Skip to main content
NCBI home page
Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease logoLink to Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease
. 2024 Jan 12;13(2):e030726. doi: 10.1161/JAHA.123.030726

Genetic Evidence for the Causal Association of Circulating Cytokines and Growth Factors With Coronary Artery Disease

Yuwei Li 1, Bin Liu 1, Ying Chen 1, Ziying Liu 1, Ding Ye 1, Yingying Mao 1,✉, Xiaohui Sun 1,✉
PMCID: PMC10926782  PMID: 38214249

Abstract

Background

Observational studies have suggested the potential role of inflammatory factors in the risk of coronary artery disease (CAD). We aimed to perform 2‐sample Mendelian randomization (MR) analyses to assess the causal association between circulating cytokines/growth factors and CAD.

Methods and Results

The instrumental variables for 28 circulating cytokines and growth factors were identified from a genome‐wide association study of 8293 European participants. Summary‐level data on CAD were derived from a large genome‐wide association study (71 602 cases and 260 875 controls). We used the inverse‐variance‐weighted and Wald ratio methods as our main MR methods. The weighted median, simple median, maximum likelihood, MR pleiotropy residual sum and outlier, and MR‐Egger methods were performed as sensitivity analyses. Genetic colocalization analyses were conducted to validate the robustness of our MR findings. We found that genetically predicted circulating levels of macrophage migration inhibitory factor were associated with an increased risk of CAD at the Bonferroni‐adjusted level of significance (P<1.79×10−3). The odds ratio was 1.20 (95% CI, 1.08–1.33; P=6.83×10−4) per 1‐SD increase in macrophage migration inhibitory factor. Colocalization analyses supported our MR findings. Additionally, we found suggestive evidence between the genetic effects of stem cell growth factor‐β and the risk of CAD (odds ratio, 0.95 [95% CI, 0.91–0.98]; P=0.007).

Conclusions

Our findings suggested a risk‐increasing effect of macrophage migration inhibitory factor level on the development of CAD. The roles of these inflammatory factors for CAD warrant further investigation.

Keywords: coronary artery disease, cytokine, inflammation, Mendelian randomization analysis

Subject Categories: Genetic, Association Studies

Nonstandard Abbreviations and Acronyms

eQTL

expression quantitative trait locus

GWAS

genome‐wide association study

IV

instrumental variable

MIF

macrophage migration inhibitory factor (glycosylation‐inhibiting factor)

MR

Mendelian randomization

pQTL

protein quantitative trait locus

SCGF‐β

stem cell growth factor beta

SNP

single nucleotide polymorphism

Clinical Perspective.

What Is New?

  • The potentially causal associations of 28 circulating cytokines/growth factors and the risk of coronary artery disease were systematically evaluated under the framework of Mendelian randomization methods.

  • The genetically predicted macrophage migration inhibitory factor levels were found to be positively associated with the risk of coronary artery disease.

What Are the Clinical Implications?

  • Our results may provide new potential therapeutic targets for coronary artery disease.

Cardiovascular disease remains the leading cause of death worldwide. Coronary artery disease (CAD) is the most common cardiovascular disease, principally driven by ischemic heart disease, which was responsible for ≈15.96% of total deaths in 2017. 1 Though several risk factors, such as family history, obesity, and unhealthy lifestyles have been identified, the pathogenesis and pathophysiology of CAD remain unclear.

Emerging evidence has demonstrated the vital role of inflammation in the pathogenesis of atherosclerosis and CAD. 2 For example, previous studies reported that elevated levels of high‐sensitivity C‐reactive protein may be a risk factor for the development of CAD. 3 , 4 In parallel, several other inflammatory factors were also identified to connect with an altered risk for CAD. Specifically, a case–control study containing 498 cases and 499 controls showed the circulating level of monocyte chemoattractant protein‐1 was significantly associated with an increased risk of CAD (odds ratio [OR], 1.33 [95% CI, 1.19–1.49]). 5 Moreover, a meta‐analysis including 17 prospective studies indicated a positive association between serum interleukin‐6 and the development of CAD. 6 Similarly, the effects of macrophage inflammatory protein‐1β, macrophage migration inhibitory factor (MIF), and monokine induced by γ‐interferon on CAD were also identified by previous epidemiological research. 7 , 8 , 9 However, observational studies may be confounded by potential biases and reverse causation. The relationships between inflammatory factors and CAD observed in the previous observational studies require further investigation.

Mendelian randomization (MR) analysis is a genetic epidemiological method to evaluate the potential causality of risk factors on an outcome by using genetic variants as instrumental variables (IVs). 10 Because parental alleles are randomly assorted during conception, MR is less likely to be affected with potential confounders. 11 Furthermore, reverse causality can be diminished, as genotype is not affected by the progression of diseases. 11 Benefitting from the aforementioned advantages, this method has been widely used to assess the potential causality of risk factors upon diseases. 12 A previous MR study has evaluated the relationships between cytokines and CAD and suggested that interleukin‐1 receptor antagonist and macrophage colony‐stimulating factor may increase the risk of CAD. 13 With the rapid expansion in sample sizes of genome‐wide association studies (GWASs), a recent larger CAD GWAS was released from the Coronary Artery Disease Genome‐Wide Replication and Meta‐analysis Plus the Coronary Artery Disease Genetics Consortium for the European population. 14 Using summary genetic data from a larger GWAS could improve the accuracy and precision of MR estimates. Therefore, we conducted a 2‐sample MR analysis to reveal the potential causal associations of cytokines and growth factors with the development of CAD using the latest and largest GWAS.

Methods

All data and materials are publicly available in the relevant articles, and the corresponding links can be found in Table S1.

Study Design and Data Source

MR analyses are conducted under 3 IV assumptions (Figure 1). First, the genetic variants used as IVs should be significantly associated with the exposure. Second, IVs should not be associated with any other confounders. Third, IVs affect the outcome only through the selected exposure, not via an alternative way. 11

Figure 1. Overview of present MR study.

Figure 1

Open in a new tab

MR indicates Mendelian randomization.

We designed a 2‐sample MR approach to evaluate the potentially causal effects of circulating cytokines and growth factors on CAD (Figure 1). 14 , 15 Genetic variants associated with CAD were obtained from a GWAS of 71 602 cases and 260 875 controls published by Nelson et al. 14 Cases were determined with a broad definition including fatal or nonfatal myocardial infarction, percutaneous transluminal coronary angioplasty, or coronary artery bypass grafting, chronic ischemic heart disease, and angina. 14 Informed consent was obtained from all participants, as well as ethical approval from their respective institutional review boards. More detailed information is displayed in Table S1.

Selection of Instrumental Variables

The genetic variants we used as IVs for cytokines were derived from a previous GWAS, which included 8293 individuals of European ethnic origin. 15 For main analyses, single nucleotide polymorphisms (SNPs) associated with circulating levels of 41 cytokines at the genome‐wide significance level (P<5×10−8) were first selected. After clumping all SNPs in linkage disequilibrium using a threshold r 2<0.01 (distance=1 Mb), we chose the SNPs with the lowest P values as IVs. To avoid the potential horizontal pleiotropy, 3 SNPs associated with >1 cytokine were excluded. Because 11 SNPs were unavailable in the CAD data set and no proxies with high linkage disequilibrium (r 2>0.8) can be found (https://ldlink.nih.gov, accessed on August 8, 2023), 111 SNPs for 28 cytokines were finally included in our MR analyses (Tables S2 and S3).

To validate our main results, we further used 2 sets of cis‐quantitative trait locus as complementary IVs to conduct MR analyses for the cytokines with significant associations with CAD. For protein quantitative trait locus (pQTL), we used genetic variants near (±500 kb) corresponding gene loci. Independent SNPs (r 2=0.01, distance=1 Mb) that met the genome‐wide statistical significance threshold (P<5×10−8) were selected as cis‐pQTL instruments. For the expression quantitative trait locus (eQTL), the genetic variants with P<5×10−8 from a ±1 Mb cis window around the transcription start site of encoded genes across 49 tissues in the Genotype‐Tissue Expression version 8 (National Institutes of Health) were retrieved and pruned to r 2=0.01 (distance=1 Mb). 16

Statistical Analysis

We conducted the MR analyses using the summarized data of associations of each genetic variant with 28 cytokines. The estimates of the causal effect were analyzed using the inverse‐variance‐weighted method, weighted median method, simple median method, maximum likelihood method, MR pleiotropy residual sum and outlier test, and MR‐Egger regression. Specifically, the inverse‐variance‐weighted method was used as a primary method, which provides a combined estimate of the causal estimate from each SNP. The Wald ratio method was applied when the MR estimate contained only a single SNP. 17 As the inverse‐variance‐weighted method can yield biased results if >1 IVs are invalid, we further performed the weighted median method. It uses inverse‐variance weights and bootstrap to calculate CIs, which can provide consistent estimates when up to 50% of the weights are derived from invalid instruments. 18 We also used the simple median method to provide an alternative estimate in which the effects of SNPs are not weighted. 19 The maximum likelihood method produces the causal estimates of exposure on the outcome by direct maximization of the likelihood under the assumption that outcome and exposure are linearly dependent and normally distributed. 20 We also performed the MR pleiotropy residual sum and outlier test to detect and correct horizontal pleiotropic outliers through outlier removal in multi‐instrument summary‐level MR testing. 21 The MR‐Egger regression test was used to evaluate the horizontal pleiotropy and investigate the null causal hypothesis under the Instrument Strength Independent of Direct Effect assumption. An intercept term that differs from zero indicates the existence of horizontal pleiotropy. 22

To validate the robustness of our main findings, we further performed several complementary analyses for the cytokines with significant associations with CAD risk. First, we conducted additional MR analyses using cis‐pQTL and cis‐eQTL IVs. Additionally, we applied 2 colocalization methods, COLOC 23 and HyPrColoc, 24 to cis‐pQTL, cis‐eQTL and CAD. The colocalization method was developed to assess whether 2 association signals are accordant with a common causal region. 23 Using COLOC, we performed colocalization analyses between pQTLs of target genes and CAD, as well as eQTLs and CAD, separately. For pQTL‐CAD colocalization, we set a region in a window of 250 kb of corresponding gene locus. For eQTL‐CAD colocalization, we extracted the cis‐eQTL signals within +/− 250kb of the transcription start site of the target gene across 49 tissues and colocalized with the CAD. We used default priors: p1=10−4, p2=10−4, p12=10−5. p1, p2, and p12 are the prior probabilities that an SNP in the tested region is significant with trait 1, trait 2, or both traits 1 and 2, respectively. The posterior probability of the association with trait 1 and trait 2 with 1 shared causal variant (PP.H4) was used to define potential colocalized signals. We considered COLOC tests with a PP.H4≥0.8 or PP.H4>0.5 as having strong or moderate evidence for colocalization. Furthermore, a multitrait (cis‐pQTL, cis‐eQTL, and CAD) colocalization analysis was performed on the basis of the Bayesian method with clustering algorithm using HyPrColoc. 24

The statistical tests of causal inference were considered statistically significant at Bonferroni‐adjusted P value <1.79×10−3 (0.05/28). We considered suggestive evidence for a potential causal association, if P value was between 1.79×10−3 and 0.05. All analyses were performed in R software version 3.6.4 (R Foundation for Statistical Computing, Vienna, Austria) using “TwoSampleMR,” “MendelianRandomization,” “MRPRESSO,” “coloc,” and “hyprcoloc” packages. Reporting follows the Strengthening the Reporting of Observational Studies in Epidemiology Using Mendelian Randomization statement (Table S4). 25 , 26

Results

The estimates of 28 cytokines on CAD risk are displayed in Table S5 and Figure 2.

Figure 2. The associations between circulating levels of 28 cytokines/growth factors and risk of coronary artery disease with inverse‐variance‐weighted/Wald ratio method.

Figure 2

Open in a new tab

*The odds ratio was estimated using Wald ratio method with only a single SNP as IV. For >1 SNP as IV, we presented an estimate from the inverse‐variance‐weighted random effect model when P<0.05 for Cochran's Q test; otherwise, the fixed‐effect model will be used. The line marked in red corresponds to the estimate reaching Bonferroni‐adjusted significance. CTACK indicates cutaneous T‐cell attracting (CCL27); GROα, growth regulated oncogene‐α (CXCL1); HGF, hepatocyte growth factor; IFN‐γ, interferon‐γ; IL‐10, interleukin‐10; IL‐12p70, interleukin‐12p70; IL‐13, interleukin‐13; IL‐16, interleukin‐16; IL‐17, interleukin‐17; IL‐18, interleukin‐18; IL‐2rα, interleukin‐2 receptor, α subunit; IL‐5, interleukin‐5; IL‐7, interleukin‐7; IP‐10, interferon‐γ–induced protein 10 (CXCL10); MCP‐1, monocyte chemotactic protein‐1 (CCL2); M‐CSF, macrophage colony‐stimulating factor; MIF, macrophage migration inhibitory factor (glycosylation‐inhibiting factor); MIG, monokine induced by interferon‐γ (CXCL9); MIP‐1β, macrophage inflammatory protein‐1β (CCL4); PDGF‐bb, platelet‐derived growth factor BB; RANTES, regulated on activation, normal T cell expressed and secreted (CCL5); SCF, stem cell factor; SCGF‐β, stem cell growth factor β; TNF‐β, tumor necrosis factor‐β; TRAIL, TNF‐related apoptosis inducing ligand; and VEGF, vascular endothelial growth factor.

We found that genetically predicted serum levels of MIF were associated with a higher risk of CAD using the Wald ratio method with 1 SNP (rs2330634). For 1‐SD increase of circulating MIF level, the OR was 1.20 (95% CI, 1.08–1.33; P=6.83×10−4) for CAD. The estimates from maximum likelihood method were consistent with the primary estimate (OR, 1.20 [95% CI, 1.06–1.35]; P=0.003). By searching the GWAS catalog, rs2330634 was not found to be susceptible to any other traits. We further performed complementary analyses focusing on cis‐pQTL and cis‐eQTL of the MIF gene. We noted that the genetic instrument (rs2330634) in the main analyses was the only variant as a cis‐pQTL instrument for MIF and yielded a same estimate. By using 1 cis‐eQTL signal (rs5751775) as the IV for MIF, however, we observed a negative association (Wald ratio method: OR, 0.97 [95% CI, 0.96–0.99]; P=8.02×10−5). The colocalization analysis showed moderate evidence to support the causal effect of MIF cis‐pQTL on CAD (PP.H4=0.553) (Table S6). A more evident signal was observed between MIF cis‐eQTL and CAD (PP.H4=0.875) (Table S6). Furthermore, MIF cis‐eQTL colocalizes for MIF cis‐pQTL and CAD with a posterior probability of a shared causal variant (rs5751775) of 75.5% (Figures S1 through S3).

There was a suggestive association between genetically predicted higher stem cell growth factor beta (SCGF‐β) concentrations and lower risk of CAD in the inverse‐variance‐weighted method (OR, 0.95 [95% CI, 0.91–0.98]; P=0.007), simple median (OR, 0.95 [95% CI, 0.90–0.99]; P=0.025) and maximum likelihood method (OR, 0.94 [95% CI, 0.91–0.98]; P=0.007), while no significant estimates were observed in the weighted median method (OR, 0.96 [95% CI, 0.91–1.01]; P=0.115). The MR pleiotropy residual sum and outlier test indicated no outliers and also showed a similar association between SCGF‐β and CAD (OR, 0.95 [95% CI, 0.92–0.97]; P=0.014). The intercept of MR‐Egger regression suggested that there is no evidence of directional horizontal pleiotropy (P for intercept, 0.601).

Discussion

In this study, by using the MR approach, we evaluated the potential causality of 28 cytokines on the risk of CAD. We found that genetically predicted circulating MIF levels were positively associated with CAD risk.

MIF, a potent proinflammatory cytokine, has been reported to be involved in the pathogenesis of cardiovascular disease. 27 , 28 One study found that patients with acute coronary syndrome had elevated plasma levels of MIF compared with stable CAD and healthy controls (P<0.001). 29 Makino et al 8 also reported a connection between a higher level of circulating MIF and an adverse long‐term outcome in patients with CAD with impaired glucose tolerance or type 2 diabetes in a prospective study (hazard ratio, 3.67 [95% CI, 2.07–5.40]). Similarly, a meta study comprising 1172 CAD cases and 1564 controls showed that C allele of the rs755622 in MIF was associated with a significantly higher risk of CAD (OR, 1.49 [95% CI, 1.22–1.81]). 30 Moreover, a multivariable MR study suggested that the causality of MIF and heart failure may be mediated by the risk of CAD. 31 Consistently, a prior MR study performed by Karhunen et al 13 evaluated the relationship between MIF and CAD and reported a nominal significant association using pQTL instruments (β=0.165, P=0.048) and eQTL (β=0.126, P=0.040). With a larger sample size of CAD GWAS, we observed a statistically significant relationship between circulating levels of MIF and CAD. However, we found that the cis‐eQTL for MIF showed an opposite direction of the effect on CAD risk. Interestingly, the multitrait colocalization analyses showed rs5751775, the lead cis‐eQTL for MIF, was the potentially shared causal variant across MIF cis‐eQTL, cis‐pQTL, and CAD. The C‐allele of rs5751775 was positively associated with expression of the corresponding mRNA across 49 tissues (data not shown in the text) but negatively associated with plasma abundance of the protein target (β=−0.135, P=8.39×10−8). The phenomenon of the uncoupling of gene and protein expression may be due to the differential translation, protein degradation, contextual confounders, or protein‐level buffering. 32 , 33 Several potential mechanisms may explain the association between MIF and CAD. 27 , 34 , 35 For instance, MIF is associated with mediators, such as E‐selectin, of leukocyte recruitment, which is a key process of atherosclerotic lesions. 27 , 35 Additionally, MIF is found as a platelet‐derived factor, regulating clot retraction properties related to atherosclerosis. 27 , 34

SCGF‐β is a newly discovered growth factor for hematopoietic progenitor cells. 36 With a combination of other cytokines, SCGF‐β shows a function in the burst‐promoting activity and granulocyte/macrophage colony stimulation activity in erythroid and granulocyte/macrophage progenitor cells. 37 , 38 A previous study demonstrated that SCGF‐β level was elevated in asymptomatic patients with unstable plaques compared with those with stable plaques, suggesting its role in the instability of plaque. 39 However, little is known about the link between SCGF‐β and CAD risk. In the present study, we observed a suggestively negative association between SCGF‐β and CAD using MR design. Considering the potential pleiotropy, one should be cautious when interpret our findings. Further work is necessary to validate this association.

For the other cytokines, Georgakis et al 40 found that a genetically predicted elevated monocyte chemotactic protein‐1 level was positively associated with the risk of CAD (OR, 1.04 [95% CI, 1.00–1.08]; P=0.04) on the basis of 60 801 CAD cases and 123 504 controls. However, it was not observed in our study, which may contribute to different strategies for IV selection. Georgakis et al set P<5×10−8 to select IVs, but they used a more liberal clumping criterion (r 2<0.1), which may include SNPs in higher linkage disequilibrium with each other and introduce a high probability of type 1 error. Yuan et al 41 also evaluated the associations between circulating interleukins and CAD and observed that interleukin‐6 and interleukin‐1 receptor antagonist were associated with altered risk of CAD (OR, 0.64 [95% CI, 0.54–0.76] for interleukin‐6; OR, 1.36 [95% CI, 1.14–1.63] for interleukin‐1 receptor antagonist). Moreover, Sjaarda et al 42 found that genetically predicted stromal cell‐derived factor 1 (CXCL12) level could increase the risk of CAD (OR, 1.69 [95% CI, 1.40–2.05]; P=6.2×10−8). Nevertheless, we did not obtain valid IVs for interleukin‐6, interleukin‐1 receptor antagonist, and CXCL12 at the threshold for the genome‐wide significance (P<5×10−8). Further work is warranted to be carried out to evaluate their associations with CAD.

The major strength of present study is the MR design, which can minimize residual confounding and reverse bias by using genetic variants as proxies of lifelong level of cytokines and growth factors. However, the present study also has some limitations. First, we were unable to investigate the nonlinear effects of cytokines on risk of CAD due to the assumption of 2‐sample MR analyses. Second, there may be minimal overlapped samples between cytokines and CAD, which may lead to weak instrument bias. However, the F statistics calculated in the present study are all >10, which may indicate the bias introduced by sample overlap should be minimal. Third, the data set of cytokines was derived from Finnish population with limited sample size, which may not represent the general European population and affect the generalizability of our findings. Finally, for several cytokines, such as β‐nerve growth factor and interleukin‐7, we have only a single genetic variant as IV, which might lead to lower power and therefore less precise estimates.

Conclusions

The current study provided genetic evidence that MIF is positively associated with risk of CAD, suggesting its potential role in CAD treatment. However, the mechanism warrants further investigation.

Sources of Funding

This work was supported by grants from the National Natural Science Foundation of China (82103936) and the Natural Science Foundation of Zhejiang Province (LQ21H260001).

Disclosures

None.

Supporting information

Tables S1–S6

Figures S1–S3

JAH3-13-e030726-s001.doc (1.2MB, doc)

Acknowledgments

The authors sincerely thank the researchers and participants of the original GWASs for the collection and management of the large‐scale data resources. Y. Li wrote the main manuscript text. B. Liu, Y. Chen, and Z. Liu performed main statistical analyses and prepared the additional materials. Drs Ye, Mao, and Sun conceptualized. Dr Sun revised the manuscript. All authors reviewed the manuscript.

This manuscript was sent to Julie K. Freed, MD, PhD, Associate Editor, for review by expert referees, editorial decision, and final disposition.

For Sources of Funding and Disclosures, see page 6.

Contributor Information

Yingying Mao, Email: myy@zcmu.edu.cn.

Xiaohui Sun, Email: 20191030@zcmu.edu.cn.

References

  • 1. Collaborators GBDCoD . Global, regional, and national age‐sex‐specific mortality for 282 causes of death in 195 countries and territories, 1980‐2017: a systematic analysis for the global burden of disease study 2017. Lancet. 2018;392:1736–1788. doi: 10.1016/S0140-6736(18)32203-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Libby P. Inflammation in atherosclerosis. Arterioscler Thromb Vasc Biol. 2012;32:2045–2051. doi: 10.1161/ATVBAHA.108.179705 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Yang X, Zhang D, Zhao Y, Liu D, Li Q, Guo C, Tian G, Han M, Qie R, Huang S, et al. Association between serum level of C‐reactive protein and risk of cardiovascular events based on cohort studies. J Hum Hypertens. 2021;35:1149–1158. doi: 10.1038/s41371-021-00546-z [DOI] [PubMed] [Google Scholar]
  • 4. Hemingway H, Philipson P, Chen R, Fitzpatrick NK, Damant J, Shipley M, Abrams KR, Moreno S, McAllister KS, Palmer S, et al. Evaluating the quality of research into a single prognostic biomarker: a systematic review and meta‐analysis of 83 studies of C‐reactive protein in stable coronary artery disease. PLoS Med. 2010;7:e1000286. doi: 10.1371/journal.pmed.1000286 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Li J, Zhang Y, Guo X, Wu Y, Huang R, Han X. Circulating level of monocyte chemoattractant protein‐1 and risk of coronary artery disease: a case‐control and Mendelian randomization study. Pharmgenomics Pers Med. 2021;14:553–559. doi: 10.2147/PGPM.S303362 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Danesh J, Kaptoge S, Mann AG, Sarwar N, Wood A, Angleman SB, Wensley F, Higgins JP, Lennon L, Eiriksdottir G, et al. Long‐term interleukin‐6 levels and subsequent risk of coronary heart disease: two new prospective studies and a systematic review. PLoS Med. 2008;5:e78. doi: 10.1371/journal.pmed.0050078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Waehre T, Damas JK, Gullestad L, Holm AM, Pedersen TR, Arnesen KE, Torsvik H, Froland SS, Semb AG, Aukrust P. Hydroxymethylglutaryl coenzyme a reductase inhibitors down‐regulate chemokines and chemokine receptors in patients with coronary artery disease. J Am Coll Cardiol. 2003;41:1460–1467. doi: 10.1016/s0735-1097(03)00263-8 [DOI] [PubMed] [Google Scholar]
  • 8. Makino A, Nakamura T, Hirano M, Kitta Y, Sano K, Kobayashi T, Fujioka D, Saito Y, Watanabe K, Watanabe Y, et al. High plasma levels of macrophage migration inhibitory factor are associated with adverse long‐term outcome in patients with stable coronary artery disease and impaired glucose tolerance or type 2 diabetes mellitus. Atherosclerosis. 2010;213:573–578. doi: 10.1016/j.atherosclerosis.2010.09.004 [DOI] [PubMed] [Google Scholar]
  • 9. Liang Y, Yang C, Zhou Q, Pan W, Zhong W, Ding R, Wang A. Serum monokine induced by gamma interferon is associated with severity of coronary artery disease. Int Heart J. 2017;58:24–29. doi: 10.1536/ihj.15-472 [DOI] [PubMed] [Google Scholar]
  • 10. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23:R89–R98. doi: 10.1093/hmg/ddu328 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Davies NM, Holmes MV, Davey SG. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362:k601. doi: 10.1136/bmj.k601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Bennett DA, Holmes MV. Mendelian randomisation in cardiovascular research: an introduction for clinicians. Heart. 2017;103:1400–1407. doi: 10.1136/heartjnl-2016-310605 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Karhunen V, Gill D, Huang J, Bouras E, Malik R, Ponsford MJ, Ahola‐Olli A, Papadopoulou A, Palaniswamy S, Sebert S, et al. The interplay between inflammatory cytokines and cardiometabolic disease: bi‐directional mendelian randomisation study. BMJ Med. 2023;2:e000157. doi: 10.1136/bmjmed-2022-000157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Nelson CP, Goel A, Butterworth AS, Kanoni S, Webb TR, Marouli E, Zeng L, Ntalla I, Lai FY, Hopewell JC, et al. Association analyses based on false discovery rate implicate new loci for coronary artery disease. Nat Genet. 2017;49:1385–1391. doi: 10.1038/ng.3913 [DOI] [PubMed] [Google Scholar]
  • 15. Ahola‐Olli AV, Wurtz P, Havulinna AS, Aalto K, Pitkanen N, Lehtimaki T, Kahonen M, Lyytikainen LP, Raitoharju E, Seppala I, et al. Genome‐wide association study identifies 27 loci influencing concentrations of circulating cytokines and growth factors. Am J Hum Genet. 2017;100:40–50. doi: 10.1016/j.ajhg.2016.11.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Consortium GT . The genotype‐tissue expression (GTEx) project. Nat Genet. 2013;45:580–585. doi: 10.1038/ng.2653 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37:658–665. doi: 10.1002/gepi.21758 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Bowden J, Smith GD, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genetic Epidemiology. 2016;40:304–314. doi: 10.1002/gepi.21965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Burgess S, Bowden J, Fall T, Ingelsson E, Thompson SG. Sensitivity analyses for robust causal inference from Mendelian randomization analyses with multiple genetic variants. Epidemiology. 2017;28:30–42. doi: 10.1097/EDE.0000000000000559 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Pagoni P, Dimou NL, Murphy N, Stergiakouli E. Using Mendelian randomisation to assess causality in observational studies. Evid Based Ment Health. 2019;22:67–71. doi: 10.1136/ebmental-2019-300085 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693–698. doi: 10.1038/s41588-018-0099-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through egger regression. Int J Epidemiol. 2015;44:512–525. doi: 10.1093/ije/dyv080 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Giambartolomei C, Vukcevic D, Schadt EE, Franke L, Hingorani AD, Wallace C, Plagnol V. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 2014;10:e1004383. doi: 10.1371/journal.pgen.1004383 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Foley CN, Staley JR, Breen PG, Sun BB, Kirk PDW, Burgess S, Howson JMM. A fast and efficient colocalization algorithm for identifying shared genetic risk factors across multiple traits. Nat Commun. 2021;12:764. doi: 10.1038/s41467-020-20885-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Skrivankova VW, Richmond RC, Woolf BAR, Davies NM, Swanson SA, VanderWeele TJ, Timpson NJ, Higgins JPT, Dimou N, Langenberg C, et al. Strengthening the reporting of observational studies in epidemiology using mendelian randomisation (STROBE‐MR): explanation and elaboration. BMJ. 2021;375:n2233. doi: 10.1136/bmj.n2233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Skrivankova VW, Richmond RC, Woolf BAR, Yarmolinsky J, Davies NM, Swanson SA, VanderWeele TJ, Higgins JPT, Timpson NJ, Dimou N, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE‐MR statement. JAMA. 2021;326:1614–1621. doi: 10.1001/jama.2021.18236 [DOI] [PubMed] [Google Scholar]
  • 27. van der Vorst EP, Doring Y, Weber C. MIF and CXCL12 in cardiovascular diseases: functional differences and similarities. Front Immunol. 2015;6:373. doi: 10.3389/fimmu.2015.00373 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Zernecke A, Bernhagen J, Weber C. Macrophage migration inhibitory factor in cardiovascular disease. Circulation. 2008;117:1594–1602. doi: 10.1161/CIRCULATIONAHA.107.729125 [DOI] [PubMed] [Google Scholar]
  • 29. Muller II, Muller KA, Schonleber H, Karathanos A, Schneider M, Jorbenadze R, Bigalke B, Gawaz M, Geisler T. Macrophage migration inhibitory factor is enhanced in acute coronary syndromes and is associated with the inflammatory response. PLoS One. 2012;7:e38376. doi: 10.1371/journal.pone.0038376 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Li DY, Zhang JY, Chen QJ, Liu F, Zhao Q, Gao XM, Li XM, Yang YN. MIF ‐173G/C (rs755622) polymorphism modulates coronary artery disease risk: evidence from a systematic meta‐analysis. BMC Cardiovasc Disord. 2020;20:300. doi: 10.1186/s12872-020-01564-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Moncla LM, Mathieu S, Sylla MS, Bosse Y, Theriault S, Arsenault BJ, Mathieu P. Mendelian randomization of circulating proteome identifies actionable targets in heart failure. BMC Genomics. 2022;23:588. doi: 10.1186/s12864-022-08811-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Pietzner M, Wheeler E, Carrasco‐Zanini J, Cortes A, Koprulu M, Worheide MA, Oerton E, Cook J, Stewart ID, Kerrison ND, et al. Mapping the proteo‐genomic convergence of human diseases. Science. 2021;374:eabj1541. doi: 10.1126/science.abj1541 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Buccitelli C, Selbach M. mRNAs, proteins and the emerging principles of gene expression control. Nat Rev Genet. 2020;21:630–644. doi: 10.1038/s41576-020-0258-4 [DOI] [PubMed] [Google Scholar]
  • 34. Wirtz TH, Tillmann S, Strussmann T, Kraemer S, Heemskerk JW, Grottke O, Gawaz M, von Hundelshausen P, Bernhagen J. Platelet‐derived MIF: a novel platelet chemokine with distinct recruitment properties. Atherosclerosis. 2015;239:1–10. doi: 10.1016/j.atherosclerosis.2014.12.039 [DOI] [PubMed] [Google Scholar]
  • 35. Cheng Q, McKeown SJ, Santos L, Santiago FS, Khachigian LM, Morand EF, Hickey MJ. Macrophage migration inhibitory factor increases leukocyte‐endothelial interactions in human endothelial cells via promotion of expression of adhesion molecules. J Immunol. 2010;185:1238–1247. doi: 10.4049/jimmunol.0904104 [DOI] [PubMed] [Google Scholar]
  • 36. Wang Y, Khan A, Heringer‐Walther S, Schultheiss HP, Moreira Mda C, Walther T. Prognostic value of circulating levels of stem cell growth factor beta (SCGF beta) in patients with Chagas' disease and idiopathic dilated cardiomyopathy. Cytokine. 2013;61:728–731. doi: 10.1016/j.cyto.2012.12.018 [DOI] [PubMed] [Google Scholar]
  • 37. Zuk PA, Zhu M, Ashjian P, De Ugarte DA, Huang JI, Mizuno H, Alfonso ZC, Fraser JK, Benhaim P, Hedrick MH. Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell. 2002;13:4279–4295. doi: 10.1091/mbc.e02-02-0105 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Hiraoka A, Sugimura A, Seki T, Nagasawa T, Ohta N, Shimonishi M, Hagiya M, Shimizu S. Cloning, expression, and characterization of a cDNA encoding a novel human growth factor for primitive hematopoietic progenitor cells. Proc Natl Acad Sci USA. 1997;94:7577–7582. doi: 10.1073/pnas.94.14.7577 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Schiro A, Wilkinson FL, Weston R, Smyth JV, Serracino‐Inglott F, Alexander MY. Elevated levels of endothelial‐derived microparticles, and serum CXCL9 and SCGF‐beta are associated with unstable asymptomatic carotid plaques. Sci Rep. 2015;5:16658. doi: 10.1038/srep16658 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Georgakis MK, Gill D, Rannikmae K, Traylor M, Anderson CD, Lee JM, Kamatani Y, Hopewell JC, Worrall BB, Bernhagen J, et al. Genetically determined levels of circulating cytokines and risk of stroke. Circulation. 2019;139:256–268. doi: 10.1161/CIRCULATIONAHA.118.035905 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Yuan S, Lin A, He QQ, Burgess S, Larsson SC. Circulating interleukins in relation to coronary artery disease, atrial fibrillation and ischemic stroke and its subtypes: a two‐sample Mendelian randomization study. Int J Cardiol. 2020;313:99–104. doi: 10.1016/j.ijcard.2020.03.053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Sjaarda J, Gerstein H, Chong M, Yusuf S, Meyre D, Anand SS, Hess S, Pare G. Blood CSF1 and CXCL12 as causal mediators of coronary artery disease. J Am Coll Cardiol. 2018;72:300–310. doi: 10.1016/j.jacc.2018.04.067 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Tables S1–S6

Figures S1–S3

JAH3-13-e030726-s001.doc (1.2MB, doc)

Articles from Journal of the American Heart Association: Cardiovascular and Cerebrovascular Disease are provided here courtesy of Wiley

RESOURCES