|Home | About | Journals | Submit | Contact Us | Français|
Ambient air pollution, including particulate matter, and gaseous pollutants, are important environmental exposures that adversely affect human health. Because of their heritable and reversible nature, epigenetic modifications provide a plausible link between environment and alterations in gene expression that might lead to disease. Epidemiologic evidence supports that environmental exposures in childhood impact susceptibility to disease later in life, supporting that epigenetic changes can impact ongoing development and promote disease long after the environmental exposure has ceased. Indeed, allergic disorders often have their roots in early childhood and early exposure to particulate matter has been strongly associated with the subsequent development of asthma. The purpose of this review is to summarize recent findings on the genetic and epigenetic regulation of responses to ambient air pollutants, specifically respirable particulate matter, and their association with the development of allergic disorders. Understanding these epigenetic biomarkers and how they integrate with genetic influences to translate the biologic impact of particulate exposure is critical to developing novel preventative and therapeutic strategies for allergic disorders.
Environmental air pollution is a serious public health concern throughout the world. Numerous studies have demonstrated a strong link between exposure to ambient air pollution and human morbidity and mortality1-3. Major air pollutants include particulate matter, ozone and nitrogen dioxide. The regulation of cellular responses in response to pollutant gases including ozone and sulphur dioxide have been previously reviewed4-6. In this review, we will focus on particulate matter (PM). Chronic exposure to particulate matter is associated with increased risk of cardiovascular disorders, chronic obstructive pulmonary disease, cancer, neurological disease and asthma7-10. Increased asthma incidence and hospitalization rates for asthma, and higher rates of allergic sensitization have been observed in children exposed to high levels of PM11. Pollutant-induced inflammation and oxidative stress have been implicated as contributors to these negative health effects12, but the mechanisms remain unclear.
Ambient PM is a complex mixture of chemicals and particles, of which the largest single source is traffic related and derived from diesel exhaust. Diesel exhaust particles (DEP) consist of an elemental carbon core with a large surface area to which hundreds of chemicals and transition metals are attached. Particulate matter is classified as coarse (PM10: particles of 10μm or less in aerodynamic diameter), fine (PM2.5: 2.5-0.1μm) or ultrafine (PM0.1: <0.1μm). The majority of DEP are fine or ultrafine particles. Studies have revealed profound adjuvant effects of these particles on the development and intensity of allergic inflammation, which is regulated via genetic and epigenetic mechanisms. In this review, we will summarize recent genetic studies identifying candidate genes conferring susceptibility to allergic diseases in response to particulate matter especially DEP. Furthermore, we will review epigenetic changes induced by DEP exposure, the epigenetic regulation of biological pathways mediating responses to DEP, and their association with clinical outcomes. Finally, we will discuss and propose potential approaches for future genetic and epigenetic studies related to PM exposure.
Several lines of evidence have shown that PM can act both on the upper and lower airways to initiate and exacerbate cellular inflammation. There are increased levels of proinflammatory cytokines and chemokines, and increased numbers of neutrophils in the bronchoalveolar lavage fluid of healthy individuals exposed to DEP, and in nasal washes from subjects following nasal provocation challenge with DEP13-15. Acute DEP exposure induced NF-κB-related inflammatory and cytokine gene expression including TNFa, IL8 and IL616.
DEP can also exacerbate in vivo allergic responses. When house dust mite (HDM) was co-administered with DEP into dust mite sensitive human subjects, nasal histamine levels increased three-fold compared to HDM alone and only 20% of the amount of intranasal dust mite normally required resulted in a symptomatic response17. Furthermore, DEP dramatically increased the production of allergen-specific IgE18. In mice, DEP delivered intranasally or intratracheally with allergen results in increased airway inflammation, goblet cell hyperplasia and airway hyperresponsiveness (AHR)19, 20. In summary, DEPs can promote sensitization and aggravate airway responses.
DEP can target multiple cell types including human B cells21, eosinophils22, pulmonary alveolar macrophages23, 24, airway epithelial cells and endothelial cells25. DEP can act directly on cells or indirectly by alternatively activating macrophages or dendritic cells (DC). Indeed, mice instilled with DEP increased the number of activated DC in the bronchoalveolar lavage and in the lung26; and DEP-exposed DCs induced Th2 polarization27, 28.
The mechanism by which DEP exposure modulates innate and adaptive immune responses remains unclear. Many of the effects of DEPs are attributed to chemicals in the DEP, including metals, organic compounds and biological fractions29. A hierarchical oxidative stress model has been postulated such that low dose DEP exposure leads to the formation of reactive oxygen species (ROS), which activate an antioxidant response and upregulate detoxifying enzymes12. A higher dose exposure would overwhelm the cell’s cytoprotective system and activate NF-κB/AP-1 mediated signaling, which results in increased expression of proinflammatory cytokines (TNF-α, IL-8 and IL-6) and adhesion molecules30. This enhanced inflammation would lead to additional generation of ROS. In support of this model, in vitro studies demonstrated that macrophages, neutrophils, eosinophils, and epithelial cells generated ROS after stimulation by DEP or its chemical constituents11. Furthermore, the proinflammatory and adjuvant effects of DEP in animal and cell culture models were neutralized in part by thiol antioxidants31, 32. Thus, it has been proposed that oxidative stress-induced inflammation is one of the contributors to PM-induced respiratory diseases33. The chemical composition of DEP also affects the immune responses that are induced. Samples with different organic content induce distinct inflammatory allergic phenotypes34, 35, possibly through activating distinct transcriptional regulatory pathways (e.g., AP-1 or NF-κB). Another factor that modifies the biological effects of DEP is particle size. Smaller particles (PM0.1) are more likely to penetrate deeper into the respiratory tract and translocate to extrapulmonary organs through blood stream36.
There is large variation between individuals in their response to PM exposure and genetic association studies have compared the adverse effect of PM between subjects with specific genotypes in biologically relevant candidate genes. Genes encoding enzymes involved in oxidative stress including GSTM1, GSTP1 and NQO1 were identified as genes conferring genetic susceptibility to ozone exposure with the outcome of asthma4. Polymorphisms in these genes were also identified as risk factors for allergic diseases in response to DEP. In the Cincinnati Childhood Allergy and Air Pollution study (CCAAPS) birth cohort of 570 children, DEP exposure was estimated using a land-use regression mode and it was discovered that high DEP exposure during infancy conferred increased risk for wheezing among GSTP1 Val105 (rs947894) carriers37, 38. In another study, individuals who were sensitive to ragweed and carried the GSTM1 null or the GSTP1 I105 genotype showed the greatest nasal allergic responses after challenge with ragweed and DEP39. These individuals also had greatest response to ragweed after second hand smoke exposure40. When using proximity to high traffic roads as a surrogate measure of exposure, a significant interaction was observed between EPHX1, the microsomal epoxide hydrolase involved in xenobiotic metabolism, GSTP1 genotype, and distance to road. Children carrying EPHX1 genotypes with high predicted enzymatic activity, who also carried the GSTP1 Val/Val genotype and had high traffic exposure were at the highest risk of asthma41.
Genes in inflammatory pathways have also been implicated as genes relevant to the host response to PM. In the Prevention and Incidence of Asthma and Mite Allergy (PIAMA) birth cohort, individuals with increased levels of PM2.5 in conjunction with at least one copy of the TLR2 rs4696480 A allele or any of the TLR4 genotypes (rs2770150 TC, rs10759931 GG, rs6478317 GG, rs10759932 CT or CC, and rs1927911 TT) had increased risk of doctor-diagnosed asthma from birth up to 8 years of age42. School children homozygous for TGF-β1 rs1800469T (associated with increased gene expression) had a significantly increased risk of asthma and this association was modified by residential proximity to freeway43.
One major limitation of candidate gene approach is that it relies on previous known functions of genes and pathways44. Also, these studies often have small sample sizes and examine a limited number of variants. In contrast, the genome-wide association method (GWA) is unbiased and examines SNPs across the genome. However, most common variants identified by GWAS confer small incremental risk individually or in combination and explain only a small portion of heritability. Factors contributing to this “missing” heritability have been suggested and include variants with small effects yet to be found, rare variants that are poorly detected with current assays, structural variants such as copy number variants that are not well studied, low power to detect gene-gene interaction, and environment factors that remain undefined 44. Many currently available genome-wide arrays have limited coverage, and pre-designed SNP chips might not contain SNPs in coding or regulatory regions or rare SNPs with significant effects. Thus, combining GWAS and candidate gene approaches may generate more comprehensive and consistent results45.
Epigenetics refers to information heritable through cell division other than the DNA sequence itself. During development, cell identities are maintained through epigenetic mechanisms. The developmental state of a cell can be reprogrammed by either somatic cell nuclear transfer or by specific genes in combination, suggesting that the information specifying cell identity is not only heritable, but ultimately reprogrammable46,47. Epigenetic mechanisms include DNA methylation, post-translational histone modification, histone variation, chromatin remodeling and non-coding RNA (Figure 1). DNA methylation refers to the covalent addition of a methyl group to a cytosine (C) residue, and promoter methylation is correlated with gene expression silencing. It occurs mostly in the context of CG dinucleotide, however non-CG methylation has been recently described at CHG and CHH sites (where H is A, C or T)48, 49. In mammalian cells, DNA methyltransferase 1 (DNMT1) methylates hemi-methylated parent-daughter duplexes during DNA replication. Methyltransferases DNMT3a and DNMT3b de novo methylate DNA, setting up the methyl-CG landscape of the genome early in development. The mechanism for demethylating DNA remains controversial50. Recently, Ten-Eleven-Translocation (TET) proteins and 5-hydroxymethylcytosine (5hmC) were proposed as a plausible mechanism as the generation of 5hmC by TET proteins is associated with large scale erasure of methylation in primordial germ cells and early embryo51, 52. Hypermethylation of CpG sites located at gene promoters are commonly associated with transcriptional silencing, possibly through precluding the binding of transcription factors to their target sequences and increasing affinity to methylated DNA binding proteins that further recruit other epigenetic modifiers and corepressors53. However, it has been recently demonstrated that during cellular differentiation, reprogramming and cancer development, DNA methylation alterations happen not only at promoters, but also at other regions that are far away from transcription start sites54-56. The biological significance of these DNA methylation changes remains unclear.
Modification of histones and other epigenetic mechanisms can regulate gene expression conjointly with DNA methylation (Figure 1). The carboxyl ends of histones have specific amino acids that are sensitive to posttranslational modifications including methylation, acetylation, phosphorylation, sumoylation and ubiquitination. Open chromatin or euchromatin is characterized by high levels of acetylation and trimethylation of Histone 3 lysine 4 (H3K4me3), lysine 36 (H3K36me3) and lysine 79 (H3K79me3). On the other hand, compact chromatin or heterochromatin is characterized by low levels of acetylation and high levels of Histone 3 lysine 9 (H3K9me3), lysine 27(H3K27me3) and Histone 4 lysine 20 methylation (H4K20me3)57. Different histone variants are found at specific genomic locations and implicated in gene expression regulation and epigenetic memory of cellular states58, 59. Several groups of large chromatin remodeling complexes (i.e. SWI/SNF, ISWI, CHD and INO80) are known to move, destabilize, eject or restructure nucleosomes and modulate gene expression60.
Heterochromatin formation can also be mediated through interactions between long intergenic non-coding RNA (LincRNAs) and chromatin remodeling complexes, resulting in gene expression silencing. Small noncoding RNAs (miRNAs) directly target mRNAs and regulate cognate target gene expression61, 62. All epigenetic mechanisms closely intertwine with each other and regulate the gene expression program of a cell to ensure its proper cellular function.
The epigenome is an important target of environment-induced modification and may serve as an interface between inherited genome and dynamic environment. Heavy metals, dietary choices and maternal behavior alter DNA methylation and chromatin, and some of these epigenetic changes can be transmitted to subsequent generations6. For example, transmission of asthma risk after maternal exposure to environmental tobacco smoke (ETS) may continue across two generations63.
Epidemiological studies have linked early environmental exposures, such as in utero starvation and exposure to mold, to long-term health consequences64-66, supporting the theory of the early origin of adult diseases. Epigenetic alterations can occur prenatally, perinatally, and later in life during developmental stages with unique susceptibility to the effects of environmental exposures. Children with in utero exposure to maternal smoking exposure had different repeat element and gene specific methylation compared to unexposed children67, 68. A post-weaning diet lacking in methyl-donors (folic acid, Vitamin B12, methionine and choline) can affect the CpG methylation status of specific genes, and the resulting methylation changes persisted following a return to normal diet, consistent with the stable and heritable nature of DNA methylation69.
The impact of environment on the genome may be best illustrated by twin studies. Remarkably, human monozygotic twins are indistinguishable in their overall content and genomic distribution of global 5mC and histone acetylation during the early years of life (3 years old), whereas older monozygous twins (~50 years old) exhibited remarkable differences, which correlated with differences in gene expression70, highlighting the impact environment has on epigenome.
Multiple studies have revealed the effects of PM on the epigenome including human, in vitro, and animal studies (Table 1 and Figure 1). In the umbilical blood DNA of subjects from Columbia Center for Children’s Environmental Health study, the promoter of ACSL3 gene was hypermethylated and this correlated with increased maternal polyaromatic hydrocarbons (PAHs) exposure71. Furthermore, the increased DNA methylation correlated with silencing of ACSL3, which encodes an isozyme of the long-chain fatty-acid-coenzyme A ligase family and plays a key role in lipid biosynthesis and fat acid degradation. Exposing human airway epithelial cell lines to PM10 or DEP induced a global increase in histone H4 acetylation72, 73. Increased histone H4 acetylation at the promoters of IL8 and COX2 was associated with increased expression of these two genes. PM10-enhanced H4 acetylation was mediated in part by oxidative stress as thiol antioxidant inhibition by N-acetyl-L-cysteine ameliorated the effects of PM treatment72. DEP exposure also caused alteration of miRNA expression in human airway epithelial cells grown at air-liquid interface74; 197 of the 313 detectable miRNAs (62.9%) were either up-or down-regulated ≥ 1.5 fold, including many miRNAs associated with responses in inflammatory pathways. Mir-222 and mir-21, which have been implicated in redox signaling and inflammation, were upregulated after exposure to PM10. 75
CD4+CD25+Foxp3+ Treg cells regulate effector T cells and have been shown to limit allergic inflammation in response to inhaled allergen76, 77. Foxp3 is the major transcription factor that mediates Treg development and its expression is epigenetically regulated78, 79. In a recent study of children with asthma exposed to high vs. low levels of ambient air pollution, hypermethylation of the FOXP3 locus was observed in the peripheral blood Tregs of asthmatic children exposed to high levels of ambient air pollution compared to the low exposure group. Furthermore, FOXP3 hypermethylation was associated with decreased Treg function and increased asthma morbidity80.
Human population studies have shown that methylation of repeat elements LINE1 and Alu are negatively associated with PM exposure81, 82(Table 1). However, in another study of 718 elderly participants in the Boston area Normative Aging Study, the methylation level of LINE1 but not Alu, decreased after recent exposure to black carbon and PM2.583 ; and an increase of LINE1 and Alu methylation in association with higher exposure to PAH was observed in the peripheral blood lymphocyte DNA from non-smoking coke-oven workers compared to matched control individuals84. Thus, although methylation changes are taking place, the specific changes in methylation may be dependent on factors such as timing and length of exposure, co-exposures, the route(s) of exposure, the target tissue/cell, and host development and genetics. These contributory factors and the mechanisms need to be better clarified.
Animal studies have been useful to evaluate epigenetic modifications following pollutant exposure. Male C57BL/CBA mice breathing ambient air near two steel mills and a major highway for 10 weeks or 16 weeks demonstrated global hypermethylation in sperm DNA compared to those breathing filtered air, and this epigenetic change persisted following removal from the environmental exposure85. In BALB/c mice sensitized by Aspergillus fumigatus, chronic inhaled exposure to DEP promoted Th2 differentiation through hypermethylation of the IFNγ promoter and hypomethylation of IL4 promoter in CD4+ T cells86. Furthermore, altered methylation of promoters of both genes was significantly correlated with changes in IgE levels.
In summary, exposure to DEP or DEP chemicals results in epigenetic changes at repeat elements and specific candidate genes, some of which are strongly correlated with alteration in expression levels. The significance of these epigenetic changes as biomarkers for DEP exposure and the underlying mechanisms require further investigations. Given that the epigenetic changes caused by DEP are inheritable during cell division and can potentially transfer to future generations, it is very important to study the role of epigenetics in the association of early life DEP exposure to allergic disorders later in life.
In this review, we discussed the genetic and epigenetic influences on the cellular response to PM, DEP in particular. Although progress has been made, many questions remain (Table 2).
Genome wide association studies (GWAS) represent a powerful tool for investigating the genetics of common diseases and have successfully identified several genetic variants associated with exposure to air particulates. However, findings from these studies are largely limited to common or imputed single nucleotide polymorphisms, which likely confer small increases in risk87. Rarer variants present in less than 5% of population that are poorly detected by available genotyping arrays possibly have larger effects. In order to more fully define the genetic influence that modifies responses to environmental pollutants, longitudinal studies designed to assess early life exposure are required. Non-SNP variants including copy number variation and copy neutral variation need to be included in future studies. Identification of rare variants by exome or whole genome sequencing, use of shared datasets provided by the 1000 Genomes Project, and the design of meta-studies with consistently well-defined phenotypes across large population sets will greatly advance the development of genetic studies.
How PM causes epigenetic changes is an intriguing question. Genetic variations in glutathione-s-transferases (GSTP1 and GSTM1) have been identified as potential modifiers of the responses to PM. GSTs represent a major group of detoxification enzymes, and their substrate glutathione (GSH) is produced through transsulfuration pathway from cysteine, which is connected with the methylation cycle and the folate cycle (Figure 2). In conditions with high oxidative stress and low GSH, such as autism88 and severe asthma89, the enhanced need to synthesize GSH could potentially impair biosynthesis of SAM (S-adenoylmethionine, the major methyl donor for most methyltransferases that modify DNA, RNA, histones and other proteins) and perturb DNA methylation90. In support of this, diets lacking sources of methyl-groups (folic acid, B12, choline and betaine) result in global and gene-specific DNA hypomethylation91. DEP exposure itself can contribute to oxidative stress; in vitro exposure to DEP induced a decrease in the cellular GSH:GSSG ratio31. Thus in early life, this may be a mechanism by which PM could affect global and specific gene methylation. One might postulate that PM may influence the methylation cycle by depleting glutathione, thus promoting epigenetic changes. It is also been proposed that oxidative stress can affect global epigenetic patterns by interfering with metabolism, and thereby activating TET and other chromatin modifiers92. Moreover, histone acetylation changes induced by DEP in vitro can be countered in part by the addition of antioxidants72, thus, there may be an opportunity for targeted intervention in high risk infants aimed at disease prevention.
An epigenetic origin of disease has been suggested based on the observations linking early life exposure with later disease. Birth cohort studies have revealed that early life exposure (before age 1) to DEP leads to persistent wheezing37, 38. It will be necessary to determine the developmental origins of this susceptibility in order to develop interventions aimed at prevention of disease during this critical window in early life. Epigenetic patterns associated with specific environmental pollutants may be useful biomarkers to determine the long-term effects of early exposure to environmental pollutants on human health, although many questions remain to be answered6. Epigenetic modifications may turn on or off gene expression inappropriately; or may mask or unmask DNA sequence variation that has disease consequences. Studies in twins and healthy controls identified genetic variations that were associated with nearby differences in DNA methylation93. In the same study, the heritability estimate in DNA methylation at 96 CpG sites out of 431 CpG sites examined was as high as 94%. Therefore, integrating epigenetics into genetics studies to study their influences on responses to environmental pollutants will greatly help to identify risk factors and better understand the pathogenesis of diseases associated with environmental pollution.
To study epigenetic makers associated with environmental pollutants, the most optimal epigenetic methodology has to be utilized. Research to date has been utilizing methylation of repeat elements as indicators of global methylation. Whether methylation of repeat elements can be used as markers for pathological diseases or specific exposure requires further investigation because the methylation status of these elements is influenced by inter-individual variability and gender specific variation94, 95. With the development of newer technologies, more discovery-oriented approaches can be utilized to identify epigenetic responses to air pollutants. Statistical tools and concepts are being developed to analyze, interpret and compare population-level epigenetic data. Another limitation of environmental epigenomic studies is sample choice. The most common samples studied include blood, buccal cells, cord blood and skin cells. Whether these are legitimate surrogate tissues for airway cells needs to be carefully evaluated. Continued research is necessary to understand how DEP is recognized in the airways and what cellular pathways initiate and mediate innate and adaptive immune responses to DEP that promote adverse health effects.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1This work was supported by National Institutes of Health grants U19A170235 (GKKH) and R01HL098134 (GKKH).