Bacteria use carbohydrate-binding proteins (CBPs), such as lectins and carbohydrate-binding modules (CBMs), to anchor to specific sugars on host surfaces. CBPs in the gut microbiome are well studied, but their roles in the vagina microbiome and involvement in sexually transmitted infections, cervical cancer and preterm birth are largely unknown. We established a classification system for lectins and designed Hidden Markov Model (HMM) profiles for data mining of bacterial genomes, resulting in identification of >100,000 predicted bacterial lectins available at unilectin.eu/bacteria. Genome screening of 90 isolates from 21 vaginal bacterial species shows that those associated with infection and inflammation produce a larger CBPs repertoire, thus enabling them to potentially bind a wider array of glycans in the vagina. Both the number of predicted bacterial CBPs and their specificities correlated with pathogenicity. This study provides new insights into potential mechanisms of colonisation by commensals and potential pathogens of the reproductive tract that underpin health and disease states.

细菌使用碳水化合物结合蛋白 (CBP)，例如凝集素和碳水化合物结合模块 (CBM) 来锚定宿主表面上的特定糖。肠道微生物组中的 CBP 已得到充分研究，但它们在阴道微生物组中的作用以及与性传播感染、宫颈癌和早产的关系在很大程度上是未知的。我们为凝集素建立了分类系统，并设计了用于细菌基因组数据挖掘的隐马尔可夫模型 (HMM) 配置文件，从而在 unilectin.eu/bacteria 上鉴定了 >100,000 种预测的细菌凝集素。对来自 21 种阴道细菌物种的 90 种分离物的基因组筛选表明，与感染和炎症相关的那些会产生更大的 CBP 库，从而使它们有可能在阴道中结合更广泛的聚糖。预测的细菌 CBP 的数量及其特异性都与致病性相关。这项研究为共生体和生殖道潜在病原体定植的潜在机制提供了新的见解，这些病原体支持健康和疾病状态。