Biochimica et Biophysica Acta 1777 (2008) 763–771

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / b b a b i o

Review

Cellular energetic metabolism in sepsis: The need for a systems approach

Jane E. Carré ⁎, Mervyn Singer
Wolfson Institute for Biomedical Research, University College London, Gower Street, London WC1E 6BT, UK

A R T I C L E I N F O A B S T R A C T

Article history: Sepsis is a complex pathophysiological disorder arising from a systemic inflammatory response to infection.
Received 4 February 2008 Patients are clinically classified according to the presence of signs of inflammation alone, multiple organ
Received in revised form 4 April 2008 failure (MOF), or organ failure plus hypotension (septic shock). The organ damage that occurs in MOF is not a
Accepted 5 April 2008
direct effect of the pathogen itself, but rather of the dysregulated inflammatory response of the patient.
Available online 23 April 2008
Although mechanisms underlying MOF are not completely understood, a disruption in cellular energetic
Keywords:
metabolism is increasingly implicated. In this review, we describe how various factors affecting cellular ATP
Mitochondria supply and demand appear to be altered in sepsis, and how these vary through the timecourse. We will
Sepsis emphasise the need for an integrated systems approach to determine the relative importance of these factors
Inflammation in both the failure and recovery of different organs. A modular framework is proposed that can be used to
Multiple organ failure assess the control hierarchy of cellular energetics in this complex pathophysiological condition.
Mitochondrial dysfunction © 2008 Elsevier B.V. All rights reserved.
Top–down metabolic control
Cellular energetic metabolism

1. Introduction
Sepsis describes a combination of clinical manifestations of sys-
temic inflammation specifically related to an infectious insult from
agents such as bacteria, pathogenic fungi, yeasts and viruses. Other
non-infectious insults such as trauma, burn injury or haemorrhage can
cause a similar response, suggesting a common final pathway. Clinical
presentations include pyrexia, variable changes in cardiac output (the
product of heart rate and ejected ventricular stroke volume),
hypotension that is non-responsive to vasopressor agents, immuno-
suppression, and evidence of failure of one or more organ systems
including haemodynamic, respiratory, neurological, renal and hepatic
systems. The severity of illness depends on the extent to which these
disturbances occur and has led to the classification of sepsis into sepsis
syndrome, severe sepsis and septic shock ([1], Table 1).
The course of sepsis is variable and often unpredictable, and
depends on factors such as age and pre-existing morbidities. Genetic
polymorphisms also play a part in determining susceptibility to sepsis
and the severity of the inflammatory response. Mortality rates are
decreasing due largely to improved supportive care, however, the
incidence of sepsis is rising [2,3]. Between 30–50% of critically ill
patients in intensive care units (ICUs) develop sepsis [2,4], with an
average early mortality risk rising from 4% after 24 h to approximately
30% by day 28 [5]. Mortality increases with the number of failed organs,
with approximately one-third of patients with single organ failure
⁎ Corresponding author.
E-mail address: j.carre@ucl.ac.uk (J.E. Carré).
dying, compared to 90% of those with four or more failures [6,7].
Patients who survive their septic episode often have long-term
morbidities, including weakness, fatigue, neuropathies and psycholo-
gical problems [2]. To what extent these problems are due to the
disease process itself and how much to sequelae and complications
related to the treatment and support given in intensive care is cur-
rently a major topic of investigation.
1.1. Pathophysiology of sepsis
Several comprehensive reviews of systemic immune [8,9], endo-
crine [10,11] and metabolic [12–14] responses in sepsis are available,
and only a general overview will be given here (Fig. 1). The
inflammatory response to infection is initiated by the binding of
microbial products such as the bacterial cell wall components
lipopolysaccharide (Gram-negative) and peptidoglycan (Gram-posi-
tive) to Toll-like and other recognition receptors on immune and
endothelial cells. Subsequent activation of nuclear transcription factor
κB (NF-κB) by its translocation to the nucleus leads to expression of
significant circulating levels of pro-inflammatory cytokines including
tumor necrosis factor-α (TNF-α) and interleukin 1-β (IL-1β). Together
with activation of complement pathways, these proinflammatory
cytokines stimulate a pro-coagulatory state. Cytokine stimulation
of circulating and resident immune cells, endothelial and some
epithelial cells results in increased production of reactive oxygen
intermediates (ROS) such as superoxide, and nitric oxide (NO).
Excessive NO production is considered a major cause of the vaso-
relaxation and decreased responsiveness to vasopressor agents

0005-2728/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbabio.2008.04.024
764 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771

Table 1 “stress” hormones including cortisol, catecholamines, vasopressin, and

Definitions of systemic inflammation and sepsis
Condition Definition and clinical description
Systemic inflammatory The systemic inflammatory response to various clinical
response syndrome (SIRS) insults (infection, trauma, haemorhage).
Requires at least two of:
• core temperature b 36 °C or N 38 °C
• heart rate N 90 beats min− 1
• respiratory rate 20 breaths min− 1or PaCO2 4.3 kPa
• white blood cell count b4000 mm− 3 or N12 000 mm− 3 or
immature neutrophil content N 10%
Sepsis The systemic response to infection.
• SIRS criteria PLUS
• proved or suspected infection
Severe sepsis Severe sepsis associated with:
• organ dysfunction
• hypoperfusion or hypotension
Septic shock Sepsis associated with:
• hypotension that is unresponsive to administered
vasopressors (despite adequate fluid resuscitation)
• perfusion abnormalities
Summarised from [1].
(vascular hyporeactivity) underlying septic shock (reviewed in [15, 16].
Expression of adhesion molecules results in increased infiltration of
immune cells into tissues both local to, and distant from, the cause of
infection. In line with the pro-inflammatory response, a compensatory
anti-inflammatory response is also initiated to quench the flames of
inflammation. After an indeterminate and variable period of hours to
days, this may even shift the balance of the inflammatory milieu to an
immunosuppressed state characterised by high levels of anti-inflam-
matory cytokines such as IL-10, T-cell anergy, accelerated apoptosis of
circulating B-lymphocytes, delayed apoptosis of neutrophils, and
decreased phagocytic and bactericidal activity [9].
Besides the alterations in cytokine profiles, the acute phase
response to infection is accompanied by a massive release of various
both insulin and glucagon [10,11]. Together, these hormones act to
maintain oxygen delivery to the tissues (q.v.) and also result in
mobilisation of carbohydrate, fat and protein stores in muscle and liver
to fuel the synthesis of acute phase proteins. Resting energy ex-
penditure (REE), as measured by calorimetry, is increased during this
period, although the more severely affected patients do not increase
REE to the same extent [17]. Indeed, septic shock patients have REE
values similar to healthy people, often despite marked pyrexia. The
acute phase endocrine response is predictive of outcome with a
suboptimal or impaired response portending a poor prognosis [18].
During prolonged sepsis, circulating hormone levels return to
normal, or even sub-normal levels. The “low T3” syndrome is often
manifest with decreases in thyroxine (T4) and the more active tri-
iodothyronine (T3), and an increase in the inactive reverse T3 (rT3).
REE shows a significant rebound during the recovery phase [17].
1.2. Current treatments for sepsis
While many immunomodulatory therapies have been trialled,
none has equivocally demonstrated an overall reduction in mortality.
Even the use of activated protein C, the only immunomodulatory agent
with a current licensed indication for sepsis, is now being formally
readdressed. Most therapy is supportive rather than curative, propping
up the failing organs until they start functioning. These include anti-
biotics to remove the initiating trigger, fluid resuscitation to maintain
organ perfusion, vasopressor administration to maintain an adequate
blood pressure, and mechanical support of failing lungs (mechanical
ventilation) and kidneys (dialysis-like techniques). Despite these
technical advances, sepsis still remains a major killer worldwide.
1.3. Energetic failure in sepsis?
The concept of cellular energetic dysfunction in septic organ failure
can be traced back some 40 years, with observations of swollen and

Fig. 1. Systemic interactions in the pathophysiology of sepsis. NF-κB, nuclear factor κB. Reproduced from [8] with permission.
 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
distorted mitochondria in livers of mice injected with endotoxin [19].
Mitochondrial respiration, respiratory complex enzyme activities and
nucleotide levels have been studied in tissue homogenates, cells or
mitochondria isolated from a variety of animal or cell models to which
endotoxin or bacterial insults had been administered. Highly variable
changes in mitochondrial function are reported in models lasting up to
16 h, but a much more consistent finding of mitochondrial dysfunction
and ultrastructural damage is seen thereafter (reviewed in detail in
[20]). Several factors are likely to be responsible for this variability,
including the type and degree of insult, the animal, organ or cell
species selected, and the adequacy or otherwise of resuscitation.
Differences in tissue ATP content have been reported in muscle
biopsies from patients [21,22] and in different tissues from animal
models [23–25]. We reported depressed ATP levels in muscle biopsies
taken from critically ill patients who went on to die, while eventual
survivors demonstrated maintained/elevated ATP [21]. In combina-
tion with several reports of elevated, rather than depressed, tissue
oxygen tensions (tPO2), these findings of mitochondrial abnormalities
and changes in ATP led to the proposal that septic organ failure
represents a state of “tissue dysoxia” or “cytopathic hypoxia”, i.e. a
cellular inability to use oxygen rather than a lack of its availability [26].
Studies in septic patients widely report depletion of glutathione
and ascorbate in both tissues and blood [27,28]. We also found a
correlation between the severity of sepsis and muscle glutathione
levels in critically ill ICU patients [21]. Apart from the afore mentioned
link between ATP levels and subsequent death, we also reported other
prognostic features including depleted muscle levels of reduced
glutathione, impaired complex I activity, elevated nitrate/nitrite
(indicator of RNS) [21], increased EPR-detected radical species, and
decreased complex I iron–sulphur centres [29]. Plasma levels of lipid
peroxide products were elevated in patients with multi-organ failure
(MOF), particularly in those with renal failure [30]. Plasma from septic
patients induced a severity-linked oxidative stress when applied to
cells in vitro [31]]. Functional implications of such oxidative stress has
been inferred by the presence of mitochondrial damage and depletion
in patient samples of both leg muscle [21,22,28]and diaphragm [22].
1.4. Energetic failure or adaptive metabolic control? The need for a systems
approach
Together, such data as described above have led to the formulation
of the hypothesis that septic organ failure results from a depression
in mitochondrial function, with key effectors being cytokine-induced
oxidant and nitrosative species, and changes in hormone levels
[20,26,32]. To some, this represents a damage-induced energetic
failure of mitochondria. Based on the relatively rapid recovery of
organs and the lack of evidence of significant cell death in failed organs
[33], an alternative hypothesis proposes that the condition manifest as
“organ failure” actually represents an adaptive and functional meta-
bolic suppression, similar to aestivation or hibernation, that enables
long-term survival if the victim survives the initial insult [34].
It is inevitably problematic to obtain tissue samples from patients
with MOF to perform mechanistic studies, especially from deep “vital”
organs such as liver and kidney. It is thus not surprising that human
literature is limited to snapshots of cellular events. However, these
steady-state measurements reveal simply that a process has been
affected by an insult, and reveal little about whether affecting that
process at that time will have any influence on cellular behaviour. For
example, nitrosative damage to respiratory complex I is likely to be
detrimental to cellular energetics only at times where supply of ATP is
unable to meet cellular ATP demand (see [35], discussed below).
Therefore, determining the cause and effect of energetic disturbances
requires a systems approach, whereby the impact of sepsis on
energetic supply and demand is determined. We anticipate that by
employing the top–down metabolic control approach described
below, we will be able to determine to what extent the long-held
765
concept of mitochondrial dysfunction in sepsis impacts upon cellular
energetics.
2. Control of cellular energetics: ATP supply and demand
2.1. Top–down metabolic control analysis and the “modular kinetic” approach
Our understanding of the control of respiration (see [36]) has
benefited significantly from the development of metabolic control
analysis [37,38] and the subsequent extension of top–down or
modular control analysis [39–41]. These approaches have led to the
appreciation that control of cellular respiration is shared between
processes that provide ATP and those that consume it and that the
strength of control that one process has over respiration varies under
different conditions [42].
Compared to the bottom-up approach, which establishes the
control exerted by individual enzymes over flux through a pathway
[38,42], the top–down approach divides systems into blocks of
reactions, consisting of either individual enzymes or, usefully, groups
of enzyme and intermediate interactions (i.e. pathways). In a top–
down (or “modular-kinetic”) analysis, the responses of the steady-
state flux of consumer and producer pathways to changes in con-
centration of their linking intermediate(s) are measured. From these
responses (i.e. elasticities), the influence of a change in activity of these
producers or consumers on both overall steady state flux (i.e. flux
control) and intermediate concentration (i.e. concentration control)
can be calculated. A pathway having large elasticity towards an
intermediate has, in general, low flux control, and vice versa. The
overall control structure of a process can thus be described quan-
titatively. This will predict whether and, if so, how a system changes in
response to acute changes in intermediate or enzyme activity (for
example, resulting from hypoxia or allosteric modulations such as
phosphorylation). This modular-kinetic approach may be exploited
further to assess (1) where and to what extent a system is affected by
administration of effector molecules (e.g. hormones, kinases, NO) and
(2) how the control between two or more systems (e.g. cell types,
pathophysiological states) differs.
2.1.1. A modular kinetic approach to cellular energetics in sepsis
In Fig. 2, we outline a simple modular framework that can be used
to determine how sepsis can affect the control of cellular energetic
metabolism. The system is centred around ATP as an intermediate
with a block each for supply and demand. By modulating the activity
of one block and determining the effect on the activity of the other, a
description of the responses of ATP-producing and ATP-consuming
pathway fluxes to a change in ATP is obtained [39–41]. From this
kinetic description, the elasticities of supply and demand blocks
towards ATP can be calculated. Subsequently, these elasticities can be
used to calculate the control of each block over the overall steady-state
flux of ATP supply and demand, and also the control exerted by each
block over the concentration of ATP. This initial analysis reveals
broadly whether or not supply and/or demand pathways have been
affected by the pathophysiological insult. Once this is ascertained, the
relevant block can be further subdivided, and a more detailed modular
kinetic analysis (or, if more appropriate, a bottom–up analysis) can
then performed to identify more precisely the site(s) of modification.
The value of understanding how and where changes in the dis-
tribution of control of a pathophysiological system have occurred
becomes apparent when deciding where to target therapeutic efforts.
For example, if at a particular time-point, control of ATP supply
and demand is found to have shifted towards ATP-producing pathways
(e.g. oxidative phosphorylation), therapies aimed towards increasing
ATP production (e.g. substrate availability, mitochondrial functional
capacity) are more likely to be successful than during periods where
control lies with ATP-consuming pathways. In this respect, further
analysis of the effects of external modulators including hormones
766 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
Fig. 2. Modular separation of energetic metabolism around ATP for analysis of cellular
energetics in sepsis. Initial analysis of Supply/Demand flux and [ATP] at different
modulations of flux reveal whether Supply block and/or Demand block have been
altered by the septic insult. Subsequently, the relevant blocks can be sub-divided and
appropriate intermediates (e.g. O2, substrate, Δψ, Ca2+) and fluxes (arrows: e.g.,
glycolysis, substrate oxidation, phosphorylation, proton leak, individual demand
pathways) can be analysed to pin-point where control distribution of the system has
been affected. A comparison of control distribution for different time points in sepsis can
be performed for acute, established and recovery phases, in different tissues, or at
different severities of insult. The site of action of various effectors (hormones, NO/ROS,
mediators of mitochondrial biogenesis, temperature) can be determined. See [38–40] for
details of the modular kinetic approach to top–down metabolic control analysis.
[43,44] or even drugs [45] used in patient management (e.g. catecho-
lamines) can be determined for the initially characterised system.
2.2. Cell models of sepsis for systems approaches to energetic metabolism
The modular kinetic approach outlined would be suitable to study
ex vivo cells that have been freshly isolated from established animal
models of sepsis. In this respect, polymicrobial models of peritoneal
sepsis lasting several days have been developed for rodent and large
animal studies. Methods used for the induction of sepsis in these
models include caecal ligation and puncture (reviewed in [46]),
injection of faecal slurry (e.g. [23] or implantation of bacterially-
infected fibrin clots (e.g. [47]). The provision of adequate fluid
resuscitation is considered to result in more clinically-relevant long-
term models. Cells can be isolated from different organs at different
stages of disease progression and recovery, and analysed ex vivo.
Alternatively, in vitro models can be used where cultured cells are
subjected to a septic insult of varying dose or duration, followed by a
washout period to follow recovery. In this respect, incubation with
either LPS alone or in combination with cytokines has been used to
investigate cellular energetic aspects of sepsis (e.g. [48,49]). Co-culture
and Transwell culture of hepatocytes and Kupffer cells has been used to
investigate the mechanism of sepsis-induced down-regulation of
peptides involved in vasodilation [50]. A further in vitro approach
investigated the effect of plasma taken from control patients or those
with sepsis on the oxygen consumption rates of peripheral blood
mononuclear cells (PBMCs) isolated from septic or control patients [51].
The choice of system for a modular kinetic analysis will depend on
a balance between (1) the clinical relevance, but inherently variable
response, of animal models, and (2) the more easily-controlled, but
potentially less-directly translatable, in vitro cell model.
3. Cellular energetics in sepsis
The discussion below concentrates on aspects of cellular energetic
metabolism that, to date, have received most attention in sepsis: ATP
production and consumption, oxygen availability, and also mitochon-
drial functional capacity. However, consideration should also be given
to the processes directly affecting Δψ, Ca2 and ROS production, as
these factors play significant roles both in energy metabolism and in
mediating the mitochondrial permeability transition and cell death.
3.1. ATP production and consumption
3.1.1. Control of ATP supply and demand
The control distribution for oxidative phosphorylation has been
found to be similar for isolated liver mitochondria [52] and resting
hepatocytes [53], with approximately 50% of control over respiratory
rate lying with phosphorylating reactions (equivalent to ATP turn-
over), 30% with the electron transfer chain and 20% with proton
cycling (leak). Most (~ 80%) of the control over phosphorylation lay
with the phosphorylating system itself, indicating that cellular
energetics in hepatocytes are largely demand-driven. On the other
hand, control over phosphorylation in brain mitochondria was found
to lie largely with substrate oxidation reactions (~ 70%) [54].
The sensitivity of supply and demand pathways to small changes
in [ATP] (or ATP/ADP, energy charge or the phosphorylation potential
ΔGp) will reflect the elasticities of all the components in a given
block. For the ATP-supply pathways, this includes the sensitivities of
both oxidative phosporylation and glycolysis. Several studies have
shown an increase in the rate of lactate production upon stimulation
of the Na+/K+-ATPase, or a decrease in lactate production by inhibition
of this transporter [55], combined with a relative insensitivity of
oxygen consumption (e.g. [56]). This behaviour suggests that these
changes in flux are due to a higher elasticity of glycolysis, compared to
that of oxidative phosphorylation, towards ATP.
The ATP-demand block comprises those processes involved in
maintenance of ion gradients (Na+/K+-ATPase, Ca2+-ATPase), and
anabolic functions including macromolecule synthesis (DNA, RNA,
protein) and tissue-specific ATP-dependent processes (gluconeogen-
esis, ureagenesis). Atkinson [57] first proposed the concept of hierarchy
among different ATP-demanding processes where less essential
anabolic functions are more sensitive to changes in energetic status
compared to essential activities such as ion homeostasis. This concept
has been demonstrated in several experimental systems including
immune-activated thymocytes [58] and responses of hepatocytes to
hypoxia [59,60]. Whether the difference in elasticities of reaction blocks
to ATP is a direct allosteric effect of nucleotides on enzymes controlling
these processes or an indirect effect via allosteric regulation by, for
example, AMP-activated protein kinase (AMPK) [61] is, at present,
unclear. In some systems (e.g. muscle), changes in [ATP] are buffered by
activity of the creatine kinase system. This system is also involved in
effectively shuttling ATP between compartments and within the cytosol
[62].
3.1.2. ATP and supply-demand interactions in sepsis
As discussed earlier, a low muscle ATP content at study entry was
associated with mortality in a critically ill group of patients [21]. In a
follow-up study [unpublished data], there was also a significant
depression of phosphocreatine (PCr) content and, in survivors, a lower
PCr/ATP ratio. Similar findings were reported in a separate study of leg
muscle, although neither PCr nor ATP content was affected in inter-
costal (respiratory) muscle [22]. Measurement of leg muscle lactate by
microdialysis in a group of patients in early septic shock revealed a high
lactate content that decreased over the following 24 h. Parallel
microdialysis with an infusion of ouabain indicated that muscle lactate
production rate was sensitive to this Na+K+-ATPase inhibitor [63].
Although a comparison with control patients is unavailable, these data
suggest an enhanced glycolytic rate in muscle in early sepsis linked to
an increase in an ATP demanding process.
Together, these data suggest a stimulation of ATP turnover in leg
muscle during the early phase of sepsis. In an investigation of septic
muscle energetics over time, 31P NMR analysis of PCr, Pi and ATP in
 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
septic rats indicated that mitochondrial activity was enhanced early
in sepsis, but diminished later, although two separate models were
used in this comparison [25]. Time-dependent studies in patients
are lacking, however our preliminary (unpublished) data suggest that
PCr/ATP in surviving patients increases over time and, after two weeks
of illness, is equal to, or indeed greater than, control values.
3.1.3. Efficiency of ATP production in sepsis: proton leak
One aspect that remains to be resolved in sepsis is the extent to
which REE represents metabolism that is uncoupled from ATP
production. Under normal physiological conditions, it is estimated
that approximately 20% of the standard metabolic rate of oxygen
consumption is not coupled to ATP production, but is used to drive the
proton leak across the mitochondrial membrane [64]. The extent to
which this changes in sepsis is uncertain, but it has been argued that
metabolic inefficiency is a major contributor to pyrexia (see [64]). One
key hormone that affects metabolic efficiency is thyroid hormone,
which both increases the rate of respiration and the proton leak rate
[44,65]. The modulations in thyroid status at different stages of critical
illness may have a major impact on the efficiency of ATP production.
Thyroid hormone is significantly elevated during the acute phase of
sepsis, whilst sub-normal levels persist during established illness
(the low T3 syndrome described earlier).
Proton leak across the inner mitochondrial membrane to the
matrix can be modulated by the regulated engagement of protein-
dependent pathways, such as activation of uncoupling proteins (UCPs)
(see [66,67]), or by increased basal leak via the adenine nucleotide
translocase [68]. Although the physiological role and function of
the more novel members of the UCP family remains to be firmly
established, UCP2 activity of immune cells seems to be involved
inflammation [69,70].
In addition to proton leak, any changes in the stoichiometry of
respiratory proton pumps (redox slip) could contribute to decreased
efficiency of ATP production (see [71,72]), although it is uncertain as to
what extent this process occurs (patho)physiologically.
3.1.4. Substrate-dependent effects on ATP production efficiency and rate
Depending on the individual tissue and on nutritional status, the
main respiratory substrates used by the body are glucose and fatty
acids, although significant amounts of lactate are also oxidised by
heart, liver and brain. Each substrate has a different efficiency of ATP
production; for example, oxidation of palmitate yields approximately 3
times the amount of ATP per mole as does complete oxidation of
glucose, resulting in maximum amounts of ATP synthesised per oxygen
atom consumed (P/O) that differ by some 14% (2.1 vs 2.4, respectively,
assuming an H+/ATP ratio of 13/3, see [73]). The effective P/O of
substrate oxidation varies with the extent of proton leak, increasing
towards state 3 as the proton leak rate declines [74]. The oxidation of a
particular substrate will affect NADH/NAD+ redox state, membrane
potential (Δψ), and respiration (including proton leak) rate to different
extents. As discussed in [74], hormones that increase Ca2+, stimulate
respiration and/or increase proton leak may also influence the effective
P/O and the rate of ATP production. The relative contribution of each
the factors above to the in vivo rate of ATP production is unclear.
However, according to the relative control of each of these parameters
over respiration under the prevailing conditions, the effective P/O and
the rate of ATP production may indeed vary (see [74]).
As control of respiration shifts towards substrate oxidation with
increasing demand for ATP, substrates that increase the rate of
respiration when demand is high are likely to increase ATP production
rate. Equally, under similar conditions of increasing demand for ATP,
substrate deprivation could have a detrimental effect on cellular
energetics, particularly under conditions of low oxygen.
Other substrate-dependent effects on cellular energetics include,
for example, direct inhibition of the adenine nucleotide translocase
(ANT) by long-chain acyl-CoA esters which, given the significant
767
proportion of control of the phosphorylating system over respiration,
could significantly affect cellular energetics [75].
3.1.5. Substrate-dependent effects in sepsis
In sepsis, changes in substrate-dependent effects on ATP
production rate and efficiency may be reflected at a number of the
levels discussed above. Firstly, circulating cytokines and catechola-
mines cause a shift in metabolism towards the stress state: acti-
vation of glycogenolysis and hepatic gluconeogenesis, together with
insulin resistance lead to increased circulating levels of glucose;
hepatic lipolysis increases plasma free fatty acids, while greatly
increased muscle protein catabolism and lactate release also occur
(see [12,14]). While the levels of these circulating substrates are
increased, the extent to which the substrates are oxidised or taken
up by cells and mitochondria of different tissues may be variable. For
example, insulin-independent glucose uptake may be inhibited in
sepsis in some cells by down-regulation of the glucose transporter,
GLUT-4, whilst expression of insulin-dependent glucose transpor-
ters is increased in the presence of cytokines. The disruptions in
circulating levels of Ca2+-modulating hormones discussed above
could also have significant effects on the activation state of
mitochondrial dehydrogenases, thus affecting the availability of
respiratory substrates. Other mechanisms by which substrate
availability may be affected in sepsis are by depletion of NAD+ via
peroxynitrite-induced activation of PARP (poly-(ADP ribose) poly-
merase), a DNA-repair enzyme [76].
3.2. Oxygen availability
3.2.1. Cell responses to tissue hypoxia
The extent to which tissue hypoxia causes ATP depletion and
associated cell damage in a particular tissue will depend on (1) the
metabolic response of the system to short-term changes in O2 tension/
concentration [O2] and (2) the sensitivity of the system to any changes
in [ATP], both in terms of short-term changes in control of metabolic
flux and longer-term adaptive responses to these changes (see [77]).
3.2.2. Short-term effects of hypoxia
Whereas some cells (e.g. kidney) are highly sensitive to changes in
[O2] and reduce their consumption as [O2] decreases [78], the oxygen
uptake rate of many other cells is relatively insensitive until O2 levels
as low as 3–8 µM are reached [79–81]. The apparent KmO2 in
hepatocytes is reported to be approximately 2 µM [82]. Changes in
intermediate concentrations (e.g. NADH/NAD+, Δψ, reduced cyto-
chrome c, phosphorylated metabolites) can occur without a change in
respiratory rate or [ATP] ([82], discussed in [36]). Below these oxygen
tensions, there are changes in both respiratory rate and levels of
intermediates. Gluconeogenesis and urea synthesis – pathways in
hepatocytes that both consume and produce ATP – exhibit a similar
sensitivity to changes in O2 (apparent KmO2 5 µM; [83]). As respiratory
oxygen consumption decreases in low O2, glycolysis increases in rate
(the Pasteur effect).
Factors that influence the availability of O2 include (1) the effect of
vasoactive hormones on oxygen delivery by heart, lungs and
circulation to the tissues (potentially in addition to direct endocrine
effects on the energetics of the tissue of interest) and (2) nitric oxide,
NO, the effects of which could include (a) cGMP-dependent vaso-
relaxation, affecting oxygen delivery and (b) competitive inhibition of
cytochrome c oxidase that, besides inhibiting the respiratory rate, may
effectively increase the local [O2] within the cell. Such an effect of NO
may be relevant to the availability of O2 to enable the seemingly
paradoxical increase in mitochondrial ROS production during hypoxia.
However, the quantities of NO produced in inflammation are likely to
be detrimental to the cell, especially in combination with hypoxia. For
example, hypoxia has been shown to sensitize mitochondria to
inhibition by NO in isolated inflamed aorta [84], neuronal [85,86]
768 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
and activated macrophage cell lines [49], resulting in increased
necrotic cell death. Inhibition by NO was reversible if NO was removed
from the medium, but irreversible following prolonged exposure to
NO, probably reflecting a reversible competitive inhibition by NO at
complex IV followed by more stable inhibition of complex I by
peroxynitrite [49,84].
3.2.3. Longer-term responses to hypoxia
Cellular responses to hypoxia depend on the tissue itself, and the
duration and extent of hypoxia. For example, hepatocytes exhibit
progressive decreases in [ATP] and oxygen consumption rate as
hypoxia continued over several hours [59,87]. These decreases were
fully reversible upon reoxygenation, normalising more rapidly than
for cells exposed to a shorter period of hypoxia [59]. Prolonged
moderate hypoxia was associated with maintained Na+/K+- ATPase
activity while other, non-essential ATPase activities were suppressed
[59]. Such a “hibernation” response to hypoxia is also seen in
ischaemic myocardium. The mechanism of such adaptive responses
is unclear [77,88]. Possible candidates for molecular regulation include
hypoxia-inducible transcription factor-1 (HIF-1a) [89] or AMP-
activated protein kinase [61,90].
3.2.4. Oxygen availability in sepsis
A major point of contention in the sepsis field is the extent to
which tissue oxygen delivery compromises the mitochondrial func-
tion. Early studies interpreted a high blood lactate level (usually
associated with a metabolic acidosis) as indicative of tissue hypoxia,
representing anaerobic glycolytic metabolism [91]. As metabolic
acidosis was positively correlated with mortality in sepsis [92], this
finding was perceived as deleterious. Despite early suggestions that
hyperlactataemia could occur in the absence of oxygen delivery
defects, the tissue hypoxia theory prevailed through the 1990s and
prompted two large clinical trials aiming to optimise tissue perfusion
of septic patients in established multi-organ failure. Alas, these
showed either no benefit [93] or harm [94].
Low cardiac output states such as cardiogenic, hypoxic, or
haemorrhagic shock are characterised by decreases in tissue oxygen
delivery and consequent tissue hypoxia [95]. In contrast, sepsis is
generally manifest as a high cardiac output state — despite hypoten-
sion, global oxygen supply is often reasonably well maintained. There
is, however, lactic acidaemia and evidence of local impairment of
microvascular blood flow, predominantly through vasoconstriction. It
has thus been argued that the resulting localised tissue hypoxia will
lead to organ failure [96]. However, several animal and human studies
suggest that, despite indications of microcirculatory abnormality,
peripheral organs do not become hypoxic during sepsis after initial
resuscitation has been performed. Elevated tissue oxygen tensions
(PO2) have been reported for bladder [95,97,98] and gut [99] in
volume-resuscitated septic models, as well as in the muscle of septic
patients [100,101]. Recently, using a short-term (3 h), fluid-resusci-
tated endotoxaemic rat model, there was an elevation in bladder PO2
and relatively maintained muscle and renal PO2 despite a significant
reduction in cardiac output [95]. On the other hand, liver PO2 fell both
early and profoundly, paralleling changes seen with severe haemor-
rhage and hypoxaemia. Similar findings have been seen after 6 h in a
faecal peritonitis septic rat model yet the tissue PO2 values normalized
by 24 h, despite severe clinical and biochemical manifestations of
organ dysfunction becoming apparent (Dyson and Singer, unpub-
lished data). Thus, the organ tissue PO2 response to sepsis – which
reflects the local O2 supply-demand balance – varies, even in the face
of decreased oxygen delivery and potential micorovascular abnorm-
alities. A reconciling view is that both macro- and microcirculatory
abnormalities occur early in sepsis and, if not corrected sufficiently
quickly, can amplify the inflammatory response with a possibly
greater impact on later mitochondrial dysfunction. However, this
needs to be confirmed and the debate continues.
3.3. Functional capacity
The concept of “reserve capacity” in mitochondria is well
recognised, in that the respiratory rate can be stimulated above
state 3 in isolated mitochondria, or over the endogenous rate in cells
upon dissipation of Δψ by the addition of a chemical uncoupler. Such
a reserve might be important in some tissues when the demand for
cellular ATP acutely rises and the proportion of control over
respiration exerted by the respiratory chain increases ([52], dis-
cussed in [35]). In this respect, under conditions of elevated ATP
demand, systems which are unable to maintain functional mito-
chondria might be expected to compromise their function and even
viability. Depending on the system, short-term adjustment in the rate
of “non-essential” (e.g. macromolecule synthesis) or facultative
(metabolic) ATP demand pathways, may initially occur to allow the
cell to prioritise essential processes such as maintenance of ion
gradients. In the longer term, failure to maintain mitochondrial
functional capacity may result in ATP depletion and this will trigger
cell death pathways. These two responses — a decrease in metabolic
function and increased cell death— could conceivably be manifested
clinically as organ failure.
Mitochondrial functional capacity in the cell is determined by
several factors including: (1) the extent to which existing mitochondrial
enzymes retain functionality, including (a) the extent of oxidative/
nitrosative damage and (b) the ability of enzymes to respond to
allosteric activators; (2) the rate of synthesis and assembly of proteins
in the mitochondria; and (3) the rate of proteolytic degradation. The
extent to which changes in functional capacity are important in
defining overall energetic metabolism in sepsis will depend on the
hierarchy of control over the system in the septic condition.
3.3.1. Impaired mitochondrial function in sepsis
Reports of altered respiratory capacity in isolated mitochondria in
sepsis date back nearly 40 years, with conflicting findings [reviewed in
[20] depending on the duration and severity of the model and the
tissue studied. The majority of reports using isolated mitochondria or
permeabilised cells and muscle fibres point towards an early increase
in capacity (4–18 h) followed by a later depression (24–48 h). A
decrease in complex I-linked respiration in isolated muscle fibres was
associated with the severity of sepsis in a long-term rat model, while
succinate-dependent oxidation rates were similar between control
and septic groups [102]. Differences in respiration between intact
hepatocytes (decreased endogenous rate) and mitochondria
(increased state 3 rate) isolated from livers of endotoxic rats have
been explained by the existence of variably affected mitochondrial
populations [24]. Electron micrographs indicate a population of
swollen mitochondria amidst more normal looking, electron-dense
mitochondria. Potentially, these swollen organelles may be more
susceptible to rupture and removal during the mitochondrial isolation
procedure.
3.3.2. Mitochondrial biogenesis
As described earlier, limited patient studies report decreases in
mitochondrial enzyme activities [21,22] that are associated with
oxidative/nitrosative damage [27]. Similar findings have been made in
animal models [23–25,103], and may be related to a functional
inhibition and/or damage/loss of active respiratory complex protein,
particularly complex I. The recovery of mitochondrial function
requires the process of organelle repair/regeneration. This process is
called mitochondrial biogenesis.
Mitochondrial biogenesis describes the co-ordination of nuclear-
and mitochondrial gene expression, protein import and assembly into
mitochondrial membranes and the changes in mitochondrial mass and
morphology. This process can be regulated, in theory, at each level,
although the relative importance remains to be established. Both ROS
and NO have been implicated as signalling molecules for regulation of
 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
biogenesis pathways [104–106]. A key regulatory molecule in
mitochondrial biogenesis is peroxisome proliferator-activated recep-
tor gamma (PPARγ) co-activator-1α (PGC-1α), which is highly
expressed, in oxidative tissues including heart, kidney, skeletal muscle
and brain [107–109]. This nuclear transcriptional co-activator speci-
fically interacts with, and enhances, activity of transcription factors
that act on promoters of genes involved in energy metabolism. Specific
targets of PGC-1α common to different tissues include the nuclear
respiratory factors (NRF)-1 and NRF-2; these are co-activated to bind to
promoter regions of nuclear-encoded mitochondrial genes and to
those of the gene encoding mitochondrially-targeted transcription
factor A (Tfam, or mTFA). Another target of PGC-1α is the (o)estrogen-
related receptor-α (ERRα). The net effect of PGC-1α co-activation is the
upregulation of large groups of mitochondrial and nuclear-encoded
genes involved in respiration and β-fatty-acid oxidation. In the liver,
PGC-1α activates gluconeogenesis and fasting-related genes. PGC-1α
is subject to complex regulation at the transcriptional and post-
translational level (reviewed in [110]).
3.3.3. Mitochondrial biogenesis in sepsis
Cytokines increase stability and activity of PGC-1α through MAP
kinase-dependent phosphorylation in muscle cells [111]]. In the same
study, 12 h after administration of LPS, transgenic mice over-
expressing PGC-1α linked to the muscle creatine kinase (MCK)
promoter showed a near-doubling of whole body oxygen consump-
tion with both increased state 3 and state 4 respiration in isolated
muscle mitochondria. These increases in respiration were related to
elevated expression of some, but not all, of the PGC-1α downstream
targets measured. Levels of the PGC-1α protein itself increased in
response to LPS, but it is unclear to what extent this reflects the
endogenous protein or effects on the MCK promoter.
To investigate a potential role of mitochondrial biogenesis in recovery
from sepsis, a 3 day mouse model of bacterial sepsis was developed that
exhibited a reduction in whole body oxygen consumption and REE,
together with increased mRNA levels of mitochondrial superoxide
dismutase and a depression in mitochondrial DNA copy number
(mtDNA) [47]. Recovery was associated with increased expression of
NRF-1, Tfam and PGC-1α transcripts and a progressive increase in
cytochrome and mtDNA, followed by recovery of oxygen consumption
and REE. This study suggests a link between recovery from sepsis and
mitochondrial biogenesis. Interestingly, our preliminary (unpublished)
data in muscle biopsies taken from critically ill patients indicate that
transcript levels of PGC1α, Tfam and NRF-1 were significantly elevated in
patients on entry to the ICU, but only in those who went on to survive. A
similar pattern was seen in ATP and phosphocreatine content of these
biopsies [Brealey 2003 and unpublished data]. With the inherent
difficulties in obtaining samples from “vital” organs, it is uncertain
whether this situation is unique to muscle but together, these data
suggest that an ability to maintain functional mitochondria early in critical
illness, or an ability to rapidly regenerate, may be crucial to recovery.
4. Conclusions/future perspectives
The controversy of to what extent tissue hypoxia, mitochondrial
dysfunction and ATP depletion contribute to organ dysfunction in sepsis
has been ongoing for many years. From the discussion above, it is clear
that cellular energetic metabolism in the pathophysiological syndrome
of sepsis could be affected at multiple sites. However, understanding the
ATP supply-demand interaction is crucial for determining the relative
importance of these factors in organ failure, yet few data are available
that describe the effect of sepsis on the cellular demand for ATP. Limited
information can be gained from measurement of steady state metabolite
content or metabolic activities alone. Instead, a systems approach is
required to determine the relative importance of each of the factors
discussed above in the development and severity of MOF, and in the
subsequent recovery period.
769
By establishing the control distribution of a system, the effect of
modulations at different points in the system can be predicted. This
aspect is fundamental to the success of therapeutic interventions, as
targeting a part of the system that has little control at a particular time
will be unlikely to be of any benefit. We anticipate that the application
of a modular kinetic analysis [39–41] will enable the role of different
candidate effectors in the causality of organ dysfunction and recovery
to be determined for different tissues and immune cells throughout
the clinical course of the disease. Such an insight may prove useful
in determining the application of therapies aimed at stimulating
energetic metabolism at one time point, whilst suppressing it at
others.
Acknowledgements
We would like to acknowledge the Medical Research Council and
European Intensive Care Society for research funding.
References
[1] R.C. Bone, R.A. Balk, F.B. Cerra, R.P. Dellinger, A.M. Fein, W.A. Knaus, R.M. Schein,
W.J. Sibbald, Definitions for sepsis and organ failure and guidelines for the use of
innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Commit-
tee, Am. College of Chest Physicians/Soci. Critical Care Med. Chest 101 (1992)
1644–1655.
[2] D.C. Angus, R.S. Wax, Epidemiology of sepsis: an update, Crit Care Med. 29 (2001)
109–116.
[3] G.S. Martin, D.M. Mannino, S. Eaton, M. Moss, The epidemiology of sepsis in the
United States from 1979 through 2000, N. Engl. J. Med. 348 (2003) 1546–1554.
[4] D.A. Harrison, C.A. Welch, J.M. Eddleston, The epidemiology of severe sepsis in
England, Wales and Northern Ireland, 1996 to 2004: secondary analysis of a high
quality clinical database, the ICNARC Case Mix Programme Database, Crit. Care
(London, England) 10 (2006) R42.
[5] J.C. Marshall, D.J. Cook, N.V. Christou, G.R. Bernard, C.L. Sprung, W.J. Sibbald,
Multiple organ dysfunction score: a reliable descriptor of a complex clinical
outcome, Crit. Care Med. 23 (1995) 1638–1652.
[6] C. Brun-Buisson, F. Doyon, J. Carlet, P. Dellamonica, F. Gouin, A. Lepoutre, J.C.
Mercier, G. Offenstadt, B. Régnier, Incidence, risk factors, and outcome of severe
sepsis and septic shock in adults. A multicenter prospective study in intensive
care units. French ICU Group for Severe Sepsis, JAMA 274 (1995) 968–974.
[7] T.M. Perl, L. Dvorak, T. Hwang, R.P. Wenzel, Long-term survival and function after
suspected gram-negative sepsis, JAMA 274 (1995) 338–345.
[8] E. Abraham, M. Singer, Mechanisms of sepsis-induced organ dysfunction, Crit.
Care Med. 35 (2007) 2408–2416.
[9] R.S. Hotchkiss, I.E. Karl, The pathophysiology and treatment of sepsis, N. Engl. J.
Med. 348 (2003) 138–150.
[10] B. Ellger, Y. Debaveye, G. Van den Berghe, Endocrine interventions in the ICU, Eur.
J. Intern. Med. 16 (2005) 71–82.
[11] E.S. Nylen, B. Muller, Endocrine changes in critical illness, J. Intensive Care Med.
(2004).
[12] H.R. Michie, Metabolism of sepsis and multiple organ failure, World J. Surgery 20
(1996) 460–464.
[13] K. Träger, D. DeBacker, P. Radermacher, Metabolic alterations in sepsis and
vasoactive drug-related metabolic effects, Curr. Opinion in Crit. Care 9 (2003)
271–278.
[14] R.R. Wolfe, Sepsis as a modulator of adaptation to low and high carbohydrate and
low and high fat intakes, European J. Clin. Nutrition 53 (Suppl 1) (1999) 136–142.
[15] D.D. Rees, S. Cellek, R.M. Palmer, S. Moncada, Dexamethasone prevents the
induction by endotoxin of a nitric oxide synthase and the associated effects on
vascular tone: an insight into endotoxin shock, Biochem. Biophys. Res. Commun.
173 (1990) 541–547.
[16] J. Carré, M. Singer, S. Moncada, Nitric Oxide, in: J.-L. Vincent (Ed.), Mechanisms of
sepsis-induced organ dysfunction and recovery, vol. 44, Springer, Heidelberg,
2006, pp. 77–95.
[17] G. Kreymann, S. Grosser, P. Buggisch, C. Gottschall, S. Matthaei, H. Greten, Oxygen
consumption and resting metabolic rate in sepsis, sepsis syndrome, and septic
shock, Crit. Care Med. 21 (1993) 1012–1019.
[18] P.M. Rothwell, P.G. Lawler, Prediction of outcome in intensive care patients using
endocrine parameters, Crit. Care Med. 23 (1995) 78–83.
[19] E. Levy, R.J. Slusser, B.H. Ruebner, Hepatic changes produced by a single dose of
endotoxin in the mouse. Electron microscopy, Am. J. Pathol. 52 (1968) 477–502.
[20] M. Singer, D. Brealey, Mitochondrial dysfunction in sepsis, Biochem. Soc. Symp.
66 (1999) 149–166.
[21] D. Brealey, M. Brand, I. Hargreaves, S. Heales, J. Land, R. Smolenski, N.A. Davies, C.E.
Cooper, M. Singer, Association between mitochondrial dysfunction and severity
and outcome of septic shock, Lancet 360 (2002) 219–223.
[22] K. Fredriksson, F. Hammarqvist, K. Strigård, K. Hultenby, O. Ljungqvist, J.
Wernerman, O. Rooyackers, Derangements in mitochondrial metabolism in
intercostal and leg muscle of critically ill patients with sepsis-induced multiple
organ failure, Am. J. Physiol. Endocrinol. Metab. 291 (2006) 1044–1050.
770 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
[23] D. Brealey, S. Karyampudi, T.S. Jacques, M. Novelli, R. Stidwill, V. Taylor, R.T.
Smolenski, M. Singer, Mitochondrial dysfunction in a long-term rodent model of
sepsis and organ failure, Am. J. Physiol. Regul. Integr. Comp. Physiol. 286 (2004)
491–497.
[24] S.P. Kantrow, D.E. Taylor, M.S. Carraway, C.A. Piantadosi, Oxidative metabolism
in rat hepatocytes and mitochondria during sepsis, Arch. Biochem. Biophys.
345 (1997) 278–288.
[25] Y. Mizobata, D. Prechek, J.D. Rounds, V. Robinson, D.W. Wilmore, D.O. Jacobs, The
duration of infection modifies mitochondrial oxidative capacity in rat skeletal
muscle, J. Surg. Res. 59 (1995) 165–173.
[26] M.P. Fink, Bench-to-bedside review: cytopathic hypoxia, Crit. Care (London,
England) 6 (2002) 491–499.
[27] U.B. Fläring, O.E. Rooyackers, C. Hebert, T. Bratel, et al., Temporal changes in
whole-blood and plasma glutathione in ICU patients with multiple organ failure,
Intensive Care Med. (2005).
[28] U.B. Fläring, O.E. Rooyackers, J. Wernerman, F. Hammarqvist, Temporal
changes in muscle glutathione in ICU patients, Intensive Care Med. 29 (2003)
2193–2198.
[29] D.A. Svistunenko, N. Davies, D. Brealey, M. Singer, C.E. Cooper, Mitochondrial
dysfunction in patients with severe sepsis: an EPR interrogation of individual
respiratory chain components, Biochim. Biophys. Acta 1757 (2006) 262–272.
[30] G.H. Metnitz, M. Fischer, C. Bartens, H. Steltzer, T. Lang, W. Druml, Impact of acute
renal failure on antioxidant status in multiple organ failure Acta anaesthesiolo-
gica Scandinavica, 44, 2000, pp. 236–240.
[31] O. Huet, R. Obata, C. Aubron, A. Spraul-Davit, J. Charpentier, C. Laplace, T. Nguyen-
Khoa, M. Conti, E. Vicaut, J.P. Mira, J. Duranteau, Plasma-induced endothelial
oxidative stress is related to the severity of septic shock, Crit. Care Med. 35 (2007)
821–826.
[32] E.D. Crouser, Mitochondrial dysfunction in septic shock and multiple organ
dysfunction syndrome, Mitochondrion 4 (2004) 729–741.
[33] R.S. Hotchkiss, P.E. Swanson, B.D. Freeman, K.W. Tinsley, J.P. Cobb, G.M.
Matuschak, T.G. Buchman, I.E. Karl, Apoptotic cell death in patients with sepsis,
shock, and multiple organ dysfunction, Crit. Care Med. 27 (1999) 1230–1251.
[34] M. Singer, V. De Santis, D. Vitale, W. Jeffcoate, Multiorgan failure is an adaptive,
endocrine-mediated, metabolic response to overwhelming systemic inflamma-
tion, Lancet 364 (2004) 545–548.
[35] D. Nicholls, Mitochondrial bioenergetics, aging, and aging-related disease Science
of aging knowledge environment: SAGE KE, 2002, 2002, pe12.
[36] G.C. Brown, Control of respiration and ATP synthesis in mammalian mitochondria
and cells, Biochem. J. 284 (Pt 1) (1992) 1–13.
[37] R. Heinrich, T.A. Rapoport, Linear theory of enzymatic chains; its application for
the analysis of the crossover theorem and of the glycolysis of human
erythrocytes, Acta Biol. Med. Ger. 31 (1973) 479–494.
[38] H. Kacser, J.A. Burns, The control of flux, Symp. Soc. Exp. Biol. 27 (1973) 65–104.
[39] M.D. Brand, Top down metabolic control analysis, J. Theor. Biol.182 (1996) 351–360.
[40] S. Schuster, D. Kahn, H.V. Westerhoff, Modular analysis of the control of complex
metabolic pathways, Biophys. Chem. 48 (1993) 1–17.
[41] G.C. Brown, R.P. Hafner, M.D. Brand, A qtop–downq approach to the determination
of control coefficients in metabolic control theory, Eur. J. Biochem. 188 (1990)
321–325.
[42] A.K. Groen, R.J. Wanders, H.V. Westerhoff, R. van der Meer, J.M. Tager,
Quantification of the contribution of various steps to the control of mitochondrial
respiration, J. Biol. Chem. 257 (1982) 2754–2757.
[43] E.K. Ainscow, M.D. Brand, The responses of rat hepatocytes to glucagon and
adrenaline. Application of quantified elasticity analysis, Eur. J. Biochem. 265
(1999) 1043–1055.
[44] M.E. Harper, M.D. Brand, Hyperthyroidism stimulates mitochondrial proton leak
and ATP turnover in rat hepatocytes but does not change the overall kinetics of
substrate oxidation reactions, Can. J. Physiol. Pharmacol. 72 (1994) 899–908.
[45] F. Buttgereit, M.D. Brand, M. Müller, Effects of methylprednisolone on the energy
metabolism of quiescent and ConA-stimulated thymocytes of the rat, Biosci. Rep.
13 (1993) 41–52.
[46] W.J. Hubbard, M. Choudhry, M.G. Schwacha, J.D. Kerby, L.W. Rue, K.I. Bland, I.H.
Chaudry, Cecal ligation and puncture, Shock 24 (Suppl 1) (2005) 52–57.
[47] D.W. Haden, H.B. Suliman, M.S. Carraway, K.E. Welty-Wolf, A.S. Ali, H. Shitara, H.
Yonekawa, C.A. Piantadosi, Mitochondrial biogenesis restores oxidative metabo-
lism during Staphylococcus aureus sepsis, Am. J. Respir. Crit. Care Med. 176 (2007)
768–777.
[48] S. Berg, P.L. Sappington, L.J. Guzik, R.L. Delude, M.P. Fink, Proinflammatory
cytokines increase the rate of glycolysis and adenosine-5'-triphosphate turnover
in cultured rat enterocytes, Crit. Care Med. 31 (2003) 1203–1212.
[49] M.T. Frost, Q. Wang, S. Moncada, M. Singer, Hypoxia accelerates nitric oxide-
dependent inhibition of mitochondrial complex I in activated macrophages, Am.
J. Physiol. Regul. Integr. Comp. Physiol. 288 (2005) 394–400.
[50] A. Jacob, M. Zhou, R. Wu, V.J. Halpern, T.S. Ravikumar, P. Wang, Pro-inflammatory
cytokines from Kupffer cells downregulate hepatocyte expression of adrenome-
dullin binding protein-1, Biochim. Biophys. Acta 1772 (2007) 766–772.
[51] I. Belikova, A.C. Lukaszewicz, V.F. Ass, C. Damoisel, M. Singer, D. Payen, Oxygen
consumption of human peripheral blood mononuclear cells in severe human
sepsis, Crit. Care Med. (2007).
[52] R.P. Hafner, G.C. Brown, M.D. Brand, Analysis of the control of respiration rate,
phosphorylation rate, proton leak rate and protonmotive force in isolated
mitochondria using the 'top–down' approach of metabolic control theory, Eur. J.
Biochem. 188 (1990) 313–319.
[53] G.C. Brown, P.L. Lakin-Thomas, M.D. Brand, Control of respiration and oxidative
phosphorylation in isolated rat liver cells, Eur. J. Biochem. 192 (1990) 355–362.
[54] D.F. Rolfe, A.J. Hulbert, M.D. Brand, Characteristics of mitochondrial proton leak
and control of oxidative phosphorylation in the major oxygen-consuming tissues
of the rat, Biochim. Biophys. Acta 1188 (1994) 405–416.
[55] J.H. James, C.H. Fang, S.J. Schrantz, P.O. Hasselgren, R.J. Paul, J.E. Fischer, Linkage of
aerobic glycolysis to sodium–potassium transport in rat skeletal muscle. Implica-
tions for increased muscle lactate production in sepsis, J. Clin. Invest. 98 (1996)
2388–2397.
[56] J.D. Campbell, R.J. Paul, The nature of fuel provision for the Na+,K+-ATPase in
porcine vascular smooth muscle, J. Physiol. (Lond) 447 (1992) 67–82.
[57] D. Atkinson, Cellular energy metabolism and its regulation, Academic Press,
New York, 1977.
[58] F. Buttgereit, M.D. Brand, A hierarchy of ATP-consuming processes in mammalian
cells, Biochem. J. 312 (Pt 1) (1995) 163–167.
[59] R.M. Subramanian, N. Chandel, G.R. Budinger, P.T. Schumacker, Hypoxic confor-
mance of metabolism in primary rat hepatocytes: a model of hepatic hibernation,
Hepatology 45 (2007) 455–464.
[60] W. Wieser, G. Krumschnabel, Hierarchies of ATP-consuming processes: direct
compared with indirect measurements, and comparative aspects, Biochem. J. 355
(2001) 389–395.
[61] D.G. Hardie, J.W. Scott, D.A. Pan, E.R. Hudson, Management of cellular energy by
the AMP-activated protein kinase system, FEBS Lett. 546 (2003) 113–120.
[62] U. Schlattner, M. Tokarska-Schlattner, T. Wallimann, Mitochondrial creatine
kinase in human health and disease, Biochim. Biophys. Acta 1762 (2006)
164–180.
[63] B. Levy, S. Gibot, P. Franck, A. Cravoisy, P.E. Bollaert, Relation between muscle
Na+K+ ATPase activity and raised lactate concentrations in septic shock: a
prospective study, Lancet 365 (2005) 871–875.
[64] D.F. Rolfe, G.C. Brown, Cellular energy utilization and molecular origin of standard
metabolic rate in mammals, Physiol. Rev. 77 (1997) 731–758.
[65] V. Nogueira, L. Walter, N. Avéret, E. Fontaine, M. Rigoulet, X.M. Leverve, Thyroid
status is a key regulator of both flux and efficiency of oxidative phosphorylation
in rat hepatocytes, J. Bioenerg. Biomembr. 34 (2002) 55–66.
[66] D.G. Nicholls, Forty years of Mitchell's proton circuit: From little grey books to
little grey cells, Biochim. Biophys. Acta — Bioenergetics. 1777 (2008) 550–556.
[67] C. Affourtit, P.G. Crichton, N. Parker, M.D. Brand, Novel uncoupling proteins,
Novartis Found Symp. 287 (2007) 70–80 discussion 80–91.
[68] M.D. Brand, J.L. Pakay, A. Ocloo, J. Kokoszka, D.C. Wallace, P.S. Brookes, E.J.
Cornwall, The basal proton conductance of mitochondria depends on adenine
nucleotide translocase content, Biochem. J. 392 (2005) 353–362.
[69] D. Arsenijevic, H. Onuma, C. Pecqueur, S. Raimbault, B.S. Manning, B. Miroux, E.
Couplan, M.C. Alves-Guerra, M. Goubern, R. Surwit, F. Bouillaud, D. Richard, S.
Collins, D. Ricquier, Disruption of the uncoupling protein-2 gene in mice reveals a
role in immunity and reactive oxygen species production, Nat. Genet. 26 (2000)
435–439.
[70] S. Rousset, Y. Emre, O. Join-Lambert, C. Hurtaud, D. Ricquier, A.M. Cassard-
Doulcier, The uncoupling protein 2 modulates the cytokine balance in innate
immunity, Cytokine 35 (2006) 135–142.
[71] M.P. Murphy, Slip and leak in mitochondrial oxidative phosphorylation, Biochim.
Biophys. Acta 977 (1989) 123–141.
[72] B. Kadenbach, Intrinsic and extrinsic uncoupling of oxidative phosphorylation,
Biochim. Biophys. Acta 1604 (2003) 77–94.
[73] M.D. Brand, The efficiency and plasticity of mitochondrial energy transduction,
Biochem. Soc. Trans. 33 (2005) 897–904.
[74] M.D. Brand, M.E. Harper, H.C. Taylor, Control of the effective P/O ratio of oxidative
phosphorylation in liver mitochondria and hepatocytes, Biochem. J. 291 (Pt 3)
(1993) 739–748.
[75] J. Ciapaite, G. van Eikenhorst, S.J. Bakker, M. Diamant, R.J. Heine, M.J. Wagner, H.V.
Westerhoff, K. Krab, Modular kinetic analysis of the adenine nucleotide
translocator-mediated effects of palmitoyl-CoA on the oxidative phosphorylation
in isolated rat liver mitochondria, Diabetes 54 (2005) 944–951.
[76] C. Szabó, B. Zingarelli, M. O'Connor, A.L. Salzman, DNA strand breakage, activation
of poly (ADP-ribose) synthetase, and cellular energy depletion are involved in the
cytotoxicity of macrophages and smooth muscle cells exposed to peroxynitrite,
Proc. Natl. Acad. Sci. U. S. A. 93 (1996) 1753–1758.
[77] R.G. Boutilier, Mechanisms of cell survival in hypoxia and hypothermia, J. Exp.
Biol. 204 (2001) 3171–3181.
[78] C. Bauer, A. Kurtz, Oxygen sensing in the kidney and its relation to erythropoietin
production, Annu. Rev. Physiol. 51 (1989) 845–856.
[79] D.I. Edelstone, M.E. Paulone, I.R. Holzman, Hepatic oxygenation during arterial
hypoxemia in neonatal lambs, Am. J. Obstet. Gynecol. 150 (1984) 513–518.
[80] J. Lutz, H. Henrich, E. Bauereisen, Oxygen supply and uptake in the liver and the
intestine, Pflugers Arch. 360 (1975) 7–15.
[81] K. Nagano, S. Gelman, D.A. Parks, E.L. Bradley, Hepatic oxygen supply-uptake
relationship and metabolism during anesthesia in miniature pigs, Anesthesiology
72 (1990) 902–910.
[82] D.P. Jones, H.S. Mason, Gradients of O2 concentration in hepatocytes, J. Biol. Chem.
253 (1978) 4874–4880.
[83] T. Kashiwagura, D.F. Wilson, M. Erecińska, Oxygen dependence of cellular
metabolism: the effect of O2 tension on gluconeogenesis and urea synthesis in
isolated rat hepatocytes, J. Cell Physiol. 120 (1984) 13–18.
[84] V. Borutaite, S. Moncada, G.C. Brown, Nitric oxide from inducible nitric oxide
synthase sensitizes the inflamed aorta to hypoxic damage via respiratory inhibition,
Shock 23 (2005) 319–323.
[85] A. Jekabsone, J.J. Neher, V. Borutaite, G.C. Brown, Nitric oxide from neuronal nitric
oxide synthase sensitises neurons to hypoxia-induced death via competitive
inhibition of cytochrome oxidase, J. Neurochem. 103 (2007) 346–356.
 J.E. Carré, M. Singer / Biochimica et Biophysica Acta 1777 (2008) 763–771
[86] P. Mander, V. Borutaite, S. Moncada, G.C. Brown, Nitric oxide from inflammatory-
activated glia synergizes with hypoxia to induce neuronal death, J. Neurosci. Res.
79 (2005) 208–215.
[87] P.T. Schumacker, N. Chandel, A.G. Agusti, Oxygen conformance of cellular respi-
ration in hepatocytes, Am. J. Physiol. 265 (1993) 395–402.
[88] P.T. Schumacker, Current paradigms in cellular oxygen sensing, Adv. Exp. Med.
Biol. 543 (2003) 57–71.
[89] G.L. Wang, G.L. Semenza, General involvement of hypoxia-inducible factor 1 in
transcriptional response to hypoxia, Proc. Natl. Acad. Sci. U. S. A. 90 (1993)
4304–4308.
[90] D.G. Hardie, The AMP-activated protein kinase pathway—new players upstream
and downstream, J. Cell Sci. 117 (2004) 5479–5487.
[91] G. Broder, M.H. Weil, Excess lactate: an index of reversibility of shock in human
patient, Science 143 (1964) 1457–1459.
[92] U.G. Hodgin, J.P. Sanford, Gram-negative rod bacteremia. An analysis of 100
patients, Am. J. Med. 39 (1965) 952–960.
[93] L. Gattinoni, L. Brazzi, P. Pelosi, R. Latini, G. Tognoni, A. Pesenti, R. Fumagalli, A trial
of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collabora-
tive Group, N. Engl. J. Med. 333 (1995) 1025–1032.
[94] M.A. Hayes, A.C. Timmins, E.H. Yau, M. Palazzo, C.J. Hinds, D. Watson, Elevation of
systemic oxygen delivery in the treatment of critically ill patients, N. Engl. J. Med.
330 (1994) 1717–1722.
[95] A. Dyson, R. Stidwill, V. Taylor, M. Singer, Tissue oxygen monitoring in rodent
models of shock, Am. J. Physiol. Heart Circ. Physiol. 293 (2007) 526–533.
[96] D. De Backer, M.J. Dubois, Assessment of the microcirculatory flow in patients in
the intensive care unit, Curr. Opin. Crit. Care 7 (2001) 200–203.
[97] D.M. Rosser, M. Manji, H. Cooksley, G. Bellingan, Endotoxin reduces maximal
oxygen consumption in hepatocytes independent of any hypoxic insult, Intensive
Care Med. 24 (1998) 725–729.
[98] D.M. Rosser, R.P. Stidwill, D. Jacobson, M. Singer, Cardiorespiratory and
tissue oxygen dose response to rat endotoxemia, Am. J. Physiol. 271 (1996)
891–895.
[99] T.J. VanderMeer, H. Wang, M.P. Fink, Endotoxemia causes ileal mucosal acidosis in
the absence of mucosal hypoxia in a normodynamic porcine model of septic
shock, Crit. Care Med. 23 (1995) 1217–1226.
771
[100] P. Boekstegers, S. Weidenhöfer, G. Pilz, K. Werdan, Peripheral oxygen availability
within skeletal muscle in sepsis and septic shock: comparison to limited infection
and cardiogenic shock, Infection 19 (1991) 317–323.
[101] M. Sair, P.J. Etherington, C. Peter Winlove, T.W. Evans, Tissue oxygenation and
perfusion in patients with systemic sepsis, Crit. Care Med. 29 (2001) 1343–1349.
[102] A. Protti, J. Carré, M.T. Frost, V. Taylor, R. Stidwill, A. Rudiger, M. Singer, Succinate
recovers mitochondrial oxygen consumption in septic rat skeletal muscle, Crit.
Care Med. (2007) Publish Ahead of Print.
[103] E.D. Crouser, M.W. Julian, D.V. Blaho, D.R. Pfeiffer, Endotoxin-induced mitochon-
drial damage correlates with impaired respiratory activity, Crit. Care Med. 30
(2002) 276–284.
[104] E. Nisoli, E. Clementi, S. Moncada, M.O. Carruba, Mitochondrial biogenesis as a
cellular signaling framework, Biochem. Pharmacol. (2004).
[105] E. Nisoli, E. Clementi, C. Paolucci, V. Cozzi, C. Tonello, C. Sciorati, R. Bracale, A.
Valerio, M. Francolini, S. Moncada, M.O. Carruba, Mitochondrial biogenesis in
mammals: the role of endogenous nitric oxide, Science 299 (2003) 896–899.
[106] H.B. Suliman, K.E. Welty-Wolf, M.S. Carraway, L. Tatro, et al., Lipopolysaccharide
induces oxidative cardiac mitochondrial damage and biogenesis, Cardiovascular
Res. (2004).
[107] H. Esterbauer, H. Oberkofler, F. Krempler, W. Patsch, Human peroxisome
proliferator activated receptor gamma coactivator 1 (PPARGC1) gene: cDNA
sequence, genomic organization, chromosomal localization, and tissue expres-
sion, Genomics 62 (1999) 98–102.
[108] D. Knutti, A. Kaul, A. Kralli, A tissue-specific coactivator of steroid receptors,
identified in a functional genetic screen, Mol. Cell Biol. 20 (2000) 2411–2422.
[109] Z. Wu, P. Puigserver, U. Andersson, C. Zhang, G. Adelmant, V. Mootha, A. Troy, S.
Cinti, B. Lowell, R.C. Scarpulla, B.M. Spiegelman, Mechanisms controlling mito-
chondrial biogenesis and respiration through the thermogenic coactivator PGC-1,
Cell 98 (1999) 115–124.
[110] C. Handschin, B.M. Spiegelman, Peroxisome proliferator-activated receptor
gamma coactivator 1 coactivators, energy homeostasis, and metabolism, Endocr.
Rev. 27 (2006) 728–735.
[111] P. Puigserver, J. Rhee, J. Lin, Z. Wu, J.C. Yoon, C.Y. Zhang, S. Krauss, V.K. Mootha, B.B.
Lowell, B.M. Spiegelman, Cytokine stimulation of energy expenditure through p38
MAP kinase activation of PPARgamma coactivator-1, Mol. Cell 8 (2001) 971–982.