Oral Administration of Compound Probiotics Ameliorates HFD ......Oral Administration of Compound Probiotics Ameliorates HFD-Induced Gut Microbe Dysbiosis and Chronic Metabolic Inflammation

Oral Administration of Compound Probiotics Ameliorates HFD-InducedGut Microbe Dysbiosis and Chronic Metabolic Inflammation via the GProtein-Coupled Receptor 43 in Non-alcoholic Fatty Liver Disease Rats

Yinji Liang1& Shu Liang2

& Yupei Zhang2& Yuanjun Deng2

& Yifang He2& Yanning Chen2

& Chan Liu3& Chenli Lin3

&

Qinhe Yang2

# Springer Science+Business Media, LLC, part of Springer Nature 2018

AbstractThe aim of this study was to investigate how the effects of compound probiotics modulate the gut microbiota, short-chain fattyacid (SCFA), body composition, serum and liver lipids, and inflammatory markers in non-alcoholic fatty liver disease (NAFLD)rats. Twenty-four male SD rats were randomly divided into 3 groups: normal control group (standard feed), high-fat diet (HFD)feeding group (83% standard feed + 10% lard oil + 1.5% cholesterol + 0.5% cholate + 5% sucrose), and compound probioticsintervention group (HFD + 0.6 g × kg−1 × d−1 compound probiotics). The microbial population was assessed by 16S rDNAamplification and sequence analysis. Body composition, serum and liver lipids, serum inflammatory markers, colonic SCFAs,and relative proteins were assessed. The results showed that compound probiotics significantly reduced bodyweight, visceral andtotal fat mass, and the levels of hepatic TC and TG and serum TG, FFA, ALT, LPS, IL-1β, and IL-18 (P < 0.05). The proportionsof TM7 phylum (0.06 vs 1.57%, P< 0.05) clearly increased, while that of Verrucomicrobia phylum (5.69 vs 2.61%, P< 0.05)clearly decreased. Compound probiotics also increased the representation of Ruminococcus genus (0.95 vs 1.83%, P< 0.05),while the proportion ofVeillonella genus decreased (0.10 vs 0.03%,P < 0.05). The levels of colonic SCFAs and GPR43, NLRP3,ASC, and CASPASE-1 proteins also changed significantly (P < 0.05). Compound probiotics modulated gut microbiota, SCFAs,and their receptor GPR43 in NAFLD rats. These changes might inhibit lipid deposition and chronic metabolic inflammation inresponse to the insult of HFD.

Keywords Non-alcoholic fatty liver disease . Probiotics . Gut microbiota . Inflammation . G protein-coupled receptor 43

Introduction

At present, non-alcoholic fatty liver disease (NAFLD) isamong the most common liver diseases affecting the healthof both adults and children worldwide [1]. Modern medicine

has considered NAFLD to be a manifestation of metabolicsyndrome in the liver, often associated with obesity, dyslipid-emia, and insulin resistance [2]. According to the latest statis-tics, the incidence of NAFLD in adults was approximately25.24% (N = 8,515,431) worldwide [3], and the number ofcases was approximately 1 billion [4], approximately 30–40% in men and 15–20% in women [5]. The incidence ofNAFLD in China was 15 to 30% [6], and the trend was in-creasing annually.

Increasing evidence suggests that NAFLD was a multisys-tem disease. In addition to the progression of end-stage liverdisease, NAFLD is also closely related to many diseases, suchas cardiovascular disease (CVD), type 2 diabetes mellitus(T2DM), chronic kidney disease (CKD), and cancer [7].More than 70% of T2DM patients have been found to haveNAFLD [8, 9]. A recent meta-analysis showed that the totalmortality of NAFLD increased by 57% due to NAFLD-related CVD [10]. In addition, NAFLD increases the

Electronic supplementary material The online version of this article(https://doi.org/10.1007/s12602-017-9378-3) contains supplementarymaterial, which is available to authorized users.

* Qinhe [email protected]

1 School of Nursing, JinanUniversity, No. 601 HuangpuAvenueWest,Guangzhou, Guangdong 510632, China

2 School of Traditional Chinese Medicine, Jinan University, No. 601Huangpu Avenue West, Guangzhou, Guangdong 510632, China

3 School of Basic Medicine, Jinan University, No. 601 HuangpuAvenue West, Guangzhou, Guangdong 510632, China

Probiotics and Antimicrobial Proteinshttps://doi.org/10.1007/s12602-017-9378-3

incidence and severity of CKD [11]. A comprehensive under-standing of the harm to human health caused by NAFLD hasbeen obtained. However, the effective prevention and treat-ment of NAFLD in clinical practice remain lacking, and re-search on the pathogenesis of NAFLD is particularly urgent.

Recent research has led to widespread acceptance ofthe Bmultiple hit^ hypothesis regarding the pathogenesisof NAFLD [12]. The multiple hit factors mainly refer tothe interaction of host genes, the environment (especiallydiet), and intestinal microorganisms, which could lead togut microbiota dysbiosis, cause intestinal mucosa barrierdysfunction, and induce the translocation of the gut mi-crobiota and its metabolites [13, 14]. These factors inter-act to cause liver fat degeneration and activate the innateimmune system, especially Toll-like receptors (TLRs) andNOD-like receptors (NLRs), inducing inflammation, celldeath, or progressive liver damage [12]. Professor JeffreyGordon’s team conducted a study on gut microbiota thatconfirmed gut microbiota imbalance to be closely relatedto metabolic diseases such as malnutrition, obesity, anddiabetes [15]. However, the composition of the gut micro-bial population may change dynamically in different en-vironments, nutrition conditions, or immune states. In ad-dition, disorders of the gut microbiota exert a profoundimpact on the host health and a variety of diseases [16]. Arandomized controlled trial showed that the populations ofaerobic bacteria such as Enterobacteriaceae andEnterococcus faecalis were significantly increased in pa-tients with NASH compared with the healthy controlgroup, while Lactobacillus, Bifidobacteria, and other an-aerobic bacteria were decreased [17]. The severity ofNAFLD is also closely related to the abundance of γ-Proteus in the intestine [18]. Despite some controversialresults, these studies indicated that certain specific phy-lum, class, order, family, genus, or species populationsmay be beneficial or harmful to NAFLD patients.Therefore, the regulation of the intestinal ecological bal-ance may be an important means of preventing andtreating NAFLD. The intestinal ecological balance de-pends on the stability of the intestinal epithelial mucosa,the gut microbiota, and the regulation of its metabolites.Recently, some studies found that the gut microbiota, as-sociated high concentrations of intestinal SCFAs and Gprotein-coupled receptors (GPCRs), played a key role inNAFLD [19–22], but it remained unclear how the gutmicrobiota communicated with the liver. Thus, due tothe close relationship between the liver and gut, the ques-tion of whether compound probiotics could regulate thegut microbiota and its metabolites in NAFLD is worthy offurther study.

Our study investigated the effects of compound probioticson the gut microbiota and its impact on body composition,serum and liver lipids, serum inflammatory markers, short-

chain fatty acid (SCFA) production, and G protein-coupledreceptor 43 (GPR43) in NAFLD rats.

Materials and Methods

Animals and Dietary Treatments

Twenty-four male Sprague Dawley rats (6 weeks old) wereobtained from the Laboratory Animal Research Center ofGuangzhou University of Traditional Chinese Medicine(Animal License No. SCXK (Yue) 2013-0034). The basic dietand high-fat diet (HFD) were provided, respectively, by theAnimal Administration Laboratory of Jinan University andthe Center of Laboratory Animal Science of Guangdong(License No. SCXK (Yue) 2013-0002). Compound probioticswere provided by Professor Heping Zhang from InnerMongolia Agricultural University [23]. After adaptive feedingfor 1 week, the rats were randomly divided into three groupswith 8 rats per group, namely, the normal control group (NCgroup), the high-fat diet feeding group (HFD group), and thecompound probiotics intervention group (CP group). This an-imal experiment protocol was approved by the LaboratoryAnimal Ethics Committee of Jinan University. NC group ratswere fed standard feed, and the other groups were fed a HFD(83% standard feed + 10% lard oil + 1.5% cholesterol + 0.5%cholate + 5% sucrose). At the same time, each rat was, respec-tively, given distilled water or the relevant medicine by ga-vage. The HFD group rats were given deionized distilled wa-ter, and the CP group rats were given compound probiotics(more than 600 billion CFU/100 g) at 0.6 g × kg−1 × d−1. Thecompound probiotics contained 9 strains of probiotics, includ-ing 6 strains of Lactobacillus and 3 strains of Bifidobacteriumcombined with 15 g/100 g of the prebiotic galacto-oligosaccharide (GOS). Composition of feed used in the studywas shown in Supplemental Table 1–3. Body weight and foodintake were measured weekly. The rats were allowed to drinkwater freely and were kept in separate cages in an SPF animallaboratory with 12-h alternating periods of dark and light at18–22 °C for 16 weeks.

Gut Microbiota Sequencing Analyses

At 1–2 days before euthanasia, fresh stool samples (n = 8rats/group) were collected in sterile tubes and stored in liquidnitrogen. Fecal sample DNA was extracted with QiaAmpDNA Mini Kits (Qiagen, Valencia, CA, USA), and the bacte-rial genomic DNA was detected by gel electrophoresis. Thesample DNA was amplified to enrich for the bacterial 16SV3–V4 rDNA region with a specific primer with a barcode[ 2 4 ] . T h e p r i m e r s e q u e n c e s w e r e 3 4 1 F :C C T A Y G G G R B G C A S C A G a n d 8 0 6 R :GGACTACNNGGGTATCTAAT. Then , the PCR

Probiotics & Antimicro. Prot.

amplification product was recovered and quantified using aQuantiFluorTM fluorometer. The purified amplification prod-ucts were mixed in equal amounts, ligated to sequencingadapters to construct a sequencing library, and sequenced ona HiSeq2500 system using the PE250 run mode. For taxo-nomic classification, tag reads were grouped into operationaltaxonomic units (OTUs) at a sequence similarity level of 97%.The taxonomic classification of the OTUs was based on theannotation of the tags they contained according to the modeprinciple; that is, the taxonomic classification containing morethan 66% of the tags was considered to be the taxonomicclassification of the OTU, and if no such classification existedat one rank, a higher rank would be considered. The taxonom-ic ranks in descending order of size are domain, phylum, class,order, family, genus, and species. The differences in genusbetween the two groups of samples were identified.

SCFA Analyses

At the end of the 16-week intervention period, the rats weredeprived of food for 12 h, weighed, and injected with 2%sodium pentobarbital solution (8mg/kg). The colonic contentswere frozen immediately after collection at the end of treat-ment and were stored at − 80 °C for later analyses. ColonicSCFAs were rapidly determined according to a method previ-ously described by Guohua Zhao [25]. One gram of the co-lonic contents was thawed and suspended in at least 5 mL ofdeionized water and mixed for approximately 5 min, resultingin a 17% (w/w) colonic suspension. The pH of the resultinghomogenate was adjusted to 2–3 by the addition of 5 M HCland then kept at room temperature for 10min with intermittentshaking by a vortex mixer. The homogenate was transferredinto a 2-mL EP tube and centrifuged for 20 min at 5000 rpm,giving a clear supernatant. The internal standard, 2-ethylbutyric acid solution, was spiked into the homogenatesat a final concentration of 1 mm. Three milliliters of ethersolution was added for extraction, and the supernatant wasinjected into a gas chromatograph (GC) for analysis. GC anal-ysis was performed using an Agilent 6890 N GC system withan automatic sampler and a flame ionization detector. TheSCFAs and the internal standard were separated using a fusedsilica capillary column (DB-FFAP 125-3237 J&W Scientific,Agilent Technologies Inc., USA). The standard curves ofacetic acid, propionic acid, i-butyric acid, n-butyric acid, i-valeric acid, and n-valeric acid (Aladdin®) were drawn usingthe six-point standard curve method.

Evaluation of Body Fat and Intrahepatic Fat by CTScanning

Animal microCT was used to quantify the body fat andintrahepatic fat of the rats. The LCT-200 MicroCT automati-cally recognizes fat and muscle; it can measure the muscle

mass, fat volume, average CT value, and fat ratio and canobtain the CT value ratio and the fat ratio of the liver to deter-mine the degree of deterioration of the fatty liver [26].

Biochemical Measurements in Serum and HepaticTissue

Each blood specimen (3–5 mL) was centrifuged at 3000 rpmand 4 °C for 10 min, and the upper layer of serum was col-lected in a 1.5-mL EP tube. The changes in serum lipid levelsand liver function (TC, TG, LDL-C, HDL-C, FFA, ALT, andAST) were determined using an automatic biochemical ana-lyzer. Enzyme-linked immunosorbent assay was used for thequantitative detection of rat TNF-α, IL-1β, and IL-18. Theendpoint chromogenic assay was used to detect the LPS con-tent (No. 160525, Xiamen Bioendo Technology, Co., Ltd.) inthe portal vein serum. Hepatic tissue (0.1 g) was placed inisopropanol (0.9 mL) and homogenized with a TissueLyser-II homogenizer, then centrifuged at 3500 rpm, 4 °C for10 min, and the clear supernatants were collected. The levelsof hepatic TC and TG were determined using an automaticbiochemical analyzer.

Histopathological Observation of Liver Tissue

Hepatic tissue was fixed in formalin and sectioned for stainingwith hematoxylin-eosin (HE) and oil red O. Steatosis andinflammation were identified by light microscopy. Oil RedO Staining Kits (No. 20131212) were made by NanjingJiancheng Science and Technology Company in China.

Western Blotting Analyses

Western blotting was used to detect the changes in proteinexpression of GPR43, NLRP3, ASC, and CASPASE-1 inthe colonic tissues of rats. Colonic tissues were sliced verythinly and thawed in RIPA buffer containing protease inhibi-tors and phosphatase inhibitors. The tissue was furtherdisrupted and homogenized with a dounce homogenizer at4 °C throughout all procedures. The homogenate was thentransferred to microcentrifuge tubes and centrifuged at×10,000g for 10 min at 4 °C. The supernatant was removedand the centrifugation repeated. The total protein was electro-phoretically separated by SDS-PAGE and transferred toPVDF membranes (MILLIPORE, IPVH00010), which werethen blocked with 5% nonfat dried milk (SIGMA) followedby overnight incubation at 4 °C with the following rabbit orgoat polyclonal antibodies: NLRP3 (Novusbio, NBP2-12446,1:100), ASC (Novusbio, NBP1-78977, 1:200), CASPASE-1(Abcam, ab108326, 1:500), GPR43 (Santa Cruz, sc-32903,1:100), and β-actin (KangChen Bio-tech Inc., KC-5A08,1:2000). After incubation with ECLWestern blotting substrate(MILLIPORE, WBKLS0500), the blots were viewed with

Probiotics & Antimicro. Prot.

developing film (Kodak, XBT-1) in darkrooms. The imageswere scanned using aMicrotek scanner, and the grayscale wasanalyzed using the Quantity-One software (BIO-RADCompany).

Statistical Analyses

SPSS forWindows 19.0 was used for the statistical analysis ofdata; GraphPad Prism 6.0 was used to draw diagrams. Theresults were expressed as themean ± standard deviation (mean± SEM). Statistical analysis was performed by one-way anal-ysis of variance (ANOVA) or metastat. P values of < 0.05were considered statistically significant.

Results

Dietary Compound Probiotics Reduced Body Weightand Body Fat

Male Sprague Dawley rats were fed a HFD for 16 weeks toestablish the NAFLDmodel. Compared with the values in theNC group, the body weight (g), subcutaneous fat mass (g),visceral fat mass (g), total fat mass (g), and fat ratio (%) in theHFD group were significantly increased (P< 0.05). In the CPgroup, those indices were all significantly lower than those inthe HFD group (P < 0.05). The results showed that compoundprobiotics significantly reduced HFD-induced weight gainand fat accumulation (Fig. 1, Supplemental Table 5).

Dietary Compound Probiotics Reduced Both LiverWeight and Liver Fat

Compared with the values in the NC group, the liver weight(g), liver index, and liver fat ratio (%) in the HFD group weresignificantly increased (P < 0.05). Compared with the valuesin the HFD group, those indices were lower in the CP group,but the difference was not statistically significant (P> 0.05)(see Fig. 1). Compared with the values in the HFD group,serum TC (mmol/L), TG (mmol/L), LDL (mmol/L), and FFA(mmol/L) in the CP group were decreased by different degrees.In particular, the differences in serum TG and FFA had statis-tical significance (P< 0.05). In addition, the TC and TG con-centrations in the hepatic tissue in the CP group clearly de-creased from 9.11 ± 0.72 and 9.16 ± 1.68 to 6.59 ± 0.56 and4.20 ± 0.96 (P< 0.05), respectively. The results showed thatthere was clear fat accumulation in NAFLD rats, especially inthe liver, while compound probiotics decreased this effect(Fig. 2, Supplemental Table 6). In the HFD group, the liverwas swollen, khaki in color, and less elastic with a strainingcapsule. The liver also exhibited lipopexia with hemorrhagicspots on the surface. In the HFD group, the rats’ hepatic lobuleand cord structure were not clear, and the radial arrangements

of the hepatocytes were not obvious. The hepatocytes wereswollen like balloons with rounded fat vacuoles. Large fat vac-uoles even pressed the nuclei to the side, as shown by HEstaining. In the CP group, the hepatic sinusoid and cord struc-ture were both improved, the changes in cellular morphologywere relieved, and the fat vacuoles were distinctly diminished.In the HFD group, oil red O staining showed diffuse reddish-orange lipid droplets, swollen hepatic cells, and a large numberof reddish-orange lipid droplets inside cells, and the cell nucleiwere pressed to one side. In the CP group, the reddish-orangelipid droplets were clearly decreased, the hepatic cord structurebecame clearer, and the cellular morphology was improved tovarying degrees (Supplemental Fig. 1, Supplemental Table 7).

Dietary Compound Probiotics Ameliorated SerumInflammatory Cytokines

Compared with the values in the NC group, serum LPS (EU/mL), TNF-α (pg/mL), IL-1β (pg/mL), IL-18 (pg/mL), AST(U/L), and ALT (U/L) concentrations in the HFD group wereall significantly increased (P< 0.05). These indices were re-duced to varying degrees in the CP group: TNF-α was a cer-tain decrease, but not significantly different (P = 0.059, HFDvs CP), while LPS, IL-1β, and IL-18 were significantly lowerthan those in the HFD group (P < 0.05 or P < 0.01). Theresults showed that compound probiotics significantly amelio-rated the inflammatory response and hepatic injury in HFD-induced NAFLD rats (Fig. 2, Supplemental Table 6).

Dietary Compound Probiotics Beneficially Modulatedthe Gut Microbiota

First, at the phylum classification stage, three phyla, Firmicutes(58.63%), Bacteroidetes (36.62%), and Proteobacteria(2.66%), made up 97.9% of the total in the NC group feces,holding an absolute advantage in the rat gut microbiota(Supplemental Table 4). The phyla TM7, Actinobacteria,Bacteria, Cyanobacteria, Spirochaetes, Tenericutes,Elusimicrobia, Fusobacteria, Deferribacteres, andVerrucomicrobia accounted for 0.01–0.63% of the total. TheHFD group mainly contained four phyla: Firmicutes (48.8%),Bacteroidetes (39.99%), Verrucomicrobia (5.69%), andProteobacteria (3.67%). Compared with NC group, the propor-tions of TM7 (0.63 vs 0.06%, P< 0.05) in the HFD groupclearly decreased, whereas the proportion of Verrucomicrobia(0.01% vs 5.69%, P< 0.05) clearly increased, which showsthat the HFD had a remarkable effect on the abovementionedtwo phyla. In the CP group, which was administered dietarycompound probiotics for 16 weeks, the TM7 (0.06 vs 1.57%, P< 0.05) and Verrucomicrobia (5.69 vs 2.61%, P< 0.05) phy-lum showed significant changes compared with their propor-tions in the HFD group (Supplemental Fig. 2). It is currentlybelieved that certain gut microbial metabolites are the

Probiotics & Antimicro. Prot.

A B

NC HFD CP

C D

NC HFD CPFig. 1 Compound probiotics reduced body shape and adiposity, liverweight, and fat ratio in NAFLD rats: a Evaluation of body shapes andabdominal adiposity by CT scanning: sky blue, lean tissue; purple,visceral fat; and yellow, subcutaneous fat. b The increased volume ofvisceral, subcutaneous, and total abdominal fat evaluated by CT

scanning. c Evaluation of liver fat ratio by CT scanning. d The levels ofliver weight, liver index, and liver TC and TG. #P< 0.05 NC vs HFD,##P< 0.01 NC vs HFD, *P < 0.05 HFD vs CP, **P< 0.01 HFD vs CP,n = 6–8

Fig. 2 Compound probiotics reduced the levels of serum liver enzymes and lipids (a), LPS, and inflammatory cytokines (b) in NAFLD rats. #P< 0.05NC vs HFD, ##P< 0.01 NC vs HFD, *P< 0.05 HFD vs CP, **P< 0.01 HFD vs CP, n = 8

Probiotics & Antimicro. Prot.

pathogenic factors of NAFLD, especially lipopolysaccharide(LPS). A total of 10 bacterial phyla were observed in all samplesfrom the three groups. Among them, the phyla Bacteroidetes,Proteobacteria , Deferribacteres , Cyanobacteria ,Verrucomicrobia, and Fusobacteria are Gram-negative bacteria;Firmicutes, Tenericutes, Actinobacteria, and TM7 are Gram-positive bacteria. Compared with the NC group, the proportionsof Gram-negative bacteria in the gut of the HFD group showedincreasing trends. In particular, theVerrucomicrobia phylumwassignificantly increased (P< 0.05). However, the proportions ofGram-positive bacteria were all reduced, especially that of theTM7 phylum (P< 0.05). Contrary to our expectations, com-pound probiotics treatment did not cause a significant increasein the abundance of Gram-negative bacteria or a reduction in theratio of Firmicutes to Bacteroidetes in the gut microbiome(Supplemental Fig. 2). Second, in the family classification stage,23 bacterial families including Lactobacillaceae,Clostridiaceae,F16, and Enterobacteriaceae showed obvious differences be-tween the NC group and the HFD group (P< 0.05). In addition,13 bacterial families, including F16, WD2101, andSolirubrobacteraceae, showed statistically significant differ-ences between the HFD group and the CP group (P< 0.05).Third, in the genus classification stage, 34 bacterial genera in-cluding Ruminococcus, Veillonella, Anaeroplasma, andCupriavidus showed statistically significant changes betweenthe NC group and the HFD group (P< 0.05). A total of 27bacterial genera, such as Ruminococcus, showed significantchanges between the HFD group and the CP group (P< 0.05).Furthermore, the relative abundance of genera that showed Pvalues < 0.05 by differential expression analysis (nonparametricANOVA with FDR correction) was expressed as a heat map(Fig. 3) including hierarchical clustering (HCN). Further studiesshowed that the genera Akkermansia, Bacteroides,Bifidobacterium, Prevotella, Ruminococcus, Clostridium,Streptococcus, Phascolarctobacterium, Dailister, Veillonella,Anaerostipes, Coprococcus, Salmonella, Roseburia, andFaecalibacterium are known for their production of SCFAs[22]. Examination of the microbial composition revealed thatcompound probiotics treatment elevated the representation ofBifidobacterium, Ruminococcus, Clostridium, and Anaerostipesin the gut, especially Ruminococcus (0.95 vs 1.83%, P< 0.05).In contrast, Akkermansia, Bacteroides, Prevotella, Veillonella,Coprococcus and Roseburia were decreased in the gut of theCP group, especially Veillonella (0.10 vs 0.03%, P< 0.05),while other known SCFA genera were barely detected.

Colonic SCFAs and GPR43 Receptor Were Involvedin the Effects of Compound Probiotics

Compared with those in the HFD group, the SCFA levels in theCP group all increased. In particular, the acetic acid, propionicacid, and butyric acid levels increased significantly (P< 0.05),but therewas no significant difference in SCFAs between theNC

group and the HFD group (Table 1). TheWestern blotting resultsshowed that compared with those in the NC group, the expres-sion levels of GPR43, NLRP3, ASC, and CASPASE-1 in theHFD group were significantly increased (P< 0.01). Comparedwith those in the HFD group, the expression levels of GPR43,NLRP3, ASC, and CASPASE-1 were downregulated by com-pound probiotics (P< 0.01) (see Fig. 4, Supplemental Table 8).

Discussion

NAFLD is a multisystem disease, often accompanied by in-sulin resistance, obesity, hyperlipidemia, and metabolic syn-drome [7]. Liver lipid accumulation is the primary risk factorof NAFLD. A number of factors affect the accumulation oflipids in the liver, such as fatty acids from foods, FFA redis-tribution, and de novo lipid synthesis [27]. Early clinical andanimal experiments confirmed that probiotics and prebioticscould alleviate HFD-induced NAFLD and its associated lipidmetabolic disorders [28]. This study further found the follow-ing. First, compound probiotics significantly inhibited theHFD-induced weight gain in NAFLD rats, and microCT con-firmed that compound probiotics could inhibit the increase invisceral fat and total fat in NAFLD rats. Therefore, compoundprobiotics may reduce weight gain by inhibiting the increasein body fat. In contrast, serum TC, TG, LDL, FFA, and liverTC and TGwere decreased in the CP group, especially the TCand TG contents in liver tissue and serum FFA. Lipid metab-olism disorder is one of the important mechanisms ofNAFLD. In rats, 80% of TG is hydrolyzed by lipoproteinlipase (LPL) in the chylomicrons. FFA is released and thenabsorbed by peripheral tissues, while 20% is transported to theliver. Donnelly et al. found the lipid accumulation in the liverto be mostly from serum FFA in NAFLD patients [29]. FFAwas one of the causes of lipotoxicity and might promote anLPS-induced NAFLD inflammatory response through theNLR family pyrin domain containing 3 (NLRP3) pathways.Therefore, severe lipid metabolism disorder occurred in theNAFLD rats, and compound probiotics reduced serum TC,TG, LDL, and FFA levels and inhibited the deposition of lipidin the liver to ameliorate fatty liver disease. Third, lipid depo-sition and lipid toxicity could also cause liver parenchymalcell inflammatory response and liver injury. ALT and ASTare the clinical markers used to determine the degree of liverdamage biomarkers [30]. We found that the levels of serumAST and ALTwere decreased after treatment with compound

�Fig. 3 Heat map of bacteria with change between each two groups at thegerms. Hierarchical clustering with a heat map shows the relativeabundance of representative OTUs (i.e., samples with the greatestdifference among the three group means from each family) selected forP < 0.05, obtained with differential expression analysis of the threegroups. The OTUs are shown as phylum, class, order, family, and genus

Probiotics & Antimicro. Prot.

NC1

NC2

NC3

NC4

NC5

NC6

NC7

NC8

HFD1

HFD2

HFD3

HFD4

HFD5

HFD6

HFD7

HFD8

CP1

CP2

CP3

CP4

CP5

CP6

CP7

CP8

Bacteria;Firmicutes;Clostridia;Clostridiales;Veillonellaceae;Dialister

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Salmonella

Bacteria;Synergistetes;Synergistia;Synergistales;Dethiosulfovibrionaceae;Pyramidobacter

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Klebsiella

Bacteria;Firmicutes;Bacilli;Lactobacillales;Aerococcaceae;Facklamia

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Rikenellaceae;Rikenella

Bacteria;Firmicutes;Bacilli;Bacillales;Staphylococcaceae;Macrococcus

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Alcaligenaceae;Oligella

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Corynebacteriaceae;Corynebacterium

Bacteria;Firmicutes;Bacilli;Bacillales;Staphylococcaceae;Jeotgalicoccus

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;Holdemania

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Porphyromonadaceae;Parabacteroides

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Prevotellaceae;Prevotella

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Blautia

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Proteus

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Dermabacteraceae;Dermabacter

Bacteria;Actinobacteria;Coriobacteriia;Coriobacteriales;Coriobacteriaceae;Collinsella

Bacteria;Firmicutes;Bacilli;Lactobacillales;Streptococcaceae;Streptococcus

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;Allobaculum

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Hyphomicrobiaceae;Rhodoplanes

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Dorea

Bacteria;Firmicutes;Clostridia;Clostridiales;Veillonellaceae;Veillonella

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Bacteroidaceae;Bacteroides

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Brenneria

Bacteria;Actinobacteria;Coriobacteriia;Coriobacteriales;Coriobacteriaceae;Coriobacterium

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Pelomonas

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Oxalobacteraceae;Ralstonia

Bacteria;Firmicutes;Clostridia;Clostridiales;Eubacteriaceae;Anaerofustis

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;Coprobacillus

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Paucibacter

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Dermabacteraceae;Brachybacterium

Bacteria;Firmicutes;Bacilli;Lactobacillales;Streptococcaceae;Lactococcus

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micrococcaceae;Kocuria

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Microbacteriaceae;Curtobacterium

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Phyllobacteriaceae;Phyllobacterium

Bacteria;Firmicutes;Clostridia;Clostridiales;Peptococcaceae;rc4−4

Bacteria;Proteobacteria;Gammaproteobacteria;Pasteurellales;Pasteurellaceae;Actinobacillus

Bacteria;Proteobacteria;Gammaproteobacteria;Pasteurellales;Pasteurellaceae;Haemophilus

Bacteria;Firmicutes;Clostridia;Clostridiales;Dehalobacteriaceae;Dehalobacterium

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;cc_115

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Morganella

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Porphyromonadaceae;Dysgonomonas

Bacteria;Firmicutes;Bacilli;Lactobacillales;Lactobacillaceae;Lactobacillus

Bacteria;Fusobacteria;Fusobacteriia;Fusobacteriales;Fusobacteriaceae;Fusobacterium

Bacteria;Firmicutes;Clostridia;Clostridiales;Ruminococcaceae;Ruminococcus

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Providencia

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Anaerostipes

Bacteria;Proteobacteria;Gammaproteobacteria;Pseudomonadales;Pseudomonadaceae;Pseudomonas

Bacteria;Proteobacteria;Gammaproteobacteria;Pseudomonadales;Moraxellaceae;Enhydrobacter

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Actinomycetaceae;Actinomyces

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Actinosynnemataceae;Actinoalloteichus

Bacteria;Proteobacteria;Alphaproteobacteria;Sphingomonadales;Sphingomonadaceae;Sphingobium

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Oxalobacteraceae;Cupriavidus

Bacteria;Proteobacteria;Gammaproteobacteria;Pseudomonadales;Moraxellaceae;Acinetobacter

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Pseudonocardiaceae;Amycolatopsis

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;p−75−a5

Bacteria;Firmicutes;Clostridia;Clostridiales;Christensenellaceae;Christensenella

Bacteria;Firmicutes;Clostridia;Clostridiales;Ruminococcaceae;Faecalibacterium

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Xanthomonas

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;RFN20

Bacteria;Proteobacteria;Deltaproteobacteria;Bdellovibrionales;Bdellovibrionaceae;Bdellovibrio

Bacteria;Firmicutes;Bacilli;Lactobacillales;Enterococcaceae;Vagococcus

Bacteria;Firmicutes;Bacilli;Lactobacillales;Enterococcaceae;Enterococcus

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Limnobacter

Bacteria;Firmicutes;Bacilli;Turicibacterales;Turicibacteraceae;Turicibacter

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Comamonas

Bacteria;Proteobacteria;Gammaproteobacteria;Pseudomonadales;Pseudomonadaceae;Azomonas

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Acidovorax

Bacteria;Proteobacteria;Alphaproteobacteria;Caulobacterales;Caulobacteraceae;Caulobacter

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Trabulsiella

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Stenotrophomonas

Bacteria;Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodobacteraceae;Paracoccus

Bacteria;Firmicutes;Clostridia;Clostridiales;Ruminococcaceae;Anaerofilum

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Methylobacteriaceae;Methylobacterium

Bacteria;Firmicutes;Bacilli;Bacillales;Planococcaceae;Kurthia

Bacteria;Proteobacteria;Alphaproteobacteria;Sphingomonadales;Sphingomonadaceae;Sphingomonas

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Delftia

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Roseateles

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micrococcaceae;Micrococcus

Bacteria;Planctomycetes;Planctomycetia;Gemmatales;Isosphaeraceae;Singulisphaera

Bacteria;Actinobacteria;Coriobacteriia;Coriobacteriales;Coriobacteriaceae;Adlercreutzia

Bacteria;Firmicutes;Clostridia;Clostridiales;Veillonellaceae;Sporomusa

Bacteria;Actinobacteria;Actinobacteria;Bifidobacteriales;Bifidobacteriaceae;Bifidobacterium

Bacteria;Firmicutes;Bacilli;Bacillales;Staphylococcaceae;Staphylococcus

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Coprococcus

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Hyphomicrobiaceae;Devosia

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Rubrivivax

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Streptomycetaceae;Streptomyces

Bacteria;Planctomycetes;Planctomycetia;Pirellulales;Pirellulaceae;Pirellula

Bacteria;Firmicutes;Bacilli;Lactobacillales;Aerococcaceae;Aerococcus

Bacteria;Verrucomicrobia;Verrucomicrobiae;Verrucomicrobiales;Verrucomicrobiaceae;Akkermansia

Bacteria;Proteobacteria;Gammaproteobacteria;Pseudomonadales;Moraxellaceae;Psychrobacter

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Butyrivibrio

Bacteria;Nitrospirae;Nitrospira;Nitrospirales;Nitrospiraceae;Nitrospira

Bacteria;Firmicutes;Erysipelotrichi;Erysipelotrichales;Erysipelotrichaceae;Erysipelothrix

Bacteria;Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodobacteraceae;Amaricoccus

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Hyphomicrobiaceae;Pedomicrobium

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Actinosynnemataceae;Lentzea

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Brevibacteriaceae;Brevibacterium

Bacteria;Firmicutes;Clostridia;Clostridiales;Ruminococcaceae;Oscillospira

Bacteria;Acidobacteria;Acidobacteriia;Acidobacteriales;Koribacteraceae;CandidatusKoribacter

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Erwinia

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Lachnospira

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Moryella

Bacteria;Proteobacteria;Epsilonproteobacteria;Campylobacterales;Helicobacteraceae;Flexispira

Bacteria;Proteobacteria;Betaproteobacteria;Thiobacterales;Thiobacteraceae;Thiobacter

Bacteria;Deferribacteres;Deferribacteres;Deferribacterales;Deferribacteraceae;Mucispirillum

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micromonosporaceae;Virgisporangium

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Rhodobiaceae;Afifella

Bacteria;Firmicutes;Clostridia;Clostridiales;Ruminococcaceae;Anaerotruncus

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Luteibacter

Bacteria;Planctomycetes;Planctomycetia;Gemmatales;Gemmataceae;Gemmata

Bacteria;Proteobacteria;Alphaproteobacteria;Sphingomonadales;Sphingomonadaceae;Kaistobacter

Bacteria;Firmicutes;Clostridia;Clostridiales;Clostridiaceae;02d06

Bacteria;Firmicutes;Clostridia;Clostridiales;Clostridiaceae;CandidatusArthromitus

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Sinobacteraceae;Steroidobacter

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Lysobacter

Bacteria;Proteobacteria;Gammaproteobacteria;Xanthomonadales;Xanthomonadaceae;Thermomonas

Bacteria;Proteobacteria;Gammaproteobacteria;Alteromonadales;Alteromonadaceae;Cellvibrio

Bacteria;Proteobacteria;Deltaproteobacteria;Myxococcales;Myxococcaceae;Corallococcus

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Propionibacteriaceae;Propionibacterium

Bacteria;Chloroflexi;Anaerolineae;Ardenscatenales;Ardenscatenaceae;Ardenscatena

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Microbacteriaceae;Agromyces

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micromonosporaceae;Solwaraspora

Bacteria;Proteobacteria;Deltaproteobacteria;Myxococcales;Nannocystaceae;Nannocystis

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Oxalobacteraceae;Janthinobacterium

Bacteria;Bacteroidetes;Cytophagia;Cytophagales;Cytophagaceae;Adhaeribacter

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Bradyrhizobiaceae;Balneimonas

Bacteria;Firmicutes;Clostridia;Clostridiales;Clostridiaceae;SMB53

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Roseburia

Bacteria;Firmicutes;Clostridia;Clostridiales;Clostridiaceae;Clostridium

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Comamonadaceae;Mitsuaria

Bacteria;Proteobacteria;Deltaproteobacteria;Desulfovibrionales;Desulfovibrionaceae;Bilophila

Bacteria;Firmicutes;Bacilli;Bacillales;Bacillaceae;Bacillus

Bacteria;Fusobacteria;Fusobacteriia;Fusobacteriales;Fusobacteriaceae;Cetobacterium

Bacteria;Proteobacteria;Epsilonproteobacteria;Campylobacterales;Helicobacteraceae;Helicobacter

Bacteria;Actinobacteria;Coriobacteriia;Coriobacteriales;Coriobacteriaceae;Eggerthella

Bacteria;Proteobacteria;Deltaproteobacteria;Desulfovibrionales;Desulfovibrionaceae;Desulfovibrio

Bacteria;Firmicutes;Clostridia;Clostridiales;Lachnospiraceae;Shuttleworthia

Bacteria;Bacteroidetes;Flavobacteriia;Flavobacteriales;Flavobacteriaceae;Flavobacterium

Bacteria;Proteobacteria;Alphaproteobacteria;Sphingomonadales;Erythrobacteraceae;Erythrobacter

Bacteria;Spirochaetes;Spirochaetes;Spirochaetales;Spirochaetaceae;Treponema

Bacteria;Proteobacteria;Betaproteobacteria;Burkholderiales;Alcaligenaceae;Sutterella

Bacteria;Proteobacteria;Gammaproteobacteria;Pasteurellales;Pasteurellaceae;Aggregatibacter

Bacteria;Proteobacteria;Gammaproteobacteria;Pasteurellales;Pasteurellaceae;Bibersteinia

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micrococcaceae;Rothia

Bacteria;Firmicutes;Clostridia;Thermoanaerobacterales;Caldicellulosiruptoraceae;Caldicellulosiruptor

Bacteria;Tenericutes;Mollicutes;Anaeroplasmatales;Anaeroplasmataceae;Anaeroplasma

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Bradyrhizobiaceae;Bosea

Bacteria;Proteobacteria;Alphaproteobacteria;Rhizobiales;Bradyrhizobiaceae;Bradyrhizobium

Bacteria;Proteobacteria;Alphaproteobacteria;Rhodobacterales;Rhodobacteraceae;Rubellimicrobium

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Streptosporangiaceae;Nonomuraea

Bacteria;Planctomycetes;Planctomycetia;Planctomycetales;Planctomycetaceae;Planctomyces

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Nocardioidaceae;Nocardioides

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Mycobacteriaceae;Mycobacterium

Bacteria;Firmicutes;Clostridia;Clostridiales;Veillonellaceae;Phascolarctobacterium

Bacteria;Proteobacteria;Betaproteobacteria;Rhodocyclales;Rhodocyclaceae;Dechloromonas

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Thermomonosporaceae;Actinocorallia

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Micromonosporaceae;Pilimelia

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Intrasporangiaceae;Phycicoccus

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Actinosynnemataceae;Kibdelosporangium

Bacteria;Actinobacteria;Actinobacteria;Actinomycetales;Cellulomonadaceae;Cellulomonas

Bacteria;Bacteroidetes;Cytophagia;Cytophagales;Flammeovirgaceae;Persicobacter

Bacteria;Bacteroidetes;Bacteroidia;Bacteroidales;Rikenellaceae;AF12

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Citrobacter

Bacteria;Proteobacteria;Gammaproteobacteria;Enterobacteriales;Enterobacteriaceae;Gluconacetobacter

−4

−2

0

2

4

Probiotics & Antimicro. Prot.

probiotics for 16weeks, and the level of ALTwas significantlydecreased, which indicated that compound probiotics couldreduce the damage to liver cells and protect the liver function.A large number of studies have confirmed inflammation to bean important pathogenetic process for the progression ofNAFLD into non-alcoholic steatohepatitis (NASH) [31].Intestinal pathogens andmetabolites (e.g., LPS) reach the liverthrough the portal vein and activate the innate immune recep-tors in hepatocytes and Kupffer cells, resulting in the recruit-ment of a large number of inflammatory factors, such asTNF-α, IL-1β, and IL-18, that are involved in the formationof NAFLD. This study also found that serum LPS levels weresignificantly increased in NAFLD rats, whereas compoundprobiotics reduced serum LPS and the liver inflammatory re-sponse. In general, compound probiotics significantly reducedthe visceral and total fat, dyslipidemia, hepatic TC and TG,and serumALT, LPS, IL-1β, and IL-18. Compound probioticsclearly alleviated the progress of NAFLD disease, but theexact mechanism still needs further study.

After 16 weeks of HFD feeding, the imbalance of the gutmicrobiota in NAFLD rats was manifested primarily in the

abundance of two phyla. Compared with those in the NCgroup, the proportions of TM7 in the HFD group clearly de-creased, and the proportion of Verrucomicrobia clearly in-creased. The TM7 and Verrucomicrobia phyla in the rat fecesalso showed significant differences in the CP group. The TM7phylum was a mysterious type of bacteria, and He X et al.found for the first time that TM7x could inhibit the productionof TNF-α in macrophages, suggesting that it has a potentialimmunosuppressive effect [32]. Therefore, TM7 was likely tobe associated with the pathogenesis of inflammation inNAFLD. In this study, we found that the relative abundanceof TM7 cells in the CP group was significantly higher than thatin the HFD group, and the serum concentration of TNF-αwasdecreased to some degree. Therefore, compound probioticsmay repress the inflammatory response in NAFLD throughthe regulation of the TM7 phylum. A large number of studieshave found that the production, degradation, and regulation ofmetabolites of the gut microbiota may be a key link betweenthe host and the gut microbiota that participates in regulatingthe immune and metabolic processes [33]. At present, someintestinal microbial metabolites are considered risk factors inNAFLD. For example, LPS, also known as endotoxin, is amajor component of the cell walls of Gram-negative bacteria.[34] LPS mainly activates the innate immune receptor TLR4,which causes serious inflammatory liver injury in NAFLD. Atthe family level of the microbiota, we found that approximate-ly 13 strains showed significant decreases (P < 0.05) after16 weeks of compound probiotics intervention in NAFLDrats. Among them, the abundance of the Prevotellaceae familydecreased by more than 50% (14.33 vs 6.97%). Recent obser-vations show that the Prevotellaceae family was clearly rich inobese patients. Prevotellaceae is one of a family in the Gram-negative Bacteroidetes phylum and was a good source of LPS[35]. In this study, it was found that compound probiotics

Table 1 Colonic SCFAs in SD rats fed a normal control diet or an HFdiet containing compound probiotics for 16 weeks

SCFAs (mM) NC HFD CP

Acetic acid 0.100 ± 0.036 0.158 ± 0.062 0.299 ± 0.121*

Propionic acid 0.056 ± 0.016 0.056 ± 0.008 0.096 ± 0.025*

i-Butyric acid 0.029 ± 0.010 0.030 ± 0.011 0.037 ± 0.014

n-Butyric acid 0.187 ± 0.066 0.136 ± 0.032 0.359 ± 0.145*

i-Valeric acid 0.063 ± 0.017 0.052 ± 0.010 0.069 ± 0.045

n-Valeric acid 0.063 ± 0.011 0.062 ± 0.016 0.092 ± 0.044

* Significant difference from HFD group rats for colonic SCFAs usingANOVA as described in the Methods. P < 0.05 HFD vs CP, n = 8

NC HFD CP

GPR43

β-actin

NC HFD CP

NLRP3

β-actin

NC HFD CP

ASC

β-actin

NC HFD CP

Caspase-1

β-actin

Fig. 4 Compound probiotics reduced the expression of GPR43, NLRP3,ASC, andCASPASE-1 proteins inNAFLD rats. These proteins were quantifiedvia Western blotting in the colon tissue. #P< 0.05 NC vs HFD, ##P< 0.01 NC vs HFD, *P< 0.05 HFD vs CP, **P < 0.01 HFD vs CP, n = 4

Probiotics & Antimicro. Prot.

reduced the abundance of Prevotellaceae family and de-creased the concentration of LPS in the portal vein, whichmay be one of the intestinal mechanisms by which compoundprobiotics alleviate metabolic endotoxemia in NAFLD rats.

Current evidence suggests that high levels of dietary fiberconsumption and associated high concentrations of intestinalSCFAs are beneficial to host health [19–21]. SCFAs are in-volved in weight regulation, maintaining intestinal homeosta-sis, and increasing the glucose and lipid metabolism, therebyaffecting many diseases, such as obesity and NAFLD [36, 37].Obesity is the most important risk factor for NAFLD [38].Some research has shown that modulation of the gut microbi-ota alone could lead to obesity due to the role of SCFAs [39,40]. Ley RE et al. found that a prolonged HFD could causepersistent changes in the intestinal microecology, which couldbe manifested as a significant reduction in the diversity of thegut microbiota and in the abundance of SCFAs [41]. Furtherstudies showed that the production, degradation, and regula-tion of gut metabolites might also be involved in regulatingthe immune and metabolic processes [33]. As the main prod-ucts of bacterial fermentation of undigested dietary fiber (e.g.,plant polysaccharides) and resistant starch in the colon,SCFAs, mainly including acetic acid, propionic acid, i-butyric acid, n-butyric acid, i-valeric acid, and n-valeric acid,may ameliorate the intestinal barrier function through aglucagon-like peptide 2 (GLP-2)-mediated intestinotrophic ef-fect or through the GPR43-mediated suppression of obesity[42]. Our results showed that compound probiotics signifi-cantly increased the levels of intestinal SCFAs in NAFLD rats,especially those of acetic acid, propionic acid, and butyric acid(P < 0.05). In addition, at the microbial genus level, we foundthat the abundance of Veillonella in the gut microbiota of theHFD group was significantly increased, but Ruminococcuswas significantly decreased. Ruminococcus and Veillonellaare also considered to produce SCFAs [43]. The differencein the Ruminococcus genus arose mainly from theRuminococcaceae family. The abundance of theRuminococcaceae family in the fecal samples of NAFLDand NASH patients is lower than that of healthy subjects[44]. A study confirmed that butyric acid or butyrate-producing bacterium significantly inhibited the progressionof NAFLD by activating the AMPK signaling pathway aswell as regulating lipid and energy metabolism, oxidativestress, and insulin sensitivity [45]. Other study found thatweight loss and improved glucose tolerance were found inobese mice fed with acetic acid [46]. Shulman GI et al. showthat increased production of acetate by an altered gut micro-biota in rodents leads to activation of the parasympatheticnervous system, which, in turn, promotes increased glucose-stimulated insulin secretion, increased ghrelin secretion, hy-perphagia, obesity, and related sequelae [47]. Propionic acidhas also been found to significantly inhibit the synthesis ofcholesterol and fatty acids in the liver of mice [48]. Our study

focused on the relationship between chronic metabolic inflam-mation of the intestine and NAFLD. In the next experiment,the effect of compound probiotics on liver enzymes will becarefully investigated. Therefore, compound probiotics canregulate intestinal SCFAs, but the different SCFA changesare inconsistent. The content and the ratio of the SCFAs maydepend on differences in the gut microbiota composition andother targets.

SCFAs can directly activate G protein-coupled receptors(GPCRs), especially GPR43, GPR41, and GPR109A [22].We found that a HFD can significantly increase in GPR43protein expression in colon tissues, consistent with the find-ings of LaurenM. Cornall et al. [49]. The Lu Y research groupfound the expression of GPR43 and GPR41 in colon tissue tobe increased after a 16-week HFD, leading to the secretion ofPYY and an increase in GLP-1, which might be due to thereduction in body weight caused by the inhibition of foodintake [42]. In addition, we further demonstrated that the ex-pression of GPR43 protein in the colon tissue of the CP groupwas significantly decreased. A recent study showed that high-fiber diets can activate GPR43 on intestinal colonic epithelialcells, which maintains epithelial integrity via activation of theNLRP3 inflammasome, a critical process for maintaining in-testine homeostasis [50]. The expression of NLRP3, ASC, andCASPASE-1 in colonic tissues was significantly higher thanthat in the NC group (P < 0.01). A study found that NLRP3inflammasomes (including NLRP3, ASC, and CASPASE-1)and the effector protein IL-18 negatively regulate NAFLD/NASH progression, as well as multiple aspects of metabolicsyndrome via modulation of the gut microbiota [51].Although the results remain controversial, these findings sug-gested that GPR43 and NLRP3 might play an important rolein the identification and metabolic regulation of gut microbi-ota. Some research showed that probiotics may directly orindirectly ameliorate NAFLD by regulating the structure ofgut microbiota and its metabolites. One of the most importantmetabolites was SCFAs. Intestinal mucosal epithelium wasone of the main sites where butyric acid works. It was foundthat SCFAs, including butyric acid, not only provide energyfor intestinal mucosal cells, but also inhibit the production ofinflammatory cytokines and reduce the occurrence of chronicmetabolic inflammation through the GPR43 pathway. Therole of other SCFAs, such as acetic acid and propionic acid,may directly affect the liver, adipose tissue, or central nervoussystem [52], which depends on our further study. Therefore,given all the above results, compound probiotics may regulatethe integrity of the gut microbiota and of the intestinal epithe-lial mucosa through SCFAs and their receptor GPR43 inNAFLD rats [50, 51].

In summary, this study found that a HFD induced partialreconstruction of the gut microbiota. The above changes in thegut microbiota structure may include increased intestinal per-meability, metabolic imbalance, and innate immune system

Probiotics & Antimicro. Prot.

activation in NAFLD rats. The gut microbiota is closely relat-ed to inflammatory response, and chronic metabolic inflam-mation may lead to obesity, T2DM and NAFLD. Moreover,compound probiotics reduce the deposition of lipids and alle-viate chronic metabolic inflammation. Compound probioticsmight regulate the gut microbiota through SCFAs and theirreceptor GPR43, which probably accounts for the protectiveeffect of compound probiotics in HFD-induced NAFLD rats.

Acknowledgments We thank Pinghe Yin, Yang Hu, and Huanyong Lifrom Jinan University Analytical and Testing Center for technical assis-tance and Guangzhou Genedenovo Biotechnology Co. Ltd. for providingthe methods for partial bioinformatics analysis.

Funding This work was supported in part by the National NaturalScience Foundation of China (no. 81774165, 81573844), the NaturalScience Foundation of Guangdong in China (no. 2016A030313824),Traditional Chinese Medicine Bureau of Guangdong in China (no.20161065), and the National Health and Family Planning Commissionof Guangdong in China (no. A2016583 and A2017228).

Compliance with Ethical Standards Animal experiments were approvedby the animal care and use committee of Jinan University and were incompliance with the guidelines for the BCare and Use of Animals.^

Conflict of Interests The authors declare that they have no conflict ofinterest.

References

1. Wree A, Broderick L, Canbay A, Hoffman HM, Feldstein AE(2013) From NAFLD to NASH to cirrhosis-new insights into dis-ease mechanisms. Nat Rev Gastroenterol Hepatol 10(11):627–636.https://doi.org/10.1038/nrgastro.2013.149

2. Rinella M, Charlton M (2016) The globalization of non-alcoholicfatty liver disease-prevalence and impact on world health.Hepatology 64(1):4–22. https://doi.org/10.1002/hep.28524

3. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, WymerM (2016) Global epidemiology of nonalcoholic fatty liver disease-meta-analytic assessment of prevalence, incidence, and outcomes.Hepatology 64(1):73–84. https://doi.org/10.1002/hep.28431

4. Loomba R, Sanyal AJ (2013) The global NAFLD epidemic. NatRev Gastroenterol Hepatol 10(11):686–690. https://doi.org/10.1038/nrgastro.2013.171

5. Browning JD, Szczepaniak LS, Dobbins R, Nuremberg P, HortonJD, Cohen JC, Grundy SM, Hobbs HH (2004) Prevalence of he-patic steatosis in an urban population in the United States: impact ofethnicity. Hepatology 40(6):1387–1395. https://doi.org/10.1002/hep.20466

6. Farrell GC, Wong VW, Chitturi S (2013) NAFLD in Asia—ascommon and important as in the west. Nat Rev GastroenterolHepatol 10(5):307–318. https://doi.org/10.1038/nrgastro.2013.34

7. Byrne CD, Targher G (2015) NAFLD: a multisystem disease. JHepatol 62:S47–S64. https://doi.org/10.1016/j.jhep.2014.12.012

8. Williams CD, Stengel J, Asike MI, Torres DM, Shaw J, ContrerasM, Landt CL, Harrison SA (2011) Prevalence of nonalcoholic fattyliver disease and nonalcoholic steatohepatitis among a largelymiddle-aged population utilizing ultrasound and liver biopsy: aprospective study. Gastroenterology 140(1):124–131. https://doi.org/10.1053/j.gastro.2010.09.038

9. Loomba R, AbrahamM, Unalp A,Wilson L, Lavine J, Doo E, BassNM (2012) Association between diabetes, family history of diabe-tes, and risk of nonalcoholic steatohepatitis and fibrosis.Hepatology 56(3):943–951. https://doi.org/10.1002/hep.25772

10. Musso G, Gambino R, Cassader M, Pagano G (2011) Meta-analy-sis: natural history of non-alcoholic fatty liver disease (NAFLD)and diagnostic accuracy of non-invasive tests for liver disease se-verity. Ann Med 43(8):617–649. https://doi.org/10.3109/07853890.2010.518623

11. Musso G, Gambino R, Tabibian JH, Ekstedt M, Kechagias S,Hamaguchi M, Hultcrantz R, Hagstrom H, Yoon SK,Charatcharoenwitthaya P, George J, Barrera F, Hafliethadottir S,Bjornsson ES, Armstrong MJ, Hopkins LJ, Gao X, Francque S,Verrijken A, Yilmaz Y, Lindor KD, Charlton M, Haring R, LerchMM, Rettig R, Volzke H, Ryu S, Li G, Wong LL, Machado M,Cortez-Pinto H, Yasui K, Cassader M (2014) Association of non-alcoholic fatty liver disease with chronic kidney disease: a system-atic review and meta-analysis. PLoS Med 11(7):e1001680. https://doi.org/10.1371/journal.pmed.1001680

12. Tilg H, Moschen AR (2010) Evolution of inflammation in nonal-coholic fatty liver disease: the multiple parallel hits hypothesis.Hepatology 52(5):1836–1846. https://doi.org/10.1002/hep.24001

13. Bechmann LP, Hannivoort RA, Gerken G, Hotamisligil GS,Trauner M, Canbay A (2012) The interaction of hepatic lipid andglucose metabolism in liver diseases. J Hepatol 56(4):952–964.https://doi.org/10.1016/j.jhep.2011.08.025

14. Wree A, Kahraman A, Gerken G, Canbay A (2011) Obesity affectsthe liver—the link between adipocytes and hepatocytes. Digestion83(1-2):124–133. https://doi.org/10.1159/000318741

15. Rawls JF, Mahowald MA, Ley RE, Gordon JI (2006) Reciprocalgut microbiota transplants from zebrafish and mice to germ-freerecipients reveal host habitat selection. Cell 127(2):423–433.https://doi.org/10.1016/j.cell.2006.08.043

16. Turnbaugh PJ, Hamady M, Yatsunenko T, Cantarel BL, Duncan A,Ley RE, Sogin ML, Jones WJ, Roe BA, Affourtit JP, Egholm M,Henrissat B, Heath AC, Knight R, Gordon JI (2009) A core gutmicrobiome in obese and lean twins. Nature 457(7228):480–484.https://doi.org/10.1038/nature07540

17. Yang LH, Cai J, Chen DF (2012) Alteration and significance ofintestinal flora in patients with NASH. J Clin Hepatol 28:124–126

18. Spencer MD, Hamp TJ, Reid RW, Fischer LM, Zeisel SH, FodorAA (2011) Association between composition of the human gastro-intestinal microbiome and development of fatty liver with cholinedeficiency. Gastroenterology 140(3):976–986. https://doi.org/10.1053/j.gastro.2010.11.049

19. Liang Y, Lin C, Wang S, Z Yupei, H Li, Wang G, Li Y, Deng Y, HeY, He Q, C Xiaoshuang, Y Qinhe, New targets for prevention andtreatment of obesity related NAFLD: short-chain fatty acids andshort-chain fatty acid receptors signaling pathways, Journal ofChongqing Medical University, (2016) 628–631

20. Thorburn AN, Macia L, Mackay CR (2014) Diet, metabolites, andBwestern-lifestyle^ inflammatory diseases. Immunity 40(6):833–842. https://doi.org/10.1016/j.immuni.2014.05.014

21. Tan J, McKenzie C, Potamitis M, Thorburn AN, Mackay CR,Macia L (2014) The role of short-chain fatty acids in health anddisease. Adv Immunol 121:91–119. https://doi.org/10.1016/B978-0-12-800100-4.00003-9

22. Koh A, De Vadder F, Kovatcheva-Datchary P, Backhed F (2016)From dietary fiber to host physiology: short-chain fatty acids as keybacterial metabolites. Cell 165(6):1332–1345. https://doi.org/10.1016/j.cell.2016.05.041

23. Wang Y, Xie J, Li Y, Dong S, Liu H, Chen J, Wang Y, Zhao S,Zhang Y, Zhang H (2016) Probiotic lactobacillus casei Zhang re-duces pro-inflammatory cytokine production and hepatic inflamma-tion in a rat model of acute liver failure. Eur J Nutr 55(2):821–831.https://doi.org/10.1007/s00394-015-0904-3

Probiotics & Antimicro. Prot.

24. Siddiqui H, Nederbragt AJ, Lagesen K, Jeansson SL, Jakobsen KS(2011) Assessing diversity of the female urine microbiota by highthroughput sequencing of 16S rDNA amplicons. BMC Microbiol11(1):244. https://doi.org/10.1186/1471-2180-11-244

25. Zhao G, Nyman M, Jonsson JA (2006) Rapid determination ofshort-chain fatty acids in colonic contents and faeces of humansand rats by acidified water-extraction and direct-injection gas chro-matography. Biomed Chromatogr 20(8):674–682. https://doi.org/10.1002/bmc.580

26. Ogasawara M, Hirose A, Ono M, Aritake K, Nozaki Y, TakahashiM, Okamoto N, Sakamoto S, Iwasaki S, Asanuma T, Taniguchi T,Urade Y, Onishi S, Saibara T, Oben JA (2011) A novel and com-prehensive mouse model of human non-alcoholic steatohepatitiswith the full range of dysmetabolic and histological abnormalitiesinduced by gold thioglucose and a high-fat diet. Liver Int 31(4):542–551. https://doi.org/10.1111/j.1478-3231.2010.02443.x

27. Perla FM, Prelati M, Lavorato M, Visicchio D, Anania C (2017)The role of lipid and lipoprotein metabolism in non-alcoholic fattyliver disease. Children (Basel) 4(6):1–14. https://doi.org/10.3390/children4060046

28. Shapiro H, Suez J, Elinav E (2017) Personalized microbiome-basedapproaches to metabolic syndrome management and prevention. JDiabetes 9(3):226–236. https://doi.org/10.1111/1753-0407.12501

29. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD,Parks EJ (2005) Sources of fatty acids stored in liver and secretedvia lipoproteins in patients with nonalcoholic fatty liver disease. JClin Invest 115(5):1343–1351. https://doi.org/10.1172/jci23621

30. Harrison SA, Oliver D, Arnold HL, Gogia S, Neuschwander-TetriBA (2008) Development and validation of a simple NAFLD clin-ical scoring system for identifying patients without advanced dis-ease. Gut 57(10):1441–1447. https://doi.org/10.1136/gut.2007.146019

31. Poeta M, Pierri L, Vajro P (2017) Gut-liver axis derangement innon-alcoholic fatty liver disease. Children (Basel) 4(8):1–19.https://doi.org/10.3390/children4080066

32. He X, McLean JS, Edlund A, Yooseph S, Hall AP, Liu SY,Dorrestein PC, Esquenazi E, Hunter RC, Cheng G, Nelson KE,Lux R, Shi W (2015) Cultivation of a human-associated TM7 phy-lotype reveals a reduced genome and epibiotic parasitic lifestyle.Proc Natl Acad Sci U SA 112(1):244–249. https://doi.org/10.1073/pnas.1419038112

33. Rooks MG, Garrett WS (2016) Gut microbiota, metabolites andhost immunity. Nat Rev Immunol 16(6):341–352. https://doi.org/10.1038/nri.2016.42

34. Porras D, Nistal E, Martinez-Florez S, Pisonero-Vaquero S, OlcozJL, Jover R, Gonzalez-Gallego J, Garcia-Mediavilla MV, Sanchez-Campos S (2017) Protective effect of quercetin on high-fat diet-induced non-alcoholic fatty liver disease in mice is mediated bymodulating intestinal microbiota imbalance and related gut-liveraxis activation. Free Radic Biol Med 102:188–202. https://doi.org/10.1016/j.freeradbiomed.2016.11.037

35. Graham C, Mullen A, Whelan K (2015) Obesity and the gastroin-testinal microbiota: a review of associations and mechanisms. NutrRev 73(6):376–385. https://doi.org/10.1093/nutrit/nuv004

36. De Vadder F, Kovatcheva-Datchary P, Goncalves D, Vinera J,Zitoun C, Duchampt A, Backhed F, Mithieux G (2014)Microbiota-generated metabolites promote metabolic benefits viagut-brain neural circuits. Cell 156(1-2):84–96. https://doi.org/10.1016/j.cell.2013.12.016

37. Puertollano E, Kolida S, Yaqoob P (2014) Biological significanceof short-chain fatty acid metabolism by the intestinal microbiome.Curr Opin Clin Nutr Metab Care 17(2):139–144. https://doi.org/10.1097/mco.0000000000000025

38. Fabbrini E, Sullivan S, Klein S (2010) Obesity and nonalcoholicfatty liver disease: biochemical, metabolic, and clinical implica-tions. Hepatology 51(2):679–689. https://doi.org/10.1002/hep.23280

39. Zhu L, Baker RD, Baker SS (2014) Gut microbiome and nonalco-holic fatty liver diseases. Pediatr Res 77(1-2):245–251. https://doi.org/10.1038/pr.2014.157

40. Endo H, Niioka M, Kobayashi N, Tanaka M, Watanabe T (2013)Butyrate-producing probiotics reduce nonalcoholic fatty liver dis-ease progression in rats: new insight into the probiotics for the gut-liver axis. PLoS One 8(5):e63388. https://doi.org/10.1371/journal.pone.0063388

41. Ley RE, Turnbaugh PJ, Klein S, Gordon JI (2006) Microbial ecol-ogy: human gut microbes associated with obesity. Nature444(7122):1022–1023. https://doi.org/10.1038/4441022a

42. Lu Y, Fan C, Li P, Lu Y, Chang X, Qi K (2016) Short chain fattyacids prevent high-fat-diet-induced obesity in mice by regulating Gprotein-coupled receptors and gut microbiota. Sci Rep 6(1):37589.https://doi.org/10.1038/srep37589

43. Louis P, Hold GL, Flint HJ (2014) The gut microbiota, bacterialmetabolites and colorectal cancer. Nat Rev Microbiol 12(10):661–672. https://doi.org/10.1038/nrmicro3344

44. Zhu L, Baker SS, Gill C, Liu W, Alkhouri R, Baker RD, Gill SR(2013) Characterization of gut microbiomes in nonalcoholicsteatohepatitis (NASH) patients: a connection between endogenousalcohol and NASH. Hepatology 57(2):601–609. https://doi.org/10.1002/hep.26093

45. Tanaka N, Matsubara T, Krausz KW, Patterson AD, Gonzalez FJ(2012) Disruption of phospholipid and bile acid homeostasis inmice with nonalcoholic steatohepatitis. Hepatology 56(1):118–129. https://doi.org/10.1002/hep.25630

46. Yamashita H, Fujisawa K, Ito E, Idei S, Kawaguchi N, Kimoto M,Hiemori M, Tsuji H (2007) Improvement of obesity and glucosetolerance by acetate in Type 2 diabetic Otsuka Long-EvansTokushima Fatty (OLETF) rats. Biosci Biotechnol Biochem 71:1236–1243

47. Perry RJ, Peng L, Barry NA, Cline GW, Zhang D, Cardone RL,Petersen KF, Kibbey RG, Goodman AL, Shulman GI (2016)Acetate mediates a microbiome-brain-beta-cell axis to promotemetabolic syndrome. Nature 534(7606):213–217. https://doi.org/10.1038/nature18309

48. Wright RS, Anderson JW, Bridges SR (1990) Propionate inhibitshepatocyte lipid synthesis. Proc Soc Exp Biol Med 195:26–29

49. Cornall LM, Mathai ML, Hryciw DH, McAinch AJ (2011) Diet-induced obesity up-regulates the abundance of GPR43 andGPR120 in a tissue specific manner. Cell Physiol Biochem 28(5):949–958. https://doi.org/10.1159/000335820

50. Macia L, Tan J, Vieira AT, Leach K, Stanley D, Luong S, MaruyaM, Ian McKenzie C, Hijikata A, Wong C, Binge L, Thorburn AN,Chevalier N, Ang C, Marino E, Robert R, Offermanns S, TeixeiraMM, Moore RJ, Flavell RA, Fagarasan S, Mackay CR (2015)Metabolite-sensing receptors GPR43 and GPR109A facilitate die-tary fibre-induced gut homeostasis through regulation of theinflammasome. Nat Commun 6:6734. https://doi.org/10.1038/ncomms7734

51. Henao-Mejia J, Elinav E, Jin C, Hao L, Mehal WZ, Strowig T,Thaiss CA, Kau AL, Eisenbarth SC, Jurczak MJ, Camporez JP,Shulman GI, Gordon JI, Hoffman HM, Flavell RA (2012)Inflammasome-mediated dysbiosis regulates progression ofNAFLD and obesity. Nature 482:179–185. https://doi.org/10.1038/nature10809

52. He M, Shi B (2017) Gut microbiota as a potential target of meta-bolic syndrome: the role of probiotics and prebiotics. Cell Biosci7(1):54. https://doi.org/10.1186/s13578-017-0183-1

Probiotics & Antimicro. Prot.