Berberine improved experimental chronic colitis by regulating interferon-γ- and IL-17A-producing lamina propria CD4+ T cells through AMPK activation

The herbal medicine berberine (BBR) has been recently shown to be an AMP-activated protein kinase (AMPK) productive activator with various properties that induce anti-inflammatory responses. We investigated the effects of BBR on the mechanisms of mucosal CD4+T cell activation in vitro and on the inflammatory responses in T cell transfer mouse models of inflammatory bowel disease (IBD). We examined the favorable effects of BBR in vitro, using lamina propria (LP) CD4+ T cells in T cell transfer IBD models in which SCID mice had been injected with CD4+CD45RBhigh T cells. BBR suppressed the frequency of IFN-γ- and Il-17A-producing LP CD4+ T cells. This effect was found to be regulated by AMPK activation possibly induced by oxidative phosphorylation inhibition. We then examined the effects of BBR on the same IBD models in vivo. BBR-fed mice showed AMPK activation in the LPCD4+ T cells and an improvement of colitis. Our study newly showed that the BBR-induced AMPK activation of mucosal CD4+ T cells resulted in an improvement of IBD and underscored the importance of AMPK activity in colonic inflammation.

Several reports have already shown the utility of BBR in the treatment of IBD [17][18][19][20] . Various mechanisms, such as repair of the epithelial barrier function and the regulation of innate and adoptive immune responses, have been indicated in these reports. However, detailed analyses, including the mechanisms of the AMPK activity against CD4 + T cells, especially LP CD4 + T cells, which are considered to be involved in the pathogenesis of IBD 2 , have not been performed. In addition, the widely used models of colitis are chemically induced, which is not appropriate for representing the chronic inflammatory features of IBD.
In this report, we investigated the effect and inhibitory mechanisms of BBR on LP CD4 + T cells of IBD-in terms of the cellular energy metabolism related to AMPK activity-using a T cell transfer colitis model involving the transfer of naïve (CD4 + CD45RB high ) T cells into congenic immunodeficiency mice to induce CD4 + T cell-specific colitis. Our results provide new insight into how BBR regulates the LP CD4 + T cell function as well as underscore the importance of AMPK activity in colonic inflammation. In the future, given these results, the involvement of AMPK activity in colonic inflammation may lead to the development of new therapeutic targets that are safe and useful for treating IBD.

BBR suppressed inflammatory cytokines of LP CD4 + t cells from colitis SciD mice in vitro.
We first assessed the inhibitory effect of BBR on LP CD4 + T cell activation in vitro. LP CD4 + T cells of colitis SCID mice (colitis LP CD4 + T cells) injected with CD4 + CD45RB high T cells were isolated and stimulated with PMA plus ionomycin mixed with BBR or not. In this colitis model, a large amount of the Th1-related cytokine IFN-γ and a small amount of the Th17-related cytokine IL-17A are known to be secreted by LP CD4 + T cells 21,22 . Therefore, we analyzed the production of IFN-γ and IL-17A. BBR exerted a dose-dependent inhibitory effect but showed no cytotoxic effect at concentrations of 100 µM (Fig. 1A, Supplementary Fig. S1). Therefore, we performed subsequent experiments with BBR 100 µM.
As shown in Fig. 1B,C, the frequency of IFN-γ-producing CD4 + T cells in BBR-treated colitis LP CD4 + T cells was significantly lower than in non-BBR-treated colitis LP CD4 + T cells. The frequency of IL-17A-producing CD4 + T cells was also lower in BBR-treated cells than in untreated cells. These results indicated that BBR directly affected the colitis LP CD4 + T cells, thereby reducing the Th1/Th17 responses. BBR suppressed the Th1/Th17-related JAK/STAT pathway of LP CD4 + t cells collected from colitis SCID mice. We further investigated the expression of signal molecules correlated with Th1/Th17 activation. The JAK/STAT pathway is a major signal that regulates Th cell differentiation and their function 23,24 . The Th1-related signal transduction molecules JAK1, JAK2, and STAT1 were all suppressed by BBR addition. The Th17-related molecule STAT3 was also suppressed by BBR addition ( Fig. 2A-D). These results further confirmed the direct effect of BBR on the immune responses in colitis LP CD4 + T cells.
BBR increased the AMPK activity and regulated the IFN-γ and IL-17A secretion from colitis LP CD4 + T cells. Given that recent reports indicated that AMPK plays a key role in T cell activation and that BBR increased AMPK activity, we hypothesized AMPK activity as a vital mechanism on BBR effects. Earlier studies had established that phosphorylation of AMPK at Thr172 correlates with AMPK activity 25 . We therefore first compared the AMPK expression by Western blotting between BBR-treated colitis LP CD4 + T cells and non-BBR-treated colitis LP CD4 + T cells. As expected, the phosphorylation state of AMPK at Thr172 (p-AMPK) expressions was significantly increased in BBR-treated colitis LP CD4 + T cells (Fig. 3A).
Next, we examined the relationship between the AMPK activity and inflammatory cytokines. We measured IFN-γ-and IL-17A-producing colitis LP CD4 + T cells using AMPK agonist, 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR) and antagonist, Compound C (C.C.). The addition of AICAR resulted in the suppression of the frequency of IFN-γ-and IL17A-producing colitis LP CD4 + T cells. In contrast to the effect of AICAR on colitis LP CD4 + T cells, the addition of C.C. increased the frequency of IFN-γ-producing colitis LP CD4 + T cells as well as that of IL-17A-producing LP CD4 + T cells (Fig. 3B,C). Furthermore, the addition of mixing with AICAR and C.C. increased the frequency of IFN-γ-and IL-17A-producing colitis LP CD4 + T cells in comparison to the addition of AICAR alone and decreased the frequency of IFN-γ-and IL-17A-producing colitis LP CD4 + T cells in comparison to the addition of C.C. alone (Fig. 3B,C).
Western blotting analysis showed that the p-AMPK activity of colitis LP CD4 + T cells was significantly increased by AICAR treatment and significantly suppressed by C.C. treatment (Fig. 3D,E). Furthermore, the p-AMPK activity of colitis LP CD4 + T cells was significantly suppressed by combination AICAR and C.C. treatment in comparison to AICAR and increased in comparison to C.C. (Fig. 3D,E).
These data strongly indicated that the production of IFN-γ and IL-17A by colitis LP CD4 + T cells was regulated through AMPK activities.
BBR affected the oxidative phosphorylation and decreased the total adenosine triphosphate production. AMPK is activated in response to stresses that deplete cellular adenosine triphosphate (ATP), such as low glucose, hypoxia, ischemia and heat shock [26][27][28] . Therefore, we compared the ATP production in BBR-treated colitis LP CD4 + T cells to that of non-BBR-treated colitis LP CD4 + T cells. As expected, the ATP production of BBR-treated colitis LP CD4 + T cells was significantly decreased relative to that of non-BBR-treated colitis LP CD4 + T cells (Fig. 4A).
ATP is produced mainly by two routes, glycolysis and oxidative phosphorylation (OXPHOS). Glycolysis is the metabolic pathway that extracts energy from glucose, which is used to reform ATP. OXPHOS is the metabolic pathway by which cells use enzymes to oxidize nutrients (glucose, fat, protein), thereby releasing energy, which is used to reform ATP. This takes place inside mitochondria. We therefore examined whether or not BBR worked on these two routes. We measured the glycolysis marker extracellular acidification rate (ECAR) and the OXPHOS marker oxygen consumption rate (OCR). BBR-treated colitis LP CD4 + T cells showed suppressed OCR and ATP productions ( Fig. 4B-D). In contrast, BBR-treated colitis LP CD4 + T cells showed elevated ECAR (Fig. 4E). This was due to the effect of increased AMPK activation compensating for the loss of ATP production in OXPHOS. However, according to the total ATP production in our experiment, glycolysis did not sufficiently compensate for ATP production.
These experiments suggest that BBR suppressed OXPHOS and total ATP production, resulting in AMPK activation. The cytotoxicity analysis of BBR. Colitis LP CD4 + T cells were stimulated with PMA plus ionomycin mixed with BBR (BBR) or not (Control). The bar graphs show number of LP CD4 + T cells. (B) Colitis LP CD4 + T cells were stimulated with PMA plus ionomycin mixed with BBR (BBR) or not (Control) for 8 h. After that, cells were collected, and intracellular staining was performed to analyze the CD3 + CD4 + IFN-γ + -or IL-17A +producing cells by flow cytometry. Representative flow cytometry images are shown. Flow cytometry showed the percentage of CD3 + CD4 + IFN-γ + -or IL-17A + -producing cells in colitis LP CD4 + T cells. (C) The bar graphs show the percentage of IFN-γ-and IL-17A-producing cells in LP CD4 + T cells. All data are reported as the mean ± SEM. N is 6 in each group. N.S. not significant. *P < 0.05, **P < 0.01. BBR ameliorated experimental colitis in vivo with changes in gut microbiota and decreased the production of ifn-γ and IL-17A. Next, we assessed the BBR effects on SCID colitis mouse model which injected with CD4 + CD45RB high T cells. We gave colitis mice a diet mixed with BBR or not and evaluated   Supplementary Fig. S3. All gels were run in the same experimental conditions (see material and methods for details). All data are reported as the mean ± SEM. N is 5 in each group. *P < 0.05, **P < 0.01. www.nature.com/scientificreports www.nature.com/scientificreports/ the severity of colitis. BBR at a concentration of 0.12% was not effective ( Supplementary Fig. S4), but that at 0.35% was effective. The clinical and histological scores of the BBR diet group were lower than those of the non-BBR diet group (Fig. 5A-D).
To further confirm the effect of BBR on experimental colitis, changes in the immune reactions were examined. The numbers of CD4 + T cells in LP were reduced by BBR administration (Fig. 5E). The numbers of IFN-γ-or IL-17A-producing LP CD4 + T cells were lower in the BBR diet group than in the non-BBR diet group (Fig. 5F). We evaluated the AMPK expression of LP CD4 + T cells in both groups; the expression in the BBR-treated group was significantly higher than in the non-BBR-treated group (Fig. 5G).
Given that BBR has been shown to regulate Treg cells by modifying the microbiota in a chemically induced IBD model and to regulate Bcl-2 in an in vitro autoimmune disease model 29,30 , we next examined the change in the frequency of Foxp3 + Treg cells and the Bcl-2 expression. The frequency of Foxp3 + Treg of LP CD4 + T cells were higher and Bcl-2 + of LP CD4 + T cells were lower in the BBR diet group than in the non-BBR diet group (Fig. 5H,I).
These data indicated that BBR reduced the T cell inflammatory responses and ameliorated experimental colitis.  Gut microbiota disorder is a known pathogenesis of IBD 1 . Therefore, we also examined the gut microbiota using fecal samples of colitis mice at the time of sacrifice. In both groups, the gut community was mainly built by four major phyla, and significant differences were noted in the rates of the Bacteroides and Firmicutes phyla (Fig. 6A). Regarding lower classifications, genus-level comparisons showed statistically significant dissimilarities in Lactobacillus, Bacteroides, unidentified genus of S24-7 family and Sutterella found in the top 10 most abundant genera of both groups (Fig. 6B).
The alpha diversity was evaluated to assess the species richness within the gut microbiota community diversity of the two groups (Fig. 6C). There were no considerable differences in the number of observed OTU, Faith's phylogenetic diversity whole tree and the diversity of the species estimated by Chao1. However, interestingly, the species richness and evenness calculated by the Shannon index were significantly reduced with BBR supplementation.
The gut microbiota community was measured not only based on the diversity within each group but also between the two groups to provide an overview of the gut microbiota structure. An unweighted and weighted UniFrac Principle Coordinate Analysis (PCoA) and the UniFrac distance of gut microbiota community in both groups were analyzed (Fig. 6D). Both scatter plots demonstrated that gut microbiota structure was changed in response to BBR administration. The gray dots representing the BBR diet group gathered closely, while the black dots representing the non-BBR diet group seemed scattered. The largest principle coordinates (PC1) were quite high, accounting for 35.86% and 79.73% of the total variation in the unweighted and weighted analyses, respectively. In addition, 20.27% and 8.99% of the total variation were the values of second principle coordinate (PC2), corresponding to the unweighted and weighted analyses. The shifting of the gut composition under BBR intake was indicated in bar plots of the unweighted and weighted UniFrac distance analyses (Fig. 6E). These data suggested that BBR affected the gut microbiota composition, and this change might be one of the factors ameliorating colitis.

Discussion
In this report, we showed that BBR ameliorated CD4 + T cell-related chronic colitis in a mouse model and revealed the relationship between BBR and AMPK, including its metabolic pathway, through in vitro and in vivo experiments.
Our message based on the present data can be summarized in the following three points: 1) We adapted the CD4 + T cell-transfer model, a chronic colitis model that clearly differs from the widely used chemically induced acute colitis models; 2) BBR affected the AMPK-related metabolic pathway and reduced the inflammatory responses; and 3) BBR reduced the microbiota diversity, possibly by its antimicrobial action.
We used CD4 + T cell-transfer colitis models in our in vitro and in vivo experiments, although most previous reports used colitis models in which colitis was induced by chemicals, such as DSS [17][18][19][20] . Such chemically induced models showed acute colitis that did not accurately reflect the chronic inflammatory process observed in IBD patients. CD4 + T cell-transfer colitis models show chronic colitis with T cell infiltration, dependent on the microbiota, and may be superior to chemically induced models. The use of CD4 + T cell-transfer colitis models revealed a new inhibitory mechanism of BBR, particularly in relation to colitis LP CD4 + T cells.
The beneficial effects of BBR have been considered to occur due to the activation of AMPK, an energy metabolic sensor crucial for cellular energy homeostasis 9,10 . Recently, many reports have revealed that the immune metabolism is a key factor for controlling the immune cell function [11][12][13][14][15][16]31,32 . We therefore focused on AMPK and attempted to reveal its immunological effects, particularly in relation to colitis LP CD4 + T cells, data on which are scarce. Our results showed that BBR increased the AMPK activity in colitic LP CD4 + T cells. In addition, we showed that the AMPK activity regulated IFN-γ-and IL-17A-producing LP CD4 + T cells. These data suggested that the regulation of AMPK in colitis LPCD4 + T cells was important to control mucosal inflammation. We further investigated the relationship between AMPK and metabolic pathways. It is well-established that AMPK is activated under conditions of ATP shortage in order to restore the decreased intracellular energy and thereby ensure the cell survival and maintenance of the cell function. After AMPK activation, ATP-consuming anabolic pathways are turned off, and ATP-producing catabolic processes are stimulated. Our results showed BBR decreased the total ATP content, suggesting that the increase in the AMPK activity was induced by BBR through a reduction in the ATP content.
BBR has been shown to inhibit OXPHOS, an oxygen-dependent ATP-producing pathway. OXPHOS is a highly efficient way of generating ATP, and almost all aerobic organisms carry out OXPHOS. BBR has been shown to inhibit OXPHOS 33,34 . We therefore further investigated the effects of BBR on the ATP production of colitis LPCD4 + T cells. In our study, as expected, BBR inhibited the OXPHOS of colitis LPCD4 + T cells. In our consecutive AMPK metabolic pathway experiments, we hypothesized that one of the anti-inflammatory mechanisms of BBR was the inhibition of OXPHOS, followed by ATP reduction, AMPK activation, and finally the reduction of inflammatory cytokines. Another study showed that AMPK activity was not involved in the reduction of inflammatory cytokines, such as IFN-γ and TNF-α, using WT-spleen CD4 + and CD8 + T cells in AMPK conditional knockout mice 15 ; however, a precise analysis of the T cells infiltration in the immune mechanism-related IBD colon has not been performed. Thus, our data showing the mechanism of AMPK-related mucosal inflammation inhibition are novel findings and may highlight a new therapeutic target. The scavenging radical effect of BBR is All gels were run in the same experimental conditions (see Materials and methods for details). The numerical values represent the mean values of 6 samples per group. Pictures and dot plots of flow cytometry show representative samples from each group. All data are shown as the mean ± SEM for 6 mice per group. *P < 0.05, **P < 0.01. www.nature.com/scientificreports www.nature.com/scientificreports/ likely to be a key mechanism underlying the anti-colitis effect, as the reduction of OCR and ATP was evident 35,36 . However, our present results indicated the strong effect of BBR on AMPK activation, and an AMPK agonist such as BBR may be expected to reduce inflammatory cell activation. We therefore emphasize the importance of AMPK in our immune-mediated colitis model as well as the effects of BBR on AMPK activation.

Scientific RepoRtS
Our results also showed that BBR accelerated glycolysis. It is possible that this activation of glycolysis occurred as compensation for the loss of ATP by inhibiting OXPHOS. However, as we did not investigate the www.nature.com/scientificreports www.nature.com/scientificreports/ role of glycolysis in relation to LPCD4 + T cells, the mechanism was not clearly defined. Previous reports have shown that inflammatory cells (e.g. effector memory T cells [T EM cells]) tend to perform glycolysis. In contrast, anti-inflammatory cells, such as Foxp3 + regulatory T cells, more often perform OXPHOS 37 . Further studies are needed to gain more insight into the glycolysis of colitis LPCD4 + T cells.
In our in vivo experiments, BBR increased the AMPK activity and suppressed inflammatory cytokines in colitic LP CD4 + T cells, suggesting that the same mechanism for suppressing inflammation as in in vitro experiments was involved. Concerning inflammatory cytokine control mechanisms by BBR, previous reports have shown that Bcl-2 and Foxp3 + Treg cells of LP CD4 + T cells were involved in regulating inflammation 38,39 . As previously reported, BBR reduced the expression of Bcl-2, suggesting the induction of apoptosis of inflammatory cells 30 . Given that our present model contained no natural Tregs, we could investigate only the induced Tregs 40 . The induced Tregs in our present model were increased by BBR administration, in line with the findings of a previous report 29 (Fig. 5H,I). We also did not examine the detailed relationships between the effects of BBR and the characteristics of these cells, including AMPK activity, but the effects of BBR on these cells might offer further insight into the regulation of mucosal inflammation.
Our data obtained from in vivo experiments showed that BBR affected the gut microbiota. The gut microbiota have been postulated to induce IBD, including CD and UC 41 . The modulatory action of BBR was shown by changing the bacterial taxonomic structure and reducing the bacterial diversity. Bacteroidetes, a Gram-negative phylum, were significantly dominant under BBR supplementation, while the rate of Firmicutes, a Gram-positive phylum, was extremely low. A higher response to BBR of Gram-positive bacteria than of Gram-negative bacteria was reported in previous studies 42 . This phenomenon may be due to the permeability barrier of Gram-negative bacteria, which can eject both synthetic and natural toxins, including BBR alkaloids, through the activities of their multidrug resistance pumps. The penetration of this compound may also be limited due to the outer membranes of these bacteria 43,44 .
BBR had no effect on regulating the richness of gut microbiota in colitis mice, but this reagent seemed to cause a remarkable reduction in the gut microbiota diversity. The wide antimicrobial spectrum of BBR has been shown to decrease the diversity of intestinal flora 45,46 . Our data revealed that BBR reduced the severity of colitis and that the change in the gut microbiota was due to the anti-inflammatory effects of BBR. We therefore considered the change in the gut microbiota in our experiment to be reasonable.
In summary, we showed for the first time that BBR ameliorated CD4 + T cell-related chronic colitis in a mouse model with changes in gut microbiota via AMPK activity, possibly induced by the inhibition of OXPHOS, according to in vitro and in vivo experiments. The AMPK activity of colonic inflammation may represent a new therapeutic target for IBD, and medicines targeting the AMPK activity, such as BBR, may be recognized as new candidates for use in IBD therapy.

Materials and Methods
Animals. Balb/c and CB17-icr SCID mice were purchased from Japan CLEA (Tokyo, Japan). Mice were maintained under specific-pathogen-free conditions in the Animal Care Facility of Okayama University. The Balb/c donors and recipients were used at 5 to 7 weeks of age. For in vivo experiments of the BBR effect, donors and CB17-icr SCID recipients were used at 5 weeks of age.
All protocols and procedures conformed to the guidelines of the Okayama University Committee for Care and Use of Laboratory Animals and were approved by the Animal Experiments Ethics Committee of Okayama University. Induction of CD4 + CD45RB high T cell-transfer mouse colitis. CD4 + T cells were positivity isolated from the spleens of Balb/c mice using anti-CD4 (L3T4) (Miltenyi Biotec) according to the manufacturers' instructions. Enriched CD4 + T cells were labeled with APC anti-mouse CD4 and FITC anti-mouse CD45RB. The CD4 + CD45RB high T cells were isolated using a FACS Aria I (Becton Dickinson). These cells were 98.0%-100% pure on a reanalysis. The isolated CD4 + CD45RB high T cells were intraperitoneally injected into the recipient CB17-icr SCID mice with 3 × 10 5 cells per animal. After transfer, the recipient mice were monitored for clinical signs, including a hunched posture, piloerection, diarrhea, and blood in the stool. At autopsy, their clinical scores were assessed as the sum of three parameters 2 : hunching and wasting, 0 or 1; colon thickening, 0-3 (0, none; 1, mild; 2, moderate; or 3, extensive); and stool consistency, 0-3 (0, normal beaded stool; 1, soft stool; 2, diarrhea; 3, bloody stool). www.nature.com/scientificreports www.nature.com/scientificreports/ A histological examination Tissue samples were fixed in phosphate-buffered saline containing 6% neutral-buffered formalin. Paraffin-embedded sections (5 μm) were stained with hematoxylin and eosin. Three tissue samples from the proximal and distal parts of the colon were prepared. The sections were analyzed without prior knowledge of the type of T cell reconstitution or treatment. The mean degree of inflammation in the colon was calculated using a modification of a previously described scoring system 2 , consisting of the sum of three parameters: crypt elongation, 0-3; mononuclear cell infiltration, 0-3; and frequency of crypt abscesses, 0-3.

Isolation of mononuclear cells from murine organs.
Single-cell suspensions were prepared from the LP as described 2 . To isolate LP CD4 + T cells, the entire length of the colon was opened longitudinally and cut into small pieces. The dissected mucosae were incubated for 45 min with Ca 2+ -, Mg 2+ -free Hanks' balanced salt solution containing 1 mM dithiothreitol (Sigma-Aldrich) to remove mucus and then treated with 3.0 mg/ml collagenase (Roche Diagnostics GmbH, Berlin Germany) for 2 h. The cells were pelleted twice through a Ca 2+ -, Mg 2+ -free Hanks' balanced salt solution and then subjected to Ficoll-Paque density gradient centrifugation. Enriched LP CD4 + T cells were obtained by positive selection using anti-CD4 (L3T4) MACS magnetic beads. Of the resultant cells, > 95% were CD4 + T cells, as determined by flow cytometry.
Cell culture and treatments and intracellular staining of cytokines. LP CD4 + T cells from colitis SCID mice were cultured overnight with ionomycin (500 ng/ml), PMA (50 ng/ml) and BD GolgiStop (0.65 l/ml BD Pharmingen) mixed with 100 µM of sterile BBR 2 . After that, cells were collected, and their surface molecules were stained. Following cell fixation using a Cytofix/Cytoperm Kit (BD Pharmingen), the cells were incubated with PE anti-mouse IL-17A, FITC anti-mouse IFN-γ for 20 min. In some experiments, the designated concentrations (in text and figures) of Compound C (C.C.) and 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR) (Selleck Chemicals, Houston, TX, USA) were used to investigate the cellular functions of AMPK 15 . Each experiment contained a group with equal concentrations of DMSO as a control.
Metabolic measurements. LP CD4 + T cells enriched from colitis SCID mice were cultured by the method mentioned previously. The extracellular acidification rate (ECAR) and oxygen consumption rate (OCR) were measured using with an xXF96 Extracellular Flux Analyzer under mitochondrial stress test conditions (Seahorse). Assay buffer was made from non-buffered RPMI 1640 medium supplemented with 2.5 mM dextrose, 2 mM glutamine and 1 mM sodium pyruvate (all from Sigma). The baseline ECAR and OCR values were averaged among at least two technical replicates per sample for the first three successive time intervals 47 .

Measuring the intracellular ATP concentrations. LP CD4 + T cells enriched from colitis SCID mice
were cultured by the method described previously. After that, cells were collected and plated at 5 × 10 5 cells in 96-well plates. The cells were lysed with 100 µl lysis buffer and placed directly into the chamber of a luminometer (Flex Station3, Molecular Devices, CA, USA). Light emission was recorded after the addition of 100 µl of luciferin-luciferase solution (TOYO B-NET, Tokyo, Japan). The ATP values were averaged among at least two technical replicates per sample.
Western blotting analyses. Total proteins were extracted from cultured LP CD4 + T cells using a protein extraction kit (Minute Protein Extraction Kit; Invent Biotechnologies, Inc., Eden Prairie, MN, USA). Extracted proteins were quantitated by the BCA method (BCA protein Assay Kit; Takara Bio, Inc., Otsu, Japan) and diluted to the same protein concentrations. Equivalent amounts of each group were run on SDS polyacrylamide gels. Proteins were transferred electrically to a PVDF membrane (Bio-Rad, Hercules, CA, USA). The membranes were blocked using polyvinylidene fluoride Blocking Reagent (Toyobo, Osaka, Japan) for 1 h. After that, the membranes were incubated with the following rabbit monoclonal antibodies overnight at 4 °C: anti-AMPK(Thr172), anti-stat1(Tyr701), anti-stat3(Tyr705), anti-JAK1(Tyr 1034/1035), anti-JAK2 (Tyr221) (Cell Signaling Technology, Inc., Beverly, MA, USA). After three washes with Tris-buffered saline with Tween-20 (Sigma), the membranes were incubated with secondary antibodies for 1 h at room temperature. The rabbit monoclonal antibodies were used at a 1:1000 dilution and the anti-rabbit secondary antibodies (Cell Signaling Technology) at a 1:2000 dilution. Bands were detected by using a LAS 4000 imager (GE Healthcare Life Sciences, Pittsburgh, PA, USA). β-actin was used as a loading control. Contrast was adjusted with Adobe Photoshop Elements 14. The quantity in each sample was analyzed using an Image J64 (NIH, Bethesda, MD, USA).
Gut microbiome analyses of mice. Fresh fecal samples of mice were collected, immediately frozen by liquid nitrogen, and stored at −80 °C until use. The DNA purification of fecal samples was conducted according to the previous methods with slight modification 48 . The fecal DNA was sequenced using Illumina Miseq platforms according to the manufacturer's recommendation (San Diego, CA, USA).
The raw sequences of both the BBR diet group and non-BBR diet group were processed using the QIIME software package, version 1.9.1 (http://qiime.org/) _ENREF_41 49 . After quality filtering, all sequences were normalized to the smallest number of reads (5324 reads) before being assigned operation taxonomic units (OTUs) using the Greengenes reference database with a 97% similarity threshold, and the alpha and beta diversity were compared. The relative abundance of OTUs in each group was evaluated at the phylum and genus levels. Chao1, observed species, Faith's phylogenetic diversity whole tree and the Shannon diversity index were determined to estimate the alpha diversity. The weighted and unweighted UniFrac distance in beta diversity were evaluated to compare the microbial diversity between the BBR diet group and non-BBR diet group. All DNA sequences are available on the Sequence Read Archive under the relevant project identification number.