BACKGROUND: PCSK9 (proprotein convertase subtilisin/kexin type 9) is a polymorphic gene whose protein product regulates plasma LDL cholesterol (LDLC) concentrations by shuttling liver LDL receptors (LDLRs) for degradation. PCSK9 variants that cause a gain or loss of PCSK9 function are associated with hyper- or hypocholesterolemia, which increases or reduces the risk of cardiovascular disease, respectively. We studied the clinical and molecular characteristics of a novel PCSK9 loss-of-function sequence variant in a white French-Canadian family.
METHODS: In vivo plasma and ex vivo secreted PCSK9 concentrations were measured with a commercial ELISA. We sequenced the PCSK9 exons for 15 members of a family, the proband of which exhibited very low plasma PCSK9 and LDLC concentrations. We then conducted a structure/function analysis of the novel PCSK9 variant in cell culture to identify its phenotypic basis.
RESULTS: We identified a PCSK9 sequence variant in the French-Canadian family that produced the PCSK9 Q152H substitution. Family members carrying this variant had mean decreases in circulating PCSK9 and LDLC concentrations of 79% and 48%, respectively, compared with unrelated noncarriers (n=210). In cell culture, the proPCSK9-Q152H variant did not undergo efficient autocatalytic cleavage and was not secreted. Cells transiently transfected with PCSK9-Q152H cDNA had LDLR concentrations that were significantly higher than those of cells overproducing wild-type PCSK9 (PCSK9-WT). Cotransfection of PCSK9-Q152H and PCSK9-WT cDNAs produced a 78% decrease in the secreted PCSK9-WT protein compared with control cells.
CONCLUSIONS: Collectively, our results demonstrate that the PCSK9-Q152H variant markedly lowers plasma PCSK9 and LDLC concentrations in heterozygous carriers via decreased autocatalytic processing and secretion, and hence, inactivity on the LDLR.
Familial hypercholesterolemia (FH)6 is characterized by high circulating concentrations of LDL particles and increases the risk of cardiovascular disease 1. Mutations in the LDLR7 (low density lipoprotein receptor), APOB [apolipoprotein B (including Ag(x) antigen)], and PCSK9 (proprotein convertase subtilisin/kexin 9) genes are linked with FH1, FH2, and FH3, respectively 1, 2. Recently, 2 other loci for hypercholesterolemia have been identified on chromosome 16q22.1 (FH4) and 8q24.22, although the cognate genes have not yet been defined 3, 4. The LDL receptor (LDLR) clears LDL particles from the circulation, whereas apolipoprotein B100 is the protein component of the LDL particle that interacts with the LDLR 5. PCSK9 is a secreted glycoprotein 6 that interacts with the LDLR and mediates its lysosome-dependent degradation 7–10. Largely produced in the liver and the intestine 6, PCSK9 is synthesized in the endoplasmic reticulum (ER) as a preproprotein of 692 amino acid residues 6. As a prerequisite for its exit from the ER, PCSK9 autocatalytically cleaves its prodomain at VFAQ152↓SIP10, 11. The approximately 14-kDa propeptide and the approximately 62-kDa PCSK9 then form a heterodimer, which transits through the Golgi apparatus, is secreted, and interacts with the LDLR 6, 10, 12.
The gene encoding PCSK9 is highly polymorphic. Two categories of PCSK9 sequence variants produce mild to moderate (and opposing) phenotypes. Gain-of-function sequence variants cause a reduction in the LDLR that leads to hypercholesterolemia 13 or to autosomal dominant hypercholesterolemia in cases of severe phenotypic variants 2, 14. PCSK9 loss-of-function sequence variants decrease LDLR degradation, thereby reducing LDL cholesterol (LDLC) concentrations 15–17. Cell culture and animal studies have established that the LDLR is the downstream effector for PCSK9 gain-of-function and loss-of-function activities at the protein level 10, 18–20. Longitudinal population studies have shown significant reductions in the risk of coronary artery disease in carriers of loss-of-function PCSK9 variants (88% and 47% for PCSK9-C679X and -R46L heterozygotes, respectively) 21.
The characterization of naturally occurring gain-of-function and loss-of-function human PCSK9 variants has increased our understanding of the cell biology and function of this secreted glycoprotein. This work has included identification of the amino acid residues important for PCSK9 autocatalytic processing, secretion, and biological activity—information that provides insight into the mechanism by which PCSK9 mediates LDLR degradation. The mechanisms of many PCSK9 variants remain unknown, however. In this study, we identified a novel PCSK9 sequence variant in a white French-Canadian family that is associated with low circulating LDLC concentrations. We carried out cell culture studies to characterize the molecular mechanism behind this loss-of-function PCSK9 phenotype.
Study Participants and Methods
POPULATION STUDY INDIVIDUALS
After obtaining written informed consent, we collected blood samples from all study participants after a 12-h fast and made clinical measurements according to study protocols approved by the ethics committees of the Ottawa Hospital Research Institute and Clinical Research Institute of Montreal. We obtained blood samples from 15 participants recruited to the Clinical Research Institute of Montreal. The comparison group consisted of 210 individuals recruited to the Ottawa Hospital Lipid Clinic. Participants underwent anthropometric measurements, including height and weight measurements. Body mass index was calculated as the weight in kilograms divided by the square of the height in meters.
MEASUREMENT OF PLASMA LIPIDS AND LIPOPROTEINS
Blood was collected into EDTA-containing Vacutainer tubes (BD), and plasma and blood leukocytes were obtained by centrifuging blood samples at 1560g for 10 min at 22 °C. Serum for lipid measurements was obtained by collecting blood into BD SST™ Vacutainer tubes, allowing the blood sample to clot at room temperature for 20 min, and centrifuging the sample at 1560g for 10 min at 22 °C. Total cholesterol and triglycerides were measured with enzymatic methods on an Ortho Clinical Diagnostics Vitros 250 analyzer. HDL cholesterol was measured by a direct enzymatic method (Beckman Coulter) on the Synchron LX20 Pro analyzer (Beckman Coulter); the LDLC concentration was calculated with the Friedewald equation.
Genomic DNA was isolated from blood leukocytes with the QIAamp DNA Blood Kit (Qiagen). The primers and PCR conditions used for amplifying individual PCSK9 exons were as described by Abifadel et al. 2. Standard DNA-sequencing services were carried out by Bio Basic.
MEASUREMENT OF PLASMA PCSK9
The plasma PCSK9 concentration was quantified with a human PCSK9 ELISA from CycLex. This assay has an intraassay CV of 1.5%–2.6% and an interassay CV of 2.9%–7.1%. All samples were quantified 4 times.
CONSTRUCTS AND ANTIBODIES
The cDNA of human PCSK9 was cloned into the pIRES2-EGFP vector with a C-terminal V5 tag, as previously described 6. Mutations were introduced by site-directed mutagenesis, also as described 22. The mouse anti-V5 IgG used for immunoprecipitation and immunoblotting of V5-tagged recombinant PCSK9 was obtained from Invitrogen. The anti-PCSK9 antibody used for immunoblotting (anti–IB PCSK9 Ab) was produced by recombinant PCSK9 vaccination 7. The rabbit anti–human LDLR IgG and the mouse anti–transferrin receptor IgG were from Cedarlane. Secondary antimouse and antirabbit IgGs were from Amersham/GE Healthcare Life Sciences.
CELL CULTURE, TRANSFECTION, AND SAMPLE COLLECTION
HuH7 cells were grown at 37 °C in Dulbecco's modified Eagle medium containing 100 mL/L fetal bovine serum and 28 mg/L gentamicin. At 24 h after plating, we transiently transfected 3 × 105 cells with cDNA encoding human PCSK9 (pIRES2) according to the standard protocol for Effectene® (Qiagen) 6. Media from cultures of HuH7 cells transiently transfected with cDNA encoding human PCSK9 or a nonrelevant DNA control were collected 48 h later in the presence of a Complete Mini Protease Inhibitor Cocktail (Roche) and a phosphatase inhibitor (200 μmol/L sodium orthovanadate) and centrifuged at 13 000g for 3 min. Cell lysis was carried out in 1× RIPA buffer (50 mmol/L Tris, pH 7.6, 150 mmol/L NaCl, 10 mL/L NP-40, 5 g/L deoxycholate, 1 g/L SDS) in the presence of the inhibitors mentioned above. The protein concentrations of total cell lysates were measured with the Bradford dye-binding method (Bio-Rad Protein Assay Kit; Bio-Rad Laboratories).
Proteins were electrophoresed through a 7% NuPAGE Tris-acetate gel (Invitrogen), electroblotted onto nitrocellulose, and immunoblotted according to a standard protocol. The primary anti-PCSK9 antibody used for immunoblotting (anti–IB PCSK9 Ab-03) 7 was raised in rabbits against recombinant PCSK9 amino acid residues 31–454 and used at a dilution of 1 part in 2000. The primary anti-V5 antibody (Invitrogen) was used at the same dilution, and the secondary antibody was used at a dilution of 1 part in 5000. Immunoblots were revealed by chemiluminescence with Western Lightning Plus (PerkinElmer) on X-Omat film (Kodak). The Chemigenius 2XE imager and GeneTools software (Syngene) were used for densitometric quantification of signals.
The results of quantification of secreted PCSK9 by ELISA were expressed as the mean and SE (n=4, human plasma; n=3, spent media). LDLR was quantified via immunoblotting followed by densitometry analysis (n=3). Representative immunoblots are provided (see below). The unpaired Student t-test was used for statistical analyses of differences. P values <0.05 were considered statistically significant.
IDENTIFICATION OF THE NOVEL PCSK9-Q152H LOSS-OF-FUNCTION VARIANT IN A FRENCH-CANADIAN FAMILY
We identified an individual in a French-Canadian family with a very low circulating PCSK9 concentration (Fig. 1; see Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol57/issue10), compared with concentrations in a general white Canadian population (n=210; 68.5 μg/L vs 326.9 μg/L), as well as low LDLC concentrations (14th percentile, adjusted for age and sex). We sequenced the 12 PCSK9 exons and the exon–intron boundaries in this individual. She carried a missense mutation at base pair 456 (G→C) in exon 3 that yielded a proPCSK9 amino acid substitution (Q152H) at the P1 site of autocatalytic cleavage (Fig. 2A). We recruited members of the family and sequenced their PCSK9 exons and exon–intron boundaries (Fig. 1; see Table 1 in the online Data Supplement). This table also shows the fasting plasma lipid concentrations and PCSK9 concentrations for these individuals, as well as the mean values for the 210 white Canadians. The data for additional characteristics (age, sex, glucose concentration, body mass index) are also presented in Table 1 in the online Data Supplement.
Four individuals in the pedigree carried the Q152H mutation (Fig. 1; see Table 1 in the online Data Supplement). Their LDLC concentrations were 64.4% (III.2), 68.8% (III.3), 25.7% (II.4), and 33.1% (II.9) lower than those of unrelated individuals (see Table 1 in the online Data Supplement). These individuals' circulating PCSK9 concentrations were 84.3% (III.2), 87.3% (III.3), 79.0% (II.4), and 64.7% (II.9) lower than those of the general white Canadian population (see Table 1 in the online Data Supplement). Although the Q152H carriers and the general population had mean plasma triglyceride and HDL cholesterol concentrations that did not differ significantly [triglycerides, 135.8 mg/dL and 138.2 mg/dL (1.53 mmol/L vs 1.56 mmol/L), respectively (P=0.94); HDL cholesterol, 54.7 mg/dL and 46.4 mg/dL (1.42 mmol/L vs 1.20 mmol/L), respectively (P=0.14)], Q152H carriers had significantly lower plasma concentrations of total cholesterol and LDLC than the general population [154.6 mg/dL vs 213.6 mg/dL (4.00 mmol/L vs 5.53 mmol/L) (P=0.0069) and 72.8 mg/dL vs 140.1 mg/dL (1.91 mmol/L vs 3.63 mmol/L) (P=0.0031), respectively]. After adjustment for age and sex, the plasma LDLC concentrations for 3 of the 4 Q152H carriers were below the fifth percentile, and the proband was at the 14th percentile (Fig. 1; see Table 1 in the online Data Supplement). Noncarriers of the Q152H mutation within the family had LDLC concentrations that ranged from the fifth percentile to the 90th percentile, after adjustment for age and sex (Fig. 1; see Table 1 in the online Data Supplement).
Sequencing of the 12 PCSK9 exons in this family also revealed that 6 members (II.2, II.5, II.7, II.8, III.5, III.6; Fig. 1) carried a Leu insertion (c.43_44insCTG and denoted L10ins) within a stretch of 9 Leu residues in the signal peptide for PCSK9. This insertion was associated with lower LDLC concentrations in a white population 17. Two other members (II.6, III.4) carried both the L10ins and the R46L sequence variants within the PCSK9 propeptide, which are also associated with PCSK9 loss of function 21 (Fig. 1; see Table 1 in the online Data Supplement). In fact, several population studies have found that the risk of cardiovascular disease is decreased by approximately 50% for PCSK9-R46L heterozygotes, the LDLC concentrations of which are reduced by approximately 14% on average, compared with age- and sex-matched controls 21, 23, 24. The PCSK9-L10ins variant is associated with an approximately 14% reduction in LDLC concentrations in white populations 17, but it is not associated with a significant lowering in the LDLC concentration in individuals of African descent 25. Three individuals (II.2, III.2, III.3) also carried the PCSK9-I474V variant, which has been found in several other populations. The I474V variant is not associated with any changes in the LDLC concentration 25, 26. The fact that several members of this family carried multiple loss-of-function PCSK9 variants may indicate that these variants may be more frequent in some French-Canadian cohorts than in the general population.
BIOSYNTHESIS AND FUNCTIONAL ANALYSIS OF THE LOSS-OF-FUNCTION PCSK9-Q152H VARIANT
Fig. 2A depicts the domain structure of preproPCSK9, its sites of posttranslational modifications (sulfation at Y38, phosphorylation at S47 and S688, and glycosylation at N533), the residues surrounding the site of prosegment cleavage, the catalytic residues (D186, H226, S386), and the oxyanion hole residue (N317). Secreted PCSK9 can interact with the LDLR and enter with it into the endocytic recycling pathway, thereby decreasing the rate of LDLR recycling and increasing lysosome-dependent LDLR degradation 7–9, 27.
To compare the clinical lipoprotein profiles and plasma PCSK9 measurements of individuals carrying the novel PCSK9-Q152H variant, we investigated the biosynthesis and secretion of PCSK9 and its effect on LDLR degradation. Western blotting results for V5-tagged PCSK9 (Fig. 2B) show the distribution of proPCSK9 and intracellular PCSK9 (PCSK9 in panel B) in total cell lysates from liver HuH7 cells transiently transfected with wild-type (WT) PCSK9 or PCSK9-Q152H (lanes 1 and 2, respectively). The Q152H amino acid substitution at position 152 greatly reduced the ability of proPCSK9 to undergo autocatalytic cleavage, compared with WT (lane 2). Immunoprecipitation and subsequent immunoblotting procedures revealed secretion of PCSK9-WT into the medium but detected no PCSK9-Q152H secretion into the medium (Fig. 2B, lanes 3 and 4, respectively). This reduced proprotein processing and loss of secretion produced LDLR concentrations (Fig. 2C) that were not significantly higher (relative concentration, approximately 1.4) than those in mock-transfected control cells (relative concentration, 1; P=0.25); however, these LDLR concentrations were significantly higher (relative concentration, 1.4) than those in cells transfected with PCSK9-WT (relative concentration, approximately 0.4; P=0.03). The inset in Fig. 2C shows representative immunoblots for the LDLR and the transferrin receptor control.
THE PCSK9-Q152H VARIANT DECREASES WT SECRETION IN CELL CULTURE
In addition to blocking cleavage of the proPCSK9 zymogen, the Q152H variant also affected PCSK9-WT processing and secretion (Fig. 3, A and B, respectively), and this effect protected LDLR from degradation in cell culture (Fig. 3C). Fig. 3A shows equal amounts of PCSK9 secreted from HuH7 cells transfected with 250 ng of either PCSK9-WT (untagged) or PCSK9-WT-V5 (V5-tagged) cDNA (lanes 1 and 2, respectively). Lanes 1 and 2 of Fig. 3B show the intracellular processing of proPCSK9 to PCSK9. In contrast, PCSK9-Q152H-V5 was not secreted (Fig. 3A, lane 3). Cotransfection of PCSK9-WT with increasing amounts of PCSK9-WT-V5 was recompensed with increased PCSK9 secretion (Fig. 3A, lanes 4–6), and both proforms were processed as in single transfections (compare lanes 1 and 2 with lanes 3–5 in Fig. 3B). In contrast, cotransfection of PCSK9-WT with increasing amounts of PCSK9-Q152H-V5 significantly decreased PCSK9-WT secretion (Fig. 3A, lanes 7–9), even at the highest ratio of WT cDNA to Q152H cDNA (4:1) (Fig. 3A; compare lanes 1 and 7; P=0.002). Cotransfection of equal amounts of PCSK9-Q152H-V5 with PCSK9-WT significantly decreased PCSK9 secretion by 78% and 90% (Fig. 3A; compare lane 1 or 6 with lane 9; P=0.0002 and 0.0004, respectively), whereas immunoblotting showed that the intracellularly processed form of PCSK9-WT was reduced (Fig. 3B; compare iPCSK9-no tag lane 1 or 6 with lane 9). Lanes 1–9 of Fig. 3C show a representative immunoblot of LDLR in HuH7 cells transiently transfected with PCSK9-WT, PCSK9-WT-V5, or PCSK9-Q152H-V5. As expected, the LDLR concentration decreased with cotransfection of increasing amounts of PCSK9-WT-V5 with PCSK9-WT (Fig. 3C, lanes 4–6); however, lanes 7–9 of Fig. 3C show that cotransfection of increasing amounts of PCSK9-Q152H-V5 protected the LDLR from degradation by PCSK9-WT. Transfections with 250 ng of PCSK9-WT significantly decreased LDLR concentrations relative to those for control cells (Fig. 3C, lane 1 vs lane 3; P=0.05); however, cotransfection in the presence of either 125 ng or 250 ng of Q152H (Fig. 3C, lanes 8 and 9) significantly increased the relative LDLR concentration to 0.94 (P=0.03) and 1.1 (P=0.02), respectively, compared with PCSK9-WT (0.79; lane 1). This approximately 40%–50% increase in relative LDLR concentration from that expected for WT can be attributed at least partially to the decrease in secreted PCSK9-WT in the presence of the Q152H variant. In Fig. 1 in the online Data Supplement, we demonstrate the linearity of PCSK9 secretion for the amounts of PCSK9-WT cDNA (untagged) and PCSK9-WT-V5 cDNA used (see above). Fig. 2 in the online Data Supplement shows that endogenous concentrations of intracellular PCSK9 were decreased upon overproduction of PCSK9-Q152H, compared with mock-transfected cells, and there was a corresponding significant upregulation of LDLR in these cells (P=0.02). Fig. 3 in the online Data Supplement illustrates the cotransfection of our plasmids by immunocytochemistry.
The loss-of-function PCSK9-Q152H variant is the first described in a white Canadian population to have such a profound effect on plasma cholesterol concentrations. Several other loss-of-function PCSK9 variants with strong phenotypic effects on the LDLC concentration have been described for other populations. Two such variants occur in carriers of African descent, and both are nonsense variants. One of the variants occurs in the prodomain of PCSK9. No PCSK9 is produced, owing to an early truncation (PCSK9-Y142X). The other variant is a C-terminal nonsense mutation (PCSK9-C679X) that causes the retention of autocatalytically cleaved PCSK9 in the ER 16, 21. The LDLC concentrations in carriers of these variants range from the first percentile to the 50th percentile, with a mean lowering in the LDLC concentration of 40%, after adjustment for age- and sex-matched controls 16. The plasma PCSK9 concentrations in carriers of the C679X and Y142X variants are approximately 60% lower than those in their control population 28. The third variant, a compound mutation (R104C/V114A) found in a French family and associated with familial hypobetalipoproteinemia, exhibits a dominant negative effect on PCSK9 secretion 29. In Fig. 3A, we show that cotransfection of equal amounts of PCSK9-Q152H-V5 and nontagged PCSK9-WT decreased WT secretion by approximately 80% (from 175 μg/L to 35 μg/L). This decrease was not a general effect of cotransfection, because transfection of equal amounts of PCSK9-WT-V5 and nontagged PCSK9-WT (Fig. 3A, lane 6) increased the concentration of secreted PCSK9 to 275 μg/L. Therefore, in the ex vivo cell culture conditions of cotransfection, our PCSK9-Q152H variant does have a dominant negative effect on PCSK9-WT secretion. Whether this effect also occurs in vivo is not known; however, persons carrying this variant have 79% less circulating PCSK9 than unrelated noncarriers. If this effect does occur in vivo, we expect that such an effect would amplify the loss-of-function phenotype of the Q152H variant.
Conversely, there are some gain-of-function PCSK9 variants—S127R and D127G—that display decreased autocatalytic cleavage and secretion but are associated with hypercholesterolemic phenotypes 10, 30. In vitro binding assays have shown 5-fold increased binding of the PCSK9-S127R variant to the LDLR compared with WT, a result that could partly account for its gain of function 31. Alternatively, these variants may mediate intracellular LDLR degradation more efficiently. This pathway has been described by Poirier et al. 32, although the relative contributions of the intracellular and extracellular degradation routes with respect to PCSK9-mediated LDLR degradation are not fully understood.
Several studies have reported a positive correlation between the plasma PCSK9 concentration and the LDLC concentration in general populations 28, 33–36. Other studies have documented variable changes in plasma PCSK9 for carriers of PCSK9 loss-of-function and gain-of-function variants compared with control individuals (noncarriers of a particular PCSK9 variant) 34, 36, 37. In fact, circulating PCSK9 may differentially affect plasma LDLC concentrations, depending on whether an individual carries a PCSK9 variant that alters its LDLR-degrading activity and depending on the mode of action of that particular PCSK9 mutation and/or variant. Overall, this variability means that the plasma PCSK9 concentration as measured by ELISA does not necessarily predict the LDLC concentration, because the LDLC concentration can, in some instances, be strongly influenced by the mode of action of a PCSK9 variant 36. For instance, carriers of the gain-of-function PCSK9-D374Y variant, which is associated with autosomal dominant hypercholesterolemia and very high LDLC concentrations, have lower plasma PCSK9 concentrations than the general population 37, a finding that conflicts with the reports of a positive correlation between the plasma PCSK9 concentration and the LDLC concentration 28, 33–36. The mode of action of the PCSK9-D374Y variant has been well studied, however. It binds 10 times better to the liver LDLR than PCSK9-WT, thereby decreasing the plasma PCSK9 concentrations in D374Y carriers compared with noncarriers and augmenting LDLR degradation 38. On the other hand, plasma PCSK9 concentrations are also reduced in carriers of the loss-of-function PCSK9-R46L variant, which is associated with reduced LDLC concentrations 24, 36, 37, although the reason for this observation is less clear. Our studies with HEK293 cell cultures showed no change in the rate of PCSK9-R46L secretion (data not shown). Our previous studies with cell cultures have shown that the propeptide of the R46L variant displays a greater susceptibility to cleavage, which may affect the half-life of the circulating PCSK9 complex 39. Others have shown that the PCSK9-R46L variant binds less strongly to the LDLR, although how this binding would affect the plasma PCSK9 concentration is unclear 38. The primary mode of action (or inaction) of the PCSK9-Q152H mutation identified and characterized in this study is clear, however. The protein is not secreted, and therefore plasma concentrations are low, protecting the liver LDLRs from PCSK9-mediated LDLR degradation (Figs. 1 and 2).
In a previous report, we described our investigation of the effect of amino acid substitutions around the autocatalytic cleavage site of PCSK9 (VFAQ152↓SIP) 10. We showed that PCSK9 tolerates an Ala substitution for Glu at P1 and that its relative processing and secretion are approximately 80% of PCSK9-WT 10. In the present study, we have shown that the naturally occurring amino acid substitution of His for Glu prevents autocatalytic cleavage by PCSK9 in the ER, thereby precluding PCSK9 secretion (Figs. 2 and 3). Consequently, this PCSK9 variant no longer affects the downregulation of cell surface LDLR through the endosomal/lysosomal pathway, a major route for PCSK9-mediated LDLR degradation 40. Our cell biology findings are consistent with the findings of our human studies, in which individuals carrying the PCSK9-Q152H mutation showed very low plasma PCSK9 concentrations (>79% reduction compared with the general Canadian population) and consequently low concentrations of circulating LDLC owing to the upregulation of liver LDLR. This report is the first of a loss-of-function mutation in PCSK9 within the white Canadian population that displays such a profound effect on cholesterolemia. It also reinforces the suggestion that lowering the PCSK9 concentration by blocking PCSK9 synthesis, processing, or secretion could be an effective therapeutic strategy to complement current lipid-lowering drugs.
We are indebted to the family members who participated in this study, as well as to the persons recruited from the general Canadian population. The authors thank Pavel Milman for his assistance with immunocytochemistry.
↵6 Nonstandard abbreviations:
- familial hypercholesterolemia;
- LDL receptor;
- endoplasmic reticulum;
- LDL cholesterol;
- anti–IB PCSK9 Ab,
- anti-PCSK9 antibody used for immunoblotting;
- wild type
↵7 Human genes: LDLR, low density lipoprotein receptor; APOB, apolipoprotein B; PCSK9, proprotein convertase subtilisin/kexin type 9.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest:
Employment or Leadership: J. Davignon, Quebec Consortium on Drug Discovery (CQDM).
Consultant or Advisory Role: J. Davignon, Abbott/Solvay, AstraZeneca, Ascasi, Cortria, Genzyme, McCain, Merck, Pfizer, and Roche.
Stock Ownership: None declared.
Honoraria: J. Davignon, Abbott/Solvay, AstraZeneca, Ascasi, Cortria, Genzyme, McCain, Merck, Pfizer, and Roche.
Research Funding: J. Mayne, Canadian Institutes of Health Research (CIHR) (CTP 82946 and MOP 102741), and the Heart and Stroke Foundation of Ontario; T.C. Ooi, the Heart and Stroke Foundation of Ontario; J. Davignon, Merck Canada, Pfizer, and AstraZeneca; N.G. Seidah, Canadian Institutes of Health Research (CIHR) (CTP 82946 and MOP 102741) and the Richard and Edith Strauss Foundation; M. Mbikay, Canadian Institutes of Health Research (CIHR) (CTP 82946 and MOP 102741), the Richard and Edith Strauss Foundation, and the Heart and Stroke Foundation of Ontario; M. Chrétien, Canadian Institutes of Health Research (CIHR) (CTP 82946 and MOP 102741), the Richard and Edith Strauss Foundation, the Fondation Jean-Louis Lévesque, and the Heart and Stroke Foundation of Ontario.
Expert Testimony: None declared.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
- Received for publication March 11, 2011.
- Accepted for publication July 22, 2011.
- © 2011 The American Association for Clinical Chemistry