GGG034702 3525..3532 INVESTIGATION Effect of Two Lipoprotein (a)-Associated Genetic Variants on Plasminogen Levels and Fibrinolysis Hong Wang,*,1 Chan E. Hong,*,1 Joshua P. Lewis,* Yanbei Zhu,* Xing Wang,*,† Xin Chu,‡ Joshua Backman,* Ziying Hu,* Peixin Yang,§ Christopher D. Still,‡ Glenn S. Gerhard,‡,** and Mao Fu*,2 *Division of Endocrinology, Diabetes, and Nutrition and §Department of Obstetrics, Gynecology, & Reproductive Sciences, University of Maryland School of Medicine, Baltimore, Maryland 21201, †Department of Orthopedic Surgery, Second Affiliated Hospital of Chongqing Medical University, 400010, China, ‡Geisinger Obesity Institute, Geisinger Clinic, Danville, Pennsylvania 17822, and **Penn State Institute for Personalized Medicine, Penn State College of Medicine, Hershey, Pennsylvania 17033 ORCID ID: 0000-0002-9904-5537 (M.F.) ABSTRACT Two genetic variants (rs3798220 and rs10455872) in the apolipoprotein (a) gene (LPA) have been implicated in cardiovascular disease (CVD), presumably through their association with lipoprotein (a) [Lp(a)] levels. While Lp(a) is recognized as a lipoprotein with atherogenic and thrombogenic characteristics, it is unclear whether or not the two Lp(a)-associated genetic variants are also associated with markers of thrombosis (i.e., plasminogen levels and fibrinolysis). In the present study, we genotyped the two genetic variants in 2919 subjects of the Old Order Amish (OOA) and recruited 146 subjects according to the carrier and noncarrier status for rs3798220 and rs10455872, and also matched for gender and age. We measured plasma Lp(a) and plasminogen levels in these subjects, and found that the concentrations of plasma Lp(a) were 2.62- and 1.73-fold higher in minor allele carriers of rs3798220 and rs10455872, respectively, com- pared with noncarriers (P = 2.04 · 10217 and P = 1.64 · 1026, respectively). By contrast, there was no difference in plasminogen concentrations between carriers and noncarriers of rs3798220 and rs10455872. Furthermore, we observed no association between carrier status of rs3798220 or rs10455872 with clot lysis time. Finally, plasminogen mRNA expression in liver samples derived from 76 Caucasian subjects was not significantly different between carriers and noncarriers of these two genetic variants. Our results provide further insight into the mechanism of action behind two genetic variants previously implicated in CVD risk and show that these polymorphisms are not major modulating factors for plasma plasminogen levels and fibrinolysis. KEYWORDS lipoprotein (a) plasminogen fibrinolysis genetics thrombogenicity Cardiovascular disease is one of the leading causes of morbidity and mortality in the world. While progression of CVD is multifactorial, substantial evidence has shown that lipoprotein (a) [Lp(a)] is a signif- icant and independent risk factor in the development of cardiovascular diseases (Berglund and Ramakrishnan 2004; Danesh et al. 2000; Kamstrup et al. 2009). Plasma Lp(a) concentration varies up to 1000- fold among individuals, is highly heritable, and is influenced minimally by environmental factors (Hobbs and White 1999; McCormick 2004). Levels of plasma Lp(a) are regulated, in part, by the LPA gene located on chromosome 6q26–27, which encodes for apo(a) (Clarke et al. 2009). The number of kringle 4 type 2 (KIV-2) repeats can vary from 12 to 51 resulting in 34 apo(a) isoforms with different sizes, and the size of apo(a) is inversely related to the plasma Lp(a) concentration (Gavish et al. 1989; Lackner et al. 1993). In addition, single nucleotide poly- morphisms (SNPs) in the LPA gene are associated with Lp(a) levels. Among them, a nonsynonymous SNP that results in an isoleucine to methionine substitution at position 1891 (rs3798220) and an intronic Copyright © 2016 Wang et al. doi: 10.1534/g3.116.034702 Manuscript received June 9, 2016; accepted for publication August 25, 2016; published Early Online September 6, 2016. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/ licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Supplemental material is available online at www.g3journal.org/lookup/suppl/ doi:10.1534/g3.116.034702/-/DC1. 1These authors contributed equally to this work. 2Corresponding author: University of Maryland School of Medicine, 660 West Redwood Street, HH492, Baltimore, MD 21201. E-mail: mfu@medicine.umaryland. edu Volume 6 | November 2016 | 3525 http://orcid.org/0000-0002-9904-5537 http://creativecommons.org/licenses/by/4.0/ http://creativecommons.org/licenses/by/4.0/ http://www.g3journal.org/lookup/suppl/doi:10.1534/g3.116.034702/-/DC1 http://www.g3journal.org/lookup/suppl/doi:10.1534/g3.116.034702/-/DC1 mailto:mfu@medicine.umaryland.edu mailto:mfu@medicine.umaryland.edu variant (rs10455872) have been confirmed to be strongly associated with increased levels of plasma Lp(a) and the risk of cardiovascular disease (Clarke et al. 2009; Li et al. 2011; Thanassoulis et al. 2013). Indeed, a recently published study by our group has provided addi- tional evidence that these variants are significantly associated with Lp(a)-cholesterol levels independently of each other and KIV-2 repeat number (Lu et al. 2015). The elucidation of a potential mechanism of action behind these Lp(a)-associated variants for CVD could lead to novel targets for treatment and/or prevention of CVD. Lp(a) is recognized as a lipoprotein with atherogenic and thrombo- genic characteristics (Caplice et al. 2001; Grainger et al. 1993; Hajjar et al. 1989; Marcovina and Koschinsky 2003). Structurally, Lp(a) is a lipoprotein particle consisting of apolipoprotein (a) [apo(a)] covalently bound to the apolipoprotein (B) (apoB) of an LDL-like particle (Koschinsky et al. 1993). Previous studies have shown that apo(a) shares structural homology with plasminogen, including a kringle 4 do- main, a kringle 5 domain, and an inactive protease domain (Hancock et al. 2003; McLean et al. 1987). Plasminogen, a critical protein in fibrinolysis, binds to lysine residues present on fibrin via its kringle domains and gets activated to plasmin by tissue plasminogen activator (tPA) or urokinase (Plow and Hoover-Plow 2004). Due to the struc- tural similarity to plasminogen and the lack of proteolytic activity, it has been suggested that Lp(a) competes with plasminogen for fibrin bind- ing, ultimately resulting in impaired fibrinolysis (Atsumi et al. 1998; Hervio et al. 1995). Furthermore, plasma plasminogen concentrations vary by about twofold among healthy individuals. The heritability of plasminogen is estimated to range from 0.48 to 0.68 (Ma et al. 2014). The plasminogen gene (PLG) and LPA gene are located on chromo- some 6q26 within �40 kb of each other (Crawford et al. 2008). A recent genome-wide association study (GWAS) has identified nine SNPs within the LPA and PLG gene region on chr6q26 to be significantly associated with plasminogen levels (Ma et al. 2014). Moreover, Lp(a) and plasminogen are primarily produced in the liver and transported into the circulation (Koschinsky et al. 1993; Rainwater and Lanford 1989; Saito et al. 1980). We speculate that LPA and PLG mRNA ex- pressions are likely to be coregulated in the liver. The genetic variants within the LPA and PLG gene region on chr6q26 may regulate Lp(a) and plasminogen levels to contribute to thrombogenicity. This pro- motes persistence of the clot and the thrombotic process that may contribute to thrombo-atherogenic diseases. LPA SNPs rs3798220 and rs10455872 are consistently associated with Lp(a) levels and result in increased risk for cardiovascular diseases (Arsenault et al. 2014; Clarke et al. 2009; Kutikhin et al. 2014; Thanassoulis et al. 2013). However, the mechanism of action behind these two genetic variants previously implicated in CVD risk is not known. In an attempt to better understand the relationship between Lp(a)-associated genetic variants and thrombogenesis we genotyped the two variants in 2919 Amish subjects and recruited 146 age- and sex-matched subjects by rs3798220 and rs10455872 genotype. We mea- sured the levels of plasma Lp(a) and plasminogen and compared the levels of Lp(a) and plasminogen based on rs3798220 and rs10455872 genotypes. In addition, we also assessed genotype-specific differences in fibrinolysis using a euglobulin clot lysis assay (ECLA). Finally, geno- type-specific differences in PLG mRNA expression were evaluated in 76 liver samples derived from Caucasian subjects. MATERIALS AND METHODS Subjects and genotyping The subjects were from the Old Order Amish community (OOA) of Lancaster, PA and were drawn from participants of our previous Lp(a) study (Lu et al. 2015), the Amish Family Diabetes Study (AFDS) (Fu et al. 2004), and Pharmacogenomics of Antiplatelet Intervention (PAPI) (Lewis et al. 2013). Details of study design, recruitment, and phenotyping have been previously described (Fu et al. 2004; Lewis et al. 2013; Lu et al. 2015). SNPs (rs3798220 and rs10455872) were geno- typed in 2919 Amish subjects using TaqMan Allelic Discrimination Assay (Applied Biosystems) and 146 subjects were recruited into the present study (Figure 1). Since carriers of minor alleles for both rs3798220 and rs10455872 are infrequent (MAF: 0.009 and 0.022, re- spectively), we recruited 31 carriers of rs3798220 and 42 carriers of rs10455872. Noncarrier control subjects were matched, as closely as possible, to the carriers according to gender and age (63 yr). To ex- clude double mutation of rs3798220 and rs10455872 in recruited sub- jects, all of the carriers and noncarriers of rs3798220 had the same genotype (AA) for rs10455872. Likewise, all of the rs10455872 individ- uals had the same genotype (TT) for rs3798220. The study was ap- proved by the institutional review board of the University of Maryland, Baltimore, and all participants provided written informed consent. The methods were carried out in accordance with the approved guidelines. Plasma samples Venousbloodwasdrawnafteranovernightfast.Fourmillilitersofblood wascollectedfromeachindividualinvacutainertubescontainingEDTA [for Lp(a) and plasminogen measurement] or 3.2% sodium citrate (for ECLA), respectively. Plasma samples were subsequently separated by centrifugation at 2000 · g for 15 min at 4�. The plasma supernatant for ECLA measurement was recentrifuged at 2000 · g for 15 min at 4� to remove any residual platelets. Multiple aliquots of plasma were stored at 280� until assays were performed. ELISA Quantitative determination of Lp(a) and plasminogen antigen levels in human plasma samples was performed by an enzyme-linked immuno- sorbent assay (ELISA) according to the manufacturer’s instruction (Assaypro, St. Charles, MO). Fifty microliters of standard or plasma using EDTA as an anticoagulant, diluted with dilution buffer (1:8000 dilution for Lp(a); 1:20,000 dilution for plasminogen), were added into each well of 96-well plates precoated with a polyclonal antibody specific for human Lp(a)/plasminogen. Wells were then incubated with the biotinylated polyclonal antibody specific for human Figure 1 A flowchart of study design. 3526 | H. Wang et al. Lp(a)/plasminogen, and was recognized by a streptavidin-peroxidase conjugate, and then a peroxidase enzyme substrate. After the reaction was stopped, the absorbance at 450 nm was read on a VICTOR X3 Multilabel Plate Reader (PerkinElmer, Waltham, MA). The standard curves were generated by polynomial regression analysis. Lp(a) and plasminogen levels were expressed in micrograms per milliliter. Euglobulin clot lysis assay ECLAwasmodifiedfromtheassaydescribedbyManco-Johnson(Smith et al. 2003). Briefly, the euglobulin fraction was prepared by adding 350 ml of plasma into 6.3 ml of 0.016% acetic acid solution, and then samples were incubated in an ice bath for 10 min. The precipitated euglobulin fraction was then resuspended with 350 ml of borate buffer (154 mM sodium chloride, 2.6 mM sodium borate, pH 9.0). One hun- dred microliters of sample was pipetted in triplicate (two wells for measurement and one well for blank) into wells of prewarmed 37� untreated 96-well plates, followed by addition of 100 ml of 0.025 mM CaCl2 to each well (100 ml of ddH2O added to the blank wells). The plate was read on a VICTOR X3 Multilabel Plate Reader (PerkinElmer) at 405 nm at 3-min intervals for 10 hr and the temperature was main- tained at 37�. Liver samples Seventy-six wedge biopsy liver samples were obtained from Caucasian patients undergoing open or laparoscopic Roux-en-Y gastric bypass operations or laparoscopic adjustable gastric banding procedures for extreme obesity or its comorbid medical problems at Geisinger Medical Center, Danville, PA (Lu et al. 2015; Still et al. 2011). Simply, we genotyped 1328 Caucasian patients for rs3798220 and rs10455872 us- ing TaqMan Allelic Discrimination Assay (Applied Biosystems). These liver samples were selected according to the carrier and noncarrier status for rs3798220 (N = 19: 19 for genotypes TT and CT) and rs10455872 (N = 20: 20: 3 for AA: AG: GG), and also matched for gender, age, and BMI (Figure 1) (Lu et al. 2015). qRT-PCR Total liverRNAwas extractedbyTRIzol(Invitrogen,GrandIsland,NY) according to the manufacturer’s instructions. The resulting RNA was subjected to DNAse digestion using RNase-free DNase Set (Qiagen, Valencia, CA) and purification using RNeasy MinElute Cleanup Kit (Qiagen). The RNA quantity and quality were determined using a ND1000 nanodrop spectrophotometer, and 1 mg of total RNA was reverse-transcribed using a Transcriptor First Strand cDNA Synthesis Kit (Roche, Indianapolis, IN). Real-time PCR was performed using TaqMan gene expression assay primers and probes [assay ID: Hs00264877_m1 for human PLG; Hs99999903_m1 for human b-actin (ACTB)]. Steady-state mRNA levels were determined by two-step quantitative real-time PCR (qRT-PCR) using the LightCycler 480 (Roche) and TaqMan probe/primer sets (Applied Biosystems, Grand Island, NY). Relative expression of mRNAs was determined after normalization with the reference gene ACTB levels using LightCycler 480 software 1.5. Statistical analysis Data were presented as mean 6 SE for the clinical profile, qRT-PCR, ELISA, and ECLA. A two-tailed Student’s t-test was used to evaluate the statistical significance for all assays except for the PLG mRNA expres- sion of rs10455872 in which ANOVA was applied (GraphPad Software, La Jolla, CA). A P value ,0.05 was considered statistically significant. Pairwise linkage disequilibrium (LD) statistics (D9 and r2) were calculated with Haploview version 4.2 (https://www.broadinstitute.org/scientific- community/science/programs/medical-and-population-genetics/haploview/ haploview) (Barrett et al. 2005). All statistical tests were two-sided. Power estimates in 42 (rs10455872) and 31 (rs3798220) pairs of subjects were calculated using the Power and Sample Size Calculation software (version 3.1.2) (Dupont and Plummer 1990). Prior data indicate that the difference in the response of matched pairs is normally distributed with stan- dard deviation. We will be able to detect a true difference in the mean response of matched pairs of 20.443 or 0.443 and 20.520 or 0.520, respectively, with probability (power) 0.8. The type I error probability associated with this test of the null hypothesis that this response difference is zero is 0.05. Data availability Both the genotypic and phenotypic information for rs3798220 and rs10455872 used in this study are listed in Supplemental Material, Table S1 and Table S2, respectively. RESULTS Clinical characteristics of the OOA subjects To evaluate whether LPA genetic variants affect plasma plasminogen levels and fibrinolysis, we successfully genotyped rs3798220 and n Table 1 Clinical profile for the OOA subjects by genotype rs3798220 rs10455872 TT (n = 31) CT (n = 31) P Value AA (n = 42) AG (n = 42) P Value Gender Male: 14 Male: 14 Male: 24 Male: 24 Female: 17 Female: 17 Female: 18 Female: 18 Age (years) 47.87 6 2.34 46.32 6 2.42 0.659 46.62 6 2.13 46.38 6 2.13 0.937 BMI (kg/m2) 29.24 6 1.04 28.41 6 1.11 0.594 28.38 6 0.92 26.70 6 0.63 0.138 SBP (mmHg) 117.07 6 2.29 119.55 6 2.93 0.511 120.19 6 1.62 118.71 6 1.63 0.523 DBP (mmHg) 75.07 6 1.59 72.74 6 1.83 0.346 76 6 1.17 74.31 6 1.30 0.337 HDL (mg/dl) 48.53 6 2.32 53.97 6 2.29 0.103 52.95 6 2.52 58.17 6 2.42 0.142 LDL (mg/dl) 127.77 6 6.86 145.35 6 7.32 0.087 130.6 6 6.69 137.82 6 6.83 0.454 TG (mg/dl) 86.17 6 8.86 83.13 6 9.72 0.820 77.93 6 6.76 78.93 6 10.09 0.935 Hct (%) 40.07 6 0.63 41.03 6 0.55 0.279 41.24 6 0.50 40.64 6 0.50 0.411 RBC count (n · 100,000) 4.52 6 0.07 4.64 6 0.07 0.278 4.56 6 0.05 4.48 6 0.06 0.377 WBC count (n ·1000) 5.42 6 0.25 6.18 6 0.31 0.066 5.29 6 0.18 5.62 6 0.22 0.252 Platelet count (n · 1000) 240.8 6 9.88 236.77 6 8.63 0.773 235.8 6 7.82 253.56 6 8.43 0.135 Data are presented as mean 6 SE. Volume 6 November 2016 | Lp(a)-Associated SNP and Thrombosis | 3527 https://www.broadinstitute.org/scientific-community/science/programs/medical-and-population-genetics/haploview/haploview https://www.broadinstitute.org/scientific-community/science/programs/medical-and-population-genetics/haploview/haploview https://www.broadinstitute.org/scientific-community/science/programs/medical-and-population-genetics/haploview/haploview http://www.g3journal.org/lookup/suppl/doi:10.1534/g3.116.034702/-/DC1/TableS1.xlsx http://www.g3journal.org/lookup/suppl/doi:10.1534/g3.116.034702/-/DC1/TableS1.xlsx http://www.g3journal.org/lookup/suppl/doi:10.1534/g3.116.034702/-/DC1/TableS2.xlsx rs10455872 in 2919 Amish subjects and identified 76 carriers of rs3798220 and 125 carriers of rs10455872. We recruited 146 age- and sex-matched subjects according to their genotype status. The mean age was 46.74 yr old, and 46.76% of subjects were female. Since carriers of minor alleles for both rs3798220 and rs10455872 were infrequent (MAF: 0.009 and 0.022, respectively) in OOA, no minor allele homozygotes were recruited. The clinical characteristics of the 146 OOA subjects are summarized in Table 1. Briefly, study subjects were relatively healthy adults, and the mean levels of age, BMI, blood pressure, HDL, LDL, TG, hematocrit, red blood cell count, white blood cell count, and platelet count were not significantly different between the carriers and noncarriers of the two LPA genetic variants rs3798220 and rs10455872. Association between LPA genetic variants and plasma Lp(a) levels To confirm whether LPA genetic variants rs3798220 and rs10455872 significantly influence plasma Lp(a) levels, we measured plasma Lp(a) levels by an ELISA (Assaypro) in our 146 Amish participants. We observed that plasma Lp(a) levels were significantly higher for the carriers of both rs3798220 and rs10455872 compared to the noncar- riers. Specifically, the Lp(a) levels in carriers of rs3798220 were 2.62 times higher than the Lp(a) levels in noncarriers (carriers: 230.96 6 10.06 mg/ml vs. noncarriers: 88.23 6 6.55 mg/ml, P = 2.04 · 10217) (Figure 2A). Similarly, the Lp(a) levels in carriers of rs10455872 were 1.73 times higher than the Lp(a) levels in noncarriers (carriers: 154.01 6 9.00 mg/ml vs. noncarriers: 88.98 6 8.76 mg/ml ; P = 1.64· 1026) (Figure 2B). Relationship between LPA genetic variants and plasma plasminogen levels To investigate whether LPA genetic variants rs3798220 and rs10455872 significantly affect plasma plasminogen levels, we measured plasma plasminogen levels by ELISA (Assaypro) in the Amish cohort. We observed that the plasminogen levels in the carriers of rs3798220 and rs10455872 were very close to the levels in the noncarriers (carriers: 217.02 6 11.34 mg/ml vs. noncarriers: 219.47 6 14.49 mg/ml, and carriers: 223.96 6 13.02 mg/ml vs. noncarriers: 232.09 6 10.99 mg/ml, respectively). The plasminogen levels in the carriers were not significantly different from the levels observed in the noncarriers of rs3798220 (P = 0.89) (Figure 3A) and rs10455872 (P = 0.63) (Figure 3B). Figure 2 Plasma Lp(a) levels among the OOA subjects for the rs3798220 genotype (A) and the rs10455872 genotype (B). A scatter- gram is shown for individuals of each genotype with the median represented by a black line respectively. Figure 3 Plasma plasminogen levels among the OOA subjects for the rs3798220 genotype (A) and the rs10455872 genotype (B). A scatter- gram is shown for individuals of each genotype with the median represented by a black line respectively. 3528 | H. Wang et al. Relationship between LPA genetic variants and fibrinolysis In addition to assessing genotype-specific differences in Lp(a) and plasminogen levels, we investigated whether LPA SNPs rs3798220 and rs10455872 were associated with fibrin clot lysis. Fibrinolysis was assessed using a euglobulin clot lysis time (ECLT) and maximum ab- sorbance (OD at 405 nm) by the ECLA. We observed no difference in ECLT between carriersand noncarriersof rs3798220 (CT vs. TT, 357.19 6 21.68 min vs. 347.26 6 15.36 min; P = 0.71, Figure 4A) or rs10455872 (AG vs. AA, 315.91 6 16.16 min vs. 334.08 6 13.67 min; P = 0.39, Figure 4B), respectively. In addition, the maximum absorbance in carriers of rs3798220 and rs10455872 were not significantly different from the maximum absorbance in the noncarriers [rs3798220 CT vs. TT, 0.84 6 0.027 vs. 0.89 6 0.023; P = 0.18 (Figure 4C) and rs10455872 AG vs. AA, 0.84 6 0.024 vs. 0.88 6 0.022; P = 0.26 (Figure 4D)]. PLG mRNA expression in the liver Given the close proximately of rs3798220 and rs10455872 with the PLG gene, we tested whether these variants had any influence on PLG mRNA expression in the liver. Total RNA was extracted from liver samples derived from Caucasian patients undergoing bariatric weight loss procedures for extreme obesity or related comorbid medical con- ditions (Lu et al. 2015). Patients were selected according to the genotype for rs3798220 (N = 19 and 19 for genotypes TT and CT, respectively) and rs10455872 (N = 20, 20, and 3 for genotypes AA, AG, and GG) and matched for age and gender. We measured the PLG mRNA expression by qRT-PCR and found that there was no statistically significant dif- ference in expression between carriers of rs3798220 and noncarriers (TT, 6.00 6 0.59; CT, 4.82 6 0.44; P = 0.12) (Figure 5A). Similarly, no difference in plasminogen RNA expression was observed between rs10455872 genotype groups (AA: 7.35 6 0.58, AG: 8.87 6 0.55, GG: 5.97 6 0.52, P = 0.06; AG + GG vs. AA, P = 0.15) (Figure 5B). Linkage disequilibrium analysis WeperformedLDanalysisforrs3798220,rs10455872,andthetopSNPs associated with plasminogen on chr.6q26 in previous GWAS (Ma et al. 2014). The result of LD analysis showed these two SNPs did not have high LD with any of the top SNPs for plasminogen (r2 from 0 to 0.227, Figure 6). DISCUSSION Lp(a) has a causal role in the development of multiple cardiovascular disorders (Arsenault et al. 2014; Clarke et al. 2009; Kamstrup et al. 2009) and is recognized as having both atherogenic and thrombogenic characteristics (Caplice et al. 2001; Enas et al. 2006; Gavish et al. 1989; Marcovina and Koschinsky 2003). LPA SNPs rs3798220 and rs10455872 are consistently associated with Lp(a) levels and result in increased risk for cardiovascular diseases (Arsenault et al. 2014; Clarke et al. 2009; Kutikhin et al. 2014; Thanassoulis et al. 2013). In the present study, we investigated the effects of these variants not only on their impact on Lp(a) levels but also on their potential effect on thrombo- genicity in order to provide novel insights regarding the mechanism(s) by which LPA variants confer CVD susceptibility. To our knowledge this study represents the first investigation to simultaneously assess the impact of these variants on both Lp(a) levels and markers of thrombosis (i.e., plasminogen levels and fibrinolysis). Consistent with previous results, we observed that both of these SNPs significantly impact Lp(a) levels; however, we observed no evidence to suggest that these variants influence plasminogen levels. In addition, we extend these finding to show that neither rs3798220 nor rs10455872 impact fibri- nolytic activity or PLG mRNA expression in the liver. Figure 4 Results of ECLA for the OOA subjects with differ- ent genotypes. Clot lysis times measured in minutes for the rs3798220 genotype (A) and for the rs10455872 genotype (B). Maximum absorbance (OD at 405 nm) measured for the rs3798220 genotype (C) and the rs10455872 genotype (D). A scattergram is shown for individ- uals of each genotype with the median represented by a black line respectively. Volume 6 November 2016 | Lp(a)-Associated SNP and Thrombosis | 3529 The OOA community is a genetically well-defined Caucasian foun- der population. More than 95% of the current Lancaster Amish pop- ulation can trace their ancestry to one of seven founder couples (Agarwala et al. 1999, 2001). The Amish today are a rural, mostly farming, community, and strong religious beliefs help them to maintain the sect as a distinct and closed entity (Cross 1976). They are relatively homogenous in terms of genetic ancestry, environment, and lifestyle characteristics, which minimizes the risk of potentially confounding variables and makes the OOA a particularly advantageous group for genetic studies. In this investigation, we have genotyped the two genetic variants of rs3798220 and rs10455872 in 2919 Amish subjects. To maximally limit confounding factors, we recruited subjects according to carrier and noncarrier status for rs3798220 and rs10455872, and matched for gender and age. The minor allele frequencies of rs3798220 and rs10455872 in the OOA are 0.9% and 2.2%, respectively (Lu et al. 2015), which is relatively lower than the minor allele frequencies (2 and 7%, respectively) in outbred European populations (Clarke et al. 2009). We did not recruit subjects who were homozygous for the minor allele of these variants in OOA. Fortunately, these two genetic variants are dom- inant genetic disorders meaning that a single allele can control whether the disease develops. We can recruit heterozygotes or/and minor allele homozygotes to study biological function. Consistent with previous stud- ies, the plasma Lp(a) levels were significantly higher for carriers of rs3798220 or rs10455872 compared to noncarriers in the OOA subjects (Clarke et al. 2009; Laschkolnig et al. 2014; Lu et al. 2015). Plasminogen is the proenzyme precursor of the primary fibrinolytic protease plasmin, which has an important role in tissue remodeling and blood clot removal after injury (Chapin and Hajjar 2015). Genetic variants in the LPA gene were identified as significant contributors to plasminogen levels by GWAS (Ma et al. 2014). Our data showed that rs3798220 and rs10455872 are not significantly associated with the plasma plasminogen levels. This finding is consistent with a previous report that rs10455872 was the most significantly associated SNP with Lp(a) levels but was not significantly associated with plasminogen levels in healthy subjects (Ma et al. 2014). Furthermore, our present study clarified for the first time that another Lp(a)-associated SNP, rs3798220, also was not significantly associated with plasminogen lev- els. Linkage disequilibrium analysis for rs3798220 and rs10455872 and the top SNPs associated with plasminogen on Chr. 6q26 in previous GWAS showed that the investigated SNPs did not have high LD with any PLG-associated SNPs. Our data suggest that the Lp(a)-associated variants rs3798220 and rs10455872 in the LPA gene are not the major genetic determinants of plasma plasminogen levels. Since apo(a) can bind to fibrin but has no proteolytic activity, it has been hypothesized that apo(a) competes with plasminogen in circula- tion and may attenuate fibrinolytic function (Angles-Cano et al. 2001; Loscalzo et al. 1990). We investigated whether genetic variation in LPA (i.e., rs3798220 and rs10455872) influences fibrinolysis and found no difference in clot lysis time and maximum absorbance between carriers and noncarriers of both rs3798220 and rs10455872, indicating that clot formation and clot lysis are not affected by these two SNPs. Of note, however, Undas et al. reported decreased clot permeability and longer clot lysis time in both healthy subjects and patients with myocardial infarction that have increased Lp(a) levels [a cutoff value of Lp(a) 300 mg/ml] as a result of small apo(a) size isoforms (Undas et al. 2006). Recently, Rowland et al. found that the LPA genetic variant rs3798220 was associated with decreased clot permeability and longer clot lysis time among Caucasians, but was associated with increased clot permeability and shorter clot lysis among non-Caucasians (Rowland et al. 2014). Mansson et al. however, reported that Lp(a) plasma levels had no effect on clot lysis time in diabetic subjects and normal controls (Mansson et al. 2014). These discordant results may be caused by: (1) differing study designs; (2) different measure Figure 5 Plasminogen mRNA expression in the liver. (A) PLG mRNA expression in the subjects with the rs3798220 genotype; (B) PLG mRNA expression in the subjects with the rs10455872 genotype. Data are presented as mean 6 SE. Figure 6 Linkage disequilibrium pattern of rs3798220, rs10455872, and top PLG association SNPs in chromosome 6q25–26. 3530 | H. Wang et al. methods; (3) small sample size; and/or (4) differences in population characteristics. In the present study, we chose relatively homogenous healthy OOA subjects that were matched with regards to sex, age, and the other lipid traits. We had a relatively large sample size and mea- sured fibrinolysis using an ECLA method without addition of human thrombin and recombinant tissue plasminogen activator (rtPA). We identified that rs3798220 and rs10455872 were not major genetic determinants for fibrinolysis. The LPA and PLG genes are adjacently located on chromosome 6 and have a high degree of structural homology, with the LPA gene believed to be generated from the duplication of the PLG gene (McLean et al. 1987). Since the liver is the major site of both LPA and PLG mRNA synthesis, it is possible that there is coregulation between he- patic LPA and PLG mRNA expression. We determined PLG mRNA levels using total RNA extracted from liver samples as described in our previous study (Lu et al. 2015). Previously, we determined that levels of LPA mRNA were higher in carriers of rs10455872 as compared to noncarriers, and were not different between the carriers and noncar- riers of rs3798220 (Lu et al. 2015). In the present study, no significant differences in PLG mRNA levels were observed between the carriers and noncarriers of rs10455872 and rs3798220. In addition, a previous animal study reported that there was no correlation between hepatic LPA and PLG mRNA levels, suggesting an independent regulation for LPA and PLG mRNA expression (Ramharack et al. 1996). There are some limitations of this study that we acknowledge. According to power estimation, we have nearly 100% power to detect alargeeffectsizeand80%powertodetectamediumeffectsize,butwedo not have power to detect small effect of rs3798220 and rs10455872 on variation of plasma plasminogen, plasminogen mRNA expression, and changeoffibrinolysis,whichmayleadtothenegativefindings.However, our data show these two variants are not the major genetic determinants of plasminogen levels and fibrinolysis although it is known that they contribute to the risk of cardiovascular diseases. In conclusion, our data indicate that two genetic variants in the LPA gene, rs3798220 and rs10455872, are significantly associated with ele- vated Lp(a) concentration, but not significantly associated with varia- tion in plasma plasminogen concentration. The increased levels of plasma Lp(a) with these genetic variants do not affect fibrinolysis in healthy subjects. Moreover, these two Lp(a)-associated genetic variants were not associated with PLG mRNA expression in the liver. These two Lp(a)-associated variants of rs3798220 and rs10455872 in the LPA gene are not the major genetic determinants of fibrinolysis, PLG mRNA expression, and plasminogen levels. ACKNOWLEDGMENTS The authors thank Alan R. Shuldiner for his contribution to Amish study and the Amish Research Clinic Staff for their great efforts in study subject recruitment and characterization as well as all the volunteers in the Amish community for their participation in these studies. This work was supported in part by an American Heart Association grant (10SDG2690004), an American Diabetes Associa- tion grant (7-12-CT-26), the National Institutes of Health (U01 HL072515, U01 HL105198, R01 AG018728, R01 DK088231), and the Mid-Atlantic Nutrition Obesity Research Center (grant P30 DK072488). Institutional review board: HP-00053256. LITERATURE CITED Agarwala, R., L. G. Biesecker, J. F. Tomlin, and A. A. Schaffer, 1999 Towards a complete North American Anabaptist genealogy: a systematic approach to merging partially overlapping genealogy resources. Am. J. Med. Genet. 86(2): 156–161. Agarwala, R., A. A. Schaffer, and J. F. Tomlin, 2001 Towards a complete North American Anabaptist Genealogy II: analysis of inbreeding. Hum. Biol. 73(4): 533–545. Angles-Cano, E., A. de la Pena Diaz, and S. Loyau, 2001 Inhibition of fibrinolysis by lipoprotein(a). Ann. N. Y. Acad. Sci. 936: 261–275. Arsenault, B. J., S. M. Boekholdt, M. P. Dube, E. Rheaume, N. J. Wareham et al., 2014 Lipoprotein(a) levels, genotype, and incident aortic valve stenosis: a prospective Mendelian randomization study and replication in a case-control cohort. Circ. Cardiovasc. Genet. 7(3): 304–310. Atsumi, T., M. A. Khamashta, C. Andujar, M. J. Leandro, O. Amengual et al., 1998 Elevated plasma lipoprotein(a) level and its association with im- paired fibrinolysis in patients with antiphospholipid syndrome. J. Rheumatol. 25(1): 69–73. Barrett, J. C., B. Fry, J. Maller, and M. J. Daly, 2005 Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21(2): 263– 265. Berglund, L., and R. Ramakrishnan, 2004 Lipoprotein(a): an elusive car- diovascular risk factor. Arterioscler. Thromb. Vasc. Biol. 24(12): 2219– 2226. Caplice, N. M., C. Panetta, T. E. Peterson, L. S. Kleppe, C. S. Mueske et al., 2001 Lipoprotein (a) binds and inactivates tissue factor pathway in- hibitor: a novel link between lipoproteins and thrombosis. Blood 98(10): 2980–2987. Chapin, J. C., and K. A. Hajjar, 2015 Fibrinolysis and the control of blood coagulation. Blood Rev. 29(1): 17–24. Clarke, R., J. F. Peden, J. C. Hopewell, T. Kyriakou, A. Goel et al., 2009 Genetic variants associated with Lp(a) lipoprotein level and coronary disease. N. Engl. J. Med. 361(26): 2518–2528. Crawford, D. C., Z. Peng, J. F. Cheng, D. Boffelli, M. Ahearn et al., 2008 LPA and PLG sequence variation and kringle IV-2 copy number in two populations. Hum. Hered. 66(4): 199–209. Cross, H. E., 1976 Population studies and the Old Order Amish. Nature 262 (5563): 17–20. Danesh, J., R. Collins, and R. Peto, 2000 Lipoprotein(a) and coronary heart disease. Meta-analysis of prospective studies. Circulation 102(10): 1082– 1085. Dupont, W. D., and W. D. Plummer, Jr., 1990 Power and sample size calculations. A review and computer program. Control. Clin. Trials 11(2): 116–128. Enas, E. A., V. Chacko, A. Senthilkumar, N. Puthumana, and V. Mohan, 2006 Elevated lipoprotein(a) – a genetic risk factor for premature vas- cular disease in people with and without standard risk factors: a review. Dis. Mon. 52(1): 5–50. Fu, M., C. M. Damcott, M. Sabra, T. I. Pollin, S. H. Ott et al., 2004 Polymorphism in the calsequestrin 1 (CASQ1) gene on chromosome 1q21 is associated with type 2 diabetes in the old order Amish. Diabetes 53(12): 3292–3299. Gavish, D., N. Azrolan, and J. L. Breslow, 1989 Plasma Ip(a) concentration is inversely correlated with the ratio of Kringle IV/Kringle V encoding domains in the apo(a) gene. J. Clin. Invest. 84(6): 2021–2027. Grainger, D. J., H. L. Kirschenlohr, J. C. Metcalfe, P. L. Weissberg, D. P. Wade et al., 1993 Proliferation of human smooth muscle cells promoted by lipoprotein(a). Science 260(5114): 1655–1658. Hajjar, K. A., D. Gavish, J. L. Breslow, and R. L. Nachman, 1989 Lipoprotein(a) modulation of endothelial cell surface fibrinolysis and its potential role in atherosclerosis. Nature 339(6222): 303–305. Hancock, M. A., M. B. Boffa, S. M. Marcovina, M. E. Nesheim, and M. L. Koschinsky, 2003 Inhibition of plasminogen activation by lipoprotein(a): critical domains in apolipoprotein(a) and mechanism of inhibition on fibrin and degraded fibrin surfaces. J. Biol. Chem. 278(26): 23260–23269. Hervio, L., V. Durlach, A. Girard-Globa, and E. Angles-Cano, 1995 Multiple binding with identical linkage: a mechanism that explains the effect of lipoprotein(a) on fibrinolysis. Biochemistry 34(41): 13353–13358. Hobbs, H. H., and A. L. White, 1999 Lipoprotein(a): intrigues and insights. Curr. Opin. Lipidol. 10(3): 225–236. Kamstrup, P. R., A. Tybjaerg-Hansen, R. Steffensen, and B. G. Nordestgaard, 2009 Genetically elevated lipoprotein(a) and increased risk of myocar- dial infarction. JAMA 301(22): 2331–2339. Volume 6 November 2016 | Lp(a)-Associated SNP and Thrombosis | 3531 Koschinsky, M. L., G. P. Cote, B. Gabel, and Y. Y. van der Hoek, 1993 Identification of the cysteine residue in apolipoprotein(a) that mediates extracellular coupling with apolipoprotein B-100. J. Biol. Chem. 268(26): 19819–19825. Kutikhin, A. G., A. E. Yuzhalin, E. B. Brusina, A. V. Ponasenko, A. S. Golovkin et al., 2014 Genetic predisposition to calcific aortic stenosis and mitral annular calcification. Mol. Biol. Rep. 41(9): 5645–5663. Lackner, C., J. C. Cohen, and H. H. Hobbs, 1993 Molecular definition of the extreme size polymorphism in apolipoprotein(a). Hum. Mol. Genet. 2(7): 933–940. Laschkolnig, A., B. Kollerits, C. Lamina, C. Meisinger, B. Rantner et al., 2014 Lipoprotein (a) concentrations, apolipoprotein (a) phenotypes, and peripheral arterial disease in three independent cohorts. Cardiovasc. Res. 103(1): 28–36. Lewis, J. P., K. Ryan, J. R. O’Connell, R. B. Horenstein, C. M. Damcott et al., 2013 Genetic variation in PEAR1 is associated with platelet aggregation and cardiovascular outcomes. Circ. Cardiovasc. Genet. 6(2): 184–192. Li, Y., M. M. Luke, D. Shiffman, and J. J. Devlin, 2011 Genetic variants in the apolipoprotein(a) gene and coronary heart disease. Circ. Cardiovasc. Genet. 4(5): 565–573. Loscalzo, J., M. Weinfeld, G. M. Fless, and A. M. Scanu, 1990 Lipoprotein(a), fibrin binding, and plasminogen activation. Arteriosclerosis 10(2): 240–245. Lu, W., Y. C. Cheng, K. Chen, H. Wang, G. S. Gerhard et al., 2015 Evidence for several independent genetic variants affecting lipoprotein (a) choles- terol levels. Hum. Mol. Genet. 24(8): 2390–2400. Ma, Q., A. B. Ozel, S. Ramdas, B. McGee, R. Khoriaty et al., 2014 Genetic variants in PLG, LPA, and SIGLEC 14 as well as smoking contribute to plasma plasminogen levels. Blood 124(20): 3155–3164. Mansson, M., I. Kalies, G. Bergstrom, C. Schmidt, A. Legnehed et al., 2014 Lp(a) is not associated with diabetes but affects fibrinolysis and clot structure ex vivo. Sci. Rep. 4: 5318. Marcovina, S. M., and M. L. Koschinsky, 2003 Evaluation of lipoprotein(a) as a prothrombotic factor: progress from bench to bedside. Curr. Opin. Lipidol. 14(4): 361–366. McCormick, S. P., 2004 Lipoprotein(a): biology and clinical importance. Clin. Biochem. Rev. 25(1): 69–80. McLean, J. W., J. E. Tomlinson, W. J. Kuang, D. L. Eaton, E. Y. Chen et al., 1987 cDNA sequence of human apolipoprotein(a) is homologous to plasminogen. Nature 330(6144): 132–137. Plow, E. F., and J. Hoover-Plow, 2004 The functions of plasminogen in cardiovascular disease. Trends Cardiovasc. Med. 14(5): 180–186. Rainwater, D. L., and R. E. Lanford, 1989 Production of lipoprotein(a) by primary baboon hepatocytes. Biochim. Biophys. Acta 1003(1): 30–35. Ramharack, R., M. A. Spahr, J. S. Kreick, and C. S. Sekerke, 1996 Expression of apolipoprotein[a] and plasminogen mRNAs in cynomolgus monkey liver and extrahepatic tissues. J. Lipid Res. 37(9): 2029–2040. Rowland, C. M., C. R. Pullinger, M. M. Luke, D. Shiffman, L. Green et al., 2014 Lipoprotein (a), LPA Ile4399Met, and fibrin clot properties. Thromb. Res. 133(5): 863–867. Saito, H., S. M. Hamilton, A. S. Tavill, L. Louis, and O. D. Ratnoff, 1980 Production and release of plasminogen by isolated perfused rat liver. Proc. Natl. Acad. Sci. USA 77(11): 6837–6840. Smith, A. A., L. J. Jacobson, B. I. Miller, W. E. Hathaway, and M. J. Manco- Johnson, 2003 A new euglobulin clot lysis assay for global fibrinolysis. Thromb. Res. 112(5–6): 329–337. Still, C. D., G. C. Wood, X. Chu, R. Erdman, C. H. Manney et al., 2011 High allelic burden of four obesity SNPs is associated with poorer weight loss outcomes following gastric bypass surgery. Obesity (Silver Spring) 19(8): 1676–1683. Thanassoulis, G., C. Y. Campbell, D. S. Owens, J. G. Smith, A. V. Smith et al., 2013 Genetic associations with valvular calcification and aortic stenosis. N. Engl. J. Med. 368(6): 503–512. Undas, A., E. Stepien, W. Tracz, and A. Szczeklik, 2006 Lipoprotein(a) as a modifier of fibrin clot permeability and susceptibility to lysis. J. Thromb. Haemost. 4(5): 973–975. Communicating editor: R. Cantor 3532 | H. Wang et al.