Unraveling New Therapeutic Targets of Coronary Artery Disease by Genetic Approaches

Abstract

Coronary artery disease (CAD) is the most common cause of death and physical disabilities in developed countries, even though efforts to identify and target causal factors such as hypertension and dyslipidemia have brought tremendous improvements in prevention and treatment. A rapid advance in technology has unraveled new genetic variants associated with CAD and also provided great opportunities to identify novel pathogenic mechanisms and to develop new drugs with higher specificity. Whole-genome sequencing and whole-exome sequencing has made it possible to find rare alleles that are responsible for CAD in small, affected families and case-control studies in a very efficient manner. At present, genome-wide association studies have identified more than 50 loci that explain approximately 10% of the heritability of CAD, most of which is unrelated to traditional risk factors. Mendelian randomization studies enable identification of causal factors among numerous biomarkers and to narrow down promising therapeutic targets. This review highlights new genetic approaches and demonstrates the extent to which the outcome contributes to the finding of new therapeutic targets. (Circ J 2015; 79: 8–14)

Coronary artery disease (CAD) is the leading cause of mortality and morbidity worldwide. It is a complex disease resulting from the interplay between multiple genetic variants and environmental factors, and genetic variability explains only a small part of the whole pathogenic mechanism of the disease. Nevertheless, in the goal of detecting new therapeutic targets, reliable identification and thorough understanding of potential causal genetic variants can be crucial. For example, in the case of low-density lipoprotein (LDL) cholesterol and statins, the discovery of mutations affecting the LDL receptor and causing hypercholesterolemia and early-onset myocardial infarction (MI) enabled the use of LDL cholesterol-lowering therapies that remarkably reduce the risk of cardiovascular events.¹ In 2003, a finding that gain-of-function mutations in the proprotein, converase stubilisin/kexin9 (PCSK9) is a causal event in 2 families with autosomal dominant hypercholesterolemia turned out to have a positive correlation with the incidence of CAD.² PCSK9 was then identified as binding the LDL receptor for degradation and reducing the capacity of the liver to bind and remove LDL cholesterol.^3,4 Subsequent studies revealed that some patients with PCSK9 loss-of-function mutations had a low LDL cholesterol level and a reduced incidence of CAD.^5–8 These study results raised the possibility that pharmacologic inhibition of PCSK9 might lower the LDL cholesterol level in patients with hypercholesterolemia. In fact, recent clinical trials proved that a human monoclonal antibody to PCSK9 resulted in a significant reduction of LDL cholesterol levels in patients.^9–11

The results from genetic studies can be applied to further research to improve disease risk prediction, pharmacogenomics and identifying new therapeutic targets. This review will mainly deal with the therapeutic aspect: to be more specific, how recent technical advances in the study of genetics help identify novel causative biochemical pathways and therapeutic targets in CAD.

Whole-Genome and Whole-Exome Sequencing in Linkage Analyses

According to the pattern of inheritance, any genetic disease can be divided into 2 classes. A monogenic or Mendelian disease is caused by mutations in 1 gene and follows Mendel’s law of inheritance, whereas a polygenic or complex disease is characterized by complex patterns of inheritance caused by interaction between multiple genetic variants and environmental factors. The latter type of CAD is common; however, there are a few cases of CAD showing a monogenic inheritance pattern and identification of genes associated with this monogenic disease has led to breakthrough understanding of the mechanism of the disease. A representative example is the Nobel Prize-winning discovery that mutations affecting the LDL receptor also cause hypercholesterolemia and early-onset ischemic heart disease (IHD).¹

Traditionally, linkage analysis (parametric) was adopted for Mendelian disease and gave researchers a powerful technique for mapping the location of disease-causing loci by identifying genetic markers that are co-inherited with a phenotype of interest (segregation).¹² First, a particular region within the genome responsible for a disease of interest must be identified using DNA markers, then a chromosomal region that segregates with the disease can be pinpointed to identify the causal gene and its variant. This type of study also successfully identified the association between ALOX5AP and coronary and cerebrovascular disease,¹³ between MEF2A and CAD,¹⁴ and between genetic variation of PCSK9 and cholesterol metabolism.¹⁵

Although previous linkage analyses in affected families followed the sequence of finding the regions first and then identifying the causal gene and the mutation by fine-mapping (positional cloning), so-called next-generation sequencing (NGS) now allows direct screening of the candidate variants and performance of linkage analyses. The NGS technologies are based on parallel sequencing of millions of DNA strands simultaneously. It dramatically decreases the cost and time to sequence the entire human genome. Less than $1,000 and 1 week is all that is needed to sequence a whole genome. Compared with whole-genome sequencing, whole-exome sequencing focuses on an alignment of exons, collectively called ‘the exome’, which constitutes only approximately 1% of the human genome but harbors 85% of occurring mutations that affect disease-related traits.^16,17 Detailed technical review of the existing platforms can be found elsewhere.¹⁸ A major advantage of these approaches is that less family members are required to have their genome sequenced to identify a particular genetic variation linked to a disease phenotype compared with conventional linkage analysis with positional cloning.^19–21

Using whole-exome sequencing and linkage analysis, Erdmann et al²² identified 2 heterozygous mutations in 2 functionally related genes, GUCY1A3 and CCT7, which impair nitric oxide signaling in an extended MI family. Only 3 distantly related members of the affected family underwent whole-exome sequencing and it was enough to screen out 4 candidate variants. Another study by Keramati et al also used the same method to identify a founder mutation of DYRK1B in 3 large families with coinheritance of early-onset CAD, central obesity, hypertension, and diabetes.²³ These pioneering studies highlight the opportunities of applying NGS to severely affected families to identify new mechanism of the disease and thereby new therapeutic targets.

Genome-Wide Association Studies and Whole-Exome Sequencing in Association Study

Performing a genome-wide association study (GWAS) is a genetic approach to a common disease based on a ‘common disease-common variant hypothesis’ that complex diseases result from cumulative and interactive effects of a large number of loci, each imparting a modest marginal effect to the expression of a phenotype.²¹ It compares the frequency of repeatedly appearing genetic variants in case-control studies and if significantly more common in cases, it implies that the variant is in a DNA location associated with an increased risk for the disease. The sequencing of the complete human genome and the rise of high-throughput technologies such as DNA microarray of more than 500,000 single-nucleotide polymorphisms (SNPs)²⁴ have triggered an explosion of GWAS aiming to identify disease-associated genetic loci. Furthermore, meta-analyses of multiple GWAS data sets have considerably increased the accuracy of detecting genetic disease associations.²⁵ In this way, many susceptible loci have been proven to be associated with CAD and MI. The advantages of this approach are that it is unbiased and requires no presupposition of which marker or gene is more likely to be a risk candidate.

To date, more than 50 genetic loci associated with CAD have been reported and they only explain approximately 10% of the heritability of CAD.^26–29 The most interesting finding is that more than two-thirds of the genetic loci are unrelated to traditional risk factors, which highlights that several pathways contributing to the pathogenesis of CAD are yet to be appreciated and there will be more therapeutic targets to unravel²⁶ (Table 1).

Table 1. Genetic Loci Associated With Coronary Artery Disease in Genome-Wide Association Studies

Chromosomal location	SNP	Related gene(s)	Risk allele	OR (95% CI)	P value
1p13.3	rs599839	SORT1	A	1.29 (1.18–1.40)	4×10−9
1p13.3	rs602633	SORT1	T	1.11 (1.08–1.15)	1×10−8
1p13.3	rs646776	CELSR2, PSRC1, SORT1	T	1.19 (1.13–1.26)	8×10−12
1p32.2	rs17114036	PPAP2B	A	1.17 (1.13–1.22)	4×10−19
1p32.3	rs11206510	PCSK9	T	1.15 (1.10–1.21)	1×10−8
1q21.3	rs4845625	IL6R	T	1.04 (1.02–1.07)	4×10−8
1q41	rs17465637	MIA3	C	1.14 (1.10–1.19)	1×10−9
2p11.2	rs1561198	VAMP5-VAMP8-GGCX	A	1.05 (1.03–1.07)	4×10−9
2p21	rs6544713	ABCG5-ABCG8	T	1.06 (1.04–1.09)	9×10−10
2p24.1	rs2123536	TTC32-WDR35	T	1.12 (1.08–1.16)	7×10−11
2p24.1	rs515135	APOB	G	1.08 (1.05–1.11)	5×10−10
2q22.3	rs2252641	ZEB2-AC074093.1	G	1.04 (1.02–1.06)	4×10−8
2q33.2	rs6725887	WDR12	C	1.17 (1.11–1.23)	1×10−8
3q22.3	rs2306374	MRAS	C	1.12 (1.07–1.16)	3×10−8
4q31.22	rs1878406	EDNRA	T	1.06 (1.02–1.11)	3×10−8
4q32.1	rs7692387	GUCY1A3	G	1.06 (1.03–1.09)	5×10−9
4q32.1	rs1842896	GUCY1A3	T	1.14 (1.10–1.19)	1×10−11
5p15.33	rs11748327	IRX1-LINC01020	N/A	1.25 (1.18–1.33)	5×10−13
5q31.1	rs273909	SLC22A4-SLC22A5	C	1.09 (1.05–1.12)	1×10−8
6p21.2	rs10947789	KCNK5	T	1.06 (1.03–1.08)	2×10−8
6p21.31	rs17609940	ANKS1A	G	1.07 (1.05–1.10)	1×10−8
6p21.32	rs9268402	C6orf10-BTNL2	G	1.16 (1.12–1.20)	3×10−15
6p21.33	rs3869109	HLA-C, HLA-B, HCG27	G	1.14 (N/A)	1×10−9
6p24.1	rs12526453	PHACTR1	C	1.12 (1.08–1.17)	1×10−9
6p24.1	rs9369640	PHACTR1	C	1.10 (1.08–1.14)	3×10−11
6p24.1	rs9349379	PHACTR1	G	1.15 (1.10–1.21)	2×10−9
6p24.1	rs6903956	C6orf105	A	1.65 (1.44–1.90)	3×10−13
6q23.2	rs12190287	TCF21	C	1.08 (1.06–1.10)	1×10−12
6q25.3	rs3798220	LPA	C	1.51 (1.33–1.70)	3×10−11
6q26	rs4252120	PLG	T	1.06 (1.03–1.09)	5×10−9
7p21.1	rs2023938	HDAC9	G	1.07 (1.04–1.11)	5×10−8
7q22.3	rs10953541	BCAP29	C	1.08 (1.05–1.11)	3×10−8
7q32.2	rs11556924	ZC3HC1	C	1.09 (1.07–1.12)	9×10−18
8p21.3	rs264	LPL	G	1.05 (1.02–1.08)	5×10−9
8q24.13	rs2954029	TRIB1	A	1.04 (1.02–1.06)	5×10−8
9p21.3	rs1333049	CDKN2A, CDKN2B	C	1.37 (1.26–1.48)	2×10−14
9p21.3	rs10757274	CDKN2A, CDKN2B	G	1.37 (1.31–1.43)	8×10−45
9q34.2	rs579459	ABO	C	1.10 (1.07–1.13)	4×10−14
10p11.23	rs2505083	KIAA1462	C	1.07 (1.04–1.09)	4×10−8
10q11.21	rs1746048	CXCL12	C	1.17 (1.11–1.24)	7×10−9
10q23.31	rs1412444	LIPA	T	1.09 (1.07–1.12)	3×10−13
10q24.32	rs12413409	CYP17A1, CNNM2, NT5C2	G	1.12 (1.08–1.16)	1×10−9
11q22.3	rs974819	PDGFD	T	1.07 (1.04–1.09)	2×10−9
11q23.3	rs964184	ZNF259, APOA5-A4-C3-A1	G	1.13 (1.10–1.16)	1×10−17
12q21.33	rs7136259	ATP2B1	T	1.11 (1.08–1.15)	6×10−10
12q24.11	rs3782889	MYL2	C	1.26 (1.19–1.34)	4×10−14
12q24.12	rs3184504	SH2B3	T	1.13 (1.08–1.18)	9×10−8
12q24.12	rs11066015	ACAD10, ALDH2, C12orf51, RPL6-PTPN11	A	1.41 (1.27–1.56)	5×10−11
12q24.13	rs11066280	C12orf51		1.19 (1.13–1.25)	2×10−11
13q12.3	rs9319428	FLT1	A	1.05 (1.03–1.08)	1×10−8
13q34	rs4773144	COL4A1, COL4A2	G	1.07 (1.05–1.09)	4×10−9
14q32.2	rs2895811	HHIPL1	C	1.07 (1.05–1.10)	1×10−10
15q25.1	rs3825807	ADAMTS7	A	1.08 (1.06–1.10)	1×10−12
15q26.1	rs17514846	FURIN-FES	A	1.05 (1.03–1.08)	4×10−10
17p11.2	rs12936587	RASD1, SMCR3, PEMT	G	1.07 (1.05–1.09)	4×10−10
17p13.3	rs216172	SMG6, SRR	C	1.07 (1.05–1.09)	1×10−9
17q21.32	rs46522	UBE2Z, GIP, ATP5G1, SNF8	T	1.06 (1.04–1.08)	2×10−8
19p13.2	rs1122608	LDLR, SMARCA4	G	1.15 (1.10–1.20)	2×10−9
19q13.32	rs2075650	APOE/TOMM40	G	1.14 (1.09–1.19)	3×10−8
21q22.11	rs9982601	KCNE2	T	1.20 (1.14–1.27)	6×10−11

Only P-value <5×10⁻⁸. Data from www.genome.gov. CI, confidence interval; OR, odds ratio.

Unfortunately, however, there is a high risk of false-positive results because of multiple comparisons. To alleviate the problem, a stringent statistical significance threshold P-value of 5×10⁻⁸ is used and the results must be replicated in an independent population.³⁰ Another drawback is that the identified susceptible loci mostly do not reveal direct information on disease-causing genes or affected pathways, which is essential for a therapeutic drug discovery. Furthermore, it is possible that GWAS just cannot detect rare variants that may explain the missing heritability of CAD. To overcome this limitation, whole-exome sequencing technique is being adopted in a case-control association study of complex diseases. The TG and HDL Working Group of the Exome Sequencing Project identified an aggregate of rare mutations in the gene encoding apolipoprotein C3 (APOC3). They found that the aggregate is associated with a lower plasma triglyceride level and a risk of CAD among the carriers, which was 40% lower than among the noncarriers.³¹

Mendelian Randomization (MR)

Epidemiologic research has discovered hundreds of biomarkers that associate with CAD.³² Because observational studies cannot draw conclusions about causal relationships, the potential mechanism linking the biomarkers and CAD has been investigated by basic experiments and bench studies.^33,34 If the biomarkers are proven to be determinants of the disease in the basic research, the pharmacologists or biotechnologists scrutinize the targets, develop therapeutic agents, and try clinical application. Nevertheless, researchers in field of atherosclerosis biology have rarely succeeded in clinically applying therapeutic agents. Unfortunately, it is only after spending immense time and money that we can judge the validity of the therapeutic agent(s).

A new approach, referred to as MR, can help overcome this limitation and possibly can determine the causal relationship between biomarkers and clinical outcome. It also gives a clue to whether the intervention targeting the biomarker will be effective prior to clinical trials. MR is an observational study that compares the disease state between individuals with and without a certain genotype that predisposes to a phenotype for which we are trying to assess a cause-and-effect relationship. Because genotypes are naturally randomized during sexual reproduction based on Mendel’s second law of independent assortment, the technique is named MR. For example, Timpson et al used SNPs governing the abundance of C-reactive protein (CRP) to assess whether associations between CRP and components of the metabolic syndrome are causal.³⁵

Recent large-scale genetic research including GWAS has created the foundation for MR studies. MR is now rapidly being applied to CAD and many biomarkers have been investigated (Table 2). The results are astonishing enough to challenge previous notions of what is causing CAD. Here are some examples.

Table 2. Mendelian Randomization Studies for CAD

Biomarker	SNPs	Genes	Effect of SNP on CAD	Reference
Positive causal relationship
Blood pressure	Multiple (n=30)	Multiple	+	36
Fetuin-A	Multiple (n=5)	AHSG	+	37
IL-6	rs2228145	IL-6R	+	38
	rs7529229	IL-6R	+	39
	rs1800795	IL-6	+	40
LDL	rs2228671	LDLR	+	41
	Multiple (n=3)	PCSK9	+	42
Myeloperoxidase	Multiple (n=8)	Multiple	+	43
Nonfasting glucose	Multiple (n=5)	Multiple	+	44
PAI-1	rs1799889	PAI-1	+	45
Telomere length	Multiple (n=7)	Multiple	+	46
Triglycerides	Multiple (n=44)	Multiple	+	47
Negative causal relationship
25-hydroxyvitamin D	Multiple (n=6)	GC, DHCR7	−	48
Bilirubin	rs6742078	UGT1A1	−	49
CRP	Multiple (n=5)	Multiple	−	50
	Multiple (n=4)	CRP	−	51
	Multiple (n=3)	CRP	−	52
Fibrinogen	Multiple (n=24)	Multiple	−	53
HDL	rs61755018	LIPG	−	54
	Multiple (n=4)	ABCA1	−	55
	rs4986970	LCAT	−	56
Homocysteine	Multiple (n=13)	Multiple	−	57
Lp-PLA2	rs1051931	PLA2G7	−	58
PON1	rs2267829	PON 1	−	59
PTX3	Multiple (n=3)	Multiple	−	60
sPLA2-IIA	rs11573156	PLA2G2A	−	61
Uric acid	rs7442295	SLC2A9	−	62
	rs13129697	SLC2A9	−	63
Uncertain relationship
LPA	Repeat polymorphism	KIV-2	+	64
	rs783147/genetic score	LPA gene	−	65
Adiponectin*	Multiple (n=17)	Multiple	+	66
Body mass index	Multiple (n=14)	Multiple	−	67
	Multiple (n=3)	Multiple	+	68

*Not a true Mendelian randomization trial because pleiotrophic effects of the genetic variants cannot be excluded.

CAD, coronary artery disease; CRP, C-reactive protein; HDL, high-density lipoprotein; IL, interleukin; LDL, low-density lipoprotein; LPA, Lipoprotein (a); SNP, single-nucleotide polymorphism.

High-Density Lipoprotein (HDL) Cholesterol

The most striking example is HDL cholesterol. Although a low serum HDL cholesterol level has been repeatedly observed with the incidence of CAD,^69–71 the causal relationship has never been established. Fibrates, nicotinic acid, and CETP-inhibitors have been used to increase the HDL cholesterol level in patients in the hope of preventing CAD, yet HDL has not been successfully exploited as therapy.⁷² The clinical trial and MR study results consistently suggested there is not a causal relationship between the HDL cholesterol level and CAD. Frikke-Schmidt et al used 4 ABCA1 mutations (P1065S, G1216V, N1800H, R2144X) associated with a 17 mg/dl reduction in HDL cholesterol but found no association with IHD.⁵⁵ The same researchers tested MI using a variant (rs4986970, S208T) in lecithin-cholesterol acyltransferase and had comparable conclusions to their previous study.⁵⁶ A recent finding is that a SNP in the endothelial lipase gene (LIPG N396S) and a genetic score consisting of 14 common SNPs, which was related to a higher HDL cholesterol level, are not associated with alower risk of MI.⁵⁴ In contrast to other SNPs specifically related to HDL, cholesterol ester transfer protein (CETP) has reproducibly shown its association with CAD according to the MR study; CETP facilitates the transfer of cholesteryl esters from HDL to LDL cholesterol and dictates the plasma concentration of the 2 lipoproteins.^73,74 Therefore, it is unclear at this time whether the effect of lower CETP activity on CAD is related to the change in HDL or LDL cholesterol level.

CRP

Another important biomarker whose causal effect has been thrown into doubt by MR study is CRP, which is an acute-phase protein that increases in response to inflammation. CRP has been shown in several large studies to be strongly associated with IHD and vascular death in a variety of clinical settings.^75–79 Furthermore, many of the results from basic experiments have suggested a causal role of CRP in the pathogenesis of atherosclerosis and vascular events.^80–85 Thus, some notable efforts to develop CRP inhibitors to reduce IHD are being made. However, the results from MR studies do not support such efforts. Zacho et al report that genotypic combinations of the 4 CRP polymorphisms (rs3091244, rs1130864, rs1205, and rs3093077) were associated with an increase in CRP levels up to 64%, but that these genotype combinations were not associated with an increased risk of ischemic vascular disease.⁵¹ Another study by the C Reactive Protein Coronary Heart Disease Genetics Collaboration (CCGC) used 4 similar CRP genes tagging SNPs (rs3093077, rs1205, rs1130864, rs1800947), each associated with up to 30% per allele difference in the concentration of CRP, and confirmed the genetic variations were not associated with the risk of CAD.⁵² These findings indicate that CRP alone does not cause cardiovascular disease (CVD) and that lowering CRP will not lessen the occurrence of CVD.

Interleukin-6 (IL-6)

Although CRP is not independently associated with IHD in MR studies, inflammation is considered to play a critical role in atherothrombosis, so several antiinflammatory agents are undergoing evaluation as potential atheroprotective drugs.⁸⁶ IL-6 is a key target of antiinflammatory treatment for atherosclerosis. Epidemiological data consistently reflect a correlation between a high circulating level of IL-6 and an increased risk of IHD.^87,88 IL-6 binds to its receptor (IL-6R) and therefore genetic variation in IL-6R can modulate the inflammatory effect in IHD. IL-6R SNP (rs7529229) marking a non-synonymous IL-6R variant (rs8192284; D358A) was associated with increased circulating levels of IL-6 and also associated with a decreased odds of CAD.³⁹ Another group examined the rs2228145 variant that affects the same amino acid (D358A) and had a consistent result.³⁸ These findings together strongly support the concept that targeting IL-6 signaling may be a novel method of preventing CAD.

Study Limitations

There are key assumptions for MR studies and they must be strictly met for an appropriate interpretation. First, there should be no pleiotropy of a variant. A genetic variant of interest must have no other function than to affect the level of the biomarker of intereset. If it does, the outcome must be independent from the influence of the variant’s other functions. For example, Dastani et al tested whether adiponectin-decreasing alleles influenced the probability of CAD occurrence in the CARDIoGRAM consortium. Among 145 SNPs related to adiponectin level, they found 5 variants that were associated with an increased risk of CAD. However, the study was unable to take into account the difference between SNPs that directly influence CAD through the adiponectin level and those that influence CAD directly but also influence the adiponectin level. Thus, one can say there certainly is evidence of a shared allelic architecture between adiponectin and CAD but never a causal relationship. Second, linkage disequilibrium with other functional variants must be taken into account. Genomic loci that are in close proximity on a given chromosome are more likely to be inherited together, resulting in linkage disequilibrium. If SNP “A” affecting the expression of biomarker A is in linkage disequilibrium with SNP “B” that affects the disease outcome, one may easily conclude that the SNP “A” is responsible for the disease outcome. However, it may or may not be the case. Therefore, when performing a MR study, it is ideal to use specific SNPs that are positioned in genomic regions where the linkage disequilibrium effect can be ruled out.⁸⁹

Third, there should be no hidden population strata. If a population under investigation is not homogeneous, any disease that runs at a higher prevalence in a subpopulation may display an association with all SNPs that are predominantly found in that subgroup. For example, in a North American cohort a variant may seem to be associated with type 2 diabetes because of Native American descent, which is known to have a greater incidence of the disease.⁹⁰ Therefore, it is ideal for a MR study to be performed in the same population, although practically it is very hard. Lastly, the sample size must be considered when there is a negative finding. The effect of a SNP on a biomarker can be small because there are multiple genetic variants associated with the biomarker, together with numerous environmental factors influencing the variability of the biomarker. Therefore, a negative result might simply imply that it is hard to say a small difference in the phenotype of a limited sample size is statistically significant.

Conclusions

The traditional approach to developing a new drug mostly used time-consuming epidemiologic studies or laboratory experiments to create a list of potential drug targets. However, those costly methods were fraught with a high risk of failure because the studies can not provide a direct cause of a disease and fail to capture inherent genetic differences between the subjects of interest. However, a ground-breaking advance in technology has brought a new era in genetic studies, providing a reliable list up of potential therapeutic targets of CAD. Specifically, the NGS platform used in linkage analysis or case-control studies and GWAS have offered promising candidate determinants of CAD. Furthermore, MR study has facilitated the unraveling process, with considerable evidence of causality of biomarkers for CAD in humans. These powerful genetic tools are synergistically providing great opportunities to uncover the pathogenic mechanism of CAD and to identify novel therapeutic targets, not only for the satisfaction of geneticists but also for general medical researchers who are sincerely interested in overcoming CAD.

Funding

This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI12C16910000).

References

• 1.    O’Donnell  CJ,  Nabel  EG. Genomics of cardiovascular disease. N Engl J Med 2011; 36: 2098–2109.
• 2.    Abifadel  M,  Varret  M,  Rabes  JP,  Allard  D,  Ouguerram  K,  Devillers  M, et al. Mutations in PCSK9 cause autosomal dominant hypercholesterolemia. Nat Genet 2003; 34: 154–156.
• 3.    Park  SW,  Moon  YA,  Horton  JD. Post-transcriptional regulation of low density lipoprotein receptor protein by proprotein convertase subtilisin/kexin type 9a in mouse liver. J Biol Chem 2004; 279: 50630–50638.
• 4.    Maxwell  KN,  Breslow  JL. Adenoviral-mediated expression of Pcsk9 in mice results in a low-density lipoprotein receptor knockout phenotype. Proc Natl Acad Sci USA 2004; 101: 7100–7105.
• 5.    Cohen  J,  Pertsemlidis  A,  Kotowski  IK,  Graham  R,  Garcia  CK,  Hobbs  HH. Low LDL cholesterol in individuals of African descent resulting from frequent nonsense mutations in PCSK9. Nat Genet 2005; 37: 161–165.
• 6.    Berge  KE,  Ose  L,  Leren  TP. Missense mutations in the PCSK9 gene are associated with hypocholesterolemia and possibly increased response to statin therapy. Arterioscler Thromb Vasc Biol 2006; 26: 1094–1100.
• 7.    Kotowski  IK,  Pertsemlidis  A,  Luke  A,  Cooper  RS,  Vega  GL,  Cohen  JC, et al. A spectrum of PCSK9 alleles contributes to plasma levels of low-density lipoprotein cholesterol. Am J Hum Genet 2006; 78: 410–422.
• 8.    Yue  P,  Averna  M,  Lin  X,  Schonfeld  G. The c.43_44insCTG variation in PCSK9 is associated with low plasma LDL-cholesterol in a Caucasian population. Hum Mutat 2006; 27: 460–466.
• 9.    Roth  EM,  McKenney  JM,  Hanotin  C,  Asset  G,  Stein  EA. Atorvastatin with or without an antibody to PCSK9 in primary hypercholesterolemia. N Engl J Med 2012; 367: 1891–1900.
• 10.    Blom  DJ,  Hala  T,  Bolognese  M,  Lillestol  MJ,  Toth  PD,  Burgess  L, et al. A 52-week placebo-controlled trial of evolocumab in hyperlipidemia. N Engl J Med 2014; 370: 1809–1819.
• 11.    Stein  EA,  Mellis  S,  Yancopoulos  GD,  Stahl  N,  Logan  D,  Smith  WB, et al. Effect of a monoclonal antibody to PCSK9 on LDL cholesterol. N Engl J Med 2012; 366: 1108–1118.
• 12.    Botstein  D,  Risch  N. Discovering genotypes underlying human phenotypes: Past successes for mendelian disease, future approaches for complex disease. Nat Genet 2003; 33(Suppl): 228–237.
• 13.    Helgadottir  A,  Manolescu  A,  Thorleifsson  G,  Gretarsdottir  S,  Jonsdottir  H,  Thorsteinsdottir  U, et al. The gene encoding 5-lipoxygenase activating protein confers risk of myocardial infarction and stroke. Nat Genet 2004; 36: 233–239.
• 14.    Wang  L,  Fan  C,  Topol  SE,  Topol  EJ,  Wang  Q. Mutation of MEF2A in an inherited disorder with features of coronary artery disease. Science 2003; 302: 1578–1581.
• 15.    Varret  M,  Rabes  JP,  Saint-Jore  B,  Cenarro  A,  Marinoni  JC,  Civeira  F, et al. A third major locus for autosomal dominant hypercholesterolemia maps to 1p34.1-p32. Am J Hum Genet 1999; 64: 1378–1387.
• 16.    Ng  SB,  Turner  EH,  Robertson  PD,  Flygare  SD,  Bigham  AW,  Lee  C, et al. Targeted capture and massively parallel sequencing of 12 human exomes. Nature 2009; 461: 272–276.
• 17.    Choi  M,  Scholl  UI,  Ji  W,  Liu  T,  Tikhonova  IR,  Zumbo  P, et al. Genetic diagnosis by whole exome capture and massively parallel DNA sequencing. Proc Natl Acad Sci USA 2009; 106: 19096–19101.
• 18.    Metzker  ML. Sequencing technologies: The next generation. Nat Rev Genet 2010; 11: 31–46.
• 19.    Ng  SB,  Buckingham  KJ,  Lee  C,  Bigham  AW,  Tabor  HK,  Dent  KM, et al. Exome sequencing identifies the cause of a mendelian disorder. Nat Genet 2010; 42: 30–35.
• 20.    Musunuru  K,  Pirruccello  JP,  Do  R,  Peloso  GM,  Guiducci  C,  Sougnez  C, et al. Exome sequencing, ANGPTL3 mutations, and familial combined hypolipidemia. N Engl J Med 2010; 363: 2220–2227.
• 21.    Marian  AJ,  Belmont  J. Strategic approaches to unraveling genetic causes of cardiovascular diseases. Circ Res 2011; 108: 1252–1269.
• 22.    Erdmann  J,  Stark  K,  Esslinger  UB,  Rumpf  PM,  Koesling  D,  de Wit  C, et al. Dysfunctional nitric oxide signalling increases risk of myocardial infarction. Nature 2013; 504: 432–436.
• 23.    Keramati  AR,  Fathzadeh  M,  Go  GW,  Singh  R,  Choi  M,  Faramarzi  S, et al. A form of the metabolic syndrome associated with mutations in DYRK1B. N Engl J Med 2014; 370: 1909–1919.
• 24.    LaFramboise  T. Single nucleotide polymorphism arrays: A decade of biological, computational and technological advances. Nucleic Acids Res 2009; 37: 4181–4193.
• 25.    Altshuler  D,  Daly  MJ,  Lander  ES. Genetic mapping in human disease. Science 2008; 322: 881–888.
• 26.    Roberts  R. Genetics of coronary artery disease. Circ Res 2014; 114: 1890–1903.
• 27.    Lieb  W,  Vasan  RS. Genetics of coronary artery disease. Circulation 2013; 128: 1131–1138.
• 28.    Schunkert  H,  Konig  IR,  Kathiresan  S,  Reilly  MP,  Assimes  TL,  Holm  H, et al. Large-scale association analysis identifies 13 new susceptibility loci for coronary artery disease. Nat Genet 2011; 43: 333–338.
• 29.    Erdmann  J,  Grosshennig  A,  Braund  PS,  Konig  IR,  Hengstenberg  C,  Hall  AS, et al. New susceptibility locus for coronary artery disease on chromosome 3q22.3. Nat Genet 2009; 41: 280–282.
• 30.    Risch  N,  Merikangas  K. The future of genetic studies of complex human diseases. Science 1996; 273: 1516–1517.
• 31.   TG and HDL Working Group of the Exome Sequencing Project, National Heart, Lung, and Blood Institute,  Crosby  J,  Peloso  GM,  Auer  PL,  Crosslin  DR,  Stitziel  NO,  Lange  LA, et al. Loss-of-function mutations in APOC3, triglycerides, and coronary disease. N Engl J Med 2014; 371: 22–31.
• 32.    Lee  SE,  Kim  HS. Human resistin in cardiovascular disease. J Smooth Muscle Res 2012; 48: 27–35.
• 33.    Cho  Y,  Lee  SE,  Lee  HC,  Hur  J,  Lee  S,  Youn  SW, et al. Adipokine resistin is a key player to modulate monocytes, endothelial cells, and smooth muscle cells, leading to progression of atherosclerosis in rabbit carotid artery. J Am Coll Cardiol 2011; 57: 99–109.
• 34.    Lee  S,  Lee  HC,  Kwon  YW,  Lee  SE,  Cho  Y,  Kim  J, et al. Adenylyl cyclase-associated protein 1 is a receptor for human resistin and mediates inflammatory actions of human monocytes. Cell Metab 2014; 19: 484–497.
• 35.    Timpson  NJ,  Lawlor  DA,  Harbord  RM,  Gaunt  TR,  Day  IN,  Palmer  LJ, et al. C-reactive protein and its role in metabolic syndrome: Mendelian randomisation study. Lancet 2005; 366: 1954–1959.
• 36.    Lieb  W,  Jansen  H,  Loley  C,  Pencina  MJ,  Nelson  CP,  Newton-Cheh  C, et al. Genetic predisposition to higher blood pressure increases coronary artery disease risk. Hypertension 2013; 61: 995–1001.
• 37.    Fisher  E,  Stefan  N,  Saar  K,  Drogan  D,  Schulze  MB,  Fritsche  A, et al. Association of AHSG gene polymorphisms with fetuin-A plasma levels and cardiovascular diseases in the EPIC-Potsdam study. Circ Cardiovasc Genet 2009; 2: 607–613.
• 38.   IL6R Genetics Consortium Emerging Risk Factors Collaboration,  Sarwar  N,  Butterworth  AS,  Freitag  DF,  Gregson  J,  Willeit  P,  Gorman  DN, et al. Interleukin-6 receptor pathways in coronary heart disease: A collaborative meta-analysis of 82 studies. Lancet 2012; 379: 1205–1213.
• 39.   The Interleukin-6 Receptor Mendelian Randomisation Analysis (IL6R MR) Collaboration,  Hingorani  AD,  Casas  JP. The interleukin-6 receptor as a target for prevention of coronary heart disease: A mendelian randomisation analysis. Lancet 2012; 379: 1214–1224.
• 40.    Niu  W,  Liu  Y,  Qi  Y,  Wu  Z,  Zhu  D,  Jin  W. Association of interleukin-6 circulating levels with coronary artery disease: A meta-analysis implementing mendelian randomization approach. Int J Cardiol 2012; 157: 243–252.
• 41.    Linsel-Nitschke  P,  Gotz  A,  Erdmann  J,  Braenne  I,  Braund  P,  Hengstenberg  C, et al. Lifelong reduction of LDL-cholesterol related to a common variant in the LDL-receptor gene decreases the risk of coronary artery disease: A mendelian randomisation study. PLoS One 2008; 3: e2986, doi:10.1371/journal.pone.0002986.
• 42.    Cohen  JC,  Boerwinkle  E,  Mosley  TH Jr,  Hobbs  HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. N Engl J Med 2006; 354: 1264–1272.
• 43.    Scharnagl  H,  Kleber  ME,  Genser  B,  Kickmaier  S,  Renner  W,  Weihrauch  G, et al. Association of myeloperoxidase with total and cardiovascular mortality in individuals undergoing coronary angiography: The LURIC study. Int J Cardiol 2014; 174: 96–105.
• 44.    Benn  M,  Tybjaerg-Hansen  A,  McCarthy  MI,  Jensen  GB,  Grande  P,  Nordestgaard  BG. Nonfasting glucose, ischemic heart disease, and myocardial infarction: A Mendelian randomization study. J Am Coll Cardiol 2012; 59: 2356–2365.
• 45.    Nikolopoulos  GK,  Bagos  PG,  Tsangaris  I,  Tsiara  CG,  Kopterides  P,  Vaiopoulos  A, et al. The association between plasminogen activator inhibitor type 1 (PAI-1) levels, PAI-1 4G/5G polymorphism, and myocardial infarction: A Mendelian randomization meta-analysis. Clin Chem Lab Med 2014; 52: 937–950.
• 46.    Codd  V,  Nelson  CP,  Albrecht  E,  Mangino  M,  Deelen  J,  Buxton  JL, et al. Identification of seven loci affecting mean telomere length and their association with disease. Nat Genet 2013; 45: 422–427, 427e1–e2, doi:10.1038/ng.2528.
• 47.    Do  R,  Willer  CJ,  Schmidt  EM,  Sengupta  S,  Gao  C,  Peloso  GM, et al. Common variants associated with plasma triglycerides and risk for coronary artery disease. Nat Genet 2013; 45: 1345–1352.
• 48.    Strawbridge  RJ,  Deleskog  A,  McLeod  O,  Folkersen  L,  Kavousi  M,  Gertow  K, et al. A serum 25-hydroxyvitamin D concentration-associated genetic variant in DHCR7 interacts with type 2 diabetes status to influence subclinical atherosclerosis (measured by carotid intima-media thickness). Diabetologia 2014; 57: 1159–1172.
• 49.    Stender  S,  Frikke-Schmidt  R,  Nordestgaard  BG,  Grande  P,  Tybjaerg-Hansen  A. Genetically elevated bilirubin and risk of ischaemic heart disease: Three Mendelian randomization studies and a meta-analysis. J Intern Med 2013; 273: 59–68.
• 50.    Elliott  P,  Chambers  JC,  Zhang  W,  Clarke  R,  Hopewell  JC,  Peden  JF, et al. Genetic Loci associated with C-reactive protein levels and risk of coronary heart disease. JAMA 2009; 302: 37–48.
• 51.    Zacho  J,  Tybjaerg-Hansen  A,  Jensen  JS,  Grande  P,  Sillesen  H,  Nordestgaard  BG. Genetically elevated C-reactive protein and ischemic vascular disease. N Engl J Med 2008; 359: 1897–1908.
• 52.   Collaboration C Reactive Protein Coronary Heart Disease Genetics Collaboration (CCGC),  Wensley  F,  Gao  P,  Burgess  S,  Kaptoge  S,  Di Angelantonio  E,  Shah  T, et al. Association between C reactive protein and coronary heart disease: Mendelian randomisation analysis based on individual participant data. BMJ 2011; 342: d548.
• 53.    Sabater-Lleal  M,  Huang  J,  Chasman  D,  Naitza  S,  Dehghan  A,  Johnson  AD, et al. Multiethnic meta-analysis of genome-wide association studies in >100 000 subjects identifies 23 fibrinogen-associated Loci but no strong evidence of a causal association between circulating fibrinogen and cardiovascular disease. Circulation 2013; 128: 1310–1324.
• 54.    Voight  BF,  Peloso  GM,  Orho-Melander  M,  Frikke-Schmidt  R,  Barbalic  M,  Jensen  MK, et al. Plasma HDL cholesterol and risk of myocardial infarction: A mendelian randomisation study. Lancet 2012; 380: 572–580.
• 55.    Frikke-Schmidt  R,  Nordestgaard  BG,  Stene  MC,  Sethi  AA,  Remaley  AT,  Schnohr  P, et al. Association of loss-of-function mutations in the ABCA1 gene with high-density lipoprotein cholesterol levels and risk of ischemic heart disease. JAMA 2008; 299: 2524–2532.
• 56.    Haase  CL,  Tybjaerg-Hansen  A,  Qayyum  AA,  Schou  J,  Nordestgaard  BG,  Frikke-Schmidt  R. LCAT, HDL cholesterol and ischemic cardiovascular disease: A Mendelian randomization study of HDL cholesterol in 54,500 individuals. J Clin Endocrinol Metab 2012; 97: E248–E256, doi:10.1210/jc.2011-1846.
• 57.    van Meurs  JB,  Pare  G,  Schwartz  SM,  Hazra  A,  Tanaka  T,  Vermeulen  SH, et al. Common genetic loci influencing plasma homocysteine concentrations and their effect on risk of coronary artery disease. Am J Clin Nutr 2013; 98: 668–676.
• 58.    Casas  JP,  Ninio  E,  Panayiotou  A,  Palmen  J,  Cooper  JA,  Ricketts  SL, et al. PLA2G7 genotype, lipoprotein-associated phospholipase A2 activity, and coronary heart disease risk in 10 494 cases and 15 624 controls of European ancestry. Circulation 2010; 121: 2284–2293.
• 59.    Tang  WH,  Hartiala  J,  Fan  Y,  Wu  Y,  Stewart  AF,  Erdmann  J, et al. Clinical and genetic association of serum paraoxonase and arylesterase activities with cardiovascular risk. Arterioscler Thromb Vasc Biol 2012; 32: 2803–2812.
• 60.    Barbati  E,  Specchia  C,  Villella  M,  Rossi  ML,  Barlera  S,  Bottazzi  B, et al. Influence of pentraxin 3 (PTX3) genetic variants on myocardial infarction risk and PTX3 plasma levels. PLoS One 2012; 7: e53030, doi:10.1371/journal.pone.0053030.
• 61.    Holmes  MV,  Simon  T,  Exeter  HJ,  Hingorani  AD,  Sabatine  MS,  Mallat  Z, et al. Reply: Limits of Mendelian randomization analyses in selection of secretory phospholipase A2-IIA as a valid therapeutic target for prevention of cardiovascular disease. J Am Coll Cardiol 2014; 63: 943.
• 62.    Oikonen  M,  Wendelin-Saarenhovi  M,  Lyytikainen  LP,  Siitonen  N,  Loo  BM,  Jula  A, et al. Associations between serum uric acid and markers of subclinical atherosclerosis in young adults: The Cardiovascular Risk in Young Finns study. Atherosclerosis 2012; 223: 497–503.
• 63.    Palmer  TM,  Nordestgaard  BG,  Benn  M,  Tybjaerg-Hansen  A,  Davey Smith  G,  Lawlor  DA, et al. Association of plasma uric acid with ischaemic heart disease and blood pressure: Mendelian randomisation analysis of two large cohorts. BMJ 2013; 347: f4262.
• 64.    Kamstrup  PR,  Tybjaerg-Hansen  A,  Steffensen  R,  Nordestgaard  BG. Genetically elevated lipoprotein(a) and increased risk of myocardial infarction. JAMA 2009; 301: 2331–2339.
• 65.    Kivimaki  M,  Magnussen  CG,  Juonala  M,  Kahonen  M,  Kettunen  J,  Loo  BM, et al. Conventional and Mendelian randomization analyses suggest no association between lipoprotein(a) and early atherosclerosis: The Young Finns Study. Int J Epidemiol 2011; 40: 470–478.
• 66.    Dastani  Z,  Johnson  T,  Kronenberg  F,  Nelson  CP,  Assimes  TL,  Marz  W, et al. The shared allelic architecture of adiponectin levels and coronary artery disease. Atherosclerosis 2013; 229: 145–148.
• 67.    Holmes  MV,  Lange  LA,  Palmer  T,  Lanktree  MB,  North  KE,  Almoguera  B, et al. Causal effects of body mass index on cardiometabolic traits and events: A Mendelian randomization analysis. Am J Hum Genet 2014; 94: 198–208.
• 68.    Nordestgaard  BG,  Palmer  TM,  Benn  M,  Zacho  J,  Tybjaerg-Hansen  A,  Davey Smith  G, et al. The effect of elevated body mass index on ischemic heart disease risk: Causal estimates from a Mendelian randomisation approach. PLoS Med 2012; 9: e1001212, doi:10.1371/journal.pmed.1001212.
• 69.    Lin  GM,  Li  YH,  Lin  CL,  Wang  JH,  Han  CL. Low high-density lipoprotein cholesterol and low/normal body mass index are associated with increased mortality in coronary artery disease patients in Taiwan. Circ J 2013; 77: 2079–2087.
• 70.    Linton  JA,  Kimm  H,  Ohrr  H,  Park  IS,  Jee  SH. High-density lipoprotein-cholesterol and ischemic heart disease risk in Korean men with cardiac risk. Circ J 2009; 73: 1296–1301.
• 71.    Annema  W,  von Eckardstein  A. High-density lipoproteins: Multifunctional but vulnerable protections from atherosclerosis. Circ J 2013; 77: 2432–2448.
• 72.    Zhang  B,  Kawachi  E,  Miura  S,  Uehara  Y,  Matsunaga  A,  Kuroki  M, et al. Therapeutic approaches to the regulation of metabolism of high-density lipoprotein: Novel HDL-directed pharmacological intervention and exercise. Circ J 2013; 77: 2651–2663.
• 73.    Hiura  Y,  Shen  CS,  Kokubo  Y,  Okamura  T,  Morisaki  T,  Tomoike  H, et al. Identification of genetic markers associated with high-density lipoprotein-cholesterol by genome-wide screening in a Japanese population: The Suita study. Circ J 2009; 73: 1119–1126.
• 74.    Thompson  A,  Di Angelantonio  E,  Sarwar  N,  Erqou  S,  Saleheen  D,  Dullaart  RP, et al. Association of cholesteryl ester transfer protein genotypes with CETP mass and activity, lipid levels, and coronary risk. JAMA 2008; 299: 2777–2788.
• 75.    Koenig  W,  Sund  M,  Frohlich  M,  Fischer  HG,  Lowel  H,  Doring  A, et al. C-Reactive protein, a sensitive marker of inflammation, predicts future risk of coronary heart disease in initially healthy middle-aged men: Results from the MONICA (Monitoring Trends and Determinants in Cardiovascular Disease) Augsburg Cohort Study, 1984 to 1992. Circulation 1999; 99: 237–242.
• 76.    Ridker  PM,  Buring  JE,  Shih  J,  Matias  M,  Hennekens  CH. Prospective study of C-reactive protein and the risk of future cardiovascular events among apparently healthy women. Circulation 1998; 98: 731–733.
• 77.    Ridker  PM,  Glynn  RJ,  Hennekens  CH. C-reactive protein adds to the predictive value of total and HDL cholesterol in determining risk of first myocardial infarction. Circulation 1998; 97: 2007–2011.
• 78.    Liuzzo  G,  Biasucci  LM,  Gallimore  JR,  Grillo  RL,  Rebuzzi  AG,  Pepys  MB, et al. The prognostic value of C-reactive protein and serum amyloid a protein in severe unstable angina. N Engl J Med 1994; 331: 417–424.
• 79.   Emerging Risk Factors Collaboration,  Kaptoge  S,  Di Angelantonio  E,  Lowe  G,  Pepys  MB,  Thompson  SG,  Collins  R, et al. C-reactive protein concentration and risk of coronary heart disease, stroke, and mortality: An individual participant meta-analysis. Lancet 2010; 375: 132–140.
• 80.    Teoh  H,  Quan  A,  Lovren  F,  Wang  G,  Tirgari  S,  Szmitko  PE, et al. Impaired endothelial function in C-reactive protein overexpressing mice. Atherosclerosis 2008; 201: 318–325.
• 81.    Verma  S,  Kuliszewski  MA,  Li  SH,  Szmitko  PE,  Zucco  L,  Wang  CH, et al. C-reactive protein attenuates endothelial progenitor cell survival, differentiation, and function: Further evidence of a mechanistic link between C-reactive protein and cardiovascular disease. Circulation 2004; 109: 2058–2067.
• 82.    Pasceri  V,  Willerson  JT,  Yeh  ET. Direct proinflammatory effect of C-reactive protein on human endothelial cells. Circulation 2000; 102: 2165–2168.
• 83.    Chang  MK,  Binder  CJ,  Torzewski  M,  Witztum  JL. C-reactive protein binds to both oxidized LDL and apoptotic cells through recognition of a common ligand: Phosphorylcholine of oxidized phospholipids. Proc Natl Acad Sci USA 2002; 99: 13043–13048.
• 84.    Bhakdi  S,  Torzewski  M,  Paprotka  K,  Schmitt  S,  Barsoom  H,  Suriyaphol  P, et al. Possible protective role for C-reactive protein in atherogenesis: Complement activation by modified lipoproteins halts before detrimental terminal sequence. Circulation 2004; 109: 1870–1876.
• 85.    Park  KW,  Kwon  YW,  Cho  HJ,  Shin  JI,  Kim  YJ,  Lee  SE, et al. G-CSF exerts dual effects on endothelial cells: Opposing actions of direct eNOS induction versus indirect CRP elevation. J Mol Cell Cardiol 2008; 45: 670–678.
• 86.    Ridker  PM,  Luscher  TF. Anti-inflammatory therapies for cardiovascular disease. Eur Heart J 2014; 35: 1782–1791.
• 87.    Kaptoge  S,  Seshasai  SR,  Gao  P,  Freitag  DF,  Butterworth  AS,  Borglykke  A, et al. Inflammatory cytokines and risk of coronary heart disease: New prospective study and updated meta-analysis. Eur Heart J 2014; 35: 578–589.
• 88.    Omicron Hartaigh  B,  Thomas  GN,  Bosch  JA,  Hemming  K,  Pilz  S,  Loerbroks  A, et al. Evaluation of 9 biomarkers for predicting 10-year cardiovascular risk in patients undergoing coronary angiography: Findings from the LUdwigshafen RIsk and Cardiovascular Health (LURIC) study. Int J Cardiol 2013; 168: 2609–2615.
• 89.    Jansen  H,  Samani  NJ,  Schunkert  H. Mendelian randomization studies in coronary artery disease. Eur Heart J 2014; 35: 1917–1924.
• 90.    Burgess  S,  Butterworth  A,  Malarstig  A,  Thompson  SG. Use of Mendelian randomisation to assess potential benefit of clinical intervention. BMJ 2012; 345: e7325, doi:10.1136/bmj.e7325.