Toll-like receptor heterodimer variants protect from childhood asthma

Article Outline

• Abstract
• Methods
  • Population
  • Genetics and molecular biology
  • Statistical analyses
• Results
• Discussion
• Methods
  • Genetic analysis
  • Quantitative real-time RT-PCR
  • Flow cytometry
  • Statistical analysis
  • Cell stimulation and ELISA
• References
• Fig E1. 
• Fig E2. 
• Fig E3. 
• Table E1. 
• Table E2. 
• Table E3. 
• Table E4. 
• References
• Copyright

Background

Early exposure to microbes reduces the risk for asthma. Toll-like receptors (TLRs) represent a major group of receptors for the specific recognition of pathogen-associated molecular patterns of microbes capable of activating innate and adaptive immunity.

Objective

Because TLRs can influence key events in the induction and perpetuation of asthma and atopy, we sought to determine whether genetic alterations in TLR genes affect asthma risk.

Methods

We systematically evaluated putatively functional genetic variants in all 10 human TLR genes for their association with different asthma phenotypes in a case-control study (n = 1872) by using matrix-assisted laser desorption/ionization time-of-flight genotyping. For polymorphisms showing association with atopic asthma, effects on gene and protein expression were studied by means of RT-PCR and flow cytometry ex vivo. T-cell cytokine production was evaluated by means of ELISA after stimulation of the respective TLRs with specific ligands.

Results

Protective effects on atopic asthma were identified for single nucleotide polymorphisms in TLR1 (odds ratio [OR], 0.54; 95% CI, 0.37-0.81; P = .002), TLR6 (OR, 0.54; 95% CI, 0.37-0.79; P = .003), and TLR10 (OR, 0.58; 95% CI, 0.39-0.86; P = .006), all capable of forming heterodimers with TLR2. Effects remained significant after correction for multiple comparisons. PBMCs of minor allele carriers showed increased levels of the respective TLR mRNA and proteins, augmented inflammatory responses, increased T_H1 cytokine expression, and reduced T_H2-associated IL-4 production after specific stimulation.

Conclusion

These results suggest that functional relevant TLR1 and TLR6 variants are directly involved in asthma development.

Key words: Asthma, atopic asthma, Toll-like receptor, heterodimer, polymorphism

Abbreviations used: BHR, Bronchial hyperresponsiveness, ISAAC, International Study of Asthma and Allergies in Childhood, LD, Linkage disequilibrium, OR, Odds ratio, PE, Phycoerythrin, SNP, Single nucleotide polymorphism, TLR, Toll-like receptor

Chronic inflammatory diseases occurring at the interface of the organism with the environment, such as atopic asthma, inflammatory bowel disease, and eczema, increased steeply in many countries over the last decades.¹ According to new insights, skin and mucosa do not represent mere passive barriers but are centrally involved in a complex and highly sophisticated system for the recognition of (and interaction with) the environment. However, if the interaction with the environment is disturbed, immune responses can become imbalanced, insufficient, or exaggerated. Critical alterations can occur on both sides of the interface on the environmental side or within intrinsic components of the mucosal barrier. Thus changes in the composition of the environmental microbial exposure or in the recognition of this exposure can have profound effects and contribute to the development of different diseases.

In the case of asthma, the most common chronic disease in childhood, a deregulation of the adaptive immune system toward a T_H2 pattern, involving T-cell activation and cytokine release, as well as allergen-specific immunoglobulin production by B cells (atopy), results in chronic inflammation of the airways. However, the initial triggers for the induction of atopic immune responses are still unknown. It is hypothesized that substantial changes on both sides of the environmental interface might play a crucial role for the onset of atopy and childhood asthma. Many epidemiologic studies support this hypothesis by showing that exposure to an environment rich in microbial exposure (eg, farming) during early childhood protects against the later development of asthma and atopy.² In contrast, the change to a modern Western lifestyle with high hygiene standards and reduced contact with microbes increases the risk for these diseases.³, ⁴

Genetically induced variations in intrinsic components of barrier organs can also alter the capability to recognize microbes, leading to inadequate immune responses and increasing the susceptibility for asthma and atopy. Toll-like receptors (TLRs) represent one such major family of intrinsic microbial recognition receptors expressed in all barrier organs. By forming either homodimers or heterodimers, TLRs are capable of recognizing molecular pathogen patterns of bacteria, viruses, and fungi and subsequently inducing direct and indirect immunologic effects.⁵ TLRs also influence T-cell polarization and development,⁵, ⁶ a key event in the induction and perpetuation of asthma and atopy.⁷ Because TLRs are so centrally involved in the recognition of the microbial environment, genetic changes in these molecules can have profound effects. Therefore we systematically studied all 10 human TLR genes for polymorphisms and the effect of these polymorphisms on receptor expression and T cell–specific immune responses and their involvement in the development of asthma.

Back to Article Outline

Methods 

For further details, see the Methods section in the Online Repository at www.jacionline.org.

Population 

Based on the Munich and Dresden part of the International Study of Asthma and Allergy in Childhood (ISAAC) phase II study⁸ and a very similar study with comparable study tools from Leipzig,⁹ a nested case-control design was applied. All children with asthma, bronchial hyperresponsiveness (BHR), or both were selected (n = 624) and matched with a stratified random selection of healthy, nonasthmatic, nonatopic children without BHR (n = 1248; age, 9-11 years) at a 1:2 ratio from the same population as previously described (see Fig E1 in the Online Repository at www.jacionline.org).¹⁰ Polymorphisms associated with atopic asthma in the nested case-control population were also genotyped in the original cross-sectional population from Munich and Dresden (total: n = 3099; Munich: n = 1159; Dresden: n = 1940) to investigate the influence of single nucleotide polymorphisms (SNPs) on atopy independent of asthma status. Consequently, overlaps between the case-control and the cross-sectional study populations exist.

All children whose parents reported that a doctor diagnosed “asthma” at least once or “asthmatic, spastic, or obstructive bronchitis” more than once were defined as having asthma.⁸ A number of objective measurements, such as atopy skin tests and pulmonary function tests, have been performed to validate the accuracy of the epidemiologic asthma definition of a “doctor's diagnosis of asthma.” As reported previously, a large number of the children given diagnoses of asthma by doctors according to the questionnaire had positive skin test responses and BHR,¹¹ confirming the specificity and sensitivity of the questionnaire-based diagnosis.

A child was considered atopic with a positive skin prick test response (≥3 mm) to at least 1 of the following aeroallergens: Dermatophagoides pteronyssinus, Dermatophagoides farinae, Alternaria tenuis, cat dander, and mixed grass and tree pollen.⁸, ⁹ Atopic asthma was defined as asthma with atopy, whereas nonatopic asthma was defined as asthma without atopy. In the case-control population 171 cases of atopic asthma and 171 cases of nonatopic asthma could be distinguished; in all additional asthmatic subjects, atopy status was not determined. Informed written consent was obtained from all parents of children included in the study, and all study methods were approved by the local ethics committees.

Genetics and molecular biology 

Genotyping was performed by means of matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, as described previously (for primers, see Table E1 in the Online Repository at www.jacionline.org).¹⁰ Derived genotype frequencies were compared with the expected allelic population equilibrium based on the Hardy-Weinberg equilibrium test to control for technical genotyping errors. cDNA was amplified by using an iCycler (Bio-Rad, Hercules, Calif) in duplicate, with 18s as a reference gene (for primers, see Table E2 in the Online Repository at www.jacionline.org). Relative expression levels of TLR1, TLR2, TLR6, and TLR10 mRNA were measured in carriers of specific alleles selected from adult volunteers, as indicated in the Methods section of the Online Repository. Cycle threshold values were obtained by using iCycler iQ software 3.1 (Bio-Rad), which automatically calculated baseline cycles and threshold positions. PBMCs were gated due to their forward-scatter/side-scatter characteristics. At least 10,000 cells per sample were analyzed for TLR1, TLR2, TLR6, and TLR10 protein expression by means of flow cytometry (FACSCalibur; Becton Dickinson, Heidelberg, Germany). Results are reported as mean fluorescence intensity. Freshly isolated PBMCs were stimulated for 24 hours with specific TLR1/2/6 ligands, according to methods previously described.¹² After the incubation period, the PBMCs were separated by means of centrifugation, and the secretion of IL-6, TNF-α, IL-12, IFN-γ, IL-10, and IL-4 was measured in triplicate in culture supernatants by using ELISA kits according to the manufacturer's instructions (R&D Systems, Minneapolis, Minn). All SNP data were based on the results of the National Institute of Health mutation screening program of innate immunity genes¹³ performed in a standard set of immortalized human samples from the Coriell Institute for Medical Research, Camden, New Jersey (ccr.coriell.org). FastSNP (fastsnp.ibms.sinica.edu.tw) was used to determine SNP-induced changes in putative transcription factor binding sites in promoter and intronic regulatory sites in silico.

Statistical analyses 

Pairwise linkage disequilibrium (LD) was assessed between single SNPs by using the r² statistic with Haploview.¹⁴ Deviations of Hardy-Weinberg equilibrium were investigated by using the χ² statistic. Associations between SNPs and qualitative outcomes were tested by using Pearson χ² tests.¹⁵ Odds ratios (ORs) and 95% CIs are reported for the dominant model, in which the wild-type is compared with heterozygote and homozygote variants. For every phenotype analyzed, multiple testing was controlled for by calculating the false discovery with the method of Benjamini.¹⁶ Haplotype frequencies were estimated by using the expectation-maximization algorithm.¹⁷ Global tests for haplotype differences between cases and control subjects were calculated. Haplotype trend regressions were performed, in which the estimated probabilities of the haplotypes are modeled in a logistic regression as independent variables, to specify the effects of individual haplotypes.¹⁸ TLR and cytokine expression ex vivo was analyzed with the exact Wilcoxon–Mann-Whitney test. Differences in mRNA expression between groups were analyzed by using the pairwise fixed reallocation randomization test with REST 2005 software (Corbett Life Science, Sydney, Australia).¹⁹, ²⁰

Back to Article Outline

Results 

Polymorphisms for genotyping in this study were systematically selected from the 259 polymorphisms previously identified in the 10 human TLR genes by using the Innate Immunity Program for Genomic Applications mutation screen.¹³ Polymorphisms with a minor allele frequency of at least greater than 0.03, causing amino acid changes (n = 9) and thus altering protein structure and function, as well as frequent SNPs (minor allele frequency >0.1) leading to altered transcription factor binding in regulatory regions of TLRs (n = 9), as predicted by bioinformatics with FastSNP,²¹ were selected. In addition, 3 TLR SNPs previously described to be associated with asthma were also included for genotyping, irrespective of their concordance with the described selection process.²², ²³ Polymorphisms tagged by the selected SNPs were not considered for genotyping separately. Thus a total of 21 polymorphisms covering all 10 human TLR genes were studied (Table I and see Fig E2, Fig E3 in the Online Repository at www.jacionline.org).

Table I. Biologic characteristics of the analyzed TLR1 through TLR10 SNPs

MAF, Minor allele frequency.

Carriers of the TLR4_a and TLR9_a minor alleles showed an increased risk for asthma in the unadjusted analysis (Table II). However, these effects did not remain significant after correction for multiple testing. TLR2_c was inversely associated with nonatopic asthma (OR, 0.69; 95% CI, 0.50-0.96; P = .027), and the overall effect of TLR4_a on asthma was also stronger on nonatopic asthma (OR, 1.41; 95% CI, 1.01–1.96; P = .045). Again, those results did not remain significant after correction for multiple testing.

Table II. Association of TLR SNPs with asthmatic phenotypes in the case-control population

ORs (95% CIs) for a dominant model are shown. For X-chromosomal TLR7 and TLR8 SNPs, “f” indicates effect in female subjects, and “m” indicates effect in male subjects.

In contrast, polymorphisms TLR1_b, TLR1_c, TLR6_a, and TLR10_b showed highly significant inverse effects with atopic asthma, reducing the risk by almost half, whereas TLR6_b increased the risk for atopic asthma (Table II). The protective effects of TLR1_b, TLR1_c, and TLR6_a remained significant after correction for multiple testing, whereas the association of TLR10_b slightly failed the specified significance threshold.

In further experiments, the effects of TLR1, TLR6, and TLR10 polymorphisms on atopic asthma were studied in detail. An analysis of homogeneity was performed to assess reproducibility of the observed results, confirming consistency of effects and effect directions among all 3 study centers (data not shown). Furthermore, these SNPs were regenotyped in the complete and unselected cross-sectional populations from Munich and Dresden (n = 3099) to determine whether the associations of polymorphisms TLR1_b, TLR1_c, TLR6_a, TLR6_b, and TLR10_b with atopic asthma are linked to or caused by the concomitant presence of atopy in these cases. (In the case-control population this question could not be addressed sufficiently because control subjects had been selected for their absence of atopy.) Although we also found significant associations of TLR1_b, TLR1_c, TLR6_a, and TLR10_b with atopy, the association with atopic asthma was stronger than the effect on atopy alone and remained significant after correction for multiple testing (see Table E3 in the Online Repository at www.jacionline.org). Moreover, the effect on atopy was stronger in asthmatic than in nonasthmatic atopic subjects. Additionally, when haplotypes in TLR genes with at least 2 genotyped SNPs were evaluated in the case-control population, specific haplotypes in TLR1, TLR6, and TLR10 were again significantly associated with atopic asthma (see Table E4 in the Online Repository at www.jacionline.org). A global test for common haplotypes in TLR1 and TLR6 confirmed the significance of this association. Therefore it was concluded that these polymorphisms indeed influence the specific risk for atopic asthma and that effects are not only due to atopic sensitization.

Intriguingly, all 3 TLRs showing associations with atopic asthma form receptor complexes (heterodimers) with TLR2. To study direct functional effects of these SNPs on the TLR2 heterodimer system, SNP-dependent gene expression was investigated at the mRNA and protein level (Fig 1). Individuals homozygous for the rare alleles in all 3 SNPs in TLR1_b, TLR6_a, and TLR10_b (n = 5) were compared with subjects homozygous for wild-type alleles (n = 9). Significantly increased TLR1, TLR6, and TLR10 mRNA expression was found in the group with the respective minor alleles, whereas TLR1, TLR6, and TLR10 polymorphisms did not influence TLR2 mRNA levels, as expected (Fig 1, A). Percentages of TLR-expressing cells did not differ between the studied groups. However, as assessed by means of flow cytometry, 1.5-fold higher levels of TLR1 (P = .009) and TLR6 (P = .003) receptor expression on PBMCs were observed in individuals homozygous for the rare alleles. When differentiating by cell type, we found TLR6 expression to deviate significantly only in the B-cell population and not on monocytes, whereas TLR1 expression differed significantly in both cell populations between allelic groups (Fig 1, B). TLR10 expression was only moderately increased, and TLR2 expression did not differ significantly (as expected; Fig 1, B). TLR1 and TLR6 were expressed on lymphocytes and monocytes, whereas TLR10 was exclusively expressed on B cells, as previously shown.²⁴, ²⁵

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  TLR1, TLR6, and TLR10 SNPs increase mRNA (A) and protein (B) expression on B cells (upper panel) and monocytes (lower panel) of respective TLRs when individuals carrying the homozygous rare alleles (n = 5) for TLR1_b, TLR6_a, and TLR10_b are compared with wild-type individuals (n = 9). Fig 1, A: Boxes represent medians ± interquartile range. Whiskers represent the minimum and maximum observations. Fig 1, B: Bars represent medians ± interquartile range of different donors. MFI, Median fluorescence intensity.

In a next step, the influence of genetic changes in TLR1/TLR6/TLR10 on cytokine responses was studied (Fig 2). PBMCs from the same individuals as before were stimulated with TLR2 ligands (lipoteichoic acid [LTA] and peptidoglycan [PGN]) and ligands specific for TLR2/TLR1 (Pam3CSK4) and TLR2/TLR6 (zymosan) heterodimers, whereas TLR10 ligands have not been identified thus far and could thus not be tested. IL-1β was used as a control (representing a TLR-independent pathway). After stimulation, the subsequent production of proinflammatory and T_H1/T_H2-associated cytokines was analyzed in cell-culture supernatants, as described previously.¹² PBMCs from homozygote rare allele carriers of TLR1_b, TLR6_a, and TLR10_b expressed significantly higher levels of the proinflammatory cytokine TNF-α (Fig 2, A) and the T_H1-related cytokines IL-12 and IFN-γ (Fig 2, B and C) on stimulation with TLR2/1 and TLR2/6 ligands compared with that seen in wild-type individuals. Although IL-1β led to a statistically significant difference in IL-12 expression between both groups, the relative differences in medians are low compared with zymosan and Pam3CSK4 stimulation. In contrast, levels of the T_H2 signature cytokine IL-4 were significantly less in homozygous rare allele individuals after stimulation with TLR2/1 and TLR2/6 compared with homozygous wild-types (Fig 2, D). These effects were not observed after stimulation with the TLR2 ligands LTA and PGN, which act independently of heterodimerization with TLR1 or TLR6. Thus polymorphisms in TLR1 and TLR6 seem to modify specific T cell–dependent cytokine expression after stimulation of the respective heterodimers.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  TLR1, TLR6, and TLR10 SNPs increase proinflammatory and T_H1-associated cytokine levels and reduce T_H2-associated IL-4 expression after TLR2/1 and TLR2/6 stimulation. The secretion of TNF-α (A), IL-12 (B), IFN-γ (C), and IL-4 (D) was measured in culture supernatants by means of ELISA in triplicate. Bars represent medians ± interquartile range between different donors. LTA, Lipoteichoic acid; PGN, peptidoglycan.

Back to Article Outline

Discussion 

In this genetic analysis of the TLR system, a distinct role of polymorphisms in specific TLR genes on the development of different asthma phenotypes was demonstrated. Although only weak effects were found for TLR2, TLR4, and TLR9 polymorphisms, strong protective effects of genetic variants in the TLR2-associated heterodimer network consisting of TLR1, TLR6, and TLR10 on atopic asthma were newly identified. Polymorphisms in TLR1, TLR6, and TLR10 were also found to be associated with increased mRNA expression of the respective TLR genes. For TLR1 and TLR6, receptor chain expression was also influenced by the presence of these polymorphisms. However, the size and significance of these effects on receptor chain expression were cell-type dependent, showing greater differences in B cells than in monocytes.

Furthermore, primary cells derived from carriers of protective TLR1, TLR6, and TLR10 variants showed augmented inflammatory responses, increased T_H1 cytokine expression, and reduced T_H2-associated IL-4 production after specific stimulation. This might lead to a better understanding of the mechanisms of protection against atopic sensitization and the subsequent development of atopic asthma acting through TLR2 heterodimers.

Polymorphisms in TLR2 and TLR4 showed only borderline associations not withstanding correction for multiple comparisons with nonatopic asthma, a form of childhood asthma mostly triggered by infection and exercise. Both these TLRs are capable of contributing to the recognition of common viruses,²⁶ providing a potential explanation why SNPs in these TLR genes might be linked to nonatopic asthma. However, amino acid changes in TLR4 (Asp299Gly and Thr399Ile) could not be directly associated with asthma itself,²⁷ and neither could we find a significant effect for Asp299Gly (TLR4_c), which is in high LD with TLR4_Thr399Ile (r² > 0.8).

In contrast, all polymorphisms significantly associated with atopic asthma are found in TLR genes that predominantly recognize bacteria-derived ligands. It is intriguing how the observed association results cluster, considering the broad spectrum of signals that are recognized by the TLR system. Both TLR2/1 and TLR2/6 heterodimers mediate responses to certain lipopeptides derived from various microbes, increasing the specificity of TLR-mediated pathogen recognition. TLR2/1 is activated by triacylated lipoproteins from different bacteria and lipoarabinomannan from mycobacteria, whereas TLR2/6 specifically recognizes diacyl lipopeptides from mycoplasma, zymosan from fungi, and lipoteichoic acid from group B streptococcus and staphylococcus.²⁸ All these microorganisms are capable of inducing severe pulmonary infections but are also frequently found in the upper respiratory tract in the absence of infection. How genetic changes in the receptors for these potential pathogens affect the capability of the organism to mount adequate defenses against these microbes is not yet clear. Based on our results, it might be speculated that these polymorphisms do not lead to a loss of function of the respective TLR but to a modification of effects associated with the TLR2 heterodimer network. Thus severe genotype effects on infection susceptibility for the TLR2 heterodimer network associated pathogens are not expected but cannot yet be excluded.

It has to be noted that the polymorphisms in TLR1, TLR6, and TLR10 studied in depth in this analysis do not act in isolation but that correlations with other SNPs within the studied genes and between TLR1, TLR6, and TLR10 exists because all 3 genes are located within close vicinity on chromosome 4p14 (see Fig E3 in the Online Repository). The TLR1_b variant is in LD with TLR6_a (r² = 0.800) and TLR10_b (r² = 0.865), whereas TLR6_a and TLR10_b show an r² of 0.690 in the cross-sectional sample. Although it could be argued that intergene LD, which is naturally occurring in many individuals in the population, might influence the observed associations on a population level, this cannot explain the distinct and independent TLR1 and TLR6 SNP effects in cytokine expression observed after TLR1- and TLR6-specific heterodimer stimulation. Functional experiments have only been conducted in samples from adults for ethical reasons. Even though it cannot be excluded that SNP effects on TLR function might differ between childhood and adulthood, this seems extremely unlikely. Taken together, association results and functional analyses indicate that TLR1 and TLR6 polymorphisms independently show significant ex vivo effects that can magnify when occurring together. Because of the high LD between asthma-associated SNPs within the TLR1, TLR6, and TLR10 gene cluster on chromosome 4 (see Figure E3 in the Online Repository), the difficulty to dissect the contribution of individual SNPs to changes in TLR expression and TLR signaling by in vivo experiments lies in the very nature of their concomitant occurrence. To further resolve those complex causal relationships at the genome level, the respective TLR cluster will subsequently have to be resequenced by using next-generation sequencing technology, which will also allow us to analyze the role of private mutations in this area. Findings of these in-depth mutation screens will have to be followed up by functional analyses. Only then will we be able to conclusively identify the contributions of single SNPs in this TLR cluster on the overall effects on disease development.

Because no specific artificial or natural ligand is currently known for TLR10, a more complex picture evolves with this receptor. TLR10 is capable of homodimerization but can also form heterodimers with TLR1 and TLR2.²⁹ An influence on asthma development of the TLR10 variant studied here was also reported previously.²³ The investigated SNP induces intracellular Ile to Val transition and might putatively affect ligand binding and signaling, but its functional consequences cannot be determined in the absence of a ligand. In this context it is interesting that significant differences in TLR10 mRNA expression were linked to this SNP (most likely because of LD with a promoter or regulatory SNP), but these differences were not reflected on the protein level (Fig 1). Thus it could be speculated that TLR10 expression is not regulated in a linear manner and not exclusively at the genome level but that strong regulatory effects might also be present at the mRNA level.

Our observations offer new insights into the mechanisms by which exposure to microbial stimuli influence the risk for atopic sensitization and the subsequent development of asthma. At the interface with the environment, not just any microbial exposure but specific ligands stimulating certain TLRs and TLR heterodimers might protect from the development of various asthma phenotypes in childhood. This new knowledge might trigger the development of new strategies to modify specific TLR signaling and thereby protect children from atopic asthma.

Back to Article Outline

Methods 

Genetic analysis 

Mass spectra resulting from matrix-assisted laser desorption/ionization time-of-flight analyses were analyzed automatically for peak identification with SpectroTYPER RT 3.4 software (Sequenom, San Diego, Calif). For genotyping TLR1_b and TLR5_a in the case-control population, a solid-phase oligonucleotide ligation assay was performed, according to the manufacturer's protocol (Variom Biotechnology, Berlin, Germany), and detected with avidin-peroxidase and colorimetric substrate on a Wallac Victor² 1420 (PerkinElmer, Waltham, Mass). Allele frequencies of control subjects were compared with those of other white populations wherever this information was available from previous studies, showing no significant deviations. Derived genotype frequencies were compared with the expected allelic population equilibrium based on the Hardy-Weinberg equilibrium test to control for technical genotyping errors. When significant deviations from Hardy-Weinberg equilibrium were detected (TLR1_a, TLR1_b, TLR2_b, TLR5_b, TLR6_a, TLR8_a, and TLR10_b), genotyping results were reproducible with different assay conditions, and samples were resequenced with native DNA on an ABI 3730 sequencer (Applied Biosystems, Foster City, Calif). Thus genotyping errors were excluded in all these cases. Completion rates for genotyping ranged from 89% to 99%.

Quantitative real-time RT-PCR 

Forty adult volunteers donating blood for immunologic experiments were genotyped for respective TLR polymorphisms, and within this population, 5 homozygous carriers of TLR1_b/TLR6_a/TLR10_b rare alleles and 9 wild-type controls were selected for mRNA quantification experiments. Within both groups, allergy and atopic diseases were equally distributed, and individuals were free of any respiratory infection for at least 1 week before the drawing of blood for the functional experiments.

Flow cytometry 

PBMCs were isolated from the same individuals as described before by means of Ficoll-Hypaque density gradient centrifugation (Biochrom, Berlin, Germany) from whole blood anticoagulated with ethylenediamine tetraacetic acid. The following antibodies were used (1 μg/10⁶ cells): TLR1-phycoerythrine (PE) mouse IgG1, TLR2-PE mouse IgG2a, TLR6-biotin rat IgG2a, and streptavidin-PE (all from eBioscience, San Diego, Calif); TLR10 mouse IgG1 and anti-mouse-PE goat IgG (Imgenex, San Diego, Calif); mouse IgG1-PE (BD, Heidelberg, Germany); and anti-human CD19 mouse IgG1-fluorescein isothiocyanate (BD). For TLR10, surface and intracellular staining were performed. B cells were detected with CD19-fluorescein isothiocyanate. TLR1 and TLR6 were differently expressed on lymphocytes (mean fluorescence intensity of 45 [interquartile range, 41-52] and 40 [interquartile range, 32-47], respectively) and on monocytes (mean fluorescence intensity of 80 [interquartile range, 72-89] and 60 [interquartile range, 52-68], respectively).

Statistical analysis 

Statistical significance was defined as a P value of less than .05. Analyses were carried out with SAS (version 9.1.3), SAS/Genetics, or Cocaphase.^E1 GraphPad Prism 4.0 (GraphPad Software, San Diego, Calif) was used to create the figures of TLR expressions and TLR ligand stimulation.

Cell stimulation and ELISA 

PBMCs were cultured in 96-well plates at 3 × 10⁵ cells/200 μL per well in RPMI supplemented with 10% FCS and stimulated for 24 hours with specific TLR1/2/6 ligands according to the previously described methods,^E2 with the following ligands: Pam3CSK4 (10 μg/mL; TLR1/2 agonist [EMC Microcollections, Tuebingen, Germany]), PGN (1 μg/mL; TLR2 and nucleotide-binding oligomerization domain containing 2 agonist [InvivoGen, San Diego, Calif]), LTA (10 μg/mL; TLR2 agonist [Sigma-Aldrich, Munich, Germany]), zymosan (10 μg/mL; TLR2/6 agonist [Sigma-Aldrich]), and IL-1β (50 ng/mL [PeproTech, Rocky Hill, NJ]) as non-TLR ligand. All reagents, buffers, and media were free of LPS (<0.01 ng/mL), as assessed by using the Limulus assay (Sigma-Aldrich). After the incubation period, PBMCs were isolated, and the secretion of TNF-α, IL-12, IFN-γ, and IL-4 was measured in triplicate in culture supernatants by using ELISA kits, according to the manufacturer's instructions (R&D Systems, Minneapolis).

Back to Article Outline

References 

Back to Article Outline

Fig E1. 

• 
  • View Large Image
  • Download to PowerPoint

• Overview of cross-sectional and case-control selection based on children from ISAAC II (Munich and Dresden) and Leipzig. The original populations, all children with questionnaire data available,^E3 that gave rise to both selections are shown.

Back to Article Outline

Fig E2. 

• 
  • View Large Image
  • Download to PowerPoint

• Scheme of the selection algorithm used to select 21 putative functional SNPs in TLR1 through TLR10 for genotyping. ΔAA, Leads to amino acid change; ΔTF, affects transcription factor binding site; R², correlation coefficient. ^∗Previously reported to be directly or indirectly associated with the development of asthma.

Back to Article Outline

Fig E3. 

• 
  • View Large Image
  • Download to PowerPoint

• TLR1/TLR6/TLR10 gene loci on chromosome 4p14 and description of intragenetic and intergenetic LD patterns (r² plots). Below are the positions of all frequent TLR1/TLR6/TLR10 SNPs (minor allele frequency > 0.1) and r² plot in the European samples of the Innate Immunity Program for Genomic Applications mutation screen.^E4 SNPs that were further investigated in this article are shown in bold red letters. CDS, Coding sequence; UTR, untranslated region.

Back to Article Outline

Table E1. 

SNP positions of genotyped SNPs, rs numbers, and primers used for matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

Back to Article Outline

Table E2. 

Primers used for real-time RT-PCR

Back to Article Outline

Table E3. 

TLR1, TLR6, and TLR10 SNPs differ in association with atopic asthma in the cross-sectional analysis (n = 3099)

ORs (95% CIs) for dominant effects are shown.

Back to Article Outline

Table E4. 

Association of TLR haplotypes with asthmatic phenotypes in the case-control population

ORs (95% CIs) for haplotype trend regression are shown.

Back to Article Outline

References 

1. Bach JF. The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med. 2002;347:911–920
   • View In Article
   • CrossRef
2. Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M, Grize L, et al. Environmental exposure to endotoxin and its relation to asthma in school-age children. N Engl J Med. 2002;347:869–877
   • View In Article
   • CrossRef
3. Strachan DP. Hay fever, hygiene, and household size. BMJ. 1989;299:1259–1260
   • View In Article
   • MEDLINE
   • CrossRef
4. Matricardi PM, Rosmini F, Riondino S, Fortini M, Ferrigno L, Rapicetta M, et al. Exposure to foodborne and orofecal microbes versus airborne viruses in relation to atopy and allergic asthma: epidemiological study. BMJ. 2000;320:412–417
   • View In Article
   • MEDLINE
   • CrossRef
5. Akira S, Takeda K, Kaisho T. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat Immunol. 2001;2:675–680
   • View In Article
   • MEDLINE
   • CrossRef
6. Medzhitov R, Preston-Hurlburt P, Janeway CA. A human homologue of the Drosophila Toll protein signals activation of adaptive immunity. Nature. 1997;388:394–397
   • View In Article
   • MEDLINE
   • CrossRef
7. Cohn L, Tepper JS, Bottomly K. IL-4-independent induction of airway hyperresponsiveness by Th2, but not Th1, cells. J Immunol. 1998;161:3813–3816
   • View In Article
   • MEDLINE
8. Weiland SK, von Mutius E, Hirsch T, Duhme H, Fritzsch C, Werner B, et al. Prevalence of respiratory and atopic disorders among children in the East and West of Germany five years after unification. Eur Respir J. 1999;14:862–870
   • View In Article
   • MEDLINE
   • CrossRef
9. von Mutius E, Weiland SK, Fritzsch C, Duhme H, Keil U. Increasing prevalence of hay fever and atopy among children in Leipzig, East Germany. Lancet. 1998;351:862–866
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (198 KB)
   • CrossRef
10. Kormann MS, Carr D, Klopp N, Illig T, Leupold W, Fritzsch C, et al. G-protein-coupled receptor polymorphisms are associated with asthma in a large German population. Am J Respir Crit Care Med. 2005;171:1358–1362
    • View In Article
    • MEDLINE
    • CrossRef
11. Nicolai T, Mutius EV, Reitmeir P, Wjst M. Reactivity to cold-air hyperventilation in normal and in asthmatic children in a survey of 5,697 schoolchildren in southern Bavaria. Am Rev Respir Dis. 1993;147:565–572
    • View In Article
    • MEDLINE
12. Renner ED, Pawlita I, Hoffmann F, Hornung V, Hartl D, Albert M, et al. No indication for a defect in toll-like receptor signaling in patients with hyper-IgE syndrome. J Clin Immunol. 2005;25:321–328
    • View In Article
    • MEDLINE
    • CrossRef
13. Innate Immunity PGA (Program for Genomic Applications), National Heart, Lung, and Blood Institute Program in Genomic Applications. Available at: http://innateimmunity.net. Accessed May 30, 2008.
    • View In Article
14. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–265
    • View In Article
    • MEDLINE
    • CrossRef
15. Sasieni PD. From genotypes to genes: doubling the sample size. Biometrics. 1997;53:1253–1261
    • View In Article
    • MEDLINE
    • CrossRef
16. Benjamini Y, Drai D, Elmer G, Kafkafi N, Golani I. Controlling the false discovery rate in behavior genetics research. Behav Brain Res. 2001;125:279–284
    • View In Article
    • MEDLINE
    • CrossRef
17. Excoffier L, Slatkin M. Maximum-likelihood estimation of molecular haplotype frequencies in a diploid population. Mol Biol Evol. 1995;12:921–927
    • View In Article
    • MEDLINE
18. Zaykin DV, Westfall PH, Young SS, Karnoub MA, Wagner MJ, Ehm MG. Testing association of statistically inferred haplotypes with discrete and continuous traits in samples of unrelated individuals. Hum Hered. 2002;53:79–91
    • View In Article
    • MEDLINE
    • CrossRef
19. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 2002;30:e36
    • View In Article
20. Dudbridge F. Pedigree disequilibrium tests for multilocus haplotypes. Genet Epidemiol. 2003;25:115–121
    • View In Article
    • MEDLINE
    • CrossRef
21. Yuan HY, Chiou JJ, Tseng WH, Liu CH, Liu CK, Lin YJ, et al. FASTSNP: an always up-to-date and extendable service for SNP function analysis and prioritization. Nucleic Acids Res. 2006;34:W635–W641
    • View In Article
    • CrossRef
22. Eder W, Klimecki W, Yu L, von Mutius E, Riedler J, Braun-Fahrlander C, et al. Toll-like receptor 2 as a major gene for asthma in children of European farmers. J Allergy Clin Immunol. 2004;113:482–488
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (104 KB)
    • CrossRef
23. Lazarus R, Raby BA, Lange C, Silverman EK, Kwiatkowski DJ, Vercelli D, et al. TOLL-like receptor 10 genetic variation is associated with asthma in two independent samples. Am J Respir Crit Care Med. 2004;170:594–600
    • View In Article
    • MEDLINE
    • CrossRef
24. Hornung V, Rothenfusser S, Britsch S, Krug A, Jahrsdorfer B, Giese T, et al. Quantitative expression of toll-like receptor 1-10 mRNA in cellular subsets of human peripheral blood mononuclear cells and sensitivity to CpG oligodeoxynucleotides. J Immunol. 2002;168:4531–4537
    • View In Article
    • MEDLINE
25. Bourke E, Bosisio D, Golay J, Polentarutti N, Mantovani A. The toll-like receptor repertoire of human B lymphocytes: inducible and selective expression of TLR9 and TLR10 in normal and transformed cells. Blood. 2003;102:956–963
    • View In Article
    • MEDLINE
    • CrossRef
26. Vaidya SA, Cheng G. Toll-like receptors and innate antiviral responses. Curr Opin Immunol. 2003;15:402–407
    • View In Article
    • MEDLINE
    • CrossRef
27. Raby BA, Klimecki WT, Laprise C, Renaud Y, Faith J, Lemire M, et al. Polymorphisms in toll-like receptor 4 are not associated with asthma or atopy-related phenotypes. Am J Respir Crit Care Med. 2002;166:1449–1456
    • View In Article
    • MEDLINE
    • CrossRef
28. Ishii KJ, Coban C, Akira S. Manifold mechanisms of toll-like receptor-ligand recognition. J Clin Immunol. 2005;25:511–521
    • View In Article
    • MEDLINE
    • CrossRef
29. Hasan U, Chaffois C, Gaillard C, Saulnier V, Merck E, Tancredi S, et al. Human TLR10 is a functional receptor, expressed by B cells and plasmacytoid dendritic cells, which activates gene transcription through MyD88. J Immunol. 2005;174:2942–2950
    • View In Article
    • MEDLINE