15 Biological Chemistry Part (ii) Nucleic Acids (b) Oligonucleotides and Polynucleotides By M. J. GAIT Medical Research Council Laboratory of Molecular Biology Hills Road Cambridge CB2 2QH 1 Introduction In 1980 an important change of emphasis took place in the field of nucleic acids. On both sides of the Atlantic there was a considerable easing of the containment regulations under which DNA could be cloned and this led to a rush to isolate genes directly from chromosomal DNA and to set up systems for the expression of DNA in bacteria and other micro-organisms. Whereas between 1977 and 1979 much effort had gone into the total synthesis of genes now it became possible to derive these sequences from relatively impure preparations of mRNA via cloning of its complementary DNA (cDNA).First signs of the change in emphasis appeared in late 1979 with a report on the expression in bacteria of a gene for human growth hormone.’ Only the terminal 84 base-pairs corresponding to the first 24 amino-acids were synthetically derived. The remaining 551 base-pairs (167 amino-acids) were obtained from a restriction fragment isolated from cloned cDNA. Then early in 1980 amid a storm of publicity the sequences of the cDNAs corresponding to human leukocyte interferon (inter- feron a)*and human fibroblast interferon (interferon p)”‘“were announced. Within a few months interferon cDNAs had been expressed in ba~teria.”~ By the end of the year much information had already been accumulated on the structure of interferon genes,7 which appeared to be arranged in multiple sets.The frenetic pace of development of the subject was undoubtedly due to the high level of D. V. Goeddel H. L. Heyneker T. Hozumi R. Arentzen K. Itakura D. G. Yansura M. J. ROSS G. Miozzari R. Crea and P. H. Seeburg Nature (London) 1979,281 544. ’N. Mantei M. Schwarzstein M. Streuli S. Panem S. Nagata and C. Weissmann Gene 1980,10 1. T. Taniguchi S. Ohno Y. Fujii-Kurijama and M. Murumatsu Gene 1980 10 11; R. Derynck J. Content E. DeClercq G. Volckaert J. Tavernier R. Devos and W. Fiers Nature (London) 1980 285 542. M. Houghton M. A. W. Eaton A. G. Stewart J. C. Smith S. M. Doel G. H. Catlin H. M. Lewis T. P. Patel J. S. Emtage N. H. Carey and A. G. Porter Nucleic Acids Res. 1980,8,2885. R. Derynck E.Remaut E. Saman P. Stanssens E. DeClercq J. Content and W. Fiers Nature (London),1980,287,193. D. V. Goeddel E. Yelverton A. Ullrich H. L. Heyneker G. Miozzari W. Holmes P. H. Seeburg T. Dull L. May N. Stebbing R. Crea S. Maeda R.McCandliss A. Shoma J. M. Tabor M. Gross, P.C. Familletti and S. Pestka Nature (London) 1980 287,411. ’ S. Nagata N. Mantei and C. Weissmann Nature (London) 1980 287 401; G. Allen and K. H. Fantes ibid. p. 408. 310 Biological Chemistry -Part (ii)b Oligo- and Poly-nucleotides 311 industrial interest in the possible exploitation of interferon as antiviral and anti- tumour agents but at the end of 1980 the clinical value of interferon still remained in doubt.8 Another major highlight of 1980 was the award of the Nobel prize in Chemistry to Walter Gilbert from Harvard and Frederick Sanger from Cambridge for their studies in sequence determination of DNA together with Paul Berg of Stanford for his original work on recombinant DNA.This award clearly marks the world-wide impact that their discoveries have had on the advancement of science. 2 Determination of Base Sequences of Nucleic Acids Extremely detailed protocols have been published recently for both the Maxam- Gilbert' and the Sanger 'dideoxy' lo methods of DNA sequencing the basic prin- ciples of which were outlined in last year's Report. Some refinements have been made to the chemical reactions involved in the Maxam-Gilbert approach.' A virtually Gua-specific reaction is now obtained by N-7-methylation with dimethyl sulphate followed by base displacement and strand scission using piperidine.Removal of Gua + Ade is achieved by depurination with formic acid. The DNA is then broken at these sites with piperidine. Conditions for pyrimidine-specific modification are unaltered. Both Cyt and Thy are modified by using hydrazine but only Cyt residues react when the hydrazinolysis is carried out in the presence of 1M sodium chloride Again base displacement and strand scission occur on treatment of the DNA with piperidine. An Ade > Cyt specific cleavage is also recommended. Here the bases are ring-opened by treatment with sodium hydroxide. Once more reaction of the DNA with piperidine completes the cleavage. The only major change in the Sanger approach lies not with the method of sequence determination" but with the increasing use of bacteriophage M13 as a vector for cloning of DNA fragments." These are usually generated by the use of yestriction enzymes to cut the DNA of interest the pieces being inserted by ligation into the double-stranded replicative form of phage M13 which has been pre-cut with the same restriction enzyme.After transformation of Escherichia coli cell clones are selected at random and the M13 DNA is isolated. This is single-stranded and the sequence of the inserted DNA is determined." DNA polymerase mediates the synthesis of radioactive complementary strands of DNA that each terminate in a 2',3'-dideoxynucleoside.Synthesis is directed by the use of a specific primer at a site on M13 immediately prior to the insert.The primer can be derived from a restriction fragment or can be a synthetic oligodeoxyribonucleotide (see Section 3). The method has been used to obtain the sequences of both human and beef mitochondria1 DNA,".'* each being ca. 15 000 base-pairs long. Whereas the Maxam-Gilbert method is perhaps most often chosen for DNA of a few hundred 'J. L. Maux Science 1980 210,998. A. M. Maxam and W. Gilbert in 'Methods in Enzymology' ed. L. Grossman and K. Moldave Academic Press New York 1980 Vol. 65 p. 499. 10 A. J. H. Smith in 'Methods in Enzymology' ed. L. Grossman and K. Moldave Academic Press New York 1980 Vol. 65 p. 560. l1 F. Sanger A. R. Coulson B. G. Barrell A. J. H. Smith and B. A. Roe J. Mol. Biol.,1980 143,161. B.G. Barrell S.Anderson A. T. Bankier M. H. L. DeBruijn E. Chen A. R. Coulson J. Drouin I. C. Eperon B. A. Roe F. Sanger P. H. Schreier A. J. H. Smith R. Staden and I. G. Young Proc. Natl. Acad. Sci. USA,1980 77 3164. 312 M. J. Gait base-pairs the Sanger approach seems more adaptable for longer stretches of DNA because of the greater incorporation of radioactivity per nucleotide and this method will undoubtedly predominate in years to come as projects become yet more ambitious. Two very similar methods have been published for determining the sequence of tRNA and short ribosomal RNA that allow minor bases to be more easily identified. Both are based on the observation of Stanley and Va~silenko’~ that RNA can be partially hydrolysed under ‘single hit’ conditions by formamide or water to give two sets of fragments one of which contains 5‘-hydroxyl groups.These latter fragments can be 32P-labelled by the polynucleotide kinase reaction and fractionated by polyacrylamide gel electrophoresis to give a set of radioactive bands each differing in length by one nucleotide. In the new procedures the bands are transfer- red on to polyethyleneimine-cellulose’4or diethylaminoethyl-cellulosels plates and the polynucleotides are digested in situ with RNase T2 to release their 5’-terminal nucleosides as radioactive 3’,5’-diphosphates In the first method the diphosphates are chromatographed in unbuffered ammonium sulphate or ammonium formate solution (pH3.5),14 whereas in the latter method they are separated by electro- phoresis at pH 2.3.15 The nucleoside derivatives are identified by their mobility and the RNA sequence is ‘read off’ along the width of the plate.The sequences of other RNAs are most often determined via their complemen- tary DNA (cDNA) transcripts. Commonly this involves cloning of the DNA as described in last year’s Report; however this is not always necessary especially if specific DNA primers are available to select the species of RNA to be copied out of an impure mixture. The primers can be synthetic oligodeoxyribonucleotides16 (see also Section 3) or restriction fragments,” and the sequencing of the DNA is carried out by the Maxam-Gilbert method or by the ‘dideoxy’ approach. A novel alternative has recently been described where 32P-labelled oligoribonucleotides generated by digestion of the unknown RNA with RNase T followed by labelling of their 5’-ends are used to prime the DNA-polymerase-mediated synthesis of the second strand of DNA.Once again chain terminators are used to generate a series of DNA fragments that are resolvable by gel electrophoresis. The method has been used to obtain the sequence of 1060 bases at the 3’-end of poliovirus RNA.I8 3 Synthetic Oligonucleotides in Molecular Biology The use of synthetic oligonucleotides featured prominently in two approaches to the isolation of interferon cDNAs. Based on protein sequence information only the sequence of a pentadecadeoxyribonucleotidewas deduced. The oligonucleo- tide was found successfully to prime the reverse-transcriptase-mediated synthesis of the cDNA of human interferon p.The sequence of the cDNA was determined l3 J. Stanley and S. Vassilenko Nature (London) 1978,274 87. l4 K.Randerath R. C. Gupta and E. Randerath in ‘Methods in Enzymology’ ed. L. Grossman and K. Moldave Academic Press New York 1980,Vol. 65,p. 638. Is Y. Tanaka T. A. Dyer and G. G. Brownlee Nucleic Acids Res. 1980.8 1259. 16 G. G.Brownlee and E. M. Cartwright J. Mol. Biol.,1977 114 93; P. H. Hamlyn G. G. Brownlee C.-C. Cheng M. J. Gait and C. Milstein Cell 1978,15 1067. e.g. P.K.Ghosh V.B. Reddy M. Piatik P. Lebowitz and S.M. Weissman in ‘Methods in Enzymology’ ed. L. Grossman and K. Moldave Academic Press New York 1980,Vol. 65,p. 580 and references therein. N. Kitamura and E. Wimmer Proc.Natl. Acad. Sci. USA 1980 77 3196. Biological Chemistry -Part (ii)b Oligo- and Poly-nucleotides 313 directly without cloning and this information allowed further specific primers to be designed to obtain the remainder of the coding ~equence.~ In an alternative procedure several synthetic DNA primers were prepared based on all of the possible DNA sequences corresponding to part of the known protein sequence of interferon. A mixture of these oligonucleotides was used as a hybridization probe to screen cloned plasmid DNA containing interferon cDNA inserts. Under stringent conditions only the exactly matched probe hybridized thus enabling correct clone selectiom6 A colony-hybridization procedure involving synthetic oligo-deoxyribonucleotides has been used for screening clones containing genomic DNA19 and synthetic polymeric genes coding for poly(L-Asp-L-Phe).” A quick way of determining sequences adjacent to the 3’-poly(A) tail in mRNA has been described.Each of twelve dodecamers of the form d(pT,-N-N’) was used to prime the reverse-transcriptase-directed synthesis of cDNA corresponding to the mRNA of bovine growth hormone. Only the correctly matched sequence hybridized and acted as a primer.” Cloning and characterization of DNA sequences corresponding to RNA of the influenza virus has been facilitated by using a specific dodecadeoxyribonucleotide to prime the synthesis of cDNA at the 3’-end of the RNA.22 Although the influenza genome is segmented into eight separate RNA chains cloning of the entire genome is simplified in that all eight chains contain identical 3‘- and 5’-leader sequences.The use of this dodecamer and another oligonucleotide for the 5’-end of the cDNA to prime the DNA-polymerase-directed synthesis of the second DNA strand has greatly aided cloning of full-length cDNA.’~ The use of a specific oligodeoxyribonucleotideprimer is an essential element in the determination of DNA sequences cloned in bacteriophage M13 (see Section 2). A thirty-base cloned primer prepared by a combination of chemical and enzymatic synthesis has been shown to be extremely efficient in this respe~t.’~ When required it is isolated from plasmid DNA by digestion with restriction enzymes. A fully synthetic nonadecadeoxyribonucleotide has been shown to be of comparable value as a primer.25 A synthetic primer has the advantage that it may be prepared on substantially larger scale than a cloned primer.Of great potential is a recently described method for the directed deletion of intervening sequences (introns). A 21-unit oligonucleotide was synthesized that was complementary to the bases on either side of an intervening sequence of fourteen bases of a yeast tRNA gene contained in a plasmid. The intron was presumably looped out when the oligonucleotide was hybridized to the plasmid template. The oligonucleotide now acted as a primer for preparation of the second DNA strand in uitru and the heteroduplex was used to transform Escherichia coli l9 J. W. Szostak J. 1. Stiles B.-K. Tye P. Chiu F. Siterman and R.Wu in ‘Methods in Enzymology’ ed. R. Wu Academic Press New York Vol. 68,p. 419. 2o M. T. Doel M. Eaton E. A. Cook H. Lewis T. Patel and N. H. Carey Nucleic Acids Res. 1980 8,4575. 21 N. L. Sasavage M. Smith S. Gillam. C. Astell J. H. Nilson and F. Rottman Biochemistry 1980,19 1737. 22 A. R. Davis A. L. Hiti and D. P. Nayak Gene 1980,10,205. 23 G. Winter S. Fields M. J. Gait and G. G. Brownlee Nucleic Acids Res. 1981,9,237;M.Baez R. Taussig J. T. Zazra J. F. Young P. Palese A. Reisfeld. and A. M. Skalka ibid. 1980,8. 5845. 24 S. Anderson M. J. Gait L. Mayol and I. G. Young Nucleic Acids Res. 1980,8,1731. 25 S. A. Narang R. Brousseau H. M. Hsiung W. Sung R. Scarpulla G. Ghangas L. Lau B. Hess and R. Wu Nucleic Acids Res. Symp. Ser.No. 7 1980 377. 3 14 M. J. Gait cells. Clones with plasmid DNAs containing the deletion were selected by colony hybridization using radiolabelled synthetic oligonucleotides.26 X-Ray structural determination of crystals of self-complementary oligonucleo- tides has been an area of great activity in 1980. A major surprise was that both the hexanucleotide d(C-G-C-G-C-G)*’ and the tetranucleotide d(C-G-C-G),28 under conditions of high salt concentrations crystallized as similar but hitherto unknown left-handed double-helixes. This ‘Z-DNA’ differed from the B form not only in handedness but also in having twelve base-pairs per turn. Whereas the deoxycytidine residues had a conformation similar to that found in the B form the deoxyguanosine residues showed an unusual pucker of the deoxyribose ring and a high angle of rotation about the phosphate residue and the heterocyclic base was found to be in the ‘syn’rather than in the normal ‘anti’orientation.The Z-structure has also been found as an occasional form in fibres of p~ly[d(G-C)].poly[d(G-C)].~~ However it seems unlikely that the Z-structure has any physio-logical significance since it is formed only in solutions in which there are high salt concentrations. Moreover the self -complementary dodecamer d(C-G-C-G-A-A- T-T-C-G-C-G) was found to crystallize in a B form,30 suggesting that the Z-form is incompatible with other base sequences. 4 Synthesis of Oligonucleotides It is now almost universally agreed that chemical synthesis of oligonucleotides is best accomplished uia phosphotriester intermediates.Significant advances have been made recently in phosphorylation methods in the use of coupling agents and protecting groups and in solid-phase synthesis. When used in excess bifunctional 0-or p-chlorophenylphosphorodi-( 1,2,4-triazolide) (1) is effectively mon~functional,~~ and it swiftly phosphorylates the 3‘-hydroxyl group of a 5’-protected 2’-deoxyribonucleoside (2). The resultant monotriazolide (3) is a useful active intermediate (see Scheme 1).It is hydrolysed by water to give the corresponding 3’-aryl phosphate (4) which may be isolated as its triethylammonium or as its barium Alternatively the monotriazolide (3) may be allowed to react with 2-cyanoethanol in the presence of a catalyst such as N-methylirnida~ole~~*~~ to form the or of excess tria~ole~~ corresponding 3’-phosphotriester (5).The phosphotriester (5)may also be prepared from the aryl phosphate (4) by reaction with 2-cyanoethanol in the presence of a 26 R.B. Wallace P.F. Johnson S. Tanaka M. Schold K. Itakura and J. Abelson Science 1980,209,1396. ” A. H.-J. Wang G. J. Quigley F. J. Kolpak J. A. Crawford J. H. van Boom G. van der Marel and A. Rich Nature (London) 1979,282,680. H. Drew T. Takano S. Tanaka K.Itakura and R. E. Dickerson Nature (London) 1980 286 567; J. L. Crawford F. J. Kolpak A. HA. Wang G. J. Quigley J. H. van Boom G. van der Marel and A. Rich Proc. Natl. Acad. Sci. USA 1980,77,4016. *’ S. Arnott R. Chandrasekaran D. L. Birdsall A. G. W. Leslie and R.L. Ratliff. Nature (London) 1980,283,743. 30 R. Wing H.Drew T. Takano C. Broka S. Tanaka K. Itakura and R. E. Dickerson Nature (London) 1980,287.755. 31 J. B. Chattopadhyaya and C. B. Reese Tetrahedron Lett. 1979,5059. 32 G. R. Gough K. J. Collier H. L. Weith. and P. T. Gilham Nucleic Acids Res. 1979 7 1955. 33 P. Cashion K. Porter T. Cadger G. Sathe T. Tranquilla H. Notman and E. Jay Tetrahedron Lett. 1976,3769. 34 C. Broka T. Hozumi R. Arentzen and K. Itakura Nucleic Acids Res. 1980 8 5461. 35 M. J. Gait S. G. Popov M. Singh and R. C. Titmas Nucleic Acids Res. Symp. Ser. No. 7,1980,243. Biological Chemistry -Part (ii)b:Oligo-and Poly-nucleotides 0 I "'7 O=P-N I bN 0 Cl(o-or p-) 0 I O=P-OCH,CH,CN I 0 Cl(o- or p-) B =Thy 6-N-PhCO-Ade 4-N-PhCO-Cyt or 2-N-isobutyryl-Gua (MeO)2Tr = dimethoxytrityl Reagents i (1);ii H20; iii HOCH2CH2CN;iv HOCH2CH,CN coupling agent; v Et,N; vi H' Scheme 1 316 M.J. Gait coupling agent. After detritylation under acidic conditions the 5‘-hydroxy-compound (6) may then be coupled with the aryl phosphate (4) to give fully protected dinucleotides. All sixteen of these have been prepared via the barium salt route in good yield.’* The reaction of the phosphomonotriazolide (3)with the hydroxy- compound (6) also gives a din~cleotide.~~ In another variation the phosphotriester (3)may be prepared directly from the 2-deoxynucleoside derivative (2) by reaction with 0-or p-chlorophenyl phosphorodichloridite in tetrahydrofuran at -78 “C followed by addition of 2-cyanoethanol and then oxidation with iodine in water.36 During phosphorylation reactions involving 2-N-protected deoxyguanosine derivatives 0-6-phosphorylation of the guanine moiety is possible depending on the choice of solvent and of basic catalyst.This side-reaction is supposedly reversible by treatment with aqueous ~yridine,~~.” but an irreversible modification of guanine has been obtained in at least one case,34 and the formation of inter-nucleotide bonds by this route clearly requires great care. In the case of 6-N-benzoyl-2‘-deoxyadenosine derivatives there is great danger of loss of N-benzoyladenine during acidic removal of 5’-dimethoxytrityl groups. Reagents of choice for this reaction are conventionally benzenesulphonic acid in chloroform-methanol (7 :3)38 or more recently trichloroacetic acid in chlor~form’~ or zinc br~rnide.’~ A saturated solution of this latter reagent in nitromethane has been used as a deprotecting agent in the solid-phase synthesis of oligodeoxyribo- n~cleotides.~’No depurination was observed during the deprotection steps.The reaction mechanism is thought to involve chelation between the Lewis acid and the 1’-endocyclic and 5’-exocyclic oxygen atoms of the deoxyribose moiety. l-(Mesitylene-2-sulphonyl)-3-nitro-1,2,4-triazole(MSNT) (7a)41 and 1-(2,4,6- tri-isopropylbenzenesulphonyl)-3-nitro-l,2,4-triazole(TPSNT) (7b)42 are new coupling agents that are useful for the rapid formation of inter-nucleotide bonds. Although rates of are slightly lower than for the corresponding tetrazole derivatives which have hitherto been popular,38 there are fewer side-reacfion~.~~ Uracil and 2-N-benzoylguanine residues are particularly susceptible to modification.For example MSNT has been shown to react with 2’,3’,5’-tri-O-acetyluridineand 2-N-benzoyl-3’,5’-di-O-acetyl-2’-deoxyguanosine to give (8) and (9) respectively. Fortunately these modifications are reversible by mild alkaline treatment. Base modification by arenesulphonyl-tetrazolesis faster and more complex.41 Much debate has centred also on the amount of unwanted 5’-0-sulphonation that is caused by arenesulphonyl-azolides during coupling reactions. Whereas arenesulphonyl-tetrazoles and l-arenesulphonyl-3-nitro-1,2,4-triazoles are undoubtedly powerful sulphonating agents in the absence of a phosphodiester component the amount of sulphonated product obtained under coupling conditions 36 D.Molko R. B. Derbyshire A. Guy A. Roget and R. Teoule Tetrahedron Lett. 1980,21 2159. 37 H. P.Daskalov M. Sekine and T. Hata Tetrahedron Lett. 1980.21 3899. W. L. Sung H. M. Hsiung R. Brousseau J. Michniewicz R. Wu and S. A. Narang Nucleic Acids Res. 1979,7,2199. 39 M. D.Matteucci and M. H. Caruthers Tetrahedron Lett. 1980 21 3243; V.Kohli H. Blocker and H. Koster ibid. p. 2683 40 M. H. Caruthers S. L. Beaucage J. W. Efcavitch E. F. Fisher M. D. Matteucci and Y. Stabinsky Nucleic Acids Res. Symp. Ser. No. 7 1980,215. 41 C.B. Reese and A. Ubasawa. Tetrahedron Lett. 1980,21,2265. 42 J. F.M.de Rooij G. Wille-Hezeleger P. H. van Deursen J. Serdijn and J. H. van Boom Recl. Trao. Chim. Pays-Bas 1979,98,537. 317 Biological Chemistry -Part (ii)b Oligo- and Poly-nucleotides SO* R (7)a; R = Me b R = Pr’ OAc OAc (8) 0 RO-P-N I 0 N=N Cl(o-or p-) OAc (9) is effectively very small when the nucleoside phosphodiester derivative is aryl rather than alk~l~~ and when the concentration of coupling agent is not excessive.43 In the case of arenesulphonyl-tetrazoles the major phosphorylating intermediate in coupling reactions is thought to be the tetrazole derivative Two alternative phosphate-protecting groups have been proposed which in contrast to aryl groups may be removed by reactions that do not involve nucleophilic attack at the phosphorus atom.The p-nitrophenylethyl group is removed by DBU (1,8-diazabicyclo[5.4.0]undec-7-ene)or DBN (1,5-diazabicyclo[4.3.0]non-5-ene) via a p-elimination me~hanism.~~ The methyl group is removed by t-b~tylamine~’ or by thiophenol and trieth~lamine.~~ However since both these protecting groups are alkyl esters the use of arenesulphonyl derivatives as coupling agents for the formation of inter-nucleotide bonds would not be recommended in these cases (see above). Methyl seems particularly suited as a protecting group when a ‘phosphite triester’ coupling procedure is used. This route was originally proposed for thymidyl oligomers in 1975.49A 5’-protected 2’-deoxynucleoside is allowed to react with an alkyl or aryl phosphorodichloridite in tetrahydrofuran at -78 “C,in the presence of a hindered base.The product is allowed to react (without isolation) with a second 43 M. J. Gait and S. G. Popov Tetrahedron Lett. 1980.21 2841. A. Kraszewski J. Stawinski and M. Wiewiorowski Nucleic Acids Res. 1980,8 2301. D. G. Knorre and V. F. Zarytova in ‘Phosphorus Chemistry directed towards Biology’ ed. W. J. Stec Pergamon Press 1980,p. 13. 46 E. Uhlman and W. Pfleiderer TetrahedronLett. 1980.21. 1181. 47 D. J. H. Smith K. K. Ogilvie and M. F. Gillen. TetrahedronLett. 1980,21. 861. G.W.Daub and E. E. van Tamelen J. Am. Chem. Soc. 1977,99,3526. 49 R.L.Letsinger and W. B. Lunsford J. Am. Chem. SOC. 1975.97.3655. 318 M. J. Gait deoxynucleoside unit to give the dinucleoside phosphite triester which may be oxidized to the corresponding phosphotriester by its reaction with iodine in water (see Scheme 2).This method has now been extended for general use in oligodeoxyribonucleotidesynthesis," where it is claimed that there is much less tendency for side-reactions to occur particularly those involving the heterocyclic bases. OH 0 I P-CI I OR \ xoPxoP' 0 0 I I OY OY Reagents i ROPCI, THF at -78°C; ii (11);iii I, H,O Scheme 2 Developments in the oligoribonucleotide field have paralleled those of the deoxy- series. An excellent assembly of a decaribonucleotide corresponding to the 3'-terminus of yeast tRNAA'" has been de~cribed.~~ The methoxytetrahydropyranyl group was used for protection of the 2'-hydroxyl group and inter-nucleotide phos- photriester bonds were formed rapidly and in high yield using MSNT as the coupling agent.5'-0-Sulphonation was not observed during the coupling steps. An added point of interest is that the 5'-hydroxyl component in each coupling reaction contained an unprotected 3'-hydroxyl group. Not only is the rate of formation of 3'-3' linkages considerably slower but they are also extremely prone to alkaline hydrolysis at rates more than three orders of magnitude faster than for the corres- ponding 3'-5' linkages.52 In contrast oligoribonucleotide assembly involving the use of the 2,2,2- trichloroethyl protecting group for phosphate and the t-butyldimethylsilyl group for protection of the 2'-hydroxyl gave poor yields in the coupling steps when arenesulphonyl-azolides were used as coupling agents.53 Much greater success was 50 J.L. Finnan A. Varshney and R. L. Letsinger Nucleic Acids Res. Symp. Ser. No. 7 1980 133. 51 S. S. Jones B. Rayner C. B. Reese A. Ubasawa and M. Ubasawa Tetrahedron 1980,36,3075. 52 B. Rayner C. B. Reese and A. Ubasawa J. Chem. SOC. Chem. Commun. 1980,972. 53 K. K. Ogilvie and R. T. Pon Nucleic Acids Res. 1980,8 2105. Biological Chemistry -Part (ii) b Oligo- and Poly-nucleotides 3 19 obtained by using this combination of protecting groups when the phosphite triester coupling procedure was although the methyl group is now preferred for phosphate protection (see above).47 Considerable progress was made during 1980 in solid-phase synthesis of oligo- nucleotides especially due to the use of phosphotriester coupling methods.In the phosphodiester route the only successful syntheses had made use of relatively polar supports such as polyamide resins. In contrast a number of different supports have found utility in oligonucleotide synthesis via phosphotriester intermediates viz. p~lydimethylacrylamide,~~ cellu-polyacryloylmorpholide,56~57poly~tyrene,~**~~ lose,6o and ~ilica.~’.~~ Several of the routes for synthesis of oligodeoxyribonucleo- tides relied on sequential coupling of 5’-O-dimethoxytrityl-2’-deoxynucleoside 3’-arylphosphates or corresponding di- or tri-nucleotide derivatives (in pyridine in the presence of arenesulphonyl-azolides) to deoxynucleoside derivatives that were joined to the support via an alkali-labile 3’-0-~uccinate~~*~~*~~ or 3’4-~hthalate~~ linkage (see Scheme 3).Oligodeoxyribonucleotidesup to 21units long 0 0 I I o=p-o-0 I I 0 c=o I CH, I CH,CNH-II C1(o-or 0 CH, I CH,~NH-@ II (MeO)2Tr= dimethoxytrityl 0 Reagents i coupling agent pyridine Scheme 3 have been prepared by this Yields approached those obtained by comparable solution methods. It has been observed that in the case of polystyrene resins the yields in the first coupling steps were unusually poor.63 However consistently high yields were obtained with polystyrene that was suspended in tetrahydrofuran using a stepwise phosphorylation approach involving addition of 54 K.K.Ogilvie and M. J. Nemer Can. J. Chem.1980 58 1389. ” M. J. Gait M. Singh. R. C. Sheppard M. D. Edge A. R. Greene G. R. Heathcliffe. T. C. Atkinson C. R. Newton and A. F. Markham Nucleic Acids Res. 1980 8 1081. 56 K.Miyoshi. T. Huang and K.Itakura Nucleic Acids Res. 1980,8,5491. 57 K.E. Norris F. Norris and K.Brunfeldt Nucleic Acids Res. Symp. Ser. No. 7 1980 233. K.Miyoshi and K.Itakura Nucleic Acids Res. Symp. Ser. No. 7 1980,281. 59 V. N. Dobrynin B. K.Chernov and M. N. Kolosov Bioorg. Khim. 1980,6 138. 6o R. Crea and T. Horn,Nucleic Acids Res. 1980,8 2231. 61 K.K.Ogilvie and M. J. Nemer. Terrahedron Leu. 1980 21 4159. 62 A. F. Markham M. D. Edge T. C. Atkinson A. R. Greene G. R. Heathcliffe C. R. Newton and D. Scanlon Nucleic Acids Res. 1980,8 5193. 63 K.Miyoshi R. Arentzen T. Huang and K.Itakura Nucleic Acids Res. 1980,8 5507. 320 M. J. Gait compounds of type (3) in the presence of 4-(dimethylarnin0)pyridine.~*Similarly good coupling yields have been obtained by use of the phosphite triester route using silica-gel supports. These may be used in a flow system of operation. Oligodeoxyribonucleotidesup to thirteen units long4’ and oligoribonucleotides up to six units long61 have been prepared. At least in the case of oligodeoxyribonucleo- tides solid-phase synthesis has now emerged as the most efficient method for preparations on a scale suitable for most biological applications (i.e. up to a few milligrams). Two reviews on solid-phase synthesis of oligonucleotides have been published.64 Two useful mass spectral methods have been described for the analysis of protected oligodeoxyribonucleotides containing phosphotriester linkages.Use of pyrolysis-mass spectrometry has allowed characteristic ions of the protected bases and terminal substituents to be Much less destructive is 252Cf plasma desorption mass spectrometry. The molecular weight of large fragments of oligo- nucleotides can be determined and this has been used for sequence analysis of protected intermediates up to hexanucleotides.66 The ions Pb2+ and Zn” are efficient catalysts for the poly(C)-directed polymeriz- ation of an activated guanylic acid derivative guanosine 5’-phosphoroimidazolide to give oligomers up to 40 units long. When Pb2+ was the catalyst 2‘-5’ linkages predominated whereas 3’-5‘linkages were mostly formed in the presence of Zn2’ ion6’ In contrast the Pb2’-catalysed polymerization of adenosine 5‘-phosphoro- imidazolide using poly(U) as a template gave preferentially 3’-5’linkages.68.5 Analogues of Oligo-and Poly-nucleotides Poly(5-ethynyluridylic acid) prepared by enzymatic polymerization of 5-ethynyl- uridine 5‘-diphosphate has an unusually stable secondary structure with a T of 76°C. It forms a 1:l complex with poly(A) that has a T of 78”C which is considerably higher than that of the unsubstituted duplex poly(U) -p~ly(A).~~ Although 8-azidopurine mononucleotides have been used effectively as photo-affinity reagents the corresponding homopolymers have not yet been prepared because the syn conformation of the nucleoside makes enzymatic polymerization difficult.Copolymers of poly(A,8-azidoadenylic acid) have recently been prepared however. Those with low 8-azido-Ade content formed 1 2 complexes with poly(U). It is believed that the azidopurine nucleotides are constrained intrahelically and possibly adopt the anti conformation. The copolymers have been used to study the subunit topology of RNA polymerase from E. coli.” An analogue of (U)4,where the inter-nucleotide linkages are replaced by 3’-5‘ phosphite esters has been prepared via the corresponding 2,2,2-trichloroethyl 64 M. J. Gait in ‘Polymer-supported reactions in Organic Synthesis’ ed. P. Hodge and D. C. Sherrington John Wiley 1980,p. 435;N. K.Mathur C. K.Narang and R. E. Williams ‘Polymers as aids in Organic Synthesis’ Academic Press 1980,p.81. J. Ulrich M.T. Bobenrieth R. Derbyshlre F. Finas A. Guy F. Odin M. Polverelli and M. Teoule 2. Nuturforsch. Teil. B 1980,35,212. 66 C. J. McNeal S. A. Narang R. D. Macfarlane H. M. Hsiung and R. Brousseau Roc. Nut/. Acud. Sci. USA 1980,77,735. 67 R. Lohrmann P.K. Bridson and L. F. Orgel. Science 1980,208. 1464. H. Sawai Nucleic Acids Res. Symp. Ser. No. 8 1980,77. 69 E.Biala A. S.Jones and R. T. Walker Tetrahedron 1980.36,155. ’O I. L. Cartwright and D. W. Hutchinson Nucleic Acids Res. 1980,8 1675. Biological Chemistry -Part (ii)b Oligo- and Poly-nucleotides 321 phosphite triester~.~~ These latter compounds are useful intermediates also. Reac- tion with iodine and a primary alkylamine gives the corresponding alkyl-phos- phoramidate.Alternatively reaction with selenium or sulphur gives the phos- phoroselenate or phosphorothioate re~pectively.~~ Similar analogues containing 2,2,2-trichloroethyl phosphite triesters have been prepared in the deoxy-~eries.~~ A number of dinucleoside phosphates containing acyclic analogues of 2’-deoxyadenosine and 2‘-deoxythymidine which lack the 2’-methylene of the sugar have also been ~ynthesized.~~ A comprehensive review of synthetic analogues of nucleic acids has been p~blished.~’ 6 Chemical Reactions on Polynucleotides Formaldehyde undergoes two different reactions with nucleic acids and their com- ponents. In aqueous solution a rapid and reversible one causes the hydroxymethyla- tion of imino- and amino-groups. A slower reaction results in formation of cross-links.Cross-linked nucleosides have now been isolated from formaldehyde-treated RNA and DNA and it has been confirmed that the cross-links are methylene bridges exclusively between exocyclic amino-groups of Cyt Gua and Ade.76 2’,3’- 0-Isopropylidene-adenosine -cytidine and -guanosine react with aqueous formal- dehyde in ethanol to give the corresponding N-ethoxymethyl derivatives which are more stable than their N-hydroxymethyl counterparts. N-Ethoxymethylation may prove useful for base-modification of single-stranded regions of nucleic Two classic pyrimidine-specific reactions have been proposed as alternatives to treatment with hydrazine for the Maxam-Gilbert procedure for sequencing DNA. Potassium permanganate preferentially oxidizes the 5-6 double-bond of thymine residues and hydroxylamine hydrochloride preferentially adds across the 5-6 double-bond of ~yfosines.~~ The reaction of DNA with a mixture of methoxyamine and sodium bisulphite also causes modification of cytosine residues.Base-pairing is disrupted and the effect of secondary structure on the mobility of DNA fragments in gels is eliminated. This reaction has been useful in the ‘dideoxy’ method of DNA seq~encing.~’ DNA may be labelled at its 3’-end to high specific activity by addition of [a-32P]cordycepin 5’-triphosphate the reaction being catalysed by terminal trans- ferase. The labelled DNA is now resistant to degradation by 3’-exonu~lease.~~ Two chemical reagents have been used to probe the secondary and tertiary interactions of RNA.Dimethyl sulphate alkylates the N-7 position of guanosines and the N-3 position of cytidines; diethyl pyrocarbonate carbethoxylates the N-7 position of adenines. However the reactions only occur if sites are not involved in structural interactions. 32P-Labelled RNA chains may be broken at the base- 71 K. K. Ogilvie and M. J. Nemer Tetrahedron Lett. 1980 21 4145. 72 M. J. Nemer and K. K. Ogilvie Tetrahedron Lett. 1980,21,4149. 73 B. P.Melnick J. L. Finnan and R. L. Letsinger J. Org. Chem. 1980,45 2715. 74 K. K. Ogilvie and M. F. Gillen. Tetrahedron Lett. 1980 21 327. 75 A. S.Jones Int. J. Biol. Macromol. 1979 1 194. 76 Y.F. M. Chaw L. E. Crane P.Lange and R. Shapiro Biochemistry 1980,19,5525. 77 P. K.Bridson J.Jirickny 0.Kemal and C. B. Reese J. Chem. SOC. Chem. Commun. 1980 208. 78 C. M. Rubin and C. W. Schmid Nucleic Acids Res. 1980,8,4613. 79 N. S.Ambartsumyan and A. M. Mazo FEBSLett. 1980,114,265. C.-P. D. Tu and S. N. Cohen Gene 1980,lO. 177. 322 M. J. Gait modified sites and the fragments separated on polyacrylamide gels to locate the positions of modification. Probing reactions carried out at different temperatures have been used in parallel with chemical sequencing reactions to locate regions of higher-order structure in tRNA.8’ Some useful reagents for RNA-RNA and RNA-protein cross-linking have been described. Like other glyoxal-type reagents N-acetyl-N-(p-glyoxylylbenzoy1)-cystamine reacts with guanine and arginine derivatives to give acid-stable adducts.Reduction of the disulphide bonds liberates free thiol groups which can be used as sites for cross-linking to RNA or protein.82 Adenines and cytosines react rapidly below pH 6 with p-nitrophenyl or N-hydroxysuccinamido 3-(4-bromo-3-oxo- butanesulphony1)propionateat the a-halogenocarbonyl moiety. At a higher pH the activated carboxylic acid may then be used for acylation of proteins.83 The mechanism of binding of the anti-cancer drug ci~-[Pt(NH~)~Cl~l to DNA is the subject of much intense investigation. Recent evidence suggests that the drug binds selectively to (dG) (dC) sequences (n s 4). A model involving intra-strand cross-linking via co-ordination of the platinum to the N-7 positions of adjacent guanines is Another report suggests that such co-ordination could result in unusual Gua-Gua mi~pairing.~’ N.m.r.and X-ray structural studies of crystals of short duplexes that are currently under way should clarify the situation. Mechan- isms of reaction between chemical carcinogens and nucleic acids have been well reviewed recently.86 D. A. Peattie and W. Gilbert Proc. Natl. Acad. Sci. USA 1980,77,4679. ’* A. Expert-Bezancon and D. Hayes Eur. J. Biochem. 1980,103,365. 83 G. Fink H. Fasold W. Rommer and R. Brimacombe Anal. Biochem. 1980,108 394. 84 G. L. Cohen J. A. Ledner W. R. Bauer H. M. Ushay C. Caravana and S. J. Lippard J. Am. Chem. SOC.,1980,102,2487. ” R. Faggiani C. J. L. Lock and B. Lippert J. Am Chem. Soc. 1980,102 5418. 86 D. E. Hathway and G. F. Kolar. Chem.SOC. Rev. 1980,9,241.