Novel factors affecting clearance of triacylglycerol-rich lipoproteins from blood

Abstract: Apolipoprotein (apo) A-V is the most recently discovered member of a protein family responsible for the structure and metabolic fate of plasma lipoproteins. While most of the apolipoproteins are well characterized with regard to structure, interactions and function, the role of apoA-V is not well understood. ApoA-V is synthesized only in liver and is present in blood at much lower concentration than the other apolipoproteins. Although apoA-V is firmly established as an important determinant for plasma triacylglycerol (TG) metabolism, the mechanism is unclear. ApoA-V has been suggested to act through 1) an intracellular mechanism affecting lipoprotein assembly and secretion, 2) direct or indirect activation of lipoprotein lipase (LPL), or 3) interaction with endocytotic lipoprotein receptors.Two other novel players involved in the clearance of lipoproteins are angiopoietin-like protein (ANGPTL) 3 and 4. Previous studies have shown that the coiled-coil domain (ccd) of ANGPTL3 and -4 can inactivate lipoprotein lipase (LPL). The functional site of action of LPL is at the capillary endothelium, but the enzyme is synthesized mostly in adipocytes and myocytes and has to be transported by trancytosis to the luminal side of endothelial cells. Both ANGPTLs are present in tissues and in the circulating blood, but it is not known were the inactivation of LPL normally takes place.The aim of this thesis was to investigate the mechanism by which apoA-V exerts its effect on TG metabolism and to investigate in further detail how ANGPTLs act on the LPL system.Binding of apoA-V to receptors involved in lipoprotein metabolism was investigated by surface plasmon resonance technique (SPR). ApoA-V was found to bind to the LDL receptor related protein 1 (LRP1) and to the mosaic type 1 receptor sorLA. Binding could be competed by receptor associated protein (RAP) or by heparin, and was calcium dependent. We concluded that apoA-V binds to the LA-repeats of these receptors. In further experiments apoA-V was shown to increase binding of TG-rich chylomicrons to the receptors. This demonstrated a possible mechanism for the TG-lowering effect of apoA-V in vivo.A putative binding region in apoA-V for heparin and receptors was investigated by site-directed mutagenesis. Two positively charged amino acid residues were changed (Arg210Glu/Lys211Gln), resulting in decreased binding to heparin and to LRP1 and thus the localization of one important functional region in apoA-V.Since the receptor sorLA also contains a Vsp10p domain, another Vsp10p domain family member, sortilin, was investigated. ApoA-V was found to interact also with this receptor. In experiments with human embryonic kidney cells transfected with sorLA or sortilin, apoA-V was found to bind to cell surfaces and to be rapidly internalized while co-localized with the receptors on the way to lysosomes for degradation.Additional apoA-V mutants, identified in patients with severe hypertriglyceridemia, were investigated with regard to effects in vitro on LPL activity and receptor binding. The most severe mutants displayed null binding to LRP1, whereas the effect on LPL activity was retained. These results suggest that lack of receptor interaction mirrors the loss of biological function in a better way than the in vitro effect on LPL activity.We noted that ccd-ANGPTL3 and -4 did not prevent the LPL-mediated uptake of chylomicron-like lipoproteins in primary murine hepatocytes. Therefore LPL activity was measured after pre-incubation with ccd-ANGPTL3 or 4 in the presence or absence of TG-rich lipoproteins. Physiological concentrations of lipoproteins were found to protect LPL from inactivation by ccd-ANGPTLs. Investigation by SPR demonstrated that the ccd-ANGPTLs did not bind to the lipoproteins. Other experiments showed that less than 1% of ANGPTL4 in human serum was bound to TG-rich lipoproteins. This implies that the known binding of LPL to TG-rich lipoproteins stabilizes the enzyme and protects it from inactivation by ANGPTLs. We conclude that the normal levels of ANGPTLs in plasma are too low to affect the LPL-system and that inactivation of the enzyme by ANGPTLs is more likely to occur locally in the extracellular interstitium of tissues where LPL is en route to its endothelial binding sites and where the concentrations of the TG-rich lipoproteins are low.