2094 BIOCHEMISTRY: GOLDBERG ET AL. PROC. N. A. S.

1I Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem., 193, 265(1951).

11 Maxwell, E. S., these PROCEEDINGS, 48, 1639 (1962).12 Katchalski, E., Advances in Protein Chemistry, 6, 123 (1951).13 Crick, F. H. C., Progress in Nucleic Acid Research (New York: Academic Press, in press).

BASIS OF ACTINOMYCIN ACTION, I. DNA BINDING AND INHIBITIONOF RNA-POLYMERASE SYNTHETIC REACTIONS BY ACTINOMYCIN

BY I. H. GOLDBERG, M. RABINOWITZ, AND E. REICHIDEPARTMENTS OF MEDICINE AND BIOCHEMISTRY, 'UNIVERSITY OF CHICAGO; ARGONNE CANCERRESEARCH HOSPITAL;* LABORATORY OF BIOCHEMICAL GENETICS, THE ROCKEFELLER INSTITUTE

Communicated by E. L. Tatum, September 4, 1962

Actinomycin inhibits DNA-dependent RNA synthesis of intact cells' and enzymesfrom mammalian2 and bacterial' sources but not the growth of RNA-viruses inanimal cells." 4 The inhibition of RNA synthesis appears to account completelyfor the antibiotic and cytotoxic action of actinomycin and is attributable to thefact that the antibiotic is tightly bound' I to DNA.

Structural changes in the actinomycin molecule which affect its biological activityalso influence, in a parallel way, its binding to DNA and its inhibition of the RNApolymerase reaction.3 The experiments described below suggest a basis for theeffect of actinomycin in that the base composition of DNA primers may influencethe extent of actinomycin binding and thereby the degree of inhibition of the dif-ferent reactions catalyzed by RNA-polymerase. The binding of actinomycin ap-pears to be absolutely dependent on the presence of guanine residues in the DNAprimers. Kersten7 showed that high concentrations of deoxyguanosine and severalother guanine and adenine compounds could alter the spectral properties of actino-mycin solutions. Our data in this and a subsequent paper' show that only guanineresidues in DNA are indispensable for actinomycin action.

Recently, Hurwitz et al. have reported on the effect of actinomycin on some as-pects of bacterial RNA-polymerase activity.9 Several of the findings of these au-thors appear to differ from those recorded below.

Materials and Methods.-The standard assay conditions used in the experiments and the isola-tion of RNA-polymerase from E. coli B were as described by Chamberlin and Berg,10 and theincubations were carried out for 15' at 370 unless otherwise noted.

Incorporation of precursors into RNA under standard conditions in the absence of DNA wasless than 0.02 m/Amoles/15' for 4 NTPf and smaller for single NTP; these control values werealways subtracted from the results recorded below. All reactions contained 4.1 ,ug enzyme pro-tein/0.25 ml. Synthesis of acid-insoluble polyribonucleotides was measured by the incorporationof one of the following precursors: C14-ATP, obtained from Schwarz Bioresearch; UTP32, 4pUTP82, CTP32, GTP32, synthesized as previously described," in which the proximal phosphatewas radioactive. Nucleic acid concentrations were based on the phosphorus content determinedaccording to Fiske and Subba-Row."2 Protein was measured according to Lowry et al."s Actino-mycin concentrations were based on the absorption of a standard solution of 25 ug/ml, giving anO.D. (440 mMA) of 0.358 in 0.01 M Tris pH 7.4 with 1 cm light path. Absorption of pure actino-mycin is appreciably greater than this value, which was established several years ago for a prep-aration containing inert material.

Dow

nloa

ded

by g

uest

on

June

8, 2

020

VOL. 48, 1962 BIOCHEMISTRY: GOLDBERG ET AL. 2095

DNA preparations: All DNA solutions were made in 0.01 M Tris (pH 7.4)-0.01 M NaCl, ex-cept for crab dAT, which was dissolved in 0.75 M NaCl-0.05M phosphate (pH 6.7). Highly poly-merized calf thymus DNA was purchased from Sigma Chemical Co., St. Louis, Mo.; DNA fromMicrococcus lysodeikticus and Tetrahymena pyriformis was prepared by the method of Marmur,14according to a modification suggested by N. Sueoka. Apyrimidinic DNA was prepared accordingto Habermann,'5 and apurinic DNA according to Tamm et al."1 from calf thymus DNA. Syn-thetic deoxyadenylic-deoxythymidylic (dAT) and deoxyguanylic-deoxycytidylic polymers (dGC),crab dAT,'7 a preparation of apyrimidinic DNA, and DNA from bacteriophage OX-174 weregenerously provided by A. Kornberg, N. Sueoka, V. Habermann, and R. L. Sinsheimer respec-tively. Nonradioactive NTP were obtained from Pabst Laboratories and Sigma Chemical Co.We thank Dr. S. B. Weiss for a gift of GTP32.

Results.-Binding of actinomycin to DNAs of varying base composition: Guanineresidues are indispensable for actinomycin binding to DNA. The data in Figure1 show that synthetic dAT copolymer and apurinic acid do not bind actinomycin.However, the naturally occurring dAT copolymer of crab testis, which containsapproximately 2.5 per cent GC,18 does cause slight optical changes in actinomycinsolutions which qualitatively resemble thosenormally produced by native DNA. Since sto Coff thymus DNAapyrimidinic DNA and dGC polymer both 005_complex actinomycin, binding requires only the _005guanine residue of the GC base pair. -0 a EThe base composition of DNA may affect its +0 10- Apyrimidinic DNA

binding of actinomycin as shown in Figure 2. +005

M. lysodeikticus DNA (G content 37%) 17 binds °actinomycin much more effectively than does -0 05

T. pyriformis DNA (G content 12.5%),19 calfd

thymus DNA (G content 22%)17 being inter- ADD + dGCmediate. All three DNA preparations were -005equally "native" as judged by "hyperchromic +0 04 Crab dATeffect" following acid hydrolysis or alkali +0 f.treatment. Binding by nucleic acid is influ- -002 Eenced also by the intactness of its helical struc- -004 -ture as shown by two facts: (a) heating the +002 Synthetic dAT0 ~~~~~~~~~~~~~~~~~~~~~0DNA (1000, 10'), followed by rapid cooling, sig- -

nificantly decreases the amount of actinomycin +005 Apurinic DNAbound; (b) despite its higher G content (24%) -005 -LO]0 42i5 450 45 50

single-stranded20 4X-174 DNA binds actinomy- a 2 4Z5 500 mp

cin no more effectively than T. pyriformis DNA. FIG. 1-Difference spectra, ob-tained in Cary SpectrophotometerInhibition of polyribonucleotide synthesis by Model 11, of actinomycin solutions

actinomycin: (a) Heteropolymer formation- (O.D. 440 my = 0.78, 1 cm light path)read against the same solutions con-The extent of incorporation into RNA of a taining the DNA preparations shown.given NTP by RNA-polymerase reflects the Note the differences in scale of the ordi-

nates. The observed differences forbase composition of the DNA primer. For ex- synthetic and crab dAT were obtainedample, DNA from M. lysodeikticus primes the under similar conditions but with a5 cm light path. The concentrationsincorporation of CTP more efficiently than of DNA-P were: calf thymus 108 mjudoes calf thymus DNA. As shown in Figure moles/ml; apyrimidinic 1.5umoles/ml;3,theincorprationof bot UTP ad CTP synthetic dAT 1.65 O.D. units (260 m1A3, the incorporation of both UTP and CTP estimated 276 mM moles/ml; crab dAT

into RNA is more sensitive to actinomycin 0.429 O.D. units (260 m1A)/ml, esti-mated 71.6 mgsmoles/ml; dGC 158

when the reaction is primed with M. lyso- mjumoles/ml; apurinic 1.61.umoles/ml.

Dow

nloa

ded

by g

uest

on

June

8, 2

020

2096 BIOCHEMISTRY. GOLDBERG ET AL. PROC. N. A. S.

deikticus DNA than with DNA from0.25-e e f 4 calf thymus or T. pyriformis. This

// / holds throughout a wide range of actin-0.20 X// omycin concentrations and also (vide

EL / /infra) for the NTP-PP exchange reac-

0O 5 // * / tion. This finding is noteworthy since:C _ (a) at the concentrations of DNA used,A90.|G-/uf/ its capacity for binding actinomycin is

not saturated below antibiotic levels of0.05 approximately 9 ,jg/ml; (b) on the basis

of arguments presented elsewhere,3 it50 000 0 sCo 4000 0oo 600 is unlikely that more than a very small

13DNA-FP mA~moles/ml fraction of actinomycin is not bound inFIG. 2.-Actinomycin binding by DNA the presence of adequate concentrations

different base composition, measured by change of native DNA. The equal amountsin absorbance at 425 mpA (initial OGD. 425 m1A owithout DNA = 0.635). All solutions 0.01 M of the three types of DNA employed inNaCl - 0.01 M Tris (pH 7.45). x-x M. this experiment, corresponding to posi-lysodeikticus DNA; 0-heated M. lysodeikticus * *i*-* calf thymus DNA; 0 heated calf thymus tions of identical slope on the respec-DNA; O-O T. pyriformis DNA; O-DNA from tive velocity/DNA concentrationbacteriophage OX-174. curves, must accordingly be presumedto have bound the same quantity of actinomycin. Therefore, it is of interestthat the antibiotic impairs the priming action of M. lysodeikticus DNA for RNAsynthesis to a significantly and reproducibly greater extent than that of calfthymus or T. pyriformis DNA. This observation might result if the binding sitesfor actinomycin, of which guanine is in some way an indispensable component,were evenly distributed throughout M. lysodeikticus DNA so that no regions of ap-preciable length are spared the inhibitory action of the drug. DNA from calf thy-mus and T. pyriformis could then be expected to be more heterogeneous with respectto structural requirements for actinomycin binding and consequent inhibition ofRNA synthesis. This expectation may be related to the capacity of these DNApreparations to prime the synthesis of poly A and poly U.

(b) Effect of actinomycin on dAT-primed RNA synthesis-Synthetic dAT copoly-mer consists of perfectly alternating deoxyadenylic and thymidylic acids,21 and,

100

.C a.0CS

E A B FIG. 3.-Actinomycin C1 inhibi-T1\@X < tion of incorporation of CMP32 (A)- 50 and UMP32 (B) into RNA. Stand-

ard conditions, 196 m/umoles DNA-o P/ml of each preparation. 100 per° 50 cent incorporation with CTP'2 (sp.

.act. 9.1 X 105 cpm/iimole) was2.99, 2.36, and 2.13 mlsmoles, and

0 N 0.2 0.4 0.6 0.8 with UTP3' (sp. act. 4.2 X106Actinomycin C,] s9g ml cpm/umole) 3.10, 0.67, and 4.09mIumoles for calf thymus (0-O),

c 2 M. lysodeikticus (*-*), and T.23 4 5 pyriformis (x-x) DNA respec-[Actinomycin Cl] ,g/ml tively.

Dow

nloa

ded

by g

uest

on

June

8, 2

020

VOL. 48, 1962 BIOCHEMISTRY: GOLDBERG ET AL. 2097

TABLE 1EFFECT OF ACTINOMYCIN C1 ON RNA SYNTHESIS PRIMED BY SYNTHETIC AND CRAB "dAT"

Actinomycin Cl mpmoles UMP32 Per centPrimer (pg/ml) incorporated inhibition

Synthetic dAT 2.15it "( 100 2.14 0.5

Crab dAT 1.98it it 100 1.50 24.3The reaction mixture contained 0.005 O.D. units (260 m;&) of the relevant primer in a final

volume of 0.25 ml UTP32 (1.16 X 106 cpm/pmole) was the radioactive precursor. All four NTPswere present at 400 m;tmoles/ml each throughout to maintain identical reaction conditions forboth primers. Each vessel also contained NaCl 33.6 umoles/ml and potassium phosphate bufferpH 6.7, 2.4 pmoles/ml 15', 37°.

when used to prime RNA synthesis, the product contains alternating adenylic anduridylic residues.22 The data in Table 1 show that this RNA synthesis is totallyresistant to actinomycin, reflecting the absence of actinomycin binding to the primer(Fig. 1). The lack of effect of actinomycin on dAT-primed RNA synthesis is incontrast to the action of proflavine, which can strongly inhibit the same reaction(50% inhibition at 1.5 X 10-6 M proflavine).26The light component of the Cancer borealis testis DNA contains predominantly

adenine and thymine.18 In an experiment in which this DNA was used to primethe DNA-polymerase reaction, dGTP and dCTP were required for completion ofthe synthetic reaction, the product contained approximately 2.6 per cent GC, andthe GC components were scattered through an otherwise almost perfectly alternat-ing sequence of deoxyadenylic and thymidylic acids. 18 We have made similar ob-servations in using this substance to prime RNA synthesis. The omission ofGTP and CTP, which does not affect the reaction primed by synthetic dAT, mark-edly reduced the RNA synthesis directed by crab dAT. tWhen UMP32 was incorporated into RNA in the presence of crab dAT and the

product hydrolyzed in alkali, the distribution of radioactivity in the 2' (3') mono-nucleotides isolated following paper electrophoresis (95.2, 2.9, 1.7, 0.3% in A, U,G. C, respectively) closely resembled that found for DNA replication with TTP32as radioactive precursor.18 In addition, the small amount of GC present in crabdAT suffices to render its priming of RNA synthesis partially, but significantly,susceptible to actinomycin, in contrast to the absolute resistance of the reactionprimed by synthetic dAT (Table 1). Actinomycin lowers the rate of synthesis butdoes not alter the base composition of the RNA produced with crab dAT as primer.The low frequency of guanine residues in crab dAT implies the existence of longguanine-free sequences incapable of binding and therefore insensitive to inhibitionby actinomycin. For this reason, it was expected that the RNA produced in theactinomycin-inhibited reaction might contain a lower than normal proportion ofC and G, but this was found not to be the case.

(c) Effect of actinomycin on dGC-primed RNA synthesis-Synthetic dGC consistsof a mixture of homopolymers of deoxyguanylic and deoxycytidylic acids,23 andprimes the synthesis, by RNA-polymerase, of polyribonucleotides containing onlyG and C.10 We have found that the incorporation of GMP32 and CMP32 into RNAsynthesized in the presence of dGC proceed independently (Table 2), in contrast toRNA synthesized with dAT as primer. That is, dGC can prime the synthesis ofRNA homopolymers, whereas dAT, whose nucleotide sequence is an alternating

Dow

nloa

ded

by g

uest

on

June

8, 2

020

2098 BIOCHEMISTRY: GOLDBERG ET AL. PROC. N. A. S.

TABLE 2POLY C AND POLY G SYNTHESIS PRIMED BY dGC AND M. lysodeikticus DNA

mpmolesPrimer Precursors incorporated

dGC GTP32 1.07dGC GTP32 + CTP 0.99dGC CTP32 0.16dGC CTP32 + GTP 0.18M. lysodeikticus DNA CTP32 0.44M. lysodeikticus DNA CTP32 + ATP + GTP X UTP 3.52

Standard reaction conditions, incubation time 30', 8.2 ,ug enzyme protein/ml. 7.4 mpumoles of dGCand 49 mpmoles M. lysodeikticus DNA were used. The specific activity of CTP32 was 2.61 X l10cpm/pmole and of GTP82 2.38 X 106 cpm/Mmole. With dGC, CTP32 and GTP32 were present at 200mpsmoles/0.25 ml when used alone; otherwise, all NTPs were present at 100 mjumoles/0.25 ml.

one, cannot.10 It follows that these homopolymer syntheses are associated with thepresence of long repeating sequences in the primer. dGC-primed synthesis of bothhomopolymeric polyribonucleotides may be inhibited by actinomycin (Fig. 4)although the concentration required is greater than that required for comparableinhibition of heteropolymer reactions primed by DNAs containing all four bases.

(d) Effect of actinomycin on homopolymer synthesis primed by various DNA prepa-rations-Chamberlin and Berg reported'0 that RNA-polymerase is capable of catalyz-ing the synthesis of homopolymeric ribopolynucleotides when certain DNA prepara-tions (high in AT) are used to prime the reaction. The data in Figure 4 show theeffect of actinomycin on the calf thymus DNA-primed synthesis of polyadenylic,polyuridylic, and polypseudouridylic acids. The formation of homopolymers issubstantially more resistant to actinomycin than that of heteropolymers. Veryhigh concentrations of actinomycin do significantly inhibit homopolymer synthesisin contrast to a previous report.9 This inhibition must be specific for a biologically

,00' active actinomycin since (a)cr&* polyCdGC)Wequivalent concentrations of

* C (M.lysodeikticus DNA) biologically inactive deriva-E

80o U (colf thymus DNA)

0) _N \ \ ^" iU(@' 1q tives (desamino-actinomycino " A " and N-p-aminophenylactino-

0 mycin) which do not bind toDNA3, 24 and which do notaffect heteropolymer synthesisare totally innocuous to homo-polymer formation, and (b) abiologically active actinomycin

20 40 60 80 100 120 140 which interferes with homo-[Actinomycin Cl] 1jq/ml

FIG. 4.-Effect of actinomycin C1 on homopolyribo- polymer synthess does so bynucleotide synthesis, primer DNA preparation indicated binding to the DNA primerin parentheses in the figure. The reactions primed with and not by nonspecific inhibi-dGC contained 400 mlumoles/ml of each of the four NTP;this addition does not affect the nature of the product, but tion of the enzyme (viz., syn-increases the reaction rate somewhat. All other reactions thetic dAT).contained 400 mimoles/ml of the single relevant NTPunder standard conditions. 100 per cent incorporation (e) Effect of actinomycin onwas as follows: poly C (dGC), 0.10 mymoles; poly G RNA synthesis primed by(dGC) 0.80 mjsmoles; poly C§ (M. lysodeikticus) 0.44mjumoles; poly A 1.48 mumoles; poly U 0.196 mjimoles; ;X-174 DNA-Heated DNA,poly &, U 0.138 m/umoles. The concentrations of DNA-P a portion of which has lost itswere: calf thymus DNA, 1.72 mjAmoles/ml; M. lyso-deikticus, 196 mumoles/ml; dGC, 29.6m~umoles/ml. helical configuration, and X

Dow

nloa

ded

by g

uest

on

June

8, 2

020

VOL. 48, 1962 BIOCHEMISTRY: GOLDBERG ET AL. 2099

174 DNA, known to be single-stranded,20 bind actinomycin less effectively thandouble-stranded DNA of identical base composition (Fig. 2). RNA-polymerase issaturated by a low level of OX-DNA (12 m/hmoles DNA-P/ml), as has been shownfor heated DNA.25 Below this level, the velocity/DNA concentration curves forDNAs from 4X-174 and calf thymus are identical and permit a comparison oftheir actinomycin-susceptibility. The actinomycin-sensitivity of the reactionprimed with OX-DNA is approxi- 2.5mately half that found for calfthymus DNA (Fig. 5).The concentration of 4X-DNA O 2.0 /

required to saturate the synthetic E4/reaction is considerably higher than E/for the exchange reaction. 1Discussion.-Any hypothesis con- /

cerning the mode of action of ac-tinomycin should be required to *5accommodate the following facts:(a) actinomycins which inhibit c 0.5 -

RNA-polymerase do so by binding 'to DNA; derivatives which are notbound fail to inhibit RNA poly- 40 80 1 0 160 200merase and are biologically inactive; [DNA-P] m,4moles/mi(b) DNAs (synthetic dAT) lackingguanine and cyntotin do notcbind FIG. 5.-Comparison of reaction velocity at vary-guanine and cytosine do not bind ing DNA concentrations, and sensitivity to actino-biologically active actinomycins and mycin of calf thymus (-e*) and 4X-174 DNAprime RNA synthesis with complete (X X). Standard conditions. Actinomycin C,immunity to actinomycin; (c) crab at 0.2 sg/ml. Calf thymus with actinomycindAT resembles synthetic dAT but (O-O). OX with actinomycin (A A). UTP32

was the radioactive precursor.contains approximately 2.5 per centGC; it binds actinomycin slightly, and its priming of RNA synthesis is inhibited byactinomycin; (d) apurinic DNA does not bind actinomycin, whereas apyrimidinicDNA does, though less efficiently than the corresponding native DNA; (e) dGCpolymer binds actinomycin, and RNA synthesis primed by dGC is sensitive toactinomycin; (f) a portion of DNA-dependent poly A synthesis is resistant toactinomycin, as is oligothymidylate-directed poly A formation,8 whereas poly Cand poly G formation is completely suppressed by actinomycin; (g) the ability ofdifferent naturally occurring native DNA preparations to bind actinomycin reflectstheir guanine content.The following conclusions can be drawn: (1) The presence of guanine in a poly-

deoxyribonucleotide is indispensable for actinomycin binding to DNA and forinhibition of RNA synthesis. The finding' that adenine compounds may alsocause spectral shifts in actinomycin solutions appears unrelated to the binding ofactinomycin by DNA. (2) While guanine in a DNA preparation is required foractinomycin binding, optimal binding is favored by the native helical configuration.(3) Since dGC-directed reactions are more resistant to actinomycIn than thosedirected by calf thymus or M. lysodeikticus DNA, it appears likely that optimalbinding is influenced by the nucleotide sequences in the immediate environment of

Dow

nloa

ded

by g

uest

on

June

8, 2

020

2100 BIOCHEMISTRY: GOLDBERG ETr AL. PROC. N. A. S.

the guanine residues, perhaps in the contralateral strand, and that long sequencesof deoxyguanylic acid do not present the most, effective binding sites for actinomycin.This conclusion is also supported by the actinomycin-sensitivity of Al. lysodeikticusDNA-direeted poly C formation. (4) Since apurinic DNA does not bind actino-mycin while dGC does, actinomycin binding to dGC probably involves only onestrand the polydeoxyguanylic strand. The independence of dGC-directedpoly G and poly C formation suggests that, RNA-polymerase replicates each strandof the primer independently. However, actinomycin can block both poly G andpoly C formation directed by dGC. Therefore, actinomycin bound to only onestrand of the primer can inhibit the synthesis of polyribonucleotides from bothstrands. This may be due to the bulky peptide chains of actinomycin which mayalso account for the fact that this agent can suppress RNA synthesis much moreeffectively than proflavin.2 (5) DNAs of high AT content prime the synthesis ofpoly A and poly U but not of poly C, whereas AM. lysodeikticus DNA (74% GC)and dGC can prime the formation of poly C but not poly A or poly U. Therefore,ribonucleotide homopolymer formation is likely to reflect complementary homo-polymeric regions in the primer. The partial sensitivity to actinomycin of polyA and poly U synthesis suggests that these regions may vary in length and containan occasional binding site of low affinity for actinomycin requiring a high concen-tration of the antibiotic for its saturation. Alternatively, the high concentrationsof actinomycin required for inhibition of homopolymer synthesis might act byinducing profound configurational distortions in the primer.Although some experiments supporting the above conclusions have been per-

formed with concentrations of actinomycin higher than required to inhibit theformation of acid-insoluble polyribonucleotides, the effects of these concentrationsretain the full specificity of actinomycin action. The relative actinomycin-resist-ance of the NTP-PP8 I exchange is not unexpected, since concentrations of the anti-biotic which could prevent the growth of polynucleotide chains to the size requiredfor acid-insolubility would still leave many sequences free for priming the exchangereaction. Acid-soluble oligonucleotides may be formed under these conditions.

We thank E. L. Tatum for his interest and encouragement, the American Cancer Society (I. H.G.), the Helen Hay Whitney Foundation (E. R.), the National Science Foundation, and the Na-tional Institutes of Health for support. We thank 1I. Jaffe for collaboration in the spectropho-tometry of the dAT preparations.

* Operated by the University of Chicago for the U.S. Atomic Energy Commission.t Abbreviations: NTP-ribonucleoside triphosphate; ATP, CTP, GTP, UTP; AMP, CMP,

G(MP, UMP; poly A, poly C, poly G, poly U-5'-triphosphates, 3'(2') monophosphate, 3'-5' poly-meric monophosphates respectively of adenosine, cytidine, guanosine, uridine. TTP-thymidinetriphosphate. DNA-deoxyribonucleic acid: RNA-ribonucleic acid.

$ This inhibition could be overcome by doubling the concentration of ATP and UTP32 nor-mally used with enzyme assay. However, on electrophoresis of the alkaline hydrolysate of theRNA produced in this manner, radioactivity was detectable in the areas corresponding to GMPand CMP, showing that the commercially available NTP contained impurities.

§ We have found that Al. lysodeikticuls DNA can prime the formation of poly C (but not poly0, poly A, or poly U) whereas calf thymus DINA cannot. This homopolymer synthesis is moresusceptible to actinomycin than those primed by calf thymus DNA and resembles the reactionsprimed by dGC. The poly C was characterized as follows: paper electrophoresis of an alkalinehvdrolv^sate of the product showed 97.5 per cent of the radioactivity associated with the CMPregion, 2.5 per cent with the UMP region (probably the result of deamination occurring during

Dow

nloa

ded

by g

uest

on

June

8, 2

020

VOL. 48, 1962 BIOCHEMISTRY: GOODMAN AND RICH 2101

hydrolysis). When the CMP region was eluted and chromatographed on paper in a borate buffersystem, all the radioactivity moved as 2'(3') CMP.

'Reich, E., R. M. Franklin, A. J. Shatkin, and E. L. Tatum, Science, 134, 556 (1961).2 Goldberg, I. H., and M. Rabinowitz, Science, 136, 315 (1962).Reich, E., I. H. Goldberg, and M. Rabinowitz, Nature, in press.

4Reich, E., R. M. Franklin, A. J. Shatkin, and E. L. Tatum, these PROCEEDINGS, 48, 1238 (1962).5 Kirk, J. M., Biochim. Biophys. Acta, 42, 167 (1960).6 Rauen, H. M., H. Kersten, and W. Kersten, Z. physiol. Chem., 321, 139 (1960).7 Kersten, W., Biochim. Biophys. Acta, 47, 610 (1961).8 Goldberg, I. H., M. Rabinowitz, and E. Reich, these PROCEEDINGS, in press (vol. 49, no. 1,

1963).9 Hurwitz, J., J. J. Furth, M. Malamy, and M. Alexander, these PROCEEDINGS, 48, 1222 (1962).10 Chamberlin, M., and P. Berg, these PROCEEDINGS, 48, 81 (1962).1' Goldberg, I. H., and M. Rabinowitz, Biochim. Biophys. Acta, 54, 202 (1961).12 Fiske, C. H., and Y. Subba-Row, J. Biol. Chem., 66, 375 (1925).13 Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem., 193, 265

(1951).14 Marmur, J., J. Mol. Biol., 3, 208 (1961).15 Habermann, V., Coll. Czechosl. Chem.. Comms., 26, 3147 (1961).'6Tamm, C., M. E. Hodes, and E. Chargaff, J. Biol. Chem., 195, 49 (1952).'7Sueoka, N., J. Mol. Biol., 3, 31 (1961).1&Swartz, M. N., T. Trautner, and A. Kornberg, J. Biol. Chem., 237, 1961 (1962).19 Sueoka, N., in Cellular Regulatory Mechanisms, Cold Spring Harbor Symposia on Quantitative

Biology, vol. 26 (1961), p. 35.20 Sinsheimer, R. L., J. Mol. Biol., 1, 43 (1959).21 Schachman, H. K., J. Adler, C. M. Radding, I. R. Lehman, and A. Kornberg, J. Biol. Chem.,

235, 3242 (1960).22 Furth, J. J., J. Hurwitz, and M. Goldmann, Biochem. Biophys. Res. Comm., 4, 431 (1961).23Josse, J., A. D. Kaiser, and A. Kornberg, J. Biol. Chem., 236, 864 (1961).24 Muller, W., Naturwiss., 49, 156 (1962).2 Weiss, S. B., Fed. Proc., 21, 120 (1962).26 Unpublished observations.

FORMATION OF A DNA-SOLUBLE RNA HYBRID AND ITSRELATION TO THE ORIGIN, EVOLUTION, AND DEGENERACY OF

SOLUBLE RNA

BY HOWARD M. GOODMAN AND ALEXANDER RICH

DEPARTMENT OF BIOLOGY, MASSACHUSETTS INSTITUTE OF TECHNOLOGY

Communicated by Paul Doty, September 25, 1962

It has been known for a long time that transfer or soluble RNA (sRNA*) mole-cules play a central role in the organization of amino acids into polypeptide chainsduring protein synthesis. Individual sRNA molecules combine with a particularamino acid to produce a complex which is active on the ribosomal particle. Recentexperiments' make it likely that a sequence of nucleotides in sRNA carry thespecificity for determining the position of the amino acid in the polypeptide chain.However, as yet little is known regarding the origin of sRNA. These moleculescould arise from DNA in a manner similar to the production of messenger RNA.On the other hand, it has been demonstrated that the sRNA molecule is largelyfolded back upon itself with a regular system of hydrogen bonding,2 and this has

Dow

nloa

ded

by g

uest

on

June

8, 2

020