Proceedings of the National Academy of Sciences Vol. 68, No. 2, pp. 481-485, February 1971

Reinitiation of RNA Synthesis on Transcription of Chromatin DNA by Escherichia coli RNA Polymerase MARGARET E. MORRIS AND HANNAH GOULD M. R. C. Metabolic Reactions Research Unit, Department of Biochemistry, Imperial College, London, S.W.7 and Department of Biophysics, King's College, 26-29 Drury Lane, London, W.C.2, England.

Communicated by Paul Doty, December 3, 1970 The transcription of DNA in chromatin ABSTRACT by Escherichia coli RNA polymerase resembles that of isolated DNA in two important respects: the release of nascent RNA and reinitiation of RNA synthesis is dependent on the salt concentration, and RNA synthesis is markedly stimulated by the addition of ribosomes.

Recent experiments have established that the release of nascent RNA and reinitiation of RNA synthesis by Escherichia coli RNA polymerase with DNA templates occurs only in favorable ionic conditions (1-8). The ability of ribosomes to bind to nascent RNA (9-11) and stimulate the transcription of DNA (12-17) in vitro is also well documented. The possibility exists that these particular factors might also govern the transcription of chromatin templates, and we show here that this is indeed the case in an in vitro system.

represses the stimulation by KCL. The increase in RNA synthesis is thus a direct consequence of reinitiation. RNA synthesis at high, but not at low, ionic strength is further stimulated by the addition of ribosomes. At 0.25 M KCl, in the experiment shown in Fig. 6, we obtained more than a tenfold increase in RNA synthesis at the optimal ribosome concentration. Whereas the salt effect becomes increasingly more apparent as a function of incubation time, the ribosome effect is manifested early in the reaction and is relatively constant throughout the course of RNA synthesis (Fig. 7).

45

40 35

7

W* 30-

RESULTS

c

The effect of variation in the KCl concentration in the assay mixture on the incorporation of labeled UTP into RNA, directed by chicken kidney chromatin, is shown in Fig. 1. The maximum rate of RNA synthesis measured in a 20-min incubation was observed at 0.25 M KCl. The optimum divalent cation concentration at 0.25 M KCl is 5-10 mM as shown in Figure 2. The rates of UTP incorporation at 0 and 0.25 M KCl are similar in the first few minutes of the incubation. Thereafter, as may be seen in Fig. 3, RNA synthesis in the absence of salt ceases, while that in the presence of salt continues for about 90 min, although at a diminishing rate. The overall stimulation of RNA synthesis in the presence of salt is generally found to be 2-4 fold. The following experiment shows that the stimulatory effect of KCl is accompanied by the release of nascent chains from the enzyme and the template. The synthesized RNA was fractionated by filtration on Millipore filters. Released RNA would be expected to flow through the filter, whereas RNA bound to chromatin or RNA polymerase would not (6, 18, 19). The results illustrated in Fig. 4 show that the RNA synthesized in the absence of salt is all retained by a Millipore filter. The excess RNA produced by the presence of salt passes through the filter. In the absence of salt, RNA polymerase cannot reinitiate RNA synthesis. This was shown with the antibiotic rifampicin, which blocks initiation but allows the synthesis of initiated chains to go to completion (6, 20). Fig. 5 shows that rifampicin added 2 min after the start of incubation totally

25-

v 20 15

10, 5

-

0

01

0-2

0-3

KCI (M)

04

0-5

5 10 MgCI2+MnC[2 [mM]

(Left) FIG. 1. The effect of KCl concentration on RNA synthesis using Escherichia coli RNA polymerase and chicken kidney chromatin. Duplicate incubation mixtures contained either no chromatin or 60 ;g of chromatin DNA and 10 units of E. coli RNA polymerase, 0.9 mM each of ATP, CTP, GTP, and [3H]UTP (9.9 Ci/mol; Amersham) 4 mM MgCl2, 1 mM MnCl2, 5 mM putrescine dihydrochloride, 1 mg/mil of washed and fractionated bentonite (sediments in 15 min at 10,000 X g but not at 600 X g) (British Drug Houses, Ltd.), and various concentrations of KCl, as indicated, in a total volume of 0.25 ml. Samples were incubated for 20 min at 30'C. Values obtained in the absence of chromatin (lower curve, 0) were subtracted from those obtained with chromatin at the same KCl concentration to give the net synthesis of chromatin-primed RNA (upper curve, 0). (Right) FIG. 2. The effect of divalent cation concentration on RNA synthesis using E. coli RNA polymerase and chicken-kidney chromatin. MgCl2 and MnCl2 were used in the ratio 4:1 as originally recommended by Konrad and Stevens (25) and found suitable in chromatin-transcription studies (73). Other authors have found MnCl2 slightly inhibitory in the presence of MgCl2 (1, 2). Incubation was as described in Fig. 1, except that the KCl concentration, 0.25 M, was held constant. Values obtained without chromatin (lower curve, A) were subtracted to give the net synthesis of chromatin-primed RNA (upper curve, A).

481

482

Biochemistry: Morris and Gould

Proc. Nat. Acad. Sci. USA I

00(D

-

A-

3001 E E

200C

100t

Time (minutes) TIME I MINUTES }

0

2

3

RIBOSOME (mg/ml)

(Left) FIG. 3. The effect of KCL on the kinetics of RNA synthesis using E. coli RNA polymerase and rat liver chromatin. Duplicate samples contained no chromatin (open symbols) or 20 ,gg of chromatin (filled symbols), either no KCl (0) or 0.25 M KCl, (A) and other ingredients described in Fig. 1. Values obtained in the absence of chromatin, and for chromatin-primed net synthesis of RNA (see Fig. 2) are shown. (Center) FIG. 4. The effect of KCI on the kinetics of synthesis and release of nascent RNA. Both total incorporation (A) and nascent RNA retained by Millipore filters (0), synthesized in the presence (0.25 M) and the absenceof KC1, wereestimated; no-chromatin "blanks" shown (corresponding open symbols) were subtracted as described in Fig. 1 to give the net synthesis of chromatin-directed RNA. (Right) FIG 6. The effect of ribosome concentration on RNA synthesis using E. coli RNA polymerase and rat liver chromatin in the presence and absence of KC1. 80S rabbit reticulocyte monosomes were added in various amounts to the standard incubation mixture (Fig. 1) containing 2.8 units of E. coli RNA polymerase and 10 ,ug of rat liver chromatin and no salt (0) or 0.25 M KC1 (A). Samples were incubated for 95 min at 30°C and no-chromatin "blanks" were subtracted.

DISCUSSION

Earlier work employing chromatin templates has emphasized the importance of intrinsic factors, in particular the structural proteins (21, 22) or RNA (23), in regulating transcription by E. coli RNA polymerase. In both laboratories in which these experiments were performed chromatin was incubated in the absence of salts other than divalent cations and 2-20 times more enzyme relative to DNA than in the present study (2123). Our present work shows that at limiting enzyme concentrations, extrinsic conditions such as the concentration of salt or ribosomes in the incubation mixture may influence both the rate and the extent of RNA synthesis. The stimulation of chromatin-DNA transcription by salt has been reported (24). We have explored the basis for this effect by methods analogous to those used in the study of DNA transcription. When phage DNA serves as a template for E. coli RNA polymerase in vitro at low ionic strength, the rate of RNA synthesis is rapid at first but reaches a plateau value after a few minutes (25-30). This plateau corresponds to the synthesis of one chain of RNA per enzyme molecule (4, 6, 7). The cessation of RNA synthesis is accounted for by the inability of the enzyme to detach from the complex with nascent RNA and template prior to reinitiation. At the optimal KCl concentration, about 0.2 M, reinitiation of RNA synthesis occurs (1-8), and with the synthesis of new chains, a corresponding amount of completed RNA is released from the template (7). It has been estimated, depending on the template and the laboratory, that the enzyme functions 1.7 (T4 phage DNA; ref. 7), 2.3 (T6 phage DNA; ref. 7), 4 (T2DNA; ref. 5), or at least 20 (T4 phage DNA; ref. 2) times. RNA synthesis in the

of salt characteristically proceeds for a longer time (up to 5 hr) than in its absence (1-8). In our experiments with chromatin, a similar early plateau in the reaction kinetics at low ionic strength is observed (Figs. 3-5 and 7). At the optimal salt concentration, around 0.25 M KCl, the extent of incorporation is increased by a factor of 2-4, and the reaction proceeds for a longer time (up to 90 min). Nascent chains are released (Figs. 3 and 4) and a corresponding number of new chains are synthesized (Fig. 5). The effects of salt on the transcription of chromatin DNA and isolated DNA are thus qualitatively and quantitatively similar. It has been pointed out that potassium is the most prevalent monovalent cation in E. coli (31, 32), and that its intracellular concentration is about 0.3 M (31, 32). The enzyme is therefore most active in vitro in conditions that approach the intracellular environment. In E. coli, ribosomes are bound to nascent RNA both in vivo (35, 36) and in vitro (9-11) and stimulate DNA transcription in vitro (12-17). The degree of stimulation of DNA transcription by E. coli (12-15) or rabbit reticulocyte (16) ribosomes is approximately twofold at 0.05 M KCl. Whereas it was first reported that salt-washed ribosomes alone produced this effect (12), later work indicated the requirement for a protein factor that is removed from ribosomes during the washing procedure (13, 14). It has been claimed more recently that the protein fraction by itself produces a fiftyfold stimulation and that salt-washed ribosomes are inhibitory (17). All of the above cited experiments with ribosomes were performed with suboptimal KCl concentrations. At 0.25 M KCl, we observe a tenfold stimulation of RNA synthesis by ribosomes in the chromatin system (Fig. 6). The presence

Initiation of DNA Transcription

Vol. 68,1>971

optimal concentration is in the same range used in similar studies with DNA templates (12-14). In the presence of 0.15 mg of ribosomes, 3.5 ;g of RNA is synthesized. This quantity of ribosomes is not sufficient to form polysomes with all of the nascent RNA (37, 38). Protein synthesis, required for polysome formation, probably cannot in any case occur due to the lack of added amino acids and transfer factors. However, translation of nascent messenger RNA is not required for the stimulation of DNA transcription by ribosomes (14). Indeed, protein synthesis may be accompanied by ribosome-dependent messenger degradation (39). Presumably ribosomes in our system only bind to the initiation sites for protein synthesis on nascent RNA. We do not know whether the synthesis of a specific class of RNA molecules, i.e., messenger RNA, is preferentially stimulated by adding ribosomes. Predominantly nuclear RNA has been shown to be synthesized in vitro at low ionic strength in the absence of ribosomes (21-23). The hybridization tests used in these studies, however, might not be sufficiently sensitive to detect cytoplasmic messenger RNA (40). Definitive evidence for the synthesis of messenger RNA in vitro comes from the coupled DNA transcription-translation studies with systems comprising both optimal salt and ribosome concentrations (41-47). In the absence of salt, ribosomes have no effect on the level of RNA synthesis with chromatin (Fig. 6). This may be for a trivial reason; for example, it is possible that ribosome binding to messenger RNA is dependent on the concentration of monovalent cations. If protein synthesis is important for the stimulatory effect of ribosomes, this would certainly be the case (48). Alternatively, it is possible that salt is necessary but not

TIME

(MINUTES)

FIG. 5. The effect of KCl on the kinetics of synthesis, release, and reinitiation of synthesis of RNA. Rifampicin, 10 pg/ml, was added after 2 min of incubation in the normal reaction mixture, to the indicated samples to block further initiation of RNA synthesis. Total (A) and nonfilterable (0) RNA were estimated, and no-chromatin "blank" values (corresponding open symbols) were estimated as described (Fig. 1).

483

a2000

1000

0

60 TIME

120 0 (MINUTES)

60

120

FIG. 7. The effect of ribosomes on the kinetics of RNA synthesis in the presence and absence of KC1. Samples were incubated either with ribosomes (1.5 mg/ml) or without ribosomes, and either no KCl (0) or 0.25 M KC1 (A), for various times. Values obtained in the absence of chromatin (open symbols) are subtracted from those obtained in its presence (closed symbols) to give the net synthesis of chromatin-primed RNA.

sufficient to attain the maximum rate of RNA synthesis, and that ribosomes act upon a step which becomes rate limiting at high ionic strength. Several studies have shown that ribosomes effect the release of nascent RNA from DNA templates (1215). RNA which tends to "stick" to the template may retard further synthesis. The question of whether the effect of ribosomes on RNA synthesis in vitro has a counterpart in the regulation of RNA synthesis in vivo has often been debated. While in bacteria ribosomes translate the nascent message as fast as it is synthesized (49, 50), there is no decisive evidence that protein synthesis is required for messenger RNA synthesis. If the cellular concentration of ribosomes is depleted by a limitation of Mg++ in the culture medium, the rate of messenger RNA synthesis is proportionately reduced (51). However, if the same effect is achieved by limiting the amino acid supply, messenger RNA synthesis continues unabated (52). In animal cells, most protein synthesis occurs in the cytoplasm. Nevertheless, it is thought that ribosomal subunits (53-55), and possibly other proteins (56-60), may be involved in the transport of messenger RNA from the nucleus to the cytoplasm. The rate of messenger RNA synthesis might be controlled by the rate at which it is removed from the template. It has been argued from a considerable weight of circumstantial evidence that ribosom3 biosynthesis precedes, and is prerequisite for, messenger RNA synthesis during growth and differentiation in animal cells (61). It is perhaps necessary to justify our use of experimental systems containing such widely heterologous elements (E. coli RNA polymerase, rat liver, or chicken kidney chromatin, rabbit reticulocyte ribosomes). Others have shown that E. coli RNA polymerase makes tissue-specific RNA on chromatin templates (21-23) and amphibian ribosomal RNA when the corresponding DNA is used (62, 63). The hybridization techniques used to characterize the transcription products in these experiments cannot, unfortunately, distinguish between RNA fragments and correctly initiated and terminated RNA

484

Biochemistry: Morris and Gould

chains. Homologous animal polymerases may be required to recognize the appropriate start and stop signals for RNA synthesis on chromatin. For this reason, comparative studies with animal polymerases when these become more readily available will be of great interest. The stimulation of E. coli phage DNA transcription by rabbit reticulocyte ribosomes has been noted previously (16). However, the current notion of ribosome specificity in messenger RNA binding (64), and translation (65-67), suggests that a study in greater depth of this aspect may also be rewarding. METHODS Preparation of E. coli RNA polymerase E. coli RNA polymerase was prepared from E. coli MRE 600

by the method of Maryanka and Johnston (68) through the DEAE-cellulose step. The specific activity of the different enzyme preparations was 250-270 units/mg protein (1 unit = 1 nmol of ["4C ]ATP incorporated into RNA in 10 min at 370C using salmon-sperm DNA as template as described by Chamberlin and Berg (69)). RNA synthesis in the absence of added DNA was less than 1% of that in the presence of excess salmon-sperm DNA. Preparation of rat liver chromatin Nuclei were prepared according to the method of Mertlesmann and Matthaei (70). The nuclei were sedimented by centrifugation and resuspended in 10 volumes of distilled water by means of a vortex mixer. The lysed nuclei were centrifuged at 600 X g for 10 min. The pellet was resuspended in the same volume of water as before by means of a Potter homogenizer (about 25nm clearance). The suspension was again centrifuged at 600 X g for 10 min. The pellet was suspended with the Potter homogenizer in a small volume of water and diluted to 1 mg/ ml of DNA, estimated from the absorbance of a sample at 260 and 280 nm (71). Preparation of rabbit reticulocyte monosomes Total ribosomes from rabbit reticulocytes, prepared as described (72), were fractionated by centrifuging for 1 hr on a gradient of 15-30% sucrose containing 25 mM KQl-50 mM Tris buffer-1.5 mM MgCl2(pH 7.6) using an MSE BXIV zonal rotor at 45,000 rpm at 0°C. The 80S monosome fraction was precipitated by adding 1 M acetic acid to pH 5; after centrifugation at 600 X g, the ribosome pellet was redissolved in the same buffer plus 10% sucrose. The concentration of the stock solution, measured from the absorbancy of a sample at 260 and 280 nm, was 15 mg/ml (71).

Assay for RNA synthesis and release from chromatin After incubation at 30°C, RNA in the reaction mixtures was precipitated with 4 ml of ice-cold 0.5 M HCl04 containing

0.125 M tetrasodium pyrophosphate (PPi), thoroughly mixed, and incubated at 0°C for 30 min. The precipitates were filtered onto Whatmann GF/C glass fiber filters, and were washed successively with 15 ml of HCl04-PPi, 10 ml of ice-cold 0.5 M HCl04, and finally 20 ml of water. The filters were dried in an oven and placed in scintillation vials with 10 ml of toluenePBD scintillation fluid (30 g butyl-PBD from CIBA and 250 g naphthalene in 5 liters of toluene). The samples were counted for 10 min in a Beckman liquid scintillation spectrometer. To assay labeled RNA bound to chromatin, 4 ml of ice-cold 0.01 M Tris buffer, pH 7.8, was added to 0.1-ml samples, and

Proc. Nat. Acad. Sci. USA

the solution was filtered through Sartorious membrane filters, 0.45 /Am pore diameter. Filters were washed with 20 ml of the same buffer, dried, and counted as above. Released RNA is the difference between total and chromatin-bound RNA. We thank the Medical Research Council for a Studentship for M. E. M. and for financial support for this work, and Sir Ernst Chain for his interest. We are also indebted to Miss Jennifer Shearman and Dr. H. S. W. King for help in the preparation of ribosomes. 1. So, A. G., E. W. Davie, R. Epstein, and A. Tissieres, Proc. Nat. Acad. Sci. USA, 58, 1739 (1967). 2. Fuchs, E., R. L. Millette, W. Zillig, and G. Walter, Europ. J. Biochem., 3, 183 (1967). 3. Walter, G., W. Zillig, P. Palm, and E. Fuchs, Eur. J. Biochem., 3, 194 (1967). 4. Millette, R. L., Fed. Proc., 28, 659 (1969). 5. Maitra, U., and F. Barash, Proc. Nat. Acad. Sci. USA, 64, 779 (1969). 6. Richardson, J. P., Prog. Nucleic Acid Res., ed. Davidson and Cohn (Academic Press, N. Y., 1969), p. 75. 7. Richardson, J. P., Nature, 225, 1109 (1970). 8. Richardson, J. P., J. Mol. Biol., 49, 235 (1970). 9. Byrne, R., J. G. Levin, H. A. Bladen, and M. W. Nirenberg, Proc. Nat. Acad. Sci. USA, 52, 140 (1964). 10. Bladen, H. A., R. Byrne, J. G. Levin, and M. W. Nirenberg, J. Mol. Biol., 11, 78 (1965). 11. Palm, P., W. Doerfler, P. Traub, and W. Zillig, Biochim. Biophys. Acta, 91, 522 (1964). 12. Shin, D. H., and K. Moldave, J. Mol. Biol., 21, 231 (1966). 13. Revel, M., and F. Gros, Biochem. Biophys. Res. Commun., 27, 12 (1967). 14. Revel, M., M. Herzberg, A. Becarevic, and F. Gros, J. Mol. Biol., 33, 231 (1968). 15. Jones, 0. W., M. Dieckmann, and P. Berg, J. Mol. Biol., 31, 177 (1968). 16. Hunt, D., and H. G. Klemperer, Biochem. J., 108, 10p (1968). 17. Davison, J., L. M. Pilarski, and H. Echols, Proc. Nat. Acad. Sci. USA, 63, 168 (1969). 18. Nygaard, A. P., and B. D. Hall, Biochem. Biophys. Res. Commun., 12, 98 (1963). 19. Jones, 0. W., and P. Berg, J. Mol. Biol., 22, 199 (1966). 20. Wehrli, W., J. Ntesch, F. Knusel, and M. Staehelin, Biochim. Biophys. Acta, 157, 215 (1968). 21. Paul, J., and R. S. Gilmour, J. Mol. Biol., 34, 305 (1968). 22. Gilmour, R. S., and J. Paul, J. Mol. Biol., 40, 137 (1969). 23. Bekhor, I., G. M. Kung, and J. Bonner, J. Mol. Biol., 39, 351 (1969). 23 b. Huang, R. C. C., and P. C. Huang, J. Mol. Biol., 39, 365 (1969). 23 c. Smith, K. D., R. B. Church, and B. J. McCarthy, Biochemistry, 8, 4271 (1969). 24. DeBellis, R. H., W. Benjamin, and A. Gellhorn, Biochem. Biophys. Res. Commun., 36, 166 (1969). 25. Bremer, H., and M. W. Konrad, Proc. Nat. Acad. Sci. USA, 51, 801 (1964). 26. Ochoa, S., D. P. Burma, H. Kroger, and J. D. Weill, Proc. Nat. Acad. Sci. USA, 47, 670 (1961). 27. Hurwitz, J., J. J. Furth, M. Anders, and A. Evans, J. Biol. Chem., 237, 3752 (1962). 28. Stevens, A., and J. Henry, J. Biol. Chem., 239, 196 (1964). 29. Fox, C. F., and S. B. Weiss, J. Biol. Chem., 239, 175 (1964). 30. Krakow, J. S., Biochim. Biophys. Acta, 72, 566 (1963). 31. Solomon, A. K., Biophys. J. Suppl. 2, 79 (1962). 32. Lubin, M., and H. L. Ennis, Biochim. Bwphys. Acta, 80, 614 (1964). 33. Deleted in proof. 34. Deleted in proof. 35. Naono, S., J. Rouviere, and F. Gros, Biochim. Biophys. Acta, 129, 271 (1966).

Vol. 68, 1971 36. Das, H. K., A. Goldstein, and L. I. Lowney, J. Mol. Biol., 24, 231 (1967). 37. Warner, J. R., P. M. Knopf, and A. Rich, Proc. Nat. Acad. Sci. USA, 49, 122 (1963). 38. Staehelin, T., F. 0. Wettstein, H. Oura, and H. Noll, Nature, 201, 264 (1964). 39. Kuwano, M., D. Schlessinger, and D. Apirion, J. Mol. Biol., 51, 75 (1970). 40. Britten, R. J., and D. Koehne, Science, 161, 529 (1968). 41. Wood, W. B., and P. Berg, Proc. Nat. Acad. Sci. USA, 48, 94 (1962). 42. Gold, L. M., and M. Schweiger, Proc. Nat. Acad. Sci. USA, 62, 892 (1969). 43. Schweiger, M., and L. M. Gold, Proc. Nat. Acad. Sci. USA, 63, 1351 (1969). 43b. Gelfand, D. H., and M. Hayashi, Proc. Nat. Acad. Sci., USA, 63, 135 (1969). 44. Gold, L. M., and M. Schweiger, J. Biol. Chem., 245, 2255 (1970). 45. DeVries, J. K., and G. Zubay, Proc. Nat. Acad. Sci. USA, 57, 1010 (1967). 46. Lederman, M., and G. Zubay, Biochem. Biophys. Res. Commun., 32, 710 (1968). 47. Zubay, G., and M. Lederman, Proc. Nat. Acad. Sci. USA, 62, 550 (1969). 48. Nair, K. G., and H. R. V. Arnstein, Biochem. J., 97, 595 (1965). 49. Alpers, D. H., and G. M. Tomkins, J. Biol. Chem., 241, 4434 (1966). 50. Leive, L., and V. Kollin, J. Mol. Biol., 24, 247 (1967). 51. Naono, S., J. Rouviere, and F. Gros, Biochim. Biophys. Ada, 129, 271 (1966). 52. Morris, D. W., and N. 0. Kjeldgaard, J. Mol. Biol., 31, 145 (1968). 53. Girard, M., H. Latham, S. Penman, and J. E. Darnell, J. Mol. Biol., 11, 187 (1965).

Initiation of DNA Transcription

485

54. Henshaw, E. C., M. Revel, and H. H. Hiatt, J. Mol. Biol., 14, 241 (1965). 55. McConkey, E. H., and J. W. Hopkins, J. Mol. Biol., 14, 257 (1965). 56. Georgiev, -G. P., Prog. Nucleic Acid Res., 6, 259 (1967). See p. 15. 57. Samarina, 0. P., E. M. Lukanidin, J. Molnar, and G. P. Georgiev, J. Mol. Biol., 33, 251 (1968). 58. Perry, R. P., and D. E. Kelly, J. Mol. Biol., 35, 37 (1968). 59. KohIer, K., and S. Arends, Eur. J. Biochem., 5, 500 (1968). 60. Cartouzou, G., J. C. Pour6e, and S. Lissitzky, Eur. J. Biochem., 8, 357 (1969). 61. Tata, J. R., Biochem. J., 104, 1 (1967). 62. Reeder, R. H., and D. D. Brown, Fed. Proc., 28, 349 (1969). 63. Reeder, R. H., and D. D. Brown, J. Mol. Biol., 51, 361 (1970). 64. Naora, H., and K. Kodaira, Biochim. Biophys. Acta, 209, 196 (1970). 65. Weiss, S. B., and W. Hsu, Proc. Nat. Acad. Sci. USA, 64, 345 (1969). 66. Lodish, H. F., Nature, 224, 867 (1969). 66b. Lodish, H. F., Nature, 226, 705 (1970). 67. Argetsinger-Steitz, J., H. K. Dube, and P. S. Rudland, Nature, 226, 824 (1970). 68. Maryanka, D., and I. R. Johnston, FEBS Lett., 7, 125 (1970). 69. Chamberlin, M., and P. Berg, Proc. Nat. Acad. Sci. USA, 48, 81 (1962). 70. Mertlesmann, R., Eur. J. Biochem., 9, 311 (1969). 71. Warburg, O., and W. Christian, Biochem. Z., 310, 384 (1942). 72. Arnstein, H. R. V., R. A. Cox, H. J. Gould, and H. Potter, Biochem. J., 96, 500 (1965). 73. Marushige, K., and J. Bonner, J. Mol. Biol., 15, 160 (1966).