Journal of General Microbiology (1982), 128,49-59.
Printed in Great Britain
The Role of Oxygen in the Regulation of Glucose Metabolism, Transport
and the Tticarboxylic Acid Cycle in Pseudomonas aeruginosa
By COLIN G. MITCHELL? AND EDWIN A. DAWES*
Department o f Biochemistry, University o f Hull, Hull HU6 7RX, U.K.
(Received 21 May 1981)
The effect of dissolved oxygen concentration on the metabolism of glucose in Pseudomonas
aeruginosa was studied with chemostat cultures using both single-step and gradual transitions
from either ammonium or glucose limitation to oxygen limitation and studying transient and
steady states. The pathway of glucose metabolism was regulated by the availability of
oxygen. The organism responded to oxygen limitation by adjusting its metabolism of glucose
from the extracellular direct oxidative pathway, which produces gluconate and 2-
oxogluconate, to the intracellular phosphorylative route. This change was a consequence of
decreased activities of glucose dehydrogenase and gluconate dehydrogenase and of the
transport systems for gluconate and 2-oxogluconate, and an increased activity of glucose
transport, while relatively high activities of hexokinase and glucose-6-phosphate dehydro-
genase were maintained. Citrate synthase, isocitrate dehydrogenase and malate dehydro-
genase activities responded to changes in dissolved oxygen concentration rather than to
changes i n the glucose or ammonium concentrations. The effect of oxygen limitation on the
0x0-acid dehydrogenases and aconitase was probably due, wholly or in part, to repression by
glucose consequent upon the increase in residual glucose concentration. Succinate
dehydrogenase was repressed by an increase in ammonium concentration under an oxygen
Studies on the effects of dissolved oxygen concentration on the enzymic activities and
metabolic versatility of obligate aerobes are relatively meagre compared with those on
facultative anaerobes. This is understandable in view of the diversity of fermentation patterns
encountered with the latter organisms on their transition from aerobic to anaerobic conditions
(for recent reviews, see Harrison, 1976; Stouthamer, 1978; Jones, 1979). However, in their
natural habitat obligate aerobes encounter a wide range of oxygen concentrations.
Nitrogen-fixing azotobacters exhibit the phenomenon of respiratory protection; in these
organisms high oxygen concentrations are toxic, inhibiting nitrogenase (Parker, 1954;
Phillips & Johnson, 1961; Dalton & Postgate, 1969). Studies on the effect of oxygen on the
enzymic complement of Azotobacter chroococcum and Azotobacter vinelandii have been
carried out (Drozd & Postgate, 1970; Haaker & Veeger, 1976). We have previously shown
that the energy-reserve polymer poly-/I-hydroxybutyrate (PHB) accumulates in Azotobacter
beijerinckii when cultures become oxygen-limited; the reductive stage of polymer synthesis
serves as an alternative electron acceptor when oxygen is no longer so readily available, and
thus permits the organism to grow under such conditions (Senior et al., 1972; Senior &
Dawes, 1973; Ward et al., 1977). The activities of key enzymes of PHB metabolism and of
certain enzymes of the tricarboxylic acid cycle responded to changes in oxygen concentration
Present address: Department of Biochemistry, University of Bath. Claverton Down, Bath BA2 7AY. U.K.
0022-1287/82/0000-9962 $02.00 O 1982 SGM
C . G . MITCHELL AND E. A. DAWES
(Jackson & Dawes, 1976) but the Entner-Doudoroff enzymes were not affected (Stephenson
et al., 1978; Carter & Dawes, 1979).
The observed regulation of tricarboxylic acid cycle activity by oxygen concentration in A.
beijerinckii posed the question of whether this behaviour was representative of obligate
aerobes in general or whether it was a manifestation of respiratory protection and therefore
characteristic only of nitrogen-fixing organisms. As we have previously studied the aerobe
Pseudomonas aeruginosa, which is a denitrifier that utilizes glucose via the Entner-
Doudoroff and tricarboxylic acid cycle pathways, we chose this organism for a comparative
study of oxygen effects. Pseudomonas aeruginosa also metabolizes glucose via an
extracellular direct oxidative pathway to gluconate and 2-oxogluconate (Ng & Dawes, 1973;
Midgley & Dawes, 1973; Roberts et al., 1973; Whiting et al., 1976a, b). Transport of
glucose, gluconate and 2-oxogluconate occurs by independently regulated systems (Whiting
et al., 1976 a) and the extracellular oxidative enzymes and associated transport systems are
repressed when the organism is transferred from ammonium to glucose limitation (Whiting et
al., 1976b). Hunt & Phibbs (1977) observed that the extracellular route was also repressed
when P. aeruginosa was grown anaerobically with nitrate as electron acceptor in batch
culture. It was of interest to examine the effect of varying oxygen concentration on the
pathways of glucose metabolism in P. aeruginosa under the controlled conditions of the
chemostat. We have thus examined the effect of transitions from animonium or glucose
limitation to oxygen limitation on various enzymes of glucose metabolism and the
tricarboxylic acid cycle, and also on the transport systems for glucose, gluconate and
Organism and growth. Pseudomonas aeruginosa PA0 1 was kindly provided by Professor B. W. Holloway.
Routine maintenance, batch growth of the organism and harvesting procedures were as previously described (Ng
& Dawes, 1973; Midgley & Dawes, 1973). For chemostat inocula, the organism was subcultured at least ten times
in the appropriate medium. All cultures were grown at 37 "C.
The chemostat vessel (2.5 1 volume) was built in this Department. It was fitted with automatic pH control
(E.I.L. Instruments, Richmond, Surrey; with pH electrodes supplied by Activion, Kinglassie, Fife), temperature
control (Fielden Electronics, Manchester) and oxygen control (Leeds & Northrup, Birmingham; Precision
Products and Controls, Tulsa, U.S.A.). The oxygen electrodes were made in this Department. COz was measured
with an infrared analyser (Mine Safety Appliances, Glasgow) and oxygen with a paramagnetic oxygen analyser
(Servomex OA 13 7; Servomex Controls, Crowborough, Sussex). Continuous readout of gas composition was
obtained with a Kent chart recorder (George Kent, Luton, Beds). The total gas flow was 11 min-' and the re-
quired dissolved oxygen tension (d.0.t.) was secured by adjusting the proportion of oxygen in the inflowing oxygen/
nitrogen gas mixture. Oxygen limitation (undetectable d.0.t.) occurred at 4.5 % (v/v) oxygen in the inflowing gas.
Medium for continuous cultivation was prepared in 40 1 batches which were sterilized by filtration through a
Sartorius filter (142 mm diam., pore size 0.25 pm) at 292 kPa. The medium contained (per litre): KH,PO,, 5.4 g;
nitrilotriacetic acid, 0.286 g; (NH,),SO,, 0.9 g (for ammonium-limited growth) or 1-8 g (for glucose- and
oxygen-limited growth); trace metal solutions 1 (5 ml), 2 (5.25 ml) and 3 (0.1 ml) (Ng & Dawes, 1973); glucose,
4.0 g (for carbon-limited growth) or 8.11 g (for ammonium-limited growth). For most of the single-step transitions
from ammonium- or glucose-limited growth to oxygen limitation, the inflowing medium was simultaneously
changed to furnish excess ammonium or glucose. Initial studies were carried out to ascertain the effect of such
additions of ammonium or glucose, as noted in the Results. In the gradual transition experiments with
ammonium-limited cultures, the inflowing medium was changed to provide excess ammonium as soon as
ammonium ions could be detected in the medium supernatant; this occurred at approximately 9% oxygen in the
inflowing gas. In similar experiments with glucose-limited Cultures, glucose was detected in the supernatant at
between 6 and 8 % oxygen in the inflowing gas and the medium was changed at 4 % oxygen to ensure an excess of
Six to ten vessel volumes were allowed to pass through the fermenter between each steady state before sampling
except in experiments to study transient responses.
Transport studies. These were performed by the methods of Midgley & D8wes (1973) using the substrate
concentrations and specific radioactivities specified by Whiting et al. (1976 6). The radioactivity was assayed as
described by Midgley & Dawes (1973).
Eflect o f oxygen on glucose metabolism
Enzyme assays. Bacterial extracts were prepared by the methods of Ng & Dawes (1973). All enzyme assays
were performed in cuvettes of 1 cm light-path at 37 "C with a Pye Unicam SP1800 recording spectrophotometer
under previously determined optimum conditions. Specific activities are recorded as pmol h-' (mg protein)-' and
are the means of at least two assays which did not differ by more than 5 %.
The enzymes of glucose and 2-oxogluconate metabolism [glucose dehydrogenase (EC 1. I . 1.47), gluconate
dehydrogenase (EC 1 . 1 .99.3), glucose-6-phosphate dehydrogenase (EC 1.1.1 .49), hexokinase (EC 2.7.1 . I )
and gluconate kinase (EC 2.7.1 . 12)] were assayed under the conditions described by Ng dc Dawes (1973) except
that glucose dehydrogenase was assayed in the presence of phenazine methosulphate (2 mM). The combined
activity of 2-oxogluconate kinase and 2-0x0-6-phosphogluconate reductase (referred to as 2-oxogluconate
enzymes) was assayed by the method of Whiting et al. (1976b). Isocitrate dehydrogenase (EC 184.108.40.206) and
aconitase (EC 4.2.1 .3) were assayed as described by Ng & Dawes (1973); fumarate hydratase (EC 220.127.116.11) by
the method of Racker (1950); citrate synthase (EC 18.104.22.168) according to Weitzman & Dunmore (1969); malate
dehydrogenase (EC 22.214.171.124) by the method of von Tigerstrom & Campbell (1966) except that 5 mM-MgC1, was
included and the pH was 8.6; pyruvate dehydrogenase (EC 1 .2.4.1) and 2-oxoglutarate dehydrogenase (EC
1 .2.4.2) by the method of von Tigerstrom dc Campbell (1966) except that Mg2+ was omitted in the latter assay;
succinate dehydrogenase (EC 1 .3.99.1) by the method of Veeger et al. (1969); and NADH oxidase according to
Jackson & Dawes (1976). Both pyruvate and 2-oxoglutarate dehydrogenases were assayed under anaerobic
conditions in cuvettes fitted with side-arms, on account of the high NADH oxidase activities of bacterial extracts.
Cuvettes were flushed with oxygen-free nitrogen for 5 min and an airtight seal was achieved with Subaseal caps.
Nitrogen was flushed for a further 5 min via syringe needles penetrating the Subaseals. Cuvettes were incubated for
10 min in the spectrophotometer at 37 "C and the reaction was started by tipping the substrate from the side-arm.
In control experiments for endogenous NAD reduction, water replaced the substrate. Combined Entner-
Doudoroff enzymes [6-phosphogluconate dehydratase (.EC 4.2.1 .12) and 2-keto-3-deoxy-6-phosphogluconate
aldolase (EC 4.1 .2.14)1 were jointly assayed by measuring the 6-phosphogluconate-dependent production of
pyruvate coupled to the oxidation of NADH at 340 nm using an excess of commercial lactate dehydrogenase.
Cuvettes contained (total volume 3 ml): 0.1 M-triethanolamine/HCl buffer, pH 7.8, 2 ml; 80 mM-dithiothreitol,
0.2 ml; 10 mM-NADH, 0-2 ml; 0.2 ~-6-phosphogluconate, 0.1 ml; lactate dehydrogenase (2 mg ml-'), 0-2 ml;
cell extract, 100 pl.
Analyses. Protein was estimated by the methods of Gornall et al. (1949) and Lowry. Glucose was determined
with a glucose oxidase kit (Boehringer) and 2-oxogluconate by the method of Lanning & Cohen (195 1). Gluconate
was determined by coupling gluconate kinase with gluconate-6-phosphate dehydrogenase according to the
Boehringer Handbook. Ammonia was determined by the method of Chaney & Marbach (1962).
Chemicals. [U-'4ClGlucose and [U-14C]gluconate were obtained from Amersham. 2-Oxo[ ''Clgluconate was
prepared by the method of Whiting et al. (1976b). The following chemicals were obtained from Sigma:
DL-isocitrate, phenazine methosulphate, DL-dithiothreitol, L-malate, 2-oxoglutarate (monosodium and mono-
potassium), 5,5'-dithiobis-(2-nitrobenzoic acid), 6-phosphogluconate, NAD (reduced and oxidized forms),
NADPH, coenzyme A, S-acetyl-coenzyme A and glucose 6-phosphate. NADP was obtained from Boehringer.
L-Cysteine. HCl and oxaloacetic acid were from BDH. All other chemicals were of AnalaR standard or the highest
Enzymes. All commercial enzymes were obtained from Sigma.
RESULTS A N D DISCUSSION
Two types of transition to oxygen limitation were used. The first, referred to as a single-step
transition, subjected ammonium- or glucose-limited cultures to a sudden decrease from a high
to an undetectable dissolved oxygen tension (d.0.t.). In the second, the d.0.t. was lowered in
stages, and is termed a gradual transition.
Eflects o f oxygen concentration: single-step transitions
The transition from ammonium limitation to oxygen limitation resulted in decreases in the
specific activities of glucose, gluconate and glucose-6-phosphate dehydrogenases, gluconate
kinase and the 2-oxogluconate enzymes, while hexokinase increased in activity (Table 1).
With the exception of gluconate kinase and the 2-oxogluconate enzymes, new steady-state
activities were reached within 40 h of the transition, as were the residual nutrient
concentrations. The activity of glucose dehydrogenase decreased 20 to 30 % faster than that
of gluconate dehydrogenase.
Table 1. Eflect of single-step transition to a differed limitation on steady-state activities of enzymes of glucose metabolism and the
tricarboxylic acid cycle and on residual concentrations of nutrients
In the transition from ammonium to glucose limitation, 15 vessel volumes were allowed to pass through the fermenter before sampling. Ammonium-
and glucose-limited cultures were grown at 80% of air saturation (23% oxygen in the inflowing gas) while oxygen limitation (undetectable d.0.t.) was
secured with 2.5 % oxygen in the inflowing gas; the total gas flow rate (oxygenhitrogen) was 11 min-'. In the transition from ammonium or glucose
limitation to oxygen limitation, the ammonium or glucose concentration in the inflowing medium was increased as described in Methods to ensure that
only oxygen was growth-limiting. The values for oxygen limitation refer to 120 h after the transition unless otherwise indicated. Each value in the table
represents the average of duplicate assays, which did not vary by more than 5 %, from one particular steady state.
Enzyme specific activity I pmol h-' (mg protein)-' I
Ammonium c_j_ Oxygen
( D = 0.36 hk')
(D = 0.35 h-')
( D = 0.35 h-')
19.5 19.2 15.9
Succinate dehydrogenase Fumarate hydratase
Bacterial dry weight (mg ml-*)
Residual ammonium (mM)
Residual glucose (mM)
Residual gluconate (mM) Residual 2-oxogluconate (mM)
-, Not determined; ND, not detected.
* Oscillations; no steady state achieved in oxygen limitation.
Damped oscillations for 180 h before attaining steady state.
$ No change in activity until 120 h; steady state not attained in oxygen limitation.
Eflect o f oxygen on glucose metabolism
Time after transition (h)
Fig. 1. Effect of single-step transition to oxygen limitation (undetectable d.0.t.) df an ammonium-limited
culture on the activities of gluconate kinase (0) and the 2-oxogluconate enzymes (0). At the time
indicated by the arrow, the oxygen supply rate in the oxygenhitrogen mixture was decreased from 230
to 25 ml min-' (total gas flow 1 I min-') and the (NH,),SO, concentration in the inflowing medium was
increased from 0.9 to 1-8 g 1-I: D = 0.35 h-'. Enzyme specific activities are expressed as pmol h-' (mg
Table 2. Effect o f a single-step transition to oxygen limitation o f an ammonium-limited
culture on the steady-state activities o f the trcrnsport systems for glucose, gluconate and
2 -0xoglu co nu t e
The conditions were as described in Fig. 1. The activities for oxygen limitation are steady-state values
obtained 110 h after the transition; activities oscillated during the initial 100 h period.
Activity [ p o l min-' (g dry wt)-'I
limitation Transport system
A similar response for hexokinase and glucose-6-phosphate dehydrogenase was observed
in the transition from glucose limitation to oxygen limitation (Table 1). The Entner-
Doudoroff enzymes were unchanged in transitions from ammonium or glucose limitation to
oxygen limitation. In the latter transition, gluconate dehydrogenase was unchanged while
glucose dehydrogenase decreased slightly in activity (Table 1). Gluconate, which is known to
induce gluconate dehydrogenase (Whiting et al., 1976a), was not detected in the medium
during the transition which probably accounts for the invariance of this enzyme. This
interpretation is supported by the anaerobic batch culture studies of Hunt & Phibbs (1977).
In transitions from ammonium to oxygen limitation the 2-oxogluconate enzymes increased
markedly in activity up to 80 h, before falling to a new steady-state level (Fig. 1). A similar
response was observed for an initially glucose-limited culture. The reasons for this peak are
not apparent but it may reflect transient changes in intracellular metabolite concentrations as
a consequence of abrupt oxygen limitation. This response was observed on each occasion the
experiment was performed, although the peak height did vary.
Transition from ammonium to oxygen limitation also resulted in an increase in the initial
rate of glucose transport and decreases in those of gluconate and 2-oxogluconate (Table 2).
In contrast, transition from glucose to oxygen limitation gave no change in these initial rates.
This correlates with the virtual absence from the culture of gluconate and 2-oxogluconate
(Table l), compounds which induce their respective transport systems, while gluconate