Plasma metabolites mediates the causal effect of inflammatory proteins on Alzheimer's disease: a Mendelian randomization study

doi:10.12122/j.issn.1673-4254.2026.02.05

Abstract

Abstract:

Objective To investigate the causal relationship between inflammatory proteins and Alzheimer's disease (AD) and the mediating role of plasma metabolites therein. Methods Using Mendelian mandomization (MR) methods and publicly available genome-wide association study (GWAS) data, we selected 91 single nucleotide polymorphisms (SNPs) that were strongly linked to inflammatory proteins without reverse causality with AD as the outcome. A bidirectional two-sample MR analysis was performed. Inflammatory proteins with causal links to AD were identified via inverse variance weighted (IVW) analysis. A mediation MR analysis was then performed using 1400 plasma metabolites to assess their mediating role in this causal pathway. Results The preliminary bidirectional MR analysis identified 3 inflammatory proteins that had a potential positive causal association with AD without reverse causality: Axin-1, C-X-C motif chemokine ligand 11 (CXCL11), and interleukin-12β (IL-12β). Elevated levels of Axin-1 were positively causally associated with AD risk (OR=1.082, 95% CI: 1.009-1.159; P=0.026), while CXCL11 (OR=0.951, 95% CI: 0.914-0.990; P=0.026) and IL-12β (OR=0.959, 95% CI: 0.926-0.994; P=0.026) were negatively causally associated with AD risk. Sensitivity analyses indicated that these causal effects exhibited no heterogeneity or pleiotropy. In the mediation analysis, 18 plasma metabolites with causal links to AD were identified, among which 3 plasma metabolites exhibited mediating effects in the inflammatory protein-AD causal relationship. Methyl-4-hydroxybenzoate sulfate partially mediated the effect between Axin-1 and AD (20.10%). Pregnenetriol sulfate partially mediated the effect between CXCL11 and AD (18.20%). The ratio of spermidine to ornithine played an important mediating role between IL-12β and AD, with a mediating effect of 43.40%. Conclusion This study reveals how specific inflammatory proteins influence AD risk via plasma metabolites and provides genetic evidence for inflammatory-metabolic interactions in AD to facilitate the identification of potential biomarkers and targets for early detection and intervention of AD.

Key words: inflammatory proteins, plasma metabolites, mendelian randomization, Alzheimer's disease, mediating effect

Meimei CHEN, Ruina HUANG, Zhaoyang YANG. Plasma metabolites mediates the causal effect of inflammatory proteins on Alzheimer's disease: a Mendelian randomization study[J]. Journal of Southern Medical University, 2026, 46(2): 278-285.

Figures/Tables 9

References 36

[1]	Scheltens P, De Strooper B, Kivipelto M, et al. Alzheimer’s disease[J]. Lancet, 2021, 397(10284): 1577-90. doi：10.1016/s0140-6736(20)32205-4
[2]	GBD 2019 Dementia Forecasting Collaborators. Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the Global Burden of Disease Study 2019[J]. Lancet Public Health, 2022, 7(2): e105-25.
[3]	Leng FD, Edison P. Neuroinflammation and microglial activation in Alzheimer disease: where do we go from here[J]? Nat Rev Neurol, 2021, 17(3): 157-72. doi：10.1038/s41582-020-00435-y
[4]	Arnold SE, Arvanitakis Z, Macauley-Rambach SL, et al. Brain insulin resistance in type 2 diabetes and Alzheimer disease: concepts and conundrums[J]. Nat Rev Neurol, 2018, 14(3): 168-81. doi：10.1038/nrneurol.2017.185
[5]	Brandl S, Reindl M. Blood-brain barrier breakdown in neuroinflammation: current in vitro models[J]. Int J Mol Sci, 2023, 24(16): 12699. doi：10.3390/ijms241612699
[6]	Ponce-Lopez T. Peripheral inflammation and insulin resistance: their impact on blood-brain barrier integrity and Glia activation in Alzheimer's disease[J]. Int J Mol Sci, 2025, 26(9): 4209. doi：10.3390/ijms26094209
[7]	Amelimojarad M, Amelimojarad M, Cui XN. The emerging role of brain neuroinflammatory responses in Alzheimer's disease[J]. Front Aging Neurosci, 2024, 16: 1391517. doi：10.3389/fnagi.2024.1391517
[8]	Hegazy SH, Thomassen JQ, Rasmussen IJ, et al. C-reactive protein levels and risk of dementia-Observational and genetic studies of 111, 242 individuals from the general population[J]. Alzheimers Dement, 2022, 18(11): 2262-71. doi：10.1002/alz.12568
[9]	Shao YP, Ouyang Y, Li TB, et al. Alteration of metabolic profile and potential biomarkers in the plasma of Alzheimer's disease[J]. Aging Dis, 2020, 11(6): 1459-70. doi：10.14336/ad.2020.0217
[10]	Borkowski K, Seyfried NT, Arnold M, et al. Integration of plasma and CSF metabolomics with CSF proteomic reveals novel associations between lipid mediators and central nervous system vascular and energy metabolism[J]. Sci Rep, 2023, 13(1): 13752. doi：10.1038/s41598-023-39737-8
[11]	Shen M, Zhang LL, Chen C, et al. Investigating the causal relationship between immune cell and Alzheimer's disease: a mendelian randomization analysis[J]. BMC Neurol, 2024, 24(1): 98. doi：10.1186/s12883-024-03599-y
[12]	Li YC, Yao YL, Wu Y. Causal relationships between plasma proteins and Alzheimer's disease using bidirectional Mendelian rando-mization[J]. J Alzheimers Dis, 2025, 106(2): 765-73. doi：10.1177/13872877251345151
[13]	Zhao JH, Stacey D, Eriksson N, et al. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets[J]. Nat Immunol, 2023, 24(9): 1540-51. doi：10.1038/s41590-023-01588-w
[14]	Chen YH, Lu TY, Pettersson-Kymmer U, et al. Genomic atlas of the plasma metabolome prioritizes metabolites implicated in human diseases[J]. Nat Genet, 2023, 55(1): 44-53. doi：10.1038/s41588-022-01270-1
[15]	Papadimitriou N, Dimou N, Tsilidis KK, et al. Physical activity and risks of breast and colorectal cancer: a Mendelian randomisation analysis[J]. Nat Commun, 2020, 11(1): 597.
[16]	李文婕, 洪耀南, 黄蕊, 等. 自身免疫性疾病是再生障碍性贫血的危险因素: 一项孟德尔随机化分析[J]. 南方医科大学学报, 2025, 45(4): 871-9.
[17]	Carter AR, Sanderson E, Hammerton G, et al. Mendelian randomisation for mediation analysis: current methods and challenges for implementation[J]. Eur J Epidemiol, 2021, 36(5): 465-78. doi：10.1007/s10654-021-00757-1
[18]	Qiu L, Sun YX, Ning HM, et al. The scaffold protein AXIN1 gene ontology, signal network, and physiological function[J]. Cell Commun Signal, 2024, 22(1): 77. doi：10.1186/s12964-024-01482-4
[19]	Tapia-Rojas C, Inestrosa NC. Wnt signaling loss accelerates the appearance of neuropathological hallmarks of Alzheimer’s disease in J20-APP transgenic and wild-type mice[J]. J Neurochem, 2018, 144(4): 443-65. doi：10.1111/jnc.14278
[20]	Tapia-Rojas C, Burgos PV, Inestrosa NC. Inhibition of Wnt signaling induces amyloidogenic processing of amyloid precursor protein and the production and aggregation of Amyloid-β (Aβ)₄₂ peptides[J]. J Neurochem, 2016, 139(6): 1175-91. doi：10.1111/jnc.13873
[21]	Koper OM, Kamińska J, Sawicki K, et al. CXCL9, CXCL10, CXCL11, and their receptor (CXCR3) in neuroinflammation and neurodegeneration[J]. Adv Clin Exp Med, 2018, 27(6): 849-56. doi：10.17219/acem/68846
[22]	Wojcieszak J, Kuczyńska K, Zawilska JB. Role of chemokines in the development and progression of Alzheimer's disease[J]. J Mol Neurosci, 2022, 72(9): 1929-51. doi：10.1007/s12031-022-02047-1
[23]	Wang XM, Zhang YY, Wang S, et al. The role of CXCR3 and its ligands in cancer[J]. Front Oncol, 2022, 12: 1022688. doi：10.3389/fonc.2022.1022688
[24]	Moore TC, Hasenkrug KJ. B-cell control of regulatory T cells in friend virus infection[J]. J Mol Biol, 2021, 433(1): 166583. doi：10.1016/j.jmb.2020.06.022
[25]	Ciccocioppo F, Lanuti P, Pierdomenico L, et al. The characterization of regulatory T-cell profiles in Alzheimer's disease and multiple sclerosis[J]. Sci Rep, 2019, 9: 8788. doi：10.1038/s41598-019-45433-3
[26]	Machhi J, Yeapuri P, Lu YM, et al. CD4⁺ effector T cells accelerate Alzheimer's disease in mice[J]. J Neuroinflammation, 2021, 18(1): 272. doi：10.1186/s12974-021-02308-7
[27]	Rentzos M, Paraskevas GP, Kapaki E, et al. Interleukin-12 is reduced in cerebrospinal fluid of patients with Alzheimer's disease and frontotemporal dementia[J]. J Neurol Sci, 2006, 249(2): 110-4. doi：10.1016/j.jns.2006.05.063
[28]	Yang HS, Zhang C, Carlyle BC, et al. Plasma IL-12/IFN‑γ axis predicts cognitive trajectories in cognitively unimpaired older adults[J]. Alzheimers Dement, 2022, 18(4): 645-53. doi：10.1002/alz.12399
[29]	Zhu XC, Tan L, Jiang T, et al. Association of IL-12A and IL-12B polymorphisms with Alzheimer's disease susceptibility in a Han Chinese population[J]. J Neuroimmunol, 2014, 274(1/2): 180-4. doi：10.1016/j.jneuroim.2014.06.026
[30]	Breitkreuz-Korff O, Tscheik C, Del Vecchio G, et al. M01 as a novel drug enhancer for specifically targeting the blood-brain barrier[J]. J Control Release, 2021, 338: 137-48. doi：10.1016/j.jconrel.2021.08.014
[31]	Wei DF, Xu M, Cui WT, et al. Research progress on the role of Claudin family proteins in mediating blood-brain barrier selective permeability in tumor metastasis[J]. Am J Transl Res, 2025, 17(4): 2411-21. doi：10.62347/gggx3909
[32]	Vallée M, Mayo W, Le Moal M. Role of pregnenolone, dehydroepiandrosterone and their sulfate esters on learning and memory in cognitive aging[J]. Brain Res Brain Res Rev, 2001, 37(1/2/3): 301-12. doi：10.1016/s0165-0173(01)00135-7
[33]	王丹, 鲁映青, 于榕. 神经甾体的合成和代谢及其对神经系统作用[J]. 中国临床药理学与治疗学, 2004, 9(7): 725-9.
[34]	Ghosh I, Sankhe R, Mudgal J, et al. Spermidine, an autophagy inducer, as a therapeutic strategy in neurological disorders[J]. Neuropeptides, 2020, 83: 102083. doi：10.1016/j.npep.2020.102083
[35]	Ju YH, Bhalla M, Hyeon SJ, et al. Astrocytic urea cycle detoxifies Aβ-derived ammonia while impairing memory in Alzheimer's disease[J]. Cell Metab, 2022, 34(8): 1104-20.e8. doi：10.1016/j.cmet.2022.05.011
[36]	Mein H, Jing Y, Ahmad F, et al. Altered brain arginine metabolism and polyamine system in a P301S tauopathy mouse model: a time-course study[J]. Int J Mol Sci, 2022, 23(11): 6039. doi：10.3390/ijms23116039

Type	Exposure	Outcome	Nsnp	Beta	SE	P	OR (95% CI)
Inflammation proteins	Axin-1	AD	13	0.079	0.035	0.026	1.082 (1.009-1.159)
	C-X-C motif chemokine 11	AD	39	-0.050	0.020	0.026	0.951 (0.914-0.990)
	Interleukin-12 subunit beta	AD	31	-0.041	0.018	0.026	0.959 (0.926-0.994)
Blood metabolites	Methyl-4-hydroxybenzoate sulfate	AD	21	0.083	0.025	0.003	1.086 (1.034-1.141)
	Epiandrosterone sulfate	AD	24	-0.038	0.014	0.006	0.963 (0.938-0.989)
	11beta-hydroxyandrosterone glucuronide	AD	29	-0.062	0.019	0.003	0.940 (0.906-0.976)
	Pregnenetriol sulfate	AD	8	0.092	0.033	0.006	1.096 (1.027-1.169)
	Palmitoleoylcarnitine (C16:1)	AD	19	0.083	0.026	0.003	1.087 (1.034-1.143)
	Spermidine to ornithine ratio	AD	4	-0.168	0.057	0.006	0.845 (0.756-0.945)
	Phosphate to threonine ratio	AD	7	0.141	0.050	0.006	1.151 (1.043-1.271)
	Citrulline to phosphate ratio	AD	33	-0.053	0.021	0.010	0.948 (0.911-0.987)

Type	Exposure	Outcome	Nsnp	Beta	SE	P	OR (95% CI)
Inflammation proteins	Axin-1	AD	13	0.079	0.035	0.026	1.082 (1.009-1.159)
	C-X-C motif chemokine 11	AD	39	-0.050	0.020	0.026	0.951 (0.914-0.990)
	Interleukin-12 subunit beta	AD	31	-0.041	0.018	0.026	0.959 (0.926-0.994)
Blood metabolites	Methyl-4-hydroxybenzoate sulfate	AD	21	0.083	0.025	0.003	1.086 (1.034-1.141)
	Epiandrosterone sulfate	AD	24	-0.038	0.014	0.006	0.963 (0.938-0.989)
	11beta-hydroxyandrosterone glucuronide	AD	29	-0.062	0.019	0.003	0.940 (0.906-0.976)
	Pregnenetriol sulfate	AD	8	0.092	0.033	0.006	1.096 (1.027-1.169)
	Palmitoleoylcarnitine (C16:1)	AD	19	0.083	0.026	0.003	1.087 (1.034-1.143)
	Spermidine to ornithine ratio	AD	4	-0.168	0.057	0.006	0.845 (0.756-0.945)
	Phosphate to threonine ratio	AD	7	0.141	0.050	0.006	1.151 (1.043-1.271)
	Citrulline to phosphate ratio	AD	33	-0.053	0.021	0.010	0.948 (0.911-0.987)

Outcome	Exposure	MR Egger		Inverse variance weighted
Outcome	Exposure	Q	Q_P	Q	Q_P
AD	Axin-1	11.388	0.411	11.440	0.492
AD	C-X-C motif chemokine 11	40.637	0.313	41.066	0.338
AD	Interleukin-12 subunit beta	30.999	0.365	31.852	0.374

Outcome	Exposure	MR Egger		Inverse variance weighted
Outcome	Exposure	Q	Q_P	Q	Q_P
AD	Axin-1	11.388	0.411	11.440	0.492
AD	C-X-C motif chemokine 11	40.637	0.313	41.066	0.338
AD	Interleukin-12 subunit beta	30.999	0.365	31.852	0.374

Outcome	Exposure	MR-Egger intercept	SE	P
AD	Axin-1	-0.002	0.009	0.826
AD	C-X-C motif chemokine 11	-0.004	0.006	0.536
AD	Interleukin-12 subunit beta	-0.004	0.004	0.379