We present the techniques required to isolate the stromal vascular fraction (SVF) from mouse inguinal (subcutaneous) and perigonadal (visceral) adipose tissue depots to assess their gene expression and collagenolytic activity. This method includes the enrichment of Sca1high adipose-derived stem cells (ASCs) using immunomagnetic cell separation.
The isolation of adipose-derived stem cells (ASCs) is an important method in the field of adipose tissue biology, adipogenesis, and extracellular matrix (ECM) remodeling. In vivo, ECM-rich environment consisting of fibrillar collagens provides a structural support to adipose tissues during the progression and regression of obesity. Physiological ECM remodeling mediated by matrix metalloproteinases (MMPs) plays a major role in regulating adipose tissue size and function1,2. The loss of physiological collagenolytic ECM remodeling may lead to excessive collagen accumulation (tissue fibrosis), macrophage infiltration, and ultimately, a loss of metabolic homeostasis including insulin resistance3,4. When a phenotypic change of the adipose tissue is observed in gene-targeted mouse models, isolating primary ASCs from fat depots for in vitro studies is an effective approach to define the role of the specific gene in regulating the function of ASCs. In the following, we define an immunomagnetic separation of Sca1high ASCs.
Stem cell antigen 1 (Sca1, or Ly6A/E) was first identified as a cell surface marker expressed by hematopoietic and mesenchymal stem cells5,6. The stromal vascular fraction (SVF) of adipose tissue obtained from mouse fat depots is a heterogeneous population of cells comprising of fibroblasts, macrophages, vascular endothelial cells, neuronal cells, and adipocyte progenitor cells7. Adipocyte progenitor cells, or adipose-derived stem cells (ASCs) are non-lipid-laden cells that reside in the collagen-rich perivascular extracellular matrix (ECM)8. Approximately 50% of the SVF consist of ASCs, which are characterized as lineage-negative (Lin–) and CD29+: CD34+: Sca1+ 9. Most of these cells are Sca1+: CD24– adipocyte progenitors, which are capable of adipocyte differentiation in vitro; however, only a fraction of cells (0.08% of SVF) constitutes Sca1+: CD24+ cells that are fully capable of proliferating and differentiating into adipocytes in the in vivo conditions9. Despite the potential caveat of using Sca1+ SVF without discriminating CD24+ cells from CD24– cells, isolating Sca1+ ASCs from fat depots using immunomagnetic cell separation is an efficient and practical approach to determine the cell-autonomous phenotype of primary adipocyte progenitor cells.
In the field of obesity and diabetes, tissue fibrosis and inflammation play a critical role in the development and maintenance of type-2 diabetes3. Recently, Tokunaga et al. showed that Sca1high cells isolated from inguinal (or subcutaneous, SQ) and perigonadal (or visceral, VIS) C57BL6/J fat depots exhibit different gene signatures and ECM remodeling in vitro10. MMP14 (MT1-MMP), a prototypical member of the membrane-type matrix metalloproteinase (MMP) family mediates the development of white adipose tissue (WAT) through its collagenolytic activity1.
Examples of experiments that may be conducted with the cells isolated and enriched through the following protocol include three-dimensional culture, differentiation studies, collagen degradation assays, and RNA sequencing10,11. Degradation assays should be conducted with acid-extracted collagen to ensure the preservation of telopeptide11,12. The following protocol will demonstrate the methods to isolate primary vascular stromal cells from different fat depots and enrich adipocyte progenitor cells using immunomagnetic cell separation. The validity of the cell sorting will be assessed with flow cytometry and through using Sca1-GFP mice that express GFP in Sca1+ cells, driven by a Sca1 promoter13.
Ethics Statement: The University of Michigan Committee on Use and Care of Animals (UCUCA) has approved all methods and protocols in accordance with the Guide for the Care and Use of Laboratory Animals (Institute for Laboratory Animal Research, National Research Council). Mice are maintained in a University of Michigan vivarium and are given free access to food and water and kept on a 12 hr dark/light cycle.
1. Preparations
2. Isolation of Subcutaneous (SQ) Fat Pads
3. Isolation of Visceral (VIS) Fat Pads
4. Collagenase Digestion of Fat Pads
5. Magnetic Cell Separation
6. Verification of Immunomagnetic Separation of Sca1high ACSs with Flow Cytometry
Enrichment of Sca1high ASCs from Different Fat Pads.
The vascular stromal cells isolated from SQ fat display fibroblast-like, stretched cell shape regardless of Sca1 expression level (Figure 1A). On the other hand, VIS (eWAT-derived) Sca1high and Sca1low cells demonstrate distinct difference in their cell shape. Like SQ (iWAT-derived) Sca1high cells, VIS (eWAT-derived) Sca1high cells display stretched, fibroblast-like cell shape, whereas VIS Sca1low cells demonstrate epithelioid shape. Sca1high cells isolated from Sca1-GFP mice are easily identified as GFP-positive cells in tissue culture (Figure 1B). When these cells were assessed with flow cytometry, most of the GFP positive cells were confirmed to express Sca1 proteins on the cell surface as detected with Anti-Sca1 antibody (Figure 1C). Sca1high cells derived from inguinal fat pad maintain increased capacity of adipocyte differentiation whereas eWAT-derived Sca1high cells are more difficult to differentiate into adipocyte with conventional adipogenic mix10.
Fat Depot-dependent Gene Expression of Sca1highASCs (Figure 2).
Genome-wide transcriptome analyses with RNA sequencing demonstrated the enrichment of genes related to extracellular matrix proteins and modifiers (GO:0031012, GO:0005578) in those Sca1high ASCs10. Coupled with real-time PCR analyses, we were able to demonstrate the differential expression of the collagenolytic MMPs (MMP2, MMP8, MMP13, MMP14) between iWAT- and eWAT-derived Sca1high ASCs. When fluorescein-labeled type I collagen gels were used to assess pericellular degradation activity, we observed the markedly increased collagen remodeling activity mediated by VIS Sca1high ASCs10.
Figure 1: Immunomagnetic Separation of Murine Sca1high ASCs from Different Fat Depots. (A) Sca1high and Sca1low ASCs isolated from SQ (iWAT) and VIS (eWAT). Scale = 100 µm. (B) Sca1-GFP cells isolated from iWAT of Sca-GFP mice. Scale = 100 µm. (C) Cell surface expression of Sca1 in Sca1-GFP-positive cells assessed with flow cytometry. (left) control rat IgG (right) anti-Sca1 antibody. X-axis, GFP intensity. Y-axis, Alexa-Fluor 647. Panel A shown previously in Tokunaga, M. et al., (2014). Please click here to view a larger version of this figure.
Figure 2: Fat Depot-Dependent Expression of Collagenolytic MMPs, TIMPs, and Collagens. (A) Differential gene expression of ECMs and ECM modifiers in Sca1high ASCs isolated from different fat depots. (B) Increased collagen degradation activity of VIS (eWAT-derived) Sca1high ASCs. Degradation of collagen is shown as the disappearance of fluorescent signals (arrows and arrowheads). Inset is an enlarged image of focused collagen degradation mediated by individual cell of SQ ASCs. Cells were cultured for 72 hr. Data shown previously in Tokunaga, M. et al., (2014). Please click here to view a larger version of this figure.
Herein we demonstrate the isolation and immunomagnetic cell separation of murine ASCs from different fat pads and their use for in vitro experiments. The presented method is effective for the quick isolation of large number of Sca1-positive ASCs, which is advantageous over the technically complex and expensive FACS-mediated isolation of ASCs9,14. Unlike FACS, immunomagnetic cell separation does not allow the use of multiple antigen for the identification of a target cell population. Nonetheless, if the surface antigen is well-characterized, the use of immunomagnetic separation increases the number of cells to be analyzed without relying on the use of FACS equipment15, which is still not readily accessible to many biological researchers at small institutes without core facilities.
There are some critical aspects of the procedure that must be observed to ensure a successful outcome. The procurement of adipose-derived stem cells requires the surgical isolation of mouse adipose tissue depots. Therefore, the speed and accuracy of tissue retrieval is imperative to yield a high number of viable cells. Additionally, the maintenance of clean surgical fields and instruments are vital to the outcome of the procedure by preventing microbial contamination of cell and tissue samples. Dissected adipose tissues must be enzymatically dissociated in a type II collagenase solution. ECM composition differs between SQ and VIS fat pads10,16, where VIS contain less collagens than SQ. Therefore, the duration of VIS fat pad digestion requires approximately half the amount of time as SQ. It is acceptable to stop tissue digestion after the stated incubation period even if small particles of tissue are still present in the collagenase solution. These pieces may be mechanically dissociated with pipetting once culture media has been added to inhibit collagenase activity. While it is possible to proceed directly to immunomagnetic sorting following SVF isolation, our group seeds 1 x 106 unsorted cells on plastic plates for about 4 to 6 hr before continuing with cell sorting. Despite multiple filtration steps during SVF isolation, debris will still be present within the cell suspension. Plating the cells before proceeding to immunomagnetic cell separation provides an opportunity to wash debris and unattached cells away, thus allowing magnetic sorting to only be applied to adherent cells that contain adipogenic progenitor cells.
While Sca1 is not found in human genome, the identification and validation of alternative cell surface antigens expressed on fat depot-dependent human adipose stem cells17, when coupled with this cell separation technique, may help us define the biology of human adipose stem cells.
The authors have nothing to disclose.
This work is supported by NIH DK095137 (to THC). We thank the current and former lab members who contributed to the development and sophistication of the described methods.
Type 3 Collagenase | Worthington Biochemical | LS004182 | Tissue digestion |
DMEM | Gibco | 11965-092 | High-glucose culture medium |
Pen/Strep/Glutamine (100x) | Gibco | 10378-016 | Media antibiotic |
Anti-anti (100x) | Gibco | 15240-062 | Media antifungal |
FBS | Gibco | 16000-044 | |
PBS (1x, pH 7.4) | Gibco | 10010-023 | |
Trypsin (0.05%) | Gibco | 25300-054 | |
Cell strainer | BD Bioscience | 352360 | 100-μm cell strainer |
60mm plates | BD Falcon | 353004 | |
Scissors | FST | 14001-12 | Large |
Scissors | FST | 14091-11 | Fine, curved tip |
Large Forceps | FST | 11000-12 | |
Fine Forceps | Any vendor | ||
25G 5/8” needles | BD | 305122 | |
22G 1.5” needles | BD | 305159 | |
15 ml conical tubes | BD Falcon | 352097 | |
50 ml conical tubes | BD Falcon | 352098 | |
MACS separation columns | Miltenyi Biotec | 130-042-201 | |
Anti-Sca1 microbead kit (FITC) | Miltenyi Biotec | 130-092-529 | FITC-anti-Sca1 1ºAb and anti-FITC microbeads 2ºAb |
AutoMACS running buffer | Miltenyi Biotec | 130-091-221 | |
MiniMACS separator | Miltenyi Biotec | 130-042-102 | |
MACS MultiStand | Miltenyi Biotec | 130-042-303 | |
Blue chux pads | Fisher | 276-12424 | |
Absorbent pads | Fisher | 19-165-621 | |
Styrofoam board | Use from 50ml tubes | ||
70% ethanol | |||
Isoflurane | Any vendor | ||
Rat IgG2a Alexa Fluor 647 | Invitrogen | R2a21 | |
Rat IgG2a anti-mouse Sca1 Alexa Fluor 647 | Invitrogen | MSCA21 | |
Rat IgG2a R-PE | Invitrogen | R2a04 | |
Rat IgG2a anti-mouse F4/80 R-PE | Invitrogen | MF48004 | |
Round-bottom tube | BD Falcon | 352058 | |
HBSS (–Ca, –Mg) | Gibco | 14175-095 | |
HBSS (+Ca, +Mg) | Gibco | 14025-092 | For collagenase solution |
Type I collagen (2.7 mg/ml in 37mm acetic acid | Prepare in house12 | ||
10x MEM | Gibco | 11430-030 | |
1M HEPES | Gibco | 15630-080 | |
0.34N NaOH | Prepare in house | ||
Cover slips | Corning | 2870-22 | |
Alexa Fluor 594 carboxylic acid, succinimidyl ester, mixed isomers | Invitrogen | A-20004 | |
0.89M NaHCO3 | Gibco | 25080-094 |