Effects of phycocyanin in modulating the intestinal microbiota of mice

Abstract The health‐promoting effects of phycocyanin (PC) have become widely accepted over the last two decades. In this study, we investigated the effects of different doses of PC in modulating the intestinal microbiota and the intestinal barrier in mice. Six‐week‐old male C57BL/6 mice were treated with PC for 28 days. Fecal samples were collected before and after PC intervention, and the microbiota were analyzed by 16S rRNA high‐throughput sequencing. Bacterial abundance and diversity increased after PC intervention. Saccharolytic bacteria of the families Lachnospiraceae and Ruminococcaceae, which can produce butyric acid, increased after PC treatment. The family Rikenellaceae, which contains hydrogen‐producing bacteria, also increased after PC intervention. The PC treatment reduced intestinal permeability and increased the intestinal barrier function, as demonstrated by hematoxylin–eosin staining and reduced serum lipopolysaccharide levels. The modulating effects on the intestinal microbiota were more favorable in the low‐dose PC group.

gut environment. Diet is a key factor influencing the intestinal environment (Power, O'Toole, Stanton, Ross, & Fitzgerald, 2014). The intake of the three major dietary nutrients (carbohydrates, proteins, and fats) can significantly affect the composition of the microbiota (Scott, Gratz, Sheridan, Flint, & Duncan, 2013).

Phycocyanin (PC) is a light-harvesting protein in the genus
Arthrospiraplatensis, which participates in algal photosynthesis. It is an excellent natural dietary pigment (Li et al., 2017). The role of PC in health promotion has become widely accepted over the past two decades (Wu et al., 2016). Phycocyanin reportedly has many biological functions, including as an antitumor, anti-inflammatory, and immunity-enhancing agent (Jiang et al., 2017). Undigested proteins can be fermented by the intestinal microbiota. According to previous studies, proteins promote the growth of intestinal bacteria, and many of the nutrients available to these bacteria in the intestine derive from undigested proteins from the host's diet (Consortium et al., 2012).
The moderate restriction of dietary protein altered the composition of the gut microbiota and improved the ileal barrier function of adult pigs (Chen et al., 2018). The microbial composition and a wide range of microbial metabolites played a complex role in various host processes, such as energy harvesting, recovery from inflammation and infection, resistance to autoimmunity, and endocrine signaling, which affect brain function through the intestinal-brain axis (Han et al., 2016;Hollister, Gao, & Versalovic, 2014). The small intestinal and colonic microbes are also considered potential sources of amino acids in animals (Miller & Ullrey, 1987).
Mouse and human genes have a high degree of similarity (99% of genes in mice have homologues in the human genome). In this study, male C57BL/6 mice were selected as an experimental model and 16S rRNA high-throughput sequencing was used to analyze the effects of PC treatment on the intestinal microbiota.

| Materials
Phycocyanin was purchased from King Dnarmse Spirulina Company (Fuqing, China). Maintenance diets for the mice were purchased from Pengyue Laboratory Animal Company (Jinan, China). Table 1 shows the daily diet of the mice.

| Animals and sample collection
A total of 27 male, 6-week-old, specific-pathogen-free C57BL/6 mice were used in this study. The study design was approved by the Animal Care and Maintenance Committee of Shandong International Biotechnology Park (SCXK 20140007), and adhered to the guidelines of the Institute of Health/Institutional Animal Care and Use Committee of Binzhou Medical University. All animals were allowed 1 week to acclimatize to 23 ± 2°C and 50%-60% humidity, with free access to food.
After acclimatization for 1 week, the mice were randomly separated into three groups: control group (n = 9), 50 mg/kg PC group (n = 9), and 100 mg/kg PC group (n = 9), and each group of mice was housed in a single cage. The 50 mg/kg PC group was given 5 mg L −1 day −1 PC by gavage; the 100 mg/kg PC group was given 10 mg L −1 day −1 PC by gavage; and the control group was given an equivalent volume of deionized water by gavage, once a day. Mouse bodyweights were measured at the beginning (0 weeks) and end (4 weeks) of the experiment. At the start and end of the experimental period (28 days), the mice were transferred individually to separate sterilized cages and their feces were collected. The samples were immediately stored at −80°C for the subsequent sequencing of the V3 and V4 regions of the bacterial 16S rRNA. At the end of the experiment, blood samples were collected, allowed to stand at 4°C for 30 min, and then centrifuged at 5,000g for 10 min at 4°C. The supernatants were collected and stored at −80°C. At the end of the experiment, ileal and colonic samples were collected from the mice and stored at −80°C until analysis.

| Extraction of genomic DNA
The fecal samples were thawed on ice and the total genomic DNA was extracted with the QIAamp DNA Stool Mini Kit (Qiagen, Germany), according to the manufacturer's instructions. Agarose gel electrophoresis was used to determine the purity and concentration of the DNA. An appropriate amount of sample was diluted to 1 ng/ μl with sterile water.

| Amplicon generations
The bacterial universal V3-V4 region of the 16S rRNA gene was amplified. Genomic DNA was used as the template for PCR, with specific primers (338F and 806R) containing a barcode, Phusion®

| Data analysis
Fragments of paired-ended sequencing were spliced with FLASH, and the clustering of operational taxonomic units (OTUs) and species classification were performed with a 97% similarity threshold. The RDP classifier was used to annotate the representative sequences of each OTU to obtain the corresponding species information and species-based abundance distribution.
The sequences were analyzed with the QIIME software, and αand β-diversities were analyzed with the Perl and R software to determine the species richness and the evenness of the samples.
The samples were weighted with principal coordinates analysis (PCoA) based on the UniFrac distances, and were clustered to identify the differences in the community structures of the different samples and groups.

| Intestinal morphology
The intestinal morphology was analyzed with hematoxylin-eosin (HE) staining. Ileal and colonic tissue samples were immediately fixed in polyoxymethylene after collection, for intestinal morphometry. The tissues were dehydrated and embedded according to standard procedures. The tissues in the paraffin blocks were cut into 4-µm sections, stained with HE, and examined with a conventional BX-51M microscope (Olympus). Image-Pro Plus 5 software was used to observe the intestinal and colonic intestinal morphology.

| Estimation of serum lipopolysaccharide (LPS) levels
The levels of serum LPS were determined with an enzyme-linked immunosorbent assay (ELISA; Shanghai ELISA Kit). The procedure was performed according to the manufacturer's guidelines. The optical density of each well was determined at 450 nm with a M200 automatic enzyme label analyzer (Tecan, USA).

| Statistical analysis
A homogeneity test of variance was performed before the difference analysis. All data are shown as means ± SDs. The differences between the different groups were tested by one-way ANOVA (Tukey's test). All statistical analyses were performed with the SPSS 19.0 software, and p < 0.05 was considered to indicate significant differences.

| Mouse bodyweights
The average bodyweights of the three groups of mice increased in the 4-week experimental period, but there was no significant difference between the control and experimental groups ( Figure 1).  (f) Bacterial richness in the gut estimated with the ACE value. CON-0w, blank control group, week 0; CON-4w, blank control group, week 4; LPC-0w, 50 mg/kg PC group, week 0; LPC-4w, 50 mg/kg PC group, week 4; HPC-0w, 100 mg/kg PC group, week 0; HPC-4w, 100 mg/kg PC group, week 4 (n = 6 per group). All data are shown as means ± SDs. Data were analyzed with one-way ANOVA (Tukey's test), *indicates a significant difference compared with the control group (p < 0.05) At the genus level, the relative abundance of Alistipes (phylum Bacteroidetes, family Rikenellaceae) increased significantly (p < 0.05) after PC interventions for 4 weeks ( Figure 6).

| PCoA after administration of PC
To explain the effects of PC on the variations in the intestinal microbiota, we conducted PCoA. A PCoA plot was generated based on the bacterial genera (Figure 7). The percentage of variation (81.13%) indicated that the separation of the fecal microbiota between the groups was caused by PC intervention. The composition of the fecal microbiota in each group was similar before PC intervention, and the differences between the groups occurred after PC treatment for 4 weeks.

| Intestinal morphology
The effect of dietary PC on the intestinal morphology was assessed by histological examination (Figure 8). The ileal and colonic tissues from the PC groups were more integrated than those from the control group. Compared with the control group, the ileal and colonic tissues in the PC dietary treatment groups showed greater villus height and more intensive goblet cells.

| Detection of serum LPS levels
The levels of serum LPS were detected by an ELISA. The average serum LPS level was modestly but not significantly reduced in the PC groups (Figure 9), and these reductions may have correlated with colon permeability.

| Correlations of bacterial abundance and mouse phenotypes (LPS level and villus height)
Spearman

| D ISCUSS I ON
Phycocyanin, a pharmaceutically active compound, has attracted increasing attention in recent years. Many studies have shown that PC has many health benefits, including antioxidant, immunomodulatory, and anti-inflammatory activities (Wu et al., 2016). However, its influence on the intestinal microbiota is still unclear. In this study, PC was administered to mice to study its effects on the intestinal microbiota and the gut barrier. The most abundant phyla in the gut were Bacteroidetes and Firmicutes, which accounted for more than 90% of the total bacteria, consistent with previous reports (Zhai, Zhu, Qin, & Li, 2018).
Changes in the ratio of Bacteroidetes to Firmicutes significantly affect the host's health (Pyo, Pajarillo, Kyoung, Heebal, & Dae-Kyung, 2016). In this study, the ratio of Bacteroidetes to Firmicutes decreased significantly after PC interventions for 4 weeks and was lower in the low-dose PC group than in the high-dose PC group.
Research has shown that the ratio of Bacteroidetes to Firmicutes is lower in obese individuals than in non-obese individuals (Ley, Turnbaugh, Klein, & Gordon, 2006). This suggests that PC may lead to weight gain. However, we observed no weight differences between the PC groups and the control group, although the 4week experimental period may have been too short to observe such differences.
The relative abundance of bacteria in the phylum Deferribacteres increased significantly after PC treatment for 4 weeks (p < 0.05).
Members of Deferribacteres gain energy through obligate or facultative anaerobic metabolism, and all use iron, manganese, or nitrate for anaerobic respiration. They can also produce energy for their host by fermentation (Huber, Foesel, Pascual, & Overmann, 2017).
Bacteria belonging to the Deferribacteraceae family also increased more significantly in the low-dose PC group than in the high-dose PC group.
Phycocyanin intervention significantly increased the abundance of the family Lactobacillaceae (in the low-dose PC group), which had a bifidogenic effect. The family Lactobacillaceae contains wellknown probiotic bacteria, which benefit the host's health. The bifidogenic effect correlated inversely with the PC dose. However, the effects of PC were far more complex than merely a bifidogenic effect, and other microbes in the gut were also affected by PC.
Phycocyanin treatment caused a dramatic increase in the abundance of the families Lachnospiraceae and Ruminococcaceae, and the low-dose PC group showed more significant effects. Each of these bacterial taxa ferments carbohydrates and produces butyric acid (Zackular, Rogers, & Schloss, 2014) and they play important roles in maintaining gut health (Louis & Flint, 2010). Butyrate is produced by intestinal microbial fermentation (Hamer et al., 2008), and can provide energy directly to the gut epithelium, improving intestinal digestion, the absorption of nutrients, and intestinal immunity (Shangari et al., 2007;Trivedi & Jena, 2013;Vital, Karch, & Pieper, 2017). Members of the family Lachnospiraceae in the genus Clostridium are reported to protect their host against colon cancer by producing butyric acid (Meehan & Beiko, 2014). The family F I G U R E 6 Mean changes in the relative abundance of the genus Alistipes in fecal samples after PC intervention. CON-4w, blank control group, week 4; LPC-4w, 50 mg/kg PC group, week 4; HPC-4w, 100 mg/kg PC group, week 4 (n = 6 per group). All data are shown as means ± SDs. Data were analyzed by one-way ANOVA (Tukey's test); *indicates a significant difference compared with the control group (p < 0.05) F I G U R E 7 Separation of intestinal microbiota based on PC interventions. Principal coordinates analysis (PCoA) of the control group and PC groups at time point 0 and week 4 based on weighted UniFrac distances. (a) PCoA analysis of different groups at the beginning of the experiment. (b) PCoA analysis of the different groups at the end of the experiment. CON-0w, blank control group, week 0; CON-4w, blank control group, week 4; LPC-0w, 50 mg/kg PC group, week 0; LPC-4w, 50 mg/kg PC group, week 4; HPC-0w, 100 mg/kg PC group, week 0; HPC-4w, 100 mg/kg PC group, week 4 (n = 9 per group) Ruminococcaceae, which is associated with butyrate production, is one of the most abundant families in the order Clostridiales, and reportedly maintains intestinal health (Biddle, Stewart, Blanchard, & Leschine, 2013). Therefore, PC may stimulate butyrate production and thus potentially improve gut health, and low-dose PC had a greater effect than high-dose PC.
A significant increase in bacteria of the family Rikenellaceae was observed in the PC-treated groups. Members of the family Rikenellaceae are hydrogen-producing bacteria that selectively neutralize cytotoxic reactive oxygen species (ROS) and protect cells from oxidative stress (Chen, Zuo, Hai, & Sun, 2011). It has been reported that endogenous hydrogen reduces oxidative stress and ameliorates the symptoms of inflammatory bowel disease (Si, Cheng, Wyckoff, & Qiang, 2016), which would benefit the host's health. Low-dose PC had a greater effect on the level of Rikenellaceae than the higher dose.
At the genus level, PC intervention significantly increased the abundance of Alistipes. Alistipes produces succinic acid as the main metabolic end product of glucose fermentation and uses iso-C15:0 as its main long-chain fatty acid (Song et al., 2006).

| CON CLUS IONS
In this study, we evaluated the effects of PC treatment on the intestinal microbiota and gut permeability in mice. The administration of PC increased the richness and diversity of the intestinal microbiota, perhaps because PC promoted the growth of the intestinal villi, enhancing the digestion and absorption of food, and thereby increasing the nutrients available to the intestinal microbiota. Phycocyanin intervention significantly increased the abundance of the family Lactobacillaceae, which exerts a bifidogenic effect. The family Lactobacillaceae contains well-known probiotic bacteria, which benefit the host's health.
Dietary PC also increased the carbohydrate decomposition and the numbers of short-chain fatty acid-producing bacteria. Therefore, PC may improve gut health by stimulating the production of short-chain fatty acids. The family Rikenellaceae, which contains hydrogen-producing bacteria that can selectively neutralize cytotoxic ROS and protect cells from oxidative stress, increased after the PC interventions.
Dietary PC increased the colonic epithelial barrier function and prevented endotoxins entering the systemic circulation. These modulatory effects were related to the PC dose, with the low-dose PC group showing better effects than the high-dose PC group. Further studies are required to determine the roles and underlying mechanisms of PC F I G U R E 9 Effect of dietary PC on the serum LPS levels in C57BL/6 mice. CON-4w, blank control group, week 4; LPC-4w, 50 mg/kg PC group, week 4; HPC-4w, 100 mg/kg PC group, week 4 (n = 5 per group) F I G U R E 1 0 Correlation analysis of bacterial abundance and mouse phenotypes (LPS level and villus height). Red represents a positive correlation and blue represents a negative correlation. The intensity of the colors indicates the degree of correlation between the abundant OTUs and the host parameters. *p < 0.05，**p < 0.01 in modulating other intestinal microbiota. However, our data provide a basis for the future modulation of intestinal microbes by treatment with PC.

ACK N OWLED G EM ENTS
This work was supported by The National Key Research and Development Program of China (2018YFD0901102).

CO N FLI C T O F I NTE R E S T S
The authors confirm that they have no conflicts of interest.

E TH I C S S TATEM ENT
The study design was approved by the Animal Care and Maintenance Committee of Shandong International Biotechnology Park (SCXK 20140007), and adheres to the guidelines of the Institute of Health/ Institutional Animal Care and Use Committee.

DATA ACCE SS I B I LIT Y
The raw data were uploaded to the National Center for