Abstract
Rationale: Epigenetic changes may play a role in the occurrence of asthma-related phenotypes.
Objectives: To identify epigenetic marks in terms of DNA methylation of asthma-related phenotypes in childhood, and to assess the effect of prenatal exposures and genetic variation on these epigenetic marks.
Methods: Data came from two cohorts embedded in the Infancia y Medio Ambiente (INMA) Project: Menorca (n = 122) and Sabadell (n = 236). Wheezing phenotypes were defined at age 4–6 years. Cytosine-guanine (CpG) dinucleotide site DNA methylation differences associated with wheezing phenotypes were screened in children of the Menorca study using the Illumina GoldenGate Panel I. Findings were validated and replicated using pyrosequencing. Information on maternal smoking and folate supplement use was obtained through questionnaires. Dichlorodiphenyldichloroethylene was measured in cord blood or maternal serum. Genotypes were extracted from genome-wide data.
Measurement and Main Results: Screening identified lower DNA methylation at a CpG site in the arachidonate 12-lipoxygenase (ALOX12) gene in children having persistent wheezing compared with those never wheezed (P = 0.003). DNA hypomethylation at ALOX12 loci was associated with higher risk of persistent wheezing in the Menorca study (odds ratio per 1% methylation decrease, 1.13; 95% confidence interval, 0.99–1.29; P = 0.077) and in the Sabadell study (odds ratio, 1.16; 95% confidence interval, 1.03–1.37; P = 0.017). Higher levels of prenatal dichlorodiphenyldichloroethylene were associated with DNA hypomethylation of ALOX12 in the Menorca study (P = 0.033), but not in the Sabadell study (P = 0.377). ALOX12 DNA methylation was strongly determined by underlying genetic polymorphisms.
Conclusions: DNA methylation of ALOX12 may be an epigenetic biomarker for the risk of asthma-related phenotypes.
Environmental factors have been associated with the development of different asthma-related phenotypes. Epigenetic changes including DNA methylation might link environmental exposures with changes in gene expression.
Data from two independent pregnancy cohorts showed that DNA hypomethylation at ALOX12 cytosine-guanine (CpG) dinucleotide sites, which was in part genetically determined, was associated with a higher risk of persistent wheezing in childhood. The DNA methylation status of the ALOX12 gene could be an epigenetic biomarker of susceptibility to asthma-related phenotypes in childhood.
Developmental environment and underlying genotype can influence epigenetic variation, including DNA methylation, histone modifications, and microRNA expression, which can alter genome transcription and predetermine life-long phenotypic consequences (1–5). Asthma is a complex disease determined by a combination of genetic and nongenetic factors. Developmental environment and genetic variants have been reported to increase risk of asthma-related phenotypes in childhood (6–8), but most of the susceptibility still remains unexplained. Epigenetic mechanisms have emerged as a factor that also may contribute to the risk of asthma-related phenotypes and mediate environmental effects (9, 10).
Recent studies show that prenatal exposures linked with an increased risk of childhood asthma-related phenotypes, such as maternal diet (11), smoking (12), and stress (13), and pollutants including polycyclic aromatic hydrocarbons (14) and persistent organic pollutants (POPs), such as dichlorodiphenyldichloroethylene (DDE) (15, 16), are also associated with gene-specific and global DNA methylation changes (17–24). The identification of epigenetic marks associated with risk of asthma-related phenotypes in childhood could provide more insight into the complex pathophysiology of asthma and help develop early interventions to prevent disease.
In this study we aimed to identify epigenetic marks in terms of DNA methylation of asthma-related phenotypes in childhood, and to assess the effect of prenatal exposures (including maternal smoking, maternal folate supplement use, and DDE) and underlying genetic variation on these epigenetic marks. We first conducted a discovery screening using the Illumina GoldenGate Cancer Panel I array with subsequent validation by pyrosequencing using data from a pregnancy cohort embedded in the Infancia y Medio Ambiente (INMA) Project (the INMA Menorca cohort). Second, a replication study by pyrosequencing was performed using data from an independent cohort of the INMA Project (the INMA Sabadell cohort). Some of the results of these studies have been previously reported in the form of an abstract at the International Society for Environmental Epidemiology 2011 Annual Conference (25).
Written informed consent was obtained from all participants and the studies were approved by the Clinical Research Ethical Committee of the Municipal Institute of Health Care, Barcelona, Spain.
Participants in the first study were members of the INMA Menorca cohort, a population-based pregnancy cohort included in the INMA Project (26). A total of 482 children (94% of those eligible) were enrolled in 1997. For the purposes of this study, a subset of 141 children was selected for DNA methylation analysis chosen to ensure a balanced distribution of prenatal exposures, and finally 122 with complete data at age 6 years were included in the present study. This subsample was representative of the whole cohort population in terms of percentage of children with asthma-related phenotypes, and main characteristics (see Table E1 in the online supplement).
Wheezing was described on each interviewer-led annual questionnaire based on the validated ISAAC questionnaire (27) as “whistling or wheezing from the chest, but not noisy breathing from the nose.” One or more episodes of wheezing over 12 months constituted wheezing during any given year. At age 6 years, children were assigned to four categories according to their history of wheeze: (1) those who had no recorded wheeze during the first 3 years of life and had no wheeze at 4 and 6 years of life (never wheeze); (2) those who had recorded wheeze during the first 3 years of life but no wheeze at 4 or 6 years (transient wheeze); (3) those who did not record wheeze during the first 3 years of life but had wheeze at 4 or 6 years (late-onset wheeze); and (4) those who had at least one wheeze episode in the first 3 years of life and had wheeze at 4 or 6 years of life or those doctor-diagnosed with asthma by age 6 years (persistent wheeze) (28).
DNA was obtained from whole blood collected at 4 years of age using the Chemagic Magnetic Separation Module I (Chemagen Biopolymer-Technologie AG, Baesweiler, Germany) at the Barcelona-CRG Node of the Spanish Genotyping Centre. For the screening of cytosine-guanine (CpG) dinucleotide sites differently methylated according to children wheezing phenotypes, the Illumina GoldenGate Methylation Cancer Panel I array was used (Illumina, San Diego, CA). Briefly, genomic DNA (600 ng) was treated with sodium bisulfite using the EZ-96 DNA Methylation Kit (Zymo Research, Irvine, CA). Methylation levels were analyzed in two independent Sentrix Array Matrixes. Three control samples were included in the array: fully methylated genomic DNA and partially methylated genomic DNA from Jurkat cells (New England Biolabs Inc., Ipswich, MA), and unmethylated DNA obtained by whole genome amplification using GenomiPhi DNA amplification kit (GE Healthcare, Piscataway, NJ). These control samples and two additional DNA samples from the study were bisulphite converted twice and included in both arrays used in the study. Control samples were used to estimate the array discrimination threshold using the total deviation index (29). The minimal detectable methylation difference observed within array was approximately 7% and between arrays was approximately 12%. See the online supplement for additional details. Lists of CpGs with differential methylation in “Beta” values at a P value less than 0.01, a pragmatic threshold for selecting CpG sites for further study, were generated for children classified as having transient and persistent wheezing using children never wheezing as the reference group (see Tables E2 and E3). Additionally, results were prioritized based on differential methylation in “Beta” values above 12% (Table 1), the minimal detectable methylation difference between arrays, and on biologic functional information. Because of its relation to airway inflammation (30–32) the arachidonate 12-lipoxygenase (ALOX12) gene was selected as a candidate for subsequent validation and replication.
| Never Wheezing (n = 61) | Persistent Wheezing (n = 17) | |||||||||
| Gene | Name | Chr | CpG ID | Position | Distance to TSS | CpG Island | Mean (SE) | Mean (SE) | Diff. | P Value |
| ZNF264 | Zinc finger protein 264 | 19 | ZNF264_P397_F | 62394284 | −397 | Y | 78.3 (4.6) | 64.6 (5.3) | −13.7 | 1.4 × 10−5 |
| ALOX12 | Arachidonate 12-lipoxygenase | 17 | ALOX12_E85_R | 6840213 | 85 | Y | 34.4 (3.8) | 21.5 (5.2) | −12.9 | 0.003 |
| EPO | Erythropoietin | 7 | EPO_P162_R | 100156197 | −162 | Y | 15.4 (9.4) | 30.8 (10.4) | 15.3 | 0.006 |
| PDGFB | Platelet-derived growth factor beta polypeptide | 22 | PDGFB_E25_R | 37970911 | 25 | Y | 16.5 (19.8) | 38 (20.8) | 21.4 | 0.007 |
DNA methylation in the first exon of ALOX12 was validated using the PSQ 96MA Pyrosequencing System (Biotage AB, Uppsala, Sweden). The pyrosequencing assay was designed to validate the CpG site (E85) previously identified on the array and three flanking CpG sites (CpG1, CpG3, and CpG4) located in the exon 1 of ALOX12 (Figure 1). See the online supplement for additional details.
Figure 1. Location of cytosine-guanine (CpG) dinucleotide sites analyzed in ALOX12 gene. The CpGs in ALOX12 gene included in the array were CpG P223 and CpG E85, located in the promoter region and in exon 1, respectively, both in a 1.6-kb CpG island that spans from the promoter region up to the second exon of the gene. CpG E85 plus three flanking CpGs (CpG1, CpG3, and CpG4) in exon 1 were validated by pyrosequencing. Location of the polymerase chain reaction and the sequencing primers used in the pyrosequencing assay (arrows) and the pyrosequenced fragment (in bold italics) are shown.
[More] [Minimize]Participants in the second study were members of the INMA Sabadell cohort, a pregnancy cohort established in Sabadell (Catalonia, Spain) between 2004 and 2006 (26). A total of 657 women were enrolled at the first trimester of pregnancy and 622 (94%) were followed until the child's birth. Overall, 236 (45%) out 519 children with available DNA extracted from whole cord blood were included in the replication study. Compared with excluded participants, mothers of children included in the present analysis had higher educational level but did not differ in other main baseline characteristics (see Table E1).
Information on wheeze episodes, defined as in the discovery study, was obtained from interviewer-led annual questionnaires from birth until age 4 years.
DNA samples were extracted from whole cord blood using the same protocol as in the discovery study. In addition, a subset of subjects were contacted again at 4 years and total blood was obtained for DNA extraction (n = 19). The same pyrosequencing assay described previously was used to determine DNA methylation levels in the four CpG sites located in the exon 1 of ALOX12.
DDE was measured by gas chromatography in cord blood serum of children participating in the Menorca study and in maternal blood serum extracted during the first trimester of pregnancy of women participating in the Sabadell study (33). Information on maternal smoking habits and use of folate supplements during pregnancy was obtained through questionnaires in both studies.
Genotypes in single-nucleotide polymorphisms (SNPs) in ALOX12 gene and surrounding region (5 kb upstream and downstream) were extracted from genome-wide genotypic data. Genotyping was performed using the Human Omni1 array (Illumina). See the online supplement for additional details. Genotyping data was available in 174 children participating in the Menorca study (85 of them with methylation data) and in 396 children participating in the Sabadell study (204 of them with methylation data).
In the discovery round, to identify CpGs that showed significant differences in methylation levels among children having had transient or persistent wheezing by using children never wheeze as the reference group, data were analyzed by modeling the methylation at each individual CpG site as quantified by the Illumina “Beta” value by mixed linear regression models. Wheezing phenotypes and potential covariates (including child sex, gestational age, maternal prepregnancy body mass index, child z-score body mass index at age 6 yr, and percentage of eosinophils at age 4 yr) were included as fixed coefficients and array effects as random coefficients. Results were summarized as means and standard errors.
The association between wheezing phenotypes and DNA methylation levels at ALOX12 CpG sites obtained by pyrosequencing was also analyzed using linear mixed models. Here, individual random effects were specified to take into account the correlation between replicates from the same individual and the information of the working plate. Similar statistical models were performed to analyze the association between prenatal exposures (i.e., maternal smoking and use of folate supplements, and prenatal DDE levels) and genetic polymorphisms with DNA methylation levels of ALOX12.
Multivariable logistic regression models were performed to evaluate the association between asthma-related phenotypes (dependent variable) and methylation status of the selected candidate gene (independent variable). Odds ratios (OR) are presented for a decrease in 1% of DNA methylation levels.
Mixed linear models were used to evaluate the association between genetic polymorphism and DNA methylation at ALOX12 CpG sites. Multivariable logistic regression models were used to explore the association between genetic polymorphisms in ALOX12 and risk of wheezing phenotypes.
Analyses were conducted using STATA 10.1 statistical software (Stata Corporation, College Station, TX) and R statistical package version 2.9.1 (R Foundation for Statistical Computing, Vienna, Austria).
Among the 122 subjects included in the discovery round, 61 (50%) reported never wheezing at age 6 years; 41 (34%) were classified as having transient wheezing; 3 (2%) as having late-onset wheezing; and 17 (14%) as having persistent wheezing. Main characteristics of study population are shown in Table E1.
Mixed linear regression models showed differences at a P value less than 0.01 in DNA methylation levels at age 4 years in 54 CpG sites comparing children having persistent wheezing with those who never wheezed at age 6 years (see Table E2). Among these CpGs, four showed a difference above 12% (Table 1). Because of its biologic function the CpG located in ALOX12 gene was selected for subsequent validation and replication of DNA methylation differences between children persistent wheezing and those who never wheezed.
Overall, no CpG site showed DNA methylation differences above 12% between children classified as having transient wheezing compared with those never wheezing (see Table E3).
In the Menorca study, pyrosequencing results featured lower methylation levels at ALOX12 CpG sites than the Illumina Golden Gate, but the overall pattern of group differences with lower methylation level at age 4 years in children having persistent wheezing at age 6 years compared with children never wheezing was found (Table 2).
| Menorca Cohort* | Sabadell Cohort† | |||||||
| Never/Persistent | Never Wheezing | Persistent Wheezing | Never/Persistent | Never Wheezing | Persistent Wheezing | |||
| CpG | N | Mean (SE) | Mean (SE) | P Value | N | Mean (SE) | Mean (SE) | P Value |
| CpG1 | 60/17 | 14.7 (0.9) | 12.2 (2.3) | 0.114 | 109/37 | 5.5 (0.1) | 4.9 (0.2) | 0.168 |
| CpG2 (E85) | 60/17 | 17.3 (1.6) | 14.6 (4.1) | 0.212 | 109/37 | 6.3 (0.2) | 5.1 (0.3) | 0.028 |
| CpG3 | 59/17 | 21.2 (1.8) | 18.1 (4.7) | 0.177 | 107/34 | 9.6 (0.3) | 8.2 (0.6) | 0.078 |
| CpG4 | 59/14 | 12.3 (0.9) | 10.6 (2.7) | 0.333 | 108/33 | 4.4 (0.1) | 3.3 (0.2) | 0.019 |
In the Sabadell study, similar to other ALOX12 CpGs DNA methylation levels in ALOX12 CpG2 (E85) site in cord blood were overall lower compared with levels detected in blood at 4 years of age (see Table E4), but still the methylation levels of each CpG site measured at these different ages showed a strong correlation ranging from 0.70–0.99. DNA hypomethylation at ALOX12 CpG sites in children having persistent wheezing at age 4 years compared with children who never wheezed could be replicated (Table 2).
In the Menorca study results from the array showed that lower levels of DNA methylation at ALOX12 CpG2 (E85) at age 4 years were significantly associated with a higher risk of persistent wheezing at age 6 years (OR per a decrease in 1% of methylation, 1.07; 95% confidence interval [CI], 1.02–1.12) (Table 3). Although the association with the four pyrosequenced CpGs in ALOX12 was not statistically significant, the direction and magnitude of the association was essentially the same (Table 3). Accordingly, in the Sabadell study, DNA hypomethylation at ALOX12 CpG2 (E85) at birth was associated with higher risk of having persistent wheezing at age 4 years (OR per a decrease in 1% of methylation, 1.19; 95% CI, 1.03–1.37; P = 0.017) (Table 3).
| Menorca Cohort* | Sabadell Cohort† | ||||||||
| Method | CpG | N (Never/Persistent) | Odds Ratio | 95% Confidence Interval | P Value | N (Never/Persistent) | Odds Ratio | 95% Confidence Interval | P Value |
| Array | CpG2 (E85) | 61/17 | 1.07 | 1.02–1.12 | 0.006 | — | — | — | |
| Pyrosequencing | CpG1 | 60/17 | 1.13 | 0.99–1.29 | 0.077 | 109/37 | 1.16 | 0.97–1.38 | 0.111 |
| Pyrosequencing | CpG2 (E85) | 60/17 | 1.07 | 0.97–1.17 | 0.165 | 109/37 | 1.19 | 1.03–1.37 | 0.017 |
| Pyrosequencing | CpG3 | 59/17 | 1.07 | 0.98–1.16 | 0.134 | 107/36 | 1.10 | 0.99–1.22 | 0.062 |
| Pyrosequencing | CpG4 | 59/14 | 1.08 | 0.95–1.22 | 0.254 | 109/34 | 1.25 | 1.04–1.50 | 0.015 |
In the Menorca study, higher levels of DDE in cord blood were associated with a decrease in DNA methylation at ALOX12 CpG2 (E85) (upper vs. lower tertile of DDE; P = 0.033) (Table 4). However, in the Sabadell study prenatal exposure to DDE, measured as DDE levels in maternal serum in first trimester of pregnancy, was not significantly associated with DNA methylation at ALOX12 CpG2 (E85) site in cord blood (Table 4).
| Menorca Cohort* | Sabadell Cohort† | |||||||
| N (%) | Mean | SE | P Value | N (%) | Mean | SE | P Value | |
| DDE tertile‡ | ||||||||
| Low | 40 (34.8) | 19.3 | 1.9 | 80 (33.3) | 6.4 | 0.2 | ||
| Medium | 37 (32.2) | 17.3 | 1.6 | 0.244 | 80 (33.3) | 6.6 | 0.2 | 0.750 |
| High | 38 (33) | 15.8 | 1.9 | 0.033 | 80 (33.3) | 5.9 | 0.2 | 0.377 |
| Maternal smoking | ||||||||
| No | 85 (69.7) | 17.6 | 1.1 | 205 (86.5) | 6.2 | 0.1 | ||
| Yes | 37 (30.3) | 18.2 | 2 | 0.688 | 32 (13.5) | 6.6 | 0.4 | |
| Folate use | ||||||||
| No | 48 (39.3) | 18.3 | 1.7 | 14 (5.8) | 6.6 | 0.8 | ||
| Yes | 74 (60.7) | 17.5 | 1.2 | 0.571 | 227 (94.2) | 6.2 | 0.1 | 0.687 |
We found no evidence of an association between maternal smoking or folate supplement use during pregnancy and DNA methylation at ALOX12 CpG sites (Table 4).
We first tested if SNPs in ALOX12 gene and surrounding regions (50 kb upstream and downstream) were associated with DNA methylation at CpG sites of ALOX12 (see Tables E5 and E6). We found that genotypes of the rs312466 (A/G), a haplotype-tagging SNP that lies within the CpG island (see Figure E1), were significant associated with DNA methylation at ALOX12. Children carrying the minor allele (A) showed lower DNA methylation levels at ALOX12 CpG2 (E85) site compared with those carrying the major allele (G) (Figure 2). Other SNPs in the region, especially those tagged by rs312466, were also associated with DNA methylation levels of ALOX12. Moreover, in the Sabadell study, children carrying the minor allele of the rs312466 tended to have higher risk of persistent wheezing in childhood than those homozygote for the major allele (GA/AA vs. GG genotype; OR, 2.62; 95% CI, 1.02–6.75; P = 0.036). Although statistically nonsignificant, similar results were found in the Menorca cohort (GA/AA vs. GG genotype; OR for persistent wheezing, 3.12; 95% CI, 0.37–26.41; P = 0.274). The results of the pooled analysis of the Menorca and Sabadell studies showed a statistically significant association between rs312466 genotypes and persistent wheezing at age 4 years (GA/AA vs. GG genotype; OR for persistent wheezing, 2.49; 95% CI, 1.19–5.23; P = 0.011).
Figure 2. DNA methylation levels in cytosine-guanine (CpG) dinucleotide E85 site at ALOX12 gene by genotype for haplotype-tagging single-nucleotide polymorphism rs312466. Bars represent means and standard errors. *P value under an additive model. 1Adjusted for child sex, gestational age, maternal prepregnancy body mass index, eosinophil count at age 4 years, and child body mass index z-score at age 6 years. 2Adjusted for child sex, gestational age, and maternal prepregnancy body mass index.
[More] [Minimize]The discovery study in the Menorca cohort identified lower DNA methylation levels at a CpG site in ALOX12 at age 4 years between children having persistent wheezing and those who never wheezed at age 6 years. This finding was validated by pyrosenquencing and replicated in an independent pregnancy cohort (the Sabadell cohort) in which DNA methylation was assessed at birth and wheezing phenotypes at age 4 years. In the Menorca study, higher DDE levels in cord blood were associated with DNA hypomethylation of ALOX12 gene at age 4 years; however, we failed to find an association between prenatal DDE levels (measured in maternal blood serum during pregnancy) and DNA methylation of ALOX12 at birth in the Sabadell cohort. We did not find evidence of an association between maternal smoking or folate use during pregnancy with DNA methylation of ALOX12. In contrast, genetic polymorphisms within or near ALOX12 were strongly associated with DNA methylation and with risk of persistent wheezing in childhood.
DNA hypomethylation at ALOX12 CpG sites in relation to persistent wheezing was found as early as birth, which suggests that methylation status of ALOX12 gene could be a potential early epigenetic biomarker of susceptibility to asthma-related phenotypes in childhood. Key elements in asthma pathophysiology include the chronic airway inflammation and remodeling associated with exaggerated Th2 over Th1 responses to allergic and nonallergic stimuli. ALOX12 gene (also known as 12S-LOX or platelet-type lipoxygenase 12) encodes for the 12-LOX enzyme that is involved in the metabolism of arachidonic acid leading to the generation of the inflammatory eicosanoids 12(S)-hydroxyeicosatetraenoic acid (12[S]-HETE) (34). Mice 12-LOX and 15-LOX are referred as 12/15-LOX because of their high homology and the fact that they can form 12(S)-HETE and 15(S)-HETE from arachidonic acid (35). The 12/15-LOX is constitutively expressed at high levels in immature red blood cells, eosinophils, and airway epithelial cells (30). Interestingly, experimental studies have shown that 12/15-LOX knockout mice were almost completely protected from the induction of specific IgE antibodies after airway exposure to allergens; were at least fourfold more protected from the induction of specific IgG1 antibodies and from allergen-induced increases in expression of Th2 cytokines (IL-4 and IL-13); and had a lower number of eosinophils in bronchoalveolar lavage fluid (31). Moreover, increased levels of 15(S)-HETE are found in bronchoalveolar lavage fluid from persistently wheezing children (32) and adults with asthma (36). Finally, genetic polymorphisms in ALOX12 gene and other family members have been investigated in relation to asthma with inconclusive results (37, 38).
Environmental insults during development could leave permanent epigenomic marks that might lead to increased risk of developing asthma phenotypes later in life. Current evidence of effects of early life environmental exposures on asthma risk in childhood mediated by epigenetic changes has so far been limited to air pollutants, such as tobacco smoke (17, 18) and polycyclic aromatic hydrocarbons (24). We previously reported that DDE levels at birth but not DDE measured at 4 years of age increased the risk of physician-diagnosed asthma and of persistent wheezing at age 6 years (16), which supports the hypothesis of developmental programming of asthma by DDE. Here, results from the Menorca study suggest a link between higher prenatal DDE levels to DNA hypomethylation of ALOX12 gene; however, this was not replicated in an independent cohort. There may be several possible explanations for this. First, assessment of DDE exposure differed between both cohorts in regards to the cohort set up, developmental timing at assessment, and biologic matrices of measurement. There is a 7-year gap between the two cohort recruitments, which may explain why levels of DDE tended to be lower in the Sabadell than in the Menorca cohort. In addition, temporal and local differences in agriculture and industrial activities could determine differences in exposure to a diverse mixture of POPs between cohorts. Second, different timing and biologic matrix of exposure assessment could also explain lack of replication. Also, differences between cohorts in relation to maternal age and maternal obesity (see Table E1) may account for different levels of exposure in utero. Third, DNA methylation levels of ALOX12 were measured at different ages and using different tissues between both cohorts that can limit direct comparisons. In the Sabadell cohort we found lower methylation levels in ALOX12 CpGs in cord blood compared with levels in blood at age 4 years, although a strong correlation was found. Because cell count information was not available, we cannot rule out whether the change in absolute values between these two time points is caused by an increase in ALOX12 methylation over time or a change in cell count between cord blood and whole blood at age 4 years. Further research is needed to confirm these findings and elucidate potential biologic mechanisms involved in DDE impact on DNA methylation.
It is also known that underlying genotype influences epigenetic variation. Thus, we evaluated the possible cis genetic effects on ALOX12 DNA methylation. We found that methylation differences at ALOX12 CpG sites were highly attributable to genetic variation acting in cis, which is consistent with previous studies (39). In addition, we found some evidence that genetic variants in ALOX12 were associated with the risk of persistent wheezing. These results strengthen the idea of using epigenomic studies to investigate genetic predispositions that could exert their action through epigenetic mechanisms. In addition, assessment of associations between genetic variants (which are not related to confounding factors, and not altered by disease processes and thus subject to reverse causation) and epigenetic patterns may elucidate causal pathways in epidemiologic studies (40).
The strengths of this study are the prospective nature and the population-based cohort design. In addition, assessment of DNA methylation changes was performed before the occurrence of asthma-related phenotypes that precludes reverse causation. The study has some limitations. First, characterization of asthma-related phenotypes was based on questionnaires used to ascertain whether subjects have had symptoms of asthma (i.e., wheezing) or have ever received a diagnosis of asthma from a physician. However, children with persistent wheezing at age 6 years had significantly reduced lung function, which is consistent with the substantial deficits in lung function reported in older children with asthma (28). Further research is warranted to assess if DNA hypomethylation of ALOX12 is also observed among children with physician-diagnosed asthma or reduced lung function. Second, dramatic differences in methylation levels are reported in cancerous versus noncancerous tissue but not in other complex diseases and phenotypes, where methylation at any given CpG island or specific CpG sites in affected versus unaffected individuals may vary by less than 10% (3). Smaller effect sizes may be a general phenomenon because they have been reported in other human epigenetic studies involving famine exposure, neuropsychiatric disorders, and assisted reproductive technologies (3, 41, 42). Moreover, for some genes, evidence exists that a small change in the level of DNA methylation, especially in the lower range, can dramatically alter gene expression levels (43, 44). Third, we did not assess if changes in DNA methylation levels at ALOX12 CpG sites affected gene transcription; thus, evidence from human- or animal-based designs is necessary to demonstrate that demethylation of the ALOX12 enhances gene expression. Fourth, we used whole blood cells as a surrogate to predict the pathophysiologic changes in the target tissues (airway and the immune cells). However, previous studies suggest that whole blood is a good surrogate because of its high levels of T cells, which are important producers of cytokines and other asthma mediators (45). Finally, we used the HumanMethylation Cancer Panel I (Illumina) for screening of DNA regions differently methylated. The CpG sites included in this array are based on their different methylation in tumor tissues and cancer processes, which in some aspects can be related to fetal development. However, this array was not designed for epigenome-wide analyses, and methylation changes at other loci and genes related with asthma may have been overlooked. Future larger prospective cohort studies and genome-wide epigenetic screening are essential to unravel the role of epigenetic mechanisms in programming susceptibility to asthma and asthma-related phenotypes and in mediating effects of environmental early life exposures.
In conclusion, DNA hypomethylation at ALOX12 CpG sites, which was highly determined by underlying genetic variation, was associated with higher risk of persistent wheezing in childhood. DNA methylation status of ALOX12 gene could be an epigenetic biomarker of susceptibility to asthma-related phenotypes in childhood. Whether effects of prenatal exposure to DDE on risk of asthma-related symptoms in childhood are mediated by epigenetic changes deserves further investigation.
The authors acknowledge all teachers and parents of the children from Menorca Island for patiently answering the questionnaires; the nurses and administrative personal from the Primary Health Care Center of Maó for administrative, technical, and material support; and the CeGen-ISCIII Barcelona-CRG Node for technical assistance. The authors are also grateful to Silvia Fochs, Anna Sànchez, Maribel López, and Nuria Pey for their assistance in contacting the families and administering the questionnaires in the Sabadell study and all the participants for their generous collaboration.
| 1. | Dolinoy DC, Weidman JR, Jirtle RL. Epigenetic gene regulation: linking early developmental environment to adult disease. Reprod Toxicol 2007;23:297–307. |
| 2. | Gluckman PD, Hanson MA, Cooper C, Thornburg KL. Effect of in utero and early-life conditions on adult health and disease. N Engl J Med 2008;359:61–73. |
| 3. | Heijmans BT, Tobi EW, Stein AD, Putter H, Blauw GJ, Susser ES, Slagboom PE, Lumey LH. Persistent epigenetic differences associated with prenatal exposure to famine in humans. Proc Natl Acad Sci USA 2008;105:17046–17049. |
| 4. | Bell JT, Pai AA, Pickrell JK, Gaffney DJ, Pique-Regi R, Degner JF, Gilad Y, Pritchard JK. DNA methylation patterns associate with genetic and gene expression variation in HapMap cell lines. Genome Biol 2011;12:R10. |
| 5. | Gibbs JR, van der Brug MP, Hernandez DG, Traynor BJ, Nalls MA, Lai SL, Arepalli S, Dillman A, Rafferty IP, Troncoso J, et al.. Abundant quantitative trait loci exist for DNA methylation and gene expression in human brain. PLoS Genet 2010;6:e1000952. |
| 6. | Eder W, Ege MJ, von Mutius E. The asthma epidemic. N Engl J Med 2006;355:2226–2235. |
| 7. | Holloway JW, Yang IA, Holgate ST. Genetics of allergic disease. J Allergy Clin Immunol 2010;125(Suppl 2):S81–S94. |
| 8. | Ober C, Yao TC. The genetics of asthma and allergic disease: a 21st century perspective. Immunol Rev 2011;242:10–30. |
| 9. | Miller RL, Ho SM. Environmental epigenetics and asthma: current concepts and call for studies. Am J Respir Crit Care Med 2008;177:567–573. |
| 10. | Ho SM. Environmental epigenetics of asthma: an update. J Allergy Clin Immunol 2010;126:453–465. |
| 11. | Håberg SE, London SJ, Nafstad P, Nilsen RM, Ueland PM, Vollset SE, Nystad W. Maternal folate levels in pregnancy and asthma in children at age 3 years. J Allergy Clin Immunol 2011;127:262–264, 264.e1. |
| 12. | Hylkema MN, Blacquière MJ. Intrauterine effects of maternal smoking on sensitization, asthma, and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2009;6:660–662. |
| 13. | Cookson H, Granell R, Joinson C, Ben-Shlomo Y, Henderson AJ. Mothers’ anxiety during pregnancy is associated with asthma in their children. J Allergy Clin Immunol 2009;123:847–853. |
| 14. | Miller RL, Garfinkel R, Horton M, Camann D, Perera FP, Whyatt RM, Kinney PL. Polycyclic aromatic hydrocarbons, environmental tobacco smoke, and respiratory symptoms in an inner-city birth cohort. Chest 2004;126:1071–1078. |
| 15. | Karmaus W, Kuehr J, Kruse H. Infections and atopic disorders in childhood and organochlorine exposure. Arch Environ Health 2001;56:485–492. |
| 16. | Sunyer J, Torrent M, Muñoz-Ortiz L, Ribas-Fitó N, Carrizo D, Grimalt J, Antó JM, Cullinan P. Prenatal dichlorodiphenyldichloroethylene (DDE) and asthma in children. Environ Health Perspect 2005;113:1787–1790. |
| 17. | Li YF, Langholz B, Salam MT, Gilliland FD. Maternal and grandmaternal smoking patterns are associated with early childhood asthma. Chest 2005;127:1232–1241. |
| 18. | Breton CV, Byun HM, Wenten M, Pan F, Yang A, Gilliland FD. Prenatal tobacco smoke exposure affects global and gene-specific DNA methylation. Am J Respir Crit Care Med 2009;180:462–467. |
| 19. | Hoyo C, Murtha AP, Schildkraut JM, Jirtle RL, Demark-Wahnefried W, Forman MR, Iversen ES, Kurtzberg J, Overcash F, Huang Z, et al.. Methylation variation at IGF2 differentially methylated regions and maternal folic acid use before and during pregnancy. Epigenetics 2011;6:928–936. |
| 20. | Gheorghe CP, Goyal R, Mittal A, Longo LD. Gene expression in the placenta: maternal stress and epigenetic responses. Int J Dev Biol 2010;54:507–523. |
| 21. | Rusiecki JA, Baccarelli A, Bollati V, Tarantini L, Moore LE, Bonefeld-Jorgensen EC. Global DNA hypomethylation is associated with high serum-persistent organic pollutants in Greenlandic Inuit. Environ Health Perspect 2008;116:1547–1552. |
| 22. | Shutoh Y, Takeda M, Ohtsuka R, Haishima A, Yamaguchi S, Fujie H, Komatsu Y, Maita K, Harada T. Low dose effects of dichlorodiphenyltrichloroethane (DDT) on gene transcription and DNA methylation in the hypothalamus of young male rats: implication of hormesis-like effects. J Toxicol Sci 2009;34:469–482. |
| 23. | Kim KY, Kim DS, Lee SK, Lee IK, Kang JH, Chang YS, Jacobs DR, Steffes M, Lee DH. Association of low-dose exposure to persistent organic pollutants with global DNA hypomethylation in healthy Koreans. Environ Health Perspect 2010;118:370–374. |
| 24. | Perera F, Tang WY, Herbstman J, Tang D, Levin L, Miller R, Ho SM. Relation of DNA methylation of 5′-CpG island of ACSL3 to transplacental exposure to airborne polycyclic aromatic hydrocarbons and childhood asthma. PLoS ONE 2009;4:e4488. |
| 25. | Abstracts of the 23rd Annual Conference of the International Society of Environmental Epidemiology (ISEE). September 13–16, 2011, Barcelona, Spain. Environ Health Perspect (In press) |
| 26. | Guxens M, Ballester F, Espada M, Fernández MF, Grimalt JO, Ibarluzea J, Olea N, Rebagliato M, Tardón A, Torrent M, et al.. The INMA–INfancia y Medio Ambiente–(Environment and Childhood) Project. Int J Epidemiol (In press) |
| 27. | Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC. The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Lancet 1998;351:1225–1232. |
| 28. | Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan WJ. Asthma and wheezing in the first six years of life. The Group Health Medical Associates. N Engl J Med 1995;332:133–138. |
| 29. | Escaramís G, Ascaso C, Carrasco JL. The Total Deviation Index estimated by tolerance intervals to evaluate the concordance of measurement devices. BMC Med Res Methodol 2010;10:31. |
| 30. | Nadel JA, Conrad DJ, Ueki IF, Schuster A, Sigal E. Immunocytochemical localization of arachidonate 15-lipoxygenase inerythrocytes, leukocytes, and airway cells. J Clin Invest 1991;87:1139–1145. |
| 31. | Hajek AR, Lindley AR, Favoreto S, Carter R, Schleimer RP, Kuperman DA. 12/15-Lipoxygenase deficiency protects mice from allergic airways inflammation and increases secretory IgA levels. J Allergy Clin Immunol 2008;122:633–9.e3. |
| 32. | Krawiec ME, Westcott JY, Chu HW, Balzar S, Trudeau JB, Schwartz LB, Wenzel SE. Persistent wheezing in very young children is associated with lower respiratory inflammation. Am J Respir Crit Care Med 2001;163:1338–1343. |
| 33. | Sala M, Ribas-Fitó N, Cardo E, de Muga ME, Marco E, Mazón C, Verdú A, Grimalt JO, Sunyer J. Levels of hexachlorobenzene and other organochlorine compounds in cord blood: exposure across placenta. Chemosphere 2001;43:895–901. |
| 34. | Pidgeon GP, Lysaght J, Krishnamoorthy S, Reynolds JV, O'Byrne K, Nie D, Honn KV. Lipoxygenase metabolism: roles in tumor progression and survival. Cancer Metastasis Rev 2007;26:503–524. |
| 35. | Yamamoto S, Takahashi Y, Hada T, Hagiya H, Suzuki H, Reddy GR, Ueda N, Arakawa T, Nakamura M, Matsuda S, et al.. Mammalian arachidonate 12-lipoxygenases. Adv Exp Med Biol 1997;400A:127–131. |
| 36. | Chu HW, Balzar S, Westcott JY, Trudeau JB, Sun Y, Conrad DJ, Wenzel SE. Expression and activation of 15-lipoxygenase pathway in severe asthma: relationship to eosinophilic phenotype and collagen deposition. Clin Exp Allergy 2002;32:1558–1565. |
| 37. | Lee HY, Kim SH, Ye YM, Choi GS, Park HS. Lack of association of ALOX12 and ALOX15 polymorphisms with aspirin-exacerbated respiratory disease in Korean patients. Ann Allergy Asthma Immunol 2009;103:84–86. |
| 38. | Tremblay K, Daley D, Chamberland A, Lemire M, Montpetit A, Laviolette M, Musk AW, James AL, Chan-Yeung M, Becker A, et al.. Genetic variation in immune signaling genes differentially expressed in asthmatic lung tissues. J Allergy Clin Immunol 2008;122:529–36.e17. |
| 39. | Pedersen KS, Bamlet WR, Oberg AL, de Andrade M, Matsumoto ME, Tang H, Thibodeau SN, Petersen GM, Wang L. Leukocyte DNA methylation signature differentiates pancreatic cancer patients from healthy controls. PLoS ONE 2011;6:e18223. |
| 40. | Relton CL, Davey Smith G. Epigenetic epidemiology of common complex disease: prospects for prediction, prevention, and treatment. PLoS Med 2010;7:e1000356. |
| 41. | Katari S, Turan N, Bibikova M, Erinle O, Chalian R, Foster M, Gaughan JP, Coutifaris C, Sapienza C. DNA methylation and gene expression differences in children conceived in vitro or in vivo. Hum Mol Genet 2009;18:3769–3778. |
| 42. | Dempster EL, Pidsley R, Schalkwyk LC, Owens S, Georgiades A, Kane F, Kalidindi S, Picchioni M, Kravariti E, Toulopoulou T, et al.. Disease-associated epigenetic changes in monozygotic twins discordant for schizophrenia and bipolar disorder. Hum Mol Genet 2011;20:4786–4796. |
| 43. | Hsieh CL. Dependence of transcriptional repression on CpG methylation density. Mol Cell Biol 1994;14:5487–5494. |
| 44. | Lillycrop KA, Phillips ES, Torrens C, Hanson MA, Jackson AA, Burdge GC. Feeding pregnant rats a protein-restricted diet persistently alters the methylation of specific cytosines in the hepatic PPAR alpha promoter of the offspring. Br J Nutr 2008;100:278–282. |
| 45. | Stock W, Hoffman R. White blood cells 1: non-malignant disorders. Lancet 2000;355:1351–1357. |
Supported by grants from the Spanish Ministry of Health (FIS-PI041436, FIS-PI041705, and FIS-PI051187); Instituto de Salud Carlos III (Red INMA G03/176 and CB06/02/0041); Ciber en Epidemiología y Salud Pública; the Generalitat de Catalunya-CIRIT (1999SGR 00241 and 2009-SGR-1502); the Spanish Ministry of Science and Innovation (SAF-2008-00357); the Seventh Framework Program of the European Commission through the ENGAGE Project; and Fundación Roger Torné. N.V. was funded by an FPI grant (BES-2009-023933) from the Spanish Ministry of Science and Innovation.
Author Contributions: E.M., M.B., and N.V. contributed equally to this work. All authors were involved in the study design and analysis, participated in the preparation and writing of the manuscript, and approved the final version. J.S. is guarantor.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Originally Published in Press as DOI: 10.1164/rccm.201105-0870OC on February 9, 2012
Author disclosures
