JOURNAL OF BACTERIOLOGY, May 1975, p. 425-432 Copyright 0 1975 American Society for Microbiology

Vol. 122, No. 2 Printed in U.S.A.

Stability of Ribosomal and Transfer Ribonucleic Acid in Escherichia coli B/r After Treatment with Ethylenedinitrilotetraacetic Acid and Rifampin DOROTHY YUAN' AND VICTOR SHEN* Institute for Molecular Biology, University of Texas at Dallas, Richardson, Texas 75080 Received for publication 17 December 1974

A short treatment with ethylenedinitrilotetraacetic acid to permeabilize bacteria for various antibiotics or treatment with the ribonucleic acid (RNA) synthesis inhibitor rifampin causes a slow degradation of 50S and 30S ribosomal particles and of the corresponding 23S and 16S ribosomal RNA species (about 25% in 1 h). The effects are additive such that the decay is about 50%o/h if rifampin is employed after permeabilization by ethylenedinitrilotetraacetic acid. The 5S ribosomal RNA and transfer RNA are essentially stable under these conditions.

The antibiotic rifampin inhibits initiation of ribonucleic acid (RNA) chains in Escherichia coli without affecting the completion of nascent RNA chains (21). By combining rifampin action and radioactive labeling methods, the synthesis rate of unstable messenger RNA (mRNA) (2, 7, 19) and the chain elongation rate of ribosomal RNA (rRNA) (3, 6) have been estimated. Those estimates, which have implications for the control of rRNA synthesis (2), depend upon a complex evaluation of measured data and could be invalid if rRNA synthesized (completed) in the presence of rifampin is unstable. For this reason we have studied the stability of rRNA and transfer RNA (tRNA) in the presence of rifampin and after an ethylenedinitrilotetraacetic acid (EDTA) treatment such as was previously used to permeabilize bacteria for the antibiotic. The results indicate that rifampin as well as EDTA treatment induces decay of rRNA. The extent of this decay does not invalidate the previous measurements of the rRNA chain elongation rate, but it may have produced an overestimate of the rate of mRNA synthesis. MATERIALS AND METHODS Bacterial growth and radioactive labeling. E. coli B/r (ATCC 12407) were grown exponentially in minimal medium C (10) supplemented with 0.2% glucose and 20 jg of each standard L-amino acid per ml. At an absorbancy at 460 nm of 0.2 (1-cm light path), the culture was divided into two portions which were then labeled with [14C uracil (0.67 MM; 150 mCi/mmol) or [3HJuracil (0.14 MM; 27 Ci/mmol), respectively. Thirty to 40 min later (absorbancy at 460 nm = 0.5) the cells were collected by centrifugation

and suspended in a 1:10 volume of either (14Clabeled culture) EDTA-PO4 buffer (6, 15) to make them permeable to rifampin or (3H-labeled culture) in PO4 buffer without EDTA. After 2 min at 37 C both cultures were diluted with 9:10 volume of prewarmed medium, and 2 min later (zero time) rifampin (Mann Research Labs) was added to the "4C-labeled culture (final concentration 60 ig/ml). In the experiment with rifampin alone, no centrifugation and resuspension procedures were used in exponential and treated cultures. Samples (1 ml) from both cultures were mixed, concentrated (20:1) by centrifugation, and lysed by addition to an equal volume of lysis mixture (2) at 100 C. After 30 s the clear lysates were cooled to and kept at 25 C. Permeabilization by EDTA. The effectiveness of the EDTA treatment depends upon the ionic strength and the pH of the phosphate buffer. The concentration of EDTA is unimportant, but the absence of Mg2+, not the presence of EDTA, appears to be important since washing of bacteria with Mg2+-free buffer has the same but slower effect. Also the duration of the EDTA-treatment (between 1 and 4 min) has no significant effect. With decreasing ionic strength or increasing pH the bacteria become increasingly more permeable, but also increasingly more damaged. The damage is evident as spontaneous cell lysis and as reduced accumulation of RNA (RNA was measured as the absorbancy at 260 nm in alkali-soluble material after acid precipitation). To determine which conditions rendered the cells sufficiently permeable but had minimal deleterious effects on the bacteria, the pH was kept at 6.8 (pH of medium) and the concentration of the buffer was varied. For our bacteria and growth medium, a concentration of 0.1 M P04 buffer was found to be a compromise. At this concentration the time until all RNA chain initiation stops in the presence of 60 Mg rifampin per ml is 3 to 5 s (3), and the rate of accumulation of stable RNA (measured as percent increase per unit of time) is 10 to 15% below the

'Present address: Department of Microbiology, University of Texas Health Science Center, Dallas, Tex. 75235. 425

426

YUAN AND SHEN

steady-state rate during the first 10 to 15 min after termination of the EDTA treatment (by dilution with pre-warmed medium). Later the rate temporarily decreases further, presumably as a result of ribosome breakdown (see Table 2), until the culture recovers. With this treatment full permeabilization lasts for about 10 min. Electrophoresis. Lysate (40 ul) from the mixed sample was electrophoresed (3 h at 12.5 V/cm on a 12-cm 2.3 to 8.0% polyacrylamide gradient gel in TEB buffer [20] at 4 C). The RNA in 1-mm slices of the gel was hydrolyzed in 0.5 ml of 0.2 N NaOH, and the radioactivity was counted after addition of 5 ml of a Biosolv 3-toluene mixture (Beckman Instruments) in a liquid scintillation spectrophotometer. Sedimentation analysis. The bacteria in a portion of the mixed sample were suspended in 3.0 ml of

Tris(hydroxymethyl)aminomethane (Tris)-magnesium buffer at 0 C (0.01 M Tris, 0.01 M MgCl, to

pH 8.0) and disrupted by decompression in a French press. After addition of 60Mg of pancreatic deoxyribonuclease per ml, the debris was removed by centrifugation, and the resulting extract was dialyzed against a solution of 0.01 M Tris and 0.1 mM MgCl,, pH 8.0. The dialyzed extract was analyzed by zone sedimentation (6 to 30% sucrose gradient) in the same buffer. The acid-insoluble materials of each 1-ml fraction were collected on glass-fiber filters, and radioactivity on the filters was counted in 5 ml of toluene-based scintillation fluid. Determination of RNA. Samples (1 ml) of the labeled bacteria were added to 2 ml of 1 M trichloroacetic acid at 0 C and collected on glass-fiber filters, and the acid-insoluble RNA was hydrolyzed with 1 ml of 0.2 N NaOH at 25 C overnight. A 1-ml volume of 0.5 M perchloric acid was added to each hydrolysate, and alkali-resistant materials (including deoxyribonucleic acid) were removed by membrane filtration. The radioactivity in 0.5 ml of filtrate was counted in 5 ml of a Biosolv 3-toluene mixture in a liquid scintillation spectrophotometer.

RESULTS Electrophoretic separation of stable RNA species. Four species (23, 16, 5, and 4S) of stable RNA were separated by electrophoresis as shown in Fig. 1 (exponentially growing cells represented by 3H label). During cell lysis, between 5 and 30% (average 15%) of 23S rRNA was cleaved; the cleavage products contaminated the 16S fraction (Table 1, ratio 23S/16S). In the experiments described below, experimental (treated) cells were always lysed together with control (untreated) cells (both labeled with different isotopes). Hence, the isotope ratio in the 23S fraction was independent of this cleavage. The isotope ratios were also independent of differences in the shape of the peaks (e.g., Fig. lc) in different gels. The distributions were not influenced by differential yields or losses of RNA species during sample preparation, since the electrophoresed samples were obtained by a method (2) that completely

J. BACTERIOL.

solubilizes bacteria, such that removal of debris, phenol extraction, or precipitation was not necessary. Labeled deoxyribonucleic acid is excluded by the gels. Stability of rRNA during exponential growth. In the experiments described below, the relative stability of "stable" RNA species in rifampin- or EDTA-treated bacterial cultures was estimated by comparison with the (unknown) stability of these species in exponentially growing cultures. The absolute stability of RNA during exponential growth is not exactly known since it cannot be measured easily by current methods. For example, RNA turnover does not result in the immediate disappearance of labeled RNA during a "chase" (addition of excess unlabeled RNA precursors to the growth medium), since the labeled breakdown products are effectively recycled (re-incorporated). Although recycling can be prevented by inhibiting RNA synthesis, this could not be done for exponential cultures since inhibition of RNA synthesis induces breakdown of ribosomes (see below). Hence, only the differential stability of RNA species could be determined: during a chase, radioactive nucleotides decreased in the less stable and accumulated in the more stable RNA species. Under our conditions of growth, the ratio of labeled tRNA to labeled rRNA did not significantly change during a chase lasting for two generations (Table 1; compare ratio of 4S/total at 30 and 90 min after labeling). This indicates that both species have either equal or negligible tumQver rates. Since tRNA was stable even under conditions where extensive breakdown of rRNA was observed (see below), it is presumed that the turnover of rRNA and tRNA in exponential cultures is negligible. This is consistent with the previously observed constancy during a chase of the ratio of labeled RNA to labeled DNA (5). Effect of rifampin after permeabilization with EDTA. The stability of the different stable RNA species ("4C labeled during exponential growth) in the presence of rifampin after EDTA treatment was compared with the stability of 3H-labeled RNA during exponential growth (Fig. 1). For this comparison, bacteria of the treated and untreated (control) cultures were lysed together, and the isotope ratios were then determined in the different RNA species after electrophoretic separation. Immediately after EDTA treatment and the addition of rifampin the isotope ratio (14C/3H) was the same for all species of RNA (Fig. la), as expected. An extract from another portion of the mixed samples (treated plus untreated), dialyzed against 10-1 M Mg2+ buffer to dissoci-

Slice number b b

2.5

6

16 S

23S

70 50 30

20

2.0 _-

5-..

I+ I

N

0

I0

X

4 X

a. 2 z I 0 0

10

.5 _-

JI0 6 PO

0

.o _-

10

0

0-

3

-~~~

z

20

2

Froct. no.

0.51_

5S

4S

I 0

-

c)

i -

51 v

o

a

10

20

30

40

60

50

80

70

90

Slice number 401

a.

)

-C

5S

4SN

SS

O

S

00

MIN

30(

a-

C-) 4

4t

z r_,204

z

: 04:

5

I)

II 104

0

___

TV

c 60

70

80

60 90 SLICE NUMBER

70

80

90

FIG. 1. Degradation of ribosomal particles and rRNA in the presence of rifampin administered after permeabilization by a 2-min EDTA treatment. (a) Electrophoretic analysis of RNA 1 min after rifampin addition 3H label, untreated control culture; 14C label, rifampin-treated culture. Insert: Sedimentation analysis. (b) Same analysis as shown in (a) for sample taken at 60 min. (c) 4 to 5S regions from (a) and (b); enlarged ordinate scale. For experimental procedures see Materials and Methods. 427

428

YUAN AND SHEN

J. BACTERIOL.

TABLE 1. Electrophoretic separation of stable RNA species labeled during exponential growtha Ratio

Radioactivity in RNA (3H counts/min)

Time (min) after labeling

23Sb

16Sb

5Sc

4Sc

Total

23S/16Sd

4S/totale

1

30 90

5,680 5,075

3,100 3,294

270 270

1,380 1,315

10,430 9,954

1.82 1.54

0.132 0.132

2

30 90

4,875 5,366

3,473 3,755

292 385

1,455 1,495

10,095 11,001

1.40 1.42

0.144 0.136

3

30 90

1,458 1,903

1,126 1,393

120 128

460 595

3,164 4,119

1.30 1.36

0.145 0.144

Expt

aSee legend of Fig. 1 for procedures. b Due to some breakdown of 23S RNA during bacterial lysis, the 23S RNA is underestimated and the 16S rRNA is overestimated (see footnote d below). c 5S and 4S RNA species were not completely separated (Fig. lc). The expected ratio 5S/(23S + 16S) is 0.026. Accordingly, in experiment 1 (30 min), 228 counts/min were expected to be in 5S RNA (observed, 270 counts/min), suggesting that the 5S RNA is somewhat contaminated with tRNA, which causes an overestimate of 5S RNA but only a small underestimate (3 to 5%) of tRNA. d Assuming the expected ratio 23S/16S to be 2.0, a ratio of 1.7 corresponds to 5% breakdown of 23S RNA during lysis [(2.0 - 0.1)/(1.0 + 0.1) = 1.71. eThe average observed ratio from the six distributions shown is 0.139. The average from 12 further distributions (not shown here) was 0.145. Thus, 14% of stable RNA is tRNA and 86% is rRNA. ' Distributions of experiment 2 are those illustrated in Fig. 1 (3H label).

ate ribosomal particles and analysed by zone sedimentation, showed the same isotope ratio for both ribosomal particles and tRNA (Fig. la, insert). Sixty minutes later, during which the 3Hlabeled control bacteria were allowed to grow while the "4C-labeled bacteria were kept in the presence of rifampin, the isotope ratio in rRNA decreased by more than 50% (Fig. lb), indicating significant rRNA degradation in the rifampin-treated culture. '4C label, which evidently represents rRNA degradation products, was found at positions between 23S and 5S RNA, suggesting that the rRNA had been degraded into fragments many of which were still acid precipitable. The degradation of 23S and 16S rRNA during rifampin treatment was paralleled by a disintegration of the whole ribosomal particle as seen from the sedimentation pattern (Fig. lb, insert). Sedimentation analyses (Fig. la and b inserts) with the label in protein ( [3H] and ["4C ileucine) rather than in RNA showed a similar decrease in the isotope ratio of 50S and 30S particles (not shown). The isotope ratio in 5S rRNA and tRNA (Fig. lc) was much less reduced, indicating that these species are more stable. Comparison of results from several similar experiments (Table 2) shows that the isotope ratio in tRNA decreased, on the average, by 5% within 60 min no matter how the cells were treated. This value may not be experimentally significant. (The

greater decrease in the results illustrated in Fig. 1 is an exception [see footnote g to Table 2].) If tRNA is stable, the breakdown of rRNA in the treated culture (relative to the untreated exponential culture) can be given by the quotient of the isotope ratio 23S/4S (Table 2) at 60 min, as well as by the quotient 23S (60 min)/23S (1 min). The first quotient requires electrophoresis of only a single sample (60 min), whereas the second quotient requires electrophoresis of two samples (1 and 60 min). With one exception the agreement between these two quotients is within 10% (Table 2). Although the rate of rRNA breakdown varied from experiment to experiment (Table 2; Fig. 2t, closed symbols), it is clear that rRNA breaks down at a significant rate while tRNA is essentially stable. In addition to being degraded into large fragments (as seen in Fig. lb), rRNA from mature ribosomal particles was slowly degraded to acid solubility during rifampin treatment (Fig. 3). Again there was considerable experimental variability. Stability of rRNA in newly assembled ribosomes. To determine whether newly assembled, and thus perhaps incomplete, ribosomes are less stable, RNA was labeled immediately before and during min 1 after rifampin treatment. The results of such experiments give no indication of a different stability of new and old ribosomes (Fig. 2), suggesting that after addition of rifampin enough ribosomal protein is made to form

429

STABILITY OF rRNA AND tRNA IN E. COLI

VOL. 122, 1975

TABLE 2. Stability of 23S rRNA and tRNA during 60 min of treatment with rifampin with and without permeabilization by EDTA and after EDTA treatment onlya 4S (peak fraction)

23S (peak fraction)

Time Treatment

(mim) after

Radioactivity

Ratio'

Radioactivity Ratiob (counts/min)

3H

"C

4"C/'H

3H

0.24 0.085

(counts/mmn)

treatment

Quotient of isotope ratios (60 min/lmin)

|"C

4"C/'H

23S

241 317

66 66

0.27e 0.21

03'

4S

23S/4S

07'0.89e 040.36t 0078 040'

rifampin

lC 60d

1,852 2,183

443 185

EDTA + rifampin

60

1

6,651

6,133

3,198 1,422

0.48 0.23

1,126 1,087

569 522

0.50e 0.48

0.48

0.96

0.96e 0.48

Rifampin

1 60

6,875

5,595

2,910 1,979

0.42 0.35

996 1,103

464 611

0.46e 0.55

0.83

1.19

0.91e 0.64

Rifampin

601

6,443 6,032

2,435 1,612

0.38 0.27

2,130 1,775

905 736

0.43e 0.41

0.71

0.95

0.88e 0.66

EDTA

601

6,651 6,505

3,198 2,653

0.48 0.41

1,126 991

569 467

0.50e 0.47

0.85

0.94

0.96e 0.87

1,585 60 11,135 ~~~1 60|1

252 140

0.16 0.12

EDTA

+

EDTA EDTA

0.75 0.75

a Experimental procedures as described for Fig. 1, except that in the rifampin (only) and EDTA (only) experiments, EDTA treatment and rifampin, respectively, were omitted. 14C label represents treated bacteria; 3H label represents exponentially growing bacteria. b The isotope ratio method is most accurate if it is applied to a purified fraction of the RNA. Here the peak fraction is assumed to be the purest fraction. Determination of the isotope ratio from the integrated peaks would introduce greater errors due to greater background contribution and inclusion of 14C breakdown products in the trail of the peak as shown in Fig. lb. c Distribution illustrated in Fig. la. dDistribution illustrated in Fig. lb. e In the 1-min distributions the ("4C/3H) ratio in tRNA (4S) is about 10% higher than the ratio in rRNA (23S), i.e., the quotient of the ratios (23S/4S) is about equal to 0.9 rather than 1.0. This is presumed to reflect substitution of 5-3H label in some uracil residues of tRNA (e.g., pseudo-uracil formation). 'The italicized values of the quotients 23S (60 min)/23S (1 min) and 23S (60 min)/4S (60 min) are both a measure of the stability of rRNA; they would be equal if tRNA were completely stable. The differences may be experimental errors. 'In this experiment the isotope ratio in tRNA decreased by 22% during 60 min, whereas in the other experiments the decrease was only 4 to 6%. This greater decrease may be due to volume errors (when 3H- and "4C-labeled cultures were mixed) or partial lysis of "4C-labeled cells caused by the EDTA treatment.

complete ribosomal particles from the rRNA finished at the beginning of the treatment (assuming absence of ribosomal protein causes more rapid degradation of rRNA as in chloramphenicol-treated bacteria [13]). Effect of rifampin concentration. With 1 ,ug of rifampin per ml (after permeabilization), RNA synthesis was 90% inhibited; above 5 ,ug/ml no RNA synthesis was detectable (Fig. 4a). In all other experiments reported here, 60 ;g of rifampin per ml was used. Above 10 ,qg/ml the rRNA breakdown was independent of the rifampin concentration (Fig. 4b), making it unlikely that the experimental variability in the stability of rRNA (Fig. 2, 3) is due to a variation in residual RNA synthesis (leakiness) in the presence of rifampin. The difference between

the zero points of the control (no rifampin) and the rifampin curve (Fig. 4b) is due to the EDTA treatment which was not used in the control (see below). Effect of rifampin without EDTA and of EDTA treatment alone. When bacteria were treated with 60 gg of rifampin per ml without prior EDTA treatment, RNA synthesis decreased to near zero within 5 to 10 min as judged by the decreasing rate of incorporation of radioactive uracil (2-min pulse labels: about 50% decrease every 2 min). (In contrast, with EDTA treatment and 60 jsg of rifampin per ml initiation of RNA chains ceases within 5 s [4]. This time increases linearly with the reciprocal of the rifampin concentration [11].) Under these conditions (rifampin alone, without EDTA treat-

430

YUAN AND SHEN

0

00

J. BACTERIOL.

0.4

0.6

FIG. 2Stbiyo2SrAad23SrRNA Ei 0.4

OoZ C

N

cL IU.

0.2

6,

0 41

0.01 0

w 0. cn. 0

1

tO

20

30

40

50

60

0

U) Time after rifampicin (min) RIFAMPICIN CONCENTRATION(pg/ml) FIG. 2. Stability of 23S rRNA and tRNA during inhibition of RNA synthesis. Elect rophoretic analysis FIG. 4. Effect of rifampin concentration and as for Fig. Ib; the isotope ratio ("4C/3H) from the EDTA treatment. (a) Permeabilized bacteria were 1-min sample is set at 1.0. Symbols: 0 and A, RNA

labeled 40 min before rifampin addition, as in Fig. 1; 0, label added 2 min prior to rifampin (about 1/'3 of the label is incorporated after addition of rifampin, reflecting the completion of nascent rRNA). Lines connect the average values from six experiments. 4

0

.02

' 0' 0 -o 0 40 50 20 50

Time after rifampicin (min)

FIG. 3. Degradation of rRNA during rifampin treatment, measured by acid precipitability. Bacteria were grown, pre-labeled (14C), and rifampin treated (time zero), and RNA was then measured.

ment) rRNA was degraded at a rate of about 25% in 60 min (28% = average from four experiments; two are shown in Table 2), i.e., half as fast as with EDTA treatment. EDTA treatment alone also produced a degradation rate of about 25% rRNA per h (23% = average

treated with various concentrations of rifampin as described in the legend of Fig. 1. They were labeled for 2.5 min with [3H]uracil either 10 min (0) or 60 min (0) after addition of rifampin. The higher values at 60 min presumably reflect the disappearance of the EDTA (permeabilization) effect, which can be overcome by using more than 10 ug of rifampin per ml. Radioactive RNA was determined as described in the legend of Fig. 3. Incorporation in untreated bacteria was set at 100%. (b) Experiment as described in the legend of Fig. 1 ('IC-labeled culture), except that various concentrations of rifampin were used. For zero rifampin concentration two points are shown, one with EDTA treatment (0) and one without EDTA treatment (0). The isotope ratio ("4C/3H) in this latter sample was set at 100%. Electrophoretic analysis (60 min after rifampin) was as described in the legend of Fig. lb.

from four experiments; two are shown in Table 2). Hence, the effects of EDTA and rifampin are roughly additive.

DISCUSSION Breakdown of ribosomes. Blundell and Wild (1) observed that ribosomes in rifampintreated E. coli MRE600 are altered. During 3 h of rifampin treatment, the polysomes decayed and dissociated to particles sedimenting more slowly than the normal 50S and 30S subunits. They concluded that the ribosomal particles had lost some protein, but that their rRNA,

VOL. 122, 1975

STABILITY OF rRNA AND tRNA IN E. COLI

even after 3 h in the presence of rifampin, was intact. This result, which is in contrast to our fimdings, may be due to the absence of ribonuclease I (RNase I) in strain MRE600 (9). Upon in vitro incubation of the altered ribosomal particles with commercial RNase (presumably pancreatic RNase I), Blundell and Wild (1) observed a rapid degradation of the rRNA under conditions that left the rRNA in particles from untreated cells intact. This means that during rifampin treatment the ribosomes are altered such that they become more susceptible to RNase. Since the rRNA in the RNase IMRE600 strain remains stable but is degraded in the RNase I+ B/r strain, we suspect that RNase I is responsible for the degradation of rRNA observed in strain B/r, but it may also result from other strain differences. Degradation of rRNA has also been observed when RNA synthesis was inhibited by actinomycin (18) or during starvation for uracil (13). It is unlikely, therefore, that the degradation observed here is rifampin specific. We suspect that the degradation is related to the inhibition of RNA synthesis or, in particular, to the absence of mRNA and, thus, to the continuous dissociation of ribosomal particles. In addition, it was observed here that a short EDTA treatment, as is widely used to increase the permeability of the bacterial membrane for antibiotics (15), has also a long-lasting effect on the stability of ribosomes. The effect may result from the short magnesium starvation only, and may be related to the depletion of ribosomes during long magnesium starvation (16), perhaps by releasing membrane-bound RNase I into the cytoplasm. Alternatively, it may be due to the continued presence of low concentrations of EDTA (10-4 M) in the growth medium during further incubation, which might affect some trace elements that may be required for ribosome stability. Since we were not primarily interested in the mechanism of RNA degradation we did not attempt to answer this question. Implications for the estimate of rRNA chain growth and mRNA synthesis rate. In the previous experiments to measure the rRNA chain elongation (see Introduction), the radioactivity in ribosomal particles was analyzed 15 min after the addition of rifampin to the bacteria. The results of that analysis were not affected by ribosome degradation if both 50S and 30S particles were degraded at the same rate (see Discussion in reference 6). Figure lb (insert) indicates that this is essentially the case. This and the small extent of the breakdown within 15 min (about 15%, Fig. 2) suggest that the previous estimate of rRNA chain growth need not be corrected.

431

Previous estimates of the mRNA synthesis rate (2, 7, 19), however, may be affected significantly by ribosome degradation. A 15% loss of acid precipitability of stable RNA after 15 min, for example, means that a previously observed ratio of stable to unstable RNA synthesis of 1:1 would be 1.15:0.85 without this breakdown, i.e., the mRNA synthesis rate would be 0.85/1.15 = 0.74, or 74% of the stable RNA synthesis rate when corrected for stable RNA breakdown, rather than being equal to stable RNA synthesis when uncorrected. Thus, the mRNA synthesis rates may have been overestimated previously. The exact magnitude of this error is difficult to assess from the results of Fig. 3 because of the experimental variability. Comparison of values for the relative rate of mRNA synthesis obtained with the rifampin method (2, 7, 19) with the values obtained by other methods that do not involve EDTA and rifampin treatment (12, 17) gives no indication for an overestimation by the rifampin method (7; Fig. 3 of reference 8); If acid precipitability of rRNA does not decrease linearly with time, then the loss of precipitable stable RNA observed here may be insignificant during the 15 min of rifampin treatment. ACKNOWLEDGMENTS This work was supported by Public Health Service grant GM 15132 from the National Institute of General Medical Sciences and by The University of Texas at Dallas Research Fund. We thank H. Bremer for helpful discussions and preparation of this manuscript. LITERATURE CITED 1. Blundell, M. R., and D. G. MTild. 1970. Altered ribosomes

2.

3. 4.

5. 6.

7.

8. 9.

after inhibition of Escherichia coli by rifampicin. Biochem. J. 121:391-398. Bremer, H., L. Berry, and P. Dennis. 1973. Regulation of ribonucleic acid synthesis in Escherichia coli B/r: an analysis of a shift-up. II. Fraction of RNA polymerase engaged in the synthesis of stable RNA at different steady-state growth rates. J. Mol. Biol. 75:161-179. Bremer, H., J. Hymes, and P. Dennis. 1974. Ribosomal RNA chain growth rate and RNA labeling patterns in Escherichia coli B/r. J. Theor. Biol. 45:379-404. Bremer, H., and D. Yuan. 1968. Chain growth rate of messenger RNA in Escherichia coli infected with bacteriophage T4. J. Mol. Biol. 34:527-540. Dennis, P. 1972. Regulation of ribosomal and transfer RNA synthesis in Escherichia coli B/r. J. Biol. Chem. 247:2842-2845. Dennis, P., and H. Bremer. 1973. Regulation of ribonucleic acid synthesis in Escherichia coli B/r: an analysis of shift-up. I. Ribosomal RNA chain growth rate. J. Mol. Biol. 75:145-159. Dennis, P., and H. Bremer. 1973. A method of determination of the synthesis rate of stable and unstable ribonucleic acid in Escherichia coli. Anal. Biochem. 56:489-501. Dennis, P. P., and H. Bremer. 1974. Macromolecular composition during steady-state growth of Escherichia coli. J. Bacteriol. 119:270-281. Gesteland, R. F. 1966. Isolation and characterization of ribonuclease I mutants of E. coli. J. Mol. Biol.

432

YUAN AND SHEN

16:67-84. 10. Helmstetter, C. E. 1967. Rate of DNA synthesis during the division cycle of Escherichia coli B/r. J. Mol. Biol. 24:417-427. 11. Hirsh, J., and R. Schleif. 1973. In vivo experiments on the mechanism of action of L-arabinose C gene activator and lactose repressor. J. Mol. Biol. 80:433-444. 12. Kennel, D. 1968. Titration of the gene sites on DNA by DNA-RNA hybridization. II. The Escherichia coli chromosome. J. Mol. Biol. 34:85-104. 13. Lazzarini, R. A., K. Nakata, and R. M. Winslow. 1969. Coordinate control of ribonucleic acid synthesis during uracil deprivation. J. Biol. Chem. 244:3092-3100. 14. Lazzarini, R., and E. Santagelo. 1968. Effect of chloramphenicol on the synthesis and stability of ribonucleic acid in Bacillus subtilis. J. Bacteriol. 95:1212-1220. 15. Leive, L., 1965. Actinomycin sensitivity in Escherichia coli produced by EDTA. Biochem. Biophy. Res. Commun. 18:13-17.

J. BACTERIOL.

16. McCarthy, B. J. 1962. The effects of magnesium starva-

17.

18. 19.

20. 21.

tion on the ribosome content of Escherichia coli. Biochim. Biophys. Acta 55:880-888. Nierlich, D. P. 1972. Regulation of ribonucleic acid synthesis in growing bacterial cells. II. Control over the composition of the newly made RNA. J. Mol. Biol. 72:765-777. Ohnishi, Y., and D. Schlessinger. 1972. Total breakdown of ribosomal and transfer RNA in a mutant of E. coli. Nature (London) New Biol. 238:228-231. Pato, M., and K. von Meyenburg. 1970. Residual RNA synthesis in Escherichia coli after inhibition of initiation of transcription by rifampicin. Cold Spring Harbor Symp. Quant. Biol. 35:497-504. Peacock, A. C., and C. W. Dingman. 1967. Resolution of multiple ribonucleic acid species by polyacrylamide gel electrophoresis. Biochemistry 6:1818-1827. Wehrli, W., and M. Staehelin. 1971. Actions of the rifamycins. Bacteriol. Rev. 35:290-309.

r after treatment with ethylenedinitrilotetraacetic acid and rifampicin.

JOURNAL OF BACTERIOLOGY, May 1975, p. 425-432 Copyright 0 1975 American Society for Microbiology Vol. 122, No. 2 Printed in U.S.A. Stability of Ribo...
1MB Sizes 0 Downloads 0 Views