LAUSR

The elegant study of Kopperud et al. in this issue of Acta Physiologica entitled ‘Increased microvascular permeability in mice lacking Epac1 (Rapgef3)’ addresses the role of Epac1 for maintenance of endothelial barrier properties. During the last decade, the regulation of endothelial barrier properties has turned out to be complex. Nevertheless, it is well established that the second messenger cAMP is required and protective against barrier breakdown induced by inflammatory mediators in vivo (Michel & Curry 1999). A central mechanism underlying cAMP-mediated permeability regulation is activation of the small GTPase Rac1, which has evolved as a key regulator of endothelial adherens junctions (AJ) (Schlegel & Waschke 2014). Therefore, it is not surprising that inflammatory mediators compromise endothelial barrier integrity at least in part by reducing endothelial cAMP. Since 2005, it is known that, besides cAMP-dependent protein kinase (PKA), the exchange protein activated by cAMP (Epac1) acting as an GTP exchange factor (GEF) to activate Rap1 GTPase also is involved in endothelial barrier stabilization and is capable of stabilizing the binding of the main endothelial AJ adhesion protein VE-cadherin (Kooistra et al. 2005). In addition, Rac1 activation was shown to be mediated by PKAand Epac1dependent mechanisms via recruitment of the Rac GEFs Tiam1 and Vav2 to AJs (Schlegel & Waschke 2014). However, so far it was not entirely clear whether Epac1 was required for the maintenance of endothelial barrier functions in vivo. Therefore, the main finding of the study of Kopperud et al. is that mice lacking Epac1, but not mice deficient for Epac2, have increased albumin clearance in several tissues and organs and display enhanced gadomer-17 clearance in microvessels of masseter muscles indicative of compromised endothelial barrier integrity. As the phosphodiesterase 4 (PDE4) inhibitor rolipram was not sufficient to reduce permeability in Epac1-deficient mice compared with wild-type littermates, Epac1-mediated barrier stabilization appears to be critical and not to be outbalanced by PKA. This is interesting because PKA and Epac1 both form signalling complexes with VE-cadherin, suggesting that spatiotemporal regulation of AJ by cAMP signalling is important. In VE-cadherin-based signalling complexes, Epac1 was shown to associate with PKA, A-kinase-anchoring protein AKAP220 and phosphodiesterase 4D (PDE4D) (Rampersad et al. 2010, Radeva et al. 2014). In this complex, Epac1 appears to be required for the formation or maintenance of intercellular junctions because the number of junctions in Epac1-deficient endothelia was reduced. This observation is in line with a previous study showing that the activation of Epac1 signalling by O-Me-cAMP enhanced the complexity of intercellular overlap and the number of junctional plaques (Spindler et al. 2011). In contrast to experiments when cAMP levels were enhanced by treatment with rolipram in combination with adenylyl cyclase activation by forskolin, the effect of Epac1 stimulation was found to be strictly dependent on the activation of the GTPase Rac1. Taken together, all these data outline a scenario in which cAMP via Epac1-mediated activation of the GTPase Rac1 is critical for endothelial barrier maintenance by the formation of complex intercellular junctions and strengthening of VE-cadherinmediated adhesion. It needs to be determined why PKA cannot compensate for Epac1 deficiency although it is capable of activating Rac1. This may be mediated by Rac1-independent mechanisms specifically engaged by Epac1. The second important result of the study is that in Epac1-deficient mice ANP fails to further enhance baseline microvascular permeability. Both albumin clearance in different tissues such as skin and skeletal muscle and gadomer-17 clearance in masseter muscle were significantly increased following the application of ANP in wild-type but not Epac1-deficient animals. As ANP enhances permeability via cGMP formation (Kuhn et al. 2009), this observation can be caused by a mechanism in which ANP, by increasing endothelial cGMP levels, activates PDE2 and thereby reduces local cAMP and Epac1 activities. This also would explain why Epac1 deficiency increases permeability to a similar extent as ANP.