FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2004453424> ?p ?o ?g. }

• W2004453424 endingPage "2028" @default.
• W2004453424 startingPage "2019" @default.
• W2004453424 abstract "The RegB endoribonuclease participates in the bacteriophage T4 life cycle by favoring early messenger RNA breakdown. RegB specifically cleaves GGAG sequences found in intergenic regions, mainly in translation initiation sites. Its activity is very low but can be enhanced up to 100-fold by the ribosomal 30 S subunit or by ribosomal protein S1. RegB has no significant sequence homology to any known protein. Here we used NMR to solve the structure of RegB and map its interactions with two RNA substrates. We also generated a collection of mutants affected in RegB function. Our results show that, despite the absence of any sequence homology, RegB has structural similarities with two Escherichia coli ribonucleases involved in mRNA inactivation on translating ribosomes: YoeB and RelE. Although these ribonucleases have different catalytic sites, we propose that RegB is a new member of the RelE/YoeB structural and functional family of ribonucleases specialized in mRNA inactivation within the ribosome. The RegB endoribonuclease participates in the bacteriophage T4 life cycle by favoring early messenger RNA breakdown. RegB specifically cleaves GGAG sequences found in intergenic regions, mainly in translation initiation sites. Its activity is very low but can be enhanced up to 100-fold by the ribosomal 30 S subunit or by ribosomal protein S1. RegB has no significant sequence homology to any known protein. Here we used NMR to solve the structure of RegB and map its interactions with two RNA substrates. We also generated a collection of mutants affected in RegB function. Our results show that, despite the absence of any sequence homology, RegB has structural similarities with two Escherichia coli ribonucleases involved in mRNA inactivation on translating ribosomes: YoeB and RelE. Although these ribonucleases have different catalytic sites, we propose that RegB is a new member of the RelE/YoeB structural and functional family of ribonucleases specialized in mRNA inactivation within the ribosome. Control of messenger RNA decay is a key factor in the regulation of gene expression. Indeed, the expression level of a protein is strongly influenced by the concentration of its mRNA, which, in turn, depends on its synthesis and degradation rates (1Wilusz C.J. Wilusz J. Trends Genet. 2004; 20: 491-497Abstract Full Text Full Text PDF PubMed Scopus (216) Google Scholar). Perturbation of mRNA stability can lead to dramatic effects on cell physiology and in some cases can induce cancer (2Audic Y. Hartley R.S. Biol. Cell. 2004; 96: 479-498Crossref PubMed Scopus (179) Google Scholar). However, the way the cell controls the lifetime of its mRNA is still a poorly understood aspect in the biology of both prokaryotic and eukaryotic organisms. In Escherichia coli, it is generally agreed that the initiation of mRNA degradation is induced by an endoribonucleolytic cleavage mediated by ribonucleases E and G with the occasional involvement of RNases III and P (3Kushner S.R. J. Bacteriol. 2002; 184: 4658-4665Crossref PubMed Scopus (199) Google Scholar, 4Deutscher M.P. Nucleic Acids Res. 2006; 34: 659-666Crossref PubMed Scopus (315) Google Scholar). These ribonucleases all generate 5′-phosphate RNA fragments (RNases E, G, and III are metallo-enzymes; RNase P is a ribozyme), have broad specificity, and are involved in other cellular functions. RNases E and G, for example, participate in the maturation of 9 S and 16 S rRNA (5Li Z. Pandit S. Deutscher M.P. EMBO J. 1999; 18: 2878-2885Crossref PubMed Scopus (237) Google Scholar), and RNase E is necessary for tRNA processing (6Li Z. Deutscher M.P. RNA. 2002; 8: 97-109Crossref PubMed Scopus (165) Google Scholar).Ribonuclease RegB is radically different from the ribonucleases described above. RegB is involved in the control of the bacteriophage T4 multiplication cycle. It is produced in the early stage of viral infection and facilitates the transition between the early and middle phases of the viral cycle by inactivating early mRNAs (7Uzan M. Favre R. Brody E. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8885-8889Crossref PubMed Scopus (81) Google Scholar). RegB cleaves its substrates with high specificity in the middle of the GGAG motif found in the ribosome-binding site of most prokaryotic mRNAs (Shine-Dalgarno sequence). Strikingly, it remains active throughout the entire lytic cycle but does not act on middle or late mRNA (8Sanson B. Hu R.-M Troitskayadagger E. Mathy N. Uzan M. J. Mol. Biol. 2000; 297: 1063-1074Crossref PubMed Scopus (33) Google Scholar). In contrast with RNases E, G, III, and P, RegB is only involved in this task and is certainly one of the most specific ribonucleases described to date. It only cleaves GGAG, or to a lesser extent and with reduced efficiency, GGAU sequences (7Uzan M. Favre R. Brody E. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8885-8889Crossref PubMed Scopus (81) Google Scholar, 9Ruckman J. Parma D. Tuerk C. Hall D.H. Gold L. New Biol. 1989; 1: 54-56PubMed Google Scholar). Moreover, it does not cleave all such sequences: those present in coding sequences and in middle or late T4 mRNAs are generally resistant (10Sanson B. Uzan M. FEMS Microbiol. Rev. 1995; 17: 141-150Crossref PubMed Google Scholar). By transforming E. coli cells with two plasmids, one coding for RegB, the other for a natural T4 mRNA, it was demonstrated that the pattern of sensitivity observed in vivo can be reproduced in the absence of any other T4 factor, suggesting that additional cleavage determinants are intrinsic to the RNA. In the absence of evidence for additional sequence determinants, we postulated that RegB may recognize a particular RNA structure, and indeed have shown that a particular secondary structure correlates with RNA susceptibility to cleavage (11Lebars I. Hu R.-M Lallemand J.-Y. Uzan M. Bontems F. J. Biol. Chem. 2001; 276: 13264-13272Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar). We have also shown that RegB acts as a transphosphorylase, producing 3′-phosphate oligoribonucleotides (7Uzan M. Favre R. Brody E. Proc. Natl. Acad. Sci. U. S. A. 1988; 85: 8885-8889Crossref PubMed Scopus (81) Google Scholar, 12Saïda F. Uzan M. Bontems F. Nucleic Acids Res. 2003; 31: 2751-2758Crossref PubMed Scopus (23) Google Scholar). Finally, RegB activity as measured in vitro is very low: ∼10-3 mol of substrate are cleaved per mole of enzyme per minute, but this activity is enhanced by a factor up to 100 in the presence of the ribosomal subunit 30 S or of the ribosomal protein S1 (13Ruckman J. Ringquist S Brody E Gold L. J. Biol. Chem. 1994; 269: 26655-26662Abstract Full Text PDF PubMed Google Scholar, 11Lebars I. Hu R.-M Lallemand J.-Y. Uzan M. Bontems F. J. Biol. Chem. 2001; 276: 13264-13272Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar).These particular features of RegB, especially the involvement of ribosome components in its activity, motivated us to determine its three-dimensional structure and to study its mechanism of action using a combination of NMR spectroscopy and biochemical methods. RegB is a highly toxic protein. It cannot be overproduced in E. coli for structural purposes, especially when isotopic enrichment is required (14Saïda F. Odaert B. Uzan M. Bontems F. Protein Expr. Purif. 2004; 34: 158-165Crossref PubMed Scopus (7) Google Scholar). We overcame this obstacle by isolating a non-toxic point mutant (RegB-H48A) that retains RegB structural integrity and ability to bind substrates (12Saïda F. Uzan M. Bontems F. Nucleic Acids Res. 2003; 31: 2751-2758Crossref PubMed Scopus (23) Google Scholar). Here, we report the three-dimensional structure of RegB and present a map of the interaction sites between the enzyme and two different RNA substrates. We also took advantage of RegB toxicity to identify, by random and site-directed mutagenesis, residues involved in RNA binding or processing. Our results show that, despite the absence of any sequence similarities and a different organization of the active site residues, RegB is structurally related to ribonucleases YoeB and RelE. Interestingly, YoeB and RelE are involved in mRNA inactivation, after nutritional stress, and act on translating messengers. Accordingly, we propose the existence of a new specialized family of transphosphorylases involved in messenger inactivation within the ribosome.EXPERIMENTAL PROCEDURESProtein Expression and Purification—Typically, a 50-ml overnight culture of the E. coli strain BL21(DE3)-pLysS, transformed with plasmid pEH48A (14Saïda F. Odaert B. Uzan M. Bontems F. Protein Expr. Purif. 2004; 34: 158-165Crossref PubMed Scopus (7) Google Scholar), was used to inoculate 1 liter of M9 minimal medium (Na2HPO4, 6 g·liter-1 ;KH2PO4, 3 g·liter-1; NaCl, 0.5 g·liter-1; MgSO4, 1 mm; CaCl2, 1.10-4 m) supplemented with 1.0 g·liter-115NH4Cl and 4 g·liter-1 glucose or 2.0 g·liter-1 [13C]glucose, leading to uniformly labeled 15N or 15N/13C samples. Protein expression was induced at A600 = 0.6 using 1.0 mm isopropyl-β-d-thiogalactopyranoside. Cells were harvested 3 h later and disrupted by sonication, and the protein was purified on a Talon resin (Clontech) following the manufacturer's recommended protocol. The protein was then concentrated to ∼0.7 mm, and the elution buffer was exchanged against 50 mm sodium citrate (pH 6.0), 300 mm NaCl, 2.0 mm dithiothreitol using Micro Bio-spin chromatography columns (Bio-Rad).RNA Synthesis—The S22loop and S26 oligoribonucleotides were synthesized chemically on an Amersham Biosciences LKB Gene Assembler Plus apparatus using Amersham Biosciences polystyrene beads for primer support and phenoxyacetyl β-RNA phosphoramidites. The protocol was adapted from the one used for DNA synthesis, the main differences being the increase in coupling time (20 min) and the use of 5-ethylthio-1H-tetrazol as activator (15Snoussi K. Nonin-Lecomte S. Leroy J.L. J. Mol. Biol. 2001; 309: 139-153Crossref PubMed Scopus (66) Google Scholar). The terminal 5′-O-dimethoxytrityl group was removed on the synthesizer. The oligoribonucleotides were cleaved from the support and unprotected according to Leroy and co-workers (15Snoussi K. Nonin-Lecomte S. Leroy J.L. J. Mol. Biol. 2001; 309: 139-153Crossref PubMed Scopus (66) Google Scholar). Full-length product final yield was >95% in both cases. Accordingly, the two oligoribonucleotides were used without purification.NMR Experiments—All NMR experiments were carried out on a Bruker DRX600 spectrometer equipped with a triple resonance Z-gradient cryoprobe. They were processed using XWINNMR 3.0 (Bruker) and analyzed with Sparky software (University of California San Francisco, Thomas L. Goddard). 1H and 13C chemical shifts were referenced to 3-(trimethylsilyl)-propionic acid-D4. 15N chemical shifts were indirectly referenced from the γ15N/γ1H ratio (16Wishart D.S. Bigam C.G. Yao J. Abildgaard F. Dyson H.J. Oldfield E. Markley J.L. Sykes B.D. J. Biomol. NMR. 1995; 6: 135-140Crossref PubMed Scopus (2054) Google Scholar). All spectra were measured at 305 K. The backbone assignment was performed on samples of 13C/15N-labeled protein by recording HNCO, HNCA, HN(CO)CA, HNCACB, CBCA(CO)NH, HNHA, and HBHA(CBCACO)NH triple resonance experiments. Most aliphatic side-chain 1H and 13C frequencies were assigned using an HCCH-TOCSY 5The abbreviations used are: TOCSY, total correlation spectroscopy; NOE, nuclear Overhauser effect; NOESY, nuclear Overhauser effect spectroscopy; HSQC, heteronuclear single quantum coherence. spectrum recorded on an identical sample dissolved in D2O. A similar experiment was also recorded to try to determine the aromatic 1H and 13C frequencies, but the signal-to-noise ratio was too low to allow its exploitation. Thus, most aromatic 1H and 13C frequencies were identified by using an aromatic 13C-NOESY-HSQC, recorded in D2O on a 13C/15 N sample and two TOCSY and NOESY spectra recorded in D2O on a 15N-labeled sample. The backbone assignment was further verified, and RegB secondary structure and topology were identified through the analysis of the NOE correlations observed in two 15N-NOESY-HSQC experiments (120- and 60-ms mixing time, recorded on an 15N-labeled sample), an aliphatic 13C-NOESY-HSQC experiment (60-ms mixing time, recorded on a 13C/15N-labeled sample dissolved in D2O) and two two-dimensional-NOESY experiments (120- and 60-ms mixing time, recorded on the 15N-labeled sample after H2O to D2O exchange). The NOE correlations used to calculate the structure were picked from the 60-ms 15N-NOESY-HSQC spectrum, from two 60-ms aliphatic 13C-NOESY-HSQC spectra recorded in D2O and H2O and from the 60-ms two-dimensional-NOESY spectrum.RNA binding studies were performed on a 0.2 mm15N-labeled RegB sample in which aliquots of lyophilized RNA were dissolved. Prior to lyophilization, the RNA was dialyzed against pure water and pH neutralized. Each titration point was done by removing the NMR sample from the tube, mixing it with lyophilized RNA and returning it to the tube. Eight 1H-15N-HSQC were recorded for each series, corresponding to RNA/protein ratios of 0, 0.25, 0.5, 0.75, 1, 1.5, 2, and 3. All recording parameters were kept rigorously identical; the only modifications concerned probe tuning and field shimming. All spectra were processed identically.Structure Calculation—First, the 15N- and 13C-NOESY-HSQC spectra were analyzed manually to determine the secondary structure elements and the topology of the RegB structure (see supplemental Fig. S1). The RegB structure was then calculated with the INCA software (17Savarin P. Zinn-Justin S. Gilquin B. J. Biomol. NMR. 2001; 19: 49-62Crossref PubMed Scopus (24) Google Scholar) that used mainly unassigned NOE cross-peaks to perform the structure determination. Typically, the software carried out 22 calculation cycles, each corresponding to an automatic assignment, a structure calculation, and an analysis step. The assignment step converts each NOE cross-peak to a list of possible constraints based on the distances observed in the structures calculated in the preceding cycle. This list is reduced at each cycle, ending, as often as possible, with a single possibility. The calculation step carries out the calculation of 500 structures from the constraint lists by simulated annealing. The 20 best are selected during the analysis step. The program was fed with two lists of unassigned NOE correlations picked from the 60-ms 13C- and 15N-NOE-HSQC spectra, with the two corresponding lists of 1H, 13C, and 15N chemical shifts, with a list of distance constraints derived from the identified hydrogen bonds and from manually assigned NOEs, and finally with a list of phi and psi dihedral angle constraints derived from the TALOS analysis of the backbone chemical shifts (18Cornilescu G. Delaglio F. Bax A. J. Biomol. NMR. 1999; 13: 289-302Crossref PubMed Scopus (2730) Google Scholar). In the first runs, the assigned distance constraints were derived from the NOE characteristic of the secondary structure elements and of the topology, from several aromatic proton-aromatic proton and aromatic proton-aliphatic proton NOEs and finally from a small number of hydrogen bonds (identified from an H2O-D2O exchange experiment). Similarly, only the phi and psi angles of residues belonging to the unambiguously identified helix and strands were constrained. To favor the convergence of the process (the number of correct structures over the total number of calculated structures), we added, in the last runs, several constraints corresponding to the hydrogen bonds clearly formed in the secondary structure elements and to phi-psi angles predicted by TALOS and compatible with the structures obtained in the previous runs. The coordinates of 15 structures have been deposited at the Protein Data Bank with accession code 2HX6.Random Mutagenesis—Error-prone PCR-mediated mutagenesis was performed using the Taq polymerase in the presence of a balanced desoxyribonucleoside pool, but witha1to10 Mn2+/Mg2+ ratio. 300 pm DNA fragments were amplified by 22 PCR cycles using 3 Taq units in the presence of 0.4 pm oligonucleotides, 0.2 mm each dNTP, 3 mm Mg2+, and 0.3 mm Mn2+. The mutagenesis protocol was independently applied to the 5′-terminal (nucleotides 1-240 and amino-acids 1-80) and 3′-terminal (nucleotides 241-459 and amino-acids 81-153) regions of the gene. After digestion by NdeI and EcoRI (5′-terminal region) or by EcoRI and XhoI (3′-terminal region), the fragments were ligated, respectively, into a pET15b derivative containing the wild-type 3′-terminal region of regB or a pET28b derivative containing the wild-type 5′-terminal region of regB. The ligation mixture was directly used to transform an XL1-Blue E. coli strain, and the viable clones were selected and amplified. Protein expression was induced in the XL1-Blue strain with a λCE6 phage (Stratagene) following the manufacturer's protocol. The genes yielding full-length proteins were subsequently sequenced.Site-directed Mutagenesis—Seven regB point mutants (R11Q, E19A, E19V, E21V, D51G, R56L, and E105A) were obtained from the pARNU2 plasmid, with the QuikChange II XL site-directed mutagenesis kit and by using the manufacturer's (Stratagene) recommended protocol. After the PCR, the mutated plasmid was amplified in the XL10-Gold strain provided in the kit and sequenced. An alteration of the enzyme toxicity was then tested by transforming a BL21(DE3) strain (for which the wild-type pARNU2 plasmid is lethal, even in the absence of induction) with 6 ng of each mutated plasmid except E19A and E19V, for which 60 ng was used. The ability of the transformants to overproduce the protein was verified, and minipreps were analyzed by sequencing to check the absence of secondary mutations.RESULTSChoice of the Mutant Protein and the RNA Substrates—RegB is highly toxic to E. coli, preventing its overproduction from standard expression vectors (19Saïda F. Uzan M. Lallemand J.-Y. Bontems F. Biotechnol. Prog. 2003; 19: 727-733Crossref PubMed Scopus (6) Google Scholar, 20Uzan M. Methods Enzymol. 2001; 342: 467-480Crossref PubMed Scopus (17) Google Scholar). We assumed that this toxicity was likely due to its ribonuclease activity, the GGAG motif being present in ∼30% of the Shine-Dalgarno sites of E. coli mRNAs (21Barrick D. Villanueba K. Childs J. Kalil R. Schneider T.D. Lawrence C.E. Gold L. Stormo G.D. Nucleic Acids Res. 1994; 22: 1287-1295Crossref PubMed Scopus (132) Google Scholar). We also assumed that RegB catalytic mechanism involves at least one histidine residue. Accordingly, as previously reported, we mutated each RegB histidine residue to alanine and obtained two non-toxic mutants, H48A and H68A (12Saïda F. Uzan M. Bontems F. Nucleic Acids Res. 2003; 31: 2751-2758Crossref PubMed Scopus (23) Google Scholar). Comparison of the HSQC spectra of these two mutants with that of another independently identified inactive mutant (R52L) indicates that these proteins have the same three-dimensional structure. In addition, the H86A mutant conserves 20% of the wild-type activity, and the H48A and H68A (but not R52L) were shown to bind RegB substrates with the same affinity as the wild-type protein. All this strongly suggests that the three mutants are good structural models of the wild-type protein (12Saïda F. Uzan M. Bontems F. Nucleic Acids Res. 2003; 31: 2751-2758Crossref PubMed Scopus (23) Google Scholar). We found that the H68A mutant was difficult to overproduce due to its residual activity but obtained a good production yield (∼20 mg·liter-1) for the H48A mutant (14Saïda F. Odaert B. Uzan M. Bontems F. Protein Expr. Purif. 2004; 34: 158-165Crossref PubMed Scopus (7) Google Scholar), which was thus chosen for further analysis.All the T4 Shine-Dalgarno sequences cleaved by RegB have now been identified, 6M. Uzan, unpublished data. but the RegB recognition site remains unknown. All natural substrates tested to date require the presence of the S1 ribosomal protein for efficient cleavage. On the other hand, we already studied several non-natural RegB substrates from a SELEX analysis (11Lebars I. Hu R.-M Lallemand J.-Y. Uzan M. Bontems F. J. Biol. Chem. 2001; 276: 13264-13272Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar). We decided to use two of them, namely the S26 (CGUGGAGACAGACCACACCACG) and the S22loop (GGUGCGAGAAAACGGAGCACC). Both are cleaved by RegB alone but with different efficiencies. The S22loop RNA is cleaved well in the absence of S1 (turnover 20 × 10-3 min-1), whereas S26 requires S1 to reach the same level (8 × 10-3 min-1 in the absence of S1 and 20 × 10-3 min-1 in its presence). The secondary structures of these two RNAs are known (11Lebars I. Hu R.-M Lallemand J.-Y. Uzan M. Bontems F. J. Biol. Chem. 2001; 276: 13264-13272Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar), and their molecular masses (8 kDa) are low compared with that of the protein (18 kDa). Thus, even in the event of slow ligand exchange, we anticipated that the quality of the protein NMR spectra would not be significantly affected by substrate binding.RegB Structure—The superposition of 15 structures obtained at the end of the last computation run is presented in Fig. 1. These structures present no systematic distance violations larger than 0.5 Å or systematic dihedral violations larger than 5°. The covalent geometry is correct (Table 1). 93% of the residues are in the most favorable and additional allowed regions of the Ramachandran plot. The core of the protein, formed by the first four β-strands and the three α-helices, is correctly defined (root mean square deviation: 0.72 ± 0.11 Å). The loop regions and the last β-strand are more dispersed, but this is mainly due to the fact that we were not able to collect enough data on them (see supplemental Fig. S1 for details about the structure calculation process).TABLE 1NMR and refinement statisticsNMR constraintsDistance constraintsA priori assigned NOEaNumber of constraints assigned by the user at the beginning of the calculation process.282A posteriori unambiguous NOEbNumber of constraints assigned by the program, at the end of the calculation process, from the peaks identified in the 15N and 13C NOE-HSQC experiments.1162A posteriori ambiguous NOEcNumber of constraints still ambiguous at the end of the calculation process.311Hydrogen bonds59Distance repartitionIntra-residual634Sequential359Medium range (i - j ≤ 4)295Long range215Dihedral angle constraintsPhi93Psi93Structure statisticsdMean (maximal - minimal) values.ViolationseNone of the observed violations are systematic. The violations are probably due to a lack of convergence of the calculation due to the relatively small number of constraints we were able to determine in certain regions (see supplemental Fig. S1 for a discussion of the calculation process).Number of distance violations >0.5 Å2.5 (1-4)Maximal distance violation (Å)0.6 (0.51-0.80)Number of dihedral violations > 10°0.6 (0-3)Maximal dihedral violation (°)10.0 (5.8-15.0)Deviation from idealized geometryBond lengths, root mean square deviation (Å)0.019 (0.018-0.019)Bond angles, root mean square deviation (°)4.0 (3.3-4.2)Improper angles, root mean square deviation (°)3.7 (3.3-4.1)a Number of constraints assigned by the user at the beginning of the calculation process.b Number of constraints assigned by the program, at the end of the calculation process, from the peaks identified in the 15N and 13C NOE-HSQC experiments.c Number of constraints still ambiguous at the end of the calculation process.d Mean (maximal - minimal) values.e None of the observed violations are systematic. The violations are probably due to a lack of convergence of the calculation due to the relatively small number of constraints we were able to determine in certain regions (see supplemental Fig. S1 for a discussion of the calculation process). Open table in a new tab As shown in Fig. 1, the three α-helices and the β-sheet are amphipathic. In particular, one side of the β-sheet, except for one residue (Arg132), is formed only of hydrophobic residues. The three helices are associated through a series of hydrophobic contacts. A first cluster involves Leu14, Phe18, Phe22, and Ile25 (helix 1); Leu50 and Ile54 (helix 2); and Val63, Phe64, and Phe67 (helix 3). A second cluster is formed of Ile29, Ala32, and Ala35 (helix 1) and Val74 and Leu81 (helix 3). These helices are packed against the hydrophobic side of the β-sheet. The first helix interacts mainly with the first strand through interactions between Leu25, Leu29, Ala32, and Ala35 (helix 1) and Phe41, Leu63, and Tyr65 (strand 1), the second helix is associated to the fourth strand through contacts between Leu49 (helix 2) and Ala134 (strand 4), and the third helix is associated with the second, third, and fourth strands through contacts between Leu66, Ile70, Val77, and Phe80 (helix 3); Leu104 and Ile106 (strand 2); Leu111, Leu113, Phe115, and Val117 (strand 3); and Leu129, Cys131, and Ala134 (strand 4). In contrast, the exposed surface is mainly hydrophilic, with the noticeable exception of the Trp112, Cys118, Cys131, Met133, and Ile135 region of the β-sheet. Most of the surface residues are charged, resulting in a large negative potential surface on the helical side of the protein and a completely positive potential surface on the opposite side (Fig. 1).RNA Binding Study—The 1H-15N HSQC experiment provides a fingerprint of the protein structure, because any structural or environmental change will modify the spectrum. Fig. 2 displays zooms of superposed HSQC spectra of RegB at different RNA/protein ratios. The left and right panels correspond to the RegB/S26 and RegB/S22 titration, respectively. Side chains Asn and Gln NH2 correlations are identified by a line (upper panels).FIGURE 2Analysis of the interactions between RegB and two substrates. The four panels represent zooms on superimposed HSQC recorded at different RNA/protein ratios. The left (A and B) and right (C and D) panels correspond to the RegB/S26 and RegB/S22loop titrations, respectively. The Asn and Gln NH2 side-chain correlations are identified by lines (A and C). The reference experiment (RNA/protein ratio of zero) is in red. The last experiment (RNA/protein ratio of three) is in green. The interpretation of the observed frequency and intensity variations is detailed under “Results.”View Large Image Figure ViewerDownload Hi-res image Download (PPT)Clearly, many peaks (e.g. Arg56, Ser82, Asn12-NH2, and Asn137-NH2 (upper panels) or Leu75 and Phe115 (lower panels)) are invariant in all experiments, and their intensities in the reference (in red) or in the last titration map (in green) appear similar, although they vary by 1.5- to 2-fold. Others peaks show moderate (Phe41, Asn78-NH2, Phe18, Cys118, and Cys131) or large (Gln47-NH2, Asn121-NH2, His42, and Leu43) chemical shift changes as substrate concentrations increase. These changes are correlated with larger variations of intensity (3- to 4-fold), particularly notable for His42 and Leu43 (lower panels) and Ser46 (upper panels), for example. Interestingly, all modifications appear very similar in the RegB/S26 and RegB/S22 titrations. The good superposition of most peaks shows that all spectra were actually recorded in reproducible conditions. The intensity variation of these peaks is probably due to the system size increase (∼18 kDa for the free protein, 25 kDa for the complex). This is supported by the observation that intensity changes are generally less pronounced for the side chains. However, we could not exclude the possibility that RNA addition induces partial aggregation of RegB. The continuous chemical shift variation indicates that the exchange frequency between protein free and bound states is larger than the difference in frequencies characterizing the two states: 2πνex ≫ |ωfree - ωbound|. The large intensity variation corresponds to the case where |ωfree - ωbound| comes close to 2πνex, i.e. to the peaks presenting the largest frequency variation. Accordingly, considering that the largest peak shift is ∼0.25 ppm (Asn121), we calculate that the exchange frequency is larger than 150 Hz and the substrate-enzyme complex lifetime smaller than 7 ms. We can also fit the shift function of the protein/RNA ratio to a simple equilibrium model E + S ↔ ES, giving rise to an equilibrium constant of ∼30 μm.The chemical shift differences between the first (reference) and the third (RNA/RegB ratio of 0.5) 1H-15N-HSQC spectra are reported for both substrates in Fig. 3 and indicate that the perturbed amino acids are clustered in three main regions. The first region (Asn4-Ile29) corresponds to the N terminus and the beginning of the first α-helix. The second cluster is formed by the first β-strand and the second α-helix. The third cluster is composed of the fourth β-strand and the preceding loop. As previously shown, these elements are close in space. The first and fourth strands are connected in the β-sheet. The first and second helices are associated and make contact with the first and fourth strands. In addition, the loop between the first helix and the first strand is in the vicinity of the loop preceding the fourth strand (Fig. 1). The figure also shows that the perturbation pattern is globally similar for the two substrates. Some details are even well conserved, such as the succession of perturbed (Ile25, Asn27, and Ile29) and non-perturbed (Asn26 and Glu28) residues at the end of the first α-helix, and the involvement of Ile58 and Glu60. However, there are also some variations: the relative shift intensities of Leu43 and Ser46 and the involvement of Asn78 are different. This suggests that, although the two substrates are bound similarly, the details of the intera" @default.
• W2004453424 created "2016-06-24" @default.
• W2004453424 creator A5007057583 @default.
• W2004453424 creator A5013104317 @default.
• W2004453424 creator A5023903527 @default.
• W2004453424 creator A5027009993 @default.
• W2004453424 creator A5043668210 @default.
• W2004453424 creator A5044707698 @default.
• W2004453424 creator A5048706352 @default.
• W2004453424 creator A5072263380 @default.
• W2004453424 creator A5088065353 @default.
• W2004453424 date "2007-01-01" @default.
• W2004453424 modified "2023-09-30" @default.
• W2004453424 title "Structural and Functional Studies of RegB, a New Member of a Family of Sequence-specific Ribonucleases Involved in mRNA Inactivation on the Ribosome" @default.
• W2004453424 cites W1487571658 @default.
• W2004453424 cites W1498454063 @default.
• W2004453424 cites W1589295940 @default.
• W2004453424 cites W1604915411 @default.
• W2004453424 cites W1606205321 @default.
• W2004453424 cites W1820567505 @default.
• W2004453424 cites W183590699 @default.
• W2004453424 cites W1914760274 @default.
• W2004453424 cites W1964051028 @default.
• W2004453424 cites W1966020270 @default.
• W2004453424 cites W1970491226 @default.
• W2004453424 cites W1975012965 @default.
• W2004453424 cites W1989869773 @default.
• W2004453424 cites W1997976504 @default.
• W2004453424 cites W2005087717 @default.
• W2004453424 cites W2006785307 @default.
• W2004453424 cites W2006923455 @default.
• W2004453424 cites W2014933415 @default.
• W2004453424 cites W2023925347 @default.
• W2004453424 cites W2028602838 @default.
• W2004453424 cites W2030361819 @default.
• W2004453424 cites W2040114290 @default.
• W2004453424 cites W2046456420 @default.
• W2004453424 cites W2069602336 @default.
• W2004453424 cites W2075875645 @default.
• W2004453424 cites W2079317055 @default.
• W2004453424 cites W2090908102 @default.
• W2004453424 cites W2093814623 @default.
• W2004453424 cites W2101058939 @default.
• W2004453424 cites W2106315897 @default.
• W2004453424 cites W2113047904 @default.
• W2004453424 cites W2122386566 @default.
• W2004453424 cites W2136124870 @default.
• W2004453424 cites W2138808082 @default.
• W2004453424 cites W2141365940 @default.
• W2004453424 cites W2160166027 @default.
• W2004453424 cites W2164512635 @default.
• W2004453424 cites W2165182519 @default.
• W2004453424 cites W2167420567 @default.
• W2004453424 cites W2749830096 @default.
• W2004453424 cites W342219782 @default.
• W2004453424 doi "https://doi.org/10.1074/jbc.m608271200" @default.
• W2004453424 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/17046813" @default.
• W2004453424 hasPublicationYear "2007" @default.
• W2004453424 type Work @default.
• W2004453424 sameAs 2004453424 @default.
• W2004453424 citedByCount "18" @default.
• W2004453424 countsByYear W20044534242012 @default.
• W2004453424 countsByYear W20044534242013 @default.
• W2004453424 countsByYear W20044534242015 @default.
• W2004453424 countsByYear W20044534242019 @default.
• W2004453424 crossrefType "journal-article" @default.
• W2004453424 hasAuthorship W2004453424A5007057583 @default.
• W2004453424 hasAuthorship W2004453424A5013104317 @default.
• W2004453424 hasAuthorship W2004453424A5023903527 @default.
• W2004453424 hasAuthorship W2004453424A5027009993 @default.
• W2004453424 hasAuthorship W2004453424A5043668210 @default.
• W2004453424 hasAuthorship W2004453424A5044707698 @default.
• W2004453424 hasAuthorship W2004453424A5048706352 @default.
• W2004453424 hasAuthorship W2004453424A5072263380 @default.
• W2004453424 hasAuthorship W2004453424A5088065353 @default.
• W2004453424 hasBestOaLocation W20044534241 @default.
• W2004453424 hasConcept C104317684 @default.
• W2004453424 hasConcept C105580179 @default.
• W2004453424 hasConcept C149364088 @default.
• W2004453424 hasConcept C185592680 @default.
• W2004453424 hasConcept C2778112365 @default.
• W2004453424 hasConcept C54355233 @default.
• W2004453424 hasConcept C55493867 @default.
• W2004453424 hasConcept C67705224 @default.
• W2004453424 hasConcept C70721500 @default.
• W2004453424 hasConcept C86803240 @default.
• W2004453424 hasConcept C88478588 @default.
• W2004453424 hasConceptScore W2004453424C104317684 @default.
• W2004453424 hasConceptScore W2004453424C105580179 @default.
• W2004453424 hasConceptScore W2004453424C149364088 @default.
• W2004453424 hasConceptScore W2004453424C185592680 @default.
• W2004453424 hasConceptScore W2004453424C2778112365 @default.
• W2004453424 hasConceptScore W2004453424C54355233 @default.
• W2004453424 hasConceptScore W2004453424C55493867 @default.
• W2004453424 hasConceptScore W2004453424C67705224 @default.
• W2004453424 hasConceptScore W2004453424C70721500 @default.
• W2004453424 hasConceptScore W2004453424C86803240 @default.
• W2004453424 hasConceptScore W2004453424C88478588 @default.

• next