Culture and Differentiation of Primary Preadipocytes from Mouse Subcutaneous White Adipose Tissue

Abstract

Adipocytes are terminally differentiated cells derived from fibroblastic preadipocyte precursors. Here, we describe a method for the isolation and proliferation of preadipocytes from murine subcutaneous white adipose tissue, followed by differentiation in culture to mature adipocytes; we refer to these cells as primary preadipocytes differentiated in vitro (PPDIVs). Compared to adipogenic cell lines, PPDIV metabolism and adipokine secretion more closely resemble in vivo adipocyte biology. While primary mature adipocytes have the greatest in vivo relevance, their fragility and buoyancy make them unsuitable for many cell culture-based methods. PPDIVs can also take advantage of transgenic and knockout mouse models to produce genetically modified adipocytes. Thus, PPDIVs are a valuable resource for studying adipocyte biology in cell culture.

1. Introduction

Adipocytes are key mediators of metabolism and metabolic health. They are unique in their ability to store large amounts of nutrients in a neutral lipid droplet that comprises the majority of the cell volume. Adipocytes are able to mobilize stored nutrients via lipolysis in response to sympathetic nervous system activation of β-adrenergic signaling. In addition to nutrient storage and mobilization, adipocytes also play an important endocrine role via the secretion of adipokines. In obesity, adipocyte function is disrupted and adipokine secretion is altered. Obesity causes adipocyte hypertrophy and hyperplasia as well as adipose tissue inflammation. In the context of obesity, adipocytes are metabolically inflexible and exhibit resistance to both insulin and catecholamine signaling. Furthermore, when the nutrient storage capacity of the adipose tissue is overwhelmed, ectopic lipid deposition occurs in other tissues such as the liver, where lipid accumulation promotes nonalcoholic fatty liver disease.

Given the critical role that adipocytes play in regulating metabolism in health and disease, it is important to investigate and explicate their biology. The adipose tissue is comprised of mature adipocytes, as well as many other cell types, including preadipocytes, immune cells, and endothelial cells. When adipose tissue is extracted and digested with collagenase, the mature adipocytes float to the top, owing to their lipid content, while the other cell types can be pelleted at the bottom and are collectively referred to as stromal vascular fraction (SVF). Techniques for collagenase digestion of adipose tissue were developed in the 1960s [1]. Primary mature adipocytes, especially from rat adipose tissue, have been used extensively to investigate adipocyte metabolism. However, mature adipocytes are prone to lysis, which releases lipids and damages neighboring cells. While mature adipocyte buoyancy facilitates their isolation, it complicates downstream utilization. Some techniques requiring adherent adipocytes can utilize ceiling culture, where the adipocytes are allowed to adhere to a surface at the top of the culture [2, 3]. However, ceiling culture is not compatible with many assays.

An alternative source of adipocytes from adipose tissue is the preadipocyte subpopulation of the SVF. The existence of preadipocytes in adipose tissue was observed in the late 1960s [4], and adipogenesis in SVF from rat and human adipose tissue was reported in the 1970s [5-7]. Meanwhile, Green et al. developed a clonal line of mouse embryonic fibroblasts, 3T3-L1s, which are prone to adipogenic differentiation [8-10]. Quickly, the 3T3-L1 cell line became one of the most popular tools for studying adipogenesis and adipocyte metabolism. Some interest remained in differentiation of SVF. These studies primarily used human or porcine adipose tissue as a source of these cells, due to their translational potential and larger volume adipose tissue as compared to murine sources. In recent decades, there has been increasing interest in the differentiation of the multipotent SVF into adipocytes, as well as other cell types [11]. At the turn of the millennium, it was observed that the SVF include multipotent cells, known as adipose-derived stem cells (ASCs), which can be induced to differentiate into cardiomyocytes, osteoblasts, and other mesenchymal cells as well as adipocytes [12-14]. The SVF also contains committed preadipocytes.

Mice are uniquely suited to genetic manipulations and the generation of whole body and tissue specific knockouts, making them a vital resource in many basic science and translational research labs [15]. PPDIVs generated from mouse adipose tissue can take advantage of these genetic models to study adipocyte biology. Whole body knockout animals can have complex phenotypes due to effects of the knockout in multiple cell types and secondary paracrine and endocrine effects. By studying PPDIVs from a knockout mouse, one can determine the cell autonomous effects of that knockout in adipocytes to deconvolute the in vivo phenotype. Additionally, PPDIVs from adipocyte specific knockout mice (as mediated by adiponectin or FABP4 promoter driven Cre expression) [16] can be very useful in studying adipocytes without systemic influences. These cells are especially useful for studying the role of a gene in mature adipocyte biology without complicating effects on adipogenesis, as the knockout occurs late in adipogenesis, after commitment. PPDIVs are more similar to in vivo adipocytes as compared to 3T3-L1 adipocytes and can be used for all the same assays involving adherent cells. Additionally, PPDIVs from mice of different ages, sexes, and diets can be investigated to determine how changes in the adipose tissue environment affect this preadipocyte population and their adipogenic potential.

Here, we present a method for the isolation of SVF from the subcutaneous inguinal fat of young mice. The following proliferation steps enrich the preadipocyte population, which can then be induced to undergo adipogenic differentiation. These PPDIVs differentiate well, allowing for the investigation of mature adipocyte metabolism without complications resulting from incomplete or unequal adipogenesis.

2. Materials

2.1. Stromal Vascular Cell Isolation

1. Inguinal white adipose tissue from four to six mice aged 6 to 12 weeks (see Note 1).

2. Tools: scissors and curved forceps for dissection of tissue and razor blades, glass plate and petri dish for mincing the tissue.

3. 150 mL of 70% ethanol in a 250 mL beaker.

4. 70% ethanol in spray bottle.

5. Cell strainer, 100 μm.

6. Wash buffer: DMEM with 2% BSA, filter sterilized (see Note 2).

7. 2× digestion buffer: Wash buffer with 2 mg/mL Type II collagenase (collagenase from Clostridium histolyticum Sigma-Aldrich cat# 6885), filter sterilized (see Note 3).

8. Sterile 50 mL conical tubes.

2.2. Plating and Proliferation

1. Culture media: 15% fetal bovine serum (FBS) and 1× penicillin-streptomycin-glutamine in DMEM/F12.

2. Amphotericin B: 5000× stock = 10 mg/mL in DMSO, working concentration = 2 μg/mL.

3. 10 mL serological pipets.

4. Dulbecco’s phosphate-buffered saline (DPBS) no calcium, no magnesium.

5. Tissue culture treated plates: 10 cm and 15 cm.

6. Trypsin-EDTA 0.05%, phenol red.

2.3. Adipogenic Differentiation

1. Culture media (same as above).

2. 1000× dexamethasone stock: 5 mM in ethanol.

3. 1000× 3-isobutyl-1-methylxanthine (IBMX) stock: 0.5 M in DMSO.

4. 1000× insulin stock: 1 mg/mL in acidified water.

5. 10,000× TZD (rosiglitazone or troglitazone) stock: 10 mM in DMSO.

6. Differentiation media I: culture media supplemented with 5 μM dexamethasone, 0.5 mM IBMX, 1 μg/mL insulin, and 1 μM thiazolidinedione (TZD) (see Note 4).

7. Differentiation media II: culture media supplemented with 1 μg/mL insulin.

8. Final assay plate, e.g., 6- to 96-well cell culture plates (depending on downstream application).

2.4. Equipment

1. Pipets and pipet aid.

2. Isoflurane administration system.

3. Shaking water bath (37 °C).

4. Centrifuge (fits 50 mL conical tubes and spins at 440 × g).

5. CO₂ incubator at 10% CO₂.

6. Microscope for viewing and counting cells.

7. Hemocytometer.

8. Tissue-culture hood.

3. Methods

3.1. SVF Isolation

Before starting the isolation, ensure that the shaking water bath is at 37 °C and filled with clean water. Culture media should be warmed to room temperature for use.

3.1.1. Preparation for SVF Isolation

1. Prepare the culture media, and wash buffer (see Note 2) and 2× digestion buffer (5 mL for each isolation, plus extra to account for losses during filtration) (see Note 2).

2. Place 5 mL of wash buffer into a 50 mL conical tube (one for each isolation from four to six mice), and place on ice alongside digestion buffer.

3. Prepare for tissue collection by setting up the following in a clean, low traffic area: sacrifice tools, isoflurane chamber, beaker with 70% ethanol, clean sacrifice surface such as WyPall L40 or sacrifice tray, and razor blade and petri dish for each isolation on ice. A biological safety cabinet, or snorkel, can be used but is not strictly necessary.

3.1.2. Mouse Sacrifice

1. Sacrifice the mice by cervical dislocation following anesthetization with isoflurane.

2. Submerge the whole mouse in 70% ethanol, holding by the tail.

3. Place the mouse supine on sacrifice area, and make a horizontal cut in the midline of the abdominal skin inferior to the ribcage. Holding each side of the cut, tear the skin apart by pulling away toward the head and tail, respectively. Carefully fold the lower abdominal skin over gloved fingers to access the inguinal fat, avoiding contamination from fur. Holding the skin taught, peel off the inguinal fat pad with forceps starting at the bottom, near the testis on a male. Place in petri dish on ice. Repeat on the other side, collecting both fat pads from each mouse (see Note 5).

3.1.3. Tissue Digest

1. Move the adipose tissue from the petri dish to the glass plate on the bench top. Use the razor blade to finely mince the tissue by chopping, then rotating 90°, and chopping again. Repeat until a smooth slurry is obtained. If the tissue is very sticky during the mincing (more like gum than liquid slime), add 500 μL of the 5 mL of aliquoted wash buffer to the tissue and continue mincing, repeat if necessary.

2. Place the minced tissue into the wash buffer and swirl (see Note 6).

3. Add 5 mL of 2× digestion buffer to the minced tissue submerged in wash buffer, and place immediately into the 37 °C shaking water bath. The shaking speed should be sufficient to keep the tissue mixed in the solution and not floating on the top. (Proceed at this time to preparations in Subheading 3.1.4, step 2, while digestion is in progress.)

4. Check the digestion after 15 min. The tissue chunks should be fully digested and no visible chunks of adipose tissue remaining, unless there were larger chunks of tissue that were not properly minced (see Note 6). Longer digestion periods may be needed. Check on the digestion progress every 5 min. Digestion time should be less than 30 min. Do not exceed 40 min of digestion.

3.1.4. Washing of SVF Cells

1. This step is done in a sterile cell culture hood. Make sure to sterilize exterior of the tubes by wiping down with 70% ethanol when transferring into the hood after spins and especially after digestion in the shaking water bath.

2. While cells are digesting, prepare the following in the tissue-culture hood: a 50 mL conical tube with 100 μm cell strainer. If performing the optional wash, also prepare a 50 mL conical tube with 30 mL wash buffer (see Note 7). If doing multiple isolations, for example, from mice of different genotypes, prepare one set for each isolation. Be sure to label all tubes.

3. Once digestion is complete, immediately inactivate the collagenase by adding 35 mL of culture media.

4. Filter cell slurry through 100 μm cell strainer into the 50 mL conical tube (see Note 8). Spin at 440 × g for 5 min to separate floating adipocytes and SVF (see Note 9). (Optional: collect floating adipocytes using a 10 mL serological pipette, and transfer to the 50 mL tube containing 30 mL wash buffer). Suction off all but 3 mL of the supernatant being careful not to disrupt the cell pellet. Add 10 mL of culture media to the pellet, and resuspend.

5. (Optional: spin the floating adipocytes from step 4 at 440 × g for 5 min (see Note 7). Suction off all but 3 mL of the supernatant, being sure to remove all floating cells and not disrupt the cell pellet. Add 10 mL of culture media to the pellet, and resuspend and combine with the first pellet). Spin the cell pellet(s) at 440 × g for 5 min, and then suction off all but 2 mL of the supernatant being careful not to disrupt the cell pellet (see Note 10).

3.2. Plating and Proliferation

These steps are carried out in a tissue-culture hood using sterile technique. The incubator should be set to 10% CO₂.

1. Add 10 mL culture media plus amphotericin B (1:5000) to the cell pellet from Subheading 3.1.4, step 5, and resuspend using a 10 mL serological pipette and ejecting onto the sidewall of the conical tube at least ten times to break up clumps (see Note 11). Plate cells on a 10 cm dish (see Note 12). Leave the cells in the incubator for 48–72 h (see Note 13).

2. Two to three days after isolation, wash the cells gently with DPBS three times. Replace media with fresh culture media. The media need to be replenished every 2–3 days during proliferation.

3. Once the cells reach 70%–80% confluence (see Note 14), split the 10 cm plate into a 15 cm plate: Wash cells twice with DPBS. Add 1 mL of 0.05% Trypsin-EDTA to the plate, and tilt plate to coat evenly. Place plate in incubator for 3–5 min. Check dissociation using an inverted microscope (see Note 15). Resuspend cells in 10 mL of DPBS and transfer the cells to a 50 mL conical tube containing 20 mL culture media. Spin cells down at 440 × g for 5 min. Suction off all but 2 mL of the supernatant and resuspend the cell pellet in 25 mL culture media and transfer to a labeled 15 cm plate. Move the 15 cm plate back and forth and side to side to evenly distribute the cells.

4. Once the cells reach 70%–80% confluence in the 15 cm plate, split cells, and plate for differentiation (see Note 16): Wash cells twice with DPBS. Add 2.5 mL of 0.05% Trypsin-EDTA to the plate, and tilt plate to coat evenly. Place plate in incubator for 3–5 min. Check dissociation using an inverted microscope. Resuspend cells in 10 mL of DPBS and transfer the cells to a 50 mL conical tube containing 20 mL culture media. Place 10 μL of the cell suspension onto a hemocytometer, and spin the remaining cells down at 440 × g for 5 min while counting cells. Resuspend cells at 5 × 10⁵⁴ cells/mL in culture media (see Note 17).

5. Plate 2 mL per well in a 12-well plate, or equivalent confluence, i.e., 4 mL per well in a 6-well plate, 1 mL per well in a 24-well plate (see Notes 18-20).

3.3. Differentiation (see Note 21)

These steps are carried out in a tissue-culture hood using sterile technique. The incubator should be set to 10% CO₂.

1. Make differentiation media I and II.

2. Check cells (plated in step 5 of Subheading 3.2) daily. Once the cells reach 100% confluence, this is Day −2 of differentiation. The cells should reach confluence within 3 days.

3. On Day 0 (48 h after the cells reach 100% confluence), change the media to differentiation media I (see Note 22). Provide double the usual media volume during and after differentiation, i.e., 2 mL per well in a 12-well plate.

4. On Day 3, change media to differentiation media II (see Note 22).

5. Differentiation media II should be changed every 2–3 days. Cells can be used for experiments on Day 7 to Day 14 of differentiation.

6. Change cells to culture media without insulin 24 h prior to use. Note that adherence in FBS is particularly fickle.

4. Notes

1. Preadipocytes from younger animals have greater differentiation potential, while older animals have larger adipose depots containing more preadipocytes [17-19]. The ideal age for preadipocyte isolation is 6–12 weeks. Cells can be isolated from older animals; however, differentiation rates will likely be lower.

2. Wash buffer can be made from any culture media, e.g., DMEM or DMEM/F12. BSA for wash buffer does not need to be fatty acid-free. Be patient dissolving the BSA. Add BSA to media and use a magnetic stir bar to mix gently. Do not shake as this will cause bubbles to form, which can make the BSA take longer to dissolve. The wash buffer can be made in advance, but digestion buffer should be made immediately before isolation.

3. It is important to filter sterilize the collagenase solution. Use of a filter with capacity above the volume being filtered is recommended. This solution can easily clog filters.

4. Addition of a TZD such as rosiglitazone or troglitazone maximizes differentiation efficiency and promotes a beige phenotype in the adipocytes [20, 21]. If beiging is not desired, the cells will differentiate without added TZD, albeit at a lower efficiency.

5. Here, we use the term inguinal to refer to the entire posterior subcutaneous depot, consisting of the contiguous dorsolumbar, inguinal, and gluteal depots [22]. Removal of the lymph node from the fat pad is optional. This method can be used to isolate SVF from any fat pad. Preadipocytes isolated from the inguinal depot have the greatest differentiation potential and are thus recommended [17, 23-26]. Additionally, subcutaneous fat contains fewer immune cells than visceral depots, such as gonadal fat.

6. Take note if large chunks of tissue are present, adjustment to the mincing technique will need to be made in the future (Fig. 1). Thorough mincing is ideal; however, time off of the ice and out of solution must be minimized; 5–10 min of vigorous mincing should be sufficient. Note that the mincing process can be physically taxing. It is recommended to have a second individual to mince the tissue, while the first individual continues to harvest. This speeds up the process, especially when multiple genotypes are being utilized.

7. Prepare one 50 mL conical tube for each wash of the floating cells in Subheading 3.1.4, step 5. This wash increases the yield of SVF and can be repeated, if deemed necessary. Up to three washes may be performed.

8. The cell strainer can easily become clogged. Do not overload it. Use the pipettor to mix the floating faction with the media before placing onto the cell strainer. If the strainer becomes clogged, add a few milliliters of media to the strainer and pipet up and down to mix. With mixing and dilution, most of the floating faction will pass through the strainer.

9. The floating cells can be collected for RNA or protein isolation after the first spin. However, if it is desired to utilize the floating mature adipocytes for a live cell experiment, do not spin the samples as this will promote lysis and damage the adipocytes. The cells can either be allowed to settle by gravity or manually spun at very low speeds. After the floating cells are collected, SVF from the remaining solution (combined with any eluate from washing the adipocytes) can be pelleted; however, yields will be minimal.

10. This step washes the cells and can optionally be repeated once or twice more; however, each spin increases clumping of the cells.

11. It is normal for some clumps to remain in the solution; cells in these clumps may still adhere to the plate so it is recommended to plate everything including the clumps. Some downstream applications for the SVF, such as fluorescence activated cell sorting, require that the cells be filtered through a 40 μM cell strainer. This is not recommended for PPDIV production, due to cell loss.

12. SVF from four to six mice should result in an appropriate cell density for a 10 cm plate. Cells plated too sparsely will proliferate slowly, while cells plated at too high a confluence will begin to differentiate prematurely.

13. During this time, there are immune cells and non-adherent cells including red blood cells that make it very difficult to determine the yield of your isolation. It is best to leave the cells for a couple days without washing. Red blood cell lysis is not necessary and not recommended.

14. Confluence is defined as coverage of the surface of the plate (Fig. 2), not the percentage of cells touching one another. The cells should reach 70% confluence 4–7 days after plating.

15. Note that some cells such as immune cells will not dissociate from the plate. This is favorable, as they will not end up in the final culture. Over 90% of the fibroblastic cells should dissociate. Cells that have rounded up, but are still adhered, usually come off the plate when rinsed with DPBS. Do not exceed 5 min in 0.05% Trypsin. If many cells remain, repeat Trypsin treatment and collect cells in another 5–10 mL of DPBS.

16. One or two more 1:3 splits can be done to increase yield; however, differentiation potential can be reduced by additional passaging. If additional cells are needed, it is recommended to plate two-thirds of the cells for differentiation and split the remaining third into a 15 cm plate, which can be used for a second round of differentiation when it reaches 70% confluence in 3 to 5 days.

17. A concentration of up to 3 × 10⁵ cells/mL can be used to plate the cells at close to 100% confluence. If two isolations with different proliferation rates are being compared, this is recommended. Otherwise, the lower plating density is recommended to maximize yield.

18. If comparing two different isolations, e.g., WT and KO, count the cells a second time after resuspension. It is critical that the plating density be equal for differentiation to be comparable. The plated cell concentration can be normalized by either spinning down and resuspending again or dilution and adjustment of the plating volume.

19. Smaller wells tend to differentiate better. Wells in a 12-well plate are sufficient for RNA or protein isolation for gene expression analysis and Western blotting, respectively. Lipolysis assays and quantification of other secreted factors can be done in 24- or 48-well plates.

20. While not necessary, collagen coating can help with cell adherence and may affect adipogenesis. However, collagen coating can interfere with protein measurement and normalization and may be incompatible with certain downstream applications.

21. Appearance of cells during differentiation (Fig. 3): From Day −2 to Day 0, the cells should change from fibroblasts with filipodia to compact cells with a honeycomb-like appearance. In differentiation media I, the cells will initially contract, and spaces between the cells may be visible, which is normal. Tiny lipid droplets should be visible on the last day of differentiation media I. During the insulin treatment in differentiation media II, lipid droplets should grow, and some cells will develop lipid droplets. Cells that do not exhibit lipid droplets by the third day of insulin will not differentiate. Longer incubation in differentiation media II will increase the lipid droplet size, and some adipocytes may become unilocular; however, adherence to the plate becomes an issue, and cells will come off in sheets more easily the longer differentiation is continued. Detached cells will clump together into what looks like a small chunk of adipose tissue, which can be transplanted to generate a viable adipose depot in vivo [27, 28].

22. The differentiation media I may be viscous. Remove it from the plate gently. For smaller plates, removing the media by inversion and blotting on a sterile paper towel is preferable to suction. If small lipid droplets are not visible in many of the cells after 3 days of differentiation media I, leave the cells in differentiation media I for a fourth day.

Fig. 1.

Mincing and digestion of adipose tissue. (a) Appearance of adipose tissue after complete and incomplete mincing. (b) Appearance of adipose tissue mixed with wash buffer after complete and incomplete mincing. (c) After digestion, completely minced tissue will no longer be visible, while incompletely minced tissue chunks will be visible when the mixture is swirled. (d) After filtration, incompletely minced tissue chunks will remain in the filter

Fig. 2.

Confluency of preadipocytes. Images of proliferating preadipocytes at (a) 95%, (b) 60%, (c) 15%, and (d) 5% confluency taken using a 10× objective. Scale bar is 100 μm

Fig. 3.

Lipid droplet development with and without rosiglitazone. Days 3, 6, and 14 of differentiation in the presence and absence of rosiglitazone. Nuclei stained with Hoechst shown in blue and Bodipy stained lipid droplets shown in green. Scale bar is 10 μm