Download Physiological and pathophysiological roles of extracellular ATP in

1264
Biochemical Society Transactions (2007) Volume 35, part 5
Physiological and pathophysiological roles of
extracellular ATP in chemosensory control
of breathing
G.L. Ackland*†1 , V. Kasymov* and A.V. Gourine*2
*Department of Physiology, University College London, Gower Street, London WC1E 6BT, U.K., and †Centre for Anaesthesia, Critical Care and Pain
Management, University College London, Gower Street, London WC1E 6BT, U.K.
Abstract
The purine nucleotide ATP mediates several distinct forms of sensory transduction in both the peripheral
and central nervous systems. These processes share common mechanisms that involve the release of
ATP to activate ionotropic P2X and/or metabotropic P2Y receptors. Extracellular ATP signalling plays an
important role in ventilatory control, mediating both peripheral and central chemosensory transduction
to changes in arterial levels of oxygen and carbon dioxide. New data also suggest that extracellular ATP
may play an important role in mediating certain neurophysiological responses to systemic inflammation.
Here, we propose the novel concept that both peripheral and central neurophysiological effects of ATP may
contribute to alterations in ventilatory control during inflammatory pathophysiological states.
Introduction
Homoeostasis is critically dependent on afferent (sensory)
mechanisms to enable adaptive behavioural and physiological
responses. Many peripheral sensory processes share a
common mechanism that involves the release of ATP to excite
afferent fibres via activation of ligand-gated (P2X) and/or
G-protein-coupled (P2Y) cell-surface receptors [1]. ATP
has emerged as a critical mediator in several physiological
processes, including the peripheral and central chemosensory
control of ventilation [2]. ATP and its metabolite adenosine
also play important roles in the peripheral and central inflammatory responses, through modulation of both immune and
neural responses to acute inflammation [3]. It is well known
that alterations in ventilatory control are common in many
disparate pathophysiological states. In heart failure, the peripheral carotid body chemoreceptors, which serve to increase
ventilation through detection of low levels of pO2 (partial
pressure of oxygen) (hypoxia) and other blood-borne stimuli
[4], exhibit much higher basal activity under normoxic conditions. This enhanced peripheral chemoreflex contributes
to the tonic increase in overall sympathetic activity, which
is associated with poorer clinical outcome [5]. Similarly,
functional studies in humans with established and borderline
essential hypertension [6,7], and experimental studies in
spontaneously hypertensive rats, demonstrated higher basal
levels of ventilatory drive and enhanced respiratory and
sympathetic activities under hypoxic conditions [8]. The
carotid chemoreceptor response to hypoxia was found to be
Key words: adenosine, ATP, chemoreceptor, inflammation, sepsis, ventilation.
Abbreviations used: CNS, central nervous system; IL-1β, interleukin-1β.
1
To whom correspondence should be addressed (email [email protected]).
2
Owing to exceptional unforeseen circumstances, this speaker was unable to give this
presentation at the meeting. This paper is included in the interest of completeness of the
session.
C The
C 2007 Biochemical Society
Authors Journal compilation markedly exaggerated in spontaneously hypertensive rats [9].
In acute inflammation, for example during bacterial sepsis,
marked increases in ventilatory drive in critically ill patients
occur often despite normal arterial pO2 , relative hypocapnia
and absence of metabolic acidosis. These clinical observations
have been confirmed by experimental studies investigating
the effects of the bacterial endotoxin, lipopolysaccharide, on
ventilation and breath variability in humans [10]. Ongoing
inflammation and the release of key inflammatory mediators,
such as cytokines and prostaglandins, may provide a common
basis for these conditions. In this brief review, we first
consider the role of ATP in the development of inflammatory
response. Through preliminary results, we speculate as to
whether ATP extends its immunological role during pathophysiological states as a mediator between ventilatory control
and activity of the immune system. We propose the novel
concept that both peripheral and central neurophysiological
effects of ATP may contribute to alterations in ventilatory
control during inflammatory pathophysiological states.
ATP and inflammation
Extracellular ATP plays an important role in the activity of the
immune system. Acting through P2X7 receptors, ATP may
cause a variety of effects, including permeabilization of the
plasma membrane, cell death, cell proliferation, shedding of
cell-adhesion molecules, killing of intracellular pathogens or
secretion of various inflammatory cytokines, including IL-1β
(interleukin-1β) and tumour necrosis factor α [3]. Extracellular ATP also causes the generation of reactive oxygen species
when applied to macrophages [11]. Peripheral blockade of
P2X receptors during systemic inflammation reduces fever
and decreases plasma levels of major inflammatory cytokines
[12]. During systemic inflammation P2X7 receptors are
Central Nervous System
markedly up-regulated in the CNS (central nervous system)
where extracellular ATP triggers cytokine release from activated microglia [13]. Most importantly, there is evidence that
activated immune cells such as lymphocytes and macrophages
release large amounts of ATP into the extracellular space [14–
16]. In conscious rabbits, using amperometric ATP biosensors
placed in the third brain ventricle, we detected significant
increases in ATP concentration in the cerebrospinal fluid
following endotoxin challenge (Figure 1A).
Thus it appears that during inflammatory pathophysiological states, extracellular concentrations of ATP increase
both in the periphery and in the CNS. Below, we discuss
recent results from our laboratory demonstrating the
importance of ATP-mediated purinergic signalling in central
and peripheral chemosensory control of breathing and how
this control may be affected by extra ATP released under
pathological conditions.
Figure 1 Role of ATP in mediating CNS response to systemic
inflammation
(A) Increase in ATP concentration in the cerebrospinal fluid of the
third ventricle during the development of the systemic inflammatory
response in conscious rabbits. The graph shows representative raw
data illustrating changes in ATP level (measured in real time using
amperometric ATP biosensors) and development of fever [increase in
core body temperature (T b )] following intravenous administration of
Escherichia coli endotoxin (0.5 µg/kg). The arrow indicates time of
endotoxin injection. (B) ATP-induced increases in intracellular Ca2+
in Rhod-2-loaded transverse slices of the rat medulla oblongata.
Fluorescent sequential pseudocoloured images of a slice at the level
of the rostral ventrolateral medulla in control (arrow 1) and after the
application of ATP (arrow 2). The trace represents relative changes in
intensity of fluorescence over time in the corresponding images.
ATP and peripheral chemoreceptors
In adult mammals, type I (glomus) cells of the carotid body
are the main peripheral O2 sensors. A variety of stimuli, but
most importantly hypercapnia [high pCO2 (partial pressure
of carbon dioxide)], acidosis (low pH) and hypoxia (low pO2 ),
trigger glomus cells to release neurotransmitters that activate
afferent nerve fibres of the carotid sinus nerve, thereby
transducing chemosensory information to the respiratory
centres in order to evoke adaptive changes in breathing [4].
Mitochondrial cytochrome aa3 , which has very low oxygen
affinity [17], and non-mitochondrial cytochromes a592 [18]
and cytochromes b558 [19] have been implicated in the glomus
cell’s chemosensory mechanism. While the exact mechanism
of chemosensory transduction still remains incompletely
understood, two important players that may link the activity
of the immune systems with respiratory control have to
be considered in the context of the present review. First,
increased generation of reactive oxygen species has been
implicated in chronic intermittent hypoxia [20], which
results in increased chemosensitivity and sympathetic drive
across species including humans [21]. Secondly, ATP has
recently been identified as a key mediator of carotid body
chemosensory transduction [22]. ATP receptors have been
detected on the afferent terminals of the carotid sinus nerve
surrounding individual glomus cells or their clusters [22].
ATP evoked a dramatic increase in the carotid sinus nerve
chemoafferent discharge in a superfused in vitro murine
carotid body–carotid sinus nerve preparations. P2 receptor
antagonists dose-dependently reduced hypoxia-evoked activation of carotid chemoafferents. Critically, P2X2 receptorknockout mice demonstrated a markedly decreased hypoxic
ventilatory responses and up to 80% reduction in baseline
and hypoxia-stimulated carotid sinus nerve discharge. Given
that CO2 -evoked carotid chemoreceptor discharge is similarly reduced in P2X2 -knockout mice (A.V. Gourine,
unpublished work), ATP clearly plays a key role in the
carotid body process of conveying information about arterial
pO2 , pCO2 and pH to the CNS.
How could extracellular ATP link
inflammation with changes in ventilatory
control at the level of the carotid body?
There are striking immunological features of the carotid body
that merit consideration. The carotid body is a rich, although
previously unrecognized, abundant site of monocytes
and macrophages within its perivascular and connective
tissue spaces [23]. Spectrophotometric recordings reveal
that monocytes/macrophages represent the dominating cell
C The
C 2007 Biochemical Society
Authors Journal compilation 1265
1266
Biochemical Society Transactions (2007) Volume 35, part 5
type through cytochrome b558 activity, which generates superoxide anions (O2 − ) through the NADPH oxidase system.
Hypoxia-induced up-regulation of several inflammatory and
hypoxia-related genes, including hypoxia-inducible factors,
in macrophages [24] in the vicinity of the carotid body glomus
cells may provide several stimuli (in addition to superoxide
anions) that could increase carotid sinus nerve discharge. Both
IL-1 [25] and IL-6 [26] receptors have been detected in the rat
carotid body, both of which are pivotal cytokines responsible
for shaping the inflammatory response and mediating adaptive autonomic and behavioural responses during infection.
Preliminary work from our laboratory, using superfused
ex vivo murine carotid body–carotid sinus nerve
preparations, supports numerous clinical observations that
systemic inflammation results in an increase in respiratory
activity. After mice had been subjected to systemic
inflammation (peritonitis) for 4 h, both baseline carotid sinus
discharge and chemosensory responses to hypoxia were
dramatically increased (Figure 2). This occurs in the absence
of circulating inflammatory mediators or acidosis but in
the presence of hyperoxia, which would normally minimize
carotid sinus discharge. These observations are consistent
with the concept that cellular dysoxia [27] occurs during
inflammation through several separate cellular mechanisms
including mitochondrial dysfunction and generation of free
radicals.
Several potential mechanisms exist that may explain how
immune activation may affect carotid body function. First,
local and/or circulating mediators of inflammation may
induce transcriptional changes in the glomus cells or directly
stimulate peripheral terminals of the carotid sinus nerve. A
precedent already exists for cytokines (e.g. IL-1β) that are
able to trigger increases in afferent vagal nerve activity, via
activation of macrophages and dendritic cells that have been
detected around vagal abdominal paraganglia of the rat [28].
This is one mechanism through which sickness behaviour
may be induced during compartmentalized inflammation.
Secondly, generation of superoxide anions contributes to
enhancing peripheral chemosensitivity in pathophysiological
states. For example, in pacing-induced heart failure, the
expression of NADPH oxidase, production of superoxide
anion and carotid sinus nerve discharge are all elevated [29].
Finally, taking into the account the key role played by
ATP in mediating carotid body chemosensory transduction
we propose that ATP may be responsible for increases in
baseline carotid sinus discharge, augmented responses to
chemosensory stimulation and high respiratory drive under
pathological conditions. Indeed, as discussed above, during
development of the systemic inflammatory response, activated immune cells release large amounts of ATP, resulting
in increases in tissue and plasma levels [30]. If this extra
ATP reaches peripheral terminals of the carotid sinus nerve
chemoafferent fibres the activities of the latter would undoubtedly increase via interaction of ATP with P2X2 , P2X3
and P2X2 /P2X3 receptors [22]. Interestingly, plasma levels
of the ATP breakdown product adenosine have also been
shown to be elevated during sepsis [31]. Adenosine not
C The
C 2007 Biochemical Society
Authors Journal compilation Figure 2 Whole carotid sinus chemoafferent nerve responses to
hypoxia in the isolated carotid body–sinus nerve preparations
excised from mice treated with either intraperitoneal injections of
sterile saline (control) or zymosan (500 mg/kg; sepsis)
Zymosan elicits a profound systemic inflammatory response characterized by release of cytokines and other inflammatory mediators.
Basal carotid sinus nerve discharge (normoxia 95% oxygen/5% carbon
dioxide) in preparations taken from the zymosan-treated mouse is
substantially higher compared with that in the controls. Chemosensory
response to hypoxia (95% nitrogen/5% carbon dioxide for 3 min) has
been found to be dramatically potentiated during zymosan-induced
sepsis.
only exhibits powerful immunosuppressive effects, but also
directly stimulates carotid chemoreceptors through postsynaptic A2a adenosine receptors [32,33] and presynaptic A2b
adenosine receptors [33]. Some other key mediators in
systemic inflammation such as nitric oxide, which is increased
markedly during sepsis [34], interestingly reduce carotid
body chemosensitivity [35].
ATP, central chemosensitivity and
inflammation (Figure 3)
The brainstem respiratory network, which generates
respiratory activity, is located just above the classical
CO2 /[H+ ] chemosensitive areas that were identified by
Hans Loeschcke and Robert Mitchell on the ventral surface
of the medulla oblongata. Compelling data now suggest that
ATP also plays an important role in central chemosensory
transduction, mediating the effects of increased arterial
pCO2 on breathing [36]. Blockade of ATP receptors within
the ventral respiratory network decreases resting respiratory
activity and attenuates the increase in ventilation induced by
CO2 . At the cellular level, blockade of ATP receptors reduces
baseline firing as well as CO2 -induced increases in the discharge of pre-inspiratory and inspiratory neurons of the
medullary respiratory network. Experiments, in which ATP
was applied exogenously demonstrated marked increases in
intracellular calcium on the ventral surface and adjacent areas
of the medulla oblongata (Figure 1B) as well as augmenting
central respiratory drive both in neonatal [37] and adult
[36] experimental animals. Furthermore, experiments using
amperometric ATP biosensors sited on the ventral medullary
surface CO2 /[H+ ] chemosensitive areas revealed a strong
temporal and quantitative correlation between ATP release and resulting changes in breathing. CO2 -evoked ATP
Central Nervous System
Figure 3 Hypothetical scheme illustrating potential interactive
sites where direct or indirect release of ATP by the cells of the
immune system may act to increase ventilatory drive
In the carotid body, a decrease in pO2 or an increase in pCO2 /[H+ ]
activates glomus cells, which release ATP as the main transmitter to
stimulate afferent terminals of the carotid sinus nerve via interaction
with P2X receptors that contain the P2X2 subunit, with or without
P2X3 subunit. Local release of extracellular ATP, other inflammatory
factors and endothelial-derived mediators increases carotid sinus nerve
chemoafferent discharge either directly or via cellular/genomic changes
within the glomus cells of the carotid body. On the ventral surface of the
medulla, an increase in pCO2 /[H+ ] activates primary chemosensors,
which release ATP to act via P2 receptors on ventrally projecting
dendrites of more dorsally located secondary chemosensitive neurons
and/or respiratory neurons. The activity of these neurons feeds into
the respiratory network and evokes adaptive increases in breathing.
Release of ATP in the CNS during systemic inflammation may increase
ventilation directly, through activation of the medullary respiratory
network. Descending excitatory projections to the respiratory centre
from the warm-sensitive neurons of the anterior hypothalamus may
further contribute to an increase in respiratory drive to meet metabolic
demands of the developing febrile response.
surface chemosensitive areas are likely to be exposed to
elevated levels of extracellular ATP even when arterial and
brain pCO2 and pH remain normal. We suggest that this may
contribute to the increases in ventilation often observed in
many disparate pathophysiological states that include an inflammatory component. Whether an increase in cerebrospinal
ATP concentration of this magnitude is indeed responsible for
changes in central respiratory drive remains to be determined.
A third pathway responsible for increases in ventilation
during systemic inflammation may originate directly from the
thermosensitive regions of the hypothalamus. We have recorded a marked increase in extracellular ATP in the preoptic
area/anterior hypothalamus during systemic inflammation
induced in rabbits by small amounts of endotoxin (A.V.
Gourine, unpublished work). In our early in vitro experiments, we observed profound excitatory effects of ATP on the
activity of hypothalamic warm-sensitive neurons [38]. The
preoptic area also mediates additional respiratory drive, observed at raised body temperature in anaesthetized rats [39].
During systemic inflammation, this pathway may play a similar role and contribute to an increase in respiratory activity to
meet metabolic demands of the developing febrile response.
Conclusion
ATP is a ubiquitous cellular energy source and intercellular
messenger molecule that plays an important role in coordinating many homoeostatic mechanisms, potentially
linking ventilatory control with the activity of the immune
system. We propose the novel concept that peripheral
and/or central neurophysiological effects of ATP, released
by activated cells of the immune system in the brain and the
periphery, may contribute to increases in ventilation during
inflammatory pathophysiological states.
A.V.G. is a Wellcome Trust Senior Research Fellow. Our experimental
work described in this review has been supported by The Wellcome
Trust, BBSRC (Biotechnology and Biological Sciences Research
Council) and a Young Investigator Award from the Intensive Care
Society to G.L.A.
References
release from the ventral surface of the medulla is site-specific
(no ATP release was detected on the dorsal surface of the
medulla) and does not require inputs from the peripheral
chemoreceptors.
These results suggest a second pathway through which
ATP can increase respiratory activity during inflammatory
pathophysiological states. Indeed, we have shown that
during systemic inflammation the level of ATP in circulating
cerebrospinal fluid increases, at a similar level to that detected
by amperometric ATP biosensors on the ventral medullary
surface chemosensitive areas (Figure 1). Therefore the ventral
1 Burnstock, G. (2007) Physiol. Rev. 87, 659–797
2 Gourine, A.V. (2005) J. Physiol. 568, 715–724
3 Bours, M.J., Swennen, E.L., Di Virgilio, F., Cronstein, B.N. and Dagnelie,
P.C. (2006) Pharmacol. Ther. 112, 358–404
4 Lahiri, S., Roy, A., Baby, S.M., Hoshi, T., Semenza, G.L. and Prabhakar,
N.R. (2006) Prog. Biophys. Mol. Biol. 91, 249–286
5 Schultz, H.D., Yu, L. and Ding, Y. (2007) Hypertension 50, 1–8
6 Habeck, J.O. (1991) J. Auton. Nerv. Syst. 34, 1–7
7 Somers, V.K., Mark, A.L. and Abboud, F.M. (1988) Hypertension 11,
608–612
8 Ohta, H. and Talman, W.T. (1995) Clin. Exp. Pharmacol. Physiol. 22,
S60–S61
9 Fukuda, Y., Sato, A. and Trzebski, A. (1987) J. Auton. Nerv. Syst. 19, 1–11
10 Preas, II, H.L., Jubran, A., Vandivier, R.W., Reda, D., Godin, P.J., Banks,
S.M., Tobin, M.J. and Suffredini, A.F. (2001) Am. J. Respir. Crit. Care Med.
164, 620–626
C The
C 2007 Biochemical Society
Authors Journal compilation 1267
1268
Biochemical Society Transactions (2007) Volume 35, part 5
11 Cruz, C.M., Rinna, A., Forman, H.J., Ventura, A.L., Persechini, P.M. and
Ojcius, D.M. (2007) J. Biol. Chem. 282, 2871–2879
12 Gourine, A.V., Poputnikov, D.M., Zhernosek, N., Melenchuk, E.V.,
Gerstberger, R., Spyer, K.M. and Gourine, V.N. (2005) Br. J. Pharmacol.
146, 139–145
13 Choi, H.B., Ryu, J.K., Kim, S.U. and McLarnon, J.G. (2007) J. Neurosci. 27,
4957–4968
14 Beigi, R., Kobatake, E., Aizawa, M. and Dubyak, G.R. (1999)
Am. J. Physiol. 276, C267–C278
15 Dubyak, G.R. and El-Moatassim, C. (1993) Am. J. Physiol. 265,
C577–C606
16 Filippini, A., Taffs, R.E., Agui, T. and Sitkovsky, M.V. (1990) J. Biol. Chem.
265, 334–340
17 Mills, E.E. and Jöbsis, F.F. (1970) Nature 225, 1147–1149
18 Acker, H., Dufau, E., Huber, J. and Sylvester, D. (1989) FEBS Lett. 256,
75–78
19 Streller, T., Huckstorf, C., Pfeiffer, C. and Acker, H. (2002) FASEB J. 16,
1277–1279
20 Prabhakar, N.R., Dick, T.E., Nanduri, J. and Kumar, G.K. (2007)
Exp. Physiol. 92, 39–44
21 Lusina, S.J., Kennedy, P.M., Inglis, J.T., McKenzie, D.C., Ayas, N.T. and
Sheel, A.W. (2006) J. Physiol. 575, 961–970
22 Rong, W., Gourine, A.V., Cockayne, D.A., Xiang, Z., Ford, A.P., Spyer, K.M.
and Burnstock, G. (2003) J. Neurosci. 23, 11315–11321
23 Dvorakova, M., Hohler, B., Vollerthun, R., Fischbach, T. and Kummer, W.
(2000) Brain Res. 852, 349–354
24 Murdoch, C., Muthana, M. and Lewis, C.E. (2005) J. Immunol. 175,
6257–6263
25 Wang, X., Wang, B.R., Duan, X.L., Zhang, P., Ding, Y.Q., Jia, Y., Jiao, X.Y.
and Ju, G. (2002) J. Histochem. Cytochem. 50, 1677–1684
C The
C 2007 Biochemical Society
Authors Journal compilation 26 Wang, X., Zhang, X.J., Xu, Z., Li, X., Li, G.L., Ju, G. and Wang, B.R. (2006)
Anat. Rec. A Discov. Mol. Cell. Evol. Biol. 288, 292–296
27 Singer, M. (2005) Crit. Care Med. 33, S539–S542
28 Goehler, L.E., Gaykema, R.P.A., Nguyen, K.T., Lee, J.E., Tilders, F.J.H.,
Maier, S.F. and Watkins, L.R. (1999) J. Neurosci. 19, 2799–2806
29 Li, Y.L., Gao, L., Zucker, I.H. and Schultz, H.D. (2007) Cardiovasc Res. 75,
546–554
30 Pasque, M.K., Murphy, C.E., Van Trigt, P., Pellom, G.L., Currie, W.D. and
Wechsler, A.S. (1983) Arch. Surg. 118, 1437–1440
31 Martin, C., Leone, M., Viviand, X., Ayem, M.L. and Guieu, R. (2000)
Crit. Care Med. 28, 3198–3202
32 Kobayashi, S., Conforti, L. and Millhorn, D.E. (2000) Am. J. Physiol. Lung
Cell. Mol. Physiol. 279, L273–L282
33 Conde, S.V., Obeso, A., Vicario, I., Rigual, R., Rocher, A. and Gonzalez, C.
(2006) J. Neurochem. 98, 616–628
34 Hotchkiss, R.S. and Karl, I.E. (2003) N. Engl. J. Med. 348, 138–150
35 Prabhakar, N.R., Kumar, G.K., Chang, C.H., Agani, F.A. and Haxhiu, M.A.
(1993) Brain Res. 625, 16–22
36 Gourine, A.V., Llaudet, E., Dale, N. and Spyer, K.M. (2005) Nature 436,
108–111
37 Lorier, A.R., Huxtable, A.G., Robinson, D.M., Lipski, J., Housley, G.D. and
Funk, G.D. (2007) J. Neurosci. 27, 993–1005
38 Gourine, A.V., Dale, N., Gourine, V.N. and Spyer, K.M. (2004)
Front. Biosci. 9, 1011–1022
39 Boden, A.G., Harris, M.C. and Parkes, M.J. (2000) Exp. Physiol. 85,
527–537
Received 7 June 2007
doi:10.1042/BST0351264