Published ahead of print on May 18, 2005, doi:10.1164/rccm.200407-864OC
© 2005 American Thoracic Society doi: 10.1164/rccm.200407-864OC
Genomewide Screen for Pulmonary Function in 200 Families Ascertained for AsthmaDepartment of Pulmonology, and Beatrix Children's Hospital, University Hospital, Groningen; Department of Pulmonary Rehabilitation, Beatrixoord, Haren, The Netherlands; and Center for Human Genomics, Departments of Pediatrics and Medicine, Wake Forest University School of Medicine, Winston-Salem, North Carolina Correspondence and requests for reprints should be addressed to D. S. Postma, M.D., Ph.D., Department of Pulmonology, University Hospital, Hanzeplein 3, 9731 GZ Groningen, The Netherlands. E-mail: d.s.postma{at}int.azg.nl
Changes in pulmonary function are important in determining asthma outcome. Genetic factors may influence airway obstruction in asthma. We performed a genomewide screen in 200 families of probands objectively diagnosed with asthma in the 1960s to identify chromosomal regions related to changes in pre- and postbronchodilator lung function (FEV1, VC, and FEV1%VC) and assess influences of early-life smoke exposure. Smoking (pack-years), age, sex, and height were covariates in variance component analyses. Significant evidence for linkage of pre- and postbronchodilator FEV1%VC was obtained for chromosome 2q32 (LOD,4.9, increasing to 6.03 with additional fine-mapping markers, and 3.2, respectively). Linkage existed for chromosome 5q for pre- and postbronchodilator VC (likelihood of disease [LOD], 1.8 and 2.6, respectively). Results for pre- and postbronchodilator FEV1 were less significant (LOD, 1.5 and 1.6, chromosomes 11p and 10q, respectively). Results were not affected by passive smoke exposure. There is significant evidence for linkage of FEV1%VC to chromosome 2q32 in families of probands with asthma, 35 cM proximal from linkage previously observed in families of probands with early-onset chronic obstructive pulmonary disease. Thus, there may be multiple genes on chromosome 2q that are important in determining presence and degree of airflow limitation in families ascertained for obstructive airway disease.
Key Words: asthma function genes linkage lung Asthma is a common respiratory disorder that is caused by the complex interactions between susceptibility genes and exposures to a diverse group of environmental stimuli (1). Most patients with asthma have a disease that is characterized by mild, intermittent, reversible airway obstruction. However, already in childhood (2) and in severe disease, there may exist more progressive and/or less reversible airway obstruction (3, 4). Delineating the biologic mechanisms underlying persistent changes in pulmonary function is important to guide asthma therapy (5). Therefore, it is of interest to assess which genes are associated with alterations in pulmonary function in asthma. Pulmonary function has an important role in health and disease and is an important predictor of survival in the general population (6). Family studies have suggested that 20 to 60% of the total variance in lung function may be accounted for by familial factors (712). Furthermore, Redline and coworkers (13) reported that monozygotic twins have a significantly higher concordance for spirometric indices (0.72) than nonidentical twins (0.27). However, estimates of heritability have varied largely, ranging from just over 0 to 100% (1017). Three genomewide screens in population-based families (9, 11, 18) and one in a founder population (12) have shown linkage of either FEV1, FEV1 or the ratio of FEV1 and FVC (FEV1%FVC) with chromosomal regions. Some chromosomal regions provided replication: for example, chromosome 5 (11, 12) and chromosome 4 (9, 18), a region where linkage with FEV1%FVC in families from probands with early-onset chronic obstructive pulmonary disease (COPD) also had been found (19). Finally, a genome scan of data from Chinese patients with asthma suggested several regions of linkage (e.g., chromosome 10 and 22 for FEV1 and chromosome 16 for FVC) (20). These results provide further evidence for a heritable component of lung function in the general population, families with asthma, and families with COPD (although with different or overlapping chromosomal regions). It is now accepted that asthma is caused by the interaction of genetic susceptibility coupled with exposure to environmental factors (1). Smoking is an important environmental stimulus affecting the development of both asthma and reduced lung function (2128). In utero cigarette smoke exposure and exposure in the first few months of life appear to be important risk factors for the development of atopy and asthma (2124) and with persistent reduced lung function (26). It therefore seems important to investigate whether a genesmoking interaction exists with respect to alterations in lung function in asthma. Our cohort of Dutch families with asthma presents a unique opportunity to further analyze susceptibility genes for asthma and related pulmonary function phenotypes. The 200 families were ascertained through a parent (proband) who had been objectively diagnosed with asthma in the 1960s and who was reevaluated with his or her family members between 1990 and 2000. Pulmonary function testing was performed in all family members, and data are available on passive smoke exposure in the children as well as on current and past smoking habits of each individual, which allows us to evaluate this geneenvironment interaction. The purposes of the current study are to report the results of a genomewide screen with linkage analyses for FEV1, VC, and the ratio of FEV1/VC, and to assess the effects of passive smoke exposure and individual smoking history as interacting factors.
Family Ascertainment Two hundred families (1,183 individuals) were ascertained through probands who were initially studied between 1962 and 1975 at Beatrixoord Hospital, Haren, The Netherlands, a regional referral center for patients with obstructive airway diseases (28). Patients with symptomatic asthma who were not experiencing a current asthma exacerbation were referred to this hospital and admitted for a standardized, comprehensive evaluation for asthma and atopy. At the time of initial testing, all probands had asthma symptoms, were hyperresponsive to histamine (provocation concentration causing a fall of 20% in FEV1 [PC20] histamine < 32 mg/ml), and were younger than 45 years. The families of 200 of these original probands, with their spouses, children, children's spouses, and grandchildren older than 6 years, were recruited and evaluated in the 1990s (28, 29). The Medical Ethics Committee of the University Hospital Groningen and Wake Forest University School of Medicine approved this study, and all participants provided informed consent.
Clinical Evaluation
Genotyping
Statistical Methods
Demographics The characteristics of the family members are reported for the probands, their spouses, and their first- and second-degree offspring (Table 1). As reported previously (31), all probands had initially symptomatic asthma with reversible airway obstruction accompanied by airway hyperresponsiveness. The probands showed evidence of airway obstruction with a postbronchodilator FEV1%VC of 64.7 and a reduced FEV1%predicted of 82.4. On average, spouses and offspring had normal pulmonary function.
FEV1 Heritability was estimated at 0.33 and 0.29 for pre- and postbronchodilator FEV1, respectively (Table 2). LOD score values greater than 1.5 were observed for chromosomes 11p and 10q: specifically, prebronchodilator FEV1 with an LOD of 1.50 to D11S1392 and postbronchodilator FEV1 with an LOD of 1.64 to D10S677 (Figure 1). Individual smoking was not a significant covariate in these models. Passive smoke exposure in early childhood did not explain the observed linkages: all LOD scores were less than 1.5 in the above-mentioned regions in the families with smoke exposure in the offspring. In families without passive smoke exposure, an LOD score of 1.55 was found for prebronchodilator FEV1 with chromosome 5q (marker D5S820) and 1.59 for postbronchodilator FEV1 with chromosome 3 (marker D3S2387).
VC Heritability was estimated at 0.29 and 0.32 for pre- and postbronchodilator VC, respectively. LOD score values greater than 1.5 were observed for chromosome 5q; with an LOD equal to 1.80 for the prebronchodilator VC at D5S816 and 2.58 for the postbronchodilator VC at the same map position (Figure 1, Table 2). Age, sex, and height were significant covariates in the resulting model, but individual smoking was not. No LOD scores greater than 1.5 were observed in the families without passive smoke exposure. In the families with smoke exposure in the offspring, an LOD of 1.75 for postbronchodilator VC on chromosome 7 (marker D7S2195) was observed (Table 2).
FEV1%VC
The results of this study report a high level of evidence for linkage of airway obstruction as assessed by FEV1%VC to chromosome 2q32 in families ascertained through a proband diagnosed with asthma in the 1960s. The LOD scores were 4.9 and 3.2 for pre- and postbronchodilator FEV1%VC, respectively, approximately 6 cM from the region where we have previously observed linkage with total serum IgE in the same families (27). Furthermore, the addition of seven fine-mapping markers increased the LOD score for postbronchodilator FEV1%VC to 6.03. These findings strongly support the presence of gene(s) that modulate airway obstruction in this region of chromosome 2q. Another important region for linkage of lung function variables was chromosome 5q32. LOD scores of pre- and postbronchodilator VC were 1.8 and 2.6, respectively. Interestingly, this was in the same chromosomal region we found suggestive evidence for linkage with postbronchodilator FEV1%VC (LOD score, 1.77) in families with children not exposed to cigarette smoke. Finally, the knowledge of personal smoking habits of all participants and of passive smoke exposure during early life in the offspring of the probands with asthma allowed us to evaluate the effects of this environmental factor on linkage of lung function parameters. Thus, we showed that these linkage results were not influenced by active and passive smoke exposure. We have previously reported that there is evidence of linkage for total serum IgE levels in this same region on chromosome 2q32 as for FEV1%VC. The maximum LOD score for IgE was 3.16 near marker D2S2314 (36, 37), whereas the maximum LOD score for FEV1%VC was 4.92 near marker D2S1384. This suggests that there may be gene(s) related to both atopy and pulmonary function located in this region. Interestingly, we found a significant but modest correlation (r = 0.33, p < 0.01) between FEV1%VC and total serum IgE levels in all family members, suggesting the presence of similar genes affecting IgE and FEV1%VC. Alternatively, because the explained variance is small, different genes or multiple genes may interact in various ways with each other or with environmental stimuli to regulate the expression of these two asthma phenotypes. In performing genetic studies of parameters related to pulmonary function, it is crucial to understand the ascertainment, clinical characterization, and type of family population that are studied to compare results. In this study, we are not only evaluating decreases in lung function due to normal aging only but also those due to the presence of obstructive airway disease primarily caused by asthma. Therefore, we would expect our results to be more similar to those reported by Silverman and coworkers (19), who ascertained families through a proband with early-onset COPD, rather than similar to those obtained by population-based family studies (9, 11, 12, 18). Indeed, three (9, 12, 18) of these four articles did not report strong evidence for linkage of FEV1%FVC (no significance [12, 18] or highest LOD = 1.4 on 6q [9]) in their genome screen, whereas Silverman and colleagues (19) demonstrated strong evidence for linkage of both pre- and postbronchodilator FEV1%FVC to chromosome 2q. Furthermore, Allen and coworkers (38) found an association with asthma on 2q14.1 and, more specifically, of DPP10, a gene encoding a homolog of dipeptidyl peptidases (DPP), which cleaves terminal dipeptides from cytokines and chemokines. The highest LOD score for FEV1%VC we found was approximately 35 cM proximal to the peak reported by Silverman and colleagues (19) and approximately 60 cM distal to the DPP10 locus. Because the latter two studies reported only genomewide screen markers and used the FVC, whereas we used slow vital capacity, it is difficult to determine whether these represent actual different linkage peaks. Moreover, it was not stated whether pre- or postbronchodilator values for spirometry were used by Silverman and colleagues. A later publication (39) showed that post-bronchodilator FEV1%FVC was significantly associated with the same location. Thus, the remaining obstruction after inhalation of a bronchodilator was linked, which the authors suggest reflects the presence of fixed airflow limitation that is characteristic of COPD. We extend this observation in that postbronchodilator FEV1%VC is also associated with genes on chromosome 2q in families of a proband with asthma. Although the linkage peaks observed in the COPD study and this study appear to be distinct, it is possible that they represent linkage to the same locus. In this case, the similar linkage region for both FEV1%VC and FEV1%FVC in families with asthma and COPD may lend support to the hypothesis that the same gene(s) could be important in the development of airway obstruction in both of these airway diseases (40). Silverman and coworkers reported that personal smoking was a significant covariate for the linkage of postbronchodilator FEV1%FVC (19), possibly due to the fact that the families were ascertained through a family member with extensive smoking history. Linkage for the prebronchodilator ratio was not influenced in our population by passive tobacco exposure. This is of interest because of the different characteristics of airway obstruction caused by asthma and COPD in these two populations. It is possible that both studies support the presence of similar genes that cause airway obstruction and modulate pulmonary function in these two disorders. However, this requires further fine mapping and gene identification in families with asthma and with COPD. In contrast to the results on FEV1%VC, linkage of pre- and postbronchodilator FEV1, a major determinant for alterations in FEV1%VC, showed more variability and less significant results, with LOD scores in the range of 1.50 to 1.64. FEV1 is highly influenced by airway obstruction, whereas VC is less affected. FEV1 and VC are only modestly related in the general population, suggesting interindividual differences in airway size relative to lung volume (41). Thus, it is not surprising that different linkage positions in the genome have been identified for the various ways of expressing airflow limitation. FEV1%VC is considered to be a measure of obstructive airway disease (42). It is therefore of interest that we found the strongest evidence for linkage with FEV1%VC and not with FEV1, the most frequently used lung function parameter to assess severity in asthma and COPD. For FEV1 in the families with early-onset COPD, the highest LOD score was observed on chromosome 12p, where the LOD increased to 2.43 at 37 cM after the addition of fine-mapping markers (19). In the general Framingham population (9), the highest LOD was observed on chromosome 6 (LOD, 2.4). Neither of these regions was observed in the current study. This may signify differences in the recruitment of populations, age, ethnic background, or other interacting factors, including the presence of concomitant allergic factors in our families with asthma. The same may be true for the observed linkage of lung function with chromosome 5q in our families with asthma. This linkage was found in the same region with which we have previously found linkage with bronchial hyperresponsiveness (29). This is not surprising because bronchial hyperresponsiveness to a direct stimulus like histamine is determined partially by prechallenge lung function. Whether this implies that the observed linkage reflects the same gene(s) for bronchial hyperresponsiveness as for airway obstruction requires investigation. In summary, we have shown for the first time that significant linkage exists of FEV1%VC to a region on chromosome 2q32 in families of a proband diagnosed with asthma in the 1960s. Other investigators have reported linkage with FEV1%FVC in an adjacent chromosomal region in a population of families of probands with early-onset COPD. This suggests that there may exist a general underlying mechanism of impaired lung function that may be independent of the obstructive airway disease under study. We could not find linkages in pulmonary function that were previously reported in general populations. This may well be due to the fact that we investigated a population ascertained through a parent with asthma and many of the offspring in these families also had this disease. Epidemiologic data suggest an effect of maternal smoking during pregnancy on the development of asthma and atopy. Our results do not suggest an interactive effect of a genetic background and smoke exposure in utero on the development of lung function impairment. We hypothesize that gene(s) on chromosome 2q32, as well as in other linked chromosomal regions reported in this study, may be important in the development of airway remodeling and the severity of airway obstruction in asthma. Further identification of the genes in the regions identified in the current study may help us to better understand some of the underlying biology of the severity of asthma.
Supported by the Dutch Asthma Funds grant AF 95.09 and National Institutes of Health grants R01HL/48341 and R01HL/66393. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form July 6, 2004; accepted in final form May 15, 2005
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||