Abstract
目的
基于孟德尔随机化研究方法探究肠道菌群与色素沉着绒毛结节性滑膜炎之间的因果关系。
方法
利用孟德尔随机化分析的3种主要方法,对211个肠道菌群类群与色素沉着绒毛结节性滑膜炎之间进行双向双样本孟德随机化分析,基于GWAS汇总数据以阐明两者间的因果关系。以逆方差加权分析方法(IVW)作为主要结果,其他方法均作为补充分析。最后使用Cochran's Q检验、MR-Egger回归法、MR- PRESSO法以及条件孟德尔随机化分析(cML-MA)验证结果的可靠性。
结果
巴恩斯氏菌属(OR=3.12,95% CI:1.15~8.41,P=0.025)和瘤胃球菌科UCG010(OR=4.03,95% CI:1.19~13.68,P=0.025)的丰度升高可能增加色素沉着绒毛结节性滑膜炎的发病风险;毛螺菌科(OR=0.33,95% CI:0.12~0.91,P=0.032)、另枝菌属(OR=0.16,95% CI:0.05~0.53,P=0.003)、经黏液真杆菌属(OR=0.20,95% CI:0.06~0.61,P=0.005)、毛螺菌科FCS020(OR=0.38,95% CI:0.15~0.94,P=0.036)和瘤胃球菌科UCG014(OR=0.36,95% CI:0.14~0.94,P=0.037)的丰度升高与色素沉着绒毛结节性滑膜炎的发病风险降低相关。敏感性分析均支持该研究结果。反向孟德尔随机化分析并未发现两者间存在反向因果关联。
结论
巴恩斯氏菌属和瘤胃球菌科UCG010可能是色素沉着绒毛结节性滑膜炎的潜在危险因素,毛螺菌科、另枝菌属、经黏液真杆菌属、毛螺菌科FCS020和瘤胃球菌科UCG014可能是潜在的保护因素,肠道菌群在其发病机制中的重要作用,提供了潜在干预措施的新思路。
Keywords: 肠道菌群, 色素沉着绒毛结节性滑膜炎, 孟德尔随机化, 因果关系
Abstract
Objective
To investigate the causal relationship between gut microbiota and pigmented villonodular synovitis using Mendelian randomization analysis.
Methods
We conducted a two-sample Mendelian randomization analysis to investigate the causal relationship between 211 gut microbiome taxa and pigmented villonodular synovitis based on GWAS summary data, with inverse variance weighted (IVW) analysis as the primary result and the other methods as supplementary analyses. The reliability of the results was tested using Cochran's Q test, MR-Egger regression, MR-PRESSO method and conditional Mendelian randomization analysis (cML-MA).
Results
The increased abundance of Barnesiella (OR=3.12, 95% CI: 1.15-8.41, P=0.025) and Rumatococcaceae UCG010 (OR=4.03, 95% CI: 1.19-13.68, P=0.025) may increase the risk of pigmented villous nodular synovitis, and elevated abundance of Lachnospiraceae (OR=0.33, 95% CI: 0.12-0.91, P=0.032), Alistipes (OR=0.16, 95% CI: 0.05-0.53, P=0.003), Blautia (OR=0.20, 95% CI: 0.06-0.61, P=0.005), and Lachnospiraceae FCS020 group (OR=0.38, 95% CI: 0.15-0.94, P=0.036) and Ruminococcaceae UCG014 (OR=0.36, 95% CI: 0.14-0.94, P=0.037) were all associated with a reduced risk of pigmented villonodular synovitis, which were supported by the results of sensitivity analyses. Reverse Mendelian randomization analysis did not reveal any inverse causal association.
Conclusion
Increased abundance of specific intestinal microorganisms is associated with increased or decreased risks of developing hyperpigmented villonodular synovitis, and gut microbiota plays an important role in the pathogenesis of this disease.
Keywords: gut microbiota, pigmented villonodular synovitis, Mendelian randomization, causality
色素沉着绒毛结节性滑膜炎(PVNS)是一种罕见的、侵袭性的滑膜增生性疾病,主要特征为关节腱鞘和滑囊内的含铁血黄素沉积以及绒毛结节样增生[1],在发达国家的发病率为1.8/100万[2]。PVNS常可发生在手、肘、髋、膝和踝等关节部位,并且明显偏向于膝关节,由于其较强的侵袭性,可累及关节内外,导致关节肿胀、积液以及进行性骨质破坏,严重时需要进行关节置换手术,却仍有一定的复发率[3]。目前对于PVNS多种治疗方法的疗效存在争议,其已经成为临床治疗的难点,备受关注;且对于PVNS的病因及发病机制尚不完全清楚,但研究推测其与脂质代谢紊乱、创伤出血、炎症过程和肿瘤相关。
肠道菌群(GM)作为人体内重要的微生物系统,其组成以及多样性可影响多种疾病(如炎症性疾病、肥胖等)进程,与人类健康息息相关[4]。目前随着对GM研究的不断深入,国外有学者研究GM与骨骼健康之间的复杂机制,揭示了GM通过微生物衍生代谢产物在骨代谢与骨形成中充当着重要角色[5]。例如,肠道微生物群与滑膜、软骨和骨关节的炎症密切联系,可以增强软骨愈合、降低滑膜炎症,是骨关节炎的潜在治疗与预防方式之一[6, 7]。尚没有研究证实GM与PVNS发病机制之间的联系。
孟德尔随机化(MR)是一种利用遗传变异作为工具变量的有效分析方法,通过对微生物组群数据和全基因组关联研究数据(GWAS)的综合分析,从而筛选与GM组成紧密相关的遗传变异,克服传统研究的局限性,降低混杂因素及反向因果所造成的偏倚,为因果推论研究提供更可靠的证据[8]。本研究旨在揭示GM与PVNS之间的因果关系,探究其潜在机制,为PVNS的预防和治疗方案提供新思路。
1. 资料和方法
1.1. 研究设计
本研究设计采用两样本双向MR分析数据,为GM和PVNS之间的因果关系提供强有力的支持。利用来自公共数据的GWAS荟萃分析,选择GM为暴露因素,筛选与GM显著相关的单核苷酸多态性(SNP)作为工具变量(IVs);PVNS作为结局变量,而后进行MR分析,并通过Cochran's Q与MR Egger法进行异质性和水平多效性检验,最后进行敏感性分析以验证结果的可靠性。为把控文章质量,获取更可靠的数据,严格遵循满足MR的3个基本假设:选择与GM显著关联的SNPs作为工具变量;SNPs应是相对独立的,与已知的混杂因素(影响GM和PVNS)无关;SNPs只通过暴露 (GM或PVNS)影响结局(PVNS或GM)(图1)。
图1.

肠道菌群孟德尔随机化研究原理及数据来源图
Fig.1 Principle and data source of Mendelian randomization analysis of gut microbiota.
1.2. 数据来源
本研究与GM相关的遗传信息数据来源于国际MiBioGen联盟的一项研究[9],该研究包括24个队列的18 340个个体,主要分析了参与者粪便菌群16S rRNA 测序图谱来查明影响肠道微生物群相对丰度的遗传位点,最终检查评估了211个门、纲、目、科和属。与PVNS相关的遗传信息数据来源于芬兰数据库R9发布的数据集(https://storage.googleapis.com/finngen-public-data-r9/summary_stats/finngen_R9_M13_VILLONODULAR.gz),其中包括212例PVNS病例和240 862例作为对照。为证明本研究结果的可靠性和可重复性,我们选取了来源于芬兰数据库R10发布的最新PVNS数据集按相同条件进行验证分析。
1.3. 工具变量选择
严格控制数据质量,选择合适的工具变量,以保障有关GM组成与PVNS之间因果关系推测的完整性和准确性。 (1)本研究为满足相关性假设,在GWAS数据集中筛选与GM显著相关的SNPs(P<1×10-5)[10,11],以确保SNPs与工具变量显著相关。(2)设置参数k=10 000,r 2=0.001,以去除连锁不平衡的影响,避免遗传变异残留所造成的误差[12]。(3)为使数据协调,从分析中删除所有的回文SNP。(4)通过Pheno Scanner[13](http://www.phenoscanner.medschl.cam.ac.uk/),我们已经对所有SNP在二级表型进行检索,以尽可能排除混杂因素的影响。(5)使用F统计值评估IV的强度,设置F>10作为强IV的统计阈值,否则视为IV与暴露的相关性较差,应予以排除[14]。
1.4. 统计学分析
本研究使用RStudio软件(4.3.2版)中TwoSampleMR、 MR-PRESSO和MRcML软件包进行数据分析。
1.4.1. 孟德尔随机化分析
本研究使用逆方差加权分析方法(IVW)为分析GM与PVNS之间因果关系的主要方法[15],该方法是根据方差的倒数计算加权平均值,当所有IV有效且不存在异质性及多效性时,选定IVW的结果是最可靠的。同时采用MR-Egger回归和加权中位数(WME)作为补充分析方法。MR-Egger回归也可以用来检验因果关系,但本研究结合MR-Egger回归排除一些结果[16]。WME只有当存在至少50%有效的IV时,才能够获得较为准确的估计结果[17]。为了更好的分析GM与PVNS之间的因果关系,本研究还采用同样的数据源和参数设置,将正向MR中发现的与结局显著相关菌群和PVNS进行反向MR分析。
1.4.2. 敏感性分析
本研究进行敏感性分析以评估结果的稳健性,包括逐一剔除检验、异质性检验和水平多效性检验。采用留一法逐个剔除每一个SNP,然后重新进行IVW分析,以评估结果是否存在偏倚或单个SNP是否对结果产生特异性影响[18]。使用Cochran's Q统计量和MR-Egger回归评估异质性,若P<0.05则认为检验具有统计学意义,存在异质性。MR-Egger截距用来评估多效性,以确定IV不会通过混杂因素影响结局变量,若MR-Egger截距分析中P<0.05,则认为检验具有统计学意义,具有水平多效性。同时本研究还采用了MR- PRESSO法,以分析和校准IVs中显著异常值,以降低水平多效性对因果推论的影响[10];采用一种基于约束最大似然和模型平均的MR方法cML-MA,用于控制相关和不相关的多效性效应,有助于提供更可靠的结果[19]。
2. 结果
2.1. 工具变量筛选结果
根据IVW结果显示,9种不同的肠道菌群(1目、1科和7属)可能与PVNS之间可能存在因果关联,同时使用MR-Egger回归和加WME进行补充分析。其中放线菌门(Actinobacteria)和Family XIII AD3011 group由于可能存在潜在的离群值,MR-Egger回归法与IVW法分析结果方向相悖,为确保数据的准确性故排除。最终筛选出7种(1科和6属)可能与PVNS之间可能存在因果关联。最后SNPs的筛选结果和结果如表1所示,最终筛选出80个SNPs,且统计量F均>10。
表1.
与PVNS风险相关的7种肠道菌群相关的SNPs特征
Tab.1 Characteristics of SNPs associated with 7 gut microbiota related with pigmented villonodular synovitis (PVNS) risk
| Exposures | SNPs | EA | Beta | SE | P | F |
|---|---|---|---|---|---|---|
| Lachnospiraceae | rs10402491 | C | 0.066170067 | 0.014850954 | 7.58E-06 | 19.85246518 |
| rs11139361 | T | 0.049446281 | 0.011050331 | 4.26E-06 | 20.02242572 | |
| rs112040820 | A | 0.054958365 | 0.011708564 | 2.42E-06 | 22.03232553 | |
| rs11841382 | G | -0.071812797 | 0.017461974 | 9.58E-06 | 16.91285828 | |
| rs11979110 | T | -0.050092357 | 0.010506859 | 1.82E-06 | 22.72987911 | |
| rs1205443 | A | 0.0501415 | 0.011215719 | 7.29E-06 | 19.98666826 | |
| rs12760724 | A | -0.048456589 | 0.010793397 | 7.27E-06 | 20.15530508 | |
| rs13005175 | A | 0.099348285 | 0.021839504 | 8.37E-06 | 20.69355985 | |
| rs2159863 | A | -0.058583102 | 0.012874212 | 3.70E-06 | 20.70634355 | |
| rs2910921 | T | 0.160302066 | 0.035832011 | 8.42E-06 | 20.01409117 | |
| rs3127230 | C | -0.050353667 | 0.011220058 | 6.20E-06 | 20.1405814 | |
| rs35524804 | T | -0.060748124 | 0.012521578 | 2.45E-06 | 23.53681074 | |
| rs7359994 | T | -0.050314217 | 0.01124817 | 5.36E-06 | 20.0086457 | |
| rs79086868 | T | 0.077672344 | 0.016447234 | 3.01E-06 | 22.30216787 | |
| rs959845 | C | -0.049366584 | 0.010780126 | 5.17E-06 | 20.97096808 | |
| rs9929145 | G | -0.125721183 | 0.024523339 | 2.84E-07 | 26.28195786 | |
| Alistipes | rs1107244 | G | 0.075855776 | 0.017118221 | 3.59E-06 | 21.22625881 |
| rs11769002 | G | -0.052879265 | 0.010938851 | 1.45E-06 | 22.22625881 | |
| rs11958296 | A | -0.098122204 | 0.021827201 | 9.30E-06 | 23.22625881 | |
| rs12990744 | C | -0.077661791 | 0.017298561 | 8.21E-06 | 24.22625881 | |
| rs1689282 | A | -0.052006142 | 0.011394196 | 5.28E-06 | 25.22625881 | |
| rs2290844 | C | 0.081396908 | 0.019163185 | 9.10E-06 | 26.22625881 | |
| rs2450745 | A | -0.080523959 | 0.018470178 | 7.12E-06 | 27.22625881 | |
| rs2875322 | T | -0.058094464 | 0.013143238 | 8.78E-06 | 28.22625881 | |
| rs34417064 | A | -0.04820195 | 0.010686843 | 7.01E-06 | 29.22625881 | |
| rs4810359 | A | -0.065242793 | 0.014615527 | 7.50E-06 | 30.22625881 | |
| rs7129639 | C | -0.052495538 | 0.010958377 | 1.78E-06 | 31.22625881 | |
| rs8130320 | A | -0.049001972 | 0.01071701 | 4.84E-06 | 32.22625881 | |
| Barnesella | rs11155559 | T | 0.095602282 | 0.021296077 | 8.92E-06 | 20.15288845 |
| rs12909713 | C | -0.055071556 | 0.012003894 | 4.95E-06 | 21.04797848 | |
| rs13242616 | T | -0.058374651 | 0.012317572 | 2.29E-06 | 22.45941164 | |
| rs199035 | G | 0.055949483 | 0.011971558 | 3.00E-06 | 21.84191978 | |
| rs2276875 | A | -0.069717565 | 0.013953704 | 4.65E-07 | 24.96349615 | |
| rs2428166 | G | -0.165858717 | 0.033730907 | 8.51E-07 | 24.17801066 | |
| rs35177866 | A | 0.091712545 | 0.019013343 | 2.95E-06 | 23.26700721 | |
| rs62251337 | A | -0.069076268 | 0.014937412 | 4.24E-06 | 21.38488958 | |
| rs72684847 | T | -0.114369278 | 0.025387327 | 6.76E-06 | 20.29480136 | |
| rs76181748 | C | -0.077855934 | 0.017163996 | 6.78E-06 | 20.57532145 | |
| rs77455852 | T | -0.089149084 | 0.019560057 | 3.16E-06 | 20.77272818 | |
| rs79795328 | A | -0.081864467 | 0.01764407 | 4.23E-06 | 21.52748687 | |
| Blautia | rs11149971 | C | 0.117604747 | 0.023389975 | 1.04E-06 | 25.28076366 |
| rs115043014 | G | -0.206605438 | 0.043991974 | 5.19E-06 | 22.05649991 | |
| rs117001700 | T | 0.196414252 | 0.044112197 | 8.84E-06 | 19.82570402 | |
| rs12453000 | C | 0.062533456 | 0.012998711 | 1.26E-06 | 23.14324677 | |
| rs16892041 | T | -0.062260496 | 0.014180398 | 8.82E-06 | 19.27739347 | |
| rs2788271 | T | -0.057559633 | 0.013349169 | 7.16E-06 | 18.59206247 | |
| rs3005511 | A | 0.050108496 | 0.01107676 | 6.19E-06 | 20.46431647 | |
| rs4926264 | T | 0.082620681 | 0.017823907 | 5.10E-06 | 21.48679922 | |
| rs67794373 | C | 0.060170889 | 0.012344281 | 1.00E-06 | 23.75971668 | |
| rs682885 | A | -0.049337611 | 0.010735883 | 4.49E-06 | 21.11935708 | |
| rs72973581 | A | 0.125169075 | 0.026540793 | 1.74E-06 | 22.24161383 | |
| rs7860714 | A | -0.050217775 | 0.010985269 | 4.09E-06 | 20.89746184 | |
| Lachnospiraceae FCS020 | rs10093861 | G | -0.056886893 | 0.012115188 | 3.06E-06 | 22.04774176 |
| rs1254846 | G | 0.105962631 | 0.023250118 | 5.60E-06 | 20.7708918 | |
| rs1363769 | T | -0.200628437 | 0.044937217 | 1.58E-06 | 19.93299851 | |
| rs2322265 | C | -0.0666305 | 0.014157654 | 5.21E-06 | 22.14948107 | |
| rs2862811 | T | 0.056492728 | 0.012173604 | 3.92E-06 | 21.53509309 | |
| rs35035870 | T | -0.190615272 | 0.041440269 | 2.62E-06 | 21.15778768 | |
| rs3999074 | G | -0.05505865 | 0.012184677 | 6.55E-06 | 20.41846429 | |
| rs4452603 | T | 0.060420825 | 0.013596453 | 8.98E-06 | 19.74795143 | |
| rs7249113 | G | 0.067948308 | 0.013349587 | 3.72E-07 | 25.90726893 | |
| rs72793667 | A | -0.116880798 | 0.024658906 | 1.63E-06 | 22.46667102 | |
| rs9308097 | A | 0.05538101 | 0.012365944 | 7.47E-06 | 20.05705626 | |
| rs9788306 | C | -0.062803973 | 0.013074422 | 1.39E-06 | 23.07433744 | |
| Ruminococcaceae UCG010 | rs12597105 | G | 0.067085456 | 0.014441482 | 4.87E-06 | 21.57909494 |
| rs2820282 | A | -0.059230421 | 0.012591704 | 2.85E-06 | 22.12690276 | |
| rs682403 | A | -0.058816059 | 0.01246713 | 2.37E-06 | 22.25660232 | |
| rs6958419 | C | -0.058571831 | 0.012499376 | 2.84E-06 | 21.95841235 | |
| rs73218807 | G | -0.166210686 | 0.036793834 | 6.43E-06 | 20.40646575 | |
| rs7441445 | C | -0.056949974 | 0.012650232 | 6.80E-06 | 20.26702856 | |
| Ruminococcaceae UCG014 | rs10495392 | C | -0.082485954 | 0.018719441 | 9.96E-06 | 19.41664775 |
| rs10791168 | A | -0.066462357 | 0.015006876 | 9.76E-06 | 19.61421311 | |
| rs10941294 | C | -0.122057405 | 0.026001669 | 2.40E-06 | 22.03564737 | |
| rs115777838 | T | -0.188349525 | 0.038664295 | 4.62E-07 | 23.7306081 | |
| rs12638134 | T | 0.058254795 | 0.011965732 | 1.21E-06 | 23.70199016 | |
| rs34402072 | C | -0.068802318 | 0.015608285 | 9.80E-06 | 19.43102754 | |
| rs56105232 | G | 0.139275689 | 0.029913177 | 2.91E-06 | 21.67831635 | |
| rs72809222 | T | 0.067177506 | 0.013983796 | 2.41E-06 | 23.07796941 | |
| rs853612 | A | -0.05282572 | 0.011936668 | 9.75E-06 | 19.58504711 | |
| rs995642 | C | 0.060047967 | 0.012641687 | 1.90E-06 | 22.56246524 |
SNPs: Single nucleotide polymorphisms; EA:Effect allele;Eaf:Effect allele frequency;Beta:Effect allele value;SE:Standard error; F: F-statistic.
1
2.2. 肠道菌群与色素沉着绒毛结节性滑膜炎的因果效应
图2.
基于MR分析GM与PVNS因果关系的森林图
Fig.2 Forest map of Mendelian randomization analysis of causal relationship between gut microbiota and PVNS.
表2.
MR分析的主要结果
Tab.2 Main results of Mendelian randomization analysis
| Exposures | Method | Beta | OR | 95%CI | P |
|---|---|---|---|---|---|
| Lachnospiraceae Inverse variance weighted -1.10 0.33 0.12-0.91 0.032 | |||||
| MR Egger | -2.46 | 0.09 | 0.003-2.85 | 0.191 | |
| Weighted median | -0.73 | 0.48 | 0.12-1.90 | 0.297 | |
| Alistipes | Inverse variance weighted | -1.83 | 0.16 | 0.05-0.53 | 0.003 |
| MR Egger | -4.40 | 0.01 | 0.00004-4.10 | 0.169 | |
| Weighted median | -1.69 | 0.18 | 0.04-0.94 | 0.041 | |
| Barnesella | Inverse variance weighted | 1.14 | 3.12 | 1.15-8.41 | 0.025 |
| MR Egger | 3.16 | 23.68 | 0.41-1365.64 | 0.157 | |
| Weighted median | 1.22 | 3.40 | 0.94-12.28 | 0.061 | |
| Blautia | Inverse variance weighted | -1.61 | 0.20 | 0.06-0.61 | 0.005 |
| MR Egger | -1.11 | 0.33 | 0.02-5.92 | 0.469 | |
| Weighted median | -1.12 | 0.33 | 0.07-1.48 | 0.146 | |
| Lachnospiraceae FCS020 | Inverse variance weighted | -0.98 | 0.38 | 0.15-0.94 | 0.036 |
| MR Egger | -0.58 | 0.56 | 0.03-9.67 | 0.697 | |
| Weighted median | -0.96 | 0.38 | 0.10-1.51 | 0.171 | |
| Ruminococcaceae UCG010 | Inverse variance weighted | 1.39 | 4.03 | 1.19-13.68 | 0.025 |
| MR Egger | 1.74 | 5.71 | 0.20-164.23 | 0.367 | |
| Weighted median | 1.59 | 4.91 | 1.02-23.68 | 0.047 | |
| Ruminococcaceae UCG014 | Inverse variance weighted | -1.02 | 0.36 | 0.14-0.94 | 0.037 |
| MR Egger | -1.07 | 0.34 | 0.04-3.13 | 0.370 | |
| Weighted median | -0.73 | 0.48 | 0.12-1.9 | 0.302 | |
Beta: Effect allele value; OR: Odds ratio; CI: Confidence interval.
本研究发现了个是PVNS风险因素的微生物群,包括巴恩斯氏菌属(Barnesiella)(OR=3.12,95% CI:1.15~8.41,P=0.025)和瘤胃球菌科UCG010(RuminococcaceaeUCG010)(OR=4.03,95% CI:1.19~13.68,P=0.025)。此外。我们还发现5个微生物群是PVNS的保护因素,包括毛螺菌科(Lachnospiraceae)(OR=0.33,95% CI:0.12~0.91,P=0.032)、另枝菌属(Alistipes)(OR=0.16,95% CI:0.05~0.53,P=0.003)、经黏液真杆菌属(Blautia)(OR=0.20,95% CI:0.06~0.61,P=0.005)、毛螺菌科FCS020(LachnospiraceaeFCS020group)(OR=0.38,95% CI:0.15~0.94,P=0.036)和瘤胃球菌科UCG014(RuminococcaceaeUCG014)(OR=0.36,95% CI:0.14~0.94,P=0.037)。GM与PVNS的MR分析的3种主要分析方法结果的散点图(图3)。

2.3. 敏感性分析
结果显示所有的Q_pval均>0.05,表明各微生物群分类组的全部SNP均不存在异质性。用MR-Egger回归评估SNPs与结果之间的多效性,MR-Egger回归中截距项的P值始终超过0.05,表明不存在水平多效性。同时,在MR-PRESSO分析中,如果相关微生物群群落的全局检验P值大于0.05,则表明不存在水平多效性。多效性检验是MR分析最重要的环节,其确保了结果的准确性和稳健性,MR-Egger截距回归分析较为合理的解决了不相关水平多效性的偏倚,但仍无法控制相关水平多效性的影响。故本研究最后采用RStudio软件的MRcML包对GM与PVNS进行分析,结果显示GM与PVNS结果不存在相关水平多效性(表3)。留一法检验结果证实了本研究MR结果的稳健性,未见单个SNP对研究结果产生较大偏倚(图4)。


2.4. 反向MR分析的结果
采用IVW方法进行反向MR分析。结果显示在相同的筛选条件下(P<1×10-5,k=10 000,r2=0.001),未鉴定出与PVNS密切相关的工具变量。
2.5. 可重复性验证
选择芬兰数据库最新的PVNS数据集再次分析,结果显示,毛螺菌科(OR=0.32,95% CI:0.12~0.85,P=0.022)、另枝菌属(OR=0.25,95% CI:0.08~0.76,P=0.014)、经黏液真杆菌属(OR=0.19,95% CI:0.07~0.54,P=0.002)是PVNS的保护因素;瘤胃球菌科UCG010(OR=6.57,95% CI:2.03~21.67,P=0.002)是风险因素(表4)。
表4.
肠道菌群与PVNS(R10)因果分析结果
Tab.4 Results of causal analysis of gut microbiota and PVNS (R10)
| Exposure | Method | Beta | OR | 95% CI | P |
|---|---|---|---|---|---|
| Lachnospiraceae | Inverse variance weighted | -1.13 | 0.32 | 0.12-0.85 | 0.022 |
| MR Egger | -2.29 | 0.10 | 0.002-4.59 | 0.191 | |
| Weighted median | -0.82 | 0.44 | 0.11-1.83 | 0.297 | |
| Alistipes | Inverse variance weighted | -1.38 | 0.25 | 0.08-0.76 | 0.014 |
| MR Egger | -4.63 | 0.01 | 0.00003-2.97 | 0.141 | |
| Weighted median | -1.96 | 0.14 | 0.03-0.67 | 0.013 | |
| Blautia | Inverse variance weighted | -1.66 | 0.19 | 0.07-0.54 | 0.002 |
| MR Egger | -1.82 | 0.16 | 0.07-1.18 | 0.220 | |
| Weighted median | -1.27 | 0.28 | 0.07-1.48 | 0.082 | |
| Ruminococcaceae UCG010 | Inverse variance weighted | 1.88 | 6.57 | 2.03-21.27 | 0.002 |
| MR Egger | 1.66 | 5.26 | 0.21-132.01 | 0.370 | |
| Weighted median | 1.77 | 5.88 | 1.18-29.37 | 0.031 |
Beta: Effect allele value; OR: Odds ratio;CI: Confidence interval.
3. 讨论
本研究旨在探究GM 与PVNS之间的因果关系,通过MR研究全面评估211种肠道微生物,鉴别出9种GM与PVNS相关,对研究结果进行严格的质量控制,避免混杂因素及反向因果所造成的影响,最终显示与PVNS患病风险存在因果关系的GM有7种。其中,增加PVNS患病风险的菌群有2种,包括巴恩斯氏菌属和瘤胃球菌科UCG010;有5种菌群对PVNS的发病存在保护作用,包括毛螺菌科、另枝菌属、经黏液真杆菌属、毛螺菌科FCS020和瘤胃球菌科UCG014。
目前尚未有研究能完全阐述GM与PVNS之间的作用机制,但仍有现存的临床研究表明两者存在一定的联系,这可以为本研究的结果提供一定的证据支撑。巴恩斯氏菌属归属于拟杆菌门,是一种近些年新发现的菌属,大多研究认为其作为一类肠道益生菌,可以帮助维持肠道菌群的平衡,促进消化和吸收营养物质,它还可以抑制有害菌的生长,提高免疫系统的功能[20-22]。有研究发现其可以作为一种“抗肿瘤益生菌”存在,通过促进特异性T细胞的浸润,改善抗肿瘤调节剂的疗效[23]。然而,另外有研究表明巴恩斯氏菌属可能与癌症的发生有关。国内有研究表明,对正常小鼠和原发性肝癌小鼠肠道菌群进行多样性研究,二者间巴恩斯氏菌属存在显著性差异,推测其与肝癌发生存在可能的相关性[24]。而恰恰有“肿瘤源”学说认为PVNS可能属于一种肿瘤样病变,可广泛侵袭滑膜组织,并且恶变的PVNS与肿瘤疾病的一般特性更加相似[25, 26],这表明巴恩斯氏菌属在PVNS中可能也具有尚未发现的影响,增加其潜在的发生风险。另枝菌属也属于拟杆菌门,是一类常见的肠道细菌,有研究证实其被视为潜在的短链脂肪酸(SCFAs)生产者而具有潜在的抗炎作用,可以在癌症免疫治疗中通过调节肿瘤微环境发挥有益作用[27]。Iida等[28]进行的一项动物研究认为,另枝菌属与Toll样受体4(TLR4)启动和肿瘤坏死因子(TNF)产生的作用之间存在正相关联系,可与TLR4相结合从而诱导肿瘤相关髓样细胞启动TNF表达,从而改变微环境,使肿瘤发生细胞凋亡。同时结合我们的研究结果,我们推测另枝菌属可通过调节免疫应答从而抑制滑膜肿瘤样病变,成为PVNS的保护因素。但是目前现存研究并未有明确的结论,要明确两者之间的因果关系还需要更多的研究。在我们的研究中显示巴恩斯氏菌属对罹患PVNS起促进作用,另枝菌属则降低PVNS的发生风险,这可能为我们以后阐述PVNS的发病机制及更加有效的治疗方案提供新思路。
瘤胃球菌科UCG010和瘤胃球菌科UCG014同属于瘤胃球菌科,作为一种革兰阳性厌氧菌,可以通过酵解宿主消化系统中的纤维素、葡萄糖以及木糖醇等获取能量,在人体新陈代谢中发挥着至关重要的作用。本研究结果显示两个菌属对于PVNS的发生产生不同的影响,这可能是因为瘤胃球菌科既包括有益菌又包括有害菌。目前尚未有研究明确瘤胃球菌科与PVNS之间的因果关系。但有研究显示,瘤胃球菌科UCG010可与血脂异常呈正相关,并且在肥胖易感的小鼠中,发现富集的瘤胃球菌科可以促进脂肪的合成[29, 30]。如今有学者认为脂质代谢紊乱可能是PVNS的病因之一[31],所以我们推测肠道中异常的瘤胃球菌科UCG010积累,可导致人体脂肪代谢的紊乱,从而增加发生PVNS的风险。另外有研究显示,瘤胃球菌科能够产生SCFAs,尤其是丁酸盐,是肠上皮细胞的主要能量来源,同时抑制促炎细胞因子的信号通路,具有抗炎作用,在调节肠道炎症方面具有重要作用[32];同时,也有研究发现瘤胃球菌科与关节炎类疾病,如膝骨关节炎等有一定的相关性,在一项观察艾灸治疗膝骨关节炎大鼠肠道菌群及炎症影响的研究中,发现艾灸可以明显调节体内炎症因子水平,降低促炎因子白细胞介素-1β(IL-1β)和肿瘤坏死因子-α(TNF-α),同时相应地发现瘤胃球菌科UCG014丰度增加,证实瘤胃球菌科UCG014与炎症因子水平的相关性[33]。有国内学者探究PVNS的发病机制,发现PVNS患者滑液中的炎症因子(IL-1β和TNF-α)显著上调和钙粘蛋白-11的高表达,肯定了炎症因子在PVNS发病中的重要影响[34]。所以我们认为瘤胃球菌科UCG014可能通过降低炎症因子表达来减小PVNS的易感风险。
经黏液真杆菌属和毛螺菌科FCS020同属于毛螺菌科,是人体最丰富的肠道菌群之一,存在于大多数健康人群的肠道中,同样可以参与人体碳水化合物的代谢,水解淀粉以及其他多糖产生SCFAs和丁酸盐等[35,36]。根据现有研究显示[37],丁酸盐能够显著降低软骨炎症因子的水平,包括 IL-1β、基质金属蛋白酶-13 (MMP-13)和 TNF-α 等,还能够降低受SCFAs调节的iNOS水平,抑制氧化应激反应,并恢复紧密连接蛋白、改善肠道组织形态,抑制肠道损伤,通过减少炎症细胞死亡和改善自噬过程保护软骨细胞。这在维持骨与软骨的完整性和防御机制方面起着至关重要的作用。若肠道菌群紊乱,肠道炎症发生,部分肠道免疫炎症因子如IL-1β、MMP13 和 TNF-α 等可以通过体液循环直接从肠道转移到骨关节,从而影响骨系统的免疫活性。此外,肠道微生物组对免疫细胞的成长和分化有一定影响,肠道细菌及其代谢产物可以与免疫细胞相互作用。一项动物研究显示[38],丁酸可以调节调节性T细胞(Tregs)的数量和功能,并减少辅助性T细胞17(Th17)的数量且不促进其极化。合理的肠道菌群组成有助于提升对骨关节疾病产生有效免疫反应的能力并维持全身免疫稳态。根据本研究结果,我们可以深入探究毛螺菌科降低PVNS潜在发生风险的保护机制,探索新的干预或治疗措施如增加益生菌或粪便微生物群移植等,这可能成为一种新的有效预防或治疗PVNS的临床策略。
本研究仍存在一定的局限性。首先,MR基于作为工具变量的遗传变异仅通过研究的暴露因素影响结果这个假设。虽然尽可能地剔除混淆因素影响,并进行异质性和敏感性分析来进一步确保研究结果的稳定性,但仍然可能存在未纳入的混淆因素。其次,本研究数据集主要来自于欧洲人种,对亚洲人种可能不具有广适性。对于亚洲人群而言,可能存在未知的肠道菌群对色素沉着绒毛结节性滑膜炎存在影响。此外,本研究为了确保结果的准确性和外推性,拟采用多队列GWAS数据对本研究结果进行验证,但现在鲜针对有PVNS的全基因组关联测序研究,故本研究验证队列数据来自于芬兰R10数据库,但两版本的结局变量数据集的遗传相关性研究在统计学准许范围内,样本重叠对本研究主要结果不存在较大偏倚影响。
综上所述,本研究通过MR探讨GM与PVNS之间的因果关系,发现7种特定肠道微生物的丰度变化与PVNS的发病风险增加或降低相关,可能有助于阐明GM对炎症因子和免疫调节作用的影响,并激发对PVNS预防和治疗领域医学的发展。
基金资助
国家自然科学基金(81873316);天津市科技计划项目(23KPXMRC00170);天津市卫健委津门医学英才项目(TJSJMYXYC-D2-028)
Supported by National Natural Science Foundation of China (81873316).
参考文献
- 1. Smith SC, Snyder GM. Orthopedic management of a patient with pigmented villonodular synovitis[J]. JAAPA, 2022, 35(11): 1-4. [DOI] [PubMed] [Google Scholar]
- 2. Myers BW, Masi AT. Pigmented villonodular synovitis and tenosynovitis: a clinical epidemiologic study of 166 cases and literature review[J]. Medicine, 1980, 59(3): 223-38. [PubMed] [Google Scholar]
- 3. Tan YC, Tan JY, Tsitskaris K. Systematic review: total knee arthroplasty (TKA) in patients with pigmented villonodular synovitis (PVNS)[J]. Knee Surg Relat Res, 2021, 33(1): 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Izda V, Schlupp L, Prinz E, et al. Murine cartilage microbial DNA deposition occurs rapidly following the introduction of a gut microbiome and changes with obesity, aging, and knee osteoarthritis[J]. Geroscience, 2024, 46(2): 2317-41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Han D, Wang WJ, Gong JP, et al. Microbiota metabolites in bone: shaping health and Confronting disease[J]. Heliyon, 2024, 10(7): e28435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Guido G, Ausenda G, Iascone V, et al. Gut permeability and osteoarthritis, towards a mechanistic understanding of the pathogenesis: a systematic review[J]. Ann Med, 2021, 53(1): 2380-90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Berthelot JM, Sellam J, Maugars Y, et al. Cartilage-gut-microbiome axis: a new paradigm for novel therapeutic opportunities in osteoarthritis[J]. RMD Open, 2019, 5(2): e001037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Boehm FJ, Zhou X. Statistical methods for Mendelian randomization in genome-wide association studies: a review[J]. Comput Struct Biotechnol J, 2022, 20: 2338-51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kurilshikov A, Medina-Gomez C, Bacigalupe R, et al. Large-scale association analyses identify host factors influencing human gut microbiome composition[J]. Nat Genet, 2021, 53(2): 156-65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Li PS, Wang HY, Guo L, et al. Association between gut microbiota and preeclampsia-eclampsia: a two-sample Mendelian randomi-zation study[J]. BMC Med, 2022, 20(1): 443. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Liu X, Qi XS, Han RS, et al. Gut microbiota causally affects cholelithiasis: a two-sample Mendelian randomization study[J]. Front Cell Infect Microbiol, 2023, 13: 1253447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hemani G, Zheng J, Elsworth B, et al. The MR-Base platform supports systematic causal inference across the human phenome[J]. Elife, 2018, 7: e34408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Kamat MA, Blackshaw JA, Young R, et al. PhenoScanner V2: an expanded tool for searching human genotype-phenotype associations[J]. Bioinformatics, 2019, 35(22): 4851-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Zhuang ZH, Yu CQ, Guo Y, et al. Metabolic signatures of genetically elevated vitamin D among Chinese: observational and Mendelian randomization study[J]. J Clin Endocrinol Metab, 2021, 106(8): e3249-60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Wang Q, Song YX, Wu XD, et al. Gut microbiota and cognitive performance: a bidirectional two-sample Mendelian randomization[J]. J Affect Disord, 2024, 353: 38-47. [DOI] [PubMed] [Google Scholar]
- 16. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression[J]. Int J Epidemiol, 2015, 44(2): 512-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Bowden J, Davey Smith G, Haycock PC, et al. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted Median estimator[J]. Genet Epidemiol, 2016, 40(4): 304-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Gronau QF, Wagenmakers EJ. Limitations of Bayesian leave-one-out cross-validation for model selection[J]. Comput Brain Behav, 2019, 2(1): 1-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Xue H, Shen X, Pan W. Constrained maximum likelihood-based Mendelian randomization robust to both correlated and uncorrelated pleiotropic effects[J]. Am J Hum Genet, 2021, 108(7): 1251-69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Mancabelli L, Milani C, Lugli GA, et al. Meta-analysis of the human gut microbiome from urbanized and pre-agricultural populations[J]. Environ Microbiol, 2017, 19(4): 1379-90. [DOI] [PubMed] [Google Scholar]
- 21. Yan H, Qin Q, Yan S, et al. Comparison of the gut microbiota in different age groups in China[J]. Front Cell Infect Microbiol, 2022, 12: 877914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Chen Q, Hu D, Wu XT, et al. Dietary γ-aminobutyric acid supplementation inhibits high-fat diet-induced hepatic steatosis via modulating gut microbiota in broilers[J]. Microorganisms, 2022, 10(7): 1281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Dutta D, Lim SH. Bidirectional interaction between intestinal microbiome and cancer: opportunities for therapeutic interventions[J]. Biomark Res, 2020, 8: 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. 甄宏德, 王爱国, 钱 祥, 等. 原发性肝癌H-ras12V小鼠肠道微生态的初步研究[J]. 中国微生态学杂志, 2018, 30(2): 132-6. DOI: 10.13381/j.cnki.cjm.201802002 [DOI] [Google Scholar]
- 25. Chipman DE, Perkins CA, Lijesen E, et al. Pigmented villonodular synovitis/giant cell tumor in the knee[J]. Curr Opin Pediatr, 2024, 36(1): 78-82. [DOI] [PubMed] [Google Scholar]
- 26. Gounder MM, Thomas DM, Tap WD. Locally aggressive connective tissue tumors[J]. J Clin Oncol, 2018, 36(2): 202-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Parker BJ, Wearsch PA, Veloo ACM, et al. The genus Alistipes: gut bacteria with emerging implications to inflammation, cancer, and mental health[J]. Front Immunol, 2020, 11: 906. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Iida N, Dzutsev A, Stewart CA, et al. Commensal bacteria control cancer response to therapy by modulating the tumor microenvironment[J]. Science, 2013, 342(6161): 967-70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Li S, Zhai JY, Chu WW, et al. Alleviation of Limosilactobacillus reuteri in polycystic ovary syndrome protects against circadian dysrhythmia-induced dyslipidemia via capric acid and GALR1 signaling[J]. NPJ Biofilms Microbiomes, 2023, 9(1): 47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Gong S, Ye T, Wang M, et al. Traditional Chinese medicine formula Kang Shuai Lao Pian improves obesity, gut dysbiosis, and fecal metabolic disorders in high-fat diet-fed mice[J]. Front Pharmacol, 2020, 11: 297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Roguski M, Safain MG, Zerris VA, et al. Pigmented villonodular synovitis of the thoracic spine[J]. J Clin Neurosci, 2014, 21(10): 1679-85. [DOI] [PubMed] [Google Scholar]
- 32. Xiong Y, Ji L, Zhao Y, et al. Sodium butyrate attenuates taurocholate-induced acute pancreatitis by maintaining colonic barrier and regulating gut microorganisms in mice[J]. Front Physiol, 2022, 13: 813735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Jia YJ, Li TY, Han P, et al. Effects of different courses of moxibustion treatment on intestinal flora and inflammation of a rat model of knee osteoarthritis[J]. J Integr Med, 2022, 20(2): 173-81. [DOI] [PubMed] [Google Scholar]
- 34. Cao CX, Wu F, Niu XY, et al. Cadherin-11 cooperates with inflammatory factors to promote the migration and invasion of fibroblast-like synoviocytes in pigmented villonodular synovitis[J]. Theranostics, 2020, 10(23): 10573-88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Stadlbauer V, Engertsberger L, Komarova I, et al. Dysbiosis, gut barrier dysfunction and inflammation in dementia: a pilot study[J]. BMC Geriatr, 2020, 20(1): 248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Li L, Bao J, Chang Y, et al. Gut microbiota may mediate the influence of periodontitis on prediabetes[J]. J Dent Res, 2021, 100(12): 1387-96. [DOI] [PubMed] [Google Scholar]
- 37. Cho KH, Na HS, Jhun J, et al. Lactobacillus (LA-1) and butyrate inhibit osteoarthritis by controlling autophagy and inflammatory cell death of chondrocytes[J]. Front Immunol, 2022, 13: 930511. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Hui WP, Yu DP, Cao Z, et al. Butyrate inhibit collagen-induced arthritis via Treg/IL-10/Th17 axis[J]. Int Immunopharmacol, 2019, 68: 226-33. [DOI] [PubMed] [Google Scholar]

