@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Chemistry, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Luoma, Gregory Allan"@en ; dcterms:issued "2010-02-24T18:26:10Z"@en, "1978"@en ; vivo:relatedDegree "Master of Science - MSc"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """None of the many previously proposed secondary structures for eucaryotic 5S RNA and 5.8S RNA is consistent with all known physical properties and suspected functions of these molecules. The present Raman results for S. cerevisiae 5S RNA and 5.8S RNA require a highly ordered secondary structure. A new, highly stable "cloverleaf" secondary structure not only accounts for the Raman data, but also accomodates previously established physical and functional features. Homologous cloverleaf structures can be adapted to other 5S RNA and 5.8S RNA species including E. coli 5S RNA, again accounting for properties and functions of these molecules. The model is therefore universal for small RNA species, and provides useful insight into the evolutionary aspects of these molecules."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/20834?expand=metadata"@en ; skos:note "LASER RAMAN EVIDENCE FOR NEW UNIVERSAL CLOVERLEAF STRUCTURES FOR 5.8S RNA AND 5S RNA by GREGORY ALLAN LUOMA B. Sc., University of B r i t i s h Columbia, 1976 A THESIS SUBMITTED. IN PARTIAL FULFILLMENT FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept th i s thes is as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1978 © Gregory A l lan Luoma, 1978. In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f ree l y ava i lab le for reference and study. I further agree that permission for extensive copying of th i s thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion of th is thes is for f inanc ia l gain sha l l not be allowed without my written permission. Department of Chemistry The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 i i ABSTRACT None of the many previously proposed secondary structures for eucaryotic 5S RNA and 5.8S RNA i s consistent with a l l known physical properties and suspected functions of these molecules. The present Raman results for S. cerevis iae 5S RNA and 5.8S RNA require a highly ordered secondary structure. A new, highly stable \" c lover lea f \" secondary structure not only accounts for the Raman data, but also accomodates previously established physical and functional features. Homologous c lover leaf structures can be adapted to other 5S RNA and 5.8S RNA species including E. c o l i 5S RNA,\"again accounting for proper-t i e s and functions of these molecules. The model i s therefore universal fo r small RNA species, and provides useful ins ight into the evolutionary aspects of these molecules. TIT TABLE OF CONTENTS Page Abstract i i Table of Contents i i i L i s t of Figures v i L i s t of Tables v i i i Acknowledgement ix Chapter I: Introduction 1 A. Protein Synthesis and 5S RNA and 5.8S RNA 2 1. Protein Synthesis Overview 2. Spec i f ic Involvement of tRNA and 5S RNA B. Structural Properties of RNA Molecules 7 1. Secondary Structural Properties 2. Tert iary Structural Features C. Techniques Used to Study RNA Structure 13 1. Chemical Methods 2. Physical Methods D. Present Knowledge of the Structure and Function of 5S RNA and 5.8S RNA 23 1. Procaryotic 5S RNA and Eucaryotic 5.8S RNA 2. Eucaryotic 5S RNA E. Present Work 27 Chapter II: Experimental Methods 128 A. Growth and Maintenance of S. cerevis iae Cultures 28 1. Maintenance of Cultures 2. Growth and Harvesting of Cultures B. I so lat ion and Pu r i f i c a t i on of RNA Species 29 1. Phenol Extraction 2. Ion Exchange Chromatography 3. Sephadex G-100 Chromatography on a Large Column I V Page •4-... Pu r i f i c a t i on \"o f 5S RNA 5. Pu r i f i c a t i on of 5.8S RNA 6. Desalting and Lyophi l i zat ion 7. Other Techniques Used in I so lat ion and Pu r i f i c a t i on C. Effect of Mg + + and Renaturation on Chromatography of 5S RNA and 5.8S RNA 42 D. Preparation of Samples For Raman Spectroscopy 42 1. Preparation of Normal 5S RNA, 5.8S RNA and tRNA Samples 2. Preparation of Low Mg + + -containing RNA Samples 3. Preparation of Samples of Varying pH E. Raman Parameters 44 Chapter I I I: Results 45 A. Physical Characterization of the Structure of 5S RNA and 5.8S RNA 45 1. Chromatographic Determination of Shape Asymmetry' 2. The Effect of Mg + + and Heat Renaturation on Chromatographic Behavior B. Origins of Raman Lines 48 C. The Raman Spectra of 5S RNA and 5.8S RNA at pH 7 352 1. Raman Spectra of Yeast 5S RNA 2. Comparison of 5.8S RNA and tRNA Raman Spectra 3. Comparison of High and Low Mg + + 5.8S RNA Raman Spectra 4. The Ef fect of pH on 5.8S RNA and tRNA Raman Spectra Chapter IV: Discussion 72 A. Secondary and Tert iary Structure of 5.8S RNA 72 1. Constraints of Structure From Raman and Other Physical Data Page 2. Incompatibi l i ty of Previous Structures With. Raman Data 3. New Cloverleaf Model F i t s Raman and Other Data 4. Conserved Structural Regions and Functional Implications of the Cloverleaf B. A5Similar Cloverleaf for Ec c o l i 5S RNA 80 1. Physical Constraints on Procaryotic 5S RNA 2. The Cloverleaf Model- for E. c o l i 5S RNA C. The Cloverleaf for Eucaryotic 5S RNA 82 1. Physical Constraints of 5S RNA Secondary Structure 2. Incompatibi l i ty of Previous Structures With Raman Data 3. The Cloverleaf Structure for Yeast 5S RNA D. Comments on Tert iary Structures of 5S RNA and 5.8S RNA 89 E. Comments on Evolution in 5S RNA and 5.8S RNA 90 Chapter V: Future Considerations 94 \" Glossary-3 98 ReRefenenees 100 vi LIST \" OF FIGURES Figure Page I- 1 A Schematic View of Protein Synthesis 4 I- 2 An Actual Structure of the Procaryotic Ribosome 6 I- 3 Php A Schematic View of the 3-D Shape of tRNA 8 I- •4 A Space - f i l l i n g Model of tRNA P h e 8 I- •5 UV Hypochromicity of RNA Species 10 II- 6 Normal Base-pairing in RNA Molecules 11 i - -7 Optical Properties of 5S RNA 21 I- -8 The Origin of 5S RNA and 5.8S RNA 25 II- -1 Growth Curve for S. cerevis iae 30 II- -2 An Outline of RNA Iso lat ion and Pu r i f i c a t i on 31 II- -3 A Typical DE-32 Elution P r o f i l e 33 II- -4 Large Scale Gel F i l t r a t i o n of RNA 35 II- -5 Pu r i f i c a t i on of 5S RNA 37 II- -6 P Pu r i f i c a t i on of 5.8S RNA 38 II- -7 Gel Electrophoresis Slabs of RNA 39 III- -1 Gross Shapes of tRNA, 5S RNA, and 5.8S RNA 46 III- -2 Renaturation of Mg + +-depleted 5S RNA and 5.8S RNA 47 III- -3 H20 Raman Spectra of tRNA and 5S RNA 54 III- -4 D20 Raman Spectra of tRNA and 5S RNA 55 III- -5 H20 Raman Spectra of tRNA, 5.8S RNA, and Low Mg + + 5.8S RNA 60 III -6 D20 Raman Spectra of tRNA and 5.8S RNA 61 III -7 Raman Spectra of tRNA and 5.8S RNA at Various pH Values 68 III -8 BaekBoi!iec0rderpnlnoitRNA\"sanat5c.8SnRNAnasRaAFunction of pH 69 III -9 The Effect of pH on Base Stacking in tRNA 70 v i i Figure Page III- •10 The Effect of pH on Base Stacking in 5.8S RNA 74 ' IV-•1 Previously Proposed Structures for 5.8S RNA 75 IV-•2 The CI overleaf Structure for Yeast 5.8S RNA 76 IV-•3 The CI overleaf Structure Adapted to Rat 5.8S RNA 77 IV-•4 The CI overleaf Structure Adapted to E. c o l i 5S RNA 83 IV--5 Previously Proposed Structures for 5S RNA 86 IV--6 The CI overleaf Structure for Yeast 5S RNA 87 IV-•7 The Cloverleaf Adapted to Other 5S RNA Species 91 vi i i LIST OF TABLES Table Page 111-1 Frequencies and Intens i t ies of RNA Raman Lines 50 111-2 Frequencies and Intens it ies of Raman Lines in Yeast tRNA and 5S RNA 56 IV-1 Properties of Various Models of 5.8S RNA 78 IV-2 Comparison of Properties of Models of Eucaryotic 5S RNA With Experimental Evidence 88 IX ACKNOWLEDGEMENT The l a s t two years in the UBC Chemistry Department have been a very enjoyable experience for me, espec ia l l y because of a l l the friends I have here. Although to name everyone I would l i k e to thank would be a d i f f i c u l t task, I would l i k e to express special appreciation to a number of people with whom I have worked most c lose ly . F i r s t , I would l i k e to thank Dr. A.G. Marshal l , whose f inanc ia l support and thoughtful suggestions both in the lab and over a draft at the \" P i t \" provided the groundwork and continued guidance for my research e f f o r t . Second, I would l i k e to thank the \"lunch-hour crew\" of Drs. D.C. Roe, R.E. Bruce and G. Webb, Mnd J.L. Smith and K.M. Ilee, and Mrs. J . Carruthers who are good fr iends to have, and who were sympathetic when things were not going we l l . F i n a l l y , I would also l i k e to thank a number of other people who thoughtful ly allowed me to use the i r equipment and helped me out with many problems. These people include Drs. A.W. Addison;,and D.G. Cil.ark, Prof. G. Tener and his group in Biochemistry, and the Univers ity of Alberta Biochemistry Department. CHAPTER I Introduction The events of protein synthesis are centermost in the function of the ce l l and the organism as a whole. They are, however, an extremely complicated set of reactions involving the coordinated e f fo r t of a large number of unique molecules, of which f i ve in procar-yotes and s ix in eucaryotes are RNA molecules (1-3). Central in under-standing how these molecules come together to form the protein synthetic apparatus i s an understanding of the i r conformations, since the i r shapes w i l l determine the i r posit ions in the ribosome, much l i k e f i t t i n g the correct piece of a jrigsaw puzzle. It i s our e f f o r t , and one which i s continuing in the laboratory, to s p e c i f i c a l l y i so la te the small RNA components (5S RNA, 5.8S RNA, tRNA) 1 of the system, and to determine the i r solut ion conformations using some of the powerful spectroscopic tools ava i lab le. It i s hoped that these structural studies w i l l provide useful information about how these indiv idual RNA species f i t into the larger network of proteins and RNA in the protein synthesis system. The introduct ion, then, w i l l consist of a b r ie f overview of what i s presently known about protein synthesis as a whole, followed by a more detai led discussion of the possible events in which tRNA, 5S RNA, and 5.8S RNA par t i c ipate . The present knowledge of the structure and function of 5S RNA and 5.8S RNA w i l l be considered, followed by a descr ipt ion of the experimental work to be presented. *5S RNA i s a small RNA with a sedimentation coe f f i c i en t of 5S and~120 nucleotides; 5.8S RNA i s a small RNA which sediments at 5.8S and has 158 nucleotides; tRNA i s transfer RNA which has \" 8 0 nucleotides. 2. A. Protein Synthesis and 5S RNA, 5.8S RNA 1. Protein Synthesis Overview The events of protein synthesis f a l l into three basic subgroups: chain i n i t i a t i o n ; chain elongation; and chain termination. These three d iv i s ions w i l l be described as they apply to procaryotic organisms, since thessystem in eucaryotes may be more complicated and i s less well understood (1-3). Fig. 1-1 summarizes the events of proteinssynthesis. (a) chain i n i t i a t i o n : The small ribosomal subunit (30S subunit) combines with three i n i t i a t i o n factors (IF 1, IF 2, IF 3). To th i s complex i s added the i n i t i a t o r tRNA (fMet-tRNA f ) 2 and GTP. F i na l l y mRNA is added to give the small subunit i n i t i a t i o n complex. The tota l i n i t i a t i o n complex i s made by attaching the large ribosomal subunit. This attachment involves the cleavage of GTP to GDP, and the release of the i n i t i a t i o n factors. At th i s point the 3 fMet-tRNA^. occupies the P-s i te and i s matched with the i n i t i a t i o n codon (AUG)*. fMet-tRNAf i s the spec i f i c i n i t i a t o r tRNA in procaryotes which i s made up of a tRNA type spec i f i c for binding the amino acid methionine which has been modified by formylation. 3 P-s i te i s the s i t e of attachment of the tRNA-bound peptide as chain elongation takes place. 4 .codon* i lh t se i s f theoS nue-liebitnide uiifithonithemmRNAawtinichcis spec i f i c fLorttherarndfnomacid to be attached to the peptide. (b) chain elongation: The now-formed ribosome has the fMet-tRNA 5 in the P-s i te and a blank space at the A-s i te with the codon for the next amino ac id. The next tRNA, which has been aminoacylated in the cytoplasm, becomes bound to EF-Tu»GTP (one of 3 elongation factors) with the concomitant cleavage of GTP. This AA-tRNA»EF-Tu«GTP complex becomes bound to the A- s i te at the ribosome by matching up the codon and anticodon. At the ribosome an enzyme (peptidyl transferase) transfers the ester linkage bond of the fMet-tRNA f from the 3'OH end of the tRNA^ to the amino group of the next amino acid to form the f i r s t peptide l inkage. The deaminoacylated tRNA i s released from the P - s i te , and EF-T breaks the tRNA°EF-T complex. The tRNA from the A - s i t e , with s u r the growing peptide attached at the 3'OH end, i s then sh i f ted to the P-s ite by the action of EF-G and the cleavage of GTP. As the tRNA moves the mRNA7 does also to expose a new three base codon at the A - s i t e . The elongation process i s c y c l i c , with the above sequence of events occurring repeatedly. This i s indicated in F ig. 1-1 by the arrows showing the recycl ing of tRNA, and elongation continues unt i l a terminator codon appears at the A - s i te . 5 A-s i te i s the s i t e of attachment of the incoming aminoacylated tRNA at the ribosome. ^AA-tRNA i s aminoacylated tRNA which i s accomplished by the enzyme aminoacyl-tRNA synthetase. ?mRNA i s messenger RNA which i s the code from the DNA specifying the protein which is to be made. Figure 1-1: A schematic representation of t ranscr ipt ion and t rans lat ion as i t i s presently envisioned ( l ) . IF= in it iat ion factor, EF=elongation fac to r , RF= release factor , a.a.=amino acid. 5. (c) chain termination: When one of the terminator codons (UAA, UAG, UGA) appears at the A - s i t e , the signal to terminate peptide elongation occurs. Termination involves the release of the polypeptide, the mRNA, and the breakage of the ribosome into i t s two subunits. In th i s process the terminator codons promote the binding of release factors (RF's) to the ribosome, with UAA or UAG promoting RF 1 binding and UAA or UGA promoting RF 2 binding. This binding i s further promoted by RF 3, while GTP l a b i l i z e s the breakdown of the tota l complex. 2. Spec i f ic tRNA and 5S RNA Involvements Although the above section gives the general sequence of events of protein synthesis, i t t e l l s nothing about the mechanisms involved. In f a c t , very l i t t l e i s known about the function of various components present (1-4). However, recent experiments in a number of laborator ies have given some clues about the mechanistic ro le that tRNA and 5S RNA play in protein synthesis (5-8). Also, many of the posit ions of the protein and RNA constituents have been determined by neutron d i f f r a c t i o n and summarized in a number of review a r t i c l e s (2-4). F ig. 1-2 is the proposed structure of the procaryotic ribosome, and the posit ion of 5S RNA i s indicated on i t . 5S RNA has been located at the A - s i t e , where AA-tRNA i s bound, and i s therefore implicated in AA-tRNA binding at the A - s i t e . Also, observed conformational changes Phe8 in tRNA (5-7) indicate that three events appear to be involved in Phe binding tRNA to the ribosome. Php ( i ) tRNA , which, when free in so lut ion, i s highly compact with Q i t s T£CG loop involved in base pair ing to the DHU loop (9,10), 8 Phe tRNA i s tRNA spec i f i c for the aminoacid phenylalanine. g T^CG and DHU are two loops of tRNA named for the i r invar iant bases. 6. 00 Figure 1-2: The proposed structure of the procaryote ribosome. From Brimacombe et a l . ( 2 ) . binds phenylalanine (Fig. 1-1). The binding causes a pa r t i a l disruption of the t e r t i a r y interact ion between the T^CG loop and DHU loop, as observed by laser l i g h t scatter ing (5). ( i i ) Phe-tRNA binds EF-T-u°GTP complex and arr ives at the ribosome. The anticodon and codon pair up, causing the T^CG loop to be exposed. This has been shown by equi l ibr ium d i a l y s i s and ol igonucleotide binding (6,7). ( i i i ) The Phe-tRNA then becomes bound to the ribosome by comple-mentary pair ing to 5SRNA, which i s located at the A - s i te . This binding i s accompanied by GTP cleavage, and one of the proteins most c lose ly associated with 5S RNA has th i s GTPase a c t i v i t y (8). (a)30S Subunit 7. The above tRNA binding mechanism has a low-frequency error rate since only aminoacylated tRNA w i l l bind EF-Tu<>GTP, and only the correct tRNA which matches the codon w i l l bind to 5S RNA at the ribosome. e Also, the conformation of the tRNA affects th i s s p e c i f i c i t y . Therefore, the present Raman studies were undertaken on 5S RNA and 5.8S RNA to see i f the eluc idat ion of structural properties could provide useful ins ight into the function of these molecules, as was the case for tRNA P h e . B. Structural Properties of RNA Molecules 1. Secondary Structural Properties The two most diagnostic features of secondary structure are base stacking and base pair ing interact ions. Not only are they the largest free energy contributors to he l i ca l s t a b i l i t y and order in the RNA structure, but they also undergo eas i l y detectable changes in opt ica l and magnetic properties. (a) Base stacking interact ions. A l l the bases present in RNA except uridine are known to undergo extensive stacking in s ingle stranded homopolymers; th i s stacking involves s i gn i f i cant s t a b i l i z a t i o n energy (e.g. poly A has A H ° . . =-13 kcal (11)), and thus provides s u f f i c i en t suacKing energy to order the ribophosphate backbone. This stacking i s greatest when the bases can get closest together, i . e . when the backbone of the RNA i s arranged h e l i c a l l y , with the bases ly ing f l a t on top of one another l i k e a stack of pennies. F ig. 1-3 gives a schematic represen-tation', of a he l i ca l piece of RNA while F ig. 1-4 shows how th i s stacking Phe appears in an actual double hel ix of tRNA . The four most important features of th i s s pa ce - f i l l i n g model are: ( i ) the base pairs are stacked and are staggered due to the he l i ca l twist of the molecule; ( i i ) the base pairs are phys ica l ly touching in the s pa ce - f i l l i n g model, 8. F ig. 1-4: A s pace - f i l l i n g model of the three-dimensional structure of tRNA P h e , as per Rich et a l . (10). 9. ind icat ing the proximity of pa i r s , and allowing overlap of the n-clouds of adjacent bases to s t a b i l i z e the he l i x ; ( i i i ) the s t a b i l i t y in some single stranded regions due to stacking i s s u f f i c i en t to maintain the hel ix in the absence of base pairs (e.g. the anticodon); ( iv) the regions where the base stacking occurs are largely constrained, and freedom of movement of the sugar phosphate backbone i s l im i ted . The base stacking interact ions described above can be studied by a number of techniques because of the unique properties that they supply to the physical structure of the molecule. The f i r s t of these i s hypochromicity, which means that the absorbance in the UV region (230 nm - 280 nm) decreases as more bases become stacked for a given number of bases. If one plots the change in absorbance of an RNA species from the native to heat-denatured form of the molecule against the wavelength from 230 nm to 280 nm one gets the standard denaturation spectrum for the RNA. Also, the denaturation spectra of poly rA-poly rU strands and poly rG-poly rC strands are very d i f f e ren t , and the tota l denaturation spectrum i s the sum of the contributions due to the unstacking of AU pairs and GC pairs (13,14). Therefore, i f one compares the RNA denaturation spectrum of an unknown RNA species to the denaturation spectra of various percentage combinations of GC and AU pa i r s , one can predict the percentage of GC pairs and AU pairs in the unknown RNA (see F ig. 1-5). The tota l area under the denaturation curve i s ind icat ive of the tota l number of base pairs present in the molecule. Also, as w i l l be discussed l a t e r , hypochromicity of base stacking has an ef fect on certa in Raman l i ne i n t e n s i t i e s , and on CD, ORD, and IR spectral charac te r i s t i c s . 10. 230 250 270 Wavelength (nm) A E x l Q 60%G-C J L 230 250 Wavelength (nm) 270 Figure 1-5: (a). A standard curve of hypochromicity verses wavelength for poly rA-poly rLI ( ) and poly rG-poly rC ( — ) . (b). A superposition of the actual hypochromicity of 5.8S RNA ( ) and the best f i t combination of the two curves in (a) (---). From Van et a l . (13). The second important property of base stacking i s the r i g i d i t y i t supplies to the double stranded backbone of the RNA in combination with the base pa i r ing. The ef fect of r i g i d i t y i s observed in the Raman spectrum of RNA molecules, and can be used as an ind i rect measure of the amount of he l i ca l structure and base pair ing present. (b) base pa i r ing. The base pair ing in RNA i s normally of two types: GC pairs and AL) pa i r s , with GU pairs also being allowed. F ig. 1-6 shows the pair ing schemes as well as the attachment of the bases to the sugar phosphate backbone. The order of strength of these H-bonded pairs i s GC (-2.4 kcal/mole), AU (-1.2 kcal/mole), and GU (0 kcal/mole) 11. (a) A-U Pair to3'end to 5 ' O H 1 to. 0 H Q / - O H (b)G-C Pair \\ N H UGAR N C, SUGAR' / V H N / \\ (c) G-U Pair I \\ ^ H N N °-J N > H - < N SUGAR c \"SUGAR / \\ Figure 1-6: A schematic representation of the possible base pairs in the secondary structure of RNA. Also indicated i s the attachment of bases to the ribose and the phosphodiester backbone. 12. at 25°C (15). From these values f o r the energy of the pa i r s , a' number of p o s s i b i l i t i e s become evident: ( i ) since GC pairs are stronger, the RNA with a higher GC content w i l l usually be more stable; ( i i ) the most stable structures w i l l be ones with the maximum number of pairs and the minimum number of looped out regions, bulges, or hairpin loops; ( i i i ) the regions r i ch in GC pairs w i l l melt at a higher temperature than regions r i ch in AU pa i r s . Therefore, by fol lowing hypochromicity as a function of temperature, the presence of GC r i ch regions can be determined, since these regions only become unstacked at higher temperatures. Also, by making use of the fact that GC pairs have a greater hypochromicity a t 278 nm than at 260 nm while AU pairs are the opposite, one can pos i t i ve l y i dent i f y regions as being GC r i ch (14). Base pair ing can also be followed by other techniques. These include Raman spectroscopy, infrared spectroscopy, and NMR-spectrloscopy. A l l of these techniques take advantage of the fact that the H-bonds have special propert ies, and w i l l be considered in deta i l in the next sect ion. 2. Tert iary Structural Features Tert iary structure i s also an important factor in RNA s t a b i l i t y and funct ion, but i s more d i f f i c u l t to determine because t e r t i a r y interact ions can occur without base pair ing or base stacking (see Fig. 1-4), and the types of H-bonding are hot always of the Watson-Phe Crick type. The t e r t i a r y structure of tRNA has been determined, and can be used as a guide to possible features of t e r t i a r y structure in other RNA species. The two most eas i l y u t i l i z e d features of t e r t i a r y structure are the low melting temperature, and the hypochromism that usually accom-panies the unstacking of t e r t i a r y base pairs . A lso, the hypochromism 13. of t e r t i a r y unfolding i s usually below 250 nrn^and i s therefore the l i k e l y contributor to the 5.8S RNA curve in Fig. 1-5 for the region below 250 nm. This explains the deviation from the standard curve. However, since the hypochromism i s not due to the unstacking of normal base pa i r s , spec i f i c t e r t i a r y features cannot be eas i l y determined. The same i s also true for plots of hypochromism verses temperature, making hypochromism of l imi ted use for t e r t i a r y determination. The most r e l i a b l e physical techniques for determining t e r t i a r y structure are X-ray crystallography and NMR, where spec i f i c interact ions can be determined. These techniques w i l l be described in the next sect ion. C. Techniques Used to Study RNA Structure A large number of physical and chemical techniques have been used to study RNA structure. Some of them have already been mentioned w i th -out j u s t i f y i n g the i r use. Therefore, th i s section w i l l be devoted to describing a number of the techniques, with emphasis being placed on the type, amount and r e l i a b i l i t y of information obtained from each of them. In most cases, the r e l i a b i l i t y w i l l be compared to resu lts Phe on tRNA , since i t s crystal structure i s known. 1. Chemical Methods Chemical methods are the most sens i t ive determinants of RNA secondary and t e r t i a r y structure. The most common of these methods are chemical modif icat ion, enzymic par t i a l d igest ion, and ol igonucleotide binding studies. The s p e c i f i c i t y , r e l i a b i l i t y and u t i l i t y of each of these techniques w i l l be considered. (a) chemical modif icat ion. This technique i s probably the most sens i t ive method of determining which bases are unpaired and exposed on the surface of the RNA molecule. Furthermore, chemical modif ication . 4 4 , causes l i t t l e disruption of structure, and can be used under conditions in which the molecule i s in the native form, or under d i f fe rent conditions (higher temperature) where t e r t i a r y structure has been a l tered. Thus one can get an overal l picture of which bases are involved in t e r t i a r y structure and in secondary structure. Also, reagents are ava i lab le which react s p e c i f i c a l l y with each of the four major r ibonucleic acid bases, allowing one to independently determine unpaired A-residues, G-residues, C-residues, and U-residues. The experimental technique involves the introduction of the modifying agent (monoperphthalic acid for adenine, glyoxal or kethoxal for guanine, carbodiimide for u r a c i l , methoxyamine for cytosine) into a solut ion of the RNA. After react ion, the modified RNA i s i so lated and the posit ion of the modifying agent i s determined by pa r t i a l digestion and analysis of the fragments by radioactive counting, i f a rad io labe l led modifying agent was used;(16,17,18). Phe The results obtained for chemical modif ication of tRNA are in excel lent agreement with the crysta l structure (19), which suggests the v a l i d i t y of th i s technique for determining RNA secondary and t e r t i a r y structure. (b) pa r t i a l enzymic hydrolys is. This technique has also been used to ident i f y s ingle stranded exposed regions in RNA, based on the fact that ribonucleases w i l l p re fe rent i a l l y break phosphodiester linkages in s ingle stranded regions before attacking double stranded regions. Also, as in chemical modif icat ion, ribonucleases are known which have s p e c i f i c i t y for a given base, so one can p re fe ren t i a l l y cleave at s ingle stranded G-residues (T^-RNase), G and A-residues (T 9-RNase), or C and U-residues (Pancreatic RNase). 15. The procedure for performing pa r t i a l enzymic hydrolyses requires that the enzyme operate under unfavorable condit ions, so that only the most susceptible regions w i l l be cleaved. The three most commonly control led conditions are temperature (0-4*0), time (5 min) and the enzyme concentration ( ser ia l d i l u t i o n ) . The digests are subjected to electrophoresis in two dimensions, the f i r s t a non-denaturing one, and the second a denaturing one. Spots are counted to determine the number of cleavage points, and each of the spots can be eluted and sequenced to determine the locations of cleavage points. Phe This technique gives reasonable resu l t s for tRNA when compared to the crysta l structure (20). Also, Holley et a l . (21) showed that Al a the f i r s t point of cleavage for tRNA is in the anticodon, which is Phe the most open region in the s t ruc tu ra l l y - re l a ted tRNA . The tech-nique i s l im i t ed , however, by the fact that the f i r s t cleavage(s) may af fect the structure enough that regions not exposed in the native form become susceptible to cleavage in the part ly cleaved molecule. Therefore, only the f i r s t cleavage s i tes are r e l i ab l e indicators of the accessible regions of the molecule. (c) ol igonucleotide binding. This technique i s based on the fact that unpaired sequences of bases should be avai lable to bind the i r complementary sequences, while those that are already base paired w i l l not. Therefore by introducing a series of ol igonucleotides covering the tota l RNA sequence, one should be able to get a map of which regions are accessible and s ingle stranded. The procedure involves introducing the ol igonucleotide to a solut ion of RNA and allowing th i s solut ion to equ i l ib rate for 3-6 days. The binding of the ol igonucleotide i s then measured by equi l ibr ium d i a l y s i s (22,23). 16. The results obtained from ol igonucleotide binding are d i f f i c u l t to Phe interpret and give unrel iable results for tRNA . Oligonucleotide experiments suggest that both the variable loop and the TKG loop are accessible (24), while the X-ray structure shows they are not (9,10). This i s l i k e l y a resu l t of a conformational change such as that which i s known' to occur when the tRNA anticodon binds i t s complement (6,7). Therefore the results <3if ol igonucleotide studies should be viewed with skepticism unless other data corroborate: the i r conclusions. 2. Physical Methods In general, physical methods are less sens i t ive than chemical methods to spec i f i c structural features such as i den t i f i c a t i on of unpaired and exposed bases; exceptions to th i s rule are X-ray c r y s t a l l o -graphy and NMR. The value of physical techniques i s to determine gross structural features such as tota l number of base pa i r s , number of AU and CG pa i r s , tota l shape, and types and degrees of base stacking. The information from these techniques, when combined with the spec i f i c data from chemical methods, provides a powerful tool for the e luc idat ion of secondary and (sometimes) t e r t i a r y structure of RNA. Some of the most widely used techniques w i l l be b r i e f l y out l ined. (a) X-ray crystallography and NMR. These are the two most powerful tools in determining conformations of biopolymers. X-ray Phe crystallography of tRNA has led to the determination of the tota l structure (9,10), and th i s structure has been backed up by NMR results (25). However, crystallography of 5S RNA and 5.8S RNA has two major drawbacks: high qua l i ty crysta l s are not eas i l y obtainable, and the crysta l structure may not be the same as the solut ion structure. At present, 5S RNA and 5.8S RNA have not been successful ly c r y s t a l l i z e d , and l i k e l y w i l l hot be in the near future, since ten years of attempts 17. on the more simple tRNA molecules has led to crysta l s of high qua l i ty for only two species (26,27). NMR, on the other hand does not require c r y s t a l l i z a t i o n , and i s therefore much easier to perform experimentally. However, i t can have the drawback of too many overlapped peaks, making analysis impossible. There are, however, some useful regions in the proton spectrum, espec ia l ly the methyl proton region of the special bases (28), and the low f i e l d proton region (29). In RNA species, the most useful region i s the low f i e l d area, (-10 to -15 ppm) since th i s region contains the H-bonded protons. The integrated in tens i ty gives an ind icat ion of the number of H-bonds in the RNA molecule, while the posit ions of the peaks are sens i t ive to Phe the environments of the protons. For tRNA a l l the H-bonds predicted by the crysta l studies have been i den t i f i ed and assigned (25,29), and the spectrum has been simulated using chemical sh i f t s calculated from the base pair ing pattern.(25). In 5S RNA, the large number of H-bonds P h6 (A/40 verses 28 in tRNA ) have made both integration and assignment of peaks d i f f i c u l t (30,31), with the resu l t that l i t t l e useful structural information concerning 5S RNA has come from NMR spectroscopy. However, since the resolut ion for low f i e l d proton spectra i s d i r e c t l y proportional to magnet strength, new superconducting magnets may improve the resolut ion s u f f i c i e n t l y to make th i s a useful technique in the future. (b) opt ica l methods. A number of opt ica l methods including u l t r a v i o l e t absorption (UV), c i r c u l a r dichroism (CD), opt ica l rotary dispersion (ORD), infrared spectroscopy (IR), and Raman spectroscopy have been used to study RNA structure. The f i r s t four w i l l be 18. considered b r i e f l y while Raman spectroscopy, because of i t s pertinence to the present study, w i l l be considered in greater d e t a i l . UV, CD, ORD, and IR are a l l useful in determining tota l amounts of base pa i r ing , while some are also useful in determining the percentage GC contentoof the he l i ca l regions. A l l such spectral ca lculat ions are based on a s ingle equation: aM-2.ft ^(XH 2 ^ A ai(\\) + 2 i }j.2p.. A a > ) 1 2 3 where a ( \\ ) i s any spectroscopic var iable at an arb i t rary wavelength. Term 1 i s theocontribution to the spectroscopic var iable of each of the four bases i f they were t o t a l l y independent of one another (f. i s the f ract ion of each base i and cq(\\)is the molar residue value of the opt ica l parameter for each base i . Term 2 i s the contr ibution of the base pairs to the spectroscopic var iable, (h^ i s the mole f ract ion of base i involved in pairs and Ac^fxlis the contr ibution of each type of base pair to the opt ica l parameter). Since only AU and GC pairs are considered, h^=h^ and hg=h^; similarly&*A=Aa: u andAaa=Aac. Therefore, the tota l he l i ca l content H = 2<' hA,U + hG,C>V . Term.3 i s the contr ibution of the unpaired bases in the strand to the spectroscopic variable (aag=ay-(at+aj)/2 where ay i s the mean residue parameter for a dinucleoside composed of bases i and j while p.. i s the mole f ract ion of pairwise interact ions between adjacent bases i and j in non-helical regions). The above treatment provides a mechanism for determining he l i ca l content of RNA species by simulation of i t s spectrum. The requirements are that the spectra of the mononucleotides, paired poly rA-poly rU and poly rG-poly rC, and unpaired homopolymers are known. Then, by 19. taking non-linear combinations of these curves, the RNA spectrum i s simulated and the he l i ca l content determined. U l t r av i o l e t absorption i s the most common method of fol lowing denaturation of RNA. This i s because i t i s experimentally simple and requires very l i t t l e sample. As mentioned e a r l i e r the most useful results from th i s technique are the tota l number of base pairs and the percentage of GC and AU pairs . However, the accuracy of th i s technique i s l imi ted because of low s e n s i t i v i t y . This i s exemplified by the fact that a 1.5% increase in area of the denaturation spectrum represents a 10% decrease in GC content of the he l i ca l regions (14). A l so, the ef fect of terminal pairs and next neighbor e f f ec t s , though ignored in the standards used for comparison, can have substantial ef fects on hypochromicity in a RNA molecule. However, UV hypochromism i s a useful technique for sett ing the upper and lower l im i t s of pa i r ing. Optical rotary dispersion (ORD) or c i r cu l a r dichroism (CD) are both measurements of the same property; i . e . the ef fects on the l e f t and r ight c i r c u l a r l y polarized .components of the rad iat ion source are independent and may d i f f e r substant ia l l y . CD measures the difference in the absorption of the two components, while ORD measures the difference in the re f rac t i ve index of the two components (32). In E. c o l i 5S RNA, CD and ORD spectra give se l f -cons i s tent results for the estimation of tota l he l i ca l content (33) (Fig. 1-7 b,c). However, these two techniques are r e l a t i v e l y insens i t ive to differences between AU and GC pa i r s , and the values for these two features of RNA structure are subject to considerable error. Infrared spectra provide information on both base pair ing (1700-1400 cm - 1 ) and secondary structure of the backbone (1300-1000 cm - 1 ) (33,34). The carbonyl region i s most usefu l , with band s p l i t t i n g 20. and sh i f t s in the 1500-1700 c m - 1 region ind icat i ve of GC pairs and AU pairs (33,34) (Fig. I-7d). The f i t t i n g of standard spectra to unknown spectra for IR i s much more d i f f i c u l t than for UV, CD, or ORD. Therefore, the accuracy of IR determinations are diminished. Also, sample handling techniques are d i f f i c u l t since water cannot be used as a solvent due to i t s strong IR absorbance. (c) Raman spectroscopy. Raman spectroscopy has been used extensively as a probe of nucleic acid structure (35-44). It i s espec ia l ly sens i t ive to secondary and t e r t i a r y structure, displaying intens i ty changes in Raman l ines for both stacking of bases and disrup-t ion of \"order\" in the phosphodiester backbone. Fortunately, the Raman spectrum of RNA samples contains a POg\" v ibrat ion which i s insens i t ive to almost any treatment (40) and can be used as an internal ca l i b ra t i on for determining re l a t i ve i n ten s i t i e s . ( i ) pr inc ip les of the Raman e f fec t . When l i gh t passes through a transparent substance a certa in small amount of radiat ion is scattered at frequencies widely d i f fe rent from the incident frequency; th i s i s Raman scatter ing. The difference in frequency of incident and scattered radiat ion suggests an exchange of energy between the molecule and the photon (AE=hA.v), and quantum theory st ipulates that th i s energy can only be of d iscrete values which are exactly equal to differences in energy of v ib ra t iona l - ro ta t iona l leve l s unique to the molecule under study. The Raman ef fect can be considered as a change in the p o l a r i -z a b i l i t y of the molecule as i t undergoes internal motion. This rule for Raman a c t i v i t y i s in contrast to the complementary infrared ru l e , where molecular motion must produce a change in the e l e c t r i c dipole of the molecule (45). 21. O P T I C A L P R O P E R T I E S O F 5 S R N A (a)UV (WORD \" i r m r E * 1 0 [mlxlO4- \\ \\ 240 260 280 X nm -l 1 1 1 I 1 I L 240 260 280 300 X nm (c)CD (d)IR EJKIO > 1 1 I I 1 1 1 \\ - j / V / / \\ . ' i / / \\ 1 / / \\ / / • ' / 1 1 1 \\ \\ •' / / / \\ \\ -jZl*-^ y / 1 1 1 < 1 f 10 -i 1 1 1 1 1 r 240 260 280 300 X nm 1600 1650 1700 X _ 1 c m - 1 Figure 1-7: Graphs of the opt ica l properties of E. c o l i 5S RNA to show i t s re lat ionship to the properties of standards, (a) UV spectra, (b) ORD spectra, (c) CD spectra, (d) IR spectra. The spectra of E. co l i 5S RNA ( ), a mixture of mono-nucleotides ( ; : \" ' ) » and completely paired systems (—•—•—) are given. From Richards et a l . (33). .22. • ( i i ) advantages and disadvantages of Raman spectroscopy. Raman spectroscopy has a number of advantages in studying RNA molecules. The most important of these i s that Hrfi has r e l a t i v e l y weak Raman bands while i t s IR bands are numerous and strong. Thus, Raman spectro-scopy can be used on aqueous samples, making handling easier and making the obtained results more r e l i a b l e in terms of the native RNA species. The second advantage, and one that i s espec ia l ly useful for RNA structure determination, i s that Raman peak i n tens i t i e s exh ib i t hypochromism or hyperchromism with stacking e f fec t s , in contrast to IR which i s r e l a t i v e l y insens i t ive to stacking. The heur i s t i c explanation for the hypochromism i s the so-cal led pre-resonance e f fec t . RNA molecules exh ib i t strong hypochromism in the 230-280 nm absorption region. The laser exc i tat ion wavelength of argon (514.8 nm) i s thought to be near enough to the absorption peak to mimic the resonance Raman e f fec t . Thus, the Raman peaks not only increase in in tens i ty in general, but assume the properties of the absorption peak. As w i l l be demonstrated, the hypochromic nature of the Raman peaks i s invaluable in determining comparative amounts of spec i f i c base stacking,as well as changes in stacking patterns due to structural perturbations. The th i rd major advantage of Raman spectroscopy i s the r e l a t i v e l y small number of peaks and the fact that many are well resolved. Many of the l ines are due to s ingle base types (e.g. the 670 c m - 1 l i ne due to G-bases) which can be monitored in terms of the stacking of a s ingle base species. Also, in the carbonyl region, unl ike IR, band in tens i t i e s are dominated by carbonyl stretches of U-residues, with the resu l t that differences in th i s region are mainly due to differences in the number of AU or GU pa i r s , and not a l l base pairs as 23. in IR (34). F i n a l l y , Raman spectra of RNA species have a spec i f i c l i ne (814 cm - 1 ) whose intens i ty i s d i r e c t l y related to the. order of the backbone and can be used to predict amounts of h e l i c i t y in RNA samples. The disadvantages of .'Raman are mostly related to the sample and the handling techniques, since thepphotomultiplier tube i s detecting small amounts of rad iat ion. In the case of RNA, the major concern is that the sample is op t i c a l l y c lear to minimize Tyndall (Rayleigh) scatter ing. This i s eas i l y accomplished by centr i fugat ion of samples. The other i n te r fe r ing type of radiat ion is fluorescence. This can be avoided by using only high qua l i ty chemicals in the i so l a t i on and pu r i f i c a t i on of the RNA, or by adding a fluorescence quenching agent (e.g. KI) to the sample. Therefore, Raman spectroscopy has the advantage over other opt ica l techniques of being sens i t ive to environments of spec i f i c base types, as well as the overal l features of the molecule. The resu l t i s that Raman spectroscopy gives more usefulestructural information about RNA molecules than does any other opt ica l technique. Combined with other techniques, i t provides the key information for determining the secondary structure of RNA molecules. D. Present Knowledge on the Structure and Function of 5S RNA and 5.8S RNA 1. Procaryotic 5S RNA and Eucaryotic 5.8S RNA As mentioned e a r l i e r , 5S RNA i s one of the small RNA molecules involved in procaryotic protein synthesis. In eucaryotes both 5S RNA and 5.8S RNA have been found associated with the ribosome (46). When the present studies were begun, the structures of these three molecules had not been determined and many models had been proposed (30,31, 47-56). 24. Also, t he i r three-dimensional crysta l structures have not been reported, and conclusions based' on l ow- f i e ld hydrogen-bonded proton nuclear magnetic resonance.spectra have been severely l imi ted by poor resolut ion (30,31). Therefore, data from the two most r e l i a b l e methods for determining the correctness of the proposed structures are not yet ava i lab le . Less d i rect structural information from other physical techniques has been reviewed (48,49), and more recent resu lts are now discussed. Pene et a l . (46) demonstrated that 5.8S RNA i s strongly hydrogen bonded to the 28S RNA of the large ribosomal subunits of eucaryotes, and 5.8S RNA i s only released from the ribosome under denaturing conditions. More recent ly, Pace et a l . (57) showed that th i s i n t e r -action i s between the 3'-end of the 5.8S RNA and a complementary segment of the 28S RNA, and i s s t ab i l i zed by the presence of a GC-rich loop near the 3'-end of the 5.8S RNA. Moreover, a l l 5.8S RNA nucleotide;sequences determined to date contain a sequence of bases which i s complementary to the T^CG loop of non - i n i t i a to r tRNA (58). These facts suggest that 5.8S RNA i s bound to the ribosome and, in turn, binds non - in i t i a to r tRNA at the ribosome during t ransc r ip t ion . In add i t ion, the fol lowing facts connect the structure and function of eucaryotic 5.8S RNA to procaryotic 5S RNA, but not eucaryotic 5S RNA. F i r s t , yeast 5.8S RNA can bind the same ribosomal proteins (EL-18 and EL-25) that E. c o l i 5S RNA binds most strongly, and th i s 5.8S RNA E. c o l i protein complex exhib its GTPase and ATPase a c t i v i t i e s s imi la r to those for the homologous 5S RNA protein complex (59). Second, a l l procaryotic 5S RNA molecules contain the complement of the TSkCG loop in the same sequence posit ion as 5.8S RNA (58), and 25. (a) procaryotic 5' 3 3 0 1 3 • 16S RNA tRNA 23S RNA 5S RNA (b) eucaryotic 5' I I • I 18S RNA 5 £ S RNA 28S RNA Figure 1-8: A schematic representation of the or ig in of 5S RNA and 5.8S RNA. th i s CGAAC region can bind TSI>CG or UUCG (49). Th i rd, both the procar-yo t i c 5S RNA sequence and the eucaryotic 5.8S RNA sequence are contained in the large ribosomal RNA primary t ranscr ipt ion units (Fig. 1-8) (60,61). Therefore, since eucaryotic 5.8S RNA and procaryotic 5S RNA appear to have s imi la r or ig ins and functions, the i r secondary structures should be s im i l a r . 2. Eucaryotic 5S RNA Eucaryotic 5S RNA i s considered separately, large ly because i t has few s i m i l a r i t i e s to eucaryotic 5.8S RNA or procaryotic 5S RNA. Along with 5.8S RNA, 5S RNA i s a structural component of the large ribosomal subunit of eucaryotic c e l l s (46). Unlike procaryotic 5S RNAsand eucaryotic 5.8S RNA, eucaryotic 5S RNA appears to be a primary trans-c r i p t i on product (undergoes no processing,by cleavage of a larger 3' I o 5SRNA 26. precursor molecule), since a 5 ' -terminal triphosphate i s present and no precursor molecules have been i so lated (49). Furthermore, eucaryotic 5S RNA has a base sequence which i s d i s s im i l a r to both procaryotic 5S RNA and eucaryotic 5.8S*RNA<(58): the constant GAAC region near pos it ion 45 i s replaced by CYGAU, which i s complementary to i n i t i a t o r tRNA but not to other tRNA's (49). Eucaryotic 5S RNA also displays fewer sequence homologies between d i f fe rent species than does procaryotic 5S RNA or (to a greater extent) eucaryotic 5.8S RNA (58). Although the above facts suggest major differences between eucaryotic and procaryotic 5S RNA, a l l ex i s t ing physical data suggest s imi la r secondary structures, with possible t e r t i a r y d i f ferences. For example, low angle X-ray scatter ing resu lts indicate that yeast and E. c o l i 5S RNA each have about the same axia l r a t i o (5:1) and radius of gyration,(34.5±1.5 A) (62). Moreover, thermal melting studies on E. c o l i and sea urbfoiiin.indicate s imi la r amounts of base pair ing and 62-64% he l i ca l content for e i ther molecule (63). However, the same study also shows that sea urchin 5S RNA i s more highly ordered and asymmetrically fo lded, and has a few GC-pairs involved only in t e r t i a r y structure. Enzymic pa r t i a l hydrolyses further demonstrate the s i m i l a r i t y in structure of procaryotic and eucaryotic 5S RNA, since a l l sequences are cleavage accessible near pos it ion 40, although eucaryotic 5S RNA i s also accessible near pos it ion 90 (64-66). Therefore, although the secondary structures of procaryotic 5S RNA and eucaryotic 5S RNA are l i k e l y very s im i l a r , the i r t e r t i a r y structures may d i f f e r . The presence in eucaryotic 5S RNA of the complementary sequence to i n i t i a t o r tRNA suggests that eucaryotic 5S ,RNA may function in the spec i f i c binding of i n i t i a t o r tRNA to the ribosome. For procaryotic 27. systems no corresponding spec i f i c molecule has been found. However, Dahlberg et a l . have very recently shown that the large 23S RNA molecule has a sequence complementary to part of i n i t i a t o r tRNA from procar-yotes, which can bind i n i t i a t o r tRNA (67). Also, procaryotic i n i t i a t o r tRNA s t i l l c con ta i n s the GTtyC region which i s complementary to the GAAC region in procaryotic 5S RNA (67). Therefore, the 23S RNA may be involved in d ist inguishing i n i t i a t o r tRNA from other tRNA species in procaryotes, while eucaryotes have 5S RNA to perform th i s function. E. Present Work In th i s work, Raman spectra are reported for Saccharomyces cerevis iae (yeast) 5.8S RNA and tRNA in the presence or absence of Mg + + ions in H2C1, and in the presence of Mg + + in D2O. Raman spectra are also reported for 5.8S RNA and tRNA as a function of pH. For yeast 5S RNA, the spectra in the presence of Mg + + in H2O or D2O are given. The results of these spectra show that none of the several previously proposed structures for e i ther 5.8S RNA or 5S RNA i s correct. Further, when combined with data from other chemical and physical techniques, the resu lts indicate a new proposed \" c lover lea f \" secondary structure for both 5.8S RNA and 5S RNA from yeast, with some interest ing t e r t i a r y properties and functional impl icat ions. The new structure i s adaptable not only to 5.8S RNA and 5S RNA from other eucaryotic species, but also to E. c o l i 5S RNA. The structure for E. c o l i 5S RNA i s consistent with almost a l l pre-determined structura l propert ies, as well as the recently obtained Raman spectrum (44). 28. CHAPTER II Experimental Methods A. Growth and Maintenance of S. cerevis iae Cultures 1. Maintenance of Cultures S. cerevis iae c e l l s were obtained as a yeast peptone agar s lant consist ing of 20 gm/1 peptone, 10 gm/1 yeast extract , 20 gm/1 glucose, and 40 mg/1 adenine in a 2% solut ion which i s heated to 95°C to dissolve the agar and then autoclaved to s t e r i l i t y . The culture i s stored at 0-5°C as a s lant , and i s maintained by transfer of old cultures to new slants at interva ls of 2 months. The new slants are allowed to grow for 2-3 days at room temperature before being stored at 0-5*C. The above procedure allows c e l l s to continue in the normal growing state while minimizing the amount of genetic mutation which may occur in rapidly growing c e l l s . At interva l s of 1-2 years the culture i s plated and s ingle i so lated colonies (ar is ing from single c e l l s ) are subsequently transferred to new s lants. The l a t t e r procedure prevents contamination of pure cultures by other organisms as well as by genet ica l ly d i f fe rent strains of S. cerevis iae. 2. Growth and Harvesting of Cultures The medium used for the ce l l growth was yeast peptone broth, ident ica l in composition to the medium used for culture maintenance except for the absence of agar. The tota l volume of each culture was 320 l i t e r s , and the medium was prepared and a_utoclaved withincthe fermentor chamber through the use of steam jackets. The inoculum for the large culture was prepared by successive scale-up of the or ig ina l s lant. The f i na l inoculum was 6.7% or 20 l i t e r s , which was grown overnight in two 10 l i t e r inoculat ion tanks and pumped d i r e c t l y through s t e r i l e connections to the main fermentor, thereby preventing 29. contamination during transfer. The inoculum was de l iberate ly large so as to outcompete any contaminating organisms and reduce the incubation time. Growth was carr ied out at 30°C with a s t i r r i n g rate of 250 rpm and a high aeration rate. These factors appear important in maintaining good growth on the large scale. The pH of the incubation mixture was monitored automatically as a further precaution against contamination (pH=5.8), Growth was monitored by measuring the tu rb id i t y at 550 nm on a Spectronic 20 spectrophotometer at 30 min interva l s on a 1/5 d i l u t i on to avoid high absorption f igures. A typ ica l growth curve i s shown in F ig. 11-1. As can be seen in the f i gure , the c e l l s were harvested during mid-log phase since th i s i s the area of the curve where maximum growth i s taking place, and thus corresponds to the region where the maximum proportion of RNA molecules are ac t i ve l y engaged in protein synthesis. The result i s that the i so lated RNA w i l l most l i k e l y be in the native (active) conformation, an important consideration when conducting structural studies. The c e l l s were harvested using a Sharpies continuous flow centrifuge spinning at 20,000 rpm. The ef f luent from the centrifuge wasijmonitored op t i c a l l y to insure that the yeast c e l l s were being sedimented with the centrifuge. The centrifuged c e l l s were scraped out and stored as one pound blocks at -20°C. The tota l y i e l d from a s ingle fermentation was four pounds. In a l l three runs were completed to give a tota l of twelve pounds of c e l l s . B. I so lat ion and Pu r i f i c a t i on of RNA Species (68,69) 1. Phenol extract ion The yeast c e l l s were extracted in one pound portions using equal volumes of deionized water and water-saturated phenol (3 l i t e r s tota l Figure 11 — 1: A typ ica l growth curve for the large culture of S. cerevis iae c e l l s . Growth was monitored as t u rb id i t y at 550 nm on a 1/5 d i l u t i on of the actual cu l ture. 31. Extraction and Purification crude 5S-RNA (so mg.) (4) pure 5S-RNA (so mg.) Yeast cells (454 gm.) RNA, polysaccharide some protein (2) tRNA, 5S-RNA, rRNA 5.8 S RNA breakdown products (o.80 gm.) (3) crude 5.8S RNA (70 mg.) (5) pure5.8S RNA (so mg.) pure mixed tRNA (600,mg) crude tRNAP h e (20 mg.) pure tRNA p h < (8 mg.) Figure 11-2: A summary of the i so la t i on showing the steps involved in pur i fy ing tRNA, 5S RNA, and 5.8S RNA. volume) with s t i r r i n g for 30 min at room temperature. The whit ish mixture was then centrifuged at 10,000 g for 15 min and the aqueous layer was withdrawn from the phenol and c e l l debris. The phenol layer was again extracted with 1.5 l i t e r s of water and centrifuged as before. The aqueous layers were then combined and MgClg was addedoto make the solut ion 0.05 M in Mg + + . F i na l l y the crude RNA was prec ip i tated with 2.5 volumes of 95% ethanol pre-cooled to -20°C. The phenol extraction serves three purposes: ( i ) i t i s a mild extract ion procedure which w i l l not a l t e r or damage RNA by chemical reaction or physical damage; ( i i ) phenol extracts some proteins and l i p i d s from the c e l l membrane but not the nuclear membrane, thereby separating the small RNA from contaminating DNA and s impl i fy ing p u r i f i c a t i o n ; and ( i i i ) a l l ribonucleases which may damage the extracted RNA are contained in the phenol layer rather than the aqueous layer. Thus, in one step, the RNA can be extracted in a stable and r e l a t i v e l y pure form. 2. Ion Exchange Chromatography The prec ip i tated RNA was centrifuged and the ethanol was poured of f . The pe l l e t was suspended in 50 ml of 1.2 M NaCl and again centrifuged at 10,000 g for 10 min. The extract was withdrawn and the pe l l e t was resuspended in a further 50 ml of 1.2 M NaCl and was extracted at 65°C for 10 min before centr i fugat ion. This step further eliminated insoluble non-RNA mater ia l , as well as most of the 18S RNA and 28S RNA which are insoluble at high sa l t concentrations. The extracts were combined and d i luted to 0.3 M NaCl. An ion exchange column was prepared by packing 200 gm of precycled 33. Figure 11-3: A typ ica l e lut ion p r o f i l e from DE-32 ion exchange chromatogra The column was loaded and washed with 0.3M NaCl, then with 1.0 M NaCl (arrow) to elute the RNA. 34. DE-32 ion exchange re s i n , equi l ibrated in 1.2 M NaCI, into a 5X60 cm column. The res in was then washed with 3 l i t e r s of 1.2 M NaCI solut ion to remove any A2gQ-absorbing\" material from the res in . F i n a l l y , the res in was equi l ibrated with 6 l i t e r s of 0.3 M NaCI to prepare i t for loading with the RNA so lut ion. The RNA was applied to the column and was washed with 0.3 M NaCI unt i l the % transmittance of the eluent (as monitored automatically with a LKB 8300 Uvicord II) i s greater than 80%. This l a t t e r procedure pur i f i e s the RNA by washing o f f prote in, phenol and mononucleotides (ATP, GTP) which were present a f te r extract ion. The RNA was then eluted with 1.0 M NaCI and prec ip i tated with 0.05 M MgCl 2 and 2.5 volumes of ethanol at -20°C overnight. The DE-32 ion-exchange chromatography, therefore, achieves the pu r i f i ca t i on of low molecular weight RNA species from other contaminants resu l t ing from the-extract ion procedure. A sample of the DE-32 chromatography i s shown in F ig. 11-3. 3. Sephadex G-100 Chromatography on a Large Gel F i l t r a t i o n Column A large Sephadex G-100 gel f i l t r a t i o n column (10x150.cm) was prepared by pouring 500 gm of pre-equi1ibrated Sephadex G-100 into a p lex ig las column and equ i l i b ra t ing with 1.2 M NaCI (10 l i t e r s ) . The prec ip i tated RNA from the ion exchange chromatography was c e n t r i -fuged and the ethanolwwas poured o f f . The pe l l e t containing about 1 gm of RNA was dissolved in a minimum of 1.2 M NaCI (about 40 ml), and was applied to the top of the column. The RNA was eluted using 1.2 M NaCI, 20 ml f ract ions were co l l ec ted , and' the % transmittance at 260 nm was monitored automatically as before to give the typ ica l e lut ion p r o f i l e in F ig. 11-4. The column was standardised using a mixture of blue dextran 2000, hemoglobin, myoglobin, and K,Fe(CN) K, 35. Mixed RNA elution from a 10-liter Sephadex G-100 column -1 —*r2-20 r 40 a n s. (260nm) 60 80 h 100 100 200 fraction 300 Figure II-4: A typ ica l Sephadex G-100 e lut ion p r o f i l e from the large gel f i l t r a t i o n column. The e lut ion of a mixture of standards (---) and the RNA ( ). Peak 1 i s predominantly -rRNA, peak 2 crude 5.8S RNA, peak 3 Crude 5S RNA, peak 4 tRNA. 36. and th i s mixture is shown overlapped with the RNA p r o f i l e in F ig. I1-4. The RNA p r o f i l e was divided into four f rac t ions : Fraction 1 contained high molecular weight RNA and was discarded; f ract ion 2 contained part ly pur i f ied 5.8S RNA which was pooled and prec ip i tated with Mg + + and ethanol; f ract ion 3 contained predominantly pure 5S RNA which was pooled and prec ip i tated for further p u r i f i c a t i o n ; and f ract ion 4 contained pure tRNA which was also pooled and prec ip i tated. Fractions 2 and 3 were then subjected to further pu r i f i c a t i on as indicated schematically in F ig. 11-2 and described below. 4. Pu r i f i c a t i on of 5S RNA The pooled prec ip i tated f ract ion 3 of F ig. 11-4 was centrifuged andithe pe l l e t containing about 80 mg of RNA was suspended in a minimal amount (about 2 ml) of 1.2 M NaCI. A long Sephadex G-75 gel f i l t r a t i o n column (2*190 cm) was prepared and equi l ibrated with 1.2 M NaCI. The RNA solut ion was applied to the column and 8 ml f ract ions were co l lected to give the p r o f i l e contained in F ig. 11-5. This same solution also gave three corresponding peaks when applied to gel electrophoresis slabs prepared as per Rubin (54) (Fig. I I-6a). These three fract ions correspond to pure 5.8S RNA, pure 5S RNA, and pure tRNA when compared to gels prepared by Rubin. As a check, some of the pooled region 2 was applied to a gel electrophesis s lab, and i t produced a s ingle peak corresponding to highly pur i f ied 5S RNA (Fig. II-6b). The three regions were separated, pooled, and p r e c i -pitated as before. The y i e l d of pure 5S RNA from th i s procedure i s about 50 mg from each pound of c e l l s . 5. Pu r i f i c a t i on of 5.8S RNA Pu r i f i c a t i on of 5.8S RNA was also accomplished by gel f i l t r a t i o n 37. G-75 chromatography of impure 5S RNA NM 2 *e-3\"H 30 40 50 fraction Figure 11-5: Pu r i f i c a t i on of 5S RNA on a long Sephadex G-75 column. Peak 1 i s pure 5.8S RNA, peak 2 pure 5S RNA, peak 3 tRNA. 38. tracking dye Figure II-6a: A gel electrophoresis slab of part ly pur i f ied yeast 5S RNA. The gel 1 was 10%, and the RNA was f ract ion 3 from Fig. 11-4 which corresponds to the gel chromato-graph in Fig. 11-5. 1 2 3 pure 5S RNA tRNA tracking dye Figure II-6b: A gel electrophoresis slab of pure yeast 5S RNA which corresponds to f ract ion, r?2 in F ig. 11-5'. G-100 chromatography of impure 5.8SRNA K- 1 — 2 — . 3 — » | 40 50 fraction 60 Figure 11-7: Pu r i f i ca t i on of 5.8S RNA on a long Sephadex G-100 column Peak 1 i s rRNA, peak 2 pure 5.8S RNA, peak 3 pure 5S RNA 40. on a long column (2x190 cm), but Sephadex G-100 was used instead of G-75 to better separate the higher molecular weight components. F ig. 11-7 contains a typ ica l e lut ion p r o f i l e of the pooled f rac t ion 2 from F ig. 11-4. The peaks correspond to large molecular weight RNA (peak 1), pure 5.8S RNA (peak 2), and pure 5S RNA (peak 3). Peak 1 was discarded while peaks 2 and 3 were separated and prec ip i tated with ethanol and Mg + + . 6. Desalting and Lyophi l i zat ion The prec ip i tated pure RNA species, (tRNA, 5S RNA, 5.8S RNA) were centr i fuged, dissolved in a minimal amount of. deionised water, and desalted on a Sephadex G-25 desalt ing column. They were then l yoph i l i zed and stored as a powder at -20°C to prevent degradation from.i.any contaminating r i bonucl ease present. A l l Raman samples were prepared using these l yoph i l i zed powders. 7. Other Techniques Used in I solat ion and Pu r i f i c a t i on (a) Gel electrophoresis. To test the pur i ty of i so lated RNA species and to confirm the type of RNA present, gel electrophoresis was used. The procedure of Rubin (54) proved to be most usefu l , and i s now b r i e f l y described. The gel was composed of 10% acrylamide, 0.35% N,N'-methylene-bisacrylamide in 50 ml of 20 mM Tr i s -acetate pH 8.0, 1 mM EDTA, 4 M urea, to which was added 0.200 ml of 10% ammonium persulphate and 0.020 ml N,N,N',N'-tetramethylethylenediamine (TEMED). The ammonium persulphate was f resh ly prepared and the TEMED was added to the degassed solut ion ju s t before i t was transferred to the slab apparatus. The gel was allowed to polymerize for 2-3 hr and was then pre-run for 3-4 hr at 20 ma. The buffer reservoirs contained 20 mM Tr is -acetate pH 8.0, 41. 1 mM EDTA, and 4 M urea. The samples were prepared by d isso lv ing 0.1 mg of RNA in 100 ul of buffer containing 20% sucrose and 0.2% bromophenol blue as a tracking dye. Each sample was then heated to 65°C for 1 min and a tota l of 10 ul of th i s solut ion was loaded into each s l o t . The gel was then run for 16 hr at 10-15 ma, or unt i l the tracking dye reached the end of the ge l . The gel was removed from the apparatus, stained for approximately 45 min with a methylene blue solut ion (0.2% methylene blue, 0.02 M NaOAc, 0.02 M HOAc) and destained overnight under running water. (b) concentration of samples. P rec ip i ta t ion of low concentrations of RNA can resu l t in low y i e l d s , while concentrated solutions give about 90% y i e l d s . Also, prec ip i tat ions of large volumes of RNA solutions are awkward and wasteful. Therefore, in many cases, samples were concentrated before p rec ip i ta t ion using an Amicon Diaf lo U l t r a f i l t r a t i o n Apparatus equipped with a PM 10 membrane. Typ ica l ly solutions were concentrated to an A 2 6 Q of about 50 or 2-3 mg/ml RNA (1 mg/ml RNA in 0.1 M NaCl=22.4 A 2 6 0 un i t s ) . (c) d i a l y s i s . An a l ternat ive method to p rec ip i ta t ion and desalt ing of RNA before l yoph i l i z a t i on i s to dialyse away the NaCI by 3 successive dialyses in 20 volumes of deionized H 20. For small quantit ies of tRNA th i s procedure works we l l . However, one must be careful when d ia lys ing 5S RNA or 5.8S RNA since removal of Mg + + from them has a large e f fect of the structure of these species. The nature of th i s d isrupt ive e f fect w i l l be considered in deta i l l a t e r . 42. C. Effect of Mg^ and Renaturation of Chromatography of 5$ RNA, 5.8S RNA Since the presence of Mg + + and heat renaturation have been shown to have a markedteffect on the chromatographic behavior of procaryotic 5S RNA (22), a set of experiments were conducted on part ly pur i f ied yeast 5S RNA and 5.8S RNA to see i f a s imi la r e f fect could be observed. Yeast RNA solutions were dialysed to remove Mg + + and were then l yoph i l i zed . The RNA was then dissolved in 0.2 M NaCl containing 10 mM MgC^. Samples, were chromatographed on Sephadex G-100 (2x190 cm) under d i f fe rent conditions of pretreatment: ( i ) no pretreatment; ( i i ) renaturation at 70°C for 3 min; ( i i i ) renaturation at 75°C for 3 min; and ( iv) renaturation at 80°C for 3 min. D. Preparation of Samples For Raman Spectroscopy 1. Preparation of Normal 5S RNA, 5.8S RNA, and tRNA Samples Samples of tRNA, 5S RNA, or 5.8S RNA were dissolved in H2Q or D20 containing 10 mM phosphate, pH 7, 10 mM MgCl 2, and 100 mM NaCl. Samples were 4% wt/vol made by d i sso lv ing 1 mg RNA in 25 ul buffer. From th i s sample 10 ul was injected into a 0.8 mm i . d . glass c ap i l l a r y sample tube. The sample .•^Av was then renatured at 65° C for 5 min before spectra were recorded. A l l samples were also centrifuged jus t before spectra were recorded to lower the scatter ing background. 2. Preparation of Low Mg + + -containing RNA Samples Two methods were used to obtain low Mg + + -containing 5S RNA and 5.8S RNA; one was as per Chen et a l . (39), and one was a modified procedure. In the f i r s t procedure 5S RNA or 5.8S RNA (15 mg/ml) was heated at 90°C for 10 min in 20 mM NaEDTA (pH 7.0). Af ter the mixture had cooled to room temperature, the RNA was dialysed overnight 43. against 20 mM Tris-HCl (pH 8.0), then dialysed four times, twice each for 6 hr against 40 mM and 5 mM NaOAc (pH 5.0). A l l dialyses were done at 4°C in a p l a s t i c beaker and in the presence of \"Chelex\" res in . In the second procedure 5S RNA (15 mg/ml) was heated to 90° C for 10 min in 20 mM NaEDTA (pH 7.0). It was then cooled and desalted on a Sephadex G-25 column and l yoph i l i zed . The l yoph i l i zed powder was then dissolved in 10 mM phosphate. In both the above procedures the maximum concentration of RNA that could be achieved without v i s i b l e cloudiness was 2% (wt/vol). Even at th i s concentration and af ter centr i fugat ion, the Raman spectra displayed f l oa t i ng baselines ind icat ive of aggregation. In the case of 5S RNA th i s background scatter ing was large enough to make useful spectra unobtainable. 3. Preparation of Samples of Varying pH Raman samples of accurately determined pH are d i f f i c u l t to prepare because of the small volumes involved (15-20 ul) and the fact that the;..phosphate groups in the backbone of RNA have a large buffering capacity. Therefore a less accurate pH determination was inev i tab le . Five or.more samples of 1 mg each of tRNA or 5.8S RNA were weighed out. To each was added 25 ul of phosphate buffer as in normal samples, except in th i s case the pH of the buffer was increased or decreased by the addition of NaOH or HC1 respect ive ly. Although the pH of the buffer could be accurately determined in th i s stage, when the buffer was added to the RNA the pH could no longer be assumed to be the same. Therefore, the pH of the mixture was tested by applying 2-3 ul of the sample to. pH paper of appropriate range which had a 44. colour change for every pH change of 0.5 pH unit . This newly deter-mined pH was assumed to be correct to ±0.25 pH unit . Raman spectra were then recorded for a range of pH values from as low as pH 4 to as high as pH 12 for 5.8S RNA and for tRNA. E. Raman Parameters Raman spectra were recorded on a Spex Ramalog 4 Laser Raman System equipped with a Spectra-Physics 265 Exciter argon laser tuned to the 5148 A exc i tat ion l i n e . 90 geometry was used, with a s l i t width of 8 cm~^ and a laser power of 600 mW at the source. The scan i _ i _ i speed was 1 cm /sec at a period of 5 sec for 4% samples, and 0.5 cm /sec at 10 sec period for the 2% sample. Each reported spectrum was recorded at least 5 times to obtain better average peak i n t en s i t i e s . Also, some samples were tested for molecular i n teg r i t y by gel electrophoresis af ter laser i l luminat ion to insure that they had not been degraded by the laser treatment. 45. CHAPTER III Results A. Physical Characterization of the Structure of 5S RNA and 5.8S RNA 1. Chromatographic Determination of Shape Asymmetry Fig. 111-1 shows a plot of log molecular weight versus e lut ion volume (20 ml/fraction) on a large (10x150 cm) Sephadex G-100 gel f i l t r a t i o n column for a series of standard globular compounds (blue dextran 2000, hemoglobin, myoclobin, fe r r i cyan ide) . The stra ight l i ne indicates that globular molecules elute roughly l i nea r l y as a function of log molecular weight. Therefore, i f a compound of known molecular weight elutes in a pos it ion far from th i s l i ne i t must have a shape which deviates s i g n i f i c an t l y from that of a globular molecule. I f the p lot f a l l s on the lower l e f t side of the l i ne the molecule must appear larger than globular molecules of the same molecular weight. The s ize i s determined by how eas i l y the molecule f i t s into thejpores in the ge l , and can be affected by axia l r a t i o , shape asymmetry, etc. F ig. 111-1 indicates that tRNA roughly resembles the s ize of globular molecules because of i t s nearness to the l i n e , while 5S RNA and 5.8S RNA have a much larger apparent s ize than globular molecules. Therefore, 5S RNA and 5.8S RNA are s i g n i f i c an t l y more asymmetric in overal l shape than tRNA. As w i l l be discussed l a t e r , the difference i s l i k e l y due to var iat ion in both the amount of t e r t i a r y fo ld ing and the axia l r a t i o of the molecules. 2. The Ef fect of Mg + + and Heat Renaturation on Chromatographic Behavior As mentioned in the Experimental section the Mg + + was dialysed out of a mixture of 5S RNA and 5.8S RNA, and the RNA was then chroma-tographed by gel f i l t r a t i o n a f ter renaturation at room temperature 46. Gross Shapes of tRNA, 5S RNA.5.8S RNA 2 I- I . > i i I i . 1 1 1— i 100 200 300 400 fraction Figure I I I - l : A standard plot of log molecular weight versus e lut ion volume for yeast RNA elut ion from a Sephadex G-100 column. Standard solutions of hemoglobin, myoglobin, blue dextran, and ferr icyanide (A) were used. The RNA species are as indicated. 47. Renaturation of Mg-depleted 5S.5.8S RNAs 5.8S 5S (a)no renaturation Figure II1-2: Elut ion volumes of Mg + +-depleted 5.8S RNA, 5S RNA under various renaturation conditions. Conditions were 3 min at (a) room temperature, (b) 70°C, (c) 75°C, (d) 80°C. A l l renaturations were in the presence of Mg + + . 48. (no renaturat ion), at 70°C for 3 min, at 75*0 for 3 min, and at 80°C for 3 min. A number of other reports suggest the importance of Mg + + in maintaining native conformation, as well as the presence of stable, non-native forms for yeast 5S RNA, E. c o l i 5S RNA, and tRNA (22,39,49). F ig. 111-2 shows that yeast 5S RNA and 5.8S RNA conformations are sens i t ive to the presence of Mg + + and the renaturation procedure. In p r o f i l e (a) four peaks are present, with only two of them corres-ponding to native 5.8S RNA and 5S RNA. As the mixture 's renaturation temperature i s increased, the other two ear ly -e lu t ing peaks disappear, giving a p r o f i l e which consists largely of peaks with the same chroma-tographic properties as native 5.8S RNA and 5S RNA. This comparison suggests, therefore, that the removal of Mg + + from both 5.8S RNA and 5S RNA causes a pa r t i a l denaturation of the native conformation, and that th i s process can be reversed to a s i gn i f i can t degree by renatura-t ion at high temperatures in the presence of Mg + + . The denaturation is apparently accompanied by aggregation, as witnessed by the ear ly -el uting peaks. More w i l l be said about the importance of renaturation and Mg + + in both the Raman experiments and the discussion. B. Origins of Raman Lines Table 111-1 gives the or ig ins of the various dist inguishable l ines in an RNA spectrum between 600 cm - 1 and 1750 cm\"'''. As can be seen from the table and from Fig. I I I -3, there i s a r e l a t i v e l y small number of l i n e s , mostly due to r ing stretching and carbonyl stretching of nucleotide bases, with two s i gn i f i cant l ines due to symmetric stretching of the phosphate group in the sugar phosphate backbone of the molecule. The types of v ibrat ions giving r i se to the Raman l ines correspond perfect ly with the most important features of RNA structures: 49. the arrangement of phosphate groups into he l i ca l regions of highly ordered conformation versus unordered s ingle stranded regions, the base stacking interact ions so important in maintaining secondary and t e r t i a r y structure (9), and the carbonyl groups which are d i r e c t l y involved in H-bonding between AU, GC, and GU pa i r s . Therefore Raman spectroscopy should be a very sens i t ive probe of RNA structure when compared to RNA molecules or polynucleotides of known structure. The l ines of most interest in structural studies are those at 670 cm - ' ' ', 725 cm\" 1 , 786 c m - 1 , 814 cm\" 1 , 1100 cm\" 1 , and 1234 cm\" 1 in H 20 spectra, and the l ines at 1655 cm\" 1 and 1688 cm\" 1 in D^ O spectra. The l i ne at 1100 cm\" 1 i s used as a ca l i b ra t i on l i n e , and a l l other l i ne i n tens i t i e s are normalized to i t s i n tens i t y , since many workers have shown th i s l i ne to be insens i t ive to conformation and ind icat ive of thettota l concentration of phosphate groups present in the sample. Intens i t ies are taken as peak heights from a baseline drawn tangent to 1065 cm\" 1 and 1130 cm\" 1 for the 1100 cm\" 1 l i n e , from a baseline drawn tangent to 740 cm\" 1 and 840 c m - 1 for the 785 cm\" 1 l i n e , and 814 cm\" 1 l i n e , and -1 -1 -1 from a baseline drawn tangent to 700 cm and 740 cm for the 725 cm l i n e . _ i The Raman l i ne at 814 cm i s assigned as a symmetric stretch of the -0-P-0- linkage holding adjacent sugars together (40). The normalized intens i ty of th i s l i ne has been shown to be a sens i t ive measure of the amount of order or r i g i d i t y in the backbone structure of tRNA (40). When the intens i ty of th i s l i ne for tRNA i s compared to that of completely ordered he l i ca l systems (e.g. poly rA-poly rU or poly rC-poly rG), tRNA i s approximately 85% ordered, a number in Table 111-1: The Raman peaks H 2 0 D 2 0 S o l u t i o n s o l u t i o n 635 ( 0 ) 625 ( 0 ) 670 (2) 668 ( 2 ) 725 ( 3 ) 718 ( 3 ) 786 ( 6 ) 780 ( 6 ) 814 ( 5 ) 814 ( 6 ) 867 (2) 860 ( 2 ) 918 (1) 915 ( 1 ) 990 ( 1 ) 1003 ( 1 ) 1049 ( 2 ) 1045 ( 1 ) 1100 5 1100 4 1185 ( 2 ) 1185 ( 1 ) 1243 6 1255 ( 5 ) 1257 ( 5 ) 1300 ( 4 ) 1310 ( 7 ) 1320 (7) 1318 ( 6 ) 1340 (7) 1345 ( 7 ) 1370 ( 3 ) 1380 ( 5 ) 1390 ( 2 ) 1422 ( 2 ) 1460 2 ) 1460 ( 2 ) 1484 ( 1 0 ) 1480 8 1503 ( 2 ) 1527 ( 2 ) 1526 ( 3 ) 1560 ( 2 ) 1575 ( 8 ) 1578 ( 1 0 ) 1622 ( 3 ) 1658 ( 4 ) 1688 ( 4 ) * From Thomas (40). normally found in RNA samples and the i r o r i g i n s . * A s s i g n m e n t R e s i d u e P r o b a b l e O r i g i n ( A d e ) , U r a , C y t O u t - o f - p l a n e r i n g d e f o r m a t i o n s : C=0 d e f o r m a t i o n s , e t c . Gua R i n g s t r e t c h i n g Ade R i n g s t r e t c h i n g U r a , C y t R i n g s t r e t c h i n g P h o s p h a t e - C - O - P - O - C - S y m m e t r i c s t r e t c h i n g A d e , U r a , G u a , C y t R i n g s t r e t c h i n g S u g a r , p h o s p h a t e - C - 0 - s t r e t c h i n g S u g a r , p h o s p h a t e ( ? ) A d e , U r a , C y t S u g a r , p h o s p h a t e - C - 0 - s t r e t c h i n g P h o s p h a t e 0 = P - 0 \" S y m m e t r i c s t r e t c h i n g A d e , U r a , G u a , C y t R i n g ; e x t e r n a l C-N s t r e t c h i n g U r a , C y t R i n g s t r e t c h i n g A d e , C y t R i n g s t r e t c h i n g C y t R i n g s t r e t c h i n g A d e , U r a , C y t R i n g s t r e t c h i n g A d e , Gua R i n g s t r e t c h i n g Ade R i n g s t r e t c h i n g A d e , Gua A d e , U r a , Gua U r a A d e , Gua R i n g s t r e t c h i n g ; CH d e f o r m a t i o n s U r a , C y t R i n g s t r e t c h i n g ; CH d e f o r m a t i o n s ( A d e ) , Gua R i n g s t r e t c h i n q C y t ( A d e ) , C y t R i n g s t r e t c h i n g U r a A d e , Gua R i n g s t r e t c h i n g D o u b l e bond s t r e t c h i n g v i b r a t i o n s o f p a i r e d and u n p a i r e d b a s e s ; m a i n l y C=0 s t r e t c h i n g 51. Phe good agreement with the X-ray crysta l lographic data for tRNA -1 -1 The Raman l ines at 670 cm and 725 cm are due to r ing v ibrat ions of G-bases and A-bases respect ively. The intens i ty at 670 c m - 1 displays inverse hypochromism, i . e . in tens i ty increases' for a given number of stacked G-residues over the same number of unstacked G-residues. The greatest amount of G-stacking i s expected for a series of consecutive G-residues in a he l i ca l manner, while i t i s least when non-consecutive G-residues are arranged in random c o i l regions. Therefore, th i s l i ne can give information on the arrangement of G-residues in RNA molecules of known sequence. The Raman l i ne at 725 c m - 1 due to A-base r ing v ibrat ion i s hypochromic, as are a l l the other important Raman l ines in the spectrum (35-44). This l i ne increases in intens i ty with increased stacking of A-residues, and can be used to determine the conformation of A-residues in RNA. S im i l a r l y , the l i ne at 786 c m - 1 i s due to a combination of C and U base r ing vibrations and can be used as an estimate of the amount of stacking of these bases, while the l i ne at 1234 c m - 1 due to U-residues can be usedtto determine the re l a t i ve stacking of U-residues. The l ines in the in-plane r ing v ibrat ion area (1200-1500 cm -''') are a l l hypochromic, and appear to be unique for each RNA species, much l i k e the f ingerpr int region of infrared spectra, but the use of th i s region for structural inferences is d i f f i c u l t since the bands overlap, are usually due to more than one type of base, and are not as strongly hypochromic as those mentioned e a r l i e r . The l i ne at 1234 cm\" 1 i s the most strongly hypochromic, but i t i s part ly overlapped by the l i ne at 1255 cm\" 1 which interferes with i t s i n t e r -pretat ion. 52. In D20 spectra, the carbonyl stretch region (1550-1760 cm -''') i s no longer obscured by the solvent peak, so interpretat ion of the peak i n tens i t i e s in th i s region i s possible. By far the largest contributor is the carbonyl groups of u r a c i l , which give strong peaks at 1660 cm - ' ' ' -1 -1 and 1686 cm . The peak at 1686 cm i s due to non-H-bonded urac i l carbonyls, while that at 1660 cm-\"'' i s due to the carbonyl bonds which are weakened by H-bonding interact ions. Therefore, a comparison of .the i n tens i t i e s of these two peaks gives a semi-quantitative estimate of the percentage of H-bonded U-residues in the RNA molecule. C. The Raman Spectra of 5S RNA and 5.8S RNA at pH 7 The Raman spectra of yeast 5S RNA and 5.8S RNA were obtained and the i r features analysed using the above treatment. In th i s sect ion, to avoid confusion, the results w i l l be considered separately; however, in the d i s cus s ion l l w i l l show how these independent analyses suggest a s imi la r secondary structure for these RNA species which i s adaptable to a l l known species of 5S RNA and 5.8S RNA. In a l l cases only comparative data i s useful since the absolute assignment of any conformational property from the Raman l ines alone i s d i f f i c u l t (35,36,41). However, since the Raman spectra can be compared to that of tRNA P ' i e (whose complete structure is known), the technique becomes extremely valuable in determining structural s i m i l a r i t i e s and di f ferences. Also, when Raman data are used in cooperation with other physical data, de f in i te structural features can be extracted. However, since the Raman intens i ty i s dependent on the re l a t i ve number of a given base present in the RNA species, as well as on the conformation of those bases, one must be sure to correct the i n tens i t i e s for the re l a t i ve proportions of a given base before drawing structural inferences from the intens i ty data. 53. 1. Raman Spectra of Yeast 5S RNA F ig. IM-3,111-4 give the Raman spectra for yeast tRNA and 5S RNA in h^ O and D2O, and Table 111-2 l i s t s the normalized peak i n tens i t i e s and or ig ins of the various resolved Raman l i ne s . The normalized intens i ty at 814 cm - 1 i s s l i g h t l y larger for yeast 5S RNA than for tRNA. As mentioned above, th i s C-0-P-O-C stretch re f l ec t s the amount of order in RNA backbone structures and, since tRNA i s about 85% ordered by th i s c r i t e r i o n , yeast 5S RNA must be approximately 90% ordered in the native state. This suggests that yeast 5S RNA must be a r i g i d , largely base paired molecule which has l i t t l e freedom of motion in so lut ion. The other Raman l ines mentioned above, whose i n tens i t i e s are sens i t ive to base stacking interact ions , also give information on 5S RNA structure. After correct ion for the difference in G-base content between tRNA (\"30%) and 5S RNA (27.5%), the 670 cm\" 1 l i ne i n tens i t i e s due to G-bases (inverse hypochromic with respect to stacking) become equal within experimental uncertainty, suggesting a s imi la r extent of G-base stacking for the two RNA types. The corrected 670 cm intens i ty for 5S RNA i s nevertheless greater than that reported for tRNA P h e (39), ind icat ing greater G-stacking for yeast 5S RNA than for yeast tRNA P h e . The 725 cm\" 1 l i n e intens i ty due to A-bases (hypochromic) indicates s imi la r A-stacking in tRNA and 5S RNA from yeast, a f te r correcting for the i r r e l a t i ve A-base contents. Thus, although 5S RNA has two regions of 3 consecutive A-bases in i t s sequence, these regions evidently do not give r i se to a large A-stacking contr ibut ion, and LO 1600 1400 1200 , 1000 800 600 cm\"' Figure II1-3: The Raman spectra of (a) 5S RNA, and (b) tRNA. Solutions were 4% wt/vol in 10 mM phosphate pH 7, 10 mM MgCl ?, 100 mM NaCI in H ?0. — 5S RNA --tRNA 1700 x 1600 cm-1 1500 Figure 111-4: The carbonyl Raman region of 5S RNA ( ) and tRNA ( — ) . Solutions were 4% wt/vol in 10 mM phosphate pH 7, 10 mM MgCl 2 100 mM NaCI in D20. 5 6 . Table 111-2: Frequencies and Intens i t ies of Some Raman Lines of Yeast tRNA and 5S RNA* Frequencies (cm -*) Or ig in of Line Relative Intensity tRNA 5S RNA 6 7 0 G 0 . 5 3 0 . 5 0 7 2 5 A 0 . 5 8 0 . 7 2 7 8 5 C,U 2 . 2 4 2 . 3 2 8 1 4 - 0 P 0 - 1 . 61 1 . 7 5 1 1 0 0 P O 2 ~ 1 . 0 0 1 . 0 0 1 2 3 4 U 1 . 0 5 1 . 2 6 1 2 5 1 CA 1 . 1 0 1 . 3 1 1 3 0 0 CA 0 . 9 0 0 . 8 4 1 3 2 1 G 1 . 1 5 1 . 1 6 1 3 3 8 A 1 . 2 2 1 . 2 3 1 3 7 5 G,A 0 . 8 1 0 . 8 1 1 4 8 5 G,A 1 . 8 8 1 . 7 3 a l l spectra are averages of at least 5 spectra are therefore probably in non-helical s ingle stranded regions. The intens i ty of-the 7 8 4 cm\"* l i ne (hypochromic) due to C- and U-bases is approximately the same for yeast tRNA and 5S RNA, but since 5 S RNA has more C- and U-bases than tRNA, there must be more stacking of one or both of C and U in 5S RNA. Greater U-stacking i s more probable, since the intens i ty at 1 2 3 4 cm - ' ' ' (hypochromic) due to U-bases does not increase in proportion to the greatly increased U-base content of 5 S RNA compared to tRNA. 57. In the D^O spectra, as mentioned before, the ra t i o of the carbonyl stretch i n tens i t i e s at 1660 cm\" 1 (H-bonded) and at 1688 cm\" 1 (non-H-bonded) i s a d i rec t measure of the r e l a t i ve percentage of paired U-residues (37). Fig. 111-4 shows that tRNA and 5S RNA from yeast have peak intens i ty rat ios v i r t u a l l y ident ica l to each other as well as to tRNA P h e (not shown)(37). Therefore, yeast 5S RNA must have approximately two-thirds of i t s U-residues in base-paired configurations Phe as does tRNA . F i na l l y , since U-residues are r e l a t i v e l y evenly d i s t r ibuted throughout the 5S RNA sequence, th i s high degree of U-base pair ing implies a high overal l degree of base pa i r ing , in agreement -1 with the high degree of backbone \"order\" indicated from the 814 cm intens i ty data. 2. Comparison of 5.8S RNA and t.RNA Raman Spectra The Raman spectra of 5.8S RNA and tRNA are shown in F ig. II1-5 and F ig. 111-6 while the posit ions and i n tens i t i e s of prominent l ines a r e i l i s t e d in Table 111-3. Again the l i ne i n tens i t i e s were normalized with respect to the intens i ty of the P0 2~ l i ne at 1100 cm\" 1 . The intens i ty at 814 cm\" 1 i s ident ica l for 5.8S RNA and tRNA. Since tRNA i s 85% 'ordered at the phosphodiesterdbaGkbone,; 5..8S.'. .' . RNA must also be 85% ordered in solutions containing Mg + +i«ions. The Raman intens i ty at 670 cm\" 1 (inverse hypochromic for G-stacking) is about 35% bigger for 5.8S RNA than for the corresponding tRNA l i ne a f ter correction for ther'difference in.G-base content between tRNA (A/30%) and 5.8S RNA (23%). In f a c t , compared to tRNA P h e (39), the 5.8S RNA l i n e i s 60% bigger. Therefore, 5.8S RNA has a greater proportion of stacked G-bases than does tRNA. This stacking ef fect 58. confirms the presence of the GC-rich arm.(see F ig. IV-2), since that arm contains a segment with f i ve consecutive h e l i c a l l y stacked G-bases which should give a very large G-stacking contr ibution to the 670 c m - 1 i n tens i ty . It i s interest ing to note that the 670 c m - 1 in tens i ty for E. c o l i 5S RNA also has a large G-stacking contr ibution(44), suggesting that a s imi la r GC-rich arm i s present.in that RNA species. The Raman l ines at 725 c m - 1 due to A-bases (hypochromic) have equal i n tens i t i e s for 5.8S RNA and tRNA, a f te r correct ion for d i f fe rent A-base contents. Therefore, they both have s imi la r stacking. The 785 c m - 1 l i ne (hypochromic) should be about 20% bigger for 5.8S RNA than tRNA, i f both have s imi la r amounts of C and U stacking. However, since the intens i ty for 5.8S RNA i s only marginally larger than for tRNA, 5.8S RNA must have more C and/or U stacking. A large stacking contr ibution i s expected for the four C-bases paired to G-bases in the GC-rich arm; a l so, there are several instances of three consecutive U-bases in 5.8S RNA, with potent ia l l y large contributions to U-base stacking. Furthermore, the hypochromic Raman l i ne due to U-residues at 1234 cm\" 1 has a much greater r e l a t i ve intens i ty for tRNA than for 5S RNA, when U-base contents are considered. Therefore, the 5.8S RNA UUU sequences must be e f f i c i e n t l y stacked in he l i ca l regions or via some t e r t i a r y interact ions. In D^ O the carbonyl stretch region can again be used to estimate the % of base paired U-residues. F ig. 111-6 indicates that 5.8S RNA has proportionately more base paired U-residues than mixed tRNA. In pa r t i cu l a r , since the 1660/1688 intens i ty ra t i o i s larger for 5.8S RNA than for tRNA P h e (39), and since two-thirds of the U-residues of tRNA p h l 59. are base paired (9,10), 5.8S RNA must have more than 70% of i t s U-residues in base paired configurations. 3. Comparison of High and Low Mg + + 5.8S RNA Raman Spectra The Raman spectra of normal and low Mg + + 5.8S RNA (Fig. 111-5) and the normalized peak i n tens i t i e s (Table 111-3) indicate an important ro le for Mg + + ions in 5.8S RNA structure. F i r s t , the 12% decrease in in tens i ty of the 814 crrf* l i ne suggests a decrease in order of the phosphodiester backbone in the absence of Mg + + . Second, the 15% decrease in intens i ty of the 670 cm - ' ' l i ne indicates a small decrease in G-base stacking in 5.8S RNA lacking Mg + + , while the decrease in in tens i ty of the 725 cm\"\"'\" l i ne suggests an increase in A-stacking. These observations are confirmed by the neg l ig ib le change in the -1 -1 i n ten s i t i e s at 1375 cm and 1485 cm , because the decreased G-stacking in tens i ty contr ibution to these l ines i s largely compensated for by the increased A-stacking contr ibut ion. F i na l l y , the s l i gh t increase in in tens i ty of the 1234 cm - ' ' ' l i n e suggests a small decrease in U-stacking when Mg + + i s absent. These results indicate that the removal of Mg + + from 5.8S RNA causes a s l i gh t disordering of the backbone and a rearrangement of base stacking interact ions. These changes are ident ica l in type but less in degree than for tRNA P ' i e (39), which has three (or four) strong Mg + + binding s i tes (70,71). The structure of 5.8S RNA i s therefore expected to be less dependent on Mg + + than i s the structure of tRNA, although Mg + + i s required to maintain native conformation in both cases. 4. The Effect of pH on 5.8S RNA and tRNA Raman Spectra The Raman spectra of 5.8S RNA were obtained on a series of samples over the pH range of 3.8 to 12.2, and for tRNA from pH 7 to pH 12, Figure II1-5: The Raman spectra of (a) tRNA, (b) 5.8S RNA, (c) low Mg 5.8S RNA. Solutions were 10 mM phosphate pH 7 and 100 mM NaCI in H 20. Solutions of (a) and (b) were 4% wt/vol and contained 10 mM MgCl 2- Solution (c) was 2% wt/vol and contained no MgCl 9. 61. Figure 111-6: The carbonyl Raman region of 5.8S RNA ( — ) and tRNA ( Solutions were 4% wt/vol in 10 mM phosphate pH 7, 10 mM MgCl 2, 100 mM NaCl in D20. •). 62. Table II1-3 ++ Comparison of Raman Lines of tRNA, 5.8S-RNA, and low Mg 5.8S-RNA Frequencies (cnf^) Origin of l i ne 670 725 785 814 1100 1234 1251 1300 1321 1338 1375 1485 G A C,U -0P0-P0~ U C,A C,A G A G,A G,A tRNA 0.53£ 0.58 2.24 1.61 1.00 1.05 1.10 0.90 1.15 1.22 0.81 1.88 Relative Intensity 5.8S-RNA 0.56* 0.75 2.28 1.61 1.00 1.21 1.18 0.93 1.19 1.65 1.00 1.91 ++ low Mg 5.8S-RNA 0.48 b 0.61 2.31 1.41 1.00 1.35 1.41 0.90 1.36 1.67 1.10 1.90 Each value represents an average from at least 6 spectra; each indiv idual b normalized in tens i ty d i f f e r s by less than ±5% from the average. As in (a), with average from at least 3 spectra. to see i f conformational changes could_ be followed. However, when analysing these spectra, one must remember that certa in of the l i ne i n tens i t i e s w i l l be affected by deprotonation of the bases at the i r pK a values, as well as by changes in conformation. Only two of the bases have pK a values in the region studied (UMP pKa=9.5, GMP pKg=9.4) for the N-3 proton, and therefore, any Raman l ines involving these bonds w i l l be affected. (a) pH and the backbone order of RNA. F ig. III-7a,b show the spectra of 5.8S RNA and tRNA as a function of pH, while F ig . 111-8. shows a plot of the normalized intens i ty of the 814 cm~* l i ne verses pH. Since th i s l i ne intens i ty i s d i r e c t l y related to the backbone 63. order of the RNA, one should be able to fol low the disruption of secondary and t e r t i a r y structure by measuring the change in intens i ty of the 814 cm - 1 l i n e as a function of pH. Fig. I I I -8 shows that, for both tRNAd and 5.8S RNA, the conformation appears to be insens i t ive to pH over the range of 7 to 10, and that the molecule becomes highly disordered between pH 10 and pH 12. For tRNA, the pK of complete unfolding is about 11, while for 5.8S RNA the pK i s 10.5. These pK values correspond with the massive a l te rat ion of a l l l ines involving G- and U-bases, and therefore suggest that deprotonation of the bases i s breaking the secondary H-bonds between GC and AU pa i r s , destroy-ing secondary structure. The use of these data to determine i f more subtle changes are taking place i s , however, l imi ted by two factors : the accuracy with which Raman intens i ty can be determined (±5%), and the r e l a t i v e l y small number of points on the curve. The data do suggest, however, that the normalized in tens i ty at 814 c m - 1 i s r e l a t i v e l y insens i t ive to changes in t e r t i a r y structure, since the data to be presented next suggests that base stacking interact ions are affected by pH<10, while the order i s not appreciably changed unt i l pH>10. (b) the e f fect of pH on base stacking in tRNA. As mentioned above, a number of Raman l i ne i n tens i t i e s are very sens i t ive to base stacking interact ions. Therefore, these l ines can be diagnostic of changes in conformation, espec ia l l y t e r t i a r y structure a l te ra t ions . F ig. 111-9a-f shows a number of normalized Raman in tens i t i e s plotted verses pH. F ig. 111-9a indicates that the 725 cm -''' i n tens i ty due to A-bases increases at pH 10.5-11.5 by about 10-15%. Since th i s rang breathing 64. v ibrat ion i s in tens i ty invariant throughout the ent i re pH range for AMP (43), the change in intens i ty for tRNA must be a resu l t of unstacking of A-bases. This unstacking, however, occurs only at high pH (pK 11) where massive disruption due to deprotonation of U- and G-bases i s taking place. Therefore, the stacking of A-bases i s largely confined to secondary stacking, with a minimal amount of t e r t i a r y A-stacking. This i s in contrast to Raman data on the stacking in a spec i f i c tRNA (tRNA ), where disruption of t e r t i a r y structure by Mg + + depletion i s stated as causing an increase in A-stacking (39). However, the authors cannot explain the i r f inding Phe in terms of the known crysta l structure of tRNA . They do note Phe a large decrease in A-stacking when tRNA is melted completely at 85°C, in agreement with the present f inding for mixed tRNA. Fig. 111-9b indicates that the 670 cm-\"'' l i ne in tens i ty due to G-bases undergoes two independent decreases in in tens i ty . Previous studies suggested that th i s l i ne has a large decrease in in tens i ty when t e r t i a r y structure i s disrupted by removal of Mg + + (39). This smaller decrease has been attr ibuted to the unstacking of the conserved bases G-18, G-19, and G-57 (see F ig. 1-3) which are stacked via t e r t i a r y interact ions in the crysta l structure (9,10). Therefore, the f i r s t slope in F ig. 111-9b at pH 8.5-9.5 may be due to the disruption of t e r t i a r y interact ions of G-bases, while the much larger slope at pH 10.5-11.5 may be a resu l t of destruction of secondary structure. This predict ion i s apparently confirmed by the intens i ty changes of the 1375 cm\"* l i ne which i s due to G-bases (hypochromic) and A-bases (Fig. I I I -9e,f ) . The intens i ty of th i s l i ne increases near pH 9.5, suggesting a decrease in stacking of G- and/or A-bases, before i t decreases sharply, l i k e l y as a resu l t of deprotonation of G-bases 65. at high pH. The normalized intensities of the Raman l ines at 784 cm - 1 and 1234 c m - 1 are ind icat ive of unstacking of C-bases and/or U-bases when the H-bonds of the secondary structure are broken. Previous workers have reported that the 784 cm\" 1 l i ne for both uridine and eyt id ine i s in tens i ty invariant for changes in pH between pH 7 and pH 12.8 (43); therefore th i s large hypochromicity must indicate the presence of s i gn i f i can t C- and U-base stacking in native tRNA. However, since no in tens i ty increase is seen below pH 10, most of the U-bases and C-bases must be stacked Via secondary interact ions. This f ind ing i s -supported by the Raman l i ne at 1234 cm - 1 which has only the sharp decrease in in tens i ty at pH 10 ind icat ive of deprotonation (43)(F ig. I I I -9f). Since th i s l i n e i s very hypochromic and no in tens i ty increase occurs at pH99-10, there must be l i t t l e t e r t i a r y U-stacking. (c) the e f fect of pH on base stacking in 5.8S RNA. F ig. 111-10a,b give the i n ten s i t i e s of the 725 cm - 1 l i n e due to A-bases and the 670 c m - 1 l i ne due to G-bases as a function of pH for yeast 5.8S RNA. Unfortunately, not enough points are present for a detai led analys i s , due to the l i m i t of sample ava i l ab le , as well as the d i f f i c u l t y in obtaining spectra for 5.8S RNA over the pH range. However, the data contained in the f igure generally agree with the data obtained in the absence of Mg + + presented e a r l i e r . The 725 c m - 1 intens i ty appears to decrease around pH 9 before increasing when the secondary structure i s destroyed by deprotonation. This indicates that the A-bases may be more e f f i c i e n t l y stacked when the t e r t i a r y structure i s disrupted, in agreement with the spectrum 66. of 5.8S RNA in the absence of Mg + + . In f a c t , the in tens i ty of the low Mg + + 5.8S RNA at 725 cnf * i s ident ica l to the pH 11 in tens i ty at that frequency. The same increase in base stacking i s observed in the related E. c o l i 5S RNA when i t s t e r t i a r y structure i s perturbed (8). The 670 cm\"* intens i ty also appears to display a biphasic decrease, as in tRNA. Thus, some G-base unstacking may be occurring at about pH 9. This unstacking i s then at t r ibutab le to t e r t i a r y unfolding, and agrees n ice ly with the unstacking seen in low Mg + + 5.8S RNA. 5. Other Effects of pH on the Raman Spectra of tRNA and 5.8S RNA Since C and A do not undergo deprotonation between pH 7 and-pH 12, the changes in the i r peak posit ions and i n ten s i t i e s should be l im i ted to the in tens i ty var iat ions due to unstacking of bases as described above. However, U and G should undergo changes of in tens i ty and band pos i t i on , espec ia l ly in those bands involving the N-3-H bond. These changes have been observed for the mononucleosides (43), and the present spectra have the same features. For U-residues in RNA, the most diagnostic band i s at 1234 cm * (1230 cm\"* in ur idine) which greatly reduces in in tens i ty and s p l i t s -1 -1 -1 into three bands aroung 1200 cm , 1240 cm , and 1300 cm , the l a t t e r two overlapping with other C and A bands in the RNA spectra. The appearance of a small in tens i ty band at 1200 cm\"* i s notable in both RNA species (Fig. I I I-7a,b), as i s the large decrease in intens i ty at 1234 cm'*. For G-residues a number of bands are diagnostic of deprotonation. For guanosine, the r ing-stretching band at 1575 cm - * undergoes a frequency s h i f t to 1590 cm\"* (43). A s imi la r s h i f t can be seen for both RNA species, while the presence of the doublet at pH 11.2 in tRNA may indicate that some of the others are protected. The band at i t undergoes the large decrease in nucleoside studies. 67. G-bases are deprotonated while 1485 cm-\"'' i s also diagnost ic, and intens i ty expected from mono-Figure 111-8: The ef fect of pH ontthe order of the phosphodiester backbone as measured by the 814 cm - 1 l i n e intens i ty for (a) tRNA, and (b) 5.8S RNA. 71. Base Stacking vs.pH for5.8S RNA 72. CHAPTER IV Discussion A. Secondary and Tert iary Structure of 5.8$ RNA 1. Constraints of Structure From Raman and Other Physical Data The present Raman results suggest that yeast 5.8S RNA has the fol lowing propert ies: ( i ) a highly ordered backbone structure s imi la r to that of tRNA and ind icat ive of a high degree of base pa i r ing ; ( i i ) a GC-rich arm giving r i se to extensive G (and C) stacking; ( i i i ) a secondary and t e r t i a r y structure containing over 70% base paired U-residues which are s i g n i f i c an t l y stacked; ( iv) a smaller structural requirement for Mg + + than tRNA suggesting e i ther less extensive t e r t i a r y fo ld ing or less dependence on Mg + + for th i s f o ld ing ; (v<),only moderate A-stacking; and (vi) a t e r t i a r y structure involving s i gn i f i can t G-stacking and A-unstacking which reverses when the t e r t i a r y structure i s perturbed by removing Mg + + or ra i s ing the pH of the so lut ion. Other' physical measurements suggest the fol lowing addit ional propert ies: ( v i i ) an exposed s ingle stranded region near base 40 which undergoes enzymic pa r t i a l hydrolysis (55); ( v i i i ) a shape s imi la r to E. c o l i 5S RNA since i t binds E. c o l i ribosomal proteins; and ( ix) a large shape asymmetry as determined by chromatographic mobi l i ty . 2. Incompatibi l i ty of Previous Structures With Raman Data Neither of the previously proposed structures (Fig. IV-la,b) has a l l of the above properties. Rubin's model (Fig. IV-la) has large regions of unpaired bases which would y i e l d i n su f f i c i en t order in the molecule. Also, as shown in Table IV-1, th i s model has only 44% of i t s U-residues in base paired regions, or less than two-thirds the number predictedbby the Raman data. Therefore, Rubin's model i s too \"open\" to 73. f i t the Raman data. Although the a l ternat ive model of Nazar et a l . (Fig. IV-lb) has a less open structure than Rubin's model, the Nazar structure contains neither a high enough percentage of base paired U-residues nor an unpaired GAAC region with which to bind tRNA.^ Moreover, when adapted to d i f fe rent 5.8S RNA species, the Nazar-type structure lacks consistency in conserved base paired regions, and lacks the s t a b i l i t y expected for a sequence which shows a very high degree of homology between d i f fe rent plant and animal species (58, 72-74). Such homologous sequences imply a highly conserved pattern of base pa i r s , but Nazar's model contains only the GC-rich arm and minor homologies in the UA-rich region. Also, the s t a b i l i t y number (as defined by Tinoco et a l . (15)) for the Nazar structure for hepatoma 5.8S RNA i s 75% larger than for yeast 5.8S RNA, which seems highly un l ike ly for a largely conserved sequence. In summary, the model of Nazar et a l . has too few base paired U-residues and does not produce consistent structures when extended to other 5.8S RNA species. 3. New Cloverleaf Model F i t s Raman and Other Data Based on the Raman data for yeast 5.8S RNA, I have developed a new c lover leaf secondary structure (Fig IV-2) which exh ib i t s a l l the above properties and those from other physical studies. In add i t ion, I have adapted the new c lover leaf structure to Novikoff hepatoma 5.8S RNA (Fig. IV-3), E. c o l i 5S RNA (Fig. IV-4), and eucaryotic 5S RNA (Fig. IV-6, IV-7). Reasons for se lect ing the new model over previously proposed structures are numerous. F i r s t , only the c lover leaf model accounts for a l l the l i s t e d Raman data. The structural s i m i l a r i t y of the 74. 5.8S RNA and tRNA c lover leafs accords with the i r s imi la r backbone order, while the percentage of base paired U-residues {12% for the c lover leaf structure) agrees with the Raman determination. Further-more, the c lover leaf contains a GC-rich arm and accounts for the low level of A-stacking by forcing many A-residues into hairpin loops, bulges, or i n t e r i o r loops. Also, the removal of Mg + + from a yeast 5.8S RNA solut ion produces Raman ef fects very s im i la r to those for Phe tRNA , again supporting a s imi la r secondary structure. Second, the new c lover leaf secondary structure i s conserved among d i f fe rent 5.8S RNA species, as suggested by the base sequence homology. Not only are the s t a b i l i t y numbers for yeast and hepatoma 5.8S RNA very s im i l a r , but the to ta l numben/of base pairs also remains constant (Table IV-1). The c lover leaf structure allows many conserved regions to be base paired in the same manner among d i f fe rent types of 5.8S RNA (Fig. IV-2). Th i rd, the new c lover leaf accommodates a wide range of data reported by othernworkers. Based on ethidium bromide binding, Van et a l . (14) report that yeast 5.8S.RNA contains about twice as many UA base pairs as Novikoff hepatoma 5.8S RNA, in agreement (Table IV-1) with the cd'overleaf (23 pairs vs. 12 pa i r s ) . Those authors also predict a GC/AU ra t i o close to 1 for yeast 5.8S RNA, and a stable stem region and s imi la r degree of base pair ing for both yeast and rat 5.8S RNA, again consistent with the c lover leaf structure. Furthermore, the points of enzyme par t i a l hydrolysis for Novikoff hepatoma 5.8S RNA (Fig. IV-3) support the c lover lea f , since a l l cleavage points occur in s ingle stranded regions or in strained regions opposite bulges (55). u c (a) Rubin's model A A G A U-A A-U C-G G-C C -G A G A A C G 75. P A A A C U U U C A A C A A C G G A U C U C U U G G U U C U \" A U A C G U A A U G U G A A V U G C A ^ A U U C C G U G A C G • • -U U U A C U G C G A G U U U G U C C G U A C ' G C G ' . C , GUUACACGCAA GUUUiJAAGcUACu ,G CK fir (b)Nazar's model u u G G u u c c u u p A A • A C U U U C A A C A A C G G A U ° C G C A U C G A U G A A A A C G C A G C G U U U U G C G A G U UUGUCpJJAC-GCGU AGCUACU. . G C G U A A A ^ G - C U - A A A A N A A G \" C „ A - U G U C G - C A C C U - A A . . G G - C A V, G-U U G - U % U C - G A - U - C A A G U C - G U C - G U - A u C \" G u U \" A U A A \" f ; U A - U G G A C Figure IV-1: The previously proposed structures for 5.8S RNA. (a) from Rubin (54); (b) from Nazar et a l . (55). 4. Conserved Structural Regions and Functional Implications of the Cloverleaf The c lover leaf model possesses many conserved structura l features which may be involved in the proposed function of 5.8S RNA. Among these features are three i n te r i o r loops, a number of i n t e r i o r bulges, and three major arms and loops named for the i r probable functions or for the i r s tructural properties. The GC-rich arm, as mentioned e a r l i e r , i s l i k e l y involved in s t a b i l i z i n g the formation of the 28S RNA-5.8S RNA junction complex through co-axial he l ix stacking (57). This s t a b i l i t y i s provided by the large number of GC pairs in the arm. The anti-T loop contains the 76. ANTI-T LOOP UlACi 2o4gp6| ' 1 J 1 1 1 GC-RICH ARM R A 1 - K A U A - R I C H ARM Yeast 5.8S RNA Figure IV-2: The new proposed c lover leaf structure for yeast 5.8S RNA. The boxed-in areas are the conserved bases of the molecule. 77. , , G - C i n ' Urn _ g U A G G - C . \" A C U C - G , 40 A V . . . ^ / 2 0 A V G - C 1 4 0 ^ / A A ^ G U ( / C U G C G ^ G C U C G X ^ ? ? ? ? ? 9 9 U C C . I I I i i i i i i i \\ i i t i i i i i i o A R C G C . . . G C . G C G A G R G G C C C C G G G • • • • C G C A j — - A U U - G A - U A - U * - Ano G - C A G A C . U - A / U ~ G c a \" - ° \\ A - U A C A G - C ° ° C A U U - A AG - C U - A C - G A U c Novikof f a s c i te s hepatoma 5.8S RNA Figure IV-3: The c lover leaf structure adapted to the sequence of Novikoff hepatoma 5.8S RNA. 78. Table IV-1: Properties of Various Models for Eucaryotic 5.8S RNA Description of YEAST NOVIKOFF HEPATOMA Property Rubin* Nazar t Cloverleaf Nazar t Cloverleaf Base Composition 1*, 41 A, 43 U, 37 G, 36 C 2*, 33 U, 31 A, 46 AU Base Pairs 15 20 23 12 12 GC Base Pairs 16 21 23 33 32 Total AU, GC Pairs 31 41 46 45 44 GU Base Pairs 4 4 8 3 6 A* Base Pairs 1 1 1 2 1 Total Base Pairs 36 46 55 50 51 % Nucleotides in AU, GC Pairs 39% 52% 58% 57% 56% % U in Base Pairs 44% 56% 72% 55% 64% % Nucleotides in a l l types of pair ing 46% 58% 70% 63% 65% Sum of S t a b i l i t y Nols 16 16 22 28 21 *from Rubin (8) from Nazar et al (9) sequence GAAC in a l l known 5.8S RNA species and could function in binding tRNA to the ribosome, since tRNA has the complementary GT^C region in a l l species except the i n i t i a t o r tRNA's. Therefore, the GAAC sequence must be single-stranded when bound at the ribosome, but since there are no enzymic pa r t i a l digestion cleavage points in th i s region, i t must be protected in solut ion v ia some t e r t i a r y in teract ion. The UA-rich arm i s the least highly conserved and most weakly base paired region, espec ia l ly between bases 80 and 100. Thus, th i s region could assume d i f fe rent conformations in d i f fe rent species, and might provide for s p e c i f i c i t y in binding of 5.8S RNA to the ribosome. 79. In addit ion to the conserved major arms and loops, there are a number of conserved i n t e r i o r loops and bulges which could function as recognition s i tes or help to maintain t e r t i a r y structure. The presence of t e r t i a r y fo ld ing i s suggested by the large hyperchromicity of the thermal denaturation curve below 250 nm (14), the e f fect of Mg + + on the Raman spectrum, and the presence of s ingle stranded regions with resistance to enzymic cleavage, as well as the pH dependence of the Raman spectrum. Although the Raman data do not serve to ident i f y spec i f i c t e r t i a r y in teract ions , one interest ing p o s s i b i l i t y i s the interact ion of the GAAC region with the large 8-base bulge in the stem region ( i . e . AAGA paired with UCUU), although in Novikoff hepatoma species th i s pair ing i s not as extensive. Both of these regions are protected from hydrolys is , and the presence of the AA bulge in the anti-T arm (UA in Novikoff hepatoma 5.8S RNA;). could conceivably furnish the required bend in th i s arm to bring the two regions into close proximity. Furthermore, when the stem region becomes unpaired at the ribosome to allow pair ing of the 3'-end to the 28S RNA, the large bulge might be s u f f i c i e n t l y destab i l i zed to cause the associated GAAC region to unpair and become ava i lab le to bind tRNA. This sort of interact ion could account for the observation that ribosome-bound E. c o l i 5S RNA (having an analogous structure and interact ion) binds UUCG or T4CG more strongly than does free 5S RNA (75). F i n a l l y , the t e r t i a r y structure may also be s t ab i l i zed by interact ions involv ing other conserved bulges and i n t e r i o r loops; these include the loop of bases U-5 and C-151 (Fig. IV-2) which could be involved as a 28S RNA or ribosomal protein recognition s i t e , and the i n t e r i o r loop made by A-71, A-72 and C-106, G-107. B A S imi lar Structure 80. B. A S imi lar Structure For E. c o l i 5S RNA As pointed out e a r l i e r , procaryotic 5S RNA and eucaryotic 5.8S RNA have a number of features in common which suggest a s imi la r structure and function for the two molecules. Therefore, i f the c lover leaf model i s a useful secondary st ructure, one should be able to construct a secondary structure for E. c o l i 5S RNA which resembles the yeast 5.8S RNA structure, and which f i t s the physical constraints determined by experiments. Since E. c o l i 5S RNA has been the 5S RNA of choice for large numbers of physical studies (4.9, 76-78), and since recently we have received as yet unpublished Raman data for E. c o l i 5S RNA (44), th i s molecule w i l l now be used as a c r i t i c a l test for the v a l i d i t y of the proposed secondary structure of procaryotic 5S RNA and eucaryotic 5.8S RNA. 1. Physical Constraints on Procaryotic 5S RNA Extensive research has led to the fol lowing properties being assigned to native procaryotic 5S RNA: ( i ) X-ray scatter ing experiments at small angles at room temperature suggest that E. c o l i 5S RNA i s a prolate e l l i p s o i d with an o axia l r a t i o of 5/1 and a radius of gyration of 34.5±1.5 A (62). This high degree of asymmetry is supported by sedimentation studies, which also suggest a r i g i d structure for 5S RNA (49). ( i i ) Total hypochromicity experiments suggest that 62-65% of E. c o l i 5S RNA i s he l i ca l with about 38 base pairs (78). 0RD,CCD, UV, and infrared measurements claim 28±4 GC pairs and 13±4 AU pairs for E. c o l i 5S RNA (33). 13 ( i i i ) NMR studies using 5Tfluorouraci l and C-uraci l suggest that 75% or more of the U-residues in procaryotic 5S RNA are base paired (76,77). 8 1 . ( iv) Chemical modif ication of unpaired bases of four kinds gives information on regions of s ingle strandedness. Monoperphthalic acid reacts with ten unpaired adenines ( 7 9 ) , carbodiimide reacts with U^Q and one other U-residue (U-^, Ug^, UJJ, or U^Q^) ( 8 0 ) , glyoxal and kethoxal react with G ^ , G-^, and ( 4 8 ) , and methoxyamine reacts with Cgg and other C-residues in posit ions 3 4 - 4 1 and 4 4 - 5 1 ( 8 1 ) . Therefore, the areas around pos it ion G^, G ^ , and Cgg are s ingle stranded and exposed. (v) Enzymic pa r t i a l digestion studies show that the most susceptible area i s around G ^ ( 4 8 , 4 9 ) . (>hv).)Chen et a l . have obtained the HgO Raman spectrum of E. col i 5S RNA ( 4 4 ) . They found that E. c o l i 5S RNA i s s l i g h t l y less ordered than tRNA though i t i s s t i l l about 7 5 - 8 0 % ordered. Furthermore, Phe 5S RNA has much more G-stacking and less A-stacking than tRNA 2 . The Cloverleaf Model For E. c o l i 5 S RNA Although a number of structures have been proposed for E. c o l i 5S RNA ( 4 7 - 5 3 , 7 8 ) , none resembles the c lover leaf structure for E. c o l i 5S RNA (Fig. I V - 4 ) , which i s s t ruc tu ra l l y s imi la r to the c lover-leaf of yeast 5 . 8 S RNA. Not only does i t contain the \"procaryotic loop\" which i s analogous to the\" GC-rich loop in 5 . 8 S RNA, but i t also contains a s ingle stranded GAAC region to bind tRNA. Furthermore, the stem region contains a bulge with a conserved region complementary to the anti-T loop, while the anti^T arm contains the AA-bulge postulated to permit bending of th i s arm around to base pair to the bulge in the stem (bases 3 7 - 4 0 with 2 1 - 2 4 ) . E. c o l i 5S RNA also has a high s t a b i l i t y number (+25 as defined by Tinoco et a l . ( 1 5 ) ) . 82. The c lover leaf model also agrees very well with the physical constraints l i s t e d above. F i r s t , i t i s a r i g i d structure with potent ia l l y a large axia l ra t io through t e r t i a r y interact ions. Second, the secondary structure alone contains 37 base pa i r s , of which 22 are GC pairs and 10 are AU pa i r s , and by adding jus t the s ingle t e r t i a r y interact ion between the large bulge and the anti-T loop, one could eas i l y form two more GC pairs . Th i rd, the c lover leaf model has 75% (15 out of 20) of i t s U-residues base paired. Fourth, the structure has a s ingle stranded exposed region around U^Q and Cgg, while the points of pa r t i a l hydrolysis by Tj-ribonuclease again coincidewwith s ingle stranded or strained regions (Fig. IV-4). The c lover leaf secondary structure also f i t s the Raman data obtained by Chen et a l . (44). The procaryotic loop contains 4 consecutive he l i ca l G-residues which would give a large G-stacking contr ibut ion. Also,tthe regions of consecutive A-residues are often in bulges or loops, which could explain^the low A-stacking l e v e l . In a l l , 2/3 of the G-residues are base paired and 43% of A-residues are base paired, both in agreement with the values predicted by Chen et a l . (44). C. The Cloverleaf For Eucaryotic 5S RNA 1. Physical Constraints of 5S RNA Secondary Structure The present Raman results suggest the fol lowing physical properties for yeast 5S RNA: ( i ) a phosphodiester backbone structure more highly ordered than for tRNA, ind icat ing a high he l i ca l base paired content; ( i i ) a secondary and t e r t i a r y structure containing 65% base paired U-residues; ( i i i ) l imi ted A-stacking, ind icat ing that the two regions with 3 consecutive A-residues are probably not h e l i c a l l y arranged. Other physical measurements suggest the fol lowing 83. A G U C i i i I U p U - A G - C C - G C - G U - A G - C G - C C - G G - u IOG-C'JJ C - G A C - G (G)U G — \" \" U - A G°G Cgc-G U G G U - A C - G I O O 3 0R 4 P C A R 90 9 A ? G C G U A C C C C U •r R U C A G G U G R 6 0 A A u Aiir r r A C C G A A 50f A A ^ A A A C G - U A G so ^ C - G C - G G - U A U \" A G 5 C c Gc 7 0 \\ E. co l i 5 S - R N A Figure IV-4: The c lover leaf structure adapted to the sequence of E. c o l i 5S RNA. 84. addit ional propert ies: ( iv) exposed s ingle stranded regions near posit ions 40 and 90 (64); (v) a radius of gyration and axia l r a t i o of 34.5±1.5 A and 5/1 respect ively (the same as E. c o l i ) ; (v i ) thermal melting (for the analogous sea urchin 5S RNA) ind icat ing extensive overal l base pair ing (62-64%) but d i f fe rent t e r t i a r y fo ld ing than E. c o l i 5S RNA. 2. Incompatibi l i ty of Previous Structures With Raman Data Of the numerous secondary structures previously proposed for eucaryotic 5S RNA (49, 53), only one i s claimed to be adaptable to a l l eucaryotic and procaryotic 5S RNA species (Fig. IV-5a,b) (53). However, th i s model i s not consistent with the experimentally deter-mined physical properties l i s t e d above. For example, Table IV-2 shows that the Vigne-Jordan model has only 35% of i t s bases in paired he l i ca l conformations, in contrast to the more than 60% predicted from thermal melting p ro f i l e s (49). Low-field proton NMR analysis o r i g i n a l l y predicted 28±3 base pairs (30), but poor resolut ion and d i f f i c u l t i e s in integrat ing peak areas make th i s at best a lower l i m i t . In f a c t , the same authors predict only 4 AU pairs in E. c o l i 5S RNA, while the more recent studies suggest 15 or more (76,77). At any ra te , the 21 base pairs of the Vigne-Jordan model i s fa r below both experimentally derived predict ions, and would require an unreasonably large number of t e r t i a r y baseripairs to reach the experimental values. Moreover, the Vigne-Jordan structure i s incompatible with the present Raman resu l t s . Both^the phosphodiester backbone \"order\" and the percentage of base paired U-residues accord with the high percen-tage of base pairs predicted from thermal melting and low angle X-ray scatter ing data. The Vigne-Jordan model contains only 60% of the number of base paired U-residues needed for the Raman re su l t s , and 85. the small number of tota l base pairs cannot account for the observed high degree of \"order\" of the phosphodiester backbone. F i n a l l y , the r e l a t i v e l y small number of primary enzymic pa r t i a l hydrolysis cleavage points in many eucaryotic 5S RNA sequences (53, 64-66) suggests that most of the eucaryotic 5S RNA molecule i s double stranded, again in c o n f l i c t with the Vigne-Jordan structure. 3. The .Cloverleaf Structure For Yeast 5S RNA The Raman and other physical properties of yeast 5S RNA lead natura l ly to a c lover leaf secondary structure which i s s im i la r to those already presented, and which i s adaptable to many d i f fe rent 5S RNA sequences. (Fig. IV-6,IV-7). Other eucaryotic 5S RNA sequences not included also conform to the same structure, making i t a universal structure for a l l small RNA molecules, of which tRNA was the f i r s t example (21). Table IV-2 shows that the c lover leaf model i s the f i r s t to f i t both the Raman and other physical properties: i t possesses a large tota l percentage of paired bases (57%), the correct percentage of base paired U-residues (64%), and a high s t a b i l i t y number (+12). In add i t ion, the new c lover leaf contains s ingle stranded regions near posit ions 40 and 90 as predicted by par t i a l hydrolysis (Fig. IV-6), and forces the two regions of 3 consecutive A-residues into non-he l i ca l segments as predicted by the 725 cm - ' ' ' Raman i n ten s i t i e s . The addit ion of ju s t a few t e r t i a r y pairs would increase the percen-tages of both to ta l base pairs and base paired U-residues to the levels consistent with the remaining physical measurements. Functionally, less is known about eucaryotic 5S RNA than e i ther eucaryotic 5.8S RNA or procaryotic 5S RNA because of two factors : 86. (a)E.coli 5SRNA c f c C A u p U G C C U G G C G G C C G U A G C G C G G U G G U C C C A C C U G A G H O A C G G A C C G U C A . . G C C G C A A G A C U C A „ C H U ' A A U A A A A A ^ A A G C G G R A G y G R G C A G G A „ % A G A ^ A : U U A C - G C - G C - G u c u (b) Yeast 5S RNA p p p G G U U G C G G C C A U A U C U A C C A G A A A G C A C C G U U C U C C < r j i p i i > i i i i i i i i i i i i ' H Q U C U A A C G U C G y G A U G G U C & , C C A A U c G G . c G A A , \" - U \" A G G A A A C C U ^ U C U U G A U r G A c c A C C U G I A ' U G A U G U ' G U G G G I -Figure IV-5: A proposed structure of yeast and E. co l i 5S RNA. (a) E. c o l i 5S RNA, from Fox and Woese (50); (b) S. cerevis iae 5S RNA, from Vigne and Jordan (53). u PppG-C1 2 0 G-U U-A U-A G-C C-G G-U 1 £ C C - G A i jA -U C-G U-A A C C U ANTI-INITIATOR 2oA A tRNA LOOP Q ^ ^ ^ U U G g c A C G A A ' ' A G g f e C A U A C C A -^ G A U C A A C U W ^ A C ^ U G G ^ 4 0 A G U T. C-G U-A G 60G-C G-C U-A A G A L^ o A G C S. cerevisiae (+12) A A Figure IV-6: The new proposed c lover leaf structure for yeast 5S RNA. 88. Table IV-2: Comparison of Properties of Models of 5S RNA With Experimental Evidence Property Experiment\"'' Model k Vigne and Jordan T Cloverleaf GC pairs -- 11 16 AU pairs — 8 13 GU pairs — 2 5 Total Pairs >35 21 34 % of Bases paired >60% 35% 57% % of U-bases paired -65% 36% 64% S t a b i l i t y number -- +9 +12 * (Connors and Beeman, 1972; Erdmann, 1976) * (Vigne and Jordan, 1977) eucaryotic P iBbsomesox^ (49), and eucaryotic 5S RNA i s d i s s im i l a r enough from procaryotic 5S RNA not to be incorporated in procaryotic ribosome assembly systems (59). However, . a l l sequences determined so far (except the least evolved Chlore l la ) contain the sequence CYGAU complementary to eucaryotic i n i t i a t o r tRNA T-loops (58). Therefore, i t i s tempting to consider that eucaryotic 5S RNA functions s p e c i f i c a l l y to bind i n i t i a t o r tRNA to the ribosome in a manner analogous to 5.8S RNA binding non - in i t i a to r tRNA's. I t i s sa t i s fy ing to note that th i s CYGAU sequence in the eucaryotic 5S RNA c lover leaf occupies a pos it ion analogous to that of the procaryotic 5S RNA and eucaryotic 5.8S RNA GAAC region which bi nds non-i niitiatorMRNA. The present c lover leaf structure for eucaryotic 5S RNA also contains a number of other features in common with the above proposed c lover leafs for procaryotic 5S RNA and eucaryotic 5.8S RNA. These 89. include the large bulge in the stem region, which may part ic ipate in t e r t i a r y interact ions or as a'ribosomal recognition s i t e , and also the arm opposite the an t i ( i n i t i a to r )T - l oop which may s t a b i l i z e the interact ion of the 3'-end of 5S RNA with the large rRNA species as in the case of 5.8S RNA (57). D. Comments on Tert iary Structures of 5S RNA and 5.8S RNA The evidence to date suggests s imi la r t e r t i a r y structures for procaryotic 5S RNA and eucaryotic 5.8S RNA, while a d i f fe rent t e r t i a r y structure i s predicted for eucaryotic 5S RNA. The most compelling evidence in favor of th i s idea i s the fact that E. c o l i ribosomal proteins which bind E. c o l i 5S RNA w i l l also bind yeast 5.8S RNA but not yeast 5S RNA. Also, the present Raman results suggest a s l i g h t l y d i f fe rent order for yeast 5.8S RNA and 5S RNA, while opt ica l measure-ments indicate a d i f fe rent shape for sea urchin 5S RNA and E. c o l i 5S RNA (63). Furthermore the present chromatographic data, and the fact that low Mg + + samples of yeast 5S RNA could not be prepared, indicate that 5.8S RNA i s more asymmetric than 5S RNA and has less 'tendency to aggregate and prec ip i tate out of so lut ion. It i s > interest ing to note that denatured E. c o l i 5S RNA also tends to aggregate (22). As mentioned above, one t e r t i a r y interact ion in 5.8S RNA could be between the bulge in the stem and the anti-T loop. However, in yeast 5S RNA th i s interact ion i s not possible. An alternate pair ing in eucaryotic 5S RNA could be between bases 33-36 in the a n t i ( i n i t i a t o r ) -T-loop and bases 64-67. This would be a stronger interact ion than in 5.8S RNA and could explain the greater tendency to aggregate (dimerize?) when the t e r t i a r y structure i s destroyed by removing Mg + +, since complementary sequences on d i f fe rent molecules would be more 90. l i k e l y to pa i r up. Furthermore, th i s complementary pair ing would serve to protect the loop containing bases 64-72 from pa r t i a l hydrolys is, as seen experimentally, and could produce the large ax ia l r a t i o deduced from low-angle X-ray scatter ing (62). Thus i t i s postulated that a l l small RNA molecules (tRNA, 5S RNA, 5.8S RNA) have the same basic c lover leaf secondary structure, and that various t e r t i a r y structural features provide some of the d i f fe rent physical properties observed for these molecDles. Although the t e r t i a r y structure of one tRNA (tRNA (9,10)) has been determined, the same i s not the case for 5S RNA and 5.8S RNA, and further work w i l l be necessary in the area. E. Comments on Evolution in 5S RNA and 5.8S RNA Although 5.8S RNA sequences are very highly conserved (58), eucar-yo t i c 5S RNA species have substantial sequence variat ions along the evolutionary path from lower to higher animals (58). However, the higher animals such as X. l aev i s , HeLa c e l l , KB c e l l , and chicken 5S RNA species d i f f e r in only a few bases, suggesting that a more stable structure has been reached in higher animals. Several such eucaryotic 5S RNA evolutionary trends become evident when c lover leaf structures are drawn for several b io log ica l species (Fig. IV-7): ( i ) the s t a b i l i t y number (as defined by Tinoco et a l . (15)) increases in going from lower to higher animals and then s t ab i l i z e s at +18, ind icat ing attainment of a very stable structure; ( i i ) the anti-T arm, which i s the funct iona l ly s i gn i f i can t arm, remains r e l a t i v e l y unchanged (constant s t a b i l i t y and number of base pa i r s ) , confirming the proposed importance of th i s region in binding i n i t i a t o r tRNA; ( i i i ) both of the other arms tend to increase in number of base pairs and s t a b i l i t y at the expense of base pairs in the stem region. The 91. pppA-U U-G _ G - U C - G U - A A - U C - G _ V A C u S -A ^ C ^ G - C A C * ~ A A „ A C C C u A G A A < U A G C C - U C G G , ^ G c C A C c A A G _ AA l 'Y/>A G ^ G ^ y c c A U ^ i ACUGGGUUGgA A - U * r C - G C U - A G - U \" -G - C _ U \" A 3 J - G G -C G U - • G C \\ CtVorttliMO) « . G C U c U U G G C % ^ k U 6 \" ~ TA-u c P P P G - C C - G C - G A - U A - U C - G G - U A C C A - U U A C C - G A - U C - G G - C U - A G - C 1 A A / M A C A U A A G9<^OCGCC A 5 A U U.A A C U U A C \" G - U C \" G A A - U G - U C - G E G - C U - G C - G G C G ^UC^gC -DrosophiU ('15) u u u pppG-C C - G C - G U-A A - U C - G G -C G -U C A C C - G A c c A \" U C - G r-C-G ' U - A A A A G \" C , U C G / C U r AGu G C A U A J * G G C M : C G C C A G c.. .cUCG6AAdc A ^ s \\ U G A U 6 A A G C c /«Vk-U A - U G G C - G A - U G - C G - C G G U G X. l«WBf»18> U u u PppG-C U - G C - G U - A A - U C - G G - U G - C I J A C C - G A r C A - U u C - G V C - G Z G ^ - G * ^ G C U C U A G C C £ G C CAUAA6GGMcCGC9Ae/ ° G A A . U \\ A - U G C - G G A-U G - C G-C G G H»L»(*HH1B) (KB e»lt) Figure IV-7: The c lover leaf structure adapted to other eucaryotic 5S RNA species to indicate the evolutionary trends of the molecule. 92. most stable structure (HeLa c e l l ) has achieved three independently stable arms and a stable stem region, in reaching a structure very s imi la r to the more highly conserved procaryotic 5S RNA and eucaryotic 5.8S RNA. In eucaryotic 5.8S RNA, the highly conserved nature of the sequence suggests that^even in lower eucaryotes such as yeast, a stable structure has been achieved. The yeast 5.8S RNA structure strongly resembles the E. c o l i 5S RNA structure, with an extra series of bases in the UA-rich arm. This RNA follows the same pattern as the other ribosomal RNA's; 16S RNA in procaryotes i s equivalent to 18S RNA in eucaryotes, 23S RNA in procaryotes i s 28S RNA in eucaryotes. Therefore, making two basic postulates: i . e . the functional arm of the molecule must not change, and the 3'-end and 5'-end must pair to form the stem, a l l other aspects of the evolution of 5S RNA and 5.8S RNA can be understood in terms of three mechanisms. The f i r s t of these i s used by a l l RNA and involves increasing the number of GC pairs since they provide 2.4 kcal each while AU pairs provide only 1.2 kcal (15). The second i s used by eucaryotic 5S RNA and involves s t a b i l i z i n g the two loops not involved in tRNA binding by breaking base pairs from the stem region, rotat ing the RNA strand into these arms, and creating more base pairs in the arms to s t a b i l i z e them. This type of a l te rat ion creates the bulge in the stem which i s charac te r i s t i c of a l l 5S RNA and 5.8S RNA species. The th i rd mechanism i s seen in the transformation from procaryotic 5S RNA to eucaryotic 5.8S RNA. In th i s case the presence of the very stable GC-rich arm \" locks \" in the tota l RNA configuration, so rotat ion of one strand cannot occur. Therefore, the only way the molecule can change i s by adding a piece of RNA to the UA-rich arm as occurs in 5.8S RNA 93. r e l a t i ve to 5S RNA. Also, th i s region i s the least highly conserved region in 5.8S RNA species (58), suggesting that i t i s the region where the molecule may be evolving to adapt for a.changing structure or function. 94. CHAPTER V Future Considerations When the present work was begun, the broad scope of the project at hand had not been remotely ant ic ipated. It was true (and s t i l l i s ) that the structural determination of small RNA molecules was a r e l a t i v e l y new and rapidly expanding f i e l d , being pioneered by Holley et a l . in 1965 with the determination of the c lover leaf secondary structure from the newly sequenced tRNA^ a , and being continued for Phe tRNA by the determination of the crystal structure by two groups at M.I.T. and Cambridge. NMR has also been used extensively for the study of H-bonding in many tRNA species, and now the methyl resonances of modified bases also appear to provide useful s tructural information for tRNA molecules. By combining the above techniques with many of the more t rad i t i ona l chemical and opt ica l techniques, Phe the structure of tRNA in solut ion has been precise ly determined and v e r i f i e d , and present research i s aimed at determining how the structure i s affected by interact ions with various constituents of the protein synthetic apparatus, and whether the changes in structure are related to function. For the other small RNA components of the protein synthetic apparatus (5S RNA, 5.8S RNA) no obvious structure was evident from the primary sequence as for tRNA, yet a spec i f i c conformation must be present for these molecules to be highly conserved and funct iona l ly s im i la r in widely varying species. Also, the function of these molecules was not eas i l y determined because they are int imately associated with the ribosome, and the i r functions could not be studied unt i l the ribosome could be reconstituted in the late 1960 1s. The resu l t was wi ld speculation, and numerous structures were published on a minimal amount of experimental evidence. Many of these structures were 95. derived so le ly by computer pair ing of bases without consideration of functional requirements. Also, some of the more powerful physical techniques such as Raman spectroscopy and NMR spectroscopy were only beginning to be used as structural determinants for RNA molecules. Since none of the models were obviously correct or incorrect , the present study was undertaken to see i f physical techniques such as Raman spectroscopy or NMR spectroscopy could be used to solve the mystery and end the confusion by determining which of the structures was correct. As was hopefully expressed in the body of th i s thes i s , the s ingle technique of Raman spectroscopy provided the t o t a l l y unexpected key to a new secondary structure for 5S RNA and 5.8S RNA as they are found free in so lut ion. Probably the four most encouraging aspects of the new model are that: ( i ) i t has a very strong resemblance to the f ami l i a r c lover leaf which i s known to be the preferred conformation for tRNA; ( i i ) i t encompasses the functional requirements of 5S RNA and 5.8S RNA in the same way as tRNA by having interact ing sequences placed in hairpin loops on the ends of stable he l i ca l arms exactly as are the anticodon and T*CG regions of tRNA; ( i i i ) the structure i s conserved when adapted to widely diverse species; and ( iv) the structure s a t i s f i e s almost a l l of the previously determined structural properties of 5S RNA and 5.8S RNA. The s i tuat ion at present i s that the c lover leaf secondary structure determined by th i s work provides a handle on the structure of the molecule, much l i k e the secondary structure of tRNA provided the s tar t for the tota l determination of tRNA structure. Also, the crysta l structures and NMR spectra of 5S RNA and 5.8S RNA are not ava i lab le. Therefore, the state of research into 5S RNA and 5.8S RNA 96. is in a pos it ion analogous to 1967 for tRNA research. The above analogy allows one to eas i l y discern the future course of research into the structure and function of 5S RNA and 5.8S RNA, since the same types of experiments which proved useful for tRNA can also be performed on 5S RNA and 5.8S RNA. The crysta l structure would cer ta in l y be the ultimate method for pinpointing st ructure, yet th i s may not be feas ib le or even desir^able due to the d i f f i c u l t y in c r y s t a l l i z i n g RNA molecules and the fact that the structure of 5S RNA and 5.8S RNA i so lated from solution may not be ind icat ive of the structure when these molecules are i n t r i c a t e l y woven into the ribosomal network. In f a c t , recent evidence suggests that a t e r t i a r y s t ructura l change involving an increase in base stacking accompanies binding of ribosomal proteins to E,. c o l i 5S RNA. Therefore, the s t a t i c technique of crystallography may be less f r u i t f u l for 5S RNA and 5.8S RNA than i t was for tRNA. However, interest has been expressed in attempting to c r y s t a l l i z e our 5S RNA and 5.8S RNA. NMR, on the other hand, because of i t s dynamic nature, can be used to measure the pattern of base pairs in so lut ion, where ribosomal proteins can be bound to the RNA. Thus, as for tRNA, conformational changes associated with Mg + + or protein binding can be studied. The large number of base pairs has in the past provided d i f f i c u l t y in resolut ion of low f i e l d proton spectra, but these spectra were recorded at 220 and 300 MHz, and more powerful magnets w i l l provide s u f f i c i en t resolut ion for future studies. These studies are presently being planned. Optical and chemical methods should also supply some useful information on the structure of eucaryotic 5S RNA and 5.8S RNA. The r e l i a b i l i t y of chemical modification and opt ica l studies have been 97. Phe proven for tRNA , and have already been employed extensively on procaryotic 5S RNA, yet eucaryotic 5S RNA and 5.8S RNA remain r e l a t i v e l y untouched in these areas. Possible experiments therefore include determining the number and liypes of base pa i r s , and the chemically-modifiable bases of these molecules. The results could then be compared with results on protein bound 5S RNA and 5.8S RNA to determine where the proteins are bound and what the i r e f fect on structure i s . The opt ica l techniques include UV, CD, IR, and Raman spectroscopy. These studies are also being planned for the near future. F i n a l l y , the introduction of nuclear spin labels into spec i f i c posit ions in RNA promises to provide useful s t ructura l information. The two of these which have been used in the past (5- f luorourac i l and 13 C-urac i l ) have been shown to be incorporated into RNA molecules 19 13 with a minimal ef fect on structure. Furthermore, the F- or C-NMR spectrum;contains a number of resolved peaks character i s t i c of the spacial pos it ion of spin l abe l s , po tent ia l l y allowing structural determination from chemical s h i f t . In our lab, 5- f luorourac i l has been incorporated into E. c o l i 5S RNA, and th i s may provide a p o s s i b i l i t y for future research into 5S RNA and 5.8S RNA structure. 98. GLOSSARY OF TERMS AND ABBREVIATIONS A: adenine AA-tRNA: aminoacyl transfer RNA A - s i t e : The s i te on the ribosome where the incoming aminoacylated tRNA attaches. anticodon: The series of three nucleotides in transfer RNA which i s complementary to the codon of messenger RNA. C^ : cytosine codon: The series of three nucleotides of messenger RNA which specify the amino acid to be added next in the protein chain. DHU: One of the three loops of transfer RNA named a f te r i t s character-i s t i c invariant modified nucleotide, 5,6-dihydrouraci l . EF-T u > EF-G, EF^-T^: Three proteins involved in elongation during protein synthesis (elongation factor s ) . fMet-tRNA^: The spec i f i c transfer RNA which has been aminoacylated with formylmethionine. (5: guanine GTP: guanosine triphosphate hypochromicity: The decrease in absorbance between 230 and 300 nm when nucleotide base become stacked (antonym: inverse hypochromism). IF-1, IF-2, IF-3: Three proteins involved in the i n i t i a t i o n of protein synthesis ( i n i t i a t i o n factor s ) . mRNA: messenger RNA procaryote: Organism which does not have a nuclear \"membrane (antonym: eucaryote). P-s i te: The s i t e on the ribosome where the growing protein bound to transfer RNA resides. •99. RF—1, RF-2: Protein involved in termination of protein synthesis (release fac to r ) . ribosome: The ce l l organelle composed of protein and RNA which functions as the s i t e for protein synthesis. RNase: A protein enzyme which cleaves RNA molecules. S_: The unit of sedimentation rate (Svedgerg unit) d i r e c t l y related to molecular weight. TtyEG: One of three loops of transfer RNA named af ter the invar iant base (pseudouracil). t ransc r ip t ion : The process by which the message contained in DNA i s transformed into messenger RNA. t rans la t i on : The process by which the coded message determined by the sequence of bases in messenger RNA i s transformed into a functional protein. tRNA: transfer RNA. Phe tRNA : Transfer RNA spec i f i c for binding phenylalanine. Ik u r a c i l . 100. REFERENCES 1. P. Lengyel in \"The Ribosome\", M. Nomura, A. T i s s ie re s , and P. Lengyel, eds., Cold Spring Harbour Press, New York, 1974. 2. R. Brimacombe, K.H. Nierhaus, R.A. Garrett, and H.G. Wittmann, Prog. Nucleic Acid Res., 18, 1 (1976). 3. C G . Kurland, Ann. Rev. Biochem., 46, 173 (1977). 4. M. Nomura, A. T i s s ie res , and P. Lengyel, \"The Ribosome\", Cold Spring Harbour Press, New York, 1974. 5. R. Potts, M.J. Fournier, and N.C. rFord, j r . , Nature, 268, 563 (1977). 6. U. Schwartz, H.M. Menzel, and H.G. Gassen, Biochemistry, _15, 2484 (1976). 7. U. Schwartz and H.G. Gassen, F.E.B.S. Letters , 78, 267 (1977). 8. J.W. Fox and K.P. Wong, J . B i o l . Chem. 253, 18 (1978). 9. S.H. Kim, Prog. Nucleic Acid.Res., 17, 182 (1976).' 10. J.D. Robertus, J.E. Ladner, J.T. Finch, D. Rhodes, R.D. Brown, B.F.C. Clark, and A Klug, Nature,1250, 546 (1974). 11. V.A..-;Bloomfield, D.M. Crothers, and I. Tinoco, j r . , Physical Chemistry of Nucleic Acids, Harper and Row, New York, 1974. 12. A. Rich and S.H. Kim, S c i e n t i f i c American, 238, 52 (1978). 13. N.T. Van, J.W. Holder, S.L.C. Woo, A.R. Means, and B.W. O'Malley, Biochemistry, 15, 2054 (1976). 14. N.T. Van, R,N3«Nazar, and T.0. S i t z , Biochemistry, 16, 3754 (1977). 15. I. Tinoco, j r . , O.C. Uhlenbeck, and M.D. Levine, Nature, 230, 362 (1971). 16. A.R. Cashmore, D.M. Brown, and J.D. Smith, J . Moi. B i o l . , 59_, 359 (1971). 17. F. Cramer, H. Doepner, F. von der Haar, E. Schlimme, and H. Se ide l , Proc. Nat l . Acad. S c i . , 61, 1384 (1968). 101 . 18. M. L i t t , Biochemistry, 10, 2223 (1971). 19. J.D. Robertus, J.E. Ladner, J.T. Finch, D. Rhodes, R.S. Brown, B.F.C. Clark, and A. Klug, Nucleic Acids Res., 1, 928 (1974). 20. G. von Ehrenstein in \"Aspects of Protein Biosynthesis\", C.B. Anf ins in , j r . , ed., Academic Press, New York, 1970. 21. R.W. Holley, J . Apgar, G.A. Everett, J.T. Madison, M. Marquisee, S.H. M e r r i l l , J.R. Penswick, and A Zamir, Science, 147, 1462 (1965). 22. J.B. Lewis and P.Doty, Biochemistry, 16, 5016 (1977). 23. J.B. Lewis and P. Doty, Nature (London), 225, 510 (1970). 24. 0. Pongs, R. Bald, E. Reinwald, Eur. J . Biochem., 32, 117 (1973). 25. B.R. Reid, Accounts Chem. Res., 10, 396 (1977). 26. J.L. Sussman and S.H. Kim, Science, 192, 853 (1976). 27. R: Giege, D. Moras, and J.C. Thierry, J . Moi. B i o l . , 115, 91 (1977). 28. L.S. Kan, P.O.P. Ts '0 , M. Sp r i n z l , F. von der Haar, and F. Cramer, Biochemistry, 16, 3143 (1977). 29. D.R. Kearns, Prog. Nucleic Acid Res., 18, 91 (1976). 30. Y.P. Wong and D.R. Kearns, J . Moi. B i o l . , 72, 741 (1972). 31. D.R. Kearns and Y.P. Wong, J . Moi. B i o l . , 87, 755 (1974). 32. A.G. Marshal l , Biophysical Chemistry, Wiley and Sons, New York, 1978. 33. E.G. Richards, M.E. Geroch, H. Simpkins, and R. Lecanidou, Biopolymers, 1 1 , 1031 (1972). 34. R. Herbeck and G. Zundel, Biochim. Biophys. Acta, 418, 52 (1976). 35. W.L. Peticolas in \"Advances in Raman Spectroscopy\", J.P. Mathieu, ed., Heyden and Son, L td . , New York, 1972. 36. G.J. Thomas in \"V ibrat ional Spectra and Structure\" , J.R. Durig, ed., Elesevier Publishing Co., New York, 1975. 37. G.J. Thomas and K.A. Hartman, Biochim. Biophys. Acta, 312, 311 (1973). 38. M.C. Chen and G.J. Thomas, Biopolymers, 13, 615 (1974). 102.. 39. M.C. Chen, R. Giege, R.C. Lord, and A. Rich, Biochemistry 14, 4385 (1975). 40. G.J. Thomas, M.C. Chen, and K.A. Hartman, Biochim. Biophys. Acta, 324, 37 (1973). 41. K.A. Hartman, R.C. Lord, and G.J. Thomas, j r . , in \"Physico-chemical Properties of Nucleic Ac ids \" , J . Duchesne, ed., Academic Press, New York, 1973. 42. R.CC Lord and G.J. Thomas, j r . , Spectrochimica Acta, 23A, 2551 (1967). 43. T. O'Connor, C Johnson, and W.M. Scove l l , Biochim. Biophys. Acta, 447, 484 (1976). 44. M.C. Chen, R. Giege, R.C. Lord, and A.Rich, Biochemistry, in press. 45. C.N. Banwell, Fundamentals of Molecular Spectroscopy, McGraw-Hill Book Co., London, 1972. 46. J . J . Pene, E. Knight, j r . , and J.E. Darnel l , j r . , J . Moi. B i o l . , 33, 609 (1968). 47. G.G. Brownlee, F. Sanger, and B.G. B a r r e l l , Nature (London), 215, 735 (1967). 48. R. Monier in \"The Ribosome\", M. Nomura, A. T i s s ie re s , and P. Lengyel, eds., Cold Spring Harbour Press, New York, 1974. 49. V.A. Erdmann, Prog. Nucleic Acid Res., 18, 45 (1976). 50. G.E. Fox and C.R. Woese, Nature, 256, 505 (1975). 51. R. Osterberg, B. Sjoberg, and R.A. Garrett, Eur. J . Biochem., 68, 481 (1976). 52. B. Dubuy and S.H. Weissman, J . Biol Chem., 246, 747 (1971). 53. R. Vigne and B.R. Jordan, J . Moi. Evo l . , 10, 77 (1977). 54. G.M. Rubin, J . B i o l . Chem., 248, 3860 (1973). 55. R.N. Nazar, T.0. S i t z , and H. Busch, J . Biol.JChem., 250, 8591 (1975). 103. 56. J.T. Madison, Ann. Rev. Biochem., 37, 131 (1968). 57. N.R. Pace, T.A. Walker, and E. Schroeder, Biochemistry, 16_, 5321 (1977). 58. V.A. Erdmann, Nucleic Acids Res., 5_, r l (1978). 59. P. Wrede and V.A. Erdmann, Proc. Nat l . Acad. S c i . , 74, 2706 (1977). 60. E. Lund, J.E. Dahlberg, L. L indahl, S.R. Jaskunas, P.P. Dennis, and M. Nomura, C e l l , ]_, 165 (1976). 61. R.P. Perry, Ann. Rev. Biochem., 45, 605 (1976). 62. P.G. Connors and W.W. Beeman, J . Moi. B i o l . , 71, 31 (1972). 63. / G. Bellemare, R.J. Cedergren, and G.H. Cousineau, J . Moi. B i o l . , 68, 445 (1972). 64. R. Vigne, B.R. Jordan, and R. Monier, J . Moi. B i o l . , 76, 303 (1973). 65. M.WWegnez, R. Monier, H. Denis, F.E.B.S. Letters , 25, 13 (1972). 66. J . Benhamou, R. Jourdan, B.R. Jordan, J . Moi. Evo l . , 9_, 279 (1977). 67. J.E. Dahlberg, C. Kintner, and E.Lund, Proc. Nat l . Acad. S c i . , 75, 1071 (1978). 68. T. Schleich and J . Goldstein, J . Moi. B i o l . , 15, 136 (1966). 69. R.W. Holley, J . Apgar, B.P. Doctor, J . Farrow, M.A. Mar in i , S.H. M e r r i l l , J . B*ol. Chem., 236, 200 (1961). 70. G.J. Quigley, M.M. Teeter, and A. Rich, Proc. Nat l . Acad. S c i . , 75, 64 (1978). 71. A. Jack, J.E. Ladner, D. Rhodes, R.S. Brown, and A. Klug, J . Moi. B i o l . , I l l , 315 (1977). 72. R.N. Nazar, T.O. S i t z , and H. Busch, Biochemistry, 15, 505 (1976). 73. J . Woledge, M.J. Corry, and P.I. Payne, Biochim. Biophys. Acta, 349, 339 (1974). 74. R.N. Nazar and K.L. Roy, J . B i o l . Chem., 253, 395 (1978). 75. V.A. Erdmann, M. Sp r i n z l , and 0. Pongs, Biochem. Biophys. Res. 1 0 4 , Commun., 54, 942 (1973). 76. A.G. Marshall and J.L. Smith, J . Amer. Chem. S o c , 99, 635 (1977). 77. W.D. H a m i l l , j r . , D.M. Grant, R.B. Cooper, S.A. Harmon, J . Amer. Chem. Soc., 100, 633 (1978). 78. H. Boedtker and D.G. Ke l l i n g , Biochem. Biophys. Res. Commun., 2_9, 758 (1967). 79. F. Cramer and V.A. Erdmann, Nature (London), 218, 92 (1968). 80. J.C. LEE and V.M. Ingram, J . Moi. B i o l . , 41, 431 (1969). 81. G. Bellemare, B.R. Jordan, J . Rocca-Serra, and R. Monier, Biochimie, 54, 1453 (1972). "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0060788"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemistry"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Laser Raman evidence for new universal cloverleaf structures for 5.8s RNA and 5s RNA"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/20834"@en .