20 Nucleic Acids By R. T. WALKER Department of Chemistry Birmingham University Birmingham B 15 2TT For the first time for many years there has been no significant change in the num- ber of papers published in 1973 when compared with the previous year' and this is no doubt a result of the cut-back in basic research. This year three main subjects can be singled out for special mention re- striction repression and X-ray crystallography. The first two subjects will be dealt with later and the advance in X-ray crystallography this year is due almost entirely to Rich and his co-workers.2-8 Several papers by this team on the struc- ture of tRNA have appeared almost in the reverse order to that in which they were written because as the results got more exciting so the speed of publication was increased.The polymorphic nature of yeast tRNAPhe was made apparent when it was found to crystallize in five different crystal systems involving eight different space groups.2 The orthorhombic form was found to occur in two forms ;3 they had identical unit cell dimensions in two directions but the distances in the third direction differed by 35%. The longer form changed to the shorter when the crystal was dried and this suggested that the tRNA molecules could slide together along the c-axis without a substantial change in internal structure. The breakthrough came when isomorphous heavy-atom derivatives were obtained for these orthorhombic crystal^.^ The data obtained at 5.5 8 resolution resulted in a model for tRNA which was quickly proved by the data at 4%1to be in~orrect.~ The model predicted from the 5.5 %1 data was a long thin molecule but it was not possible to trace unambiguously the phosphate backbone and 1 'Nucleic Acids Abstracts' ed.E. S. Krudy and A. Williamson Information Retrieval Ltd. London 1973. 2 S. H. Kim G. Quigley F. L. Suddath A. McPherson D. Sneden J. J. Kim J. Weinzierl and A. Rich J. Mol. Biol. 1973 75,421. 3 S. H. Kim G. Quigley F. L. Suddath A. McPherson D. Sneden J. J. Kim J. Weinzierl and A. Rich J. Mol. Biol. 1973 75,429. 4 S. H. Kim G. Quigley F. L. Suddath A. McPherson D. Sneden J. J. Kim J. Weinzierl P. Blattmann and A. Rich Proc. Nar. Acad. Sci. U.S.A. 1973 69 3746. 5 S. H. Kim G. J. Quigley F. L. Suddath A. McPherson D. Sneden J.J. Kim J. Weinzierl and A. Rich Science 1973 179 285. 6 S. H. Kim H. M. Berman N. C. Seeman and M. D. Newton Acra Crysr. 1973 B29 703. 7 J. M. Rosenberg N. C. Seeman J. J. P. Kim F. L. Suddath H. B. Nicholas and A. Rich Nature 1973 243 150. 8 R. 0.Day N. C. Seeman J. M. Rosenberg and A. Rich Proc. Nut. Acad. Sci. U.S.A. 1973 70,849. 624 Nucleic Acids 625 with hindsight it can be seen that at one point the chains of two adjacent mole- cules were muddled. The 4A map gave what is now widely accepted to be the correct structure for this tRNA which is in the form ofan 'L'.' The result obtained by interpreting the data collected at 3%1resolution is now eagerly awaited as this should enable the position of the bases to be determined so that those bases capable of forming hydrogen bonds can be identified.Many other .people are working on the structure of this and other tRNAs and it will be interesting to see whether these results are confirmed and whether other tRNAs have similar structures because already there are signs of disagreement.' The work on the structure of dinucleoside phosphates is probably of as great a fundamental importance as the determination of tRNA structures. The seven basic conformations for these molecules are described6 and in particular adeno- sinyl-3',5'-uridine phosphate (APU)~ and guanylyl-3',5'-cytidine phosphate (GpC)* both crystallize to give right-handed antiparallel double helices in which the ribose phosphate backbones are held together by Watson-Crick hydrogen- bonding between the base residues.The resolution could be obtained at 0.84.98 and thus the atomic details of structures likely to be very similar to those of double-helical nucleic acids are now available. The unit cell of ApU contains 26 water molecules and GpC contains 36. When they have all been accurately located it should be possible to understand how hydrophobic forces lead to a stabilization of nucleic acid double helices. There is one sodium ion per ApU molecule and the two ions in each asymmetric unit occupy distinct sites. One is as expected between phosphate groups but the other lies in the minor groove of the helix and has octahedral co-ordination to two uracil oxygen atoms and to water molecules. In GpC each phosphate group is associated with two sodium ions each of which has octahedral co-ordination to additional water molecules.To complete a successful year Rich and his colleagues have reported pre- liminary X-ray analysis data from the crystalline complex formed between polypeptide elongation factor Tu and GDP." Large crystals have been obtained using the same method as was used to crystallize tRNA. The complex of phenyl- alanyl-tRNAPhe elongation factor Tu and GTP has also been crystallized,' but the crystals were too small for X-ray crystallographic analysis. To set the record straight the results reported last year," that guanine and cytosine contain appreciable quantities of minor tautomers have been shown to be incorrect and it has been suggested that the presence of paramagnetic impu- rities in the solutions led to the misinterpretation of the n.m.r.spectra.13 Inter- calation of ethidium bromide into DNA is claimed to cause the DNA helix to unwind not wind.I4 To prove a point made last year neither of the sequences of M. Levitt J. Mol. Biol. 1973 80 255. lo D. Sneden D. L. Miller S. H. Kim and A. Rich Nature 1973 241 530. I' K. Arai M. Kawakita S. Nishimura and Y. Kaziro Biochim. Biophys. Acra 1973 324,440. '*G. C. Y. Lee J. H. Prestegard and S. I. Chan J. Amer. Chem. SOC.,1972 94 951; G. C. Y. Lee and S. 1. Chan ibid. p. 321 8. l3 Y. P. Wong K. L. Wong and D. R. Kearns Biochem. Biophys. Res. Comm. 1972 49 1580; Y. P. Wong J. Amer. Chem. SOC.,1973,95 3511. l4 W. J. Pigram W. Fuller and M. E. Davies J. Mol.Biol. 1973 80 361. 626 R. T. Walker the 'renaturable' yeast tRNAL'" independently determined by two laboratories' was correct and a third version has now appeared.'6 The chromosomal location of human repetitive satellite DNA continues,' ' but work claiming to have located the structural genes for haemoglobin" has been seriously q~estioned,'~ as calculations from the specific activity of the mRNA used and the known number of cistrons for these genes per genome show that the hybrid molecules could not have been detected under the conditions used. The same group also claims to have detected regions in human metaphase chromosomes exhibiting complementarity with labelled RNA isolated from RD 114 virus'O-the virus announced in a blaze of publicity at a news conference as the prime candidate for the human cancer virus.*' Unfortunately seven papers which were published simultaneously by a total of no fewer than 22 people,22 came to the unanimous conclusion that RD 114 virus is of feline origin.It has also been pointed out2' that all this similar work was funded by the National Cancer Institute and that it would be interesting to know how much money has been wasted on this viral equivalent of a 'wild goose chase'. Viruses are however continuing to be found in association with tumour cells and other oncogenic cell lines. Rauscher leukaemia virus (RLV) RNA has been found associated with Burkitts lymphomas and nasopharyngeal carcinoma^.^^ The DNA-containing Epstein-Barr virus (EBV) has also been found associated with both these neopla~ia.~~*~~ In particular the RLV RNA is a 70s RNA25 and is associated with an RNA-instructed DNA polymerase in a particle having the density of an RNA tumour virus.However the genetic information contained in the RNA of the particle still has to be found in the DNA of the tumour cells and the aetiological role of EBV or RLV in the pathogenesis of Burkitts disease cannot be assumed. Hybridization studies have failed to show the presence of virus-related sequences in normal human cells which could be detected in human leukaemic cells.26 These experiments do not support the virogene-oncogene theory which postulates the inclusion of at least one complete copy of oncogenic information in the genomes of every normal cell and thus more Is S.Kowalski T. Yamane and J. R. Fresco Science 1971 172 385; S. H. Chang N. R. Miller and C. W. Harmon F.E.B.S. Letters 1971 17 265. l6 S. H. Chang S. Kuo E. Hawkins and N. R. Miller Biochem. Biophys. Res. Comm. 1973 51 951. K. W. Jones J. Prosser G. Corneo and E. Ginelli Chromosoma 1973 42 445. P. M. Price J. H. Conover and K. Hirschhorn Nature 1972 237 340. l9 J. 0.Bishop and K. W. Jones Nature 1972,240 149; W. Prensky and G. Holmquist Nature 1973 241. 44. 2o P. M. Price. K. Hirschhorn N. Gabelman and S. Waxman Proc. Nut. Acad. Sci. U.S.A. 1973,70 11. See Nature 1973 244 252. 22 See Nature New Biol. 1973 244 51-64; R. M. Ruprecht N. C. Goodman and S. Spiegelman Proc. Nut. Acad. Sci. U.S.A. 1973.70 1437. 23 D. Kufe R.Hehlmann and S. Spiegelman Proc. Nut. Acad. Sci. U.S.A. 1973 70 5. ?* H. Wolf H. zur Hausen and V. Becker Nature New Biol. 1973 244 245. 25 D. Kufe I. T. Magrath J. L. Ziegler and S. Spiegelman Proc. Nat. Acad. Sci. U.S.A. 1973 70 737. 26 W. G. Baxt and S. Spiegelman Proc. Nut. Acad. Sci. U.S.A. 1972 69 3737. iVucleic Acids 627 hopeful pathways leading to the control and cure of cancer can perhaps be entertained. Among the stranger reports this year is one which finds that patients with severe organic memory defects did not improve when given tRNA (no doubt microbial in rigi in).^' However a diet of DNA and ATP provides radioprotection (and gout?) for radiologists.28 The team working on the DNA content of the blood of a live coelacanth had little danger of being pre-empted2’ and finally there is a report that one team has been using its money to induce the production of monster ‘lollipops’ from T-even phages,30 no doubt as their contribution to the world protein shortage! 1 Bases Nucleosides and Nucleotides Treatment of thymine and uracil with trialkyl phosphates in molar proportions has been found to give high yields of the N-l-alkylated base.31 An excess of the phosphate results in the production of 1,3-dialkyl derivatives although the reac- tion rate decreases as the length of the alkyl chain increases.An improved pro- cedure for the synthesis of 9-alkyladenines has been reported.32 5-Vinyluracil has been prepared by no fewer than four different methods the first of which involves the decarboxylation of trans-3-(5-uracilyl)propenoicacid33 (l) but the overall yield is low (<20%).Base-catalysed elimination of esters of 5-(2-hydroxyethyl)uracil(2) failed3 as the substitution reaction was preferred. Attempts to dehydrate the alcohol under acidic conditions led to an intramol- ecular displacement reaction and the formation of the bicyclic compound (3). In asecondmethod ofpreparation,5-( l-hydroxyethyl)uraci1(4)can bedeh~drated~~ under acidic conditions to give 5-vinyluracil but the yield is low owing to the competing reaction resulting in the production of trans-1,3-bis(uracil-5-yl)but-l-ene (5). 5-Vinyluracil can also be prepared by the base-catalysed elimination of the methanesulphonyl ester of (4) in -80% yield,35 and a method involving ” A.Britton L. L. Bernstein A. J. Brunse M. W. Buttiglieri A. Cherkin J. H. McCor-mack and D. J. Lewis J. Geronroi. 1972,27 478. V. Goyanes-Villaescursa Lancet 1973 2 575. 29 K. S. Thomson J. G. Gall and L. W. Coggins Nature 1973 241 126. 30 D. J. Cummings V. A. Chapman S. S. Delong and N. L. Couse Virology 1973 54 245. 3’ K. Yamauchi and M. Kinoshita J.C.S. Perkin I 1973 391. 32 T. Fujii S. Sakurai and T. Uematsu Chem. and Pharm. Buil. (Japan) 1972 20 1334. 33 J. D. Fissekis and F. Sweet J. Org. Chem. 1973. 38 264. ” C. H. Evans A. S. Jones and R. T. Walker Tetrahedron 1973 29 161 1. 35 A. S. Jones G. P. Stephenson and R. T. Walker Nucieic Acids Res. 1974 1 105. 628 R.T. Walker synthesis of the pyrimidine ring36 has the sole merit that the starting materials are cheap as the yield is <10%.5-Vinyluracil has been incorporated into the DNA of a thymine-requiring mutant of Escherichia ~oli.~~ 5-Lithiouracil has been used as an intermediate in the preparation of several 5-alkylated uracil~.~~ Thiols particularly 2-mercaptoethanol and cy~teine,~~ dehalogenate 5-iodo- and 5-bromo-uracil under physiological conditions to give uracil while 5-bromo- 2’-deoxyuridine has been found to give 2’-deoxyuridine and s-[5-(2’-deoxyuridyl)]- cysteine (6):’ Thus 5-bromouracil may not be as stable as has previously been thought particularly when in bacterial growth media. The reaction is thought to proceed via addition of the thiol across the 5,6-double bond. 0 I 2'-Deoxyribosy l (6) A similar reaction has been used to label uracil by isotope exchange at the 5-position at a pH of 8.9 by cysteine-catalysed hydrogen-deuterium exchange.41 The reaction conditions are very mild and could be used at the nucleic acid level.The kinetics of hydrogen exchange at the 5-and 6-positions of uracil catalysed by Pt-black have been inve~tigated.~~ The 6-position of pyrimidine nucleosides can be labelled by treating them in DMSO solution with labelled base.43 In the cases where the sugar hydroxyls can easily participate in the saturation of the 5,6-double bond quantitative incorporation of the label at C-5 can also be obtained. The mechanism of H-6 exchange is thought to involve the direct 36 J. D. Fissekis and F. Sweet J. Org.Chem. 1973 38 1963. 37 E. T. J. Chelton C. H. Evans A. S. Jones and R. T. Walker Biochim. Biophys. Acta 1973 312 8. 38 D. M. Mulvey R. D. Babson S. Zawoiski and M. A. Ryder J. Heterocyclic Chem. 1973 10 79. 39 F. A. Sedor and E. G. Sander Biochem. Biophys. Res. Comm. 1973,50 328. 40 Y. Wataya and H. Hayatsu Biochemistry 1973 12 3992. 41 Y. Wataya H. Hayatsu and Y. Kawazoe. J. Biochem. 1973. 73 871. 42 T. Ido M. Tatara and Y. Kasida Internat. J. Appl. Radiation Isotopes 1973 24 81. 43 J. A. Rabi and J. J. Fox J. Amer. Chem. Sor. 1973 95 1628. Nucleic Acids 629 abstraction of H-6 by base. The reactions of thymine irradiated in aqueous solution in the presence of cy~teine,~~ and the photoaddition reactions of nucleo- philes to uracil have been studied.45 The photoaddition of bisulphite under these conditions takes place at much lower concentrations (lo-' moll-') of bisulphite than is required for the thermal addition and the addition is sterochemically random.The major product of y-irradiation of thymine in aerated aqueous solution is cis-5-hydroxy-6-hydroperoxy-5,6-dihydrothymine (7).46 Apart from the ionic reactions of bisulphite which take place more readily at high bisulphite concentrations (usually 1-3 moll-') bisulphite also reacts under free-radical conditions at concentrations of 10-* moll-'. These reactions require the presence of oxygen and are accompanied by the oxidation of bisulphite to sulphate. Several reactions of this type have previously been reported and it has now been used to convert 2-thiouracil into uracil in 60% yield.47 The bisulphite- oxygen system also results in the rapid cleavage of the glycosidic bond in uridine cytidine and their homopolynucleotides (with concomitant chain scission for the latter).48 The fate of the sugar moiety is unknown but it can be argued from the substrate specificity of the reaction that the initial attack is possibly that of a free radical on the 2'-OH group.The reaction is clearly different from that of the hydrogen peroxide or the hydroxylamine-oxygen system as deoxynucleosides purine glycosides and their phosphate esters do not react. People using bi- sulphite to modify nucleic acids chemically should be aware of this type of radical reaction which can occur unless steps are taken to prevent it.The reaction of adenine derivatives with chloroacetaldehyde to give the cor- responding 'etheno'-derivatives (E) first described by Kochetkov et ~1.:~and further exploited by Leonard and co-workers as described last year,50 has ob- viously appealed to many people. The fluorescent adenine derivatives so produced (8) are potential probes of enzymatic mechanisms and structure. The energy of &-AMP fluorescence has been shown not to be highly sensitive to the environment of the molecule so that large fluorescent shifts are unlikely. However thequantum yield of fluorescence is more sensitive and is likely to fall when the analogue is 44 '' A. J. Varghese Biochemistry 1973 12 2725. W. A. Summers C. Enwall J. G. Burr and R. L. Letsinger Phorochem.and Photobiol. 1973 17 295. 46 B. S. Hahn and S. Y. Wang Biochem. Biophys. Res. Comm. 1973,54 1224. 47 M. Sono and H. Hayatsu Chem. and Pharm. Bull. (Japan) 1973 21 995. 48 N. Kitamura and H. Hayatsu Nucleic Acids Res. 1974 1 75. 49 N. K. Kochetkov V. N. Shibaev and A. A. Kost Tetrahedron Letters 1971 1993. 50 R. T. Walker Ann. Reports (B) 1972 69 534. 630 R. T. Walker bound to a macromolecule.51 The kinetic and fluorescence behaviour of E-ADP and E-ATPwith the enzyme pyruvate kinase has been studied in detail.52 The analogues substitute very well for ADP and ATP in terms of reaction rates and the binding of the analogues to the enzyme can be observed by fluorescence polarization techniques. The same analogues have been used as substrates for photophosphorylation reactionss3 and inevitably the 3’,5‘-cyclic AMP analogue has been ~repared.’~ At a more complex level E-DPN’ and &-FAD have been prepared and their physicochemical and biological properties studied.” E-Adenylated glutamine synthetase has been used as an internal fluorescence probe for enzyme confor- mati~n.’~ Chloroacetaldehyde reacts with both adenine and cytosine bases in denatured DNA so that 1.5% of these residues are modified in 5 minutes at pH 4.5 and 53 0C.57The melting point of the DNA is lowered by 1.3 “Cfor each base modified per 100 base pairs which corresponds to a 2.8 kcal destabilizing free energy per mismatched base pair.Poly(rA) when treated with chloro- acetaldehyde can give a range of degrees of substitution of &-residues in the poly- mer.’* Highly substituted polymers with more than 80% .+A residues give no indication of a co-operative transition at acid pH.However the homopoly- nucleotide poly-( 1,N6-ethanoadenylic acid) has been prepared from E-ADP and polynucleotide phosph~rylase,’~*~~ but this polymer is claimed to possess an organized secondary structure at acidic but not at neutral PH.~’ Poly-(&-A) does not complex with poly(U) or poly(1). The polymer is more resistant to hydrolysis by pancreatic RNase and snake venom phosphodiesterase than the corresponding natural polynucleotide. 51 G. R. Penzer European J. Biochem. 1973 34 297. 52 J. R. Barrio J. A. Secrist Y.-H. Chien P. J. Taylor J. L. Robinson and N.J. Leonard F.E.B.S. Letters 1973 29 215. 53 Y. Shahak D. M. Chipman and N. Shavit F.E.B.S. Letters 1973 33 293. 54 G. H. Jones D. V. K. Murthy D. Tegg R. Golling and J. G. Moffatt Biochem. Biophys. Res. Comm. 1973 53 1338. 55 C.-Y. Lee and J. Everse. Arch. Biochem. Biophys. 1973 157 83. 56 P. B. Chock C. Y. Huang R. B. Timmons and E. R. Stadtman Proc. Nat. Acad. Sci. U.S.A. I973,70 312. 57 C. H.Lee and J. G. Wetmur Biochem. Biophys. Res. Comm. 1973,50 879. 58 R. F. Steiner W. Kinnier A. Lunasin and J. Delac Biochim. Biophys. Acta 1973,294 24. 59 H.Lehrach and K. H.Scheit Biochim. Biophys. Acta 1973,308 28. 60 B. Janik R. G. Sommer M. P. Kotik D. P. Wilson and R. J. Erickson Physiof. Chem. Phys. 1973,5 27. Nucleic Acids 631 Poly-(3,N4-ethenocytidylicacid)60 has been similarly prepared and it does not complex with poly(1).Poly-(E-C) does not significantly fluoresce. Fluorescent derivatives of cytidine have been made by reaction with pyridinium and quino- linium hydrazides at pH 4.2 and 37°C.61 The fluorescence of the modified cytidines showed structure and environment dependence. Compounds (9) and (lo) by their ultraviolet absorption and fluorescence emission characteristics Ribosyi (9) R = H (10) R = Ph present favourable possibilities for energy-transfer studies with other fluorescing molecules particularly in single-stranded oligo- and poly-nucleotides and nucleic acids. p-Nitrophenol has been found to activate the fusion of an acylated sugar with a purine derivative.62 In an attempt to clarify the mechanism of acidic catalysis of the fusion reaction p-nitrophenol was added to compete with the heterocyclic base for the acylated sugar.The result was the formation of a high yield of nucleoside at a lower temperature than usual. The a-anomer of the carbohydrate moiety gave no nucleoside at all but resulted in a quantitative yield of (11). When the acidic catalyst was omitted the reaction continued as before. However the phenol exerts no catalytic effect and has to be present in molar quantities. It was deduced that for activating agents of this type the purine has to be acti- vated by polarization bonding with the activating agent. It was also concluded that in the normal fusion reaction the catalyst resulted in the activation of the purine base (12).61 J. R. Barrio and N. J. Leonard J. Amer. Chem. SOC.,1973,95 1323. 62 M. Sekiya T. Yoshino H. Tanaka and Y. Ishido Bull. Chem. SOC.Japan 1973,46 556. 632 R.T. Walker -$A? 0-C-Me The useful intermediate in nucleoside synthesis 2,3-@isopropylidene-~- ribofuranosylamine has now been isolated in high yield in a stable crystalline form as the toluene-p-sulphonate (13) and has been used in the synthesis of nucleosides such as Sacetyluridine by reaction with ethyl-N-(a-acetyl-P-ethoxyacryloy1)carbamate (14)? Both a- and P-anomers are produced but their relative amounts depend upon the solvent used and the p-anomer cafi usually be obtained in crystalline form from the reaction mixture. M I O,SC,H,Me-p c=oI EtOCH=C -NHCO,E t 0 0 (14) MeXMe The first successful functionalization of a purine 2‘-deoxynucleoside is repor- ted.64 Previous attempts had failed owing to the acid- and heat-lability of purine 2’-deoxynucleosides but it was argued that substitution of strongly electronegative groups (trifluoroacetyl) on the sugar moiety would destabilize the postulated glycosyl carbonium ion intermediate and thus retard base cleavage.2’-Deoxyino- sine when treated with trifluoroacetic anhydride followed by thionyl chloride in refluxing methylene chloride gave 6-chloropurine-2-deoxyriboside(15)in 80% yield with only 8 % cleavage to hypoxanthine. Quantitative cleavage of the giyco- sidic bond results if 3’,5’-di-O-acety1-2’-deoxyinosine is used.This compound is a useful intermediate in the preparation of 6-substituted purine deoxynucleosides several of which are useful in cancer chemotherapy. ‘’ N. J. Cusack B. J. Hildick D. H. Robinson P. W. Rugg,and G. Shaw J.C.S. Perkin I 1973 1720. 64 M. J. Robins and G. L. Basom Canad. J. Chem. 1973,51 3161. Nucleic Acids 633 C-C-Linked p-D-ribofuranosyl nucleosides have been synthesized in four steps6’ [the yields were :good excellent almost quantitative and 65 %(ascending order?)] from 2,3-O-isopropylidene-~-ribofuranose. The preparation of 5-(2’,3’-O-isopropylidene-5’-O-trityl-~-~-ribofuranosyl)barb~turic acid (16) is re-ported. The ct :p ratio of the anomeric mixture of C-glycosides can be altered 0 0WONa 2’-Deoxyribosyi (15) Me OX0Me by prolongation of the reaction time,more of the p-isomer being produced as the time increases.The mechanism is explained in terms of the thermodynamically more stable p-(‘trans’)isomer being favoured in an equilibrium situation which is possible because of an ‘active’ proton on C-5. Nucleoside phosphates can be easily synthesized by the reaction of an inter- mediate formed between phosphorous acid mercuric chloride and N-methylimi- dazole with a suitably blocked nucleoside such as 2’,3’-O-isopropylideneadeno-sine.66 Yields are between 70 and 80% for all the common nucleotides. 8-Mercaptoadenosine nucleotides can be prepared by the reaction of the corres- ponding 8-bromo-compounds with NaSH in DMF-water at room ternperat~re.~~ 7-Methylguanine can be demethylated under conditions which should be suitable for tRNA by reaction with a new powerful nucleophilic reagent lithium 2-methylpropane-2-thiolate in hexamethylphosphoramide.68 The hazards of using diethyl pyrocarbonate as an enzyme inhibitor because of its reaction with the nucleic acid bases under physiological conditions have been re-empha~ized,~~ but to read the biochemical literature one can only conclude that either biochemists are still unaware of its effects or that they don’t care or don’t believe it.An interesting series of reactions of 2-acyloxyisobutyryl halides (17) with nucleosides has belatedly been de~cribed.~’.~~ Uridine reacts with (17a) in the 65 H. Ohrui and J. J. Fox Tetrahedron Letters 1973 1951. 66 H.Takaku Y.Shimada and H. Oka Chem. and Pharm. Bull. (Japan) 1973,21 1844. ”M. Ikehara E. Ohtsuka and S. Uesugi Chem. and Pharm. Bull. (Japan) 1973,21,444. 68 S. M. Hecht and J. W. Kozarich J.C.S. Chem. Comm. 1973 387. 69 A. Vincze R.E. L. Henderson J. J. McDonald and N. J. Leonard J. Amer. Chem. SOC.,1973,95,2677; R.E. L. Henderson L. H. Kirkegaard and N. J. Leonard Biochim. Biophys. Acta 1973 294 356. ’O S. Greenberg and J. G. Moffatt J. Amer. Chem. Sac. 1973 95,4016. ” A. F. Russell S. Greenberg and J. G. Moffatt J. Amer. Chem. SOC.,1973 95 4025. 634 R. T. Walker absence of solvent to give (18a) and in acetonitrile to give (18b) both of which on treatment with sodium methoxide give 2’-chloro-2‘-deoxyuridinein high yield.7 O The reaction proceeds by a rapid intramolecular participation of the C-2 carbonyl group of the uracil ring with the initial acetoxonium ion (19) giving the proto- nated species (20) which is then opened up by chloride ion to give (21).Me \ Me /I OAc c-cox OAc C1 (17) a; X = C1 b;X = Br II I(18) a; R = C-C-Me 0 OAc ROHZ OAc CI Adenosine when treated with (17a) in acetonitrile gives (22a) which on de- blocking gives 3’-deoxy-3’-chloro-~-~-xylofuranosyladenos~ne.~~ The reaction goes via the cyclic acetoxonium ion which in the absence of participation from the base is opened by chloride ion which on steric grounds prefers attack at C-3’. Reaction of (22a) with sodium methoxide gives (23). If the acyl bromide (17b) is used in the initial reaction a similar product (22b) is formed which can be catalytically debrominated to give 2,3’-dideoxyadenosine and 3’-deoxy-adenosine (Cordycepin) each in 40% yield.Deacetylation of (22a) followed by catalytic debromination results in a high yield of 3’-deoxyadenosine only. Similar reactions have been used to produce analogues of the drugs tubercidin and formycin. 2’,3’-O-Methoxyethylideneadenosine,when heated under reflux with pivalic acid chloride in pyridine gives a mixture from which (24a) could be isolated. On treatment with sodium methoxide 2’,3’-anhydroadeno- sine (23) was formed. Hydrogenolysis of the iodo-ester (24b) gives the correspond- ing 3’-deoxyadenosine derivative and treatment of (Nb) with the non-saponifying 72 T. C. Jain A. F. Russell and J. G. Moffatt J.Org. Chem. 1973 38 3179. 73 M. J. Robins R.Mengel and R.A. Jones J. Amer. Chem. SOC.,1973,95,4074. Nucleic Acids 635 0 \\O Me I OAC (22) a; X = C1 b;X = Br NHR I R = COCMe (24) a; X = C1 b;X=I base 1,5-diazabicyclo[4,3,0]non-5-enegives the corresponding 3’-ene-derivative. 5’-Deoxy-5’-halogenouridines can be prepared by the reaction of 2‘,3’-0-iso- propylidene-02,5’-cyclouridinewith acyl halides at room temperat~re.~~ The syntheses of 2’-0- 3’-0- and 5’-0-benzylcytidine have been described75 and the 2’-0-and 3’-O-benzyl ethers of the four common ribonucleosides can be prepared in a one-step reaction of the free nucleoside with phenyldiazomethane in the presence ofa Lewis acid catalyst.76 2’-O-Methyluridine can be synthesized directly by standard procedures from the carbohydrate l-O-acetyl-3,5-di-O- benzoyl-2-O-methyl-Q-~-ribofuranose.~ Mono- di- and tri-0-alkyl derivatives of cytidine and uridine can be prepared using dialkyl sulphates in a strongly alkaline medium.78 Treatment of nucleoside 3’,5’-cyclic phosphates with an alkyl iodide followed by cleavage of the phosphate enzymatically or chemically gives 2’-O-alkyl 3’(5’)-nucleotides in high yield.79 The 5’-nucleotides can be further phosphorylated to the 5’-diphosphates which are substrates for poly- nucleotide phosphorylase in the preparation of poly 2’-O-alkyl nucleotides.A synthesis of 8-bromo-2’-O-tosyladenosine illustrates the use of the 5‘-carboxylate group as a protecting agent.*’ Its production from and regeneration 74 Y.Fujisawa and 0. Mitsunobu J.C.S. Chem. Comm. 1973,201. 75 W. Hutzenlaub and W. Pfieiderer Chem. Ber. 1973 106 665. 76 L. F. Christensen and A. D. Broom J. Org. Chem. 1972 37 3398. ” A. H. Haines Tetrahedron 1973 29 2807. 78 J. T. Kusmierek J. Giziewicz and D. Shugar Biochemistry 1973 12 194; J. Giziewicz and D. Shugar Acta Biochim. Polon. 1973 20 73. 79 I. Tazawa S. Tazawa J. L. Alderfer and P. 0.P. Ts’o Biochemistry 1972 11 4931. B. R. Schmidt R. Machat and U. Scholz Chem. Ber. 1973 106 1256. 636 R. T. Walker to the 5’-hydroxymethyl group is nearly quantitative and its use is claimed to be more effective than that of the conventional 5’-blocking groups. The 5’-nitrates of uracil- and cytosine-containing nucleosides have been prepared by the action of 90 % nitric acid at 70 “C for 1.5 h on the free nucleosides.81 Many hundreds of papers on the synthesis of new potential drugs have been published.Among the most interesting is the synthesis of 1,2-dihydro-1- (~-deoxy-~-~-erythro-pentofuranosy~)-~-oxopyraz~ne(25) which has 4-oxide been shown to be a potent deoxyuridine analogue.’* The base moiety is an anti- bacterial compound in its own right but the deoxynucleoside is lo5times more effective an inhibitor of Streptococcus faecium and E. coli than either the base or the ribonucleoside. If the weight of publications dealing with Virazole (1-P-~-ribofuranosyl-1,2,4-triazole-3-carboxamide) (26) is any g~ide,~~-~ this compound could be very 0 0 t II 0111 I I 2’-Deoxyribosy1 Ribosyl useful to the medical profession as it has broad-spectrum antiviral proper tie^.^^ X-Ray crystallographic data indicate that Virazole is similar in structure to g~anosine,~~ and results of biochemical experiments to determine its mechanism of action suggest that this is due to the inhibition of GMP biosynthesis at the step involving the conversion of IMP into xanthosine S-pho~phate.~~ 02,2’-Cyclocytidine has antitumour activitys6 due it is thought to its trans- formation in solution into ma-cytidine and it can be synthesized in 33% yield by the action of toluene-p-sulphonyl chloride on cytidylic acid.8 3’,5‘-Cyclic ma-CMP is also an active antiviral compound and in oiuo is more reactive than ara-C probably because it cannot be deaminated by deoxycytidylic acid de- aminase which is responsible for the loss of activity of ~ra-C.~~ Attempts are being made to synthesize derivatives of biologically active compounds which can F.W. Lichtenthaler and H. J. Muller Angew. Chem. Internat. Edn. 1973 12 752. 82 P. T. Berkowitz T. J. Bardos and A. Block J. Medicin. Chem. 1973 16 183. 83 J. H. Huffman R. W. Sidwell G. P. Khare J. T. Witkowski L. B. Allen and R. K. Robins Antimicrob. Agents Chemother. 1973 3 235. 84 P. Prusiner and M. Sundaralingam Nature New Biol. 1973,244 116. 85 D. G. Streeter J. T. Witkowski G.P. Khare R. W. Sidwell R. J. Bauer R. K. Robins and L. N. Simon Proc. Nut. Acad. Sci. U.S.A. 1973,70 1 174. 86 A. Hoshi M. Yoshida F. Kanzawa K. Kuretani T.Kanai and M. Ichino Chem. and Pharm. Bull. (Japan) 1972 20 2286. M. Ikehara and S. Uesugi Chem. and Pharm. Bull. (Japan) 1973 21 264. R. W. Sidwell L. N. Simon J. H. Huffman L. B. Allen R. A. Long and R. K. Robins Nature New Biol. 1973 242 204. Nucleic Acids 637 penetrate cells and liberate the compound once inside the cell over a period of time. 3’,5’-Cyclic phosphoramidates have been synthesized but these appear to be too stable.89 The water-soluble N6-dimethylaminomethyleneanalogue of ara-A has been used to provide the relatively insoluble ara-A in a sustained fashion.’O 2 Oligonucleotidesand Polynucleotide Analogues The year has been one of consolidation for those working on oligonucleotide synthesis with the methods becoming more widely used and the technique of isolation” and the condensation yields 92 having been improved.The announce- ment of the synthesis of the gene for tRNATy‘ from Escherichia coli is still awaited but the aim is now the synthesis of not only the DNA corresponding to the pre- cursor tRNA but also that for the promoter and terminator regions as well. Much interest has centred on the development of new methods for sequencing oligonucleotides on a micro scale. Low-resolution mass spectrometry has been used to determine the sequence of oligodeoxynucleotides.93 The 3’- and 5’- termini of the oligonucleotides can be identified and in combination with limited enzymatic digestion the sequence of a hexanucleotide has been obtained. A continuous directional degradation in the 3‘ +5’ direction for polyribonucleo- tides involves incubating the polynucleotide (10-’moll-’) with periodate and alkaline phosphatase at an alkaline pH.94 Dialdehyde derivatives are sequentially liberated from the 3’-terminus and these are removed and reduced with [3H]KBH4.Successive treatment of oligodeoxyribonucleotides with phosphatase and then phosphodiesterase can give information about the chain length base composition and identity of the terminal nu~leotides.~~ Polynucleotide kinase can be used to label the 5’-terminus of oligo- rib^-,^^ and de~xyribo-~’ nucleotides. Polynucleotide phosphorylase has been used to label the 3’-end of oligoribonucleotides either by phosphorolysing the oligo- nucleotides in the presence of [32Pi] to nucleoside diphosphates9* in a non- processive manner or by using oligoribonucleotides obtained from a T1 digestion as primers for the reaction of primer-dependent polynucleotide phosphorylase 89 R.B. Meyer D. A. Shuman and R. K. Robins Tetrahedron Letters 1973 269. 90 S. Hanessian J. Medicin. Chem. 1973 16 290. P. J. Cashion M. Fridkin K. L. Agarwal E. Jay and H. G. Khorana Biochemistry 1973,12 1985; N. Katagiri C. P. Bahl K. Itakura J. Minchniewicz and S. A. Narang J.C.S. Chem. Comm. 1973 803. 92 Y.A. Berlin 0. G. Chakhmakhcheva V. A. Efimov M. N. Kolosov and V. G. Korobko Tetrahedron Letters 1973 1353. 93 J. L. Wiebers Analyt. Biochem. 1973 51 542. 94 K. Randerath F.E.B.S. Letters 1973,33 143. 95 N. W. Y. Ho and P. T. Gilham Biochim.Biophys. Acta 1973 308 53. 96 M. Simsek J. Ziegenmeyer J. Heckman and U. L. RajBhandary Proc. Nut. Acud. Sci. U.S.A. 1973,70 1041. 97 K. Murray Biochem J. 1973 131 569. 98 G. Kaufmann H. Gosfeld and U. Z. Littauer F.E.B.S. Letters 1973 31 47. 638 R.T. Walker with [LZ-~~PIGDP This results in the 3'-hydroxy- in the presence of T1 RN~s~.~~ group of each fragment becoming phosphorylated (Scheme). This method99 and the previously mentioned kinase method" for the labelling of the 5'-end of PY A G OH Phosphatase HO OH P' HO Py = pyrimidinyl Scheme oligonucleotides ought to complement each other in the sequence determination of tRNAs from a wide variety of sources which cannot be sufficiently labelled in uiuo so that a complete sequence might be obtainable with only 2mg of tRNA.Much work continues on oligonucleotide synthesis on a polymer support but the first requirement seems to be a knowledge of German! The search for a suit- able support continues and candidates this year have included highly cross- linked polystyrene-containing supports,"' soluble polymers,"' and poly-peptides."* Although the polymer-support methods suffer from low yields at the condensation steps which lead to chains of different lengths and sequences being produced it is now realized that unlike polypeptide synthesis only short oligonucleotides need be prepared by chemical synthesis as these can be joined enzymatically. It is also possible to separate the different short-chain oligomers synthesized.The advantages of the method include the ease of manipulation of reactants and the speed at which successive condensations can be performed. A new approach to the sequential synthesis of oligodeoxyribonucleotides involves the selective removal of an acid-labile 5'-protecting group phosphory- lation with POCl in 2,6-lutidine and tetrahydrofuran followed by condensation with a deoxynucleoside containing a free 3'-OH group and an acid-labile 5'-protecting group.lo3 Other oligodeoxynucleotides to be synthesized include an octanucleotide complementary to a section of 16s rRNA,lo4 four undecanucleotides comple- 99 K. S. Szeto and D. So11 Nucleic Acids Res. 1974 1 171. 100 R. Glaser U. Sequin and C. Tamm Helv. Chim. Acta 1973 56 654; H. Koster and F.Cramer Makromol. Chem. 1973 167 171; V. K. Potapov V. V. Zvezdina M. N. Kochetkova Z. A. Shabarova and M. A. Prokofiev Doklady Akad. Nauk S.S.S.R. 1973,209 364. 101 F. Brandstetter H. Schott and E. Bayer Tetrahedron Letters 1973 2997; H. Seiiger and G. Aumann ibid. p. 291 1. 102 T. M. Chapman and D. G. Kleid J.C.S. Chem. Comm. 1973 193. 103 H. Koster and W. Heidmann Angew. Chem. Internat. Edn. 1973 12 859. I04 V. N. Kagramanov V. D. Smirnov A. A. Bogdanov Z. A. Shabarova and M. A. Prokofiev Doklady Akad. Nauk S.S.S.R. 1973 208 858. Nucleic Acids 639 mentary to a region of the coat protein cistron of fd phage,lo5 and a dodecanu- cleotide complementary to a section of the lysozyme gene of phage T4.'06 It is reportedlo7 that only the 2'-O-(a-methoxyethyl)nucleoside5'-diphosphates are substrates for polynucleotide phosphorylase in the enzymatic synthesis of oligoribonucleotides described two years ago.O8 A chemical synthesis of poly- riboguanylic acid is rep~rted,''~ although it appears that 84 % of the products have a chain length of between four and six units and it hardly qualifies for the title 'poly'. The 3'-terminal nonanucleotide and 5'-terminal hexanucleotide of yeast tRNAA'" have been synthesized by conventional techniques.' ' ' The triester approach to oligonucleotide synthesis is gradually gaining favour as the advantages of the method particularly the ease of isolation of the products are realized. A hexadecamer of thymidylic acid has been prepared" ' and the techniques used are now almost at the stage where they can be applied to the synthesis of oligoribonucleotides of defined sequence in high yield.The phenyl group is used to make the phosphotriester."' Other people have used the tri-chloroethyl group,' although this becomes increasingly more difficult to remove as the length of the chain increases. In the analogue field poly-( 1-vinyluracil) can complex with poly(A) and com- pletely block poly(A)-stimulated formation of polylysine in an E. coli subcellular protein synthesis system.'13 Poly-(3'-O-carboxymethyl-2'-deoxyadenosine)'14 has been shown to have a similar effect to poly-(9-vinyladenine)' l3 and reduces the poly(U)-stimulated formation of polyphenylalanine. Poly-(1-vinyluracil) has been found to inhibit the in uitro translation of rabbit globin mRNA pre- sumably by interacting with the poly(A) sequences on the mRNA as poly- (9-vinyladenine) has no effect.' l5 Both vinyl polymers inhibit acute murine leukaemia virus infection in mouse embryo cells.' l6 3 Photochemistry and Physical Chemistry of Polynucleotides There is an increasing interest in the influence of conformation of a polynucleotide chain upon its photochemistry and the related subject of how the presence of different photoproducts alters the conformation of polynucleotide structures.From a comparison of the effects of pyridimine dimerization and ethidium bromide H. Schott and H. Kossel J. Amer. Chem. SOC.,1973 95 3778. lo6 M. T. Doe1 and M. Smith F.E.B.S. Letters 1973 34 99. lo' G.N. Bennett J. K. Mackey J. L. Wiebers and P. T. Gilham Biochemistry 1973 12 3956. R. T. Walker Ann. Reports(B) 1971 68 420. log D. B. Strauss and J. R. Fresco .I.Amer. Chem. Soc. 1973,95 5025. 'lo E. Ohtsuka M. Ubasawa S. Morioka and M. Ikehara J. Amer. Chem. SOC.,1973,95 4725. I" N. J. Cusack C. B. Reese and J. H. van Boom Tetrahedron Letters 1973 2209. 'I2 J. C. Catlin and F. Cramer J. Org. Chem. 1973 38 245. 'I3 F. Reynolds D. Grunberger J. Pitha and P. M. Pitha Biochemistry 1972 11 3261. 'I4 A. S. Jones M. MacCoss and R. T. Walker Biochim. Biophys. Acta 1973 294 365. 'Is F. Reynolds D. Grunberger P. M. Pitha and J. Pitha in the press. P. M. Pitha N. M. Teich D. R. Lowy and J. Pitha Proc. Nat. Acad. Sci. U.S.A. 1973 70 1204. 640 R.T.Walker intercalation on the sedimentation coefficient of superhelical DNA,' '' it has been concluded that each dimer unwinds the helix by 54".From the effect of a central thymine photodimer on the stability of the double helix dAlo.dTlo it has been estimated that each dimer disrupts four base pairs those of the dimerized thymines and those of their immediate neighbours.' '* The influence of the conformation of poly(U) and poly(C) on their photochemistry has been studied' l9 and the photo- hydration and photodimerization reactions of uracil have been used to charac- terize the local environment of extra-helical uridine residues in the copolymer strand of poly(AU).poly(U) helices.'2o The occurrence and stability of cytosine photohydrates in DNA have been studied12' using a newly developed reduction assay.These lesions are sufficiently stable to interfere with the in viuo function of DNA. Hitherto purines in comparison with pyrimidines have proved very resistant to photochemical change and accordingly have been considered to have little importance in photobiology. The discovery' 22 that poly(dA) [but not poly(A)] undergoes a specific photoreaction with high quantum efficiency upon irradiation at 248 nm should herald a new interest in the photochemistry of purines. The contrast in the reactivity of the two polyadenylic acids is ascribed to the known difference in their single-stranded conformations. Investigations into the effects of heavy atoms on the photochemistry of nucleic acids have been initiated.Thymine dimerization and biological inactivation of transforming DNA are enhanced by complexed silver ions'23 but reduced by complexed mercuric ions.' 24 The enzymology of cellular mechanisms for repairing photolesions in DNA is being actively explored. Photoreactivation which is the simplest cellular repair mechanism is achieved by an enzyme which monomerizes pyrimidine photodimers in DNA upon illumination at wavelengths of ca. 400nm. The photoreactivating enzymes from blue-green alga' 25 and yeast' 26 have been extensively purified and a fluorescent cofactor has been implicated in their func- tion. Now that milligram amounts of the E. coli en~yrne'~'can be prepared from a strain carrying the photoreactivation gene as part of a transducing phage the detailed characterization of this remarkable enzyme will be possible.'H N.m.r. (220 and 300 MHz) has been used to study hydrogen-bonded ring protons in Watson-Crick base pairs in oligonucleotide model systems,' 28 11' D. T. Denhardt and A. C. Kato J. Mol. Biol. 1973 77 479. 118 F. N. Hayes D. L. Williams R. L. Ratliff A. J. Varghese and C. S. Rupert J. Amer. Chem. SOC. 1971 93,4940. A. J. Lomant and J. R. Fresco J. Mol. Biol. 1972 66 49. A. J. Lomant and J. R. Fresco J. Mol. Biol. 1973 77 345. Iz1 J. Y. Vanderhoek and P. A. Cerutti Biochem. Biophys. Res. Comm. 1973,52 1156. D. Porschke Proc. Nut. Acad. Sci. U.S.A. 1973 70 2683. Iz3 R. 0.Rahn and L. C. Landry Photochem. and Photobiol. 1973 18 29. lz4 R. 0.Rahn J. K. Setlow and L. C. Landry Photochem.and Photobiol. 1973 18 39. * '' S. Minato and H. Werbin Photochem. and Photobiol. 1972 15 97. lZ6 J. S. Cook and T. E. Worthy Biochemistry 1972 11 388. l '' B. M. Sutherland M. J. Chamberlin and J. C. Sutherland Biochemistry 1973,12,4200. 12* D. M. Crothers C. W. Hilbers and R. G. Shulman Proc. Nut. Acad. Sci. U.S.A. 1973 70 2899. Nucleic Acids 641 5s RNA,12' and tRNA.'30-'34 tRNA?" from bakers' yeast in going from the denatured to the native form gains 3-5 G-C pairs and loses 0-2 A-U pairs.'30 No evidence can be found for additional Watson-Crick base pairs apart from those occurring in the clover-leaf In a detailed study of tRNAPhe from yeast the CCA stem is shown to form a continuous helix with the T+C stem and this result agrees with the structure determined from X-ray studies.'33 No difference in conformation between the charged and uncharged tRNAPhe can be detected and it is concluded that the secondary and tertiary structures of these molecules are probably the same.134 Conformational changes in tRNA have also been investigated by temperature-jump techniques.'35 Polynucleotide complexes involving non-Watson-Crick base-pairing and possibly left-handed helices have been described. Poly-(2-dimethylaminoadenylic acid) forms hydrogen bonds through N6and N-7 with poly(U) and poly-(5- brom~uracil).'~~ Wobble base-pairing between G and T occurs in the stable helical secondary structure of poly(dT-G).' 37 The hydrogen-bonding pattern in the hairpin secondary structure of poly(U) has been established by i.r.spectro- scopy.' 38 Poly-(2-thiouridylic acid) forms a similar but much more stable hairpin helix.' 39 Although the spectroscopic and hydrodynamic properties of the alter- nating copolymer poly[d(A-s4T).d(A-s4T)] are consistent with a left-handed double helix incorporating reversed Hoogsteen base-pairing between adenine and 4-thi0thyrnine,'~~ this is not supported by a small-angle X-ray scattering study.141 Poly-(2-methyladenylic acid) forms a 1 1 complex with poly(U) to give a double-stranded complex with Hoogsteen-type h~dr0gen-bonding.l~~ Oligonucleotides containing 6,2'-anhydro-6-oxy-1-/3-~-arabinofuranosyluracil (Uo)have been prepared143 and have been found to give a 1 :1 complex with oligo 8,2'-S-cycloadenosine (A').The duplex probably has a left-handed con- formation and since poly(A') could not form complexes with poly(U) nor could poly(Uo) form complexes with poly(A) it seems that the torsion angle of the bases has to be identical before a double-stranded complex can form. Anomalous 0.r.d. lZ9 Y. P. Wong D. R. Kearns B. R. Reid and R. G. Shulman J. Mol. Biol. 1972,72 741. I3O Y. P. Wong D. R. Kearns R. G. Shulman T. Yamane S. Chang J. G. Chibikjian and J. R. Fresco J. MoI. Biol. 1973 74 403. '" R. G. Shulman C. W. Hilbers D. R. Fearns B. R. Reid and Y. P. Wong J. Mol. Biol. 1973 78 57. 132 D. R. Lightfoot K. L. Wong D. R. Kearns B. R. Reid and R. G. Shulman J. Mol. Biol. 1973 78 71. 133 R. G. Shulman C. W. Hilbers Y. P. Wong K. L. Wong D. R. Lightfoot B.R. Reid and D. R. Kearns Proc. Nut. Acad. Sci. U.S.A. 1973 70 2042. 134 Y. P. Wong B. R. Reid and D. R. Kearns Proc. Nat. Acad. Sci. U.S.A. 1973,79,2193. 13' P. E. Cole and D. M. Crothers Biochemisrry 1972 11 4368; S. K. Yang and D. M. Crothers. ibid.. p. 4375. 136 F. Ishikawa J. Frazier F. B. Howard and H. T. Miles J. Mol. Biol. 1972 70 475; F. Ishikawa J. Frazier and H. T. Miles Biochemistry 1973 12 4790. 13' A. G. Lezius and E. Domin Nature New Biol. 1973 244 169. 13' D. Bode M. Heinecke and U. Schernau Biochem. Biophys. Res. Comm. 1973,52,1234. W. Bahr P. Faerber and K. Scheit European J. Biochem. 1973,35 535. I4O E. M. Gottschalk E. Kopp and A. G. Lezius European J. Biochem. 1971 24 168. 14' P. Zipper European J. Biochem. 1973,39,493. 14' M.Ikehara M. Hattori and T. Fukui European J. Biochem. 1972 31 329. '43 M. Ikehara and T. Tezuka J. Amer. Chem. Soc. 1973,95,4054. 642 R.T. Walker spectra suggestive of unusual polynucleotide conformation are shown by the complexes of poly(A) with formycin and 7-methyl~anthine.'~~ X-Ray diffraction studies have shown the triple-stranded polynucleotides p~ly(U).poly(A).poly(U)'~~and poly(I).poly(A).poly(I)'4' to have diameters only very slightly greater than those of the corresponding double helices. In agreement with other physical studies,'48 the two poly(U) [and poly(I)] strands are antiparallel. The properties of the triplex dT .dA .rU have been investi- gated149 and the structures of p~ly(I),'~~ poly-(1-deaza- 3-deaza- and 7-deaza- adenylic acid),I5 'and poly-(7-deazainosinic acid)' 52 determined.A method for estimating the most stable secondary structure of an RNA mole- cule from its sequence has been refined and used to predict the secondary structure of a 55-base fragment of R17 RNA.ls3 The free energy of the RNA conformation is considered to be determined by contributions from helical base-paired regions internal loops bulge loops hairpin loops and single-stranded regions. Recent thermodynamic investigations of model systems incorporating internal loops,' 54 bulge 100ps,'55 and hairpin 100ps'~~*'~' have allowed the contributions of these features to the over-all free energy of the nucleic acid structure to be estimated more accurately. However a study of the thermodynamic kinetic and optical properties of the double helices formed by the series of self-complementary oligonucleotides (Ap),GpC(pU) shows that the shortest helix containing just six base pairs is much less stable than would be predicted from the properties of larger rn~lecules.'~~ Apparently one cannot assume that base-pair free energies and extinction-coefficient changes are independent of size.The temperature-jump relaxation method has been used to study the transition in a variety of oligonucleotides containing G-C base pair^."^+'^^ The initial nucleation of about three base pairs is rate-limiting for helix growth in oligo- nucleotides containing A-A A-U and A-T base pairs. Once such a stable nucleus has formed the helix 'zips up' at a rate of 10'-lo8 base pairs per second.With the oligonucleotides containing G-C base pairs formation of one or perhaps two base pairs is sufficient for rapid helix growth and the rate of reaction 144 R. J. H. Davies J. Mol. Biol. 1973 73 317. 14' R. J. H. Davies Biochem. Biophys. Res. Comm. 1973 52 1115. 146 S. Arnott and P. J. Bond Nature New Biol. 1973 244 99. 14' S. Arnott and P. J. Bond Science 1973 181 68. 148 J. Thrierr and M. Leng Biochim. Biophys. Acta 1972 272 238. 149 N. L. Murray and A. R. Morgan Canad. J. Biochem. 1973,51,436. 'O D. Thiele and W. Guschlbauer Biophysik 1973 9 277. L. Hagenberg H. G. Gassen and H. Matthaei Biochem. Biophys. Res. Comm. 1973 50 1104. lS2 M. Ikehara T. Fukui T. Koide and J. Inaba Nucleic Acids Res. 1974 1 53. I. Tinoco P.N. Borer B. Dengler M. D. Levine 0.C. Uhlenbeck D. M. Crothers and J. Gralla Nature New Biol. 1973 246 40. J. Gralla and D. M. Crothers J. Mol. Biol. 1973,78 301. lS5 T. R. Fink and D. M. Crothers J. Mol. Biol. 1972,66 1. lS6 0.C. Uhlenbeck P. N. Borer B. Dengler and I. Tinoco J. Mol. Biol. 1973 73 483. Is' J. Gralla and D. M. Crothers J. Mol. Biol. 1973,73 497. "* J. Ravetch J. Gralla and D. M. Crothers Nucleic Acids Res. 1974 1 109. S. K. Podder European J. Biochem. 1971 22,467. D. Porschke 0. C. Uhlenbeck and F. H. Martin Biopolymers 1973 12 1313. Nucleic Acids 643 is markedly dependent upon the base sequence. The importance of single- stranded stacking interactions in determining the stability of these complexes has been stressed’60 and it is therefore appropriate that investigations of the dynamics of the conformational changes of single-stranded polynucleotides have been initiated with a study of oligoadenylates and poly(A).16’ 4 tRNA tRNA is still the only readily available source of a homogeneous nucleic acid and thus there are many hundreds of papers each year dealing with various aspects of the chemistry and biochemistry of these molecules.Happily the number of such reports dealing with unfractionated tRNA is now very small and there really is little excuse for not first fractionating the tRNA now that an RPC-5 column (1 1) has been used to fractionate 1.5g of a partially-purified tRNA in 4.5 h with a flow rate of 10ml min-1.’62 One of the most interesting experiments reported this year has at last solved the problem as to which hydroxy-group of the 3’-terminal adenosineof tRNA is aminoacylated.Using nucleotidyl transferase 2‘-and 3’-deoxyadenosine were in turn incorporated into this terminal position in the tRNAPhe from yeast.’63 Only the 3’-deoxyadenosine-terminated molecule could be aminoacylated (i.e.on the 2’-OH group) but the product could not take part in protein ~ynthesis.’~~ The tRNA terminating in 2‘-deoxyadenosine could not be aminoacylated but acts as a competitive inhibitor of the phenylalanyl-tRNA synthetase and it can be concluded that tRNA is normally aminoacylated on the 2’-OH group which is followed by a rapid acyl-group migration to the 3’-position before it can be active in protein biosynthesis.3’-Deoxy-3’-aminoadenosine has also been incor- porated into the 3’-terminal position of tRNAPh‘ from yeast.’65 This tRNA can be aminoacylated; the amino-acid is attached to the 3’-NH2 group and the product is very stable. Thus the tRNA has acceptor but not donor activity and the kinetics for the aminoacylation are the same as for a normal tRNA which is no doubt due to the fact that it is the acylation of the common 2’-OH group which is the rate- determining step. It has been shown that mischarged initiator tRNAs such as Phe-tRNAp‘ can be formylated showing that the formylation depends upon the tRNA and not the amino-acid.’66 Several more tRNA sequences have been published during the year and they include for E. coli tRNAAsp,167 tRNAggG,168 and tRNAS,”;16’ for yeasts 16’ D.Porschke European J. Biochem. 1973 39 117. ’‘* B. Roe K. Marcu and B. Dudock Biochim. Biophys. Am 1973 319 25. M. Sprinzl K. H. Scheit H. Sternbach F. von der Haar and F. Cramer Biochem. Biophys. Res. Comm. 1973,51 881. 164 M. Sprinzl and F. Cramer Nature New Biol. 1973 245 3. 165 T. H. Frazer and A. Rich Proc. Nat. Acad. Sci. U.S.A. 1973 70 2671. R. Giege J. P. Ebel and B. F. C. Clark F.E.B.S. Letters 1973 30,291. F. Harada K. Yamaizumi and S. Nishimura Biochem. Biophys. Res. Comm. 1973 49 1605. C. W. Hill G. Combriato W. Steinhart. D. L. Riddle and J. Carbon J. Bioi. Chem. 1973 248,4252. 169 (u)D. Ish-Horowicz and B. F. C. Clark J. Bid. Chem. 1973,248,6663;(6)Y.Yamada and H. Ishikura F.E.B.S. Letters 1973. 29 231.644 R.T. Walker tRNALys,190 tRNAGIy,' 71 and tRNA;4i'g;172 for rabbit liver tRNAPhe for Salmonella typhimurium tRNAGIy;168 for phage T4 coded tRNAG'y,' 74 tRNALe",'75 and tRNASer.176 When bacteriophage T4177 (and probably T2 and T6)178infect E. coli,the phage genome directs the synthesis of at least eight tRNAs and two stable RNA species of low molecular weight whose function is unknown. Two groups of workers have determined the sequences of several of these tRNAs and have compared them with those of the corresponding E. coli tRNAs. In particular tRNAGIY from the phage probably recognizes both GGG and GGA and contains a 5-methylamino-2-thiouridineresidue at the 5'-end of the anticodon.' 74 This tRNA has a very high mobility on gel electrophoresis suggesting a chain length of only 66 nucleotides but in fact the molecule contains 74 nucleotides and it may have a more compact shape than other tRNAs.The T4 system has been used to synthesize tRNA in uitro in a p~rified'~' and in a crude system.'80 The tRNAs are identical to those synthesized in uiuo except for the absence of modified bases. An RNA molecule of 140 nucleotides has also been sequenced from T4- infected cells and the nucleotides 46-140 can be arranged in the form of a clover leaf with a CCA 3'-termin~s.'~~ Precursors of the T4-coded tRNAs have also been found which consist of polynucleotide chains containing two tRNA species. At least six of the tRNAs can be accounted for in three such dimeric species.'81 Eukaryotic initiator tRNA from mouse myeloma cells lacks Tp$p in loop IV,18* and in tRNAp' from wheat germ,96 rabbit li~er,'~*'~~ and sheep mammary gland,96 the normal GT$CG is replaced by a sequence GAUCG so that the lack of rT is not because of the non-modification of a U residue.An oligonucleotide fragment from the 3'-terminus to the 7-methylguanosine residue of these tRNAs can be obtained following a specific mild alkaline cleavage at m7Gand separation of the products. The base sequences in loop IV of yeast mouse myeloma rabbit liver wheat germ and sheep mammary gland are identical'84 and it is probable 170 C. J. Smith H.-S. Teh A. N. Ley and P. D'Obrenan J. Biol. Chem. 1973 248,4475. M. Yoshida Biochem. Biophys. Res. Comm. 1973 50 779. J. Weissenbach R.Martin and G. Dirheimer F.E.B.S. Letters 1972 28 353. G. Keith F. Picaud J. Weissenbach J. P. Ebel G. Petrissant and G. Dirheimer F.E.B.S.Letters 1973 31 345. 174 S. Stahl G. Paddock and J. Abelson Biochem. Biophys. Res. Comm. 1973 54 567; B. G. Barrell A. R. Coulson and W. H.McClain F.E.B.S. Letters 1973 37 64. 175 T. C. Pinkerton G. Paddock and J. Abelson J. Biol. Chem. 1973 248,6348. 176 See reference 169a. 177 W. H. McClain C. Guthrie and B. G. Barrell Proc. Nut. Acad. Sci. U.S.A. 1972 69 3703. 178 G. Paddock and J. Abelson Nature New Biol. 1973 246 2. 179 D. P. Nierlich H. Lamfrom A. Sarabhai and J. Abelson Proc. Nut. Acad. Sci. U.S.A. 1973,70 179. H.Lamfrom A. Sarabhai D. P. Nierlich and J. Abelson Nature New Biol. 1973,246 1 I. 181 C.Guthrie J. G. Seidman S. Altman B. G. Barrell J. D. Smith and W. H.McClain Nature New Biol. 1973 246 6. P. W. Piper and B. F. C. Clark F.E.B.S.Letters 1973 30 265. G. Petrissant Proc. Nut. Acad. Sci. U.S.A. 1973 70 1046. M. Simsek G. Petrissant and U. L. RajBhandary Proc. Nat. Acad. Sci. U.S.A. 1973 70 2600. Nucleic Acids that the total sequences of the cytoplasmic initiator tRNAs from mouse rabbit and sheep have the same sequence. The sequence T+CG in non-initiator tRNAs has been directly implicated in the binding of the tRNA to the ribosome.'85 Other tRNA species have been found to lack rT. A tRNATy' from a mutant E. cofistrain lacking rT is able to support protein synthesis at a rate equivalent to that observed with normally methylated tRNA.'86 Using an E.cofi uracil methylase assay five specific tRNAs in wheat embryo and also tRNAs in foetal calf calf liver and beef liver have been found to lack rT.ls7 In tRNALY" from rabbit liver 2'-O-methylribothymidine has been identified.'88 This tRNA is fully active in protein synthesis. When Bacillus stearothermophilus is cultured at 70" C a three-fold increase in the amount of 2'-O-methylation of tRNA has been found compared with that when the bacterium is cultured at 50 "C. Little difference in the amount of base methylation was found.189 Two particularly interesting tRNA molecules have been found in mutant bacteria. In frameshift mutants of Salmonelfa typhimurium a new tRNAGIy has been found which has an identical sequence to the tRNAGIy from the wild-type strain except for an additional C in the anticodon l00p.'~~ The anticodon now comprises the quadruplet CCCC and presumably the mutant suppresses frame- shift mutations by reading a quadruplet codon.A new series of 'Smith' mutants has been These are Su; mutants of E. coli selected for their ability to suppress an amber mutation in the p-galactosidase gene of E. cofi fac~ooo which is suppressed by the glutamine-inserting suppressor Sui. Five such mutants have been isolated the tRNATY' isolated and all have been found to contain a different single base substitution in the amino-acid stem. Four of the five mutant tRNAs have been shown to accept glutamine and to insert this in vivo at an UAG site. The results clearly implicate the amino-acid acceptor stem of the tyrosine tRNA as a site of synthetase recognition.The modification of 5-methylaminomethyl-2-thiouridine in the anticodon of E. coli tRNAG1" with BrCN results in a much lowered reaction of the tRNA with its cognate synthetase.' 93 The incorporation of a 2-thiocytidine into the penulti- mate position of the 3'-end of tRNAPhe from yeast does not affect the amino- acylation of the tRNA.Ig4 The addition of UGG to bind to the CCA stem of a tRNA prevents aminoacylation whereas oligonucleotides binding to the anticodon loop do not prevent aminoacylation.' 95 An investigation of homologous and D. Richter V. A. Erdmann and M. Sprinzl Nature New Biol. 1973 246 132. S. Yang E. R. Reiniz and M. L. Gefter Arch. Biochem. Biophys. 1973 157 55.I87 K. Marcu R. Mignery R. Reszelbach B. Roe M. Sirover and B. Dudock Biochem. Biophys. Res. Comm. 1973 55 477. H. J. Gross M. Simsek M. Raba K. Limburg J. Heckman and U. L. RajBhandary, Nucleic Acids Res. 1974 1 35. P. F. Agris H. Koh and D. Soll Arch. Biochem. Biophys. 1973 154 277. I9O D. L. Riddle and J. Carbon Nature New Biof. 1973 242 230. 19' J. D. Smith and J. E. Celis Nature New Biol. 1973 243 66. 19* J. E. Celis M. L. Hooper and J. D. Smith Nature New Biof. 1973 244 26i. 193 P. F. Agris D. So11 andT. Seno Biochemistry 1973 12 4331. M. Sprinzl K. H. Scheit and F. Cramer European J. Biochem.. 1973 34 306. 19' C. J. Bruton and B. F. C. Clark Nucleic Acids Res. 1974 I 217. 646 R. T. Walker heterologous aminoacylation with the yeast phenylalanyl-tRNA synthetase (PRS) has shown that 11 tRNAs can be amin~acylated.'~~ These fall into three classes fast intermediate and slow.Those in the slow class all contain nine bases in the dihydrouridine loop whereas the other two classes all contain eight. All the members of the fast group contain an N'-methylguanosine at position 10 from the 5'-end and those in the intermediate group lack this modification. Specific methylation of one of the intermediate class tRNAPhe from E. coli at position 10 by an enzyme preparation from rabbit liver results in an increased rate of aminoacylation of this tRNA by the PRS confirming that the synthetase requires this position to be methylated for rapid aminoacylation to occur.197 However specific methylation of tRNAp' from E.coli has no effect upon its aminoa~ylation'~~ and it is going to be very important that all experiments in this field are done on pure tRNAs that the modifications claimed are specific and that the positions of modification are identified by sequencing the tRNAs so modified. A three-point attachment for the synthetase has been proposed'99 which includes the amino-acid stem the anticodon loop and the extra loop in tRNAy' but although much progress is being made in this field it is clearly going to be some time if ever before all the apparently conflicting data are rationalized and a uniform recognition system is obtained. Little work of much significance involving chemical modification of tRNAs has appeared during the year.Exceptions to this include the experiments in which tRNAF' from E. coli undergoes photo-oxidation in the presence of methy- lene blue which results in the modification of two guanosine residues at positions 2 and 71.'" It has been shown that the lethal event is the modification at position 71 whereas molecules only containing the modification at position 2 are active in the aminoacylation reaction. Methoxyamine modification of cytidine residues has been used to show that the anticodon loops of bacterial and mammalian initiator tRNAs have similar tertiary structures2" and also that precursor tRNATyr adopts a clover-leaf configuration before maturation.'" Affinity chromatography has been used to isolate purified aminoacyl-tRNA synthetases (or ligases).Columns of Sepharose-phenylalaninezo3and Sepharose- methionine204 have been used to purify their cognate synthetases. In the reverse type of reaction the crude synthetases have been adsorbed on phosphocellulose columns205 and the individual synthetases have been eluted with a pure tRNA solution. 196 B. Roe M. Sirover. and B. Dudock Biochemistry 1973 12 4146. 19' B. Roe M. Michael and B. Dudock Nature New Biol. 1973,246 137. 19' L. P. Shershneva T. V. Venkstern and A. A. Bayer Nucleic Acids Res. 1974 1 235. 199 S. K. Dube Nature New Biol. 1973 243 103. 2oo L. H. Schulman Proc. Nut. Acad. Sci. U.S.A. 1972,69 3594. 201 P. W. Piper and B. F. €. Clark Nucleic Acids Res. 1974 1,45. '02 S. E. Chang and J. D. Smith Nature New Biol. 1973 246 165. '03 P.I. Forrester and R. L. Hancock Canad. J. Biochem. 1973 51 231. '04 M. Robert-Gero and J. P. Waller European J. Biochem. 1972 31 315. '05 F. von der Haar European J. Biochem. 1973 34 84; H. Yamada J. Biochem. 1973 74 187. Nucleic Acids 647 5 RNA Oneof the most interesting results in the viral RNA field is the report by Spiegelman and hisco-workers ofthe complete base sequence ofareplicating RNA molecule.206 It is probably worth describing the history of this RNA molecule which started in 1965 when a template-specific RNA replicase from E. cofi infected with the RNA phage QP was isolated2" and it was established that the enzyme preparation could mediate a virtually indefinite and catalytic synthesis of biologically com- petent and infectious RNA.This system was then used to investigate evolution in the test-tube that is conditions under which the QP-RNA molecules are liberated from many of the restrictions stemming from the requirements of a complete viral life cycle. It was argued that as replicase was provided and the RNA molecules did not have to infect cells for replication the RNA sequences coding for coat protein and replicase might be dispensable. A serial transfer experiment was devised in which the selective advantage depended upon rapid completion of synthesis. After 74 transfers a lower limit of 550 nucleotides had been reached corresponding to an 88 %rejection of the phage genetic information. The report at the end of last year2'* showed that further experiments had resulted in the isolation of a replicating molecule of only 218 nucleotides and this has now been sequenced.The length of this molecule is such that its sequence and that of mutants which it produces in response to external stimuli can be determined without too much effort. The sequence of the RNA is given in Figure 1. The RNA replicates by way of a two-stranded intermediate and it is argued that if a given variant is superior because of a particular sequence in its plus strand the advantage would be increased if the same sequence could occur in the minus strand. However the plus strand must then contain the antiparallel complement of the advantageous sequence and the formation of intra-strand helices should occur. As can be seen the sequence does contain several complementary runs of nucleotides capable of forming helical regions.Now experiments can be devised so that base changes which have occurred in the RNA mutating from one pheno- type to another can be determined. There is now a computer algorithm for generating all possible sequences of an RNA given the sequences of pancreatic and T1 RNase fragment^.^" Several sequences of R17 RNA have been determined.210 The sequence determination of QP RNA continues now that it has been possible to resynchronize RNA synthesis at an internal site of the RNA template.211 In MS2 phage the gene coding for the A-protein has been shown to start at nucleotide 130 from the '06 D. R. Mills F. R. Kramer and S. Spiegelman Science 1973 180 916. 20' I. Haruna and S. Spiegelman Proc.Nut. Acad. Sci. U.S.A. 1965 54 579. 208 D. L. Kacian D. R. Mills F. R. Kramer and S. Spiegelman Proc. Nut. Acad. Sci. U.S.A. 1972 69 3038. 209 L. L. King Math. Bioscience 1973 16 273. 210 U. F. E. Rensing and J. G. G. Schoenmakers European J. Biochem. 1973 33 8; U. F. E. Rensing Biochem. J. 1973 131 593; U. F. E. Rensing A. Coulson and E. 0.P. Thompson Biochem. J. ibid. p. 605. 211 D. Kalakofsky M. A. Billeter H. Weber and C. Weissmann J. Mol. Biol. 1973 76 271. G U/A A GA CG U UA uc CG CG CG GC GA GA GC UA AA GA CG CG CG GG CG AU AC GC CG CG GC GG CG CG CG UA UA AU CG CG GC GC CG CG CG CG GC CG CG CG CG AU GC AU pppGGGGA’ I/ACGGGAGUUCGA’ ‘GCU) {CUCC’ I{GAACCI \CCU\ /GGUGI \UCCCCOH AU AU UA UA CG GC C.G CG GC CG GC GC AU GC UA CG GC CG GC GC GC uc CG CG UA uu GC AU GC UA GC CG UG CG GG U C cc CG G Figure1 The complete sequence of the plus strand of MDV-1 RNA Nucleic Acids 649 5'-end and the initiating codon is GUG.2'2 Another fragment of the MS2 RNA genome 73 nucleotides long has been described and it shows a striking similarity to some ribosomal binding B.stearothermophilus ribosomes have been shown to protect a 38-nucleotide fragment of Qp RNA but the region contains no AUG or GUG initiator ~odon.~' The nucleotide sequence of a ribosome binding site of RNA synthesized in uitro from coliphage T7 a source evolutionally unrelated to sources from which previous sequences have been obtained is found to contain similar sequences to those found for R17 and QB sites.215 The probable initiator codon is in the sequence AACAUGAGG which also occurs in the R17 replicase cistron ini- tiation sequence.The pentanucleotide sequence GAGGU also occurs in the R17 A-protein ribosome binding site and the Qp A-protein site contains the sequence GAGG. It would appear that the primary sequence around the AUG codon does indeed have a major role in specifying the point of ribosome attachment for the initiation of protein synthesis. "he length of RNA fragments necessary for the formation of complexes with ribosomes has been investigated.216 Once again the report of a symposium on protein synthesis in reproductive tissue has been printed and issued in book form within lOOdays of the meeting2 "-at least some people can get their work printed quickly! Many interesting papers and discussions on eukaryotic mRNA are included.HnRNA or its constituent mRNA can be isolated using ordinary cellulose apparently without the need for the oligo dT which is usually at- tached to the cellulose. mRNA has also been isolated using poly(U) covalently linked to Sepharose" and mouse heavy-chain immunoglobulin mRNA has been purified by specifically binding it to complete myeloma protein and pre- cipitating the resulting RNA-protein complex with antiserum against the myeloma protein.220 The link between HnRNA and mRNA continues to be established. The poly(A) sequence in both RNAs has been shown to be the same,221 although HnRNA contains a poly(U) segment 30 nucleotides long which is absent in the mRNA ;this is possibly transcribed from the repeated regions of the DNA.222 DNA/RNA hybridization has been used to establish the identity of sequences of globin mRNA in giant nuclear precursor RNA of avian erythr~blasts.~' 212 G.Volckaert and W. Fiers F.E.B.S. Letters 1973 35 91. 'I3 G. Haegeman and W. Fiers European J. Biochem. 1973,36 135. 214 J. A. Steitz J. Mol. Biol. 1973 73 1. J. R. Arrand and J. Hindley Nature New Biol. 1973 244 10. A. G. Porter and J. Hindley F.E.B.S. Letters 1973 33 339. ' See Acta Endocrinologica 1973 Suppl. 180. 'I8 C. J. Larsen M. Marty R. Emanoil-Ravicovitch and M. Boiron F.E.B.S. Letters 1973 33 61. 219 U. Lindberg and T. Persson European J.Biochem. 1972 31 246. *" R. H. Stevens and A. R. Williamson Proc. Nut. Acad. Sci. U.S.A. 1973,70 1 127. ''I G. R. Molloy and J. E. Darnell Biochemistry 1973 12 2324. 222 G. R. Molloy W. L. Thomas and J. E. Darnell Proc. Nut. Acad. Sci. U.S.A. 1972 69 3684. 223 T. Imaizumi H. Diggelmann and K. Scherrer Proc. Nor. Acad. Sci. U.S.A. 1973.70 1122. 650 R.T. Walker More evidence is available which suggests that the poly(A) sequences in mammalian cell mRNA originate in the nucleus224 and that once in the cytoplasm they become smaller with age.225 A partial base sequence of an mRNA has been reported.226 The immuno- globulin light-chain mRNA of mouse myeloma cells contains a ‘poly(A)’ sequence composed entirely of A residues. The sequence of four oligonucleotides each about 20 residues long can be correlated with the known amino-acid sequence of t’he protein and a fifth oligonucleotide must be derived from a region of the mRNA which is not translated.The determination of some nucleotide sequences for HeLa cell HnRNA confirms previous observations that the GC doublet occurs very infrequently in the vertebrate genome.227 The Xenopus oocyte continues to be used for the micro-injection of eukaryotic mRNAZ2* and precursor mRNAZz9 in which it is translated very efficiently. However a word of caution has been given in the interpretation of some of the results obtained with precursor mRNA because of the sensitivity of the It is claimed that only 1% of an mRNA contaminating the high molecular weight RNA which it is desired to show contains mRNA sequences would account for some of the published results.It is the duty of people using this system to use controls which demonstrate that this is not the case. The fidelity of translation of some mRNAs has been checked under very critical eonditions such that the amino-acid sequence of the product has been compared with that of the in vim pr~duct.’~’ The translation of mRNA in heterologous in uitro systems has continued.232 One such system described consists of mouse liver ribosomes rabbit reticulocyte initiation factors and rat liver pH 5en~ymes.2~~ This directs the synthesis of all known duck globin chains from duck globin mRNA but if given a saturating mixture ofeach of rabbit and duck globin mRNAs only rabbit globin synthesis takes place.Rabbit globin mRNA has been translated on ribosomes from trout liver and kidney bean root tips.234 In the ribosomal RNA field major sequence heterogeneity has been observed between the 5s RNA from the kidneys and ovaries of Xenopus laeui~.~~~ The nucleotide sequence of a 5.8s rRNA from Saccharomyces cereuisiae has been 224 W. Jelinek M. Adesnik M. Salditt D. Sheiness R. Wall G. Molloy L. Philipson and J. E. Darnell J. Mol. Biol. 1973 75 515. 225 D. Sheiness and J. E. Darnell Nature New Biol. 1973 241 265. 226 G. G. Brownlee E. M. Cartwright N. J. Cowan J. M. Jarvis and C. Milstein Nature New Biol. 1973 244 236. 227 N. W. Frazer B. E. H. Maden and R. H. Burdon F.E.B.S. Letters 1973 36 257. 228 C.D. Lane and C. M. Gregory European J. Biochem. 1973 34 219. 229 R. Williamson C. E. Drewienkiewicz and J. Paul Nature New Biol. 1973 241 66. 230 C. D. Lane C. M. Gregory T. Iyazumi and K. Scherrer Nature New Biol. 1973 243 78. 231 N. J. Cowan and C. Milstein European J. Biochem. 1973,36 1; R. F. Jones and J. B. Lingrel J. Biol. Chem. 1972 247 7951. 232 A. G. Stewart E. S. Gander C. Morel B. Luppis and K. Scherrer European J. Biochem. 1973 34 205. 233 M. H. Schreier T. Staehelin A. Stewart E. Gander and K. Scherrer European J. Biochem. 1973 34 21 3. 234 C. Vaquero L. Reibel J. Delaunay and G. Shapira Biochem. Biophys. Res. Comm. 1973 54 1171. 235 P. J. Ford and E. M. Southern Nature New Biol. 1973 241 7. Nucleic Acids 651 determined236 and the deliberations of a Cold Spring Harbor Symposium on the ribosome have appeared under the title ‘Ribosome Model to Take Home to Ma.’237 While it undoubtedly provides an up-to-date summary of our present knowledge of ribosomal structure and function it is highly doubtful whether in the majority of cases ‘Ma’ would be either interested or impressed! 6 DNA An excellent review entitled ‘The Primary Structure of DNA’ has appeared.238 This deals with methods of sequencing DNA and in particular the interaction of restriction enzymes and other proteins with DNA and some of the literature covered is still in the press at the end of 1973.Bernardi and his co-workers have been investigating the specificity of the DNases :239 Helix aspersa DNase pancreatic DNase E.coli DNase I and spleen DNase have been studied. It is concluded that these enzymes are not as non- specific as has previously been assumed. The snail enzyme is thought to recognize a sequence at least two nucleotides long the pancreatic and E. coli enzymes recog- nize a sequence of at least three nucleotides and the spleen DNase one of at least four nucleotides. The specificity of the DNases remains constant throughout the course of digestion and these enzymes may not be too unlike therestriction enzymes except that the nucleotide sequences they recognize are much shorter. A method of using these enzymes in nucleotide sequence studies has been proposed.240 Neutron activation analysis has been used to measure the total and terminal phosphorus contents of oligonucleotide^,^^^ and thus their molecular weights can be calculated.The method is claimed to use less than a milligram of polymer of molecular weight lo7. Two satellite DNAs have been investigated. The original satellite the ‘poly- (dAT)’ from crabs has been shown to possess 89 % of the thymine residues (and hence also the adenine residues) as XpTpTpX and a further 7 %as XpTpTpTpX.242 This non-random base arrangement explains the anomalously low density of this satellite DNA. A satelIite DNA from kangaroo rat has been found to have the sequence (27),243each diploid cell containing 80 x lo6 repetitions. 5‘A C A C A G C G G G 3’ 3’T GT GT C GC C C “5’ ( 1 (27) 236 G. M. Rubin J. Biol. Chem. 1973 248 3860. 23’ See Nature New Biol.1973 246 129. 238 K. Murray and R. W. Old Progr. Nucleic Acid Res. Mol. Biol. 1973,14 to be published. 239 A. Devillers-Thiery S. D. Ehrlich and G. Bernardi European J. Biochem. 1973 38 416 and the following fouY papers; J. Laval J. P. Thiery S. D. Ehrlich C. Paoletti and G. Bernardi European J. Biochem. 1973,40 133 and the following three papers. 240 G. Bernardi S. D. Ehrlich and J. P. Thiery Nature New Biol. 1973 246 36; S. D. Ehrlich J. P. Thiery R. Devillers-Thiery and G. Bernardi Nucleic Acids Res. 1974 1 87. 241 L. Clerici E. Sabbioni F. Campagnari S. Spadari and F. Girardi Biochemistry 1973 12 2887. 242 S. D. Ehrlich J. P. Thiery and G. Bernardi Biochem. Biophys. Acta 1973 312 633. 243 K. Fry R. Poon P. Whitcome J. Idriss W. Salser J.Mazrimas and F. Hatch Proc. Nut. Acad. Sci. U.S.A. 1973,10 2642. 652 R. T. Walker Modern methods available for obtaining the primary sequence of DNA include:238 (i) the use of the repair reaction of DNA polymerase I for single- stranded DNA regions (the length of DNA repaired can be restricted by providing only two or three deoxynucleoside triphosphates and a ribonucleotide can also be incorporated in the presence of Mn2 +);(ii) the specific labelling of the 5‘- or the 3’-termini;(iii) the use of enzymes to produce double-strand breaks at specific sites ;(iv) the in nitro transcription reaction to analyse regions of DNA around the promoters; (v) the protection of a region of DNA by a protein or ribosomes followed by analysis of the protected fragment; and (vi) the use of partial endonuclease digests of uniformly labelled DNA to give large fragments which can subsequently be analysed by chemical and enzymatic degradative methods.DNA polymerase I has been used to show that the cohesive ends of phage $80 have identical sequences to those of phage A.244 DNA polymerase I has also been primed by a synthetic octadeoxynucleotide complementary to a sequence of phage f DNA and the adjacent DNA sequence of 50 nucleotides has been deter- mined.245 Despite the fact that the sequence of the octanucleotide was chosen to be complementary to a DNA sequence coding for the coat protein the sequence of the DNA obtained shows that it is probably an inter-cistronic region. The sequence is very T-rich and contains several short repeated sequences.The amino-acid sequence of the coat protein is now thought to be incorrect. Khorana has put his ability to synthesize oligodeoxynucleotides (22 units in length) to good use by determining the sequence of DNA in the phage 480,which contains the tRNATy‘ gene past the precursor tRNA sequence into the promoter and terminator regions of the phage genome.246 Using a restricted number of deoxyribonucleoside 5’-triphosphates and also the ribosubstitution technique a sequence of 23 nucleotides into the terminator region beyond the DNA coding for the CCA end has been determined. This sequence contains two palindromic sequences and also several elements of symmetry. The sequence of an octanucleo- tide of the promoter region beyond the 5’-end of the terminal pppG of the pre- cursor tRNA has been found to be ACGCGGGG.The properties required of primers used in the repair reaction of DNA poly- merase I have been investigated.247 Oligonucleotides with a high purine content (needless to say the most difficult to synthesize) are the best primers and usually a length of 8-12 nucleotides is required for specific initiation and primer activity. Some activity is detected with pieces as short as trimers and tetramers indicating that the enzyme must play a role in stabilizing the primer on the template. Polynucleotide kinase has been to label the 5’-termini of the DNAs from phage h and phage 424 and the sequences of the cohesive ends of these and 244 R. Bambara R. Padmanabhan and R.Wu J. Mol. Biol. 1973,75 741. 24s F. Sanger J. E. Donelson A. R. Coulson H. Kossel and D. Fischer Proc. Nar. Acad. Sci. U.S.A. 1973 70 1209. 246 P. C. Loewen and H. G. Khorana J. Biol. Chem. 1973 248 3489. 247 M. Goulian S. H. Goulian E. E. Codd and A. Z. Blumenfield Biochemistry 1973 12 2893; W. Oertel and H. Schaller European J. Biochem. 1973,35 106; G. G. Peters and R. S. Hayward ibid. 1973 31 360. 24a K. Murray Biochem. J. 1973 131 569. Nucleic Acids 653 that of several other lambdoid and non-lambdoid DNA phages249 have been determined. The sequences of all the lambdoid phages are identical and the sequence between the single-strand breaks has a two-fold rotational axis of sym-metry and is probably involved in the recognition of the site by the enzyme re- sponsible for the formation of the cohesive ends.The 3’-terminal nucleotide sequences of phage A have also been determined following the labelling of the 3’-end~~~’ and together with the known 5’-terminal sequences a sequence of 25 base pairs in the vicinity of the termini is now known. The cohesive ends of the non-lambdoid phages are quite long (19 bases) and they contain no apparent rotational axis of symmetry.249 Much more is now known about the restriction system used by many bacteria to recognize and destroy foreign DNA. Each system consists of two enzymes. One the restriction enzyme is an endonuclease which cleaves foreign but not host DNA at specific sites often by means of two single-strand breaks a short distance apart to create cohesive ends of characteristic sequence for each enzyme.The second enzyme is a methylase which methylates the host DNA to prevent its degradation. Many restriction enzymes each of which will be accompanied by a methylase enzyme have been discovered and a system of nomenclature has been proposed251 which it is hoped will be adopted by everyone in this field before the present chaos is further increased. The proposed nomenclature is given in the Table together with a list of the known sites of action of the restriction enzymes. The use of these enzymes in DNA sequence determination is already apparent and they have also been used to obtain a genetic map of the SV40,251,252 4X174,253 and A,254 genomes. A simple assay for restriction endonucleases has been proposed.255 The DNA sequences recognized by the restriction enzymes have so far all been shown to possess a high degree of symmetry238 and the one methylase enzyme which has been extensively studied has been shown very pro- bably to methylate that sequence in the host DNA which would have been cleaved by the endonuclease thus confirming the one substrate-two enzyme concept for this system.256 In rather an ominous development a biologically functional bacterial plasmid has been constructed in uitro by joining restriction endonuclease-generated fragments of separate plasm id^.^^' This hybrid plasmid DNA has been inserted 249 K.Murray and N. E. Murray Nature New Biol. 1973 243 134. P. H. Weigel P. T. Englund K. Murray and R. W. Old Proc.Nut. Acad. Sci. U.S.A. 1973 70 1151 ;G. S. Ghangas E. Jay R. Bambara and R. Wu Biochem. Biophys. Res. Comm. 1973 54 998. 2s’ K. J. Danna G. H. Sack jun. and D. Nathans J. Mol. Biol. 1973 78 363. 2s2 G. H. Sack and D. Nathans Virology 1973,51 517. 2s3 C. Y.Chen C. A. Hutchinson and M. H. Edgell Nature New Biol. 1973 243 233; M. Sclair M. E. Edgell and C. A. Hutchinson J. Virol. 1973 11 378. 2s4 B. Allet P. G. N. Jeppesen K. J. Katagiri and H. Delius Nature 1973 241 120. *” F. M. DeFilippes Analyt. Biochem. 1973 52 637. 2s6 H. W. Boyer L. T. Chow A. Dugaiczyk J. Hedgpeth and H. M. Goodman Nature New Biol. 1973 244 40. ” S. N. Cohen C. Y. Chang H. W. Boyer and R. B. Helling Proc. Nut. Acad. Sci. U.S. A. 1973,70 3240. 654 R. T. Walker Table Abbreviations and DNA sequences at sites of restriction for restriction enzymes Source of enzyme Abbreviat iona DNA sequence at restriction siteb Reference Haemophilus injluenzae R.Hin,I HinJI Unknown - strain Ra Haemophilus injluenzae R.HindII 45’ pG-T-YTR-A-C 260 strain Rd C-A-R~Y-T-~p 5’ Haemophilus injuenzae R.HindIII 5’ PA’A-G~C-T-T 238 strain Rd T-T-C~G-A-AP3.5’ Escherichia coli R.Eco, 5‘ PG~A-A~T-T-c 26 1 resistance transfer factor RI C-T-TLA-A-G~I f 5’ Escherichia coli R.Eco,, + I 5’ pC-C-A-G-G 262 resistance transfer factor RII G-G-T-C-CP5‘ I t Haemophilus aegyptius R.Hae C-CGG-G~ 4 5‘ PG-GTC-C 5’ 238 Haemophilus aphirophilus R.Hap 5‘ ~C~C~G-G 263 G-GIC-CP 5’ Haemophilus parainjuenzae R.Hpa1 HpaII t &known Restriction enzymes are designated R and modification enzymes M followed by the three-letter code describing the organism from which they were obtained.Subscripts describe the strain or a plasmid carried by the organism and where more than one enzyme is produced by a given strain these are distinguished by Roman numerals written in line with the code for the enzyme. The broken line drawn through the sequence around the sites of restriction represents an axis of two-fold rotational symmetry and the arrows show the position of hydrolysis of the phosphodiester bonds. Y is a pyrimidine and R a purine residue. into E. coli by transformation and has been shown to be biologically functional so that the bacterium possesses genetic properties and base sequences from both parent DNA molecules.This genetic engineering requires little skill as the restric- tion enzyme produces complementary cohesive ends from the two plasmids which can then be joined by ligase. It has already been suggested that it would be ‘interesting’ to incorporate a tetracycline-resistance gene from a plasmid into bacteria and no doubt the medical profession will find it most ‘interesting’ to be confronted with a whole new series of bacterial strains which possess specific drug resistance. Further implications of the technique are obvious and one can only hope that workers in this field will ‘restrict’ their own enthusiasm in design- ing experiments to see what happens when various DNA molecules are linked together particularly where the DNA from oncogenic viruses is concerned.Nucleic Acids 655 h I I dI-e U W c < u u < 0 8I-U c u u < u < < U % U u U v) I I 4 U c U U c (3 u c u U I-< 0 u c u I-I-e u V k-W c W W 3 U 3 V 0 d u < U U 0 u U u U u U u 3 3 *3 U u 4 U 3 4 0 3 u 3 3 U "< b, ou n d & E 656 R. T. Walker Finally more is now understood about the interaction of DNA operators with protein repressors. In the lactose operon of E. coli the DNA sequence protected by the repressor has been isolated and its sequence determined.The DNA was transcribed into RNA and a sequence of 24 base pairs258 was deter- mined. Using a mutant E. coli strain the 5'-end of the Zac mRNA has been transcribed in vitro and its initial sequence was found to match the sequence of the operator fragment which was followed by an AUG starting at residue 39 and continuing with an RNA sequence which codes for the amino-terminal sequence of B-galactosidase. This sequence is shown in Figure 2. The operator fragment has a symmetrical arrangement of bases and the method by which the repressor and operator interact is being investigated. However all repressor-operator interactions are not as simple as this system. In bacteriophage A,259 two lambda operators exist to which the lambda repressor protein binds to maintain the phage in an inactive state.More than one molecule of repressor binds per operator molecule and eventually two sequences of 100 base pairs each are covered. A model is suggested in which the operator consists of a sequence of some 35 base pairs which binds a dimeric repressor and four additional monomer units may then add to sequences ca. 15 base pairs long adjacent to the primary sequence. The two lambda operators are known to have different DNA sequences and the system is obviously very complicated. It is interesting that the operator sequences were obtained following the action of restriction endonucleases which act at symmetrical sequences and this suggests that the sequence of this promoter may also possess elements ofsymmetry.Thanks are due to Dr. R. J. H. Davies who collected much of the material for the section on Photochemistry and Physical Chemistry of Polynucleotides. 258 W. Gilbert N. Maizels and A. Maxam Cold Spring Harbor Symp. Quantitative Biol. 1973 38 845. 2s9 T. Maniatis and M. Ptashne Nature 1973 246 133; T. Maniatis M. Ptashne and R. Maurer Cold Spring Harbor Symp. Quantitative Biol. 1973 38 857. "O T. J. Kelly and H. 0.Smith J. Mol. Biol. 1970 51 393. 261 J. Hedgpeth H. M. Goodman and H. W. Boyer Proc. Nut. Acad. Sci. U.S.A. 1972 69 3448. 262 C. H. Bigger K. Murray and N. E. Murray Nature New Biol. 1973 244 7. 263 H. Sugisaki and M. Takenami Nature New Biol. 1973 246 140.