Evolution of vertebrate E protein transcription factors: comparative analysis of the E protein gene family in Takifugu rubripes and humans
Jun-ichi Hikima1,
Mara L. Lennard1,
Melanie R. Wilson2,
Norman W. Miller2,
L. William Clem2 and
Gregory W. Warr1
1 Marine Biomedicine and Environmental Sciences Center and Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, South Carolina
2 Department of Microbiology, University of Mississippi Medical Center, Jackson, Mississippi
 |
ABSTRACT
|
---|
E proteins are essential for B lymphocyte development and function, including immunoglobulin (Ig) gene rearrangement and expression. Previous studies of B cells in the channel catfish (Ictalurus punctatus) identified E protein homologs that are capable of binding the µE5 motif and driving a strong transcriptional response. There are three E protein genes in mammals, HEB (TCF12), E2A (TCF3), and E2-2 (TCF4). The major expressed E proteins found in catfish B cells are homologs of HEB and of E2A. Here we sought to define the complete family of E protein genes in a teleost fish, Takifugu rubripes, taking advantage of the completed genome sequence. The catfish CFEB (HEB homolog) sequence identified homologous E-protein-encoding sequences in five scaffolds in the Takifugu genome database. Detailed comparative analysis with the human genome revealed the presence of five E protein homologs in Takifugu. Single genes orthologous to HEB and to E2-2 were identified. In contrast, two members of the E2A gene family were identified in Takifugu; one of these shows the alternative processing of transcripts that identifies it as the ortholog of the E12/E47-encoding mammalian E2A gene, whereas the second Takifugu E2A gene has no predicted alternative splice products. A novel fifth E protein gene (EX) was identified in Takifugu. Phylogenetic analysis revealed four E protein branches among vertebrates: EX, E2A, HEB, and E2-2.
comparative immunology; phylogeny; gene structure; catfish
 |
INTRODUCTION
|
---|
TRANSCRIPTIONAL CONTROL of the IGH locus in a teleost fish differs greatly from that described in the mammals. In the channel catfish, Ictalurus punctatus, the major enhancer (Eµ3') lies between the µ and
genes (11), and its function is dependent on transcription factors binding to two variant octamer motifs and to a single µE5 site (4). The prototype µE2 and µE5 E-box motifs (consensus sequence CANNTG) were originally discovered within the mammalian immunoglobulin intronic enhancer (Eµ) by A. Ephrussi, after whom they are named (2, 8). E proteins are class I basic helix-loop-helix (bHLH) transcription factors that are ubiquitously expressed but are nevertheless involved in many cell type-specific functions, including the development of the lymphoid system and the transcriptional control of many T- and B-cell-specific genes.
The mammalian E protein family consists of three members: E2A (with alternative splice products E12 and E47), HEB, and E2-2 (6, 10). In mammals, these proteins play major roles in several essential lymphoid processes such as immunoglobulin and T-cell antigen receptor (TCR) gene rearrangement as well as the expression of activation-induced cytidine deaminase, which mediates somatic hypermutation and class-switch recombination (6, 16). The Id proteins, which are class V bHLH transcription factors, also play an important role as negative regulators of gene transcription driven by E proteins, since they can heterodimerize with E proteins to inhibit their binding to the DNA (13, 19).
In the channel catfish significant effort has been made to understand the nature of the Oct and E protein transcription factors through which the Eµ3' enhancer operates (5, 9, 14, 15). In catfish B cells the major expressed E protein was shown to be an ortholog of HEB. This transcription factor is unique for a HEB factor in that it is expressed as two different isoforms (CFEB1 and 2) that are derived by alternative splicing of a primary transcript (9). Both isoforms are capable of binding to the µE5 motif, forming homo- or heterodimers, and strongly driving transcription from the Eµ3' core enhancer in a µE5-dependent manner (9). Recently, catfish homologs of E2A have been discovered (AY770493 and AY860223). One of these, E2A1, has similar functional properties to CFEB, although it is expressed at lower levels (9a).
Projects designed to investigate gene families, such as the E proteins of catfish, must always consider the question "How do we know that we have identified all the members of the family?" This is a question that can be answered definitively only by analysis of a complete genome sequence. Although a project to sequence the catfish genome has yet to be initiated, several teleost fish genome projects are currently underway, including the Japanese pufferfish (Takifugu rubripes), fresh water pufferfish (Tetraodon nigroviridis), zebrafish (Danio rerio), and medaka (Oryzias latipes). Thus we sought to utilize these resources to investigate the genomic representation and structure of E protein genes in teleost fish. The focus was placed on the Japanese pufferfish (Takifugu), whose compact genome (400 Mb) makes it a powerful model for understanding the structure of genes, the transcripts they encode, and their evolutionary histories (1, 7). Unfortunately, little is known about the function of most proteins encoded by Takifugu genes, due to the lack of a workable in vitro system for conducting functional studies. However, examination of the Takifugu genome can provide phylogenetically important information about the extent, nature, and transcription of gene family members. In addition, the Takifugu genomic resources can be used to identify genes in other teleost species (such as the catfish) where in vitro functional studies can be performed to give valuable information concerning the functional evolution of gene families within the vertebrates.
To this end the study reported here uses knowledge of catfish E proteins to interrogate the Takifugu genome database to define the E protein family members in this teleost fish and to gain insights into the evolution of the vertebrate class I bHLH transcription factors.
 |
MATERIALS AND METHODS
|
---|
Computational identification of Takifugu E protein genes.
The genomic sequences of T. rubripes were obtained from the fugu genomic sequence databases ver. 3.0 at the United States Department of Energy Joint Genome Institute (JGI; http://genome.jgi-psf.org/) formatted for National Institute for Biotechnology Information (NCBI) BLAST analysis. Protein homology searches were performed using the deduced amino acid sequence of catfish CFEB1 (GenBank accession no. AY528668 ) as a probe in the BLAST platform to obtain overlapping sequences (scaffolds) that include putative E protein genes. The analyses of ortholog identification and transcript structures were performed using the Ensembl utilities at http://www.ensembl.org and NCBI Map Viewer at http://www.ncbi.nlm.nih.gov/mapview.
Phylogenetic analysis.
The alignment of full-length E proteins or their bHLH domains was performed using inferred amino acid sequences of catfish CFEB1 and CFEB2 (GenBank accession nos. AY528668 and AY528669), catfish E2A1 (AY770493), catfish E2A2 (AY860223), human E12 (AAA52331), human E47 (S10099), mouse E12 (AAH18260), mouse E47 (AAK18618), hamster E2A (P98180), rat E2A (P21677), chicken E12 (NP_989817), chicken E47 (CAE30454), Xenopus E2
(Q01978), Xenopus E12 (S23391), zebrafish E12 (I50518), carp E12 (BAA78382), human HEB (M80627), mouse ALF1A (C45020), mouse ALF1B (S19958), mouse TF12 (NP_035674), rat TF12 (NP_037308), rat REB
(AAB25129), rat REBß (AAB25128), chicken CTF4 (S22611), chicken TF4 (I50636), chicken TF12 (P30985), human E2-2 (NP_003190), mouse TF4 (NP_038713), mouse MITF2A (AAC52414), mouse SEF2 (CAA62868), dog TF4 (P15881), Takifugu S233a (Ensembl transcript ID SINFRUT00000142959), Takifugu S233b (SINFRUT00000169151), Takifugu S233c (SINFRUT00000172919), Takifugu S233d (SINFRUT00000174413), Takifugu S810 (SINFRUT00000133258), Takifugu S1214a (SINFRUT00000132796), Takifugu S1214b (SINFRUT00000132797), Tetraodon E2-2 (GSTENG00019575001), Takifugu S1335a (SINFRUT00000162900), Takifugu S1335b (SINFRUT00000162903), Takifugu S1539a (SINFRUT00000133007), Takifugu S1539b (SINFRUT00000133008), Tetraodon E2A1 (GSTENT00017714001), zebrafish E2Aa (ENSDART00000015857), zebrafish E2Ab (ENSDART00000018197), zebrafish E2Ac (ENSDART00000023718), and zebrafish HEB (ENSDART00000009938 or zgc:85956). The alignment was performed using the Clustal W program in the MegAlign program of DNASTAR software (DNASTAR, Madison, WI) with PAM 250 residue weight table, gap penalty of 10, and gap-length penalty of 10. The alignment was used to generate three types of phylogenetic tree, using the PAUP program v4.0 beta (17). These were 1) a neighbor-joining tree with distance-based methods (with 1,000 bootstrap replicates), 2) a maximum parsimony heuristic tree (with 1,000 bootstrap replicates), and 3) a quartet-puzzling tree with 5,000 puzzling steps. The class I HLH protein, daughterless (Da, Ref. 18) of Drosophila melanogaster was used as an outgroup. Bootstrap values of greater than 50 are shown in support of each node.
 |
RESULTS AND DISCUSSION
|
---|
Identification of Takifugu E proteins.
To search for E protein homologs in the Takifugu genome, the inferred amino acid sequence of CFEB1, the catfish ortholog of HEB (9), was used in the translated BLAST search (tblastn) to interrogate the Takifugu v3.0 sequence using the JGI browser (http://genome.jgi-psf.org/cgi-bin/runAlignment?db=fugu6). Five scaffolds (scaffold_233, scaffold_1335, scaffold_1214, scaffold_810, and scaffold_1539) containing putative E protein gene sequences were identified by this search, and these all have conspicuously conserved bHLH domain sequences (Fig. 1A). A summary of the ortholog predictions (by reciprocal BLAST analysis) of the five probable class I bHLH transcription factor genes of Takifugu is shown in Table 1. Two HEB homologs (scaffold_233 and scaffold_1335), two E2A homologs (scaffold_810 and scaffold_1539), and one E2-2 homolog (scaffold_1214) were predicted (Table 1). Reciprocal BLAST analysis searches the genes of one organism against the genome of a second, repeats the BLAST searches in the opposite direction, and identifies putative orthologs as those genes from the two genomes that show the best combined BLAST scores. However, although these ortholog predictions from reciprocal BLAST analysis served as a useful starting point, the assignment of the individual Takifugu genes as orthologs of specific mammalian E proteins benefited from additional analyses. These analyses combined considerations of alternative RNA processing pathways to generate mature transcripts of these genes (Fig. 2), exon structure comparisons (Figs. 1 and 3), and sequence alignments and phylogenetic tree construction (Figs. 4 and 5). The combination of these analytical approaches identified two of the Takifugu sequences as orthologs of HEB and E2-2, respectively, two as homologs of E2A, and one (scaffold_1335) as a novel class I bHLH gene, which is termed EX.

View larger version (46K):
[in this window]
[in a new window]
|
Fig. 1. Takifugu homologs of the catfish CFEB1. A: Takifugu scaffold sequences were identified with catfish CFEB1 amino acid sequences using a BLAST search of the JGI Fugu v3.0 browser (http://genome.jgi-psf.org/) as described. The shaded boxes indicate the regions of high similarity (hsp scores >80) between CFEB1 and the conceptual translation of the sequences identified within the scaffolds. The basic helix-loop-helix (bHLH) domain of CFEB1 was highly conserved in all five Takifugu E protein genes. B: comparison of the CFEB1 cDNA sequence with the structure of the Takifugu HEB gene ortholog within scaffold_233, showing predicted exons within the CFEB1 sequence. C: table showing the predicted exon correlation and percent identities between the Takifugu HEB ortholog of scaffold_233 and CFEB1.
|
|

View larger version (44K):
[in this window]
[in a new window]
|
Fig. 2. Comparison of the structure of E protein transcripts in Takifugu and human. A: these sequence data were analyzed from all mRNA data including ESTs through Ensembl Genome Browser (http://www.ensembl.org/). The schematics of transcript exon structures were based on the results of the Ensembl Genome Browser for Takifugu or human. Because some of the introns are short, the full number of exons is not always resolved in these diagrams. The arrow at the bottom indicates transcriptional orientation, and the triangles indicate the bHLH domain positions in the E protein transcripts. Asterisks indicate that translation initiation codons remain unidentified. B: Takifugu scaffold number and Human Genome Organization (HUGO) ID number, exon structure, and size of the transcripts analyzed in A.
|
|

View larger version (40K):
[in this window]
[in a new window]
|
Fig. 3. Structure of the human E2A gene compared with two Takifugu E2A genes and their transcripts. A: the structure of the human E2A gene. Exons 17 and 18 are expressed by alternative splicing of the primary transcript to generate mRNAs encoding E12 or E47. B: exon structure of the Takifugu E2A gene encoded on scaffold_1539 related to the structure of its two alternatively processed transcripts, S1539a (SINFRUT00000133007) and S1539b (SINFRUT00000133008). C: exon structure of the Takifugu E2A gene present on scaffold_810 related to its transcript S810 (SINFRUT00000133258), showing that no alternative splicing of the primary transcript occurs.
|
|

View larger version (45K):
[in this window]
[in a new window]
|
Fig. 4. Phylogeny of vertebrate E proteins derived by three tree-building methods. A: neighbor-joining tree based on the distance method (1,000 bootstrap replicates). B: heuristic tree generated by the maximum parsimony method (1,000 bootstrap replicates). C: quartet-puzzling tree (5,000 puzzling steps). Trees are shown with bootstrap values or puzzling support values for each node. E2A, E2-2, and HEB branches are circumscribed with dotted lines. Within the E2A branch, the genes with single spliced transcript and two alternatively spliced transcripts (E12/E47 orthologs) are indicated as groups 1 and 2, respectively. Drosophila Da was used as an outgroup.
|
|

View larger version (86K):
[in this window]
[in a new window]
|
Fig. 5. Sequence alignment of the bHLH domain of vertebrate E proteins. The aligned length of all bHLH domains was 53 amino acids. Amino acid residues that differ from the consensus are highlighted by gray boxes.
|
|
A Takifugu ortholog of HEB.
The sequence in Takifugu scaffold_233 (position 182219 to 220024) showed the longest region(s) of similarity to CFEB1 (Fig. 1A), and the inferred Takifugu gene could easily be aligned with the CFEB1 sequence, allowing putative exons within the CFEB1 sequence to be readily deduced. The results shown in Fig. 1B indicate that 19 exons are predicted to be conserved within the catfish CFEB1 gene. Exon-by-exon comparisons of Takifugu HEB and CFEB1 showed high sequence conservation at the amino acid level, ranging from 53% to 95% (Fig. 1C). Exon 18, in which the bHLH domain is encoded, showed the highest identity (95%). Since CFEB has been identified previously as the catfish ortholog of HEB (9), it was thus concluded that the Takifugu HEB ortholog is present within scaffold_233 (Fig. 1A).
E2A genes in Takifugu: an ortholog and a paralog.
Two of the Takifugu scaffolds (810 and 1539) contained sequences whose ortholog predictions included E12, which is one of the alternative splice products of the mammalian E2A gene (Table 1). Thus these two genes were examined more closely as potential E2A genes. The gene present in scaffold_1539 was particularly interesting since it includes two exons that each encodes apparently complete bHLH domains and which are expressed in two alternatively spliced transcripts of the gene (transcripts 9 and 10 in Fig. 2). Such a pattern where two distinct bHLH-encoding exons are present in the gene and are expressed by alternative splicing in separate transcripts is typical of the mammalian E2A gene (Ref. 12; transcripts 11 and 12 in Fig. 2). Thus the Takifugu gene on scaffold_1539 is very similar to the human E2A gene in both structure and expression (Fig. 3, A and B). Although it is one exon shorter than the human gene, the 3' coding region of the Takifugu and human genes shows remarkable similarity. The two subterminal exons of both genes (exons 17 and 18 in human, exons 16 and 17 in Takifugu) encode bHLH domains that are expressed in two alternatively spliced transcripts (Fig. 3, A and B). Thus the Takifugu gene in scaffold_1539 is identified as the ortholog of mammalian E2A and is referred to as E2A2. The second Takifugu gene (scaffold_810) that showed close similarity to mammalian E2A contains 19 exons, as does the mammalian E2A gene (compare Fig. 3, A and C), but only one of these exons encodes a bHLH domain, and there are no alternatively spliced transcripts (Figs. 2 and 3C). Thus it appears that the Takifugu gene on scaffold_810 is a second E2A gene. This second E2A gene is an ortholog of the previously described catfish E2A1 gene (accession no. AY770493). This second E2A gene identified in teleost fish (Takifugu 810, catfish E2A1) is most readily accounted for as a product of the duplication of the orthologous teleost E2A gene (Takifugu 1539, catfish E2A2), of which it is therefore a paralog.
A Takifugu ortholog of E2-2.
One of the Takifugu sequences (scaffold_1214, Table 1), although clearly a class I bHLH family member, gave no ortholog predictions in reciprocal BLAST analysis (Table 1). However, the ortholog of the Takifugu gene was identified in T. nigroviridis (GSTENG00019575001), and when the T. nigroviridis sequence was searched against the databases, it identified two mammalian E2-2 sequences: ENSG00000141654 from human, and ENSMUSG00000053477 from the mouse. Thus the Takifugu gene 1214 and the T. nigroviridis gene 19575001 are most likely the orthologs of mammalian E2-2.
A novel class I bHLH gene in Takifugu.
Although the sequence on scaffold_1335 was initially identified as a likely HEB gene (Table 1), its transcript structure is very different from that of the identified HEB ortholog (Fig. 2). A sequence alignment focused on the bHLH domains of Takifugu E proteins shows that the bHLH region encoded in scaffold_1335 differs from those of HEB orthologs by two amino acid residues (Fig. 5). These differences in bHLH sequence and transcript structure are, however, insufficient, in themselves, to draw reliable conclusions about the nature of the Takifugu E protein encoded on scaffold_1335. Thus a phylogenetic analysis, described below, was undertaken to help assign more definitively the relationships of this gene, as well as the other four Takifugu E protein genes, identified in this study.
Diversity of E proteins in Takifugu.
The five Takifugu E protein sequences were aligned with other selected vertebrate E proteins, as described in MATERIALS AND METHODS. Three types of phylogenetic trees were generated by the alignment of E protein sequences (Fig. 4, AC). Phylogenetic analysis, using the Drosophila E protein homolog daughterless (Da) as an outgroup, predicted three distinct lineages of vertebrate E proteins, i.e., E2A, HEB, and E2-2 (Fig. 4). However, the relationship between these three major branches is unclear. Maximum parsimony analysis did not resolve the relationship between the three branches (Fig. 4B), and neighbor-joining analysis placed HEB and E2-2 on one branch with a high supporting bootstrap value (Fig. 4A; see also Ref. 9). Quartet puzzling linked E2-2 and E2A in a single branch, although with a low bootstrap value (Fig. 4C). However, from this overall phylogenetic analysis we can draw several conclusions. First, the "non-splicing" form of E2A1 seen in teleost fish has clearly arisen after the separation of the bony fish lineage from that of the tetrapods, perhaps as the outcome of a whole-genome duplication event that occurred early in the evolution of the ray-finned fishes (3). Second, the assignment of the Takifugu 1214 gene as an ortholog of E2-2 is strongly supported by these phylogenetic analyses. Third, the identification of the Takifugu 233 gene as an HEB ortholog is also strongly supported by the analyses. Fourth, the Takifugu 1335 gene, which was identified by BLAST search as an HEB homolog, but which differed from the Takifugu HEB ortholog (233) in its transcript structure and bHLH sequence, occupied two different positions in the phylogenetic analyses. It was basal to the HEB/E2-2 branch in the neighbor-joining analysis (Fig. 4A) and basal to all vertebrate E protein lineages in the maximum parsimony and quartet-puzzling trees(Fig. 4, B and C). However, the basal nodes for the 1335 gene (EX) are poorly supported (highlighted with dotted circles in Fig. 4, AC). Thus, although it is possible that EX arose by gene duplication early in vertebrate evolution, the current analysis is insufficient to strongly support this conclusion.
To summarize the results of this study, and to place the results obtained in Takifugu in context with what is known about T. nigroviridis and catfish, the number and relationships of E proteins in six species of fish and mammals are shown in Table 2. There are a total of five E protein genes in both Takifugu and T. nigroviridis, compared with only three E protein genes in mammals. The difference in number has arisen from an apparent duplication of the E2A gene and the presence of a novel E protein (EX) that appears to have survived (and may even have arisen) only in the teleost lineage. Although the number of E protein genes in catfish is not yet completely known, the finding to date of one HEB ortholog and two E2A genes in catfish suggests that the expansion of the E protein family in the teleost fish is not specific to the tetraodontiforms. Although it is not known what biochemical functions the five Takifugu E proteins identified here may have in driving transcription, the genome database resources available for this species are clearly of great utility in defining the diversity and relationships of homologous genes in other species (such as catfish) where functional studies are feasible.
 |
GRANTS
|
---|
This work was supported by National Science Foundation Award MCB9807531 and by National Institutes of Health Grants R01-GM-62317 and R01-AI-19530.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: G. W. Warr, Hollings Marine Laboratory, 331 Fort Johnson Road, Charleston SC 29412 (E-mail: warrgw{at}musc.edu).
10.1152/physiolgenomics.00312.2004.
 |
REFERENCES
|
---|
- Brenner S, Elgar G, Sandford R, Macrae A, Venkatesh B, and Aparicio S. Characterization of the pufferfish (Fugu) genome as a compact model vertebrate genome. Nature 366: 265268, 1993.[CrossRef][ISI][Medline]
- Calame K and Sen R. Diverse forms of immunoglobulin genes in lower vertebrates. In: Molecular Biology of B Cells, edited by Honjo T, Alt FW, and Neuberger M. San Diego, CA: Elsevier, 2004.
- Christoffels A, Koh EG, Chia JM, Brenner S, Aparicio S, and Venkatesh B. Fugu genome analysis provides evidence for a whole-genome duplication early during the evolution of the ray-finned fishes. Mol Biol Evol 21: 11461151, 2004.[Abstract/Free Full Text]
- Cioffi CC, Middleton DL, Wilson MR, Miller NW, Clem LW, and Warr GW. An IgH enhancer that drives transcription through basic helix-loop-helix and Oct transcription factor binding motifs. Functional analysis of the Eµ3' enhancer of the catfish. J Biol Chem 276: 2782527830, 2001.[Abstract/Free Full Text]
- Cioffi CC, Pollenz RS, Middleton DL, Wilson MR, Miller NW, Clem LW, Warr GW, and Ross DA. Oct2 transcription factor of a teleost fish: activation domains and function from an enhancer. Arch Biochem Biophys 404: 5561, 2002.[CrossRef][ISI][Medline]
- Engel I and Murre C. The function of E- and Id proteins in lymphocyte development. Nat Rev Immunol 1: 193199, 2001.[CrossRef][Medline]
- Elgar G, Sandford R, Aparicio S, Macrae A, Venkatesh B, and Brenner S. Small is beautiful: comparative genomics with the pufferfish (Fugu rubripes). Trends Genet 12: 14550, 1996.[CrossRef][ISI][Medline]
- Ephrussi A, Church GM, Tonegawa S, and Gilbert W. B lineage-specific interactions of an immunoglobulin enhancer with cellular factors in vivo. Science 227: 134140, 1985.[ISI][Medline]
- Hikima J, Cioffi CC, Middleton DL, Wilson MR, Miller NW, Clem LW, and Warr GW. Evolution of transcriptional control of the IgH locus: characterization, expression, and function of TF12/HEB homologs of the catfish. J Immunol 173: 54765484, 2004.[Abstract/Free Full Text]
- Hikima J, Middleton DL, Wilson MR, Miller NW, Clem LW, and Warr GW. Evolution and diversity of E-protein transcription factors: identification, expression, and functional properties of E2A in the channel catfish. Immunogenetics. In press.
- Kee BL, Quong MW, and Murre C. E2A proteins: essential regulators at multiple stages of B-cell development. Immunol Rev 175: 138149, 2000.[CrossRef][ISI][Medline]
- Magor BG, Wilson MR, Miller NW, Clem LW, Middleton DL, and Warr GW. An Ig heavy chain enhancer of the channel catfish Ictalurus punctatus: evolutionary conservation of function but not structure. J Immunol 153: 55565563, 1994.[Abstract/Free Full Text]
- Murre C, McCaw PS, and Baltimore D. A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins. Cell 10: 777783, 1989.
- Nagata Y and Todokoro K. Activation of helix-loop-helix proteins Id1, Id2 and Id3 during neural differentiation. Biochem Biophys Res Commun 199: 13551362, 1994.[CrossRef][ISI][Medline]
- Ross DA, Lyles M, Ledford BE, Magor BG, Wilson MR, Norman WM, Clem LW, Middleton DL, and Warr GW. Catfish Oct2 binding affinity and functional preference for octamer motifs, and interaction with OBF-1. Dev Comp Immunol 23: 199211, 1999.[CrossRef][ISI][Medline]
- Ross DA, Magor BG, Middleton DL, Wilson MR, Norman WM, Clem LW, and Warr GW. Characterization of Oct2 from the channel catfish: functional preference for a variant octamer motif. J Immunol 160: 38743882, 1998.[Abstract/Free Full Text]
- Sayegh CE, Quong MW, Agata Y, and Murre C. E-proteins directly regulate expression of activation-induced deaminase in mature B cells. Nat Immunol 4: 586593, 2003.[CrossRef][ISI][Medline]
- Swofford DL. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Sunderland, MA: Sinauer, 2002, version 4.
- Wubelbeck C, Fromental-Ramain C, and Campos-Ortega JA. The HLH domain of a zebrafish HE12 homologue can partially substitute for functions of the HLH domain of Drosophila DAUGHTERLESS. Mech Dev 46: 7385, 1994.[CrossRef][ISI][Medline]
- Yokota Y. Id and development. Oncogene 20: 82908298, 2001.[CrossRef][ISI][Medline]
Copyright © 2005 by the American Physiological Society.