Association of a Toll-like receptor 1 polymorphism with heightened Th1 inflammatory responses and antibiotic-refractory Lyme arthritis

Abstract

Objective: Single-nucleotide polymorphisms (SNPs) that alter immune function, inflammatory responses, and disease susceptibility have been identified in several genes encoding Toll-like receptors (TLRs). The TLR SNPs with the best evidence of an effect on immune function are those in TLR1 (1805GG), TLR2 (2258GA), and TLR5 (1174CT). This study was undertaken to assess the frequency and functional outcomes of these polymorphisms in patients with Lyme disease.

Methods: SNP frequencies and functional outcomes were assessed in 248 patients with Lyme disease. Cytokine and chemokine levels were determined using multiplex assays in the serum of patients with erythema migrans (EM), joint fluid of patients with Lyme arthritis, and supernatants of Borrelia burgdorferi-stimulated peripheral blood mononuclear cells (PBMCs) from patients with Lyme arthritis.

Results: The frequency of the TLR1-1805GG polymorphism was greater in patients with antibiotic-refractory arthritis compared with patients with EM or those with antibiotic-responsive arthritis. Early in the illness, patients with EM carrying 1805GG, primarily those infected with B burgdorferi 16S-23S ribosomal spacer RNA intergenic type 1 (RST1) strains, had higher serum levels of interferon-γ (IFNγ), CXCL9, and CXCL10 and had more severe infection than EM patients carrying the 1805TG/TT polymorphism. These inflammatory responses were amplified in patients with Lyme arthritis, and the highest responses were observed in patients with 1805GG in the antibiotic-refractory group who had been infected with RST1 strains. When PBMCs from patients with Lyme arthritis were stimulated with a B burgdorferi RST1 strain, the 1805GG group had a significantly larger fold increase in the levels of IFNγ, CCL2, CXCL9, and CXCL10 compared to the 1805TG/TT group. In contrast, the TLR2 and TLR5 polymorphisms did not vary in frequency or function among the groups.

Conclusion: The TLR1-1805GG polymorphism in B burgdorferi RST1-infected patients was associated with stronger Th1-like inflammatory responses, an environment that may set the stage for antibiotic-refractory arthritis.

Figure 1

Frequency of selected TLR polymorphisms in patients with Lyme disease. Frequencies of polymorphisms in panel A, TLR1 (T1805G), panel B, TLR2 (G2258A), and panel C, TLR5 (C1174T) are shown in 248 patients with Lyme disease, 71 with EM, 76 with antibiotic-responsive, and 101 with antibiotic-refractory Lyme arthritis and in normal Caucasian populations. As reported in the literature, the 46 normal subjects for TLR1 were from Seattle (15); the 349 normal subjects for TLR2 were from Germany (17), and the 505 normal subjects for TLR5 were primarily from the Netherlands (22). Each polymorphism was determined by PCR amplification and restriction fragment length polymorphism (RFLP) techniques.

Figure 2

Cytokine and chemokine levels in serum samples from patients with EM stratified by the TLR1-1805GG polymorphism. Protein levels of 7 chemokines and 5 cytokines were determined in sera from 66 patients with EM using bead-based multiplex assays. In panel A, the patients were first stratified according to TLR1-1805GG or 1805TG/TT genotypes. In panel B, the patients were further subdivided based on the B. burgdorferi RST genotype of the infecting strain. Of the 34 patients with 1805GG, 18 had RST1 infection, 14 RST2 infection, and 2 RST3 infection; of the 32 patients with 1805GT/TT, 12 had RST1 infection, 15 RST2 infection, and 5 RST3 infection. Bars show the median values and I-bars represent the third quartile values. For comparison of 1805GG versus 1805TG/TT, *P ≤ 0.05, **P ≤ 0.01.

Figure 3

Clinical correlations in patients with EM. In panel A, protein levels of 7 chemokines and 5 cytokines were compared in 66 patients according the presence or absence of symptoms. In panel B, the proportion of 66 patients with or without associated symptoms was correlated with the 1805GG polymorphism and RST genotype of the infecting strain. The differences between RST1-infected patients with 1805GG compared with RST1-, 2-, or 3-infected patients with 1805 TG/TT were statistically significant, whereas the differences among RST1-, 2-, or 3-infected patients with 1805GG were not. In panel C, the pathogen burden in EM skin lesions from 50 patients, quantified by QPCR detection of the B. burgdorferi flaB gene, was stratified by the 1805GG polymorphism and RST genotype. In panels A and C, bars show median values and I-bars represent the third interquartile range. For comparison of 1805GG versus 1805TG/TT, *P ≤ 0.05, **P ≤ 0.01.

Figure 4

Cytokine and chemokine levels in joint fluid of patients with Lyme arthritis. Protein levels of 7 chemokines and 5 cytokines were measured in joint fluid samples from 49 patients with Lyme arthritis using bead-based multiplex assays. In panel A, the patients were stratified in antibiotic-responsive or antibiotic-refractory groups. In panel B, the 10 patients with antibiotic-refractory arthritis, all of whom had 1805GG, were stratified according to the infecting strain. Because of small numbers, the RST2 and RST3 groups were combined for comparison with the RST1 group. Bars show median values and I-bars represent third interquartile ranges. For each comparison, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001.

Figure 5

Stimulation of PBMC from patients with Lyme arthritis. Protein levels of 6 chemokines and 6 cytokines in culture supernatants were assessed in PBMC from 43 patients with Lyme arthritis; 25 were 1805GG and 18 were TG/TT, using bead-based multiplex assays. Cells were stimulated with Pam3CSK4 (300 ng/ml) (panel A), or with an RST1 (OspC type A) isolate of B. burgdorferi at an MOI of 25 spirochetes per cell (panel B), in each instance for 5 days. Data are calculated as fold increase in levels of cytokines and chemokines (stimulated values divided by unstimulated values). Bars show median values and I-bars represent third interquartile ranges. For comparison of 1805GG versus 1805TG/TT, *P ≤ 0.05, ***P ≤ 0.001.