Here, we present a detailed protocol outlining the use of microvascular fragments isolated from rodent or human fat tissue as a straightforward approach to engineer functional, vascularized beige adipose tissue.
Engineering thermogenic adipose tissue (e.g., beige or brown adipose tissues) has been investigated as a potential therapy for metabolic diseases or for the design of personalized microtissues for health screening and drug testing. Current strategies are often quite complex and fail to accurately fully depict the multicellular and functional properties of thermogenic adipose tissue. Microvascular fragments, small intact microvessels comprised of arteriole, venules, and capillaries isolated from adipose tissue, serve as a single autologous source of cells that enable vascularization and adipose tissue formation. This article describes methods for optimizing culture conditions to enable the generation of three-dimensional, vascularized, and functional thermogenic adipose tissues from microvascular fragments, including protocols for isolating microvascular fragments from adipose tissue and culture conditions. Additionally, best practices are discussed, as are techniques for characterizing the engineered tissues, and sample results from both rodent and human microvascular fragments are provided. This approach has the potential to be utilized for the understanding and development of treatments for obesity and metabolic disease.
The goal of this protocol is to describe an approach for developing vascularized beige adipose tissue from a single, potentially autologous source, microvascular fragment (MVF). Brown and beige adipose tissues have been demonstrated to display beneficial properties related to metabolic regulation; however, the small volume of these adipose tissue depots in adults limits the potential impact on systemic metabolism, particularly in diseased conditions such as obesity or type 2 diabetes1,2,3,4,5,6,7. There is significant interest in brown/beige fat as a therapeutic target for preventing the harmful metabolic effects linked with obesity and its comorbidities8,9,10,11,12.
MVFs are vessel structures that can be directly isolated from adipose tissue, cultured, and maintained in a three-dimensional configuration for extended periods of time13,14,15. Previous work from our group, and others, have begun to exploit the multicellular and multipotent capacity of MVFs, specifically as it relates to adipose tissue formation16,17,18. As a buildup of this work, we recently demonstrated that MVFs derived from rodent models of healthy and type 2 diabetes19 and from human subjects (adults over 50 years of age)20 contained cells capable of being induced to form thermogenic, or beige, adipose tissue.
Herein is an innovative approach from which a single source MVF is utilized, not only capable of creating beige adipose tissue but also its associated and critical vascular component21. The use of this technique could be of great value for studies looking for a straightforward tissue-engineered approach for thermogenic adipose tissue formation. Unlike other methods aspiring to engineer beige adipose tissue22,23,24,25,26,27,28, the process described in this study does not require using multiple cell types or complex induction regimens. Vascularized beige and white fat models can be created with MVFs originating from rodent and human sources, demonstrating great translation potential. The end product of this protocol is an engineered beige thermogenic fat tissue with a structure and metabolic function comparable to brown adipose tissue. Overall, this protocol presents the idea that an easily accessible and possibly autologous source MVF may be a worthwhile therapeutic intervention and tool for studying metabolic disorders.
This study was 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. All animal procedures were approved by the Institutional Animal Care and Use Committee at the University of Texas at San Antonio.
NOTE: For the steps described below, male Lewis Rats are utilized. Slight protocol adjustments must be made for a female, as well as mouse microvascular fragment (MVF) collection29. For protocols using human MVFs (h-MVFs), the only steps required are the resuspension of h-MVFs following the manufacturer's protocol, growth media preparation (1.3), formation of fibrin hydrogels (5), and culturing (6). For an overview of the protocol, please see Figure 1.
Figure 1: Experimental outline. Breakdown of six key steps, prior to analysis, for the formation of thermogenic adipose tissue using MVF. Please click here to view a larger version of this figure.
1. Reagent preparation
NOTE: Reagents below correspond to one rat, weighed and made inside a biohood.
2. Tool/materials preparation
NOTE: All instrumentation should be autoclaved/sterilized prior to use.
3. Fat isolation protocol
Figure 2: Isolation of different adipose tissue depots. (A) Initial incisions needed for excision of the inguinal adipose tissue. (B) Location of the inguinal fat depot. (C) Location of the epididymal fat depot, noting incision of the outer skin needed for access. (D) Additional incisions needed once the mice are placed prone to access additional fat. (E) Location of the posterior subcutaneous fat depot. Please click here to view a larger version of this figure.
4. Microvascular fragment isolation protocol
Figure 3: Isolation of MVFs. (A) Post digestion of the adipose tissue, depiction of the separation of MVFs containing pellet and supernatant following a spin-down. (B) Layout of supplies for filtration and entrapment of MVFs. (C) Illustration of the concentric circle method for filtration/washing steps. Please click here to view a larger version of this figure.
5. Formation of fibrin hydrogels
Figure 4: Formation of MVF fibrin gels. (A) A 5/7-part thrombin mixture gets pipetted into the corresponding well. (B) Next, with a clipped pipet tip (to not disturb MVFs), a 2/7-part fibrinogen+MVF mixture is pipetted into the well and gently mixed. (C) Lastly, all completed gels are placed into an incubator at 37 °C, allowing hydrogel to solidify fully before the media is placed on top. Please click here to view a larger version of this figure.
6. Culturing conditions of MVFs
Figure 5: Timing for non-vascularized adipose tissue formation. This figure has been modified from Acosta et al.19. Please click here to view a larger version of this figure.
Figure 6: Timing for vascularized adipose tissue formation. This figure has been modified from Acosta et al.19. Please click here to view a larger version of this figure.
There are a few key phenotypic morphological characteristics of beige/brown adipose tissue: it is multilocular/contains small lipid droplets, possesses a large number of mitochondria (the reason for its characteristically "brownish" appearance in vivo), correspondingly has a high oxygen consumption rate/mitochondrial bioenergetics, is highly vascularized, has increased lipolysis/insulin-stimulated glucose uptake, and, most notoriously, expresses high levels of uncoupling protein 1 (UCP1), a mitochondrial protein involved in thermogenic respiration19,30.
Accordingly, when characterizing the ability of MVFs to differentiate into beige adipose tissue, an analysis was conducted which would allow us to visualize (Figure 7, Figure 8), genetically confirm (Figure 9, Figure 10), and functionally measure (Figure 11) the transformation of the MVF.
In Figure 7 and Figure 8, through the use of BODIPY, a lipid stain, and imaging of the hydrogels using confocal microscopy, the visualization of the sizing and location of the lipids in the differentiated adipocytes was seen16,17,19. Notably, in this analysis, especially in comparison between conditions, the BAM groups should display smaller lipid sizes (indicative of beige adipose tissue formation), quantifiable through NIH ImageJ19,20.
Using RT-qPCR16,19,20, in Figure 9 and Figure 10, most distinctively, the expression of UCP1 is, across the board, increased significantly upon MVF exposure to BAM.
Lastly, looking at mitochondrial bioenergetics (Figure 11), it is evident that BAM groups have characteristically higher oxygen consumption rate (OCR) levels, measured using a Seahorse mito stress test19,20.
Figure 7: r-MVF histological evaluation. Lean or type 2 diabetic derived r-MVFs were exposed to either direct (upper panel, ADIPO) or indirect (lower panel, ANGIO+ADIPO) WAM or BAM adipogenic media to obtain non-vascularized or vascularized white or beige adipose tissue, respectively (scale bars = 200 µm). This figure has been modified from Acosta et al.19. Please click here to view a larger version of this figure.
Figure 8: h-MVF histological evaluation. h-MVF were exposed to direct (ADIPO) WAM or BAM to obtain non-vascularized white or beige adipose tissue, respectively (scale bars = 200 µm). This figure has been modified from Gonzalez Porras et al.20. Please click here to view a larger version of this figure.
Figure 9: r-MVF evaluation through RT-qPCR. Lean (L) or type 2 diabetic (Db) derived r-MVF exposed to either (A–C) direct or (E–G) indirect WAM or BAM were evaluated for (A,E) adipogenesis (Adiponectin), (B,F) thermogenesis (UCP1), and (C,G) angiogenesis (ANGPT1). Results are reported as mean ± standard error of two experimental replicates (n = 4). * = p < 0.05, ** = p < 0.01, *** = p < 0.001, and **** = p < 0.0001. D0 = Day 0. Two-way analysis of variance (ANOVA) tests with Holm-Sidak's multiple comparison analyses to determine differences between groups. Statistical significance was defined as p < 0.05. This figure has been modified from Acosta et al.19. Please click here to view a larger version of this figure.
Figure 10: h-MVF evaluation through RT-qPCR. h-MVFs exposed to direct WAM or BAM were evaluated for (A) adipogenesis (Adiponectin) and (B) thermogenesis (UCP1). Results are reported as mean ± standard error of two experimental replicates (n = 4). * = p < 0.05, ** = p < 0.01, *** = p < 0.001, and **** = p < 0.0001. One-way analysis of variance (ANOVA) tests with Holm-Sidak's multiple comparison analyses to determine differences between groups. Statistical significance was defined as p < 0.05. This figure has been modified from Gonzalez Porras et al.20. Please click here to view a larger version of this figure.
Figure 11: r-MVF and h-MVF functional assessment. Lean (L) or type 2 diabetic (Db) derived r-MVFs exposed to either (A) direct or (B) indirect WAM or BAM or (C) h-MVFs exposed to direct WAM or BAM were evaluated functionally through measurement of oxygen consumption rate (OCR). Results are reported as mean ± standard error of two experimental replicates (n = 4). This figure has been modified from Acosta et al.19. and Gonzalez Porras et al.20. Please click here to view a larger version of this figure.
The field of brown/beige adipose tissue engineering is largely immature22,23,24,25,26,27,28, with the bulk of adipose models being developed for white adipose tissue8,22,31. Engineered brown/beige microtissues typically consist of multiple cell sources or genetic alterations to obtain a subset of the phenotypic characteristics of native brown adipose tissue8,11,32. The approach described herein presents a single-sourced, potentially autologous18,33 way of creating functional, structurally relevant, and vascularized beige fat utilizing microvascular fragments (MVFs). MVFs are most notably known as a source of white adipose tissue formation14,16,17,18,34, though we have recently demonstrated their ability for beige fat formation stemming from rodent, human, and diseased sources, as shown here19,20. Given the considerable interest in utilizing beige/brown fat for its therapeutic or disease modeling potential, this technique has far-fetching applications in the fields of metabolism, obesity, and related disorders.
There are several key points described in the protocol. First, there are differences between the utilization of rat versus human MVFs. The use of rodent-derived (either from mice29 or rats) has, thus far, largely dominated research with MVFs, with work looking at them in a multitude of conditions such as obesity35, type 1 diabetes36,37, type 2 diabetes19, and aging38, and even differences between adipose depots39 or gender40. Though MVFs, given their origin, can be isolated autologously from the subcutaneous adipose depots of adults using standard minimally-invasive procedures41, MVF-based vascularization has not been performed in clinical practice. However, preclinical studies where human MVFs were harvested from lipoaspirate have demonstrated its possibility18,33. For our group specifically, as shown in the representative data, the formation of vascularized beige adipose tissue is currently limited to MVFs derived from rodents. As previously demonstrated by our group and others18, obtaining the balance of vessel growth and adipocyte differentiation is highly sensitive, proven to be dependent on the factors introduced42, and timepoint differentiation is provoked16. A limitation of the protocol described is that further development is needed to optimize the conditions conducive to vascularized h-MVF beige adipose tissue formation. Additionally, further work looking at the response of these scaffolds in vivo and deriving from other diseased states, along with associated optimization steps are needed.
Moreover, described here is a protocol for the isolation of r-MVFs from three different adipose tissue depots in male rats. Previous work from Später et al.39 discussed differences between the vascularization ability of visceral versus subcutaneous depot-derived MVFs, noting that subcutaneous depot MVFs had a decreased ability to vascularize, a feature they attributed to excess connective tissue contamination. It should be noted that, for our studies, as presented here in the "representative data," only the inguinal subcutaneous and posterior subcutaneous depots were utilized. The choice to solely use subcutaneous-derived MVFs was made to more closely mimic translational studies where lipoaspirate, or similar procedures, collect exclusively subcutaneous adipose tissue. Additionally, the fact that in vivo studies pin-pointing beige adipose tissue are being developed within subcutaneous adipose tissue, which contains a distinct subset of preadipocytes or white adipocytes that transdifferentiate, provided a further rationale for our decision43. Previous work from our group showed no discernable differences among the MVFs originating from either visceral or subcutaneous depots of healthy rodents to undergo both angiogenesis and white adipose tissue formation16. All these variables should be considered when designing future studies.
Lastly, when attempting to either modify or troubleshoot the method described, a few essential points should be deliberated. First, the step of enzymatically digesting the adipose tissue is extremely important; special care should be taken and optimization ensured to consistently reproduce MVFs of similar size and quality. Given the large variation between fat types and fat volumes (highly dependent on animal age, size, health, and care taken at the time of fat isolation [avoidance of contaminants and extraction efficiency]), the time of digestion can vary, thus ranges, which best fit with our equipment/animals are provided. However, customization should be considered for optimal results. Additionally, when handling the MVF, during post-enzymatic digestion, special care should be taken to avoid unnecessary roughness and further breaking up fragments. Finally, one should be aware that media formulations, the hydrogel of choice44, and culture conditions are highly customizable based on intended outcomes. As shown here, MVFs derived from different sources (e.g., lean vs. diabetic MVFs) have different degrees of differentiation, so when designing experiments, proper controls and experimental groups should be included.
In conclusion, as the field of engineering thermogenic adipose tissue grows, it is critical to construct biological-relevant systems that structurally, genetically, and functionally mimic native beige/brown adipose tissue. MVFs present an exciting and unique approach to this challenge, as they, as described here, provide a simple sole-source method to create biological imitations of beige fat. Therefore, they hold significant potential for utilization of the understanding or development of treatments for obesity and metabolic disease.
The authors have nothing to disclose.
Dr. Acosta is supported by the National Institutes of Health grants CA148724 and TL1TR002647. Dr. Gonzalez Porras is supported by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health, under Award Number F32-0DK122754. This work was supported, in part, by the National Institutes of Health (5SC1DK122578) and the University of Texas at San Antonio Department of Biomedical Engineering. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Figures were partially created with Biorender.com.
Aminocaproic Acid | Sigma Aldrich | A2504-100G | Added in DMEM at the concentration of 1 mg/mL |
Blunt-Tipped Scissors | Fisher scientific | 12-000-172 | Sterilize in autoclave |
Bovin Serum Albumin (BSA) | Millipore | 126575-10GM | Diluted in PBS to 4 mg/mL and 1 mg/mL |
Collagenase Type 1 | Fisher scientific | NC9633623 | Diluted to 6 mg/mL in BSA 4 mg/mL, Digestion of minced fat |
Dexamethasone | Thermo Scientific | AC230302500 | Diluted in ethanol at a 2 mg/ml stock concentration |
Disposable underpads | Fisher scientific | 23-666-062 | For fluid absorption during surgery |
Dissecting Scissors | Fisher scientific | 08-951-5 | Sterilize in autoclave |
Dulbecco′s Modified Eagle′s Medium (DMEM) | Fisher scientific | 11885092 | |
Dulbecco′s Modified Eagle′s Medium/Nutrient Mixture F-12 Ham (DMEM/F12) | Sigma Aldrich | D8062 | |
Fetal Bovine Serum | Fisher scientific | 16140089 | Added in DMEM to 20% v/v. |
Fibrinogen | Sigma Aldrich | F8630-25G | Solubilized in DMEM at the concentration of 20 mg/mL, Protein found in blood plasma and main component of hydrogel |
Flask, 250 mL | Fisher scientific | FB500250 | Allows for digestion of fat using a large surface area |
Forceps | Fisher scientific | 50-264-21 | Sterilize in autoclave, For handling of tissue and filters |
Forskolin | Sigma Aldrich | F6886 | Diluted in ethanol at a 10 mM stock concentration |
Human MVF | Advanced Solutions Life Scienes, LLC | https://www.advancedsolutions.com/microvessels | Human MVFs (hMVFs) isolated from three different patients (52-, 54-, and 56-year old females) were used in the current study. |
Indomethacine | Sigma Aldrich | I7378 | Diluted in ethanol at a 12.5 mM stock concentration |
Insulin from porcine pancreas | Sigma Aldrich | I5523 | Diluted in 0.01 N HCl at a 5 mg/ml stock concentration |
MycoZap | Fisher scientific | NC9023832 | Added in DMEM to 0.2% w/v, Mycoplasma Prophylactic |
Pennycilin/Streptomycin (10,000 U/mL) | Fisher scientific | 15140122 | Added in DMEM to 1% v/v. |
Petri dishes, polystyrene (100 mm x 15 mm). | Fisher scientific | 351029 | 3 for removal of blood vessels and mincing, 8 (lid) for presoaking of screens & 8 (dish) for use when filtering with 500 or 37 µM screens |
Petri dishes, polystyrene (35 mm x 10 mm). | Fisher scientific | 50-202-036 | For counting fragments |
Phosphate Buffer Saline (PBS) | Fisher scientific | 14-190-250 | Diluted to 1x with sterile deionized water. |
Rat Clippers (Andwin Mini Arco Pet Trimmer) | Fisher scientific | NC0854141 | |
Rosiglitazone | Fisher scientific | R0106200MG | Diluted in DMSO at a 10 mM stock concentration |
Scissors | Fine Science Tools | 14059-11 | 1 for initial incision, 1 for epididymal incision, 1 for tip clipping |
Screen 37 µM | Carolina Biological Supply Company | 652222R | Cut into 3" rounded squares and sterilized in ethylene oxide, Fragment entrapment and removal of very small fragments/single cells and debris |
Screen 500 µM | Carolina Biological Supply Company | 652222F | Cut into 3" rounded squares and sterilized in ethylene oxide, Removes larger fragments/debris |
Serrated Hemostat | Fisher scientific | 12-000-171 | Sterilize in autoclave, For clamping of skin before incision |
Steriflip Filter 0.22 μm | Millipore | SE1M179M6 | |
Thrombin | Fisher scientific | 6051601KU | Diluted in deionzed water to 10 U/mL, Used as a clotting agent turning fibrinogen to fibrin |
Thyroid hormone (T3) | Sigma Aldrich | T2877 | Diluted in 1N NaOH at a 0.02 mM stock concentration |
Zucker diabetic fatty (ZDF) rats – obese (FA/FA) or lean (FA/+) male | Charles River | https://www.criver.com/products-services/find-model/zdf-rat-lean-fa?region=3611 https://www.criver.com/products-services/find-model/zdf-rat-obese?region=3611 |
Obtained from Charles River (Wilmington, MA). Rats were acquired at 4 weeks of age and fed Purina 5008 until euthanasia (15-19 weeks of age). Glucose levels (blood from the lateral saphenous vein) were greater than 300 mg/dL in all FA/FA rats used in the study. All animals were housed in a temperature-controlled environment with a 12-h light-dark cycle and fed ad libitum. |