14 Nucleic Acids By R. J. H. DAVIES Biochemistry Department Medical Biology Centre The Queen‘s University of Belfast Belfast BT9 7BL 1 Introduction The advent of a new journal is usually regarded ambivalently by those reviewing the literature! However the appearance of Nucleic Acids Research in 1974 is to be welcomed for concentrating many of the contributions in this field into a single source. The first two volumes in a new series entitled ‘Basic Principles in Nucleic Acid Chemistry’ contain a number of advanced reviews on physico- chemical aspects of nucleic acids.’ The latest additions to the series ‘Methods in Enzymology’ give practical details for many of the established procedures used in the study of nucleic acid and protein synthesis.2 A much-needed author- itative textbook on the physical chemistry of nucleic acids has been published which affords an excellent comprehensive treatment of this subject3 On a more relaxing note ‘The Path to the Double Helix’4 provides an interesting and readable historical perspective of the great discovery.It is particularly fitting that the year which marked the twenty-first anniversary of Watson and Crick’s announcement of the structure of DNA should see another triumph for X-ray crystallographers in the determination of the three-dimensional structure of a tRNA molecule. The honours for the first correct structure at 3 A resolution appeared to have been won by Rich and his group,’ at MIT when they published their results for yeast tRNAPhe in March. However in August Robertus and co-workers6 at the MRC Laboratory in Cambridge who had been working on another crystal form of the same tRNA proposed a model differing significant- ly from that of the MIT group.The outstanding feature of the MRC model was ‘ ‘Basic Principles in Nucleic Acid Chemistry’ ed. P. 0. P. Ts’O Academic Press London 1974 vols. 1 and 2. * ‘Methods in Enzymology’ ed. L. Grossman and K. Moldave Academic Press London 1974 vols. 29 and 30. V. A. Bloomfield D. M. Crothers and I. Tinoco ‘Physical Chemistry of Nucleic Acids’ Harper and Row London 1974. R. Olby ‘The Path to the Double Helix’ Macmillan London 1974. F. L. Suddath G. J. Quigley A. McPherson D. Sneden J. J. Kim S. H. Kim and A. Rich Nature 1974 248 20. J. D. Robertus J. E.Ladner J. T. Finch D. Rhodes R. S. Brown B. F. C. Clark and A. Klug Nature 1974 250 546; A. Klug J. D. Robertus J. E. Ladner R. S. Brown and J. T. Finch Proc. Nut. Acad. Sci. U.S.A. 1974 71 371 I. 383 R.J. H. Darks that it identified in addition to the base pairs of the familiar cloverleaf a number of base pairs and base triplets -mostly not of the Watson-Crick type which ~ help to maintain the tertiary structure of the molecule. Most of these additional interactions can be related to constant features of tRNA sequences and the structure as a whole accords well with the chemical reactivity of various tRNAs. The MIT workers have since revised their original structure7 so that it is now in close agreement with that of the MRC group apart from some minor differences in interpretation.This new model for tRNA structure. which is discussed further in the section on tRNA will no doubt inspire detailed investigations by n.m.r. spectroscopy. Thanks to recent advances in instrumentation this technique can now be effec- tively applied to conformational studies of tRNA molecules in solution.8 It should be possible to compare the crystal structure of yeast tRNAPhe with that in solution and to determine which features of the secondary and tertiary structure are common to other tRNAs. Perhaps another twenty one years will see the unveiling of a three-dimensional map of the ribosome. This year has seen some more elegant experiments involving affinity-labelling probes (see p. 400),which have begun to define the location of the ribosomal proteins and RNA molecules around the peptidyl-transferase centre.Some indication of the progress made in the sequence analysis of nucleic acids over the past few years can be gained from a recent handbook' containing all the sequences published up to March 1974. This year alone many more tRNA sequences have been reported together with substantial sequence data on a variety of mRNAs and viral RNAs. While the methodology for sequencing small RNA molecules is now well established that for DNA is still being actively explored and refined. The available methods for DNA sequencing have been reviewed," and a number of DNA sequences have appeared during the year. The longest of these is a tract of 89 nucleotides in phage fl DNA." It seems clear that the routine sequencing of DNA molecules containing up to several hundred nucleotides should be achieved in the near future.This will generally be sufficient to analyse the individual cleavage fragments produced from larger DNA molecules by the action of restriction endonucleases. Already the cleavage fragments produced by different restriction enzymes have been ordered for DNA molecules containing up to tens of thousands of base pairs and the complete sequencing of such molecules may now be anticipated. As will be discussed in the relevant sections the available sequences of RNA and DNA frequently reveal homologies and elements of symmetry which have profound implications for the evolution and function of the nucleic acid molecules.' S. H. Kim F. L. Suddath G. J. Quigley A. McPherson J. L. Sussman A. H. J. Wang N. C. Seeman and A. Rich Science 1974 185 435. D. R. Kearns and R. G. Shulman Accounts Chem. Res. 1974 7 33. B. G. Barrel1 and B. F. C. Clark 'Handbook of Nucleic Acid Sequences' Joynson- Bruvvers Oxford 1974. lo K. Murray and R. W. Old Progr. Nucleic Acid Res. Mol. Biol. 1974 14 117; W. A. Salser Ann. Reo. Biochem. 1974 43 923. '*F. Sanger J. E. Donalson A. R. Coulson H. Kossel and D. Fischer J. Mol. Biol. 1974 90 315. Nucleic Acids 385 Last year's Annual Report12 alluded to growing concern over 'plasmid engineering' experiments whereby restriction cleavage fragments of DNAs from different sources can be joined by ligase to produce biologically functional hybrid DNA molecules.Earlier this year the National Academy of Sciences in the United States took the unique step of calling for a worldwide moratorium on certain types of experiments in this area of research because of their potential danger to human health.13 In particular it was suggested that all experiments involving the introduction into bacterial plasmids or viral DNA of (i) genetic determinants for antibiotic resistance or bacterial toxin formation and (ii) of segments of DNA from oncogenic or other animal viruses be deferred until their hazards can be reliably assessed. Although this step undoubtedly excludes many experiments of great intrinsic interest and potential benefit it can only be a wise and responsible decision in our present state of knowledge.In the mean- time there is plenty of scope for 'safer' endeavour in this field. For example fragments of X. laeuis DNA coding for the 18s and 28s ribosomal RNAs and generated by EcoRI restriction endonuclease have been linked in vitro to a bacterial plasmid. l4 When introduced into E. coli by transformation the recom- binant plasmid can replicate stably and act as a template for the synthesis of RNA complementary to the X. lae~isribosomal DNA. Much of the DNA in the 3 phage genome is not essential for the viabiIity of the phage. However if too much of this non-essential DNA is deleted infection of non-restricting E. coli cells will not result in plaque formation because the phage DNA molecules are too short to be packaged into virus particles.The incorporation of restriction cleavage fragments from foreign DNA sequences can restore the infectivity of such heavily deleted A genomes and this system offers an effective assay for the production of viable hybrid DNA molecules. It has been used15 to produce a large number of functional hybrid iphages which contain DNA sequences derived from either prokaryotic (E. coli) or eukaryotic (D.mehogaster) sources. Instead of inserting DNA into plasmid and viral genomes the action of restric- tion nucleases may be exploited to construct defined deletion mutants lacking one or more restriction cleavage fragments from their DNA. This procedure constitutes a powerful new method of genetic mapping and it has been applied to the production of deletion mutants of SV40I6 and phage 2 Bases and Nucleosides A new oxidizing agent for nucleic acid bases has been investigated :I8 potassium peroxodisulphate K,S,O,.Reaction occurs only with the anionic forms of the '' R. T. Walker Ann. Reports (B),1973 70 623. l3 See Nature 1974 250 175. I4 J. F. Morrow S. N. Cohen A. C. Y. Chang H. W. Boyer H. M. Goodman and ' R.B. Helling Proc. Nar. Acad. Sci. U.S.A.,1974 71 1743. M. Thomas J. R. Cameron and R. W. Davis Proc. Nat. Acad. Sci. U.S.A.,1974 71 * 4579. l6 C.-J. Lai and D. Nathans J. Mol. Biol. 1974 89 179. N. E. Murray and K. Murray Nature 1974 251 476. R. C. Moschel and E. J. Behrman J. Org. Chem. 1974 39 1983 2699. R. J. H. Davies bases and guanine is oxidized much more rapidly than adenine thymine uracil or cytosine.In 1 M-potassium carbonate at 40 “C,guanosine and deoxyguano- sine are at least 500 times more reactive than any other nucleoside and are converted mainly into guanidine urea and ribosyl- or deoxyribosyl-urea. The reagent may therefore be useful for the selective alteration and cleavage of polydeoxyribonucleotides at guanine sites. 1,N6-ethenoadenosine (1) is defor- mylated at C-2 by the action of O.05N-alkali. l9 The product (2)reacts with nitrous acid to give the fluorescent 1,N6-etheno-2-aza-adenosine(3). Instead of carbon dioxide and oxides of nitrogen treatment of adenine with sulphuric acid potas- sium bromide potassium permanganate and hydrogen peroxide yields 8,8’-dioxo-6,6’-azapurine (4). This bright orange compound” is an extremely potent noncompetitive inhibitor of xanthine oxidase (Ki = 2.8 x mol I-’) which may prove useful in the therapy of gout.(4) The synthesis of a number of purine nucleoside-6-sulphonates has been des- cribed.21 This provides a simple route to the ribonucleoside and 5’-nucleotide derived from 6-aziridinylpurine (5) which are potential affinity-labelling analogues of adenosine and 5’-AMP. The synthesis of some tricyclic nucleoside analogues of the ‘Y’base found in tRNA has been reported.22 Stemming from work on synthetic analogues of CAMP a useful new procedure for purine ring closure has been developedz3 which permits introduction of alkyl and aryl sub- stituents into the 2-position. CAMP can be easily converted24 in four steps via l9 K.C. Tsou K. F. Yip E. E. Millar and K. W. Lo Nucleic Acids Res. 1974 1 531. 2o J. R. Davis A. L. Jadhav and J. Fareed J. Medicin. Chem. 1974 17 639. ” H.-R. Rackwitz and K.-H. Scheit Chem. Ber. 1974 107 2284. ’’ G. L. Anderson B. H. Rizkalla and A. D. Broom J. Org. Chem. 1974 39 937. ” R. B. Meyer jun. D. A. Shuman and R. K. Robins J. Amer. Chem. SOC.,1974 96 4962. 24 R. B. Meyer jun. D. A. Shuman R. K. Robins J. P. Miller and L. N. Simon J. Medicin. Chem. 1973. 16 1319. Nucleic Acids 387 the l-N-oxide into the 5-amino-4-carboxamidine-imidazolenucleotide (6). This reacts with aldehydes under mild oxidative conditions to give a variety of 2-substituted derivatives (7). The reaction sequence should be quite generally applicable to adenine-containing compounds.A detailed of the mechan- ism of the intramolecular rearrangement of 06-allylic guanines to 8-allylic guanines indicates that a double [3,3] sigmatropic shift via C-5 is involved. R1= Ribosyl 3’,5‘-CyclicPhosphate The acetone-photosensitized addition of vinylene carbonate to uracil26 gives a mixture of the cis-and trans-[2 + 21 photoadducts (8) which on treatment with triethylamine in DMF is converted into 5-(formylmethyl)uracil (9) in 30 7; overall yield. The latter compound is a useful intermediate for the introduction of extended carbon chains into the 5-position of uracil and its versatility has been demonstrated in the synthesis of (f)-5-(4’,5’-dihydroxypentyl)uracil.The ribo- nucleoside2’ and deoxyribonucleoside28 of ‘uracil anhydride’ (10) have both 0 0 .i’&x 0 N H (9) been synthesized.These analogues are of interest because due to their reactive anhydride function they may be effective as active-site-directed irreversible inhibitors of enzymes involved in pyrimidine metabolism. The first examples of 1’,2’-unsaturated pyrimidine29 (of uracil) and purine3’ (of adenine) nucleosides (11) have been synthesized. The stannic chloride-catalysed reaction of silylated purines with fully acetylated sugars3 is claimed to be equivalent if not superior to other procedures for the preparation of adenine nucleosides. 25 N. J. Leonard and C. R. Frihart J. Amer. Chern. SOC.,1974,96 5894. 2b D. E. Bergstrom and W. C. Agosta Tetrahedron Letters 1974 1087.21 T. Ling Chwang and C. Heidelberger Tetrahedron Letters 1974 95. M. Bobek A. Bloch and S. Kuhar Tetrahedron Letters 1973 3493. 29 M. J. Robbins and E. M. Trip Tetrahedron Letters 1974 3369. ’O M. J. Robbins and R. A. Jones J. Org. Chem. 1974 39 115. 31 F. W. Lichtenthaler P. Voss and A. Heerd Tetrahedron Letters 1974 2141. R. J. H. Davies II Bu9, Bu B = U. C. A. or Hypoxanthine (12) The reaction of ribonucleosides with dibutyltin oxide32 gives crystalline 2‘,3’-O-(dibutylstannylene)nucleosides (12) in high yield. The dibutylstannylene function serves as an activating group for the 2‘-and 3’-oxygen functions treatment of (12) with phosphorus oxychloride leads to the formation of mixed 2’(3‘)-phosphates acyl chlorides and anhydrides give 3’-O-acyl derivatives while reaction with tosyl chloride selectively yields the 2’-tosyl-nucleosides.This method has been used to prepare 2’-0-nitrobenzyl-uridine.~~ The o-nitro- benzyl group constitutes a photolabile protecting group (removable by irradiation at 320 nm) and it has been utilized in the synthesis of UpU and UpA. The use of alkysilyl groups appears to represent a major breakthrough in the development of protecting groups for the hydroxyl functions of ribon~cleosides~~ and deoxyribon~cleosides.~~ The crystalline alkylsilyl derivatives are readily prepared in good yield by the reaction of the nucleoside with the alkylsilyl chloride and imidazole in DMF. For ribonucleosides the most useful reagents are t-butyldimethylsilyl chloride and tri-isopropylsilyl chloride.With uridine it is a simple matter to prepare the 2-and 2‘,5’-protected nucleosides which have been used in a synthesis of UpU. Removal of the silyl groups is achieved by the action of Bu”,NF in THF which in general does not affect other acid- or base- labile protecting groups. The authors promise to describe the synthesis of some important oligoribonucleotides using only alkylsilyl groups for the protection of sugar hydroxyl functions. For the deoxyribonucleoside deoxythymidine specific protection of the 5’-hydroxy-group is possible with t-butyldimethylsilyl- imidazole in pyridine. Alternatively the 3’,5’-disilylated derivative (13) may be prepared which because of the difference in the rate of hydrolysis of the two silyl groups can be converted in 50% yield into the 3’-protected deoxyribo- nucleoside (14) by treatment with acetic acid.ROQ’ ---+ HOA~ HOGT Me I R = Me3C-Si-I OR OR Me (13) (14) 32 D. Wagner J. P. H. Verheyden and J. G. Moffatt J. Org. Chem. 1974 39 24. 33 E. Ohtsuka S. Tanaka and M. Ikehara Nucleic Acids Res. 1974 1 1351. 34 K. K. Ogilvie K. L. Sadana E. A. Thompson M. A. Quilliam and J. B. Westmore Tetrahedron Letters 1974 2861. 35 K. K. Ogilvie E. A. Thompson M. A. Quilliam and J. B. Westmore Tetrahedron Letters 1974 2865. Nucleic Acids 389 A high-pressure liquid chromatography system has been described36 which is capable of resolving each of the common purine and pyrimidine bases and their ri bonucleosides.Spectroscopic studies have shown that in aqueous solution in~sine~~ is at least 85 ”/ in the keto-form while 4-thiouracil 4-thiouridine and 5,6-dihydro-4- thiouracil all exist in the 2-keto-4-thione form.38 N.m.r. measurements have shown that for Nh-methyladenine and N4-methylcytosine there is restricted rotation about the nitrogen-ring carbon bond.39 The two possible rotamers in each case lie in the plane of the ring and show a 20 1 preference for the syn-position in which the methyl group prevents normal Watson-Crick hydrogen bonding. It is estimated that methylation of adenine or cytosine should destabilize a Watsonxrick double helix by approximately 1.4 kcal (mol of methyl sub- stituent)-‘. Rotation about the C-2-N-10 bond in methyl- and dimethyl- guanine does not appear to be restricted.A simple new method for determining the anomeric configuration of ~-ribofuranonucleosides~~ is based on the n.m.r. spectra of their 2’,3’-isopropylidene derivatives. For /I-anomers the chemical shifts of the two isopropylidene methyl groups differ by more than 0.18 p.p.m. while for a-anomers the difference is less than 0.10 p.p.m. In recent years a number of bivalent metal ions have been reported to complex with guanosine in DMSO. These conclusions were based mainly on observations of changes in the chemical shifts of the guanosine base protons in the presence of the metallic chlorides. Strong evidence has now been produced41 that the species being complexed in these cases is not the metal cation but the chloride anion! It is believed that the charge-reversed chelate complex (15) is formed in which two positive centres chelate the negative ion.This finding may be Ribosyl (15) relevant to the well-known destabilizing effect of many anions on nucleic acid secondary structure. A further cautionary note derives from the observation4’ that co-ordination of the methylmercury(I1) cation at N-7 of inosine or of 5’-GMP catalyses very rapid exchange of the proton at C-8 with solvent water leading to broadening or disappearance of this proton resonance in the n.m.r. spectrum. Binding sites of paramagnetic cations have often been assigned on the 36 P. R. Brown S. Bobick and F. L. Hanley J. Chromatog. 1974 99 587. 37 F. E. Evans and R. H.Sarma J. Mol. Biol. 1974 89 249. 38 A. Psoda Z. Kazimierczuk and D. Shugar J. Amer. Chem. Soc. 1974,96 6832. 39 J. D. Engel and P. H. von Hippel Biochemistry 1974 13 4143. 40 J.-L. Imbach J.-L. Barascut B. L. Kam and C. Tapiero Tetrahedron Letters 1974 129. 41 Chien-Hsing Chan and L. G. Marzilli J. Amer. Chem. Soc. 1974 96 3656. 42 S. Mansy and R. S. Tobias J.C.S. Chem. Comm. 1974 957. R.J. H. Davies basis of the broadening of proton resonances caused by the cation-assuming it to arise solely from dipolar spin-lattice relaxation. Metal-ion-catalysed exchange with solvent must be ruled out before this assumption is justified. The structure of coformycin (16) an unusual antibiotic from Streptomyces has been established by X-ray crystallography and confirmed by an elegant synthesis from neb~larine.~~ Another research isolated the correspond- ing deoxyribonucleoside and again determined its structure by X-ray diffraction.Both compounds are very powerful inhibitors of adenosine deaminase because it is believed they may be structural analogues of the transition-state intermediate in the deamination of adenosine. The crystal structures of at least twenty other bases base analogues and assorted nucleosides have been published this year.4s Of particular interest are 6-aza-cytidine and 8-aza-adenosine both of which adopt a ‘high-anti’ conf~rmation,~~ i.e. the torsional angle between the base plane and the sugar lies outside the ranges conventionally ascribed to either the syn-or anti-conformation.This probably results from removal of the rotational barrier caused by interactions between the sugar C-2’ and H-2’ with H-6 of normal pyrimidine bases and H-8 of normal purines. 3 Nucleotides and Oligonucleotides A simple acetoxymercuration reaction4’ has been used to prepare the 5-mer- curiacetate derivatives of UTP (17) CTP dUTP and dCTP as well as the 7-mercuriacetate of 7-deaza-ATP. These compounds are mostly potent in- hibitors of polymerase enzymes. However on conversion into the corresponding mercurithio-compounds (18) by the action of thiols they become excellent substrates for all the polymerases tested. Potential applications of these mer- curinucleotides include their use as (i) heavy-atom reagents for crystallographic purposes (ii) affinity probes for enzymes containing active-site sulphydryl groups and (iii) reagents for forming protein-polynucleotide complexes.The 43 H. Nakamura G. Koyama Y. Iitaka M. Ohno N. Yagisawa S. Kondo K. Maeda and H. Umezawa J. Amer. Chem. SOC. 1974 96 4327; M. Ohno N. Yagisawa S. Shibara S. Kondo K. Maeda and H. Umezawa ibid.,p. 4326. 44 P. W. K. Woo H. W. Dion S. M. Lange L. F. Dahl and L. J. Durham J. Heterocyclic Chem. 1974 11 641. 45 Mainly in Acja Cryst. 1974. VOl. B30. 46 P. Singh and D. J. Hodgson J. Amer. Chem. SOC. 1974 96 1239 5276. 4’ R. M. K. Dale D. C. Livingston and D. C. Ward Proc. Nar. Acad. Sci. U.S.A. 1974 70 2238. Nucleic Acids 391 0 0 II HIAyHgX Adenosine-S-O-P=O/o-pTo-ON \ /O 0-P I I1\ 0-(17)(17) XX == OACOAC 0 ‘0-(18)(18) XX = SR (1% = long-suspected formation of monoadenosine 5‘-trimetaphosphate(19) when ATP salts are treated with DCC in anhydrous pyridine has been confirmed by 31P n.m.r.measurement^.^^ The molecule is highly reactive and rapidly reverts to ATP on contact with water. For nearly forty years it has been recognized that NADH (20) is unstable in dilute acid and slowly rearranges to a modified pyr-idine coenzyme known as the primary acid product. By a combination of 220 MHz n.m.r. spectroscopy c.d. measurements chemical modifications and specific deuterium labelling Oppenheimer and Kaplan4’ have now successfully resolved the identity of the product (21). The rearrangement reaction is shown to be stereospecific and its mechanism has possible implications for the function of the coenzymes in dehydrogenases.HO OH (20) A method has been described” for the rapid and complete separation of the mono- di- and tri-phosphates of A G U and C on a conventional anion-exchange column operating at low pressure. Nucleotides can be efficiently separated by anion-exclusion chromatography ;5 its effectiveness has been critically compared with that of anion-exchange chromatography. A promising new chromatographic material having a combination of reversed-phase and ion-exchange characteristics has been useds2 for the high-speed resolution and analysis of mono-and oligo-nucleotides at the nanomole level. 48 T. Glonek R. A. Kleps and T. C. Myers Science 1974 185 352.49 N. J. Oppenheimer and N. 0. Kaplan Biochemistry 1974 13 4675. 50 J. X. Khym J. Chromatog. 1974,97 277. R. P. Singhal European J. Biochem. 1974 43 245. ’’ R. A. Holton D. M. Spatz E. E. van Tamelen and W. Wierenga Biochem. Biophys. Res. Comm. 1974 58 605. R.J. H. Davies Cyclic Nuc1eotides.-The central importance of the 3'.5'-cyclic purine ribo- nucleotides CAMP and cGMP in the regulation of metabolic processes is now well recognised. This year cCMP has been reported53 to regulate the initiation of growth of leukaemia cells in culture. Furthermore cCMP has been detected in extracts from these cells though it remains to be established that the compound is not an artefact of the isolation procedure. A number of ribonucleoside 3'3'- cyclic phosphates including CAMP have been prepared in good yield by the action of base on the corresponding 2'-protected 3'-diphenyl phosphate^.'^ Adenosine 3',5'-cyclic phosphorothioate (22) has been synthesized;55 both diastereomers are hydrolysed by diesterases much more slowly than CAMP.I I 0-0-(22) (23) The analogue 3'-amido-3'-deoxyadenosine3',5'-cyclic phosphate (23) has also been prepared. 56 The 2',5'-cyclic phosphate derived from cordycepin (3'-deoxyadenosine) contains a seven-membered cyclic phosphate ring." Synthesis and Sequencing of Oligonucleotides.-It has been pointed out5* that when phenyl is used as the protecting group for internucleotide linkages in oligonucleotide synthesis by the phosphotriester approach a substantial amount of internucleotide cleavage may occur during unblocking.However this problem appears to be surmountable by the introduction of chloro-substituents into the phenyl residue and the use of milder conditions for phosphotriester hydrolysis. Arylsulphonyl triazolides (24) have been proposed59 as alternatives to aryl- R = Me or Me,CH (24) 53 A. Bloch Biochem. Biophys. Res. Comm. 1974,58,652; A. Bloch G. Dutschman and R. Maue ibid. 1974 59 955. 54 J. H. van Boom P. M. J. Burgers P. van Deursen and C. B. Reese J.C.S. Chem. Comm. 1974 618. 55 F. Eckstein L. P. Simonson and H.-P. Bar Biochemistry 1974 13 4675. 5b M. Morr M.-R. Kula G. Roesler and B. Jastorff Angew. Chem. Internat. Edn. 1974 13 280. " M. Ikehara and J. Yano Nucleic Acids Res.1974 1 1783. J. H. van Boom P. M. J. Burgers P. H. van Deursen R. Arentzen and C. B. Reese Tetrahedron Letters 1974 3785. 59 N. Katagiri K. Itakura and S. A. Narang J.C.S. Chem. Comm. 1974 325. Nucleic A cids 393 sulphonyl chlorides as condensing agents. Although the rate of reaction is slower the yields are considerably higher. Further studies6' on the synthesis of oligodeoxyribonucleotides on polymer supports using modified polystyrene resins have been described. The pentamer and hexamer of 2'-O-methylinosine 3'-phosphate have been complexed with poly(C) as template and then poly- merized" with a water-soluble carbodi-imide. The chain length of the product depends upon the stability of the polymer-oligomer complex but a 157; yield of the 30-mer has been realized.Khorana" has devised an effective method for linking oligodeoxyribo- nucleotides to cellulose through a poly(dT) spacer attached to the 3'-OH ter- minus (25). Replication of the oligonucleotide by DNA polymerase I is then Cellulose-p(dT),,,-dCCACCCC (5') (25) possible when poly(dA) or oligo(rA) as primer is hybridized to the poly(dT) region. The cellulose-linked template is easily separated from the product of replication by denaturation and may then be used for further cycles of replication. So far the method has been investigated only for the oligonucleotide dCCCCACC -a sequence from the E. coli tRNATy' gene. The results are very promising and the method should be equally applicable to the linkage and replication of poly- deoxyribonucleotides.If so it should greatly simplify the problem of efficiently replicating synthetic DNA molecules. Polynucleotide phosphorylase has been used previously to add a single protected nucleotide residue to an existing oligo- ribonucleotide. A disadvantage of the method is that partial phosphorolysis of the acceptor molecule normally occurs. This has now been overcome63 by the expedient of removing (enzymatically) inorganic phosphate from the reaction mixture as it is formed. The terminal addition of nucleotides to (PA) has been achieved in 90 "/ yield with essentially no side-product formation. Poly- nucleotide phosphorylase also appears to be capable of catalysing the addition of a single deoxyribonucleotide to oligonucleotide primers.64 The unique sequence from the T4 lysozyme gene which was determined by Streisinger from the genetic analysis of frameshift mutants continues to inspire oligonucleotide synthesis.A nona- and a dodeca-deoxyribonucleotide corres-ponding to sections of the gene ~equence,~' and a tetradecadeoxyribonucleotide segment from the complementary strand,66 have been synthesized ; the latter " H. Koster and F. Cramer Annalen 1974 946; H. Koster A. Pollak and F. Cramer ibid. 1974 959; H. Sommer and F. Cramer Chem. Ber. 1974 107 24. 6' S. Uesugi and P. 0.P. Ts'o Biochemistry 1974 13 3142. '' A. Panet and H. G. Khorana J. Biol. Chem. 1974 249 5213. " J. J. Sninsky G. N. Bennett and P. T. Gilham Nucleic Acids Res. 1974 1 1665. 64 S. Gillam K. Waterman M.Doel and M. Smith Nucleic Acids Res. 1974 I 1649. 65 S. A. Narang K. Itakura C. P. Bahl and N. Katagiri J. Amer. Chem. Soc. 1974 96 7074. '' R. Padmanabhan E. Jay and R. Wu Proc. Nut. Acad. Sci. U.S.A. 1974 71 2510. 394 R.J. H. Dauies has been investigated as a primer for DNA sequencing purposes. A dodeca-nucleotide complementary to a defined region of the endolysin gene of phage A has also been prepared and ~haracterized.~~ The enzymatic synthesis of the double-stranded block polymer d(C 5A15).d(T1 5G1 5)-which corresponds to three double-helical turns -has been reported.68 Oligoribonucleotides may be sequentially degraded69 by the action of either alkaline phosphatase or snake-venom phosphodiesterase in Combination with periodate to give a series of oligonucleotide dialdehydes which can be reduced with tritiated sodium borohydride to give 3’-terminal trialcohols.Subsequent separation of these oligonucleotide trialcohols according to chain length and identification of the labelled end-groups provides a very sensitive post-labelling technique for sequencing oligoribonucleotides. As little as 0.01 Az6* units of a non-radioactive decanucleotide are sufficient for its sequence determination. Spectroscopic and Structural Studies.-Details have been published” of the sophisticated method for determining the conformation of nucleotides in solution which relies on the use of lanthanide ions as paramagnetic n.m.r. probes. The geometry of nucleotide-lanthanide complexes is deduced from perturbations of the nucleotide n.m.r.spectrum caused by the complexing of a combination of relaxation (e.g.Gd3+) and shift (e.g.Eu3+) probes. The procedure for selecting molecular conformations which fit the n.m.r. data is entirely computerized and for AMP a small family of closely related conformations -very similar to the crystal structure-is obtained. A study of CAMP has also been made and the results have been independently ~onfirmed.~ A more conventional approach to the conformational analysis of nucleotides in solution is based on a com- prehensive knowledge of the chemical shifts and coupling constants of all the protons in the molecule. This method has been widely applied; the systems studied include all the common 5’-ribo- and -deoxyribo-nu~leotides,~~ and the dinucleoside phosphates ~TPT,’~ N6-methyladenylyl-uridine,and APU.~~ The low-field resonances of the N-3-H proton of thymine and the N-1-H proton of guanine in base-paired oligodeoxyribonucleotides have been identified and the perturbations of their chemical shifts due to nearest-neighbour base pairs have been eval~ated.~ ” R.Wu C. D. Tu and R. Padmanabhan Biochem. Biophys. Res. Comm. 1974,55 1092. J. F. Burd and R. D. Wells J. Biof. Chem. 1974 249 7094. b9 K. Randerath E. Randerath L. S. Y. Chia R. C. Gupta and M. Sivarajan Nucfeic Acids Res. 1974 1 1121 1329. 70 C. D. Barry J. A. Glasel R. J. P. Williams and A. V. Xavier f. Mof.Biol. 1974 84 471. 71 C. D. Barry D. R. Martin R. J. P. Williams and A. V. Xavier J. Mof.Biof.,1974 84,491 ;D.K. Lavallee and A. H. Zeltmann f. Amer. Chem. Soc. 1974 96 5552. 72 D. B. Davies and S. S. Danyluk Biochemistry 1974 13 4417. 73 D. J. Wood F. E. Hruska and K. K. Ogilvie Canad. J. Chem. 1974,52 3353. ” C. C. Altona J. H. van Boom J. de Jager H. J. Koerners and G. van Binst Nature 1974 247 558; Rec. Trau. chim. 1974 93 169. ” D. J. Pate1 and A. E. Tonelli Biopofymers 1974 13 1943. Nucleic Acids 395 The helix-coil transition has been monitored for nineteen double-helical oligoribonu~leotides~~ containing between six and fourteen base pairs. Helix stability is strikingly dependent on sequence and thermodynamic parameters have been derived to predict the T of any RNA double helix of known sequence. Two have independently published the crystal structure of the di- sodium salt of 5'-dGMP.The conformation about the C-5'-C-4 bond is gauchr-trans in contrast to gauche-gauche for all the other nucleotides whose crystal structures are known. The structure of the sodium salt of cGMP~~ has been determined. X-Ray diffraction photographs of crystalline dTpdA 79 indicate that the molecules are base-paired and stacked on top of each other to produce an extended helical structure resembling the A-form of DNA. 4 Polynucleotides Poly(2-methyl-N6-methyladenylic acid)" is strongly stacked in neutral solution does not form a regular helix in acid solution but forms a 1 :1 complex with poly(U) by Hoogsteen or reverse Hoogsteen base-pairing. The effects of 0-alkylation on the stacking of single-stranded poly(A) the stability of its double- helical acid form and its complexes with complementary polynucleotides have been investigated by two groups.81 In contrast to poly(A) the fluorescent analogue poly( 1,N6-etheno-2-aza-adenylicacid) appears to have a random structure in solution.82 Both poly(5-fluorocytidylic acid)83 and poly(5-ethyl- cytidylic have been synthesized and shown to form 1 :1 complexes with POlY(1).DNA polymerase I has been used to prepare d(TCC),.d(GGA) and d(TTG),. d(CAA),.8s A persistent problem in such syntheses has been that the enzyme frequently introduces covalent links between the two complementary strands. A protein factor has now been identified whose presence prevents the formation of these interstrand linkages.The self-complementary alternating copolymer of guanosine and 2-thio-cytosine poly[r(G-s2C)] has exceptional thermal stability,86 in common with other helical polynucleotides incorporating 2-thioketo-pyrimidine bases. In " P. N. Borer B. Dengler I. Tinoco jun.+ and 0.C. Uhlenbeck J. Mol. Biof.,1974 86 843. " D. W. Young P. Tollin and H. R. Wilson Acta Cryst. 1974 B30 2012; M. A. Viswamitra and T. P. Seshadri Nature 1974 252 176. '8 A. K. Chwang and M. Sundaralingam Acta Cryst. 1974 B30 1233. 79 H. R. Wilson Nature 1974 251 735. M. Hattori M. Ikehara and H. T. Miles Biochemistry 1974 13 2754. *' J. L. Alderfer 1. Tazawa S. Tazawa and P. 0. P. Ts'o Biochemistry 1974 13 1615; F. Rottman K. Friderici P. Comstock and M. K. Khan ibid. 1974 13 2762.n2 K. C. Tsou and K. F. Yip Biopolymers 1974 13 987. J. 0. Folayan and D. W. Hutchinson Biochim. Biophys. Acta 1974 340 194. 84 T. Kulikowski and D. Shugar Biochim. Biophys. Acta 1974 374 164. n5 A. R. Morgan M. B. Coulter W. F. Flintoff and V. H. Paetkau Biochemistry 1974 13 1596; M. Coulter W. Flintoff V. Paetkau D. Pulleyblank and A. R. Morgan ibid. p. 1603; W. F. Flintoff and V. H. Paetkau ibid. p. 1610. 86 P. Faerber. F.E.B.S. Letters 1974 44 11 I. 396 R.J. H. Davies lop3M-sodium cacodylate buffer it remains undenatured even at 98 "C! The secondary structure formed by poly(2-thiouridylic acid) (T = 68 "C) is considerably more stable than that of poly(U) (T = 8 "C under the same con- ditions). X-Ray diffraction8' has revealed that it comprises a double helix with non-equivalent polynucleotide chains and an unusual base-pairing scheme N-3-H--0-4 and S-2..H-N-3. Double-stranded polydeoxyribonucleotides containing alternating purine and pyrimidine bases such as poly(dA-dT).poly(dA-dT),have a distinctive double- helical structure.88 There are only eight base pairs per helical turn and the bases themselves are highly tilted with respect to the helix axis. This unusual molecular geometry may have implications for the biological function of some satellite DNAs. The structure of poly(dA).(dT) closely resembles that of B-DNA with C-3-exo-furanose ring pucker while poly(dG).poly(dC) preferentially adopts the A-conformation with C-3-endo-furanose rings.89 In poly(dT). poly(dA).poly(dT) each of the chains is in a twelve-fold helix of the A-t~pe.'~ A new molecular model" has been proposed for the secondary structure of poly(1).It consists of four identical poly(1) chains related to one another by a four-fold rotation axis. Each hypoxanthine base is linked to two others by hydrogen bonds involving 0-6 and N-1. An isogeometrical structure can be built for poly(G) with additional hydrogen bonds between every N-2 and N-7. 5 tRNA The proposed model' for the tertiary structure of yeast tRNAPhe based on X-ray crystallographic analysis at 3A resolution is shown in Figure 1. The ribose- phosphate backbone has been traced through the whole molecule with the excep- tion of a small corner and the positions of most of the bases have been determined.The model confirms the presence of the base pairs incorporated in the convention- al cloverleaf formula as well as identifying a number of extra interactions respon- sible for maintaining the tertiary structure of the molecule. These additional interactions involve base pairs and base triplets which are stacked or intercalated so as to make their hydrogen bonds inaccessible to water molecules. They are especially prominent in the central region of the molecule where four ribose- phosphate chains are in close proximity. Consideration of the tRNA sequences known at the present time shows that these interactions which are vital to the integrity of the unique tertiary structure are highly conserved." For example a trans-bonded base pair between G15 and C48 occurs in thirty-nine tRNAs.The importance of this interaction is emphasized by the fact that in eleven other tRNAs there have been co-ordinated base changes to A15.U48 so as to conserve complementarity at these points. A base triplet between the invariant bases *'-S. K. Mszumdar W. Saenger and K. H. Scheit J. Mol. Biol. 1974,85 213. S. Arnott R. Chandrasekaran D. W. L. Hukins P. J. C. Smith and L. Watts J. Mol. Biol. 1974 88 523. 89 S. Arnott and E. Selsing J. Mol. Biol. 1974 88 509,551. 90 S. Arnott R. Chandrasekaran and C. M. Marttila Biochem. J. 1974 141 537. 9' A. Klug J. Ladner and J. D. Robertus J. Mol. Biol. 1974 89 51 1. Nucleic Acids 397 -a c stem '-plus18a 19 Figure 1. A schematic diagram of the tertiary structure of yeast tRNAPhe.The ribose- phosphate backbone is represented by a continuous dark line except where there is ambiguity when it is shown dashed. Base pairs in the double-helical stem are repre- sented by long light lines and non-paired bases by shorter lines. Many of the latter stack as indicated those which do not are drawn at an angle to the backbone. Base pairs additional to those in the cloverleaf formula are indicated by dotted lines. f denotes the region containing residues 16 17 and 20. (Reproduced by permission from J. Mol. Biol. 1974,89 51 1) U8 (or s4U8),A14 and A21 is expected to occur in all tRNA species. Two further base triplets C1 3.G22.m7G46 and U12.A23.A9 are replaced in a co-ordinated fashion in other tRNAs. The insight into the structure of tRNA afforded by this model will no doubt provide the inspiration for much future research work and the explanation for much that is past.Already n.m.r. evidence has been presented92 for the ter- tiary base-pairing of the 4-thiouracil base at position 8 in many E. coli tRNAs with the adenine at position 14. The location of the heavy-metal-atom binding sites in isomorphous derivatives of tRNA crystals is vital to the successful interpretation of X-ray diffraction data. K. L. Wong and D. R.Kearns Nature 1974 252 738. 398 R.J. H. Davies Conventional oligonucleotide-fractionation procedures have been used to iden- tify C38 as the site of osmium attachment93 in a crystalline isomorphous derivative of yeast tRNA,M" and to show that when trans-dichlorodiammineplatinum(1r) reacts with yeast tRNAPhe the platinum is bound to the oligonucleotide G,AAYAl//p which contains the anticodon When the lanthanide ions Tb3+ and Eu3+ are complexed with E.coli tRNA their fluorescence is enhanced several hundred-fold9' due to energy transfer from 4-thiouridine residues to bound lanthanide ions. The binding of lanthanide ions to tRNA also has important applications in n.m.r. studies. From changes in the n.m.r. spectrum induced by Eu3+ it has been deduced96 that the folding of yeast tRNAPhe in solution brings the phosphate backbone of the CCA ter- minus and the dihydrouridine stem into close contact -a result in agreement with the crystal structure. The thermal melting of the cloverleaf arms of E.coli tRNA:"' has been followed by a combination of temperature-jump relaxation measurements and high- resolution n.m.r. spectro~copy.~' The results indicate that the dihydrouridine helix melts first followed by the Tl//Chelix the anticodon helix and finally the acceptor-stem helix. A similar involving only n.m.r. has been carried out for E. coli tRNAG'". This approach which allows the individual resonances of all the hydrogen-bonded ring NH protons in the tRNA molecule to be assigned has been further exploited in an e~amination~~ of yeast tRNAk'". This tRNA is particularly interesting because it can exist in a biologically active (native conformer) and an inactive state (denatured conformer). An analysis of the tem- perature dependence of the n.m.r.spectra of the two conformers has confirmed the cloverleaf structure for the native conformer and led to a structural model for the denatured conformer. In this model only the acceptor stem the T$C stem and the minor stem from the cloverleaf are retained and a new helix is formed by pairing bases from the anticodon loop with bases in the T$C loop. These conclusions are supported O0 by the equilibrium binding patterns of complemen- tary oligonucleotides to the denatured conformer. A very rapid (and economical) procedure"' for isolating a specific tRNA from a crude preparation involves firstly amino-acylation with the specific amino-acid followed by treatment with p-chloromercuribenzenesulphonyl chloride to form the sulphonamide of the amino-acyl group.This derivative reacts covalently with a sulphydryl-resin. After removal of the unreacted t-RNA the specific tRNA is released from the resin by deacylation. 93 J. J. Rosa and P. B. Sigler Biochemistry 1974 13 5102. 94 D. Rhodes P. W. Piper and B. F. C. Clark J. Mol. Biol. 1974 89 469. 95 M. S. Kayne and M. Cohn Biochemistry 1974 13 4159. 96 C. R. Jones and D. R. Kearns Proc. Nat. Acad. Sci. U.S.A. 1974 71 4237. 97 D. M. Crothers P. E. Cole C. W. Hilbers and R. G. Shulman J. Mol. Biol. 1974,87 63. 98 C. W. Hilbers and R. G. Shulman Proc. Nat. Acad. Sci. U.S.A. 1974 71 3239. 99 D. R. Kearns Y. P. Wong S. H. Chang and E. Hawkins Biochemistry 1974,13,4736. loo 0.C. Uhlenbeck. J. G. Chirikjian and J. R. Fresco J. Mol. Biol. 1974 89 495.* ' D. J. Goss and L. J. Parkhurst Biochem. Biophys. Res. Comm. 1974 59 181. Nucleic Acids 399 Two laboratorieslo2 have identified the modified nucleoside in E. coli tRNAPhe that is responsible for its reaction with phenoxyacetoxysuccinimide as 3-(3- amino-3car boxypropy1)uridine. Streptococcusfuecalis grown in the absence of folate initiates protein synthesis with non-formylated tRNA,M"' which differs from the normal formylated tRNA,M"' only in the T$C loop where uridine replaces thymidine.lo3 The tRNA from an extremely thermophilic and aptly named bacterium -Thermus thermophilus HB8 -which can grow at 85"C contains 2-thio-thymidine in place of thy- midine' O4 in the GT$CG sequence. Introduction of the 2-thiopyrimidine base should considerably stabilize the interaction between this sequence and the complementary CGAAC sequence of 5s RNA in the ribosomal A-site for which there is very good evidence.lo5 In fact binding of the tRNA fragment T$CGp to a ribosome-mRNA complex is sufficient to induce the synthesis of ppGpp and pppG~p.'~~ Only a selection of the tRNA sequences published this year can be mentioned.The sequence of one of the dimeric tRNA precursor molecules coded for by bacteriophage T4 has been determined. lo' Besides the nucleotide sequences for tRNASe' and tRNAPr0 it contains at least thirteen precursor-specific nucleotides located at the 5' and 3' termini and in the interstitial region. All the minor bases present in the mature tRNAs are found except for a 2'-O-methylguanosine residue from tRNAS".The 3'-CCA, termini of both tRNAs are absent so these nucleotides must be added enzymatically at a later stage of maturation. The complete sequences of the mammalian cytoplasmic initiator tRNAs (tRNA?'') from rabbit liver sheep mammary gland and mouse myeloma P3 cells are iden- tical.'o8 A cytidine residue replaces the uridine that has hitherto always been found to precede the 5'-end of the anticodon in the tRNA sequences. The unusual loop IV (normally T$CG loop) sequence AUCGm'AAA may be responsible for some of the distinctive functional properties of mammalian initiator tRNAs. The sequences of tRNA?"' and the major tRNAVa' from mouse myeloma cells have also been determined. lo9 The tRNAp' functions only in protein elongation in contrast to the tRNAF'' whose sole function is initiation.Comparison of the sequences of these two methionyl tRNAs shows no striking homologies. The Io2 Z. Ohashi M. Maeda. J. A. McCloskey and S. Nishimura. Biochemistry 1974 13. 2620; S. Friedman H. J. Li K. Nakanishi and G. Van Lear ibid. p. 2932. lo' A. S. Delk and J. C. Rabinowitz Nature 1974. 252 106. lo* K. Watanabe. T. Oshima M. Saneyoshi and S. Nishimura F.E.B.S. Letters 1974,43. 59. lo' F. Grummt I. Grummt H. J. Goss M. Sprinzl D. Richter and V. A. Erdmann F.E.B.S. Letters 1974 42 15. D. Richter V. A. Erdmann and M. Sprinzl Proc. Nar. Acad. Sci. U.S.A. 1974 71 3226. lo' B. G. Barrell J. G. Seidman C. Guthrie and W. H. McClain Proc. Nat. Acad. Sci. U.S.A. 1974 71 413.P. W. Piper and B. F. C. Clark Nature 1974 247 516; Errropeun J. Biochem. 1974 45 589; M. Simsek U. L. Rajbhandary M. Boisnard and (3. Petrissant Nature 1974 247 518. Io9 P. W. Piper and B. F. C. Clark F.E.B.S. Letters 1974 47 56. R. J. H. Dauies tRNAp' contains fifteen modified nucleosides and is unique in having only four base pairs in the anticodon stem. Another highly modified eukaryotic tRNA (tRNAys from yeast) contains sixteen modified nucleosides,' lo including a 5'-terminal pseudouridine and 2-thio-5-carboxymethyluridine methyl ester in the first position of the anticodon. It is reported' that the nucleoside composition of the tRNA from mice and mosquitoes does not vary with their age. Hopefully this extends to other or- ganisms too! 6 Arrinity Labelling of Ribosomes Affinity labelling with reactive substrate analogues has proved a powerful method for identifying amino-acid residues adjacent to or forming part of the active sites of enzymes.There has recently been growing interest in this technique as a means of investigating the structure of ribosomes and it has been applied successfully in a number of instances. Most studies have employed alkylating acylating or photoalkylating derivatives of the wamino-group of Phe-tRNAPhe. Cantor and his colleagues' ' have proved that N-bromoacetyl-phenylalanyl-tRNAPhe can bind normally to the peptidyl (P) site on the E. coli ribosome and react specifically with proteins L2 and L26-L27 of the large (L) subunit. Under different experimental conditions,' ' it binds preferentially to the amino-acyl (A) site and then labels protein L16 in addition to L2 and L26-L27.The protein L16 is thus identified exclusively as an A-site protein and it is concluded that all three proteins are located close to one another in the peptidyl-transferase centre. This reagent has been modified by the introduction of up to sixteen glycyl residues to give a series of peptidyl tRNA analogues' l4 with varying peptide chain lengths. These all react covalently with certain ribosomal proteins and the extent of reaction depends on the peptide chain length. The results favour the existence of a ribosomal binding site for the growing polypeptide chain and indicate that the proteins in this site are ordered with L2 closest to the 3'-end of the tRNA followed by L26-L27 L32-L33 and then L24.It has been inferred' ' from affinity-labelling experiments with p-nitrophenylcarbamyl-phenylalanyl-tRNAPhe that the proteins L27 L15 L2 L16 and L14 are all at or near tRNA binding sites on the E. coli 50s subunit. Other investigators utilizing reactive N-bromoacetyl analogues of puromycin which is a substrate for the ribosomal 'lo J. T. Madison S. J. Boguslawski and G. H. Teetor Biochemistry 1974 13 518; J. T. Madison and S. J. Boguslawski ibid.. p. 524. J. L. Hoffman and M. T. McCoy Nature 1974 249 558. M. Pellegrini H. Oen D. Eilat and C. R. Cantor J. Mol. Bid. 1974 88 809. lI3 D. Eilat M. Pellegrini H. Oen N. De Groot Y. Lapidot and C. R. Cantor Nature 1974,250 514. l4 D.Eilat M. Pellegrini H. Oen Y.Lapidot and C. R. Cantor J. Mol. Biol. 1974 88 831. A. P. Czernilofsky E. E. Collatz G. Stoffler and E. Kuechler Proc. Nut. Acad. Sci. U.S.A. 1974 71 230. Nucleic Acids 40 1 A-site have observed affinity labelling of the 23s RNA molecule in E. coli ribosomes," and proteins L27 and L29 of rat-liver ribosomes.' The photoaffinity reagent N-(2-nitro-4-azidophenyl)glycyl-phenylalanyl-t-RNAPhe is bound specifically to the P-site of E. coli ribosomes,"* in the presence of poly(U). Upon irradiation it reacts covalently with proteins L11 and L18 which are therefore presumed to be part of or adjacent to the peptidyl-trans- ferase centre. These proteins have not been labelled by a-halogenocarbonyl affinity probes possibly because they lack easily accessible nucleophilic groups.The phenacyl-p-azide of 4-thiouridine in E. coli tRNAVa' has been prepared.' l9 Photolysis of p-azidophenacyl-valyl-tRNAval,bound to the ribosomal P-site causes covalent linking exclusively to the 16s RNA of the 30s subunit. Poly- (4-thiouridylic acid) which can act as an mRNA for polyphenylalanine synthesis in vim becomes covalently bound'20 to the 30s subunit protein S1 when a complex of E. coli ribosomes poly(4-thiouridylic acid) and phenylalanyl-tRNAPhe is irradiated at 30WOO nm. This finding supports current evidence that protein S1 is an important constituent of the mRNA ribosomal binding site. Affinity- labelling with an azidophenyl derivative of GDP strongly implicates proteins L5 L11 L18 and L30 in the elongation-factor G-dependent binding of GDP to E.coli ribosomes.121 7 RNA General.-The radioiodination of RNA with labelled sodium iodide in the presence of thallic trichloride can give activities in excess of lo7c.p.m. pg- '. A systematic evaluation of this reaction has been carried out122 and has led to a satisfactory procedure for the iodination of very small samples of RNA. An improved high-resolution fractionation procedure for the preparative isolation of RNA by polyacrylamide gel electrophoresis has been developed.'23 When RNA is attached to a solid support for affinity chromatography or hybridization studies a stable covalent linkage is desirable. This may be achieved124 by coupling the RNA to phosphocellulose using carbonyldi-imidazole.Treatment of per- iodate-oxidized RNA with several newly described hydrazide compounds' introduces a fluorescent label at the 3'-terminus. The 3'-terminal fragments of radioactively labelled RNA molecules produced by nuclease digestion may be selectively isolated in excellent yield on columns of cellulose derivatives containing dihydroxyboryl groups.' 26 Using partial digests overlapping terminal fragments containing up to sixty nucleotides have P. Greenwell R. J. Harris and R. H. Symons European J. Biochem. 1974 49 539. 'I7 J. Stahl K. Dressler and H. Bielka F.E.B.S. Letters 1974 47 167. 'Is N. Hsiung and C. R. Cantor Nucleic Acids Res. 1974,1 1753. ' I. Schwartz and J. Ofengand Proc. Nut. Acad. Sci. U.S.A. 1974 71 3951. 1. Fiser K.H. Scheit G. Stoffler and E. Kuechler Biochent. Biophys. Res. Comm. 1974,60 1112. J. A. Maassen and W. Moller Proc. Nut. Acad. Sci. U.S.A. 1974 71 1277. 122 N. H. Scherburg and S. Refetoff J. Biol. Chem. 1974 249 2143. 123 F. S. Hagen and E. T. Young Biochemistry 1974 13 3394. 24 T. Y.Shih and M. A. Martin Biochemistry 1974 13 341 1. ''' S. A. Reines and C. R. Cantor Nucleic Acids Res. 1974 1 767. 126 M. Rosenburg Nucleic Acids Res. 1974 1 653. 402 R.J.H. Davies been successfully isolated and sequenced. In the presence of manganese ions RNA polymerase can incorporate deoxyribonucleotides into an RNA transcript of either single-stranded or double-stranded DNA. 12' In a normal transcription reaction it is possible to replace one of the ribonucleoside triphosphate precursors entirely by the corresponding deoxyribonucleoside triphosphate.This finding opens up exciting new possibilities for nucleic acid sequencing. For instance by using U,-RNase with dG-RNA and pancreatic RNase with dC-RNA or dU-RNA base-specific cleavage is now possible at any of the four nucleotide residues (TI-RNase for G residues). In this way overlaps of at least six residues can be obtained for any site in the RNA. Some preliminary and promising applications of this method are reported. A new method for the preparative stepwise degradation of polyribonucleotides128 is based on the periodate oxidation-fl elimination+nzymatic dephosphorylation reaction sequence. The conditions for quantitative reaction in each step have been perfected and up to nine degradative cycles have been carried out successfully on tRNA fragments.Meanwhile the established RNA sequencing procedures have been used to determine the primary structure of the U-1 RNA of Novikoff hepatoma ascites cells which contains 171 nucleotide residues. 129 Ribosomal RNA.-A new model for the secondary structure of 5s RNA has been proposed on the basis of high-resolution n.m.r. ~pectroscopy.~ The room- 30 temperature structure contains five helical regions with a total of 28 base pairs of which 17 are GC. 32 The 3'-terminal sequence of E.coIi 16SrRNA which has been determined as -AUCACCUCCUUA,H suggests that it may play a direct part in the ter- mination and initiation of protein ~ynthesis.'~ The terminal trinucleotide UUA, is complementary to the terminator codon UAA while the sequence ACCUCC complements a conserved sequence found in the ribosome-binding sites of coliphage mRNAs.In eukaryotes,' 33 the corresponding 3'-sequence of the 18s rRNA is -GAUCAUUA, and it appears to be invariant. The terminal hexanucleotide is complementary to two termination codons and could recognize a third (UAG) if 'wobble' occurred. As no tRNA species appear to bind termination triplets it is possible that 18s RNA fulfils this function. The distribution of the polypyrimidine sequences in 28s rRNA from rat hepatoma cells has been studied.134 Of the 118 fragments identified with chain lengths between 5 and 21 nucleotides 94 (comprising a total of 667 nucleotides) have been sequenced.In addition a sequence of 99 nucleotides has been e~tablished'?~ 12' A. Van de Voorde R. Rogiers J. Van Herreweghe H. Van Heuverswyn G. Volckaert and W. Fiers Nucleic Acids Res. 1974 1 1059; G. V. Paddock H. C. Heindell and W. Salser. Proc. Nat. Acad. Sci. U.S.A. 1974 71 5017. 12* G. Keith and P. T. Gilham Biochemistry 1974 13 3601. Iz9 R. Reddy T. S. Ro-Choi D. Henning and H. Busch. J. Biol. Chem. 1974 249 6486. D. R. Kearns and Y. P. Wong J. Mol. Biol. 1974 87 755. ' 31 J. Shine and L. Dalgarno Proc. Nat. Acad. Sci. U.S.A. 1974 71 1342. C. Ehresmann P. Stiegler and J.-P. Ebel F.E.B.S. Letters 1974 49 47. 133 J. Shine and L. Dalgarno Biochem. J. 1974 141 609. 134 R. N. Nazar and H. Busch J. Biol. Chem. 1974 249 919. 135 R.Kanamaru Y. C. Choi and H. Busch J. Biol. Chem. 1974 249 2453. Nucleic Acids 403 which is remarkable in having just a single adenylic acid residue -located at its 5'-end. The methylation pattern of HeLa cell rRNA and its precursors has been in- ~estigated.'~~ The 18s RNA and 28s RNA contain approximately 46 and 70 methyl groups respectively whereas the 5.8s RNA contains only one. The transcribed spacer regions of the precursor molecules are unmethylated. When precursor and mature ribosomal RNA molecules are examined by electron microscopy,' 'a reproducible arrangement of hairpin loops reflecting discrete areas of secondary structure may be observed. Maps based on this secondary structure for X. laevis and mouse L-cell ribosomal precursors have revealed the arrangement of the mature RNA sequences within the precursors and have been used to follow the steps in the maturation process.mRNA and Viral RNA.-A brief review' 38 dealing primarily with the isolation and structure of eukaryotic mRNA has appeared. Despite intensive research and speculation the function(s) of the 3'-terminal poly(A) sequence found in eukaryotic mRNA molecules remains a mystery. Several reports this year have complicated rather than clarified the issue. Contrary to earlier opinion it now seems that a substantial proportion of cytoplasmic mRNA molecules at least from HeLa cells139 and from sea-urchin embryo^,'^' may lack poly(A). Although in mammalian cells poly(A) is added to the mRNA precursor (HnRNA) in the nucleus cytoplasmic polyadenylation of mRNA has now been demonstrated in sea-urchin embryos.14' The relative rates of translation (in Xenopus oocytes) of rabbit globin mRNA with and without the terminal poly(A) sequence have been measured.'42 Initially the rates of translation are comparable but at longer incubation times the poly(A)-free mRNA is translated much less rapidly. This finding suggests that the presence of the poiy(A) increases the functional stability of the mRNA. A similar study utilizing cell-free protein-synthesizing systems supports this view.'43 Perhaps this explains why the poly(A) sequence is necessary for the infectivity of poliovirus RNA. 144 Other representatives of the picornavirus group namely the cardioviruses and foot-and-mouth disease viruses contain poly(C)-rich tracts up to 100 nucleotides long in their RNA.14' It is not yet established whether these poly(C) sequences are terminally situated.136 B. E. H. Maden and M. Salim J. Mol. Biol. 1974 88 133. 13' P. K. Wellauer and I. B. Dawid J. Mol. Biof.,1974,89,379; P. K. Wellauer I. B. Dawid D. E. Kelley and R. P. Perry ibid. p. 397. 38 G. Brawermann Ann. Rev. Biochem. 1974 43 621. L39 C. Milcarek R. Price and S. Penman Cell 1974 3 1. I4O M. Nemer M. Graham and L. M. Dubroff J. Mof. Biol. 1974,89,435. 14' I. Slater and D. W. Slater Proc. Nur. Acad. Sci. U.S.A. 1974 71 1103. 142 G. Huez G. Marbaix E. Hubert M. Leclercq U. Nudel H. Soreq R. Salomon B. Lebleu M. Revel and U. Z. Littauer Proc. Nar. Acud. Sci. U.S.A. 1974 71 3143.143 A. E. Sippel J. G. Stavrianopoulos G. Schutz and P. Feigelson Proc. Nut. Acad. Sci. U.S.A. 1974 71 4635. L44 D. H. Spector and D. Baltimore Proc. Nut. Acad. Sci. U.S.A. 1944 71 2983. '45 F. Brown. J. Newman J. Stott A. Porter D. Frisby C. Newton N. Carey and P. Fellner Nature 1974 251 342. 404 R. J. H. Davies Methylated nucleosides have been identified in the mRNA from rat hep- at~rna'~~ and mouse L-~ells.'~' In the rat hepatoma mRNA all four 2'-0-methyl-nucleosides occur but base methylation is mainly confined to N6-methyladenosine. As with rRNA methylation may be important in the matura- tion of eukaryotic mRNA from its precursors. Bacteriophage T 7 early mRNAs are generated from a large precursor molecule by RNase I11 cleavage at specific base sequences.48 The sequencing of mRNA has been greatly advanced by methods involving the use of reverse transcriptase enzymes to produce faithful cDNA transcripts. A number of oligonucleotide sequences in human globin mRNA thus established have been shown to match unique amino-acid sequences in the a-or B-globin chains. 149 Sequence analysis of immunoglobulin light-chain mRNA has proved that a single mRNA molecule codes for both the variable and constant regions of the antibody light chain.'" In addition an untranslated sequence of 52 nucleotides adjacent to the poly(A) segment of this mRNA has been determined. This shows striking homologies with the corresponding sequence of 52 nucleo-tides from rabbit P-globin mRNA.' 51 In both cases identically positioned double-hairpin-loop secondary structures can be drawn which may have an important bearing on the biological function of these and other mRNA molecules.5'-Terminal sequences exceeding 60 nucleotides and including the initiation codons have been determined for the mRNAs from the lactose'52 and galactose'53 operons of E. coli. An examination of the sequences at the ribosome-binding sites of a number of RNA molecules' 54 has revealed some remarkable symmetries which may have implications for the evolution of the genetic code. True palindromic sequences (reading the same forwards and backwards) frequently occur overlapping with or immediately following the initiation codon. In addition sequences from this region often contain one or more out-of-phase termination codons.Sequences of 83 nucleotides from the replicase cistron'55 of phage Qp and of 54 nucleotides from the A protein ci~tron'~~ of phage R17 have been reported. The in oitro preparation of QP RNA with a G to A transition 16 bases from the 3'-terminus has been described.' 57 Further studies involving 'evolution in a 146 R. Desrosiers K. Friderici and F. Rottman Proc. Nut. Acad. Sci. U.S.A. 1974 71 397 1. R. P. Perry and D. E. Kelley Cell 1974 1 37. 148 R. A. Kramer M. Rosenburg and J. A. Steitz J. Mot. Biol. 1974 89 767 777. 149 C. A. Marotta B. G. Forget S. M. Weissman I. M. Verma R. P. McCaffrey and D. Baltimore Proc. Nut. Acad. Sci. U.S.A. 1974 71 2300. lSo C. Milstein G. G. Brownlee E. M. Cartwright J.M. Jarvis and N. J. Proudfoot Nature 1974. 252 354. lS1 N. J. Proudfoot and G. G. Brownlee Narure 1974 252 359. 152 N. M. Maizels Proc. Nut. Acad. Sci. U.S.A. 1973 70 3585. 153 R. E. Musso B. DeCrombrugghe I. Pastan J. Sklar P. Yot and S.Weissman Proc. Nat. Acad. Sci. U.S.A. 1974 71 4940. L54 G. Pieczenik P. Model and H. D. Robertson J. Mol. Biol.,1974 90 191. Is' A. G. Porter J. Hindley and M. A. Billeter European J. Biochem. 1974 41 413. IS' U. F. E. Rensing A. Coulson and J. G. G. Schoenmakers European J. Biochem. 1974 41 431. R. A. Flavell D. L. Sabo E. F. Bandle and C. Weissman J. Mol. Biol. 1974,89 255. Nucleic Acids 405 test tube' have been carried on the MDV-1 RNA mentioned in last year's Report." The sequence of a mutant RNA whose replication is resistant to ethidium bromide inhibition and which evolved during a serial transfer experi- ment differs from that of wild-type MDV-1 RNA at only three positions out of 218.The secondary structure of a 59-nucleotide fragment from phage R 17 mRNA has been e~tablished"~ by a combination of different physical techniques. mRNA is expected to form stable secondary structures computer-generated random base sequences can be arranged into thermodynamically stable conformations with about 50'1 of the bases paired.16' 8 DNA General.-Methods of gene isolation have been reviewed. lhl Solid supports bearing DNA sequences for specific genes can be preparedLh2 by hybridizing mRNA to oligo(dT)-cellulose by means of its poly(A) tail and then using reverse transcriptase to synthesize cDNA which is covalently attached to the cellulose.This provides a highly specific chromatographic medium for the isolation and purification of complementary nucleic acid sequences. The transformation of cells in uitro can be effected by a segment of human adenovirus type-5 DNA which contains about 1500 base pairs and which is located close to one end of the DNA molecule. 63 Excluded-volume interactions of DNA with other macromolecules such as polyethylene oxide in solution cause it to collapse into a compact relatively dense state known as $ DNA. Production of $ DNA in viw may be relevant to chromosome structure and the packaging of DNA in viruses. X-Ray scattering studies have now shown164 that $ DNA adopts a folded-chain structure com- parable to that of many crystalline linear polymers.Despite its highly anomalous c.d. spectrum the secondary structure of DNA differs little from that of the B-form of DNA to which it reverts on heating.'65 Lnteraction with Small Molecules.-Supercoiled DNA is twice as twisted as was thought! A careful study'66 of the unwinding of helical DNA by ethidium based on a new theoretical and experimental approach has shown that each bound ethidium molecule unwinds the helix by 26 _+ 3 degrees instead of the previously accepted value of 12". The antibiotic echinomycin which comprises two quinoxaline rings linked by a cyclic octapeptide is an extremely potent 15* F. R. Kramer.D. R. Mills P. E. Cole T. Nishihara and S. Spiegelman J.Mo/. Biol. 1974 89 719. J. Gralla J. A. Steitz and D. M. Crothers Nature 1974 248 204; C. W. Hilbers R. G. Shulman T. Yamane and J. A. Steitz ibid. p. 225. '(I" J. Gralla and C. DeLisi Nature 1974 248 330. Ih' D. D. Brown and R. Stern Ann. Rev. Biochem. 1974 43 667. 16' P. Venetianer and P. Leder Proc. Nut. Acad. Sci. U.S.A. 1974 71 3892. F. L. Graham A. J. van der Eb and H. L. Heijneker Nature 1974 251 687. T. Maniatis J. H. Venable jun. and L. S. Lerman J. Mol. Biol. 1974 84 37. 165 S.-M. Cheng and S. C. Mohr F.E.B.S. Letrers 1974 49 37. lh6 J. C. Wang J. Mol. Biol. 1974 89 783. R.J. H. Davies inhibitor of RNA synthesis. On binding to DNA both the quinoxaline rings of the antibiotic are intercalated into the helix.'67 Intercalative binding to DNA is also exhibited by a cationic terpyridyl platinum complexL68 and by the anti- tumour agent 9-hydroxyellipticene.'69 The ability of acridine dyes to cause frame- shift mutations appears to correlate with the quenching of their fluorescence when intercalated close to a GC base pair.'70 A phenazinium dye with a high specificity for complexing GC base pairs elutes DNA molecules from hydroxy- apatite according to their base comp~sition.'~ ' Preferential binding to GC-rich DNA is shown' 72 by the antitumour compound C~S-P~(NH,)~CI,. Tritium-displacement studies' 73 indicate that the binding of benzpyrene to DNA does not involve the K-region but probably occurs by covalent substitution at position 1 or 6. Sequence Studies.-A detailed account has now appeared' 74 of the methodology used in the direct determination of the sequence (48 bases) of a fragment from phage 4x174 DNA.DNA polymerase I repair synthesis with ribosubstitution and primed by appropriate oligodeoxyribonucleotides,has been used in estab- lishing the sequences complementary to a stretch of 81 nucleotides in phage fl DNA," and of 29 nucle~tides'~~ from the 1-strand of phage $8Opsu,,,+ DNA. The latter sequence (26) immediately precedes the starting point of transcription of the E. coli tRNATyr gene. It shows remarkable elements of two-fold symmetry (boxed in bases) about its central base pair and can loop out from the regular DNA helix to form cruciform arms in which 18 out of 25 residues are base- paired.RNA transcripts complementary to both the strands of SV40 DNA in the region immediately preceding the preferred site of initiation by E. coli RNA polymerase have been isolated and sequenced.' 76 The DNA template in this promotor region contains a palindromic sequence of 17 bases. '67 M. J. Waring and L. P. G.Wakelin Nature. 1974 252 653. 16' K. W. Jennette S. J. Lippard G. A. Vassiliades and W. R. Bauer Proc. Nar. Acad. Sci. U.S.A. 1974 71 3839. 169 J.-B. Le Pecq N.-D. Xuong C. Gosse and C. Paoletti Proc. Nat. Acad. Sci. U.S.A. 1974 71 5078. "* J. P. Schreiber and M. P. Daune. J. Mol. Biol. 1974 83 487. W. Pakroppa and W. Miiller Proc. Nat. Acad. Sci. U.S.A. 1974 71 699. 17' P. J. Stone A. D. Kelman and F. M. Sinex Nature 1974 251 736. 173 G.M. Blackburn P. E. Taussig and J. P. Will J.C.S. Chem. Comm. 1974 907. 17' F. Galibert J. Sedat and E. Ziff J. Mol. Bid 1974 87 377. 17' T. Sekiya and H. G. Khorana Proc. Nut. Acad. Sci. U.S.A. 1974 71 2978. B. s. Zain S. M. Weissman. R. Dhar and J. Pan Nucleic Acids Res. 1974 1 577; R.Dhar S. M. Weissman B. S. Zain J. Pan and A. M. Lewis jun. ibid. p. 595. Nucleic Acids 407 The specific cleavage of DNA molecules by restriction endonucleases and the ordering of the resulting fragments into physical maps has found wide application during the year. Several such studie~'~~-'~~ have been made on SV40 DNA and the results have proved very helpful in understanding certain aspects of the expression of this genome.' 79 Restriction-cleavage maps have also been reported for the replicative form of 6x174 DNA,'80 polyoma virus DNA,'" and mito- chondrial DNA.'82 1 1 (5') N-G-T-T-A-A-C-N (3') (5') N-C-C-G-G-N (3') (3')-N-C-A-A-J-T-G-N (5') (3') N-G-G-C-C-N (5') The symmetrical sequences at the cleavage sites of two restriction endo- nucleases from H.parainjluenzae Hpa I (27) and Hpa I1 (28) have been deter- mined.'83 The phage P1 modification enzyme methylates the central adenine in the sequence pAGATCT.la4 The duplex formed by this hexanucleotide with its complementary sequence resembles the recognition sites for several restriction enzymes in having a two-fold axis of symmetry. A review of DNA modification and restriction has been published.'8s Satellite DNA.-Restriction endonucleases have proved very useful in the identification and analysis of satellite DNAs.Because of their repetitive base sequence many satellite DNAs are resistant to the action of restriction enzymes and this may be used as a criterion for their identification and subsequent isolation.'86 In other cases such as some rodent satellite DNAs,lg7 a highly regular arrangement of restriction-endonuclease-sensitivesites has been ob-served. By its susceptibility to different restriction enzymes bovine satellite I DNA has been proved'88 to consist of direct tandem repeats 1400 base pairs in length. Renaturation kinetics indicate that this repeated sequence is itself internally repetitious. Three satellite DNAs from Drosophila virilis have repeating heptanucleotide sequences la9 which are related to one another by simple base- P.Lebowitz W. Siegel and J. Sklar J. Mol. Biol. 1974 88 105. K. N. Subramanian J. Pan S. Zain and S. M. Weissman Nucleic Acids Res. 1974 1 727. D. Nathans S. P. Adler. W. W. Brockman K. J. Danna T. N. H. Lee and G. H. Sack jiin. Fed. Pror. 1974 33 1135. A. S. Lee and R. L. Sinsheimer Proc. Nut. Acad. Sci. U.S.A. 1974 71 2882. B. E. Griffin M. Fried and A. Cowie Proc. Nar. Acad. Sci. U.S.A. 1974 71 2077. W. M. Brown and J. Vinograd Proc. Nar. Acad. Sci. U.S.A. 1974 71 4617. D. E. Garfin and H. M. Goodman Biochem. Biophys. Res. Comm. 1974,59 108. J. P. Brockes P. R. Brown and K. Murray J. Mol. Biol. 1974 88 437 W. Arber Progr. Nucleic Acid Res Mol. Biol. 1974 14 1. G. Roizes Nucleic Acids Res.1974 1 1099. W. Horz I. Hess and H. G. Zachau European J. Biochem. 1974 45 501. M. R. Botchan Nature 1974 251. 288. IB9 J. G. Gall and D. Atherton J. Mol. Biol. 1974 85 633 408 R.J. H. Ducks pair changes satellite I poly(dACAAACT).poly(dAGTTTGT) ; satellite 11 poly(dATAAACT).poly(dAGTTTAT);satellite 111 poly(dACAAATT).poly-(dAATTTGT). Hermit-crab satellite I DNA is p~ly(dTAGG).poly(dCCTA).'~~ Chromosome Structure.-Investigations of the organization of the DNA sequences in eukaryotic genomes have concentrated on the readily available ribosomal DNA from Xenopus. The 5s ribosomal DNA consists of repeated units of 750 base pairs of which 120 base pairs represent the actual gene for 5s RNA while the remainder is termed spacer DNA.Twelve closely related oligomers about 15 nucleotides long have now been identified'" from the spacer region and their tandem arrangement has been demonstrated. In the DNA containing the genes for Xenopus 18s and 28s rRNA the repeating units are arranged in head-to-tail fashion'92 and the spacer lengths are hetero-geneous.193 The individual strands of eukaryotic DNAs contain large numbers of self-complementary sequences which are capable of forming hairpin loops containing thousands of base pairs.'94 A new model for the structure of chromatin'95 is based on a repeating unit of two of each of the four main types of histone and about 200 base pairs of DNA. It is supported by endonuclease digestion studies.'96 A comprehensive survey of the structure and function of chromosomes has appeared.I9' Enzymes.-Many of the latest developments in the field of DNA replication are described in a collection of papers from a recent syrnp~siurn.'~~ The role of multi-enzyme systems in DNA replication has been reviewed.'99 At high tem- peratures cytosine bases in single-stranded DNA are deaminated to uracil at an appreciable rate whereas in native DNA they are well protected.200 An N-glycosidase activity has been detected,20' in extracts from E.coli which releases free uracil from single- and double-stranded DNA containing deaminated cytosine residues. An exonuclease from calf thymus has been characterized202 which introduces single-strand breaks into double-stranded DNA specifically Iyo D. M. Skinner W. G. Beattie F.R. Blattner B. P. Stark and J. E. Dahlberg Bio-chrmistrj,. 1974 13 3930. 19' G. G. Brownlee E. M. Cartwright and D. D. Brown J. Mol. Biol. 1974 89 703. Iy2 S. Henikoff J. Heywood and M. Meselson J. Mol. Biol. 1974 85 445. 19' P. K. Wellauer R. H. Reeder D. Carroll D. D. Brown A. Deutch.T. Higashinakagawa and 1. B. Dawid Proc. Nut. Acad. Sci. U.S.A. 1974,71. 2823. '94 D. A. Wilson and C. A. Thomas jun. J. Mol. Biol. 1974 84 115. Iy5 R. D. Kornberg Science 1974 184 868. '96 M. Noll Nature 1974 251 249. 19' Cold Spring Harbor Symp. Quant. Biol.,1974 vol. 38. 19' 'DNA Synthesis in Vitro' ed. R. D. Wells and R. B. Inman. Medical and Technical '" Publishing Lancaster 1973. R. Schekman A. Weiner and A. Kornberg Science 1974 186 987. 2oo T. Lindahl and B.Nyberg Biochemistry 1974 13 3405. T. Lindahl Proc. Nut. Acad. Sci. U.S.A. 1974 71 3649. 202 S. Ljundquist and T. Lindahl J. Biol. Chem. 1974 249 1530; S. Ljundquist A. Andersson and T. Lindahl ibid. p. 1536. Nucleic Acids 409 at apurinic sites. The covalent linking of polyribonucleotides to polydeoxyribo- nucleotides by DNA ligase has been dem~nstrated.”~ Microwave-induced emission spectrometry which can detect as little as 10-l4 (g atom) of metal in a 1 p1 sample has been used to prove that reverse transcriptase from avian myeloblastosis virus contains two atoms of zinc per enzyme molecule.204 This observation may account for demonstrated dif- ferences in zinc metabolism between normal and leukaemic leukocytes. Con-ditions have been estab1ished2O5 for using E.coli DNA polymerase I as a reverse transcriptase to produce cDNA from an RNA template; the molar ratio of enzyme to template is critical. The reconstitution of the RNA polymerase core and holoenzymes from their respective subunits has been achieved.206 A new form of RNA polymerase RNA polymerase 111 has been discovered207 which has an extra subunit of low molecular weight bound to the holoenzyme RNA polymerase I. lo3 K. Nath and J. Hurwitz J. Biol. Chem. 1974 249 3680. ‘04 D. S. Auld H. Kawaguchi D. M. Livingston and B. Vallee Proc. Nut. Acud. Sci. U.S.A..1974 71 2091. ’*’S. C. Gulati D. L. Kacian and S. Spiegelman Proc. Nut. Acud. Sci. U.S.A. 1974 71 1035; E. C. Travaglini and L. A. Loeb Biochemistry 1974 13 3010. ’04 L.R. Yarborough and J. Hurwitz J. Biol. Chem. 1974 249 5394. ‘07 W. Wickner and A. Kornberg Proc. Nat. Acad. Sci. U.S.A. 1974 71 4425.