Genetic variation in Interleukin-17A is associated with increased risk and severity of pneumococcal meningitis in children

Background Streptococcus pneumonia is a leading cause of bacterial meningitis in children. Interleukin-17 (IL-17) promotes the host’s defense against pneumococcal infections. We investigated whether single nucleotide polymorphisms (SNPs) of IL -17A were associated with pneumococcal meningitis (PM) in Chinese children.Methods Three SNPs of IL-17A promoter rs4711998 (-877A>G), rs8193036 (-737C>T) and rs2275913 (-197G>A) were studied in 95 laboratory-confirmed PM pediatric patients and 330 healthy controls by PCR-based sequencing. Luciferase assays were used to determine the effects of IL-17A haplotypes on expression. Serum concentrations of IL-17A and C-X-C motif ligands 1 (CXCL-1) were determined.Results Frequency of the haplotype A-C-A differed significantly between cases and controls (43.6% vs. 35.7%, OR: 1.397, 95% CI: 1.004-1.942, P =0.047). The haplotype had an increased expression of IL-17A compared to wild haplotype A-C-G (P<0.001). Moreover, it produced a higher level of serum C-reactive protein (>10mg/L) (46.6% vs. 28.4%, OR: 2.276, 95%CI: 1.011-5.126, P =0.043). This haplotype was associated with low level of cerebrospinal fluid (CSF) glucose (≤1.5mmol/L) (52.8% vs.25.1%, OR: 3.350, 95%CI: 1.787-6.280, P <0.001) and high level of CSF protein (45.3% vs.29.2%, OR: 2.010, 95%CI: 0.991-4.076, P =0.05).Conclusions Our result indicated that haplotype A-C-A of IL-17A was associated


Background
Bacterial meningitis (BM) among the young children is one major public health concern around the world. Streptococcus pneumoniae (S. pneumoniae) is a leading cause of BM.
Although pneumococcal conjugate vaccines have introduced in industrial countries, the global case-fatality ratio of PM in children younger than 5 years was 59% despite the mortality rate of 10 (4-13) per 100,000 [1]. Cognitive impairments, hearing loss and psychological distress are the major sequelae among survivors of PM [2]. In China, during 2012-2017, of all the 1138 children with invasive pneumococcal disease, 39.2% (n=446) were diagnosed with meningitis. [3].
The pathogenesis of bacterial meningitis involves a complex interplay between pathogens and the host defense. Interleukin-17A (IL-17A) is an important proinflammatory cytokine, which is mainly produced by a subset of CD4+T cells named Th17 cells [4,5]. It plays an important role in host innate immune responses against bacterial colonization and infection. Studies in animal models have demonstrated that IL-17A can promote production and recruitment of neutrophils to infection sites via inducing the expressions of various chemokines, inflammatory factors and adhesion molecules [6][7][8]. In most cases, nasopharyngeal colonization of S. pneumoniae is followed by its invasion into bloodstream. After crossing the blood-brain barrier (BBB), bacteria multiply and trigger activation of host immune defense [9,10]. This response includes expression of inflammatory cytokines such as IL-6 and IL-8 through MyD88 signaling. However, these cytokines are not capable of clearing the pathogens effectively, only when IL-17A recruits effector cells such as neutrophils and macrophages can the pathogens be cleaned [10,11]. A recent study demonstrated that downstream chemokines of IL-17A signaling pathway were significantly increased in cerebrospinal fluid (CSF) of BM patients which might serve as a neurological prognostic index [12]. Another recent study reported that IL-17 signaling could recruit neutrophils by regulating the secretion of CXCL-1 in mycobacteria-infected mice [13]. IL-17 could be induced by heat-killed pneumococci and pneumococcal peptidoglycan (PGN) when co-cultured with CD4+ T cells and monocytes isolated from healthy volunteers, suggesting a Th17 response to S. pneumonia [14] Single nucleotide polymorphisms of IL-17A may lead to altered production of IL-17A.
Gene encoding human IL-17A is located on chromosome 6. Studies have shown that SNPs in IL-17A are associated with a majority of allergic and autoimmune diseases such as asthma, rheumatoid arthritis and ankylosing spondylitis [15][16][17]. SNPs of IL-17A rs4711998 (-877A>G), rs8193036 (-737C>T), rs2275913 (-197G>A) are found in the upstream of the transcriptional factor binding site (TFBS). Theoretically, these SNPs can influence the expression of nuclear factor of activated T cells (NFAT), which is known to regulate transcription activity of IL-17 [18,19]. Rasouli et al [ 20] reported that Iranian people with IL-17A rs4711998 (-877A>G) AA genotype had an increasing risk to brucellosis. Stappers et al [21] found that Eastern European individuals with IL-17A rs8193036(-737C>T) TT genotype were less susceptible to complicated skin and skin structure infections caused by Staphylococcus aureus, β-haemolytic Streptococcus group A-G, Enterococcus faecalis or Escherichia coli. Another study carried out in Finland showed that healthy children at two years of age who had undetectable or low serum concentration of IL-17A with rs2275913 (-197G>A) AA genotype were more likely to be colonized by S. pneumoniae [22]. These studies suggest that genetic variations of IL-17A can influence susceptibility and outcome of bacterial colonization and infections.
The aim of this study was to explore the possible associations between gene polymorphisms in IL-17A and susceptibility, severity and prognosis of PM in Chinese children. A total of 3 SNPs: IL-17A rs4711998 (-877A>G), rs8193036 (-737C>T) and rs2275913 (-197G>A) ( Figure 1) were analyzed. The selection of these functional SNPs was based on already published data of in vitro and in vivo studies [19,21,22].  [23]. In addition, 330 healthy adults (168 males and 162 females) who attended annual medical examination in June 2015 were randomly selected as the control group. The age of the control group ranged from 21 to 60 years with a median age of 38 years. All study subjects were Chinese from different families and thus considered as unrelated individuals.

Study subjects
Clinical data collection Clinical characteristics of the study subjects were summarized in Table 1 Duplicate wells for each sample were did and a volume of 50μL was used per well.
Samples were diluted to 1:2 with sample diluent provided by the manufacturer before added into wells. The detection limit of IL-17A and CXCL-1 were 2.3pg/mL and 3.0pg/mL, respectively.
Statistical analysis Allelic and genotypic frequencies of each SNP studied were calculated. All of the SNPs analyzed were found to be in HWE (P>0.05). Categorical variables were analyzed by χ2 test or Fisher's exact test as appropriate. Univariate analysis was performed for continuous variables with Mann-Whitney U test or Student's ttest, The Odds Ratio (OR) and 95% confidence interval (CI) were calculated using unconditional binary logistic regression. Genotypes found to be statistically significant by univariate analysis were further analyzed by logistic regression. A two-tailed P value less than 0.05 was considered as significant. Bonferroni correction was employed for multiple testing correction and P < 0.05/n (n =the number of comparisons) was considered significant. All the statistical analyses were conducted using the SPSS software, version 23.0 (IBM, Armonk, NY, USA). Estimation of haplotype frequencies and haplotype association analysis with the χ2 test were performed with SHEsis (http://analysis.bio-x.cn).
The statistical analyses were carried out by using Graph-Pad Prism 6 (San Diego, CA). were CSF culture positive for S. pneumoniae, 35 were CSF positive for duplicate specific pneumococcal antigen and 9 were only blood culture positive for S. pneumoniae. All 95 PM patients had fever at admission.

Comparison of IL-17A promoter variants between patients and controls
The location and distributions of genotypes and haplotype frequencies of three SNPs studied between PM patients and healthy controls (HCs) were summarized in Figure 1 and Table 2.
The genotypes and allele frequencies of 3 SNPs between cases and controls did not differ significantly. However, a significant difference was found for haplotype A-C-A, with 43.6%  (Table 3). However, after logistic regression analyses, significant difference observed between variant genotype GA of rs4711998 (-877A>G) and CSF glucose disappeared.
Frequency of haplotype A-C-A was significantly higher in PM patients with an initial CSF glucose concentration less than 1.5mmol/L compared with those with greater than pneumoniae is an important commensal resident of the human nasopharynx. Although the carriers have no symptoms at first, S. pneumoniae can become invasive and spread from upper respiratory tract to the brain through bloodstream [26]. The mechanism of entering the blood-brain barrier is probably via platelet-activating factor receptor-mediated transcytosis across endothelial and epithelial cell layers [27]. In mouse experiment, S. pneumoniae can also transmigrate from nasopharyngeal epithelium to the central nervous system through olfactory nerves [28]. It is known that host defense has a crucial effect on the incidence and severity of infections caused by this particular pathogen [22]. The IL-17 family includes six members (IL-17A to F) and five receptors (IL-17RA to E) and plays an active role in host immune responses. IL-17A, expressed as a subset of CD4+ effector T (Th17) cells, has a strong biological activity. It has been shown that children with BM exhibit high fractions of γδ IL-17A producing cells in CSF [29]. When the IL-17A downstream signaling pathways which are common to nuclear factor-kappa B (NF-κB) and mitogen-activated protein kinase (MAPK) signaling are activated, the pro-inflammatory factors like granulocyte colony-stimulating factor (G-CSF), IL-6 and CXCL-1 are upregulated. As a result, the neutrophils are induced and recruited to the sites of inflammatory infiltration, where neutrophil cytotoxic and phagocytic activities are strengthened, thereby providing an amplification loop for neutrophil pro-inflammatory responses [30,31]. Previously, Wright et al [ 32] showed that IL-17A production accelerated the clearance of Pneumococcal bacteria by enhancing the anti-pneumococcal response of human alveolar macrophages. Animal experiments also demonstrated that streptococcal-specific Th17 cells could migrate from nasal-associated lymphoid tissue to the brain by traveling along olfactory sensory axons [33]. All these suggest that IL-17A signaling has an effector role in pneumococcal immunity in humans.
The main result of the present study was that the haplotype A-C-A of IL-17A rs4711998 (-877A>G), rs8193036 (-737C>T) and rs2275913 (-197G>A) was more common in PM patients than in the Chinese population controls. And the haplotype may release excessive inflammatory factors by up-regulating the expression of IL-17, resulting in more severe illness. CRP, one of the acute-phase proteins, is primarily produced in the liver during episodes of acute inflammation or infection [34]. It is widely requested by the physicians in order to estimate the presence and severity of infectious and inflammatory diseases. IL-17A, however, plays a crucial role in the neutrophil expansion of progenitor cells and mature neutrophils in bone marrow, spleen and peripheral blood by inducing other inflammatory mediators such as IL-6, G-CSF, CXCL-1 and CXCL-2 [35]. In our study, although we failed to detect IL-17A in serum, we found that the frequency of haplotype A-C-A was higher in patients whose serum CXCL-1 was high (top 25%). Similar trend was also observed for serum CRP.
Multicenter studies have illustrated that low CSF glucose and high CSF protein are strong predictors for severity of patients with BM [36,37]. It is known that the increased number of invasive S. pneumoniae destroy the cells in brain, thus the enzymes related to glycolysis are released, leading to an enhancement of glycolysis activity. Then the insufficient glucose in the CSF can greatly affect neuron energy supply and cause damage to the brain. Furthermore, low levels of glucose in the central nervous system have been associated with increased inflammation and cytokine levels in the CSF which may consequently results in the pathogenesis of meningitis-associated brain injury, leading to certain neurological sequelae [38,39]. As for changes in CSF protein, it is often considered an indicator of blood-brain barrier permeability. Indeed, Asgari et al [40] have shown that high level of albumin was likely to originate from a barrier dysfunction.
It is known that IL-17A can damage BBB function by disrupting tight-junctions in vitro and in vivo through the generation of ROS in endothelial cells [41,42]. In this study, we found a high frequency of PM patients who carry an IL-17A high-producing haplotype A-C-A. Our finding is consistent with those reported above. A recent study conducted among patients with tuberculosis in Argentina found that subjects with AA genotype of rs2275913 SNP, AA individuals displayed significantly higher concentration of serum IL-17A compared with those with GG genotype. Moreover, these subjects with AA genotype had higher disease severity such as displaying the highest bacilli burden, showing severe pulmonary conditions and exhibiting weak cell-mediated immunity against Mtb-Ag [43]. Our result is in agreement with the finding reported among TB patients.
There were certain limitations in this study. First, the gender distribution of patients with PM was not equal. We found the male gender might be a risk factor for susceptibility to PM (OR=1.834, 95%CI: 1.141-2.947, P=0.012). However, contradictory results have been reported. One post hoc analysis showed that gender had no significant effect on prognosis [44]. Another meta-analysis indicated the male gender might be one of the prognostic factors for sequelae and mortality after PM, but no explanations were provided [24]. In our study, we found frequency of haplotype A-C-A significantly increased in PM with lower CSF glucose in both male and female gender, indicating no gender differences in disease severity and prognosis (Supplementary Figure S1). It should be kept in mind that genes encoding IL-17 family members are not located on the sex chromosome of human. Secondly, the healthy adults were used as controls in this study. We recognize that the clinical characteristics and prognosis of PM in pediatric and adult patients are different. However, as stated above, the purpose to use adult as controls was to compare frequencies of different SNPs between patients and population controls and was not meant to compare the clinical characteristics of PM between the two groups. Genotypes do not alter with age, which is widely accepted. Thirdly, we did not find an association between SNPs of IL-17A and proportion of polynuclear cells in CSF or proportion of neutrophils in blood. The explanation could be that the patients with PM included in this study were caused by different serotypes of S. pneumoniae. As shown in animal experiments, IL-17 can have an adverse effect on S. pneumoniae with high capsular thickness and low invasive potential such as serotypes 3 and 6B [45]. Unfortunately, we did not have information on serotypes of pneumococcal strains isolated in our cohorts. On the other hand, in most cases a critical procedure for S. pneumonia to cause BM is to invade the blood-brain barrier through blood flow [46]. As the highly encapsulated pneumococcal strains very seldom could traverse the alveolar epithelial barrier by neutrophil extracellular traps [45]. Fourthly, we did not determine level of CSF cytokines including IL-17A, because the samples were not available. Last, due to the limited sample size, we were unable to find out whether the less frequent haplotypes such as G-C-A, G-T-G, etc.
were related to the susceptibility and severity of PM. In this study, serum IL-17A was not detected in most of patients. One explanation might be that these patients had already antibiotic treatments before admission to the hospital.

Conclusions
Our results indicate that the haplotype A-C-A of IL-17A was associated with susceptibility

Declarations
This study was approved by the Ethics Committee of the Capital Medical University and the Beijing Children's Hospital, Beijing, China. The methods were carried out in accordance with the relevant guidelines, including any relevant details. Written informed consent was taken prior to the study enrollment from their guardians of children.

Availability of data and material
The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflict of interests
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.   There were no statistically significant differences in genotype and allele frequencies of the SNPs studied between cases and controls   Student-t test was used for group comparison (***P <0.001).