Tim JOURNAL OF BIOLOQICAL CAEMISTRY

Vol.246, No. 14,Issue of July 25, pp. 4381-4385, 1971

Printed

in

U.S.A.

On the Regulation
of Guanosine
Tetraphosphate
Levels in
Stringent and Relaxed Strains of Escherichia
coli
(Received for publication,
ROBERT

A. LAZZARINI

AND

MICHAEL

January 25, 1971)

CASHEL

From the Laboratory of Molecular Biology, National Institute of Neurological Diseasesancl Stroke, ilational
and Weuare, BeInstitutes of Health, Public Health Service, TJnited States Department of Health, Education,
thesda,Maryland 20014
JONATHAN

GALLANT

From the Department

of Genetics, University of Washington, Seattle, Washington 98105

We have examined the intracehular

levels of the guanosine
tetraphosphate,
guanosine
5-diphosphate
2- or 3-diphosphate (ppGpp), in an isogenic pair of stringent and relaxed
strains of Escherichia coli. As previously observed, relaxed
strains do not accumulate ppGpp during amino acid deprivation, while stringent strains do. However,
under all other
culture conditions
examined,
relaxed
and stringent
cells
regulate their ppGpp levels similarly.
During carbon deprivation or step-down
transitions
both strains accumulate
ppGpp.
Neither
strain accumulates
appreciable
amounts
during phosphate deprivation
or after treatment with colicin
El. The basal lev*ls of ppGpp observed in both stringent
and relaxed strains during balanced
growth are, in most
cases, the same, and vary inversely with the growth rate and
RNA content of the cells.

When Escl~erichia coli cells are deprived of an essential amino

acid or arc unable to activate one, there is a severe and preferential restriction of ribosomal and transfer RNA synthesis
(l-5).
This major adjustment
of biosynthetic
activity which
serves to maintain the RNA to protein ratio of the cell is termed
stringent control of RNA synthesis. This control requires t,he
product of the rel gene: rel- i strains are unable to curtail ribosomal RNA synthesis and maintain the RNA to protein ratio
Within a few
of the cell during amino acid deprivation
(6-8).
seconds of the imposition of amino acid deprivation,
relf cells
begin to accumulate an unusual tetraphosphate
of guanosine,
guanosine ii-diphosphate
2- or 3-diphosphate
(pp\+p)
above
the basal levels found in growing cells (9, 10). Although a basal
level of ppGpp is demonstratable
in rel- cells (ll), this level
does not increase appreciably during amino acid deprivation.
The coincidence of high intracellular
ppGpp levels and the
failure of cells to accumulate RNA during amino acid depriva* The abbreviat,ions
gent control.

used are: rel-, relaxed control;

leE+, strin-

tion and the kinetics with which changes in ppGpp levels are
effected by the cell suggests that this compound participates in
the amino acid control of RNA synthesis (9). Direct support
for the causal role of ppGpp in the inhibition of ribosomal RNA
synthesis was obtained by Travers, Kamen, Schleif, and Cnshel
(12, 13). They found that ribosomal RNA synthesis required,
in addition to the core RNA polymerase and u factor, a protein
factor, $r. They further showed that the #r-mediated
stimulation of ribosomal RNA synthesis was preferentially
inhibited
by PPGPP.
Under conditions other than amino acid deprivation,
both
relaxed and stringent cells are able to regulate the rate of ribosomal RNA synthesis and adjust the RNA:DNA
ratio of the
cell (7). When transferred from a rich medium to a poor one,
both relaxed and stringent cells sharply restrict their synthesis
of ribosomal RNA while allowing continued protein accumulation (14). When the RNA to protein ratio characteristic of the
new medium is achieved, ribosomal RNA synthesis resumes at a
rate characteristic of cells cultured on the poorer medium.
The
mechanism controlling RNA synthesis during growth rate transitions is not understood at present.
However, the fact that it
shares several characteristics with that operating during amino
acid starvation of a stringent strain suggests that they may be
related (14-16).
In the present communication
we show that during growth
rate transitions
both relaxed and stringent
cells accumulate
ppGpp in amounts approaching those observed during the stringent response to amino acid starvation.
Similarly,
during a
carbon source deprivation
both relaxed and stringent cells accumulate large amounts of ppGpp.
The basal levels of ppGpp
found in relaxed and stringent cells during balanced growth are,
in most cases, the same and vary inversely with growth rate and
RNA content of the cells. On the basis of these observations
we suggest that the intracellular
ppGpp levels and the rate of
ribosomal RNA synthesis are interrelated.
EXPERIMENTAL

PROCEDURE

Bacteria and Culture Conditions-E. coli NFl61 (filet, arg,

reZ+) and NE162 (met, urg, rel-) were obtained front Dr. Nils
Fiil.
These strains are of the K-12 lineage and arc believed to

4381

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

SUMMARY

4382

was grown in Tris glucose rnedium supplemented with Casamino

acids to a density of 2 x lo* cells per ml. Colicin production was
induced with 0.2 kg per ml of mitomycin, as described by Maeda
and Nornura (23). After induction, the cells were harvested by
centrifugation,
washed twice, and ruptured by passage through a
French pressure cell, as described by Fields and Luria (24). The
extract was clarified by centrifugation
at 105,000 x g for 30 min
and frozen in small portions. The colicin titer (2 x 1Ol2 per ml)
was determined from the fraction of cells surviving exposure to
the colicin preparation, with the equation b/b,, = e-, where b and
bo are the number of viable bacteria after and before exposure to
the colicin preparation and m is the multiplicity.
In all cases the
test organism was E. coli NF161.
Bacterial titers were determined by plating sufficiently diluted samples on nutrient broth
agar plates.
Materials--Silicate-free
32P-orthophosphoric
acid was obtained
from Tracerlab,
Division- of Laboratory
for Electronics,
Inc.
Iolyethylcncimine
cellulose sheets (MN-polygram
cell 300 polyethylenc~iminc) were the product of Brinkman
Instruments,
Inc.
Nitroccllulosc
filters (Metricel, GA-6, 47 mm diameter, 0.45 ~1
pore size) and glass filters (25 rnm diameter, type A, withoutj
binder) were purchased from Gelman Instrurnent
(ompany.
Liquiflor was a product of New England Nuclear.
RE:SULTS

oj ppGpp during Amino Acid Deprivation-In

a
previous publication
ppGpp was shown to accumulate in stringent strains but not in relaxed strains during amino acid deprivation (9). In the present report we show the accumulation
of
ppGpp in relaxed as well as stringent strains under several conditions of nutritional
deprivation.
For the purposes of comparison, the changes in the intracellular
concentration
of ppGpp in
the isogenic pair NF161 and NFl62 during methionine deprivation are presented in Fig. 1. Upon exhaustion of the methionine
in the rnedium the stringent strain abruptly accumulates large
amounts of ppGpp above the basal level of 0.04 to 0.05 nmole per
AGoo norrnally found in this strain.
In contrast the relaxed
strain, NF162, after amino acid deprivation shows no detectable
Accumdation

TIME (hours)
FIG. 1. Accumulation
of ppGpp
by Escherichia
coli during
methionine
deprivation.
At zero time growing
cultures
of E. coli
NF161 and NF162 were filtered,
washed,
and resuspended
in fresh
medium
containing
3 pg per ml of methionine.
[32P]PO(
(400
&i per ml) was added to a l-ml portion
of each culture.
Growth
(- - -) was monitored
turbidimetrically
in the unlabeled
culture
while
ppGpp
accumulation
(--)
was measured
in aliquots
in
the labeled
culture.

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

be isogcnic except for a small region of the chromosome which

iuclutlcs the rel locus. E. coli P687-54 (01 E1 (17) was obtained
fro111 1)r. John Inselburg.
Bacteria were cultured with the Tris
minimal medium described by Kaempfer and Magasanik
(18),
except that the KII~lOl concentration was 10e3 M. The carbon
sources for the media were glucose (0.25c/,), succinate (0.3%),
or alanine (0.50;).
Minimal
media contained 50 pg per ml of
each required amino acid. Where noted in the text, minimal
medium was supplemented
with 0.47;) vitamin-free
Casamino
a&Is.
Bacteria were cultured with reciprocal shaking in a
water bath at 37. Growth was measured by following the absorbance at 600 nm with a Zeiss PMQ II spectrophotometer.
1Jntlcr the conditions used, a cell concentration of 5 X lo* cells
per 1111yielded a turbidity of 1 at 600 Tim.
Determination of Intracellular
ppGpp Levels-The
ppGpp levcls ill [Y] phosphate labeled cells were determined radiochemicajly after extraction of the acid-soluble pools with formic acid
and i;epnrntion of ppGpp by thin layer chromatography
(9).
In :I typical csl)crimcnt,
cells were cultured in the requisite
metlium to a density of 0.5 x lo8 cells per ml and [32P]phosphoric acid added (final specific activity 200 to 500 &i per
pmole) to a portion (1 tb 2 ml) of the culture.
Growth was
nlolritored in the unlabeled portion of the culture.
When high levels of ppGpp were expected, a one-dimensional
separation with 1.25 or 1.5 M KIMOl
as the developing solution
(9, 19) was used. When basal or very low levels of ppGpp were
mc:lsured, a two-dimensional
separation was used to reduce the
background radioactivity.
The chromatograms
were developed
with the step-formate system in the first dimension and with
I .25 M KR&O~ in the second (19). ppGpp was located by
radioautography
and its radio content determined
by liquid
scintillation counting.
The [32P]-phosphate specific activity of the media was determined for each individual experiment.
A portion (5 ~1) of the
suitably diluted sample of medium was spotted on a circle of
polycthyleneiminc
cellulose, dried, and counted in Liquiflor
scintillation fluid. The phosphate concentration of the medium
at the tirne of the [Y]phosphate addition was assumed to be the
~amc as that of fresh media (lob3 M), although the bacteria had
been cultured in it to a density of approximately
0.5 x lo* cells
per ml. The error introduced by this assumption is approximately a 55% overestimation.
In the case of rnedia that contained Casamino acids, the phosphate concentration at the time
of [92P]phosphate addition was determined with the calorimetric
assay of Arnes and Dubin (20).
Calorimetric Rstimation of RNA and DNA-To
determine the
RXA and DNA content of cells, duplicate samples (25 ml) were
removed from a growing culture and rnixed with 2.5 ml of 500/,
trichloroacetic
acid at 0. After at least 15 rnin at 0 the cells
were collected on glass fiber filters and washed with 45 ml of 5%
trichloroacetic
acid. The filters containing the entrapped cells
were folded and placed in the bottom of a test tube, 13 x 100 mm,
and covered with 2 ml of trichloroacetic
acid (5yi).
The tube
was cal)pcd with a marble and placed in a 90-95 wat,er bath for
20 mill. After cooling the tube, the glass fibers were sedimentecl
by centrifugation and portions of the supernatant fluid were used
in either the orcinol calorimetric assay for RNA (21) or the diphenylamine assay for DN.4 (22). When only RNA was to be
determined the initial volume of cells removed from t,he culture
was reduced to 5 ml.
Preparation
of Colicin K,-The
colicinogenic
strain P687-54

Vol. 246, No. 14

Issue of July

R. A. LaxxariG,

25, 1071

IV.

Cashel,

and

J. Gallant

4383

TABLE

in E. coli NF162 during single and multiple

amino
deprivations
acid
A growing culture of E. coli NF162 was filtered and resllspcndcd
in complete medium which was lacking amino acids. Portions of
the culture were either left unsupplemented,
or supplemented
ppGpp

with

levels

arginine

together.

alone,

methionine

alone,

[32P]orthophosphate

each culture,
and samples
after 30 min.
Labeling

were

or arginine

and methionine

(500 ,uCi per pmole) was added to

removed

for ppGpp

determinations

condition

Complete..
- Methionine.
..
. . .
- Arginine. .
- Methionine
and arginine. . .

14
14.6
12.1
39.4

TIME

might

in fact

be related

to an endogenous

that is occasioned by the eshaustion

e R,. A. Lazzarini,

unpublished

amino

acid

deprivation

of the easily utilizable

experiments.

glu-

FIG.
2. Accumulation
of ppGpp by Escherichia
coli NFlGl
md
NF162 during
glucose deprivation.
At zero time growing
cultures
of each organism
were filtered,
washed,
and resuspended
in fresh
medium
containing
0.025%
glucose.
[ZP]P04
(200 pCi per ml)
was added to l-ml portions
of each culture.
Growth
(- - -) and
ppGpp
accumulation
(---)
were monitored
as in Fig. 1.

500,

TIME (hours)

FIG.
3. Accumulation
of ppGpp
by Escherichia
coli NFl61
and
NF162
during
a glucose
to succinate
step-down
kansition.
1Xach
strain
of E. coli was cultured
in Tris minimal
medium
containing
0.025%
glucose
and 0.394 succinate
as sole carbon
sources.
At
zero time [32P]POd was added to l-ml portions
of each culture
to
yield a final concentration
of 100 &i
per pmole.
Growth
(-- - -)
and ppGpp
accumulation
(--)
were monitored
during
the collrse
of the step-down
transition
as in Fig. 1.

case. Thus, the rapid accumulation of ppGpp in NFl61 rn:ly reflect a response to amino acid starration, superimposed on c::wbon
source deprivation.
The relased strain being insensitive
to
amino

acid

deprivation

however

would

synthesize

ppGp1)

ill re-

sponse to only one stimulus.

Furthermore, the difference in rate
of ppGpp accumulation apparent in Fig. 3 is not observed in all
step-down transitions.
With another relaxed and stringent pair
of strains, E. coli EAl and EA2, a carbon source step-down from
glucose to proline elicits approximately
the same rate of ppGpp
accumulation in both strains, although the stringent strain crentually accumulates slightly more. In still another isogenic pair
of E. coli strains, CP78 and CP79, undergoing a step-down from
glucose to lactate, ppGpp accumulates in both strains at approximately the same rate.3 In this latter case, the final lerel of
ppGpp in both strains is considerably less (0.2 nmole per AGO,,)
3 I<. M. Winslow,

unpublished

experiments.

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

increase above its basal level of 0.01 to 0.02 nmole per AeoO. The
failure of NF162 to increase its ppGpp level cannot be attributed
to a lack of severity of the amino acid starvation, since deprivation of arginine alone or simultaneously
with methionine only
marginally
increases the intracellular
concentration
of ppGpp
(Table I).
Accumulation
of ppGpp during Carbon LIeprivation and StepDown Transitions-The
ability of relaxed strains to accumulate
substantial amounts of ppGpp can be shown during a simple
carbon source starvation.
In Fig. 2 the effects of glucose starvation on the ppGpp levels of NF161 and NF162 are shown. The
stringent strain, NF161, upon carbon deprivation
accumulates
amounts of ppGpp that are equivalent to those found in amino
acid-deprived
cells. The relaxed counterpart,
NF162, also accumulates ppGpp upon carbon deprivation, but at a slower rate:
after 1 hour of deprivation the level of ppGpp in the relaxed
strain is approximately one-half that found in t,he stringent strain
after 10 min of deprivation.
Qualitatively,
the same results can
be observed during a step-down transition between a rich and a
poor
medium.
In the example shown in Fig. 3, NF161 and
NF162 exhaust the glucose present in the medium at a cell density
of 2 x lo8 and are forced to use succinate that is also present in
the medium as a source of carbon. As pointed out by Niedhardt
(7), both cultures traverse the step-down transition in a nearly
identical manner.
After depleting the glucose both strains enter
a diauxic lag during which the turbidity of the cultures slowly increases, presumably as the result of the synthesis and accumulation of protein and DNA.
RNA accumulation abruptly ceases
at the break in the growth curve and does not resume until the
turbidity of the culture increases 1.35.fold above that observed at
glucose depletion.2
During this period the RNh:DNA
ratio of
the cells is adjusted to that characteristic of cells cultured on the
poorer medium (15). In the present case the RNb1:DNA
ratio
changes from 7.4 to 5.5 (see below and Fig. 5).
depletion of the glucose both strains behave as they do
Upon
during
a simple
glucose
deprivation
and accumulate ppGpp, although a utilizable carbon source is present in the media. The
difference in the kinetics of appearance of ppGpp in the two
strains during the carbon deprivation
or step-down transition

(hours)

4384

Regulation o;f ppGpp Levels

than is observed in NF161 undergoing the glucose to succinate

shift.
Limited ppGpp Accumulation
during Restricted Phosphate or
Nucleotide Metabolism-Several
experimental or nutritional
conditions which restrict RNA accumulation presumably by altering
the availability
of nucleotide triphosphates
elicit little or no
ppGpp accumulation.
Phosphate deprivation
of NF161 and
NF162 leads to only a marginal increase in the ppGpp level of
both strains (Fig. 4). Upon depletion of phosphate in the medium the turbidity of the cultures continue to increase but at a
much reduced rate. Within 5 min of the break in the growth
curve the major purine nucleotide pools, ATP and GTP, contract

(hours)

Fro. 4. Accumulation of ppGpp by Escherichia coli NF162 and

NF161 during phosphate deprivation.
Growing cultures of both
strains of E. coli were filtered and resuspended
in fresh media
containing
of each

1 X 10-4 M phosphate.

[32P]P0,

was added

culture to yield a final concentration

Growth (- - -) and ppGpp accumulation
(-)
as in Fig. 1.

to portions

of 40 pCi per ml.

were monitored

246, No. 14

to approximately
one-third of their normal size but do not undergo any further change. The int.racellular
concentration
of
ppGpp, on the other hand, first expands slightly, and then contracts upon continued phosphate deprivation.
Neither uracil nor guanine deprivation of stringent auxotrophs
of E. coli elicits ppGpp accumulation.
Treatment
of E. coli
with colicin El leads to the rapid inhibition of RNA, protein, and
DNA syntheses (25). It has been proposed that the inhibition of
these diverse synthetic processes is a secondary effect of a primary inhibition
of oxidative phosphorylation
(25). Treatment
of NF161 and NF162 with colicin preparations at a multiplicity
of
5 leads to an abrupt reduction in the intracellular
levels of ATP
and GTP but less than a 2-fold increase in the intracellular
ppGpp concentration.
Intracellular
Levels of ppGpp during Balanced Growth-Under
all of the experimental
and nutritional
conditions examined except amino acid deprivation, relaxed and stringent cells similarly
regulate their intracellular
ppGpp levels. We were led to investigate whether the similarity in the control of ppGpp levels
extended to cells in balanced growth as well. For this purpose,
ppGpp levels were measured in relaxed and stringent cells growing on several media which support growth at widely different
rates. Since the RNA content of a cell varies regularly with
growth rate (26,27), this comparison also can reveal a systematic
variation of ppGpp level with RNA content.
The results, shown
in Fig. 5, indicate that relaxed and stringent cells contain approximately the same level of ppGpp when cultured in the same
medium, except for glucose minimal medium.
In that case the
relaxed contains about one-third the ppGpp level of the stringent (11).
Furthermore, it is clear that with the same exception,
the levels of ppGpp vary regularly and approximately
linearly
with the RNA content of the cell.
DISCUSSION

,\

IO

12

RNA/DNA

5. Basal levels of ppGpp and RNA content of Escherichia

coli cultured
on different
growth
media.
E. coli NF161 (0) and
FIG.

NF162

(A)

were

cultured

on minimal

media

containing

as a sole

carbon source 0.5% alanine, 0.3% succinate, 0.25% glucose or

0.25y0 glucose plus 0.40/, Casamino acids. The cells were cultured
for four to six doublings
in the appropriate
medium
prior
to the
addition
of [3zP]POc
(400 &i
per ml).
After
one doubling
in
the presence
of [32P]P0, samples
were removed
for ppGpp
estima-

tion; RNA and DNA were determined in the same sample of a

parallel unlabeled culture.
The doubling times for both E. coli
NF161 and NFl62 were 205 min with alanine, 90 min with suceinate,

55 min with

glucose,

and 40 min with

acids as sole sources of carbon.

glucose

plus

Casamino

The nucleotide ppGpp was originally

observed as material
which accumulated in stringent but not in relaxed cells during
amino acid deprivation
(9). The present observations provide
a more comprehensive picture of the conditions governing the
accumulation
of ppGpp.
It is clear that relaxed cells are by no
means unable to produce the nucleotide: relaxed cells contain a
significant basal level of ppGpp, which varies with growth rate,
and they are able to accumulate r&her high levels during carbon
source starvation or carbon source downshift.
We may therefore reject the simplest interpretation
of earlier results, namely
that the rel- mutation inactivates the enzyme responsible for
ppGpp biosynthesis.
The rel- mutation seems rather to affect
the relationship between ppGpp formation and some specific consequence of amino acid deprivation.
Since relaxed mutants do
accumulate ppGpp during downshift, but do not during amino
acid deprivation, it follows that these two nutritional
conditions
regulate the nucleotides accumulation
through different mechanisms, or by means of different signals.
Neidhardt (7) has reported that several different stringent and
relaxed strains show a similar variation in RNA content with
growth rate. The RNA:DNA
ratios presented in Fig. 5 extend
this conclusion to an isogenic pair of strains, confirming the conclusion that growth rate control of RNA accumulation
is unimpaired by the rel- mutation.
The basal levels of ppGpp also
vary systematically with growth rate in both members of the isogenie pair, leading to a striking inverse correlation
between
ppGpp concentration
and RNA:DNA
ratio (Fig. 5). If the

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

TIME

Vol.

lssuc

of *July 25, 1971

R. A. Laxzarini, M. Cashel, and J.

4 H. Erlich, T. Laffler, and J. Gallant,

unpublished

experiments.

4385

In summary, the results presented here indicate that relaxed

mutants have lost the ability to produce ppGpp in response to
amino acid deprivation,
but retain the ability to produce the
nucleotide in response to other conditions.
The conditions under
which ppGpp accumulates suggest that it participates in a mechanism governing the synthesis of ribosomal RXA in both stringent and relased cells.
REFERENCES
1. STENT, G. S., AND BRENNER,
S., Proc. Nat. Acacl. Xci. U. S. A.,
47, 2005 (1!161).
2. FANGMAN,
W. L., AND NEIDH.~RDT,
F. C., J. Biol. Chem., 239,
1844 (1964).
3. FORCHHAMMER,
J., AND KJELDGAARD,
N. O., J. Mol. Biol.,
37,
245 (1968).
4. LAZZBRINI,.
R. -4., AND DAHLBERG,
A. E., J. Biol.
Chem., 246,
420 (1971).
5. L~v.YLL~,
It.,
.ZND DE H.~u~EI~,
G., J. Mol.
Biol.,
37, 269
11968,
6. BORF,K,
E., R.OCI<F:NBACK,
J., AND RYAN,
A., J. Bacterial.,
71,
318 (1956).
7. NEIDIIUDT,
F. C., Biochim.
Biophys.
Acta, 68, 365 (1963).
8. NEIDIXAI~D~P,
F. C., Biochim.
Biophys.
Acta, 68, 380 (1963).
9. CASHEEL,
M., J. Hiol. Chem., 244, 3133 (1969).
10. CASIIEL,
M., AND KALIUCHER,
B., J. Biol.
Chem.,
246, 2309
(1970).
11. GALLANT,
J., ERLICH,
H., HALL,
B., AND LAFFLER,
T., Cold
Spring
Harbor
Symp.
Quant.
Biol.,
36, 397 (1971).
12. TRAVERS,
A., KAMEN,
R., AND SCHLIEF,
R. F., Nature,
228, 748
(1970).
13. TRAVERS,
A., KAMEN,
R., AND CASHEL,
M., Cold Spring Harbor
Symp.
Quant.
Biol.,
36, 415 (1971).
14. LAZZARINI,
R. A., AND WINSLOW,
R. M., Cold Spring Harbor
Symp.
Quant.
Biol.,
36, 383 (1971).
15. NIXDHARDT,
F. C., AND FRAENKEL,
D. G., Cold Spring
Harbor
Symp.
Quant.
Biol.,
26, 63 (1961).
16. MAAL@,
O., Twenty-eighth
Symposium
of Society for Developmental
Biology:
Communication
in Development,
University
of Colorado,
1969.
17. INSELBURG,
J., J. Bacterial.,
102, 642 (1970).
18. KAEMPFER,
R. 0. R., AND MAGASANIK,
B., J. Mol. Biol., 27,
475 (1967).
19. CASH&+
M., LAZZAIEINI,
R. A., AND KALBACHER,
B., J. Chromatogr.,
40, 103 (1969).
20. AMES, B. N., AND DUBIN,
D. T., J. Biol. Chem., 236,769
(1960).
21. ScHNmDk:R, W. C., J. Biol.
Chem., 161, 293 (1945).
22. BURTON,
K. A., Biochem.
J., 62, 315 (1956).
23. MAEDA,
A., AND NOMURA,
M., J. Bacterial.,
91, 685 (1966).
K. L., AND LURIA,
S. E., J. Bacterial.,
97, 57 (1969).
24. FIELDS,
25. FIELDS,
K. L., AND LURIA,
S. E., J. Bacterial.,
97, 64 (1969).
26. NEIDIURDT,
F. C., AND MAGASANIK,
B., Biochim.
Biophys.
Acta, 42, 99 (1960).
27. SCHAECHTICR,
M., MAAL$E,
O., AND KJELDG~ARD,
N. O., J.
Gen. Microbial.,
19, 592 (1958).
R. A., NAKATA,
K., AND WINSLOW,
It. M., J. Biol.
28. LAZZARINI,
Chem., 244, 3092 (1969).
J., ROSSET,
R., AND MONIER,
R., Bull.
SOC. Chim.
29. JULIEN,
Biol.,
49, 131 (1967).
T., HORIUCHI?
S., AND MIZUNO,
D., Biochim.
Bio30. HORIUCHI,
phys. Acta, 31, 570 (1959).
M., Cold Spring
Harbor
Symp.
Quant.
Biol.,
36, 407
31. CASH~L,
(1971).
J. T.-F.,
AND NAZAR:
R. N.: J. Biol. Chem., 246, 4591
32. WONG,
(1970)
D. H., AND ELKINS,
B. N., Biochim.
Biophys.
Acta,
33. EZEKIEL,
166, 466 (1968).
\----I

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

line:lr relationship displayed in Fig. 5 extends to very low levels of

ppC;pp, one would expect a limiting RNA : DNA ratio of about 11,
B. coli cells cultured on nutrient broth, at near maximum growth
rate, have an RNA:DNA
ratio of 10. The one departure from
this rqulnr relationship, that of relaxed cells growing in glucose
minimal Incdium, is at present an enigma.
There seem to be at least two physiological mechanisms by
which net synthesis of RNd is curtailed.
During amino acid
deprivation of a stringent strain (4), and during carbon source
downshift of a relaxed (presumably also of a stringent) (14), the
synthesis of ribosomal and transfer RNA is preferentially
inhibited.
On the other hand, during deprivation of uracil (28) or
phosphate (29, 30), ribosomal RNA appears to be made, but
fails to accumulate because synthesis is offset by abnormally
At present, there is not sufficient evidence
rapid degradation.
available to permit conclusions as to the mode of RNA control
exercised during guanine deprivation, colicin treatment, or balanced growth at different rates. Nonetheless, the rudimentary
correlation that emerges from this comparison is that ppGpp
accumulates under two conditions where ribosomal RNA synthesis is preferentially
inhibited, while under at least two other conditions RNA accumulation is restricted by rapid degradation and
ppGpp does not accumulate.
This correlation suggests that the
control of ribosomal RN.4 synthesis involves pl)Gpp while the
control of ribosomnl RNB degradation does not.
Recent studies on transcription
in vitro suggest a plausible
mechanism for the preferential inhibition of ribosomnl RNA synthesis by ppGpp.
Cashel (31) has shown that ppGpp inhibits
core and u-supplemented RNA polymerase in a specific although
limited manner.
Moreover, Travers, et al. (13) have shown that
ppGpp preferentially
inhibits the production of ribosomal RNA
in vi&o by RNA polymerase in the presence of #r factor. These
observations suggest that ppGpp governs the specificity of RNA
polymerase, and in particular the ability of the enzyme plus $r
factor to transcribe the ribosomal RNA cistrons.
Since the
RNA content of cells reflects predominantly
the quantity
of
ribosomal RNA, the relationship
shown in Fig. 5 is consistent
with this model.
Our results are also consistent with a different interpretation
Wong
of the relationship between ppGpp and RNA synthesis.
and Nazar have observed that the addition of rifampicin to cells
which have accumulated ppGpp during amino acid deprivation
(32) results in a reduction in ppGpp concentration, which appears
The eventual disapto precede an effect on protein synthesis.
pearance of ppGpp after the inhibition of protein synthesis is to
be expected, since inhibitors of protein synthesis antagonize the
stringent response (33). These results suggest that the RNA
polymerase itself may be responsible for the synthesis of ppGpp
and that the accumulation of large amounts of ppGpp may be a
consequence rather than a cause of the restricted RNA accumulation.
However, the significance of the rifampicin effect is difficult to assess, since the effect may be an indirect one. Experiments to be reported elsewhere4 will show ppGpp formation can
proceed in the presence of rifampicin under certain conditions,
suggesting that the effect of rifampicin on ppGpp accumulation
is indirect.

Gallant

CONTROL MECHANISMS AND

BIOCHEMICAL GENETICS:
On the Regulation of Guanosine
Tetraphosphate Levels in Stringent and
Relaxed Strains of Escherichia coli
Robert A. Lazzarini, Michael Cashel and
Jonathan Gallant
J. Biol. Chem. 1971, 246:4381-4385.

Find articles, minireviews, Reflections and Classics on similar topics on the JBC Affinity Sites.
Alerts:
When this article is cited
When a correction for this article is posted
Click here to choose from all of JBC's e-mail alerts
This article cites 0 references, 0 of which can be accessed free at
http://www.jbc.org/content/246/14/4381.full.html#ref-list-1

Downloaded from http://www.jbc.org/ by guest on August 25, 2015

Access the most updated version of this article at http://www.jbc.org/content/246/14/4381