Structural features of proteins encoded by novel genes. The full length coding regions of eight new proteins were determined and the domain structure of the encoded proteins was determined using SMART and Profilescan. The structure of the predicted human protein is shown except for 7012 where no motifs were detected and 15037 where the mouse protein is shown. The size of each protein and its closest relative in protein databases is shown along with a putative function where one could be determined based on structural motifs. Names approved by the HUGO nomenclature committee are indicated for KIF27 and SMYD2, the other clones are referred to as hypothetical proteins in current database entries. The structural motifs detected are indicated on each linear protein. Key: kinesin motor: kinesin motor domain; neck: kinesin neck domain; TOPEUc; topoisomerase domain; CC: predicted coiled coil segment; TM transmembrane region; cNMP binding: cyclic nucleotide binding domain; V-ATPase: V-type ATPase catalytic domain; S and ET: interrupted Su(var)3–9, E(z), trithorax domain; MYND: myeloid transcription factor, nervy, DEAF-1 domain.

Alignment of human, macaque, mouse and rat KIF27 sequences. The derived amino acid sequences of human, macaque, mouse and rat KIF27 were aligned using the AlignX program from the Vector NTI suite of sequence analysis programs. The KIF27A sequence from human is shown and identical amino acids within the orthologous proteins are indicated with a dash. Gaps are indicated as spaces within the sequence. The motor domain is indicated with a red underline, the neck domain with a blue underline and the topoisomerase domain with a green underline. The segment corresponding to the original hamster clone is indicated with a red dotted underline. Key: Hs: homo sapiens; Mf: macaca fascicularis; Mm: mus musculus; Rn: rattus norvegicus.

Alignment of human, macaque, mouse and rat 7012 sequences. The sequenc es of predicted peptides for human (Hs), macaque (Mf), mouse (Mm) and rat (Rn) orthologues of 7012 were compared using AlignX. Portions of the human sequence were obtained from XP_059659, a peptide predicted from annotation of the human genome at NCBI. The macaque sequence was derived from Genbank entry BAB63112, the predicted translation product of a cDNA isolated from macaque testis. The mouse sequence was obtained by mapping the exon structure of the orthologous gene on the region of mouse chromosome 15. The rat coding sequence was constructed by assembling conserved exons on contigs RNOR01068092, RNOR01068093 and RNOR01068094 (Accession numbers AABR02034780 and AABR02122391). The sequence corresponding to the hamster clone obtained in this study is indicated with a dotted red underline in the human sequence.

Alignment of human, macaque, mouse and rat 13043 sequences. This alignment was anchored on a single cDNA isolated from macaque (AB0770087). This sequence was compared to the human (Hs), mouse (Mm) and rat (Rn) genomic sequences at Ensembl and individual exons were assigned. One additional potential exon was assigned based on similarity between genomic sequences based on PipMaker analysis. The human and mouse sequences listed here match partially with predicted proteins in both human and mouse (ENSP0000306627 and XP_156183, respectively) but conform more to the template predicted by the macaque sequence. The rat sequence was constructed from genomic sequences AABR02142366, AABR02119574, AABR02127843, AABR02117677 and AABR02097473. Predicted transmembrane domains are underlined in red in the human sequence and a potential nucleotide binding domain is underlined in blue. A putative cation channel in the human, mouse and rat proteins is double underlined in the mouse sequence. The region of similarity to the original hamster clone is indicated with a red dotted underline.

Alignment of human and rat 15018 sequences. The human (Hs) sequence is derived from the predicted protein FLJ30655 and the rat (Rn) sequence from SCAP1A (NP_714962). The similarity to the hamster clone is underlined.

Alignment of human, mouse, rat and pufferfish SMYD2 sequences. The human (Hs) sequence was derived from HSKM-B (NM_020197) and the mouse (Mm) sequence from BC023119. The rat (Rn) sequence was established by mapping exons in rat genomic sequences (RNOR01035016 and RNOR01035017; Accession numbers: AABR02111753 and AABR02114803) and the Fugu (Fr, fugu rupripes) sequence using a similar strategy in Fugu genomic sequences (Scaffold _6078.1 and Scaffold 1370.1). The interrupted SET domain (S-ET) is indicated with a blue underline and the MYND domain with a red underline. The Cys and His residues in the MYND domain predicted to form two zinc fingers are indicated in red type and with an asterisk and the Cys-rich region following the SET domain is underlined in black.

Northern analysis of expression patterns of putative novel genes in hamster tissues. 20 μg of total RNA from the indicated hamster tissues were electrophoresed on formaldehyde-agarose gels, transferred to Nylon membranes and hybridized with ³²P-labeled DNA probes derived from putative novel genes. The clone number of each gene is listed to the left of each panel and the positions of the 18S and 28S ribosomal RNAs are indicated to the right. Key: K, kidney; T, testis; SV, seminal vesicle; B, brain; S, spleen; H, heart; L, liver; I, intestine.

Clone 1030 is derived from a new member of the kinesin superfamily. A. The genomic structure of 1030-derived genes on chromosomes 9, 13 and 17 of human, mouse and rat, respectively, are depicted with numbered vertical bars representing exons. The rat sequence was construc ted from four separate contigs (RNOR01052409, RNOR01052410, RNOR01052411 and RNOR01052412, Accession numbers: AABR02079933, AABR02039869, AABR02135902 and AABR02012418) and the arrowheads indicate the positions of short gaps in the genomic sequence. B. PCR analysis of human testis cDNA reveals the presence of three alternatively spliced isoforms. Human testis cDNA was amplified with primers designed to amplify bases 2118–3129 of the predicted full length cDNA based on the genomic structure described above. Three discrete products were observed and are indicated with asterisks. These products were isolated, subcloned and sequenced and corresponded to cDNAs resulting from alternatively spliced mRNAs. The relative proportions of the three products varied in different cDNA preparations although the shortest product was generally the predominant product (data not shown). Key: M: 100bp marker ladder; T: human testis cDNA; T/10: one-tenth of human testis cDNA loaded in previous lane. C. The structure of transcripts derived from the human gene was analyzed by RT-PCR yielding three mRNAs that result from alternative splicing. The exon distribution of each mRNA is indicated (exon 1 is non-coding and was omitted) along with the size of the predicted protein. This protein has now been named KIF27 with a prefix to indicate the species and a letter to indicate the splice variant. Kinesins are characterized by a ~300 amino acid motor domain that is located at the N-terminus of KIF27. The N-5 subgroup of kinesins also possesses a characteristic neck domain at the C-terminal end of the motor domain. A comparison of the KIF27 neck domain with the corresponding domain of KIF4 is shown, with conserved amino acids indicated between the sequences and in upper case letters within each sequence. Key: φ: hydrophobic residue.

Alignment of the motor domain of KIF27 and other members of kinesin N-5 subgroup. The motor domain of human KIF27 was aligned with the corresponding domains of human KIF4A (NP_046332) and mouse KIF21A (NP_057914) and KIF21B (NP_064346). The motor and neck domains are underlined in red and blue, respectively. The sequence of each protein is shown and blocks of similarity are shaded.

Alignment of human and macaque 9004 sequences. Clone 9004 was found to display significant similarity to cDNAs derived from human (Hs) and macaque (Mf). The human protein is the predicted translation product of AL832216 and the macaque protein is the predicted product of AB074456. The complete human cDNA mapped to a 15 exon gene on chromosome 1. Only eleven of these fifteen exons could be detected in the mouse and rat genome sequences (not shown). The sequence corresponding to the hamster clone is underlined.

Alignment of human, macaque and mouse 15014 sequences. The sequences of human (Hs), mouse (Mm) and macaque (Mf) 15014 orthologues were compared in Align X. The human sequence was derived from XP_089976 (LOC159989). The mouse sequence was obtained from a combination of XP_134746 (LOC234964), a hypothetical protein predicted from the NCBI annotation process along with additional exons predicted from comparisons between the human cDNA and the mouse genomic sequence. The macaque sequence was obtained from Genbank entry BAB63125. Gaps in the human sequence indicate the likely presence of alternative exo ns that may undergo alternative splicing. This hypothesis was supported by exon mapping in both the mouse and rat genome sequences (data not shown). We were unable to detect additional exons in the mouse genomic sequence that might encode the N-terminal sequences in the human protein sequence. The similarity to the hamster clone is underlined.

Alignment of human and mouse 15037 sequences. The human (Hs) sequence is based partly on cDNA sequences AL834368, AF273054 and AK057508 with 5' coding sequences based on conservation to mouse cDNA and genomic sequences. The mouse (Mm) sequence is derived from oocyte-testis gene 1 (Otg1, NM_170757). The region of similarity to the hamster clone is underlined.

Comparison of SMYD2 to members of the SMYD1/BOP family of chromatin remodeling proteins. The human SMYD2 sequence was aligned with the sequence of mouse BOP1 (AAC53021) and chicken BOP1 (AAL31880). The S-ET and MYND domains and the Cys-rich region are labelled as in figure 12. Conserved residues are shaded and the sequences lacking from the BOP2 splice variants in mouse and chicken are indicated with a double green underline.