Journal of Biological Chemistry · 2006 · 30 citations · 22 references
Life SciencesGeneticsMolecular BiologyMicrobial GeneticsRna BiologyStructural BiologyRibosome FunctionRna StructureTranslation In EukaryotesTranslation In ProkaryotesTranslation RegulationTranslational RegulationBacterial Ribosome FunctionEscherichia ColiHairpin MotifsPseudoknot StructureRational DesignRibosome Biogenesis In EukaryotesRibosome Biogenesis In ProkaryotesRna Translation Control
tmRNA rescues stalled ribosomes in eubacteria by forcing the ribosome to abandon its mRNA template and resume translation with tmRNA itself as a template. Pseudoknot 1 (pk1), immediately upstream of this coding region in tmRNA, is a structural element that is considered essential for tmRNA function based on the analysis of pk1 mutants in vitro. pk1 binds near the ribosomal decoding site and may make base-specific contacts with tmRNA ligands. To study pk1 structure and function in vivo, we have developed a genetic selection that ties the life of Escherichia coli cells to tmRNA activity. Mutation of pk1 at 20% per base and selection for tmRNA activity yielded sequences that retain the same pseudoknot fold. In contrast, selection of active mutants from 106 completely random sequences identified hairpin structures that functionally replace pk1. Rational design of a hairpin with increased stability using an unrelated sequence yielded a tmRNA mutant with nearly wild-type activity. We conclude that the role of pk1 in tmRNA function is purely structural and that it can be replaced with a variety of hairpin structures. Our results demonstrate that in the study of functional RNAs, the inactivity of a mutant designed to destroy a given structure should not be interpreted as proof that the structure is necessary for RNA function. Such mutations may only destabilize a global fold that could be formed equally well by an entirely different, stable structure. tmRNA rescues stalled ribosomes in eubacteria by forcing the ribosome to abandon its mRNA template and resume translation with tmRNA itself as a template. Pseudoknot 1 (pk1), immediately upstream of this coding region in tmRNA, is a structural element that is considered essential for tmRNA function based on the analysis of pk1 mutants in vitro. pk1 binds near the ribosomal decoding site and may make base-specific contacts with tmRNA ligands. To study pk1 structure and function in vivo, we have developed a genetic selection that ties the life of Escherichia coli cells to tmRNA activity. Mutation of pk1 at 20% per base and selection for tmRNA activity yielded sequences that retain the same pseudoknot fold. In contrast, selection of active mutants from 106 completely random sequences identified hairpin structures that functionally replace pk1. Rational design of a hairpin with increased stability using an unrelated sequence yielded a tmRNA mutant with nearly wild-type activity. We conclude that the role of pk1 in tmRNA function is purely structural and that it can be replaced with a variety of hairpin structures. Our results demonstrate that in the study of functional RNAs, the inactivity of a mutant designed to destroy a given structure should not be interpreted as proof that the structure is necessary for RNA function. Such mutations may only destabilize a global fold that could be formed equally well by an entirely different, stable structure. tmRNA 2The abbreviations used are: tmRNA, transfer messenger RNA; pk1, pseudoknot 1; A-site, ribosomal aminoacyl site; P-site, ribosomal peptidyl site; EOP, efficiency of plating. plays an important role in quality control of protein synthesis in eubacteria (1Karzai A.W. Roche E.D. Sauer R.T. Nat. Struct. Biol. 2000; 7: 449-455Crossref PubMed Scopus (344) Google Scholar, 2Withey J.H. Friedman D.I. Annu. Rev. Microbiol. 2003; 57: 101-123Crossref PubMed Scopus (126) Google Scholar). This dual-function RNA acts as both tRNA and mRNA to rescue ribosomes stalled on broken templates or at certain nascent peptide sequences (3Hayes C.S. Bose B. Sauer R.T. J. Biol. Chem. 2002; 277: 33825-33832Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). Aminoacylated tmRNA enters the A-site of stalled ribosomes and transfers alanine to the growing peptide chain as would a normal tRNA. The ribosome then resumes translation with tmRNA as the template, adding a 10-amino-acid tag to the nascent polypeptide. As a result of tmRNA action, stalled ribosomes are released and recycled, and the aborted protein product is marked for destruction by proteases (1Karzai A.W. Roche E.D. Sauer R.T. Nat. Struct. Biol. 2000; 7: 449-455Crossref PubMed Scopus (344) Google Scholar, 2Withey J.H. Friedman D.I. Annu. Rev. Microbiol. 2003; 57: 101-123Crossref PubMed Scopus (126) Google Scholar). In the trans-translation model of tmRNA function described above, the ribosome switches templates while synthesizing a single polypeptide. How does tmRNA position itself inside the ribosome for translation to resume in the correct frame? The global fold of tmRNA places the first codon near the decoding center of the ribosome; a short single-stranded sequence immediately upstream of this initial codon determines the precise frame (4Williams K.P. Martindale K.A. Bartel D.P. EMBO J. 1999; 18: 5423-5433Crossref PubMed Scopus (90) Google Scholar, 5Lee S. Ishii M. Tadaki T. Muto A. Himeno H. RNA (N. Y.). 2001; 7: 999-1012Crossref PubMed Scopus (51) Google Scholar, 6Lim V.I. Garber M.B. J. Mol. Biol. 2005; 346: 395-398Crossref PubMed Scopus (13) Google Scholar). The global structure of tmRNA is dominated by several pseudoknot structures surrounding the tag-encoding sequence (7Williams K.P. Bartel D.P. RNA (N. Y.). 1996; 2: 1306-1310PubMed Google Scholar, 8Felden B. Himeno H. Muto A. McCutcheon J.P. Atkins J.F. Gesteland R.F. RNA (N. Y.). 1997; 3: 89-103PubMed Google Scholar). The Escherichia coli tmRNA, for example, has four pseudoknots (see Fig. 1). Only 11 nucleotides upstream of the resume codon, pseudoknot 1 (pk1), may be involved with positioning the template in the ribosomal A-site. The cryoelectron microscopy structure of the 70 S ribosome bound to tmRNA, and its partner protein SmpB reveals that although pk2-4 are looped around the beak of the 30 S subunit without forming extensive contacts with the ribosome, pk1 is bound intimately between the beak and the decoding center on the 30 S subunit (9Valle M. Gillet R. Kaur S. Henne A. Ramakrishnan V. Frank J. Science. 2003; 300: 127-130Crossref PubMed Scopus (126) Google Scholar). Valle et al. (9Valle M. Gillet R. Kaur S. Henne A. Ramakrishnan V. Frank J. Science. 2003; 300: 127-130Crossref PubMed Scopus (126) Google Scholar) speculate that pk1 is pulled toward the decoding center as tmRNA transitions from the initial binding complex visualized by cryoelectron microscopy to full accommodation in the A-site resulting in Ala transfer and tmRNA translocation to the ribosomal P-site. In vitro studies support the conclusion that pk1 plays a crucial role in tmRNA function. Although pk2-4 are dispensable for trans-translation in vitro, replacing pk1 with single-stranded sequence destroys tmRNA function (10Nameki N. Tadaki T. Himeno H. Muto A. FEBS Lett. 2000; 470: 345-349Crossref PubMed Scopus (58) Google Scholar). Mutations designed to disrupt the base pairing of pseudoknot 1 helices reduce tmRNA-mediated tagging dramatically (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). The alteration of single-stranded loop sequences lowers activity, leading to proposals that these nucleotides form Mg2+ binding sites (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar) or make base-specific contacts with the ribosome (12Nameki N. Chattopadhyay P. Himeno H. Muto A. Kawai G. Nucleic Acids Res. 1999; 27: 3667-3675Crossref PubMed Scopus (36) Google Scholar). One limitation in the study of the trans-translation system has been the lack of a robust in vivo selection for tagging activity. The ssrA gene that encodes tmRNA in E. coli is not essential for growth under laboratory conditions (2Withey J.H. Friedman D.I. Annu. Rev. Microbiol. 2003; 57: 101-123Crossref PubMed Scopus (126) Google Scholar). One report of a genetic selection takes advantage of tmRNA-mediated tagging and proteolysis of the Arc repressor, allowing derepression of the kanR gene and survival on kanamycin (4Williams K.P. Martindale K.A. Bartel D.P. EMBO J. 1999; 18: 5423-5433Crossref PubMed Scopus (90) Google Scholar). The authors characterized 2451 colonies by replica plating and reported high background (∼0.1%); this screen is not robust enough to search through large collections of mutants (libraries) of tmRNA or other components of the tagging machinery. To extend the study of tmRNA structure and function to a relevant in vivo context, we have created a genetic selection that ties tmRNA function to the life of an E. coli cell. This selection allows the characterization of millions of mutants of tmRNA in a single experiment. Here we describe the application of this selection to identifying the structural and functional requirements for pseudoknot 1. How does the pk1 sequence determine its structure and what role does pk1 play in positioning tmRNA correctly inside the ribosome for resumption of translation on tmRNA? DNA oligonucleotides were synthesized on an ABI Expedite 8909 or purchased from Sigma-Genosys. Enzymes were purchased from New England Biolabs. Plasmid Construction—The tmRNA-K1 plasmid expresses the altered tmRNA gene from its own (ssrA) promoter on a low copy (p15A) origin with a tetracycline resistance marker. It was created by altering the tag sequence (ANDENYALAA) to code for the last 14 amino acids of KanR (ANKLQFHLMLDEFF). Nucleotides 90-137 of tmRNA, from the resume codon to the end of helix 5, were replaced with the sequence GCAAATAAACTGCAGTTTCATTTGATGCTCGATGAGTTCTTCTAATAACAGAATCTCATC. This sequence creates a stem-loop structure similar to helix 5 following the tag template. It contains a PstI cleavage site for cloning mutant pk1 regions immediately upstream. The tmRNA-K1 plasmid was created by PCR amplification of pKW11 (13Roche E.D. Sauer R.T. J. Biol. Chem. 2001; 276: 28509-28515Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar) using the primers GAGCATCAAATGAAACTGCAGTTTATTTGCGACTATTTTTTGCGGCTTTTTAC and GATGAGTTCTTCTAATAACAGAATCTCATCCCTCTCTCCCTAGCCTCC, followed by phosphorylation and blunt end ligation. Truncated kanR constructs were expressed from the araBAD promoter on a multicopy (pBR322 origin) vector with a β-lactamase marker. The first version of the selection, used in the 20% mutagenesis of wild-type pk1, induced stalling with the sequence Pro255-Pro256-Opal. The kanR-DPP plasmid was cloned by the amplification of the kanR gene with primers CATATGGCTAGCATGAGCCATATTCAACGGGAAAC and CGGCTTTTTCAAAAATATGGTATTGATCCGCCTTGAGAATTCGAGCTC, digestion with NheI and EcoRI, and ligation into the pBAD-GFP vector. To combine the two plasmids into a single selection vector, the tmRNA expression cassette was PCR-amplified from tmRNA-K1 with primers CCGCTACGGTCCGAGAACTGTGAATGCGCAAACC and TAGCGAAGATCTTAAATCCTGGTGTCCCTGTTG. The kanR-DPP plasmid was PCR-amplified with primers and were with and and mutagenesis and of and PstI sites from the resulting these sites surrounding the pseudoknot 1 sequence to be used for The sequence was then between and replacing wild-type pk1, and tmRNA This and sites to in active version of the selection vector was by replacing the amino acids with in the kanR This sequence was by cloning random of the gene in and for kanamycin resistance under and of a resistance cassette into the site a to reduce the plasmid tmRNA expression for the were created by PCR amplification of the pKW11 plasmid and blunt end The was similar to that described to make that was to the tag the only from wild-type tmRNA is the pseudoknot 1 The control was by the tmRNA sequence by digestion of pKW11 with and EcoRI, followed by with the of DNA to blunt for The pk1 sequences are as of pk1 mutagenesis of pk1 was by the 30 nucleotides of pk1 at 20% per The was synthesized with into the of the wild-type base and of the other short bound to a was by the of DNA to This DNA was by and PstI and into the vector, and the plasmid was into by plasmid the resulting was into the selection Sauer R.T. Science. 1996; PubMed Scopus Google Scholar). The cells were induced with for at 30 and and for at The with pk1 was in a similar it in of the sequence This was cloned into the selection vector and as The with 20% mutagenesis of replaced the sequence with as based on the The mutant were cloned into and on and kanamycin at from these were into from single and under selection conditions to that the tmRNA mutant in is for the tagging activity. of cells a pk1 mutants of tmRNA were in with tetracycline at The cells were in and in with for at to an of The cells were then with to an of and into 1 of at was to an at for with of and and at on tmRNA mutants were formed on wild-type cells as reported J. Friedman J. 1999; PubMed Google Scholar). The of were for four as reported in Fig. of a for tmRNA function of tmRNA in vivo is to tag with an this tag could be altered to protein function the activity of the protein would the of tmRNA function. To a genetic selection, the protein be for survival and have essential in its amino In the of tmRNA activity, the protein would be and the cells would in the of an active tmRNA the protein could be with the essential amino acids to form a functional and survival The selection protein we is the E. coli kanamycin resistance protein of the structure reveals that the helix of in plays an important role both and in binding the EMBO J. 2002; PubMed Scopus Google Scholar). of these to of cells the KanR kanamycin resistance cells the kanR tmRNA sequences the peptide tag in and and can be altered without of function (13Roche E.D. Sauer R.T. J. Biol. Chem. 2001; 276: 28509-28515Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar, P. K.P. Nucleic Acids Res. PubMed Google Scholar). We replaced the tag template with a sequence and expressed tmRNA from its own promoter from the tmRNA-K1 transfer of Ala from the tmRNA the of these 14 by tmRNA the amino acids that make the helix of We that the of the resume codon was an important was active the wild-type kanR codon not tmRNA template sequences resume P. K.P. Nucleic Acids Res. PubMed Google Scholar). at this position is important for the of the frame for of translation by the the altered tag to the KanR protein ribosomes at the position on the at as the codon the protein sequence in (3Hayes C.S. Bose B. Sauer R.T. J. Biol. Chem. 2002; 277: 33825-33832Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). The E. coli for example, in and is with efficiency to by tmRNA at the position of the codon (3Hayes C.S. Bose B. Sauer R.T. J. Biol. Chem. 2002; 277: 33825-33832Abstract Full Text Full Text PDF PubMed Scopus (132) Google Scholar). mutations in and a sequence in the kanR mRNA that acts as a to ribosome the selection on the structure of the KanR protein is to have a loop of in the helix EMBO J. 2002; PubMed Scopus Google Scholar). This loop the of the by these and transfer of Ala from tmRNA without KanR function. of the KanR on stalled ribosomes by the altered tmRNA functional KanR kanamycin resistance to We an E. coli tmRNA with the and tmRNA-K1 of the at the same with the tmRNA-K1 at the of 5 forming in The is a of the genetic selection for tmRNA function (4Williams K.P. Martindale K.A. Bartel D.P. EMBO J. 1999; 18: 5423-5433Crossref PubMed Scopus (90) Google Scholar). of genetic selection for a to active tmRNA sequences from of millions of in the sequences of active mutants for function. This of mutagenesis and selection was used to the of the pk1 sequence to its structure and function. In a of pk1 we the sequences toward the wild-type pk1 sequence to retain the pk1 fold to be for tmRNA function. We the sequence of pk1 at the of 20% per base (see and of mutants to selection that an of mutations were at random in the 30 nucleotides of pk1. The of tmRNA mutants was with the selection gene into an E. coli selection at on colonies The of pk1 sequences from the a of what are to The of mutations per pk1 sequence was the of in This was given the of the pk1 structure to The results in Fig. the at position by the position has low and is of the mutagenesis at position is the sequence is toward the wild-type The is and the are The two 1 and and and are not equally important in the pk1 structure 1 is with the base The by the is mutations at the other are by mutations that the base and In contrast, only of the base of are mutations of the and base are to based on in vitro (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar, N. Chattopadhyay P. Himeno H. Muto A. Kawai G. Nucleic Acids Res. 1999; 27: 3667-3675Crossref PubMed Scopus (36) Google loop can be in vivo without The that loop mutants form pseudoknots activity to the that these form an Mg2+ or ribosome binding The for or at position and at in vitro was not in in vivo and mutants were is in the or not the first in loop with to extend by base (7Williams K.P. Bartel D.P. RNA (N. Y.). 1996; 2: 1306-1310PubMed Google Scholar, 8Felden B. Himeno H. Muto A. McCutcheon J.P. Atkins J.F. Gesteland R.F. RNA (N. Y.). 1997; 3: 89-103PubMed Google Scholar, N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). that is is from (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). that can be to the other without of is We that is and is of loop The of the in loop is at the of that of the pseudoknot base that is in loop of pseudoknots at the D.P. Nucleic Acids Res. 2005; PubMed Scopus Google Scholar). may make P. N. S. A. 2001; PubMed Scopus Google Scholar) with the 1 as in the structures of several pseudoknots M. A. Nat. Struct. Biol. 1999; PubMed Scopus Google Scholar, J. Mol. Biol. 2001; PubMed Scopus Google Scholar). Such are with the in vitro that mutations in loop destabilize 1 in tmRNA pk1 (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). of may on the sequence of the and are in of of a mutant that these in 1 not activity in vitro (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). of what the requirements are for pk1 to fold we mutations could form structures that function in of pk1. To active sequences in sequence from wild-type pk1, we created a of 1 106 mutants in the pk1 sequence was completely This a of the of or selection at on colonies of tmRNA sequences that these mutants to the wild-type pk1 form pseudoknot structures by or analysis by the J. R. PubMed Scopus Google Scholar). the mutant structures are by the M. Nucleic Acids Res. 2003; PubMed Scopus Google Scholar) to form structures mutant Fig. and structural were at The activity of mutants that replace pk1 with a hairpin that pk1 is not for tmRNA activity. is active wild-type tmRNA to resistance to high kanamycin or at conditions under cells wild-type tmRNA To a we the sequence at 20% per base the 30 nucleotides replacing the pk1 of 106 was to selection on kanamycin at the is the with survival on at and survival on at The structure of contains two base its and is to be stable of pk1 initial of and function were with the KanR and in the tmRNA tag in the mutants were To with in the and to of the activity of the we the pk1 sequences into wild-type tmRNA the template tmRNA activity was by the efficiency of by the This only on with trans-translation (2Withey J.H. Friedman D.I. Annu. Rev. Microbiol. 2003; 57: 101-123Crossref PubMed Scopus (126) Google Scholar, J. Friedman J. 1999; PubMed Google cells wild-type tmRNA from a low copy plasmid support the growth of cells In to the wild-type and we two pk1 mutants that were characterized in vitro. The pk1 with a single-stranded sequence This mutant is from the study that although and were dispensable for tmRNA function in vitro, pk1 was essential (10Nameki N. Tadaki T. Himeno H. Muto A. FEBS Lett. 2000; 470: 345-349Crossref PubMed Scopus (58) Google Scholar). The the pk1 sequence is designed to destroy the of 1 by the from The mutant a in activity in vitro (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar). that these two mutants are from the tmRNA As by this a or pk1 region tmRNA completely in The from random sequence is active the or pk1 that its hairpin structure is to tmRNA stability and function wild-type The an activity The for survival in the high KanR between and activity. The that the mutant has base and stability to the that in stability of the hairpin could activity. a hairpin structure is that is to replace the wild-type pk1, is the that we and stable hairpin structure could functionally replace pk1. To these we designed an with a a sequence from the and This designed is to be stable the hairpin M. Nucleic Acids Res. 2003; PubMed Scopus Google Scholar). that is nearly as active as wild-type tmRNA, active that the stability of the structure replacing pseudoknot 1 plays an role in the function of tmRNA mutants that the sequence used is not as in vitro studies that tmRNA pseudoknot 1 is essential to the tagging function in It form a pseudoknot structure (10Nameki N. Tadaki T. Himeno H. Muto A. FEBS Lett. 2000; 470: 345-349Crossref PubMed Scopus (58) Google Scholar, N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google may make contacts to the ribosome (12Nameki N. Chattopadhyay P. Himeno H. Muto A. Kawai G. Nucleic Acids Res. 1999; 27: 3667-3675Crossref PubMed Scopus (36) Google and may play a role in positioning the template region of tmRNA near the decoding center for of translation (9Valle M. Gillet R. Kaur S. Henne A. Ramakrishnan V. Frank J. Science. 2003; 300: 127-130Crossref PubMed Scopus (126) Google although the precise of frame is by the single-stranded sequence between the resume codon and pk1 (4Williams K.P. Martindale K.A. Bartel D.P. EMBO J. 1999; 18: 5423-5433Crossref PubMed Scopus (90) Google Scholar, 5Lee S. Ishii M. Tadaki T. Muto A. Himeno H. RNA (N. Y.). 2001; 7: 999-1012Crossref PubMed Scopus (51) Google Scholar, 6Lim V.I. Garber M.B. J. Mol. Biol. 2005; 346: 395-398Crossref PubMed Scopus (13) Google Scholar). To study the requirements for pk1 structure and its to tmRNA function in vivo, we developed a genetic selection that ties the life of an E. coli to tmRNA activity. KanR an essential sequence are stalled on by an altered tmRNA that for the amino the ribosomes functional and the on kanamycin tmRNA activity are This selection the to active mutants large of tmRNA, of active is to this genetic selection to other components of the trans-translation The of two with of mutagenesis The first by a mutagenesis per the in vitro that mutations that destabilize the pseudoknot structure of pk1 destroy tmRNA activity. mutations were the 1 region mutations of its base the of its two base The of loop are to that not form Mg2+ binding sites or as ribosomal components as (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar, N. Chattopadhyay P. Himeno H. Muto A. Kawai G. Nucleic Acids Res. 1999; 27: 3667-3675Crossref PubMed Scopus (36) Google Scholar). does not with and other at the of loop form with the end of 1. the sequence requirements of the pk1 structure a of to what was are with in vitro the that a well pseudoknot 1 structure is essential to tmRNA function. In contrast, results from the in pk1 was replaced by completely random sequence support the conclusion that stable structure is mutants were that structures in of pk1. of mutagenesis and selection the activity of the of these by hairpin stability Rational design of an stable using an unrelated created a mutant is nearly as active as wild-type The sequence of the hairpin to be important its results were with a using This that results are not of an of the KanR or the altered tmRNA tag sequence used in the is the of these of mutagenesis of the wild-type pk1 sequence destroys or the pk1 allows stable structures in sequence to be We that the only for the pk1 region of tmRNA is that it a stable structure that other global stable were by and in vitro (11Nameki N. Felden B. Atkins J.F. Gesteland R.F. Himeno H. Muto A. J. Mol. Biol. 1999; 286: 733-744Crossref PubMed Scopus (61) Google Scholar, N. Chattopadhyay P. Himeno H. Muto A. Kawai G. Nucleic Acids Res. 1999; 27: 3667-3675Crossref PubMed Scopus (36) Google the role of stable pseudoknots is to these structures from pseudoknot structure is not necessary This has important for of structures in functional The inactivity of a mutant designed to destroy a given structure does not that the structure is necessary for as is The may global that in the This does not that the given structure could not be replaced by a stable structure from In the inactivity of a mutant pseudoknot does not that the structure has to be a it only that the region in form a stable fold. and global should be an important in mutations to the role of structures in RNA and in Our the of a large of mutants and the of genetic to in an We Sauer and for the pKW11 plasmid and the and Friedman for the
22
WebLogo: A Sequence Logo Generator: Figure 1
Gavin E. Crooks, Gary C. Hon, John‐Marc Chandonia et al. · Genome Research · 2004 · 13K citations · Full text
Signal Transduction by DR3, a Death Domain-Containing Receptor Related to TNFR-1 and CD95
Arul M. Chinnaiyan, Karen O’Rourke, Guo-Liang Yu et al. · Science · 1996 · 653 citations