Browning Effects of a Chronic Pterostilbene Supplementation in Mice Fed a High-Fat Diet

To define the most suitable dose of pterostilbene to be used during in vivo experiments, we performed preliminary tissue distribution experiments. Mice were fed with a high fat diet (HFD) supplemented with three different dosages of pterostilbene for 2 weeks (see Materials and Methods for details). The higher dosage (352 µmol/kg/day, corresponding to 90 mg/kg/day) was selected to be the most suitable one, since it allowed the achievement of µM levels of the compound in all major organs and also in both epididymal and inguinal adipose tissues (Figure 1 and Table S1).

The ability of Pt to increase the expression of (at least some) browning genes was tested in vitro on cultured 3T3-L1 adipocytes; cells were grown and differentiated following standard experimental protocols [9]. On day 12 they can be considered fully differentiated and they showed typical accumulation of large lipid droplets in the cytoplasm. Mature 3T3-L1 adipocytes were incubated for 24 h with 5 µM pterostilbene, which is the mean concentration we observed in both epididymal and inguinal adipose tissues after chronic administration of 352 µmol/kg/day Pt to mice.

RT-qPCR of 3T3-L1 adipocytes showed an increase in the transcription of Cidea and Fgf21 and a decrease in PPARγ after Pt treatment. Ucp1 transcript levels were not affected (Figure 2), but UCP1 protein levels were increased (Figure 3).

Weight gain of the mice receiving the HFD from weaning was higher compared to that of mice fed a standard diet (p ≤ 0.05 at all time points for both females and males). Notably, the administration of Pt slowed down weight increase; the effect was somewhat different between males and females. Both in males and females, for the first 8 weeks of treatment the body weight increase of the HFD + Pt group was not significantly different from that of the standard diet (STND) group. From week 9 (and until the end of the experiment), the weight increase of HFD + Pt groups became significantly higher than that of STND mice (p ≤ 0.01 and p ≤ 0.05 at all time points for males and females, respectively), but for males the curve remained also significantly lower than the HFD group (p ≤ 0.002 at all time points) (Figure 4a). On the other hand, the body weight increase of HFD + Pt females became more similar to that of the HFD group (differences were not significant at weeks 9, 10, 12–15, 21, and 22) (Figure 4b).

The glucose tolerance test was performed in fasted animals after 18 and 28 weeks of treatment. Altered glucose homeostasis was observed in both males and females of the HFD group at both times (Figure 5). Pt treatment slightly improved glucose homeostasis in obese mice, but only after 18 weeks of treatment. For both males and females, glucose levels at 120 min were not significantly different from those measured in the corresponding STND groups (Figure 5a,b). At 28 weeks this effect was no longer observed, and no significant differences were detected between HFD and HFD + Pt groups (Figure 5c,d).

The insulin tolerance test was also performed at 18 and 28 weeks. No significant effects were associated with Pt supplementation in any case (). Figure S1

We analyzed the effect of Pt treatment on inguinal WAT since this is known from literature to be the WAT depot most prone to undergo browning in mice [30]. RT-qPCR analysis showed a general increase of browning markers in the HFD + Pt group, compared to the HFD one: a significant increase (p ≤ 0.05) in transcript levels was observed for Cidea, Ebf2, Pgc1a, PPARγ, Sirt1, and Tbx1 (Figure 6a). This trend towards browning was also observed considering data from males and females separately (Figure 6b,c). In accordance with our in vitro experiments, Ucp1 mRNA levels did not change with Pt supplementation, while its protein expression seems to be upregulated (p = 0.09; Figure 7a). Interestingly, if data from males and females are considered separately, Pt seems to affect females (p = 0.07; Figure 7) more than males (p = 0.62; Figure 7). Histological analysis on a representative set of samples showed a slight decrease in adipocyte size in the HFD + Pt group compared to the HFD one (Figure 8).

3T3-L1 preadipocytes were cultured in DMEM containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (10,000 U/mL). Two days after confluence (day 0 of differentiation), the cells were stimulated to differentiate by the addition into the culture medium of 1% biotin, 1% panthothenic acid, 10 μg/mL insulin, 0.5 mM isobutylmethylxanthine (IBMX), and 1 μM dexamethasone. After 2 days, the differentiation medium was replaced by a medium containing FBS, penicillin, streptomycin, 1% biotin, 1% pantothenic acid, and 10 μg/mL insulin, and from day 4 onward by the same medium containing 0.2 μg/mL insulin (instead of 10 μg/mL). This medium was changed every 2 days until cells were harvested. Cells were maintained at 37 °C in a humidified 5% CO₂ atmosphere. For the 3T3-L1 adipocyte treatment, cells grown in 6-well plates were incubated with pterostilbene 5 μM (0.1% final DMSO) for 24 h on day 12 after differentiation. By that day, >90% of cells were mature adipocytes with visible lipid droplets.

C57BL/6 mice were housed in the facility of the Department of Pharmaceutical and Pharmacological Sciences (University of Padova, Padova, Italy); food and water were provided ad libitum. All procedures were approved by the University of Padova Ethical Committee for Animal Welfare (OPBA) and by the Italian Ministry of Health (Permit Number 211/2015-PR, approved on 7 April 2015), and conducted with the supervision of the Central Veterinary Service of the University of Padova, in compliance with Italian Law DL 26/2014, embodying UE Directive 2010/63/EU (approved on 14 March 2014).

Preliminary tissue distribution studies were performed to assess the most suitable dose of pterostilbene to be used. Mice were fed for 2 weeks with a high-fat diet (60% calories from fat; OpenSource Diets, cat. n. D12492) supplemented with pterostilbene (Wonda Science, Changzhou, China; cat. n. 25992) (88, 176, or 352 µmol/kg body weight/day, respectively). At the end of the treatment, animals were anesthetized with isoflurane and sacrificed (n = 4 for each condition). Blood was collected in heparinized tubes, kept in ice and treated within 10 min. Skeletal muscle (gastrocnemius), liver, brain, kidneys, visceral adipose tissue, and subcutaneous adipose tissue were explanted, weighed, and immediately frozen in liquid nitrogen. Pterostilbene, pterostilbene-sulfate, and pterostilbene-glucuronide were extracted from blood and major organs (skeletal muscle, liver, brain, and kidneys) as described in [5]. Recovery yields of pterostilbene-glucuronide from blood and tissues were assumed to be the same as for pterostilbene-sulfate [5]. Adequate recovery yields from adipose tissue were obtained modifying this extraction protocol, as described in [53]. Analytes in the extracts were finally quantified through HPLC/UV analysis (see below, Section 4.4).

Samples (2 µL) were analyzed by HPLC/UV (1290 Infinity LC System, Agilent Technologies, Cernusco sul Naviglio (Milano), Italy) using a reverse phase column (Zorbax Extend-C18, 1.8 µm, 50 × 3.0 mm i.d.; Agilent Technologies, cat. n. 727975-302) and a UV diode array detector (190–500 nm). Solvents A and B were water containing 0.1% trifluoroacetic acid (TFA) and acetonitrile, respectively. The gradient for B was as follows: 10% for 0.5 min, then from 10% to 100% in 3.5 min, 100% for 1 min; the flow rate was 0.6 mL/min and the column compartment was maintained at 35 °C. Eluate was preferentially monitored at 286, 300, and 320 nm (corresponding to absorbance maxima of the internal standard, derivatives/metabolites, and pterostilbene, respectively).

After weaning (about 5 weeks from birth), C57BL/6 mice were randomly divided into three 16 member groups, each containing an equal number of males and females: STND, receiving a standard diet; HFD, receiving a high-fat diet (60% calories from fat; OpenSource Diets, New Brunswick, NJ, USA; cat. n. D12492); HFD + Pt, receiving a high-fat diet supplemented with pterostilbene (352 µmol/kg body weight/day). Mice were maintained under these three different dietetic regimens for 30 weeks.

Body weight was measured weekly; glucose and insulin tolerance were monitored twice (after 18 and 28 weeks; seeand) in order to evaluate the establishment of the diet-induced obesity phenotype and the efficacy of the Pt treatment. Section 4.6 Section 4.7

After 30 weeks, animals were fasted for 4 h and then sacrificed. Inguinal adipose tissue was collected, immediately frozen in liquid nitrogen, and then stored at −80 °C.

GTT and ITT were performed after 18 and 28 weeks of treatment. Animals were fasted for 6 h prior to the assay. Blood was collected from the tail vein, and blood glucose levels were measured with an Accu-Chek glucometer (Roche Diagnostics, Monza, Italy) at time 0 and 15, 30, 45, 60, 90, and 120 min after an intraperitoneal injection of D-glucose (1 g/kg) or of bovine insulin (0.75 U/kg; from Sigma-Aldrich, Milano, Italy; cat. n. I0516).

The frozen inguinal adipose tissue from control and treated mice was pulverized in liquid nitrogen using a mortar and a pestle, and the resulting frozen powder was divided evenly into two test tubes, for RNA and protein extraction respectively.

Total RNA was extracted using TRIzol reagent (Thermo Fisher, Waltham, MA, USA) according to the manufacturer’s instructions, adding a centrifugation step (5 min at 12,000× g at 4 °C) after TRIzol lysing to eliminate most of the fat content of the tissue.

Protein extraction was performed by resuspending 3T3-L1 cells (washed twice with PBS) or the frozen tissue powder in RIPA lysis buffer containing protease inhibitors, incubating for 30 min on ice, and then physically disaggregating the tissue with a potter. The lysate was then centrifuged at 20,000× g for 15 min at 4 °C to remove fat and transferred to a new tube.

From the adipose tissue, 400 ng of total RNA was retrotranscribed with the Superscript VILO reverse transcriptase kit (Thermo Fisher) according to the manufacturer’s protocol, and the expression levels of the genes of interest were analyzed by quantitative real time PCR, using the iQ SYBR Green Supermix (Bio-Rad, Segrate, Italy) kit on a Bio-Rad CFX96 Real-Time System and C1000 Touch Thermal Cycler instrument. All samples were run in triplicate, and the housekeeping gene GAPDH was used as an internal control to normalize the expression levels, applying the 2^−ΔΔCt method [54]. The primers used were designed using the Primer3 software, and all the sequences can be found in Table S2.

Proteins were separated by SDS-PAGE (Pre-cast NuPAGE Bis-Tris Gels, Invitrogen-Life Technologies, Monza, Italy). After electrophoretic separation, proteins were transferred to PVDF membranes (immobilion-FL). The membranes were saturated with 5% BSA (Sigma-Aldrich) in TBS-T buffer (TBS supplemented with 0.1% Tween-20; Sigma Aldrich) for 1 h and then incubated overnight at 4 °C with the primary antibody. Primary antibodies used (all from Cell Signaling, Danvers, MA, USA) were anti-β-actin (#8457S) and anti-UCP1 (#14670). Secondary goat anti-rabbit antibody (Cell Signaling, #7074) was horseradish peroxidase-conjugated and was used with chemiluminescence detection (Pierce) using digital imaging by a UVITEC Eppendorf apparatus.

Immediately after excision, adipose tissue samples were fixed in 4% paraformaldehyde (overnight, 4 °C). They were then sequentially transferred in a 30% ethanol solution (EtOH; 30 min), followed by 50% EtOH (30 min), then 70% EtOH; samples were then stored in a 70% EtOH solution at 4 °C, until paraffin embedding.

Paraffin embedding: samples were sequentially transferred in 70% EtOH (30 min), 95% EtOH (1 h), 100% EtOH (1 h; twice), xylene (1.5 h; twice), then paraffin (1.5 h, 60 °C; four times), and they were finally embedded in paraffin.

Tissue was then sliced into 8 µm thick sections with a microtome, mounted on coverslips, and stained with hematoxylin/eosin.

Hematoxylin/eosin staining: slides were sequentially immersed in xylene (15 min), 100% EtOH (10 min), 95% EtOH (10 min), 70% EtOH (10 min), 50% EtOH (10 min), distilled water (5 min), ematoxylin (10 min), tap water (10 min), distilled water (2 min), 1% eosin (5 min), tap water (5 min), 50% EtOH (2 s), 70% EtOH (2 s), 95% EtOH (2 s), 100% EtOH (2 s), xylene (10 min), and were finally mounted and examined with a Leitz Dialux 22 microscope, equipped with an OPTIKA microscopes C-P8 camera, using a 10X objective.

Statistical analysis was performed using GraphPad Prism Software, version 6. Significance in comparisons was assessed using Student’s t-test (changes in gene expression or protein levels) or two-way ANOVA for repeated measures (body weight increase, GTT curves). Tukey’s test was used for multiple comparisons.