Exploring Intervention of Selenium on Gut Microbiota Homeostasis and Corresponding Genetic Metabolism in Mice with Breast Cancer on a High-fat Diet Using Metagenomics Sequencing

doi:10.21203/rs.3.rs-2237461/v1

Download PDF

Research Article

Exploring Intervention of Selenium on Gut Microbiota Homeostasis and Corresponding Genetic Metabolism in Mice with Breast Cancer on a High-fat Diet Using Metagenomics Sequencing

https://doi.org/10.21203/rs.3.rs-2237461/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: The trace element selenium has an important effect on gut microbial homeostasis and the gut microbiota is closely related to breast cancer and its high-fat status. The effect of selenium on the gut microbiota of breast cancer in the high-fat state is unknown.

Materials and Methods: Twelve female BALB/c mice were randomly divided into two groups (4T1+selenium+fat diet group, 4T1+high fat diet group). To establish an obese mouse model, cultured 4T1 cells were transplanted on the right 4th mammary fat pad under anesthesia with 10⁵4T1 cells /50µl /mouse, and were given a high-fat diet for feeding. DNA was extracted from mouse fecal samples for meta-genomics sequencing and bioinformatics analysis. The relevant target genes and pathways were annotated and metabolically analyzed to explore the intervention effect of selenium on breast cancer in the high-fat state.

Results: Compared with the control group, the gut microbiota distribution and diversity were changed in breast cancer tumor bearing mice on a high-fat diet with selenium intervention. The gut microbial composition was significantly different in the selenium intervention group, with Proteobacteria, Actinobacteria, Verrucomicrobia phylum as well as the Helicobacter_ganmani，Helicobacter_japonicus and Akkermansia_muciniphila several species were increased and The phyla represented by Bacteroidetes, Firmicutes, Deferribacteres, Spirochaetes, as well as the Prevotella_sp_MGM2，Muribaculum_intestinale，Lactobacillus_murinus and Prevotella_sp_MGM1 several species of microbes were decreased. By functional analysis, a total of 21 significantly different functional genes associated with carbohydrate active enzymes were predicted, 455 significant genes of pathogen host interactions, 66 genes significantly associated with cell communication and cell auto-induction, 834 genes associated with membrane transporters, 220 genes related to virulence factors that were significantly different after selenium intervention. 37 cogs were predicted. 48 metabolites with rising metabolic potential in the selenium intervention group, such as L-alanine, SO2, and O2, among others.

Conclusions: Selenium can affect the homeostasis of gut microbiota by affecting the structure and abundance and associated metabolism of gut microbiota in mice with breast cancer on a high-fat diet. The mechanism may be through interfering with gut microbiota homeostasis, further affecting the synthesis of tumor associated proteins and fatty acids, and inducing tumor cell apoptosis and pyroptosis.

selenium

gut microbiota

genetic functions

breast cancer

high fat status

metagenomics sequencing

In 2020, female breast cancer (BC) overtook lung cancer as the most common cancer with 2,261,419 new cases (11.7%) and it is the fifth leading cause of cancer mortality worldwide, with 685,000 deaths (Sung et al., 2021). Obesity is a high-risk factor for the development and recurrence of breast cancer as well as metastasis(Picon-Ruiz et al., 2017), and it is closely related to breast cancer specific mortality, and the mechanism may be related to the secretion of hormones such as estrogen and insulin and the enzymatic activity such as aromatase (Engin, 2017). Changes in gut microbiota are often observed in patients with breast cancer. A case-control study investigating the association between fecal microbiota and BC in postmenopausal women showed that the diversity of fecal microbiota was changed in BC patients, especially Clostridiaceae, Faecalibacterium, Ruminococcaceae, Dorea and Lachnospiraceae (Goedert et al., 2015). Gut microbiota is closely related to the onset and prognosis of breast cancer, and can affect the occurrence, development, and metastasis of breast cancer through a variety of mechanisms. Drives epithelial cell transformation by affecting genomic stability, impeding apoptosis, and releasing cell proliferation signals. Gut microbiota also plays an important role in host oncogenic pathways versus chemoresistance. Class alterations in gut microbial populations affect hormone levels and may lead to higher estrogen levels, therefore it would increase breast cancer risk, and as some bacterial types in the gut increase and others decrease, the diversity of the gut microbiota is significantly affected and may be related to catabolism by the gut microbiota. This may affect estrogen release from the enterohepatic circulation, leading to increased systemic estrogen levels (Feng et al., 2021). Gut microbiota homeostasis restoration may become a novel therapy for breast cancer.

Selenium, an essential trace element, plays an important role in biological growth and development, has antioxidant, anti-inflammatory and immune functions, and is involved in the metabolism of thyroid hormones. Selenium and gut microbiota are very complex, on the one hand, it will affect the absorption of selenium. The study by Jovana Knezevic et. al found that gut microbiota may affect selenium absorption through the strong thyroid-gut axis, and similarly, selenium intake affects gut flora through the release of hormones (Knezevic et al., 2020). Studies have found that selenium intake can change the composition of the gut microbiota to some extent, and significantly affect the function and metabolism of the gut microbiota, for instance the proportion of Bacteroidetes and Proteobacteria was significantly different between the high- and low-selenium area (Zhang et al., 2021). At present, many experiments have shown that selenium can reduce the risk of cancer, such as breast, bowel, prostate and lung cancer, while modulating the immune system and many other effects (Hatfield et al., 2014; Vinceti et al., 2018). Kamil demircan et al investigated the association between pre diagnostic selenium intake and breast cancer prognosis by a prospective cohort study and showed that pre diagnostic selenium levels are strongly associated with low mortality and recurrence of invasive breast cancer (Demircan et al., 2021). Therefore, it is urgent to explore the changes of gut microbiota in high fat diet breast cancer bearing mice after intervention with organic selenium and reveal the role of the relationship between organic selenium and gut microbiota in breast cancer patients. In spite of this, the relationship between the gut microbiota and organic selenium in breast cancer mice has rarely been investigated using metagenomics sequencing analysis. In this study, to further investigate the relationship between organic selenium and gut microbiota composition, gene function and related metabolites in breast cancer patients predicted from the gut microbiota composition in high fat diet breast cancer bearing mice after organic selenium intervention, the profiles and composition of gut microbiota in fecal samples of high fat diet breast cancer bearing mice and controls under organic selenium intervention were compared using metagenomics sequencing analysis. Investigating the intervention effects of organic selenium on gut microbiota dysbiosis in breast cancer may contribute to our understanding of breast cancer etiology and pathomechanisms and facilitate targeted new prediction and treatment.

2.1 Animals Used in Experiments

The current investigation involved female Balb/c mice that were 8 weeks old (Beijing HFK Bioscience Co., Ltd., Beijing, China). In accordance with standard laboratory procedures, the animal models were established in Beijing Viewsolid Biotechnology Co., Ltd., and all animals were housed at the Beijing Hospital of Traditional Chinese Medicine (22-24 ℃, 40-60% relative humidity) food and water were freely available, and the light/dark-cycle was maintained for 12/12 hours, with the light turned on at 6:00 am. The selenium-enriched microalgal protein contained Se (VI): N. D., Se (IV):14.975, SeCs2:173.433, MeSeCys:14.468, SeMet:5.104, GSSeGS:18.249(μg/g), was provided by Enshi Zaoyuan Selenopeptide Biotechnology Co.Ltd.

2.2 Selenium Therapy in An Animal Model

Female Balb/c mice were divided into 2 groups (n=6 in each group) according to their body weight, as shown below: (I) a high-fat diet (45% kcal fat, 35% kcal carbohydrates, and 20% kcal proteins) was provided to the 4T1 + high-fat diet group (control group); (Ⅱ) 4T1 + selenium + fat diet group was fed with a high-fat diet and selenium-enriched microalgal protein 0.4mg/kg intragastric injection administration(IG) daily (selenium group). All mice were subcutaneously xenografted with 1 × 10⁵ 4T1 cells/50 µl each in the right fourth mammary fat pad under anesthesia. The observation period began with the injection of 4T1 cells for 4 weeks.

The selenium-enriched microalgal protein was dissolved in deionized water make out to a product with a stock solution that could be used to prepare treatment media at intragastric injection administration. The volume of the IG solution was performed at 10ml/kg according to mice body weight.

2.3 Sample Collection and DNA Extraction

Stool samples were collected in a specimen collection kit and stored at -20°C immediately after defecation and then at -80°C before further manipulation in the laboratory. DNA was extracted from samples using the stool DNA extraction Mini Kit. 0.2g fecal sample was added to Glass Beads Tube I, after which 600 µl Buffer STL and 20 µl Proteinase K were added. Samples were vortexed for 5min at maximum speed and further lysed by heating at 70 ° C for 15 min, after the samples had returned to room temperature, 200 µl Buffer IRP was added to the samples, which were vortexed thoroughly to mix. After the sample was centrifuged at 13,000g for 3 min on ice for 5 min, 450µl of supernatant was taken into a 2 ml centrifuge tube, after addition of 450µl Buffer MBL invert 3~4 times, vortex for 15s, 55˚C for 10min, during which mix several times by inversion, add 450µl absolute ethanol, vortex for 15s and centrifuge briefly. The extracted supernatant was purified following the instructions of DNA Extraction Mini Columns II to obtain sample DNA and stored at -20°C. DNA was fragmented to a fragment size of 200-300 BP for library construction, paired end sequencing was performed using PE150, and metagenomics was performed on the Illumina HiSeq platform following the manufacturer's instructions. Sequencing instrument model: novaseq6000.

2.4 Quality Control of Integrated Data

All raw metagenomics sequencing data were quality-controlled by MOCAT2 software(Kultima et al., 2016). All the original sequencing reads were disjointed by Cutadapt software(Kechin et al., 2017)and were trimmed by SolexaQA package with a quality of less than 20 and a length of less than 30 bp(Cox et al., 2010). Clean reads were obtained by quality control. We used SOAPaligner to compare the filtered reads with host reads whose genome was decontaminated to obtain high-quality clean data(Li et al., 2009).

2.5 Microbial Community Structure and Diversity Analysis

The CleanReads were used to be the input data and we used MetaPhlAn3(Segata et al., 2012)to make a Classification of species,thus the relative abundance of the sample bacteria from species level to phylum level was obtained.

2.5.1 Species Accumulation Curves

The Species Accumulation curve describes how species increase as the sample size increases condition, which can be used to determine the adequacy of sample size and estimate species richness. The species accumulation curve can not only judge whether the sample size is sufficient, but also predict the species richness if the sample size is sufficient. We used specaccum in R package Vegan to calculate the cumulative species curve Based on the species annotation results of MetaPhlAn3(L. Simpson, 2017). Alpha diversity includes richness, evenness and diversity Shannon index, which can be used to describe genes and functions in samples and the diversity of species. We used diversity in R package Vegan Function evaluates these indices based on the species annotation results of MetaPhlAn3. The clean reads obtained from quality control were assembled by De Novo using SOAPdenovo to obtain scaftigs with a length greater than 500 bp: Reads were interrupted into K-mer to construct de Bruijn graph and Eulerian Path was searched to assemble into contigs. Then scaffolds are connected according to the location relationship of pin-end reads, we select successive contigs from scaffolds to get scaftigs. Among them, k-mer of each sample was calculated by MOCAT2 according to the length and quantity of sample reads. Based on scaftigs obtained by assembly, we used MetaGeneMark for gene structure prediction, and CD-HIT was used for clustering to eliminate redundancy after gene set was obtained (Fu et al., 2012). If the sequence consistency of two genes is greater than 95%, and the overlap area covers more than 90% of the short sequences, then the two genes will be clustered into a cluster. The longest sequence in the cluster was selected as the representative sequence of the cluster to construct a non-redundant gene set with gene length greater than 100bp. Finally, we compared high-quality reads to the constructed non-redundant reference gene set by BWA, and the reads with length less than 30bp and consistency less than 95% were removed to obtain the reads count of each gene. We downsized count data by using the Rrarefy function in the vegan package of R language to obtain the corrected gene abundance.

2.5.2 Dimensionality Reduction Cluster Analysis

Principal component analysis (PCA), a simplified analysis of data that effectively finds the most "dominant" elements and structures in the data, is based on the original species composition matrix derived from MetaPhlAn3 using the RDA function of R language and the eigenvalue decomposition method. The differences of multiple sets of data are reflected on the two-dimensional principal component diagram. The more similar the sample composition is, the closer the distance is in the PCA diagram. Principal co-ordinates analysis (PCoA) is a visualization method to study data similarity or difference, which is similar to PCA analysis. The difference is that PCoA uses the Bray-Curtis distance. We used the PCoA function of R language to sort the eigenvalues and eigenvectors, and selected the eigenvalues ranked at the top to represent the distance difference between samples. Permutational multivariate analysis of variane (PERMANOVA) was used to assess whether there were significant community differences among different groups. Among them, PERMANOVA was calculated by adonis2 function in the Vegan package, and the P value was obtained by 1000 permutations. Nonmetric Multidimensional Scale analysis (NMDS) is a data analysis method which can simplify the study objects (samples or variables) in multidimensional space to a low-dimensional space for positioning, analysis and classification, while preserving the original relationships between objects. Based on Bray-Curtis distance, we used metaMDS function of R language for NMDS analysis. In addition, PERMANOVA was used to evaluate whether there were significant community differences among different groups.

2.5.3 Enterotypes Analysis

Enterotypes are clusters divided according to the composition of microbial communities, which is a very effective method to distinguish human intestinal microbes. JSD distance was calculated according to the genus level abundance, and CH index was used to calculate the optimal cluster K value after clustering by PAM method. Finally, Between Class analysis (BCA) was used for visualization.

2.5.4 Analysis of Differences between Group

Multiple Response Permutation Procedure, often combined with dimension reduction analysis such as PCoA and NMDS, is a statistical method used to test whether the differences between groups (two or more groups) are significantly greater than the differences within groups. Based on the relative species abundance data of MetaPhlAn3, we used the MRPP function of R package Vegan to compute A semi-metric matrix such as Bray-Curtis to evaluate the differences between and within groups, and finally performed significance analysis of the groups using the substitution test to obtain the P value. Based on the species composition results obtained from metaphlan3, we scored the units that were significantly different by comparing the relative abundances between the two groups using the Wilcoxon rank-sum permutation test, set the P value at 0.05, and corrected the P value using the Benjamini-Hochberg method. Comprehensive antibiotic resistance database (CARD), the core of which is ARO, the antibiotic resistance ontology, which contains 4,498 terms related to antibiotic resistance genes, resistance mechanisms, antibiotics, and targets. In addition, 2,984 reference sequences, 1,337 SNPs, 3,030 AMR detection models, and 2685 relevant references (Alcock et al., 2020) were included in the database. We used the CARD (v3.0.8) official software RGI (resistance gene identifier, v4.0.2) to annotate nonredundant gene sets for resistance genes. Evolutionary gene of genes: non conserved orthologous groups, the specialized annotation databases of orthologous groups (Egg-NOG), is a public data resource collecting information on orthologous relationships, gene evolutionary history and functional annotations, which also includes information on functional classification such as COG/KOG. The version5.0 incorporates 5,090 organisms, including 4,445 representative bacterial species, 168 archaeal species, 477 eukaryotic species, and also the whole genome protein sequences of 2502 viruses. Among them, the species belonged to 379 different taxonomic levels, and the corresponding genes were classified into 4.4M orthologous genes (orthologous group, OG) (Huerta-Cepas et al., 2019). The data were generated by aligning non redundant reference gene sets to the Egg-NOG database using EggNOG-mapper V2. Clusters of orthologous groups is a genome scale database for protein functional and evolutionary analysis, we aligned nonredundant reference gene sets to the COG database using Egg-NOG mapper V2 produce the data. Carbohydrate-Active enzymes database (CAZy) is a database specifically designed to describe enzymes of the degradable, modified or synthetic glycosidic bond class, which contains information on the species origin of carbohydrate enzyme classes, enzyme function EC classification, gene sequences, protein sequences and their structures. Annotation of carbohydrate active enzymes was performed using EggNOG-mapper V2 by aligning non redundant reference gene sets to the CAZy. MvirDB integrates the sequence information of DNA and protein in Tox-Prot, SCORPION, the PRINTS virus factors, VFDB, TVFac, Islander, ARGO and VIDA (part), and can provide rapid information retrieval of protein toxins, virulence factors, antibiotic resistance and other related genes for biological defense and medical fields. However, the database had been abandoned, and we obtained the 2015 updated sequence files as well as annotation information for the site using wayback machine. Annotation of biodefense related genes was performed by aligning non redundant reference gene sets to the MvirDB database using DIAMOND (v0.7.9.58, parameters: blastp-v-sensitive-k 10) (Buchfink et al., 2015). In this study, after obtaining the alignment results, if there was more than one alignment result for each sequence, one optimal alignment result shall be retained as the gene annotation result. Alignment parameters set expect values E-value as 1e-5. The Pathogen host interaction database (PHI), primarily collects interaction information from fungal, oomycetal, and bacterial pathogens, as well as from infected hosts (including animals, plants, fungi, insects, etc.) that has been experimentally validated. Version 4.8 contained information on 6,780 genes, 13,801 pathogen host interaction relationships, 268 pathogens, 210 hosts, etc(Urban et al., 2017). Annotation information such as pathogen host interactions was obtained by aligning non redundant reference gene sets to the PHI (v4.8) database using DIAMOND. Quorum sensing is a type of signal communication mechanism between microorganisms, also known as cell communication or autoinduction, by secreting, releasing some specific signaling molecules, sensing concentration changes, detecting the density of bacteria, regulating the physiological functions of bacteria, and thus adapting to environmental conditions. We obtained a QS database of 23,225 associated genes by a variety of means such as searching in the Uniprot database using QS related keywords (Barriuso and Martinez, 2018). Annotation of quorum sensing associated genes was performed by aligning non redundant reference gene sets to QS databases using DIAMOND. Transporter classification database (TCDB) details a classification system for membrane transporters known as the transporter classification (TC) system, which is similar to the Enzyme Commission (EC) system, in addition to integrating functional as well as phylogenetic information. These contain a total of 1474 descriptions of transporter families, TC numbers, as well as gene information, and the entire system is divided into 5 tiers(Saier et al., 2016), such as classes, subclasses, families, subfamilies, and transport systems, each corresponding to a letter in TC#. Taxonomic annotation of transporters was performed by aligning nonredundant reference gene sets to the TCDB database (update: 2015) using DIAMOND. Virulence factor database (VFDB)(Chen et al., 2016) is an integrated and integrated web resource database for managing virulence factors. Among them, there were 74 bacterial pathogens (32 with complete annotation information), 1081 virulence factors (576 experimental validation), 951 bacterial species (529 complete genomes), 32522 VF related non redundant genes (3228 experimental validation), and 2612 related literatures. Non redundant reference gene set ratios were individually aligned to the VFDB (update: 2020.4) database using DIAMOND.

2.5.5 DeSeq2 Analysis

The amount of sequencing will be different in different samples, and conventional normalization methods, RPKM, FPKM, etc., are CPM (count per million) normalization according to the length of genes or transcripts, but their disadvantages are easy to be affected by extremely highly expressed and differentially expressed genes in different samples. DeSeq2 was developed to address this concern by using a scale factor that accounts for differences in sequencing depth. First, based on clean reads data, we used Kraken2 (2.0.8-beta) for bacterial species annotation to obtain species count abundances, followed by the DeSeq2 function of the R language for differential species analysis, and the Benjamini-Hochberg method was used to correct P values.

2.5.6 LEfSe Analysis

LEfSe analysis enables comparison between two or more groups to find biomarkers with significant differences in abundance between groups. Based on MetaPhlAn3 based species relative abundance data, LEfSe was used to first detect species with significantly different abundances between different groupings using the nonparametric Kruskal Wallis rank-sum test for two or more groups, and then significant differential species obtained in the previous step were analyzed using the Wilcoxon rank-sum test for groups, Finally, linear discriminant analysis (LDA) was used to dimensionally reduce the data and assess the influence of the significantly different species (i.e., LDAscore).

2.6 Multiple Linear Association Analysis

Due to the presence of other factors affecting the microbiota, such as age, gender, BMI, etc., we analyzed the relationships between these factors and the microbiota as well as the functions of the microbiota using multiple linear association models (MaAsLin). This method employs a generalized linear model, while MaAsLin screens for potential, microbiota related phenotypic variables with a boosting approach due to the sparsity of metagene data. We performed multivariate association analysis with the results of MetaPhlAn3.

2.7 KEGG Pathway Enrichment Analysis

After aligning clean reads to their customized database with HUMAnN and obtaining the relative abundances at the level of metabolic pathways in the gene families from the UniProt Reference Cluster and the MetaCyc database, the resulting data were analyzed using the humann_regroup_table merges annotation results, resulting in KEGG KO annotation results. For relative abundance of KEGG orthology (KO), KEGG pathway enrichment analysis was finally performed with the GAGE package from R/Bioconductor.

2.8 Metabolic Potential Analysis

The microbiota is able to generate and consume many metabolites, and for each sample and metabolite, we used the PRMT (predicted relative metabolic turnover) method to calculate the CMPs (community metabolic potential) after predicting the resulting relative abundances of genes and to estimate the differences in metabolic potential between the different groups (Noecker et al., 2016).

3.1 Microbial Diversity

The species accumulation curves of each group leveled off, indicating that the sample biodiversity was adequately covered by the applied sequencing depth (Figure 1). We used Shannon, Simpson, inversesimpson, richness as well as evenness indexes to represent alpha diversity of the samples, and we can see that alpha diversity was lower in the selenium intervention group than in the control group, in which the Simpson index showed significant difference (Figure 2). The results of gene richness analysis showed that selenium intervention group was lower than the control group, but there was no significant difference between the two groups (Figure 3). In terms of dimensionality reduction cluster analysis, we performed PCA, PCoA and NMDS analysis, which revealed significant differences in the composition of gut microbiota between the two groups (Figure 4). In addition, we tested whether the difference between the groups was significantly greater than that within the group by multiple response permutation procedure, and the results showed that the difference between the groups was smaller and significant than that between the groups, indicating that the grouping was homogeneous and that the effects of interventions were obvious (Table 1).

Table 1: Results of MRPP analysis. The smaller the Observed-delta value, the smaller the within group difference, and the larger the Expect-delta value illustrates the larger the between group difference. A-score greater than 0 indicates that the difference between groups is greater than the difference within groups, and A-score less than 0 indicates that the difference between groups is greater than the difference between groups.

	Group	Distance	A_score	Observe_delta	Expect_delta	P_value	Sig
1	Control_VS_Selenium	Bray-Curtis	0.254	0.254	0.581	0.005	**

3.2 Gut Microbial Alterations

According to the different relative abundances in the selenium intervention and control groups, seven dominant phyla were identified. The phyla Bacteroidetes, Firmicutes, Deferribacteres, Spirochaetes were decreased and Proteobacteria, Actinobacteria, Verrucomicrobia were increased in the selenium intervention group compared to the control group (Figure 5A, B). At the species level, Helicobacter_ganmani，Helicobacter_japonicus and Akkermansia_muciniphila were increased and Prevotella_sp_MGM2，Muribaculum_intestinale，Lactobacillus_murinus and Prevotella_sp_MGM1 were decreased in the selenium intervention group compared with the control group (Figure 5C, D). In addition, we performed enterotype analysis for the two groups of samples, and the best cluster K value was calculated by Calinski-Harabasz (CH) index (Figure6A) and then visualized by between-class analysis, which showed no significant difference in enterotype between the two groups (Figure 6B). Wilcoxon rank-sum permutation test was performed to further compare the significant differences in the composition of gut microbiota between the two groups. The top 20 items with the most significant differences are shown in table (Table 2).

Table 2: Taxonomic differences between the selenium intervention group and control group. (top20) (W indicates the statistical value of the Wilcoxon rank sum test. The P.adj indicates the P values corrected by multiple testing).

	taxonomy	Control	Selenium	W	pvalue	p.adj
1	Species Parabacteroides_goldsteinii	0	0.282	0	0.006	0.090
2	Species Bacteroides_stercorirosoris	0.004	0.630	0	0.009	0.090
3	Phylum Bacteroidetes	49.814	8.821	24	0.010	0.090
4	Class Bacteroidia	49.814	8.821	24	0.010	0.090
5	Order Bacteroidales	49.814	8.821	24	0.010	0.090
6	Family Muribaculaceae	17.558	1.533	24	0.010	0.090
7	Genus Muribaculum	14.845	0.920	24	0.010	0.090
8	Species Muribaculum_intestinale	14.845	0.920	24	0.010	0.090
9	Family Prevotellaceae	29.019	0.060	24	0.010	0.090
10	Genus Prevotella	29.019	0.060	24	0.010	0.090
11	Phylum Proteobacteria	25.797	70.937	0	0.010	0.090
12	Class Epsilonproteobacteria	25.717	70.732	0	0.010	0.090
13	Order Campylobacterales	25.717	70.732	0	0.010	0.090
14	Family Helicobacteraceae	25.717	70.732	0	0.010	0.090
15	Genus Helicobacter	25.717	70.732	0	0.010	0.090
16	Species Helicobacter_ganmani	20.157	61.864	0	0.010	0.090
17	Species Parabacteroides_gordonii	0.002	0.189	0	0.011	0.101
18	Species Prevotella_sp_MGM2	23.214	0.049	24	0.014	0.117
19	Family Tannerellaceae	0.145	1.115	1	0.023	0.177
20	Genus Parabacteroides	0.145	1.115	1	0.023	0.177

The selenium intervention group showed a significant increase in the species parabacteroides goldsteinii, Bacteroides stercorirosoris, Helicobacter ganmani and parabacteroides gordonii and a decrease in Muribaculum-stercorirosoris, Prevotella-sp-MGM2, compared to the control group. Increases in the genus Muribaculum and Prevotella, decreases in the genus Helicobacter and Parabacteroides were observed in the selenium intervention group. In selenium intervention group, the families helicobacteraceae, tannerellaceae were more abundant, whereas muribaculaceae, prevotellaceae were less abundant. The results of DeSeq2 differential analysis by count abundance of species in the two groups showed no significant differences between the groups at the gene level for various species (Table 3).

Table 3: Species count data DeSeq2 comparison results between groups (top 5). The results showed no significant differences between the two groups at the gene level across species.

	ID	Control	Selenium	baseMean	log2FoldChange	lfcSE	stat	pvalue	padj
1	Species Abelson murine leukemia virus	1.167	1.750	1.430	0.531	0.900	0.590	0.555
2	Species Abisko virus	1	1	1.030	-0.056	1.062	-0.053	0.958
3	Species Acanthocystis turfacea chlorella	1.167	1	1.130	-0.277	1.033	-0.268	0.789
4	Species Acaryochloris marina	1	1	1.030	-0.056	1.062	-0.053	0.958
5	Species Acetoanaerobium sticklandii	4.500	6	5.119	0.386	0.698	0.554	0.580	0.943

LEfSe was used to further determine the specific significantly different bacterial taxa between the two cohorts (Figure 7A). Several species, Helicobacter-ganmani, Akkermansia-muciniphila, Bacteroides-faecichinchillae, Bacteroides-faecichinchillae, Lactobacillus-reuteri, Parabacteroides-goldsteinii and Bacteroides-stercorirosoris, were significantly enriched in the selenium intervention group. The identified taxa were highlighted on a cladogram to indicate significant differences in phylogenetic distribution as well as their LDA scores (Figure 7B). These results indicated that there was significant gut microbiota alteration between the two groups. MaAsLin results similarly showed that the species parabacteroides-gordonii, parabacteroides-goldsteinii, Bacteroides-stercorirosoris, and Lactobacillus-reuteri were increased in selenium intervention group compared to control group, while lachnospiraceae-bacterium-28-4 was decreased (Table 4).

Table 4: The results of multivariate linear correlation analysis of gut microbiota indicated that there was significant gut microbiota alteration between the two groups(top5).

	Variable	Feature	Value	Coefficient	N	N.not.0	P-value	Q-value
1	group	Species Parabacteroides_gordonii	Selenium	0.413	10	6	0.0001	0.003
2	group	Species Parabacteroides_goldsteinii	Selenium	0.551	10	4	0.0001	0.003
3	group	Species Bacteroides_stercorirosoris	Selenium	0.285	10	5	0.0001	0.003
4	group	Species Lachnospiraceae_bacterium_28_4	Selenium	-0.365	10	5	0.0099	0.332
5	group	Species Lactobacillus_reuteri	Selenium	0.442	10	6	0.0135	0.361

3.3 Gene Function and Metabolic Potential Analysis

By CARD annotation we found that 24 antibiotic resistance associated genes were significantly different between the two groups (Table 5).

Table 5: Results of intergroup comparisons of card annotated outcomes. (top20)

	feature	Control	Selenium	W	pvalue	p.adj
1	gimA family macrolide glycosyltransferase	0	11.031	0	0.006	0.207
2	AAC (3)	23.353	5.797	24.000	0.010	0.207
3	DHA beta-lactamase	28.560	1.248	24.000	0.010	0.207
4	SPG beta-lacatamase	11.169	0.315	24.000	0.010	0.207
5	TLA beta-lactamase	13.349	0.753	24.000	0.010	0.207
6	subclass B3 PEDO beta-lactamase	16.583	0.610	24.000	0.010	0.207
7	van ligase	7.580	17.618	0	0.010	0.207
8	CepS beta-lactamase	2.778	0.123	24.000	0.014	0.265
9	tetracycline inactivation enzyme	2.476	14.092	1.000	0.019	0.279
10	vanR	7.227	1.599	23.000	0.019	0.279
11	vanS	6.847	0.521	23.000	0.025	0.279
12	aminocoumarin resistant parY	0	0.061	3.000	0.026	0.279
13	aminocoumarin self resistant parY	0	0.061	3.000	0.026	0.279
14	OCH beta-lactamase	5.100	0	22.000	0.031	0.279
15	subclass B1 Vibrio cholerae varG beta-lactamase	0.421	3.317	2.000	0.031	0.279
16	AAC (6')	8.843	2.099	22.000	0.038	0.279
17	resistance-nodulation-cell division (RND) antibiotic	7.198	1.217	22.000	0.038	0.279
18	OKP beta-lactamase	2.378	14.226	2.000	0.038	0.279
19	ole glycosyltransferase	8.766	27.710	2.000	0.038	0.279
20	aminocoumarin self resistant parY	0.461	2.500	2.000	0.040	0.279

The results of the gene functional annotation of Egg NOG revealed 1,674 genes with significant differences between the two groups (Table S1).

While a total of 37 functional COGs were significantly different between the selenium intervention group and the control group (Table S2).

A total of 21 significantly different functional genes associated with carbohydrate active enzymes between two groups were predicted (Table S3).

The two groups of genes were annotated by MvirDB, leading to the prediction of selenium effects on biological defense. (Table 6) PHI results revealed a total of 455 significant pathogen host interactions (Table S4).

A total of 66 genes significantly associated with cell communication and cell auto-induction by QS analysis (Table S5).

The TCDB results revealed a total of 834 genes associated with membrane transporters (Table S6). VFDB analysis identified a total of 220 genes related to virulence factors that were significantly different after selenium intervention (Table S7).

We found that several pathways related to RNA anabolism were enriched after selenium intervention by KEGG pathway enrichment analysis, unfortunately none of these pathways were significantly different (Table 7).

Table 7: Based on the differential results from Ko, enrichment analysis was used to obtain the KEGG pathways that were significantly different among the different groups (top 10). All pathway differences were not significant.

	pathway	p.geomean	stat.mean	p.val	q.val	set.size	Selenium
1	ko03018.RNA.degradation	0.062	1.575	0.062	0.672	22	up
2	ko00970. Aminoacyl.tRNA.biosynthesis	0.081	1.424	0.081	0.672	31	up
3	ko00195.Photosynthesis	0.093	1.346	0.093	0.672	22	up
4	ko00061. Fatty.acid.biosynthesis	0.094	-1.349	0.094	0.854	17	down
5	ko00240. Pyrimidine.metabolism	0.095	1.318	0.095	0.672	73	up
6	ko03030.DNA.replication	0.098	-1.316	0.098	0.854	26	down
7	ko00670.One.carbon.pool.by.folate	0.116	1.231	0.116	0.672	17	up
8	ko03020.RNA.polymerase	0.117	1.228	0.117	0.672	11	up
9	ko00190. Oxidative.phosphorylation	0.124	1.163	0.124	0.672	72	up
10	ko00770.Pantothenate.and.CoA.biosynthesis	0.127	1.163	0.127	0.672	22	up

Based on the KEGG annotation results, the CMPs were calculated for the metabolites of the two groups of samples, and the differences in metabolic potential between the different groups were estimated, which showed that 48 metabolites including L-alanine, SO2, O2, d-biotin, L-asparagine, cephalin and pyridoxine phosphate were significantly different (Table 8).

Table 8: CMPs were calculated using the PRMT method, and analysis of differences between groups revealed several significant differences in metabolic potential(top10).

		Metabolite	Control	Selenium	W	pvalue	p.adj
1	C00041	L-Alanine	0	215,078.68	0	0.006	0.151
2	C09306	Sulfur dioxide	0	215,078.68	0	0.006	0.151
3	C00007	Oxygen	115,274.72	601,031.50	0	0.010	0.151
4	C00120	Biotin	314,575.83	637,720.75	0	0.010	0.151
5	C00152	L-Asparagine	-119,335.17	1,066,719.55	0	0.010	0.151
6	C00350	Phosphatidylethanolamine	10,649.08	541,085.00	0	0.010	0.151
7	C00627	Pyridoxine phosphate	215,444.17	625,641.00	0	0.010	0.151
8	C02972	Dihydrolipoylprotein	125,261.67	693,816.00	0	0.010	0.151
9	C04631	UDP-N-acetyl-3-(1-carboxyvinyl)-D-	-64,879.00	688,963.75	0	0.010	0.151
10	C04851	MurAc-diphospho-undecaprenol	-18,425.33	726,519.50	0	0.010	0.151

Breast cancer is the most common malignancy in women. Recent relevant studies have also illustrated that high-fat status is a high-risk factor for breast cancer development, meanwhile, the content of cholesterol, low-density lipoprotein and so on in the human body are regulated by dietary patterns, and the gut microbiota plays an important role in metabolic processes. Studies linking the gut microbiota to high-fat status breast cancer are increasing, with studies showing that alterations in breast cancer gut microbiota in humans and mice in response to obesity and high-fat dietary intake are similar (Soto-Pantoja et al., 2021). Metagenomics sequencing and 16S rRNA sequencing are widely used in the analysis of the defined composition of microbial communities, and can both be used to study the species composition of a community, the evolutionary relationships among species, and the diversity of the community, but many of the sequences obtained from 16S sequencing are poorly annotated at the species level, and metagenomics sequencing on the basis of 16S sequencing also allows for indepth studies at the genetic and functional levels, such as the GO, KEGG pathway, and so on, At the same time, microbes can be identified to the species level(New and Brito, 2020). In this study, we tried to deeply explore the mechanism of selenium on breast cancer and its high-fat status by examining the metabolic function of gut microbiota and related target genes KEGG, COG pathways in breast cancer tumor bearing mice on high-fat diet after intervention with selenium.

Our results showed that the diversity of the gut microbiota in the high-fat status breast cancer bearing mice after selenium intervention was significantly different under the premise that the sample biodiversity was sufficiently covered at the sequencing depth, and the distribution of the gut microbiota was also altered. The phyla represented by Bacteroidetes, Firmicutes, Deferribacteres, Spirochaetes, as well as the Prevotella_sp_MGM2，Muribaculum_intestinale，Lactobacillus_murinus and Prevotella_sp_MGM1 several species of microbes were decreased in the gut microbiota of the selenium intervention group, whereas the Proteobacteria, Actinobacteria, Verrucomicrobia phylum as well as the Helicobacter_ganmani，Helicobacter_japonicus and Akkermansia_muciniphila several species were increased. Same results were shown in the lefse analysis. The analysis found no significant difference in enterotype and no significant difference in gene richness between the two groups, which revealed that selenium intervention did not significantly affect the gut microbiota genes as well as enterotype in breast cancer bearing mice on a high-fat diet. Studies have shown that breast cancer tumor growth can be inhibited by regulating the homeostasis of gut microbiota, producing short chain fatty acids (SCFAs), and then interfering with the expression of tumor associated proteins (Han et al., 2021). This was also confirmed by the elevated Firmicutes/Bacteroidetes (F/B) ratio in our study, selenium can alter the production of medium and short chain fatty acids, such as acetate and butyrate, which are essential for the maintenance of colonic mucosal integrity, by regulating Bacteroides and Prevotella abundance(Christensen et al., 2018). Similar to the conclusion of Xuepeng Chi et al.(Chi et al., 2022), the relative abundance of Lachnospiraceae and Prevotellaceae increased and the abundance of Helicobacterceae decreased after increasing the content of selenium in the ingested food, and the correlation analysis suggested that selenium could optimize the functional network of gut microbiota and optimize the interaction of gut microbiota and host. It has been reported that Faecalibacterium prausnitzii supernatant can inhibit the growth of breast cancer cells by inhibiting the IL-6/STAT3 pathway(Ma et al., 2020), suggesting that this class of bacteria may contribute to breast cancer prevention, and that reduction of this class of bacteria may promote breast cancer progression, and our previous study found that decreasing STAT3 phosphorylation can effectively induce apoptosis and pyroptosis in breast cancer cells exposed to high-fat conditions(Liu et al., 2022b). There was a trend toward enrichment in several KEGG pathways related to RNA anabolism in the selenium intervention group, although not significant, and notably, there was a trend toward decreased Fatty-acid-biosynthesis related to fatty acid anabolism, perhaps suggesting a possible role for selenium in the biosynthesis of fatty acids in breast cancer bearing mice fed a high-fat diet, selenium can inhibit fat accumulation and has anti-inflammatory activity which has been confirmed by the study of Liu et al(Liu et al., 2022a). Our results annotated by CAZy also confirm that selenium can interfere with carbohydrate active enzymes in breast cancer bearing mice fed a high-fat diet, which is produced by intervening the gut microbiota to limit lipogenesis. Selenium also exerts an effect on antibiotic resistance, and the Se NP-ε-PL prepared by Tao Huang et al. (Huang et al., 2020) showed good antibacterial activity against eight different bacteria, including partially drug-resistant strains. Based on the results of CARD annotation, we found that selenium can modulate antibiotic resistance in high-fat status breast cancer tumor bearing mice by intervening in the gut microbiota. Not only that, Egg Nog analysis results showed that selenium can have an effect on pathogen biological attack factors, cell autoinduction and membrane transporters by intervening gut microbiota. Prediction of the metabolic potential of the samples from the two groups revealed that selenium intervention may improve the metabolic capacity of breast cancer bearing mice on a high-fat diet to L-Alanine, Sulfur dioxide, Oxygen, Biotin and L-Asparagine et al., in which biotin plays an important role in biochemical pathways such as fat synthesis, gluconeogenesis, and the combination of biotin and prebiotic supplementation may help prevent the deterioration of metabolic status in severely obese patients(Belda et al., 2022).

Potential limitations of the present study are as follows: (1) The gut epithelium and fecal supernatant were helpful for further studies, but related samples were not collected. (2) The sample marker gene database used when analyzing the gut microbiota, genetic functions, and related metabolites in mice using metagenomics sequencing may include most species- and strain level information rather than 100% of all existing species or organisms. (3) Metabolomics was not used when performing microbial metabolite analysis, but rather was analyzed in terms of DNA and gut microbiota abundance. The actual microbial metabolite profile may differ from it during DNA transcription and translation. (4) The experimental model mice and the 4T1 breast cancer tumor cell line used for research have shed new light on breast cancer research, but this does not represent the actual situation of the gut microbiota in all high-fat state breast cancer patients in the clinic.

The results of the present study suggest that gut microbial composition and associated gene functions may affect fatty acid synthesis, regulate the synthesis of tumor associated proteins, and induce tumor cell pyroptosis. Selenium can affect the homeostasis of gut microbiota by affecting the structure and abundance of gut microbiota, as well as the related metabolism, in mice with breast cancer on a high-fat diet. Intervention of gut microbiota by selenium this approach may become a new prospect in the treatment of high-fat status breast cancer. The results of this study are validated and contribute to the development of new predictive and therapeutic approaches for breast cancer in the high-fat state in the future.

ETHICS STATEMENT

All experimental processes were carried out in compliance with the Guidelines for the Care and Use of Laboratory Animals, which were developed by the Chinese Ministry of Science and Technology (Beijing, China). Experiments were performed under a project license granted by ethics board of Beijing Viewsolid Biotechnology Co., Ltd. (No. 202000038).

CONFLICT OF INTEREST

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

AUTHOR CONTRIBUTIONS

Y Li, G Zhang and Q Zhang designed the study. Y Li, B Kong performed the experiments. M Liu collected and prepared samples for sequencing. Y Li and B Kong performed the statistical analysis and interpretation with the assistance of M Liu, and all the authors participated in the writing and review of the manuscript.

FUNDING

This study was funded by Enshi Prefecture Science and Technology Program Research and Development Project (No. JCY2019000040). The funders had no stakes in the design, collection of data and analysis, decisionmaking of publishing, or the writing of the manuscript.

ACKNOWLEDGMENTS

We gratefully acknowledge the assistance provided by the Central Laboratory of Beijing Traditional Chinese medicine hospital, Capital Medical University.

SUPPLEMENTARY MATERIAL

Please see the Supplementary Material section of the Author guidelines for details on the different file types accepted.

DATA AVAILABILITY STATEMENT

The datasets presented in this study can be found in online repositories. The names of the repository/repositories andaccession number(s) can be found below: https://www.ncbi.nlm.nih.gov/sra/PRJNA857801.

Alcock, B.P., Raphenya, A.R., Lau, T.T.Y., Tsang, K.K., Bouchard, M., Edalatmand, A., et al. (2020). CARD 2020: antibiotic resistome surveillance with the comprehensive antibiotic resistance database. Nucleic Acids Res 48(D1), D517-D525. doi: 10.1093/nar/gkz935.
Barriuso, J., and Martinez, M.J. (2018). In Silico Analysis of the Quorum Sensing Metagenome in Environmental Biofilm Samples. Front Microbiol 9, 1243. doi: 10.3389/fmicb.2018.01243.
Belda, E., Voland, L., Tremaroli, V., Falony, G., Adriouch, S., Assmann, K.E., et al. (2022). Impairment of gut microbial biotin metabolism and host biotin status in severe obesity: effect of biotin and prebiotic supplementation on improved metabolism. Gut. doi: 10.1136/gutjnl-2021-325753.
Buchfink, B., Xie, C., and Huson, D.H. (2015). Fast and sensitive protein alignment using DIAMOND. Nat Methods 12(1), 59-60. doi: 10.1038/nmeth.3176.
Chen, L., Zheng, D., Liu, B., Yang, J., and Jin, Q. (2016). VFDB 2016: hierarchical and refined dataset for big data analysis--10 years on. Nucleic Acids Res 44(D1), D694-697. doi: 10.1093/nar/gkv1239.
Chi, X., Liu, Z., Wang, H., Wang, Y., Xu, B., and Wei, W. (2022). Regulation of a New Type of Selenium-Rich Royal Jelly on Gut Microbiota Profile in Mice. Biol Trace Elem Res 200(4), 1763-1775. doi: 10.1007/s12011-021-02800-4.
Christensen, L., Roager, H.M., Astrup, A., and Hjorth, M.F. (2018). Microbial enterotypes in personalized nutrition and obesity management. Am J Clin Nutr 108(4), 645-651. doi: 10.1093/ajcn/nqy175.
Cox, M.P., Peterson, D.A., and Biggs, P.J. (2010). SolexaQA: At-a-glance quality assessment of Illumina second-generation sequencing data. BMC Bioinformatics 11, 485. doi: 10.1186/1471-2105-11-485.
Demircan, K., Bengtsson, Y., Sun, Q., Brange, A., Vallon-Christersson, J., Rijntjes, E., et al. (2021). Serum selenium, selenoprotein P and glutathione peroxidase 3 as predictors of mortality and recurrence following breast cancer diagnosis: A multicentre cohort study. Redox Biol 47, 102145. doi: 10.1016/j.redox.2021.102145.
Engin, A. (2017). Obesity-associated Breast Cancer: Analysis of risk factors. Adv Exp Med Biol 960, 571-606. doi: 10.1007/978-3-319-48382-5_25.
Feng, Z.P., Xin, H.Y., Zhang, Z.W., Liu, C.G., Yang, Z., You, H., et al. (2021). Gut microbiota homeostasis restoration may become a novel therapy for breast cancer. Invest New Drugs 39(3), 871-878. doi: 10.1007/s10637-021-01063-z.
Fu, L., Niu, B., Zhu, Z., Wu, S., and Li, W. (2012). CD-HIT: accelerated for clustering the next-generation sequencing data. Bioinformatics 28(23), 3150-3152. doi: 10.1093/bioinformatics/bts565.
Goedert, J.J., Jones, G., Hua, X., Xu, X., Yu, G., Flores, R., et al. (2015). Investigation of the association between the fecal microbiota and breast cancer in postmenopausal women: a population-based case-control pilot study. J Natl Cancer Inst 107(8). doi: 10.1093/jnci/djv147.
Han, B., Jiang, P., Jiang, L., Li, X., and Ye, X. (2021). Three phytosterols from sweet potato inhibit MCF7-xenograft-tumor growth through modulating gut microbiota homeostasis and SCFAs secretion. Food Res Int 141, 110147. doi: 10.1016/j.foodres.2021.110147.
Hatfield, D.L., Tsuji, P.A., Carlson, B.A., and Gladyshev, V.N. (2014). Selenium and selenocysteine: roles in cancer, health, and development. Trends Biochem Sci 39(3), 112-120. doi: 10.1016/j.tibs.2013.12.007.
Huang, T., Holden, J.A., Reynolds, E.C., Heath, D.E., O'Brien-Simpson, N.M., and O'Connor, A.J. (2020). Multifunctional Antimicrobial Polypeptide-Selenium Nanoparticles Combat Drug-Resistant Bacteria. ACS Appl Mater Interfaces 12(50), 55696-55709. doi: 10.1021/acsami.0c17550.
Huerta-Cepas, J., Szklarczyk, D., Heller, D., Hernandez-Plaza, A., Forslund, S.K., Cook, H., et al. (2019). eggNOG 5.0: a hierarchical, functionally and phylogenetically annotated orthology resource based on 5090 organisms and 2502 viruses. Nucleic Acids Res 47(D1), D309-D314. doi: 10.1093/nar/gky1085.
Kechin, A., Boyarskikh, U., Kel, A., and Filipenko, M. (2017). cutPrimers: A New Tool for Accurate Cutting of Primers from Reads of Targeted Next Generation Sequencing. Journal of Computational Biology 24(11), 1138-1143. doi: 10.1089/cmb.2017.0096.
Knezevic, J., Starchl, C., Tmava Berisha, A., and Amrein, K. (2020). Thyroid-Gut-Axis: How Does the Microbiota Influence Thyroid Function? Nutrients 12(6). doi: 10.3390/nu12061769.
Kultima, J.R., Coelho, L.P., Forslund, K., Huerta-Cepas, J., Li, S.S., Driessen, M., et al. (2016). MOCAT2: a metagenomic assembly, annotation and profiling framework. Bioinformatics 32(16), 2520-2523. doi: 10.1093/bioinformatics/btw183.
L. Simpson, P.S., M. H. H. Stevens, E. Szoecs, J. O. Helene Wagner, F. G. Blanchet,M. Friendly, R. Kindt, P. Legendre, D. McGlinn, P. R. Minchin, R. B. O’Hara, and Gavin.vegan (2017). Community ecology package.
Li, R., Yu, C., Li, Y., Lam, T.W., Yiu, S.M., Kristiansen, K., et al. (2009). SOAP2: an improved ultrafast tool for short read alignment. Bioinformatics 25(15), 1966-1967. doi: 10.1093/bioinformatics/btp336.
Liu, G., Li, J., Pang, B., Li, Y., Xu, F., Liao, N., et al. (2022a). Potential role of selenium in alleviating obesity-related iron dyshomeostasis. Crit Rev Food Sci Nutr, 1-15. doi: 10.1080/10408398.2022.2074961.
Liu, M., Li, Y., Kong, B., Zhang, G., and Zhang, Q. (2022b). Polydatin down-regulates the phosphorylation level of STAT3 and induces pyroptosis in triple-negative breast cancer mice with a high-fat diet. Ann Transl Med 10(4), 173. doi: 10.21037/atm-22-73.
Ma, J., Sun, L., Liu, Y., Ren, H., Shen, Y., Bi, F., et al. (2020). Alter between gut bacteria and blood metabolites and the anti-tumor effects of Faecalibacterium prausnitzii in breast cancer. BMC Microbiol 20(1), 82. doi: 10.1186/s12866-020-01739-1.
New, F.N., and Brito, I.L. (2020). What Is Metagenomics Teaching Us, and What Is Missed? Annu Rev Microbiol 74, 117-135. doi: 10.1146/annurev-micro-012520-072314.
Noecker, C., Eng, A., Srinivasan, S., Theriot, C.M., Young, V.B., Jansson, J.K., et al. (2016). Metabolic Model-Based Integration of Microbiome Taxonomic and Metabolomic Profiles Elucidates Mechanistic Links between Ecological and Metabolic Variation. mSystems 1(1). doi: 10.1128/mSystems.00013-15.
Picon-Ruiz, M., Morata-Tarifa, C., Valle-Goffin, J.J., Friedman, E.R., and Slingerland, J.M. (2017). Obesity and adverse breast cancer risk and outcome: Mechanistic insights and strategies for intervention. CA Cancer J Clin 67(5), 378-397. doi: 10.3322/caac.21405.
Saier, M.H., Jr., Reddy, V.S., Tsu, B.V., Ahmed, M.S., Li, C., and Moreno-Hagelsieb, G. (2016). The Transporter Classification Database (TCDB): recent advances. Nucleic Acids Res 44(D1), D372-379. doi: 10.1093/nar/gkv1103.
Segata, N., Waldron, L., Ballarini, A., Narasimhan, V., Jousson, O., and Huttenhower, C. (2012). Metagenomic microbial community profiling using unique clade-specific marker genes. Nat Methods 9(8), 811-814. doi: 10.1038/nmeth.2066.
Soto-Pantoja, D.R., Gaber, M., Arnone, A.A., Bronson, S.M., Cruz-Diaz, N., Wilson, A.S., et al. (2021). Diet Alters Entero-Mammary Signaling to Regulate the Breast Microbiome and Tumorigenesis. Cancer Res 81(14), 3890-3904. doi: 10.1158/0008-5472.CAN-20-2983.
Sung, H., Ferlay, J., Siegel, R.L., Laversanne, M., Soerjomataram, I., Jemal, A., et al. (2021). Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71(3), 209-249. doi: 10.3322/caac.21660.
Urban, M., Cuzick, A., Rutherford, K., Irvine, A., Pedro, H., Pant, R., et al. (2017). PHI-base: a new interface and further additions for the multi-species pathogen-host interactions database. Nucleic Acids Res 45(D1), D604-D610. doi: 10.1093/nar/gkw1089.
Vinceti, M., Filippini, T., Del Giovane, C., Dennert, G., Zwahlen, M., Brinkman, M., et al. (2018). Selenium for preventing cancer. Cochrane Database Syst Rev 1, CD005195. doi: 10.1002/14651858.CD005195.pub4.
Zhang, Z.X., Xiang, H., Sun, G.G., Yang, Y.H., Chen, C., and Li, T. (2021). Effect of dietary selenium intake on gut microbiota in older population in Enshi region. Genes Environ 43(1), 56. doi: 10.1186/s41021-021-00220-3.

Table 6 is not available with this version

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Exploring Intervention of Selenium on Gut Microbiota Homeostasis and Corresponding Genetic Metabolism in Mice with Breast Cancer on a High-fat Diet Using Metagenomics Sequencing

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1 Animals Used in Experiments

2.2 Selenium Therapy in An Animal Model

2.3 Sample Collection and DNA Extraction

2.4 Quality Control of Integrated Data

2.5 Microbial Community Structure and Diversity Analysis

2.5.1 Species Accumulation Curves

2.5.2 Dimensionality Reduction Cluster Analysis

2.5.3 Enterotypes Analysis

2.5.4 Analysis of Differences between Group

2.5.5 DeSeq2 Analysis

2.5.6 LEfSe Analysis

2.6 Multiple Linear Association Analysis

2.7 KEGG Pathway Enrichment Analysis

2.8 Metabolic Potential Analysis

3. Results

3.1 Microbial Diversity

3.2 Gut Microbial Alterations

3.3 Gene Function and Metabolic Potential Analysis

4. Discussion

5. Conclusion

Declarations

References

Table

Additional Declarations

Status:

Version 1