Abstract
It has been suggested that interleukin (IL)-18 plays a role in the development of inflammatory and fibrosing lung diseases.
Associations of polymorphisms in the genes coding for IL-18 (IL18 /G-656T, C-607A, G-137C, T113G, C127T) and its receptor (IL18R1 /C-69T) with coal workers' pneumoconiosis (CWP) were studied in 200 miners who were examined in 1990, 1994 and 1999. Coal-dust exposure was assessed according to job history and ambient measures. The main health outcome was lung computed tomography (CT) score in 1990. Internal coherence was assessed by studying CT score in 1994, 4-yr change in CT score and CWP incidence and prevalence.
CT score in 1990 was a good predictor of radiographic grade in 1999 and, therefore, an appropriate subclinical quantitative trait. The IL18 -137C allele was associated with lower CT score in 1990 and 1994 (1.24 versus 1.69 and 1.57 versus 2.46, respectively), slower progression of CT score between 1990 and 1994 and lower pneumoconiosis prevalence in 1999 relative to the G allele (0.33 versus 0.77 and 8.2 versus 19.6%, respectively). Smoking- or dust-adjustment, and stratification on IL18R1 genotype and adjustment for haplotype effects did not change the conclusions.
In conclusion, the results of the present study suggest a role for IL18 in reducing the development of this fibrosing lung disease.
Interleukin (IL)-18 is a recently described lymphokine involved in neutrophil activation, reactive oxygen species (ROS) synthesis 1, pro-inflammatory cytokine production, nuclear factor (NF)-κB activation and degranulation 2. A role for IL-18 in pulmonary inflammation has been suggested by studies in rodent models, but its importance is not clearly understood. In epidemiological genetic studies, the IL18 C-607A single nucleotide polymorphism (SNP) was significantly associated with higher prevalence of sarcoidosis in Japanese subjects 3 but not in Dutch subjects 4. The IL18 A105C SNP was significantly associated with asthma 5, the G-allele of the IL18 promoter variant (-137G/C) was associated with an increased risk of atopic asthma in the Swiss Study on Air Pollution and Lung Disease in Adults (SAPALDIA) Cohort Study 6, and the IL18 G-656A, G-137C, T-133G, T113G and C127T SNPs were significantly associated with high immunoglobulin (Ig)E levels, specific sensitisation to common allergens and seasonal allergic rhinitis in 105 Caucasian families 7. However, these findings have not been replicated 8.
Davis et al. 9 proposed that IL-18 plays an early role in the reiterative process of macrophage−lymphocyte interaction following silica exposure in mice, leading to chronic inflammation, tissue injury and collagen production. Kitasato et al. 10 reported markedly elevated levels of IL-18 in the serum and bronchoalveolar lavage of patients with idiopathic pulmonary fibrosis compared with controls. Coal workers' pneumoconiosis (CWP) is another inflammatory and fibrosing lung disease caused by chronic inhalation of particles. The overall hypothesis of the present study was that IL18 and IL18R1 polymorphisms contribute to the pathogenesis of CWP. To test this hypothesis, the present authors investigated the associations of IL18 and IL18R1 SNPs with computed tomography (CT) score, a quantitative subclinical phenotype predicting the occurrence and the evolution of the disease 11–13, and with disease prevalence. The primary health outcome for the study was the CT score at the first survey, when nearly all miners were active. The internal coherence of the results was tested by studying associations with CT score at the second survey and with CWP prevalence at the end of the follow-up. CT score change between 1990 and 1994 was largely considered as a measure of the activity of the disease between both surveys. The present authors also tested the effects of the interactions between polymorphisms in IL18 and its receptor and coal-dust exposure (i.e. gene X−environment interaction), and between IL18 and IL18R1 polymorphisms (i.e. gene X−gene interaction) on lung CT score and disease prevalence.
METHODS
Study sample
The study design has been described elsewhere 14. Briefly, unrelated coal miners (aged 34–50 yrs) were recruited in 1990 to be contrasted according to exposure and chest radiography. They included: 80 subjects heavily exposed to underground coal-dusts (≥10 yrs at the coal face), with chest radiography classified 0/1 or 1/0 and without health alteration from any other diseases; 80 healthy subjects exposed to underground coal-dusts, with normal chest radiography classified 0/0; and 80 healthy subjects slightly exposed, with normal chest radiography. The three groups were matched for age and smoking habits 12. Miners were re-examined in 1994 and in 1999.
The present study sample included 200 coal miners for whom genetic, environmental and health data were available from 1990 to 1999; no differences regarding genotype, exposure and health data were found between the study sample and those not included in the analyses (n = 40). In 1990, 96% of the miners were active; the proportion of retired miners rose from 24 to 88% between 1994 and 1999. The appropriate ethical committee approved the study and written consent was obtained from all subjects.
Environment
Besides smoking, detailed information on high or low current exposure based on job description and cumulative exposure was recorded 15. High current exposure refers to miners working at the coal face, mining, stope or drift advance. Low current exposure refers to those working in ventilation maintenance, pumping, haulage, shaft, stock equipment or safety. Cumulative personal exposure to dust was estimated from each person's job history and from dust measurements at various sites of the coal mine. The estimates were summations of each dust measurement (mg·m-3) for the respective time spent in each job. Estimated cumulative exposures to dust were calculated until 1999 and expressed as mg·m-3·yr-1.
Radiographic examination
CT scans were performed in 1990 and 1994 for all subjects, as already described elsewhere 12. Briefly, the lungs were divided into six areas: the upper zones above the carina, the middle zones between the level of the carina and the lower pulmonary veins, and the lower zones below the level of the lower pulmonary veins. Micronodules were defined as opacities <7 mm in diameter and nodules were defined as opacities 7−20 mm in diameter. Micronodules, nodules and other abnormalities, such as emphysema and profusion of shadows, were recorded according to the criteria of the consensus meetings for CT scans established by the Society for Thorax Imagery and the Thorax Club in September 1989 and June 1990 16, and according to the pathology standards for Coal Workers' Pneumoconiosis defined by the College of American Pathologists 17. Profusion of abnormalities was graded 0−3 (absent, rare, intermediate, high profusion), and was estimated for each of the six lung zones (upper, middle and lower parts of both lungs), giving a total CT score of 0−18. Analyses were based on the mean CT score calculated for each individual by dividing their total score by six. The readings of the scans for the first and second examinations (1990 and 1994) were performed at the same time by two experienced radiologists blinded to individual exposure and radiographic findings. Where discrepancies occurred, a consensus was reached.
At each survey (1990, 1994 and 1999), chest radiographs taken in the yearly medical examination were interpreted by two independent and experienced physicians, according to the International Labour Office (ILO) standardised classification of radiographs of pneumoconiosis 18. The films were presented in random order, without any information about the occupational and medical history of the subjects. For statistical analyses, the 12-point ILO profusion grades were reduced to three points: 0/- and 0/0, 0/1 and 1/0, and ≥1/1. Subjects with a profusion grade of <1/1 were considered as not having pneumoconiosis, those with a profusion grade of 0/- or 0/0 were considered as having normal chest radiography, and those with a profusion grade of 0/1 or 1/0 were suspected to be in the process of evolution to pneumoconiosis. Pneumoconiosis was defined by a grade ≥1/1.
Genotyping
IL18 -137, -607, +113, and +127 SNPs were genotyped in 2004, as previously described 19. IL18 -656 and IL18R1 -69 SNPs were genotyped by allelic discrimination using TaqMan probes. Primers and TaqMan probes were designed using Primer Express v2.0 (Applied Biosystems, Foster City, CA, US). The forward and reverse primers for IL18R1 are 5'-tttttttaaaaatctgtgtgccagaa-3' and 5'-cagccaaagctttcaaacaaaa-3', respectively and TaqMan probes are 5'-ttatgaaAgtttaaaaatc-6-carboxyfluorescein(Fam)-3' and 5'-ttatgaaGgtttaaaaat-Vic-3'. The forward and reverse primers for IL18 -656 are 5'-taggtcagtctttgctatcattcca-3' and 5'-acactttctgcaacagaaagtaagct-3', respectively and TaqMan probes are 5'-aattttggtaTccctctc-Fam-3' and 5'-aattttggtaGccctct-Vic-3'. For both assays, primers, probes, and TaqMan Universal MasterMix with no AmpErase UNG (Applied Biosystems) were utilised according to the manufacturer's standard protocol in a Prism 7000 Sequence Detection System (Applied Biosystems). Genotypes were determined by manual clustering using Prism 7000 sequence detection software version 1 (Applied Biosystems).
Statistical methods
Standard statistical tests (Chi-squared or Fisher exact test when appropriate, and logistic regression for qualitative variables; ANOVA and multiple regression analysis for quantitative variables) were performed. Significance was assessed at the 5% two-sided level.
All analyses were first conducted considering each IL18 SNP separately and the main outcome CT score in 1990, where 96% of the miners were active. Associations of each SNP with CT score in 1994, change in CT score between 1990 and 1994, and 1999 CWP incidence and prevalence were also investigated to test the coherence of the results and the activity of the disease (change in CT score between 1990 and 1994). No a priori adjustment was performed. CT score was not normally distributed, but nonparametric Kruskal−Wallis and standard parametric tests (ANOVA and unpaired t-test) gave similar p-values. Analyses were conducted considering subjects heterozygous and subjects homozygous for the variant allele, and variant allele carriers, as in previous studies 3–7. IL18 haplotype analysis was then performed using a maximum likelihood method for haplotype−phenotype association as implemented in the Testing Haplotype Effects In Association Studies (THESIAS) program 20, 21. The most frequent haplotype (IL18 -607C/IL18 -137G (CG)) was used as the reference.
Interaction between genetic polymorphisms and exposure to coal-mine dusts, or between both genetic polymorphisms (IL18 and its receptor IL18R1) on health outcomes (CT score and pneumoconiosis prevalence) were statistically tested using multivariate linear or logistic regression models.
RESULTS
The characteristics of the 200 miners included in the analyses are summarised in table 1⇓. The mean age of the miners in 1990 was 43 yrs. More than half of the miners were current smokers, and 48.5% were highly exposed to coal-mine dusts in 1990. Almost 68% were born in France and only 1.5% had their geographical origin in non-European countries. Among all coal miners, the CT score increased by ∼40% between 1990 and 1994, and the prevalence of pneumoconiosis rose from 3.5% to 14% between 1994 and 1999.
Characteristics of coal miners
CT score and radiographic grade were highly associated in 1990 and in 1994. Mean±sd CT scores in 1990 were 0.71±1.19, 2.98±2.20 and 3.09±2.61 in miners with radiographic grades of 0/0, 0/1 and 1/0, respectively (trend test, p<0.0001). In 1994, mean±sd CT scores were 1.07 ±1.38, 3.87±2.95 and 4.39±3.60, with the seven pneumoconiotic miners (≥1/1) having a mean±sd score of 7.43±4.20 (trend test, p<0.0001). CT score and cumulative coal-dust exposure (mg·m-3·yr-1) were highly correlated in 1990 and in 1994 (r = 0.35; p<0.0001 and r = 0.29; p<0.0001, respectively).
In 1999, 26 (92.9%) of the 28 pneumoconiotic miners were those heavily exposed to underground coal-dusts with chest radiographs classified 0/1 or 1/0 in 1990, two (7.1%) were those exposed to underground coal-dusts with normal chest radiographs classified 0/0 in 1990. No miner slightly exposed with normal chest radiograph in 1990 had pneumoconiosis in 1999. Figure 1⇓ shows the relationship between radiographic grade in 1999 and CT score in 1990. At the end of the 10-yr follow-up, 46 subjects had worsened radiographic findings and 28 of them were pneumoconiotic. CT score in 1990 was significantly higher in miners whose radiographic findings had worsened in 1999, compared with those who had not (3.67±2.39 (n = 46) versus 0.81±1.22 (n = 154); p<0.0001), and in miners who developed pneumoconiosis compared with others (4.18±2.68 (n = 28) versus 1.03±1.41 (n = 172); p<0.0001).
Box plots showing computed tomography (CT) score, a quantitative trait predicting the evolution to pneumoconiosis, in 1990 versus chest radiographic grade in 1999. Plots show the median (|), the first and third quartile (░), the first and last decile (−−−) and the maximum (•) of CT score for each radiographic grade category. Mean±sd CT scores for each category are 0.72±1.19, 1.35±1.00, 2.76±1.70, 3.58±2.22 and 7.75±2.63, respectively (p<0.0001).
Genotype and allele frequencies
Minor allele frequencies were: 0.42 for IL18 G-656A and C-607A; 0.285 for IL18 G-137C, T+113G, and C+127T; and 0.347 for IL18R1 C-69T. All of these fit predictions for Hardy−Weinberg equilibrium (all p>0.6). Complete linkage disequilibrium was observed between the IL18 -656 and -607 genotypes, and between IL18 -137, +113 and +127 genotypes. Three haplotypes were found: IL18 -607C/IL18 -137G (CG), 58.0%; AC, 28.5%; and AG, 13.5%. All miners homozygous for the IL18 -137C allele were homozygous for the IL18 -607A allele, and 63.7% of miners homozygous for the IL18 -137G allele were homozygous for the IL18 -607C allele (p<0.0001 for association). No differences in genotype or allele distributions were observed according to the geographical origin of the miners (data not shown).
Association of IL18 -607, IL18 -137 and IL18R1 SNPs with stage of pneumoconiosis
No significant association was found between IL18 or IL18R1 SNPs and CT score in 1990 (table 2⇓). Furthermore, no significant associations were found between IL18 C-607A or IL18R1 C-69T genotype and CT score in 1994, change in CT score between 1990 and 1994, or pneumoconiosis incidence or prevalence.
Association of polymorphisms in the genes coding for interleukin-18 (IL18) and its receptor (IL18R1) with stages of pneumoconiosis in coal miners
Lower CT score in 1990, significantly lower CT score in 1994 and slower progression of CT score were found in IL18 -137C carriers (i.e. miners homozygous or heterozygous for -137C). No IL18 -137C carrier had pneumoconiosis in 1994, and the IL18 -137C allele was significantly associated with lower disease prevalence in 1999.
No interaction was observed between SNPs and coal-dust exposure on CT score in 1990 and in 1994, change in CT score or disease prevalence in 1999 (data not shown). Analysing smoking- or dust-adjusted CT score and disease prevalence, or stratifying on IL18R1 C-69T genotype did not change the conclusions.
Haplotype analyses
No association between IL18 -607/-137 haplotype AC and CT score in 1990 was found (table 3⇓). The IL18 -607/-137 haplotype AC was associated at borderline significance with CT score in 1994, and was significantly associated with a slower progression of CT score between 1990 and 1994 and with a lower prevalence of disease in 1999. No significant association was found with haplotype AG.
Association of haplotypes of the gene coding for interleukin-18 (IL18) with stages of pneumoconiosis in coal miners
DISCUSSION
The present study tested the hypothesis that polymorphisms in IL18 and IL18R1 contribute to the pathogenesis of CWP, an inflammatory and fibrosing lung disease. The results show significant associations of the IL18 -137C allele with CT score in 1994, slower progression of CT score between 1990 and 1994 and lower pneumoconiosis prevalence in 1999 relative to the G allele. Adjusting for haplotype effects confirmed the results. Furthermore, analysis of smoking- or dust-adjusted CT score or disease prevalence, or stratifying on IL18R1 genotype, did not change the conclusions.
CT is not used as a standard method to assess pneumoconiosis, although it is a sensitive tool for the evaluation of lung parenchyma 11. The determination of genetic factors is greatly enhanced by considering subclinical quantitative phenotypes 22. CT has been proposed as a screening method to distinguish normal from early pneumoconiosis 13. In the present study, the predictive value of CT score as an appropriate subclinical quantitative phenotype was confirmed. Results with a 10-yr follow-up confirm and extend those reported in a 4-yr period 12. As mines have been totally closed in France, it was not possible to build a replication sample. Furthermore, the use of CT in coal mining is still limited at an international level. CWP, however, remains a major occupational disease in terms of public health burden worldwide and it is important to better understand the genetic modifiers of this environmental disease.
The present study has some limitations. Not all of the IL18 and IL18R1 SNPs were considered, nor were other genes involved in the IL-18 pathway which might contribute, alone or in combination, to IL-18 variability and in turn affect CWP susceptibility. However, the SNPs analysed were those with functional significance that have been previously published. Furthermore, the sample size of the population studied was small. A precise assessment of the power could not be performed prior to the study; variations of CT score across the general population are not known, as it is not possible to perform CT scanning on a large scale for ethical and technical reasons. However, it was anticipated that the contrasted disease status chosen by design and the availability of such a sensitive score would increase the power to detect differences compared with classical designs based on random samples using only chest radiographic measures.
Few epidemiological genetic studies have examined the associations of the IL18 A-607C and G-137C SNPs with chronic inflammatory and fibrosing lung diseases, and none have simultaneously investigated the role of environmental factors and polymorphism in IL18R1. Furthermore, none has considered CT score. Kruse et al. 7 reported a significant association of the IL18 137C allele with high serum IgE levels, specific sensitisation to common allergens and seasonal allergic rhinitis in 105 German families. The IL18 A-607C SNP was unrelated to these phenotypes. In populations recruited in the same areas, the IL18 A-607C and G-137C SNPs were unrelated to bronchial asthma in 230 children compared to 270 controls 8. Recently, a significant association of IL18 -137G allele with increased risk for atopic asthma in the SAPALDIA Cohort Study 6 was found. Takada et al. 3 reported a significantly higher frequency of the IL18 -607C allele in 119 Japanese sarcoidosis patients compared with 130 controls, and no association was found for the IL18 G-137C SNP. The significant association was not replicated in a population of Dutch Caucasians, where 133 sarcoidosis patients were compared with 103 controls 4. The inconsistency between studies may be due to differences in IL18 -607A and -137C allelic frequencies between populations, or in asthma onset or biological pathways during disease progression. Furthermore, the pleiotropic role of IL-18 with varying effects according to the cytokine milieu (i.e. T-helper (Th) type 2 cytokines when considering atopic phenotypes, or Th2 and Th1 cytokines when considering asthma 23) could also partly explain these differences.
In the present study, only the IL18 G-137C SNP was significantly associated with pneumoconiosis phenotype and prevalence. Taking into account coal-dust exposure, which is the first cause of the disease, and stratifying on IL18R1 genotype did not change the conclusions. Genotype and allelic frequencies were very close to those reported in previous studies in Caucasians 4, 6–8. Haplotypes found in the present study were identical to those reported by Giedraitis et al. 24 in a Swedish population. Giedraitis et al. 24 found that the haplotype -656T/-607A/-137C/+113G/+127T (-607/-137 AC in the present study) was clearly associated with lower promoter activity and lower IL18 gene expression than haplotypes CG and AG. Furthermore, in C carriers at position -137, no correlation was found between IL-18 and interferon (IFN)-γ mRNA levels, whereas a strong correlation was found in those homozygous wildtype, with GG at -137.
The results of the present study were also consistent with the study of Wei et al. 25, in which markedly reduced incidence and severity of collagen-induced arthritis was found in IL-18-/- mice compared with wildtype mice. This was accompanied in vitro by significantly reduced production of pro-inflammatory cytokines, including IFN-γ. Further, significantly reduced lung collagen was observed in IFN-γ-deficient mice exposed to silica compared with wildtype mice 9. Pneumoconiosis is another collagen-related disease, including the activation of alveolar macrophages, ROS synthesis and the production of pro-inflammatory cytokines such as tumour necrosis factor and NF-κB. IL-18 is involved in all of these steps 1–2.
In conclusion, it was found that the IL18 G-137C single nucleotide polymorphism was associated with lower computed tomography score, slower progression of computed tomography score and lower pneumoconiosis prevalence. Furthermore, smoking- or dust-adjustment, stratification on IL18R1 genotype and adjustment for haplotype effects did not change these conclusions. The results are consistent, support the biological and functional significance of IL18 and suggest its potential role in reducing the development of this inflammatory and fibrosing lung disease. However, it is premature to consider any clinical application of the findings and replication of these findings in additional populations is warranted.
Acknowledgments
The authors would like to thank the medical staff, especially J-P. Bertrand, and all the French coal mine workers who made the present study possible.
- Received March 3, 2006.
- Accepted August 21, 2006.
- © ERS Journals Ltd
References
