Plasma PCSK9 preferentially reduces liver LDL receptors in mice

Abstract

Proprotein convertase subtilisin/kexin type 9 (PCSK9) is a secreted protein that regulates the expression of LDL receptor (LDLR) protein. Gain-of-function mutations in PCSK9 cause hypercholesterolemia, and loss-of-function mutations result in lower plasma LDL-cholesterol. Here, we investigate the kinetics and metabolism of circulating PCSK9 relative to tissue levels of LDLRs. The administration of recombinant human PCSK9 (32 microg) to mice by a single injection reduced hepatic LDLRs by approximately 90% within 60 min, and the receptor levels returned to normal within 6 h. The half-life of the PCSK9 was estimated to be approximately 5 min. Continuous infusion of PCSK9 (32 microg/h) into wild-type mice caused a approximately 90% reduction in hepatic LDLRs within 2 h and no associated change in the level of LDLR in the adrenals. Parallel studies were performed using a catalytically inactive form of PCSK9, PCSK9(S386A), and similar results were obtained. Infusion of PCSK9(D374Y), a gain-of-function mutation, resulted in accelerated clearance of the mutant PCSK9 and a greater reduction in hepatic LDLRs. Combined, these data suggest that exogenously administrated PCSK9 in plasma preferentially reduces LDLR protein levels in liver at concentrations found in human plasma and that PCSK9's action on the LDLR is not dependent on catalytic activity in vivo.

Fig. 1.

Hepatic LDL receptor (LDLR) protein levels in mice injected with recombinant human proprotein convertase subtilisin/kexin type 9 (PCSK9). A: Immunoblot analysis of the LDLR and transferrin receptor (TFR) in liver membranes of male wild-type mice 1, 2, or 4 h following the injection of 0, 2, 4, 8, 16, or 32 μg of recombinant human PCSK9 protein. Membrane proteins were isolated, equal aliquots from three mouse livers in each group were pooled, and a total of 5 μg protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR and TFR as described in Materials and Methods. B: Immunoblot analysis of the LDLR and TFR protein in liver membranes of wild-type male mice 5, 10, 30, 60, 120, 240, 360, and 480 min after the injection of 32 μg of recombinant human PCSK9. Membrane proteins were isolated, equal aliquots from three mouse livers at each time point were pooled, and immunoblots were performed as described in A. C: Clearance of ¹²⁵I-PCSK9 in wild-type and Ldlr^−/− mice. Recombinant human PCSK9 labeled with ¹²⁵I was produced as described in Materials and Methods. Six mice of each genotype were injected with 4 μg of the ¹²⁵I-labeled PCSK9 protein, and blood samples were obtained at the indicated times following the injection. Values are averages ± SD. Insert: Plasma half-life (min) of ¹²⁵I-PCSK9 calculated from the slope during the first 10 min. * P < 0.05 versus wild-type. D: ¹²⁵I-labeled PCSK9 uptake in livers of wild-type and Ldlr^−/− mice. Seven male mice of each genotype were administered 4 μg of the ¹²⁵I-labeled PCSK9 protein, and the mice were euthanized 15 min after injection. Livers were harvested after whole-body perfusion with saline for 30 min, and radioactivity was counted. Values are averages ± SD. * P < 0.05 versus wild-type.

Fig. 2.

Plasma PCSK9 concentrations and hepatic LDLR protein levels in wild-type mice infused with recombinant human PCSK9. A: Human PCSK9 plasma concentrations in male wild-type mice during the 3 h infusion of 8, 16, 24, or 32 μg/h of recombinant human PCSK9. Plasma human PCSK9 concentrations were measured by ELISA (10). Data are averages ± SEM, n = 4. B: Immunoblot analysis of the LDLR and TFR in liver membranes of male wild-type mice infused for 3 h with saline vehicle or 8, 16, 24, or 32 μg/h of recombinant human PCSK9 protein. A liver sample was obtained prior to infusion (−) and at the end of the infusion (+). Membrane proteins were isolated, and 5 μg protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. TFR was used as a control for loading. Immunoblots show protein levels from individual mice before and after infusion.

Fig. 3.

Plasma PCSK9 concentrations and hepatic LDLR protein levels in Pcsk9^−/− mice infused with recombinant human PCSK9. A: Human PCSK9 plasma concentrations in male Pcsk9^−/− mice during a 3 h infusion of 4, 8, 16, or 32 μg/h of recombinant human PCSK9. Plasma human PCSK9 concentrations were measured by ELISA (10). Data are averages ± SEM, n = 3. B: Immunoblot analysis of the LDLR and TFR protein in liver membranes of male Pcsk9^−/− mice infused for 3 h with 4, 8, 16, or 32 μg/h of recombinant human PCSK9. A liver sample was obtained prior to infusion (−) and at the end of the infusion (+). Membrane proteins were isolated, and 5 μg protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. TFR was used as a control for loading. Immunoblots show protein levels from individual mice before and after infusion.

Fig. 4.

Liver and adrenal LDLR protein levels in wild-type mice infused with recombinant human PCSK9 and in noninfused wild-type and Pcsk9^−/− mice. A: Immunoblot analysis of the LDLR and TFR in liver membranes of male wild-type mice infused for 3 h or 6 h with saline vehicle or 32 μg/h of recombinant human PCSK9 protein. Membrane proteins were isolated, and 5 μg of protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. TFR protein was used as a control for loading. Immunoblots show protein levels from individual mice after infusion. B: Immunoblot analysis of the LDLR and TFR in membranes from adrenal glands of male wild-type mice infused for 3 h or 6 h with saline vehicle or 32 μg/h of recombinant human PCSK9 protein. Adrenal gland membrane proteins were isolated as described in Materials and Methods. Immunoblots show protein levels from pooled adrenal glands from three mice after infusion. C: Quantitative immunoblot analysis of the LDLR and TFR in liver and adrenal gland membranes of wild-type (^+/+) and Pcsk9^−/− (^−/−) mice. Membrane proteins were isolated, and 5 μg was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. TFR protein was used as a control for loading. Secondary antibodies labeled with IRDye 680 were used for visualization. Detection and quantification of LDLR and TFR protein were performed with the LI-COR Odyssey Infrared Imaging System. Ratiometric analysis of wavelength intensities was used to quantify the LDLR and TFR membrane protein. Data are from pooled membrane fractions from 18 male wild-type and 18 male Pcsk9^−/− mice. Similar results were obtained from two smaller studies that used livers and adrenals pooled from six mice per genotype.

Fig. 5.

Plasma half-life and activity of recombinant gain-of-function human PCSK9(D374Y) protein in mice. A: Clearance of ¹²⁵I-wild-type-PCSK9 and gain-of-function ¹²⁵I-PCSK9(D374Y) in wild-type mice. Recombinant human wild-type-PCSK9 or PCSK9(D374Y) labeled with ¹²⁵I was produced as described in Materials and Methods. Five mice of each genotype were injected with 4 μg of the ¹²⁵I-labeled protein, and blood samples were obtained at the indicated times following the injection. Values are averages ± SD. Insert: Plasma half-life (min) of ¹²⁵I-wild-type-PCSK9 and ¹²⁵I-PCSK9(D374Y) calculated from the slope during the first 10 min. * P < 0.05, PCSK9(D374Y) versus wild-type-PCSK9. B: Human PCSK9 plasma levels during a 3 h infusion of 4, 8, or 32 μg/h of recombinant human PCSK9(D374Y) in male Pcsk9^−/− mice. Plasma human PCSK9 concentrations were determined by ELISA (10). Data are averages ± SEM, n = 3. C: Immunoblot analysis of the LDLR and TFR protein in liver membranes of male Pcsk9^−/− mice infused with 4, 8, or 32 μg/h of recombinant human PCSK9(D374Y) protein for 3 h. Liver membrane proteins were isolated, and 5 μg protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. TFR was used as a control for loading. Immunoblots show protein levels from individual mice before and after infusion.

Fig. 6.

Hepatic LDLR protein levels in mice infused with catalytically inactive recombinant trans-PCSK9(S386A) protein. Male Pcsk9^−/− mice were infused with saline, 32 μg/h of recombinant human wild-type-PCSK9, or 32 μg/h of trans-PCSK9(S386A) for 3 h. A liver sample was obtained prior to infusion (−) and at the end of the infusion (+). Liver membrane proteins were isolated, and 5 μg protein was subjected to 8% SDS-PAGE for immunoblot analysis of the LDLR as described in Materials and Methods. The TFR protein was used as a control for loading. Immunoblots show protein levels from individual mice before and after infusion. Plasma human PCSK9 concentrations were determined by ELISA from blood samples obtained at the end of the 3 h infusion (10). ND, not detected.