Drinks preparation
The orange flavor occupies the first position with 26.2% of the market share [65]. Hence, we chose 2 kinds of popular drinks from the local market as the objects of our study. For FB, its label claims no less than 10% orange juice content. For FJ, its label claims 100% orange juice content. All drinks samples were stored at room temperature and shaken before use.
Measurement of 3 primary free sugar in drinks
The contents of glucose, fructose, and sucrose were determined according to the previous method with some modifications [66]. The juice was diluted 5000 times and filtered using a 0.22-µm polytetrafluoroethylene membrane. Ion chromatography (ICS 3000+, Thermo Fisher Scientific, USA) system, consisted of (i) an ICS-3000 dual pump, (ii) an automated sample injector (Dionex AS autosampler), (iii) working electrode: Gold (Au), reference electrode: pH electrode (Dionex, Germany), (iv) PA20 anion analysis column (3 mm × 150 mm), (vi) PA20 guard column (4×50 mm), was used. Mobile phase A was water and B was 200 mM NaOH. The elution procedure was 0 − 20.00 min, 5% B; 20.00 − 20.01 min, 5%−20% B; 20.01 − 30.00 min, 20% B; 30.00 − 30.01 min, 20%−100% B; 30.01 − 40.00 min, 100% B; 40.00 − 40.01 min, 100%−5% B; 40.01 − 50.00 min, 5% B. The injection volume, flow rate, and column temperature were 10 µL, 0.400 mL/min, and 30°C, respectively.
Animal experiment
Six- to eight-week-old SD (Rattus norvegicus) rats (male; SPF; the mean weight of 209 g) were purchased from Vital River Laboratory Animal Technology Co. Ltd (Beijing, China). The animal feeds were purchased from Keao Xieli Feed Co., Ltd (Beijing, China), and the primary nutrients of animal feed are shown in Table S4. The rats were kept under controlled light conditions (12 h light-dark cycle) with free access to feed and water throughout the entire intervention.
The design of drink intervention study is shown in Figure S5. The rats were acclimatized for 7 days, and the weight, fasting blood glucose (FBG), total cholesterols (TC), total triglycerides (TG), high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were recorded on the last day of acclimatization. When the rats were adapted to the experiment condition during the first week, body weight, FBG, TC, TG, HDL-C, and LDL-C were measured to divide these rats into 4 groups, denoted as C0, control, FJ-intake, and FB-intake, respectively. After grouping, there are no significant differences in these indexes among groups (Figure S6).
For C0, cecum, and serum samples were directly collected after 7 days of acclimatization and stored at -80°C until analysis. Then, rats were dissected to obtain kidney, brain, liver, fat, colon, intestine, and cecum. The organ mass was recorded and the organs were stored at -80°C until analysis.
The control and 2 intervention groups (control, FJ-intake, and FB-intake) proceeded with formal intervention for 28 days. Under the regular diet, the rats were gavaged with normal saline, FJ, and FB for control, FJ-intake, and FB-intake, respectively (twice per day). Each rat was given the equivalent of one adult (60 kg) drinking 300 mL of drinks a day. The specific volume of gavage was calculated as 31.5 times the volume of each rat's last measured body weight (for example, if the last measured weight of a rat is 0.2 kg, its gavage volume is 6.3 mL per day). During the 28-day diet, the weight, FBG, TC, TG, HDL-C, and LDL-C were recorded once per week. On the last day, samples from the 3 groups were collected and stored under the same condition as C0. The organ index was calculated based on the ratio of organ mass to body weight.
Gut microbiota analysis
Eight rats were randomly selected from each group for 16S rRNA sequencing. The corresponding cecum sample from the chosen rat was snap-frozen with liquid nitrogen following collection and was stored in a -80°C refrigerator. Then, 32 cecum samples were transported to Majorbio Bio-pharm Technology Co., Ltd (Shanghai, China) on dry ice.
The genomic DNA of cecum samples was extracted using the DNA Extract Kit (Omega Bio-Tek, USA). The DNA was checked on 1% agarose gel, and the concentration and purity of DNA were determined using NanoDrop 2000 UV-vis spectrophotometer (Thermo Scientific, USA). The 16S rRNA genes of V3–V4 regions were amplified with specific primers (338F and 806R) using ABI GeneAmp® 9700 PCR thermocycler (ABI, USA). After initial denaturation at 95°C for 3 min, 30 cycles of amplification were performed as follows: 95°C for 30 s, 55°C for 30 s, and 72°C for 45 s followed by a single extension at 72°C for 10 min to amplify the 16S rRNA gene. The PCR product was verified on a 2% agarose gel and purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, USA) and quantified using Quantus™ Fluorometer (Promega, USA). Purified amplicons were pooled into equimolar and paired-end sequenced on an Illumina MiSeq PE300 platform (Illumina, USA). The raw 16S rRNA gene sequencing reads were demultiplexed, quality-filtered by fastp version 0.20.0, and merged by FLASH version 1.2.7. Operational taxonomic units (OTUs) with a 97% similarity cutoff were clustered using UPARSE version 7.1, and chimeric sequences were identified and removed. The taxonomy of each OTU representative sequence was analyzed by RDP Classifier version 2.2 against the Silva v138 16S rRNA database using a confidence threshold of 0.7.
Metabolite analysis
The ultrahigh-pressure liquid chromatography coupled with a quadrupole time-of-flight mass (UHPLC-QTOF MS) was performed. The samples were subjected to a reversed-phase chromatography using an LC system (SCIEX, USA). Mass spectrometric analysis was performed with a QTOF MS with a DuoSpray ion source (TripleTOF 6600, SCIEX, USA). Quality control (QC) samples were made by pooling equal aliquots of each sample together in an analytical run [67]. The QC sample was injected and analyzed at regular intervals of every 6 samples in UHPLC-QTOF MS acquisition. These QC sample data were used to monitor the reproducibility and stability of the analytes in samples during the analysis [68].
Drinks
The metabolomics was conducted on 2 drinks according to the previous method with modifications [69]. The 2 drinks were filtered using a 0.22-µm polytetrafluoroethylene membrane and 2 drinks with 6 replicates each were obtained. The separation column is ACQUITY UPLC HSS T3 (1.8 µm, 2.1×100 mm, Waters, USA), maintained at 40°C, and the autosampler was maintained at 4°C. The injection volume and flow rate were 2 µL and 0.300 mL/min. Mobile phases A and B were 0.2% formic acid in water and acetonitrile, respectively. Solvent gradient was as follow: 0 − 11.50 min, 5%−30% B; 11.50 − 11.51 min, 30 − 100% B; 11.51 − 15.00 min, 100% B; 15.00 − 15.01 min, 100%-5% B; 15.01 − 18.00 min, 5% B. The SWATH with a cycle time of 545 ms was composed of a TOF MS scan (accumulation time, 50 ms; CE, 10 eV) and a series of product ion scans (accumulation time, 30 ms each; CE, 35 eV, CES 15 eV) of 15 Q1 windows of 60 Da from m/z 100–1000.
Kidney lipidomic samples
The lipidomics was conducted on kidney samples according to previous methods with modifications [70–72]. Thirty-mg frozen kidney tissue was weighed and homogenized in 600-µL extract (MTBE: MeOH = 5:1, v/v) using the homogenizer, and 1-µL internal standard of Cer-d7 (100 µg/mL) was added. The solution was vortexed for 30 s, followed by 10 min of sonication, and centrifugation at 3000 rpm for 15 min. The upper organic layer was collected. An additional 600-µL MTBE was added to the bottom layer for re-extraction. The re-extraction was repeated twice. The pooled organic layer was evaporated using a termovap sample concentrator. Before subsequent analysis, the dry extract was reconstituted using 100-µL DCM: MeOH (1:1, v/v). The separation column is Phenomenex Kinetex C18 column (1.7 µm, 2.1×100 mm, Waters, USA), maintained at 55°C, and the autosampler was maintained at 4°C. The injection volume was 2 µL and 6 µL in positive and negative modes, respectively. The mobile phase consisted of A: 5 mM H3CCOONH4 + 40% H2O + 60% CAN, B: 5 mM H3COONH4 + 10% ACN + 90% IPA was carried with elution gradient as follows: 0 min, 40% B; 12 min,100% B; 13.5 min, 100% B; 13.7 min, 40% B; 18 min, 40% B, which was delivered at 0.3 mL/min. The Triple TOF mass spectrometer was used for its ability to acquire MS/MS spectra on an IDA mode. The acquisition software (Analyst TF 1.7, AB SCIEX) continuously evaluates the full scan survey MS data in this mode. It collects and triggers the acquisition of MS/MS spectra depending on preselected criteria. In each cycle, 11 precursor ions whose intensity greater than 100 were chosen for fragmentation at collision energy (CE) of 35 V (15 MS/MS events with product ion accumulation time of 50 ms each). ESI source conditions were set as follows: Ion source gas 1 as 60, Ion source gas 2 as 60, Curtain gas as 30, source temperature 550℃, Ion Spray Voltage Floating (ISVF) 5500 V or − 4500 V in positive or negative modes, respectively.
Cecum content samples
The cecum content samples were performed according to previous methods with modifications [73, 74]. And 300-mg of cecum contents was weighed with 1.5-mL methanol-water (v:v = 1:1, 4°C) solution added. The mixture was vibrated for 30 s, frozen in liquid nitrogen for 30 s, then thawed in water at room temperature, and repeated 3 times. After the last thawing, an ultrasound was performed for 30 s followed by 10-s vibration and 30-s standing; then, the process was repeated 5 times. Centrifugation was performed at 15000 g for 5 min at 4°C. Finally, the supernatant was filtered using a 0.22-µm nylon membrane for analysis. The separation column is C18 (1.7 µm, 2.1×100 mm, Waters, USA), maintained at 40°C, and the autosampler was maintained at 4°C. The injection volume and flow rate were 2 µL and 0.300 mL/min. Mobile phases A and B were 0.1% formic acid in water, and 0.1% formic acid in acetonitrile, respectively. For cecum samples, solvent gradient was as follow: 0 − 0.10 min, 5% B; 0.10 − 2.00 min, 5 − 13% B; 2.00 − 4.00 min, 13 − 28% B; 4.00 − 8.50 min, 28 − 40% B; 11.50 − 13.00 min, 40 − 70% A; 13.00 − 13.50 min, 73 − 100% B; 13.50 − 16.00 min, 100% B; 16.00 − 20 min, 5% B. The SWATH with a cycle time of 550 ms was composed of a TOF MS scan (accumulation time, 50 ms; CE, 10 eV) and a series of product ion scans (accumulation time, 30 ms each; CE, 35 eV, CES 15 eV) of 15 Q1 windows from m/z 50-1200.
Serum content samples
The serum content samples were performed according to a previous method with modifications [75]. The serum was thawed in a refrigerator at 4°C. After thawing, the serum was rotated for 5 s. The 400-µL methanol was added to 200-µL serum, and the mixture was rotated for 10 s and stood at room temperature for 15 min. Then, centrifugation was performed at 10000 g for 20 min at 4°C. Finally, the supernatant was filtered using a 0.22-µm nylon membrane for analysis. Totally, 120 samples of the cecum and serum samples (60 samples each) from all the rats were obtained. The separation column is C18 (1.7 µm, 2.1×100 mm, Waters, USA), maintained at 40°C, and the autosampler was maintained at 4°C. The injection volume and flow rate were 2 µL and 0.300 mL/min. Mobile phases A and B were 0.1% formic acid in water, and 0.1% formic acid in acetonitrile, respectively. For serum samples, solvent gradient was as follow: 0 − 0.10 min, 5% B; 0.10 − 1.50 min, 5 − 13% B; 1.50 − 2.50 min, 13 − 28% B; 2.50 − 5.50 min, 28 − 40% B; 5.50 − 7.50 min, 40 − 70% A; 7.50 − 15.00 min, 70 − 90% B; 15.00 − 16.00 min, 90 − 100% B; 16.00 − 19.00 min, 100% B; 19.01-22.00 min, 5%B. The SWATH with a cycle time of 550 ms was composed of a TOF MS scan (accumulation time, 50 ms; CE, 10 eV) and a series of product ion scans (accumulation time, 30 ms each; CE, 35 eV, CES 15 eV) of 15 Q1 windows from m/z 50-1200.
Data processing and statistical analysis
Mean values with relative standard deviation (RSD) were calculated, and statistical analyses were performed using R 4.0.3 (Inc., MA, USA). To compare glucose, sucrose, fructose contents, and the fructose-to-glucose ratio in 2 drinks, the student’s t-test was conducted. For comparing food intake, FBG, TC, TG, HDL-C, LDL-C, body weight change, and organ mass index in rats from different groups, ANOVA followed by the Bonferroni test was conducted.
For gut microbiota analysis, the alpha diversity (ACE, Chao, and Sobs estimator) and beta diversity of PCoA based on weighted UniFrac distances were analyzed using R 4.0.3. To compare alpha diversity, the Kruskal-Wallis rank-sum test followed by Game-Howell was conducted. Linear discriminant analysis (LDA) effect size (LEfSe) was used to identify the microbial taxa that significantly differed between FJ-intake and FB-intake groups, with a threshold of LDA score > 2.0. Functional pathway analysis of gut microbiota was performed using the PICRUSt2 approach, based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database with the 16S rRNA gene sequencing data (Majorbio Cloud Platform). Differences in pathways between FJ-intake and FB-intake groups were identified using LEfSe with a threshold LDA score of > 2.0.
For metabolite analysis, UHPLC-QTOF MS data were processed by MS-DIAL 4.60 to get the MS1 raw peak list and intensity. The parameters were set as follows: MS1 and MS2 tolerance were 0.01 Da and 0.025 Da in peak extraction, and the maximum charge number was 2. MS1 and MS2 tolerance were 0.01 Da and 0.05 Da in qualitative extraction, and the retention time deviation in peak alignment was 0.2 min. The MS1 deviation was 0.015 Da, the detection rate reached 80% in at least one group, and the S/N ratio (maximum in the sample/mean in the blank sample) was 5. The exported MS1 peak list was analyzed by Microsoft Office Excel 2020 (Microsoft Corporation, USA) to remove the MS1 features with a detection rate was < 80% or RSD > 30% in the QC group [76]. For kidney lipidomics samples, the MS1 features with p-value < 0.05 and fold change (FC) > 1.2 were selected as characteristic markers for further identification with a built-in lipid database in MS-DIAL [77]. For drinks, the MS1 features with p-value < 0.05, FC > 5, and variable importance for projection (VIP) > 1.5 based on the orthogonal partial-least squares discriminant analysis (OPLS-DA) model were selected as characteristic markers for further identification. For cecum and serum samples, all the MS1 features were further identified. Specifically, the MS1 features that met the above conditions were identified by exporting their MS2 spectra from MS-DIAL 4.60 into MS-FINDER 3.50 for putative annotation by matching the databases, such as HMDB, lipid maps, PubChem, and FooDB. The parameters, MS1 tolerance, and MS2 tolerance were set as 10 ppm and 15 ppm, respectively.
For correlation analysis, spearman correlations were conducted on GraphPad Prism 8 (GraphPad Software, USA). The correlation with coefficient > 0.5 and p-value < 0.05 was considered a significant correlation and then plotted by Microsoft Office Excel 2020 or Cytoscape (v3.4.0).