Colonic Microflora Protagonist of Liver Metabolism and Gut Permeability: Study on Mice Model

Abstract

Alteration of gut microflora results in a metabolic imbalance in the liver. In the present study, we investigate the reversal potential of alteration of the colonic microflora via improving metabolism balance and regulating the altered tight junction of the intestinal tract. Animals were fed with high sugar diet to mimic the onset of the pathophysiological conditions of diabetes. Following induction, animals were divided into two reversal groups i.e., crude cefdinir and colon-specific formulated cefdinir, to alter the gut microflora. In the present study, we have tried to quantify the microbial content via metagenome analysis to provide an actual picture of the alteration and subsequent reversal. Expression of mRNA of junctional protein and parameters involved in liver metabolism was determined using qPCR. Results indicated direct effect of altered composition of gut microflora on the gut permeability and metabolic alteration. Metagenomic analysis showed least evenness and richness in the HSD group whereas antibiotic-treated groups showed reversal of microflora towards control group with increased richness, evenness and decreased distance on PCoA plot. This changes in gut microflora composition changes expression of metabolic markers and thus insulin sensitivity. Targeting colonic microflora to have a reversal effect on T2D pathogenesis, found to have a positive impact on liver metabolic state with improved permeability markers of gut with SCFA alteration.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12088-022-01032-x.

Introduction

Gut microflora was identified as a key intermediate for various metabolic processes. The Short Chain Fatty Acids (SCFAs), other metabolic by-products [1], and bacterial cell wall components like Lipopolysaccharides (LPS) [2] are end products. Deviation in the state called ‘healthy microbiota’ can relate to the etiopathogenesis of illness. The Metagenomic approach is the current standard to study microflora profiles.

Type 2 Diabetes (T2D) is considered a lifestyle disease where the adoption of the western diet, over-consumption of calories, and sedentary life are the major cause of gut microflora shift [3]. The upsurge in the Gram-negative bacterial population during T2D ceases gut-permeability via disproportion of SCFAs [4]. Gut permeability is defined by tight-junction (TJ) integrity which involves the interaction of claudins with zonal occludin (ZO) and F-actin. Junctional Adhesion Molecules (JAM) encompass barrier function and cell-cell adhesion of gut epithelial cells [5, 6]. Intestinal permeability is considered as an indicator for intestinal health, increased permeability or leaky gut is the sign of inflammation caused by bacteria or its cell wall component i.e., LPS [7].

Sugar rich diet influences gut microflora profile [8]. Reversal of dysbiosis examined using antibiotic approach in this study. Cefdinir, is the Gram-negative antibiotic chosen as it is having least side effects. Our earlier work on preventive strategy using Cefdinir along with High Sugar Diet (HSD) showed decrease in Gram-negative phyla with decreased inflammatory marker gene expression such as TLR4 with improved butyrate content in fecal sample of rat animals [9]. The treatment strategies showed decreased liver inflammation with an improved Th17/Treg ratio in the liver. The immunological studies showed on elevated inflammatory environment in ‘HSD’ fed animal with increased Th17 and decreased Treg response in spleen and mesenteric lymph nodes. T2D and inflammatory conditions act as feed-forward loop [10]. Over-nutrition provided by HSD alters gut microflora which led to increase LPS in the circulatory system causing systemic inflammation. In the present study, the Metagenomic analysis and metabolite quantification has been emphasized to correlate the same with the permeability of the Gastrointestinal tract and liver metabolism. Enzymes of glycolysis and gluconeogenesis are regulated by ChREBP. Liver X receptor (LXR) regulates liver metabolism and inflammatory signaling. It further activates SREBP1c in hepatocytes which regulates fatty acid synthesis. β-oxidation pathway is regulated by PPARα. It also involved in anti-inflammatory response.

Alteration of gut microflora leads to inflammation and imbalance in liver metabolism. Here, the use of an antibiotic is to decrease the load of pathogenic bacteria and dysbiosis. By conventional cultivation methods, one cannot find out the number of species present in the gut. The Metagenomic approach used to analyse microbiome, identify the abundance of bacteria, their classification and compare diversity in different study groups [11]. In the present study, we analyzed the bacterial population through the metagenomic approach in all the experimental groups at the end of the experimentation.

Materials and Methods

Test Animals and Experimental Design

Male C57BL/6J mice having 10–12 weeks of age, were procured from Zydus Research Centre (ZRC), Ahmedabad. They were co-housed and divided into four groups with 4 animals in each group. All group of animals were given ab libitum food (chow pellet from Amrut agro foods, Mumbai) and water. The Other three groups of animals were supplemented with ab libitum High Sugar Diet (HSD) having a 65% Carbohydrate source in the form of Sucrose (20%) and Fructose (45%) to induce diabetes. Diet composition is given in supplementary table 1. After 200 days animals were found to be hyperglycaemic with blood glucose level above 150 mg/dl, the treatment was administered for 30 days, Gram-negative antibiotic- Cefdinir (10mg/kg body weight, daily) in two forms, along with HSD diet; a) the first group was given API form of cefdinir, (crude group) and b) the Second group of animals was given colon targeted cefdinir (CR group); and c) the third group remained as HSD with same diet pattern [12].

Antibiotic Preparation

Crude cefdinir is an API from. API form of cefdinir formulated using Eudragit polymer ES100. Microsphere was prepared based on oil/water emulsification method. Polymer and drug were used in a ratio of 3:1. Antibiotic was dissolved in a small amount of water, added to 5 ml of acetone organic phase (internal aqueous phase) that has 10% w/v Eudragit S100 with constant agitation. This solution was slowly injected into an external aqueous phase containing water and was emulsified by an emulsifier, Tween-20. The system was stirred continuously using a mechanical stirrer at 1000–1200 rpm for 3hrs to form a uniform emulsion. The microsphere was collected on filter paper after vacuumed based filtration and finally dried in a hot air oven at 65 °C for 15mins. Thus microsphere, thus prepared using ELS00 would provide specific release of cefdinir in Colon.

Blood Collection and Biochemical Assays

Retro orbital plexus blood collected from mouse under mild anesthetic state after model induction and treatment. Tubes were centrifuged for serum collection. The levels of glucose, total cholesterol and triglyceride, Serum Glutamate pyruvate transaminase (SGPT) and Serum Glutamate oxaloacetate transaminase (SGOT) by kit (Accucare, India) from serum. FFA estimation was performed by a pre-coated ELISA kit (Merck, Germany).

SCFA Estimation

High-Performance Liquid Chromatography (HPLC) was performed from fecal samples of animals to measure SCFA content. 100mg of excreted fecal was minced and vortexed in 1ml of acidic buffer (0.05% of H₃PO₄), which was centrifuged at 11200g for 15 minutes at 4˚C. The supernatant was collected. The final volume was made up to 1ml. 20µl of the syringed filtered sample was injected in HPLC for the analysis. For standards, 10mg/ml of acetate, propionate and butyrate (Sigma Aldrich) were prepared in the same buffer solution. Further dilutions of 1mg/ml, 0.1mg/ml, 0.01mg/ml and 0.001mg/ml were prepared for quantification of peak area. Samples were run in phenyl-hexyl column (Agilent) for 15min at 1ml/min flow rate with mentioned buffer as mobile phase [9].

Gene Expression of Gut Permeability Marker and Metabolic Marker

Gene expression studies were performed on q-PCR (Applied Bioscience) using the power-up Syber green master mix (Applied Bioscience). From the cDNA, Q-PCR cycle was carried out to check the expression pattern of the genes. Mice-specific primer for SREBP1c, CHREBP, PPAR-α, LXR for the liver; ZO, Occludin, JAM for colon were designed using BLAST (Supplementary Table 1). β-Actin was used as endogenous control; comparative quantification of the expressions was calculated using the ΔΔCT method.

Metagenome Analysis

Fresh fecal samples excreted prior to 24 hours of the experiment were collected from each group of cages, each sample thus representing the four animals, which were caged together. DNA was extracted using the Qiagen Bacterial DNA isolation kit. The DNA sample was then subjected to Metagenome analysis.

16S rRNA sequence analysis was done using a TruSeq Small RNA Sample Kit (Illumina) and was sequenced with standard sequencing oligos on the Illumina MiSeq platform with 301x2 read length. FASTQC toolkit was used for Stringent Quality control of the Illumina reads and to evaluate the base call errors. Further, reads from all the samples were combined and subjected to De-replication and De-noising followed by Clustering of operational taxonomic units (OTUs) after that Chimera Filtering. Thus, gained non-redundant and illustrative OTUs were annotated up to Species-level subsequently individual sample quantitation, as reads are mapped back to the final OTUs to assign abundances to each OTU. The above analysis was done using UPARSE Tool [13] and they were processed and analyzed further.

Sequences were denoised and trimmed to eliminate barcodes and primers. The obtained sequences were assigned into OTUs with 97% identity clustering, and the most abundant sequence from each OTU was selected as the representative sequence for that OTU. Based on the results of OTU, bacterial taxonomy was assigned using Ribosomal Database Project Classifier at a confidence level of 80%. The Complete Linkage Clustering tool of RDP was utilized to define OTUs at the species level (3% sequence divergence)

Microbiome Analyst software (http://www.microbiomeanalyst.ca/) was used for further data visualization and statistical analysis. The OTU file, metadata file and taxonomy files (.csv file format) were generated using UPARSE tools; were then uploaded to Marker data profiling (MDP) module. MDP module of Microbiome Analyst is intended for analysis of 16S rRNA marker gene survey data. Community diversity profiling was performed at different taxonomy levels. Alpha-diversity analysis was performed at the species level with Mann-Whitney/Kruskal-Wallis statistical analysis. The beta-diversity analysis was done at the species level by Principle Coordinate Analysis (PCoA) and statistical significance assessed by Bray–Curtis dissimilarity method with Permutational Multivariate Analysis of Variance (PERMANOVA) as the default option [14].

Statistical Analysis

Statistical analysis is done using GraphPad Prism v.6. The Graph represents values as mean± SEM. The significance level measured between experimental groups at significant interval of 95% using one-way ANOVA with turkey’s multiple comparison test. * In all figure indicates significance between two group connected with line, *p<005, **p<0.01, ***p<0.001, ****p<0.0001.

Results

Animal Model and Diabetic Induction

Animals were kept on an HSD diet for 200 days and given antibiotic treatment for 30 days. Post-treatment, blood was collected from the retro-orbital plexus and various parameters were checked as shown in Table 1. Biochemical parameters such as glucose, cholesterol, triglyceride, SGPT, SGOT, and Free Fatty acid (FFA) of the HSD group of animals were found to be elevated significantly compared to “Control”. The glucose level in HSD was 180±11.12 mg/dl compared to the normal range of 80-100mg/dl. Upon treatment, these values of glucose and other parameters were markedly decreased in “CR” and “Crude” compared to “HSD”. Here, the reversal groups show data comparable to that of the control group, indicating the reversal trends with CR effectively doing so, highlighting the need for formulation rather than using the crude antibiotic.

Table 1.

Biochemical parameters

Biochemical parameter	Glucose	Cholesterol	Triglyceride	SGOT	SGPT	FFA
	(mg/dl)	(mg/dl)	(mg/dl)	(IU/ml)	(IU/ml)	(μMol/L)
Control	111.79 ± 10.58	61.33 ± 8.7	55.0 ± 21.1	10.0 ± 0.3	12.3 ± 0.3	93.0 ± 6.0
HSD	180.0 ± 11.12**	135.9 ± 12.6**	95.0 ± 12.0*	78.02 ± 21.0**	75.0 ± 21.0***	143.5 ± 5.5**
Crude	119.16 ± 27.0#	52.14 ± 26.33###	46.12 ± 18.0##	18.0 ± 12##	33 ± 9##	75.0 ± 4.0##
CR	100.24 ± 41.93#	77.5 ± 8.2##	56.6 ± 18.4#	22.6 ± 12.4##	69.6 ± 4.4	62.5 ± 3.5##

Data represented here as mean ± SEM. *comparison of HSD group with Control group. #comparison with HSD group. */# p < 0.05, **/## p < 0.01, ***/### p < 0.001

SCFA Profiling

Acetate, propionate, and butyrate are major SCFA present in the stool samples. The level of acetate and propionate from stool samples was increased significantly in ‘HSD’ compared to ‘Control’ (Fig. 1). But there is no notable difference in the butyrate concentration among ‘HSD’ and ‘Control’. Gut microflora management with supplementation with crude and CR form of cefdinir resulted in a marked decrease in fecal acetate and propionate concentration compared to the ‘HSD’. ‘CR’ group showed a noteworthy increase in fecal butyrate concentration.

Fig. 1.

SCFA profiling: HPLC based quantification of Acetate, butyrate and propionate level from fecal content of mice. Two-way ANOVA was performed using GraphPad Prism. Data represented as mean ± SEM. *comparison of two group connected with line, * p < 0.05

Gut Permeability Markers

The expression of tight junction may be modulated by a colonic organisms through a various mechanisms, which is harmful to the host. Here mRNA expression of markers was checked, IHC staining could not be done. In colonic epithelium cells, expression of tight junction proteins was reduced drastically in the ‘HSD’ group. Expressions of Occludin and JAM (p<0.05) were increased in ‘CR’. The effect of a crude antibiotic on tight junction proteins’ expression was not significant (Fig. 2).

Fig. 2.

Gut permeability checking: mRNA expression of Occludin, JAM and ZO from LI region of mice. Two-way ANOVA was performed using GraphPad Prism. Data represented as mean ± SEM. *comparison of two group connected with line, * p < 0.05

Liver Metabolic Markers

The lipogenic condition can be correlated with the expression of PPAR, LXR, ChREBP and SREBPs. Carbohydrate-rich diet mainly alters their expression and hepatic insulin sensitivity [15]. Their expression in the liver was checked by qPCR. HSD group showed a reduced levels of LXR, ChREBP, SREBP and PPARα. LXR regulates the transcription of SREBP and ChREBP. Thus the reduced levels of ChREBP and SREBP can be correlated to low levels of LXR and same way increased upon ‘CR’ treatment (Fig. 3).

Fig. 3.

Metabolic parameters: mRNA expression of LXR, ChREBP, SREBP 1c and PPARα from Liver tissue of mice. Two-way ANOVA was performed using GraphPad Prism. Data represented as mean ± SEM. *comparison of two group connected with line, * p < 0.05, **p < 0.01, ***p < 0.001.

Diversity Analysis

Diversity analysis can be explained by alpha and beta diversity analysis. Alpha diversity measures richness and evenness within a sample [16]. Chao1 index measures richness which is a number of different species present in a sample. Shannon index is used to measure the evenness of a sample which is homogeneity in an abundance of the different species in a sample. Beta diversity measures microbiome composition between samples [16]. However, there was no significant differences found using the Mann-Whitney U test but R-value. The value 1 means communities are essentially high separable (Fig 4c).

Fig. 4.

Gut Microbiota diversity analysis: Alpha diversity at species level measured using a Chao1 and b Shannon estimator and were evaluated with a Mann–Whitney U test, P value = 0.39163, c 3D PCoA plot for Beta diversity of samples at species level. The analysis was performed using “Microbiome Analyst”. Dissimilarity of bacterial communities was resolute by using Principal Component Analysis (PCoA) on bray curtis distances among all groups. (Unweighted UniFrac uses phylogenetic information to compare group, PCoA is visualization tool of this information). ANOSIM was used to evaluate the UniFrac distances of all treatment vs controls. Although the P value is < 1, the R value = 1 suggests that the communities are essentially high separable. PC: principal coordinate

Animals fed with HSD showed their gut micro-floral composition is different from the control group with decreased evenness and richness of the species [fig 4a, b]. The change in microflora by colon release antibiotic (CR group) has reconstituted microflora with the highest evenness and richness of different species and the PCoA plot shows the comparison was becoming more similar to that of the animals on the control diet (Fig.4c).

Antibiotic treatment changes gut microbial composition and reduces Bacteroidetes vs Firmicute ratio in comparision to HSD (Fig 5b). Relative abundance of Bacteroidetes increased in ‘HSD’ compared to ‘Control’. Relative abundance of Lachnospiraceae, Porphyromonadaceae family decreased in HSD group which re-established upon CR treatment. Relative abundance of Bacteroidaceae, Prevotellaceae, Helicobacteraceae, Staphylococaceae, Flavobacteriaceae, Deferribactereaceae and Eubacteriaceae family outgrow in the colon of HSD group which were decreased upon CR treatment (Fig 5c).

Fig. 5.

Fecal Microbiota Composition. 16S rRNA sequencing done using MiSeq illumine System. a Percentage abundance of phyla. b Ratio of Bacteroidetes and Firmicrutes in each group calculated from the abundance table c Percentage abundance of family. “Microbiome Analyst” tool was used for visualization of bacterial abundance between study groups

Discussion

It is fascinating that the individual gut microbiota shows more diversity under healthy conditions than during illness when the diversity within the community diminishes. The disease advancement is accepted to be vital, because of the modifications in the proportion of plausibly pathogenic to useful commensal microorganisms, as opposed to the presence of a specific organism or a group [17].

Altered intestinal permeability upon dysbiosis of microflora following administration of HSD, alters metabolism and finally enhance the diabetic progression. Altered permeability of the gut, leads to infiltration of dietary fatty acids and bacterial components. Infiltrated bacterial products and FFA impair liver metabolic activities through liver receptors [18]. Acetate, propionate and butyrate are major SCFA metabolites of gut flora and act as the perfect mediators between the host and the gut microflora [19]. Level of SCFA, the diversity of microflora as well as their activities are interlinked. Bifidobacterium phyla are known to produce Acetate and propionate whereas Firmicutes are butyrate producers [20]. This fact is further confirmed in our current studies. Targeting colonic microflora using antibiotic formulation has reduced the level of acetate and propionate producers, thus reducing their level compared to ‘HSD’. Butyrate is known to increase insulin sensitivity [21, 22]. Our approach towards colonic bacteria (CR group), also improves the level of Firmicutes with improved butyrate levels, thus reducing glucose tolerance. Improved glucose sensitivity of hepatocytes led to the improved expression of PPARα, ChREBP and SREBP1c.

In this study, metagenomic analysis has revealed that Lachnospiraceae family members of Firmicutes phylum are abundant in the human gut where they ferment polysaccharides to acetate, butyrate and alcohol. Family Cryptosporangiaceae, Bifidobacteriacea and Coiobacteriacea that belong to Actinobacteria phylum was not much stimulated in any group of animals. Aneroplasmatacea family, Tenericutes phylum of bacteria increased in ‘Crude’ and ‘CR’ group of animals as compared to ‘HSD’ whereas Mycoplasmatacea class from the same phylum was decreased as compared to HSD. Deferribacteraceae family belongs to Deferribactes phylum, are flagellated and associated with some disease, were found to be high in ‘HSD’ but decreased in ‘Crude’ group of animals. Defebrribactereaceae family increase Th17 response in Gut. T cell phenotyping also showed significantly increased Th17 response in mesenteric lymph node and spleen of HSD group of mice with the release of IL17 in serum compared to control mice [10]. Class Bacteroidea, under Bacteroidetes phyla, was much increased in ‘HSD’ group of animals as compared to the ‘Control’, upon treatment with ‘Crude’ and ‘CR’ their level again returned to the level of ‘Control’ group. Few family members such as Porphyromonadaceae, Rikenellaceae from Bacteroidetes phylum are commensal whereas Bacteroidaceae and Prevotellaceae family members belong to the same phylum are opportunistic pathogens. Helicobacteraceae Family from Proteobacteria phylum are pathogenic and known to cause peptic ulcer and stomach cancer [23]. Eubacteriaceae are found in healthy individuals but their abundance increase during intestinal bowel disease [24]. Porphyromonadeceae, Bacteroidaceae, Flavobacteriaceae, Sphingobacteriaceae were increased in the ‘HSD’ group of animals. Rikenellaceae, Prevotellaceae family members are part of a healthy gut which was found to be decreased in HSD animals [25]. The Gram-positive Firmicutes phylum was decreased in HSD group and its level was controlled with the treatment of ‘CR’ and ‘Crude’ (Fig 5a). From the Firmicutes phylum ‘Control’ group of animal’s intestine was found to be rich with Lachnospiraceae family members (Roseburia intestinalis) which are gram-positive and species from this family known to produce butyrate and are involved in healthy mucosa [26]. These family members were reduced in the ‘HSD’ group upon treatment. Lactobacillus spp. and Bifidobacterium spp. are known to induce TJ barrier function [27]. Metagenomic data showed an increase in these bacteria upon treatment.

TJ barrier disruption augment paracellular permeability and leads to infusion of luminal pro-inflammatory molecules. This can activate the mucosal immune system, inflammation, and tissue damage [28]. Colon targeted microspheres (CR group) had reduced Bacteroidetes to Firmicute ratio and had increased abundance of SCFA producing genus (Bifidobacterium, various clusters of Clostridium), increased butyrate production and thus improved colon integrity with increased JAM and Occludin expression.

The metagenome approach has given insight into the significant role of microbes and their mutual association with the host. SCFAs also appear to play an important role in regulating the integrity of the epithelial barrier through the regulation of tight junction proteins. Restoration of gut microflora can be the treatment for a range of metabolic disorders. For the same knowledge of dysbiosis is a must, where metagenome approach is boon. Metagenome, with its high specificity in decoding diversity, makes it the top-tier path to achieve personalized medicine and nutrition.

As evident from the findings, CR is more effective in reversing the altered parameters comparable to that of control as against the crude. justifies the need of using a formulated approach of using an antibiotic as the test compound. Crude exerts its effect in all GI tract where as CR targets colonic microflora only. This indicates restoring colonic microflora is more effective. The primary emphasis of the study was the use of an agents to cause the restoration of induced dysbiosis. As a future prospect, colonic targeted formulation with an agent for restoration of dysbiosis can be employed, which need not be an antibiotic.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contribution

Sweta Patel has done investigation, writing of original draft and project administration. Dipeeka Mandaliya has helped in formal analysis and data curation. Sriram Seshadri has conceptualized and guided the project work.

Funding

This research work was not supported by any Funding.

Data availability

The datasets generated during the current study have been deposited in the EBI Sequence. Study accession number is PRJEB44934. https://www.ebi.ac.uk/ena/browser/view/PRJEB44934

Declarations

Conflict of interest

All The authors also declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical Approval

Animal-associated protocol was sanctioned by the ‘Institutional Animal Care and Use Committee’, Nirma University, Ahmedabad under the CPCSEA guidelines of Ministry of Environment and Forest, New Delhi (Protocol No: IS/PHD/18/024).

Consent for Publication

All the authors have gone through the manuscript and have consented to proceed with publication in this journal.