免疫细胞表型与梅尼埃病的因果关联：一项双向双样本孟德尔随机化分析与动物实验研究

doi:10.12122/j.issn.1673-4254.2026.07.20

摘要/Abstract

摘要：

目的通过孟德尔随机化与动物实验探讨731种免疫细胞表型与梅尼埃病（MD）的因果关系。方法基于免疫细胞表型和MD的全基因组关联研究（GWAS）数据分别为暴露因素与结局变量，主要用逆方差加权法（IVW）分析，辅以MR-Egger、加权中位数等方法，并行异质性、多效性及留一法敏感性分析。动物实验中，将30只C57BL/6小鼠随机分为对照组、模型组及抗CD20单抗组，10只/组。采用耳后皮下注射脂多糖（LPS）制备膜迷路积水模型，抗CD20单抗组造模前尾静脉注射药物干预。通过听性脑干反应（ABR）、水迷宫及旷场实验评估小鼠听觉及前庭功能；HE染色观察内耳组织病理变化；免疫荧光检测CD19、CD27、CD3、CD39等免疫细胞标志物；qPCR、Western blotting、免疫组化检测TLR4/NF-κB通路及AQP3、ZO-1表达；ELISA检测血清中TNF-α、IL-6、IL-1β炎症因子水平。结果 MR分析显示，CD19⁺B细胞亚群（OR=1.123~1.204）、CD27⁺IgD⁻CD38⁺B细胞（OR=1.248，95% CI：1.096~1.421，P=8.14×10^-4）、CD3⁺T细胞亚群（OR=1.089~1.123）为主要危险因素；CD39⁺ CD4⁺ T细胞（OR=0.902，95% CI：0.838~0.972，P=6.46×10^-3）为核心保护因素，敏感性分析提示结果稳健。且不存在异质性或水平多效性（P<0.05）或反向因果关系。动物实验中，模型组小鼠表现出明显听觉及前庭功能障碍；内耳膜迷路扩张积水；耳蜗组织中CD19、CD27、CD3表达增多，CD39、AQP3、ZO-1表达减少；TLR4/NF-κB通路激活，血清TNF-α、IL-6、IL-1β水平升高（P<0.05）。抗CD20单抗组上述异常得到改善。结论 MR分析与动物实验均证实免疫细胞表型异常与MD存在因果关联，B细胞异常活化介导的免疫失衡是MD的重要因素，其通过激活TLR4/NF-κB通路，引发内耳炎症，下调AQP3、ZO-1表达，致水液代谢及屏障功能异常，导致听觉与前庭功能损伤；抗CD20单抗通过耗竭异常活化B细胞，逆转免疫失衡及下游分子通路异常，改善MD小鼠病理损伤与功能障碍，为MD免疫靶向治疗提供新依据。

关键词: 免疫细胞表型, 梅尼埃病, 孟德尔随机化, 因果关联

Abstract:

Objective To explore the causal relationship between 731 immune cell phenotypes and Ménière's disease (MD) using Mendelian randomization (MR) analysis and experimental validation in mice. Methods Based on GWAS data and with immune cell phenotypes and MD as the exposure and outcome variables respectively, we conducted MR analysis using inverse variance weighting (IVW) supplemented by MR-Egger and weighted median methods. Heterogeneity, pleiotropy and leave-one-out sensitivity analyses were performed. Thirty C57BL/6 mice were randomly divided into control, model and anti-CD20 mAb groups, and in the latter two groups, endolymphatic hydrops models were established by LPS injection behind the ear; treatment with the anti-CD20 mAb was administered via tail vein injection before modeling. Auditory and vestibular functions of the mice were evaluated by auditory brainstem response (ABR), water maze and open field tests. HE staining, immunofluorescence, qPCR, Western blotting, immunohistochemistry and ELISA were used to examine inner ear pathology, expressions of immune markers, TLR4/NF-κB pathway, AQP3, and ZO-1, and serum levels of inflammatory factors. Results MR analysis identified CD19⁺B cells, CD27⁺IgD⁻CD38⁺B cells and CD3⁺T cells as the major risk factors, and CD39⁺CD4⁺T cells as the core protective factor. Sensitivity analysis confirmed robustness of the results without heterogeneity, pleiotropy or reverse causality. The mouse models of endolymphatic hydrops showed obvious auditory/vestibular dysfunction, endolymphatic hydrops, upregulated CD19, CD27 and CD3 expressions, downregulated CD39, AQP3 and ZO-1 expressions, activated TLR4/NF-κB pathway, and elevated serum levels of inflammatory factors. All these abnormalities were significantly improved in anti-CD20 mAb treatment group. Conclusion Immune cell phenotype abnormality is causally associated with MD, and B cell activation-mediated immune imbalance contributes to MD via the TLR4/NF-κB pathway, inducing inner ear inflammation and abnormal fluid/barrier function. Anti-CD20 mAb ameliorates MD pathology and dysfunction by depleting activated B cells, which provides new evidence for MD immunotherapy.

Key words: immune cell phenotype, Ménière's disease, Mendelian randomization, causal association

杨琳, 于嘉祥, 韩磊, 戴俭宇. 免疫细胞表型与梅尼埃病的因果关联：一项双向双样本孟德尔随机化分析与动物实验研究[J]. 南方医科大学学报, 2026, 46(7): 1671-1684.

Lin YANG, Jiaxiang YU, Lei HAN, Jianyu DAI. Causal associations between immune cell phenotypes and Ménière's disease: a bidirectional two-sample Mendelian randomization study and validation in mice[J]. Journal of Southern Medical University, 2026, 46(7): 1671-1684.

图/表 14

图1 孟德尔随机化分析假设以及流程图

Fig.1 Hypotheses and flowchart of Mendelian randomization analysis.

表2 每一步筛选的单核苷酸多态性（SNP）数量

Tab.2 Number of single-nucleotide polymorphisms (SNPs) at each screening step

Immune traits	ID	No. SNPs after LD	No. SNPs after F>10	No. final IVs
IgD⁺ B cell absolute count	GCST90001400	13	13	13
IgD⁺ CD24⁺ B cell absolute count	GCST90001412	22	22	22
IgD^- CD24^- B cell absolute count	GCST90001414	15	15	15
IgD^- CD24^- B cell %lymphocyte	GCST90001440	21	21	21
CD62L^- dendritic cell absolute count	GCST90001462	14	14	14
CD86⁺ plasmacytoid dendritic cell %dendritic cell	GCST90001467	24	24	24
Activated CD4 regulatory T cell %CD4 regulatory T cell	GCST90001487	24	24	24
CD33^- HLA DR^- absolute count	GCST90001522	20	20	20
CD45RA^- CD4⁺ T cell %CD4⁺ T cell	GCST90001535	25	25	25
Naive CD4⁺ T cell absolute count	GCST90001540	22	22	22
Naive CD4⁺ T cell %CD4⁺ T cell	GCST90001541	30	30	30
Terminally differentiated CD4^- CD8^- T cell absolute count	GCST90001572	20	20	20
Terminally differentiated CD4^- CD8^- T cell %T cell	GCST90001574	21	21	21
CD39⁺ CD4⁺ T cell absolute count	GCST90001660	25	25	25
CD28⁺ CD45RA^- CD8dim T cell %CD8dim T cell	GCST90001668	20	20	20
CD28⁺ CD45RA^- CD8dim T cell absolute count	GCST90001669	16	16	16
CD25⁺⁺ CD8⁺ T cell %T cell	GCST90001679	15	15	15
CD19 on CD20^- B cell	GCST90001721	18	18	18
CD19 on IgD⁺ CD38^- B cell	GCST90001726	26	26	26
CD19 on IgD^- CD24^- B cell	GCST90001731	19	19	19
CD19 on IgD^- CD38^- B cell	GCST90001733	20	20	20
CD27 on IgD^- CD38⁺ B cell	GCST90001803	15	15	15
CD3 on effector memory CD8⁺ T cell	GCST90001839	18	18	18
CD3 on CD28⁺ CD45RA^- CD8⁺ T cell	GCST90001863	20	20	20
CD34 on hematopoietic stem cell	GCST90001870	13	13	13
CD16^- CD56 on HLA DR⁺ natural killer	GCST90001885	20	20	20
CCR7 on naive CD4⁺ T cell	GCST90001907	22	22	22
CD127 on CD8⁺ T cell	GCST90001927	23	23	23
CD127 on CD28^- CD8⁺ T cell	GCST90001930	19	19	19
FSC-A on HLA DR⁺ CD8⁺ T cell	GCST90001978	12	12	12
PDL-1 on CD14^- CD16⁺ monocyte	GCST90001999	18	18	18
CD4 on CD4⁺ T cell	GCST90002022	22	22	22
CD4 on naive CD4⁺ T cell	GCST90002024	19	19	19
CD4 on effector memory CD4⁺ T cell	GCST90002025	21	21	21
CD4 on terminally differentiated CD4⁺ T cell	GCST90002026	22	22	22
CD4 on CD45RA⁺ CD4⁺ T cell	GCST90002027	26	26	26
CD80 on myeloid dendritic cell	GCST90002035	24	24	24
CD80 on monocyte	GCST90002039	20	20	20
CD45 on CD33^- HLA DR^-	GCST90002045	15	15	15
CD45 on granulocytic myeloid-derived suppressor cells	GCST90002047	17	17	17
CD11c on CD62L⁺ myeloid dendritic cell	GCST90002088	22	22	22
HLA DR on myeloid dendritic cell	GCST90002104	14	14	14
HLA DR on plasmacytoid dendritic cell	GCST90002105	19	19	19
HLA DR on HLA DR⁺ CD8⁺ T cell	GCST90002115	17	17	17

表2 每一步筛选的单核苷酸多态性（SNP）数量

Tab.2 Number of single-nucleotide polymorphisms (SNPs) at each screening step

Immune traits	ID	No. SNPs after LD	No. SNPs after F>10	No. final IVs
IgD⁺ B cell absolute count	GCST90001400	13	13	13
IgD⁺ CD24⁺ B cell absolute count	GCST90001412	22	22	22
IgD^- CD24^- B cell absolute count	GCST90001414	15	15	15
IgD^- CD24^- B cell %lymphocyte	GCST90001440	21	21	21
CD62L^- dendritic cell absolute count	GCST90001462	14	14	14
CD86⁺ plasmacytoid dendritic cell %dendritic cell	GCST90001467	24	24	24
Activated CD4 regulatory T cell %CD4 regulatory T cell	GCST90001487	24	24	24
CD33^- HLA DR^- absolute count	GCST90001522	20	20	20
CD45RA^- CD4⁺ T cell %CD4⁺ T cell	GCST90001535	25	25	25
Naive CD4⁺ T cell absolute count	GCST90001540	22	22	22
Naive CD4⁺ T cell %CD4⁺ T cell	GCST90001541	30	30	30
Terminally differentiated CD4^- CD8^- T cell absolute count	GCST90001572	20	20	20
Terminally differentiated CD4^- CD8^- T cell %T cell	GCST90001574	21	21	21
CD39⁺ CD4⁺ T cell absolute count	GCST90001660	25	25	25
CD28⁺ CD45RA^- CD8dim T cell %CD8dim T cell	GCST90001668	20	20	20
CD28⁺ CD45RA^- CD8dim T cell absolute count	GCST90001669	16	16	16
CD25⁺⁺ CD8⁺ T cell %T cell	GCST90001679	15	15	15
CD19 on CD20^- B cell	GCST90001721	18	18	18
CD19 on IgD⁺ CD38^- B cell	GCST90001726	26	26	26
CD19 on IgD^- CD24^- B cell	GCST90001731	19	19	19
CD19 on IgD^- CD38^- B cell	GCST90001733	20	20	20
CD27 on IgD^- CD38⁺ B cell	GCST90001803	15	15	15
CD3 on effector memory CD8⁺ T cell	GCST90001839	18	18	18
CD3 on CD28⁺ CD45RA^- CD8⁺ T cell	GCST90001863	20	20	20
CD34 on hematopoietic stem cell	GCST90001870	13	13	13
CD16^- CD56 on HLA DR⁺ natural killer	GCST90001885	20	20	20
CCR7 on naive CD4⁺ T cell	GCST90001907	22	22	22
CD127 on CD8⁺ T cell	GCST90001927	23	23	23
CD127 on CD28^- CD8⁺ T cell	GCST90001930	19	19	19
FSC-A on HLA DR⁺ CD8⁺ T cell	GCST90001978	12	12	12
PDL-1 on CD14^- CD16⁺ monocyte	GCST90001999	18	18	18
CD4 on CD4⁺ T cell	GCST90002022	22	22	22
CD4 on naive CD4⁺ T cell	GCST90002024	19	19	19
CD4 on effector memory CD4⁺ T cell	GCST90002025	21	21	21
CD4 on terminally differentiated CD4⁺ T cell	GCST90002026	22	22	22
CD4 on CD45RA⁺ CD4⁺ T cell	GCST90002027	26	26	26
CD80 on myeloid dendritic cell	GCST90002035	24	24	24
CD80 on monocyte	GCST90002039	20	20	20
CD45 on CD33^- HLA DR^-	GCST90002045	15	15	15
CD45 on granulocytic myeloid-derived suppressor cells	GCST90002047	17	17	17
CD11c on CD62L⁺ myeloid dendritic cell	GCST90002088	22	22	22
HLA DR on myeloid dendritic cell	GCST90002104	14	14	14
HLA DR on plasmacytoid dendritic cell	GCST90002105	19	19	19
HLA DR on HLA DR⁺ CD8⁺ T cell	GCST90002115	17	17	17

图2 逆方差加权法的正向孟德尔随机化分析结果

Fig.2 Results of forward Mendelian randomization analysis using the inverse variance weighted method.

图3 各组小鼠 ABR 检测结果

Fig.3 Results of auditory brainstem response (ABR) test of the mice in each group (Mean±SD, n=20). *P<0.05 vs control group, #P<0.05 vs Model group.

图4 各组小鼠旷场实验结果比较

Fig.4 Comparison of open field test results among the 3 groups of mice. A: Open field trajectory plot. B: Open field heatmap. C: Total activity distance. D: Activity duration (Mean±SD, n=10). *P<0.05 vs control group, #P<0.05 vs Model group.

图5 各组小鼠Morris水迷宫实验结果比较

Fig.5 Comparison of Morris water maze test results among the 3 groups of mice. A: Morris Water Maze Trajectory Plot. B: Morris Water Maze Heatmap. C: Latency to platform. D: Distance to platform (Mean±SD, n=10).*P<0.05 vs control group, #P<0.05 vs Model group.

图6 各组小鼠耳蜗组织 HE 染色情况

Fig.6 HE staining of cochlear tissues in each group (Original magnification: ×200).

图7 各组小鼠耳蜗组织中CD19、CD27、CD3、CD39的表达

Fig.7 Expression of CD19, CD27, CD3, and CD39 in the cochlear tissues of mice in each group (×200). A: CD19 immunofluorescence staining. B: CD27 immunofluorescence staining. C: CD3 immunofluorescence staining. D: CD39 immunofluorescence staining. E,F: CD19, CD27, CD3 and CD39⁺ average fluorescence intensity (Mean±SD, n=8). *P<0.05 vs control group; #P<0.05 vs Model group.

图8 各组小鼠耳蜗组织中TLR4、NF-κB、AQP3、ZO-1 mRNA相对表达

Fig.8 Relative expressions of TLR4 (A), NF-κB p65 (B), AQP3 (C) and ZO-1 (D) mRNA in the cochlear tissue of the mice in each group (Mean±SD, n=8). *P<0.05 vs Control group; #P<0.05 vs Model group.

图9 Western blotting 检测各组小鼠耳蜗组织中TLR4、AQP3、ZO-1蛋白相对表达及p-NF-κB/NF-κB比值的比较

Fig.9 Relative expressions of TLR4, AQP3, ZO-1 proteins and the ratio of p-NF-κB/NF-κB in cochlear tissues of the mice in each group detected by Western blotting. A: Results of Western blotting of the proteins. B-D: Relative expressions of TLR4, AQP3 and ZO-1 proteins. E: Ratio of p-NF-κB/NF-κB. Data are presented as Mean±SD (n=8). *P<0.05 vs control group; #P<0.05 vs Model group.

图10 各组小鼠耳蜗组织中AQP3、ZO-1的表达

Fig.10 Expression of AQP3 and ZO-1 in mouse cochlear tissues in each group (×200). A: AQP3 immunohistochemistry. B: ZO-1 immunohistochemistry. C-D: A450 nm values of AQP3 and ZO-1 (Mean±SD, n=8). *P<0.05 vs control group; #P<0.05 vs Model group.

图11 各组小鼠血清中TNF-α、IL-6、IL-1β的含量

Fig.11 Serum levels of TNF-α (A), IL-6 (B) and IL-1β (C) in each group (Mean±SD, n=10). *P<0.05 vs control group; #P<0.05 vs Model group.

参考文献 32

[1]	Basura GJ, Adams ME, Monfared A, et al. Clinical practice guideline: ménière's disease[J]. Otolaryngol--head Neck Surg, 2020, 162(S2): S1-55. doi：10.1177/0194599820909438
[2]	中华耳鼻咽喉头颈外科杂志编辑委员会, 中华医学会耳鼻咽喉头颈外科学分会. 梅尼埃病诊断和治疗指南(2017)[J]. 中华耳鼻咽喉头颈外科杂志, 2017, 52(3): 167-72. doi：10.3760/cma.j.issn.1673-0860.2017.03.002
[3]	Thompson-Harvey A, Friedland DR, Adams JA, et al. The demographics of menière's disease: selection bias or differential susceptibility?[J]. Otol Neurotol, 2023, 44(2): e95-102. doi：10.1097/mao.0000000000003780
[4]	于慧前, 李华伟, 李庆忠. 2020版梅尼埃病临床实践指南解读[J]. 临床耳鼻咽喉头颈外科杂志, 2021, 35(5): 385-90. doi：10.13201/j.issn.2096-7993.2021.05.001
[5]	Webster KE, Galbraith K, Harrington-Benton NA, et al. Systemic pharmacological interventions for Ménière's disease[J]. Cochrane Database Syst Rev, 2023, 2(2): CD015171. doi：10.1002/14651858.cd015171.pub2
[6]	Lee A, Webster KE, George B, et al. Surgical interventions for ménière's disease[J]. Cochrane Database Syst Rev, 2023, 2(2): CD015249. doi：10.1002/14651858.cd015249.pub2
[7]	Lucas JC, Hong RS. Recent surgical advances and continued controversies in medically refractory Meniere's disease[J]. Curr Opin Otolaryngol Head Neck Surg, 2023, 31(5): 325-31. doi：10.1097/moo.0000000000000907
[8]	Lopez-Escamez JA, Liu Y. Epidemiology and genetics of Meniere's disease[J]. Curr Opin Neurol, 2024, 37(1): 88-94. doi：10.1097/wco.0000000000001227
[9]	Yuan VG, Xia AP, Santa Maria PL. Immunological mechanisms in Meniere's disease[J]. Front Immunol, 2025, 16: 1639916. doi：10.3389/fimmu.2025.1639916
[10]	Takeda T, Takeda S, Kakigi A. The clinical manifestation and treatment of Meniere's Disease from the viewpoint of the water homeostasis of the inner ear[J]. Auris Nasus Larynx, 2024, 51(6): 905-10. doi：10.1016/j.anl.2024.08.002
[11]	Lopez-Escamez JA, Vela J, Frejo L. Immune-related disorders associated with ménière's disease: a systematic review and meta-analysis[J]. Otolaryngol-head Neck Surg, 2023, 169(5): 1122-31. doi：10.1002/ohn.386
[12]	Jiang HL, Zhan YX, Zhao ML, et al. Immunological heterogeneity in Ménière's disease: CD4⁺ T cell subset profiling reveals three distinct Immunophenotypes[J]. J Neuroimmunol, 2025, 409: 578743. doi：10.1016/j.jneuroim.2025.578743
[13]	Flook M, Escalera-Balsera A, Rybakowska P, et al. Single-cell immune profiling of Meniere Disease patients[J]. Clin Immunol, 2023, 252: 109632. doi：10.1016/j.clim.2023.109632
[14]	Papadopoulou A, Åsvold BO, Burgess S, et al. Height, autoimmune thyroid disease, and thyroid cancer: a Mendelian randomization study[J]. Thyroid®, 2023, 33(12): 1476-82. doi：10.1089/thy.2023.0272
[15]	Larsson SC, Butterworth AS, Burgess S. Mendelian randomization for cardiovascular diseases: principles and applications[J]. Eur Heart J, 2023, 44(47): 4913-24. doi：10.1093/eurheartj/ehad736
[16]	潘兴旭, 张秉祺, 张智华, 等. 戈登杆菌属丰度降低与肾结石风险增加相关: 一项孟德尔随机化分析与动物实验研究[J]. 南方医科大学学报, 2025, 45(11): 2405-15.
[17]	Long YW, Tang LH, Zhou YY, et al. Causal relationship between gut microbiota and cancers: a two-sample Mendelian randomisation study[J]. BMC Med, 2023, 21(1): 66. doi：10.1186/s12916-023-02761-6
[18]	Kintu C, Soremekun O, Kamiza AB, et al. The causal effects of lipid traits on kidney function in Africans: bidirectional and multivariable Mendelian-randomization study[J]. EBioMedicine, 2023, 90: 104537. doi：10.1016/j.ebiom.2023.104537
[19]	Zhang DG, Yu WQ, Liu JH, et al. Serum/glucocorticoid-inducible kinase 1 deficiency induces NLRP3 inflammasome activation and autoinflammation of macrophages in a murine endolymphatic Hydrops model[J]. Nat Commun, 2023, 14(1): 1249. doi：10.1038/s41467-023-36949-4
[20]	石科, 李悦淇, 王君, 等. 100 dB SPL 8~16 kHz倍频带噪声暴露对小鼠听觉功能的影响[J]. 中华耳科学杂志, 2025(7): 863-8.
[21]	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.
[22]	张仪铃, 周梓意, 郭佳, 等. 免疫学视角下梅尼埃病的研究新进展[J]. 中华耳科学杂志, 2026, 24(2): 154-8. doi：10.3969/j.issn.1672-2922.2026.02.011
[23]	Mougiakakos D, Krönke G, Völkl S, et al. CD19-targeted CAR T cells in refractory systemic lupus erythematosus[J]. N Engl J Med, 2021, 385(6): 567-9. doi：10.1056/nejmc2107725
[24]	Chung MKY, Gong LQ, Kwong DL, et al. Functions of double-negative B cells in autoimmune diseases, infections, and cancers[J]. EMBO Mol Med, 2023, 15(9): e17341. doi：10.15252/emmm.202217341
[25]	Timperi E, Barnaba V. CD39 regulation and functions in T cells[J]. Int J Mol Sci, 2021, 22(15): 8068. doi：10.3390/ijms22158068
[26]	Liu DY, Hu XX, Chen ZY, et al. Key links in the physiological regulation of the immune system and disease induction: T cell receptor-CD3 complex[J]. Biochem Pharmacol, 2024, 227: 116441. doi：10.1016/j.bcp.2024.116441
[27]	Ly S, Nedosekin D, Wong HK. Review of an anti-CD20 monoclonal antibody for the treatment of autoimmune diseases of the skin[J]. Am J Clin Dermatol, 2023, 24(2): 247-73. doi：10.1007/s40257-022-00751-7
[28]	Monaghan NP, Shah S, Keith BA, et al. Proinflammatory cytokine profiles in menière's disease and vestibular migraine[J]. Otol Neurotol, 2025, 46(1): 88-95. doi：10.1097/mao.0000000000004372
[29]	Song J, Huang XH, Luo Y, et al. Short-term exposure to a high-humidity environment triggers intestinal inflammation via AQP3[J]. Front Immunol, 2025, 16: 1563602. doi：10.3389/fimmu.2025.1563602
[30]	陈俊宏, 江英, 周利, 等. 连接蛋白在小鼠膜迷路积水模型中对耳蜗血-迷路屏障的影响[J]. 四川大学学报: 医学版, 2022, 53(2): 291-6. doi：10.12182/20220360207
[31]	任银先, 肖芝, 于亚峰. 梅尼埃病与免疫细胞表型之间的关系: 双向孟德尔随机化研究[J]. 中华耳科学杂志, 2025, 23(5): 637-42. doi：10.3969/j.issn.1672-2922.2025.05.003
[32]	Yatsunenko T, Rey FE, Manary MJ, et al. Human gut microbiome viewed across age and geography[J]. Nature, 2012, 486(7402): 222-7. doi：10.1038/nature11053