Genetic Evidence of Causal Effect between C1q/TNF-Related Protein-1 and Atherosclerosis: a Bidirectional and Multivariate Mendelian Randomization Study

Abstract

Aims: To investigate the causal relationship between C1q/TNF-related protein-1 (CTRP1) and atherosclerosis across various vascular sites, informed by studies connecting CTRP1 to coronary artery disease.

Methods: Summary statistics of CTRP1 from the available genome-wide association studies and atherosclerosis in classic vascular sites (including cerebral, coronary, and other arteries) from the FinnGen biobank were extracted for a primary MR analysis, and the analysis was replicated using Ischemic Stroke cohort (large artery atherosclerosis) for validation. The inverse variance-weighted method was used for primary assessment. Sensitivity analysis was performed by Cochrane’s Q test and leave-one-out analysis. Potential pleiotropic effects were assessed by MR-Egger intercept and MR-PRESSO global test. Additionally, multivariable MR (MVMR) analysis was performed to investigate the independent effect of CTRP1 on atherosclerosis after removing confounding factors.

Results: Reliable causal evidence was found for CTRP1 involvement in three atherosclerosis endpoints: causal effects of CTRP1 on cerebral atherosclerosis (OR=1.31, CI:1.04–1.66; FDR_P=0.0222)], coronary atherosclerosis (OR=1.13, CI: 1.08–1.19; FDR_P=2.86e-07), and atherosclerosis at other sites (OR=1.06, CI:1.02–1.11; FDR_P=0.0125). The validation cohort further confirmed its causal effect on large-artery atherosclerosis (OR=1.10, CI:1.03–1.18; FDR_P=0.0115). The reverse MR analysis did not support the causal effect of atherosclerosis on CTRP1. Moreover, the MVMR analysis, adjusting for confounders (CTRP3, CTRP5, and CTRP9A), highlighted a significant independent causal effect of CTRP1 remaining on atherosclerosis.

Conclusion: CTRP1 may represent a promising target for preventing and treating systemic atherosclerosis.

Bo Hu and Qing Zhou are joint senior authors.

Juhong Pan and Jia Huang contributed equally to this work.

Introduction

Cardiovascular and cerebrovascular diseases, along with peripheral arterial diseases, are major global health concerns and represent the leading causes of morbidity, mortality, and disability worldwide¹⁾ Atherosclerosis, the primary pathological basis of these conditions, is a progressive, chronic inflammatory disease of medium-sized and large arteries, characterized by endothelial dysfunction, lipid deposition, inflammatory cell infiltration, foam cell formation, and smooth muscle cell proliferation and migration^{2, 3)} which lead to progressive thickening and hardening of the arterial wall and gradual narrowing of the lumen or plaque rupture, resulting in thrombosis and vascular occlusion, consequently causing hypoxia in the heart, brain, and other vital organs, leading to fatal and disabling diseases⁴⁾. Despite decades of research aimed at reducing the burden of atherosclerotic cardiovascular disease (ASCVD) through the control of traditional risk factors, significant residual risk persists, and atherosclerosis has progresses insidiously over the decades. This ongoing challenge underscores the urgent need for new and effective clinical risk assessment/monitoring of biomarkers and therapeutic targets for the early diagnosis and treatment of atherosclerosis.

To explore potential biomarkers and therapeutic targets for atherosclerosis, researchers have conducted numerous studies on proteins⁵⁾, exosomes⁶⁾, metabolites⁷⁾, and circulating nucleic acids⁸⁾. Recent studies have highlighted the potential role of a class of proteins mainly secreted by adipose tissue, known as Complement C1q tumor necrosis factor-related proteins (CTRPs), in the occurrence and progression of atherosclerosis. These findings suggest that CTRPs may serve as effective diagnostic biomarkers and therapeutic targets, offering new avenues for addressing the residual risk of ASCVD. Among these, C1q/TNF-related protein-1 (CTRP1)⁹⁾ has been widely studied owing to its pleiotropic biological functions, including lipid metabolism^{10, 11)}, insulin sensitivity¹²⁾, inflammation regulation¹³⁾, and endothelial function¹⁴⁾. The serum level of CTRP1 is reported to be elevated in patients with coronary artery diseases^15-17). Additionally, Aleksandra et al. reported that the CTRP1 concentration in ischemic stroke (IS) patients was higher than in controls¹⁸⁾.These outcomes prompt us to investigate CTRP1 in the context of systemic atherosclerosis affecting various anatomical sites.

Although previous observational studies have found a link between elevated CTRP1 levels and atherosclerosis, establishing causality remains challenging owing to potential reverse causation and residual confounding factors. Mendelian randomization (MR) is a data analysis technique that uses genetic variants as instrumental variables to assess the observed causal relationships with heritable risk factors and diseases¹⁹⁾. In this study, we utilized a bidirectional two-sample MR and multivariate MR approach to evaluate the potential causal associations between serum CTRP1 levels and the risk of atherosclerosis across different vascular beds, including the cerebral, coronary, and other sites, thereby clarifying whether CTRP1 can serve as a diagnostic biomarker and therapeutic target for ASCVD.

Methods

Study Design

A bidirectional two-sample MR analysis was conducted to identify any potential causal correlation between CTRP1 and atherosclerosis at three sites, including CEREBATHER (cerebral atherosclerosis), CORATHER (coronary atherosclerosis), and ATHSCLE (atherosclerosis, excluding cerebral, coronary, and PAD), and further validated using Ischemic Stroke cohort (large artery atherosclerosis). The methodology employed for this MR analysis was based on three fundamental assumptions¹⁹⁾. Subsequently, a multivariable MR analysis was performed to control for confounding variables and evaluate the independent causal effect of CTRP1 on atherosclerosis. The whole research framework is illustrated in Fig.1.

Fig.1. Study overview

Overview of major analytical methods and data sources used in the present study to identify the causality between CTRP1 and atherosclerosis, as well as MR Assumptions. CTRP1, Complement C1q tumor necrosis factor-related protein 1; IVs, instrumental variables; SNPs, single nucleotide polymorphisms; LD, linkage disequilibrium; CEREBATHER, cerebral atherosclerosis; CORATHER, coronary atherosclerosis; ATHSCLE, atherosclerosis, excluding cerebral, coronary and PAD; IS ischemic stroke (large artery atherosclerosis); AS, atherosclerosis; CTRP3, Complement C1q tumor necrosis factor-related protein 3; CTRP5, Complement C1q tumor necrosis factor-related protein 5; CTRP9A, Complement C1q tumor necrosis factor-related protein 9A. Image prepared By Figdraw.

Study Samples

According to Mendel’s law of inheritance, genes carried by chromosomes undergo random assignment during meiosis, facilitating the simulation of randomized controlled experiments without confounding factors. Here, we utilized GWAS to identify genetic variables associated with both CTRP1 and atherosclerosis. To reduce potential bias stemming from population heterogeneity, all participants in this study were of European descent. Detailed information on each cohort is provided in Table 1.

Table 1.Detailed information for the GWAS data

Exposures/ Outcomes	ICD-10 code	GWAS ID	Data Source	Sample Size	Cases/Controls	PMID	Year	Population
CTRP1	/	prot-a-303	Sun BB et al.	3,301	/	29875488	2018	European
CEREBATHER	I67. 2	finngen_R10_I9_CEREBATHER	FinnGen	412,181	342/411,839	36653562	2023	European
CORATHER	I24, I25, T82.2, Z95.1	finngen_R10_I9_CORATHER	FinnGen	394,668	51,589/343,079	36653562	2023	European
ATHSCLE	I70	finngen_R10_I9_ATHSCLE	FinnGen	398,220	16,243/381,977	36653562	2023	European
IS (val)	I70.91	ebi-a-GCST005840	Malik R et al.	150,765	4,373/406,111	29531354	2018	European

Notes: CTRP1 Complement C1q tumor necrosis factor-related protein 1, CEREBATHER Cerebral atherosclerosis, CORATHER Coronary atherosclerosis, ATHSCLE Atherosclerosis, excluding cerebral, coronary, and PAD, IS Ischemic stroke (large artery atherosclerosis); ICD-10 code International Classification of Diseases, 10th Revision code, GWAS genome-wide association study.

Genetic Association Datasets for CTRP1

We selected SNPs closely linked to CTRP1 as instrumental variables (IVs) based on summary data retrieved from the IEU OpenGWAS project (https://gwas.mrcieu.ac.uk/). The CTRP1 cohort (prot-a-303) from the INTERVAL study contained 3301 samples, with 10,534,735 SNPs acquired through SOMAscan aptamer-based proteomics²⁰⁾.

Genetic Association Datasets for Atherosclerosis

Primary cohorts of atherosclerosis were obtained from the latest release of the FinnGen consortium, updated on December 18, 2023 (https://r10.finngen.fi). The FinnGen study is a large-scale genomics initiative that has analyzed over 500,000 Finnish biobank samples and correlated genetic variation with health data to understand disease mechanisms and predispositions²¹⁾. A total of 412,181 subjects with cerebral atherosclerosis were included in this study, with 342 cases and 411,839 controls (GWAS ID: Finngen _R10_I9_CEREBATHER). A total of 394,668 subjects with coronary atherosclerosis were included, with 51,589 cases and 343,079 controls (GWAS ID: Finngen _R10_I9_CORATHER). A total of 398,220 subjects with atherosclerosis, excluding cerebral, coronary, and PAD, were included in 16,243 cases and 381,977 controls (GWAS ID: Finngen _R10_I9_ATHSCLE). For the validation analysis, we extracted ischemic stroke (ebi-a-GCST005840) data, which contained 150,765 samples with 4,373 cases and 406,111 controls²²⁾, to check the consistency of the findings across different datasets.

As all studies that contributed to this MR study were reviewed and approved by the ethics review boards of the relevant institutions, no additional consent or ethics approval was required for the current study.

Instrumental Variable (IV) selection

MR uses genetic variation as an IV to estimate the causal impact of risk factors on multiple complex diseases²³⁾. The effectiveness of the MR analysis in this study relied on the availability of sufficient genetic variants associated with CTRP1, which is also related to atherosclerosis. Single nucleotide polymorphisms (SNPs) were filtered using the genome-wide significance criterion of P＜5＜10⁻⁸ in all two-sample univariable MR (UVMR) and multivariable MR (MVMR) analyses, except for a lower threshold²⁴⁾ of P＜5＜10⁻⁶ for CEREBATHER as exposure in the reverse MR, as only one SNP was identified at P＜5＜10⁻⁸ in the CEREBATHER GWAS summary statistics. Subsequently, we selected independent SNPs (kb=10,000, r²＜0.001) according to the linkage disequilibrium (LD) pattern in European populations. Finally, we calculated the F statistic in univariable two-sample MR analyses to assess the overall strength of the selected SNPs in explaining phenotypic variation using the formula

$F=\frac{N-K-1}{K}\times \frac{R^2}{1-R^2}$ ,

where F ＞10 indicates substantial efficacy of the instrumental variables in mitigating potential biases²⁵⁾. The IVs of UVMR and MVMR are listed in Supplementary Table 1.

Supplementary Table 1.IVs of UVMR and MVMR

UVMR CTRT1-ATHSLCE
	SNP	effect_allele. exposure	other_allele. exposure	effect_allele. outcome	other_allele. outcome	beta.exposure	beta.outcome	eaf.exposure	eaf.outcome
1	rs10762481	C	T	C	T	-0.1846	-0.0090	0.3827	NA
2	rs112635299	T	G	T	G	-0.584	-0.0930	0.0226	NA
3	rs5848	T	C	T	C	-0.2483	0.0096	0.2874	NA
4	rs646776	T	C	T	C	0.6399	0.0442	0.7765	NA
5	rs72802395	A	G	A	G	-0.2547	-0.0246	0.0859	NA

	remove	palindromic	ambiguous	id.outcome	pval.outcome	se.outcome	outcome	mr_keep. outcome	pval_origin. outcome
1	0	0	0	kJnKkR	0.4527	0.0120	outcome	1	reported
2	0	0	0	kJnKkR	0.0300	0.0428	outcome	1	reported
3	0	0	0	kJnKkR	0.4435	0.0125	outcome	1	reported
4	0	0	0	kJnKkR	0.0022	0.0145	outcome	1	reported
5	0	0	0	kJnKkR	0.2627	0.0220	outcome	1	reported

	chr.exposure	pos.exposure	se.exposure	pval.exposure	samplesize. exposure	id.exposure	exposure	mr_keep. exposure	pval_origin. exposure
1	10	73538164	0.0254	3.72E-13	3301	prot-a-303	Complement C1q tumor necrosis factor- related protein 1 || id:prot-a-303	1	reported
2	14	94838142	0.0838	3.16E-12	3301	prot-a-303	Complement C1q tumor necrosis factor- related protein 1 || id:prot-a-303	1	reported
3	17	42430244	0.0282	1.26E-18	3301	prot-a-303	Complement C1q tumor necrosis factor- related protein 1 || id:prot-a-303	1	reported
4	1	109818530	0.0281	1.82E-114	3301	prot-a-303	Complement C1q tumor necrosis factor- related protein 1 || id:prot-a-303	1	reported
5	16	75286484	0.0444	9.77E-9	3301	prot-a-303	Complement C1q tumor necrosis factor- related protein 1 || id:prot-a-303	1	reported

	data_source. exposure	action	SNP_index	mr_keep	samplesize. outcome	PVE	FSTAT
1	igd	2	1	1	3301	0.0157	52.7877
2	igd	2	1	1	3301	0.0145	48.5372
3	igd	2	1	1	3301	0.0229	77.4804
4	igd	2	1	1	3301	0.1358	518.2608
5	igd	2	1	1	3301	0.0099	32.8873

UVMR CTRT1-CEREBATHER
	SNP	effect_allele. exposure	other_allele. exposure	effect_allele. outcome	other_allele. outcome	beta.exposure	beta.outcome	eaf.exposure	eaf.outcome
1	rs10762481	C	T	C	T	-0.1846	-0.0469	0.3827	NA
2	rs112635299	T	G	T	G	-0.584	-0.2612	0.0226	NA
3	rs5848	T	C	T	C	-0.2483	-0.1120	0.2874	NA
4	rs646776	T	C	T	C	0.6399	0.1503	0.7765	NA
5	rs72802395	A	G	A	G	-0.2547	-0.0322	0.0859	NA

	remove	palindromic	ambiguous	id.outcome	pval.outcome	se.outcome	outcome	mr_keep. outcome	pval_origin. outcome
1	0	0	0	aBbIxB	0.5475	0.0780	outcome	1	reported
2	0	0	0	aBbIxB	0.3479	0.2783	outcome	1	reported
3	0	0	0	aBbIxB	0.1691	0.0815	outcome	1	reported
4	0	0	0	aBbIxB	0.1059	0.0929	outcome	1	reported
5	0	0	0	aBbIxB	0.8213	0.1427	outcome	1	reported

	chr.exposure	pos.exposure	se.exposure	pval.exposure	samplesize. exposure	id.exposure	exposure	mr_keep. exposure	pval_origin. exposure
1	10	73538164	0.0254	3.72E-13	3301	prot-a-303	|| id:prot-a-303	1	reported
2	14	94838142	0.0838	3.16E-12	3301	prot-a-303	|| id:prot-a-303	1	reported
3	17	42430244	0.0282	1.26E-18	3301	prot-a-303	|| id:prot-a-303	1	reported
4	1	109818530	0.0281	1.82E-114	3301	prot-a-303	|| id:prot-a-303	1	reported
5	16	75286484	0.0444	9.77E-9	3301	prot-a-303	|| id:prot-a-303	1	reported

	data_source. exposure	action	SNP_index	mr_keep	samplesize. outcome	PVE	FSTAT
1	igd	2	1	1	3301	0.0157	52.7877
2	igd	2	1	1	3301	0.0145	48.5372
3	igd	2	1	1	3301	0.0229	77.4804
4	igd	2	1	1	3301	0.1358	518.2608
5	igd	2	1	1	3301	0.0099	32.8873

UVMR CTRT1-CORATHER
	SNP	effect_allele. exposure	other_allele. exposure	effect_allele. outcome	other_allele. outcome	beta.exposure	beta.outcome	eaf.exposure	eaf.outcome
1	rs10762481	C	T	C	T	-0.1846	-0.0227	0.3827	NA
2	rs112635299	T	G	T	G	-0.584	-0.0369	0.0226	NA
3	rs5848	T	C	T	C	-0.2483	-0.0104	0.2874	NA
4	rs646776	T	C	T	C	0.6399	0.0891	0.7765	NA
5	rs72802395	A	G	A	G	-0.2547	-0.0660	0.0859	NA

	remove	palindromic	ambiguous	id.outcome	pval.outcome	se.outcome	outcome	mr_keep. outcome	pval_origin. outcome
1	0	0	0	uAaR9R	0.0036	0.0078	outcome	1	reported
2	0	0	0	uAaR9R	0.1791	0.0274	outcome	1	reported
3	0	0	0	uAaR9R	0.1990	0.0081	outcome	1	reported
4	0	0	0	uAaR9R	3.84E-21	0.0094	outcome	1	reported
5	0	0	0	uAaR9R	4.51E-6	0.0144	outcome	1	reported

	chr.exposure	pos.exposure	se.exposure	pval.exposure	samplesize. exposure	id.exposure	exposure	mr_keep. exposure	pval_origin. exposure
1	10	73538164	0.0254	3.72E-13	3301	prot-a-303	|| id:prot-a-303	1	reported
2	14	94838142	0.0838	3.16E-12	3301	prot-a-303	|| id:prot-a-303	1	reported
3	17	42430244	0.0282	1.26E-18	3301	prot-a-303	|| id:prot-a-303	1	reported
4	1	109818530	0.0281	1.82E-114	3301	prot-a-303	|| id:prot-a-303	1	reported
5	16	75286484	0.0444	9.77E-9	3301	prot-a-303	|| id:prot-a-303	1	reported

	data_source. exposure	action	SNP_index	mr_keep	samplesize. outcome	PVE	FSTAT
1	igd	2	1	1	3301	0.0157	52.7877
2	igd	2	1	1	3301	0.0145	48.5372
3	igd	2	1	1	3301	0.0229	77.4804
4	igd	2	1	1	3301	0.1358	518.2608
5	igd	2	1	1	3301	0.0099	32.8873

MVMR CTRT1, 3, 5, 9A-ATHSLCE
SNP	exposure	id. exposure	effect_allele. exposure	other_allele. exposure	eaf. exposure	beta. exposure	se. exposure	pval. exposure
rs10762481	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	C	T	0.3827	-0.1846	0.0254	3.72E-13
rs112635299	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	G	0.0226	-0.584	0.0838	3.16E-12
rs11955347	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.4719	0.0126	0.0249	0.6166
rs2248863	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.1418	0.0488	0.035	0.1622
rs5848	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.2874	-0.2483	0.0282	1.26E-18
rs61932023	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	C	A	0.3319	-0.0118	0.0262	0.6457
rs646776	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.7765	0.6399	0.0281	1.82E-114
rs7259081	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	T	0.7513	0.1739	0.0314	3.16E-8
rs72802395	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.0859	-0.2547	0.0444	9.77E-9
rs74480769	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	G	A	0.0317	0.0402	0.0737	0.5888
rs7539005	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	A	0.6103	0.0197	0.0256	0.4467
rs840382	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.3941	-0.0469	0.0255	0.0661
rs9266141	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	G	C	0.2340	-0.0653	0.0312	0.0363
rs10762481	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	C	T	0.3827	0.026	0.0256	0.3090
rs112635299	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	G	0.0226	0.1854	0.0843	0.0275
rs11955347	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.4719	0.0349	0.0249	0.1622
rs2248863	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.1418	0.0107	0.035	0.7586
rs5848	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.2874	-0.024	0.0285	0.3981
rs61932023	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	C	A	0.3319	-0.0004	0.0262	0.9772
rs646776	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.7765	0.0376	0.0303	0.2138
rs7259081	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	T	0.7513	0.1016	0.0315	0.0013
rs72802395	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.0859	-0.0176	0.0446	0.6918
rs74480769	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	G	A	0.0317	-0.0373	0.0738	0.6166
rs7539005	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	A	0.6103	0.0844	0.0256	0.0010
rs840382	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.3941	0.1411	0.0254	2.82E-8
rs9266141	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	G	C	0.2340	0.0231	0.0312	0.4571
rs10762481	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	C	T	0.3827	-0.0018	0.0256	0.9333
rs112635299	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	G	0.0226	0.1363	0.0843	0.1072
rs11955347	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.4719	-0.2464	0.0245	9.12E-24
rs2248863	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.1418	0.3261	0.0345	3.55E-21
rs5848	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.2874	0.036	0.0285	0.2042
rs61932023	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	C	A	0.3319	-0.1884	0.026	3.89E-13
rs646776	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.7765	0.0015	0.0303	0.9550
rs7259081	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	T	0.7513	0.1134	0.0315	0.0003
rs72802395	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.0859	-0.0027	0.0446	0.9550
rs74480769	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	G	A	0.0317	0.0334	0.0738	0.6457
rs7539005	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	A	0.6103	-0.0079	0.0256	0.7586
rs840382	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.3941	0.0066	0.0255	0.7943
rs9266141	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	G	C	0.2340	0.1755	0.031	1.58E-8
rs10762481	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	C	T	0.3827	-0.0304	0.0256	0.2344
rs112635299	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	G	0.0226	0.0764	0.0844	0.3631
rs11955347	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.4719	0.0276	0.0249	0.2692
rs2248863	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.1418	0.0231	0.035	0.5129
rs5848	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.2874	-0.0142	0.0285	0.6166
rs61932023	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	C	A	0.3319	-0.0347	0.0262	0.1862
rs646776	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.7765	0.0179	0.0303	0.5495
rs7259081	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	T	0.7513	0.0865	0.0315	0.0060
rs72802395	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.0859	0.0029	0.0446	0.9550
rs74480769	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	G	A	0.0317	-0.6455	0.0729	8.32E-19
rs7539005	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	A	0.6103	-0.1899	0.0254	7.41E-14
rs840382	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.3941	0.0092	0.0255	0.7244
rs9266141	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	G	C	0.2340	0.0192	0.0312	0.5370

MVMR CTRT1, 3, 5, 9A-CEREBATHER
SNP	exposure	id. exposure	effect_allele. exposure	other_allele. exposure	eaf. exposure	beta. exposure	se. exposure	pval. exposure
rs10762481	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	C	T	0.3827	-0.1846	0.0254	3.72E-13
rs112635299	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	G	0.0226	-0.584	0.0838	3.16E-12
rs11955347	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.4719	0.0126	0.0249	0.6166
rs2248863	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.1418	0.0488	0.035	0.1622
rs5848	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.2874	-0.2483	0.0282	1.26E-18
rs61932023	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	C	A	0.3319	-0.0118	0.0262	0.6457
rs646776	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.7765	0.6399	0.0281	1.82E-114
rs7259081	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	T	0.7513	0.1739	0.0314	3.16E-8
rs72802395	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	A	G	0.0859	-0.2547	0.0444	9.77E-9
rs74480769	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	G	A	0.0317	0.0402	0.0737	0.5888
rs7539005	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	A	0.6103	0.0197	0.0256	0.4467
rs840382	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	T	C	0.3941	-0.0469	0.0255	0.0661
rs9266141	Complement C1q tumor necrosis factor-related protein 1 || id:prot-a-303	prot-a-303	G	C	0.2340	-0.0653	0.0312	0.0363
rs10762481	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	C	T	0.3827	0.026	0.0256	0.3090
rs112635299	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	G	0.0226	0.1854	0.0843	0.0275
rs11955347	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.4719	0.0349	0.0249	0.1622
rs2248863	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.1418	0.0107	0.035	0.7586
rs5848	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.2874	-0.024	0.0285	0.3981
rs61932023	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	C	A	0.3319	-0.0004	0.0262	0.9772
rs646776	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.7765	0.0376	0.0303	0.2138
rs7259081	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	T	0.7513	0.1016	0.0315	0.0013
rs72802395	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	A	G	0.0859	-0.0176	0.0446	0.6918
rs74480769	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	G	A	0.0317	-0.0373	0.0738	0.6166
rs7539005	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	A	0.6103	0.0844	0.0256	0.0010
rs840382	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	T	C	0.3941	0.1411	0.0254	2.82E-8
rs9266141	Complement C1q tumor necrosis factor-related protein 3 || id:prot-a-304	prot-a-304	G	C	0.2340	0.0231	0.0312	0.4571
rs10762481	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	C	T	0.3827	-0.0018	0.0256	0.9333
rs112635299	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	G	0.0226	0.1363	0.0843	0.1072
rs11955347	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.4719	-0.2464	0.0245	9.12E-24
rs2248863	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.1418	0.3261	0.0345	3.55E-21
rs5848	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.2874	0.036	0.0285	0.2042
rs61932023	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	C	A	0.3319	-0.1884	0.026	3.89E-13
rs646776	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.7765	0.0015	0.0303	0.9550
rs7259081	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	T	0.7513	0.1134	0.0315	0.0003
rs72802395	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	A	G	0.0859	-0.0027	0.0446	0.9550
rs74480769	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	G	A	0.0317	0.0334	0.0738	0.6457
rs7539005	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	A	0.6103	-0.0079	0.0256	0.7586
rs840382	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	T	C	0.3941	0.0066	0.0255	0.7943
rs9266141	Complement C1q tumor necrosis factor-related protein 5 || id:prot-a-305	prot-a-305	G	C	0.2340	0.1755	0.031	1.58E-8
rs10762481	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	C	T	0.3827	-0.0304	0.0256	0.2344
rs112635299	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	G	0.0226	0.0764	0.0844	0.3631
rs11955347	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.4719	0.0276	0.0249	0.2692
rs2248863	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.1418	0.0231	0.035	0.5129
rs5848	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.2874	-0.0142	0.0285	0.6166
rs61932023	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	C	A	0.3319	-0.0347	0.0262	0.1862
rs646776	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.7765	0.0179	0.0303	0.5495
rs7259081	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	T	0.7513	0.0865	0.0315	0.0060
rs72802395	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	A	G	0.0859	0.0029	0.0446	0.9550
rs74480769	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	G	A	0.0317	-0.6455	0.0729	8.32E-19
rs7539005	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	A	0.6103	-0.1899	0.0254	7.41E-14
rs840382	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	T	C	0.3941	0.0092	0.0255	0.7244
rs9266141	Complement C1q and tumor necrosis factor-related protein 9A || id:prot-a-306	prot-a-306	G	C	0.2340	0.0192	0.0312	0.5370

MVMR CTRT1, 3, 5, 9A-CORATHER
SNP	exposure	id.exposure	effect_allele. exposure	other_allele. exposure	eaf.exposure	beta.exposure	se.exposure	pval.exposure
rs10762481	CTRP1	prot-a-303	C	T	0.3827	-0.1846	0.0254	3.72E-13
rs112635299	CTRP1	prot-a-303	T	G	0.0226	-0.584	0.0838	3.16E-12
rs11955347	CTRP1	prot-a-303	A	G	0.4719	0.0126	0.0249	0.6166
rs2248863	CTRP1	prot-a-303	A	G	0.1418	0.0488	0.035	0.1622
rs5848	CTRP1	prot-a-303	T	C	0.2874	-0.2483	0.0282	1.26E-18
rs61932023	CTRP1	prot-a-303	C	A	0.3319	-0.0118	0.0262	0.6457
rs646776	CTRP1	prot-a-303	T	C	0.7765	0.6399	0.0281	1.82E-114
rs7259081	CTRP1	prot-a-303	A	T	0.7513	0.1739	0.0314	3.16E-8
rs72802395	CTRP1	prot-a-303	A	G	0.0859	-0.2547	0.0444	9.77E-9
rs74480769	CTRP1	prot-a-303	G	A	0.0317	0.0402	0.0737	0.5888
rs7539005	CTRP1	prot-a-303	T	A	0.6103	0.0197	0.0256	0.4467
rs840382	CTRP1	prot-a-303	T	C	0.3941	-0.0469	0.0255	0.0661
rs9266141	CTRP1	prot-a-303	G	C	0.2340	-0.0653	0.0312	0.0363
rs10762481	CTRP3	prot-a-304	C	T	0.3827	0.026	0.0256	0.3090
rs112635299	CTRP3	prot-a-304	T	G	0.0226	0.1854	0.0843	0.0275
rs11955347	CTRP3	prot-a-304	A	G	0.4719	0.0349	0.0249	0.1622
rs2248863	CTRP3	prot-a-304	A	G	0.1418	0.0107	0.035	0.7586
rs5848	CTRP3	prot-a-304	T	C	0.2874	-0.024	0.0285	0.3981
rs61932023	CTRP3	prot-a-304	C	A	0.3319	-0.0004	0.0262	0.9772
rs646776	CTRP3	prot-a-304	T	C	0.7765	0.0376	0.0303	0.2138
rs7259081	CTRP3	prot-a-304	A	T	0.7513	0.1016	0.0315	0.0013
rs72802395	CTRP3	prot-a-304	A	G	0.0859	-0.0176	0.0446	0.6918
rs74480769	CTRP3	prot-a-304	G	A	0.0317	-0.0373	0.0738	0.6166
rs7539005	CTRP3	prot-a-304	T	A	0.6103	0.0844	0.0256	0.0010
rs840382	CTRP3	prot-a-304	T	C	0.3941	0.1411	0.0254	2.82E-8
rs9266141	CTRP3	prot-a-304	G	C	0.2340	0.0231	0.0312	0.4571
rs10762481	CTRP5	prot-a-305	C	T	0.3827	-0.0018	0.0256	0.9333
rs112635299	CTRP5	prot-a-305	T	G	0.0226	0.1363	0.0843	0.1072
rs11955347	CTRP5	prot-a-305	A	G	0.4719	-0.2464	0.0245	9.12E-24
rs2248863	CTRP5	prot-a-305	A	G	0.1418	0.3261	0.0345	3.55E-21
rs5848	CTRP5	prot-a-305	T	C	0.2874	0.036	0.0285	0.2042
rs61932023	CTRP5	prot-a-305	C	A	0.3319	-0.1884	0.026	3.89E-13
rs646776	CTRP5	prot-a-305	T	C	0.7765	0.0015	0.0303	0.9550
rs7259081	CTRP5	prot-a-305	A	T	0.7513	0.1134	0.0315	0.0003
rs72802395	CTRP5	prot-a-305	A	G	0.0859	-0.0027	0.0446	0.9550
rs74480769	CTRP5	prot-a-305	G	A	0.0317	0.0334	0.0738	0.6457
rs7539005	CTRP5	prot-a-305	T	A	0.6103	-0.0079	0.0256	0.7586
rs840382	CTRP5	prot-a-305	T	C	0.3941	0.0066	0.0255	0.7943
rs9266141	CTRP5	prot-a-305	G	C	0.2340	0.1755	0.031	1.58E-8
rs10762481	CTRP 9A	prot-a-306	C	T	0.3827	-0.0304	0.0256	0.2344
rs112635299	CTRP 9A	prot-a-306	T	G	0.0226	0.0764	0.0844	0.3631
rs11955347	CTRP 9A	prot-a-306	A	G	0.4719	0.0276	0.0249	0.2692
rs2248863	CTRP 9A	prot-a-306	A	G	0.1418	0.0231	0.035	0.5129
rs5848	CTRP 9A	prot-a-306	T	C	0.2874	-0.0142	0.0285	0.6166
rs61932023	CTRP 9A	prot-a-306	C	A	0.3319	-0.0347	0.0262	0.1862
rs646776	CTRP 9A	prot-a-306	T	C	0.7765	0.0179	0.0303	0.5495
rs7259081	CTRP 9A	prot-a-306	A	T	0.7513	0.0865	0.0315	0.0060
rs72802395	CTRP 9A	prot-a-306	A	G	0.0859	0.0029	0.0446	0.9550
rs74480769	CTRP 9A	prot-a-306	G	A	0.0317	-0.6455	0.0729	8.32E-19
rs7539005	CTRP 9A	prot-a-306	T	A	0.6103	-0.1899	0.0254	7.41E-14
rs840382	CTRP 9A	prot-a-306	T	C	0.3941	0.0092	0.0255	0.7244
rs9266141	CTRP 9A	prot-a-306	G	C	0.2340	0.0192	0.0312	0.5370

Univariate Bidirectional MR Analysis

For univariate bidirectional two-sample MR analyses, we used five MR methods to estimate the causal effect between CTRP1 levels and atherosclerosis outcomes: the random effect inverse variance weighting (IVW) method was the main MR analysis method, and the MR Egger, weighted median, simple mode, and weighted mode methods were used as complementary methods^{26, 27)}.

Sensitivity Analysis

The sensitivity analysis encompassed leave-one-out sensitivity, heterogeneity, and pleiotropy tests. The leave-one-out sensitivity test evaluates the MR results by recalculating them after removing each instrumental variable individually. Significant discrepancies between the recalculated and final results indicate the sensitivity of the MR analysis to that particular instrumental variable²⁸⁾. Heterogeneity among the IVs was analyzed using MR-Egger and IVW in Cochran’s Q statistic, and P＞0.05 indicated that there was no significant heterogeneity. Additionally, to investigate potential pleiotropy of relevant SNPs, we conducted the MR-Egger intercept test and Mendelian Randomization Pleiotropy Residual Sum and Outlier (MR-PRESSO) analysis, where a p-value and global test P value ＞0.05 indicates the absence of horizontal pleiotropy.

Multivariable MR (MVMR)

MVMR extends the capabilities of MR by simultaneously investigating causal relationships between multiple exposures and outcomes. This method requires that all chosen instrumental variables (IVs) be conditionally independent of the outcome, given both exposures and confounding variables²⁹⁾. For significant causality in the univariate MR analysis, we further employed a multivariate inverse variance-weighted (IVW) analysis to evaluate the independent causal effects of CTRP1 on atherosclerosis outcomes. This approach aimed to adjust for potential confounding factors including CTRP3, CTRP5, and CTRP9A. These proteins, from the CTRPs superfamily also secreted by adipose tissue, may have potentially similar or contradictory biological functions to CTRP1 and risks associated with atherosclerosis, thus serving as control exposures alongside CTRP1 in the MVMR analysis, according to the description of Zhao et al.³⁰⁾. The causal relationship between each variable and atherosclerosis at different sites was assessed using an IVW model.

Statistical Analysis

All statistical tests were conducted between two samples using the “TwoSampleMR” (version 0.5.11) and “MR-PRESSO” packages in R (version 4.3.2). The MVMR analysis was performed using the “MVMR” package (version 0.4). For the univariate MR analysis, combining the strengths of each MR method, the five approaches complemented one another and furnished robust causal effects in our study, especially when the IVW P value was ＜0.05, and the direction of β values remained consistent. To address multiple comparisons across atherosclerosis outcomes, we corrected the P-values derived from the IVW approach using the Benjamini-Hochberg method, also known as the false discovery rate (FDR) method. Statistical significance was defined as an FDR-adjusted P value of ＜0.05. Effect values from MR analyses are represented by odds ratios (ORs) and 95% confidence intervals (CI). P values of ＜0.05 were considered to indicate statistical significance. In the reverse MR analysis, a P value ＞0.05 indicated no evidence of reverse causality between exposures and outcomes.

Results

Univariable Analysis

Causal Effect of CTRP1 on Atherosclerosis Endpoints

In the primary cohort, our MR estimate based on the IVW method yielded positive results. Genetically predicted higher CTRP1 levels were associated with an increased risk of cerebral atherosclerosis (OR=1.31, 95% CI: 1.04–1.66, P=2.22×10⁻²), increased risk of coronary atherosclerosis (OR, 1.13; 95% CI: 1.08–1.19, P=7.15×10⁻⁸), and increased risk of atherosclerosis, excluding cerebral, coronary, and PAD (OR=1.06, 95% CI: 1.02–1.11], P=8.64＜10⁻³) (Fig.2).

Fig.2. Forest plot of the association of CTRP1 with atherosclerosis

Different methods were used to obtain OR values and 95% confidence intervals. CTRP1, Complement C1q tumor necrosis factor-related protein 1; CEREBATHER, cerebral atherosclerosis; CORATHER, coronary atherosclerosis; ATHSCLE, atherosclerosis; excluding cerebral, coronary, and PAD; IS, ischemic stroke (large artery atherosclerosis); val, validation; CI, confidence interval; OR, odds ratio; FDR-P, P value of false discovery rate.

To test for consistency across datasets, we replicated the MR analysis using a validation cohort from a large meta-analysis. We observed a similar significant causal relationship between CTRP1 and ischemic stroke (OR=1.10, 95%CI: 1.03–1.18, P=6.24×10⁻³).

The remaining four methods all showed a trend towards an increasing association between the circulating CTRP1 level and the risk of atherosclerosis, although not all reached statistical significance. The results are shown in Fig.2. Scatter plots further illustrated that increasing circulating CTRP1 was correlated with an increased risk of atherosclerosis (Fig.3).

Fig.3. Scatter plots for the association between CTRP1 and atherosclerosis

a CTRP1 on CEREBATHER. b CTRP1 on CORATHER. c CTRP1 on ATHSCLE. d IS (val). SNP effects are plotted into lines for the inverse-variance weighted test (blue line), MR-Egger (red line), weighted median (light blue line), simple mode (green line), and weighted mode (purple line). The slope of the line corresponds to the causal estimation.

To address the potential for false positives in the multiple comparisons conducted in the univariate two-sample MR analyses across both the primary and validation cohorts, where CTRP1 levels were considered as the exposure and atherosclerosis at various sites was considered as the outcome, we applied FDR correction to the IVW results. The analysis yielded FDR-corrected P values for four distinct atherosclerotic cohorts (FDR_P=0.0222 for CEREBATHER, FDR_P=2.86＜10⁻⁷ for CORATHER, FDR_P=0.0125 for ATHSCLE, and FDR_P=0.0115 for IS), all of which were ＜0.05. This provides robust evidence to support the notion that CTRP1 acts as a risk factor for atherosclerosis (Fig.2).

Causal Effect of Different Atherosclerosis Endpoints on CTRP1

Reverse Mendelian randomization (MR) was used to investigate whether genetically predicted atherosclerosis caused alterations in CTRP1 levels. In the reverse MR across the five methodologies, no significant causal effect of genetic predisposition to the three atherosclerosis endpoints on the risk of CTRP1 was observed in the primary cohort. This includes CEREBATHER (β=-0.03, 95% CI: -0.09–0.02, P=0.2201), CORATHER (β=0.06, 95% CI: -0.07–0.19, P=0.3881), and ATHSCLE (β=-0.01, 95% CI: -0.21–0.18, P=0.8884), which were all based on the IVW method. Similarly, there was no evidence to support a significant causal link of genetically predicted IS on evaluated serum CTRP1 levels (β=0.07, 95% CI: -0.12–0.26, P=0.4782) in the reverse MR analysis conducted on the validation cohort (Fig.4).

Fig.4. Forest plot of the reverse MR analysis of the association of atherosclerosis with CTRP1

Different methods were used to obtain beta values and 95% confidence intervals. CTRP1, Complement C1q tumor necrosis factor-related protein 1; CEREBATHER, cerebral atherosclerosis; CORATHER, coronary atherosclerosis; ATHSCLE, atherosclerosis, excluding cerebral, coronary and PAD; IS, ischemic stroke (large artery atherosclerosis); val, validation; CI, confidence interval; β effect size for the effect allele.

Sensitivity Analysis

Upon completion of the main MR analysis, a sensitivity analysis was performed on the results of the MR analysis. The “Leave-one-out plot” revealed that none of the SNPs exert a dominant influence on the estimated causal association between CTRP1 and the risk of the four atherosclerosis endpoints (Fig.5).

Fig.5. Leave-one-out plots for the causal effect of CTRP1 on atherosclerosis

a CTRP1 on CEREBATHER. b CTRP1 on CORATHER. c CTRP1 on ATHSCLE. d IS (val). CTRP1 Complement C1q tumor necrosis factor-related protein 1; CEREBATHER, cerebral atherosclerosis; CORATHER, coronary atherosclerosis; ATHSCLE, atherosclerosis, excluding cerebral, coronary, and PAD; IS, ischemic stroke (large artery atherosclerosis); val, validation; SNPs, single nucleotide polymorphisms. The black line represents the deviation of the 95% confidence interval corresponding to the estimate of the SNPs. The red line corresponds to the estimated value from the IVW test. After removing the SNPs one by one, there was no difference in the final results.

Cochran’s Q test indicated no significant heterogeneity between CTRP1 levels and the risk of atherosclerosis, with p-values ＞0.05 for both the IVW and MR Egger models. However, some heterogeneity was observed in certain, associations such as CTRP1-CORATHER (Table 2). Nevertheless, the use of the random-effects IVW method in our study ensured the validity of the MR estimates, potentially mitigating any pooled heterogeneity.

Table 2.Heterogeneity and pleiotropy test of CTRP IVs in atherosclerosis endpoints GWAS

Exposure	Atherosclerosis endpoints	Heterogeneity test				Pleiotropy test
		IVW		MR‐Egger		MR‐Egger			MR‐PRESSO Global test P
		Q‐statistics	P	Q‐statistics	P	Intercept	SE	P
CTRP1	CEREBATHER	0.567810	0.966573	0.508058	0.917117	0.022518	0.092118	0.822653	0.930000
CTRP1	CORATHER	14.776677	0.005188	13.892259	0.003056	-0.008686	0.019875	0.691641	0.148000
CTRP1	ATHSCLE	6.039546	0.196214	4.350088	0.226058	-0.018453	0.017095	0.359453	0.359000
CTRP1	IS (Val)	4.083618	0.394808	2.594327	0.458485	-0.029111	0.023854	0.309502	0.406000

Notes: CTRP1 Complement C1q tumor necrosis factor-related protein 1, CEREBATHER Cerebral atherosclerosis, CORATHER Coronary atherosclerosis, ATHSCLE Atherosclerosis, excluding cerebral, coronary, and PAD, IS Ischemic stroke (large artery atherosclerosis)

Val validation. P＞0.05 indicates no significant pleiotropy. Q‐statistics_P value ＞0.05, indicating no significant heterogeneity.

The pleiotropy test further confirmed the causal relationship between CTRP1 levels and atherosclerosis for both P values of the MR-Egger intercept and MR-PRESSO Global test ＞0.05 (Table 2). MR-PRESSO did not yet reveal any outliers.

Multivariable Mendelian Randomization (MVMR) Analysis of Confounding Factors

The MVMR analysis aimed to isolate the independent effect of CTRP1 on atherosclerosis by adjusting for potential confounders, such as CTRP3, CTRP5, and CTRP9A. The dataset information and MVMR analysis results are listed in Table 3. Remarkably, the findings from the MVMR analysis were consistent with those from the univariate MR analysis, with CTRP1 exhibiting a significance level of P＜0.05 across all three cohorts (Table 3). Interestingly, only CTRP9A in the CEREBATHER cohort displayed a significance level of P＜0.05 (Table 3), suggesting that its concurrent presence with CTRP1 may contribute to the development of cerebral atherosclerosis. Conversely, CTRP3 and CTRP5 did not appear to significantly influence the causal relationship between CTRP1 and atherosclerosis.

Table 3.Analysis of confounders

Outcomes	GWAS ID	Confounders	nSNP	beta	SE	P	OR	or_lci95	or_uci95
CEREBATHER
	prot-a-303	CTRP1	5	0.281331	0.045058	4.27E-10	1.324892	1.212903	1.447221
	prot-a-304	CTRP3	1	0.089953	0.192440	0.640189	1.094123	0.750340	1.595417
	prot-a-305	CTRP5	3	0.010125	0.075709	0.893612	1.010176	0.870868	1.171769
	prot-a-306	CTRP9A	1	-0.402766	0.082313	0.000001	0.668469	0.568872	0.785502
CORATHER
	prot-a-303	CTRP1	5	0.124246	0.020002	5.24E-10	1.132295	1.088763	1.177568
	prot-a-304	CTRP3	1	-0.024984	0.084420	0.767269	0.975325	0.826589	1.150826
	prot-a-305	CTRP5	3	-0.019609	0.033360	0.556672	0.980582	0.918516	1.046841
	prot-a-306	CTRP9A	1	0.038314	0.036175	0.289545	1.039058	0.967936	1.115405
ATHSCLE
	prot-a-303	CTRP1	5	0.056930	0.020367	0.005186	1.058582	1.017156	1.101695
	prot-a-304	CTRP3	1	0.040516	0.086150	0.638142	1.041348	0.879556	1.232901
	prot-a-305	CTRP5	3	-0.049464	0.033999	0.145706	0.951740	0.890385	1.017322
	prot-a-306	CTRP9A	1	-0.021748	0.037044	0.557134	0.978486	0.909961	1.052172

Notes: CEREBATHER Cerebral atherosclerosis, CORATHER Coronary atherosclerosis, ATHSCLE Atherosclerosis, excluding cerebral, coronary, and PAD; GWAS genome-wide association study, CTRP1 Complement C1q tumor necrosis factor-related protein 1, CTRP3 Complement C1q tumor necrosis factor-related protein 3, CTRP5 Complement C1q tumor necrosis factor-related protein 5, CTRP9A Complement C1q tumor necrosis factor-related protein 9A, nSNP number of single-nucleotide polymorphisms, beta effect size for the effect allele, SE standard error, OR odds ratio.

Discussion

In the present study, we used univariate bidirectional MR analyses to investigate the genetic evidence of a causal relationship between CTRP1 levels and classical atherosclerotic diseases (CEREBATHER, CORATHER, and ATHSCLE) in the primary cohort, with subsequent validation in another atherosclerosis cohort (IS). A significant causal link between the genetic susceptibility of circulating CTRP1 levels and the probability of developing atherosclerosis was detected and further confirmed through FDR correction. However, we did not observe a causal relationship between atherosclerosis and serum CTRP1 levels in reverse MR analyses. Additionally, the sensitivity analyses provided further support for the robustness of our results. Finally, by employing an MVMR analysis, we found substantial evidence supporting an independent causal effect of genetically predicted CTRP1 levels on the risk of atherosclerosis after adjusting for confounding factors.

Many observational studies have provided findings that are consistent with our Mendelian randomization (MR) results regarding coronary atherosclerosis. CTRP1, a member of the CTRPs family, may influence the development of atherosclerosis by regulating immune inflammation, glycolipid metabolism, and the endothelial function. It has garnered widespread attention, particularly in the field of cardiovascular disease, with a specific emphasis on coronary atherosclerosis, which is also known as coronary artery disease¹⁵⁾.

Yuasa et al. first suggested that increased circulating CTRP1 levels are associated with the prevalence of CAD³¹⁾. A study by Lu et al. included 451 patients with CAD and 686 non-CAD controls and found that CTRP1 levels were elevated in patients with CAD and were associated with the severity of CAD¹³⁾. A small sample of case-control clinical studies found that circulating levels of CTRP1 were associated with obesity indices and carotid intima-media thickness (cIMT) values in patients with type 2 diabetes³²⁾. Given that the onset and progression of atherosclerosis are often asymptomatic and that cIMT serves as a marker of subclinical atherosclerosis, these findings suggest that serum CTRP1 levels can be utilized as biomarkers for the early diagnosis of atherosclerosis. Another study by Axel et al.³³⁾ assessed CTRP1 serum levels in 539 patients undergoing coronary angiography for established or suspected stable CAD by prospectively recording major adverse cardiovascular events (MACE) over an 8-year follow-up period. The study suggested a significant association between elevated serum CTRP1 levels and future MACE. Similarly, according to a recent study³⁴⁾ that included 200 patients with acute myocardial infarction (AMI) who underwent percutaneous coronary intervention (PCI), CTRP1 may serve as a target for diagnostic and prognostic biomarkers of myocardial remodeling following acute myocardial infarction. These studies support the notion that CTRP1 can act as a prognostic factor for coronary atherosclerosis.

A growing number of observational studies have also reported an association between elevated CTRP1 and cerebral atherosclerosis, along with complications, such as neurological damage and cognitive impairment. Consistent with our genetic findings, a study involving 43 stroke patients³⁵⁾ found that higher CTRP1 was associated with an increased risk of ischemic stroke (IS), which also showed a positive correlation with high-sensitivity C-reactive protein (hs-CRP). Similarly, another study including 46 patients with ischemic stroke and 32 controls showed that CTRP1 could be an independent predictive factor of ischemic stroke¹⁸⁾. Moreover, Li et al. conducted an investigation³⁶⁾ that enrolled 452 patients with IS and 403 healthy subjects, and implied that CTRP1 can effectively predict the severity of neurological deficits in patients with acute IS.

In contrast to coronary and cerebral atherosclerosis, relatively few studies have investigated the link between serum CTRP1 levels and atherosclerosis at other sites, such as the aorta, renal artery, or mesenteric artery. In an animal experiment, Liu et al. observed a notable increase in the expression of CTRP1 in vascular endothelial cells subjected to disturbed flow, as demonstrated by the positive staining of the mouse aorta. This finding indicated that CTRP1 acts as a mechanosensitive proinflammatory factor in the early stages of arteriosclerosis¹⁴⁾, which is consistent with our results investigating the causal effect of genetically predicted CTRP1 on ATHSCLE.

Our findings, which indicate that genetically-predicted higher serum levels of CTRP1 are associated with an increased risk of atherosclerosis, are further supported by several mechanistic studies. CTRP1 secreted by vascular cells, leukocytes, and adipose tissue accelerates the development of atherosclerosis by promoting vascular inflammation, macrophage foam cell formation, and endothelial barrier dysfunction. Lu et al. demonstrated that CTRP1 is a pro-atherogenic adipokine in mice through modulation of arginase 1 expression and enzyme activity via the p38MAPK/NF-κB pathway¹³⁾. Macrophages are typical cellular components within atherosclerotic plaques, where they accumulate in large quantities by phagocytizing oxidized low-density lipoproteins to form the core of the plaque, promoting its progression. Simultaneously, they contribute to arterial wall tissue repair and remodeling by clearing necrotic cells. Contributing to its pro-inflammatory and pro-atherogenic effects, CTRP1 plays a role in lipid metabolism and macrophage inflammation during the onset and development of atherosclerosis³⁷⁾. Additionally, CTRP1 restrained cholesterol efflux and promoted macrophage lipid accumulation through the miR-424-5p/FoxO1/ABCA1 ³⁸⁾ signaling pathway, thereby facilitating the development of atherosclerosis. It is well established that structural changes in arteries, such as stenosis, bifurcation, or curvature, can cause the blood flow pattern to transition from laminar to turbulent. In addition, turbulence results in irregular fluctuations in wall shear stress. Persistent or recurrent alternation between regions of high and low shear stress imposes mechanical stress on endothelial cells, ultimately leading to endothelial dysfunction. Liu et al. found that CTRP1 contributes to the vascular hyperpermeability induced by turbulent flow through the activation of VEGFR2 ¹⁴⁾.

Considering that atherosclerosis is a chronic inflammatory disease³⁹⁾ and that CTRP1 levels may increase in such states⁴⁰⁾, we performed a reverse Mendelian randomization (MR) analysis to investigate whether genetically determined atherosclerosis was causally associated with circulating CTRP1 levels. Surprisingly, the results showed nonsignificant p-values for this association, indicating the absence of a causal effect. A possible explanation for this finding may be that atherosclerosis is a multifactorial disease that manifests in individuals after many years of exposure to various risk factors. Numerous confounding factors, including diabetes, obesity, and hypertension, which are common traditional risk factors for cardiovascular disease and may affect CTRP1 levels, are also associated with atherosclerosis. Our MR analysis rigorously excluded the interference of these multiple confounding factors.

It is also noteworthy that CTRPs have a significant impact on various cell types involved in atherosclerosis, including endothelial cells, macrophages, and vascular smooth muscle cells (VSMCs). However, individual members may have similar or opposite effects on the development of atherosclerosis⁴¹⁾. An MVMR analysis, derived from two-sample Mendelian randomization (TSMR), offers a more robust framework for assessing causality and controlling for confounding factors²⁹⁾. After adjusting for MVMR, CTRP1 still exhibited a significant independent causal effect on coronary, cerebral, and other sites of atherosclerosis, consistent with our results of the univariate two-sample MR analyses. Notably, CTRP9A has been identified as a confounder, specifically in the relationship between CTRP1 and cerebral atherosclerosis. These findings suggest that while atherosclerosis is multifactorial, CTRP1 retains an independent causal role with potential interactions with CTRP9A in cerebral artery health, which is supported by previous findings^{42, 43)}. Rigorous exclusion of confounding factors enhances the robustness of these findings, highlighting the importance of further research to elucidate the biological mechanisms underlying these associations and their clinical implications.

Strengths and Limitations

The strength of this study lies primarily in its research design. First, we comprehensively extracted gene-outcome associations from three distinct GWAS datasets on classical atherosclerotic diseases obtained from the FinnGen database (R10) release. Subsequent validation was conducted using an ischemic stroke (large artery atherosclerosis) dataset from a large-scale meta-analysis to confirm the stability of the results. The effect estimates from different data sources consistently pointed in the same direction and the FDR-adjusted p-values remained statistically significant, enhancing the robustness of our findings. Second, we conducted reverse Mendelian randomization analyses to explore the genetic determinants of atherosclerosis on CTRP1, which offers a more comprehensive understanding of causality and aids in delineating the causal pathways between the two entities. Third, a series of sensitivity analyses, including heterogeneity tests, leave-one-out analyses, and MR-PRESSO yielded consistent results, thereby enhancing the reliability of our study. Furthermore, to mitigate the influence of confounding factors, we performed multivariable analyses by jointly assessing the association of CTRP1 with multiple members of the CTRP family at different sites of atherosclerosis, strongly explaining the independent role of CTRP1 in atherosclerosis.

The present study was also associated with several limitations. First, the exclusive use of summary data prevented us from evaluating potential effect modifications according to sex, age, or previous disease. Second, the data in this study were derived solely from European populations, which, while reducing the population stratification bias, necessitates further validation for generalizability to other populations.

Conclusion

Our study confirms that CTRP1 is an independent risk factor for atherosclerosis. The association between genetic variations in CTRP1 and disease demonstrated a stronger causal relationship, surpassing traditional correlation studies. This provides robust evidence for its potential use as a diagnostic marker and a therapeutic target. This finding not only aids in the early detection and precise treatment of atherosclerosis, but also offers new directions for future clinical translational research related to individualized prevention and intervention. This study may have a positive impact on the patient prognosis and quality of life. Further research is needed—including analyses of the 15 human CTRPs—to elucidate the full underlying biological mechanisms and explore the therapeutic potential of targeting these proteins in systemic atherosclerotic diseases.

Abbreviations

ATHSCLE Atherosclerosis, excluding cerebral, coronary and PAD

CEREBATHER Cerebral atherosclerosis

CORATHER Coronary atherosclerosis

CTRP1 Complement C1q tumor necrosis factor-related protein 1

CTRP3 Complement C1q tumor necrosis factor-related protein 3

CTRP5 Complement C1q tumor necrosis factor-related protein 5

CTRP9A Complement C1q tumor necrosis factor-related protein 9A

GWAS Genome-wide association study

nSNP Number of single-nucleotide polymorphisms

MR Mendelian randomization

MVMR Multivariable mendelian randomization

OR Odds ratio

SE Standard error

Acknowledgements

We extend our gratitude to the participants and investigators of the FinnGen Biobank, MEGASTROKE Consortium, and INTERVAL study, and Figdraw.

Data Availability

The datasets for CTRP1, CTRP3, CTRP5 and CTRP9A and Ischemic stroke are available from the IEU OpenGWAS project (https://gwas.mrcieu.ac.uk/). The CEREBATHER, CORATHER, and ATHSCLE datasets are available from the FinnGen consortium (DF10, released December 18 2023, https://r10.finngen.fi/).

Author Contributions

Juhong Pan: Conceptualization, writing the first draft of the manuscript, and writing review; Jia Huang: Data curation, statistical analysis, and editing; Yueying Chen, Nan Jiang, and Yuxin Guo: Investigation and editing; Ji Zhang, Shiyuan Zhou, and Huan Pu: Data curation and validation. Qing Deng: critical review. Bo Hu: Designed the study and supervised and revised the manuscript. Qing Zhou: Conceptualization, study design, manuscript supervision and revision, and funding. All authors contributed to the article and the final version was approved by all authors.

Funding

This study was supported by the National Natural Science Foundation of China (Grant Nos. 81971624 and 82271999) and the Cross-Innovation Talent Project of the People’s Hospital of Wuhan University (JCRCYR-2022-011).

Ethics Approval and Consent to Participate

Not applicable.

All data utilized in this study were publicly available and anonymized. Each contributing study received appropriate ethical approval and patient consent from the respective original GWAS study sites. The detailed information can be found in the original studies.

Consent for Publication

Not applicable.

Conflicts of Interest

The authors declare no conflicts of interest.

References

• 1)  Vaduganathan M, Mensah GA, Turco JV, Fuster V and Roth GA: The Global Burden of Cardiovascular Diseases and Risk: A Compass for Future Health. J Am Coll Cardiol, 2022; 80: 2361-2371
• 2)  Lusis AJ: Atherosclerosis. Nature, 2000; 407: 233-241
• 3)  Forstermann U, Xia N and Li H: Roles of Vascular Oxidative Stress and Nitric Oxide in the Pathogenesis of Atherosclerosis. Circ Res, 2017; 120: 713-735
• 4)  Libby P: The changing landscape of atherosclerosis. Nature, 2021; 592: 524-533
• 5)  Hoefer IE, Steffens S, Ala-Korpela M, Back M, Badimon L, Bochaton-Piallat ML, Boulanger CM, Caligiuri G, Dimmeler S, Egido J, Evans PC, Guzik T, Kwak BR, Landmesser U, Mayr M, Monaco C, Pasterkamp G, Tunon J, Weber C, Atherosclerosis ESCWG and Vascular B: Novel methodologies for biomarker discovery in atherosclerosis. Eur Heart J, 2015; 36: 2635-2642
• 6)  Wang C, Li Z, Liu Y and Yuan L: Exosomes in atherosclerosis: performers, bystanders, biomarkers, and therapeutic targets. Theranostics, 2021; 11: 3996-4010
• 7)  Sardar SW, Nam J, Kim TE, Kim H and Park YH: Identification of Novel Biomarkers for Early Diagnosis of Atherosclerosis Using High-Resolution Metabolomics. Metabolites, 2023; 13:
• 8)  Kianmehr A, Qujeq D, Bagheri A and Mahrooz A: Oxidized LDL-regulated microRNAs for evaluating vascular endothelial function: molecular mechanisms and potential biomarker roles in atherosclerosis. Crit Rev Clin Lab Sci, 2022; 59: 40-53
• 9)  van Hinsbergh VW and Eringa EC: C1q/TNF-related protein 1: a novel link between visceral fat and athero-inflammation. Eur Heart J, 2016; 37: 1772-1774
• 10)  Si Y, Fan W and Sun L: A Review of the Relationship Between CTRP Family and Coronary Artery Disease. Curr Atheroscler Rep, 2020; 22: 22
• 11)  Chu LY, Ramakrishnan DP and Silverstein RL: Thrombospondin-1 modulates VEGF signaling via CD36 by recruiting SHP-1 to VEGFR2 complex in microvascular endothelial cells. Blood, 2013; 122: 1822-1832
• 12)  Xin Y, Zhang D, Fu Y, Wang C, Li Q, Tian C, Zhang S and Lyu X: C1qTNF-related protein 1 improve insulin resistance by reducing phosphorylation of serine 1101 in insulin receptor substrate 1. Endocr J, 2017; 64: 787-796
• 13)  Lu L, Zhang RY, Wang XQ, Liu ZH, Shen Y, Ding FH, Meng H, Wang LJ, Yan XX, Yang K, Wang HB, Pu LJ, Zhang Q, Chen QJ, De Caterina R and Shen WF: C1q/TNF-related protein-1: an adipokine marking and promoting atherosclerosis. Eur Heart J, 2016; 37: 1762-1771
• 14)  Liu ZH, Li C, Chen JW, Shen Y, Gao J, Shen WF, Zhang RY, Wang XQ and Lu L: C1q/TNF-related protein 1 promotes endothelial barrier dysfunction under disturbed flow. Biochem Biophys Res Commun, 2017; 490: 580-586
• 15)  van Rosendael AR, van den Hoogen IJ, Lin FY, Gianni U, Lu Y, Andreini D, Al-Mallah MH, Cademartiri F, Chinnaiyan K, Chow BJW, Conte E, Cury RC, Feuchtner G, de Araujo Goncalves P, Hadamitzky M, Kim YJ, Leipsic JA, Maffei E, Marques H, Plank F, Pontone G, Raff GL, Villines TC, Lee SE, Al’Aref SJ, Baskaran L, Cho I, Danad I, Gransar H, Budoff MJ, Samady H, Virmani R, Min JK, Narula J, Berman DS, Chang HJ, Shaw LJ and Bax JJ: Age related compositional plaque burden by CT in patients with future ACS. J Cardiovasc Comput Tomogr, 2022; 16: 491-497
• 16)  Shen Y, Lu L, Liu ZH, Wu F, Zhu JZ, Sun Z, Zhang RY, Zhang Q, Hu J, Chen QJ, Wu ZG and Shen WF: Increased serum level of CTRP1 is associated with low coronary collateralization in stable angina patients with chronic total occlusion. Int J Cardiol, 2014; 174: 203-206
• 17)  Zhang Y, Liu C, Liu J, Guo R, Yan Z, Liu W, Lau WB, Jiao X, Cao J, Xu K, Jia Y, Ma X and Wang Y: Implications of C1q/TNF-related protein superfamily in patients with coronary artery disease. Sci Rep, 2020; 10: 878
• 18)  Kazimierczak-Kabzinska A, Marek B, Borgiel-Marek H, Kajdaniuk D and Kos-Kudla B: Assessing the blood concentration of new adipocytokines in patients with ischaemic stroke. Endokrynol Pol, 2020; 71: 504-511
• 19)  Larsson SC, Butterworth AS and Burgess S: Mendelian randomization for cardiovascular diseases: principles and applications. Eur Heart J, 2023; 44: 4913-4924
• 20)  Sun BB, Maranville JC, Peters JE, Stacey D, Staley JR, Blackshaw J, Burgess S, Jiang T, Paige E, Surendran P, Oliver-Williams C, Kamat MA, Prins BP, Wilcox SK, Zimmerman ES, Chi A, Bansal N, Spain SL, Wood AM, Morrell NW, Bradley JR, Janjic N, Roberts DJ, Ouwehand WH, Todd JA, Soranzo N, Suhre K, Paul DS, Fox CS, Plenge RM, Danesh J, Runz H and Butterworth AS: Genomic atlas of the human plasma proteome. Nature, 2018; 558: 73-79
• 21)  Kurki MI, Karjalainen J, Palta P, Sipila TP, Kristiansson K, Donner KM, Reeve MP, Laivuori H, Aavikko M, Kaunisto MA, Loukola A, Lahtela E, Mattsson H, Laiho P, Della Briotta Parolo P, Lehisto AA, Kanai M, Mars N, Ramo J, Kiiskinen T, Heyne HO, Veerapen K, Rueger S, Lemmela S, Zhou W, Ruotsalainen S, Parn K, Hiekkalinna T, Koskelainen S, Paajanen T, Llorens V, Gracia-Tabuenca J, Siirtola H, Reis K, Elnahas AG, Sun B, Foley CN, Aalto-Setala K, Alasoo K, Arvas M, Auro K, Biswas S, Bizaki-Vallaskangas A, Carpen O, Chen CY, Dada OA, Ding Z, Ehm MG, Eklund K, Farkkila M, Finucane H, Ganna A, Ghazal A, Graham RR, Green EM, Hakanen A, Hautalahti M, Hedman AK, Hiltunen M, Hinttala R, Hovatta I, Hu X, Huertas-Vazquez A, Huilaja L, Hunkapiller J, Jacob H, Jensen JN, Joensuu H, John S, Julkunen V, Jung M, Junttila J, Kaarniranta K, Kahonen M, Kajanne R, Kallio L, Kalviainen R, Kaprio J, FinnGen, Kerimov N, Kettunen J, Kilpelainen E, Kilpi T, Klinger K, Kosma VM, Kuopio T, Kurra V, Laisk T, Laukkanen J, Lawless N, Liu A, Longerich S, Magi R, Makela J, Makitie A, Malarstig A, Mannermaa A, Maranville J, Matakidou A, Meretoja T, Mozaffari SV, Niemi MEK, Niemi M, Niiranen T, CJ OD, Obeidat ME, Okafo G, Ollila HM, Palomaki A, Palotie T, Partanen J, Paul DS, Pelkonen M, Pendergrass RK, Petrovski S, Pitkaranta A, Platt A, Pulford D, Punkka E, Pussinen P, Raghavan N, Rahimov F, Rajpal D, Renaud NA, Riley-Gillis B, Rodosthenous R, Saarentaus E, Salminen A, Salminen E, Salomaa V, Schleutker J, Serpi R, Shen HY, Siegel R, Silander K, Siltanen S, Soini S, Soininen H, Sul JH, Tachmazidou I, Tasanen K, Tienari P, Toppila-Salmi S, Tukiainen T, Tuomi T, Turunen JA, Ulirsch JC, Vaura F, Virolainen P, Waring J, Waterworth D, Yang R, Nelis M, Reigo A, Metspalu A, Milani L, Esko T, Fox C, Havulinna AS, Perola M, Ripatti S, Jalanko A, Laitinen T, Makela TP, Plenge R, McCarthy M, Runz H, Daly MJ and Palotie A: FinnGen provides genetic insights from a well-phenotyped isolated population. Nature, 2023; 613: 508-518
• 22)  Malik R, Chauhan G, Traylor M, Sargurupremraj M, Okada Y, Mishra A, Rutten-Jacobs L, Giese AK, van der Laan SW, Gretarsdottir S, Anderson CD, Chong M, Adams HHH, Ago T, Almgren P, Amouyel P, Ay H, Bartz TM, Benavente OR, Bevan S, Boncoraglio GB, Brown RD, Jr., Butterworth AS, Carrera C, Carty CL, Chasman DI, Chen WM, Cole JW, Correa A, Cotlarciuc I, Cruchaga C, Danesh J, de Bakker PIW, DeStefano AL, den Hoed M, Duan Q, Engelter ST, Falcone GJ, Gottesman RF, Grewal RP, Gudnason V, Gustafsson S, Haessler J, Harris TB, Hassan A, Havulinna AS, Heckbert SR, Holliday EG, Howard G, Hsu FC, Hyacinth HI, Ikram MA, Ingelsson E, Irvin MR, Jian X, Jimenez-Conde J, Johnson JA, Jukema JW, Kanai M, Keene KL, Kissela BM, Kleindorfer DO, Kooperberg C, Kubo M, Lange LA, Langefeld CD, Langenberg C, Launer LJ, Lee JM, Lemmens R, Leys D, Lewis CM, Lin WY, Lindgren AG, Lorentzen E, Magnusson PK, Maguire J, Manichaikul A, McArdle PF, Meschia JF, Mitchell BD, Mosley TH, Nalls MA, Ninomiya T, O’Donnell MJ, Psaty BM, Pulit SL, Rannikmae K, Reiner AP, Rexrode KM, Rice K, Rich SS, Ridker PM, Rost NS, Rothwell PM, Rotter JI, Rundek T, Sacco RL, Sakaue S, Sale MM, Salomaa V, Sapkota BR, Schmidt R, Schmidt CO, Schminke U, Sharma P, Slowik A, Sudlow CLM, Tanislav C, Tatlisumak T, Taylor KD, Thijs VNS, Thorleifsson G, Thorsteinsdottir U, Tiedt S, Trompet S, Tzourio C, van Duijn CM, Walters M, Wareham NJ, Wassertheil-Smoller S, Wilson JG, Wiggins KL, Yang Q, Yusuf S, Consortium AF, Cohorts for H, Aging Research in Genomic Epidemiology C, International Genomics of Blood Pressure C, Consortium I, Starnet, Bis JC, Pastinen T, Ruusalepp A, Schadt EE, Koplev S, Bjorkegren JLM, Codoni V, Civelek M, Smith NL, Tregouet DA, Christophersen IE, Roselli C, Lubitz SA, Ellinor PT, Tai ES, Kooner JS, Kato N, He J, van der Harst P, Elliott P, Chambers JC, Takeuchi F, Johnson AD, BioBank Japan Cooperative Hospital G, Consortium C, Consortium E-C, Consortium EP-I, International Stroke Genetics C, Consortium M, Neurology Working Group of the CC, Network NSG, Study UKYLD, Consortium M, Sanghera DK, Melander O, Jern C, Strbian D, Fernandez-Cadenas I, Longstreth WT, Jr., Rolfs A, Hata J, Woo D, Rosand J, Pare G, Hopewell JC, Saleheen D, Stefansson K, Worrall BB, Kittner SJ, Seshadri S, Fornage M, Markus HS, Howson JMM, Kamatani Y, Debette S and Dichgans M: Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet, 2018; 50: 524-537
• 23)  Cheng Q, Zhang X, Chen LS and Liu J: Mendelian randomization accounting for complex correlated horizontal pleiotropy while elucidating shared genetic etiology. Nat Commun, 2022; 13: 6490
• 24)  Zhang H, Zhan J, Jin J, Zhang J, Lu W, Zhao R, Ahearn TU, Yu Z, O’Connell J, Jiang Y, Chen T, Okuhara D, andMe Research T, Garcia-Closas M, Lin X, Koelsch BL and Chatterjee N: A new method for multiancestry polygenic prediction improves performance across diverse populations. Nat Genet, 2023; 55: 1757-1768
• 25)  Burgess S, Foley CN, Allara E, Staley JR and Howson JMM: A robust and efficient method for Mendelian randomization with hundreds of genetic variants. Nat Commun, 2020; 11: 376
• 26)  Qin J, Zhang L, Ke B, Liu T, Kong C and Jin C: Causal relationships between circulating inflammatory factors and IgA vasculitis: a bidirectional Mendelian randomization study. Front Immunol, 2023; 14: 1248325
• 27)  Taschler B, Smith SM and Nichols TE: Causal inference on neuroimaging data with Mendelian randomisation. Neuroimage, 2022; 258: 119385
• 28)  Dai Z, Xu W, Ding R, Peng X, Shen X, Song J, Du P, Wang Z and Liu Y: Two-sample Mendelian randomization analysis evaluates causal associations between inflammatory bowel disease and osteoporosis. Front Public Health, 2023; 11: 1151837
• 29)  Sanderson E, Davey Smith G, Windmeijer F and Bowden J: An examination of multivariable Mendelian randomization in the single-sample and two-sample summary data settings. Int J Epidemiol, 2019; 48: 713-727
• 30)  Yang Z, Schooling CM and Kwok MK: Credible Mendelian Randomization Studies in the Presence of Selection Bias Using Control Exposures. Front Genet, 2021; 12: 729326
• 31)  Yuasa D, Ohashi K, Shibata R, Takeshita K, Kikuchi R, Takahashi R, Kataoka Y, Miyabe M, Joki Y, Kambara T, Uemura Y, Matsuo K, Hayakawa S, Hiramatsu-Ito M, Ito M, Ikeda N, Murohara T and Ouchi N: Association of circulating C1q/TNF-related protein 1 levels with coronary artery disease in men. PLoS One, 2014; 9: e99846
• 32)  Majidi Z, Emamgholipour S, Omidifar A, Rahmani Fard S, Poustchi H and Shanaki M: The circulating levels of CTRP1 and CTRP5 are associated with obesity indices and carotid intima-media thickness (cIMT) value in patients with type 2 diabetes: a preliminary study. Diabetol Metab Syndr, 2021; 13: 14
• 33)  Muendlein A, Leiherer A, Saely C, Ebner J, Geiger K, Brandtner EM, Vonbank A, Fraunberger P and Drexel H: The novel adipokine CTRP1 is significantly associated with the incidence of major adverse cardiovascular events. Atherosclerosis, 2019; 286: 1-6
• 34)  Wang J, Ding Y, Yao YR, Liu HY and Gu Y: Effect of azilsartan on myocardial remodeling after acute myocardial infarction. Eur J Clin Pharmacol, 2024; 80: 223-230
• 35)  Wang H, Liu Q and Zhang X: C1q/tumor necrosis factor-related protein-1 attenuates microglia autophagy and inflammatory response by regulating the Akt/mTOR pathway. Life Sci, 2020; 256: 117992
• 36)  Suping L, Yan X, Fei X, Yi L, Baoming H and Hongbin S: The expression of C1q/TNF - related protein - 1 in patients with acute ischemic stroke and its predictive value for the severity of neurological deficits. The Journal of Practical Medicine 2018; 34: 1672-1675
• 37)  Wang XQ, Liu ZH, Xue L, Lu L, Gao J, Shen Y, Yang K, Chen QJ, Zhang RY and Shen WF: C1q/TNF-related protein 1 links macrophage lipid metabolism to inflammation and atherosclerosis. Atherosclerosis, 2016; 250: 38-45
• 38)  Zhang ZZ, Chen JJ, Deng WY, Yu XH and Tan WH: CTRP1 decreases ABCA1 expression and promotes lipid accumulation through the miR-424-5p/FoxO1 pathway in THP-1 macrophage-derived foam cells. Cell Biol Int, 2021; 45: 2226-2237
• 39)  Kong P, Cui ZY, Huang XF, Zhang DD, Guo RJ and Han M: Inflammation and atherosclerosis: signaling pathways and therapeutic intervention. Signal Transduct Target Ther, 2022; 7: 131
• 40)  Kim KY, Kim HY, Kim JH, Lee CH, Kim DH, Lee YH, Han SH, Lim JS, Cho DH, Lee MS, Yoon S, Kim KI, Yoon DY and Yang Y: Tumor necrosis factor-alpha and interleukin-1beta increases CTRP1 expression in adipose tissue. FEBS Lett, 2006; 580: 3953-3960
• 41)  Zhang H, Zhang-Sun ZY, Xue CX, Li XY, Ren J, Jiang YT, Liu T, Yao HR, Zhang J, Gou TT, Tian Y, Lei WR and Yang Y: CTRP family in diseases associated with inflammation and metabolism: molecular mechanisms and clinical implication. Acta Pharmacol Sin, 2023; 44: 710-725
• 42)  Yang C, Fan F, Sawmiller D, Tan J, Wang Q and Xiang Y: C1q/TNF-related protein 9: A novel therapeutic target in ischemic stroke? J Neurosci Res, 2019; 97: 128-136
• 43)  Yang C, Xin JY, Liu ZL, Fan F, Li YM, Jin F, Wang QS, Guo FQ, Yu NW, Le WD and Xiang Y: Association Between Serum C1q Tumor Necrosis Factor-Related Protein 9 and the Clinical Characteristics and Prognosis of Ischemic Stroke. Neurol Ther, 2022; 11: 87-101