Inulin-type Fructans Protect Against High Fat-sucrose Diet-induced Gestational Diabetes Mice in Association with Gut Microbiota Regulation

Background: Inulin-type fructans have been used as prebiotics to alleviate glucose and lipid metabolism disorders. However, few studies evaluate the microbial mechanism of inulin in improving maternal metabolic status during pregnancy. Methods: We use high fat-sucrose diet-induced gestational diabetes mice as a model to study the effect of inulin-type fructans on the cecal microbiota by 16S rRNA gene amplicon sequencing. Results: The inulin-type fructans-supplemented diet improved glucose and lipid metabolism disorder parameters in high-fat/sucrose diet (HFD) induced gestational diabetes mice, alleviating fat accumulation and glucose intolerance. The α diversity of gut microbial community of HFD mice was reduced after inulin treatment, while the β diversity tended to return to the level of normal chow diet (NCD) mice. Interestingly, Verrucomicrobia, Bi�dobacterium and Akkermansia were obviously enriched, while Dubosiella was obviously lessened after inulin treatment. Further analysis indicated that Dubosiella was positively correlated with markers of glycolipid metabolism disorders, whereas the inulin-type fructans-supplemented diet partially reversed the changes. Conclusions: Our results suggest that the inulin-type fructans-supplemented diet alleviates glucose and lipid metabolism disorders mediated by gut microbiota.


Background
Gestational diabetes mellitus (GDM), carbohydrate intolerance and insulin resistance during pregnancy are serious problems that are increasing worldwide [1], which carry signi cant short-term and long-term adverse health outcomes in both mother and offspring [2][3][4].In GDM women, the physiological changes in insulin resistance and lipid pro les are exaggerated and may indicate an underlying metabolic dysfunction that transiently manifests during pregnancy [5,6] .
Gut dysbiosis plays a vital role in abnormal host metabolism, as recently demonstrated in studies of type 2 diabetes (T2D) and obesity [7].Prevotella and Bacteroides have been identi ed as the main species contributing to insulin resistance and glucose intolerance [8].While the impact of gut microbiota on host metabolism and metabolic diseases is well-documented [9], only recently, studies have focused on microbiota changes to in uence metabolic mechanisms during pregnancy [10].Parabacteroides are signi cantly more abundant in GDM women than in healthy pregnant women [11].Novel relationship between gut microbiome composition and the metabolic hormonal environment in overweight and obese pregnant women at the rst trimester has also been described [12].These studies suggest that major shifts in the gut microbiome during pregnancy may play a crucial part in the development of GDM.
Dietary intervention to modulate the gut microbiota has become a potentially effective strategy to improve host health [13].Dietary bers, like inulin-type fructans (ITF), are present in many vegetables, and can also be extracted and isolated from chicory roots, to be used as food ingredients [14].Isolated ITF have been considered to be typical prebiotics [15].Prebiotics are de ned as non-digestible compounds that through fermentation by the microorganisms in the gut, modulate the composition and/or activity of the gut microbiota, thereby conferring a bene cial physiological effect on the host [16,17].In vitro studies and randomized controlled trials have shown that ITF can stimulate the growth of Bi dobacterium populations [18,19] and certain butyrate-producing species [20,21] as well as reduce the abundance of Firmicutes [22][23][24].In addition, numerous randomized controlled trials have demonstrated direct health bene ts from ITF, including the inhibition of pathogens, protective effects against cardiovascular disease, and the improvement of mineral bioavailability [25][26][27].However, the relationships among dietary ITF, GDM and gut microbiota are still not clear.
Given that there are few studies aiming to evaluate the microbial mechanism of soluble dietary ber in improving maternal metabolic status during pregnancy, the current research was undertaken to investigate effects of adding ITF to high-fat/sucrose diet (HFD), on the composition and metabolites of fecal microbiota from 4 weeks before conception and throughout gestation as well as maternal and neonatal health parameters in GDM mice model.It is supposed to provide some microbial mechanistic insights into the application of ITF to a typical gestational diet characterized by high fat/sucrose and energy intake for improving maternal and neonatal health.

Materials
Six-week-old C57BL/6 J mice were purchased from Vital River Laboratory Animal Technology Co., Ltd.

Animal treatment and experiment design
Mice were housed in a temperature-and humidity-controlled laboratory.Approval of this animal experiment was approved by the Animal Protection Ethics Committee of Women's Hospital of Nanjing Medical University (No.2018-49).All animal experiments were performed in accordance with Chinese national regulations on the administration of animal experimentation as well as international guidelines on animal experimentation.After one week of acclimatization, mice were randomly divided into three groups (n=5): control (normal chow diet (NCD)+vehicle, n=5), HFD (high-fat/sucrose diet (HFD)+vehicle, n=5) and inulin-type fructans treatment (HFD+ITF, N=5).In order to compare the changes of intestinal ora before and after pregnancy, the three groups were renamed normal chow diet in (NCDG) group, HFDG group and ITFG group after mating.Fasting blood glucose ( FBG) and Oral glucose tolerance test (OGTT) Blood samples were collected from the tail vein and blood glucose levels were measured with a glucose meter (Roche Accu-Chek Active, Mannheim, Germany).Fasting blood glucose (FBG) were monitored at different time points, including before dietary intervention, after 4 weeks of HFD, and on GD 0, 10, 14 and 18. Oral glucose tolerance test (OGTT) was performed on GD14.The animals were fasted for 6 h, and then gavaged with 2 g/kg glucose.The blood glucose levels at 0, 30, 60, 90 and 120 min were determined.

Detection of biochemical indexes
Mice were euthanized by CO 2 inhalation on GGD18 (or equivalent) after fasting for 6 h from 8 AM and blood sample was collected.Blood was centrifugated at 3,000 g for 15 min at 4℃ and serum was isolated.The levels of fasting serum insulin (FINS), triglyceride (TG), total cholesterol (TC), low-density lipoprotein (LDL) and high-density lipoprotein (HDL) were measured using a commercial detection kit (NJJCBIO Co., Ltd, Nanjing, China) according to the kit instructions.
Based on the measured content of FBG and FINS, the homeostasis model of assessment (HOMA) for IR index (HOMA-IRI) was calculated and compared.HOMA-IR was calculated as (fasting glucose (mmol/L) x fasting insulin (mU/L))/22.5.Meanwhile, area under the curve (AUC) of blood glucose was calculated [28].

Haematoxylin-eosin Staining
Liver and inguinal fat tissues were xed in 4% paraformaldehyde, decalci ed, para n-embedded and stored at 4°C.After tissues were sliced into 4 µm sections, haematoxylin-eosin staining was performed.First, sections were stained with haematoxylin for 5-10 minutes, immersed in 70% ethanol for 30 minutes to remove cytoplasm colouring, alkalized with alkaline solution and washed with distilled water for 1 minute.Second, sections were stained with eosin for 30-60 seconds, dehydrated with gradient ethanol, cleared two times with xylene, dried and mounted.Finally, the morphological structures of the liver and inguinal fat tissues were observed under an optical microscope.

Fecal DNA extraction
On 1 day prior to mating and GD18, fecal samples were collected in individual sterilized cages and immediately frozen in liquid nitrogen.About 100 mg of stool samples were used to extract total genome DNA according to the DNA extraction kit (DP328, Tiangen Company, Beijing, China).The concentration and purity of the extracted bacterial DNA were detected using Qubit 2.0 Fluorometer (Thermo Scienti c, USA).The 16S rRNA gene V4 region-speci c primer are 515F (GTGCCAGCMGCCGCGGTAA) and 806R GGACTACHVGGGTWTCTAAT.The PCR products of sterile water were considered as the negative control for 16S rRNA seq.The PCR products were puri ed using the Gene JET Gel Extraction Kit (Thermo Scienti c).The library was constructed using Ion Plus Fragment Library Kit 48 reactions (TermoFisher, USA).Aſter Qubit quanti cation and testing, the library was sequenced by TermoFisher's Ion S5 TM XL.

Gut microbiota analysis
Raw data were obtained aſter data were processed using Cutadapt (V1.9.1, http:// cutadapt.readthedocs.io/en/stable/).Ten, chimera sequences were removed to obtain clean reads.OTUs were assigned for sequences with ≥ 97% similarity.OTUs were annotated using the SILVA132 database (http://www.arb-silva.de/).The taxonomic information was obtained, and the community composition was counted at seven taxonomic levels: kingdom, phylum, class, order, family, genus and species.Alphadiversity was analyzed by Chao 1 (http://scikit-bio.org/docs/latest/generated/skbio.diversity.alpha.chao1.html#skbio.diversity.alpha.chao1)with QIIME soſtware (version 1.9.1).Betadiversity metrics were calculated by the NMDS model based on Bray-Curtis distance.One-way ANOSIM analysis with multiple pairwise post-tests on all groups at the same time was performed to test whether the difference between the extra groups was greater than that between the intra groups and to assess the signi cance of the difference in separation.The Chao1, Bray-Curtiss indexes, NMDS and ANOSIM were calculated at OTU level.Differentially abundant genera were analyzed by meta stats (https://omictools.com/metastats-tool ) with a nonparametric test, followed by the Benjamini and Hochberg false discovery rate approach to lter relevant p-values.

Statistical analysis
Data represent mean ± standard error of the mean.For parametric variables, the unpaired two-tailed Student t-test was used to assess the differences in mean values between two groups.For three groups, statistical analysis was performed with ANOVA with Tukey post hoc test.For nonparametric variables, the statistical signi cance of the differences was evaluated by the Mann-Whitney test or Kruskal-Wallis test.For the repeated measures in case of growth and OGTT results, multivariate ANOVA was performed with a post hoc test using Bonferroni method.P<0.05 was considered statistically signi cant.GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA) was used to do the statistical analyses.

Results
The inulin-supplemented diet can improve metabolic disorder related symptoms in high fat-sucrose diet-induced gestational diabetes mice To investigate the effect of inulin treatment on glycolipid metabolism disorders in high-fat/sucrose diet induced gestational diabetes mice, we examined the body weight, daily food intake and glycolipid metabolism related parameters.The body weight, FBG, FINS, TG, TC, LDL-C and the area under the curve (AUC) of the oral glucose tolerance test (OGTT) of the HFDG group mice were signi cantly elevated compared with those of NCDG group mice (Figure 1A-G, Figure 2A-D), indicating severe glucose intolerance and dyslipidemia.
In contrast, ITFG group mice fed the ITF-supplemented diet showed improved metabolic parameters (Figure 1A-G, Figure 2A-D).After ITF intervention, body weight, serum TG, TC and LDL-C on GD18 reduced signi cantly by 4.54 g, 0.48 mmol/l, 1.04 mmol/l and 0.494 mmol/l (p< 0.05, vs HFDG group) (Figure 1C, Figure 2A, B, D).Additionally, the AUC of OGTT on GD 14, FBG and serum insulin on GD 18 were lowered by 7.95 mmol/L/h, 2.04 mmol/l and 3.46 mIU/L (p< 0.05, vs HFDG group), indicating a signi cant improvement in glucose tolerance (Figure 1G, D, E).According to hepatic and adipose tissues staining (Figure 2E), HFDG group mice exhibited severe hepatic lipid droplets and adipocyte hypertrophy, which were alleviated after ITF treatment.Overall, the above results indicate that ITF have a bene cial effect that ameliorates glycolipid metabolism disorders in HFD mice.

Changes of Fecal Microbial Diversity
We used the 16S rRNA gene amplicon sequencing method (V4 region) and generated 2,131,728 reads for a total of 25 samples, with an average of 85269± 22,171 reads per sample.At each stage, NCD-HFD-ITF and NCDG-HFDG-ITFG pairs shared less common OTUs with each other, respectively.The Venn graph exhibited common OTUs for NCD-HFD-ITF and NCDG-HFDG-ITFG pairs decreased from 579 before mating to 438 on GD18 of gestation, respectively (Figures 3A).
To assess fecal microbial community structure, richness (Chao 1 index) and diversity (Simpson index) were calculated (Figure 3B, C).For the Chao 1 index, the data of ITF group exhibited signi cantly higher than NCD and HFDG groups (p<0.05,p<0.01).A remarkable increment in Simpson index with ITF supplementation was found compared with HFD and HFDG groups in the present study (p<0.05).All the results above provide the view that ITF addition could effectively improve the decline of Chao 1 index and Simpson index induced by high fat addition.
We then used Principal Co-ordinate Analysis to categorize the Operational Taxonomic Units (OTU) data into two main factors that explained 64.42% of the variance (Figure 3E), which showed that the microbiome in NCD (NCD and NCDG), HFD (HFD and HFDG) and ITF (ITF and ITFG) treatment groups signi cantly differed from one another while the two groups of the same treatment shared some overlapping regions before and after conception, which indicated that the overall gut microbial community had been signi cantly modi ed.The four groups exhibited signi cant, tight clustering according to NCD or ITF diet.Independent biological replicates were generally consistent, but more variable among mice fed by HFD (Figure 3D).

Changes of the Relative Abundance at Phylum Level
The phylum Bacteroidetes was dominant among the 9 phyla (>1% in at least one sample) present in the gut microbiota from the six groups of mice, and the ratio of Firmicutes/ Bacteroidetes was increased in HFD and HFDG mice over NCD and NCDG groups, but lower in the ITF and ITFG groups compared with HFD and HFDG mice (Figure 4).The gut microbiota in obese individuals has usually shown an increased Firmicutes/Bacteroidetes ratio [21].Therefore, the decreased Firmicutes/Bacteroidetes ratios of ITF and ITFG mean that this feature in obesity could be reversed by the ITF-supplemented diet.High fat treatment decreased the relative abundance of Proteobacteria before mating (p<0.01).ITF supplementation increased the relative abundance of Verrucomicrobia compared with HFD before mating and on GD18 of gestation (p<0.01).Relative abundances of Deferribacteres group of HFD and Cyanobacteria group of NCD were not detected in fecal samples on GD18 of gestation.Moreover, relative abundances of Actinobacteria decreased in HFD before mating, while increased substantially when reaching the perinatal period.The majority of genera were affected by gestation stage indicating that their relative abundances changed greatly over the pregnancy progress.

Changes of the Relative Abundance at Genus Level
The relative abundances at genus level (>1% in at least one sample) were present in Figure 5. Fat addition (HFD and HFDG) increased the relative abundances of Dubosiella and Lactobacillus and decreased that of Romboutsia and Alloprevotella compared to the NCD (NCD and NCDG).The abundance of Bi dobacterium increased, whereas that of Dubosiella decreased with the intervention of ITF before and after conception.Our results also indicated that the abundance of Akkermansia was signi cantly higher in the ITF-treated (ITF and ITFG) groups than any other group.The heat map analysis of microbial community composition at the family level con rmed that the abundance of Dubosiella that cause obesity and metabolic syndrome-related in ammation were reduced (Figure 6).
Next, to identify the changes in speci c bacterial taxa after the inulin-supplemented diet intervention before and after conception, we utilized the linear discriminant analysis (LDA) effect size (LEfSe) to compare the cecal microbiota composition between the NCD, HFD and ITF groups, LDA score was selected to discriminate speci c taxa in different groups.Compared with the HFD group, the ITF mice had a higher abundance of f-Ruminococcaceae, f-prevotellaceae, o-Verrucomicrobiales, g-Akkermansia, c-Verrucomicrobiae, p-Verrucomicrobia and f-Akkermansiaceae but lower abundance of g-Unidenti ed clostridiales, f-Unidengti ed Clostridiales, g-Dubosiella, c-Erysipelotrichia, o-Erysipelotrichales and f-Erysipelotrichaceae (Figure 7A-E).Correspondingly, g-Bacteroides, f-Ruminococcaceae and f-Bacteroidaceae were enriched in ITFG group on GD18 of gestation (Figure 8A-C).

Correlations between glycolipid metabolism indicator and bacterial abundance
At phylum level, we analyzed the correlations between signi cant glycolipid metabolism indicator and gut microbiota on GD18 of gestation.Bacteroidetes abundance was negatively correlated with FBG, FINS, TG and TC, whereas Firmicutes abundance was positively correlated with FBG, FINS and TG (Figure 9A).Moreover, Actinobacteria abundance was positively correlated with FINS and TC (Figure 9A).
At genus level, the relative abundance of Dubosiella showed positively correlated with FBG, FINS and TC (Figure 9B).Romboutsia abundance was positively correlated with FBG (Figure 9B).

Discussion
Disorder of the gut microbiota has been considered one of the reasons for metabolic disorders.The composition of the microbiome also changes during pregnancy.It has recently been proposed that intestinal micro ora and their metabolic activities (intestinal dysbiosis) may play a critical role in body weight control, energy homeostasis, fermentation, and absorption of non-digestible carbohydrate, and also in the development of IR; therefore, may also participate in the pathogenesis of several metabolic disorders, such as obesity, diabetes mellitus, and GDM [29][30][31][32].Prebiotics can exert positive effects on the maintenance of host metabolic homeostasis, which are mainly mediated by gut microbiota [33,34].ITF, one of the crucial prebiotics, has been demonstrated the effectiveness in treatment of T2DM [35,36], while data on the effects of synbiotic supplementation on markers of insulin metabolism and lipid concentrations in GDM are scarce.The aim of this study was to determine whether ITF taken before and during pregnancy, would impact the development of HFD-induced glucose intolerance during pregnancy.
To induce features of gestational diabetes, mice were fed a high-fat/sucrose diet (HFD) for 4 weeks before and during pregnancy.This model has previously been used to induce features of GDM in mice, such as insulin resistance and dyslipidemia [37][38][39].A period of only 4 weeks of HFD exposure before pregnancy is not su cient to cause a diabetic phenotype; however, continued feeding throughout pregnancy leads to progressive glucose intolerance and insulin resistance, mimicking human disease.This mouse model allowed a factorial design to determine the interaction of treatments, as well as more thorough examination of potential mechanisms and whole-tissue analysis, which would not be possible in human trials.
Our results demonstrated that ITF possessed the therapeutic efforts on HFD-induced GDM mice.The decrease of BWs after treatment of ITF indicated that dietary inulin may potentially improve the nutrition absorption and energy metabolism in the gastrointestinal tract.The metabolic routine indicators including FBG, insulin, OGTT, TG, TC and LDL-C were totally alleviated in the treatment of ITF, revealing that ITF can regulate the metabolic disorders.In consistent with our ndings, inulin-fed mice increased the production of short chain fat acids (SCFAs) to bene t the balance of gut microbiota in the alleviation of diabetic mice [40].Inulin dose-dependently decreased caloric intake, improved glucose tolerance, increased the abundance of Bacteroidetes and Bi dobacterium spp.with metabolic effects being largely independent of caloric restriction [41].A randomized-controlled clinical trial revealed that inulin administration signi cant decreased in the levels of FBG, GHb, IL-6, TNF-α and plasma LPS in T2DM patients [42,43].
Accumulating studies have been performed to reveal the underlying mechanisms of e cient treatment of ITF in GDM.The majority of mechanisms are attributed to gut microbiota alteration, immune in ammation, abnormal lipid metabolism and oxidative stress.During these mechanisms, growing evidences have demonstrated the gut microbiota play a critical role in the development of GDM [44][45][46][47].
In the present study, the Alpha diversity index reduced by HFD could be effectively improved by inulin addition.Stanislavski et al reported gestational weight gain has been associated with lower alpha diversity [48].The Beta diversity analysis of unweighted UniFrac illustrated the distinct clustering of the relative abundances of operational taxonomic units (OTUs) after ITF treatment.Moreover, similar results from PCoA analysis were obtained.
At phylum level, a higher ratio of Firmicutes to Bacteroidetes (F/B) was observed in the HFD group, which was supported by a study showing that the F/B ratio in overweight human adults was lower than that in lean controls [49].An imbalance in the Firmicutes/Bacteroidetes ratio has already been related to dysbiosis conditions [50,51].The decreased Firmicutes/Bacteroidetes ratios of ITF and ITFG mean that this feature in obesity could be reversed by the ITF-supplemented diet.Our analyses showed that after ITF treatment, an enhancement of relative abundance of Verrucomicrobia in HFD group before mating and on GD18 of gestation, as well as an obviously lessened Actinobacteria on GD18 of gestation.The Verrucomicrobia is a member of the PVC (Planctomycetes-Verrucomicrobia-Chlamydiae) superphylum which includes phylogenetically related bacteria with unusual characteristics such as the existence of a complex and dynamic endomembrane system that, in some aspects, makes them closer to eukaryotic cells.A recent study showed that the healthy chilean subjects reveals a high abundance of the phylum Verrucomicrobia [52].Positive correlations of Actinobacteria with FINS caused aggravation of insulin resistance in the disease was reversed by inulin intervention.
At genus level, ITF supplementation showed signi cant effect on increasing the abundance of Bacteroides which have been demonstrated to ameliorate in ammation in recent studies [53][54][55].SCFAs derived from gut microbiome, including acetates, propionates and butyrate, are pivotal for rectifying host metabolism and immunity [56].Signi cant elevation of SCFAs-generating Bacteroides revealed that our ITF treatment may restore gut dysbiosis by promotion of Bacteroides.Another genus that we found in abundance in fecal samples of the controls was Akkermansia.Recent studies described this as an important eubiotic genera, with systemic bene cial effects to the host [57, 58], including the control of metabolic syndromes [59,60].In rodents, probiotics supplementation with Akkermansia improved glucose tolerance and insulin sensitivity [61].Our results suggest that Akkermansia might have another impact on host physiology during pregnancy than otherwise described or that we nd another subspecies of Akkermansia.The applied 16S rRNA gene amplicon sequencing methods does, however, not make it possible to investigate this nding at a deeper taxonomic resolution.We observed that HDF mice have increased Dubosiella which has been previously described in dysbiotic conditions such as GDM and obesity [62][63][64][65][66]. Positive correlations of Dubosiella with FBG, FINS and TC demonstrated that these bacteria may promote the glycolipid metabolism disorders, but could be reversed by ITF treatment.

Conclusions
GDM is accompanied by alterations in gut microbiota composition and increasing in ammatory cascade.Dietary inulin ameliorates GDM via suppressing in ammation and modulating gut microbiota, especially in pre-diabetic and early diabetic stages, thus potentially serves as an inexpensive intervention for the prevention and treatment of T2DM patients.
Prebiotic effect of ITF might be involving in the host transcriptome changing mediated by altering the gut microbiota.Based on the results of 16S rRNA gene amplicon analysis and transcriptomic analysis, we explored the complex interplay between the cecal microbiota and host glycolipid metabolism.Giving ITF (3.33 g/kg/day) to high fat-sucrose diet-induced gestational diabetes mice can alleviate glucose and lipid metabolism disorders, increase the abundance of Verrucomicrobia, Bi dobacterium and Akkermansia, and reduce the abundance of Dubosiella.However, further studies are required to reveal the precise mechanism(s) behind these effects.A previous study suggested that the AMPK signaling pathway not only plays a pivotal role in regulating metabolism in adipose tissue, liver, and skeletal muscle but also plays a vital role in the metabolic regulation mediated by gut microbiota [67].The alteration of this similar pathway in the cecum may also play an indispensable role in the interaction between the cecal microbiota and the host metabolism system.Overall, signaling pathways related to energy metabolism in the gut are not negligible.

Figures
Figures

Figure 1 Improvement
Figure 1

Figure 2 Improvement
Figure 2

Figure 5 Relative
Figure 5

Figure 9 Correlations
Figure 9 5 .Meijer K, de Vos P, Priebe MG: Butyrate and other short-chain fatty acids as modulators of immunity: what relevance for health?Current opinion in clinical nutrition and metabolic care 2010, 13(6):715-721.57.Cani PD, de Vos Next-Generation Bene cial Microbes: The Case of Akkermansia muciniphila.
2. Li Y, Yan H, Zhang Y, Li Q, Yu L, Li Q, Liu C, Xie Y, Chen K, Ye F et al: Alterations of the Gut Microbiome Composition and Lipid Metabolic Pro le in Radiation Frontiers in cellular and infection microbiology 2020, 3. Bai YF, Wang SW, Wang XX, Weng YY, Fan XY, Sheng H, Zhu XT, Lou LJ, Zhang F: The avonoid-rich