July 2021
Volume 62, Issue 9
Jump To...
Open Access
Immunology and Microbiology  |   July 2021
Next-Generation HLA Sequence Analysis Uncovers Shared Risk Alleles Between Clinically Distinct Forms of Childhood Uveitis
Roos A. W. Wennink; Joke H. de Boer; Sanne Hiddingh; Anne-Mieke J. W. Haasnoot; Viera Kalinina Ayuso; Talitha de Hoop; Jessica van Setten; Eric Spierings; Jonas J. W. Kuiper
Author Affiliations & Notes
  • Roos A. W. Wennink
    Department of Ophthalmology, University Medical Center Utrecht, Utrecht University, The Netherlands
    Center of Translational Immunology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Joke H. de Boer
    Department of Ophthalmology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Sanne Hiddingh
    Center of Translational Immunology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Anne-Mieke J. W. Haasnoot
    Department of Ophthalmology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Viera Kalinina Ayuso
    Department of Ophthalmology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Talitha de Hoop
    Center of Translational Immunology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Jessica van Setten
    Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Eric Spierings
    Center of Translational Immunology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Jonas J. W. Kuiper
    Department of Ophthalmology, University Medical Center Utrecht, Utrecht University, The Netherlands
    Center of Translational Immunology, University Medical Center Utrecht, Utrecht University, The Netherlands
  • Correspondence: Jonas J.W. Kuiper, University Medical Center Utrecht, Department of Ophthalmology, E03.136, PO Box 85500, 3508 GA Utrecht, The Netherlands; j.j.w.kuiper@umcutrecht.nl. 
Investigative Ophthalmology & Visual Science July 2021, Vol.62, 19. doi:https://doi.org/10.1167/iovs.62.9.19
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      User Alerts
      You are adding an alert for:
      Next-Generation HLA Sequence Analysis Uncovers Shared Risk Alleles Between Clinically Distinct Forms of Childhood Uveitis
      You will receive an email whenever this article is corrected, updated, or cited in the literature. You can manage this and all other alerts in My Account
      The alert will be sent to:
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation
      Citation

      Roos A. W. Wennink, Joke H. de Boer, Sanne Hiddingh, Anne-Mieke J. W. Haasnoot, Viera Kalinina Ayuso, Talitha de Hoop, Jessica van Setten, Eric Spierings, Jonas J. W. Kuiper; Next-Generation HLA Sequence Analysis Uncovers Shared Risk Alleles Between Clinically Distinct Forms of Childhood Uveitis. Invest. Ophthalmol. Vis. Sci. 2021;62(9):19. https://doi.org/10.1167/iovs.62.9.19.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

      ×
    • Get Permissions
  • Supplements
Abstract

Purpose: Classical alleles of the human leukocyte antigen (HLA) complex have been linked to specific entities of pediatric noninfectious uveitis, yet genetic predisposition encoded by the HLA super-locus across the patient population remains understudied.

Methods: We performed next-generation full-length sequencing of HLA-A, HLA-B, HLA-C, HLA-DPB1, HLA-DQB1, and HLA-DRB1 in 280 cases. Dense genotype data from 499 Dutch controls from Genome of the Netherlands were imputed using an HLA-specific reference panel (n = 5225 samples from European ancestry). Cases and controls were compared using logistic regression models adjusting for sex.

Results: In total, 179 common and rare alleles were detected. Considering all cases and controls, HLA-DQB1*04:02 and HLA-DRB1*08:01 were identified as the principal HLA association, which was mainly driven by 92 cases with juvenile idiopathic arthritis-associated uveitis (JIA-U). The HLA-DQB1*04:02-HLA-DRB1*08:01 haplotype was also the primary association for the phenotypically similar idiopathic chronic anterior uveitis without arthritis (CAU). Also, HLA-DQB1*05:03 was an independent risk allele for CAU, but not in JIA-U. Analysis of 185 cases with other forms of uveitis revealed HLA-wide associations (P < 2.79 × 10−4) for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03, which could be primarily attributed to cases with panuveitis. Finally, amino acid substitution modeling revealed that aspartic acid at position 57 that distinguishes the risk allele HLA-DQB1*05:03 (for CAU and panuveitis) from nonrisk alleles, significantly increased the binding capacity of naturally presented ligands to HLA-DQ.

Conclusions: These results uncovered novel shared HLA associations among clinically distinct phenotypes of pediatric uveitis and highlight genetic predisposition affecting the antigen presentation pathway.

Noninfectious pediatric uveitis poses a significant burden on the health and quality of life of affected children, because of its high risk of visual impairment and its lifelong dependency on specialized care.1,2 Despite the variety of clinical phenotypes, little is understood of the underlying disease mechanisms that render some children prone to develop uveitis. Consequently, due to lack of understanding their causes, at least half of cases are considered idiopathic and therefore left to be classified according to the affected areas (i.e., anterior uveitis describes cases with a predominant involvement of the anterior segment of the eye).3 Regardless, immune-mediated etiology is strongly suspected because pediatric uveitis occurs often in cases suffering from autoimmune diseases (e.g., rheumatic disease). In fact, the largest category (12%–30%) of pediatric uveitis is cases suffering from anterior uveitis associated with preexisting juvenile idiopathic arthritis (JIA).4 In addition, immunosuppressive therapy regimes are effective in dampening eye inflammation and further support that derailed immunity is an important contributor of disease.5–8 
Genetic studies that strongly linked several alleles of the family of human leukocyte antigen (HLA) genes on chromosome 6 to noninfectious uveitis add to the body of evidence.9–15 HLA molecules are critically involved in adaptive immunity by presenting protein fragments sampled from the cellular (class I HLA molecules) and extracellular (class II HLA molecules) environment to surveilling T and natural killer cells to orchestrate antigen-specific immunity.16 Genetic studies of pediatric uveitis have mostly been conducted in JIA-U cases of Western European ancestry, which mapped the primary genetic risk for uveitis in JIA to amino acid positions 10 to 12 encoded by the HLA–DRB1 gene.14 This uveitis motif exhibited a disease risk in females which was independent from the known genetic association of JIA with other amino acid positions in HLA-DRB1.14 This example demonstrated that different polymorphisms in the same gene may confer independent disease risks and supported that uveitis is a distinct trait of the pediatric population.14,15 Other HLA typing studies have been conducted in disease-specific contexts, such as sequence specific typing association studies in small cohorts of cases with tubulointerstitial nephritis and uveitis (TINU) syndrome.17–22 At present, HLA typing studies across a representative sample of pediatric uveitis population are sparse. Here, we performed allelic resolution next-generation sequencing across all classical class I and class II HLA alleles to refine HLA associations and amino acids polymorphisms at the population level and stratify HLA haplotypes according to anatomical categories and specific entities. 
Methods
Patient and Material Collection
The study was approved by the Medical Ethical Research Committee of the University Medical Center Utrecht in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients ≥ 18 years, from both parents and patients in cases between 12 to 18 years of age, and from parents in cases < 12 years old. Peripheral blood samples (EDTA tubes) were obtained from patients with juvenile idiopathic arthritis-associated uveitis (JIA-U, n = 92), idiopathic chronic anterior uveitis (i.e., no JIA) (CAU, n = 50), (HLA-B27-positive) acute anterior uveitis (n = 18), intermediate uveitis (IU, n = 39), and panuveitis (PAN, n = 81). The diagnosis of uveitis was established by a trained uveitis specialist according to the SUN criteria.23 All cases were recruited at the University Medical Center Utrecht, the Netherlands, a tertiary referral center. The study only considered patients with a diagnosis of noninfectious uveitis before the age of 18 years. Juvenile idiopathic arthritis was diagnosed according to the criteria of the International League of Associations for Rheumatology or by former criteria (e.g., European League Against Rheumatism).24,25 Patients with JIA were screened by an ophthalmologist according to the guidelines of the Academy of Pediatrics.26 Patients with 1+ cells or more in the anterior chamber and treated with at least topical steroids during ophthalmologic examinations, were diagnosed with JIA-U. The diagnosis for (probable or possible) TINU syndrome was made by an ophthalmologist specialized in pediatric uveitis and a pediatric nephrologist based on clinical criteria.17 
HLA Typing by High-Resolution Next-Generation Sequencing
DNA was isolated from 300 µL of peripheral blood (EDTA tubes) using the MagNa Pure Compact (Roche Diagnostics, the Netherlands). From genomic DNA, coding sequences of the HLA region were enriched by PCR using multiplexed HLA amplification by NGSgo-MX6-1 PCR (GenDx, Utrecht, the Netherlands). Next-generation sequencing was performed on all samples using the Illumina sequencing system. The DNA fragments were analyzed with the NGSEngine software package version 2.14.0 (GenDx, Utrecht, The Netherlands) using the IPD-IMGT/HLA database version 3.36.0. After inspection of the full-length sequence, the genotype data for 179 distinct HLA alleles were reduced to second field nomenclature to reflect the unambiguous phenotype. The corresponding amino acid sequences for each of the 179 alleles was obtained from the the Immuno Polymorphism Database (IPD) and the international ImMunoGeneTics (IMGT)/HLA release 3.2927 (Supplementary Table S1A-F). A total of 293 amino acid positions were considered polymorphic and used for fine-mapping of amino acid associations (Supplementary Table 6). 
Reference Cohort
Genotype data from 499 unrelated and unaffected Dutch subjects from the Genome of the Netherlands (GoNL) were used as population controls.28 The GoNL project is a human genetic project based on a representative sample of Dutch citizens from all provinces in the Netherlands, which captures the genetic variation in the Dutch population (for more information see http://www.nlgenome.nl). Next, we used BEAGLE package to prephase the GoNL single-nucleotide polymorphism (SNP) data for imputation. In other words, we estimated the haplotypes for each individual GoNL sample for efficient imputation of ungenotyped variants based on linkage disequilibrium.29 We imputed classical HLA alleles using SNP data and analysis with the SNP2HLA pipeline, which given the strong linkage disequilibrium in the MHC region, can be used to tag the majority of HLA alleles. The SNP2HLA pipeline uses a reference panel assembled through HLA typing of 5225 European-ancestry individuals collected by the Type 1 Diabetes Genetics Consortium to ascertain accurate HLA imputation.30 
HLA Association Analysis
Sequence data were converted to dosage data by coding the presence or absence of each classical allele or encoded amino acid residue at indicated positions to 0,1, or 2. This generated three possible genotypes: homozygous positive (2 alleles), heterozygous positive (1 allele), and homozygous negative (0 alleles) for any of the classical HLA alleles or amino acid residues at a given position. To visualize the ethnicity of the cases and controls, we used Uniform Manifold Approximation and Projection (UMAP)31 based on the top 60 principal components calculated using the dosage data. For this analysis we also included HLA genotypes from populations of the 1000 Genomes Project (Supplementary Table 2).32,33 HLA association testing was performed using PLINK 2.0 using an additive logistic regression model with sex as a covariate.34 To identify potential independent associations for HLA alleles and amino acid residues, we performed a series of association analyses adjusting for the primary association (i.e., conditional analysis), by adding each significantly associated allele as a covariate to the model. Our data enabled multiple comparisons: 1) pediatric noninfectious uveitis versus controls, 2) JIA-U versus controls, 3) non-JIA-U versus controls, 4) CAU versus controls, 5) IU versus controls, and 6) PAN versus controls. A P value threshold of 2.79 × 10−4 (Bonferroni correction; 0.05/179 detected HLA alleles) was used as a threshold for significance to prioritize reporting of HLA associations and termed this threshold “HLA-wide” significance. 
Amino Acid Substitution Modeling
To explore whether the presence of naturally occurring amino acids at position 57 of the different HLA-DQB1*05 alleles affected the binding of peptide cargo to the HLA molecule, we extracted 157 ligands identified by mass spectrometry (from the HLA ligand atlas) from elutions of immunoprecipitations of HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03.35 Replacement of position 57 with the amino acids at that position in the HLA-DQB1*05 alleles was done in the NetMHCIIpan 4.0 server using the amino acid sequence of HLA-DQB1*05 (UniProt accession: Q9GIK3) and the binding of each of the 157 predicted ligands using default settings.36 Differences in the median binding score were calculated using a Wilcoxon signed rank test. We computed the peptide motif of the 157 peptide ligands using the artificial neural network model of NNAlign-2.0, which used the amino acid sequences and binding scores for each peptide for HLA-DQB*05:03 as input (trained with a motif length of 9 and default settings for “MHC CLASS II ligands”) to return a peptide core sequence alignment.37 We also did this for a subset of the 157 peptides with a relatively stronger binding score for HLA-DQB*05:03 (D57) compared to DQB1:05:02 (V57), by computing the ratio of the binding scores (the log-transformed IC50 affinity scores from NetMHCIIpan 4.0) for all 157 peptides for those two alleles and extracting the top 50 peptides with the largest ratio. 
Results
In total 280 cases with uveitis were genotyped of which 160 (57%) patients were diagnosed with anterior uveitis, 39 (14%) patients with intermediate uveitis, and 81 (29%) patients with panuveitis (Table 1). Summary of the HLA genotypes at four-digit resolution revealed 110 unique common alleles (≥1% detected in controls) and 69 rare alleles (<1% in controls) of which included 32 HLA-A alleles, 51 HLA-B alleles, 22 HLA-C alleles, 23 HLA-DQB1 alleles, 13 HLA-DPB1 alleles, and 38 HLA-DRB1 alleles (Supplementary Table 3). Uniform manifold approximation and projection was used to visualize the ethnicity of the cases and controls (Supplementary Table 2). As expected, cases and controls were predominantly mapped to the European population (Fig. 1). 
Table 1.
 
View Table
 
Diagnosis of the 280 Uveitis Cases
Table 1.
 
Diagnosis of the 280 Uveitis Cases
  Diagnosis of the 280 Uveitis Cases
Figure 1.
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
View OriginalDownload Slide
 
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
Figure 1.
 
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
View OriginalDownload Slide
HLA-DRB1 and HLA-DQB1 Are Risk Alleles for Pediatric Uveitis
Next, we tested for HLA associations considering all uveitis cases (n = 280). We identified seven associated HLA alleles and mapped the primary association to HLA-DQB1*04:02 (odds ratio [OR] = 5.27, 95% confidence interval [95% CI] 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8; Fig. 2A). The results from the association testing of the classical HLA alleles in all cases and controls are presented in Supplementary Table 3. All of the associated alleles were common (allele frequency in controls > 1%). Since the haplotype DRB1*08:01-DQB1*04:02 has been previously associated with polyarticular rheumatoid factor-negative JIA,38,39 we extracted the data from the 95 JIA cases and conducted association testing. For JIA-U versus controls, we ascertained the strong association to known JIA risk alleles HLA-DRB1*08:01 (OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12; Fig. 2B). Next, we investigated all independent HLA associations by using conditional analysis. To this end we adjusted for the strongest HLA association by adding the allele as a covariate to the regression model. Adjusting for HLA-DRB1*08:01 mitigated the association for HLA-DQB1*04:02 due to strong linkage disequilibrium (i.e., correlation [LD]) between these alleles (D′  =  0.97, r2  =  0.86 in our data, Table 2). However, adjusting for HLA-DRB1*08:01 revealed an independent association for HLA-DRB1*11:04 (Table 2). Finally, after adjusting for all three HLA alleles (DRB1*08:01-DQB1*04:02-HLA-DRB1*11:04), no residual HLA-wide association signal was observed (P > 2.79 × 10−4). 
Figure 2.
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
 
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Figure 2.
 
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
Table 2.
 
View Table
 
HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
Table 2.
 
HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
  HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
Association testing in the 185 uveitis cases without JIA revealed a primary association for HLA-DRB1*01:02 (OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6; Fig. 2C). Adjusting for HLA-DRB1*01:02 revealed an independent association for HLA-DRB1*04:03 (OR = 11.67, 95% CI 3.82–35.66; P = 1.62 × 10−5). Adjusting for both DRB1*01:02 and DRB1*04:03 revealed an additional independent association for HLA-DQB1*04:02 (OR = 3.82, 95% CI 2.01–7.25; P = 4.21 × 10−5). Next, adjusting for DRB1*01:02, DRB1*04:03, and DQB1*04:02, revealed an independent association for HLA-DQB1*05:03 (OR = 3.31, 95% CI 1.87–5.83; P = 3.65 × 10−4). And finally, adjusting for DRB1*01:02, DRB1*04:03, DQB1*04:02, and DQB1*05:03, identified an independent association for HLA-DRB1*14:01 (OR = 0.01, 95% CI 0.002–0.12; P = 8.96 × 10−5) (Table 2).After adjusting for DRB1*01:02, DRB1*04:03, HLA-DQB1*04:02, DQB1*05:03, and DRB1*14:01, no additional HLA-wide associations were detected (P > 2.79 × 10−4). 
The JIA Risk HLA-DRB1*08:01-HLA-DQB1*04:02 Haplotype Is Associated with Idiopathic Chronic Anterior Uveitis
Next, we stratified patients according to the anatomical location of uveitis. The frequency of the classical HLA alleles in all cases according to the anatomical location of their inflammation and controls are presented in Supplementary Table 4. 
Idiopathic chronic anterior uveitis (CAU) is phenotypically similar to JIA-U, albeit without clinical evidence of arthritis. Remarkably, when testing for 50 CAU cases versus controls, we observed an HLA-wide significant association for the JIA risk loci HLA-DQB1*04:02 (OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5) and HLA-DRB1*08:01 (OR = 5.30, 95% CI 2.26–12.44; P = 1.30 × 10−4; Table 3, Fig. 3A, and Supplementary Table 4). As expected, after adjusting for HLA-DQB1*04:02, the association signal for HLA-DRB1*08:01 in CAU was lost, but in contrast to JIA-U revealed independent association for HLA-DQB1*05:03 (OR = 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5; Fig. 3B). Further adjusting for HLA-DQB1*05:03 removed the remainder of the signal (Fig. 3C). Note that although we considered the number of cases with acute anterior uveitis too low (n = 18) to be meaningful for disease-specific testing, the allele frequency of HLA-DQB1*04:02 or HLA-DRB1*08:01 was nearly identical to CAU, thus also higher compared to controls (Supplementary Table 4). In summary, these data show that the HLA-DRB1*08:01-HLA-DQB1*04:02 haplotype is associated with pediatric anterior uveitis. 
Table 3.
 
View Table
 
Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
Table 3.
 
Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
  Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
Figure 3.
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
 
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Figure 3.
 
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 Are Risk Alleles for Pediatric Panuveitis
We did not observe an HLA-wide significant association for intermediate uveitis—most likely due to the limited number of cases with this subtype in this study (n = 39)—but detected HLA-DRB1*15:01 (OR = 2.18, 95% CI 1.26–3.77; P = 0.005) as the lead allele, which is in line with previous observations11,12 (Table 3 and Supplementary Table 4). We detected HLA-DRB1*01:02 (OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8) as the primary association in patients with panuveitis (Table 3, Fig. 3D, and Supplementary Table 4). We then adjusted for HLA-DRB1*01:02, which revealed independent association for the rare allele HLA-DRB1*04:03 (OR = 29.61, 95% CI 7.96–110.22; P =4.35 × 10−7; Fig. 3E). Adjusting for both DRB1*01:02 and DRB1*04:03, we discovered an additional independent association for HLA-DQB1*05:03 (OR = 4.51, 95% CI: 2.27–8.96; P = 1.72 × 10−5; Fig. 3F). Finally, conditioning on all three risk HLA alleles removed the remainder of the signal (P > 2.79 × 10−4; Fig. 3G). The haplotype HLA-DRB1*01:02 and HLA-DQB1*05 have been previously reported in patients suspected of TINU syndrome.19–22 However, removing these patients (n = 15) from the analysis modestly affected the signal (HLA-DRB1*01:02, P = 1.61 × 10−6, and HLA-DQB1*05:03, P = 8.59 × 10−4) and indicates that the association signal at DRB1*01:02 and DQB1*05:03 represents a signal beyond the association with TINU. 
Sex-Specific Differences in Associations with HLA Alleles
Because we previously detected sexual dimorphism in uveitis associated with JIA,14 we were interested to test whether males and females showed differential association with HLA alleles (Supplementary Table 5). The distribution of male and females across each group is shown in Table 4. Several notable observations were made; Considering 27 males with JIA-U versus 250 male controls, we detected association for a novel risk allele HLA-DRB1*11:01 (OR = 6.25, 95% CI 2.49–15.72; P = 9.71 × 10−5), which was not associated with females (P = 0.32). In analysis of 65 females with JIA-U, we observed a significant association for HLA-DRB1*11:04 (OR = 10.16, 95% CI 3.88–26.58; P = 2.32 × 10−6), which was not associated with males (P = 0.21). 
Table 4.
 
View Table
 
The Distribution of Males and Females in the Investigated Groups
Table 4.
 
The Distribution of Males and Females in the Investigated Groups
  The Distribution of Males and Females in the Investigated Groups
Testing female patients (n = 33) with CAU versus 249 female controls revealed HLA-DQB1*05:03 (OR= 13, 95% CI 4.17–40.49; P = 9.75 × 10−6) as the primary association over HLA-DQB1*04:02 (P = 4.95 × 10−4). However, our cohort of CAU cases may not have sufficient power to accurately assign one risk allele over another. Therefore, we ran the association test at HLA-DQB1*05:03 for CAU and included sex as an interaction term in the model. With this test you can examine if sex influences the HLA association in general. This analysis showed a moderate, but significant effect for the interaction term (Pinteraction = 0.03), indicating that the HLA-DQB1*05:03 signal was indeed sex specific. We also ran interaction tests for the two JIA-U sex-specific alleles HLA-DRB1*11:01 and HLA-DRB1*11:04, which revealed similar interactions with sex (Pinteraction = 0.0006 and Pinteraction = 0.06, respectively). These data show sex-specific differences in the association with risk HLA alleles. 
HLA Amino Acid Associations With Uveitis
Next, we analyzed the association of HLA amino acid polymorphisms with uveitis to assess the contribution of specific amino acid positions (Supplementary Table 1A-F and Supplementary Table 6). Univariate analysis identified 12 amino acids at 12 positions for JIA-U and CAU, eight amino acids at eight positions for IU, and two amino acids at two positions for panuveitis (Supplementary Table 7A and 7B). However, none of the amino acids showed a stronger association compared to the classical alleles. (Supplementary Table 7A). In other words, no polymorphic amino acid position could explain the data better than the classical HLA alleles. 
Aspartic Acid (D) at Position 57 Increases the Peptide Binding Affinity to HLA-DQ
We aimed to explore the functional implications of risk HLA alleles in pediatric uveitis. The amino acid position 86 distinguishes the JIA-U associated alleles HLA-DRB1*11:01 (86-Glycine) in males, HLA-DRB1*11:04 (86-Valine) in females (Supplementary Table 1E). This position has been shown to control DR alpha/beta dimer stability, but also peptide binding for the panuveitis-associated HLA-DRB1*04:03, and is therefore likely to influence antigen presentation.40,41 Therefore, we asked whether additional allele-defining positions affect the binding of peptides to HLA allotypes encoded by risk HLA alleles. The HLA-DQB1*05:03 was found to be associated with two distinct forms of uveitis, CAU and PAN, and this allele is functionally understudied compared to other uveitis risk alleles detected in this study. This allele harbors an aspartic acid (D) at position 57 while other nonrisk alleles for HLA-DQB1*05 have a serine (S) or valine (V) at position 57 (Fig. 4A). Curiously, the amino acid position 57 of the DQ betachain is located at the edge of the peptide binding groove (Fig. 4A) and confers risk to several autoimmune conditions suggesting it to be involved in peptide binding of HLA-DQ.42–46 To assess whether the presence of aspartic acid at position 57 functionally affects binding of peptide cargo to HLA-DQ, we extracted ligands (n = 157 from the HLA ligand atlas) from immunopeptidomes of HLA-DQB1*05:01 (57-S), HLA-DQB1*05:02 (57-V), and HLA-DQB1*05:03 (57-D) that only differ in the amino acid residue at position 57 (Supplementary Table 8).35 Next, we modeled the binding affinity of the naturally occurring ligands to HLA-DQ, by using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0 and performed amino acid substitutions at position 57.36 We observed that substitution of position 57 from S or V (nonrisk alleles) to the uveitis-associated amino acid aspartic acid (D) (equivalent to DQB1*05:03) resulted in a significant increased binding affinity score (S versus D; P = 1.67 × 10−27 and V versus D; P = 1.64 × 10−27; Fig. 4B). In more detail, major binding pockets for HLA-DQ are position P4, P6, and P9 in the peptide motif.47 Therefore, we computed the peptide motif of the 157 HLA-DQB1:05 ligands and the top 50 peptides with the largest difference in binding scores between HLA-DQB1*05:02 (57-V) and HLA-DQB1*05:03 (57-D) and compared the core sequence alignments to identify changes in specific amino acid position. This analysis revealed, as expected, a marked change in amino acid residue preference in particular at anchor position P9, which directly interacts with position 57 (Fig. 4C). Collectively, these data show that position 57 in the HLA-DQ beta chain modulates the peptide binding capacity and indicates risk HLA alleles associated with pediatric uveitis show altered antigen presentation capacity. 
Figure 4.
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36 C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
View OriginalDownload Slide
 
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
Figure 4.
 
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36 C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
View OriginalDownload Slide
Discussion
The role of the human MHC locus in determination of the susceptibility to uveitis is evident from the routine screening for HLA alleles (more specifically HLA-B27 and HLA-A29) to support the diagnosis of specific entities of noninfectious uveitis.48 One of the outcomes of this study is that we detected a predominant contribution of common alleles to disease risk that allowed accurate comparison with available data from population controls,28 but also detected the less common or rare allele HLA-DRB1*04:03 (allele frequency of 0.7% in controls) as a novel risk allele for panuveitis. This supports that common and rare alleles define the susceptibility to uveitis in children. 
Previous genome-wide association studies have consistently identified HLA class I and class II genes as the primary genetic risk factors for noninfectious uveitis.10–15 However, these studies were based primarily on HLA imputation methods that despite their overall high accuracy to identify common alleles may be less adequate for the identification of rare or poorly characterized alleles. We therefore used sequencing data to comprehensively determine the full sequence at high resolution of class I and class II HLA alleles for a large cohort of patients with pediatric uveitis. Note imputed HLA alleles from population controls were nearly identical to the allele frequencies in Europe (Supplementary Table 9).49 
Our study revealed that HLA-DQB1*04:02 and HLA-DRB1*08:01 (in strong LD) are risk alleles for both JIA-U and CAU. Idiopathic chronic anterior uveitis is clinically similar to JIA-U uveitis, but these patients do not present with any evidence for inflammatory arthritis.50 A possible explanation for the shared genetic alleles could be that in the minority of JIA-U cases, uveitis may first manifest before the onset of arthritis (classified as CAU). A fraction of CAU cases may later develop arthritis and will be eventually classified as JIA-U.50,51 However, in some of them, arthritis development might be suppressed by immunomodulating treatment for uveitis. In contrast to a previous GWAS in JIA-U, here we did not include JIA patients without uveitis and therefore could not formally test whether the identified associations were related to uveitis or JIA. However, several lines of evidence link DQB1*04:02-DRB1*08:1 to pediatric anterior uveitis. First, we showed that these alleles are genetically linked to CAU. Second, the allele frequency of these alleles in a previous GWAS was higher in patients with JIA-U compared to JIA patients without uveitis (0.19 vs. 0.11, respectively).14 Finally, we showed that the allele frequency of HLA-DQB1*04:02 and HLA-DRB1*08:01 was also increased in patients with acute anterior uveitis, in line with previous observations.52 These observations suggests that despite the association of these alleles with JIA in general these alleles may contribute to a common molecular mechanism that promotes anterior uveitis, perhaps also in JIA-uveitis.38,39 Future studies that use additional genotype data to fine-map the MHC will provide sufficient resolution (extended haplotypes including noncoding polymorphisms) to perform bivariate analysis to determine the genetic correlation between these different types of anterior uveitis. 
In addition, such studies should also aim to expand the cohort to represent non-European populations. A limitation of the current study is that cases and controls of the study were near exclusively from Western European ancestry as visualized by UMAP. Note that because UMAP dedicates relatively more visual space on populations with larger sample size, this may exaggerate the variation within cases and controls in the projection. Regardless, it remains to be determined if the here identified risk loci for anterior uveitis are shared with other patient populations. For example, in East Asian populations the prevalence of uveitis in the JIA populations may be different compared to Europe.53 
We identified three new independent HLA-wide associations HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 as being strongly associated with panuveitis. HLA-DRB1*01:02 and HLA-DQB1*05 associations have been previously reported in patients suspected for TINU syndrome by Levinson and co-workers.20 A follow-up study from the same group, however, showed that the HLA-DRB1*0102 and HLA-DRB1*08:01 alleles were associated with patients with isolated acute anterior uveitis, but not in those with isolated biopsy-proven TIN without uveitis.21 Perasaari et al. identified HLA-DRB1*08:01 and HLA-DRB1*14:01 as risk alleles rather than HLA-DRB1*01:02 in patients with biopsy-proven TINU.19 Our data also showed that in patients with acute anterior uveitis the allele frequency of HLA-DRB1*08:01 was higher compared to controls (Supplementary Table 4). Of interest, work by Reddy et al. demonstrated that 14 out of 15 children with unexplained panuveitis also carried the HLA-DRB1*01-HLA-DQB1*05 haplotype.22 In our study, 15 of 81 patients with panuveitis were suspected for TINU syndrome. However, removing these patients (n = 15) from the analysis modestly affected the signal and indicates that the association signal at DRB1*01:02 and DQB1*05:03 might represent a signal beyond the association with TINU. In other words, we speculate that the alleles HLA-DRB1*01 and HLA-DQB1*05 might convey risk for panuveitis and HLA-DRB1*08:01 may convey risk for anterior uveitis. Though, it should be noted that TINU is often asynchronous and therefore TINU might be underestimated. We further note that the diagnosis of some subtypes of uveitis may be prone to unintentional heterogeneity due to the intrinsic dependency to classify pediatric uveitis based on clinical expert opinion. Regardless, the diagnosis of uveitis according to the SUN criteria (as done in this study) may allow comparison between studies. These classification criteria appeared to perform well enough for use in clinical and translational research.54 We detected HLA-DRB1*15:01 as the lead allele in cases with IU, which is in line with previous observations.11,12 Remarkably, this allele is also the primary association in multiple sclerosis, an inflammatory demyelinating disease of the central nervous system and is often accompanied by IU.55 Furthermore, IU patients also show demyelinating lesions by MRI, supporting the close relation of IU and MS.56 In our study, two cases that carried the HLA-DRB1*15:01 allele showed demyelinating lesions by MRI. 
We previously showed that uveitis in JIA is genetically linked to other amino acid positions by comparing to JIA without uveitis rather than an individual HLA allele.14 In the current study, we demonstrated that the disease risk is detected best by the classical HLA alleles instead of specific amino acid positions, which are often shared by multiple HLA alleles. Alternatively, moderate-to-strong LD between HLA alleles and nearby non-HLA loci may suggest that risk HLA alleles serve as a proxy for variants in non-HLA genes implicated in pediatric uveitis. For example, HLA-DQB1:04 is in LD with the TAP1 gene involved in antigen presentation.57 
Further analyses revealed that the association signals in JIA-U (HLA-DQB1*04:02), CAU (HLA-DQB1*05:03), and panuveitis (HLA-DRB1*01:02) showed evidence for sexual dimorphism. We identified male-specific associations for HLA-DRB1*11:01 in JIA-U, which showed a large difference in frequency with females. Conversely, HLA-DRB1*11:04 was only associated with females. Our data supported sex-specific effects based on interaction of sex with the risk for these alleles. Sex-specific HLA allele associations have been reported in other chronic inflammatory conditions.14,58–60 This is in line with functional data on sex differences in the adaptive immune responses and the degree to which HLA molecules interact with T cells (i.e., selection and expansion).61–63 
Other functional implications of our data became evident by functional modeling of allele-defining amino acid positions. Although such studies have been done more extensively for HLA-DR.40,41 Since we previously showed that amino acid positions 10 to 12 encoded by the HLA-DRB1gene are the primary genetic risk factors for uveitis in JIA,14 we focused in our study on HLA DQB1*05:03—especially since it revealed association with multiple clinically distinct types of uveitis. We demonstrated similar direct effects on peptide presentation for HLA-DQ. We showed that the risk allele HLA-DQB1*05:03 was found to be associated with two forms of uveitis CAU and PAN. This allele harbors an aspartic acid (D) at position 57 while the other HLA-DQB1*05 alleles have a serine (S) or valine (V) at that position. We observed that amino acid substitution of position 57 from S or V (in the nonuveitis associated allele HLA-DQB1*05:01 and HLA-DQB1*05:02) to the uveitis-associated amino acid aspartic acid (D) in DQB1*05:03 increased the binding of peptides, underscoring the critical involvement of this position in antigen presentation by HLA-DQ. Alternatively, although not evaluated in this study, this position may also influence the flexibility and possibility to accommodate longer peptides protruding out of the groove. 
In conclusion, we conducted a comprehensive analysis of HLA alleles implicated in uveitis susceptibility in children. We ascertained previous and identified novel independent HLA associations. We further demonstrated that the phenotypically similar CAU and JIA-U share HLA associated alleles. This suggests a common molecular basis for these types of uveitis and might help to understand the molecular mechanism driving pediatric uveitis. 
Acknowledgments
Supported by the Dutch Ophthalmology Foundation “UitZicht” and ODAS Stichting. The sponsor or funding organization had no role in the design or conduct of this research. 
Disclosure: R.A.W. Wennink, None; J.H. de Boer, None; S. Hiddingh, None; A.M.J.W. Haasnoot, None; V. Kalinina Ayuso, None; T. de Hoop, None; J. van Setten, None; E. Spierings, None; J.J.W. Kuiper, None 
References
de Boer J, Wulffraat N, Rothova A. Visual loss in uveitis of childhood. Br J Ophthalmol. 2003; 87(7): 879–884. [CrossRef] [PubMed]
McDonald J, Cassedy A, Altaye M, et al. Comprehensive assessment of quality of life, functioning and mental health in children with juvenile idiopathic arthritis and non-infectious uveitis [published online ahead of print, 2021 Jan 9]. Arthritis Care Res (Hoboken). 2021.
Maccora I, Sen ES, Ramanan AV. Update on noninfectious uveitis in children and its treatment. Curr Opin Rheumatol. 2020; 32(5): 395–402. [CrossRef] [PubMed]
Sen ES, Ramanan AV. Juvenile idiopathic arthritis-associated uveitis. Clin Immunol. 2020; 211: 108322. [CrossRef] [PubMed]
Walscheid K, Neekamp L, Heiligenhaus A, Weinhage T, Heinz C, Foell D. Increased circulating proinflammatory T lymphocytes in children with different forms of anterior uveitis: results from a pilot study. Ocul Immunol Inflamm. 2019; 27(5): 788–797. [CrossRef] [PubMed]
Walscheid K, Neekamp L, Heiligenhaus A, et al. Peripheral blood monocytes reveal an activated phenotype in pediatric uveitis. Clin Immunol. 2018; 190: 84–88. [CrossRef] [PubMed]
Haasnoot AJW, Kuiper JJW, de Boer JH. Predicting uveitis in juvenile idiopathic arthritis: from biomarkers to clinical practice. Expert Rev Clin Immunol. 2019; 15(6): 657–666. [CrossRef] [PubMed]
Kalinina Ayuso V, Makhotkina N, van Tent-Hoeve M, et al. Pathogenesis of juvenile idiopathic arthritis associated uveitis: the known and unknown. Surv Ophthalmol. 2014; 59(5): 517–531. [CrossRef] [PubMed]
Karnes JH, Bastarache L, Shaffer CM, et al. Phenome-wide scanning identifies multiple diseases and disease severity phenotypes associated with HLA variants. Sci Transl Med. 2017; 9(389): eaai8708. [CrossRef] [PubMed]
Shi T, Lv W, Zhang L, Chen J, Chen H. Association of HLA-DR4/HLA-DRB1*04 with Vogt-Koyanagi-Harada disease: a systematic review and meta-analysis. Sci Rep. 2014; 4: 6887. Published Nov 10, 2014. [CrossRef] [PubMed]
Márquez A, Cordero-Coma M, Martín-Villa JM, et al. New insights into the genetic component of non-infectious uveitis through an Immunochip strategy. J Med Genet. 2017; 54(1): 38–46. [CrossRef] [PubMed]
Petrushkin H, Thomas D, Vaughan R, et al. Possession of the HLA-DRB1*1501 allele and visual outcome in idiopathic intermediate uveitis. JAMA Ophthalmol. 2015; 133(4): 482–483. [CrossRef] [PubMed]
Kuiper JJ, Van Setten J, Ripke S, et al. A genome-wide association study identifies a functional ERAP2 haplotype associated with birdshot chorioretinopathy. Hum Mol Genet. 2014; 23(22): 6081–6087. [CrossRef] [PubMed]
Haasnoot AJW, Schilham MW, Kamphuis S, et al. Identification of an amino acid motif in HLA-DRβ1 that distinguishes uveitis in patients with juvenile idiopathic arthritis. Arthritis Rheumatol. 2018; 70(7): 1155–1165. [CrossRef] [PubMed]
Angeles-Han ST, McCracken C, Yeh S, et al. HLA associations in a cohort of children with juvenile idiopathic arthritis with and without uveitis. Invest Ophthalmol Vis Sci. 2015; 56(10): 6043–6048. [CrossRef] [PubMed]
Rock KL, Reits E, Neefjes J. Present yourself! By MHC class I and MHC class II molecules. Trends Immunol. 2016; 37(11): 724–737. [CrossRef] [PubMed]
Mandeville JT, Levinson RD, Holland GN. The tubulointerstitial nephritis and uveitis syndrome. Surv Ophthalmol. 2001; 46(3): 195–208. [CrossRef] [PubMed]
Okafor LO, Hewins P, Murray PI, Denniston AK. Tubulointerstitial nephritis and uveitis (TINU) syndrome: a systematic review of its epidemiology, demographics and risk factors. Orphanet J Rare Dis. 2017; 12(1): 128. Published Jul 14, 2017. [CrossRef] [PubMed]
Peräsaari J, Saarela V, Nikkilä J, et al. HLA associations with tubulointerstitial nephritis with or without uveitis in Finnish pediatric population: a nation-wide study. Tissue Antigens. 2013; 81(6): 435–441. [CrossRef] [PubMed]
Levinson RD, Park MS, Rikkers SM, et al. Strong associations between specific HLA-DQ and HLA-DR alleles and the tubulointerstitial nephritis and uveitis syndrome. Invest Ophthalmol Vis Sci. 2003; 44(2): 653–657. [CrossRef] [PubMed]
Mackensen F, David F, Schwenger V, et al. HLA-DRB1*0102 is associated with TINU syndrome and bilateral, sudden-onset anterior uveitis but not with interstitial nephritis alone. Br J Ophthalmol. 2011; 95(7): 971–975. [CrossRef] [PubMed]
Reddy AK, Hwang YS, Mandelcorn ED, Davis JL. HLA-DR, DQ class II DNA typing in pediatric panuveitis and tubulointerstitial nephritis and uveitis. Am J Ophthalmol. 2014; 157(3): 678–86.e862. [CrossRef] [PubMed]
Jabs DA, Nussenblatt RB, Rosenbaum JT; Standardization of Uveitis Nomenclature (SUN) Working Group. Standardization of uveitis nomenclature for reporting clinical data. Results of the First International Workshop. Am J Ophthalmol. 2005; 140(3): 509–516. [PubMed]
Petty RE, Southwood TR, Manners P, et al. International League of Associations for Rheumatology classification of juvenile idiopathic arthritis: second revision, Edmonton, 2001. J Rheumatol. 2004; 31(2): 390–392. [PubMed]
Wood PHN . Diagnosis criteria, nomenclature, classification. In: Munthe E, ed. The Care of Rheumatic Children. Basel: EULAR Publishers; 1978: 42–50.
Cassidy J, Kivlin J, Lindsley C, Nocton J. Ophthalmologic examinations in children with juvenile rheumatoid arthritis. Pediatrics. 2006; 117(5): 1843–1845. [CrossRef] [PubMed]
Robinson J, Barker DJ, Georgiou X, Cooper MA, Flicek P, Marsh SGE. IPD-IMGT/HLA Database. Nucleic Acids Res. 2020; 48(D1): D948–D955. [PubMed]
Boomsma DI, Wijmenga C, Slagboom EP, et al. The Genome of the Netherlands: design, and project goals. Eur J Hum Genet. 2014; 22(2): 221–227. [CrossRef] [PubMed]
Browning BL, Zhou Y, Browning SR. A one-penny imputed genome from next-generation reference panels. Am J Hum Genet. 2018; 103(3): 338–348. [CrossRef] [PubMed]
Jia X, Han B, Onengut-Gumuscu S, et al. Imputing amino acid polymorphisms in human leukocyte antigens. PLoS One. 2013; 8(6): e64683. Published Jun 6, 2013. [CrossRef] [PubMed]
McInnes L, Healy J, Saul N, Großberger L. UMAP: Uniform Manifold Approximation and Projection. Journal of Open Source Software. 2018; 3(29): 861. [CrossRef]
Gourraud PA, Khankhanian P, Cereb N, et al. HLA diversity in the 1000 genomes dataset. PLoS One. 2014; 9(7): e97282. Published 2014 Jul 2. [CrossRef] [PubMed]
1000 Genomes Project Consortium, Auton A, Brooks LD, et al. A global reference for human genetic variation. Nature. 2015; 526(7571): 68–74. [PubMed]
Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015; 4: 7. Published Feb 25, 2015. [CrossRef] [PubMed]
Marcu A, Bichmann L, Kuchenbecker L, et al. The HLA ligand atlas. A resource of natural HLA ligands presented on benign tissues. bioRxiv 778944.
Reynisson B, Alvarez B, Paul S, Peters B, Nielsen M. NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Res. 2020; 48(W1): W449–W454. [CrossRef] [PubMed]
Nielsen M, Andreatta M. NNAlign: a platform to construct and evaluate artificial neural network models of receptor-ligand interactions. Nucleic Acids Res. 2017; 45(W1): W344–W349. [CrossRef] [PubMed]
Hollenbach JA, Thompson SD, Bugawan TL, et al. . Juvenile idiopathic arthritis and HLA class I and class II interactions and age-at-onset effects. Arthritis Rheum, 2010; 62(6): 1781–1791. [CrossRef] [PubMed]
Hinks A, Cobb J, Marion MC, et al. Dense genotyping of immune-related disease regions identifies 14 new susceptibility loci for juvenile idiopathic arthritis. Nat Genet. 2013; 45(6): 664–669. [CrossRef] [PubMed]
Anderson KM, Roark CL, Portas M, Aubrey MT, Rosloniec EF, Freed BM. A molecular analysis of the shared epitope hypothesis: binding of arthritogenic peptides to DRB1*04 alleles. Arthritis Rheumatol. 2016; 68(7): 1627–1636. [CrossRef] [PubMed]
Verreck FA, Termijtelen A, Koning F. HLA-DR beta chain residue 86 controls DR alpha beta dimer stability. Eur J Immunol. 1993; 23(6): 1346–1350. [CrossRef] [PubMed]
Morel PA, Dorman JS, Todd JA, McDevitt HO, Trucco M. Aspartic acid at position 57 of the HLA-DQ beta chain protects against type I diabetes: a family study [published correction appears in Proc Natl Acad Sci USA 1989 Feb;86(4):1317]. Proc Natl Acad Sci USA. 1988; 85(21): 8111–8115. [CrossRef]
Delgado JC, Baena A, Thim S, Goldfeld AE. Aspartic acid homozygosity at codon 57 of HLA-DQ beta is associated with susceptibility to pulmonary tuberculosis in Cambodia. J Immunol. 2006; 176(2): 1090–1097. [CrossRef] [PubMed]
Ikegami H, Tahara Y, Cha T, et al. Aspartic acid at position 57 of the HLA-DQ beta chain is not protective against insulin-dependent diabetes mellitus in Japanese people. J Autoimmun. 1990; 3(2): 167–174. [CrossRef] [PubMed]
Awata T, Kuzuya T, Matsuda A, Iwamoto Y, Kanazawa Y. Genetic analysis of HLA class II alleles and susceptibility to type 1 (insulin-dependent) diabetes mellitus in Japanese subjects [published correction appears in Diabetologia 1992 Sep;35(9):906]. Diabetologia. 1992; 35(5): 419–424. [CrossRef] [PubMed]
Pettersen EF, Goddard TD, Huang CC, et al. UCSF Chimera–a visualization system for exploratory research and analysis. J Comput Chem. 2004; 25(13): 1605–1612. [CrossRef] [PubMed]
Bondinas GP, Moustakas AK, Papadopoulos GK. The spectrum of HLA-DQ and HLA-DR alleles, 2006: a listing correlating sequence and structure with function. Immunogenetics. 2007; 59(7): 539–553. [CrossRef] [PubMed]
Zamecki KJ, Jabs DA. HLA typing in uveitis: use and misuse. Am J Ophthalmol. 2010; 149(2): 189–193.e2 [CrossRef] [PubMed]
Hurley CK, Kempenich J, Wadsworth K, et al. Common, intermediate and well-documented HLA alleles in world populations: CIWD version 3.0.0. HLA. 2020; 95(6): 516–531. [CrossRef] [PubMed]
Heiligenhaus A, Klotsche J, Niewerth M, et al. Similarities in clinical course and outcome between juvenile idiopathic arthritis (JIA)-associated and ANA-positive idiopathic anterior uveitis: data from a population-based nationwide study in Germany. Arthritis Res Ther. 2020; 22(1): 81. Published 2020 Apr 15. [CrossRef] [PubMed]
Kalinina Ayuso V, Ten Cate HA, van der Does P, Rothova A, de Boer JH. Male gender and poor visual outcome in uveitis associated with juvenile idiopathic arthritis. Am J Ophthalmol. 2010; 149(6): 987–993. [CrossRef] [PubMed]
Pimentel-Santos FM, Matos M, Ligeiro D, et al. HLA alleles and HLA-B27 haplotypes associated with susceptibility and severity of ankylosing spondylitis in a Portuguese population. Tissue Antigens. 2013; 82(6): 374–379. [CrossRef] [PubMed]
Tanya M, Teh KL, Das L, Hoh SF, Gao X, Arkachaisri T. Juvenile idiopathic arthritis in Southeast Asia: the Singapore experience over two decades. Clin Rheumatol. 2020; 39(11): 3455–3464. [CrossRef] [PubMed]
Jabs DA, McCluskey P, Oden N, et al. Development of classification criteria for the uveitides [published online ahead of print, 2021 Apr 10]. Am J Ophthalmol. 2021; S0002-9394(21)00186-0
Hollenbach JA, Oksenberg JR. The immunogenetics of multiple sclerosis: a comprehensive review. J Autoimmun. 2015; 64: 13–25. [CrossRef] [PubMed]
Petrushkin H, Kidd D, Pavesio C. Intermediate uveitis and multiple sclerosis: to scan or not to scan. Br J Ophthalmol. 2015; 99(12): 1591–1593. [CrossRef] [PubMed]
Ivansson EL, Magnusson JJ, Magnusson PK, Erlich HA, Gyllensten UB. MHC loci affecting cervical cancer risk: distinguishing the effects of HLA-DQB1 and non-HLA genes TNF, LTA, TAP1 and TAP2. Genes Immun. 2008; 9(7): 613–623. [CrossRef] [PubMed]
Sayad A, Akbari MT, Pajouhi M, Mostafavi F, Kazemnejad A, Zamani M. Investigation the role of gender on the HLA-DRB1 and -DQB1 association with type 1 diabetes mellitus in Iranian patients. Cell J. 2013; 15(2): 108–115. [PubMed]
Meyer JM, Han J, Singh R, Moxley G. Sex influences on the penetrance of HLA shared-epitope genotypes for rheumatoid arthritis. Am J Hum Genet. 1996; 58(2): 371–383. [PubMed]
Celius EG, Harbo HF, Egeland T, Vartdal F, Vandvik B, Spurkiand A. Sex and age at diagnosis are correlated with the HLA-DR2, DQ6 haplotype in multiple sclerosis. J Neurol Sci. 2000; 178(2): 132–135. [CrossRef] [PubMed]
Fish EN. The X-files in immunity: sex-based differences predispose immune responses. Nat Rev Immunol. 2008; 8(9): 737–744. [CrossRef] [PubMed]
Amadori A, Zamarchi R, De Silvestro G, et al. Genetic control of the CD4/CD8 T-cell ratio in humans. Nat Med. 1995; 1(12): 1279–1283. [CrossRef] [PubMed]
Schneider-Hohendorf T, Görlich D, Savola P, et al. Sex bias in MHC I-associated shaping of the adaptive immune system. Proc Natl Acad Sci USA. 2018; 115(9): 2168–2173. [CrossRef] [PubMed]
Figure 1.
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
View OriginalDownload Slide
 
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
Figure 1.
 
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
Uniform manifold approximation and projection of the HLA genotypes data. Ancestry populations from the 1000 genome dataset32,33 (Supplementary Table 2) and Dutch subjects from the Genome of the Netherlands was used as reference data to analyze ethnicity. The UMAP projection mapped cases and controls predominantly to the European population. A) Visualization of data from uveitis samples (n = 280) used in this study. The different colors represent the five different forms of pediatric uveitis. B) Visualization of data from the control samples (n = 499) from the Genome of the Netherlands. C) Visualization of data from the 1000 genome per population code. D) Visualization of data from the 1000 genome per super population. AAU, acute anterior uveitis; CAU, idiopathic chronic associated uveitis; JIA-U, juvenile idiopathic arthritis associated uveitis; IU, intermediate uveitis; PAN, panuveitis; African (LWK, Luhya in Kenya; YRI, Yoruba in Nigeria; ASW Americans of African Ancestry in Southwest US); European (CEPH, Utah residents with Northern/Western European ancestry; FIN, Finnish in Finland; GBR, British in England and Scotland; TSI, Tuscans in Italy); East Asian (CHB, Han Chinese in Beijing; CHS, Southern Han Chinese; JPT, Japanese in Tokyo, and CHD, Chinese in Metropolitan Denver, CO, USA); and Mixed American (CLM, Colombian in Medellin, Colombia; MXL, Mexican in Los Angeles, California, USA, and PUR, Puerto Rican).
View OriginalDownload Slide
Figure 2.
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
 
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Figure 2.
 
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Association testing for human leukocyte alleles in all uveitis cases (n = 280), JIA-uveitis (n = 92), and uveitis without JIA-uveitis (n = 185) versus controls (n = 499). A) Initial association testing for all uveitis cases revealed a primary association for HLA-DQB1*04:02 (AF = 0.09, OR = 5.27, 95% CI 3.04–9.12; P = 2.98 × 10−9), closely followed by HLA-DRB1*08:01 (AF = 0.08, OR = 5.21 95% CI; 2.94–9.23; P = 1.53 × 10−8). B) Initial association testing for juvenile idiopathic associated uveitis cases (n = 92) showed a primary association for HLA-DRB1*08:01 (AF = 0.16, OR = 11.12, 95% CI 5.76–21.43; P = 6.51 × 10−13) and HLA-DQB1*04:02 (AF = 0.16, OR = 10.11, 95% CI 5.32–19.22; P = 1.69 × 10−12). C) Initial association testing for uveitis cases without juvenile idiopathic associated uveitis (n = 185) revealed a primary association for HLA-DRB1*01:02 (AF = 0.06. OR = 5.58, 95% CI 2.70–11.55; P = 3.60 × 10−6). All indicated HLA alleles are associated at HLA-wide significance (P < 2.79 × 10−4, dashed dark gray line). The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. The light gray dotted line indicates nominal significance (P = 0.05). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
Figure 3.
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
 
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Figure 3.
 
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
Independent HLA allele associations (i.e., conditional testing) in cases with idiopathic chronic anterior uveitis and panuveitis. A) Initial association testing in 50 chronic anterior uveitis cases detected the primary association for HLA-DQB1*04:02 (AF = 0.10, OR = 5.37, 95% CI 2.38–12.09; P = 5.07 × 10−5). B) Adjusting for lead HLA allele (HLA-DQB1*04:02) revealed independent association for HLA-DQB1*05:03 (AF = 0.10, OR 5.86, 95% CI 2.52–13.62; P = 3.92 × 10−5). C) After adjusting for HLA-DQB1*04:02-HLA-DQB1*05:03 no additional independent HLA associations were detected. D) Initial association testing in 81 panuveitis cases mapped the primary association to HLA-DRB1*01:02 (AF = 0.11, OR = 10.18, 95% CI 4.55–22.76; P = 1.58 × 10−8). E) After adjusting for HLA-DRB1*01:02 an independent association for HLA-DRB1*04:03 (AF = 0.07, OR = 29.61, 95% CI 7.96–110.22; P = 4.35 × 10−7) was detected. F) Adjusting for HLA-DRB1*01:02 and HLA-DRB1*04:03 discovered independent association for HLA-DQB1*05:03 (AF = 0.10, OR = 4.51, 95% CI 2.27–8.96; P = 1.72 × 10−5). G) Adjusting for HLA-DRB1*01:02, HLA-DRB1*04:03, and HLA-DQB1*05:03 revealed no additional independent HLA associations. The size and color of the dots represent the allele frequency and effect size (odds ratio), respectively. The gray dots represent the HLA alleles that did not reach significance. HLA-wide significance threshold (P = 2.79 × 10−4) is indicated as a dashed dark gray line in the association plots). AF = allele frequency; OR = odds ratio; CI = confidence interval.
View OriginalDownload Slide
Figure 4.
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36 C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
View OriginalDownload Slide
 
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
Figure 4.
 
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
The effect of amino acid substitutions for position 57 on predicted binding affinity for natural ligands of HLA-DQ*05 alleles. A) View into the peptide-binding groove of a three-dimensional ribbon model for HLA-DQB1 (based on Protein Data Bank entry 4D8P).46 Position 57 that distinguishes DQB1*05 alleles is highlighted in red. B) The effect of amino acid substitution for position 57 on predicted binding affinity for naturally presented ligands eluted from HLA-DQB1*05:01, HLA-DQB1*05:02, and HLA-DQB1*05:03 (n = 157, see Supplementary Table 8). Binding affinity scores were predicted using the full amino acid sequence of HLA-DQB1*05 in NetMHCIIpan 4.0.36 C) A sequence logo representation of the peptide motif of the 157 HLA-DQB1:05 peptides from the HLA atlas as computed by the NNAlign 2.0 server.37 The relative height of color-coded amino acid (letter code) at each position represent the information content in bits. Positive values show favored amino acids at that position and negative bits indicate amino acids that are not preferred at that position. The anchor position P4, P6, and P9 of HLA-DQ are highlighted.
View OriginalDownload Slide
Table 1.
 
View Table
 
Diagnosis of the 280 Uveitis Cases
Table 1.
 
Diagnosis of the 280 Uveitis Cases
  Diagnosis of the 280 Uveitis Cases
Table 2.
 
View Table
 
HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
Table 2.
 
HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
  HLA Associations Before and After Adjusting for the Primary Association in Juvenile Idiopathic Arthritis Associated Uveitis and Nonjuvenile Idiopathic Arthritis Associated Uveitis Versus Controls (n = 499)
Table 3.
 
View Table
 
Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
Table 3.
 
Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
  Primary HLA Associations in Juvenile Idiopathic Arthritis Associated Uveitis, Idiopathic Chronic Anterior Uveitis (No JIA), Intermediate Uveitis, and Panuveitis Versus Controls (n = 499)
Table 4.
 
View Table
 
The Distribution of Males and Females in the Investigated Groups
Table 4.
 
The Distribution of Males and Females in the Investigated Groups
  The Distribution of Males and Females in the Investigated Groups
Supplement 1
i1552-5783-62-9-19_s1_1626089963.92535.xlsx
Copyright 2021 The Authors
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Attribution-NonCommercial-NoDerivatives
Copyright © 2024 Association for Research in Vision and Ophthalmology.
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy.Accept