Genetic architecture and mechanisms shared between kidney and ureteral stones, cardiovascular diseases, and metabolic syndrome: A comprehensive GWAS analysis

Absctract

Background

Our study aims to investigate the shared genetic architecture between kidney and ureteral stones (KUS) and cardiovascular diseases (CVDs), as well as metabolic syndrome (MetS), and explore the shared risk loci, potentially critical tissues and relevant genetic mechanisms.

Methods

Dependent on large-scale genome-wide association study (GWAS) summary-level data sets, we observed genetic correlations between KUS and CVDs, as well as MetS, and cross-diseases pleiotropic analysis was conducted to identify shared pleiotropic loci and genes. Furthermore, we performed functional annotation and tissue-specific analysis to detect potential relationships between complex traits. We performed heritability enrichment analysis to determine potentially critical tissues. At last, we investigate the causal effects between KUS and other traits using bidirectional Mendelian randomization (MR).

Results

Our findings underlined shared genetic architecture between three CVDs, two MetS and KUS. We identified 937 pleiotropic loci at the genome-wide significance level (p < 5 × 10⁻⁸), 35 of which were annotated as genomic risk loci. Among them, 4 had strong evidence of colocalization (PP.H4 > 0.7). In addition, a total of 163 unique pleiotropic genes (pFDR <0.05) were recognized at the gene level, including FTO, NEK4, GNL3, GLT8D1, SMIM4, PBRM1 and TFAP2B. Pathway analysis illustrated the essential biological process including metabolic processes, transcriptional regulation processes, transmembrane transport of drugs, and cardiac structure development were involved in these diseases. Analysis of tissue enrichment at single nucleotide polymorphism (SNP) level and gene level indicated pleiotropic mechanisms may engage in prostate, pancreas, adipose subcutaneous, and muscle skeletal. HyPrColoc method and metabolite enrichment analysis revealed tryptophan metabolism might be a crucial shared metabolic pathway in two different diseases. At last, bidirectional MR analysis demonstrated no strong evidence of causal associations between KUS and CVDs, as well as MetS.

Conclusions

Our study determined shared genetic architecture between KUS and CVDs, as well as MetS, and unraveled underlying genetic mechanisms.

Highlights

• •Genetic CorrelationsGWAS analysis identified significant genetic correlations between kidney stones (KUS) and cardiovascular diseases (CVDs) and metabolic syndrome (MetS), including coronary atherosclerosis, angina pectoris, myocardial infarction, hypertension, and obesity.
• •Pleiotropic SNPs and Genes937 pleiotropic SNPs and 35 genomic risk loci were identified, with 4 loci showing strong colocalization (PP.H4 > 0.7). Key genes included FTO, NEK4, GNL3, GLT8D1, SMIM4, PBRM1, and TFAP2B.
• •Metabolite PathwaysTryptophan metabolism was identified as a key shared metabolic pathway between KUS and hypertension. Amino acids and lipid metabolism were highlighted as shared mechanisms.
• •Potential Drug TargetsGenes such as NEK4, GNL3, GLT8D1, PBRM1, and SMIM4 were identified as potential drug targets through SMR analysis.
• •ConclusionsThe study reveals shared genetic architecture between KUS, CVDs, and MetS, providing insights into potential therapeutic targets and shared metabolic pathways.

1. Introduction

KUS is one of the most common diseases of the urinary system caused by the formation of concretions within the pelvis and calyces of the kidney [1]. Due to its high prevalence and recurrence rate, KUS could cause chronic kidney damage in individuals and impose a considerable financial burden on the nation's healthcare system [2,3]. Cardiovascular diseases (CVDs) constitute a category of clinical conditions that affect the heart and the vascular system. Ischemic heart disease and stroke, both linked to atherosclerosis, are the leading causes of cardiovascular deaths, making up 84.9 % of such fatalities [4,5]. Researches have demonstrated that CVDs are closely related to metabolic syndrome (MetS) [6,7], a chronic condition characterized by a series of vascular risk factors such as dyslipidemia, hyperglycemia, obesity (OB), and hypertension [8]. KUS is progressively being recognized as a systemic condition, and growing evidence links kidney stone formation to CVDs and MetS [9]. While the exact reasons are unclear, possible factors include metabolic responses that encourage stone formation, environmental influences like diet, oxidative stress, inflammation, and molecular changes affecting urine analyte transport [[10], [11], [12]].

Epidemiological studies have demonstrated correlations between CVDs and MetS with the occurrence of KUS. In the context of cardiovascular diseases, a current meta-analysis has identified that patients with KUS have a 25 % increased risk of developing coronary artery disease (CAD) and a 17 % higher risk of experiencing stroke or transient ischemic attacks (TIA) [13]. Meanwhile, a recent systematic review has indicated that patients with hypertension and dyslipidemia exhibit a 61.3 % and 58.6 % increased risk of developing KUS respectively [14]. Patients diagnosed with diabetes mellitus and OB also exhibit an increased risk of developing KUS by 55.2 % and 53.1 % [14]. Over the past decades, MetS has long been seen as an effective predictor of coronary heart disease (CHD) and stroke. [15]. A systematic review and meta-analysis of longitudinal studies found that MetS is associated with a relative risk (RR) of 1.78 for cardiovascular events and mortality (95 % confidence interval: 1.58–2.00) [16]. Nevertheless, current study on the potential associations between stone formation and these systemic conditions predominantly concentrates on analogous metabolic reactions, dietary practices and lifestyle habits. This leaves a distinct gap in comprehending the shared pathophysiological and pleiotropic genetic mechanisms between KUS and CVDs, as well as MetS. Only Asokan Devarajan have reported the shared mechanism between CVDs and KUS, highlighting the crucial role of reactive oxygen species (ROS) in injuring the urothelium and vascular endothelium [10]. Moreover, the polarization of M1 macrophages and the differentiation of osteoblasts are also regarded as potential shared mechanisms linking CVDs and KUS [10]. There are considerable gaps existing in the field of genetics, emphasizing the urgent requirement to identify shared risk loci between KUS and these two systemic conditions. It is important to note that traditional clinical or epidemiological research may struggle to ensure statistical efficiency in these analyses.

Currently, methodologies such as linkage disequilibrium score regression (LDSC) and high-definition likelihood (HDL), which are grounded in genome-wide association studies (GWAS), have been designed to assess the presence of genetic correlations between KUS and CVDs, as well as MetS [17,18]. It's uncertain if the genetic correlation is due to a few loci or the whole genome. Up to now, few studies have systematically assessed the overlapping risk loci, shared pathogenic genes, and causal relationships between KUS and CVDs, as well as MetS. The cross-trait analyses of pleiotropic genes or loci through the examination of GWAS signal correlations have been demonstrated to effectively identify shared loci between different diseases [19,20]. These pleiotropic loci present opportunities for targeted interventions, which may facilitate the deeply understanding or effectively treatment of these diseases. A novel approach, named pleiotropic analysis under composite null hypothesis (PLACO), has been recently developed to identify pleiotropic genetic loci at the SNP level [21]. Consequently, identifying the unique genetic variants or loci that contribute to genome-wide genetic correlations is vital, as is exploring the overlapped genetic pathogenesis underlying these diseases.

2. Methods

2.1. GWAS summary data source

The GWAS summary data for six cardiovascular diseases are sourced from publicly available data with European ancestry: coronary atherosclerosis (CAS), ischemic stroke (IS), atrial fibrillation and flutter (AF/AFL), angina pectoris (AP), myocardial infarction (MI), and venous thromboembolism (VTE). Additionally, we obtained the GWAS summary data for four MetS traits from publicly available datasets of European ancestry: essential hypertension (HPT), obesity (OB), type 2 diabetes (T2D), and hyperlipidaemia (HL).

GWAS summary statistics for KUS were obtained from FinnGen study including a total of 11,650 cases and 441,039 controls of European ancestry. The relationship between KUS and SNP genotypes was evaluated through logistic regression, incorporating genetic principal components as covariates in the association study. The details of these datasets are summarized in Additional fle 2: Table S1.

2.2. Quality control

In this study, a series of rigorous quality control measures were implemented to ensure the accuracy and reliability of the GWAS data. Initially, SNPs located within the major histocompatibility complex (MHC) region, ranging from the 25 Mb–35 Mb interval on chromosome 6, were excluded. Due to its complex genetic architecture and high linkage disequilibrium, this region is often excluded in most GWAS researches to prevent the occurrence of false positive results. We retained only those SNPs with a minor allele frequency (MAF) greater than 0.01, indicating that the frequency of the minor allele exceeded 1 %. This filtering step ensures a primary focus on common variants, thereby enhancing statistical power and diminishing the probability of false positives.

Furthermore, we conducted additional quality control for both samples and markers, including filtering criteria based on sample call rate and marker call rate. For samples, a call rate exceeding 95 % is required, whereas for markers, the call rate must greater than 99 %. Samples and markers that did not satisfy these criteria were excluded to enhance the overall quality of the data and ensure the reliability of the analytical results.

2.3. Genetic association at the genome-wide level

To evaluate the shared genetic architecture between KUS and CVDs, as well as MetS, we employed the LDSC method [18]. The linkage disequilibrium (LD) scores in LDSC were estimated utilizing samples of European ancestry from the 1000 Genomes Project [22]. The leave-one-out method was employed to estimate the standard errors (SE) in LDSC, and it was subsequently utilized to correct for attenuation bias. The LDSC intercept was utilized to evaluate the potential overlaps of population across various studies. In our analysis, there was no population overlap among CVDs, MetS and KUS, which enhances the reliability of the results.

To further validate the LDSC results, we employed the HDL method, a likelihood-based analytical tool that more effectively use GWAS summary statistics to estimate genetic associations [17]. Compared to the LDSC method, HDL can decrease the variance of genetic association estimates by almost 60 %, thereby significantly enhancing the precision and robustness of the estimates [17]. Through validation with HDL, we ensured the credibility of genome-wide genetic overlap analysis.

2.4. Tissue-specific heritability enrichment

To examine the extent of association between CVDs, MetS traits and KUS in different tissues and organs, we further investigated the enrichment of SNP heritability in specific tissues using stratified-LDSC (S-LDSC). We obtained 54 human tissue datasets from the GTEx database to evaluate the significance of SNP heritability enrichment across various tissues [23].

2.5. Identifcation of pleiotropic SNPs and risk loci

We employed PLACO approach to systematically identify genetic associations at the SNP level between KUS and CVDs, as well as MetS. PLACO is a novel statistical method specifically developed to detect genetic pleiotropy, with the capability to identify shared genetic variations across multiple phenotypic traits [21]. With this method, we can effectively identify SNPs that present significant associations in multiple diseases and traits. Genome-wide significant (p < 5 × 10⁻⁸) SNPs were defined as pleiotropic variants. This indicated that these SNPs exhibit robust genetic associations across various phenotypes, suggesting an important role in the pathogenesis of these diseases. Identifying these pleiotropic variants was highly significant for elucidating the shared genetic architecture underlying various systemic diseases and KUS. Additionally, to further detect the biological significance of these pleiotropic SNPs, we performed the functional mapping and annotation (FUMA) to map the risk variants onto the genomic region [24]. Ultimately, we performed a Bayesian colocalization analysis to identify the pleiotropic risk loci predominantly shared between KUS and CVDs, as well as MetS [25].

2.6. Identifcation of pleiotropic genes

Initially, we mapped nearby genes based on the lead SNPs at each locus in order to investigate their shared mechanisms. In addition, we performed a Multi-Marker Analysis of Genomic Annotation (MAGMA) to elucidate the biological functions of these pleiotropic loci. Pleiotropic genes and multi-marker effects were identified by MAGMA gene analysis (pFDR<0.05) [26]. Furthermore, to explore the biological function of lead SNPs, MAGMA gene-set analysis was conducted based on c2 curated gene sets and go terms (c5.bp, c5.cc, and c5.mf) from Molecular Signatures Database (MSigDB) [27]. To avoid false positive results, all tested gene sets were corrected using Bonferroni correction. At last, based on 54 GTEx tissues, we performed genome-wide tissue-specific enrichment analysis on PLACO polygenic results [28], and calculated the average log2 expression of all pleiotropic genes in these tissues.

2.7. Exploration of potential gene targets by summary-based mendelian randomization

We utilized summary-based Mendelian randomization (SMR), a method integrating GWAS summary-level data with expression quantitative trait loci (eQTL) studies to identify genes whose expression is associated with complex traits [29]. eQTL are genetic variations that affect gene expression, causing individuals with different genotypes to have different expression levels [30]. Furthermore, the Heterogeneity in Dependent Instrument (HEIDI) method was used to test if the association is due to colocalization, meaning the same causal variant affected both SNPs' impact on gene expression and complex traits. Therefore, SMR method identifies genes associated complex traits and uncovers their regulatory mechanisms, offering crucial insights for target discovery.

2.8. Multi-trait colocalization analysis for plasma metabolites

We employed the hypothesis prioritization for multi-trait colocalization (HyPrColoc) method [31] to conduct a multi-trait colocalization analysis, aiming to elucidate the significant roles that plasma metabolites played in the pathogenesis of CVDs, MetS and KUS. The metabolite GWAS dataset comprises a total of 1400 plasma metabolites [32], which could be publicly available from the GWAS catalog (GCST 90199621 ∼ GCST 90201020). Notably, 528 metabolites classified as "Ratio" and "Unknown" have been excluded from this dataset. Then, we used MetaboAnalyst 6.0 (https://www.metaboanalyst.ca/) to conduct enrichment analysis based on the Small Molecule Pathway Database (SMPDB).

2.9. Causal association analysis

A bidirectional two-sample Mendelian Randomization (MR) analysis was performed to evaluate the potential causal relationship between KUS and CVDs, as well as MetS. We used clumping procedure of the PLINK 1.9 software to obtain independent significance SNPs for CVDs, MetS traits and KUS (p < 5 × 10⁻⁸), with r² set to 0.001 and a window size of 10,000 KB. To verify the causal relationships between these traits, we used five MR methods for each set of instrumental variables: Inverse Variance Weighted (IVW), Weighted Median, Wald Ratio, Simple Mode, and MR-Egger. Cochran's Q statistic tested effect size heterogeneity among instrumental variables, while the MR-Egger regression intercept detected horizontal pleiotropy.

3. Results

3.1. Shared genetic architecture between KUS and CVDs, as well as MetS

Initially, we performed LDSC and HDL methods to evaluate the genetic correlation between CVDs, MetS traits and KUS (Table 1). Using the LDSC method, six traits were found to be genetically correlated with KUS, including CAS, AF/AFL, AP, MI, HPT, and OB (p < 0.05). The results of HDL method indicated that KUS was genetically correlated with CAS, IS, AP, MI HPT, and OB (P < 0.05). However, AF/AFL was not observed genetically correlated with KUS in HDL. Consequently, we selected five traits that are significant in both methods for further analysis (see Fig. 1).

Table 1.

Genetic correlation between KUS and CVDs, as well as MetS.

Trait pairs	LDSC		HDL
	rg(SE)	P	rg(SE)	P
KUS & CAS	0.141 (0.046)	0.002	0.133 (0.040)	7.58 × 10−4
KUS & IS	0.038 (0.047)	0.426	0.192 (0.087)	0.027
KUS & AF/AFL	0.214 (0.059)	3 × 10−4	0.079 (0.049)	0.106
KUS & AP	0.228 (0.063)	3 × 10−4	0.223 (0.042)	1.40 × 10−7
KUS & MI	0.139 (0.057)	0.014	0.175 (0.053)	8.64 × 10−4
KUS & VTE	0.0494 (0.073)	0.496	0.088 (0.075)	0.242
KUS & HPT	0.135 (0.037)	3 × 10−4	0.182 (0.038)	1.71 × 10−6
KUS & OB	0.2260 (0.073)	0.002	0.204 (0.055)	2.26 × 10−4
KUS & T2D	0.082 (0.051)	0.107	0.075 (0.061)	0.223
KUS & HL	−0.168 (0.111)	0.130	0.015 (0.062)	0.812

LDSC, linkage disequilibrium score regression; HDL, high-definition likelihood; SE, standard error; KUS, kidney and ureteral stones; CAS, coronary atherosclerosis; IS, ischemic stroke; AF/AFL, atrial fibrillation and flutter; AP, angina pectoris; MI, myocardial infarction; VTE, venous thromboembolism; HPT, hypertension; OB, obesity; T2D, type 2 diabetes; HL, hyperlipidaemia.

Fig. 1.

The flow chart for the research design.

3.2. Tissue heritability enrichment analysis

We further used S-LDSC method to investigate whether the SNP heritability of each trait was enriched in specific tissues. S-LDSC was used on GWAS summary data from multiple tissues to evaluate the significance of genetic enrichment for specific traits. S-LDSC results indicated that SNP heritability for KUS exhibited significant enrichment in the kidney cortex, prostate and pancreas. On the other hand, SNP heritability for CVDs, including conditions such as CAS, AP and MI, was mainly enriched in the coronary or adipose tissues. For MetS, OB showed significant enrichment in pancreas. (Fig. 2 and Additional file 2: Table S2).

Fig. 2.

S-LDSC tissue heritability enrichment analysis. KUS, kidney and ureteral stones; CAS, coronary atherosclerosis; AP, angina pectoris; MI, myocardial infarction; HPT, hypertension; OB, obesity.

3.3. Pleiotropic SNPs and risk loci identified for CVDs, MetS and KUS

Given the genetic correlation between KUS and CAS, AP, MI, HPT, and OB, we employed an innovative pleiotropy analysis (PLACO) to identify potential pleiotropic SNPs associated with KUS and these traits. We totally identified 937 pleiotropic SNPs between KUS and CAS, AP, MI, HPT, and OB at the genome-wide significance level (p < 5 × 10⁻⁸) (Additional file 1: Fig. S1 and Additional file 2: Table S3). The QQ plots showed no early deviation between the observed and expected values, effectively eliminating the likelihood of group stratification (Additional file 1: Fig. S2). Among these pleiotropic SNPs obtained by PLACO, we further annotated 35 pleiotropic genomic risk loci associated with KUS and CAS, AP, MI, HPT, and OB using FUMA (Fig. 3 and Additional file 2: Table S4). Through colocalization analysis, 4 out of 35 potential pleiotropic risk loci (11.4 %) were ultimately identified that had a posterior probability of H4 (PP.H4) exceeding 0.7 (Table 2 and Additional file 1: Fig. S3∼S6), and 16q12.2 was the shared risk loci among KUS, HPT and OB. It was worth noting that several pleiotropic genomic risk regions were overlapped between multiple traits. For instance, 3p21.1, 17q23.2, and 20q13.2 were observed in four traits (Additional file 2: Table S5).

Fig. 3.

The circular diagram presents pleiotropic risk loci and nearby genes identified by FUMA between KUS and other five traits. Note: Colocalized loci identified by colocalization analysis (PP.H4 > 0.7) were highlighted in green; Pleiotropic genes identified by MAGMA analysis were highlighted in red. KUS, kidney and ureteral stones; CAS, coronary atherosclerosis; AP, angina pectoris; MI, myocardial infarction; HPT, hypertension; OB, obesity.

Table 2.

4 colocalized loci identified by colocalization analysis conducted on 35 pleiotropic loci (PP.H4 > 0.7).

Trait pairs	Locus boundary	Region	Nearest genes	Lead SNPs	p	PP.H4
KUS&HPT	16:20350459–20392332	16p12.3	UMOD, PDILT, SERPINB2	rs77924615	4.60 × 10−11	0.969
KUS&HPT	16:53797908–53845487	16q12.2	TUBB4B, FTO, ZNF480	rs11642015	1.92 × 10−13	0.968
KUS&OB	16:53797908–53848561	16q12.2	TUBB4B, FTO, ZNF480, RBL2	rs11642015	1.01 × 10−20	0.966
KUS&OB	16:46179043–46202172	19q13.32	VASP, GIPR, SNRPD2, QPCTL, FBXO46, DMWD, SYMPK	rs11672660	4.71 × 10−10	0.861

Lead SNPs represents the top SNPs (smallest P-values) in genomic loci. PP.H4 represents the posterior probability of H4 calculated by coloc analysis; the Locus boundary was defined as chromosome: start-end.

PP.H4, the posterior probability of H4, KUS, kidney and ureteral stones; HPT, hypertension; OB, obesity.

3.4. MAGMA gene-level enrichment analysis

Based on genome-wide pleiotropic results performed by PLACO, we conducted MAGMA gene analysis to identified significant pleiotropic genes between KUS and CAS, AP, MI, HPT, and OB. In this analysis, we identified 3859 (p < 0.05) significant pleiotropic genes, among which 163 are unique pleiotropic genes (pFDR <0.05) (Additional file 2: Table S6). MAGMA gene analysis identified 52 repeat unique pleiotropic genes across various trait pairs (Additional file 2: Table S7). For instance, TFAP2B was identified as a unique pleiotropic appearing in five trait pairs: KUS, CAS, MI, HPT, and OB, followed by NEK4, GNL3, APEH, FES, GLT8D1, MST1, PBRM1, NT5DC2, SMIM4, SPCS1, Furin, and NRF123 in four trait pairs. Through further analysis of these pleiotropic genes, we discovered that they were involved in multiple critical biological processes, including transcriptional regulation processes, metabolic processes, transmembrane transport of drugs, and cardiac structure development (Fig. 4A and Additional file 2: Table S8). Subsequent tissue-specific analyses revealed these pleiotropic genes were mainly enriched in subcutaneous adipose and skeletal muscle (Fig. 4B and Additional file 2: Table S9). These results revealed the shared genetic mechanisms between KUS and CVDs, as well as MetS.

Fig. 4.

Bar plot of (A) MAGMA gene-set analysis and (B) tissue-specific analysis for pleiotropic genes. Note: The red dotted line represents the Bonferroni adjusted P values, and the blue represents the significance of 0.05. KUS, kidney and ureteral stones; CAS, coronary atherosclerosis; AP, angina pectoris; MI, myocardial infarction; HPT, hypertension; OB, obesity.

3.5. Drug targets of European population

Initially, employing the SMR method, we identified 6663 potential candidate drug targets that associated with complex traits (p SMR<0.05, p HEIDI>0.05) (Additional file 2: Table S10). Based on the enrichment analysis of pleiotropic genes in tissues, we further summarized the pleiotropic genes identified in different methods, tissues, and traits (Fig. 5 and Additional file 2: Table S11). We overserved several pleiotropic genes such as NEK4, GNL3, GLT8D1, PBRM1, SMIM4, and SPCS1 were significantly mapped in different tissues using different methods. Moreover, eQTL and SMR analyses confirmed these genes' pleiotropic effects on at least four traits and offered precise chromosomal location annotations.

Fig. 5.

Overview of pleiotropic genes between KUS and other traits using different methods. eQTL, expression quantitative trait loci; SMR, summary-based Mendelian randomization.

3.6. Metabolite-related mechanisms shared between CVDs, MetS and KUS

The involvement of subcutaneous adipose tissue and skeletal muscle, along with metabolic processes, indicated a significant role of metabolic mechanisms in these diseases. We conducted a multi-trait colocalization analysis employing HyPrColoc to identify key plasma metabolites (Additional file 2: Table S12). Results identified three pleiotropic loci (rs77924615, rs11642015, and rs56094641) supporting the significance of 49 unique plasma metabolites shared between KUS and HPT through common causal variants. Notably, amino acids constituted the predominant portion of these plasma metabolites. Enrichment analysis for these plasma metabolites revealed tryptophan metabolism (p = 6.37 × 10⁻⁴) might be a crucial shared metabolic pathway in KUS and HPT (Fig. 6 and Additional file 2: Table S13).

Fig. 6.

Metabolite enrichment analysis for KUS and HPT.

3.7. The causal relationship between KUS and CVDs or MetS estimated by MR

MR analyses using the methods of IVW, Weighted median, MR Egger, Simple mode and Weighted mode did not show significant casual effects of CAS, AP, MI, HPT and OB on KUS (Fig. 7A and Additional fle 2: Table S14). However, reverse MR analysis demonstrated significant causal effects of KUS on risks of AP (OR = 1.002, 95 %CI = 1.000–1.003, p = 0.015), CAS (OR = 1.003, 95 %CI = 1.001–1.005, p < 0.001) and HPT (OR = 1.010, 95 %CI = 1.006–1.014, p < 0.001) by using IVW method (Fig. 7B and Additional file 2: Table S15). Nevertheless, the odds ratios were very close to 1, indicating that these traits provided limited evidence of causal associations.

Fig. 7.

The forest plot shows causal associations between KUS and CVDs, as well as MetS by using (A) MR analysis and (B) reverse MR analysis. IVW, Inverse Variance Weighted; KUS, kidney and ureteral stones; CAS, coronary atherosclerosis; AP, angina pectoris; MI, myocardial infarction; HPT, hypertension; OB, obesity.

4. Discussion

Considering the similar metabolic responses and common pathophysiologic mechanisms, KUS may exhibit a complex relationship with CVDs and MetS [9]. The study utilized extensive genetic methods to examine the genetic association between CVDs, MetS, and KUS. Through genetic correlation analysis, we identified significant genetic correlations between KUS and CVDs as well as MetS, including conditions such as CAS, AP, MI, HPT, and OB. We presented compelling evidence supporting the genetic correlation between MI and KUS, along with HDL results indicating that patients with should be warned about the risk of developing MI, aligning with previous studies [33,34]. Additionally, obesity, as the predominant component of MetS, had a significantly higher incidence of stone formation compared to individuals with normal weight [11]. However, it is crucial to acknowledge that correlation does not imply causation. At present, there is a lack of research examining the genetic connections related to these conditions from a hereditary standpoint. Investigating shared mechanisms, particularly at the molecular level, seems necessary.

We identified a series of genomic risk loci associated with KUS and CVDs, as well as MetS, and several pleiotropic risk loci were found across multiple traits. For instance, 16q12.2 was linked to KUS, HPT and OB, which had a PP.H4 exceeding 0.7 through colocalization analysis. The gene FTO situated at this locus, which was proved to have a significant association with CVDs and MetS, including atherosclerosis, hypertension and obesity [[35], [36], [37], [38]]. Besides, we identified several pleiotropic risk loci shared among more traits, such as 3p21.1, which presented in KUS, HPT, AP, and CAS, and contained the NEK4, GNL3, GLT8D1, PBRM1, and SMIM4 genes [39]. NEK4 and SMIM4 were both proved crucial for regulating mitochondrial function [[40], [41], [42], [43]], and dysfunction of mitochondrial was closely related to CVDs [44], MetS [45], and KUS [46,47]. GLT8D1 was a member of glycosyltransferase, capable of catalyzing the glycosylation process, which was crucial for tissue and cellular homeostasis [48]. Glycosylation was widely reported to be associated with kidney diseases [49], CVDs [50,51], and metabolic diseases [52]. GNL3, also known as nucleostemin (NS), participated in cardiomyocytes apoptosis induced by ischemia-reperfusion through regulating the P53 signaling pathway [53,54]. As a key component of nucleolar stress response, GNL3 was critically involved in the pathogenesis of ischemic heart disease [55]. Notably, these genes also passed SMR screening as potential drug targets, potentially serving as common intervention points for KUS, CVDs, and MetS. The roles of NEK4 and SMIM4 in mitochondrial function were well elucidated, and mitochondrial-targeted drugs are being trialed in the cardiovascular [56,57] field. Therefore, these genes have the potential for clinical translation as drug targets.

We identified some risk loci that were consistent with previous researches. For instance, Hao et al. reported rs2206271 was a novel loci in KUS [58]. TFAP2B, the mapped gene in this loci, was identified as a unique pleiotropic gene appearing in KUS, CAS, MI, HPT, and OB through MAGMA gene analysis. As a member of the AP-2 transcription factors family, TFAP2B played a role in regulating the distal nephron development, crucial for kidney electrolyte balance and urine concentration [59]. In this study, utilizing the MAGMA analysis enabled a comprehensive examination of the enrichment of these pleiotropic genes across specific tissues. The results revealed a significant enrichment of these genes in skeletal muscle and subcutaneous adipose, aligning closely with their crucial roles in metabolic processes. Skeletal muscle was an important metabolic organ, playing a significant role in the metabolism of glucose and fatty acids [60]. Reduced skeletal muscle mass and function decreased energy utilization, leading to accumulation of subcutaneous adipose and obesity, while dyslipidemia promotes atherosclerosis, further progressing to heart disease and stroke [61]. Obese individuals have lower urine pH, increasing their risk of uric acid stones [62]. On the other hand, chronic inflammation occurring in skeletal muscle accompanied by metabolic disease could contribute to insulin resistance [63] and the progression of atherosclerosis [6]. Since insulin may boost renal calcium excretion, insulin resistance could lead to higher calcium excretion in obese stone formers [64,65]. Consequently, tissue-specific analyses support the physiological roles of these organs in CVDs, MetS, and KUS.

Through MAGMA gene-set analysis, it was observed that metabolic processes appear to be a shared characteristic among CVDs, MetS, and KUS. Firstly, organic cyclic compound catabolic process was observed to be enriched in KUS, CAS and AP. The catabolism of organic cyclic compounds, including polyphenols and flavonoids such as resveratrol [66,67] and quercetin [68,69], influenced oxidative stress and inflammation, which were critical factors in the development of atherosclerosis and kidney stone formation. Besides, lipid metabolism process might be an important biological process in KUS, CVDs, and MetS, as it was observed to appear simultaneously in different traits. For example, cholesterol efflux was observed in KUS and OB, while very long chain fatty acid CoA ligase activity was noted in KUS and MI, and lipoprotein particle receptor binding occurred in KUS and AP. Many studies demonstrated that lipid metabolism abnormality was closely related to the formation of kidney stones [[70], [71], [72]]. Lipid metabolism was widely recognized as a critical factor in the pathogenesis of CVDs, particularly atherosclerosis [[73], [74], [75]], as well as in the development of MetS [76,77]. In addition to metabolic processes, we also observed various transcriptional regulation processes were shared among these traits, highlighting the necessity for further investigation at the genetic and molecular levels.

In this study, multi-trait colocalization analysis by using HyPrColoc identified various amino acids as the main plasma metabolites shared between KUS and HPT. Further metabolite enrichment analysis revealed that tryptophan metabolism was the most significant metabolic pathway connecting both diseases. Previous study also found tryptophan metabolism was significantly enriched in KUS patients [78]. Tryptophan was metabolized via the kynurenine pathway, leading to oxidative stress and inflammation in kidney diseases, contributing to stone formation [79,80]. In addition, dietary tryptophan was shown to reduce hypertension development in hypertensive animal models [81]. It was reported that tryptophan, along with its metabolites including kynurenine and anthranilic acid, exhibited vasodilatory properties [82]. As a pleiotropic gene for KUS and HPT, FTO was also reported to play an important role in tryptophan metabolism, likely influencing the conversion of tryptophan to kynurenine [83]. On the other hand, tryptophan was a precursor of nicotinamide adenine dinucleotide (NAD), phosphorylated NAD (NADP), and NAD phosphate (NADPH), which could directly bind to FTO to enhance its demethylase activity, thereby reducing RNA m6A methylation [84]. In summary, the discovery of the tryptophan metabolism pathway as a shared pathway provided a new direction for the development of metabolic modulating drugs aimed at concurrently treating KUS and HPT.

Furthermore, a bidirectional two-sample MR analysis was conducted to explore the casual association between KUS with CVDs and MetS. We initially considered CAS, AP, MI, HPT, and OB as exposure variables and KUS as the outcome variable, but found no causal link with KUS. Reverse MR analysis showed significant causal effects of KUS on AP, CAS, and HPT risks, but the odds ratios were extremely near 1, suggesting limited evidence of causal associations. Biologically, recurrent stone episodes may promote systemic inflammation or endothelial dysfunction, indirectly contributing to cardiovascular remodeling over decades [85]. However, the observed effects could be influenced by unmeasured confounders. Dietary factors, such as high sodium or oxalate intake may simultaneously promote KUS and HPT [86], but genetic instruments for diet are poorly captured in GWAS. Therefore, we were more inclined to conclude that the associations between KUS and CAS, AP, MI, HPT, and OB were primarily driven by genetic factors rather than direct causal relationships.

In this study, we identified four high-confidence colocalized loci (PP.H4 > 0.7), including 16q12.2, which spans the FTO gene and is shared among KUS, HPT, and OB, as well as 3p21.1, which encompasses NEK4, GNL3, GLT8D1, PBRM1, and SMIM4, and is associated with four traits. Among 163 pleiotropic genes, seven core genes (FTO, NEK4, GNL3, GLT8D1, SMIM4, PBRM1, TFAP2B) were consistently observed across four or more traits and were validated as druggable targets through SMR analysis. Furthermore, the application of HyPrColoc to metabolite data identified tryptophan metabolism as a central pathway linking KUS and HPT, representing a methodological approach not previously reported in similar studies. Besides, through tissue-specific enrichment analyses, we identified skeletal muscle and subcutaneous adipose tissue as pivotal hubs in the interactions between KUS, CVDs, and MetS, offering a novel perspective on the underlying pathophysiological connections. These findings address a significant gap in the current understanding of the genetic mechanisms underlying these diseases.

Our study is subject to several limitations that warrant consideration. Firstly, the generalizability of our results may be limited because the GWAS data primarily focuses on European ancestry, which may not capture the full spectrum of genetic diversity across different populations. It is imperative to conduct future replication studies in diverse cohorts, such as African, Hispanic, and East Asian populations, to verify the universality of the identified pleiotropic loci, including 16q12.2 and 3p21.1, as well as the associated pathways, such as tryptophan metabolism. Secondly, while these analyses offer insights into potential biological mechanisms, they are inferential and based on existing databases and annotations. They may not accurately capture the complex biological relationships and need experimental validation. Future research should investigate these pleiotropic genes using in vitro methods like CRISPR/Cas9 in renal cell and cardiomyocyte lines to evaluate their impact on stone formation, lipid metabolism, or inflammation, and in vivo methods like knockout mouse models on lithogenic or atherogenic diets. These studies could determine if these genes directly influence shared pathways in KUS, CVDs, and MetS.

5. Conclusion

Our research has elucidated the genetic associations between KUS and CVDs, as well as MetS, with particular emphasis on its connections to CAS, AP, MI, HPT, and OB. We have identified 937 pleiotropic SNPs and 35 genomic risk loci, including four high-confidence loci, such as 16q12.2 and 3p21.1, showing strong colocalization (PP.H4 > 0.7). Seven core pleiotropic genes (FTO, NEK4, GNL3, GLT8D1, SMIM4, PBRM1, TFAP2B) were linked to ≥4 traits and validated as potential drug targets via SMR. Additionally, metabolic pathways like organic cyclic compound, lipid, and amino acid metabolism may be common mechanisms underlying these diseases.

CRediT authorship contribution statement

Yibo Hua: Formal analysis, Investigation, Writing – original draft. Zhengkai Huang: Investigation, Methodology. Yu Yin: Software, Data curation. Rijin Song: Data curation, Supervision. Xianghu Meng: Conceptualization, Funding acquisition.

Availability of data and materials

The public datasets were downloaded and analyzed in this study were available in the GWAS repository, including FinnGen, UK Bank, and EBI.

Ethics approval and consent to participate

The patients involved in the public database have obtained ethical approval. Users can download relevant data for free for research and publish relevant articles. Our study is based on open-source data, so there are no ethical issues and other conflicts of interest.

Consent for publication

Written informed consent was obtained from all individuals.

Funding

The present study was funded by the National Natural Science Foundation of China (grant number: 81801438).

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Xianghu Meng reports financial support was provided by National Natural Science Foundation of China. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article.

Data availability

Data will be made available on request.