Association of CLEC16A with human common variable immunodeficiency disorder and role in murine B cells

Common variable immunodeficiency disorder (CVID) is the most common symptomatic primary immunodeficiency in adults, characterized by B cell abnormalities and inadequate antibody response. CVID patients have considerable autoimmune comorbidity and we therefore hypothesized that genetic susceptibility to CVID may overlap with autoimmune disorders. Here, in the largest genetic study performed in CVID to date, we compare 778 CVID cases with 10,999 controls across 123,127 single nucleotide polymorphisms (SNPs) on the Immunochip. We identify the first non-HLA genome-wide significant risk locus at CLEC16A (rs17806056, P=2.0×10−9) and confirm the previously reported human leukocyte antigen (HLA) associations on chromosome 6p21 (rs1049225, P =4.8×10−16). Clec16a knock down (KD) mice showed reduced number of B cells and elevated IgM levels compared to controls, suggesting that CLEC16A may be involved in immune regulatory pathways of relevance to CVID. In conclusion, the CLEC16A associations in CVID represent the first robust evidence of non-HLA associations in this immunodeficiency condition.


Supplementary Figure 3. Quantile-quantile plot of observed versus expected P-values from association statistics of a list of null SNPs which are not immune related.
The Immunochip contains 3,120 SNPs that were part of a bipolar disease replication effort and other studies that were not immune related. After quality control, 2,647 of these single nucleotide polymorphisms (SNPs) were used as null markers to estimate the overall inflation of the distribution of association test statistics. Quantile-quantile plot of P-values in logistic regression analysis for these SNPs demonstrated minimal deviation from the expected distribution of the test statistics (genomic inflation factor λ=1.039). The red line is y = x.

Supplementary Figure 4. The relative expression level of SOCS1 and DEXI compared between CVID cases of different genotypes at SNP rs17806056.
The mRNA level of SOCS1 and DEXI was measured by quantitative real-time polymerase chain reaction (qRT-PCR) and normalized to GAPDH control. The resulting relative fold change (Yaxis) was plotted against the genotype at SNP rs17806056 (X-axis). Each blue dot represents the average value of three measures from each individual sample in the experiment and the black line through the dots represents the mean level among each genotype group. The number of samples in each group is n=3 AA, n=7 TA, n=11 TT.
The regional association plots show the -log 10 P-values of genotyped SNPs in logistic regression analysis plotted against the genomic position of markers. The index SNP is indicated by the purple color while the colors of the remaining SNPs correspond to the strength of linkage disequilibrium with the index SNP (for color coding see legend in the upper right corner of each plot). The recombination rate was derived from the HapMap project 1 and is represented by the light blue line. The plots were generated using the LocusZoom software 2 .
Supplementary Figure 9. Representative western blot for Clec16a in splenocytes of Clec16a knockdown (KD) and controls mice. Whole cell lysates from splenocytes of Clec16a knock down mice and the control (cntr) littermates were separated on 4-12% NuPAGE Bis-Tris gels and transferred onto nitrocellulose membrane. The membrane was probed with rabbit polyclonal CLEC16A antibody. Band intensities were measured using Image J software and normalized to β-Actin loading control. A 76% reduction of Clec16a protein was observed. Granulomas were defined as biopsy-proven unexplained granuloma, (Crohn's disease excluded).
Absolute counts and valid percent given (excluding missing data).
We observed variability in the prevalence of different phenotypes and feature like e.g. age at onset between different centers (Supplementary Table 1). The heterogeneity seen between the different study sites is consistent with observations from previous cross center studies and may be due to difference in awareness, diagnostic protocols, clinical assessment, and geography 4, 5 . Infection only: absence of known granuloma, splenomegaly, organ-specific autoimmunity, autoimmune cytopenias, enteropathy, lymphoid hyperplasia, lymphoma, nodular regenerative hyperplasia of the liver (NRH) and Lymphoid interstitial pneumonitis. Granulomas: were defined as biopsy-proven unexplained granuloma (Crohn's disease excluded). Splenomegaly was defined as 11 cm or greater on ultrasound or on palpation, including previous splenectomy (previous splenomegaly). Organ-specific autoimmunity: includes Graves' disease, primary thyroid failure, and insulin-dependent diabetes mellitus. Autoimmune cytopenias: autoimmune haemolytic anaemia, immune thrombocytopenia, persistent unexplained neutropenia. Enteropathy: was defined as diarrhea after exclusion of an infectious origin. Lymphoid hyperplasia was defined as extensive and persistent lymphadenopathy (on palpation, ultrasound or computed tomography scan. NRH liver: Nodular regenerative hyperplasia of the liver.

b)
We also performed a stratified association analysis of the most strongly associated SNP rs17806056 with four sub-phenotype groups according to Chapel's division into distinct clinical phenotypes: "Infection only", "Autoimmunity", "Polyclonal lymph proliferation", "Enteropathy" 4 . The fifth subphenotype "lymphoma" was not included due to too few cases. Significant association results are highlighted in bold.   SNP,single-nucleotide polymorphism, Chr.,chromosome; A1,minor allele; OR, odds ratio; P-values were obtained from logistic regression analysis. a Most associated SNP in each gene. b SNP rs12622799 is 30kb 5' of ICOS.

Supplementary Table 8. Stepwise conditional analysis of the HLA association.
The most significant HLA-SNP after each step of conditioning is listed in the second column.
Steps SNP, single-nucleotide polymorphism; Chr, chromosome; A1, minor allele; OR, odds ratio; CI, confidence interval; r 2 , LD measurement between the top SNP at each step and rs1049225; P-values were obtained from logistic regression conditioned on the top SNP of each previous step. a . The most associated SNP at each step of conditional association analysis. are highlighted in red and green, respectively.

b) HLA-C allele frequencies in patients with Common variable immunodeficiency disorder
(CVID) (n=886) and healthy controls (n=11,139). Alleles with a frequency ≤ 3 in the CVID population are grouped under "other". Risk alleles and protective alleles with a P-value < 5×10 -3 are highlighted in red and green, respectively.  healthy controls (n=11,139). Alleles with a frequency ≤ 3 in the CVID population are grouped under "other". Risk alleles and protective alleles with a P-value < 5×10 -3 are highlighted in red and green, respectively. CI: Confidence interval; N.A: Not applicable; Odds ratio with confidence interval and P-value were calculated directly from the 2×2 SNP.,single nucleotide polymorphisms; Chr., Chromosome; AID, associated autoimmune disease; DOE, minor allele direction of effect with major allele as the reference; MAF, minor allele frequency; OR, odds ratio; 95% CI, 95% confidence interval; LD, Linkage Disequilibrium. r 2 measurement with rs17806056; P-values shown were from logistic regressions of our genome-wide association analysis.