Mutations in PCSK9 cause autosomal dominant hypercholesterolemia

© 2003 Nature Publishing Group http://www.nature.com/naturegenetics

B R I E F C O M M U N I C AT I O N S

Mutations in PCSK9 cause autosomal dominant hypercholesterolemia Marianne Abifadel1,2, Mathilde Varret1, Jean-Pierre Rabès1,3, Delphine Allard1, Khadija Ouguerram4, Martine Devillers1, Corinne Cruaud5, Suzanne Benjannet6, Louise Wickham6, Danièle Erlich1, Aurélie Derré1, Ludovic Villéger1, Michel Farnier7, Isabel Beucler8, Eric Bruckert9, Jean Chambaz10, Bernard Chanu11, Jean-Michel Lecerf12, Gerald Luc12, Philippe Moulin13, Jean Weissenbach5, Annick Prat6, Michel Krempf4, Claudine Junien1,3, Nabil G Seidah6 & Catherine Boileau1,3 Autosomal dominant hypercholesterolemia (ADH; OMIM144400), a risk factor for coronary heart disease, is characterized by an increase in low-density lipoprotein cholesterol levels that is associated with mutations in the genes LDLR (encoding low-density lipoprotein receptor) or APOB (encoding apolipoprotein B). We mapped a third locus associated with ADH, HCHOLA3 at 1p32, and now report two mutations in the gene PCSK9 (encoding proprotein convertase subtilisin/kexin type 9) that cause ADH. PCSK9 encodes NARC-1 (neural apoptosis regulated convertase), a newly identified human subtilase that is highly expressed in the liver and contributes to cholesterol homeostasis. To identify the third locus associated with ADH (called HCHOLA3 or FH3), which we previously mapped1 to 1p34.1–p32 (OMIM 603776) and was confirmed in a large Utah kindred2, we carried out positional cloning using the family in whom linkage was originally identified and 23 French families in whom the implication of LDLR and APOB had been excluded (Supplementary Note online). Family HC92 was identified through the proband (HC92-II-7), who belongs to a multiplex pedigree affected with ADH. We obtained samples from 29 members of this family and tested them using parametric linkage analyses. We excluded linkage to LDLR and APOB (lod scores of –14.05 and –10.01 (θ = 0.0), respectively). We genotyped family members for eight Genethon markers in the 1p34–p32 region (Fig. 1) and obtained highly significant lod scores with a maximum of 4.26 (θ = 0.0) at D1S2742 that reached 4.80 in the multipoint analyses

(Supplementary Table 1 online and Fig. 2a). Haplotype analysis identified a critical interval of 5.9 Mb between D1S231 and D1S2890. The critical interval that our team had previously reported in family HC2 (ref. 1) was more distally located between markers D1S472 and D1S211. Reexamination of haplotype data (Supplementary Fig. 1 online) showed that all affected members of family HC2 also shared the same haplotype between markers D1S2722 and D1S2890, except for individual HC2-II-5. This affected individual had a recombinational event at D1S211, which formed the centromeric boundary of the region that we originally described1. Therefore, all family members were reinvestigated. Individual HC2-II-5 (who refuses treatment) was the only one who showed a substantial variation (a marked elevation of triglycerides) and thus no longer conforms with the inclusion criteria. The region between D1S197 and D1S2890 contains 41 genes, including PCSK9. It encodes NARC-1, a novel putative proprotein convertase belonging to the subtilase subfamily3. A related protein is the subtilisin kexin isoenzyme-1 (SKI-1)/site-1-protease (S1P), which has a key role in cholesterol homeostasis through processing the sterol regulatory element–binding proteins4,5. The cDNA spans 3,617 bp and encodes a protein of 692 amino acids. By sequencing the 12 exons of PCSK9, we identified in family HC92 a T→A substitution in exon 2 at nucleotide 625 predicting a substitution at codon 127 of arginine for the conserved serine (S127R), thereby creating a MnlI cleavage site (Fig. 2b and Supplementary Fig. 2 online). We tested the members of family HC92 and 100 control individuals for this substitution. It was absent in the 200 control chromosomes, indicating that it is not a polymorphism. We found it in the 12 affected family members and in individual HC92-IV-3, whose total cholesterol level is in the 90th percentile when compared with other French individuals matched by age and sex. Thus, the penetrance in the family is estimated at 0.94. Notably, the 625T→A mutation was also found in the proband of family HC2 and cosegregated with the disease except in individual HC2-II-5, confirming that he had been misclassified in the linkage analyses previously reported1. To assess the possible recurrence of this mutation, we tested flanking and intragenic polymorphic markers in both families. The same haplotype segregated with the 625T→A mutation in families HC2 and HC92 (Supplementary Table 2 online), indicating that despite different geographical origins, the families share a common ancestor. The possibility of a French founder effect was ruled out, as the mutation was not found in 22 other French probands with ADH.

1INSERM U383, Hôpital Necker-Enfants Malades, AP-HP, Université Paris V, 149-161 rue de Sèvres, 75743 Paris Cedex 15, France. 2Faculté de Pharmacie, Université Saint-Joseph, BP.11-5076 Beirut, Lebanon. 3Laboratoire de Biochimie, d’Hormonologie et de Génétique Moléculaire, CHU Ambroise Paré, AP-HP, Université VersaillesSaint-Quentin-en-Yvelines, 9 avenue Charles de Gaulle, 92104 Boulogne Cedex, France. 4Service d’Endocrinologie, CHU Hôtel Dieu, 44093 Nantes, France. 5Génoscope, Centre National de Séquençage, 2 rue G. Crémieux, CP 5706, Evry, France. 6Laboratory of Biochemical Neuroendocrinology, Clinical Research Institute of Montreal,110 Pine Ave. West, Montreal, QC H2W 1R7, Canada. 7Point Médical, Rond Point de la Nation, 21000 Dijon, France. 8Laboratoire des Lipides, Hôpital Pitié Salpétrière, 83 boulevard de l’Hôpital, 75013 Paris, France. 9Service d’Endocrinologie-Métabolisme, Hôpital Pitié Salpétrière, 83 boulevard de l’Hôpital, 75013 Paris, France. 10INSERM U505, Institut des Cordeliers, 15 rue de l’Ecole de Médecine, 75006 Paris, France. 11Service de Médecine Interne, Hôpital Henri Mondor, 51 avenue du Maréchal de Lattre de Tassigny, 94010 Créteil, France. 12Service de Médecine Interne, Hôpital Huriez, Place de Verdun, 59019 Lille, France. 13Service d’Endocrinologie et des Maladies de la Nutrition, Hôpital de l’Antiquaille, rue de l’Antiquaille, 69005 Lyon, France. Correspondence should be addressed to C.B. (e-mail: [email protected]).

154

VOLUME 34 | NUMBER 2 | JUNE 2003 NATURE GENETICS

B R I E F C O M M U N I C AT I O N S

I

1

2

Figure 1 Pedigree of family HC92 and genetic analysis with markers spanning the 1p34.1–p32 region. Affected individuals present with a history of tendon xanthomas (individuals HC92-II-7 and HC92-III-3), coronary heart disease, early myocardial infarction (individuals HC92-II-2 and HC92-II-6) and stroke (individual HC92-II-4). Filled bars indicate the mutated allele. Age (in years) at lipid measurement, total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C; in g per liter; untreated values for affected individuals) are given.

II 1

2

3

4

Age TC LDL-C

5

6

7

52 3.50 _

D1S2722 D1S211 D1S197 D1S231 D1S417 D1S200 D1S2742 D1S2890

6 4 3 3 2 3 6 5

6 6 3 3 3 5 4 9

6 4 7 3 2 4 7 9

3 1 4 1 3 3 1 9

59 5.21 4.20

3 3

6 4

_

_

1 2 3 4 9

3 2 3 6 5

Age TC LDL-C


1

2

3

4

5

6

7

8

51 1.59 0.78

42 3.44 2.76

46 4.00 _

48 2.22 1.32

49 3.34 2.63

42 2.22 1.37

42 1.75 0.90

40 3.34 2.64

6 4 6 2 2 6 4 6

1 2 6 3 3 3 7 9

6 4 4 2 1 3 4 8

6 4 3 3 2 3 6 9

6 4 7 3 2 4 7 9

6 4 3 3 2 3 6 5

2 5

6 4

_

_

2 3 2 7 2

4 3 3 4 1

3 1 4 1 3 3 1 9

4 6 7 4 7 3 3 3 2 6 4 2 7 4 9

3 2

6 6 3 3 2 3 6 5

_

_

3 3 2 7 7

6 6 3 3 3 5 4 9

6 4 7 3 2 4 7 9

6 4 3 3 2 3 6 5

9

10

11

12

36 4.66 3.89

40 3.29 2.56

38 1.89 1.00

39 3.24 _

6 4 7 3 2 4 7 9

6 4 3 3 2 3 6 5

6 4

5 1

5 1

6 1

6 4

_

_

_

_

3 2 3 6 5

3 3 6 7 9

4 2 6 1 6

3 4 3 4 3

3 2 3 6 5

_

5 1

_

3 3 6 7 9

IV Age TC LDL-C


1

2

19 1.35 0.48

17 1.74 0.81

1 2 6 3 3 3 7 9

6 6 4 _ 3 6 3 2 1 2 3 6 4 4 8 6

6 6 4 3 6 3 2 1 2 3 6 4 4 8 6

_

3

4

5

6

15 2.05 1.44

14 2.79 1.91

23 1.68 0.99

21 3.54 2.78

6 4 3 3 2 3 6 9

1 2 6 3 3 3 7 9

6 4 3 3 2 3 6 9

6 4

3 2 1 5

_

_

_

4 3 3 4 1

1 3 3 1 9

2 3 2 7 2

7 17 1.22 0.58

8

9

10

11

12

14 3.22 2.45

15 1.75 1.07

12 1.68 1.04

11 2.17 1.49

6 1.63 0.94

6 6 6 4

3 1

2 5

6 6

_

_

_

_

3 2 3 6 5

1 3 3 1 9

2 3 2 7 2

3 2 3 6 5

3 3 2 7 7

_ _

4 3 3 4 1

3 2

6 3 4 2

6 4

_ _

_

_

4 7

6 6

3 3 2 7 7

3 2 4 7 9

3 3 6 2 4

3 2 4 7 9

3 2 4 7 9

_

_

Through systematic bidirectional sequencing of the 12 exons of PCSK9 in 22 probands with ADH, we identified a second mutation (890T→C, resulting in the amino-acid substitution F216L) in the proband of family HC60 (Fig. 2c), who died from myocardial infarction at 49 years of age (Fig. 2d and Supplementary Fig. 2 online). This mutation segregated with the ADH phenotype in the family and was not found in 200 control chromosomes. No large rearrangement was found in any of the probands by Southern blotting (data not shown). Thus, mutations in PCSK9 have been found in 12.5% of the families with ADH that we tested. We also identified 25 polymorphisms present in different probands and on control chromosomes,

a

c

13

14

9 1.91 0.94

11 1.34 0.62

5 1

5 5 1 1 _

_

5 1

_

_

_

4 2 6 1 6

3 3 6 7 9

4 2 6 1 6

3 3 6 7 9

3 4 3 4 3

3 3 6 7 9

5 1

6 1

none of which gives rise to new donor or acceptor splice sites (Supplementary Table 2 online). NARC-1 is a novel proprotein convertase3. It is synthesized as a soluble zymogen that undergoes autocatalytic intramolecular processing in the endoplasmic reticulum at the primary cleavage site YVVVL↓− KEE85, indicative of the enzymatic specificity3 of NARC-1. Prosegment cleavage is necessary for NARC-1 to exit from the endoplasmic reticulum. The S127R mutation resides between the primary and putative secondary zymogen processing sites of the NARC-1 propeptide, and F216L is located close to the active site (which is at His226). Notably, the S127R mutation creates an RGD site that may

I 1

6

2

II Age TC LDL-C D1S197 D1S231 D1S417 D1S2742 D1S2890

5 lod score


III

4 3 2

2

1

3 40 2.01 1.21 2 4 1 3 5 5 3 4 8 8

49 4.41 3.56 – – 3 3 2 4 4 3 3 8

III

1

PCSK9 Age TC LDL-C D1S197 D1S231 D1S417 D1S2742 D1S2890

0 D1S2890

Exon 2 mutation control

6 D1S200 D1S417

D1S231

D1S197

b

4

D1S2742

2

0

8 cM

Individual HC 92-II-7

d

Exon 4 mutation control

N

625T→A (S127R)

NATURE GENETICS VOLUME 34 | NUMBER 2 | JUNE 2003

1 28 1.63 0.73 3 5 3 3 2 5 4 4 8 2

2

3

5 2.20 1.54 2 4 3 3 4 5 3 4 8 8

3 2.36 1.72 – – 3 1 4 5 3 3 8 8

Individual HC60-II-2 N

890T→C (F216L)

Figure 2 Genetic analysis and mutation detection in families HC92 and HC60. (a) Results of LINKMAP analyses in family HC92 indicating a maximum lod score for D1S2742 at θ = 0. PCSK9 maps 1.2 Mb away from this marker. (b) Mutation in family HC92. The proband (HC92-II-7, indicated by an arrow in Fig. 1) is heterozygous with respect to the 625T→A substitution in exon 2 (resulting in the amino-acid substitution S127R). (c) Pedigree and genetic analysis of family HC60. Age (in years) at lipid measurement, total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C; in g/L; untreated values for affected individuals) are given. (d) Sequence analysis in family HC60. The proband (HC60-II-2, indicated by an arrow) is heterozygous with respect to the 890T→C substitution in exon 4, predicting the amino-acid substitution F216L.

155


B R I E F C O M M U N I C AT I O N S be involved in integrin binding6. Although proprotein convertases activate a wide variety of proteins, none of them has yet been linked to a dominant human disease. NARC-1 is thus unique in this respect, and its identification may lead to the discovery of others. At present, the molecular mechanisms that underlie the dominance of the trait are unknown. At this time, only missense mutations have been identified in PCSK9, favoring a gain-of-function or a dominant-negative mechanism. Loss of function cannot be ruled out at this stage, however, as documented in acute intermittent porphyria (OMIM 176000), a dominant trait sometimes associated with reduction of porphobilinogen deaminase activity in the liver. The related convertase SKI-1/S1P has a key role in regulating cholesterol and fatty acid homeostasis4,5, but the precise implication of NARC-1 in cholesterol homeostasis is unknown. Notably, NARC-1 is expressed mainly in the liver and small intestine3, both of which are important in cholesterol synthesis and regulation. The crucial role of NARC-1 is indicated by the hypercholesterolemia that occurs when the gene is mutated. The identification of NARC-1 substrates may elucidate novel disease mechanisms and constitute targets for new intervention strategies to limit elevation of low-density lipoprotein particles and prevent morbidity and mortality from premature atherosclerosis.

156

Note: Supplementary information is available on the Nature Genetics website. ACKNOWLEDGMENTS We are indebted to the family members for their cooperation and to C. Mugnier for expert assistance in computing the statistics. This work was supported by grants from Progrès-Institut National de la Santé et de la Recherche Médicale, Pfizer, Fondation de France, Comité Français de Coordination des Recherches sur l’Athérosclérose et le Cholestérol and Canadian Research grants (to N.G.S.). M.A. was supported by grants from Société Française d’Athérosclérose, Comité Français de Coordination des Recherches sur l’Athérosclérose et le Cholestérol, Fondation de la Recherche Médicale, Université René Descartes, Conseil de Recherche de l’Université Saint-Joseph. M.V. was supported by a grant from Société Française d’Athérosclérose and Pfizer. L.V. and D.A. were supported by grants from Ministère de l’Education Nationale de la Recherche et de la Technologie. COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests. Received 20 February; accepted 28 March 2003 Published online 5 May 2003; doi:10.1038/ng1161 1. 2. 3. 4.

Varret, M. et al. Am. J. Hum. Genet. 64, 1378–1387 (1999). Hunt, S.C. et al. Arterioscler. Thromb. Vasc. Biol. 20, 1089–1093 (2000). Seidah, N.G. et al. Proc. Natl. Acad. Sci. USA 100, 928–933 (2003). Brown, M.S. & Goldstein, J.L. Proc. Natl. Acad. Sci. USA 96, 11041–11048 (1999). 5. Elagoz, A., Benjannet, S., Mammarbassi, A., Wickham, L. & Seidah N.G. J. Biol. Chem. 277, 11265–11275 (2002). 6. Ruoslahti, E. Annu. Rev. Cell. Dev. Biol. 12, 697–715 (1996).

VOLUME 34 | NUMBER 2 | JUNE 2003 NATURE GENETICS