Portulaca oleracea polysaccharides reduce serum lipid levels in aging rats by modulating intestinal microbiota and metabolites

Metabolic diseases characterized by dyslipidemia are common health problems for elderly populations. Dietary fiber intake is inversely associated with the risk of dyslipidemia. This study investigated the effects of Portulaca oleracea polysaccharide (POP) on the intestinal microbiota and its metabolites in aging rats using 16S rRNA sequencing and metabolomics techniques. Our results showed that POPs reduced the ratio of Firmicutes/Bacteroidetes (F/B), relative abundance of Fusobacteria, and levels of triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), alanine aminotransferase (ALT), and gamma-glutamyl transferase (γ-GT) in the serum of aging rats. POP supplementation also reduced 5beta-cholestane-3alpha,7alpha,12alpha,25-tetrol, and vaccenic acid concentrations in lipids and lipoid-like molecules, while soyasapogenol E and monoacylglycerol (MG) (24:0/0:0/0:0) levels increased. This study demonstrated that POP’s beneficial effects on lipid levels in aging rats might be partially attributable to the modification of gut microbiota and related metabolites.


Introduction
Age is one of the main physiological factors that increases serum lipid levels, which leads to increased levels of triglyceride (TG), serum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and decrease of high-density lipoprotein cholesterol (HDL-C); all of these are dyslipidemia characteristics (1,2). TC and LDL-C levels increased with normal aging, while HDL-C decreased with it. Elevated plasma LDL-C levels are one of the most important risk factors for atherosclerotic cardiovascular disease (ASCVD) (1,3,4). By 2050, China will have approximately 400 million citizens aged >65 years, including 150 million aged >80 years. It is apparent that China faces a huge healthcare challenge owing to its aging population (5, 6).
Dysregulation of the intestinal microbiota and its metabolites relates to the development of metabolic diseases, such as hyperlipidemia, diabetes, hypertension, and obesity (7,8). A study on elderly patients with coronary heart disease showed an association between increased blood lipid levels and abnormal intestinal microbiota distribution, which may influence patients' serum lipid levels (9). Animal experiments have also revealed a significant decline in the relative abundance of Bilophila, bile salt hydrolase-rich Bifidobacterium, and Lactobacillus. Meanwhile, decreases in deoxycholic acid and lithocholic acid are potential lipid-lowering mechanisms for improving lipid disorders (10).
Diet is one of the most important factors affecting gut microbiome composition, particularly indigestible food components (11,12). Dietary fiber has anti-digestion and anti-absorption properties and is mainly fermented by the intestinal microbiota, which affects its composition and function (13,14). In one study, significant differences in microbial function among adults in the United States, Malawi, and Venezuela were attributed to protein (carnivorous) and dietary fiber consumption (herbivorous), respectively (12). In Western societies, low fiber intake is a driving factor in the reduction of human gastrointestinal microbiota, increasing the risk of chronic non-communicable diseases, such as obesity, cardiovascular disease, and type 2 diabetes (14, 15). Thus, the dietary reference intake recommends a daily allowance of 38 g/day of dietary fiber in men and 25 g/day in women aged 19-50 years. Men and women aged >51 years should consume 31 g/day and 21 g/day of dietary fiber, respectively (16). The regulation of the production of gut microbiota and bacteriaderived metabolites is attributed to the beneficial effects of dietary fiber on dyslipidemia (17,18). Bioactive polysaccharides also improve dyslipidemia and reduce the risk of cardiovascular disease (16). Previous studies have found that polysaccharides from Portulaca oleracea (POPs) help regulate lipid metabolism in diabetic mice (19). However, its effect on lipid levels in elderly individuals is unclear. This study aimed to investigate the structural changes in the intestinal microbiota, map the bacteriaderived metabolites, and understand the effect of POPs on lipid levels in elderly populations.

Equipment and reagents
The equipment and reagents used in this study included: High Resolution Mass Spectrometer (QE Plus, Thermo

Animals and experiment design
Thirty 18-month-old SPF-grade SD rats were purchased from Hunan SJA Laboratory Animal Co. Ltd. [Laboratory animal production license, SCXK(Xiang)2019-0004]. All animals were acclimatized at 23 ± 1 • C, 50-60% relative humidity, and 12/12 h light/dark cycle for 1 week before the animal experiments commence. The rats (n = 10 per group) were divided into the Con (normal rat pellet feed), POP (1.0% POPs added to 99.0% pellet feed), and herb (3.5% P. oleracea powder combined with 96.5% pellet feed) groups. The rats ate and drank freely. Food intake and fasting blood glucose were recorded on the last day of each week. At the end of the 15-week experimental period, the rats were sacrificed by heart puncture after anesthesia with urethane, and biological samples were collected for further analysis. The Animal Care and Use Committee of

Metabolomic analysis of the gut microbiota
A day before the end of the experiment, the animals were held in aseptic conditions, and their anuses were squeezed to stimulate defecation. The fecal samples were collected with aseptic cryopreservation tubes and stored at −80 • C after liquid nitrogen quick-freezing.
Approximately 60 mg of intestinal content was added to 20 µL of L-2-chlorophenylalanine (0.3 mg/mL, methanol configuration) and 600 µL methanol-water (v:v = 4:1). Then, they were ground for 2 min, ultrasonically extracted for 10 min in an ice-water bath, and centrifuged for 10 min at 13,000 rpm at 4 • C. A volume of 300 µL of supernatant air dried and then added to 400 µL of methanol-aqueous solution for redissolving, followed by vortex and ultrasonication for 3 min. It was then centrifuged at 13,000 rpm for 10 min at 4 • C, and filtered (0.22 µm). UPLC-Orbitrap/MS analysis was performed later.
The acquired LC-MS raw data were analyzed using Progenesis QI software (Waters Corporation, Milford, MA, United States). Metabolites were identified using Progenesis QI (Waters Corporation, Milford, MA, United States) data processing software, based on public databases, such as http://www.hmdb.ca/, http://www.lipidmaps.org/, and self-built databases. The positive and negative data were combined to and imported into the R ropls package. Principle component analysis (PCA) and (orthogonal) partial least-squares-discriminant analysis (O) PLS-DA were carried out to visualize the metabolic alterations among the experimental groups, after mean centering (Ctr) and Pareto variance (Par) scaling, respectively. Hotelling's T2 region, shown as an ellipse in the models' score plots, defines the 95% confidence interval (CI) of the modeled variation. Variable importance in the projection (VIP) ranks the overall contribution of each variable to the orthogonal partial least squares discriminant analysis (OPLS-DA) model, and those variables with VIP > 1 are relevant for group discrimination.
In this study, the default 7-round cross validation was applied, with 1/7 of the samples being excluded from the mathematical model in each round to guard against overfitting.
The differential metabolites were selected on the basis of a combination of statistically significant thresholds of variable influence on projection (VIP) values obtained from the OPLS-DA model and P-values from a two-tailed Student's t-test on the normalized peak areas, where metabolites with VIP > 1.0 and P < 0.05 were considered differential metabolites.

Gut microbiota analysis
Total genomic DNA was extracted using a DNeasy PowerSoil Kit (QIAGEN, Germantown, MD, United States) following the manufacturer's instructions. The DNA concentration was verified using NanoDrop and agarose gel. The V3-V4 variable regions of the 16S rRNA genes were amplified using universal primers 343F (5 -TACGGRAGGCAGCAG-3 ) and 798R (5 -AGGGTATCTAATCCT-3 ) and then sequenced using the Illumina Miseq platform from Shanghai OE Biotech Co., Ltd. (Shanghai, China). Raw sequencing data were in FASTQ format. Paired-end reads were preprocessed using Trimmomatic software to detect and remove ambiguous bases (N). It also removed low-quality sequences those with average quality scores below 20 using the sliding window trimming approach. After trimming, the paired-end reads were assembled using FLASH software. The assembly parameters were as follows: 10 bp of minimal overlapping, 200 bp of maximum overlapping, and 20% of maximum mismatch rate. Further denoising of the sequences was performed as follows: reads with ambiguous homologous sequences and those below 200 bp were abandoned. Reads with 75% of the bases above Q20 were retained. Then, chimeric reads were detected and removed. These two steps were achieved using QIIME software (version 1.8.0). Clean reads were subjected to primer sequence removal and clustering to generate operational taxonomic units (OTUs) using VSEARCH software with a 97% similarity cutoff. Each OUT's representative read was selected using the QIIME package. All representative reads were annotated and blasted against the Silva database version 123 (or Greengenes) (16s/18s rDNA) using an RDP classifier (the confidence threshold was 70%). All representative reads were annotated and blasted against the UNITE database (ITSs rDNA) using BLAST.

Statistical analysis
The statistical analysis was performed using SPSS 19.0 (SPSS Inc., Chicago, IL, United States). One-way analysis of variance (ANOVA) for multiple comparisons, followed by the LSD t-test, was used to determine the significance of differences between groups. A difference of P < 0.05 was significant. The data were expressed as (x ± s), and the profiles were plotted in GraphPad Prism, version 8.0 (GraphPad Software, Inc., La Jolla, CA, United States).

Relative feed intake and blood glucose changes
As age and digestive function decline, food intake amounts in aged individuals are affected. Additionally, abnormal glucose metabolism-related diseases are common problems in these people. Our results in aging rats showed that a POPsupplemented diet could improve the dietary intake in the aging rats, and the relative feed intake from week 5 to week 11 significantly increased compared with the control (Con) group; However, the purslane-supplemented diet (Herb) had no similar effect ( Figure 1A). The results showed that POPs and the purslane could reduce blood glucose in aging rats, and the effect of the purslane was better than that of POP. Blood glucose levels FIGURE 1 Relative feed intake (A) and fasting plasma glucose level (B) measurements after supplementation with POP and purslane. The different letters indicate significant differences (P < 0.05).
Frontiers in Nutrition 04 frontiersin.org Effects of POP administration on (A) body weight, (B) organ index, (C) enzyme activity of liver function, and (D) serum lipid levels. The symbols * and ** indicate statistically significant differences among the groups (P < 0.05 and P < 0.01, respectively). H&E staining of the colon. Regular morphology and clearer and more intact structures of the colons of rats were observed in the POP and herb groups than in the Con group.
Frontiers in Nutrition 05 frontiersin.org in the herb group were significantly reduced at weeks 6, 8, and 13 compared with that of the Con group ( Figure 1B).

Body weight, organ index, blood lipid levels, and liver function index changes
The body weight, organ index, blood lipid level (TG, TC, HDL-C, and LDL-C), and liver function results are shown in Figure 2. Compared with the Con group, POPs and P. oleracea showed a trend in preventing weight loss in aging rats, but no statistical difference was observed between the groups (Figure 2A). The purslane polysaccharide diet significantly reduced the cardiac organ index levels (TG, LDL-C, ALT, and γ-GT) in aging rats, while the purslane herb diet only significantly reduced TG and ALT levels (Figures 2B-D). Figure 3 shows regular morphology and clearer and more intact structures of the colons of rats in the POP and herb groups than in the Con group, and POP supplementation promoted intestinal villi development.

Changes in the microbiota structure of fecal samples
In all samples, the coverage rate of goods was >98.3%, indicating that the sequencing depth of the fecal microbiota analysis met the experimental requirements. In the alpha diversity analysis, the Chao1 and observed indices reflected the degree of bacterial abundance, whereas Shannon and Simpson indices were used to evaluate the microbial community's diversity. As shown in Table 1, POP and P. oleracea herb supplementation slightly increased the abundance of intestinal microbiota in aging rats but did not have an effect on diversity. According to the species annotation results, the dominant bacteria in each group at the phylum level were Bacteroidetes and Firmicutes. Other phyla, including Spirochaetes, Proteobacteria, Actinobacteria, and Tenericutes, showed a low relative abundance ( Figure 4A). Compared to the Con group, POP and purslane-supplementation  significantly reduced the relative abundance of Fusobacteria and Firmicutes/Bacteroidetes (F/B) ratio ( Figure 4B). Principal component analysis showed that the clustering effect of samples in the herb and POP groups was better, while there were large differences between samples in the Con group. There was a significant difference between the treatment and Con groups ( Figure 4C).

Analysis of differences in species composition
LEfSe analysis was used to identify differences in species composition in each group's fecal microbiota. The results showed that the relative abundance of Coprococcus_2 was significantly enriched after POP supplementation, and the relative abundance of the Rikenellaceae_RC9_gut_group, Rikenellaceae, Parabacteroides, Tannerellaceae, Fusicatenibacter, and Ruminococcaceae-UCG-005 was significantly increased after purslane supplementation. However, in the Con group, the relative abundance of uncultured bacteria, Muribaculaceae, Lactobacillus, Lactobacillaceae, Lactobacillales, Lactovum, Bacilli Lachnospiraceae-NK4A136, and Ruminococcus-2 groups were significantly increased (Figure 5A,B).   Table 1. After data processing of all groups by QI, 6,293 and 8,119 ions were detected in the negative and positive modes, respectively. The score scatter plots for the OPLS-DA models showed clear differentiations among the POP, herb, and Con groups (Figures 6A,B). The OPLS-DA results indicated that the R2Y and Q2 values in the POP and Con groups were 0.994 and 0.905, respectively, while the R2Y and Q2 values in the herb and Con group were 0.992 and 0.906, respectively. This indicates that the existing model's prediction was stable and accurate. Furthermore, sevenfold cross validation and response permutation testing were used to investigate the quality of the model (Figures 6C,D). Notably, the Q2 values in both analyses were negative (−0.431 and −0.395), indicating a low risk of overfitting. Based on the VIP-plot analysis results, the ions with VIP > 1 and P < 0.05 were selected as differential metabolites, among which, there were 1,631 differences (P < 0.05) and 747 differences in lipids and lipoid-like molecule classifications in POP vs. Con analysis. There were 1,671 differential metabolites in Herb vs. Con (P < 0.05) analysis, and 684 differences in the lipid and lipoid-like molecule classifications. After adjusting the screening conditions to a score >50, VIP > 1.0, and P < 0.05, POP and purslane supplementation significantly regulated 33 lipids and function-like metabolites related to aging in rats, of which nine were up-regulated and 24 were downregulated ( Table 2).

Changes in fecal metabolites and corresponding pathways
To understand metabolic pathway changes in different samples, we conducted metabolic pathway enrichment analyses for differential metabolites based on the KEGG database. The results showed that the main changes between the POP and Con groups included linoleic acid (LA) metabolism, arachidonic acid metabolism, primary bile acid biosynthesis, steroid hormone biosynthesis, terpenoid backbone biosynthesis, serotonergic synapse, sphingolipid metabolism, and the peroxisome proliferator-activated receptor (PPAR) signaling pathways ( Figure 7A). Changes between the herb and Con groups included arachidonic acid metabolism, LA metabolism, steroid biosynthesis, the PPAR signaling pathway, primary bile acid biosynthesis, serotonergic synapse, steroid hormone biosynthesis, the Fc epsilon RI signaling pathway,    Changes in the corresponding pathways associated with fecal metabolites. (A) Major changes between the POP and Con groups (P < 0.05).
(B) Major changes between the herb and Con groups (P < 0.05).

Discussion
Aging is associated with multiple systemic dysfunctions, lipid metabolism disorders, and chronic inflammatory states, leading to ASCVD (1). Dyslipidemia is one of the most important triggers of atherosclerosis in the arterial wall and ASCVD, which is characterized by abnormal elevated levels and deposition of apolipoprotein B, including LDL-C (20). Previous studies have shown that POP has biological properties, such as anti-tumor, anti-oxidation, anti-diabetes, anti-viral, and intestinal microbiota regulation (21-24). Therefore, to investigate the mechanism by which purslane polysaccharides FIGURE 8 Correlation between serum lipids, fecal metabolites, and gut microbiota in Spearman's correlation coefficient. * and ** indicate statistically significant differences (P < 0.05 and P<0.01, respectively).
improve serum lipid levels in aging rats, we studied the intestinal metabolite profile and changes in the microbial community structure.
As expected, POPs improved dietary volume and weight loss in the aging rats. High levels of plasma glucose, TG, TC, and LDL-C reflect the state of lipid absorption and metabolism, which contribute the risk of various diseases (25). However, POPs can regulate plasma glucose, TG, TC, and LDL-C levels and reduce these potential risks. Gong et al. also found that POPs significantly reduced fasting blood glucose, TC, and TG levels, and increased HDL-C levels, body weight, and serum insulin levels in diabetic mice (26). The liver regulates lipid homeostasis and participates in glucose synthesis and metabolism (27). As important indicators of liver function, ALT and γ-GT are related to the lipid profile of patients with non-alcoholic fatty liver disease (28). Interestingly, we found that POPs significantly reduced ALT and γ-GT levels aging rats. Zheng et al. confirmed that purslane markedly decreases liver injury effects by reducing serum glucose, anti-aspartate aminotransferase, ALT, TG, and TC levels in diabetic mice (29). Our results suggest that purslane may reduce fasting blood glucose, TG, and ALT levels and alleviate the effects of weight loss in aging rats. These results suggest that POPs are one of the main active components of purslane that lowers blood lipid levels and protects the liver. Additionally, our study showed that POPs could alleviate the effects of cardiac hypertrophy in elderly rats. However, whether purslane polysaccharides can improve blood lipid levels and protect the heart requires further investigation. Therefore, POPs may be a promising factor in the prevention of dyslipidemia.
Food and intestinal microbiota interact to processes dietary nutrients into metabolites. Diet also affects intestinal microbiota composition (30). Due to the close anatomical and functional interaction of the gut-liver axis, the microbiota-gut-liverbrain axis has drawn great attention in recent years, and a growing number of studies have shown that gut microbiota is linked with many diseases, such as depression, autism, anxiety, obesity, schizophrenia, diabetes, Parkinson's disease (PD) and Alzheimer's disease (31). The gut microbiota composition in the elderly individuals is characterized by decreases in diversity and the abundance of butyrate-producing species (32). Compared with the Con group, the Chao1 and observed indices of the POP group increased by 3.32 and 2.93% in the αdiversity analysis, respectively, suggesting that POPs potentially improve the abundance of intestinal microbiota in aging rats. Moreover, β-diversity showed significant differences within the Con group, while POP and herb group cohesion was good, suggesting that P. oleracea or its polysaccharide-active substances, affects the composition of intestinal microbiota in elderly rats. At the phylum level, a higher F/B ratio is positively correlated with energy acquisition, which may be a risk factor for metabolic syndrome (33). In this study, a significant change in the gut microbiota was observed, and the F/B ratio in the Con group was significantly higher than that in the POP and herb groups, suggesting that POPs and purslane may reduce energy acquisition in aging rats and help prevent metabolic syndrome. Fusobacteria, at the phylum level, is an intestinal microbiota associated with colorectal cancer (34,35). Here, POP and purslane supplementation significantly reduced the relative abundance of Fusobacteria in aging rats. Studies in diabetic populations have shown that Fusobacteria are negatively correlated with higher dietary fiber intake (36). Notably, this risk was also reduced by POP and purslane feeding.
It is widely believed that Lactobacillaceae, at the family level, and Lactobacillus, at the genus level, are beneficial in improving human health (37). Recent studies have reported higher levels of Lactobacillaceae in obese mice and patients (38). Moreover, a high abundance of lactic acid bacteria was positively correlated with obesity-related characteristics (39). A recent study suggests that some species of lactic acid bacteria, such as Lactobacillus reuteri, are associated with obesity (40). In the present study, the relative abundance of Lactobacillaceae and Lactobacillus were significantly higher in the Con group than in the other two groups. Furthermore, the levels of Lactovum, Bacilli, Lachnospiraceae-NK4A136, and Ruminococcus-2 in Firmicutes, and Muribaculaceae in Bacteroidetes phylum in the Con group were significantly higher than those in the POP and herb groups. High-fat/carbohydrate diets program the gut microbiota to be dominated by Firmicutes (Clostridium), Prevotella, and Methanobrevibacter (41). Muribaculaceae were the dominant genus in the intestinal microbiota of mice fed a high-fat diet (42). As expected, P. oleracea and POPs inhibited the proliferation of these bacteria, which may help prevent dyslipidemia. Coprococcus spp. is fungus capable of fermenting complex plant carbohydrates in the Lachnospiraceae family (43). However, Coprococcus abundance is low in patients with PD) and children with autism spectrum disorders (44, 45). In this study, the relative abundance of Coprococcus-2 was significantly higher in the intestines of aging rats. Prior studies have found that POPs significantly improve the γ-aminobutyric acid level in juvenile rat serum (data not shown). These results suggest that POPs may be beneficial because they metabolize complex plant carbohydrates in the gut. Mushroom-enriched diets modulate the microbial-gut-brain axis in mice and reduce oxidative stress in the brain (46). Increased Tannerellaceae may positively influence the composition of the intestinal microbiota. A study of normal-weight subjects who consumed 40 g/day of resistant starch suggested that the relative abundance of Ruminococcaceae-UCG-005 was significantly increased (47). High amounts of Rikenellaceae are protective against cardiovascular and metabolic diseases related to visceral fat and, is thus a potential marker of healthy aging, and possibly, longevity (48). However, the results of bacterial community changes from PD patients' fecal samples showed that the abundance of the Rikenellaceae-RC9-gut group and Parabacteroides was relatively high (49). Frailty is a critical aging-related syndrome, but the decreased relative abundance of Fusicatenibacter was found in frailty (50). In this study, supplementation with P. oleracea increased the relative abundance of Parabacteroides, Tannerellaceae, and Fusicatenibacter, and Rikenellaceae-RC9-gut-group abundance in Ruminococcaceae-UCG-005 and Rikenellaceae, suggesting that POPs were more specific in fostering intestinal microbiota in aging rats than those fostered by P. oleracea, which may be related to the complex composition of P. oleracea.
We used the LC-MS/MS metabonomic method to analyze metabolite changes in the feces of aging rats fed POPs and P. oleracea. In the rats, lipids and lipoid-like molecules, including steroids and steroid derivatives, prenol lipids, glycerophospholipids, glycerolipids, and fatty acids, were significantly altered. Supplementation with POPs and purslane improved lipid metabolism in aging rats by regulating related pathways. Our results showed that POPs and purslane regulate related metabolic pathways in aging rats, with the PPAR signaling pathway being the most affected, followed by LA metabolism. Purslane is an excellent source of ω-3 fatty acids, including α-linolenic acid and LA, which are essential for normal growth, health promotion, and disease prevention in humans (51). The metabolites levels derived from LA, such as 10-hydroxy-cis-12-octadecenoic acid, were significantly reduced in the intestinal tract of rats fed a high-fat diet for 2 weeks, but were reversed by supplementation with LA (52). During adipocytes differentiation, some transcription factors, including CCAAT/enhancer binding proteins and PPAR-γ, activate fat production. Studies have found that citrus flavonoids inhibit intracellular TG and fat accumulation, and reduce PPAR-γ2 expression (53). The leukotriene family is a metabolite catalyzed by arachidonic acid through 5-lipoxygenase, and is an important indicator of the inflammatory response in atherosclerosis (54). In this study, 12-keto-tetrahydroleukotriene B4 was enriched in the PPAR signaling pathway in the POP and herb groups. POPs and purslane supplementation significantly reduced 12-keto-tetrahydro-leukotriene B4 levels (Supplementary Tables 2, 3). Therefore, POPs in P. oleracea may improve serum lipid levels in aging individuals mainly by regulating the PPAR signaling pathways and LA metabolism.

Conclusion
Our study investigated the effect of POPs on lipid levels in aging rats, based on its effect on intestinal microbiota and their related metabolites. The results showed that POPs can promote the growth of Coprococcus-2, a complex plant carbohydrate-degrading bacterium, and reduce the abundance of some obesity-related bacteria, including Lactobacillus and Muribaculaceae. The improvement of the gut microbiomes affects the production of bacteria-derived metabolites, such as the decrease in 5beta-cholestane-3alpha,7alpha,12alpha,25tetrol, and vaccenic acid levels in fatty acyls in steroids and steroid derivatives and increase of soyasapogenol E level in prenol lipids, and MG (24:0/0:0/0:0) in glycerolipids. Therefore, these results suggest that the effect of POPs on lipid levels in aging rats is related to the regulation of the intestinal microbiota and its metabolites.

Data availability statement
The datasets presented in this study can be found in online repositories. The names of the repository/repositories Frontiers in Nutrition 13 frontiersin.org