Abstract
Background: Previous studies have indicated that the cardioprotective effects of long-chain (LC) n–3 (ω-3) polyunsaturated fatty acids (PUFAs) may vary across various ethnic populations. Emerging evidence has suggested that the gene-environment interaction may partly explain such variations. Proprotein convertase subtilisin/kexin type 9 (PCSK9) was shown to have a mutually regulating relation with LC n–3 PUFAs and also to reduce the risk of cardiovascular diseases (CVDs). Therefore, we hypothesized that certain PCSK9 genetic variants may modify the association between LC n–3 PUFA intake and CVD risk.
Objective: We determined whether a PCSK9 variant (rs11206510), which has been identified for early onset myocardial infarction (MI), modified the association of LC n–3 PUFAs with nonfatal MI risk in Costa Rican Hispanics.
Design: We analyzed cross-sectional data from 1932 case subjects with a first nonfatal MI and 2055 population-based control subjects who were living in Costa Rica to examine potential gene-environment interactions. Two-sided P values <0.05 were considered significant.
Results: We observed a significant interaction between the PCSK9 rs11206510 genotype and LC n–3 PUFA intake on nonfatal MI risk (P-interaction = 0.012). The OR of nonfatal MI was 0.84 (95% CI: 0.72, 0.98) per 0.1% increase in total energy intake from LC n–3 PUFAs in protective-allele (C-allele) carriers, whereas the corresponding OR (95% CI) in non–C-allele carriers was 1.02 (95% CI: 0.95, 1.10). Similar results were observed when we examined the association between docosahexaenoic acid, which is one type of LC n–3 PUFA, and nonfatal MI risk (P-interaction = 0.003).
Conclusion: LC n–3 PUFA intake is associated with a lower risk of nonfatal MI in C-allele carriers of PCSK9 rs11206510 (n = 799) but not in non–C-allele carriers (n = 3188).
Keywords: Costa Rican Hispanics, gene-diet interaction, genetics, long-chain n–3, PUFAs, nonfatal myocardial infarction
See corresponding editorial on page 1029.
INTRODUCTION
Cardiovascular disease (CVD)10 is a leading cause of death globally (1). Dietary guidelines have emphasized increased intake of long-chain (LC) n–3 PUFAs as a way of reducing CVD risk (2). However, it has been noted that the cardioprotective effects of LC n–3 PUFAs vary across various ethnic populations. For example, the Multi-Ethnic Study of Atherosclerosis study showed consistent ethnic differences in the associations between circulating LC n–3 PUFA concentrations and multiple CVD outcomes (3). Emerging evidence has suggested that the gene-environment interaction may partly explain such population differences (4, 5).
Recent studies have shown a mutually regulating relation between LC n–3 PUFAs and proprotein convertase subtilisin/kexin type 9 (PCSK9). A randomized controlled trial (RCT) reported that LC n–3 PUFA intake significantly reduced plasma PCSK9 concentrations (6); evidence from a cell culture showed that PCSK9 promoted the degradation of a major receptor that is involved in LC fatty acid transportation, thereby limiting fatty acid uptake (7). PCSK9 regulates cholesterol metabolism by promoting LDL-receptor (LDLR) degradation, thereby reducing the amount of circulating LDL cholesterol, which is a key risk factor for CVD (8). A meta-analysis of 24 trials of PCSK9 monoclonal antibody therapy showed a 51% reduction in myocardial infarction (MI) risk and a 50% reduction in cardiovascular mortality (9). A recent study further linked PCSK9, LC n–3 PUFAs, and CVD together, reporting that LC n–3 PUFA intake reduced plasma PCSK9 concentrations and, thus, attenuated CVD risks (10). The common variant rs11206510 that is located in the promoter region of the PCSK9 gene has been identified for early onset MI through a genome-wide association study (11). On the basis of previous evidence, we hypothesized that the PCSK9 variant rs11206510 may interact with LC n–3 PUFAs in relation to CVD risk.
Hispanics are a genetically admixed population with diverse genetic susceptibility to chronic diseases such as CVD in their ancestral populations (4); such a unique genetic structure facilitates the detection of a gene-environment interaction (12, 13). In the current study, we examined whether the CVD-associated PCSK9 variant rs11206510 interacted with LC n–3 PUFA intake on nonfatal MI risk in a Costa Rican Hispanic population.
METHODS
Study population
Participants were from the Costa Rica Heart Study, which was a case-control study of nonfatal MI that was conducted in the Central Valley of Costa Rica from 1994 to 2004. The study has been described in details previously (14, 15). Cases were adult patients who were survivors of a first acute MI that was diagnosed at any of the 6 recruiting hospitals in the study area during the study period. Cases were ineligible if they 1) died during hospitalization, 2) were ≥75 y old on the day of their first MI, 3) were physically or mentally unable to answer the questionnaire, or 4) had a previous hospital admission that was related to CVDs. For each case, one population-based control subject was matched by age (±5 y), sex, and area of residence (county) with the use of information from the National Census and Statistics Bureau of Costa Rica. The response rate for controls was 88%. Controls were ineligible if they 1) had ever had an MI or 2) were physically or mentally unable to answer the questionnaire. The study enrolled 2273 cases and 2274 controls.
In total, 1932 incident MI cases and 2055 population-based controls with a valid case-control status, genotyping data, LC n–3 PUFA measurement, and information on matching factors were included in the current study (Supplemental Figure 1). All subjects gave informed consent on documents that were approved by the human subjects committees of both the Harvard T.H. Chan School of Public Health and the University of Costa Rica.
Assessment of nutrient and food intakes
Dietary intake was measured with the use of a 135-item semiquantitative food-frequency questionnaire (FFQ) that was specifically developed for use in the Costa Rican population and was administered by trained interviewers (16, 17). For cases, mean intakes were estimated from the year that preceded their MIs. Nutrient intakes from foods were estimated on the basis of USDA food-composition tables (18). In this study, LC n–3 PUFA intake was defined as the summation of EPA, docosapentaenoic acid (DPA), and DHA intakes from food sources. Nutrient intakes, including the aforementioned dietary fatty acids, were described as percentages of total energy (19, 20). We used raw values for food intakes.
Assessment of fatty acids in adipose tissue
Subcutaneous adipose samples were collected from the buttock with the use of a 16-gauge needle and disposable syringe after an overnight fast (21). Fatty acids from adipose tissue were quantified with the use of gas-liquid chromatography according to procedures that have been previously described (22). Fatty acids were expressed as percentages of the total fatty acids that were identified in adipose tissue. There were 2819 subjects in our study population who had valid measurements of adipose tissue LC n–3 PUFA concentrations.
Assessment of covariates
Trained personnel visited all study participants in their homes for data collection. Information on sociodemographic characteristics (age, sex, and area of residence), smoking, physical activity, and medical history was collected through an interview with the use of closed-ended questionnaires. Anthropometric measurements, including waist and hip circumferences, were collected in duplicate and the means were used for the analyses. Alcohol intake was assessed with the use of the FFQ that was previously mentioned.
Single nucleotide polymorphism selection and genotype determination
Genotyping was conducted on 4082 individuals who provided valid DNA samples (∼90% of the Costa Rica Heart Study). Blood samples were collected in the morning at the participant’s home after an overnight fast and were centrifuged to separate the plasma and leukocytes for DNA isolation via standard procedures. The PCSK9 rs11206510 polymorphism was detected with the use of a restriction fragment-length polymorphism polymerase chain reaction (11).
Statistical analyses
A chi-square test was used to assess whether the genotype frequency of rs11206510 followed Hardy-Weinberg equilibrium. Because some of the matched case-control pairs were broken as a result of missing genotype information and LC n–3 PUFA intake, ORs and Wald 95% CIs were calculated with the use of an unconditional logistic regression to determine the effect of LC n–3 PUFA intake on nonfatal MI risk. Two hierarchical models were used. Model 1 was adjusted for the matching factors age (continuous), sex (male or female), and area of residence (urban or other). Model 2 was further adjusted for the waist-to-hip ratio (continuous), daily alcohol consumption (continuous), current smoking status (yes or no), aspirin use (yes or no), physical activity (quartile of total metabolic equivalent task), family history of MI (yes or no), history of diabetes (yes or no), history of hypertension (yes or no), history of hypercholesterolemia (yes or no), and genetic admixture, which was calculated with the use of a set of 39 ancestry informative markers (proportion of European ancestral ethnicities) (23). The waist-to-hip ratio, rather than BMI, was used because of it is more appropriate for gauging obesity in the elderly (24). Because there were only a few subjects who had missing data in covariates, we used the complete case-analysis method to address missingness. Multiplicative terms of the genotype–LC n–3 PUFA interaction were included in the 2 models to test for potential gene-environment interactions. The genotype was coded as an ordinal variable in the interaction terms. We examined the relation between LC n–3 PUFA intake and nonfatal MI risk of each genotype. Results are presented with TC and CC genotypes combined because of the low frequency of minor allele C. In addition, we performed a stratified analysis by tertile categories of LC n–3 PUFA intake to investigate to what extent intake modified the relation between the rs11206510 genotype and nonfatal MI risk. The same analyses were conducted with the use of EPA, DPA, and DHA intakes as exposure variables separately. In the sensitivity analysis, we examined the effects of fish intake and adipose tissue LC n–3 PUFA concentration on nonfatal MI risk by rs11206510 genotype with adjustment for all covariates that were included in the main analysis and total energy intake (quartiles). All statistical analyses were performed with the use of SAS version 9.4 software (SAS Institute Inc.). Two-sided P values <0.05 were considered significant.
RESULTS
Demographic and risk-factor characteristics of participants on the basis of tertiles of LC n–3 PUFA intake in non-MI controls are presented in Table 1. The distribution of the PCSK9 rs11206510 genotype did not differ across tertile categories of LC n–3 PUFA intake. Participants who had higher intake of LC n–3 PUFAs were more likely to be urban residents and to drink more alcohol, and tended to have higher total energy intake than did subjects with lower intake. Characteristics of participants according to their rs11206510 genotypes are shown in Supplemental Table 1. Non–C-allele carriers tended to drink more alcohol than did C-allele carriers.
TABLE 1.
Demographic and risk factor characteristics by LC n–3 PUFA intake in controls (n = 2055)1
| Tertile of LC n–3 PUFA intake |
|||
| First (n = 685) | Second (n = 685) | Third (n = 685) | |
| Nutrient intake, % of total energy intake | |||
| LC n–3 PUFAs | 0.07 ± 0.022 | 0.12 ± 0.01 | 0.25 ± 0.13 |
| EPA | 0.02 ± 0.01 | 0.03 ± 0.01 | 0.07 ± 0.06 |
| DPA | 0.02 ± 0.01 | 0.03 ± 0.01 | 0.04 ± 0.01 |
| DHA | 0.03 ± 0.01 | 0.06 ± 0.01 | 0.14 ± 0.08 |
| Fish intake, g/d | 6.88 ± 5.90 | 14.12 ± 6.19 | 31.62 ± 19.73 |
| Adipose tissue LC n–3 PUFAs, % of total fatty acids | 0.35 ± 0.10 | 0.36 ± 0.10 | 0.39 ± 0.11 |
| Genotype, n (%) | |||
| rs11206510 TT | 549 (80.15) | 518 (75.62) | 538 (78.54) |
| rs11206510 TC | 129 (18.83) | 160 (23.36) | 132 (19.27) |
| rs11206510 CC | 7 (1.02) | 7 (1.02) | 15 (2.19) |
| Age,3 y | 59.32 ± 11.01 | 57.62 ± 11.76 | 58.01 ± 11.16 |
| Sex,3 F, n (%) | 218 (31.82) | 157 (22.92) | 170 (24.82) |
| Urban residence,3 n (%) | 247 (36.06) | 266 (38.83) | 295 (43.07) |
| Waist:hip ratio | 0.94 ± 0.07 | 0.95 ± 0.08 | 0.96 ± 0.08 |
| Alcohol consumption, g/d | 5.51 ± 15.3 | 5.90 ± 13.27 | 6.91 ± 14.76 |
| Current smoker, n (%) | 157 (23.05) | 156 (22.77) | 132 (19.41) |
| Aspirin use, n (%) | 106 (15.47) | 122 (17.81) | 118 (17.23) |
| Physical activity, METs/wk | 36.18 ± 17.36 | 36.02 ± 16.45 | 34.47 ± 14.39 |
| Family history of MI, n (%) | 47 (6.86) | 46 (6.72) | 56 (8.18) |
| History of diabetes, n (%) | 102 (14.93) | 96 (14.01) | 93 (13.58) |
| History of hypertension, n (%) | 199 (29.14) | 210 (30.7) | 205 (29.93) |
| History of hypercholesterolemia, n (%) | 181 (26.50) | 183 (26.79) | 188 (27.45) |
| Individual European admixture proportion, % | 57.36 ± 7.54 | 57.11 ± 7.84 | 58.71 ± 8.43 |
| Total energy intake, kcal/d | 2402 ± 761 | 2464 ± 704 | 2471 ± 826 |
DPA, docosapentaenoic acid; LC, long chain; MET, metabolic equivalent; MI, myocardial infarction.
Mean ± SD (all such values for continuous variables).
Matching variable.
Table 2 shows risk of nonfatal MI associated with LC n–3 PUFAs, EPA, DPA, and DHA intakes for all participants and by rs11206510 genotype. Per 0.1% increase in total energy intake from LC n–3 PUFAs, the multivariable-adjusted OR of nonfatal MI was 0.99 (95% CI: 0.92, 1.05) in the fully adjusted model (model 2). In C-allele carriers, the OR of nonfatal MI was 0.84 (95% CI: 0.72, 0.98) per 0.1% increase in total energy intake from LC n–3 PUFAs; in contrast, the association between LC n–3 PUFA intake and nonfatal MI risk was NS in non–C-allele carriers (P-interaction = 0.012). We observed similar results and a significant interaction when examining the association of DHA intake with nonfatal MI risk. ORs of nonfatal MI in C-allele carriers were 0.68 (95% CI: 0.43, 1.08) for EPA and 0.71 (95% CI: 0.55, 0.91) for DHA. The corresponding ORs (95% CIs) in subjects without the C allele were 0.91 (95% CI: 0.73, 1.12) for EPA and 1.03 (95% CI: 0.92, 1.16) for DHA (P-gene-EPA and P-gene-DHA interactions were 0.162 and 0.003, respectively). However, DPA intake was shown to be associated with an increased risk of nonfatal MI in our study population with and OR of 2.74 (95% CI: 1.65, 4.57).
TABLE 2.
ORs of myocardial infarction per 0.1% increase in total energy intake from LC n–3 PUFAs, EPA, DPA, and DHA and per 10-g increase in fish intake by rs11206510 (PCSK9) genotypes1
| Model 1 |
Model 2 |
|||
| Type of intake | OR (95% CI) | P-interaction | OR (95% CI) | P-interaction |
| LC n–3 PUFAs | 0.027 | 0.012 | ||
| Total population | 0.97 (0.91, 1.03) | 0.99 (0.92, 1.05) | ||
| *1T/*1T | 1.00 (0.94, 1.07) | 1.02 (0.95, 1.10) | ||
| *1C/*1T+*1C/*1C | 0.86 (0.74, 0.99) | 0.84 (0.72, 0.98) | ||
| EPA | 0.174 | 0.162 | ||
| Total population | 0.83 (0.69, 0.98) | 0.87 (0.72, 1.05) | ||
| *1T/*1T | 0.86 (0.71, 1.05) | 0.91 (0.73, 1.12) | ||
| *1C/*1T+*1C/*1C | 0.67 (0.44, 1.03) | 0.68 (0.43, 1.08) | ||
| DPA | 0.099 | 0.129 | ||
| Total population | 3.36 (2.09, 5.40) | 2.74 (1.65, 4.57) | ||
| *1T/*1T | 3.95 (2.33, 6.71) | 3.16 (1.79, 5.58) | ||
| *1C/*1T+*1C/*1C | 1.70 (0.57, 5.08) | 1.21 (0.37, 4.02) | ||
| DHA | 0.013 | 0.003 | ||
| Total population | 0.94 (0.85, 1.03) | 0.97 (0.87, 1.08) | ||
| *1T/*1T | 0.99 (0.89, 1.11) | 1.03 (0.92, 1.16) | ||
| *1C/*1T+*1C/*1C | 0.74 (0.59, 0.93) | 0.71 (0.55, 0.91) | ||
All intakes were energy adjusted. Model 1 (n = 3987) was adjusted for matching variables (age, sex, and area of residence). Model 2 (n = 3813) was adjusted as for model 1 and for waist-to-hip ratio, alcohol consumption, current smoking status, aspirin use, physical activity, family history of myocardial infarction, history of diabetes, history of hypertension, history of hypercholesterolemia, and genetic admixture. DPA, docosapentaenoic acid; LC, long chain; PCSK9, proprotein convertase subtilisin/kexin type 9.
Table 3 shows the association of rs11206510 genotype with nonfatal MI risk by tertiles of LC n–3 PUFAs, EPA, DPA, and DHA intakes. We showed that the association between the rs11206510 genotype and nonfatal MI risk appeared to be stronger in participants with higher LC n–3 PUFA intake. The OR of the fully adjusted model (model 2) for the increment of 1 C allele was 0.66 (95% CI: 0.51, 0.86) in subjects in the highest tertile of LC n–3 PUFA intake; in contrast, the OR of nonfatal MI per minor C allele of rs11206510 was 0.97 (95% CI: 0.74, 1.27) in subjects in the lowest tertile. In subjects in the highest tertile of intake, ORs of nonfatal MI per minor C allele of rs11206510 were 0.60 (95% CI: 0.46, 0.78) for EPA, 0.65 (95% CI: 0.51, 0.84) for DPA, and 0.61 (95% CI: 0.47, 0.81) for DHA. The corresponding ORs in subjects in the lowest tertile were 0.82 (95% CI: 0.62, 1.07) for EPA, 0.88 (95% CI: 0.67, 1.16) for DPA, and 0.87 (95% CI: 0.67, 1.13) for DHA.
TABLE 3.
ORs of myocardial infarction per minor allele of rs11206510 (PCSK9) by LC n–3 PUFAs, EPA, DPA, DHA, and fish intakes1
| Model 1 |
Model 2 |
||||
| Type and tertile of intake | Mean ± SD2 | OR (95% CI) | P-interaction | OR (95% CI) | P-interaction |
| LC n–3 PUFA, % of total energy intake | 0.027 | 0.012 | |||
| First | 0.07 ± 0.02 | 1.05 (0.82, 1.35) | 0.97 (0.74, 1.27) | ||
| Second | 0.12 ± 0.01 | 0.71 (0.56, 0.91) | 0.75 (0.58, 0.98) | ||
| Third | 0.24 ± 0.12 | 0.71 (0.56, 0.91) | 0.66 (0.51, 0.86) | ||
| EPA, % of total energy intake | 0.174 | 0.162 | |||
| First | 0.02 ± 0.004 | 0.86 (0.67, 1.11) | 0.82 (0.62, 1.07) | ||
| Second | 0.03 ± 0.004 | 0.95 (0.75, 1.21) | 0.97 (0.74, 1.25) | ||
| Third | 0.07 ± 0.05 | 0.64 (0.50, 0.82) | 0.60 (0.46, 0.78) | ||
| DPA, % of total energy intake | 0.099 | 0.129 | |||
| First | 0.01 ± 0.004 | 0.93 (0.71, 1.20) | 0.88 (0.67, 1.16) | ||
| Second | 0.03 ± 0.003 | 0.86 (0.67, 1.11) | 0.90 (0.68, 1.19) | ||
| Third | 0.04 ± 0.01 | 0.69 (0.54, 0.86) | 0.65 (0.51, 0.84) | ||
| DHA, % of total energy intake | 0.013 | 0.003 | |||
| First | 0.03 ± 0.01 | 0.91 (0.71, 1.15) | 0.87 (0.67, 1.13) | ||
| Second | 0.06 ± 0.01 | 0.85 (0.66, 1.09) | 0.87 (0.67, 1.13) | ||
| Third | 0.14 ± 0.08 | 0.68 (0.53, 0.88) | 0.61 (0.47, 0.81) | ||
Model 1 (n = 3987) was adjusted for matching variables (age, sex, and area of residence). Model 2 (n = 3813) was adjusted as for model 1 and for waist-to-hip ratio, alcohol consumption, current smoking status, aspirin use, physical activity, family history of myocardial infarction, history of diabetes, history of hypertension, history of hypercholesterolemia, and genetic admixture. DPA, docosapentaenoic acid; LC, long chain; PCSK9, proprotein convertase subtilisin/kexin type 9.
For continuous variables.
Supplemental Figures 2 and 3 show risk of nonfatal MI associated with fish intake and adipose tissue LC n–3 PUFAs by rs11206510 genotype. Per 10-g increase in fish intake, the multivariable-adjusted OR of nonfatal MI was 0.94 (95% CI: 0.85, 1.04) in C-allele carriers and was 1.01 (95% CI: 0.97, 1.06) in non–C-allele carriers. Per 1% increase in total fatty acids from LC n–3 PUFAs in adipose tissue, the multivariable-adjusted OR was 0.57 (95% CI: 0.10, 3.28) in C-allele carriers and was 0.94 (95% CI: 0.40, 2.22) in other subjects.
DISCUSSION
In this Costa Rican Hispanic population, we showed a significant interaction between LC n–3 PUFA intake and PCSK9 rs11206510 genotype on nonfatal MI risk. LC n–3 PUFA intake was inversely associated with nonfatal MI risk in C-allele carriers of PCSK9 rs11206510 but not in non–C-allele carriers. Similar results were observed when examining the effects of DHA. In addition, we observed a stronger association between single nucleotide polymorphism rs11206510 and nonfatal MI risk in participants with higher LC n–3 PUFA intake.
Compared with the extensive literature on fatal MI, there have been far fewer epidemiologic studies regarding the association between LC n–3 PUFAs (25–28) and nonfatal MI risk. In most of the studies, higher LC n–3 PUFAs (25–28) were not associated with lower risk of nonfatal MI, and our results of the total population were consistent with the previous observations. However, when stratified by PCSK9 rs11206510 genotype, we observed a significant inverse association between LC n–3 PUFA intake and nonfatal MI risk in C-allele carriers but not in non–C-allele carriers. We also showed that the protective effect of the C allele on nonfatal MI risk was strongest in subjects in the highest tertile of LC n–3 PUFA intake.
PCSK9 is a secreted protein that binds to cell-surface LDLR and directs it toward lysosomal degradation, thereby elevating the plasma LDL-cholesterol concentration and CVD risk (29). Genetic-association studies have consistently shown that PCSK9 loss-of-function variants were associated with lower CVD risk through a lifelong reduction in the plasma LDL cholesterol concentration (30). In addition to LDL cholesterol, circulating PCSK9 has been reported to play a role in the modulation of total cholesterol, VLDL, plasma triglycerides (31), HDL cholesterol, apolipoproteins, and insulin (32).
Recent studies have indicated a mutually regulating relation between PCKS9 and LC n–3 PUFAs, thereby further bridging the functional relation between them. Evidence from a cell culture showed that, besides enhancing the degradation of LDLR, PCSK9 also promoted the degradation of CD36, which is a major receptor that is involved in LC fatty acid transportation, thereby limiting fatty acid uptake (7); one RCT showed that the consumption of 2.2 g marine n–3 PUFAs/d for 12 wk significantly reduced plasma PCSK9 concentrations in women (6); another RCT showed that dietary DHA intake attenuated CVD risk via the lowering of the plasma PCSK9 concentration (10). Taken together, the results lend support to the biological plausibility of interactions between n–3 PUFA and the PCSK9 rs11206510 genotype.
We examined gene-diet interactions for EPA, DPA, and DHA intakes separately. DHA intake was the only intake that showed similar results and a significant interaction with the PCSK9 rs11206510 genotype as LC n–3 PUFA intake did, which suggested the gene–LC n–3 PUFAs interaction was mainly driven by DHA intake. The different results between DHA intake and the other 2 types of LC n–3 PUFAs may partly have been because of their different metabolic pathways in the human body. Tissue and circulating DHA concentrations nearly entirely depend on direct dietary consumption and correlate well with DHA intake (33). In contrast, EPA is principally derived from direct consumption but also can be synthesized from fatty acid precursors in small amounts (33), and DPA concentrations in the human body are predominantly determined by endogenous metabolism and are not correlated with the dietary consumption of LC n–3 PUFAs (33, 34). We examined the effect of adipose tissue LC n–3 PUFAs on nonfatal MI risk. However, because the measurement of adipose tissue LC n–3 PUFAs was only available in a subset of our study population, we did not have enough power to detect a significant interaction between it and the genotype.
Our study has several strengths. We used an FFQ that was designed and validated specifically for Hispanics who were living in Costa Rica, which likely captured accurate intakes of nutrients and foods (16, 17). Our study also benefits from a large sample size of Costa Rican Hispanics and detailed information of the participants. The main limitation is the cross-sectional nature of our analysis, and we could not exclude the possibility of reverse causations. However, because the causal effect of LC n–3 PUFAs on nonfatal MI has been well established in previous RCTs, and the reverse causality between genetic variants and nonfatal MI risk was biologically impossible, it was less of a concern. Other limitations include the absence of information on hypocholesterolemic drug use, not having enough statistical power to detect a significant interaction between adipose tissue LC n–3 PUFAs and the genotype, and the lack of generalizability because we included only Costa Rican Hispanics. In addition, because cases in our study population were defined as individuals who experienced a nonfatal MI, we could not exclude the possibility that the observed interaction may affect the survival after an acute nonfatal MI.
In conclusion, there is a significant interaction between LC n–3 PUFA intake and the PCSK9 rs11206510 genotype on nonfatal MI risk in Costa Rican Hispanics. LC n–3 PUFA intake is associated with a lower risk of nonfatal MI in C-allele carriers of PCSK9 rs11206510 but not in non–C-allele carriers.
Acknowledgments
We thank Francisco Cai from the Harvard T.H. Chan School of Public Health for his support.
The authors’ responsibilities were as follows—ZY: analyzed the data; ZY, TH, and LQ: designed the study and wrote the first draft of the manuscript; ZY and LQ: were guarantors of the study; YZ, TW, YH, DS, HC, and LQ: were involved in the data collection; HC: provided statistical expertise; and all authors: contributed to the interpretation of the results and to the critical revision of the manuscript for important intellectual content and read and approved the final version of the manuscript. None of the authors reported a conflict of interest related to the study.
Footnotes
Abbreviations used: CVD, cardiovascular disease; DPA, docosapentaenoic acid; FFQ, food-frequency questionnaire; LC, long chain; LDLR, LDL receptor; MI, myocardial infarction; PCSK9, proprotein convertase subtilisin/kexin type 9; RCT, randomized controlled trial.
REFERENCES
- 1.Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, et al. Executive summary: heart disease and stroke statistics–2013 update: a report from the American Heart Association. Circulation 2013;127:143–52. [DOI] [PubMed] [Google Scholar]
- 2.McGuire S. US Department of Agriculture and U.S. Department of Health and Human Services, Dietary Guidelines for Americans, 2010. 7th Edition. Washington (DC): US Government Printing Office, January 2011. Adv Nutr 2011;2:293–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.de Oliveira Otto MC, Wu JH, Baylin A, Vaidya D, Rich SS, Tsai MY, Jacobs DR Jr, Mozaffarian D. Circulating and dietary omega-3 and omega-6 polyunsaturated fatty acids and incidence of CVD in the Multi-Ethnic Study of Atherosclerosis. J Am Heart Assoc 2013;2:e000506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Qi L, Parast L, Cai T, Powers C, Gervino EV, Hauser TH, Hu FB, Doria A. Genetic susceptibility to coronary heart disease in type 2 diabetes: 3 independent studies. J Am Coll Cardiol 2011;58:2675–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Qi L. Gene-environment interaction and human diseases. Hauppauge (NY): Nova; Science Publishers Inc.; 2015. [Google Scholar]
- 6.Graversen CB, Lundbye-Christensen S, Thomsen B, Christensen JH, Schmidt EB. Marine n-3 polyunsaturated fatty acids lower plasma proprotein convertase subtilisin kexin type 9 levels in pre- and postmenopausal women: a randomised study. Vascul Pharmacol 2016;76:37–41. [DOI] [PubMed] [Google Scholar]
- 7.Demers A, Samami S, Lauzier B, Des Rosiers C, Sock ET, Ong H, Mayer G. PCSK9 induces CD36 degradation and affects long-chain fatty acid uptake and triglyceride metabolism in adipocytes and in mouse liver. Arterioscler Thromb Vasc Biol 2015;35:2517–25. [DOI] [PubMed] [Google Scholar]
- 8.Li S, Li JJ. PCSK9: a key factor modulating atherosclerosis. J Atheroscler Thromb 2015;22:221–30. [DOI] [PubMed] [Google Scholar]
- 9.Navarese EP, Kolodziejczak M, Schulze V, Gurbel PA, Tantry U, Lin Y, Brockmeyer M, Kandzari DE, Kubica JM, D’Agostino RB Sr. , et al. Effects of proprotein convertase subtilisin/kexin type 9 antibodies in adults with hypercholesterolemia: a systematic review and meta-analysis. Ann Intern Med 2015;163:40–51. [DOI] [PubMed] [Google Scholar]
- 10.Rodríguez-Pérez C, Ramprasath VR, Pu S, Sabra A, Quirantes-Pine R, Segura-Carretero A, Jones PJ. Docosahexaenoic acid attenuates cardiovascular risk factors via a decline in proprotein convertase subtilisin/kexin type 9 (PCSK9) plasma levels. Lipids 2016;51:75–83. [DOI] [PubMed] [Google Scholar]
- 11.Kathiresan S, Voight BF, Purcell S, Musunuru K, Ardissino D, Mannucci PM, Anand S, Engert JC, Samani NJ, Schunkert H, et al. Genome-wide association of early-onset myocardial infarction with single nucleotide polymorphisms and copy number variants. Nat Genet 2009;41:334–41. Erratum in: Nat Genet 2009;41:762. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Qi L, Ma J, Qi Q, Hartiala J, Allayee H, Campos H. Genetic risk score and risk of myocardial infarction in Hispanics. Circulation 2011;123:374–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zheng Y, Li Y, Huang T, Cheng HL, Campos H, Qi L. Sugar-sweetened beverage intake, chromosome 9p21 variants, and risk of myocardial infarction in Hispanics. Am J Clin Nutr 2016;103:1179–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Baylin A, Kabagambe EK, Ascherio A, Spiegelman D, Campos H. Adipose tissue alpha-linolenic acid and nonfatal acute myocardial infarction in Costa Rica. Circulation 2003;107:1586–91. [DOI] [PubMed] [Google Scholar]
- 15.Campos H, Siles X. Siesta and the risk of coronary heart disease: results from a population-based, case-control study in Costa Rica. Int J Epidemiol 2000;29:429–37. [PubMed] [Google Scholar]
- 16.El-Sohemy A, Baylin A, Ascherio A, Kabagambe E, Spiegelman D, Campos H. Population-based study of alpha- and gamma-tocopherol in plasma and adipose tissue as biomarkers of intake in Costa Rican adults. Am J Clin Nutr 2001;74:356–63. [DOI] [PubMed] [Google Scholar]
- 17.El-Sohemy A, Baylin A, Kabagambe E, Ascherio A, Spiegelman D, Campos H. Individual carotenoid concentrations in adipose tissue and plasma as biomarkers of dietary intake. Am J Clin Nutr 2002;76:172–9. [DOI] [PubMed] [Google Scholar]
- 18.Baylin A, Kabagambe EK, Ascherio A, Spiegelman D, Campos H. High 18:2 trans-fatty acids in adipose tissue are associated with increased risk of nonfatal acute myocardial infarction in Costa Rican adults. J Nutr 2003;133:1186–91. [DOI] [PubMed] [Google Scholar]
- 19.Kabagambe EK, Baylin A, Allan DA, Siles X, Spiegelman D, Campos H. Application of the method of triads to evaluate the performance of food frequency questionnaires and biomarkers as indicators of long-term dietary intake. Am J Epidemiol 2001;154:1126–35. [DOI] [PubMed] [Google Scholar]
- 20.Willett W. Nutritional epidemiology. New York: Oxford University Press; 1998. [Google Scholar]
- 21.Beynen AC, Katan MB. Rapid sampling and long-term storage of subcutaneous adipose-tissue biopsies for determination of fatty acid composition. Am J Clin Nutr 1985;42:317–22. [DOI] [PubMed] [Google Scholar]
- 22.Baylin A, Kabagambe EK, Siles X, Campos H. Adipose tissue biomarkers of fatty acid intake. Am J Clin Nutr 2002;76:750–7. [DOI] [PubMed] [Google Scholar]
- 23.Ruiz-Narváez EA, Bare L, Arellano A, Catanese J, Campos H. West African and Amerindian ancestry and risk of myocardial infarction and metabolic syndrome in the Central Valley population of Costa Rica. Hum Genet 2010;127:629–38. [DOI] [PubMed] [Google Scholar]
- 24.Srikanthan P, Seeman TE, Karlamangla AS. Waist-hip-ratio as a predictor of all-cause mortality in high-functioning older adults. Ann Epidemiol 2009;19:724–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hu FB, Bronner L, Willett WC, Stampfer MJ, Rexrode KM, Albert CM, Hunter D, Manson JE. Fish and omega-3 fatty acid intake and risk of coronary heart disease in women. JAMA 2002;287:1815–21. [DOI] [PubMed] [Google Scholar]
- 26.Mozaffarian D, Ascherio A, Hu FB, Stampfer MJ, Willett WC, Siscovick DS, Rimm EB. Interplay between different polyunsaturated fatty acids and risk of coronary heart disease in men. Circulation 2005;111:157–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Dietary supplementation with n-3 polyunsaturated fatty acids and vitamin E after myocardial infarction: results of the GISSI-Prevenzione trial. Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico. Lancet 1999;354:447–55. [PubMed] [Google Scholar]
- 28.Lemaitre RN, King IB, Mozaffarian D, Kuller LH, Tracy RP, Siscovick DS. n –3 Polyunsaturated fatty acids, fatal ischemic heart disease, and nonfatal myocardial infarction in older adults: the Cardiovascular Health Study. Am J Clin Nutr 2003;77:319–25. [DOI] [PubMed] [Google Scholar]
- 29.Giugliano RP, Sabatine MS. Are PCSK9 inhibitors the next breakthrough in the cardiovascular field? J Am Coll Cardiol 2015;65:2638–51. [DOI] [PubMed] [Google Scholar]
- 30.Lambert G, Sjouke B, Choque B, Kastelein JJ, Hovingh GK. The PCSK9 decade. J Lipid Res 2012;53:2515–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Druce I, Abujrad H, Ooi TC. PCSK9 and triglyceride-rich lipoprotein metabolism. J Biomed Res 2015;354:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Ferri N, Corsini A, Macchi C, Magni P, Ruscica M. Proprotein convertase subtilisin kexin type 9 and high-density lipoprotein metabolism: experimental animal models and clinical evidence. Transl Res 2016;173:19–29. [DOI] [PubMed] [Google Scholar]
- 33.Mozaffarian D, Wu JH. (n –3) fatty acids and cardiovascular health: are effects of EPA and DHA shared or complementary? J Nutr 2012;142:614S–25S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sun Q, Ma J, Campos H, Rexrode KM, Albert CM, Mozaffarian D, Hu FB. Blood concentrations of individual long-chain n –3 fatty acids and risk of nonfatal myocardial infarction. Am J Clin Nutr 2008;88:216–23. [DOI] [PMC free article] [PubMed] [Google Scholar]