Association of human TLR1 and TLR6 deficiency with altered immune responses to BCG vaccination in South African infants

Abstract

The development of effective immunoprophylaxis against tuberculosis (TB) remains a global priority, but is hampered by a partially protective Bacillus Calmette-Guérin (BCG) vaccine and an incomplete understanding of the mechanisms of immunity to Mycobacterium tuberculosis. Although host genetic factors may be a primary reason for BCG's variable and inadequate efficacy, this possibility has not been intensively examined. We hypothesized that Toll-like receptor (TLR) variation is associated with altered in vivo immune responses to BCG. We examined whether functionally defined TLR pathway polymorphisms were associated with T cell cytokine responses in whole blood stimulated ex vivo with BCG 10 weeks after newborn BCG vaccination of South African infants. In the primary analysis, polymorphism TLR6_C745T (P249S) was associated with increased BCG-induced IFN-γ in both discovery (n = 240) and validation (n = 240) cohorts. In secondary analyses of the combined cohort, TLR1_T1805G (I602S) and TLR6_G1083C (synonymous) were associated with increased IFN-γ, TLR6_G1083C and TLR6_C745T were associated with increased IL-2, and TLR1_A1188T was associated with increased IFN-γ and IL-2. For each of these polymorphisms, the hypo-responsive allele, as defined by innate immunity signaling assays, was associated with increased production of TH1-type T cell cytokines (IFN-γ or IL-2). After stimulation with TLR1/6 lipopeptide ligands, PBMCs from TLR1/6-deficient individuals (stratified by TLR1_T1805G and TLR6_C745T hyporesponsive genotypes) secreted lower amounts of IL-6 and IL-10 compared to those with responsive TLR1/6 genotypes. In contrast, no IL-12p70 was secreted by PBMCs or monocytes. These data support a mechanism where TLR1/6 polymorphisms modulate TH1 T-cell polarization through genetic regulation of monocyte IL-10 secretion in the absence of IL-12. These studies provide evidence that functionally defined innate immune gene variants are associated with the development of adaptive immune responses after in vivo vaccination against a bacterial pathogen in humans. These findings could potentially guide novel adjuvant vaccine strategies as well as have implications for IFN-γ-based diagnostic testing for TB.

Figure 1. Polymorphism TLR6_C745T is associated with BCG-induced whole blood cytokine production.

Whole blood drawn 10 weeks after BCG vaccination of newborns was re-stimulated with BCG for 7h ex vivo and plasma levels of IFN-γ, IL-2, and IL-13 were measured. A general linear model was used to examine whether polymorphism TLR6_C745T was associated with BCG-induced cytokine levels after subtraction of unstimulated control values. IFN-γ (A), IL-2 (B), and IL-13 (C) levels are shown with unstimulated control values subtracted for each genotype of polymorphism TLR6_C745T. The number of individuals for each genotype is shown in parentheses. P values represent significant differences as assessed using the general linear model. *P<0.05, **P<0.01, ***P≤0.001.

Figure 2. Polymorphisms TLR6_G1083C and TLR1_T1805G are associated with BCG-induced whole blood cytokine production.

Whole blood drawn 10 weeks after BCG vaccination of infants was re-stimulated with BCG for 7h ex vivo and plasma levels of IFN-γ, IL-2, and IL-13 were measured. A general linear model was used to examine whether TLR polymorphisms were associated with BCG-induced cytokine levels after subtraction of unstimulated control values. IFN-γ, IL-2, and IL-13 levels are shown with unstimulated control values subtracted for each genotype of polymorphism TLR1-T1805G (A–C) and TLR6_G1083C (D–F). The number of individuals for each genotype is shown in parentheses. P values represent significant differences as assessed using the general linear model. *P<0.05, **P<0.01.

Figure 3. Haplotype tagging SNPs and pair-wise linkage disequilibrium analysis of TLR1-6-10 gene region.

A. The arrangement of TLR1 TLR6 and TLR10 on chromosome 4 and the position of the haploytype-tagging SNPs are shown. B. Linkage disequilibrium analyses between the four TLR SNPs with functional associations in this study are expressed as D' and R² for pair-wise comparisons. The minor allele frequencies (MAF) are listed in the top row of boxes and highlighted in bold.

Figure 4. TLR1 polymorphism A1188T is associated with increased Th1 cytokine production in response to BCG stimulation. A-C.

As in Figure 1, whole blood drawn 10 weeks after BCG at birth was re-stimulated with BCG for 7h ex vivo. Plasma IFN-γ (A), IL-2 (B), and IL-13 (C) levels are shown for individuals with TLR1 polymorphism A1188T (rs3923647) with the number of individuals for each genotype shown in parentheses. D. PBMCs from individuals vaccinated with BCG were stimulated with media or TLR ligands (LPS (100 ng/mL), PAM₂CSK₄ (250 ng/mL), PAM₃CSK₄ (250 ng/mL)) for 24 hours. TNF-α production was quantified by ELISA. **P<0.01, ***P≤0.001.

Figure 5. Two TLR polymorphisms are associated with increased expression of cytotoxic molecules after ex vivo BCG stimulation.

Whole blood drawn 10 weeks after BCG at birth was re-stimulated with BCG ex vivo for 3 days. Granulysin, granzyme B and perforin production were measured by intracellular cytokine staining in (A) CD4-positive and (B) CD8-positive T cells. A general linear model was used to examine whether 4 TLR polymorphisms (TLR1_T1805G, TLR1_A1188T, TLR6_G1083C, and TLR6_C745T) were associated with BCG-induced cytotoxicity markers after subtraction of unstimulated control values. *P<0.05, **P<0.01.

Figure 6. TLR1/6 polymorphisms are associated with decreased lipopeptide-induced IL-6 and IL-10 production in PBMCs.

PBMCs from individuals with polymorphisms in TLR6_C745T and TLR1_T1805G were stimulated for 20 hours with TLR ligands LPS (10 ng/mL), PAM₂CSK₄ (250 ng/mL), or PAM₃CSK₄ (250 ng/mL). Supernatant IL-6 and IL-10 levels were quantified by ELISA. TLR1 and TLR6 genotypes are designated based on functionality as: TLR1_low (genotype TLR1_1805GG, n = 22), TLR1_high (1805GT/TT, n = 25), TLR6_low (TLR6_745CT/TT, n = 24), or TLR6_high (745CC, n = 23). TLR1/6 diplotypes are designated as TLR1/6_low (TLR1_1805GG and TLR6_745CT/TT, n = 18) or TLR1/6_high (TLR1_1805GT/TT and TLR6_745CC, n = 18). Panels: For IL-6, TLR1/6_low vs high diplotypes (A), TLR6_low versus high genotypes for (B), and TLR1_low versus high genotypes for IL-6 (C). For IL-10, TLR1/6_low vs high diplotypes for media, PAM2, and PAM3 (D) or LPS (E); TLR6_low versus high genotypes for media, PAM2, and PAM3 (F) or LPS (G); TLR1_low versus high genotypes for media, PAM2, and PAM3 (H) or LPS (I). Diplotypes and genotypes with normal signaling (high) designated with a box with dots while deficient signaling (low) designated with clear box. P values represent significant differences as assessed by a Mann-Whitney U-test for comparison of low versus high groups. *P<0.05, **P<0.01, ***P≤0.001.