Using Genetics to Inform The Role of Hemostatic Factors in COVID-19: A Mendelian Randomization Study

Background: Severe corona virus disease 19 (COVID-19) is challenging to treat as it presents with thrombotic as well as inammatory features. Specically, endothelial cells are thought to be damaged in COVID-19. Here we assessed whether key druggable targets related to endothelial function, i.e., nitric oxide and platelets, are affected specically in severe COVID-19, using a Mendelian randomization study. Methods: We compared genetically predicted circulating nitric oxide, proxied by three relevant NOS3 functional variants (rs2070744, rs1799983 and rs3918226), and genetically predicted platelets, proxied by a functional variant related to platelet reactivity from GRK5 (rs10886430) and by platelet count (based on 243 variants), because they share a phenotype, in people with severe COVID-19 with the general population (cases=536, non-cases=329,391) largely based on studies from Northern Europe. We made a similar comparison for any COVID-19. Results: Nitric oxide was inversely associated with severe COVID-19 (odds ratio (OR) 0.84 per raising allele, 95% condence interval (CI) 0.75 to 0.95), but was not associated with any COVID-19. Associations of both platelet reactivity and platelet count with severe COVID-19 were in a harmful direction (OR 1.20, 95% CI 0.95 to 1.50 and OR 1.23, 95% CI 0.97 to 1.56 respectively) with combined OR 1.21, 95% CI 1.03 to 1.42, but associations with any COVID-19 were null. Conclusions: Genetic validation of the role of nitric oxide and possibly of platelets in specically severe COVID-19 suggests they could be targets of intervention for which well-established treatments exist.


Background
Treatment of severe corona virus disease  has improved substantially over the course of the pandemic, informed by clinical insights and ndings from trials. Nevertheless, COVID-19 presents a complex challenge as a respiratory disease with thrombotic features, whose origins are not entirely clear, but may involve a range of factors including the endothelium. 1 COVID-19 damages endothelial cells; 2 a key part of the hemostatic system. Speci cally, endothelial cells regulate arterial tone by generating contracting factors such as thromboxane A2 and relaxing factors, such as nitric oxide. 3,4 Nitric oxide, formerly known as endothelium-derived relaxing factor, can act as a vasodilator and anti-thrombotic agent. 3,5 Several trials are currently investigating the role of inhaled nitric oxide as a treatment for COVID-19, https://clinicaltrials.gov/ct2/results/details?term=nitric+oxide+COVID, but . systemic nitric oxide has had less attention https://racetoacure.stanford.edu/clinical-trials/991, although it is the target of longstanding cardiovascular drugs, such as nitroglycerin.
Endothelial cells also modulate platelet aggregation, 6, 7 possibly via thrombin, 8 which is the key coagulation factor associated with ischemic heart disease. 9 Thrombin induced platelet reactivity and aggregation has recently been shown to be relevant to a range of cardiovascular diseases, 10 consistent with the well-established role of platelets in thrombosis. 11 Platelet reactivity and aggregation is also the target of long-standing cardiovascular drugs, such as aspirin. The relevance of platelets to COVID-19, particularly severe COVID-19, is unclear, although more platelet aggregation has been seen in more severe cases. 12 However, observational studies in selected samples, such as patients, can be di cult to interpret, given potential selection bias, confounding, and typically small sample sizes.
As COVID-19 may impact endothelial function, investigating factors relevant to haemostatic processes in severe COVID-19 may provide additional insight. In this study, we used mendelian randomization, a design robust to confounding, to assess whether nitric oxide or platelets contributed to very severe COVID-19. We used any COVID-19 as a control outcome as we would not expect factors involved in the response to COVID-19 to necessarily to be the same as the factors that result in COVID-19 infection, particularly as severe COVID-19 is only experienced by a minority of those infected. We included hospitalized COVID-19 for completeness.

Methods
This is a two-sample mendelian randomization study. We obtained genetic instruments for nitric oxide and platelets from the largest publicly available published studies. Preferentially we selected genetic instruments on functional grounds, because functional variants known physiologically to correspond to a speci c phenotype are most likely to capture it both comprehensively and exclusively. Otherwise, we selected genetic variants statistically (as reaching genome wide signi cance. We applied these genetic predictors of nitric oxide and platelets to the largest publicly available genome wide association study (GWAS) of COVID-19 to determine if people with genetically different levels of nitric oxide or platelets also differed in their vulnerability to very severe COVID-19, and obtained mendelian randomization estimates. We similarly considered any and hospitalized COVID-19.

Platelets
We used both platelet reactivity in response to thrombin and platelet count, because few genetic predictors of platelet reactivity are available, and platelet count may be phenotypically similar to platelet reactivity. 10 One genetic variant (rs10886430 (GRK5)) functionally and statistically relevant to thrombin induced platelet reactivity is available. 10 We extracted genetic variants for platelet count from a published GWAS giving independent genome-wide signi cant genetic predictors. 15 We presented platelet reactivity estimates in terms of a square root transform, which was used to normalize the distribution, 10 and platelet count in effect sizes. 15 Outcome We obtained associations with COVID-19 from the latest publicly available GWAS summary statistics.

Statistical Analysis
We aligned genetic variants for the exposures (eNOS or platelets) and genetic associations with COVID-19 on the same effect allele. There were no palindromic SNPs for eNOS or platelet reactivity. For the genetic variants predicting platelet count we did not drop palindromic SNPs because the GWAS for platelet count and COVID-19 both used the same strand direction. For genetic variants not in the COVID-19 GWAS, proxies (r 2 >0.8) were identi ed. We meta-analyzed genetic variant speci c Wald estimates (genetic variant on COVID-19 divided by genetic variant on exposure) with the standard error obtained from the rst term of Fieller's theorem. 16 We used inverse variance weighting, with multiplicative random effects for three or more genetic variants. We also used the weighted median and MR-Egger estimates as sensitivity analysis, 17 where possible (i.e., when three or more independent genetic predictors of each exposure were available). Since variants for nitric oxide are correlated, we obtained their correlations using LdLink (https://ldlink.nci.nih.gov/), and included them in the analysis. We obtained mendelian randomization estimates using the MendenlianRandomization R package. All analyses were performed using R Version 3.6.1 (R Development Core Team, Vienna, Austria).

Ethics approval
This study only used publicly available summary statistics and hence no ethics approval is required. This study complies with the Declaration of Helsinki.

Nitric Oxide
Each of the three eNOS genetic variants (rs2070744 (NOS3), rs1799983 (NOS3) and rs3918226 (NOS3)) associated with greater eNOS activity was also associated with a lower risk of very severe COVID-19 (Supplementary Table 1). Taken together, accounting for correlations, genetically predicted nitric oxide was inversely associated with very severe COVID-19 but was not associated with hospitalized or any COVID-19 (Figure 1 and supplementary Table 2).

Platelets
The one genetic variant (rs10886430) functionally and statistically relevant to thrombin induced platelet reactivity 10 was available for each COVID-19 case de nition. Of 287 published variants predicting platelet count, 10 230 were available for very severe COVID-19, based on position or rsid, and a further 13 had proxies (r 2 >0.8), giving 243 genetic predictors in total for platelet count, more variants were available for hospitalized (n=263) and severe COVID-19 (n=268). Both platelet reactivity and platelet count had associations in the direction of higher risk of severe COVID-19 ( Figure 1 and Supplementary Table 3).

Discussion
Consistent with the role of endothelial cells in severe COVID-19, 2 we present genetic validation of a protective role of a druggable factor, nitric oxide, that regulates endothelial function, 3,4 in speci cally severe COVID-19. In addition, we show that a platelet phenotype might also be relevant speci cally to severe COVID-19.
Nitric oxide and platelets are the targets of some of the oldest treatments for cardiovascular disease.
Nitric oxide forms the basis of a very long-standing treatment for angina and acute myocardial infarction, i.e., nitroglycerin, 18 while aspirin is one of the most effective anti-platelet agents in secondary prevention. 19 Another, very well-established treatment for cardiovascular disease, statins, may improve endothelial function, 20 and increase eNOS 21 as well as reducing platelet count. 15 These ndings also have some consistency with men being more vulnerable to severe COVID-19 than women. 22 Typically men have lower nitric oxide than women, 23 and lower endothelial function, 24 but also lower platelet counts with little difference in platelet reactivity. 25 Whether any common underlying factor, such as hormones, is driving associations of haemostatic and coagulation factors with COVID-19, might bear consideration.
Despite providing some genetic validation for the role of key druggable haemostatic factors in severe COVID-19, this study has limitations. First, valid genetic instruments need to ful l three assumptions, i.e., relate strongly to the exposure, not be associated with potential confounders of the effects of the exposure on the outcome and only affect COVID-19 via the relevant exposure. The genetic variants for nitric oxide are well-established functionally relevant variants from a protein coding gene (NOS3). 14 The variant for platelet reactivity was partly selected on functional grounds. 10 The sensitivity analysis for platelet count did not give any indication of pleiotropy (Supplementary Table 3). Confounding by population strati cation could create bias, although genetic associations for both the exposures (nitric oxide and platelets) 10,13,15 and the outcomes, COVID-19, largely concern people of European ancestry (https://www.covid19hg.org/results/). Mendelian randomization studies can be open to selection bias, particularly if the genetic variants affect survival and other causes of survival and the outcome, here COVID-19, exist. 26 However, whether nitric oxide or platelets affect survival has not been clearly established. The underlying GWAS for COVID-19 obtained the non-cases from the same population as the cases (https://www.covid19hg.org/results/) which also reduces any possible selection bias, as does comparing with the general population to avoid index event bias. 27 Buffering of genetic effects could occur during development, i.e., canalization, however prior adaption to COVID-19 is unlikely. Findings, largely in Europeans, may not apply to other populations. Causes would be expected to act consistently, although may be less relevant in different settings. 28 We were not able to assess vulnerability to severe COVID-19 by sex, because no sex-speci c GWAS exists. Exploration of differences by sex could be important, given response to viral infections, including COVID-19, differs by sex. 29

Conclusions
This study suggests that the haemostatic factors, nitric oxide and a platelet phenotype, may underlie some of the thrombotic features of severe COVID-19. Investigation of the role in COVID-19 of wellestablished, widely available essential medicines targeting nitric oxide and platelets, such as nitrogylcerin, aspirin and statins, might be worthwhile.  Mendelian randomization estimates of nitric oxide (per allele higher*) based on rs2070744, rs1799983 and rs3918226, platelet reactivity (per square root transform) based on rs10886430, and platelet count (per effect size) based on 243+ variants on severe, hospitalized and any COVID-19 (*For rs1799983 carriers of the nitric oxide lowering allele have nitric oxide lower by -0.27 mol/L 30).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.