Real-time monitoring of cAMP in brown adipocytes reveals differential compartmentation of β(1) and β(3)-adrenoceptor signalling

OBJECTIVE: 3′,5′-Cyclic adenosine monophosphate (cAMP) is a central second messenger governing brown adipocyte differentiation and function. β-adrenergic receptors (β-ARs) stimulate adenylate cyclases which produce cAMP. Moreover, cyclic nucleotide levels are tightly controlled by phosphodiesterases (PDEs), which can generate subcellular microdomains of cAMP. Since the spatio-temporal organisation of the cAMP signalling pathway in adipocytes is still unclear, we sought to monitor real-time cAMP dynamics by live cell imaging in pre-mature and mature brown adipocytes. METHODS: We measured the real-time dynamics of cAMP in murine pre-mature and mature brown adipocytes during stimulation of individual β-AR subtypes, as well as its regulation by PDEs using a Förster Resonance Energy Transfer based biosensor and pharmacological tools. We also correlated these data with β-AR stimulated lipolysis and analysed the expression of β-ARs and PDEs in brown adipocytes using qPCR and immunoblotting. Furthermore, subcellular distribution of PDEs was studied using cell fractionation and immunoblots. RESULTS: Using pre-mature and mature brown adipocytes isolated from transgenic mice expressing a highly sensitive cytosolic biosensor Epac1-camps, we established real-time measurements of cAMP responses. PDE4 turned out to be the major PDE regulating cytosolic cAMP in brown preadipocytes. Upon maturation, PDE3 gets upregulated and contributes with PDE4 to control β(1)-AR-induced cAMP. Unexpectedly, β(3)-AR initiated cAMP is resistant to increased PDE3 protein levels and simultaneously, the control of this microdomain by PDE4 is reduced upon brown adipocyte maturation. Therefore we postulate the existence of distinct cAMP pools in brown adipocytes. One cAMP pool is formed by β(1)-AR associated with PDE3 and PDE4, while another pool is centred around β(3)-AR and is much less controlled by these PDEs. Functionally, lower control of β(3)-AR initiated cAMP by PDE3 and PDE4 facilitates brown adipocyte lipolysis, while lipolysis activated by β(1)-AR and is under tight control of PDE3 and PDE4. CONCLUSIONS: We have established a real-time live cell imaging approach to analyse brown adipocyte cAMP dynamics in real-time using a cAMP biosensor. We showed that during the differentiation from pre-mature to mature murine brown adipocytes, there was a change in PDE-dependent compartmentation of β(1)-and β(3)-AR-initiated cAMP responses by PDE3 and PDE4 regulating lipolysis.