Fig. 4. The medaka mespb PSM enhancer is functionally equivalent to its counterpart in the mouse. (A) A comparison of the medaka mespb and mouse Mesp2 PSME regions. Black and gray boxes represent presumptive T-box binding sites. The numbers above the boxes represent the nucleotide positions from the first ATG. The nucleotide sequences of the putative T-box binding sequences are shown beneath. Consensus Tbx6 binding sequences and their directions are indicated by arrows. The dashed arrow in Site B of the Mesp2 PSME depicts an incomplete Tbx6 binding sequence that only binds to Tbx6 if an adjoining complete Tbx6 binding sequence is present. The T-box proteins that might bind to these sequences are indicated. (B) EMSA analysis of the T-box binding sites in the medaka mespb PSME. T-box binding site T1 associates with a single Tbx24 molecule and T2 and T3 with two Tbx24 molecules, which is consistent with their nucleotide sequences as shown in A. (C) The targeting strategy used to generate the medaka mespb PSME knock-in mouse (medakaP2). A 2.8-kb fragment of mespb genomic DNA that is required for PSM-specific mespb expression was substituted for Mesp2 PSME by homologous recombination. The neoR selection marker was removed by recombination using the Cre-loxP system. (D) medakaP2 homozygotes are viable and have normal external features. (E) Homozygotes are indistinguishable from heterozygotes and wild-type littermates in skeletal preparations.