Journal of General Microbiology (1982), 128,705-712.
Printed in Great Britain
705
Cyclic AMP and Cyclic GMP Control of Synthesis of Constitutive
Enzymes in Escherichia coli
By P E T E R H . C A L C O T T t
Department of Biological Sciences, Wright State University, Dayton, Ohio 45435, U.S.A.
(Received 5 May 1981; revised 14 July 1981)
Escherichia coli was grown in chemostat culture under glycerol-limited and ammoniumlimited conditions at growth rates between 0.1 and 0.5 h-l. At steady state, the
concentrations of cyclic AMP and cyclic GMP and the activities of four constitutive enzymes
(glucose-6-phosphate dehydrogenase, isocitrate dehydrogenase, NADH oxidase and cyclic
phosphodiesterase) were determined in the organism. Addition of exogenous cyclic AMP,
cyclic GMP or phencyclidine perturbed the steady state and caused inhibition or stimulation
of synthesis of phosphodiesterase and isocitrate dehydrogenase. A novel hypothesis is
proposed to account for the ability of bacteria to regulate the synthesis of constitutive
enzymes with cyclic nucleotides and possibly other small molecules.
INTRODUCTION
When bacteria grow, they can modulate their enzyme complement to adjust to changing
environments. With a few exceptions, the precise factors which stimulate or repress synthesis
of enzymes are unknown (Chock et al., 1980; Drew & Demain, 1977; Matin, 1981; Nierlich,
1978; Pastan & Adhya, 1976; Rickenberg, 1974). Matin (1981), while reviewing control of
enzyme synthesis in microbes grown in continuous culture, concluded that small molecules
acted as effectors in controlling the synthesis of a variety of constitutive enzymes. He
postulated that cyclic AMP might be one of these small molecules. Cyclic AMP is a potential
effector since it has been implicated in the control of numerous inducible enzymes and to a
lesser degree in constitutive enzymes (Aono et al., 1978; Calcott & Calvert, 1981; Dietzler et
al., 1977; Pastan & Adhya, 1976; Peterkofsky, 1976; Rickenberg, 1974). Cyclic AMP
concentrations and adenylate cyclase activities have been shown to be modulated by growth
rate in chemostat populations of Escherichia coli (Botsford & Danley, 1975; Harman &
Botsford, 1979). More recently, cyclic GMP has been implicated in cell metabolism, though
the evidence is less strong (Black et al., 1980; Cook et al., 1980; Goldberg et al., 1975;
Gonzalez & Peterkofsky, 1975; Pastan & Adhya, 1976; Peterkofsky, 1976).
In the work described here, the relationship was investigated between the activities of four
constitutive enzymes and the concentration of cyclic AMP and cyclic GMP in E. culi grown
under a variety of growth conditions in continuous culture. In addition, the effect of exogenous
cyclic nucleotides and phencyclidine on the synthesis of two of the enzymes was investigated.
METHODS
Organism and cultural conditions. Escherichia coli (45 1-B variant) was maintained on nutrient agar at room
temperature after growth at 37 OC for 24 h. Organisms were grown aerobically in chemostats of effective volume
500 ml (except in certain experiments where it was lowered to 50 ml) at 37 "C, at growth rates from 0.10 to
t Present address: CR-Bioproducts Laboratory, The Dow Chemical Company, Midland, MI 48640, U.S.A.
Downloaded from www.microbiologyresearch.org by
0022-1287/82/0000-9935 $02.00 O 1982 SGM
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
706
P. H . C A L C O T T
0.5 h-' (Calcott & MacLeod, 1974) in a simple salts/ammonium/glycerol-based medium (Postgate & Hunter,
1962). Glycerol-limiting medium contained 2 g glycerol 1-' and 50 mM-ammonium, while ammonium-limiting
contained 10 g glycerol 1-' and 6 mwammonium ions. Cultures were grown in the chemostat to steady state (as
judged by a constant turbidity over a 24 h period) before samples were removed for analysis. In all experiments,
steady states were obtained at the lowest growth rate first (0.10 h-'). Once analysis had been performed, a new
steady state was established at a specific growth rate ( f i ) of 0.20 h-l, then at 0.30 h-' and finally at 0.50 h-l. After
analysis, the growth rate was then decreased to 0.10 h-' and the steady state reestablished. The data for this
steady state never deviated by more than 10%from those obtained from the population initially at p = 0.1 h-l. In
general, steady states required more than seven replacements of the culture to establish. Chemostats were never
operated for more than one month. In certain experiments, steady states were perturbed by addition of cyclic AMP
(10 mM), cyclic GMP (10 mM) or phencyclidine (0.1 mM) to the culture vessel and by changing the inflow medium
to one with the additive at the appropriate concentradon. At various times, samples were removed for analysis.
Under no growth condition did the bacterial carbohydrate content determined by the anthrone method (Calcott &
MacLeod, 1974) exceed 10% dry weight. This prevents complications in interpreting the concentrations of
nucleotides and enzymes in the organism.
Enzyme and protein assays. Isocitrate dehydrogenase (EC 1.1.1.42), glucose-6-phosphate dehydrogenase (EC
1.1.1.49), 2' :3'-cyclic nucleotide 2'-phosphodiesterase (EC 3 . 1.4.16) and NADH oxidase (EC 1.6.99.3) were
assayed as described before (Calcott, 1981a; Calcott & Calvert, 1981) on extracts prepared by ultrasonic
disruption. Protein was determined by the Lowry method with bovine serum albumin as standard.
Cyclic nucleotide extraction and determination. Samples of the culture were centrifuged and washed in growth
medium without carbon source before the organisms were resuspended in 5 % (w/v) trichloroacetic acid (Gersch ef
al., 1978). The preparations were then frozen at -80 "C, thawed and sonicated for 6 min (Gersch et al., 1978).
The extract was centrifuged to remove debris and washed thrice in diethyl ether to remove lipids; the fluid was then
applied to a Dowex (1 x 2-200) column. The column was eluted sequentially with distilled water, 50 mM-HCl (to
remove cyclic AMP) and then 500mM-HCI (to removc cyclic GMP). The acid fractions were collected and
freeze-dried before reconstitution with boiled, glass-distilled water. Recove6y of cyclic nucleotides, which was
between 80 and 90%, was determined by monitoring recovery of [3Hlcyclic AMP or [3Hlcyclic GMP from
parallel samples taken through the freezing, sonication, ether washing, chromatography and freeze-drying steps.
No degradation of the cyclic nucleotide could be detected in the extraction and purification procedures as
determined by paper chrfmatography (Calcott & Calvert, 1981). Cyclic AMP and cyclic GMP were determined
by protein-binding assay systems (Boehringer-Mannheim);0.5 pmol cyclic AMP and 0.1 pmol cyclic GMP could
be detected. The authenticity of the extracted nucleotides was confirmed since they co-chromatographed with
authentic nucleotide and were at least 90 % degraded'by commercial cyclic phosphodiesterase (Sigma: no. PO134).
Cyclic nucleotide contents of bacteria were expressed as pmol or fmol (mg dry wt)-'. Bacterial dry weights were
determined turbidometrically from a calibration curve.
Chemicals. All chemicals were purchased from Sigma or from Fisher Chemical Company, Cincinnati, Ohio,
except phencyclidine ( 1-phenylcyclohexylpiperidine)which was purchased from Applied Science Division of
Milton Roy Company, State College, Pennsylvania.
RESULTS
When Escherichia coli was grown in chemostat culture, the concentrations of cyclic AMP
and cyclic GMP were dependent on both growth rate and nutrient limitation (Figs 1 and 2).
For glycerol-limited cultures the concentrations of cyclic AMP and cyclic GMP decreased as
growth rate increased (Fig. l), while for ammonium-limited cultures, the cyclic GMP
increased and the cyclic AMP was unchanged (Fig. 2). The concentrations of cyclic
nucleotides were higher in glycerol-limited cultures than in ammonium-limited ones. The
concentration of cyclic AMP also decreased with increasing growth rate in glucose-limited and
lactose-limited cultures in a similar fashion, with glucose-limited cultures containing low
(16 pmol mg-') and lactose-limited cultures intermediate (52 pmol mg-') concentrations of
cyclic AMP at a specific growth rate of ,.u = 0.1 h-l.
The activities of four constitutive enzymes, isocitrate dehydrogenase, glucose-6-phosphate
dehydrogenase, NADH oxidase and phosphodiesterase, were investigated in organisms
grown in continuous culture (Figs 1 and 2). In general, highest activities were detected in
glycerol-limited organisms grown at the highest rates. In ammonium-limited cultures, a
similar trend was noted. Under both nutrient limitations, each enzyme appeared to be
regulated independently.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
c
0
0
N
P
I
I
Cyclic GMP
I fmol (mg dry wt)-'
0
0
0
0
e
0
m
Cyclic AMP
lpmol (mg dry wt)-'l
I
n
0
0
'
c
0
I
o
e
I
m
c
o
-
v
Glucose-6-phosphate
dehydrogenase
v,
Isocitrate dehydrogenase
0
w
,
c
-
I
m
Phosphodiesterase
0
N
0
L
NADH oxidase
8
w
L
0
w
8
Cyclic GMP
fmol (mg dry wt)-*
c
8
0
v,
Cyclic AMP
[pmol (mg dry
708
P. H. CALCOTT
Table 1. Statistical relationships between the activities of four constitutive enzymes and cyclic
nucleotide concentrations in E. coli
Enzyme activities and cyclic AMP and cyclic GMP concentrations were determined in steady-state
populations of bacteria grown in chemostat. The data were fitted to the equation z = c + a,x + a g ,
where z represents enzyme activity, x and y represent the concentrations of cyclic AMP and cyclic
GMP, respectively, and c, a, and a, are constants (see text). The statistical analysis was made by
multiple linear regression analysis by the methods of least squares; r2, degrees of freedom (DF) and t
values were determined and the t value was converted to probability with Student t-tables.
Model parameters
I
Enzyme
Phosphodiesterase
Isocitrate dehydrogenase
Glucose-6-phosphate dehydrogenase
NADH oxidase
0
*
1
C
a,
4.0
4.4
2.43
-2.22
-0.032
-0.007
2
+Om019
-0.02
3
0
Time (h)
a2
P
DF
r2
+0*01
<0.088
(0.0076
8
8
8
8
0.60
0.86
0.78
0.99
-0.023
-0.065
+0.03
1
2
(0.058
<0*0001
3
Fig. 3. Effect of exogenous cyclic AMP and cyclic GMP on the activities of isocitrate dehydrogenase
and phosphodiesterase in chemostat cultures of Escherichiu coli. The bacteria were grown to steady
state in glycerol-limiting chemostat culture of 50 ml effective volume at a growth rate of 0.50h-l. At
time 0, cyclic AMP (a) or cyclic GMP (b) was added to the growth vessel to a final concentration of
1 0 m and
~ the inflow medium changed to one containing the cyclic nucleotide. At various times
The
samples were removed and assayed for phosphodiesterase (a)and isocitrate dehydrogenase (0).
dotted lines represent variation (range of values) experienced in a parallel unperturbed culture. The
mean control activities were: phosphodiesterase, 3 -2 units (mg dry wt)-'; isocitrate dehydrogenase, 9
units (mg dry wt)-'. The bars represent the range of results for each point.
Since there appeared to be a linear relationship between cyclic nucleotide concentrations
and certain of the enzyme activities, a linear model, as a first-order approximation, was tested.
Multiple regression analysis of the relationships between the concentrations of the two cyclic
nucleotides and the activities of the four enzymes yielded regression equations of the general
form z = c + a l x + a,y, where z was the enzyme activity (pmol substrate consumed or
product formed min-' mg-l), c (pnol substrate consumed or product formed min-' mg-'), a,
[pmol substrate consumed or product formed (pmol cyclic AMP)-'] and a, [pmol substrate
consumed or product formed (pmol cyclic GMP)-'I were constants and x and y were the
concentrations of the cyclic AMP (pmol mg-') and cyclic GMP (fmol mg-'), respectively
(Table 1). The relationships were statistically significant for each set of eight data points at
least at the 0-1 level. From this analysis, between 60 and 99% of the variation in enzyme
activity could be explained by altered cyclic nucleotide concentrations. Although the activities
of the four enzymes could be predicted by the concentrations of the two cyclic nucleotides,
each enzyme exhibited a different relationship with different values of a, and a,, both positive
and negative.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
Cyclic AMP and cyclic GMP in E. coli
0
1
2
Time (h)
3
"
0
1
2
Time (h)
709
3
Fig. 4. Effect of phencyclidine on the activities of isocitrate dehydrogenase and phosphodiesterase, and
the bacterial concentration of cyclic AMP and cyclic GMP, in chemostat cultures of Escherichiu coli
grown as described for Fig. 3. At time 0, phencyclidine was added to the growth vessel to a final
concentration of 0-1 mM and the inflow medium changed to one containing the drug. At various times
and phosphodiesterase (0)
samples were removed and assayed (a) for isocitrate dehydrogenase (0)
activities, and (b) for the concentrations of cyclic AMP (A) and cyclic GMP (A). The dotted lines
represent variation (range of values) experienced in a parallel unperturbed culture. The mean control
activities were: isocitrate dehydrogenase, 8.6 units (mg dry wt)-'; phosphodiesterase, 3-5units (mg dry
wt)-'. The bars represent the range of results for each point.
While relationships were found between enzyme activities and cyclic nucleotide
concentrations, causality could not be determined with the data presented. To determine
whether the observed enzyme activities were direct results of the cyclic nucleotide
concentrations, cyclic AMP or cyclic GMP was added to steady-state cultures of E. coli and
cyclic phosphodiesterase and isocitrate dehydrogenase were monitored (Fig. 3). When the
culture was perturbed by cyclic AMP, there was an inhibition in the synthesis of the
phosphodiesterase while the synthesis of isocitrate dehydrogenase was not affected. In a
similar fashion, perturbation of the culture with cyclic GMP caused a stimulation in synthesis
of the phosphodiesterase and an inhibition in synthesis of the isocitrate dehydrogenase. This
is consistent with the values in Table 1, where phosphodiesterase activity was negatively
related to cyclic AMP and positively related to cyclic GMP concentrations. Similarly, the
effects of perturbation by cyclic nucleotides on isocitrate dehydrogenase activities were as
predicted from Table 1, where enzyme activities were negatively related to both cyclic AMP
and cyclic GMP concentrations, with the effect of cyclic AMP, though statistically
significant, being smaller than that for cyclic GMP (Table 1). Additions of cyclic nucleotides
did not in themselves stimulate or inhibit the observed enzyme activities in cell extracts
(results not shown). This indicated that the cyclic nucleotide concentrations could be directly
controlling the synthesis or degradation of the enzymes.
To test further the relationships between enzyme activities and cyclic nucleotide
concentrations, the effect of inhibitors of cyclic nucleotide metabolism on enzyme synthesis
was sought. While methyl xanthines are effective inhibitors of'cyclic phosphodiesterases in
animal systems, Klebsiella and Serratia, they are not in E. coli (Calcott & Calvert, 1981;
Peterkofsky, 1976). However, phencyclidine has been shown to inhibit guanylate cyclase in
animal cells (Vesely, 1979). Phencyclidine was added to a steady-state culture of E. coli and
the concentrations of the two cyclic nucleotides and the activities of the two enzymes,
phosphodiesterase and isocitrate dehydrogenase were followed (Fig. 4). Phencyclidine caused
a rapid and similar decrease in the concentrations of both cyclic nucleotides, which appeared
in the growth medium indicating that the mechanism of action of the drug might be by
stimulating excretion. During these decreases in cyclic nucleotide concentration, isocitrate
dehydrogenase synthesis was stimulated and that of phosphodiesterase inhibited. This result
is consistent with the values in Table 1.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
710
P . H. C A L C O T T
DISCUSSION
Matin (1981) has reviewed the control of enzyme synthesis in microbes grown in
continuous culture. Of 5 1 bacterial enzyme systems studied, 95 % showed enzyme synthesis
influenced by growth rate and/or nutrient limitation, whereas of 47 eukaryotic systems
studied, 77% were influenced by these two parameters. Matin (1981) predicted that small
molecules, such as cyclic AMP, could play crucial roles in regulating enzyme synthesis. In
fact, cyclic AMP concentrations and the activity of adenylate cyclase were growth-rate
dependent in chemostat cultures of E. coli (Botsford & Danley, 1975; Harman & Botsford,
1979). The observation of modulation of cyclic AMP and cyclic GMP concentrations by
growth rate and nutrient limitation in chemostat cultures reported here not only confirms
Botsford’s studies and Matin’s predictions, but extends them to include cyclic GMP.
Pastan & Perlman (1967) proposed four criteria to implicate cyclic nucleotides in
mediation of cell phenomena, namely: ( 1) nucleotide concentrations must correlate with
observed cell function; (2) addition of exogenous nucleotides must affect in a predictable
manner the cell response; (3) inhibitors of cyclic nucleotide metabolism must similarly affect
the phenomenon; and (4) the cyclic nucleotide synthesis and degradation enzymes should
also be modulated during the phenomenon. This study has satisfied the first three but has not
tested the fourth criterion. Thus, the four enzyme activities or their rates of synthesis in
chemostat populations were the direct result of the concentrations of the cyclic nucleotides in
the organisms. While these two cyclic nucleotides have been implicated in the synthesis of the
enzymes, each enzyme shows a different relationship between enzyme activity and cyclic
nucleotide concentration (Table 1). Although 60-99 % of the variability in enzyme activity
could be explained by fluctuations in cyclic nucleotide concentrations, 1-40 % of the
variability must be due to other factors, such as other small molecules. Nevertheless, the two
cyclic nucleotides play important roles in governing the levels of four distinct constitutive
enzymes in E . coli and logically might also play roles in general metabolism of this organism.
The precise steps in enzyme synthesis that are controlled by cyclic AMP and cyclic GMP
cannot be determined from these data.
With these data, it is possible to construct a working hypothesis, which I term the
‘permutation hypothesis’, to relate cyclic nucleotide concentration to cell metabolism. In
bacteria, there are blocks of constitutive enzymes, which may or may not represent metabolic
pathways. The synthesis of these blocks of enzymes is controlled at the transcriptional level
by the concentration of at least two small molecules, cyclic AMP and cyclic GMP. For
instance, synthesis of one block might be stimulated by low cyclic AMP and high cyclic GMP
concentrations and repressed by the opposite. An example of an enzyme such as this would
be phosphodiesterase. On the other hand, the synthesis of another block might be stimulated
by both high cyclic AMP and cyclic GMP concentrations. The example in this system would
be glucose-6-phosphate dehydrogenase. The synthesis of another enzyme might be relatively
unaffected by cyclic AMP yet be stimulated by cyclic GMP (e.g. isocitrate dehydrogenase).
With two nucleotides, e.g. cyclic AMP and cyclic GMP, and three possible effects, a positive,
a negative and a neutral, nine permutations or blocks of enzymes under separate control are
possible. Most probably, the cell contains more than nine blocks of enzymes. By introducing
a third regulator, the number would increase to 27; a fourth would increase it to 8 1. The other
regulators could be other cyclic nucleotides such as cyclic TMP, cyclic IMP, cyclic UMP,
cyclic CMP, or highly phosphorylated compounds such as guanosine tetraphosphate or
pentaphosphate (Gallant, 1979; Silverman & Atherley, 1979) or catabolite modulator factor
(Dessein et al., 1978). Thus these small molecules could play roles in control of enzyme
synthesis with other, cruder, control being accomplished by intermediates of metabolism or
inducers, e.g. lactose for the lac operon (Pastan & Adhya, 1976; Pastan & Perlman, 1967;
Peterkofsky, 1976; Rickenberg, 1974). Once the absolute concentration of enzyme has been
defined, the organisms’ fine control mechanisms, e.g. allostery, could be used to produce in
vivo activity.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
Cyclic AMP and cyclic GMP in E. coli
711
The data and the hypothesis presented in this paper pertain to this population of E. coli
when growing at steady state in either glycerol- or ammonium-limited chemostat culture.
Several questions arise, namely: is the hypothesis relevant (1) for other nutrient limitations
such as sulphate, potassium or magnesium, (2) for growth conditions where the carbon
source is changed, for example to glucose or lactose, and (3) for other strains of bacteria? At
present no data are available to answer any of these questions. However, enzyme activities
are modulated in bacteria grown under nutrient limitations other than carbon or ammonium
(see Matin, 198 1) but no data are available on the effects on cyclic nucleotide concentrations.
When bacteria are transferred from one carbon source to another, large changes in cyclic
nucleotide, particularly cyclic AMP, concentration are observed for batch-grown (Gonzalez
BE Peterkofsky, 1975) and chemostat-grown (Wright et al., 1979; this paper) cultures. Under
conditions when one cyclic nucleotide concentration is abruptly changed, the hypothesis
would require the concentration of the other effectors to be modulated to compensate;
Gonzalez & Peterkofsky (1975) have observed this modulation of cyclic AMP and cyclic
GMP in E. coli during carbon source change. While this hypothesis was developed to
describe the response of one particular strain of E. coli grown in chemostat culture, the study
of the response of other strains, for instance those defective in cyclic nucleotide metabolism,
should be fruitful. While bacterial physiology is rigorously controlled by the constant
environment attained at steady state in a chemostat culture, batch-culture grown organisms
are in an ever-changing environment whizh directly alters their physiology (Calcott, 198 1 b).
Thus interpretation of the relationship between enzyme activities and cyclic nucleotide
concentrations might be extremely difficult in batch-grown organisms.
When the concentration of carbon source was very high (ammonium-limiting conditions)
the cyclic AMP concentration was very low. Under these growth conditions, growth rate does
not influence the concentration of carbon source in the growth vessel (Calcott, 1981 a) and
did not affect the cyclic AMP concentrations in the organism. However, under glycerol
limitation, the concentration of carbon source in the growth vessel would be much lower than
that in ammonium-limited cultures and was also growth rate dependent. When organisms
grew in chemostat at slow rates, the concentration of carbon source would be low while the
cyclic AMP concentration was high. As growth rate increased, the concentration of carbon
source would increase and cyclic AMP concentration decreased. Not only was cyclic AMP
concentration apparently governed by sugar concentration, but also by the nature of the
sugar. Glucose-limited cultures contained lower concentrations of the nucleotide than
glycerol-limited cultures, with lactose-limited cultures containing intermediate concentrations.
This correlated with the relative efficiency of the sugars as catabolite repressors, with glucose
being the most effective, glycerol the least and lactose intermediate (Pastan & Adhya, 1976;
Pastan & Perlman, 1967; Peterkofsky, 1976; Rickenberg, 1974). With the data presented, it
is impossible to predict whether cyclic AMP concentrations were controlled by adenylate
cyclase activity as proposed by Peterkofsky and colleagues (Peterkofsky, 1976, 1977;
Peterkofsky & Gazdar, 1979), by the phosphodiesterases (Buettner et al., 1973; Calcott &
Calvert, 1981) and/or by excretion of the nucleotide (Goldenbaum & Hall, 1979). From the
data presented, it is apparent that the control of cyclic GMP concentrations in E. coli was
more complex than that of cyclic AMP, with both nutrient concentrations and limitations, as
well as growth rates, playing roles.
This research was supported by a research grant from the U.S. Army Research Office (Grant no.
DRXRO-CB- 15525-L).I acknowledge excellent technical assistance of Rita S. Petty and Donna L. Williams in
parts of this study and the aid of Dr W. J. Freeland and Dr A. Peterkofsky for critically reviewing the manuscript.
REFERENCES
AONO, R., YAMASAKI,
M. & TAMURA,G. (1978).
Changes in composition of envelope proteins in
adenylate cyclase and cyclic AMP receptor defi-
cient mutants of Escherichia coli. Journal of
Bacteriology 136,812-8 14.
BLACK,R. A., HOBSON,A. C. & ADLER,J. (1980).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27
712
P. H. C A L C O T T
Involvement of cyclic GMP in intracellular signalling in the chemotactic response of Escherichia coli.
Proceedings of the National Academy of Sciences of
the United States of America 71,3879-3883.
BOTSFORD,
J. L. & DANLEY,D. E. (1975). Induction of
tryptophanase and accumulation of cyclic AMP in
Escherichia coli grown in continuous culture.
Abstracts of the Annual Meeting of the American
Societyfor Microbiology, 123.
BUETTNER,M. J., Sprrz, E. & RICKENBERG,
H. V.
(1973). Cyclic adenosine 3 ’ 3 ’ monophosphate in
Escherichia coli. Journal of Bacteriology 114, 10681073.
CALCOIT,P. H. (1981~).Transient loss of plasmidmediated mercuric ion resistance after stress in
Pseudomonas aeruginosa. Applied and Environmental Microbiology 41, 1348-1354.
CALCOIT,P. H. (1981 b). Continuous culture: where it
came from and where it’s going. In Continuous
Culture of Cells, vol. 1, pp. 1-11. Edited by P. H.
Calcott. Boca Raton, Florida: CRC Press.
CALCOTT,P. H. & CALVERT,T. J. (1981). Characterization of 3’,5’-cyclic AMP phosphodiesterase
in Klebsiella aerogenes and its role in substrateaccelerated death. Journal of General Microbiology
122,313-321.
CALCOIT,P. H. & MACLEOD,R. A. (1974). Survival
of Escherichia coli from freeze-thaw: influence of
nutritional status and growth rate. Canadian Journal of Microbiology 20,67 1-68 1.
CHOCK,P. B., RHEE,S. G. & STADTMAN,
E. R. (1980).
Interconvertible enzyme cascades in cellular
regulation. Annual Review of Biochemistry 49,
8 13-843.
COOK, W. R., KALB, V. F., PEACE, A. A. &
BERNLOHR,R. W. (1980). Is cyclic guanosine
3’3‘-monophosphate a cell cycle regulator? Journal
of Bacteriology 141, 145&1453.
DESSEIN,A., TILLIER,F. & ULLMANN,A. (1978).
Catabolite modulator factor: physiological properties and in vivo effects. Molecular and General
Genetics 162, 89-94.
DIETZLER,D. N., LECKIE,M. P., STERNHEIM,
W.L.,
TAXIMAN,T. L., UNGAR,J. M. & PORTER,S. E.
(1977). Evidence for the regulation of bacterial
glycogen synthesis by cyclic AMP. Biochemical and
Biophysical Research Communications 77, 14681477.
DREW,S. W. & DEMAIN,A. L. (1977). Effect of
primary metabolites on secondary metabolism.
Annual Review of Microbiology 31,343-356.
GALLANT,J. A. (1979). Stringent control in E. coli.
Annual Review of Genetics 13,393-415.
GERSCH,D., ROMER, W., BOCKER,H. & THRUM,H.
(1978). Variations in cyclic adenosine 3’,5’ monophosphate and cyclic guanosine 3’,5’ monophosphate
in antibiotic producing strains of Streptomyces
hygroscopicus. FEMS Microbiology Letters 3,
3941.
GOLDBERG,N. D., HADDOX,M. K., NICOL, S. E.,
GLASS,D. B., SANFORD,C. H., KUEHL,F. A. &
ESTENSEN,R. (1975). Biologic regulation through
opposing influences of cyclic GMP and cyclic AMP:
the Yin-Yang Hypothesis. Advances in Cyclic
Nucleotide Research 5,307-330.
GOLDENBAUM,
P. E. & HALL,G. A. (1979). Transport
of cyclic adenosine 3’3’ monophosphate across
Escherichia coli vesicle membranes. Journal of
Bacteriology 140,459-467.
GONZALEZ,
J. E. & PETERKOFSKY,
A. (1975). Diverse
directional changes in cGMP relative to CAMP in
E. coli. Biochemical and Biophysical Research
Communications 67,190-197.
HARMAN,J. G. & BOTSFORD,
J. L. (1979). Synthesis
of adenosine 3’,5’-cyclic monophosphate in
Salmonella typhimurium growing in continuous
culture. Journal of General Microbiology 110,
243-246.
MATIN,A. (1 98 1). Microbial regulatory mechanisms
as studied in continuous culture. In Continuous
Culture of Cells, vol. 11, pp. 69-97. Edited by P. H.
Calcott. Boca Raton, Florida: CRC Press.
NIERLICH,D. P. (1978). Regulation of bacterial
growth, RNA and protein synthesis. Annual Review
of Microbiology 32,393-432.
PASTAN,I. & ADHYA,S. (1976). Cyclic adenosine
5’-monophosphate in Escherichia coli. Bacteriological Reviews 40,527-55 1.
PASTAN,I. & PERLMAN,D. (1967). Cyclic AMP in
metabolism. Nature, London 229,5-7.
PETERKOFSKY,
A. (1976). Cyclic nucleotides in bacteria. Advances in Cyclic Nucleotide Research 7,
1-48.
PETERKOFSKY,
A. (1977). Regulation of Escherichia
coli adenylate cyclase by phosphorylation-dephosphorylation. Trends in Biochemical Sciences 2,
12-14.
PETERKOFSKY,
A. & GAZDAR,C. (1979). Escherichia
coli adenylate cyclase complex: regulation by the
proton electrochemical gradient. Proceedings of the
National Academy of Sciences of the United States
ofAmerica 76,1099-1 103.
POSTGATE,
J. R. & HUNTER,J. R. (1962). The survival
of starved bacteria. Journal of General Microbiology 29,233-263.
RICKENBERG,H. V. (1974). Cyclic AMP in procaryotes. Annual Review of Microbiology 28,
353-369.
SILVERMAN,
R. H. & ATHERLY,A. G. (1979). The
search for guanosine tetraphosphate (ppGpp) and
other unusual nucleotides in eukaryotes. Microbiological Reviews 43, 2 7 4 1.
VESELY,D. L. (1979). Phencyclidine stimulates guanylate cyclase activity. Biochemical and Biophysical
Research Communications 88,1244-1 248.
WRIGHT, F. L., MILNE, D. P. & KNOWLES,C. J.
(1979). The regulatory effects of growth rate and
cyclic AMP levels on carbon catabolism and
respiration in Escherichia coli K 12. Biochimica et
biophysica acta 583,73-80.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 17:39:27