Role of Low Density Lipoprotein Receptor in the Clearance of Coagulation Factor VIII In Vivo.

Abstract

Low Density Lipoprotein Receptor (LDLR) is the archetype of the family of endocytic receptors that also includes Low-density lipoprotein Receptor-related Protein (LRP), Megalin, and Very Low Density Lipoprotein Receptor (VLDLR). While most of these receptors bind a variety of ligands, LDLR has restricted specifity. Its known ligands are apolipoprotein E (apoE) and apolipoprotein B100 (apoB100). Ligand binding to the LDL-receptor family is inhibited by the Receptor-Associated Protein (RAP). We have previously reported that RAP overexpression elevates plasma levels of factor VIII (FVIII) in both normal mice and mice with hepatic LRP deficiency [Bovenschen et al. (2003) Blood 101, 3933–3939]. This implies that LRP, but also another RAP-sensitive mechanism contributes to the regulation of FVIII in vivo. This study addresses the question whether LDLR, despite its restricted ligand specificity, binds FVIII and contributes to its clearance. In vitro binding was studied using a recombinant LDLR fragment spanning the extracellular domain of complement-type repeats 1–7. The purified fragment efficiently bound to immobilized ApoE and apoB-100 containing LDL. The immobilized LDLR fragment also bound human FVIII, with half-maximal binding at 156 nM FVIII, and binding was inhibited by RAP. Human von Willebrand Factor (VWF) or non-activated factor IX did not bind to the LDLR fragment. The relevance of the FVIII-LDLR interaction was assessed in vivo employing LDLR−/− mice, cre/loxP-mediated conditional LRP-deficient mice (LRP−), and mice with the combined deficiency. Plasma FVIII levels of controls, LDLR−/− and LRP− mice were 1.1, 0.9 and 1.8 U/ml, respectively. This suggests that LRP, but not LDLR regulates FVIII in plasma. Surprisingly, however, mice that combined LDLR deficiency with hepatic LRP deficiency displayed much higher FVIII levels (median value 4.6 U/ml) than mice lacking LRP alone. This suggests that LDLR does have the potential of regulating FVIII levels. LDLR−/− LRP− mice further displayed elevated levels of VWF (median value 3.3 U/ml), but not of factor V or factor IX. The possibility was considered that FVIII levels were elevated secondary to the profound changes in lipoprotein profiles. To this end, we also examined ApoE deficient mice, which have reduced LDL, and mice that overexpress ApoC1, which is associated with elevated levels of cholesterol- and triglyceride VLDL. ApoE−/− LDLR−/−LRP− mice had a median FVIII level of 4.2 U/ml, which is close to that of LDLR−/−LRP− mice. Mice that overexpressed human ApoC1 had elevated levels of cholesterol and triglycerides, but 0.5 U/ml FVIII. This demonstrates that elevated FVIII levels were independent of lipoprotein levels. The role of LDLR and LRP in FVIII clearance were further studied by analyzing the pharmacokinetics of human FVIII. In normal mice the Mean Residence Time (MRT) was 160 min [68% confidence intervals (CI) 117–218 min]. MRT was 200 [CI 154–259] min in LDLR−/− mice, and 263 [CI 206–336] min in LRP− mice. This confirms the previously described role of LRP in FVIII clearance. Strikingly, in LDLR−/−LRP− mice the MRT of FVIII was 760 [691–826] min, which is approximately 5-fold longer than in control mice. These data demonstrate that LRP and LDLR act in concert in regulating FVIII levels in plasma. In the absence of LDLR, LRP maintains normal FVIII levels, while hepatic LRP deficiency is largely compensated by LDLR. This regulatory role of LDLR represents a novel link between LDLR and the hemostatic system.