Heme oxygenase 1 variations and lung function decline in smokers: proof of replication
- 1Department of Epidemiology, University Medical Center Groningen, University of Groningen, the Netherlands
- 2Department of Pulmonology, University Medical Center Groningen, University of Groningen, the Netherlands
- Dr H M Boezen, Department of Epidemiology, University Medical Center Groningen, E3.29, PO Box 30.001, 9700 RB Groningen, The Netherlands;
- Received 14 February 2008
- Revised 14 February 2008
- Accepted 17 February 2008
We provide supportive evidence for a role of the promoter polymorphism (GT-repeat) in heme oxygenase 1 (HO-1) in relation to lung function loss over time. This observation has been made by Guénégou and colleagues in a Caucasian population, and a call for replication of these results in a larger and independent cohort was made.1 The relevance of HO-1 has been widely acknowledged.2 Chronic obstructive pulmonary disease (COPD)—often a consequence of abnormally accelerated lung function decline3—has been widely studied in relation to HO-1 promoter GT-repeat, using a case–control approach. So far these studies provided contradictory findings.4–7
We genotyped the HO-1 GT-repeat in our Dutch general population based Vlagtwedde-Vlaardingen cohort (n = 1390, age median (range): 52 (35–79) years, 51% males, packyears median (range): 9 (0–262), 67.9% ever smokers), that was followed for 25 years (1965–1990) with forced expiratory volume in 1 s (FEV1) measurements every 3 years. We analysed effects of this repeat on FEV1 change using linear mixed effect models as described previously.8 HO-1 GT-repeat polymorphism was genotyped using MegaBACE 1000 (GE Healthcare, UK) system and alleles were defined as published previously (that is, short (S), medium (M), long (L)).1 We additionally extended our study to a separate analysis of all prevalent HO-1 genotypes—that is, S/S (n = 183), S/M (n = 571), S/L (n = 70), M/M (n = 456), and M/L+L/L genotypes pooled together (n = 93+7).
We found the M/L+L/L genotype pool constituted a risk factor for accelerated FEV1 decline in the total population compared with any other genotype. The mean adjusted change in FEV1 for the M/L+L/L genotype pool was −24.2 ml/year. S/L, S/S, S/M and M/M genotypes provided, respectively, 6.3 (SE = 2.8, p = 0.025), 3.6 (SE = 2.3, p = 0.120), 5.7 (SE = 2.0, p = 0.005), and 4.9 (SE = 2.0, p = 0.017) ml/year less decline compared to the M/L+L/L pool. Since mentioned associations remained significant in ever smokers and furthermore in heavy smokers exclusively (packyears smoked >9; except for the M/M genotype where p = 0.08), we confirm an existence of the interaction between this polymorphism and smoking as observed previously.1
To our knowledge this is the first successful replication of a candidate polymorphism in relation to FEV1 decline in the general population. We additionally narrow the risk HO-1 genotypes to the M/L+L/L pool. Furthermore, we confirm that GT-repeat variation in the HO-1 plays a particular role in smokers, which likely is related to the unique role of this gene in regulation of smoking induced oxidative stress.2 We believe that further studies on the HO-1 GT-repeat will provide more insight into the role of this gene in lung function loss and eventually in the pathogenesis of COPD.
Competing interests: None.
Funding: Dutch Asthma Foundation (grant 3.2.02.51), The Netherlands; University of Groningen, the Netherlands.