We have isolated and characterized a plasmid (pChug 20.1) that contains the cDNA of a nuclear uracil DNA glycosylase (UDG) gene Isolated from normal human placenta. This cDNA directed the synthesis of a fusion protein (Mr 66,000) that exhibited UDG activity. The enzymatic activity was specific for a uracil-containing polynucleotide substrate and was inhibited by a glycosylase antibody or a (3-galactosidase antibody. Sequence analysis demonstrated an open reading frame that encoded a protein of 335 amino acids of calculated Mr 36,050 and pI 8.7, corresponding to the Mr 37,000 and pI 8.1 of purified human placental UDG. No homology was seen between this cDNA and the UDG of herpes simplex virus, Escherichia coil, and yeast; nor was there homology with the putative human mitochondrial UDG cDNA or with a second human nuclear UDG cDNA. Surprisingly, a search of the GenBank data base revealed that the cDNA of UDG was completely homologous with the 37-kDa subunit of human glyceraldehyde-3-phosphate dehydrogenase. Human erythrocyte glyceraldehyde-3-phosphate dehydrogenase was obtained commercially in its tetrameric form. A 37-kDa subunit was isolated from it and shown to possess UDG activity equivalent to that seen for the purified human placental UDG.The multiple functions of this 37-kDa protein as here and previously reported indicate that it possesses a series of activities, depending on its oligomeric state. Accordingly, mutation(s) in the gene of this multifunctional protein may conceivably result in the diverse cellular phenotypes of Bloom syndrome.Human cells contain two major DNA excision-repair pathways to remove DNA lesions (1). Bulky DNA adducts are eliminated by the nucleotide-excision pathway, whereas most alkylated bases and alterations due to spontaneous damage are removed by the base-excision pathway. DNA glycosylases remove modified bases in the latter pathway by cleaving the base-sugar bond. Uracil present in DNA as a result of utilization of dUTP during DNA synthesis (2) or by deamination of existing cytosine residues (3, 4) is removed by the uracil DNA glycosylase (UDG).In an examination of the molecular mechanisms involved in expression of human nuclear DNA-repair genes, we isolated a normal human placental cDNA that hybrid-selected the mRNA encoding the nuclear UDG (5). Northern (RNA) blot analysis revealed the presence of a 1.6-kilobase (kb) RNA transcript. In this study we report that the nucleotide sequence of this human glycosylase cDNA* and the deduced amino acid sequence of the glycosylase are identical to those reported for the 37-kDa subunit of the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (G3PD). This finding reveals an unusual and different biochemical function ofthe monomeric form ofG3PD in normal human cells as that of a UDG that functions in the base-excision repair of DNA.

At least three types of opioid receptors have been identified in the nervous system. In this paper we report molecular cloning and expression of a rat kappa opioid receptor. PCR was performed on double-stranded cDNA derived from poly(A)+ RNA of the rat striatum with primers similar to those of Libert and co-workers [Libert, Parmentier, Lefort, Dinsart, Van Sande, Maenhaut, Simons, Dumont and Vassart (1989) Science 244, 569-572]. One of the PCR products, which had 65% sequence similarity to the mouse delta opioid receptor, was used to screen a rat striatum cDNA library. Two positive clones were isolated and found to be identical. The clone had a 2.1-kb insert, which was termed RKOR-1. RKOR-1 has an open reading frame of 1140 bp and encodes a 380-amino-acid protein. Hydropathy analysis indicates that RKOR-1 has seven putative transmembrane domains with short intra- and extra-cellular loops. Membranes of Cos-7 cells transfected with RKOR-1 exhibited high specific binding for [3H]diprenorphine ([3H]DIP), a non-selective opioid ligand. Naloxone inhibited [3H]DIP binding with stereospecificity. [3H]DIP binding was potently inhibited by selective kappa opioid ligands, with Ki values in the nanomolar or subnanomolar range, but much less potently inhibited by drugs selective for mu or delta receptors. Thus, RKOR-1 represents an opioid receptor with kappa characteristics.

Double stranded human globin cDNA was synthesized by use of viral reverse transcriptase from globin mRNA of cord blood of premature infants requiring exchange transfusions. The cDNA was introduced into plasmids and the recombinant DNA plasmids used to transform E. coli X1776. A number of transformants were obtained. Plasmid DNA from selected colonies was isolated and characterized for the type of globin cDNA it contained by three types of procedures: 1) hybridization to previously characterized 3H-labeled a,B and y cDNA; 2) analysis of the size and nature of fragments produced by digestion of the plasmid DNA by different restriction endonucleases; and 3) by rapid DNA sequence analysis of selected DNA fragments produced by restriction endonuclease digestion. Analysis by these techniques of plasmid DNA from different colonies has definitively identified the presence of human a, a or y cDNA sequences in different plasmids.

The human 3 globin gene encodes the ,3 globin chain of normal adult hemoglobin A (subunit structure a2(2). A large number of genetic disorders of the ,3 globin gene are known: hemoglobinopathies, characterized by synthesis of qualitatively abnormal ( globin chains, and the f3-thalassemias, characterized by quantitatively deficient synthesis of (3 globin chains.The /3-thalassemias are a clinically and biochemically heterogeneous group of disorders, which probably result from a wide variety of molecular defects. In 3+-thalassemia, the most common type of 03-thalassemia, /3 globin chain synthesis is decreased to approximately 5-30% ofnormal in the erythroid cells of affected individuals (1, 2). Those 3 chains that are produced appear normal by both carboxymethylcellulose column chromatography and peptide analysis (reviewed in ref.3). Moreover, there is a corresponding deficiency of P globin mRNA in (8+-thalassemic erythroid cells (4-6). This mRNA also appears normal by cDNA-RNA hybridization criteria and by its ability to direct translation ofnormal ,3 globin chains in vitro (2). These features suggest that the l3+-thalassemias may result from abnormalities of ( globin gene transcription or of processing, transport, or stability of /3 globin mRNA. Apparent abnormal processing or instability of nuclear /3 globin mRNA precursor molecules has been observed in some cases (7-10).Family studies have indicated that most /3-thalassemia mutations are allelic with, or tightly linked to, the ,B globin structural locus (1, 2), suggesting that structural analysis of thalassemic /3 globin genes might yield insight into the molecular basis of these disorders. Accordingly, we have cloned , globin gene fragments from a patient with /3+-thalassemia, and we have determined the complete nucleotide sequence of this thalassemic 3 globin gene. Comparison with the nucleotide sequence of a normal human f3 globin gene (11) reveals only a single divergent nucleotide, which occurs in the internal region of the small intervening sequence. This sequence difference suggests a possible mechanism for abnormal splicing of the thalassemic nuclear P3 globin mRNA precursor. MATERIALS AND METHODSMolecular Cloning. Total DNA was prepared from the spleen of a 12-year-old Greek Cypriot girl with typical transfusion-dependent 83+-thalassemia. The f3/a globin chain synthetic ratio obtained after incubation ofher ervthroid cells with [3H]leucine was 0. 14. The erythrocytes ofboth her parents had elevated Hb A2 levels of4.2% and 4.9%, respectively, as well as hypochromia and microcytosis, characteristic of heterozygous thalassemia; the Hb F values in the parents' erythrocytes were 1.3% and 1.7%, respectively. The spleen DNA was digested to completion with EcoRI, which cleaves the f3 globin gene at codon 122, and the fragments were electrophoresed in agarose. A strip of the gel was transferred to nitrocellulose (12), and the 5.2-kilobase (kb) 5' and 3.6-kb 3' , globin DNA fragments were identified by hybridization to 32P-labeled nick-translated (13) P globin c...

The denV gene from bacteriophage T4, which codes for endonuclease V, a small DNA repair enzyme, has been cloned and identified by an approach combining DNA sequencing and genetics, independent of the phenotypic effect of the cloned gene. Appropriate DenV+ and DenV- deletion mutants were mapped physically to define precisely a region encompassing the denV gene. This region was sequenced in order to identify a protein-coding sequence of the correct size for the denV gene (400-500 bp). Finally, identification was confirmed by sequencing the corresponding fragments cloned from four genetically and phenotypically well-characterized denV mutants. The denV gene is located at 64 kb on the T4 genome, adjacent to the ipII gene, and codes for a basic protein of 138 amino acids with a deduced molecular weight of 16,078.