Gestational diabetes mellitus (GDM), a prevalent complication in pregnancy, represents a disorder of glucose metabolism during pregnancy [1]. With economic development and the increasing number of people with overweight and obesity worldwide, approximately 7% of all pregnant women suffer from GDM. The incidence of GDM has escalated significantly in China, reaching 17.5%, partly due to the implementation of the two-child and three-child policies and partly because of the increasing incidence of geriatric pregnancy in recent years [2]. GDM has adverse outcomes for mothers, such as preeclampsia, abortion, cardiovascular disease, and type 2 diabetes mellitus (T2DM). In addition, it casts short- and long-term effects on the health of the offspring too [3], including macrosomia, preterm birth, deformity, overweight and obesity, insulin resistance, and neurocognitive decline. Therefore, pregnant women must undergo an oral glucose tolerance test (OGTT) during mid-pregnancy. Once diagnosed, doctors can immediately provide professional dietary and exercise guidance to such patients. Notably, patients with GDM tend to have good control of blood glucose. At the same time, there was no reduction in the incidence of adverse pregnancy outcomes, suggesting that other factors may also be responsible for adverse pregnancy outcomes. However, this is not the situation when patients have high blood glucose levels during their pregnancies.

Some recent studies based on metagenomics and metabolomics have shown that intestinal flora imbalance is related to many metabolic disorders, such as T2DM and GDM [4, 5]. Koren et al. [6] studied gut microbiota across different pregnancy stages in 91 healthy pregnant women and found that the overall content of Proteobacteria and Actinobacteria increased in late pregnancy while the overall abundance of the microbiota decreased. Such alterations may contribute to metabolic dysfunction and GDM. Crusell et al. [7] reported higher levels of Actinobacteria, Collinsella, Rothia, and Desulfovibrio during the third trimester of GDM compared to those in healthy controls.

However, a study has reported conflicting results regarding the gut microbiota in patients with GDM. One case-control study showed that, compared to pregnant women with normal blood sugar, those with GDM had decreased abundance of Bacteroidota and Actinobacteria [8]. In contrast, another study found no statistically significant difference in the gut microbiota of the two sets of pregnant women [9]. The previous research results are inconsistent or contradictory because of the significant differences in the design and sample size of the studies. Gut microbiota composition is influenced by various factors, including region, age, dietary habits, economic status, and gestational age. Therefore, in studies investigating the gut microbiota of patients with GDM, minimizing the impact of potential confounding factors is crucial for the accurate interpretation of results. Given the marked rise in the incidence of GDM in China, it is essential to understand the intestinal flora characteristics of patients with GDM.

In this study, we employed high-throughput sequencing technology to analyze the gut microbiota, aiming to compare the microbial composition between age-, body weight-, and gestational age-matched Chinese women with GDM and healthy pregnant controls in late pregnancy, providing a reference for subsequent interventional research on gut microbiota in pregnant women with GDM.

In this study, we analyzed the composition of the gut microbiota in women with GDM and healthy pregnant women during late pregnancy. Distinct differences were observed between the two groups. Specifically, GDM patients exhibited increased abundances of Firmicutes, Bacteroidota, Escherichia–Shigella, Romboutsia, and Paraprevotella, alongside a decreased abundance of Bifidobacterium. These alterations suggest that the aforementioned taxa may play pivotal roles in the onset and progression of GDM.

Currently, community diversity is most commonly assessed using diversity indices, with $\alpha$-diversity metrics employed to evaluate the diversity information contained within bacterial 16S rDNA gene sequences. Although several studies have reported reduced $\alpha$-diversity in GDM patients [6, 11]. For instance, a systematic review of 23 studies demonstrated an association between the gut microbiota and GDM, showing that certain bacterial taxa were correlated with altered glucose metabolism and that both $\alpha$- and $\beta$-diversity were generally reduced [12]. However, the review also highlighted a key limitation—most studies included participants with heterogeneous BMI [12]. In contrast, more recent evidence indicates that this association is not consistent. For example, Ye et al. [13] conducted a study in China examining the gut microbial and metabolic profiles of GDM patients and found no significant differences in the Shannon, Chao, or Ace indices between groups. Similarly, Crusell et al. [7], in a Danish cohort, observed that while gut microbial composition ($\beta$-diversity) differed between GDM and control groups, $\alpha$-diversity (Shannon index) remained comparable. These findings suggest that alterations in microbial composition, rather than overall diversity, may be more informative in distinguishing GDM from non-GDM pregnancies. The variability among published studies can at least partly be attributed to differences in participants’ age, body weight, and gestational age. Our findings align with these emerging reports, suggesting that GDM-related dysbiosis is more likely characterized by shifts in microbial composition ($\beta$-diversity) and fluctuations in specific bacterial taxa, rather than a consistent reduction in $\alpha$-diversity. Such inconsistencies across studies may arise from differences in population characteristics, dietary habits, diagnostic criteria, and the inherent heterogeneity of GDM itself. In our study, the two groups of pregnant women were strictly matched for age, body weight, and gestational age. Moreover, dietary assessment revealed no significant differences in total energy or macronutrient intake between the groups. Therefore, potential confounding factors were minimized as much as possible, lending greater robustness to our conclusions.

This study undertook a comparative analysis of the composition of the microbial community between the group affected by GDM and the healthy control group, revealing that at the phylum level, there existed 209 microbial communities common to both groups. A subsequent cluster analysis elucidated that the gut microbiota of both groups mainly comprised the following four phyla: Firmicutes, Proteobacteria, Actinobacteria, and Bacteroidota. Blautia, Bifidobacterium, Escherichia, Shigella, Faecalibacterium, Romboutsia, Collinella, Eubacterium_hallii_group, and Subdoligranulum are the predominant genera within the two sample groups, with Brucella, Clostridium difficile, and Romboutsia being notably dominant. These results align with those of the majority of current studies on the composition of dominant microbial communities [14, 15]. Ma et al. [16] identified the following four as the predominant bacterial phyla: Firmicutes, Bacteroidota, Proteobacteria, and Tenericutes. In the current study, the relative abundance of Firmicutes was observed to be 1%, which can be attributed to both the sample type and the study scope.

Previous studies regarding metabolic disorders, such as obesity, GDM, and metabolic syndrome, have reported inconsistent findings concerning disparities in gut microbiota species. Firmicutes and Bacteroides have been linked with obesity and T2DM [17]. Empirical evidence has demonstrated that a high-fat diet can elevate the abundance of Firmicutes, whereas a fiber-rich diet can diminish it, thereby reducing obesity and weight gain [18]. The phylum Bacteroidota is identified as the dominant gut microbiota in humans and other mammals. In terms of polysaccharide metabolism, Bacteroidota have a potent fermentation capacity, potentially supplying an additional 10–15% energy to the human body. Bäckhed et al. [19] determined that inoculating rats with Bacteroides resulted in a 23% increase in body fat and the onset of insulin resistance, which suggests that this microbial community may promote fat accumulation. This may be related to its proficiency in decomposing plant polysaccharides to facilitate nutrient absorption and energy reserves of the host. Collado et al. [20] also ascertained that Bacteroidota are prevalent among pregnant women that are overweight or have obesity. It is well-recognized that being overweight or obese is a risk factor for GDM. Thus, based on the discussion so far, we can suggest that a boom in the abundance of Firmicutes and Bacteroidota may be associated with the emergence of GDM.

This study demonstrated a marked increase in the prevalence of Firmicutes and Bacteroidota within the GDM cohort. Within the phylum Bacteroidota, both Bacteroides and prevotella showed a significant rise in the GDM cohort relative to that in healthy pregnant women. Concurrently, the levels of fasting glucose, insulin, OGTT and triglycerides were considerably elevated in the GDM cohort compared to those in healthy pregnant women. These findings further suggest that the pronounced proliferation of Firmicutes and Bacteroidota may be associated with dysregulated glucose and lipid metabolism. In fact, it was reported that those who eat plenty of protein and animal fats have Bacteroides bacteria predominantly [21]. However, our comparative analysis of three-day average dietary intake revealed no significant differences in total fat or protein consumption between the two groups. This may be attributed to dietary management interventions for patients with GDM post-diagnosis, potentially obscuring the authentic dietary patterns of both groups. Given that long-term dietary habits are strongly associated with the gut microbiome composition [22], this confounding factor deserves careful consideration. Our findings are in line with a parallel study examining the correlations between the GDM gut microbiota and blood glucose levels [23]. In contrast, another study revealed a diminished abundance of Firmicutes in the GDM group, as the study participants were in the early pregnancy stage [24]. Actinobacteria, encompassing filamentous Gram-positive bacteria, comprise various beneficial short-chain fatty acid-producing bacteria. In this study, the gut microbiota of patients with GDM showed a significant reduction in Actinobacteria and the genus Bifidobacterium compared to that in the healthy group, indicating that a decline in the beneficial Bifidobacterium coincides with the onset of GDM. An increase in the proportion of actinomycetes within the altered gut microecology of patients with GDM has also been reported [25]. This might be ascribed to the augmented presence of Collinsella bacteria within actinomycetes, which exerts a potentially adverse impact on the regulation of glucose metabolism during mid- and late-pregnancy stages.

Our study demonstrated that Shigella, Clostridium, and Romboutsia exhibited increased prevalence in the GDM cohort. Shigella, a member of the phylum Proteobacteria, primarily produces lipopolysaccharide (LPS) as its toxic agent. The LPS shows its effect by activating the Toll-like receptor-4 along its downstream pathways, thereby facilitating the translocation of endotoxins into the circulatory system and initiating persistent inflammatory reactions. Additionally, the LPS can upregulate the expression of pro-inflammatory cytokines and chemokines within the adipose tissue, impairing normal cellular functions in humans and contributing to metabolic endotoxemia, obesity, and insulin resistance [26, 27]. Two independent studies have reported enrichment of LPS biosynthesis pathways in women with GDM [28, 29]. Enhanced gut permeability allows LPS to enter the systemic circulation, eliciting inflammatory responses that impair insulin signaling and lead to glucose intolerance during pregnancy. Romboutsia is a Gram-positive bacterium capable of compromising intestinal mucosal integrity. A study has established a positive correlation between the prevalence of this bacterium in the gut microbiota and ulcerative colitis [30]. The highly virulent exotoxins and invasive enzymes synthesized by Clostridium are also capable of inducing inflammatory processes [31]. Prolonged and subclinical inflammatory states may compromise the architecture and functionality of pancreatic islet $\beta$-cells, resulting in insufficient insulin secretion and eventual progression to T2DM or GDM.

This preliminary investigation focused on bacterial colonies. A diversity analysis indicated that gut microbiota abundance was heightened in the GDM cohort. The study demonstrated a comparatively greater abundance of gut microbiota in women with GDM. In addition, it reported notable distinctions in the gut microbiota of women with GDM and healthy pregnant women. Future research should integrate multi-omics data—such as metagenomics and metabolomics—and leverage longitudinal cohorts as well as animal models to further elucidate the causal relationships and specific mechanisms linking these differential microbial communities to the pathogenesis of GDM.

ASV, amplicon sequence variant; BMI, body mass index; GDM, gestational diabetes mellitus; IDF, International Diabetes Federation; LDA, linear discriminant analysis; LEfSe, linear discriminant function analysis; LPS, lipopolysaccharide; OGTT, oral glucose tolerance test; PCoA, Principal component analysis; PE, paired-end; T2DM, type 2 diabetes mellitus.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, PRJNA1338062.

BZ and JY participated in study design, data acquisition, statistical analysis, and literature search and also drafted the manuscript. JG participated in study design, performed the research, and analyzed the data. XQW conceived the study and participated in its design and fund collection. All authors contributed to editorial changes in the manuscript. All the authors have read and approved the final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted in accordance with the Declaration of Helsinki. The research protocol was approved by the Ethics Committee of The Third Affiliated Hospital of Zunyi Medical University (The First People’s Hospital of Zunyi) (Ethic Approval Number: 2021-1-77), and all of the participants provided signed informed consent.

We would like to express our gratitude to the Department of Gynecology and Obstetrics of the First People’s Hospital of Zunyi, Guizhou, China, for their assistance in data collection, as well as to all the participants in the study.

This study was supported by the Scientific Research Project of the Guizhou Provincial Health Commission (zwkj2021-390).

The authors declare no conflict of interest.

During the preparation of this work the authors used ChatGpt-3.5 in order to check spelling and grammar. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.