南方医科大学学报

南方医科大学学报 ›› 2020, Vol. 40 ›› Issue (12): 1768-1775.doi: 10.12122/j.issn.1673-4254.2020.12.11

• • 上一篇    下一篇

生育晚期各级别宫颈病变患者宫颈微生态的特征

翟青枝，任腾杰，付玉荣，张 哲，李立安，李亚里，孟元光   

  • 出版日期:2020-12-20 发布日期:2020-12-29

Characteristics of cervical microecology in late reproductive-age women with different grades of cervical lesions

  • Online:2020-12-20 Published:2020-12-29

RichHTML

27

HTML

PDF

480

摘要： 目的 探讨生育晚期宫颈病变患者宫颈微生态特征，寻找阻断宫颈病变的新途径。方法 根据入组及排除标准，纳入健康女性25例，高危型人乳头瘤病毒感染者24例，低级别宫颈上皮内瘤变（LSIL）27例，高级别宫颈上皮内瘤变（HSIL）36例，宫颈癌35例，合计147例。通过对所取的宫颈分泌物进行核酸提取，针对16S rRNA的V3+V4区域进行PCR扩增及测序。样本稀释曲线及物种箱型图证明本研究测序深度充分，样本合格。对有效数据进行OTUs分析，了解物种表达情况，并以此为基础进行α和β多样性分析。α多样性分析主要采用Chao1、Simpson和Shannon指数。β多样性分析主要采用T检验条形图、基于Weighted Unifrac Beta多样性的箱形图以及MetaStat分析。结果 5组受试者OTUs分析显示，宫颈菌落属水平丰度均以乳酸杆菌、加德纳菌、普氏菌为主，随着病变的加重乳酸杆菌丰度降低，加德纳菌属和普氏菌属丰度增高。Chao1、Simpson和Shannon指数差异均无统计学意义（P>0.05）。T检验条形图中，与健康组相比，其余各组中差异具有统计学意义的菌属为9~15个。除LSIL组外，差异菌种丰度最高的3个菌属为乳酸杆菌属（P1-2=0.025，P1-3=0.025，P1-4<0.001）、加德纳菌属（P1-2=0.01，P1-3=0.001，P1-4<0.001）和普氏菌属（P1-2=0.047，P1-3=0.023，P1-4=0.048）。LSIL组与健康人群比较差异菌属丰度最高的3个菌属分别为加德纳菌属（P1-3=0.021）、依格纳季氏菌属（P1-3=0.015）、链球菌属（P1-3=0.041）。MetaStat综合分析5组数据，其中差异具有统计学意义的菌属中丰度最高的为乳酸杆菌属（P 1-4=0.025）、加德纳菌属（P 1-2=0.004、P1-4=0.002、P1-5=0.001）及普氏菌属（P3-5=0.005）。结论 宫颈群落属水平表达均以乳酸杆菌、加德纳菌属、普氏菌属为主，并随疾病进展而变化；α多样性分析无明显差异，证实除病理因素外，生理因素也可导致α多样性的差别。LSIL组是所有病变中与健康人群最为相近的一组，这符合其疾病预后，进一步证实宫颈菌群表达与疾病预后相关，有望成为良好预后的生物学评估指标。

关键词: 宫颈菌群；宫颈病变；微生物组学；16S rRNA测序

Abstract: Objective To analyze the characteristics of cervical microecology in late reproductive-age women with cervical lesions and explore new methods for preventing cervical lesions. Methods Cervical smears were obtained from a total of 147 women of late reproductive age, including 24 with high-risk HPV infection (HR-HPV), 27 with low-grade squamous intra-epithelial lesions (LSIL), 36 with high-grade squamous intra-epithelial lesions (HSIL), 35 with cervical cancer (CC) and 25 healthy women. llumina MiSeq sequencing of V3-V4 region of the 16S rRNA gene amplicons was used to characterize the vaginal microbiota of the women. OTUs analysis of the valid data was performed, and the α-diversity (Chao1, Simpson's Index and Shannon Index) and β-diversity (T-test, weighted UniFrac β diversity, and MetaStat analysis) were evaluated. Results Dilution curve and species accumulation boxplot validated the quality of the samples. OTUs analysis of the 5 groups demonstrated that cervical bacterial genus consisted primarily of Lactobacillus, Garrotella and Prussiella. With the aggravation of the lesions, the expression abundance of Lactobacillus was decreased, and Gardnerella and Prussiella were increased. The Chao1, Simpson and Shannon indexex showed no significant difference. T test indicated that 9 to 15 genera from 4 groups showed significant difference from the healthy control group. In all but the LSIL group, Lactobacillus (P1-2=0.025, P1-3=0.025, P1-4<0.001), Gardiner (P1-2=0.01, P1-3=0.001, P1-4<0.001), and Pruella (P1-2=0.047, P1-3=0.023, P1-4=0.048) showed the highest abundance in the cervical smears. The abundance of Gardiner (P1-3=0.021), Ignatius (P1-3=0.015) and Streptococcus (P1-3=0.041) was the highest in women with LSIL as compared with healthy women. In all the 5 groups, MetaStat analysis showed that lactobacillus (P1-4=0.025), gardnella (P1-2=0.004, P1-4=0.002, P1-5=0.001) and proctella (P3-5=0.005) had the highest abundance in the cervical flora. Conclusion The abundance of Lactobacillus, Gardnella and Proctella is the highest in cervical bacteria at the genus level and may vary with disease progression. The α-diversity does not differ significantly, suggesting that apart from pathological factors, physiological factors also contribute to the difference in α-diversity. Women with LSIL have the most similar cervical flora to healthy women, which is consistent with the prognosis of the disease and confirms that the expression of cervical microecology is related to disease prognosis and may serve as a biological indicator for favoralble prognosis.

Key words: cervical microecology; cervical lesions; microbiome; 16S rRNA sequencing analysis