TLR5 rs5744174 gene polymorphism is associated with the virus etiology of infant bronchiolitis but not with post‐bronchiolitis asthma

Abstract Background and aim Bronchiolitis is a leading cause of hospitalization in infants and is associated with a risk of subsequent asthma. The innate immunity genes, such as those encoding toll‐like receptors (TLRs), are likely to play a role in bronchiolitis and post‐bronchiolitis outcome. Thus far, only one study has considered TLR5 genes in respiratory syncytial virus (RSV) bronchiolitis. The aim of this study was to investigate the association of TLR5 gene polymorphism with virus etiology and severity of bronchiolitis, and with post‐bronchiolitis asthma. Methods We recruited 164 infants (age < 6 months) hospitalized for bronchiolitis in this study and determined TLR5 rs5744174 (C > T) single nucleotide polymorphism, virus etiology and severity markers of bronchiolitis, and presence of post‐bronchiolitis asthma until age 11 to 13 years. Results RSV was detected in 113 (68.9%), rhinovirus in 19 (11.6%), and some other virus in 20 (12.2%) cases. Non‐RSV etiology was more common among infants with the variant CT or TT genotype in the TLR5 rs5744174 gene than in those with the CC genotype (89.7% vs 71.7%, P = 0.03). TLR5 rs5744174 polymorphism was not associated with the need of supplementary oxygen or feeding support, with the length of hospital stay, or with post‐bronchiolitis asthma at any age. Conclusion The TLR5 rs5744174 variant genotype may increase the susceptibility to bronchiolitis not caused by RSV.

Bronchiolitis is the most common lower respiratory tract infection requiring hospitalization in young children. 3 Among the various respiratory viruses causing bronchiolitis, respiratory syncytial virus (RSV) is the single most important one, especially in the youngest children. 4,5 The clinical course of primary RSV infection is highly variable, and genetic variations in genes regulating the immune response certainly influence disease severity. 6 Bronchiolitis in infancy is associated with subsequent wheezing in early childhood and asthma in later childhood. [7][8][9] The link between bronchiolitis and subsequent asthma remains unclear, but it has been suggested that the causative virus modifies the immature immunity towards a Th2-oriented cytokine profile. 10 Toll-like receptors (TLRs) are innate immune molecules that recognize conserved structures of microbial pathogens. 11 They activate immediate and early mechanisms of innate host defense as well as initiate and orchestrate adaptive immune responses. 11 Several single-nucleotide polymorphisms (SNPs) within the TLR genes have been associated with altered susceptibility to infectious, inflammatory, and allergic diseases. 11,12 TLRs 1, 2, 4, 5, 6, and 10 are located on the cell surface, directly recognizing microbial components, whereas TLRs 3, 7, 8, and 9 are located inside the cell, 13 recognizing microbial components after endocytosis. TLR5 recognizes bacterial flagellin, which is an important virulence factor of many bacteria 14 and is also found in house dust. 15 The effect of bacterial products in house dust is mainly protective for allergy and asthma. 16 In prior studies, activation of TLR5 by flagellin has been associated with atopic eczema 17 and also with asthma, 15 but there are studies with contradictory results reporting that asthmatic patients have decreased expression 18,19 or impaired function of TLR5. 19 Moreover, polymorphisms in the TLR5 gene have been described to be associated with both acute and chronic lung diseases, eg, pneumonia caused by Legionella pneumophila 20 and bronchopulmonary dysplasia in preterm infants. 21 TLR5, as a receptor of bacterial flagellin, is known to be involved in allergy development and subsequent asthma. 15 A Dutch study found preliminary evidence that the association of TLR5 rs5744174 (Phe616Leu) polymorphism with bronchiolitis risk may be divergent in preterm and term infants. 22 We have previously investigated the associations of SNPs of 9 TLR genes, not including the TLR5 gene, with bronchiolitis and post-bronchiolitis outcome. [23][24][25][26][27] The aim of this study was to complete our previous exploratory studies by evaluating the association of TLR5 rs5744174 gene polymorphism with bronchiolitis in infancy, with post-bronchiolitis wheezing until the age of 18 months, with preschool asthma at the age of 5 to 7 years, and with childhood asthma at the age of 11 to 13 years.

| Design
The study was conducted at the Department of Pediatrics, Tampere University Hospital, Finland, and the design has previously been described. 9 In brief, 187 previously healthy, full-term infants hospitalized for bronchiolitis at less than 6 months of age in 2001 to 2004 were eligible for the study. Of them, 139 attended a follow-up visit in 2002 to 2004 when the mean age of the children was 18 months. The second follow-up visit was arranged in 2008 to 2010, at 5 to 7 years of age, to which 166 attended, and the third took place in 2014 to 2015, at the age of 11 to 13 years, to which 138 attended. Whole blood samples were obtained for genetic studies in infancy, supplemented at age 5 to 6 years, if needed. Overall, data on the TLR5 rs5744174 (C > T) gene polymorphism were available from 164 infants. In infancy, bronchiolitis was defined as an acute lower respiratory illness characterized by rhinorrhea, cough, and diffuse wheezes or crackles. 28 The viral etiology of bronchiolitis was studied by antigen detection and polymerase chain reaction (PCR) in nasopharyngeal aspirates, as described previously. 29 The studied viruses were RSV, rhinovirus, human metapneumovirus, influenza A and B virus, parainfluenza type 1, 2, and 3 viruses, bocavirus, and adenovirus. Data on disease severity, such as the need for supplementary oxygen and feeding support, and the length of hospital stay (LOS), were recorded during the inpatient care. 27 After hospitalization for bronchiolitis, the children were invited to a follow-up visit at, on average, 1.5 years of age. At the follow-up visit, the parents were interviewed using a structured questionnaire on the occurrence of atopic eczema, wheezing episodes, and the use of corticosteroids for wheezing after hospitalization for bronchiolitis. 30 The parents had recorded all infections and wheezing periods diagnosed by a family doctor or a pediatrician and all given treatments in a diary during the 1.5 years post-bronchiolitis follow-up period.
Only doctor-diagnosed episodes were included in the analyses.
Repeated wheezing was defined as 2 or more wheezing episodes during the 1.5-year post-bronchiolitis follow-up period. 30 The follow-up study when the children were 5 to 7 years old, consisted of medical histories since the last visit at the age of 1.5 years.
The data were collected using structured questionnaires, which the parents had completed before the visit, and a clinical examination by a doctor that included an interview, checking the questionnaire data, and an exercise challenge test with impulse oscillometry for bronchial hyper-reactivity. 9 Current asthma was defined as the current use of continuous maintenance medication with inhaled corticosteroids (ICS) for asthma, or suffering from doctor-diagnosed episodes of wheezing after the age of 1.5 years, or from prolonged or night cough during the preceding 12 months, with a diagnostic finding in the exercise challenge test with impulse oscillometry. 9 Previous asthma before the control visit was defined as the previous use of ICSs as continuous or intermittent maintenance medication for asthma. 9 The follow-up study, when the children were 11 to 13 years old, consisted of medical histories since the last visit at the age of 5 to 7 years. As in previous visits, data were collected using structured questionnaires, which the parents had completed before the visit, and a clin- An increase of 12% in FEV1 was considered as diagnostic for asthma.
Current asthma was considered to be present if the child had used ICSs continuously during the last 12 months. It was also considered to be present if the child had suffered from repeated wheezing or from a prolonged cough or night cough for 4 or more weeks during the last 12 months, and, in addition, had a diagnostic increase of FEV1 in the bronchodilation test. Persistent asthma was considered to be present if the child with current asthma had also had asthma at the last control visit at the age of 5 to 7 years.
At both the 5 to 7-year and the 11 to 13-year visit, the questionnaire consisted of questions concerning doctor-diagnosed atopic eczema, allergic rhinitis, and food allergies at different ages since the last control visit. It also covered the current use of asthma medication, current symptoms that could suggest asthma, and the current asthma diagnosis, if present. were run at 95°C for 10 minutes followed by 45

| Ethics
The study was carried out in accordance with the approved guidelines of the WMA Declaration of Helsinki. The protocol of the study was approved by the Ethics Committee of the Tampere University Hospital District, Tampere, Finland. Before enrolling the children, we obtained informed parental consent, including the use of samples for genetic studies on bronchiolitis and asthma risk, both during hospitalization and at the control visit. The personal data of the study subjects were not given to the laboratory that performed the genetic studies, the Department of Medical Microbiology and Immunology, University of Turku, Finland.

| Statistics
Statistical analyses were carried out with SPSS version 23.0 software (IBM Corp, NY, USA). The Chi-square test and Fisher's exact test were used, as appropriate, for categorized variables. Logistic regression with adjustments for sex and age was used to analyze the association between the TLR5 rs5744174 genotypes and virus etiology of bronchiolitis. The results were expressed as odds ratios (OR) and 95% confidence intervals (95% CI).
The FINETTI program, version 3.0.5 (http://ihg.gsf.de/cgi-bin/hw/ hwa1.pl) was used to evaluate the Hardy-Weinberg equilibrium of the studied TLR5 alleles, and they were in the Hardy-Weinberg equilibrium. There were no significant associations between theTLR5 rs5744174 genotypes and the severity markers of bronchiolitis, such as the need of feeding support, oxygen supplementation, or the LOS (Table 2.).

| Comparison with FIN data
The MAF (allele T) in this study population was 0.51 and did not differ from the general Finnish population according to FIN data of the 1000 Genomes Project, 33 where the MAF of the TLR5 rs5744174 (C > T) was 0.50. In the RSV group, the MAF was 0.47 (P = 0.47 vs the FIN data), and in the non-RSV group, the MAF was significantly higher, at 0.64 (P = 0.03 vs the FIN data) ( Table 1).  (Table 3).

| 5 to 7-year follow-up visit
There were 139 children with clinical data available from the 5 to 7year follow-up visit. The TLR5 rs5744174 genotype was CC in 36 (25.9%) cases and CT or TT (variant) in 103 (74.1%) cases. There were no significant associations between the TLR5 rs5744174 genotype and prolonged cough, ICS use, current asthma, or presence of atopic eczema or allergic rhinitis (Table 4).

| 11 to 13-year follow-up visit
There were 123 children with clinical data available from the 11 to 13year follow-up visit. The TLR5 rs5744174 genotype was CC in 32 (26.0%) cases and CT or TT (variant) in 91 (74.0%) cases. There were no significant associations between the TLR5 rs5744174 genotypes and prolonged cough, ICS use, current asthma, persistent asthma continuing from preschool age until the latest follow-up visit, or presence of atopic eczema or allergic rhinitis (Table 5.).

| DISCUSSION
There are 3 main results in this study concerning the TLR5 rs5744174 (C > T) gene polymorphism in bronchiolitis and post-bronchiolitis outcome. First, the TLR5 rs5744174 variant genotype was associated with non-RSV etiology of bronchiolitis. Second, TLR5 rs5744174 polymorphism was not associated with severity of bronchiolitis. Third, TLR5 rs5744174 polymorphism was not associated with post-bronchiolitis wheezing, preschool asthma, or current asthma in 11 to 13-year-old children after bronchiolitis in infancy.    There is some evidence that polymorphisms in the TLR5 gene are associated with bacterial infections, 20,36 but their role in viral infections is less studied. So far, there has been only one study on TLR5 in infant bronchiolitis, where TLR5 rs5744174, the same SNP as in the present study, did not have any significant effect on the risk of RSV bronchiolitis. 22 In stratified analyses, however, preliminary evidence was found that the influence may be divergent in term and preterm infants with bronchiolitis. No studies are available on the association between TLR5 genes and bronchiolitis caused by other viruses, or between TLR5 genes and the post-bronchiolitis outcome.
The TLR5 rs5744174 site is located in the coding region of the gene, and the SNP results in a missense mutation, where the 616 th amino acid of the TLR5 protein, phenylalanine (Phe), is substituted by a leucine (Leu). 37 Depending on its location, polymorphisms in the encoding gene of TLRs may affect the level of expression or alternatively, lead to an altered binding affinity or to an altered downstream signaling. 38 Subsequently, these changes in the primary defense may lead to attenuated immune responses and, further, to increased susceptibility to infections. In primary human cell cultures, TLR5 rs5744174 polymorphism was found to attenuate TLR5 signaling in response to bacterial flagellin. 39 In addition to recognizing flagellin, there is preliminary evidence that TLR5 might have immune modulating effects in virus infections. A recent study found that flagellininduced activation of TLR5 prevented rotavirus infection in mice through activating innate immunity. 40 Moreover, flagellin has been shown to be effective as an adjuvant in influenza vaccines by triggering TLR5 activity and boosting immune responses. 41,42 In the present study, the TLR5 rs5744174 variant genotype was associated with an increase in susceptibility to non-RSV bronchiolitis, which may take place via an attenuated function of TLR5.
The data on the role of TLR5 in asthma is inconsistent, and even less is known about the effect of the genetic variants of the TLR5 rs5744174 gene on asthma susceptibility. In the present study, TLR5 rs5744174 polymorphism had no significant association with recurrent wheezing or asthma during the longer than 10 years follow-up after bronchiolitis in infancy. This is in agreement with a study of a German population in which no significant associations between TLR5 rs5744174 polymorphism (the polymorphism studied here), or 2 other TLR5 gene polymorphisms, rs5744168 and rs2072493, and childhood asthma, were found. 38 It has been suggested, that in healthy subjects, the activation of TLR5 induces Th1-type immune responses, but in asthmatic patients, the expression and activation of TLR5 and subsequent release of Th1-type and anti-inflammatory cytokines are decreased, and the normal function of TLR5 probably is protective from asthma. 19 In another study, TLR5 expression was decreased only in severe asthmatics. 18 A different study reported contradictory results, proposing that the activation of TLR5 leads to Th2-oriented responses in mice. 43 Further, a few studies have implicated that the effect of contact with bacterial flagellin may be dose dependent: a small amount of flagellin may promote asthma by priming allergic responses, 15 whereas higher amounts may suppress allergic asthma. 44 Compared with other viruses, RSV is associated with more severe symptoms of bronchiolitis, whereas rhinovirus is more commonly associated with post-bronchiolitis asthma than is RSV. 45 In the present study, TLR5 rs5744174 genotypes did not associate with severity markers of bronchiolitis during hospitalization, although non-RSV bronchiolitis was more common among children with variant genotypes of the TLR5 rs5744174 gene. This may be due to the small sample size, as only 39 children had bronchiolitis caused by some other virus than RSV. It is also notable, that a loss-of-function mutation in the ligandbinding domain of TLR5, 392STOP, was associated with defective responses to flagellin and increased susceptibility to Legionnaire's disease. 20 This polymorphism is present in 5% to 10% of Europeans and in 23% of some other populations, and when present in both alleles, leads to a complete TLR5 defect without any general immunodeficiency. 46 Thus, TLR5 seems not to be crucial for host defense, and in case of attenuated or even complete loss of function, other molecules of innate immunity are able to compensate for its lacking function. 46 The strengths of this study are the prospective design and carefully collected data during both the hospitalization and repeated follow-up visits, and a long follow-up time of over 10 years. The homogeneous ethnic background of the study children is also a benefit in genetic studies. One clear limitation of this study, however, is the fact that we were not able to measure the expression of TLR5 or the functionality of the TLR5 rs5744174 gene. However, the functionality of this polymorphism has been proven in previous studies. 39 The small sample size is a limitation for genetic studies and, therefore, the findings need to be repeated in larger study populations.
In conclusion, the TLR5 rs5744174 variant genotypes were associated with non-RSV etiology of bronchiolitis, but not with the severity of bronchiolitis or with the incidence of post-bronchiolitis asthma. Thus, the findings of this exploratory study suggest that TLR5 may