Targeting Compartmentalization of PDE4

The existence of a myriad of PDE4 isoforms that are targeted to discrete sub-cellular locations and that have distinct regulatory properties means that PDE4s are key players in compartmentalizing cAMP signaling and regulating input from other signaling pathways (Baillie and Houslay 2005). The ability of PDE4s to be strategically targeted and anchored throughout the cell can be largely attributed to associations with other proteins. These protein-protein interactions are mediated mainly, but not exclusively, by the unique N-terminal regions of PDE4 isoforms (Houslay and Adams 2003; Huston et al. 2006b).

2.4.1 Role of the N-Terminal Region of PDE4s in Intracellular Targeting and cAMP Compartmentalization

As discussed in Sect. 2.1, a unique N-terminal region defines each PDE4 isoform. Evidence from our laboratory has shown unequivocally that this N-terminal region is essential in the intracellular targeting of PDE4 isoforms. Pivotal in this work has been the study of the super-short PDE4A, PDE4A1. PDE4A1 is unique in the PDE4 family in that it is entirely membrane bound and found, in a number of cell types, to be predominantly localized to the Golgi and to vesicles underlying the plasma membrane (Shakur et al. 1993, 1995; Pooley et al. 1997). However, removal of its isoform-unique 25-amino-acid N-terminal region generates a fully active and entirely soluble cytosolic form (Shakur et al. 1993). This discovery led to the proposal that intracellu-lar targeting was pivotal to the PDE4 family and so formed an inherent part of the observed isoform multiplicity. As such, PDE4A1 has provided a particularly useful model in underpinning the notion of PDE4 intracellular targeting.

NMR analysis has revealed the N-terminal region of PDE4A1 is composed of 25 amino acids that form two distinct helical structures bound by a flexible and mobile hinge region (Smith et al. 1996). It was observed in the original work by Shakur et al. (1993) that engineered constructs lacking this N-terminal region formed entirely soluble, fully active species. In addition, it was shown that when the N-terminal region of PDE4A1 was bound to an entirely cytosolic protein, such as CAT or GFP, the resultant chimeric protein was entirely membrane bound (Scotland and Houslay 1995; Smith et al. 1996; Baillie et al. 2002). These studies formed the paradigm for the importance of the N-terminal regions of the PDE4 enzymes in directing intracellular targeting and thereby allowing exquisite control of the shape and formation of gradients of cAMP within the cell, the basis of cAMP compartmentalization. The identification of the role of the N-terminal region of PDE4s in targeting and the knowledge that this region provides the signature of each individual PDE4 isoform led to the realization that this region could allow tailored compartmentalization of the cAMP signal through selective expression of PDE4 isoforms. Indeed, further studies identified the N-terminal regions of PDE4A5, PDE4A4 and PDE4D4 to interact with members of the SRC tyrosyl kinase family (O'Connell et al. 1996; Beard et al. 1999; McPhee et al. 1999), PDE4A5 to interact with the immunophilin XAP2 (Bolger et al. 2003b), PDE4D5 to interact with both RACK1 and P-arrestin (Yarwood et al. 1999; Steele et al. 2001; Bolger et al. 2002; Bolger et al. 2003a), PDE4B1 to interact with DISC1 (Millar et al. 2005) and PDE4D3 to interact with myomegalin, mAKAP and AKAP450 (Dodge et al. 2001; Tasken et al. 2001; Verde et al. 2001; McCahill et al. 2005). These interactions are discussed in Sect. 2.5 (and see Fig. 3).

Fig. 3 Summary of the known protein-protein interactions of various PDE4 isoforms shown in their appropriate locations within a hypothetical cell

The interactions of PDE4 N-terminal regions are not confined to protein-protein interactions. Indeed, within PDE4A1 it is a microdomain within the second helical structure named TAPAS1 that confers Ca2+-gated membrane insertion and shows a preferential binding for net -2 charge phosphatidic acid (Baillie et al. 2002). More recently, it has been shown that this TAPAS1/PA interaction is not sufficient for the targeting of PDE4A1 in intact cells. Indeed, while TAPAS1 provides the core insertion domain, a separate site located within helix 1 is responsible for efficiency of the insertion and also for the retention of PDE4A1 to the Golgi (Huston et al. 2006a). It is suggested that this second site within helix1 may interact with another protein, as this site is not a lipid interaction site. The presence of more than one binding site for efficient targeting of proteins is synonymous with a number of other PDE4 protein-protein interactions where a second site of interaction within the PDE enzyme has also been identified. Binding of PDE4A4 to LYN SH3 has been shown to be dependent on the interaction of LYN-SH3 with sites located within both the N-terminus of PDE4A4 and within the LR2 region of the protein (McPhee et al. 1999). Likewise, binding sites for both P-arrestin and RACK1 have recently been found within both the N-terminal regions and catalytic region of PDE4D5 (Bolger et al. 2006). That such a level of complexity exists within the interactions of PDE4 isoforms with their myriad of binding partners indicates that absolute definition of the molecular basis of these interactions will be of great value in the development of powerfully specific drug therapies.

0 0

Post a comment