Fig. 1.
JAK genes from the zebrafish. (a) Alignment of deduced amino acid sequences for the PTKs identified by PCR from zebrafish cDNA. The alignment was generated using CLUSTAL.50 The clone name is listed as HD on the right of the sequences, followed by the name of the gene that gave the highest BLAST score in database similarity searches.49 The JAK-specific sequence elements in Hanks motif VIII are boxed. (b) Comparison of the zebrafish jak2a and human JAK2 protein. The amino acid sequence (single-letter code) of the jak2a protein from zebrafish and the JAK2 protein from human88 is numbered from the putative initiation methionine. The JAK homology (JH) domains are indicated by brackets and are labeled JH1 through JH7 according to Harpur et al.3 Conserved motifs in the catalytic JH1 domain are labeled in roman numerals. Conserved motifs in the kinase related domain (JH2) are labeled in roman numerals with the subscript a. A series of conserved residues mutated in murine Jak2 without phenotypic effect in IFN signaling8 are indicated by an asterisk above the amino acid, and are labeled according to Kohlhuber et al8 by capital letters. The site of the E665K mutation within JH2 that hyperactivates the catalytic activity31 of murine Jak2 and D melanogaster HOP is marked with a solid arrowhead. The autophosphorylation site in the JH1 domain of murine Jak2, which is required for catalytic activity,89 is marked by an open arrowhead. The structure of the variant jak2aβcDNA is marked by an arrow at the site of predicted translational termination due to alternate splicing.62 (c) Phylogeny of the JAK gene family. The amino acid sequences of zebrafish (Danio rerio, Dr) jak2a and jak2b were aligned across the known region of jak2b, consisting of most of JH2 and all of JH1, with JAK2 proteins from human (Homo sapiens, Hs),88 pig (Sus scrofa,Ss),90 mouse (Mus musculus,Mm),3 and rat (Rattus norvegicus,Rn)91; all other members of the JAK family from human (Hs),10,41,42 the zebrafish (Dr) jak1 sequence (this study), and the sequence of the D melanogaster (Dm) JAK homolog,42 hopscotch, using the CLUSTAL alignment algorithm.50This alignment was used to construct a dendrogram with the maximum parsimony options of the DNAstar MegALIGN application to infer the likely genealogy of the JAK family. The names of the sequences are displayed to the right of the dendrogram; zebrafish sequences are in bold.