128
Biochimico @ Elsevier
BRA
et Biophyrica
Scientific
Acta,
395
Publishing
(1975)
Company,
128-135
Amsterdam
--. Printed
in The
Netherlands
98319
RIBOSOMAL
ALBERT
J.J.
Biochcmisch
(Received
AND NON-RIBOSOMAL
VAN
OOYEN,
Laboraforium.
December
13th,
HERMAN %erni/:elaan,
RNA SYNTHESIS
A. DE BOER, Groningerz
GEERT (The
AB and
IN VITRO
MAX
GRUBER
Netherlandsj
1974)
Summary The synthesis of total and ribosomal RNA using nucleoids of Escherichiu coli as template was measured; of the total RNA synthesized by endogenous RNA polymerase which only completes chains, and added RNA polymerase which initiates new chains, 50.~--70 and 3---570, respectively, was rRNA. Total RNA synthesis by added enzyme, however, was lo----20 times higher than endogenous RNA synthesis; thus rRNA was synthesized at the same rate by the endogenous and the added enzyme. We conclude that the percentage rRNA in vitro is no measure of the rate of rRNA synthesis. Furthermore, it follows that the added enzyme, like the endogenous one, is packed at the physical limit on the ribosomal cistrons. Consequently, initiation of ribosomal cistrons by added enzyme was at or near the maximal rate possible for this system in which the elongation rate is 10. -20% of that in vivo. When RNA synthesis was assayed at various ratios of RNA polymerase to phenol-extracted DNA, the amount of rRNA made per DNA, which is a measure of the frequency of transcription of ribosomal cistrons, varied. The ratio of rRNA synthesis relative to total RNA synthesis also varied, but in a different way, again leading to the conclusion that this ratio, as determined in vitro, does not reflect the efficiency of transcription of the ribosomal cistrons.
Introduction In vivo, the rate of ribosomal RNA synthesis in Escherichia coli is closely correlated with the growth rate of the cells. The rate of rRNA synthesis as a percentage of total RNA made ranges from 15 to 60% [ 11. In vitro, values between 0 and 34% for the percentage rRNA have been found, depending on the experimental conditions used. In purified systems containing only DNA and RNA polymerase the percentage was always lower than 15% [2-g]. The percentage rRNA found in vitro has generally been taken as a measure of the efficiency of rRNA synthesis. However, the percentage rRNA found is the
129
resultant of ribosomal and non-ribosomal starts. So initiation of ribosomal cistrons in vitro might still be high, but the percentage value low due to a relative abundance of non-ribosomal starts. We studied this problem by using nucleoids from Escherichia coli prepared by methods as developed by Pettijohn et al. [lo]. The RNA polymerase present on these nucleoids is distributed on ribosomal and non-ribosomal cistrons according to the in vivo situation and also has the same packing as in vivo. Furthermore this RNA polymerase does not reinitiate, presumably because of lack of u factor [lo]. If RNA polymerase is added it initiates chains. These features allow a comparison of the rates of rRNA and non-rRNA synthesis by the endogenous RNA polymerase, representing the in vivo situation to those of exogenous RNA polymerase, representing the in vitro situation. The rates of rRNA and non-rRNA synthesis were also compared using phenol-extracted DNA at different RNA polymerase to DNA ratios. Materials and Methods Bacterial strains. E. coli NP2, CP, x (leu’, his; arg; tlzreo-, B;, RCStr), NFI,y, D,” (met B-, Arg A; rel A, ; a gift from Dr N.P. Fiil) and Proteus vulgaris (a gift from Dr W.N. Konings) were used in this study. Chemicals. Unlabelled nucleotides and egg-white lysozyme were from Boehringer, Mannheim, G.F.R. Labelled nucleotides were from The Radiochemical Centre, Amersham, U.K. Proteinase K was from Merck, Darmstadt, G.F.R. DNA, used as template for RNA synthesis, was prepared from E. coli strain NP, by the method of Miura [ 111 . The purification step was carried out twice, and the treatment with the two ribonucleases was followed by incubation at 37°C for 2 h with 200 pg/ml proteinase K. Membrane-bound and membrane-free nucleoids were prepared from E. coli strain NF, 7 essentially according to published procedures [ 12--141 with the following modifications: The 4000 X g step was omitted. Membrane-bound nucleoid was obtained as follows: cells were exposed to egg-white lysozyme at 0.8 mg/ml for 45 s at O”C, and lysis was carried out in high salt for 25 min at 0°C. Membrane-free nucleoid: egg-white lysozyme treatment occurred at 0.08 mg/ml for 45 s at O”C, and lysis occurred in high salt for 20 min at 26°C. Sucrose gradients contained KC1 instead of NaCl. Gradient profiles of both lysates are shown in Fig. 1. In this experiment all fractions were also assayed for endogenous RNA polymerase. As can be seen RNA polymerase and DNA profiles coincide, and the membrane-bound nucleoid sediments about three times as fast as the membrane-free nucleoid. The yield of membrane-bound nucleoid (as judged by recovery of labelled DNA) usually was between 60 and loo%, and for membrane-free nucleoid between 20 and 40%. DNA, used in the hybridisation test, was isolated from P. vulgaris cells by the method of Miura [ll] . After denaturation by alkali [ 151, the DNA was put on a methylated albumin kieselguhr column. Preparation of the column was according to the method of Mandell and Hershey [16], except that the column consisted of only one layer of coated kieselguhr. The denatured DNA was fractionated by the stepwise elution technique (0.1 M NaCl steps) which
130
Emin
A
B
21 mln
103
Fracilon
top Fraction
Fig. 1. Sedimentation properties of nucleoids. Cells were grown in the presence of [ .‘Hl thymidine for more than two generations. and lysed at 0°C (A) and 26OC (B), yielding membrane-bound (A) and membrane-free (B) nucleoids. The lysates were layered on lo-30% sucrose gradients containing 10 mM Tris HCl, pH 8.1 (4°C); 1 M KCl; 1 mM EDTA: 0.1 mM dithiothreitol. Centrifugation (at 4>C) was for 8 (A) and 21 (B) min at 17 000 rev./min in a SW 50 rotor (Spinco). Fractions of 8 drops were collected. ) was determined by radioactivity measurement. Endogenous RNA polymerase DNA ( a(.--0) was measured by the addition of 50 ~1 of a fraction to 100 1.11 of a reaction mixture containing 40 mM Tris HCl, pH 7.9 125’C); 0.4 mM potassium phosphate; 10 mM MgClz: 0.3 mM each of ATP. CTP and UTP; 0.1 rnM dithiothreitol; 0.3 mM [a-“?PlGTP having a specific activity of 61.8 Cilmol (A) or 34.1 Ci/mol (B). Incubation
was for 15 min at 37’C.
results in separation of the complementary ribosomal strands [17]. Enrichment of DNA complementary for rRNA was about lo-fold. RNA polymeruse was purified by the method of Burgess and Travers [18] from E. coli strain CP, x through fraction 4. This was followed by chromatography on a DNA-cellulose column [19]. The enzyme obtained in this way was more than 90% pure as judged by polyacrylamide gel electrophoresis in the presence of dodecylsulphate [20] . Protein was measured by the method of Lowry et al. [Zl]. In vitro (“H] RNA synthesis was carried out as stated in the legends of the figures. At the times indicated aliquots were precipitated with trichloroacetic acid to measure total RNA synthesis. For the measurement of rRNA synthesis, one quarter volume of 1.5 M NaCl containing 0.15 M sodium citrate, pH 7, was added. This solution was extracted once with the phenol solution of Kirby [22] and, without further treatment, used in the hybridisation reaction. The hybridisation-competition test for rRNA was carried out essentially by the method of Haseltine (Fig. 6 of ref. 2). Hybridisation efficiences assessed by an .’’ P-labelled rRNA internal standard were between 60 and 80%, with the P. vulgar-is complementary rDNA in 5-lo-fold excess over the amount of labelled rRNA in the tube. Competition by 200-500-fold excess of unlabelled over labelled rRNA was more than 96%. Because of the heterologous DNA used, only a small fraction (l-3%) of the non-rRNA hybridized. This value is probably an overestimate since not all rRNA-incompetible radioactivity will be in RNA. The percentage rRNA made was calculated after correction of the radioactivity competed by excess unlabelled rRNA, for hybridisati~on efficiency,
131
competition efficiency, and the difference in base composition between and non-rRNA (assumed to contain 20 and 25% UMP, respectively). All hybridisation assays were done in triplicate.
rRNA
Results and Discussion Ribosomal RNA synthesis by the nucleoids In all our experiments the two types of nucleoid that can be prepared, the membrane free and membrane bound, behaved identically. They will therefore not be treated separately. With respect to RNA synthesis from the nucleoids a distinction must be made between RNA synthesized by RNA polymerase present in the nucleoids and RNA polymerase which is added. During incubation at 38°C in the presence of the four nucleoside triphosphates endogenous RNA polymerase only completes RNA chains initiated in vivo and does not reinitiate at all as is evident from experiments with rifampicin (Fig. 2). If purified RNA polymerase is added to the nucleoids it initiates RNA synthesis and total synthesis is stimulated manyfold (Fig. 2). These results are in agreement with those of others [10,13]. In four different experiments a stimulation of total RNA synthesis by added RNA polymerase of 10 to 20 times was found in the first 10 min of synthesis. While 50----70% of the RNA made by the endogenous enzyme is rRNA, this value for the added enzyme is 3-5s (Table I). The latter value was obtained by subtracting the value of rRNA endogenously synthesized from that of the total rRNA synthesized. Hence, the added RNA polymerase synthesizes about as much rRNA as the endogenous one, but at the same time non-rRNA at a much higher rate. As the RNA polymerase molecules present on the ribosomal cistrons are packed close to the physical limit [lo] , the same must hold for the added enzyme. From this it can be concluded that initiation of ribosomal chains is at
15
30
45
60 minutes
Fig. 2. Kinetics nucleoids was mM potassium mM [“HI UTP of 50 g1 were in the presence
of RNA synthesis by endogenous and added enzyme. 0.1 ml of sucrose gradient-purified added to 0.9 ml of a reaction mixture containing: 40 mM Tris . HCl. pH 7.9 (25°C); 0.4 phosphate; 10 mM M&12: 0.3 mM each of ATP, CTP and GTP; 0.1 mM dithiothreitol; 0.3 at a specific activity of 80 Ci/mol. The reaction temperature was 38°C. Duplicate samples taken for analysis. o----x ). no RNA polymerase was added. The same curve was obtained of 5 pg/ml of rifampicin. lL, 12.6 /.@ RNA polymerase was added at 0 min.
132
TABLE rRNA
1 SYNTHESIS
BY NUCLEOIDS
0.05 ml of sucrose gradient-purified nucleoids were added to 0.45 ml reaction mixture containing 40 mM Tris . HCl. PH 7.9 (25 ’ C): 0.4 mM potassium phosphate: 10 mM MgClz; 0.3 mM each of ATP. CTP and GTP; 0.1 mM dithiotlreitol, 0.05 mM t3Hl UTP (9 Cilmmol, a) or 0.1 mM [3Hl UTP (1 Ciimmol, b, c). Incubation was for 15 min at 38 a C (a) or the nucleoids were incubated for 10 min at 38 ’ C to measure endogenous RNA synthesis (b.c). At this moment purified RNA polymerase was added to a final concentration of 25 pg/ml and incubation continued for 30 min. In (b) endogenous rRNA synthesis has been subtracted. In (c) the endogenous RNA polymerase was completely inactivated by the heat treatment. Means and standard deviations of the mean are given. Template
RNA polymerase
Number of experiments
Percentage
(a) Untreated nucleoid (b) Untreated nucleoid (c) Nucleoid after
Endogenous Exogenous Exogenous
4 5 4
57 3.40 4.34
5 min at 60’
rRNA
?;4 + 0.20 ? 0.58
C
or near the maximum that can be obtained in vitro under the conditions of our experiments, and that the low percentage found is only due to the abundance of non-ribosomal starts compared to maximal growth in vivo. The abundance may be caused by a larger number of initiation sites or a higher initiation frequency on the same sites or both. Under our conditions only the elongation rate determines the frequency of initiation of RNA polymerase on the ribosomal cistrons. This rate is about 10 nucleotides per s on the nucleoids (see Addendum) which is 10.--20% of the rate in vivo [1,23,24] . Heating the nucleoids at 60°C which results in a strong viscosity increase and completely suppresses the activity of endogenous RNA polymerase did not significantly change the amount of rRNA made after the addition of RNA polymerase. Total RNA synthesis was decreased to a variable extent, averaging 20% (data not given) resulting in a slightly higher percentage rRNA made (Table I). The presence of transcribing RNA polymerase is no precondition for the maximal activity of added enzyme as can be seen in Fig. 3 when RNA
30
60
90 minutes
Fig. 3. Kinetics of RNA synthesis by RNA polymerase added at different times. E’or experimental condi): 8 min (‘\-A’ ): 17 min tions see Fig. 2. 12.6 /a RNA ~olwnerase was added at 0 min (‘l---i’ (O--0): and 26 min (m----A). Endogenous activity was subtracted.
133
polymerase was added after endogenous synthesis had ceased. The decrease in total activity after heating thus must be caused by some other effect. No rRNA synthesis from the nucleoids by added RNA polymerase was found by Pettijohn et al. [lo]. This is in agreement with our result since their detection limit is about 5% rRNA. Ribosomal RNA synthesis at different RNA polymerase to DNA ratios Fig. 4 shows the effect of the RNA polymerase to DNA ratio on total and ribosomal RNA synthesis. The experiments were carried out with DNA prepared by phenol extraction. The curve for the “absolute rate” of rRNA synthesis. shows a kind of saturation for the ribosomal cistrons at a RNA polymerase to DNA ratio where the non-ribosomal cistrons are not yet saturated. As a consequence the percentage rRNA is lowered at higher ratios. We have no satisfactory explanation for the decrease in rRNA synthesis at the highest ratios, nor for the relatively low transcription of ribosomal cistrons compared to non-ribosomal DNA at low ratios. Pettijohn [5] did not observe the latter phenomenon. He, however, used different reaction conditions. A detailed investigation of these conditions will be necessary to compare results obtained by different investigators in purified systems without added factors [ 2,3,5--81. It has been shown in the previous section, that the initiation frequency of the ribosomal cistrons can be at the maximum possible under our conditions,
10 rotlo
20
, , ,
30
LO
RNA polymerose:
50
60 CINA(w/w)
Fig. 4. rRNA synthesis at different RNA polvmerase to DNA ratios. K. coli DNA. prepared as described in Materials and Methods was used. Complete reaction mixtures contained: 40 mM Tris . HCI, pH 7.9 (25OC); 0.4 mM potassium phosphate; 10 mM MgClZ; 100 mM KCI; 0.3 mM each of ATP, CTP and GTP; 0.1 mM EDTA; 0.1 mM dithiothreitol; 0.1 mM c3H1 UTP 1 Ci/mmol; DNA in varying amounts. The reaction mixtures were preincubated at 38°C for 10 min RNA polymerase was added and incubation continued for 30 min. rRNA synthesis expressed in arbitrary units per weight of DNA (o.), and. expressed in percentage of total RNA synthesis ((l---iJ).
134
when the nucleoids are used as template. We suggest that this may also hold when phenol-extracted DNA is used as template at high enzyme to DNA ratios. These experiments confirm the conclusion from those with the nucleoids that the percentage rRNA found is no measure for the frequency of transcription of the ribosomal cistrons, but also contains the capacity of the system for non-ribosomal RNA formation. It can be predicted that in crude systems which are more likely to reflect the in vivo situation than purified systems, the number of non-ribosomal starts will be reduced resulting in an increased percentage rRNA made. The results of Murooka and Lazzarini [ 41 are in agreement with this prediction. The relatively weak stimulation by added RNA polymerase of ribosomal RNA synthesis, compared to non-ribosomal RNA synthesis, observed by these authors, may be due to saturation of rRNA cistrons. Since the number of total starts in vitro depends upon many variables the study of effecters of rRNA synthesis should preferably be done by comparing absolute rates of synthesis. Addendum We assume that the RNA polymerase molecules present on a ribosomal cistron in the nucleoid are equally spaced across the entire cistron, and after resumption of synthesis will move with equal rate to the terminator, and terminate correctly. If IZ is the number of enzyme molecules per cistron the amount of RNA synthesized per scripton in vitro will be e.qual to l/zn RNA molecules since there is no reinitiation of chains. If we call the length of the entire scripton 1, in a certain time in which all enzyme molecules have travelled a distance x (X < 1) along the scripton, each of the terminator-distal n(l -x) enzyme molecules will have synthesized x scripton length while each of the nx terminator-proximal molecules will synthesize a different length, averaging to Yti per enzyme molecule. Thus, when molecules move a distance X, n( 1 -- x)x + ‘hlX’ is synthesized, i.e. n(x -- %x’ ). Half of total in vitro synthesis will be reached when n(x _ ‘/UC’ ) = %n. x then is 1 -- 2-“, i.e. 0.29. The total length of a ribosomal scripton is about 6000 nucleotides, and we know from our experiments that 50% of the RNA is synthesized in 3 min. The elongation rate thus is 0.29. 6000/180, i.e. nearly 10 nucleotides per s. We have assumed that the length of the non-ribosomal scriptons is of the same size as that of the ribosoma1 ones. Because of the preponderance of rRNA, however, the calculated elongation rate would not be far different even if the average non-ribosomal transcripts were half or twice as long as the ribosomal ones. Acknowledgements We would like to thank Miss Aly technical assistance. The present investigations were Netherlands Foundation of Chemical the Netherlands Organization for the
van Goor for her excellent
and dedicated
carried out under the auspices of the Research (S.O.N.) with financial aid from Advancement of Pure Research (Z.W.O.).
135
References 1
Pato.
2
Haseltine,
M.L.
and
Von
W.A.
Meyenburg,
(1972)
K.
Nature
(1970)
235,
Cold
Spring
3
Murooka.
Y. and
Lazzarini.
R.A.
(1972)
Proc.
4
Murooka.
Y. and
Lazzarini.
R.A.
(1973)
J. Biol.
D.E.
New
Biol.
5
Pettijohn,
6
Birnbaum,
7
Travers,
(19’72)
L.S. A.
and
(1973)
Nature D.L.
8
Travers,
A.
Bailhe.
9
Travers.
A.
and
10
Pettijohn.
Nat.
Kaplan, and
Nat].
Qua&.
Biol.
35,
497-504
Acad.
Chem.
Sci.
248,
U.S.
69,
2336-2340
62484250
204-206
J. Mol.
Pedersen,
Biol.
S. (1973)
R. (1974)
Clarkson.
Symp.
75.
73-81
15-18
Cukier-Kahn,
D.E..
235,
S. (1973) 244,
Harbor
329-333
C.R..
FEBS
Kossman.
Nat.
New
Lett.
43,
C.R.
and
Biol.
243,
161-163
86-88 Stonington,
O.G.
(1970)
J.
Mol.
Biol.
52.
Part A,
pp.
281-300 11
Miura.
K.I.
(1967)
543-545. 12
Stonington,
13
Pettijohn,
O.G.
R.D.
14
Worcel.
A.
Ehrlich,
P. and
16
MandelI.
J.D.
17
Udvardy.
A.
18
Burgess.
R.R.
19
eds),
Alberts, Symp.
20
S.N.,
K., eds)
21
Lowry,
22
Kirby.
87-99,
23
Manor,
24
Dougan,
fiershey, Venetianer,
2, pp.
F.J..
pp.
and
and
82,
NatI. and
Sci.
eds).
Vol.
XII,
Stamato.
Anal.
80,
Procedures
and
68.
6-9
(1973)
Technical
DNA
synthesis
Publishing
Co,
in
vitro
Lancaster
4251-4255
Biochem.
(1972)
U.S.
T.D.
91-105
Sot.
Biochim.
Jenkins,
M.,
Osborn,
1. 66-77
Biophys.
Acta
in Nucleic
Gutmann.
Press,
E.D.
Farr,
New
D.A.
Methods
Press,
A.L.
and
in Enzymology
D. and
Glaser,
M. (1972)
Academic
N.J..
Methods
Goodman,
A.H.
Biol. Chem.
(1960)
3-27,
Rosebrough, (1968)
Acad.
K.,
308,
Acid
234-241
Research
(Cantoni,
G.L.
and
Davies,
and
Ferris.
F.L.
(1968)
Cold
Spring
Harbor
289-305
J.R.
XXVI,
Moldave.
Medical
P. (1973) A
L. and
851-863
33,
Academic
H..
A.D.
Travers,
Proc. O.G.
145-162,
J. Mol.
Pringle,
K.S.
pp.
J. Am.
Amadio.
Vol.
eds)
E. (1974)
Biol.
O.H..
R.B.,
P. (1958)
and
B.M.,
(1971)
Stonington.
Doty,
and Vol.
D.E.
R.M.,
(Grossman,
York
Burgi,
and
Quant
Weber,
pp.
and
New
Pettijohn,
Hecht.
and Inman.
15
in Enzymology
Press,
and
D.E.,
(Wells,
D.R..
Methods
Academic
New
Randall. (Grossman,
in Enzymology R.J. L.
(1951) and
J. Biol. Moldave.
York
Stent, (1974)
G.S.
(1969)
J. Mol.
J. Mol. Biol.
87,
(Hirs,
C.H.W.
and
Timasheff.
York
Biol. 775-795
39,
l-29
Chem. K..
eds).
193. Vol.
265-275 XII,
part
B,
Biochimico @ Elsevier
BRA
et Biophyrica
Scientific
Acta,
395
Publishing
(1975)
Company,
128-135
Amsterdam
--. Printed
in The
Netherlands
98319
RIBOSOMAL
ALBERT
J.J.
Biochcmisch
(Received
AND NON-RIBOSOMAL
VAN
OOYEN,
Laboraforium.
December
13th,
HERMAN %erni/:elaan,
RNA SYNTHESIS
A. DE BOER, Groningerz
GEERT (The
AB and
IN VITRO
MAX
GRUBER
Netherlandsj
1974)
Summary The synthesis of total and ribosomal RNA using nucleoids of Escherichiu coli as template was measured; of the total RNA synthesized by endogenous RNA polymerase which only completes chains, and added RNA polymerase which initiates new chains, 50.~--70 and 3---570, respectively, was rRNA. Total RNA synthesis by added enzyme, however, was lo----20 times higher than endogenous RNA synthesis; thus rRNA was synthesized at the same rate by the endogenous and the added enzyme. We conclude that the percentage rRNA in vitro is no measure of the rate of rRNA synthesis. Furthermore, it follows that the added enzyme, like the endogenous one, is packed at the physical limit on the ribosomal cistrons. Consequently, initiation of ribosomal cistrons by added enzyme was at or near the maximal rate possible for this system in which the elongation rate is 10. -20% of that in vivo. When RNA synthesis was assayed at various ratios of RNA polymerase to phenol-extracted DNA, the amount of rRNA made per DNA, which is a measure of the frequency of transcription of ribosomal cistrons, varied. The ratio of rRNA synthesis relative to total RNA synthesis also varied, but in a different way, again leading to the conclusion that this ratio, as determined in vitro, does not reflect the efficiency of transcription of the ribosomal cistrons.
Introduction In vivo, the rate of ribosomal RNA synthesis in Escherichia coli is closely correlated with the growth rate of the cells. The rate of rRNA synthesis as a percentage of total RNA made ranges from 15 to 60% [ 11. In vitro, values between 0 and 34% for the percentage rRNA have been found, depending on the experimental conditions used. In purified systems containing only DNA and RNA polymerase the percentage was always lower than 15% [2-g]. The percentage rRNA found in vitro has generally been taken as a measure of the efficiency of rRNA synthesis. However, the percentage rRNA found is the
129
resultant of ribosomal and non-ribosomal starts. So initiation of ribosomal cistrons in vitro might still be high, but the percentage value low due to a relative abundance of non-ribosomal starts. We studied this problem by using nucleoids from Escherichia coli prepared by methods as developed by Pettijohn et al. [lo]. The RNA polymerase present on these nucleoids is distributed on ribosomal and non-ribosomal cistrons according to the in vivo situation and also has the same packing as in vivo. Furthermore this RNA polymerase does not reinitiate, presumably because of lack of u factor [lo]. If RNA polymerase is added it initiates chains. These features allow a comparison of the rates of rRNA and non-rRNA synthesis by the endogenous RNA polymerase, representing the in vivo situation to those of exogenous RNA polymerase, representing the in vitro situation. The rates of rRNA and non-rRNA synthesis were also compared using phenol-extracted DNA at different RNA polymerase to DNA ratios. Materials and Methods Bacterial strains. E. coli NP2, CP, x (leu’, his; arg; tlzreo-, B;, RCStr), NFI,y, D,” (met B-, Arg A; rel A, ; a gift from Dr N.P. Fiil) and Proteus vulgaris (a gift from Dr W.N. Konings) were used in this study. Chemicals. Unlabelled nucleotides and egg-white lysozyme were from Boehringer, Mannheim, G.F.R. Labelled nucleotides were from The Radiochemical Centre, Amersham, U.K. Proteinase K was from Merck, Darmstadt, G.F.R. DNA, used as template for RNA synthesis, was prepared from E. coli strain NP, by the method of Miura [ 111 . The purification step was carried out twice, and the treatment with the two ribonucleases was followed by incubation at 37°C for 2 h with 200 pg/ml proteinase K. Membrane-bound and membrane-free nucleoids were prepared from E. coli strain NF, 7 essentially according to published procedures [ 12--141 with the following modifications: The 4000 X g step was omitted. Membrane-bound nucleoid was obtained as follows: cells were exposed to egg-white lysozyme at 0.8 mg/ml for 45 s at O”C, and lysis was carried out in high salt for 25 min at 0°C. Membrane-free nucleoid: egg-white lysozyme treatment occurred at 0.08 mg/ml for 45 s at O”C, and lysis occurred in high salt for 20 min at 26°C. Sucrose gradients contained KC1 instead of NaCl. Gradient profiles of both lysates are shown in Fig. 1. In this experiment all fractions were also assayed for endogenous RNA polymerase. As can be seen RNA polymerase and DNA profiles coincide, and the membrane-bound nucleoid sediments about three times as fast as the membrane-free nucleoid. The yield of membrane-bound nucleoid (as judged by recovery of labelled DNA) usually was between 60 and loo%, and for membrane-free nucleoid between 20 and 40%. DNA, used in the hybridisation test, was isolated from P. vulgaris cells by the method of Miura [ll] . After denaturation by alkali [ 151, the DNA was put on a methylated albumin kieselguhr column. Preparation of the column was according to the method of Mandell and Hershey [16], except that the column consisted of only one layer of coated kieselguhr. The denatured DNA was fractionated by the stepwise elution technique (0.1 M NaCl steps) which
130
Emin
A
B
21 mln
103
Fracilon
top Fraction
Fig. 1. Sedimentation properties of nucleoids. Cells were grown in the presence of [ .‘Hl thymidine for more than two generations. and lysed at 0°C (A) and 26OC (B), yielding membrane-bound (A) and membrane-free (B) nucleoids. The lysates were layered on lo-30% sucrose gradients containing 10 mM Tris HCl, pH 8.1 (4°C); 1 M KCl; 1 mM EDTA: 0.1 mM dithiothreitol. Centrifugation (at 4>C) was for 8 (A) and 21 (B) min at 17 000 rev./min in a SW 50 rotor (Spinco). Fractions of 8 drops were collected. ) was determined by radioactivity measurement. Endogenous RNA polymerase DNA ( a(.--0) was measured by the addition of 50 ~1 of a fraction to 100 1.11 of a reaction mixture containing 40 mM Tris HCl, pH 7.9 125’C); 0.4 mM potassium phosphate; 10 mM MgClz: 0.3 mM each of ATP. CTP and UTP; 0.1 rnM dithiothreitol; 0.3 mM [a-“?PlGTP having a specific activity of 61.8 Cilmol (A) or 34.1 Ci/mol (B). Incubation
was for 15 min at 37’C.
results in separation of the complementary ribosomal strands [17]. Enrichment of DNA complementary for rRNA was about lo-fold. RNA polymeruse was purified by the method of Burgess and Travers [18] from E. coli strain CP, x through fraction 4. This was followed by chromatography on a DNA-cellulose column [19]. The enzyme obtained in this way was more than 90% pure as judged by polyacrylamide gel electrophoresis in the presence of dodecylsulphate [20] . Protein was measured by the method of Lowry et al. [Zl]. In vitro (“H] RNA synthesis was carried out as stated in the legends of the figures. At the times indicated aliquots were precipitated with trichloroacetic acid to measure total RNA synthesis. For the measurement of rRNA synthesis, one quarter volume of 1.5 M NaCl containing 0.15 M sodium citrate, pH 7, was added. This solution was extracted once with the phenol solution of Kirby [22] and, without further treatment, used in the hybridisation reaction. The hybridisation-competition test for rRNA was carried out essentially by the method of Haseltine (Fig. 6 of ref. 2). Hybridisation efficiences assessed by an .’’ P-labelled rRNA internal standard were between 60 and 80%, with the P. vulgar-is complementary rDNA in 5-lo-fold excess over the amount of labelled rRNA in the tube. Competition by 200-500-fold excess of unlabelled over labelled rRNA was more than 96%. Because of the heterologous DNA used, only a small fraction (l-3%) of the non-rRNA hybridized. This value is probably an overestimate since not all rRNA-incompetible radioactivity will be in RNA. The percentage rRNA made was calculated after correction of the radioactivity competed by excess unlabelled rRNA, for hybridisati~on efficiency,
131
competition efficiency, and the difference in base composition between and non-rRNA (assumed to contain 20 and 25% UMP, respectively). All hybridisation assays were done in triplicate.
rRNA
Results and Discussion Ribosomal RNA synthesis by the nucleoids In all our experiments the two types of nucleoid that can be prepared, the membrane free and membrane bound, behaved identically. They will therefore not be treated separately. With respect to RNA synthesis from the nucleoids a distinction must be made between RNA synthesized by RNA polymerase present in the nucleoids and RNA polymerase which is added. During incubation at 38°C in the presence of the four nucleoside triphosphates endogenous RNA polymerase only completes RNA chains initiated in vivo and does not reinitiate at all as is evident from experiments with rifampicin (Fig. 2). If purified RNA polymerase is added to the nucleoids it initiates RNA synthesis and total synthesis is stimulated manyfold (Fig. 2). These results are in agreement with those of others [10,13]. In four different experiments a stimulation of total RNA synthesis by added RNA polymerase of 10 to 20 times was found in the first 10 min of synthesis. While 50----70% of the RNA made by the endogenous enzyme is rRNA, this value for the added enzyme is 3-5s (Table I). The latter value was obtained by subtracting the value of rRNA endogenously synthesized from that of the total rRNA synthesized. Hence, the added RNA polymerase synthesizes about as much rRNA as the endogenous one, but at the same time non-rRNA at a much higher rate. As the RNA polymerase molecules present on the ribosomal cistrons are packed close to the physical limit [lo] , the same must hold for the added enzyme. From this it can be concluded that initiation of ribosomal chains is at
15
30
45
60 minutes
Fig. 2. Kinetics nucleoids was mM potassium mM [“HI UTP of 50 g1 were in the presence
of RNA synthesis by endogenous and added enzyme. 0.1 ml of sucrose gradient-purified added to 0.9 ml of a reaction mixture containing: 40 mM Tris . HCl. pH 7.9 (25°C); 0.4 phosphate; 10 mM M&12: 0.3 mM each of ATP, CTP and GTP; 0.1 mM dithiothreitol; 0.3 at a specific activity of 80 Ci/mol. The reaction temperature was 38°C. Duplicate samples taken for analysis. o----x ). no RNA polymerase was added. The same curve was obtained of 5 pg/ml of rifampicin. lL, 12.6 /.@ RNA polymerase was added at 0 min.
132
TABLE rRNA
1 SYNTHESIS
BY NUCLEOIDS
0.05 ml of sucrose gradient-purified nucleoids were added to 0.45 ml reaction mixture containing 40 mM Tris . HCl. PH 7.9 (25 ’ C): 0.4 mM potassium phosphate: 10 mM MgClz; 0.3 mM each of ATP. CTP and GTP; 0.1 mM dithiotlreitol, 0.05 mM t3Hl UTP (9 Cilmmol, a) or 0.1 mM [3Hl UTP (1 Ciimmol, b, c). Incubation was for 15 min at 38 a C (a) or the nucleoids were incubated for 10 min at 38 ’ C to measure endogenous RNA synthesis (b.c). At this moment purified RNA polymerase was added to a final concentration of 25 pg/ml and incubation continued for 30 min. In (b) endogenous rRNA synthesis has been subtracted. In (c) the endogenous RNA polymerase was completely inactivated by the heat treatment. Means and standard deviations of the mean are given. Template
RNA polymerase
Number of experiments
Percentage
(a) Untreated nucleoid (b) Untreated nucleoid (c) Nucleoid after
Endogenous Exogenous Exogenous
4 5 4
57 3.40 4.34
5 min at 60’
rRNA
?;4 + 0.20 ? 0.58
C
or near the maximum that can be obtained in vitro under the conditions of our experiments, and that the low percentage found is only due to the abundance of non-ribosomal starts compared to maximal growth in vivo. The abundance may be caused by a larger number of initiation sites or a higher initiation frequency on the same sites or both. Under our conditions only the elongation rate determines the frequency of initiation of RNA polymerase on the ribosomal cistrons. This rate is about 10 nucleotides per s on the nucleoids (see Addendum) which is 10.--20% of the rate in vivo [1,23,24] . Heating the nucleoids at 60°C which results in a strong viscosity increase and completely suppresses the activity of endogenous RNA polymerase did not significantly change the amount of rRNA made after the addition of RNA polymerase. Total RNA synthesis was decreased to a variable extent, averaging 20% (data not given) resulting in a slightly higher percentage rRNA made (Table I). The presence of transcribing RNA polymerase is no precondition for the maximal activity of added enzyme as can be seen in Fig. 3 when RNA
30
60
90 minutes
Fig. 3. Kinetics of RNA synthesis by RNA polymerase added at different times. E’or experimental condi): 8 min (‘\-A’ ): 17 min tions see Fig. 2. 12.6 /a RNA ~olwnerase was added at 0 min (‘l---i’ (O--0): and 26 min (m----A). Endogenous activity was subtracted.
133
polymerase was added after endogenous synthesis had ceased. The decrease in total activity after heating thus must be caused by some other effect. No rRNA synthesis from the nucleoids by added RNA polymerase was found by Pettijohn et al. [lo]. This is in agreement with our result since their detection limit is about 5% rRNA. Ribosomal RNA synthesis at different RNA polymerase to DNA ratios Fig. 4 shows the effect of the RNA polymerase to DNA ratio on total and ribosomal RNA synthesis. The experiments were carried out with DNA prepared by phenol extraction. The curve for the “absolute rate” of rRNA synthesis. shows a kind of saturation for the ribosomal cistrons at a RNA polymerase to DNA ratio where the non-ribosomal cistrons are not yet saturated. As a consequence the percentage rRNA is lowered at higher ratios. We have no satisfactory explanation for the decrease in rRNA synthesis at the highest ratios, nor for the relatively low transcription of ribosomal cistrons compared to non-ribosomal DNA at low ratios. Pettijohn [5] did not observe the latter phenomenon. He, however, used different reaction conditions. A detailed investigation of these conditions will be necessary to compare results obtained by different investigators in purified systems without added factors [ 2,3,5--81. It has been shown in the previous section, that the initiation frequency of the ribosomal cistrons can be at the maximum possible under our conditions,
10 rotlo
20
, , ,
30
LO
RNA polymerose:
50
60 CINA(w/w)
Fig. 4. rRNA synthesis at different RNA polvmerase to DNA ratios. K. coli DNA. prepared as described in Materials and Methods was used. Complete reaction mixtures contained: 40 mM Tris . HCI, pH 7.9 (25OC); 0.4 mM potassium phosphate; 10 mM MgClZ; 100 mM KCI; 0.3 mM each of ATP, CTP and GTP; 0.1 mM EDTA; 0.1 mM dithiothreitol; 0.1 mM c3H1 UTP 1 Ci/mmol; DNA in varying amounts. The reaction mixtures were preincubated at 38°C for 10 min RNA polymerase was added and incubation continued for 30 min. rRNA synthesis expressed in arbitrary units per weight of DNA (o.), and. expressed in percentage of total RNA synthesis ((l---iJ).
134
when the nucleoids are used as template. We suggest that this may also hold when phenol-extracted DNA is used as template at high enzyme to DNA ratios. These experiments confirm the conclusion from those with the nucleoids that the percentage rRNA found is no measure for the frequency of transcription of the ribosomal cistrons, but also contains the capacity of the system for non-ribosomal RNA formation. It can be predicted that in crude systems which are more likely to reflect the in vivo situation than purified systems, the number of non-ribosomal starts will be reduced resulting in an increased percentage rRNA made. The results of Murooka and Lazzarini [ 41 are in agreement with this prediction. The relatively weak stimulation by added RNA polymerase of ribosomal RNA synthesis, compared to non-ribosomal RNA synthesis, observed by these authors, may be due to saturation of rRNA cistrons. Since the number of total starts in vitro depends upon many variables the study of effecters of rRNA synthesis should preferably be done by comparing absolute rates of synthesis. Addendum We assume that the RNA polymerase molecules present on a ribosomal cistron in the nucleoid are equally spaced across the entire cistron, and after resumption of synthesis will move with equal rate to the terminator, and terminate correctly. If IZ is the number of enzyme molecules per cistron the amount of RNA synthesized per scripton in vitro will be e.qual to l/zn RNA molecules since there is no reinitiation of chains. If we call the length of the entire scripton 1, in a certain time in which all enzyme molecules have travelled a distance x (X < 1) along the scripton, each of the terminator-distal n(l -x) enzyme molecules will have synthesized x scripton length while each of the nx terminator-proximal molecules will synthesize a different length, averaging to Yti per enzyme molecule. Thus, when molecules move a distance X, n( 1 -- x)x + ‘hlX’ is synthesized, i.e. n(x -- %x’ ). Half of total in vitro synthesis will be reached when n(x _ ‘/UC’ ) = %n. x then is 1 -- 2-“, i.e. 0.29. The total length of a ribosomal scripton is about 6000 nucleotides, and we know from our experiments that 50% of the RNA is synthesized in 3 min. The elongation rate thus is 0.29. 6000/180, i.e. nearly 10 nucleotides per s. We have assumed that the length of the non-ribosomal scriptons is of the same size as that of the ribosoma1 ones. Because of the preponderance of rRNA, however, the calculated elongation rate would not be far different even if the average non-ribosomal transcripts were half or twice as long as the ribosomal ones. Acknowledgements We would like to thank Miss Aly technical assistance. The present investigations were Netherlands Foundation of Chemical the Netherlands Organization for the
van Goor for her excellent
and dedicated
carried out under the auspices of the Research (S.O.N.) with financial aid from Advancement of Pure Research (Z.W.O.).
135
References 1
Pato.
2
Haseltine,
M.L.
and
Von
W.A.
Meyenburg,
(1972)
K.
Nature
(1970)
235,
Cold
Spring
3
Murooka.
Y. and
Lazzarini.
R.A.
(1972)
Proc.
4
Murooka.
Y. and
Lazzarini.
R.A.
(1973)
J. Biol.
D.E.
New
Biol.
5
Pettijohn,
6
Birnbaum,
7
Travers,
(19’72)
L.S. A.
and
(1973)
Nature D.L.
8
Travers,
A.
Bailhe.
9
Travers.
A.
and
10
Pettijohn.
Nat.
Kaplan, and
Nat].
Qua&.
Biol.
35,
497-504
Acad.
Chem.
Sci.
248,
U.S.
69,
2336-2340
62484250
204-206
J. Mol.
Pedersen,
Biol.
S. (1973)
R. (1974)
Clarkson.
Symp.
75.
73-81
15-18
Cukier-Kahn,
D.E..
235,
S. (1973) 244,
Harbor
329-333
C.R..
FEBS
Kossman.
Nat.
New
Lett.
43,
C.R.
and
Biol.
243,
161-163
86-88 Stonington,
O.G.
(1970)
J.
Mol.
Biol.
52.
Part A,
pp.
281-300 11
Miura.
K.I.
(1967)
543-545. 12
Stonington,
13
Pettijohn,
O.G.
R.D.
14
Worcel.
A.
Ehrlich,
P. and
16
MandelI.
J.D.
17
Udvardy.
A.
18
Burgess.
R.R.
19
eds),
Alberts, Symp.
20
S.N.,
K., eds)
21
Lowry,
22
Kirby.
87-99,
23
Manor,
24
Dougan,
fiershey, Venetianer,
2, pp.
F.J..
pp.
and
and
82,
NatI. and
Sci.
eds).
Vol.
XII,
Stamato.
Anal.
80,
Procedures
and
68.
6-9
(1973)
Technical
DNA
synthesis
Publishing
Co,
in
vitro
Lancaster
4251-4255
Biochem.
(1972)
U.S.
T.D.
91-105
Sot.
Biochim.
Jenkins,
M.,
Osborn,
1. 66-77
Biophys.
Acta
in Nucleic
Gutmann.
Press,
E.D.
Farr,
New
D.A.
Methods
Press,
A.L.
and
in Enzymology
D. and
Glaser,
M. (1972)
Academic
N.J..
Methods
Goodman,
A.H.
Biol. Chem.
(1960)
3-27,
Rosebrough, (1968)
Acad.
K.,
308,
Acid
234-241
Research
(Cantoni,
G.L.
and
Davies,
and
Ferris.
F.L.
(1968)
Cold
Spring
Harbor
289-305
J.R.
XXVI,
Moldave.
Medical
P. (1973) A
L. and
851-863
33,
Academic
H..
A.D.
Travers,
Proc. O.G.
145-162,
J. Mol.
Pringle,
K.S.
pp.
J. Am.
Amadio.
Vol.
eds)
E. (1974)
Biol.
O.H..
R.B.,
P. (1958)
and
B.M.,
(1971)
Stonington.
Doty,
and Vol.
D.E.
R.M.,
(Grossman,
York
Burgi,
and
Quant
Weber,
pp.
and
New
Pettijohn,
Hecht.
and Inman.
15
in Enzymology
Press,
and
D.E.,
(Wells,
D.R..
Methods
Academic
New
Randall. (Grossman,
in Enzymology R.J. L.
(1951) and
J. Biol. Moldave.
York
Stent, (1974)
G.S.
(1969)
J. Mol.
J. Mol. Biol.
87,
(Hirs,
C.H.W.
and
Timasheff.
York
Biol. 775-795
39,
l-29
Chem. K..
eds).
193. Vol.
265-275 XII,
part
B,
