Targeted disruption of dematin gene. (A) Targeting vector. Top panel depicts WT mouse dematin (Epb4.9/Epb49/Dmtn/ AI325486) gene. The exon 5 to 8 region is targeted for deletion and the flanking regions upstream (5.7 kb) and downstream (3.9 kb) that constitute the 5′-homology arm and 3′-homology arm of the targeting vector (middle). Lower panel represents the correctly targeted dematin gene designated as knockout-first allele. (B) Western blotting and Coomassie blue–stained SDS/polyacrylamide gel electrophoresis profile of RBC ghosts from WT (^+/+), heterozygous (^+/−), dematin FLKO (^−/−) mice, and phenylhydrazine-treated WT (PHZ) erythrocytes/reticulocytes ghosts. Western blotting of RBC ghosts was performed using dematin monoclonal antibody C18 against the core domain of dematin (upper). Similar results were obtained using a polyclonal antibody raised against the full-length dematin. The absence of both 48- and 52-kDa isoforms of dematin in the knockout samples was confirmed with no trace of any residual truncated polypeptide in the expanded gels. In the Coomassie blue–stained SDS/polyacrylamide gel electrophoresis (10%) profile (lower), band 3 was used to normalize equal amounts of total protein. The high reticulocyte count in FLKO samples (∼66%) and PHZ-treated WT (∼60%) samples contributed to distinct profile of membrane polypeptides in the homozygous and WT mice. (C) Pups at day 1. Note the paleness of dematin FLKO (^−/−) compared with WT (^+/+) and heterozygous (^+/−) pups. (D) Splenomegaly observed in dematin FLKO mice. (E) Reticulocyte counts in peripheral blood from WT (^+/+), heterozygous (^+/−), and dematin FLKO (^−/−) mice. Reticulocytes were stained by Thiazole orange and counted by flow cytometry. Data were compared by unpaired t test. Mean ± standard deviation (SD): (^+/+), n = 3; (^+/−), n = 3; (^−/−), n = 3. ***P < .0001.