473种肠道菌群与阿尔茨海默症：基于233种循环代谢物的孟德尔随机化中介分析

doi:10.12122/j.issn.1673-4254.2026.08.20

摘要/Abstract

摘要：

目的孟德尔随机化（MR）探究肠道菌群（GM）与阿尔茨海默症（AD）的因果关联及循环代谢物的介导作用，为AD相关生物标志物和干预靶点提供遗传学依据。方法基于全基因组关联分析数据，采用两样本MR设计开展研究。筛选473种GM及233种循环代谢物的有效工具变量，通过R软件进行统计分析，以逆方差加权法作为主要分析方法，MR-Egger法和加权中位数法作为补充验证；结合多种敏感性分析评估结果稳定性，并通过反向MR分析验证因果关系方向；再进一步采用中介MR分析明确循环代谢物的潜在介导效应。结果马赛阴性杆菌丰度与AD发病风险之间存在正向因果关联（OR=1.204，95% CI：1.020-1.421，P=0.028），敏感性分析结果稳定可靠（P>0.05），反向MR分析未发现AD对马赛阴性杆菌丰度存在显著因果效应（P=0.678）。中介MR分析显示，在马赛阴性杆菌对AD的总效应中，经极小型极低密度脂蛋白中游离胆固醇与总脂质比率（v-VLDL FC/TL）介导的间接效应占6.63%。结论从遗传因果推断层面提示，马赛阴性杆菌可能与AD发病风险升高相关，且v-VLDL FC/TL可能在其中发挥部分中介作用，这为AD的预防和治疗提供了潜在靶点。

关键词: 肠道菌群, 循环代谢物, 阿尔茨海默症, 孟德尔随机化, 中介效应

Abstract:

Objective To investigate the causal associations between gut microbiota (GM) and Alzheimer's disease (AD) and the mediating role of circulating metabolites using Mendelian randomization (MR) analysis. Methods A two-sample MR analysis was conducted based on genome-wide association study (GWAS) summary data. Valid instrumental variables for 473 GM taxa and 233 circulating metabolites were selected, and the inverse-variance weighted (IVW) method was used as the primary analytical approach, with MR-Egger regression and weighted median method as the complementary analyses. Multiple sensitivity analyses were conducted to assess the robustness of the results, and reverse MR analyses were used to verify the direction of causality. Mediation MR analyses were performed to determine the mediating effects of the circulating metabolites. Results A positive causal association was identified between the abundance of Negativibacillus massiliensis and the risk of AD (OR=1.204, 95% CI: 1.020-1.421,P=0.028), and the results were stable and reliable as confirmed by sensitivity analyses (P>0.05). Reverse MR analysis revealed no significant causal effect of AD on the abundance of Negativibacillus massiliensis (P=0.678). Mediation MR analysis showed that the indirect effect mediated by free cholesterol to total lipid ratio in very small very-low-density lipoprotein (v-VLDL FC/TL) accounted for 6.63% of the total effect of Negativibacillus massiliensis on AD. Conclusion From a genetic causal inference perspective, Negativibacillus massiliensis is likely associated with an increased risk of AD, and v-VLDL FC/TL may partially mediate this association, suggesting their potential as targets for AD prevention and treatment.

Key words: gut microbiota, circulating metabolites, Alzheimer's disease, Mendelian randomization, mediation effect

李德臣, 李彦杰, 杨新宇, 王亚文, 渠燕飞. 473种肠道菌群与阿尔茨海默症：基于233种循环代谢物的孟德尔随机化中介分析[J]. 南方医科大学学报, 2026, 46(8): 1926-1935.

Dechen LI, Yanjie LI, Xinyu YANG, Yawen WANG, Yanfei QU. Association between 473 gut microbiota and Alzheimer's disease: a Mendelian randomization mediation analysis of 233 circulating metabolites[J]. Journal of Southern Medical University, 2026, 46(8): 1926-1935.

图/表 9

参考文献 38

[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]	Twarowski B, Herbet M. Inflammatory processes in Alzheimer's disease-pathomechanism, diagnosis and treatment: a review[J]. Int J Mol Sci, 2023, 24(7): 6518. doi：10.3390/ijms24076518
[3]	Ossenkoppele R, van der Kant R, Hansson O. Tau biomarkers in Alzheimer’s disease: towards implementation in clinical practice and trials[J]. Lancet Neurol, 2022, 21(8): 726-34. doi：10.1016/s1474-4422(22)00168-5
[4]	Rostagno AA. Pathogenesis of Alzheimer's disease[J]. Int J Mol Sci, 2022, 24(1): 107. doi：10.3390/ijms24010107
[5]	Bomasang-Layno E, Bronsther R. Diagnosis and treatment of Alzheimer's disease: : an update[J]. Dela J Public Health, 2021, 7(4): 74-85. doi：10.32481/djph.2021.09.009
[6]	Li SP, Han WF, He QC, et al. Relationship between intestinal microflora and hepatocellular cancer based on gut-liver axis theory[J]. Contrast Media Mol Imaging, 2022, 2022: 6533628. doi：10.1155/2022/6533628
[7]	Socała K, Doboszewska U, Szopa A, et al. The role of microbiota-gut-brain axis in neuropsychiatric and neurological disorders[J]. Pharmacol Res, 2021, 172: 105840. doi：10.1016/j.phrs.2021.105840
[8]	O’Riordan KJ, Moloney GM, Keane L, et al. The gut microbiota-immune-brain axis: Therapeutic implications[J]. Cell Rep Med, 2025, 6(3): 101982. doi：10.1016/j.xcrm.2025.101982
[9]	Lu X, Xue ZY, Qian Y, et al. Changes in intestinal microflora and its metabolites underlie the cognitive impairment in preterm rats[J]. Front Cell Infect Microbiol, 2022, 12: 945851. doi：10.3389/fcimb.2022.945851
[10]	Silva YP, Bernardi A, Frozza RL. The role of short-chain fatty acids from gut microbiota in gut-brain communication[J]. Front Endocrinol, 2020, 11: 25. doi：10.3389/fendo.2020.00025
[11]	He M, Wei WQ, Zhang YC, et al. Gut microbial metabolites SCFAs and chronic kidney disease[J]. J Transl Med, 2024, 22(1): 172. doi：10.1186/s12967-024-04974-6
[12]	Wang YL, Li L, Zhao XD, et al. Intestinal microflora changes in patients with mild Alzheimer's disease in a Chinese cohort[J]. J Alzheimers Dis, 2022, 88(2): 563-75. doi：10.3233/jad-220076
[13]	Elmas A, Spehar K, Do R, et al. Associations of circulating biomarkers with disease risks: a two-sample Mendelian randomization study[J]. Int J Mol Sci, 2024, 25(13): 7376. doi：10.3390/ijms25137376
[14]	Bhattacharyya U, John J, Lam M, et al. Circulating blood-based proteins in psychopathology and cognition: a Mendelian randomization study[J]. JAMA Psychiatry, 2025, 82(5): 481-91. doi：10.1001/jamapsychiatry.2025.0033
[15]	Lv YC, Cheng X, Dong Q. SGLT1 and SGLT2 inhibition, circulating metabolites, and cerebral small vessel disease: a mediation Mendelian Randomization study[J]. Cardiovasc Diabetol, 2024, 23(1): 157. doi：10.1186/s12933-024-02255-6
[16]	Huang SY, Yang YX, Zhang YR, et al. Investigating causal relations between circulating metabolites and Alzheimer's disease: a Mendelian randomization study[J]. J Alzheimers Dis, 2022, 87(1): 463-77. doi：10.3233/jad-220050
[17]	Tettevi EJ, Simpong DL, Maina M, et al. Antibiotic-induced gut dysbiosis modulates Alzheimer's disease-associated gene expression and protein aggregation in 3xTg-AD mice via the gut–brain axis[J]. Brain Behav, 2025, 15(10): e70946. doi：10.1002/brb3.70946
[18]	雷雯慧, 高婕, 徐鹏, 等. 阿尔茨海默病肠道细菌和真菌菌群结构和组成改变研究[J]. 中国微生态学杂志, 2024, 36(5): 577-83. doi：10.13381/j.cnki.cjm.202405013
[19]	Loh JS, Mak WQ, Tan LKS, et al. Microbiota-gut-brain axis and its therapeutic applications in neurodegenerative diseases[J]. Signal Transduct Target Ther, 2024, 9(1): 37. doi：10.1038/s41392-024-01743-1
[20]	Qin YW, Havulinna AS, Liu Y, et al. Combined effects of host genetics and diet on human gut microbiota and incident disease in a single population cohort[J]. Nat Genet, 2022, 54(2): 134-42. doi：10.1038/s41588-021-00991-z
[21]	Karjalainen MK, Karthikeyan S, Oliver-Williams C, et al. Genome-wide characterization of circulating metabolic biomarkers[J]. Nature, 2024, 628(8006): 130-8.
[22]	Li PS, Wang HY, Guo L, et al. Association between gut microbiota and preeclampsia-eclampsia: a two-sample Mendelian random-ization study[J]. BMC Med, 2022, 20(1): 443. doi：10.1186/s12916-022-02657-x
[23]	Wu J, Li CS, Huang WY, et al. Gut microbiota promote the propagation of pathologic α-syn from gut to brain in a gut-originated mouse model of Parkinson's disease[J]. Brain Behav Immun, 2025, 128: 152-69. doi：10.1016/j.bbi.2025.04.001
[24]	Yadav S, Raj RG. Parkinson's disease and the gut microbiota connection: unveiling dysbiosis and exploring therapeutic horizons[J]. Neuroscience, 2025, 581: 1-15. doi：10.1016/j.neuroscience.2025.07.003
[25]	Yan ZZ, Yang F, Cao JW, et al. Alterations of gut microbiota and metabolome with Parkinson's disease[J]. Microb Pathog, 2021, 160: 105187. doi：10.1016/j.micpath.2021.105187
[26]	Kalyanaraman B, Cheng G, Hardy M. Gut microbiome, short-chain fatty acids, alpha-synuclein, neuroinflammation, and ROS/RNS: Relevance to Parkinson's disease and therapeutic implications[J]. Redox Biol, 2024, 71: 103092. doi：10.1016/j.redox.2024.103092
[27]	Aho VTE, Houser MC, Pereira PAB, et al. Relationships of gut microbiota, short-chain fatty acids, inflammation, and the gut barrier in Parkinson's disease[J]. Mol Neurodegener, 2021, 16(1): 6. doi：10.1186/s13024-021-00427-6
[28]	Chen Z, Chen ZY, Jin XL. Mendelian randomization supports causality between overweight status and accelerated aging[J]. Aging Cell, 2023, 22(8): e13899. doi：10.1111/acel.13899
[29]	Lv XF, Liang FQ, Liu SS, et al. Causal relationship between diet and knee osteoarthritis: a Mendelian randomization analysis[J]. PLoS One, 2024, 19(1): e0297269. doi：10.1371/journal.pone.0297269
[30]	Liang Z, Zhao L, Lou Y, et al. Causal effects of circulating lipids and lipid-lowering drugs on the risk of epilepsy: a two-sample Mendelian randomization study[J]. QJM, 2023, 116(6): 421-8. doi：10.1093/qjmed/hcad048
[31]	Zhu X, Huang SJ, Kang WY, et al. Associations between polyunsaturated fatty acid concentrations and Parkinson's disease: a two-sample Mendelian randomization study[J]. Front Aging Neurosci, 2023, 15: 1123239. doi：10.3389/fnagi.2023.1123239
[32]	Ndongo S, Dubourg G, Bittar F, et al. Marseillibacter massiliensis gen. nov., sp. nov., a new bacterial genus isolated from the human gut[J]. New Microbes New Infect, 2017, 16: 30-1. doi：10.1016/j.nmni.2016.12.006
[33]	Tran VTA, Kang YJ, Kim HK, et al. Oral pathogenic bacteria-inducing neurodegenerative microgliosis in human neural cell platform[J]. Int J Mol Sci, 2021, 22(13): 6925. doi：10.3390/ijms22136925
[34]	Wu LC, Du LL, Ju QQ, et al. Silencing TLR4/MyD88/NF‑κB signaling pathway alleviated inflammation of corneal epithelial cells infected by ISE[J]. Inflammation, 2021, 44(2): 633-44. doi：10.1007/s10753-020-01363-1
[35]	Shi J, Wang F, Tang L, et al. Akkermansia muciniphila attenuates LPS-induced acute kidney injury by inhibiting TLR4/NF‑κB pathway[J]. FEMS Microbiol Lett, 2022, 369(1): fnac103. doi：10.1093/femsle/fnac103
[36]	Wu XY, Wei JJ, Ran W, et al. The gut microbiota-xanthurenic acid-aromatic hydrocarbon receptor axis mediates the anticolitic effects of trilobatin[J]. Adv Sci, 2025, 12(10): 2412234. doi：10.1002/advs.202412234
[37]	Yu X, Shen GX, Zhang Y, et al. Genetically predicted small dense low-density lipoprotein cholesterol and ischemic stroke subtype: multivariable Mendelian randomization study[J]. Front Endocrinol, 2024, 15: 1404234. doi：10.3389/fendo.2024.1404234
[38]	Hashemi M, Banerjee S, Lyubchenko YL. Free cholesterol accelerates aβ self-assembly on membranes at physiological concentration[J]. Int J Mol Sci, 2022, 23(5): 2803. doi：10.3390/ijms23052803

Rank	Outcome	Method	β	SE	P
Kingdom	Cyanobacteria	IVW	-0.016	0.009	0.082
Phylum	Actinobacteriota	IVW	0.011	0.018	0.548
Order	Francisellales	IVW	-0.005	0.007	0.447
Order	Parachlamydiales	IVW	-0.007	0.006	0.223
Order	Sporomusales	IVW	0.004	0.005	0.382
Family	Desulfovibrionaceae	IVW	0.008	0.015	0.608
Family	Treponemataceae	IVW	-0.011	0.016	0.513
Genus	Eubacterium Q	IVW	-0.006	0.020	0.766
Genus	Klebsiella	IVW	-0.002	0.006	0.690
Genus	Spirillospora	IVW	0.016	0.014	0.254
Genus	Azorhizobium	IVW	-0.001	0.006	0.850
Species	Bifidobacterium adolescentis	IVW	-0.012	0.008	0.135
Species	Negativibacillus massiliensis	IVW	0.006	0.015	0.678
Species	CAG-110	IVW	-0.004	0.028	0.898
Species	CAG-449	IVW	-0.006	0.009	0.496
Species	CAG-488	IVW	0.001	0.033	0.996
Species	Massiliomicrobiota sp002160815	IVW	-0.001	0.013	0.974
Species	Bifidobacterium pseudocatenulatum	IVW	0.001	0.005	0.807
Species	GCA-900066135 sp900066135	IVW	0.001	0.007	0.990

Rank	Outcome	Method	β	SE	P
Kingdom	Cyanobacteria	IVW	-0.016	0.009	0.082
Phylum	Actinobacteriota	IVW	0.011	0.018	0.548
Order	Francisellales	IVW	-0.005	0.007	0.447
Order	Parachlamydiales	IVW	-0.007	0.006	0.223
Order	Sporomusales	IVW	0.004	0.005	0.382
Family	Desulfovibrionaceae	IVW	0.008	0.015	0.608
Family	Treponemataceae	IVW	-0.011	0.016	0.513
Genus	Eubacterium Q	IVW	-0.006	0.020	0.766
Genus	Klebsiella	IVW	-0.002	0.006	0.690
Genus	Spirillospora	IVW	0.016	0.014	0.254
Genus	Azorhizobium	IVW	-0.001	0.006	0.850
Species	Bifidobacterium adolescentis	IVW	-0.012	0.008	0.135
Species	Negativibacillus massiliensis	IVW	0.006	0.015	0.678
Species	CAG-110	IVW	-0.004	0.028	0.898
Species	CAG-449	IVW	-0.006	0.009	0.496
Species	CAG-488	IVW	0.001	0.033	0.996
Species	Massiliomicrobiota sp002160815	IVW	-0.001	0.013	0.974
Species	Bifidobacterium pseudocatenulatum	IVW	0.001	0.005	0.807
Species	GCA-900066135 sp900066135	IVW	0.001	0.007	0.990

Exposure	Outcome	Methods	β	SE	P
AD	v-VLDL FC/TL	MR Egger	-0.009	0.012	0.457
AD	v-VLDL FC/TL	Weighted median	-0.016	0.007	0.135
AD	v-VLDL FC/TL	IVW	-0.005	0.008	0.547

Exposure	Outcome	Methods	β	SE	P
AD	v-VLDL FC/TL	MR Egger	-0.009	0.012	0.457
AD	v-VLDL FC/TL	Weighted median	-0.016	0.007	0.135
AD	v-VLDL FC/TL	IVW	-0.005	0.008	0.547

Exposure	SNPs	Beta	P	F
Negativibacillus massiliensis	rs117450486	0.013	4.1e-08	30.106
Negativibacillus massiliensis	rs1242927	0.002	1.9e-06	22.731
Negativibacillus massiliensis	rs1326749	0.018	8.5e-06	19.813
Negativibacillus massiliensis	rs1497993	0.043	8.7e-06	19.786
Negativibacillus massiliensis	rs16851304	-0.009	7.3e-06	20.108
Negativibacillus massiliensis	rs17552264	0.018	6.6e-06	20.294
Negativibacillus massiliensis	rs188774576	0.029	5.1e-06	20.801
Negativibacillus massiliensis	rs2641348	-0.021	5.5e-06	20.648
Negativibacillus massiliensis	rs2763223	-0.017	8.7e-06	19.774
Negativibacillus massiliensis	rs28634441	0.009	4e-06	21.271
Negativibacillus massiliensis	rs36124608	0.039	9e-06	19.719
Negativibacillus massiliensis	rs4898867	-0.007	6.6e-06	20.299
Negativibacillus massiliensis	rs4979255	-0.016	5.3e-06	20.716
Negativibacillus massiliensis	rs5771271	-0.023	3.7e-06	21.393
Negativibacillus massiliensis	rs73279647	-0.116	4.7e-06	20.969
Negativibacillus massiliensis	rs74032121	-0.001	2.7e-06	22.055
Negativibacillus massiliensis	rs79831182	-0.016	9.4e-06	19.621
Negativibacillus massiliensis	rs9384293	0.039	8.4e-06	19.845