Bronchial Nitric Oxide Flux May Be Better Associated with Inducible Nitric Oxide Synthase Promoter Methylation

To the Editor:

We read with great interest the article by Breton and colleagues (1) reporting the association between DNA demethylation of arginase (ARG)1 and ARG2, but not inducible nitric oxide synthase (iNOS), gene promoters, and fractional exhaled nitric oxide (Fe_NO) in children with asthma. The authors are the first to link epigenetic variation in buccal DNA with increases in Fe_NO, a biomarker associated with acute bronchial hyperresponsiveness, asthma relapse, and respiratory infection (2–4). They shed new light on biological mechanisms that may underlie NO production during asthma exacerbations. We commend the authors’ collection of buccal cells as noninvasive, readily accessible cells that represent aerodigestive tract cells and can share similar patterns of gene expression with bronchial epithelial cells (5). This approach may have great advantages in longitudinal pediatric research.

Our group at the Columbia Center for Children's Environmental Health also was interested in investigating the association between buccal cell iNOS promoter demethylation and Fe_NO, and evaluated the association between alveolar and conducting airway contributions of Fe_NO with iNOS demethylation (6). Buccal cells were collected from 9- to 11-year-old urban children twice, 4–7 d apart (n = 57 subjects). Similarly, iNOS CpG-359 (corresponding to position 3 of NOS2A) was moderately methylated with a mean methylation level of 53.8% (SD 5.43; interquartile range 5.00). This intermediate level of methylation provides opportunity for changes in methylation patterns, presumably in association with environmental exposures.

We also examined the association between methylation of iNOS and Fe_NO, and its postulated components bronchial NO flux (J_NO) and alveolar NO (C_alv) with methodology that takes advantage of multiple flow rates during collection of samples (6). Using generalized estimating equations to model the repeated measures, we found that iNOS methylation was not significantly associated with Fe_NO. However, when compartmentalizing NO production into its proximal (bronchial) and distal (alveolar) airway components, iNOS methylation was associated inversely with J_NO (β = −3.68, 95% confidence interval: −6.68 to 0.67, P = 0.016), but not C_alv (P > 0.1). Previously we showed that J_NO was highly correlated with Fe_NO and may be a better indicator of seroatopy than C_alv. In contrast, C_alv may be a better indicator of current wheeze (6).

Breton and colleagues’ study and our pilot study suggest that measures of buccal cell DNA methylation may be informative biomarkers of airway inflammation in pediatric cohorts. Additionally, our pilot data suggest that iNOS demethylation may be more closely associated with proximal NO source components. Breton and colleagues may be underestimating the iNOS effect estimate because they have an NO measure from both proximal and distal sources in the lung. To strengthen these conclusions, both studies would benefit from buccal RNA expression analysis. Nonetheless, both studies reveal some new understanding on the immunopathogenesis of airway inflammation. Further studies are needed to examine the association of environmental asthma triggers with methylation patterns and clinical outcomes.