Cell transformation by avian defective leukaemia viruses

A comparative study of seven independently isolated defective leukaemia viruses has been carried out. Phenotypic analysis of the chicken bone marrow cells transformed in vitro allowed the separation of these seven viruses into three groups based on the differentiation phenotype of the transformed cell. Nucleic acid hybridization studies revealed that these seven viruses had acquired cellular sequences. Interestingly, these studies also showed that the viruses within the same biological grouping had acquired related sequences. This indicates that viruses that have acquired the same or similar cellular sequences have very similar oncogenic capabilities. Analysis of proteins expressed in cells transformed by these viruses demonstrated that the cellular sequences were usually inserted within the gene for the viral core proteins, gag. Therefore the cellular sequences are expressed as a gag-related fusion protein which has an amino-terminal region derived from the gag gene and a carboxy-terminal half derived from the cellular sequences. Two exceptions to this are discussed. The general conclusion from these studies is that defective leukaemia viruses transform cells by virtue of acquired host cellular sequences. The ability of these viruses to transform cells and the target cell specificity of the transformation depends on these cellular sequences.

The structure and expression of the integrated viral DNA in mouse cells transformed by simian virus 40

Mouse cells stably transformed by simian virus 40 (SV40) contain viral DNA covalently integrated into their chromosomal DNA. We have used molecular cloning techniques to isolate and characterize the integrated viral DNA, together with the flanking cellular sequences, from several lines of transformed cells. We have identified in SV40-infected mouse cells a novel form of viral DNA, which we believe to be the precursor to the integrated DNA. The mRNAs transcribed from the integrated viral DNA templates have been characterized. Integrating the viral genome into the cellular chromosome can significantly alter its transcription pattern and in several cases the altered mRNAs encode novel forms of viral tumour antigen.

Differential expression of H-2 gene products in tumour cells in associated with their metastatogenic properties

Many neoplasms seem to be heterogeneous in nature, producing metastases by pre-existing variant cells with inherent biochemical and biological properties. The survival and proliferation of metastatic cells depend on various biological properties, such as enzymes which degrade basement membranes, resistance to various host defence systems, association with host cellular components, adhesiveness and expression of certain membrane glycoproteins. Recent studies have indicated that metastatic cells may differ from the local tumour cells in the expression of immune recognizable membrane-associated antigens. Such antigenic differences may result from an immunoselection of cells with distinct antigenic properties due to a specific immune response evoked against the local tumour. In view of the role of the major histocompatibility complex (MHC) system in controlling and restricting the function of immune effector cells against modified self-components, one could assume that the modulation of the expression of MHC-encoded antigens on the membrane of tumour cells influenced the interclonal relationship within a local heterogeneous tumour cell population and the subsequent generation of metastasis. The modulation of the expression of H-2 antigens on several murine tumours is well documented; however, practically no attempts were made to relate H-2 modulation with invasiveness. We now describe principal differences in the expression of H-2 parental haplotypes between a local F1 methylcholanthrene-induced tumour and its descendant pulmonary metastases. These results suggest that both the expression and the immunogenicity of MHC products strongly influence the immune relationship between the tumour and the host's immune system, thus determining the generation and dissemination of metastases.

X-chromosome inactivation in extra-embryonic membranes of diploid parthenogenetic mouse embryos demonstrated by differential staining

In somatic cells of female mammals one of the two X chromosomes is genetically inactive and heterochromatic, resulting in dosage compensation for X-linked genes. In marsupials the paternally derived X chromosome is preferentially inactivated. In eutherian mammals, although either X chromosome can be inactivated at random in somatic cells, preferential inactivation of the paternally derived X chromosome has been demonstrated cytologically in mouse and rat yolk sac and mouse chorion and biochemically in mouse yolk sac, chorionic ectoderm and trophoblast. In mouse yolk sac the non-random element has been shown both biochemically and cytologically to be confined to the endoderm layer in which there is almost total paternal X-chromosome activity in the separated yolk sac layers of diploid parthenogenetic mouse embryos in which both X chromosomes are maternally derived. Kaufman et al. have demonstrated X inactivation in somatic cells of diploid parthenogenetic embryos, and we have used a modification of Kanda's method, which renders the presumptive inactive X dark staining, to reveal an inactive X chromosome in both endoderm and mesoderm layers of separated yolk sacs from parthenogenones. Thus even in tissues in which there is normally total non-random paternal X inactivation, in the absence of a paternally derived X chromosome a maternally derived X can be inactivated.

Structure and developmental expression of the chick alpha-actin gene

Recombinant DNA clones containing chick alpha-actin mRNA sequence have been isolated and used as probes to analyze the structure and developmental expression of the chick alpha-actin gene. The full length, 2000 nucleotide alpha-actin mRNA is detected in poly(A) RNA at early and late stages of in vivo leg muscle development. As expected, the alpha-actin mRNA is present at very low levels at early myogenic stages but is a high abundance species in terminally differentiated muscle. However, most of the alpha-actin mRNA from fused leg muscle is shorter than 2000 nucleotides, and occurs in relatively discrete size classes. An alpha-actin-like mRNA can be detected in poly(A) RNA from early embryonic brain, indicating that transcription of the alpha-actin gene may not be strictly muscle-specific at all stages of development. We have identified at least 3, very short (< 100 base pairs) intervening sequences in the alpha-actin gene which was isolated from a chick genomic library. The structure of the chick alpha-actin gene differs, therefore, from the structures of actin genes from yeast and Drosophila, both of which contain a single, relatively long, intervening sequence.

Sequence organization of the repeating units in the nucleus of wheat which contain 5S rRNA genes

Examples of both the 410 and 500 bp size classes of repeating units containing wheat 5S rRNA genes have been cloned in plasmid pBR322 and sequenced. The structural genes showed sequence microheterogeneity. Also the gene in the 500 bp repeat which was sequenced had a 15 bp tandem duplication within it and appears to be representative of non-transcribed subfamily of repeating units. The transcription terminators comprise 14-17 A.T bp immediately preceded by a region of weak dyad symmetry. The spacer regions adjacent to the transcription terminators in the two different size repeat units have interspersed oligonucleotides of high and low homology. The central spacer regions of the two size classes have very different sequences. The only repeated sequence in the spacers has undergone extensive divergence. In contrast to most of the spacer, the 70 bp region preceding the genes of each type of repeat show high homology, suggesting that it has functional importance. The transcription start point obeys the pyrimidine-1 purine+1 rule.

DNA methylation and gene function

In most higher organisms, DNA is modified after synthesis by the enzymatic conversion of many cytosine residues to 5-methylcytosine. For several years, control of gene activity by DNA methylation has been recognized as a logically attractive possibility, but experimental support has proved elusive. However, there is now reason to believe, from recent studies, that DNA methylation is a key element in the hierarchy of control mechanisms that govern vertebrate gene function and differentiation.

Transcriptional fidelity of histone genes injected into Xenopus oocyte nuclei

Previous work has indicated that at least some of the genetic information required for the expression of sea urchin histone genes is recognized following injection of the gene repeat (h22) into Xenopus oocyte nuclei. The ability to elicit the expression of cloned genes and their sequence-manipulated counterparts is proving invaluable in analysing the molecular details of gene expression. Direct injection of such genes into Xenopus oocyte nuclei remains one of the simplest methods for obtaining such expression and a remarkable degree of transcriptional fidelity has been demonstrated using this system with RNA polymerase III genes, and to a lesser extent with rDNAs transcribed by RNA polymerase I. In the case of polymerase II genes there is ample evidence for coupled transcription-translation, but the degree of transcriptional fidelity involved may, as has recently been shown for the ovalbumin gene, be minimal. However, clearly if the oocyte is to be used to investigate transcriptional regulation of such genes, transcriptional fidelity defined as the production of correct RNA termii, rather than the production of 'functional mRNAs' (ref. 15), must pertain. Here we demonstrate such fidelity in the expression of all five Psammechinus miliaris histone genes comprising a repeat unit. However, we find large quantitative variations in the levels of synthesis of the individual correct termini and hence of the mRNAs. In addition to the mRNAs, species with no detectable counterparts in the sea urchin are generated off the coding strand, as are heterogeneous noncoding species.

Mutations in nif genes that cause Klebsiella pneumoniae to be derepressed for nitrogenase synthesis in the presence of ammonium

Four Nif+ revertants from strains with polar insertions in nifL, were insensitive to ammonium and amino acid repression of nitrogenase synthesis. These strains have mutations located in or near the nifL region. The derepressed phenotype was dominant in a merodiploid containing a nif+ plasmid. These nif regulatory mutations also suppressed the Nif- phenotype of Gln- strains. Thus, regulation by fixed nitrogen (possible via glutamine synthetase) occurs on the nifLA operon but not on the other six nif operons.

Tissue-specific DNA methylation in a cluster of rabbit beta-like globin genes

The relationship between DNA methylation and differential expression of rabbit beta-like globin genes was studied by using restriction enzymes that cleave the sequence C-C-G-G but are differentially inhibited by the presence of 5-methylcytosine. The methylation frequency of 13 C-C-G-G sites that flank a set of four closely linked rabbit beta-like globin genes was determined. This analysis revealed that certain sites surrounding embryonic and adult globin genes are relatively undermethylated in DNA from embryonic and adult erythroid tissues, respectively. This pattern is most pronounced for three sites that are undermethylated in erythroid cells but are totally methylated in nonerythroid cells. We conclude that the degree of CpG methylation in the rabbit beta-like globin gene cluster is correlated with gene activity, but the effect is confined to relatively small regions of DNA.

Cell commitment and gene expression in the axolotl embryo

In the axolotl embryo the somitic mesoderm passes through a reversible and then an irreversible phase of commitment with respect to its later differentiation into muscle. We show that the commencement of alpha-actin synthesis and the first appearance of thin myofilaments occur at the the same developmental stage as the transition between these phases. In intact embryos beta-and gamma-actin are made at all stages of development and in all tissues. alpha-actin, however, first appears at the late head process stage and is confined to the somites and tailbud, the regions of the embryo which later form the myotomal muscles. The thin filaments first appear near the myocoel of the somites in association with a novel organelle of mottled appearance and are quite distinct in structure from the epidermal microfilament networks found in embryos of the same stage. Cell fusion occurs somewhat later, although before the appearance of muscle striations in the light microscope. Well formed sarcomeres are not found until some time after the onset of motility. Presumptive somitic mesoderm isolated at a stage before the transition and cultured in a buffered salt solution for six days autonomously begins to synthesize alpha-actin and develop sarcomeres with the normal structure.

Physiological characterization of a Neurospora crassa mutant with impaired regulation of nitrate reductase

This report describes the isolation and characterization of a Neurospora crassa mutant with an impaired regulation of nitrate reductase. Glutamine, which prevents the induction of nitrate reductase in N. crassa, did so relatively ineffectively in this mutant. The mutation did not affect the regulation of all enzymes regulated by "nitrogen metabolite regulation"; it did affect the regulation of nitrate reductase, nitrite reductase, histidase, and acetamidase, as well as that of thiourea sensitivity. The mutation was not allelic with nit-2, the gene controlling a general positive effector of nitrogen metabolite-regulated enzyme formation.

Novel surface antigen expressed on dividing cells but absent from nondividing cells

Radioimmunoprecipitation followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis was used to study the distribution on human lymphoid cells of a previously undescribed surface antigen recognized by several heteroantisera. A glycoprotein with a 90,000 mol wt (under reducing conditions) was detected on all cell lines tested including T, B, null, and myeloid cell lines, although the amount of antigen present varied considerably. The antigen was absent from normal peripheral blood lymphocytes (PBL), B and T cells, monocytes, granulocytes, thymocytes, and erythrocytes. After stimulation with lectins or allogeneic B cells, the antigen was induced on PBL T cells. A limited number of leukemic T cells tested all expressed the antigen, as did melanoma cell line and human embryonic lung fibroblasts. Hence, the antigen was present only on dividing lymphoid cells and absent from nondividing cells, but was also present on the two examples of dividing non-lymphoid cells tested. Under nonreducing conditions, the 90,000-mol wt band normally present disappeared to be replaced by another at approximately 200,000 mol wt. The glycoprotein bound to lectins from lentil and ricin, but not to wheat germ agglutinin. It could be readily labeled metabolically by [35S]methionine or by surface iodination, and appeared to be a major membrane protein on some cell lines.

A theory of modular evolution for bacteriophages

The modular theory of virus evolution has clear experimental support among the temperate bacteriophages of the enteric bacteria. However, there is also similar genetic and DNA heteroduplex evidence for such evolution among other families of bacteriophages: the virulent bacteriophages of the enterics comprise several families: the T-even group, the T3-T7 group (which has many members among different species of bacteria, including bacteria as widely divergent as E. coli and Caulobacter crescentus. It nicely explains the diffusion of very similar homologous bacteriophages into hosts whose own DNAs have diverged very greatly from each other in nucleotide sequence. It also accounts for the rigorous maintenance of regulatory schemes while units of function (including regions coding for proteins) diverge more rapidly. It should also be noted that the considerations that make modular evolution seem advantageous for bacteriophages apply equally well to viruses of higher organisms. Furthermore, the kinds of heteroduplex similarity observed among animal viruses are reminiscent of what is found for bacteriophages. Viruses found in widely divergent hosts show much greater similarity than would be expected; quite possibly animal viruses also evolve as a population of interchangeable modules.

Developmental regulation of laccase levels in Aspergillus nidulans

Asexual spores (conidia) of Aspergillus nidulans contain a dark green pigment which is not present in other cell types. Synthesis of this pigment is catalyzed, in part, by a developmentally controlled p-diphenol oxidase, or laccase, encoded at the gamma A genetic locus (A. J. Clutterbuck, J. Gen. Microbiol. 70:423-435, 1972). We have investigated the mechanisms regulating expression of the gamma A gene of A. nidulans. Vegetative hyphae grown in submerged culture lacked detectable laccase enzyme activity and neither contained nor synthesized immunoprecipitable laccase protein. When such cultures were induced to conidiate by harvesting the cells onto filter papers and aerating them, laccase levels began to increase after 10 to 16 h, reached a peak at 20 to 36 h, and then declined slowly. Immunological assays showed that increases in laccase enzyme activity were (i) proceded by a transient rise in the relative rate of laccase protein synthesis and (ii) closely paralleled by increases in the amount of laccase protein. Addition of cycloheximide to cultures at any time after inducing conidiation inhibited further accumulation of laccase enzyme activity. These data are most consistent with increases in laccase levels being due to regulated, de novo synthesis of laccase protein. Addition of inhibitors of ribonucleic acid synthesis to conidiating cultures also inhibited further accumulation of laccase, suggesting that laccase expression is regulated by alterations in the transcriptional activity of the gamma A locus.

Synthesis of ribosomal proteins in merodiploid strains of Escherichia coli

The regulation of the synthesis of r-proteins in Escherichia coli was investigated by increasing the dosage of the genes for a limited number of ribosomal proteins (r-proteins) using either transducing phage lambda fus 3 (Lindahl et al. 1977) or lambda rifd 18 (Kirschbaum and Konrad 1973). During exponential growth the presence in the cell of either lysogenised transducing phage did not increase the rate of synthesis or degradation of any of the 31 r-proteins whose genes are duplicated. Experiments were also performed to determine whether r-protein synthesis during the period of unbalanced r-protein synthesis that follows nutritional enrichment was sensitive to an increase in gene dosage. Duplication of the 27 r-protein genes on lambda fus 3 did not alter the rate of synthesis of any of the r-proteins after enrichment. However, gene dosage effects were detected for at least 3 of the r-proteins whose genes were duplicated on lambda rifd 18.

Simultaneous expression of mouse immunoglobulins M and D is determined by the same homolog of chromosome 12

A hamster-mouse hybrid cell line expressiong both murine IgM and murine IgD on the membrane was shown to have only one copy of mouse chromosome 12. This chromosome is known to carry the structural genes for the immunoglobulin heavy chains. Cloning of populations selected for loss of mouse membrane IgM yielded cells that had also lost expression of membrane IgD, but not the expression of hamster immunoglobulin heavy chain. Karyotype analysis of these subclones demonstrated the concurrent loss of the chromosome 12 present in the parental hybrid. Absence of this chromosome was confirmed by use of the isozyme acid phosphatase 1. The results of the genetic analysis prove that the coexpression of mu and delta immunoglobulin heavy chains is not due to long-lived immunoglobulin mRNA nor to the transcription of genes on homologous chromosomes. We conclude that the genetic information for IgM and IgD expressed by a single cell lies on the same chromosome.

Instability in tyrR strains of plasmids carrying the tyrosine operon: isolation and characterization of plasmid derivatives with insertions or deletions

The transformation of tyrR strains of Escherichia coli with multicopy plasmids which carry the tyrosine operon gave rise to modified plasmids with either insertions or deletions. The effect of each of these insertions or deletions was to decrease the level of expression of this operon. It is proposed that plasmid instability arose as a direct consequence of the metabolic effects of an overproduction of the enzymes coded for by the tyrosine operon. The results have significant implications for the cloning of genes that are repressed by the product of a regulatory gene. Since the predominant plasmid modification observed was the insertion of an IS1 element near the regulatory region of the tyrosine operon, the results also suggest a role for IS1 elements in the regulation of gene expression.

Identification of the dnaA and dnaN gene products of Escherichia coli

A specialized transducing lambda phage carrying the dnaN genes of Escherichia coli specifies two proteins of about 41 and 48 kilodaltons (kd). The temperature-sensitive mutations, dnaN59 and dnaA167, were found to result in altered isoelectric points of the 41 and 48 kd proteins, respectively. Thus the dnaN gene product was identified as a weakly acidic 41 and 48 kd protein. The synthesis of the dnaN gene product is greatly reduced by insertion of a transposon Tn3 in the dnaA gene and by deletion in the gene at the distal end to the dnaN gene. Temperature-sensitive dnaA mutations, on the dnaN gene product. These results indicate that the synthesis of the dnaN gene product is dependent on the structural integrity of the dnaA gene.

Mating signals control expression of mutations resulting from insertion of a transposable repetitive element adjacent to diverse yeast genes

ROAM mutations cause overproduction in S. cerevisiae. Overproduction of ROAM mutant gene products is less in MATa/MAT alpha diploid strains which cannot conjugate than in haplolid strains which can. Overproduction occurs in diploid strains capable of mating whether or not they are capable of sporulating. Overproduction decreases when haploid ROAM mutants also contain the ste7 mutation which prevents conjugation; other ste mutations do not affect the expression of ROAM mutations. Cloning of the ROAM mutant gene CYC7-H2 shows that a 5.5 kb sequence homologous to a transposable and reiterated Ty1 element is inserted in the 5' noncoding region of the CYC7 structural locus. The similar genetic properties of other ROAM mutations suggest that they each contain an inserted Ty element. These results also suggest that ROAM mutations respond to signals normally directed toward genes controlling conjugation functions, and that sequences present in Ty elements may be adjacent to structural loci and are the normal receptors for these signals.

Isolation of yeast genes with mRNA levels controlled by phosphate concentration

A library of DNA from the yeast, Saccharomyces cerevisiae, was constructed in phage lambda Charon 4 vector and then screened by differential plaque filter hybridization for genes induced by phosphate starvation. Two EcoRI fragments of 7.9 and 5.0 kilobase pairs that contained such genes were isolated. These cloned fragments may each carry one of the several copies of the genes for the repressible acid phosphatase of yeast. The fragments were use to examine mRNA levels of these genes in regulatory mutants of acid phosphatase.

Expression of an X-linked gene from an inactive human X chromosome in mouse-human hybrid cells: further evidence for the noninactivation of the steroid sulfatase locus in man

Somatic cell hybrid clones were derived from the fusion of hypoxanthine phosphoribosyltransferase (HPRT; EC 2.4.2.8)-deficient mouse cells and two different human fibroblast strains, each carrying an X chromosome-autosome translocation. One of these had an X/11 translocation [46,X,t(X;11)(p21;q13)] and the other had an X/19 translocation [46,X,t(X;19)(q22;q13)]. The structurally normal human X chromosome is the late-replicating (genetically inactive) chromosome in these two cell strains; the rearranged X chromosome is early replicating (genetically active). One primary hybrid clone carrying both the translocated X chromosome and the structurally normal X chromosome was isolated in hypoxanthine/aminopterin/thymidine medium from each of these two cell fusion experiments. These clones were then selected in medium containing 8-azaguanine to achieve the loss of the active human HPRT locus. Five subclones from the cell hybrid with the X/11 translocation failed to express two known human X-chromosome markers [glucose-6-phosphate dehydrogenase (G6PD; EC 1.1.1.49) and phosphoglycerate kinase (PGK; EC 2.7.2.3)] but did express human microsomal steroid sulfatase (STS; sterol-sulfate sulfohydrolase, EC 3.1.6.2). Three of these were cytogenetically analyzed and found to contain a structurally normal human X chromosome but not the X/11 translocation. Two subclones were isolated in 8-azaguanine from the hybrid with the X/19 translocation. Cytogenetic analysis of these two clones showed the presence of a structurally normal human X chromosome; the X/19 translocation was not present. They did not express human G6PD, PGK, or HPRT but did express human STS. These results indicate that human STS is expressed from a locus on the inactive human X chromosome and support our earlier finding that the STS locus escapes X-inactivation in man.

Large sized nascent protein as dominating component during protein synthesis in Chironomus salivary glands

The main secretory protein fractions from Chironomus tentans have been investigated with particular emphasis on the dominant fraction, component 1, here designated I (Grossbach, 1969). This polypeptide was suggested to be the translatory product of 75SS RNA from Balbiani ring 2 (BR2) because of its size and quantitative prominence. Its molecular weight was estimated by gel filtration in 8 M urea at 850,000 + 101,000 D. During short pulses with radioactive amino acids a large fraction of the label was found in a population of polypeptide chains suggestive of molecules continuously growing to the size of compoenet I. Populations of nascent large protein chains of similar size distribution were dominant in the polysomes and constituted the only population present in the largest polysomes, known to contain 75S RNA from BR2 (and BR1) as predominant or only component (Daneholt et al., 1977; Wieslander and Daneholt, 1977). These data indicate strongly that the large size of component I is not a result of posttranslational modifications. No sequence similarities, using limited proteolysis, were found between component I and component II, both of which have been considered to the BR2 products. There was, furthermore, no detectable immunological identity between component I and smaller secretory protein fractions. The data support Grossbach's and Daneholt's suggestion that component I is closely related to the primary translation product of 75S RNA from the large Balbiani rings.

[Genome instability]

The paper deals with the various manifestations and the role of genome changes, mainly in eukaryotes. I. Hereditary changes: 1) multiplication of genes in animals; 2) multiplication of genes in prokaryotes; 3) microsymbionts, mobile elements and supermutability; 4) extrachromosomal element delta in Drosophila; 5) hybrid dysgenesis in Drosophila; 6) mobile and unstable genes in drosophila; 7) controlling elements in planrs. II. Genome changes as a regulatory factor of genetic activity: 1) immunoglobulin genes; 2) mating types in yeast; 3) the inversion mechanism of switching gene activity in prokaryotes; 4) ribosomal and histone genes; 5) changes of DNA fractions in ontogenesis; 6) regulation of genome changes.

Association of a 5S gene transcription factor with 5S RNA and altered levels of the factor during cell differentiation

We have previously purified from Xenopus ovaries a protein factor (TF IIIA) which is necessary for the accurate in vitro transcription of 5S RNA genes. We now report that this factor (a 5S gene transcription effector) is identical by immunological, chemical and functional criteria to the protein associated with 5S RNA (the gene product) as a 7S ribonucleoprotein (RNP) complex in immature oocytes. After oocyte maturation, TF IIIA is no longer detectable functionally or immunologically in unfertilized eggs, which do not synthesize 5S RNA in vitro. Moreover, we cannot detect TF IIIA immunologically in extracts fron Xenopus somatic cells which, nevertheless, efficiently transcribe 5S genes.

Control of serum protein production in hepatocyte hybridomas: immortalization and expression of normal hepatocyte genes

"Hepatocyte hybridomas" have been isolated after fusion of adult hepatocytes and alpha-fetoprotein (AFP)-producing mouse hepatoma cells. The yield of viable hybrid clones was low but could be increased by culturing the cells in the presence of insulin. On the basis of their chromosome constitution, the hybrids were classified into two groups characterized by either a single or a double set of mouse (hepatoma) chromosomes. The hybrids segregated rat chromosomes and thus constitute an excellent material for gene mapping studies in the rat. Most of the hepatocyte hybridomas retained the production of one or more rat serum proteins, indicating that the corresponding structural genes, contributed by the normal hepatocyte parent, have been immortalized and maintained in the active state after fusion. However, these hybrids do not produce rat AFP, although mouse AFP synthesis is maintained. This result strongly suggests that silent rat (hepatocyte) AFP genes coexist in hepatocyte hybridoma nuclei with active mouse (hepatoma) AFP genes. Finally, on the basis of certain properties of these hepatocyte-hepatoma hybrids, we suggest that the nondividing state of the hepatocytes is actively controlled by a regulatory mechanism which prevents DNA synthesis or entry into mitosis or both.

Expression of alcohol dehydrogenase during pearl danio development

The ontogeny of alcohol dehydrogenase (ADH) was examined in the pearl danio (Brachydanio albolineatus) by the utilization of vertical slab acrylamide gel electrophoresis. The restriction of ADH activity to liver tissue in B. albolineatus is consistent with observations reported for other species of brachydanio. At 118 hours posthatching, a time when larval fishes were found to undergo functional maturation of the liver, the ADH homopolymer appeared. The enzyme was observed to increase in activity concurrently with decreasing yolk reserves. It is suggested that the state of liver cell differentiation and/or the physiological function of liver tissue provides the stimulation required to activate the Adh locus in Brachydanio.

Transposed LEU2 gene of Saccharomyces cerevisiae is regulated normally

The repression of beta-isopropylmalate dehydrogenase, the LEU2 gene product, by leucine and leucine plus threonine was unaffected by the transposition of LEU2 from its original locus on chromosome III to a new locus within the ribosomal deoxyribonucleic acid gene cluster on chromosome XII. Since the expression of the LEU2 gene is probably controlled at a pretranslational level, we conclude that the recombinant plasmid used for transformation carries regulatory information in addition to LEU2 structural information.

Characterization of a HindIII-generated DNA fragment carrying the glutamine synthetase gene of Salmonella typhimurium

The glnA gene, encoding glutamine synthetase in Salmonella typhimurium, has been cloned into the plasmid pBR322. One hybrid plasmid, pJB1, containing an 8.5 kb insert generated by a HindIII digest, was analyzed using eleven different restriction enzymes. Evidence that the region controlling glutamine synthetase expression remained on the insert was obtained by showing that the regulation is normal in cells carrying plasmids with the insert in the original and reversed orientation. Several new plasmids derived from pJB1 following SalI and EcoRI digestions were examined for their ability to complement a glnA202 mutation in order to locate the DNA segment needed for glutamine synthetase expression. The results show that cells containing plasmid pJB8, which has a 21 kb deletion, produce and regulate glutamine synthetase normally, whereas cells with a plasmid (pJB11) similar to pJB8, but lacking a 0.25 kb EcoRI fragment, do not exhibit glutamine synthetase activity. The analysis of proteins produced in minicells containing pJB8 and pJB11 show that they both produce a protein that migrates with the glutamine synthetase subunit. Because pJB11 makes an inactive protein of similar size to the glutamine synthetase subunit, the 0.25 kb deletion may encode only the C-terminus of this protein. Consistent with this finding is the presence of a strong RNA polymerase-binding site on pJB8 to the right of the 0.25 kb EcoRI that could correspond to a promoter near the N-terminus of the glnA gene.

Isolation of E.coli promoters from the late region of bacteriophage T7 DNA

Promotor sequences recognized by Escherichia coli RNA polymerase have been isolated from bacteriophage T7 DNA using the plasmid pBRH4. T7 DNA was digested with the restriction endonuclease Hae III, Alu I, and Eco RI* and the products of these digestions were ligated into the EcoRI site of pBRH4. Cloning of Hae III and Alu I-digested T7 DNA was achieved by blunt-end ligation of these fragments to the polymerized ends of Eco-RI-cleaved pBRH4. This converts blunt-end Eco RI fragments of T7 DNA into cohesive-end EcoRI fragments. Promoter-containing T7 restriction fragments were selected by activation of the tetracycline-resistance gene located on the plasmid vector. The genomic location of each T7 insert was determined and Hpa I-cleaved T7 DNA. Two promoter-active restriction fragments are thought to contain the C and E promoters of T7. However, the majority, of the promoter-active fragments cloned map within the late gene region of T7. In vitro binding studies indicate that E. coli RNA polymerase can form heparin resistant complexes with the cloned T7 DNA promoter fragments. These results suggest that while E. coli RNA polymerase may not participate directly in the transcription of late T7 genes, promoters for this enzyme are present in this region of the DNA.

Transcriptional control regions of the adenovirus VAI RNA gene

By constructing deletion mutations in cloned adenovirus types 2 and 5 VAI genes and measuring the ability of altered templates to direct transcription of VAI RNA in HeLa cell extracts, we have located two transcriptional control regions. The first is an intragenic region located between positions +9 and +72 relative to the 5' end of the VAI(A) RNA. Those deletions examined within these sequences abolished the transcription of mutant templates in HeLa cell extracts. The second control region includes 5' flanking sequences which abut the VAI coding region. Mutations here can reduce the efficiency with which the VAI gene is transcribed. Nucleotide sequence similarities were noted on comparison of the VAI intragenic control region to tRNA sequences, which lead us to speculate that the transcriptional regulation of these two types of genes may be quite similar; the adenovirus VA genes may even have evolved from a tRNA gene(s).

DNA replication and the early to late transition in adenovirus infection

Expression of the late genes of adenovirus is only detectable after virus DNA synthesis has occurred. Using a superinfection protocol, we show that replication of the template per se is required for expression of late regions L2--L5 (mapping to the right of position 39) and that the accumulation of early gene products does not suffice. This regulation is probably exerted at the level of transcription rather than by control of processing or selective stabilization of late mRNA or its precursors. The promoter-proximal late gene block L1, however, appears to be subject to processing control. At early times a single member of this gene family (tripartite leader plus coordinates 29--39, encoding the 52,55K polypeptide pair) is expressed, whereas at late times an additional, differently spliced mRNA species is generated from this region (tripartite leader plus coordinates 34--39, encoding polypeptide IIIa).

Regulation of adenovirus mRNA synthesis

The lytic cycle of adenovirus is a tightly regulated sequence of stages. When this regulation is studied at the level of mRNA production, the most significant step in controlling gene expression is initiation of transcription. Thus in preceding from one stage of expression to another, viral factors seem to turn on transcription of new sets of genes. At the moment, it is thought that viral mRNA synthesis involves initiation of transcription at ten different promoter sites. It is likely that in some manner the frequency of an initiation of transcription at nine of these sites is affected by one or more viral gene products. With the recent development of soluble in vitro transcription systems that respond to exogenously added DNA, it should be possible to begin to study regulation of gene expression at this stage of transcription. At present, these systems yield the paradoxical observation that extracts prepared from uninfected human cells more efficiently recognize the late promoter as compared to the early promoter of adenovirus. As more is learned about regulation of synthesis of viral mRNAs, examples will surely be found where RNA processing and RNA turnover play a critical role in determining the level of mRNAs. Such cases are more likely to appear in the balancing of synthesis of different mRNAs derived from one transcriptional unit. Few experiments have been directed to this possibility and the study of adenovirus molecular biology is only now entering the age of maturity where these experiments are feasible.

Expression of a transposable antibiotic resistance element in Saccharomyces

Some eukaryotic genes can be expressed in bacteria but there are few examples of the expression of prokaryotic genes in eukaryotes. Antibiotic G418 is a 2-deoxystreptamine antibiotic that is structurally related to gentamicin but has inhibitory activity against a much wider variety of pro- and eukaryotic organisms. In bacteria, resistance to G418 can be determined by several plasmid-encoded modifiying enzymes and, in view of the broad spectrum of activity of G418, we considered that this antibiotic might be useful as a selective agent for the introduction of these antibiotic resistance genes into a eukaryotic organism such as Saccharomyces cerevisiae. Additional impetus for these experiments came from the knowledge that certain of the G418-resistance determinants in bacteria are carried on transposable elements; a study of the properties of these elements in eukaryotes would be intriguing.

Mutations affecting segment number and polarity in Drosophila

In systematic searches for embryonic lethal mutants of Drosophila melanogaster we have identified 15 loci which when mutated alter the segmental pattern of the larva. These loci probably represent the majority of such genes in Drosophila. The phenotypes of the mutant embryos indicate that the process of segmentation involves at least three levels of spatial organization: the entire egg as developmental unit, a repeat unit with the length of two segments, and the individual segment.

Transforming genes of neoplasms induced by avian lymphoid leukosis viruses

Oncogenic avian retroviruses can be classified into three groups: sarcoma viruses, acute leukaemia viruses and lymphoid leukosis viruses (LLVs). Sarcoma and acute leukaemia viruses transform fibroblasts and/or haematopoietic cells in culture and induce tumours with short latent periods in infected birds. In contrast, LLVs do not transform cells in vitro and require long latent periods before formation of neoplasms in vivo. The most frequent neoplasm induced by LLVs is malignant lymphoma of the bursa of Fabricius, but LLVs also induce other neoplasms, including sarcomas, nephroblastomas and erythroblastosis. The genomes of both sarcoma and acute leukaemai viruses contain specific genes responsible for viral oncogenicity, whereas the genome of LLVs apparently includes only genes required for virus replication. The genetic basis for the low oncogenic potential of LLVs is therefore obscure. The present experiments indicate that LLV-induced tumours contain transforming genes that can be detected by transfection of NH 3T3 mouse cells. These transforming genes are not linked to LLV DNA sequences, suggesting that oncogenesis by LLVs may result from indirect activation of cellular transforming genes.