UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functions and properties of RNase G and RNase E from Escherichia coli Briant, Douglas James 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2003-859991.pdf [ 12.43MB ]
JSON: 831-1.0091231.json
JSON-LD: 831-1.0091231-ld.json
RDF/XML (Pretty): 831-1.0091231-rdf.xml
RDF/JSON: 831-1.0091231-rdf.json
Turtle: 831-1.0091231-turtle.txt
N-Triples: 831-1.0091231-rdf-ntriples.txt
Original Record: 831-1.0091231-source.json
Full Text

Full Text

FUNCTIONS AND PROPERTIES OF RNase G and RNase E FROM  Escherichia coli.  by Douglas James Briant Hon. B. Sc., The University of Waterloo, 1993 M. Sc., The University of Waterloo, 1995 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Biochemistry and Molecular Biology We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA July, 2003 © Douglas James Briant, 2003  UBC Rare Books and Special Collections - Thesis Authorisation Form  Page 1 of 1  In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Columbia, I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the head o f my department o r by h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n .  Department of BiflCKf nAvCT<M  I MULCfv/Lfll  "IQuptV^j  The U n i v e r s i t y o f B r i t i s h Columbia Vancouver, Canada Date  http://www.library.ubc.ca/spcoll/thesauth.html  7/18/03  Abstract  Ribonucleic acid (RNA) is a vital molecule in the cell. Messenger RNA (mRNA) serves as the intermediate between DNA and protein while ribosomal RNA (rRNA) and transfer RNA (tRNA) catalyze translation. RNA is processed and ultimately degraded by ribonucleases. The majority of the endoribonucleolytic activity in Escherichia coli is derived from RNase E. This 1061 amino acid protein associates with at least three other proteins to form a complex called the RNA degradosome. This work established that the degradosome does not assemble de novo with successive rounds of degradation. It also determined that RNase E lacks 5'-phosphatase activity. Studies into the activity of RNase E are confounded by the fact that it associates into a complex with other RNA processing enzymes. We therefore utilized RNase G, which shares 35% amino acid sequence identity (50% similarity) to the catalytic domain of RNase E, as a model for RNase E. RNase G is the endonuclease responsible for forming the mature 5'-end of 16S rRNA. Non-denaturing purifications for RNase G were developed and the correct Nterminal sequence of the protein unambiguously identified. Through crosslinking studies, sucrose gradient centrifugation and gel filtration, it was determined that RNase G, and by inference, RNase E, exists primarily as a dimer. Site-directed mutagenesis was utilized to elucidate the role of six cysteine residues, including two highly conserved cysteines, of RNase G. None of the mutations resulted in a loss of activity, although subtle influences on structure and activity were observed with the RNase G variants. The S1 domain, which potentially binds RNA, was deleted without inactivating the enzyme. Further ii  studies are required to determine if the S1 domain plays a role in substrate recognition. Finally,  examinations  using  synthetic,  chimeric  RNA-DNA  oligonucleotides revealed the chemical requirements for recognition and cleavage of a substrate by RNase E or RNase G. I concluded that a single 2'OH group 5' to the site of cleavage was sufficient for endoribonucleic activity. This work established RNase G as a model for investigating the activity and structure ofthe catalytic domain of RNase E.  in  Table of contents  Abstract List of figures List of tables List of abbreviations Acknowledgements 1 INTRODUCTION 1.1 Importance of mRNA decay 1.2 Endoribonucleases 1.2.1 RNaseE Specificity of RNase E RNase E and the degradosome Architecture of RNase E Intracellular localization of the degradosome Importance of the degradosome 1.2.2 RNase G Identification of RNase G as an endonuclease Role for RNase G in mRNA decay Comparison of the activities and specificities of RNase G and RNase E 1.2.3 Other endonucleases 1.3 Exonucleases 1.3.1 Exonucleases involved in mRNA decay 1.3.2 Revised model of mRNA decay 1.3.3 Other exoribonucleases 1.4 Aims of this project 2 MATERIALS AND METHODS 2.1 Source of reagents 2.2 Bacterial strains and culture techniques  iv  Page ii vii x xi xiii 1 1 2 5 6 12 15 18 19 20 22 27 28 32 33 33 34 37 38 39 39 39  2.3 2.4 2.5 2.6  Common buffers Polymerase chain reaction Other molecular biological methods Enzyme purifications 2.6.1 Untagged RNase G 2.6.2 RNase G mutants 2.6.3 His-tagged RNase G 2.6.4 Purification of pnpl3 degradosomes 2.6.5 Electropurification of RNase E 2.7 Protein analysis 2.7.1 SDS-polyacrylamide gel electrophoresis 2.7.2 Western blotting for RNase G 2.7.3 N-terminal protein sequencing 2.7.4 Mass spectrometry 2.7.5 Protein cross-linking experiments 2.7.6 Circular dichroism 2.7.7 Sucrose gradient centrifugation 2.7.8 Gel filtration size determination 2.8 RNA substrates and endoribonucleases 2.8.1 2.8.2 2.8.3 2.8.4  39 41 42 42 42 43 45 46 48 49 49 50 51 51 51 52 52 53 53  Full-length substrates 5'-end labelling of oligonucleotides Standard endoribonuclease assays Ribonucleoprotein assays  53 57 57 58  2.8.5 Competition assays 2.8.6 Phosphatase assays and thin-layer chromatography 3 ACTIVITY AND QUATERNARY STRUCTURE OF RNase G  58 59 60  3.1 Introduction 3.2 Results 3.2.1 Sequence comparison 3.2.2 Purification of RNase G and identification of its translational start site  v  60 62 62 67  3.2.3 3.2.4  4  5  6 7 8  Oligomerization of RNase G Effect of cysteine to serine mutations in RNase G Physical properties of RNase G cysteine to serine mutations Activity of wild type RNase G and mutants 3.2.5 Effect of S1 deletion on RNase G activity 3.2.6 rRNA processing activity of recombinant RNase G 3.3 Discussion RELATIONSHIPS WITHIN THE RNA DEGRADOSOME 4.1 Introduction 4.2 Results 4.2.1 Purification of degradosomes from a pnp13 strain 4.2.2 Endonucleolytic cleavages by wild type and pnp13 degradosomes 4.2.3 PNPase and the pnp13 degradosome 4.2.4 Phosphatase activity of the degradosome 4.3 Discussion ACTIVITY STUDIES OF RNase E AND RNase G USING SYNTHETIC OLIGONUCLEOTIDES 5.1 Introduction 5.2 Results 5.2.1 Design of chimeric oligonucleotides 5.2.2 Activity of RNase E against chimeric substrates 5.2.3 Activity of RNase G against chimeric substrates 5.3 Discussion PERSPECTIVES REFERENCES APPENDIX  vi  74 79 84 93 100 102 108 115 115 116 116 117 119 129 133 141 141 143 143 144 149 150 154 165 187  List of Figures  Figure 1.1 Figure 1.2 Figure 1.3 Figure 1.4 Figure 1.5 Figure 1.6 Figure 1.7 Figure 1.8 Figure 2.1 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11 Figure 3.12 Figure 3.13 Figure 3.14  The Apirion model for mRNA decay Sequence requirements for RNase E recognition and cleavage 5'-Tethering model of RNA decay Protein interaction domains of RNase E The mre gene cluster Processing of 16S rRNA rne-lacZ fusion reporter gene for examining RNase E autoregulation Current model of mRNA decay Templates for runoff RNA transcription Sequence alignment of RNase G and RNase E Potential translation initiation sites of the rng gene Purification of RNase G Construction of pDB1 Oxidative cross-linking of RNase G Sedimentation velocity analysis of RNase G Gel filtration analysis of RNase G Structural comparison of RNase E and RNase G, and the location of cysteine residues Purity of RNase G mutants Western blotting of RNase G variants with cysteine to serine mutations Crosslinking of RNase G mutants Sedimentation velocity analysis of RNase G mutants Comparison of rpsT mRNA and 9S processing sites Endonuclease activity of RNase G variants with cysteine to serine mutations  vii  Page 3 8 10 16 21 23 29 35 54 63 68 70 72 76 78 80 82 85 86 89 91 94 97  Figure 3.15 Effect of S1 deletion on RNase G endoribonucleolytic activity Figure 3.16 Reconstitution of RNP substrate processing Figure 3.17 Primer extension analysis of in vitro processing reactions Figure 3.18 Processing sites observed with reconstituted RNP substrate Figure 4.1 Purification of pnp13 degradosomes Figure 4.2 Endonuclease activity of wild type and pnp13 degradosomes on rpsTmRNA Figure 4.3 Phosphate-stimulated activation of PNPase in degradosomes Figure 4.4 Secondary structure of the malEF RNA substrate Figure 4.5 PNPase activity of pnp 13 degradosome Figure 4.6 Phosphatase activity of degradosomes Figure 4.7 Phosphatase activity of pnp13 degradosomes on ethanol-precipitated oligoribonucleotides Figure 4.8 Model for malEF mRNA degradation by wild type and pnp13 degradosomes with active PNPase Figure 5.1 RNA1 and RNA oligonucleotides Figure 5.2 Synthetic oligonucleotide sequences Figure 5.3 Activity of degradosomes and RNase G on oligonucleotides Figure 5.4 Requirements for RNase E and RNase Gmediated cleavage of BR10 Figure 6.1 Schematic structure of RNase G Figure 6.2 Two-metal-ion mechanism for endonucleolytic cleavage Figure 6.3 Oligonucleotide competitors of RNase G and RNase E activity  viii  101 103 106 113 118 120 122 125 127 130 134 138 142 145 146 151 155 158 162  Appendix Figure 1 Appendix Figure 2  Column elution profiles for the purification of untagged RNase G Gel filtration size determination of RNase G samples  ix  187 189  List of Tables  Table 1.1 Ribonucleases involved in RNA processing and degradation in E. coli Table 2.1 Bacterial strains Table 2.2 Primers and templates used for construction of RNase G clones Table 3.1 Features of RNase G and RNase E Table 3.2 Mass spectrometer analysis and sequencing of RNase G mutants Table 3.3 Comparison of RNase G mutants  4 40 44 66 75 88  List of Abbreviations AQOO  aa ATP C-terminal BSA DEPC DTT EDTA ETS fMet HSR 9 IPTG kDa KH domain LB Mg-acetate MW MWCO NaF Na-phosphate NH -acetate N-terminal nt(s) P/C/l PCR P-DNA PH domain 4  absorbance at 600 nm amino acid adenosine 5'-triphosphate carboxyl-terminal bovine serum albumin diethyl pyrocarbonate dithiothreitol ethylenediamine tetraacetate externally transcribed spacer N-formylmethionine high similarity region gravity isopropyl-p-D-thiogalactopyranoside kilo Daltons K-homology domain, RNA binding domain Luria-Bertani Broth magnesium acetate molecular weight molecular weight cut-off sodium fluoride sodium phosphate ammonium acetate amino-terminal nucleotide(s) phenol/chloroform/isoamyl alcohol polymerase chain reaction 5'-monophosphorylated DNA oligonucleotide homologous to RNase PH, no relation to Pleckstrin Homology domain  XI  Pi PMSF ppte psi RNP S1 domain SDS TAE TBE TCA TLC Tm tris ts U unppte UTR w/v  inorganic phosphate phenylmethanesulfonyl fluoride precipitate(d) pounds per square inch ribonucleoprotein RNA binding domain homologous to ribosomal protein S1 sodium dodecyl sulphate tris-acetate-sodium EDTA tris-boric acid-sodium EDTA trichloroacetic acid thin layer chromatography midpoint unfolding temperature tris(hydroxymethyl)-aminomethane temperature-sensitive units unprecipitated untranslated region weight/volume  xii  Acknowledgements  Many people have helped me during the course of my studies at UBC. Most importantly, I'd like to thank George Mackie for giving me the opportunity to work in his lab. George has been a fantastic supervisor, who somehow always found time to discuss my project despite an oppressive schedule of meetings. He instills both knowledge and enthusiasm into everyone that has the pleasure of working with him. I would also like to thank my committee members, Ross MacGillivray and Lawrence Mcintosh, for valuable input and direction. Thanks also to George Jones, whom I had the opportunity to work alongside. George introduced our lab to the wonderful and highly aromatic world of  Streptomyces.  A lot of people have come and gone through the lab over the years (how many years?) who have all contributed in some way to my project. Thanks to Glen, X-man, Steph, Anand, Kenny G (the original cast), Jan, Annie, Rob, Kristian, Catherine, Mike and countless undergrads now littering med schools and decontamination units around the country. Thanks to Cam and Rich for their sage advice, without whom I would still be trying to figure out how to purify my proteins. What is life without the occasional distraction? These were kindly provided in-house by those mentioned above as well as Angus, Loewen, Nelson, Rohde and Brunstein. I wish to thank my friends in the outside world,  Xlll  Ying and Trish, and their burgeoning family, Paul and Laurie (who aren't really dating), Damir, Holly, Chris, Ron and Storm Brewing. Of course, none of this would have been possible without the love and support of my parents, Jim and Edna, my sister Laura and her husband Gary. I have also received much support and encouragement from Yoshiro and Emiko Arakawa. Finally, I want to thank my wonderful wife Toshie. She gets to end this list, because she is always there to anchor me when I need her.  Slainte,  Douglas J. Briant  xiv  1. INTRODUCTION  RNA plays a central role in the metabolism of a cell. Messenger RNA (mRNA) acts as an intermediate between DNA and protein. mRNA is inherently unstable within the cell due to the activity of ribonucleases. Ribosomal RNA (rRNA) and transfer RNA (tRNA), which are essential to translation, are stabilized by association with auxilliary proteins and/or folding into stable structures. Both rRNA and tRNA, however, require processing by ribonucleases prior to maturation. It follows that a strong comprehension of ribonucleolytic events is important for understanding the expression of protein from genes.  1.1. Importance of mRNA decay  The degradation of mRNA is an important metabolic process, as it contributes to the steady-state levels of mRNA. This in turn helps set the amount of protein produced from an mRNA template. In Escherichia  coli,  the half-lives of most  mRNAs range from less than 30 seconds to greater than 15 minutes, with an average half life of 90 seconds (for recent reviews, see Coburn and Mackie, 1999; Grunberg-Manago, 1999). Rates of transcription and translation are roughly equal in  E. coli  (reviewed by Richardson and Greenblatt, 1996). Thus,  rapid mRNA turnover limits the translational yield per mRNA and allows the bacteria to respond rapidly to environmental changes and, most importantly, to negative regulatory signals. Without this rapid decay, negative regulatory signals would block the production of new mRNAs, but the existing mRNAs would persist and could still be translated. If mRNA turnover is rapid, however, the amount of protein which could successfully be translated following 1  transcriptional inactivation would be minimized. The decay of mRNA is also important for differential expression of polycistronic messages (Newbury et al., 1987) and the recycling of ribonucleotides, because approximately 60% of RNA being synthesized at any time is mRNA, while mRNA only accounts for 5% of the total cellular RNA (Deutscher and Reuven, 1991). Therefore, without efficient recycling of nucleotides, the bacterium would not be capable of maintaining RNA levels. A simple model of prokaryotic mRNA decay was initially proposed by Apirion (Apirion, 1973) and is outlined in Fig. 1.1. In this model, mRNA is protected from exoribonuclease attack at the 3'-end by RNA polymerase. Decay is initiated by one or more endonucleases. The products of endonucleolytic attack are subsequently degraded to individual ribonucleotides by 3' to 5'exonucleases. Exoribonucleases, however, are impeded by secondary structure (McLaren et al., 1991), and this model fails to address how stable 3'-secondary structure is degraded. Updated models of mRNA decay will be presented in sections and 1.3.2. The endoribonucleases will be introduced in section 1.2, and the exoribonucleases in section 1.3.  1.2. Endoribonucleases  Endoribonucleases cleave RNA at internal sites. Eight endoribonucleases have been identified to date in E.  coli,  and these are listed in Table 1.1. Of these, only  RNase E, RNase G and RNase III have been found to play a role in mRNA  2  endonucleases  5'  PPP  I endonucleases expose new I 3'-ends, which are attacked A by exonucleases  5'  PPP  exonucleases  Figure 1.1. The Apirion model for mRNA decay. In this cartoon,  mRNA is represented as a straight line, with the 5'-triphosphorylated end indicated at left. The scissors represent endonucleolytic cleavages, while the "pacman" represents an exoribonuclease. The arrow indicates the 3' to 5' direction of exonucleolytic decay.  3  Table 1.1. Ribonucleases involved in RNA processing and degradation in E. coli. Enzyme  comments  endonucleases  RNase E RNase G RNase III RNase P RNase I RNase H RNase HI Rel E  mRNA degradation, rRNA processing, tRNA processing maturation ofthe 5'-end of 16S rRNA, limited role in mRNA decay, homologous to RNase E double-strand specific, rRNA processing, limited role in mRNA decay catalytic component is a ribozyme, forms mature 5'end of tRNA periplasmic, scavenger removal of RNA primers following DNA replication removal of RNA primers following DNA replication, less active than RNase H cleaves mRNA in ribosomal A site  exonucleases  PNPase RNase II oligoribonuclease RNase R RNase BN RNase D RNase PH RNase T  mRNA decay, associated with RNA degradosome, phosphorylytic hydrolytic, mRNA decay degrades limit oligonucleotides, essential involved in virulence, homologous to RNase II, possible role in degrading rRNA tRNA 3'-end processing tRNA 3'-end processing tRNA 3'-end processing, phosphorylytic tRNA 3'-end processing, forms mature 3'-ends of 5S and 23S rRNA  4  decay (for reviews, see Coburn and Mackie, 1999; Condon and Putzer, 2002; Grunberg-Manago, 1999; Steege, 2000). This introduction will focus mainly on RNase E and RNase G.  1.2.1. RNase E  RNase E was initially identified as the activity that produces pre-5S rRNA from a 9S RNA precursor, itself the product of RNase III cleavages of pre-rRNA (see Fig. 1.6; Misra and Apirion, 1979). The gene encoding RNase E was separately identified as ams, me, and hmp1 (Casaregola et al., 1992; Ghora and Apirion, 1978; Misra and Apirion, 1979; Ono and Kuwano, 1980). The hmp1 gene (high molecular weight protein) was identified by its sequence similarity to yeast heavy chain myosin (Casaregola et al., 1990), but its relationship to mRNA stability was not immediately realized. The  rne-3071  gene contains a temperature  sensitive mutation (Fig. 3.1) that leads to the accumulation of pre-5S rRNA at non-permissive temperature (Ghora and Apirion, 1978), while ams-1 contains a temperature sensitive mutation that resulted in the accumulation of mRNA decay fragments (Ono and Kuwano, 1980). Attempts were made to identify the me gene by complementation of the temperature sensitivity of  rne/ams.  None of  these attempts successfully led to the cloning of the rne gene for two reasons. First, only the N-terminal half of RNase E is required for activity and viability (Kido et al., 1996; McDowall and Cohen, 1996). Second, at non-permissive temperatures, Rne/Ams protein cannot autoregulate its own synthesis, resulting in significant overexpression of an active, yet labile enzyme. As a consequence, RNase E activity can be "rescued" in high copy suppression experiments (Jain 5  and Belasco, 1995b; Mudd and Higgins, 1993). Further complicating the issue, RNase E is very sensitive to proteolysis and aggregation. Together, these factors led to problems identifying the rne/ams/hmp1 gene product (Coburn and Mackie, 1999). Full-length RNase E was eventually successfully cloned, overexpressed and purified from an SDS-polyacrylamide gel followed by renaturation. This led to its identification as the 1061 amino acid product ofthe rne gene (Cormack et al., 1993). The renatured protein could reproduce the processing of 9S RNA and rpsT mRNA in vitro, demonstrating its independence of cofactors (Cormack et al., 1993). In E. coli, RNase E is an important endoribonuclease. It is essential for initiating the degradation of mRNA (for reviews, see Coburn and Mackie, 1999; Grunberg-Manago, 1999; Regnier and Arraiano, 2000; Steege, 2000), processing of ribosomal RNA (Li et al., 1999b; Misra and Apirion, 1979; Wachi et al., 1999) and the maturation of transfer RNA (Li and Deutscher, 2002; Ow and Kushner, 2002). RNase E also autoregulates the stability of its own mRNA through a target in the 5'-untranslated region (Diwa et al., 2000; Jain and Belasco, 1995a; Mudd and Higgins, 1993). Specificity of RNase E  Two experiments demonstrated that RNase E is specific for single-stranded RNA. In the first, structure-mapping ofthe model substrate rpsT mRNA provided evidence that RNase E cleavage sites were primarily located in single-stranded regions. In the second experiment, rpsT mRNA was denatured prior to the assay. Subsequent refolding resulted in several potential RNase E sites 6  becoming double-stranded. These sites were no longer recognized and cleaved by RNase E (Mackie, 1992). RNase E has no strict recognition sequence although certain features are often associated with a cleavage site. These features are illustrated in Fig. 1.2. RNase E shows a preference for cleavage 5' to an AU (Cormack and Mackie, 1992; Ehretsmann et al., 1992; Mackie, 1992; Mackie and Genereaux, 1993; McDowall et al., 1994). In several cases, a G residue two nucleotides upstream of the scissile bond enhances the efficiency of cleavage (Walsh et al., 2001). Mutation in identified cleavage sequences resulted in altered rates of hydrolysis by RNase E, implying that this process is directed, rather than completely random (Lin-Chao et al., 1994; McDowall et al., 1994). Furthermore, cleavage sequences are often flanked by secondary structure (Cormack and Mackie, 1992; Mackie, 1992; Mackie and Genereaux, 1993), although removal of stem-loops downstream of cleavage sites of 9S RNA has little effect on its processing by RNase E. Since a synthetic 10 nucleotide oligonucleotide based on a known RNase E cleavage site is cleaved efficiently, it was concluded that secondary structure is not necessary for recognition and cleavage of RNA by RNase E (McDowall et al., 1995). Rather, it appears that flanking secondary structure may play a role in anchoring recognized cleavage sites in single-stranded form (Mackie and Genereaux, 1993). Stable stem-loops at the extreme 5'-end of an mRNA confer stability against attack by RNase E (Chen et al., 1991; Hansen et al., 1994). Such protection is lost if as few as 3 to 5 single-stranded nucleotides precede the stable stem-loops (Bouvet and Belasco, 1992). This observation was explained by G. A. Mackie, who utilized circularized RNA and partial heteroduplexes to 7  o 5—\\-  G x A u—ti  \\  3'  RNase E  Fig. 1.2. Sequence requirements for RNase E recognition  and cleavage. RNase E  prefers to cleave 5' to an AU  dinucleotide. This is indicated by the arrow. A G residue 2 nts upstream of the cleavage site enhances cleavage. Secondary structures are thought to anchor the cleavage site in a singlestranded form, which can be recognized by the enzyme.  8  demonstrate that RNase E was strongly inhibited by the lack of a free singlestranded 5'-end either in vivo or in vitro (Mackie, 1998; Mackie, 2000). In vitro studies revealed that RNase E showed a strong preference for 5'monophosphorylated, over 5'-triphosphorylated ends (Mackie, 1998). This has important ramifications, as intact mRNA is triphosphorylated at the 5'-end and products of endoribonucleolytic cleavage retain 5'-monophosphorylated termini. Previously cleaved RNA will, therefore, be strongly preferred as a substrate for subsequent cleavage. This explains the usual lack of observable degradative intermediates in E. coli and led to development of a model for mRNA called the "5'-tethering model". The "5'-tethering model" (or "alternating sites model") is depicted in Fig. 1.3. This model requires the combined action of at least two subunits of RNase E, which are shown as components of the degradosome (section in Fig. 1.3. In the initial step, RNase E binds at or near the 5'-end. There may, or may not, be an endoribonucleolytic cleavage at this stage. Following the initial interaction of the degradosome with the mRNA, the second subunit of RNase E binds to the second site, which is cleaved. RNase E cleavage sites are described in section, and depicted in Fig. 1.2. The segment of mRNA 5' to the cleavage site is now rapidly degraded exonucleolytically, either by PNPase associated with the degradosome or by RNase II. The first subunit then releases any uncleaved mRNA, and loops to the next available RNase E target site. This process is repeated until the mRNA is completely degraded. This processive model can explain the "all or none" phenomenon of mRNA decay,  9  Figure 1.3. 5'-Tethering model of RNA decay. Triphosphorylated mRNA substrate is represented as a line, with the 5' and 3' ends indicated. Sequential RNase E cleavage sites are shown as boxes. RNase E sites are described in section, and shown in Fig. 1.2. The degradosome consists of three proteins in this figure, with a dimer of RNase E indicated as two circles, one unshaded (Subunit I) and one shaded (Subunit II). Each subunit of RNase E contains an RNA binding site, indicated by an open box. There are two trimers of PNPase (striped circles) and two units of RhIB (black triangles). Enolase is omitted for simplicity. The predicted sequence of events proceeds from top to bottom, with arrows separating each step. The diagonal arrow shows the direction of exonucleolytic degradation from 3' to 5'. This is adapted from a figure that originally appeared in Coburn and Mackie (Coburn and Mackie, 1999)  10  \ pppj  3' E1  E2  E3  R N a s e E in the degradosome binds 5'-end. Cleavage may occur.  PPP. E1  E2  E3  i  second subunit of R N a s e E binds and cleaves Site 2 and remains bound to the newly exposed 5' P  3' E3  5'-end fragment containing E1 is released and degraded to limit oligonucleotides by exonucleases  i  5'E2  3' E3  subunit 1 now binds and cleaves site 3; this cycle continues until the m R N A is completely degraded  11  since decay intermediates are not usually detectable in E. coli (Coburn and Mackie, 1999). This model implies that RNase E exists as at least a dimer. It is important, therefore, to determine the multimeric state of this enzyme. This will be addressed in Chapter 3. While circularized substrates, which lack a 5'-end, are stabilized against RNase E-mediated endonucleolytic attack, they are still degraded slowly over time (Mackie, 1998; Mackie, 2000). Work performed in this lab revealed that RNase E could also recognize internal sites, independently of interaction with the 5'-end. This has been termed the "bypass" or "internal entry" pathway (Baker and Mackie, 2003). This pathway is significantly slower than the 5'-end recognition pathway. RNase E and the degradosome  During attempts to purify RNase E by non-denaturing methods, it was found that RNase E copurified with three additional polypeptides that migrated as 85, 50 and 48 kDa bands (Carpousis et al., 1994). This same study utilized limited proteolysis to identify the 85 and 48 kDa bands as the PNPase a and p subunits, respectively. The 85 kDa protein was also isolated from an SDS-PAGE gel, renatured, and assayed to show that this protein had PNPase activity (Carpousis et al., 1994). PNPase is further discussed in section 1.3.1. While PNPase was initially described as a two subunit protein, with the 85 kDa a subunit containing 3' to 5' single-strand specific exoribonuclease activity (Portier, 1975), the N-terminal sequence of the p subunit revealed it to be enolase (Py et 12  al., 1996). Enolase is the glycolytic enzyme responsible for the dehydration of 2phosphoglycerate to phosphoenolpyruvate. Only 5-10% of cellular enolase copurifies with the degradosome (Py et al., 1996) and the role of enolase in the degradosome is unclear. The 50 kDa protein, in turn, was identified as RhIB by N-terminal sequencing and ATPase activity (Py et al., 1996). RhIB is a DEADbox helicase, named because of a conserved aspartic acid (D), glutamic acid (E), alanine (A), aspartic acid (D) domain (Kalman etai., 1991). The presence of RhIB in the degradosome is provocative, as the helicase may unwind secondary structures, which would otherwise impede single-strand specific ribonucleases. RhIB-mediated ATPase and helicase activity has been identified by several groups (Coburn et al., 1999; Py et al., 1996; Vanzo et al., 1998). Initial reports indicated that this activity was dependent upon interaction with RNase E, a novel phenomenon for DEAD-box helicases (Coburn et al., 1999; Vanzo et al., 1998). More recently, Liou et al (2002) have claimed that RhIB and PNPase may interact directly to degrade double stranded RNA. The size and stoichiometry of the degradosome remains elusive. The degradosome is now estimated to be between 1.5 and 2.4 million Da (reviewed by Steege (2000)), although initial estimations based on glycerol sedimentation ranged from 160 to 460 kDa (Carpousis et al., 1994). Similarly, discrepancies surround the stoichiometry of the degradosome, with reports of molar ratios of RNase E, PNPase, RhIB and enolase of 1: 1.5: 0.6: 1.7 (Carpousis et al., 1994), 1: 2.7: 1.6: 2.7 (Py et al., 1994; Py et al., 1996) and 1: 0.9: 0.9: 1.8 (Miczak et al., 1996). Data from our lab estimate the ratio of RNase E: PNPase:  13  RhIB to be 1: 1.5: 1 (Coburn et al., 1999). These differences may arise from differences in isolation procedures (Grunberg-Manago, 1999). Three additional proteins have been identified to interact, albeit substoichiometrically, with the degradosome. Two of these, DnaK and GroEL, are chaperones that were identified by N-terminal sequencing of proteins associated with FLAG-tagged RNase E (Miczak et al., 1996). These proteins are not required for the activity of either RNase E or PNPase (Coburn and Mackie, 1998; Cormack et al., 1993) and are likely contaminants of the degradosome. Indeed, GroEL only associates with the temperature-sensitive RNase E mutant, rne-3071 (Miczak et al., 1996). The third sub-stoichiometrically-associated  protein is polyphosphate kinase (PPK), a non-essential protein that polymerizes long-chain polyphosphate at the terminal phosphate of ATP (Blum et al., 1997; Miczak et al., 1996). While a PPK deletion has no effect on bulk mRNA decay, the addition of polyphosphate inhibits the degradation of both malEF and ompA mRNA in vitro. Addition of ADP in the presence of PPK generates ATP from the polyphosphate and alleviates the inhibition of degradation of both ompA and malEF mRNA (Blum et al., 1997). While the exact implications of PPK in the  degradosome are not clear, it has been postulated that it may help maintain a localized concentration of ATP. A minimal degradosome was assembled in vitro, and required only RNase E, PNPase and RhIB to maintain function (Coburn et al., 1999). This demonstrated that the degradosome could assemble and function in the absence of cellular factors. It also revealed that enolase was not required for degradosome function. These experiments also demonstrated that the minimal 14  degradosome could be assembled in the absence of catalytically active RNase E, as long as the C-terminal half of RNase E remained intact. These experiments will be addressed further in Chapter 4. Assembly of proteins into ribonucleotide-degrading machines is not unique to E.  coli.  An RNase E-based degradosome, consisting of two DEAD-box  helicases and the transcription terminator Rho was isolated from capsulatus,  Rhodobacter  although no exonuclease was associated with the complex (Jager  et  al., 2001). Conversely, mitochondria, yeast and higher organisms possess "exosomes", which consist of several exonucleases associated with RNA helicases. Exosomes degrade RNA in a 3' to 5' direction (Margossian et al., 1996; Mitchell et al., 1997). RNA processing complexes with both exoribonuclease and endoribonuclease activity have also been isolated from chloroplasts (Hayes et al., 1996) although these data have been challenged recently (Baginsky et al., 2001). Architecture of RNase E  During attempts to discover suppressors of a temperature-sensitive mutation in a strain that does not partition chromosomes properly at 44°C, six suppressors were revealed, all of which lay within the rne gene (Kido et al., 1996). These genes encoded RNase E proteins that lacked the C-terminal "half (Fig. 1.4), but were capable of rescuing rne-1 temperature-sensitive mutants at non-permissive temperature. Significantly, these mutants did not copurify with PNPase (Kido et al., 1996) or the other components of the degradosome (Kaberdin et al., 1998; Vanzo et al., 1998). Other investigations showed that the 15  1  |  1  catalytic domain  35 135 S1 domain  ^j*  I  8  IIIHHIII  600 750 ARRBD  EM  1061  self interaction  1-528  self interaction  500-752  RhIB  734-738  enolase  739-845  PNPase  844-1045  Figure 1.4. Protein interaction domains of RNase E. Intact RNase E  (1061 amino acids) is represented by a large rectangle, with the Nterminal catalytic domain shaded in grey, and the C-terminal scaffolding domain represented by the open rectangle. The S1 domain is shown as a black box, and the arginine-rich RNA binding domain (ARRBD) by a vertically striped box. Blocks below RNase E illustrate regions of RNase E which interact with the degradosomal proteins. Amino acid residues of RNase E involved in each interaction are indicated to the right of each block. Modified from Vanzo, et al. (1998).  16  N-terminal 498 amino acid residues retained the catalytic activity of RNase E, albeit with reduced efficiency. Further deletion studies revealed that the first 321 residues did not retain activity (McDowall and Cohen, 1996). Only the N-terminal 450 amino acid residues of Rne are required to support growth (Ow et al., 2000). These results imply that the N-terminal half of RNase E contains the catalytic activity, while the C-terminal half is important for interactions within the degradosome. A detailed deletion analysis (Vanzo et al., 1998) of RNase E revealed the important features of the C-terminal domain, outlined in Figure 1.4. Coimmunopurification, far Western blotting and the yeast two-hybrid system were all employed to delineate the different protein-protein interactions. RhIB was shown indirectly to interact with residues 734-738, PNPase with residues 844-1045 and enolase with 739-845. Thus, the C-terminal domain of RNase E, while not required for endoribonucleolytic function, serves as a scaffold that is required for formation of the degradosome (Vanzo et al., 1998). Two regions mediating RNase E self-interaction were also found between residues 1 and 528 and between residues 500 and 752. The locations of the two potential RNAbinding domains are indicated in Fig. 1.4. The first potential binding domain, a previously characterized arginine-rich RNA binding domain (ARRBD; Cormack et al., 1993; McDowall and Cohen, 1996; Taraseviciene et al., 1995) lies between residues 600 and 684. More precise analysis places this between residues 608 and 632 (Miao, 2002). A second potential RNA-binding domain, the S1 domain, is located between residues 35 and 135 (Bycroft et al., 1997).  17  This domain was discovered in ribosomal protein S1 (reviewed in Coburn and Mackie (1999)), and will be discussed further in Chapters 3. A number of investigations have been performed to determine the significance of the C-terminal domain for RNase E activity. Although still not definitive, these studies revealed many potential modulators of RNase E activity. Removal of the ARRBD inhibited the activity of RNase E on some substrates (Kaberdin et al., 2000; Leroy et al., 2002; Ow et al., 2000), as did removal of residues 734-1045, which are required for formation of the degradosome. Likewise, removal of the RhIB and enolase binding region (residues 728 to 845) led to stabilization of a 11-lacZ reporter (which is normally very sensitive to cleavage by RNase E). Removal of the ARRBD, RhIB and enolase-binding region led to the greatest increase in stabilization (over six fold) of the J7-lacZ mRNA. Somewhat surprisingly, however, removal of the PNPase binding region (residues 844 to 1045) destabilized the 11-lacZ model substrate. This was hypothesized to be due to the removal of an adjacent acidic domain, which may inhibit the RNA-binding domains (Leroy et al., 2002). From these data, it can be concluded that disruption of degradosomal protein contacts, as well as loss of direct RNase E-mRNA interactions, subtly influence mRNA decay. Intracellular localization of the degradosome  RNase E activity has been detected in E. coli inner membrane preparations (Miczak et al., 1991). This implies that RNase E localizes to the inner membrane. Further indirect evidence for the mRNA decay apparatus associating with the membrane was found in a screen for temperature-sensitive alleles of 18  pnp. Inactivation of a gene identified as mrsC was found to stabilize mRNA. An mrsC/rne-1 strain was examined, and found to display much greater stabilization  of mRNA than in strains with single mutations at non-permissive temperature (Granger et al., 1998). The product of the mrsC gene was subsequently identified as hfIB, which encodes a membrane-associated ATP-dependent protease (Wang et al., 1998). Steege (2000) proposed that RNase E may associate with HfIB at the membrane, and that membrane localization may affect mRNA decay. Immunostaining experiments also provided direct evidence that both RNase E and RhIB localized mainly to the inner membrane in situ. Freezefracture electron microscopy showed that all four degradosomal proteins were present in membranes (Liou et al., 2001). A truncated RNase E, consisting of the N-terminal 603 amino acids was sufficient for membrane association, although this led to loss of RhIB's association with the membrane. Thus, it appears that RhIB, at least, localizes to the membrane indirectly via its association with the C-terminus of RNase E (Liou et at., 2001). The location of PNPase and enolase was difficult to determine, as only a small fraction of these proteins associates with RNase E in the degradosome. Free enolase and PNPase are located in the cytoplasm. Importance ofthe degradosome  In independent experiments, the C-terminal scaffolding domain of RNase E (approximately residues 600 to 1061) was eliminated, yet E. coli retained its viability (Kido et al., 1996; Ow et al., 2000). These strains, which no longer form 19  intact degradosomes, exhibited increased bulk mRNA stability. They were also unable to rescue rne-1 temperature sensitive mutants at non-permissive temperatures (Lopez et al., 1997; Ow et al., 2000). The processing of 5S rRNA is unaffected, possibly because rRNA is a higher affinity substrate than mRNA (Georgellis et al., 1992). The most compelling evidence, to date, for the degradosome playing an important role in cellular processes lies in the fact that strains that fail to form degradosomes due to truncation of RNase E cannot compete with wild-type strains when co-cultured (Leroy et al., 2002).  1.2.2. RNase G  RNase G is a member of a gene cluster originally identified as the mre region (murein pathway cluster e). The mre gene cluster (depicted in Fig. 1.5), located at 71 min on the E. coli chromosome, is composed of five proteins, and was deemed to be responsible for determination of rod shape and sensitivity to the antibiotic mecillinam (Wachi et al., 1987). The first gene of this cluster, mreB, encodes a protein with close similarity to ftsA, a protein involved in cell division (Doi et al., 1988). Overexpression of MreB resulted in filamentous growth (Wachi and Matsuhashi, 1989). The final gene in the cluster, originally termed orfF, encodes a 489 aa, 55 kDa protein (Wachi et al., 1991). The OrfF protein  was consequently found to share sequence homology with RNase E (McDowall et al., 1993). Overproduction of the protein encoded by orfF led to the formation  of chained cells and minicells, along with anucleated cells. Significantly, these cells contained long filamentous axial bundles, leading to renaming of the gene as cafA (cytoplasmic axial filaments; Okada et al., 1994). 20  rng  55 kDa  orfE  mreD  mreC  mreB  J ^ < = = = r < ^ = r - / / 22 kDa 19 kDa 40 kDa 37 kDa  Figure 1.5. The mre gene cluster. Chromosomal DNA is indicated by a horizontal line, with DNA extending both upstream and downstream of the cluster. The five mre genes are represented by arrows, indicating the direction of transcription. Identification of the genes is given above the corresponding arrow, and the size of each gene product is indicated below. The black arrow represents rng, which encodes RNase G. This figure is adapted from Wachi et al. (1997).  21 Identification of RNase G as an endonuclease  Ribosomal RNA transcription units encode the genes for 16S, 23S and 5S rRNAs, along with several tRNAs, as shown in Fig. 1.6. Mature rRNA is produced from the 30S primary transcript through a series of endo- and exoribonucleolytic processing events. RNase III, a double-strand specific endoribonuclease, cleaves the 30S transcript releasing 17S RNA. In turn, 17S RNA is processed at both its 5' and 3' ends to form mature 16S rRNA (Li et al., 1999b). There are seven rRNA gene clusters in E. coli, and all consist of two promoters, followed by 16S rRNA, one or two tRNAs, 23S rRNA, 5S rRNA. Some clusters encode another tRNA following 5S rRNA and the cluster ends with a terminator (for reviews, see Apirion and Miczak, 1993; Srivastava and Schlessinger, 1991). Folding of the pre-rRNA and binding of ribosomal proteins occurs co-transcriptionally. The initial step in processing involves two RNase III cleavages, releasing 17S and 23S pre-rRNA and 9S RNA (reviewed by Apirion and Miczak, 1993)). The 3'-end of 17S RNA is then processed by an unknown enzyme to yield 16.5S RNA. This step is slowed considerably when RNase E is inactivated, and the data imply that 3'-maturation is the result of a single endonucleolytic cleavage (Li etal., 1999b). The pre-16S rRNA is then cleaved in the 5'-externally transcribed spacer (ETS) region by RNase E, producing 16.3S RNA. Finally, the mature 5'-end of 16S rRNA is produced by an endonucleolytic cleavage by RNase G (Li etal., 1999b; Wachi etal., 1999).  22  Figure 1.6.  Processing of 16S rRNA.  This figure schematically depicts the 30S  rRNA precursor. Mature 16S rRNA is represented by a box with diagonal stripes, tRNA by an open box, 23S rRNA by a shaded box and 5S rRNA by a black box. Known processing sites are indicated by arrows, with III representing RNase III cleavages; E, RNase E; G, RNase G; P, RNase P; T, RNase T and exo representing cleavages by RNase II, PNPase, RNase BN, D, PH or T. Cleavages  made by unknown  ribonucleases are identified by "?".  23  24  Two cleavages are made to form the mature 5'-end of tRNA (Fig. 1.6), although there are exceptions. Typically, RNase E makes the initiating cleavage 5' to the mature 5'-end, which affects the rates of subsequent steps (Li and Deutscher, 2002; Ow and Kushner, 2002). The mature 5'-end is the result of a single endonucleolytic cut by RNase P (for recent review, see Gopalan et al. (2002)). The 3'-end can be processed by a combination of the 3'-5' exoribonucleases RNases BN, D, PH, ll,T and PNPase, with RNases PH and T forming the mature 3'-end most efficiently (Li and Deutscher, 1996). RNase E cleaves 9S RNA twice releasing pre-5S rRNA (Misra and Apirion, 1979). This is further processed by RNase T to form its mature 3'-end (Li and Deutscher, 1995) and by an unidentified nuclease to form the mature 5'end (Fig. 1.6). RNase T is also responsible for forming the mature 3'-end of 23S rRNA, among others (Li et al., 1999a), although the activity that forms the mature 5'-end remains unidentified. Over two decades ago, a mutant was isolated that accumulated 16.3S rRNA containing approximately 60 additional nucleotides at its 5'-end (Dahlberg et al., 1978). It appeared as if some residual mature 16S rRNA could still form in this mutant that displayed increased temperature-sensitivity. The 16.3S rRNA retained the ability to associate with all 21 small subunit ribosomal proteins, and ribosome function did not appear to be impaired in the mutant strain. A fraction from a cell lysate contained an endoribonucleolytic activity that could restore 16S rRNA processing. Since the impairment appeared to be due to an inactivated endoribonuclease, the strain was designated as BUMMER (Brown University Medical Mutant Endo-Ribonuclease; Dahlberg etai., 1978). 25  Wachi et al. (1999) discovered that immature 16S rRNA also accumulated in a cafA::cat mutant. This defect could be complemented by introducing cafA on a low copy plasmid (pMEL2). This demonstrated that the defect was due to disruption of the cafA gene, rather than the flanking genes. The immature precursor contained 66 extra nts at its 5'-end (Li et al., 1999b; Wachi et al., 1999). In view of the similar accumulation of 16.3S rRNA observed in both strains, it was reasoned that a mutation in cafA may be responsible for the altered 16S rRNA processing observed in the BUMMER strain. To test this idea, the plasmid pMEL2 (which contains intact cafA) was introduced into the BUMMER strain and was capable of complementing the mutation. The cafA gene in the BUMMER strain was sequenced and found to contain an 11 base pair deletion that led to a truncation at amino acid residue 267 of 489 (Wachi et al., 1999). The role of RNase G in 16S rRNA maturation was further defined by Li et al. (1999). In a cafA mutant, 16.3S rRNA was slowly converted to a product with  4 or 5 extra nts at the 5'-end, presumably due to cleavage by RNase E. In an rne-1, cafA  +  strain, mature 16S rRNA could form very slowly without a transient  accumulation of 16.3S rRNA. This indicated that RNase G could cleave the prerRNA at the mature 5'-end directly, without the participation of RNase E, but that the presence of RNase E greatly increased the rate of 16S rRNA maturation (Li etal., 1999b).  26 Role for RNase G In mRNA decay  Functional RNase G is not a requirement for cell viability (Wachi et al., 1997); therefore, it is unlikely to play an important role in overall mRNA metabolism. The similarity of RNase G to RNase E, and the fact that it behaves as an endonuclease, however, beg the question of whether RNase G may affect the decay of some mRNAs. To screen for potential targets of RNase G, total cellular proteins were isolated from an rngr.cat strain and compared to total proteins from a wild-type strain. Two proteins, enolase and alcohol dehydrogenase, were moderately overexpressed in the rngr.cat strain. This was due to increased stability of their mRNAs and appeared to be independent of RNase E, as no increase in protein levels was observed for either protein in an rne-1 mutation at non-permissive temperature (Umitsuki et al., 2001, Kaga et al., 2002). Microarray analysis of mRNAs from strains that were depleted of RNase E, deleted for RNase G or which overexpressed RNase G confirmed that abundances of enolase {end) and alcohol dehydrogenase {adhE) mRNA were controlled by RNase G (Lee et al., 2002). In addition to eno and adhE, nine other mRNAs with steady-state levels dependent upon RNase G were identified; of these eleven mRNAs, five are associated with utilization of energy sources {eno, adhE, pgi (glucosephosphate isomerase 1), ndh (respiratory NADH  dehydrogenase) and tpiA (triosephosphate isomerase 1)). The significance of this observation is unclear, but suggests that RNase G may play a specific role in RNA metabolism under certain growth conditions (Lee et al., 2002).  27 Comparison of the activities and specificities of RNase G and RNase E  RNase G displays 35% sequence identity (50% similarity) with the catalytic domain of RNase E (McDowall et al., 1993) and both act as endoribonucleases (Wachi et al., 1999). It was inevitable, therefore, that the activity of these two enzymes be compared. RNase G was purified, and its activity against model RNA substrates tested (Jiang et al., 2000; Tock et al., 2000). Based on these studies, it was concluded that RNase G can cleave both ompA mRNA and RNA 1, as well as substrates based on the RNase E cleavage site of RNA 1 (Jiang et al., 2000; Tock et al., 2000). It is significant to note, however, that while RNase G also displays a propensity for cleavage in AU-rich sequences, the actual site of cleavage is not identical to that catalyzed by RNase E. RNase G cleaves an oligonucleotide based on the 5'-end of RNA 1 not only at the RNase E site, but also at sites one or two nucleotides 3' to the expected cleavage (Tock et al., 2000). RNase G is also ineffective at cleaving 9S RNA (see Fig. 3.13a) at the "a" site. RNase G, however, is capable of cleaving an oligonucleotide based on the "a" site of 9S RNA in vitro. This indicates that RNase G, like RNase E, is sensitive to secondary structure (Tock et al., 2000). An important feature of RNase E is that it autoregulates its own accumulation. This was illustrated, in part, by constructing a reporter fusion gene that comprised the 5'-UTR and first 128 codons of rne with lacZ (Jain and Belasco, 1995b). As depicted in Fig. 1.7, this construct was introduced into the chromosome of a temperature-sensitive RNase E strain. In this system, (328  Figure 1.7. rne-lacZ fusion reporter gene for examining RNase E  autoregulation. The 5'-UTR and first 128 codons of rne (diagonally striped box) were fused with lacZ (shaded box) which encodes pgalactosidase. This was introduced into the chromosome of a strain encoding the temperature-sensitive mutant RNase E, ams-1 (pixilated box). Plasmids encoding proteins of interest were introduced into the cells and tested for their ability to regulate pgalactosidase activity at non-permissive temperatures (Jian and Belasco, 1995; Jiang etai., 2000).  29  galactosidase activity was inversely proportional to autoregulation. By introducing plasmids expressing different nucleases or regions of RNase E, other proteins which could potentially regulate RNase E were identified. This approach was used to determine if RNase G could also regulate RNase E (Jiang et al., 2000). Plasmids encoding full-length RNase E, RNase G or the first 498 amino acids of RNase E, which cannot form degradosomes, were tested for their ability to restore or mimic RNase E autoregulation at non-permissive temperatures. Neither RNase G nor RNase E 1-498 could repress the reporter p-galactosidase activity, which implies that the C-terminal domain of RNase E is essential for autoregulation. Interestingly, two other RNases, RNase III (Bardwell et al., 1989) and PNPase (Robert-Le Meur and Portier, 1992) also regulate their own synthesis. RNase E displays a strong 5'-end dependence for monophosphorylated over triphosphorylated substrates (Mackie, 1998). The activity responsible for 5'end dependence is localized to the N-terminal 498 amino acids of RNase E (Jiang et al., 2000). RNase G activity on mono- and triphosphorylated model RNA substrates was examined in vitro. RNase G, like RNase E, was found to strongly  prefer  5'-monophosphorylated  substrates to  triphosphorylated  substrates (Jiang et al., 2000). In contrast to 5'-phosphorylated substrates, 5'unphosphorylated substrates were impervious to cleavage (Tock et al., 2000). This property will be developed further in Chapter 6. The phosphorylation state of the 3'-end had no effect on endonuclease activity (Tock et al., 2000). Since  30  RNase G also displays 5'-end dependence, it is useful as a model to investigate the mechanism(s) that underly this property. In three separate studies, the ability of RNase G to rescue an RNase E mutation was investigated (Jiang et al., 2000; Lee et al., 2002; Wachi et al., 1997). It was initially noted that introduction of cafA::cat to an  ams-1  strain  increased the resultant strain's temperature-sensitivity (Wachi et al., 1997). While ams-1 could grow at 30°C and 39°C, the double mutant was incapable of growth at 39°C. This deficiency was reversed by introduction of the rng gene on a plasmid, showing that RNase G was capable of partially suppressing, but not complementing, the ams-1 temperature-sensitive allele of rne. These results were confirmed by Jiang et al. (2000), who found that RNase G, even at six-fold elevated levels relative to normal, could not rescue an RNase E mutant. Both of these studies utilized rng with an intact 5'-leader sequence. In contrast, a separate study found that 50-fold overexpression of rng could complement an RNase E mutation (Lee et al., 2002). This study, however, engineered the rng gene to select an upstream, unused start codon (see Chapter 3) that resulted in insertion of six additional amino acids at the N-terminus. Experiments in J. Belasco's lab have found that the presence of the additional 6 amino acids at the N-terminus allows RNase G to complement RNase E mutants (personal communication). Therefore, it is important to unambiguously identify the translational start site of rng to determine if RNase G is capable of rescuing an RNase E mutant.  31  1.2.3. Other endonucleases  The third endoribonuclease that plays a role in the decay of a small set of mRNAs is the double strand-specific RNase III (see Table 1.1). RNase III is mainly involved in rRNA processing, specifically in the processing of pre-16S and pre-23S rRNA from a 30S precursor. Surprisingly, disruption of RNase III function has very little effect on bulk mRNA decay (for reviews, see Coburn and Mackie, 1999; Grunberg-Manago, 1999; Steege, 2000; Condon and Putzer, 2002) or viability (Apirion and Gitelman, 1980; Babitzke et al., 1993). In an interesting twist, RNase III processing of the adhE mRNA is required for efficient translation of the message (Aristarkhov et al., 1996). Thus it appears that two of the RNA targets of RNase G identified to date are initially processed by RNase III. RNase P is an essential enzyme that creates the mature 5'-end of tRNA through an endonucleolytic cleavage (for recent review, see Condon and Putzer, 2002). The active ribonucleoprotein consists of a small protein component, involved in substrate recognition, and a catalytic RNA of 377 nts called M1 RNA. Recent work has revealed that RNase E may play a role in the 3'-end processing of the catalytic M1 RNA (Sim etal., 2001). The remaining endoribonucleases found in Table 1.1 are RNase I, a periplasmic endoribonuclease which acts as an RNA scavenger, RNases H and HI, which remove RNA primers following DNA replication (reviewed by Coburn and Mackie, 1999; Condon and Putzer, 2002) and Rel E. Rel E is the most recently identified endoribonuclease, and it appears to cleave mRNA in the ribosomal A site (Pedersen et al., 2003). There may be other, as yet unidentified 32  endoribonucleases, as the activities that generate the 5'-end of 5S and 23S rRNA and the 3'-end of 16S rRNA are unknown (see section, above).  1.3. Exonucleases  Exoribonucleases remove single nucleotides successively from the ends of RNA (see Fig. 1.1). Eight exoribonucleases have been identified in  E. coli,  and are  listed in Table 1.1. All eight degrade RNA in the 3' to 5' direction. Eukaryotes, in contrast, encode both 3-5' and 5'-3' exoribonucleases. Of the eight exoribonucleases  in E.  coli,  only  three,  PNPase,  RNase  II and  oligoribonuclease, are involved in mRNA decay (for recent reviews, see Condon and Putzer, 2002; Zuo and Deutscher, 2001).  1.3.1. Exoribonucleases involved in mRNA decay  Polynucleotide phosphorylase (PNPase) is a 711 residue protein originally identified as an RNA polymerase, but now recognized as an exoribonuclease that cleaves RNA phosphorylytically, releasing nucleoside diphosphates. PNPase is a trimer, whose x-ray structure has recently been determined (Symmons et al., 2000). PNPase contains four identifiable domains: a PH domain (region similar to RNase PH), a PH' domain, which is a diverged PH-like domain, as well as KH and S1 domains, both of which are RNA binding domains (for reviews, see Coburn and Mackie, 1999; Grunberg-Manago, 1999; Steege, 2000; Zuo and Deutscher, 2001). Of particular interest to this project, however, was the discovery that PNPase associates with RNase E in the degradosome (Carpousis et al., 1994). This association will be addressed further in Chapter 4. 33  RNase II is a 72.3 kDa hydrolytic exoribonuclease. An investigation into the overall contribution of RNase II to exoribonucleolytic decay revealed that RNase ll-associated hydrolytic decay accounts for 90% of the exonuclease activity in E.  coli  (Deutscher and Reuven, 1991), despite the fact that it cannot  associate with RNase E in the degradosome. Mutations in either PNPase or RNase II alone do not lead to lethality. Double mutants, however, are inviable (Donovan and Kushner, 1986). The activity of both enzymes is strongly inhibited by secondary structure of the substrate. This will be addressed in Section 1.4. Neither RNase II nor PNPase is capable of completely degrading an RNA as they lose processivity and dissociate from substrates smaller than 12 to 15 residues. The resultant limit oligonucleotides are, in turn, degraded by the essential enzyme, oligoribonuclease (Ghosh and Deutscher, 1999).  1.3.2. Revised model of mRNA decay  A model to explain mRNA decay through secondary structures is presented in Figure 1.8. For simplicity, RNase E is represented as a monomer and only additional aspects beyond the 5'-tethering model are shown. In this model, mRNA decay is initiated by RNase E. The newly formed cleavage products are now susceptible to exoribonucleolytic attack by either PNPase, either free or associated with the degradosome (as shown), or by RNase II. Following the initial  endonucleolytic  cleavage,  monophosphorylated, which  the  is highly  distal preferred  RNA  product  over the  is  5'-  initial 5'-  triphosphorylated end (Mackie, 1998). Consequently, the distal product is highly preferred for subsequent cleavages by RNase E, resulting in the "all-or-none" 34  Figure 1.8.  Current model of mRNA decay.  mRNA is represented as a solid  black line, with the 5' and 3' ends indicated, as well as the 5'-triphosphate group. Secondary structure is indicated in the form of a hairpin, with a double-stranded region and a loop near the 3'-end. In the degradosome, RNase E is indicated by a shaded circle, RhIB by a black triangle and PNPase a trimer of striped circles. Enolase has been omitted from the degradosome for simplicity. The 3'-5' direction of exonucleolytic activity is indicated by a horizontal arrow, and the string of A's represents a poly(A) tail. These models originally appeared in Coburn and Mackie (1999) and Coburn etai. (1999).  35  initial cleavage  SLOW  PPP-  /  iL_.  i  subsequent cleavages  FAST  O  5' (PP)P-  I  3'  exonuclease activity P N P a s e , R N a s e II  limit oligonucleotides  \ 5L  3'  51  oligoribonuclease  FREE NUCLEOTIDES exonuclease stalling  36  phenomenon. This phenomenon describes the lack of observable mRNA decay intermediates in E.  coli  (Coburn and Mackie, 1999). Fig. 1.8 also depicts how  the exonucleases, RNase II and PNPase, respond to secondary structure (Coburn and Mackie, 1998; Coburn etal., 1999). In the first pathway (Fig. 1.7a), secondary structure may be unwound by the RNA helicase RhIB, dependent on ATP hydrolysis and RNase E. In the second pathway (Fig. 1.8b), the RNA undergoes one or several rounds of polyadenylation, catalyzed by Poly(A) polymerase (PAP 1). This provides a single-stranded RNA platform for binding of RNase II, PNPase or the degradosome (shown in Fig. 1.8). In the absence of PAP 1, some polyadenylation still occurs  in vivo,  and this may be due to the  synthetic activity of PNPase (Mohanty and Kushner, 2000). Indeed, a recent study has revealed that PNPase is the primary poly(A) polymerase in chloroplasts (Ruth et al., 2003).  1.3.3. Other exoribonucleases  The other exoribonucleases from E.  coli  are listed in Table 1.1. RNase R, the  product of a gene originally identified for its involvement in the virulence of Shigella  and enteropathic  E. coli,  is not essential, although RNase R/PNPase  double mutants are reportedly inviable (Cheng et al., 1998). The sequence of the gene encoding RNase R (rnr) is homologous to rnb, which encodes RNase II (see above; Cheng et al., 1998). RNase R is non-specific, and may be involved in degrading  rRNA (Cheng and Deutscher, 2002). The remaining  exoribonucleases, RNase BN, D, PH and T, are all involved in tRNA processing. There is considerable functional redundancy among these enzymes. Mutational 37  inactivation of these enzymes is only lethal when also combined with an RNase II mutant. Viability could be restored and maintained by any of these five exonucleases separately (Kelly and Deutscher, 1992). RNase T is also the enzyme responsible for formation of the mature 3'-ends of 5S and 23S rRNA (Li and Deutscher, 1995; Li etai., 1999a).  1.4. Aims of this project  This thesis has been divided into three chapters of Results. Chapter 3 will deal with RNase G as a surrogate for RNase E activity. I developed simple purification techniques for both His-tagged and untagged versions of RNase G, and explored its biochemical properties. I determined the multimeric state of RNase G, and the effects of multimerization on its activity. Finally, I reconstituted a model RNP substrate that was processed by the enzymatic interaction of RNase E and RNase G. Chapter 4 examines relationships within the degradosome, as well as the activity of RNase E. Specifically, this research established that the degradosome likely does not dissociate and reform following each cycle of attack on an mRNA. Potential 5'-phosphatase activity of RNase E was also investigated. RNase E was not found to possess 5'-phosphatase activity. In Chapter 5, I explored the specificity of recognition of RNA by both RNase E and RNase G and their ability to cleave potential sites. The requirements for 2'-OH groups for recognition and cleavage were established, and the search for a strong inhibitor of cleavage was sought.  38  2. MATERIALS AND METHODS 2.1. Source of reagents  All chemicals were purchased from commercial sources and were of reagent grade.  2.2. Bacterial strains and culture techniques  Luria-Bertani broth (LB; 86 mM NaCl, 0.5% w/v yeast extract, 1% w/v peptone) was the rich medium used for growing liquid cultures, while agar plates (LB supplemented with 2% w/v agar) were utilized as solid media. Bacterial cultures grown in minimal media utilized M9ZB (42 mM Na HP0 »H 0, 22 mM KH P0 , 2  4  2  2  4  18 mMI NH CI, 1% (w/v) N-Z-amine A, 86 mM NaCl). Bacterial manipulations 4  were performed aseptically, as described (Sambrook, 1989). Bacterial strains used are listed in Table 2.1.  2.3. Common buffers  RNase Assay Buffer- 25 mM Tris-HCl pH 7.5, 100 mM NH CI, 60 mM KCI, 5 mM 4  MgCI , 0.1 mM DTT, 0.1 mM EDTA, 5% glycerol, pH 7.7 2  Buffer I- 25 mM Tris-HCl pH 7.6, 50 mM NaCl, 7 mM fresh (5-mercaptoethanol Buffer II- 25 mM Tris-HCl pH 7.5, 50 mM NaCl, 5 mM fresh DTT Buffer III- (1 x One-Phor-AII, Pharmacia) 10 mM Tris-acetate pH 7.9, 10 mM Mgacetate, 50 mM K-acetate  39  Table 2.1. Bacterial strains  Strain  Genotype  BL21(DE3)  F" ompT nsdS (r ", m ") gal, dcm A(DE3)  CF881  F" Alac argA trp recB1009 A(xthA -pnc)Arna  DH5a  F" OdlacZAMW A(lacZYA -argF) U169 deoR recA1 endA1  e  B  B  hsdR17(rK-, mK+) phoA supE44 A- thi-1 gyrA96 relA1  JM110  rpsL (Str) thr leu thi-1 lacY galK galT ara touA tsx dam dcm r  supE44 A(lac -proAB)[F' traD36 proAB lacl°Z AM15]  RD100  rna pnp-13 sup met relA trpD9778 lacZ  40  Buffer IV- 50 mM Tris-HCl pH 7.5, 100 mM NaCl, 5% glycerol, 1 mM DTT, 1 mM PMSF, 2 ug/ml aprotinin and 0.8 ug/ml each of leupeptin and pepstatin A (Sigma). Buffer V- 10 mM Tris-HCl pH 7.5, 5% glycerol, 0.5% genepol-X80, 1 mM EDTA) with 50 mM NaCl, and PMSF, aprotinin, pepstatin A and leupeptin (at the same concentrations as Buffer IV, listed above). Buffer D1- 100 mM Tris-HCl pH 7.5, 10% glycerol, 128 mM NH CI, 0.2 mM 4  EDTA, 1 mM MgCI 0.1 mM DTT, 0.2 mM PMSF. 2>  Buffer D2- 100 mM Tris-HCl pH 7.5, 10% glycerol, 68 mM NH CI, 0.2 mM 4  EDTA, 1 mM MgCI , 0.1 mM DTT, 0.2 mM PMSF. 2  Buffer D3- 50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1 mM DTT, 0.1 mM EDTA, 20% glycerol. Tris/Boric acid/EDTA (TBE) buffer- 90 mM Tris, 90 mM boric acid, 2 mM EDTA PTBN- 20 mM Na-phosphate pH 7.0, 0.05% Tween 20, 0.1 mM bovine serum albumin, 0.85% NaCl, 1 mM NaN . 3  Phosphate Buffered Saline (PBS)- 137 mM NaCl, 2.7 mM KCI, 10 mM Na HP0 , 1.8 mM KH P0 , pH to 7.4 with HCI. 2  4  2.4. Polymerase  2  4  chain reaction  DNA was amplified using the polymerase chain reaction (Mullis etai., 1986). For site-directed mutations, the Stratagene QuikChange kit, which employs PfuTurbo DNA polymerase. Taq DNA polymerase (Life Technologies) was used  for all other applications. The manufacturers' protocols were followed for both  41  enzymes, with incubations performed in a Perkin Elmer GeneAmp 2400 thermocycler.  2.5. Other molecular biological methods  Standard molecular biology techniques, including restriction endonuclease treatments, ligation, subcloning, agarose gel electrophoresis and generation of RNase-free water were performed as previously described, unless otherwise noted (Sambrook, 1989).  2.6. Enzyme purifications  Note: for all purifications, the ratio of protein of interest to contaminant was determined by SDS-polyacrylamide gel electrophoresis, which is described in Section 2.6.1.  2.6.1. Untagged RNase G  BL21 (DE3) transformed with pDB6 (strain DB6) resulting in strain DB6. DB6 was grown at 37° in 600 ml LB with ampicillin (100 ug/ml) to an Aoo of 0.4 - 0.6 6  and then for approximately 16 hours at 20° with 0.1 mM IPTG. Cells were harvested by centrifugation and resuspended in 5 ml Buffer II (section 2.3), with aprotinin, leupeptin, pepstatin A (all to a final concentration of 0.02 ug/ml). Cells were lysed at 12,000 psi in a French Pressure cell (Aminco) and one additional volume of Buffer II was added. The diluted lysate was incubated on ice for 20 min with1.5 ug/ul DNase I. Lysates were cleared by centrifugation at 30,000 x g 42  to form an S30 (Fraction 2) and a pellet (Fraction 1), which was resuspended in one volume (this and subsequent volumes are relative to the S30) of Buffer II. RNase G was precipitated from the S30 (Fraction 2) by the addition of 26% w/v ammonium sulphate followed by centrifugation at 17,500 x g. Precipitated proteins were taken up in approximately 10 volumes of Buffer II (to give Fraction 3) and loaded onto a 25 ml Heparin-agarose column. Proteins were eluted with a gradient of 50 mM to 2 M NaCl in Buffer II. Fractions containing RNase G were pooled and diluted 1:1 with 25 mM Tris-HCl pH 7.6, 5 mM DTT (forming Fraction 4) prior to loading on a 1 ml Mono Q column (Pharmacia). Proteins were eluted with a gradient (25 mM Tris-HCl pH 7.6, 5 mM DTT, 50 mM - 2 M NaCl). Fractions containing RNase G were pooled (Fraction 5), divided into aliquots, frozen in liquid N and stored at -70°. Protein was quantified by staining with 2  SYPRO Red (Pharmacia) and analysis on ImageQuant software (Molecular Dynamics).  2.6.2. RNase G mutants  Cysteine to serine mutations were introduced into C-terminally His-tagged RNase G using the PCR-based Stratagene QuickChange kit. The six cysteines of RNase G are located at residues 79, 162, 402, 405, 408 and 421 (Fig. 3.1). Templates and primers used for mutagenesis are listed in Table 2.2. Plasmids designated pDB were constructed by myself, while those designated pJH were made by Janet S. Hankins. To perform the mutagenesis, 0.1-1 ng/ul template DNA, 25 ng/ul each of forward and reverse primers, 0.3 ng/ul each of ATP, CTP,  43  Table 2.2. Primers and templates used for construction of RNase G clones  Plasmid  Mutation  Template  Primer Sequence  pDB6  wild-type  chromosomal  For-CCGTAGTCGGATCCCCGCTGGTTG  DNA pDB1  pDB4  C-terminal  His chromosomal  tag  DNA  AS1  pDB1  Rev-CTCAAAAACCCGGATCCGGATGGCGG For-CCGTAGTCGGATCCCCGCTGGTTG Rev-CGCGGATCCTCGAGCATCATTACGACGTCAAACTGC ForCATATTGAACGTGAGGCGCGACGCGCTTCTCACGTTGGGGTTT  (A36-135)  cc RevGGAAACCCCAACGTGAGAAGCGCGTCGCGCCTCACGTTCAAT ATG pJH1b  C421S  pDB1  For-CCGTGGAAACGGTAAGCTATGAAATCATGCGCGAG Rev-CTCGCGCATGATTTCATAGCTTACCGTTTCCACGG  pJH4  C402S  pDB1  For-GCACGTACTGTCTAACGAATGCCC Rev-GGGCATTCGTTAGACAGTACGTGC  pJH5  C162S  pDB1  For-CGCAGAGTATTCCGAACGAGCAGGG Rev-CCCTGCTCGTCGGAATACTCTGCG  pDB5  C402S, C405S  pDB1  For-CTGTGTAACGAATCCCCAACCTCCCACGGTCGCGG Rev-CCGCGACCGTGGGAGGTTGGGGATTCGTTACACAG  pJH3  C402S, C421S  C405S,pDB5  For-GCACGTACTGTCTAACGAAATCCC Rev-GGGATTCGTTAGACAGTACGTGC  44  GTP and CTP, were mixed with 0.05 U/ul PfuTurbo DNA polymerase (Stratagene) in 10 mM KCI, 10 mM (NH )2S0 , 20 mM Tris-HCl pH 8.8, 2 mM 4  4  MgS0 , 0.1% Triton X-100 and 0.1 ug/ul bovine serum albumin. As directed by 4  the manufacturer's instructions, 18 cycles of PCR were performed on a Perkin Elmer GeneAmp 2400. Products of PCR (5 pi each in 0.5 pi 50% glycerol, 50 mM EDTA, 0.25% bromophenol blue, 0.25% xylene cyanol) were analyzed on a 0.8% agarose gel in TBE buffer (section 2.3). Successfully amplified products were treated with 5U Dpnl (Stratagene) at 37°C for 1 hr to remove template DNA. The mutant plasmids were then transformed into Gibco Subcloning Efficiency DH5a Chemically Competent Cells, following the manufacturer's instructions. Plasmids were isolated from DH5a clones, and the mutation confirmed by sequencing or restriction enzyme analysis. The desired plasmids were transformed into BL21(DE3) cells, to allow protein overexpression. RNase G from mutant strains was purified and quantified as described for wild-type Histagged RNase G (Section 2.6.3).  2.6.3. His-tagged RNase G  Plasmid pDB1 was transformed into E. coli BL21(DE3) to form strain DB1. Purification of His-tagged RNase G was similar to the method of Jiang, et al (Jiang et al., 2000). In outline, 2 I cultures of DB1 were grown at 37°C in LB containing 75 u.g/ml kanamycin to an Aoo of 0.4 - 0.6 and then overnight at 20°C 6  with 0.1 mM IPTG. Cells were harvested at 3,000 x g, resuspended in 10 ml binding buffer (25 mM Tris-HCl pH 7.6, 500 mM NaCl, 7 mM p-mercaptoethanol) 45  with 0.5 mM PMSF, and lysed at 12,000 psi in a French Press. Lysates were diluted with one additional volume of binding buffer and treated with 1.5 ug/ul DNase I. Insoluble material was removed by centrifugation at 30,000 x g and the resulting supernatant (S30) was diluted with one volume binding buffer. The dilute S30 was mixed with an appropriate volume (2 to 2.5 ml) of Talon resin (Clonetech) in a 50 ml screw-cap tube (Falcon) for 30 min at 4°C on a Nutator orbital shaker (Rose Scientific). Following binding, the resin was poured into a 10 ml column and washed with 3 volumes of binding buffer containing 5 mM imidazole. Proteins were eluted in three steps with four column volumes each of Buffer I (section 2.3) containing 10 mM, 50 mM or 500 mM imidazole. Elution fractions 2 and 3 (50 mM and 500 mM imidazole respectively) were pooled and loaded onto a 15 ml SOURCE Q column (Pharmacia) in Buffer I. RNase G was eluted with a 50 mM - 2 M NaCl gradient in the same buffer. Fractions containing the highest ratio of RNase G to contaminants were pooled, aliquots were frozen in liquid N , and stored at -70°C. Protein preparations were 2  quantified by staining with SYPRO Red (Pharmacia) and analysis on ImageQuant software (Molecular Dynamics).  2.6.4. Purification  ofpnp13  degradosomes  Purification of pnp13 degradosomes has been previously described (Coburn et al., 1999). E. coli strain RD100 (constructed by R. P. Dottin, a gift from M. L.  Pearson) was grown overnight by the Fermentation Pilot Plant Facility, Biotechnology Laboratory, University of British Columbia. The resulting cell  46  pellet (approximately 25 g) was resuspended in 50 ml Buffer IV (section 2.3) with 3 mM EDTA, 1.5 mg/ml lysozyme (Sigma), and incubated on ice for 70 min with intermittent stirring. An additional 20 ml Buffer IV with 30 mM Mg-acetate, 3% Triton X-100 and 20 ug/ml DNase I (Sigma) were added, followed by 4° C incubation on ice. NH CI was added to a final concentration of 1M and the 4  sample was centrifuged for at 4°C for 60 min at 30,000 x g. The resulting supernatant (S30) was centrifuged at 200,000 x g at 4°C for 2 hr in a Ti 60 rotor (Beckmann). Degradosomes were precipitated from the supernatant (S200) by addition of 26% w/v ammonium sulphate with stirring on ice for 40 min and centrifugation at 10,000 x g at 4°C for 1 hr. The AS26 pellet was resuspended in Buffer V (section 2.3). The preparation was loaded onto a 25 ml SP-Sepharose FF column (BioRad) followed by two stepwise washes with four column volumes each of Buffer V with 50 mM NaCl or 300 mM NaCl. Proteins were eluted with Buffer V containing 1% genepol-X80 and 1 M NaCl and fractions collected. Fractions with the highest ratio of degradosomal proteins to contaminants (as determined by SDS-polyacrylamide gel electrophoresis followed by staining with Coomassie Brilliant Blue R-250) were pooled and loaded onto a 10 ml Affi-Blue gel column (BioRad). The column was washed with 10 volumes of Buffer V with 50 mM KCI. Proteins were eluted in Buffer V with a 300 mM to 3 M KCI gradient, and fractions collected. Fractions with the highest ratio of degradsosomal proteins to contaminants were pooled, and concentrated by precipitation with 26% ammonium sulphate, as previously described. The pellet was taken up in 1 ml Buffer V with 50 mM NaCl and loaded onto a 100 ml Bio-Gel A5m column (BioRad). Fractions with the highest ratio of degradosomal proteins to 47  contaminants were pooled and concentrated to approximately 0.5 mg/ml in a 15 ml Ultrafree Biomax-5K centrifugal filter device (Millipore).  2.6.5. Electropurification of RNase E  Electroelution of RNase E was performed as previously described (Cormack et al., 1993) with some modifications. Briefly, 500 ml of M9ZB containing 50 ug/ul ampicillin was inoculated with strain GM402, which expresses full-length RNase E under the control of a T7 promoter. The culture was grown to an Aoo of 0.4, 6  then induced with 1 mM IPTG and shaken at 30° for 5 hrs. Cells were pelleted by centrifugation, resuspended in 5 ml Buffer II (section 2.3), with 7.5 % glycerol and aprotinin, leupeptin, pepstatin A (all to a final concentration of 0.02 u.g/ml) and lysed a French Pressure cell (Aminco) at 8,000 psi. The lysate was incubated on ice for 10 min with1.5 pg/ul DNase I. Lysates were cleared by centrifugation at 30,000 x g to form an S30. Three additional volumes of Buffer II with 5% glycerol were added to the S30 and RNase E was precipitated by the addition of 26% w/v ammonium sulphate followed by centrifugation at 17,500 x g. The pellet was redissolved in Buffer II with 5% glycerol, 0.02 ug/ml leupeptin, and dialzyed in Spectra/Por dialysis tubing (MWCO 12,000 - 14,000; Spectrum Laboratories) at 4° for 1 hr in 400 mis Buffer D1 (section 2.3). The RNase E sample was subsequently dialyzed for 2 hrs in 800 mis Buffer D2 (section 2.3). Following dialysis, one volume of 2 x SDS-sample buffer was added to each sample (section 2.7.1) which was then loaded into a single well, the width ofthe gel, and subjected to electrophoresis on a 6% SDS-polyacrylamide gel (49:1  48  acrylamide:bis-acrylamide) (modified from 10% SDS-polyacrylamide gel, described in section 2.6.1). Following electrophoresis, slices were removed from each side of the gel and stained with Coomassie Brilliant Blue R-250 to visualize the location of the RNase E protein band. The band was then excised from the gel, placed in dialysis tubing (Spectra/Por, MWCO 12,000 - 14,000; Spectrum Laboratories) with a minimal volume of Laemmli buffer (section 2.7.1) and electroeluted at 100 volts for 7 hrs in a horizontal electrophoresis unit containing Laemmli buffer. Following electroelution, the liquid (approximately 2 ml) was removed and protein precipitated overnight at -20° in 5 volumes of acetone with 0.2 mM DTT. Protein was recovered by centrifugation followed by three 10 ml washes with 80% acetone, 0.2 mM DTT. The RNase E pellet was dissolved in 350 ul Buffer D3 (section 2.3) with 6 M guanidine hydrochloride and diluted with 5 ml Buffer D3. The sample was then dialyzed in two 400 ml changes of Buffer D1 at 4° for 6 and 4 hours, respectively. The sample was then concentrated to approximately 300 pi in a 4 ml Ultrafree Biomax-5K centrifugal filter device (Millipore).  2.7. Protein analysis 2.7.1. SDS-polyacrylamide gel electrophoresis  One volume of 2 x SDS-sample buffer (120 mM Tris-HCl pH 6.8, 3% sodium dodecylsulphate (SDS), 50 mM DTT, 10% glycerol, 0.1% bromophenol blue) was added to an appropriate volume of protein sample (generally 4-7 pi) and the mixture boiled for 2 min. Samples were loaded onto a 10% polyacrylamide gel (36:1 acrylamide:bis-acrylamide) containing 1% SDS and separated at 15049  200V in Laemmli buffer (25 mM Tris-HCl, 192 mM glycine, 0.1% SDS)(Laemmli, 1970) along with Broad Range SDS-page standards (BioRad). Proteins were visualized by staining either with Coomassie Brilliant Blue (0.5 mg/ml Coomassie Brilliant Blue R-250 (BioRad), 45% methanol, 10% acetic acid) followed by destaining with 5% acetic acid, 5% ethanol, or by staining with SYPRO Red (diluted 1:5000 in 7.5% as recommended by the manufacturer; Pharmacia) followed by a rinse with distilled water and imaging on a Typhoon 8600 imager (Molecular Dynamics).  2.7.2. Western blotting for RNase G  Proteins were separated by SDS-polyacrylamide gel electrophoresis (section 2.7.1). Gels were not stained following electrophoresis, but were instead blotted to Trans-Blot Transfer Medium nitrocellulose paper (BioRad) at 250 mA for 2 hrs in transfer buffer (3 mM Na C0 , 10 mM NaHC0 , 20% methanol; Dunn, 1986). 2  3  3  Blots were shaken at room temperature for 1 hr in PTBN (Section 2.3) with 5% casein. Blots were subsequently incubated with shaking in PTBN with 5% casein and the primary antibody, rabbit anti-RNase G (gift from Janet S. Hankins). Blots were exposed to three 5 min washes with PBS followed by shaking for 45 minutes in PBS containing the secondary antibody (goat anti-rabbit; Amersham, 1:3000 dilution). Washes with PBS were repeated and bands visualized by the addition of ECL (Pharmacia) chemiluminescent reagent and exposure to x-ray film.  50  2.7.3. N-terminal protein sequencing  A sample of wild type RNase G was subjected to SDS-polyacrylamide gel electrophoresis and blotted as described in section 2.7.2, with the exception that proteins were blotted onto Immobilon-P PVDF membrane (Millipore). Following blotting, the membrane was washed 3 times with H2O, and stained for 2 min in Coomassie R-250 Brilliant Blue (see section 2.7.1). The blot was destained in several changes of 50% methanol, 10% acetic acid, and allowed to air dry. Bands of interest were excised, and gas phase sequencing was performed at the University of Victoria-Genome B. C. Proteomics Centre.  2.7.4. Mass spectrometry  50 pi samples of protein (0.5-1 pg/ul) were analyzed using electrospray mass spectrometry by Dr. Shuming Hu at the U. B. C. Laboratory of Molecular Biophysics.  2.7.5. Protein cross-linking experiments  Cross-linking was performed as previously described (Klingenberg and Appel, 1989). Briefly, RNase G (0.08-0.16 ug/uJ) was incubated at 4° for 25 minutes with 12.5-30 pM CuS0 and 0.65 mM 1,10-phenanthroline in 30 mM Na S0 , 4  2  4  0.1% Triton X-100, 0.1 mM EDTA, 5 mM Tris-HCl pH 6.8. Reactions were quenched by addition of 2 mM N-ethylmaleimide, 15 mM EDTA followed by incubation at 25°. Cross-linked protein samples were separated on a 7.5% SDSpolyacrylamide gel (Laemmli, 1970) and visualized by staining with Coomassie  51  Brilliant Blue R-250. For 2-D analysis, sample lanes were sliced from the gel and soaked in either non-reducing sample buffer or reducing sample buffer (sample buffer +100 mM DTT) for twenty minutes at ambient temperature. This gel slice was then placed horizontally on top of a second 7.5% gel and subjected to electrophoresis. Proteins were visualized with Coomassie Brilliant Blue R-250.  2.7.6. Circular dichroism  Purified RNase G was diluted to 4.6 pM (0.26 pg/pl) in 25 mM Tris-HCl pH 7.6, 300 mM NaCl, and 400 pi was loaded into a 0.2 mm quartz cuvette (Hellma). Wavelength scanning from 190 nm to 300 nm in a Jasco J-810 Spectropolarimeter (Jasco) was performed on each sample. Maximum CD values occurred at approximately 218 nm, so this wavelength was chosen for the CD melting studies. Sample temperature was increased from 20°C to 70°C at 1°C per minute. The melting temperature for each protein corresponded to the inflection point in the CD scan, and was calculated using Spectra Manager software (Jasco).  2.7.7. Sucrose gradient centrifugation  2 ml gradients were poured in 11 x 34 mm Ultra-Clear tubes (Beckmann) in a stepwise fashion by layering 500 pi each of 20%, 15%, 10% and 5% w/v sucrose in 25 mM Tris-HCl pH 7.6, 300 mM NaCl, 10 mM DTT, followed by incubation at room temperature for 1 hr, as previously described (Loewen and Molday, 2000). Purified RNase G (30 ug) and 4 pg each of bovine serum albumin and aldolase standards (Pharmacia) were layered on top ofthe gradient 52  and the samples were centrifuged at 4°C for 6 hours at 50,000 rpm in a Beckmann TLS-55 rotor. Four drop fractions were collected by puncturing the bottom of the tube. 6 pi portions of each fraction were mixed with SDS-sample buffer (section 2.7.1) and products were separated on a 10% SDSpolyacrylamide gel. Proteins were visualized by staining with SYPRO Red (Pharmacia) and quantified using ImageQuant software (Molecular Dynamics).  2.7.8. Gel filtration size determination  A Superdex S200 column was calibrated by measuring the elution volume of thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), BSA (67 kDa), ovalbumin (45 kDa) and chymotrypsin (25 kDa)(all standards purchased from Pharmacia) on an Akta Explorer (Pharmacia), with the elution volumes determined using Unicorn software (Pharmacia). Samples were run in 25 mM Tris-HCl pH 7.6, 320 mM NaCl. A standard curve was constructed from this data. Samples (amounts listed in the figure legend) were loaded and passed through the column in Buffer I (section 2.3). The elution volume analyzed using Unicorn software (Pharmacia). Sizes of the samples were extrapolated from the standard curve. Calibration of the column was performed by R. Pfeutzner.  2.8. RNA substrates and endoribonuclease assays 2.8.1. Full-length substrates  Full-length RNA substrates were prepared as previously described (Cormack and Mackie, 1992; Mackie, 1998). Briefly, plasmid DNA (see Fig. 2.1 for 53  Figure 2.1. Templates for runoff RNA transcription. The transcription reaction is described in section 2.8.1. Panel a depicts the template for 9S RNA, which is transcribed from an SP6 promoter at the +1 site and extends the full length of the gene to residue 246 corresponding to a cleavage by >Acc/ or Hindi. In panel b, the coding region for S20 is indicated as an open box. Transcription from an SP6 promoter begins at the native P2 promoter, and the template is linearized with Dral at nucleotide 447, which corresponds to the natural termination 3' to the rho-independent terminator (see also Fig. 3.11b). Panel c shows the rrnB 16S rRNA cistron. The two native promoters (P1 and P2) are indicated with arrows, and the mature rRNA sequence is represented with and open box. Panel d represents the template derived from plasmid pGM119, which can be linearized with either BstUI or BamHI to yield a truncated pre-16S rRNA. Processing sites are indicated as follows: RNase III at residue -115, III; RNase E at residue -66, E; RNase G at the mature 5'-end, +1, G. The circle represents the approximate binding site of ribosomal protein S20. The horizontal line indicates the annealing site of oligo GMV 2011, which was used to direct RNase H cleavage at the RNase III processing site. Panel e shows the template derived from plasmid pGM122, which uses a T7 promoter to initiate transcription at the P2 site (see panel c). RNA transcripts corresponding to those transcribed from pGM122 could also be derived from cleaving substrates transcribed from pGM119 cleaved by RNase H directed by oligo GMV 2011, described above.  54  Accl/Hincll +246 +1  9SRNA  |  r  SP6^  T  E  _2_  PJG9S.2  E  _2_  246 nt  Dral  +92  ••"«> SPb  E 372 nt  P1 P2  ilL  +447  frpsr;  S20  T  _L. PGM79  *  SH  rrnB 16S rRNA cistron  E  • A -115 -66  "1  +1  BstUI +404  576 nt 728 nt  671 nt  55  f  BamHI +556  ->| PGM119 >l  templates) was linearized with an appropriate endonuclease. Template DNA (0.6 pg) was combined with 500 pM each of ATP, GTP and UTP, 100 pM CTP, 30 pCi a P-CTP, 0.8 U/pl RNA guard, 6 U/pl SP6 or T7 RNA polymerase 32  (depending on the template promoter) in 40 mM Tris-HCl pH 7.9, 6 mM MgCb, 2 mM spermidine, 10 mM NaCl and incubated for one hour at 37°C. The reaction was quenched by dilution and addition of NH -actetate to 2M, EDTA to 5 mM. 4  RNA was extracted with phenol/chloroform/isoamly alcohol (20:19:1) and precipitated overnight at -20°C with four volumes of ethanol. RNA was quantified by determining the ratio of total to TCA-precipitable scintillation counts (Sambrook,  1989).  Following  quantification,  half  the  product  (5'-  triphosphorylated RNA) was diluted to a final concentration of 0.2 pmol/uJ in H 0. The remainder was dephosphorylated in Buffer III (section 2.3) with 0.02 2  U/pl calf intestinal alkaline phosphatase. The phosphatase was inactivated byincubation in 8 ng/pl proteinase K in 0.1% SDS for 10 min at room temperature. This reaction was quenched with EDTA (9 mM) and the RNA was extracted with phenol/chloroform/isoamyl alcohol (20:19:1) and precipitated with ethanol. The dephosphorylated RNA was resuspended in Buffer III and rephosphorylated at its 5'-end by addition of 0.3 U/uJ polynucleotide kinase in the presence of 0.9 mM ATP and 0.8 U/uJ RNAguard (Amersham-Pharmacia). Samples of mono- and triphosphorylated RNA were separated simultaneously on a 6% urea-polyacrylamide gel for quantitation.  56  2.8.2. 5'-end labelling of oligonucleotides  200 pmoles oligonucleotide were incubated at 37°C for 30 mins in 20 ul Buffer III (section 2.3) with 0.5 U/ul polynucleotide kinase (Fermentas), 25 pCi y P-ATP 32  and 10 pmol/pl ATP. For unlabelled oligonucleotides, y P-ATP was omitted and 32  the concentration of ATP was increased to 60 pmol/pl. The reaction was quenched with 160 mM NH -acetate, 8 mM EDTA and the reaction volume 4  brought up to 50 pi with DEPC-treated water. The kinase was inactivated by heating the mixture at 65°C for 10 mins, according to the manufacturer's instructions. To precipitate oligonucleotides, the reaction volume was increased to 100 pi with 300 mM Na-acetate, 0.5 pg/pl yeast RNA instead of quenching the reaction. Residual proteins were extracted with phenol/chloroform/isoamyl alcohol (20:19:1) and the oligonucleotides were precipitated overnight at -20°C with 4 volumes of ethanol. The pellet was taken up in an appropriate volume of DEPC-treated water.  2.8.3. Standard endoribonuclease assays  Labelled RNA (amounts indicated in Figure Legends) was denatured by heating in RNase Assay Buffer (section 2.3) for 2 min at 50°C, for 10 min at 37°C, and chilling on ice. Processing was initiated by the addition of purified RNase G, RNase E or purified degradosomes, with amounts of enzyme listed in the Figure Legends for each experiment, in a final volume of 30pl (Coburn et al., 1999). Incubation was continued at 30°C. Samples of 4 pi were withdrawn at appropriate times, quenched with 12 pi 90% formamide, 22 mM Tris, 22 mM boric acid, 0.5 mM EDTA, 0.1% each of bromophenol blue and xylene cyanol, 57  heat denatured, and separated on polyacrylamide gels containing 8 M urea, and separated in TBE buffer (Section 2.3). For assays using oligonucleotide substrates, products were analyzed on 15% polyacrylamide gels while larger RNAs, unless otherwise indicated, were separated on 8% polyacrylamide gels. Products were visualized by phosphorimaging.  2.8.4. Ribonucleoprotein assays  Substrates based on pre-16S rRNA are depicted in Fig. 2.1. In each assay, unless  otherwise  indictated  in  the  Figure  Legend,  0.6  pmoles  monophosphorylated RNA (see section 2.8.1, above) was incubated in RNase Assay Buffer (section 2.3) at 50°C for 2 min, followed by a 10 min incubation at 37°C. RNA was chilled on ice, followed by incubation at 30°C for 30 min with 2.4 pmoles purified ribosomal proteins S20 (gift from G. A. Mackie) and S4 (gift from Drs. Kellie Horn and David Draper) as indicated in the Figure Legends. Reactions were initiated by addition of 18 pmoles purified RNase G and allowed to proceed as outlined in section 2.8.3.  2.8.5. Competition assays  Reactions were performed as described in Section 2.8.3, but unlabelled oligonucleotide competitor (quantities listed in Figure Legends) was added concurrently  with enzyme. Competitors  were  either  used in their  unphosphorylated condition (5'-OH group) or the oligonucleotides were 5'monophosphorylated with cold ATP, as described in Section 2.8.2. Competitor sequences are listed in Fig. 5.2. 58  2.8.6. Phosphatase assays and thin-layer chromatography  5 uCi Y  3  2  P - A T P ,  8 pmol A T P were incubated at 50°C for 2 min, 37°C for 10 min  in RNase Assay Buffer (Section 2.3). The reaction was initiated by the addition of 400 ng pnp13 degradosome in a final volume of 30 ul. At 0, 30 and 90 min, 8 ul aliquots were removed and heated at 100°C for 2 min. To examine phosphatase activity on labelled 5'-monophosphorylated oligonucleotides, assay reactions were performed as described in Section 2.8.3, but at each time point, 6 pi of the assay volume was removed and heated at 100°C for 2 min. As a positive control, 8 pmol ATP containing 5 pCi [y P]-ATP was incubated at 37°C 32  for 45 min with 0.1 U/pl calf intestinal phosphatase (Pharmacia) in Buffer III (Section 2.3) in a final reaction volume of 10 pi. Heating at 75°C for 10 mins inactivated calf intestinal phosphatase. Polyethyleneimine-impregnated 5x10 cm thin layer chromatography plates (Macherey-Nagel) were rinsed in dH 0, and 2  allowed to dry. 1-1.5 pi of each sample was spotted on to the plate approximately 1 cm from the bottom of the plate. The plates were placed in a chromatography chamber containing approximately 0.5 cm 0.375 M KH P0 2  4  (pH adjusted to 3.5 with HCI) and the solvent run until within 2 cm of the top of the plate. Plates were dried, and products visualized by phosphorimaging (Molecular Dynamics).  59  3. ACTIVITY AND QUATERNARY STRUCTURE OF RNase G 3.1. Introduction  RNase G is the endonuclease responsible for formation of the mature 5'-end of 16S rRNA from a larger precursor (Li et al., 1999b; Wachi et al., 1999). In addition, RNase G may control the stability of alcohol dehydrogenase and enolase mRNAs (Kaga et al., 2002; Umitsuki er al., 2001). Moreover, array data suggest that RNase G can also act on a limited set of additional mRNAs (Lee et al., 2002). Curiously, the gene encoding RNase G was originally identified as Caf A which when overexpressed resulted in the formation of cytoplasmic axial filaments in E. coli (Okada et al., 1994). Importantly, RNase G exhibits 35% sequence identity and 50% similarity to the catalytic domain of RNase E (McDowall and Cohen, 1996; McDowall etai., 1993; Taraseviciene etai., 1995). RNase G can mimic some RNase E cleavages in vitro, and these cleavages are also 5'-end-dependent (Jiang et al., 2000; Tock et al., 2000), a characteristic of RNase E activity (Mackie, 1998). RNase E is the major endonuclease in E. coli and has been implicated in rRNA processing (Li era/., 1999b; Misra and Apirion, 1979; Wachi etai., 1999), tRNA maturation (Li and Deutscher, 2002; Ow and Kushner, 2002) and bulk mRNA decay (see reviews (Coburn and Mackie, 1999; Grunberg-Manago, 1999; Regnier and Arraiano, 2000; Steege, 2000)). RNase E associates in vivo with PNPase, a 3' to 5'-exonuclease, Rhl B, a DEAD-box RNA helicase, and enolase, a glycolytic enzyme, to form the RNA degradosome (Carpousis et al., 1994; Py et al., 1996). The assembly of the degradosome requires the Cterminal domain of RNase E (residues 645 to 1045) to serve as a scaffold 60  (Vanzo et al., 1998). However, this domain has no equivalent in RNase G and is not required for endonucleolytic activity. Several important questions still surround RNase E and RNase G. First, the active site in either enzyme has yet to be identified. Second, the basis of substrate recognition by both enzymes remains elusive. Both cleave RNA at single-stranded sites rich in A and U residues (Ehretsmann et al., 1992; Jiang et al., 2000; Mackie, 1991; Mackie, 1992; McDowall etai., 1994; Tock etai., 2000).  RNase G, however, appears to cleave somewhat more promiscuously within a given sequence than does RNase E (Tock et al., 2000). The molecular basis for these findings is unknown. Finally, the tertiary and quaternary structures of both enzymes are unknown. Potential self-interactions within RNase E have been demonstrated with yeast two-hybrid experiments (Fig. 1.4; Vanzo et al., 1998), implying that RNase E may form homodimers, but this has not been confirmed by other means. RNase G shares significant similarity to the N-terminal domain of RNase E, the region associated with endonucleolytic activity and self-interaction (McDowall and Cohen, 1996; McDowall etai., 1993; Taraseviciene etai., 1995; Vanzo et al., 1998). This implies that the properties of RNase G will likely also pertain to the catalytic domain of RNase E, allowing RNase G to act as a surrogate for investigating the structure and activity of RNase E.  61  3.2. Results 3.2.1. Sequence comparison  A number of regions of high similarity (90% or greater) are visible in the alignment (Fig 3.1) of RNase G and RNase E from a number of different species. Regions of interest are described in Table 3.1. The first region of high similarity is located between residues 100 and 110 (with respect to the residues of RNase G from E. coli). This is located in the S1 domain, which extends from approximately residue 35 to 135. This is a potential RNA binding domain, which was originally identified in ribosomal protein S1 (reviewed in Coburn and Mackie, 1999). Two temperature-sensitive alleles (marked with "t") of RNase E also map to this region, ams-1 (G66S) and rne-3071 (L68F)(McDowall et al., 1993). Residue 267 is marked with an arrow, corresponding to the truncated RNase G protein which is produced by the BUMMER strain (Wachi et al., 1999). The second region of high similarity is located between residues 279 and 351, and includes the highly conserved aspartic acids at residues 304, 347 and 350, which are marked with "#". Recent findings in our lab have revealed that introducing D304A or D347A mutations lead to a loss of activity, while a D350A mutation leads to a decrease in activity (V. Kunanithy and G. A. Mackie, unpublished). The third region of high similarity spans residues 377 to 400. The cysteine residues are indicated with "@" symbols, and it is interesting to note that apart from C405 and C408, which are highly conserved, there is very little conservation among cysteine residues.  62  Figure 3.1. Sequence alignment of RNase G and RNase E. Organism names  are listed to the left of each sequence, and the sequences are grouped into RNase G and RNase E homologues. Sequences were aligned using ClustalW. Numbers at the top correspond to amino acid residues in E. coli RNase G, and numbers to the right of each sequence indicate the actual residue number for each sequence. Residues highlighted in black are similar in over 90% of the proteins, dark grey residues are similar in 75-90% of proteins and light grey residues represent 50-75% similarity. The six cysteine residues found in E. coli RNase G are marked with "@" above the sequence. Aspartic acid residues which have been mutated to alanines by this lab (V. Kunanithy and G. A. Mackie, unpublished) are marked with "#". Two temperature sensitive mutations, ams-1 (G66S) and rne-3071 (L68F) are market by "t". The three high similarity  regions (HSR1-3) are marked with horizontal arrows, as is the S1 domain. A vertical arrow marks the position of the C-terminus of the truncated RNase G encoded by the BUMMER strain.  63  S1 80  halodurans  C. H. W. P. V. X. A. N. A.  trachomatis influenzae meningitidis aeruginosa cholerae fastidiosa aeolicus gonorrhoeae actinomycetem*  E. coli R . .ipii t [j i , \ ' : .1 H. influenzae M. tuberculosis N. meningitidis P. aeruginosa F. prowazekil Synechocystis sp V. cholerae X. fastidiosa A. thaliana G. theta N.  Oil eli  [REARRGI' ;RPVEERIVfS| j f e P K K I RQLKGW|  SJ^tPHTECVAGEEQKQFTVRDj S F H L S N E E E D E K K - - K R N | | S 0 F •QEG 1( ENSKKFEQMFDIDTSEAr?; KELgK:..;  JTHEP.KKVP. Q LKGN| gljJEBKKIPQLKra| LRQAKPGIVG!l| NSEHSLVSMg [PTQRPGIvq^ LEAKRGI' ;RDGCP.(  ENSKKFEQMFDMDSEEAP|EEF»I|LD ENSKKFEQMFDIDTSKAP||EELBKLD fVSHTECVDENEQKQFKVKsffsEIflREG EQPRNPEE jSN REGSAVES S|SA£lHE<S PHT E CVAE S E KQQ FQ VRDj|s B I f l R QG .SAPADS 3 V I D T D L S EI P§\ EKKNCKN JRIS  ;  FK^ERRKEKYPTGAF:  ||1NATQQE|LRVM JGQ] p A T Q Q E | L R V M 3GQ! p A T Q K E | L R V A . I . DGQ] ^ A R R E A V g T O I A V EDGf  DgjS5PGHEQKKAM| [SGSEQKKs|fe SPGHEQKKAII SAASASLVSKI  EYFPANY3AHGRis K N Y F P E N H I E T L G ft ? !! EYFPDDYVFOGR IAAGLGGAD R K - ISRSYFQD—YEGGR-REYFKK- - S PEGR |KS$|: FS^HPNYYNLPTAESTFYKPYKSi Vi 'PVRLRRTAGS 'iKgAREYFPDGYSYQGRPCKgSKDYFQAG-LDHHKDJjfKSNREPFI F P P F C D G S K S 5 K PFTD^IKKSNQQ J HfS|TGPLKKKYYINN- • DFATTHELTGEN YAGgjFEALGMANGPRR flES, KtsSgNEAYRKTVLKGQ- • 1  'LGKEQRK||| ;GAREQKKA1| |EFQTTVRQQNKGM| §3!AEKHQ---VAA "]STGSQQVG--D| SNATQKE|LRVAI JGQR! ISPGHESKKAKj] ! N A T Q A E | L R V A 1 DGVS: PSKEQKKSI ' NSSICTQKT.AV IGGI PVKTNVQCDSj iLKELG F S I FQSKCE IFQKEQCGLNDIYFGFIPPQSJ ; I S C L K N - - - I A A LCGQSJQKS' '2JANAHYQVS--Dj£gj] jDKLAPGLGIAFVSWT] ¥AP.REAV|TQ;IG\IJEDGV|V|H' Y^KEQATSYVGBJ jNATOPE|i.RVAl | $ D A N F P S B'T R W I  3GQRJ  G3NNIJEE;  -MDDGJJLEEVFFD-  EIETIABKI  . Br ;. .  coli holodurans muridarum pneumoniae  180  3 "•  *  1  200  ,I^GEA^AS|AAY|KRVITKJME R K  " E f i | E R ^ S EWEQ&LLWDEJj K K E V R H B A E Q I I LF F R Q J E E K J K R G IMSDQBQfelRS F E M P Q ^ I SA£T&A1I I^QD|LNT1QSILE KF •n^EcKoLi RS F E M P Q M | 1 | | •"^^JtJRSFEMPQM^® TTA5TEAHI1^AHI^LLTBKTBLEKF A L V E P FC DE L G & F g $ | A S I S A S T g T a i N g A Q I ^ I ^ T i Q s l L E K F | A T E G A S EE£|P Q | A E F|KR LIR VMLERK MQEQA " "^AGEOB|IA^IRYWRR JDQIAAOI E • IVSHYCDEHGj ; i A N E K i | T Q f A A F i K P I L B S KB I ERR --•'ISRIIGVIERRG T K E . . . I FTED^KILKEELNDR J p 3 g  influenzae  !AENATDEoiQs|xDYiTKvfc  cholerae fa :: aeolicus gonorrhoeae actinomycet*  M. tube. N. meningitidis P. aeruginosa R. prowazekii Synechocystis sp V. cholerae X. fastidiosa A. thaliana  FSHQIGQKINNYIKVP : ( l .:LKKI:VPEDA^ ^~1AF::L:  B U M M E R  FADAELDRi  C. H. N. P. v. X. A. N. A.  JFPLD«VGEJj 'CVDKSYK ITCIDKNYK 'CVDKSYK IGTNLEK 'GCOTQK  muridarum trachomatis influenzae meningitidis aeruginosa cholerae fastidiosa aeolicus gonorrhoeae actinomycetem*  E.  coli aphidicola H. tflu M. tuberculos N. memngitid R. prowazekii Synechocystis V. cholerae X. fastidiosa A. thaliana G. theta P. purpurea  pRLTYEAlLErrsEYIPEMTSKLEHYTGRQ^ bDYATYQKCKRL£S-KYSPDTAVKIEY'i" i'J E' S : ; Y A T Y O K C K l f M t . K - KYSPDAS I KTEYYDS ' " * IDYATYQKCKRL^G-KYSPDTTVKIEYYDS IS KLCFGEVKEFTDE FMPELS DKLVLTF SGN is TVNHGRJJT R FAE q YVHGAL GRI ELF KGE JDSRENFQKITSFVEELMPEISDPLEHYPGE: i H D S P . Q E F E N I j K E F T S E Y ' u ' P E L T A K L F , LYHGDKpK^ MpS K S AFAQLQ G F V T KYM PVLAD K L E L Y V G D H ^ ^ 'fpNGDVWGEJVEFFGDSIKHKLRYVNTID-ELfKGE: EFIPNLTDKLILYAGNQP^  ESRPAKF| KS RT Aip F g QSPPA|F^ QEAGAAV KAHHDfY GEBGAPF IKFSAPC MTRSAPM DSNPAfF LSKPAPL K S G A I PAi LRKKMfQF IKMCAP-v' g  i v V ' i - R . Y: P  ikPKVLELARQHTAALGRPDFSSKIKLYGEI: KPKILDIARKHITFLGPPDFVNKLKLYGEV^ SPKIFEKAKEH|K'LVRPDFINRVKLYGEV»|S: GDEASMT:I:NEYVNSVAPELVSKLTKYESAPDJJ  DNOE'/VDCVAEEMS-YVMPGNIGBLKLYDHTJ B 5 I DAQEEALNffrR - QVMPQYAS K V K L Y D S V ^ ^GQKAYEDAS D L L P S ELAKEKEHNKT3 3TPAGM!<T(3KQQ£MNWDQGRLPEGViIDESLSJ DSNSiYEPALEHIR-LVRPDFVNRVKKYGEVP^^SI y:y.KMYCI:AHEFM?-KVMP*r. ;« • 3SPRTYKEVTHKLQDMAPDLCNRVELHD--I^D: SSYQRAVYERH^ITHYFTIKQTDYRILTAYKSMQI . . r YDNNTKNg QSNLGLKQiiNYilNTWQCNFSSTIPSLQSKKcf^T IETGISY^SMTEKFVSPVCAgWFQHPLSV GCNLQIRESMTKWMITHVPHKSRHIEIT-LDAWKNV SGDDAiffQTil-HG^VS HVAPDLAE RLSKMTSEV1 :YNDRSLLETlQKiiQ NLPKKPEIEVEG1  g:  64  EI 2QQA SA KQAS RA AEAA KN TDEA 5 N FTKL YFfQKSA  :: :-CNFERALEI#IQ|LQI  E  2 30 ; 93 : 9:  187  KSAL5S&ELPQ1  • ^ E A t E K t H I PN1 ^ GE&1AAB3E: g •"I A K T L Q P Q G F | l i -  |QKQ Lra F^LHA LT YYS KEO  RNase  2 •; •  200  I  G  200  ULAVLIKP?'  ™  RNase  197 200  <EAIJ\SLELPEH^Ll\ffiAGVCKEAEA|QW^SFPLKH§E KKAA <ELLTLLELPEMS^3^|RGASKSIEs|pwil,3LR IIA E A t S S t D V P D V j y a f f i i G V G K S r E E .W i l QAS E H R E W PS D A ^ p ^ f t S E G V K E O I ^ R A | V A P | AKA AAMAEiiD I PNMSjjj JEEAG EALNGtNAP. 'WpiiDYSLQlJJKA KEAS DliNKJVSIQYS^ ;YNY](|ARI NK RKST  _..C£ITL:  C.  200  194  192  194 193 198 177 187  <E. B. C. C. C. H. N. P. V.  coli halodurans muridarum pneumoniae trachomatis influenzae meningitidis aeruginosa cholerae  A", A. N. A.  fastidiosa aeolicus gonorrhoeae actinomycetei  E. B. H. M. N.  coli aphidicola influenzae tuberculosis meningitidis  DHRRR t H s | E Q A | s i A v | T s L . . „ , ... : g \S :& •:.  AFVGHRNIj fSKFIGKQDti  BRSTQLECTBEIISVOII GR5TQLEG1  .. . g  G FV GARM E^gfllFM^ EjAFVC IGAFV:  DAEHRCC GFVGARNFlllLR! AFVGHJ  PVRHvRA ENR|REA|RI; PIRHQRE EMI Hi', ,ESNRDL &RR||T E SaARret'  T  LATRGA-  PAKNQRA j.SEAKM RRI BSHKDQLC. PVRHQRE SSKHQRE 'ESNKfii IHQKDQIH  ATKGG-  1KSTSEK-FTH5NS ATKGG|SRATKGN [FGQGSQgKAlLE'  --HSSVSQMLSL1 -FNNSSAR§^VL1  glLEIGS pKFTGQG]  E. B.  coli aphidicola  H. influenzae M. tuberculosis N. meningitidis P. a e r u g i n o s a P.. prowazekii Synechocystis ap. V. cholerae X. fastidiosa A. thaliana G. theta  IiCME fcSKP LFTT tFTL IFTT tCDE £CEP LCEP fcCST fiSET  : |GET : J.CK.F :  : SHHV :  SHHI  SHHI  : : : : : :  .SC^QLC^EHFDKE|SL|D^PC^C41E:  KTC,  rDIDLEKLgVMilQFTSLt  ~  r  489 R N a s e G  Ei|w-EYRGSDAE^VIELPSHmPVlFGQDGGHLKRLETAlGSR|LAVPKKTLKEBt KIIKHKEHTHLCLVVHPESWYMKOJEQDDELVHLAKQIiKAKgClNTSDSIHLNI VI NHKEHSHtC tVVHPEJJAS Y K K Q E N D D E M I NLAKQL KAKSQINTS D S VHLI ™JtIIKHKEHTNI^tVVHPESWYMKOJ!QDDELIRIJVKQtKAKaQIMTSDSIHLh  ; :  I A S V T Y E S ^ H D W T R A V R Q F D A V R L L V I A S PN^SRWTDiBDSAAVAELEAViGKAgRFQADPHYLQE^EVVL^  :  »\flji[^FEll^  iKWFlLHTgEBUSEKTRGKK  ....  rvl.KVHPS|ftKT  EEFINEK^KDWpTlKEVWEERSOY^QtsYS  'PQTVC YESO§EgVREARR YDAE S FRI LAAPN«I DL F I . D E E S Q S LAMLIDFIGK P B S LA VET AYTQ(< ",-grVC Y EWfeEBl RVHHVFAS EQFWYAS H A » D YtlNEES HLLAEL EVFXGKCJWQVKT EVF YTCE' ll^SLSLSglJl^EEALKEOTQEVHAIVPVPgKYlX^KRSVUAIETRQDGW  •  INSSLS L S S J L ^ E E A L K E N T YEVRAI VPVE»C YUJJE KRDVHAIE KRQAGGKTIIVPSKKMKTPI 'N^LSLS^gL^EEALKENTKQVHTIVPVO§^YLI.NEKRKISNIEKRHN-vI)§IVAPNEAMETPI lAftPVDSAAATGRKSEPGARSSDRSMVAKVPX/HAPGlJ^QAGDPPTDLDDTAQADFEDTEDTDEDEDELDADED > I|«5TALH|Lgl|QEEAMKDNTGEVRAQVPVD®»T FtLl*EKRAL FAME ERLD - V N S V L I P N I H L E N P H J & I N R O iVHfliS L A E i L B E E E A L K D RTAEVRARVP F(3glAFliNEKRNITKIELRTR-AJ^Fn.PDDHLETPH&VQR5  FST3 SHVS SGIV N5EV  -is: 478  ^SANAMLJ^iT|ENEIFEERIDin^TNIVSVIYLIJ4NKRAIKFIEEKY^  PGQP SHHI  ISEP  .LPGEKGFgpLSPTAWSSIKVNDlPKKEEAKfeSP-tDLLFHNYQEOGDRDSNRRRRRRRGSEFSEKENXKSg iNRSLAL SALSLgEEEAL K D N T S CVLAVVPVSgftS YIiNEKRRINHIEKAQQ-VR?fTXVPNSDMETPH^VIRa lIpLSLS|l|l|EEHAMKENTGQVLVQTPVE^YLL^KRRINEIEKRHD-APfiIIIADEO |EAI^TFSK|ECE|CRQLVSV IIsrj$FCYFSFQPNEEHlJ ISKSSFLYDR|LSLNKNLFKTN LIIKYTLFSNVKLIYNY  -  :  1  -DSNYFHFFIQL I  ESNRDL^RR||LEC]  JEHIRVYHLFSSEQFVVYASPASSEYLINIESHLLPEV^^ —^JEMVIUIARRYDAESFRXIAAPNGLDLFLDEESQSIAMLIDFIGK !AGTTC YE§FSE«LREARAYQADS YLVLANQKFFIVDRL&DE ES GTWADLEAFIGRT«KFQVEAOTSQEI™.  : SQIV  :  EER^R] ERSFKE Fi JBHFNRC ENI ET1 rai tAFFRYLL  ESVRVHHAYDSDRFLVYASPAKU^KGEESHSIJ^  IiCGi:  :  LP'/f  JSAI5HQRA ENRpRElARI  jSKFTGAGNlijj  : : : : : : : : : :  JTRESBEIP/ LQ3§QAS  "  KEEHRTF FGIIAGSPCRRPGENOCQWFHS  TRGC[SRST5GT URSTRGG  YTDAE-CJ|SLAL:  coli halodurans raunddrum pneumoniae trachomatis influenzae meningitidis aeruginosa cholerae fastidiosa aeolicus gonorrhoeae actinomycetemco mitans  B  KSPKNQRR EiERiKEHi "ITDEHRNR L , .. I: . .QESHREA IQEF JEDEEHRRC LRT;; ,SDEHRKR L  AFVGHRNt|»KcFN!  F. . pi o v / . - i : j Synechocysti sp. V. cholerae X. fastidios, A. thaliana G. theta P. purpurea N. olivacea S. coelicolo  E. B. C. C. C. H. N. P. V. X. A. N. A.  HSR3  HSR2  B' B ' ' ' " H '  ;  n  EQLDSIKP5SYKS  KRisSSVKS^FSE-lDMINRFFKK--  QKILfK|gIREDLKKLKDFEEVrIiKVHPN'|SGYFKRED  65  480  -LSNF§YARN^  'fl™L||QP3SAGGGGGKRRKRARETEVAAEvAEPVALPAKAEAAPAAPTAQDVTVREE|ERPV^  tFTR§? jJs^T;|KfLS.  474 479  IKKLQKEFK|KLNLDYGWHDPNSYEIKA  ; :  Table 3.1. Features of RNase G and RNase E. Several features were noted when comparing the amino acid sequence alignment of RNase G and RNase E from a variety of prokaryotic organisms (Fig. 3.1). The amino acid residue numbers are based on the sequence of RNase G from E. coli. HSR is used to describe High Similarity Regions. Feature  residue number(s)  importance  S1 domain  35-135  potential  RNA-binding  domain,  possible role in 5'-end dependence HSR1  100-110  unknown  HSR2  279-351  possible role in endonuclease activity  HSR3  377-400  unknown  ams-1  66  Gly—>Ser  results  in  temperature  sensitivity in RNase E rne-3071  68  Leu-»Phe  leads  to  temperature  sensitivity in RNase E BUMMER  267  truncation at this residue results in a loss of activity mutant of RNase G  cysteine  79, 162, 402, 405, 408, unknown function, useful as a tool for  residues  421  elucidating quaternary structure  aspartate  304, 347, 350  304 and 347 required for activity of  residues  RNase  G, 350  affects  activity.  Possible metal binding residues  66  3.2.2. Purification of RNase G and identification of its translational sta site.  Purification of His-tagged RNase G has been described previously (Jiang et al., 2000; Tock et al., 2000). Neither report, however, resolved the ambiguity of the two potential sites for translation initiation that reside 18 nucleotides apart (Li et al., 1999b; Wachi etal., 1999) in the rng gene, as shown in Fig. 3.2. The  cloning strategies used by others have utilized artificial upstream regions, thereby selecting for a particular start site (Lee et al., 2002; Tock et al., 2000). Furthermore, the purification of RNase G by Tock et al. (2000) was performed under denaturing conditions. Two separate cloning and purification strategies were undertaken to determine the correct N-terminus of RNase G. To examine full length, untagged RNase G, the PCR strategy outlined in Fig. 3.2 (also see Materials and Methods, section 2.6.1) was used to maintain the integrity of both potential Ntermini of RNase G. By retaining 64 nucleotides upstream of the second potential start site in the cloned fragment, the bacterial host was allowed to select the preferred N-terminus. The rng gene was amplified by PCR from genomic DNA from strain CF881 (Table 2.1) using the primers listed in Table 2.2 which add BamHI sites to the 5'- and 3'- ends. The amplified DNA was cut with BamHI and ligated into the corresponding site of pET11. Following transformation and isolation of clones, the correct orientation of the insert was confirmed by restriction endonuclease mapping and the resulting plasmid was  67  >  amplified fragment  -18 +1 -64  51  +1469 +1507  rng UAA  gene  M T A E L L V N... 55364 Da M R K G I N M T... 56064 Da  Figure 3.2. Potential translation initiation sites of the rng gene.  The rng open reading frame is represented by an open box together with two possible translation initiation sites shown by horizontal arrows. Nucleotides are numbered beginning with the first residue of the open reading frame determined in our experiments (+1, see Results 3.2.2). The site at +1 corresponds to an AUG start codon, with the N-terminal sequence and predicted molecular weight listed below the open reading frame. The second potential site of initiation is a GUG codon which begins 18 nts 5' to the +1 site and would generate the N-terminal sequence and molecular weight indicated at the bottom of the diagram. The region of the rng gene amplified by PCR (see Section 2.6.1) encompasses 64 nts 5' to the +1 site, and 23 nts 3' of the translational stop codon.  68  designated pDB6. In an attempt to preserve both the structure and activity of wild type, untagged RNase G, denaturation steps were avoided during the purification. The purification is described in section 2.6.1. Fractions from successive steps of the purification are shown in Fig. 3.3, with each lane representing roughly equivalent amounts of protein. The initial step in the purification was a 30,000 x g centrifugation to remove insoluble proteins and cellular debris. The insoluble fraction (Fraction 1) contained a significant portion of the overexpressed RNase G, presumably due to misfolding or association of the protein with the cellular membrane (Fig. 3.3, lane 2). In this regard, the Nterminal region of RNase E, which shares a high similarity to RNase G, also interacts with the membrane (Liou et al., 2001). This fraction of the protein may also represent RNase G associated with cytoplasmic axial filaments (cited in Tock et al., 2000). RNase G was the most abundant protein in the soluble S30 fraction (Fraction 2), although there were a number of contaminants (Fig. 3.3, lane 3). RNase G in Fraction 2 was precipitated with 26% w/v ammonium sulfate (approximately 40% of saturation) leading to the removal of several visible contaminants, including a prominent 31 kDa band (Fig. 3.3; compare lanes 3 and 4). Further impurities were largely removed by affinity chromatography on a Heparin-agarose column, and the resulting fractions were pooled to maximize purity, rather than yield, forming Fraction 3 (Fig. 3.3, lane 5). The final step in purification was anion exchange chromatography (see Materials and Methods, section 2.6.1). The resulting fractions were also pooled to maximize purity yielding Fraction 4 (Fig. 3.3, lane 6). This step removed most of the remaining  69  kDa  1  2  3  4  5  6  20011697.466 - \ 'RNase G 45 H  3H  Figure 3.3. Purification of RNase G. Purification of untagged RNase G is described in Materials and Methods, section 2.6.1. Each lane contains approximately 500 ng of total protein. Lane 1 contains size markers, with sizes indicated to the left of the panel. The remaining lanes are as follows: lane 2, protein from the 30,000 x g pellet (Fraction 1); lane 3, Fraction 2, 30,000 x g supernatant; lane 4, Fraction 3, AS-26; lane 5, Fraction 4, heparin column; lane 6, Fraction 5, final pooled fractions from anion exchange chromatography on a Source Q (Pharmacia) column. Proteins were separated on a 10% SDS-polyacrylamide gel and visualized by staining with Coomassie Brilliant Blue R-250.  70  impurities. Yields of up to 6 mg/l culture were obtained by this method. The gain in specific activity during purification could not be determined reliably. Representative elution profiles for the heparin and anion exchange columns are shown in Appendix Figure 1, panels a and b, respectively. The second cloning strategy was used to overexpress and ultimately purify RNase G with a C-terminal 6xHis tag. This strategy is depicted in Fig. 3.4. To construct the overexpression vector, pET24b (Novagen) was isolated from the dam' strain JM110 and digested with Xbal (dam methylation-sensitive) and Nhel and religated to remove the ribosome binding site, thus creating pET24bmod. The rng gene was amplified by PCR from genomic DNA from E. coli strain CF881 (Table 2.1) using the primers listed in Table 2.2. This introduced BamHI and Xhol sites into the 5'- and 3'- ends, respectively. The resulting fragment was cleaved with Xhol and BamHI and ligated between the corresponding sites of pET24b-mod producing the recombinant plasmid pDB1. As a consequence of these manipulations, the RNase G enzyme encoded by pDB1 contains a noncleavable His6 tag at its C-terminus. His-tagged RNase G was purified under non-denaturing conditions, as described in section 2.6.2. As shown in Fig. 3.2, translation of RNase G could be initiated at either of two in-frame sites. The potential N-terminal sequences are MRKGINM and MTAELLV resulting from initiation at the -18 or +1 sites in the rng gene, respectively. Two lines of evidence show that only the second site is authentic. First, purified RNase G was subjected to electrophoresis and blotting to a PVDF membrane (see section 2.7.3). The region containing RNase G was excised and gas phase sequencing was performed (University of Victoria-Genome B. C. 71  Figure 3.4. Construction of pDB1. Plasmid pDB1 was constructed as described in the text, section 2.6.3. Panel a shows the intact plasmid pET24b. The region of interest is expanded in panel b, with the T7 RNA polymerase promoter indicated with an arrow facing the direction of transcription, the ribosome binding site indicated as a box with diagonal stripes, and the region encoding the six histidine tag as a shaded box. Religation of the plasmid cut with Xbal and Nhel results in plasmid pET24b-mod, which is shown in panel c. The "X" indicates the loss of the Xbal and Nhel restriction endonuclease sites at the ligated junction of the plasmid. Panel d shows the plasmid pDB1 with the rng coding sequence inserted between the BamHI and Xhol sites. Expression adds six histidines to the C-terminus of RNase G.  72  cut with Xbal/Nhel religate  X  _l_  BamHI I  Xhol  T7  Histag  promoter cut with BamHI/Xhol add insert ligate  X J  BamHI  Xhol  rng  BamHI  Xhol  rng  promoter  Hstag  73  Proteomics Centre). The sequence obtained, TAELLV, corresponds to translational initiation at the second of two potential start sites (+1) in the rng gene followed by removal of the N-terminal methionine by methionyl amino peptidase (Ben-Bassat et al., 1987). Interestingly, earlier attempts to characterize Caf A/RNase G resulted in purification of a 51 kDa protein whose N-terminus was TAELLVNVTP (Okada et al., 1994). Second, the expected mass of RNase G translated from the -18 site is 56,064 while it would be 55,364 if translated from the +1 site. If the N-terminal methionine is cleaved from the protein, the expected mass in the second case would be 55,233. The mass determined by electrospray mass spectrometry (Table 3.2, section 2.7.4) was 55,228, which corresponds to initiation at +1 followed by cleavage of the Nterminal methionine. No species of mass corresponding to initiation at the potential translational start site at the -18 position were detected.  3.2.3. Oligomerization of RNase G  The quaternary structure of RNase G was probed because the resultant information would also provide insight into the structure of RNase E. Preliminary results showed that RNase E could be cross-linked by mild oxidation (J. S. Hankins and G. A. Mackie, unpublished data), which indicates that cysteines are located in close proximity between subunits at the protein-protein interface. We employed Cu -phenanthroline to introduce inter-molecular cross-links into His2+  tagged RNase G and resolved the products by two-dimensional electrophoresis (section 2.7.5). The resultant partially oxidized species were separated in the first dimension under non-reducing conditions (Fig. 3.5a, b). The prominent band 74  Table 3.2. Mass spectrometer analysis and sequencing of RNase G  mutants. Mass spectrophotometry was performed as described in Materials and Methods, section 2.7.4. The expected mass corresponds to the mass calculated, using ProtParam (http://us.expasy.org/tools/protparam.html), for each RNase G mutant with the N-terminal methionine (fMet) cleaved. Potential RNase G cysteine to serine mutants were sequenced through the region of interest to confirm success of the site-directed mutagenesis (Materials and Methods, sections 2.4, 2.6.2). expected  determined  mass  mass  ENZYME  (-fMet)  (+/- 0.02%)  difference  sequencing  wt  55,233 56,298 56,265 56,282  55,228 56,312 56,273 nd  56,250 56,282 56,282 45,401  56,266 56,304 nd 45,415  +5 +14 +8 nd +16 +22 nd +14  yes yes yes yes yes yes yes yes  wt, His-tagged C405S/C408S C421S C402S/C405S/C408S C402S C162S AS1  75  confirmed  by  1  1 1  200-  1  w \  1 1  [*  M  116- mm 97.4- *•)  1  66- H *  45dimer monomer  reduced dimer  monomer  Figure 3.5. Oxidative cross-linking of RNase G. Samples (3 pg) of His-tagged RNase G were oxidatively cross-linked by C u 2+  phenanthroline as described in Materials and Methods, section 2.7.5. Panels a and b depict 2-dimensional analysis of 3 pg of each cross-linked His-tagged RNase G. In both panels, the first dimension is non-reducing. Monomeric RNase G is denoted by *. The second dimension gel in Panel a is non-reducing, and in Panel B the second dimension gel contains 100 mM DTT, and is thus reducing. Proteins in Panels a and b were visualized by staining with Coomassie Brilliant Blue R250, and monomeric and dimeric species are indicated by arrows. These experiments were performed by Janet S. Hankins and were published in Briant et al. (2003).  76  at 56 kDa (denoted by *) represents monomeric RNase G. The higher molecular mass bands represent intermolecularly cross-linked RNase G, including a distinct band migrating at the position corresponding to RNase G dimers. A number of higher multimers, some the size of tetramers, were also observed in Fig. 3.5a. These results suggested that RNase G exists as at least a dimer. In addition to these more slowly migrating species, a second, faster running band was observed just ahead of the monomer (Fig. 3.5a, b). This apparently resulted from the formation of intramolecular disulphide bonds (see below). In Fig. 3.5a, the second dimension polyacrylamide gel was non-reducing and all detectable species separated along a diagonal, consistent with maintenance of disulphide bonds. Due to incomplete stacking, the samples were streaked. In Fig. 3.5b, cross-linked samples were initially separated under non-reducing conditions, but reduced in DTT prior to application to the second dimension. On the second dimension gel, all species from the first dimension, including that migrating faster than the monomer, migrated with an apparent mass of 56 kDa, rather than on a diagonal. This result demonstrates the reversibility of the oxidative crosslink and clearly shows that the mobilities of the various species in Fig. 3.5a must result from differences in disulphide bonding. Sucrose gradient sedimentation velocity centrifugation was also employed to determine the multimeric state of wild-type RNase G (see Materials and Methods, section 2.7.7). Fig. 3.6 shows the sedimentation profile of wildtype RNase G relative to markers in a 5-20% w/v sucrose gradient in the presence of DTT. The bulk of RNase G sedimented at a velocity which was consistent with a mass of around 130 kDa, which corresponds to a dimer. Static 77  158kDa aldolase 1.2  j  T  9  67kDa BSA  10  11  12  1—  13  14  15  16  17  18  19  20  fraction  Figure 3.6. Sedimentation velocity analysis of RNase G. 30  ug of purified RNase G (untagged; •) was analyzed by sedimentation through a 2 ml 5-20% w/v sucrose gradient as described in Materials and Methods, section 2.7.7. Fraction numbers are indicated at the bottom of the panel. The elution profiles of 4 pg each of aldolase (158 kDa, • ) and bovine serum albumin (BSA, 67 kDa, •) are represented as dotted lines. The peak fraction for each standard is indicated with a vertical line. Sedimentation is from right to left.  78  light scattering of RNase G also demonstrated that RNase G was multimeric, but heterogeneity of the sample prevented an accurate determination of its size (data not shown). An estimate of the size of RNase G was also made by gel filtration (Materials and Methods, section 2.7.8; Appendix Fig. 2). The majority of RNase G eluted from a calibrated Superdex 200 column (Pharmacia) at an elution volume corresponding to approximately 210 kDa (Fig. 3.7). This indicated that RNase G was likely tetrameric under the experimental conditions. The double peak following elution of each sample corresponds to components of the buffer (Appendix Fig. 2).  3.2.4. Effect of cysteine to serine mutations in RNase G  RNase G and RNase E from different organisms share a number of conserved residues. Strikingly, those corresponding to cysteine residues 405 and 408 of E. coli RNase G are conserved in 26 of the 29 RNase E and RNase G protein  sequences aligned (see Figure 3.1, and section 3.2.1). Moreover, the oxidative cross-linking of RNase G shows that cysteines on different subunits must be in proximity. RNase G from E. coli contains six cysteine residues. Figure 3.8 is a schematic diagram of RNase E and RNase G, with the relative positions of their cysteine residues indicated. Cysteines at position 405 and 408 (relative to RNase G) are highly conserved. RNase G contains one cysteine (C79) in the S1 domain. Each cysteine was systematically changed to serine alone or in combination (see Materials and Methods, section 2.6.2), creating the mutants listed in Table 2.2. This was done to determine which cysteines were responsible for crosslinking, and thus which cysteines lie in close proximity. 79  Figure 3.7. Gel filtration analysis of RNase G. 640 ug wt RNase G, 960 ug  RNase G C402S/C405S/C408S, 150 ug ug RNase G-His6 and 70 ug RNase G C405S/C408S were examined by gel filtration in 25 mM Tris-HCl pH 7.6, 320 mM NaCl on a calibrated Superdex S200 column (Pharmacia) as described in Materials and Methods, section 2.7.8. The standards, marked by •, were thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), aldolase (158 kDa), BSA (67 kDa), ovalbumin (45 kDa) and chymotrypsin (25 kDa) (all standards from Pharmacia). Sizes of the RNase G mutants were estimated from their elution volume (Appendix Fig. 2), and are marked with dotted lines. Wildtype RNase G eluted at 11 mis, RNase G C405S/C408S at 12.7 mis, RNase GHis6 at 12.9 mis and RNase G C402S/C405S/C408S at 13.4 mis. The column was calibrated by R. Pfeutzner.  80  Fig 3.7  Figure 3.8. Structural comparison of RNase E and RNase G, and the  locations of cysteine residues. A schematic of RNase E is shown at the top of the figure, and RNase G at the bottom. The C-terminal scaffolding domain of RNase E is shaded, and the S1 domains of both RNase E and RNase G are indicated by diagonal striping. The numbering above each diagram corresponds to the amino acid residue. All cysteines are identified with stars. In RNase E, cysteines are found at residues 404, 407, 471 and 832. In RNase G, cysteines are located at residues 79, 162, 402, 405, 408 and 421. For complete sequences of RNase G and the N-terminus of RNase E, see Fig. 3.1.  82  83 Physical properties of RNase G cysteine to serine mutations  Each mutant protein was expressed and tested for solubility following centrifugation at 30,000 x g. All RNase G mutants listed in Table 2.2 produced soluble proteins. In contrast, RNase G C79S could not be overexpressed while RNase G C402S/C405S/C408S/C421S and RNase G C405S/C408S/C421S were insoluble or marginally soluble. Introduction of the desired mutations was confirmed by sequencing through the region of interest; several purified RNase G mutants were also analyzed by mass spectrometry (Table 3.2). Oddly, the mass of each His-tagged RNase G determined by mass spectrometry is higher than the expected mass by approximately 14 Da, the mass of a methyl group. Presumably a mutation was been introduced into the parent strain, pDB1, during the PCR-based site-directed mutagenesis reaction (Materials and Methods, sections 2.4 and 2.6.2). The parent plasmid (pDB1) was completely sequenced through the rng coding region, and no mutations were observed. The increased mass could, therefore, be due to a mutation in the C-terminal tag, which was then carried over to the subsequent plasmids encoding RNase G C405S/C408S, C402S and C402S/C405S/C408S. Samples of each soluble mutant were separated on an SDSpolyacrylamide gel (Fig. 3.9). While each preparation was close to homogeneous, each contained traces of contaminating proteins. The increased mobility of wild type RNase G, Fig. 3.9, lanes 14, 15, was due to its smaller size, since it lacked a His6 tag. Western analysis (Fig. 3.10) was used to illustrate that each mutant was recognized by an anti-RNase G polyclonal antibody, while the negative controls, catalase (Fig 3.10, lane 1) and aldolase (Fig 3.10, lane 2) did 84  O  O'  1  2  3  4  5  6  0  7  8  ©  9  10  O  O  &  11 12 13 14 15  Figure 3.9. Purity of RNase G mutants. Duplicate samples of each RNase G mutant preparation (section 2.6.2) were separated on a 10% SDS-polyacrylamide gel, and stained with SYPRO red (Pharmacia) for quantitation. The position of RNase G is indicated to the right of the gel with an arrow. Lane 1 contains standards; lanes 2 and 3, 100 ng RNase GHis6 (section 2.6.3); lanes 4 and 5, 260 ng RNase G C405S/C408S; lanes 6 and 7, 210 ng RNase G C421S; lanes 8 and 9, 200 ng C402S/C405S/C408S; lanes 10 and 11, RNase G C402S; lanes 12 and 13, RNase G C162S; lanes 14 and 15, wild type RNase G (section 2.6.3).  85  ^56 kDa  1 2  3 4 5 6  7 8  Figure 3.10. Western blotting of RNase G variants with  cysteine to serine mutations. 64 ng of each protein sample were separated on a 10% SDS-polyacrylamide gel, blotted to nitrocellulose and probed with anti-RNase G antibodies as described in Materials and Methods, section 2.7.2. Lanes 1 and 2 contain the negative controls catalase and aldolase (Pharmacia).  Lane 3, RNase G C405S/C408S;  lane 4,  C402S/C405S/C408S; lane 5, wild-type RNase G; lane 6, Histagged RNase G; lane 7, RNase G C421S; lane 8, RNase G C402S. The position of migration of RNase G is indicated to the right by an arrow. 86  not bind the antibody. RNase G C421S (Fig. 3.10, lane 7) was either underloaded or poorly recognized by the antibody. To ensure that no gross significant structural changes had been introduced into the mutant RNase G proteins, the midpoint unfolding temperature (Tm) of each RNase G variant was determined by circular dichroism (see Materials and Methods, section 2.7.6). The Tms for each mutant are compiled in Table 3.3. Unfolding was irreversible. All His-tagged RNase G proteins examined displayed significantly lower Tm values compared to wildtype RNase G (Tm=54.8°). RNase G-His6 and RNase G C405S/C408S exhibited Tm values just above 50°. In contrast, RNase G C402S and RNase G C402S/C405S/C408S showed Tm values just below 49°, while RNase G C402S and RNase G C162S had a Tm below 47°. This lowering of the Tm may reflect either a decrease in the stability of the enzyme, or a change in the multimeric state of the protein. Therefore, it can be inferred that addition of the His6 tag reduced the stability of RNase G, and that each of the examined cysteine to serine mutations also led to a modest decrease in structural stability, or these changes affected the multimeric state. Each RNase G mutant was also subjected to oxidative cross-linking and separated  on  a  one-dimensional,  non-reducing  polyacrylamide  gel.  Representative samples are shown in Fig. 3.11 and lane 2 contains a sample of RNase G-His6 for comparison. The pattern of cross-linking was equivalent to that observed in the first dimension of Fig. 3.5a. A cross-linked sample subsequently treated with 0.05 M DTT migrated as a sharp, homogeneous band 87  Table 3.3. Comparison of RNase G mutants. Tm were determined by circular  dichroism and activity was determined as described in Experimental Procedures. The 671 nt and 728 nt pre-rRNA substrates were transcribed from pGM122 and pGM119 linearized with BamHI, respectively (Figure 2.2e, d).  ENZYME  Tm (°C)  Relative Activity  a  b  671 nt  728 nt  9S  wt no tag  54.8 +/- 0.2  100  100  100  wt, His-tagged  50.3+/-0.1  110  190  60  C405S/C408S  50.1 +/- 0.1  160  190  55  C421S  45.1 +/-0.3  40  45  25  C402S/C405S/C408S 48.6+/-0.1  20  45  25  C402S  48.7 +/-0.1  50  60  60  C162S  46.8 +/- 1.0  35  45  25  a  b  Tm values are the average of duplicate experiments Relative Activities are the average of triplicate experiments  88  1  2 3 4  5 6 7  8  Figure 3.11. Crosslinking of RNase G mutants. A onedimensional separation of cross-linked mutant RNase G proteins was performed on a non-reducing SDS-polyacrylamide gel stained with SYPRO Red (Pharmacia). Lane 1 contains size markers; lanes 2-7 contain, respectively: RNase G-His6; RNase G  C405S/C408S;  RNase  G  C421S;  RNase  G  C 4 0 2 S / C 4 0 5 S / C 4 0 8 S ; RNase G C 4 0 2 S and RNase G C 1 6 2 S .  Lane 3 contains approximately 6 pg protein, while lanes 2 and 4 - 7 contain 3 ug. The size of monomeric RNase G (56 kDa) is indicated at left, and the two intramolecular species, bands i and ii, are identified to the right of the panel. The experiment was performed by Janet S. Hankins, and is described in Materials and Methods, section 2.7.5.  89  with an apparent size of 56 kDa (data not shown). Thus, the band migrating slightly faster than the 116 kDa marker in Fig. 3.11, lane 2 would represent apparent dimers. Bands i and ii, migrating slightly faster than monomers (56 kDa), represent intramolecularly cross linked RNase G. RNase G C405S/C408S (Fig. 3.11, lane 3) also formed at least two different intramolecular cross-linked bands. Both migrated to the same position as bands i and ii in Fig. 3.11, lane 2. A series of multimers was also formed by oxidation of RNase G C405S/C408S, notably, apparent dimers that migrated slightly faster than the 116 kDa marker and apparent tetramers (slightly slower than the 200 kDa marker; Fig. 3.11, lane 3). Intermediate bands between dimers and tetramers likely resulted from a mixture of inter- and intramolecular bridged species. The distribution and relative intensities of cross-linked species formed from RNase G C402S did not differ from those in wild-type RNase G and RNase G C405S/C408S (Fig. 3.11; compare lane 6 with lanes 2 and 3). Likewise, RNase G C402S/C405S/C408S, which combines the previous mutations, retained the ability to undergo oxidative dimerization as evidenced by a faint band just below the 116 kDa marker (Fig. 3.11, lane 5). However, bands i and ii migrating ahead of the 56 kDa monomeric band could not be detected. Oxidative cross-linking of RNase G C162S revealed apparent dimers and band ii (Fig. 3.11, lane 7). Taken together, these data imply that upon oxidation different cysteine residues within a given RNase G monomer are capable of forming intermolecular and intramolecular disulphide bonds. Useful information was not obtained for RNase G C421S as it was poorly soluble in the buffer used for cross-linking (Fig. 3.11, lane 4). An interpretation of these data is provided in the Discussion (section 3.3). 90  Figure 3.12. Sedimentation velocity analysis of RNase G mutants. 30 ug of  purified RNase G (untagged; •), RNase G-His6 (•), RNase G C405S/C408S (A) or RNase G C421S (•) were analyzed by sedimentation through a 2 ml 520% w/v sucrose gradient as described in Materials and Methods (section 2.7.6). Panel a depicts the elution profile. Fraction numbers are indicated at the bottom of the panel, and the position of sedimentation of the standards aldolase (158 kDa) and BSA (67 kDa) are indicated with vertical lines. Sedimentation is from right to left. In Panel b, the activity was tested for each mutant. For each sample, 2 pi of wt RNase G (Fraction 14), RNase G-His (Fraction 15), RNase G C405S/C408S (Fraction 15) or RNase G C402S/C405S/C408S were incubated with 60 fmoles labelled monophosphorylated 519 nt RNA (Fig. 2.1) for 4 hrs, as described in Materials and Methods, section 2.8.3. Products of the assay were separated on an 8% polyacrylamide gel. Lanes 1 and 2 are negative controls, showing the substrate at 0 hrs and 4 hrs (T=0 and T=4, respectively). Lane 3 is a sample from the RNase G-His6 assay, lane 4 from the RNase G C405S/C408S assay, lane 5 from the RNase G C402S/C405S/C408S assay and lane 6 from the wt RNase G assay. Intact substrate is indicated by "S", and the major products of endoribonucleolytic cleavage are marked with arrows to the right of the gel (P).  91  92  The sedimentation rates of RNase G, RNase G-His6, RNase G C405S/C408S and RNase G C402S/C405S/C408S were compared to provide an independent measure of their sizes (Fig. 3.12a). RNase G-His6 and RNase G C405S/C408S both sedimented at a rate that would correspond to dimers. RNase G C402S/C405S/C408S sedimented just ahead of BSA, indicating that it forms monomers. These RNase G mutants were also sized by gel filtration chromatography. In Fig. 3.7, RNase G-His6 was estimated to be 105 kDa, and RNase G C405S/C408S as 110 kDa, which both correspond to dimers. In contrast, RNase G C402S/C405S/C408S was estimated to be 85 kDa from Fig. 3.7, which confirmed that this mutant behaved as a monomer. It is important to note that both sedimentation velocity centrifugation and gel filtration separate proteins based on both size and shape. The samples may also undergo some auto-oxidation during the course of the experiment. Sizes of complexes are, therefore, estimates. Activity of wild type RNase G and mutants  The ability of degradosomes, wild type RNase G or RNase G C405S/C408S to cleave two model substrates, 9S RNA and rpsT mRNA, was determined (Fig 3.13). Assays using 9S RNA as substrate are shown in Panel c. Fig. 3.13c lanes 5 and 6 contain degradsosomes, and are a positive control. Bands corresponding to products of cleavage at the "a" site accumulate, along with a fainter band containing pre-5S rRNA, which is the final product of RNase E processing. Despite their strong sequence conservation, mutation of C405 and C408 actually increased activity on these substrates RNase G-His6 (compare 93  Figure 3.13. Comparison of rpsT mRNA and 9S processing sites. Panel a is  a schematic of 9S RNA. 9S RNA undergoes RNase E-mediated cleavage at sites a and b to release pre-5S RNA (stem-loop III). RNase E makes three major cleavages in the rpsT mRNA, and these are indicated in Panel b as sites a-c (minor cleavage sites are not shown). Panels c and d shows the products of endonuclease assay (section 2.8.3) of 0.4 pmoles 9S and rpsT mRNA, respectively, following separation on an 8% urea-polyacrylamide sequencing gel. In panels c and d, lane 1 is the 0 min control. Lane 2 contains products following treatment with 14 pmol RNase G-His6; lanes 3 and 4, 14 pmol RNase G C405S/C408S. Lanes 5 and 6, containing 100 ng wild-type degradosome serve as positive controls. Panel c shows the processing of 9S RNA, and panel d shows the products of cleavage of rpsT mRNA.  94  95  Fig 3.13c, compare lanes 2 and 4). The products resulting from cleavage at the "a" site and pre-5S rRNA are both visible over the course of the assay. Interestingly, significant amounts of pre-5S rRNA product did not accumulate over the course of the assay with ether RNase G-His6 or RNase G C405S/C408S, although products corresponding to cleavage at the "a" site were observed (Fig. 3.13c, lanes 2 and 4). Assays on rpsT mRNA are shown in Fig. 3.13d. Lanes 5 and 6, the positive controls, contain degradosomes. The final 147 nt product of RNase E processing is visible. RNase G and RNase G C405S/C408S showed similar patterns of cleavage of rpsT mRNA to degradosomes(Fig. 3.13d, compare lane 2 and lanes 3 and 4 to 5 and 6). Importantly, all three preparations produced the cleavage at site "c" (Fig. 3.13b) leading to an accumulation ofthe 147 nt product. A more detailed examination of the activity of the RNase G mutants was undertaken, with the results shown in Fig. 3.14 and Table 3.3. Activity was assayed using both monophosphorylated 671 nt and 728 nt substrates (see Fig. 2.1d, e) as well as full length 9S RNA (Fig. 2.1a). The relative activities of each RNase G variant were determined as described in Materials and Methods, section 2.8.1, with results graphed in Fig. 3.14a-c and compiled in Table 3.3. Although none ofthe cysteine to serine mutations resulted in a significant loss of activity, the various forms of RNase G could be sorted into two groups based on their rates of cleavage relative to wild-type RNase G (Table 3.3, line 1). RNase G-His6 and RNase G C405S/C408S form the first group. Both enzymes exhibited moderately increased rates of cleavage of reconstituted pre-16S rRNA substrates relative to wt RNase G, but reduced rates on 9S RNA (Table 2.3, 96  Figure 3.14. Endonuclease activity of RNase G variants with cysteine to  serine mutations. Activities of RNase G preparations (see Table 2.2) on three model RNA substrates were determined. Each assay contained 0.6 pmoles RNA and approximately 18 pmoles RNase G preparation. Panel a shows the rate of disappearance of 5'-monophosphorylated 9S RNA substrate. The substrate is described in Fig. 2.1, and the assay is described in Materials and Methods, section 2.8.3. Panels b and c illustrate the rate of disappearance of the 728 nt (Fig. 2.1) and 671 nt (Fig. 2.1) RNP substrates (section 2.8.1). These data were used to construct Table 3.3, with the rates of each mutant on 9S determined by the time at which 75% of the substrate remained and the time at which 50% of the 728 and 671 nt substrates remained relative to wild-type RNase G. The data points for each mutant are as follows: wild type, untagged RNase G, +; Histagged RNase G, •; RNase G C162S, •; RNase G C402S, *; RNase G C421S, A; RNase G C405S/C408S, • ;  RNase G C402S/C405S/C408S, X.  97  98  lines 2,3). The RNase G variants in the second group (last four entries in Table 2.3), display reduced activities (20-60% of wt) on all three substrates. It is important to note that the rate of substrate disappearance was much lower for allthe RNase G preparations with 9S RNA as the substrate, and enzyme was used in excess over substrate to achieve observable activity. This agrees with earlier reports that found RNase G is incapable of simulating RNase E-mediated processing of 9S RNA when enzyme is not present in excess over substrate (Jiang et al., 2000; Tock et al., 2000). This excess of enzyme may be required if the turnover of the enzyme is slow and/or only a fraction of the preparation is active. The effect of oligomerization on the activity of different RNase G variants was examined. Fractions from the sucrose gradients in Fig. 3.12a were assayed for activity, including wt RNase G (fraction 13), RNase G-His6 (fraction 14), RNase G C405S/C408S (fraction 14) and RNase G C402S/C405S/C408S. The assay is depicted in Fig. 3.12b. It should be noted that an additional band, smaller than the intact substrate, but larger than the primary product (resulting from cleavage at the "E" site, Fig. 2.1), transiently accumulates. This product was observed when the 519 nt RNA was not pre-digested with RNase H and oligo GMV 2011, which mimics the RNase III cleavage (Fig. 2.1). This band is presumably due to accessibility of an additional cleavage site. Each fraction exhibited activity (Fig. 3.12b, compare lanes 3 to 6). Combining this data and the results discussed in section, two interpretations can be made. In the first, dimerization is not required for activity. Dimerization, rather, increases the rate 99  of endoribonucleolytic cleavage. In the second interpretation, dimerization is required for activity. RNase G may exist in a monomer/dimer equilibrium. With the active RNase G variants, RNase G-His6 and RNase G C405S/C408S, the equilibrium favours dimers. This results in a comparitivley high activity. The equilibrium for RNase G C402S/C405S/C408S, however, favours monomers. This results in a smaller proportion of the protein existing in the active, dimeric state, and lowered activity.  3.2.5. Effect ofS1 deletion on RNase G activity  Both RNase E and RNase G contain an S1 domain. This potential RNA binding domain (reviewedby (Coburn and Mackie, 1999)) spans approximately residues 35 to 135 (see Fig. 3.1). Residues 36 to 135 were removed from RNase G (Chapter 2, section 2.4.3) to create a His-tagged protein lacking the S1 domain. RNase GAS1-His6 was purified as described in Chapter 2, section 2.6.3. Due to a lack of solubility, however, the final step in RNase GAS1-His6 purification was the Talon column. This yielded a pool of proteins enriched for RNase GAS1His6, which accounted for approximately half of the protein in the sample. Successful introduction of the deletion was confirmed by sequencing and mass spectrometry (Table 3.2). Assays of RNase G-His6 and RNase GAS1-His6 with triphosphorylated 519 nt substrate (Fig. 2.1) were performed as described (Chapter 2, section 2.8.3). Products of the assay were separated on an 8% polyacrylamide gel, as shown in Fig. 3.15. Activity of RNase GAS1-His6 does not appear to be diminished, compared to RNase G-His6 (compare lanes 1-6 100  RNase G-His6 0  5  RNase GAS1-His6  15 30 45 60  0  5  15 30 45 60 min  ^^^^^ .  ! - » 1 1 f I I 1 2  Figure  3 4 5 6  3.15.  Effect  of  7 8 9 10 11 12  S1  endoribonucleolytic activity.  <4P  deletion  on  RNase  G  Assays combining 0.6 pmoles P 32  labelled, triphosphorylated 519 nt substrate (Fig. 2.1) with 0.55 pmoles RNase G-His6 or RNase GAS1-His6 were performed as described in Chapter 2, section 2.8.3. Products were separated on an 8% polyacrylamide gel and visualized by Phosporimaging. Intact substrate (S) and product (P) are marked with arrows, and time points are indicated along the top ofthe gel.  101  with 7-12) on triphosphorylated substrate. Both enzyme preparations yield the product (P) resulting from cleavage at the "E" site (Fig. 2.2), and RNase GAS1His6 even accumulates products which appear to correspond to cleavage at the "G" site (band below "P" in lanes 10 through 12). The band just below intact substrate likely results from cleavage at a site present in the 519 nt substrate (see section  3.2.6. rRNA processing activity of recombinant RNase G  Examination of the processing of precursors to16S rRNA in vivo has shown that prior cleavage of rRNA precursors by RNase E facilitates the subsequent action of RNase G. RNase G, in turn, forms the mature 5'-end of 16S rRNA (Li et al., 1999b). Templates were prepared for the transcription of derivatives of pre-16S rRNA that would include the 5'-external transcribed sequence (5'-ETS) from the P2 promoter to the mature 5'-end (-173 nt) and the full 5'-domain of 16S rRNA (residues 1-556) as described in Materials and Methods (section 2.8.1) and shown in Figs. 2.1d and 3.16c. Ribonucleoprotein substrates, mimicking the in vivo target of RNase G, were produced by binding the RNA to either ribosomal  protein S20 or S20 and S4 combined. To produce a substrate resembling the product of an initial RNase III cleavage, we used oligonucleotide GMV2011 to direct RNase H to cleave the RNA at or near residues -115/-116 (Fig. 3.16c; Fig 3.16a, compare lanes 1 and 2). This treatment also produces a 5'monophosphorylated terminus on the resultant 519 nt RNA. 102  Figure 3.16. Reconstitution of RNP substrate processing. Panel a shows a 6% polyacrylamide urea-polyacrylamide gel of the processing of a 576 nt RNP. The 576 nt RNP is represented in Panel c. Experiments were performed as described in section 2.8.3, but the initial substrate was 0.6 pmoles triphosphorylated 576 nt RNA arising from transcription of pGM119 linearized with BstUI (see Fig. 2.1). Lanes 2-5 included oligonucleotide GMV2011 (indicated as a horizontal line above the -115 site in Panel c) in the heating step, and 1 U RNase H during the 30 min incubation at 30°C with ribosomal proteins S4 and S20. Lanes 4 and 5 include 18 pmoles RNase G (His-tagged) and lanes 3 and 5 contain 200 ng wild-type degradosomes from strain CF881. The arrows to the left of the panel indicate the position of the starting substrate (576 nt) and the 519 nt product that arises from the oligo-directed cleavage by RNase H (see above), which mimics the RNase III processing event (Panel c). Arrows to the right of the panel indicate the products arising from processing at the "E" ( -430450 nt) and "G" (-375-400 nt) sites (Panel c). Panel b shows the kinetics of formation of the -375-400 nt product from the 519 nt RNA ("precleaved" at the RNase III site): •, RNase G alone; •, degradosomes alone; • , RNase G + degradosomes. This experiment was performed by G. A. Mackie.  103  + + + + RNase H - + - + D-somes - - + + RNase G  1 2  3 4 5  b +->  700  t3> 600 -Q O i- 500 D-4 0 0 tflD < 300 O  |  200 100 0 20  40  60  80  Time (min)  •172 f-  III A -115  G E -r "A~^ A 7T -66  +1  104  +404  Using the ribonucleoprotein substrates described above, the ability of RNase E and RNase G to process the 5'-spacer region was tested. As shown in Fig. 3.16a, lane 3, degradosomes catalyzed cleavage of the 519 nt substrate (115 to +404 in Fig. 3.16c) to yield a major product of -430-450 nt and an unexpected minor product of -375-400 nt. The larger product would correspond to cleavage at or near the known RNase E site at residues -67/-66 in the 5'-ETS (Li et al., 1999b), while the smaller product approximated the size of RNA processed near the mature 5'-end. Digestion of the same substrate with RNase G produced qualitatively similar results, although the rate was slower (Fig. 3.16a, lane 4). In contrast, when present simultaneously, degradosomes and RNase G efficiently produced the smaller product of -375-400 nt (Fig. 3.16a, lane 5). The kinetics of cleavage showed that the rate of accumulation of the smaller product was 8-10-fold faster in the presence of both enzymes (Fig. 3.16b) than with either enzyme alone, suggestive of a cooperative interaction between these enzymes. To improve the resolution and assess the accuracy of cleavage, primer extension was performed on RNA products obtained in an experiment similar to that in Fig. 3.16. With the triphosphorylated 576 nt RNA (-172 to +404, Fig. 3.16c) as substrate, two prominent 5'-ends were detected. The first mapped to A -35 with minor products extending between residues -26 to -43. These products were produced by degradosomes (Fig. 3.17, lane 2), and by RNase G (Fig. 3.17, lane 3), or by both enzymes in concert (Fig. 3.17, lane 4). This major terminus would correspond to a 439 nt product, consistent with the major 105  Figure 3.17. Primer extension analysis of in vitro processing reactions.  This experiment was performed by George A. Mackie. Briefly, 0.6 pmole of the 576 nt RNA transcribed from pGM119 linearized with BstU1 (-172 to +404; Fig. 2.2.E) was assayed as described in Materials and Methods, section 2.8.3. Following 45 min of digestion, reactions were quenched with 0.3 M Na-acetate, 0.01 M EDTA, 0.2% SDS, and 3 pg yeast RNA, prior to extraction with phenolchloroform. Following precipitation and recovery, ~80 fmoles of RNA were annealed to 5'- P-oligonucleotide GMV2013 (5'-TGTTACCGTTCGACTTGC) 32  and the primer was extended using M-MuLV reverse transcriptase (Gibco-BRL), as previously described (Mackie, 1991). The cDNA products were resolved on an 8% sequencing gel containing 8 M urea, together with a sequencing ladder derived from pGM119 primed with 5'-phosphorylated GMV2013. GMV2011 and RNase H were omitted in the digestions analyzed in lanes 1-4 but were present in lanes 5-8 where the substrate would be about 519 nt in length. Processing reactions contained the following enzymes: lanes 1, 5: zero time samples; lanes 2, 6: degradosomes; lanes 3, 7: RNase G; lanes 4, 8: degradosomes and RNase G. The numbering in the right margin gives the nucleotide sequence of the rrnB gene where +1 corresponds to the 5'-end of mature 16S rRNA. Arrows in the left margin give the positions of new 5'-ends created by in vitro processing.  106  f +10  h+20  H+30 +33^  107  product in Fig. 3.16a, lanes 3 and 4. A second prominent 5'-end mapped to residue +33 and was produced by degradosomes either alone or in combination with RNase G (Fig.3.17, lanes 2 and 4). This end would generate a 371 nt product, also consistent with the data in Fig. 3.16a lanes 3 and 5. When the monophosphorylated 519 nt RNA produced by RNase H cleavage (residues 115 to +404) was employed as substrate, the 5'-termini generated by degradosomes or RNase G alone became more heterogeneous (Fig. 3.17, lanes 6 and 7). Following treatment with degradosomes, prominent 5-ends mapped to residues +33, -26, -27, -28, -35, -37 and -39 (Fig. 3.17, lane 6). Although RNase G produced most of the same cleavages, albeit with slightly altered intensities, exceptions include no cleavage at -39 and three additional cleavages at residues -19, -20, and -21. In the presence of both enzymes, the yield of the 5'-end at residue +33 increased and the intensity of other products decreased (Fig. 3.17, lane 8). In no case was a 5'-end detected at the sites of authentic RNase E (-66) or RNase G (+1) cleavage.  3.3. Discussion  Comparative analysis of RNase G and RNase E sequences from a variety of organisms revealed a number of conserved motifs (Fig. 3.1, Table 3.1). The presence of an S1 RNA binding domain is provocative, but this region does not appear to be required for activity at 30° (see below). Loss of the S1 domain, however, may affect cleavage at elevated temperatures. Two well defined temperature-sensitive alleles of RNase E, ams-1 and rne-3071, map to the S1 domain (McDowall et al., 1993). Residue 68, which is changed from leucine to 108  phenylalanine in rne-3071, was conserved in all S1-containing RNase E and RNase G proteins aligned. Residue 66, which is changed from glycine to serine in ams-1, is conserved in 17 of the 27 sequences with S1 domains. While G66 is not completely conserved, RNase G and RNase E sequences lacking a glycine at this position have an alanine instead. Alanine is another small amino acid that differs only by the addition of a methyl group on its side chain, and is likely functionally interchangeable with glycine. The second region of high similarity, which extends from residue 279 to 351, contains three aspartic acid residues that are of interest since they are highly conserved. These are D304 and D347, which were conserved in all of the sequences compared, and D351, which was conserved in 27 of the 29 sequences aligned. Mutational analysis (V. Kunanithy and G. A. Mackie, unpublished) revealed that mutation of D304 or D347 resulted in loss of activity. Mutation of D351 led to a decrease, but not loss, of activity and decrease in Tm. While further analysis is required, it is possible that these acidic residues play a role in coordinating metal ions. This will be discussed further in Chapter 6. In this work, two relatively mild methods for purifying RNase G were developed, both of which yielded nearly homogeneous material. N-terminal sequencing and mass spectrometric analysis of the purified enzyme permitted us to resolve the site of translation initiation for RNase G. Previous investigations employed RNase G artificially initiated from the -18 initiation site and thus added six additional residues to its N-terminus (Lee et al., 2002). Reports of successful high copy suppression of a deletion in rne by rng employed a construct in which RNase G was initiated from the -18 site (Lee et 109  al., 2002). Overexpression of RNase G from the native initiation site, however, allowed only partial suppression of a temperature-sensitive rne allele (Wachi et al., 1997). It was observed previously that RNase G had a tendency to aggregate when its concentration exceeded 1 mg/ml (Tock et al., 2000). I also found that not all of the overexpressed RNase G was soluble and that purified RNase G was somewhat prone to aggregation, as shown by light scattering. These properties can explain why RNase G participates in the formation of cytoplasmic axial filaments when overexpressed in vivo (cited in Tock et al. (2000)). It is also possible that RNase G did not fold correctly, or that it interacts either directly, or indirectly, with the membrane. In this regard, the N-terminal region of RNase E, which shares 35% identity (50% similarity) to RNase G, associates with the membrane in vivo (Liou et al., 2001). Our data indicate that RNase G is a multimer, existing primarily in a dimeric form. Oxidative cross-linking, in particular, showed that RNase G readily formed dimers and higher multimers via surface-exposed cysteine residues. However, the environment within E. coli is a reducing one. Cysteine residues, therefore, would not be involved in covalent protein-protein interactions in vivo. Sedimentation velocity experiments revealed that RNase G sedimented as a dimer. This suggests that RNase E might do likewise, with important implications for the size ofthe RNA degradosome and the relative stoichiometry of its components (Carpousis et al., 1994; Miczak et al., 1996; Py et al., 1994; Py et al., 1996). Our data suggests that when the equilibrium of RNase G is shifted towards the monomeric form, as with RNase G C402S/C405S/C408S, 110  there is a concomitant loss of activity. We conclude, therefore, that dimerization is required for RNase G activity. Cysteine scanning mutagenesis was undertaken in order to map the cysteines involved in inter- and intra-molecular cross-link formation. Cysteines 162, 402, 405 and 408 appear to function largely interchangeably. However, differences in intramolecular disulphide bond formation were detected among the various cysteine to serine mutants tested. Formation of bands i and ii (Fig. 3.5c) required C402, C405 and C408 while band i also required C162. We propose that band i represents a cross-link between one of the three clustered cysteines (C402, C405 and C408) and C162. Band ii would represent the oxidation of one of the three clustered cysteines and an acceptor, potentially C421. There is clearly redundancy among C402, C405 and C408, as loss of C402 or C405 and C408 did not result in any of either intramolecular crosslinking. Although cysteine to serine mutations ought to be highly conservative, most such changes produced a measurable reduction in stability (as measured by Tm), coupled with a partial loss of activity and changes in the monomerdimer-tetramer distribution. These observations, together with the results of oxidative cross-linking, suggest that many of the cysteine residues in RNase G lie in or near subunit interfaces and are adjacent to cysteine residues in a second subunit of RNase G. Data presented in this thesis indicates that the S1 domain of RNase G is dispensable for endoribonucleolytic activity. It has been proposed, however, that the S1 domain may play a role in the 5'-end-dependent activity observed with both RNase E and RNase G (X. Miao and G. A. Mackie, unpublished results). 111  The 5'-end-dependent activity has been ascribed to a hypothetical phosphate pocket, a feature which was first proposed by Mackie (1998). This region could be responsible for the 5'-end dependence of RNase E. Since RNase G also displays 5'-end dependence (Jiang et al., 2000; Tock et al., 2000), it is predicted that RNase G would also contain a phosphate pocket. It has been proposed that the phosphate pocket may lie either within the S1 domain (X. Miao and G. A. Mackie, unpublished results) or in the highly conserved region lying just downstream of the S1 domain (see Fig. 3.2) in both RNase G and RNase E (Gadi Schuster, personal communication). The HSR1 region, however, is lost in RNase GAS1-His6, and this mutant protein may prove valuable in identifying the region responsible for 5'-end-dependence. The presence of degradosomes significantly stimulated the activity of RNase G. This raises the possibility that RNase E and RNase G interact physically during the processing of pre-16S rRNA and is consistent with the observed dependence of 16S rRNA maturation on RNase E (Li et al., 1999b; Wachi et al., 1999). Remarkably, although the sites of cleavage by both RNase E and RNase G on the 576 nt substrates were displaced 3' relative to the physiological sites, the distance between them was relatively well maintained (68nt rather than 66nt). This is consistent with a cooperative interaction between these enzymes. The displacement of cleavage sites 3' to the authentic sites likely reflects misfolding of the RNA template, as illustrated in Fig. 3.18. The 576nt substrate (-172 to +404) would lack helices I, 2 (the central pseudoknot) and 3 (Brimacombe et al., 1988; Schaferkordt and Wagner, 2001). Moreover, the 576 nt substrate is predicted to be single-stranded from residues -9 to 112  5'-PPP(-172)  Figure 3.18. Processing sites observed with reconstituted  RNP substrate. This diagram is an expanded view of pre-16S rRNA from Fig. 1.8, with secondary structure added. Regions of the RNA represented by dotted lines represent sequence omitted for clarity. Numbering is in relation to the mature 5'-end of 16S rRNA. The authentic sites of processing by RNase III (III), RNase E (E) and RNase G (G) are shown by solid arrows. The site for RNase E cleavage is predicted to be occluded by secondary structure in the substrates of Figure 2.1. The positions of the 3' ends of the substrates at +404 or +556 are indicated. Experimentally observed processing sites (E* and G*) are shown by open arrows.  113  beyond +33 using RNAdraw V1.1 (rnadraw.base8.se/). The RNase G cleavage between +32 and +33 was completely suppressed in a 728nt substrate extending from -172 to +556 (data not shown). Helix 3 is predicted to form in this RNA, thereby "burying" the +32/+33 site. Our data suggest, therefore, that RNase G possesses minimal primary sequence specificity, in agreement with Tock et al. (Tock et al., 2000). The specificity of RNase G must, therefore, be dictated by RNA structure and its accessibility to a single-stranded RNA target. The sensitivity of processing by RNase G to RNA folding may also explain early findings that pre-16S rRNA had to be incorporated into ribosomes for correct 5'processing to mature 16S rRNA (Dahlberg et al., 1978).  114  4. RELATIONSHIPS WITHIN THE RNA DEGRADOSOME 4.1. Introduction  A thorough understanding of RNase E is vital to elucidating cellular processes such as gene expression since it provides the bulk of the endoribonuclease activity in E. coli. We have used RNase G as a surrogate for investigating the activity and quaternary structure of RNase E (Chapter 3). RNase G, however, lacks the C-terminal scaffolding domain of RNase E (see Fig. 3.8) that is required for assembly of the RNA degradosome (Vanzo et al., 1998). The RNA degradosome is composed of RNase E, an endoribonuclease, the 3'- to 5'-exonuclease PNPase, RhIB, which is an RNA helicase and enolase, a glycolytic enzyme of unknown function in RNA metabolism (Carpousis et al., 1994; Py et al., 1996). The importance of the degradosome has been a topic of debate. Previous studies illustrated that only the N-terminal catalytic region of RNase E is needed to maintain cell viability (Jain and Belasco, 1995b; Kido et al., 1996; Taraseviciene et al., 1995). However, deletion of the C-terminal half of RNase E, and thus the loss of formation of the degradosome, led to a significant decrease in the decay rate of RNA I and some other mRNAs (Leroy et al., 2002). Recent studies have also demonstrated that mutant strains that are incapable of forming the degradosome could not compete with a wild type strain (Leroy et al., 2002). Thus, while formation of the degradosome is not required for viability, it does confer a significant advantage to cells capable of forming this macromolecular machine.  115  The physical interactions between RNase E and the components of the degradosome were delineated previously (Vanzo et al., 1998). At the inception of these studies, however, little was known about the functional interaction ofthe individual enzymes in the complex with each other. These interactions were investigated by isolating degradosomes from RD100, an E. coli strain which carries the pnp13 allele, resulting in the formation of a full-length, but functionally deficient PNPase (Reiner, 1969). This allowed investigation ofthe importance of both the endonucleolytic and exonucleolytic functions of the RNA degradosome. In particular, it enabled distinction between 5'-modes of decay (described in Chapter 1, section 1.2.1) and 3'-modes of decay. In a 3'-tethering model of mRNA decay, the initial step of mRNA decay would be recognition of the 3'-end of the mRNA by PNPase, dependent upon a poly (A) tail. PNPase would then progress until stalling at secondary structure, at which point RNase E would loop around, interact with the 5'-end and initiate endonucleolytic decay (Hajnsdorf et al., 1996; Ingle and Kushner, 1996; reviewed in Coburn and Mackie, 1999). The main weakness of this model is that it would result in the accumulation of truncated protein products (Coburn and Mackie, 1999). Availability of a degradosomal preparation lacking active PNPase also allowed a study of the impact of exonucleolytic activity on the function of the RNA helicase, RhIB.  4.2. Results 4.2.1. Purification of degradosomes from a pnp13 strain  Degradosomes that were deficient in PNPase activity were purified in order to distinguish between endoribonucleolytic and exoribonucleolytic processing 116  events. Degradosomes were isolated from the E. coli strain RD100, which carries the pnp13 allele. This is a missense mutation that reduces PNPase activity to less than 5% of wild type (Reiner, 1969). The purification of the pnp13 degradosomes is described in section 2.6.4, and is identical to the purification used for wt degradosomes from strain CF881 (Coburn et al., 1999). An electrophoretic analysis of the final pnp13 degradosome preparation is shown in Fig. 4.1, lane 3. The composition and apparent stoichiometry of the four major RNA degradosomal components are shared between degradosomes isolated from wild type (CF881) and pnp13 (RD100) strains (Fig. 4.1, compare lanes 2 and 3). Both preparations contain RNase E, (118 kDa; but migrates at 180 kDa on SDS-polyacrylamide gels (McDowall and Cohen, 1996)); PNPase (77 kDa; compare also to lane 6); RhIB (47 kDa; compare to lane 8); and enolase (46 kDa). There is an extra, faint band present in the wild type preparation (Fig. 4.1 lane 2 between the 116 and 97 kDa size markers). This band is variable from preparation to preparation and is likely a proteolytic fragment of RNase E (Mudd and Higgins, 1993).  4.2.2. Endonucleolytic cleavages by wild type and pnp13 degradosomes  Endoribonucleolytic assays (section 2.8.3) were performed on rpsT mRNA to ensure that the endonucleolytic activity of pnp13 degradosomes was unaltered from that of wild type degradosomes. Processing products were separated on a  117  Mr (kDa)  1  2  3  4  5  6  7  8  Figure 4.1. Purification of pnp13 degradosomes. Samples of  wild type degradosomes, pnp13 degradosomes and purified degradosomal proteins were separated on a 10% SDSpolyacrylamide gel and visualized with Coomassie Brilliant Blue R250 (section 2.7.1). This figure originally appeared in Coburn et al., (1999) and is reprinted with permission. Lane 1 contains size markers, with sizes indicated to the left of the panel, lane 2 contains 15 u.g wild type degradosomes; lane 3, 12 u.g pnp13 degradosomes; lane 4, 1.5 u.g purified RNase E; lane 5, 1.5 uxj RNase E A208 (lacks aa 1-207); lane 6, 1.5 ug PNPase; lane 7, 1.5 u.g RNase II and lane 8 contains 1.5 u.g RhIB.  118  polyacrylamide gel, with the results shown in Fig. 4.2b. The intact substrate is 372 nt in length, and appears in the 0 min lanes of both wild type and pnp13 processing reactions (Fig. 4.2b lanes 1 and 9, respectively). The processing products generated by wild type and pnp13 degradosomes appear to be very similar. In particular, the 257 nt transient intermediate that results from cleavage at residue 190/191 (Mackie, 1992)(Fig. 4.2a, site "a"), accumulates in lanes 4 to 6 (wild type) and lanes 10 to 12 (pnp13). Disappearance of this product was concomitant with the accumulation of a 147 nt product (Fig. 4.2b lanes 7 - 8 and 13-16). This 147 nt band is the result of cleavage at residues 300/301/302 (Fig. 4.2a, "b" site; Mackie, 1992). Accumulation of these products indicated that endonucleolytic activity was preserved in both degradosomal preparations, and that they appeared to process RNA substrates at similar rates. Since the rates of decay of rpsT mRNA were not decreased by the presence of an inactive form of PNPase, it follows that decay of this mRNA can be initiated by degradosomes independently of 3' to 5' exonuclease activity.  4.2.3. PNPase and the pnp13 degradosome  Phosphate is required for PNPase activity (Deutscher and Reuven, 1991). The effect of adding sodium phosphate (Na-phosphate) to endonuclease assays was examined using BR13, a 13 residue oligoribonucleotide (see Fig. 5.2), in the presence of degradosomes isolated from wild type or pnp13 strains. Fig. 4.3a illustrates the effect of Na-phosphate on wild type degradosomes. In lanes 1 through 6, which lack Na-phosphate, product slowly accumulates. Smearing of the bands is likely due to salts carried over from the assay. This, and insufficient 119  Figure 4.2. Endonuclease activity of wild type and pnp13 degradosomes on rpsT mRNA. Endoribonuclease assays were performed on rpsT mRNA as  described in section 2.8.3 using 150 ng of degradosomes. A schematic of the rpsT mRNA, with RNase E processing sites indicated is found in Panel a.  Products ofthe assay were separated on a 15% polyacrylamide sequencing gel, which is shown in Panel B. The corresponding products of cleavage are indicated to the right of the panel. Lanes 1 through 8 contained wild type degradosomes, while lanes 9 through 16 contained pnp13 degradosomes.  120  121  Figure 4.3. Phosphate-stimulated activation of PNPase in degradosomes.  PNPase activity of wild type and pnp13 degradosomes was assessed on 2 pM 5'-end labelled oligonucleotide BR13 (see section 2.8.2 and Figure 5.1). Endoribonuclease assays were performed as described in section 2.8.3 and products separated on a 15% polyacrylamide gel. Panel a shows the assay products following digestion with 200 ng of wild type degradosomes at the times indicated above each lane, while Panel b follows an assay containing 2 pg of pnp13 degradosomes. In both panels, assay samples represented in lanes 1  through 6 were performed as described in section 2.6.3, while samples in lanes 7 through 12 are from an endonuclease assay which also contained 30 mM Naphosphate. Substrate and endonucleolytic products are indicated to the right of the panel with "S" and "P".  122  a 0  - phosphate 5 10 20 40 90  0  1  2  7  0  - phosphate 5 10 20 40 90  0  + phosphate 5 10 20 40 90  1  2  7  8  3  4  5  6  + phosphate 5 10 20 40 90  8  mins  9 10 11 12  b  3  4  5  6  123  9 10 11 12  mins  time of separation, make an estimation of the product size difficult. Lanes 7 through 12 (30 mM Na-phosphate) display a rapid disappearance of substrate, without a concomitant accumulation of endonucleolytic products. Thus, it appears the substrate is rapidly degraded to small products that run off of the gel during electrophoresis. In contrast, in experiments performed with pnp13 degradosomes as the source of ribonucleases, there is no obvious distinction between lanes 1 - 6 , which lack Na-phosphate, and lanes 7 - 1 2 , which contain Na-phosphate (Fig. 4.3b). In both assays with pnp13 degradosomes, substrate disappeared over time and the 8 nt product of endoribonucleolytic cleavage accumulated. Additionally, the rates of disappearance of BR13 in the presence of pnp13 degradosomes either with or without Na-phosphate are comparable (Fig. 4.3b, compare lanes 1-6 with 7-12). Glen A. Coburn subsequently performed experiments on a second RNA substrate, malEF RNA (Fig. 4.4), to verify the lack of PNPase activity in pnp13 degradosomes. These results appeared originally in Coburn et al. (1999). Fig. 4.5 Panels a and b compare the action of wild type and mutant degradosomes on malEF RNA. In Fig. 4.5a, the two intermediates," * "(Py et al., 1996) and  RSR (McLaren et al., 1991) transiently accumulated, but were ultimately degraded. Control experiments (Coburn et al., 1999) showed that formation of both intermediates required PNPase and phosphate, while their disappearance required ATP and RhIB in addition to PNPase and Pi. In contrast, Fig. 4.5b shows that pnp13 degradosomes were incapable of processing the malEF substrate, although a faint product visible at 45 or 60 min of digestion is likely 124  Figure 4.4. Secondary structure of the malEF RNA substrate. The secondary  structure of malEF RNA, determined using RNAdraw, has been published previously (Coburn et al., 1999). The 5' and 3' ends are identified, and residues 1 through 207 are represented with a line. The large stem loop to the left is the repetitive extragenic p_alindrome (REP) stem-loop structure (Py etal., 1996). The two sites of PNPase stalling are indicated with "*" and RSR, where RSR represents the REP-stabilized RNA (McLaren et al., 1991; Py et al., 1996).  125  REP AA G A C C 250— C - G A-U A  A  A  A  A - U —260 C-G A-U U-A C-G C-G 240— G - C G-C C U C-G U - A —270 G U U-A U-A  I  G C-G 230— G " C C - G —280 A-U A A G C U G G-C C-G G U G-C 220— C " G G-C -290 U-A  320  I  U  I GA  A ^ A "'C-G' G - C  G"C. ft-U A U-A C-G 310—A-U C-G G - C G-C G - C AAAU - A G U U G U U G U C - G G  I  210  I  I  300  340  g:g  >~  \f u A U Q C G U-A C - G C - G C - G C - G C - G —360 U"A AGAA - U A C C G A G C U C G A A U U ^  A G G  A  I  |  350  RSR  126  I  370  Figure 4.5. PNPase activity of pnp13 degradosomes. PNPase activity was  assessed for wild type and pnp13 degradosomes. Experiments are described in Coburn et al. (1999). Assays were performed as described in section 2.8.3, with the exceptions that the final assay volume was 40 pi; the reactions were carried out in the presence of 10 mM Na-phosphate and 3 mM ATP. A schematic of the substrate, malEF RNA, appears in Fig. 4.4. Processing products are indicated to the right of each panel. Panel a depicts the processing of malEF RNA by 200 ng/ml wild type, and Panel b depicts the processing of malEF RNA by 200 ng/ml pnp13 degradosomes. Panel c shows the products of processing of 200 ng/ml pnp13 degradosomes with 60 ng/ml PNPase added in trans. These assays were  performed by Glen A. Coburn, and the figure is reprinted with permission from Coburn etai. (1999).  127  3  wt Degradosomes +ATP 0 2.5 5 10 15 20 30 45 60  §  (min)  § • „ m m m  ^ malEF mm mm mm  -  ^  R  S  R  pnp-13 Degradosomes +ATP 0 2.5 5 10 20 30 45 60  (min)  4 malEF  pnp-13 Degradosomes +Pnp +ATP 0  1  2  5 10 20 30 45 60  m *B m ~ w v  w  128  (min)  < ma/EF -4 * RSR  due to RNase E activity. To determine if this lack of exonucleolytic processing could be complemented, purified wild type PNPase was added to the reaction in trans (Fig. 4.5c). Over the course of the assay, the first intermediate  (*)  accumulated transiently (Fig. 4.5. Panel a 2.5 and 5 min, and Panel c 1 and 2 min) and the RSR intermediate accumulated stably over the course of the reaction. In contrast to wild type degradosomes, however, this was the end product of the reaction, and complete degradation of the RSR intermediate was not observed. PNPase could not complement the pnp13 mutant in trans. It appears that PNPase must be coordinated with RhIB in degradosomes to activate RhIB, and that PNPase does not cycle on and off of the degradosome, since functional PNPase could not displace inactive pnp13 PNPase in the reaction.  4.2.4. Phosphatase activity of the degradosome  During these studies of the degradosome-directed endonucleolytic cleavage of 5'-end labelled oligoribonucleotides (section 2.6.2), it was noted that label was lost during endoribonuclease assays. Originally, this was assumed to be due to exonucleolytic degradation of products. Isolation of pnp13 degradosomes, which lack PNPase activity (see above, section 4.2.3) was undertaken to remove exonucleolytic activity from the degradosome. Endoribonuclease assays (section 2.8.3) were performed with pnp13 degradosomes and 5'-end labelled oligonucleotide BR10 (see Fig. 5.1, 5.2). Products of the reaction are depicted in Fig. 4.6a, lanes 1-6. The amount of P label remaining was determined by 32  quantifying total label in the substrate (S; 10 nt) and product (P; 5 nt) bands by 129  Figure 4.6. Phosphatase activity of degradosomes. Degradosomes were analyzed for 5'-dephosphorylation activity. In Panel a, products arising from an endonuclease assay (section 2.8.3) containing 1 pg pnp13 degradosomes, 0.2 pM 5'-monophosphorylated BR10 RNA oligonucleotide substrate (see section 2.8.2 for substrate preparation, and Fig. 5.1 for BR10) and 0 (lanes 1-6), 5 mM (lanes 7-12) or 1 mM sodium fluoride (NaF; lanes 13-18). Separation was on a 15% polyacrylamide sequencing gel. Substrate (10 nt) and cleavage products (5 nt) are indicated to the right of the panel with "S" and "P" respectively. Remaining label was determined by summing the recoveries of substrate and product using a phosphorimager and ImageQuant software (Molecular Dynamics). These data are found in Panel b: •, 0 NaF; •, 5 mM NaF; A, 1 mM NaF. In Panel c, y P-ATP (0.17 pCi/pl) was incubated with either 0.1 U/pl calf 32  intestinal phosphatase or 400 ng pnp13 degradosomes as described in section 2.8.6. Products (8 pi) were then separated by thin layer chromatography on polyethyleneimine-impregnated plates in 0.375 M KH P0 (section 2.6.6). The 2  4  origin is marked to the right by "O" and the products by "P". Lane 1 is the positive control, which followed incubation for 45 min in the presence of 0.1 U/pl calf intestinal phosphatase; lane 2 is the negative, time zero control and lane 3 contains the products resulting from incubation with 400 ng pnp13 degradosomes for 10 min (section 2.8.6).  130  a  c  131  phosphorimaging and subsequent analysis using ImageQuant software (Molecular Dynamics). These data are plotted in Fig. 4.6b. Despite the lack of exonuclease activity, there was still significant loss of label over the course of the assay (Fig. 4.6b, •). The oligonucleotide substrate was labelled with P at 32  its 5'-end, and it was therefore possible that the loss of label was due to the action of a phosphatase. Since RNase E is 5'-end dependent (Mackie, 1998), it was possible that RNase E removed the 5'-phosphate. Phosphatase activity is inhibited by the addition of sodium fluoride (NaF)(Gilboe and Nuttall, 1976; Yeaman and Cohen, 1975), so endonuclease digestions were performed in the presence of 5 and 1 mM sodium fluoride (Fig. 4.6a, lanes 7-12 and 13-18, respectively). Comparing these to the control, it was noted that addition of NaF did not inhibit the loss of label. It was, therefore, unlikely that RNase E was behaving as a phosphatase. This was confirmed by comparing the products of Y P-ATP treated with calf intestinal alkaline phosphatase (New England 32  Biolabs) or pnp13 degradosomes by thin layer chromatography (Fig. 4.6c and section 2.8.6). Fig. 4.6c lane 1 is the positive control, and it shows the liberation of free phosphate (P) by calf intestinal alkaline phosphatase. Fig. 4.6c, lane 2 is the time zero negative control, and two spots are visible. The first is intact v P32  ATP, and is located at the origin (O), and the second is a faint band just below free phosphate (P). This second band was also observed with thin layer chromatography  of products from the reaction between BR10 and  degradosomes, as shown in Fig. 4.7c, lanes 6-10. Importantly, this second band is also present in the negative control (Fig. 4.6c lane 2) that contained degradosomes and y P-ATP, but lacked any incubation time. This second band 32  132  is therefore likely a contaminant of the y P-ATP and the loss of label is not due to degradosome-mediated phosphatase activity liberating free phosphate from Y P-ATP. 32  The source of the contaminating band found in Fig. 4.6c lanes 2 and 3 was sought by precipitating the oligonucleotides following labelling, but prior to performing the endoribonuclease assay (section 2.8.2). Fig. 4.7a confirmed that precipitation had no effect on the products of the endonuclease assays (Fig. 4.7a, compare lanes 1-5 and 6-10). Quantification of the amount of label remaining (Fig. 4.6b) revealed that precipitation also had no effect on the loss of 5'-label, as expected (compare unprecipitated, "•" and precipitated "A"). Assay products were then separated by thin layer chromatography (Fig. 4.7c). Precipitation of the oligonucleotide (Fig. 4.7c lanes 1-5) removed the contaminating band, which is still visible in lanes 6-10, where the oligonucleotides were not precipitated prior to assay. By comparing Fig. 4.7c lanes 6-10 and Fig. 4.6c lane 2 (y P-ATP) it appears that a radioactive 32  contaminant was present in the labelled ATP. Possible explanations for the loss of label will be described in the Discussion, section 4.3.  4.3. Discussion  Degradosomes were successfully isolated from E. coli strain RD100, which carries the pnp13 allele that leads to the inactivation of PNPase (Reiner, 1969). The composition of these degradosomes was equivalent to those isolated from strain CF881 "wild type" (Fig. 4.1 lanes 2 and 3). The endonucleolytic activity of  133  Figure 4.7. Phosphatase activity of pnp13 degradosomes on ethanol-  precipitated oligoribonucleotides. Panel a is a 15% sequencing gel of the products arising from endonuclease assays (section 2.8.3) containing 0.4 pM BR13 (see Fig. 5.1) and 50 ng pnp13 degradosomes. In lanes 1-5 (ppte), BR13 was precipitated with ethanol (section 2.8.2) prior to assaying while the BR13 in lanes 6 through 10 (unppte) was not precipitated. Intact substrate (S) and products (P) are indicated to the right of the panel. As with Fig. 4.5, total label remaining was calculated for each time point in Panel a, and graphed in Panel b. Data points for precipitated samples (A) and unprecipitated (•) samples are as indicated. Assay samples were also prepared and separated by thin layer chromatography using polyethyleneimine impregnated plates and K HP0 as 2  4  the solvent (section 2.8.6). The direction of migration is indicated with an arrow. Each lane contains 5 pi of sample, with lanes 1 - 5 containing samples from assays using precipitated BR13, and lanes 6 - 1 0 samples using unprecipitated BR13.  134  c  Precipitation  i  0  1  5  2  10 15 30  3  4  5  0  6  5  7  10 15 30 min  8  9  Precipitation  +  II  1  0 5 101530 0 5 10 1530 min  10  b O)  c c  100  'co  E  CD  i—  75  "55  n <S  50 0  10  20  30  40  Time (min)  135  1 2 3 4 5 6 7 8 9  10  pnp13 degradosomes was fully preserved, with cleavage patterns mimicking  those of wild type degradosomes on both rpsT mRNA (Fig. 4.2) and 9S RNA (data not shown). These results illustrated that active PNPase associated with the degradosome had no effect on the overall rate of rpsT mRNA cleavage. This is a strong argument against the 3'-tethering model, which postulates that the initial step in mRNA decay is recognition of the 3'-end (see section 4.1). This model, which depends upon addition of single-stranded poly (A) tails, is further weakened by several arguments. First, the major enzyme involved in polyadenylation, PAP I is not an essential enzyme and is not a component of purified degradosomes (Masters et al., 1993). Secondly, the rate of decay of mRNA, both in vivo and in vitro, is strongly dependent upon the state of the 5'end (Mackie, 1998; Mackie, 2000). Finally, the stability conferred upon a model mRNA substrate by blocking the 5'-end with a stable stem loop can be overcome by introducing an efficient RNase E cleavage site within the message, whereupon cleavages proceed in a 5' to 3' manner (Baker and Mackie, 2003). Two separate sets of experiments were performed to verify the inactivity of PNPase in pnp13 degradosomes. In the first set of experiments, phosphate was added to endoribonuclease assays with oligoribonucleotides and degradosomes. Addition of phosphate led to a greatly increased rate of disappearance. No endoribonuclease cleavage products accumulated when wild type degradosomes were the source of ribonuclease activity. In contrast, phosphate had no effect on either rates or product accumulation with pnp13 degradosomes. The loss of product accumulation observed with wild type degradosomes  was  presumably  due 136  to  the  degradation  of  the  oligoribonucleotide to mononucleotides and limit oligos (Niyogi and Datta, 1975; Yu and Deutscher, 1995) that would migrate out of the gel. Since the accumulation of product was not phosphate-dependent, it was concluded that pnp13 degradosomes were devoid of PNPase exonuclease activity, despite the  presence of a full complement of PNPase in the degradosome (Fig. 4.1, lane 3). In the second set of experiments, Glen A. Coburn examined the degradosomemediated 3'-degradation of malEF RNA (Coburn et al., 1999). The interpretation of these experiments is depicted Fig. 4.8. It was previously established that PNPase processed malEF to produce two major intermediates, both of which are caused by exonuclease stalling at stable stem-loop structures. The first intermediate (*) (Py ef al., 1996) is transient, and the second intermediate, RSR (McLaren et al., 1991), is stable in the absence of RhIB and ATP. In the presence of ATP, wild type degradosomes were capable of completely degrading malEF RNA (Fig. 4.8, pathway 1; Coburn et al., 1999). In contrast, pnp13 degradosomes did not appreciably degrade malEF (Fig. 4.8, pathway 2).  Addition of exogenous active PNPase would have two possible outcomes. If the degradosome disassembles when not engaged on a substrate, and reassembles de novo with each round of decay, wild type PNPase could become incorporated into the complex and pathway 3, which ultimately leads to complete degradation, would be followed. The other possibility is that degradosomes remain intact (Fig. 4.8, pathway 4). If the pnp13 degradosomes remained intact, PNPase would only be capable of degrading malEF to the stable RSR intermediate because RhIB requires association with PNPase  137  Figure 4.8. Models for malEF mRNA degradation by wild type and pnp13 degradosomes with active PNPase. The intact malEF substrate is depicted in  the top left of the Panel, with the two sites of PNPase stalling marked by "*" and "RSR" (see Fig. 4.4). In the degradosomes, RNase E is represented by a shaded circle, RhIB by a black triangle, wild type PNPase by a trimer of striped circles and pnp13 by a trimer of speckled circles. Enolase has been omitted from the degradosomes for simplicity and the model is illustrated independently of the 5'-tethering model (section, Fig. 1.3). Pathway 1 follows the complete degradation of malEF by wild type degradosomes, while Pathway 2 illustrates that pnp13 degradosomes are incapable of degrading the RNA (see also Fig. 4.5, Panels a and b). The pnp13 degradosomes may then take one of two different pathways, both of which involve complementation with active PNPase. In Pathway 3, pnp13 degradosomes fall apart, and new degradosomes, some of which contain active PNPase assemble either in the buffer (shown) or on the RNA itself. These new degradosomes then follow Pathway 1. In Pathway 3, however, pnp13 degradosomes remain intact, and malEF RNA can only be degraded to the stable RSR intermediate.  138  REP  REP  pnp13  Pathway 2 no decay Pathway 4  Pathway  T degradosome remains intact, falls off message  degradosome disassembles  •  wt PNPase  REP  o  degradosome stalls  5'.. RSR  REP  degradosome reassembly  y  5'..  REP  o  /  RSR  5'...._ RhIB unwinds stable structure  RSR PNPase stalls, RhIB cannot unwind secondary structure in trans  139  through the C-terminus of RNase E for activation (Coburn et al., 1999; Vanzo et al., 1998). Our experiments showed that the RSR intermediate was, in fact, stable upon addition of wild type PNPase to pnp13 degradosomes (Fig. 4.5c). Since wild type PNPase cannot functionally replace the inactive pnp13 mutant on the degradosome, it follows that degradosomes are very stable, and do not disassemble and reassemble during successive rounds of processing. RNase E is an end-dependent endonuclease, with a strong preference for 5'-monophosphorylated ends. Theoretically, mRNA decay could be initiated by a phosphatase or pyrrophosphase acting to remove the 5'-terminal y and P phosphates thus increasing RNase E activity on the substrate (Mackie, 1998). When assaying 5'-end labelled oligoribonucleotides, a loss of label over time was noted (Fig. 4.6b). It was hypothesized that this might be due to phosphatase activity. Data, however, suggested that degradosomes do not act as phosphatases. The cause of the label loss could not be determined from the experiments which were performed. All buffers used for assays and RNA preparation were treated with DEPC, which inhibits RNases (Sambrook, 1989), although it is still possible that these were a source of contamination. Another possibility is that the labelled RNA, which is fairly small, displayed affinity for the reaction tube over the course of the assay or precipitated after boiling prior to loading the gel. The substrate may also have become bound to a component of the degradosome and was not released. The resulting oligonucleotide-protein complex was then incapable of migrating into the gel, and the label was therefore not recovered.  140  5. ACTIVITY STUDIES OF RNase E AND RNase G USING SYNTHETIC OLIGONUCLEOTIDES 5.1. Introduction  Studies of the specificity of RNase E have not resolved all the ambiguities of what constitutes an RNase E cleavage site. Earlier investigations using natural substrates of 250 to 400 nts, summarized in Chapter 1, section, revealed that RNase E prefers to cleave 5' to AU dinucleotides (Cormack and Mackie, 1992; Ehretsmann et al., 1992; Mackie, 1992; Mackie and Genereaux, 1993; McDowall et al., 1994). It was also established that flanking secondary structure affected cleavage efficiency and specificity (Cormack and Mackie, 1992; Mackie, 1992; Mackie and Genereaux, 1993). The influence of secondary structure on site recognition and cleavage was subsequently examined using a synthetic decaribonucleotide (BR10), based on the 5'-end of RNA1, that contains an RNase E cleavage site (McDowall et al., 1995)(Fig. 5.1). This work established that secondary structure was not required for recognition and cleavage of an RNA substrate. Thus, it is believed flanking secondary structure serves to anchor cleavage sites in a single-stranded form which is recognizable by the enzyme (Mackie and Genereaux, 1993). These experiments also revealed that synthetic oligonucleotides could provide a powerful tool for investigating the activity of processing ribonucleases. Interestingly,  however, very few  investigators since have utilized oligoribonucleotides to examine properties of endoribonucleases from E. coli (Feng et al., 2002; Tock et al., 2000). Nonetheless, cleavage of oligonucleotides illustrated that RNase G, like RNase  141  RNase E 5'  A C A G U A U U U G G U A U C U G G C U  RNase 5' T 3' A C A G U A U U U G  C U G  BR 10 also  5  CUUC  G G GA CA G U AU UU G V  - C -A - U U - G -A - C - C - G -A  A G U „ U G  RNA 1  3  G G A A A A A G A G U „  G G C AC AC AU C A AG AU AU C cU CC AG cA cU GC U  G U U U U U U U U U G G U G G C G A . U  3'  BR 13  Figure 5.1. RNA1 and RNA oligonucleotides. The complete sequence of RNA1 is outlined with the secondary structure indicated. RNase E cleaves precisely 5 nt from the 5'-end as indicated. The synthetic oligonucleotide BR10 was constructed based on the first 10 nucleotides of the 5' sequence of RNA1, while BR13 contained three additional G residues at the 5'-end which would mimic transcription from a T7 promoter (McDowall et al., 1995). This sequence was used as a basis for the construction of the oligonucleotides listed in Fig. 5.2.  142  E, does not require secondary structure for endonucleolytic processing (Tock et al., 2000). These experiments also established that RNase G and RNase E might exhibit slightly altered sequence specificities. Studies using chimeric synthetic oligonucleotides incorporating ribonucleotides and modified 2'-0methyl ribonucleotides established that RNase E could cleave a sequence with a single ribonucleotide on either side of the cleavage site (Feng et al., 2002). These chimeric oligos were also used in attempts to determine the directionality of RNase E and RNase G (Feng et al., 2002). Using synthetic oligonucleotides, I sought to refine previous experiments and determine the minimum requirement for recognition and cleavage by both RNase E and RNase G. A strong competitor for endoribonuclease activity that could be utilized in future experiments to probe the active site of either RNase E or RNase G was also sought.  5.2. Results 5.2.1. Design of chimeric oligonucleotides  The ability of RNase E to cleave an oligonucleotide based on a known cleavage site was demonstrated previously (McDowall et al., 1995). In their study, which is outlined in Fig. 5.1, a 10 nt oligoribonucleotide (BR10) was constructed based on the first 10 nucleotides of RNA1, a naturally occurring antisense RNA whose abundance is controlled by RNase E (Lin-Chao and Cohen, 1991). The first 10 nt of RNA1 contain an RNase E cleavage site between residues 5 and 6. A second oligoribonucleotide, BR13, which incorporates 3 extra G residues at the 5'-end of BR13, corresponding to transcription from a T7 promoter, was also 143  constructed. RNase E was capable of cleaving both BR10 and BR13 at the same location observed in full length RNA1 (McDowall et al., 1995). In contrast to RNase E-mediated cleavage of BR10, RNase G processes BR10 not only at the predicted site of cleavage, but also 3' to residues 6 and 7; i.e., 1 and 2 nucleotides 3' to the RNase E cleavage site (Tock et al., 2000). These additional cleavages are indicated in Fig. 5.4. Utilizing this information, a series of synthetic oligonucleotides based on the sequence of BR10, which incorporated both ribo- and deoxyribonucleotides, were constructed. Using these oligonucleotides, the chemical features of a substrate required for recognition and cleavage by both RNase E and RNase G were sought. Specifically, I sought to establish how many ribonucleotides are required at a site of RNase E/G cleavage, and if ribonucleotides must be located 5'-, 3'- or both 5'- and 3'- to the scissile bond. The oligonucleotide sequences are listed in Fig. 5.2. Ribo 3 and 4 contain ribonucleotides straddling the cleavage site, Ribo 5 consists of ribonucleotides 3' to the putative cleavage site, Ribos 6 and 7 contained ribonucleotides 5' to the cleavage site and Ribo 8 has a single deoxynucleotide at the extreme 3'-end to minimize the action of 3' to 5'-exonucleases.  5.2.2. Activity of RNase E against chimeric substrates.  The activity of RNase E on Ribos 3-8 was determined, with the results illustrated in Fig. 5.3a, b. Ribo 8 differs from BR10 by incorporation of a single deoxynucleotide at its 3'-end but otherwise serves as a positive control. This modification did not affect cleavage of Ribo 8, and the cleavage product (P) clearly accumulated during the reaction (Fig. 5.3a, lanes 1-6). In this reaction, no 144  RNase E  3'  BR 10  a c a gu a u u ug  BR 13  g g g a c a g u a u u ug  RIB0 3  AC AGu aTTTG  RIB0 4  A CA gu a uTTG  RIB0 5  AC A G T a u u ug  RIB0 6  a ca gu A T T T G  RIB0 7  AC AGu A T T T G  RIB0 8  a c a gu a u u uG  Figure 5.2. Synthetic oligonucleotide sequences. A number  of oligonucleotides were constructed based on the sequence of BR10 (see Fig. 5.1). The location of the RNase E cleavage site of BR10 is indicated with an arrow and a vertical line. Upper case letters are deoxyribonucleotides, while the lower case shaded letters are ribonucleotides. The oligonucleotide name is indicated to the left.  145  Figure 5.3. Activity of degradosomes and RNase G on oligonucleotides.  Endonuclease assays (section 2.6.3) were performed on 5'-end labelled oligonucleotides (section 2.6.2). The sequence of the oligonucleotide is indicated above each assay set, with the sequences also depicted in Fig. 5.2. Time of digestion is indicated above each lane. Approximate positions of migration of the substrate (S) and products (P) are located to the right of the panels. Panels a and b contained 2 pg pnp13 degradosomes (section 2.4.4) and 0.2 pM oligonucleotide. In Panel c, 0.4 pM oligonucleotides were digested with 200 ng His-tagged RNase G in a final volume of 30 pi (section 2.4.2).  146  Ribo 8 acaguauuuG 0  5  Ribo 3 A C A G u aTTTG  10 20 60 120  0  5 10 20 6 0 1 2 0  Ribo 4 ACAguauTTG 0  5 10 20 6 0 1 2 0 min  H P mm  1  2  3  4  5  6  7  Ribo 7 A C A G u ATTTG  b  8  9 10 11 12  ~  13 14 15 16 17 18  Ribo 6 acagu ATTTG  Ribo 5 ACAGTauuug  0  5 10 20 6 0 1 2 0 0  5 10 20 6 0 1 2 0  0  1  2  8  13 14 15 16 1 7 1 8  3  4  5  6  7  Ribo 3 A C A G u aTTTG  c  9 10 11 12  Ribo 5 ACAGTauuug  0  5  10 20 4 5 90  0  5 10 20 4 5 90 0  1  2  3  7  8  4  5  6  9  147  5 10 20 6 0 1 2 0  min  Ribo 7 A C A G u ATTTG 5  10 20 4 5 90 min  10 11 12 13 14 15 16 17 18  intact substrate was detectable at 20 min, and recovered product was stable, although the amount of label recovered was less than the initial label in the reaction. The loss of label over the course of endonuclease assays with oligonucleotides was addressed in Chapter 4. It appears that products are either lost through precipitation, interacting with the surface of the plasticware or associating with proteins in the assay. Ribos 3 and 4 are largely DNA but contain two or four ribonucleotides, respectively, which straddle the expected cleavage site (Fig. 5.2). RNase E was able to process both of these substrates resulting in the formation of product (Fig. 5.3a lanes 7-12 and 13-18). Processing of these substrates occurs at a slightly lower rate than that of Ribo 8; however, little or no intact substrate was visible after 20 min. This implies that ribonucleotides straddling the cleavage site are sufficient to allow RNase E cleavage. The ribonucleotide requirement of RNase E for recognition and cleavage was further refined using Ribo 7, which contains a single ribonucleotide 5' to the potential site of cleavage (Fig. 5.2). The cleavage of Ribo 7 was very slow, with no significant accumulation of 5 nt product until about 60 min (Fig. 5.3b, lanes 16). Ribo 5 and Ribo 6 were constructed to determine the effect of deoxynucleotides 5' (Ribo 5) and 3' (Ribo 6) of the potential RNase E cleavage site (Fig. 5.2). With ribonucleotides 5' to the cleavage site (Ribo 6), intact substrate disappeared at a rate which was slower than that observed with Ribo 8 (Fig. 5.3, compare panel b lanes 7-12 with panel a lanes 1-6). The loss of substrate cannot be unambiguously ascribed to RNase E activity as the  148  recovery of product was low in the assay. No cleavage of Ribo 5 was observed (Fig. 5.2b, lanes 13-18). It is important to note that RNase E does not cleave at any secondary sites when the position 5' to the expected cleavage site is blocked by addition of deoxynucleotides.  5.2.3. Activity of RNase G against chimeric substrates.  Processing of Ribos 3, 5 and 7 by RNase G was also examined (Fig. 5.2c). RNase G was capable of cleaving Ribo 3, which has a single ribonucleotide on either side of the site of RNase E cleavage (Fig. 5.3c, lanes 1-6). However, close examination reveals that the major (lower) product band is, in fact, two separate cleavage products. These are shown in Fig. 5.4. The major product (5 nts), which is the only product observed upon RNase E treatment, arises from cleavage between residues 5 and 6 and the minor product (6 nts) from processing between residues 6 and 7 (G and +1 from Fig. 5.4, respectively). Ribo 5 was also fully processed by RNase G to yield a 6 nt product (Fig. 5.3c, lanes 7-12). In contrast, Ribo 5 was resistant to RNase E (compare to Fig. 5.3b, lanes 13-18). This result is consistent with previous findings that RNase G cleaves BR10 not only at the same position as RNase E, but also at positions 1 and 2 nucleotides 3' of the RNase E cleavage site (Tock et al., 2000). We did not detect a 7 nt product of Ribo 5 cleavage corresponding to cleavage at +2 in Fig. 5.4. These products are shown in Fig. 5.4 (+1 and +2, respectively). The size of the product observed following Ribo 5 treatment with RNase G is larger than the major product observed with Ribo 3 (Fig. 5.3, compare lane 6 with lane 12), and it likely corresponds to cleavage at the +1 site (Fig. 5.4) yielding a 6 nt 149  product. Ribo 7 (which has a single ribonucleotide 5' to the site of RNase E cleavage) was almost completely resistant to RNase G (Fig. 5.3c, lanes 13-18), although the assay was only continued for 90 min, compared to 120 min for the RNase E assay.  5.3. Discussion  The role of the 2'-OH of ribose for substrate recognition and cleavage by both RNase E and RNase G was examined in this study. These findings are illustrated in Fig. 5.4. In another study, published after these experiments had been completed, it was found that ribonucleotides straddling the cleavage site were sufficient for RNase E-mediated cleavage of an oligonucleotide (Feng et al., 2002). These results were confirmed and the minimal requirements for recognition and cleavage by both RNase E and RNase G further elucidated. It was discovered that RNase E requires at least a single ribonucleotide immediately 5' to the site of cleavage (Fig 5.4, residue 5). However, the rate of cleavage of substrates with a single ribonucleotide 5' to the site of cleavage was very low relative to the positive control (Fig. 5.2, compare panels a and b, lanes 1-6). RNase G cleaves this substrate extremely slowly (Fig. 5.3c lanes 13-18), which implies it is more sensitive to ribonucleotide content. The rate of cleavage for both RNase E and RNase G increased when two ribonucleotides flanked the cleavage site (Fig. 5.4, residues 5 and 6). As discussed in Chapter 6, a 2'-OH  150  additional RNase G sites E/G 7^-; • +1 +2 5  A  C A G(Q) A U U U G ' 3  8 must be a ribonucleotide for cleavage at "E/G" site  9  10  a ribonucleotide 3' of cleavage site enhances the rate of the upstream cleavage  Figure 5.4. Requirements for RNase E and RNase Gmediated cleavage of BR10. The 5' and 3' ends of the BR10  synthetic oligoribonucleotide (Fig. 5.1) are indicated, with residue numbers located below. The RNase E cleavage site, which can also be cleaved by RNase G, is labelled "E/G" and marked with a large arrow. Smaller arrows identify RNase G cleavage sites at +1 and +2 (numbered in relation to the E/G site).  151  should not be necessary for catalysis. Thus, the role of the 2'-OH may be in controlling sugar conformation or in direct recognition by RNase E/G. In addition to cleaving at the RNase E site of BR10, RNase G can also cleave at two additional sites (see Fig. 5.4, sites +1 and +2)(Tock et al., 2000). When the nt 5' of the major RNase E cleavage site (Fig. 5.4, residue 5) is replaced with a deoxynucleotide (Ribo 5), RNase G cleaves the +1 site. Interestingly, when RNase G is assayed with Ribo 3, which has ribonucleotides flanking the E/G site, the major product is 5 nts This corresponds to cleavage at the E/G site. However, a minor product of 6 nts is also visible in the reaction, and this corresponds to processing at the +1 site. These data affirm the earlier observation that sites which are flanked by ribonucleotides are cleaved more efficiently than those with a single ribonucleotide 5' of the cleavage site. To more accurately assess the requirements for RNase G cleavage, a substrate with a single processing site would be preferred. In the course of their studies, Feng et al. (2002) found that 2'-0methylated nucleotides straddling an expected RNase E site would prevent cleavage by RNase E. Our data imply that a single deoxynucleotide 5' to the scissile bond is sufficient to prevent cleavage. This hypothesis could be tested with a chimeric oligo based on BR10 (Fig. 5.1) with a single deoxynucleotide replacing the ribonuleotide at position 5. These studies have defined the chemical requirements for recognition and cleavage by RNase E of a model substrate. Minimally, a single ribonucleotide 5' to the site of cleavage is sufficient, but the efficiency of cleavage improves when ribonucleotides straddle the scissile bond. These 2'152  OH groups at the scissile bond are likely involved in recognition and/or binding of substrate, rather than contributing directly to the cleavage reaction. A possible mechanism for activity will be described in Chapter 6.  153  6. PERSPECTIVES  This thesis has addressed several fundamental aspects concerning the structure and activity of RNase E and RNase G. Two simple purification techniques for RNase G have been developed, and established the value of this system as a model for RNase E activity and structure. Findings revealed that RNase G, and by implication RNase E, exists primarily as a dimer. This has ramifications for the stoichiometry ofthe degradosome. Two subunits of RNase E would exhibit a molecular weight of approximately 240 kDa; therefore, early size estimates of the degradosome of between 160 and 460 kDa are likely underestimates (Carpousis etai., 1994). Determination ofthe stoichiometry ofthe degradosome, however, is required to make more accurate approximations of its size. Dimerization of RNase E supports the validity of the "5'-tethering model", which is described in Chapter 1, section This model requires RNase E to exist as at least a dimer, which our findings support, although the data does not conclusively validate the model. The predicted structure of RNase G, based on findings presented in this thesis, is shown in Fig. 6.1. All six cysteine residues (indicated in Fig. 6.1) were mutated, in turn, to serine. None of these residues, including two highly conserved cysteine residues (C405 and C408) was catalytically significant, although cross-linking data revealed they lie near the interface of protein subunits. Significantly, C402, C405, C408 and C421 form a cluster, which can form intramolecular oxidative crosslinks, and any one of these residues can also form an intramolecular disulphide bond with C162. Intramolecular crosslinks are  154  Figure 6.1. Schematic structure of RNase G. In Panel a, the proposed N and C terminal domains of RNase G are shaded in grey, and the S1 domain is represented by an open oval. Cysteine residues are presented in striped circles labelled with the corresponding residue numbers. The two aspartic acid residues (D304 and D347) implicated in endonucleolytic activity are represented as open boxes. The site of trypsin cleavage at arginine 314 (R314) is located in a linker region. The complete sequence of RNase G is shown in Fig. 3.1. Panel b shows the interaction between two subunits of RNase G, each identical to that in Panel a, but shown in cross-section. Subunit I is shaded, and subunit two represented as open ovals. Only the five cysteine residues implicated in crosslinking are depicted. Cysteine residue numbers are indicated in the open boxes.  155  a NH  402, 405, 408,421  156  + 3  shown in Fig. 6.1a. In Fig. 6.1b, a second subunit of RNase G interacts with the domain illustrated in Fig. 6.1a in such a way that the cysteines of the second subunit are in very close contact to those of the first. This allows intermolecular disulphide bond formation under oxidative conditions. Limited trypsin digestion of RNase G resulted in cleavage at arginine residue 314 (M. A. Cook., J. S. Hankins and G. A. Mackie, unpublished results). This residue must, therefore, be accessible, and we propose that it is located in a linker region, which separates two domains of RNase G. Future work should concentrate on cyrstallizing these regions and determining their structure. Recently in our lab, V. Kunanithy used the approach of mutation and purification established in this thesis to show that two conserved aspartic acid residues in RNase G, D304 and D347 (Fig. 6.1) are required for activity (V. Kunanithy and G. A. Mackie, unpublished results). Her findings are consistent with the hypothesis that RNase G utilizes a two-metal ion mechanism for catalysis. This mechanism was proposed for the cleavage reactions observed with RNA splicing and RNase P hydrolysis of precursor RNA. The model, illustrated in Fig. 6.2, was based on the crystallographic structure of the DNA polymerase I 3'- 5' single-strand specific exonuclease from E. coli (Steitz and Steitz, 1993). Four acidic amino acid residues are required to stabilize two metal ions that, in turn, activate the attacking water and stabilize the transition state. Importantly, this substrate cleavage yields a 3'-product with a 5'monophosphorylated end and 5'- product with a 3'-OH. Similar products are observed following RNase E cleavage (Misra and Apirion, 1979). Analysis of the products of RNase G activity have not been examined but should be identical. 157  Figure 6.2. Two-metal-ion mechanism for endonucleolytic cleavage. The  two nucleotides surrounding the scissile bond of an RNA substrate are indicated by base 1 and base 2, and the substrate continues both 5' and 3' of the cleavage site (curvy lines). The two metal ions are represented by "M " and the ++  four acidic residues are indicated in boxes beside the carboxyl groups (D or E). The top panel shows the mechanism of cleavage, and the bottom panel the final cleavage products. This figure is modified from Steitz and Steitz (1993).  158  DorE  159  This mechanism differs from that of RNase A. RNase A utilizes a 2'-OH in the reaction, but yields a 3'-product with an unphosphorylated 5'-end (reviewed by Voet and Voet, 1990). While not directly involved in the catalytic mechanism, the 2'-OH group may aid in substrate recognition or coordination within the active site. Future analyses to determine the presence of metal ions in RNase G would either support or disprove the two-metal-ion mechanism of action for the enzyme. This mechanism requires four acidic residues, two of which (D304 and D347) have been mutated resulting in decreased enzyme activity. There are two strong candidates for the other acidic acid residues, E298 and E326 (see Fig. 3.1). These two glutamic acid residues should be mutated in turn and the resulting proteins assayed for endoribonuclease activity. The analogous mutations could then be introduced into RNase E to examine the effect on viability, growth and activity. Isolation and analysis of a pnp13 degradosome, which contains a mutant, inactive PNPase, allowed us to show clearly that degradosomes do not disassemble with each new round of 3'-degradation. The recycling of intact degradosomes, rather than de novo assembly on each new substrate, strengthens the argument for a 5'-tethering model of decay. In the "5'-tethering model" (also know as the "alternating sites" model) RNase E must be at least a dimer, either alone, or in the degradosome complex. The active site and phosphate-binding pocket are identical, but function alternatively. Once RNase E is bound to a substrate, it does not dissociate or, if it does, it rebinds preferentially to the cleavage product (Chapter 1, section, Fig. 1.2) (Coburn and Mackie, 1999). If the enzyme complex fully dissociates after each 160  cleavage, support for this model is weakened. However, we have also presented evidence that RNase G, and by extension RNase E, exists as at least a dimer, and these two findings strengthen the argument for a 5'-tethering model. Chapter 3 presented preliminary activity studies of an S1 deletion mutant of RNase G. It was established that the mutant protein retained activity. Unpublished evidence indicates that the S1 domain may contain the phosphate pocket, which confers 5'-end dependence to RNase E (X. Miao and G. A. Mackie, unpublished data; G. Schuster, personal communication). In future studies, RNase GAS1-His6 should be assayed on 5'-tri-, mono- and dephosphorylated RNA substrates and the rates compared, to determine if this is in fact, the region responsible for 5'-end-dependence. In Chapter 5, chimeric DNA/RNA oligonucleotides were used to establish the chemical requirements for recognition and cleavage of substrates by RNase G and RNase E. Another goal of this analysis was to identify potential inhibitors of RNase E/G activity. Fig. 6.3 shows the findings when DNA oligonucleotides were added to endoribonuclease assays using labelled 9S RNA substrate. In Fig. 6.2 panels a and b, gels of assay products are shown, and the corresponding amounts of substrate remaining at each time point are indicated in panels c (RNase G) and d (RNase E). Accumulation of pre-5S RNA was limited over the course of the assay (indicated by arrow), although a product, marked by "P", accumulated that corresponds to cleavage at the "a" site of 9S RNA (Fig. 3.13a). The DNA oligonucleotide utilized in these reactions was present in large excess to enzyme. Its sequence was based on that of BR10,  161  Figure 6.3. Oligonucleotide competitors of RNase G and RNase E activity.  Competition assays were performed (section 2.8.5) with 0.45 pmol labelled 9S and 69 pmol of the DNA competitor oligonucleotide which was based on the sequence of BR10 (5'-ACAGTATTTG-3\ see Fig. 5.2) in a final reaction volume of 30 pi. Products of the assay were separated on 8% polyacrylamide gels. The band corresponding to intact 9S RNA substrate (depicted in Fig. 3.13a) is indicated (S) to the right of each panel, as is the position of pre-5S RNA and the product (P) of a single cleavage at the "a" site of 9S RNA. The time of each assay point is indicated above each lane. P-DNA lanes followed competition with cold 5'-monophosphorylated DNA (P-DNA; section 2.8.2) and the HO-DNA lanes represented assays which contained DNA with no 5'-phosphate. In Panel a, the source of endonuclease was 50 ng His-tagged RNase G, while in Panel b it was 20 ng gel purified RNase E. The amount of substrate remaining at each time point was calculated and graphed in Panel c, for the RNase G assays, and in Panel d for the RNase E assays.  162  RNase G 0  30 60  P-DNA 0 30 60  • •  RNase E  HO-DNA 0 30 60  min  0  30 60  m » m m m>Ms  0  P-DNA 30 60  Hfe  0  HO-DNA 30 fin  min  mm. mm, mmt *m+ mm  mm mm  Mm* ^mw  mm  mr  ^S  IP I pre-5S  1 2  3  4  8  pre-5S  1  9  2  3  8  9  100  100 Dl C  _£ '£  're E  re E £ o  5  •»E -» 05  CA .Q 3  .C  3) HO-DNA 2) P-DNA no competitor  3) HO-DNA 2) P-DNA 1) no competitor  ~s  Time (min)  Time (min)  163  which is in turn based on the 5'-end of RNA 1 (Fig. 5.1 and 5.2)(McDowall et al., 1995). From these data, it was determined that 5'-OH DNA oligonucleotides are very strong inhibitors of RNase E/G activity. Similar inhibition was observed with all DNA oligonucleotides examined, and inhibition was observed with as little as a three-fold excess of inhibitor to enzyme (data not shown). Further experiments are required to determine the mode and efficiency of inhibition. If these oligonucleotides are established as strong inhibitors of RNase E/G activity, they may aid in crystallographic efforts and/or illuminating the active site or nucleotide binding regions. In the course of this thesis, some of the mysteries concerning the structure and specificity of RNase E and RNase G were solved. By establishing RNase G as a model for RNase E activity, the groundwork for future projects has been laid. In the short term, the focus should be on mutating highly conserved residues in RNase G and determining their effects on endonucleolytic processing. Mutations that affect substrate processing can then be introduced into RNase E, and the effects on activity examined. This would enable us to utilize RNase G as a simple model for screening residues of interest. Longer term efforts will also focus on determining the crystal structure of RNase G. A structure should shed light on the catalytic region of RNase E.  164  7. REFERENCES  Apirion D. (1973) Degradation of RNA in Escherichia coli. A hypothesis. Moi Gen Genet 122, 313-322.  Apirion D. and D. R. Gitelman. (1980) Decay of RNA in RNA processing mutants of Escherichia coli. Moi Gen Genet  177,  339-343.  Apirion D. and A. Miczak. (1993) RNA processing in prokaryotic cells. Bioessays 15, 113-120.  Aristarkhov A., A. Mikulskis, J. G. Belasco and E. C. Lin. (1996) Translation of the adhE transcript to produce ethanol dehydrogenase requires RNase III cleavage in Escherichia coli. J Bacteriol  178,  4327-4332.  Babitzke P., L. Granger, J. Olszewski and S. R. Kushner. (1993) Analysis of mRNA decay and rRNA processing in Escherichia coli multiple mutants carrying a deletion in RNase III. J Bacteriol  175,  229-239.  Baginsky S., A. Shteiman-Kotler, V. Liveanu, S. Yehudai-Resheff, M. Bellaoui, R. E. Settlage, J. Shabanowitz, D. F. Hunt, G. Schuster and W. Gruissem. (2001) Chloroplast PNPase exists as a homo-multimer enzyme complex that is distinct from the Escherichia coli degradosome. RNA 7, 1464-1475.  165  Baker K. E. and G. A. Mackie. (2003) Ectopic RNase E sites promote bypass of 5'-end-dependent mRNA decay in Escherichia coli. Moi Microbiol 47, 7588.  Bardwell J. C , P. Regnier, S. M. Chen, Y. Nakamura, M. Grunberg-Manago and D. L. Court. (1989) Autoregulation of RNase III operon by mRNA processing. EMBO J 8, 3401-3407.  Ben-Bassat A., K. Bauer, S. Y. Chang, K. Myambo, A. Boosman and S. Chang. (1987) Processing of the initiation methionine from proteins: properties of the Escherichia coli methionine aminopeptidase and its gene structure. J Sacfe/Vo/169, 751-757.  Blum E., B. Py, A. J. Carpousis and C. F. Higgins. (1997) Polyphosphate kinase is a component of the Escherichia coli RNA degradosome. Moi Microbiol 26, 387-398.  Bouvet P. and J. G. Belasco. (1992) Control of RNase E-mediated RNA degradation by 5-terminal base pairing in E. coli. Nature 360, 488-491.  Briant, D. J., J. S. Hankins and G. A. Mackie (2003) Purification, properties and function of RNase G from Escherichia coli. Manuscript in revision.  166  Brimacombe R., J. Atmadja, W. Stiege and D. Schuler. (1988) A detailed model ofthe three-dimensional structure of Escherichia coli 16 S ribosomal RNA in situ in the 30 S subunit. J Moi Biol 199, 115-136.  Bycroft M., T. J. Hubbard, M. Proctor, S. M. Freund and A. G. Murzin. (1997) The solution structure of the S1 RNA binding domain: a member of an ancient nucleic acid-binding fold. Ce//88, 235-242.  Carpousis A. J., G. Van Houwe, C. Ehretsmann and H. M. Krisch. (1994) Copurification of E. coli RNAase E and PNPase: evidence for a specific association between two enzymes important in RNA processing and degradation. Cell 76, 889-900.  Casaregola S., A. Jacq, D. Laoudj, G. McGurk, S. Margarson, M. Tempete, V. Norris and I. B. Holland. (1992) Cloning and analysis of the entire Escherichia coli ams gene, ams is identical to hmp1  and encodes a 114  kDa protein that migrates as a 180 kDa protein. J Moi Biol 228, 30-40.  Casaregola S., V. Norris, M. Goldberg and I. B. Holland. (1990) Identification of a 180 kD protein in Escherichia coli related to a yeast heavy-chain myosin. Moi Microbiol 4, 505-511.  167  Chen L. H., S. A. Emory, A. L. Bricker, P. Bouvet and J. G. Belasco. (1991) Structure and function of a bacterial mRNA stabilizer: analysis of the 5' untranslated region of ompA mRNA. J Bacteriol 173, 4578-4586.  Cheng Z. F. and M. P. Deutshcer. (2002) Purification and characterization of the Escherichia coli exoribonuclease RNase R. Comparison with RNase II. J Biol Chem 277, 21624-21629  Cheng Z. F., Y. Zuo, Z. Li, K. E. Rudd and M. P. Deutscher. (1998) The vacB gene required for virulence in Shigella flexneri and Escherichia coli encodes the exoribonuclease RNase R. J Biol Chem 273, 14077-14080.  Coburn G. A. and G. A. Mackie. (1998) Reconstitution of the degradation of the mRNA for ribosomal protein S20 with purified enzymes. J Moi Biol 279, 1061-1074.  Coburn G. A. and G. A. Mackie. (1999) Degradation of mRNA in Escherichia coli: an old problem with some new twists. Prog Nucleic Acid Res Moi Biol 62, 55-108.  Coburn G. A., X. Miao, D. J. Briant and G. A. Mackie. (1999) Reconstitution of a minimal RNA degradosome demonstrates functional coordination between a 3' exonuclease and a DEAD-box RNA helicase. Genes Dev 13, 2594-2603. 168  Condon C. and H. Putzer. (2002) The phylogenetic distribution of bacterial ribonucleases. Nucleic Acids Res 30, 5339-5346.  Cormack R. S., J. L. Genereaux and G. A. Mackie. (1993) RNase E activity is conferred by a single polypeptide: overexpression, purification, and properties of the ams/rne/hmp1 gene product. Proc Natl Acad Sci USA 90, 9006-9010.  Cormack R. S. and G. A. Mackie. (1992) Structural requirements for the processing of Escherichia coli 5 S ribosomal RNA by RNase E in vitro, d Moi Biol 228, 1078-1090.  Dahlberg A. E., J. E. Dahlberg, E. Lund, H. Tokimatsu, A. B. Rabson, P. C. Calvert, F. Reynolds and M. Zahalak. (1978) Processing ofthe 5' end of Escherichia coli 16S ribosomal RNA. Proc Natl Acad Sci U S A 75, 3598-  3602.  Deutscher M. P. and N. B. Reuven. (1991) Enzymatic basis for hydrolytic versus phosphorylytic mRNA degradation in Escherichia coli and Bacillus subtilis. Proc Natl Acad Sci USA88,  169  3277-3280.  Diwa A., A. L. Bricker, C. Jain and J. G. Belasco. (2000) An evolutionarily conserved RNA stem-loop functions as a sensor that directs feedback regulation of RNase E gene expression. Genes Dev 14, 1249-1260.  Doi M., M. Wachi, F. Ishino, S. Tomioka, M. Ito, Y. Sakagami, A. Suzuki and M. Matsuhashi. (1988) Determinations of the DNA sequence of the mreB gene and of the gene products of the mre region that function in formation of the rod shape of Escherichia coli cells. J Bacteriol 170, 4619-4624.  Donovan W. P. and S. R. Kushner. (1986) Polynucleotide phosphorylase and ribonuclease II are required for cell viability and mRNA turnover in Escherichia coli K-12. Proc Natl Acad Sci U S 4 83, 120-124.  Dunn S. D. (1986) Effects of the modification of transfer buffer composition and the renaturation of proteins in gels on the recognition of proteins on Western blots by monoclonal antibodies. Anal Biochem 157, 144-153.  Ehretsmann C. P., A. J. Carpousis and H. M. Krisch. (1992) Specificity of Escherichia coli endoribonuclease RNase E: in vivo and in vitro analysis  of mutants in a bacteriophage T4 mRNA processing site. Genes Dev 6, 149-159.  170  Feng Y., T. A. Vickers and S. N. Cohen. (2002) The catalytic domain of RNase E shows inherent 3' to 5' directionality in cleavage site selection. Proc Natl Acad Sci U SA99, 14746-14751.  Georgellis D., S. Arvidson and A. von Gabain. (1992) Decay of ompA mRNA and processing of 9S RNA are immediately affected by shifts in growth rate, but in opposite manners. J Bacteriol 174, 5382-5390.  Ghora B. K. and D. Apirion. (1978) Structural analysis and in vitro processing to p5 rRNA of a 9S RNA molecule isolated from an rne mutant of E. coli. Ce//15, 1055-1066.  Ghosh S. and M. P. Deutscher. (1999) Oligoribonuclease is an essential component of the mRNA decay pathway. Proc Natl Acad Sci USA 96, 4372-4377.  Gilboe D. P. and F. Q. Nuttall. (1976) Phosphohistone and glycogen synthase D phosphatase activities in a liver glycogen pellet preparation. Physiol ChemPhysB, 405-415.  Gopalan V., A. Vioque and S. Altman. (2002) RNase P: variations and uses. J Biol Chem 277, 6759-6762.  171  Granger L. L, E. B. O'Hara, R. F. Wang, F. V. Meffen, K. Armstrong, S. D. Yancey, P. Babitzke and S. R. Kushner. (1998) The Escherichia coli mrsC gene is required for cell growth and mRNA decay. J Bacteriol 180,  1920-1928.  Grunberg-Manago M. (1999) Messenger RNA stability and its role in control of gene expression in bacteria and phages. Annual Review of Genetics 33, 193-227.  Hajnsdorf E., F. Braun, J. Haugel-Nielsen, J. Le Derout and P. Regnier. (1996) Multiple degradation pathways of the rpsO mRNA of Escherichia coli. RNase E interacts with the 5' and 3' extremities of the primary transcript. Biochimie 78, 416-424.  Hansen M. J., L. H. Chen, M. L. Fejzo and J. G. Belasco. (1994) The ompA 5' untranslated region impedes a major pathway for mRNA degradation in Escherichia coli. Moi Microbiol 12, 707-716.  Hayes R., J. Kudla, G. Schuster, L. Gabay, P. Maliga and W. Gruissem. (1996) Chloroplast mRNA 3'-end processing by a high molecular weight protein complex is regulated by nuclear encoded RNA binding proteins. EMBO J 15, 1132-1141.  172  Ingle C. A. and S. R. Kushner. (1996) Development of an in vitro mRNA decay system for Escherichia coli: poly(A) polymerase I is necessary to trigger degradation. Proc Natl Acad Sci U S A 93, 12926-12931.  Jager S., O. Fuhrmann, C. Heck, M. Hebermehl, E. Schiltz, R. Rauhut and G. Klug. (2001) An mRNA degrading complex in Rhodobacter capsulatus. Nucleic Acids Res 29, 4581-4588.  Jain C. and J. G. Belasco. (1995a) Autoregulation of RNase E synthesis in Escherichia coli. Nucleic Acids Symp Ser 33, 85-88.  Jain C. and J. G. Belasco. (1995b) RNase E autoregulates its synthesis by controlling the degradation rate of its own mRNA in Escherichia coli: unusual sensitivity of the rne transcript to RNase E activity. Genes Dev 9, 84-96.  Jiang X., A. Diwa and J. G. Belasco. (2000) Regions of RNase E important for 5'-end-dependent RNA cleavage and autoregulated synthesis. J Bacteriol 182, 2468-2475.  Kaberdin V. R., A. Miczak, J. S. Jakobsen, S. Lin-Chao, K. J. McDowall and A. von Gabain. (1998) The endoribonucleolytic N-terminal half of Escherichia coli RNase E is evolutionarily conserved in Synechocystis sp.  173  and other bacteria but not the C-terminal half, which is sufficient for degradosome assembly. Proc Natl Acad Sci U S A 95, 11637-11642.  Kaberdin V. R., A. P. Walsh, T. Jakobsen, K. J. McDowall and A. von Gabain. (2000) Enhanced cleavage of RNA mediated by an interaction between substrates and the arginine-rich domain of E. coli ribonuclease E. J Moi B/o/301, 257-264.  Kaga N., G. Umitsuki, K. Nagai and M. Wachi. (2002) RNase G-dependent degradation of the eno mRNA encoding a glycolysis enzyme enolase in Escherichia coli. Biosci Biotechnol Biochem 66, 2216-2220.  Kalman M., H. Murphy and M. Cashel. (1991) rhIB, a new Escherichia coli K-12 gene with an RNA helicase-like protein sequence motif, one of at least five such possible genes in a prokaryote. New Biol 3, 886-895.  Kelly K. O. and M. P. Deutscher. (1992) The presence of only one of five exoribonucleases is sufficient to support the growth of Escherichia coli. J Bacteriol MA, 6682-6684.  Kido M., K. Yamanaka, T. Mitani, H. Niki, T. Ogura and S. Hiraga. (1996) RNase E polypeptides lacking a carboxyl-terminal half suppress a mukB mutation in Escherichia coli. J Bacteriol 178, 3917-3925.  174  Klingenberg, M., and Appel, M. (1989) The uncoupling protein dimer can form a disulfide cross-link between the mobile C-terminal SH groups. Eur J Biochem  180: 123-131.  Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.  Lee K., J. A. Bernstein and S. N. Cohen. (2002) RNase G complementation of rne null mutation identifies functional interrelationships with RNase E in Escherichia coli. Moi Microbiol 43,  1445-1456.  Leroy A., N. F. Vanzo, S. Sousa, M. Dreyfus and A. J. Carpousis. (2002) Function in Escherichia  coli of  the non-catalytic part of RNase E: role in  the degradation of ribosome-free mRNA. Moi  Microbiol 45,  1231-1243.  Li Z. and M. P. Deutscher. (1995) The tRNA processing enzyme RNase T is essential for maturation of 5S RNA. Proc  Natl Acad Sci U S A  92, 6883-  6886.  Li Z. and M. P. Deutscher. (1996) Maturation pathways for E. precursors: a random multienzyme process in  vivo. Cell  coli  tRNA  86, 503-512.  Li Z. and M. P. Deutscher. (2002) RNase E plays an essential role in the maturation of Escherichia  coli tRNA  175  precursors. RNA 8, 97-109.  Li Z., S. Pandit and M. P. Deutscher. (1999a) Maturation of 23S ribosomal RNA requires the exoribonuclease RNase T. RNA 5, 139-146.  Li Z., S. Pandit and M. P. Deutscher. (1999b) RNase G (CafA protein) and RNase E are both required for the 5' maturation of 16S ribosomal RNA. EMBO J 18, 2878-2885.  Lin-Chao S. and S. N. Cohen. (1991) The rate of processing and degradation of antisense RNAI regulates the replication of ColE1-type plasmids in vivo. Cell 65, 1233-1242.  Lin-Chao S., T. T. Wong, K. J. McDowall and S. N. Cohen. (1994) Effects of nucleotide sequence on the specificity of rne-dependent and RNase Emediated cleavages of RNA I encoded by the pBR322 plasmid. J Biol Chem 269, 10797-10803.  Liou G. G., W. N. Jane, S. N. Cohen, N. S. Lin and S. Lin-Chao. (2001) RNA degradosomes exist in vivo in Escherichia coli as multicomponent complexes associated with the cytoplasmic membrane via the N-terminal region of ribonuclease E. Proc Natl Acad Sci U S A 98, 63-68.  Loewen C. J. and R. S. Molday. (2000) Disulfide-mediated oligomerization of Peripherin/Rds and  Rom-1  in photoreceptor 176  disk membranes.  Implications for photoreceptor outer segment morphogenesis and degeneration. J Biol Chem 275, 5370-5378.  Lopez P. J., J. Guillerez, R. Sousa and M. Dreyfus. (1997) The low processivity of T7 RNA polymerase over the initially transcribed sequence can limit productive initiation in vivo. J Moi Biol 269, 41-51.  Mackie G. A. (1991) Specific endonucleolytic cleavage of the mRNA for ribosomal protein S20 of Escherichia coli requires the product of the ams gene in vivo and in vitro. J Bacteriol  173,  2488-2497.  Mackie G. A. (1992) Secondary structure of the mRNA for ribosomal protein S20. Implications for cleavage by ribonuclease E. J Biol Chem 267, 10541061.  Mackie G. A. (1998) Ribonuclease E is a 5'-end-dependent endonuclease. Nature 395, 720-723.  Mackie G. A. (2000) Stabilization of circular rpsT mRNA demonstrates the 5'end dependence of RNase E action in vivo. J Biol Chem 25072.  177  275,  25069-  Mackie G. A. and J. L. Genereaux. (1993) The role of RNA structure in determining RNase E-dependent cleavage sites in the mRNA for ribosomal protein S20 in vitro. J Moi Biol 234, 998-1012.  Margossian S. P., H. Li, H. P. Zassenhaus and R. A. Butow. (1996) The DExH box protein Suv3p is a component of a yeast mitochondrial 3'-to-5' exoribonuclease that suppresses group I intron toxicity. Cell 84, 199-209.  Masters M., M. D. Colloms, I. R. Oliver, L. He, E. J. Macnaughton and Y. Charters. (1993) The pcnB gene of Escherichia coli, which is required for ColE1 copy number maintenance, is dispensable. J Bacteriol 175, 44054413.  McDowall K. J. and S. N. Cohen. (1996) The N-terminal domain ofthe rne gene product has RNase E activity and is non-overlapping with the argininerich RNA-binding site. J Moi Biol 255, 349-355.  McDowall K. J., R. G. Hernandez, S. Lin-Chao and S. N. Cohen. (1993) The ams-1 and rne-3071 temperature-sensitive mutations in the ams gene are  in close proximity to each other and cause substitutions within a domain that resembles a product of the Escherichia coli mre locus. J Bacteriol 175, 4245-4249.  178  McDowall K. J., V. R. Kaberdin, S. W. Wu, S. N. Cohen and S. Lin-Chao. (1995) Site-specific RNase E cleavage of oligonucleotides and inhibition by stem-loops. Nature 374, 287-290.  McDowall K. J., S. Lin-Chao and S. N. Cohen. (1994) A+U content rather than a particular nucleotide order determines the specificity of RNase E cleavage. J Biol Chem 269, 10790-10796.  McLaren R. S., S. F. Newbury, G. S. Dance, H. C. Causton and C. F. Higgins. (1991) mRNA degradation by processive 3-5' exoribonucleases in vitro and the implications for prokaryotic mRNA decay in vivo. J Moi Biol  221,  81-95.  Miao X. (2002) Modular Domain Organization of RNase E and PNPase. In Biochemistry and Molecular Biology. University of British Columbia,  Vancouver.  Miczak A., V. R. Kaberdin, C. L. Wei and S. Lin-Chao. (1996) Proteins associated with RNase E in a multicomponent ribonucleolytic complex. Proc Natl Acad Sci USA93,  3865-3869.  Miczak A., R. A. Srivastava and D. Apirion. (1991) Location of the RNAprocessing enzymes RNase III, RNase E and RNase P in the Escherichia coli cell. Moi Microbiol 5, 1801 -1810.  179  Misra T. K. and D. Apirion. (1979) RNase E, an RNA processing enzyme from Escherichia coli. J Biol Chem 254, 11154-11159.  Mitchell P., E. Petfalski, A. Shevchenko, M. Mann and D. Tollervey. (1997) The exosome: a conserved eukaryotic RNA processing complex containing multiple 3'->5' exoribonucleases. Ce//91, 457-466.  Mohanty B. K. and S. R. Kushner. (2000) Polynucleotide phosphorylase functions both as a 3' ->• 5' exonuclease and a poly(A) polymerase in Escherichia coli. Proc Natl Acad Sci U S A 97, 11966-11971.  Mudd E. A. and C. F. Higgins. (1993) Escherichia coli endoribonuclease RNase E: autoregulation of expression and site-specific cleavage of mRNA. Moi Microbiol 9, 557-568.  Mullis K., F. Faloona, S. Scharf, R. Saiki, G. Horn and H. Erlich. (1986) Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harb Symp Quant Biol 51, 263-273.  Newbury S. F., N. H. Smith and C. F. Higgins. (1987) Differential mRNA stability controls relative gene expression within a polycistronic operon. Cell 1131-1143. 180  51,  Niyogi S. K. and A. K. Datta. (1975) A novel oligoribonuclease of Escherichia coli. I. Isolation and properties. J Biol Chem 250, 7307-7312.  Okada Y., M. Wachi, A. Hirata, K. Suzuki, K. Nagai and M. Matsuhashi. (1994) Cytoplasmic axial filaments in Escherichia coli cells: possible function in the mechanism of chromosome segregation and cell division. J Bacteriol 176, 917-922.  Ono M. and M. Kuwano. (1980) Chromosomal location of a gene for chemical longevity of messenger ribonculeic acid in a temperature-sensitive mutant of Escherichia coli. J Bacteriol 142, 325-326.  Ow M. C. and S. R. Kushner. (2002) Initiation of tRNA maturation by RNase E is essential for cell viability in E. coli. Genes Dev 16, 1102-1115.  Ow M. C , Q. Liu and S. R. Kushner. (2000) Analysis of mRNA decay and rRNA processing in Escherichia coli in the absence of RNase E-based degradosome assembly. Moi Microbiol 38, 854-866.  Pedersen K., A. V. Zavialov, M. Y. Pavlov, J. Elf, K. Gerdes and M. Ehrenberg. (2003) The bacterial toxin RelE displays codon-specific cleavage of mRNAs in the ribosomal A site. Ce//112, 131-140.  181  Portier C. (1975) Quaternary structure of polynucleotide phosphorylase from Escherichia coli: evidence of a complex between two types of polypeptide  chains. Eur J Biochem 55, 573-582.  Py B., H. Causton, E. A. Mudd and C. F. Higgins. (1994) A protein complex mediating mRNA degradation in Escherichia coli. Moi Microbiol  14,  717-  729.  Py B., C. F. Higgins, H. M. Krisch and A. J. Carpousis. (1996) A DEAD-box RNA helicase in the Escherichia coli RNA degradosome. Nature 381, 169-172.  Regnier P. and C. M. Arraiano. (2000) Degradation of mRNA in bacteria: emergence of ubiquitous features. Bioessays 22, 235-244.  Richardson, J. P. and J. Greenblatt . (1996). Control of RNA chain elongation and termination. In Neidhardt, F. C. (ed.) Escherichia coli and Salmonella. ASM Press. Washington. 822-848.  Reiner A. M. (1969) Isolation and mapping of polynucleotide phosphorylase mutants of Escherichia coli. J Bacteriol 97, 1431-1436.  Robert-Le Meur M. and C. Portier. (1992) E.coli polynucleotide phosphorylase expression is autoregulated through an RNase Ill-dependent mechanism. EMBO J 11, 2633-2641.  182  Ruth R., G. Zipor, V. Portnoy, V. Liveanu and G. Schuster. (2003) RNA polyadenylation and degradation in cyanobacteria are similar to the chloroplast but different from E. coli. J Biol Chem 21, 21.  Sambrook J., E. F. Fritsch and T. Maniatis. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory.  Schaferkordt J. and R. Wagner. (2001) Effects of base change mutations within an Escherichia coli ribosomal RNA leader region on rRNA maturation and ribosome formation. Nucleic Acids Res 29, 3394-3403.  Sim S., S. Kim and Y. Lee. (2001) Role of the sequence of the rne-dependent site in 3' processing of M1 RNA, the catalytic component of Escherichia coli RNase P. FEBS Lett 505, 291-295.  Srivastava A. K. and D. Schlessinger. (1991) Structure and organization of ribosomal DNA. Biochimie 73, 631-638.  Steege D. A. (2000) Emerging features of mRNA decay in bacteria. RNA 6, 1079-1090.  Steitz T. A. and J. A. Steitz. (1993) A general two-metal-ion mechanism for catalytic RNA. Proc Natl Acad Sci USA90, 6498-6502. 183  Symmons M. F., G. H. Jones and B. F. Luisi. (2000) A duplicated fold is the structural  basis for polynucleotide  phosphorylase catalytic  activity,  processivity, and regulation. Structure Fold Des 8, 1215-1226.  Taraseviciene L, G. R. Bjork and B. E. Uhlin. (1995) Evidence for an RNA binding region in the Escherichia coli processing endoribonuclease RNase E. J Biol Chem 270, 26391-26398.  Tock M. R., A. P. Walsh, G. Carroll and K. J. McDowall. (2000) The CafA protein required for the 5'-maturation of 16 S rRNA is a 5'-end-dependent ribonuclease that has context-dependent broad sequence specificity. J Biol Chem 275, 8726-8732.  Umitsuki G., M. Wachi, A. Takada, T. Hikichi and K. Nagai. (2001) Involvement of RNase G in in vivo mRNA metabolism in Escherichia coli. Genes Cells 6, 403-410.  Vanzo N. F., Y. S. Li, B. Py, E. Blum, C. F. Higgins, L. C. Raynal, H. M. Krisch and A. J. Carpousis. (1998) Ribonuclease E organizes the protein interactions in the Escherichia coli RNA degradosome. Genes Dev 12, 2770-2781.  184  Voet, D. and J. G. Voet (1990) Biochemistry. 2  na  ed. John Wiley and Sons.  Toronto. 356-359.  Wachi M., M. Doi, S. Tamaki, W. Park, S. Nakajima-lijima and M. Matsuhashi. (1987) Mutant isolation and molecular cloning of mre genes, which determine cell shape, sensitivity to mecillinam, and amount of penicillinbinding proteins in Escherichia coli. J Bacteriol 169, 4935-4940.  Wachi M., M. Doi, T. Ueda, M. Ueki, K. Tsuritani, K. Nagai and M. Matsuhashi. (1991) Sequence of the downstream flanking region of the shapedetermining genes mreBCD of Escherichia coli. Gene 106, 135-136.  Wachi M. and M. Matsuhashi. (1989) Negative control of cell division by mreB, a gene that functions in determining the rod shape of Escherichia coli cells. J Bacteriol 171, 3123-3127.  Wachi M., G. Umitsuki and K. Nagai. (1997) Functional relationship between Escherichia coli RNase E and the CafA protein. Moi Gen Genet 253, 515-  519.  Wachi M., G. Umitsuki, M. Shimizu, A. Takada and K. Nagai. (1999) Escherichia coli cafA gene encodes a novel RNase, designated as RNase G, involved  in processing of the 5' end of 16S rRNA. Biochem Biophys Res Commun  259, 483-488. 185  Walsh A. P., M. R. Tock, M. H. Mallen, V. R. Kaberdin, A. von Gabain and K. J. McDowall. (2001) Cleavage of poly(A) tails on the 3'-end of RNA by ribonuclease E of Escherichia coli. Nucleic Acids Res 29, 1864-1871.  Wang R. F., E. B. O'Hara, M. Aldea, C. I. Bargmann, H. Gromley and S. R. Kushner. (1998) Escherichia coli mrsC is an allele of hfIB, encoding a membrane-associated ATPase and protease that is required for mRNA decay. J Bacteriol 180, 1929-1938.  Yeaman S. J. and P. Cohen. (1975) The hormonal control of activity of skeletal muscle phosphorylase kinase. Phosphorylation of the enzyme at two sites in vivo in response to adrenalin. Eur J Biochem 51, 93-104.  Yu D. and M. P. Deutscher. (1995) Oligoribonuclease is distinct from the other known exoribonucleases of Escherichia coli. J Bacteriol 177, 4137-4139.  Zuo Y. and M. P. Deutscher. (2001) Exoribonuclease superfamilies: structural analysis and phylogenetic distribution. Nucleic Acids Res 29, 1017-1026.  186  8. APPENDIX  Appendix Figure 1. Column elution profiles for the purification of untagged  RNase G. The purification was performed as described in Materials and Methods, section 2.6.1. Each lane contains 7 pi of a collected fraction separated on a 10% SDS-polyacrylamide gel (section 2.7.1). Proteins were visualized by staining with Coomassie Brilliant Blue R-250. The concentration of NaCl in the running buffer increases from left to right. Panel a shows the collected fractions from the Heparin-agarose column. Panel b shows fractions from the anion exchange column. The lanes containing molecular weight standards are indicated on both gels. The size of each standard is indicated to the left of the panel.  187  188  Appendix Figure 2. Gel filtration size determination of RNase G samples.  Samples were eluted in 25 mM Tris-HCl pH 7.6, 320 mM NaCl on a calibrated Superdex S200 column as described in Materials and Methods, section 2.7.8. The volume of injection for each sample is marked by "t", and the volumes of the corresponding peaks for each sample are indicated above each peak. Volumes were determined using Unicorn software (Pharmacia).  189  190  191  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items