Abstract

Background. The factors that predispose to pulmonary tuberculosis (PTB) are not fully understood. Previous studies have shown that cytokine gene polymorphisms were associated with PTB. Objectives. In this study, we have investigated the relationship between ILB, IL6, and TNFα polymorphisms and a predisposition to Mycobacterium tuberculosis (MTB) infection and PTB. Methods. A total of 209 cases of PTB, 201 subjects with latent TB infection (LTBI), and 204 healthy controls (HCS) were included in this study. Logistic regression analyses under allelic, homozygous, and heterozygous models were used to calculate values, odds ratios (ORs), and 95% confidence intervals (CIs) for assessing the association between single nucleotide polymorphisms (SNPs) and disease risk, adjusting for sex and age. Genotyping was conducted using the improved multiplex ligase detection reaction (iMLDR) method. Results. When comparing PTB patients with LTBI subjects, significant associations with disease development were observed for SNPs of IL6 and TNFα. When comparing LTBI subjects with HCS, IL1B polymorphisms were significantly associated with LIBI. Haplotype analyses suggested that the CGG haplotype of IL1B was associated with an increased risk of PTB (, OR = 1.34, 95% CI: 1.01–1.76), while the TTGCG haplotype of TNFα was a protective factor against PTB (, OR = 0.66, 95% CI: 0.44–0.98). Conclusion. Our study demonstrated that IL1B variants were related to LTBI and IL6 and TNFα variants were associated with PTB.

1. Introduction

Tuberculosis (TB) is still a major cause of mortality and morbidity in developing countries. In 2015, there were an estimated 9.2 million new TB cases [1]. India, Indonesia, and China had the largest number of patients with TB: 23%, 10%, and 10% of the global total, respectively. It is reported that one-third of the world’s population is infected with Mycobacterium tuberculosis (MTB); however, only 10% of the MTB infected population will develop TB disease during their lifetime. TB is mainly caused by MTB that primarily invades the lung phagocytes such as macrophages, neutrophils, monocytes, and dendritic cells [2]. In 90% of subjects infected with MTB (LTBI), the ensuing immune response arrests further the development of TB disease. However, the predisposing factors for TB remain unclear. Previous studies have demonstrated that various host genetic polymorphisms play important roles in determining TB susceptibility [3, 4].

Investigating the various host genes involved in susceptibility and resistance to TB may enhance understanding of the pathogenesis of TB and development of treatment or prophylactic strategies.

Previous studies have increased our understanding of the roles of host genetics in susceptibility to MTB infection [4]. There are several cytokines and their receptors expressed by macrophages that are important for host responses to TB. They include tumor necrosis factor (TNF), interleukin-1β (IL1B), and IL6 [5]. When MTB invades the lungs, these cytokines are released at the site of disease after the interaction between T lymphocytes and infected macrophages. Macrophages are essential for the induction of the immune response against MTB. When MTB enters the lungs, macrophages are among the first to sense it. After the phagocytosis of MTB by lung macrophages, the process generates an inflammatory response and antigen presentation to effector T cells [6]. Furthermore, macrophages can kill phagocytosed MTB and are involved in tissue repair as a consequence of the inflammatory process [7].

Several cytokine gene polymorphisms have been reported to affect gene transcription and influence serum cytokine production [8, 9]. Polymorphisms in cytokine gene have been suggested to play key roles in the susceptibility, development, and severity of infectious diseases [10]. Among the inflammatory cytokine genes, IL1B, IL6, and TNFα were demonstrated to be associated with TB infection or TB disease [1113].

We therefore investigate the impact of the most widely studied SNPs of the three genes on LTBI and/or PTB. Participants in our study formed three groups: healthy control subjects without MTB infection (HCS), latent TB infection (LTBI), and bacteria-confirmed active pulmonary TB disease (PTB).

2. Methods

2.1. Study Population

A total of 209 patients with PTB and 415 close contacts of people with active PTB were enrolled between 2013 and 2014 in the West China Hospital of Sichuan University, Chengdu, China. These subjects were all Chinese Han individuals. Close contacts of active PTB cases included spouses, accompanying people, and coworkers. Some medical staff of the TB ward was also included. After one year of follow-up, several close contacts developing into PTB were excluded. We divided the remaining 405 close contacts into LTBI and HCS according to interferon gamma release assay (IGRA) results at enrolment, chest X-ray, and symptoms at enrolment and at one-year follow-up.

The definition of close contacts is as follows: (1) shared airspace with PTB for at least 15 hours per week for at least one week during an infectious period and (2) shared airspace with a PTB patient for at least 180 hours during an infectious period. The infectious period was 3 months before the collection of the first culture-positive sputum or the date of first signs of cough (whichever was earlier) until two weeks after the TB combination treatment therapy. LTBI controls had no TB-related symptoms but a positive IGRA test. HCS without MTB infection were IGRA negative and without evidence of active TB and history of TB. PTB patients were diagnosed by sputum positive and/or through bronchoscopic biopsy and pathological examinations. IGRA was performed according to the manufacturer’s instructions (Beijing Wantai DRD Co., Ltd., Beijing, China). All participants in our study were older than 18 years and younger than 80 years. All participants with blood relationship were excluded from this genetic association study. Subjects with autoimmune diseases, chronic obstructive pulmonary disease, diabetes, hepatitis, and those tested positive for HIV were excluded from the study. All participants were required to provide their written informed consent to participate in this study. Specialized nurses drew 2–5 mL of blood with anticoagulant from all participants and then the blood was preserved in an 80°C freezer for DNA extraction and genotyping.

2.2. SNP Selection and Genotyping

SNPs were selected based on a literature review of previous studies and in silico functional prediction from the NIH FuncPred website (https://snpinfo.niehs.nih.gov/snpinfo/snpfunc.html). Only those SNPs with potential effects on function and significant disease associations were selected in our study. As a result, there were 11 SNPs from 3 target genes that were selected for this study. Among the 11 SNPs, rs1143623 [14], rs2069837 [15], rs1799724 [16], rs1800629 [16], and rs1800630 [16] were associated with transcriptional activity. rs1143634 [17], rs16944 [18], and rs1800795 [19] polymorphisms were found to regulate cytokine levels. rs1799964 [20] and rs361525 [21] were found to be contributory factors to TNFα gene expression; rs17147230 was predicted to have functional significance by the FuncPred algorithm.

Genomic DNA was isolated using a genomic DNA purification kit (Axygen Scientific Inc., Union City, CA, USA) according to the manufacturer’s instructions. Genotyping was conducted using the improved multiplex ligase detection reaction (iMLDR), with technical support from the Shanghai Genesky Biotechnology Company. To validate the genotype results, 5% of the samples were repeated by iMLDR.

2.3. Statistical Analysis

Hardy-Weinberg equilibrium (HWE) was estimated using the test. Differences in age between cases and controls were compared by an unpaired -test. Differences in the distribution of sex between the two groups were assessed using the test. Multiple logistic regression under allelic (minor allele versus major allele), homozygous (homozygous minor genotype versus homozygous major genotype), and heterozygous (heterozygous genotype versus homozygous major genotype) models was used to calculate values, odds ratios (ORs), and 95% confidence intervals (CIs) for assessing the association between SNPs and disease risk, adjusting for sex and age. Haplotype association was calculated by SHEsis online software (http://analysis.bio-x.cn/). values < 0.05 were considered statistically significant. Statistical analyses were performed using SPSS (SPSS Inc., Chicago, IL, USA) 17.0 software.

3. Results

3.1. Demographic and Clinical Characteristics of Study Subjects and Results of Quality Control

The study subjects consisted of 209 PTB patients, 201 LTBI, and 204 HCS (Table 1). The distributions of sex in two groups had no significant difference. However, the mean ages of the three groups were significantly different (Table 1).

The characteristics of all the SNPs were showed in Table 2. One SNP (rs1800795 of IL6) was out of HWE in the HCS group and was excluded from further analysis. The remaining 10 SNPs did not deviate from HWE in the control subjects ().

3.2. Association between 10 SNPs in the IL1B, IL6, and TNFα Genes and LTBI or TB Susceptibility

Ten SNPs in the IL1B, IL6, and TNFα genes were evaluated for associations with LTBI or PTB. The genotype distributions in LTBI were compared with HCS and PTB cases, respectively. Differences in genotype distribution between the study groups are shown in Table 3. Significant associations were found for IL1B rs1143634 between LTBI and HCS. Compared with the IL1B rs1143634G allele, allele A had an 82% decreased risk for development LTBI (, OR = 0.18, 95% CI: 0.039–0.84). The rs1143634GA genotype was also a protective factor for LTBI (, OR = 0.18, 95% CI: 0.08–0.83). We found the rs2069837 polymorphism of IL6 was associated with PTB. Compared with AA, GA had a significantly decreased risk of LTBI developing into PTB (, OR = 0.62, 95% CI: 0.4–0.98). A significantly decreased risk of PTB development was also observed for the TNFα rs1799724T allele compared with C allele (, OR = 0.64, 95% CI: 0.42–0.98).

3.3. Association between Haplotypes of IL1B, IL6, TNFα, and LTBI or TB Susceptibility

Haplotype association analyses suggested that the rs1143623-rs1143634-rs16944 CGG haplotype of IL1B was a risk factor for PTB development (, OR = 1.34, 95% CI: 1.01–1.76), while the rs1799724-rs1799964-rs1800629-rs1800630-rs361525 TTGCG haplotype of TNFα was a protective factor against PTB development (, OR = 0.66, 95% CI: 0.44–0.98) (Table 4).

4. Discussion

Most previous studies focused on the association between gene polymorphisms and TB using control subjects including LTBI and no TB infection individuals. However, few studies have identified candidate genes related to LTBI and/or TB. In this study, we designed three study groups including HCS without TB infection, LTBI, and PTB to identify risk factors for both LTBI and TB and hence find genetic markers specific for TB development stages. When comparing PTB with LTBI controls, significant relationships were detected for IL6 rs2069837 G/A and TNFα rs1799724 C/T. When comparing LTBI with HCS, significant relationships were detected for IL1B rs1143634GA.

It is well recognized that proinflammatory cytokines, including IL1B, IL6, and TNFα, have been implicated in the host defense against MTB infection. IL1B, produced by macrophages and monocytes, is involved in the immune response and is an important proinflammatory cytokine [22]. IL1B plays a critical role in the early stages of antimycobacterial immune response and protects the tissue by activating epithelial antimicrobial peptides [23]. IL1B not only regulates some immune cells but also stimulates expression of other inflammation-associated proteins [24]. Previous studies have revealed that high levels of IL1B were detected in alveolar macrophages from PTB patients and these levels were related to significant tissue necrosis in lung lesions of PTB cases with a large cavity [25]. Thus, polymorphisms of IL1B have become biologically probable candidate markers associated with PTB susceptibility. Association studies of IL1B polymorphisms and TB have previously been conducted. Hall et al. researched the relationship between 29 candidate genes related to innate immune responses and suggested that rs1143643 and rs1143633 in IL1B were associated with TB [13]. A meta-analysis including 3327 participants indicated that rs16944 and rs1143634 alleles/genotypes in IL1B have no relationship with TB [26]. However, these studies only researched the genetic polymorphisms and the development of TB utilizing controls without differentiating HCS from LTBI. In the present study, we identified rs1143634 allele/genotype was associated with LTBI development compared with HCS but not with LTBI. This inconsistent result may be attributed to the different type of control group. A previous study suggested that the use of different controls could result in contrary conclusions and it is important to divide the controls into LTBI and HCS to understand the role of these genes in TB pathogenesis and progression [4].

IL6 is a pleiotropic proinflammatory cytokine, and high levels of IL6 were observed in TB patients [27]. To show the importance of IL6 in TB, murine models with mycobacterial infections and models with a TB subunit vaccine indicated that IL6 was essential for the development of an optimal T-cell response [28]. IL6 mRNA and protein expression were significantly elevated in the serum of PTB patients [29]. MTB stimulates the secretion of IL6 which inhibits IFN-signaling, thus enabling MTB to evade eradication by a cellular immune response [30]. Previous studies have investigated the relationship between IL6 polymorphisms and TB; however, the results were inconsistent [30, 31]. A meta-analysis indicated that IL6 rs1800795 was associated with decreased risk of TB [32]. Unfortunately IL6 rs1800795 diverged from the HWE in our data and as a result was excluded from our study. Our study revealed that rs2069837GA was associated with decreased risk of TB. Moreover, our study revealed that this polymorphism was associated with PTB when compared with LTBI. SNP rs2069837 is located within the second intron of IL6 and was not in linkage disequilibrium (LD) with any other SNPs in our study. LD was also not found in the LD data from the HapMap project (http://hapmap.ncbi.nlm.nih.gov/). This explains why only a single variant of IL6 was found as a protective factor for PTB. rs2069837 is located within a regulatory genetic region [33], indicating this genetic variant may modify risk of PTB. The rs2069837 polymorphism has been suggested to be related to inflammatory diseases such as Takayasu arteritis [33]. Consistent with this observation, our results indicated that this genetic variant was related to PTB which is also an inflammatory disease. However, we demonstrated that this SNP was a protective factor for PTB. This phenomenon may be due to the different MAF between the two ethnic groups.

When infected with MTB, a strong inflammatory cell-mediated immune response appears, with elevated expression TNFα [34]. TNFα was not only an important component of the immune response to TB but was also involved in granuloma formation [35]. There have been a number of studies of the association of TNFα polymorphisms and TB, however; the results were inconsistent [36]. It was reported that the TNFα rs1800629 and rs361525 polymorphisms were not related to TB [36, 37]. These two genetic variants were also found to have no relationship with PTB/HCS in our study. Our study demonstrated that the TNFα rs1799724 allele T was associated with decreased risk of PTB. Since TNFα rs1799724T had higher transcriptional activity and an increase in TNFα cytokine production [16], we speculated this genetic variant was associated with TNFα gene expression. Lindenau et al. showed that the TNFα rs1799724 polymorphism had no significant difference in allele/genotype distribution between TB patients and non-TB controls. However, they observed an association between tuberculin skin test (TST) reactivity and haplotypes derived from TNFα rs1799724 [38]. Further studies are warranted to verify our results.

Previous studies have mainly focused on the association between gene polymorphisms and TB. However, most of these investigations employed healthy controls without identifying MTB infection status. Recent research has found that some genes may be associated with LTBI but not active TB [36, 39], and our results have also shown that the IL1B polymorphism was only associated with LTBI. In addition, other studies have reported that some genes may differentiate between LTBI and active TB disease [40, 41]. The aforementioned results are important in order to further understand the important role of SNPs in TB progression. Such information may also help in the differentiation between LTBI and PTB and hence affect therapeutic regimen. We reasoned that the genetic susceptibility to TB is important not only for the general population but also for the patients with autoimmune diseases under immunosuppressive treatments such as treatment with anti-TNFα or anti-IL6 antibodies. However, most researchers excluded participants with autoimmune diseases from their study [36], and thus it is difficult to directly evaluate the role of genetic variants in the diagnosis of LTBI for patients under immunosuppressive treatment. Therefore, more studies regarding the association of SNPs with LTBI in patients treated with such biological therapies are needed to verify our results.

Nevertheless, our research has several limitations. First, functional validation of the associated SNPs was not conducted. As a result, the true causal allele underlying the genetic association result is still unknown. Second, the results have not been validated in a replication cohort, which may increase the chance of type I errors. And finally, we did not analyze the association between SNPs and detailed clinical characteristics.

In conclusion, our study demonstrated that IL6 and TNFα genetic polymorphisms were significantly associated with susceptibility to PTB in LTBI subjects. We also identified that IL1B polymorphisms were associated with LTBI. Our results may reflect the genetic aspects of a predisposition to different stages of developing PTB and these results warrant verification by larger studies.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

This work was supported by the Research Fund for the Doctoral Program of Higher Education of China (Grant no. 20130181110068), the National Natural Science Foundation of China (Grants nos. 81170042 and 81370121), and the National Scientific and Technological Major Project of China (Grant no. 2012ZX10004-901).