Etiology of Bronchopulmonary Dysplasia: Before Birth

Abstract

Bronchopulmonary dysplasia affects thousands of infants annually with high attendant morbidity, mortality, and healthcare costs. Respiratory outcomes for preterm infants may be widely disparate even within the same medical center for infants of similar gestational ages. Given the high estimates for the heritability for bronchopulmonary dysplasia as measured at 36 weeks of postmenstrual age, the course of disease for a particular infant is likely heavily influenced by factors that operate prenatally or in the early postnatal period. In this review we discuss the etiologies of lung disease in the premature infant, including in utero, genetic, and epigenetic factors that may influence pulmonary outcomes.

Introduction

Northway et al. first described bronchopulmonary dysplasia (BPD) in 1967, attributing the late sequelae of hyaline membrane disease to a combination of preterm delivery and assisted mechanical ventilator injury with high FiO₂.¹ In addition to postnatal lung developmental abnormalities, including asthma and an emphysema-like picture, BPD came to be recognized as a systemic syndrome with associated complications of neurodevelopmental deficits, cerebral palsy, cognitive impairments, failure to thrive, cor pulmonale, and pulmonary hypertension. Over subsequent decades, surfactant replacement, technologic advances, and improvements in delivery of care in the neonatal intensive care unit have led to improved survival and outcomes among premature infants.

Despite these improvements in care, the incidence of BPD has not substantially declined in the United States, as resuscitation of earlier gestational age infants has led to a new population of preterm infants found to have BPD.² In current practice, it is not uncommon for 23-week gestation infants to be delivered. These earlier gestational age infants treated with surfactant continue to suffer from BPD, but this new BPD differs substantially from the syndrome described by Northway et al. over 40 years ago. Infants delivered earlier than 26 weeks of gestation are born with lungs in the canalicular stage of fetal lung development. Undergoing this normally in utero growth and differentiation ex utero can lead to abnormal lung development and lung disease. Indeed, the highest rates of BPD occur in the most immature infants. In preterm infants with birth weights of 501–1,500 g, the risk of developing BPD was 30%, although this ranged between 4% and 58% across centers that contributed data in 2003.³ From a pathophysiological perspective, the new BPD lacks the prominent interstitial fibrosis, patchy alveolar over-distention and atelectasis, and excessive airway smooth muscle hypertrophy originally described by Northway et al. Instead, the histology of new BPD is more compatible with developmental arrest, inadequate alveolarization, and impaired angiogenesis.⁴

Most preterm infants with BPD and some without the formal diagnosis of BPD are at considerable risk for recurrent wheezing and an exaggerated response to respiratory infections in childhood and even asthma later in life.⁵ Although gestational age and intrauterine growth are significant predictors of the development of persistent chronic lung disease in premature infants, no single etiologic factor has been identified to predict the incidence and severity of disease. In this review, we will discuss the competing theories for the explanation of the new BPD, focusing on in utero, epigenetic, and genetic factors.

Definitions of BPD: Old Versus New

The definition of disease is critical, not only for diagnosis, but also for establishing phenotypes for epigenetic and genetic studies. Defining disease is extremely difficult for a disease process like BPD where the disease has been evolving over time with advances in care and changes in the distribution of gestational ages of premature infants. Furthermore, unlike many other chronic childhood and adult diseases, the manifestations of BPD may change very rapidly in an individual infant from requiring mechanical ventilation at birth to breathing room air at 6 months of life. Knowing the definitions of BPD leads to understanding of some of the limitations of the theories for the development of BPD.

Northway et al. characterized the old BPD as having 4 different stages based on clinical outcome, histology, and chest radiography findings.¹ Stage I occurred in the first several days postdelivery. Before the recognition and supplementation of exogenous surfactant, this stage is best described as respiratory distress syndrome (ie, hyaline membrane disease). Stages II (days 4–10) and III (days 11–20) were described as regenerative and transitional in nature, respectively. Stage IV was defined as chronic lung disease with supplemental oxygen dependence at 1 month of life. In contrast to the definition for old BPD, the definition for new BPD as developed by the National Institute of Child Health and Human Development is dependent solely on oxygen use.^6,7 Mild BPD (≥28 days supplemental oxygen use) is assigned to infants <32 weeks of gestation and breathing room air at 36 weeks postmenstrual age (PMA) or discharge, and to infants born ≥32 weeks of gestation and breathing room air by 56 days postnatal age or discharge, whichever comes first. At the same timepoints as mild BPD, infants on supplemental oxygen <30% FiO₂ are defined as having moderate BPD, and those infants with oxygen requirements ≥30% FiO₂ and/or positive pressure ventilation are defined as having severe BPD.

The use of a respiratory phenotype based solely on oxygen use may capture components of decreased pulmonary reserve in the form of inadequate alveolarization, or vascular disease in the form of oxygen-dependent pulmonary hypertension, but not necessarily small airway disease. Thus, the use of the new BPD phenotype may predispose to discovering only genetic and environmental factors that influence alveolarization or vascular disease. There are certainly other diagnostic elements that could be helpful for future definitions of BPD, such as chest radiography, and perhaps eventually tracheal aspirate or blood biomarkers and infant pulmonary function testing.

Finally, it is worth noting 2 other aspects of BPD definitions that may contribute to the limited findings of candidate gene studies in BPD to date. First, it is important to recognize that the definitions of BPD described above and others may not completely overlap. In a study of extremely low gestation births at 20 U.S. academic centers,⁸ data were collected for 9,575 infants (22–28 weeks of gestation and 401–1,500 g). Of the survivors, 68% were classified with BPD by the new severity-based system,⁹ but only 42% by traditional BPD criteria (any oxygen use at 36 weeks PMA), and 40% by a physiological definition.¹⁰ The implication of this finding is that any epigenetic or genetic study intended to replicate the findings of a previous study must utilize the exact same phenotype of BPD used by the first study, which is not always the case. Second, the use of a dichotomous (BPD versus no BPD) or trichotomous (mild BPD versus moderate versus severe) phenotype may not provide enough differentiation between subjects in a small study to detect small effects; again, infant lung function testing could provide the means for assessing the phenotype quantitatively.

Recent randomized multi-center trials designed to compare the effectiveness of different treatment strategies on neonatal outcomes have shown little substantial difference between strategies with respect to the incidence or severity of BPD. For example, the SUPPORT trial enrolled infants born at 24–28 weeks of gestation to compare intubation with surfactant dosing in the delivery room to continuous positive airway pressure initiation in the delivery room with a limited ventilator strategy.¹¹ The primary outcome was death or BPD as broadly defined as the use of supplemental oxygen at 36 weeks PMA. Neither mortality rate nor BPD differed between the 2 treatment arms. The lack of a specific therapeutic intervention or combination of interventions to alter the development of BPD suggests that genetic and developmental factors are critically important.¹²

In Utero Factors

Several in utero exposures or pressures have been associated with increased incidence of BPD in some studies. Maternal smoking, chorioamnionitis, colonization and pre-eclampsia alone or in combination have been hypothesized to predispose the infant to a more complicated postnatal course and the subsequent development of BPD.^13–15 The development of BPD may be influenced by pre-eclampsia, which occurs hours to days before delivery, as Hansen et al. reported an odds ratio of 18.7 risk for the development of BPD associated with pre-eclampsia in 107 mother/preterm infant (23–32 weeks of gestation) dyads in a multivariate risk model for which clinical chorioamnionitis, intrauterine growth restriction, and gestational age were adjusted.¹³ Pre-eclampsia and other in utero exposures may lead to impaired lung growth prior to delivery and in the early postnatal period where other postnatal environmental factors such as oxygen toxicity, barotrauma, and infection contribute to development of BPD.

The pathological mechanism of pre-eclampsia, and perhaps other in utero exposures, may be related to the promotion of an anti-angiogenic environment with parallel impairment of airway and capillary bed growth. The vascular hypothesis for the development of BPD challenges the older concept that vascular development follows the airways passively, and instead suggests that there is a dynamic feedback-driven interaction between the developing vascular bed and airways.¹⁶ Based on laboratory and clinical observations it has been postulated that disruption of angiogenesis during fetal lung development impairs alveolarization.¹⁶ Subsequent lung development in animals and humans with BPD is associated with an abnormal vascular bed with respect to numbers and distributions of vessels as well as reduced signaling molecules such as vascular endothelial growth factor (VEGF) and nitric oxide.^17,18 In these relatively small studies with many competing factors, it is possible that pre-eclampsia needs to progress to intrauterine growth restriction in order to promote later BPD.

Other in utero exposures that may lead to BPD include infection (eg, chorioamnionitis) or colonization and its subsequent inflammation. For example, Ureaplasma respiratory tract colonization, which may begin in utero and can be exacerbated postnatally, is one potential risk factor.¹⁹ Ureaplasma species are atypical mycoplasma of low virulence that are commensals in the adult female reproductive tract. Premature rupture of membranes can be the initiating pathway for invasion of the fetal respiratory tract. Other explanations for pathogen entry for which evidence exists include amniocentesis and prolonged vaginal delivery. Chorioamnionitis remains a major diagnostic problem since subclinical chorioamnionitis is common and preterm births often occur in the absence of culture-positive or polymerase chain reaction-detectable colonization of amniotic fluids.²⁰ Strategies that might eliminate Ureaplasma infection such as azithromycin treatment of the mother prior to birth and the baby after birth have failed to prevent BPD.^19–22

Pre-eclampsia and infections such as Ureaplasma are not the only in utero risk factors for the development of BPD. Prolonged maternal smoking may contribute to early lung injury during development and complicate the later development of BPD in the very low birthweight infants in particular.¹⁵ Other anomalies such as the presence of a patent ductus arteriosus have been implicated as well.²³ In contrast, maternal diabetes has not been associated with respiratory morbidity in very low birth weight (<1,500 g) infants.²⁴ Work with fetal sheep models suggests that any antenatal inflammatory insult may be detrimental to lung growth and development (eg, Escherichia coli endotoxin, resuscitation-based barotrauma, or other intrauterine pressure).^20,25

Genetic Susceptibility

Genetic factors have long been suspected to contribute to the risk for BPD. Studies of siblings and twins suggest a high concordance of BPD with shared genetic inheritance.^12,26–28 Potential modifier genes could include asthma-related genes based on family history, genes that encode the surfactant proteins,^29–32 and genes that encode for inflammatory cytokines and growth factors based on specimens from infants with BPD.^12,33 To date no genome-wide association studies for the development of BPD have been published; the literature is limited to candidate gene studies.¹² The candidate gene approach has been used most notably to examine polymorphism variants in inflammatory mediator genes, including tumor necrosis factor-α (TNF-α), mannose binding lectin (MBL2), VEGF, and a matrix metalloproteinase (MMP16).¹² However, perhaps secondary to small sample sizes, there has been an inability to replicate initial findings in subsequent independent cohorts.¹²

Another logical modifier for the development of BPD is pulmonary surfactant. Pulmonary surfactant is critical for perinatal respiratory adaptation by lowering surface tension and preventing end expiratory atelectasis. Premature infants needing respiratory support beyond the first week of life may have reduced surfactant functional activity and decreased amounts of surfactant components in their tracheal aspirates.³⁴ Persistent surfactant deficiency could thus contribute to lung injury and the development of BPD.^35,36 The hydrophilic surfactant proteins, SP-A and SP-D, also have important roles in innate immunity and in modulating local immunity. Thus, variants in the genes encoding proteins critical for surfactant function are attractive candidates for influencing risk for development of BPD. A polymorphism in intron 4 of the SFTPB gene due to variable numbers of repeat sequence motifs has been examined in several studies. Shorter variants of this polymorphism were weakly associated with risk for developing BPD in 2 association studies,^37,38 but only with BPD defined as need for supplemental oxygen at 28 days postnatally, and not at later time points or using a physiologic definition. A different SFTPB gene SNP, located in the promoter region, was linked to BPD risk in a small study from Greece using family-based association testing, and a protective allele of the SFTPA SFTPD locus was also identified.³¹ However, the number of subjects studied was small for the large number of potential associations examined, and these findings have yet to be replicated in other populations with larger sample sizes.

In addition, member A3 of the ATP binding cassette family (ABCA3) has been shown to have a critical role in surfactant metabolism, with loss-of-function mutations on both alleles causing severe lung disease. As with the surfactant proteins, the expression of ABCA3 is developmentally regulated and increases with advancing gestation. Associations of common polymorphisms in this gene with risk for BPD have not been examined. The rate of heterozygosity for a single disease-causing ABCA3 mutation (p.Glu292Val) was higher in a group of relatively older preterm (28–34 weeks of gestation) infants with respiratory distress syndrome than in a control population (3.8% versus 1%).³⁹ These observations suggest the hypothesis that the combination of heterozygosity for a loss of or reduced function ABCA3 allele and prematurity may lead to severe perinatal lung disease.³⁶ The potential contribution of rare but potentially damaging variants in genes encoding surfactant components to the development of BPD has not been examined.

Epigenetics

In addition to genetic factors, more recent research has focused on the role of epigenetic factors on the development of BPD. The term epigenetics is used to define the influence of environment on expression of the genetic material. Epigenetic events are modifications of the DNA or associated proteins that interact with the gene, short of changing the nucleotide sequence.⁴⁰ Development can be viewed as a series of epigenetic re-programmings that guide the growth and differentiation of the fetus. Epigenetic patterning is frequently inherited during somatic cell division, but also can be a function of external exposures.

The most well-studied epigenetic modification to date is methylation of the nucleotide cytosine. Cytosine guanine dinucleotides (CpG) maintain heritable methylation via DNA Methyl Transferase 1, which adds a methyl group to the daughter strand of DNA during cell division. From observations in tumor cells, DNA hypermethylation has been seen to lead to reduced gene expression and gene silencing, whereas DNA undermethylation or hypomethylation involves repeated DNA sequences, transcriptional activation, and disruption of adjacent gene expression.⁴¹ Dietary methionine and a cofactor synthesized from folic acid contribute to maintain this programming. Dietary starving or supplementation with folic acid and or methionine can profoundly affect epigenetic expression in animals and humans. Examples of epigenetic processes that explain disease in humans include cancer, Rett's syndrome, and aging. Folic acid supplementation during pregnancy is an example of an intervention in an epigenetic program to decrease the incidence of a congenital malformation, myelomeningocele. Using a quantitative high resolution mapping strategy of CpG islands in 14,496 genes in each of 12 fetal cord blood samples, Fryer et al. detected an association of folate-associated changes that correlated with plasma homocysteine levels and birth weight percentile.⁴² A subgroup of 17 CpG islands were found within genes that contribute to pathways that impact on plasma homocysteine levels or perinatal birth weight, which is a remarkable clue to developmental reprogramming leading to a profound effect on outcomes.

Another means of epigenetic programming is histone modification, which changes the structure of nucleosomes of the chromatin. Histone acetylation results in opening of the chromatin structure and increases gene expression. Histone deacetylation condenses the chromatin and silence gene expression. Altered methylation of the enzyme histone deacetylase has been associated with chronic obstructive pulmonary disease, another obstructive lung disease that is often compared with BPD. Potential therapies for modifying the balance of histone modification exist. For example, histone deacetylase inhibitors such as 4-phenylbutyrate or suberoylanilide hydroxamic acid have been studied as adjunct chemotherapeutic agents in cancers as well as modifiers of the most common mutation in cystic fibrosis, F508del.^43,44

Although epigenetic analysis is still in its infancy, there are several tantalizing leads for the role of variation in epigenetic programming leading to the development of BPD. For example, surfactant protein A is a defense molecule within the alveolus whose expression is mediated by oxygen exposure. Increasing the oxygen tension in the alveolus causes epigenetic mechanisms to recruit transcription factors that open chromatin and activate SFTPA transcription.⁴⁵ Another mechanism may be through fibrosis as interstitial lung disease has been associated with hypermethylation of gene promoters in pathways relevant to fibrosis.^46,47 Additionally, animal models are providing evidence that dietary manipulation to increase methyl donor groups induces allergic disease.^48,49 Folate, vitamin B12, choline, L-methionine, zinc and betaine supplements administered to mice reduce transcription of counter-regulatory genes in the inflammatory/allergic disease pathways.^50–52 In human neonates hypomethylation of the interferon-gamma promoter in CD4+ T cells was associated with T_h1 pathway of differentiation rather than T_h2.⁵³ Lastly, as previously mentioned, maternal asthma is often treated with corticosteroids; this may reduce histone acetyltransferase activity and recruit histone deacetylases to inhibit inflammatory gene expression.⁵⁴

Future Directions

BPD as defined by oxygen use at 36-week PMA has a strong heritable component, perhaps with an epigenetic contribution, but lung development remains sensitive to environmental pressures and likely epigenetic modifications imposed during pregnancy and later in postnatal life, particularly by the therapies clinicians are forced to use in early life. Some of the risk factors for the development of BPD are apparent and are under study, but without robust biomarkers, the study of human BPD will be confounded by the variability of inheritance, differing clinical practices in the intensive care units, and small sample size. Analysis of DNA methylation and other epigenetic signatures may provide clues to pathogenesis and possibly in the end become a therapeutic target. Ultimately, we need better markers for diagnosis, prognosis, and assessing the course of therapy for BPD.

Footnotes

Author Disclosure Statement

The authors disclose that they have no financial interests in the subject of this article.

References

Northway

Jr. , Rosan

, Porter

. Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. N Engl J Med, 1967; 276:357–368.

Northway

Jr.

Prologue: advances in bronchopulmonary dysplasia. Semin Fetal Neonatal Med, 2009; 14:331.

Van Marter

. Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med, 2009; 14:358–366.

Van Marter

. Epidemiology of bronchopulmonary dysplasia. Semin Fetal Neonatal Med, 2009; 14:358–366.

Baraldi

, Carraro

, Filippone

. Bronchopulmonary dysplasia: definitions and long-term respiratory outcome. Early Hum Dev, 2009; 85,10 Suppl:S1–S3.

Jobe

, Bancalari

. Bronchopulmonary dysplasia. Am J Respir Crit Care Med, 2001; 163:1723–1729.

Ehrenkranz

, Walsh

, Vohr

et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics, 2005; 116:1353–1360.

Stoll

, Hansen

, Bell

et al. Neonatal outcomes of extremely preterm infants from the NICHD Neonatal Research Network. Pediatrics, 2010; 126:443–456.

Jobe

, Bancalari

. Bronchopulmonary dysplasia. Am J Respir Crit Care Med, 2001; 163:1723–1729.

10.

Walsh

, Yao

, Gettner

et al. Impact of a physiologic definition on bronchopulmonary dysplasia rates. Pediatrics, 2004; 114:1305–1311.

11.

Finer

, Carlo

, Walsh

et al. Early CPAP versus surfactant in extremely preterm infants. N Engl J Med, 2010; 362:1970–1979.

12.

Lavoie

, Dube

. Genetics of bronchopulmonary dysplasia in the age of genomics. Curr Opin Pediatr, 2010; 22:134–138.

13.

Hansen

, Barnes

, Folkman

, McElrath

. Maternal preeclampsia predicts the development of bronchopulmonary dysplasia. J Pediatr, 2010; 156:532–536.

14.

Ryan

, Ahmed

, Lakshminrusimha

. Inflammatory mediators in the immunobiology of bronchopulmonary dysplasia. Clin Rev Allergy Immunol, 2008; 34:174–190.

15.

Antonucci

, Contu

, Porcella

, Atzeni

, Chiappe

. Intrauterine smoke exposure: a new risk factor for bronchopulmonary dysplasia? J Perinat Med, 2004; 32:272–277.

16.

Thebaud

, Abman

. Bronchopulmonary dysplasia: where have all the vessels gone? Roles of angiogenic growth factors in chronic lung disease. Am J Respir Crit Care Med, 2007; 175:978–985.

17.

Thebaud

, Ladha

, Michelakis

et al. Vascular endothelial growth factor gene therapy increases survival, promotes lung angiogenesis, and prevents alveolar damage in hyperoxia-induced lung injury: evidence that angiogenesis participates in alveolarization. Circulation, 2005; 112:2477–2486.

18.

Maniscalco

, Watkins

, Pryhuber

, Bhatt

, Shea

, Huyck

. Angiogenic factors and alveolar vasculature: development and alterations by injury in very premature baboons. Am J Physiol Lung Cell Mol Physiol, 2002; 282:L811–L823.

19.

Viscardi

, Hasday

. Role of Ureaplasma species in neonatal chronic lung disease: epidemiologic and experimental evidence. Pediatr Res, 2009; 65,5 Pt 2:84R–90R.

20.

Kallapur

, Jobe

. Contribution of inflammation to lung injury and development. Arch Dis Child Fetal Neonatal Ed, 2006; 91:F132–F135.

21.

Waites

, Katz

, Schelonka

. Mycoplasmas and ureaplasmas as neonatal pathogens. Clin Microbiol Rev, 2005; 18:757–789.

22.

Ballard

, Shook

, Bernard

et al. Use of azithromycin for the prevention of bronchopulmonary dysplasia in preterm infants: a randomized, double-blind, placebo controlled trial. Pediatr Pulmonol, 2011; 46:111–118.

23.

Farstad

, Bratlid

, Medbo

, Markestad

. Bronchopulmonary dysplasia - prevalence, severity and predictive factors in a national cohort of extremely premature infants. Acta Paediatr, 2011; 100:53–58.

24.

Rehan

, Moddemann

, Casiro

. Outcome of very-low-birth-weight (<1,500 grams) infants born to mothers with diabetes. Clin Pediatr (Phila), 2002; 41:481–491.

25.

Kallapur

, Ikegami

. Physiological consequences of intrauterine insults. Paediatr Respir Rev, 2006; 7:110–116.

26.

Bhandari

, Gruen

. The genetics of bronchopulmonary dysplasia. Semin Perinatol, 2006; 30:185–191.

27.

Bhandari

, Bizzarro

, Shetty

et al. Familial and genetic susceptibility to major neonatal morbidities in preterm twins. Pediatrics, 2006; 117:1901–1906.

28.

Lavoie

, Pham

, Jang

. Heritability of bronchopulmonary dysplasia, defined according to the consensus statement of the national institutes of health. Pediatrics, 2008; 122:479–485.

29.

Makri

, Hospes

, Stoll-Becker

, Borkhardt

, Gortner

. Polymorphisms of surfactant protein B encoding gene: modifiers of the course of neonatal respiratory distress syndrome? Eur J Pediatr, 2002; 161:604–608.

30.

Rova

, Haataja

, Marttila

, Ollikainen

, Tammela

, Hallman

. Data mining and multiparameter analysis of lung surfactant protein genes in bronchopulmonary dysplasia. Hum Mol Genet, 2004; 13:1095–1104.

31.

Pavlovic

, Papagaroufalis

, Xanthou

et al. Genetic variants of surfactant proteins A, B, C, and D in bronchopulmonary dysplasia. Dis Markers, 2006; 22:277–291.

32.

Garmany

, Wambach

, Heins

et al. Population and disease-based prevalence of the common mutations associated with surfactant deficiency. Pediatr Res, 2008; 63:645–649.

33.

Hayes

Jr. , Feola

, Murphy

, Shook

, Ballard

. Pathogenesis of bronchopulmonary dysplasia. Respiration, 2010; 79:425–436.

34.

Merrill

, Ballard

, Cnaan

et al. Dysfunction of pulmonary surfactant in chronically ventilated premature infants. Pediatr Res, 2004; 56:918–926.

35.

Ballard

, Merrill

, Truog

et al. Surfactant function and composition in premature infants treated with inhaled nitric oxide. Pediatrics, 2007; 120:346–353.

36.

Hamvas

, Nogee

, Wegner

et al. Inherited surfactant deficiency caused by uniparental disomy of rare mutations in the surfactant protein-B and ATP binding cassette, subfamily a, member 3 genes. J Pediatr, 2009; 155:854–859.

37.

Makri

, Hospes

, Stoll-Becker

, Borkhardt

, Gortner

. Polymorphisms of surfactant protein B encoding gene: modifiers of the course of neonatal respiratory distress syndrome? Eur J Pediatr, 2002; 161:604–608.

38.

Rova

, Haataja

, Marttila

, Ollikainen

, Tammela

, Hallman

. Data mining and multiparameter analysis of lung surfactant protein genes in bronchopulmonary dysplasia. Hum Mol Genet, 2004; 13:1095–1104.

39.

Garmany

, Wambach

, Heins

et al. Population and disease-based prevalence of the common mutations associated with surfactant deficiency. Pediatr Res, 2008; 63:645–649.

40.

Feinberg

. Genome-scale approaches to the epigenetics of common human disease. Virchows Arch, 2010; 456:13–21.

41.

Bowman

, Wright

, Davidson

, Francis

, Yang

, Fong

. Epigenomic targets for the treatment of respiratory disease. Expert Opin Ther Targets, 2009; 13:625–640.

42.

Fryer

, Emes

, Ismail

et al. Quantitative, high-resolution epigenetic profiling of CpG loci identifies associations with cord blood plasma homocysteine and birth weight in humans. Epigenetics, 2011; 6:86–94.

43.

Wang

, Venable

, LaPointe

et al. Hsp90 cochaperone Aha1 downregulation rescues misfolding of CFTR in cystic fibrosis. Cell, 2006; 127:803–815.

44.

Zeitlin

, Diener-West

, Rubenstein

, Boyle

, Lee

, Brass-Ernst

. Evidence of CFTR function in Cystic Fibrosis after systemic administration of 4-phenylbutyrate. Mol Ther, 2002; 6:119–126.

45.

Islam

, Mendelson

. Permissive effects of oxygen on cyclic AMP and interleukin-1 stimulation of surfactant protein A gene expression are mediated by epigenetic mechanisms. Mol Cell Biol, 2006; 26:2901–2912.

46.

Sanders

, Pardo

, Selman

et al. Thy-1 promoter hypermethylation: a novel epigenetic pathogenic mechanism in pulmonary fibrosis. Am J Respir Cell Mol Biol, 2008; 39:610–618.

47.

Coward

, Watts

, Feghali-Bostwick

, Jenkins

, Pang

. Repression of IP-10 by interactions between histone deacetylation and hypermethylation in idiopathic pulmonary fibrosis. Mol Cell Biol, 2010; 30:2874–2886.

48.

Hollingsworth

, Maruoka

, Boon

et al. In utero supplementation with methyl donors enhances allergic airway disease in mice. J Clin Invest, 2008; 118:3462–3469.

49.

Prescott

, Clifton

. Asthma and pregnancy: emerging evidence of epigenetic interactions in utero. Curr Opin Allergy Clin Immunol, 2009; 9:417–426.

50.

Rosenfeld

. Animal models to study environmental epigenetics. Biol Reprod, 2010; 82:473–488.

51.

Adcock

, Tsaprouni

, Bhavsar

, Ito

. Epigenetic regulation of airway inflammation. Curr Opin Immunol, 2007; 19:694–700.

52.

Hollingsworth

, Maruoka

, Boon

et al. In utero supplementation with methyl donors enhances allergic airway disease in mice. J Clin Invest, 2008; 118:3462–3469.

53.

White

, Hollams

, Yerkovich

et al. CpG methylation patterns in the IFNgamma promoter in naive T cells: variations during Th1 and Th2 differentiation and between atopics and non-atopics. Pediatr Allergy Immunol, 2006; 17:557–564.

54.

Barnes

. Histone deacetylase-2 and airway disease. Ther Adv Respir Dis, 2009; 3:235–243.