精神分裂症和双相障碍躁狂发作患者肠道菌群的对照研究

A comparative study of gut microbiota in patients with Schizophrenia and manic episode of bipolar bisorder

目的描述精神分裂症和双相情感障碍(躁狂症)两种严重精神疾病患者的肠道微生物菌群特征,探讨肠道微生物群作为精神分裂症和躁狂状态诊断生物标志物的潜力。方法于2019年3—8月招募精神分裂症(SCZ)、躁狂症(BD-M)患者和健康对照者(HCs)作为研究对象,采用Bech-Rafaelsen躁狂症评定量表(BRMS)、精神分裂症阳性和阴性症状量表(PANSS)评定精神症状的严重程度。入院后3天内收集患者组(SCZ和BD-M)和HCs粪便标本,采用高通量的16SrRNA微阵列技术分析,比较严重精神疾病患者组和HCs的肠道微生物组成和多样性的差异。采用各种操作分类单元(OTUs)计算多个α多样性指标,利用Bray-Curtis相异度和加权UniFrac距离进行了主坐标分析(PCoA),比较肠道细菌群落的总体多样性。通过LEfSe软件应用线性判别分析方法(LDA)进行组间差异物种筛选。采用随机森林算法对每个样本的变量在属水平进行聚类,并确定了前10个(平均基尼重要性)标记,用于预测不同属的重要性。结果92份粪便标本中包括32份BD-M、28份SCZ和32份HCs标本。肠道细菌群落的总体多样性:SCZ、BD-M和HCs三组粪便细菌群落α多样性没有显著差异(P>0.05);PCoA分析显示SCZ和BD-M均与HCs不同,SCZ和BD-M在点图(pot)上重叠更多,表现出更高的相似性。丰度差异:在属水平上发现了一系列的丰度差异的细菌。大肠埃希菌、瘤胃球菌、纤毛杆菌、坎德勒氏菌、原发性细菌、扁平单孢菌在严重精神疾病患者组(SCZ和BD-M)中的表达显著高于棒状杆菌在HCs中的表达(P<0.05)。随机森林算法,分别确定前10个菌属用于鉴别三组之间差异,SCZ和HCs之间,迪茨氏菌属在平均下降准确度得分和平均下降基尼得分最高,ROC曲线下面积(AUC)为0.792(95%CI:0.679-0.906,P<0.05)。BD-M和HCs之间,大肠志贺氏杆菌得分最高,AUC为0.804(95%CI:0.695-0.913,P<0.05)。SCZ和BD-M之间,乳酸杆菌属在平均下降准确度和丙酸杆菌属在平均下降基尼得分最高AUC为0.605(95%CI:0.459-0.751,P<0.05)。结论SCZ和BD-M患者的肠道微生物群几个类群相对丰度发生变化。有些改变的类群如,大肠埃希菌、瘤胃球菌、纤毛杆菌、坎德勒氏菌、原发性细菌、扁平单孢菌在两种严重精神疾病患者组中都存在,有些改变的在BD-M或SCZ中是独特的,如SCZ中迪茨氏菌属的得分最高,而BD-M中大肠志贺氏杆菌的得分最高。SCZ乳酸杆菌属和丙酸杆菌属得分较BD-M高。以肠道微生物群为基础的生物标志物可能有助于SCZ和BD-M的诊断。

Objective To characterize the gut microbiota in patients with two types of severe mental illnesses(SMI)schizophrenia and bipolar affective disorder(mania),and to explore the potential of microbiota as a biomarker for schizophrenia and mania diagnosis.Methods Patients with schizophrenia(SCZ)orbipolar disorder-mania(BD-M)and healthy controls(HCs)were recruited from March to August 2019.The severity of mental symptoms was assessed by Bech-Rafaelsen Mania Rating Scale(BRMS)and schizophrenia positive and negative symptom scale(PANSS).Stool samples of patients and HCs were collected within 3 days after admission,and high-throughput 16 S rRNA microarray technology was used to analyze the differences of gut microbial composition and diversity between severe mental illness(SMI)and HCs.Operational taxonomic units(OTUs)are used to calculate multipleαdiversity index,principal co-ordinates analysis(PCoA)was carried out by using Bray Curtis dissimilarity and weighted UniFrac distance to compare the overall diversity of gut microbiota.ThroughLEfSe software,linear discriminant analysis(LDA)was used to screen the different species among groups.The random forest algorithm was used to cluster the variables of each sample at the genus level,and the first 10(average Gini importance)markers are determined to predict the importance of different genus.Results A total of 92 stool samples were collected,including 32 manic patients(BD-M),28 schizophrenic patients(SCZ)and 32 healthy controls(HCs).Overall diversity of gut microbiota:There was no significant difference inαdiversity(P>0.05)among the three groups.PCo A analysis showed that SCZ and BD-M were different from HCs.SCZ and BD-M overlapped more on the dot map and showed higher similarity.Abundance difference:A series of bacteria with different abundance were found at the genus level.At genus level,Escherichia,Ruminococcus,Leptothrix,Kandleriavitulina,Oribacterium,Planomonosporaare significantly over-presented in the SMI group(SCZ and BD-M)were significantly higher than Corynebacterium represented in HCs(P<0.05).According to the random forest algorithm,the first 10 bacterial genera were determined to identify the differences among the three groups.Between SCZ and HCs,Dietzia had the highest average decline accuracy score and average decline Gini score,and the area under curve(AUC)of ROC was 0.792(95%CI:0.679-0.906,P<0.05).Between BD-M and HCs,Escherichia shigella had the highest average decline accuracy score and average decline Gini score,and the AUC was 0.804(95%CI:0.695-0.913,P<0.05).Between SCZ and BD-M,Lactobacillus had the highest average decline accuracy score and Propionibacteriumhadthe highestaverage decline Gini score,with an AUC of 0.605(95%CI:0.459-0.751,P<0.05).Conclusions The results show that patients with SCZ and BD-M have alterations in their gut microbiota.Some of these changed taxa are shared between both conditions,such as Escherichia,Ruminococcus,Leptothrix,Kandleria,Oribacterium,Planomonosporaare significantly overpresented,while several taxa are uniquely altered in mania or schizophrenia,for example,the score of Dietzia in SCZ is the highest,while the score of Escherichia shigella in BD-M is the highest.The scores of Lactobacillus and Propionibacterium in SCZ were higher than those in BD-M.Therefore,biomarkers based on gut microbiota may be helpful in the diagnosis of SCZ and BD-M.