Anhidrotic ectodermal dysplasia (EDA) is a disorder of ectodermal differentiation characterized by sparse hair, abnormal or missing teeth, and inability to sweat. X-linked EDA is the most common form, caused by mutations in the EDA gene, which encodes ectodysplasin, a member of the tumor necrosis factor (TNF) family. Autosomal dominant and recessive forms of EDA have been also described and are accounted for by two genes. Mutations in EDAR, encoding a TNF receptor (EDAR) cause both dominant and recessive forms. In addition, mutations in a recently identified gene, EDARADD, encoding EDAR-associated death domain (EDARADD) have been shown to cause autosomal recessive EDA. Here, we report a large Moroccan family with an autosomal dominant EDA. We mapped the disease gene to chromosome 1q42.2-q43, and identified a novel missense mutation in the EDARADD gene (c.335T>G, p.Leu112Arg). Thus, the EDARADD gene accounts for both recessive and dominant EDA. EDAR is activated by its ligand, ectodysplasin, and uses EDARADD to build an intracellular complex and activate nuclear factor kappa B (NF-kB). We compared the functional consequences of the dominant (p.Leu112Arg) and recessive mutation (p.Glu142Lys), which both occurred in the death domain (DD) of EDARADD. We demonstrated that the p.Leu112Arg mutation completely abrogated NF-kB activation, whereas the p.Glu142Lys retained the ability to significantly activate the NF-kB pathway. The p.Leu112Arg mutation is probably a dominant negative form as its cotransfection impaired the wild-type EDARADD's ability to activate NF-kB. Our results confirm that NF-kB activation is impaired in EDA and support the role of EDARADD DD as a downstream effector of EDAR signaling.