BMC Pulmonary Medicine BioMed Central

Background Endothelin-1 (ET-1) and Nitric Oxide (NO) are crucial mediators for establishing pulmonary artery hypertension (PAH). We tested the hypothesis that their imbalance might also occur in COPD patients with PAH. Methods The aims of the study were to measure exhaled breath condensate (EBC) and circulating levels of ET-1, as well as exhaled NO (FENO) levels by, respectively, a specific enzyme immunoassay kit, and by chemiluminescence analysis in 3 groups of subjects: COPD with PAH (12), COPD only (36), and healthy individuals (15). In order to evaluate pulmonary-artery systolic pressure (PaPs), all COPD patients underwent Echo-Doppler assessment. Results Significantly increased exhaled and circulating levels of ET-1 were found in COPD with PAH compared to both COPD (p < 0.0001) only, and healthy controls (p < 0.0001). In COPD with PAH, linear regression analysis showed good correlation between ET-1 in EBC and PaPs (r = 0.621; p = 0.031), and between arterial levels of ET-1 and PaPs (r = 0.648; p = 0.022), while arterial levels of ET-1 inversely correlated with FEV1%, (r = -0.59, p = 0.043), and PaPs negatively correlated to PaO2 (r = -0.618; p = 0.032). Significantly reduced levels of FENO were found in COPD associated with PAH, compared to COPD only (22.92 ± 11.38 vs.35.07 ± 17.53 ppb; p = 0.03). Thus, we observed an imbalanced output in the breath between ET-1 and NO, as expression of pulmonary endothelium and epithelium impairment, in COPD with PAH compared to COPD only. Whether this imbalance is an early cause or result of PAH due to COPD is still unknown and deserves further investigations.


Background
Some 100,000 Iranians were exposed to chemical warfare agents during the 8-year Iraq-Iran war, with the approximately 50,000 mustard gas-affected individuals exhibiting a pattern of late respiratory, eye and skin complications [1]. Chronic bronchitis, asthma, bronchiectasis and pulmonary fibrosis account for the most frequent long-term respiratory sequelae, with progressive decline occurring over many years [2][3][4][5].
Mustard gas, bis (2-chloroethyl) sulphide, is a bifunctional alkylating agent. It is a potent vesicant, whose rapid penetration leads to extensive blistering in all epithelial tissues exposed to it. The immediate toxicity of mustard gas is thought to be due to the consequences of both DNA and protein alkylation reactions [6]. However, the factors which drive the development and progression of the long term respiratory complications remain unclear.
The circulating or endocrine renin-angiotensin system (RAS) plays a key role in circulatory homeostasis. Angiotensin Converting Enzyme (ACE) converts angiotensin I to the potent vasoconstrictor angiotensin II. However, local RAS exist in diverse human tissues, where they play proinflammatory and profibrotic roles [7][8][9]. A lung RAS is now known to exist and is implicated in the genesis of lung inflammatory and fibrotic responses [10].
Whether in the circulation or the tissues [11,12], the absence (deletion, D allele) rather than the presence (insertion, I allele) of a 287 base pair fragment in the human ACE gene is associated with increased ACE activity. In keeping with the postulated roles for ACE in the lung, the D-allele has been associated with development of the acute respiratory distress syndrome, and poorer markers of respiratory function in acute illness [13,14]. However, such detrimental impact of the D-allele in acute illness might be counterbalanced by more positive systemic affects in chronic disease, as suggested by the association of the D-allele with preserved skeletal muscle strength amongst patients with chronic pulmonary disease [15].
We thus postulated that ACE genotype might influence the severity of the late respiratory complications of mustard gas exposure, and have tested this hypothesis in a pilot study.

Methods
Thus study was approved by the Ethics Committee of the Iranian Janbazan [Veterans'] organization. Written informed consent was obtained from each subject.

Subjects
All were Kurdish civilians and had suffered high-level mustard gas exposure at Sardasht, Iran on June 29 1987, as recorded by the Janbazan veterans' organization -the official center for compensation of war disabled victims. In keeping with accepted methodologies, those with only erythema were defined as "low exposure", whilst those with erythema and edema, vesiculation, scaling, ulceration, or crusting are categorized as "high exposure" [16]. Using such criteria, 'high exposure' subjects were selected. Patients were excluded if they had a history of smoking, heart failure, occupational history of chemical agent exposure or a lung inflammatory disorder of a different origin.

Pulmonary impairment
Spirometric evaluation was performed in September 2005, with forced expiratory volume over 1 second (FEV 1 ), forced vital capacity (FVC), and the ratio between them calculated (Multi-Functional Spirometer HI-801, Chest M.I., INC, Tokyo, Japan). Spirometry was performed according to international guidelines with the best of three readings recorded.

Genotyping
Five milliliters of EDTA blood was obtained from each subject, and ACE genotype was determined using threeprimer polymerase chain reaction amplification (PCR) and subsequent agarose gel electrophoresis, as previously reported [17]. All gels included a positive heterozygous control sample, and genotypes were read by two independent observers blind to case/control status. Discrepancies were resolved by repeat PCR.

Statistical Analysis
Analysis was performed using Statview 5.0 (Abacus concepts, Inc., Berkeley, CA, USA). Deviation from Hardy-Weinberg equilibrium was considered using a chi-squared test. Disease severity was defined according to a median split of FEV 1 % predicted into better or worse lung function groups. The influence of age, gender and ACE genotype on disease severity was tested using binary logistic regression analysis comparing D homozygotes to other genotypes. Throughout, a p-value < 0.05 was considered statistically significant. Data was normally distributed and results presented as mean ± standard deviation (SD).
For technical reasons, ACE genotype could not be determined in 1 subject. In the remaining 207 subjects, ACE genotype distribution was 37 (17.9%) vs 115 (55.6%) vs 55 (26.6%) for II, ID and DD respectively, and was consistent with Hardy-Weinberg equilibrium (p value = 0.11). Neither age nor gender varied significantly between ACE genotypes.
As a continuous variable, FEV 1 % predicted tended to be higher in association with the D allele 68.03 ± 20.5%, 69.4 ± 21.4% and 74.8 ± 20.1% for II, ID and DD genotypes respectively (Figure 1) linear trend p = 0.10. Analysis of FVC % predicted demonstrated higher values for the DD genotype with FVC % predicted of 74.59 ± 17.2%, 72.9 ± 19.5% and 77.13 ± 20.3% for II, ID and DD genotypes respectively (ANOVA p = 0.43) Gender did not appear to have an influence on disease severity. Mean FEV 1 % predicted for males was 71.7 ± 21 and for females 70.0 ± 21. There was no significant association between gender and severity grouping (Chi 2 p = 0.43).
Increasing age was associated with reduced FEV 1 % predicted (r 2 0.05 p = 0.001). Mean age in the high FEV 1 % predicted group was 43.2 ± 13.2 in comparison with 50.4 ± 14.7 in the low FEV 1 % predicted group (t test p = 0.0002).
Median FEV 1 % predicted was 73 and this was taken as a cut off between groups demonstrating higher or lower spirometry readings. (Table 1).
The ACE DD genotype was overrepresented in the better spirometry group (Chi 2 4.9 (p = 0.03) ( Table 2). In a logistic regression model including age this became more significant (p = 0.011).
The presence or absence of particular symptoms and signs, or the number of them present was not associated with genotype.

Discussion
This study is the first to suggest an association between the Angiotensin Converting Enzyme Insertion/Deletion polymorphism and the severity of late respiratory complications of mustard gas exposure, as measured by FEV 1 . 33.6% of subjects in the better lung function group carry the DD-genotype, whereas only 12% in this group carry the II-genotype.
The finding that the D-allele was over-represented in subjects with less severe late pulmonary impairment following exposure to mustard gas might be considered unexpected given the higher ACE activity associated with it and the recognized pro-inflammatory and pro-fibrotic roles of increased RAS activity in the lung. In contrast to this finding, previous studies have demonstrated associations of the D-allele with increased severity of pulmonary sarcoid [18], acute lung injury responses in adults [19], non-infectious pulmonary complications of bone marrow transplantation [20] and of esophageal surgery [21], and   with the development of bronchopulmonary dysplasia after premature birth [22].
Recently a study of transbronchial lung biopsies in patients with severe mustard gas related lung disease has revealed histopathological changes diagnosable as organizing pneumonia [3,23]. Previously fibrosis had been considered to be the predominant pathology in the late respiratory complications of mustard gas exposure. The pathogenesis of organizing pneumonia may interact with the RAS in a different way to the generation of fibrosis and help explain the association of the D-allele with less severe pulmonary impairment.
Most studies examining the role of the RAS in lung inflammation and fibrosis have associated the I-allele with less severe disease in the short term response to a pathological insult. This study examined pulmonary impairment nearly 20 years after exposure to mustard gas and may suggest that the RAS has a different influence on chronic pulmonary disease.
Additionally we found the severity of pulmonary impairment following mustard gas exposure to be associated with increasing age -findings that accord with those of Zarchi et al, in a study of 1337 soldiers exposed to mustard gas [24]. The mechanism underlying this increased susceptibility with age is unclear. It could represent a greater susceptibility at a pulmonary level or could be that removal of children from exposure (both physically and the removal of clothing to prevent ongoing exposure) may have been more effective A further study is warranted -and one which not only includes far greater numbers, but which obtains far greater phenotypic detail -including detailed non-invasive pulmonary function testing, and imaging.
These findings, and the issue of further study, are of importance for a number of reasons. Certainly, gene-environment studies such as this may prove powerful in exploring the fundamental mechanisms driving progressive lung pathology of diverse origin. In addition, however, there are many tens of thousands of patients suffering the pulmonary sequelae of mustard gas exposure, for whom no specific therapeutic modality is yet available. The demonstration that ACE influences such pathology may open the way to new therapeutic options.

Conclusion
ACE genotype influences the severity of the late respiratory complications of mustard gas exposure with the D allele being associated with higher FEV 1 % predicted 18 years after exposure.