Based on a variety of single-cell measurements, the notion that cAMP microdomains exist in cells is being increasingly embraced. The cellular and molecular underpinnings of this organization are also steadily being revealed. A dependence of Ca2+-sensitive ACs (adenylate cyclases) in HEK-293 cells (human embryonic kidney cells) on capacitative Ca2+ entry is enforced by their presence in lipid rafts and protein–protein interactions. In these cells, many of the participants in the cAMP cascade, including AC, phosphodiesterase 4, cAMP-dependent protein kinase [PKA (protein kinase A)] and protein phosphatase 2A, are now seen to be involved in higher order assemblies. Moreover, the presence of Na+/H+ exchanger 1 in these domains creates a microclimate, protected against global swings in cellular pH. The Ca2+-stimulatable AC8, which is targeted to these regions, can sequester calmodulin for its own regulatory purposes. These devices are a sampling of the multiple layers of organization that are in place – even in a simple cell – to ensure faithful and economical communication of the cAMP message.

Introduction

The notion that the elements of cAMP signalling pathways are concentrated in microdomains, to allow efficient signal transduction, is superseding the previously accepted model of random collision between signalling components. The molecular details and cellular underpinnings of this organization are steadily being revealed, a growing number of participants involved in these signalling complexes are being identified, and the environmental regulatory factors that maintain a privileged microclimate in these cAMP domains are progressively being established. Below, we summarize developing concepts under these broad headings.

Dependence of Ca2+-sensitive ACs (adenylate cyclases) on CCE (capacitative Ca2+ entry)

Our interest in cAMP microdomains stemmed from our investigation of the dependence of Ca2+-sensitive ACs on discrete forms of Ca2+ entry. In non-excitable cells, Ca2+-sensitive ACs are regulated by CCE, which is triggered by the depletion of intracellular Ca2+ stores [13]. Moreover, Ca2+-sensitive ACs are selective for CCE over other types of Ca2+ entry, such as ionophore- and arachidonic acidmediated Ca2+ entry [24].

Caveolar microdomains of the plasma membrane, enriched in cholesterol and sphingolipids [5], play a part in allowing the signalling molecules of the Ca2+ and cAMP pathways to be concentrated. Fractionation studies have revealed that Ca2+-sensitive ACs are confined to caveolae [6], whereas Ca2+-insensitive ACs are excluded from these structures [7]. Residence of Ca2+-sensitive ACs, both endogenous and transfected, in cholesterol-enriched caveolae is essential, although not sufficient, for their regulation by CCE [6,8]. Thus co-localization in caveolae provides a basis for the selective regulation of Ca2+-sensitive ACs by Ca2+ entry occurring through CCE channels.

Additional evidence supports an intimate association between Ca2+-sensitive ACs and CCE channels. For example, in HEK-293 cells (human embryonic kidney cells), BAPTA (bis-(o-aminophenoxy)ethane-N,N,N′,N′-tetra-acetic acid), a fast chelator of Ca2+, prevents the stimulation of endogenously expressed Ca2+-sensitive ACs by CCE, whereas at equivalent concentrations, EGTA, a slower chelating agent, is ineffective [3]. The differential efficacy of these two chelators is indicative of a short diffusional distance between the source of Ca2+ and its target, which further reinforces the juxtaposition between Ca2+-sensitive ACs and CCE channels.

Although the molecular identity of the CCE channel remains unknown, the putative channel complex may conceivably include non-channel proteins, by analogy with the voltage-gated Ca2+ channel Cav1.2, which has been reported to form a complex with an AC, a β2-adrenergic receptor, a G-protein, cAMP-dependent protein kinase [PKA (protein kinase A)] and the counterbalancing PP2A (protein phosphatase 2A) [9]. The attractive possibility that ACs may form an integral subunit of the CCE channel complex would elegantly account for the close association of the enzyme with the source of CCE and the selectivity of CCE in regulating Ca2+-sensitive ACs (Figure 1).

Organization of a cAMP microdomain

Figure 1
Organization of a cAMP microdomain

AC8 segregates to caveolae and is closely associated with the store-operated channels (SOC). CNG channels (CNGC) are localized in caveolae and can be used to measure local cAMP concentrations. Whereas OAG-activated channels (OAC) also localize to caveolae, they are unable to activate AC8 and may segregate to separate rafts. The β2-adrenergic receptor is reported to interact with an AC, a G-protein, PKA and PP2A. Moreover, PP2A interacts with the N-terminus of AC8. AKAPs both interact with Ca2+-sensitive ACs and target PDEs to the cAMP microdomain. NHE1 creates a microclimate by protecting caveolae from swings in cellular pH. Structural barriers, such as the endoplasmic reticulum (ER), may contribute to restrict the cAMP microdomain.

Figure 1
Organization of a cAMP microdomain

AC8 segregates to caveolae and is closely associated with the store-operated channels (SOC). CNG channels (CNGC) are localized in caveolae and can be used to measure local cAMP concentrations. Whereas OAG-activated channels (OAC) also localize to caveolae, they are unable to activate AC8 and may segregate to separate rafts. The β2-adrenergic receptor is reported to interact with an AC, a G-protein, PKA and PP2A. Moreover, PP2A interacts with the N-terminus of AC8. AKAPs both interact with Ca2+-sensitive ACs and target PDEs to the cAMP microdomain. NHE1 creates a microclimate by protecting caveolae from swings in cellular pH. Structural barriers, such as the endoplasmic reticulum (ER), may contribute to restrict the cAMP microdomain.

Since Ca2+-sensitive AC activity is a discerning measure of CCE, AC8, an archetypal Ca2+-stimulable member of this family, can function as a sensitive and specific sensor for CCE. AC8 has previously been used to discriminate between the capacitative and arachidonate-activated pathways of Ca2+ entry in HEK-293 cells [4]. Thus AC8 can be used as a tool to investigate candidate CCE channels. In this regard, a subset of TRP (transient receptor potential) subunits, TRPC3, TRPC6 and TRPC7, can be activated by diacylglycerol and its synthetic analogue, OAG (1-oleyl-2-acetyl-sn-glycerol). In recent experiments, we explored the ability of these TRP subunits, reported to assemble into OAG-activated channels, to regulate AC8. It turned out that although OAG and CCE evoked a Ca2+ rise of equivalent magnitude, only the latter stimulated AC8 [9a]. The inability of OAG-activated Ca2+ entry to activate AC8 indicates that this type of Ca2+ entry is mediated by channels that are distinct from the elusive CCE channels, even though they may share common TRP subunits [10]. This observation further underlines the dependence of Ca2+-sensitive ACs on CCE.

Interaction between PP2A and AC8

The finding of a strict dependence of Ca2+-sensitive ACs on CCE led us to search for binding partners that would confer this selectivity. A recent yeast two-hybrid study using the N-terminus of AC8 as bait did not detect any interaction with expected targets, for example the candidate elements of the CCE apparatus. However, both the catalytic and the scaffolding subunit of PP2A interacted with the N-terminus of AC8 in a manner that was competitive with calmodulin [11]. Although the regulatory significance of this interaction is not immediately obvious, all of the ACs possess putative phosphorylation sites, and indeed some have been shown to be phosphorylated [12]. In this light, the presence of PP2A in close proximity to AC8 may be required to transduce the cAMP signal and suggests the existence of a regulatory cluster centred on AC.

Complex between AC, PDE4 (phosphodiesterase 4) and AKAP (A-kinase-anchoring protein)

The awareness of microdomains brings with it the implicit assumption that, in these domains, the local cAMP concentration differs from the concentration in the broad cytosol. Demonstrating this assumption experimentally requires techniques that selectively monitor cAMP in these domains. For this purpose, we developed CNG (cyclic nucleotide-gated) channels to report sub-plasmalemmal concentrations of cAMP [13,14]. By application of this technique, in combination with RNA interference methods to suppress candidate partners, we now find that, in HEK-293 cells, most of the participants in the cAMP cascade, including AC, PDE4D, PKA and an AKAP (as well as PP2A, see above), are involved in higher order assemblies (Figure 1). These findings are discussed in more detail elsewhere [14a].

Protected pH in microclimate

Cellular pH is a major issue that governs the performance of many metabolic and regulatory pathways. Particularly susceptible to pH swings are Ca2+-sensitive enzymes, such as the Ca2+-sensitive ACs, whose activity can be substantially increased or decreased by a change of 0.3 pH unit, a shift that could readily occur during cellular activity [15]. Regulation by pH of Ca2+-sensitive ACs is mediated either by the catalytic domain of the ACs (in the case of AC5 and AC6) [16] or by calmodulin (in the case of AC1 and AC8) [17], in which the degree of ionization of aspartate residues governs the ability of Ca2+ to be distinguished from Mg2+. Consequently, it is not surprising that Ca2+-sensitive ACs, while being highly sensitive to small changes in pH in in vitro measurements, are effectively insulated from such changes in the intact cell. This insulation is at least partially due to the presence of NHE1 (Na+/H+ exchanger 1) in caveolae, which buffers the climate of these microenvironments against global swings in cellular pH [18]. The expected sensitivity to small changes in pH is revealed by inhibition of NHE1 in the intact cell.

Calmodulin sequestration

The Ca2+-binding protein calmodulin mediates the regulation by Ca2+ of numerous target proteins. However, calmodulin availability in the cell is limiting, such that the different targets must compete with one another for the restricted pool [19]. Hence, it is of critical importance that AC8, which is activated by Ca2+-bound calmodulin, can sequester calmodulin for its regulatory purposes from the available cytosolic pool. In particular, the N-terminus binds calmodulin avidly in the face of competition from numerous other targets (Figure 2). When calmodulin binds Ca2+, following Ca2+ entry, it undergoes a conformational change that allows it to bind to the C-terminus of AC8, displacing the autoinhibitory domain and activating the enzyme [20]. As well as favouring AC8 over other, lower-affinity calmodulin targets, the pre-association of calmodulin with AC8 may allow the enzyme to respond rapidly to Ca2+ entry.

The N-terminus of AC8 tethers calmodulin (CaM) and sequesters it from a limited pool

Figure 2
The N-terminus of AC8 tethers calmodulin (CaM) and sequesters it from a limited pool

Calmodulin regulates many targets, such as Ca2+/CaM-dependent protein kinase II (CaMKII), neuronal nitric oxide synthase (nNOS) and calcineurin, which compete with one another for binding to the limited CaM pool. Moreover, the presence of small neuronal IQ motif proteins (SNIQ), which are devoid of catalytic activity, is thought to further regulate the availability of CaM. AC8 requires Ca2+-bound calmodulin for its activation, and, under resting conditions, the N-terminus of AC8 sequesters CaM from the available pool. Upon Ca2+ entry, for example through a voltage-gated Ca2+ channel (VGCC), CaM undergoes a conformational change. Ca2+-bound CaM is then able to bind to the C-terminus of AC8, displacing the autoinhibitory domain and resulting in activation of the enzyme.

Figure 2
The N-terminus of AC8 tethers calmodulin (CaM) and sequesters it from a limited pool

Calmodulin regulates many targets, such as Ca2+/CaM-dependent protein kinase II (CaMKII), neuronal nitric oxide synthase (nNOS) and calcineurin, which compete with one another for binding to the limited CaM pool. Moreover, the presence of small neuronal IQ motif proteins (SNIQ), which are devoid of catalytic activity, is thought to further regulate the availability of CaM. AC8 requires Ca2+-bound calmodulin for its activation, and, under resting conditions, the N-terminus of AC8 sequesters CaM from the available pool. Upon Ca2+ entry, for example through a voltage-gated Ca2+ channel (VGCC), CaM undergoes a conformational change. Ca2+-bound CaM is then able to bind to the C-terminus of AC8, displacing the autoinhibitory domain and resulting in activation of the enzyme.

Conclusion

Based on this brief overview, we may begin to understand that ‘compartmentalization’ relies on multiple layers of organization, which ensure, in concert, that the cAMP message is faithfully and economically communicated and regulated. The challenge for the future is daunting, not just in terms of identifying and validating direct protein or lipid partners for regulatory molecules, but also determining the range of elements that ensure that these microclimates function optimally to enable these interactions.

Compartmentalization of Cyclic AMP Signalling: Biochemical Society Focused Meeting held at King's College, Cambridge, U.K., 29–30 March 2006. Organized by D. Cooper (Cambridge, U.K.), M. Houslay (Glasgow) and M. Zaccolo (Padua, Italy). Edited by D. Cooper.

Abbreviations

     
  • AC

    adenylate cyclase

  •  
  • AKAP

    A-kinase-anchoring protein

  •  
  • CCE

    capacitative Ca2+ entry

  •  
  • CNG channel

    cyclic nucleotide-gated channel

  •  
  • HEK-293 cell

    human embryonic kidney cell

  •  
  • NHE

    Na+/H+ exchanger

  •  
  • OAG

    1-oleyl-2-acetyl-sn-glycerol

  •  
  • PDE

    phosphodiesterase

  •  
  • PKA

    protein kinase A

  •  
  • PP2A

    protein phosphatase 2A

  •  
  • TRP

    transient receptor potential

References

References
1
Chiono
M.
Mahey
R.
Tate
G.
Cooper
D.M.F.
J. Biol. Chem.
1995
, vol. 
270
 (pg. 
1149
-
1155
)
2
Fagan
K.A.
Mahey
R.
Cooper
D.M.F.
J. Biol. Chem.
1996
, vol. 
271
 (pg. 
12438
-
12444
)
3
Fagan
K.A.
Mons
N.
Cooper
D.M.F.
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
9297
-
9305
)
4
Shuttleworth
T.J.
Thompson
J.L.
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
31174
-
31178
)
5
Simons
K.
Ikonen
E.
Nature (London)
1997
, vol. 
387
 (pg. 
569
-
572
)
6
Smith
K.E.
Gu
C.
Fagan
K.A.
Hu
B.
Cooper
D.M.F.
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
6025
-
6031
)
7
Crossthwaite
A.J.
Seebacher
T.
Masada
N.
Ciruela
A.
Dufraux
K.
Schultz
J.E.
Cooper
D.M.F.
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
6380
-
6391
)
8
Fagan
K.A.
Smith
K.E.
Cooper
D.M.F.
J. Biol. Chem.
2000
, vol. 
275
 (pg. 
26530
-
26537
)
9
Davare
M.A.
Avdonin
V.
Hall
D.D.
Peden
E.M.
Burette
A.
Weinberg
R.J.
Horne
M.C.
Hoshi
T.
Hell
J.W.
Science
2001
, vol. 
293
 (pg. 
98
-
101
)
9a
Martin
A.C.
Cooper
D.M.F.
Mol. Pharmacol.
2006
 
10
Zagranichnaya
T.K.
Wu
X.
Villereal
M.L.
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
29559
-
29569
)
11
Crossthwaite
A.J.
Ciruela
A.
Rayner
T.F.
Cooper
D.M.F.
Mol. Pharmacol.
2006
, vol. 
69
 (pg. 
608
-
617
)
12
Lin
T.H.
Lai
H.L.
Kao
Y.Y.
Sun
C.N.
Hwang
M.J.
Chern
Y.
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
15721
-
15728
)
13
Fagan
K.A.
Rich
T.C.
Tolman
S.
Schaack
J.
Karpen
J.W.
Cooper
D.M.F.
J. Biol. Chem.
1999
, vol. 
274
 (pg. 
12445
-
12453
)
14
Rich
T.C.
Fagan
K.A.
Nakata
H.
Schaack
J.
Cooper
D.M.F.
Karpen
J.W.
J. Gen. Physiol.
2000
, vol. 
116
 (pg. 
147
-
161
)
14a
Willoughby
D.
Cooper
D.M.F.
Biochem. Soc. Trans.
2006
, vol. 
34
 (pg. 
468
-
471
)
15
Willoughby
D.
Schwiening
C.J.
J. Physiol.
2002
, vol. 
544
 (pg. 
487
-
499
)
16
Hu
B.
Nakata
H.
Gu
C.
De Beer
T.
Cooper
D.M.F.
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
33139
-
33147
)
17
Cooper
D.M.F.
Biochem. J.
2003
, vol. 
375
 (pg. 
517
-
529
)
18
Willoughby
D.
Masada
N.
Crossthwaite
A.J.
Ciruela
A.
Cooper
D.M.F.
J. Biol. Chem.
2005
, vol. 
280
 (pg. 
30864
-
30872
)
19
Persechini
A.
Stemmer
P.M.
Trends Cardiovasc. Med.
2002
, vol. 
12
 (pg. 
32
-
37
)
20
Simpson
R.E.
Ciruela
A.
Cooper
D.M.F.
J. Biol. Chem.
2006