The association of Toll-like receptor 4 gene polymorphisms with the development of emphysema in Japanese subjects: a case control study

Chronic obstructive pulmonary disease (COPD) is a major global health problem that causes 64 million patients with COPD worldwide in 2004 and more than 3 million deaths in 2005 [1]. COPD is predicted to become the third most common cause of death and the fifth most common cause of disability in the world by 2020 [2]. The inflammation, extracellular matrix synthesis, oxidative stress and apoptosis are the major pathophysiological pathways in the mechanisms of COPD [3]. Recently, an animal model of autoimmune emphysema showed that CD4+ cell-dependent mechanisms were sufficient to trigger the development of emphysema, suggesting that alveolar septal cell destruction might result from immune mechanism [4]. The development of COPD is known to be influenced by multiple genetic factors [5]. A coding variant in surfactant protein B (SFTPB Thr131Ile) and the (GT)₃₁ allele of the heme oxygenase (HMOX1) promoter short tandem repeat were evidenced to be associated with COPD in both the family-based study and case-control study [6]. In addition, TNF-α-308-1 and TNF-α-308-2 alleles, IL-13 promoter polymorphisms, metalloproteinase (TIMP)-2 polymorphisms, and β₂-adrenoceptor Gly16 polymorphism were proved to be significantly associated with the presence of smoking-related COPD [7]. We also demonstrated that the transforming growth factor beta 1 gene polymorphisms were associated with emphysema phenotype of COPD in Japanese [8]. Recently, genome-wide association study identified bicaudal D homolog 1 (BICD1) as a susceptibility gene for emphysema [9]. However, few evidences have shown that genes involved with innate immune system are associated with emphysema or COPD.

The principal role of Toll-like receptors (TLRs) is the induction of immune responses. TLRs activate both innate and adaptive immune responses; they regulate the immediate response to pathogens and antigen presentation to the adaptive system [10, 11]. Among the TLR family members, TLR4 is activated by bacterial lipopolysaccharides (LPS) [12]. LPS is a glycolipid component of Gram-negative bacteria cell walls; it is present in airborne particles, like tobacco smoke [13]. Recent evidence has demonstrated that mice deficient in the TLR4 gene developed pulmonary emphysema [14]. In that study, TLR4 knock out mice exhibited upregulation of a novel NADPH oxidase system in lungs and endothelial cells; this resulted in increased oxidant generation and elastolytic activity. However, various mediators of inflammation, including IL-1β, TNF-α, IL-6, IL-13, IFN-γ, and VEGF, were not differentially expressed. It is thought that TLR4 might maintain constitutive lung integrity by modulating oxidant generation. In another study, when TLR4 transgenic mice were exposed to hyperoxia, reduced apoptosis was observed in alveolar type I and II epithelial cells and alveolar macrophages. In addition, antiapoptotic molecules, like heme oxigenase-1, were up-regulated [15]. Those results suggested that TLR4 may function to protect against the development of emphysema through defending the oxidative stress and apoptosis.

The TLR4 gene is located on chromosome 9q32-33; it spans approximately 13 kb and contains three exons that encode a 222-amino acid protein. Single nucleotide polymorphisms (SNPs) in the TLR4 gene have been reported to be associated with endotoxin hyporesponsiveness and gram-negative infections [16–18]. TLR4 SNPs have been shown to affect the risks of various inflammatory diseases, including atherosclerosis [19], Crohn's disease [20], rheumatoid arthritis [21], and prostate cancer [22, 23]. Studies on the association of TLR4 SNPs with COPD showed that the Asp299Gly (rs4986790) and Thr399Ile (rs4986791) polymorphisms of TLR4 gene were strongly associated with Caucasian patients [24–26]. However, these two functional SNPs were absent in the Japanese population according to the HapMap Project. At present, no TLR4 SNPs have yet been reported with the emphysema phenotype of COPD in Japanese subjects. In this case-control study, we investigated the association between TLR4 SNPs and COPD in Japanese subjects, with a focus on the emphysema phenotype.

The present study showed the minor C allele and the CC genotype of the rs11536889 SNP in the TLR4 gene were likely associated with the emphysema phenotype of COPD in Japanese subjects. The strength of the association of the CC genotype with emphysema phenotype was in an OR of 2.60. Moreover, the frequency of the major haplotype (Haplotype 1), which carried the minor C allele of rs11536889, was significantly higher in the emphysema group than in the controls. Another haplotype (Haplotype 4), which carried the major G allele of the rs11536889, was significantly higher in the in the controls than the emphysema group. However, we found that the clinical features of emphysema, such as %FEV₁, %DLco and LAA scores were not significantly associated with the presence of rs11536889 in patients with emphysema.

TLRs are expressed in many cells, including airway epithelial cells, alveolar type II epithelial cells, alveolar macrophages, endothelium, fibroblasts, vascular smooth muscle cells, and T-cells [27]. TLR4s can recognize both LPS and a respiratory syncytial virus fusion protein [28]. The latter is a major respiratory pathogen in humans, infects the lower respiratory tract, and can exacerbate COPD. TLR4 can also be activated by proteins released from dead and dying cells, and tissue matrix breakdown products in the absence of infection, including high-mobility group protein 1 [29], surfactant protein A [30], fibronectin [31], fibrinogen [32], and hyaluronic acid oligosaccharides [33]. In smokers, stimulation with TLR4 agonists caused alveolar macrophages to reduce gene expression and secrete proinflammatory cytokines and chemokines [34]. Smokers and severe COPD patients displayed reduced TLR4 gene expression in the nasal epithelium. Consistent with that finding, a human epithelial cell line exposed to cigarette smoke extracts showed dose-dependent reductions in TLR4 mRNA and protein expression [35]. Those findings suggested that TLR4 might play a crucial role in the pathogenesis of COPD inflammation.

In the present study, the distribution of the rs11536889 CC genotype in control smokers was consistent with that reported for Japanese subjects in the HapMap project. In contrast, the CC genotype in the emphysema group was significantly higher than in the controls. The C allelic was also associated with emphysema phenotype. The rs11536889 C allele was previously demonstrated to be associated with moderate and severe periodontitis [36] and a high risk of gastric atrophy in Helicobacter pylori seropositive Japanese subjects [37]. However, the rs11536889 was not associated with autoimmune pancreatitis [38] or sarcoidosis-related uveitis [39] in Japanese subjects.

The rs11536889 SNP is located in the 3'-UTR of the TLR4 gene on chromosome 9q32-q33; it substitutes the ancestral guanine (G) with a mutated cytosine (C) at nucleotide position 3725. This SNP does not have any direct influence on the conformation of the TLR4 protein molecule, according to updated biogenetic data. However, it remains possible that this G/C mutation may have an effect on mRNA stability and transcription and/or translation efficiency; this might cause a dysfunction of the TLR4 molecule and interfere with the host immune system.

Two mutations in the TLR4 gene, Asp299Gly (rs4986790) and Thr399Ile (rs4986791), were reported to be highly associated with COPD in Caucasians [24–26], but these two functional SNPs were absent in the Japanese population according to the HapMap Project. At present, no other additional SNPs of the TLR4 gene have been reported to be associated with emphysema or COPD in Japanese or other populations worldwide.

The major limitation of the present case-control study was the relatively small sample sizes for both the case and control groups. The cases were strictly defined as the emphysema phenotype among the COPD patients by HRCT evaluation that was only applied to a part of the COPD patients due to several reasons of the patients. And the control group was recruited from the health centers where the apparent healthy people, mostly before retirement, voluntarily visit and take medical check-up once a year. As a result, the age and smoking history were significantly higher in the case group than the control group. When we analyzed the association of the rs11536889 SNPs with the emphysema phenotype, logistic regression was performed to adjust for age and smoking history as potential confounders. However, the significances in comparisons of the allelic frequency, genotype distribution and allele positivity of the rs11536889 between the emphysema and control smokers were not apparent. Logistic regression model allows adjustment for confounders with a representative random sample from the targeted study population assuming without bias [40], but that tends to systematically overestimate odds ratios when the sample size is less than about 500 [40]. Since our case group of emphysema phenotype was selected from the COPD population and the sample size was relatively small, we don't think logistic regression is suitable for the current adjustments. The odds ratio of 1.69 with 95% CI from 1.14 to 2.50 regarding the minor C allele and the odds ratio of 2.60 with 95% CI from 1.17 to 5.78 regarding the CC genotype of the rs11536889 provided a positive hint that the rs11536889 is important in the association with emphysema phenotype. With increasing sample size, the significances are supposed to approach the true population values [40]. We would expect replication studies for the association of the rs11536889 with emphysema in other Japanese groups. Further studies using appropriate animal models may verify the roles of TLR4 rs11536889 polymorphism in the pathogenesis of emphysema.

This study demonstrated that the rs11536889 (+3725 G/C) SNP and the haplotype that carried the minor C allele of this SNP in the TLR4 gene were likely associated with the risk of developing the emphysema phenotype in the Japanese population.