This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Farber, S. A. Articles by Halpern, M. E. Search for Related Content PubMed PubMed Citation Articles by Farber, S. A. Articles by Halpern, M. E. Social Bookmarking                   What's this?

Letter

The Zebrafish Annexin Gene Family

Department of Embryology, Carnegie Institution of Washington, Baltimore, Maryland 21210, USA

	ABSTRACT

Top ABSTRACT RESULTS DISCUSSION METHODS REFERENCES WEB SITE REFERENCES

 
The Annexins (ANXs) are a family of calcium- and phospholipid-binding proteins that have been implicated in many cellular processes, including channel formation, membrane fusion, vesicle transport, and regulation of phospholipase A₂ activity. As a first step toward understanding in vivo function, we have cloned 11 zebrafish anx genes. Four genes (anx1a, anx2a, anx5,and anx11a) were identified by screening a zebrafish cDNA library with a Xenopus anx2 fragment. For these genes, full-length cDNA sequences were used to cluster 212 EST sequences generated by the Zebrafish Genome Resources Project. The EST analysis revealed seven additional anx genes that were subsequently cloned. The genetic map positions of all 11 genes were determined by using a zebrafish radiation hybrid panel. Sequence and syntenic relationships between zebrafish and human genes indicate that the 11 genes represent orthologs of human anx1,2,4,5,6,11,13,and suggest that several zebrafish anx genes resulted from duplications that arose after divergence of the zebrafish and mammalian genomes. Zebrafish anx genes are expressed in a wide range of tissues during embryonic and larval stages. Analysis of the expression patterns of duplicated genes revealed both redundancy and divergence, with the most similar genes having almost identical tissue-specific patterns of expression and with less similar duplicates showing no overlap. The differences in gene expression of recently duplicated anx genes could explain why highly related paralogs were maintained in the genome and did not rapidly become pseudogenes.

View larger version (9K): [in this window] [in a new window]   Figure 1 Canonical ANX structure. The N-terminal domains of ANX paralogs are highly variable, with the exception of highly conserved phosphorylation domains. At least two of the four ANX repeats are required to bind Ca2+ and phospholipid.

The function of this evolutionarily conserved family of proteins remains poorly understood. Studies on cultured cells suggest that ANXs function in a broad range of physiological processes (for a review, see Seaton 1996). For example, it has long been known that ANXs inhibit phospholipases in vitro (Huang et al. 1986; Davidson et al. 1987, 1990), and more recently, two ANXs (I and V) have been shown to inhibit phospholipase A2 in living cells (Kim et al. 1994; Croxtal et al. 1996). ANXs can also form volume-activated chloride currents (Nilius et al. 1996) and playa role in membrane fusion and vesicle transport (Bandorowicz-Pikula and Pikula 1998). During exocytosis, ANXs promote the aggregation of phospholipid membranes (Emans et al. 1993; Konig et al. 1998; Raynor et al. 1999), whereas arachidonic acid, which is released from membrane phospholipids by phospholipase A2 activity, promotes membrane fusion (Creutz 1981).

A number of studies have examined the evolutionary history of this gene family(Morgan and Fernandez 1997). Analyses of the gene structure of numerous ANXs have revealed a highly conserved intron—exon organization (for a review, see Smith and Moss 1994). Phylogenetic studies indicate that vertebrate ANX paralogs emanated from a common ancestor 800 million years ago and are now dispersed throughout the genome (Morgan and Fernandez 1997). The observation that ANXs are evolving at different rates suggests that paralogs have nonoverlapping functions (Morgan and Fernandez 1997; Morgan and Pilar Fernandez 1997; Morgan et al. 1999). The zebrafish (Danio rerio) genome has been subject to a round of duplication some 100–400 million years ago, after the divergence of fish and mammalian ancestors, resulting in an 30% increase in the total number of genes (Postlethwait et al. 1999). Thus, the zebrafish provides the opportunity to examine why duplicated genes remain fixed in the genome as opposed to becoming pseudo genes.

This report describes the cloning and sequencing of zebrafish orthologs to vertebrate ANXs 1, 2, 4, 5, 6, 11, and 13. Expression of the anx genes during embryogenesis was examined by in situ hybridization not only to characterize each zebrafish ANX, but to compare the expression of recently duplicated gene paralogs. The nonoverlapping expression of duplicated genes suggests that the divergence of tissue specific functions drives the continued presence of ANX paralogs in the zebrafish genome.

	RESULTS

Top ABSTRACT RESULTS DISCUSSION METHODS REFERENCES WEB SITE REFERENCES

 
Four zebrafish anx genes (anx1a, anx2a, anx5, and anx11a) were identified by low-stringency hybridization of a postsomite cDNA library by using a Xenopus laevis anx2 probe. From comparisons with these genes, we identified seven additional anx genes by BLAST searches of zebrafish ESTs (see Table 1). Sequence alignments of the 11 zebrafish anx genes with mammalian anx genes revealed that three zebrafish anx genes were homologous with human anx1, two were homologous with human anx2, and two were homologous with human anx11 (Fig. 2).

View this table: [in this window] [in a new window]   Table 1. Zebrafish Annexin ESTs

View larger version (37K): [in this window] [in a new window]   Figure 2 Comparisons of zebrafish ANX amino acid sequences. The corresponding proteins were aligned by using CLUSTALW, revealing significant divergence in their N-terminal domains as opposed to the first core repeat.

The predicted zebrafish ANX1b and ANX1c proteins are almost identical in their core repeat domain (97% identity in the first repeat), yet highly divergent in the N-terminal domain (33% identity). Only ANX1c contains the typical conserved EGF receptor and PKC phosphorylation sites observed in the N-terminal regions of mammalian ANX1 proteins (Fig. 2). Similarly, the predicted ANX2a and ANX2b proteins show higher identity in the repeat domain than in the N-terminal domain even though both paralogs contain the conserved N-terminal PKC phosphorylation site (Fig. 2). The ANX11 paralogs contain N termini rich in glycine, glutamine, and proline, characterized by a GYPPQPG repeat that is similar to that found in mammalian ANX11 proteins (Tokumitsu et al. 1992). As is the case for the other duplicated genes, the N-terminal regions of ANX11a and ANX11b are more divergent than are the repeat domains.

Despite the prevailing view that the N-terminal regions can be used to identify specific ANX family members, our analysis indicated that only the amino acids in the repeat domain were useful for phylogenetic analysis. With the exception of ANX4, all the zebrafish ANX N-terminal domains show little or no overall homology with their mammalian counterparts. For example, in the N-terminal region of ANX5, only the GTV motif is conserved between species (zebrafish, medaka, human, mouse, and chick), whereas most of the first repeat sequence is identical. The lack of homology in the N-terminal region of even closely related ANXs indicates that comparisons of this region would be less useful in elucidating the evolutionary relationships between fish ANXs. In contrast, the slow rate of mutation accumulation in the repeat region makes it useful for locating sequence similarities.

The phylogenetic tree of zebrafish, medaka (Oryzias latipes), pufferfish (Fugu rubripes), and human ANXs constructed by using only the amino acid sequence of the first ANX repeat reveals a close relationship between several pairs of duplicated zebrafish genes (Fig. 3). The anx1b and anx1c genes are the most closely related of all the duplicated paralogs. The phylogenetic analysis also suggests that we have identified thologs to each of the four previously identified medaka ANXs. Medaka Max 1 is a likely ortholog of zebrafish ANX4, Max 2 is an ortholog of ANX5, Max 3 is an ortholog of ANX1, and Max 4 is an ortholog of ANX11.

View larger version (25K): [in this window] [in a new window]   Figure 3 Phylogenetic relationships of the human, zebrafish, pufferfish (Fugu rubripes), and medaka (Oryzias latipes) ANXs. Medaka ANX sequences were obtained from GenBank: max1, CAA72125 [GenBank] ; max2, CAA72123 [GenBank] ; max3, CAA72124 [GenBank] ; max4, CAA72122 [GenBank] .

To determine the timing of the duplication event observed in zebrafish, we searched the public databases for additional medaka anx sequences and identified 13 ESTs (data not shown). All but one of these new sequences cluster with the four previously published medaka anx genes (Max1 through Max4) already represented on the phylogenetic tree. The remaining clone was not a recent duplicate of these previously identified genes. Thus, the evolutionary relationship between the duplicated genes in these two species was not compared because of the paucity of medaka EST sequences. BLAST searches of fugu ESTs identified orthologs for zebrafish ANXs. Interestingly, phylogenetic analysis of these data (Fig. 3) suggests that fugu ANX1 and ANX2 paralogs were also duplicated, indicating that the duplication occurred in at least subgroup of the Teleostei (Clupeocephala).

Syntenic Relationships Between Human and Zebrafish anx Genes
Because the N-terminal sequences of fish ANXs are considerably different from their mammalian counterparts, it is conceivable that zebrafish anx genes could be misidentified. confirm that the 11 zebrafish anx genes were correctly assigned, we mapped the chromosomal position of each gene and, on the basis of the surrounding mapped genes, determined whether the given zebrafish anx gene showed conserved syntenies with its putative human ortholog. Ten of the 11 anx genes mapped to syntenic clusters containing the human ortholog (Fig. 4A-CFig. 4D-F). For example, zebrafish anx5 is surrounded by genes (mel1ar, fga, and sc4mol) that also map near the human anx5 ortholog (Fig. 4D), indicating that the zebrafish gene was correctly identified.

View larger version (44K): [in this window] [in a new window]   Figure 4 Map positions of human and zebrafish gene orthologs.

The mapping data for the anx1 genes indicate that three genes lie on zebrafish linkage group (LG) 5. They probably arose by tandem duplication (Fig. 4A) because anx1b and anx1c map very close to each other (e.g., beyond the resolution of the radiation hybrid panel). Zebrafish anx1c/anx1b are closely linked to the zebrafish trk gene; this anx1/trk syntenic cluster is preserved in humans (Fig. 4A). Of the ANX1 proteins, only ANX1c has both the conserved tyrosine and threonine phosphorylation sites present in the human protein (Fig. 2), suggesting that anx1c is the most closely related to human anx1.

Mapping data for the anx2 genes reveal the relationship between zebrafish and human chromosomes. The region containing anx2a is syntenic with a region of human chromo-some 15 that contains anx2 (Fig. 4C). However, both zebrafish anx2 genes are surrounded by orthologous genes that map to either human chromosome 11 or 15, suggesting that these clusters of human genes once resided on a single ancestral chromosome that split after divergence of mammalian and zebrafish genomes. The mapping results for zebrafish anx11 genes are more clear-cut, indicating that both paralogs are surrounded by genes with human orthologs that map to the same region on human chromosome 10 (Fig. 4F). These mapping results unequivocally relate mammalian anx genes to those found in zebrafish.

Expression of Zebrafish Genes
The temporal expression patterns of zebrafish anx genes were examined by whole-mount in situ hybridization and, in some instances, validated by Northern analysis. The three anx1 genes are primarily expressed in the outermost embryonic cells, known as the enveloping layer (Kimmel et al. 1995), during the period (4 to 9 h postfertilization [hpf]) when this layer is dividing and migrating over the large yolk cell in a process termed epiboly(Fig. 5A—C). After epiboly is completed and somatogenesis begins, the expression of all three anx1 genes is reduced, as exemplified by expression of anx1a at the midsomite stage (Fig. 5D) and by Northern analysis (Fig. 5K). By 24 hpf, anx1 transcripts increased primarily in the outer ectoderm, with the expression level of anx1a being the highest (Fig. 5E). It is only at this stage that divergence between the paralogous anx1 genes is observed. Not only is the expression of anx1b and anx1c reduced compared with anx1a, but anx1c is selectively expressed in a small group of cells close to the yolk and posterior to the eyes (Fig. 5F,G). After 80 hpf, all three anx genes were expressed in gill arches, epithelium, and fin buds (Fig. 5H,I), but only anx1a was expressed in the cells surrounding the swim bladder (Fig. 5I).

View larger version (95K): [in this window] [in a new window]   Figure 5 Expression of anx1 paralogs. (A—C) Lateral views of 70% epiboly stage embryos revealing enveloping layer expression of anx1a (A), anx1b (B), and anx1b (C). (D) anx1a expression is markedly reduced by the 12-somite stage and then reappears in the periderm (E) at 24 hpf. (F) Periderm expression of anx1c at 24 hpf. Arrowhead indicates a domain of anx1c expression not found with anx1a and anx1b. (G) A higher magnification dorsal view of the anx1c unique expression domain. A ventral view(H) and sagittal section (I) of 80-hpf embryos showing anx1a expression in the gills, epithelium, fin buds, and the cells of the swim bladder (arrowhead). (J) At 80 hpf, anx1c was expressed predominately in the epithelium. (K) Embryonic and adult expression of anx1a determined by Northern analysis. Total RNA was isolated from zebrafish embryos at selected stages from the one-cell stage to adult. Scale bar, 200 µm.

The zebrafish anx2 gene paralogs are significantly more divergent in their core domain as compared with anx1b and anx1c paralogs (76% versus 94%; Fig. 2). We also found that the zebrafish anx2 paralogs do not exhibit overlap in their expression patterns (Fig. 6). anx2a was expressed in the notochord (Fig. 6A) and in a subset of enveloping layer cells (Fig. 6B). Notochord expression was detected at the end of epiboly through early somite stages (Fig. 6C,D), and decreased in an anterior to posterior fashion. By the 21-somite stage, anx2a expression was observed only in the caudal-most notochord (Fig. 6E) and, after 24 hpf, in the epithelium and cells around the anus (Fig. 6F). In contrast, anx2b expression was not observed during early embryonic development (data not shown) and was first detected at 48 to 80 hpf in the cells of the intestinal epithelium (Fig. 6G). Transcripts were observed in both the proximal (Fig. 6H) and distal segments of the differentiating intestine (Fig. 6I).

View larger version (78K): [in this window] [in a new window]   Figure 6 Expression of anx2 genes. (A) Dorsal view of a 100% epiboly stage embryo revealing both notochord and enveloping layer expression (arrowhead) of anx2a. (B) Higher magnification view of embryo in A, highlighting enveloping layer expression. (C) A lateral view of notochord expression of anx2a throughout the entire anterior to posterior axis; (D) a higher power lateral view of the same embryo. (E) A lateral view of a 21-somite embryo with only posterior anx2a expression remaining. (F) At 24 hpf, anx2a expression is predominately in the periderm not notochord and then later in the skin (data not shown). (G) anx2b expression in the intestine at 72 hpf (arrowhead). Sagittal sections through both the proximal (H) and distal (I) intestine indicate anx2b expression in the apical brush border of the intestinal epithelium (arrowhead).

The zebrafish anx4 ortholog was expressed during the onset of somatogenesis in the floor plate and in two lateral stripes (Fig. 7A) that most likely correspond to the pronephric duct precursor cells (Drummond et al. 1998). Two-color in situ hybridization with anx4 and draculin, a marker of blood cell precursors (Herbomel et al. 1999), confirmed that the lateral anx4-expressing cells were not hematopoietic precursors (Fig. 7F,G). At the 21-somite stage, anx4 is predominately expressed in floor plate, hypochord, and pronephros (Fig. 7B), a pattern that continues through 24 hpf (Fig. 7C,H). Additionally, at 24 hpf, expression was observed in the otolith, glomerulus, pronephric tubules, and pronephric ducts (Fig. 7D). Tissue-specific anx4 expression was assigned on the basis of its similarity with pax2, a previously established marker of pronephros development (Drummond et al. 1998), which also labels these structures. After 70–80 hpf, anx4 was expressed predominantly in the liver and gall bladder (Fig. 7I,J). The expression of anx4 in the pronephros is consistent with its reported role in regulating Cl conductance in fluid secreting epithelia (Kaetzel et al. 1994).

View larger version (154K): [in this window] [in a new window]   Figure 7 Expression of anx4. (A) Dorsal view of a four-somite embryo revealing anx4 transcripts in the floor plate (arrowhead) and in two lateral stripes (arrow) corresponding to pronephric duct precursor cells. Lateral views of a 21-somite (B) and a 24-hpf (C) embryo with prominent anx4 expression in the floor plate, hypochord, and pronephros. (D) Dorsal view at 24 hpf also shows anx4 expression in the otolith, glomerulus (midline arrowhead), pronephric tubulues, and pronephric ducts. (E) A similar dorsal view illustrating pax2 transcripts in the otolith, pronephric tubulues, and pronephric ducts but not the glomerulus. Hematopoietic precursors were simultaneously visualized in cross sections using one riboprobe for draculin transcripts (red) and another for anx4 transcripts (blue) at both the 20-somite (F) and (G) 24-hpf stages. (H) A lateral view at 24 hpf also localizes anx4 expression to floor plate (arrowhead), hypochord (arrow), and pronephros. Cross sections of 3-hpf larvae reveal anx4 expression predominately in the gall bladder (I) and liver (J). (A—E) Bar, 200 µm; (F—J) bar, 100 µm.

Zebrafish anx5 showed the most restricted pattern of expression during embryonic development. During somatogenesis, transcripts were found in the blood islands (21 somites, Fig. 8A) and at the most anterior tip of the embryo (26 somites; Fig. 8B). Cells in the olfactory placodes also expressed anx5 (Fig. 8C). At 24 hpf, expression was mostly restricted to the olfactory placodes, hatching gland cells, and anus (Fig. 8D). After 80 hpf, expression was maintained in the developing nasal epithelium and in the anus; cells surrounding the swim bladder also expressed anx5. After 120 hpf, the cells surrounding the opening to the mouth expressed anx5 (Fig. 8E), suggesting that these cells were related to the anx5-expressing cells observed in the anterior most region at the 26-somite stage (data not shown). Consistent with the RNA in situ gene expression data, Northern analysis of anx5 transcripts revealed a 2.2-kb message that was initially detected at the 22-somite stage (Fig. 8F). Expression was also detected in RNA from 256 cell embryos, a stage prior to the onset of zygotic transcription, indicating some maternal contribution of anx5 RNA to the embryo.

View larger version (106K): [in this window] [in a new window]   Figure 8 Expression of anx5. (A) Lateral view of a 21-somite embryo revealing anx5 transcripts in blood island (arrowhead). Coronal view of a 26-somite embryo indicates anx5 expression in the most anterior domain (B) and in the olfactory placodes (C). (D) Ventral view of a 24-hpf embryo reveals intense anx5 expression in the olfactory placodes, hatching gland cells, and anus. (E) Sagittal section through an 80-hpf larvae reveals expression in the developing nasal epithelium, anus, and swim bladder. (F) Embryonic and adult expression of anx5 determined by Northern analysis. Total RNA was isolated from zebrafish embryos at selected stages from the one-cell stage to adult. (A—C) Bar, 100 µm; (D, E) bar, 200 µm.

anx11a expression was observed at the end of gastrulation predominantly in the notochord (Fig. 9A,B). Unlike anx2a, which was also expressed in the developing notochord, there was no enveloping layer expression. As somatogenesis advanced, expression was seen to shift from the axial mesoderm to the adaxial mesoderm (cells directly adjacent to the notochord; Fig. 9C). It was previously shown that in the absence of notochord, there is decreased adaxial expression of MyoD, a transcription factor that is known to play a role in myogenesis (Weinberg et al. 1996). To test whether differentiated notochord was similarly required for the adaxial expression of anx11a, no tail embryos, which are known to lack notochord (Halpern et al. 1993), were examined. The transition of anx11a expression to adaxial cells was unaffected in no tail embryos (Fig. 9D). In wild-type embryos, the shift in expression away from the midline to more lateral structures continued as development proceeded, such that by 24 hpf, expression was observed in the paraxial mesoderm, anus, and periderm (Fig. 9E,F). The anx11a paralog, anx11b, was only weakly detected in the notochord of four-somite embryos (data not shown); however, at 24 hpf, anx11b showed strong expression in the periderm similar to anx11a (Fig. 9F). Later in development, anx11a was expressed more intensely in the liver (arrowhead) than in the intestine (arrow; Fig. 9G), the opposite to the expression pattern of anx11b (Fig. 9H). Northern analysis of anx11a expression revealed a 2.2-kb message that was detected from the end of gastrulation to the adult stage (Fig. 9I).

View larger version (81K): [in this window] [in a new window]   Figure 9 Expression of anx11 genes. Dorsal (A) and lateral (B) views of notochord anx11a expression in a four-somite embryo. (C, D) Dorsal views of 15-somite embryos illustrating a shift in anx11a expression from the axial mesoderm to the adaxial mesoderm that is also observed in embryos that lack differentiated notochord (D; no tail-/-). (E) By 24 hpf, anx11a expression was observed in the paraxial mesoderm (arrowhead) and anus (arrow). (F) Lateral views of 24-hpf embryos reveal expression of both anx11a (top) and anx11b (bottom) in the periderm, arrowhead marks the anus. (G, H) Lateral views of 80-hpf larvae indicating greater levels of anx11a transcripts (G) in the liver (arrowhead) than that of anx11b (H). Embryonic and adult expression of anx11a determined by Northern analysis. (I) Total RNA was isolated from zebrafish embryos at selected stages from the one-cell stage to adult. Bar, 100 µm.

The anx6 gene was primarily expressed in somitic mesoderm from the midsomite stages (Fig. 10A) to 24 hpf (Fig. 10B,C). Little expression was detected after 72 hpf (data not shown). anx13 is detected during gastrulation in the notochord and in an area at the anterior-most portion of the embryo in a region called the polster (Fig. 10D,E). At the 21-somite stage, anx13 expression was maintained in the notochord and central nervous system (CNS; Fig. 10F,G). After 24 hpf, expression persisted in much of the CNS (retina, floor plate, etc.) and nonsomitic mesoderm (Fig. 10H,I). At later stages, anx13 was expressed throughout the embryo (data not shown), being the most widely expressed of all the anx genes.

View larger version (61K): [in this window] [in a new window]   Figure 10 Expression of anx6 and anx13 genes. (A) Dorsal view of an 18-somite embryo reveals anx6 expression in somitic mesoderm. This pattern of anx6 expression continues through to 24 hpf, lateral view(B) and dorsal view(C). Dorsal (D) and lateral (E) views of anx13 expression in both the notochord (arrow) and polster (arrowhead) of bud stage embryos. Lateral views of 21-somite (F) and 24-hpf (G) embryos reveal that anx13 expression is maintained in the notochord and expands to the CNS. (H) Dorsal view at 24 hpf indicates anx13 expression in the retina (arrowhead). (I) Lateral view of the tail at this stage indicates notochord, floor plate, and nonsomitic mesoderm. Bar, 100 µm.

	DISCUSSION

Top ABSTRACT RESULTS DISCUSSION METHODS REFERENCES WEB SITE REFERENCES

 
The ANXs comprise a family of Ca²⁺ and phospholipid binding proteins that contain four or eight highly conserved repeating domains of 70 amino acids (Seaton 1996). Data largely from cell culture studies indicate that ANXs are regulated by growth factor—mediated tyrosine phosphorylation (Brugge 1986; Huang et al. 1986) and by PKC phosphorylation (Schlaepfer et al. 1992; Minin et al. 1998; Schmitz-Peiffer et al. 1998). ANXs can inhibit phospholipase A₂ activity (Brugge 1986; Huang et al. 1986; Kim et al. 1994; Croxtal et al. 1996; Mira et al. 1997), form channels (Nilius et al. 1996), and regulate membrane trafficking (Donnelly and Moss 1997; Gerke and Moss 1997; Bandorowicz-Pikula and Pikula 1998; Kobayashi et al. 1998). It is likely ANXs play keyroles in regulating a variety of signal transduction events. ANXs are also ubiquitous in eukaryotic cells and can represent ≥1% of total cell protein (Seaton 1996). Although there is now a body of literature describing the interesting properties of ANXs, their in vivo functions remain elusive.

As a first step toward exploring the function and evolution of the vertebrate anx gene family, we cloned 11 zebrafish anx genes by cDNA library screening and the analysis of zebrafish ESTs deposited in GenBank. In studying ANX evolution, the zebrafish has the advantage that there was a genome-wide duplication some 100 million years ago, well after fish and mammals diverged, followed by selective gene loss (Postlethwait et al. 1999). Thus, it is possible to study how sibling paralogs diverged in both sequence and expression.

Phylogenetic and BLAST analysis of the 11 anx genes indicated that zebrafish contain three paralogs of anx1, and duplicates of anx2 and anx11 (Figs. 2, 3). The prevailing view based on the study of mammalian anx genes is that the N-terminal variable domain acts as a "fingerprint" to identify ANX family members (Moss 1992). The zebrafish data do not support this idea because the N-terminal domains have diverged so considerably that they were of little use in determining the correct ortholog (data not shown). This was in contrast to comparisons using the repeat domain. Once zebrafish genes were assigned to a particular ANX subtype, CLUSTAL alignments of the N-terminal domains with mammalian orthologs often revealed the conservation of key phosphorylation sites. In fact, the only recognizable homologyin the N-terminal regions was in amino acids surrounding well established regulatory domains important for phosphorylation.

To confirm that we had correctly identified a given zebrafish ANX, we localized each gene to the zebrafish genetic map. We were able to establish the identities of 10 of the 11 zebrafish genes by their position within syntenic gene clusters (Fig. 4A-C Fig. 4D-F). Four previously described medaka genes (max1, max2, max3, and max4) were also included in our phylogenetic comparisons. Sequence analysis indicated that medaka ortholog max3 corresponds to zebrafish anx1a, max1 to anx4, max2 to anx5, and max4 to anx11b. Osterloh et al. (1998) observed the variability in the N-terminal portion of the medaka proteins and concluded that max3 and max4 were "novel" ANX family members. However, the zebrafish data clearly illustrates that anx11b (max4) maps to a region syntenic with human anx11. Similarly, the synteny data also reveals that max1 and max2 are orthologs of human anx4 and anx5, respectively. From these findings, a strong argument can be made that sequence comparisons need to be coupled with gene mapping and synteny data to understand the evolution of a complex multigene family such as the ANXs.

Increases in ploidy have been found for many cyprini form fish (Aparicio 1998), a group that includes the zebrafish. Although the zebrafish is diploid, its genome has retained 30% of the duplicated genes that occurred after fish and mammalian ancestors diverged, even though these genes were initially redundant (Amores et al. 1998; Postlethwait et al. 1998; Prince et al. 1998; Force et al. 1999). It is expected that, following duplication, redundant genes accumulate mutations that would rapidly transform one of the gene paralogs into a nonfunctional pseudogene. Although this certainly occurs in many cases, the rate of gene silencing is often much less than predicted by a number of genetic models (Meyer and Schartl 1999). The duplication-degeneration-complementation (DDC) model has been proposed to explain the retention of duplicated genes (Force et al. 1999). This model considers the effect of mutations in noncoding regions, and predicts that if duplicate genes accumulate mutations in regulatory elements, then their expression might no longer overlap. Once expression between recently duplicated gene paralogs is nonredundant, it is more likely that the activity of both genes is now required to maintain the function of the single ancestral ortholog. Wagner (2000) suggests an alternative view that genes with overlapping or partially redundant functions might be maintained to offer protection from deleterious mutations. If one applies such models, then duplicated gene paralogs should have nonoverlapping aspects of their expression, and the combined expression patterns of recently duplicated paralogs should be consistent with the expression of a single mammalian ortholog.

Given the high degree of nucleotide identity and their proximity in the genome, anx1b and anx1c most likely arose from a single gene that was subject to a more recent tandem duplication event (Fig. 4A). Even though these paralogs are 94% identical in the first repeat, they have diverged extensively in their N-terminal domains (Fig. 2). This pattern was true for the other duplicated genes as well. Smith and Moss (1994) have argued that because these domains are "unique," they likely mediate the specific functions of each anx gene. If the accepted view that the ANX N-terminal domains defines ANX function holds, then these duplicated paralogs have evolved very different functions. Whether this hypothesis is correct must await direct functional analysis of the sibling genes.

An analysis of the three zebrafish anx1 genes revealed that the ancestral anx1 gene common to both fish and mammals was duplicated most likely as a result of the predicted genome duplication event >100 million years ago. This resulted in anx1a near Z20915 [GenBank] and a second anx1 gene near marker coe2. This second anx1 gene was more recently duplicated to yield an additional anx gene at the coe2 locus as evidenced by the high degree of sequence homology between anx1b and anx1c paralogs (Figs. 2A, 4A). Early in development, gene expression of all three anx1 paralogs was largely overlapping; however, by 24 hpf, differences emerged. Only anx1c was expressed in the anterior domain near the eyes and close to the yolk. However, it is possible that there are significant differences in expression levels of each anx1 ortholog (as suggested by the time needed for staining; data not shown), something that can be more accurately quantified in further studies using quantitative reverse transcription—polymerase chain reaction (RT-PCR).

The overlap in expression of the anx2 paralogs was significantly different from that observed for the anx1 genes. The anx2 paralogs had no detectable overlap in expression, in that anx2b transcripts were not detected until the onset of intestinal development (48 hpf), whereas anx2a was expressed during gastrulation in the notochord and later in the periderm (Fig. 6). Given that mammalian anx2 is expressed in both the skin and intestinal epithelium (Bastian et al. 1993; Ma and Ozers 1996; Munz et al. 1997; Dreier et al. 1998; Massey et al. 1998), it is likely that the expression of both zebrafish anx2 paralogs is required to provide the equivalent function of the human ortholog. Consistent with its selective role in intestinal physiology, preliminary data indicate that disruption of anx2b translation by antisense injections at the one-cell stage profoundly affects cholesterol uptake, whereas disruption of anx2a does not (E. Smart, R. DeRose, and S. Farber, in prep.). A major effort to identify the specific amino acids mediating this effect is underway by examining the differences between the two zebrafish ANX2 paralogs identified in this study and by comparison to murine ANX2.

As was the case with anx2, anx11 paralogs exhibited significant differences in expression. Only anx11a was expressed during gastrulation (notochord) and somatogenesis (adaxial and paraxial mesoderm), whereas periderm expression was observed for both paralogs at 24 hpf (Fig. 9A—F) After 80 hpf, anx11a was expressed more intensely in liver, whereas anx11b was greater in intestine (Fig. 9G,H). The observation that duplicated zebrafish genes often have nonoverlapping expression patterns was also observed in a recent study of Na, KATPases (Rajarao et al. 2001).

The observation that the evolutionary divergence between anx2 paralogs is clearly greater than that observed between anx1b and anx1c paralogs not only indicates that the anx1 gene duplication was a more recent event but illustrates how gene paralogs evolve. These studies have enabled us to focus our efforts on elucidating the function of both ANX2 and ANX4. Interestingly, ANX4 is expressed in the identical tissues (floor plate and kidney) of both zebrafish and mouse, in which analysis of the "knock out" phenotype is underway (J. Dedman, pers. comm.). Work is ongoing to compare the phenotypes observed with zebrafish ANX4 "morphants" with those observed in the mouse knock outs. With the availability of antisense techniques for the zebrafish, such as the injection of morpholino oligonucleotides (Nasevicius and Ekker 2000), it will be possible to assess further the redundant and nonoverlapping functions of particular ANXs.

	METHODS

Top ABSTRACT RESULTS DISCUSSION METHODS REFERENCES WEB SITE REFERENCES

 
Zebrafish
Methods for breeding and raising zebrafish were followed as described (Westerfield 1995). Embryos were obtained from natural matings of wild-type (Oregon, AB) fish and staged according to criteria previously outlined (Kimmel et al. 1995) and by hpf.

Cloning of Zebrafish anx Genes
PCR was used to amplifya 560-bp fragment of Xenopus laevis anx2 from a random primed tadpole library(gift of D. Brown, Carnegie Institution of Washington, Maryland). Primer sequences were 5'-GAGCTGAAGG CTTCAATG-3' (forward) and 5'-GTCTAACTCACTTCGTGAAAC-3' (reverse). A ³²P-labeled DNA probe, prepared by random priming of the gel-purified amplification product (Prime-It II; Stratagene), was used to screen a zebrafish postsomite stage cDNA library(gift of D. Grunwald, University of Utah Medical School, Utah) at low stringency(Sambrook et al. 1989). The protein sequences of the four zebrafish ANXs obtained from this screen were used for BLAST searches (Altschul et al. 1994) of >73,000 zebrafish EST sequences generated by the Zebrafish Genome Resources Project. The 212 anx EST sequences identified were clustered by using Assembly LIGN software (Oxford Molecular Group). From this analysis, some unique clones were obtained commercially(Research Genetics) for complete DNA sequencing.

Seven additional anx clones were identified from the EST analysis. Two potential cloning artifacts were observed in commercially obtained clones. In the case of anx2b (clone fb57f04), the 5' end was disrupted by a repetitive element. The correct missing 5' sequence was obtained by using RACE (Frohman et al. 1988). The primer sequences were 5'-CTCTTTATTGCGAGAACACACCAT-3', 5'-CATTCAACTAGT GTGAGGAAGG-3', and 5'-CTCTCTCCCCTCCATAACTCA-3'.

The commercially obtained anx13 clone (fb40a08) contained a shorter-than-expected open reading frame that appeared to be the result of a frame-shift mutation. To determine the correct anx13 sequence, PCR was used to amplify a fragment from genomic DNA to find if it contained the frame shift (primer sequences, 5'-GGAGCCGGAACCGAT GAAGAC-3' and 5'-GGAGGCGCTTG AAATCGCCTCCG-3'). After PCR, the fragment was gel purified and directly sequenced to reveal that the anx13 frame-shift mutation was a cloning artifact.

Sequences of all zebrafish anx genes described in this paper have been deposited in GenBank. The accession nos. are as follows: anx1a, AY178793 [GenBank] ; anx1b, AY178794 [GenBank] ; anx1c, AY178795 [GenBank] ; anx2a, AY178796 [GenBank] ; anx2b, AY178797 [GenBank] ; anx4, AY178798 [GenBank] ; anx5, AY178799 [GenBank] ; anx6, AY178800 [GenBank] ; anx11a, AY178801 [GenBank] ; anx11b, AY178802 [GenBank] ; and anx13, AY178803 [GenBank] .

anx Gene Evolution
A phylogenetic trees showing the evolutionary relationship of the anx genes was produced using MacVector software (Accelrys). The final tree was determined by using UPGMA with ties resolved randomly by using 1000 bootstrap replications. Percentages represent the occurrence of the nodes. Trees were calculated from comparisons of the first 73–83 amino acids of the first ANX repeat.

Mapping and Syntenic Analysis
Zebrafish anxs were mapped using the LN54 Radiation Hybrid Panel as previously described (Hukriede et al. 1999). Primers for each gene are noted in Table 2. To determine syntenic relationships between zebrafish and human genomes, mapped zebrafish genes flanking a particular zebrafish anx gene were identified (maps used were Mother of Pearl Meiotic Map, GAT Meiotic Map, LN54 Radiation Hybrid Map; for map data, see http://zfin.org/cgi-bin/webdriver?Mlval=aa-refcrosslist.apg). The protein sequences of the flanking genes were used to identify human orthologs by BLAST searches. The chromosomal locations of human orthologs were determined by searching either the Human Gene Map99 Project or the Human Gene Database.

View this table: [in this window] [in a new window]   Table 2. Mapping Primers

To confirm that the mapping of anx1b and anx1c to the same locus was not due to a PCR artifact (primers for one paralog actually amplify the other paralog), two sets of primers were used for each gene (Table 1), in which one set was designed in a region of the 3'UTR that was not similar between the sibling genes. Furthermore, a number of PCR products from the mapping reactions were cut from the mapping gel and directly sequenced, confirming that primers for anx1b were not amplifying anx1c and vice versa (data not shown).

Northern Analysis
For some anx genes (anx1a, anx2a, anx5, and anx11a), mRNA levels were detected by Northern analysis of total RNA (Sambrook et al. 1989) using a ³²P-labeled anx probe prepared by random priming of gel-purified zebrafish anx cDNAs (Prime-It II; Stratagene) and compared with expression observed with RNA in situ hybridization.

RNA In Situ Hybridization
In situ hybridization for anx expression was carried out as previously described (Thisse et al. 1993). Digoxigenin-labeled RNA probes synthesized for each gene were hybridized to embryos or larvae at 50% epiboly, 100% epiboly, tail bud, 1 somite, 5 somites, 15 somites, 20 somites, 24 hpf, 48 hpf, and 72 hpf. For double-labeled in situ hybridizations, probes were synthesized with either digoxigenin- or fluoresce in-conjugated UTP and detected as previously described (Sagerstrom et al. 1996). After visualization, some embryos older than 24 h were refixed in 4% paraformaldehyde and sectioned as described (Westerfield, 1995).

	Acknowledgements

 
S.A.F was supported by an NRSA and Barbara McClintock Fellowship (Carnegie Institution); M.E.H., by a Pew's Scholar's Award. We also appreciated the helpful comments provided by Dr. John Dedman.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.

	Footnotes

 
Sequences of all zebrafish anx genes described in this paper have been deposited in GenBank. The accession nos. are as follows: anx1a, AY178793 [GenBank] ; anx1b, AY178794 [GenBank] ; anx1c, AY178795 [GenBank] ; anx2a, AY178796 [GenBank] ; anx2b, AY178797 [GenBank] ; anx4, AY178798 [GenBank] ; anx5, AY178799 [GenBank] ; anx6, AY178800 [GenBank] ; anx11a, AY178801 [GenBank] ; anx11b, AY178802 [GenBank] ; and anx13, AY178803 [GenBank] .

Article and publication are at http://www.genome.org/cgi/doi/10.1101/gr.479603.

¹ Present address: Department of Microbiology and Immunology, Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, PA 19107, USA.

² Corresponding author.
E-MAIL sfarber{at}lac.jci.tju.edu; FAX (215) 923-7144.

	REFERENCES

Top ABSTRACT RESULTS DISCUSSION METHODS REFERENCES WEB SITE REFERENCES

 

Altschul, S., Boguski, M., Gish, W., and Wootton, J. 1994. Issues in searching molecular sequence databases. Nat. Genet. 6: 119–129.[CrossRef][Medline]

Amores, A., Force, A., Yan, Y., Joly, L., Amemiya, C., Fritz, A., Ho, R., Langeland, J., Prince, V., Wang, Y., et al. 1998. Zebrafish hox clusters and vertebrate genome evolution. Science 282: 1711–1714.[Abstract/Free Full Text]

Aparicio, S. 1998. Exploding vertebrate genomes. Nat. Genet. 18: 301–303.[CrossRef][Medline]

Bandorowicz-Pikula, J. and Pikula, S. 1998. Annexins and ATP in membrane traffic: A comparison with membrane fusion machinery. Acta Biochim. Pol. 45: 721–733.[Medline]

Bastian, B., van der Piepen, U., Romisch, J., Paques, E., and Brocker, E. 1993. Localization of annexins in normal and diseased human skin. J. Dermatol. Sci. 6: 225–234.[CrossRef][Medline]

Bonfils, C., Greenwood, M., and Tsang, A. 1994. Expression and characterization of a Dictyostelium discoideum annexin. Mol. Cell. Biochem. 139: 159–166.[CrossRef][Medline]

Brugge, J.S. 1986. The p35/p36 substrates of protein-tyrosine kinases as inhibitors of phospholipase A2. Cell 46: 149–150.[CrossRef][Medline]

Creutz, C.E. 1981. cis-Unsaturated fattyacids induce the fusion of chromaffin granules aggregated by synexin. J. Cell. Biol. 91: 247–256.[Abstract/Free Full Text]

Creutz, C.E., Snyder, S.L., Daigle, S.N., and Redick, J. 1996. Identification, localization, and functional implications of an abundant nematode annexin. J. Cell. Biol. 132: 1079–1092.[Abstract/Free Full Text]

Croxtal, J.D., Newman, S.P., Choudhury, Q., and Flower, R.J. 1996. The concerted regulation of cPLA2, COX2, and lipocortin 1 expression by IL-1 in A549 cells. Biochem. Biophys. Res. Commun. 220: 491–495.[CrossRef][Medline]

Davidson, F.F., Dennis, E.A., Powell, M., and GlenneyJr., J.R. 1987. Inhibition of phospholipase A2 by "lipocortins" and calpactins: An effect of binding to substrate phospholipids. J. Biol. Chem. 262: 1698–1705.[Abstract/Free Full Text]

Davidson, F.F., Lister, M.D., and Dennis, E.A. 1990. Binding and inhibition studies on lipocortins using phosphatidylcholine vesicles and phospholipase A2 from snake venom, pancreas, and a macrophage-like cell line. J. Biol. Chem. 265: 5602–5609.[Abstract/Free Full Text]

de Coupade, C., Gillet, R., Bennoun, M., Briand, P., Russo-Marie, F., and Solito, E. 2000. Annexin 1 expression and phosphorylation are upregulated during liver regeneration and transformation in antithrombin III SV40 T large antigen transgenic mice. Hepatology 31: 371–380.[CrossRef][Medline]

Donnelly, S. and Moss, S. 1997. Annexins in the secretory pathway. Cell. Mol. Life Sci. 53: 533–538.[CrossRef][Medline]

Dreier, R., Schmid, K., Gerke, V., and Riehemann, K. 1998. Differential expression of annexins I, II and IV in human tissues: An immunohistochemical study. Histochem. Cell. Biol. 110: 137–148.[CrossRef][Medline]

Drummond, I., Majumdar, A., Hentschel, H., Elger, M., Solnica, K.L., Schier, A., Neuhauss, S., Stemple, D., Zwartkruis, F., Rangini, Z., et al. 1998. Early development of the zebrafish pronephros and analysis of mutations affecting pronephric function. Development 125: 4655–4667.[Abstract]

Emans, N., Gorvel, J.P., Walter, C., Gerke, V., Kellner, R., Griffiths, G., and Gruenberg, J. 1993. Annexin II is a major component of fusogenic endosomal vesicles. J. Cell. Biol. 120: 1357–1369.[Abstract/Free Full Text]

Fiedler, K. and Simons, K. 1995. Annexin homologues in Giardia lamblia. Trends Biochem. Sci. 20: 177–178.[CrossRef][Medline]

Force, A., Lynch, M., Pickett, F., Amores, A., Yan, Y., and Postlethwait, J. 1999. Preservation of duplicate genes by complementary, degenerative mutations. Genetics 151: 1531–1545.[Abstract/Free Full Text]

Frohman, M., Dush, M., and Martin, G. 1988. Rapid production of full-length cDNAs from rare transcripts: Amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. 85: 8998–9002.[Abstract/Free Full Text]

Gerke, V. and Moss, S. 1997. Annexins and membrane dynamics. Biochim. Biophys. Acta 1357: 129–154.[Medline]

GlenneyJr., J.R. 1985. Phosphorylation of p36 in vitro with pp60src: Regulation by Ca²⁺ and phospholipid. FEBS Lett. 192: 79–82.[CrossRef][Medline]

Halpern, M.E., Ho, R.K., Walker, C., and Kimmel, C.B. 1993. Induction of muscle pioneers and floor plate is distinguished by the zebrafish no tail mutation. Cell 75: 99–111.[CrossRef][Medline]

Herbomel, P., Thisse, B., and Thisse, C. 1999. Ontogenyand behaviour of early macrophages in the zebrafish embryo. Development 126: 3735–3745.[Abstract]

Huang, K.S., Wallner, B.P., Mattaliano, R.J., Tizard, R., and Burne, C. 1986. Two human 35-kd inhibitors of phospholipase A2 are related to substrates of pp60v-src and of the epidermal growth factor receptor/kinase. Cell 46: 191–199.[CrossRef][Medline]

Hukriede, N., Joly, L., Tsang, M., Miles, J., Tellis, P., Epstein, J., Barbazuk, W., Li, F., Paw, B., Postlethwait, J., et al. 1999. Radiation hybrid mapping of the zebrafish genome. Proc. Natl. Acad. Sci. 96: 9745–9750.[Abstract/Free Full Text]

Johnston, P.A., Perin, M.S., Reynolds, G.A., Wasserman, S.A., and Sudhof, T.C. 1990. Two novel annexins from Drosophila melanogaster: Cloning, characterization, and differential expression in development. J. Biol. Chem. 265: 11382–11388.[Abstract/Free Full Text]

Kaetzel, M., Chan, H., Dubinsky, W., Dedman, J., and Nelson, D. 1994. A role for annexin IV in epithelial cell function: Inhibition of calcium-activated chloride conductance. J. Biol. Chem. 269: 5297–5302.[Abstract/Free Full Text]

Kim, K.M., Kim, D.K., Park, Y.M., Kim, C.K., and Na, D.S. 1994. Annexin-I inhibits phospholipase A2 by specific interaction, not by substrate depletion. FEBS Lett. 343: 251–255.[CrossRef][Medline]

Kimmel, C., Ballard, W., Kimmel, S., Ullmann, B., and Schilling, T. 1995. Stages of embryonic development of the zebrafish. Dev. Dyn. 203: 253–310.[Medline]

Kobayashi, T., Gu, F., and Gruenberg, J. 1998. Lipids, lipid domains and lipid-protein interactions in endocytic membrane traffic. Semin. Cell. Dev. Biol. 9: 517–526.[CrossRef][Medline]

Konig, J., Prenen, J., Nilius, B., and Gerke, V. 1998. The annexin II-p11 complex is involved in regulated exocytosis in bovine pulmonary artery endothelial cells. J. Biol. Chem. 273: 19679–19684.[Abstract/Free Full Text]

Ma, A. and Ozers, L. 1996. Annexins I and II show differences in subcellular localization and differentiation-related changes in human epidermal keratinocytes. Arch. Dermatol. Res. 288: 596–603.[Medline]

Massey, H.D., Mayran, N., and Maroux, S. 1998. Polarized localizations of annexins I, II, VI and XIII in epithelial cells of intestinal, hepatic and pancreatic tissues. J. Cell. Sci. 111: 3007–3015.[Abstract]

Meyer, A. and Schartl, M. 1999. Gene and genome duplications in vertebrates: The one-to-four (-to-eight in fish) rule and the evolution of novel gene functions. Curr. Opin. Cell. Biol. 11: 699–704.[CrossRef][Medline]

Minin, A.A., Zemskov, E.A., and Khaidarova, N.V. 1998. Protein kinase C and casein kinase 2 phosphorylate in vitro proteins of the annexin family from eggs of loach Misgurnus fossilis. Biochemistry (Mosc.) 63: 1074–1077.[Medline]

Mira, J.P., Dubois, T., Oudinet, J.P., Lukowski, S., Russo-Marie, F., and Geny, B. 1997. Inhibition of cytosolic phospholipase A2 by annexin V in differentiated permeabilized HL-60 cells: Evidence of crucial importance of domain I type II Ca²⁺-binding site in the mechanism of inhibition. J. Biol. Chem. 272: 10474–10482.[Abstract/Free Full Text]

Morgan, R.O. and Fernandez, M.P. 1997. Annexin gene structures and molecular evolutionary genetics. Cell. Mol. Life Sci. 53: 508–515.[CrossRef][Medline]

Morgan, R.O. and Pilar Fernandez, M. 1997. Distinct annexin subfamilies in plants and protists diverged prior to animal annexins and from a common ancestor. J. Mol. Evol. 44: 178–188.[CrossRef][Medline]

Morgan, R.O., Jenkins, N.A., Gilbert, D.J., Copeland, N.G., Balsara, B.R., Testa, J.R., and Fernandez, M.P. 1999. Novel human and mouse annexin A10 are linked to the genome duplications during early chordate evolution. Genomics 60: 40–49.[CrossRef][Medline]

Moss, S. 1992, The annexins. In: The annexins (ed. S.E. Moss), pp. 1–10. Portland Press, Chapel Hill, NC.

Munz, B., Gerke, V., Gillitzer, R., and Werner, S. 1997. Differential expression of the calpactin I subunits annexin II and p11 in cultured keratinocytes and during wound repair. J. Invest. Dermatol. 108: 307–312.[CrossRef][Medline]

Nasevicius, A. and Ekker, S.C. 2000. Effective targeted gene "knockdown" in zebrafish. Nat. Genet. 26: 216–220.[CrossRef][Medline]

Nilius, B., Gerke, V., Prenen, J., Szucs, G., Heinke, S., Weber, K., and Droogmans, G. 1996. Annexin II modulates volume-activated chloride currents in vascular endothelial cells. J. Biol. Chem. 271: 30631–30636.[Abstract/Free Full Text]

Osterloh, D., Wittbrodt, J., and Gerke, V. 1998. Characterization and developmentally regulated expression of four annexins in the killifish medaka. DNA Cell Biol. 17: 835–847.[Medline]

Postlethwait, J., Yan, Y., Gates, M., Horne, S., Amores, A., Brownlie, A., Donovan, A., Egan, E., Force, A., Gong, Z., et al. 1998. Vertebrate genome evolution and the zebrafish gene map. Nat. Genet. 18: 345–349.[CrossRef][Medline]

Postlethwait, J., Amores, A., Force, A., and Yan, Y. 1999. The zebrafish genome. Methods Cell. Biol. 60: 149–163.[Medline]

Prince, V.E., Joly, L., Ekker, M., and Ho, R.K. 1998. Zebrafish hox genes: Genomic organization and modified colinear expression patterns in the trunk. Development 125: 407–420.[Abstract]

Rajarao, S.J., Canfield, V.A., Mohideen, M.A., Yan, Y.L., Postlethwait, J.H., Cheng, K.C., and Levenson, R. 2001. The repertoire of Na,K-ATPase and subunit genes expressed in the zebrafish, Danio rerio. Genome Res. 11: 1211–1220.[Abstract/Free Full Text]

Raynor, C.M., Wright, J.F., Waisman, D.M., and Pryzdial, E.L. 1999. Annexin II enhances cytomegalo virus binding and fusion to phospholipid membranes. Biochemistry 38: 5089–5095.[CrossRef][Medline]

Sagerstrom, C., Grinbalt, Y., and Sive, H. 1996. Anteroposterior patterning in the zebrafish, Danio rerio: An explant assay reveals inductive and suppressive cell interactions. Development 122: 1873–1883.[Abstract]

Sambrook, J., Fritsch, E., and Maniatis, T., 1989. Molecular cloning: A laboratory manual. Cold Spring Harbor Press, Cold Spring Harbor, NY.

Schlaepfer, D.D., Fisher, D.A., Brandt, M.E., Bode, H.R., Jones, J.M., and Haigler, H.T. 1992. Identification of a novel annexin in Hydra vulgaris: Characterization, cDNA cloning, and protein kinase C phosphorylation of annexin XII. J. Biol. Chem. 267: 9529–9539.[Abst