Modulation of Adipocyte Size and Fat Pad Weight via Resveratrol Releasing Scaffolds Implanted into the Epididymal Adipose Tissue

Abstract

Lipid overload of the adipose tissue, which can be caused by overnutrition, underlies metabolic disease. We hypothesized that increasing the energy demand of adipose tissue is a promising strategy to combat excessive lipid accumulation. Resveratrol, a natural polyphenol, activates lipid catabolism in fat tissue; however, its clinical success is hindered by poor bioavailability. Here, we implanted resveratrol releasing poly(lactide-co-glycolide) scaffolds into epididymal fat to overcome its poor bioavailability with the goal of enhancing local lipid catabolism. In lean mice, resveratrol scaffolds decreased adipocyte size relative to scaffolds with no drug, a response that correlated with AMP kinase activation. Immunohistochemistry indicated that macrophages and multinucleated giant cells within the scaffold expressed carnitine palmitoyltransferase 1 (CPT1) at higher levels than other cells in the adipose tissue. Furthermore, resveratrol increased CPT1 levels in cultured macrophages. Taken together, we propose that resveratrol scaffolds decrease adipocyte size because resveratrol increases lipid utilization in scaffold-infiltrating immune cells, possibly through elevating CPT1 levels or activity. In a follow-up study, mice that received resveratrol scaffolds 28-days prior to a high fat diet exhibited decreased weight gain, adipose tissue expansion, and adipocyte hypertrophy compared to mice with control scaffolds. Notably, this scaffold-based strategy required a single resveratrol administration compared to the daily regiment generally needed for oral administration. These results indicate that localized delivery of metabolism modulating agents to the adipose tissue may overcome issues with bioavailability and that the role of biomaterials should be further investigated in this therapeutic strategy for metabolic disease.

INTRODUCTION

The adipose tissue functions as the body’s main energy reservoir and plays a central role in maintaining whole body energy homeostasis¹. Adipocytes, the tissue’s parenchymal cell, accomplish this by storing excess lipids in the form of triglycerides in specialized lipid droplets². As other organs in the body require energy, triglycerides are broken down into fatty acids (via lipolysis), released into the bloodstream and are ultimately taken up by the energy demanding organ³. However, the adipose tissue has a limited storage capacity; therefore, chronic overnutrition can lead to an accumulation of lipid metabolites in other tissues^4,5. Furthermore, lipid accumulation in the adipose tissue has significant consequences including adipose tissue inflammation and dysregulated insulin signaling, which are associated with metabolic diseases such as obesity and type 2 diabetes^5,6.

Current methods to mitigate this lipid accumulation include lifestyle changes, pharmaceuticals, and bariatric surgery⁷. Each of these strategies pose challenges that hinder their effectiveness. Diet and exercise alone do not typically result in long term outcomes for those with severe obesity⁸ and FDA approved drugs that improve lipid storage in the adipose tissues, for example thiazolidinediones, are complicated by off-target effects including cardiotoxicity^9,10. Finally, bariatric surgery is highly invasive and comes with high risks of post-operative or chronic complications¹¹. Therefore, a strategy that enhances lipid metabolism in the adipose tissue and prevents the side effects associated with systemically administered drugs that can also be combined with diet and exercise would be highly beneficial to combat lipid accumulation.

Biomaterials have been extensively studied to repair injured skin, bone, and muscle; however, only in recent years have researchers addressed adipose tissue with tissue engineering strategies^12-16. Our recent studies indicate that, upon implant into the epididymal adipose tissue, scaffolds made from poly(lactide-co-glycolide) (PLG) initiate remodeling of the host tissue that shares many characteristics of wound healing, including recruitment of immune cells and extracellular matrix deposition¹⁶. This remodeling serves to degrade and integrate the scaffold with the host tissue. To our knowledge, no one has studied the metabolic consequences on the adipose tissue after biomaterial implant; however, we anticipate that this response is energetically demanding, and fueled, in part, by the metabolism of lipids^17,18. In fact, our previously published results indicate that the host response to PLG scaffolds is responsible for enhancing local lipid utilization as evidenced by a decrease in epididymal fat weight in both healthy mice and mice challenged with a high fat diet¹⁶.

In the current study, we investigated the hypothesis that the scaffold’s effects on local lipid levels could be enhanced through the incorporation of resveratrol into the scaffold polymer matrix. Resveratrol, a natural polyphenol, was chosen because it enhances activity of proteins that lead to lipid mobility via lipolysis^19-22 and lipid catabolism via fatty acid oxidation^23-26. For example, resveratrol has shown to increase the expression of adipose triglyceride lipase (ATGL), and carnitine palmitoyltransferase (CPT1), the rate limiting enzymes involved in lipolysis and fatty acid oxidation, respectively. In addition, the literature indicates that these effects are a result of resveratrol’s activation of AMP kinase (AMPK)^21,27. AMPK activation has shown to regulate ATGL expression²⁸ and inhibit acetyl-coA-carboxylase (ACC) leading to an increase in CPT1 activity²⁹.

To study resveratrol scaffold’s effect, we first characterized the host response to resveratrol loaded PLG scaffolds after implant into the epididymal adipose tissue of lean mice. In addition to promoting cellular infiltration and extracellular matrix deposition, we hypothesized that resveratrol scaffolds would lead to a decrease in lipid accumulation, as observed by adipocyte area, compared to PLG implants without a drug payload due to resveratrol’s impact on metabolic programs. We chose to initially focus on the tissue remodeling 28 days after implant, a time point 2 weeks longer than our previously published report and a commonly studied time point to evaluate the host response¹⁶. We then measured resveratrol’s effect on the expression of ATGL and CPT1 as well as AMPK activation. Furthermore, to study the ability of resveratrol loaded scaffolds to prevent onset of high fat diet induced weight gain, mice received either PLG scaffolds without a drug payload or resveratrol loaded PLG scaffolds and were then challenged with a high fat diet containing 60% calories from fat, a mouse model of diet induced obesity³⁰. The epididymal fat pad was chosen as the implant site because this fat pad expands more during the initial stages of high fat diet feeding compared to other fat depots³¹; therefore, decreasing this fat pad’s mass would be a significant achievement during the study time frame. While scaffolds have been used to deliver resveratrol to modulate inflammation for several tissue engineering applications^32-37, we are the first to investigate scaffold-based delivery of resveratrol to modulate metabolic programs in the adipose tissue.

MATERIALS AND METHODS

Scaffold fabrication

Scaffolds were fabricated from PLG microparticles using a gas foaming, particulate leaching technique. Particles containing resveratrol (Sigma) were fabricated as previously described³⁸. Briefly, a 1:3 volume ratio of dichloromethane (Sigma) and ethanol (Sigma) containing 6% w/w PLG (75:25 mol ratio lactide to glycolide 0.76 dL/g, Evonik) and 10 mg/mL resveratrol was added to an aqueous solution of 1% w/v PVA (Sigma) in a 1:7 volume ratio and homogenized. The emulsion was then added to ultrapure water and stirred for 5 hours to remove the dichloromethane and promote solidification of the polymer and encapsulation of resveratrol. Particles were washed with ultrapure water and collected via centrifugation. Particles were lyophilized and stored in aluminum foil in a desiccator to ensure particles were protected from light and moisture. Particles that did not contain resveratrol were fabricated in a similar fashion except the oil phase contained only dichloromethane and 6% w/w PLG.

To fabricate scaffolds, polymer particles were mixed with a sacrificial porogen, NaCl (250-500 μm particles), in a 1:30 mass ratio and pelleted in a die. The resulting tablet was gas foamed using 800 psi CO₂ at room temperature in a custom-made pressure vessel and the salt porogen was removed by washing in ultrapure water. Complete salt removal was verified by scaffold weight and microscopy. Scaffolds containing resveratrol were made with resveratrol loaded PLG particles and scaffolds that did not contain the drug were made with particles containing only PLG. Scaffolds were fabricated so that they weighed 2.08 mg ± 0.06 mg (mean ± SEM). From here on, scaffolds that contain resveratrol are referred to as “RSV scaffolds” and scaffolds containing only the polymer are referred to as “PLG scaffolds.”

Characterizing scaffold structure and pore size

Scaffolds were added to carbon adhesive tape on aluminum stubs and gold sputtered two times for 1 minute using a Denton Desk II vacuum sputter coater. Images were taken on a Tescan Vega3 scanning electron microscopy at 5kV. The face of the scaffolds was imaged to give a top view observation. Scaffolds were also imaged using a Nikon Eclipse Ci microscope at 4x magnification. ImageJ was used to quantify the average pore diameter in both scaffold groups.

Measuring resveratrol loading in scaffolds

Resveratrol loading in scaffolds was measured as previously described³⁸. Briefly, after leaching, scaffolds were weighed and dissolved in a defined volume of DMSO (Sigma). The solution was then analyzed for absorbance at 330 nm and absorbance values were compared to a nine-point standard curve allowing for the interpolation of the unknown resveratrol concentration in the sample. Scaffold loading was calculated as the amount of resveratrol measured in the scaffold (μg) normalized by the mass of the scaffold analyzed (mg).

In vitro release assay

Resveratrol release from scaffolds over time was determined as previously described. Briefly, scaffolds were weighed and incubated in ultrapure water at 37 °C for 0.25, 1, 3, or 7 days. At each time point, scaffolds were removed from the incubator and stored in a desiccator until dry, which was confirmed by scaffold weight. The amount of resveratrol remaining in the scaffold was measured as described in the previous section.

Animal care and scaffold implant

All animal procedures were approved by the University of South Carolina Institutional Animal Care and Use Committee and NIH guidelines for the care and use of laboratory animals were observed. Six-week-old male C57BL/6 mice were purchased from the Jackson Laboratory and, upon arrival, acclimated for 2 weeks prior to the scaffold implant procedure. Prior to implant, scaffolds were washed in 70% ethanol and rinsed in sterile PBS. Mice were anesthetized with isoflurane and their abdomens were shaved and prepped with betadine and ethanol. A lower abdominal midline incision was made, and one scaffold was wrapped into each epididymal fat pad (i.e. each mouse received two PLG or RSV scaffolds, one in each epididymal fat pad). The intraperitoneal cavity was closed with a running stitch and the skin was closed with wound clips.

Animal study design

To investigate the effect of RSV scaffolds on adipocyte size and adipose tissue gene expression, three mice per group (RSV or PLG) received a scaffold into each epididymal fat pad and were then euthanized 28 days later. Three unmanipulated mice served as a naïve control. One fat pad was frozen for western blot, and one fat pad was fixed and embedded in paraffin for histology. In a second study designed to investigate adipose tissue gene expression at an earlier time point, 8 mice per group were implanted with one scaffold per epididymal fat pad. Seven or 14 days later, mice were euthanized and epididymal fat pads were harvested and frozen for western blot. Four mice per group were euthanized at either time point.

To assess the effect of RSV scaffolds on mice fed a high fat diet, five mice per group (RSV or PLG) received a scaffold into each epididymal fat pad. Twenty-eight days after scaffold implant, mice were placed on a 60% high fat diet (Research Diets D12492) and euthanized 28 days after induction of the diet. Fat pads were collected for protein and histological analysis. Mice were allowed access to the diet ad libitum during the entire study except when they were fasted prior to body weight measurements.

Sample sizes were selected using power analyses. Previous studies have demonstrated a 30% decrease in adipocyte size in lean animals with exercise training versus those that were sedentary³⁹ and we reasoned this would be a reasonable benchmark for scaffolds designed to release resveratrol compared to scaffolds without a drug payload. Thus, three mice were randomized to each group (PLG or RSV scaffolds) giving 80% power to detect a 30% decrease in adipocyte size with a significance level of 0.05%. For 7- and 14-day studies, we powered studies to see a 20% difference between the means of the groups (PLG or RSV scaffolds). If there was a difference, 4 mice per group would be required for 80% power and a significance level of 0.05%. For the high fat diet study, we increased our animal group size to 5 mice in case there were issues regarding animal health in combining the high fat diet feeding with scaffold surgery; however, none were noted. Therefore, a total of 35 mice were used in the study: 8 in the 7-day study with 4 mice per group, 8 in the 14-day study with 4 mice per group, 6 in the 28-day study with 3 mice per group, 10 in the high fat diet study with 5 mice per group, and 3 unmanipulated mice that served as a naïve control for adipocyte size quantification. Male mice were used in these studies because female rodents are protected from negative effects associated with high fat diet feeding^40,41.

Fasting weight measurements

On days 14 and 28 after initiation of the high fat diet, mice were fasted for 6 hours beginning at 7AM with access to water. This fasting program is the suggested standard practice for fasting mice prior to measuring metabolic indices⁴². Weight measurements were obtained by placing the mouse into a beaker tared on an Ohaus digital scale.

Histological analysis

After collection, fat pads were washed in sterile PBS, placed in 4% paraformaldehyde for 24 hours, and then paraffin embedded. Serial, 5 μm sections were cut from the paraffin blocks and stained with hematoxylin and eosin (VWR). Sections were imaged using a Nikon Eclipse Ci microscope at 4x or 40x magnifications. ImageJ software was used to quantify adipocyte area from 3 random 40x images per section. Three sections were quantified from 3 mice per group. All adipocytes within the field of view were quantified. The number of giant cells were quantified by counting the number of cells containing greater than or equal to 3 nuclei in at least 3 random 40x images of the implant per section. Three sections were quantified from 3 mice per group. This quantification was conducted using previously published methods^43-45.

Western blot analysis

Collected fat pads were washed in sterile PBS, frozen on dry ice and stored at −80°C. Tissues were homogenized in a defined volume of RIPA buffer (Fisher Scientific) containing Halt Protease Inhibitor Cocktail (Thermo Fisher Scientific), PhosStop (Sigma), and PMSF (Thermo Fisher Scientific) using a benchtop homogenizer. Insoluble material was removed by centrifugation and whole tissue homogenate samples were aliquoted and stored at −80°C until use.

Total protein in the homogenate was quantified using the Pierce BCA protein assay kit (Thermo Fisher Scientific). Twenty or 30 μg protein was separated by 8 or 10% polyacrylamide gel electrophoresis based on target protein molecular weight. Separated proteins were then transferred to 0.2 μm nitrocellulose membranes. The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline containing Tween-20 (TBST) for 1 hour at room temperature. Primary antibody for phosphorylated AMPK-T172 (Cell Signaling Technologies, CST), ATGL (CST), phosphorylated ACC-S79 (CST), and CPT1 (Abcam) was added to each blot at 1:1000 dilution in 5% BSA in TBST and allowed to incubate overnight at 4°C before incubation with a polyclonal secondary antibody conjugated to horseradish peroxidase (Abcam) for 1 hour at room temperature. The blots were then developed with SuperSignal enhanced chemiluminescent substrate solution (Thermo Fisher Scientific). To verify equal loading, blots were incubated with stripping buffer for 25 minutes and washed with TBST. The blot then underwent the same protocol as previously stated for total AMPK, total ACC, or GAPDH (all purchased from CST). Blots probed for AMPK were stripped a second time and underwent the same protocol to probe for GAPDH. ImageLabs software was used to analyze the relative intensity of each protein.

Immunohistochemical staining

Serial tissue sections from each mouse, 5 μm in thickness, were deparaffinized, rehydrated, and processed for antigen retrieval in citrate buffer using a vegetable steamer. Sections were blocked with a TBS solution containing Tween-20, BSA, goat serum, and anti-CD16/32 (Biolegend) and incubated with an anti-CPT1 antibody (Abcam). Sections were treated with hydrogen peroxide prior to incubation with an HRP labelled polymer conjugated to goat anti-rabbit immunoglobulins (EnVision+, Dako). HRP was detected using the Liquid DAB+ Substrate Chromogen system (Dako). Mayer’s hematoxylin (Dako) was used to counterstain. Three sections from each mouse were stained and then imaged using a Nikon Eclipse Ci microscope.  Analysis included 3 mice per group.

Raw 264.7 cell culture and resveratrol treatment

Raw 264.7 macrophages (ATCC) were cultured at 37°C with 5% CO₂ in DMEM media (Fisher Scientific) supplemented with 1 mM sodium pyruvate (Fisher Scientific), 1500 mg/L sodium bicarbonate (VWR), 10% fetal bovine serum (Fisher Scientific), and 1% penicillin/streptomycin (Fisher Scientific) at 57,000 cells/cm² in 6 well plates. Twenty-four hours after seeding, cells were treated with 0, 10 or 25 μM resveratrol dissolved in ethanol. Twenty-four hours later, cells were removed via trypsin 0.25% EDTA (Fisher Scientific) and a cell scraper. Cells were lysed in RIPA buffer supplemented with protease inhibitors via 21G needle and syringe. Protein fraction was collected after centrifugation at 10,000 xg for 10 minutes and stored in −20 °C. Western blot analysis of CPT1 in treated macrophages was conducted as previously stated using 10 μg of total protein content which was determined via BCA assay and β-actin (Sigma) served as the loading control.

Statistical analysis

Statistically significant differences between PLG and RSV implanted groups were evaluated using an unpaired student’s t-test. Statistically significant differences between three groups were determined using a one-way ANOVA with a Tukey’s multiple comparison test. Statistical significance was set to a p value less than 0.05. The number of samples analyzed is detailed in each figure legend. All analyses were completed using GraphPad Prism. In all figures, data indicates mean ± standard error of the mean (SEM).

RESULTS

Resveratrol scaffold characterization

Scaffold structure was observed using scanning electron microscopy and light microscopy and these images, shown in Figure 1 A-D, indicate that both scaffolds contained similar porous structure. Indeed, ImageJ analysis of light microscopy images, as representatively shown in Figure 1 B and D, determined the average pore diameter was 454±31.8 μm and 434±7.45 μm in PLG and RSV scaffolds, respectively, and therefore, were not significantly different (Fig 1E). Resveratrol scaffolds used in this study contained approximately 48±2 μg RSV/mg scaffold; therefore, delivering approximately 100 μg resveratrol total per 5mm x 2mm scaffold (Fig 1F-G). Release kinetics of resveratrol from scaffolds used in this study is depicted in Figure 2. One-week in vitro release characterization indicates a burst release profile where approximately 50% of the resveratrol initially loaded in the scaffold is released in the first 3 days with approximately 30% of the initial resveratrol remaining in the scaffold after 7 days.

Figure 1. Scaffold Characterization.

(A-D) Scanning electron microscopy and bright field microscopy images of PLG (A-B) and RSV (C-D) scaffolds. Scale bar represents 500 μm for B and D. (E) Quantification of pore diameter using ImageJ analysis of images depicted in B and D. Data was collected from three scaffolds per group. Statistics were determined using an unpaired t-test. (F-G) Top (F) and side (G) profile of scaffold implants measuring approximately 5mm in diameter and 2 mm in height.

Figure 2. Resveratrol release profile.

One week in vitro release profile of resveratrol loaded PLG scaffolds. Data are from 3-8 scaffolds per time point. Error bars denote SEM.

Adipose tissue composition after scaffold implant

Tissue sections from epididymal fat 28 days after PLG or RSV scaffold implant revealed extensive cellular infiltration and protein deposition not present in unmanipulated fat (Fig 3A,D, unmanipulated fat not shown). This tissue remodeling was localized around irregular shape voids indicating pieces of scaffold removed during histological processing (SC Fig 3 A,D) that is surrounded by adipocytes (AD Fig 3 A,D). At high magnification, we observe the presence of fibroblasts (oval nuclei, white arrows) and mononuclear immune cells (circular nuclei, cyan arrows), as well as extensive extracellular matrix deposition (Fig 3B,E; specific areas highlighted by black boxes in Fig 3 A,D). Furthermore, the presence of giant cells, a result of macrophage fusion⁴⁶, was evident in both scaffold groups (outlined in green dotted line Fig 3 B,E). Upon quantification, we observed that there were no significant differences in the number of giant cells present in each 40x image (Fig 3G). Furthermore, there were no obvious qualitative differences in composition or extracellular matrix deposition within the scaffold environment as indicated by H&E staining.

Figure 3: Epididymal adipose tissue remodeling after resveratrol scaffold implant.

(A-F) Histological sections of epididymal fat pads 28 days after implant of PLG (A-C) or RSV (D-F) scaffolds at 4x (A,D) or 40x (B,C,E,F) magnification. Scale bar represents 500 μm for A and D and 50 μm for B, C, E, and F. SC indicates the scaffold environment and AD indicates the surrounding adipocytes. Green dotted lines outline giant cells, cyan arrows indicate mononuclear immune cells, and white arrows indicate fibroblasts. (G) Quantification of giant cells per 40x field. Statistics were calculated by an unpaired t-test. (H) Quantification of adipocyte area. Statistics were calculated by a one-way ANOVA with a Tukey’s multiple comparison test; * p<0.05 compared to PLG. Data for both quantifications was collected from at least 3 40x images per section, with 3 sections per mouse and 3 mice per group.

Interestingly, from a metabolic perspective, adipocytes surrounding RSV scaffolds looked to be smaller in area compared to those surrounding PLG scaffolds (Fig 3 C,F; specific areas highlighted by black boxes in Fig 3 A,D). Indeed, adipocytes surrounding RSV scaffolds were approximately 900 μm² compared to those surrounding PLG scaffolds which measured approximately 1500 μm², and adipocytes in epididymal fat collected from unmanipulated animals, termed naïve, measured approximately 1200 μm² (Fig 3H). Using a one-way ANOVA with a Tukey’s multiple comparison test, resveratrol scaffolds significantly reduced adipocyte area compared to PLG scaffolds; however, adipocyte area was not significantly changed between either scaffold group and the naïve control. Taken together, scaffolds induce local tissue remodeling in the adipose tissue, while resveratrol delivery from scaffolds decreases adipocyte area compared to scaffolds with no drug payload.

Expression of key proteins involved in lipid metabolism following scaffold implant

To further investigate a possible mechanism for resveratrol’s significant impact on adipocyte area, proteins involved in lipid regulation were measured via western blot in whole tissue homogenate. Adipose triglyceride lipase (ATGL), carnitine palmitoyltransferase 1 (CPT1), phosphorylated acetyl-co-A-carboxylase (pACC), and phosphorylated AMP kinase (pAMPK) levels were quantified (Fig 4). Interestingly, we did not observe a significant difference in the expression of these proteins between scaffold groups at the 28-day timepoint. In addition, a large amount of variability was observed when pACC and ACC were measured. We then looked at protein expression 14 days after implant to determine if any differences were present at an earlier time point. Interestingly, at 14 days after scaffold implant, phosphorylated AMPK levels were significantly higher in the RSV scaffold group compared to PLG (Fig 5E). We also measured ATGL and CPT1 in whole tissue homogenate via western blot 7 days after implant to investigate a time when the in vitro release profile, shown in Figure 2, indicates that 30% of the initial amount of resveratrol loaded into the scaffold remains. Interestingly, we did not observe any changes in ATGL and CPT1 protein expression between PLG and RSV scaffold groups 7 days after implant (Supplemental Figure 1).

Figure 4: Effects of resveratrol scaffolds on key lipolysis and fatty acid oxidation protein expression 28 days after implant in lean mice.

(A) Representative western blots of ATGL, CPT1, phosphorylated ACC (Ser79), total ACC, phosphorylated AMPK (Thr172), and total AMPK protein expression in adipose homogenates 28 days after implant. (B-E) Quantification of (B) ATGL, (C) CPT1, (D) ACC, and (E) AMPK expression. Data is from 3 mice per group. Statistics were calculated using an unpaired t-test. * indicates p<0.05 compared to PLG.

Figure 5: Effects of resveratrol scaffolds on key lipolysis and fatty acid oxidation protein expression 14 days after implant in lean mice.

(A) Representative western blots of ATGL, CPT1, phosphorylated ACC (Ser79), total ACC, phosphorylated AMPK (Thr172), and total AMPK protein expression in adipose homogenates 14 days after implant. (B-E) Quantification of (B) ATGL, (C) CPT1, (D) ACC, and (E) AMPK expression. Data is from 4 mice per group. Statistics were calculated using an unpaired t-test. * indicates p<0.05 compared to PLG.

Since CPT1 is the rate limiting enzyme involved in fatty acid oxidation, we were interested in how its spatial distribution was altered after RSV scaffold implant; therefore, we stained fat pads collected 28 days after implant with CPT1 using immunohistochemistry (Fig 6). We found that CPT1 was most highly expressed in the new tissue surrounding both scaffolds (SC Fig 6A,B) compared to the adipocytes surrounding the implant (AD Fig 6A,B). Looking more closely at both scaffold environments indicated by black boxes in Figure 6A and B, there is evidence indicating that CPT1 is expressed by the giant cells (outlined in green) located at the interface of the tissue and material as well as non-fused mononuclear immune cells in the new tissue (highlighted by cyan arrows) (Fig 6C,D). This was observed in both PLG and RSV scaffolds.

Figure 6: Spatial analysis of CPT1 in the epididymal fat pad after scaffold implant.

(A-D) CPT1 stained histological sections of epididymal fat pads 28 days after implant of PLG (A,C) or RSV (B,D) scaffolds at 4x (A,B) or 40x (C,D) magnification. Scale bar represents 500 μm for A and B and 50 μm for C and D. Data is representative of 3 sections per mouse with 3 mice per group. SC indicates the scaffold environment and AD indicates the surrounding adipocytes. Green dotted lines outline CPT1 expressing giant cells and cyan arrows indicate non-fused mononuclear immune cells expressing CPT1.

Expression of CPT1 in resveratrol treated RAW 264.7 macrophages

The histology presented above indicated that macrophages in the scaffold environment are CPT1 positive; however, we did not detect an effect of resveratrol on CPT1 expression using westerns, which was surprising. It is possible that carrying out the measurements in whole tissue homogenates kept us from detecting changes that were occurring in the scaffold recruited cells. To determine if it is possible for resveratrol to modulate CPT1 levels in macrophages, we treated RAW 264.7 macrophages with resveratrol and measured CPT1 protein levels via western blot (Fig 7A,B). Twenty-four hours after initial seeding, the macrophages were treated with 10 or 25 μM resveratrol for 24 hours. Western blot indicated that 25 μM resveratrol increased CPT1 expression by 30% relative to vehicle control (Fig 7B). No difference was detected between cells treated with 10 μM and untreated cells indicating the importance of dose.

Figure 7: Effect of resveratrol on CPT1 expression in RAW 264.7 macrophages.

(A) Representative western blot CPT1 protein expression in Raw 264.7 macrophages treated with 0, 10, or 25 μM RSV for 24 hours. (B) Quantification of CPT1 expression. Data is from three 6 well plates with 2 wells per group on each plate. Statistics were calculated using a one-way ANOVA with a Tukey’s multiple comparison test. * indicates p<0.05 compared to untreated (0 μM).

Protective effect of resveratrol scaffolds in mice challenged with a high fat diet

To determine if resveratrol loaded scaffolds could prevent diet induced obesity compared to PLG scaffolds, lean mice were pre-treated with RSV scaffolds for 4 weeks prior to being placed on a high fat diet for an additional 4 weeks, as illustrated in Fig 8A. Body weight measurements indicate that 14 days after switching to the high fat diet, mice that received RSV scaffolds gained significantly less weight than mice that received PLG scaffolds (Fig 8B). This effect remained after 28 days on the high fat diet (Fig 8B). Additionally, at the end of the study, epididymal fat pads collected from mice that received RSV scaffolds weighed significantly less than those from the PLG scaffold group (Fig 8C).

Figure 8. Impact of scaffolds on weight gain and epididymal fat pad mass in mice challenged with a high fat diet.

(A) Timeline of experiment. (B) Percent increase in fasting body weight relative to the day 0 when the high fat diet began. (C) Epididymal fat pad weight 28 days after HFD challenge. n=5 mice per group. Statistics were conducted using an unpaired t-test. * indicates p<0.05 compared to PLG. HFD = high fat diet.

Histological sections of the epididymal fat pad collected at the end of this study were analyzed similarly to those in the previously discussed in the lean animal study. Adipocytes in the RSV scaffold group appeared smaller than those from the PLG scaffold group (Fig 9A) and, indeed, image analysis supported this observation. In fact, adipocyte area in the RSV scaffold group was 30% smaller compared to the PLG group (Fig 9B). Similar to results observed in the lean animal study, this observation was not accompanied by whole tissue modifications in ATGL, CPT1, pACC, or pAMPK in the RSV scaffold group (Fig 10).

Figure 9: Effects of resveratrol scaffolds on epididymal fat adipocyte size after HFD challenge.

(A-B) Histological sections of epididymal fat pads that received (A) PLG or (B) RSV scaffolds 28 days after HFD challenge taken at 40x magnification. Scale bar represents 50 μm. (C) Quantification of adipocyte area. Data collected from 3 images per section, with 3 sections per mouse and 3 mice per group. Statistics were calculated using an unpaired t-test. * indicates p<0.05 compared to PLG.

Figure 10: Effects of resveratrol scaffolds on key lipolysis and fatty acid oxidation protein expression after HFD challenge.

(A) Representative western blots of ATGL, CPT1, phosphorylated ACC (Ser79), total ACC, phosphorylated AMPK (Thr172), and total AMPK protein expression in adipose homogenates after HFD challenge. (B-E) Quantification of (B) ATGL, (C) CPT1, (D) ACC, and (E) AMPK expression. Data is from 5 mice per group. Statistics were calculated using an unpaired t-test.

DISCUSSION

Chronic overnutrition leads to lipid accumulation that has detrimental consequences such as adipose tissue inflammation and dysregulated insulin signaling resulting in the progression of metabolic diseases such as obesity and type 2 diabetes⁵. Therefore, we are pursuing an approach to combat metabolic disease by enhancing lipid catabolism in the adipose tissue⁴⁷. Extensive research shows that resveratrol has the ability to increase lipid catabolism; however, the molecule’s poor bioavailability proved to be a great obstacle during clinical trials^48,49. As a strategy to overcome this shortcoming, we incorporated resveratrol into a scaffold for direct delivery to the adipose tissue and investigated its ability to modulate fat pad size. Mice pre-treated with resveratrol scaffolds and then fed a high fat diet gained significantly less total body weight and epididymal adipose tissue mass compared to mice that received scaffolds containing only polymer. The effect on total body weight of this localized therapy might not be surprising because the epididymal fat pad is large, expands quickly, and contributes significantly to total body weight gain in this model³¹. Importantly, this scaffold-based strategy required a single administration compared to previous studies indicating that oral resveratrol delivery requires daily dosing^19,25,50. Furthermore, it is possible that this scaffold-based strategy could be easily combined with lifestyle interventions, such as calorie restriction and physical activity, which could further enhance its potential as a promising therapy for metabolic diseases.

In both lean and high fat diet fed mice, resveratrol scaffolds reduced adipocyte area compared to PLG scaffolds indicating a reduction in lipid content since adipocyte area and lipid content are positively correlated⁵¹. We propose that resveratrol scaffolds promote the utilization of lipids within the implant site and that this effect is likely due to the combination of the host response to biomaterial implant and the release of resveratrol from the scaffold. The host response exhibited here, characterized by local cellular infiltration and protein deposition, is similar to wound healing which depends on cellular activity that is largely fueled by fatty acid metabolism^52-54. Furthermore, oral delivery of resveratrol to animals as well as treatment of cultured cells impacts the activity of several proteins involved in lipid metabolism, as described in several review articles^26,55,56. For example, treatment of cultured 3T3-L1, porcine, and human adipocytes with resveratrol led to an increase in lipolysis as a result of elevated ATGL levels^19-22,57. In addition, resveratrol increased CPT1 mRNA levels in adipose tissue from pigs and rats fed a resveratrol supplemented diet and cultured adipocytes treated with resveratrol^23-26,58. Contrary to our initial hypothesis, scaffold-based resveratrol delivery did not significantly increase expression of either ATGL or CPT1 when measured in whole tissue homogenate collected 14 or 28 days after implant. We must consider that measuring these proteins in whole tissue may have decreased our sensitivity to see significant changes. For example, CPT1 is found on the mitochondrial membrane; therefore, it may be difficult to measure with high sensitivity in whole tissue homogenate. It is also possible we did not see differences in these proteins between scaffold groups at day 28 because the resveratrol payload was exhausted at that time point. However, we did see increased levels of activated AMPK at day 14 in the tissue. Literature indicates that resveratrol activates AMPK and resveratrol’s effect on ATGL and CPT1 are consequences of AMPK activation^21,27. Thus, it is possible that the decreased adipocyte size at day 28 was due to increase in ATGL and CPT1 enzyme activity at earlier time points.

Remarkably, resveratrol scaffolds protected mice from high fat died induced weight gain and epididymal fat accumulation 8 weeks after initial implant, while the in vitro release profile indicates that approximately 70% of the initial resveratrol is released from the scaffold in the first week. We speculate that as pieces of the scaffold are encapsulated by the immune system, resveratrol release is slowed, extending its resident time in the tissue. These potential areas of high resveratrol concentration would more readily impact cells that are in direct contact. Therefore, we believe that macrophages are a particularly interesting population for further study as they are recruited to the implant site³⁸ and fuse together to form multinucleated giant cells that are in direct contact with the scaffold⁴⁶. Resveratrol has extensively been shown to polarize macrophages to an anti-inflammatory phenotype^32,59-61. Interestingly, anti-inflammatory macrophages utilize fatty acid oxidation as their key energy source compared to pro-inflammatory macrophages that utilize glucose metabolism⁶². CPT1 is the rate limiting enzyme that controls fatty acid oxidation and is elevated in anti-inflammatory macrophages^63,64. Activation of peroxisome proliferator-activated receptor-α (PPARα) inhibits NF-κB signaling and also increases the expression of CPT1 in macrophages^65,66. We show that CPT1 is more highly expressed in the newly formed tissue surrounding the scaffold compared to surrounding adipocytes, and the giant cells were CPT1 positive. This data is further supported by work that demonstrated that CPT1 is more highly expressed in adipose tissue macrophages compared to adipocytes⁶⁷. However, spatial information of CPT1 expression was similar in both resveratrol and PLG scaffolds; therefore, to further study resveratrol’s potential effect on macrophages, we treated RAW 264.7 macrophages with resveratrol. We found that resveratrol treatment increased CPT1 expression. It has been shown that resveratrol decreases lipid levels in RAW 264.7 macrophages⁶⁸ and our work suggests that resveratrol decreases lipid levels in RAW 264.7 macrophages through increasing CPT1, possibly through PPARα activation. Furthermore, anti-inflammatory macrophages have shown to play a part in regulating adipocyte lipid metabolism. For example, co-culture of 3T3-L1 adipocytes and anti-inflammatory macrophages led to enhanced lipolysis in adipocytes⁶⁹. Therefore, we hypothesize that the mechanisms by which resveratrol scaffolds decrease fat pad size relative to blank scaffolds is that resveratrol polarizes macrophages recruited to the scaffold, which can fuse into giant cells, to an anti-inflammatory phenotype, leading to enhanced fatty acid utilization. This then enhances lipolysis in surrounding adipocytes resulting in the release of free fatty acids, reflected in decreased adipocyte size, which anti-inflammatory macrophages then use as their preferred energy source. Future aims will be focused on isolating adipocytes, macrophages, and other cells from the implant site to dissect cell-specific modifications that arise after scaffold-based resveratrol delivery that results in reduced lipid accumulation.

While this work demonstrates that resveratrol releasing scaffolds decrease fat pad size in mice fed a high fat diet compared to scaffolds with no drug, this study does have limitations. First, we focused on demonstrating a protective effect, which lays a promising foundation for the technology, but is not translatable to humans as this strategy will likely be used as an obesity intervention, not a prophylactic. Studies investigating resveratrol scaffold’s ability to treat high fat diet induced obese mice are necessary and is a focus of future studies. Secondly, we limited the scope of the study to RSV scaffold’s impact on the epididymal fat pad. Due to differences in composition and metabolic activity in visceral and subcutaneous fat depots, it remains unclear if this scaffold-based strategy would achieve similar results in other fat depots. Furthermore, there is a question of if this strategy would need to target multiple fat depots simultaneously to observe a significant therapeutic effect in humans. In addition, this study does not address if the use of scaffolds to enhance lipid metabolism would work in other metabolically relevant tissues, such as skeletal muscle; however, this is the focus of future work. Finally, we must consider that the extensive remodeling that occurred as a result of the implant may have negative effects on adipose tissue. Future work is focused on developing injectable technologies that would not induce extensive remodeling but would rather individually target immune cells to modulate gene programs involved in lipid catabolism.

CONCLUSION

Implant of resveratrol releasing scaffolds into epididymal fat of lean mice decreased adipocyte size, a surrogate for lipid content, compared to PLG scaffolds without a drug payload. Interestingly, the expression of CPT1 and ATGL, key proteins involved in fatty acid catabolism, whose elevated expression is associated with decreased adipocyte size, were unchanged in the epididymal fat at 7, 14 or 28 days after scaffold implant. However, AMPK activation was increased at 14 days, which indicates that ATGL and CPT1 activity levels could have been elevated, providing a plausible explanation for the decreased adipocyte size. Histology indicated that scaffold-recruited immune cells expressed CPT1 at higher levels than other cells in the fat pad and resveratrol increased CPT1 in cultured macrophages by 30%. Thus, it is possible that our study of whole tissue homogenates limited our ability to detect increases in CPT1 because it was occurring in a small population of cells. Taken together, we hypothesize that increases in CPT1 levels and/or activity in macrophages within the scaffold play a role in resveratrol’s effect on adipocyte size. We also demonstrate that mice implanted with resveratrol scaffolds are more resistant to fat gain during 4 weeks of high fat diet compared to mice that received scaffolds with no drug. Importantly, this effect was realized with a single administration of scaffolds, which is considerably less than the daily dosing regimen usually required for resveratrol’s weight loss effects on animals fed a high fat diet. We conclude that direct delivery of resveratrol to the fat tissue overcomes issues with bioavailability and that tissue engineering scaffolds can be used to achieve this goal. This work motivates further study into the use of biomaterials for localized delivery of metabolism modulating agents to key metabolic cells in adipose tissue as a treatment for metabolic disease secondary to over-expanded adipose tissue.