Upstream Flanking DNA Research Articles

The ribosomal protein L34 gene from the mosquito, Aedes albopictus: exon–intron organization, copy number, and potential regulatory elements

We describe the structural analysis of genomic DNA encoding ribosomal protein (rp) L34 from the mosquito, Aedes albopictus. Comparison of genomic DNA sequences encompassing approximately 8 kb with the rpL34 cDNA sequence showed that the gene contains three exons and two introns, encoding a primary transcript with a deduced size of 6196 nucleotides from the transcription start site to the polyadenylation site. Exon 1, which is not translated, measures only 45 bp, and is separated from Exon 2 by a 359 bp intron. Exon 2 measures 78 bp, and contains the AUG translation initiation codon 14 nucleotides downstream of its 5′-end. Downstream of Exon 2 is a 5270 bp intron, followed by the remainder of the coding sequence in Exon 3, which measures 444 bp including the polyadenylation signal. We used a novel PCR-based procedure to obtain 1.7 kb of DNA upstream of the rpL34 gene. Like the previously described Ae. albopictus rpL8 gene and various mammalian rp genes, the DNA immediately upstream of the rpL34 gene lacks the TATA box, and the rpL34 transcription initiation site is embedded in a characteristic polypyrimidine tract. The 5′-flanking DNA contained a number of cis-acting elements that potentially interact with transcription factors characterized by basic domains, zinc-coordinating DNA binding domains, helix-turn-helix motifs, and beta scaffold factors with minor groove contacts. Particularly striking was the conservation of an AP-4 binding site within 100 nucleotides upstream of the transcription initiation site in both Aal-rpL34 and Aal-rpL8 genes. Comparison of Southern hybridization signals using probes from the 5′ and 3′-ends of the 5.3 kb second intron and the cDNA suggested that the Ae. albopictus rpL34 gene most likely occurs as a single expressed copy per haploid genome with restriction enzyme polymorphisms in the upstream flanking DNA and the likely presence of one or more pseudogenes.

Detection of 28S RNA with the FGF-2 cDNA at high stringency through related G/C-rich sequences.

Fibroblast growth factor-2 (FGF-2) or basic FGF is a multifunctional protein that, through interaction with specific cell surface receptors, plays important roles in the growth and development of tissues and organs. Thus, considerable attention has focused on the control of FGF-2 gene expression, including assessments of RNA levels through blotting and the use of radiolabeled FGF-2 cDNA probes. Multiple transcripts of different sizes have been reported for FGF-2 by this approach, however, more recent evidence indicates that at least one of these RNAs of about 1.5 kb, is not an authentic FGF-2 transcript. A major band of 4.7 kb and a minor band of 6.1 kb were detected in total rat glial tumor cell RNA, using the 'intact' rat ovarian FGF-2 cDNA as a probe at high stringency. This cDNA contains both coding and 5'-untranslated sequences. Although the 6.1 kb transcript levels were increased in RNA enriched for polyadenylated species, the levels of the 4.7 kb band were decreased and also shared a mobility with 28S RNA. A truncated FGF-2 cDNA probe, containing coding but not 5'-untranslated sequences, detected the 6.1 kb transcript but failed to see the 4.7 kb band. The domain responsible for detecting the 4.7 kb band was localized to a G/C-rich region containing 5'-untranslated sequences, by using different fragments of the rat FGF-2 gene, including coding and upstream flanking DNA, as probes. The degree of similarity between sequences of this G/C-rich region of the FGF-2 gene and 28S RNA from rat, human and mouse was sufficient to predict strong cross hybridization. This was confirmed by the detection of a 4.7 kb band in mouse heart RNA with the 'intact' but not truncated rat FGF-2 cDNA probes; a 6.1 kb mouse FGF-2 transcript was detected with both probes. These data indicate that the 4.7 kb RNA detected is not a bona fide FGF-2 transcript, and most likely represents cross hybridization with abundant 28S RNA through G/C-rich non-coding sequences present in the 'intact' rat FGF-2 cDNA. However, sequence comparisons suggest that this result may be the case for other species and might not be restricted to the rat FGF-2 cDNA.

Nuclear factor binding to a DNA sequence element that represses MMTV transcription induces a structural transition and leads to the contact of single-stranded binding proteins with DNA.

NRE1 is a DNA sequence element in the long terminal repeat of mouse mammary tumor virus through which viral transcription is repressed. In addition to double-stranded DNA binding, both upper- and lower-stranded NRE1 binding activities occur in nuclear extracts. All three binding activities appear to be important for transcriptional effects. We report that occupancy of NRE1 within linear double-stranded NRE1 induces a structural transition in upstream flanking DNA that is facilitated by Mg2+. This transition was reflected by the striking DNase I sensitivity of the DNA. As Mg2+ concentration was increased, discrete DNase I hypersensitivity on one face of the DNA progressed to complete degradation of template. On the DNA face opposite the DNase I hypersensitivity, Mg2+ promoted regularly spaced cleavage by the single-strand-specific cleavage agents KMnO4 and S1 nuclease. Induction of degradation by DNase I occurred independently of MMTV sequences flanking NRE1, because nuclear extract-dependent DNase I sensitivity was conferred to an unrelated DNA fragment by introduction of a 23-bp NRE1-containing oligonucleotide. UV protein-DNA cross-linking revealed that addition of Mg2+ to a double-stranded NRE1 DNA binding assay induced conversion from a double- to a single-stranded protein-DNA cross-linking pattern. Thus, nuclear factor binding to NRE1 induces changes in DNA topology that promote the direct contact of single-stranded NRE1 binding factors with DNA.

Cloning and analysis of duplicated rfbM and rfbK genes involved in the formation of GDP-mannose in Escherichia coli O9:K30 and participation of rfb genes in the synthesis of the group I K30 capsular polysaccharide.

The rfbO9 gene cluster, which is responsible for the synthesis of the lipopolysaccharide O9 antigen, was cloned from Escherichia coli O9:K30. The gnd gene, encoding 6-phosphogluconate dehydrogenase, was identified adjacent to the rfbO9 cluster, and by DNA sequence analysis the gene order gnd-rfbM-rfbK was established. This order differs from that described for other members of the family Enterobacteriaceae. Nucleotide sequence analysis was used to identify the rfbK and rfbM genes, encoding phosphomannomutase and GDP-mannose pyrophosphorylase, respectively. In members of the family Enterobacteriaceae, these enzymes act sequentially to form GDP-mannose, which serves as the activated sugar nucleotide precursor for mannose residues in cell surface polysaccharides. In the E. coli O9:K30 strain, a duplicated rfbM2-rfbK2 region was detected approximately 3 kbp downstream of rfbM1-rfbK1 and adjacent to the remaining genes of the rfbO9 cluster. The rfbM isogenes differed in upstream flanking DNA but were otherwise highly conserved. In contrast, the rfbK isogenes differed in downstream flanking DNA and in 3'-terminal regions, resulting in slight differences in the sizes of the predicted RfbK proteins. RfbMO9 and RfbKO9 are most closely related to CpsB and CpsG, respectively. These are isozymes of GDP-mannose pyrophosphorylase and phosphomannomutase, respectively, which are thought to be involved in the biosynthesis of the slime polysaccharide colanic acid in E. coli K-12 and Salmonella enterica serovar Typhimurium. An E. coli O-:K30 mutant, strain CWG44, lacks rfbM2-rfbK2 and has adjacent essential rfbO9 sequences deleted. The remaining chromosomal genes are therefore sufficient for GDP-mannose formation and K30 capsular polysaccharide synthesis. A mutant of E. coli CWG44, strain CWG152, was found to lack GDP-mannose pyrophosphorylase and lost the ability to synthesize K30 capsular polysaccharide. Wild-type capsular polysaccharide could be restored in CWG152, by transformation with plasmids containing either rfbM1 or rfbM2. Introduction of a complete rfbO9 gene cluster into CWG152 restored synthesis of both O9 and K30 polysaccharides. Consequently, rfbM is sufficient for the biosynthesis of GDP-mannose for both O antigen and capsular polysaccharide E. coli O9:K30. Analysis of a collection of serotype O8 and O9 isolates by Southern hybridization and PCR amplification experiments demonstrated extensive polymorphism in the rfbM-rfbK region.

Organization and sequence of the HpaII restriction-modification system and adjacent genes

We report the organization of the HpaII restriction and modification (R-M) system from Haemophilus parainfluenzae (recognition sequence: 5'…CCGG…3'), the sequence of the gene coding for the HpaII restriction endonuclease, and the sequence of the upstream flanking DNA. The HpaII system comprises two genes, hpaIIM, coding for the methyltransferase (MTase; 358 amino acids (aa), 40.4 kDa: product, Cm 5CGG), and hpaIIR, coding for the restriction endonuclease (ENase; 358 aa, 40.9 kDa: product, C'CGG). The genes are adjacent, they have the same orientation, and they occur in the order hpaIIM then hpaIIR. The ENase bears little aa sequence similarity to the isoschizomeric R- BsuFI and R · MspI ENases. Upstream of, and partly overlapping hpaIIM is the coding sequence for a 141-aa protein that resembles the very-short-patch-repair endonuclease (Vsr) of Escherichia coli. Upstream ofthat is the coding sequence for a protein that resembles valyl-tRNA synthetase (ValS).

Deletion analysis and localization of SbPRP1, a soybean cell wall protein gene, in roots of transgenic tobacco and cowpea1

SbPRP1 is a member of the soybean (Glycine max L. Merr) proline-rich cell wall protein family and is expressed at high levels in root tissue. To characterize the sequences required for this expression, we have fused 1.1 kb of upstream flanking DNA sequence from an SbPRP1 genomic clone to a gene encoding beta-glucuronidase (GUS). This construct was introduced into tobacco using Agrobacterium tumefaciens-mediated transformation. Histochemical staining of GUS activity in transgenic tobacco indicated that SbPRP1 is expressed in the apical and elongating region of both primary and lateral roots, most strongly in the epidermis. A similar localization pattern was found in transformed hairy roots when this construct was introduced into cowpea (Vigna aconitifolia) using Agrobacterium rhizogenes-mediated transformation. Nested 5'-deletion analysis of the SbPRP1 promoter indicated that a minimal promoter for SbPRP1 expression in roots is located within the first 262 bases of upstream flanking DNA and that the region between -1080 and -262 is required for maximal expression of this gene. Gel retardation assays showed that nuclear factors can be detected in soybean roots which specifically bind to sequences located between -1080 and -623, a region which is needed for maximal expression of the SbPRP1 promoter. Northern hybridization analysis was also used to show that little SbPRP1 mRNA was present in roots during the first 24 h after imbibition. These studies indicate that SbPRP1 expression is localized to the actively growing region of the root and that this expression is temporally regulated during very early stages of seedling growth.

Proximal Upstream Flanking Sequences Direct Calcium Regulation of the Rat Prolactin Gene

Previous investigations have shown that Ca2+ strongly and specifically stimulates endogenous PRL gene expression by GH3 cells. In this study, addition of Ca2+ to Ca2+-deprived GH3 cells yielded a large (ca. 8-fold) stimulation of transient expression of a transfected PRL-chloramphenical acetyltransferase (CAT) construct containing ca. 1 kilo-base-pair of the PRL promoter region, but only a slight (less than or equal to 2-fold) nonspecific stimulation of CAT activity directed by any of three control promoters: dihydrofolate reductase, Rous sarcoma virus, or thymidine kinase. In GH3 cells never deprived of Ca2+, expression of a PRL-CAT construct was specifically stimulated and inhibited, respectively, by the dihydropyridine voltage-dependent Ca2+ channel modulators Bay K8644 and nimodipine; Ca2+ can thus regulate expression of an exogenous PRL promoter in cells incubated under physiological Ca2+ conditions. By employing a combined protocol, in which Ca2+-deprived cells are exposed to Ca2+ in the presence of Bay K8644, a very large (greater than 35-fold) but still promoter-specific induction of expression of a PRL-CAT construct was obtained. Analysis of 5'-deleted PRL-CAT constructs implied that the PRL gene Ca2+ response element is contained entirely within the first 174 base pairs of upstream flanking DNA sequence.

Juvenile hormone-regulated locust vitellogenin genes: Lack of expression after transfer intoDrosophila

In Locusta migratoria, vitellogenin (Vg) is normally produced only in adult female fat body under stimulation by juvenile hormone (JH). This permits study of (a) programming of genes for expression and (b) modulation of expression by JH. From L. migratoria genomic libraries, two Vg genes (A and B) have been cloned. Coding regions encompass 10-12 kbases of DNA, contain introns, and hybridize with 6,300 nucleotide mRNAs. Although the 5'-exons, coding for signal peptides, are similar in sequence, the remaining coding sequences have distinct restriction maps and show no crosshybridization. Upstream DNA from the two genes has some sequence similarity, corresponding to potential regulatory regions. Dot hybridization assays of mRNAs A and B in locust fat body after induction by methoprene show coordinate expression of the two Vg genes and also a third, unidentified JH-regulated gene. In the hope of providing a system for identification of control sequences, P element-mediated transformation has been used to transfer locust DNA into Drosophila. From Vg gene B, a central block was deleted, to give a mini-Vg gene comprising 5' and 3' terminal coding sequences plus upstream flanking DNA. This was incorporated into the P element vector Carnegie 20 and injected into Drosophila embryos. Three transformed fly lines were obtained in which the locust mini-Vg gene was integrated at different chromosomal sites. On Northern blots of fly RNA, however, no expression of the locust sequences could be detected, even though the accompanying rosy gene was expressed. This suggests that the locust regulatory sequences were not recognized in Drosophila, and current effort is directed toward developing a homologous locust transformation system for expression assay of cloned JH-regulated genes.

Mapping of DNase I-hypersensitive sites in the upstream DNA of human embryonic epsilon-globin gene in K562 leukemia cells.

We have mapped the DNase I-hypersensitive sites around the epsilon-globin and c-myc genes in two human leukemia cell lines K562 and HL60. In K562 cells in which the epsilon-globin gene is transcribed, six DNase I-hypersensitive sites are found in 6 kilobases (kb) of upstream flanking DNA; in HL60 cells in which the c-myc gene is expressed, two DNase I-hypersensitive sites are observed in 2 kb of upstream DNA. Neither the inactive epsilon-globin gene in HL60 cells nor the inactive c-myc gene in K562 cells displays such upstream DNase I-hypersensitive sites. Our results are consistent with previous studies that have shown DNase I-hypersensitive sites within 1 kb of the 5' end of other expressed genes. In addition, we have found sites displaying even more DNase I sensitivity further upstream of expressed epsilon-globin and c-myc genes. Among the six DNase I-hypersensitive sites of the expressed epsilon-globin gene in K562 cells, the most sensitive site is located about 6 kb upstream of the epsilon-globin gene. When correlated with the DNA sequence upstream of the epsilon-globin gene, this site was found to correspond to a region that contains a stretch of 28 consecutive Ts, three enhancer core-like sequences, and a stretch of consecutive (C-A)15(T-A)6 alternating purine and pyrimidine bases. These findings suggest the possibility that an enhancer element for epsilon-globin gene expression resides within this DNase I-hypersensitive site.