E

WDJSJA

Tctmm(ghoapthstto

cgwbtp(lodaiavS

pht

vA

U

Journal of the American College of Cardiology Vol. 45, No. 10, 2005© 2005 by the American College of Cardiology Foundation ISSN 0735-1097/05/$30.00P

a2otl

attddir

s(cfdFoogtatebmmtrsCii1uAacrfcPrstalsao

ublished by Elsevier Inc. doi:10.1016/j.jacc.2005.02.056

DITORIAL COMMENT

hat Promiseoes PCSK9 Hold?*

ohn J. P. Kastelein, MD, PHD,igrid W. Fouchier, MSC,oep C. Defesche, PHDmsterdam, the Netherlands

he cholesterol hypothesis has been with us for almost aentury. Nevertheless, only in the last three decades, afterhe discovery of the low-density lipoprotein (LDL) andevalonate pathways by Brown and Goldstein, lipoproteinetabolism has become the subject of intensive research

1,2). The discovery that mutations in the LDL receptorene are the cause of familial hypercholesterolemia (FH)as, among other findings, enabled us to discern FH fromther types of hypercholesterolemia, either inherited orcquired. The subsequent elucidation of the mevalonateathway, with hydroxyl-methylglutaryl coenzyme A reduc-ase as the rate-limiting enzyme in cholesterol synthesis, haseralded the development of inhibitors of this enzyme, theo-called statins (3). Statins are the drugs of choice in thereatment of elevated cholesterol levels and in the preven-ion of cardiovascular disease and will remain the backbonef any treatment regimen in this area for many years to

See page 1611

ome. In 1987, it became evident that there was a secondenetic cause for inherited hypercholesterolemia. Patientsith this disorder carry LDL particles that are not able toind to the LDL receptor because of mutations in the genehat codes for apolipoproteinB 100 (apoB), the structuralrotein component of LDL that interacts with this receptor4,5). This genetically distinct cause of inherited hypercho-esterolemia was coined familial defective apolipoprotein B,r FDB. Although from a therapeutic perspective theiscovery of FDB did not lead to new insights, it did providedditional focus on the atherogenic potential of apoB. Thisnsight has contributed to the notion that the ratio apoB/poA1 has emerged as the single best predictor for cardio-ascular disease, as firmly established by the Interhearttudy recently (6).Until very recently, it was fair to say that in most

opulations approximately 65% of the cases of inheritedypercholesterolemia could be explained by mutations inhe LDL receptor gene and in 10% by mutations in the

*Editorials published in the Journal of the American College of Cardiology reflect theiews of the authors and do not necessarily represent the views of JACC or themerican College of Cardiology.

nFrom the Department of Vascular Medicine, Academic Medical Center at theniversity of Amsterdam, Amsterdam, the Netherlands.

poB gene. The underlying genetic cause of the remaining5% of individuals suffering from inherited hypercholester-lemia now appears to be partly resolved with the findinghat proprotein convertase subtilisin/kexin 9, or PCSK9,ikely plays an important role in cholesterol metabolism.

The question now is: does PCSK9 hold the same promises apoB and the LDL receptor 15 and 30 years ago? Willhe elucidation of the role of PCSK9 equally contribute tohe improvement of diagnosis, risk estimation, and theevelopment of new therapeutic targets in cardiovascularisease prevention? Unfortunately, the situation surround-ng PCSK9 is, at minimal, conflicting and confusing. Let useview how this situation has arisen.

In 1999, a French group identified a locus on chromo-ome 1 that was linked to inherited hypercholesterolemia7) and in May 2003, the same group pinpointed PCSK9 tohromosome 1p32 (8). PCSK9 is a protease whose preciseunction is not fully understood. It probably can activate oreactivate other proteins by proteolytic cleavage. Therench group reported two mutations in the coding regionf the PCSK9 gene that segregated with a hypercholester-lemic phenotype, in which the LDL receptor and the apoBenes were excluded as the cause of inherited hypercholes-erolemia. These researchers also introduced the termutosomal-dominant hypercholesterolemia (ADH) to refero the three genetically distinct forms of hypercholesterol-mia that are inherited in an autosomal dominant fashion,ut that are clinically undistinguishable: FH caused byutations in the LDL-receptor gene, FDB caused byutations in the apoB-gene, and FH3, caused by a, up to

hat time, unknown genes. Since the time that PCSK9 waseported as the putative third locus involved in ADH,everal studies have been published, including the study byhen et al. (9) in this issue of the Journal. These colleagues

nvestigated the role of PCSK9 and the molecular variationn this gene in the metabolism of LDL-cholesterol (8,10–3). Regretfully, these recent studies have not been able tonequivocally confirm PCSK9 as the third gene causingDH, and the reported results vary from definitely caus-

tive, e.g., Timms et al. (10) and Leren (11) to definitely notausative (14). The essential difference between the firsteport by Abifadel et al. (8) and the investigations thatollowed, lies in the fact that Abifadel et al. (8) identified ahromosomal segment on chromosome 1 that harbored theCSK9 gene, plus 42 other known genes, whereas otheresearchers were focused directly onto the PCSK9 gene. Thetudy by Abifadel et al. (8) does not rule out the possibilityhat the missense mutations that were identified in PCSK9re in fact not the cause of ADH but that they might be ininkage disequilibrium with other genetic variants on theame chromosomal segment of 5.2 megabases. Such reason-ble doubt is strengthened by the fact that functional assaysf the effects of these mutations on PCSK9 function have

ot been performed so far. Nevertheless, PCSK9 has

er

csawoaerepntrotLThll

aoavEavPhaLhtLbstfistsvnhdCid

ro

RDUt

R

1

1

1

1

1

1

1

1

1621JACC Vol. 45, No. 10, 2005 Kastelein et al.May 17, 2005:1620–1 Editorial Comment

merged as a candidate for the third ADH gene because aole for PCSK9 in cholesterol metabolism is evident.

Recent gene expression data obtained from hepatic tissue ofholesterol-fed mice have shown that PCSK9 expression isignificantly down-regulated by dietary cholesterol (15). Ingreement with these findings, up-regulation becomes evidenthen human liver cells are cholesterol depleted (13). Also,verexpression of PCSK9 in mice results in a similar phenotypes in mice lacking the LDL receptor (15). This severelylevated cholesterol phenotype is supposed to be the result ofapid degradation of the LDL receptor protein because over-xpression of PCSK9 caused a 72% decrease in LDL receptorrotein, whereas LDL receptor mRNA levels were in theormal range (16,17). This notion is further strengthened byhe observation that overexpression of PCSK9 in LDLeceptor-deficient mice does not result in additional elevationf cholesterol levels (15). Despite these solid data in animals,he mechanism by which genetic variants in PCSK9 affectDL cholesterol levels in humans is still poorly understood.he obvious conflicting situation has now arisen insofar thatigh PCSK9 activity levels in mice result in elevated cholesterol

evels, whereas in contrast in humans, reduced PCSK9 activityeads to an elevated cholesterol phenotype.

Therefore, novel attempts, such as the study by Chen etl. (9) in this issue of the Journal, to further clarify the rolef PCSK9 in cholesterol metabolism have been eagerlywaited. In this study, several novel and known geneticariants in the PCSK9 gene were assessed. One variant,670G, a substitution of glutamic acid by glycin at amino

cid position 670, indeed explained as much as 3.5% of theariation in LDL cholesterol levels, confirming a role forCSK9. Nevertheless, a gene dosage effect, which mightave indicated a direct functional influence on enzymectivity in human lipoprotein metabolism, was not observed.ipoprotein levels associated with wild type (EE) and theeterozygous (EG) form were essentially similar, and onlyhe homozygous (GG) form was associated with elevatedDL cholesterol levels. However, these findings might haveeen the result of reduced power because of limited sampleize. In view of the modest elevation of LDL cholesterol byhe homozygous GG form, the clinical implications of thesendings seem to be limited. The risk of coronary athero-clerosis only showed an upward trend in GG carriers andhe response to cholesterol lowering treatment with fluva-tatin was not influenced by presence or absence of the Gariant. Therefore, the currently gathered knowledge doesot seem to provide the final answer to the questions thatave arisen since the implication of PCSK9 in hereditaryefects in human lipoprotein metabolism. The study byhen et al. (9) should be replicated in other populations,

deally with a large enough sample size to generate the

efinitive data. So far, a significant impact on cardiovascular

isk or treatment of cardiovascular disease has not been partf the PCSK9 story.

eprint requests and correspondence: Dr. John J. P. Kastelein,epartment of Vascular Medicine, Academic Medical Centre,niversity of Amsterdam, P.O. Box 22700, 1100 DE Amsterdam,

he Netherlands. E-mail: e.vandongen@amc.uva.nl.

EFERENCES

1. Goldstein JL, Brown MS. Familial hypercholesterolemia: identifica-tion of a defect in the regulation of 3-hydroxy-3-methylglutarylcoenzyme A reductase activity with overproduction of cholesterol. ProcNatl Acad Sci USA 1973;70:2804–9.

2. Brown MS, Brannan PG, Bohmfalk HA, et al. Use of mutantfibroblasts in the analysis of the regulation of cholesterol metabolism inhuman cells. J Cell Physiol 1975;85:425–36.

3. Endo A, Kuroda M, Tanzawa K. Competitive inhibition of3-hydroxy-3-methylglutaryl coenzyme A reductase by ML-236A andML-236B fungal metabolites, having hypocholesterolemic activity.FEBS Lett 1976:72:323–6.

4. Innerarity TL, Weisgraber KH, Arnold KS, et al. Familial defectiveapolipoprotein B-100: low density lipoproteins with abnormal receptorbinding. Proc Natl Acad Sci USA 1987;84:6919–23.

5. Innerarity TL, Mahley RW, Weisgraber KH, et al. Familial defectiveapolipoprotein B-100: a mutation of apolipoprotein B that causeshypercholesterolemia. J Lipid Res 1990;31:1337–49.

6. Yusuf S, Hawken S, Ounpuu S, on behalf of the INTERHEARTstudy investigators. Effect of potentially modifiable risk factors asso-ciated with myocardial infarction in 52 countries (the INTERHEARTstudy): case-control study. Lancet 2004;364:937–52.

7. Varret M, Rabes JP, Saint-Jore B. A third major locus for autosomaldominant hypercholesterolemia maps to 1p34.1-p32. Am J HumGenet 1999;64:1378–87.

8. Abifadel M, Varret M, Rabes JP, et al. Mutations in PCSK9 causeautosomal dominant hypercholesterolemia. Nat Genet 2003;34:154–6.

9. Chen SN, Ballantyne CM, Gotto AM Jr., Tan Y, Willerson JT,Marian AJ. A common PCSK9 haplotype, encompassing the E670Gcoding single nucleotide polymorphism, is a novel genetic marker forplasma low-density lipoprotein cholesterol levels and severity ofcoronary atherosclerosis. J Am Coll Cardiol 2005;45:1611–9.

0. Timms KM, Wagner S, Samuels ME, et al. A mutation in PCSK9causing autosomal-dominant hypercholesterolemia in a Utah pedigree.Hum Genet 2004;114:349–53.

1. Leren TP. Mutations in the PCSK9 gene in Norwegian subjects withautosomal dominant hypercholesterolemia. Clin Genet 2004;65:419–22.

2. Shioji K, Mannami T, Kokubo Y, et al. Genetic variants in PCSK9affect the cholesterol level in Japanese. J Hum Genet 2004;49:109–14.

3. Dubuc G, Chamberland A, Wassef H, et al. Statins upregulatePCSK9, the gene encoding the proprotein convertase neuralapoptosis-regulated convertase-1 implicated in familial hypercholes-terolemia. Arterioscler Thromb Vasc Biol 2004;24:1454–9.

4. Damgaard D, Jensen JM, Larsen ML, et al. No genetic linkage ormolecular evidence for involvement of the PCSK9, ARH or CYP7A1genes in the familial hypercholesterolemia phenotype in a sample ofDanish families without pathogenic mutations in the LDL receptorand apoB genes. Atherosclerosis 2004;177:415–22.

5. Maxwell KN, Breslow JL. Adenoviral-mediated expression of Pcsk9 inmice results in a low-density lipoprotein receptor knockout phenotype.Proc Natl Acad Sci USA 2004;101:7100–5.

6. Benjannet S, Rhainds D, Essalmani R, et al. NARC-1/PCSK9 and itsnatural mutants: zymogen cleavage and effects on the low densitylipoprotein (LDL) receptor and LDL cholesterol. J Biol Chem2004;279:48865–75.

7. Park SW, Moon YA, Horton JD. Post-transcriptional regulation of

LDL receptor protein by proprotein convertase subtilisin/kexin type 9a(PCSK9) in mouse liver. J Biol Chem 2004;279:50630–8.