A Straightforward Approach to Engineer Vascularized Adipose Tissue Using Microvascular Fragments

Abstract

There is a need to overcome the donor-site morbidity and loss of volume over time that accompanies the current clinical approaches to treat soft tissue defects caused by disease and trauma. The development of bioactive constructs that can regenerate adipose tissue have made great progress toward addressing the limitations of current therapies, but their lack of vascularization and ability to meet the significant dimension requirements of tissue defects limit their clinical translatability. Microvascular fragments (MVFs) can form extensive vascular networks and contain resident cells that have the ability to differentiate into adipocytes. Therefore, the objective of this study was to determine if vascularized adipose tissue could be engineered using a fibrin-based hydrogel containing MVFs as the sole source of microvessels and adipocyte-forming cells. The potential for MVFs from different fat depots (epididymal, inguinal, and subcutaneous) to form microvascular networks and generate adipocytes when exposed to growth media (GM), adipogenic differentiation media (ADM), or when treated with GM before adipogenic induction (i.e., they were allowed to presprout before adipogenic induction) was evaluated. MVFs treated with adipogenic induction media, both with and without presprouting, contained lipid droplets, had an increase in expression levels of genes associated with adipogenesis (adiponectin and fatty acid synthase [FAS]), and had an increased rate of lipolysis. MVFs allowed to presprout before ADM treatment maintained their ability to form vascular networks while maintaining an elevated lipid content, adipogenic gene expression, and lipolysis rate. Collectively, these results support the contention that MVFs can serve as the sole source of biologic material for creating a vascularized adipose tissue scaffold.

Impact statement

Microvascular fragments have both the ability to form extensive vascular networks and function as a source of adipocytes. These phenomena were exploited as vascularized adipose tissue was generated by first allowing for a period of angiogenesis before the adipogenic induction. This strategy has the ability to provide a means of both improving soft tissue reconstruction while also serving as a model to better understand adipose tissue expansion.

Introduction

The ability to effectively reestablish adipose tissue and protect underlying structures in composite tissues would provide a long-term benefit to patients that is not fully met by the current clinical approaches to treat tissue defects resulting from tumor resection and the treatment of congenital defects. For example, the use of autologous fat grafting is currently hindered by limited persistence and the potential for negative side effects that are attributed, at least in part, to a high rate of resorption, graft survival inconsistency, and insufficient perfusion.¹ Given the dependence of adipose tissue on a functional microvasculature in healthy tissue, it is not surprising that an insufficient vascular supply has been suggested to be responsible, at least in part, for the insufficiency of autologous fat grafting when fat grafting is used therapeutically.^1–3

The field of adipose tissue engineering is directed toward developing soft tissue replacements that overcome some of the negative outcomes associated with the current clinical approaches, to include donor-site morbidity and lack of graft persistence. However, traditional adipose tissue engineering approaches may benefit from a vascular component to support the replacement of substantial volumes of tissue, while also recapitulating the normal physiology of adipose tissue.

One way that vascularized adipose tissue engineering confronts the need for implant perfusion is through the prevascularization of biomaterials, typically achieved with stem cells, endothelial cells, or some combination thereof.^4–7 This approach is a logical progression toward improving the perfusion of tissue-engineered adipose tissue, but it still requires substantial optimization to effectively recapitulate the complexity of microvessels and address the intricate physiology of microvascular perfusion in vivo. Furthermore, the ability to concurrently develop and maintain adipocyte integrity under the constraints needed for prevascularization is not a trivial hurdle to overcome without considerable in vitro manipulation.

Microvessels derived from adipose tissue, collectively referred to as microvascular fragments (MVFs), are an alternative strategy to the common prevascularization approaches employed in tissue engineering.^8,9 Because they are isolated as intact microvessels, they require minimal manipulation to develop advanced networks with the ability to inosculate with host tissues and improve tissue perfusion.^10–14 In addition to their potential for improving tissue/implant perfusion, the observation that a variety of stem cells with multidifferentiation potential reside within microvessels, and that cells derived from MVFs have adipogenic potential, support the contention that MVFs may have a distinctive role in adipose tissue engineering.^13,15–17

Based on the physical relationship between adipocytes and microvessels and the strong correlation between adipogenesis and angiogenesis during adipose tissue expansion, it is logical to speculate that MVFs may be used as a means to exploit this natural phenomenon while addressing the clinical need for vascularized adipose tissue.^18–20 The objective of this study was to determine if MVFs could be used as a single source of cells to create a microvascular network containing adipocytes.

To gain insight into the importance of adipose tissue source in the creation of a vascularized adipose tissue, three different sources of rodent adipose tissue, namely, epididymal (EPI), inguinal (ING), and subcutaneous (SUBQ) depots, corresponding to human visceral, lower body SUBQ, and upper-body SUBQ, respectively, were used as the source of MVFs.²¹ The development of vascularized networks with functional adipocytes using MVFs provides a platform for the further development of a simple procedure to address the need for vascularized adipose tissue for treating soft tissue defects, while also presenting an opportunity to gain a deeper understanding of the mechanisms that govern adipocyte physiology in both developing and diseased tissue.^3,19

Materials and Methods

This study has been conducted in compliance with the Animal Welfare Act and the Implementing Animal Welfare Regulations in accordance with the principles of the Guide for the Care and Use of Laboratory Animals, and was approved by the Institutional Animal Care and Use Committee at the University of Texas at San Antonio.

Microvascular fragments isolation and hydrogel formation

Microvascular fragments (MVFs) were isolated from the EPI, ING, and SUBQ fat depots similar to that previously described (Fig. 1A).¹⁰ Briefly, adipose tissue from the EPI, ING, or SUBQ fat from 8- to 12-month-old male Lewis rats (Envigo, Huntingdon, United Kingdom) were subjected to an 8- to 15-min collagenase type I digestion (Worthington Biochemical Corporation, Lakewood, NJ) at 37°C with agitation. The digested material was subjected to centrifugation (400 g × 4 min), which resulted in a floating layer of adipocytes, and a pellet containing a heterogeneous mixture of cells and MVFs. The pellet was resuspended in phosphate-buffered saline containing 0.1% bovine serum albumin (Sigma-Aldrich, St. Louis, MO) and filtered through 500 and 37 μm filters (Carolina Biological Supply, Burlington, NC) to remove large debris and minimize cell contamination, respectively. MVFs were then counted, centrifuged, and resuspended in fibrinogen (Sigma-Aldrich) in Dulbecco's modified Eagle's medium (DMEM; 20 mg/mL) at a concentration of 20,000 MVFs/mL. Fibrin hydrogels (5.7 mg/mL) were formed by combining MVFs containing fibrinogen and 10 U/mL thrombin (Millipore Sigma, St. Louis, MO) in 96-well culture plates. Hydrogels for polymerase chain reaction (PCR) were 100 μL in volume, whereas gels for all other analyses were 50 μL. Isolated MVFs for use in these experiments were from four different rats, with pooling of two rats occurring for days 14 and 21 experiments. For all conditions, experiments repeated at least twice yielded similar results.

FIG. 1.

Experimental groups and design. (A) Depots of rodent adipose tissue, as the source of MVFs evaluated throughout the study. Specifically, EPI tissue (visceral adipose tissue depot), ING, and SUBQ (SUBQ adipose tissue depots) were used. (B) MVFs from the different depots were analyzed for angiogenesis in GM or ADM conditions after 7, 14, and 21 days in culture. To investigate the angiogenic and adipogenic potential of MVFs after a period of angiogenesis (presprouting), MVFs were grown in GM for 7 days followed by 14 days of ADM (with an analysis on day 21 of culture). ADM, adipogenic differentiation media; EPI, epididymal; GM, growth media; ING, inguinal; MVFs, microvascular fragments; SUBQ, subcutaneous. Color images are available online.

Experimental design and culture conditions for evaluating angiogenic and adipogenic potential of MVFs

Fibrin hydrogels containing MVFs from different fat depots (Fig. 1A) were grown for a period of 7, 14, or 21 days in either microvascular growth media (GM), adipogenic differentiation media (ADM), or GM+ADM, according to the treatment scheme as described in Figure 1B. Angiogenesis was stimulated by culturing MVFs in GM (DMEM containing 20% fetal bovine serum, 1% Pen-Strep, and 0.2% MycoZap) for 7, 14, or 21 days. To stimulate adipogenesis, MVFs were grown in adipogenic differentiation media that first consisted of a 4-day treatment with induction media (DMEM/F12 containing 20% fetal bovine serum, 1% Pen-Strep, 0.2% MycoZap, 10 μg/mL insulin, 10 μM forskolin, and 1 μM dexamethasone) followed by maintenance media (DMEM/F12, 20% fetal bovine serum, 1% Pen-Strep, 0.2% MycoZap, and 5 μg/mL insulin) for 10 days. To investigate the angiogenic and adipogenic potential of MVFs after a period of angiogenesis (presprouting), MVFs were grown in GM for 7 days followed by 14 days of ADM. For all treatments the appropriate medium (100 μL in a 96-well plate) was replaced every other day throughout the study while cultures were maintained in a humidified incubator at 37°C and 5% CO₂. All media for hydrogels was supplemented with 1 mg/mL aminocaproic acid to inhibit fibrinolysis.

Gene expression analysis

RNA was isolated from hydrogels (100 μL) containing MVFs (n = 4 individual hydrogels/group) and purified using the Qiagen RNeasy Mini Kit (Valencia, CA) according to the manufacturer's guidelines. Messenger RNA (mRNA) concentrations were measured using a Take3 Micro-Volume Plate (BioTek, Winooski, VT), then normalized to 150 ng of mRNA for its conversion to complementary DNA (cDNA). Isolated RNA was converted to cDNA using the iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA). Real-time quantitative polymerase chain reaction (qPCR) was performed using a CFX96 Touch Real-Time PCR Detection System (Bio-Rad). All primers used to carry out the qPCR analysis were predesigned primers (Sigma-Aldrich) (Table 1). Ten microliters of iTaq Universal SYBR Green Supermix (Bio-Rad) was used for each reaction. Fold expression levels were calculated using the 2^−ΔΔCt method, where the GM gels at day 1 were designated as the calibrator group and GAPDH expression was used as the endogenous control.²²

Table 1.

Polymerase Chain Reaction Primers

Target gene	Sequence (5′-3′)
rGAPDH	F—AGCCCAGAACACCATTCCTAC
	R—ATGCCTGCTTCACCACATTC
rFLK1	F—ACTCACAGTTCCCAGAGTGGTT
	R—GAATGGTGACCTGTGATCTTGA
rANGPT-1	F—TATTTTGTGATTCTGGTGATT
	R—TCGCTTTATTTTTGTAATG
rVEGF	F—CAACTTCTGGGCTCTTCTCT
	R—CTCACCCGTCCATGAGC
rAdiponectin	F—GTTGGATGGCAGGCATCC
	R—GCTCTCCTTTCCTGCCAG
rFAS	F—GCCCTGCTACCACTGAAGAG
	R—GTTGTAATCGGCACCCAAGT

ANGPT-1, angiopoietin 1; F, forward primer; FAS, fatty acid synthase; r, rat; R, reverse primer; VEGF, vascular endothelial growth factor.

Histological analyses

Hydrogels were fixed in 4% formaldehyde for 2 h at room temperature, permeabilized using 0.5% Triton-X for 20 min, blocked using 10% goat serum for 2 h, then stained using Rhodamine-labeled Griffonia (Bandeiraea) Simplicifolia Lectin I (GS-1; 1:100; Vector Laboratories, Burlingame, CA), boron-dipyrromethene (BODIPY;D3922, 1:100; Thermo Fisher), and F-Actin (R37112; Thermo Fisher). The distribution of both vessels and lipid droplets of entire wells (n = 6/group) were determined using a Leica TCS SP8 Confocal Microscope (Buffalo Grove, IL) using a rendering of 100 μm thickness/10 μm sections of the entire well coverage area. Quantification was performed using the Leica three-dimensional (3D) analysis toolkit using Otsu thresholding.

Lipolysis assay

A lipolysis assay measuring glycerol release was completed using the Lipolysis Colorimetric Assay Kit according to the manufacturer's recommendations (BioVision, Milpitas, CA) with a few modifications. Briefly, at the end of the differentiation protocol, gels (n = 12/group) were washed two times with provided Lipolysis Wash Buffer, which was then replaced with the Lipolysis Assay Buffer. Ten micromolars of Isoproterenol (final concentration 100 nM) was added to half the wells to stimulate lipolysis for 3 h. Following stimulation, 50 μL of media was collected into a 96-well plate and 50 μL reaction mix, as provided by the manufacturer, was added and incubated at room temperature for 1 h, after which absorbance was read at optical density 570 nm, with the amount of glycerol released calculated using a standard curve.

Statistical analysis

GraphPad Prism Software 6 (GraphPad Software, Inc., La Jolla, CA) was used to run two-way analysis of variance tests with Holm/Sidak's multiple comparison analyses to determine differences between groups. Statistical significance was determined when p < 0.05. All results are presented as mean ± standard error of the mean.

Results

Qualitative evaluation of angiogenesis and adipogenesis

At the completion of each time point, the fibrin hydrogels were labeled with GS Lectin I to visualize vessel formation and BODIPY to examine the presence of lipid droplets. MVFs grown only in GM conditions for 7, 14, or 21 days, did not have any visible lipid droplets within the fibrin gels (Fig. 2). MVFs treated with ADM had numerous lipid deposits within the hydrogel. When MVFs were cultured with ADM for 14 days, the amount of vessel formation was minimal and lipid deposits were numerous and scattered with minimal clustering. MVFs exposed to GM+ADM exhibited less vessel formation as compared with the day 21 GM control; however, branched hierarchical vascular structures, distinctive of vessel formation were formed alongside lipid deposits throughout the scaffold. Notably, lipid formation in the GM+ADM groups followed a more clustered arrangement with a high degree of association between droplets and vessels.

FIG. 2.

Qualitative histological analysis of vessel and lipid formation. Confocal images of MVFs grown in fibrin hydrogels (plugs) and stained with GS Lectin I (red) to visualize vessel formation and BODIPY (green) to identify the presence of lipid droplets. Groups depicted were exposed to a range of media conditions (GM, ADM, or GM+ADM) for a duration of 7, 14, or 21 total days. Additionally, MVFs were sourced from EPI, ING, or SUBQ fat. The images on the leftmost side (scale bars = 1 mm) display an overview of the distribution of the MVFs in the hydrogel, and the corresponding insets on the right were included to visualize more clearly the presence of adipocytes and their interaction with the vessels (scale bars = 200 μm). BODIPY, boron-dipyrromethene; GS, Griffonia (Bandeiraea) simplicifolia. Color images are available online.

When looking at vessel growth and lipid formation across fat types, similar observations were seen among the groups. MVFs isolated from ING depots and cultured with GM only for 7 days appeared to have a higher degree of vessel formation than MVFs isolated from other sources, suggesting a faster sprouting rate may be occurring in this group initially. Also worth noting, lipid formation appeared to be the highest for both the ADM and GM+ADM groups in the MVFs derived from EPI fat.

Quantitative evaluation of angiogenesis and adipogenesis

Quantification of vessel and lipid formation was performed on 3D depictions of entire wells containing treated MVFs (Fig. 2). The percentage well coverage of lipids (BODIPY staining) was used to indirectly quantify the extent of adipogenesis occurring in the different groups (Fig. 3A). Culturing MVFs with only GM had little effect on lipid formation, an observation that was consistent among fat sources tested, with the exception of MVFs derived from the visceral fat depot (EPI) after 21 days of treatment (Fig. 3A). MVFs isolated from EPI had a significant increase in lipid quantity when comparing GM+ADM measured on day 21 as compared with ADM treatment measured on day 14 (Fig. 3A). On the contrary, MVFs isolated from the SUBQ fat depots, ING and SUBQ, had a significant decrease or no significant change in lipid formation, respectively, when comparing GM+ADM measured on day 21 and ADM treatment measured on day 14 (Fig. 3A). Notably, lipid formation was significantly higher in all groups treated with ADM, relative to GM-only treated groups within fat sources (Fig. 3A). Among fat sources, MVFs isolated from EPI and treated for 14 days with ADM had the highest quantity of lipid as compared with ING and SUBQ (Fig. 3A). Similarly, MVFs isolated from EPI and treated with GM+ADM for 21 days had the highest level of lipid quantity, followed by SUBQ and ING, respectively (Fig. 3A).

FIG. 3.

Quantitative histological analysis of vessel and lipid formation. Effect of media condition (GM, ADM, or GM+ADM) and duration of media exposure (7, 14, or 21 total days) on (A) lipid accumulation and (B) vessel formation as determined with BODIPY accumulation or GS Lectin I (Lectin), respectively. Quantification performed as a measurement of percent well coverage within wells when EPI, ING, or SUBQ fat was used as the source of MVFs. Results are reported as mean ± standard error (n = 6). Different letters indicate statistically significant differences between treatments (p < 0.05) within a fat source. Statistically significant differences between treatments (p < 0.05) among fat sources indicated by: *different from the same treatment with EPI as the source of MVFs; ^#different from the same treatment with ING fat as the source of MVFs; ^&different from the same treatment with SUBQ as the source of MVFs.

To quantify the extent of angiogenesis and measure the vessel density among the different treatment groups, the percentage of the well covered Lectin was quantified. As shown in Figure 3B, among the GM-only treated groups, MVFs isolated from EPI had an increase in vessel density between days 14 and 21. MVFs isolated from ING had a high vessel density as indicated by lectin coverage at day 7, and no difference in vessel density between the three time points studied (Fig. 3B). MVFs isolated from SUBQ significantly increased vessel density between days 7 and 21 (Fig. 3B). Across fat types, ING showed significantly higher vessel density relative to EPI and SUBQ after 7 days of treatment with GM (Fig. 3B). After 14 days of treatment with GM, MVFs isolated from ING had a significantly higher vessel density relative to MVFs isolated from EPI, while after 21 days of treatment MVFs isolated from different fat sources were not different with regard to their vessel coverage (Fig. 3B). Within all fat types, when comparing the GM-only treated groups to their ADM-treated counterparts of that same day (i.e., day 14—GM vs. day 14—ADM and day 21—GM vs. day 21—GM+ADM) there was a significant decrease in vessel density (Fig. 3B). However, between the ADM-treated groups, within each fat type across all fat groups there was a statistically significant increase of vessel density between the day 14—ADM and day 21—GM+ADM groups (Fig. 3B). Stated another way, when considering all fat types, presprouting the MVFs in GM for 7 days (GM+ADM) increased vessel density in ADM. There were no significant differences in vessel density among MVFs isolated from the different fat sources and treated with ADM (Fig. 3B).

Lipolysis

Based on the high level of lipid quantity observed in MVFs treated with ADM for 14 days, or MVFs treated with GM for 7 days followed by ADM for 14 days relative to their GM-only counterparts (Figs. 2 and 3A), a lipolysis assay was performed to evaluate the functionality of the adipocytes under these conditions. MVFs derived from all fat sources and treated with ADM for 14 days had an increased rate of both basal and isoproterenol-stimulated release in glycerol as compared with their GM-only-treated counterparts (Fig. 4A). MVFs isolated from the EPI fat depot had the greatest amount of lipolysis when treated with isoproterenol as compared with MVFs isolated from the other sources (Fig. 4A). MVFs isolated from the different depots and treated with GM for 7 days followed by AM for 14 days (GM+ADM) followed a similar pattern as MVFs treated with ADM only for 14 days with regard to their rate of lipolysis as compared with treatment with only GM (Fig. 4A, B). Similarly, isoproterenol stimulated the release of glycerol in MVFs isolated from EPI and SUBQ, but not ING. There were no differences in the isoproterenol-stimulated release of glycerol among MVFs isolated from the different fat depots (Fig. 4B).

FIG. 4.

MVF lipolysis. The ability of MVFs to generate adipocytes was evaluated with a lipolysis assay that measured glycerol released. MVFs from EPI, ING, or SUBQ fat were (A) exposed directly to ADM for a total of 14 days or (B) were allowed to presprout forming vascular networks in GM for 7 days, after which medium was changed to ADM for an additional 14 days (GM+ADM). At the end of 14 and 21 days, respectively, lipolysis was performed ± isoproterenol, to stimulate lipolysis. Control groups, without any exposure to ADM, (GM) were included for both the 14 and 21-day time points. Results are reported as mean ± standard error (n = 4). Different letters indicate statistically significant differences between treatments (p < 0.05) within a fat source. Statistically significant differences between treatments (p < 0.05) among fat sources indicated by: *different from the same treatment with EPI as the source of MVFs; ^#different from the same treatment with ING fat as the source of MVFs; ^&different from the same treatment with SUBQ as the source of MVFs.

Expression of genes associated with angiogenesis

For day 7—GM, day 21—GM, and day 21—GM+ADM, the expression of genes associated with angiogenesis, including Flk-1, ANGPT-1, and vascular endothelial growth factor (VEGF), were evaluated. Flk-1 (Fig. 5A), an early angiogenic marker, was only increased between days 7 and 21 of GM treatment in MVFs derived from ING (Fig. 5A). There was also a significant decrease in Flk-1 expression between MVFs derived from ING and treated with GM for 21 days as compared with MVFs treated with GM for 7 days followed by ADM for 14 days (GM+ADM) (Fig. 5A). MVFs isolated from EPI and SUBQ showed an increase in expression in the late angiogenic marker ANGPT-1 between the day 7—GM and day 21—GM groups, with no significant difference between day 21—GM and day 21—GM+ADM, and day 7—GM and day 21—GM+ADM (Fig. 5B). There were no differences among MVFs isolated from the different fat depots after 7 days of treatment with GM or MVFs treated with GM for 7 days followed by ADM for 14 days (GM+ADM) (Fig. 5B). Lastly, EPI had no significant change in VEGF expression while ING and SUBQ showed a significant increase in VEGF expression between day 7—GM and day 21—GM, and a decreased expression in VEGF when comparing day 21—GM and day 21—GM+ADM (Fig. 5C). Among MVFs derived from the different fat sources, there were no differences in VEGF expression for the day 7—GM groups, whereas both ING and SUBQ had a significant increase when compared with EPI (Fig. 5C). Additionally, ING was also significantly higher than SUBQ for day 21—GM groups (Fig. 5C). Moreover, ING had a significant increase compared with EPI for the day 21—GM+ADM groups (Fig. 5C).

FIG. 5.

Expression of genes associated with angiogenesis. qPCR was performed on MVFs isolated from EPI, ING, or SUBQ fat and exposed to either GM, or GM followed by ADM (GM+ADM) with a varied duration of media exposure (7 or 21 total days). Specifically, (A) Flk-1, a tyrosine kinase receptor and early angiogenic marker, (B) angiopoietin 1 (ANGPT-1), a later maker for vascular development and angiogenesis, and (C) VEGF, a potent angiogenic factor were all analyzed. Results are normalized based on the fold expression of an untreated control group of MVFs at day 0. Results are reported as mean ± standard error (n = 4). Different letters indicate statistically significant differences between treatments (p < 0.05) within a fat source. Statistically significant differences between treatments (p < 0.05) among fat sources indicated by: *different from the same treatment with EPI as the source of MVFs; ^#different from the same treatment with ING fat as the source of MVFs; ^&different from the same treatment with SUBQ as the source of MVFs. qPCR, quantitative polymerase chain reaction; VEGF, vascular endothelial growth factor.

Expression of genes associated with adipogenesis

For day 14—GM, day 14—ADM, day 21—GM, and day 21—GM+ADM, the expression of genes associated with adipogenesis, including Adiponectin, a protein hormone produced by adipose tissue, and fatty acid synthase (FAS), a multienzyme protein that catalyzes fatty acid synthesis in adipose tissue, were evaluated. The only significant differences in adiponectin expression was between MVFs isolated from ING and SUBQ and treated with GM for 7 days followed by ADM for 14 days (GM+ADM), with their expression being significantly higher than MVFs isolated from EPI (Fig. 6A). Among MVFs isolated from the different fat sources, MVFs isolated from ING and treated with GM for 7 days followed by ADM for 14 days (GM+ADM) had greater adiponectin than EPI and SUBQ (Fig. 6A). FAS expression demonstrated a pattern similar to that of adiponectin with ADM and GM+ADM having a higher gene expression relative to the GM-only treatment (Fig. 6B). Among fat types, the only difference was a higher level of FAS expression in MVFs derived from SUBQ relative to MVFs from ING treated with GM for 7 days followed by ADM for 14 days (GM+ADM) (Fig. 6B). SUBQ showed an increase in FAS expression relative to ING for day 21—GM+ADM.

FIG. 6.

Expression of genes associated with adipogenesis. qPCR was performed on MVFs isolated from EPI, ING, or SUBQ fat and exposed to GM, ADM, or GM+ADM with a varied duration of media exposure (14 or 21 total days). Specifically, (A) Adiponectin, a protein hormone produced in adipose tissue and (B) FAS, multienzyme protein that catalyzes fatty acid synthesis in adipose tissue, were both analyzed. Results are normalized based on the fold expression of an untreated control group of MVFs at day 0. Results are reported as mean ± standard error (n = 4). Different letters indicate statistically significant differences between treatments (p < 0.05) within a fat source. Statistically significant differences between treatments (p < 0.05) among fat sources indicated by: *different from the same treatment with EPI as the source of MVFs; ^#different from the same treatment with ING fat as the source of MVFs; ^&different from the same treatment with SUBQ as the source of MVFs. FAS, fatty acid synthase.

Discussion

The development of a successful tissue-engineered therapy to treat soft tissue defects will likely be at least partially attributable to a level of graft perfusion capable of supporting viability and volume retention. This notion is not overlooked as tissue-engineered adipose tissue grafts are progressing, however, the task of providing and sustaining perfusion support concurrent with adipocytes is not trivial, and often involves the implementation of a combinatorial approach of bioactive materials and cells. The experiments described herein endeavored to achieve the goal of developing a vascularized adipose tissue construct with a more straightforward approach, using only microvessels derived from adipose tissue as the source of both microvascular support and adipocytes. The idea that microvessels have the ability to play this role is logical given that: (1) angiogenesis and adipogenesis coincide during adipocyte tissue expansion, and (2) microvessels contain a heterogeneous mixture of cells, to include stem and progenitor cells with adipogenic potential.^13,15–20 Therefore, the straightforward approach taken herein is essentially an exploitation of the inherent properties of microvessels derived from adipose tissue. Furthermore, the ability to use multiple sources of adipose tissue lends itself to a degree of flexibility with regard to both disease modeling, as well as clinical applications.

A salient finding of the current study is that the timing of adipogenic induction is critical to achieve a balance between angiogenesis and adipogenesis. Specifically, 14 days of adipogenic induction was effective at promoting adipogenesis as evidenced by high levels of lipid loading (Figs. 2 and 3A), lipolysis (Fig. 4A), and adipogenesis-related gene expression (Fig. 6), but at the expense of angiogenesis, which was slightly reduced as evidenced by a lesser network formation (Figs. 2 and 3B). A possible explanation for this phenomenon is that the commitment of stem/progenitor cells within the microvessels toward an adipogenic fate negates their contribution to angiogenesis, a contention that is supported by previous observations where it is seen that epimorphic regeneration is largely limited by an irreversible differentiation process.²³

Whether or not the potential for adipogenesis will be inhibited if a longer duration of angiogenesis is allowed before adipogenesis is induced remains to be determined. For example, there was a ∼2.5-fold increase in angiogenesis between 7 and 21 days of culture in GM only with epdidymal-derived MVFs indicating there was still room for microvascular angiogenesis at the time that adipogenesis was induced (at 7 days). When adipogenesis was induced after 7 days of GM and allowed to occur for 14 days, the lipid content was equivalent to that which occurred without the 7 days of angiogenic growth (Fig. 3A). Conversely, ING tissue-derived MVF angiogenesis was already high on day 7, and when adipogenesis was induced after 7 days of GM and allowed to occur for 14 days the lipid content was less than that which occurred without the 7 days of angiogenic growth (Fig. 3A).

Regardless, taken together with the deleterious effects of adipogenesis on angiogenesis described above, it is plausible to speculate that adipogenesis is less affected by prior angiogenesis, than is angiogenesis by prior adipogenesis. The observation that cells of the MVFs, including perivascular cells, which have been shown to have adipogenic potential, are present after 8 days of culture and increase over time when implanted, indicate that there is a source of cells available for adipogenesis even after angiogenesis has commenced.^12,24 Future experiments directed toward optimizing the timing for adipogenic induction for the maximization of either angiogenesis or adipogenesis, and the specific delineation of the cells involved will provide valuable information toward optimal therapy development. Regardless, the observation that the level of angiogenesis was similar between 7 days of GM culture and 7 days of GM culture followed by 14 days of adipogenic induction (Fig. 3B) support the contention that angiogenesis was at least maintained, and provides the impetus for future exploration of this phenomenon.

It is worth pointing out that the importance of adipose tissue extends beyond its contribution to the structural integrity of soft tissue replacements. Adipose tissue is a highly active metabolic tissue with the ability to affect a vast number of both local and distant cells and tissues.²⁵ In this regard, it is especially important to consider that in addition to a high lipid content with the addition of adipogenic induction conditions (Figs. 2 and 3A), there was a considerable increase in lipolysis when adipogenic conditions were employed (Fig. 4), which suggests functional adipocytes were generated, for lipolysis is a characteristic metabolic process of adipocytes.²⁶ Notably, lipolysis occurred in all ADM-treated groups, and was further exemplified when exposed to isoproterenol, a lipolysis stimulator.²⁷

Perhaps even more important, especially when taking into consideration the need for maintaining the functionality of adipocytes when a vasculature is present, is the observation that allowing for angiogenesis to commence before adipogenic induction had no apparent deleterious effect on adipocyte function (Fig. 4B vs. Fig. 4A). This maintenance of functionality is in agreement with the intimate association of adipocytes with microvessels in vivo, and the crosstalk between adipocytes and cells of the microvasculature.^18,20,28 Collectively, these encouraging findings suggest that further improvements in both angiogenesis and adipogenesis may lead to the generation of large volumes of metabolically active adipose tissue.

Although one of the major focuses of adipose tissue engineering is the development of therapies for the treatment of soft tissue defects, the potential for adipose tissue-engineered constructs to be used as a means to better understand disease should not be overlooked. In this regard, the findings herein support the use of a variety of adipose tissue sources for generating vascularized adipose tissue. Specifically, the SUBQ depots (ING and SUBQ) are clinically translatable given their ease of access for human tissue engineering applications. The findings herein provide the impetus for future experiments to investigate the adipogenic potential of MVFs sourced from human SUBQ depots. Conversely, the visceral depot (EPI) is implicated in the progression of metabolic disease, such as type 2 diabetes, and may provide a relevant means to model pertinent diseases in vitro.²⁹

The adipogenic induction of MVFs after a period of angiogenesis occurred (i.e., GM+ADM) resulted in few differences in adipogenesis or angiogenesis among the fat sources studied, with the most notable difference being a larger increase in adipogenesis with EPI-derived MVFs (Fig. 3). Nonetheless, differences in adipose tissue expansion among adipose tissue depots correspond with a depot's angiogenic capacity.^19,30,31 This understanding of the parallel changes in adipose tissue expansion concomitant with adipose tissue growth has fundamental implications for gaining a more comprehensive view of adipose tissue in the context of metabolic disease. Although there were no striking differences among the adipose depots in the current study where healthy animals were used as a source of MVFs, the development of the approach described herein whereby both angiogenesis and adipogenesis can be evaluated using MVFs from diseased subjects may provide a means to interrogate mechanisms that underly the progression of metabolic diseases.

Conclusion

MVFs may be used as a sole source of material for engineering vascularized adipose tissue. The generation of vascularized adipose tissue using MVFs derived from a variety of adipose tissue depots was successful when adipogenesis was initiated after the MVFs had undergone a period of angiogenesis, which suggests that cells within the MVFs retain the ability to form adipocytes. This approach to vascularized adipose tissue engineering provides an alternative strategy in the pursuit of generating large volumes of functional adipose tissue to treat soft tissue defects.

Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the National Institutes of Health (SC1DK122578 and 5R01EB020604), Veterans Administration (5 I01 BX000418-06), the University of Texas System Science and Technology Acquisition and Retention Program, the University of Texas at San Antonio Department of Biomedical Engineering, and the University of Texas at San Antonio Research Initiative for Scientific Enhancement Research Training Program (NIH GM060655).