Combination of QTL and GWAS to uncover the role of phosphodiesterases in ischemic heart disease =============================================================================================== * Jun Xiao * Ziting Gao * Hongye Wei * Yajing Wei * Ziyi Qiu * Wuqing Huang ## Abstract **Background** Phosphodiesterases (PDEs) are regarded as important therapeutic targets for multiple diseases, and the cardiovascular benefits of several PDE inhibitors have received extensive interests. **Objectives** To explore the relationship between genetically-predicted PDEs and ischemia heart disease via drug target Mendelian Randomization (MR) approach. **Methods** The expression of genes encoding PDEs was used to proxy the level of PDEs, and available quantitative trait loci of gene expression and DNA methylation (eQTLs and mQTLs) for each target gene were identified as the genetic instruments. Coronary heart disease (CHD) and myocardial infarction (MI) were the outcomes. Summary-data-based MR method was used to generate the estimates and two-step MR analysis was applied to investigate the mediation of metabolic traits. **Results** MR analyses identified two types of PDEs, PDE5 and PDE8, genetically-predicted expression in blood of the encoded genes was significantly associated with the risk of CHD (ORPDE5A=1.22,95% CI=1.06-1.40; ORPDE8A=1.26,95% CI=1.07-1.49) and MI (ORPDE5A=1.27,95% CI=1.09-1.48; ORPDE8A=1.24,95% CI=1.04-1.48). Especially, the highest expression of PDE5A was observed in artery aorta, which was also positively related to CHD (OR=1.17,95% CI=1.05-1.32) and MI (OR=1.15,95% CI=1.02-1.30). Besides, the methylation level of 12 CpG sites showed a relation with CHD or MI via affecting PDE5A expression. The observed association between PDE5A expression and outcomes were partly mediated by blood pressure and LDL cholesterol, and the association with MI were mostly mediated by CHD (Proportion-mediated: 78.84%). **Conclusions** This study provided genetic evidence about the protective role of PDE5 inhibition against ischemic heart disease, especially in preventing patients with CHD from developing MI. Keywords * phosphodiesterases * coronary heart disease * myocardial infarction * Mendelian randomization ## Introduction Phosphodiesterases (PDEs), a family of enzymes from PDE1 to PDE11, are well-known to play a critical role in the regulation of intracellular signaling via hydrolyzing cyclic nucleotides cyclic adenosine monophosphate (cAMP) and/or cyclic guanosine monophosphate (cGMP)(1). The cAMP and cGMP are the second messengers linking to a series of physiological processes, thus PDEs are regarded as important therapeutic targets for multiple diseases (2). Although the PDE families are structurally related, the molecular diversity makes them functionally distinct (2). Some PDE inhibitors have been widely used in clinical practice, and emerging evidence indicates that PDE inhibitors have the potential to be developed or repurposed for a wide range of clinical applications (2,3). The cardiovascular benefits of PDE inhibitors have received extensive interests in recent decades, for example, accumulating epidemiological studies suggested that PDE5 inhibitors could enhance contractile function for patients with heart failure, improve the prognosis in patients with coronary heart disease, reduce ventricular arrhythmias, and so on (4–16). Coronary heart disease (CHD) is the most common type of cardiovascular diseases, and myocardial infarction (MI) is a major cause of death worldwide. Since 1960s, cAMP and cGMP were found to inhibit platelet aggregation and a group of PDE inhibitors were considered as anti-platelet agents, for instance, dipyridamole (PDE3 inhibitor) has been approved for secondary prevention of ischemic stroke (17). PDE inhibitors were also reported to ameliorate some established risk factors of coronary heart disease, such as metabolic syndrome, a cluster of several metabolic disorders (18). Although these evidence have suggested the cardiovascular benefits of PDE inhibitors, there is a paucity of studies comprehensively exploring the causal effect of PDE inhibitors on CHD or MI. Randomized clinical trial (RCT) is recognized as the gold standard to confirm the causal effect, however, RCT is not always feasible due to the difficulty of implementation. Mendelian randomization (MR) is thus developed for causal inference by using genetic variants as instruments. Previous studies have applied drug-target MR method to explore the potential drug targets by using quantitative trait loci (QTLs) as instruments, which are genetic variants associated with quantitative traits, such as gene expression (eQTL), protein (pQTL), methylation (mQTL), et al(19,20). Therefore, in this study, we aimed to investigate the role of PDEs on CHD and MI by using available eQTLs and mQTLs of genes encoding PDEs as instruments via MR approach and explore how the identified PDEs affect the outcome through metabolic traits. ## Methods ### Study design The design of the study was summarized in Figure 1. In the primary analysis, we comprehensively investigated the causal effect of the expression of genes encoding PDEs on the risk of CHD or MI by using eQTLs from blood as instruments. For significant MR associations, we further conducted a series of secondary analyses. Firstly, the observed significant associations were examined by using eQTLs in artery or heart tissues as instruments. Secondly, it is known that DNA methylation plays an important role in regulating gene expression, thus we further investigated the relationship between methylation level of related genes encoding significant PDEs and the outcomes by using mQTLs in blood as instruments. Thirdly, two-step MR analysis was used to explore if metabolic traits perform a mediating role between the identified PDEs and the outcomes, including body mass index (BMI), blood pressure, blood lipids and blood glucose. ![Figure 1.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/25/2023.09.24.23296052/F1.medium.gif) [Figure 1.](http://medrxiv.org/content/early/2023/09/25/2023.09.24.23296052/F1) Figure 1. Flowchart of the study. Summary-level data from publicly available genome-wide association studies (GWASs), expression and methylation quantitative trait loci (eQTLs and mQTLs) studies, were used in this study, all of which have been approved by the relevant institutional review boards with informed consents from participants. Details of data resources are presented in **Supplementary Table 1**. ### Selection of genetic instruments We used the expression of the encoded gene to proxy the level of each PDE, and eQTLs for related genes were used as instruments of gene expression. Given that data of eQTLs in blood are most comprehensive and top eQTLs are highly correlated between blood and other tissues, eQTLs in blood were used to generate genetic instruments to identified significant PDEs in the primary analysis. For significant associations, we provided additional evidence by using eQTLs in disease-related organs (ie.,artery and heart tissues) as instruments. Summary-level data of eQTLs in blood and tissues were obtained from eQTLGen Consortium (N=31,684) and GTEx Consortium V8 (Artery Aorta: N=387; Artery Tibial: N=372; Heart Atrial Appendage: N=584), respectively (21,22). And common SNPs significantly associated with the expression and the methylation in blood for related genes were identified and used to generate instruments of the level of methylation, summary data of mQTLs were selected from a meta-analysis of two cohorts (N=1,980) (23). To reduce weak instrument bias, we only included cis-QTLs (i.e., QTLs within 1LJMb windows from the region of the encoded gene). Hence, common cis-eQTLs or cis-mQTLs (minor allele frequency (MAF) >1%) significantly associated with the expression or methylation of PDEs were identified, in which p<5.0LJ×LJ10-8 was set as the significant threshold to select cis-eQTLs or cis-mQTLs in blood and the threshold of p<1.0LJ×LJ10-5 was used to select cis-eQTLs in tissues due to the limited number of available cis-eQTLs in tissues and the relatively small sample size of original studies. The most significant cis-eQTL or cis-mQTL for each gene was finally selected as the genetic instrument. The coefficient obtained from eQTL or mQTL study is equivalent to 1-SD change of the level of gene expression or DNA methylation for each additional effect allele. The details of QTLs data are presented in **Supplementary Table 1.** ### Outcome sources The outcome is ischemic heart disease, including CHD and MI. The GWAS summary-level data for CHD (cases: N=60,801; controls: N=123,504) and MI (cases: N=43,676; controls: N=128,199) were obtained from the CARDIoGRAMplusC4D Consortium with meta-analyses of 48 GWASs, in which the study population was predominantly individuals with European ancestry (24). Metabolic traits were considered as the potential mediators: summary data of GWAS for systolic and diastolic blood pressure was obtained from the International Consortium of Blood Pressure; data for low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C) and triglycerides (TG) obtained from the Global Lipids Genetics Consortium; data for body mass index obtained from the GIANT Consortium; data for fasting blood glucose obtained from the MAGIC Consortium. The details of GWASs data are presented in Supplementary Table 1. ### Statistical analyses Effect estimates were generated by using summary-data-based MR (SMR) method, which was recently developed to evaluate the association of quantitative trait and outcome combining summary-level data from QTL and GWAS studies (25). F-statistic was calculated to assess the strength of genetic instruments, an F-statistic of >10 indicates a small role of weak instrument bias in the results (26). The heterogeneity in dependent instruments (HEIDI) test was applied to assess if linkage disequilibrium played a role in the observed association of gene expression or methylation with outcome of interest, in which p value <0.01 indicates that association may be due to linkage. One SNP probably regulates the expression of more than one genes, resulting in the presence of horizontal pleiotropy. To evaluate the potential impact of horizontal pleiotropy, genes near the instrumental variant (within a 1Mb window) were identified if the expression was significantly associated with the variant. Then SMR analysis was conducted to examine if the instrumental variant affected the risk of outcomes via regulating the expression of nearby genes. Inverse-variance weighted mendelian randomization (IVW-MR) method was applied to obtain the estimates of the causal associations between metabolic traits and the outcome. To evaluate if the significant association between PDE and outcome was mediated by metabolic traits, we conducted a two-step MR study between PDE, candidate mediators, and incident outcome. Two-step MR, also known as network MR, is similar to the approach calculating the product of coefficients (27). We first calculated two MR coefficients: the association between PDE and the mediator (beta1) and the association between the mediator and the outcome (beta2), then multiplied two coefficients to estimate the indirect effect (beta1*beta2). The total effect was estimated by MR coefficient of PDE on the outcome (beta0) and the direct effect was calculated by the difference between total effect and indirect effect (beta0-beta1*beta2). Proportion of mediated effect was calculated by the formula of (beta1*beta2)/beta0. P-value of <0.05 was considered as statistical significance. To account for multiple testing correction, Bonferroni correction was also used to adjust the thresholds of significance level, with a strong evidence of P<0.05/(number of genes) and a suggestive evidence of 0.05/(number of genes)<=P<0.05. All analyses were conducted by using SMR software (version 1.03) or R software (version 4.1.0). ## Results ### Genetic instrument selection As shown in Figure 1 **and Supplementary Table 2,** a total of 7,737 SNPs significantly associated with the expression in blood of 17 genes encoding PDE1 to PDE9 were identified. While there is a lack of available cis-eQTLs identified for PDE10 and PDE11, thus these two PDEs were not included in the analyses. The most significant cis-eQTL for all 17 encoded genes had F statistics of >10, indicating a small role of weak instrument bias in our study. Approved or investigational drugs targeting PDEs are presented in **Supplementary Table 2.** ### MR analysis between the expression of genes encoding PDEs in blood and outcomes As shown in Figure 2 **and Supplementary table 3**, the most significant association with the risk of both CHD and MI was found on PDE5A expression, in which a 1-SD increase of PDE5A expression was associated with increased risk of CHD with an OR of 1.22 (95% CI=1.06-1.40, PLJ=LJ0.0046) or MI with an OR of 1.27 (95% CI=1.09-1.48, PLJ=LJ0.0024). The observed association with MI kept significant after bonferroni correction using a significance threshold of PLJ0.01). No nearby genes were found to be significantly related to the instrumental variant of PDE5A expression in artery aorta (rs1393966) (**Supplementary table 5**). No association was observed between the expression in heart atrial appendage of gene nearby the instrumental variant of PDE5A expression (rs75098287) and the outcomes **(Supplementary Table 6)**. These results suggesting a small role of the horizontal pleiotropy in the observed associations. ![Figure 3.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/25/2023.09.24.23296052/F3.medium.gif) [Figure 3.](http://medrxiv.org/content/early/2023/09/25/2023.09.24.23296052/F3) Figure 3. Association between the expression of PDE5A or PDE8A in artery or heart tissue and coronary heart disease or myocardial infarction. ### MR analysis between the methylation of PDE5A or PDE8A in blood and outcomes There were 1,185 and 310 cis-mQTLs in blood identified to generate instruments of DNA methylation in 14 CpG sites of PDE5A and 12 CpG sites of PDE8A, respectively (Figure 1). Among these CpG sites, the methylation level of 12 CpG sites of PDE5A and 7 CpG sites of PDE8A robustly associated with the expression of the corresponding gene had significant associations with the risk of CHD and MI in the same direction, indicating that methylation in these CpG sites may affect the outcomes via regulating gene expression(Figure 4**, Supplementary table 7 and 8**). ![Figure 4.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/25/2023.09.24.23296052/F4.medium.gif) [Figure 4.](http://medrxiv.org/content/early/2023/09/25/2023.09.24.23296052/F4) Figure 4. Association between the methylation of PDE5A or PDE8A and the gene expression in blood, coronary heart disease or myocardial infarction. (A) PDE5A; (B) PDE8A. ### Two step MR analyses between PDE5A expression, metabolic traits and outcomes As shown in **Supplementary table 9,** the expression of PDE5A but not PDE8A was related to multiple metabolic traits, thus two step MR analyses were performed to explore the causal relationship between PDE5A expression, metabolic traits and the risk of CHD or MI. Figure 5 showed that the association between PDE5A expression and CHD or MI was partly mediated by diastolic blood pressure (Proportion-mediated CHD=25.97%; Proportion-mediated MI=21.42%), systolic blood pressure (Proportion-mediated CHD =15.77%; Proportion-mediated MI =11.63%) and LDL-C (Proportion-mediated CHD=15.35%; Proportion-mediated MI = 8.17%). The association between PDE5A expression and MI was mostly mediated by CHD (Proportion mediated=78.84%). No mediated role of body mass index, blood glucose or other lipids was observed between PDE5A expression and outcomes. ![Figure 5.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/25/2023.09.24.23296052/F5.medium.gif) [Figure 5.](http://medrxiv.org/content/early/2023/09/25/2023.09.24.23296052/F5) Figure 5. Two-step mendelian randomization between PDE5A expression, metabolic traits and coronary heart disease or myocardial infarction. (A) Coronary heart disease; (B) Myocardial infarction. * Percentage in plots indicates the proportion of mediated effect. The asterisk in plots indicates statistical significance (P<0.05). ## Discussion To investigate the therapeutical potential of PDE inhibitors for ischemic heart disease and the underlying pathways, we explored the association of the expression of genes encoding PDEs (i.e., PDE1-PDE9) with the risk of CHD and MI in this MR study. Two types of PDEs, PDE5 and PDE8, were identified to have the potential to be repurposed for CHD or MI prevention in this study. Specifically, a strong evidence was observed for the positive association of PDE5A expression in blood and artery aorta (the tissue with highest expression) with the risk of CHD and MI. We also identified 12 CpG sites, the methylation level of which was significantly related to the risk of CHD and MI via affecting the PDE5A expression. Mediation analysis suggested that PDE5 inhibition might reduce the incidence of ischemic heart disease partly through regulating blood pressure and LDL cholesterol. These findings provided genetic evidence that PDE5 inhibitors may be effective in the prevention of CHD and MI, especially in protecting against MI among patients with CHD. PDE5, encoded by PDE5A gene, is an enzyme selectively hydrolyzing cGMP, which is distributed across most tissues of human body. Thus, it is known that PDE5 inhibitors, including sildenafil, vardenafil and tadalafil, exert its therapeutic effect by inhibiting cGMP degradation. PDE5 inhibitors were originally developed to treat angina while failed in clinical trials, then repurposed and approved by FDA for treating erectile dysfunction in 1998, pulmonary arterial hypertension in 2005 and benign prostatic hyperplasia in 2011 (4–6). Given of the safety of PDE5 inhibitors and the involvement of cGMP in vast physiological processes, enormous interest has been raised to explore the potential new functions of PDE5 inhibitors(6). Growing preclinical studies have reported that PDE5 inhibitors play a protective role against cardiovascular diseases, such as heart failure, MI, ischemia-induced ventricular arrhythmias, myocardial hypertrophy and so on (4–6). Several clinical trials have demonstrated the benefits of PDE5 inhibitors in patients with heart failure, while the effect on ischemic heart disease is still uncertain (28). A number of observational studies have reported the protective effect of PDE5 inhibitors on ischemic heart disease (29–31). For example, a nationwide Danish cohort study including 71,710 men found that PDE5 inhibitors use was related to a decreased incidence of MI as compared to the general male population without receiving PDE5 inhibitors (RR=0.74; 95% CI=0.69-0.79), however indication bias may play a role in the observed association (31). Another case-control study including type 2 diabetic men with silent coronary heart disease found that treatment with PDE5 inhibitors and statins was negatively associated with the occurrence of major adverse cardiac events (HR=0.66, 95%CI=0.46-0.97), while in which the effect of statins and PDE5 inhibitors cannot be distinguished and indication bias cannot be eliminated as well(30). A recent real-world study based on Swedish nationwide registers investigated the protective effect by comparing the survival between men treated with PDE5 inhibitors and those treated with alprostadil in order to reduce the influence from indication bias, since erectile dysfunction is the major indication of PDE5 inhibitors and alprostadil is another treatment for erectile dysfunction(29). This study included 18,542 men with stable coronary heart disease and found that men treated with PDE5 inhibitors had a lower mortality as well as a reduced incidence of major adverse cardiac events, including MI (HR= 0.81, 95%CI=0.70–0.93). Although indication bias can be reduced as much as possible by using active comparator, it is impossible to be completely avoided and unknown confounders cannot be ruled out in observational studies. Besides, as these drugs were mainly prescribed for men to treat erectile dysfunction in real world, it is unlikely to assess the effect in general population. Our study observed consistent results that increased expression of PDE5A was related to a higher risk of CHD and MI, which provided strong supportive evidence for findings from previous observational studies. Besides, we found that the mediated proportion of CHD on the positive relation between PDE5 and MI risk was as high as round 80%, indicating that PDE5 inhibition probably contributed to preventing patients with CHD from MI. Furthermore, PDE5 has been emerged as novel target for treating metabolic disorders, which are known risk factors of ischemic heart disease (18). This study also observed a mediated effect of blood pressure and LDL cholesterol between PDE5 inhibition and ischemic heart disease, which offered some clues for the underlying mechanisms, calling for further studies. PDE8 was also identified in this study, the inhibition of which was recently found to be a potential novel therapeutic target of treating airway diseases and dementia (32,33). No studies reported the cardiovascular effect of PDE8. Although this MR study found that the expression in blood of PDE8A was related to the risk of CHD and MI, the association did not pass the bonferroni correction test. Besides, PDE8A was expressed less in cardiovascular organs and no relationship was observed between the PDE8A expression in artery and the outcomes. Therefore, whether PDE8 inhibition really had a cardiovascular benefit need to be further investigated. This is the first MR study to comprehensively explore the causal effects of PDEs on ischemic heart disease. The use of QTLs as instruments to proxy drug exposure in MR study could overcome the shortcomings of observational studies, such as confounding bias and reverse causation. Besides, several types of QTLs (eQTLs in blood, artery aorta and heart, mQTLs in blood) have provided consistent evidence, indicating the robustness of the protective effect of PDE5 inhibition on ischemic heart disease. Two-step MR analysis contributed to identifying the subgroup population that may benefit more from PDE5 inhibitors use. However, there are several limitations. Firstly, this study used the expression of drug target genes to proxy drug exposure, which might not be equivalent to clinical treatment because of a gap between drug targets and available agents, for example, there may be various effects across different types of PDE5 inhibitors. However, these findings could provide additional evidence for previous observational studies regarding the preventive effect of PDE5 inhibitors against ischemic heart disease. Hence, RCTs are required to further investigate the efficacy of different PDE5 inhibitors. Secondly, horizontal pleiotropy cannot be excluded when using eQTLs in blood as instruments for PDE5A expression in primary analysis, while consistent associations were observed using eQTLs in tissues or mQTLs in blood as instrument, suggesting the robustness of the findings. Thirdly, false negative association of other PDEs cannot be ruled out in drug target MR study as there may be unidentified eQTLs which play a more important role in gene expression than the identified top eQTL. Fourthly, this study was conducted based on summary-level data, thus we were unable to explore the specific effect in subgroup population, such as in men and women. Fifthly, genetic data of this study were mainly collected from population of European ancestry, leading to a limited generalization of these findings. ## Conclusion In conclusion, this drug-target MR study identified PDE5 that had the potential to be repurposed as novel target for CHD and MI prevention, especially in protecting against MI among patients with CHD, and the underlying mechanisms of the effect might be partly involved in regulating blood pressure and LDL cholesterol. ## Supporting information Supplementary figure 1 [[supplements/296052_file07.jpg]](pending:yes) Supplementary figure 2 [[supplements/296052_file08.jpg]](pending:yes) Supplementary tables [[supplements/296052_file09.xlsx]](pending:yes) ## Data Availability All raw data produced are available online at eQTLGen Consortium, GTEx Portal, Global Lipids Genetics Consortium, International Consortium of Blood Pressure, GIANT Consortium, MAGIC Consortium and CARDIoGRAMplusC4D Consortium. ## Ethical statement All publicly available genome-wide association studies (GWASs) and quantitative trait loci (QTLs) studies used in this MR analyses were approved by the relevant institutional review boards with informed consents from participants. ## Funding Information This work was supported by National Natural Science Foundation of China (82304213) and Start-up Fund for high-level talents of Fujian Medical University (XRCZX2021026) to Dr. Wuqing Huang. The funder had no role in study design, data collection and interpretation, or the decision to submit the work for publication. ## Author Contributors All authors were responsible for the study concept and design. WH obtained funding. JX did the statistical analysis. JX and WH drafted the manuscript, and all authors revised it for important intellectual content. ## Competing interests The authors declare no conflict of interest. ## Acknowledgments The data used for the analyses described in this manuscript were obtained from the eQTLGen Consortium, GTEx Portal, Global Lipids Genetics Consortium, International Consortium of Blood Pressure, GIANT Consortium, MAGIC Consortium and CARDIoGRAMplusC4D Consortium on 01/07/2023. We thank the patients and investigators who contributed to these studies. ## Abbreviations CHD : coronary heart disease CI : confidence interval CpG : Cytosine-guanine dinucleotides eQTLs : expression quantitative trait loci GWAS : genome-wide association study HEIDI : heterogeneity in dependent instruments mQTL : Methylation quantitative trait loci MI : myocardial infarction OR : odds ratio PDE : phosphodiesterase SNP : single-nucleotide polymorphism SMR : summary-data-based Mendelian Randomization * Received September 24, 2023. * Revision received September 24, 2023. * Accepted September 25, 2023. * © 2023, Posted by Cold Spring Harbor Laboratory The copyright holder for this pre-print is the author. All rights reserved. The material may not be redistributed, re-used or adapted without the author's permission. ## References 1. 1.Boswell-Smith V, Spina D, Page CP. Phosphodiesterase inhibitors. British journal of pharmacology 2006;147 Suppl 1:S252–7. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/sj.bjp.0706495&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16402111&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000235087400029&link_type=ISI) 2. 2.Maurice DH, Ke H, Ahmad F, Wang Y, Chung J, Manganiello VC. Advances in targeting cyclic nucleotide phosphodiesterases. Nature reviews Drug discovery 2014;13:290–314. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nrd4228&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24687066&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 3. 3.Huang W, Sundquist J, Sundquist K, Ji J. Use of Phosphodiesterase 5 Inhibitors Is Associated With Lower Risk of Colorectal Cancer in Men With Benign Colorectal Neoplasms. Gastroenterology 2019;157:672–681.e4. 4. 4.Hutchings DC, Anderson SG, Caldwell JL, Trafford AW. Phosphodiesterase-5 inhibitors and the heart: compound cardioprotection? Heart (British Cardiac Society) 2018;104:1244–1250. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiaGVhcnRqbmwiO3M6NToicmVzaWQiO3M6MTE6IjEwNC8xNS8xMjQ0IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjUvMjAyMy4wOS4yNC4yMzI5NjA1Mi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 5. 5.Ravipati G, McClung JA, Aronow WS, Peterson SJ, Frishman WH. Type 5 phosphodiesterase inhibitors in the treatment of erectile dysfunction and cardiovascular disease. Cardiology in review 2007;15:76–86. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1097/01.crd.0000233904.77128.49&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17303994&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 6. 6.Das A, Durrant D, Salloum FN, Xi L, Kukreja RC. PDE5 inhibitors as therapeutics for heart disease, diabetes and cancer. Pharmacology & therapeutics 2015;147:12–21. 7. 7.Hashimoto T, Kim GE, Tunin RS et al. Acute Enhancement of Cardiac Function by Phosphodiesterase Type 1 Inhibition. Circulation 2018;138:1974–1987. 8. 8.Nakano SJ, Miyamoto SD, Movsesian M, Nelson P, Stauffer BL, Sucharov CC. Age-related differences in phosphodiesterase activity and effects of chronic phosphodiesterase inhibition in idiopathic dilated cardiomyopathy. Circulation Heart failure 2015;8:57–63. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NjoiY2lyY2hmIjtzOjU6InJlc2lkIjtzOjY6IjgvMS81NyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzI1LzIwMjMuMDkuMjQuMjMyOTYwNTIuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 9. 9.Pagani ED, Buchholz RA, Silver PJ. Cardiovascular cyclic nucleotide phosphodiesterases and their role in regulating cardiovascular function. Basic research in cardiology 1992;87 Suppl 1:73–86. 10. 10.Kloner RA. Cardiovascular effects of the 3 phosphodiesterase-5 inhibitors approved for the treatment of erectile dysfunction. Circulation 2004;110:3149–55. [FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiRlVMTCI7czoxMToiam91cm5hbENvZGUiO3M6MTQ6ImNpcmN1bGF0aW9uYWhhIjtzOjU6InJlc2lkIjtzOjExOiIxMTAvMTkvMzE0OSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzI1LzIwMjMuMDkuMjQuMjMyOTYwNTIuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 11. 11.Vinogradova TM, Lakatta EG. Dual Activation of Phosphodiesterase 3 and 4 Regulates Basal Cardiac Pacemaker Function and Beyond. International journal of molecular sciences 2021;22. 12. 12.Leroy J, Fischmeister R. Inhibit a Phosphodiesterase to Treat Heart Failure? Circulation 2018;138:2003–2006. 13. 13.Roks AJM. Phosphodiesterase-1 in the cardiovascular system. Cellular signalling 2022;92:110251. 14. 14.Bevan S, Porteous L, Sitzer M, Markus HS. Phosphodiesterase 4D gene, ischemic stroke, and asymptomatic carotid atherosclerosis. Stroke 2005;36:949–53. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6OToic3Ryb2tlYWhhIjtzOjU6InJlc2lkIjtzOjg6IjM2LzUvOTQ5IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjUvMjAyMy4wOS4yNC4yMzI5NjA1Mi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 15. 15.Lee DI, Zhu G, Sasaki T et al. Phosphodiesterase 9A controls nitric-oxide-independent cGMP and hypertrophic heart disease. Nature 2015;519:472–6. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nature14332&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25799991&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 16. 16.Chen S, Zhang Y, Lighthouse JK et al. A Novel Role of Cyclic Nucleotide Phosphodiesterase 10A in Pathological Cardiac Remodeling and Dysfunction. Circulation 2020;141:217–233. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1161/circulationaha.119.042178&link_type=DOI) 17. 17.Smolenski A. Novel roles of cAMP/cGMP-dependent signaling in platelets. Journal of thrombosis and haemostasis : JTH 2012;10:167–76. 18. 18.Lugnier C, Meyer A, Talha S, Geny B. Cyclic nucleotide phosphodiesterases: New targets in the metabolic syndrome? Pharmacology & therapeutics 2020;208:107475. 19. 19.Gaziano L, Giambartolomei C, Pereira AC et al. Actionable druggable genome-wide Mendelian randomization identifies repurposing opportunities for COVID-19. Nat Med 2021;27:668–676. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41591-021-01310-z&link_type=DOI) 20. 20.Huang W, Xiao J, Ji J, Chen L. Association of lipid-lowering drugs with COVID-19 outcomes from a Mendelian randomization study. eLife 2021;10. 21. 21.Vosa U, Claringbould A, Westra HJ et al. Large-scale cis-and trans-eQTL analyses identify thousands of genetic loci and polygenic scores that regulate blood gene expression. Nat Genet 2021;53:1300–1310. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1177/0022146515594631.Marriage&link_type=DOI) 22. 22.Consortium GT. The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 2020;369:1318–1330. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEzOiIzNjkvNjUwOS8xMzE4IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjUvMjAyMy4wOS4yNC4yMzI5NjA1Mi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 23. 23.McRae AF, Marioni RE, Shah S et al. Identification of 55,000 Replicated DNA Methylation QTL. Sci Rep 2018;8:17605. 24. 24.Nikpay M, Goel A, Won HH et al. A comprehensive 1,000 Genomes-based genome-wide association meta-analysis of coronary artery disease. Nat Genet 2015;47:1121–1130. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/ng.3396&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26343387&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 25. 25.Zhu Z, Zhang F, Hu H et al. Integration of summary data from GWAS and eQTL studies predicts complex trait gene targets. Nat Genet 2016;48:481–7. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/ng.3538&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=27019110&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 26. 26.Burgess S, Thompson SG. Avoiding bias from weak instruments in Mendelian randomization studies. Int J Epidemiol 2011;40:755–64. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/ije/dyr036&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21414999&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000293618300027&link_type=ISI) 27. 27.Carter AR, Sanderson E, Hammerton G et al. Mendelian randomisation for mediation analysis: current methods and challenges for implementation. Eur J Epidemiol 2021;36:465–478. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10654-021-00757-1&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F25%2F2023.09.24.23296052.atom) 28. 28.Nadur NF, de Azevedo LL, Caruso L, Graebin CS, Lacerda RB, Kümmerle AE. The long and winding road of designing phosphodiesterase inhibitors for the treatment of heart failure. European journal of medicinal chemistry 2021;212:113123. 29. 29.Andersson DP, Landucci L, Lagerros YT et al. Association of Phosphodiesterase-5 Inhibitors Versus Alprostadil With Survival in Men With Coronary Artery Disease. Journal of the American College of Cardiology 2021;77:1535–1550. 30. 30.Gazzaruso C, Solerte SB, Pujia A et al. Erectile dysfunction as a predictor of cardiovascular events and death in diabetic patients with angiographically proven asymptomatic coronary artery disease: a potential protective role for statins and 5-phosphodiesterase inhibitors. Journal of the American College of Cardiology 2008;51:2040–4. [FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czo0OiJhY2NqIjtzOjU6InJlc2lkIjtzOjEwOiI1MS8yMS8yMDQwIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjUvMjAyMy4wOS4yNC4yMzI5NjA1Mi5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 31. 31.Vestergaard N, Søgaard P, Torp-Pedersen C, Aasbjerg K. Relationship between treatment of erectile dysfunction and future risk of cardiovascular disease: A nationwide cohort study. European journal of preventive cardiology 2017;24:1498–1505. 32. 32.Johnstone TB, Smith KH, Koziol-White CJ et al. PDE8 Is Expressed in Human Airway Smooth Muscle and Selectively Regulates cAMP Signaling by beta(2)-Adrenergic Receptors and Adenylyl Cyclase 6. Am J Respir Cell Mol Biol 2018;58:530–541. 33. 33.Wu XN, Zhou Q, Huang YD et al. Structure-based discovery of orally efficient inhibitors via unique interactions with H-pocket of PDE8 for the treatment of vascular dementia. Acta Pharm Sin B 2022;12:3103–3112.