Vo.. 51, 1964 BIOCHEMISTRY: D. NATHANS 585 PUROMYCIN INHIBITION OF PROTEIN SYNTHESIS: INCORPORATION OF PUROMYCIN INTO PEPTIDE CHAINS By DanieL NaTHANns DEPARTMENT OF MICROBIOLOGY, THE JOHNS HOPKINS UNIVERSITY SCHOOL OF MEDICINE Communicated by Fritz Lipmann, February 24, 1964 The similarity in structure between puromycin and aminoacyl-sRNA (Fig. 1), first noted by Yarmolinsky and de la Haba,’ led to the hypothesis that the anti- biotic inhibits protein synthesis by acting as an analogue of esterified sRNA. Puromycin blocks protein synthesis after aminoacyl-sRNA formation,» ? and at the same time it leads to the accumulation of small peptides. Both of these effects appear to be due to the splitting of ribosome-bound peptidyl-sRNA,‘ which results in release of incomplete peptide chains.5~’? If puromycin does act as an analogue of “charged” sRNA, one might anticipate the possibility that it could substitute for an incoming a b aminoacyl-sRNA as the acceptor of the carboxy]- mA wo activated peptide, forming peptidyl-puromycin, nocn <, 2 wocn < CO thus ending growth of the polypeptide.” * Such ° oy a mechanism is supported by the finding of Allen NH 3 and Zamecnik’ that when puromycin labeled in ort queen {Sony Os¢-cn-n the amino acid moiety is incubated with reticu- ‘ “ locyte ribosomes, a portion of the radioactivity myo Ta ence ot eure precipitates with the released polypeptides, the sRNA (b). amount being equivalent to the N-terminal valine of these peptides. In order to establish, however, that puromycin replaces amino- acyl-sRNA as an acceptor of the peptide chain it is desirable to show unequivo- cally that puromycin as such becomes linked to polypeptide via a peptide bond in the course of protein synthesis. The study reported here was undertaken to de- termine whether or not peptidyl-puromycin is formed when susceptible cells are ex- posed to the antibiotic. Materials and Methods.—H?-puromycin, labeled in the methoxy group, was prepared under the N(CH5), z 586 BIOCHEMISTRY: D. NATHANS Proc. N. A. 8. guidance of Dr. Amos Neidle of Columbia University, New York, by first methylating N-carbo- benzoxy-L-tyrosine (Mann Research Laboratories, New York, N. Y.) with H*-dimethyl sulfate (120 mc/mmole, New England Nuclear Corp., Boston, Mass.) according to the procedure of Pitt- Rivers and Lerman.* The resulting H*-N-carbobenzoxy-O-methyl]-L-tyrosine was then condensed with the aminonucleoside of puromycin (obtained from Dr. Leon Goldman of the Lederle Labora- tories, Pearl River, N. Y.), following the procedure of Baker et al. to give H*-N-carbobenzoxy- puromycin (mp 193-195°). H*-N-carbobenzoxypuromycin was reduced with hydrogen in the presence of palladium oxide, and the resulting H*-puromycin was crystallized from alcohol (mp 160-163°). The product was analyzed by paper chromatography in butanol-acetic acid-water (4:1:5) and by paper electrophoresis at pH 4.7 in pyridine acetate. In each case a single ultraviolet-absorbing spot was found corresponding with puromycin and containing 90-93% of the recoverable radio- activity. Ultraviolet absorption spectra of the H*-puromycin corresponded closely with that of puromycin; wavelengths giving maximal and minimal absorption at pH 1, 7, and 11 were the same as with puromycin. Prior to use in the experiments to be described the H?-puromycin was further purified by paper electrophoresis. L-phenylalanyl-puromycin (9-(3-deoxy-3-L-phenylalanyl-O-methy)-L-tyrosinamido-g-D-ribv- furanosy])-6-dimethylamino-9H-purine) was prepared by condensing N-carbobenzoxy-L-phenyla- lanine (Mann Research Laboratories) with puromycin.” Glycyl-puromycin (9-(3-deoxy-3-glycyl- O-methyl-L-tyrosinamido 6-D-ribofuranosyl)-6-dimethylamino-9H-purine) and puromycin were supplied by Dr. Goldman. O-methyl-L-tyrosine was obtained by the reduction of N-carbobenzoxy- O-methyl-L-tyrosine prepared as described above. Chymotrypsin, trypsin, and carboxypeptidase A (treated with diisopropylfluorophosphate) were purchased from Worthington Biochemical 80, 60) o q = aot $ = ei 3 $a 3 a z 3 3 et - 2 e s © 207 é "y te 30 30 60 30 spe oO 60 MINUTES Fie. 2.~The effect of dif- ferent concentrations of puro- mycin on the growth of E. colt. Aliquots of a log-phase culture were continuously aerated in bubbler tubes at 37° and samples removed at various times for optical density readings at 600 mp. Puromycin concentrations were as follows: zero (0), 3.2 X 10-§ M (ao), 8 X 10% M (@), 16 5 2 °o 1 oO BX 500 2 3 4 5 6 +6 +4 42 -2 -4. -6 -8--10 -12 -14 -16 SEGMENT NUMBER Fic. 6.— Electro) pherogram of radioactive material liberated by tryp- sin-chymotrypsin jon and having the mobility of puromycin on chromatography. E lectrophoresis was carried out at pH 4.5 in pyridine- acetic acid-butanol (2.5, D5, 5% v/v) on Whatman 3 MM paper at 30 V/em for 2.5 hr. On the left i is a tracing of the ultraviolet-absorbing spot corresponding to added puromycin (P). X represents the radio- active material. On the right are plotted the counts per minute in the various segments of paper. 590 : BIOCHEMISTRY: D. NATHANS Proc. N. A. 8. liberation of puromycin by chymotrypsin and the results with carboxypeptidase mentioned above, it appeared that the amide bond of puromycin was resistant to chymotrypsin. To demonstrate this more directly the action of chymotrypsin on synthetic aminoacy-puromycins was examined. As shown in Figure 5, when phenylalanyl-puromycin or glycyl-puromycin was treated with chymotrypsin, no aminonucleoside was formed, i.e., the amide bond between O-methyl-tyrosine and the 3’ amino group was not split. . With phenyl- alanyl-puromycin, phenylalanine and puromycin were found. Therefore, chymo- trypsin would be expected to liberate puromycin from a polypeptidyl-puromycin where the last amino acid of the polypeptide is phenylalanine or tyrosine. Simi- larly (though this was not directly tested), trypsin should lead to free puromycin wherever lysine or arginine is the last amino acid of the polypeptide chain. In order to identify the radioactive compound(s) liberated by trypsin-chymotryp- sin digestion of the H*-puromycin product and moving with puromycin on chro- matography, this material was eluted and electrophorized. As shown in Figure 6 about 65 per cent of the radioactivity migrated with puromycin. Moreover, when 100 {go 8 ia 80 13 "2 60 a7 a 3 40 COUNTS PER MINUTE Om wR Oey BOs: 0 24 6 8 10 12 14 16 HOt tt tt ore Aree SEGMENT NUMBER PA Fie. 7.—Chromatography of the acid-treated putative H* puromycin in butanol-acetic acid-water. On the left is a tracing of the chromatogram stained with ninhydrin. The line at the top represents the solvent front. Legends are the same as given under Fig. 5. A indicates acid-treated; X, the radioactive ma- terial. On the right are plotted the counts per minute in the various segments of paper. an aliquot of the material purified by consecutive chromatography and electro- phoresis was treated with 1 N HCl at 100° for 15 min to depurinate the putative puromycin, most of the radioactivity then moved on chromatography with the depurination product of puromycin (Fig. 7). Finally, the purified radioactive ma- terial was crystallized together with carrier puromycin to constant specific activity (Table 2). We conclude therefore that puromycin is incorporated as such into the polypeptide chain of protein by means of a peptidase-sensitive bond. Since the molecule has no carboxyl or similar group capable of continuing the chain, it must be present at the carboxyl terminal end of the polypeptide. Discussion.—The finding of puromycin within (and therefore at the end) of Vot. 51, 1964 BIOCHEMISTRY: D. NATHANS 591 TABLE 2 Cpm per 1000 . O.D. units Cpm O.D. units Starting material 1760 697 396 2nd crystallization 388 102 263 3rd crystallization 510 128 251 4th crystallization 329 92 279 To an alcoholic solution of the material purified by consecutive chromatography and electro- proresis (as described in the text) 87 zmoles of puromycin dihydrochloride were added, followed an equivalent amount of cyclohexylamine. The puromycin was dissolved at 65-70° and lowed to crystallize overnight at 4°. The crystals were washed with aleohol at —15°, redis- solved in warm alcohol, and aliquots taken for counting and measurement of O.D. at 267 my. Thia orzaealisedon procedure was repeated three times. The product of the first crystallization was not analyzed. polypeptide chains can readily account for the inhibition of protein synthesis by this antibiotic. Whereas peptidyl-ssRNA found on the ribosome‘ is ordinarily transferred to the amino group of the next aminoacy]-sRNA, in the presence of puromycin the carboxyl-activated peptide appears to be transferred to puromycin, thus ending the sequential extension of the protein chain: N(CHs)o N\/\N HOCH, < o SNAN S wconkman’ — —C-NH-CH-C-NH-CH-C-sRNA + we es 2 Ri Re | | cH,o —< _ S—cH,-CH-c=0 N(CHs)p a: : "YON ——C-NH-CH-C-NH-CH-€ HOCH, i ) R, Re ° + sRNA NH Nd CH,0 —/ _ S—CH,-CH-C=0 Although this mechanism can account for the inhibition of protein synthesis by puromycin, it is not clear that this is the only way in which it functions. In the aminoacy]-sRNA-ribosomal system, puromycin leads to the formation of small peptides without puromycin at the end.* (Such peptides would not have been detected in the present study.) The appearance of free peptides suggests that puromycin may bring about hydrolytic cleavage of the peptidyl-sRNA bond in addition to the transfer mechanism already discussed. However, because the ob- servations on formation of free peptides were made under conditions where pep- tidases might secondarily split peptidyl-puromycin,? this alternate reaction is not established. 592 ENGINEERING: J. SLEPIAN Proc. N. A. 8. An implication of the mode of action of puromycin presented here is the likelihood that the transfer of peptide from peptidyl-sRNA to puromycin can serve as a model for the formation of peptide bonds in protein synthesis. Thus, one would expect that the same enzyme that catalyzes peptide bond formation also catalyzes the transfer of peptide to puromycin. In line with this is the observation that release of peptides from E. colt ribosomes by puromycin is stimulated by the enzyme frac- tion which is required for polypeptide synthesis from aminoacyl-sRNA.’? Summary.—Radioactive puromycin, when added to a culture of E. coli, becomes incorporated into acid-precipitable compounds by means of a reaction that is sensitive.to chloramphenicol. Digestion of this acid-insoluble product with trypsin and chymotrypsin liberates radioactive material with chromatographic, electro- phoretic, and chemical properties of puromycin. It is concluded that growing peptide chains on ribosomes are transferred to the free amino group of puromycin in analogy with the normal reaction of peptide bond formation in protein synthesis. The generous assistance of Dr. Amos Neidle in the synthesis of radioactive puromycin and of phenylalanyl-puromycin is most gratefully acknowledged, as is the continued interest of Dr. Fritz Lipmann, in whose laboratory the author’s work on puromycin was begun. This research has been supported by a grant from the National Institutes of Health. - ! 'Yarmolinsky, M. B., and G. L. de la Haba, these Proceepines, 45, 1721 (1959). ? Nathans, D., and F. Lipmann, these PRocEEpINas, 47, 497 (1961). * Nathans, D., J. E. Allende, T. W. Conway, G. J. Spyrides, and F. Lipmann, in Informational Macromolecules, ed. H. J. Vogel, V. Bryson, and J. O. Lampen (New York: Academic Press, 1963). ‘Gilbert, W., J. Mol. Biol., 6, 389 (1963). * Morris, A., R. Arlinghaus, S. Flavelukes, and R. Schweet, Biochemistry, 2, 1084 (1963). * Rabinovitz, M., and J. M. Fisher, J. Biol. Chem., 237, 477 (1962). 7 Allen, D. W., and P. C. Zamecnik, Biochim. Biophys. Acta, 55, 865 (1962). * Nathans, D., G. von Ehrenstein, R. Monro, and F. Lipmann, Fed. Proc., 21, 127 (1962). * Pitt-Rivers, R., and J. Lerman, J. Endocrinol., 5, 223 (1947-48). © Baker, B. R., J. P. Joseph, and J. H. Williams, J. Am. Chem. Soc., 77, 1 (1955). 11 Takeda, Y., S. Hayashi, H. Nakagawa, and F. Suzuki, J. Biochem., 48, 169 (1960). 12 Nathans, D., unpublished observations.