Correlation between gut microbiota and glucagon-like peptide-1 in patients with gestational diabetes mellitus

Yun-Yi Liang, Ling-Yu Liu, Yan Jia, Yi Li, Jie-Na Cai, Yi Shu, Jing-Yi Tan, Pei-Yi Chen, Hong-Wei Li, Hui-Hua Cai, Xiang-Sheng Cai

Yun-Yi Liang, Yan Jia, Yi Li, Yi Shu, Jing-Yi Tan, Pei-Yi Chen, Health Management Center, The Sixth Affiliated Hospital, School of Medicine, South China University of Technology, Foshan 528000, Guangdong Province, China

Ling-Yu Liu, The First Affiliated Hospital, Henan University of Chinese Medicine, Zhengzhou 450000, Henan Province, China

Yi Li, Xiang-Sheng Cai, Shenzhen Hospital, University of Chinese Academy of Sciences, Shenzhen 518001, Guangdong Province, China

Jie-Na Cai, Clinical Laboratory, Puning People’s Hospital, Puning 515300, Guangdong Province, China

Hong-Wei Li, Institute of Biotherapy, Southern Medical University, Guangzhou 510515, Guangdong Province, China

Hui-Hua Cai, Department of Obstetrics and Gynecology, Guangdong Provincial People's Hospital, Guangzhou 510080, Guangdong Province, China

Abstract BACKGROUND Gestational diabetes mellitus (GDM) places both the mother and offspring at high risk of complications. Increasing evidence suggests that the gut microbiota plays a role in the pathogenesis of GDM. However, it is still unclear whether the gut microbiota is related to blood biochemical traits, particularly glucagon-like peptide-1 (GLP-1), in GDM patients.AIM To explore the correlation between the gut microbiota and blood biochemical traits, particularly GLP-1, in GDM patients.METHODS The V4 region of the 16S ribosomal ribonucleic acid (rRNA) gene was sequenced based on the fecal samples of 35 pregnant women with GDM and was compared to that of 25 pregnant women with normal glucose tolerance (NGT).RESULTS The results showed that Ruminococcaceae_UCG-002, Ruminococcaceae_UCG-005, Clostridium_sensu_stricto_1, and Streptococcus were more abundant in the NGT group than in the GDM group. Bacteroides and Lachnoclostridium were more abundant in the GDM group than in the NGT group. Spearman’s correlation analysis was performed to identify the relationships between microbiota genera and blood biochemical traits. Paraprevotella, Roseburia, Faecalibacterium, and Ruminococcaceae_UCG-002 were significantly negatively correlated with glucose. Ruminococcaceae_UCG-002 was significantly negatively correlated with hemoglobin A1c. Bacteroides was significantly positively correlated with glucose. Sutterella, Oscillibacter, and Bifidobacterium were significantly positively correlated with GLP-1. A random forest model showed that 20 specific genera plus glucose provided the best discriminatory power, as indicated by the area under the receiver operating characteristic curve (0.94).CONCLUSION The results of this study reveal novel relationships between the gut microbiome, blood biochemical traits, particularly GLP-1, and GDM status. These findings suggest that some genera are crucial for controlling blood glucose-related indices and may be beneficial for GDM treatment. Alteration in the microbial composition of the gut may potentially serve as a marker for identifying individuals at risk of GDM.

Key Words: Gut microbiome; Glucagon-like peptide-1; Gestational diabetes mellitus; Glucose

lNTRODUCTlON

Gestational diabetes mellitus (GDM) is defined as abnormal glucose tolerance in pregnancy[1]. It is one of the most common complications of pregnancy. The incidence of GDM has increased due to lifestyle changes, increasing maternal age, and changes to the GDM diagnostic criteria. The incidence of GDM is reported to be 13.20%. GDM is closely related to the occurrence of perinatal maternal and neonatal complications, and also significantly increases the risk of long-term metabolic diseases in pregnant women and newborns. The pathogenesis of GDM is complex and has not yet been comprehensively elucidated. Timely diagnosis and intervention are of great significance to the long-term health of pregnant women and their fetuses.

In recent years, with increased research and understanding of gastrointestinal hormones, their roles in the occurrence and development of GDM have attracted growing attention. Research suggests that GDM patients exhibit insufficient glucagon-like peptide-1 (GLP-1) secretion during pregnancy and after delivery relative to their blood glucose level. Bondeet al[2] found that the postprandial GLP-1 level of pregnant women was decreased, especially in GDM patients. However, with the recovery of blood glucose homeostasis after delivery, postprandial GLP-1 secretion gradually returned to normal. Kosinskiet al[3] confirmed that decreased GLP-1 in patients with GDM is reversible. Changes in GLP-1 levels may be related to insulin resistance (IR) as a result of high blood glucose levels. However, it cannot be ruled out that changes in GLP-1 levels may be involved in the occurrence and development of GDM.

Evidence indicates that the gut microbiota is closely related to GDM[4]. The mechanism underlying the effect of probiotics in diabetes has not yet been fully elucidated, but it may be related to reductions in oxidative stress, regulation of the immune response, reductions in inflammation, and regulation of the gut microbiota[5,6]. In addition, probiotics can also reduce postprandial blood lipid levels and improve the absorption of antioxidants, which are related to oxidative stress[7]. Numerous studies have demonstrated that GLP-1 has insulin tropic and antioxidant effects[8-10]. Since GLP-1 and the gut microbiota each play roles in GDM, is there a correlation between the two?

Both clinical and animal studies have reported correlations between changes in the GLP-1 level and changes in the gut microbiota after gastrointestinal bypass surgery in type 2 diabetes mellitus (T2DM) patients or mice[11]. Therefore, it is speculated that GLP-1 may regulate blood glucose by regulating the number and structure of gut microbiota. Several authors have argued that GLP-1 may play a role in regulating blood glucose by increasing the diversity of the gut microbiota[12] and increasing the proportion of probiotics. Together, the current literature provides a comprehensive explanation of the hypoglycemic mechanism of GLP-1 and a reliable experimental basis for the study of GDM therapeutic targets and therapeutic drugs based on GLP-1. Accordingly, one study found that bifidobacteria improved insulin sensitivity by increasing the production of GLP-1[13].

The rapid increase in the prevalence of GDM in recent years cannot be easily explained by genetic factors; thus, it has been hypothesized that environmental factors may play a more important role. The gut microbiota constitutes an important environmental factor. GLP-1, as the most important representative of gastrointestinal hormones, may also be involved in the pathogenesis of GDM. The gastrointestinal microbiota and gastrointestinal hormones share the same root, are inseparable, influence and restrict each other, and jointly participate in the occurrence, development, and prognosis of GDM. Thus, a comprehensive study of the correlations between changes in the gut microbiota and GLP-1 will help to further clarify the pathogenesis of GDM. This is of great significance for the prevention, treatment, and prognosis of GDM, and may provide a novel and sensitive index for the clinical evaluation of GDM.

MATERlALS AND METHODS

Subjects

Patients were screened for GDM in the obstetric outpatient department according to the GDM diagnostic criteria (2014). Thirty-five patients with GDM were randomly selected from the patients who met the diagnostic criteria for GDM; these patients formed the GDM group. A further 25 pregnant women with normal glucose tolerance (NGT) were selected as the NGT group. Each subject provided written informed consent before inclusion in the study. The study was approved and carried out in accordance with the guidelines of the Ethics Committee of Nanhai District People’s Hospital of Foshan.

The inclusion criteria were as follows: 18-45 years old, being female, and any education level. Based on the diagnostic criteria for gestational diabetes (2014), during the 24-28thweek of gestation, the 75 g oral glucose tolerance test (OGTT) was used to measure each patient’s blood glucose levels before, 1 h after, and 2 h after consuming sugar. If the patient’s blood glucose level reached or exceeded 5.1, 10.0, or 8.5 mmol/L, respectively, they were deemed to have GDM and were eligible for inclusion in the GDM group.

The exclusion criteria were as follows: (1) History of chronic digestive system disorder; (2) history of treatment with GLP-1 analogues or GLP-1 receptor agonists; (3) history of cardiac, renal, or liver dysfunction; (4) multiple pregnancy; (5) pregnancy-induced hypertension syndrome, placental insufficiency, placenta previa, placental abruption, pelvic or soft birth canal abnormalities, or other pregnancy complications; (6) history of mental disorders; (7) exposure to a large amount of radiation, chemical poisons, or drugs that can affect the fetus during pregnancy; (8) tumor history or history of radiotherapy and chemotherapy within the past 6 mo; (9) participation in other research studies; (10) patients lost to follow-up due to various reasons, including the occurrence of other serious diseases during the study; and (11) consumption of antibiotics or probiotics within 1 mo prior to admission.

Sample collection and testing

Fresh fecal samples were collected from the participants and immediately frozen in a refrigerator at -80 °C. After collection of all samples, they were sent to the Treat Gut company for 16S rDNA sequencing.

Blood samples were collected after fasting and then 1 h and 2 h following consumption of sugar. Plasma glucose (Glu), glycosylated serum protein (GSP), low-density lipoprotein (LDL), uric acid (UA), hemoglobin (HB), total cholesterol (TCH), triglyceride (TG), and high-density lipoprotein (HDL) were determined with a Beckman AU5800 fully automatic biochemical analyzer. Glycosylated hemoglobin A1c (HbA1c) was determined using an ADAMS? A1c HA-8180 automatic glycosylated hemoglobin analyzer. Insulin (INS), thyroid-stimulating hormone (TSH), and free tetraiodothyronine (FT4) were measured with a Maglumi2000plus automatic chemiluminescence immunoanalyzer.

The active forms of GLP-1 in the plasma samples of patients with and without GDM were measured using a GLP-1 (active) ELISA kit (ELabscience, Wuhan, Hubei Province, China).

Bacterial 16S rRNA gene sequencing

The total genomic DNA of each sample was extracted using a fecal genomic DNA extraction kit (Tiangen Company). Sixteen S rDNA sequencing was performed by PCR amplification of V4 variable regions (39 to 297 base pairs), and a purified product library was established. The library construction steps followed the library construction method of the Illumina sequencing platform. The sequencing analysis was as follows. First, the Illumina Miseq 2 × 300bp paired-end sequencing data were analyzed. According to the barcode information, the samples were distinguished. Then, the data were merged, spliced, and filtered, and quality control analysis was conducted, including Q20 and Q30 scores. The final clean data were analyzed by operational taxonomic unit (OTU) cluster analysis and species taxonomy.

Microbiome data

The data were filtered using Mothur software and clustered into OTUs (species) at a similarity level of 97% using Quantitative Insights into Microbial Ecology (QIIME) software version 1.80[14]. Based on the OTU analysis, the Ace, Shannon, observed species, Simpson, Chao1, and J indices were calculated as alpha diversity metrics. To compare the microbial composition between the samples, beta diversity analysis was performed using principal component analysis (PCA) and principal coordinate analysis (PCoA). Analysis of similarities (ANOSIM) was applied to evaluate the statistical significance of differences between the groups. A linear discriminant analysis (LDA) effect size (LEfSe) method was employed to evaluate any differences in the gut microbe between the groups.

Statistical analysis

GraphPad Prism (version 7.0) and R version 3.0.2 (R Foundation for Statistical Computing) were using for statistical analyses. The measurement data are expressed as the mean ± SD. Differences between groups were analyzed using oneway ANOVA. The differences were considered statistically significant atP＜ 0.05. Random-forest classification was performed for discriminating the samples from different groups using the R package “random forest”. The model was employed for five-fold cross-validation of the relative species abundance profile. Case probabilities were calculated by drawing receiver operating characteristic (ROC) curves.

RESULTS

Characteristics of the study population

GDM was diagnosed in 35 women based on fasting or oral glucose-stimulated hyperglycemia, or a combination of the two. Markers of glucose and insulin homeostasis were higher in the GDM group compared with the NGT group (Table 1). Individuals with GDM also had higher hemoglobin A1C (P= 0.003) and fasting blood glucose levels (P＜ 0.001). There were no significant differences in prepregnancy body weight, BMI, UA, TCH, TG, HDL, LDL, TSH, or FT4 between the two groups.

Table 1 Clinical variables of gestational diabetes mellitus patients and healthy controls

OTU distributions

In this study, the OTUs annotated included 14 phyla, 62 families, and 214 genera of gut microbiota; the similarity among samples was 97% (Figure 1A). The total number of OTUs of the NGT group (at the 97% similarity level) was 652, and for the GDM group, it was 619. Venn diagram shows that 560 OTUs were shared by the NGT and GDM groups (Figure 1B).

Alpha and beta diversities

The observed species index of the GDM group was significantly difference from that of the NGT group (25;P= 0.044). The Chao1 richness index of the GDM group was significantly different from that of the NGT group (43;P= 0.004). The ACE index of the GDM group differed significantly from that of the NGT group (25;P= 0.0055) (Figure 2). There were no significant differences in the Shannon, Simpson, or J indices between the GDM group and NGT group (Shannon,P= 0.65; Simpson,P= 0.9; J,P= 0.91; Figure 2A). PCA and PCoA indicated that the gut microbiota in GDM patients differed significantly from that of the NGT subjects. There was no difference in the gut microbiota structure between the groups (ANOSIM,r= 0.019,P= 0.2232). NMDS cluster analysis indicated marked differences between the GDM patients and NGT subjects (Figure 2B).

Taxonomy

The composition of gut microbiota was different between the groups at the phylum, family, and genus levels. At the phylum level,Bacteroidetes,Proteobacteria,Firmicutes,Verrucomicrobia,Actinobacteria,Fusobacteria,Synergistetes, andTenericuteswere common phyla in both the GDM group and NGT group, accounting for 98.81% and 98.58% of the gut bacteria of each group, respectively (Figure 3). The GDM group had a lower abundance ofFirmicutes(31.4%vs33.2%),Verrucomicrobia(0.18%vs1.01%,P＜ 0.05),Synergistetes(0.003%vs0.110%,P＜ 0.01), andTenericutes(0.05%vs0.08%), and a higher abundance ofBacteroidetes(63.50%vs60.81%),Proteobacteria(3.03%vs2.56%), andFusobacteria(0.33%vs0.29%), compared to the NGT group. The ratio ofFirmicutestoBacteroideteswas decreased in the GDM group compared to the NGT group (0.49vs0.54).

At the family level, a greater number of different families were identified between the two groups (Figure 4). Fifty-five and 54 of the dominant families were detected in the GDM group and NGT group, respectively.Bacteroidaceae(phylumBacteroidetes),Prevotellaceae(phylumBacteroidetes),Acidaminococcaceae(phylumFirmicutes),Veillonellaceae(phylumFirmicutes),Lachnospiraceae(phylumFirmicutes),Ruminococcaceae(phylumFirmicutes),Enterobacteriaceae(phylumProteobacteria), andTannerellaceae(phylumBacteroidetes) had the highest relative abundance in the GDM group, whileBacteroidaceae,Prevotellaceae,Acidaminococcaceae,Veillonellaceae,Lachnospiraceae,Enterobacteriaceae,Ruminococcaceae, andRikenellaceaewere the eight most abundant families in the NGT group.

The bacterial taxa whose levels differed significantly between the two groups were identified by LEfSE analysis (Figure 4). At the family level,Atopobiaceae,Eggerthellaceae,Streptococcaceae,Christensenellaceae,Clostridiaceae,Bifidobacteriaceae,Lachnospiraceae, andRuminococcaceaewere significantly more abundant in the NGT group than in the GDM group.

Figure 1 Operational taxonomic units distributions.

At the genus level, bacterial genera exhibited significant differences between the two groups (Figure 5). In the NGT group,Bacteroides(phylumBacteroidetes),Prevotella_9(phylumBacteroidetes),Phascolarctobacterium(phylumFirmicutes),Megasphaera(phylumFirmicutes),Megamonas(phylumFirmicutes),Lachnospiraceae(phylumFirmicutes),Escherichia-Shigella(phylumProteobacteria), andPrevotella_2(phylumBacteroidetes) were the eight most dominant genera. The eight most dominant genera in the GDM group wereBacteroides(phylumBacteroidetes),Prevotella_9(phylumBacteroidetes),Megamonas(phylumFirmicutes),Phascolarctobacterium(phylumFirmicutes),Lachnospiraceae(phylumFirmicutes),Megasphaera(phylumFirmicutes),Prevotella_2(phylumBacteroidetes), andParabacteroides(phylumBacteroidetes).

Ruminococcaceae_UCG-002,Ruminococcaceae_UCG-005,Clostridium_sensu_stricto_1,andStreptococcuswere more abundant in the NGT group than in the GDM group (P＜ 0.05).BacteroidesandLachnoclostridiumwere more abundant in the GDM group than in the NGT group (P＜ 0.05).Prevotella_9,Oscillibacter,Roseburia, andFaecalibacteriumwere slightly more abundant in the NGT group than in the GDM group.

Functional profiling of the gut microbiome

The cluster of orthologous groups (COG) categories and KEGG pathways were compared between the GDM and NGT groups. Figure 6A shows that three functional KEGG pathways differed between the GDM group and NGT group, including the glycosphingolipid biosynthesis-globo series, synthesis and degradation of ketone bodies, and renal cell carcinoma pathways. Figure 6B shows that 20 COG categories differed between the GDM group and NGT group, including the phosphotransferase system, galactitol-specific IIC component, metal-dependent proteases with possible chaperone activity, uncharacterized protein, homolog of phage Mu protein gp30, uncharacterized protein conserved in bacteria, Acyl-CoA dehydrogenases, putative virion core protein (lumpy skin disease virus), predicted phosphohydrolase, large-conductance mechanosensitive channel, uncharacterized conserved protein, uncharacterized protein predicted to be involved in DNA repair, predicted permease, DMT superfamily, nicotinic acid mononucleotide adenylyltransferase, amidases related to nicotinamidase, histone acetyltransferase, plasmid maintenance system antidote protein, uncharacterized conserved protein, DNA polymerase III, alpha subunit, uncharacterized protein conserved in bacteria, antirestriction protein, NA polymerase III, and alpha subunit (Gram-positive type).

Correlations between blood biochemical traits and gut composition

Spearman’s correlation analysis was performed to identify whether the different dominant genera were associated with blood biochemical traits in the second trimester of pregnancy (Figure 7).Paraprevotella,Roseburia,Faecalibacterium, andRuminococcaceae_UCG-002were negatively correlated with Glu (P＜ 0.05).Ruminococcaceae_UCG-002was negatively correlated with HbA1c (P＜ 0.05).Clostridium_sensu_stricto_1,Desulfovibrio, and (Ruminococcus)_torques_groupwere negatively correlated with pre-pregnancy body weight (P＜ 0.05).Phascolarctobacteriumwas negatively correlated with HDL (P＜ 0.05).Ruminococ-caceae_UCG-003andFaecalibacteriumwere negatively correlated with height (P＜ 0.05).LachnoclostridiumandLachnospiraceae_NK4A136_groupwere positively correlated with age (P＜ 0.05).Bacteroideswas significantly positively correlated with Glu (P＜ 0.01).Sutterella,Oscillibacter, andBifidobacteriumwere positively correlated with GLP-1 (P＜ 0.05).

Figure 2 Gut microbiota alpha and beta diversity indices in patients with gestational diabetes mellitus.

Figure 3 Taxonomy.

Roseburiawas negatively correlated with OGTT (0 h), OGTT (1 h), and OGTT (2 h) (P＜ 0.05).Faecalibacteriumwas negatively correlated with OGTT (0 h) and OGTT (1 h) (P＜ 0.05).Bacteroideswas positively correlated with OGTT (0 h), OGTT (1 h), OGTT (2 h), and GSP (P＜ 0.05).Lachnoclostridiumwas positively correlated with OGTT (1 h) and OGTT (2 h) (P＜ 0.05).Sutterellawas positively correlated with GLP-1(0 h), GLP-1(1 h), GLP-1(2 h), and pre-pregnancy BMI (P＜ 0.05).Oscillibacterwas positively correlated with GLP-1(0 h), GLP-1(1 h), and GLP-1(2 h) (P＜ 0.05).Bifidobacteriumwas positively correlated with GLP-1(0 h), GLP-1(1 h), OGTT (2 h), TG, and TCH (P＜ 0.05).

Gut microbiota-based prediction of GDM

Finally, random forest models were used to assess the ability of the genera abundance profiles to predict GDM status (Figure 8A). Twenty genera plus Glu provided the best discriminatory power, as indicated by the area under the receiver operating characteristic (AUROCC) value of 0.94. The value was higher than that achieved with a model including just 20 genera (the best AUC was 0.828) (Figure 8B). Further, models with 20 genera plus GLP-1, INS, or HbA1c had lower AUROCC values than the model with 20 genera plus Glu. The AUROCC values were 0.81, 0.8288, and 0.8502, respectively.

Figure 4 LEfSE analysis to determine which bacterial taxa differ significantly between the groups.

Figure 5 Bacterial genera exhibit significant differences between the two groups.

Figure 6 KEGG pathways and COG categories between the GDM and NGT groups.

Figure 7 Spearman’s correlations between different dominant genera and blood biochemical traits.

DlSCUSSlON

In recent years, the relationships between the gut microbiota and diabetes as well as other endocrine diseases have become research hotspots. Similarly, the characteristics of the gut microbiota among pregnant women with GDM have received widespread research attention. To date, research on GDM has focused on the correlation between the gut microbiota and blood glucose or insulin, but there is still a lack of research on the relationship between the gut microbiota and GLP-1. Many studies have reported that GLP-1 is closely related to the gut microbiota and short-chain fatty acids (SCFAs)[15-17], and changes in the gut microbiota directly affect the secretion of GLP-1, which, in turn, affects insulin and blood glucose. These are closely related to the occurrence of GDM. Therefore, we focused on the relationship between GLP-1 and the gut microbiota in GDM patients. To the best of our knowledge, it is the first report on the relationship between GLP-1 and the gut microbiota in patients with GDM.

At the phylum level, the abundance ofFirmicutesin the gut microbiota of the GDM group was lower than that in NGT group.Firmicutesare known to transform carbohydrates and undigested proteins into SCFAs, producing energy for the host organism. As a crucial SCFA, butyrate participates in the activation of multiple physiological signal pathways, including the proliferation and differentiation of regulatory T cells and anti-inflammatory activities[18,19]. Moreover, the GDM group exhibited reduced phylum levels ofVerrucomicrobia,Synergistetes, andTenericutescompared to the NGT group. Mucindegrading bacteriaVerrucomicrobiacontribute to glucose homeostasis and intestinal health, and play a key role in the interaction between the host tissues and gut microbiome[20]. The gut microbiota in the GDM patients also exhibited a higher abundance ofBacteroidetes,Fusobacteria,andProteobacteriacompared to healthy subjects. Proteobacteria is an opportunistic pathogen that creates a major structural imbalance in the gut microbiota of GDM patients. The ratio ofFirmicutes toBacteroidetesin GDM patients is lower than that of NGT individuals.

At the family level,Atopobiaceae,Eggerthellaceae,Streptococcaceae,Christensenellaceae,Clostridiaceae,Bifidobacteriaceae,Lachnospiraceae, andRuminococcaceaewere more abundant in the NGT subjects than in the GDM patients. Zhanget al[21] reported thatRuminococcaceae,Bifidobacteriaceae,Christensenellaceae,Erysipelotrichaceae,Peptostreptococcaceae, andEggerthellaceaewere more abundant in the NGT subjects, which is consistent with the current study. In line with the study of Maet al[22], the current results revealed thatRuminococcaceaewere more abundant in the NGT group than in the GDM group. However, other studies have observed the opposite result[23,24]. The mechanisms remain unclear.

At the genus level,Ruminococcaceae_UCG-002,Ruminococcaceae_UCG-005,Clostridium_sensu_stricto_1, andStreptococcuswere more abundant in the NGT group than in the GDM group.BacteroidesandLachnoclostridiumwere more abundant in the GDM group than in the NGT group. Kuanget al[25] found that the proportion ofBifidobacteriumin the gut microbiota of GDM pregnant women was significantly reduced, while the proportions ofBacteroidesandKlebsiellawere significantly increased. Liuet al[26] found that compared with normal pregnant women, the proportion ofFaecalisin GDM patients was significantly lower, while the proportion ofPrevotellawas significantly higher. In the present study, there were no significant differences inBifidobacterium,Klebsiella, orPrevotellabetween the GDM group and NGT group. Maet alfound thatRuminococcaceae_UCG-002andRuminococcaceae_UCG-005were reduced in women with GDM[22].

One study found that supplementation withLactobacillus rhamnosusin pregnant women may reduce the prevalence of GDM[27]. Another study showed that additional probiotic supplementation from pregnancy to 12 months post-delivery can reduce insulin levels and improve insulin sensitivity[28]. In this study, to identify beneficial bacteria for pregnant women, Spearman’s correlation analysis was performed to identify the relationships between bacterial genera and blood biochemical traits.Paraprevotella,Roseburia,Faecalibacterium, andRuminococcaceae_UCG-002were negatively correlated with Glu.Ruminococcaceae_UCG-002was negatively correlated with HbA1c.Clostridium_sensu_stricto_1,Desulfovibrio, and (Ruminococcus)_torques_groupwere negatively correlated with pre-pregnancy body weight.Phascolarctobacteriumwas negatively correlated with HDL.Ruminococcaceae_UCG-003andFaecalibacteriumwere negatively correlated with height.LachnoclostridiumandLachnospiraceae_NK4A136_groupwere positively correlated with age. Zhanget al[21] found thatRuminococcaceae_UCG-002was negatively correlated with fasting blood glucose levels. In the study of Crusell,Clostridium (sensu stricto)was positively correlated with gestational weight[29]. To the best of our knowledge, no studies have yet reported on the relationships betweenParaprevotella,Roseburia, andFaecalibacteriumand Glu in GDM. The current findings suggest that these genera are crucial for controlling blood glucose-related indices and may be beneficial for GDM treatment.

GLP-1 and its receptor agonist can promote insulin secretion only when the blood glucose level is elevated[30]. This safety feature makes GLP-1 and its agonist suitable for the treatment of GDM, which requires strict maintenance of blood glucose levels and stable, safe blood glucose regulation. Thus, the current study aimed to examine the correlations between the gut microbiota and GLP-1 levels, and identify beneficial bacteria that can improve the expression of GLP-1 in patients with GDM, so as to more safely control blood glucose. In the present study,Sutterellawas significantly positively correlated with GLP-1 (0 h), GLP-1 (1 h), GLP-1 (2 h), and pre-pregnancy BMI.Wanget al[31] reported that subjects taking metformin exhibited a significantly increased relative abundance ofSutterella, whereas liraglutide dosing was associated with a significant increase in the genusAkkermansia. Another study showed thatSutterellawas associated with C-reactive protein levels[32]. In the current study,Oscillibacterwas significantly positively correlated with GLP-1 (0 h), GLP-1 (1 h), and GLP-1 (2 h). One study reported thatCyclocarya paliurus polysaccharidesalleviated type 2 diabetic symptoms by increasing eleven SCFAproducing species, includingOscillibacter_valericigenesandOscillibacter_ruminantium[33]. Oscillibacterbelongs to the Clostridia class of Firmicutes, and in the human gut microbiota, this bacterium grow fermentatively, predominantly producing valerate when grown using glucose as a carbon source[34]. In the current study,Bifidobacteriumwas significantly positively correlated with GLP-1 (0 h), GLP-1 (1 h), TG, TCH, and OGTT (2 h).In the study of Zhaoet al[35],Bifidobacterium longum DD98improved the serum and intestinal cell GLP-1 levels, which protected pancreatic β-islet cells from damage induced by type 2 diabetes. To the best of our knowledge, this is the first report on the associations between GLP-1 and genera such asSutterella,Oscillibacter, andBifidobacteriumin GDM.

CONCLUSlON

In summary, this study contributes to a better understanding of the relationships between the gut microbiota and blood biochemical traits, particularly GLP-1, in individuals with GDM. The current findings suggest that some genera are crucial for controlling blood glucose-related indices and may be beneficial for GDM treatment. Alteration in the microbial composition of the gut may potentially serve as a marker for identifying individuals at risk of GDM. Future studies combining metagenomics and metabonomics would be of value for improving our understanding of the roles of specific strains and metabolites in patients with GDM and supporting precise prevention and intervention strategies for GDM.

ARTlCLE HlGHLlGHTS

Research background

Gestational diabetes mellitus (GDM) places both the mother and offspring at high risk of complications.Increasing evidence suggests that the gut microbiota plays a role in the pathogenesis of GDM.

Research motivation

To confirm whether the gut microbiota is related to blood biochemical traits, particularly glucagon-like peptide-1 (GLP-1), in GDM patients.

Research objectives

To explore the correlation between the gut microbiota and blood biochemical traits.

Research methods

The V4 region of the 16S ribosomal ribonucleic acid (rRNA) gene was sequenced based on the fecal samples of 35 pregnant women with GDM and was compared to that of 25 pregnant women with normal glucose tolerance (NGT).

Research results

The results showed thatRuminococcaceae_UCG-002,Ruminococcaceae_UCG-005,Clostridium_sensu_stricto_1, andStreptococcuswere more abundant in the NGT group than in the GDM group.BacteroidesandLachnoclostridiumwere more abundant in the GDM group than in the NGT group. Spearman’s correlation analysis was performed to identify the relationships between bacterial genera and blood biochemical traits.Paraprevotella,Roseburia,Faecalibacterium, andRuminococcaceae_UCG-002were significantly negatively correlated with glucose.Ruminococcaceae_UCG-002was significantly negatively correlated with HbA1c.Bacteroideswas significantly positively correlated with glucose.Sutterella,Oscillibacter, andBifidobacteriumwere significantly positively correlated with GLP-1. A random forest model showed that 20 specific genera plus glucose provided the best discriminatory power, as indicated by the area under the receiver operating characteristic curve (0.94).

Research conclusions

The results of this study reveal novel relationships between the gut microbiome, blood biochemical traits, particularly GLP-1, and GDM status.

Research perspectives

These findings suggest some genera are crucial for controlling blood glucose-related indices and may be beneficial for GDM treatment. Alteration in the microbial composition of the gut may potentially serve as a marker for identifying individuals at risk of GDM.

FOOTNOTES

Author contributions:Liang YY, Liu LY, and Jia Y designed the study, collected the samples, compiled the data, and drafted and critically revised the manuscript; Li Y, Cai JN, and Shu Y compiled the data and performed the statistical analysis and data interpretation; Tan JY collected the samples and compiled the data; Chen PY critically revised the manuscript; Cai HH and Cai XS designed this study, collected the samples, and drafted and critically revised the manuscript; all authors contributed to the article and approved the submitted version.

Supported bythe National Natural Science Foundation of China, No. 81701396; Foshan Science and Technology Planning Project, No.1920001001163; and Shenzhen Guangming District Soft Science Research Project, No. 2021R01002.

lnstitutional review board statement:The study was approved and carried out in accordance with the guidelines of the Ethics Committee of Nanhai District People’s Hospital of Foshan.

Conflict-of-interest statement:The authors declare that no competing interest exists.

Data sharing statement:Illumina sequencing reads were uploaded to the SRA under accession number PRJNA11381. The rest of the data that support the conclusions of this study are available from the corresponding author upon request.

Open-Access:This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BYNC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is noncommercial. See: https://creativecommons.org/Licenses/by-nc/4.0/

Country/Territory of origin:China

ORClD number:Xiang-Sheng Cai 0000-0002-6353-8188.

S-Editor:Chen YL

L-Editor:Wang TQ

P-Editor:Chen YX