A Correlation Analysis Of Distribution Characteristics Of The Vaginal Microbiome During Pregnancy And Pregnancy Outcomes For Women With Gestational Diabetes Mellitus

Research backgroundGenomes of various microorganisms, such as bacteria and viruses that live symbiotically with humans, are referred as the "microbiome," which are widely distributed in the human body. A balanced distribution of microbiomes is important for human health. At present, it is believed that an imbalanced distribution of microbiomes correlates with some of the common diseases, such as obesity, diabetes, asthma, chronic noninfectious diarrhea, inflammatory bowel disease, and irritable bowel syndrome. With the development of detection technologies, the functions of these microbiomes and their clinical relevance are receiving greater recognition. These microbes play an important role in women’s health by maintaining a homeostatic microenvironment in the vagina. Disruption of the vaginal microbiome can lead to a series of infectious diseases, such as bacterial vaginosis and fungal vaginitis, which affecting the women life quality, social interactions, and pregnancy outcomes. Lactobacillus bacteria are widely recognized as beneficial microorganisms because they inhibit the growth of other bacteria in the vagina by occupying the vaginal space, producing lactic acid, H2O2, and biotin as well as by competing for nutrition with other bacteria. At the same time, Lactobacillus populations maintain the acidic environment in the vagina (pH 3.8 to 4.4) by interacting with the immune system.Information on the vaginal microbiota is mainly obtained though microscopic examination and traditional culturing methods. On the other hand, because a large number of microbes in the vaginal microbiota belong to diverse species, some bacteria are difficult to identify by traditional culturing and isolation techniques. Nowadays, due to the rapid developments in molecular biology, molecular genetics, and second-generation high-throughput sequencing technologies, it is possible to gain a better understanding of the composition of the vaginal microbiome.Various studies have shown that the composition of the intestinal microbiome of pregnant women changes at different gestational stages due to various hormonal alterations during pregnancy. In addition, a transplant of the intestinal microbiota from late-stage pregnancy into sterile mice can induce the development of physiological characteristics of late-stage pregnancy in the mice.Composition of the vaginal microbiome is complex and is affected by a variety of factors. The traditional point of view is that vaginal pH and the levels of estrogen are important for vaginal homeostasis and for the composition of the vaginal microbiome. Drastic hormonal changes (increased levels of estrogen) during pregnancy may increase vaginal glycogen levels and cause glycogen accumulation, which is beneficial for the microbial growth that tends to disrupt a balanced vaginal microbiota. Meanwhile, other important factors for an imbalanced vaginal microbiota during pregnancy include the states of congestion and edema as well as increased permeability of the vaginal mucosa for cells, reduced functionality of the vaginal mucosal barrier, and selective suppression of the immune system for immune tolerance toward the fetus. An imbalanced vaginal microbiota is associated with the frequency of various perinatal complications, such as chorioamnionitis, premature rupture of membranes, premature birth, and puerperal infection. Thus, defining the composition and the dynamic changes of the vaginal microbiome will further the understanding of the pathogenesis of these diseases and should facilitate the development of appropriate preventive measures. Although there have been studies on the changes in the vaginal microbiome at different pregnancy stages, the following questions have yet to be answered:(1) Does sampling at different sites of the female reproductive tract affect research outcomes? (2) How do changes in the vaginal microbiome affect perinatal complications such as gestational diabetes mellitus (GDM) and its clinical significance?Research objectives1. To analyze the different characteristics of microbiomes sampled from different locations in the reproductive tract during pregnancy.2. To analyze the characteristics of microbiomes at different gestational ages. 3. To analyze the relation between the diversity of the vaginal microbiomes in pregnant women and GDM and adverse pregnancy outcomes.Research methodsThe first partEnrollment of patients:Women of childbearing age who received treatment in the maternity clinic of our hospital were randomly selected. Their age ranged from 20.1 to 38.2 years. The selected patients had not taken antibiotics or antifungal drugs 30 days prior to the sampling, and had not engaged in sexual activities, vaginal douching and/or applications of topical drugs 48 h prior to the sampling. A total of 34 patients were included in this study.Sample collection:Examination and sample collection in all cases were completed by a gynecologist, who used a sterile vaginal speculum to dilate the vaginal tract, and a vaginal swab to collect samples from the cervix (C), posterior fornix (P), and the vaginal canal (V).Experimental process:Total genomic DNA was extracted from the bacteria. A pair of bacterial 16S universal primers was used to amplify the V46 conserved region of the 16S rRNA gene from the vaginal swab samples. The total PCR products were subjected to pair end 101-bp sequencing to obtain operational taxonomic units (OTUs). Classification was carried out to determine the a-diversity and β-diversity of the samples. Based on the UniFrac distance, a phylogenetic tree of the OTUs was built. Biomarkers with substantial differences between the groups were identified, and these differences were subjected to statistical analysis.The second partPatient enrollment:Pregnant women at different pregnancy stages [middle stage (18.2±2.0 weeks) and late stage (32.3± 2.7 weeks)] and nonpregnant women were included in this study.The third partResearch subjects:GDM group:Pregnant women, who received routine obstetric examination for the first time in the maternity clinic of Nanfang Hospital, were recruited into this study. The ages ranged from 19.4 to 39.2 years (28.5±3.7 years, mean±SD). They were pregnant for 16 to 20 weeks, and a total of 340 volunteers were included. After examination by a gynecologist, patients without clinical signs or evidence of vaginal diseases were included in this study. The inclusion criteria were: ①reproductive women age between 20 to 40 years. ②Singleton pregnancies belonging to the Han population. The exclusion criteria were:①Antibiotics or antifungal drugs were taken 30 days prior to sampling, and sexual activities, vaginal douching and/or topical drug applications were performed 48 h prior to sampling. ② reproductive women who reported history of hypertension, kidney, cardiovascular, liver, and hematological diseases prior to this pregnancy. ③ The subjects had taken drugs that might interfere with glucose or lipid metabolism (for example, indomethacin, phentolamine, furosemide, thiazole diuretics, phenytoin, and hydrocortisone). ④ History of endocrine diseases (such as hyperthyroidism, hypothyroidism, or Cushing’s syndrome).⑤ Subjects who smoke nor drink.Diagnostic criteria for GDM:Fasting plasma glucose of ≥5.1 mmol/L,1-hour postprandial blood glucose of ≥10.0 mmol/L, and 2-hour postprandial blood glucose of≥8.5 mmol/L. GDM was diagnosed when one or more of these criteria were met.Control group:Pregnant women who received prenatal examination at the maternity clinic of Nanfang Hospital and did not meet the criteria for GDM were included. Other inclusion criteria were the same as those in the GDM group.Sample collection:The secretions from the posterior fornix were collected.Experimental process:As described in the first part and second part.ResultsThe first part1.1 The overall composition of the vaginal microbiome in pregnant womenLactobacillus spp. was found to be the major member of the vaginal microbiome, while Atopobium spp., Fusobacterium spp., Gardnerella spp., Hallela spp., Prevotella spp., and Streptococcus spp. constituted a small proportion of the microbiome. Among the Lactobacillus spp., the study mainly revealed four subtypes: L. crispatus, L. gasseri, L. iners, and L. jensenii.1.2 No significant difference in microbiomes at the phylum and genus levels among the three sites in the reproductive tract of pregnant womenThe reproductive tracts of the pregnant women mainly contained Firmicutes, Actinobacteria, Proteobacteria, Bacteroidetes, and Fusobacteria. The proportion of Firmicutes was more than 80%. The microbiomes from the three sampling sites, cervix (C), posterior fornix (P), and vaginal canal (V), showed no significant differences at the phylum level. At the genus level, the reproductive tracts of pregnant women mainly contained Lactobacillus:a 60% proportion of the total population. This prevalence was followed by Gardnerella, Streptococcus, Atopobium, and Prevotella. Nevertheless, the microbiomes from the three sampling sites, cervix (C), posterior fornix (P), and vaginal canal (V), showed no significant differences at the genus level.1.3 No differences in the a-diversity of microbiomes among the three sites in the reproductive tract of pregnant womenThe two indicators of a-diversity of the microbiomes from the reproductive tract, the Shannon index and the PD whole tree, were used here to estimate differences in genus diversity among the three sampling sites. The rank sum tests of the K independent samples were applied to the data, and the results revealed that the a-diversity values of the three sites showed no statistically significant differences for both the Shannon index and the PD whole tree.The second part2.1 Characteristics of the microbiome from the reproductive tract at different gestational stages and distribution of community state types (CSTs)The microbiomes from the reproductive tracts of females with middle-stage pregnancy (18.2±2.0 w) and late-stage pregnancy (32.3±2.7 w) were highly similar to those of nonpregnant (NP) women. The dominant bacteria were Lactobacillus spp.:the proportion of more than 80%. The microbiomes from the reproductive tracts of women with early-stage pregnancy (9.0±2.6 w) and those in the postpartum period (7.0±0.3 w) were characterized a greater number of cells of anaerobes such as Gardnerella, Streptococcus, Atopobium, Prevotella, and Hallella. Our data on the vaginal microbiome can be subdivided into four CSTs. CST I was dominated by L. crispatus, CST II was dominated by L. gasseri, CST III was dominated by L. iners, and CST Ⅳ-A was dominated by small numbers of Gardnerella, Atopobium, Streptococcus, Prevotella, and Hallella.2.2 β-Diversity of microbiomes from the reproductive tracts at different gestational stagesThe weighted UniFrac distance was used to express β-diversity values of the microbiomes from the reproductive tracts. Samples from the non-pregnancy period, Trimester 2, and Trimester 3 were tightly clustered, with a concentrated distribution, and hence, these data were suggestive of a high similarity. The distribution of T1 was more scattered than that of T2 and T3; this finding indicated increased bacterial diversity in T1. The distribution of bacterial populations from the postpartum group was significantly different from that in the other groups and showed a high degree of scatter, which reflected the uniqueness of the microbiome of the postpartum reproductive tract.2.3 a-Diversity of microbiomes from the reproductive tracts at different gestational stagesWe next analyzed the Shannon index at different gestational stages. In comparison with the other four groups (NP, T1,T2, and T3), the Shannon indexes of T1 and postnatal period (PP) groups were significantly higher, with the PP group showing the highest Shannon index.2.4 Species differences among the microbiomes of reproductive tracts at different gestational stagesBecause the a-diversity and P-diversity of microbiomes of reproductive tracts from different gestational stages showed significant differences, we used LEfSe to identify differences in species. In T1, the diversity of Streptococcaceae increased, while in T3, the diversity of Coriobacteriaceae and Atopobium increased. The NP group was dominated by Firmicutes, Lactobacillales, and Bacilli. In the PP group, the diversity of the dominant anaerobes such as Bacteroides, Fusobacterium, Gardnerella, Hallella, Incertae, and Prevotella increased.The third part3.1 The cluster heatmap and characteristics of vaginal microbiomes during GDMHere, all vaginal microbiomes were subjected to heatmap cluster analysis, and the CST was used to classify the vaginal microbiomes. The vaginal microbiome was subdivided into five large CSTs (CST I-V), where four types were associated to varying degrees with the genus Lactobacillus spp., namely L. crispatus (CST I,A classification analysis was carried out for 54 vaginal microbiomes from the patients with GDM. Among the these patients,13 patients showed prevalence of L iners (proportion 31.5%),17 patients showed prevalence of L. crispatus (23.5%), and 24 patients showed coexistence of anaerobes with a variety of other bacteria that accounted for the total proportion of 45.0%. The proportion of anaerobes and various other bacteria was higher in GDM vaginal microbiomes than in the non-GDM group.3.2 a-Diversity of GDM vaginal microbiomesFour indicators were used to assess the a-diversity of microbiomes, namely:the Shannon index, PD whole tree, Observed species, and Chaol. The results on the Shannon index, PD whole tree, Observed species, and Chaol showed no statistically significant differences{P> 0.05) between the GDM group and non-GDM group. This finding indicated that the diversity in the GDM vaginal microbiomes was similar to that in the non-GDM group.3.3 Factors affecting GDM vaginal microbiomesIn this study, the Pearson correlation distance was utilized to identify the factors that may be associated with GDM. CSTs, trimester stages, pH, and prepregnancy BMI (Pre-BMI) failed to show a characteristic distribution. These results revealed that these factors were not associated with GDM.3.4 Correlations between GDM-related vaginal microbiomes and adverse pregnancy outcomesCSTs during pregnancy and adverse pregnancy outcomes [which included premature rupture of membranes (PROM) and the delivery mode] did not show a characteristic clustering of vaginal microbiomes. Statistical analysis showed that the correlations (if any) of CSTs with PROM and the delivery mode were not statistically significant (χ2 test, P> 0.05).3.5 Unique species of GDM microbiomesAccording to LEfSe statistics, the diversity of Bifidobacteriaceae, Bifidobacteriales, Actinobacteria, and Lachnospiraceae was higher in the GDM group. The diversity of Pseudomonadales and Gammaproteobacteria was higher in the non-GDM group.Conclusions1. There are no significant differences among the microbiomes at different sampling sites in the female reproductive tract.2. Characteristics of the microbiomes from the reproductive tract show changes at different pregnancy stages. The microbiomes from early-stage pregnancy and from the postpartum period mainly consist of anaerobes, and the species diversity is higher than that in other periods.3. There are no significant differences in total microbiomes of reproductive tracts between the GDM and non-GDM groups, but the proportion and diversity of anaerobes are higher in the GDM group.4. There are no correlations between GDM and gestational age, prepregnancy BMI, and pH of vaginal discharge.5. There are no significant correlations between the characteristics of microbiomes from the GDM reproductive tracts and premature rupture of membranes or cesarean delivery.