Original Article
Rare Variant APOC3 R19X Is Associated With
Cardio-Protective Profiles in a Diverse Population-Based
Survey as Part of the Epidemiologic Architecture for
Genes Linked to Environment Study
Dana C. Crawford, PhD; Logan Dumitrescu, PhD, MS; Robert Goodloe, MS;
Kristin Brown-Gentry, MS; Jonathan Boston, BS; Bob McClellan Jr, BS; Cara Sutcliffe, MS;
Rachel Wiseman, BS; Paxton Baker, MS; Margaret A. Pericak-Vance, PhD;
William K. Scott, PhD; Melissa Allen, BS; Ping Mayo, BS; Nathalie Schnetz-Boutaud, PhD;
Holli H. Dilks, PhD; Jonathan L. Haines, PhD; Toni I. Pollin, PhD
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
Background—A founder mutation was recently discovered and described as conferring favorable lipid profiles and reduced
subclinical atherosclerotic disease in a Pennsylvania Amish population. Preliminary data have suggested that this null
mutation APOC3 R19X (rs76353203) is rare in the general population.
Methods and Results—To better describe the frequency and lipid profile in the general population, we as part of the
Population Architecture using Genomics and Epidemiology I Study and the Epidemiological Architecture for Genes
Linked to Environment Study genotyped rs76353203 in 1113 Amish participants from Ohio and Indiana and 19 613
participants from the National Health and Nutrition Examination Surveys (NHANES III, 1999 to 2002, and 2007 to
2008). We found no carriers among the Ohio and Indiana Amish. Of the 19 613 NHANES participants, we identified
31 participants carrying the 19X allele, for an overall allele frequency of 0.08%. Among fasting adults, the 19X allele
was associated with lower triglycerides (n=7603; β=−71.20; P=0.007) and higher high-density lipoprotein cholesterol
(n=8891; β=15.65; P=0.0002) and, although not significant, lower low-density lipoprotein cholesterol (n=6502; β= -4.85;
P=0.68) after adjustment for age, sex, and race/ethnicity. On average, 19X allele participants had approximately half the
triglyceride levels (geometric means, 51.3 to 69.7 versus 134.6 to 141.3 mg/dL), >20% higher high-density lipoprotein
cholesterol levels (geometric means, 56.8 to 74.4 versus 50.38 to 53.36 mg/dL), and lower low-density lipoprotein
cholesterol levels (geometric means, 104.5 to 128.6 versus 116.1 to 125.7 mg/dL) compared with noncarrier participants.
Conclusions—These data demonstrate that APOC3 19X exists in the general US population in multiple racial/ethnic groups
and is associated with cardio-protective lipid profiles. (Circ Cardiovasc Genet. 2014;7:848-853.)
Key Words: genetics ◼ genetic association studies ◼ high-density lipoprotein cholesterol
◼ molecular epidemiology ◼ triglycerides
B
oth common and rare genetic variation is associated with
lipid trait distributions. Candidate gene and genome-wide
association studies in populations of mostly European descent
have identified >150 common genetic variants associated with
high-density lipoprotein cholesterol (HDL-C), low-density
lipoprotein cholesterol (LDL-C), triglyceride, and total cholesterol levels.1,2 Early linkage and family-based studies have
identified rare mutations linked to extreme lipid trait profiles
associated with dyslipidemias.3 More recent population-based
sequencing studies are bridging the gap between common
genetic variation and disease-causing mutations with the discovery and catalog of additional rare and less common variation (frequency <1% in the general population) impacting
lipid trait profiles in humans.4–10
Clinical Perspective on p 853
One such rare variant discovered in a Pennsylvania Old
Order Amish population sample11 and recently described
in the National Heart, Lung, and Blood Institute Exome
Sequencing Project12,13 is APOC3 R19X (rs76353203). This
Received October 1, 2013; accepted September 16, 2014.
From the Institute for Computational Biology (D.C.C., P.M., J.L.H.), Department of Epidemiology and Biostatistics, Case Western Reserve University,
Cleveland, OH (D.C.C., J.L.H.); Center for Human Genetics Research (L.D., R.G., K.B.-G., J.B., B.M., M.A., N.S.-B.), Department of Molecular
Physiology and Biophysics (L.D.), Vanderbilt Technologies for Advanced Genomics Core Facility, Vanderbilt University, Nashville, TN (C.S., R.W., P.B.,
H.H.D.); Hussman Institute for Human Genomics, University of Miami, FL (M.A.P.-V., W.K.S.); and Division of Endocrinology, Diabetes, and Nutrition,
Department of Medicine, University of Maryland School of Medicine, Baltimore (T.I.P.).
The Data Supplement is available at http://circgenetics.ahajournals.org/lookup/suppl/doi:10.1161/CIRCGENETICS.113.000369/-/DC1.
Correspondence to Dana C. Crawford, PhD, Case Western Reserve University, Wolstein Research Building, 2013 Cornell Rd, Suite 2527, Cleveland,
OH 44106. E-mail [email protected]
© 2014 American Heart Association, Inc.
Circ Cardiovasc Genet is available at http://circgenetics.ahajournals.org
848
DOI: 10.1161/CIRCGENETICS.113.000369
Crawford et al APOC3 R19X, Cardio-Protective Profiles, and NHANES 849
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
null mutation was associated with cardio-protective profiles
in the Amish, including significantly higher levels of HDL-C
and lower levels of triglycerides and total cholesterol levels
compared with noncarriers.11 Pennsylvania Amish carriers of
19X were also less likely to have detectable coronary artery
calcification compared with noncarriers, which is consistent with their cardio-protective profiles.11 In contrast to the
Pennsylvania Amish population, no 19X carriers were identified in a preliminary study of 214 European American adults
from Baltimore, Maryland, suggesting that this variant is rare
in the general population.11
To date, the rare APOC3 R19X has not yet been characterized in a large and diverse general population or other Amish
populations. Therefore, to further characterize this variant in other populations, we as part of the Epidemiological
Architecture for Genes Linked to Environment Study, a
member of the Population Architecture using Genomics and
Epidemiology (PAGE) I Study,14 genotyped APOC3 R19X
in 19 613 Americans (including non-Hispanic whites, nonHispanic blacks, and Mexican Americans) ascertained for
the National Health and Nutrition Examination Surveys
(NHANES). We also genotyped 1113 Old Order Amish from
Ohio and Indiana ascertained for studies of aging and dementia. Overall, we found the 19X variant to be present but rare
in the general American population and on the same single
haplotype as in the Lancaster Amish; 19X was absent in this
sample of Old Order Amish from the Midwestern United
States. These data confirm the association between APOC3
R19X and cardio-protective lipid profiles and provide the first
glimpse of carrier rates in a general population of Americans.
Methods
Study Population
The National Health and Nutrition Examination Surveys are conducted by the National Center for Health Statistics at the Centers for
Disease Control and Prevention. The NHANES participants included
in this study were ascertained as part of NHANES III phase 2 (between 1991 and 1994), NHANES 1999 to 2000, NHANES 2001 to
2002, and NHANES 2007 to 2008. NHANES is a national representative survey of noninstitutionalized Americans ascertained regardless of health status. NHANES collects data on health and lifestyle
of participants via questionnaires, laboratory measures, and a physical examination administered by Centers for Disease Control (CDC)
health professionals.
We accessed data for serum HDL-C, triglycerides, and total cholesterol, which were all measured using standard enzymatic methods.
LDL-C was calculated using the Friedewald equation, with missing values assigned for samples with triglyceride levels >400 mg/
dL. Body mass index (kg/m2) was calculated from measured height
and weight as part of the physical examination in the CDC Mobile
Examination Center. A total of 19 613 samples were available for
study from consenting participants. All procedures were approved
by the CDC Ethics Review Board and written informed consent was
obtained from all participants. Because no identifying information
is available to the investigators, Vanderbilt University’s Institutional
Review Board determined that this study met the criteria of nonhuman subjects.
The Old Order Amish participants included in this study were originally ascertained as part of a population-based study of aging and
dementia conducted between 1998 and present in the Amish communities of Adams, Elkhart, and LaGrange Counties in Indiana and
Holmes County in Ohio. Study population characteristics of the Old
Order Amish have been previously described.15 Briefly, the Amish
immigrated to the United States from Europe in 2 waves. The first
wave arrived and settled in Pennsylvania in the early 1700s and some
of these families proceeded to migrate west to Ohio in the 18th and
19th centuries.16 A second wave of European Amish with distinct surnames arrived in Pennsylvania in the 1800s but continued westward
to Indiana and Ohio.17,18 A third wave of immigrants from Switzerland
settled in Adams County, Indiana in the mid-1800s. Written informed
consent was obtained for all Old Order Amish participants or their legal guardians. A total of 1113 unique DNA samples were available for
genotyping in this study, of which 143 were cases of dementia, 620
were controls, and 350 were of unknown dementia status. Biomarker
data were not available on the Midwestern Amish samples.
Genotyping
Genotyping was conducted by the Vanderbilt DNA Resources Core
using Applied Biosystems’ custom TaqMan assay with the following primers and probes: 5′-CCTCCTGGCGCTCCTG-3′ (forward), 5′-CCAAGTTGCCTCCACCCT-3′ (reverse), 5′-CAAGTGC
TTACGGGCAGA-3′ (G allele probe), and 5′-CAAGTGCTTACA
GGCAGA-3′ (A allele probe). To evaluate the assay and to assist
in clustering the rare variant, 94 deidentified samples consisting of
39 Pennsylvania Amish individuals heterozygous for R19X and 55
relatives without the mutation11 were genotyped blinded by mutation
status. In addition to experimental NHANES DNA, we genotyped
blinded duplicates provided by CDC for concordance checks and
quality control. The genotyping call rate for rs76353203 in NHANES
DNA samples and the Old Order Amish samples was ≈95% and 98%,
respectively.
For haplotype inference in the Pennsylvania Amish subjects, 211
individuals were genotyped for both rs7635320311 and the Illumina
Omni 2.5 mol/L Beadchip. Originally 1472 individuals were genotyped for 2 443 179 single nucleotide polymorphisms (SNPs). The 6
Beadchip SNPs used in the haplotype inference (using Haploview)19
and comparison with NHANES III were among the 2 391 559 passing quality control filters, which comprised exclusion of SNPs with
>2% duplicate inconsistency, >5% missing data, >5 Mendelian inconsistencies, Hardy–Weinberg Equilibrium P<10−6, mitochondrial
location, minor allele frequency <0.01, as well as duplicated and
nonuniquely mapped SNPs.
Statistical Analysis
Allele frequencies were calculated for 19X. Tests of association were
limited to fasting (≥8 hours since last meal) adults (≥18 years of
age), and study population characteristics are given in Table 1. We
performed an SNP test of association assuming a dominant genetic
model using linear regression. HDL-C, LDL-C, triglycerides each
Table 1. NHANES Study Population Characteristics for
Fasting Adults by Survey
NHANES III
(n=3501)
% female
NHANES 1999 to
2002 (n=4699)
NHANES 2007 to
2008 (n=1854)
56.4
51.2
49.4
Mean age (SD)
44.78 (19.24)
48.82 (18.58)
50.93 (17.78)
% non-Hispanic
white
39.5
54.1
56.3
% non-Hispanic
black
29.6
18.5
22.4
% Mexican American
30.9
27.4
21.3
Mean HDL-C (SD)
50.36 (15.23)
51.60 (15.68)
53.10 (15.78)
Mean LDL-C (SD)
124.79 (37.57)
123.70 (34.99)
115.35 (35.96)
Mean TG (SD)
134.85 (94.47)
145.57 (92.71)
135.06 (92.63)
HDL-C indicates high-density lipoprotein cholesterol; LDL-C, low-density
lipoprotein cholesterol; NHANES, National Health and Nutrition Examination
Surveys; and TG, triglyceride.
850 Circ Cardiovasc Genet December 2014
Table 2. Meta-Analysis Results for APOC3 R19X and Lipid
Traits in NHANES
Sample Size
β (SE)
P Value
HDL-C, mg/dL
8891
+15.65 (4.23)
0.0002
LDL-C, mg/dL
6502
−4.85 (11.65)
0.68
TG, mg/dL
7603
−71.20 (26.45)
0.007
HDL-C indicates high-density lipoprotein cholesterol; LDL-C, low-density
lipoprotein cholesterol; NHANES, National Health and Nutrition Examination
Surveys; and TG, triglyceride.
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
were the dependent variables as continuous traits. Models were adjusted for age, sex, and self-reported race/ethnicity. Logistic regressions were performed for each NHANES (III, 1999 to 2002, and 2007
to 2008) with SAS v9.2 (SAS Institute, Cary, NC) using the Analytic
Data Research by E-mail portal of the CDC Research Data Center in
Hyattsville, MD. Meta-analyses were conducted using a fixed-effects
inverse-variance weighted approach using METAL.20 METAL also
implements Cochran Q-test for heterogeneity. To facilitate comparisons of the genetic effect of 19X across lipid traits across studies,
we also calculated the geometric means by carrier status. Haplotypes
were inferred using PHASE v2.1.121,22 for 35 genetic variants in
the APOA5/A4C3/A1 gene cluster on chromosome 11. The 35 genetic variants included APOC3 R19X and genetic variants selected
as tagSNPs (based on data from Fullerton et al)23 and genotyped in
NHANES III for previous lipid trait genetic association studies24,25
(Table I in the Data Supplement). These Genetic NHANES data are
available for secondary analysis via the CDC.
Results
General Population
To estimate the prevalence of the APOC3 19X allele in a
general population from the United States, we genotyped
rs76353203 in a total of 19 613 DNA samples from participants ascertained as part of NHANES III, NHANES 1999
to 2002, and NHANES 2007 to 2008. We identified a total
of 31 carriers of the 19X allele among all participants genotyped. The overall frequency of the 19X allele was 0.08% in
this US population. Among the 3 major NHANES racial/ethnic groups, the 19X allele was observed at a higher frequency
among Mexican Americans (0.23% allele frequency) and nonHispanic whites (0.20% allele frequency) compared with nonHispanic blacks (0.0124% allele frequency).
We next characterized the haplotype background containing APOC3 19X in the 10 carriers identified in NHANES
III to establish the number of haplotype backgrounds associated with the mutation event. A total of 35 genetic variants
spanning the APOA5/A4/C3/A1 gene cluster on chromosome
11 were available for haplotype inference in NHANES III
only (Table I in the Data Supplement). All 10 19X alleles in
NHANES III were inferred on a single haplotype background,
suggesting that this mutation occurred once (Table II in the
Data Supplement). This haplotype background was identical
to the second most common haplotype in the population at
all loci except R19X. Using data from 211 individuals in the
Pennsylvania Amish genotyped for both APOC3 R19X and
the Illumina Omni 2.5 mol/L Beadchip, we identified 6 polymorphic markers genotyped in both the Amish and NHANES
samples. These 6 markers along with APOC3 R19X uniquely
tagged 8 of the 10 common haplotypes and collapsed 2 pairs
of haplotypes. The Amish haplotype containing the APOC3
R19X variant was identical to the NHANES III R19X haplotype, but found at a frequency of 0.028 versus 0.0007 in
NHANES III consistent with a founder effect and genetic drift
(Table III in the Data Supplement; also shows frequency of
haplotypes in the 1000 Genomes Project).
To assess the relationship between APOC3 R19X and
lipid profiles in the general population, we performed a test
of association between the rare variant and HDL-C, LDLC, and triglycerides among fasting adults (Table 1). Overall,
the 19X allele was associated with lower triglycerides
(β=−71.20±26.45 [SEM] ln mg/dL; P=0.007) and higher
HDL-C (β=+15.65±4.23 mg/dL; P=2.1×10−4) after adjustment for age, sex, and race/ethnicity (Table 2). The 19X allele
was not significantly associated with LDL-C in adjusted
analyses (β=−4.85±11.65 mg/dL; P=0.68). There was no evidence for heterogeneity for any of the 3 meta-analyzed lipids
traits (pheterogeneity=0.226, 0.794, and 0.997 for HDL-C, LDLC, and triglycerides, respectively). On average, participants
with the 19X allele had approximately half the triglyceride
levels, >20% higher HDL-C levels, and, although not statistically significant, had lower LDL-C levels compared with RR
homozygotes (Table 3). In contrast, there was no association
between the 19X allele and body mass index (β=2.16±1.82;
P=0.234), as was the case in the Pennsylvania Amish.11
Midwestern US Amish Population
The original discovery of APOC3 R19X involved Amish
participants from Pennsylvania.11 Given the migration patterns and history of the Amish in the Americans, we sought to
estimate the prevalence of 19X in a non-Pennsylvania Amish
population to determine if the cardio-protective variant is common in the Amish that settled in Ohio and Indiana. Among the
1113 Amish participants from Indiana and Ohio genotyped for
APOC3 R19X, no carriers were detected. Detailed examination of the pedigree connections between the sampled Ohio/
Table 3. Geometric Means and SD of HDL-C, LDL-C, and TG Levels Among Fasting Adults by APOC3
19X Carrier Status and NHANES
NHANES III
Noncarriers
Carriers
NHANES 1999 to 2002
Noncarriers
Carriers
NHANES 2007 to 2008
Noncarriers
Carriers
HDL-C, mg/dL (SD)
50.38 (0.27)
59.51 (7.31)
51.92 (0.25)
74.38 (5.47)
53.36 (0.37)
56.78 (14.47)
LDL-C, mg/dL (SD)
125.68 (0.80)
104.46 (36.26)
122.95 (0.74)
117.73 (13.09)
116.05 (0.98)
128.60 (36.07)
TG, mg/dL (SD)
135.67 (1.65)
51.30 (45.32)
141.33 (1.90)
63.68 (34.03)
134.62 (2.44)
69.67 (90.70)
HDL-C indicates high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; NHANES, National Health and
Nutrition Examination Surveys; and TG, triglyceride.
Crawford et al APOC3 R19X, Cardio-Protective Profiles, and NHANES 851
Indiana Amish and the Pennsylvania Amish indicate that only
2 founders of the sampled Ohio/Indiana Amish were directly
descended from the most recent common ancestor of all Pennsylvania Amish individuals possessing the R19X variant.11
This suggests that the 19X allele was either lost in this sample
because of chance nontransmission, the variant is much less
frequent compared with the Pennsylvania Amish, or the variant was not detected because of genotyping error.
Discussion
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
We estimated the prevalence of the APOC3 19X allele to be
0.08% in a general population from the United States ascertained regardless of health status, with allele frequencies
highest among Mexican Americans (0.23%) and non-Hispanic whites (0.20%) and lowest among non-Hispanic blacks
(0.0124%). We also replicated the association between APOC3
rs76353203 and cardio-protective lipid profiles first described
in a Pennsylvania Amish population. APOC3 19X carriers in
NHANES on average had significantly higher HDL-C (geometric means, 64 versus 52 mg/dL) and lower triglycerides
(geometric means, 62 versus 137 mg/dL) compared with noncarriers, similar to the effects reported in the Pennsylvania
Amish carriers of 19X (median HDL-C, 67 versus 55 mg/dL;
median triglycerides, 31 versus 57 mg/dL).11 We also observed
a trend for lower LDL-C levels among APOC3 19X carriers
versus noncarriers (117 versus 122 mg/dL), which, although
not statistically significant, was in the same direction as was
seen in the Pennsylvania Amish (116 versus 140 mg/dL).11
We did not identify any APOC3 19X carriers among 1113
Amish from Ohio and Indiana. These data coupled with
NHANES data of an overall frequency of 0.08% suggest that
APOC3 19X is a rare variant in most populations. Interestingly,
among the 31 carriers identified in NHANES, the overwhelming majority were either self-identified non-Hispanic white
or Mexican American. Indeed, frequency of 19X in non-Hispanic blacks is >15 times less than that for the other 2 groups.
Furthermore, the mutation was observed on a single haplotype
background in all NHANES III participants and the Amish
subjects. In the National Heart, Lung, and Blood Institute
Exome Sequencing Project, the 19X allele was less common
in European Americans than in NHANES (allele frequency
0.03% [3/8588 alleles] in 4294 individuals) but nonexistent
(0/4402 alleles) in blacks.12,13 Collectively, these data suggest
that the founding mutation event occurred once.
Since the writing of this article, Tachmazidou and colleagues26 published frequency estimates of APOC3 R19X in a
Greek population isolate. Among 1267 individuals, 3.8% were
identified as carriers of the 19X allele, with an overall allele
frequency of 1.9% in this isolated population.26 In contrast, the
19X allele was found only among 4 of the 3621 participants
whole genome sequenced from the UK10K project,27 for an
overall allele frequency of 0.05%. The allele frequency estimated by the UK10K project (0.05%) is similar to the overall
frequency estimated in NHANES (0.08%), reflecting the fact
that both surveys ascertained participants without regard to
disease unlike the Exome Sequencing Project, which ascertained cases and controls of various diseases and extremes of
quantitative trait distributions for ≥7 studies. In the case of
the UK10K project, the whole genome sequence data were
generated on well-phenotyped controls from the TwinsUK
and Avon Longitudinal Study of Parents and Children Study.27
Also like NHANES, Tachmazidou et al26 found the 19X allele
on a single haplotype background in both the Greek population
isolate and the UK10K data sets, suggesting a single origin of
the mutation. And, Tachmazidou et al26 replicated the association between 19X and lower triglyceride levels and higher
HDL-C levels compared with noncarriers. Together, these
data establish the association between 19X and cardioprotective lipid profiles in both isolate and outbred populations.
This study has many strengths, including sample size
(≈20 000 total DNA samples) and diversity (3 racial/ethnic
groups). Despite these strengths, this study also has several
weaknesses. First, given that the allele frequency in Africandescent populations is low, the sample size for non-Hispanic
blacks in NHANES is not sufficient for accurate estimates especially given the error rate of the assay is nearly equal to the
frequency estimate of the overall population. Second, the Ohio
and Indiana Amish population genotyped here was ascertained
for studies related to aging and dementia, and it is possible that
ascertainment bias may be an explanation for the lack of 19X
carriers in the non-Pennsylvania Amish. However, given that
most of the genotyped Amish were free of dementia and were
older than a general population, the sample was likely biased
toward more 19X carriers. These data coupled with examination
of the pedigrees suggest that the founder mutation described in
the Pennsylvania Amish was lost by chance nontransmission
in the Ohio and Indiana Amish (although differences in frequency of the mutation between Amish samples or genotyping
error cannot yet be ruled out). Third, while we replicated the
genetic association between APOC3 19X and cardio-protective
profiles in NHANES, we were limited to HDL-C, LDL-C, and
triglyceride levels. NHANES does not regularly measure other
lipid traits such as non-HDL or very low-density lipoprotein.
NHANES also does not determine coronary artery calcification
in participants. Finally, given the wide age range of NHANES
(children to older adults), NHANES has few cases of myocardial infarction or other clinical outcomes. Therefore, this study
was limited in exploring the extent of cardio-protection afforded
by APOC3 19X in NHANES.
There are other limitations worth noting that may have
affected the allele frequency estimations presented here. The
frequencies presented here were not weighted to account for
the complex survey design used to ascertain participants for
NHANES. The sample design differs by survey, and NHANES
provides sampling weights by survey (and by variable, if applicable). NHANES does not yet provide sampling weights for
analyses combining all Genetic NHANES (NHANES III, 1999
to 2002, and 2007 to 2008). It is, in fact, unclear if sampling
weights can be calculated for the Genetic NHANES data set
given the original surveys were conducted during different census years. Given the lack of sampling weights, we have analyzed
the data unweighted, and this may have slightly overestimated
the point estimate of the allele frequencies for all racial/ethnic
groups presented here. It is difficult, however, to assess the
extent of overestimation given that the genotyping performed
here may have also introduced both false positives and false
negatives. TaqMan genotyping assays are generally associated
with low error rates of 0.1%,28 with published error rates as
852 Circ Cardiovasc Genet December 2014
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
low as <1 in 2000 genotypes or 0.05%.29 On the basis of these
error rates, we would expect 10 to 20 genotyping errors among
19 613 NHANES participants genotyped for rs76353203 using
this assay. Given the PAGE I Study only supported genotyping,
we were unable to verify the 31 identified carriers of the 19X via
resequencing. Comparisons of the frequencies estimated here to
the UK10K Project described above and replication of the associations between 19X and HDL-C and triglycerides as reported
in several populations, however, suggest that the effect of the
absence of sample weights or genotyping error is minimal.
Despite these limitations, these data establish APOC3
R19X as a variant present in the general outbred population
at an appreciable, albeit low, frequency with favorable effects
on lipid profiles similar to that observed in the Pennsylvania
Amish. Recent Phase I clinical trials have shown that inhibition of apoC-III leads to reductions in plasma apoC-III and triglycerides in humans,30 both of which are associated with risk
of cardiovascular disease. Further studies are needed to establish the relevance of APOC3 in cardiovascular disease prediction, drug therapy, and other possible clinical applications.
Acknowledgments
We at Epidemiological Architecture for Genes Linked to Environment
would like to thank Dr Geraldine McQuillan and Jody McLean for
their help in accessing the Genetic NHANES data and Keith Tanner
for technical assistance in assembling the Pennsylvania Amish positive control samples. The Vanderbilt University Center for Human
Genetics Research, Computational Genomics Core provided computational or analytic support for this work. The findings and conclusions in this report are those of the authors and do not necessarily
represent the views of the National Institutes for Health or the Centers
for Disease Control and Prevention.
Sources of Funding
Genotyping in National Health and Nutrition Examination
Surveys (NHANES) was supported in part by the Epidemiological
Architecture for Genes Linked to Environment Study (U01HG004798
and its ARRA supplements) as part of the Population Architecture using Genomics and Epidemiology Study established by the National
Human Genome Research Institute. Select NHANES III data presented here were genotyped under funding provided by the University
of Washington’s Center for Ecogenetics and Environmental Health
supported by the National Institute of Environmental Sciences (5
P30 ES007033-12). Also, genotyping services for select NHANES
III single nucleotide polymorphism (SNPs) presented here were
provided by the Johns Hopkins University under federal contract
number (N01-HV-48195) from National Heart, Lung, and Blood
Institute. Genotyping and analysis in Pennsylvania Amish samples
supported by National Institutes of Health (NIH) R01 HL088119,
R01 AR046838, U01 HL72515, R01 AG18728, U01 HL084756,
R01 HL104193, and R01 CA122844; General Clinical Research
Centers Program, National Center for Research Resources, NIH; the
University of Maryland General Clinical Research Center, grant M01
RR 16500; University of Maryland Nutrition and Obesity Research
Center grant P30 DK072488. The collection and genotyping of
the Indiana/Ohio Amish samples was supported by the NIH grants
AG019085 (to Dr Haines and Dr Pericak-Vance) and AG019726 (to
Dr Scott). We thank Dr Julie Douglas for kindly sharing the Illumina
Omni 2.5 mol/L Beadchip SNP genotypes obtained in the Old Order
Amish as part of R01 CA122844.
Disclosures
None.
References
1. Global Lipids Genetics Consortium. Discovery and refinement of loci associated with lipid levels. Nat Genet. 2013;45:1274–1283.
2. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM,
Koseki M, et al. Biological, clinical and population relevance of 95 loci
for blood lipids. Nature. 2010;466:707–713.
3. Hegele RA. Plasma lipoproteins: genetic influences and clinical implications. Nat Rev Genet. 2009;10:109–121.
4. Cohen JC, Kiss RS, Pertsemlidis A, Marcel YL, McPherson R, Hobbs HH.
Multiple rare alleles contribute to low plasma levels of HDL cholesterol.
Science. 2004;305:869–872.
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. Kotowski IK, Pertsemlidis A, Luke A, Cooper RS, Vega GL, Cohen JC,
et al. A spectrum of PCSK9 alleles contributes to plasma levels of lowdensity lipoprotein cholesterol. Am J Hum Genet. 2006;78:410–422.
7. 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.
8. Rios J, Stein E, Shendure J, Hobbs HH, Cohen JC. Identification by
whole-genome resequencing of gene defect responsible for severe hypercholesterolemia. Hum Mol Genet. 2010;19:4313–4318.
9.Romeo S, Pennacchio LA, Fu Y, Boerwinkle E, Tybjaerg-Hansen A,
Hobbs HH, et al. Population-based resequencing of ANGPTL4 uncovers variations that reduce triglycerides and increase HDL. Nat Genet.
2007;39:513–516.
10. Lange L, Hu Y, Zhang H, Xue C, Schmidt E, Tang ZZ, et al. Whole-exome
sequencing identifies rare and low-frequency coding variants associated
with LDL cholesterol. Am J Hum Genet. 2014;94:233–245.
11. Pollin TI, Damcott CM, Shen H, Ott SH, Shelton J, Horenstein RB, et al.
A null mutation in human APOC3 confers a favorable plasma lipid profile
and apparent cardioprotection. Science. 2008;322:1702–1705.
12. Tennessen JA, Bigham AW, O’Connor TD, Fu W, Kenny EE, Gravel S,
et al. Evolution and functional impact of rare coding variation from deep
sequencing of human exomes. Science. 2012;337:64–69.
13. Fu W, ‘Connor TD, Jun G, Kang HM, Abecasis G, Leal SM, et al. Analysis
of 6,515 exomes reveals the recent origin of most human protein-coding
variants. Nature. 2013;493:216–220.
14. Matise TC, Ambite JL, Buyske S, Carlson CS, Cole SA, Crawford DC, et
al; PAGE Study. The next PAGE in understanding complex traits: design
for the analysis of Population Architecture Using Genetics and Epidemiology (PAGE) Study. Am J Epidemiol. 2011;174:849–859.
15. Edwards DR, Gilbert JR, Jiang L, Gallins PJ, Caywood L, Creason M,
et al. Successful aging shows linkage to chromosomes 6, 7, and 14 in the
Amish. Ann Hum Genet. 2011;75:516–528.
16. Walt J, Scott W, Slifer S, Gaskell PC, Martin E, Welsh-Bohmer K et al.
Maternal lineages and Alzheimer disease risk in the Old Order Amish.
Hum Genet. 2005;118:115–122.
17. Hostetler JA. Amish Society. Baltimore, MD: Johns Hopkins University
Press; 1993.
18. Hostetler JA. Hutterite Society. Baltimore, MD: Johns Hopkins University
Press; 1997.
19. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics. 2005;21:263–265.
20. Willer CJ, Li Y, Abecasis GR. METAL: fast and efficient meta-analysis of
genomewide association scans. Bioinformatics. 2010;26:2190–2191.
21.Stephens M, Smith NJ, Donnelly P. A new statistical method for
haplotype reconstruction from population data. Am J Hum Genet.
2001;68:978–989.
22. Stephens M, Scheet P. Accounting for decay of linkage disequilibrium
in haplotype inference and missing-data imputation. Am J Hum Genet.
2005;76:449–462.
23. Fullerton SM, Buchanan AV, Sonpar VA, Taylor SL, Smith JD, Carlson
CS, et al. The effects of scale: variation in the APOA1/C3/A4/A5 gene
cluster. Hum Genet. 2004;115:36–56.
24. Dumitrescu L, Carty CL, Taylor K, Schumacher FR, Hindorff LA, Ambite
JL, et al. Genetic determinants of lipid traits in diverse populations from
the population architecture using genomics and epidemiology (PAGE)
study. PLoS Genet. 2011;7:e1002138.
25. Keebler ME, Sanders CL, Surti A, Guiducci C, Burtt NP, Kathiresan S.
Association of blood lipids with common DNA sequence variants at 19
genetic loci in the multiethnic United States National Health and Nutrition
Crawford et al APOC3 R19X, Cardio-Protective Profiles, and NHANES 853
Examination Survey III/CLINICAL PERSPECTIVE. Circ Cardiovasc
Genet. 2009;2:238–243.
26. Tachmazidou I, Dedoussis G, Southam L, Farmaki AE, Ritchie GRS, Xifara DK, et al. A rare functional cardioprotective APOC3 variant has risen
in frequency in distinct population isolates. Nat Commun. 2013;4:2872.
27. Muddyman D, Smee C, Griffin H, Kaye J. Implementing a successful data-management framework: the UK10K managed access model. Genome
Med. 2013;5:100.
28. De La Vega FM, Isaac H, Collins A, Scafe CR, Halldorsson BV, Su X, et
al. The linkage disequilibrium maps of three human chromosomes across
four populations reflect their demographic history and a common underlying recombination pattern. Genome Res. 2005;15:454–462.
29. Ranade K, Chang MS, Ting CT, Pei D, Hsiao CF, Olivier M, et al. Highthroughput genotyping with single nucleotide polymorphisms. Genome
Res. 2001;11:1262–1268.
30. Graham MJ, Lee RG, Bell TA III, Fu W, Mullick AE, Alexander VJ, et
al. Antisense oligonucleotide inhibition of apolipoprotein C-III reduces
plasma triglycerides in rodents, nonhuman primates, and humans. Circ
Res. 2013;112:1479–1490.
CLINICAL PERSPECTIVE
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
Decades of genetic studies spanning linkage studies, candidate gene and genome-wide association studies, and contemporary sequencing studies have identified hundreds of rare and common genetic variants associated with lipid profiles within
the normal range as well as extreme lipid profiles associated with dyslipidemia. One recently identified variant in a Pennsylvania Old Order Amish population is rs76353203 (R19X), a rare null variant in apolipoprotein C III (APOC3) that was
associated with cardio-protective profiles including significantly higher levels of high density lipoprotein (HDL) cholesterol
and lower levels of triglycerides and total cholesterol levels compared with noncarriers of the mutation. Preliminary data in
European-descent outbred populations suggested that APOC3 19X is rare in the general population. In this study, we sought
to further characterize the frequency of APOC3 rs76353203 and replicate the reported associations in 19 613 participants
from 3 National Health and Nutrition Examination Surveys (NHANES III, NHANES 1999 to 2002, and NHANES 2007 to
2008). We estimated the prevalence of the APOC3 19X allele to be 0.08% in a general population from the United States
ascertained regardless of health status, with allele frequencies highest among Mexican Americans (0.23%) and non-Hispanic
whites (0.20%) and lowest among non-Hispanic blacks (0.0124%). We also replicated the association between APOC3
rs76353203 and cardio-protective lipid profiles. These data establish APOC3 19X is present in the general outbred population at an appreciable, albeit low, frequency with favorable effects on lipid profiles. Further studies are needed to establish
the relevance of APOC3 in cardiovascular disease prediction, drug therapy, and other possible clinical applications.
Downloaded from http://circgenetics.ahajournals.org/ by guest on June 17, 2017
Rare Variant APOC3 R19X Is Associated With Cardio-Protective Profiles in a Diverse
Population-Based Survey as Part of the Epidemiologic Architecture for Genes Linked to
Environment Study
Dana C. Crawford, Logan Dumitrescu, Robert Goodloe, Kristin Brown-Gentry, Jonathan
Boston, Bob McClellan, Jr, Cara Sutcliffe, Rachel Wiseman, Paxton Baker, Margaret A.
Pericak-Vance, William K. Scott, Melissa Allen, Ping Mayo, Nathalie Schnetz-Boutaud, Holli
H. Dilks, Jonathan L. Haines and Toni I. Pollin
Circ Cardiovasc Genet. 2014;7:848-853; originally published online November 1, 2014;
doi: 10.1161/CIRCGENETICS.113.000369
Circulation: Cardiovascular Genetics is published by the American Heart Association, 7272 Greenville Avenue,
Dallas, TX 75231
Copyright © 2014 American Heart Association, Inc. All rights reserved.
Print ISSN: 1942-325X. Online ISSN: 1942-3268
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circgenetics.ahajournals.org/content/7/6/848
Data Supplement (unedited) at:
http://circgenetics.ahajournals.org/content/suppl/2014/11/01/CIRCGENETICS.113.000369.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Cardiovascular Genetics can be obtained via RightsLink, a service of the Copyright Clearance
Center, not the Editorial Office. Once the online version of the published article for which permission is being
requested is located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation: Cardiovascular Genetics is online at:
http://circgenetics.ahajournals.org//subscriptions/
SUPPLEMENTAL MATERIAL
Supplementary Table 1: APOA5/A4/C3/A1 gene cluster SNPs available in NHANES III for
haplotype inference. SNP ID (if available), chromosome 11 position (Genome Build 19),
alleles, nearest gene, and SNP location within or near gene are given for each variant. The minor
allele in CEU or other European-descent populations (such as those in Fullerton et al21) is bolded
italicized. Names of SNPs used to compare haplotype backgrounds with Pennsylvania Amish
subjects are bolded.
SNP
Position
Alleles
Nearest Gene
Location
rs28927680
116619073
C/G
APOA5
3′ UTR
rs964184
116648917
C/G
APOA5
3′ near gene
rs12286037
116652207
C/T
APOA5
Intronic
rs2075294
116658122
A/C
APOA5
Intronic
rs33984246
116660450
A/G
APOA5
5′ near gene
rs34089864
116660768
A/G
APOA5
3′ UTR
rs619054
116660813
C/T
APOA5
3′ UTR
rs2072560
116661826
C/T
APOA5
Intronic
rs12287066
116662331
A/C
APOA5
Synonymous
rs17120035
116663851
C/T
APOA5
5′ near gene
rs10750097
116664040
C/T
APOA5
5′ near gene
rs12721040
116691480
A/G
APOA4
3′ UTR
rs5110
116691634
A/C
APOA4
Nonsynonymous
rs675
116691675
A/T
APOA4
Nonsynonymous
rs5106
116691928
C/T
APOA4
Synonymous
rs12721043
116692293
A/C
APOA4
Nonsynonymous
rs5104
116692334
A/G
APOA4
Nonsynonymous
rs5100
116692694
A/G
APOA4
Intronic
rs5096
116693095
C/T
APOA4
Intronic
rs5094
116693213
A/G
APOA4
Intronic
rs5092
116693464
A/G
APOA4
Synonymous
rs5091
116694005
A/G
APOA4
5′ UTR
1
rs5090
116694055
C/G
APOA4
5′ near gene
rs12721083
116699336
G/T
APOC3
5′ near gene
rs12721092
116700194
C/T
APOC3
5′ near gene
rs17257761a
116700860
A/G
APOC3
Intronic
rs76353203
116701353
A/G
APOC3
Nonsense
rs4520
116701535
C/T
APOC3
Synonymous
rs5142
116701850
C/T
APOC3
Intronic
rs34635405b
116703671
G/T
APOC3
5′ UTR
rs11216153
116705100
G/T
APOA1
3′ near gene
rs12721028
116705590
A/G
APOA1
3′ near gene
rs12718463
116706343
A/G
APOA1
3′ near gene
rs7116797
116707338
C/T
APOA1
Intronic
rs5071
a
merged with rs734104
b
merged with rs4225
116707734
A/G
APOA1
Synonymous
2
Supplementary Table 2. APOA5/A4/C3/A1 gene cluster inferred haplotypes in NHANES
III. Haplotypes were inferred in NHANES III using 35 SNPs from Supplementary Table 1 and
PHASE v2.1.119, 20. Major alleles for all SNPs are shown in the first row of the table and are
ordered 5′ to 3′ according to the genomic positions given in Supplementary Table 1. A total of
475 haplotypes were inferred in NHANES III for this gene region. Ten of the most common
inferred haplotypes in NHANES III are displayed (along with counts and percentage) followed
by the haplotype containing APOC3 19X. The 19X allele is bolded and shown in white text in a
dark box.
Inferred haplotype
#
chromosomes
(%)
N/a
Major alleles
GCCCAGCCCCTGCTCCAATAAGCGTAGCCGGAATG
Haplotype #1
GCCCAGCCCCTGCTCCAATAAGCGTAGCCTGAATG
3,558
(23.58)
Haplotype #2
GCCCAGTCCCTGCACCAGCAAGCGTAGCCTGAATG
961
(6.37)
Haplotype #3
CGTCAGCCACCGCTCCAATAAGCGTAGTCGTGATG
736
(4.88)
Haplotype #4
GGTCAGCTCCCGCTCCGGCAGGCGTGGTTGGAACG
651
(4.31)
Haplotype #5
GCCCAGCCCCTGATCCAATAAGCGTAGCCTGAATG
631
(4.18)
Haplotype #6
GCCCAGCCCCTGCTCCAATAAGCGTAGCCGGAATG
420
(2.78)
Haplotype #7
GCCCAGCCCCTGCTCCAATAAGCGTAGTCGTGATG
379
(2.51)
Haplotype #8
GCCCAGCCCTCGCTCCAATAAGCGTAGCCGGAATG
349
(2.31)
Haplotype #9
GCCCAGCCCCCGCTCCAATAAGCGTAGTCGTGATG
306
(2.03)
Haplotype #10
GCCCAGTCCCTGCTCCGGCAGGCGTAGTCGTGATG
301
(2.0)
Haplotype 19X
GCCCAGTCCCTGCACCAGCAAGCGTAACCTGAATG
10
(0.07)
3
Supplementary Table 3. Comparison of APOA5/A4/C3/A1 haplotype frequency among NHANES III, Amish, and 1000 Genomes
using polymorphic tagging SNPs genotyped in both NHANES III and Amish showing that APOC3 R19X is found on a haplotype
background otherwise identical to haplotype #2 in both NHANES III and Amish. Alleles departing from the major haplotype are
shown in dark boxes. SNPs are (1) rs619054; (2) rs12287066; (3) rs10750097; (4) rs5110; (5) rs34635405/ rs734104; (6) rs76353203
(R19X) and (7) rs34635405/ rs4225. The number of individuals is denoted by “n” for each sample.
7
% in
NHANES III
(n = 3,980)
% in
Amish
(n = 211)
%1000G
EUR
(n = 379)
1000G
AFR
(n = 246)
%1000G
MEX
(n = 66)
%1000G
COMBINED
(n = 691)
1
2
3
4
5
6
R19X
Hap #1
C
C
T
C
A
G
T
23.6
38.4
38.4
3.4
29.1
23.9
Hap #2
T
C
T
C
A
G
T
6.4
17.9
14.6
3.4
11.3
10.8
Hap #3
C
A
C
C
A
G
G
4.9
4.4
4.7
12.0
10.4
8.0
Hap #4
C
C
C
C
G
G
G
4.3
2.1
8.9
3.7
11.7
7.5
Hap #5
C
C
T
A
A
G
T
4.2
6.4
6.0
0.0
3.0
3.9
Hap #6-7
C
C
T
C
A
G
G
5.3
10.9
9.3
33.5
10.2
18.2
Hap #8-9
C
C
C
C
A
G
G
4.3
0.0
4.7
21.6
15.8
11.6
Hap #10
T
C
T
C
A
G
G
2.0
5.8
0.0
5.6
0.0
0.0
Hap 19X
T
C
T
C
A
A
T
0.07
2.8
0.0
0.0
0.0
0.0
4