Integrated analysis of the microbiota-gut-brain axis in response to sleep deprivation in a diet-induced obesity mouse model

doi:10.21203/rs.3.rs-2200531/v1

Download PDF

Article

Integrated analysis of the microbiota-gut-brain axis in response to sleep deprivation in a diet-induced obesity mouse model

https://doi.org/10.21203/rs.3.rs-2200531/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Sleep deprivation (SD) and obesity are common in modern societies. SD and obesity frequently coexist, but research on the combined consequences of SD and obesity has been limited. In this study, we investigated the gut microbiota and host responses to SD and high-fat diet (HFD)-induced obesity. In addition, we attempted to identify key mediators of the microbiota-gut-brain axis. C57BL/6J mice were divided into four groups based on whether they were sleep deprived and whether they were fed a standard chow diet (SCD) or HFD. We then performed fecal microbiome shotgun sequencing, gut transcriptome analysis using RNA sequencing, and brain mRNA expression analysis using the nanoString nCounter Mouse Neuroinflammation Panel. The gut microbiota was significantly altered by the HFD, whereas the gut transcriptome was primarily influenced by SD. Sleep and diet are both important in the inflammatory system of the brain. When SD and the HFD were combined, the inflammatory system of the brain was severely disrupted. In addition, inosine-5' phosphate may be the gut microbial metabolite that mediates microbiota-gut-brain interactions. To identify the major drivers of this interaction, we analyzed the multi-omics data. Integrative analysis revealed two driver factors that were mostly composed of the gut microbiota. We discovered that the gut microbiota may be the primary driver of microbiota-gut-brain interactions. These findings imply that healing gut dysbiosis may be a viable therapeutic target for enhancing sleep quality and curing obesity-related dysfunction.

Health sciences/Medical research/Experimental models of disease

Biological sciences/Molecular biology/Transcription

Gut microbiota

Gut-brain axis

Sleep deprivation

Diet induced obesity

Sleep deprivation (SD) and obesity are common in modern society. In the United States, only 65 percent of individuals report sleeping for 7 hours or more per day. The average prevalence of obesity in adults was 19.5 percent across OECD nations in 2015 ¹. These two clinical conditions are now recognized as serious problems, because sleep and diet are both important for maintaining physical and mental health. Insufficient sleep leads to metabolic imbalance and an increased risk of metabolic diseases, including cardiovascular disease (CVD), type 2 diabetes mellitus (T2DM), hypertension, obesity, and depression ^2,3. Even acute SD can cause altered glucose metabolism, changes in hormone production, and weight gain ^4–6. Similarly, obesity has been linked to a variety of modern diseases, including T2DM, CVD, and other metabolic illnesses ⁷.

Over the past decade, the microbiome has emerged as a significant component of human health ⁸. Many studies have revealed that the gut microbiome communicates with the host and participates in the regulation of the systemic immune, metabolic, and nervous systems, as well as gut metabolism ⁹. The gut-brain axis (GBA) is a bidirectional communication pathway between the gut and the brain. Recent research has revealed that gut microbiota may play a role in mediating these interactions, which have been dubbed microbiota-gut-brain interactions ^9,10.

Diet and obesity are well-recognized contributors to the gut microbiome ¹¹. Obesity is a major risk factor for gut dysbiosis, and the gut microbiota contribute to metabolic dysfunction in obese individuals ¹². SD also induces dysbiosis in the gut microbiome. Sleep disturbance and loss are linked to systemic inflammation and oxidative stress in the gut ^13,14. Insomnia leads to significant structural and functional changes in the gut microbiome ¹⁵, which fluctuates in response to disturbances in sleep and circadian rhythm ¹⁶. The gut microbiota can also influence sleep quality through the GBA ¹⁷.

In modern society, there are concurrent epidemics of SD and obesity with a potential bidirectional relationship ¹⁸. While SD is a well-known risk factor for obesity ¹⁹, obesity can also cause sleep disorders, such as insomnia, obstructive sleep apnea, and obesity hypoventilation syndrome ²⁰. Although SD and obesity frequently coexist, research on the combined consequences of SD and obesity have been limited. Furthermore, to the best of our knowledge, there have been no studies on the key regulators of microbiota-gut-brain interactions in the gut and brain. In this study, we investigated the shotgun metagenomic sequencing and host gene expression profiles in response to SD and obesity in a mouse model. In addition, we performed a multi-omics factor analysis to identify a major driver of these profile alterations in response to metabolic stress caused by SD and obesity.

Mouse experimental model

Eight-week-old C57BL6/J male mice were randomly divided into four cages and fed one of two diets: a standard chow diet or high-fat diet (D12492; Research Diets, Inc., New Brunswick, NJ, USA). Sleep conditions were either SD or exercise control (EC) after eight weeks, with one day of habituation inside the wheel and five days in the sleep environment, as stated in the graphical abstract (Fig. 1A, Supplementary Fig. 1). C57BL/6J mice were then divided into four groups based on whether they were deprived of sleep and whether they were fed a standard chow diet (SCD) or high-fat diet (HFD): EC + SCD, EC + HFD, SD + SCD, and SD + HFD (number of mice per group = 3, the minimum requirement for statistics). All mice were maintained in a 12 h dark-light cycle, with the lights turning on at 7 AM and off at 7 PM.

Table 1

The significant taxa either diets or sleep parameter.
	PARAMETERS	TAXA	W	p-value (< 0.05)
PHYLA	Diets	Bacteroidetes	1	0.008
		Firmicutes	30	0.066
		F/B	36	0.005
FAMILIES	Diets	Muribaculaceae	0	0.005
		Prevotellaceae	0	0.005
		Staphylococcaceae	34	0.013
		Enterococcaceae	32	0.025
		Streptococcaceae	36	0.003
	Sleep	Tannerellaceae	3	0.020
GENERA	Diets	Muribaculaceae (unclassified)	0	0.005
		Muribaculum	0	0.004
		Prevotella	0	0.005
		Staphylococcus	34	0.013
		Enterococcus	32	0.025
		Lactococcus	36	0.003
	Sleep	Parabacteroides	3	0.020
SPECIES	Diets	Bacteroides caecimuris	5	0.045
		Bacteroides uniformis	2	0.013
		Muribaculaceae bacterium DSM 103720	0	0.005
		Muribaculum intestinale	0	0.004
		Prevotella sp MGM1	0	0.003
		Prevotella sp MGM2	4	0.031
		Staphylococcus aureus	34	0.013
		Enterococcus gallinarum	32	0.025
		Lactococcus lactis	36	0.003
		Clostridium sp ASF356	0	0.003
		Lachnospiraceae bacterium M18 1	2	0.013
		Anaerotruncus colihominis	33	0.010
	Sleep	Bacteroides sartorii	3	0.020
		Parabacteroides distasonis	5	0.045
		Parabacteroides merdae	4.5	0.026
In 4 taxonomy classification including phyla, familes, genera, species, significant taxa were measured as wilcoxon test depending on the variable, diets and sleep. The total number of each taxanomy were 7 phyla, 22 famila, 29 genera and 53 species as metaphlan results. This table only described the p-value below 0.05.

Table 2

Firmicutes, Bacteroidetes and both ratios.
Group	n	Firmicutes		Bacteroidetes		F/B
Group	n	Median	IQR	Median	IQR	Median	IQR
SD + SCD	3	13.09	11.74–20.24	67.95	64.37–75.73	0.17	0.16–0.29
SD + HFD	3	56.23	47.96–57.47	12.47	8.09–20.62	4.51	3.28–5.75
EC + SCD	3	18.24	11.91–30.63	29.28	28.81–51.63	0.25	0.22–0.86
EC + HFD	3	25.94	20.61–50.57	0.87	0.56–5.89	62.01	32.20-74.21
The median and interquartile range (IQR) of firmicutes, bacteroidetes, and those ratios were specifically explained in the table. There were 3 in each category. SD, sleep deprivation; SCD, standard chow diet; HFD, high fat diet; EC, exercise control.

There are no inclusion/exclusion criteria for mice selection for the analysis. Because C57BL/6J mouse is genetically homogeneous inbred strain.

Sleep Deprivation

SD was achieved by placing mice individually in activity wheels (Lafeyette Instruments, Lafayette, IN, USA) that had a motorized wheel with a diameter of 6.985 cm and an internal wheel with a diameter of 5.715 cm, with free access to food and water, as previously described ²¹. First, the mice were habituated to the activity wheel conditions for 24 h. We conducted experiments on the SD mice, including the SD + SCD and SD + HFD groups, over the course of five days using a slow rotational movement of the activity wheel (programmed on a schedule of 3 s 'on' and 12 s 'off') for 18 h (from ZT 06 to ZT 24). SD in this study involved forcing the mice to move to interrupt their sleep. Exercise control is necessary to avoid confusing the interpretation of experimental results because of the nonspecific effects of movement itself. For five days, mice were provided with the same walking distance (EC + SCD and EC + HFD groups) but with a wheel on/off schedule of 15 min on/15 min off at 3 m/min speed for 7.2 h, enabling them to sleep uninterrupted for longer periods.

Tissue Harvest And Rna Precipitation

Mouse tissues were all harvested at 7 AM, and the harvest time in each mouse sample did not exceed 3 min, limiting variations and stress environments between the groups. After cervical dislocation, whole blood was collected from the left ventricle of the heart tissue and perfused with 4% paraformaldehyde (PFA). Whole blood was centrifuged for 10 min at 6000 rpm, and floating serum was collected in a clean container. Blood sera were collected and stored at -80°C. Overnight storage of brain tissue was performed in a 4% PFA solution.

RNA from colon samples was extracted using TRIzol (Invitrogen, Waltham, MA, USA) and followed up according to the manufacturer’s protocol. The extracted RNA was then stored at < -80°C. The RNA integrity number of each sample was ≥ 8. The brain was divided vertically, and the right hemisphere was used for RNA extraction with a PureLink FFPE RNA Isolation Kit (Invitrogen).

Colon Rna Sequencing

Colon RNA FASTQ was automatically counted as the transcript amount using Kallisto (version 0.45.0). DESeq2 packages were used to extract the list of differentially expressed genes (DEGs) with a design matrix for diet, sleep, and interaction terms. Volcanoplot was used to compare DEG analysis results without interaction terms, and with an adjusted P-value cutoff of 0.01 and absolute value of log2 transformed fold-change 2. For pathway enrichment analysis using clusterprofiler packages (version 4.0.5), such as Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database-based pathway enrichment, the input list had significant values that were below the adjusted P-value of 0.05 for the DEseq2 results. Input data were based on the Entrez ID labeling, which indicates that non-available gene names in the Entrez ID are automatically removed. During the analysis, the number of genes enriched in KEGG function was 499 and 557 in SD vs. EC (Ref. SCD) and SD vs. EC (Ref. HFD), respectively; however, the filtered genes were automatically recognized at 479 and 541, respectively. The results of the pathway enrichment analysis based on GO biological processes were 199 and 371 in SD vs. EC (Ref. SCD) and SD vs. EC (Ref. HFD), respectively. Gene set enrichment analysis (GSEA) was performed using the GSEA software (version 4.1.0) on 3895 gene sets. The enrichment map of GSEA was programmed in Cytoscape (version 3.8.2) using the enrichmentMap tool (version 3.3.3).

Shotgun Metagenomic Sequencing

Using sterilized forceps, internal feces from the colon were collected, and colon tissue was washed with phosphate-buffered saline to remove any remaining feces and then stored at -80°C. Shotgun sequencing was used to sequence precipitated fecal DNA. The total number of reads produced from 12 samples averaged 76,004,118 ± 636,745.9436, with at least 92% of the nucleotides having a Phred quality score ≥ 30. Overall, a total of 6 different phyla, 22 families, 29 genera, and 53 species were detected using Metaphlan (version 3.0). The HUMAnN program (version 3.0) was used to measured estimated genes, as referred by uniref90, which were quantified as copies per million, and the total sum of all genes ranged from 1,294,101 to 1,467,302, depending on the samples. STAMP (version 2.1.3) and Mofapy2 (version 1.2.2) were used to examine the metagenomic data from the relative abundance results (version 0.6.4). α- and β-diversities were calculated from species-level taxa using the vegan package (version 2.5-7). In particular, for Mofapy2, significant gene sets of positive and negative weights from colon RNA sequencing and brain nCounter were manually extracted from the GO database.

Brain Nanostring Ncounter

Brain RNA samples were loaded onto the nCounter Inflammation Panel (nanoString, Seattle, WA, USA). The cartridge counted the probe-tagged fluorescent barcode with a digital analyzer equipped with a microscope. One molecule was quantified as the count value. Gene expression levels between samples were confirmed using 13 housekeeping genes: Aars, Asb10, Ccdc127, Cnot10, Csnk2a2, Fam104a, Gusb, Lars, Mto1, Supt7l, Tada2b, Tbp, and Spnpep1. Counting data from the nanoString nCounter analysis were analyzed using R software (versions 4.1.0 and 4.1.1). The DEseq2 package was used to analyze the DEGs.

Ethical statement

All experiments were reviewed and approved by the Institutional Animal Care and Use Committee of Gwangju Institute of Science and Technology (Approval number: GIST-2021-064).

High-fat diet mainly loaded gut microbiome diversity

To determine the impact of SD and HFD-induced obesity on the baseline composition of the gut microbiome, fecal samples were collected, and shotgun metagenome sequencing was performed for taxonomic profiling and functional analysis. To calculate inter-group dissimilarity (beta-diversity), we computed the Bray-Curtis dissimilarity, and unweighted and weighted Unifrac. Beta-diversity-based principal coordinate analysis plots (Fig. 1b and Supplementary Fig. 2) showed a strong distinction by diet. This means that diets had a greater impact on gut microbial diversity than that of sleep variables. In the permutational multivariate analysis of variance test (Fig. 2c), diet showed a significant effect, but sleep did not cause significant differences between the groups (Fig. 1C).

At the phylum level, the Firmicutes/Bacteroidetes (F/B) ratio is significantly associated with intestinal homeostasis ²². The HFD significantly increased the microbiota F/B ratio by increasing the content of Firmicutes; in contrast, SD decreased the F/B ratio, but the difference was not statistically significant (Tables 1 and 2). Figure 1D shows the intragroup diversity (alpha diversity) expressed using the Shannon index. The HFD decreased alpha diversity while SD increased alpha diversity, but these changes were not statistically significant. At the genus level, the HFD decreased the relative abundance of the genera Prevotella and Muribaculum and increased that of Lactococcus. SD increased the abundance of Firmicutes (Fig. 1E and 1F).

Predictive metabolomic profiling of gut microbiota following sleep deprivation and consumption of a high-fat diet

Gut microbiota composition was then examined at the species level. Figure 2A shows the diverse compositions of the bacterial species within each group. The HFD reduced the relative abundance of Muribaculum intestinale and Prevotella MGM1 and MGM2 and increased the abundance of Anaerotruncus colihominis and Lactococcus lactis. When compared to that in the control (EC + SCD group), Muribaculaceae bacteria DSM 103720 abundance increased in the SD + SCD group but not in the SD + HFD group. When SD and an HFD were combined, Ileibacterium valens was a key player (Fig. 2B). Principal component analysis (PCA) ordination plots of the relative abundance of species indicated the main drivers of each principal component (Fig. 2C).

The gut microbiota is an important part of host digestion, and this process results in hundreds of microbial metabolites ²³. Recent findings suggest that there is a bidirectional link between the brain and intestine, the so-called gut-brain axis (GBA); microbial metabolites are major mediators of this communication ^23–25. To understand the metabolic effects of sleep and diet, we performed strain-level functional pathway-enriched pathway analysis (Fig. 2D). Among the enriched pathways, inosin-5’-phosphate (5’-IMP) biosynthesis-related strains were significantly increased in the SD + HFD group compared to those in the other groups.

Transcriptome analysis of mouse large intestine after sleep deprivation and consumption of a high-fat diet

Next, to compare the effects of sleep and diet on the gut transcriptome, we performed RNA-seq analysis of the large intestine of mice from the EC + SCD, EC + HFD, SD + SCD, and SD + HFD groups. PCA was performed to reveal the major stress on the gut transcriptome between sleep and diet. Figure 3A depicts the results. Unlike the gut microbiota, which was mostly affected by diet, the gut transcriptome was primarily affected by sleep. In a DEG analysis (Fig. 3B), 90 genes were upregulated, and 26 genes were downregulated in the SD + SCD group compared to those in the EC + SCD group. Gasdermin C-like 2 (Gsdmcl2), chymase 1 (Cma1), solute carrier family 37 member 2 (Slc37a2), Alpha-2,8-sialyltransferase 8E (St8sia5), and gasdermin C4 (Gsdmc4) genes were the top five upregulated DEGs (based on P-value). Stress-associated endoplasmic reticulum protein 1 (Serp1), death-associated protein 1(Dap), transmembrane protein 35A (Tmem35a), ubiquitin-like modifier enzyme 5 (Uba5), and bone gamma-carboxyglutamate protein 3 (Bglap3) were the top five downregulated DEGs (based on P-value). In the HFD group, only 11 genes were upregulated, and 23 genes were downregulated (SD + HFD vs. EC + HFD). The Serp1 gene was also one of the top five downregulated genes in the SD + HFD group compared to that in the EC + HFD group (Fig. 3B).

Gsdmcl2, Cma1, Slc37a2, and St8sia5 were downregulated in the SD + HFD group compared to those in the EC + SCD group, but were upregulated in the SD + SCD group compared to those in the EC + SCD group (Fig. 3B). After SD, the tumor protein D52-like 1 (Tpd52l1), cellular communication network factor 3 (Ccn3), and UDP-GlcNAc:betaGal beta-1,3-N-acetylglucosaminyltransfera 9 (B3gnt9) genes were downregulated in both SCD- and HFD-fed mice (Supplementary Fig. 3).

The KEGG enrichment analysis indicated that DEGs between the SD + SCD and EC + SCD groups were enriched in digestive system-related pathways, including “Pancreatic secretion” and “Protein digestion and absorption” (Fig. 3C). The DEGs between the SD + HFD and EC + HFD groups were enriched in nucleic acid metabolism-related and lipid metabolism-related pathways (Fig. 3C and 3D). Figure 3E shows the common GO terms related to SD. The nutrient metabolism-related terms, including “carbohydrate biosynthetic process,” “fatty acid metabolic process,” and “glycoprotein biosynthetic process,” and immune system-related terms, including “cytokine-mediated signaling pathway,” “leukocyte migration,” and “macrophage-derived foam cell differentiation,” were highly enriched in both the SCD and HFD conditions after SD (Fig. 3E).

Using DEGs among the four groups, we performed heatmap clustering analysis. Figure 3F shows a heatmap of GO terms based on the DEGs in each cluster. Cluster A, which was composed of DEGs in the EC + HFD group, included immune system-related GO terms such as “leukocyte proliferation,” “lymphocyte proliferation,” and “regulation of mononuclear cell proliferation.” Cluster D, which was composed of DEGs in the EC + SCD group, included DNA replication-related genes (Fig. 3F).

Neuroinflammatory changes in the brain after sleep deprivation and consumption of a high-fat diet

To identify neuroinflammatory changes associated with SD or an HFD in the brain, we employed a nanoString neuroinflammation panel ²⁶, which covers 770 genes related to neuroinflammation in the brain. Figure 4A shows volcano plots of the DEGs between each group. Under the SCD, cathepsin S (Ctss), endothelial cell adhesion molecule (Esam), and minichromosome maintenance complex component 6 (Mcm6) were the top three upregulated genes, and aspartate beta-hydroxylase (Asph), ribosomal protein S (Rps21), and BRCA-associated RING domain 1 (Bard1) were the top three downregulated genes after SD. The mcm6, C-C motif chemokine ligand 4 (CCl4), and interleukin 1 receptor kinase 3 (Irak3) genes were upregulated in the EC + HFD group compared to those in the EC + SCD group (Fig. 4A).

Using DEGs from the four groups, we compared the expression patterns in each group. In the SD + HFD group, Fc epsilon receptor 1 g (Fcer1g), growth arrest and DNA damage inducible alpha (Gadd45a), and RAS like proto-oncogene B (Ralb) gene expression was significantly increased compared to that in the other groups (Fig. 4B). Figure 4C shows the DEGs between each group and their roles in neuroinflammation. Under the SCD, five genes related to adaptive immune response, three genes related to microglial function, and three genes related to the cell cycle were differentially expressed after SD. Under the HFD, two genes related to cytokine signaling, two genes related to the innate immune response, and three genes related to microglial function were differentially expressed after SD. Adaptive immune response-related genes did not show significant differences between the SD + HFD and EC + HFD groups. Sialic acid-binding Ig-like lectin 1 (Siglec1), a marker for active neuroinflammation ²⁷, was highly expressed in the SD + HFD group compared to that in the SD + SCD group. Compared with that in the EC + SCD group, the SD + HFD group had the highest number of DEGs (Fig. 4A). These genes were related to inflammation, neuropathology, and microglial function (Fig. 4D).

Integration analysis of gut microbiome and host gene expression

To identify the main factors that mediate the microbiota-gut-brain interactions, we performed multi-omics factor analysis (MOFA) by integrating microbiome and gene expression data ^28,29. Figure 5A displays the four determinants discovered by the factor analysis. Among the four factors, factors 1 and 2 showed effective discriminating values (Fig. 5A and Supplementary Fig. 4). The variable with the largest weight in factors 1 and 2 was from the microbiome layer. Prevotella sp. MGM1 in factor 1 and Bacteroides satori in factor 2 showed the highest weight in this analysis (Fig. 5B and Supplementary Fig. 5).

To determine the main driver of the microbiota-gut-brain interactions, we performed factor analysis. MOFA2 revealed that the major feature of factors 1 and 2 was gut bacteria (Fig. 5 and Supplementary Fig. 5). This suggests that the main driver of microbiota-gut-brain interactions with SD and an HFD is the gut microbiome. Notably, the SRSF3 genes in the gut showed significant negative correlations with factors 1 and 2. According to recent studies, SRSF3 suppresses tumorigenesis ⁵⁴ and inhibits cellular senescence ⁵⁵. Thus, reduced SRSF3 expression might be an important contributor to gastrointestinal dysfunction caused by SD and an HFD.

Our study had several limitations. First, we performed an MOFA based on microbiome and transcriptome data. Further evidence, such as the blood metabolome and gut proteome, is required to substantiate our conclusions. Second, we only tested adult male mice. As a result, sex differences and aging were not reflected in our study.

In summary, our study revealed novel associations between the gut microbiota and host responses after SD and diet-induced obesity. Obesity with SD has deleterious effects on gut and brain health. We discovered that the gut microbiota may be the primary driver of microbiota-gut-brain interactions, and 5'-IMP may be an essential microbial metabolite that facilitates gut-brain communication. These findings imply that healing gut dysbiosis may be a viable therapeutic target for enhancing sleep quality and curing obesity-related dysfunction.

Here, we examined the effects of SD and diet-induced obesity on the gut microbiota, gut transcriptome, and brain gene expression. In addition, we integrated these data to reveal the main drivers of microbiota-gut-brain interactions. In the present study, we revealed the pleiotropic effects of SD and a HFD on the gut and brain. The HFD reduced gut microbiota biodiversity in terms of both alpha and beta diversity (Figure 1C and 1D). The gut transcriptome was primarily influenced by SD (Figure 3A and 3B), whereas brain gene expression associated with neuroinflammation was significantly altered following exposure to a HFD with SD (Figure 4A and 4D).

Gut microbiota analysis revealed that the HFD caused dramatic changes in the physiology of the gut microbiota (Figures 1 and 2). SD did not induce robust changes in the gut microbiota, as in other previous studies ^30,31. Our results showed that the HFD both alone and with SD increased Firmicutes (Table 2), which has already shown significant correlations with obesity and sleep quality ^32,33. Interestingly, the HFD alone increased the F/B ratio, whereas the HFD with SD decreased the F/B ratio by increasing Bacteroidetes (Table 2). Recent findings suggest that an increased F/B ratio leads to more effective glucose fermentation and higher nutritional absorption ³⁴. This means that the increased F/B ratio in the HFD mice may be a compensatory reaction to overeating, and sleep disrupts this compensation. Prevotella species have shown negative associations with a HFD in mice ³⁵ and significant relationships with weight change in a human randomized controlled trial ³⁶. In this study, Prevotella sp. MGM1 and MGM2 levels were decreased in both the EC+HFD and SD+HFD groups compared to those in the other groups (Figure 2B).

The abundance of Prevotella and Muribaculum was higher in the SD+SCD group than that in the EC+SCD group but did not increase in the SD+HFD and EC+HFP groups. Recently, Badran et al. ³⁷ reported that fecal microbiota transplantation using a fecal slurry, which has abundant levels of Prevotella and Muribaculaceae, improves sleep disturbances in mice. This suggests that the gut microbiota adjusts its composition to defend against SD stressors, whereas an HFD compromises these responses. To identify the possible mediators of microbiota-GBA interactions, we performed a microbial pathway analysis and found that the abundance of 5’-IMP synthesis-related species was highly increased in the SD+HFD group compared to that in the other groups (Figure 2D). 5’-IMP plays a key role in purine nucleotide synthesis and regulates various immune responses ³⁸. In addition, 5’-IMP is important for hypnotic action in the brain ³⁹. Therefore, increased 5’-IMP synthesis might be a protective response in the gut against metabolic stresses induced by SD and an HFD.

Gut transcriptome analysis revealed that SD caused robust changes in the gut transcriptome (Figure 3). In this study, the most significantly enriched pathways were related to ribosome biogenesis and nucleic acid metabolic processes (Figure 3C), which are critical for gut mucosal defense ⁴⁰ and colorectal cancer progression ⁴¹. This result elucidates the current association between sleep, obesity, and colon cancer. Both SD and obesity are high-risk factors for colorectal cancer development ^42,43. Recent studies have shown deleterious effects of SD on the gut ^44,45. For example, SD causes premature death by increasing reactive oxygen species in the gut ⁴⁴. We also found that SD enhanced the renin-angiotensin system (RAS)-associated pathways in the gut (Figure 3C). The gut RAS interacts bidirectionally with the gut microbiota and can promote intestinal inflammation and fibrosis ^46,47.

In the brain, genes related to neuroinflammation were altered by both SD and the HFD. The Mcm6 gene was highly upregulated after SD under the SCD and HFD conditions (Figure 4A). This gene encodes a protein that is a component of the MCM complex, which is required for the initiation of eukaryotic genome replication ⁴⁸. Furthermore, this gene has shown positive correlations with poor prognosis in brain and gastrointestinal tumors ^49,50. The most severe neuroinflammatory changes were observed in the SD+HFD group compared to those in the EC+SCD group. CXC motif chemokine ligand 10 (Cxcl10), insulin-like growth factor-1 (Igf1), and cluster of differentiation 70 (Cd70) were the top three highly elevated genes (Log2FC). These genes are well-known markers of proinflammatory signals ^51-53.

To determine the main driver of the microbiota-gut-brain interactions, we performed factor analysis. MOFA2 revealed that the major feature of factors 1 and 2 was gut bacteria (Figure 5 and Supplementary Figure 5). This suggests that the main driver of microbiota-gut-brain interactions with SD and an HFD is the gut microbiome. Notably, the SRSF3 genes in the gut showed significant negative correlations with factors 1 and 2. According to recent studies, SRSF3 suppresses tumorigenesis ⁵⁴ and inhibits cellular senescence ⁵⁵. Thus, reduced SRSF3 expression might be an important contributor to gastrointestinal dysfunction caused by SD and an HFD.

Funding: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2020R1C1C1004999), the Ministry of Health & Welfare (HU22C0150), Joint Research Project of Institutes of Science and Technology (2021, 2022), ‘GIST Research Institute IIBR’ grants funded by the GIST in 2022, the Bio & Medical Technology Development Program (2021M3A9G8022959), and an NRF grant (NRF-2021R1C1C1006336) from the Ministry of Science and ICT through the National Research Foundation.

Ethics approval: The handling and care of animals were approved by the Institutional Animal Care and Use Committee of the Gwangju Institute of Science and Technology (GIST-2021-064).

Data availability: The sequencing data is accessible via the Gene Expression Omnibus database (BioProject ID: PRNJA862187).

Competing interests: All the authors declare that they have no competing interests.

Authors' contributions:

S.L., C.-M.O., and T.K. contributed to the conceptual design of the project and the experiments described in the manuscript. The experiments were performed by J.L. and J.K. The data were analyzed by J.L., J.K., and Y.K. The manuscript was written by J.L., J.K., and C.-M.O. Then, the manuscript was edited and critically evaluated by S.L., C.-M.O., and T.K. All authors read and approved the final version of the manuscript.

Blüher, M. J. N. R. E. Obesity: global epidemiology and pathogenesis. 15, 288-298 (2019).
Beccuti, G. & Pannain, S. Sleep and obesity. Curr Opin Clin Nutr Metab Care 14, 402-412, doi:10.1097/MCO.0b013e3283479109 (2011).
Mullington, J. M., Haack, M., Toth, M., Serrador, J. M. & Meier-Ewert, H. K. Cardiovascular, inflammatory, and metabolic consequences of sleep deprivation. Prog Cardiovasc Dis 51, 294-302, doi:10.1016/j.pcad.2008.10.003 (2009).
Knutson, K. L., Spiegel, K., Penev, P. & Van Cauter, E. The metabolic consequences of sleep deprivation. Sleep Med Rev 11, 163-178, doi:10.1016/j.smrv.2007.01.002 (2007).
Spaeth, A. M., Dinges, D. F. & Goel, N. Effects of Experimental Sleep Restriction on Weight Gain, Caloric Intake, and Meal Timing in Healthy Adults. Sleep 36, 981-990, doi:10.5665/sleep.2792 (2013).
Spiegel, K., Leproult, R. & Van Cauter, E. Impact of sleep debt on metabolic and endocrine function. Lancet 354, 1435-1439, doi:10.1016/S0140-6736(99)01376-8 (1999).
Kinlen, D., Cody, D. & O’Shea, D. J. Q. A. I. J. o. M. Complications of obesity. 111, 437-443 (2018).
Stiemsma, L. T., Reynolds, L. A., Turvey, S. E., Finlay, B. B. J. I. & therapy. The hygiene hypothesis: current perspectives and future therapies. 4, 143 (2015).
Gwak, M.-G. & Chang, S.-Y. J. I. N. Gut-Brain connection: microbiome, gut barrier, and environmental sensors. 21 (2021).
Ma, Q. et al. Impact of microbiota on central nervous system and neurological diseases: the gut-brain axis. 16, 1-14 (2019).
Schütz, F., Figueiredo-Braga, M., Barata, P. & Cruz-Martins, N. J. P. B. J. Obesity and gut microbiome: review of potential role of probiotics. 6 (2021).
Scheithauer, T. P. et al. Gut microbiota as a trigger for metabolic inflammation in obesity and type 2 diabetes. 2546 (2020).
Irwin, M. R., Olmstead, R. & Carroll, J. E. Sleep Disturbance, Sleep Duration, and Inflammation: A Systematic Review and Meta-Analysis of Cohort Studies and Experimental Sleep Deprivation. Biol Psychiatry 80, 40-52, doi:10.1016/j.biopsych.2015.05.014 (2016).
Vaccaro, A. et al. Sleep Loss Can Cause Death through Accumulation of Reactive Oxygen Species in the Gut. Cell 181, 1307-1328 e1315, doi:10.1016/j.cell.2020.04.049 (2020).
Li, Y. et al. Gut Microbiota Changes and Their Relationship with Inflammation in Patients with Acute and Chronic Insomnia. Nat Sci Sleep 12, 895-905, doi:10.2147/NSS.S271927 (2020).
Liu, Z. et al. Acute Sleep-Wake Cycle Shift Results in Community Alteration of Human Gut Microbiome. mSphere 5, doi:10.1128/mSphere.00914-19 (2020).
Smith, R. P. et al. Gut microbiome diversity is associated with sleep physiology in humans. 14, e0222394 (2019).
Ogilvie, R. P. & Patel, S. R. J. S. h. The epidemiology of sleep and obesity. 3, 383-388 (2017).
Jean-Louis, G. et al. Associations between inadequate sleep and obesity in the US adult population: analysis of the national health interview survey (1977–2009). 14, 1-10 (2014).
Akinnusi, M. E., Saliba, R., Porhomayon, J. & El-Solh, A. A. J. E. J. o. I. M. Sleep disorders in morbid obesity. 23, 219-226 (2012).
Dispersyn, G. et al. The homeostatic and circadian sleep recovery responses after total sleep deprivation in mice. 26, 531-538 (2017).
Stojanov, S., Berlec, A. & Štrukelj, B. J. M. The influence of probiotics on the firmicutes/bacteroidetes ratio in the treatment of obesity and inflammatory bowel disease. 8, 1715 (2020).
Oliphant, K. & Allen-Vercoe, E. J. M. Macronutrient metabolism by the human gut microbiome: major fermentation by-products and their impact on host health. 7, 1-15 (2019).
Cryan, J. F. et al. The microbiota-gut-brain axis. (2019).
Banfi, D. et al. Impact of microbial metabolites on microbiota–gut–brain axis in inflammatory bowel disease. 22, 1623 (2021).
Das, M. et al. Interdependence of neural network dysfunction and microglial alterations in Alzheimer’s disease-related models. 24, 103245 (2021).
Ostendorf, L. et al. SIGLEC1 (CD169): a marker of active neuroinflammation in the brain but not in the blood of multiple sclerosis patients. 11, 1-8 (2021).
Argelaguet, R. et al. Multi‐Omics Factor Analysis—a framework for unsupervised integration of multi‐omics data sets. 14, e8124 (2018).
Argelaguet, R. et al. MOFA+: a statistical framework for comprehensive integration of multi-modal single-cell data. 21, 1-17 (2020).
Zhang, S. L. et al. Human and rat gut microbiome composition is maintained following sleep restriction. 114, E1564-E1571 (2017).
El Aidy, S., Bolsius, Y. G., Raven, F. & Havekes, R. J. J. o. s. r. A brief period of sleep deprivation leads to subtle changes in mouse gut microbiota. 29, e12920 (2020).
Daniel, H. et al. High-fat diet alters gut microbiota physiology in mice. 8, 295-308 (2014).
Triplett, J., Braddock, A., Roberts, E., Ellis, D. & Chan, V. J. S. S. Identification of sleep fragmentation-induced gut microbiota alteration and prediction of functional impact in Sprague Dawley rats harboring microbiome derived from multiple human donors. 15, 07 (2022).
Palmas, V. et al. Gut microbiota markers associated with obesity and overweight in Italian adults. 11, 1-14 (2021).
Qiao, F. et al. Alteration and the function of intestinal microbiota in high-fat-diet-or genetics-induced lipid accumulation. 12 (2021).
Christensen, L. et al. Prevotella abundance predicts weight loss success in healthy, overweight adults consuming a whole-grain diet ad libitum: a post hoc analysis of a 6-wk randomized controlled trial. 149, 2174-2181 (2019).
Badran, M., Khalyfa, A., Ericsson, A. & Gozal, D. J. E. n. Fecal microbiota transplantation from mice exposed to chronic intermittent hypoxia elicits sleep disturbances in naïve mice. 334, 113439 (2020).
Lovászi, M. et al. Inosine monophosphate and inosine differentially regulate endotoxemia and bacterial sepsis. 35, e21935 (2021).
Sánchez, C. L. et al. The possible role of human milk nucleotides as sleep inducers. 12, 2-8 (2009).
Grimble, G. J. G. Dietary nucleotides and gut mucosal defence. 35, S46-S51 (1994).
Nait Slimane, S. et al. Ribosome biogenesis alterations in colorectal cancer. 9, 2361 (2020).
Lin, C.-L., Liu, T.-C., Wang, Y.-N., Chung, C.-H. & Chien, W.-C. J. i. v. The association between sleep disorders and the risk of colorectal cancer in patients: A Population-based Nested Case–Control Study. 33, 573-579 (2019).
Bardou, M., Barkun, A. N. & Martel, M. J. G. Obesity and colorectal cancer. 62, 933-947 (2013).
Vaccaro, A. et al. Sleep loss can cause death through accumulation of reactive oxygen species in the gut. 181, 1307-1328. e1315 (2020).
Hao, G., Zhu, B., Wang, P., Li, L. & Hou, L. J. S. m. Sleep quality and disease activity in patients with inflammatory bowel disease: a systematic review and meta-analysis. 75, 301-308 (2020).
Garg, M. et al. Imbalance of the renin–angiotensin system may contribute to inflammation and fibrosis in IBD: a novel therapeutic target? 69, 841-851 (2020).
Jaworska, K., Koper, M., Ufnal, M. J. A. J. o. P.-G. & Physiology, L. Gut microbiota and renin-angiotensin system: A complex interplay at local and systemic levels. 321, G355-G366 (2021).
Bochman, M. L., Schwacha, A. J. M. & Reviews, M. B. The Mcm complex: unwinding the mechanism of a replicative helicase. 73, 652-683 (2009).
Shim, Y.-R., Kim, A., Gu, M.-J. J. I. J. o. C. & Pathology, E. Prognostic significance of MCM6 expression in gastrointestinal stromal tumor. 14, 1119 (2021).
Cai, H.-Q. et al. Overexpression of MCM6 predicts poor survival in patients with glioma. 78, 182-187 (2018).
Michlmayr, D., McKimmie, C. S. J. I. j. o. i., cytokine & research, m. Role of CXCL10 in central nervous system inflammation. 6, 1 (2014).
Wolters, T. L. C., Netea, M. G., Hermus, A. R. M. M., Smit, J. W. A. & Netea-Maier, R. T. J. J. o. m. e. IGF1 potentiates the pro-inflammatory response in human peripheral blood mononuclear cells via MAPK. 59, 129-139 (2017).
Dhaeze, T. et al. CD70 defines a subset of proinflammatory and CNS-pathogenic TH1/TH17 lymphocytes and is overexpressed in multiple sclerosis. 16, 652-665 (2019).
Xiong, J., Chen, Y., Wang, W. & Sun, J. J. O. L. Biological function and molecular mechanism of SRSF3 in cancer and beyond. 23, 1-11 (2022).
Tang, Y. et al. Downregulation of splicing factor SRSF3 induces p53β, an alternatively spliced isoform of p53 that promotes cellular senescence. 32, 2792-2798 (2013).

(Not answered)

SupplementaryFigureandtableEMM.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Integrated analysis of the microbiota-gut-brain axis in response to sleep deprivation in a diet-induced obesity mouse model

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Mouse experimental model

Sleep Deprivation

Tissue Harvest And Rna Precipitation

Colon Rna Sequencing

Shotgun Metagenomic Sequencing

Brain Nanostring Ncounter

Results

DISCUSSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1