Available online at www.sciencedirect.com
ScienceDirect Compartmentalisation of second messenger signalling pathways Kristie McCormick and George S Baillie The ability of a cell to transform an extracellular stimulus into a downstream event that directs specific physiological outcomes, requires the orchestrated, spatial and temporal response of many signalling proteins. The notion of compartmentalised signalling pathways was popularised in the 1980s by Brunton and colleagues, with their discovery that spatially segregated cAMP directs a variety of signalling responses in cardiomyocytes. It is now understood that compartmentalisation is a common mechanism used by all cells to ensure the interaction of signalling ‘second messenger’ molecules with localised ‘pools’ of appropriate effector proteins. In this way, the cell can elicit differential cellular responses by using a single, freely diffusible, molecular species. Recently, the compartmentalisation schemes employed by signalling systems involving cyclic nucleotides, calcium and nitric oxide have been elucidated and as a result, the varied range of functional consequences underpinned by such strategies can be better appreciated. Addresses Institute of Cardiovascular and Medical Sciences, CMVLS, Wolfson-Link Building, University of Glasgow, Glasgow G12 8QQ, UK Corresponding author: Baillie, George S ([email protected])
Current Opinion in Genetics & Development 2014, 27:20–25 This review comes from a themed issue on Developmental mechanisms, patterning and evolution Edited by Lee A Niswander and Lori Sussel
0959-437X/$ – see front matter, # 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.gde.2014.02.001
Introduction In order to translate the events occurring at the cell surface into an intracellular response, the cell enables specific molecules, so-called ‘second messengers’, as chemical relays to engage the signalling machinery within the cytoplasm. In short, cellular stimulation via cell surface receptors induces (often at the plasma membrane) transient production of signalling molecules that ‘find’ and bind to an appropriate effector protein. This action serves to initiate a signal cascade, which both amplifies the signal and directs it towards the alteration of a specific cellular process in response to the initial environmental threat or chemical cue on the outside of the cell. Despite the underlying mechanism being similar for each second messenger, the cellular responses are highly diverse, ranging from gene activation to cell division [1,2]. Current Opinion in Genetics & Development 2014, 27:20–25
A multiplicity of function also exists within each second messenger system, as a single chemical species may have multiple downstream effectors and regulate numerous physiological processes. Essentially, the cell achieves this feat by the spatio-temporal regulation, or compartmentalisation, of its second messengers to allow discrete pathways of localised signalling activity to occur. This review will discuss compartmentalisation with respect to four well-studied second messenger systems; cAMP, cGMP, calcium and nitric oxide (NO), concentrating on the way in which each system evokes its effects in a spatially restricted manner. The functional implications from recent groundbreaking studies will also be highlighted and discussed.
Compartmentalised cAMP signalling 30 ,50 -Cyclic adenosine monophosphate (cAMP) was the first second messenger to be discovered [3] and many of the concepts that we take for granted surrounding signal generation via effectors [4] and compartmentalisation [5] stem from seminal work on this molecule. Today, we understand that, when activated, a variety of different cell surface receptors coupled to Gas, educe equivalent increases in cAMP production that result in diverse cellular responses. This can only happen if the signalling ‘machinery’ for each process is spatially linked to the ‘pool’ of cAMP produced by adenylate cyclase in response to distinct receptors [6]. Such compartmentalisation allows anchored cAMP effectors such as Protein Kinase A (PKA) [7] and the exchange protein activated by cAMP (EPAC) [8] to sample cAMP transients and produce downstream signals when their activation threshold (in terms of cAMP concentration) is surpassed (Figure 1). The duration and strength of signals produced by cAMP effectors is heavily influenced by action of the only superfamily of enzymes that has evolved to degrade cAMP, the cAMP-phosphodiesterases (PDEs) [9]. The pivotal role of PDEs in shaping the spatial and temporal dimensions of cAMP ‘pulses’ produced in response to receptor activation was realised following the invention of a number of cAMP reporters that allowed real-time visualization of cAMP dynamics within cellular micro-domains [10]. These probes revealed spatially restricted cAMP gradients that were tailored by discretely anchored PDE populations to allow activation of small subsets of cAMP effectors. In effect, the PDEs acted as local cyclic nucleotide ‘sinks’ allowing ‘hotspots’ of cAMP to accumulate in areas of the cell that were devoid of PDEs (Figure 1). This action underpinned receptor specificity of action, as co-treatment with a non-specific PDE inhibitor and receptor agonist, allowed uniform, global increases in cAMP and concomitant activation of all cAMP effectors [11]. www.sciencedirect.com
Compartmentalised signalling McCormick and Baillie 21
Figure 1
Ligand (a)
(b)
Ligand
β
β
cAMP
cAMP
cAMP cAMP cAMP cAMP cAMP cAMP cAMP cAMP cAMP
(d)
PKA –
cAMP
cAMP cAMP
–
cAMP
–
+
os
ph cAM od P ie st er
–
EPAC
cAMP cAMP cAMP cAMP cAMP
cAMP
cAMP
+ cAMP
cAMP cAMP
cAMP cAMP cAMP
cAMP
cAMP cAMP cAMP
–
cAMP
cAMP
cAMP
terase
(c)
cAMP
(e)
Receptorspecific signal
cAMP
cAMP cAMP cAMP
–
e
s MP era cA iest d ho sp
+
c Phosp AMP hodie s
cAMP cAMP
cAMP
cAMP
PKA
cAMP cAMP cAMP
cAMP cAMP cAMP cAMP
cAMP
+
cAMP
–
cAMP
cAMP cAMP
e
–
cAMP Phosphodiesterase
cAMP
cAMP
cAMP
cAMP cAMP
+
–
as
cAMP
cAMP
–
cAMP cAMP cAMP
cAMP
cAMP
STOP
STOP
cAMP cAMP cAMP
cAMP
AC AC
cAMP
cAMP cAMP cAMP
GDP
Ph
cAMP cAMP cAMP
cAMP cAMP cAMP cAMP cAMP
–
γ αs
GTP
–
cAMP
GTP
–
cAMP
αs
+
αs
–
γ
+
cAMP
+ cAMP Phosphodiesterase
+ o
Ph
Current Opinion in Genetics & Development
Compartmentalisation of cAMP signalling. (A) The G-protein subunit Gas is released following receptor activation by the appropriate ligand. (B) The subunit binds adenylate cyclase (AC) to trigger production of cAMP. (C) Phosphodiesterases (PDEs) shape the cAMP gradient ensuring the specificity of signals emanating from defined receptors. (D) cAMP effectors, such as PKA and EPAC, that are in close proximity to a PDE ‘pool’ cannot be activated unless the associated PDE is ‘swamped’ by the cAMP produced in response to activation of the receptor. (E) The threshold of activation for cAMP effectors not in the vicinity of PDEs is more easily reached. These enzymes drive downstream signalling events that orchestrate physiological changes in response to the extracellular stimulus detected by cell-surface receptors.
Although the fundamental roles that cAMP-PDEs play in tailoring compartmentalised cAMP signals have been supported by studies using pharmacological inhibition of PDEs [12], siRNA silencing [13], transfection of dominant negative PDE isoforms [14], knockout mice [15], mathematical modelling [16] and peptide interference [17], other concepts such as local changes in viscosity and structural impediments to free cAMP diffusion have been offered as explanations for selective activation of cAMP effectors following receptor activation [18].
stimulation [19]. The stimuli for activation, however, are distinct, and include peptide hormones and nitric oxide. Peptide hormones typically activate membrane-bound GC (pGC), whereas nitric oxide activates the soluble form (sGC). Downstream effectors of this pathway include cyclic nucleotide gated cation channels and cGMP dependent protein kinase (PKG) (Figure 2). Hydrolysis of cGMP is catalysed by cGMP-PDEs. As with the cAMP system, compartmentalisation of enzymes that produce, are activated by and degrade cGMP is key to the signalling fidelity of this system [20].
Compartmentalised cGMP signalling Cyclic GMP is a similar second messenger to cAMP, in that both are cyclic nucleotides with comparable mechanisms of signalling. In an analogous manner to cAMP, cGMP is synthesised from GTP by the enzyme guanylate cyclase (GC) in response to G-protein coupled receptor www.sciencedirect.com
The action of cGMP-PDEs is pivotal for the regulation of spatial and temporal cGMP dynamics, with PDE5 and PDE2 having major roles. The PDE family member with the greatest influence appears to be determined by the cGMP source being considered. PDE5 and PDE2 limit Current Opinion in Genetics & Development 2014, 27:20–25
22 Developmental mechanisms, patterning and evolution
Figure 2
ion channels
Ras
Ras SNO
pGC
(e)
(b)
cGMP Phosphodiesterase
(c)
Active ERK 1/2
NO cGMP
NOS
(d)
(b)
sGC (a)
Protein modifications: e.g. Cysteine S-nitrosylation
PKG
Downstream events: e.g. regulation of cardiac contraction Current Opinion in Genetics & Development
Nitric oxide generation and signalling. Nitric oxide (NO) produced by nitric oxide synthase (NOS) drives downstream signalling by (A) promoting the modification of other signalling proteins or (B) via the activation of guanylate cyclase in the cytoplasm (sGC) or at the membrane (pGC). (C) sGC can result in events such as the spatially restricted activation of ERK MAP kinase cascade following the S-nitrosylation (SNO) of membrane-bound Ras [32]. pGC results in the production of cGMP, which activates (D) PKG and (E) cyclic nucleotide gated ion channels in order to trigger a variety of physiological alterations, including changes in cardiac contraction [24].
sGC and pGC-derived cGMP signals, respectively [21]. Functionally, this has had implications in understanding the regulation of cardiac contractility, as localised PDE5 activity is required to underpin specific sGC action (via PKG) in promoting b-adrenergic (b-AR) stimulation [22]. Indeed, pharmacological inhibition of this PDE5 pool promoted global cGMP increases that actually resulted in a retardation of b-AR signalling. In contrast, cGMP increases triggered by natriuretic peptide do not modulate b-AR responsiveness, once again suggesting that these different cGMP pools are relevant only in their defined location [23]. This also holds true for cGMP gradients, which are shaped by the action of PDE2. It appears that PDE2 modulates cardiac function through its localisation to lipid rafts in the plasma membrane. Such localisation, along with adenylate cyclase, b-AR, and nitric oxide synthase (NOS) is thought to underpin its coupling to b3-AR-dependent NOS/cGMP activation of PDE and subsequent cAMP hydrolysis [24]. Another level of compartmentalised cGMP regulation depends on the PKG mediated phosphorylation of Current Opinion in Genetics & Development 2014, 27:20–25
PDE and guanylate cyclase. A recent study showed that PKG activation limits the accumulation and intracellular diffusion of NO-stimulated, sGC-derived cGMP through the phosphorylation of PDE5 (an action which increases PDE5 hydrolytic activity) [25]. This mechanism is not surprising, considering the parallels with a similar ‘feedback’ loop, which involves PDE3 and PDE4 activation following phosphorylation by PKA. However, at the sarcolemmal membrane, this action can be countered by a PKG-dependent increase in ANP (natriuretic peptide receptor)-stimulated, pGC-derived, cGMP production, where PKG phosphorylates pGC directly, to trigger its activation. So in effect, PKG regulates both membrane and cytosolic cGMP dynamics but in opposite directions.
Compartmentalised NO signalling As NO is a gas, it cannot be considered as a typical second messenger, however it is an important signalling molecule involved in the regulation of a myriad of cellular processes ranging from vasodilation to respiration [26,27]. Produced by the action of NOS, NO is a lipophillic and short-lived www.sciencedirect.com
Compartmentalised signalling McCormick and Baillie 23
messenger, which mediates its responses through one of two mechanisms; cGMP production via sGC activation or chemical modification of proteins via reactive nitrogen species (e.g. S-nitrosylation) (Figure 2). Compartmentalisation of this highly diffusible molecule is conferred by the subcellular localisation of NO sources. In terms of S-nitrosylation, it has been noted that membrane localisation of NOS is one of the ways in which this signalling system is compartmentalised. eNOS, for example, traffics between the plasma membrane and Golgi apparatus in order to ensure proximity to appropriate target proteins [28]. Another level of regulation is conferred by the fact that Golgi and plasma membranelocated eNOS enzymes have differing thresholds of activation as well as NO production rates (less eNOS is released from Golgi), thus, the locality of the enzyme is doubly important in governing spatial and temporal response to NO [29,30]. Indeed, a recent study concluded that the greatest influence in the determination of functional outcomes for differently located NOS isoforms was the amount of NO produced [31]. The implications of NO compartmentalisation on cellular functioning are only now becoming realised. Indeed, it has been found recently that localised NO production ensures Ras activation is limited to the plasma membrane [32] (Figure 2). The study in question reported that bradykinin treatment in endothelial cells, selectively stimulates H-Ras at the plasma membrane via S-nitrosylation, an action which is independent of Src kinase action or EGFR activation. As only the membrane bound pool of Ras was activated, the authors suggested that the compartmentalisation of the subsequent downstream signal was underpinned by the restricted production of NO in localised areas. These findings are in agreement with other descriptions of NO sources, which have been restricted to defined organelles e.g. Golgi, nucleus [33]. Specifically, it is thought that intracellular S-nitrosylation of substrates is limited to the primary site of eNOS localisation, and this allows fine control of spatial physiological changes, for example initiation of intracellular transport processes.
development to muscle contraction [37,38]. Regulation of calcium compartmentalisation differs from that seen with cyclic nucleotides, as there is no enzymatic degradation of the molecule, so spatial/temporal control of Ca2+ concentration depends on only two parameters, site of release/ uptake and diffusion rate. Indeed, calcium can induce its own release from intracellular stores (Ca2+ induced Ca2+ release), an ability that can rapidly amplify cellular responses to small amounts of this second messenger. It can also be rapidly pumped back into calcium stores or sequestered by Ca2+ binding proteins [39]. As the diffusion rate of calcium is relatively slow (one tenth of cAMP), Ca2+ ATPase pumps/exchangers, which act to move large amounts of Ca2+ to and from intracellular locations, play the most important role in Ca2+ compartmentalisation. Their importance to life is substantiated by the fact that there are nine different types of Ca2+ ATPase pumps/ exchangers belonging to three multi-gene families and they exhibit differences in tissue/cell distribution and modes of regulation [40]. In the resting cell, concentrations of Ca2+ within the cytoplasm are kept low (
ScienceDirect Compartmentalisation of second messenger signalling pathways Kristie McCormick and George S Baillie The ability of a cell to transform an extracellular stimulus into a downstream event that directs specific physiological outcomes, requires the orchestrated, spatial and temporal response of many signalling proteins. The notion of compartmentalised signalling pathways was popularised in the 1980s by Brunton and colleagues, with their discovery that spatially segregated cAMP directs a variety of signalling responses in cardiomyocytes. It is now understood that compartmentalisation is a common mechanism used by all cells to ensure the interaction of signalling ‘second messenger’ molecules with localised ‘pools’ of appropriate effector proteins. In this way, the cell can elicit differential cellular responses by using a single, freely diffusible, molecular species. Recently, the compartmentalisation schemes employed by signalling systems involving cyclic nucleotides, calcium and nitric oxide have been elucidated and as a result, the varied range of functional consequences underpinned by such strategies can be better appreciated. Addresses Institute of Cardiovascular and Medical Sciences, CMVLS, Wolfson-Link Building, University of Glasgow, Glasgow G12 8QQ, UK Corresponding author: Baillie, George S ([email protected])
Current Opinion in Genetics & Development 2014, 27:20–25 This review comes from a themed issue on Developmental mechanisms, patterning and evolution Edited by Lee A Niswander and Lori Sussel
0959-437X/$ – see front matter, # 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.gde.2014.02.001
Introduction In order to translate the events occurring at the cell surface into an intracellular response, the cell enables specific molecules, so-called ‘second messengers’, as chemical relays to engage the signalling machinery within the cytoplasm. In short, cellular stimulation via cell surface receptors induces (often at the plasma membrane) transient production of signalling molecules that ‘find’ and bind to an appropriate effector protein. This action serves to initiate a signal cascade, which both amplifies the signal and directs it towards the alteration of a specific cellular process in response to the initial environmental threat or chemical cue on the outside of the cell. Despite the underlying mechanism being similar for each second messenger, the cellular responses are highly diverse, ranging from gene activation to cell division [1,2]. Current Opinion in Genetics & Development 2014, 27:20–25
A multiplicity of function also exists within each second messenger system, as a single chemical species may have multiple downstream effectors and regulate numerous physiological processes. Essentially, the cell achieves this feat by the spatio-temporal regulation, or compartmentalisation, of its second messengers to allow discrete pathways of localised signalling activity to occur. This review will discuss compartmentalisation with respect to four well-studied second messenger systems; cAMP, cGMP, calcium and nitric oxide (NO), concentrating on the way in which each system evokes its effects in a spatially restricted manner. The functional implications from recent groundbreaking studies will also be highlighted and discussed.
Compartmentalised cAMP signalling 30 ,50 -Cyclic adenosine monophosphate (cAMP) was the first second messenger to be discovered [3] and many of the concepts that we take for granted surrounding signal generation via effectors [4] and compartmentalisation [5] stem from seminal work on this molecule. Today, we understand that, when activated, a variety of different cell surface receptors coupled to Gas, educe equivalent increases in cAMP production that result in diverse cellular responses. This can only happen if the signalling ‘machinery’ for each process is spatially linked to the ‘pool’ of cAMP produced by adenylate cyclase in response to distinct receptors [6]. Such compartmentalisation allows anchored cAMP effectors such as Protein Kinase A (PKA) [7] and the exchange protein activated by cAMP (EPAC) [8] to sample cAMP transients and produce downstream signals when their activation threshold (in terms of cAMP concentration) is surpassed (Figure 1). The duration and strength of signals produced by cAMP effectors is heavily influenced by action of the only superfamily of enzymes that has evolved to degrade cAMP, the cAMP-phosphodiesterases (PDEs) [9]. The pivotal role of PDEs in shaping the spatial and temporal dimensions of cAMP ‘pulses’ produced in response to receptor activation was realised following the invention of a number of cAMP reporters that allowed real-time visualization of cAMP dynamics within cellular micro-domains [10]. These probes revealed spatially restricted cAMP gradients that were tailored by discretely anchored PDE populations to allow activation of small subsets of cAMP effectors. In effect, the PDEs acted as local cyclic nucleotide ‘sinks’ allowing ‘hotspots’ of cAMP to accumulate in areas of the cell that were devoid of PDEs (Figure 1). This action underpinned receptor specificity of action, as co-treatment with a non-specific PDE inhibitor and receptor agonist, allowed uniform, global increases in cAMP and concomitant activation of all cAMP effectors [11]. www.sciencedirect.com
Compartmentalised signalling McCormick and Baillie 21
Figure 1
Ligand (a)
(b)
Ligand
β
β
cAMP
cAMP
cAMP cAMP cAMP cAMP cAMP cAMP cAMP cAMP cAMP
(d)
PKA –
cAMP
cAMP cAMP
–
cAMP
–
+
os
ph cAM od P ie st er
–
EPAC
cAMP cAMP cAMP cAMP cAMP
cAMP
cAMP
+ cAMP
cAMP cAMP
cAMP cAMP cAMP
cAMP
cAMP cAMP cAMP
–
cAMP
cAMP
cAMP
terase
(c)
cAMP
(e)
Receptorspecific signal
cAMP
cAMP cAMP cAMP
–
e
s MP era cA iest d ho sp
+
c Phosp AMP hodie s
cAMP cAMP
cAMP
cAMP
PKA
cAMP cAMP cAMP
cAMP cAMP cAMP cAMP
cAMP
+
cAMP
–
cAMP
cAMP cAMP
e
–
cAMP Phosphodiesterase
cAMP
cAMP
cAMP
cAMP cAMP
+
–
as
cAMP
cAMP
–
cAMP cAMP cAMP
cAMP
cAMP
STOP
STOP
cAMP cAMP cAMP
cAMP
AC AC
cAMP
cAMP cAMP cAMP
GDP
Ph
cAMP cAMP cAMP
cAMP cAMP cAMP cAMP cAMP
–
γ αs
GTP
–
cAMP
GTP
–
cAMP
αs
+
αs
–
γ
+
cAMP
+ cAMP Phosphodiesterase
+ o
Ph
Current Opinion in Genetics & Development
Compartmentalisation of cAMP signalling. (A) The G-protein subunit Gas is released following receptor activation by the appropriate ligand. (B) The subunit binds adenylate cyclase (AC) to trigger production of cAMP. (C) Phosphodiesterases (PDEs) shape the cAMP gradient ensuring the specificity of signals emanating from defined receptors. (D) cAMP effectors, such as PKA and EPAC, that are in close proximity to a PDE ‘pool’ cannot be activated unless the associated PDE is ‘swamped’ by the cAMP produced in response to activation of the receptor. (E) The threshold of activation for cAMP effectors not in the vicinity of PDEs is more easily reached. These enzymes drive downstream signalling events that orchestrate physiological changes in response to the extracellular stimulus detected by cell-surface receptors.
Although the fundamental roles that cAMP-PDEs play in tailoring compartmentalised cAMP signals have been supported by studies using pharmacological inhibition of PDEs [12], siRNA silencing [13], transfection of dominant negative PDE isoforms [14], knockout mice [15], mathematical modelling [16] and peptide interference [17], other concepts such as local changes in viscosity and structural impediments to free cAMP diffusion have been offered as explanations for selective activation of cAMP effectors following receptor activation [18].
stimulation [19]. The stimuli for activation, however, are distinct, and include peptide hormones and nitric oxide. Peptide hormones typically activate membrane-bound GC (pGC), whereas nitric oxide activates the soluble form (sGC). Downstream effectors of this pathway include cyclic nucleotide gated cation channels and cGMP dependent protein kinase (PKG) (Figure 2). Hydrolysis of cGMP is catalysed by cGMP-PDEs. As with the cAMP system, compartmentalisation of enzymes that produce, are activated by and degrade cGMP is key to the signalling fidelity of this system [20].
Compartmentalised cGMP signalling Cyclic GMP is a similar second messenger to cAMP, in that both are cyclic nucleotides with comparable mechanisms of signalling. In an analogous manner to cAMP, cGMP is synthesised from GTP by the enzyme guanylate cyclase (GC) in response to G-protein coupled receptor www.sciencedirect.com
The action of cGMP-PDEs is pivotal for the regulation of spatial and temporal cGMP dynamics, with PDE5 and PDE2 having major roles. The PDE family member with the greatest influence appears to be determined by the cGMP source being considered. PDE5 and PDE2 limit Current Opinion in Genetics & Development 2014, 27:20–25
22 Developmental mechanisms, patterning and evolution
Figure 2
ion channels
Ras
Ras SNO
pGC
(e)
(b)
cGMP Phosphodiesterase
(c)
Active ERK 1/2
NO cGMP
NOS
(d)
(b)
sGC (a)
Protein modifications: e.g. Cysteine S-nitrosylation
PKG
Downstream events: e.g. regulation of cardiac contraction Current Opinion in Genetics & Development
Nitric oxide generation and signalling. Nitric oxide (NO) produced by nitric oxide synthase (NOS) drives downstream signalling by (A) promoting the modification of other signalling proteins or (B) via the activation of guanylate cyclase in the cytoplasm (sGC) or at the membrane (pGC). (C) sGC can result in events such as the spatially restricted activation of ERK MAP kinase cascade following the S-nitrosylation (SNO) of membrane-bound Ras [32]. pGC results in the production of cGMP, which activates (D) PKG and (E) cyclic nucleotide gated ion channels in order to trigger a variety of physiological alterations, including changes in cardiac contraction [24].
sGC and pGC-derived cGMP signals, respectively [21]. Functionally, this has had implications in understanding the regulation of cardiac contractility, as localised PDE5 activity is required to underpin specific sGC action (via PKG) in promoting b-adrenergic (b-AR) stimulation [22]. Indeed, pharmacological inhibition of this PDE5 pool promoted global cGMP increases that actually resulted in a retardation of b-AR signalling. In contrast, cGMP increases triggered by natriuretic peptide do not modulate b-AR responsiveness, once again suggesting that these different cGMP pools are relevant only in their defined location [23]. This also holds true for cGMP gradients, which are shaped by the action of PDE2. It appears that PDE2 modulates cardiac function through its localisation to lipid rafts in the plasma membrane. Such localisation, along with adenylate cyclase, b-AR, and nitric oxide synthase (NOS) is thought to underpin its coupling to b3-AR-dependent NOS/cGMP activation of PDE and subsequent cAMP hydrolysis [24]. Another level of compartmentalised cGMP regulation depends on the PKG mediated phosphorylation of Current Opinion in Genetics & Development 2014, 27:20–25
PDE and guanylate cyclase. A recent study showed that PKG activation limits the accumulation and intracellular diffusion of NO-stimulated, sGC-derived cGMP through the phosphorylation of PDE5 (an action which increases PDE5 hydrolytic activity) [25]. This mechanism is not surprising, considering the parallels with a similar ‘feedback’ loop, which involves PDE3 and PDE4 activation following phosphorylation by PKA. However, at the sarcolemmal membrane, this action can be countered by a PKG-dependent increase in ANP (natriuretic peptide receptor)-stimulated, pGC-derived, cGMP production, where PKG phosphorylates pGC directly, to trigger its activation. So in effect, PKG regulates both membrane and cytosolic cGMP dynamics but in opposite directions.
Compartmentalised NO signalling As NO is a gas, it cannot be considered as a typical second messenger, however it is an important signalling molecule involved in the regulation of a myriad of cellular processes ranging from vasodilation to respiration [26,27]. Produced by the action of NOS, NO is a lipophillic and short-lived www.sciencedirect.com
Compartmentalised signalling McCormick and Baillie 23
messenger, which mediates its responses through one of two mechanisms; cGMP production via sGC activation or chemical modification of proteins via reactive nitrogen species (e.g. S-nitrosylation) (Figure 2). Compartmentalisation of this highly diffusible molecule is conferred by the subcellular localisation of NO sources. In terms of S-nitrosylation, it has been noted that membrane localisation of NOS is one of the ways in which this signalling system is compartmentalised. eNOS, for example, traffics between the plasma membrane and Golgi apparatus in order to ensure proximity to appropriate target proteins [28]. Another level of regulation is conferred by the fact that Golgi and plasma membranelocated eNOS enzymes have differing thresholds of activation as well as NO production rates (less eNOS is released from Golgi), thus, the locality of the enzyme is doubly important in governing spatial and temporal response to NO [29,30]. Indeed, a recent study concluded that the greatest influence in the determination of functional outcomes for differently located NOS isoforms was the amount of NO produced [31]. The implications of NO compartmentalisation on cellular functioning are only now becoming realised. Indeed, it has been found recently that localised NO production ensures Ras activation is limited to the plasma membrane [32] (Figure 2). The study in question reported that bradykinin treatment in endothelial cells, selectively stimulates H-Ras at the plasma membrane via S-nitrosylation, an action which is independent of Src kinase action or EGFR activation. As only the membrane bound pool of Ras was activated, the authors suggested that the compartmentalisation of the subsequent downstream signal was underpinned by the restricted production of NO in localised areas. These findings are in agreement with other descriptions of NO sources, which have been restricted to defined organelles e.g. Golgi, nucleus [33]. Specifically, it is thought that intracellular S-nitrosylation of substrates is limited to the primary site of eNOS localisation, and this allows fine control of spatial physiological changes, for example initiation of intracellular transport processes.
development to muscle contraction [37,38]. Regulation of calcium compartmentalisation differs from that seen with cyclic nucleotides, as there is no enzymatic degradation of the molecule, so spatial/temporal control of Ca2+ concentration depends on only two parameters, site of release/ uptake and diffusion rate. Indeed, calcium can induce its own release from intracellular stores (Ca2+ induced Ca2+ release), an ability that can rapidly amplify cellular responses to small amounts of this second messenger. It can also be rapidly pumped back into calcium stores or sequestered by Ca2+ binding proteins [39]. As the diffusion rate of calcium is relatively slow (one tenth of cAMP), Ca2+ ATPase pumps/exchangers, which act to move large amounts of Ca2+ to and from intracellular locations, play the most important role in Ca2+ compartmentalisation. Their importance to life is substantiated by the fact that there are nine different types of Ca2+ ATPase pumps/ exchangers belonging to three multi-gene families and they exhibit differences in tissue/cell distribution and modes of regulation [40]. In the resting cell, concentrations of Ca2+ within the cytoplasm are kept low (
