Regulation of PURA Gene Transcription by Three Promoters
Total Page：16
File Type：pdf, Size：1020Kb
Wortman et al. BMC Molecular Biology 2010, 11:81 http://www.biomedcentral.com/1471-2199/11/81 RESEARCH ARTICLE Open Access Regulation of PURA gene transcription by three promoters generating distinctly spliced 5-prime leaders: a novel means of fine control over tissue specificity and viral signals Margaret J Wortman1*, Laura K Hanson1,2, Luis Martínez-Sobrido3, Ann E Campbell1, Jonas A Nance1, Adolfo García-Sastre4,5,6, Edward M Johnson1* Abstract Background: Pura is an evolutionarily conserved cellular protein participating in processes of DNA replication, transcription, and RNA transport; all involving binding to nucleic acids and altering conformation and physical positioning. The distinct but related roles of Pura suggest a need for expression regulated differently depending on intracellular and external signals. Results: Here we report that human PURA (hPURA) transcription is regulated from three distinct and widely- separated transcription start sites (TSS). Each of these TSS is strongly homologous to a similar site in mouse chromosomal DNA. Transcripts from TSS I and II are characterized by the presence of large and overlapping 5’-UTR introns terminated at the same splice receptor site. Transfection of lung carcinoma cells with wild-type or mutated hPURA 5’ upstream sequences identifies different regulatory elements. TSS III, located within 80 bp of the translational start codon, is upregulated by E2F1, CAAT and NF-Y binding elements. Transcription at TSS II is downregulated through the presence of adjacent consensus binding elements for interferon regulatory factors (IRFs). Chromatin immunoprecipitation reveals that IRF-3 protein binds hPURA promoter sequences at TSS II in vivo. By co-transfecting hPURA reporter plasmids with expression plasmids for IRF proteins we demonstrate that several IRFs, including IRF-3, down-regulate PURA transcription. Infection of NIH 3T3 cells with mouse cytomegalovirus results in a rapid decrease in levels of mPURA mRNA and Pura protein. The viral infection alters the degree of splicing of the 5’-UTR introns of TSS II transcripts. Conclusions: Results provide evidence for a novel mechanism of transcriptional control by multiple promoters used differently in various tissues and cells. Viral infection alters not only the use of PURA promoters but also the generation of different non-coding RNAs from 5’-UTRs of the resulting transcripts. Background zone of DNA replication upstream of the c-MYC gene The Pur protein family of sequence-specific single- [2-4]. It was quickly recognized that Pura has a high affi- stranded nucleic acid-binding proteins is extremely well nity for single-stranded oligonucleotides, DNA or RNA, conserved from bacteria through humans [1]. Human and with a purine-rich repeat, (GGN)n [3]. Pura binds the mouse Pura differ by only 2 amino acids. Human family G-rich strand of double-stranded DNA [3] with the ability member, Pura, has a primitive codon usage preference to locally separate the bound strands [5-7]. Pura’s role in resembling that in bacteria. Pura was first identified due DNA replication has been supported in several labora- to its ability to bind a purine-rich sequence in an initiation tories (8-12). Shimotai [8] determined that Pura binds an element within an autonomously replicating sequence * Correspondence: [email protected]; [email protected] 1 (ARS) in the rat aldolase B promoter that is essential for Department of Microbiology and Molecular Cell Biology, Eastern Virginia a Medical School, 700 W. Olney Road, Norfolk, VA 23507, USA replication. Liu et al. found that Pur and Cyclin A/Cdk2, Full list of author information is available at the end of the article two proteins essential for the initiation of replication, are © 2010 Wortman et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Wortman et al. BMC Molecular Biology 2010, 11:81 Page 2 of 15 http://www.biomedcentral.com/1471-2199/11/81 simultaneously bound to the replication origin near the c- Type Culture Collection (ATCC, Manassas, VA) and MYC gene [9]. Pura has also been associated with viral were maintained in RPMI medium supplemented with DNA replication. Pura binds the JC viral (JCV) origin of L-glutamine and antibiotics as described by ATCC. NIH replication and at higher concentrations, inhibits JCV 3T3, a mouse embryonic fibroblast derived cell line was DNA replication [10]. At lower concentrations Pura coop- maintained in culture and infected with mouse cytome- eratively interacts with HIV-1 Tat and large T-antigen to galovirus (MCMV) as previously described [20]. enhance JCV DNA replication [11,12]. Several papers indicate that Pura is multifunctional in RNA Analysis cell cycle regulation. During early G1, Pura associates Total RNA from normal lung tissue and lung adenocar- with hypophosphorylated retinoblastoma protein, Rb, cinoma (prepared by Clontech, Mountain View, CA) from which it is released when Rb becomes phosphory- was used to analyze PURA 5’-UTR RNA. C-DNAs were lated in mid to late G1 phase [13]. Pura functions in synthesized using the ImProm-II Reverse Transcription maintaining the integrity of replicating DNA [14]. Pura System (Promega, Madison, WI) and a primer located remains bound to Cyclin A/Cdk2 at the G2/M boundary 115 nt upstream of the translational start codon. [9,15], a checkpoint for replication fidelity. The Cyclin A/ Sequence closer to the translational start codon was Cdk2 complex is important in the repair of double-strand avoided because of its high G:C content. PURA contain- breaks [16]. Wang et al. [14] found that mouse embryo ing transcripts were amplified with Taq DNA polymer- fibroblasts (MEFs) depleted of Pura, have an increased ase as described by Promega in a Hybaid OmniGene incidence of double-strand breaks when grown in the Temperature Cycler. The initial PCR products were pur- presence of the DNA replication inhibitor hydroxyurea ified on Qiaquick PCR purification columns (Qiagen, (HU). Deletions or translocations of PURA correlate with Valencia, CA) and the DNA re-amplified with nested the occurrence of myelodysplastic syndrome and its tran- primers in a reaction mixture made according to the sition to acute myelogenous leukemia [17]. directions of the Expand HiFidelity PCR System (Roche Transcription of a given gene can be initiated at more Applied Science, Indianapolis, IN) but with the additions than one site. Analyses of full-length human cDNAs from of 5% DMSO and 1.5 μM betaine. Primers prepared by a large number of cDNA databases has revealed that Invitrogen, (Carlsbad, CA) are described in the text. human genes, especially those encoding certain proteins Positions and sequences are listed in Table 1. expressed in the brain, have multiple putative alternative The MTC™ Panel I (Clontech, Mountain View, CA) of promoters (PAPs) [18]. On average there are 3.1 PAPs per first strand cDNAs prepared from poly A+ RNA isolated gene with frequent variation in splicing of the first exon. It from eight human tissues was used to quantitate PURA is known that Pura is highly expressed in the brain [19] transcripts. Real time (RT)-PCR was performed using and that in the mouse it has one intron in its 5’ untrans- the ICycler (BioRad, Hercules, CA) with primers lated region (5’-UTR). Tissue specificity in the usage of described in the text (also see Table 1) and processed PAPs was observed [18]. Multiple PAPs allow transcription using ICycler software. RT-PCR reaction products were to be regulated by different sets of regulatory factors. In routinely analyzed by electrophoresis in 0.8% agarose the cases thus far studied, different promoters have gener- gels in 40 mM Tris-Acetate, 2 mM disodium EDTA ated different coding sequences. We hypothesize that Pura (TAE), pH 8.5, and stained with SYBR Gold nucleic acid expression is controlled differently depending on various gel stain (Molecular Probes, Invitrogen). When reactions states of cell environment, including growth-altering sig- resulted in PCR products of < 200 bp, they were also nals or viral infection and that such control is mediated analyzed in 8% polyacrylamide gels in the same buffer. effectively by independent modes of transcriptional promo- tion. Here we identify three functionally distinct PURA mPura expression in murine cytomegalovirus promoters and show that these are utilized differentially in (MCMV)-infected cells human cell types and that they respond differently to cyto- NIH 3T3 fibroblasts were propagated and infected with megalovirus infection. While all three promoters generate wild-type Smith strain MCMV as previously described the same protein coding sequence, multiple levels of con- [20,21]. For the current experiments, cells were infected trol are exerted over the sequences of transcripts produced with a multiplicity of 2 PFU/cell, and infected cell and over cellular levels of Pura protein. lysates were harvested at 1, 3, 5, and 9 hr post infection. TotalcellproteinandRNAwas isolated using TRI Methods Reagent (Molecular Research Center, Inc., Cincinnati, Cell Culture Ohio) according to manufacturer’s instructions. RNA NCI-H82 and NCI-H146 small-cell lung carcinoma was further processed using RNAeasy columns (Qiagen) (SCLC) cell lines were obtained from the American with DNase I digestion. Mock-infected NIH 3T3 cells Wortman et al.