In FY2018, we have expanded knowledge on the pathophysiology of combined immune deficiencies (CID), developed novel assays to assess pathogenicity of genetic variants in the RAG2 gene, and gained preclinical data to test safety and efficacy of gene therapy for RAG2 deficiency in particular. Infants affected with the most severe form of these disorders, also known as severe combined immune deficiency (SCID), are highly prone to serious infections since early after birth, and die within the first few years of life unless immune reconstitution is obtained, usually after hematopoietic cell transplantation (HCT). However, even after HCT a proportion of patients with SCID and related disorders continue to manifest clinical problems, including transplant-related toxicities (such as graft-versus-host disease -GvHD- and chemotherapy-induced organ damage) and manifestations related to incomplete immune reconstitution, such as infections and immune dysregulation. Defining factors that affect the quality and robustness of immune reconstitution after transplant, and developing alternative forms of treatment, such as gene therapy, that would eliminate the risk of GvHD, remain important objectives. Moreover, unexpected phenotypic heterogeneity has been reported in patients with mutations in SCID-associated genes. Defects of the RAG1 and RAG2 genes represent a prototypical example of such clinical heterogeneity, CID, with some patients presenting early in life with generalized erythrodermia and prominent lymphadenopathy and hepatosplenomegaly (Omenn syndrome), and other presenting later in life with combined immune deficiency associated with granulomas and/or autoimmunity (CID-G/AI). This phenotypic heterogeneity, and presentation with autoimmunity in particular, is not restricted to hypomorphic RAG mutations, but has been observed also in patients with CD3 deficiency. Overall, the molecular and cellular bases underlying phenotypic diversity and immune dysregulation in these conditions remain ill defined. During FY2018, the following achievements have been obtained: 1) We have identified novel hypomorphic RAG mutations associated with a spectrum of clinical phenotypes that include infections, organ-specific autoimmunity, and granulomatous lesions (1-7). In the attempt to perform genotype-phenotype correlation in RAG deficiency, have developed a cellular platform to analyze the recombination activity of naturally occurring RAG2 mutations. To this purpose, we have generated Abelson virus-transformed mouse Rag2-/- pro-B cells that contain an inverted GFP cassette flanked by recombination signal sequences. Upon retrovirus-mediated introduction of either wild-type or mutant human RAG2 into these cells, GFP expression can be used to track efficiency of the recombination activity of the mutant RAG2 protein. In addition, high throughput sequencing (HTS) of rearrangements introduced at the endogenous Ighc locus in these cells can also be used to assess RAG2 recombination activity. Using this assay we have demonstrated genotype-phenotype correlation in human RAG2 deficiency (1). The results obtained correlate well with the diversity and richness of T and B cell repertoire detected in vivo in the patients. These results complement what previously demonstrated by our group for human RAG1 deficiency. 2) HTS analysis of T cell receptor (TRB) rearrangements in sorted T cell subsets from patients with RAG deficiency has revealed a severe restriction of repertoire diversity in conventional CD4+ (Tconv), regulatory T (Treg) and CD8+ cells from patients with Omenn syndrome (2). By contrast, patients with CID-G/AI display a modest restriction of TRB repertoire in Tconv and CD8+ T cells, but their Treg cells express reduced diversity (2). This observation is consistent with faulty cross-talk between CD4+ thymocytes, medullary thymic epithelial cells (mTECs), and dendritic cells in the thymus medulla. Furthermore, based on our previous observation that mTECs from RAG-mutated patients fail to express AIRE, we had anticipated a defect in the negative selection of self-reactive thymocytes. Consistent with this hypothesis, we have observed that Tconv cells from patients with CID-G/AI have a molecular signature of self-reactivity, with increased usage of hydrophobic amino acids at positions 6 and 7 of the TRB-CDR3 (2). We have also found that similar abnormalities are shared by peripheral T cells from patients with CD3 gamma deficiency (8), a condition that is mainly characterized by autoimmunity. 3) To better characterize the mechanisms underlying immune deficiency and immune dysregulation in patients with CID-G/AI, we have generated three new mouse models with mutations in the Rag1 gene that correspond to mutations observed in patients with CID-G/AI (9). Our results have confirmed that these mutations allow residual T and B cell development, with a gradient of severity of the immunological phenotype. Mutant mice spontaneously produce autoantibodies and have impaired T-dependent antibody responses, thus mirroring what seen in patients. The study of these models has allowed us to identify novel mechanisms that account for the immune deficiency and immune dysregulation of CID-G/AI, in particular: a) impairment in sequential rearrangements at the TRB and IGHC loci, resulting in progressive cell loss and increased proportion of peripheral cells with productive rearrangements on one allele, the other being in germ-line configuration; b) peculiar utilization of immunoglobulin and T cell receptor Variable (V) and Joining (J) genes, resulting in abnormalities of T and B cell repertoire composition; c) abnormal distribution of B1 and B2 B cells, with increase of self-reactive B cells. 4) In a collaborative study exploring safety and efficacy of lentivirus-mediated gene therapy in mouse model of RAG2 deficiency, we have demonstrated that this approach can improve immune reconstitution without leading to autoantibody production (10). 5) As part of a large collaborative study with the Primary Immune Deficiency Treatment Consortium (PIDTC), we have found that the quality of reconstitution of humoral immunity after HCT in patients with X-linked and JAK3 deficiency does not correlate with the number of follicular helper T (Tfh) cells that are generated, but rather with the attainment of robust donor B cell engraftment, allowing normal response to IL-21 (11). As part of the consortium, we have also contributed to the analysis of immune reconstitution and outcome of the first 100 patients with SCID treated by HCT and enrolled in a prospective natural history study (12), and we have developed recommendations for screening and management of late effects after HCT for SCID (13). Finally, we have co-authored a position statement for the American Academy of Pediatrics on cord blood banking for potential future transplantation (14). 6) In a series of collaborative studies, we have contributed to the identification of novel genetic defects responsible for primary immune deficiencies, in particular, DBR1 mutations accounting for brainstem viral encephalitis (15) and Cadherin 17 deficiency responsible for thymic dysfunction (16). We have also helped identify a role for neutrophil-derived type 1 Interferon, inflammatory macrophages and platelet abnormalities in the autoimmunity of Wiskott-Aldrich syndrome (17-19).