TLR4/NF-κB signaling pathway gene single nucleotide polymorphisms alter gene expression levels and affect ARDS occurrence and prognosis outcomes.

BACKGROUND: To study the occurrence and prognosis of acute respiratory distress syndrome (ARDS) using single nucleotide polymorphisms (SNPs) of TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci in the TLR4/NF-κB pathway.
METHODS: Genotypes were analyzed for TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci. Plasma TNF-α and IL-6 levels and MyD88 mRNA expression in peripheral blood mononuclear cells (PBMCs) of 300 ARDS patients and 300 non-ARDS patients (control group) were examined. The patients were followed up for 60 days, and the prognosis outcome was recorded.
RESULTS: The TNF-α rs1800629 locus A allele and the IL-6 rs1800796 locus G allele were found to be risk factors for ARDS (adjusted OR = 1.452, 95% CI: 1.211-1.689, P < .001 and adjusted OR = 1.205, 95% CI: 1.058-1.358, P = .005, respectively). The G allele at MyD88 rs7744 locus was a protective factor against ARDS (adjusted OR = 0.748, 95% CI: 0.631-0.876, P < .001). Compared with the other groups, homozygotes for TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci had higher expression levels, of which homozygotes for TNF-α rs1800629 and IL-6 rs1800796 loci had lower 60-day survival rates, while MyD88 rs7744 locus homozygotes had a higher 60-day survival rate.
CONCLUSION: The effect of TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 SNPs on gene expression level is a likely cause of ARDS occurrence and poor prognosis.

Introduction

Acute respiratory distress syndrome (ARDS) is a respiratory failure characterized by progressive dyspnea and refractory hypoxemia. It is also the final stage of acute lung injury, with a high mortality rate.[ According to previous research, the incidence of ARDS in China is high, with an annual mortality rate of 52%.[

From the perspective of pathophysiology, the occurrence of ARDS is associated with abnormal inflammatory reactions, interstitial and alveolar edema, cell infiltration into the alveolar space, and endothelial damage.[ The Toll-like receptor 4 (TLR4)/nuclear factor (NF)-κB pathway may be a key target for inflammatory damage. TLR4 is a pattern recognition receptor from the TLR protein family expressed in lung cells, which activates NF-κB and induces the production of inflammatory cytokines and chemokines such as TNF-α and IL-6.[ Myeloid differentiation factor 88 (MyD88) is a key adaptor of TLR4, which leads to activation of downstream NF-κB and subsequent production of pro-inflammatory cytokines associated with neurotoxicity.[

This study investigated the effects of genetic variation in TNF-α, IL-6, and MyD88 genes on the occurrence and prognosis of ARDS in the TLR4/NF-κB pathway. The three SNP loci for TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 are located in the non-coding region of the respective genes (Table 1). These non-coding regions may be involved in the regulation of gene expression and have been shown to be sensitive to disease sites.[ This study explored the correlation between the occurrence and prognosis of these three SNPs and ARDS and analyzed the possible underlying mechanisms at the level of gene expression.

Table 1

SNP loci information.

Data and methods

General data

From May 2015 to May 2017, 300 patients with ARDS who were admitted to the ICU of the Second Affiliated Hospital of Zhejiang Chinese Medical University, Hangzhou Lin’an District People's Hospital, the Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, and Department of Emergency, the Second Affiliated Hospital of Zhejiang Chinese Medical University were enrolled in this study, including 185 men and 115 women, aged 34 to 84 years. During the same period, based on the age and sex of the patients in the ARDS group, 300 non-ARDS patients admitted to the ICU were enrolled as a control group. The inclusion criteria were

one or more risk factors for ARDS, such as sepsis (without shock), Septic shock, Pneumonia, Aspiration, Multiple transfusion, and Trauma, and

consistent diagnosis of ARDS patients with Berlin's mild, moderate, or severe ARDS criteria.[

The criteria for exclusion of patients were

age < 18 years,

other chronic lung diseases except asthma and chronic obstructive pulmonary disease (COPD),

diffuse alveolar hemorrhage,

directive to withhold intubation,

prior treatment with granulocyte colony-stimulating factor, and

immunosuppression not secondary to corticosteroid.

The study protocol was reviewed by an ethics committee, and all the subjects signed informed consent.

Genotyping

A 5-mL sample of peripheral venous blood was collected from all subjects, and within 30 minutes of the collection, the samples were centrifuged at 1500 × g for 25 minutes. The plasma was separated, and the samples were stored at −80°C in a refrigerator for testing. Peripheral blood mononuclear cells (PBMCs) were separated after centrifugation, Genomic DNA was extracted from PBMC using Puregene DNA Isolation Kit (Gentra Systems, Minneapolis, MN). The experiment was carried out according to the supplier's instructions. The extracted genomic DNA samples were sequenced by Shanghai Shenggong Bioengineering Co. Ltd. The primer sequences for TNF-α rs1800629 locus were Fw: 5′-CGG GTC AGA ATG AAA GAA GA-3′ and Rv: 5′-CTC ATC TGG AGG AAG CGG TAG-3′. The primer sequences for IL-6 rs1800796 locus were Fw: 5′-CTG CAC GAA ATT TGA GGA TGG C-3′ and Rv: 5′-CTT CTG TGT TCT GGC TCT CCC-3′. The primer sequences for MyD88 rs7744 locus were Fw: 5′-CCA AAC TCT GGA AAG GAC CCA-3′ and Rv: 5′-GCT CTC TTC CTC TCT CTG TGC-3′. All the samples were tested twice, and the results were identical for both trials (Fig. 1).

Figure 1

SNP site sequencing results. A is the sequencing result for different genotypes of TNF-α rs1800629 locus, B is the sequencing result for different genotypes of IL-6 rs1800796 locus, and C is the sequencing result for different genotypes of MyD88 rs7744 locus.

Detection of plasma TNF-α and IL-6 levels

Plasma TNF-α and IL-6 levels were measured using an ELISA kit (E-EL-H0109 for TNF-α and E-EL-H0102 for IL-6, Wuhan Elabscience Biotechnology, Wuhan, China). A semi-automated chemiluminescence enzyme-linked immunosorbent assay system was used for detection (Tecan Group Ltd, Salzburg, Austria). Each sample was tested three times, and the averages are reported.

MyD88 mRNA detection in PBMCs

Total RNA was extracted from PBMC using TRIzol (Thermo Fisher Scientific, Inc., Waltham, MA) and reverse transcribed into cDNA using the PrimeScriptTM RT reagent kit. The relative expression of MyD88 mRNA was detected using ABI 7900HT Real-Time PCR System (Applied Biosystems; Thermo Fisher Scientific, Inc.) with SYBR Premix Ex TaqTM II kit (Takara Bio, Inc.). The primer sequences for MyD88 mRNA were Fw: 5′-GCA CAG AGA GAG GAA GAG AGC-3′ and Rv: 5′-TGC TCT GGG AAG GAG AGA GG-3′. The primer sequences for GAPDH were Fw: 5′-CAC CCA CTC CTC CAC CTT TG-3′ and Rv: 5′-CCA CCA CCC TGT TGC TGT AG-3′. The expression level of MyD88 mRNA relative to GAPDH was analyzed using the 2-ΔΔCq method.

Follow-up

All the patients were followed up for 60 days and their survival data were recorded.

Statistics

SPSS20.0 software (IBM, Chicago, IL) was used for statistical analysis. The continuous variables are reported as mean ± SD. Independent sample t-test was used for statistical analysis. The categorical variables are expressed as percentage [n (%)], and the differences between groups were assessed by χ test. Hardy-Weinberg equilibrium was evaluated using χ test. The correlation between SNPs and ARDS risk was determined based on the distribution of allele frequencies and genetic models (dominant and recessive models). Odds ratio (OR) and 95% confidence interval (CI) were used in unconditional logistic regression analysis to adjust factors such as age and sex. Multifactor Dimensionality Reduction (MDR) was used to analyze gene-gene interactions.[ The correlations between TNF-alpha rs1800629, IL-6 rs1800796, MyD88 rs7744 SNP and 60-day progression-free survival (PFS) were evaluated by Kaplan-Meier curve and logarithmic rank test. All the tests were double-tailed, Kappa > 0.8, and P < .05 indicated significant difference.

Results

Demographic characteristics of ARDS and control groups

In this study, 300 patients with ARDS were selected based on the screening conditions. Based on the age and sex distribution within the ARDS group, 300 non-ARDS patients were selected as the control group. The demographic characteristics of ARDS patients and healthy subjects are presented in Table 2, in which ARDS patients had a higher APACHE III score, and the incidence of septic shock and pneumonia was significantly higher in this group than in the control group (P < .05). The proportion of patients with sepsis who developed shock was significantly lower than that of the control group (P < .05).

Table 2

Demographic characteristics.

Correlation analysis between genetic polymorphism and ARDS risk

The genotypic distributions for TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci were consistent with the Hardy-Weinberg equilibrium (P > .05) (Table 3). The risk of ARDS in TNF-α rs1800629 locus homozygotes (AA) was significantly higher than that in the wild type (GG) (adjusted OR = 2.041, 95% CI: 1.683–2.162, P < .001). The risk of ARDS in the dominant model and implicit model was significantly increased (adjusted OR = 1.452, 95% CI: 1.211–1.689, P < .001/adjusted OR = 1.988, 95% CI: 1.644–2.103, P < .001). The population carrying TNF-α rs1800629 locus A allele was more prone to developing ARDS than the population carrying the G allele (adjusted OR = 1.452567, 95%CI: 1.373–1.741, P < .001). The population carrying the homozygous IL-6 rs1800796 locus (GG) was more prone to ARDS than the wild type (CC) (adjusted OR = 1.520, 95% CI: 1.229–1.773, P < .001), and the risk of ARDS with a dominance model was significantly high (adjusted OR = 1.568, 95% CI: 1.275–1.815, P < .001). Heterozygotes (CG) and recessive models did not have a significant risk of ARDS (P > .05). The IL-6 rs1800796 G allele was a risk factor for ARDS (adjusted OR = 1.205, 95% CI: 1.058–1.358, P = .005). MyD88 rs7744 locus heterozygotes and dominant model carriers were at a lower risk of developing ARDS than the wild-type (adjusted OR = 0.497, 95% CI: 0.373–0.646, P < .001/adjusted OR = 0.619, 95% CI: 0.501–0.756, P < .001), and the G allele at the MyD88 rs7744 locus was a protective factor against ARDS (adjusted OR = 0.748, 95% CI: 0.631–0.876, P < .001).

Table 3

Correlation between ADRS risk and allele frequencies of TNF-αrs1800629, IL-6 rs1800796, and MyD88 rs7744.

Interaction of SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 with respect to ARDS risk

MDR was used to analyze the correlation of SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 with respect to ARDS risk (Fig. 2). The results showed that an SNP at the MyD88 rs7744 locus had the strongest effect on ARDS risk, (vertex effect: 4.54%), followed by SNP at TNF-α rs1800629 (vertex effect: 4.29%) and finally IL-6 rs1800796 (vertex effect: 2.21%) (Table 4). Gene-gene interaction analysis showed that there was a redundancy between TNF-α rs1800629 and IL-6 rs1800796 (edge effect: −0.48%), whereas a synergistic interaction was observed between TNF-α rs1800629 and MyD88 rs7744, as well as between IL-6 rs1800796 and MyD88 rs7744 (Edge effect: 0.59% and 0.62% respectively) (Fig. 2A). The Dendogram results also showed that the interaction between rs1800796 and MyD88 rs7744 was the strongest (Fig. 2B).

Figure 2

MDR analysis of interaction of SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 with respect to ARDS risk. A: Circle graph, red line represents synergistic interaction, blue line represents redundancy, values represent the interaction strength, and the values at the vertices represent the impact on ARDS risk. B: Dendogram. The closer the distance, the stronger the interaction between SNP sites and the farther the distance, the weaker the interaction between SNP sites.

Table 4

Stratified analysis of the relationship between TNF-α rs1800629 SNP and ARDS risk.

Stratified analysis of the relationship between genetic polymorphism and ARDS risk

According to stratified analysis of age (≤ 60, > 60) and sex (men, women), the increased ARDS risk in TNF-α rs1800629 A allele carriers (GA/AA) was found only in patients ≤ 60 years old (adjusted OR = 1.468, 95% CI: 1.170–1.759, P = .001). TNF-α rs1800629 SNP was not associated with ARDS risk in patients > 60 years of age (adjusted OR = 1.404, 95% CI: 0.986–1.809, P = .060). Similarly, an elevated ARDS risk in A allele carriers (GA/AA) was observed only in male patients (adjusted OR = 1.548, 95% CI: 1.230–1.843, P < .001), and not in female patients (adjusted OR = 1.315, 95% CI: 0.943–1.706, P = .107) (Table 4). The correlation between IL-6 rs1800796 SNP and ARDS risk was not affected by age or sex (Table 5). MyD88 rs7744 locus G allele carriers (AG/GG) had a reduced risk of ARDS only for patients aged ≤ 60 years (adjusted OR = 0.504, 95% CI: 0.383–0.653, P < .001). The MyD88 rs7744 locus G allele carriers (AG/GG) aged > 60 did not have a significant ARDS risk (P > .05), while MyD88 rs7744 locus G allele was a protective factor against ARDS in both male and female patients (adjusted OR = 0.586, 95% CI: 0.450–0.778, P < .001 / adjusted OR = 0.651, 95% CI: 0.464–0.892, P = .006) (Table 6).

Table 5

Stratified analysis of the relationship between IL-6 rs1800796 SNP and ARDS risk.

Table 6

Stratified analysis of the relationship between MyD88 rs7744 SNP and ARDS risk.

Correlation between SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci and gene expression levels

Analysis of TNF-α and IL-6 mRNA levels in plasma and MyD88 mRNA levels in PBMCs showed that the TNF-α and IL-6 mRNA levels in the plasma of ARDS patients were higher than those in the control group, whereas the MyD88 mRNA level in PBMCs was significantly lower than that in the control group (P < .001) (Fig. 3). One-way ANOVA showed that the plasma TNF-α level of TNF-α rs1800629 GG locus genotype was significantly lower than that of GA genotype, and the AA genotype was the highest (P < .05). The plasma TNF-α level of subjects with IL-6 rs1800796 site CC genotype was significantly lower than that of CG genotype subjects, and the GG genotype was the highest (P < .05). The MyD88 mRNA level of subjects with MyD88 rs7744 locus AA genotype was significantly lower than that of AG genotype, and subjects with the GG genotype was the highest. (P < .05) (Fig. 4).

Figure 3

Comparison of TNF-α and IL-6 levels in plasma and MyD88 mRNA levels in PBMCs between ARDS and control groups. A is plasma TNF-α level, B is plasma IL-6 level, and C is MyD88 mRNA level in PBMCs. The difference between the ARDS group and the control group was analyzed by t test. The data were expressed as standard deviation and maximum and minimum values.

Figure 4

Correlation of SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci with the gene expression levels. A is the correlation between TNF-α rs1800629 SNP and plasma TNF-α levels. B is the correlation between IL-6 rs1800796 SNP and plasma IL-6 levels. C is the correlation between MyD88 rs7744 locus SNP and MyD88 mRNA level in PBMCs.

SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci affect survival prognosis

The 60-day mortality (11.39%) of ARDS patients with TNF-α GG genotype at rs180062 was significantly lower than that of TNF-α GA genotype and AA genotype (22.86%, 39.13%) (P = .038) (Fig. 5A). The 60-day mortality of ARDS patients with IL-6 rs1800796 site CC genotype (10.00%) was lower than that of IL-6 rs1800796 CG genotype and GG genotype (25.00%, 26.32%) (P = .015) (Fig. 5B). The 60-day mortality of ARDS patients with MyD88 rs7744 locus AA genotype (18.42%) was higher than that of AG and GG genotypes (10.26%, 3.23%) (P = .041) (Fig. 5C).

Figure 5

Correlation between SNPs at TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci and survival prognosis of ARDS patients. A is the survival curve of patients with different genotypes for TNF-α rs1800629. B is the survival curve of patients with different genotypes for IL-6 rs1800796. C is the survival curve of patients with different genotypes for MyD88 rs7744. The Log-rank test was used to assess the correlation between TNF-α rs1800629, IL-6 rs1800796, MyD88 rs7744 SNP and 60-day progression-free survival (PFS).

Discussion

Acute respiratory distress (ARDS) is the development of acute lung injury (ALI), a common critical illness in the intensive care unit (ICU), with a mortality rate of 30% to 50%[2]. ARDS is a transitional, uncontrolled inflammatory response in the lung caused by a variety of factors. The primary clinical feature is acute and hypoxic respiratory failure.[ In recent years, studies have found that systemic inflammatory response syndrome (SIRS), caused by infection, trauma, and other factors, is the fundamental cause of ARDS.[ Therefore, studies on pathways associated with inflammatory responses may be of great importance for the prevention and treatment of ARDS.

Among the many cytokines involved in ALI, TNF-α is an early-reactive cytokine that promotes further inflammatory responses, so anti-TNF-α preparations constitute the first form of medication in ALI/ARDS clinical trials.[ TNF-α initiates a downstream signaling pathway by binding with two receptors, p55 and p57, which play different roles in the formation of ARDS pulmonary edema. Studies have shown that the deletion of p55 has a protective effect against the acute phase of ARDS induced by mechanical ventilation, whereas the effect of p57 deletion is opposite,[ demonstrating that TNF-α plays an important role in the development of ARDS. Very few studies have investigated the effect of TNF-α genetic variation on the occurrence of ARDS. In contrast with the results of this study, Azevedo et al[ had reported that TNF-α rs1800629 locus A allele has protective effects against ARDS in the Brazilian population. The researchers reported that the A allele frequency was 13% among the Brazilian population unaffected by ARDS, significantly higher than that in the Chinese Han population (5.71%), which may be one of the reasons for the difference in the results. In addition, the impact of environmental factors cannot be ignored.

IL-6 is a cytokine with multiple immunoregulatory functions and exerts a pro-inflammatory effect in the development and progression of ALI, in addition to having an anti-inflammatory effect.[ Recent studies indicate that IL-6 can enhance the damaging effect of neutrophils on tissues, in addition to participating in the body's immunosuppression response and delaying the apoptosis of neutrophils, which is a key factor in the transformation of innate immunity to acquired immunity. It also plays an important role in the occurrence and development of chronic inflammation.[ The genetic variation for IL-6 is associated with diseases caused by abnormal inflammatory responses.[ Studies have shown that IL-6 expression levels are abnormally elevated after acute inflammatory response,[ so we believe that the abnormal expression of IL-6, caused by IL-6 gene polymorphism, may be related to the occurrence and development of ARDS. It can be seen from the results of this study that the G allele at IL-6 rs1800796 is a risk factor for ARDS, and the plasma IL-6 level in the population carrying the G allele is significantly higher than in the C allele carrier population, which may be one of the main causes of ARDS.

MyD88 is a key linker in the TLR4 signaling pathway and plays an important role in upstream signaling and the development and progression of disease.[ The TLR4-mediated MyD88 signaling pathway regulates the expression of multiple genes involved in inflammation during the immune response to pathogen invasion.[ The MyD88 gene encodes a cytosolic adaptor protein that plays an important role in both innate and adaptive immune responses, as an essential signal transducer in IL-1 and TLR signaling pathways. Activation of MyD88/NF-κB signaling accelerates the inflammatory response.[ Studies have shown that the single nucleotide polymorphism at the rs7744 locus of MyD88 gene is closely related to the susceptibility of ulcerative colitis.[ This SNP locus is located in the 3’-UTR of the gene and may be involved in the regulation of gene expression. The results of this study showed that the G allele at MyD88 rs7744 locus is a protective factor against ARDS, and the MyD88 level in PBMCs in G allele carriers is higher than that in A allele carriers, which may be the reason for the decreased risk of ARDS in the presence of G allele.

In addition, we analyzed the correlation between TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 SNPs and survival prognosis in patients with ARDS. The results showed that the 60-day survival rate of wild-type ARDS patients with TNF-α rs1800629 and IL-6 rs1800796 was significantly higher than that of mutants, while that of wild-type ARDS patients with MyD88 rs7744 was significantly lower than that of mutants, indicating that TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 locus SNPs are significantly associated with survival prognosis in patients with ARDS. We believe that the cascade of reactions may be caused by inconsistent expression levels of inflammatory factors in subjects with different genotypes at SNP sites.

This study had some limitations. First, the occurrence and development of ARDS is likely the result of interactions between genes and environmental factors. Thus, the influence of single SNPs may be limited. Therefore, this study aimed to analyze the effects of SNPs on ARDS in a population with individuals of different ages and sex. The results showed that the correlation between SNP at TNF-α rs1800629 and ARDS was influenced by age and sex, whereas the effect of SNP at IL-7 rs1800796 was not dependent on age and sex. The effect of MyD88 rs7744 locus SNP on ARDS risk was age-related. There may be other factors involved, and further research is needed to uncover their effects. Second, there may be multiple SNP loci involved in the occurrence and development of ARDS. Thus, it may be more fruitful to study the combined effects of SNPs at different genetic loci. The results of this study indicate that TNF-α rs1800629 and IL-6 rs1800796 share redundant effects, whereas interactions between TNF-α rs1800629 and MyD88 rs7744, as well as IL-6 rs1800796 and MyD88 rs7744 are synergistic in nature. This suggests that there may be redundancy across some sites, and synergistic interaction across others.

Conclusion

The single nucleotide polymorphisms of TNF-α rs1800629, IL-6 rs1800796, and MyD88 rs7744 loci are associated with poor occurrence and prognosis of ARDS, likely due to the effect of gene expression levels. The combined effects of multiple genes and environmental factors may require further research to reveal the intrinsic and extrinsic mechanisms of ARDS development and progression, for better prevention and treatment of ARDS.

Author contributions

Data curation: Yueping Ding, Qijia Feng, Jianshi Chen.

Formal analysis: Yueping Ding, Qijia Feng, Jia Song.

Investigation: Yueping Ding.

Methodology: Yueping Ding, Jia Song.

Project administration: Yueping Ding, Jia Song.

Software: Qijia Feng, Jianshi Chen.

Supervision: Jia Song.

Visualization: Qijia Feng, Jianshi Chen.

Writing – original draft: Yueping Ding.

Writing – review & editing: Jia Song.