Zebrafish genetics: harnessing horizontal gene transfer.

The promiscuous spread of Tc1/mariner transposons across species implies that host factors are relatively unimportant for their transposition. Heterologous elements can integrate on expression of the corresponding transposases, an approach that should greatly facilitate genetic analysis in the zebrafish.

Sea urchin early and late H4 histone genes bind a specific transcription factor in a stable preinitiation complex.

Early embryonic H4 (EH4) and H2B (EH2B) and late embryonic H4 (LH4) histone genes were transcribed in vitro in a nuclear extract from hatching blastula embryos of the sea urchin Strongylocentrotus purpuratus. The extract was prepared by slight modifications of the methods of Morris et al. (G. F. Morris, D. H. Price, and W. F. Marzluff, Proc. Natl. Acad. Sci. USA 83:3674-3678, 1986) that have been used to obtain a cell-free transcription system from embryos of the sea urchin Lytechinus variegatus. Achievement of maximum levels of transcription of the EH4 and LH4 genes required a 5- to 10-min preincubation of template with extract in the absence of ribonucleoside triphosphates. This preincubation allowed the formation of a stable complex which was preferentially transcribed compared with a second EH4 or LH4 template that was added 10 min later. Although the EH4 gene inhibited both EH4 and LH4 gene transcription in this assay and although the LH4 gene inhibited both EH4 and LH4 genes, neither of these genes inhibited transcription of the EH2B gene. Preincubation with the EH2B gene had no effect on the transcription of subsequently added EH4 or LH4 genes. Using this template commitment assay, we showed that the site of binding of at least one essential factor required for transcription of both EH4 and LH4 genes was located between positions -102 and -436 relative to the 5' terminus of the EH4 mRNA. Moreover, deletion of this region resulted in a reduction in EH4 gene transcription in vitro. The sea urchin gene-specific trans-acting factors, in the analysis of the cis-acting sequences with which they interact, and in biochemical studies on the formation of stable transcription complexes.

UHF-1, a factor required for maximal transcription of early and late sea urchin histone H4 genes: analysis of promoter-binding sites.

A protein, denoted UHF-1, was found to bind upstream of the transcriptional start site of both the early and late H4 (EH4 and LH4) histone genes of the sea urchin Strongylocentrotus purpuratus. A nuclear extract from hatching blastulae contained proteins that bind to EH4 and LH4 promoter fragments in a band shift assay and produced sharp DNase I footprints upstream of the EH4 gene (from -133 to -106) and the LH4 gene (from -94 to -66). DNase I footprinting performed in the presence of EH4 and LH4 promoter competitor DNAs indicated that UHF-1 binds more strongly to the EH4 site. A sequence match of 11 of 13 nucleotides was found within the two footprinted regions: [sequence: see text]. Methylation interference and footprinting experiments showed that UHF-1 bound to the two sites somewhat differently. DNA-protein UV cross-linking studies indicated that UHF-1 has an electrophoretic mobility on sodium dodecyl sulfate-acrylamide gels of approximately 85 kDa and suggested that additional proteins, specific to each promoter, bind to each site. In vitro and in vivo assays were used to demonstrate that the UHF-1-binding site is essential for maximal transcription of the H4 genes. Deletion of the EH4 footprinted region resulted in a 3-fold decrease in transcription in a nuclear extract and a 2.6-fold decrease in expression in morulae from templates that had been injected into eggs. In the latter case, deletion of the binding site did not grossly disrupt the temporal program of expression from the injected EH4 genes. LH4 templates containing a 10-bp deletion in the consensus region or base substitutions in the footprinted region were transcribed at 14 to 58% of the level of the wild-type LH4 template. UHF-1 is therefore essential for maximal expression of the early and late H4 genes.

Cloning and characterization of vascular endothelial growth factor (VEGF) from zebrafish, Danio rerio.

We have cloned and sequenced a zebrafish (Danio rerio) Vascular Endothelial Growth Factor (vegf) cDNA. It encodes a precursor protein of 188 amino acids with a putative 23 amino acids signal peptide. Sequence comparison analysis indicates that the zebrafish vegf cDNA corresponds to the human VEGF165 isoform and shows about 52% identity to human VEGF165 at the amino acid level. A 2.8 kb vegf message RNA was detected in adult zebrafish by Northern blot analysis. Expression of vegf165 is also detected by RT-PCR in adult fish and throughout the zebrafish embryonic development. Whole mount in situ hybridization of zebrafish embryos indicates strong expression in four areas of the 18-19 h post-fertilization (hpf) embryo: within the anterior central nervous system in the prospective optic stalk, in mesoderm overlapping the bilaterally located merging heart fields, in mesoderm underlying and flanking the hindbrain posterior to rhombomere 4, and in medial regions of the somites. The study of vegf function in zebrafish embryonic vascular development will contribute to our understanding of the mechanisms of vertebrate endothelial cell differentiation and vasculature formation.

Expression of two zebrafish orthodenticle-related genes in the embryonic brain.

To analyze the molecular mechanism of pattern formation in the anteriormost regions of the zebrafish embryo, we isolated two zebrafish sequences, zOtx1 and zOtx2, related to the Drosophila orthodenticle (otd) and two murine Otx genes. zOtx1 and zOtx2 encode predicted gene products which are 82% and 94% identical to the corresponding mouse proteins. Transcripts of both zebrafish genes appear abruptly at high levels in a triangular patch at the animal pole of the mid-gastrula, a region which contains cells fated to become midbrain and forebrain. Between 9 and 14 h of development, zOtx transcripts disappear from forebrain regions in a manner characteristic for each gene, and from 14 to 24 h, particular regions of the forebrain and midbrain express one or both genes. The posterior limit of expression of both genes in 10-30-h embryos forms a sharp boundary at the posterior border of the midbrain. As in the mouse, the early expression patterns of the zOtx genes are consistent with a role in defining midbrain and forebrain territories. However, there are a number of interesting differences between the forebrain and midbrain regions which express the genes in the two species.

Specific localization of zebrafish hsp90 alpha mRNA to myoD-expressing cells suggests a role for hsp90 alpha during normal muscle development.

Members of the eukaryotic hsp90 family function as important molecular chaperones in the assembly, folding and activation of a select group of cellular signalling molecules and transcription factors. Several of the molecules with which hsp90 interacts, such as the bHLH transcription factor myoD, are known to be important regulators of developmental events in vertebrates. However, little information is available in support of any specific role for hsp90 in developing embryos in vivo. In this study, we provide the first in vivo evidence that the hsp90 alpha gene may play a role in the process of myogenesis. We show that constitutive hsp90 alpha mRNA in zebrafish embryos is restricted primarily to a subset of cells within the somites and pectoral fin buds which also express myoD. Furthermore, expression of the hsp90 alpha gene is down-regulated along with myoD in differentiated muscles of the trunk at a time when levels of mRNA encoding the muscle structural protein alpha-tropomyosin remain high. No hsp90 alpha mRNA is detectable within the CNS at control temperatures. In contrast, heat shock-induced expression of the hsp90 alpha gene occurs throughout the embryo at all stages of development examined. The expression patterns strongly suggest that the hsp90 alpha gene plays a specific role in the normal process of myogenesis in addition to providing protection to all cells of the embryo during periods of environmental stress.

The role of vascular endothelial growth factor (VEGF) in vasculogenesis, angiogenesis, and hematopoiesis in zebrafish development.

Vascular endothelial growth factor (VEGF, VEGF-A), a selective mitogen for endothelial cells is a critical factor for vascular development. Two isoforms that differ in the presence of exons 6 and 7, Vegf(165) and Vegf(121), are the dominant forms expressed in zebrafish embryo. Simultaneous overexpression of both isoforms in the embryo results in increased production of flk1, tie1, scl, and gata1 transcripts, indicating a stimulation of both endothelial and hematopoietic lineages. We also demonstrate that vegf can stimulate hematopoiesis in zebrafish by promoting the formation of terminally differentiated red blood cells. Simultaneous overexpression of both isoforms also causes ectopic vasculature and blood cells in many of the injected embryos as well as pericardial edema in later stage embryos. Overexpression of vegf also resulted in earlier onset of flk1, tie1, scl, and gata1 expression in the embryo, indicating a possible role of vegf in stimulating the differentiation of both vascular and hematopoietic lineages. Co-injection of RNAs for both isoforms results in increased expression of three of these markers over and above that observed when either RNA is singly injected and analysis of vegf expression in the notochord mutants no tail and floating head suggests that the notochord patterns the formation of the dorsal aorta by stimulating adjacent somite cells to express vegf, which in turn functions as a signal in dorsal aorta patterning. Finally, studies of vegf expression in cloche mutant indicate that vegf expression is generally independent of cloche function. These results show that in the zebrafish embryo, vegf can not only stimulate endothelial cell differentiation but also hematopoiesis. Moreover, these effects are most dramatic when both vegf isoforms are co-expressed, indicating a synergistic effect of the expression of the two forms of the VEGF protein.

Cytokeratin staining for intraoperative evaluation of sentinel lymph nodes in patients with invasive lobular carcinoma.

BACKGROUND: Frozen section and intraoperative imprint cytology (IIC(N)) are 2 methods used for intraoperative pathologic assessment of sentinel lymph nodes (SLNs). The SLN evaluation of patients with invasive lobular carcinoma (ILC) results in a relatively high number of false-negative results using either of these methods. The purpose of this study was to evaluate the added benefits that intraoperative immunohistochemical-cytokeratin staining (I(CK-IHC)) can bring to IIC(N) in the evaluation of SLN in patients with ILC. METHODS: A total of 59 breast cancer patients with ILC underwent an SLN biopsy evaluated by our standard IIC(N) assessment in addition to I(CK-IHC). The results of IIC(N) with I(CK-IHC) were compared with the final histopathologic assessment consisting of standard hematoxylin and eosin staining and additional cytokeratin staining of nodes. RESULTS: Intraoperative evaluation of SLN using IIC(N) and I(CK-IHC) correctly diagnosed the nodal status in 45 of 59 (76.3%) patients. On final histopathologic assessment, 31 of 59 (52.5%) patients were found to have positive nodes. Using I(CK-IHC), 17 of these 31 positive cases (54.8%) were detected. Using IIC(N) alone, without the benefit of I(CK-IHC), only 13 of 31 (41.9%) positive cases were detected intraoperatively. CONCLUSIONS: For patients with ILC, I(CK-IHC) staining in addition to IIC(N) improves accuracy over using IIC(N) alone. In this study, I(CK-IHC) staining demonstrated a 12.9% improvement in the detection of SLN metastases in patients with ILC. Cytopathologists should consider employing I(CK-IHC) staining to evaluate the touch-imprint slides of SLN in ILC patients.

Effects of chlordiazepoxide on schedule-induced water and alcohol consumption in the squirrel monkey.

Lever pressing of five squirrel monkeys was maintained by a 3-min fixed-interval schedule of food presentation. 3 monkeys had water concurrently available and, for a second pair of monkeys, initially water, then increasing concentrations of alcohol (1--3% v/v) were present. Substantial amounts of post-pellet drinking occurred with all five monkeys. The amount of water ingested was approximately 100 ml per session, that of 3% alcohol nearly 63 ml. For the monkeys drinking alcohol, increasing concentrations of alcohol decreased both the rate of lever pressing and the volume of fluid consumed. Chlordiazepoxide (1.0--17.0 mg/kg) produced increases in lever pressing and in the schedule-induced consumption of both 3% alcohol and water.

Expression of zTlxA, a Hox11-like gene, in early differentiating embryonic neurons and cranial sensory ganglia of the zebrafish embryo.

We have isolated a cDNA encoding a member of the Tlx/Hox11 family of homeodomain factors from the zebrafish, most closely related to the vertebrate Tlx-1/Hox11 and Tlx-3/Hox11L2 proteins. The gene is expressed in a set of early differentiating neurons that project to a common tract, the lateral longitudinal fascicle. We show that the gene is specifically expressed in spinal cord Rohon Beard neurons, in nucleus of the posterior commissure neurons of the midbrain, in a set of hindbrain neurons that include RoL3 reticulospinal interneurons, and in the trigeminal, statoacoustic, anterior lateral line, glossopharyngeal, and vagal cranial sensory ganglia. Timing of expression of the gene in these neurons correlates with the phase of axonal outgrowth and target innervation. Expression of the gene is also observed in several non-neural tissues, including the pharyngeal arches, budding gill filaments, outgrowing semicircular protrusions in the otic vesicle, and in the pectoral fin buds.

The expression pattern of two zebrafish achaete-scute homolog (ash) genes is altered in the embryonic brain of the cyclops mutant.

Two achaete-scute homolog sequences, Zash-1a and Zash-1b, were isolated from a zebrafish embryonic cDNA library. The Zash-1a cDNA encodes a protein very similar to rat Mash-1 and Xenopus Xash-1, with over 94% identity in the C-terminal three-fourths of all three polypeptides. The Zash-1b cDNA encodes a more distantly related protein, with 80% identity of amino acids to Mash-1 in this part of the sequence. At 24 hr, the Zash-1a transcripts are found in the hindbrain in two bilaterally symmetrical lines of cells which mark the boundary between the alar and basal plates and in rhombomere 1 in ventral cells near the floorplate. The gene is also expressed in particular regions of the telencephalon and diencephalon, in the epiphysis, the ventral tegmentum, the neural retina, and in specific cells in the spinal cord. Zash-1b transcripts are found in the hindbrain in segmentally arranged fan-like groups of cells which are located close to the anterior and posterior boundaries of each of rhombomeres 2-6 and in ventral cells close to the floor plate of most rhombomeres. The gene is also expressed at sites distinct from cells expressing Zash-1a in the tegmentum, diencephalon, telencephalon, and spinal cord. In the mutant cyclops, Zash-1a transcripts are absent from the ventral region of the tegmentum and in the ventral cells of rhombomere 1, while more dorsal expression regions are unaffected. The effects of the mutation on Zash-1b expression, however, are more complex. In the hindbrain, the ventral expression zone of this gene is absent, the more dorsal segmented expression is disorganized, and ectopic expression in the alar plate is observed. A dramatic ectopic expression is also observed in the anterior tegmentum. The cyclops gene, therefore, has both positive and negative effects on the CNS of the wild-type embryo: it is required for activation of both Zash-1a and -1b in particular ventral cells, but it also restricts the expression of Zash-1b in other ventral cells and in some dorsal regions. Zash-1a and -1b gene probes will be extremely useful in the analysis of additional mutations affecting development of the central nervous system in zebrafish embryos.

Length and sequence heterogeneity of the histone gene repeat unit of the sea urchin, S. purpuratus.

Histone gene repeats in S. purpuratus are shown to be of variable length and sequence. Two recombinant plasmids containing the full-length 6.3 kb histone repeat unit are found to differer in length at two sites in the repeating structure and in the occurrence of two restriction enzyme recognition sites. Variation in repeat length is also demonstrated in the unfractionated DNA of five sea urchins and in a sample of DNA enriched for histone gene sequences by density gradient methods. The repeats in each individual are of a very limited number of major classes, which may differ from one another in overall length or in distribution and presence of particular restriction enzyme sites. Variations are found to occur at many regions of the repeat; some have been mapped specifically to spacer regions. Repeats may differ dramatically from individual to individual since there is no one type of repeat class common to all, although the absolute length differences of the repeats that are found are small.

Histone gene transcripts in the cleavage and mesenchyme blastula embryo of the sea urchin, S. purpuratus.

Two distinct populations of histone gene transcripts have been identified in the sea urchin embryo. Both late cleavage and mesenchyme blastula stages contain histone transcripts which hybridize to a full-length histone repeat recombinant DNA, pCO1. The histone RNAs of the two stages, however, are dissimilar in sequence. While the transcripts of the cleavage embryo form well matched hybrids with the plasmid DNA which are relatively resistant to RNAase, the hybrids containing the mesenchyme blastula transcripts melt some 10 degrees C lower and are twice as sensitive to RNAase. Hybridization of the two RNA samples to the Hha I fragments of the histone DNA, or to segments of the histone repeat subcloned in other plasmids, shows that many regions scattered along the repeat are complementary to widely diverged transcripts in the mesenchyme blastula RNA. The two RNA populations consist predominantly of polysomal RNA sequences and are most probably mRNAs for the five histones. The mesenchyme blastula RNA sequences in both S. purpuratus and L. pictus form hybrids with pCO1 DNA that are less stable than those containing L. pictus cleavage RNA, indicating the wide divergence of the two histone RNA populations. The bulk of the histone genes in S. purpuratus appear to be of the type coding for the early mRNAs. Only a small percentage of the several hundred gene copies are candidates for the type coding for the late mRNAs. The melting characteristics of the hybrids and the sensitivity of RNAase provide an assay for the late embryonic histone genes. Of the total RNA labeled during a 10 min pulse in the cleavage embryo, histone transcripts represent approximately 9.7 and 6.5% of the radioactivity in S. purpuratus and L. pictus, respectively. These values fall to 0.57 and 1.4%, respectively, at the mesenchyme blastula stage. Although histone genes are transcribed during these two periods, the type of gene which is active is switched at some point prior to the mesenchyme blastula stage.

Induction of fern spore germination.

Light-induced germination of Anemia spores can be inhibited by AMO-1618, a selective inhibitor of gibberellin biosynthesis. The inhibitor has no effect on gibberellin-induced dark germination and its inhibition of light-induced germination can be reversed by supplying gibberellin. Barley-endosperm bioassay of concentrates of medium in which spores are imbibed in light reveals the presence of substances with gibberellin-like activity; assay of medium from dark-imbibed spores does not. Simultaneous exposure of spores to sub-optimal levels of light and gibberellin leads to additivity of effect on germination level. Uptake of labeled gibberellin by spores in light is similar to that in darkness. The implications of these findings for the light-dependent synthesis of a gibberellin-like germination substance are discussed. The bearing of the observations upon understanding the interaction of light and gibberellins in seeds of higher plants is considered.

Polymorphism and stability in the histone gene cluster of Drosophila melanogaster.

Histone genes in Drosophila melanogaster are organized into repeats of 4.8 and 5.0 kb (Lifton et al., 1978). We find these repeat sizes in every one of the more than 20 Drosophila strains we have examined. Strains differ in the relative amounts of the two repeat types, with ratios varying from 1:1 to 1:4, the 5.0 kb repeat always present in equal to or greater amounts than the 4.8 kb repeat. Restriction enzyme digestion and blotting analysis reveals that the strains also differ in a number of far less abundant fragments containing histone DNA sequences. In the Amherst and Samarkand strains, there are, in addition, many copies of 4.0 and 5.5 kb repeat-like fragments respectively. A series of stocks were made isogenic for single second chromosomes from the Amherst strain. The hybridization patterns of the histone DNA from these stocks containing different Amherst chromosomes are very similar but a number of differences in the minor fragments were seen. The stability but a number of differences in the minor fragments were seen. The stability of the histone locus restriction pattern was tested by following the DNA derived from a single second chromosome of the b Adhn2 pr cn strain over a two year period. The restriction pattern of major and minor bands remained identical. Finally, histone loci distinguishable by their restriction pattern on blots were recombined with visible markers. These chromosome will be useful in tracing the fate of specific histone loci during genetic manipulations.

An inverted sea urchin histone gene sequence with breakpoints between TATA boxes and mRNA cap sites.

A sea urchin histone gene fragment containing inverted regions of the normal repeat has been cloned in pBR322. Restriction enzyme mapping and homoduplex analysis of this fragment indicate that the H1-H4 spacer of one repeat is situated alongside the inverted H2A-H1 spacer of another repeat. The site of the breakpoint has been sequenced and compared with homologous stretches of the normal repeat. The breakpoints in the original duplexes were found to be within 4-6 bp of the H4 mRNA cap site and 8-10 bp of the H1 mRNA cap site in the standard repeat. The breakpoints in both original duplexes contain short direct repeats and an overlap of 3 bases. As this is the first breakpoint resulting in the apposition of inverted sequence to be analyzed at the level of DNA sequence, we speculate whether structural features described here are typical of such rearrangements. The structure observed is consistent with, but does not prove, that the sequence is the endpoint of a true inversion since only one junction has been isolated and characterized.

Sequence, organization and expression of late embryonic H3 and H4 histone genes from the sea urchin, Strongylocentrotus purpuratus.

A gene pair coding for histones H3 and H4 expressed during late embryonic development has been cloned from the S. purpuratus genome. The organization of these genes is similar to the divergently transcribed H3-H4 gene pairs of Lytechinus pictus. Whole genome Southern analysis indicates that most late H3 and H4 genes are organized as pairs in both sea urchin genomes. The nucleotide sequences of late and early S. purpuratus H3 and H4 genes differ in the coding regions by 17.0% and 15.7%, respectively. Although there is little match between early and late genes outside the transcription unit, there are short stretches of homology in the spacers 5' to the S. purpuratus early and late H3 genes and the early and late H4 genes. Sequence comparison of the late H3-H4 S. purpuratus pair with two late H3-H4 pairs from L. pictus reveals additional striking homologies in the intergenic spacer. At least four late H3 and two H4 RNAs are distinguished by hybridization to Northern electroblots with probes derived from the cloned gene pair. Although most of these RNAs appeared to accumulate coordinately, with the most dramatic increase occurring between 13 and 17 hours of development, some differences in timing of appearance of different H3 RNA species are observed.