Purpose: To investigate the potential role of gut microbiota in obesity-induced insulin resistance (IR). Methods: Four-week-old male C57BL/6 wild-type mice (n = 6) and whole-body SH2 domain-containing adaptor protein (LNK)-deficient in C57BL/6 genetic backgrounds mice (n = 7) were fed with a high-fat diet (HFD, 60% calories from fat) for 16 weeks. The gut microbiota of 13 mice feces samples was analyzed by using a 16 s rRNA sequencing analysis. Results: The structure and composition of the gut microbiota community of WT mice were significantly different from those in the LNK-/- group. The abundance of the lipopolysaccharide (LPS)-producing genus Proteobacteria was increased in WT mice, while some short-chain fatty acid (SCFA)-producing genera in WT groups were significantly lower than in LNK-/- groups (p < 0.05). Conclusions: The structure and composition of the intestinal microbiota community of obese WT mice were significantly different from those in the LNK-/- group. The abnormality of the gut microbial structure and composition might interfere with glucolipid metabolism and exacerbate obesity-induced IR by increasing LPS-producing genera while reducing SCFA-producing probiotics.
Keywords: gut microbiota; LNK deficiency; obesity; insulin resistance
Obesity is becoming a worldwide health risk factor, and obesity-induced morbidity and complications account for huge costs for affected individuals, families, healthcare systems, and society at large. Obesity is a low-grade sustained inflammatory state that alters the whole-body metabolism that frequently leads to insulin resistance (IR) [[
Gut microbiota is the general term for the microbes that inhabit the gastrointestinal tract of the human body. Around 98–99% of the intestinal microbiomes can be classified into four groups: Bacteroidetes, Firmicutes, Proteobacteria, and Actinomycetes. The balance of intestinal microbe species is the key to keeping the intestinal immune function normal and maintaining the homeostasis of the body. Breaking the balance will lead to serious pathophysiological changes, which is called gut microbiota dysbiosis [[
Metagenomic sequencing and 16S RNA sequencing were used to detect the changes in intestinal microbiota in patients with prediabetes, type 2 diabetes, and MetS. Two studies found that although the races and their diets were different, in type 2 diabetes patients, the proportion of Clostridium butyrate-producing Roche fusobacterium and Clostridium leptum decreased while the proportion of non-Clostridium butyrate increased [[
Our previous study discovered that ovarian tissues from PCOS patients with IR exhibited higher expression of the SH2 domain-containing adaptor protein (LNK) than ovaries from normal control subjects and PCOS patients without IR [[
The study protocol was approved by the Research Ethics Board of Sun Yat-sen memorial hospital of Sun Yat-sen University and Guangdong Provincial People's Hospital. All the experimental procedures were approved by the Committee for Animal Research of Sun Yat-sen University and the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals.
Four-week-old male C57BL/6 wild-type mice (n = 6) were purchased from the animal research center of Sun Yat-sen University. Whole-body LNK-deficient in C57BL/6 genetic backgrounds mice (n = 7) were created via CRISPR/Cas mediated genome engineering by Cyagen Biosciences Inc. The mouse Sh2b3 gene (GenBank accession number: NM_001306127.1; Ensembl: ENSMUSG00000042594) is located on mouse chromosome 5. Exon 1 to exon 3 were selected as target sites. Cas9 mRNA and gRNA generated using an in vitro transcription were then injected into fertilized eggs for knockout mouse production. All mice were randomly divided into different groups, housed 4 to 5 per cage, with standard laboratory conditions (12 h light:12 h darkness cycle) at a controlled temperature (23 ± 2 °C) and free access to rodent feed and water. All mice (4–5 weeks old) were fed a high-fat diet (HFD, 60% calories from fat, D12492; Research Diets Inc., New Brunswick, NJ, USA) for 16 weeks.
When mice were fed with a HFD for up to 16 weeks, fecal samples were collected and immediately kept frozen at −80 °C until processed for analysis. Total DNA was isolated from the fecal samples using the MasterPure Complete DNA&RNA Purification Kit (Epicenter) according to the manufacturer's instructions with some modifications as described previously [[
DNA was extracted using a DNA extraction kit for the corresponding sample. The concentration and purity were measured using the NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA). Next, 16S rRNA/18SrRNA/ITS genes of distinct regions (e.g., Bac 16S: V3-V4/V4/V4-V5; Fug 18S: V4/V5; ITS1/ITS2; Arc 16S: V4-V5 et al.) were amplified used specific primer (e.g., 16S: 338F and 806R/515F and 806R/515F and 907R; 18S: 528F and 706R/817F and 1196R; ITS5-1737F and ITS2-2043R/ITS3-F and ITS4R; Arc: Arch519F and Arch915R et al.) with a 12bp barcode. Primers were synthesized by Invitrogen (Invitrogen, Carlsbad, CA, USA). PCR reactions, containing 25 μL 2× Premix Taq (Takara Biotechnology, Dalian Co. Ltd., Dalian, China), 1 μL each primer (10 μM), and 3 μL DNA (20 ng/μL) template in a volume of 50 µL, were amplified via thermocycling: 5 min at 94 °C for initialization; 30 cycles of 30 s denaturation at 94 °C, 30 s annealing at 52 °C, and 30 s extension at 72 °C; followed by 10 min final elongation at 72 °C. The PCR instrument was BioRad S1000 (Bio-Rad Laboratory, Hercules, CA, USA). The length and concentration of the PCR product were detected via 1% agarose gel electrophoresis. Samples with the bright main strip between (e.g., 16S V4: 290–310 bp/16S V4V5: 400–450 bp et al.) could be used for further experiments. PCR products were mixed in equidensity ratios according to the GeneTools Analysis Software (Version 4.03.05.0, SynGene, Cambridge, UK). Then, the mixture of PCR products was purified with E.Z.N.A. Gel Extraction Kit (Omega, Bellevue, WA, USA). Next, sequencing libraries were generated using NEBNext
Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20). The primers were removed by using cutadapt software according to the primer information at the beginning and end of the sequence to obtain the paired-end clean reads. Paired-end clean reads were merged using usearch -fastq_mergepairs (V10) according to the relationship of the overlap between the paired-end reads; when with at least a 16 bp overlap, the read generated from the opposite end of the same DNA fragment, the maximum mismatch allowed in the overlap region was 5 bp, and the spliced sequences were called Raw Tags. Fastp (version 0.14.1) was used to control the quality of the raw data by sliding the window (-W 4 -M 20) to obtain the paired-end clean tags.
R software was used to count the union (pan) and intersection (core) of the target classification level in different samples to evaluate whether the sample size was sufficient. R software was used to analyze the common and endemic species, the composition of the community, and the richness of species.
A total of 13 mice (7 LNK-/-mice and 6 WT mice) were included in this study. The average body weights of 0W LNK-/-mice and WT mice were 21 g ± 2.2 g and 21.1 g ± 2.1 g, respectively, with no significance (p > 0.05). During the process of the mice fed with HFD, we observed that LNK-/- mice had a loss of appetite compared with WT mice. The food intakes of LNK-/- mice and WT mice were 18.8 g ± 1.3 g and 19.4 g ± 1.8 g, respectively, with statistical significance (p < 0.05). After 16 weeks, there was a significant difference in body weight between LNK-/- mice (47.5 g ± 4.6 g) and WT mice (52.6 g ± 3.3 g) (p < 0.05). All thirteen feces samples from seven LNK-/-mice and six WT mice were analyzed. A majority of intestinal microbe species of LNK-/-mice and WT mice were similar, however, the diversity of gut microbiomes in the WT mice group was less than that of the LNK-/- mice group (Figure 1A). The α diversity of the gut microbiota calculated using the Shannon index showed that the LNK-/- group species diversity was higher than that of the WT group at the phylum level (p < 0.05, t test) (Figure 1B,C).
To compare the composition difference in the intestinal microbiota between LNK-/- and WT mice, we next performed a Bray–Curtis-based principal coordinates analysis (PCoA) (Figure 2A). It was shown that the degree of similarity between the two groups of microbial communities was significantly different (Bray–Curtis PERMANOVA, p = 0.016). In addition, the composition of the microbiota in the samples of the LNK-/- group was more heterogeneous and significantly different from that of the WT group. The heat map showed that gut microbiota compositions between the LNK-/- and WT groups were markedly different (Figure 2B).
In the phylum-level taxonomy classification, the WT group was dominated by Proteobacteria, Verrucomicrobia, and Bacteroidetes; the LNK-/- group was dominated by Bacteroidetes, Proteobacteria, and Firmicutes (Figure 2C). Although bacteria are similar at the phylum level between the two groups, Figure 2C showed that their proportion was different. The WT group was dominated by Proteobacteria and had a relative abundance of Verrucomicrobia, with the significance compared with LNK-/- mice (p < 0.05) (Figure 2D), while the LNK-/- group has a relatively large proportion of Firmicutes (p < 0.05) and Bacteroidetes (Figure 2D).
According to the results of the linear discriminant analysis effect size (LEfSe) (LDA ≥ 2.0), the abundances of Proteobacteria, Helicobacteraceae, Epsilonproteobacteria, and Campylobacterales were significantly increased in WT mice, while the abundance of Erysipelotrichales, Allobaculum, and Bacteroidales was significantly increased in LNK-/- mice (Figure 2E).
To explore the gut microbial differences between LNK-/- and WT mice further, we used STAMP software to analyze the genera with significant differences (p < 0.05). We found that the abundances of some short-chain fatty acid (SCFA)-producing genera in the WT groups were significantly lower than in the LNK-/- groups, such as Prevotella_9, Prevotellaceae_UCG-001, Clostridium_sensu_strict_1, Ruminococcaceae_UCG-010, and Stenotrophomonas (Figure 2F).
Our previous study showed that upon the consumption of HFD, LNK-/- mice had a loss of appetite, and WT mice accumulated more intrahepatic triglyceride, TG, and FFA compared with LNK-/- mice. LNK plays a pivotal role in adipose glucose transport by regulating insulin-mediated IRS1/PI3K/Akt/AS160 signaling. In this study, we found that the abundance of Proteobacteria was significantly increased in the WT mice group, which was one of the main LPS-producing bacteria. Some SCFA-producing genera in WT groups were significantly lower than in the LNK-/- groups.
LPS is also called endotoxin. The complex of LPS and its receptor CD14 can be recognized by Toll-like receptor 4 (TLR4) on the surface of immune cells to induce an inflammatory response. When the change in diet or the use of antibiotics affects the balance of gut microbiota, the number of harmful bacteria such as G- bacteria increases, and the decomposed product LPS passes into the blood circulation through the intestinal epithelium to cause endotoxemia, which triggers a systemic inflammatory response [[
The intestinal microbiota may affect the content of circulating LPS in the following two ways to induce insulin resistance. For one thing, the structure of intestinal microbiota is unbalanced, the number of G + bacteria is decreased, the proportion of G- bacteria is increased, and the production of LPS is increased. Studies have shown that the number of G + bacteria such as Clostridium decreased and the number of LPS-containing bacteria such as Bacteroides and Proteobacteria increased in diabetic patients. Adding Lactobacillus and Bifidobacterium to the diet of high-fat-induced obese mice could help restore a balance between probiotics and pernicious bacteria in the gut and increase insulin sensitivity. The addition of prebiotic oligosaccharides to a high-fat diet-induced diabetic mouse model also increased the number of bifidobacteria, decreased the level of LPS, and improved insulin secretion and inflammation, which was significantly associated with the number of bifidobacteria [[
Probiotics such as Bifidobacterium, Lactobacillus, and Prevotella_9 can promote the release of SCFAs from the undigested soluble dietary fiber in the colon via fermentation, at the same time reducing the intestinal pH, inhibiting the growth of harmful bacteria, to reduce the production of LPS in the intestinal lumen [[
Our research explored the changes in the gut microbiota in LNK-/- and ET mice, which provided new ideas for the mechanism and treatment of MetS and IR. Although previous studies had shown that the disorder of intestine microbiota was related to MetS, the underlying mechanism remains unclear. Therefore, this study was a supplement to this research field. Nevertheless, this study still had some shortcomings. Firstly, as is known to all, sex hormones strongly influence body fat distribution and adipocyte differentiation. Estrogen and testosterone differentially affect adipocyte physiology and estrogens play a leading role in the causes and consequences of female obesity. Therefore, in this study, to avoid the influence of estrogen on the occurrence of obesity, we did not put male and female mice together to compare, and only collected fecal samples based on previous obesity-induced IR male mouse models. The sample size was not large enough, and there may be bias in the results for female mice. The results of female mice and the potential effects of sex hormones on gut microbiota need further research. Secondly, in the study, we focused on the difference in gut microbiota between LNK-/- and WT mice. We will continue relevant studies, and the indexes such as LPS, butyrate, gut permeability, and mucosal structural changes will be measured or observed in our next study. The relationship between changes in gut microbiomes and IR needs to be confirmed by further experiments. Finally, the 16S rRNA gene sequencing had some limitations, such as a short reading length, sequencing errors, and difficulty in evaluating and operating taxa. It would be important to combine signaling pathways and metabolomics analysis in the next step.
Our research described that the structure and composition of the gut microbiota community between LNK-/- and WT mice were significantly different. The change in the gut microbial structure and composition of obese WT mice might aggravate glucolipid metabolic disorder and IR by increasing the production of LPS while reducing the production of SCFAs.
DIAGRAM: Figure 1 Diversity difference in the intestinal microbiota of LNK-/- and WT mice. (A) Venn diagram of common and specific intestinal microbe species of LNK-/- and WT group. (B,C) The α diversity of the gut microbiota between the LNK-/- and WT groups, calculated using the Shannon index (* p < 0.05).
MAP: Figure 2 Composition and abundance difference in the intestinal microbiota between LNK-/- and WT mice. (A) Principal coordinate analysis (PCoA) of gut microbiota based on Bray–Curtis dissimilarity (Bray–Curtis PERMANOVA, p = 0.016). (B) The heat map of gut microbiota compositions between LNK-/- and WT group. (C) Taxonomic classification of the gut microbiota of LNK-/- and WT group at the level of phylum. (D) Boxplots of Proteobacteria, Verrucomicrobia, Firmicutes, and Bacteroidetes between LNK-/- and WT group (* p < 0.05). (E) Linear discriminant analysis of the differential abundance gut microbiota between LNK-/- and WT group. (F) Differences in bacterial genera between LNK-/- and WT group (p < 0.05).
J.C.: sample collection, data curation, formal analysis, original draft writing, review & editing. J.X., Y.S. and Y.X.: sample collection, data curation and draft review & editing. Y.Z., D.Y. and S.L.: draft review & editing and supervision. X.Z.: conceptualization, project administration, draft review & editing and supervision. All authors have read and agreed to the published version of the manuscript.
This study was approved by the Research Ethics Board of Sun Yat-sen memorial hospital of Sun Yat-sen University and Guangdong Provincial People's Hospital (2019-09). All methods were carried out in accordance with relevant guidelines and regulations.
Not applicable.
Not applicable.
The authors declare no conflict of interest.
By Jingbo Chen; Jiawen Xu; Yan Sun; Yuhuan Xue; Yang Zhao; Dongzi Yang; Shuijie Li and Xiaomiao Zhao
Reported by Author; Author; Author; Author; Author; Author; Author; Author
