ABSTRACT STUDIES ON THE INFLUENCE OF SECONDARY STRUCTURE AND 2'jQ-METHYLATION ON THE TRANSLATION OF RIBOSOMAL RNA By Brian Eric Dunlap Two factors which might influence the translation of ribosomal RNA have been examined. The first of these is secondary structure and the second 2'fQ—methylation. The effect of secondary structure was examined by using g. subtilis ribosomal RNA, which starts to thermally denature at L60 and has a Tm of about 70°, as template RNA in a thermophilic amino acid incorporating system. The incorporating system was pre- pared from g. stearothermophilus and was found to retain its incorporating activity to 70°. Controls of g. stearothermo- philus and g. subtilis whole cell RNA, and poly uridylic acid, were active as templates in the g. stearothermophilus incor- porating extract at all temperatures tested (MO-650). A fourth control, the RNA from the virus R17, had a thermal denaturation profile similar to g. subtilis ribosomal RNA, and had template activity which increased as the temperature of incorporation reaction increased. .§° subtilis ribosomal RNA, however, had no template activity at any temperature tested (MO-65° C) or Brian Eric Dunlap any Mg++ concentration tested (0-12 mM). If neomycin was included in the incorporation reaction mixture, the ribosomal RNA had template activity, which was found to increase as the temperature of the incorporation reaction increased. These results indicate that secondary structure is not the primary inhibiting factor in preventing ribosomal RNA translation, but may influence translation once the primary inhibiting factor is removed. The effect of 2'1Q-methylation on translation was investi- gated by preparing synthetic polynucleotides containing 2'j9- methylnucleosides, and testing them for template activity in an E. ggli cell free amino acid incorporating system. The heteropolymer, poly(Cm,U), directed the incorporation of sig- nificant levels of phenylalanine, serine, leucine, and proline, and small amounts of isoleucine and tyrosine. The total in- corporation of amino acids was slightly greater with poly(Cm,U) than with poly(C,U). The heteropolymer poly(Am,C) directed the incorporation of proline, threonine, and histidine, but its template activity was lower than that of poly(A,C). Poly(Cm,U) was active as a template for a longer period of time than poly(C,U) in directing the incorporation of phenyl- alanine. Both poly(Am,C) and -(Cm,U) were degraded more slowly than their unmethylated analogs when incubated in re- action mixtures used for cell-free protein synthesis. These Brian Eric Dunlap results indicate that the level of 2'fiQ-methylnucleosides (O.1-l.O%) found in ribosomal RNA would not be high enough to eliminate its template activity. STUDIES ON THE INFLUENCE OF SECONDARY STRUCTURE AND 2'197METHYLATION ON THE TRANSLATION OF RIBOSOMAL RNA BY Brian Eric Dunlap A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1971 DEDICATED to My Family ii ACKNOWLEDGMENTS The author wishes to thank Dr. Fritz Rottman for his help in this project. Thanks also go to Dr. J. A. Boezi, Dr. A. J. Morris, Dr. S. D. Aust and Dr. H. L. Sadoff for serving as members of my guidance committee. Appreciation is also extended to my co-workers, Lee Pike, Joseph Abbate, Karen Friderici, Galvin Swift and Diana Filner for helpful discussions and frequent assistance. And, of course, behind every successful (male married) Ph.D. there's a helpful, enduring wife who is this case was also a proficient typist. TABLE OF CONTENTS LIST OF TABLES . . . . . . . . LIST OF FIGURES . . . . . . . GENERAL INTRODUCTION . . . PART I: HISTORICAL . . . . . Materials and Methods RNA extractions (a) Whole cell RNA . (b) Ribosomal RNA . Measurement of melting profile of RNA . Preparation of amino acid incorporating extracts . o o Assay for template activity of RNA . RESULTS . . . . . . . . DISCUSSION . . . . REFERENCES . . . . . . . PART II . . . . . . . . . . . . . Page vi 19 20 22 25 27 3o 50 61 65 LIST OF TABLES Table _ Page 1 Incorporation of amino acids by whole cell RNA from E. coli and B. stearothermophilus 5h 2 Temperature dependence of incorporation of 140 leucine directed by B. subtilis ribosomal RNA 0 O O O O O O O O O O I C O 0 us 5 Incorporation of 140 leucine directed by B, subtilis ribosomal RNA at different Mg concentration . . . . . . . . . . . . . . . M6 h Effect of B. subtilis ribosomal RNA on incorporation of I4C leucine directed by B. subtilis whole cell RNA . . . . . . . #8 Figure LIST OF FIGURES Page Proposed secondary structure of a fragment of E. coli ribosomal RNA . . . . . . . . . 12 Thermal denaturation profile of B. subtilis ribosomal RNA and R17 RNA . . . . . . . . 52 Cell free incorporation of amino acids directed by B. stearothermophilus or ‘B. subtilis whole cell RNA . . . . . . . 57 Temperature dependence of cell free incorporation of amino acids directed by polyuridylic acid or the whole cell RNA from g. subtilis or g. stearother- mophilus . . . . . . . . . . . . . . . . #0 Temperature dependence of cell free incorporation of amino acids directed by B. subtilis rRNA in the presence or absence of neomycin, and R17 RNA . . . . L2 INTRODUCTION The question of whether or not ribosomal ribonucleic acid (rRNA) can serve as a template for in_!i££2 protein biosynthesis has not yet been conclusively answered. When early studies showed that proteins were synthesized on ribosomes, it was hypothesized that ribosomal RNA was the template. The demonstration that rRNA has a very slow turnover rate, and that the two types of rRNA are too homogeneous to code for the wide variety of proteins found in gixg, soon caused this hypothesis to be abandoned (Watson, 1965). More recently (Osawa, 1965) it has been suggested that rRNA may serve as a template only for ribosomal proteins, a hypothesis based on the following three findings; (1) during recovery of methionine starved E, coli Rcrel cells or in g, 221; cells previously treated with chloramphenicol (CM), ribosomal proteins are preferentially synthesized. During the recovery phase, accumulated precursor rRNA is converted to mature rRNA, even though further RNA synthesis is inhibited with actinomycin D (Nakada, 1965). This was interpreted as indicating that the accumulated precursor rRNA acted as a template for the ribosomal proteins. Mfldgley and Gray (1971), however, have shown that appreciable messenger RNA (mRNA) also accumulates during inhibition with GM, 1 and thus the synthesis of proteins which takes place during recovery may just be the expression of this accumulated mRNA; (2) most of the accumulated nascent rRNA appears to exist combined with ribosomes, the cellular site of protein synthesis (Mnto, gtwgl., 1966); (3) nascent ribosomal particles accumulate in g, 321; B cells during treatment with CM, and the RNA extracted from these particles has template activity in cell-free amino acid incorporating systems, while mature rRNA does not (Otaka §£_al., 196h). Attempts by others to repeat these experi- ments have led to conflicting results (Manor and Haselkorn, 1967), and several authors have pointed out that mRNA contamination is possibly the reason for the template activity. This is a pertinent criticism in light of Midgley and Gray's (1971) demonstration that GM treated cells accumulate mRNA, and Levinthal's finding that GM treatment of g, subtilis cells prevents mRNA degradation (Levinthal st 31,, 1962). Another argument against ribosomal precursor RNA acting as a template for ribosomal protein stems from the fact that the molecular weight of rRNA is too low to code for all the proteins found in ribosomes (Meore g£_al., 1968). This assumes that rRNA molecules are homogeneous, an assumption that may not be completely warranted (Santes £3 31., 1961; Aronson and Halowyczyk, 1965). It has been shown that there are several cistrons from rRNA in both E. subtilis (Smith—e9 §_1_., 1968; 019111 _e£ 31., 1965) and E, coli (Cutler and Evans, 1967) and it is not known if these are identical. Midgley and MCIlreavy, (1967) have found that the 5' terminal nucleotide sequence of 168 E, 2211 rRNA varies, depending on the growth conditions of the cells. Fellner and Sanger (1968) however, who have done extensive sequence analysis of rRNA from g, 231; have found no evidence for heterogeneity in this rRNA. An alternative approach to determining whether rRNA serves as a template in_gi!2 is to ask what prevents mature rRNA from serving as a template in 23352. Once this is known, one could then determine if the rRNA even exists in 2322 without this inhibitory factor present. Comparison of the properties of rRNA with those of various template active RNA molecules, indicates that rRNA contains more secondary structure and a number of modified nucleo- tides. This thesis will be primarily concerned with the effect of secondary structure which will be the subject of Part I, and to a more limited degree with the effect of modified nucleotides, the subject of Part II. The main types of modified nucleotides which occur in rRNA are pseudouridine, various base methylated nucleotides, and 2'7Qdmethyl nucleotides (Attardi and Amaldi, 1970). One experiment approach to determine the effect of these modifications in rRNA is to study their effect on the template activity of known synthetic polynucleotide templates. This has been done for pseudouridine (Pochon ‘25‘a1., 196k), and several base methylated nucleotides (Wahba 21; g” 1963 Ludlum g_t_ £11., 1964, McCarthy gt £11., 1966), but no work has been done to date on the effects on 2'-Qfmethylation on template activity. Thus Part II of this thesis is an investigation into the effects of 2'-Qfmethylation on the template activity of synthetic polynucleotides. The discussion following Part I will attempt to synthesize the results of the two parts. PART I - HISTORICAL The formulation of the hypothesis that secondary structure may influence the template capacity of RNA molecules was based upon the findings of two separate lines of research. The first involved studies on the species of RNA which could direct the incorporation of amino acids into proteins $2.!$£E2: and the second was a more detailed analysis of the secondary structure of the various types of natural and/or synthetic molecules of RNA. The first of these lines of research was developed during investigations on the mechanism of protein synthesis. Several workers (Nirenberg and Matthaei, 1961, Tissieres and Hepkins, 1961) reported that E, ggli_whole cell RNA added to E, ggll_cell-free extracts produced a 10 to 20 fold stimulation of protein synthesis. The stimulatory RNA was found to differ in properties from the ribosomal RNA fraction, which by itself was essentially devoid of stimulatory ability. It was subsequently postulated that the stimulatory fraction contained a specific type of RNA called "messenger RNA." In bacterial systems, this mRNA was characterized by rapid turnover and a base composition resembling that of deoxyribonucleic acid (DNA) (watson, 1965). Difficulty in separating this 5 messenger RNA from the large amounts of rRNA and transfer RNA (tRNA) in cells has precluded any further studies on purified cellular mRNA in bacteria. Whole cell RNA preparations from a wide variety of eukaryotic sources, for example liver (Barondes g£_§l,, 1962) brain (Zomzely g£_gl., 1970) reticulocytes (Arnstein §£.§l,, 196h) have been tested and found to have template activity. In the Specialized reticulocyte system it has been found that the haemoglobin mRNA is "long lived," (Nathans g£_gl., 1962) and several laboratories are engaged in isolating this mRNA. In 1961 Nirenberg and Matthaei made the importand dis- covery that the single stranded synthetic polynucleotide, poly U, was an efficient template for directing the incor- poration of phenylalanine to form polyphenylalanine. In addition to its importance in unraveling the genetic code, this result also substantially supported the hypothesis that mRNA.was a single stranded polynucleotide intermediate carrying the genetic information from DNA to ribosomes where it could be translated (the mRNA hypothesis). Soon other synthetic homopolynucleotides were found to be active as templates, but there was substantial variation in their relative activity; the order being poly U >poly A apoly C, with poly G being inactive (Speyer g£_gl,, 1965). Random copolymers of all four nucleotides were also active. However, polydeoxynucleotide polymers were inactive in extracts free of RNA polymerase activity (McCarthy g£_gl., 1966). The third type of RNA which has been found to have template activity is viral RNA. Nirenberg and Matthaei (1961) tested tobacco mosaic virus (TMV) RNA in their E. 221; B cell free extracts and found it to have template activity. Since then the RNA from several viruses, such as turnip yellow mosaic virus (Ofengand and Haselkorn, 1962), the RNA phages f2 (Nathans, 1962b), M82 (Nathans, 1965), and R17 (Gussin.g£‘2l., 1966), have been found to make excellent templates in cell-free systems. Viral RNA has the advantage that it can be obtained from preparations of pure phage, thus eliminating cellular rRNA and tRNA contamination. This has made possible analysis of both primary and secondary structure of these RNA.molecules and has led to the interesting findings that these molecules are single stranded, contain extensive secondary structure, and have long stretches of untranslated sequences at 5' and 5' ends (Cory, 1970). It remains to be seen, however, if the findings gleaned from these RNA molecules can be extrapolated to cellular mRNA. Concurrently with these studies on cell free template activity, work was also being done on the secondary structure of rRNA, the second line of research mentioned above. The potential importance of secondary structure in RNA was suggested by the dramatic correlation between structure and function of DNA as shown by Watson and Crick (1955). Much work has been done to elucidate the nature and relative contribution of the forces responsible for the secondary structure of DNA and RNA, but there are still many unanswered questions. Qualitatively there are two main factors reaponsible for the secondary structure of nucleic acids; hydrogen bonding (H-bonding) and base stacking. H-bonding occurs between the bases guanine and cytosine (G-C pairs) and between adenine and thymine (or uracil) (A-T or A-U pairs). Recent studies on these free bases, performed in organic solvents to minimize base stacking interactions, indicates that the pairings are very Specific, i.e., mixtures of the incorrect pairs (A-C or G-U) showed no tendency to interact (Felsenfeld and Miles, 1967). The importance of this Specificity for maintaining the correct sequence of bases in DNA and the genetic consequences thereof, was recognized by Wetson and Crick (1955) in their original proposal of the structure of DNA. Hydrogen bonding also occurs in the formation of complementary duplexes of synthetic polynucleotides such as poly adenylic acid (poly A) complexed with poly uradylic acid (poly U) (poly (A:U)) or poly (G:C). H-bonding can also occur between strands of the same homopolymer, such as poly G at neutral pH or poly A at acid pH. In addition to interstrand base pairing, polymers such as poly (A, U) consisting of alternating adenine and uracil bases, can exhibit intra- strand base pairing, where sections of the polymer loop back and form "hairpin" sections of H-bonded paired duplexes (Michelson g£_§l,, 1967). Experiments with single stranded synthetic polymers where H-bonding possibilities have been eliminated (such as poly N8 - dimethyladenylic acid), indicate that these polymers still possess considerable secondary structure as evidenced by their hyperchromicity and optical activity (Griffin 2£“21°: 196M). Michelson was the first to show that these properties result from the "stacking" of the bases of a single stranded polynucleotide with their planes parallel to one another, because of favorable free energy contribution from electro- static interactions or solvent exclusions (Michelson, 1965). Support for this theory comes from the analysis of the Optical Rotatory Dispersion (0RD) Spectra of dinucleotides, which indicates that even with this short oligonucleotide, base stacking occurs (DeLuca and McElroy, 1965). The hydrodynamic prOperties of non-hydrogen-bonded polymers 10 show that they are not rigid rods, however, like H-bonded duplexes, but more closely resemble flexible chains. Studies with poly A indicate that about two-thirds of the bases are in the stacked conformation at any given instant at 20° (Leng and Felsenfeld, 1966), thus giving considerable flexibility to the molecule. The exact origin of the energy of stabilization of single stranded structures is still under debate. Some of the factors which have been considered are dipole-dipole interactions of the bases, solvent exclusion and solvent interactions with the sugar-phosphate moiety (Felsenfeld and Miles, 1967). Hydrogynamic studies of the natural RNA's, rRNA, tRNA and several viral RNA's show that they are not rigid double- stranded duplexes like DNA but are flexible single-stranded chains. Analysis of ORD and circular dichroism (CD) spectra, along with hyperchromic measurements, showed the presence of considerable secondary structure, with both H—bonding and base stacking contributions present (Spirin, 196h). Cox and Kangalingan (1967) have estimated that about 75% of the secondary structure (as measured by hyperchromicity) of both rRNA and tRNA is due to H-bonding, the rest due to base stacking. The model which has emerged from these studies is shown in Figure 1. Short sections of the polynucleotide chains are seen to loop back over complementary 11 Figure 1. Proposed secondary structure of a fragment of.E; coli ribosomal RNA (Ehresmann 23 gl., 1970). Note the "folding back' of complementary sections and "looping out" of the non complementary hairpin turns. A structure similar to this has been proposed for M82 RNA. 12 cu A UA CA U c hGoC MCIG c Gm Gnu . G 6 vols . o c “.W W-“ of» . . 2-6.. . . m . 6.0 CL C mfiwwm£wucmm4wucA ”33.3.. Awmmw£££W£AGGCAAAAGAAGU UACCAC CG «cc.ccuccccuucuucc 0AACAC AUAAGcc c . a 0616 \6C ~ A a éclcw ”A ccc . use use 6:6 IV .9. GIC 6-0 0A 6 A C U c A a. e... IVA U A A A CC 66 m CA6 6A.. u a C e ‘ 15 sections and form hydrogen—bonded stretches. In these, and in the non-hydrogen—bonded sections, base stacking is present. Upon heating, the H-bonds are broken and base stacking is disrupted. Thus the increased ultraviolet absorbance (hyperchromism) represents the loss of both types of ordering forces. Because the "hairpin" loops are of variable length, and the loss of base stacking is non- cooperative, the melting profile is rather broad. The finding of considerable secondary structure in rRNA and its lack of template ability prompted Willson and Gros (196A) to hypothesize that the secondary structure of rRNA was preventing its translation. This was consistent with Nirenberg and Matthaei's (1961) finding that whereas poly U was an excellent template by itself, when complexed with poly A to form a double stranded duplex (poly AzU), it was completely inactive. Nirenberg g£_§E., (1965) had also observed that there was a correlation between synthetic homopolymers which were most active as templates and those which had the least secondary structure. Singer g£_gl,, (1965) were the first to systematically test the hypothesis that secondary structure influences template activity. They prepared synthetic polymers con- taining varying amounts of guanosine (G) and used these as templates in a cell-free system. They found that polymers 1h with large amounts of G were almost inactive as templates, had the highest thermal denaturation temperature, and were the most resistant to phosphorolysis by polynucleotide phosphorylase. They concluded that the presence of extensive secondary structure inhibits template activity. Szer and Ochoa (196h) arrived at similar conclusions by measuring the template activity of poly ribothymidylic acid (rT) both below and above its Tm of 56°. Poly IT is similar to poly U with the addition of a methyl group at position 5 of the uracil residues, and thus both have the same H-bonding Capabilities. Poly U has a Tm (the temperature at which the polymer is 50% denatured) of 8.50 while poly IT has a Tm of 56°, at pH 7.0 and 0.01 M M3012. At 570 or higher, poly rT was as effective as poly U in promoting the incorporation of phenylalanine, while at 200 the activity of poly rT was almost negligible compared to poly U. These authors also found that poly A becomes more effective for polylysine synthesis as the temperature is raised from 20 to #50. They interpret these results as indicating that extensive secondary structure of a polymer inhibits its attachments to ribosomes and hence its effectiveness as a template. A.more recent experiment of this type using natural RNA molecules was done by Lodish (1970). He used the RNA 15 from bacteriophage f2 (a virus which normally infects E. 221;) which is a very efficient template for protein synthesis in 'E. 22;; cell-free extracts (Nathans, 1962b). Analysis of peptides labelled in E, 521; B extracts with 14C-labelled N-formylmethionine under the direction of f2 RNA had shown that initiation of only two of the three genes takes place independently (coat and maturation); the initiation of the third (polymerase) gene can take place only after translation of the maturation gene. After mild formaldyhyde treatment, which partially disrupts H-bonding in the RNA, he was able to show that initiation could take place independently at all three initiation sites, thus implying that secondary structure prevented initiation at the polymerase gene. He obtained further evidence for the involvement of secondary structure in the translation of f2 RNA by measuring the incorporation of amino acids directed by f2 RNA at various temperatures, using an Eg_vitro system from E. stearothermo- philus (Lodish, 1971). At temperatures where the £2 RNA was partially unfolded, as evidenced by increasing A260, the incorporation increased with the temperature. Thus at 65°, where the £2 RNA is approximately 5% denatured, the incorporation of amino acids is 20 times greater than at 50°, where it is still in its fully-folded, compact form. The above results indicate that secondary structure 16 affects the ability of both synthetic and natural RNA polymers to function as templates in the cell-free synthesis of proteins. However, there has been no direct test of this hypothesis as it applies to rRNA. McCarthy g£_gl., (1966), attempted to test this hypothesis by including neomycin in a cell-free amino acid incorporating system from E, 221;. The neomycin stimulation of the translation of rRNA is greatly stimulated if the rRNA is first boiled. They interpret this promotion of template activity by heating to a loss of rRNA secondary structure or exposure of new ends, since they calculated that heating of the rRNA would have caused an average of one break per chain. However, they also point out that the inhibition in translation of rRNA may be due to the methylated bases it contains, and the neomycin may be overriding this inhibition. Other indirect evidence for the involvement of secondary structure in translation of rRNA comes from the work of Willson and Gros (196h) and Osawa (1965) on the template activity of precursor rRNA. These workers found that precursor ribosomal particles accumulate in E. £21; cells treated with chloramphenicol, or in methionine-requiring relaxed mutants which were starved for methionine. When the RNA was extracted from these particles, it was found to be less methylated than mature rRNA and had template activity in cell-free protein synthesizing systems. It also is reported to have less 17 secondary structure, on the basis of its slightly lower ("h0) Tm and an aggregation in Mg'II' containing buffers. Osawa (1965) has thus advanced the hypothesis that the precursor rRNA can be translated because it is not yet methylated and has a different (or less) secondary structure than mature rRNA. After it is translated, it "matures," i.e., it becomes methylated, assumes its complete or final secondary structure, and can no longer be translated. Since Lodish (1971) has shown that unfolding or reducing the secondary structure of f2 RNA affects its template activity, it seems reasonable to ask whether this holds true for rRNA as well. This hypothesis can be tested by using rRNA as a template for amino acid incorporation at a temperature sufficiently high to cause disruption of secondary structure. Although the Tm of bacterial rRNA is about 500 in NaCl-phosphate buffers (pH 7), (Spirin, 196%), the addition of Mg"+ to these buffers shifts the Tm upward almost 15°. Mg‘H' must be included in the buffers used to measure the Tm since it is a necessary factor in cell-free incorporation systems. The resulting Tm of the rRNA (6O - 70°) is above the temperature at which nesophilic bacterial cell-free amino acid incorporation can occur. Thus experiments using a homologous system of rRNA and an Eg_vitro amino acid incorporating system in E, coli is 18 not possible. However, Friedman and Wéinstein (1966), and Algranati and Lengyel (1966), have prepared amino acid incorporating systems from E. stearothermophilus which are active to almost 70°. Thus it should be possible to use mesophilic rRNA which has a Tm of 60 - 70°, and test it for template activity in E, stearothermophilus extracts at temperatures near its Tm. This will provide a direct test of the influence of secondary structure on the template activity of rRNA. If no activity is found at any temperature, under conditions in which control RNA.molecules are active, it is probable that the lack of rRNA template activity cannot be ascribed solely to its secondary structure. If incorporation takes place at temperatures where rRNA is thermally denatured, but not below these temperatures, then the hypothesis that secondary structure is preventing template activity remains valid. MATERIALS AND METHODS RNA Extractions (a) Whole Cell RNA. .2' subtilis (strain 168)1 cells were grown in one liter of media in a 5.8 liter Fernbach flask on a rotary shaker at 57°. The media consisted (per liter) of 10 gms Difco Bacto-tryptone, one gm Difco yeast extract, three gms Tris Cl pH 7.5. ‘E. stearothermophilus (strain NCA2l8h)2 cells were grown in the same media with agitation in a shaking water bath at 60°. When cells reached mid log phase (A660"0.6), they were poured over 500 gms of ice containing 15 ml 1.0 M NaN3, and then centrifuged at 7,000 x g_in a Sorvall refrigerated centrifuge for four minutes. The pellet was washed with #0 ml 0.1 M sodium acetate pH 5.0 and centrifuged at 7,000 x g for four minutes. The pellet was frozen, thawed and resuspended in six ml of 10 mM sodium acetate, pH 5. The cells were sonicated for h5 seconds at position 8 of a Bronwill Biosonik II ultrasonic tissue disruptor sonicator using a needle probe, and 60u1 of 25% sodium lauryl sulfate was added with mixing. Six ml of redistilled, water saturated, phenol was added, and the sonicate mixed for 1 2Courtesy of Dr. Alan Price, Univ. of Michigan, Ann Arbor, Mich. Courtesy of Dr. Koffler, Purdue Univ., Lafayette, Indiana 19 20 ten minutes on a vortex mixer, keeping it cool (Avho) by periodic chilling in an ice bath. The extract was centrifuged at 50,000 x g for ten minutes, the aqueous layer withdrawn and extracted twice more with an equal volume of water saturated phenol. After the third extraction, 0.10 volume of 1.0 M sodium acetate pH 5 and 2 volumes of cold 95% ethanol were added. The RNA was allowed to precipitate for several hours at -20°, and then centrifuged at 5,000 x g for five minutes. The pellet was taken up in two ml of 0.01 sodium acetate (pH 5), 0.01 M.MgC12) and approximately 20 ug of electrOphoretically purified DNA ase I (WCrthington Biochemical Company) added, and allowed to incubate at #0 for five minutes. Four ml of cold 95% ethanol was added, the RNA allowed to precipitate for one hour at -20°, and then centrifuged for five minutes at 5,000 xlg. The pellet was washed three times with 2.0 m1 of 5.0 M sodium acetate (pH 5.0) by vigorous resuSpension and recentrifugation, then rinsed twice with 95% ethanol, once with ethyl ether, dried, and dissolved in 1.0 m1 of water. The yield of total RNA was approximately 150 - 200 A260 units per gm of cells. (b) Ribosomal RNA. E, subtilis cells were grown as described in section (a). Upon reaching mid log (A660“’O.6) rifampicin (Calbiochem Company) was added to a final concentration of 50 ug/ml. Rifampicin has been shown (Di Mauro g£,§l., 1969) to Specifically inhibit RNA synthesis, and hence was added to 21 reduce the level of endogenous mRNA. The cells were incubated in the presence of rifampicin for 15 minutes, after which they were cooled, and harvested by centrifugation. All the following operations were carried out at 0 - ho. The packed cells were washed once with TKM buffer (0.05 M Tris Cl (pH 7.5), 0.05 M KCl, 0.01 M MgCla, 0.001 M dithiothreotol), centrifuged, and the pellet suspended in 10 ml TKM buffer per gm (wet weight) of packed cells. The cell suSpension was passed twice through a French press (Aminco Company, Silver Springs, Maryland) at 8,000 psi. After the first pass DNase I was added to a concentration of 10 pg per ml. The broken cell extract was centrifuged at 50,000 x‘g for 20 minutes, the supernatant removed and again centrifuged at 50,000 x g for 20 minutes. The supernatant contained in the upper three-fourths of the tube was centrifuged at 105,000 x g_for two hours to pellet the ribosomes. The supernatant (S-100) was carefully removed and the tube rinsed with TKM buffer in which the KCl con- centration had been increased from 0.05M to 2.0'M (HS-TKM) taking care not to disturb the surface of the pellet. The ribosomal pellet was gently suspended in 5 ml of the HS-TKM, allowed to stand overnight at 0°, and then centrifuged at 105,000 x g.to pellet the ribosomes. The supernatant was removed, the tube rinsed carefully with water, and the ribosomal pellet suspended in 5 m1 of 0.05 sodium acetate, 22 pH 5. The ribosomal suspension was transferred to a glass stOppered test tube, the ribosomes disrupted by adding sodium lauryl sulfate to a final concentration of 0.5% and the ribosomal proteins removed by extracting the solution three times with an equal volume of water saturated phenol. The RNA was then precipitated by addition of two volume of 95% ethanol and allowing to stand at -200 for two hours. The precipitate was washed three times with 5.0 M NaAc pH 5.0, rinsed twice with 95% EtOH, once with ether, dried and then dissolved in 1.0 ml of water. Measurement of Melting Profile of RNA RNA samples which were to be measured were diluted to a final concentration of approximately one A260 units/m1. The buffer used to dilute the RNA.was the same as that which was used in the amino acid incorporating assays, 0.05 TrisCl pH 7.5, 0.10 M NH401, 5 mM012, 5 mM Spermidine and lmM dithiothreotol. The samples were degassed by placing ig_ggggg for two minutes, placed in cuvettes, and the cuvettes sealed with plastic caps. One cuvette contained only buffer as a control blank. The temperature of the cuvettes was measured directly by placing a thermistor probe into one of the cuvettes filled with water. The temperature of the samples was raised in 5 degree increments every 15 minutes using a Haake FJ thermostatically controlled water bath which circulated water through Special heating spacers 25 fitted on a Beckman DU Spectrophotometer. After allowing the cuvettes to equilibrate for two to five minutes at each temperature, the A260 of the sample was measured. No corrections for volume changes have been made in plotting the resultant melting profile. Preparation of Amino Acid Incorporating Extracts Since it was not possible to anticipate the temperature required to demonstrate rRNA template activity, it was desirable to prepare a cell-free amino acid incorporating system which would be active at the highest temperatures possible. Two Species of bacteria, E, stearothermophilus and Thermus aquaticus were used in attempts to achieve these goals. The bacteria Bacillus stearothermophilus grows in the temperature range h5 - 68°. Cell-free amino acid incor- porating extracts have been prepared from this organism, which had endogenous activity up to 700 (Friedman and Weinstein, 1966), and were stimulated by added synthetic RNA (Algranati and Lengyel, 1966). In order to prepare extracts of high activity towards added natural mRNA an attempt was made to deplete endogenous mRNA.E2_!E!2_using the method of Forchammer and Kjeldgaard (1967). When an E. coli uracil auxotroph was starved for uracil (causing a preferential depletion of endogenous mRNA) the resulting 2h amino acid incorporating extract had very low endogenous amino acid incorporation and was more highly stimulated by added template RNA than extracts prepared in the conventional way (i.e., Nirenberg's technique (1961)). Initially, attempts were made to obtain a E, stearothermophilus uracil auxotroph using nitrosoguanidine as a mutagen; enriching for the mutant using the penicillin enrichment technique, and finally identifying the mutant by replica plating. However, after checking over 500 replica plated colonies, whflzh should have theoretically yielded 50 uracil auxotrOphs, and finding none, this procedure was abandoned. we reasoned that another theoretically attractive technique for depleting mRNA E2 2332 might employ the use of a specific inhibitor of RNA synthesis, since Levinthal 25.21:: (1962), and Forchhammer and Kjeldgaard (1967), found that the inhibition of RNA synthesis does not prevent its degradation. This would be an easier and more general approach than the use of uracil mutants, since the difficult and time consuming preparation of such mutants would be eliminated. TWO likely inhibitors were actinomycin D and rifamycin (or its derivatives). Actinomycin D is prohibitively expensive for large scale preparations, it is not effective with E, 2213, and its specificity has been questioned. Rifampicin (a derivative of rifamycin) however, has been 25 shown (Di Mauro E£.El°: 1969) to specifically inhibit initiation of RNA synthesis, it is effective against a wide spectrum of bacteria, and is inexpensive. Initially, amino acid incorporating extracts were prepared using E, stearothermophilus cells which had been incubated with rifampicin for 15 minutes after reaching mid log phase of growth. Since the half life of bacterial mRNA is about two to five minutes, this should allow time for most of the mRNA to decay (Levinthal SE 21': 1962). After incubation with the rifampicin, a crude $50 was prepared by lysing the harvested, washed cells in a French pressure cell, centrifuging the lysate at 50,000 x g for 50 minutes, and using the supernatant (S50) as the cell extract in the amino acid incorporating system. The activity of such extracts, while high with synthetic polymers, was not satisfactory with whole cell RNA (only about three to five times background). Therefore, the 850 system was fractionated into ribosomes and 105,000 x g_supernatant (8100), the ribosomes washed with high salt, and then recombined with the $100 supernatants for use in lEHXAEES amino acid incorporating studies. The disadvantage of this approach is that one might wash off the initiation factors, since this has been shown to happen with E, gglg_(Stanley E£.Elna 1966), and would result in washed ribosomes which would not 26 work with native mRNA. When this procedure was tried with 'E. stearothermophilus, the resulting reconstituted extract was found to have very high activity, both with synthetic polymers and whole cell RNA. In addition, the Mg++optimum for the template activity of whole cell RNA was low (5 mM) which indicates that the washing procedure used here does not remove the initiating factors, since high Mg++con- centrations are needed for initiation when the initiation factors are removed. Therefore, the following procedure was used to prepare the cell free extracts for the studies in this thesis. E, stearothermophilus cells were grown, harvested, lysed, separated into 8100 and washed ribosomes as described in the previous section. After the washed ribosomes had been resedimented from the high salt wash, the supernatant was removed and the tube rinsed carefully with TKM buffer. The ribosomal pellet was dissolved in 5 m1 of TKM buffer. Both ribosomes and the 8100 were frozen in small aliquots in a dry ice-acetone bath and stored frozen in liquid nitrogen. Equal volumes of the thawed ribosomes and 8100 were combined to form the reconstituted extract, which was used in the template studies. The protein content of the extracts was determined by the method of Lowry 22.213, (1951)- 27 A second Species of bacteria from which cell-free extracts were prepared was Thermus aquaticus. This extremely thermophilic bacteria which grows from #0 - 79° has been isolated and characterized by Brock and Freeze (1969). Presumably, amino acid incorporating extracts from this organism would permit testing the template activity of rRNA over a wider range of the melting profile since the extracts would be expected to be active to approximately 75 - 780, as compared to 65 - 680 for E, stearothermophilus. Cultures of this organism which were obtained from Dr. Brock (University of Indiana, Bloomington, Indiana), were grown and cell free extracts prepared. However, they were completely inactive when tested for amino acid incorporation even with the usually active template poly U. Variation of several parameters of the assay, such as Mg concentration, etc., did not stimulate activity. These bacteria produce copius Slime layers which heavily contaminated the cell free extracts and were probably responsible for its inactivity. Because of this, further experiments with this bacterial Species were discontinued. Assay for Template Activity of RNA The conditions for amino acid incorporation were similar to those of Nirenberg and Matthaei (1961) and the assay included a modification of Bollum's paper filter disc method 28 (1966). Each reaction contained the following components in a total volume of 70 pl; 0.05 M Tricine (adjusted to pH 7.8 with NH4OH), 0.10 M ammonium chloride, 5.0 mM phosphoenolpyruvate, 1.0 mM ATP, 0.2 mM GTP, 2.0 mM each of nineteen (1°C) amino acids, 2.0 mM of the 14C-labelled amino acid being studies (5-h0 m Ci/mM), 2.0 mM dithiothreitol, 1.0 A260 unit of E, stearothermophilus tRNA, 5 ug of phOSphoenolpyruvate kinase, 5 mM magnesium acetate, and 5 mM spermidine HCl. The polymer concentration is given for each experiment. Ten ul of reconstituted extract was used, composed of approximately 70 ug of ribosomal protein and 20 pg of $100 protein. When poly U was the template, the ammonium chloride concentration was reduced to 0.01 M. After addition of the reconstituted extract, the reaction mixtures were incubated at the temperatures indicated for each experiment for 20 minutes, chilled rapidly in an ice bath, and a 60 ul sample from each reaction mixture was spotted on Whatman No. 5MM disks (2.5 cm diameter). The disks were dried under a heat lamp, placed in a beaker of 5% trichloroacetic acid (approximately 10 m1/disk), and heated at 90 — 95° for 20 minutes. The disks were placed on a wet filter paper in a Buchner funnel and washed with 5% trichloroacetic acid, 95% ethanol, and finally dietyl ether. The dried disks were placed in 10 ml of 29 toluene containing 0.h% 2,5-bis (2(5-tert-butylbenzoxazolyl)) thiophene and counted in a Beckman LSlOO scintillation counter. Background incorporation of amino acids in the absence of added polynucleotide has been substracted from the values reported. RESULTS Before testing an rRNA molecule for its ability to direct the synthesis of protein at elevated temperatures in a‘E. stearothermgphilus cell-free incorporating system, three basic criteria must be fulfilled. First, under . ++ conditions of Mg , salt, etc., used in the cell-free system, the rRNA must be shown to undergo thermal denaturation at a temperature within the termperature range of the amino acid incorporating system. Second, mRNA isolated from the same bacterial source as the rRNA to be used in the experiments must be able to act as a template in the E, stearothermophilus amino acid incorporating system. The necessity for this requirement will be demonstrated later. Third, the incorporating system should be capable of detecting changes in template activity which are due to loss of RNA secondary structure. The melting profile of purified E, subtilis rRNA is shown in Figure 2. This rRNA was extracted from the ribosomes of rifampicin treated cells, as described in Materials and Methods. Gel electrophoresis of this RNA showed that only the 168 and 25S ribosomal species were present. The maximum thermal hyperchromicity of E. subtilis rRNA is about 50% (Stenesh and Holazo, 1967). On this basis the Tm or temperature at 50% maximum hyperchromicity of E, subtilis rRNA (Figure 2) is about 70°. Since the rRNA 5O 51 Figure 2. Thermal denaturation profile 0f.§° subtilis ribosomal RNA and R17 RNA. The profiles were obtained as described in Materials and Methods. The relative absorbance (the absorbance of the heated sample divided by the absorbance of the sample at 28°) is plotted against the temperature of the sample after it has reached temperature equilibrium. (—0 ), rRNA; (-—A ), R17 RNA- 32 8b mmameazE om om ON. Om Om ov on. ON .\.\ . 41.111! «\0 ‘ I \:, It)‘ cpm l. Llllll and ol'pol)(Am.C), L? X llllcpm/Ani. unit. . - . "manta—.- TI‘MI’LATI’. A("l'|Vl'| Y ()I-' Ger - ‘r _‘T‘—“"“ g 5, ./ V POLY (A,C) / ‘2’ /‘ 3 4r . ‘1 w . '— 4 a: O 0. 8 l J U 3“ o E LLJ Z 3 a 2'" .. O. i . '/——d——i '3 ‘ v 33 1 ,/POLY (Am,C) m I» - 5’ i o l /' % ,/' c... I - l--. -_. 00 so I 0 TIME (MINUTES) lltillRI 3: Time dependence of incorporation of [“(‘1proline di- _rected by poly/MC") or poly(AmL‘). Conditions of the S-30 assay are given in Materials and Methods. The reaction volume was 701) pl. which contained 2.6 /t1_-.- units of pt)l)(A.(') or I.3 Age. units of poly(AmC) and 5.8 mg of S-3t) protein. After addition of the S-30. (in-ill samples were withdrawn at the times indicated and spotted on Whatman No. 3M M disks. which were then treated as described in Materials and Methods. Incorporation is expressed as nanomoles of ammo acid incorporated per milligram of S-30 protein. poly(Am,C) and poly(A,C) are shown in I igure 28. Although poly(Am,C) was degraded at a slower rate than poly(A,C), the dill‘erences in the rates of degradation are not as pronounced as the dlII'erences between poly(Cm,U) and pub/(CU)- In determining the optimum conditions for the amino acid incorporation assays, dilferences were found between poly- (A.C) and poly(Am,C) for both the polymer saturation level and the Mg‘-H concentration. I’oly(A,(‘) saturated the reaction at 0.25 Age, unit/701d reaction, and had an optimum Mgl" contentration of ”.5 lmi, while poly(Am,C) saturated the reaction at O.l2 A-_1.,-, unit/7t) pi. and had an optimum My" concentration of I3.5 mst. A lower saturation level was also found for poly(Cm,U) (0.I2 A11. .1 unit/7t) pl) compared to POl)'((‘,U) (0.25 Air... unit '7l),1l). However. the optimum Mg“ concentration of 10 mM was the same for both poly(Cm,U) and poly(C,U ). Use of 2'-0-M(’Ih.rl‘ Ilmmi/ml_1'nu'r.s as Templates. The three. 2'-()-melhylribose homopolymers, poly(Am), poly(Cm), and poly(Um), were tested for template activity under condi- tions which were optimum for the corresponding nonmethyl- ated polymer. None of the three methylated homopolymers e\hibited any template activity under these conditions, which _W0uld have permitted detection of I 9:, of the poly(U)-directed incorporation, and 5% of the poly((‘)- or poly(A)-directed incorporation. Since template activity is highly dependent on ”‘0 M8" concentration in the assay, the 2’-0-methyl homo- IIOIYmers were also tested for template activity using a range 9f M8" concentrations from 4 to 20 mm. No activity was lnund at any Mg“ concentration. Effects o/ Neomycin on Temp/ate .4(‘Iit‘il_l‘ o/ Homo/milmers. MCCarthy cl (7!. (I966) and Morgan at al. (I967) have shown that neomycin enables DNA to act as a template for in rim) Protein synthesis. Hence it was of interest to see if neomycin Would influence the template activity of 2’-0-mcthyl-contain- 2’-0-Ml«;THvL POLYNUCLEOTIDES nMOl-ES ["c] PHE INCORPORATED/MG N 1 ./O /°’ 0 / I, / /v _ o o /v v 0 I"3'=‘-‘ = 9:? .. . 9:—-:;':_L—3;_fi 0 I5 30 45 60 TIME (MINUTES) “(illRl' 4: Time dependence of incorporation of [”C]phenylalanine directed by poly(U) or poly(Um) with and without neomycin. Con- ditions of the S-30 assay are given in Materials and Methods. The reaction volume was 350 1d. which contained 2.0 A2611 units of poly(U) or poly(Um) and 2.0 mg of S-30 protein. Neomycin con- centration was 5 pg/ml. After addition of the S-30, 30ml samples were withdrawn at the times indicated and spotted on Whatman No. 3MM disks. which were then treated as described in Materials and Methods. 0 Poly(U). (V) poly(Um). (O) poly(U) + neomycin, and (V) poly(Um) + neomycin. Incorporation is expressed as nano— moles ol' amino acid incorporated per milligram of protein. irig polymers. At neomycin concentrations of 5 pg/ml, neither poly(Cm) nor poly(Am) exhibited any template activity, but poly(Um) now served as an eflicient template for the incor- poration of phenylalanine. A time study of the template activity of poly(Um) plus neomycin as compared to poly(U) is shown in Figure 4. The results indicate that the rate of phenylalanine incorporation was slower when directed by poly(Um) plus neomycin, but incorporation continued for a longer period of time, perhaps due to the stability of the tem- plate. Thus the incorporation of phenylalanine directed by poly(U) is essentially over after 20 min, while the incorpora- tion of phenylalanine directed by‘ poly(Um) plus neomycin is still continuing after (10 min. To determine the relative ambi- guity of the neomycin-facilitated translation of poly(Um), the incorporation of the amino acids, leucine, isoleucine, serine. and tyrosine, was measured using both poly(Um) and poly(U) as templates. The results. in Table III, show that the total incorporation of the amino acids listed is about the same when the templateewas either poly(Um)with neomycin or poly(U) without neomycin. This implies that the total ambiguity in translation of poly(Um) in the presence of neo- mycin is about the same as that of poly(U) in the absence of neomycin. 'I he pattern of miscoding has changed, however, and the predominant miscoded amino acid with poly(Um) is serine as compared to leucine with poly(U). Effect ol'Neomrcin on the Template Actit‘ity o/'Heteropol,1‘- mers. Table IV lists the results of experiments to determine the elTect of neomycin on the incorporation of amino acids directed by poly(Cm,U) and poly(C,U). With poly(Cm,U) the ell'eet of neomycin depended on its concentration and the amino acid being tested. Low (0.5 15.0 pg/ml) concentrations of neomycin stimulated the incorporation of all amino acids, whereas higher concentrations of neomycin were inhibitory. The neomycin concentration for maximum stimulation of incorporation depended on the amino acid, and is given in Table IV. With poly(C,U) as a template however, the incor- poration of the amino acids listed in Table IV was inhibited at all concentrations of neomycin. The results of similar experiments using poly(Am,C) and poly(A,C) are listed in No. 13,1971 2585 BIOCHEMISTRY, VOL. 10, TABLE 111: Effect of Neomycin on the Misreading of Poly(Um) and Poly(U).a Poly(U) Poly(U n1) [”C]Amino — Neo- + New — Neo- + Neo- Acid Incorpd mycin mycin mycin mycin Leucine 510 330 ND" 260 Isoleucine 100 82 ND 51 Serine 61 150 ND 570 Tyrosine 50 ND N D N D Total 721 562 881 a Conditions for the S-30 assay are described ill Materials and Methods. Approximately 0.50 A1151, unit of polymer was used per reaction. Neomycin concentration was 5 pg/ml. Results are expressed as picomoles of amino acid incorporated per milligram of S-30 protein. b ND = none detected. Back- ground incorporation of amino acids in the absence of added polynucleotide has been subtracted from the values reported. Polynucleotidestimulated incorporation ranged from two- times background for isoleucine to fivedimes background for leucine. Table II. At a concentration of 5 ug/ml, neomycin stimulated the poly(Am,C)-directed incorporation of all the amino acids tested. This concentration of neomycin stimulated the poly- (A,C)-directed incorporation of four of the five amino acids tested, but inhibited the incorporation of the fifth, proline. Discussion Both poly(Cm,U) and poly(Am,C) served as templates for amino acid incorporation, as shown in Tables I and II. Com- pared to their nonmethylated analogs, poly(Am,C) was less active as a template than poly(Cm,U). Other experiments employing poly(Um,U), which was prepared by deamination of poly(Cm,U), indicate that poly(Um,U) also served as a template (data not shown). The effect of 2’-0-methyl groups on the template activity of a polymer may depend on which of the four nucleotides is 2’-0-methylated, the nature of adja- cent nucleotides, and the base composition of the polymer. The important conclusion to be drawn from these studies is that low leVels of 2’-0-methyl nucleotides in an RNA polymer do not eliminate its template activity and under certain condi- tions can be stimulatory. If methylation of the 2’-sugar position caused a significant change in the conformation of the bases in the polymer, it might alter the hydrogen-bonding capabilities of the com- ponent nucleotides and thus increase misreading. The results of Tables I and 11, however, indicate that both Am and Cm can replace their respective nonmethylated analogs without caus- ing an increase in misreading of the polymer. Nakada (1965) has postulated that in E. coli, methylation of nascent rRNA destroys its template activity. 2’-0-Mcthy| nucleotides comprise from 0.1 to 1.9‘7o of the component nucleotides oerNA (Starr and Sells, 1969). Our results, which show that even higher levels of 2’~0-methyl nucleotides (7 15%) do not completely inhibit template activity, suggest that the amount of 2’-0-methyl nucleotides found in rRNA would not be sullicient to prevent translation. Neomycin had no effect on poly(Am) or poly(Cm) tern— 2586 Bl(.l(lll—..slls1Ri', VOL. 11), NO. 13,1971 TABLE IV: Elfect of Neomycin on the Incorporation of Amin11 Acids Directed by Poly(Cm,U) and Poly(C,U).¢ Poly(C,U) Poly(Cm,L: ) —Neo- +Neo- —Neo- -i Nea- [”C]Amino Acid mycin mycin mycin mycin Phenylalanine I684 897 1868 21 $2 Serine 355 129 310 41: Leucine 548 330 494 1331‘) Proline » 132 77 126 2119 “ Conditions for the S-30 assay are described in Material and Methods. Approximately 0.50 A2110 unit of polymer 11.1.1 used per reaction. Neomycin concentration was 0.5 pgsllll 1’1.“ phenylalanine, 2.5 lag/ml for leucine and proline, and 5 pg/ml for serine. Incorporation is expressed as picomoles 1‘- amino acid incorporated per milligram of S-30 proteil‘ Background incorporation of amino acids in the absence 1‘ added polynucleotide has been subtracted from the value- reported. Polynucleotide stimuated incorporation range; from two-times background for proline to ten-times bael. ground for phenylalanine. plate activity, but promoted that of poly(Um) and stimulalri the template activity of both methylated hCIL'I‘OPOIanch. Tr.- promotion of poly(Um) template activity was not act-‘0‘)".- panied by any appreciable increase in miscoding. In a snail-.1: finding, Morgan et a1. (1967) reported that poly(d'l) is in- active as a template, but in the presence of neomycin it eli- ciently directed the incorporation of phenylalanine. They ills} noted that neomycin did not cause misreading of this polyntct Davis (1966) has reported that neomycin can cause inhihitm of template activity with only a slight increase in miscoding Our results are in essential agreement with the conclusiet that neomycin can affect template elliciency without signa- cantly increasing miscoding. Bobst et a1. (I969a-c) have shown that poly(Am) has mer: secondary structure than poly(A), and Zmudzka 91 ul. (1%hi obtained similar results for poly(Cm). Thus the failure of 11".. 2’-0-methyl homopolymcrs to serve as templates may be d1: to their increased secondary structure. since Szer and ()chee (1964) have reported that increased secondary structure in RNA decreases its template ability. The degradation studies using poly(Cm,U) and poly-(Amt) indicate that even in a crude cell-free protein-synthesizing system, known to contain a variety of nucleases (Barondes and Nirenberg. I962) methylated nucleotides confer DLI'L‘lLds.‘ resistance to a polymer. We have also noted that the methyl ated homopolymers are very resistant to mixtures of Illlsdllrlt' phosphatase, snake venom phosphodiesterase, and narrow;- cal nuclease. Thus 2'-()-methylation may provide a illechzmisltl for stabilizing a template without affecting its fidelity oflraln- lation. Further studies are in progress to investigate the ciful of 2’-0-methyl nucleotides on the stability of RNA 1111111111 various nucleases. Acknowledgments We thank Mr. Joseph Abbate and Mr. Lee Pike for lhcll comments and Miss Galvin Swift for expert technical assist- ance. ABSORPTION AND CIRCULAR DICHROISM STUDIES \ References Barondes, S. H., and Nirenberg, M. W. (1962), Science I38, 810. Basilio, C., Wahba, A. J., Lengyel, P., Speyer, J. F., and " Ochoa, S. (1962), Proc. Nat. Acad. Sci. U. S. 48, 613. ‘Bobst, A. M., Cerutti, P. A., and Rottman, F. (19693), J. Amer. Chem. Soc. 9], 1246. Bobst, A. M., Rottman, F., and Cerutti, . J. Amer. Chem. Soc. 91, 4603. -Bobst, A. M., Rottman, F., J. Mol. Biol. 46, 221. Bollum, F. J. (1966), in Procedures in Nucleic Acid Research, Cantoni, G. L., and Davies, D. R., Ed., New York, N. Y., Harper & Row, p 296. ""Davies, J. (1966), Cold Spring Harbor S ymp. Quant. Biol. 31, 665. Di Mauro, E., Snyder, L., Marino, P., Lamberti, A., Coppo, A., and Tocchini-Valentini, G. P. (1969), Nature (London) 222,533. Hall, R. H. (1964), Biochemistry3, 876. .Janion, C., Zmudzka, B., and Shugar, D. (1970), Acta Bio- chim. Polon. I 7 31. ——Jones, 0. W., Townsend, E. E., Sober, H. A., Heppel, L. A. P. A. (1969b), and Cerutti, P. A. (I969c), (1964), Biochemistry 3, 238. Knorre D. G., Sirotyuk, V. I., Stefanovich, L. E. (1967), Mol. Biol. 1, 837. Lowry, O. H., Rosenbrough, N. J., Farr, A. L., and Randall, R. J. (1951),]. Biol. Chem. 193, 265. McCarthy, B. J., Holland, J. J., and Buck, C. A. (1966), C old Spring Harbor 5 imp Quant. Biol. 31, 683. Morgan, A. R., Wells, R. D, and Khorana, H. G. (1967), J. Mol. Biol. 26, 477. Nakada, D. (1965), J. Mol. Biol. 12, 695. Nirenberg, M. W., and Matthaei, J. H. (1961), Proc. Nat. Acad. Sci. U. S. 47, 1588. Price, A. R., and Rottman, F. (1970), Biochemistry 9, 4524. Rottman, F., and Heinlein, K. (1968), Biochemistry 7, 2634. Rottman, F., and Johnson, K. L. (1969), Biochemistry 8, 4354. Smith, J. D., and Dunn, D. B. (1959), Biochim. Biophys. Acta 3], 573. Starr, J. L., and Sells, B. H. (1969), Physiol. Rev. 49, 623. Szer, W., and Ochoa, S. (1964), J. Mol. Biol. 8, 823. Wagner, E. K., Penman, S., and Ingram, V. M. (1967), J. Mol. Biol. 29, 371. Zmudzka, B., Janion, C., and Shugar, D. (1969), Biochem. Biophys. Res. C ommun. 37, 895. rt. . . lll' .é... lul‘t.‘ i‘lll'J‘lll‘ ll!