Zhang Kunna, Hu Menglu, Yang Wentao, Hu Zhexia, Rong Yun, Luo Biyun, Wang Mengjia, Cheng Yajuan, Zhang Rui, Lv Ning, Zhou Qian, Zhang Xueling
Department of Obstetrics, the First Hospital of Yongnian District, Handan, Hebei Province, China.
School of Medicine, Southeast University, Nanjing Province, China.
Heliyon. 2024 Sep 16;10(18):e37986. doi: 10.1016/j.heliyon.2024.e37986. eCollection 2024 Sep 30.
The composition of the gut microbiome has been recorted to be strongly associated with gestational diabetes mellitus (GDM), but mutational characterization of the microbiome in patients with GDM has been overlooked. Here, we revealed the genetic variation landscape of the gut microbiome and assessed its clinical significance in a cohort of patients with GDM.
We employed a macrogenomic dataset made up of a discovery cohort of 54 cases and a validation cohort of 220 cases to screen for high-abundance microbial flora and identified single nucleotide variants (SNVs) and insertions/deletions (indels). Subsequently, we analyzed the mutation spectra of genomes of the intestinal flora by using the previously identified SNVs and identified mutation signatures. Additionally, we utilized the Random Forest algorithm to identify key differential SNVs and elucidated their biological functions and associations with the clinicopathological parameters of GDM.
We screened 15 key microbial flora and found that the GDM group had more SNVs and indels in the intestinal flora than the control group, with a significant increase in C > T and T > C base mutations and were more susceptible to sequence mutations. Compared to the control group, the GDM group underwent a more significant evolution, as evidenced by the presence of a unique mutational spectrum and mutational characteristics. Random Forest algorithm analysis showed that the combined characterization of five gut microbial species and 21 SNV-related markers was effective in distinguishing between GDM and control subjects in both discovery (area under the curve (AUC) = 0.86) and validation (AUC = 0.73) sets. These markers also revealed that GDM is strongly associated with sphingolipids, galactose, and proteins containing the DUF structural domain.
The GDM intestinal flora has unique mutational features that correlate significantly with clinicopathological involvement and may be involved in the development of the disease.
肠道微生物群的组成已被记录为与妊娠期糖尿病（GDM）密切相关，但GDM患者微生物群的突变特征一直被忽视。在此，我们揭示了肠道微生物群的遗传变异图谱，并评估了其在一组GDM患者中的临床意义。
我们采用了一个由54例发现队列和220例验证队列组成的宏基因组数据集，以筛选高丰度微生物菌群，并识别单核苷酸变异（SNV）和插入/缺失（indel）。随后，我们利用先前识别的SNV分析肠道菌群基因组的突变谱，并识别突变特征。此外，我们利用随机森林算法识别关键的差异SNV，并阐明其生物学功能以及与GDM临床病理参数的关联。
我们筛选出15种关键微生物菌群，发现GDM组肠道菌群中的SNV和indel比对照组更多，C>T和T>C碱基突变显著增加，且更容易发生序列突变。与对照组相比，GDM组经历了更显著的进化，这体现在独特的突变谱和突变特征上。随机森林算法分析表明，五种肠道微生物物种和21个与SNV相关的标记物的联合特征在发现集（曲线下面积（AUC）=0.86）和验证集（AUC=0.73）中均能有效区分GDM患者和对照受试者。这些标记物还表明，GDM与鞘脂、半乳糖以及含有DUF结构域的蛋白质密切相关。
GDM肠道菌群具有独特的突变特征，与临床病理受累显著相关，可能参与了该疾病的发生发展。
