Final Report Summary - EURO-PADNET (The Pathophysiology and Natural Course of Patients with Primary Antibody Deficiencies (PAD))
Executive Summary:
EURO PADnet has set out to better understand primary (inborn) deficiencies of antibody production (Primary Antibody Deficiencies; PAD). We documented the exact phenotype of more than 4000 patients into a European web-based database and offered patients participation in approved research protocols. Genetic causes for PADs were sought and more than four novel monogenetic defects have been identified. Moreover, genetic risk factors for becoming antibody-deficient have been identified and better described. We set up in vitro experiments and murine models of the diseases to better understand their pathophysiology. First successful steps towards the genetic correction of the first antibody deficiency called Bruton's disease - which actually was the first primary immune deficiency to be described back in 1952 - have been conducted. Not only the role of immunoglobulin G (IgG), but also the role of IgA and IgM and their memory cells has been studied and their importance in the host defence has been highlighted. The consortium has initiated and accompanied investigator initiated observational trials and the results of these have impacted local, national and international policy making for the management and the understanding of antibody deficiencies (for examples please see www.chest-ct-group.eu or www.springerlink.com/content/l4065729w12r1114/).
During the first three years after the start of this collaborative EU project, we have published in Journals such as The Lancet and Nature Immunology and a total of more than 356 IMPACT FACOR points have been collected.
Most importantly, the consortium runs three websites and has progressed two diagnostic tests into the clinical setting: The project website www.europadnet.eu contains general information about the project and can also be accessed by the partners via a secure login to allow them to exchange information on a confidential level. We also have implemented a teaching website www.ipidnet.org for linking the clinical and cellular phenotyping, and a clinical support website www.chest-ct-group.eu for the analysis of lung complications in antibody deficiencies. Moreover, we have held educational symposia and a Winter School for physicians and researchers in the field.
Within the EURO PADnet consortium, The Binding Site (TBS; www.bindingsite.com) has facilitated the development of both Pneumococcus-specific polysaccaride (PCP) IgM and IgA ELISAs. The assays have been validated and the IgM and IgA responses to Pneumovax have been determined in normal healthy subjects. The development of both PCP IgM and IgA ELISAs has involved the identification, location and preparation of suitable materials. The optimisations of assay conditions and clinical validation have been achieved. Both assays have been initiated as new products and entered into the New Product Development system at The Binding Site (TBS). This process involves full product specification and commercial justification which was established through the EUROPADnet group.
Patients with antibody deficiency often require IgG antibody replacement made from human plasma from healthy donors. This product contains traces of IgA. Some patients with antibody deficiency unfortunately have auto-antibodies against IgA. These often lead to the intolerance of the IgG replacement product and may end up in an anaphylactic reaction. To screen for these in order to assess the risk for possible adverse side effects, an anti-IgA ELISA has been established by EURO PADnet collaborators which is now marketed by www.biovendor.com
Taken together, the EURO PADnet consortium progressed the field of primary antibody deficiency research in Europe in an unparalleled way in just 3 years.
Project Context and Objectives:
This medium-sized collaborative project will document and monitor the natural course and study the pathophysiology of Primary Antibody Deficiencies (PAD). PADs are rare inborn errors of the immune system with an estimated incidence of <1:25.000. The defective immunity in patients with PAD causes an increased susceptibility to recurrent infections of the respiratory- and gastro-intestinal tract as well as ill defined co-morbidity including granulomatous disease, lymphocytic organ infiltration and (paradoxically) autoimmunity.
Our consortium cares for more than 1000 PAD patients, representing approx. 50% of the patients registered in Europe, many of them being children. This project will:
1. Combine clinical and research data in a central online registry, complemented by a sample repository (WP1)
2. Elucidate the genetic cause of PADs by linkage analysis and candidate gene approaches (WP2)
3. Establish in vitro models for B cell differentiation steps that are defective in PADs such as in vitro class switch recombination, and siRNA knockdown models for PAD screening (WP3)
4. Create mouse models of PAD by using several technology platforms including knock-in and knock-out mice, shRNA mediated gene knock-down and humanised mice (WP4)
5. Perform translational research by taking the observations from the patients into the basic research projects and transfer these results back to the patients (WP5).
The study of these immunodeficiencies represent an "experiment of nature", uniquely enabling researchers to study the detrimental effects of mutations in specific genes involved in the immune system. Basic research results are translated into the development of new technologies, new knowledge, and new therapeutic tools. The clinical results will improve the diagnosis, management and quality of life of PAD patients, leading to future developments of new tests, preventive measures and treatment protocols.
Introduction to WP1: The Natural Course and Phenotype of PAD
In the last framework programme, the coordinator of this FP7 project, Prof. Grimbacher headed the development and the establishment of a clinical online patient registry. After 4 years already more than 9.000 patients with primary immunodeficiency diseases were registered into the ESID online database, (59%) of which belonging to the group of PAD (www.esid.org/esid_registry.php). This project will be taken further to create a platform for the exchange of clinical information and research data complemented by quality of life data from the actual patients. To this end, the database shall be complemented by a sample bank.
Projects 1-3: Clinical Registry, Research Databases and Repositories
The ESID Online Registry (www.esid.org/esid_registry.php) is capable of registering clinical and laboratory data on all 206 primary immunodeficiency diseases currently listed in the IUIS publication (1). The particular strength of this database is that follow-up data is being collected on a yearly basis. The data currently being collected however only comprises a common dataset of 22 fields which are considered to give the basic information needed for assessment of an individual patient. In this WP we will extend the datasets of each PAD to facilitate future observational trials.
Project 4: Multiple breath washout for refining lung function testing.
Lung disease is one of the most frequent complications and causes of death of the patients with a primary antibody deficiency syndrome (PAD). Longitudinal data suggest that lung disease starts during childhood and leads to progressive structural lung changes as demonstrated by computed tomography (CT) of the chest. The goals of project 4 are i) to assess lung pathology of PID patients with spirometry and computed chest tomography and ii) to determine the sensitivity and specificity of Multiple Breath Washout (MBW) technique as an alternative for detection of early lung disease in paediatric PAD patients.
Introduction to WP2: The Genetic Etiology of PADs
Positional cloning of disease causing genes has proven to be very successful in the identification of many defects in primary immunodeficiency diseases. This consortium has collected more than 150 families with PAD including informative consanguineous families, and has recently published four candidate loci linkage papers identifying genetic susceptibility loci on chromosomes 4q, 5p, 6p, and 16q, However, the disease causing genes still remain elusive. Moreover, the consortium has recently collected 300 patients with sporadic CVID and 300 patients with selective IgAD, in whom we aim to search for the aetiology of these diseases by molecular/genetic analysis in order to define the underlying molecular/genetic defect(s).
Project 1: Positional Cloning and Linkage studies
Common Variable Immunodeficiency (CVID) is a primary antibody immunodeficiency characterized by low immunoglobulin levels and low or normal B cell numbers. The genetic cause of this disease is unknown for the majority of cases. CVID may occur in an autosomal dominant (AD-CVID) trait or sporadically, but autosomal recessive (AR-CVID) families are also reported.
Project 2: Genetic association studies
The future direction of this project will be to recruit additional patients to run GWS on a sufficiently large replication cohort, aiming to identify both MHC and non-MHC susceptibility loci in patients with IgAD. Furthermore, we will try to identify multicase CVID families for mapping of susceptibility genes.
Project 3: The Candidate Gene Approach
Functional candidate genes for antibody deficiencies will be sought and evaluated on genetics and functional grounds.
Introduction to WP3: In Vitro Models of PAD
The aim of this WP is to develop new in vitro models for PAD study. Each partner has his own approach but all are complementary and converge to the same objective, i.e. the molecular characterization of PAD. By a close collaboration between all groups, 3 main objectives will be reached.
A.Better characterization of PAD, including phenotypic and functional B cell studies.
B.Validation of gene mutations
C.In vitro search for new genetic defects
Project 1: Human TACI deficiency - a T cell independent CSR defect leading to CVID?
TACI C104R accounts for about 50 % of all TACI-mutated alleles in CVID patients. One of the aims of U. Salzer's group in this study is to use existing monoclonal or polyclonal antibodies or generate these reagents to analyse the mutated C104R protein.
Project 2: Class switch recombination Defects.
Study of normal antibody maturation processes in normal and pathological conditions.
Immunoglobulin class switch recombination (CSR)-deficiencies are a rare (1/200,000 births) inherited disease characterized by increased (or normal) IgM serum levels contrasting with a strong decrease or an absence of other isotypes (IgG, IgA). According to the molecular defect, this Ig-CSR deficiency is associated or not with a defective generation of somatic hypermutations (SHM). Several molecular defects underlie this condition. We have previously contributed to the description of CD40-ligand and CD40-deficiencies which both lead to a severe cellular and humoral immunodeficiency.
Project 3: Plasma cell defects in PAD
Plasma cell development and potential defects in functions required to generate plasma cells were studied in Freiburg and in Brescia through the collaboration between the groups of A. Plebani and H. Eibel.
Project 4: CD21low B cell populations in primary human antibody deficiencies
The expansion of specific B cell subsets expressing low levels of CD21 in patients with PAD is an expression of an underlying pathogenetic dysregulation. These B cell subsets will be used to define subgroups of patients for specific functional as well as molecular characterization. Circulating B cell populations with a low expression of CD21 contain transitional B cells, activated CD21low B cells and possibly plasmablasts.
Project 5: shRNA-based knockdown of candidate genes
In order to identify transcripts affected by Btk deficiency, E. Smith's group has performed several experiments. First, they have created a new database for immune-related genes as an important tool for screening targets for shRNA-based knockdown. This database contains 1982 immune genes and combines information from 3 already existing databases: Immunogenetic Related Information Source (IRIS) - a database surveying known human immune system genes, Immunome Knowledge Base (IKB) - an integrated service for immunome research, and Gene Ontology assigned 'immune response' genes.
Introduction to WP4: Mouse Models of PAD
Animal studies make it possible to analyze several aspects of PAD, which cannot be addressed in humans. Although in vitro systems have undisputable advantages (e.g. costs, no use of animal experiments, high through-put) and go some way addressing the mechanisms of disease pathogenesis, in vitro systems lack the complexities of an in vivo system. We will generate and study a variety of murine models that will allow a greater understanding of human antibody deficiency syndromes.
The following examples will directly address key aspects of PAD in animal models:
1.When the molecular basis is not understood, such as in patients with heterozygous TACI mutations (CVID) or XLP, the generation of knock-in and knockout models will provide new insights.
2. When the defect is known, but when the precise mechanism is not understood, the generation of cell-specific in vivo knockdowns, such as Btk-knockdown in B cells will elucidate cell specific functions of a given gene product.
3. When experimental in vitro data suggest novel genetic disease mechanisms, the generation of the corresponding knock-in mouse models (membrane IgG mutations and docking molecules) will test these hypotheses.
4. Developing "humanised mice" will provide us with a model system allowing us to test the capabilities of the transferred hematopoietic cells in any given patient (e.g. for patients with CVID).
The projects are based upon the study of the following diseases: CVID, XLA, AR agamma, XLP, PAD of unknown cause The animal studies will be based upon the following technology platforms: Knockout, knock-in, shRNA (lentiviral), dominant negative (lentivir), humanized mouse
Project 1: Knock-in model of TACI deficiency in CVID
The aim of this work package was to generate a murine model of the C104R mutation and to determine whether this resulted in a murine humoral deficiency. After the successful generation of the respective mouse mutant C76R, we were able to undertake some phenotypic analysis of the mutant. In summary, our data suggest that the C76R homozygous mutant mice do show humoral abnormalities that are similar to human TACI null mutants with a hyper-expansive B cell phenotype and defects in immunoglobulin production. To date no major differences have been observed in heterozygous mutants. Further analysis including immunisation experiments will be undertaken in the latter half of the project.
Project 2: Knockout model of SAP deficiency in X-linked lymphoproliferative disease (XLP)
XLP is a complex immunodeficiency in which mutations in the gene encoding the adaptor protein SAP (SLAM associated protein) lead to severe immunodysregulatory phenomena [1]. Boys with XLP present typically but not always after EBV infection with fulminant infectious mononucleosis, B cell lymphoma or hypogammaglobulinaemia which is highly reminiscent of a CVID phenotype. The pathogenesis of hypogammaglobulinaemia in XLP is unclear and has been until very recently been thought to arise due to a lack of T cell help.
Project 3: Knock-in model for antigen receptor signalling
Establishment of IgG-switched memory B cells and their costimulation-independent activation during secondary immune responses is thought to be dependent on isotype-specific antigen receptor signals. Since the start of the project we have made significant progress in the identification of intracellular signalling proteins that are recruited to tyrosine-phosphorylated cytoplasmic domains of mIgG and mIgE. The first interaction partner that was detected by co-immunoprecipitation and confirmed by Western blot, the adaptor molecule Grb2, could already be published in Nature Immunolgy in September 2009 (Recruitment of the cytoplasmic adaptor Grb2 to surface IgG and IgE provides antigen receptor-intrinsic costimulation to class-switched B cells. Engels N, König LM, Heemann C, Lutz J, Tsubata T, Griep S, Schrader V, Wienands J., Nat Immunol. 2009 Sep;10(9):1018-25.). Based on these findings functional studies of Grb2 and its potential binding partners are in progress, which include the generation and analysis of Grb2 mutants that lack either the N- or C-terminal SH3 domain or carry amino acid substitutions in the binding sites for know interaction partners. At the same time we are trying to generate of Grb2-deficient cell lines.
Project 5: Development of knockdowns and humanized mice
Development of human B cells in reconstituted double knockout Rag2/Il2rg deficient mice
Early steps in human B cell development were analysed in humanized mice. Human B cells developing from engrafted CD34+ hematopoietic stem cells were found in bone marrow and blood but rarely in the spleen or in other secondary lymphoid organs. To improve the development of human lymphocytes in a murine cytokine environment Rag2/Il2rg mice transgenic for human IL7 were generated. To facilitate the construction of mice expressing human cytokines and growth factors supporting the development of human hematopoetic cells we also started to generate ES cell lines on the Rag2/Il2rg background. In addition, CD34+ HSC were engrafted into Rag2/Il2rg transgenic for human BAFF. The analysis of these mice is under way.
Project 6: Humanized mice to study memory and plasma cells in CVID
IgM memory B cells play a fundamental role in the defence against encapsulated bacteria producing anti-bacterial IgM antibodies [1-3]. These bind to the capsular polysaccharides of rendering them targets of monocyte-macrophage cells.
Therefore, IgM antibodies are indispensable for the removal of bacteria circulating in the blood, collected by the filtering system of the spleen and invading tissues.
Antibodies, however, also play an important role before tissue invasion, forming a protective layer on mucosal surfaces. Bronchial and intestinal epithelia are both covered by a film of antibodies of IgA isotype. Secretory IgA (sIgA) is secreted by mucosal plasma cells and then transported from the basal to the apical side of epithelial cells by the poly-Ig receptor. In the mouse most of the IgA at epithelial surfaces is secreted in a T-independent manner and originate from a defined type of B cells, the B-1a. This population is ontogenetically and functionally distinct from conventional B-2 B cells [4].
In humans the important question of the origin of mucosal plasma cells has never been addressed.
Introduction to WP5: Translational Research
Project 1 is to improve the quantity and quality of immune-phenotypic data of peripheral lymphocyte subsets from PAD patients. Currently, the flow cytometric analysis is performed by different laboratories using different panels of monoclonal antibodies. A standardized assessment is underway.
Results of this subproject can be viewed on the new website www.ipidnet.org.
Project 2 is to define genetic variations in PAD patients with agammaglobulinemia and low B cell counts who do not have a definitive genetic/molecular diagnosis by the analysis of specific points of arrest of B cell differentiation in the bone marrow.
We have undertaken bone marrow biopsy and FACS immunophenotyping in patients with CVID in order to assess if the clinical and immunological phenotype correlates with a specific developmental block/alteration in the early stages of B cell development. Our preliminary data are very interesting and promising: early B cell development in CVID is disturbed at different stages in a large percentage of CVID patients and apparently more patients share a similar developmental defect.
Project 3 is to analyse the role of the different B lymphocyte subpopulation in the protection from infections by encapsulated bacteria. We showed that IgM anti PnPS antibodies in PAD correlate with the capacity to differentiate into plasmablasts. We also demonstrated that a reduced proliferative response to CpG correlates with an increased lung infection rate. Due to the impaired proliferative response, CpG induced plasmablast differentiation is decreased in patients with CD21low (>10%).
Project 4 is to define quantitative and qualitative antibody defects in PAD patients.
Project 5 is to identify biomarkers to predict the risk of infusion reaction in PAD patients. Anti-IgA antibodies have been proposed as predictive factors for the occurrence of adverse reactions to IgG replacement therapy which contains traces of IgA in PAD patients, but their pathogenetic role and their isotype (IgG vs IgE) is still unclear.
Project Results:
THE NATURAL COURSE AND PHENOTYPE OF PADs
The objective for Workpackage 1 was to better describe the natural phenotype of primary antibody deficiencies. To this end, we used the existing database of the European Society for Primary Immune Deficiencies (ESID), which has been developed with help of the European Framework 6 Program and commercial funding during the years 2004 to 2006, and local pre-existing national registries such as the Italian IPInet registry. During the funding period of EURO-PADnet, we increased the total number of PAD patients in the ESID Registry from 3,369 in May 2008 to 5,554 on August 2009 by 2185 patients. This has further increased to 7,466 patients as of April 2011. The total increase over the three-year period hence was 4,099 patients, or 122%.
We aimed at better describing the T and B cell phenotype in antibody deficiencies, to identify pathogenic and clinically relevant patterns of dysregulation, and to improve current classifications schemes based on a large European cohort of patients with PAD. Data of 411 patients have been collected from 6 centers and are currently evaluated for the association of changes in T and B cell phenotype with the clinical manifestation of splenomegaly, lymphadenopathy, autoimmune disease, granulomatous disease, opportunistic infection, conventional infection of the respiratory and gastrointestinal tract, bronchiectasis and interstitial lung disease. Statistical evaluation is under way. Data is posted on www.ipidnet.org. As an example shall serve the following published study: We were interested in the question of whether the congenital lack of B cells actually had any influence on the development of the T cell compartment in patients with agammaglobulinaemia. The CD4 T cell memory compartment was reduced in patients with XLA of all ages. This T cell subset encompasses both CD4(+) CD45RO(+) and CD4(+) CD45RO(+) CXCR5(+) cells. This observation was confirmed in patients with CVID who had <2% B cells, suggesting that not the lack of Bruton's tyrosine kinase but the lack of B cells is most probably the cause of the impaired CD4 T cell maturation. We postulate that this defect is a correlate of the observed paucity of germinal centres in XLA. Our results support the importance of the interplay between B and T cells in the germinal centre for the activation of CD4 T cells in humans.
EURO-PADnet researchers also accompanied a prospective clinical study on the natural course of antibody deficiencies over a cumulative follow-up period of 1,365 patient-years. This study was carried out by centres in Italy. It revealed that for X-linked agammaglobulinemia, the only co-morbidity risk factor for pneumonia was the presence of bronchiectasis. In common variable immunodeficiency (CVID), the data allowed us to identify a clinical phenotype characterised by a high pneumonia risk: patients with low IgG and IgA levels at diagnosis; patients who had IgA level <7 mg/dL and who had bronchiectasis. Patients with pneumonia did not have significant lower IgG trough levels than patients without pneumonia, with the exception of patients whose IgG trough levels were persistently <400 mg/dL.
Based on a cohort of paediatric (n = 232) and adult (n = 73) patients (mean age 36.5 SD 17.7 range 1.6 to 79.3 years) with CVID who underwent chest CT, the overall prevalence of structural bronchial pathology was 72%, with bronchiectasis present in 63%. Parenchymal pathology, such as lines, bands and nodules, was detected in 76%. A pathological lung function as indicated by a forced expiratory volume in 1 second (FEV1) of < 80% predicted was present in 39.5% of the patients who underwent lung function analysis (n=185). Importantly, 55% of the patients with a normal lung function had bronchiectasis. Air trapping was evident in up to 30%.
The consortium established an international, multidisciplinary expert group, the "Chest CT in ADS Group", with 26 immunologists, 13 radiologists, and 7 pulmonologists from 20 academic centres from 9 countries, www.chest-ct-group.eu. The Chest CT in ADS Group plans to hold a European meeting consensus conference on the use of chest CT scans in primary antibody deficiencies by January, 2012.
To link the data in the clinical databases, our centres collected biological samples of the documented patients for research. The centre in Brescia has collected 150 PID patients, 150 SIgA patients, and 15 PID families; the Karolinska Institute has collected 250 CVID samples, 1200 sIgAD samples and 100 samples with other PIDs; INSERM in Paris has collected 98 samples with class switch recombination defects. Rome has collected approximately 1,450 samples, Freiburg about 660; and London more than 864 samples. These sample collections provided a huge resource for future projects and collaborations and put each of the centres into a unique position. This collection of samples in unprecedented in the world.
THE GENETIC ETIOLOGY OF PADs
Linkage analysis in antibody-deficient families:
Linkage analysis is a non-hypothesis-driven pure positional cloning approach. By linkage analysis, the consortium was successful in identifying a novel genetic cause for CVID in four autosomal recessive families from Belgium, Israel and Iran. The causative gene was previously not known to be important for the immune system and most likely affects autophagy of lymphocytes. What is especially interesting for the pathophysiological understanding is the fact that the phenotype in these patients include autoimmunity. This paper is currently under revision.
In the other CVID families we located the genetic defect to certain chromosomal regions and positional candidate genes were sequenced. The autosomal-dominant CVID family FR1 is linked to chromosomes 5, 6, and 7. The autosomal-dominant CVID family NL1 is linked to chromosome 4. The autosomal-dominant CVID family FR15 is linked to chromosome 3. The autosomal-recessive CVID family US2 is linked to chromosomes 1, 5, and 20. The autosomal-recessive CVID families Port1 and cv32 have multiple regions of homozygosity as they are quite small. Finally, the autosomal-dominant CVID family IT4 is linked to chromosome 4 and achieved a LOD score of 3.9.
However, as this purely positional approach followed by Sanger sequencing of possible candidates in the linkage regions was not fruitful in these families for the last few months. Hence we are now employing whole exome sequencing to identify the genetic defect. This will be carried out in part by the EU project EURO GENEscan where UCL and KI are partners.
Genome-wide association studies:
Tissue typing on Iranian and Swedish CVID patients/families reveals that the previously suggested HLA association in CVID/IgAD multicase families is largely restricted to those carrying the HLA B8, DR3, DQ2 haplotype and that the pattern in "sporadic" CVID patients is largely random. The addition of 100 more cases confirmed the initial findings.
Analysis of the contribution of genes within the HLA region in 100 multicase IgAD families is ongoing at KI (analysis of degree of HLA sharing among affected/non-affected siblings of the proband). This material is also currently being extended, including additional multicase families. A preliminary analysis, using the formula of Risch, has suggested that genes within the MHC region contribute approximately 40% of the disease risk in IgAD.
The candidate gene approach:
In an exceptional effort, fourteen functional candidate genes for CVID were studied by a single team in Brescia, Italy; some of which may represent disease-causing genes, others may be modifier genes for CVID.
The genetics of hyper-IgM syndromes:
Whole exome sequencing in five patients from different families with hyper IgM syndromes of unknown cause was performed. In two siblings we found a mutation in the CD40L gene, although CD40L-deficiency seemed previously excluded by normal CD40L expression. This observation highlights the fact that protein detection alone does not exclude a mutation. In two other families, two genes were found mutated on both alleles. These genes encode proteins involved in both chromatin modification and DNA repair. These mutations are found "likely damaging" in Polyphen data base and are not described as polymorphisms in data bases. No murine model is available. Mutations in both genes are now looked for in a large cohort of HIGM patients (80 patients). To validate the role of these molecules in CSR, shRNA lentiviral transduction of CSR-induced CH12 cell line is now on-going. In another non consanguineous family, with two affected children, we found in both patients a few genes carrying bi-allelic mutations, that are now under investigation. Five other informative families have been studied by exome sequencing and results are pending. In a consanguineous and informative family, in which a homozygous genetic region has been already determined, a gene was found mutated as homozygous in the two affected children. However, a normal sibling and the healthy mother present the same genotype, excluding the role of this mutation in the pathogeny of the disease. Consequently, all the exons present in this region, not completely or not perfectly covered by the exome, were sequenced, with no detected abnormalities. Expression the cDNA level of all the genes comprised in that region will be assessed as soon as material is available (EBV B cell line) in order to detect splice defects (due to intronic mutations) or defective expression (mutations in promotor).
IN VITRO MODELS OF PAD
BAFF as the survival factor for peripheral B cells:
Interaction of the ligand BAFF and its receptor (BAFFR, which is exclusively expressed on B cells) is one of the most important survival signals, if not the most important survival signal to B cells. As antibody deficiencies often lack of sufficient B cell numbers in the periphery, and also B cell differentiation defects have been demonstrated in patients with antibody deficiencies (B cell differentiation is tightly linked to proliferation during ongoing immune responses), the study of this ligand-receptor pair was a paramount interest.
We analyzed BAFF serum levels in CVID and BTK-deficient patients, 2 patients. Highest BAFF serum concentrations were found in BTK patients (10 - 100 ng/ml) followed by a group of 21 CVID patients with < 2% of B cells and > 10 ng/ml BAFF. The analysis revealed that BAFF serum concentrations correlate inversely with B cell numbers but not with autoimmunity or B cell neoplasia. Sequencing of the BAFF genes in selected patients did not reveal mutations. BAFF-R expression was studied by FACS in > 80 patients. About 15% expressed significantly reduced BAFF-R levels. Expression went back to normal in EBV lines indicating regulatory changes in CVID patients.
We developed several readout systems for BAFF-R function including analysis of the alternative NF-kB activation and p100 processing by western blotting and in vitro survival assays for human primary B cells and for murine B cells from BAFF-R deficient mice which were transduced with various BAFF-R mutants. Analysis of BAFF-R receptor structure revealed that BAFF-R associates in multimeric receptor complexes. Different from wildtype, the BAFF-R variant P21R shows impaired complex formation, ligand binding and NF-kB activation. Presently we are studying the impacts of BAFF-R multimerization on the association with downstream signalling components such as TRAF6, TRAF2 and TRAF3.
CD21low B Cell Populations In Primary Human Antibody Deficiencies:
The so called CD21-low B cells population has been primarily described in patients with CVID, but has recently also been shown to be increased in cohorts of autoimmune diseases. Interestingly, CVID patients also suffer from autoimmune phenomena. CD21(low) B cells were shown to be profoundly anergic, and defects of BCR-mediated calcium signaling and of T cells have been described in CVID. We found that also the classical naïve B cells from many CVID patients proliferated poorly. The B cells of some CVID patients had a reduced capacity to divide reminiscent of the proliferative arrest associated with replicative senescence which is known to be influenced by telomere length. Thus, we investigated whether lymphocyte dysfunction in CVID was related to telomere-dependent replicative senescence, and found that both the B and the T cells from some CVID patients had significantly shorter telomeres compared with B and T cells from other CVID patients. Telomere lengths in B and T cells were significantly correlated, indicating that the rate of telomere attrition in lymphocytes is an individual characteristic of CVID patients. Our findings suggest that telomere-dependent replicative senescence contributes to the immune dysfunction of some CVID patients, and may provide an important clue for a better understanding of the pathogenesis of CVID.
IFN Signature in CD21-low B cells:
The percentage of CD21low B cells was shown to be significantly expanded in HCV-viremic patients supporting the idea of an association of CD21low B cells to IFN rich environment. Detailed examination of IFN signature in CD21low B cells of CVID patients demonstrated increased expression of MXA, OAS1, while other targets like IFIT1 showed higher expression in naïve B cells of patients. The IFN signature was B cell specific in CVID, but IFN itself failed to induce the phenotype of CD21low B cells in vitro.
Plasma cells in antibody deficiencies:
Our research activities focused on the role of epigenetic and protein modifying histone acetylase / deacetylase activities during plasma cell development. We found that inhibition of HDAC by valproic acid and other iHDAC blocks IL21R signaling by affecting STAT5 phosphorylation The VPA-mediated inhibition of IL21R signalling impairs plasma cell formation and antibody secretion.
IgM memory B cells in antibody deficiencies:
In order to demonstrate the developmental connection between IgM memory and plasma cells, we first determined the presence of sIgA and/or IgM in jejunal biopsies of two groups of CVID patients, with or without circulating IgM memory B cells and in splenectomised patients. By confocal microscopy we analyzed IgA plasma cells and sIgA on the epithelial cells in jejunal biopsies. After splenectomy the reduction of memory B cells in the peripheral blood is associated to a significant diminution of IgA plasma cells and the disruption of the film of sIgA on the luminal side of epithelial cells. In CVID the absence of sIgA in the gut and IgA in the serum are associated to a significant reduction of memory B cells in the peripheral blood and high risk of respiratory infections. CVID patients with sIgA at mucosal sites had a normal number of circulating memory B cells, detectable serum IgA and have a low risk of respiratory infections. Memory B cells are indispensable for the production of sIgA at mucosal sites. The disruption of the sIgA film in the gut impairs the local defense against invading pathogens.
Identifying precursors of IgM memory B cells:
The expansion of specific B cell subsets expressing low levels of CD21 in patients with PAD is an expression of an underlying pathogenic dysregulation. These B cell subsets are being used to define subgroups of patients for specific functional as well as molecular characterization. We demonstrated that, upon exposure to CpG, a fraction (26%) of transitional B cells proliferate (P) and differentiate either to IgM-secreting plasma cells (19.2%) or to cells phenotypically identical to IgM memory B cells (41.3%). The remaining non-proliferating (NP) B cells down-regulate the expression of CD24 and CD38, thus resembling mature B cells. We have also demonstrated that natural antibodies, including immunoglobulins with anti-pneumoccocal specificity, can be generated "in vitro" from CpG-stimulated transitional B cells. The progenitors of IgM memory B cells are recent bone marrow emigrant transitional B cells, which through the blood arrive to the spleen, an optimal anatomical site where blood-borne pathogens and cellular debris are filtered by resident macrophages. Here, transitional B cells diversify their antigen receptors in the presence of bacterial components, e.g. CpG DNA. CpG induces somatic mutations in the Ig heavy chains (VH) regions in P B cells, as opposed to the prevalence of germline sequences in the NP cells. The mutations are confined almost exclusively to the framework regions. Moreover, the incidence of mutations is VH-specific, being VH1 and VH4 the most mutated, and VH5 the least mutated family. In summary, early recognition of bacterial DNA induces proliferation, SHM, antibody secretion or survival in transitional B cells. IgM memory B cells in one year-old children mirror the mutation pattern seen in CpG-stimulated transitional B cells.
The role of TACI mutations for antibody deficiencies:
Coding variants in TACI, aka tumor necrosis factor receptor superfamily member 13B (TNFRSF13B) have been implicated in common variable immunodeficiency (CVID), but the functional effects of such mutations in relation to the development of the disease have not been entirely established. To examine the potential contribution of TNFRSF13B variants to CVID, we studied the mutant alleles in vitro. In summary we found that each mutation acts via a different pathomechanism: I. frameshift mutations, nonsense or missense, leading to loss of protein expression (examples ins204A, S194X, L171R); II. expressed missense mutant that fails to bind ligand (example C104R, I87N); III. mutant with intracellular retention of mutated protein (examples C104R and L171R); IV. failure of trimeric assembly as proposed by Lee JJ Blood. 2009;114:2254-62 (example A181E); V. mutants with no effect at all = non pathogenic variants (examples W40R, H81N, K188M).
Most importantly, we were able to show that having a pathogenic mutation in TACI increases the risk for developing CVID by 4. Presence of a heterozygous mutation was associated with antibody deficiency (P <.001 , relative risk 3.6). Heterozygosity for the most common mutation, C104R, was associated with disease (P < .001, relative risk 4.2). Furthermore, heterozygosity for C104R was associated with low numbers of IgD-CD27+ B cells (P = .019), benign lymphoproliferation (P < .001), and autoimmune complications (P = .001). These associations indicate that C104R heterozygosity increases the risk for common variable immunodeficiency disorders and influences clinical presentation. (Salzer et al., Blood. 2009; 113:1967-1976).
Class-switch Recombination defects and Mismatch Repair defects:
This consortium described PMS2-deficiency as causative of a partial Class-Switch-Recombination (CSR) defect. The group went on studying other MisMatch-Repair-Deficiencies. Seven further PMS2-, two MLH1- and seven MSH6-deficient patients could be diagnosed. All of the patients presented with a partial CSR defect. Interestingly enough, the group provided evidence that PMS2 is involved in the CSR-induced generation of DNA double strand breaks (DSB) in switch regions, likely because of its endonuclease domain.
The group observed that both double-strand-break (DSB) generation and repair is impaired in PMS2-deficiency, we asked the question whether impaired DSB generation could influence their repair. For this purpose, we studied S?-S? junctions in B cells in which DSB occurrence is very likely (AID+/-AID Cter?/+) or certainly impaired (UNG-/-). We showed a progressive bias towards microhomology usage for switch junction repair from AID+/- to AID Cter?/+ and to UNG-/-. Our results strongly suggest for the first time that proper induction of CSR-DSB either recruits NHEJ, or excludes other repair pathways (Alternative End Joining).
In other hyper-IgM patients, we identified a transcription elongation factor, the expression of which was strongly diminished in EBV B cell lines from four patients. Although the RNA level and the genomic sequence was found normal, this molecule appears indirectly down-regulated in these patients. The role of this molecule in CSR is being studied by shRNA in CSR-induced CH12 cell line.
MOUSE MODELS FOR PAD
TACI-knock-in mice as a model for antibody deficiencies:
Mutations in TNFRSF13B, the gene encoding transmembrane activator and calcium modulator cyclophilin ligand interactor (TACI), are found in 10% of patients with common variable immunodeficiency. However, the most commonly detected mutation is the heterozygous change C104R, which is also found in 0.5% to 1% of healthy subjects. The contribution of the C104R mutation to the B-cell defects observed in patients with common variable immunodeficiency therefore remains unclear. We sought to define the functional consequences of the C104R mutation on B-cell function. A knock-in mouse with the equivalent mutation murine TACI (mTACI) C76R was generated as a physiologically relevant model of human disease. We examined homozygous and heterozygous C76R mutant mice alongside wild-type littermates and studied specific B-cell lineages and antibody responses to T cell-independent and T cell-dependent challenge. Mice heterozygous and homozygous for mTACI C76R exhibit significant B-cell dysfunction with splenomegaly, marginal zone B-cell expansion, diminished immunoglobulin production and serological responses to T cell-independent antigen, and abnormal immunoglobulin synthesis. These data show that the C104R mutation and its murine equivalent, C76R, can significantly disrupt TACI function, probably through haploinsufficiency. Furthermore, the heterozygous C76R mutation alone is sufficient to disturb B-cell function with lymphoproliferation and immunoglobulin production defects.
Overall we have been able to show significant abnormalities of B cell homeostasis and humoral function in TACI C76R deficient mice. These data clearly show that the C76R mutation alone in mice has a detrimental effect on B cell function and if extrapolated to the human setting suggests that TACI C104R affects human B cells and contributes to the B cell defects seen in CVID.
SAP knockout mice as a model for hypogammaglobulinaemia in XLP:
Vaccination studies in SAP knockout mice suggest that responses to T cell dependent antigens are severely compromised in SAP deficient mice and further support the argument that T cell help is defective in SAP deficiency rather then there being an intrinsic B cell abnormality. We then investigated whether humoral function could be restored either by standard bone marrow transplant or through gene modification of murine stem cells. SAP knockout mice were subject to lethal irradiated and then reconstituted with either BM from wild type mice or from SAP KO lin-ve cells transduced with lentiviral vectors encoding SAP or a reporter vector as a control. Mice were then sacrificed at 12 weeks and were analysed from total IgG, IGg subclasses and for reconstitution of cytokine production in splenic CD4 T cells.
Naïve total IgG levels are significantly lower in SAP KO mice than in wild type mice. Following BMT from wild type mice and gene transfer using SAP encoding vectors we saw an increase in total IgG levels that was statistically significant. No increase was seen following gene transfer from a reporter gene encoding vector. Similarly both BMT and SAP gene transfer both resulted in increase in levels of IgG1, IgG2a and IgG2b whereas no increase was seen following reporter gene transfer. To look at whether this increase may have been due to restoration of T cell help, splenic CD4 cells from the different groups of mice were either unstimulated or stimulated with an anti-CD3 antibody and analysed for IL-4 or IL-10 secretion by ELISA. In comparison to SAP KO mice, both BMT and SAP gene transfer resulted in increased levels of IL-4 and IL-10 production whereas no difference in cytokine production was seen in mice receiving reporter gene modified cells.
Together, these data suggest that there is a significant defect of Ig production both in the naïve state and in response to T dependent antigen stimulation. These and other data suggest the abnormalities are due to an absence of T cell help. Our studies show that this humoral defect can be corrected by BMT and SAP gene therapy and the correction is due to the restoration of effective function in CD4+ T cells. Further experiments vaccinating mice after reconstitution are required to demonstrate that the full humoral function can be restored in these mice by BMT or gene therapy.
Analysis of membrane-IgG1- and IgE-signalling in a murine knock-in model:
Using stable isotope labeling with amino acids in cell culture (SILAC) in B cells, we have applied this technology to analyze the Spleen tyrosine kinase (Syk) and mapped Syk phosphorylation sites and elucidate the B-lymphoid interactome of human Syk.
The major task was the generation and analysis of gene-targeted mice, which express a mutated form of the mIgG1 (IgG1-YF) and mIgE (IgE-YF) B cell receptors. Like with the IgE-YF mice, chimeric IgG1-YF mice have been generated, and germline transmission of the modified allele has been achieved. Both mouse strains have been backcrossed to remove cre or flp recombinase transgenes, transferred to the barrier facility by in-vitro fertilization and bred to homozygosity. Cohorts of both strains have been immunized with model antigens and the immune response has been measured by ELISA. Preliminary results of these studies have been presented on the PADNet meetings.
Taken together, the two successfully established mouse models will provide the means to analyse the importance of tyrosine-phosphorylated cytoplasmic domains of mIgG and mIgE in BCR signaling in a physiological setting and to unravel isotype-specific signaling cascades. These results are important for understanding of PAD and, in the case of IgE, allergic diseases.
BCR-intrinsic costimulation of memory B cells:
During the project we have identified and published the adaptor molecule Grb2 as a pivotal factor for the BCR-intrinsic costimulation of memory B cells. We then set out to analyze the Grb2 interactome and identified Bruton's tyrosine kinase (Btk) as an interaction partner of Grb2. For further analysis, we generated a Btk-deficient variant of the human B cell line DG75, which will allow us to test the signalling potential of different BCR-variants in the presence or absence of Btk.
We have also performed experiments on Stap1, which is a unique IgE binding partner discovered during the first half of the project. On the one hand, we used the Stap1-deficient human B cell line DG75 generated in the first half to analyze the composition of IgE-interacting proteins in the absence of Stap1. On the other hand, we generated cell lines expressing a modified version of Stap1, which carries an affinity tag and allowed the purification of Stap1-interacting proteins. The subsequent mass spectrometric analysis of the purified complexes revealed several unknown interaction partners, and allowed us to draw first conclusions on the role and function of Stap1 in the BCR-intrinsic costimulation of memory B cells.
Murine knock-in model for antigen receptor signalling
New: Optimized splice correcting oligonucleotides (SCO) for gene correction of XLA in a mouse model:
The ability to correct the aberrant splicing varied for different SCOs. However, splice correcting LNA based oligos are not able to efficiently be taken up by lymphoid primary cells in our system (mouse B cells and human monocytes). Only by the use of electroporation techniques we were able to successfully correct the aberrant BTK splicing in those cell types. As such a new oligonucleotide chemistry - morpholino - was used in combination with a covalently attached cell penetrating peptide. Cell penetrating peptides covalently attached to antisense morpholino splice correcting oligomers (CPP-SCO) were used without the need for electroporation and were extremely efficient in correction of BTK aberrant splicing in patient monocytes. These splice correcting oligos equiped with the CPP are able, by the use of the membrane translocation properties of the peptide, to accumulate inside the cell when added exogenously to the medium. These cell uptake properties of the peptide-oligo conjugates makes them ideal candidates for further testing in in vivo situation.
The main objective of our work within the consortium is to improve our understanding on molecular basis of the Btk deficiency. We have developed a new model: BAC-transgenic mice carrying the human genomic BTK. We have two mouse strains; one carrying the WT human BTK gene and another carrying mutated human BTK gene. At the moment the transgenic mice are bred onto the Btk-/- C57BL6 mice and we assume that they will provide a new excellent model for studying different aspects of Btk deficiency. Furthermore, we aimed to correct the mutated BTK in the BAC-transgenic mouse as a proof of concept for the treatment possibilities. We achieved gene correction of the mutated BTK gene in a transgenic mouse model. The transgenic mice are now bred on the Btk-/- mice, providing an excellent humanized model to study different aspects of Btk and XLA disease. We have been able to successfully correct the mutated human BTK gene in our animal model and we could prove that both at the RNA and protein level. Gene correction by using the same means of monocytes fom XLA patients carrying similar mutations to prove the efficacy of the method in human primary cells was achieved. Gene expression profiling in Btk-/- and WT B cells stimulated with CpG as an important component in our studies on molecular basis of Btk deficiency. Our preliminary data indicate that the CpG stimulation alters the phenotype of Btk-/- cells making them more alike WT B cells (an important factor in XLA is that B-lymphocytes do not survive). This work is submitted under: Moreno P et.al. "Correction of splicing mutations in X-linked agammaglobulinema (XLA) using an exon skipping strategy".
Importantly, XLA has a very decisive feature, namely that the end-stage cells, the plasma cells, do not express any BTK protein. This means that if regular B-lymphocytes can develop as a result of the treatment, subsequent immunization is anticipated to generate long-lived plasma cells. This also means that a short-term treatment can mediate a long-term clinical outcome. If the animal model shows an in vivo treatment effect, the personalized medicines developed in this program could therefore hopefully, in the not too distant future, also be used as experimental drugs in the affected patients.
Development of human B cells in humanised mice:
In the humanized mouse model the co-transplantation of mesenchymal stem cells (MSC) may improve engraftment and develompent of human hematopoietic cells. We found that MSCs isolated from adult bone marrow are a heterogeneous cell population and therefore we are currently trying to identify those cells which provide the strongest support for HSC.
We studied the development of human B cells in humanized mice of different ages. In total, 273 mice were transplanted at UKL-FR with human CD34+ HSC. Bone marrow engraftment was found in 1/4 of the mice. For comparison, bone marrow B cell development was analyzed in BM samples from CVID patients.
In total, bone marrow samples of 48 patients and 15 controls were analyzed by immunohistology. From 25 patient, fresh aspirates were available and studied by immunophenotyping. In 94% of the patients' BM plasma cells were either absent or significantly reduced and correlated with serum IgG levels. Biopsies from CVID patients had significantly more diffuse and nodular T cell infiltrates than biopsies from controls. These infiltrates correlated with autoimmune cytopenia but not with other clinical symptoms nor with disease duration and peripheral B cell counts. Nodular T cell infiltrates correlated significantly with circulating memory T cells, elevated soluble IL2-receptor and neopterin serum levels indicating an activated T cell compartment in most patients. Nine out of 25 patients had a partial block in B cell development at the pre-B-I to pre-B-II stage. Since the developmental block correlates with lower transitional and mature B cell counts in the periphery, we propose that these patients might form a new subgroup of CVID patients.
TRANSLATIONAL RESEARCH
Biomarkers in CVID :
The use of b2MG and sIL2R as well as free light chains as biomarkers were evaluated retrospectively in a cohort of 100 CVID patients. Both b2MG and sIL2R correlate highly significant with the appearance of splenomegaly, lymphadenopathy, autoimmune disease and granulomatous disease. At the same time there is a significant association with cellular alterations of B and T cells most pronounced in CVID Ia patients. The value of these markers at the time of malignant transition in CVID patients with secondary lymphoma is under current evaluation.
Definition of protective role of IgM memory B cells and IgM anti PnPS antibodies in PAD:
We defined that IgM anti PnPS antibodies in PAD correlate with the capacity to differentiate into plasmablasts. The residual capacity to produce IgM anti PnPS antibodies correlated with the risk to develop pneumonia and chronic lung disease. B-cell subsets and their functional development have been also analyzed in 74 newborns from birth to 6 months of life. Moreover, we evaluated natural antibody production in vitro. The results documented a predominance of naive B-lymphocytes at all time-points evaluated, decreasing from birth to 6 months (p=0.009). The percentages of CD27+IgD+ and CD27+IgDneg memory B-cells were very low at birth and significantly increased only at 6 months (p=0.02 and p less than 0.001 respectively). We found a significant increase only in in vitro stimulated IgG production at 6 months as compared to birth (p less than 0.001). Moreover, a lower secretion of anti-Pn IgM antibodies up to 6 months of age, as compared to controls was observed. Our results underline that the susceptibility and severe course of infection in the neonate can be attributed, at least in part, to the lack of pre-existing immunological memory and competent adaptive immunity.
The role of IgM memory B cells in the protection from encapsulated bacteria:
In this project we evaluated B-cell subsets and their functional development in 74 newborns from birth to 6 months of life. Moreover, we evaluated natural antibody production in vitro. The results documented a predominance of naive B-lymphocytes at all time-points evaluated, decreasing from birth to 6 months (p=0.009). The percentages of CD27+IgD+ and CD27+IgDneg memory B-cells were very low at birth and significantly increased only at 6 months (p=0.02 and p less than 0.001 respectively). We found a significant increase only in in vitro stimulated IgG production at 6 months as compared to birth (p less than 0.001). Moreover, a lower secretion of anti-Pn IgM antibodies up to 6 months of age, as compared to controls was observed. Our results underline that the susceptibility and severe course of infection in the neonate can be attributed, at least in part, to the lack of pre-existing immunological memory and competent adaptive immunity.
Definition of B cell stimulation as prognostic indicator in PAD:
Reduced proliferative response to CpG correlates to increased lung infection rate. Due to reduced proliferative response, CpG induced plasmablast differentiation is reduced in patients with CD21low cells to <10%.
Development of prototype ELISA for IgM anti PnPs response:
The aim of the subproject is the development of a reliable and easy-to-use ELISA for anti-pneumococcal polysaccharide (PnPS) antibodies of IgM isotype. The assay is intended to assess the function of B cells in patients with antibody deficiency sndrome in vivo. The project was conducted in co-operation with our industrial partner, The Binding Site (TBS), Schwetzingen, Germany. N = 20 volunteers were vaccinated with a 23 valent pneumococcal polysaccharide vaccine (Pneumovax®). Subsequently, serum and cellular components were obtained at 8 time points within 2 months. The vaccination study was performed to i) determine the optimal time point for assessment of the anti PnPS-IgM-antibody response following PnPS vaccination and ii) to obtain specimen with high levels of anti-PnPS-IgM antibody contents required for the ELISA development. The anti-PnPS-IgM-ELISA was developed on the basis of a commercial anti-PnPS-IgG-ELISA provided by our industrial partner TBS. The assay was optimised for sensitivity and retest-reliabbility by serial variations of IgG removing agents and crossreactive antibodies binding substances. The ELISA is now ready for definitive measurements of the IgM immune response of the healthy vaccinees. The antibody immune response can be compared to the detection of PnPS specific B-cells at the same time points.
Evaluation of IgM anti PnPS response in PAD patients after vaccination:
The serum level of IgM antibodies against different pneumococcal serotypes (type 1, 3, 4, 14) has been analysed according to the D. Goldbatt (WHO) recommendations before and after vaccination with unconjugated pneumococcal vaccine. In cooperation with The Binding Site (Germany) an easy-to-use ELISA for anti-PnPS antibodies of IgA and IgM isotype has been developed. This test is based on a commercially available 23-valent anti-PnPS ELISA for IgG antibodies (Pneumococcal Capsular Polysaccharide (PCP) Assay, The binding Site).
Outcome: A good correlation of the IgM 23-valent anti-PnPS with each single serotypes IgM titre (type 1, 3, 4, 14) was observed. The serum level of specific antibody IgM and IgA against single serotypes (type 1, 3, 4, 14) and the IgM and IgA 23-valent anti-PnPS have been correlated with the B lymphocyte subsets (according to the IPIDNET panel). A response to at least two serotypes, doubling the IgM or IgA level was considered a good response to vaccination. Based on this definition, a good IgM response was significantly related to higher numbers of switched memory B cells and IgM memory B cells and with lower levels of CD21low B cells. Instead a good IgA response to vaccination was significantly related to higher numbers of switched memory B cells and with lower levels of CD21low B cells. The clinical evaluation included the clinical history (frequency of pneumonias), the chest CT scan for the detection of bronchiectasis and other structural changes and the pulmonary function test (FEV1 of less than 80% [predicted]) over a period of 36 months. An inverse correlation was found between the response to the vaccination and the frequency of pneumonia and between the response to the vaccination and the presence of bronchiectasis at the chest CT scan. Conversely, no significant correlation was found between the response to immunization and the pulmonary function test (FEV1). Observing the distribution of the response to vaccination with specific IgA and IgM 23-valent anti-PnPS production, a cut-off level has been suggested in order to define a response to vaccination in subjects with PAD: IgM>20 U/ml and IgA>0.15 µg/ml. Based on this definition, three groups of patients have been described (see graph below): Patients with good IgM and IgA response (complete responders), patients with no response at all (non-responders) and patients with IgM response only (IgM only). The latter group of patients (IgM only) has defined intermediate level of switched memory B cells (lower than complete responders but higher that non-responders) and lower numbers of CD21low B cells than complete responder patients. Comparing the response to vaccination with IgA and IgM 23-valent anti-PnPS production to the clinical findings, a stronger inverse correlation was found between the complete response to the vaccination and the frequency of pneumonia and between the response to the vaccination and the presence of bronchiectasis at the chest CT scan, while an intermediate inverse correlation was observed in the IgM only group. (see Figure 1)
Assessing antibody response in patients with hypogammaglobulinaemia by ELISPOT:
Hypogammaglobulinaemic patients are often started on immunoglobulin substitution therapy before antibody production is adequately evaluated. In such a situation, it is difficult to segregate transferred from antigen-induced specific antibody. Therefore we characterized changes in B-cell subpopulations in hypogammaglobulinaemic patients, including plasmablasts, in peripheral blood by flow cytometry after in vivo antigen challenge. We investigated the specificity of antibody production on the B-cell level by ELISPOT, which is independent of substitution therapy.
Hypogammaglobulinaemic patients and healthy volunteers were immunized with tetanus toxoid (T-cell dependent) and pneumococcal polysaccharide (T-cell independent) vaccines. Specific antibody levels were measured by ELISA. ELISPOT was used before and on day 7 after vaccination. B-cell subpopulations were examined by flow cytometry. In the control group, an increase in circulating plasmablasts (IgD-CD27++, IgM-CD38++ gated from CD19+ B cells) on day 7 post immunization corresponded with the appearance of spot forming cells. In contrast, patients with Common variable immunodeficiency (CVID) who are characterized by hypogammaglobulinaemia and impaired antibody production failed to increase plasmablasts significantly in peripheral blood after antigen challenge.
Our observation that the majority of CVID patients lack antigen specific spot forming B cells and fail to increase circulating plasmablasts following in vivo antigen challenge provides a rapid screening test to demonstrate defective antibody responses in CVID patients.
Clinical Implications and Conclusion: Identification of circulating plasmablasts after vaccination is a new simple flow based test to assess antibody responses in hypogammaglobulinaemic patients, even if on immunoglobulin (IVIG or SCIG) replacement therapy.
Development of an IgM-specific anti PnPS ELISA based on 23 serotypes from S. pneumonia and
Assessment of the IgM and IgG anti PnPS antibody response after vaccination with pneumococcal unconjugated vaccine in PAD patients:
As an immunodiagnostic company the role of The Binding Site in the EUROPADnet project was to translate the observations from primary antibody deficient (PAD) patients into a medical device so the data and results could be utilised in patient care.
Since there is a high risk of bacterial infections in patients suffering from common variable immunodeficiency (CVID) we supported the investigation for the hypothesis that IgA and IgM responses to pneumovax may improve the determination of the risk of pulmonary morbidity in PAD patients (work package 5).
Along with the study group we have facilitated the development of both PCP IgM and IgA ELISAs. The assays have been validated and the IgM and IgA responses to pneumovax have been determined in normal healthy subjects.
The development of both PCP IgM and IgA ELISAs has involved the identification, location and preparation of suitable materials. The optimisations of assay conditions and clinical validation have been achieved in collaboration with the study group. Both assays have been initiated as new products and entered into the New Product Development system at the Binding Site. This process involves full product specification and commercial justification established through the EUROPADnet group.
Working with the same team of experts we are currently evaluating the measurement of both the IgA and IgM responses to pneumovax using these ELISAs in broader and wider clinical applications such as in patients diagnosed with IgA deficiency and those with combined IgA/IgG subclass deficiencies.
Predicting the risk of infusion reactions to anti-IgA antibodies:
The occurrence of anti-IgA antibodies at high titre is a risk factor associated with anaphylactoid reaction in hypogammaglobulinemic patients. The presence or absence of anti-IgA antibodies is also considered important basic information in the European ESID database of primary immunodeficiencies. Screening for anti-IgA antibodies in patients lacking IgA may be helpful for determine the best therapeutic strategy, including the route of administration of substitution immunoglobulin therapy or the transfusion of IgA-free blood products. From a hematological point of view tested IgA-deficient blood components without anti-IgA antibodies are a precious resource that should be allocated with care and reserved for IgAD patients at risk of anaphylactoid reaction. The combined use of quantitative ELISA for detection of anti-IgA antibodies and the measurement of serum IgA concentration by nephelometry provides an effective and safe strategy for the diagnosis and prevention of IgA anaphylactic transfusion reactions. We found a significant amount of anti-IgA antibodies in 10 % of patients with the most common primary hypogammaglobulinemia (selective IgA deficiency and common variable immunodeficiencies) where the serum IgA level was lower than 0.05 g/L. Analysis of anti-IgA response in PAD patients and the assessment of biomarker predicting the risk of infusion reaction in PAD demonstrated that Common Variable Immunodeficiency (CVID) patients who are seropositive for anti-IgA antibodies have a predisposition for anaphylactoid reactions to intravenous immunoglobulin replacement therapy (IVIG). A new anti-IgA ELISA has been established www.biovendor.com .
Development of a combined clinical, phenotypic and functional classification of PAD:
This subproject represents the main translational research target of WP5 and might be completed in the following months. This is a final goal of WP5. Earlier than anticipated, and thanks to the sterling work of EUROPADnet, the new classification of PAD is already in progress. A proposed algorithm for CVID diagnosis is illustrated in figure 2 (attached)ow:
Major changes of the new proposal compared to the current criteria present the definition of absolute immunoglobulin levels as cut off, the definition of vaccine responses, a more precise definition of exclusion criteria especially defining patients with combined immunodeficiency.
Potential Impact:
The consortium has produced a large number of publications already, based on the research results generated to date. At present several publications are either at the manuscript stage or already submitted. These will focus on the achievements throughout the second reporting period and cover the whole project. Publications are listed in the respective section in the reporting
system. The grant number and Acronym have been used accordingly and the support of the EC will be mentioned in the future when publications rely on results generated within EURO-PADnet.
The partners have represented the project in numerous conferences and meetings.
CONSORTIUM MEETINGS
The regular consortium meetings have taken place as planned:
- 21 January 2008, London, Royal Free Hospital
- 15 October 2008, S'Hertogenbosch, Golden Tulip Hotel
- 3-4 April 2009, Rome, Universita Sapienza
- 12 September 2009, Berlin, Art'otel Berlin Mitte
- 30 April 2010, Università degli Studi di Brescia, Spedali Civili di Brescia, Italy
- 4. and 5. November 2010, John Radcliffe Hospital Oxford, UK
- 1 April 2011, Hopital Necker, Paris, France
REPORT FROM THE EDUCATIONAL COMMITTEE
At the beginning of the project the idea was developed that several consortium meetings should be complemented by a second day allocated to teaching activities. Therefore, teaching days were organised and opened for local and any other interested person in order to provide an educational opportunity for doctoral students and undergraduates in any participating centre (mostly taken up by the local host institute). Such sessions were held at three PADnet meetings in Rome, Berlin and Oxford.
In addition, doctoral students from participating centres were encouraged to give study updates. In Rome a specific mentor was identified beforehand to help each speaker and the talks were graded formatively to give feedback. This was successful and the same forms were used at the Oxford meeting. An example is given below: The Educational Afternoon Session offered a great opportunity for junior scientists from the EURO-PADnet partners and other juniors to hear educational talks from some of the consortium's senior scientists and PIs. Furthermore, invited speakers contributed to the success of the afternoon. After the plenary session, the participants had the opportunity to ask questions or discuss problems with the speakers.
Presentations covered the following fields:
- Proliferative capacity of B cells in antibody deficiency (Guest speaker Gertjan Driessen, Erasmus MC Rotterdam)
- Pitfalls using linkage analysis (Bodo Grimbacher, UCL)
- Defects in IL10 signalling cause inflammatory bowel disease in children (Erik Glocker, UCL)
- New assays
- Berne Ferry (UOXF-HX): Salmonella antibody testing
- Isabella Quinti (Sapienza): IgM pneumococcal assay
- Vojtech Thon (MU): Anti-IgA antibody assay
- Translation of useful assays to clinical use (Hermann Eibel, UKL-FR)
Many PADnet participants and their trainees attended the educational sessions as part of the biennial meeting of the European Society for ImmuneDeficiencies [ESID], in which there was a specific session for primary antibody deficiencies run by Helen Chapel and Gavin Spickett.
Attendance was also encouraged at specific educational sessions arranged within the ECI congress, Berlin in September 2009, hosted by Reinhold Schmidt (MHH) and arranged by Helen Chapel andB.Grimbacher. Specifically two symposia on immune deficiencies were organized and sponsored with the generous help of pharmaceutical industry (CSL-Behring and Baxter). These educational events were open for EURO-PADnet members as were numerous Satellite symposia; both educational events were very well attended by over 250 persons.
One of the educational highlights of this reporting period was the PID Winter-School organised with the help of the University College London in Windsor from 28th February to March 2nd at the Cumberland Lodge.
The aim of this school was i) to educate students in the field ii) to transmit latest research data from EURO PADnet projects to the field iii) to get to know excellent young hopeful students in the field for further career support and attachment to the field of immune deficiencies.
37 students from 7 European countries (UK, Italy, Spain, Portugal, Belgium, Netherlands and Switzerland) were accepted based on a personal application with support letter and including an abstract for presentation and a personal career statement. Active participation of the students was encouraged and also well received (see evaluation attached). Faculty included approx. 20 lecturers, including Profs Chapel, Grimbacher, Gaspar and Thrasher from EURO PADnet (agenda attached). EURO PADnet provided approximately one third of the funding for this event.
REPORT FROM THE ETHICS COMMITTEE
Relevant ethical approvals as well as a declaration from each Principal Investigator, stating that the ethical approvals cover the entirework on the project and the entire period of the project have been submitted to the EC. Licences for animal experimentation have been included. The beneficiaries VUMC and TBS only work with samples that are ethically approved by the other partner institutions and beneficiary UOXF-HX stated that the work they reported in the first half of the project did not require ethical approval. The consenting process for the ESID and UKPID registry projects (WP1) is well-established with approved Patient Information Sheets and Consent Forms for adults, young adults and parents of children. In the UK a national version of the ESID registry is in use on local servers. The patient is therefore given the additional choice, whether he would like his data to remain on the UK registry only or whether he wants it to be included in the ESIID registry as well, which then takes place by an electronic import. This mechanism has been ethically approved in the UK and is running smoothly.
List of Websites:
The project website www.europadnet.eu is online and contains general information about the project and can also be accessed by the partners via a secure login to allow them to exchange information on a confidential level. Some results have been made public on the website and final results will be posted after the submission of the final report. We also have implemented a teaching website www.ipidnet.org for linking the clinical and cellular phenotyping, and a www.chest-ct-group.eu for the analysis of lung complications in antibody deficiencies.
EURO PADnet has set out to better understand primary (inborn) deficiencies of antibody production (Primary Antibody Deficiencies; PAD). We documented the exact phenotype of more than 4000 patients into a European web-based database and offered patients participation in approved research protocols. Genetic causes for PADs were sought and more than four novel monogenetic defects have been identified. Moreover, genetic risk factors for becoming antibody-deficient have been identified and better described. We set up in vitro experiments and murine models of the diseases to better understand their pathophysiology. First successful steps towards the genetic correction of the first antibody deficiency called Bruton's disease - which actually was the first primary immune deficiency to be described back in 1952 - have been conducted. Not only the role of immunoglobulin G (IgG), but also the role of IgA and IgM and their memory cells has been studied and their importance in the host defence has been highlighted. The consortium has initiated and accompanied investigator initiated observational trials and the results of these have impacted local, national and international policy making for the management and the understanding of antibody deficiencies (for examples please see www.chest-ct-group.eu or www.springerlink.com/content/l4065729w12r1114/).
During the first three years after the start of this collaborative EU project, we have published in Journals such as The Lancet and Nature Immunology and a total of more than 356 IMPACT FACOR points have been collected.
Most importantly, the consortium runs three websites and has progressed two diagnostic tests into the clinical setting: The project website www.europadnet.eu contains general information about the project and can also be accessed by the partners via a secure login to allow them to exchange information on a confidential level. We also have implemented a teaching website www.ipidnet.org for linking the clinical and cellular phenotyping, and a clinical support website www.chest-ct-group.eu for the analysis of lung complications in antibody deficiencies. Moreover, we have held educational symposia and a Winter School for physicians and researchers in the field.
Within the EURO PADnet consortium, The Binding Site (TBS; www.bindingsite.com) has facilitated the development of both Pneumococcus-specific polysaccaride (PCP) IgM and IgA ELISAs. The assays have been validated and the IgM and IgA responses to Pneumovax have been determined in normal healthy subjects. The development of both PCP IgM and IgA ELISAs has involved the identification, location and preparation of suitable materials. The optimisations of assay conditions and clinical validation have been achieved. Both assays have been initiated as new products and entered into the New Product Development system at The Binding Site (TBS). This process involves full product specification and commercial justification which was established through the EUROPADnet group.
Patients with antibody deficiency often require IgG antibody replacement made from human plasma from healthy donors. This product contains traces of IgA. Some patients with antibody deficiency unfortunately have auto-antibodies against IgA. These often lead to the intolerance of the IgG replacement product and may end up in an anaphylactic reaction. To screen for these in order to assess the risk for possible adverse side effects, an anti-IgA ELISA has been established by EURO PADnet collaborators which is now marketed by www.biovendor.com
Taken together, the EURO PADnet consortium progressed the field of primary antibody deficiency research in Europe in an unparalleled way in just 3 years.
Project Context and Objectives:
This medium-sized collaborative project will document and monitor the natural course and study the pathophysiology of Primary Antibody Deficiencies (PAD). PADs are rare inborn errors of the immune system with an estimated incidence of <1:25.000. The defective immunity in patients with PAD causes an increased susceptibility to recurrent infections of the respiratory- and gastro-intestinal tract as well as ill defined co-morbidity including granulomatous disease, lymphocytic organ infiltration and (paradoxically) autoimmunity.
Our consortium cares for more than 1000 PAD patients, representing approx. 50% of the patients registered in Europe, many of them being children. This project will:
1. Combine clinical and research data in a central online registry, complemented by a sample repository (WP1)
2. Elucidate the genetic cause of PADs by linkage analysis and candidate gene approaches (WP2)
3. Establish in vitro models for B cell differentiation steps that are defective in PADs such as in vitro class switch recombination, and siRNA knockdown models for PAD screening (WP3)
4. Create mouse models of PAD by using several technology platforms including knock-in and knock-out mice, shRNA mediated gene knock-down and humanised mice (WP4)
5. Perform translational research by taking the observations from the patients into the basic research projects and transfer these results back to the patients (WP5).
The study of these immunodeficiencies represent an "experiment of nature", uniquely enabling researchers to study the detrimental effects of mutations in specific genes involved in the immune system. Basic research results are translated into the development of new technologies, new knowledge, and new therapeutic tools. The clinical results will improve the diagnosis, management and quality of life of PAD patients, leading to future developments of new tests, preventive measures and treatment protocols.
Introduction to WP1: The Natural Course and Phenotype of PAD
In the last framework programme, the coordinator of this FP7 project, Prof. Grimbacher headed the development and the establishment of a clinical online patient registry. After 4 years already more than 9.000 patients with primary immunodeficiency diseases were registered into the ESID online database, (59%) of which belonging to the group of PAD (www.esid.org/esid_registry.php). This project will be taken further to create a platform for the exchange of clinical information and research data complemented by quality of life data from the actual patients. To this end, the database shall be complemented by a sample bank.
Projects 1-3: Clinical Registry, Research Databases and Repositories
The ESID Online Registry (www.esid.org/esid_registry.php) is capable of registering clinical and laboratory data on all 206 primary immunodeficiency diseases currently listed in the IUIS publication (1). The particular strength of this database is that follow-up data is being collected on a yearly basis. The data currently being collected however only comprises a common dataset of 22 fields which are considered to give the basic information needed for assessment of an individual patient. In this WP we will extend the datasets of each PAD to facilitate future observational trials.
Project 4: Multiple breath washout for refining lung function testing.
Lung disease is one of the most frequent complications and causes of death of the patients with a primary antibody deficiency syndrome (PAD). Longitudinal data suggest that lung disease starts during childhood and leads to progressive structural lung changes as demonstrated by computed tomography (CT) of the chest. The goals of project 4 are i) to assess lung pathology of PID patients with spirometry and computed chest tomography and ii) to determine the sensitivity and specificity of Multiple Breath Washout (MBW) technique as an alternative for detection of early lung disease in paediatric PAD patients.
Introduction to WP2: The Genetic Etiology of PADs
Positional cloning of disease causing genes has proven to be very successful in the identification of many defects in primary immunodeficiency diseases. This consortium has collected more than 150 families with PAD including informative consanguineous families, and has recently published four candidate loci linkage papers identifying genetic susceptibility loci on chromosomes 4q, 5p, 6p, and 16q, However, the disease causing genes still remain elusive. Moreover, the consortium has recently collected 300 patients with sporadic CVID and 300 patients with selective IgAD, in whom we aim to search for the aetiology of these diseases by molecular/genetic analysis in order to define the underlying molecular/genetic defect(s).
Project 1: Positional Cloning and Linkage studies
Common Variable Immunodeficiency (CVID) is a primary antibody immunodeficiency characterized by low immunoglobulin levels and low or normal B cell numbers. The genetic cause of this disease is unknown for the majority of cases. CVID may occur in an autosomal dominant (AD-CVID) trait or sporadically, but autosomal recessive (AR-CVID) families are also reported.
Project 2: Genetic association studies
The future direction of this project will be to recruit additional patients to run GWS on a sufficiently large replication cohort, aiming to identify both MHC and non-MHC susceptibility loci in patients with IgAD. Furthermore, we will try to identify multicase CVID families for mapping of susceptibility genes.
Project 3: The Candidate Gene Approach
Functional candidate genes for antibody deficiencies will be sought and evaluated on genetics and functional grounds.
Introduction to WP3: In Vitro Models of PAD
The aim of this WP is to develop new in vitro models for PAD study. Each partner has his own approach but all are complementary and converge to the same objective, i.e. the molecular characterization of PAD. By a close collaboration between all groups, 3 main objectives will be reached.
A.Better characterization of PAD, including phenotypic and functional B cell studies.
B.Validation of gene mutations
C.In vitro search for new genetic defects
Project 1: Human TACI deficiency - a T cell independent CSR defect leading to CVID?
TACI C104R accounts for about 50 % of all TACI-mutated alleles in CVID patients. One of the aims of U. Salzer's group in this study is to use existing monoclonal or polyclonal antibodies or generate these reagents to analyse the mutated C104R protein.
Project 2: Class switch recombination Defects.
Study of normal antibody maturation processes in normal and pathological conditions.
Immunoglobulin class switch recombination (CSR)-deficiencies are a rare (1/200,000 births) inherited disease characterized by increased (or normal) IgM serum levels contrasting with a strong decrease or an absence of other isotypes (IgG, IgA). According to the molecular defect, this Ig-CSR deficiency is associated or not with a defective generation of somatic hypermutations (SHM). Several molecular defects underlie this condition. We have previously contributed to the description of CD40-ligand and CD40-deficiencies which both lead to a severe cellular and humoral immunodeficiency.
Project 3: Plasma cell defects in PAD
Plasma cell development and potential defects in functions required to generate plasma cells were studied in Freiburg and in Brescia through the collaboration between the groups of A. Plebani and H. Eibel.
Project 4: CD21low B cell populations in primary human antibody deficiencies
The expansion of specific B cell subsets expressing low levels of CD21 in patients with PAD is an expression of an underlying pathogenetic dysregulation. These B cell subsets will be used to define subgroups of patients for specific functional as well as molecular characterization. Circulating B cell populations with a low expression of CD21 contain transitional B cells, activated CD21low B cells and possibly plasmablasts.
Project 5: shRNA-based knockdown of candidate genes
In order to identify transcripts affected by Btk deficiency, E. Smith's group has performed several experiments. First, they have created a new database for immune-related genes as an important tool for screening targets for shRNA-based knockdown. This database contains 1982 immune genes and combines information from 3 already existing databases: Immunogenetic Related Information Source (IRIS) - a database surveying known human immune system genes, Immunome Knowledge Base (IKB) - an integrated service for immunome research, and Gene Ontology assigned 'immune response' genes.
Introduction to WP4: Mouse Models of PAD
Animal studies make it possible to analyze several aspects of PAD, which cannot be addressed in humans. Although in vitro systems have undisputable advantages (e.g. costs, no use of animal experiments, high through-put) and go some way addressing the mechanisms of disease pathogenesis, in vitro systems lack the complexities of an in vivo system. We will generate and study a variety of murine models that will allow a greater understanding of human antibody deficiency syndromes.
The following examples will directly address key aspects of PAD in animal models:
1.When the molecular basis is not understood, such as in patients with heterozygous TACI mutations (CVID) or XLP, the generation of knock-in and knockout models will provide new insights.
2. When the defect is known, but when the precise mechanism is not understood, the generation of cell-specific in vivo knockdowns, such as Btk-knockdown in B cells will elucidate cell specific functions of a given gene product.
3. When experimental in vitro data suggest novel genetic disease mechanisms, the generation of the corresponding knock-in mouse models (membrane IgG mutations and docking molecules) will test these hypotheses.
4. Developing "humanised mice" will provide us with a model system allowing us to test the capabilities of the transferred hematopoietic cells in any given patient (e.g. for patients with CVID).
The projects are based upon the study of the following diseases: CVID, XLA, AR agamma, XLP, PAD of unknown cause The animal studies will be based upon the following technology platforms: Knockout, knock-in, shRNA (lentiviral), dominant negative (lentivir), humanized mouse
Project 1: Knock-in model of TACI deficiency in CVID
The aim of this work package was to generate a murine model of the C104R mutation and to determine whether this resulted in a murine humoral deficiency. After the successful generation of the respective mouse mutant C76R, we were able to undertake some phenotypic analysis of the mutant. In summary, our data suggest that the C76R homozygous mutant mice do show humoral abnormalities that are similar to human TACI null mutants with a hyper-expansive B cell phenotype and defects in immunoglobulin production. To date no major differences have been observed in heterozygous mutants. Further analysis including immunisation experiments will be undertaken in the latter half of the project.
Project 2: Knockout model of SAP deficiency in X-linked lymphoproliferative disease (XLP)
XLP is a complex immunodeficiency in which mutations in the gene encoding the adaptor protein SAP (SLAM associated protein) lead to severe immunodysregulatory phenomena [1]. Boys with XLP present typically but not always after EBV infection with fulminant infectious mononucleosis, B cell lymphoma or hypogammaglobulinaemia which is highly reminiscent of a CVID phenotype. The pathogenesis of hypogammaglobulinaemia in XLP is unclear and has been until very recently been thought to arise due to a lack of T cell help.
Project 3: Knock-in model for antigen receptor signalling
Establishment of IgG-switched memory B cells and their costimulation-independent activation during secondary immune responses is thought to be dependent on isotype-specific antigen receptor signals. Since the start of the project we have made significant progress in the identification of intracellular signalling proteins that are recruited to tyrosine-phosphorylated cytoplasmic domains of mIgG and mIgE. The first interaction partner that was detected by co-immunoprecipitation and confirmed by Western blot, the adaptor molecule Grb2, could already be published in Nature Immunolgy in September 2009 (Recruitment of the cytoplasmic adaptor Grb2 to surface IgG and IgE provides antigen receptor-intrinsic costimulation to class-switched B cells. Engels N, König LM, Heemann C, Lutz J, Tsubata T, Griep S, Schrader V, Wienands J., Nat Immunol. 2009 Sep;10(9):1018-25.). Based on these findings functional studies of Grb2 and its potential binding partners are in progress, which include the generation and analysis of Grb2 mutants that lack either the N- or C-terminal SH3 domain or carry amino acid substitutions in the binding sites for know interaction partners. At the same time we are trying to generate of Grb2-deficient cell lines.
Project 5: Development of knockdowns and humanized mice
Development of human B cells in reconstituted double knockout Rag2/Il2rg deficient mice
Early steps in human B cell development were analysed in humanized mice. Human B cells developing from engrafted CD34+ hematopoietic stem cells were found in bone marrow and blood but rarely in the spleen or in other secondary lymphoid organs. To improve the development of human lymphocytes in a murine cytokine environment Rag2/Il2rg mice transgenic for human IL7 were generated. To facilitate the construction of mice expressing human cytokines and growth factors supporting the development of human hematopoetic cells we also started to generate ES cell lines on the Rag2/Il2rg background. In addition, CD34+ HSC were engrafted into Rag2/Il2rg transgenic for human BAFF. The analysis of these mice is under way.
Project 6: Humanized mice to study memory and plasma cells in CVID
IgM memory B cells play a fundamental role in the defence against encapsulated bacteria producing anti-bacterial IgM antibodies [1-3]. These bind to the capsular polysaccharides of rendering them targets of monocyte-macrophage cells.
Therefore, IgM antibodies are indispensable for the removal of bacteria circulating in the blood, collected by the filtering system of the spleen and invading tissues.
Antibodies, however, also play an important role before tissue invasion, forming a protective layer on mucosal surfaces. Bronchial and intestinal epithelia are both covered by a film of antibodies of IgA isotype. Secretory IgA (sIgA) is secreted by mucosal plasma cells and then transported from the basal to the apical side of epithelial cells by the poly-Ig receptor. In the mouse most of the IgA at epithelial surfaces is secreted in a T-independent manner and originate from a defined type of B cells, the B-1a. This population is ontogenetically and functionally distinct from conventional B-2 B cells [4].
In humans the important question of the origin of mucosal plasma cells has never been addressed.
Introduction to WP5: Translational Research
Project 1 is to improve the quantity and quality of immune-phenotypic data of peripheral lymphocyte subsets from PAD patients. Currently, the flow cytometric analysis is performed by different laboratories using different panels of monoclonal antibodies. A standardized assessment is underway.
Results of this subproject can be viewed on the new website www.ipidnet.org.
Project 2 is to define genetic variations in PAD patients with agammaglobulinemia and low B cell counts who do not have a definitive genetic/molecular diagnosis by the analysis of specific points of arrest of B cell differentiation in the bone marrow.
We have undertaken bone marrow biopsy and FACS immunophenotyping in patients with CVID in order to assess if the clinical and immunological phenotype correlates with a specific developmental block/alteration in the early stages of B cell development. Our preliminary data are very interesting and promising: early B cell development in CVID is disturbed at different stages in a large percentage of CVID patients and apparently more patients share a similar developmental defect.
Project 3 is to analyse the role of the different B lymphocyte subpopulation in the protection from infections by encapsulated bacteria. We showed that IgM anti PnPS antibodies in PAD correlate with the capacity to differentiate into plasmablasts. We also demonstrated that a reduced proliferative response to CpG correlates with an increased lung infection rate. Due to the impaired proliferative response, CpG induced plasmablast differentiation is decreased in patients with CD21low (>10%).
Project 4 is to define quantitative and qualitative antibody defects in PAD patients.
Project 5 is to identify biomarkers to predict the risk of infusion reaction in PAD patients. Anti-IgA antibodies have been proposed as predictive factors for the occurrence of adverse reactions to IgG replacement therapy which contains traces of IgA in PAD patients, but their pathogenetic role and their isotype (IgG vs IgE) is still unclear.
Project Results:
THE NATURAL COURSE AND PHENOTYPE OF PADs
The objective for Workpackage 1 was to better describe the natural phenotype of primary antibody deficiencies. To this end, we used the existing database of the European Society for Primary Immune Deficiencies (ESID), which has been developed with help of the European Framework 6 Program and commercial funding during the years 2004 to 2006, and local pre-existing national registries such as the Italian IPInet registry. During the funding period of EURO-PADnet, we increased the total number of PAD patients in the ESID Registry from 3,369 in May 2008 to 5,554 on August 2009 by 2185 patients. This has further increased to 7,466 patients as of April 2011. The total increase over the three-year period hence was 4,099 patients, or 122%.
We aimed at better describing the T and B cell phenotype in antibody deficiencies, to identify pathogenic and clinically relevant patterns of dysregulation, and to improve current classifications schemes based on a large European cohort of patients with PAD. Data of 411 patients have been collected from 6 centers and are currently evaluated for the association of changes in T and B cell phenotype with the clinical manifestation of splenomegaly, lymphadenopathy, autoimmune disease, granulomatous disease, opportunistic infection, conventional infection of the respiratory and gastrointestinal tract, bronchiectasis and interstitial lung disease. Statistical evaluation is under way. Data is posted on www.ipidnet.org. As an example shall serve the following published study: We were interested in the question of whether the congenital lack of B cells actually had any influence on the development of the T cell compartment in patients with agammaglobulinaemia. The CD4 T cell memory compartment was reduced in patients with XLA of all ages. This T cell subset encompasses both CD4(+) CD45RO(+) and CD4(+) CD45RO(+) CXCR5(+) cells. This observation was confirmed in patients with CVID who had <2% B cells, suggesting that not the lack of Bruton's tyrosine kinase but the lack of B cells is most probably the cause of the impaired CD4 T cell maturation. We postulate that this defect is a correlate of the observed paucity of germinal centres in XLA. Our results support the importance of the interplay between B and T cells in the germinal centre for the activation of CD4 T cells in humans.
EURO-PADnet researchers also accompanied a prospective clinical study on the natural course of antibody deficiencies over a cumulative follow-up period of 1,365 patient-years. This study was carried out by centres in Italy. It revealed that for X-linked agammaglobulinemia, the only co-morbidity risk factor for pneumonia was the presence of bronchiectasis. In common variable immunodeficiency (CVID), the data allowed us to identify a clinical phenotype characterised by a high pneumonia risk: patients with low IgG and IgA levels at diagnosis; patients who had IgA level <7 mg/dL and who had bronchiectasis. Patients with pneumonia did not have significant lower IgG trough levels than patients without pneumonia, with the exception of patients whose IgG trough levels were persistently <400 mg/dL.
Based on a cohort of paediatric (n = 232) and adult (n = 73) patients (mean age 36.5 SD 17.7 range 1.6 to 79.3 years) with CVID who underwent chest CT, the overall prevalence of structural bronchial pathology was 72%, with bronchiectasis present in 63%. Parenchymal pathology, such as lines, bands and nodules, was detected in 76%. A pathological lung function as indicated by a forced expiratory volume in 1 second (FEV1) of < 80% predicted was present in 39.5% of the patients who underwent lung function analysis (n=185). Importantly, 55% of the patients with a normal lung function had bronchiectasis. Air trapping was evident in up to 30%.
The consortium established an international, multidisciplinary expert group, the "Chest CT in ADS Group", with 26 immunologists, 13 radiologists, and 7 pulmonologists from 20 academic centres from 9 countries, www.chest-ct-group.eu. The Chest CT in ADS Group plans to hold a European meeting consensus conference on the use of chest CT scans in primary antibody deficiencies by January, 2012.
To link the data in the clinical databases, our centres collected biological samples of the documented patients for research. The centre in Brescia has collected 150 PID patients, 150 SIgA patients, and 15 PID families; the Karolinska Institute has collected 250 CVID samples, 1200 sIgAD samples and 100 samples with other PIDs; INSERM in Paris has collected 98 samples with class switch recombination defects. Rome has collected approximately 1,450 samples, Freiburg about 660; and London more than 864 samples. These sample collections provided a huge resource for future projects and collaborations and put each of the centres into a unique position. This collection of samples in unprecedented in the world.
THE GENETIC ETIOLOGY OF PADs
Linkage analysis in antibody-deficient families:
Linkage analysis is a non-hypothesis-driven pure positional cloning approach. By linkage analysis, the consortium was successful in identifying a novel genetic cause for CVID in four autosomal recessive families from Belgium, Israel and Iran. The causative gene was previously not known to be important for the immune system and most likely affects autophagy of lymphocytes. What is especially interesting for the pathophysiological understanding is the fact that the phenotype in these patients include autoimmunity. This paper is currently under revision.
In the other CVID families we located the genetic defect to certain chromosomal regions and positional candidate genes were sequenced. The autosomal-dominant CVID family FR1 is linked to chromosomes 5, 6, and 7. The autosomal-dominant CVID family NL1 is linked to chromosome 4. The autosomal-dominant CVID family FR15 is linked to chromosome 3. The autosomal-recessive CVID family US2 is linked to chromosomes 1, 5, and 20. The autosomal-recessive CVID families Port1 and cv32 have multiple regions of homozygosity as they are quite small. Finally, the autosomal-dominant CVID family IT4 is linked to chromosome 4 and achieved a LOD score of 3.9.
However, as this purely positional approach followed by Sanger sequencing of possible candidates in the linkage regions was not fruitful in these families for the last few months. Hence we are now employing whole exome sequencing to identify the genetic defect. This will be carried out in part by the EU project EURO GENEscan where UCL and KI are partners.
Genome-wide association studies:
Tissue typing on Iranian and Swedish CVID patients/families reveals that the previously suggested HLA association in CVID/IgAD multicase families is largely restricted to those carrying the HLA B8, DR3, DQ2 haplotype and that the pattern in "sporadic" CVID patients is largely random. The addition of 100 more cases confirmed the initial findings.
Analysis of the contribution of genes within the HLA region in 100 multicase IgAD families is ongoing at KI (analysis of degree of HLA sharing among affected/non-affected siblings of the proband). This material is also currently being extended, including additional multicase families. A preliminary analysis, using the formula of Risch, has suggested that genes within the MHC region contribute approximately 40% of the disease risk in IgAD.
The candidate gene approach:
In an exceptional effort, fourteen functional candidate genes for CVID were studied by a single team in Brescia, Italy; some of which may represent disease-causing genes, others may be modifier genes for CVID.
The genetics of hyper-IgM syndromes:
Whole exome sequencing in five patients from different families with hyper IgM syndromes of unknown cause was performed. In two siblings we found a mutation in the CD40L gene, although CD40L-deficiency seemed previously excluded by normal CD40L expression. This observation highlights the fact that protein detection alone does not exclude a mutation. In two other families, two genes were found mutated on both alleles. These genes encode proteins involved in both chromatin modification and DNA repair. These mutations are found "likely damaging" in Polyphen data base and are not described as polymorphisms in data bases. No murine model is available. Mutations in both genes are now looked for in a large cohort of HIGM patients (80 patients). To validate the role of these molecules in CSR, shRNA lentiviral transduction of CSR-induced CH12 cell line is now on-going. In another non consanguineous family, with two affected children, we found in both patients a few genes carrying bi-allelic mutations, that are now under investigation. Five other informative families have been studied by exome sequencing and results are pending. In a consanguineous and informative family, in which a homozygous genetic region has been already determined, a gene was found mutated as homozygous in the two affected children. However, a normal sibling and the healthy mother present the same genotype, excluding the role of this mutation in the pathogeny of the disease. Consequently, all the exons present in this region, not completely or not perfectly covered by the exome, were sequenced, with no detected abnormalities. Expression the cDNA level of all the genes comprised in that region will be assessed as soon as material is available (EBV B cell line) in order to detect splice defects (due to intronic mutations) or defective expression (mutations in promotor).
IN VITRO MODELS OF PAD
BAFF as the survival factor for peripheral B cells:
Interaction of the ligand BAFF and its receptor (BAFFR, which is exclusively expressed on B cells) is one of the most important survival signals, if not the most important survival signal to B cells. As antibody deficiencies often lack of sufficient B cell numbers in the periphery, and also B cell differentiation defects have been demonstrated in patients with antibody deficiencies (B cell differentiation is tightly linked to proliferation during ongoing immune responses), the study of this ligand-receptor pair was a paramount interest.
We analyzed BAFF serum levels in CVID and BTK-deficient patients, 2 patients. Highest BAFF serum concentrations were found in BTK patients (10 - 100 ng/ml) followed by a group of 21 CVID patients with < 2% of B cells and > 10 ng/ml BAFF. The analysis revealed that BAFF serum concentrations correlate inversely with B cell numbers but not with autoimmunity or B cell neoplasia. Sequencing of the BAFF genes in selected patients did not reveal mutations. BAFF-R expression was studied by FACS in > 80 patients. About 15% expressed significantly reduced BAFF-R levels. Expression went back to normal in EBV lines indicating regulatory changes in CVID patients.
We developed several readout systems for BAFF-R function including analysis of the alternative NF-kB activation and p100 processing by western blotting and in vitro survival assays for human primary B cells and for murine B cells from BAFF-R deficient mice which were transduced with various BAFF-R mutants. Analysis of BAFF-R receptor structure revealed that BAFF-R associates in multimeric receptor complexes. Different from wildtype, the BAFF-R variant P21R shows impaired complex formation, ligand binding and NF-kB activation. Presently we are studying the impacts of BAFF-R multimerization on the association with downstream signalling components such as TRAF6, TRAF2 and TRAF3.
CD21low B Cell Populations In Primary Human Antibody Deficiencies:
The so called CD21-low B cells population has been primarily described in patients with CVID, but has recently also been shown to be increased in cohorts of autoimmune diseases. Interestingly, CVID patients also suffer from autoimmune phenomena. CD21(low) B cells were shown to be profoundly anergic, and defects of BCR-mediated calcium signaling and of T cells have been described in CVID. We found that also the classical naïve B cells from many CVID patients proliferated poorly. The B cells of some CVID patients had a reduced capacity to divide reminiscent of the proliferative arrest associated with replicative senescence which is known to be influenced by telomere length. Thus, we investigated whether lymphocyte dysfunction in CVID was related to telomere-dependent replicative senescence, and found that both the B and the T cells from some CVID patients had significantly shorter telomeres compared with B and T cells from other CVID patients. Telomere lengths in B and T cells were significantly correlated, indicating that the rate of telomere attrition in lymphocytes is an individual characteristic of CVID patients. Our findings suggest that telomere-dependent replicative senescence contributes to the immune dysfunction of some CVID patients, and may provide an important clue for a better understanding of the pathogenesis of CVID.
IFN Signature in CD21-low B cells:
The percentage of CD21low B cells was shown to be significantly expanded in HCV-viremic patients supporting the idea of an association of CD21low B cells to IFN rich environment. Detailed examination of IFN signature in CD21low B cells of CVID patients demonstrated increased expression of MXA, OAS1, while other targets like IFIT1 showed higher expression in naïve B cells of patients. The IFN signature was B cell specific in CVID, but IFN itself failed to induce the phenotype of CD21low B cells in vitro.
Plasma cells in antibody deficiencies:
Our research activities focused on the role of epigenetic and protein modifying histone acetylase / deacetylase activities during plasma cell development. We found that inhibition of HDAC by valproic acid and other iHDAC blocks IL21R signaling by affecting STAT5 phosphorylation The VPA-mediated inhibition of IL21R signalling impairs plasma cell formation and antibody secretion.
IgM memory B cells in antibody deficiencies:
In order to demonstrate the developmental connection between IgM memory and plasma cells, we first determined the presence of sIgA and/or IgM in jejunal biopsies of two groups of CVID patients, with or without circulating IgM memory B cells and in splenectomised patients. By confocal microscopy we analyzed IgA plasma cells and sIgA on the epithelial cells in jejunal biopsies. After splenectomy the reduction of memory B cells in the peripheral blood is associated to a significant diminution of IgA plasma cells and the disruption of the film of sIgA on the luminal side of epithelial cells. In CVID the absence of sIgA in the gut and IgA in the serum are associated to a significant reduction of memory B cells in the peripheral blood and high risk of respiratory infections. CVID patients with sIgA at mucosal sites had a normal number of circulating memory B cells, detectable serum IgA and have a low risk of respiratory infections. Memory B cells are indispensable for the production of sIgA at mucosal sites. The disruption of the sIgA film in the gut impairs the local defense against invading pathogens.
Identifying precursors of IgM memory B cells:
The expansion of specific B cell subsets expressing low levels of CD21 in patients with PAD is an expression of an underlying pathogenic dysregulation. These B cell subsets are being used to define subgroups of patients for specific functional as well as molecular characterization. We demonstrated that, upon exposure to CpG, a fraction (26%) of transitional B cells proliferate (P) and differentiate either to IgM-secreting plasma cells (19.2%) or to cells phenotypically identical to IgM memory B cells (41.3%). The remaining non-proliferating (NP) B cells down-regulate the expression of CD24 and CD38, thus resembling mature B cells. We have also demonstrated that natural antibodies, including immunoglobulins with anti-pneumoccocal specificity, can be generated "in vitro" from CpG-stimulated transitional B cells. The progenitors of IgM memory B cells are recent bone marrow emigrant transitional B cells, which through the blood arrive to the spleen, an optimal anatomical site where blood-borne pathogens and cellular debris are filtered by resident macrophages. Here, transitional B cells diversify their antigen receptors in the presence of bacterial components, e.g. CpG DNA. CpG induces somatic mutations in the Ig heavy chains (VH) regions in P B cells, as opposed to the prevalence of germline sequences in the NP cells. The mutations are confined almost exclusively to the framework regions. Moreover, the incidence of mutations is VH-specific, being VH1 and VH4 the most mutated, and VH5 the least mutated family. In summary, early recognition of bacterial DNA induces proliferation, SHM, antibody secretion or survival in transitional B cells. IgM memory B cells in one year-old children mirror the mutation pattern seen in CpG-stimulated transitional B cells.
The role of TACI mutations for antibody deficiencies:
Coding variants in TACI, aka tumor necrosis factor receptor superfamily member 13B (TNFRSF13B) have been implicated in common variable immunodeficiency (CVID), but the functional effects of such mutations in relation to the development of the disease have not been entirely established. To examine the potential contribution of TNFRSF13B variants to CVID, we studied the mutant alleles in vitro. In summary we found that each mutation acts via a different pathomechanism: I. frameshift mutations, nonsense or missense, leading to loss of protein expression (examples ins204A, S194X, L171R); II. expressed missense mutant that fails to bind ligand (example C104R, I87N); III. mutant with intracellular retention of mutated protein (examples C104R and L171R); IV. failure of trimeric assembly as proposed by Lee JJ Blood. 2009;114:2254-62 (example A181E); V. mutants with no effect at all = non pathogenic variants (examples W40R, H81N, K188M).
Most importantly, we were able to show that having a pathogenic mutation in TACI increases the risk for developing CVID by 4. Presence of a heterozygous mutation was associated with antibody deficiency (P <.001 , relative risk 3.6). Heterozygosity for the most common mutation, C104R, was associated with disease (P < .001, relative risk 4.2). Furthermore, heterozygosity for C104R was associated with low numbers of IgD-CD27+ B cells (P = .019), benign lymphoproliferation (P < .001), and autoimmune complications (P = .001). These associations indicate that C104R heterozygosity increases the risk for common variable immunodeficiency disorders and influences clinical presentation. (Salzer et al., Blood. 2009; 113:1967-1976).
Class-switch Recombination defects and Mismatch Repair defects:
This consortium described PMS2-deficiency as causative of a partial Class-Switch-Recombination (CSR) defect. The group went on studying other MisMatch-Repair-Deficiencies. Seven further PMS2-, two MLH1- and seven MSH6-deficient patients could be diagnosed. All of the patients presented with a partial CSR defect. Interestingly enough, the group provided evidence that PMS2 is involved in the CSR-induced generation of DNA double strand breaks (DSB) in switch regions, likely because of its endonuclease domain.
The group observed that both double-strand-break (DSB) generation and repair is impaired in PMS2-deficiency, we asked the question whether impaired DSB generation could influence their repair. For this purpose, we studied S?-S? junctions in B cells in which DSB occurrence is very likely (AID+/-AID Cter?/+) or certainly impaired (UNG-/-). We showed a progressive bias towards microhomology usage for switch junction repair from AID+/- to AID Cter?/+ and to UNG-/-. Our results strongly suggest for the first time that proper induction of CSR-DSB either recruits NHEJ, or excludes other repair pathways (Alternative End Joining).
In other hyper-IgM patients, we identified a transcription elongation factor, the expression of which was strongly diminished in EBV B cell lines from four patients. Although the RNA level and the genomic sequence was found normal, this molecule appears indirectly down-regulated in these patients. The role of this molecule in CSR is being studied by shRNA in CSR-induced CH12 cell line.
MOUSE MODELS FOR PAD
TACI-knock-in mice as a model for antibody deficiencies:
Mutations in TNFRSF13B, the gene encoding transmembrane activator and calcium modulator cyclophilin ligand interactor (TACI), are found in 10% of patients with common variable immunodeficiency. However, the most commonly detected mutation is the heterozygous change C104R, which is also found in 0.5% to 1% of healthy subjects. The contribution of the C104R mutation to the B-cell defects observed in patients with common variable immunodeficiency therefore remains unclear. We sought to define the functional consequences of the C104R mutation on B-cell function. A knock-in mouse with the equivalent mutation murine TACI (mTACI) C76R was generated as a physiologically relevant model of human disease. We examined homozygous and heterozygous C76R mutant mice alongside wild-type littermates and studied specific B-cell lineages and antibody responses to T cell-independent and T cell-dependent challenge. Mice heterozygous and homozygous for mTACI C76R exhibit significant B-cell dysfunction with splenomegaly, marginal zone B-cell expansion, diminished immunoglobulin production and serological responses to T cell-independent antigen, and abnormal immunoglobulin synthesis. These data show that the C104R mutation and its murine equivalent, C76R, can significantly disrupt TACI function, probably through haploinsufficiency. Furthermore, the heterozygous C76R mutation alone is sufficient to disturb B-cell function with lymphoproliferation and immunoglobulin production defects.
Overall we have been able to show significant abnormalities of B cell homeostasis and humoral function in TACI C76R deficient mice. These data clearly show that the C76R mutation alone in mice has a detrimental effect on B cell function and if extrapolated to the human setting suggests that TACI C104R affects human B cells and contributes to the B cell defects seen in CVID.
SAP knockout mice as a model for hypogammaglobulinaemia in XLP:
Vaccination studies in SAP knockout mice suggest that responses to T cell dependent antigens are severely compromised in SAP deficient mice and further support the argument that T cell help is defective in SAP deficiency rather then there being an intrinsic B cell abnormality. We then investigated whether humoral function could be restored either by standard bone marrow transplant or through gene modification of murine stem cells. SAP knockout mice were subject to lethal irradiated and then reconstituted with either BM from wild type mice or from SAP KO lin-ve cells transduced with lentiviral vectors encoding SAP or a reporter vector as a control. Mice were then sacrificed at 12 weeks and were analysed from total IgG, IGg subclasses and for reconstitution of cytokine production in splenic CD4 T cells.
Naïve total IgG levels are significantly lower in SAP KO mice than in wild type mice. Following BMT from wild type mice and gene transfer using SAP encoding vectors we saw an increase in total IgG levels that was statistically significant. No increase was seen following gene transfer from a reporter gene encoding vector. Similarly both BMT and SAP gene transfer both resulted in increase in levels of IgG1, IgG2a and IgG2b whereas no increase was seen following reporter gene transfer. To look at whether this increase may have been due to restoration of T cell help, splenic CD4 cells from the different groups of mice were either unstimulated or stimulated with an anti-CD3 antibody and analysed for IL-4 or IL-10 secretion by ELISA. In comparison to SAP KO mice, both BMT and SAP gene transfer resulted in increased levels of IL-4 and IL-10 production whereas no difference in cytokine production was seen in mice receiving reporter gene modified cells.
Together, these data suggest that there is a significant defect of Ig production both in the naïve state and in response to T dependent antigen stimulation. These and other data suggest the abnormalities are due to an absence of T cell help. Our studies show that this humoral defect can be corrected by BMT and SAP gene therapy and the correction is due to the restoration of effective function in CD4+ T cells. Further experiments vaccinating mice after reconstitution are required to demonstrate that the full humoral function can be restored in these mice by BMT or gene therapy.
Analysis of membrane-IgG1- and IgE-signalling in a murine knock-in model:
Using stable isotope labeling with amino acids in cell culture (SILAC) in B cells, we have applied this technology to analyze the Spleen tyrosine kinase (Syk) and mapped Syk phosphorylation sites and elucidate the B-lymphoid interactome of human Syk.
The major task was the generation and analysis of gene-targeted mice, which express a mutated form of the mIgG1 (IgG1-YF) and mIgE (IgE-YF) B cell receptors. Like with the IgE-YF mice, chimeric IgG1-YF mice have been generated, and germline transmission of the modified allele has been achieved. Both mouse strains have been backcrossed to remove cre or flp recombinase transgenes, transferred to the barrier facility by in-vitro fertilization and bred to homozygosity. Cohorts of both strains have been immunized with model antigens and the immune response has been measured by ELISA. Preliminary results of these studies have been presented on the PADNet meetings.
Taken together, the two successfully established mouse models will provide the means to analyse the importance of tyrosine-phosphorylated cytoplasmic domains of mIgG and mIgE in BCR signaling in a physiological setting and to unravel isotype-specific signaling cascades. These results are important for understanding of PAD and, in the case of IgE, allergic diseases.
BCR-intrinsic costimulation of memory B cells:
During the project we have identified and published the adaptor molecule Grb2 as a pivotal factor for the BCR-intrinsic costimulation of memory B cells. We then set out to analyze the Grb2 interactome and identified Bruton's tyrosine kinase (Btk) as an interaction partner of Grb2. For further analysis, we generated a Btk-deficient variant of the human B cell line DG75, which will allow us to test the signalling potential of different BCR-variants in the presence or absence of Btk.
We have also performed experiments on Stap1, which is a unique IgE binding partner discovered during the first half of the project. On the one hand, we used the Stap1-deficient human B cell line DG75 generated in the first half to analyze the composition of IgE-interacting proteins in the absence of Stap1. On the other hand, we generated cell lines expressing a modified version of Stap1, which carries an affinity tag and allowed the purification of Stap1-interacting proteins. The subsequent mass spectrometric analysis of the purified complexes revealed several unknown interaction partners, and allowed us to draw first conclusions on the role and function of Stap1 in the BCR-intrinsic costimulation of memory B cells.
Murine knock-in model for antigen receptor signalling
New: Optimized splice correcting oligonucleotides (SCO) for gene correction of XLA in a mouse model:
The ability to correct the aberrant splicing varied for different SCOs. However, splice correcting LNA based oligos are not able to efficiently be taken up by lymphoid primary cells in our system (mouse B cells and human monocytes). Only by the use of electroporation techniques we were able to successfully correct the aberrant BTK splicing in those cell types. As such a new oligonucleotide chemistry - morpholino - was used in combination with a covalently attached cell penetrating peptide. Cell penetrating peptides covalently attached to antisense morpholino splice correcting oligomers (CPP-SCO) were used without the need for electroporation and were extremely efficient in correction of BTK aberrant splicing in patient monocytes. These splice correcting oligos equiped with the CPP are able, by the use of the membrane translocation properties of the peptide, to accumulate inside the cell when added exogenously to the medium. These cell uptake properties of the peptide-oligo conjugates makes them ideal candidates for further testing in in vivo situation.
The main objective of our work within the consortium is to improve our understanding on molecular basis of the Btk deficiency. We have developed a new model: BAC-transgenic mice carrying the human genomic BTK. We have two mouse strains; one carrying the WT human BTK gene and another carrying mutated human BTK gene. At the moment the transgenic mice are bred onto the Btk-/- C57BL6 mice and we assume that they will provide a new excellent model for studying different aspects of Btk deficiency. Furthermore, we aimed to correct the mutated BTK in the BAC-transgenic mouse as a proof of concept for the treatment possibilities. We achieved gene correction of the mutated BTK gene in a transgenic mouse model. The transgenic mice are now bred on the Btk-/- mice, providing an excellent humanized model to study different aspects of Btk and XLA disease. We have been able to successfully correct the mutated human BTK gene in our animal model and we could prove that both at the RNA and protein level. Gene correction by using the same means of monocytes fom XLA patients carrying similar mutations to prove the efficacy of the method in human primary cells was achieved. Gene expression profiling in Btk-/- and WT B cells stimulated with CpG as an important component in our studies on molecular basis of Btk deficiency. Our preliminary data indicate that the CpG stimulation alters the phenotype of Btk-/- cells making them more alike WT B cells (an important factor in XLA is that B-lymphocytes do not survive). This work is submitted under: Moreno P et.al. "Correction of splicing mutations in X-linked agammaglobulinema (XLA) using an exon skipping strategy".
Importantly, XLA has a very decisive feature, namely that the end-stage cells, the plasma cells, do not express any BTK protein. This means that if regular B-lymphocytes can develop as a result of the treatment, subsequent immunization is anticipated to generate long-lived plasma cells. This also means that a short-term treatment can mediate a long-term clinical outcome. If the animal model shows an in vivo treatment effect, the personalized medicines developed in this program could therefore hopefully, in the not too distant future, also be used as experimental drugs in the affected patients.
Development of human B cells in humanised mice:
In the humanized mouse model the co-transplantation of mesenchymal stem cells (MSC) may improve engraftment and develompent of human hematopoietic cells. We found that MSCs isolated from adult bone marrow are a heterogeneous cell population and therefore we are currently trying to identify those cells which provide the strongest support for HSC.
We studied the development of human B cells in humanized mice of different ages. In total, 273 mice were transplanted at UKL-FR with human CD34+ HSC. Bone marrow engraftment was found in 1/4 of the mice. For comparison, bone marrow B cell development was analyzed in BM samples from CVID patients.
In total, bone marrow samples of 48 patients and 15 controls were analyzed by immunohistology. From 25 patient, fresh aspirates were available and studied by immunophenotyping. In 94% of the patients' BM plasma cells were either absent or significantly reduced and correlated with serum IgG levels. Biopsies from CVID patients had significantly more diffuse and nodular T cell infiltrates than biopsies from controls. These infiltrates correlated with autoimmune cytopenia but not with other clinical symptoms nor with disease duration and peripheral B cell counts. Nodular T cell infiltrates correlated significantly with circulating memory T cells, elevated soluble IL2-receptor and neopterin serum levels indicating an activated T cell compartment in most patients. Nine out of 25 patients had a partial block in B cell development at the pre-B-I to pre-B-II stage. Since the developmental block correlates with lower transitional and mature B cell counts in the periphery, we propose that these patients might form a new subgroup of CVID patients.
TRANSLATIONAL RESEARCH
Biomarkers in CVID :
The use of b2MG and sIL2R as well as free light chains as biomarkers were evaluated retrospectively in a cohort of 100 CVID patients. Both b2MG and sIL2R correlate highly significant with the appearance of splenomegaly, lymphadenopathy, autoimmune disease and granulomatous disease. At the same time there is a significant association with cellular alterations of B and T cells most pronounced in CVID Ia patients. The value of these markers at the time of malignant transition in CVID patients with secondary lymphoma is under current evaluation.
Definition of protective role of IgM memory B cells and IgM anti PnPS antibodies in PAD:
We defined that IgM anti PnPS antibodies in PAD correlate with the capacity to differentiate into plasmablasts. The residual capacity to produce IgM anti PnPS antibodies correlated with the risk to develop pneumonia and chronic lung disease. B-cell subsets and their functional development have been also analyzed in 74 newborns from birth to 6 months of life. Moreover, we evaluated natural antibody production in vitro. The results documented a predominance of naive B-lymphocytes at all time-points evaluated, decreasing from birth to 6 months (p=0.009). The percentages of CD27+IgD+ and CD27+IgDneg memory B-cells were very low at birth and significantly increased only at 6 months (p=0.02 and p less than 0.001 respectively). We found a significant increase only in in vitro stimulated IgG production at 6 months as compared to birth (p less than 0.001). Moreover, a lower secretion of anti-Pn IgM antibodies up to 6 months of age, as compared to controls was observed. Our results underline that the susceptibility and severe course of infection in the neonate can be attributed, at least in part, to the lack of pre-existing immunological memory and competent adaptive immunity.
The role of IgM memory B cells in the protection from encapsulated bacteria:
In this project we evaluated B-cell subsets and their functional development in 74 newborns from birth to 6 months of life. Moreover, we evaluated natural antibody production in vitro. The results documented a predominance of naive B-lymphocytes at all time-points evaluated, decreasing from birth to 6 months (p=0.009). The percentages of CD27+IgD+ and CD27+IgDneg memory B-cells were very low at birth and significantly increased only at 6 months (p=0.02 and p less than 0.001 respectively). We found a significant increase only in in vitro stimulated IgG production at 6 months as compared to birth (p less than 0.001). Moreover, a lower secretion of anti-Pn IgM antibodies up to 6 months of age, as compared to controls was observed. Our results underline that the susceptibility and severe course of infection in the neonate can be attributed, at least in part, to the lack of pre-existing immunological memory and competent adaptive immunity.
Definition of B cell stimulation as prognostic indicator in PAD:
Reduced proliferative response to CpG correlates to increased lung infection rate. Due to reduced proliferative response, CpG induced plasmablast differentiation is reduced in patients with CD21low cells to <10%.
Development of prototype ELISA for IgM anti PnPs response:
The aim of the subproject is the development of a reliable and easy-to-use ELISA for anti-pneumococcal polysaccharide (PnPS) antibodies of IgM isotype. The assay is intended to assess the function of B cells in patients with antibody deficiency sndrome in vivo. The project was conducted in co-operation with our industrial partner, The Binding Site (TBS), Schwetzingen, Germany. N = 20 volunteers were vaccinated with a 23 valent pneumococcal polysaccharide vaccine (Pneumovax®). Subsequently, serum and cellular components were obtained at 8 time points within 2 months. The vaccination study was performed to i) determine the optimal time point for assessment of the anti PnPS-IgM-antibody response following PnPS vaccination and ii) to obtain specimen with high levels of anti-PnPS-IgM antibody contents required for the ELISA development. The anti-PnPS-IgM-ELISA was developed on the basis of a commercial anti-PnPS-IgG-ELISA provided by our industrial partner TBS. The assay was optimised for sensitivity and retest-reliabbility by serial variations of IgG removing agents and crossreactive antibodies binding substances. The ELISA is now ready for definitive measurements of the IgM immune response of the healthy vaccinees. The antibody immune response can be compared to the detection of PnPS specific B-cells at the same time points.
Evaluation of IgM anti PnPS response in PAD patients after vaccination:
The serum level of IgM antibodies against different pneumococcal serotypes (type 1, 3, 4, 14) has been analysed according to the D. Goldbatt (WHO) recommendations before and after vaccination with unconjugated pneumococcal vaccine. In cooperation with The Binding Site (Germany) an easy-to-use ELISA for anti-PnPS antibodies of IgA and IgM isotype has been developed. This test is based on a commercially available 23-valent anti-PnPS ELISA for IgG antibodies (Pneumococcal Capsular Polysaccharide (PCP) Assay, The binding Site).
Outcome: A good correlation of the IgM 23-valent anti-PnPS with each single serotypes IgM titre (type 1, 3, 4, 14) was observed. The serum level of specific antibody IgM and IgA against single serotypes (type 1, 3, 4, 14) and the IgM and IgA 23-valent anti-PnPS have been correlated with the B lymphocyte subsets (according to the IPIDNET panel). A response to at least two serotypes, doubling the IgM or IgA level was considered a good response to vaccination. Based on this definition, a good IgM response was significantly related to higher numbers of switched memory B cells and IgM memory B cells and with lower levels of CD21low B cells. Instead a good IgA response to vaccination was significantly related to higher numbers of switched memory B cells and with lower levels of CD21low B cells. The clinical evaluation included the clinical history (frequency of pneumonias), the chest CT scan for the detection of bronchiectasis and other structural changes and the pulmonary function test (FEV1 of less than 80% [predicted]) over a period of 36 months. An inverse correlation was found between the response to the vaccination and the frequency of pneumonia and between the response to the vaccination and the presence of bronchiectasis at the chest CT scan. Conversely, no significant correlation was found between the response to immunization and the pulmonary function test (FEV1). Observing the distribution of the response to vaccination with specific IgA and IgM 23-valent anti-PnPS production, a cut-off level has been suggested in order to define a response to vaccination in subjects with PAD: IgM>20 U/ml and IgA>0.15 µg/ml. Based on this definition, three groups of patients have been described (see graph below): Patients with good IgM and IgA response (complete responders), patients with no response at all (non-responders) and patients with IgM response only (IgM only). The latter group of patients (IgM only) has defined intermediate level of switched memory B cells (lower than complete responders but higher that non-responders) and lower numbers of CD21low B cells than complete responder patients. Comparing the response to vaccination with IgA and IgM 23-valent anti-PnPS production to the clinical findings, a stronger inverse correlation was found between the complete response to the vaccination and the frequency of pneumonia and between the response to the vaccination and the presence of bronchiectasis at the chest CT scan, while an intermediate inverse correlation was observed in the IgM only group. (see Figure 1)
Assessing antibody response in patients with hypogammaglobulinaemia by ELISPOT:
Hypogammaglobulinaemic patients are often started on immunoglobulin substitution therapy before antibody production is adequately evaluated. In such a situation, it is difficult to segregate transferred from antigen-induced specific antibody. Therefore we characterized changes in B-cell subpopulations in hypogammaglobulinaemic patients, including plasmablasts, in peripheral blood by flow cytometry after in vivo antigen challenge. We investigated the specificity of antibody production on the B-cell level by ELISPOT, which is independent of substitution therapy.
Hypogammaglobulinaemic patients and healthy volunteers were immunized with tetanus toxoid (T-cell dependent) and pneumococcal polysaccharide (T-cell independent) vaccines. Specific antibody levels were measured by ELISA. ELISPOT was used before and on day 7 after vaccination. B-cell subpopulations were examined by flow cytometry. In the control group, an increase in circulating plasmablasts (IgD-CD27++, IgM-CD38++ gated from CD19+ B cells) on day 7 post immunization corresponded with the appearance of spot forming cells. In contrast, patients with Common variable immunodeficiency (CVID) who are characterized by hypogammaglobulinaemia and impaired antibody production failed to increase plasmablasts significantly in peripheral blood after antigen challenge.
Our observation that the majority of CVID patients lack antigen specific spot forming B cells and fail to increase circulating plasmablasts following in vivo antigen challenge provides a rapid screening test to demonstrate defective antibody responses in CVID patients.
Clinical Implications and Conclusion: Identification of circulating plasmablasts after vaccination is a new simple flow based test to assess antibody responses in hypogammaglobulinaemic patients, even if on immunoglobulin (IVIG or SCIG) replacement therapy.
Development of an IgM-specific anti PnPS ELISA based on 23 serotypes from S. pneumonia and
Assessment of the IgM and IgG anti PnPS antibody response after vaccination with pneumococcal unconjugated vaccine in PAD patients:
As an immunodiagnostic company the role of The Binding Site in the EUROPADnet project was to translate the observations from primary antibody deficient (PAD) patients into a medical device so the data and results could be utilised in patient care.
Since there is a high risk of bacterial infections in patients suffering from common variable immunodeficiency (CVID) we supported the investigation for the hypothesis that IgA and IgM responses to pneumovax may improve the determination of the risk of pulmonary morbidity in PAD patients (work package 5).
Along with the study group we have facilitated the development of both PCP IgM and IgA ELISAs. The assays have been validated and the IgM and IgA responses to pneumovax have been determined in normal healthy subjects.
The development of both PCP IgM and IgA ELISAs has involved the identification, location and preparation of suitable materials. The optimisations of assay conditions and clinical validation have been achieved in collaboration with the study group. Both assays have been initiated as new products and entered into the New Product Development system at the Binding Site. This process involves full product specification and commercial justification established through the EUROPADnet group.
Working with the same team of experts we are currently evaluating the measurement of both the IgA and IgM responses to pneumovax using these ELISAs in broader and wider clinical applications such as in patients diagnosed with IgA deficiency and those with combined IgA/IgG subclass deficiencies.
Predicting the risk of infusion reactions to anti-IgA antibodies:
The occurrence of anti-IgA antibodies at high titre is a risk factor associated with anaphylactoid reaction in hypogammaglobulinemic patients. The presence or absence of anti-IgA antibodies is also considered important basic information in the European ESID database of primary immunodeficiencies. Screening for anti-IgA antibodies in patients lacking IgA may be helpful for determine the best therapeutic strategy, including the route of administration of substitution immunoglobulin therapy or the transfusion of IgA-free blood products. From a hematological point of view tested IgA-deficient blood components without anti-IgA antibodies are a precious resource that should be allocated with care and reserved for IgAD patients at risk of anaphylactoid reaction. The combined use of quantitative ELISA for detection of anti-IgA antibodies and the measurement of serum IgA concentration by nephelometry provides an effective and safe strategy for the diagnosis and prevention of IgA anaphylactic transfusion reactions. We found a significant amount of anti-IgA antibodies in 10 % of patients with the most common primary hypogammaglobulinemia (selective IgA deficiency and common variable immunodeficiencies) where the serum IgA level was lower than 0.05 g/L. Analysis of anti-IgA response in PAD patients and the assessment of biomarker predicting the risk of infusion reaction in PAD demonstrated that Common Variable Immunodeficiency (CVID) patients who are seropositive for anti-IgA antibodies have a predisposition for anaphylactoid reactions to intravenous immunoglobulin replacement therapy (IVIG). A new anti-IgA ELISA has been established www.biovendor.com .
Development of a combined clinical, phenotypic and functional classification of PAD:
This subproject represents the main translational research target of WP5 and might be completed in the following months. This is a final goal of WP5. Earlier than anticipated, and thanks to the sterling work of EUROPADnet, the new classification of PAD is already in progress. A proposed algorithm for CVID diagnosis is illustrated in figure 2 (attached)ow:
Major changes of the new proposal compared to the current criteria present the definition of absolute immunoglobulin levels as cut off, the definition of vaccine responses, a more precise definition of exclusion criteria especially defining patients with combined immunodeficiency.
Potential Impact:
The consortium has produced a large number of publications already, based on the research results generated to date. At present several publications are either at the manuscript stage or already submitted. These will focus on the achievements throughout the second reporting period and cover the whole project. Publications are listed in the respective section in the reporting
system. The grant number and Acronym have been used accordingly and the support of the EC will be mentioned in the future when publications rely on results generated within EURO-PADnet.
The partners have represented the project in numerous conferences and meetings.
CONSORTIUM MEETINGS
The regular consortium meetings have taken place as planned:
- 21 January 2008, London, Royal Free Hospital
- 15 October 2008, S'Hertogenbosch, Golden Tulip Hotel
- 3-4 April 2009, Rome, Universita Sapienza
- 12 September 2009, Berlin, Art'otel Berlin Mitte
- 30 April 2010, Università degli Studi di Brescia, Spedali Civili di Brescia, Italy
- 4. and 5. November 2010, John Radcliffe Hospital Oxford, UK
- 1 April 2011, Hopital Necker, Paris, France
REPORT FROM THE EDUCATIONAL COMMITTEE
At the beginning of the project the idea was developed that several consortium meetings should be complemented by a second day allocated to teaching activities. Therefore, teaching days were organised and opened for local and any other interested person in order to provide an educational opportunity for doctoral students and undergraduates in any participating centre (mostly taken up by the local host institute). Such sessions were held at three PADnet meetings in Rome, Berlin and Oxford.
In addition, doctoral students from participating centres were encouraged to give study updates. In Rome a specific mentor was identified beforehand to help each speaker and the talks were graded formatively to give feedback. This was successful and the same forms were used at the Oxford meeting. An example is given below: The Educational Afternoon Session offered a great opportunity for junior scientists from the EURO-PADnet partners and other juniors to hear educational talks from some of the consortium's senior scientists and PIs. Furthermore, invited speakers contributed to the success of the afternoon. After the plenary session, the participants had the opportunity to ask questions or discuss problems with the speakers.
Presentations covered the following fields:
- Proliferative capacity of B cells in antibody deficiency (Guest speaker Gertjan Driessen, Erasmus MC Rotterdam)
- Pitfalls using linkage analysis (Bodo Grimbacher, UCL)
- Defects in IL10 signalling cause inflammatory bowel disease in children (Erik Glocker, UCL)
- New assays
- Berne Ferry (UOXF-HX): Salmonella antibody testing
- Isabella Quinti (Sapienza): IgM pneumococcal assay
- Vojtech Thon (MU): Anti-IgA antibody assay
- Translation of useful assays to clinical use (Hermann Eibel, UKL-FR)
Many PADnet participants and their trainees attended the educational sessions as part of the biennial meeting of the European Society for ImmuneDeficiencies [ESID], in which there was a specific session for primary antibody deficiencies run by Helen Chapel and Gavin Spickett.
Attendance was also encouraged at specific educational sessions arranged within the ECI congress, Berlin in September 2009, hosted by Reinhold Schmidt (MHH) and arranged by Helen Chapel andB.Grimbacher. Specifically two symposia on immune deficiencies were organized and sponsored with the generous help of pharmaceutical industry (CSL-Behring and Baxter). These educational events were open for EURO-PADnet members as were numerous Satellite symposia; both educational events were very well attended by over 250 persons.
One of the educational highlights of this reporting period was the PID Winter-School organised with the help of the University College London in Windsor from 28th February to March 2nd at the Cumberland Lodge.
The aim of this school was i) to educate students in the field ii) to transmit latest research data from EURO PADnet projects to the field iii) to get to know excellent young hopeful students in the field for further career support and attachment to the field of immune deficiencies.
37 students from 7 European countries (UK, Italy, Spain, Portugal, Belgium, Netherlands and Switzerland) were accepted based on a personal application with support letter and including an abstract for presentation and a personal career statement. Active participation of the students was encouraged and also well received (see evaluation attached). Faculty included approx. 20 lecturers, including Profs Chapel, Grimbacher, Gaspar and Thrasher from EURO PADnet (agenda attached). EURO PADnet provided approximately one third of the funding for this event.
REPORT FROM THE ETHICS COMMITTEE
Relevant ethical approvals as well as a declaration from each Principal Investigator, stating that the ethical approvals cover the entirework on the project and the entire period of the project have been submitted to the EC. Licences for animal experimentation have been included. The beneficiaries VUMC and TBS only work with samples that are ethically approved by the other partner institutions and beneficiary UOXF-HX stated that the work they reported in the first half of the project did not require ethical approval. The consenting process for the ESID and UKPID registry projects (WP1) is well-established with approved Patient Information Sheets and Consent Forms for adults, young adults and parents of children. In the UK a national version of the ESID registry is in use on local servers. The patient is therefore given the additional choice, whether he would like his data to remain on the UK registry only or whether he wants it to be included in the ESIID registry as well, which then takes place by an electronic import. This mechanism has been ethically approved in the UK and is running smoothly.
List of Websites:
The project website www.europadnet.eu is online and contains general information about the project and can also be accessed by the partners via a secure login to allow them to exchange information on a confidential level. Some results have been made public on the website and final results will be posted after the submission of the final report. We also have implemented a teaching website www.ipidnet.org for linking the clinical and cellular phenotyping, and a www.chest-ct-group.eu for the analysis of lung complications in antibody deficiencies.