Increased soluble urokinase plasminogen activator levels modulate monocyte function to promote atherosclerosis

People with kidney disease are disproportionately affected by atherosclerosis for unclear reasons. Soluble urokinase plasminogen activator receptor (suPAR) is an immune-derived mediator of kidney disease, levels of which are strongly associated with cardiovascular outcomes. We assessed suPAR’s pathogenic involvement in atherosclerosis using epidemiologic, genetic, and experimental approaches. We found serum suPAR levels to be predictive of coronary artery calcification and cardiovascular events in 5,406 participants without known coronary disease. In a genome-wide association meta-analysis including over 25,000 individuals, we identified a missense variant in the plasminogen activator, urokinase receptor (PLAUR) gene (rs4760), confirmed experimentally to lead to higher suPAR levels. Mendelian randomization analysis in the UK Biobank using rs4760 indicated a causal association between genetically predicted suPAR levels and atherosclerotic phenotypes. In an experimental model of atherosclerosis, proprotein convertase subtilisin/kexin–9 (Pcsk9) transfection in mice overexpressing suPAR (suPARTg) led to substantially increased atherosclerotic plaques with necrotic cores and macrophage infiltration compared with those in WT mice, despite similar cholesterol levels. Prior to induction of atherosclerosis, aortas of suPARTg mice excreted higher levels of CCL2 and had higher monocyte counts compared with WT aortas. Aortic and circulating suPARTg monocytes exhibited a proinflammatory profile and enhanced chemotaxis. These findings characterize suPAR as a pathogenic factor for atherosclerosis acting at least partially through modulation of monocyte function.


Introduction
People with chronic kidney disease (CKD) are disproportionately affected by cardiovascular disease (CVD), with two-thirds of patients having at least one form of CVD, atherosclerosis being the most common (1). Conversely, over 40% of patients with CVD have signs of kidney disease (1). The reasons for which a large proportion of patients with CKD have concomitant CVD are unknown. Little progress has been made in understanding the nontraditional contributors of CKD to CVD risk, and the development of therapies targeting purported mechanisms such as vitamin D deficiency, hyperuricemia, and hyperphosphatemia have largely been ineffective (2).
Study cohort characteristics. Overall, Multi-Ethnic Study of Atherosclerosis (MESA) participants in this analysis had a mean (SD) age of 63 (10) years, 48.5% were male, 51.5% were female, and 28%, 11%, 22%, and 38% were Black, Chinese American, Hispanic American, and White, respectively. The median baseline suPAR level was 2.5 (IQR 2.0, 3.1) ng/mL. With increasing suPAR categories, participants were older, consisted of a larger proportion of women, and had a history of smoking, diabetes, and hypertension (P < 0.001 for all) (Supplemental Table 1; supplemental material available online with this article; https:// doi.org/10.1172/JCI158788DS1). The median CAC for the cohort was 1.91 Agatston units (AU), and a total of 2,636 (48.8%) participants had a CAC greater than 0.
Association between baseline CAC scores and suPAR. Participants with CAC greater than 0 had significantly higher suPAR levels than those with CAC equal to 0 (2.56 ng/mL, IQR 2.05-3.23, compared with 2.34 ng/mL, IQR 1.89-2.95, respectively, P < 0.001). Baseline suPAR levels correlated modestly with baseline CAC (Spearman's rank r = 0.14, P < 0.001). After adjustment for demographics, cardiac risk factors, and laboratory data, higher baseline suPAR levels were significantly associated with a higher CAC score at baseline: for a 2-fold difference in suPAR levels, baseline CAC scores were higher on average by 28.7 AU (95% CI 8.0-49.5) (Supplemental Table 2).
Baseline suPAR and longitudinal CAC. The median time between baseline and initial follow-up measurement of CAC was 2.5 years. For all suPAR categories, the median CAC at followup was higher than baseline: participants with suPAR of less than 2.0 ng/ml had a 103% increase in CAC at follow-up compared with 229% for those with suPAR greater than 3.0 ng/ml ( Figure 1 and Supplemental Figure 1). In multivariable analysis, higher baseline suPAR levels were associated with a greater increase in CAC scores over time (Supplemental Table 3), with a yearly increase in CAC score of 15.0 AU (95%CI 6.6-23.4) per 2-fold higher suPAR levels.
suPAR also appears to have an important role in CVD. Its levels are tightly correlated with the most prominent risk factors for atherosclerosis: smoking, diabetes mellitus, and CKD (14,22,23). suPAR levels are elevated within atherosclerotic plaque and correlate with intraplaque proinflammatory cytokines, including monocyte chemoattractant protein 1, TNF-α, IL-1β, and IL-66 (24). Patients with peripheral arterial disease have very high plasma suPAR levels that correlate positively with the number of clinically significant atherosclerotic sites and are predictive of vascular events (25). suPAR levels have been consistently associated with incident CVD and poor outcomes in various patient groups, including critically ill patients, those with HIV, cancer, and kidney disease, as well as the general population (24,(26)(27)(28)(29)(30)(31)(32)(33)(34). The association between suPAR and cardiovascular outcomes is independent of its impact on the kidneys, as even patients with end-stage renal disease can be risk stratified using suPAR (35).
Thus, suPAR is an immune-derived pathogenic factor for kidney disease and a strong biomarker of CVD, notably atherosclerosis. Here, we used epidemiologic, genetic, and experimental approaches to assess whether suPAR is causally involved in atherosclerosis. We first examined the relationship between suPAR levels and coronary artery calcifications (CAC) -a surrogate of atherosclerosis -and CVD events in 5,406 participants without known preexisting CVD. We sought genetic evidence of a causal role for suPAR in CVD by performing a genome-wide association study (GWAS) meta-analysis for suPAR levels in over 24,000 individuals, confirming experimentally a missense variant that specifically led to higher suPAR levels and using Mendelian randomization (MR) and rare variant association to leverage the genotypes and disease phenotypes in 500,000 participants of the UK Biobank (https://www.ukbiobank.ac.uk/). Finally, using a well-established murine model of atherosclerosis, we assessed whether mice overexpressing suPAR were more prone to atherosclerotic disease compared with WT mice and examined the impact of suPAR on monocyte profile and function.

Results
suPAR levels associated cross-sectionally with CAC and predicted CAC progression in the Multi-Ethnic Study of Atherosclerosis J Clin Invest. 2022;132(24):e158788 https://doi.org/10.1172/JCI158788 The regional association plots for all 8 loci from the European ancestry meta-analysis are illustrated in Supplemental Figure 7. The variance of suPAR levels explained by a weighted genetic risk score of all independent variants was 3%. Using sequential conditional analysis, 3 of the top 6 variants at the PLAUR locus (rs4760, rs2302524, and rs36229204) remained independent (Supplemental Figure  8). Bayesian fine mapping of the PLAUR locus resulted in 5 credible sets with both missense variants, rs4760 and rs2302524, capturing the top 2 of the 5 sets with very high posterior inclusion probabilities (Supplemental Table 8). The other 3 sets were captured by 3 sentinel variants, rs4251824, rs117564136, and rs400058.

Impact of PLAUR missense variants on suPAR levels in vitro and in vivo
To support the GWAS findings at the PLAUR locus, we assessed experimentally whether the PLAUR missense variants rs2302524 and rs4760 led to altered suPAR levels compared with the reference allele. We transfected HEK293 cells with plasmid DNA encoding either reference cDNA or the missense variants and measured suPAR levels in the cell media 48 hours later. The supernatant of cells transfected with the rs4760 variant had higher suPAR levels compared with reference, while we observed no increase in suPAR in the medium of cells transfected with rs2302524 ( Figure 3A) compared with reference. We did not find significant differences in PLAUR gene expression or differing patterns of cellular distribution of uPAR on immunostaining among the reference and the rs4760 and rs2302524 variants in HEK293 cells (Supplemental Figure 9), suggesting that the increase in suPAR levels is caused by increased secretion and not mediated by an increase in expression or cellular redistribution.
Expression of rs4760 (p.317Pro) in vivo using mouse hydrodynamic tail-vein injection of plasmid DNA similarly demonstrated a significant increase in serum suPAR levels 24 hours after injection (Supplemental Figure 10), while rs2302524 had no significant difference compared with the reference sequence ( Figure 3B). These findings confirm that the rs4760 variant, but not rs2302524, has a significant impact on suPAR levels, likely through increased secretion.
Genetically predicted suPAR level and atherosclerotic disease in the UK Biobank To assess whether suPAR levels are causally linked to CVD, we performed MR using the experimentally validated PLAUR rs4760 missense variant and the following cardiovascular phenotypes: aortic valve stenosis, atrial fibrillation, coronary artery disease, heart failure, hypertension, intracerebral hemorrhage, ischemic stroke, myocardial infarction, peripheral artery disease, pulmonary embolism, stroke, subarachnoid hemorrhage, and venous thromboembolism. We found that a genetically predicted 1 SD increment in suPAR was specifically associated with atherosclerotic phenotypes: 55% higher odds of coronary artery disease (P adjusted = 0.0002), 75% higher and Supplemental Tables 5 and 6) and did not differ according to the presence (CAC > 0) or absence of CAC (CAC = 0) at baseline (P interaction = 0.31) or baseline eGFR (P interaction = 0.98).

GWAS meta-analysis of suPAR
We performed a multi-ancestry GWAS meta-analysis of suPAR levels (Supplemental Figure 4) on 16.6 million variants in 12,937 participants of 4 cohort studies with European (n = 9,869), African (n = 1,363), East Asian (n = 623), and Hispanic (n = 1,082) ancestries. Fifteen independent signals in 8 loci were associated with suPAR levels at a genome-wide significance level (P < 5 × 10 -8 ) ( Table 1 and Supplemental Figure 5). A meta-analysis limited to the European ancestry sample included 9.9 million variants and identified 12 independent signals in 8 loci at genome-wide significance (Supplemental Figure 5 and Supplemental Table 7).
The strongest signals in both GWAS analyses were in or near the genes encoding suPAR (PLAUR) and its canonical ligand uPA (PLAU). There were 6 independent signals at the PLAUR locus, and 2 of them included missense variants in the PLAUR gene (Supplemental Figure 6): rs2302524 (p.Lys220Arg in domain III of uPAR), with each minor C allele associated with a 0.21 SD increase in suPAR levels (P = 1 × 10 -35 ); and rs4760 (p.Leu317Pro in the C-terminal portion of the proprotein uPAR form), with each minor G allele associated with a 0.11 SD increase in suPAR levels (P = 8 × 10 -9 ). Four other putatively independent signals were tagged by top SNPs in the noncoding sequence of the PLAUR locus (Table 1). At the PLAU locus, the A allele of rs2633321 was associated with higher levels of suPAR (β = 0.10 SD, P = 6 × 10 -15 ). Associations between suPAR levels and 12 of the 15 signals, including the 2 PLAUR missense variants, were replicated in the DBDS cohort (36) (n = 12,177) ( Table 1 and Supplemental Table 7). function of variants of PLAUR are protective against ischemic heart disease ( Figure 4B).

suPAR overexpression exacerbates atherosclerosis in a murine model
We then sought to determine whether experimentally raising levels of suPAR would exacerbate atherosclerosis. We induced atherosclerosis using proprotein convertase subtilisin/kexin-9-adeno-associated virus (Pcsk9-AAV) transfection in transgenic mice overexpressing full-length suPAR (suPAR Tg ) and WT C57BL/6J mice. The suPAR Tg mice had total cholesterol levels similar to those of WT mice at baseline and after D377Y-m Pcsk9 overexpression coupled with Western diet feeding for 10 weeks (Supplemental Figure 14). suPAR levels were significantly higher in suPAR Tg mice compared with WT at baseline (2.4 μg/mL versus 0.005 μg/mL, respectively) and at 10 weeks (25.2 μg/mL versus 0.01 μg/mL, respectively; Supplemental Figure 14).
All suPAR Tg (n = 21) mice developed larger plaques in the aortic root compared with the WT group (n = 18), with a mean plaque volume of 1.55 mm 3 in the suPAR Tg and 0.90 mm 3 in the WT group ( Figure 5). Atherosclerotic plaques of the suPAR Tg mice had significantly increased necrotic core areas compared with WT mice, with a mean volume of 0.18 mm 3 compared with 0.05 mm 3 , respectively. Observations were consistent when female and male mice were analyzed separately (Supplemental Figure 15). Furthermore, the atherosclerotic plaques of the suPAR Tg mice had a significantly higher percentage of macrophage-positive areas by Mac2 staining of 47.3% on average compared with 27.6% in the WT mice ( Figure 5). suPAR was detectable in both WT and suPAR Tg atherosclerotic plaques, with increased deposition in suPAR Tg plaques compared with those in WT (Supplemental Figure 16).

Elevated suPAR levels induce proatherogenic changes in monocyte profiles
Given the urokinase receptor system's known role in the regulation of innate immune system physiology, notably efferocytosis, we sought to assess whether suPAR overexpression altered the profile and function of monocytes and macrophages. We first examined aortas isolated from WT and suPAR Tg mice that did not undergo Pcsk9-AAV transfection to avoid the confounding effects of atherosclerosis and hyperlipidemia. We found that nonatherosclerotic suPAR Tg aortas secreted significantly higher levels of C-C motif chemokine ligand 2 (CCL2), one of the primary monocyte chemoattractants implicated in atherosclerosis, compared with WT aortas ( Figure 6A) (40). Flow cytometry of aortic cell suspensions revealed a 2-fold higher count of monocytes in suPAR Tg aortas compared with WT ( Figure 6B). The suPAR Tg monocytes isolated from aortas exhibited higher expression of C-C chemokine receptor type 2 (CCR2), the receptor for CCL2, compared with WT monocytes ( Figure 6B). Circulating monocytes and bone marrow-derived macrophages exhibited a similarly proinflammatory phenotype, with higher expression of CCR2 and lower expression of major histocompatibility complex class 2 (MHCII) and membrane-bound odds of myocardial infarction (P adjusted = 0.0002), and 71% higher odds of peripheral arterial disease (P adjusted = 0.03) after adjusting for multiple comparisons ( Figure 4A). Associations with coronary artery disease and peripheral arterial disease were replicated in independent cohorts (CARDIoGRAM C4D, ref. 37; and the Million Veterans Program, ref. 38) (Supplemental Figure 11). We did not observe an association between rs2302524 variant suPAR and any of the cardiovascular phenotypes (Supplemental Figure 12).
We also found that higher suPAR levels predicted by rs4760 were associated with lower creatinine-derived glomerular filtration rate (1% decrease per 1 SD higher suPAR; P = 0.001) and increased risk for CKD (OR = 1.24 per 1 SD higher suPAR, P = 0.02) in the UK Biobank and CKDGen (39) consortium, supporting the hypothesis of suPAR being a common pathogenic factor between cardiovascular and kidney disease (Supplemental Figure 13).
To support the findings in MR, we performed a collapsing analysis of rare variants in PLAUR. Our hypothesis was that rare damaging variants in PLAUR would lead to reduced plasma suPAR levels and be associated with a reduced risk for the cardiovascular phenotypes implicated by MR. We examined the more than 280,000 exomes in the UK Biobank and found that individuals with rare nonbenign coding variants in PLAUR had a lower risk of ischemic heart disease. Aggregate burden of rare damaging coding variants was associated with 41% lower odds of ischemic heart disease (95% CI 7%-63%), suggesting that heterozygous loss of

Figure 2. Cumulative incidence of any CVD event by suPAR categories. Unadjusted
Kaplan-Meier curves for the cumulative incidence of CVD events stratified by suPAR categories: 0-2.0 ng/mL (red), 2.0-2.5 ng/mL (green), 2.5-3.0 ng/mL (blue), >3 ng/mL (purple). The difference in cumulative incidence curves between suPAR categories was tested using the log-rank test. A CVD event was defined as the composite of myocardial infarction, resuscitated cardiac arrest, angina, revascularization, stroke (excluding transient ischemic attack), and death due to CVD. J Clin Invest. 2022;132(24):e158788 https://doi.org/10.1172/JCI158788 Monocytes from suPAR Tg mice exhibited a proatherogenic profile and altered function even prior to the induction of atherosclerosis. Overall, chronically elevated suPAR levels appear to promote atherosclerosis at least partially through priming the immune system to a dysregulated response. These findings dovetail extensive experimental and clinical data on suPAR's role in kidney disease and place high suPAR levels as a shared risk factor and potential therapeutic target for CVD and CKD.
Systemic inflammation is recognized as a key process common to CVD and CKD, with suPAR levels traditionally perceived as biomarkers of chronic inflammation related to activation of the innate immune system (7,43,44). suPAR levels are induced by shared risk factors for CKD and CVD, such as smoking, hypertension, and diabetes mellitus (14,19,22), associated with coronary and peripheral atherosclerotic disease (23,25,28,(45)(46)(47) and are predictive of incident kidney disease and CVD outcomes across age, sex, race, and clinical settings, independently of the aforementioned risk factors (11, 12, 14-16, 18, 19, 24, 26-34). To determine whether high suPAR levels precede CVD, we leveraged MESA -a cohort in which clinical CVD was an exclusion criterion at enrollment -and found that high suPAR levels at baseline predicted accelerated atherosclerosis as measured by CAC and incident CVD events even in participants with CAC equal to 0 and normal kidney function. Other biomarkers of inflammation have not exhibited a similar relationship with CAC (48,49), which has prompted us to further explore suPAR's singular role in atherosclerosis -now supported by our genetic and experimental analyses.
GWAS have revealed connections between common genetic variants and the risk for complex disease traits and quantitative traits such as plasma protein concentrations (50). These genetic variants, which are inherited independently of other disease risk modifiers, can be used in MR studies to determine whether a specific protein plays a causal role in a complex disease (51). We report a single instrument MR (rs4760) that supports a causal role for suPAR in the pathogenesis of atherosclerosis. MR studies rely uPAR ( Figure 6C and Supplemental Figure 17). Circulating monocytes from suPAR Tg mice also exhibited increased expression of C-X3-C motif chemokine receptor 1 (CX3CR1), another chemokine receptor that has been implicated in atherosclerosis (41), compared with WT ( Figure 6C).
We next assessed whether monocyte chemotaxis in suPAR Tg is altered as measured by migratory potential using a Transwell assay. Significantly more suPAR Tg monocytes migrated through the Transwell membrane compared with WT monocytes in response to both basal cell culture media and cell culture media with added recombinant CCL2 ( Figure 6D). Overall, these data indicate that suPAR acts on monocytes and myeloid cells in general to render these cells more atherogenic (40)(41)(42).

Discussion
We report epidemiologic, genetic, and experimental evidence of a causal role for suPAR in atherosclerosis. In a multi-ethnic cohort of over 5,000 participants without known CVD, we found high suPAR levels to be strongly associated with incident CVD and accelerated atherosclerosis as measured by serial CAC scores independently of decline in kidney function and established risk factors. In genetic analyses, we identified 2 independent common missense variants in PLAUR associated with higher plasma suPAR levels. One variant (rs4760, p.Leu317Pro) was confirmed experimentally in vitro and in vivo to lead to higher suPAR levels. Using that variant as an instrument in MR, we found that increased suPAR levels were causally linked to atherosclerotic phenotypes in the UK Biobank, notably coronary artery disease, myocardial infarction, and peripheral arterial disease in addition to kidney disease. Conversely, rare, damaging variants of PLAUR were associated with lower risk of ischemic disease. Experimentally, overexpression of suPAR in a murine model of atherosclerosis using Pcsk9-AAV led to a 2-fold increase in atherosclerotic plaque size with large necrotic cores and macrophage infiltration in suPAR Tg mice compared with WT mice. on the assumption that the instruments (SNPs) used in the analysis are a genetic proxy for only one action, namely, altered levels of the protein being tested. This assumption is very likely to be valid with rs4760. This SNP is located only in the proprotein form of uPAR and results in higher levels of full-length (DI-DII-DIII) reference sequence suPAR in circulation. Thus, even though rs4760 may be associated with other traits, these are likely mediated by altered suPAR levels and not through pleiotropic effects on genes other than PLAUR. The heterogeneity that we noted between rs2302524 and rs4760 may relate to the functional consequences of the missense variant on suPAR in the pathogenesis of atherosclerosis. Although the rs2302524 variant was the top signal in GWAS, it did not lead to an increase in levels when expressed experimentally and was not found to be linked to CVD phenotypes in a previous study (52). The resulting amino acid change encoded by rs2302524 (p.Lys220Arg) is located in the DIII domain of suPAR and is associated with levels of variant suPAR in humans. Query of the Genotype-Tissue Expression (GTEx) database (https:// gtexportal.org/home/gene/PLAUR) for PLAUR expression in human tissues revealed that the rs2302524 C allele is associated with lower PLAUR gene expression and that this association is the opposite of that observed with plasma suPAR levels. According to GTex, the rs2302524 is a splicing quantitative trait locus associated with higher levels of alternative splicing of PLAUR transcripts, resulting in the expression of different suPAR isoforms. The lack of colocalization of this variant and the cardiovascular phenotypes could be due to impaired function of the p.Lys220Arg suPAR secondary to the missense variant or due to altered circulating isoforms of suPAR. Conversely, the rs4760 variant is associated with increased plasma suPAR levels without altering the structure of the circulating protein, as the p.Leu317Pro variant is located only in a proprotein form of uPAR, suggesting that full-length (DI-DII-DIII) reference suPAR is the pathogenic form.
To confirm whether high levels of full-length suPAR accelerate atherosclerosis, we used a murine Pcsk9-AAV model of induced atherosclerosis, which allows for the study of immunometabolic processes without the confounding effects of germline alterations seen with the apoE knockout and LDL receptor knockout models (53). We found that overexpression of suPAR led to a 2-fold increase in total atherosclerotic plaque size, a 3.5fold increase in necrotic core size, and a 2-fold increase in lesional macrophage infiltration in suPAR Tg compared with WT mice, without differences in cholesterol levels. Given the urokinase receptor system's known role in modulation of immune cell motility and efferocytosis (3-9, 54, 55), we sought to determine whether chronically elevated suPAR levels affect monocyte profile and function in suPAR Tg mice that did not undergo induction of atherosclerosis. We found that nonatherosclerotic aortas of suPAR Tg mice excreted substantially higher CCL2 levels and contained more monocytes compared with aortas from WT mice.
Circulating monocytes of suPAR Tg mice exhibited higher expression of CCR2 and CX3CR1. The CCL2/CCR2 and CX3CR1 pathways have important roles in orchestrating monocyte recruitment into the vessel wall by chemotaxis, which we have found to be enhanced in suPAR Tg monocytes compared with WT (40,41,56). Moreover, monocytes from suPAR Tg also had reduced uPAR and MHCII expression compared with monocytes from WT mice. A decrease in MHCII expression and subsequent ability to present antigens has recently been linked to atherosclerosis through impairment of regulatory T cell activation (42), while a reduction in cell membrane-bound uPAR expression inhibits the "don't eat me" signal, resulting in enhanced phagocytosis and efferocytosis (54,57,58). Conversely, macrophages derived from uPAR-knockout mice have impaired phagocytic ability (50). Overall, these data suggest that chronically elevated suPAR levels prime myeloid cells to be more atherogenic, leading to accelerated atherosclerosis in the setting of additional injurious stimuli (such as hyperlipidemia in this case). suPAR's role in atherosclerosis may also be related to its binding of integrins (59). Integrins, notably α v β 3 , are crucial in initiation of atherosclerosis in endothelial cells and promote inflammation through the NF-κB pathway (59)(60)(61). Activation of integrins can also facilitate immune cell homing to the aorta and vascular remodeling (62). Future research will determine whether suPAR drives atherosclerosis through other mechanisms.
Overall, the greatest strength of this study is the multipronged approach to identifying a role for suPAR in atherosclerosis, leveraging epidemiologic and genetic analysis of large, well-characterized cohorts, and using an experimental murine model of atherosclerosis that does not involve germline alterations. We acknowledge certain limitations. The use of CAC as a surrogate for atherosclerosis may have led to an underestimation of the strength of its association with suPAR in MESA, given CAC scoring does not identify noncalcified plaque or coronary stenoses. Our genetic findings are disparate from a recent study that inferred genetic determinants of suPAR levels measured using proteomics platforms (63). The correlation between suPAR levels measured with proteomics platforms and immunoassays is, however, poor (r = 0.2-0.5), and their associations with outcomes vary greatly (64-68), explaining why our results differed from those obtained using proteomics platforms. Our GWAS analysis encompasses over 24,000 individuals in whom suPAR levels were measured using the suPARnostic (ViroGates) immunoassay used in the seminal studies on suPAR, kidney disease, and CVD outcomes (14,18,28,34,67). While our analysis included participants of mostly European ethnicity, a smaller GWAS in Black individuals also identified rs4760 as a determinant of suPAR levels measured using immunoassay (69). Finally, our experimental approach relies on the use of murine models of atherosclerosis, which cannot recapitulate all the features of the human disease, with major differences in lipoprotein metabolism and bile acid absorption (70). Nevertheless, these models are commonly used and have provided valuable insights into the pathophysiology of atherosclerosis (53).
Our findings may have important implications. suPAR's role in modulating the inflammatory profile and function of myeloid cells likely extends beyond atherosclerosis and may represent a common mechanism underlying suPAR's role as a predisposing factor in other chronic diseases, such as kidney, rheumatologic, and inflammatory bowel diseases. Targeting inflammation as a strategy for decreasing the risk of CVD has been shown to be viable in recent trials using monoclonal antibodies to IL-1β and IL-6 (40, 71, 72). suPAR has been targeted successfully in experimental models; bone marrow ablation (10), monoclonal antibodies directed to suPAR (8,18), or small molecule inhibitors of suPAR can prevent or reverse kidney injury (21). In patients with focal segmental glomerulosclerosis, plasmapheresis reduces suPAR levels, decreases β 3 integrin activity, and stabilizes the disease (73)(74)(75). The aggregate of epidemiologic, genetic, and experimental evidence we provide and the advent of anti-suPAR therapies strongly support exploring suPAR as a target for the prevention and treatment of CVD.
Methods suPAR, CAC, and incident cardiovascular events Study cohort. MESA is a multicenter observational cohort designed to identify risk factors for the incidence and progression of CVD. A detailed Measurement of suPAR. suPAR was measured using a commercially available ELISA (suPARnostic, ViroGates) in serum samples. The lower limit of detection of the assay is 100 pg/mL; however, all measurements were above the lower limit of detection. The interassay coefficient of variation determined using blinded replicate samples from participants ranged from 8% to 11%, depending on the cohort. suPAR description of the study design and methods has been published previously (76). In summary, 6,814 (3,601 women; 3,213 men) participants aged 45 to 84 years who identified as either White, Black, Hispanic, or Chinese were enrolled between 2000 and 2002 at 6 participating communities across the US. Participants were eligible if they were free of clinical CVD at enrollment. For the present study, we included all participants who provided serum samples for suPAR biomarker measurements at enrollment (n = 5,406).
Measurement of CAC. A detailed description of the methodology for the acquisition and interpretation of CAC scores in MESA has been published previously (77). Briefly, CT scanning of the chest was performed using either electron-beam CT (Chicago, Los Angeles, and New York field centers) or using a multidetector CT system (Baltimore, Maryland, USA; Forsyth County, Georgia, USA; and St. Paul, Minnesota, USA, field centers). CAC scores were calculated using the Agatston and CAC scores at baseline using Spearman's rank. To determine whether suPAR levels (log-transformed base 2) were independently associated with CAC at baseline, we used linear regression with CAC as the dependent variable adjusted for CVD risk factors including age, sex, race, BMI, history of smoking, eGFR using the Chronic Kidney Disease levels are stable in stored serum samples, with levels reproducible in samples stored for over 5 years at -80°C (79).
suPAR and CAC. Clinical characteristics for the cohort are reported stratified by suPAR categories (0-2.0 ng/mL, 2.0-2.5 ng/mL, 2.5-3.0 ng/mL, and > 3.0 ng/mL). We examined the correlation between suPAR Figure 6. suPAR overexpression in mice leads to proatherosclerotic phenotype in circulating and aortic monocytes. Aortas and blood were harvested from disease-free C57BL/6 WT and suPAR overexpressing mice (suPAR Tg mice). (A) Aortas from WT (n = 11) and suPAR Tg (n = 11) mice were excised, cleaned of fat, and cultured for 24 hours. At this point, the conditioned culture medium was isolated and CCL2 level was assessed by ELISA. (B) Aortas from WT (n = 6) and suPAR Tg (n = 6) mice were isolated, cleaned of fat, digested, stained with fluorescently labeled antibodies, and analyzed by flow cytometry. Quantification of F4/80-Ly-6G-CD11b + monocytes from WT and suPAR Tg mice as a percentage of live CD11b + CD45 + cells and median fluorescent intensity (MFI) of CCR2 expression from WT on F4/80-Ly-6G-CD11b + monocytes. (C) Blood from WT and suPAR Tg mice was isolated and red blood cells were lysed, stained with fluorescently labeled antibodies, and analyzed by flow cytometry. MFI on live CD45 + CD11b + monocytes for expression of CCR2, MHCII, and CX3CR1, and percentage of uPAR + cells of live CD45 + CD11b + cells. CCR2: n = 16 WT and n = 15 suPAR Tg , compared by Student's t test. MHCII: n = 6 WT and n = 6 suPAR Tg . CX3CR1: n = 4 WT and n = 4 suPAR Tg . uPAR + cells: n = 6 WT and n = 5 suPAR Tg . For MHCII, CX3CR1, and uPAR + cells, Mann-Whitney U test was used. (D) Monocytes were isolated from spleens of WT and suPAR Tg mice and cultured in Transwell assays with either control cell culture media or cell culture media with CCL2 added. Quantification of fluorescent intensity of cellular dye was compared by 2-way ANOVA. n = 6 for each group. Each data point represents a biological replicate. suPAR levels. We then leveraged the UK Biobank to perform MR and assess for a causal link between genetically determined suPAR levels and CVD (n = 408,894) (85).
Quality control measures were performed to exclude low-quality samples and low-quality variants within each study prior to imputation to reference genomes. In general, samples were excluded if they showed discordance between genetically inferred and reported sex, low call rate, and duplications. Variants were excluded if they deviated from the Hardy-Weinberg equilibrium.
Imputation was done to predict nongenotyped variants. The TSS, GABC, and MESA were imputed using TOPMed Freeze 5b (GRCh 38). The MDCS was imputed using the Haplotype Reference Consortium reference panel (GRCh 37) (86). The build was liftover to GRCh 38 using Cross-Map (87). The DBDS was imputed using 1 KG phase 3, HapMap, and a data set consisting of more than 6,000 Danish whole-genome sequences.
GWAS analyses. GWAS analyses were performed with natural log suPAR levels adjusted for age, sex, and the first 10 principal components of ancestry followed by inverse-normal transformation within each study and ancestry combination using array data imputed to reference genomes. Single-variant association analyses were performed using linear regression in PLINK, version 2.0 (88), within each study-ancestry combination. For GABC, linear mixed models incorporating a kinship matrix were performed using RVTESTS (89). Overall, our analyses resulted in genome-wide summary data from European ancestry data sets from MDC (n = 4,735), TSS (n = 2,179), MESA (n = 2,024), and GABC (n = 931), and African (n = 1,363), East Asian (n = 623) and Hispanic (n = 1,082) populations from MESA. We performed quality control measures on each of the summary association data sets prior to meta-analysis (90,91). Within each data set, we filtered out variants with minor allele count of less than 20, Hardy-Weinberg equilibrium P value of less than 5 × 10 -6 , low imputation quality (INFO < 0.6), multiallelic variants, and palindromic variants (A/T or C/G) with minor allele frequency above 0.4.
Meta-analysis. We performed multi-ancestry and European ancestry-specific inverse-variance weighted fixed effects meta-analyses using METAL software (90). We generated quantile-quantile plots to assess for genomic control and structure within our data (Supplemental Figure  18). To identify leading and independent variants from each meta-analysis, we performed pruning and thresholding using the "clump" flag in PLINK. PLINK implements an iterative multistep process in which variants are sorted by their P values and those in linkage disequilibrium are removed (r 2 < 0.05 and within 250 kilobases from the lead variant). The process is repeated until the genome-wide significance threshold of 5 × 10 -8 is reached. The PLAUR locus was further finemapped using the SuSie Iterative Bayesian Stepwise Selection procedure (92). Top variants Epidemiology Collaboration equation (80), LDL levels, HDL levels, C-reactive protein, hypertension (use of antihypertensives or systolic blood pressure ≥140/90 at enrollment), and diabetes mellitus. We then visualized the median CAC scores at baseline and initial follow-up stratified by suPAR categories using bar graphs. Additionally, we examined the adjusted difference in CAC scores between baseline and initial follow-up by calculating the mean predicted change in CAC score for each suPAR category accounting for age, sex, race, BMI, history of smoking, eGFR, LDL levels, HDL levels, C-reactive protein, and diabetes mellitus.
To determine whether suPAR levels at baseline were associated with an increase in CAC over time, we generated generalized estimating equations modeling with CAC as a continuous and longitudinal variable using all CAC scores measured after baseline and examined the interaction term suPAR × follow-up time. The model was adjusted for the aforementioned variables in addition to baseline CAC.
suPAR and cardiovascular events. We then assessed whether suPAR levels were predictive of CVD events. A CVD event was defined in MESA as the composite of myocardial infarction, resuscitated cardiac arrest, angina, revascularization, stroke (excluding transient ischemic attack), or death due to CVD (76,77). We used stepwise multivariable-adjusted Cox's proportional hazards modeling to assess the contribution of relevant factors such as eGFR and CAC to the association between suPAR and CVD events. Model 0 (suPAR alone) was unadjusted; model 1 was adjusted for age, sex, race, BMI, history of smoking, LDL, HDL, C-reactive protein, hypertension, and diabetes mellitus; model 2 included all variables in model 1 in addition to baseline eGFR; and model 3 included the variables in model 2 with the addition of baseline CAC. We explored eGFR as a time-varying covariate in a separate model including the covariates from model 3. In MESA, eGFR was measured at baseline and at exam 5 (April 2010-February 2012). suPAR was modeled as a continuous (log-transformed base 2) and categorical variable (0-2.0 ng/mL, 2.0-2.5 ng/mL, 2.5-3.0 ng/mL, and > 3.0 ng/mL) in all models. Additionally, we conducted a sensitivity analysis, further adjusting for baseline high-sensitivity troponin T and NT-proBNP in addition to the variables in model 2. Follow-up time was up to the first CVD event, death, last contact with the research team, or end of study period. Unadjusted and adjusted Kaplan-Meier cumulative incidence curves for CVD events were generated. Adjusted Kaplan-Meier curves were calculated using inverse probability weighting for suPAR categories with propensity scores estimated using generalized boost models adjusted for age, sex, race, BMI, history of smoking, eGFR, LDL levels, HDL levels, C-reactive protein, and diabetes mellitus (81). A complete case analysis was performed. A 2-sided P value of less than 0.05 was used to determine statistical significance. Analyses were performed using R, version 4.1.0, (R Foundation for Statistical Computing).

Genetic determinants of suPAR and the link to atherosclerosis
We measured plasma suPAR levels using immunoassay (suPARnostic,-ViroGates) in 4 different cohorts: the Trinity Student Study (TSS) (82), the Genes and Blood-Clotting cohort (GABC) (83), MESA, and the Malmo Diet and Cancer Study (MDCS), totaling 12,937 participants (84). We performed GWAS and meta-analysis to identify genetic determinants of suPAR levels and replicated our findings in 12,177 healthy participants of the DBDS where suPAR levels were measured using the same immunoassay. The top 2 significantly associated missense variants of PLAUR were then expressed in human embryonic kidney cells (HEKs) (CRL-3216; ATCC) and in C57BL/6J mice (000664; Jackson Laboratory) to determine which variants led to significant increases in defined as those with a REVEL score of 0.25 or more and a maximum minor allele frequency of 0.0005 (96).
suPAR overexpression in a Pcsk9-AAV murine model of atherosclerosis A total of 39 mice, 12 to 16 weeks of age, including n = 18 C57BL/6J WT mice (000664, Jackson Laboratory), of which 7 were female, and n = 21 suPAR Tg mice, of which 4 were female, overexpressing the soluble form of mouse full-length suPAR (corresponding to NP_035243, DI-DII-DIII without GPI anchor) in adipose tissue using the adipocyte fatty acid binding protein (AP2) promoter on C57BL/6 background, were used (10). All mice were maintained on a 12-hour light/12-hour dark cycle with free access to food and water.
To induce hypercholesterolemia, we administered an i.p. injection of recombinant AAV8-D377Y-murine Pcsk9 (5 × 10 6 viral genomes/kg body weight), which was previously described (97). After 1 week, the diet was switched to a Western diet (42% calories from fat, Teklad, catalog 88137) for 10 weeks and all 39 mice completed the study.
Cholesterol and suPAR measurements. Plasma was collected via tailvein puncture in heparin-coated tubes. Fasting cholesterol levels were measured by colorimetric assay (STA-384; Cell Biolabs). Plasma levels of suPAR were measured using R&D DuoSet ELISA antibodies and Ancillary Reagent Kit 2 for development of a sandwich ELISA (DY531, R&D Systems). The ELISA has a detection range of 78 to 5000 pg/mL.
Atherosclerotic lesion analysis, histology, and immune histochemistry. Mice were euthanized via carbon dioxide overdose. Blood was harvested by right ventricular puncture and the vasculature perfused with icecold PBS. The heart and brachiocephalic artery (BCA) were harvested from all 39 mice, placed in 4% paraformaldehyde, and embedded in paraffin. Sixty sections (6 μm each) were cut through the aortic root as the primary site of atherosclerosis, and 30 sections (6 μm each) were cut through the BCA as a secondary anatomic site from each mouse, as recommended (98). For morphometric analysis, 30 sections from the aortic root and 15 sections from the BCA were stained with H&E and assessed for total lesion size and necrotic core size (acellular lesion area) as previously described (99), for a total coverage of 360 μm of the aortic root. Paraffin-embedded sections of the aortic sinus were deparaffinized and rehydrated. After blocking, sections (6 μm each) were incubated at room temperature for 2 hours with Mac2 (sc-81728; Santa Cruz Biotechnology Inc.). Mac2 slides were counterstained with hematoxylin and coverslipped. Images were captured with an Olympus LC30 camera mounted on an Olympus CX41 microscope. For the Mac2 + area, all images were obtained with the same light source at the same time. The Mac2 + area was determined using the threshold function in ImageJ (NIH) and normalized to total nonnecrotic lesion area. Results were reported as percentage of lesion area. Sectioning and staining were performed by the In Vivo Animal Core Laboratory technicians at the Unit for Laboratory Animal Medicine, University of Michigan. Technicians in this laboratory were blinded to experimental identity. Atherosclerotic plaque size was calculated using ImageJ software and graphed by section number.
Flow cytometry of aortic cell suspension and circulating cells. Approximately 50 to 100 μl whole blood was harvested via tail vein, and the were defined as those with a P value of less than 5 × 10 -8 and were independent of each other. We then investigated the identified variants in the DBDS cohort. Functional annotations for top variants were obtained from the Ensemble Variant Effect Predictor (91).
MR analysis. We leveraged the UK Biobank for MR analysis in 408,894 participants of European ancestry (UK Biobank resource, application number 59206) (93). Details of measures for variant and sample quality control have been previously reported (94). We used rs4760, the PLAUR missense variant confirmed to alter suPAR levels in both in vitro and in vivo models, as an instrument for MR analyses of 13 cardiovascular phenotypes from the UK Biobank (Supplemental Table 9). Significant associations were replicated using publicly available summary GWAS data from the CARDIoGRAM C4D consortium for coronary artery disease (60,801 cases and 123,504 controls) and the Million Veterans Program for peripheral arterial disease (31,307 cases and 211,753 controls) (37,38). Wald ratios were used to derive the odds ratio per 1 SD increments in suPAR levels instrumented by rs4760. Similar analyses were performed using the rs2302524 missense variant as an instrument. Finally, we obtained summary-level data from the CKDGen consortium to perform MR and assessed for a causal link between genetically determined suPAR levels by rs4760 and (a) kidney function as measured by creatinine-derived eGFR (n = 567,460) (39) and (b) CKD (41,395 cases, 439,303 controls), defined as an eGFR of less than 60 ml/min/1.73 m 2 (95). The MR was then replicated in the UK Biobank (eGFR, n = 387,937; CKD, 8,031 cases and 400,863 controls) (85). MR analyses were performed using the TwoSampleMR package in R, version 4.0.
To assess whether rare coding variations with damaging consequences on the suPAR protein are associated with ischemic heart disease, we performed a lookup in a previously published exomesequenced analysis of more than 280,000 UK Biobank participants (http://azphewas.com/). Both rare protein truncating variants and rare damaging missense variants in the PLAUR gene were selected for studying the impact of attenuated PLAUR function on coronary heart disease. In brief, protein-truncating variants are defined as variants that are predicted to truncate a protein and with a maximum minor allele frequency of 0.001. Rare damaging missense variants were