银屑病患者肠道菌群多样性分析:单中心前瞻性研究

Diversity Analysis of Intestinal Microbiota in Psoriasis Patients:A Single-center Prospective Study

目的探讨银屑病患者与健康人群肠道菌群多样性的差异。方法收集2017年5月至2018年6月间在中国医学科学院皮肤病研究所住院治疗的银屑病患者(银屑病组)及同期本院健康体检人群(健康组)的新鲜粪便标本,分析受试者相关临床资料。提取肠道菌群DNA,采用16S r DNA基因扩增和Illumina平台双端2×300策略测序,基于Gold数据库按>97%的相似性聚类操作分类单元(operational taxonomic unit,OTU),对照Silva数据库进行物种注释及分类,各层级样本采用轶和检验分析物种差异; QIIME软件计算α多样性主要指数、β多样性分析,t检验分析指数差异,P<0. 05为差异有统计学意义。相关研究结果通过R及GraphPad Prism作图展示。结果符合入选和排除标准的11例银屑病患者及21例健康对照者入选本研究,两组研究对象的性别构成比、年龄和体质量指数无统计学差异(P均>0. 05)。DNA测序分析显示样本测序覆盖深度> 0. 99。银屑病组肠道菌群的OTU数量(278. 18±89. 75比722. 95±152. 81,t=10. 36,P<0. 01)、赵氏指数(433. 38±147. 47比1156. 08±292. 50,t=9. 291,P<0. 01)、香农指数(3. 56±0. 87比5. 73±0. 78,t=6. 972,P<0. 01)和辛普森指数(0. 79±0. 15比0. 94±0. 04,t=3. 287,P <0. 01)均显著低于健康组。RankAbundance曲线显示银屑病组肠道菌群均匀程度较低。PCoA分析(unweighted)显示银屑病组与健康组在第一主成分(24. 35%)可显著分离,Weighted Uni Frac分析可见银屑病组混杂在健康样本中无法区分,且与银屑病亚型无关。样本聚类分析显示,银屑病组肠道菌群与健康组有一定重叠性,银屑病样本特异性菌群少;在门层级,健康组中可检测到TM7,相对丰度为0. 000 066 9 (0. 000 033 4～0. 000 200 5),而银屑病组仅在个别样本中显示有微量存在,相对丰度为0(P<0. 05);在属层级,银屑病组双歧杆菌属[0. 000 033 4 (0. 000 016 7～0. 000 100 3)比0. 000 401 1 (0. 000 200 5～0. 001 337 0)]、布劳特氏菌属[0. 000 467 9 (0. 000 183 8～0. 000 434 5)比0. 002 206 0 (0. 000 935 9～0. 005 582 0)]、粪球菌属[0. 000 033 4 (0～0. 000 401 1)比0. 000 902 5 (0. 000 334 2～0. 005 315 0)]较健康组显著降低,健康组中的戴阿利斯特杆菌属和嗜血菌属仅在银屑病组个别样本中微量存在,克雷伯氏杆菌属在健康组和银屑病组个别样本中存在,但在银屑病组的相对丰度更低,接近0 (P均<0. 05)。个体样本高丰度菌群分析显示,银屑病组个别样本中多糖及短链脂肪酸代谢相关菌群的相对丰度降低。结论银屑病患者肠道菌群多样性低于健康人群。

Objective The aim of this study was to investigate the difference in the diversity of intestinal microbiota of psoriasis patients compared with that of healthy people. MethodsFresh fecal samples and clinical data of hospitalized psoriasis patients from May 2017 to June 2018 in the Institute of Dermatology of Chinese Academy of Medical Sciences & Peking Union Medical College were prospectively collected. Healthy people undergoing physical examination during the same period were selected as the healthy control. After DNA extraction of intestinal microbiota and the gene amplification of 16S rDNA, the paired-end 2×300 strategy was used to sequence by Illumina platform. According to Gold database and the similarity-clustering operational taxonomic unit (OUT)>97%, intestinal microbiota was annotated and classified by Silva database. The species difference of samples at all levels was analyzed using Rank sum test. The main index of α diversity and the analysis of β diversity were calculated using QIIME software. All indexes were analyzed by Welch s t test and P <0.05 was considered statistically significant. The results were shown by R and GraphPad Prism plotting. ResultsTotally 11 psoriasis patients and 21 healthy controls were eligible for this study. Between the psoriasis group and the healthy group, the ratios of gender, age, and body mass index showed no statistical significance (all P > 0.05). The DNA sequencing showed that coverage index of samples sequencing was >0.99. The OTU numbers (278.18±89.75 vs. 722.95±152.81, t =10.36, P <0.01), Zhao index (433.38±147.47 vs. 1156.08±292.50, t =9.291, P <0.01), Shannon index (3.56±0.87 vs. 5.73 ±0.78, t =6.972, P <0.01), and Simpson index (0.79± 0.15 vs. 0.94±0.04, t =3.287, P <0.01) of the particular species of intestinal microbiota in the psoriasis group were significantly lower than those of the healthy group. The Rank-abundance curve showed that the psoriasis group had a lower evenness of intestinal microbiota. PCoA analysis (unweighted) showed that the first principal component (24.35%) was significantly separated between the two groups;however, Weighted UniFrac analysis showed that it was not clearly distinguished when the samples of the psoriasis group were mixed with the healthy group, which was independent of the subtypes of psoriasis. Through the sample clustering analysis, the intestinal microbiota partly overlapped between the two groups and the special microbiota were less in the psoriasis group. In the phylum level, TM7 could be detected in the healthy group and its relative abundance was 0.000 066 9 (0.000 033 4~ 0.000 200 5), while it was at micro-level in a few samples of the psoriatic group with a relative abundance of 0(Rank sum test, P <0.05). On the genus level, the psoriasis group had significantly lower bifidobacterium [0.000 033 4(0.000 016 7~0.000 100 3) vs. 0.000 401 1 (0.000 200 5~ 0.001 337 0 )], brauteria [0.000 467 9 (0.000 183 8~0.000 434 5) vs. 0.002 206 0 (0.000 935 9~ 0.005 582 0 )], and fecal coccus [0.000 033 4 (0~0.000 401 1) vs. 0.000 902 5 (0.000 334 2~ 0.005 315 0 )] compared to the healthy group. Dialisteria and haemophilus appearing in the healthy group were only found in a few samples in the psoriasis group. Klebsiella existed in a few samples of the two groups but was lower in the psoriasis group (relative abundance was close to 0)(Rank sum test, all P <0.05). The high abundance analysis of individual samples showed that in some psoriasis samples, the abundance of the taxa related to the metabolism of polysaccharides and short-chain fatty acids was decreased. ConclusionDiversity of intestinal microbiota in psoriasis patients is lower than that of healthy people.