We present a protocol for the isolation, culture, and adipogenic induction of neural crest derived adipose-derived stem cells (NCADSCs) from the periaortic adipose tissue of Wnt-1 Cre+/-;Rosa26RFP/+ mice. The NCADSCs can be an easily accessible source of ADSCs for modeling adipogenesis or lipogenesis in vitro.
Cite this ArticleCopy Citation | Download Citations | Reprints and Permissions
Qi, Y., Miao, X., Xu, L., Fu, M., Peng, S., Shi, K., Li, J., Ye, M., Li, R. Isolation, Culture, and Adipogenic Induction of Neural Crest Original Adipose-Derived Stem Cells from Periaortic Adipose Tissue. J. Vis. Exp. (157), e60691, doi:10.3791/60691 (2020).
Translate text to:
An excessive amount of adipose tissue surrounding the blood vessels (perivascular adipose tissue, also known as PVAT) is associated with a high risk of cardiovascular disease. ADSCs derived from different adipose tissues show distinct features, and those from the PVAT have not been well characterized. In a recent study, we reported that some ADSCs in the periaortic arch adipose tissue (PAAT) descend from the neural crest cells (NCCs), a transient population of migratory cells originating from the ectoderm.
In this paper, we describe a protocol for isolating red fluorescent protein (RFP)-labeled NCCs from the PAAT of Wnt-1 Cre+/-;Rosa26RFP/+ mice and inducing their adipogenic differentiation in vitro. Briefly, the stromal vascular fraction (SVF) is enzymatically dissociated from the PAAT, and the RFP+ neural crest derived ADSCs (NCADSCs) are isolated by fluorescence activated cell sorting (FACS). The NCADSCs differentiate into both brown and white adipocytes, can be cryopreserved, and retain their adipogenic potential for ~3–5 passages. Our protocol can generate abundant ADSCs from the PVAT for modeling PVAT adipogenesis or lipogenesis in vitro. Thus, these NCADSCs can provide a valuable system for studying the molecular switches involved in PVAT differentiation.
The prevalence of obesity is increasing worldwide, which increases the risk of related chronic diseases, including cardiovascular disease and diabetes1. PVAT surrounds blood vessels and is a major source of endocrine and paracrine factors involved in vasculature function. Clinical studies show that high PVAT content is an independent risk factor of cardiovascular disease2,3, and its pathological function depends on the phenotype of the constituent adipose-derived stem cells (ADSCs)4.
Although ADSC cell lines like the murine 3T3-L1, 3T3-F442A, and OP9 are useful cellular models to study adipogenesis or lipogenesis5, regulatory mechanisms for adipogenesis differ between cell lines and primary cells. The ADSCs in the stromal vascular cell fraction (SVF) isolated directly from adipose tissues and induced to differentiate into adipocytes most likely recapitulate in vivo adipogenesis and lipogenesis6. However, the fragility, buoyancy, and the variations in size and immunophenotypes of the ADSCs make their direct isolation challenging. In addition, the different isolation procedures can also significantly affect the phenotype and adipogenic potential ability of these cells7, thus emphasizing the need for a protocol that maintains ADSC integrity.
Adipose tissue is typically classified as either the morphologically and functionally distinct white adipose tissue (WAT), or the brown adipose tissue (BAT)8, which harbors distinct ADSCs9. While ADSCs isolated from perigonadal and inguinal subcutaneous WATs have been characterized in previous studies9,10,11,12, less is known regarding the ADSCs from PVAT that is mainly composed of BAT13.
In a recent study, we found that a portion of the resident ADSCs in the periaortic arch adipose tissue (PAAT) are derived from neural crest cells (NCCs), a transient population of migratory progenitor cells that originate from the ectoderm14,15. Wnt1-Cre transgenic mice were used for tracing neural crest cell development16,17. We crossed Wnt1-Cre+ mice with Rosa26RFP/+ mice to generate Wnt-1 Cre+/-;Rosa26RFP/+ mice, in which NCCs and their descendants are labeled with red fluorescent protein (RFP) and are easily tracked in vivo and in vitro15. Here, we describe a method for isolating neural crest derived ADSCs (NC-derived ADSCs, or NCADSCs) from mouse PAAT and induce the NCADSCs to differentiate into white adipocytes or brown adipocytes.
The animal protocol has been reviewed and approved by the Animal Care Committee of Shanghai Jiao Tong University.
1. Generation of Wnt-1 Cre+/-;Rosa26RFP/+ Mice
- Cross Wnt-1 Cre+/- mice16 with Rosa26RFP/+ mice18 to generate Wnt-1 Cre+/-;Rosa26RFP/+ mice. House mice under a 12 h light/dark cycle in a pathogen-free facility at 25 °C and 45% humidity until they are 4–8 weeks old.
2. Dissection of the PAAT
NOTE: See Figure 1.
- Sterilize all surgical tools (e.g., surgical scissors, standard forceps, and microsurgical scissors and forceps) by autoclaving at 121 °C for 30 min.
- Prepare the digestion medium (High glucose Dulbecco's Modified Eagle Medium [HDMEM] containing 2 mg/mL collagenase type I). Prepare the culture medium (HDMEM containing 10% fetal bovine serum [FBS] and 1% v/v penicillin-streptomycin [PS]). Sterilize using a 0.22 μm syringe filter before use.
- Sterilize the cell culture reagents using UV, ethanol, filtration, or steam, as appropriate.
- Prepare a Petri dish with Hanks' Balanced Saline Solution (HBSS) and a 15 mL conical tube with 10 mL of HBSS supplemented with 1% v/v of PS solution. Keep both on ice.
- Anesthetize the mice15 with isoflurane and sacrifice by cervical dislocation. Immerse the bodies in a beaker filled with 75% alcohol (200 mL) for 5 min to sterilize the skin surface.
- Snip and separate the skin on the abdomen and cut along the ventral midline from the pelvis to the neck. Open the abdomen and move the liver to expose the diaphragm19.
- Cut the diaphragm and the ribs on both sides of the midline and expose the heart and lungs by peeling back the ribs.
- Remove the lung and thymus and extract the PAAT along with the aorta and heart.
- Cut off the aorta at the aortic root to remove the heart. Make a cut between the aortic arch and descending aorta and carefully separate the adipose tissue surrounding both structures and the left and right common carotid arteries from the posterior chest wall. Transfer the tissue to the ice-cold HBSS buffer in the Petri dish.
- Using sterile forceps, remove as much of the vasculature (e.g., aorta, common carotid arteries, and other small vasculature) and fascia as possible, and transfer the adipose tissue into the 2 mL Eppendorf tubes contain 0.5 mL ice-cold HBSS buffer on ice.
3. Isolation of the SVF
Collect the PAAT of 5-6 mice into one 2 mL microcentrifuge tube containing 1 mL freshly prepared digestion medium and mince the tissue using surgical scissors in an Eppendorf tube at room temperature (RT).
- Transfer the mix into 50 mL tubes containing 9 mL of the digestion medium. Homogenize the tissues by pipetting up and down with a 1 mL pipette 10x.
- Incubate the tubes at 37 °C with constant shaking at 100 rpm for 30–45 min and check every 5–10 min to prevent overdigestion. This is critical to improving cell viability and yield.
NOTE: Good tissue digestion will result in a homogenous, light yellow adipose tissue that is visible to the naked eye upon gently swirling the tube.
- Stop the digestion by adding 5 mL of HDMEM containing 10% FBS and 1% v/v PS at RT and mix well by pipetting.
- Centrifuge the cell suspension at 500 x g for 5 min at RT. The SVF will be visible as a brownish pellet. Carefully aspirate the floating adipocytes and decant the remaining supernatant without disturbing the SVF. Dissolve the SVF pellet in 10 mL of culture medium and filter through a 70 µm cell strainer.
- Centrifuge the cell suspension at 500 x g for 5 min, remove the supernatant, and gently resuspend the pellet in 5 mL of erythrocyte lysis buffer in a 15 mL conical tube for 10 min at RT.
- Stop the reaction by adding 10 mL of 1x PBS containing 1% FBS. Centrifuge the cell suspension at 500 x g for 5 min at 4 °C, remove the supernatant, and resuspend the pellet in 10 mL of 1x PBS containing 1% FBS.
- Centrifuge the cells again at 500 x g for 5 min at 4 °C. Remove the supernatant and resuspend the pellet in 5 mL of culture medium in a 15 mL conical tube at 4 °C.
- After a final round of centrifugation (500 x g for 5 min at 4 °C), resuspend the pelleted cells in 5 mL of FACS buffer (PBS containing 10% FBS, 100 units/mL DNA I, and 1% v/v PS) on ice, and count the cells with a hemocytometer.
4. Isolation of NCADSCs by FACS
- Set up and optimize the cell sorter following the instruction manual. Select the 100 µm nozzle, sterilize the collection tubes, install the required collection device, and set up the side streams20.
- A 561 nm yellow/green laser and optical filter 579/16 are recommended for sorting RFP+ cells. Perform the compensation using the negative control and the single-stained positive controls. See Figure 2A for the gating scheme.
- Filter the cells through a 40 µm strainer, centrifuge at 500 x g for 5 min, and resuspend the cells in 2 mL of FACS buffer at a density of 0.5–1 x 107/mL. Transfer the cells to clearly labeled 5 mL round bottom polystyrene tubes, and load into the sorter.
- Run the experimental sample tube at 4 °C, turn on the deflection plates, and sort into a 15 mL conical tube precoated with RPMI containing 1% FBS and 1% v/v PS.
NOTE: Protect the samples from strong light to minimize RFP quenching.
5. Culture of NCADSCs
- Plate the sorted cells at a density of 5,000 cells/cm2 in a 12 well culture plate in complete culture medium and incubate at 37 °C in a humid atmosphere with 5% CO2 for 20–24 h.
- Remove the culture medium, wash the cells with prewarmed (37 °C) PBS to remove cell debris and add fresh culture medium.
- Once the cells are 80–90% confluent, digest the monolayer using a 0.25% trypsin EDTA solution at 37 °C in an incubator for 3–5 min, and neutralize with 2 mL of culture medium.
- Centrifuge the harvested cells for 15 min at 250 x g at RT, remove the supernatant, and resuspend the cells in 1 mL of culture medium. Count the cells with a hemocytometer.
- Seed the cells in a 12 well culture plate at the density of 5,000 cells/cm2.
- Resuspend the remaining cells in culture medium containing 10% DMSO, freeze, and store in liquid nitrogen.
6. Adipogenic Induction of NCADSCs
- Induce adipogenic differentiation of the NCADSCs at 80–90% confluency and standard culture conditions21.
- For brown adipogenic induction, first treat the cultured cells with brown adipogenic induction medium (HDMEM, 10% FBS, 1% v/v PS, 0.5 mM/L IBMX, 0.1μM/L dexamethasone, 1 μM/L rosiglitazone, 10 nmol/L triiodothyronine, and 1 μg/mL insulin) for 2 days. Wash the cells with PBS 2x and replace with fresh medium (HDMEM, 10% FBS, 1% v/v PS, 1 μM/L rosiglitazone, 10 nmol/L triiodothyronine, and 1 μg/mL insulin). Change this medium every 2 days for a total of 3–5x.
- For white adipogenic induction, first treat the cells with white adipogenic induction medium (HDMEM, 10% FBS, 1% v/v PS, 0.5 mM/L IBMX, 0.1 μM/L dexamethasone, and 1 μg/mL insulin) for 2 days. Wash the cells with PBS 2x and replace with fresh medium (HDMEM, 10% FBS, 1% v/v PS, and 1 μg/mL insulin). Change this medium every 2 days for a total of 3-5x.
- Analyze the adipogenic cells as appropriate.
NOTE: Be gentle when washing the cells with PBS. Differentiated adipocytes can easily wash away.
Using the protocol described above, we obtained ~0.5–1.0 x 106 ADSCs from 5–6 Wnt-1 Cre+/-;Rosa26RFP/+ mice (48 weeks old, male or female).
The flow chart of collection of PAAT from mice is presented in Figure 1. The morphology of the NCADSCs was similar to the ADSC from other mice adipose tissues. The cultured NCADSCs reached 80–90% confluency after 7–8 days of culture, and the NCADSCs had an expanded fibroblast-like morphology (Figure 2B,C).
To further confirm that NCADSCs had adipogenic potential, differentiation of the NCADSCs into white or brown adipocytes was induced. Oil red staining was used to detect the mature adipocytes (Figure 2). The NCADSCs exhibited strong adipogenic potential for both white and brown adipocytes after induction. Mature adipocytes were observed after 8 days of white or brown adipogenic induction, with over 60% of the NCADSCs showing adipogenic differentiation (Figure 2D,F,H). Prolonging adipogenic induction time improved the harvesting rate of mature adipocytes (data not shown). NCADSCs had greatly reduced adipogenic potential after passaging (Figure 2E,G,H).
Immunoblotting and quantitative real-time PCR (qRT-PCR) (see Supplemental File 1 for primers used) proved that the expression levels of adipocyte-specific relative proteins and genes (Perilipin, PPARγ, Cebp/α) in the adipogenically differentiated NCADSCs significantly increased after 8 days of white adipogenic induction (Figure 3A,B). The qRT-PCR results showed that the induction of adipocyte-specific genes (Perilipin, PPARγ, Cebp/α) and brown adipocyte-specific genes (Pgc1α, UCP-1, PPARα, PRDM16) significantly increased in 8 days of brown adipogenic induction of NCADSCs (Figure 3B,C,D).
Figure 1: Flow chart of collection of PAAT from mice. (A) Anesthetize and sacrifice the Wnt-1 Cre+/-;Rosa26RFP/+ mice and perform longitudinal dissection of the mouse to expose heart and lungs; (B) Remove the lungs and thymus; (C) Expose PAAT, aorta arch, and heart; (D) Remove PVAT, aorta, and heart into precooled HBSS Buffer; (E) Harvest PAAT and transfer into precooled HBSS buffer. H = Heart; AA = Aorta arch; T = Thymus; L = Lung. Please click here to view a larger version of this figure.
Figure 2: Adipogenic differentiation of NCADSCs isolated from PAAT. (A) General gating scheme for characterizing and sorting NCADSCs (RFP) populations. (B) Fluorescence microscope images show that the NCADSCs adhered and expanded after 96 h seeding on a 12 well culture plate. (C–G) Representative images showing that oil red O stained NCADSCs from PAAT after adipogenic induction. (C) Control (no induction). (D) Primary NCADSCs and (E) 3x-passaged NCADSCs after 10 days of white adipogenic induction. (F) Primary NCADSCs and (G) 3x-passaged NCADSCs after 10 days of brown adipogenic induction. (H) Statistical results of the oil red staining area of primary and 3x passaged NCADSCs from PAAT after 8 days of adipogenic induction. n = 6. Values are expressed as mean ± standard deviation (SD). Scale bar = 50 µm. Please click here to view a larger version of this figure.
Figure 3: Characterization of the white and brown adipogenic induction of NCADSCs. (A) Immunoblot showing expression levels of adipocyte-specific proteins (Perilipin, PPARγ, Cebp/α) in the white adipogenically differentiated NCADSCs. (B) qRT-PCR results showing the induction of adipocyte-specific genes, Cebp/α, PPARγ, Perilipin, Fabp4 in the white and brown adipogenically differentiated NCADSCs. (C) qRT-PCR results showing the induction of brown adipocyte-specific genes, Pgc1α, UCP-1, PPARα, PRDM16 in the white and brown adipogenically differentiated NCADSCs. The expression levels were normalized against HPRT and measured by the Ct (∆∆Ct) method. Representative result of n = 3 independent experiments. Values are expressed as mean ± SD. Unpaired 2-tailed Student t-test was used for comparisons between the two groups. *P < 0.05. Please click here to view a larger version of this figure.
Supplemental File 1. Please click here to view this file (Right click to download).
In this study, we present a reliable method for the isolation, culture, and adipogenic induction of NCADSCs extracted from the PVAT of Wnt-1 Cre+/-;Rosa26RFP/+ transgenic mice designed to produce RFP+ ADSCs. Previous reports show that there is no significant difference in the expression of general multipotent mesenchymal stem cells (MSCs) markers in NCADSCs and non NCADSCs22, and that NCADSCs have a strong potential to differentiate into adipocytes in vitro15,22,23. Thus, the NCADSCs isolated with this protocol should be suitable for most ADSC studies.
The advantage of the present method is that the inherent fluorescent reporter in the transgenic NCADSCs makes the isolation process simple and economical without the need for antibodies or probe-based FACS, or magnetic activated cell sorting24. In addition, the fluorescence intensity of RFP is stronger than FITC, which further improves the efficiency of the FACS.
The key to this protocol is the utilization of young mice. Although older and larger mice can yield a greater amount of adipose tissue, the proportion of NC-derived adipocytes in the PAAT decreases with age because the NCCs primarily contribute to the early development of PAAT15. Thus, the adipogenic potential of these cells declines with age. Based on our experiments, the optimal time window for NCADSC isolation in mice is 4–8 weeks.
Our method is simple, practical, and can generate abundant ADSCs for the study of PVAT adipogenesis or lipogenesis in vitro, and to test novel drugs against obesity and cardiovascular disease. Moreover, the NCADSCs of Wnt-1 Cre+/-;Rosa26RFP/+ mice can also be an effective in vitro system for other research fields. However, several caveats remain: First, these cells are more sensitive and fragile than immortalized adipocyte lines. Second, their high proliferation rate and adipogenic differentiation is counteracted by the fact that they tend to lose their adipogenic potential after a maximum of five passages.
The authors have nothing to disclose.
National Key R&D Program of China (2018YFC1312504), National Natural Science Foundation of China (81970378, 81670360, 81870293), and Science and Technology Commission of Shanghai Municipality (17411971000, 17140902402) provided the funds for this study.
|4% PFA||BBI life sciences||E672002-0500||Lot #: EC11FA0001|
|Agarose||ABCONE (China)||A47902||1% working concentration|
|Anti-cebp/α||ABclonal||A0904||1:1000 working concentration|
|Anti-mouse IgG, HRP-linked||CST||7076||1:5000 working concentration|
|Anti-perilipin||Abcam||AB61682||1 μg/mL working concentration; lot #: GR66486-54|
|Anti-PPARy||SANTA CRUZ||sc-7273||0.2 μg/mL working concentration|
|Anti-rabbit IgG, HRP-linked||CST||7074||1:5000 working concentration|
|Anti-β-Tubulin||CST||2146||1:1000 working concentration|
|BSA||VWR life sciences||0332-100G||50 mg/mL working concentration; lot #: 0536C008|
|Collagenase, Type I||Gibco||17018029|
|Dexamethasone||Sigma-Aldrich||D4902||0.1 µM working concentration|
|Erythrocyte Lysis Buffer||Invitrogen||00-4333|
|FBS||Corning||R35-076-CV||50 mg/mL working concentration; lot #: R2040212FBS|
|HDMEM||Gelifesciences||SH30243.01||Lot #: AD20813268|
|IBMX||Sigma-Aldrich||I7018||0.5 mM working concentration|
|Insulin||Sigma-Aldrich||I3536||1 μg/mL working concentration|
|Microsurgical forceps||Suzhou Mingren Medical Equipment Co.,Ltd. (China)||MR-F201A-1|
|Microsurgical scissor||Suzhou Mingren Medical Equipment Co.,Ltd. (China)||MR-H121A|
|Oil Red O solution||Sigma-Aldrich||O1516||0.3% working concentration|
|PBS (Phosphate buffered saline)||ABCONE (China)||P41970|
|PrimeScript RT reagent Kit||TAKARA||RR047A||Lot #: AK4802|
|RNeasy kit||TAKARA||9767||Lot #: AHF1991D|
|Rosa26RFP/+ mice||JAX||No.007909||C57BL/6 backgroud; male and female|
|Rosiglitazone||Sigma-Aldrich||R2408||1 μM working concentration|
|Standard forceps||Suzhou Mingren Medical Equipment Co.,Ltd. (China)||MR-F424|
|Surgical scissor||Suzhou Mingren Medical Equipment Co.,Ltd. (China)||MR-S231|
|SYBR Premix Ex Taq||TAKARA||RR420A||Lot #: AK9003|
|Triiodothyronine||Sigma-Aldrich||T2877||10 nM working concentration|
|Wnt1-Cre+;PPARγflox/flox mice||JAX||No.009107||C57BL/6 backgroud; male and female|
- Afshin, A., et al. Health Effects of Overweight and Obesity in 195 Countries over 25 Years. New England Journal of Medicine. 377, (1), 13-27 (2017).
- Brown, N. K., et al. Perivascular adipose tissue in vascular function and disease: a review of current research and animal models. Arteriosclerosis, Thrombosis, and Vascular Biology. 34, (8), 1621-1630 (2014).
- Britton, K. A., et al. Prevalence, distribution, and risk factor correlates of high thoracic periaortic fat in the Framingham Heart Study. Journal of the American Heart Association. 1, (6), 004200 (2012).
- Police, S. B., Thatcher, S. E., Charnigo, R., Daugherty, A., Cassis, L. A. Obesity promotes inflammation in periaortic adipose tissue and angiotensin II-induced abdominal aortic aneurysm formation. Arteriosclerosis, Thrombosis, and Vascular Biology. 29, (10), 1458-1464 (2009).
- Farmer, S. R. Transcriptional control of adipocyte formation. Cell Metabolism. 4, (4), 263-273 (2006).
- Aune, U. L., Ruiz, L., Kajimura, S. Isolation and differentiation of stromal vascular cells to beige/brite cells. Journal of Visualized Experiments. (73), e50191 (2013).
- Ruan, H., Zarnowski, M. J., Cushman, S. W., Lodish, H. F. Standard isolation of primary adipose cells from mouse epididymal fat pads induces inflammatory mediators and down-regulates adipocyte genes. Journal of Biological Chemistry. 278, (48), 47585-47593 (2003).
- Cinti, S. Between brown and white: novel aspects of adipocyte differentiation. Annals of Medicine. 43, (2), 104-115 (2011).
- Van Harmelen, V., Rohrig, K., Hauner, H. Comparison of proliferation and differentiation capacity of human adipocyte precursor cells from the omental and subcutaneous adipose tissue depot of obese subjects. Metabolism. 53, (5), 632-637 (2004).
- Rodeheffer, M. S., Birsoy, K., Friedman, J. M. Identification of white adipocyte progenitor cells in vivo. Cell. 135, (2), 240-249 (2008).
- Church, C. D., Berry, R., Rodeheffer, M. S. Isolation and study of adipocyte precursors. Methods in Enzymology. 537, 31-46 (2014).
- Chen, Y., et al. Isolation and Differentiation of Adipose-Derived Stem Cells from Porcine Subcutaneous Adipose Tissues. Journal of Visualized Experiments. (109), e53886 (2016).
- Ye, M., et al. Developmental and functional characteristics of the thoracic aorta perivascular adipocyte. Cellular and Molecular Life Sciences. 76, (4), 777-789 (2019).
- Medeiros, D. M. The evolution of the neural crest: new perspectives from lamprey and invertebrate neural crest-like cells. Wiley Interdisciplinary Reviews. Developmental Biology. 2, (1), 1-15 (2013).
- Fu, M., et al. Neural Crest Cells Differentiate Into Brown Adipocytes and Contribute to Periaortic Arch Adipose Tissue Formation. Arteriosclerosis, Thrombosis, and Vascular Biology. (2019).
- Danielian, P. S., Muccino, D., Rowitch, D. H., Michael, S. K., McMahon, A. P. Modification of gene activity in mouse embryos in utero by a tamoxifen-inducible form of Cre recombinase. Current Biology. 8, (24), 1323-1326 (1998).
- Tamura, Y., et al. Neural crest-derived stem cells migrate and differentiate into cardiomyocytes after myocardial infarction. Arteriosclerosis, Thrombosis, and Vascular Biology. 31, (3), 582-589 (2011).
- Madisen, L., et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nature Neuroscience. 13, (1), 133-140 (2010).
- Tan, P., Pepin, É, Lavoie, J. L. Mouse Adipose Tissue Collection and Processing for RNA Analysis. Journal of Visualized Experiments. (131), e57026 (2018).
- Basu, S., Campbell, H. M., Dittel, B. N., Ray, A. Purification of specific cell population by fluorescence activated cell sorting (FACS). Journal of Visualized Experiments. (41), e1546 (2010).
- Gupta, R. K., et al. Zfp423 expression identifies committed preadipocytes and localizes to adipose endothelial and perivascular cells. Cell Metabolism. 15, (2), 230-239 (2012).
- Sowa, Y., et al. Adipose stromal cells contain phenotypically distinct adipogenic progenitors derived from neural crest. PLoS One. 8, (12), 84206 (2013).
- Billo, N., et al. The generation of adipocytes by the neural crest. Development. 134, (12), 2283-2292 (2007).
- Thelen, K., Ayala-Lopez, N., Watts, S. W., Contreras, G. A. Expansion and Adipogenesis Induction of Adipocyte Progenitors from Perivascular Adipose Tissue Isolated by Magnetic Activated Cell Sorting. Journal of Visualized Experiments. (124), e55818 (2017).