Construction and recovery of viable retroviral genomes carrying a bacterial suppressor transfer RNA gene.

The integration of retroviral genomes into cellular DNA can induce mutations by altering the expression of nearby cellular genes and can serve to identify the gene affected. The construction of a retrovirus that stably carries a suppressor transfer RNA gene from Escherichia coli has allowed facile recovery of the viral genome in vectors marked with amber mutations. This virus can be used for rapid isolation of cellular sequences at the site of proviral insertion.

A mouse homeo box gene is expressed in spermatocytes and embryos.

The MH-3 gene, which contains a homeo box that is expressed specifically in the adult testis, was identified and mapped to mouse chromosome 6. By means of in situ hybridization with adult testis sections and Northern blot hybridization with testis RNA from prepuberal mice and from Sl/Sld mutant mice, it was demonstrated that this gene is expressed in male germ cells during late meiosis. In the embryo, MH-3 transcripts were present at day 11.5 post coitum, a stage in mouse development when gonadal differentiation has not yet occurred. The MH-3 gene may have functions in spermatogenesis and embryogenesis.

Insertion mutagenesis of embryonal carcinoma cells by retroviruses.

Mutagenesis was studied in cultured F9 embryonal carcinoma cells infected with a variant of Moloney murine leukemia virus. Proviral insertion induced the inactivation of the hypoxanthine phosphoribosyltransferase locus, and the virus was used to isolate the mutated genes rapidly. Mutagenesis by these methods may be useful for the genetic dissection of the various mammalian cell phenotypes.

Spatial regulation of homeobox gene fusions in the embryonic central nervous system of transgenic mice.

The spatially restricted expression of mammalian homeobox genes in teh embryonic central nervous system (CNS) provides an opportunity to study the basis of spatial gene regulation in mammalian development. Here, we define a regulatory region of the murine Hox 1.3 gene that mediates such a region-specific expression pattern. The Hox 1.3 gene contains two exons, encodes a putative protein of 270 amino acids, and is expressed preferentially in the spinal cord at midgestation. We have analyzed transgenic mice containing various Hox 1.3 DNA fragments fused to reporter sequences, such as a human growth hormone gene fragment or the E. coli lacZ structural gene. As shown by RNAase protection assays or by in situ analyses of beta-galactosidase activity, several hybrid genes are expressed in the embryonic central nervous system in a spatially restricted manner, along both the rostrocaudal and dorsoventral axes. A 912 nucleotide sequence located immediately upstream of the Hox 1.3 coding sequence is sufficient to direct expression to the dorsolateral cells of the brachial spinal cord.

Murine Hox-1.7 homeo-box gene: cloning, chromosomal location, and expression.

A new murine homeo-box, called Hox-1.7, has been identified in a rare cDNA from F9 teratocarcinoma stem cells. The Hox-1.7 homeo-box is 68 and 72% homologous to the Drosophila antennapedia (Antp) and iab-7 homeo-boxes, respectively. A major 2.5-kilobase transcript and several minor transcripts were detected by Northern blot (RNA blot) analysis in adult tissues as well as in midgestational embryos. The posterior spinal cord was found to be a major site of Hox-1.7 expression in 12.5-day-old embryos. Somatic cell hybrids were used to map the Hox-1.7 gene to mouse chromosome 6. Restriction fragment length polymorphisms associated with either the Hox-1.7 gene or the previously known Hox-1 complex were identified. Their distribution patterns in recombinant inbred mouse strains were used to determine the linkage between the two loci as well as to other loci on chromosome 6. This maps Hox-1 and Hox-1.7 close to two mouse loci that affect morphogenesis, postaxial hemimelia (px) and hypodactyly (Hd).

Genetic analysis and developmental regulation of testis-specific RNA expression of Mos, Abl, actin and Hox-1.4.

The pattern of Mos proto-oncogene RNA expression in the gonads of the sterile mouse mutants, dominant spotting (W), sex reversal (Sxr), testicular feminization (Tfm), hypogonadal (hpg), quaking (qk), two t-haplotypes, three X-autosomal translocations, and the YPOS strain, is consistent with its presence in haploid spermatids in the testes and in oocytes in the ovaries. In the male-sterile mouse mutants the pattern of expression of the testis-specific transcripts for Abl, actin, and the mouse homeobox Hox-1.4 genes is identical to that observed for Mos. However, during the postnatal onset of normal spermatogenesis we detected differences in the time of the appearance of the four transcripts. We detected Hox-1.4 transcripts at day 20, Mos at day 25, and Abl and actin at day 30, demonstrating a specific regulation of expression of each of these genes during haploid spermatid maturation in the mouse. Furthermore, comparison of Mos, Abl and actin RNA expression in mouse and rat testes revealed species-specific variations in the regulation of gene expression.

Cellular retinoic acid- and cellular retinol-binding proteins: complementary deoxyribonucleic acid cloning, chromosomal assignment, and tissue specific expression.

Both cellular retinoic acid-binding protein (CRABP) and cellular retinol-binding protein (CRBP) are widely distributed in mammalian tissues and display different patterns of tissue distribution. We have isolated cDNA clones for bovine CRABP and for human CRBP from a bovine adrenal gland cDNA library and from a human liver cDNA library, respectively, by probing with synthetic oligonucleotides. The primary structures of the two proteins inferred from the DNA sequences were identical to the previously reported amino acid sequences. The cDNA probes were used to obtain some information about the genes for CRABP and CRBP, including their chromosomal localization, and about the tissue specific expression of these two genes. Southern blot analyses under highly stringent conditions, of genomic DNA from both the bovine and the mouse, showed that for each retinoid-binding protein, specific DNA sequences are sufficiently conserved across species to allow cross-hybridization to occur. Under the same conditions, however, the DNA sequences for CRABP and CRBP within each species appear to be sufficiently different that cross-hybridization was not observed between the two cDNA probes. Using mouse-hamster somatic cell hybrids, it was demonstrated that these genes map to mouse chromosome 9 or 10. Northern blot analyses of RNA from six tissues from both the bovine and the mouse showed marked differences in the levels of the specific transcript for each binding protein among the different tissues, and in the tissue distribution of CRABP mRNA as compared to that of CRBP mRNA. In both species, CRBP mRNA was detected in more tissues than was CRABP mRNA. For both proteins, the relative tissue distribution of the mRNAs appeared to both resemble and to differ from the reported distribution of the binding proteins themselves. The factors that regulate the tissue specific expression of the genes for CRABP and CRBP remain to be determined.

Regulation of the cellular retinoid-binding proteins and their messenger ribonucleic acids during P19 embryonal carcinoma cell differentiation induced by retinoic acid.

P19 embryonal carcinoma (EC) cells can be induced to differentiate in vitro into a variety of cell types by treatment with different concentrations of retinoic acid (RA). A study was conducted to explore the regulation of expression of the genes for cellular retinoic acid-binding protein (CRABP) and cellular retinol-binding protein (CRBP) in P19 cells induced to differentiate by RA. For each retinoid-binding protein, both the level of specific mRNA and of immunoreactive protein were measured, respectively, by RNase protection assay and by a specific RIA. Dramatic increases in CRABP and CRBP were seen, at both the mRNA and protein levels, during the RA-induced differentiation. CRBP induction differed from that of CRABP in several major ways. 1) Induction of CRBP occurred at lower concentrations of RA (10(-9) M) than did that of CRABP (10(-8)-10(-7) M). 2) CRBP induction was an early response (within 3 h) to RA treatment, whereas CRABP induction occurred at a later time (12-24 h). 3) Induction of CRABP mRNA by RA was blocked by the protein synthesis inhibitor cycloheximide, whereas induction of CRBP mRNA was not. 4) Several differentiation inducers were tested for their effects on the expression of CRABP and CRBP in P19 cells. CRBP induction occurred with a wider spectrum of inducers than did that of CRABP. 5) In addition, the induction of CRABP and CRBP mRNAs by RA was examined in six different cell lines, including three EC lines. CRBP induction occurred in a wider spectrum of cell lines than did that of CRABP. The induction of CRABP in EC cells seems, in general, to correlate with their differentiation into neuron-like cells. Taken together, our results suggest that CRBP induction may be a direct response to RA and represent a general event in RA-induced cell differentiation, whereas CRABP induction may be an indirect response and represent a later event restricted to only certain differentiation pathways. CRBP may be an early response gene induced by RA.

Molecular cloning and transcriptional mapping of the mouse cellular retinoic acid-binding protein gene.

The gene encoding the mouse cellular retinoic acid-binding protein (CRABP) has been isolated from a mouse genomic library and its structure has been determined. This gene spans approximately 10.5 kb and consists of four exons encoding 24, 59, 38, and 16 amino acid residues, respectively. This gene structure is very similar to the structures of other related genes belonging to the same protein family such as the human cellular retinol-binding protein, the rat cellular retinol-binding protein II, the rat fatty acid-binding protein, and the mouse adipocyte P2 protein. The site for transcription initiation has been mapped to the 93rd nucleotide upstream from the translation initiation codon ATG using both primer extension and RNase protection assays. From the DNA sequence, the promoter of the CRABP gene resembles those found in the "housekeeping" genes in that it is very G/C rich, lacks a TATA box, and contains multiple copies of the sequence GGGCGG. The deduced amino acid sequence of the translated region is identical to the amino acid sequence of the known bovine CRABP, and the DNA sequence of the transcribed region from the mouse gene shows approximately 78% homology to that of the bovine cDNA.

Murine embryonic spinal cord and adult testis Hox-1.4 cDNAs are identical 3' to the homeo box.

The Hox-1.4 murine homeo box containing gene is expressed in both embryonic spinal cord and adult testis where different sized Hox-1.4 transcripts are detected. In order to compare the sequences of these transcripts, Hox-1.4 cDNA clones were isolated from cDNA libraries prepared from mouse embryonic spinal cord and mouse adult testis. Sequence analysis showed these clones to be identical and extend from an internal EcoRI site at position 55 in the homeo box to a poly (A) stretch 20 nucleotides 3' to the polyadenylation signal. These data indicate that murine Hox-1.4 mRNAs expressed in embryonic spinal cord and adult testis are identical 3' to the homeo box and utilize the same polyadenylation signal.

Alternatively spliced Hox-1.7 transcripts encode different protein products.

Two Hox-1.7 cDNAs, GPK5 and GPK6, were isolated from an adult guinea-pig kidney cDNA library by hybridization at low stringency using a Hox-1.7 cDNA probe, MH-1, cloned from mouse F9 teratocarcinoma cells. Sequence analysis of these two Hox-1.7 cDNAs showed that (a) GPK5 contains a putative initiation codon preceding an open reading frame which includes the homeo box, and may represent the complete protein coding region for the corresponding Hox-1.7 transcript; (b) the amino acids encoded by GPK6 and MH-1 are nearly identical (with two changes); (c) both adult guinea-pig kidney cDNA clones share identical homeo domains with the mouse Hox-1.7 cDNA; and (d) both adult guinea-pig kidney cDNA clones are identical in the homeo box region and in the 3' untranslated region but differ significantly starting from the 12th codon upstream from the homeo box. These data, supported by Southern blot analysis, indicate that a splice site is present 5' to the homeo box and that alternative splicing results in transcripts encoding different protein products.

Expression and regulation of a human metallothionein gene carried on an autonomously replicating shuttle vector.

A human metallothionein (MT) gene was inserted into a bovine papillomavirus (BPV) vector. The chimeric vector (pMTII-BPV) transforms rodent fibroblasts to a cadmium-resistant phenotype. The resistance is due to the high level of expression of human MT-II in those cells. The vector is maintained in the cells as a free replicating plasmid, present at about 10--15 copies per cell. Transcription of the episomal human MT-IIA gene is initiated from its authentic start sites and is regulated by the level of cadmium in the growth medium. The presence of the human MT-IIA gene allows the BPV replicon to function even though it is ligated to an intact copy of pBR322. Due to the presence of plasmid origins of replication and dominantly acting selective markers functional in both Escherichia coli and mammalian cells, pMTII-BPV can be used as a shuttle vector.

Defective lens fiber differentiation and pancreatic tumorigenesis caused by ectopic expression of the cellular retinoic acid-binding protein I.

All-trans retinoic acid, a metabolite of retinol, is a possible morphogen in vertebrate development. Two classes of cellular proteins, which specifically bind all-trans retinoic acid, are thought to mediate its action: the nuclear retinoic acid receptors (RAR alpha, beta, gamma), and the cytoplasmic binding proteins known as cellular retinoic acid-binding proteins I and II (CRABP I and II). The function of the retinoic acid receptors is to regulate gene transcription by binding to DNA in conjunction with the nuclear retinoid X receptors (RXR alpha, beta, gamma), which in turn have 9-cis retinoic acid as a ligand. Several lines of evidence suggest that the role of the cellular retinoic acid-binding proteins is to control the concentration of free retinoic acid reaching the nucleus in a given cell. Here, we have addressed the role of the cellular retinoic acid-binding protein I in development by ectopically expressing it in the mouse lens, under the control of the alpha A-crystallin promoter. We show that this ectopic expression interferes with the development of the lens and with the differentiation of the secondary lens fiber cells, causing cataract formation. These results suggest that correct regulation of intracellular retinoic acid concentration is required for normal eye development. In addition, the generated transgenic mice also present expression of the transgene in the pancreas and develop pancreatic carcinomas, suggesting that overexpression of the cellular retinoic acid-binding protein is the cause of the tumors. These results taken together provide evidence for a role of the cellular retinoic acid-binding protein in development and cell differentiation. The relevance of these findings to the possible role of the cellular retinoic acid-binding proteins in the transduction of the retinoic acid signal is discussed.

Region-specific expression of mouse homeobox genes in the embryonic mesoderm and central nervous system.

The homeobox is a 180-base-pair sequence characteristically found in homeotic and segmentation genes in Drosophila. Several copies of homeoboxes are also found in the mammalian genome, but it is not known whether these are components of morphogenetic loci in mammals as well. As a step toward understanding the function of mammalian homeoboxes, we have used in situ hybridization to define the spatial pattern of expression of two mouse homeobox genes in the midgestational mouse embryo. The two mouse homeoboxes studied here, Hox 1.2 and Hox 1.4, are located 20 kilobases apart on mouse chromosome 6. Our results demonstrate the following: (i) Hox 1.2 transcripts are localized mainly in the posterior myelencephalon, in the cervical central nervous system (CNS), and in several thoracic prevertebrae; (ii) Hox 1.4 transcripts are localized mainly in the posterior myelencephalon and in the cervical CNS; (iii) within the CNS region expressing Hox 1.4, the level of Hox 1.4 transcripts is higher in the mantle layer than in the ependymal layer and higher in the dorsal than in the ventral area. The specific localization of Hox 1.2 and Hox 1.4 transcripts in the embryonic CNS and the restricted pattern of expression along the rostrocaudal axis are strikingly reminiscent of the expression pattern of Drosophila homeoboxes in the fly embryo and larvae. Despite the different developmental strategies adopted by Drosophila and mammals, functional similarities may exist between Drosophila and mammalian homeobox genes.

Preferential transcription of the ovalbumin gene in isolated hen oviduct nuclei by RNA polymerase B.

The synthesis of ovalbumin mRNA sequences was studied in isolated nuclei from hen oviduct. Two different methods of analysis were used to distinguish in vitro synthesized from preexisting mRNA sequences: (i) Mercurated ribonucleotides were used for in vitro RNA synthesis, and the newly synthesized RNA was purified by chromatography on sulfhydryl-agarose and hybridized to radioactive ovalbumin cDNA. (ii) [3H]UTP was used to label the in vitro synthesized RNA. Hybridization to unlabeled mercurated cDNA, RNase A digestion, and subsequent purification of the hybrids on SH-agarose allowed the quantitation of newly synthesized ovalbumin mRNA sequences. Approximately 0.1% of the newly synthesized RNA was identified as ovalbumin RNA by both methods. The synthesis of ovalbumin RNA progressed during the incubation of nuclei and was sensitive to actinomycin D and low concentrations of alpha-amanitin. The preferential in vitro transcription of the ovalbumin gene (1000-fold over random transcription of the chicken genome) by RNA polymerase B (nucleosidetriphosphate:RNA nucleotidyltransferase, EC 2.7.7.6) suggests that the specificity of in vitro RNA synthesis is retained in isolated nuclei.

Region-specific enhancers near two mammalian homeo box genes define adjacent rostrocaudal domains in the central nervous system.

To gain insight to the mechanisms underlying region-specific gene expression in mammalian development, we investigated the regulatory DNA associated with the proximal promoter of two homeo box genes, murine Hox-1.3 and human Hox-5.1. Using lacZ gene fusions in transgenic mice, we identified regulatory elements in the 5'-flanking sequences of the Hox-1.3 and the Hox-5.1 genes that specifically direct beta-galactosidase expression to the brachial and the upper cervical regions (respectively) of the central nervous system (CNS). These two elements act at the transcriptional level, are active in either orientation, and confer region-specific expression to unrelated promoters, satisfying the criteria for enhancer elements. The two spatial domains defined by these enhancers are directly adjoining, extend along the rostrocaudal axis for the same span of 6-7 metameres, and represent specific subsets of the overall CNS regions expressing all endogenous Hox-1.3 or Hox-5.1 transcripts. The adjacent domains in the developing murine CNS that express Hox-1.3 and Hox-5.1 gene fusions are strikingly reminiscent of the adjacent stripes of expression in Drosophila embryos seen with Sex combs reduced and Deformed, the two Drosophila homeotic genes most homologous to Hox-1.3 and Hox-5.1, respectively. These findings represent the first demonstration of region-specific mammalian enhancers and raise the possibility that the mammalian CNS may be subdivided into a series of rostrocaudal domains on the basis of the activity of enhancers near homeo box genes.

Spatial and temporal pattern of expression of the cellular retinoic acid-binding protein and the cellular retinol-binding protein during mouse embryogenesis.

Retinol (vitamin A) and retinoic acid are potent teratogens and also represent good candidates for normal morphogens during development. Their actions may be mediated by the cellular retinoic acid-binding protein (CRABP) and the cellular retinol-binding protein (CRBP). As a step towards understanding the possible function for CRABP and CRBP in morphogenesis, we have used in situ hybridization to analyze their expression during mouse development. Both CRABP and CRBP transcripts were detected at embryonic days 9.5-14.5. (i) In the nervous system, CRABP transcripts were found in the mantle layer of the dorsal spinal cord and hindbrain and in the marginal layer of the midbrain, whereas CRBP transcripts were found in the ependymal and mantle layer of the ventral spinal cord and of the forebrain as well as in the spinal nerves and the roof plate of the spinal cord. (ii) In the eye, CRABP is expressed in the retinal layer, and CRBP is expressed in both retinal and pigmented layers. (iii) In the craniofacial region, CRABP transcripts were found in the mesenchyme of the frontonasal mass and mandible, while CRBP transcripts were found in the mesenchyme of the nasolachrymal duct and surrounding the auditory vesicle. Two general conclusions can be made. First, all of the tissues that are known to be teratogenic targets of retinoic acid and retinol also express CRABP and CRBP transcripts. Second, the specific expression of CRABP and CRBP in numerous developing tissues indicates that these proteins may perform specific functions during morphogenesis of a broad variety of embryonic structures.

Transcription of the chicken ovalbumin and conalbumin gene during early secondary induction with estrogens.

Estrogens stimulate the rate of transcription of the ovalbumin and conalbumin gene in the chicken oviduct. The synthesis of ovalbumin and conalbumin mRNA was studied in isolated nuclei. RNA synthesized in vitro was distinguished from preexisting nuclear RNA by affinity labeling the in vitro products with a mercurated nucleotide and subsequent purification of the Hg-RNA on SH-agarose. The content of ovalbumin and conalbumin mRNA sequences in the in vitro transcripts was determined by hybridization to cDNA. After the withdrawal of implanted hormones from chickens, the synthesis of conalbumin and ovalbumin RNA increased 2.5- and at least 20-fold, respectively, by treatment with estrogens. The maximal rate of transcription of the conalbumin gene is achieved within 2 h after estrogen induction, whereas the rate of transcription of the ovalbumin gene becomes maximal after a lag of several hours. These results demonstrate that estrogens affect two genes in the same target cell differently.

Hoxa5 gene regulation: A gradient of binding activity to a brachial spinal cord element.

The Hox genes cooperate in providing positional information needed for spatial and temporal patterning of the vertebrate body axis. However, the biological mechanisms behind spatial Hox expression are largely unknown. In transgenic mice, gene fusions between Hoxa5 (previously called Hox-1.3) 5' flanking regions and the lacZ reporter gene show tissue- and time-specific expression in the brachial spinal cord in day 11-13 embryos. A 604-bp regulatory region with enhancer properties directs this spatially specific expression. Fine-detail mapping of the enhancer has identified several elements involved in region-specific expression, including an element required for expression in the brachial spinal cord. Factors in embryonic day 12.5 nuclear extracts bind this element in electrophoretic mobility shift assays (EMSA) and protect three regions from DNase digestion. All three sites contain an AAATAA sequence and mutations at these sites reduce or abolish binding. Furthermore, this element binds specific individual embryonic proteins on a protein blot. The binding activity appears as a gradient along the anterior-posterior axis with two- to threefold higher levels observed in extracts from anterior regions than from posterior regions. In parallel with the EMSA, the proteins on the protein blot also show reduced binding to probes with mutations at the AAATAA sites. Most importantly, transgenic mice carrying Hoxa5/lacZ fusions with the three AAATAA sites mutated either do not express the transgene or have altered transgene expression. The brachial spinal cord element and its binding proteins are likely to be involved in spatial expression of Hoxa5 during development.