Compartmentation is a strategy that enables cells to produce multiple distinct signals using a single second messenger. Cyclic AMP is one example of a second messenger that exhibits local signals (restricted to the site of production) and global signals that propagate throughout the cell. In the cardiac myocyte, β1 and β2 adrenoceptors (AR) signal via global and local cAMP signals respectively (Steinberg, 1999). Through cAMP-dependent protein kinase A (PKA), β1 AR simulation promotes phosphorylation of a range of targets including those distant from the sarcolemma such as phospholamban (PLB) and troponin I (TnI). These targets are not accessible to local cAMP signals arising from β2 AR stimulation. It is generally accepted that coupling of the β2 AR with Gi limits the size and propagation of the β2 AR cAMP signal (Steinberg, 1999). We have shown that caveolae, specialised invaginated lipid rafts, are central to the size and spatial characteristics of β2 AR cAMP-dependent signals in the adult cardiac myocyte. When caveolae are disrupted by the cholesterol-depleting agent methyl-β-cyclodextrin (MBCD), the inotropic response to β2 (but not β1) AR stimulation is markedly enhanced (Calaghan & White, 2006). This suggests that caveolae are responsible for restriction of the magnitude/spread of the β2 AR cAMP signal. Furthermore, caveolar disruption and abolition of Gi signalling with pertussis toxin (PTX) have identical, non-additive effects on β2 AR responsiveness, implying that caveolar control is mediated via Gi (Calaghan & White, 2006). Our current work focuses on characterising the extent of the spatial control of cAMP by caveolae, and identifying the caveolae-dependent components which are responsible for this. We have employed genetically encoded FRET-based cAMP biosensors to index cAMP in different cellular compartments: a PKA type II probe to detect cAMP in (AKAP-tethered) PKA II domains and a probe based on the nucleotide binding domain of Epac2, which is expressed throughout the cytosol, to detect cAMP in all compartments (Warrier et al., 2007). Using 10 µM zinterol with 300 nM CGP20712A to selectively stimulate β2 AR, we have shown that disruption of caveolae with MBCD results in an increase in PKA probe activity in response to β2 AR stimulation, with no effect on Epac2 probe responses. In parallel, we have mapped the pattern of protein phosphorylation. In control cells, β2 AR stimulation produces no detectable increase in phosphorylation at Ser16 PLB, Ser2809 RyR or Ser22,23 TnI. However, when caveolae are disrupted, β2 AR stimulation promotes a robust increase in phosphorylation of PLB, with no change in phosphorylation of RyR or TnI. We conclude that caveolae disruption increases cAMP in some domains containing PKA II (which include PLB) but that this represents only a very small proportion of the total global compartment. Our data suggests that caveolae preferentially control access of the β2 AR cAMP signal to a sub-compartment of the sarcoplasmic reticulum containing PLB. We are currently exploring the contribution of specific phosphodiesterases to this. Alternations in spatial control of cAMP can have profound consequences for cardiac myocyte function. Our work in this field led us to consider whether the HMG CoA inhibitors statins may regulate cAMP signalling through effects on cholesterol-dependent caveolae. Culturing adult cardiac myocytes in the presence of simvastatin (10 µM) for 2 days depletes cellular/caveolar cholesterol and caveolin 3, and reduces caveolae density by 30%. Statin-treated cells show enhanced β2 (but not β1) AR responsiveness and PTX does not further potentiate the β2 AR response, implying that statins’ effects are mediated through Gi. Furthermore, simvastatin treatment promotes phosphorylation of PLB in response to β2 AR stimulation (2.5-fold over baseline). This is the first description of the effects of statins on the β-responsiveness of the adult cardiac myocyte and demonstrates that, in vitro, these drugs have the capacity to increase the spread of the β2 AR cAMP signal. Our data suggest that this is due, in part at least, to disruption of normal caveolar control of cAMP (although a contribution from the isprenoid pathway has not been excluded). This mechanism could contribute to pleiotropic effects of statins by enhancing the contractile reserve of the failing heart in which the functional relevance of β1/β2 AR signalling shifts in favour of β2 AR. In conclusion, the caveolar signalosome plays a key role in defining the spatial characteristics of cAMP in the adult cardiac myocyte. This highlights a potential mechanism by which changes in caveolae (seen in cardiac disease, with drug treatment) could have profound consequences for the chronotropic, inotropic and lusitropic function of the heart.