Phosphoinositides and Other Glycerophospholipids Interact with Protein C Inhibitor (PCI) and Modify Its Activity towards Activated Protein C (APC).

Abstract 2122

Poster Board II-99

Protein C Inhibitor (PCI, SERPINA5, PAI3) is a non-specific, secreted serine protease inhibitor (serpin) which circulates at low levels (5μg/mL or 90nM) in blood plasma (review: Geiger 2007, Suzuki 2008). First described as an inhibitor of activated protein C (APC) an anticoagulant serine protease in human plasma, it has been shown that PCI inactivates a variety of other proteases and has a wide tissue distribution.

Some compounds like glycosaminoglycans (e.g. heparin, heparan sulfate) and certain phospholipids can modify PCI activity. Recently it was shown that single-stranded DNA aptamers stimulates the inhibitory activity of PCI towards APC in a glycosaminoglycan-like fashion (Müller 2009).

In 2007 Malleier et al. analyzed the interaction of PCI with phosphatidylserine (PS), oxidized PS (OxPS) and oxidized phosphatidylethanolamin (OxPE). PS, OxPS and OxPE bind to PCI and enhance the stimulation of APC-inhibition 130 to 190-fold. In addition, PE supports the internalization of PCI by cells, and internalized PCI promotes phagocytosis of bacteria (Baumgärtner et al. 2007). JFC1 (synaptotagmin-like protein 1) was identified by our group as a new intracellular interaction partner of PCI and colocalization of PCI and CSN6, a subunit of the COP9 signalosome could be observed in lymphocytes.

Here, we analyzed the interaction of PCI with phosphatidic acid (PA), phosphatidylglycerol (PG), cardiolipin (CL), phosphoinositides and derived second messengers like inositol-1,4,5-trisphosphate (IP₃) and diacylglycerol (DAG). To identify lipid-regions which are important for PCI binding we were focusing on differences in fatty acid composition and headgroup phosphorylation. The binding was studied by native PAGE, protein overlay assays (dot blot analysis) or ELISA. The stimulation of PCI activity towards APC was analyzed in functional assays using a low molecular weight substrate (S-2366).

IP₃ and inositol-1,3,4,5-tetrakisphosphate (IP₄) did neither interact with PCI nor stimulate its activity towards APC.

PCI bound to saturated, unsaturated and oxidized PA. The oxidized form of 1-palmitoyl-2-arachidonoyl-phosphatidic acid (OxPAPA) exhibited lower binding to PCI, but higher stimulatory activity on APC inhibition, as compared to unoxidized PAPA. Saturated dipalmitoyl-PA did not modulate PCI activity.

From all studied lipids 1-palmitoyl-2-arachidonoyl-PG (PAPG) had the strongest stimulatory effect on APC-inhibition similar to 0.1μM low molecular weight heparin. Oxidation of PAPG led to a slight decrease and saturation to a complete loss in stimulatory activity.

Also tetra-oleoyl-CL bound to PCI with high affinity and had a similar effect as oxidized PAPG and OxPAPA.

All mono- and diphosphorylated phosphoinositides as well as phosphatidylinositol-3,4,5-triphosphate (PI3,4,5P₃) bound to PCI as judged from binding assays. A mobility shift of PCI antigen on native PAGE was observed when PCI was incubated with phosphatidylinositol-3,5-diphosphate and phosphatidylinositol-4,5-diphosphate (PI4,5P₂).

Therefore different phospholipids modulate the activity of PCI, but on the other hand PCI may as well effect lipid signaling. As PI4,5P₂ plays an important role as substrate for PI3-kinase we will take a closer look at the effect of PCI on the PI3K/PTEN system and on the activation of AKT (PKB) by PDK1. We will also study the influence of PCI on the generation of IP₃ and DAG and its possible role in calcium signaling and protein kinase activation.

So far we conclude that phosphoinositides and other glycerophospholipids may function as additional intracellular interaction partners of PCI.