obtaining better estimates of the population-attributable risk for different genetic and environmental risk factors by accounting for their joint interactions
  • • 
    Strengthening observed associations between environmental factors and diseases by examining these factors in genetically susceptible individuals and the utilization of Mendelian randomization strategies to infer causality for environmental exposures
  • • 
    Identification of key biological pathways that underlie response to environmental exposures providing both insight into which components of complex mixtures of pollutants cause disease as well as better understanding of pathogenesis allowing development of therapeutic strategies

A large number of studies have now examined the interaction between candidate gene polymorphisms, air pollution exposure and the onset and severity of respiratory disease. One focus of research has been the role of antioxidant gene polymorphism. A number of such polymorphisms have shown to exert gene-only effects on asthma and COPD susceptibility. For example, polymorphism of members of the GST gene family have shown to be associated with risk of both asthma35 and COPD.36 Therefore, given the known oxidative stress effects of air pollution exposure, many groups have now examined the interaction between such polymorphisms, air pollution exposure and onset of these diseases. A recent systematic review by Minelli et al. identified 12 such studies together with 3 experimental studies of controlled exposure, with 12 studies supporting the presence of interaction.37 These included studies of genes including GSTM1, GSTP1, GSTT1, NQO1, SOD2 (MnSOD), GPX1 and NQO1. For example, several studies have showed an interaction among GST genes, asthma and/or childhood wheezing, and exposure to ozone,38–40 particulate exposure39–41 oxides of nitrogen,42 and ambient pollution.43 Furthermore, a clinical trial performed in Mexico City, where outdoor concentrations of ozone are high throughout the year, showed that asthmatic children with the low antioxidant GSTM1 genotype had a fall in lung function with increasing ozone concentration, unless protected by antioxidant supplementation with vitamins C and E,44 illustrating how identification of gene-environment interactions may enable targeted chemoprevention for asthmatics who are genetically susceptible to the adverse respiratory effects of ozone.

Despite the undoubted success of the GWAS approach in identifying genetic susceptibility variants for respiratory diseases, as for other complex diseases, the identified variants explain only a small proportion of the estimated heritability of these conditions. This unexplained heritability could be partly due to gene–environment (G × E) interactions or more complex pathways involving multiple genes and exposures. The standard approach to analysis of genome-wide association data scans for main effects and ignores the potentially useful information in the available environmental exposure data. There has therefore been considerable interest in exploring G × E interactions in a genome-wide hypothesis independent manner in studies where exposure information is available. Such studies have been termed genome-wide interaction studies (GWIS).45

While such an approach is of interest to detect novel G × E interactions, it comes with a number of challenges.46 First, there is the issue of accurate exposure assessment, which is a relatively difficult task.47 Various approaches have been used to measurement have been employed, including self-reported exposure (e.g. to farm animals or environmental tobacco smoke), correlation with epidemiological data (e.g. concurrent measurement of house dust endotoxin or the utilization of geographic information systems and spatial mapping of pollution exposures) and controlled exposure in the laboratory. Nonetheless, accurate exposure assessment remains a challenge for future studies.

Second, consideration needs to be made to the issue of sample size and power. Estimates of sample-size requirements for G × E studies can be huge, with typically detection of an interaction requiring a sample size at least four times larger than that required for the detection of a main effect of similar size. This suggests that while GWAS typically involve analysis of thousands to tens of thousands of cases, assessment of G × E interaction at the genome-wide level could easily require analysis of hundreds of thousands of cases.46 A number of potential methods have been proposed to reduce the number of comparisons needed in such analyses by prioritizing the large number of SNP tested to highlight those most likely to be involved in a G × E interaction.48–50

Finally, there is the issue of heterogeneity between studies. As for gene-only effects, failure to replicate G × E interactions may result from genetic heterogeneity (from differing ethnicity with differing genetic background), exposure heterogeneity (e.g. different sizes or chemical constituents of particulate air pollution across regions), differing methods of exposure assessment and differing confounders between study populations.

Despite these challenges, investigators have begun to utilize GWIS to identify genetic variants underlying complex diseases. For example, Beaty et al. undertook a GWAS of non-syndromic cleft palate, a common birth defect with a complex and heterogeneous aetiology involving both genetic and environmental risk factors. While no SNP achieved genome-wide significance when considered alone, markers in several genes attained or approached genome-wide significance when G × E interaction with three common maternal exposures (maternal smoking, alcohol consumption, and multivitamin supplementation) was included.51 This illustrates the need to consider G × E interaction when searching for genes influencing risk to complex and heterogeneous disorders, as many genetic effects may not be detectable in unexposed populations.

While to date there have been no reported GWIS investigating air pollution exposure and respiratory disease, a GWIS investigating the effect of farming environments on the risk of childhood asthma has recently been undertaken.52 In this study, data on 500 000 SNP were assessed for interaction with seven farm-related exposures (living on a family-run farm, mother grew up on a farm, regular consumption of raw farm milk, regular contact with cows, regular contact with straw, regular contact with hay, coincidence of cow and straw exposure) in 1708 children that formed part of the larger GABRIEL study into the genetics of asthma.53 The GWIS did not reveal any significant interactions with common SNP for which the study had >50% power. However, among rarer SNP, 15 genes with crossover interactions or effect concentrations in the exposed group for asthma or atopy in relation to farming, consumption of farm milk, and contact with cows and straw, were identified, many showing showed a ‘flip-flop’ pattern of association. While these interactions deserve further investigation, of more interest, no interactions were observed involving SNP in genes previously identified as showing interactions with farming exposures in previous candidate gene studies such as the endotoxin receptors CD14 and TLR4. This may reflect issues with exposure assessment, as endotoxin levels were not directly measured in the population, and farming exposure, while correlated with endotoxin exposure, is nonetheless a surrogate measure of exposure. This highlights the critical need for accurate exposure assessment in such studies.

In conclusion, the future utilization of a GWIS approach to identify interactions between genetic susceptibility factors and air pollution exposure has undoubted potential to increase our understanding of the mechanisms by which exposure causes respiratory disease. However, even more so than studies of genetic and environmental factors in isolation, issues of statistical power, accurate exposure assessment, and controlling for confounding factors will be critical to the success of such studies.

EPIGENOME

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Epigenetic regulation of gene expression

Epigenetic changes are heritable changes in gene expression and regulation that occur without alteration of the DNA sequence. The epigenome is the cell's complete collection of changes that transmit epigenetic information.54 Epigenetic processes, such as DNA methylation and histone modification, have been implicated in a wide range of chronic lung diseases, including lung cancer, asthma and COPD.55–61 DNA methylation mainly involves the addition of a methyl group (-CH3) to cytosine contained within cytosine-guanine (CpG) dinucleotides, at the fifth carbon of the pyrimidine ring.62 CpG methylation is involved in several processes including tissue-specific silencing of gene expression, genomic imprinting, X chromosome inactivation, chromosome stabilization and chromatin condensation.63 Environmental factors—including air pollution, diet, infection and smoking—can all affect DNA methylation patterns.58

Exposure to air pollution and changes in gene-specific methylation

Exposure to environmental agents such as cigarette smoke and air pollutants has shown to induce changes in DNA methylation. Prenatal cigarette smoke exposure in utero leads to hypomethylation of repetitive elements and alterations in gene-specific methylation,64 which has implications for the pathogenesis of asthma. Exposure to air pollutants in animal has also been demonstrated to alter methylation of genes. Inhalation of diesel exhaust particles (DEP) in BALB/c mice sensitized to Aspergillus fumigatus resulted in hypermethylation of the interferon-gamma promoter and hypomethylation of the IL4 promoter in CD4+ T lymphocytes, leading to altered IgE production.65

Several human studies have focused on gene-specific methylation, showing methylation changes with air pollutant exposure. Individuals exposed to smoky coal, used in some countries for cooking and heating indoors, are at higher risk for the development of lung cancer. A study performed in China showed that sputum DNA from people chronically exposed to smoky coal emissions had a higher frequency (51%) of promoter methylation of p16, a tumour suppressor gene.66 This finding was not altered by the smoking status of the participants in this study, some of whom were never smokers. Smoky coal emissions contain polycyclic aromatic hydrocarbons, among other pollutants, which may be involved in the process of hypermethylation. A study correlating wood smoke exposure, COPD risk and gene promoter methylation of a panel of eight lung cancer-related genes showed that p16 methylation in sputum was associated with a significantly lower predicted FEV1 in subjects with prior wood smoke exposure.67 In addition, higher odds of airflow obstruction and lower percent of FEV1 predicted were observed in the presence of aberrantly methylated GATA4.67 These epidemiological studies support the relevance of air pollutant exposure, especially coal and wood smoke, in aberrant methylation of key lung cancer-associated genes.

Methylation changes with exposure to cigarette smoke have been found in peripheral blood DNA. In a study of 177 participants (current smokers, former smokers and never smokers), microarray methylation profiling of peripheral blood lymphocyte DNA revealed that in smokers, there was lower methylation of the coagulation factor II receptor-like 3 (F2RL3) gene, which has a role in platelet activation.68

Use of global methylation profiling to discover epigenomic effects of air pollution

To progress beyond gene-specific investigations of methylation status for single genes, environmental studies have used epigenome-wide approaches. Methods employed include measurement of DNA methylation of repetitive DNA elements (giving a global measure of the amount of DNA methylation in the epigenome) and profiling using microarrays. In animal models of exposure (modelling human exposure over long periods), mice exposed to particulate air pollution in an urban environment, near steel mills and a major highway, had higher levels of global DNA methylation in sperm DNA during exposure, compared with mice breathing filtered clean air. This hypermethylation persisted after exposure ceased.69 Higher frequencies of DNA mutations and DNA strand breaks also occurred, potentially increasing mutagenicity.

Exposure studies of workers and residents have been useful in characterizing the effects of acute and chronic exposure to air pollutants on DNA methylation of genes. For example, 63 steel production plant workers had DNA methylation measured in lymphocyte DNA from peripheral blood taken on their first day back from work (pre-exposure) and after 3 days of work (post-exposure).70 Steel workers are exposed to particulate matter, especially PM10. Testing of Alu and long interspersed nuclear element-1 (LINE-1) repeats, which are repetitive DNA sequences in the genome, showed no changes in global DNA methylation content with short-term exposure. However, higher chronic PM10 exposure was associated with global hypomethylation, indicating long-term genomic changes with chronic exposure. Chronic PM10 exposure was also associated with decreased gene-specific promoter methylation of the iNOS gene, an inflammatory pathway gene, consistent with previous reports of increased inducible nitric oxide synthase (iNOS) expression after air pollution exposure.70

Other studies have found changes in methylation of repeat elements of DNA sequence with air pollutant exposure. In a large cohort study of over 700 participants in the normative aging study, changes in DNA methylation of LINE-1 and the short interspersed Alu repeats were correlated with acute or chronic exposure to ambient air pollution.71,72 These repeats are retrotransposons, which are highly mobile and variable repeats that alter the structure of intergenic regions and lead to genomic diversity.73 The exposures measured were ambient particles (particulate matter ≤ 2.5 µm in aerodynamic diameter, PM2.5), and two components of PM2.5: black carbon and sulphates. There were rapid DNA methylation changes in LINE-1 repetitive elements after short-term (up to 7 days) exposure to particulate matter.71 With chronic exposures, an increase in black carbon over a 90-day exposure period was associated with a decrease of 5-methylcytosine (methylation mark) in Alu repeats.72 In addition, an increase in sulphates over a 90-day period was associated with a decrease of 5mC in LINE-1. The authors also genotyped GSTM1, a xenobiotic metabolizing enzyme, and found that the null genotype, which has impaired function, enhanced the association between black carbon and Alu hypomethylation.72 Given that LINE-1 and Alu repetitive elements may be involved in inflammation and immune responses, this study of epigenetic changes in an air pollution cohort has demonstrated that prolonged exposure to particulate air pollutants may lead to hypomethylation of repetitive elements in DNA.

These epigenetic studies demonstrate that exposure to air pollutants induces gene-specific and global methylation changes in the lungs. Our challenge will be to determine the functional importance of these epigenetic changes, which are involved in gene silencing. The potential now for epigenome-wide association studies to be conducted opens up the opportunity to combine epigenomic and genomic (GWAS) information in population studies of the effects of air pollution.54

Histone modification and air pollution exposure

Post-translational modification of histones in the nucleosome of chromatin regulates gene expression. Histone deacetylases (HDAC) act in counterbalance with histone acetyltransferases (HAT) to regulate histones and chromatin structure, and alter the rate of transcription. Exposure of BEAS-2B bronchial epithelial cells to DEP resulted in the selective degradation of HDAC1, and activation of HAT p300, a transcriptional coactivator.74 In addition, there was increased acetylation of histone H4, a component of the nucleosome, in the promoter region of the cyclooxygenase-2 (COX-2) gene, an inflammatory mediator whose expression was increased with DEP. This study provides evidence for inflammation arising through histone modification, upon exposure to particulate matter.

TRANSCRIPTOME

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Potential applications of transcriptomics to studying the response to air pollution

The total RNA expression of cells (transcriptome) can be profiled using microarrays. Such an approach has been applied with success to cell culture models, animal models and ex vivo lung and other samples to study susceptibility and progression of lung disease. Much progress has been made in studying the transcriptomics of lung disease, including (from many available studies) the identification of molecular gene expression signatures of smoking and its effects in the lung,75 lung cancer and its recurrence,76–78 and progression of COPD.79–81

Transcriptomic responses to air pollutants in the lungs

Specific patterns of transcriptional profile have been identified in in vitro and in vivo studies of gene expression responses to exposure to diesel exhaust, urban air pollution, cigarette smoke and other complex mixtures.82 The use of animal models or in vitro cell culture systems has been the most common study designs for transcriptional profiling.

Particulate matter

Mouse models have been useful in vivo exposure models to study particulate matter. In a murine model of asthma, ovalbumin-sensitized mice were exposed to ambient particulate matter.83 This exposure induced eosinophilic and neutrophilic responses in the airways, and increased bronchial hyper-responsiveness in the mice. By day 4 of exposure to particulate matter, microarrays detected 436 differentially expressed genes, with activated pathways including innate immunity, allergic inflammation, chemotaxis, complement system, inflammation, host defence and signal transduction. This study of gene expression profiling confirmed the pro-inflammatory and allergic effects of particulate matter exposure, implicating this air pollutant exposure to susceptibility and severity of asthma.83 Similar results have been found by other groups using allergic mouse models.84 In a mouse model using microarrays to measure gene expression, inhalation of ultrafine carbon particles resulted in an initial (4-h) upregulation of heat shock proteins, and subsequently (24-h) upregulation of host defence and adhesion pathways (osteopontin, lipocalin-2 and galectin-3).85

However, not all preclinical murine models have shown widespread gene expression responses to particulate matter. A study of transgenic mice constitutively overexpressing the pro-inflammatory cytokine, TNF-α, showed upregulation of gene expression of CYP1A1, endothelin-1 and metallothionein-II (indicating changes in metabolism, endothelial function and metal toxicity, respectively).86 In contrast, there was no major change detected using microarrays in genes expressed upon particulate matter, between the TNF and the wild-type mice. To explain the lack of effect on global gene expression, the investigators concluded that physiological responses were more likely to be important, or that relatively greater statistical power, site-specific lung measurement (vs global measurement) and more sensitive animal models may have yielded more divergent gene expression patterns.86

Several studies have measured gene expression using human cells lines. BEAS-2B, a respiratory epithelial cell line, was exposed to PM2.5.87 Microarray profiling showed that there was upregulation of inflammatory cytokines and mediators, including IL-1R1 and IL-6R, and activation of the STAT3 pathway. In another study, BEAS-2B cells exposed to DEP showed upregulation of inflammatory, chemokine, metabolizing enzyme and stress response pathways by microarray analysis of gene expression.88 A549 cells, which are alveolar type-II cancer cells, similarly express a number of pathways (e.g. oxidative stress) in response to DEP.89 To date, no studies have measured gene expression globally in human exposure studies; this remains a potential area of discovery.

Ozone

Microarray profiling of ozone exposure has been performed in rodent models. Rat lungs exposed to toxic concentrations of ozone (2 ppm (parts per million) and 5 ppm) for 2 h display an acute response transcriptome that is predominated by fatty acid metabolism (resulting from lipid ozonation), cell proliferation, stress response and adhesion molecules.90 A study of ozone exposure in toll-like receptor 4 (tlr4) mutant and wild-type mice found that an intact TLR4 pathway was needed for ozone to induce inflammation in the lungs.91 Gene expression profiling showed that heat shock proteins were major mediators of the effects of ozone exposure, in the TLR4 downstream pathway.

Biological studies of respiratory cell lines and animal models have shown that there are distinct gene expression signatures arising from exposure to particulate matter or ozone (summarized in Table 1). Longer term exposure studies of transcriptional responses are now required, as are further human studies.

Table 1.  Examples of pathways with upregulated gene expression, detected by microarray profiling in animal and human models, in response to air pollutant exposure (particulate matter or ozone)
PathwaysExamples of genes
Particulate matter 
 InflammationInterleukin (IL)-6, IL-8, IL-6R, IL-1R, IL-2087,88
Chemokine (C-X-C motif) ligand 1 and 288 IL-4, IL-5, eotaxin83
 ApoptosisCaspase-3, caspase-1287
 Innate immunityToll-like receptor (TLR)587
 Oxidative stressHeme oxygenase-189
 Metabolic enzymesCytochrome P450 CYP1A1 and CYP1B188
 Stress response and host defenceHeat shock proteins85,88
Osteopontin, lipocalin-285
 Cell adhesionGalectin-385
Ozone 
 Fatty acid metabolismFatty acid amide hydrolase, phospholipase A2–activating protein90
 Cell proliferationG1/S specific cyclin (cyclin E1)90
 Stress responseTranscription factor AP190
 Adhesion moleculesL-selectin90
 Host defenceToll-like receptor 4, heat shock proteins91
 Metal bindingMetallothioneins91

MicroRNAome

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Regulatory functions of microRNA

MicroRNA (miRNA) are short non-coding RNA that function as gene expression regulators. miRNA are capable of switching genes ‘off’ in response to endogenous and exogenous signals, including adverse environmental challenges such as air pollution and cigarette smoke. The mechanisms for gene silencing include translational repression, mRNA degradation or mRNA cleavage.92 Over 1000 human miRNA have been discovered, and these are able to regulate over 60% of human genes. miRNA are important mediators of normal biological functions of development and function, and their dysregulation has been implicated in a wide range of diseases including malignancies, heart disease, inflammation and lung disease.93

MicroRNA response to air pollutant exposure

Exposure to air pollution has the potential to dysregulate normal miRNA functioning.94 In a study of miRNA profiling, human bronchial epithelial cells (HBEC), obtained from non-smoking adult donors, were exposed to DEP in vitro.95 This exposure altered the expression of 313 miRNA in the HBEC, with several miRNA up- or downregulated by >1.5-fold. Putative gene targets were identified for three of the miRNA—miR-513a-5p, miR-494 and miR-96—and these genes were involved in pathways enriched for inflammation (IL-8 and CXCR4 signalling) and tumourigenesis. These results indicate that bronchial epithelial cells, as the first structural cells that encounter inhaled air pollutants, respond not just with altered gene expression but also with dysregulated miRNA expression, which may affect cellular homeostasis. Dysregulation of miRNA expression may also extend systemically, as shown in peripheral blood mononuclear cells in steel plant workers exposed to particulate matter.96

Exposure to other environmental toxins (besides traffic-related air pollutants) has been studied in relation to miRNA. The expression of miRNA was found to be altered in a mouse model exposed to the toxin hexahydro-1,3,5-trinitro-1,2,3-triazine (RDX), an environmental contaminant from various sources, including explosives.97 A total of 113 miRNA were altered in the brain and liver of mice as a result of RDX exposure. The miRNA expression was tissue-specific, with differences between the brain and the liver expression. Although lung expression of miRNA was not studied, this study supports the notion that environmental toxins affect the expression of miRNA and that this expression is specific to the tissue type. For lung-origin cells, exposure of A549 type II alveolar cells to gaseous formaldehyde98 and exposure of mouse lungs to titanium dioxide nanoparticles99 altered expression of miRNA, as determined by miRNA microarray profiling.

Potential similarities between miRNA responses to air pollution and cigarette smoke exposure

Cigarette smoke is a major contributor to miRNA dysregulation, which has been implicated in the development of lung cancer, COPD and airway inflammation. Several studies have reported found differentially expressed miRNA associated with lung cancer samples.100–102 In addition, Schembri et al. performed a functional study to analyse the effect of cigarette smoke on miRNA involved in lung tumourigenesis.103 They identified 28 miRNA that were differentially expressed in bronchial epithelial cells of current or former smokers, versus never smokers, and that the expression of a number of these miRNA were inversely correlated with expression of their predicted target mRNA in vitro. Selecting one of these miRNA for further functional studies, they were able to demonstrate that altering miR-218 expression in vitro, with anti- or pre-miR oligos in bronchial epithelial cells exposed to cigarette smoke, changed the expression of the mRNA targets of miR-218.103 MiRNA have also been implicated in COPD development and severity, with decreased expression of two miRNA (let-7c and miR-125b) in induced sputum of patients with COPD compared with healthy smokers.104 As miRNA can be detected in body fluids such as sputum, bronchial washings, serum and blood, miRNA would potentially be excellent biomarkers for the early detection of lung disease and a promising portal for novel therapy. There are similarities between toxic exposure to cigarette smoke and air pollutants, providing synergy for the development of models of genomic responses to these environmental agents.

Collectively, these studies highlight that miRNA are likely to be key regulators of the toxicogenomic response to environmental challenges in the lungs and deserve further study in relation to the adverse health effects of air pollutants. Functional studies of the importance of miRNA in the lung are needed, including studies in exposure cohorts, in vitro studies of altered miRNA expression, target mRNA validation, and comparisons of miRNA expression in patients with lung disease and healthy controls.94

CONCLUSIONS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

Air pollution remains an important public health problem worldwide. The ‘omics’ revolution has provided tools to systematically study biological responses to air pollution exposure in the lungs. There is much promise from characterizing how air pollution interacts with genetic variants (genome), methylation and other gene regulation mechanisms such as histone and chromatin changes (epigenome), gene expression (transcriptome), and expression of miRNA that regulate genes (miRNAome; Fig. 1). Evidence to date from profiling indicates that genomic alterations play an important role in mediating pathogenic mechanisms involved in response to air pollutants. For example, the biology of the inflammatory response to air pollutants is regulated by SNP in the TNF gene in subjects exposed to ozone, histone modifications that increase COX-2 expression in bronchial epithelial cells exposed to DEP, and dysregulated miRNA expression in bronchial epithelial cells challenged with DEP.

image

Figure 1. ‘Omics’ approaches to studying the respiratory effects of air pollutant exposure. Response to air pollutant exposure in the lungs can be studied systematically at the genomic level. The response may vary based on SNP in DNA comprising the genome, differences in DNA methylation and other epigenomic changes, mRNA expression in the transcriptome and altered miRNA expression in the miRNAome. These genomic changes, arising as a result of gene-environment interaction, influence protein expression and function in a number of biological pathways, and ultimately influence cellular function in response to air pollution.

Download figure to PowerPoint

Using microarrays and emerging methods such as next generation sequencing will enable investigators to profile these changes across the entire genome. Although relatively few studies have been performed to date, those that have been undertaken, in a number of experimental settings, demonstrate the potential importance of genomic responses in regulating the respiratory responses to air pollutants.

There is now a strong imperative to use the best air pollution models in vitro and in vivo, combined with genomics, to identify the key pathways involved in mechanisms of lung injury from air pollution, and then undertake further functional validation of the most important genes and mediators. While there is clearly an important public health initiative to contain rising levels of air pollution, it is also important that strategies be developed to minimize the damaging effects of air pollutant exposure on the lung, especially for patients with chronic lung disease and for people at risk of future lung disease after exposure to air pollution. The study of genomic responses will improve understanding of disease mechanisms and enable future clinical testing of interventions against the toxic effects of air pollutants.

ACKNOWLEDGEMENTS

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES

This work was supported by Asthma, Allergy and Inflammation Research (J.W.H.), National Health and Medical Research Council of Australia (NHMRC) Career Development Award (I.Y.), NHMRC Practitioner Fellowship (K.F.), NHMRC Biomedical Scholarship (S.S.), Cancer Council Queensland Senior Research Fellowship (K.F.), Australian Lung Foundation/Boehringer Ingelheim COPD Research Fellowship (I.Y.), and project grants from NHMRC, Queensland Health Smart State, The Prince Charles Hospital Foundation and Asthma Foundation of Queensland.

REFERENCES

  1. Top of page
  2. ABSTRACT
  3. INTRODUCTION
  4. GENOME AND GENE–ENVIRONMENT INTERACTION
  5. EPIGENOME
  6. TRANSCRIPTOME
  7. MicroRNAome
  8. CONCLUSIONS
  9. ACKNOWLEDGEMENTS
  10. REFERENCES