SYNTRETIC POLYRUCLEOTIDES AND TRS AMINO ACID cope, v®, aX GARLCS BASELIO’, ALBERT J. WARERS, PETER LENGYEI, JOSEPH F. SPEYER, AND SEVERO OCHOA DEPART AENT Cy BUOQOCBEMISTRY, NEW YORR UNIVERSITY SCHOGL GF MEDICINE Gemmunicated February, 1962, In the preceding paper of this series substantial agreement was found between tie experimentally deCertained letters of the genctic code! “* and mine acid Tenlacements in nitrous acid mutants ef tobacco mesaic vine, i220 fo deaminetion of certain bases, treatmens of nucleic acids with HO, -onda to baze substitutions in the polynucicotide chains end hence te ahangea of thotr nuclectide esquence®®?, Guanine is converted te santhine, adenine te hypecmnthine, and cytosine te uracil, Uracil is nat changed, In the cage of tebaceca mosaic virus RNA, controlied HNO, treatment can lead fo deamination cf a aingle base and, as a result of the base change, to the substitution of 6n cinino acid for another at a specific location in the poly- 6 that the conversion peptide chain of the protein coat. I has been assumed of guanine to xanthine, a base which is not present in nucleic acids, is isvarted {guanine taking the place of xanthine) when the RNA is replicated 80 chat no mutation occurs. Ia a similer way, when adenine is deaminated ts hypoxanthine, which like xanthine {s not normally present in nucleic acids, cuenine 4s suppoged to teke She place of hypoxanthine after replication of the RNA. In this case, however, guanine is substituted for the original adenine (AG replacement) and mutation occurs. Deamination of cytosine is uracil leads directly 20 a C—U replacement. Hypoxanthine can be substituted for guanine, i.e. inosinic acid (2 for qguanylic acid (G), in synthetic polynucleotides with retention of their coding chsrecteristics for, as previously noted*, poly UI (5:1) was equivalent 35 poly UG (5:1) in regard to these amino acids (cysteine, glycine, tryntc= phan, ond valine) which are coded by U= and G-containing letters. In iiae with this observation, treatzon? with HNO, conferred upon poly UA (£:1} the coding characteristics of poly UG (3, o.g. cysteine and vali incorporation was substituted fer isoleucine and tyrosine incerpor- Silon. This is the test tube counterpart of amino acid replacements in HNC2 mutants of tobacco moseic virus, On the other hand, deamination of guanine to xanthine by treatmant of poly UG (5+3} with HNO, sliminated the stimulation of valine incorporation caused by the untreated pelymer. aus, contrary to hypoxanthine, xanthine cannot replace guanine in the ganstic code. An account of these experiments is given in this paper, Preparetions end methods.- These were the same as in previous work! unless otherwise specified. Poly UI (5:1) was propared with Azotohscter polyaucleotide phosphorylase! from a mixture of uridine 5"=diphosphate snd inosine 5°*=diphosphate in molar ratio 5:3, Its sedimentation coefficient was 6.33. Deamination of polynucleotides. - 720 mg of scdium nitrite wore added to a solution of 6 mg of polymer in 9 mi of 20% acetic acid. After standing at room temperature with occesiona! shaking for one hour (experiment 2, Table 2) or two hours (experiment 1, Table z), the solution was dialyzed for 5-6 aours against distilled water, with several changes, and the polymer re~ covered by lyophilization. In the case of poly UA (521), deamination of the adsaine residues appeared te be neer completion in 35 minutes as judged by extensive loss of the capacity te stimuinte the incorporation of isc- leucine info acid-inecluble products in the Escherichia coli system, Results. Experiments with poly Uk As shown in Tubie 1! poly U! {41} stimulated the incorporaticn of phenylalanine, cysteine, valinc, gly- cine, and tryptophan (experiment 1) end leucine (axperiment 2). Although she sctivity of this polymer was somowhet lower than that of poly UG (5:85 (c%,Tabie 2), the phe/cys, phe/vel, arefgly, phe/try, and pho/lev incr porstior faiiog, given in the last column of the table, were in reasonable agreement with the comesponding ratios for poly UG (5:1) (cf. Table 2 of preceding paper}. This proves that hypoxanthine can replace guanine in the genetic cacie, UNO 2 treated polymers: The results of oxperimenis wits poly UA, UG, and UC are shown in Table 2. In addition te the changes that can be specifically ascribed to deamination, treatment with HNO» re~ sulted in a pronounced overall decrease in the activity of the polymers, This wae reflected by an 85% and 70% decrease of the capacity of poly UA (5:1) to stimulate the incorporation of phenyalanine after treatment with HNO, for 2 hours (experiment 1) and 1 hour (experiment 2}, respectively. The reason for this decrease in activity is unknown. Nevertheless, the chengs in coding characteristics of poly UA, due to A--I conversion by deo~ anination, was readily epparent. Before deamination, poly UA promoted the 4acorporation of isaicucina, leucine, and tyrosine but not that of cysteine end valinc. Deaminetion largely aliminated tha capacity to stimulate in- cesporetion cf isaleucine and tyzosing and brought forth stimulation af the incorporation of cysteine sad valine. Stiovalation of leucine incorporation was largely retained in agreement with the tinding* that ioucine is ceded by 2UIA and 2UIG (er 2UM, Fable 1) besides Z2UIC letters. Treatment of paly UG (5:3) with HNC, (guantne~xanthine conversion} virtually eliminated all activity of this palymer, Phenyalanine incerpore~ Mon was drastically decreased and valine incorporation was wiped out!®, Thus, in sharp contrast to hypoxanthine, xanthine is unable te substitute for guanine in the genetic code, The experiment with pely UC (5:1) showed retention of phenyalanine (coce letter UUU) and marked Joss of serine (code letter 2U1C) incorporetion activity following treatment with HNO2 for 1 hour (C-~U conversion}, Since the activity toward phenylalanine was retaincd, contrary to the marked drop observed with poly UA, it is Ukely that a non=spectfic decrease in activity, caused by HNO treatment, wes compensated by an increase due to conver- sion of poly UC, which codes for phenylalanine and other amino acids, %0 poly U which codes for phenylalanine only. Discuesion,~ As showa in this paper HNO2 treatment of synthetic poly- nucleotides, used es artificial messengers for protein synthesis in the £s_c9li system, leads to amino acid replacements like those observed in HNO, mutants of tobaccs mosaic virus, However, multiplication cf the virus involves two processes viz, replication of the RNA and transecrip~ tion. of its code inte a polvecuitde sequence, Our model experiments with palyarcieotides relate only to the affect of HNO on the transcription ef the mesycga. fhe finding that hypoxanthine cen replace guanine in amins acid coding is nat surprising tn view of the similarity of thesa two bases with regard to hydrogen bonding, Poly I has becn shown to form DNA-ile, double-stranded helical complexes with poly C2412, the stability of the hypoxanthine~cytesine pair ia of the order of magnitude of that of the adentne-uracil pair o6 the melting out temperature of poly A + U (61° in 9.15 M NeCl-0, 235 M sodium citrate) is only about 10 degrees higher than that of poly 1+ C!3, therefore, hypoxanthine in poly UI triplets 6.9. UUN would pair with cytosine in complementary “adapter” !* griptets (AAC) af cysteine or valine transfer RNA. However, the guenine-cytosine pair, with three hydrogen bonds, is held together more tightly than the hypo<- xanihnine~cytosine pair with only two hydrogen bonds. The lower efficiency of poly Ul es compared with poly UG, noted in a previous section, might be a reflection of this difference. Our observations are in line with the finding!5 that deoxy ITP could replace (with 25% efficiency) deoxy GTP in DNA synthesis by DNA polymerese. The further finding reported in this paper that xanthine cannot replece guanine in coding, although epperentiy net explainable in terms ef hydrogen bonding propertica (an xanthine Is similar to guanine in thig respect}, 4s also in line with the failure of decxyxanthosine tiphospaate to replace deoxy GTP in the DNA polymerces vystem!>, in view of these resulta, reversal of the gusuine—xanthine conversion {guanine taking the place of xenthine) on replication of BNO>" treuted tobacco moszeic virus RNA, is unlikely, Deamination of guanine is raere likely to yield an RNA that is uneble to replicate (iethal miutatioas}. Modification of the coding charmeteristics of synthetic polynucleotides with agents ofer than HNO; might throw Moht on the mode of action of certain mutagens. Several amino acid replacements in tebecco mosaic virus protein have been brought about by treatment of the virus with bromi- nating and alkylating agenta?*?, However, the relationship between the chasrved replacements and the chemical effects of these mutagens on ths nucleic acid bases is obscure. Sugunary.- Hypoxanthine can replace guanine in amino acid coding for, like poly UG (511), poly UI ¢5:1) stimulated the incorporation of cysteine, glycine, leucine, tryptophan, and valine into acid-insoluble products in the Fe_cohi systam to the same relative extent. Treatment of synthetic polyau- cleatides with nitrous acid modiiied their ceding characteristics as expected from the deamination af adenine to hypoxanthine and cytosine to uracil. Poly UA lost its coding specificity and acquired thet of poly UI, and poly UG beat ite activity to stimulete the incorporation of serine but not thet of phenylalanine. Deamtnetion of gusnins to xanthine, by treatment of poly UG with nitrous acid, wiped out the activity of this polymer te stiraulate valine incerparation, Thus, contrary to hypoxanthine, uanthine cannot vaplacea guanine in amino acid coding. We are indalted to Horece Lozina for ckillful technical assisianca, "aided by grante from the Netional Institute of Arthritis and Matabeitc Dissases (Grant A-1845) of the U.S. Public Health Service and from the jano Goffin Childs Fund fer Medical Research, The akbreviatioas uzed ia thie paper ere the same as in previcus papers of this series. The standard shoarevistions are used for amino acids, *intemationel Postdoctoral Fellow of the National Institutes of Health, U. 8. Public Health Service, Permanent address: Institute de Quimica Fisiclégica y Patoldgica, Universidad de Chile, Santiago, Chils. +F allow of the Jane Coffin Childs Fund for Medical Research. ILengyel, Po» JeFeSpeyer, and 3.Ochoa, these PROCEEDINGS, 42, 493% (1963). “speyer,JoFe, PeLengyel, C. Basilio, and $3. Ochoa, these PROCEEDINGS, 48, 63 (1962), *Lengyel, Pee JeFo Speyer, C. Basilio, and . Ochoa, these PROCEEDINGS, 48, 282 (1962). *speyor, JF.» Po Lengyel, C. Basilio, and S. Ochoa, these PROCEEDINGS, 48, 000 (1962), *Teugita, As, Protein, Nucleic Acid, Enzyme {Tokys, 6 385 (1961). Gwittmann, H.Ge, Maturwissenschaften, 49, 729 (1961). TE eugita, A., and H, Fraenkel-Conrat, 4,.Mgqi. Bigl,, in press. Sschuster, Ha, and G.Schramm, Z,Netucioysch., 13b, 697 (1958). IGierer, A., and ¥W.Mundry, Nature, 182, 1437 (1958), 301, Schuster (Zp Naturforsch, , 15b, 293 (1960)) has reported that the giycosidic bond betwsen pentose and xanthine in RNOgtreated DNA, is more labile to acid than the pentose-guanino bond in untreated DNA. Were this the case for RNA, hydrolytic lose of xanthine could contribute to some extent to the inectivation of poly UG following treatment with HNO2. This will be investigated with poly uridylicexanthylic acid (poly UX) prepared from uridine 5°-diphosphate and xanthogine 5*=diphosphate with polynu- cleotide phosphorylase, 'IDevis, D.R., and A. Rich, Je Agi Chemin, Soc., 30, 1003 (1958). l2pavis, DR.» Nature, 186, 1030 (2960), 13poty, P., H.Boedtker, J.R. Fresco, R. Haselkorm, end M. Litt, these PROCEEDINGS, 43, 482 (1959). M4Crick, Fo H.C., in “The Biological Replication of Macromolecules", Symposin Soc, Exptl. Biol,» No. 12 (1958). 15g essman, Mo Jeo. LR. Lehman, J. Adler, S,B. Zimmerman, £. 5S. Simms, and A. Komnberg, these PROCEEDINGS, 44, 633 (1958). TABLE 1 EFFECT OF POLY UI(5: 1) ON AMINOAGID INCORPORATION IN E. COLISYSTEM® Expert- Amino acid Without With Net Ratio* ment poly UI poly UI No. i Phenylalanine 0,18 9.14 8.96 oo Cysteine 0.15 1, 28 1,13 79 Valine 0,22 2. 24 2.03 4.4 Glycine 0.17 0, 56 0. 39 23.0 Tryptophan 0.32 0. 53 0. 39 23,0 2 Phenylalanine 0.10 4.93 4.83 renene Leucine 0. 26 1,5) 1, 25 3.9 *mumoles/mg ribosomal protein. *Ratio of phenylalanine incorporation to that of the amino acid in question. TABLE 2 EPPEGT OF VARIOUS POLYNUCLEOTIDES, BEFORE AND AFTER TREATMENT WI ANO5Z. ON AMINO ACID INCORPORATION IN E, GOI! system* Exgeri~ Ardins &oid Polynucleatide ment VA (523) UG (5:3) US ¢5: 4) Nos Before After Before After Before After HNO2 4HNO2 HNO, HNO: HNOz HNO, 1 Phenylelenine 15.9 2.4 22.4 0,07 mee mouse isoleucine 4:3 2495 “me =~e ome ace Valine a On4 Sak Q one wren z Phenylajanine 12.8 3,6 wwe “em 15.8 14,2 Isoleucine 1,5 0,07 “ae --- cow ee Leucine 1.7? 0, 30 moe -—- —— a Tyroains 2x}, 9232 oman ol cones ame Cysteine g On16 am “oe one orem Valine 9 0,33 ~— = —- oe Serine < “ee 3n3 On7 “values given refer to net incorporation (in mumoles/mg ribosomal protein) after subtraction of smell blank incorporation without polynuclectide,