ars.els-cdn.com · Web viewIndividual level genetic data was available from 134,941 individuals in UK Biobank, after excluding related samples, individuals whose genetic sex did not

ONLINE APPENDIX

Supplemental Methods

Non-invasive determination of coronary atherosclerosisCoronary computed tomography angiography was performed on a 128-row detector scanner

(Siemens Definition Edge, Siemens, Erlangen Germany) in two phases: a non-contrast acquisition for coronary artery calcium scoring and an angiographic phase for plaque characterization and evaluation of coronary artery stenosis.

The images for the coronary artery calcium score were acquired with a gantry rotation time of 0.33 ms utilizing 120 kVp and 80 mAs. Images were acquired with prospective cardiac gating to minimize radiation exposure. A collimation of 0.6 mm and pitch of 1 was utilized for scanning. Images were reconstructed and obtained at 55% of the cardiac cycle (mid diastolic phase) at 3 mm slice thickness with a reconstruction interval of 3 mm. The CTDIvol was 4.3. Images for the coronary angiography were acquired following the intravenous administration of 100 mL Optiray 350 contrast agent at 5 mL/s. The images were acquired utilizing retrospective cardiac gating with the application of 4D CareDose to minimize radiation exposure. The gantry rotation time was 0.33 ms utilizing 120 kVp and a reference mAs of 770. The pitch was 0.2. Images were reconstructed in mid diastolic phase from 50-75% of the RR interval at 5% increments. Reconstruction slice thickness was 0.75 mm with a reconstruction interval of 0.4 mm. The CTDIvol was 61.8.

The coronary artery calcium score and coronary angiography images were transferred and analyzed on a Vital workstation utilizing VitraCore v6.7 (Vital Images, Minnetonka, MN, USA). Utilizing Vscore on the workstation, the calcium score was calculated utilizing the Agatston method(1-4). Calcium in each of the coronary arteries was manually selected and then summed for all segments to provide a composite calcium score for the entire coronary artery tree. Plaque characterization was performed utilizing SurePlaque on the Vital workstation. The characterization included determination of plaque composition, as well as, the percent plaque of the entire coronary arteries. The plaque composition determination was based upon Hounsfield units, such that soft plaque was determined that the Hounsfield units (HU) of -150 to 30, fibrous plaque was 30 HU to 149 HU and calcified plaque was greater than 140 HU (see Supplementary Figure 2). In addition, the overall percentage of plaque burden was determined. This is based upon the total wall area divided by total vessel area averaged over all segments of the coronary arteries(5-7).

A single board-certified radiologist with clinical expertise in coronary imaging interpreted all images while blinded to participant mutation status.

DNA sequencing and genotype imputationSequence data for the seven protein coding exons of the ANGPTL3 gene were extracted from

whole exome sequences in 11 previously described studies of cases with coronary artery disease and coronary artery disease free controls as reported previously(8). The studies included: 1) the Italian Atherosclerosis Thrombosis and Vascular Biology (ATVB) study (dbGaP Study Accession phs000814.v1.p1); 2) the Exome Sequencing Project Early-Onset Myocardial Infarction (ESP-EOMI) study(9); 3) a nested case-control cohort from the Jackson Heart Study (JHS); 4) the South German Myocardial Infarction study (dbGaP Study Accession phs000916.v1.p1); 5) the Ottawa Heart Study (OHS) (dbGaP Study Accession phs000806.v1.p1); 6) the Precocious Coronary Artery Disease (PROCARDIS) study (dbGaP Study Accession phs000883.v1.p1) ; 7) the Pakistan Risk of Myocardial Infarction Study (PROMIS) (dbGaP Study Accession phs000917.v1.p1); 8) the Registre Gironi del

COR (Gerona Heart Registry or REGICOR) study (dbGaP Study Accession phs000902.v1.p1); 9) the Leicester Myocardial Infarction study (dbGaP Study Accession phs001000.v1.p1); 10) the BioImage study (dbGaP Study Accession phs001058.v1.p1); 11) and the North German Myocardial Infarction study (dbGaP Study Accession phs000990.v1.p1). Participants of the Atherosclerosis Risk in Communities population-based prospective cohort study underwent exome sequencing as reported previously(10). Data from targeted sequencing of ANGPTL3 exons using the Illumina TrueSeq Custom Amplicon approach was available in participants of the Duke CATHGEN study. Additional details for each study are shown in Table S1. Loss-of-function mutations in ANGPTL3 included those leading to a premature stop codon, variants that disrupted exon splicing, or frameshift mutations that introduce premature truncation due to an alteration in the open reading frame.

Imputation of an intronic splice variant in ANGPTL3Additional data for rs372257803, an intronic splice region variant in ANGPTL3 previously

linked to significantly reduced circulating triglyceride levels(11) was obtained by imputation in a nested case-control study derived from the UK Biobank(12,13). Individual level genetic data was available from 134,941 individuals in UK Biobank, after excluding related samples, individuals whose genetic sex did not match self-reported sex and extreme outliers. Of these individuals, 43,839 were genotyped using the Affymetrix UK BiLEVE array and 91,102 were genotyped using the Affymetrix UK Biobank Axiom Array. Phasing and imputation were performed centrally, by UK Biobank, using a reference panel combining UK10k and 1000 Genomes samples. 73,355,667 variants were imputed. rs372257803 was successfully imputed in 133,886 individuals with an INFO score of 0.6; these individuals (111,417 of European ancestry and 22,469 of non-European ancestry) formed the sample for analysis.

rs372257803 was similarly imputed in individuals of European ancestry with available genotype information derived from the PennCath study (genotyped using the Affymetrix 6.0 gene chip)(14) and the Wellcome Trust Case Control Consortium study (genotyped using the Affymetrix GeneChip Human Mapping 500K Array Set)(15). Genotype data was then phased with the Eagle phasing algorithm(16) and imputed with Minimac3 software using the Haplotype Reference Consortium reference panel(17) through the Michigan Imputation server. High quality imputation was performed (R2 of 0.74 and 0.62 in PennCath and WTCCC respectively) and most likely genotype assigned to each individual for subsequent analyses.

Mouse modeling of ANGPTL3 missense variants of unknown significance.All animal procedures performed in this study were reviewed and approved by the pertinent

Institutional Animal Care and Use Committees at the University of Pennsylvania and Harvard University and were consistent with local, state, and federal regulations as applicable. Euthanasia in all instances was via terminal inhalation of carbon dioxide, consistent with the 2013 AVMA Guidelines on Euthanasia. All mice were fed a standard chow diet and maintained on a 12 hour light/12 hour dark cycle.

In order to generate an Angptl3 knockout mouse model, we designed two clustered regularly interspaced short palindromic repeats (CRISPR) guide RNAs with cleavage sites flanking the entire Angptl3 gene, with protospacer sequences 5’-GAGTCACACTCTTAGTAATT-3’ and 5’-TAATGCCAATCCACGAGCAT-3’. These guide RNA sequences were PCR amplified, and the purified PCR products were used as templates for in vitro transcription using the MEGAshortscript T7 kit (Life Technologies). The transcribed RNAs were purified by phenol/chloroform extraction, ethanol precipitated, and resuspended in injection buffer (5 mM Tris-HCl pH 7.6, 0.1 mM EDTA). One-cell

embryo injections were performed by the Genome Modification Facility at Harvard University. C57BL/6J females (The Jackson Laboratory) were superovulated and mated with C57BL/6J males, and fertilized embryos were harvested from the oviducts. One-cell embryos were injected with a mixture of 100 ng/μL Cas9 mRNA (TriLink BioTechnologies) and 25 ng/μL of each guide RNA. Injected embryos were implanted into the uteri of pseudopregnant foster mothers. DNA was prepared from tail biopsies of 3-week-old founder mice by the hot hydroxide method, and genotyping was performed with PCR primers flanking the cleavage sites around the Angptl3 gene. Founder mice with Angptl3 deletion were bred with wild-type C57BL/6J mice, and the resulting progeny were intercrossed for one to two generations to breed the knockout allele to homozygosity. Genotyping of progeny was performed in the same manner. The Angptl3 knockout mice displayed decreased circulating triglyceride levels (~60%) and decreased total cholesterol (~30%) levels compared to wild-type littermates, as expected.

Adeno-associated virus 8 (AAV8) and adenoviral vectors expressing the human ANGPTL3 cDNA from the CAG promoter were produced by Vector BioLabs. We tested various dosages of the two viruses administered via intraperitoneal and retro-orbital injections to wild-type and Angptl3 knockout mice. We found that AAV8 at high dosages resulted in only ~50% increased triglyceride levels in knockout mice at 1 week. In contrast, adenovirus resulted in consistent, >100% increased triglyceride levels in both knockout and wild-type mice at 4 days when administered with retro-orbital injections, although concomitantly increased total cholesterol levels were only observed in knockout mice, not wild-type mice. We accordingly chose to use knockout mice for all further experiments. We found that administration of 2.5 × 108 infectious units (IFU) to knockout mice was sufficient to rescue the lipid phenotypes and approximated the triglyceride and total cholesterol levels of wild-type mice, and so we used that “physiological” dosage for further experiments.

Adenoviral vectors expressing wild-type and mutant ANGPTL3 alleles were produced by Vector BioLabs using plasmids generated by PCR mutagenesis and standard molecular biology techniques. Groups of 5 to 6 mice at about 8 weeks of age were used for studies with each adenovirus. Assignment to groups (wild-type virus, each mutant virus) was performed to achieve age- and sex-matched groups of littermates or colonymates with shared housing. Mice received 2.5 × 108 IFU of adenovirus via retro-orbital injection under inhaled isoflurane anesthesia. Blood was collected prior to injection and at 4 days post-injection via tail vein sampling following a 4-hour fast. Samples were kept on ice and then centrifuged for 10 minutes at 2000 × g in a centrifuge at 4°C. The plasma levels of triglyceride and total cholesterol were measured using Infinity Triglycerides Reagent (Thermo Fisher) and Infinity Cholesterol Reagent (Thermo Fisher), respectively, according to the manufacturer’s instructions. Mice were sacrificed at 4 days post-injection, and livers were collected. Hepatic ANGPTL3 expression was confirmed in all mice by qRT-PCR.

All animal procedures performed in this study were reviewed and approved by the pertinent Institutional Animal Care and Use Committees at the University of Pennsylvania and Harvard University.

Supplemental Figure 1. Pedigree of the family in the current study. Square shapes indicate males; circles indicate females. Shading indicates presence of a loss of function mutation in ANGPTL3 as indicated in the inset.

Supplemental Figure 2. Example of coronary artery plaque composition characterization

(A) Representative sagittal CT angiogram image of the right coronary artery in an individual with complete ANGPTL3 deficiency. Cross-sectional image of the coronary artery demonstrates no calcifications, fatty plaque or fibrous plaque. (B) Representative sagittal CT angiogram image of the right coronary artery in a matched first degree relative with normal ANGPTL3 levels a cross-section of the plaque demonstrates the composition and distribution of the atherosclerotic plaque. Red identifies the fatty plaque, yellow identifies the calcified plaque, and blue is the fibrous plaque (Supplemental Methods Section).

A B

Supplemental Figure 3. In vivo characterization of ANGPTL3 missense mutations in Angptl3 knockout mouse model

(A) Triglyceride (TG) and total cholesterol (TC) levels in Angptl3 knockout (KO) mice before and after receiving adenoviruses (AdV) expressing wild-type or mutant ANGPTL3 alleles (five to six mice in each group). Asterisks indicate groups of mice demonstrating statistically significant changes in the lipid level with AdV treatment (P < 0.05 by two-sided paired t-test). The mutant ANGPTL3 alleles were tested in batches in three distinct experiments. (B) The effects of mutant ANGPTL3 alleles expressed from AdV in KO mice (five to six in each group) on TC levels, shown as percentages of the effect of AdV expressing the wild-type ANGPTL3 allele on TC.

Supplemental Table 1: Sources of cases and controls for identifying loss-of-function mutations in ANGPTL3

Study Total number of individuals studied Case Definition Control Definition Reference

BioImage 405 Incident MI observed during follow-up Free of prevalent or incident CAD or MI (19)United Kingdom Biobank

133886Prevalent MI, CABG, or coronary revascularization at time of enrollment or hospitalization for acute myocardial infarction or

revascularization procedure during follow-up

No history of hospitalization for MI or revascularization (12)

WTCCC 4819 MI or coronary revascularization at < 66 years of ageSelected from the British 1958 Birth Cohort

and blood donors from the UK Blood Services cohort

(15)

REGICOR 784 MI in men ≤ 50 years of age or women ≤ 60 years of ageControls from a population-based study; free of MI, coronary revascularization; ≥

55 years of age and < 80 years of age(20)

South German MI 799 MI in men ≤ 40 years of age or women ≤ 55 years of age Controls without CAD, men ≥ 65 years of

age and women ≥ 75 years of age (21)

PennCath 1363 Angiography (>1 coronary vessel with > 50% stenosis); < 55 years of age for males and < 60 years of age for females

No or minimal (<10% stenosis) on angiography; men > 40 years of age and

women > 45 years of age(14)

Jackson Heart Study 2176

Combination of prevalent CHD (self-reported or electrocardiographic evidence of MI) and incident CHD (MI

or coronary revascularization)Free of CHD during follow-up (22)

ESP-EOMI 2819 MI (men ≤ 50 years of age or women ≤ 60 years of age) Hospital-based, no report of MI by history (9)

North German 1728 MI in men and women≤ 60 years of age

Controls without CAD; men and women ≤ 65 years of age (23)

Ottawa Heart Study 1922 MI or CABG or angiographic disease (>50% stenosis) in men

≤ 50 years of age or women ≤ 60 years of age) Asymptomatic (24)

ARIC 10245 Incident definite or probable MI or death related to coronary causes

Controls free of prevalent or incident MI during follow-up (10)

Duke CATHGEN 1187 MI; coronary stenosis

≥ 50%

Patients > 50 years old; no coronary stenosis greater than 30%; No history of

ICC/PCI, CABG, MI or transplant(25)

PROCARDIS 1817 MI (men ≤ 50 years of age or women ≤ 60 years of age) No history of CAD (26)ATVB 3468 MI in men or women ≤ 45 years of age No history of thromboembolic disease (27)

Leicester 2316 MI (men ≤ 50 years of age or women ≤ 60 years of age) Controls ≥ 64 years of age without reported CAD history (23)

PROMIS 10446 MI, age 30-80 years of age No history of MI or CVD (28)WTCCC: Wellcome Trust Case Control Consortium Coronary Artery Disease study; REGICOR: Registre Gironi del COR (Gerona Heart Registry); ESP-EOMI: Exome Sequencing Project Early-Onset Myocardial Infarction study; ARIC: Atherosclerosis Research in Communities study; PROCARDIS: Precocious Coronary Artery Disease Study; ATVB: Italian Atherosclerosis Thrombosis and Vascular Biology Study; PROMIS: Pakistan Risk of Myocardial Infarction Study

Supplemental Table 2: Clinical characteristics and summary results from coronary computed tomography angiography of family members with and without ANGPTL3 loss-of-function mutationsPedigree ID II-1 II-2 II-4 II-7* II-8* II-10ANGPTL3 mutation status

Compound heterozygote

Compound heterozygote

Compound heterozygote None None None

Pedigree identifier from Musunuru et al(18) II-2 II-4 II-5 II-12 II-14 II-17

Gender Male Male Female Male Male FemaleAge 62 42 61 58 50 52Total Cholesterol, mg/dl 68 59 63 133 159 151LDL, mg/dL 44 38 40 71 73 78HDL, mg/dL 15 15 18 41 33 63Triglycerides, mg/dl 47 31 30 103 263 52Lp(a), mg/dL <6 <6 <6 <6 <6 <6Glucose 157 110 99 137 102 99Type 2 DM Yes No No Yes No NoHTN Yes No No Yes No NoTobacco use Former

smoker No No Former smoker No No

Calcium score (Agatston units) 0 0 0 610 6 0

Plaque burden 0% 0% 0% 64% 53% 0%Percent of plaque fat 0% 0% 0% 16.6% 13.3% 0%All laboratory values were obtained after an overnight fast. *=Participants taking cholesterol lowering medications.

8

Supplemental Table 3: Loss-of-function mutations identified in ANGPTL3Variant(CHR:POS_REF/ALT) Consequence Amino Acid Change Control

CarriersNumber of Controls*

CAD Carriers

Number of CAD Cases*

1:63063413_AC/A Frameshift Phe60LeufsTer40 1 26198 0 139141:63063592_GAACTC/G Frameshift Asn121LysfsTer3 7 26198 6 139141:63063667_CAACT/C Frameshift Asn147Ter 18 26198 5 139141:63064445_CA/C Frameshift Gln192ArgfsTer5 19 26198 3 139141:63066786_AT/A Frameshift Ser215LeufsTer17 1 26198 0 139141:63066839_GA/G Frameshift Asn232Fs 1 26198 0 139141:63068034_GT/G Frameshift Phe306LeufsTer12 0 26198 1 139141:63069905_AG/A Frameshift N/A 0 26198 2 139141:63070368_AG/A Frameshift Ala422GlnfsTer23 2 26198 0 139141:63070433_C/CT Frameshift Ile444TyrfsTer17 1 26198 0 139141:63070434_TATAA/T Frameshift Ile444AsnfsTer6 1 26198 0 139141:63070464_AAC/A Frameshift Thr454ArgfsTer6 1 26198 0 139141:63063361_G/A Missense Asp42Asn 13 26198 6 139141:63069855_A/T Missense Thr383Ser 6 26198 4 139141:63067955_G/C Splice-site N/A 1 26198 0 139141:63063738_T/C Splice-site N/A 82 157559 7 214341:63063486_T/A Premature Stop Tyr83Ter 0 26198 1 139141:63064429_T/A Premature Stop Tyr186Ter 1 26198 0 139141:63064445_C/T Premature Stop Gln192Ter 3 26198 0 139141:63067982_C/T Premature Stop Arg288Ter 1 26198 0 139141:63069749_T/G Premature Stop Tyr347Ter 1 26198 0 139141:63070316_G/A Premature Stop Trp404Ter 1 26198 1 139141:63070483_G/T Premature Stop Glu460Ter 0 26198 1 13914Combined¶ 161 158200 37 21980Loss-of-function mutations in ANGPTL3 included those leading to a premature stop codon, a disruptive frameshift or splice-site variant, and missense variants conferring <25% of wild-type activity in a mouse model. CHR: Chromosome; POS: Chromosomal positions based on the hg19 build of the human reference genome; REF: Reference allele; ALT: Alternate allele. CAD: coronary artery disease. *Number of Controls and Number of Cases refer to the total available cases and controls according to each variant.

9

†Carriers of this variant included one homozygote. ¶Combined refers to independent numbers of Cases and Controls

10

Supplemental References

1. Carr JJ, Crouse JR, 3rd, Goff DC, Jr., D'Agostino RB, Jr., Peterson NP, Burke GL. Evaluation of subsecond gated helical CT for quantification of coronary artery calcium and comparison with electron beam CT. AJR American Journal of Roentgenology 2000;174:915-21.

2. Hoff JA, Chomka EV, Krainik AJ, Daviglus M, Rich S, Kondos GT. Age and gender distributions of coronary artery calcium detected by electron beam tomography in 35,246 adults. Am J Cardiol 2001;87:1335-9.

3. Agatston AS, Janowitz WR, Hildner FJ, Zusmer NR, Viamonte M, Jr., Detrano R. Quantification of coronary artery calcium using ultrafast computed tomography. Journal of the American College of Cardiology 1990;15:827-32.

4. Becker CR, Kleffel T, Crispin A et al. Coronary artery calcium measurement: agreement of multirow detector and electron beam CT. AJR American Journal of Roentgenology 2001;176:1295-8.

5. Sun J, Zhang Z, Lu B et al. Identification and quantification of coronary atherosclerotic plaques: a comparison of 64-MDCT and intravascular ultrasound. AJR American Journal of Roentgenology 2008;190:748-54.

6. Rinehart S, Vazquez G, Qian Z, Murrieta L, Christian K, Voros S. Quantitative measurements of coronary arterial stenosis, plaque geometry, and composition are highly reproducible with a standardized coronary arterial computed tomographic approach in high-quality CT datasets. Journal of cardiovascular computed tomography 2011;5:35-43.

7. Brodoefel H, Burgstahler C, Heuschmid M et al. Accuracy of dual-source CT in the characterisation of non-calcified plaque: use of a colour-coded analysis compared with virtual histology intravascular ultrasound. The British Journal of Radiology 2009;82:805-12.

8. Khera AV, Won HH, Peloso GM et al. Diagnostic Yield and Clinical Utility of Sequencing Familial Hypercholesterolemia Genes in Patients With Severe Hypercholesterolemia. Journal of the American College of Cardiology 2016;67:2578-89.

9. Do R, Stitziel NO, Won HH et al. Exome sequencing identifies rare LDLR and APOA5 alleles conferring risk for myocardial infarction. Nature 2015;518:102-6.

10. Li AH, Morrison AC, Kovar C et al. Analysis of loss-of-function variants and 20 risk factor phenotypes in 8,554 individuals identifies loci influencing chronic disease. Nat Genet 2015;47:640-2.

11. Helgadottir A, Gretarsdottir S, Thorleifsson G et al. Variants with large effects on blood lipids and the role of cholesterol and triglycerides in coronary disease. Nat Genet 2016;48:634-9.

12. Sudlow C, Gallacher J, Allen N et al. UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Medicine 2015;12:e1001779.

13. UK Biobank Analysis Team. Genotype imputation and genetic association studies of UK Biobank (http://wwwukbiobankacuk/wp-content/uploads/2014/04/imputation_documentation_May2015pdf). 2015.

11

14. Reilly MP, Li M, He J et al. Identification of ADAMTS7 as a novel locus for coronary atherosclerosis and association of ABO with myocardial infarction in the presence of coronary atherosclerosis: two genome-wide association studies. Lancet 2011;377:383-92.

15. Samani NJ, Erdmann J, Hall AS et al. Genomewide association analysis of coronary artery disease. N Engl J Med 2007;357:443-53.

16. Loh PR, Palamara PF, Price AL. Fast and accurate long-range phasing in a UK Biobank cohort. Nat Genet 2016;48:811-6.

17. McCarthy S, Das S, Kretzschmar W et al. A reference panel of 64,976 haplotypes for genotype imputation. Nat Genet 2016.

18. Musunuru K, Pirruccello JP, Do R et al. Exome sequencing, ANGPTL3 mutations, and familial combined hypolipidemia. N Engl J Med 2010;363:2220-7.

19. Baber U, Mehran R, Sartori S et al. Prevalence, impact, and predictive value of detecting subclinical coronary and carotid atherosclerosis in asymptomatic adults: the BioImage study. Journal of the American College of Cardiology 2015;65:1065-74.

20. Senti M, Tomas M, Marrugat J, Elosua R, Investigators R. Paraoxonase1-192 polymorphism modulates the nonfatal myocardial infarction risk associated with decreased HDLs. Arterioscler Thromb Vasc Biol 2001;21:415-20.

21. Crosby J, Peloso GM, Auer PL et al. Loss-of-function mutations in APOC3, triglycerides, and coronary disease. N Engl J Med 2014;371:22-31.

22. Peloso GM, Lange LA, Varga TV et al. Association of Exome Sequences With Cardiovascular Traits Among Blacks in the Jackson Heart Study. Circ Cardiovasc Genet 2016;9:368-74.

23. Stitziel NO, Stirrups KE, Masca NG et al. Coding Variation in ANGPTL4, LPL, and SVEP1 and the Risk of Coronary Disease. N Engl J Med 2016;374:1134-44.

24. McPherson R, Pertsemlidis A, Kavaslar N et al. A common allele on chromosome 9 associated with coronary heart disease. Science 2007;316:1488-91.

25. Davies RW, Wells GA, Stewart AF et al. A genome-wide association study for coronary artery disease identifies a novel susceptibility locus in the major histocompatibility complex. Circ Cardiovasc Genet 2012;5:217-25.

26. Clarke R, Peden JF, Hopewell JC et al. Genetic variants associated with Lp(a) lipoprotein level and coronary disease. N Engl J Med 2009;361:2518-28.

27. Atherosclerosis, Thrombosis, and Vascular Biology Italian Study Group. No evidence of association between prothrombotic gene polymorphisms and the development of acute myocardial infarction at a young age. Circulation 2003;107:1117-22.

28. Saleheen D, Zaidi M, Rasheed A et al. The Pakistan Risk of Myocardial Infarction Study: a resource for the study of genetic, lifestyle and other determinants of myocardial infarction in South Asia. European Journal of Epidemiology 2009;24:329-38.

12