Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (237K) Freely available
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Hanna, M. M.
Right arrow Articles by Riggs, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hanna, M. M.
Right arrow Articles by Riggs, D. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 1369-1376  

Probing the environment of nascent RNA in Escherichia coli transcription elongation complexes utilizing a new fluorescent ribonucleotide analog
Introduction
Materials And Methods
   DNA templates, enzymes and chemicals
   Synthesis and structural characterization of 5-SF-UTP
   Transcription with T7 RNA polymerase
   Transcription with yeast RNA polymerases I and III
   Transcription with E.coli RNA polymerase
   Fluorescence polarization of RNA in E.coli transcription complexes
   Modeling of RNA and RNA-DNA hybrids containing 5-SF-UMP
Results
   Synthesis and structural characterization of 5-SF-UTP
   Transcription with T7 RNA polymerase
   Transcription with yeast RNA polymerases I and III
   Transcription with E.coli RNA polymerase
   Fluorescence polarization of RNA in E.coli transcription complexes
   Modeling of RNA and RNA-DNA hybrids containing 5-SF-UMP
Discussion
   Transcription properties of 5-SF-UTP
   Probing the environment of the RNA in E.coli elongation complexes: polarization data and molecular modeling
   Summary
Acknowledgements
References

Probing the environment of nascent RNA in Escherichia coli transcription elongation complexes utilizing a new fluorescent ribonucleotide analog

Probing the environment of nascent RNA in Escherichia coli transcription elongation complexes utilizing a new fluorescent ribonucleotide analog

Michelle M. Hanna*, Elizabeth Yuriev, Juan Zhang and Daniel L. Riggs

Department of Chemistry and Biochemistry, University of Oklahoma, 620 Parrington Oval Room 208, Norman, OK 73019-0370, USA

Received October 30, 1998; Revised and Accepted January 9, 1999

ABSTRACT

We report the synthesis and characterization of 5-thioacetamidofluorescein-uridine 5[prime]-triphosphate (5-SF-UTP), and its application to the characterization of the environment of the nascent RNA during trans-cription. This analog specifically replaced UTP as a transcription substrate for Escherichia coli and T7 RNA polymerases, and yeast RNA polymerase III. Escherichia coli transcription complexes containing analog incorporated at only position +21 of the RNA were prepared. The RNA was then elongated in the absence of analog, moving the fluorescent group further away from the enzyme active site, and the fluorescence polarization was measured. Analog positioned near the 3[prime] end of the transcript exhibited significantly increased polarization relative to that of free probe, consistent with the constrained environment of the RNA in the DNA-RNA hybrid. Analog positioned 14 nucleotides from the 3[prime] end exhibited significantly decreased polarization relative to that at the 3[prime] end of the RNA, but only slightly above that of free RNA, suggesting that the probe was on the solvent-exposed surface of the polymerase. Molecular modeling of these analog-substituted RNAs produced structures consistent with the experimental data. The excellent substrate properties of this analog make it useful for the characterization of the environment of RNA not only during transcription and translation, but in any type of ribonucleoprotein complex.

INTRODUCTION

RNA molecules function as structural components of ribonucleoprotein complexes, as messengers in the transmission of genetic information, and as ribozymes that catalyze chemical reactions. Our work has focused primarily on characterization of the molecular mechanisms of transcription, with a focus on the contacts made by the RNA in the transcription complex, and how these are altered by various transcription factors. For this purpose, we have developed a variety of ribonucleotide analogs that contain base modifications at positions that do not interfere with normal Watson-Crick base-pairing. These analogs have proved to be excellent substrates for a variety of RNA polymerases, and they can be specifically incorporated at one or more positions in RNA during transcription in vitro.

Here we describe the third in a series of analogs that have different modifications at the 5 position of UTP. The first, 5-[4-(azidophenacyl)thio]-UTP (5-APAS-UTP), contained an aryl azide group that can be photo-crosslinked to adjacent molecules (1). This analog was used to identify the RNA polymerase subunits and transcription factors that make direct contact with the RNA during transcription (2-5). These experiments provided only minimal information about the environment of the RNA with regard to surface exposure and solvent accessibility, however. The second analog, 5-mercapto-UTP (5-SH-UTP), contained an unmodified thiol group on the base that can be modified post-transcriptionally with a variety of reagents (6), which allowed us to begin examining the solvent accessibility of the RNA local environment.

We have now synthesized a related fluorescent analog, 5-thioacetemidofluorescein-uridine 5[prime]-triphosphate (5-SF-UTP), that contains a fluorescein group attached by a short, flexible linker to the 5 position of UTP. This analog now allows us to compare the degree of hindrance of nucleotides at different positions in the RNA in the Escherichia coli transcription complex by comparing the fluorescent polarization of analog in different transcription complexes. The data from the polarization experiments are in agreement with the conformations predicted by molecular mechanics simulations.

MATERIALS AND METHODS

DNA templates, enzymes and chemicals

Plasmid pAR1707 (7), which contains the E.coli RNA polymerase promoter A1 from bacteriophage T7 and the Rho-independent te terminator, and plasmid pKK 34-121 (8), which contains the E.coli rrnB P1 and P2 promoters and Rho-independent t1 terminator, were used as transcription templates for E.coli RNA polymerase. Escherichia coli RNA polymerase (holoenzyme) was purified from MRE 600 cells (Grain Processing Corp.) using published procedures (9,10). T7 RNA polymerase and a linear DNA template which contains the bacteriophage T7 [phis]10 promoter were kindly provided by Dr William McAllister (SUNY, Brooklyn, NY). Ultrapure ribonucleoside triphosphates used for transcription reactions were purchased from Pharmacia LKB Biotechnology Inc., Tris (carboxyethyl)phosphine (TCEP) and 5-iodoacetamidofluorescein (5-IAF) were from Molecular Probes Inc., and ultralow fluorescence Tris was from Pan Vera Corp.All RNA samples were analyzed by denaturing electrophoresis on polyacrylamide (acrylamide:bisacrylamide = 19:1) gels containing 7 M urea. The RNA sample load buffer was 80% (v/v) formamide, 0.1% (w/v) bromophenol blue, 0.1% (w/v) xylene cyanol, 10 mM Na2EDTA.

Synthesis and structural characterization of 5-SF-UTP

5,5[prime]-Dithiobis(uridine 5[prime]-triphosphate) (bis-5-S-UTP; Fig. 1, I) was synthesized and isolated as described (1,11,12). Bis-5-S-UTP (10 µmol) was dissolved in 5 ml distilled water and the pH was adjusted to 4.0 with glacial acetic acid. The 5-SH-UTP (Fig. 1, II) was formed upon addition of TCEP (20 µmol) and stirring at 4°C for 1 h (13). The pH was adjusted to 8.0 with ammonium hydroxide, and the formation of 5-SH-UTP was verified spectrophotometrically by the appearance of a new absorbance peak at 330 nm ([epsis]330 = 8 × 103 M-1 cm-1 at pH 8.0) (14). All remaining manipulations were carried out in reduced light. Immediately after reduction, 5-thioacetemidofluorescein-uridine 5[prime]-triphosphate (5-SF-UTP; Fig. 1, III) was prepared by addition of 100 µmol 5-iodoacetamidofluorescein (15,16). The mixture was stirred overnight at 4°C in the dark. The pH of the reaction was then adjusted to 5.0 with glacial acetic acid and the red precipitate formed was removed by filtration through a 0.45 µm filter. The filtrate was lyophilized overnight and the solid was redissolved in 0.5 ml water and adjusted to pH 8.0 with ammonium hydroxide. The sample was then loaded onto a 10 × 1 cm silica gel column equilibrated in 95% (v/v) ethanol and eluted with 30 ml of 95% (v/v) ethanol, followed by 20 ml of 1 µM NH4OH, pH 8.0. Fractions (1 ml) were collected and an aliquot was analyzed on a PEI-cellulose F TLC plate, as described previously (1)


Figure 1. Synthesis of 5-SF-UTP. 5-SF-UTP was synthesized by reduction of 5,5[prime]-dithiobis(uridine 5[prime]-triphosphate) (I, bis-5-S-UTP) with Tris (carboxyethyl)-phosphine (TCEP, Step 1) to 5-mercapto-UTP (II, 5-SH-UTP). The fluorescent group was introduced by alkylation of 5-SH-UTP with 5-iodoacetamidofluorescein (5-IAF, Step 2) to give 5-thioacetamidofluorescein-UTP (III, 5-SF-UTP). The related photocrosslinking nucleotide analog, 5-APAS-UTP (IV) was prepared from II by alkylation with APB (azidophenacylbromide).

5-SF-UTP was separated from bis-5-S-UTP on a Beckman Ultraprep C18 column (21.2 mm × 15 cm) and eluted with a 17 min linear gradient of 5-33% (v/v) acetonitrile in 50 mM triethylammonium bicarbonate, pH 7.0, followed by a 3 min wash with 33% (v/v) acetonitrile. The flow rate was 5 ml/min and 5-SF-UTP eluted at 13 min. The solvent was removed by lyophilization, and the resulting product was dissolved in a minimum volume of water. The product concentration was determined spectrophotometrically ([epsis]491 = 8.2 × 104 M-1 cm-1 for fluorescein). The excitation and emission spectra of 5-SF-UTP were obtained with a Hitachi Fluorescence Spectrophotometer F-2000. The measurements were performed in 20 mM ultralow fluorescence Tris-HCl buffer (pH 8.9) at room temperature.

Transcription with T7 RNA polymerase

Transcription reactions contained 20 nM T7 [phis]10 promoter DNA, 40 nM T7 RNA polymerase, 40 µM GpG, 20 µM ATP, 20 µM CTP, 2 µM [[alpha]-32P]GTP (5 × 104 c.p.m./pmol), and the indicated concentrations of UTP or 5-SF-UTP in 40 mM Tris-HCl (pH 7.6), 15 mM MgCl2, 10 mM [beta]-mercaptoethanol, and 50 µg/ml acetylated bovine serum albumin (AcBSA). The reactions were incubated at 37°C for 10 min and the RNA was precipitated with ethanol, dissolved in 15 µl RNA load buffer and analyzed by electrophoresis on a 5% gel.

Transcription with yeast RNA polymerases I and III

Preparation of the yeast RNA polymerase I and III transcription extracts and the rDNA template pDR10 and tDNA template pTZ1 have been described previously (17,18). RNA polymerase I transcription assays contained: 20 mM Tris-acetate (pH 7.5), 10 mM magnesium acetate, 1 mM dithiothreitol, 200 mM potassium glutamate, 0.125 µg/ml creatine kinase, 30 µM creatine phosphate, 200 µM ATP and CTP, 15 µM [[alpha]-32P]GTP (4 × 104 c.p.m./pmol), UTP and/or 5-SF-UTP, as indicated, and 2.5 µg/ml of DNA template in a final volume of 20 µl. The assays were incubated at 30°C for 5 min, then extracted with phenol:chloroform:isoamyl alcohol (25:24:1) and the RNA was precipitated with ethanol. The dried pellets were then resuspended in 3 µl RNA load buffer and analyzed by electrophoresis on a 5% gel.

Transcription with E.coli RNA polymerase

Transcription complexes were formed at the bacteriophage T7 A1 promoter by incubation of 20 nM plasmid pAR1707, 20 nM E.coli RNA polymerase holoenzyme and 100 µM ApUfor 1 min at 30°C in 20 µl of 20 mM Tris-acetate (pH 7.8), 10 mM magnesium acetate, 50 mM potassium glutamate, 4% (v/v) glycerol, 0.04 mM EDTA, 0.04 mg/ml AcBSA. RNA synthesis was initiated by addition of ultrapure ATP, CTP (1 µM final) and [[alpha]-32P]GTP (0.33 µM final, 6.7 × 107 c.p.m./pmol), and the reaction was incubated at 30°C for 2 min. The RNA polymerase in the resulting ternary transcription complexes was paused after incorporation of AMP into the 3[prime] end of the transcript at+20 because the next encoded nucleotide is UMP (RNA sequence is 5[prime] AUCGA GAGGG-10 ACACG GCGAA-20 UAGCC AUCCC-30 AAUCG A....). These A-20 ternary complexes (so named because the 3[prime] nucleotide is A and the RNA is 20 nt long), were separated from the unincorporated nucleotides by gel exclusion chromatography at room temperature on a 15 × 1 cm Sepharose 6B column equilibrated in a buffer containing 20 mM Tris-acetate (pH 7.8), 10 mM magnesium acetate, 50 mM potassium glutamate, 4% (v/v) glycerol, 0.04 mM EDTA.

The A-20 RNA was extended by 1 nt to U-21 by incubation in the presence of either UTP (1 µM final), or various concentrations of 5-SF-UTP, for 1 min at 30°C. The A-20 RNA was extended 2 nt to A-22 by incubation at 30°C for 1 min with ATP (1 µM final) and either UTP (1 µM final) or 5-SF-UTP (100 µM final). An equal volume of RNA load buffer was added, and the RNA was analyzed by electrophoresis on a 40 × 0.75 mm, 20% denaturing polyacrylamide gel.

To incorporate analog throughout the RNA, transcription was initiated from the rrnB P1 and P2 promoters on the plasmid pKK 34-121 with all four ribonucleoside triphosphates present. The transcription assays were incubated at 37°C for 10 min andeach contained 200 µM ATP and CTP, 40 µM [[alpha]-32P]GTP(2.5 × 104 c.p.m./pmol), 5 nM plasmid pKK 34-121 DNA, 160 nM E.coli RNA polymerase holoenzyme, and UTP or 5-SF-UTP in 20 µl of a buffer containing 44 mM Tris-HCl (pH 8.0), 14 mM [beta]-mercaptoethanol, 14 mM MgCl2, 20 mM NaCl, 2% (v/v) glycerol, 40 µM EDTA, 0.04 mg/ml AcBSA. Transcription was stopped by addition of an equal volume of RNA load buffer, and the RNA was analyzed by electrophoresis on a 10% denaturing polyacrylamide gel.

Fluorescence polarization of RNA in E.coli transcription complexes

For polarization studies, A-20 ternary transcription complexes were assembled at the T7 A1 promoter (see the RNA sequence above) with 100 nM pAR1707, 300 nM E.coli RNA polymerase, 100 µM ApU, 5 µM ATP and CTP, and 2 µM [[alpha]-32P]GTP in 420 µl of 20 mM Tris-acetate (pH 7.8), 10 mM magnesium acetate, 50 mM potassium glutamate, 4% (v/v) glycerol, 0.04 mM EDTA and 0.04 mg/ml AcBSA. After incubation at 30°C for 3 min, the ternary complexes were purified by gel exclusion chromatography, as described above. Complexes containing U-21 RNA with analog at the 3[prime] end were prepared by incubating A-20 complexes with 30 µM 5-SF-UTP for 10 min at 30°C. Complexes containing C-24 RNA with analog at U-21 were prepared by incubating A-20 complexes with 30 µM 5-SF-UTP, 10 µM ATP and GTP, and 100 µM 3[prime]-OMe-CTP (an RNA chain terminator) for 10 min at 30°C. For both the U-21 and C-24 complexes, the unincorporated 5-SF-UTP was removed by gel exclusion chromatography, as described above, before polarization measurements were taken. Transcription complexes containing G-35 or the full-length 161 nt transcript were made by first preparing transcription complexes elongated to position +23 (G-23) that contained the analog at U-21. This was done by incubating A-20 complexes with 30 µM 5-SF-UTP, and 10 µM ATP and GTP for 10 min at 30°C. After gel exclusion chromatography, the G-23 RNA was elongated to the 35mer (G-35) or to the te terminator (at +161 nt) by incubation at 30°C for 10 min in the presence of either ATP, CTP, UTP and 3[prime]-OMe-GTP or ATP, CTP, GTP, UTP (each at 10 µM), respectively. Aliquots were removed for analysis of the RNA by electrophoresis on a 20% gel, and the remainder of the transcription complexes were used for the polarization studies.

The emission polarization of the RNA was measured immediately after the transcription complexes were prepared and separated from unincorporated analog. The measurement was performed on a Hitachi F-2000 fluorescence spectrophotometer at room temperature under the following conditions: excitation and emission wavelengths of 491 and 515 nm, respectively, integration time of 10 s, response time of 10 s, bandpass of 10 nm for both excitation and emission, and a PM voltage of 700 V. Three or more replicates were measured and the average fluorescence polarization and average deviation were determined.

Modeling of RNA and RNA-DNA hybrids containing 5-SF-UMP

Structures of analog-containing oligonucleotides were modeled using the molecular mechanics facilities of HyperChem (version 5.1, Hypercube Inc., Waterloo, ON, Canada), running on a COMPAQ PentiumII computer. Structure optimization was carried out using the modified AMBER force field (19). See supplementary material for force field modification, charge derivation method and optimization protocol details. For oligonucleotides containing 5-SF-U or 5-APAS-U, input structures for conformational searching were prepared by (i) constructing a single or double stranded RNA helix (A-form, 3[prime]-endo), (ii) mutating the complementary strand from RNA to DNA for hybrids and (iii) optimizing structures to a delta RMS gradient of 0.1 kcal mol-1 Å-1. Following this, the U residue at position 21 in the RNA transcript was modified to replace the H5 atom with the SF or APAS group. For each oligonucleotide, Monte Carlo ‘usage directed’ conformational search (conformations within 20 kcal/mol above lowest were sampled) was carried out. During the search, the torsion angles around rotatable bonds in the SF and APAS groups were varied (see supplementary material), and the SF group was optimized while the rest of the molecule was kept frozen. The sequences of the RNAs synthesized during the polarization experiments were truncated for modeling (without the loss of important intramolecular contacts) to speed up the simulation: for the 21mer (U-21) and 24mer (C-24) RNAs, oligonucleotides containing 7 nt were used (from position G15 to U21 and G18 to C24, respectively); for G-35, the oligonucleotide 11 residues long, representing one helix turn, was used (nt G16-A26).

RESULTS

Synthesis and structural characterization of 5-SF-UTP

The fluorescent UTP analog, 5-thioacetamidofluorescein-UTP (5-SF-UTP, Fig. 1, III), was synthesized in 38% yield (based on bis-5-S-UTP) by alkylation of 5-SH-UTP (Fig. 1, II, step 1). 5-SF-UTP was isolated by HPLC purification and eluted from the column at ~13 min. The product had an absorbance maximum at 491 nm, and an emission maximum at 515 nm. The product structure was verified as described in (1). Thin-layer chromatography on PEI-cellulose F plates indicated a single product with Rf = 0.07. After enzymatic digestion with calf intestine alkaline phosphatase, this product was converted to a new product withRf = 0.17 (data not shown). This change in mobility is consistent with conversion of the nucleoside 5[prime]-triphosphate to the nucleoside.

Transcription with T7 RNA polymerase

Transcription with bacteriophage T7 RNA polymerase was compared in reactions containing either UTP or 5-SF-UTP. In the presence of only 5-SF-UTP and no UTP, only short RNAs were synthesized (Fig. 2A, lanes 3 and 4). These were found to be abortive initiation products which could not be elongated to full-length RNA, and they were also observed when UTP, but no analog was present (lane 1), and also when only ATP, CTP and GTP were present (lane 2). To obtain the full-length, runoff transcripts containing the analog, a small amount of UTP (1 or 2 µM) was added to the reactions (lanes 6 and 8). The full-length RNA produced could only result from incorporation of the analog, as these concentrations of UTP are too low to produce full-length RNA (lanes 5 and 7).


Figure 2. Transcription with T7 and yeast RNA polymerases. (A) Evaluation of 5-SF-UTP as a substrate for T7 RNA polymerase. Autoradiogram showing the RNA synthesized in the presence of GTP, ATP, CTP and the following concentrations of UTP and/or analog: 20 µM UTP (lane 1), no UTP or analog (lane 2), 0.5 mM 5-SF-UTP (lane 3), 1 mM 5-SF-UTP (lane 4), 1 µM UTP (lane 5), 0.5 mM 5-SF-UTP and 1 µM UTP (lane 6), 2 µM UTP (lane 7), 0.5 mM 5-SF-UTP and 2 µM UTP (lane 8). (B) Evaluation of 5-SF-UTP as a substrate for yeast RNA polymerases. Autoradiogram showing the RNA synthesized in an RNA polymerase III extract containing a tDNA template (lanes 1-8), or an RNA polymerase I extract containing an rDNA template (lanes 9-12). The assays contained ATP and CTP (100 µM each), GTP (15 µM) and the following concentrations of UTP or analog: 100 µM UTP (lane 1), 1 mM 5-SF-UTP (lane 2), 5 µM UTP (lane 3), 5 µM UTP and 1 mM 5-SF-UTP (lane 4), 12.5 µM UTP (lane 5), 12.5 µM UTP and 1 mM 5-SF-UTP (lane 6), 25 µM UTP (lane 7), 25 µM UTP and 1 mM 5-SF-UTP (lane 8), 100 µM UTP (lane 9), no UTP or analog (lane 10), 5 µM UTP (lane 11), 5 µM UTP and 1 mM 5-SF-UTP (lane 12).

Transcription with yeast RNA polymerases I and III

The transcription substrate properties of 5-SF-UTP for eukaryotic RNA polymerases were examined with yeast whole cell extracts that support either RNA polymerase I and III transcription. The analog incorporation by RNA polymerase III was found to be similar to that by T7 RNA polymerase. It could not completely replace UTP for synthesis of full-length RNA (Fig. 2B, compare lanes 1 and 2), but when the assays were supplemented with as little as 12.5 µM UTP, which alone could not support transcription (lane 5), the analog was incorporated into full-length tRNA (lane 6). Addition of 25 µM UTP stimulated even greater production of full-length RNA (lanes 7 and 8).

Yeast RNA polymerase I requires lower concentrations of UTP for transcription of full-length RNA than does RNA polymerase III, producing full-length RNA with UTP concentrations as low as 5 µM (Fig. 2B, lane 11). However, unlike RNA polymerase III, the addition of 5-SF-UTP to the RNA polymerase I reaction completely inhibited production of full-length RNA, even in the presence of UTP (lane 12).

Transcription with E.coli RNA polymerase

Specific replacement of UTP by the analog as a substrate for E.coli RNA polymerase was verified by transcription from the T7 A1 promoter (RNA sequence: 5[prime]-AUCGA GAGGG ACACG GCGAA UAG-3[prime], Fig. 3A). Ternary transcription complexes in which the polymerase is paused after incorporation of an A residue at nucleotide position +20 (A-20 complexes) were prepared by initiating transcription with the dinucleotide ApU and omitting UTP from the reaction (Fig. 3B, lane 1 and C, lane 1). A-20 complexes were incubated with increasing concentrations of 5-SF-UTP (Fig. 3B, lanes 2-13), and the apparent Km for 5-SF-UTP at this position was 15 µM. To determine if the analog could be incorporated at internal positions in the RNA, A-20 RNA was first extended to U-21 with UTP (Fig. 3C, lane 2) or 5-SF-UTP (lane 4), and then extended to A-22 with ATP (lanes 3 and 5). Incubation of A-20 complexes with ATP in the absence of UTP or analog resulted in no elongation of the A-20 RNA(data not shown).


Figure 3. Transcription with E.coli RNA polymerase. (A) The RNA sequence from the bacteriophage T7 A1 promoter, which is transcribed by E.coli RNA polymerase. (B) Titration of 5-SF-UTP. Autoradiogram of an RNA gel showing the specific incorporation of 5-SF-UMP at the 3[prime] end of RNA. A-20 complexes (lane 1) were incubated with the following concentrations of 5-SF-UTP: 1 µM (lane 2), 2 µM (lane 3), 3 µM (lane 4), 4 µM (lane 5), 10 µM (lane 6), 15 µM (lane 7), 20 µM (lane 8), 30 µM (lane 9), 40 µM (lane 10), 50 µM (lane 11), 75 µM (lane 12), 100 µM (lane 13). The positions of the A-20 (no analog), U-21 and A-22 RNAs are indicated at the right. (C) Incorporation of 5-SF-UMP internally 1 nt from the 3[prime] end of RNA. RNA in A-20 complexes (lane 1) was extended in the presence of either 1 µM UTP (lane 2), 1 µM UTP and ATP (lane 3), 100 µM 5-SF-UTP (lane 4), or 100 µM 5-SF-UTP and 1 µM ATP (lane 5). The positions of the U-21 and A-22 RNAs are indicated at the right. (D) Incorporation of 5-SF-UMP internally at multiple positions in RNA. Full-length terminated transcripts were synthesized from the tandem E.coli rrnB P1 and P2 promoters in the presence of GTP, ATP, CTP and the following concentrations of UTP and/or analog: 200 µM UTP (lane 1), no UTP or analog (lane 2), 500 µM 5-SF-UTP (lane 3), 1 mM 5-SF-UTP (lane 4), 2 mM 5-SF-UTP (lane 5), 1 µM UTP (lane 6), 200 µM 5-SF-UTP and 1 µM UTP (lane 7). The 260 and 380 nt transcripts which are produced by termination at the t1 terminator are indicated to the right of the autoradiogram.

To evaluate the characteristics of 5-SF-UTP as an elongation substrate for production of full-length transcripts, a truncated E.coli ribosomal RNA transcription unit containing the rrnB P1 and P2 promoters upstream of the t1 terminator was used as a DNA template. When 0.5 mM 5-SF-UTP replaced UTP in the transcription assay, most of the transcription complexes were stalled within the transcription unit, and few full-length trans-cripts were produced (Fig. 3D, lane 3). Increasing the analog concentration, even in the absence of UTP, increased the synthesis of full-length transcripts (lanes 4 and 5). Supplementation of the 5-SF-UTP-containing assays with low concentrations of UTP, which alone could not support transcription (lane 6), increased the production of full-length transcripts significantly (lane 7).

Fluorescence polarization of RNA in E.coli transcription complexes

To make transcription complexes for the polarization studies, 5-SF-UMP was incorporated by E.coli RNA polymerase at position +21 of the T7 A1 transcript (Fig. 4A). The A-20 RNA was extended in the absence of analog to either 24 (C-24), 35 (G-35), or 161 nt (Fig. 4A and B), moving the fluorescence probe at position +21 away from the 3[prime] end of the RNA and the active site of the enzyme. Unlike the other RNAs, which are still part of transcription complexes, the 161 nt RNA is a released transcript produced by termination at the te terminator. The fluorescence polarization of either the free analog, or the analog in different positions within the RNA transcript was measured (Fig. 4C). The polarization of free analog, not incorporated into RNA, was0.047 ± 0.008. When the analog was positioned at the 3[prime] end of the transcript, or 3 nt from the 3[prime] end, the polarization was much higher (0.196 ± 0.025 and 0.217 ± 0.024, respectively). When positioned 13 nt from the 3[prime] end of the transcript in the 35mer, the analog showed a decrease in polarization (0.129 ± 0.013). When the RNA was further extended and released from the trans-cription complex, an additional decrease in polarization resulted(0.105 ± 0.016), but it still remained higher than that of free analog. The differences between the polarization of the free analog and of analog at different positions in the RNA are plotted in Figure 4C.

   A, B, C
   D

Figure 4. Preparation of 5-SF-UMP-containing RNAs for fluorescent polarization studies. (A) The sequence of the RNA molecules containing analog atposition +21. The position of the analog at +21 is indicated by the asterisk. (B) Autoradiogram showing the RNAs produced for polarization studies. RNA in A-20 ternary complexes (lane 1) was extended in the presence of analog to produce RNA containing 5-SF-UMP at position +21 (lane 2). After purification to remove unincorporated analog, the RNAs were then extended to either +24, +35 or +161 as described in Materials and Methods (lanes 3-5, respectively). (C) Relative fluorescent polarization of RNA in transcription complexes or free in solution. The change in polarization of the analogs at position U-21 of the 21, 24, 25 and 161 nt RNAs compared to free 5-SF-UTP is shown. The average of three experiments is represented. (D) Proposed location of the fluorescent groups in the RNAs with respect to the DNA template and the RNA polymerase. Current models for E.coli RNA polymerases elongation complex structure include a DNA-RNA hybrid at the 3[prime] end of the RNA and a tunnel through the polymerase which functions as an exit port for the RNA.

Modeling of RNA and RNA-DNA hybrids containing 5-SF-UMP

We have modeled oligonucleotides containing 5-SF-UMP to investigate the conformational flexibility of the 5-SF group when in the DNA-RNA (U-21) hybrid (Fig. 5B), the DNA-RNA (C-24) hybrid (Fig. 5C), or in single-stranded RNA (G-35, Fig. 5D). The DNA-RNA hybrid that contains 5-APAS-UMP at position 21 was modeled for comparison (Fig. 5A). The lowest energy conformation (yellow) and other low energy conformations are shown.


Figure 5. Conformations of the analogs in single-stranded RNA and in DNA-RNA hybrids. The low energy conformations for oligonucleotides U-21, C-24 and G-35, which contain analog at position U-21, are shown. The hydrogens have been omitted for clarity. The U-21 and C-24 RNAs are shown in DNA-RNA hybrids (A-C). (A) The analog at position U21 is 5-APAS-UMP. (B-D) The analog at position U21 is 5-SF-UMP. The G-35 is shown as single-stranded RNA. The RNA strand is shown in cyan and the DNA strand in white, with the arrow tip corresponding to the 3[prime] end of the DNA or RNA chain. The ladder representation (top) is complemented by the visualization of the Connolly solvent accessible surface (bottom) to demonstrate the orientation of the fluorescein moiety and the contacts it makes in the low energy conformations. Analog conformation color coding: lowest energy, yellow; second in energy, red; third, green; fourth, blue; fifth, magenta.

For the DNA-RNA (U-21) hybrid containing 5-APAS-UMP (Fig. 5A), in the lowest energy conformation and in conformation #5 (which differ with respect to the orientation around CH2), the APAS group is restricted by the helix. In conformations #2, #3 and #4, the APAS group is facing the solution and is [pi]-stacked with nucleotide rU21 in the RNA. In the DNA-RNA hybrids containing 5-SF-UMP (Fig. 5B and C), the lowest energy conformations are stabilized by hydrogen bonding between the oxygen(s) of the carboxylic group of the fluorescein moiety and the hydrogen of the exocyclic amino group of a nucleobase. In U-21, the fluorescein group is buried in the duplex, in close proximity to rC16 and the rA19/dT19 base pair. In the second lowest energy conformation (red), which is >2 kcal/mol above that of the lowest conformation, the fluorescein moiety is more exposed to solution, but it is still restrained by the hybrid and makes contacts near nucleotides rA19 and rA20. In the lowest energy conformation (yellow) of C-24 (Fig. 5C), the three-ring part of the fluorescein moiety is [pi]-stacked outside of the guanine nucleobase of dG24. In the second energy conformation (red), the fluorescein moiety is again buried by the hybrid, with close contacts to nucleobases in bp 20-24. In the low energy conformations of the G-35 RNA (Fig. 5D), the fluorescein moiety is exposed to the solution and approaches both the RNA 3[prime] (lowest energy, yellow) and 5[prime] ends (third lowest energy conformation, green), and can also orient parallel to the helix axis (second energy conformation, red).

DISCUSSION

Transcription in E.coli is carried out by a single RNA polymerase core enzyme with subunit composition [alpha]2[beta][beta][prime], with specific promoter recognition involving an additional subunit, the [sigma] factor. Current models suggest that in E.coli elongation complexes (RNA longer than ~20 nt), the bases nearest the 3[prime] end of the transcript are part of a DNA-RNA hybrid of ~8-10 bp in length (for reviews see 20-24). The precise length of the hybrid may differ in different elongation complexes, and the hybrid length may be slightly longer (10-12 bp) in initial transcribing complexes which have not fully matured to elongation complexes (RNA 10-20 nt). Some data now suggest that after the RNA is displaced from the DNA template, it moves through an RNA binding channel or tunnel on the RNA polymerase, eventually exiting this tunnel and leaving the surface of the enzyme via an RNA exit port. Such a port could serve as a regulatory point, presenting the nascent RNA to regulatory proteins, such as NusA for attenuation, or to the ribosome for translation.

To characterize the RNA-binding domain(s) on E.coli RNA polymerase, we have developed and utilized a variety of photocrosslinking ribonucleotide analogs (1,6,11,12). When placed at the 5[prime] end, at internal positions, or at the 3[prime] end of the RNA, we observe RNA crosslinking first to [sigma], [beta] and [beta][prime] (25,26), then to only [beta] and [beta][prime] (3,5,27,28), and eventually to the C-terminal domain (CTD) of [alpha] (2,4,29). Our crosslinking data suggest that if an RNA exit port exists, it may involve the CTD of [alpha] (Fig. 4D) and lie near the binding site for the transcription factor NusA, which removes the nascent RNA from its binding site on the polymerase [alpha] subunit. To further characterize the environment of the nascent RNA as it moves through the transcription complex, and to specifically probe for an RNA tunnel, we have now developed a related fluorescent analog. This analog, 5-thioacet-amido-fluorescein-uridine 5[prime]-triphosphate (5-SF-UTP) contains a fluorescein group at the 5 position of the nucleotide base (Fig. 1, III). Modification of the 5 position of UTP does not interfere with normal Watson-Crick base-pairing and preserves the free3[prime]-hydroxyl group on the sugar, which is required for further elongation of the RNA after the analog is incorporated. The attachment of the fluorescent group utilizes a linker in which a sulfur atom is attached directly to the nucleotide base, and the first three bonds attached to the base are single bonds capable of free rotation.

Transcription properties of 5-SF-UTP

The 5-SF-UTP was tested as a substrate for transcription by the bacteriophage T7 RNA polymerase, the bacterial E.coli RNA polymerase and the eukaryotic yeast RNA polymerases I and III. These polymerases differ significantly in their subunit compositions and complexity. The bacteriophage RNA polymerase contains a single polypeptide which is capable of promoter recognition, elongation and specific termination of transcription. The yeast RNA polymerases, by contrast, contain many subunits, and very little is known as yet about the polymerase structures or their regulation. The RNA polymerase from E.coli falls between the phage and yeast polymerases in both subunit complexity and regulation by ancillary factors.

The photocrosslinking analog 5-APAS-UTP (Fig. 1, IV) functions well with both T7 and RNA polymerases. However, neither polymerase could incorporate two analogs in adjacent positions in the nascent RNA. We have shown previously that 5-APAS-UTP, which contains an azide group in approximately the same position as the fluorescein moiety in 5-SF-UTP, causes transcriptional pausing after incorporation of the first of two or more sequentially encoded UMP residues (28). Transcription through sequential UMPs requires the addition of low concentrations of UTP. We have found this to be true for most of the nucleotide analogs we have developed previously. It was therefore quite surprising that the 5-SF-UTP proved to be such an excellent substrate for E.coli RNA polymerase. Full-length RNA could be synthesized in the absence of any UTP, indicating that two analogs could be incorporated sequentially in the RNA. This is particularly interesting because the side chain on the base is larger in 5-SF-UTP than in 5-APAS-UTP, suggesting that side chain size alone does not determine the substrate properties of these nucleotide analogs.

In contrast, and more typical, neither yeast RNA polymerase III nor bacteriophage T7 RNA polymerase were able to synthesize full-length RNA unless a small amount of UTP was also present (Fig. 2). For T7 RNA polymerase, this may be related to the high level of abortive initiation observed with this enzyme. Once the enzyme has escaped the abortive stage and the transcription complex has matured to a stable elongation complex, the analog may be incorporated throughout the RNA. With related analogs, the inability to incorporate analog sequentially has been indicated by the appearance of multiple transcription pause sites on the DNA template when only low concentrations of UTP were added, corresponding to regions where several analogs are encoded. No such pauses were observed in these experiments (Fig. 2A, lane 11 versus 12).

The behavior of yeast RNA polymerase I was different than any of the other RNA polymerases. Not only could this enzyme apparently not incorporate analog into full-length RNA, but the analog inhibited transcription even when UTP was present. This suggests that the 5-SF-UTP may bind strongly to the enzyme active site, thereby blocking access of UTP and RNA synthesis. Alternatively, the analog might be incorporated into the RNA but act as an RNA chain terminator, thereby producing only very short oligonucleotides on this template. We are unable to distinguish between these two possibilities with the experiments presented here. Whatever the reason, 5-SF-UTP does not function well as a substrate for synthesis of long RNA by yeast RNA polymerase I.

Probing the environment of the RNA in E.coli elongation complexes: polarization data and molecular modeling

To probe the environment of the RNA as it is translocated from the active site of the polymerase to the surface of the enzyme and into the solvent, we assembled transcription complexes which contained the fluorescent analog at nucleotide position +21 in the RNA. Three different complexes were prepared, containing RNA either 21, 24 or 35 nt long (Fig. 4). In addition, RNA containing analog at position +21 was elongated to the end of the transcription unit to produce full-length RNA of 161 nt, which was released from the RNA polymerase and DNA. The fluorescence polarization of the analog in each of these complexes was determined after removal of unincorporated analog. We have also carried out molecular modeling to predict the low energy conformations for the analog in these RNAs, in both single-stranded RNA and as part of DNA-RNA hybrids. The results are represented in Figure 5, where the lowest energy conformations are shown (see Table 2 of the supplementary material).

When the analog was positioned at the 3[prime] end of the RNA (U-21) or 4 nt from the 3[prime] end of the RNA (C-24), its polarization was significantly greater than that of free analog (Fig. 4C). Further elongation of the RNA to place the analog 14 nt from the 3[prime] end of the RNA (35mer) produced a marked decrease in polarization, which was only slightly greater than that for analog in the released 161 nt RNA. We believe that these data are consistent with the transcription structures shown in Figure 4D, and our modeling data further support this interpretation. We believe that the greatly increased polarization in the U-21 and C-24 RNAs is due to the involvement of the RNA in a DNA-RNA hybrid. This hybrid would both constrain the movement of the RNA and limit the available space that the fluorescent group could occupy (Fig. 5). Our modeling data indicate that when the fluorescent probe is at the 3[prime] end of the RNA (U-21) or when it is moved further from the 3[prime] end of the RNA (C-24), and the RNA is part of a DNA-RNA hybrid, there are two low energy conformations for the analog. It may be seen from Figure 5 that, while moving between these two conformations, the fluorescent group is more free to rotate in the 21mer than it is in the 24mer. If the 24mer was not part of a DNA-RNA hybrid, as one model for the elongation complex suggests (24), then the analog would have significantly greater freedom to move, and thus, should have a lower polarization value than the 21mer. Our experimental data, however, show that there is an increase in polarization from the 21mer to the 24mer (Fig. 4C), consistent with the 24mer being base-paired to the DNA. When analog is in a single-stranded RNA (G-35, analog is 14 nt away from the 3[prime] end of the RNA), there are several low energy conformations predicted by the modeling of the representative stretch of RNA, of which only the three lowest are shown in Figure 5. Figure 5D shows that the analog in this oligonucleotide has a greater freedom of motion than in the DNA-RNA hybrid.

Our data are consistent with the model for the path of the nascent RNA in E.coli transcription complexes which is shown in Figure 4D. Analysis of several different transcription complexes indicate that ~8-10 nt at the 3[prime] end of the nascent RNA are part of a DNA-RNA hybrid with the template DNA and make contacts with both the [beta] and [beta][prime] subunits of the polymerase (27,30-35). Nucleotides at the 5[prime] end or 3[prime] end of the transcript also contact the [sigma] subunit when the RNA is very short (25,26). In the region 13-18 nt from the 3[prime] end of the RNA, transcripts are shielded from both enzymatic digestion and hybridization to complementary oligonucleotides (31,33,34), presumably through interaction with the RNA polymerase. This interaction has been proposed to involve a channel or tunnel through E.coli RNA polymerase which is ~8-10 Å wide (reviewed in 22). This tunnel may direct the RNA from the active site of the polymerase to the surface of the enzyme, and RNA may enter this tunnel at or near the point that it separates from the DNA template.

Using these parameters, a nucleotide within the RNA might be expected to exit the tunnel when it is ~13-18 nt from the 3[prime] end of the RNA. The crosslinking and fluorescent analogs probe almost identical regions, as shown in Figure 5A and B. Our crosslinking studies with 5-APAS-UTP suggest that such a tunnel would be composed of both the [beta] and [beta][prime] subunits, and that upon exiting the tunnel, the RNA would then encounter the [alpha] subunit of the polymerase. In the 35mer synthesized in these studies, the fluorescent probe is positioned 14 nt from the 3[prime] end of the RNA. Its low polarization value compared to that of the 21mer and 24mer RNA, and its only slightly greater value than the free 161mer RNA, suggest to us that if such a shielded RNA exit channel or tunnel exits, that this nucleotide has already exited, as we predict that nucleotides would have little freedom of rotation within the tunnel if its diameter is only ~8-10 Å.

Summary

We have described here a new fluorescent UTP analog which is an excellent substrate for a variety of RNA polymerases, including bacteriophage, bacterial and eukaryotic polymerases. We have utilized fluorescence polarization and molecular modeling data to evaluate the environment of the analog free in solution and as part of several different RNA molecules. By placing the analog at single positions in different RNA molecules, we have been able to evaluate the ability of the nucleotide to rotate freely, and thereby obtain information about the environment of the RNA. We have presented experiments which suggest that by carefully positioning this analog at increasing distances from the 3[prime] end of the RNA in active transcription complexes, it should be possible to determine when the RNA is part of a DNA-RNA hybrid, hindered by a protein binding pocket, or on the surface of the enzyme.

ACKNOWLEDGEMENTS

This work was supported by NIH grant RO1 GM47493, NSF grant MCB-9509132, and an award from the Oklahoma Center for the Advancement of Science and Technology (project number AR6060). Supplementary material is available at http://faculty-staff.ou.edu/Y/Elizabeth.Yuriev-1/NAR1/top.html or the Hanna laboratory (http://cheminfo.chem.ou.edu/faculty/mmh.html ).

See supplementary material available in NAR Online (2.63 MB PDF file).

REFERENCES

1. Hanna,M.M., Dissinger,S., Williams,B.D. and Colston,J.E. (1989) Biochemistry, 28, 5814-5820. MEDLINE Abstract

2. Liu,K. and Hanna,M.M. (1995) Proc. Natl Acad. Sci. USA, 92, 5012-5016. MEDLINE Abstract

3. Liu,K. and Hanna,M.M. (1995) J. Mol. Biol., 247, 547-558. MEDLINE Abstract

4. Liu,K., Zhang,Y., Severinov,K., Das,A. and Hanna,M.M. (1996) EMBO J., 15, 150-161. MEDLINE Abstract

5. Zhang,Y. and Hanna,M. (1994) J. Bacteriol., 176, 1787-1789. MEDLINE Abstract

6. He,B., Riggs,D.L. and Hanna,M.M. (1995) Nucleic Acids Res., 23, 1231-1238. MEDLINE Abstract

7. Studier,F.W. and Rosenberg,A.H. (1981) J. Mol. Biol., 153, 503-527. MEDLINE Abstract

8. Thayer,G.C. and Brosius,J. (1985) Mol. Gen. Genet., 199, 55-58. MEDLINE Abstract

9. Burgess,R.R. and Jendrisak,J.J. (1975) Biochemistry, 14, 4634-4638. MEDLINE Abstract

10. Gonzalez,N., Wiggs,J. and Chamberlin,M.J. (1977) Arch. Biochem. Biophys., 182, 404-408. MEDLINE Abstract

11. Hanna,M.M. (1989) Methods Enzymol., 180, 383-409. MEDLINE Abstract

12. Hanna,M.M., Zhang,Y.Y., Reidling,J.C., Thomas,M.J. and Jou,J. (1993) Nucleic Acids Res., 21, 2073-2079. MEDLINE Abstract

13. Burns,J.A., Butler,J.C., Moran,J. and Whitesides,G.M. (1991) J. Org. Chem., 56, 2648-2650.

14. Bardos,T.J. and Kalman,T.I. (1966) J. Pharm. Sci., 55, 606-609. MEDLINE Abstract

15. Hartig,P.R., Bertrand,N.J. and Sauer,K. (1977) Biochemistry, 16, 4275-4282. MEDLINE Abstract

16. Takashi,R. (1979) Biochemistry, 18, 5164-5169. MEDLINE Abstract

17. Riggs,D.L., Peterson,C.L., Wickham,J.Q., Miller,L.M., Crowell,J.A., Clarke,E.M. and Serge,J.-C. (1995) J. Biol. Chem., 270, 6205-6210. MEDLINE Abstract

18. Clarke,E.M., Peterson,C.L., Brainard,A.V. and Riggs,D.L. (1996)J. Biol. Chem., 271, 22189-22195. MEDLINE Abstract

19. Weiner,S.J., Kollman,P.A., Case,D.A., Singh,U.C., Ghio,C., Alagona,G., Profeta,S.J. and Weiner,P. (1984) J. Am. Chem. Soc., 10, 765-784.

20. von Hippel,P.H. (1998) Science, 281, 660-665. MEDLINE Abstract

21. Gelles,J. and Landick,R. (1998) Cell, 93, 13-16. MEDLINE Abstract

22. Ebright,R.H. (In Press) Cold Spring Harbor Symposium.

23. Nudler,E., Gusarov,I., Avetissova,E., Kozlov,M. and Goldfarb,A. (1998) Science, 281, 424-428. MEDLINE Abstract

24. Uptain,S.M., Kane,C.M. and Chamberlin,M.J. (1997) Annu. Rev. Biochem., 66, 117-172. MEDLINE Abstract

25. Hanna,M.M. and Meares,C.F. (1983) Biochemistry, 22, 3546-3551. MEDLINE Abstract

26. Bowser,C. and Hanna,M.M. (1991) J. Mol. Biol., 220, 227-239. MEDLINE Abstract

27. Hanna,M.M. and Meares,C.F. (1983) Proc. Natl Acad. Sci. USA, 80, 4238-4242. MEDLINE Abstract

28. Dissinger,S. and Hanna,M.M. (1990) J. Biol. Chem., 265, 7662-7668. MEDLINE Abstract

29. Hanna,M.M. (1996) Methods Enzymol., 274, 403-418. MEDLINE Abstract

30. Melnikova,A.F., Beabealashvilli,R. and Mirzabekov,A.D. (1978)Eur. J. Biochem., 84, 301-309. MEDLINE Abstract

31. Lee,D.N. and Landick,R. (1992) J. Mol. Biol., 228, 759-777. MEDLINE Abstract

32. Nudler,E., Mustaev,A., Lukhtanov,E. and Goldfarb,A. (1997) Cell, 89, 33-41. MEDLINE Abstract

33. Komissarova,N. and Kashlev,M. (1997) Proc. Natl Acad. Sci. USA, 94, 1755-1760. MEDLINE Abstract

34. Komissarova,N. and Kashlev,M. (1997) J. Biol. Chem., 272, 15329-15338. MEDLINE Abstract

35. Sidorenkov,I., Komissarova,N. and Kashlev,M. (1998) Mol. Cell, 2, 55-64.

*To whom correspondence should be addressed. Tel: +1 405 325 1678; Fax: +1 405 325 6111; Email: mhanna@chemdept.chem.ou.edu

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 11 Feb 1999
Copyright©Oxford University Press, 1999.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
C. Costas, E. Yuriev, K. L. Meyer, T. S. Guion, and M. M. Hanna
RNA-protein crosslinking to AMP residues at internal positions in RNA with a new photocrosslinking ATP analog
Nucleic Acids Res., May 1, 2000; 28(9): 1849 - 1858.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (237K) Freely available
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (6)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Hanna, M. M.
Right arrow Articles by Riggs, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hanna, M. M.
Right arrow Articles by Riggs, D. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?