OR13-001 Loss-of-function mutations in CECR1, encoding adenosine deaminase 2 (ADA2), cause recurrent fevers and early onset strokes

Results The 5 unrelated patients shared 3 missense mutations in CECR1, encoding adenosine deaminase 2 (ADA2), with the genotypes A109D/Y453C, Y453C/G47A, G47A/ H112Q, R169Q/Y453C, and R169Q/28kb genomic deletion encompassing the 5’UTR and first exon of CECR1. All of the mutations are either novel or present at low frequency (<0.001) in several large databases. Computer modeling based on the crystal structure of the human ADA2 protein suggests that CECR1 mutations either disrupt protein stability or impair ADA2 enzyme activity. Whereas the ADA1 protein, which is mutated in some patients with severe combined immunodeficiency disease, is an intracellular protein that catalyzes the conversion of adenosine to inosine, ADA2 is expressed predominantly in myeloid cells and secreted into the blood, and has a lower affinity for adenosine than ADA1. Western blots showed a decrease in ADA2 protein in supernatants of PBMCs from patients relative to healthy controls. All patients had at least 10-fold diminished serum and plasma concentrations of ADA2 protein, and reduced ADA2-specific adenosine deaminase activity. However, whereas ADA1 deficiency leads to the accumulation of deoxyadenosine nucleotides and lymphocytotoxicity, such toxic metabolites were not found in the blood of patients with ADA2 deficiency. Animal models suggest that ADA2 is the prototype for a family of proteins with growth factor activity (adenosine deaminase growth factors, ADGF). There is no mouse homolog of CECR1, but there are 2 homologs in the zebrafish, Cecr1a and Cecr1b. While a zebrafish line with a hypomorphic retroviral insertion in Cecr1a did not exhibit an obvious phenotype, when we used 2 different morpholinos to knock down the expression of Cecr1b, at 48 hours post fertilization we observed intracranial hemorrhage in approximately 50% of zebrafish embryos, but in only 3% of control embryos. Utilizing zebrafish lines with GFP-tagged leukocytes, we found that Cecr1b knockdown also led to the near absence of neutrophils, but not monocytes. These observations suggest that the ADA2 homolog may be necessary both for vascular integrity and leukocyte development in the zebrafish, and that the near absence of ADA2 in humans may lead to strokes and autoinflammation by similar mechanisms.