The extended phenotype of LRBA deficiency
Laura Gámez-Díaz, MSc1, Dietrich August, cand. MD1, PolinaStepensky, MD2, Shoshana Revel-Vilk, MD, MSc2, Markus G. Seidel, MD3, Mitsuiki Noriko, MD4, Tomohiro Morio, MD, PhD4, Austen JJ Worth, MD, PhD5, Jacob Blessing, MD, PhD6, Frank Van de Veerdonk, MD, PhD7, Tobias Feuchtinger, MD8, Maria Kanariou, MD, PhD9, Annette Schmitt-Graeff, MD10, SuranjithSeneviratne, MD, PhD11, Siobhan Burns, MD11, Bernd H Belohradsky, MD12, NimaRezaei, MD, PhD13, ShahrzadBakhtiar, MD14, Carsten Speckmann, MD1,15, Michael Jordan, MD6 and Bodo Grimbacher, MD1,11, #
1Center for Chronic Immunodeficiency, University Medical Center Freiburg, Freiburg, Germany
2Pediatric Hematology-Oncology and Bone Marrow Transplantation, Hadassah Hebrew University Hospital, Israel
3Department of Pediatrics and Adolescent Medicine, Division of Pediatric Hematology-Oncology, Medical University Graz, Austria
4Department of Pediatrics and Developmental Biology Graduate School of Medical and Dental Sciences Tokyo Medical and Dental University, Tokyo, Japan
5Department of Immunology, Great Ormond Street Hospital for Children, London, UK
6Cincinnati Children's Hospital Medical Center, University of Cincinnati Medical School, Cincinnati Ohio, USA
7Department of Internal Medicine, Radboud University Nijmegen Medical Centre, Nijmegen, The Netherlands
8Pediatric Hematology, Oncology and Stem Cell Transplantation, Dr. von Hauner University Children´s Hospital, Ludwig-Maximilians-University, Munich, Germany
9Department of Immunology, “Aghia Sophia” Children´s Hospital, Athens, Greece
10Department of Pathology, University Hospital Freiburg, Freiburg, Germany
11UCL Centre for Immunodeficiency, Royal Free Hospital Foundation Trust, London, UK
12Division of Immunology and Infectious Disease, University Childrens Hospital Munich, Munich, Germany
13Research Center for Immunodeficiencies, Children’s Medical Center; Department of Immunology, School of Medicine Tehran University of Medical Sciences, Tehran, Iran
14Division for Stem Cell Transplantation and Immunology, Department for Children and Adolescents Medicine, University Hospital Frankfurt, Goethe University Frankfurt am Main, Frankfurt, Germany
15Center for Pediatrics and Adolescent Medicine, University Medical Center Freiburg, Freiburg, Germany
#Corresponding author: Prof. Dr. Bodo Grimbacher. Address of correspondence: Engesserstraße 4, second floor. Post code: 79108, Freiburg, Germany. Phone: +49 76127077732
ONLINE REPOSITORY
Material and Methods
Study design
Cross-sectional study to describe the clinical, laboratory and genetic characteristics of patients with a diagnosis of LRBA deficiency.
Study Setting and Population
From January 2012 to May 2015, we included all patients referred to us who were suspected of having LRBA deficiency. According to our experience and to previous published reports on LRBA deficiency (2, 3,5), our inclusion criteria were: I. Unknown genetic cause of their disease and any of the following diagnosis: IIa. ALPS-phenotype (Autoimmune lymphoproliferative syndrome of undetermined genetic cause), IIb. CVID (Common Variable Immunodeficiency), IIc. Child-onset hypogammaglobulinemia,IId: early-onset Inflammatory Bowel Disease, i.e. age of onset <10 years (10), and IIe: Autoimmune cytopenias of unknown etiology.III. Signed patient consent. IV. Availability of samples.
Study Protocol
All patients referred to us who met the clinical inclusion criteria were classified as “possible” LRBA deficiency. The protocol for each possible LRBA-deficient patient included: Collection of i) baseline data and ii) blood sample for LRBA protein testingby Western blotting or flow cytometry. In case of an absence of LRBA protein, patients were classified as “probable” LRBA deficiency, and LRBAwas sequenced. Patients with clearly reduced LRBA protein expression received targeted re-sequencing of 77 CVID-associated candidate genes. Single nucleotide variants which were called more than 10-times, were filtered by removal of non-functional intronic and synonymous mutations. Single heterozygous mutations with a MAF of >1% as reported in the dbSNP138 and 1000 genome databases were also removed. Remaining homozygous or compound heterozygous coding or splicing variants which were predicted as pathogenic by either Mutation tester, Polyphen, or Swift software were considered as deleterious mutations and further validated in the patients’ parents.Patients with biallelic mutationsin LRBA were classified as “definitive” LRBA deficiency. Clinical and laboratory findings were entered into a standardized case report form. The local scientific Ethics committees approved this research on the basis of written informed consent to every participating Institution (N° 499/11 at the Medical University of Graz, Austria, N° 0306-10-HMO at theHadassah Medical Center of Jerusalem, Israel, N° 290/13 and N°40/08 at the University Medical Center of Freiburg, Germany, N°92 and N° 103 at the Tokyo Medical and Dental University Hospital, Japan, N° 04/Q0501/119 at the Royal Free Hospital, London, UK, N°2013/194 at the Radboud University Nijmegen Medical Centre in The Netherlands, N°89-04-80-11945 at the Tehran University of Medical Sciences in Iran, and the samples collection protocol at the Cincinnati Children´s Hospital in Cincinnati, USA, was approved by the CCHMC-IRB).
Determination of LRBA protein expression
Isolated PBMCs were stimulated with either 10ng/µl of PHA (cat. L1668, SIGMA) for 72 hours or 20ng/µl of PMA (cat.P1585, SIGMA) and ionomycin (cat.IO634, SIGMA) for 16 hours, or with anti-CD3(1μg/ml)/antiCD28 (5μg/ml) along with 100 units of IL-2 for 72 hours.
Western Blotting
Protein lysates were size-fractionated by SDS-PAGE (12%, 8% and 4% gradient gel), electrotransferred to a PVDF membrane for 2 hours at 45V and immunodetected by rabbit polyclonal anti-LRBA antibody (cat. HPA023597, Sigma). The specific binding of the antibody to the protein band was detected with a secondary HRP-rabbit IgG (cat. 7074, Cell signaling). Tubulin (cat. 4074, abcam), β-actin (cat. 4970, cell signaling), GAPDH (cat. 2118, cell signaling), and mTOR (cat. 2972, cell signaling)were used as protein loading control, and were detected as a band of ~50KDa, 48KDa, 39KDa and 210KDa, respectively.
Flow cytometry
Cellular expression of LRBA in unstimulated and stimulated PBMCs was determined by flow cytometry. Briefly, 300.000 cells were permeabilized and fixed using BD Cytofix/Cytoperm solution (cat.554715, BD), and then stained with rabbit polyclonal anti-LRBA antibody (cat.HPA023597, Sigma). Subsequently, a PE- secondary antibody against the LRBA antibody (PE (ab´)2 Donkey anti-rabbit IgG, (cat.558416, BD) was added. Cells were then washed and analyzed on a FACS Canto II. Data analysis was performed using the software FlowJo version 10 (TreeStarInc, Ashland, OR, USA) as follows, cells were first gated for singlets (FSC-H vs. FSC-A) and lymphocytes (SSC-A vs. FSC-A). Living lymphocytes were gated based on the negative staining for the fixable viability dye (cat. 65-0863-14, eBioscience), and analyzed for LRBA expressionas an univariate histogram. Mean fluoresce intensities (MFI) were calculated and compare with healthy non-related controls (Supplementary Figure S3, panel B). Activation of stimulated-cells was confirmed by the expression of CD69.
LRBA Mutation Detection
The identification of the genetic defect was performed on genomic DNA extracted from peripheral blood leukocytes from patients lacking or with reduced LRBA protein. In addition, seven patients were diagnosed directly by whole exome sequencing or targeted next generation re-sequencing without prior protein expression testing (patient codes: 105-3, 553-2, 656, 657-1, 657-2 and 773-1).
Next Generation Sequencing (NGS)
Briefly, 225ng of gDNA from “probable” LRBA-deficient patients was digested and hybridized with a HaloPlex biotinylated probe library in presence of an indexing primer cassette for enrichment. After capturing and ligating the circularized target DNA-probe hybrids with streptavidin beads, amplification of targeted fragments was done by PCR. Sample barcodes were introduced during amplification for precise tracking. After elution, PCR-amplification and pooling ofequimolar amounts of indexed targeted-samples were prepared for multiplexed sequencing on the Illumina MiSeq platform. Having run samples of different quantity and quality, we averaged 99% of the bases in the exons covered with a depth of at least 38-fold. The sequences were aligned to the human genome using the Agilent SureCall.
Sanger Sequencing
All variants detected by NGS or WES were validated using Sanger sequencing. LRBA exons were amplified by PCR from gDNA according to standard protocols. Chromatograms are shown in supplementary material Figure S4. Primer sequences and PCR conditions are available upon request. The PCR products were sequenced in both directions by Sanger sequencing and analyzed with Sequencher version 4.10.1.
Whole exome sequencing (WES)
About 1 to 3 µg of genomic DNA was fragmented to about 200bp using ultrasonication. Then, specific adapters (for sequencing on the Illumina HiSeq) were ligated to the fragments and amplified by PCR to generate the genomic library. DNA baits complementary to the target sequence and containing biotin-labeled uridine were hybridized with the genomic library. A physical pull-down was then performed using streptavidin-coated magnetic beads. After elution and PCR amplification the fragments were then sequenced using the Illumina HiSeq 2000 instrument. The mean depth of coverage varied between 86X to 108X, as well, as the quality threshold from 97.6% to 99.1%.