Role of the IL33 and IL1RL1 pathway in the pathogenesis of Immunoglobulin A vasculitis

Cytokines signalling pathway genes are crucial factors of the genetic network underlying the pathogenesis of Immunoglobulin-A vasculitis (IgAV), an inflammatory vascular condition. An influence of the interleukin (IL)33- IL1 receptor like (IL1RL)1 signalling pathway on the increased risk of several immune-mediated diseases has been described. Accordingly, we assessed whether the IL33-IL1RL1 pathway represents a novel genetic risk factor for IgAV. Three tag polymorphisms within IL33 (rs3939286, rs7025417 and rs7044343) and three within IL1RL1 (rs2310173, rs13015714 and rs2058660), that also were previously associated with several inflammatory diseases, were genotyped in 380 Caucasian IgAV patients and 845 matched healthy controls. No genotypes or alleles differences were observed between IgAV patients and controls when IL33 and IL1RL1 variants were analysed independently. Likewise, no statistically significant differences were found in IL33 or IL1RL1 genotype and allele frequencies when IgAV patients were stratified according to the age at disease onset or to the presence/absence of gastrointestinal (GI) or renal manifestations. Similar results were disclosed when IL33 and IL1RL1 haplotypes were compared between IgAV patients and controls and between IgAV patients stratified according to the clinical characteristics mentioned above. Our results suggest that the IL33-IL1RL1 signalling pathway does not contribute to the genetic network underlying IgAV.

www.nature.com/scientificreports/ IgAV patients stratified according to the clinical characteristics mentioned above. Our results suggest that the IL33-IL1RL1 signalling pathway does not contribute to the genetic network underlying IgAV.
Immunoglobulin-A vasculitis (IgAV), formerly called Henoch-Schönlein purpura (HSP), is an inflammatory small-sized blood vessel disease, more common in children and rarer but more serious in adults [1][2][3] . The defining pathophysiologic feature of this vasculitis is the IgA1-predominant immune deposits in the vessel walls 1 . Although IgAV usually involves the skin, joints and the gastrointestinal (GI) tract [4][5][6][7][8] , nephritis is also common in affected patients and constitutes the most feared complication of this vasculitis [1][2][3][4][5] . IgAV has a multifactorial aetiology in which genetics plays a relevant role in both the predisposition and clinical course of the disease 6,9,10 . In this regard, outside the human leukocyte antigen region, cytokines signalling pathway genes constitute a key component of the genetic network underlying IgAV 6,[11][12][13][14] . Interleukin (IL)-33 is a cytokine that belongs to the IL-1 family 15 . This molecule exerts its biological functions by binding to its receptor, IL-1 receptor like 1 (IL-1RL1) (also known as suppression of tumorigenicity 2 or ST2), a member of the Toll/IL-1 receptor superfamily 16 . This binding leads to the activation of mast cells and Th2 lymphocytes and, consequently, to the production of chemokines, pro-inflammatory and Th2-associated cytokines, as well as increased serum immunoglobulin levels 15 . In accordance with that, a pathogenic role of the IL-33 axis in autoimmunity has been described [17][18][19] . In addition, several lines of evidence demonstrate that genetic variants located both in the IL33 and IL1RL1 genes are implicated in the increased risk of numerous immune-mediated diseases [20][21][22][23][24][25][26][27][28] .
Taking into account all these considerations, this study aimed to determine, for the first time, the potential implication of the IL33-IL1RL1 signalling pathway in the pathogenesis of IgAV. For this purpose, we genotyped three tag genetic variants within IL33 and three tag polymorphisms within IL1RL1, which cover the major variability of these loci and that were previously associated with several inflammatory diseases, in the largest series of Caucasian IgAV patients ever assessed for genetic studies.  Table 1. For GI manifestations, bowel angina was considered present if there was diffuse abdominal pain that worsened after meals or bowel ischemia usually with bloody diarrhoea. GI bleeding was defined as the presence of melena, haematochezia, or a positive test for occult blood in the stool. Renal manifestations were defined to be present if at least one of the following findings was observed: haematuria, proteinuria, or nephrotic syndrome at any time over the clinical course of the disease and/or renal sequelae (persistent renal involvement) at last follow-up.

Patients and methods
In addition, a set of 845 sex and ethnically matched healthy controls without history of cutaneous vasculitis or any other autoimmune disease, constituted by blood donors from Hospital Universitario Marqués de Valdecilla (Santander) and National DNA Bank Repository (Salamanca), was also included in this study.
For experiments involving humans and the use of human blood samples, all the methods were carried out in accordance with the approved guidelines and regulations, according to the Declaration of Helsinki. All experimental protocols were approved by the Single nucleotide polymorphisms selection and genotyping methods. Three tag genetic variants within IL33 (rs3939286, rs7025417 and rs7044343) and three tag polymorphisms within IL1RL1 (rs2310173, rs13015714 and rs2058660) were selected in this study. Regarding IL33, rs3939286 is situated in an intergenic region close to IL33, whereas rs7025417 and rs7044343 are located in the promoter and in intron 5 of the gene, respectively. With respect to IL1RL1, both rs2310173 and r2058660 are placed close to IL1RL1, while rs13015714 is downstream of the gene. Interestingly, these six polymorphisms map and tag both IL33 and IL1RL1 regions as previously described 20 . In addition, the IL33 and IL1RL1 genetic variants selected in our study were previously associated with several immune-mediated diseases [20][21][22][23][24][25][26][27][28] . In particular, IL33 rs3939286 was previously implicated in the development of subclinical atherosclerosis in patients with rheumatoid arthritis (RA) 20 as well as in the pathogenesis of inflammatory bowel (IBD) disease 21 and asthma 22 , whereas IL33 rs7025417 was previously associated with coronary artery disease 23  www.nature.com/scientificreports/ in ankylosing spondylitis 25 and ulcerative colitis 26 , as well as an association of rs13015714 with celiac disease 27 and IBD 21 , and an influence of rs2058660 on the development of Crohn's disease 28 were previously reported. Genomic deoxyribonucleic acid (DNA) from all the individuals included in the study was extracted from peripheral blood using standard procedures.
Genotyping was performed in a QuantStudioTM 7 Flex Real-Time polymerase chain reaction system, according to the conditions recommended by the manufacturer (Applied Biosystems, Foster City, CA, USA).
Negative controls and duplicate samples were included to check the accuracy of the genotyping.
Statistical analyses. Genotype data were checked for deviation from Hardy-Weinberg equilibrium (HWE). Differences in IL33 and IL1RL1 frequencies were evaluated between patients with IgAV and healthy controls and between patients with IgAV stratified according to specific clinical characteristics of the disease (age at disease onset or presence/absence of GI or renal manifestations).
First, each IL33 and IL1RL1 polymorphism was analysed independently. Both genotype and allele frequencies were calculated and compared between the groups mentioned above by chi-square test. Strength of association was estimated using odds ratios (OR) and 95% confidence intervals (CI).
Then, allelic combinations (haplotypes) of both IL33 and IL1RL1 polymorphisms were carried out. Haplotype frequencies were calculated by the Haploview v4.2 software (http:// broad. mit. edu/ mpg/ haplo view) and then compared between the groups mentioned above by chi-square test. Strength of association was estimated by OR and 95% CI.
P-values were two-tailed and those lower than 0.05 were considered as statistically significant. All analyses were performed with the STATA statistical software 12/SE (Stata Corp., College Station, TX, USA).
Genotype and allele frequencies of each IL33 and IL1RL1 polymorphism were similar to that reported for populations of European origin in the 1000 Genomes Project (http:// www. inter natio nalge nome. org/).

Differences in IL33
and IL1RL1 genotype and allele frequencies between patients with IgAV and healthy controls. In a first step, we compared genotype and allele frequencies of each IL33 and IL1RL1 variant assessed independently between patients with IgAV and healthy controls.
As shown in Table 2, similar genotype and allele IL33 and IL1RL1 frequencies were observed in patients with IgAV when compared to healthy controls ( Table 2).
Differences in IL33 and IL1RL1 genotype and allele frequencies between patients with IgAV stratified according to specific clinical characteristics of the disease. Then, potential differences www.nature.com/scientificreports/ in the genotype and allele frequencies of each IL33 and IL1RL1 genetic variant were evaluated between patients with IgAV stratified according to specific clinical characteristics of the disease. Given that IgAV is often a benign and self-limited pathology in children and a more severe condition in adults, we assessed if potential differences in IL33 and IL1RL1 could exist in IgAV patients stratified according to the age at disease onset. With this respect, and as shown in Table 3, similar genotype and allele IL33 and IL1RL1 frequencies were disclosed when children (age ≤ 20 years) were compared to adults (age > 20 years) ( Table 3).
In addition, we compared genotype and allele frequencies of IL33 and IL1RL1 in patients with IgAV stratified according to the presence/absence of GI or renal manifestations. Accordingly, similar genotype and allele IL33 and IL1RL1 frequencies were disclosed when patients with IgAV who developed GI manifestations were compared to those who did not exhibit these complications (Table 3). This was also the case when patients with IgAV were stratified according to the presence/absence of renal manifestations (Table 3).
Haplotype analyses of IL33 and IL1RL1. In a further step, we also examined whether haplotype frequencies of both IL33 and IL1RL1 differed between patients with IgAV and healthy controls as well as between patients with IgAV stratified according to the specific clinical characteristics of the disease mentioned above.
The haplotype analysis did not yield additional information since IL33 and IL1RL1 haplotypes frequencies were similar in patients with IgAV when compared to healthy controls (Table 4). This was also the case when patients with IgAV were stratified according to the age at disease onset and the presence/absence of GI or renal manifestations (Supplementary Tables S1-S3 online).
Based on these considerations, in this study we aimed to determine whether the IL33-IL1RL1 signalling pathway is also implicated in the pathogenesis of IgAV, a small-vessel vasculitis in which cytokines signalling pathway genes are considered as an essential component in its etiology 6,[11][12][13][14] . For that purpose, six polymorphisms within the IL33-IL1RL1 signalling pathway, which cover most of the variability of both IL33 and IL1RL1 regions and that were previously related to numerous immune-mediated diaseases [20][21][22][23][24][25][26][27][28] , were evaluated in the largest series of Caucasian patients with IgAV ever assessed for genetic studies. Our results showed no influence of the IL33-IL1RL1 signalling pathway on the susceptibility to IgAV when we studied each of the polymorphisms separately. Moreover, when we analysed IL33 genetic variants as well as IL1RL1 polymorphisms together conforming haplotypes, our results revealed a lack of association between the IL33-IL1RL1 signalling pathway and IgAV susceptibility. Furthermore, no specific association of IL33 and IL1RL1 polymorphisms (assessed independently or combined conforming haplotypes) with clinical features of IgAV was observed in our study, indicating that this pathway does not represent a risk factor for the severity of the disease.
Data on the potential effect of the IL33-IL1RL1 signalling pathway on vasculitis were scarce so far. A study performed in Caucasians evaluated the role of this pathway in the susceptibility to giant cell arteritis (GCA) 34 , another primary systemic vasculitis that involves large and middle-sized blood vessels 31,35 . In keeping with our results, no influence of IL33 rs3939286, IL33 rs7044343, IL1RL1 rs2310173, IL1RL1 rs13015714 and IL1RL1 rs2058660 on GCA was detected 34 . By contrast, a consistent association between IL33 rs7025417 and GCA was Table 2. Genotype and allele frequencies of IL33 and IL1RL1 genes in patients with IgAV and healthy controls. IgAV: IgA vasculitis; SNP: single nucleotide polymorphism. No statistically significant differences in the IL33 and IL1RL1 genotype and allele frequencies were disclosed when patients with IgAV were compared to healthy controls (p ≥ 0.05 in all the cases).

Change
Genotypes, % (n) Alleles, % (n) www.nature.com/scientificreports/ found, pointing out this polymorphism as a novel genetic risk factor for GCA susceptibility 34 . In addition, the potential role of an IL33 polymorphism (rs1342326) in the pathogenesis of Behçet's disease, a condition affecting blood vessels of all sizes and types 31 , was assessed in Iranians 36 . With this respect, an effect of this genetic variant in the susceptibility to this disease was disclosed 36 .
The results derived from our study are of potential clinical relevance. Vasculitides comprises a heterogeneous group of autoimmune diseases that often have overlapping clinical and pathological manifestations 37 . Although their complex aetiology is far from being completely understood, differences among them in molecular terms have been postulated 38 . Our data support this hypothesis since no implication of the IL33-IL1RL1 signalling pathway in the pathogenesis of IgAV in Caucasians was disclosed, unlike GCA 34 .  www.nature.com/scientificreports/