Kshif Hussin, Yingying Yng, Ji Wng, Hngjun Bin, Xi Li,Junji Chn, Qinying Li, Li Wng, Qingping Zhong, Xing Fng,Yuto Wng, Hong Wi, Yigng Tong, Zhnlin Lio,,*

a College of Food Science, South China Agricultural University, Guangzhou 510642, China

b Department of Food Science, University of Guelph, Guelph, Ontario N1G 2W1, Canada

c College of Life and Geographic Sciences, Key Laboratory of Biological Resources and Ecology of Pamirs Plateau in Xinjiang Uygur Autonomous Region,Kashi University, Kashi 844000, China

d Institute of Precision Medicine, First Affiliated Hospital of Sun Yat-Sen University, Guangzhou 510080, China

e College of Life Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China

Keywords:

High fat diet

Pseudo germ-free mice

Tea polyphenol

Obesity

Lipid metabolism

A B S T R A C T

The role of gut microbiota in terms of host health is becoming increasingly important.In this study, the comparative effects of tea polyphenols (TPs) on weight loss and lipid metabolism on conventionalized mice(CVZ) and pseudo germ-free (PGF) mice (treated with antibiotics) were investigated.Our findings revealed that high fat (HF) diet considerably increased the body weight, total fat and upsurge lipid indices in CVZ mice but PGF mice were not sensitive to the effect of HF diet as CVZ mice.After the dietary administration of TP,body weight, perirenal fat and epididymal fat, liver weight, glucose (GLU) level, total chloestrol (TC level),high density lipoprotein-cholesterol (HDL-C) level significantly lowered in PGF mice as compared to CVZ mice group.However, the area of fat cells and triglyceride (TG) level were significantly increased in PGF mice.In CVZ mice, TP intervention resulted in a considerable drop in the Firmicutes/Bacteroides ratio as compared to PGF mice.The intestinal flora of PGF mice was severely reduced after antibiotic treatment, while TP administration restored intestinal diversity; the abundance of Akkermansia and Lactobacillus increased,whereas the abundance of Enterobacteriaceae and Prevotella reduced.Overall, we can assume that PGF obese mice administered with TP have less anti-obesity effects compared to obese CVZ mice.

1.Introduction

The gut microbiota and human body have an inseparable and mutually beneficial symbiotic relationship; it participates in many metabolic processes such as nutrient absorption, energy utilization and demonstrates immunomodulatory effects [1].The modulation of the gut microbiota caused by the improper diet pattern is a critical factor in developing many gastrointestinal diseases [2].Obesity is currently the most common nutritional disorder in developed and developing countries due to genetic and environmental factors playing a significant role in the onset of obesity [3].The onset of obesity is linked with the diversity and function of gut microbiota in the body [4].Gut microbiome has a significant role in energy harvesting from the dietary substances and also assist in the accumulation of these calories in adipose tissues.Metabolism is affected by the dietary intervention which results in the manipulation of host gut microbiota [5,6].The potential cause which triggers obesity is due to the disturbance of gut microbiota and alteration in intestinal permeability.Studies have confirmed that animals without gut microbes have better glucose tolerance and insulin sensitivity, as well as a low-fat and low-weight phenotype [7].

Pseudo germ-free (PGF) mouse animal models are widely used to study the relationship between the gut microbiome and host health,and it is obtained by using antibiotic mixtures to remove the majority of intestinal microorganisms in wild-type mice.Typical combinations of antibiotic mixtures are neomycin, streptomycin [8]; ampicillin,neomycin [9]; vancomycin, neomycin, metronidazole, ampicillin [10,11];where ampicillin and neomycin antibiotic combination can eliminate 90% of the microbes in the intestine [9].The appearance,developmental status, organ toxicity and physiological changes of PGF mice are not significantly different from those of conventionalized mice.PGF mouse models are economical, convenient and easy to obtain,and the research results using PGF mouse models are often consistent with the research results of germ-free mouse models.Therefore, the PGF mouse model is frequently used to predict the research results of the germ-free mouse model [12,13].

Tea polyphenols (TPs) are essential functional compounds in tea.The primary green TPs are catechin (C), epicatechin (EC),gallocatechin (GC), epicatechin gallate (ECG), epigallocatechin(EGC), and epigallocatechin gallate (EGCG) [14].The EGCG portion of TPs has an anti-obesity effect in obese mice models by targeting adipose tissues directly [15,16].In clinical [17,18]and animal model experiments [19]consumption of green tea resulted in weight reduction.Green TP extracts significantly lowered total visceral and liver fat weight [20].Furthermore, green TP considerably improves total serum cholesterol, triglycerides (TG), and low-density lipoprotein cholesterol (LDL-C) content [21].Green tea intervention increased the intestinal abundance ofAkkermansia,Bifidobacteria,andLactobacillusspp., returned the Firmicutes/Bacteroidetes ratio,and improved thePrevotella/Bacteroidesratios [22].Seo et al.[23]studied the effect of fermented green tea extract on obese mice gut micro flora and deduced that its intervention for 8 weeks might restore gut microbiota disrupted by HF diet.

Although TPs has been proven productive for weight loss and reducing body fats [24], increasing insulin sensitivity [25], lowering serum cholesterol [26], and inhibiting inflammation level [27].Yet only a few researches have been conducted to demonstrate the anti-obesity effect of TPs in the absence of intestinal flora.It is well-known that the absorption rate and bioavailability of TPs in human body is very low, and maximum of tea polyphenol reach to gastrointestinal tract where gut microbiome exist [28,29].In this study, we tried to investigate the relationship between tea polyphenol,gut microflora and obesity.Therefore, we developed PGF mice by administrating antibiotics to conventionalized mice and compared the TP’s impact (low and high doses) on CVZ and PGF mice body weight, adipose tissue physiological indexes, biomarkers for lipid metabolism and gut microbiota composition.

2.Materials and methods

2.1 Experimental animals and feed

Eighty 8-week-old healthy CVZ-class C57BL/6J male mice were provided by the Experimental Zoology Department of the Third Military Medical University (TMMU), Chongqing, China.CVZ class C57BL/6J Mouse License Number: SYXK (Chongqing) 2012-0002.AIN93 standard MD12031 conventional diet (CK) feed and MD12032 high-fat (HF) purified feed were obtained from Jiangsu Medison Biopharmaceutical Co., Ltd.(Table 1).TPs were delivered by Guangzhou Pubo Biological Co., Ltd., China.The defined composition of TP is given in (Table 2).

Table 1Composition of diets in the study.

Table 2Composition of green tea polyphenols.

2.2 Model construction and grouping of PGF and CVZ grade mice

For the construction of pseudo germ-free mouse model, CVZ C57BL/6J male mice (n= 40) were administrated a combination of antibiotics – vancomycin (200 mg/kg), neomycin (200 mg/kg),metronidazole (200 mg/kg) and ampicillin (200 mg/kg) as described by previous research [30,31].Briefly, the antibiotics mixtures were added into drinking water and fed on every 3 days during whole 8 weeks to establish pseudo germ-free mice model.These mice were then randomly categorized into 4 groups, conventional diet (PGF-CK)group, high-fat diet (PGF-HF) group, PGF HF + 0.8% TP group, and PGF HF + 1.6% TP group, each group containing 10 mice.Meanwhile,CVZ C57BL/6J male mice (n= 40) were also randomly alienated into 4 similar groups used as control group.The model design and grouping of both CVZ and PGF mice are shown in Fig.1.

Fig.1 Model construction and grouping of CVZ and PGF mice.

The study was approved and reviewed by TMMUs Ethics Committee.All groups of mice were housed in a controlled environment (the temperature at 22 °C to 24 °C) and a light period of 12-h in sterile insulators.All feeds were sanitized with 60 kGy60Co gamma rays whereas drinking water, cages, and bedding were autoclaved for 45 min at 121 °C, 110 kPa.The mice were allowed free access to eat and drink, and the cage, water, feed, and litter changed twice a week.From 0 to 8thweek of experiment, the feces were collected in sterile 1.5 mL Eppendorf (EP) tubes and immediately frozen at –80 °C.

2.3 Measurement of CVZ and PGF mice body weight index

During the experiment, the health status of each mouse was observed regularly.The CVZ and PGF mice’s weight was recorded weekly.

2.4 Measurement of fasting blood glucose and lipid indexes

After 8 weeks of regular drinking, the mice were fasted for 12 h.The mice’s tails were wiped with a 75% alcohol cotton ball, and the tail tips of about 1 mm in length were cut with sterile scissors to measure their fasting blood glucose using Steady Blood Glucose Meter (OneTouch?, USA).Mice were then anesthetized by using ethyl ether and sacrificed.The serum samples were collected with a 1.5 mL sterile EP tube and stored in the refrigerator at 4 °C overnight.The next day, samples were centrifuged at 1 274 ×gfor 20 min and stored at –80 °C.The blood concentration of total cholesterol (TC),triglycerides (TG), high-density lipoprotein cholesterol (HDL-C),low-density lipoprotein cholesterol (LDL-C) and inflammatory level responses (IL) were measured in the Clinical Laboratory of Southwest hospital TMMU, Chongqing, China.

2.5 Measurement of fat and liver indexes

The mice were sacrificed after the termination of 8 weeks.Mice body was fixed in the anatomical disk, and its abdominal chest cavity was opened using sterilized surgical scissors and tweezers.The mice liver, epididymal fat pad, perirenal fat were collected and tissue index is calculated by following formula.

2.6 Histological analysis of adipose and liver tissues

The adipose tissues were placed in the 4% polyformaldehyde solution for 24 h for fixing.The tissues were then covered in paraffin,and 3 paraffin slices were cut evenly for each sample and stained with hematoxylin-eosin.The prepared pathological slices were placed under an optical microscope, and the representative areas were photographed with image Pro plus 6.0 software and measured the area of the fat cells along with the short and long diameter of each fat cell.

2.7 DNA extraction and 16S rDNA amplicon sequencing

Mice fecal samples were enclosed in a pre-aseptic 1.5 mL EP tube with 1 mL of Phosphate buffer saline solution.Specimens were homogenized with a (Mini Vortex Mixer Shanghai Huxi Analytical Instrument Factory) for 30 s, and continued until a uniform bacterial suspension was prepared.The bacterial suspensions were then incubated in an ice-water bath, and performed intermittent ultrasonic treatment with 20% ultrasonic power for 5 min.200 μL of ultrasonically treated bacterial suspension were taken in a sterile 1.5 mL EP tube, centrifuge at 22 673 ×gfor 5 min, and then added 200 μL of PBS to re-suspend the bacterial suspension.After that, 5 μL of lysozyme were added and incubated at 37 °C for 15 min; then added 200 μL binding buffer, 40 μL proteinase K and incubated at 70 °C for 10 min.Added 100 μL of isopropanol to mixture, and assembled the filter tube and collector, aspirated the sample and centrifuge at 22 673 ×gin the filter tube for 1 min.After centrifugation, collection tubes were discarded and inserted the filter tube into a new collection tube, then added 500 μL inhibitor buffers to the filter tube, and centrifuge at 22 673 ×gfor 1 min and repeated twice.Next, we removed the waste liquid from filter tube and added elusion buffer.The DNA was collected in EP tube and its concentration was measured with a Nano Drop ND-1000 spectrophotometer (USA),and DNA integrity was examined by 0.1% (m/V) agarose gels electrophoresis.For drying the DNA, we added 1/10 of the volume of sodium acetate (3 mol/L, pH = 5.2) and mixed thoroughly in the DNA solution to make the final concentration 0.3 mol/L.Pre-chilled anhydrous ethanol was added, mixed it thoroughly and placed in refrigerator at –20 °C for 30 min and then centrifuged it at 22 673 ×gfor 10 min.Added half the volume of the centrifuge tube with 70% (V/V)ethanol, centrifuged at 22 673 ×gfor 2 min.Finally, the uncapped EP tube was placed on the laboratory table at room temperature to evaporate the remaining liquid to dryness.The dried DNA samples were sent to the Institute of Microbiology and Epidemiology, Beijing Academy of Military Medical Sciences for 16S rDNA amplicon sequencing.

2.8 Statistical analysis

All data of this experiment employed GraphPad Prism 8.0 statistical software for statistical data analysis.The student’st-test was used to compare the differences between groups.The test level ofP＜ 0.05 was considered to be significant, andP< 0.01 referred highly significant.All values are calculated as mean ± SEM.

3.Results

3.1 Body weight comparison between CVZ and PGF mice

After 8 weeks administration of CK diet, the average body weight gain observed in CVZ-CK was significantly higher than PGF-CK group (P＜ 0.05).HF dietary supplementation increased body weight gain in both CVZ-HF (5.6 ± 0.5) g and PGF-HF (3.5 ± 0.4) g mice, but PGF-HF mice weight gain was significantly lower than CVZ-HF mice group (P< 0.01).After the administration of TP, the average loss in body weight gain by HF diet of the mice groups were(4.6 ± 0.5) g in CVZ HF + 0.8% TP, (3.3 ± 0.3) g in CVZ HF + 1.6% TP,(3.1 ± 0.7) g in PGF HF + 0.8% TP, and (3.0 ± 0.6) g in PGF HF + 1.6%TP.It is noted that 0.8% TP intervention reduced the body weight gain in CVZ (1.1 g) and PGF mice (0.4 g), whereas 1.6% TP reduced the body weight of CVZ and PGF mice by 2.1 g and 0.5 g, respectively.The weight loss in PGF mice by administrating 0.8% and 1.6% TP was significantly lowered (P< 0.01 andP＜ 0.05) than those in CVZ mice,respectively.There was a significant difference between CVZ-HF and CVZ HF + 0.8% TP (P< 0.01) and CVZ HF + 1.6% TP(P< 0.001) group; however, there was no significance difference, on comparing PGF-HF and PGF low and high dose TP mice groups (Fig.2).

Fig.2 Differences in body weight between PGF mice and CVZ mice;*P ＜ 0.05; **P < 0.01; **P < 0.001.Values are presented as mean ± SEM,n = 10 in each group.

3.2 Lipid metabolism comparison between CVZ and PGF mice

TP administration resulted in the reduction of serum glucose and lipid parameters (Fig.3).After 8 weeks, the serum GLU level in PGFCK ((0.9 ± 0.6) mmol/L) and PGF-HF ((3.0 ± 1.1) mmol/L) were significantly lower (P < 0.01) than CVZ-CK ((4.6 ± 0.4) mmol/L)and CVZ-HF ((8.0 ± 0.8) mmol/L), respectively.Meanwhile, the low and high TP intervention steadily ameliorated GLU levels.According to the results, PGF HF + 0.8% TP and PGF HF + 1.6% TP groups showed significantly lower (P< 0.000 1 andP< 0.01) serum GLU levels than their respective CVZ mice groups.There was a significant difference (P< 0.01) in serum GLU level between CVZ-HF and CVZ HF + 1.6% TP mice group; while, the GLU of PGF-HF was significantly higher than PGF HF + 0.8% TP (P< 0.01) and PGF HF + 1.6% TP (P＜ 0.01) mice groups (Fig.3a).Next, the HF diet increased TC in both CVZ and PGF mice groups, but there was no significant difference between them.After the TP intervention, the TC level in PGF HF + 0.8% TP ((3.5 ± 0.2) mmol/L) and PGF HF+ 1.6% TP ((3.4 ± 0.2) mmol/L) was lower than its corresponding CVZ mice groups.However, the TC level in CVZ HF + 1.6% TP was significantly higher (P＜ 0.05) than the PGF HF + 1.6% TP mice group (Fig.3b).

Fig.3 The difference in blood glucose and lipid metabolism between PGF mice and CVZ mice.(a) Fasting glucose levels, (b) serum TC levels, (c) serum triglyceride(TG) level, (d) serum HDL-C level, and (e) LDL-C level; *P ＜ 0.05; **P < 0.01; ***P < 0.001; ****P < 0.000 1.Values are mean ± SEM, n = 10 in each group.

There was a significant difference (P＜ 0.05) between PGF-HF and CVZ-HF mice’s TG level.Low and high TP doses were highly operational for dropping TG levels in CVZ compared to PGF mice.TG levels in PGF HF + 1.6% TP mice were significantly higher(P＜ 0.05) than the CVZ HF + 1.6% TP mice group.TG level of CVZ-HF group was significantly higher (P＜ 0.05) than CVZ HF +1.6% TP group, similarly the level of TG in PGF-HF mice group was significantly higher (P＜ 0.05) than PGF HF + 1.8% and PGF HF + 1.6% TP mice group (Fig.3d).Furthermore, the HF diet had little effect in increasing HDL-C and LDL-C in CVZ and PGF mice groups.TP administration showed no significant changes in blood HDL-C and LDL-C in PGF mice; however, the extent of HDL-C and LDL-C raised in CVZ mice, and the level of HDL-C was significantly higher in CVZ 0.8% HF + TP mice than its corresponding PGF mice group.However, compare to PGF-HF, the level of HDL-C significantly increased (P＜ 0.05) in both PGF HF + 0.8% and 1.6%TP mice group.

3.3 Differences in inflammation levels between PGF and CVZ mice

After the 8 weeks of ingesting conventional diet, the levels of IL-1 and IL-10 were higher in PGF mice, whereas the levels of IL-6 and TNF-α were slightly higher in CVZ mice, but there was no significant difference between both mice groups.Meanwhile, HF diet intervention rapidly increased the levels of IL-1, IL-6, and TNF-α,while the level of IL-10 decreased in both mice groups.The levels of IL-1, IL-6, and TNF-α were slightly higher in PGF mice, while the level of IL-10 was higher in CVZ mice.

After the TPs administration, IL-1, IL-6, and TNF-α levels decreased and IL-10 level increased in both CVZ and PGF mice,but there was no significant difference between PGF and CVZ mice groups.High and low doses of TPs showed the same effect on decreasing IL-1 level in PGF mice, while HF + 0.8% TP had a stronger effect on reducing IL-1 level in CVZ mice (Fig.4a).The level of IL-6 was slightly higher in PGF mice treated with HF +0.8% TP, and it was slightly lower in PGF HF + 1.6% TP group than in CVZ mice group (Fig.4b).On the other hand, HF + 0.8% TP intervention triggered an increase of IL-10 level, and after HF + 1.6%TP intervention, the IL-10 level in CVZ mice returned to the levels of inflammation similar to CK control group.The levels of IL-10 in PGF mice were slightly lower than that in CVZ mice, with no significant difference (Fig.4c).It indicated that TPs could reduce inflammation in the body, and its effect on CVZ mice is better than that on PGF mice.The high and low TP doses steadily reduced the TNF-α, indicating that TP was very effective in relieving TNF-α levels in the body (Fig.4d).

Fig.4 Differences in inflammation levels between PGF and CVZ mice.Comparison of (a) IL-1, (b) IL-6, (c) IL-10, and (d) TNF-α levels of PGF and CVZ mice.Values are mean ± SEM, n = 10 in each group.

3.4 Fat weight index comparison between PGF and CVZ mice

The dietary intervention of TP on CVZ mice caused a reduction of the epididymal fat pad and perirenal fat, particularly with increasingly more substantial effects at higher doses (Fig.5).The results showed no significant difference in the weight of perirenal and epididymal fat pad in PGF and CVZ mice treated with the conventional diet.After 8 weeks, in the CVZ mice group, CVZ-HF showed the highest mean weight of perirenal fat ((0.035 ± 0.11) g) and fat pad ((0.013 ± 0.41) g).Subsequently, the TP administration to the CVZ-HF group, the weight of both fats, decreased in a dose-dependent manner.Similarly,in the PGF mice group, PGF HF had the highest mean weight of perirenal and fat pad.Low and high doses of TP in CVZ mice significantly reduced the weight of epididymal fat pad and perirenal fat as compared to PGF mice group.Furthermore, the HF + 0.8% TP administration group, PGF mice perirenal and epididymal fat pad exhibited significantly lower weights than CVZ mice (P< 0.000 1).In the HF + 1.6% TP groups of both mice, the weight of perirenal and epididymal fat pad of PGF mice were significantly lower than in CVZ mice (P＜ 0.01).There was a significant difference (P＜ 0.05) in epididymal fat index between CVZ-HF and CVZ high dose TP mice group; however, TP administration did not decrease epididymal fat in PGF mice groups.After the intervention of high dose of TP, the perirenal fat index significantly decreased (P< 0.01) as compare to CVZ-HF mice group.TP doses significantly reduced (P< 0.01 andP< 0.001) perirenal fat weight in CVZ-HF.The results revealed that the effects of low and high TP doses on the weight of PGF mice fats were not significantly different.However, the low dose TP group had a broader difference than the high dose group in CVZ mice; it indicated that the higher dosage of TPs might have a more substantial effect on fat reduction.

Fig.5 Differences in fat weight indexes between PGF and CVZ mice.(a) Perirenal fat, (b) epididymal fat pad.**P < 0.01; **** P < 0.000 1.Values are mean ±SEM, n = 10 in each group.

3.5 Fat cells morphology in mice with intervention of HF diet and TP

The changes in the fat cells’ morphology by the intervention of the HF diet and TP were shown in (Fig.6).Adipocytes were significantly more abundant in both CVZ and PGF mice fed with the HF diet, resulting in fat cell hypertrophy, and the numbers of fat cells were lessened under the same field view of the microscope compared to the CK group.In CVZ mice group, after fed with 0.8% and 1.6%TPs, the size of fat cells was dwindled does-dependently, indicating that TPs could effectively ameliorate fat cell hypertrophy, whereas,in PGF mice group, a little variation in the size of adipocytes was observed and it indicated that tea TPs slightly inhibited the swelling of adipocytes.

Fig.6 Cross-sections of fat cell morphology of (a) CVZ and (b) PGF mice by intervention of HF diet and TPs.

3.6 The difference between the maximum diameter, minimum diameter and area of fat cells

There was no significant difference in the maximum and minimum diameters of CVZ and PGF mice’s fat cells treated with a conventional diet.In both CVZ and PGF mice groups, the HF diet significantly increased cellular maximum and minimum diameters, but PGF mice showed significantly lower maximum and minimum diameters of fat cells than CVZ mice withP< 0.01,P＜ 0.05, respectively.It might indicate that the HF diet-induced giant fat cells in larger sizes in CVZ mice than those in PGF mice.After the administration of HF + 0.8% TP and HF + 1.6% TP, the maximum and minimum diameters of adipocytes of CVZ mice turned significantly smaller (P< 0.000 1) than those of PGF mice, indicating that TP might effectively inhibit the enlargement of fat cells in CVZ mice.However, TP rarely showed any effects on PGF mice’s fat cells, according to Figs.7a, 7b.On comparing CVZ-HF with low and high TP doses group, the fat cell maximum and minimum diameter significantly reduced (P< 0.001,P< 0.001), respectively; whereas, in PGF-HF the fat cell maximum diameter slightly reduced (P< 0.01) in PGF high dose TP group.

Fig.7 The difference between the diameter and area of fat cells after HF diet and TP intake.(a) Long diameter, (b) short diameter, (c) area of adipocytes.*P ＜ 0.05;**P < 0.01; ****P < 0.000 1.Values are mean ± SEM, n = 10 in each group.

The fat cell area of PGF mice treated with a conventional diet was significantly lesser than that of CVZ mice (P＜ 0.05).After the HF diet, the fat cell area became larger in both PGF and CVZ mice.Nonetheless, PGF mice’s enlarged fat cell area was significantly smaller than CVZ mice (P＜ 0.05), which suggested that the HF diet was probably less effective for PGF mice in the induction of obesity than in CVZ mice (Fig.7c).In the meantime,treatment with TP decreased the area of fat cells in both groups of mice; however, the area of fat cells in PGF mice was larger than that of CVZ mice (P< 0.000 1), which indicated that TP could inhibit the swelling of adipocytes in PGF mice more significantly than CVZ mice.The area of fat cells of CVZ-HF mice group significantly reduced after the intervention of low and high doses of TP (P< 0.000 1,P< 0.000 1), respectively; while after the TP administration, there was no significant decreased in the area of fat cells of PGF-HF mice group.

3.7 Liver weight index comparison between PGF and CVZ mice

The fluctuations in liver weight after 8 weeks of different dietary treatments were highlighted in Fig.8.In the CK group, there was no significant difference between both CVZ and PGF mice’s liver weight.On the other hand, the HF diet had little effect on PGF mice’s liver weight, whereas it significantly induced an increase in the liver weight of CVZ mice (P＜ 0.05).Meanwhile, in the CVZ mice group,0.8% and 1.6% doses of TPs notably change the liver weight in a dose dependent manner.On comparing with CVZ-HF mice, CVZ-HF 1.6%TP group liver index significantly reduced (P＜ 0.05); however,in the PGF mice group, the administration of 0.8% and 1.6% TP showed minor changes in liver weight.

Fig.8 Differences in liver weight index between PGF mice and CVZ mice;*P < 0.05.Values are mean ± SEM, n = 10 in each group.

3.8 Liver cells morphology by intervention of HF diet and TP

Fig.9 showed the morphology of hepatocytes near the portal vein of CVZ and PGF mice.In CVZ-CK mice group, the morphology of liver cells was complete, the cytoplasm was stained red, blue cells were the nucleus, and the hepatocytes were radially distributed around the central vein (Fig.9a).In the CVZ-HF group, large white fat vacuoles and blue nuclei at the center were observed in the mice’s hepatocytes,and the cytoplasm became shorter and squeezed to the edge of the cells.The extent of hepatic steatosis (adipogenesis) in the CVZ HF + 0.8%TP group was less than that in the CVZ-HF group, and the intracellular fat vacuoles were fewer than that in the CVZ-HF group.Likewise, the extent of hepatocyte steatosis in the CVZ HF + 1.6% TP group was less than that in the CVZ-HF group.Statistically, there was no significant difference between the CVZ HF + 0.8% TP and CVZ HF + 1.6% TP mice group.The results indicated that TP could alleviate the hepatic steatosis induced by a HF diet in CVZ mice.

Fig.9 Cross-sections of liver cell morphology of (a) CVZ and (b) PGF mice after HF diet and TP intake.

On the other hand, the morphology of hepatocytes in the PGF-CK group was intact.There were white fat vacuoles in the PGF-HF group in hepatocytes with blue nuclei in the cells’ center (Fig.9b).The cytoplasm decreased, and the cells were about to be squeezed to the edge of the cells.After the dietary intake of 0.8% and 1.6% TP, liver fats’ concentration was retained, and the intracellular fat vacuoles were still prominent.Thus, it is indicated that TP showed no significant effect on hepatic steatosis in PGF mice induced by a HF diet.

3.9 Changes in the phylum level abundance of the intestinal flora of CVZ mice

The analysis of the changes in bacterial abundance at the phylum level of the intestinal flora of CVZ mice is shown in Fig.10.After 8 weeks of continuous intake of a HF diet, the bacterial abundance of Actinomycete in the CVZ-HF group was significantly higher(P< 0.001) than CVZ-CK group.Dietary intervention of 0.8% TP and 1.6% TPsignificantly decreased (P＜ 0.05 andP< 0.001) the bacterial abundance of the intestinalActinobacteria.The HF diet significantly reduced (P< 0.01) the bacterial abundance of the intestinal Bacteroides in the CVZ-HF group.After the intervention of 1.6% TP,the abundance ofBacteroides significantly increased (P< 0.001) than HF diet group.In addition, the HF diet significantly increased the bacterial abundance of the intestinal Firmicutes and Proteobacteria in mice (P< 0.01,P< 0.001).After the intervention of 1.6% TP, the abundance of Firmicutes and Proteobacteria significantly reduced(P< 0.001,P＜ 0.05).This showed TP intervention can improve the intestinal flora disorder caused by HF diet.

Fig.10 Difference in the phylum level abundance of intestinal flora of CVZ mice at week 0 and week 8.The gut microbiota composition of (a, e) CVZ-CK,(b, f) CVZ-HF, (c, g) CVZ HF + 0.8% TP, (d, h) CVZ HF + 1.6% TP group at the phylum level at (a–d) week 0 and (e–h) week 8.

Fig.10 (Continued)

3.10 Changes in the phylum level abundance of intestinal flora in PGF mice

After 8 weeks, the bacterial abundance of phylumBacteroides in the intestinal tract of PGF-CK mice was significantly lower than before treatment at week 0; and after HF diet administration, the abundance of Bacteroides was significantly increased (P< 0.001), but its abundance was still lower than before antibiotic treatment at week 0.The intervention of 1.6% TP increased the bacterial abundance of the intestinal Bacteroides, but there was no significant difference between PGF-HF diet group and 1.6% TP + HF diet group.After PGF mice ingested HF diet for 8 weeks, the bacterial abundance of the intestinal Firmicutes was significantly higher(P＜ 0.05) than that of PGF-CK mice.After the intervention of 0.8% and 1.6% TP, the bacterial abundance of the intestinal Firmicutes in the obese mice was significantly reduced (P<0.001).Data analysis showed that the bacterial abundance ratio of the intestinal Firmicutes/Bacteroides of PGF mice that ingested a HF diet under the regulation of 0.8% and 1.6% TPs gradually decreased, of which 1.6% TPs intervention showed effective results.Antibiotic treatment caused an extremely significant increase in the bacterial abundance of the intestinal Proteobacteria of PGF mice.After a HF diet, the abundance of the Proteobacteria in the PGF-HF group was significantly lower than the PGFCK group (P< 0.001).After TP intervention, its abundance significantly increased (P< 0.01) (Fig.11).

Fig.11 Differences in the phylum level abundance of intestinal flora of PGF mice at week 0 and week 8.The gut microbiota composition of (a, e) PGF-CK, (b, f)PGF-HF, (c, g) PGF HF + 0.8% TP, (d, h) PGF HF + 1.6% TP group at the phylum level at (a–d) week 0 and (e–h) week 8.

3.11 Changes in the genus level abundance of intestinal flora in CVZ mice

Before the experiment, the main bacterial genera in the intestine of CVZ mice wereAkkermansia,Allobaculum,Bacteroides,Clostridium,Desulfovibrionaceae,p-TM7,Lactococcus,Parabacteroides, andPrevotella, which accounted for 90% of the total bacterial population(Fig.12).After 8 weeks of the experiment, more than 90% of the intestinal bacteria of CVZ mice wereAkkermansia,Allobaculum,Bacteroidales,Bifidobacterium,Clostridiales,Dorea,Lactococcus,Lactobacillus, and Lachnospiraceae.After 8 weeks of HF diet ingesting in CVZ mice, the abundance ofAkkermansia,Dorea,andRuminococcaceae in the intestines were significantly reduced;however the administration of TP results significant increase in their abundance.The HF diet also caused a significant increase in the abundance ofBifidobacterium,Desulfovibrionaceae,Lactobacillus,andRuminococcusin the intestines of CVZ-HF mice respectively,and the abundance ofPrevotellawas higher than that of CVZ-CK diet group.The abundance ofBifidobacterium, Desulfovibrionaceae,RuminococcusandPrevotellasignificantly decreased under the intervention of 0.8% TP.Under the intervention of 1.6% TP dosage the abundance ofBifidobacteriumandDesulfovibrionaceae significantly decreased.

Fig.12 Differences in abundance of gut microbiota at genus level in CVZ mice at week 0 and week 8.The gut microbiota composition of (a, e) CVZ-CK,(b, f) CVZ-HF, (c, g) CVZ HF + 0.8% TP, (d, h) CVZ HF + 1.6% TP group at the genus level at (a–d) week 0 and (e–h) week 8.

3.12 Changes in the genus level abundance of intestinal flora in PGF mice

Before the experiment, the main bacterial genera in the intestines of PGF mice were Bacteroidales,Akkermansia,Lactobacillus,Clostridiales,Prevotella,Ruminococcus,Allobaculum,and Lachnospiraceae, which accounts for 80% of the total bacterial population.After the antibiotic treatment, the mainly genus Bacteroidales, Clostridiales,Prevotella,Allobaculum,Ruminococcuswere largely reduced in PGF-CK diet group; correspondingly,EnterobacteriaceaeandStuterellawere colonized.Among them,the abundance ofAkkermansiaandLactobacilluswere significantly increased after antibiotic treatment, as shown in Fig.13.After the PGF mice ingested a HF diet, the abundance ofBacteroides,Lactobacillus,Clostridium,Prevotella,Allobaculum,andp-TM7were significantly reduced than before the experiment; however,Akkermansia,Erysipelotrichaceae,Sutterella,Bacteroides,Parabacterioideswere significantly increased, and Enterobacteriaceaedisappeared in the intestine.After the intervention of 0.8% and 1.6% TP, the abundance ofAkkermaniaandLactobacillusincreased with the increase in the dosage of TP dose.The abundance of Erysipelotrichaceae,Sutterella,Helicobacter, andBacteroidesgradually decreased with the increase of TP dosage.After antibiotic treatment, the main bacteria genre that increased wereAkkermaniaandSutterellain both PGF-CK and PGF-HF group.In addition, HF diet increased Eryipelotrichaceae,BacteroidesandParabacteroides; however, after the intervention of low and high doses of TPs, the abundance of these three bacteria gradually decreased.The administration of TP also promoted the colonization ofParabacteroides(Fig.14).

Fig.13 Changes in the genus level abundance of intestinal flora in PGF-CK mice under conventional diet at (a) week 0 and (b) week 8.The microbiota name legends are same in the two figures.

Fig.14 Differences in abundance of gut microbiota genus levels in PGF mice at week 8. (a) PGF-CK (b) PGF-HF (c) PGF HF + 0.8% TP (d) PGF HF + 1.6% TP.The microbiota name legends are same in a-d.

3.13 Alpha and beta diversity analysis of intestinal microbial community

Alpha diversity was observed in both CVZ and PGF mice, CK group had the highest abundance and diversity of gut microbiota, and the HF group had the lowest.Meanwhile, treatment with 0.8% and 1.6% TP recovered the abundance of both CVZ and PGF HF mice,compared to HF group (Fig.15).On the other hand, according to the results of beta diversity (Fig.16), in CVZ mice, each group showed relatively more different diversity of their gut microbiome compared to other groups.Interestingly, in PGF mice, all groups had more similar diversity of gut microbiota between each other.These results indicated that a HF diet might reduce the intestinal microbial diversity of mice,while treatment with TP at different doses probably had a significant effect on the recovery of gut microbiota of mice.Since the treatment of TP exhibited less changes of the intestinal microbial diversity of mice,and slight obesity was induced by HF diet in PGF mice.

Fig.15 Alpha diversity of CVZ and PGF mice.(a) The rarefaction curve, (b) The species accumulation curves.

Fig.16 (a) Beta diversity of CVZ and PGF mice, (b) weighted principal coordinate analysis (PCoA) of intestinal microbial community, (c) Un-weighted PCoA of intestinal microbial community.

4.Discussion

This research aimed to investigate the potential effects of TP on diet-induced obese PGF and CVZ mice for 8 weeks.The establishment of obese mice model was the basis of our research.C57BL/6J mice can induce obesity by ingesting high fat diet;additionally, it takes short time with obesity characteristics remain stable for a long time [32].In our study, after feeding HF diet to CVZ mice group, the changes in blood glucose and lipid profiles indicated the development of successful obese mice model.The mouse obesity model established in this experiment was consistent with previous research results [33].We used PGF mice in order to study the relationship between gut microbiota and weight loss mechanism by TP.The antibiotics did not remove the entire micro flora of PGF mice;however, the number and species are greatly reduced.As a result of giving PGF mice a HF diet, their body and fat weight increased.However, we noted that the effects of a HF diet on body phenotypes differed between CVZ and PGF mice.Earlier studies have explained that the intervention of HF diet and high-sugar Western diet to germfree mice did not induce obesity, indicating that high-calorie diets alone cannot cause obesity.The upsurge in AMPK activity and quantity of fasting induced-adipose factor are the two mechanisms behind the germ free mice shielded from diet induced obesity [34].

The differences between CVZ and PGF mice after TP intervention were investigated further on the basis of obesity caused by a HF diet.In obese PGF mice, the effect of dietary intervention of TP on fat weight, blood glucose, blood lipid index, and inflammation level was less significant than in CVZ mice.The fat reducing ratio of TP in CVZ mice was greater than that of PGF mice.The two abundant bacteria in the intestine are Firmicutes and Bacteroides which are collectively involve in host energy and storage [34].When the intake of polyphenols increases, the bacterial abundance of the intestinal Bacteroides increases and promotes the fermentation of phenolic substances, and it is widely believed that a decreased ratio of the major phylum Firmicutes/Bacteroides can prevent the development of obesity in both dietary and genetic models of obesity in mice [35].Similarly, we found that the bacterial abundance ratio of Firmicutes/Bacteroides together with Actinobacteria in the digestive tract of CVZ mice decreased by following consumption of low and high doses of TP.However, unlike the CVZ mice groups, the ratio of Firmicutes/Bacteroides did not decrease much in the PGF mice group following TP intervention, resulting in very weak TP fermentation ability, and the activity of polysaccharide degrading enzymes indicating that TP action on PGF mice is less effective.We found that TP dietary intervention significantly decreased the upsurge in serum GLU by the HF diet.The extent of lowering blood GLU levels by TP was highly significant in CVZ mice.Studies found that the endocannabinoid system is involved in glucose and energy metabolism regulation, and that certain gut bacteria; like,Akkermansiaspp.may turn it up or down [36].In our study, most of the micro flora eliminated following antibiotic treatment; however, TP administration improved the abundance ofAkkermansiaandLactobacillus.TP extract epigallocatechin gallate (EGCG) drops lipid indexes by opposing the enzymes responsible for the breakdown and absorption of lipids [38,39].Our results demonstrated that TP administration was significantly effective in reducing serum TC and TG levels in CVZ mice, whereas, in PGF mice, there were no significant changes in lipid indexes.

Our result in altering liver weight by feeding the HF diet was in agreement with the study by Xu et al.[40].We found that TP administration reduced liver weight; however, in CVZ mice, more significant effects were observed than in PGF mice.Our results demonstrated histologically that the degree of hepatic steatosis was significantly reduced by the supplementation of TP in CVZ mice.A previous study demonstrated that germ-free mice fed on an HF diet were resistant to hepatic steatosis and other metabolic diseases [41].Similarly, in PGF mice, TP was not effective for reducing hepatic steatosis induced by the HF diet.In our research, the impact of the HF diet on inflammatory levels was close to that Feuerer et al.[42]reports that the levels of pro-inflammatory cytokines significantly increased in the liver and adipose tissues of mice taking a HF diet;including IL-1, IL-6, and TNF-α, while IL-10 decreased.Prevotellais an important genus of bacteria in the phylum Bacteroides and has a potential pro-inflammatory effect, and the colonization ofPrevotellais linked with weight loss effect in the body [43].Compare to CVZ mice, we found that the abundance ofPrevotellaincreased at first by HF diet, but then decreased after consuming TPs in PGF mice;thus the inflammation in the body and the inhibition of the growth ofPrevotellais related to the consumption of TPs.

5.Conclusion

To sum up, the administration of a HF diet and TP affect differently on body phenotypes and gut microbiome of CVZ and PGF mice.It is reasonable to conclude that the gut microbiome is an indispensable factor in the anti-obesity effect of TP.TP can alleviate obesity by reshaping the structure of intestinal flora and promoting the colonization of beneficial bacteria.However, a potential mechanism of TP in reducing body weight and fat still requires further investigation to understand the gut microbiome and host health relationship.

Conflict of interest

The authors have declared no conflict of interest.

Acknowledgement

The work was financially support by the Key-Area Research and Development Program of Guangdong Province (2020B020226008,2018b020206001), the National Science Foundation of China(NSFC31171673).