© 2004 American Thoracic Society
Cystic Fibrosis and NOS3To the Editor:The recent article by Grasemann and colleagues (1) examines NOS3, located on 7q35–7q36 downstream of the CFTR gene, as a modulator of cystic fibrosis lung disease. We would like to provide additional information on the genomic region targeted by Grasemann and colleagues and its role for cystic fibrosis disease severity. The chromosomal region 7q31–7qtel, encompassing the NOS3 gene, encodes quantitative trait loci determining growth and stature (2, 3), part of which are paternally imprinted (4). Currently, the underlying genes that determine these phenotypes are unknown and it is unclear how many loci within that region are involved (2–4). We have studied the respective chromosomal region in a set of F508del homozygous cystic fibrosis siblings for an association of marker genotype with the pulmonary and gastrointestinal disease phenotype (5). Our findings show that the disease manifestation in cystic fibrosis is modulated by loci in the partially imprinted region 3' of CFTR as a parent-of-origin effect is evident within our cystic fibrosis sibling sample (5). We have read the publication of Grasemann and colleagues (1) with curiosity, wondering whether the authors have evidence for an association of NOS3 genotypes and measures of stature and body weight that can easily be retrieved from the patients' files. A positive finding in that matter would help to unravel the relative impact of NOS3 and the aforementioned growth and stature gene(s) on the sex-stratified risk for airway colonization with Pseudomonas aeruginosa. As it is, one cannot decide whether NOS3 variants are causative modulators or just hitchhiking with phenotypically relevant polymorphisms in other genes of the linkage group. Linkage to neighboring genes is widely known as a resource for false positives in candidate gene approach–based association studies, and this has also been recognized within the cystic fibrosis field (6). We consider it vital to take the genomic context of a candidate region into account when describing the phenotype. We would like take this opportunity to join Accurso and Sontag (6) who emphasized in their editorial that a detailed phenotypic characterization facilitates the interpretation of the outcome of an association study. In this context, phenotypes considered to be of minor importance in view of the trait highlighted for a specific study might provide an opportunity to discriminate the influence of a candidate gene on the trait under consideration from a neighboring gene that might act on another facet of the multiorgan disease cystic fibrosis.
Medizinische Hochschule Hannover Hannover, Germany REFERENCES
From the Authors: We thank Mekus and Tümmler for their interest in our article on the association of NOS3 genotypes with exhaled NO (FENO) and colonization of airways in patients with cystic fibrosis (CF) (1). We agree that the genomic context of a candidate region is important for the interpretation of association studies aiming to identify modulator genes for CF phenotypes. The NOS3 gene is located in a region on chromosome 7q, downstream of the CFTR gene, which has recently been shown by Mekus and colleagues (2) to contain loci that modulate phenotypic aspects of CF including their nutritional status. The extent of pulmonary involvement and its progression is the most relevant clinical manifestation of CF. We have, therefore, focused our study on the suspected role of NOS3 in NO-related lung pathophysiology. We found that the NOS3 894T variant was associated with higher FENO and a decreased rate of airway colonization with P. aeruginosa in female patients with CF. This observation supported previous results from our group in which we had shown an association of the neuronal NOS gene (NOS1), which is localized on chromosome 12q, and CF pulmonary phenotype (3, 4). In those studies we had found a close relation of the size of a repeat polymorphism in NOS1 with FENO and colonization of airways (3, 4). Because NOS is enrolled in host defense against bacterial pathogens, this led to the hypothesis that patients harboring variants in genes encoding for constitutive NOS variants associated with low FENO have a greater risk for infections with pathogens such as P. aeruginosa.
An analysis of anthropometric data, as suggested by Mekus and colleagues, comparing the patients with CF from our cohort that are homozygous for the NOS3 894G wild-type allele and those carrying one or two copies of the 894T mutant allele revealed no differences in body weight (41.2 vs. 42.5 kg), height (150.7 vs. 153.7 cm), or body mass index (17.4 vs. 17.3 kg/m2). The proportion of subjects with a body weight for height less than the 10th percentile also was not different (13.8% vs. 19.5%, p = 0.5) among groups. The same analysis in patients homozygous for the CFTR
Children's Hospital University of Essen Essen, Germany REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
F508 mutation, as well as in our male and female subpopulation, also did not show differences in growth and stature among the NOS3 genotypes, respectively. Therefore, our data provide evidence that variants in the NOS3 gene are causative modulators of pulmonary disease but not of growth and stature in our CF population.