Translational Control of Specific Genes during Differentiation of HL-60 Cells
1999; Elsevier BV; Volume: 274; Issue: 20 Linguagem: Inglês
10.1074/jbc.274.20.14295
ISSN1083-351X
AutoresAnna M. Krichevsky, Esther Metzer, Haim Rosen,
Tópico(s)RNA modifications and cancer
ResumoEukaryotic gene expression can be regulated through selective translation of specific mRNA species. Nevertheless, the limited number of known examples hampers the identification of common mechanisms that regulate translation of specific groups of genes in mammalian cells. We developed a method to identify translationally regulated genes. This method was used to examine the regulation of protein synthesis in HL-60 cells undergoing monocytic differentiation. A partial screening of cellular mRNAs identified five mRNAs whose translation was specifically inhibited and five others that were activated as was indicated by their mobilization onto polysomes. The specifically inhibited mRNAs encoded ribosomal proteins, identified as members of the 5′-terminal oligopyrimidine tract mRNA family. Most of the activated transcripts represented uncharacterized genes. The most actively mobilized transcript (termed TA-40) was an untranslated 1.3-kilobase polyadenylated RNA with unusual structural features, including two Alu-like elements. Following differentiation, a significant change in the cytoplasmic distribution of Alu-containing mRNAs was observed, namely, the enhancement of Alu-containing mRNAs in the polysomes. Our findings support the notion that protein synthesis is regulated during differentiation of HL-60 cells by both global and gene-specific mechanisms and that Alu-like sequences within cytoplasmic mRNAs are involved in such specific regulation. Eukaryotic gene expression can be regulated through selective translation of specific mRNA species. Nevertheless, the limited number of known examples hampers the identification of common mechanisms that regulate translation of specific groups of genes in mammalian cells. We developed a method to identify translationally regulated genes. This method was used to examine the regulation of protein synthesis in HL-60 cells undergoing monocytic differentiation. A partial screening of cellular mRNAs identified five mRNAs whose translation was specifically inhibited and five others that were activated as was indicated by their mobilization onto polysomes. The specifically inhibited mRNAs encoded ribosomal proteins, identified as members of the 5′-terminal oligopyrimidine tract mRNA family. Most of the activated transcripts represented uncharacterized genes. The most actively mobilized transcript (termed TA-40) was an untranslated 1.3-kilobase polyadenylated RNA with unusual structural features, including two Alu-like elements. Following differentiation, a significant change in the cytoplasmic distribution of Alu-containing mRNAs was observed, namely, the enhancement of Alu-containing mRNAs in the polysomes. Our findings support the notion that protein synthesis is regulated during differentiation of HL-60 cells by both global and gene-specific mechanisms and that Alu-like sequences within cytoplasmic mRNAs are involved in such specific regulation. Translational control of eukaryotic gene expression can be broadly classified into two major categories: global control affecting the overall rate of protein synthesis, and selective control in which the translation rate of mRNA subsets varies in response to biological stimuli (1Hershey J.W.B. Annu. Rev. Biochem. 1991; 60: 717-755Crossref PubMed Scopus (845) Google Scholar, 2Merrick W.C. Microbiol. Rev. 1992; 56: 291-315Crossref PubMed Google Scholar, 3Mathews M.B. Sonenberg N. Hershey J.W.B. Hershey J.W.B. Mathews M.B. Sonenberg N. Translational Control. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996: 1-29Google Scholar). While much data has been accumulated in the last two decades regarding global control of protein synthesis (for review, see Refs. 1Hershey J.W.B. Annu. Rev. Biochem. 1991; 60: 717-755Crossref PubMed Scopus (845) Google Scholar and 2Merrick W.C. Microbiol. Rev. 1992; 56: 291-315Crossref PubMed Google Scholar), little was published on selective control systems until recently (4Bernstein J. Shefler I. Elroy-Stein O. J. Biol. Chem. 1995; 270: 10559-10565Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar, 5Fu L. Benchimol S. EMBO J. 1997; 16: 4117-4125Crossref PubMed Scopus (108) Google Scholar, 6Hengst L. Reed S.I. Science. 1996; 271: 1861-1864Crossref PubMed Scopus (823) Google Scholar, 7Imataka H. Nakayama K. Yasumoto K. Mizuno A. Fujii-Kuriyama Y. Hayami M. J. Biol. Chem. 1994; 269: 20668-20673Abstract Full Text PDF PubMed Google Scholar, 8Ostareck D.H. Ostareck-Lederer A. Wilm M. Thiele B.J. Mann M. Hentze M.W. Cell. 1997; 89: 597-606Abstract Full Text Full Text PDF PubMed Scopus (430) Google Scholar, 9Rousseau D Kaspar R. Rosenwald I. Gehrke L. Sonenberg N. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 1065-1070Crossref PubMed Scopus (362) Google Scholar, 10Vagner S. Gensac M.C. Maret A. Bayard F. Amalric F. Prats H. Prats A.C. Mol. Cell. Biol. 1995; 15: 35-44Crossref PubMed Scopus (289) Google Scholar). Regulation of eukaryotic gene expression during developmental processes, such as germ cell maturation and early embryonic development in Xenopus, Caenorhabditis elegans, andDrosophila, was demonstrated to involve selective translation of certain mRNA species (see Ref. 11Wickers M. Kimble J. Strickland S. Hershey J.W.B. Mathews M.B. Sonenberg N. Translational Control. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996: 433-450Google Scholar). In mammalian cells, however, the translation of only few mRNAs was shown to be regulated by gene-specific mechanisms during differentiation. The most characterized example of this type of regulation is the translation of 15-lipoxygenase mRNA during red blood cell differentiation (8Ostareck D.H. Ostareck-Lederer A. Wilm M. Thiele B.J. Mann M. Hentze M.W. Cell. 1997; 89: 597-606Abstract Full Text Full Text PDF PubMed Scopus (430) Google Scholar). Previously, two approaches were used to identify and isolate specific mRNAs translationally regulated during differentiation. One approach was based on the pulse-chase labeling of cells with radioactive amino acids, followed by two-dimensional gel electrophoresis to identify relative changes in the translation pattern. Candidate proteins were further characterized using various biochemical methods, and their corresponding cDNAs were cloned and used to quantify their respective mRNA levels (12Jefferies H.B.J. Thomas G. Thomas G. J. Biol. Chem. 1994; 269: 4367-4372Abstract Full Text PDF PubMed Google Scholar, 13Thomas G. Thomas G. J. Cell Biol. 1986; 103: 2137-2144Crossref PubMed Scopus (89) Google Scholar). The second approach was based on the assumption that translationally inactive mRNAs are present as free cytoplasmic mRNPs, whereas actively translated mRNAs are contained within polysomes. To isolate and identify mRNAs selectively mobilized from mRNPs 1The abbreviations used are: RNP, ribonucleoprotein; 5′-TOP, 5′-terminal oligopyrimidine tract; RT-PCR, reverse transcriptase-polymerase chain reactio; DDRT-PCR, differential display RT-PCR; TPA, 12-O-tetradecanoyl-1-phorbol-13-acetate; BMV, Brome mosaic virus to polysomes, cDNA probes were prepared from mRNPs and polysomes and hybridized against cDNA libraries (14Loreni F. Francesconi A. Jappelli R. Amaldi F. Nucleic Acids Res. 1992; 20: 1859-1863Crossref PubMed Scopus (11) Google Scholar, 15Yenofsky R. Bergmann I. Brawerman G. Biochemistry. 1982; 21: 3909-3913Crossref PubMed Scopus (10) Google Scholar). These studies, although limited to the relatively abundant mRNAs, led to the identification of a group of mRNAs encoding ribosomal proteins, elongation factors for protein synthesis, and proteins of as yet unknown function (14Loreni F. Francesconi A. Jappelli R. Amaldi F. Nucleic Acids Res. 1992; 20: 1859-1863Crossref PubMed Scopus (11) Google Scholar). More recent studies have established that translational efficiency of these mRNAs correlates with the rate of cell proliferation (16Avni D. Shama S. Loreni F. Meyuhas O. Mol. Cell. Biol. 1994; 14: 3822-3833Crossref PubMed Scopus (135) Google Scholar). The cis element in this type of translationally controlled mRNAs was found to be a 5′-terminal oligopyrimidine tract (5′-TOP), and they were termed 5′-TOP mRNAs (16Avni D. Shama S. Loreni F. Meyuhas O. Mol. Cell. Biol. 1994; 14: 3822-3833Crossref PubMed Scopus (135) Google Scholar). In this study a new protocol based on the separation of polysomes from mRNPs using sucrose gradient centrifugation followed by differential display RT-PCR analysis is presented. This methodology identifies new mRNAs specifically mobilized from cytoplasmic free mRNPs onto polysomes and vice versa in the promyelocytic leukemia cell line HL-60 during cellular differentiation. Our findings suggest that Alu-like elements within cytoplasmic mRNAs are involved in gene-specific translation regulation. Sera, cell culture medium, and antibiotics were provided by Biological Industries (Beit-Haemek, Israel); standard chemicals, phorbol ester 12-O-tetradecanoyl-1-phorbol-13-acetate (TPA), dimethyl sulfoxide were purchased from Sigma. 1,25-Dehydroxyvitamin D3 was a kind gift from Zvi Bar-Shavit (Faculty of Medicine, HU, Jerusalem, Israel). Restriction endonucleases and Random Priming Reagent kit were provided by New England BioLabs (Beverly, MA); AMV-Reverse Transcriptase, Taq DNA-polymerase, RiboMAXTM Large Scale RNA Production System, wheat germ extract, and rabbit reticulocyte nuclease-treated lysates were obtained from Promega (Madison, WI). The rapid amplification of 5′ cDNA ends kit was obtained from Roche Molecular Biochemicals (Mannheim, Germany). Nylon membranes and radiochemicals were purchased from NEN Life Science Products Inc. (Boston, MA). The human cell lines K562 and HL-60 were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum,l-glutamine, penicillin, and streptomycin. Cells were routinely passaged twice a week to a density of 0.5–1.0 × 106 cells/ml and grown at 37 °C in 5% CO2. Induction of differentiation (30 ml of 4 × 105cells/ml for 140-mm plate) was performed by the addition of 50 nm TPA, 100 nm 1,25-dehydroxyvitamin D3, or 1.25% dimethyl sulfoxide. Cell harvesting and sucrose gradient separation between polysomal and subpolysomal fractions were performed as described previously (17Meyuhas O. Thompson J.R. Perry R.P. Mol. Cell. Biol. 1987; 7: 2691-2699Crossref PubMed Google Scholar). Briefly, cells were collected by centrifugation, washed with ice-cold phosphate-buffered saline, lysed at 4 °C, and the postmitochondrial supernatant was overlaid onto a 15–45% sucrose gradient and spun at 26,000 rpm for 4 h at 4 °C. Fractions of polysomes (up to disomes) and subpolysomes (from 80 S to RNPs) were collected from the bottom of each gradient with continuous monitoring at 260 nm and precipitated with 0.1 mNaCl and 2.5 volumes of ethanol, the pellets were dissolved in guanidine thiocyanate and further purified by centrifugation through CsCl (18Chirgwin J.M. Przybyla A.E. MacDonald R.J. Rutter W.J. Biochemistry. 1979; 18: 5294-5299Crossref PubMed Scopus (16652) Google Scholar). Purified RNA preparations were quantified by spectrophotometry. For Northern blot analysis, RNA was denatured in glyoxal and subjected to electrophoresis on 1.5% agarose gel in 10 mm sodium phosphate buffer. The RNA was transferred to a nylon-based membrane, hybridized with the indicated32P-random primed labeled probe, washed, and exposed to x-ray film as described (19Rosen H. Polakiewicz R. Brain Res. Dev. 1989; 46: 123-129Crossref PubMed Scopus (26) Google Scholar). The DNA probes were specific cDNAs, prepared by PCR and labeled by random priming. For RT-PCR analysis first-strand cDNA was synthesized with avian myeloblastosis virus-RT in the presence of 1 μg of total RNA, 1.5 μm oligo(dT) primer, 0.2 mm of each dNTP, 50 mm Tris, pH 8.3, 40 mm KCl, 6 mmMgCl2, 10 units of RNase inhibitor, and 10 units of avian myeloblastosis virus-RT in a final volume of 10 μl. Reaction was carried out for 45 min at 42 °C followed by heat inactivation for 5 min in 75 °C. For each PCR reaction we used 2 μl of the cDNA reaction mixture, 0.2 mm of each dNTP, 2 units ofTaq polymerase, 1 times the corresponding reaction buffer provided with the enzyme, and a pair of PCR-specific primers (50 pmol of each). The amplification was performed for 15–40 cycles (each cycle consisted of 1 min at 93 °C, 1 min at 58 °C, and 1 min at 72 °C). The PCR products were then separated on 2.5% NuSieve 3:1 agarose gel. The following gene-specific primers were used (the sequences are written from 5′ to 3′): GGAACATGCTGAGAAACTG and CAGGGTGTGCTTGTCAAAG for H-ferritin; TCACCAACTGGGACGACATG and GTACAGGGATAGCACAGCCT for b-actin; ATCCCCAGAAGCAGCATGAC, TGCAGAGTGCCGAGAAATCC and CCCATTTCATGCCGTCCTG for TA 90 (fibulin 1D); CTCACACACACAACCATCC and GGAGTCTGGCTGTATTGTCC for TA 40; GGGAAAGCTTTCTAGTCTA and GTACTTAGGATAGCAGTAC for TA 10; GCTTGGGCCTTCCTCCATC and CTCTGTACACCCAGGAGTC for TA 12; CTGTTCCTAATGAGCAGGGC and TCAAAACTCACACGGCAGC for TA 20; GTTAGACCCCAATGAGACC and CACATTCCCCTTCACCTTC for TR (TR 11); GTCTCAAGGTGTTTGACGG and AGGAGTCCGTGGGTCTTGA for TR 10; ATCTCCTTCATCCCTCTCC and GCTTTTAGTGCTGCTTCCTC for TR 13; AGACCCTCACTGGCAAAAC and TGACCTTCTTCTTGGGACG for TR 40; CCGCAAACTCTGTCTCAAC and GGCCTCCTCTTTGCTGATT for TR 80; GTAACTCCACCAAGCCCATC and CCCTCTTTTCTTTTCCTCCC for VDR; TGCAGGTGGCAGAGTGAATG and CAAGAGATTGGGGGGTGAAG for PAI-1; GATTATGTCCGGCCACGTTC and ACCCAGCCCCACAAAAAAAG for acyl-CoA; TCTCTCCTCCCTTTCTTCCC and TCCTTCCTCTGCTTCTCACC for Il-6R; CTGCGGAGATCACACTGAC and GCTCTTCCTCCTACACATC for HLA; GAGTCTGCTGAAGCTATCC and AGTCTACACCACAACCACC for RepA. For the semiquantitative RT-PCR the number of amplification cycles was limited in order to maintain the PCR reaction within the linear range. The optimal number of cycles for each pair of primers was determined by withdrawn aliquots of the corresponding reactions for agarose gel analysis after various number of cycles. The number of cycles sufficient for detecting clearly visible bands was chosen. The relative intensity of the DNA bands remained the same for at least four additional cycles. Differential display of 0.5 μg of total RNA from polysomes and from subpolysomes before and after cell treatment was performed essentially as described by Liang and Pardee (20Liang P. Pardee A.B. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, Inc., New York1995: 15.8.1-15.8.8Google Scholar, 21Liang P. Pardee A.B. Science. 1992; 257: 967-971Crossref PubMed Scopus (4707) Google Scholar). For reverse transcription, 13 units/μl of Super Script II RT (Life Technologies, Inc., Bethesda, MD) were used. PCR reactions were performed with combinations of 10 different arbitrary 10-mer primers (50% GC, sense primers) and 5 different 5′-T12MN-3′ primers (antisense primers, where MN were GG, GC, CG, GA, or AT). The PCR products were labeled with 50 nm32P (3000 Ci/mmol), separated by 6% polyacrylamide gel electrophoresis, and detected by autoradiography. DNA was eluted from the differentially amplified bands, reamplified by PCR as described (20Liang P. Pardee A.B. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, Inc., New York1995: 15.8.1-15.8.8Google Scholar), cloned into pCR-Script Amp SK(+) (Stratagene, La Jolla, CA), and sequenced by an automated ABI PRISM 377 DNA Sequencer. Sequence analysis was carried out with GCG supplied programs. The HL-60 unidirectional cDNA library (obtained from Invitrogen, Carlsbad, CA, number 950-07) was constructed in pcDNA I vector usingNotI dT primer and transfected into the Escherichia coli MC1061/PC. Screening of the HL-60 cell line cDNA library was performed by in situ hybridization with a radioactive labeled probe (22Strauss W.M. Ausubel F.M. Brent R. Kingston R.E. Moore D.D. Seidman J.G. Smith J.A. Struhl K. Current Protocols in Molecular Biology. John Wiley and Sons, Inc., New York1995: 6.3.1-6.3.6Google Scholar). The DNA templates for in vitro transcription and translation were either the TA-40 pcDNA I plasmid linearized at the XhoI site (S1) or TA-40 derived fragments. These fragments were obtained by PCR reactions with T7 promoter primer TAATACGACTCACTATAGGG in combination with each of the following TA-40 specific primers: CTCACACACACAACCATCC (S2), CGGATTTGGGAAACTTTTCTATAAA (S3), and AGCTGTGCTTTACATAGCAATCTT (S4). Transcription reactions with T7 RNA polymerase were performed using the Ribo MAXTM Large Scale RNA Production System as recommended by the supplier. The quality of the synthesized RNA was determined by agarose gel electrophoresis. Protein synthesis assays were carried in a standard micrococcal nuclease-treated rabbit reticulocyte lysate reaction mixture according to the manufacturer's instructions. Reactions containing BMV or luciferase transcripts as templates were incubated for 40 min at 30 °C in the presence of [35S]methionine. Aliquots were spotted onto 1-cm square filters and hot trichloroacetic acid percipitable radioactivity was determined. Where indicated, polyinosinic-polycytidylic acid (poly[I]·poly[C]) or TA-40 derived transcripts were added to the translation reaction prior to the addition of BMV or luciferase transcripts. All results are representative of at least three independent experiments. To identify genes whose expression is translationally regulated, we combined a sucrose gradient separation of polysomes from mRNPs with DDRT-PCR analysis and cDNA cloning. Polysomal and subpolysomal RNAs were purified from cells before and after treatment. These RNA samples were subjected to a DDRT-PCR analysis and the radioactive labeled cDNA products were separated by polyacrylamide gel electrophoresis. This protocol is illustrated schematically in Fig. 1. Distribution of RNAs between polysomal and subpolysomal fractions is indicated by the relative intensity of the bands of the same electrophoretic mobility in adjacent lanes of the autoradiogram. Differences in the distribution of the intensities of bands found in treated cells relative to nontreated cells pointed toward candidates for translationally regulated transcripts (see arrows in Fig. 1). The bands' intensity was changed linearly in the range of 28–40 PCR cycles and thus, even small differences in mRNA distribution could be detected (data not shown). To confirm the validity of this experimental approach, the regulation of ferritin expression was re-examined. Ferritin biosynthesis is a classical model for gene-specific translational control (23Klausner R.D. Rouault T.A. Harford J.B. Cell. 1993; 72: 19-28Abstract Full Text PDF PubMed Scopus (1055) Google Scholar, 24Melefors O. Hentze M.W. BioEssays. 1993; 15: 85-89Crossref PubMed Scopus (107) Google Scholar). This model of iron-dependent regulation was thoroughly investigated and established in a number of cell lines, including the K562 human erythroleukemia cells (25Mattia E. Josic D. Ashwell G. Klausner R. van Renswoude J. J. Biol. Chem. 1986; 261: 4587-4593Abstract Full Text PDF PubMed Google Scholar) that were used in this study as a test model. K562 cells were grown in suspension to a density of 5 × 105 cells/ml and then divided to untreated control culture and to experimental culture that was exposed to 100 mmhemin-iron for 4 h. Cytoplasmic RNA was separated by centrifugation on sucrose gradients and analyzed (Fig.2 A). The RNAs from polysomal and subpolysomal fractions were extracted separately. 10 μg of RNA from each fraction were subjected to a Northern blot analysis with32P-labeled human H-ferritin cDNA probe (Fig.2 B). In the untreated cells most of the H-ferritin mRNA sedimented with the subpolysomal fraction (Fig. 2 B, lanes 1and 2), whereas, in the hemin-treated cells H-ferritin mRNA was equally distributed between the polysomal and subpolysomal fractions (Fig. 2 B, lanes 3 and 4). Similar results were obtained by semiquantitative RT-PCR analysis performed with specific human H-ferritin primers (Fig. 2 B, lower panel). These results confirmed the iron-dependent mobilization of H-ferritin mRNA onto polysomes in K562 cells. It is worthwhile to mention that in these cells ferritin biosynthesis was reported to be regulated at the mRNA level as well (26Mattia E. den-Blaauwen J. Ashwell G. van-Renswoude J. Proc. Natl. Acad. Sci. U. S. A. 1989; 86: 1801-1805Crossref PubMed Scopus (35) Google Scholar). RNAs from the same samples were subjected to DDRT-PCR analysis (as described under “Experimental Procedures”); the 10-mer arbitrary oligonucleotides upstream primers used were: 1) 5′-AAGGTAGTGC-3′, 2) 5′-CTTGATTGCC-3′, 3) 5′-GCTATCACAG-3′; and the T12MN anchored primers were: 4) 5′-T12CG-3′, 5) 5′-T12AG-3′, and 6) 5′-T12AT-3′. The sequences of primers 1 and 4 were derived from the 3′-untranslated region of the human H-ferritin mRNA (27Chou C.C. Gatti R.A. Fuller M.L. Concan P. Wong A. Chada S. Davis R.C. Salser W.A. Mol. Cell. Biol. 1986; 6: 566-573Crossref PubMed Scopus (74) Google Scholar). The sequence of primer 1 is located 120 nucleotides upstream to the poly(A) addition site, whereas primer 4 is the H-ferritin mRNA-specific anchored primer. For a given RNA preparation, the DDRT-PCR reactions generated highly reproducible mobility patterns of 30–100 distinct bands, depending on the combination of primers. The relative intensity patterns of the bands of the same electrophoretic mobility in adjacent lanes (from control and treated cells) were reproducible only in 70–80%. Therefore, it was necessary to run the DDRT-PCR analysis in duplicate. An autoradiogram of the DDRT-PCR analysis, performed with the combination of primers numbers 1 and 4 and exposed for 2 h, is shown in Fig.2 C. An additional 25 bands appeared after a 12-h exposure (data not shown). The arrow on the autoradiogram points at bands with an estimated size of 130 nucleotides. Isolation and sequencing of these DNA fragments confirmed that they corresponded to the H-ferritin cDNA. Comparison between the signals in lanes 1 and 2 (control) and those in lanes 3 and4 (hemin-treated) indicated that the DDRT-PCR analysis produced similar results to those obtained by Northern blot and RT-PCR analyses (Fig. 2, B and C). This demonstrated the feasibility to identify mobilization of specific mRNAs onto polysomes using the DDRT-PCR analysis. Promyelocytic leukemia cell line HL-60 is one of the best studied models of cell differentiation (28Birnie G.D. Br. J. Cancer. 1988; 58: 41-45Google Scholar, 29Tsiftsoglou A.S. Robinson S.H. Int. J. Cell. Cloning. 1985; 3: 349-366Crossref PubMed Scopus (43) Google Scholar). This culture comprises 90–95% of cells with myeloblastic/promyelocytic morphology. A variety of agents can induce differentiation of these cells either to granulocyte-like (dimethyl sulfoxide) or to monocyte/macrophage-like (TPA or 1,25-dihydroxyvitamin D3) cells (28Birnie G.D. Br. J. Cancer. 1988; 58: 41-45Google Scholar). Differentiation of HL-60 cells into monocyte/macrophage-like cells was induced with 50 nm TPA. After 24 h of induction, proliferation was arrested, and almost all the cells adhered in aggregates to the plastic dish; then they spread out and acquired a spindle-shaped morphology and prominent pseudopodia. This series of events was confirmed microscopically in all experiments. To compare the polysomal profiles of HL-60 cells during their differentiation, cytoplasmic RNA from noninduced control, 12-, 24-, and 48-h treated cells was analyzed following centrifugation on sucrose gradients (Fig. 3 A). The relative amount of polysomes decreased gradually during the course of cellular differentiation (Fig. 3 A), reflecting the growth arrest and inhibition of overall protein synthesis associated with cellular differentiation. To identify transcripts whose translation was specifically regulated, we isolated RNAs from polysomal and subpolysomal fractions of control and TPA-treated cells, and examined them by differential display analysis (Figs. 3 B and5 A). Thirty DDRT-PCR reactions, employing various combinations of primers on each RNA sample, produced a total of about 2500 different bands. Assuming that an average of 15,000 different species of mRNAs are expressed per cell, our analysis covered about 15% of the cellular RNAs. Most bands in this analysis represented RNA species with unaltered polysomal/subpolysomal distribution and thus provided multiple internal controls. Some mRNA species (type A) were found both in polysomal and subpolysomal fractions, both before and after treatment (Fig. 3 B, m and p, in bothcon and 24-h lanes). Other species (type B) were present in subpolysomes only (about 10%, in m lanes only), or were found exclusively in polysomes (type C; about 5%, in p lanes only). Type B represents untranslated RNAs, whereas type C consists of RNAs that are apparently translated very efficiently, and therefore likely to be translationally regulated. In this study, however, only specific mRNAs whose translation was affected by cellular differentiation were further investigated.Figure 5Identification of mobilized transcripts during differentiation of HL-60 cells. A, differential display RT-PCR analysis of polysomal (p) and nonpolysomal (m) fractions of 48-h TPA-treated cells (48 h) or control untreated cells (con). The 10-mer arbitrary oligonucleotides upstream primers used in these experiments were: (i) 5′-GCTATCACAG-3′; (ii) 5′-GTCATGCACA-3′; (iii) 5′-TGGATCCAAG-3′; (iv) 5′-ATTGGGAAGG-3′. The open arrows point at the bands corresponding to candidates for translationally activated RNAs (TAs).B, RT-PCR analysis of TA transcripts, performed on nonpolysomal (lanes 1 and 3) and polysomal (lanes 2 and 4) fractions of control (lanes 1 and 2) or 48-h TPA-treated cells (lanes 3and 4) with TA 10, 12, 20, and 90 specific primers.C, RT-PCR analysis of nonpolysomal (lanes 1, 3, 5, 7, and 9) and polysomal (lanes 2, 4, 6, 8, and 10) fractions of control (lanes 1 and2) or 12- (lanes 3 and 4), 24- (lanes 5 and 6), 48- (lanes 7 and8), and 72- (lanes 9 and 10) h TPA-treated cells. TA-40 (TA40), actin (AC), and H-ferritin (FE) specific primers were used in these reactions (see “Expermental Procedures” for details).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Analysis of RNA from 24-h treated cells revealed few candidates for specifically repressed RNAs (arrows in Figs. 3 Band 4 a). In these cases bands derived from control cells exhibited usually similar intensities in subpolysomes and polysomes, whereas in treated cells the intensity of the bands from subpolysomes was much higher than that of the polysomes (arrows in Figs.3 B and 4 A). These DNA bands (designated TR10, 11, 13, 40, and 80) were purified from the gels, amplified, cloned, and used as probes for RNA blot analysis of 10 μg of total RNA from different fractions. Autoradiograms of these analyses are presented in Fig. 4 B. It is evident that most of the corresponding mRNAs in the 24-h treated cells were found in subpolysomal fractions (Fig. 4 B, lanes 3 and4). Comparison between the signals in lanes 1 and2 (control) and those in lanes 3 and 4(treated) indicates that the later mRNAs were no longer associated with the polysomes upon differentiation (Fig. 4 B). Semiquantitative analysis revealed no significant changes in the total amount of TR mRNAs during the early stages of HL-60 cellular differentiation (data not shown). The kinetics of translational repression of these TR genes was studied by semiquantitative specific RT-PCR. An example of such analysis is shown in Fig. 4 C. The release of TR-RNA from the polysomes was detected 12 h after TPA treatment (compare lanes 1 and 2, control, tolanes 3 and 4, 12-h treated). Expression of actin mRNA (AC) was determined by Northern blot (Fig. 4 B) and RT-PCR (Fig. 4 C) analysis to test the specificity of the observed phenomena. It is worthwhile to mention that actin mRNA was always found predominantly in polysomal fractions. The TR clones (TR 10, 11, 80, 13, and 40) were sequenced. Search for homologous sequences in the Gene Banks revealed that these clones derived from the 3′ termini of the human mRNAs for ribosomal proteins L13a, L19, L11, S27 and the mRNA for ubiquitin 52 amino acid fusion protein (the natural hybrid between ubiquitin monomer and ribosomal protein), respectively. All five transcripts were relatively short mRNAs (from ∼340 to ∼700 nucleotides) with polypyrimidines at their 5′ end, indicating the possibility that they were members of the 5′-TOP mRNA family. These results are in agreement with the accepted concept of translational repression of 5′-TOP's mRNAs in nonproliferating cells (for review, see Ref. 30Meyuhas O. Avni D. Shama S. Hershey J.W.B. Mathews M.B. Sonenberg N. Translational Control. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1996: 363-388Google Scholar). Further analysis of the 48-h TPA-treated cells revealed, in addition to the TR genes, 10 specific bands with relative enhanced intensity in the polysomal fraction (some of them are labeled by arrows in Fig. 5 A). These bands represented mRNA candidates for translationally activated genes and were designated TA genes. They were isolated from the gels, amplified, cloned, sequenced, and used as probes for RNA blot analysis. We could not detect the corresponding RNAs in the Northern blot analysis of HL-60 total RNA (20 μg), probably due to their low abundance in these cells. Therefore, eight pairs of TA-specific primers were used in semiquantitative RT-PCR analysis (Fig. 5, B andC). This analysis confirmed that five out of the eight tested TA transcripts represented mRNAs that were mobilized onto polysomes upon differentiation (see TA-12, 20, 90, and 40 in Fig. 5,B and C; TA 10*
Referência(s)