Introduction
CAD is one of the most prevalent complex diseases [
1]. Its pathogenesis is influenced by an interplay of genetics, diet, lifestyle, environmental and socioeconomic factors [
2]. Regional differences in CAD prevalence have been observed globally, continentally and even among adjacent countries. For example, as compared to the Netherlands and the United Kingdom (UK), Spain had a constantly lower CAD rate throughout the past 20 years [
3]. The same phenomenon can be observed among populations within the UK. In the last 15 years Scotland had constantly higher CAD prevalence compared to England, Wales, and Northern Ireland—the underlying reasons being largely unclear [
4].
In principle, both environmental factors or genetics may contribute to the local disparities in CAD [
5,
6]. Compared to the European population, Burokienė et al. found that high BMI and poor plasma lipid profiles are primarily responsible for higher cardiovascular disease (CVD) mortality in Lithuania whereas no difference was found for a genetic risk score based on 60 CVD-associated Single-nucleotide polymorphisms (SNPs) [
7]. Indeed, exogenous risk factors affected by culture, lifestyle, or socioeconomics can undergo rapid changes on the individual, familial, and population level leading to marked temporal changes in CAD prevalence [
6,
8].
Evolutionary genetics determine the allele frequency in a population, which is modulated by natural selection and stochastic forces such as genet drift [
9]. These and other factors contribute to variation among individuals in the same population and across populations [
8‐
10]. While mutations causing monogenic disorders are under evolutionary pressure, this applies, to a lesser extent to, common risk alleles with small effect sizes [
10]. Indeed, genome-wide association studies (GWAS) revealed that most common cardiometabolic conditions like hypertension, diabetes mellitus, or hyperlipidemia are affected by hundreds of risk alleles, most of which are common [
11]. The high number of susceptibility variants and their high allele frequencies jointly contribute to the genetic architecture of disease [
9,
12].
Lately, genetic risk scoring has been found to be useful in CAD risk prediction as well as therapeutic and lifestyle guidance. Using a GRS based on 27 SNPs, Mega et al. observed that individuals at high genetic risk have greater benefit from statin therapy [
13]. Moreover, Khera et al. showed that a healthy lifestyle drastically reduces risk of incident CAD events among individuals at high genetic risk [
14]. Besides for individual disease risk prediction, GRS are also used to assess and compare the risk allele burden between populations with different disease prevalence. Keaton et al. found ethnic-specific differences in the genetic architecture in the context of type 2 diabetes (T2D) between African- and European-Americans [
15], whereas Werissa et al. found no such difference between the Roma and the Hungarian general population [
16]. Pima Indians in Arizona have the highest prevalence and incidence of non-insulin-dependent diabetes of any geographically defined population [
17], but Hanson et al. found that this is not attributable to allele frequency differences at 63 diabetes loci [
18].
In this study, we explored whether the higher CAD prevalence in the Scottish population could be explained by traditional risk factors and / or common genetic variants. We used a traditional scoring model, the FRS, and a GRS model based on 163 established common risk alleles.
Discussion
The prevalence of CAD is higher in Scotland than in England for largely unexplained reasons [
4,
30]. This observation was also evident in the UK Biobank participants studied here. The traditional risk factors included in the FRS hardly explained the difference in CAD prevalence between the two countries. Out of 163 genome-wide significant risk alleles studied, 35 had higher RAF in Scotland whereas 37 had higher RAF in England. However, overall, these differences appeared to neutralize each other since there was no significant difference in the means and distributions of both weighted and unweighted GRS based on 163 CAD SNPs.
According to the ancestral-complex disease susceptibility model, genetic variations existed before the human spreading out of Africa and evolved with an extremely slow speed [
31,
32]. However, nowadays environment and lifestyle are remarkably different from that of our ancestors. A mismatch between the ancestral variants and current environment might contribute to the development of some of non-communicable, complex diseases [
2,
33].
It is unclear as to whether differences in ancestral variants contributing to CAD risk explain regional differences in CAD prevalence. With respect to England and Scotland, we observed that about 40% of genome-wide significant variants displayed significant differences in allele frequencies. It is remarkable to find that many significant differences in allele frequencies of disease relevant genes in such closely related populations. However, the balanced effect—35 variants had higher RAF in Scotland and 37 had higher RAF in England—suggests that this is not driven by any selection pressure on these risk alleles, which is in line with findings of Keyue and Iftikhar, who did not observe significant differences in the distribution of Fst values at 158 CVD-associated SNPs compared to background SNPs [
34]. In fact, the net effects of these differences at multiple loci seem to neutralize each other, since we observed no differences in the CAD risk based on polygenic risk scores.
Thus, genetic susceptibility to CAD—based to common risk alleles—appears to be rather similar in England and Scotland. The same applies to traditional risk factors for CAD, since the present as well as previous studies failed to demonstrate profound differences between these two countries [
35,
36]. In 1989, Carstairs and Morris reported that Scotland suffers from more severe deprivation than England and Wales [
37], In 2011, the same pattern of deprivation was still observed between the countries of Scotland and England [
38]. In 2013, Newton et al. reported that significant health inequalities remain between the poorest and most deprived areas [
39]. Thus, social deprivation might be one of the explanations for Scotland´s higher CAD rates. In order to lower CAD rates in Scotland, it seems to be reasonable to intensify preventive measures to be delivered at the most deprived.
A limitation of our study may be the fact that the lead SNPs we used to represent risk at a given genome-wide significant locus might not be the causal ones. However, these variants were associated with the strongest risk such that the causal variants are likely to be in very high LD. Moreover, the estimation of risk based on polygenic risk scores is unlikely to be affected by lack of knowledge on the causal variant. Another limitation of our study could be that we did not explore rare variants, gene–gene interactions, gene-environment, and exposure to epigenetic factors. All of these can modulate genetic risk [
2,
40,
41] but are challenging to investigate in a study like ours. As for the traditional factors analysis, we only included the major risk factors for CAD (sex, age, BMI, HDL-C, TC, SBP, antihypertensive medication, smoking status and diabetes), while other important factors such as physical activity, family history and socioeconomic status are not included in the Framingham risk model [
42]. Finally, the UKB population has been considered to represent a relatively low risk. As such, the data may not be representative for the entire population spectrum [
43]. Nevertheless, the repeatedly observed differences in CAD prevalence between Scotland and England were apparent in UKB as well.
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (
http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.