Search

DNA from lambdagt-lambdaB bacteriophage was cleaved with EcoRI endonuclease and fragments from EcoRI-digested E. coli DNA were inserted. This DNA was used to infect E. coli, and phages containing the gene for DNA ligase were isolated by genetic selection. Two different hybrids were found with the same E. coli segment inserted in opposite orientations. Both hybrids produced similar levels of ligase as measured in crude extracts of infected cells.

A selective scheme is presented for isolating sequences capable of replicating autonomously in the yeast Saccharomyces cerevisiae. YIp5, a vector that contains the yeast gene ura3, does not transform a ura3 deletion mutant to Ura+. Hybrid YIp5-Escherichia coli DNA molecules also fail to produce transformants. However, collections of molecular hybrids between YIp5 and DNA from any of six eukaryotes tested (S. cerevisiae, Neurospora crassa, Dictyostelium discoideum Ceanhorabditis elegans, Drosophila melanogaster, and Zea mays) do transform the deletion mutant. The Ura+ transformants grow slowly, are unstable under nonselective conditions, and carry the transforming DNA as autonomously replicating, supercoiled circular molecules. Such a phenotype is qualitatively identical to that of strains transformed by molecules containing a yeast chromosomal origin of replication. Thus, these DNA hybrid molecules may contain eukaryotic origins of replication. The isolated sequences may be useful in determiing the signals controlling DNA replication in yeast and in studying both DNA replication and transformation in other eukaryotic organisms.

The RecA protein of Escherichia coli is important for genetic recombination in vivo and can promote synapsis and strand exchange in vitro. The DNA pairing and strand exchange reactions have been well characterized in reactions with circular single strands and linear duplexes, but little is known about these two processes using substrates more characteristic of those likely to exist in the cell. Single-stranded linear DNAs were prepared by separating strands of duplex molecules or by cleaving single-stranded circles at a unique restriction site created by annealing a short defined oligonucleotide to the circle. Analysis by gel electrophoresis and electron microscopy revealed that, in the presence of RecA and single-stranded binding proteins, a free 3' homologous end is essential for stable joint molecule formation between linear single-stranded and circular duplex DNA.

A technique is described for physically positioning any cloned DNA on a native or artificial Saccharomyces cerevisiae chromosome. The technique involves splitting a chromosome at a specific site by transformation with short linear molecules containing the cloned DNA at one end and telomeric sequences at the other. Recombination between the end of the linear molecules and homologous chromosomal sequences gives rise to chromosome fragments comprising all sequences distal or proximal to the mapping site depending on the orientation of the cloned DNA. The recombinant products are recovered by screening for stabilization of a suppressor tRNA on the linear molecules using a colony color assay. The cloned DNA is positioned relative to the chromosome ends by sizing the chromosomal fragments using alternating contour-clamped homogeneous electric field gel electrophoresis. Application of this technique to organisms other than S. cerevisiae and to the analysis of exogenous DNA cloned in yeast is discussed.

We present evidence that a subset of mRNAs in the human parasitic trematode Schistosoma mansoni contain an identical 36-nucleotide spliced leader (SL) sequence at their 5' termini. The SL is derived from a 90-nucleotide nonpolyadenylylated RNA (SL RNA), presumably by trans-splicing. Neither the SL nor the SL RNA share significant sequence identity with previously described trans-spliced leaders and SL RNAs in trypanosomatid protozoans or nematodes. However, several features, such as predicted secondary structure, trimethylguanosine cap, and potential Sm binding site, suggest similarities among SL RNAs in widely divergent organisms. Our evidence also indicates that the exon 3 acceptor site of the 3-hydroxy-3-methylglutaryl-CoA reductase gene can be spliced either to the SL by trans-splicing or to an upstream exon, 2, by cis-splicing. The presence of a SL sequence in S. mansoni, a member of the phylum Platyhelminthes, suggests that transplicing may be a common feature of other lower invertebrates.

The product of the human c-myc protooncogene (Myc) is a sequence-specific DNA binding protein. Here, we demonstrate that the placement of the specific Myc DNA binding site CACGTG upstream of a luciferase reporter gene conferred Myc-stimulated expression that was inhibited by the overexpression of the basic-helix-loop-helix/leucine zipper protein Max. It was observed that Myc was phosphorylated in vivo within the NH2-terminal domain at Thr-58 and Ser-62. Replacement of these phosphorylation sites with Ala residues caused a marked decrease in Myc-stimulated reporter gene expression. In contrast, the replacement of Thr-58 or Ser-62 with an acidic residue (Glu) caused only a small inhibition of transactivation. Together, these data demonstrate that the NH2-terminal phosphorylation sites Thr-58 and Ser-62 are required for high levels of transactivation of gene expression by Myc.

Previous work identified a protein kinase activity that phosphorylates the epidermal growth factor (EGF) receptor at Thr669. An assay for this protein kinase activity present in homogenates prepared from A431 human epidermoid carcinoma cells was developed using a synthetic peptide substrate corresponding to residues 663-681 of the EGF receptor (peptide T669). Here we report that a greater initial rate of T669 phosphorylation was observed in experiments using homogenates prepared from EGF- or phorbol ester-treated cells compared with control cells. EGF and 4 beta-phorbol 12-myristate 13-acetate (PMA) caused a 6-fold and a 2-fold increase in protein kinase activity, respectively. A kinetic analysis of T669 phosphorylation demonstrated that the increase in protein kinase activity observed was accounted for by an increase in Vmax. To examine the interaction between protein kinase C and signal transduction by the EGF receptor, the effect of pretreatment of cells with PMA on the subsequent response to EGF was investigated. Treatment of cells with PMA caused greater than 90% inhibition of the EGF-stimulated tyrosine phosphorylation of the EGF receptor and abolished the EGF-stimulated formation of soluble inositol phosphates. In contrast, PMA was not observed to inhibit the stimulation of T669 protein kinase activity caused by EGF. Thus, the apparent functional desensitization of the EGF receptor caused by PMA does not inhibit signal transduction mediated by the T669 protein kinase. Our results demonstrate that EGF receptor transmodulation alters the pattern of signal-transduction pathways that are utilized by the EGF receptor.

A recombinant DNA strategy has been used systematically to survey the Mycobacterium tuberculosis genome for sequences that encode specific antigens detected by monoclonal antibodies. M. tuberculosis genomic DNA fragments with randomly generated endpoints were used to construct a large lambda gt11 recombinant DNA expression library. Sufficient numbers of recombinants were produced to contain inserts whose endpoints occur at nearly every base pair in the pathogen genome. Protein antigens specified by linear segments of pathogen DNA and produced by the recombinant phage of Escherichia coli were screened with monoclonal antibody probes. This approach was coupled with an improved detection method for gene isolation using antibodies to clonally isolate DNA sequences that specify polypeptide components of M. tuberculosis. The methodology described here, which is applicable to other pathogens, offers possibilities for the development of more sensitive and specific immunodiagnostic and seroepidemiological tests for tuberculosis and, ultimately, for the development of more effective vaccines.

Experimental allergic encephalomyelitis (EAE) is an autoimmune disease of the central nervous system that serves as a model for the human disease multiple sclerosis. We evaluated rolipram, a type IV phosphodiesterase inhibitor, for its efficacy in preventing EAE in the common marmoset Callithrix jacchus. In a blinded experimental design, clinical signs of EAE developed within 17 days of immunization with human white matter in two placebo-treated animals but in none of three monkeys that received rolipram (10 mg/kg s.c. every other day) beginning 1 week after immunization. In controls, signs of EAE were associated with development of cerebrospinal fluid pleocytosis and cerebral MRI abnormalities. In the treatment group, there was sustained protection from clinical EAE, transient cerebrospinal fluid pleocytosis in only one of three animals, no MRI abnormality, and marked reduction in histopathologic findings. Rolipram-treated and control animals equally developed circulating antibodies to myelin basic protein. Thus, inhibition of type IV phosphodiesterase, initiated after sensitization to central nervous system antigens, protected against autoimmune demyelinating disease.

A cloned segment of yeast DNA containing the structural gene for imidazoleglycerolphosphate dehydratase (D-erythro-imidazoleglycerolphosphate hydro-lase, EC 4.2.1.19) is transcribed and translated in Escherichia coli with sufficient fidelity to produce functional enzyme. This segment of yeast DNA was isolated as a viable molecular hybrid of bacteriophage lambda (lambdagt-Sc2601) which complements a nonrevertible hisB auxotroph of E. coli lacking dehydratase activity. The equivalent segments of DNA cloned from two independent his3 mutants of yeast lacking IGP dehydratase activity do not complement the hisB auxotroph. The two nonfunctional his3 alleles cloned in bacteriophage lambda can be recombined in E. coli to generate a hybrid phage which complements the hisB auxotroph. The dehydratase activity produced in E. coli by the cloned segment of yeast DNA strongly resembles the activity found in yeast.

Investigation of the efficacy of pharmacologic agents affecting myocardial infarct size after coronary artery occlusion is complicated by the variability of collateral flow among experimental animals which results in variability of infarct size. To overcome this difficulty, we developed an autoradiographic method to delineate the ischemic area at risk of necrosis after coronary artery occlusion and we invetigated the potential protective effect of a calcium antagonist verapamil. The left anterior descending coronary arteries of 25 barbiturate-anesthetized dogs were occluded. Thirty minutes later, highly radioactive human albumin microspheres labeled with 99mTc were injected into the left atrium. One hour after coronary artery occlusion, dogs were randomized to control or treated groups; the latter received a 0.2 mg/kg loading dose and 0.6 mg/kg per hr maintenance dose of verapamil intravenously. Eight hours after coronary artery occlusion, the dogs were killed, the hearts were excised, and the left ventricle was sectioned parallel to the atrioventricular groove; infarct size was determined planimetrically after incubation in triphenyl tetrazolium chloride. The slices were then exposed to high-speed x-ray film with image-enhancing screens. The percentage of left ventricle that was ischemic, as determined by planimetry of autoradiographs, was similar in treated and control animals (36.6 +/- 2.0% compared to 37.3 +/- 2.8%, respectively). Of the ischemic area, 92.0 +/- 4.3% was infarcted in control animals and 70.5 +/- 5.1% was infarcted in treated animals (P < 0.01). Thus, this autoradiographic method using 99mTc-labeled human albumin microspheres is useful in delineating the area of ischemia after coronary artery occlusion and in evaluating the efficacy of pharmacologic agents designed to protect ischemic myocardium. Verapamil, administered 1 hr after coronary artery occlusion, is effective in limiting infarct size.

cDNA clones encoding the 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase [(S)-mevalonate:NADP+ oxidoreductase (CoA-acylating), EC 1.1.1.34] from the human parasite Schistosoma mansoni have been isolated and characterized. The composite 3459 base pairs of cDNA sequence contains a 2844-base-pair open reading frame corresponding to a protein of 948 amino acids. The predicted S. mansoni HMG-CoA reductase protein contains a hydrophobic amino terminus consisting of seven potential transmembrane domains that are structurally conservative but are not identical in amino acid sequence with HMG-CoA reductases from other species. The hydrophilic carboxyl terminus of the S. mansoni HMG-CoA reductase protein, however, shares 48-52% sequence identity with the carboxyl termini of other HMG-CoA reductases in a region that contains the catalytic domain. When expressed as a fusion protein in Escherichia coli, the carboxyl-terminal domain of the schistosome protein exhibits HMG-CoA reductase enzyme activity.

We wished to identify metabolic signals governing changes in ornithine decarboxylase (L-ornithine carboxy-lyase, EC 4.1.1.17) activity in Neurospora crassa. By manipulations of the ornithine supply and by the use of inhibitors of the polyamine pathway, we found that spermidine negatively governs formation of active ornithine decarboxylase and that putrescine promotes inactivation of the enzyme. Direct addition of putrescine or spermidine to cycloheximide-treated cells confirmed the role of putrescine in enzyme inactivation and showed that spermidine had no effect on this process. Increases in ornithine decarboxylase activity caused by blocking spermidine synthesis occurred prior to a significant decrease in the spermidine pool. This is consistent with our previous finding that only 10-20% of the spermidine pool is freely diffusible within N. crassa cells. We presume that only this small fraction of the pool is active in regulation.

Addition of platelet-derived growth factor (PDGF) to quiescent WI-38 human fetal lung fibroblasts mimics the effect of tumor-promoting phorbol diesters to inhibit the high-affinity binding of 125I-labeled epidermal growth factor (125I-EGF). PDGF, like phorbol diesters, was found to increase the phosphorylation state of EGF receptors immunoprecipitated from intact fibroblasts that were labeled to equilibrium with [32P]phosphate. Phosphoamino acid analysis of the EGF receptors indicated that both PDGF and phorbol diesters increased the level of [32P]phosphoserine and [32P]phosphothreonine. Phosphopeptide mapping of the EGF receptor demonstrated that PDGF increased the phosphorylation of several sites and induced the phosphorylation of a site that was not observed to be phosphorylated on EGF receptors isolated from control cells. This latter phosphorylation site on the EGF receptor was identified as threonine-654, previously shown to be phosphorylated in response to phorbol diesters in intact cells or by purified protein kinase C in vitro. Further, it was observed that PDGF mimicked the action of phorbol diesters to inhibit the EGF-dependent tyrosine phosphorylation of the EGF receptor in [32P]phosphate-labeled fibroblasts. These results are consistent with the hypothesis that increases in diacylglycerol and Ca2+ levels caused by addition of PDGF to fibroblasts activate protein kinase C and that this kinase, at least in part, mediates the effect of PDGF on the phosphorylation of the EGF receptor. The data further suggest that protein kinase C may play an important role in the regulation of cellular metabolism and proliferation by PDGF.

Evidence from prokaryotic systems suggests that enzymes of dNTP synthesis are organized near the DNA replication apparatus, allowing direct utilization of dNTPs at their sites of synthesis. To investigate whether similar interactions exist within a eukaryotic environment, we have prepared anti-idiotypic antibodies to the small subunit of vaccinia virus ribonucleotide reductase, and we used these antibodies to search for proteins that interact with this enzyme. This approach identified a 34-kDa viral phosphoprotein, which, like ribonucleotide reductase itself, is localized within infected cells at DNA replication sites. After expression of its structural gene in Escherichia coli, the recombinant protein was purified and found (i) to bind tightly to single-stranded DNA and (ii) to stimulate enzymatic activity of vaccinia ribonucleotide reductase. These observations suggest a physical association between dNTP synthesis and DNA replication in this viral system.

This work reports the identification, characterization, and nucleotide sequence of STE20, a newly discovered gene involved in the Saccharomyces cerevisiae mating response pathway, to date one of the best understood signal transduction pathways. STE20 encodes a putative serine/threonine-specific protein kinase with a predicted molecular mass of 102 kDa. Its expression pattern is similar to that of several other protein kinases in the mating response pathway. Deletion of the kinase domain of STE20 causes sterility in both haploid mating types. This sterility can be partially suppressed by high-level production of STE12 but is not suppressible by high levels of STE4 or a dominant STE11 truncation allele. A truncation allele of STE20 was isolated that can activate the mating response pathway in the absence of exogenous mating pheromone. This allele causes dominant growth arrest that cannot be suppressed by deletions of STE4, STE5, STE7, STE11, or STE12. The allele is able to suppress the mating defect of a strain in which the STE20 kinase domain has been deleted, but not the mating defects of strains carrying mutations in STE4, STE5, STE7, STE11, or STE12.

We have explored the mechanism of genomic rearrangement in a hamster fibroblast cell culture system in which rearrangements are induced 5' to the endogenous thymidine kinase gene by chemical carcinogen treatment. The wild-type region around one rearrangement breakpoint was cloned and sequenced. With this sequence information, the carcinogen-induced rearrangement was cloned from the corresponding rearranged cell line by the inverse polymerase chain reaction. After the breakpoint fragment was sequenced, the wild-type rearrangement partner (RP15) was isolated by a second inverse polymerase chain reaction of unrearranged DNA. Comparison of the sequence of the rearrangement breakpoint with the wild-type RP15 and 5' thymidine kinase gene regions revealed short repeats directly at the breakpoint, as well as nearby A + T-rich regions in each rearrangement partner. Pulsed-field electrophoresis analysis demonstrated that this rearrangement is an interstitial deletion of 35 kilobases. Southern blot analysis of the RP15 region in unrearranged parental cells showed a demethylated CpG island and a complex of DNase I-hypersensitive sites adjacent to the breakpoint in the region deleted by the rearrangement. Therefore, these studies reveal interesting sequence and chromatin features near the rearrangement breakpoints and suggest a role for nuclear organization in the mechanism of carcinogen-induced genomic rearrangement.

Promoters that allow the selective induction of gene expression in vivo constitute an important methodology in eukaryotic organisms such as yeast and the fruit fly, but to date no such system has been described for higher plants. Given the fact that mammalian steroid receptors can function as hormone-dependent inducers of gene expression in heterologous systems, the feasibility of using mammalian steroid hormones as selective inducers of plant gene expression was investigated. Here it is shown that the glucocorticoid receptor expressed in plant cells is capable of activating a test gene linked to glucocorticoid response elements, providing the transfected plant cells are treated with glucocorticoid hormone. Nanomolar concentrations of glucocorticoids are sufficient to induce gene expression more than 150-fold, without causing detectable alterations in the physiology of the cultured plant cells. These findings indicate that glucocorticoid induction of steroid-responsive promoters should provide a general method for regulating gene expression in plant cells and imply that such a system might ultimately function in whole plants such as Arabidopsis thaliana.

A technique employing ferritin-conjugated antibody has been developed to visualize specific protein-DNA complexes in the electron microscope and has been used to demonstrate the preferential binding of simian virus 40 (SV40) T antigen at or near the origin of replication of SV40 DNA, 0.67 fractional length clockwise from the EcoRI restriction endonuclease cleavage site. urified covalently closed supercoiled circles of SV40DNA were treated with partially purified T antigen and the complex was stabilized by crosslinking with glutaraldehyde. Hamster antiT antigen gamma-globulin, ferritin-labeled goat anti-hamster gamma-globulin, and glutaraldehyde were then added sequentially. The location of the bound ferritin cores was measured with respect to the EcoRI cleavage site and the orientation of the cores relative to the ends of the DNA was determined with respect to the locations of Escherichia coli DNA unwinding protein, which binds to covalently closed supercoiled SV40 DNA at either of two preferred sites, 0.46 or 0.90 fractional length clockwide from the EcoRI cleavage site.

A set of vector DNAs (Y vectors) useful for the cloning of DNA fragments in Saccharomyces cerevisiae (yeast) and in Escherichia coli are characterized. With these vectors, three modes of yeast transformation are defined. (i) Vectors containing yeast chromosomal DNA sequences (YIp1, YIp5) transform yeast cells at low frequency (1--10 colonies per microgram) and integrate into the genome by homologous recombination; this recombination is reversible. (ii) Hybrids containing endogenous yeast plasmid DNA sequences (YEp2, YEp6) transform yeast cells at much higher frequency (5000--20,000 colonies per microgram). Such molecules replicate autonomously with an average copy number of 5--10 covalently closed circles per yeast cell and also replicate as a chromosomally integrated structure. This DNA may be physically isolated in intact form from either yeast or E. coli and used to transform either organism at high frequency. (iii) Vectors containing a 1.4-kilobase yeast DNA fragment that includes the centromere linked trp1 gene (YRp7) transform yeast with an efficiency of 500--5000 colonies per microgram; such molecules behave as minichromosomes because they replicate autonomously but do not integrate into the genome. The uses of Y vectors for the following genetic manipulations in yeast are discussed: isolation of genes; construction of haploid strains that are merodiploid for a particular DNA sequence; and directed alterations of the yeast genome. General methods for the selection and the analysis of these events are presented.

We have developed a method that may be of general application for the stable introduction of foreign sequences or deletions, constructed in vitro, into the chromosomes of Saccharomyces cerevisiae. No vector sequences are present in the final strains. Ability to transform cells with DNA, availability of a single selective marker, and integration of the transforming DNA by homologous recombination into the chromosomes are the requirements of the system. Any isolated gene can be deleted or altered and then be used to replace the wild-type chromosomal copy. An internal deletion mutant of the his3 gene and a transposition of a galactose-inducible region into chromosome XV have been generated by using the ura3 gene as the selective marker.

Homeobox genes are present in both plants and animals. Homeobox-leucine zipper genes, however, have been identified thus far only in the small mustard plant Arabidopsis thaliana. This observation suggests that homeobox-leucine zipper genes evolved after the divergence of plants and animals, perhaps to mediate specific regulatory events. To better understand this gene family, we isolated several sequences containing the homeobox-leucine zipper motif and carried out a comparative analysis of nine homeobox-leucine zipper genes (HAT1, HAT2, HAT3, HAT4, HAT5, HAT7, HAT9, HAT14, and HAT22). Gene structures, sequence comparisons, and chromosomal locations suggest a simple model for the evolution of these genes. The model postulates that a primordial homeobox gene acquired a leucine zipper by exon capture. The nascent homeobox-leucine zipper gene then appears to have undergone a series of gene duplication and chromosomal translocation events, leading to the formation of the HAT gene family. This work has general implications for the evolution of regulatory genes.

In Saccharomyces cerevisiae, numerous genes have been identified by selection from high-copy-number libraries based on "multicopy suppression" or other phenotypic consequences of overexpression. Although fruitful, this approach suffers from two major drawbacks. First, high copy number alone may not permit high-level expression of tightly regulated genes. Conversely, other genes expressed in proportion to dosage cannot be identified if their products are toxic at elevated levels. This work reports construction of a genomic DNA expression library for S. cerevisiae that circumvents both limitations by fusing randomly sheared genomic DNA to the strong, inducible yeast GAL1 promoter, which can be regulated by carbon source. The library obtained contains 5 x 10(7) independent recombinants, representing a breakpoint at every base in the yeast genome. This library was used to examine aberrant gene expression in S. cerevisiae. A screen for dominant activators of yeast mating response identified eight genes that activate the pathway in the absence of exogenous mating pheromone, including one previously unidentified gene. One activator was a truncated STE11 gene lacking approximately 1000 base pairs of amino-terminal coding sequence. In two different clones, the same GAL1 promoter-proximal ATG is in-frame with the coding sequence of STE11, suggesting that internal initiation of translation there results in production of a biologically active, truncated STE11 protein. Thus this library allows isolation based on dominant phenotypes of genes that might have been difficult or impossible to isolate from high-copy-number libraries.

We have isolated a segment of DNA from the eukaryote Saccharomyces cerevisiae (baker's yeast) as a viable molecular hybrid of bacteriophage lambda DNA which, when integrated into the chromosome of an E. coli histidine auxotroph, allows this bacterium to grow in the absence of histidine. The nonrevertable, histidine auxotroph lacks the enzymatic activity of imidazole glycerol phosphate (IGP) dehydratase (EC 4.2.1.19). From genetic experiments, we conclude that expression of the segment of yeast DNA results in the production of a diffusible substance and that transcription necessary for the complementation is most likely initiated from the segment of eukaryotic DNA.

RNA can hybridize to double-stranded DNA in the presence of 70% formamide by displacing the identical DNA strand. The resulting structure, called an R-loop, is formed in formamide probably because of the greater thermodynamic stability of the RNA-DNA hybrid when it is near the denaturation temperature of duplex DNA. The rate of R-loop formation is maximal at the temperature at which half of the duplex DNA is irreversibly converted to single-stranded DNA (the strand separation temperature of tss) of the duplex DNA and falls precipitously a few degrees above or below that temperature. This maximal rate is similar to the rate of hybridization of RNA to single-stranded DNA under the same conditions. At temperatures above the tss the rate is proportional to the RNA concentration. However, at temperatures below tss the rate of R-loop formation is less dependent upon the RNA concentration. Once formed, the R-loops display considerable stability; the formamide can be removed and the DNA can be cleaved with restriction endonucleases without loss of R-loop structures.

Homeobox genes encode a large family of homeodomain proteins in animal systems. To test whether such genes are also abundant in higher plants, degenerate oligonucleotides complementary to sequences encoding the recognition helix (helix three) of the homeodomain were used to screen genomic and cDNA libraries from the plant Arabidopsis thaliana. Analysis of 8 of the 41 cDNAs isolated revealed that each encodes a presumptive homeodomain; interestingly, most of these clones also contain a leucine zipper motif tightly linked to the homeodomain. It is concluded that Arabidopsis encodes a large family of homeodomain proteins, including members that contain a homeodomain/leucine-zipper (HD-Zip) motif.

An automated oligonucleotide synthesizer has been developed that can simultaneously and rapidly synthesize up to 96 different oligonucleotides in a 96-well microtiter format using phosphoramidite synthesis chemistry. A modified 96-well plate is positioned under reagent valve banks, and appropriate reagents are delivered into individual wells containing the growing oligonucleotide chain, which is bound to a solid support. Each well has a filter bottom that enables the removal of spent reagents while retaining the solid support matrix. A seal design is employed to control synthesis environment and the entire instrument is automated via computer control. Synthesis cycle times for 96 couplings are < 11 min, allowing a plate of 96 20-mers to be synthesized in < 5 hr. Oligonucleotide synthesis quality is comparable to commercial machines, with average coupling efficiencies routinely > 98% across the entire 96-well plate. No significant well-to-well variations in synthesis quality have been observed in > 6000 oligonucleotides synthesized to date. The reduced reagent usage and increased capacity allow the overall synthesis cost to drop by at least a factor of 10. With the development of this instrument, it is now practical and cost-effective to synthesize thousands to tens of thousands of oligonucleotides.

We describe a genetic selection method designed to facilitate the cloning of genes encoding sequence-specific DNA-binding proteins. The strategy selects for clones expressing particular sequence-specific DNA-binding activities from a library of clones encoding other, nonspecific proteins. Specific DNA-binding sites have been placed near the start of transcription of the strong synthetic conII promoter to create promoters that can be repressed by the corresponding sequence-specific DNA-binding proteins. Transcription from the conII derivatives in the absence of repression interferes with the phenotypic expression of an adjacent drug-resistance gene, aadA. Sequence-specific DNA-binding proteins are shown to repress these promoters and alleviate transcriptional interference of aadA, resulting in drug resistance in cells expressing the appropriate DNA-binding protein.

A probe representing the Drosophila dunce+ (dnc+) gene, the structural gene for a cAMP phosphodiesterase (PDEase), detects homologous sequences in many different organisms, including mouse, rat, and human. Genomic and cDNA clones representing a homolog of the Drosophila dnc+ gene were isolated from rat libraries and characterized. This gene has been named ratdnc-1. One cDNA clone defines a large open reading frame of approximately 1.8 kilobases (kb), predicting a protein sequence of 610 amino acids with significant homology to a conserved domain of approximately 275 residues found in most other PDEases. The amino acid identity value to the Drosophila cAMP PDEase within this domain is a striking 75%. Other cDNA clones show blocks of sequence divergence from this cDNA clone close to the predicted N terminus, indicating the potential existence of a family of related enzymes encoded by alternatively spliced messenger RNAs from ratdnc-1. Genomic blotting experiments suggest the existence of at least one other rat gene with homology to ratdnc-1. RNAs homologous to ratdnc-1 are heterogeneous in size between tissues, with heart containing a major transcript of 4.4 kb and brain one of 4.0 kb. The potential identity of the product of the ratdnc-1 gene with known PDEases is discussed.

beta-Actin mutations in chemically transformed human cell lines have been associated with tumorigenicity, an association consistent with other evidence suggesting that altered cytoskeletal proteins may have an important role in cancer initiation or progression. From a human promyelocytic leukemia cell line, we have isolated a gamma-actin cDNA clone with amino acid substitutions in a region highly conserved in the many actins analyzed. To our knowledge, this is the first example of a variant gamma-actin in a human neoplasm. A separate finding from the analysis of this clone is that the gamma-actin 3'-untranslated region is among the most highly conserved of all 3'-untranslated sequences so far reported, but is entirely different from the beta-actin 3'-untranslated region. The high degree of evolutionary conservation suggests that the 3'-untranslated regions of these two mRNAs have important and distinct functional roles that were already fully differentiated more than 100 million years ago. Mutations affecting four major cytoskeletal components have now been identified in human neoplastic cells. These findings suggest that mutated cytoskeletal genes may be members of a class of oncogenes, fundamentally different from both the nuclear-acting (e.g., myc and simian virus 40 large tumor antigen) and growth factor/receptor/protein kinase-related (e.g., sis, erbB, and ras) types of oncogenes.

The sequence of nitrate reductase (EC 1.6.6.1) mRNA from the plant Arabidopsis thaliana has been determined. A 3.0-kilobase-long cDNA was isolated from a lambda gt10 cDNA library of Arabidopsis leaf poly(A)+ RNA. The cDNA hybridized to a 3.2-kilobase mRNA whose level increased 15-fold in response to treatment of the plant with nitrate. An open reading frame encoding a 917 amino acid protein was found in the sequence. This protein is very similar to tobacco nitrate reductase, being greater than 80% identical within a section of 450 amino acids. By comparing the Arabidopsis protein sequence with other protein sequences, three functional domains were deduced: (i) a molybdenum-pterin-binding domain that is similar to the molybdenum-pterin-binding domain of rat liver sulfite oxidase, (ii) a heme-binding domain that is similar to proteins in the cytochrome b5 superfamily, and (iii) an FAD-binding domain that is similar to NADH-cytochrome b5 reductase.

A "transplason" mutagenesis procedure was developed for the dual purposes of low resolution mapping of antigenic coding regions (using transposons) and constructing insertion mutations in yeast genes (by transplacement). Mini-Tn10 transposon derivatives containing both Escherichia coli and yeast selectable markers have been constructed. These elements are used to mutagenize lambda gt11 clones that express foreign antigens in E. coli. The transposition events are first selected in E. coli, and the effect of these insertions on antigen expression is used to locate the antigenic coding regions on the cloned DNA. Insertion mutations located within a desired yeast sequence are then substituted for the genomic copies by one-step gene transplacement. This provides a powerful method for rapidly mapping antigenic coding sequences of cloned genes and inactivating these genes in yeast to help determine their function. Several examples using this technique are presented.

Dicentric plasmids containing either two copies of centromere 4 or one copy of centromere 4 and one copy of centromere 3 in the yeast plasmid vector YRp17 were constructed in vitro and introduced into yeast cells by DNA transformation. The resulting colonies were heterogeneous for a mixed population of rearranged plasmids. The rearrangements always involved deletion of one or both centromere sequences originally present on the plasmid. Heterogeneity was due to the continued production of deleted plasmids from a pool of unrearranged dicentric plasmids maintained within some of the yeast cells in the colony. The RAD52 gene product is known to be required for the repair of DNA double-strand breaks in yeast. Transformation of rad52 mutant yeast cells with dicentric plasmids gave rearranged plasmids similar to those observed with RAD+ yeast cells, but the transformation frequency was only 5-10% compared to transformation with monocentric plasmids. Also, the ratio of unrearranged dicentric plasmid to deleted plasmids was greatly reduced in the rad52-transformed cells. These observations are consistent with a model in which centromeric DNA sequences can interact independently with the yeast cell spindle apparatus. Occasional movement of centromeres to opposite poles may result in mechanical breakage of plasmid sequences. Plasmids deleted for one or both centromere sequences can be obtained from these broken molecules and are resistant to further rearrangement.

Retrovirus-transformed rat cells that actively express transforming growth factor type alpha (TGF-alpha) release into the medium a soluble protein of 17-19 kDa that has been identified as a precursor for TGF-alpha. The identification of this protein as a TGF-alpha precursor is based on its recognition by specific antibodies and by the ability of elastase to convert this protein into a 6-kDa polypeptide with the properties of mature TGF-alpha. This TGF-alpha precursor binds to epidermal growth factor/TGF-alpha receptors and activates the receptor-associated tyrosine kinase activity in intact cells. The biological potency of this precursor is not markedly increased by conversion into mature TGF-alpha in vitro. These studies demonstrate the ability of transformed cells to release a TGF-alpha precursor capable of strong mitogenic action in vivo.

The effect of tumor-promoting phorbol diesters to potentiate the action of epidermal growth factor (EGF) on cell proliferation is associated with phosphorylation of EGF receptors, acute depression of EGF binding, and inhibition of EGF receptor tyrosine kinase activity. In the present studies, normal human fibroblasts and A431 carcinoma cells were labeled with [32P]phosphate and treated with and without 10 nM 4 beta-phorbol 12 beta-myristate 13 alpha-acetate (PMA). The EGF receptors then were isolated by immunoprecipitation and digested with trypsin. Analysis of the labeled receptor phosphopeptides by reversed-phase HPLC revealed that PMA induces the phosphorylation of a unique phosphopeptide containing [32P]phosphothreonine. Comparison of several chemical and physical properties of the 32P-labeled phosphopeptide with the primary structure of the EGF receptor suggested the identify Lys-Arg-Thr(P)-Leu-Arg. This was confirmed by direct demonstration that a synthetic peptide of this structure comigrates during HPLC and electrophoresis with the 32P-labeled phosphopeptide isolated from the EGF receptors of normal human fibroblasts. The phosphorylated site on the peptide corresponds to threonine-654 of the EGF receptor, which is located on the cytoplasmic side of the plasma membrane nine residues distant from the transmembrane domain. These data indicate that phosphorylation of the EGF receptor in human fibroblasts and A431 cells at threonine-654 may regulate the EGF receptor tyrosine kinase activity and the binding of EGF.

Three recombination events, reciprocal recombination, sister-chromatid recombination, and gene conversion, were studied using substrates designed in vitro. Each type of recombination event can be monitored at any chromosomal location. We have shown that sister-chromatid recombination is induced mitotically by DNA damaging agents, such as methyl methanesulfonate and gamma-rays, but is decreased mitotically in strains defective in rad52. Reciprocal recombination by which circular plasmids integrate into the genome is unaffected by rad52 defective alleles and occurs by a different recombination pathway. Mechanisms are suggested by which gene conversion between sister chromatids can generate chromosome rearrangements.

MyoD is a master regulatory gene for myogenesis. Under the control of a retroviral long terminal repeat, MyoD was expressed in a variety of differentiated cell types by using either a DNA transfection vector or a retrovirus. Expression of muscle-specific proteins was observed in chicken, human, and rat primary fibroblasts and in differentiated melanoma, neuroblastoma, liver, and adipocyte lines. The ability of MyoD to activate muscle genes in a variety of differentiated cell lines suggests that no additional tissue-specific factors other than MyoD are needed to activate the downstream program for terminal muscle differentiation or that, if such factors exist, they are themselves activated by MyoD expression.

The RNA homologous to the yeast transposable element Ty1 is one of the more abundant poly(A)+ RNAs in many strains of the yeast Saccharomyces cerevisiae. The 5' and 3' ends of Ty1 RNA have been determined from analysis of cDNA. The 5' end is 245 bases into the left delta sequence measured from the left side of the Ty1 element. The delta sequence is a direct repeat of about 340 base pairs present at each end of the Ty1 element. The Ty1 transcription includes 93-97 bases of the left delta sequence and continues through the entire internal portion of the element and through about 295 bases of the right delta sequence before reaching the 3' end located 38-46 bases from the right side of the right delta sequence. Because the delta sequences present at each end of a single Ty1 element have identical or very similar DNA sequences, these end points for Ty1 RNA raise several questions about the expression of Ty1 elements. First, what are the initiation and termination signals, because the Ty1 transcript must read through a DNA sequence that is identical to the 3' end at about 50 bases from the 5' end? Second, why is the direction of transcription of the Ty1 element opposite to that of genes that are overexpressed after the insertion of a Ty1 element? Third, because the Ty1 RNA itself has direct repeats of about 45 bases, a structure analogous to retrovirus RNAs, is the Ty1 RNA an intermediate in the transposition of Ty1?

A sensitive and general technique has been devised for the dual purposes of cloning genes by using antibodies as probes and isolating unknown proteins encoded by cloned DNA. The method uses an expression vector, lambda gt11 (lac5 nin5 cI857 S100), that permits insertion of foreign DNA into the beta-galactosidase structural gene lacZ and promotes synthesis of hybrid proteins. Efficient screening of antigen-producing clones in lambda gt11 recombinant cDNA libraries is achieved through lysogeny of the phage library in hflA (high-frequency lysogeny) mutant cells of Escherichia coli; lysogens produce detectable quantities of antigen on induction, even when plated at high cell densities. The vector is also designed to facilitate the isolation of proteins specified by previously cloned gene sequences. Hybrid proteins encoded by recombinant phage accumulate in strains defective in protein degradation (lon mutants) in amounts amenable to large-scale purification. Antibodies produced against the portion of the hybrid encoded by foreign DNA could in turn be used to isolate the native polypeptide from eukaryotic cells.

Hepatic lipase and lipoprotein lipase hydrolyze fatty acids from triacylglycerols and are critical in the metabolism of circulating lipoproteins. The two lipases are similar in size and amino acid sequence but are distinguished by functional differences in substrate preference and cofactor requirement. Presumably, these distinctions result from structural differences in functional domains. To begin localization of these domains, a chimeric lipase was constructed composed of the N-terminal 329 residues of rat hepatic lipase linked to the C-terminal 136 residues of human lipoprotein lipase. The chimera hydrolyzed both monodisperse short-chain (esterase) and emulsified long-chain (lipase) triacylglycerol substrates with catalytic and kinetic properties closely resembling those of native hepatic lipase. However, monoclonal antibodies to lipoprotein lipase inhibited the lipase activity, but not the esterase function, of the chimera. Therefore, the chimeric molecule is a functional lipase and contains elements and characteristics from both parental enzymes. It is proposed that the N-terminal domain, containing the active center from hepatic lipase, governs the catalytic character of the chimera, and the C-terminal domain is essential for hydrolysis of long-chain substrates.

We have isolated and sequenced cDNA clones representing portions of the polyadenylylated transcripts of the dunce+ gene. These define an open reading frame of 1086 bases and some of the 5'- and 3'-untranslated regions of the transcripts. The deduced amino acid sequence is strikingly homologous to the amino acid sequence of a Ca2+/calmodulin-dependent cyclic nucleotide phosphodiesterase isolated from bovine brain and more weakly related to the predicted amino acid sequence of a yeast cAMP phosphodiesterase. These homologies, together with prior genetic and biochemical studies, provide unambiguous evidence that dunce+ codes for a phosphodiesterase. In addition, the dunce+ gene product shares a seven-amino acid sequence with a regulatory subunit of cAMP-dependent protein kinase that is predicted to be part of the cAMP binding site. We also identify a weak homology between a region of the dunce+ gene product and the egg-laying hormone precursor of Aplysia californica. The open reading frame is divided in the genome by four introns.

MyoD is a master regulatory gene for myogenesis. Its product, the MyoD protein, appears to act by binding to muscle-specific enhancer sequences. We show that MyoD binds cooperatively to two sites in the muscle-specific creatine kinase enhancer; this is dramatically reflected in dissociation-rate measurements. A deletion of the acidic N terminus (residues 3-56) results in a protein that binds normally to single sites but fails to bind cooperatively to two adjacent sites, suggesting a role of the N terminus in cooperative interactions. In transfection assays, a reporter gene flanked by a single MyoD binding site fails to be activated by cotransfected MyoD expression vectors. In contrast, a reporter with two or more MyoD binding sites is activated by wild-type MyoD but not by N-terminally deleted MyoD. A reporter gene with a single binding site, although not activated by MyoD, can nonetheless compete for expression with a reporter gene containing three sites. Thus, in vivo, a single site can bind MyoD, but occupancy of two or more sites is required for subsequent transcriptional activation.

To explore the process of lipoprotein assembly, plasmids encoding truncated forms of apolipoprotein B (apoB) were transfected into Chinese hamster ovary (CHO) fibroblasts. (One, encoding apoB53, the N-terminal 53% of apoB100, can direct the assembly and secretion of lipoproteins when expressed in hepatoma cells, while the other, encoding the shorter apoB15, does not direct lipoprotein assembly.) Expression of apoB15 in CHO cells resulted in the accumulation of apoB15 protein in both medium and cells. In contrast, apoB was not detectable in medium or within CHO cells transfected with the plasmid encoding apoB53, despite the expression of apoB53 mRNA. ApoB53 did accumulate within transfected cells incubated with the thiol protease inhibitor N-acetylleucylleucylnorleucinal (ALLN), suggesting that it is synthesized but completely degraded in the absence of the inhibitor. ApoB53 was not secreted despite its presence within ALLN-treated cells. Essentially all the apoB53 that accumulated in microsomes from ALLN-treated cells was associated with the membrane and was susceptible to degradation by exogenous trypsin, indicating exposure on the cytoplasmic face of the membrane. Thus, translocation of apoB53 across the endoplasmic reticulum membrane is blocked. However, the apoB53 bound to concanavalin A, suggesting that it is glycosylated and therefore partly exposed to the lumen as well. ApoB requires a unique process, not expressed in CHO fibroblasts, for its complete translocation and entrance into the secretory pathway. This process might account for the inability of abetalipoproteinemic patients to secrete apoB.

This work describes a multifunctional phage lambda expression vector system, lambda YES, designed to facilitate gene isolation from eukaryotes by complementation of Escherichia coli and Saccharomyces cerevisiae mutations. lambda YES vectors have a selection for cDNA inserts using an oligo adaptor strategy and are capable of expressing genes in both E. coli and S. cerevisiae. They also allow conversion from phage lambda to plasmid clones by using the cre-lox site-specific recombination system, referred to here as automatic subcloning. A simple method has been developed for the conversion of any plasmid into a phage lambda cDNA cloning vector with automatic subcloning capability. cDNA libraries constructed in these vectors were used to isolate genes from humans and Arabidopsis thaliana by complementation of yeast and bacterial mutations, respectively.

The availability of multicopy plasmid vectors for the yeast Saccharomyces cerevisiae allows the selective amplification of individual segments of the genome. Increased dosage of particular genes results in overproduction of specific gene products and thereby confers resistance to certain metabolic inhibitors. Advantage was taken of this fact to isolate recombinant clones that increase the activities of the enzymes UDP-N-acetylglucosamine-1-P transferase and 3-hydroxy-3-methylglutaryl-CoA reductase.

A yeast artificial chromosome (YAC) library in Saccharomyces cerevisiae consisting of 30,000 clones with an average insert size of 0.1 megabase pair of human DNA has been generated from primary fibroblast DNA. A YAC vector was modified to enable the recovery of both ends of a human DNA insert in plasmids in Escherichia coli and to confer G418 resistance to mammalian cells. A rapid method for yeast colony hybridization was used that exploits the ability of yeast spheroplasts to regenerate in a thin layer of calcium alginate. This method permits direct replica plating and processing of colonies from the primary transformation plate to nitrocellulose filters. Yeast colony hybridization conditions have been established to identify, within a YAC library of human genomic DNA, artificial chromosomes with homology to human DNA probes of unique single-copy sequence. An artificial chromosome with a 0.1-megabase-pair insert from the human Xq28 region has been identified by hybridization to a DNA probe that detects a unique sequence near the 3' end of the factor VIII gene.