Fig. 2. ULTRAPETALA1 cloning and sequence analysis. (A) Schematic of the positional cloning of the ULT1 locus and the structure of the ULT1 gene. The region of chromosome 4 containing BACs T29A15 to F19B15 is represented. The CAPS markers designed for mapping ult1-2 are shown in black boxes and the frequency of recombinant chromosomes is indicated for each marker. The exon/intron structure of the ULT1 gene is shown along with the positions of the ult1-1 and ult1-2 mutations. (B) Alignment of the conceptual translation products of the Arabidopsis ULT1 and ULT2 genomic sequences with conceptually translated consensus EST sequences from four other plant species. The sequences compared are from Arabidopsis thaliana ULT1 (AtULT1, At4g28190), Arabidopsis thaliana ULT2 (AtULT2, At2g20825), Glycine max (GmULT, BM524875.1), Lycopersicon esculentum (LeULT, EST357945), Oryza sativa (OsULT, CA763280.1) and Triticum aestivum (TaULT, BG604592). Identical amino acids are boxed and blocks of similar amino acid residues are shaded. The positions of the mutations in ULT1 and ULT2 are shown above the sequences, the SAND domain is boxed in red and the B box-like motif is boxed in green. Stars indicate the amino acid substitution in the ult1-1 (C173T) and ult1-2 (S83F) alleles. Arrowheads denote the position of the T-DNA insertion in the ult1-3 and the ult2-1 allele. Arginine/lysine rich nuclear localization signal (NLS) candidate polypeptides are underlined in white. (C) Multiple sequence alignment of AtULT1, AtULT2 and animal SAND domains from CeC44F1.2 (Caenorhabditis elegans, Z49067), CeC25G4.4 (Caenorhabditis elegans, Z70680), HsSp100b (Homo sapiens, U36501), HsNucP41 (Homo sapiens, Q14976), HsNUDR (Homo sapiens, AF049459), DmDEAF-1 (Drosophila melanogaster, AAC47040, HsGMEB2 (Homo sapiens, NM031803), HsGMEB1 (Homo sapiens, NM006582), AIRE-1 (Homo sapiens, AB006682). The alignment was obtained with the ClustalW 1.82 program and manually refined using the calculated two-dimensional structure. Secondary structure elements are shown above the multiple alignment. Period, semicolon and asterisk mark partial to full residue conservation. Color-coding reflects the conservation of amino acid types. Background colors reveal their physiochemical properties (green: hydrophobic; red: positively charged residues; blue: negatively charged residues), while foreground colors mark identical (red) and similar (blue) amino acids. The amino acid corresponding to the position of ult1-2 mutation is underlined. (D) Alignment of the AtULT1 and AtULT2 B box-like domains with B-box proteins from animals: CeLIN-41 (Caenorhabditis elegans, NP492488), CeNCL-1 (Caenorhabditis elegans, P34611), DmDAPPLED (Drosophila melanogaster, Q9V4M2), HsTIF-1 (Homo sapiens, NP003843), HsPML (Homo sapiens, P29590). The conserved cysteine/histidine residues are boxed. Below the sequence alignment, the conserved spacing of the B2 B-box consensus (Torok and Etkin, 2000) and the ULT B-box consensus are compared.