Phytochemical composition of Tibetan tea fermented by Eurotium cristatum and its effects on type 1 diabetes mice and gut microbiota

Abstract

“Golden-flower” Tibetan tea (GTT) is an innovative dark tea fermented via fungus Eurotium cristatum. To study GTT effects on alleviating the symptoms of type 1 diabetes mellitus (T1DM), GTT's extract (GTTE) was prepared. GTTE chemical compositions were analyzed via HPLC, pyrolysis-gas chromatography-mass (Py-GC-MS) spectrometry analysis, and chemistry analyses. GTTE effects on T1DM were explored on T1DM mice model induced by streptozotocin (STZ). GTTE was composed mainly of tea pigment theabrownin (TB) (49.18%), with high percentages of polysaccharide (16.93%), protein (10.15%), polyphenols (13.90%), amino acids (5.89%), caffeine (1.83%), and flavonoids (0.67%). Py-GC-MS results exhibited that GTTE constituted of phenols, lipids, sugars, and proteins. GTTE attenuated T1DM conditions of mice, relieved their liver and pancreatic injury, restored damaged islet cells, decreased oxidative stress by increasing superoxide dismutase (SOD) and catalase (CAT) levels, modulated cytokine expression leading to the decreasing pro-inflammatory cytokines TNF-α and IL-6, increased anti-inflammatory cytokines IL-4 to improve inflammatory responses, and optimized gut microbiota composition and structure based on high-throughput 16S rDNA sequencing, suggesting multi-channel anti-diabetes mechanisms.

1. Introduction

Diabetes mellitus (DM) is a complex metabolic disorder characterized by hyperglycemia caused by the defects in insulin secretion, insulin action, or both. Long-time complications caused by DM increase the hazards of this disease as well, such as the long-term damage, dysfunction, and failure of different organs, especially the eyes, kidneys, nerves, heart, and blood vessels [1]. There are two types of diabetes mellitus, Type 1 and Type 2. Type 1 diabetes mellitus (T1DM) is caused by the absolute insulin deficiency associated with autoimmune-mediated destruction of pancreatic β-cells islets. T1DM patients regularly take exogenous insulin to prevent the development of ketoacidosis for survival [2]. Although T1DM composes of about ten percent of the diabetic population, it is more difficult for people to overcome or alleviate T1DM than Type 2 diabetes mellitus (T2DM). There are many challenges for the prevention and treatment to T1DM as according to the recent report in The Lancet [3].

In addition to the variety of medicines being used for the treatment of diabetes, herbs and botanicals, such as dietary supplements, are also used for glycemic control and management of diabetic complications [4]. Most plants generate bio-active natural substances including polyphenols, flavonoids, terpenoids, coumarins, and glycosides, some of which were reported to have anti-diabetic effects [5].

Tea (Camellia sinensis L.), derived from the leaves or buds of tea trees, has a long history of more than 5000 years in dietary and medicinal applications [6]. Nowadays, tea has become more and more popular due to its attractive aroma, taste and health-promoting benefits, such as antioxidant, anticancer, anti-inflammatory, immunomodulatory, and hypoglycemic activities [[7], [8], [9], [10]]. According to the post-harvest processing and the degree of fermentation, tea products are generally classified into five major categories in China, i.e., unfermented (green tea), slight-fermented (white tea and yellow tea), semi-fermented (Oolong tea), fermented (black tea), and the post-fermented (dark tea) [11]. The difference lies in the degree of fermentation, which results from oxidative, enzymatic changes, or microbial fermentation within the tea leaves during the manufacturing process [12]. There are significant differences in the active components in the various types of teas. Catechins are the predominant polyphenolic compounds in green tea. Tea pigments theaflavin and thearubigin are present in high concentrations in black tea whereas, theabrownin (TB), is the main tea pigment with various bio-activities in dark tea [13,14].

Tibetan tea is a dark tea in China originally made in Ya'an city area of Sichuan Province. This tea has existed for thousands of years and is deeply loved by the Tibetans. Tibetans live in the Qinghai-Tibet Plateau, an area with extreme cold, hypoxia, and intense UV radiation. The Tibetan eating habits consist of long time high-fat foods and less dietary fiber. Because Tibetan tea is rich in polyphenols, polysaccharides, trace elements, free amino acids and other components [15]. Tibetan tea is helpful for Tibetan digestion and health, as the saying “the energy of meat can't be digested without Tibetan tea, the heat of hulless barley can't be dispelled without Tibetan tea” [16]. In recent decades, the application of an innovative post-fermentation process has caused dominate fungi probiotic Eurotium cristatum yellow flora to grow on both brick and loose forms of Tibetan tea, causing the tea to possess many specific “golden flowers” in the dark tea. Nowadays, drinking the “golden flower” Tibetan tea (GTT) has become popular among Chinese people. However, there are few research reports of its antidiabetic effect, especially on its effects on the treatment of T1DM. The aims of this paper are to study and elucidate the phytochemical compositions of the extract of the “golden flower” Tibetan tea (GTTE), and to investigate its effects on T1DM mice model and the influence on gut microbiota profiles. This research provides deep insights of the phytochemical components of drinking from GTT, potential supplementary health benefits on T1DM, and the possible mechanisms.

2. Materials and methods

The “golden-flower” Tibetan tea (GTT) bricks (batch no. 20100619) were made from tea leaves and hubs (Camellia sinensis L.) that were post-fermented for 10 years with the growing Eurotium cristatum yellow flora (from Ya'an Golden Flower Tibetan Tea Corporation).

2.1. Reagents and standards

Metformin (DMBG) was purchased from Sino-American Shanghai Squibb Pharmaceuticals Co., Ltd. The superoxide dismutase (SOD) kit and catalase (CAT) kit were obtained from Nanjing Jiancheng Biotechnology Co., Ltd. Multi-factor detection kit was purchased from American R&D company. Insulin ELISA kit was purchased from American Millipore Co., Ltd. Paraformaldehyde (4%) was purchased from Beijing Soleibo Technology Co., Ltd. Streptozotocin (STZ) was obtained from American Sigma Company. Gallic acid (GA), Catechin (C), Epicatechin (EC), Gallocatechin (GC), Epigallocatechin (EGC), Catechin gallate (CG), Epicatechin gallate (ECG), Gallocatechin gallate (GCG), Epigallocatechin gallate (EGCG), purity of up to 98%, were purchased from Beijing Solarbio Science & Technology Co., Ltd.

Instruments and apparatuses were used in the studies included the following: CR21 N high-speed refrigerated centrifuge (Hitachi Co., Ltd., Japan), RM2235 paraffin microtome (Leica Co., Ltd., Germany), Eclipse Ci- L upright fluorescence photo microscope (Nikon Co., Ltd., Japan), MV752 N UV–Vis spectrophotometer (Shanghai Youke instrument Co., Ltd. China), Analysis platform (Millipore Co., Ltd., USA), Eclipse Ci-L upright white light photo microscope (Nikon Co., Ltd., Japan), Eclipse Ci-L upright polarized light photo microscope (Nikon Co., Ltd., Japan), Eclipse Ti-E laser confocal microscope (Nikon Co., Ltd., Japan), PANNORAMIC DESK/MIDI/250/1000 panoramic slide scanner (3DHISTECH Ltd., Hungary); 2.5\10\20\200\1000\5000 μL pipette (Eppendorf Ltd., Germany).

2.2. Preparation of “golden-flower” Tibetan tea extract (GTTE)

The dry powder of “golden-flower” Tibetan tea (5875 g) was homogenized with 58.75 L water and extracted at 75 °C for 2 h. The supernatant was decanted and retained. The residue was further extracted twice at 75 °C for 1 h and 0.5 h, respectively. Then, the three extracted supernatants were combined and evaporated under vacuum at 50 °C on a rotary evaporator (Hei-VAP; Heidoiph Instruments, Schwabach, Germany), until the volume was reduced to 1/5 of the original volume. The remaining aqueous phase was mixed with anhydrous ethanol (1/4, v/v), and then precipitated at 4 °C for 24 h. The precipitate was collected and dissolved in distilled water. The solution was finally frozen-dried to obtain 682.5 g of brown extract powder with the yield of 11.62%. This water-extraction alcohol-precipitation extract of “golden-flower” Tibetan tea (GTTE) was used for the following experiments.

2.3. Phytochemical composition analyses of GTTE

2.3.1. Determination of polysaccharides, total phenols, total flavonoids, amino acids, protein and tea pigments

The polysaccharide content was determined by the phenol-sulfuric acid method using dextrose as a standard [17]. The protein content was measured by Coomassie Brilliant Blue method using bovine serum albumin as reference. The amino acids were determined according to a national standard GB5009.124–2016 method of China. The total polyphenols content was determined by the Folin-Ciocalteu method using gallic acid as a standard [18]. The total flavonoids content was determined by the AlCl3 colorimetric method with rutin as a standard [19]. The contents tea pigments, i.e., theaflavin, thearubigin, and theabrownin (TB) were determined using Roberts colorimetric method [17].

The analyses of nine phenolic compounds (GA, C, EC, GC, EGC, CG, ECG, GCG and EGCG) and caffeine were performed using an Agilent LC-1290 UHPLC system (Agilent, USA, Santa Clara, CA, USA) equipped with a diode array detector (DAD). Chromatographic separation was conducted on a Waters BEH C18 (2.1 × 100 mm, 1.7 μm; Waters Corporation; Milford, MA, USA). The mobile phase consisted of water soluble with 1% formic acid (A) + acetonitrile (B). The gradient program was set at 0–2 min (5%–10% B), 2–10 min (10%–20% B), 10–15 min (20%–40% B), 15–17 min (40%–70% B), 17–20 min (70% B–95% B) using a flow rate of 0.3 mL/min. The column temperature was 30 °C; the detection wavelength was 280 nm. Samples were filtered using a 0.22 μm membrane filter and the injection volume was 1 μL. The calibration for the quantification of the nine phenolic compounds and caffeine is exhibited in Supplementary Material S1 Table 1.

Table 1.

The phytochemical component of GTTE.

Conventional composition （g/100g）		Tea pigment（g/100g）		Phenolic compound（g/100g）		Amino acid（g/100g）
Tea polysaccharide	16.93 ± 0.2	Theabrownin	49.18 ± 0.28	GA	0.48 ± 0.00	Aspartate	0.75 ± 0.01
Protein	10.15 ± 0.11	Theaflavin	0.25 ± 0.03	EGC	0.06 ± 0.00	Threonine	0.36 ± 0.01
Total amino acids	5.89 ± 0.01	Thearubigin	ND	C	0.05 ± 0.00	Serine	0.46 ± 0.00
Total polyphenols	13.90 ± 0.65	–	–	EGCG	0.03 ± 0.00	Glutamate	1.04 ± 0.02
Total flavonoids	0.67 ± 0.00	–	–	GCG	0.01 ± 0.00	Glycine	0.44 ± 0.01
Caffeine	1.83 ± 0.00	–	–	EC	0.11 ± 0.00	Alanine	0.23 ± 0.01
		–	–	ECG	0.06 ± 0.00	Valine	0.28 ± 0.01
–	–	–	–	CG	0.01 ± 0.00	Methionine	0.16 ± 0.00
–	–	–	–	GC	0.51 ± 0.00	Isoleucine	0.15 ± 0.00
–	–	–	–	–	–	Leucine	0.27 ± 0.01
–	–	–	–	–	–	Tyrosine	0.19 ± 0.00
–	–	–	–	–	–	Phenylalanine	0.17 ± 0.01
–	–	–	–	–	–	Lysine	0.32 ± 0.01
–	–	–	–	–	–	Histidine	0.29 ± 0.01
–	–	–	–	–	–	Arginine	0.14 ± 0.01
–	–	–	–	–	–	Proline	0.34 ± 0.02

ND, not detected.

2.3.2. Pyrolysis-gas chromatography-mass (Py-GC-MS) spectrometry analysis of GTTE

The Pyrolysis-gas chromatography-mass (Py-GC-MS) (Shimadzu PY2020, Japan) was used to determine the thermal decomposition fragments of GTTE, especially that of tea pigment TB. The Py-GC-MS conditions were programed as following: pyrolysis time 10 s, at the pyrolysis chamber temperature 380 °C and 600 °C, respectively, injection needle temperature 380 °C. J&W DB-5 quartz capillary column (30 m) was used to separate the compounds. The analysis was conducted via programmed temperature: 60 °C (5 min), 60–280 °C at 10 °C/min (8 min), 280–380 °C at 5 °C/min (2 min). Carrier was helium gas, and flow rate was 5.0 mL/min. MS conditions: ion source temperature 220 °C, transmission line temperature 270 °C, electron ionization source and electron energy 70 eV, scanning time 0.2 s, quality scan range 50–500. The area normalization method was used to determine the relative content of decomposing substances.

2.4. Animals and experiments

2.4.1. Animal experiments

The GTTE was used as test sample. A total of 70 male mice from Institute of Cancer Research (ICR), weighing 20–24 g (7–8 weeks old), were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Certificate Number SCXK [zhe] 2019-0001; Beijing, China). They were adaptive having free access to feed (normal feed nutrition see Supplementary Material S2 Table 2) and water (one mouse per cage) in an environmentally controlled room (temperature 22 ± 1 °C, relative humidity 55%–60%, 12 light/dark cycle) for two weeks. The study was approved by Animal Care and Use Committee (Approval No. KS20201023), Sichuan University, China.

Table 2.

Pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS) analysis of GTTE at 380 °C.

No.	Rt (min)a	Compound	CAS No.	Content (%) b	SI c
1	1.342	Ethylamine	75-12-7	16.73	94
2	1.415	5-Amino-6-nitroso-2,4(1H,3H)-pyrimidin	0-00-0	7.02	95
3	1.466	Ethanol	64-17-5	10.45	97
4	1.565	3-oxo-Butanenitrile	0-00-0	9.97	92
5	1.879	2,3-Butanedione	431-03-8	2.04	98
6	2.048	Acetic acid	64-19-7	11.82	97
7	2.340	Furan	1708-29-8	0.45	79
8	2.390	1-hydroxy-2-Propanone	116-09-6	4.77	98
9	2.717	2,3-Pentanedione	600-14-6	0.48	95
10	2.810	Propanoic acid	79-09-4	0.26	83
11	2.878	3-hydroxy-2-Butanone	513-86-0	0.43	95
12	2.920	2,4-Dimethylfuran	3710-43-8	0.43	89
13	5.622	2-Furanmethanol	98-00-0	0.91	96
14	5.904	1-(acetyloxy)-2-Propanone	592-20-1	0.60	97
15	9.273	Phenol	108-95-2	0.58	95
16	10.055	5-Methyl-5,6-dihydro-2(1H)-pyridinone	0-00-0	0.16	86
17	11.544	Cyclopropyl carbinol	2516-33-8	0.32	88
18	12.778	Xanthine riboside	146-80-5	0.36	89
19	14.444	Methyl pentadecyl ether	0-00-0	0.40	72
20	14.552	2-[(tetrahydro-2H-pyran-2-yl)oxy]- Propanoic acid	87930-86-7	0.50	78
21	14.850	2,5-Monomethylene-l-rhamnitol	0-00-0	0.31	68
22	17.425	1,1′-oxybis-Octane	629-82-3	0.33	92
23	17.563	1-Tetradecanol	112-72-1	0.37	86
24	18.291	Octanoic acid, octyl ester	2306-88-9	0.11	92
25	19.013	Caffeine	58-08-2	29.47	95
26	19.595	Dibutyl phthalate	84-74-2	0.39	95
27	20.412	1-Hexadecanol	36653-82-4	0.11	92
28	20.935	Propionic acid, 3,3′-thiodi-, didodecyl	123-28-4	0.03	68
29	21.256	Octanoic acid, 4-pentadecyl ester	0-00-0	0.03	76
30	22.192	Hexanedioic acid, bis(2-ethylhexyl) ester	103-23-1	0.04	90
31	22.813	Oxalic acid, 2-ethylhexyl isohexyl ester	0-00-0	0.04	83
32	23.050	Diisooctyl phthalate	131-20-4	0.05	92
33	23.418	Heptyl tetracosyl ether	0-00-0	0.04	78

Note.

^aRt, retention time of pyrolytic product.

^bThe relative contents of pyrolytic products were calculated by peak area normalization, and results are expressed as the ratio of single constituent area to total area.

^cSI, index of similarity.

The type 1 diabetes in mice was induced via intraperitoneal administration of Streptozotocin (STZ) (130 mg/kg BW) in citric acid/sodium citrate buffer (0.1 mol/L, pH 4.2–4.5) [20]. Simultaneously, the normal mice were administered equivalent citrate buffer. The normal control (NC) group and the high-dose normal group were given equal doses of citrate buffer. After STZ administration for 7 days, fasting blood glucose (FBG) was detected via a tail vein nick. The mice were considered as successful diabetic models when their FBG was greater than 8.3 mmol/L [21], and with typical diabetes symptoms of polyuria, polydipsia, polyphagia, and weight loss. The diabetic model mice were divided randomly into five groups: diabetic control group (DC, n = 7), positive control group (PC, 333 mg/kg BW Metformin, n = 7), and three dosage groups, i.e., diabetic mice with high dose GTTE group (TH, 900 mg/kg BW GTTE, n = 7), diabetic mice with medium dose GTTE group (TM, 450 mg/kg BW GTTE, n = 7), and diabetic mice with low dose GTTE group (TL, 225 mg/kg BW GTTE, n = 7). The dosages were designed according to a reference of tea pigment capsule study [22]. Meanwhile, normal mice were randomly divided into two groups: normal control group (NC, n = 7), nondiabetic with GTTE group (NDT, 900 mg/kg BW GTTE, n = 7). During the experimental period, fasting blood glucose, body weights and feed intake were measured weekly. Their body weights were weighed weekly to adjust gavage dose, and their behaviors were observed. At the end of the 28 days of experimental period, feces were collected on the day before their sacrifice and stored at −80 °C. Mice were fasted for 12 h and then sacrificed. Blood samples were collected and centrifuged at 3500 rpm for 15 min to gain serum for subsequent analyses. In addition, kidneys, livers, and pancreas of the mice were immediately taken and stored at −80 °C.

2.4.2. Measurement of body weight, fasting blood glucose (FBG), and oral glucose tolerance test (OGTT)

Body weight of each group was measured once a week (0 W, 1 W, 2 W, 3 W, 4 W, and 5 W) in 5 weeks. The fasting blood glucose (FBG) was determined using Accu-Chek Performa (Roche Diagnostics, Mannheim, Germany) blood glucose meter from the tail vein after overnight fasting at 1 W and 5 W.

The oral glucose tolerance test (OGTT) was carried on the day of mice euthanized. The mice were fasted for 4 h (without water) after the last intragastric administration. The 10 mL of glucose (Sigma-Aldrich Co.) solution (2 g/kg) was performed by gastric perfusion. Subsequently, the blood sample were taken from their tail vine, and blood glucose level was determined using Roche blood glucose meter at 30, 60, 90, and 120 min.

2.4.3. Serum insulin test

The blood samples were centrifuged for 15 min at 3500 rpm to obtain the serum for following assays. The content of insulin in serum was detected according to the enzyme-linked immunosorbent assay in the ELASA kit (Millipore Inc., USA).

2.4.4. Organ index of kidney, liver, and pancreas

Weights of kidney, liver, and pancreas were measured to calculate the indexes of these organs, as the ratio of organ mass (g) to body mass (g).

2.4.5. Liver and serum antioxidant assay

Liver tissues were homogenized with physiological saline (1/9, w/v), then centrifuged by 3000 rpm for 10 min at 4 °C. The supernatants were used to determine the liver biochemical indicators, i.e., superoxide dismutase (SOD) and catalase (CAT) levels, for assays of antioxidant, using commercial kits (Nanjing Jiancheng Bioengineering Institute, China). Serum antioxidant assays were also using the kits.

2.4.6. Histopathological observation of liver and pancreas

Hepatic tissue and pancreatic tissue were fixed with paraformaldehyde, dehydrated by gradient ethanol, wrapped in wax, sectioned into 5 μm sections, dewaxed in xylene, and hydrated by gradient ethanol. All specimens were stained through hematoxylin-eosin staining, and sealed with neutral resin to observe histomorphological changes using optical microscopy.

2.4.7. Determination of serum inflammatory factors

Three serum samples were randomly selected from each group. The levels of pro-inflammatory cytokines TNF-α, IL-6, and anti-inflammatory cytokines IL-4 in the serum samples were detected using multi-factor detection kits (US R&D Company).

2.4.8. Determination of islet β-cell area

Three pancreatic tissues were randomly selected from each group, which were embedded with wax and sliced for immunofluorescence double staining. The sections were incubated with specific primary antibodies against insulin and glucagon. Subsequently, sections were incubated with FITC-labeled goat anti-Rabbit IgG + IgM and TRIT-Labeled Goat anti-mouse IgG + IgM, and then observed with a fluorescence microscope (Eclipse Ci-L, NIkon LTD., Japan).

2.4.9. High-throughput 16S rDNA sequencing of gut microbiota

The feces of mice were collected on the day before sacrifice and stored at −80 °C for 16S rDNA sequencing. Total genomic DNA from samples was extracted via CTAB/SDS method. DNA concentration and purity were monitored on 2% agarose gels. Then, the V3–V4 variable regions of the bacterial 16S rDNA gene were amplified with universal primers (ACTCCTACGGGAGGCAGCA and GGACTACHVGGGTYGTTECTAAT). The PCR products were recovered for fluorescence quantification using AxyprepDNA Gel Extraction Kit (AXYGEN). Sequencing libraries were generated using NEB Next®Ultra™DNA Library Prep Kit for Illumina (NEB, USA) following manufacturer's recommendations and index codes were added. The library quality was assessed on the Qubit@ 2.0 Fluorometer (Thermo Scientific) and Agilent Bioanalyzer 2100 system. At last, the library was sequenced on an Illumina NovaSeq 6000 platform and 250bp paired-end reads were generated by Shanghai Applied Protein Technology Corp. (Shanghai, China).

The 16S sequences analysis was performed by UPARSE software package using the UPARSE-OTU and UPARSE-OTUref algorithms. In-house Perl scripts were used to analyze alpha (within samples) and beta (among samples) diversity. Sequences with ≥97% similarity were assigned to the same OTUs. We pick a representative sequences for each OTU and use the RDP classifier to annotate taxonomic information for each representative sequence. In order to compute Alpha Diversity, we rarify the OTU table and calculate three metrics: Chao1 estimates the species abundance; Observed Species estimates the amount of unique OTUs found in each sample, and Shannon index.

Graphical representation of the relative abundance of bacterial diversity from phylum to species can be visualized using Krona chart. Cluster analysis was preceded by principal component analysis (PCA), which was applied to reduce the dimension of the original variables using the QIIME software package. QIIME calculates both weighted and unweighted unifrac distance, which are phylogenetic measures of beta diversity. We used unweighted unifrac distance for Unweighted Pair Group Method with Arithmetic mean (UPGMA) Clustering，which is a type of hierarchical clustering method using average linkage and can be used to interpret the distance matrix.

To identify differences of microbial communities between the two groups, ANOSIM and ADONIS were performed based on the Bray-Curtis dissimilarity distance matrices. Based on the relative abundance of identified differential abundant bacterial group, gut microbiota composition were analyzed at phylum level and genus level.

2.5. Statistical analysis

All results were expressed as mean ± standard deviation (SD). One-way analysis of variance (ANOVA) was performed to analyze statistical differences. When appropriate, student's test was used to identify the significant difference between different groups. A P value < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS software (Version 22.0, SPSS Inc., Chicago, IL, USA).

3. Results

3.1. Phytochemical compositions of extract GTTE

The phytochemical components of GTTE are shown in Table 1. It contented high percentage of polysaccharide (16.93%), protein (10.15%), total polyphenols (13.90%), total amino acids (5.89%), caffeine (1.83%), and small amount of flavonoids (0.67%). For the free amino acids in GTTE, the contents of glutamate (1.04%), aspartate (0.75%), serine (0.46%), and glycine (0.44%) were relatively higher than that of other free amino acids (0.14%–0.36%). However, almost half of GTTE contained tea pigments, i.e., theabrownin (TB) (49.18%), theaflavin (0.25%), but no thearubigin.

3.2. Pyrolysis-gas chromatography-mass (Py-GC-MS) spectrometry analysis of GTTE

The Py-GC-MS analyses results are shown in Tables 2 and 3. As shown in Table 2, ethylamine (16.73%), acetic acid (11.82%), ethanol (10.45%), 3-oxo-butanenitrile (9.97%), 5-Amino-6-nitroso-2,4(1H,3H)-pyrimidin (7.02%), 1-hydroxy-2-propanone (4.77%), and 2,3-butanedione (2.04%) were the high content compounds produced from the decomposition at the pyrolysis temperature 380 °C and 600 °C. The ethylamine should be derived from the decompositions of free amino acids, for example, the decarboxylation of alanine generates ethylamine. Amino acids could also be generated from the pyrolysis of protein and TB. The content of caffeine was the highest at 29.47%. As caffeine is heat stable with very little decomposition resulting high percentage, caffeine should be the major component. When the sample was heated at 600 °C (Table 3), ethylamine (31.79%), ethanol (8.82%), dimethylamine (4.82%), acetic acid (4.16%), 2-methyl-furan (3.57%), 1-hydroxy-2-propanone (3.33%), and 2,3-butanedione (1.98%) were the highest content compounds from the heat decomposition. Ethylamine production was almost doubled at this high pyrolysis temperature, indicating the further decompositions of protein and amino acids in the complex. Caffeine (27.48%) was decomposed only slightly at 600 °C.

Table 3.

Pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS) analysis of GTTE at 600 °C.

No.	Rt (min) a	Compound	CAS No.	Content (%) b	SI c
1	1.326	Ethylamine	75-12-7	31.79	95
2	1.454	Dimethylamine	124-40-3	4.82	92
3	1.541	Ethanol	64-17-5	8.82	93
4	1.727	1-Propen-2-ol, acetate	108-22-5	0.61	92
5	1.770	2-methyl-Propanal	78-84-2	0.36	82
6	1.815	2-Butene-1,4-diol	6117-80-2	0.42	83
7	1.852	2,3-Butanedione	431-03-8	1.98	97
8	1.929	2-methyl-Furan	534-22-5	3.57	95
9	1.984	Acetic acid	64-19-7	4.16	97
10	2.080	2-methyl-Propanenitrile	78-82-0	0.72	87
11	2.180	1,3-Cyclohexadiene	592-57-4	0.77	94
12	2.310	Methoxyacetonitrile	1738-36-9	0.58	83
13	2.367	1-hydroxy-2-Propanone	116-09-6	3.33	96
14	2.600	3-methyl-2-Butanone	563-80-4	0.68	93
15	2.697	2,3-Pentanedione	600-14-6	0.46	96
16	2.802	2,5-dimethyl-furan	625-86-5	0.39	88
17	2.860	3-hydroxy-2-Butanone	513-86-0	0.27	89
18	2.907	2,4-Dimethylfuran	3710-43-8	0.34	92
19	3.046	2,5-dimethy-furan	625-86-5	0.29	88
20	3.165	3-methyl-Butanenitrile	625-28-5	0.27	96
21	3.261	1-methyl-Pyrrole	96-54-8	0.20	85
22	3.489	Pyrrole	109-97-7	0.62	96
23	3.489	Toluene	108-88-3	0.81	97
24	4.155	n-Hexyl acetate	142-92-7	0.38	84
25	5.050	2-Cyclopenten-1-one	930-30-3	0.31	95
26	5.266	cis-3-Hepten-1-ol	1708-81-2	0.41	81
27	5.706	Ethylbenzene	100-41-4	0.36	92
28	5.928	Methyl benzyl ketone	103-79-7	0.72	80
29	6.510	Phenethylene	100-42-5	0.23	86
30	9.242	Phenol	108-95-2	0.37	95
31	9.305	1-Octanol	111-87-5	0.24	88
32	10.062	1,2,5,6-Tetrahydropyridin-2-one	0-00-0	0.24	83
33	11.565	Heptanal	111-71-7	0.21	78
34	12.774	Xanthosine	146-80-5	0.38	80
35	12.904	1-Tetradecene	1120-36-1	0.20	89
36	14.575	2,3-dimethyl-Cyclohexanol	1502-24-5	0.22	65
37	15.099	1-Tetradecene	1120-36-1	0.03	90
38	16.015	1-Hexadecene	629-73-2	0.03	91
39	16.863	1-Pentadecene	13360-61-7	0.04	91
40	17.425	n-Octyl ether	629-82-3	0.11	93
41	18.291	n-Octyl caprylate	2306-88-9	0.17	92
42	18.990	Caffeine	58-08-2	27.48	96
43	19.057	1-Hexadecanol	36653-82-4	0.22	95
44	19.592	Dibutyl phthalate	84-74-2	0.26	95
45	19.640	Hexadecamethylheptasiloxane	541-01-5	0.24	65
46	20.416	1-Eicosanol	629-96-9	0.20	94
47	20.937	Octanoic acid, tetradecyl ester	16456-36-3	0.04	63
48	21.254	1-Propyldodecyl octanoate	0-00-0	0.03	74
49	21.387	Hexadecamethylheptasiloxane	541-01-5	0.04	78
50	21.877	2-Propenoic acid, 3-(4-methoxyphenyl)-, 2-ethylhexyl est Parsol MCX	5466-77-3	0.12	80
51	22.644	Hexadecyl 2-ethylhexanoate	59130-69-7	0.06	89
52	22.810	Carbonic acid, decyl 2-ethylhexyl ester	0-00-0	0.14	85
53	23.048	Di-n-octyl phthalate	117-84-0	0.14	95
54	23.416	1-Ethylhexanoic anhydride	36765-89-6	0.09	78
55	24.571	Bis(2-ethylhexyl) isophthalate	137-89-3	0.06	87

Note.

^aRt, retention time of pyrolytic product.

^bThe relative contents of pyrolytic products were calculated by peak area normalization, and results are expressed as the ratio of single constituent area to total area.

^cSI, index of similarity.

3.3. Effect of GTTE on feed intake, fasting blood glucose (FBG), body weight, glucose tolerance, serum insulin level, organ index, anti-oxidation ability and pro-inflammatory cytokines

The feed intake data was attached in Supplementary Material S2 Table 3 & Fig. 1. In the adaptive week and creating T1DM week, feed intake amount of all the groups were not different. In the first week after GTTE administration, DC group's feed intake showed significant (P < 0.05) more than that of NC group. TH group showed significant (P < 0.05) less than DC group. In the second week after administration, PC group showed higher (P < 0.05) feed intake than NC group. TM group showed less (P < 0.05) feed intake than DC group. In the third week after administration, PC group showed significant higher (P < 0.01) feed intake than NC group, and higher (P < 0.05) than DC group. Maybe metformin treated T1DM mice’ conditions were improved, resulting more feed eating. TL group ate less (P < 0.05) than DC group, yet TM group ate more (P < 0.05) than NC group in this period. In the fourth week after administration, both PC group (P < 0.05) and DC group (P < 0.01) showed more feed intake than NC group. Yet TL and TH groups took less (P < 0.01) feed than DC group. Generally, DC group T1DM mice took more feed than NC group, significantly and not significantly (in the second and third weeks after administrations). Yet GTTE administration groups (TL, TM, and TH) showed less feed intake than DC group, significantly and not significantly. GTTE administration groups' feed intake was not different to NC group (except TM group in the third week). These feed intake results implied that GTTE could attenuate polyphagia symptom of T1DM mice.

Fig. 1.

GTTE attenuated STZ-induced diabetes-related symptoms (n = 7). (a) FBG. (b) Body weight. (c) OGTT. (d) Serum insulin. (e) Liver index. (f) Kidney index. (g) Pancreas index. (h) Liver SOD. (i) Liver CAT. (j) Serum SOD. (k) Serum CAT. (l) Serum TNF-α. (m) Serum IL-6. (n) Serum IL-4. Data are expressed as mean ± standard deviation (SD). ANOVA was performed to analyze statistical differences. *P < 0.05, **P < 0.01 vs NC group (contained NDT vs NC, DC vs NC), #P < 0.05, ##P < 0.01 vs DC group (contained PC vs DC, TL vs DC, TM vs DC, TH vs DC). Note: DC = diabetic control group; PC = positive control group; NC = normal control group; TH = diabetic with high dose GTTE group; TM = diabetic with medium dose GTTE group; TL = diabetic with low dose GTTE group; NDT = nondiabetic with GTTE group; the same in Fig. 2.

The dynamic monitoring of fasting blood glucose (FBG) levels is presented in Fig. 1a. Compared with the NC group, FBG levels in the diabetic groups were significantly increased (P < 0.01) in the first week, indicating that the type 1 diabetic mice model was successfully established. At the beginning, the same FBG states were exhibited in DC group and different treatment groups. The FBG levels of the groups with different dosages of GTTE and PC group were gradually decreased from week 1 to week 5. On week 2, 3, 4 and 5, PC, TL, TM and TH groups showed significant reduction effects of the FBG levels (P < 0.05), compared with that of DC group. This was especially evident for the TM group, which displayed much lower FBG levels than that of other treatment groups. This result showed that GTTE could cause the decrease of the FBG levels of diabetic mice. Furthermore, the medium dose of GTTE (450 mg/kg BW, TM) had the best effect to reduce FBG levels of diabetic mice.

As shown in Fig. 1b, the same body weight trends are exhibited in 7 groups before pharmaceutical interventions. From week 0–5, the body weight of NC mice was continuously increased, while that of mice in all other groups was increased slowly compared with those in the NC group. From week 0–5, a significant decrease in the body weight of mice in DC group was observed compared with the NC group (P < 0.01). However, TM and TH groups showed higher body weight than those of DC group from week 2–5, especially in the TM group in week 5 (P < 0.01). These results indicate that GTTE can significantly alleviate the weight loss caused by type 1 diabetes.

The oral glucose tolerance test (OGTT) of GTTE on diabetic mice is shown in Fig. 1c. The blood glucose of DC group mice was significantly increased within 30 min after oral gavage of glucose, yet TM and TH group significantly reduced the blood glucose level in comparison to DC group (P < 0.01). After oral gavage glucose for 120 min, the blood glucose level of TM and TH group were all significantly lower than that of DC group (P < 0.01). These results indicated that GTTE could significantly improve glucose intolerance in diabetic mice.

As shown in Fig. 1d, the DC group showed a significant drop in the serum insulin level compared with the NC group (P＜0.01). On the contrary, PC, TM, TL, and TH groups remarkably increased the serum insulin levels comparative to the DC group (P ＜ 0.01).

The effects of GTTE on the organ indexes (including liver, kidney and pancreas) of diabetic mice were determined. The organ indexes of liver, kidney and pancreas of DC group were all significantly increased compared with NC group (Fig. 1e,f,g, P < 0.01). Compared with DC group, the liver indexes of the TM group and TH group were significantly decreased (P < 0.01), and kidney and pancreas index in TH group were statistically different (P < 0.05). These results indicated that GTTE had a protective effect on the liver, kidney and pancreas of mice.

The effects of GTTE on the antioxidant enzymes in the liver and serum were measured, as shown in Fig. 1h,i,j,k. The DC group reduced the SOD and CAT activities in the liver and serum as compared with that of the NC group. Compared with the DC group, the levels of SOD and CAT in liver and serum in TL, TM, and TH treatment groups were increased with varying degrees, indicating that GTTE had positive effects on antioxidant capacity.

In order to investigate the effect of GTTE on pro-inflammatory cytokines, serum levels of tumor necrosis factor-α (TNF-α), interleukin-4 (IL-4), and interleukin-6 (IL-6) were detected using reagent test kit (Fig. 1l,m,n). Compared with the NC group, TNF-α in the DC group was significantly increased (P < 0.01), and IL-4 was significantly reduced (P < 0.01). Compared with DC group, TNF-α in the PC group and the TL, TM and TH groups was dramatically reduced (P < 0.01), IL-4 in PC and TM groups was increased (P < 0.01), and IL-6 in PC group was significantly reduced (P < 0.05). Although not statistically significant, the TM and TH group reduced the IL-6 level compared with the DC group. The results showed that the GTTE could significantly improve the inflammatory conditions of diabetic mice.

3.4. Effects of GTTE on liver and pancreatic pathology

The liver tissue status of the mice was observed under a microscope at 400 times (Supplementary Material S3 Fig. 2a). The NC group showed that the liver lobules were clearly structured clearly with neatly arranged liver cords, and liver cells were filled with cytoplasm and exhibited normal morphology and structure. In addition, liver sinusoids were not significantly expanded or compressed. In the DC group, the fibrosis of liver tissue was increased, with the central vein of the liver lobule and the nearby hepatic sinusoids highly dilated and congested, and the liver cells displayed balloon-like degeneration. Compared with DC group, the PC and GTTE administration groups apparently alleviated the fibrosis of liver tissue and improved the morphology of the liver cells.

Fig. 2.

Effects of GTTE on alpha diversity. (a) Shannon index, (b) Simpson index, (c) Ace index, (d) Chao1 index. (e) PCoA. (f) HCA. Data are expressed as mean ± standard deviation (SD). Student's test was used to identify the significant difference between different groups. *P < 0.05, **P < 0.01 vs NC group (contained NDT vs NC, DC vs NC), #P < 0.05, ##P < 0.01 vs DC group (contained PC vs DC, TL vs DC, TM vs DC, TH vs DC).

The morphology of mice pancreas was observed under a microscope at 400 times (Supplementary Material S3 Fig. 2b). Both the NC and NDT group showed that the pancreatic acinar and pancreatic islets were normal. The pancreatic islet tissue was intact with complete morphology. The pancreatic islet cells were cord-shaped, round or oval, with abundant cytoplasm, clear borders, centered nucleus and clear borders between cells. The number of pancreatic islet cells in DC group decreased, with irregularly shaped cells and disordered and scattered cell arrangement. The edges of the cells were uneven, with blurred boundaries, enlarged cell volume, and pancreatic acinar. The PC and GTTE administration groups exhibited better repair alterations of islet cells and pancreas. Compared with DC group, the number of islet cells in the GTTE administration group increased, and the state of pancreatic tissue was significantly improved.

3.5. Effects of GTTE on islet β-cells areas

The effects of GTTE on the insulin and glucagon in the pancreas are shown in immunoreaction images in Supplementary Material S3 Fig. 3a,b,c. Compared with normal NC group, the area of islet α-cells in the DC group was significantly increased, and the area of islet β-cells was significantly reduced (P < 0.01). Compared with the DC group, the PC and TL group decreased the area of islet α-cells and increased the area of islet β-cells, but the difference was not statistically significant (P > 0.05). The TM and TH groups had significantly decreased islet α-cells area and increased islet β-cells area (P < 0.01).

3.6. Effects of GTTE on gut microbiota

3.6.1. Community diversity analysis

The gut microbiota community diversity of the seven groups’ feces samples was assessed based on alpha diversity and beta diversity. Alpha diversity reflects the richness and diversity of microbial community, including Ace index, Chao1 index, Shannon index and Simpson index (Fig. 2a,b,c,d). Among them, the Ace index and Chao1 index, as the abundance index, reflect the number of bacteria in the feces samples. The Shannon index and Simpson index reflect the index of bacterial diversity in the feces samples. The DC group elevated the Ace, Chao1, Shannon, and Simpson indexes compared with NC group, and the difference was statistically significant (P ≥ 0.05), except in Ace index. Compared with DC group, the Ace and Chao1 indexes were increased in TL, TM, TH groups. Although these differences were not statistically significant (P ≥ 0.05), but TM group was significant (P < 0.05). The indexes of Shannon and Simpson of TH group showed an upward trend, but TL group showed a downward trend (P ≥ 0.05). These trends indicated that the effects of GTTE on bacterial diversity could be relevant to its proper dosage. The upward trends indicated that GTTE administrations can increase the diversity and abundance of the gut microbiota of diabetic mice.

Beta diversity can reflect differences in the microbial communities of samples. UniFrac distance-based principal coordinate analysis (PCoA) was used to reveal the distinct clustering of intestinal microbe communities for each experimental group. The DC group was deviated from the NC group according to the microbiota composition, indicating that the composition of the gut microbiota of diabetic model mice had changed (Fig. 2e). Additionally, both the PC and GTTE administration groups had the trend closing to the NC group, and they all departed from DC group, which indicated that GTTE administration could reduce the damage of gut microbiotia of diabetic mice induced by STZ. Furthermore, TH group was more close to the NC group than that of TL and TM groups. This indicated that high doses of GTTE administration had a more significant effect on improving gut microbiota imbalance. The results of PCoA largely agreed with that of the hierarchical clustering analysis (Fig. 2f).

3.6.2. Gut microbiota composition

To further investigate the specific changes in bacterial communities, the relative abundance of the predominant phyla and genera was compared across seven groups, especially taxa responding to GTTE administration (Supplementary Material S3 Fig. 4). At the phylum level, the structures of seven groups of microbial communities were mainly composed of bacteroidetes and Firmicutes, fllowed by Actinobacteria and Proteobacteria (Supplementary Material S3 Fig. 4a). The contents of Bacteroidetes and Firmicutes in the total flora were 90%, 66%, and 42%–87% in the NC, DC, and GTTE groups, respectively. Compared with NC group, the ratio of Firmicutes to Bacteroidetes (F/B) and the abundance of Firmicutes were significantly reduced (P < 0.01, Supplementary Material S3 Fig. 4b–d). The abundance of Bacteroidetes and Actinobacteria were increased in DC group (Supplementary Material S3 Fig. 4c and d). In TL and TM groups, the F/B ratio and the abundance of Actinobacteria showed an upward trend compared with DC group (Supplementary Material S3 Fig. 4d and e). Additionally, the abundance of Bacteroidetes was increased, while Actinobacteria was decreased, along with the increasing of GTTE administration dose in treatment groups. These opposite trend maybe reflect the dosage effect of GTTE. The abundance of Proteobacteria showed no significant difference among NC, DC, TL, TM and TH (Supplementary Material S3 Fig. 4f). At the genus level, the dominant bacteria genera in NC and NDT groups were Lactobacillus, Alistipes, Ruminococcaceae UCG-014, and Psychrobacter (Supplementary Material S3 Fig. 4g). In the DC group, the dominant bacteria were Corynebacterium, Lactobacillus, Psychrobacter, and Lachnospiraceae NK4A136 group. The abundance of Corynebacterium, Lactobacillus, Lachnospiraceae NK4A136 group in GTTE administration groups were more than other bacteria. However, a little content of Corynebacterium was observed in NC and NDT groups. Compared with NC group, the abundance of Corynebacterium and Psychrobcuter were increased and the content of Lactobacillus and Alistipes reduced in DC group (Supplementary Material S3 Fig. 4h,i,j,l). The abundance of Lactobacillus and Lachnospiraceae NK4A136 group of GTTE administration groups were increased in comparison with DC group but did not reach statistical significance (Supplementary Material S3 Fig. 4i–k). Psychrobacter was increased in TL group and reduced in TL and TH group. These results showed that the changes in various gut microbiota were caused in the DC group, but the administration of GTTE can improve this trend.

4. Discussion

“Golden-flower” Tibetan tea (GTT) is an innovative dark tea fermented via fungus Eurotium cristatum. It belongs to the category of post-fermentation tea, which generates abundant phytochemicals in the fermentation stage. TB is a major component in GTTE. In recent years, studies have exhibited that TB possesses inhibitory activity on α-glucosidase in vitro, reductes blood glucose levels and improves glucose tolerance in mice under hyperglycemic stress [23], indicating its health benefits to diabetes mellitus (DM). Theaflavins show inhibitory effects on α-glucosidase and α-amylase activity resulting in decreased plasma levels of glucose [24]. Thus, tea pigments may play an important role for the antidiabetic effects of GTTE. Polysaccharides should not only directly play a role for the health benefits of the dark tea but also be used by gut microbiota to turn into short-chain fatty acids (SCFAs) via saccharolytic fermentation [25]. Polyphenols also contribute to health benefits. Although total polyphenols content was high, there were only a total of 1.32% of specific tea phenolic compounds in GTTE, including GC (0.51%), GA (0.48%), and EC (0.11%). This was in line with the literature's reports [26] that tea phenolic compounds were transformed to tea pigments or other complex in the fermentation processes of dark tea.

In addition, caffeine can increase glucose transporter IV expression [27]. A meta-analysis of prospective studies reported that caffeine intake might significantly reduce the incidence of T2DM [28]. Thus, caffeine was also a contributor to the antidiabetic effects.

The Py-GC-MS analyses results further exhibit the insight that the large complex substance, such as theabrownin, in GTTE contains phenols, acids, organic hydrocarbons, aromatic compounds, ketones, alcohols, aldehydes, and nitrogen compounds, mainly composed of phenols, lipids, sugars, and proteins.

Overall, it is reasonable to conclude that all these phytochemical components could synergistically contribute to the antidiabetic effects of GTTE.

Type 1 diabetes is caused by an absolute deficiency of insulin secretion. FBG level, OGTT, and insulin resistance are important indicators of diabetes alleviation [29,30]. In this study, at some point in the beginning, DC group exhibited more feed intake, higher level of FBG and blood glucose, and lower serum insulin compared with those of the NC group. These results were in agreement with another report [31] that STZ rat models with raised blood glucose levels coexists with either the same, lower or higher levels of insulin as compared with control. After a period of GTTE administrations, the GTTE groups reduced the elevated FBG and blood glucose level, and increased the depressive serum insulin levels. It is reported that drugs with hypoglycemic effect usually had obvious effects on reducing the FBG, improving glucose intolerance and increasing the serum insulin levels [32,33]. The reversal effects were especially evident on FBG level for TM group, on glucose tolerance for TM and TH group, on serum insulin for TM and TH group, respectively. These results indicate that GTTE could improve the symptoms of diabetes.

The STZ induced diabetic mice group (DC) showed drastic weight loss after STZ induction when compared with that of normal diet group (NC) group. This may be due to defects in glucose metabolism and excessive breakdown of tissue protein and fat [31]. However, the diabetic mice treated with GTTE (TM and TH groups) significantly prevented the body weight loss. The liver, kidney, and pancreas indexes of the TH group were significantly decreased. The organ weights could reflect the body's pathological conditions and inflammation [34]. These results indicated that the GTTE could ameliorate the decrease of body weight and alleviate the pathological changes of organs of diabetic mice.

Various studies have shown that oxidative stress plays an important role in the pathology of diabetes [35]. It is not just because of its role in the development of complications, but also because persistent hyperglycemia, secondary to insulin resistance, may induce oxidative stress and contribute to β-cell destruction in diabetes [36,37]. In this study, the antioxidant potency of GTTE was investigated via estimating the antioxidant enzymes SOD and CAT. These two enzymes can reduce oxidative stress, inhibit the proliferation of peroxisome and improve islets β-cell damage [38,39]. Compared with the DC group, the levels of SOD and CAT in the liver and blood of GTTE groups increased in varying degrees, indicating that the extract from the “Golden-flower” Tibetan tea could stimulate the defense mechanism of oxidative stress and reduce oxidative stress in vivo.

Inflammatory cytokines play an important role in the pathogenesis of insulin resistance. As the intermediate hub of the immune response, pro-inflammatory cytokines directly produce toxic effects on pancreatic β-cells and affect the body by activating the inflammatory cytokines signaling pathway [40,41]. As such, decreasing of inflammatory factors and the increasing of anti-inflammatory factors can effectively alleviate the occurrence and development of diabetes. Furthermore, disturbance of the homeostasis between the pro- and anti-inflammatory cytokines leads to T1DM [42]. In this study, the results showed that the proinflammatory cytokines levels of TNF-α were down-regulated, and the anti-inflammatory level of IL-4 was up-regulated, which indicated that the GTTE might have the potential to alleviate the development of diabetes. There was also report [43] that inhibiting the release of inflammatory factors could repair insulin signaling pathways and reduce blood sugar.

According to the histopathological analysis, diabetes causes the destruction of the pancreas [41]. In addition to the pancreas, the liver plays an important role in maintaining blood glucose balance by storing glucose as glycogen and producing glucose through glycogen degradation and gluconeogenesis [44]. In the present study, GTTE showed a favorable protective effect on the liver and repair of islet cells and pancreas. Compared with DC group, the PC and GTTE groups alleviated the fibrosis of liver tissue, improved the morphology of the liver cell, increased the number of islet cells, and improved the state of pancreatic tissue.

In order to further understand the mechanism of GTTE on the blood glucose regulation in diabetic mice, this study observed the relative positive area of islet α-cells and islet β-cells. Previous studies have shown that islet α-cells secrete glucagon, which affect glycogen breakdown and gluconeogenesis, and interact with insulin secreted by islet cells to maintain the body's blood sugar balance [45]. The body's high blood sugar will destroy the islet β-cells, increase glucagon secretion and decrease insulin secretion [29]. After 4 weeks of administrations, TM and TH groups significantly increased the area of pancreatic islet β-cells and decreased the area of pancreatic islet α-cells (P < 0.05), which should be led by the continued low blood sugar levels to protect their islet β-cells.

The gut microbiota plays a vital role in the physiology and pathology of the host, especially associated with the pathogenesis of metabolic diseases such as obesity, insulin resistance and diabetes [46]. Nowadays, more and more evidence has shown that dietary factors profoundly modulate gut microbiota, which further mediates the biological function of the dietary factors [47]. Therefore, the 16S rDNA amplicon sequencing was performed to analyze the changes in gut microbiota of different GTTE intervention groups. The upward trends of alpha diversity in GTTE groups reflect that GTTE administration could increase the diversity and abundance of gut microbiota of mice. Similarly, beta diversity analysis demonstrated a clear separation of the DC group from the NC group. The GTTE group also deviated from the DC group, indicating that GTTE could regulate the intestinal microbial community to develop toward the NC group.

Results showed that gut microbiota composition was improved. Bacteroidetes and Firmicutes are the dominant flora, followed by Actinobacteria and Proteobacteria. Firmicutes and Bacteroidetes can convert indigestible carbohydrates into short-chain fatty acids (SCFAs) in the intestines, mainly propionic acid and butyric acid [48]. SCFAs are a class of beneficial metabolites produced by gut microbiota. The deficiency of SCFAs could be associated with metabolic diseases, including obesity and diabetes [49]. The Firmicutes to Bacteroidetes (F/B) ratio is associated with the development of metabolic disorder. Bacteroidetes was reported in children who developed T1DM [50]. Patients with T1DM were also found the high abundance of Bacteroides [51,52]. In present study, the relevant abundance of Bacteroides was 28%, 37% and 14%, in NC, DC, and TL groups, respectively. This indicates that low dose GTTE administration induced a lower abundance of Bacteroides. Additionally, the F/B ratio decreased in DC group, and increased in TL, and TM groups. Giongo et al. [53] reported that the F/B ratio was decreased with the lapse of time in the T1DM cases. Previous research has found that the higher F/B ratio was conducive to the decomposition of polysaccharides [54]. Furthermore, the relevant abundance of Bacteroidetes increased and Actinobacteria (related with a series of inflammatory reactions [55]) reduced as with the increasing of GTTE dosage. This dose effect indicated that different dosages of GTTE might have different alleviating effects on diabetes. Proteobacteria includes a series of pathogenic bacteria such as Salmonella and Vibrio cholerae, which may cause disturbance of the intestinal microflora and affect the health of the host [56]. Here, the relevant abundance of Proteobacteria was 7.6%, 7.2%, 5.7%, 2.8% and 2.9% in NC, DC, TL, TM and TH groups, respectively. The results indicate that GTTE administration can inhibit the proliferation of Proteobacteria. In the study, the GTTE with different dosages exhibited different effect on the same bacteria. When compared with DC group, the relevant abundance of Corynebacterium 1 and Psychrobacter was increased in TL group, and decreased in TH group. These results further suggested that the dosage of GTTE was an impact factor of the effects on the alleviation of diabetes.

The study shows comprehensive insight of GTTE phytochemical compositions. GTTE showed significant effects to alleviate the general conditions of T1DM mice. As discussed above, many of the compositions may synergistically contribute to the antidiabetic effects. Systematic mechanism was not studied in this paper. Network pharmacology approach will be useful to understand the complex pharmacological mechanisms of GTTE. The alleviating effects of GTTE to T1DM mice model can only be considered preliminary results. Other T1DM animal model, such as BB rat, could be used for further studies.

5. Conclusions

Extracted from “golden-flower” Tibetan tea, GTTE contains pigment theabrownin (TB) and a high percentage of tea polysaccharide, protein, polyphenols, amino acids, caffeine, and a small amount of flavonoids. GTTE can effectively alleviate the conditions of T1DM mice induced by STZ, i.e., reducing their FBG, elevating body weight, improving glucose intolerance, increasing serum insulin levels, relieving liver and pancreatic injury, restoring damaged islet cells. GTTE also improves the gut microbiota of diabetic mice by increasing the diversity and abundance of the gut microbiota, modulating gut microbiota composition by resisting changes of flora structure, increasing the abundant of beneficial bacteria such as Firmicutes, increasing the proper Firmicutes/Bacteroidetes (F/B) ratio, and reducing harmful bacteria content such as Proteobacteria. The findings suggest multi-channel mechanisms could be involved in the synergistic effects on T1DM. The pathways include decreasing oxidative stress by increasing SOD and CAD levels, modulating cytokine expression leading to the decreasing pro-inflammatory cytokines TNF-α and IL-6, increasing anti-inflammatory cytokines IL-4 to improve inflammatory responses, and improving of gut microbiota. Further study of the mechanisms is needed in the future works. These findings not only provide meaningful understandings for the health benefits of long term daily drinking “golden-flower” Tibetan tea but also provide scientific basis for the research and development of functional foods and dietary supplements originating from the “golden-flower” Tibetan tea.

Ethics statement

The study was approved by the Institutional Animal Care and Use Committee of Sichuan University (KS20201023).

Consent for publication

All authors consent for publication.

Funding statement

This study was supported by the grant from Key Research and Development Program of Science and Technology Foundation of Sichuan Province, China (No. 2019YFN0178, No.2023YFN0012); 1 + 9 Unveiling of for tackling key scientific and technological problems - functional food core technology, Sichuan Academy of Agricultural Sciences (1+9KJGG007), China.

Data availability statement

Data available within the article and its supplementary materials.

CRediT authorship contribution statement

Junlin Deng: Writing – original draft, Visualization, Formal analysis. Kebin Luo: Writing – original draft, Software, Methodology, Data curation. Chen Xia: Supervision, Methodology, Investigation, Funding acquisition. Yongqing Zhu: Resources, Investigation, Funding acquisition. Zhuoya Xiang: Software, Formal analysis. Boyu Zhu: Software, Formal analysis. Xiaobo Tang: Software, Formal analysis. Ting Zhang: Software, Formal analysis. Liugang Shi: Resources. Xiaohua Lyu: Supervision, Conceptualization. Jian Chen: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Jian Chen reports financial support was provided by Sichuan Science and Technology Agency, China. Chen xia reports financial support was provided by Sichuan Science and Technology Agency, China. Yongqing Zhu reports financial support was provided by Sichuan Academy of Agricultural Sciences, China. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: