Reprinted from the Proceedings of the Nationa, ACADEMY OF SCIENCES Vol. 47, No. 10, pp. 1580-1588. October, 1961. CHARACTERISTICS AND STABILIZATION OF DNAASE-SENSITIVE PROTEIN SYNTHESIS IN E. COLI EXTRACTS By J. Hemvrich Marruarr* Anp MARSHALL W. NIRENBERG NATIONAL INSTITUTES OF HEALTH, BETHESDA, MARYLAND Communicated by Joseph E. Smadel, August 3, 1961 It has been assumed for many years that in protein synthesis the base sequence of DNA specifies the base sequence of RNA and that RNA in turn controls the amino acid sequence of protein. In accord with this notion, several groups recently have observed an inhibition of amino acid incorporation into protein by DNAase in cell-free extracts.'~* One object of the present investigation was to study this phenomenon further. A major difficulty in the study of cell-free protein synthesis in H. coli systems has been the necessity for preparing fresh enzyme extracts for each experiment. Techniques have not been available for stabilization and storage of enzyme ex- tracts comparable to the techniques available for mammalian systems.‘ In the present communication, an amino acid-incorporating system stable to storage for several months will be described. The characteristics of amino acid incorporation into protein by the stored extracts were investigated also. A part of these data has been presented in a preliminary report.? Methods and Materials.—E. coli W3100 cells, harvested in early log phase, were washed by centrifugation and disrupted by grinding with twice their wet weight of alumina A301 (Aluminum Corporation of America) for 5 min at 5°. All subsequent steps were performed at this tempera- ture. The enzymes were extracted with buffer containing 0.01 M Tris(hydroxymethy])amino- methane, pH 7.8; 0.01 M magnesium acetate; 0.06 I KCl; and 0.006 M mercaptoethanol (standard buffer) equivalent to two or three times the wet weight of cells. The extract was centrifuged three times at 30,000 x g for 20, 20, and 60 minutes, respectively. The pellets were discarded after each centrifugation. The final supernatant fluid (S-30) was centrifuged at 105,000 X g for 2 hr in the Spinco Model L ultracentrifuge to sediment the ribosomes. The supernatant solution (S-100) was aspirated, and the ribosomes were suspended in standard buffer by gentle homogenization in a Potter-Elvehjem homogenizer and were washed by centrifuging again at 105,000 & g for 2 hr. The supernatant fluid was decanted and discarded, and the ribosomes (W-Rib) were suspended in the original volume of standard buffer. Fractions 8-30, 8-100, and W-Rib were dialyzed against 60 volumes of standard buffer overnight at 5° and were stored in small aliquots at — 15° until needed. DNAase I, RNAase, and trypsin were crystalline preparations obtained from the Worthington Vow. 47, 1961 BIOCHEMISTRY: MATTHAEI AND NIRENBERG 1581 Biochemical Co. The sodium salt of polyadenylic acid (Batch BC-5815-2, Miles Chemical Co.) had a molecular weight of approximately 30,000. U-C1+L-valine (6.5 mC/mM) was obtained from Nuclear-Chicago Corp. Polyglucose carboxylic acid (molecular weight approximately 30,000) was a gift from Dr. Peter Mora. Puromycin was a gift obtained from Dr. Arthur Weiss- bach. Chloramphenicol was obtained from Parke Davis & Co., and highly polymerized salmon sperm DNA from California Foundation for Biochemical Research. Proteolytic enzymes as contaminants of the DNAase were assayed by the method of Anson‘ using denatured hemoglobin (Nutritional Biochemical Co.) as the substrate. RNAase contamina- tion of DNAase was assayed by incubation of DNAase with purified RNA, precipitation with cold trichloroacetic acid, and determination of the absorbancy at 260 my of the supernatant solu- tion obtained after centrifugation. The DNAase digest of DNA was prepared by incubating 5 mg/ml salmon sperm DNA with 10 pg/ml DNAase for 6 hr at 35° in 20 wmoles/m] phosphate buffer, pH 7.0, and 20 umoles/ml magnesium acetate. DNAase was destroyed by three deproteinizations according to the method of Sevag,7 and traces of solvents were removed by bubbling N, through the solution. Protein was determined by a micro modification of the method of Lowry. The synthetic amino acid solution used contained 1 wmole/m] of each of the following L-amino acids: glycine, alanine, serine, aspartic acid, asparagine, glutamic acid, glutamine, isoleucine, leucine, cysteine, cystine, histidine, tyrosine, tryptophan, proline, threonine, methionine, phenylalanine, arginine, and lysine. The complete reaction mixture is shown in the legend for Table 1. TABLE 1 CHARACTERISTICS OF C!4L-VaLiInE INCORPORATION INTO ProTEIN By CELL-FREE, L. colt ENnzyYME PREPARATIONS STORED FOR SEVERAL WEEKS AT — 15° Experiment Counts/min/mg no. Additions protein 1 Complete 679 — 105,000 X g supernatant solution 74 “ ~— Ribosomes 15 “ — ATP, PEP, PEP kinase 38 “ + 10 we RNAase 9 “ — Amino acid mixture 365 “ + 0.02 wmole Puromycin 38 7 + 0.30 »pmole Chloramphenicol 82 “ — 0.03 umole GTP, CTP, UTP 583 “ — 0.03 pmole CTP, UTP 652 “ Deproteinized at zero time 4 2 Complete 2078 a + 10 ug DNAase 1223 . Deproteinized at zero time 5 The reaction mixtures contained the following in ~xmole/ml: 100 Tris (hydroxymethyl)aminomethane, pH 7.8; 10 magnesium acetate; 50 KCl; 6.0 mereaptoethanol; 1.0 ATP; 5.0 phosphoenolpyruvate, K salt; 20ug¢ phos- phoenolpyruvate kinase, crystalline; 0.05 each of 20 L-amino acids minus valine; 0.03 each of GTP, CTP, and UTP; 0.015 C!41-valine (~70,000 counts); 1.0 and 0.7 mg 8-100 and W-Rib protein, respectively were present per reaction mixture in Experiment 1. 2.1 mg 8-30 protein were present in Experiment 2. Total volume was 1.0 ml. Samples were incubated at 35° for 60 min, were deproteinized with 10 per cent trichloroacetic acid, and the precipi- tates were washed and counted by the method of Siekevitz.13 Planchets were counted in a Nuclear-Chicago gas flow counter with a Micromil window and a counting efficiency of approximately 30 per cent. All assays were performed in duplicate. Resulis.—Stabilization of cell-free extracts: The effects of dialysis and freezing upon the ability of the $-30 fraction to incorporate C'+-L-valine into protein are presented in Figure 1. No enzymatic activity was lost after overnight dialysis. If mercaptoethanol was omitted from the dialyzing buffer, rapid loss of activity was observed. Reduced glutathione was not quite as effective as mercaptoethanol in preventing loss of activity. Dialyzed 8-30 fractions were divided into aliquots and were stored at —15°. The results of Figure 1 demonstrate that the enzyme, after storage for twenty- four hr at —15°, was as active as fresh 8-30. Frozen preparations lost less than 10 per cent activity per month. 1582 BIOCHEMISTRY: MATTHAEI AND NIRENBERG Proc. N. A.S. 2400 DIALYZED 4 a FRESH DIALYZED FROZEN T T T T T 2000 [ T T T T i 1 | 1600 0 Qo °° ° + FRESH 1200 DIALYZED + 1600 DIALYZED 800 FROZEN 4 '200 400 COUNTS/MINUTE /MG. PROTEIN a Qo ° COUNTS/MINUTE/MG. PROTEIN 400 7 MINUTES | Fig. 2.—The effects of dialysis and o | bow L 1 freezing upon C!*-L-valine incorporation ° 1s 30 45 60 7s into protein using washed. ribosomes (W- MINUTES Rib) and 105,000 x g supernatant solu- tions (S-100). @ Fresh W-Rib and 8- Fig. 1.—The effects of dialysis and freezing 100; A W-Rib and 8-100 dialyzed sepa- upon C!_L-valine incorporation into protein in rately for 12 hr; oO W-Rib and S-100 30,000 x g supernatant layer fractions (S-30). after dialysis and storage separately at @ Fresh 8-30; oO 8-30 after 12 hr of dialysis: —15° for 24 hr. The composition of the A 58-30 after dialysis and storage at —15° for 24 reaction mixtures is presented in Table 1. hr. The composition of the reaction mixtures is 0.9 and 1.0 mg W-Rib and 8-100 protein presented in Table 1. 2.1 mg 8-30 protein were respectively were present in each reaction added to each reaction mixture. mixture. When washed ribosomes (W-Rib) and 105,000 x g supernatant fluid (S-100) were stored separately or recombined at —15°, some activity was lost compared to the 8-30 fraction (Fig. 2). No loss in enzymatic activity of fractions S-100 or W-Rib was observed after overnight dialysis. Again, addition of mercapto- ethanol prevented rapid inactivation. Storage of the fractions separately at —15° resulted in a loss of approximately 25 per cent of the activity. The enzyme fractions lost less than 5 per cent of their activity per week, and fractions stored for several months were routinely used in these studies. Figures 1 and 2 also demonstrate that of the total incorporation obtained after incubation for one hour, approximately 50 per cent occurred within the first 15 min. The characteristics of C'*-L-valine incorporation into protein by S-100 and W-Rib fractions stored at —15° for several weeks are presented in Table 1. The rate of incorporation in the absence of either S-100 or W-Rib fractions was negligible. Experiments of this type strongly suggested that W-Rib fractions were not con- taminated with intact cells, and this check was routinely performed with each enzyme preparation. Effects of additions and deletions: Incorporation was dependent upon addition of ATP and an ATP-generating system, and incorporation was completely inhibited by the addition of RNAase. Omitting a mixture of 20 L-amino acids from the reaction mixture resulted in a 46 per cent decrease in incorporation of C\-valine into protein, suggesting de novo synthesis of protein. The dependence of C'4-L- valine incorporation upon addition of the amino acid mixture was observed only when well-dialyzed preparations were used. Addition of 0.02 umoles puromycin or 0.30 umoles chloramphenicol/ml to the reaction mixture depressed amino acid incorporation. Puromycin was a better inhibitor of C-valine incorporation Vou. 47, 1961 BIOCHEMISTRY: MATTHAEI AND NIRENBERG 1583 than chloramphenicol. Omission of guanosine-5’-triphosphate (GTP), cytidine- 5’-triphosphate (CTP), and uridine-5’-triphosphate (UTP) resulted in a slight inhibition of C'-valine incorporation. However, addition of GTP alone largely replaced the mixture of three triphosphates. Experiment 2, also in Table 1, demonstrates that addition of 10 yg of crystalline DNAase markedly inhibited Ci-valine incorporation. Further experiments dealing with this effect will be discussed later. That the incorporation of C1*-valine into protein depends upon the presence of the S-100 fraction is further documented in Figure 3. Little C'-valine was in- corporated into protein when 0.7 mg W-Rib protein alone was used. The incorpo- ration was proportional to the amount of 8-100 fraction added up to 2.5 mg S-100 protein. In Figure 4 are presented data demonstrating the dependence of C#4-valine PI 23800 - 4 2600 }- 4 2400 [- 4 2200 F 4 2000 - 4 w ~ 1800 4 2a < 5 1600 + 4 2 1400 + 4 < Ld 3 1200 > . 4 2 tooo 4 = a Boo 4 S Zz 600 4 Ss 400 + 4 200 4 oO « L L { 1 oO 10 20 3.0 4.0 1 MG.105,000XG SUPERNATANT 6 \ i Lt L FLUID PROTEIN 9 ! 2 3 4 MG. RIBOSOMAL PROTEIN Fic. 3.—The dependence of C1L- . valine incorporation into protein upon _ Fig. 4.—The dependence of incorpora- the amount of 105,000 X g supernatant tion of C1*-L-valine inte protein upon E. solution (8-100). The composition of colt ribosome concentration. The compo- the reaction mixtures is presented in sition of the reaction mixtures is presented Table 1. 0.7 mg ribosome protein in Table 1. 1.0 mg 105,000 x g super- (W-Rib) were added to each sample. natant solution protein (S-100) was added Reaction mixtures were incubated for to each sample. Reaction mixtures were 60 min at 35°. incubated for 60 min at 35°. incorporation upon W-Rib fractions. No incorporation by 1.0 mg 8-100 protein alone was observed. C1+valine incorporation was proportional to the amount of W-Rib added within the range of 0-1.0 mg ribosomal protein. The effect of pH upon C'-L-valine incorporation into protein is presented in Figure 5. A sharp pH optimum was observed with maximal incorporation at pH 7.8. DN Aase effect: The effect of DNAase upon C!-valine incorporation over a 90-min incubation period is presented in Figure 6. In the absence of DNAase, 1584 BIOCHEMISTRY: MATTHAEI AND NIRENBERG Proc. N. A. 8. COUNTS/MINUTE/MG. PROTEIN 1600 1400 1200 1000 800 400 200 L 4 Mo 7.0 7.4 7.8 8.2 86 3.0 pH Fig. 5.—The effect of pH upon C!*-L-valine incorporation into protein. The composi- tion of the reaction mixtures is presented in Table 1. 100 wmoles Tris(hydroxymethy!)- aminomethane were present in each reaction mixture. Samples were incubated for 10 min at 35°. COUNTS/MINUTE /MG.PROTEIN 2000 1500 500 ro ~ ~~ T 1 To TO ™. - DNAase Let. dod Fe Ld 20 30 +40 50 60 70 80 30 MINUTES Fic. 6.—The effect of DNAase upon C!*-L-valine incorporation into protein. © Minus DNAase, A + 10 ug DNAase. 2.1 mg 30,000 g supernatant solution protein (8-30) were present in each reaction mixture. The composition of the reaction mixtures is pre- sented in Table 1. Vou. 47, 1961 BIOCHEMISTRY: MATTHAEI AND NIRENBERG 1585 POT TTt 2200 100 4 2000 z S = 90 + 1800 < S _e—— COUNTS / MINUTE /MG. PROTEIN & 80 + 1600 3 ye z ° ao 7O 41400 © S 3 Ss a 6 4 ~ g 60 1200 = = Zz a c 4 = uw 80 1000 a z z S 3 i 40 Tf —~PERCENT INHIBITION = 800 |, a 2a z g Z 30 4600 m K 2 z WW & 20 4 400 ut a 10 4 200 Ll 1 | tf | | | | S6 3 1 23 45 6 7 8 9 10 #o DNAase Fig. 7.—The effect of increasing concentrations of DNAase upon C!-L-valine incorpora- tion inte protein. A Counts/min/mg protein, @ Percent inhibition of amino acid incor- poration. The composition of the reaction mixtures is presented in Table 1. 2.1 mg 30,000 X g supernatant solution protein (S-30) were present in each reaction mixture. Samples were incubated at 35° for 60 min. the incorporation was more rapid during the first 30 min of incubation than during the next 60 min. At the end of 90 min of incubation, however, incorporation had not stopped. Addition of 10 ug crystalline DNAase/ml of reaction mixture did not affect. the initial rate of incorporation, but incorporation ceased after 30 min. These results demonstrate that reaction mixtures must be incubated for more than 20 min to obtain reproducibly the DNAase effect. The sensitivity of the system to DNAase is presented in Figure 7. When 0.1 ug. DNAase were added, approximately 70 per cent of the maximum DNAase inhibition was obtained. An inhibition that was almost maximal was obtained with 1.0 yg DNAase/ml. Increasing the concentration of DNAase 10-fold beyond this did not appreciably increase the inhibitory effect of DNAase, thus negating the presence of a contaminant inhibitor in the crystalline DNAase preparation. Contamination of DNAase with traces of proteolytic enzymes and RNAase seemed likely, since commercial DNAase is prepared from pancreas. Therefore, different preparations of crystalline DNAase were tested for both proteolytic and RNAase activity. Their purity varied widely. Crystalline DNAase obtained from the Worthington Biochemical Company (Lot No. D692-95-7) was the purest prepara- tion tested and contained less than 0.3 per cent by weight of material which had a proteolytic enzyme activity and less than 0.001 per cent RNAase activity, cor- responding to less than 0.05 yg trypsin and 0.0001 ug RNAase per 10 yg DNAase. In Table 2, the effects of these concentrations of RNAase and trypsin, singly and combined, upon C-valine incorporation into protein are presented. The system 1586 BIOCHEMISTRY: MATTHAEI AND NIRENBERG Proc. N. A. 8. TABLE 2 Tue Errect or RNAasE AND TRYPSIN UPON C-L-VALINE INCORPORATION INTO PROTEIN Experiment Counts/min/mg no. Additions protein L Complete 374 “ + 0.0001 ug RN Aase 352 “ + 0.001 yg RNAase 206 “ +0.01 yg RNAase 69 “ + 0.1 pg RN Aase 18 “ + 1.0 ug RNAase 9 “ Deproteinized at zero time 8 2 Complete 374 “ + 10 ug DNAase 234 “ + 0.05 ug Trypsin 394 “ + 0.05 ng Trypsin + 0.0001 pg RNAase 402 “ Deproteinized at zero time 8 The composition of the reaction mixtures is presented in Table 1. 2.8 mg S-30 protein were added to each reac- tion mixture. Samples were incubated at 35° for 60 min. TABLE 3 Tue Errrecr or PoLyANIONS UPON THE DNAass-INHIBITED INCORPORATION OF C44-L-VALINE INTO PROTEIN AND THE Errect oF 4 DNAass Dicest or DNA upon INCORPORATION Experiment Counts/min/mg no Additions protein 1 Complete 2,040 “ + 10 we DNAase 825 “ + “ “ + 100 xg polyadenylic acid 685 “ “ “ + 100 ug polyglucose car- boxy] derivative 810 “ + 100 ug polyadenyliec acid 2,430 “ + 100 pg polyglucose carboxyl derivative 2,150 “ Deproteinized at zero time 7 2 Complete 1,842 “ + 100 wg DNAase digest of salmon sperm DNA _ 1,825 3 Complete + 10 wg DNAase 604 “ + 1.0 ml reaction mixture incubated with 10 ug DNAase for 60 minutes 661 The components of the reaction mixtures and the incubation conditions are presented in Table 1. 2.0 and 1.0 mg W-Rib and S-100 protein were present in Experiments 1 and 2. In Experiment 3, 2.1 mg 8-30 protein was present in complete systems. 2.1 mg 8-30 protein was also present in reaction mixture incubated with DNAase for 60 min. Samples were incubated at 35° for 20 min. was extremely sensitive to RNAase. As little as 0.001 wg RNAase/ml of final reaction mixture depressed amino acid incorporation. The addition of 0.05 gg of trypsin and 0.0001 hg RNAase (the amount of trypsin and RNAase in 10 pg of the DNAase preparation used) had no effect upon the incorporation. This DNAase preparation was used for all subsequent work with DNAase. It should be emphasized that crystalline DNAase obtained from commercial sources varies widely in its content of trypsin and RNAase and thus should be assayed before use. The data of Table 3 demonstrate that addition of polyanions such as polyadenylic acid and a polymer of glucose carboxylic acid did not reverse the inhibition ob- tained upon addition of DNAase. Higher concentrations of polyanions than those shown in Table 3 were inhibitory. Addition of a DNAase digest of highly poly- merized salmon sperm DNA had no effect upon amino acid incorporation. In Experiment 3, Table 3, an additional experiment of this sort is presented. A reaction mixture containing 2.1 mg S-30 protein was incubated with 10 ug DNAase for 60 min. During this period the DNA contained in the enzyme would have been largely digested. Also, after 60 min, amino acid incorporation into protein had completely ceased. This reaction mixture, containing the endogenous, di- Vou. 47, 1961 BIOCHEMISTRY: MATTHAEI AND NIREN BERG 1587 gested DNA products, was then added to a fresh reaction mixture to see whether products of DNAase digestion were inhibitory. The results show that the prod. ucts of DN Aase digestion were not inhibitory. Discusston.—A considerable amount of evidence has been obtained indicating that C'-valine incorporation into protein in this system was not due to the pres- ence of contaminating intact H. coli cells. The extracts were repeatedly centrifuged at 30,000 X g, and the pellet containing intact cells and debris was discarded. No intact cells or protoplast-like bodies were microscopically visible in the supernatant fluid. Neither ribosomes nor 105,000 < g supernatant solution alone incorporated appreciable quantities of C'-valine. Both fractions were needed. In addition, combined extracts were almost totally inactive if ATP and an ATP-generating system were omitted. The rate of amino acid incorporation procecded rapidly for approximately 30 min and then gradually decreased. The incorporation had many characteristics expected of de novo protein synthesis. It required ATP and an ATP-generating system, was stimulated by a mixture of other L-amino acids, and was strongly inhibited by low concentrations of puromycin, chloramphenicol, and RNAase. In addition, the incorporation could be partially inhibited by addition of DN Aase. The initial rate of incorporation was not inhibited by DNAase, in contrast to the extremely sensitive inhibition of the portion of the incorporation occurring after 20 min of incubation. As low as 0.1 ug DNAase per ml of reaction mixture greatly inhibited the incorporation occurring after 20 min of incubation. Various com- mercial preparations of crystalline DNAase were assayed for contamination with proteolytic enzyme activity and RNAase. Some were heavily contaminated. The maximum amount of trypsin and RNAase present as contaminants in the crystalline DNAase preparation used in this study had little effect upon C1+valine incorporation when added to the system. Furthermore, if a trace contaminant in the DNAase were responsible for inhibiting amino acid incorporation, a cor- respondingly greater inhibition of amino acid incorporation would be expected when high concentrations of DNAase were used. The data of Figure 7 demon- strate that almost maximal inhibition was obtained with 1 yg DNAase per ml of reaction mixture. Increasing the DNAase concentration 10-fold did not appreci- ably increase the inhibition. The data of Table 3 demonstrate that the products of DNAase digestion were not inhibitory in this system and that polyanions cannot non-specifically reverse a DNAase-inhibited meorporation system, in contrast to reports of such non-specific reversal in thymus nuclei.® Although the mechanism of synthesis of template or “messenger” RNA has remained an enigma, enzymes possibly involved in this process are being studied, 8-2 It 1s not possible to say whether intact DNA is necessary for amino acid incorpora- tion into protein in the later stages of incubation in this system. One possibility, however, which is consonant with all of the known facts is that the initial rate of amino acid incorporation is primarily due to the completion of partially finished peptides linked to RNA templates. If template RNA were used only once, amino acid incorporation would cease as soon as the peptide chains were finished. Inhibition by DNAase observed in this cell-free system may be due to the destruc- tion of DNA and its resultant mability to serve as templates for the synthesis of 1588 BIOCHEMISTRY: MATTHAEI AND NIRENBERG Proc. N. A. 8. template RNA. Other explanations, however, are fully plausible, and it is not possible at this state to rule out alternative interpretations. In the following paper, further experiments on amino acid incorporation using the system described here are presented. It will be shown that in addition to the usual requirements, the system is stimulated by template RNA. Summary.—Cell-free H. cola extracts have been obtained which actively in- corporate amino acids into protein. Methods were devised whereby these extracts could be dialyzed and stored for long periods of time at —15° without undue loss of activity. The characteristics of amino acid incorporation by such stored ex- tracts were strongly suggestive of de novo protein synthesis, for incorporation required both ribosomes and 105,000 X g supernatant fractions, ATP and an ATP- generating system, was stimulated by a mixture of other L-amino acids, and was markedly inhibited by puromycin, chloramphenicol, and RNAase. The initial rate of amino acid incorporation was not inhibited by DNAase; subsequent in- corporation was greatly inhibited. The possible relationship of the DNAase inhibition of amino acid incorporation into protein to the synthesis of messenger” RNA was briefly discussed. * Supported by a NATO Postdoctoral Research Fellowship. 1 Tissiéres, A., D. Schlessinger, and F. Gros, these ProcrEepines, 46, 1450 (1960). 2 Matthaei, J. H., and M. W. Nirenberg, Ped. Proc., 20, 391 (1961). 3 Kameyama, T., and G. D. Novelli, Biochem. Biophys. Res. Comm., 2, 393 (1960). 4 Kirsch, J. F., P. Siekevitz, and G. E. Palade, J. Biol. Chem., 235, 1419 (1960). 5 Mora, P. T., E. Merler, and P. Maury, J. Am. Chem. Soc., 81, 5449 (1959). 6 Anson, M. L., J. Gen. Physiol., 22, 79 (1938). 7 Sevag, M. G., D. B. Lackmann, and J. Smolens, J. Biol. Chem., 124, 425 (1938). 8 Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall, zbid., 193, 265 (1951). 9 Allfrey, V. G., and A. E. Mirsky, these Procrgpinas, 44, 981 (1958). 0 Hurwitz, J., A. Bresler, and R. Diringer, Biochem. Biophys. Res. Comm., 3, 15 (1960). 4} Stevens, A., ibid., 3, 92 (1960). 12 Weiss, S. B., and T. Nakamoto, J/. Biol. Chem., 236, PC18 (1961). 18 Siekevitz, P., ibid., 195, 549 (1952).