Loss-of-function mutations in the IL-21 receptor gene cause a primary immunodeficiency syndrome

Abstract

Primary immunodeficiencies (PIDs) represent exquisite models for studying mechanisms of human host defense. In this study, we report on two unrelated kindreds, with two patients each, who had cryptosporidial infections associated with chronic cholangitis and liver disease. Using exome and candidate gene sequencing, we identified two distinct homozygous loss-of-function mutations in the interleukin-21 receptor gene (IL21R; c.G602T, p.Arg201Leu and c.240_245delCTGCCA, p.C81_H82del). The IL-21R(Arg201Leu) mutation causes aberrant trafficking of the IL-21R to the plasma membrane, abrogates IL-21 ligand binding, and leads to defective phosphorylation of signal transducer and activator of transcription 1 (STAT1), STAT3, and STAT5. We observed impaired IL-21-induced proliferation and immunoglobulin class-switching in B cells, cytokine production in T cells, and NK cell cytotoxicity. Our study indicates that human IL-21R deficiency causes an immunodeficiency and highlights the need for early diagnosis and allogeneic hematopoietic stem cell transplantation in affected children.

Figure 1.

Clinical and immunological phenotype, identification of IL-21R deficiency, and protein structure analysis in family A. (A) Pedigree of family A. All affected children died secondary to infections and/or therapy-associated complications before the identification of the molecular genetic defect. (B) Abdominal magnetic resonance image revealing hepatomegaly and dilatation of intra- and extrahepatic bile ducts (*) in P2. (C) Endoscopic retrograde cholangiopancreatography demonstrated marked dilatation and irregularities of bile ducts in P2. (D) Histopathological analysis of liver biopsy from P2 showed that fibrous expansion of portal tracts with formation of septa accompanied by prominent ductular proliferation in the portal/lobular interphase was evident. (E) Duodenal biopsy from P2 showed numerous microorganisms ∼2 µm in size and confined to the luminal surface of enterocytes, consistent with cryptosporidiosis. (F) FACS analysis shows accumulation of CD19⁺IgD^highCD27⁻IgG⁻ or IgD^highIgM^high naive B cells in PBMCs isolated from P2 in comparison to a healthy donor. Plots are representative of 3 independent experiments. (G) Scheme of filtering approach for variant calls (VC, single-nucleotide variants and indels [insertion and deletions] <20 bp) from high-throughput sequencing of family members A.II-1, A.II-5, and A.II-6. All variants were filtered according to an autosomal recessive model of inheritance in a consanguineous family. Restricting the candidate set to rare, homozygous, nonsynonymous (NS), or splice site–affecting (SS) variants in the 38 MB CCDS exome resulted in identification of two potentially disease-causing variants: RAB3IL1 and IL21R. (H) DNA Sanger sequencing of the RAB3IL1 and IL21R genes confirmed segregation of the IL-21R^Arg201Leu mutation with the disease phenotype in family A. (I) Structural analysis of wild-type (Arg201, R201; top) and mutant (Arg201Leu, L201; bottom) IL-21R based on the recently published Protein Data Bank structure 3TGX, illustrating IL-21 complexed with the extracellular domain of the IL-21R (Hamming et al., 2012). Hamming et al. number the amino acids in IL-21R with respect to the first amino acid in chain A of the structure 3TGX, while we number the amino acids in IL-21R in respect to the start of translation. Accordingly, the mutated Arg201 in P1 and P2 refers to Arg182 in (3TGX, Hamming et al., 2012). (top) The neighborhood of wild-type Arg201 displaying putative hydrogen bonds with a sugar chain and the Glu157 reside (green dashed lines). (bottom) The same region in the mutated IL-21R structure, illustrating that the Arg201Leu substitution is predicted to break these putative hydrogen bonds. Steric clashes are shown in purple.

Figure 2.

Defective IL-21R subcellular distribution, N-linked glycosylation, and signal transduction in family A. (A) High-resolution APD imaging in HeLa cells expressing wild-type (wt, Arg201) or mutant (mut, Arg201Leu) C-terminal, eGFP-fused, IL-21R demonstrated distinct subcellular distribution. In contrast to the IL-21R^wt-eGFP, which was expressed at the plasma membrane (PM; arrowheads), the IL-21R^mut-eGFP protein was dispersed in the endoplasmic reticulum (ER; arrow heads) and did not traffic to the plasma membrane. Nuclei were stained using DRAQ5. (B) High-resolution APD imaging in HeLa cells expressing wild-type (wt, Arg201) or mutant (mut, Arg201Leu) IL-21R-eGFP illustrates incomplete colocalization of IL-21R^Arg201Leu with transiently expressed RFP-tagged JAK3. (C) FACS analysis gated on intact HeLa cells that were lentivirally engineered to overexpress either γ_c/IL-21R^wt-eGFP (Arg201) or γ_c/IL-21R^mut-eGFP (Arg201Leu). The top image reveals that IL-21-Atto647N ligand binding correlates with the magnitude of eGFP expression in HeLa cells expressing γ_c/IL-21R^wt-eGFP (Arg201), but not in γ_c/IL-21R^mut-eGFP (Arg201Leu)–transduced HeLa cells. The bottom image demonstrates that IL-21R surface staining correlates with the eGFP expression in γ_c/IL-21R^wt-eGFP (Arg201)–transduced HeLa cells, whereas HeLa cells expressing γ_c/IL-21R^mut-eGFP (Arg201Leu) showed aberrant IL-21R surface expression. (D) Western blot analysis of IL-21R expression in EBV-BCL from P2 in comparison to healthy donors. Protein lysates were treated with PNGaseF (top; 60 min, 30 IU/µl) or Endo H (bottom; 60 min, 100 IU/µl) before electrophoresis. Data shown are representative of three independent experiments. D, denaturation buffer; R, reaction buffer. (E) Western blot analysis of STAT signaling in EBV-BCL from P2 and healthy donor (HD) upon stimulation with indicated γ_c-related cytokines: IL-2 (100 ng/ml), IL-4 (100 ng/ml), IL-7 (100 ng/ml), IL-15 (10 ng/ml), and IL-21 (10 ng/ml). Please note that for detection of STAT3 phosphorylation (Tyr705, Y705) upon stimulation with IL-2, IL-4, IL-7, and IL-15, a longer exposure was used compared with stimulation with IL-21. Data are representative of two independent experiments, and defective STAT phosphorylation in P2 upon stimulation with IL-21 was confirmed in three additional experiments. (F) Western blot analysis of STAT3 phosphorylation (Tyr705, Y705) in IL-21–stimulated HeLa cells that were lentivirally transduced with γ_c along with either IL-21Rwt or IL-21Rmut. Data are representative of three independent experiments. (G) Lentiviral gene transfer of wild-type IL-21R in patient’s fibroblasts induces IL-21–mediated STAT3 signaling. Western blot analysis of STAT3 phosphorylation (Tyr705, Y705) in IL-21–stimulated dermal fibroblasts that were lentivirally transduced with wild-type or mutated IL-21R (Arg201Leu). Data depicted are representative of four independent experiments. GAPDH or actin were used as loading controls in D–G.

Figure 3.

Functional consequences of aberrant IL-21R signaling in B and T cells from P2 of family A. (A) Representative FACS analysis showing proliferation of IL-21R–deficient naive B cells that were co-cultured with 3T3 CD40L-expressing cells and stimulated with anti-IgM beads (50 ng/ml), CpG (50 nM), and rh IL-2 (20 U/ml) in the presence or absence of rh IL-21 (30 ng/ml) for 6 d. Histograms illustrate proliferation of CFSE-labeled B cells isolated from healthy donors (HD) and patient (P2, A.II-5). Gray shaded, unstimulated B cells; gray line, CD40L-stimulated B cells; red line, CD40L-/IL-21–stimulated B cells. (B) FACS analysis of immunoglobulin class-switch in naive IL-21R–deficient B cells upon stimulation with rh IL-21 for 6 d. (C) Expression of AICDA and PRDM1 mRNA was determined in CD40L-/IL-21–stimulated B cells after 6 d using Real-time PCR analysis. Values represent the expression of respective genes in relation to RPS9 (ribosomal protein S9). (D) Analysis of NK cell cytotoxicity and antibody-dependent cellular cytotoxicity (ADCC) using ⁵¹Cr-release assay. (E) T cell proliferative response to stimulation with antigens was determined by [³H]-thymidine incorporation assay. PBMCs isolated from healthy donors (n = 4), father, and P2 were stimulated with C. albicans (Cand), DTT, and TT. Depicted data represent average of duplicates and are representative of two independent experiments. (F) Secretion of multiple cytokines in IL-21R–deficient T cells measured by protein array. Data shown are compiled from two independent experiments. The bars render logarithms (to the base of 2) of fold changes between unstimulated cells and cells upon stimulation with anti-CD3/anti-CD28 for 18 h. For each bar, the 95% confidence interval is given. Asterisks indicate cytokines for which the 95% (*), 99% (**), and 99.9% (***) confidence intervals do not overlap.

Figure 4.

Clinical and immunological phenotype and identification of IL-21R deficiency in family B. (A) Pedigree of family B. (B) Chest computed tomography (CT) showing extensive bronchiectasis. P. aeruginosa was cultured from bronchoalveolar lavage after this CT. (C) Magnetic resonance imaging/magnetic resonance cholangiopancreatography of the liver showing extensive intra and extrahepatic biliary duct dilatation. (D) Modified acid fast staining of bronchoalveolar lavage showing abundant neutrophils and Cryptosporidium. (E) FACS analysis of STAT phosphorylation (p-STAT1 Y701, p-STAT3 Y705, and p-STAT5 Y694) in PBMCs isolated from healthy donors (HD), mother (B.I-1), and P3 upon stimulation with rh IL-21. Black line, HD, rh IL-21; dark gray line, B.I-1, rh IL-21; light gray shaded; P3, unstimulated; red line; P3, rh IL-21. (F) DNA Sanger sequencing revealed a homozygous deletion in the IL21R gene (c.240_245delCTGCCA, p.C81_H82del) in P3 and P4. In contrast, the parents showed a heterozygous genotype. (G) FACS analysis of B cell proliferation and class-switch in P3 and P4 upon stimulation with rh IL-21.