The authors have no conflict of interest.
A surface-tethered spheroid model for functional evaluation of 3T3-L1 adipocytes
Article first published online: 30 SEP 2013
© 2013 Wiley Periodicals, Inc.
Biotechnology and Bioengineering
Volume 111, Issue 1, pages 174–183, January 2014
How to Cite
Turner, P. A., Harris, L. M., Purser, C. A., Baker, R. C. and Janorkar, A. V. (2014), A surface-tethered spheroid model for functional evaluation of 3T3-L1 adipocytes. Biotechnol. Bioeng., 111: 174–183. doi: 10.1002/bit.25099
Paul A. Turner and Lacey M. Harris contributed equally to this work.
- Issue published online: 22 NOV 2013
- Article first published online: 30 SEP 2013
- Accepted manuscript online: 23 AUG 2013 07:40AM EST
- Manuscript Accepted: 13 AUG 2013
- Manuscript Revised: 7 AUG 2013
- Manuscript Received: 19 JUN 2013
- School of Dentistry and the University of Mississippi Medical Center
- 3D culture model;
In order to effectively treat obesity, it must be better understood at the cellular level with respect to metabolic state and environmental stress. However, current two-dimensional (2D) in vitro cell culture methods do not represent the in vivo adipose tissue appropriately due to the absence of complex architecture and cellular signaling. Conversely, 3D in vitro cultures have been reported to have optimal results mimicking the adipose tissue in vivo. The main aim of this study was to examine the efficacy of a novel conjugate of a genetically engineered polymer, elastin-like polypeptide (ELP) and a synthetic polymer, polyethyleneimine (PEI), toward creating a 3D preadipocyte culture system. We then used this 3D culture model to study the preadipocyte differentiation and adipocyte maintenance processes when subjected to various dosages of nutritionally relevant free fatty acids with respect to total DNA and protein content, cell viability, and intracellular triglyceride accumulation. Our results showed that 3T3-L1 preadipocytes cultured on the ELP-PEI surface formed 3D spheroids within 72 h, whereas the cells cultured on unmodified tissue culture polystyrene surfaces remained in monolayer configuration. Significant statistical differences were discovered between the 3D spheroid and 2D monolayer culture with respect to the DNA and protein content, fatty acid consumption, and triglyceride accumulation, indicating differences in cellular response. Results indicated that the 3D culture may be a more sensitive modeling technique for in vitro adipocyte culture and provides a platform for future evaluation of 3D in vitro adipocyte function. Biotechnol. Bioeng. 2014;111: 174–183. © 2013 Wiley Periodicals, Inc.
In the past few decades, the prevalence of overweight and obese individuals has increased steadily within the United States, affecting men and women of all age groups and all racial and ethnic groups (Flegal et al., 2010; Moldket et al., 1999; Wang et al., 2008). Obesity is a contributing factor to many systemic and metabolic diseases including type-2 diabetes, hypertension, and coronary heart disease (Pi-Sunyer, 2002). Effective treatments for this disease must be developed that address pathologies at a cellular level with respect to metabolic state and environmental stress (Kang et al., 2007). In vivo animal models used for this purpose are often costly and have many confounding factors, while current two-dimensional (2D) monolayer in vitro models do not represent the adipose tissue appropriately due to the absence of complex architecture and extracellular signaling. Studies have shown that 3D spheroid in vitro cultures effectively mimic the in vivo adipose tissue structure, function, and gene expression to help bridge the gap between in vivo animal models and in vitro cell cultures (Kang et al., 2007; Wang et al., 2009; Yamada and Cukierman, 2007). A variety of materials, including polycaprolactone (PCL), chitosan, polyester, hyaluronic acid, collagen, polyethylene glycol, and chemically modified alginate, in the form of porous scaffolds or hydrogels have been used in 3D modeling techniques to culture preadipose cells (Cheng et al., 2012; Gomillion and Burg, 2006; Kang et al., 2007; Yamada and Cukierman, 2007). Unfortunately, porous materials often exceed the acceptable scaffold rigidity and preadipocytes embedded directly in hydrogels may remain functionally impaired (Gomillion and Burg, 2006).
Spheroids are cellular aggregates which allow individual cells to interact through junctional complexes (Mueller-Klieser, 1997). Previously, we have conjugated a genetically engineered polymer, elastin-like polypeptide (ELP), to a synthetic polymer, polyethyleneimine (PEI), and demonstrated that primary rat hepatocytes cultured atop ELP-PEI coated substrates formed adherent spheroids (Janorkar et al., 2008). In this ELP-PEI copolymer, the PEI component induced spheroid formation, while the ELP component allowed surface attachment of the formed spheroids. Furthermore, we have optimized reaction conditions and surface coating concentration of the ELP-PEI conjugate for spheroid formation without adversely affecting cell viability (Turner et al., 2013). Hepatic and model cell lines have been found to maintain sufficient interaction with the underlying ELP-PEI substrate allowing cell motility without causing detachment. We propose that this adhesion allowed by our ELP-PEI culture system may result in spheroids that are easier to culture and functionally superior compared to conventional spheroid-forming suspension techniques (Gottschling-Zeller et al., 1999; Weeks et al., 2013) or “ceiling” adipose culture techniques (Zhang et al., 2000). Therefore, the main aim of this study was to examine the efficacy of the ELP-PEI conjugate toward creating a 3D adipocyte spheroid culture system and evaluate markers indicative of adipogenesis. In this 3D culture, adipocyte spheroids formed through self-aggregation are expected to remain tethered to the ELP-PEI coating, thereby lending them long-term stability.
Adipocytes are complex cells with many functions in addition to storage of intracellular lipids and metabolism. Adipocytes aid in homeostasis and play role in the regulation of blood pressure, immune function, angiogenesis, and energy levels (Lau et al., 2005). Increased dietary fats alter differentiated gene expression and lead to dysregulation of these functions (Lopez et al., 2003; Todoric et al., 2006). However, in addition to the quantity of the fat present, the composition and nature of the fatty acids appear to influence human health. The physical and chemical properties of fatty acids are dependent on chain-length, position of double bonds, and number of double bonds, which ultimately affect the metabolism of a particular fatty acid (Madsen et al., 2005). For this study, two different classes of nutritionally relevant fatty acids, linoleic acid (C18:2, LA) and oleic acid (C18:1, OA), were selected. LA is an essential fatty acid and must be obtained by dietary means (Burr et al., 1930). It is a polyunsaturated ω-6 fatty acid that may be converted into arachidonic acid, suggesting that consumption of LA may influence inflammatory processes (Calder, 2006). OA is a monounsaturated ω-9 fatty acid that occurs naturally in various animal and vegetable fats and oils. It is the most abundant fatty acid in human adipose tissue (Kokatnur et al., 1979) and served as a non-inflammatory comparison in our study. Therefore, we aimed to utilize the 3D spheroid culture system created atop the ELP-PEI coating to study adipocyte behavior when subjected to various dosages of fatty acids with respect to total DNA and protein content, cell viability, fatty acid consumption, and intracellular triglyceride accumulation.
Materials and Methods
Expression, Purification, and Chemical Modification of ELP
A detailed account of the procedure has been previously published (Janorkar et al., 2008; Turner et al., 2013). The ELP, with primary sequence [VGVPG]40, where G = glycine, P = proline, and V = valine, was produced using E. coli BLR (DE3) and purified by inverse phase transition purification (repeated solubilization at 4°C and precipitation at 40°C). ELP was activated with N-hydroxysuccinimide and 1-ethyl-3-[3-dimethyl aminopropyl] carbodiimide hydrochloride (Sigma, St. Louis, MO) for 20 min and subsequently reacted with PEI for 18 h. Unreacted PEI was removed by the inverse phase transition purification. Total conjugate (ELP-PEI) concentration was measured and adjusted to 5 mol% for subsequent coating and cell culture experiments (Turner et al., 2013).
ELP-PEI was adsorbed to a 24-well tissue culture polystyrene (TCPS) plate (Costar, Corning Life Sciences, Lowell, MA) by placing 200 µL of a 5 mg/mL solution in each well. The plate was incubated at 37°C for 48 h in a dry incubator to remove all solvent.
3T3-L1 Cell Culture
The 3T3-L1 mouse preadipocyte cell line used in our study was purchased from American Type Culture Collection (ATTC, Manassas, VA) and cultured according to the provider's protocol as described below. This cell line has been used for years as a model system for studying the in vitro differentiation process of fibroblastic preadipocytes into bulbous mature adipocytes (Rosen et al., 1979). 3T3-L1 mouse preadipocytes were multiplied in high glucose (4.5 g/L) Dulbecco's modified Eagle's medium (DMEM, Invitrogen, Carlsbad, CA) supplemented with 10% calf serum and 100 U/mL penicillin–100 µg/µL streptomycin at 5% CO2 and 37°C and harvested before they reached 70% confluence. Preadipocytes were seeded at 5 × 104 cells per well into 24-well uncoated or ELP-PEI coated TCPS plates and were cultured using above conditions for 72 h. Subsequently, cells were differentiated for 72 h using DMEM supplemented with 10% fetal bovine serum (FBS), 1 µM dexamethasone, 0.5 mM IBMX, and 0.1 U/mL insulin. Control medium was prepared by supplementing DMEM with 10% FBS and 2% bovine serum albumin, BSA (Sigma). Experimental media were prepared by supplementing control medium with 0.5 and 0.7 mM of OA or LA (Sigma). The control and experimental media were sonicated for 45 min at 40°C, cooled to 4°C, sterile filtered, and added with 0.2 U/mL insulin and 7 ng/mL glucagon. After differentiation, cells were fed control or experimental medium up to 120 h. Half of the media volume was changed every 48 h.
Live/Dead and Oil Red O Staining
Cells were exposed to following media for 72 h (i.e., 216-h time point): control, 0.5 mM OA, and 0.5 mM LA and the cell viability was analyzed using a Live/Dead assay performed per manufacturer's protocol (Invitrogen). Oil Red O staining was completed using same growth conditions after 120 h of exposure (i.e., 264-h time point). Oil Red O stain (0.35 g; Sigma) was dissolved in 100 mL of isopropanol overnight and filtered using a 0.22 µm filter. Working solution was six parts of this stock solution mixed with four parts sterile deionized (DI) water. 3D cultures were first embedded in collagen (Invitrogen) for 30 min for the purpose of preserving spheroid architecture and structure prior to staining. Adipocytes were rinsed with PBS and fixed with 4% paraformaldehyde for 1 h. Cells were then rinsed once with PBS and twice with sterile DI water. Finally, cells were stained with the working solution for 2 h at room temperature and rinsed thrice with sterile DI water. 2D cultures were imaged using VHX-600 Digital Microscope (Keyence Corp, Osaka, Japan). Collagen-embedded 3D spheroids were imaged using a Nikon Eclipse 50i microscope with Niko DS Fi-1 digital camera (Nikon, Inc., Melville, NY).
Optical and Fluorescence Microscopy
Cells were observed using an Olympus IX-81 microscope (Olympus, Center Valley, PA) equipped with an environmental chamber (LiveCell, Pathology Devices, Westminster, MD), which maintained normal culture conditions (37°C temperature, 70% relative humidity, and 5% CO2). Bright field, FITC, and TRITC filters were used to capture the cellular growth, lipid storage phases, and Live/Dead assay. Slidebook software (Intelligent Imaging, Denver, CO) was used for image analysis. Spheroid dimensions were tracked by time-lapse imaging and measured with ImageJ digital analysis software.
Quantification of Total DNA, Protein, and Intracellular Triglyceride Content
The 3T3-L1 cells were rinsed with PBS and removed from the TCPS via trypsin digestion for 2 min. All aliquots were centrifuged for 2 min at 1,000 rpm, resuspended in DI water, and sonicated for 30 s at 10% amplitude using a Branson Digital Sonifier 450 (Danbury, CT). Total DNA, protein, and intracellular triglyceride content were analyzed using assays performed per manufacturers' protocols (Sigma, Thermo Scientific, and Sigma, respectively) and measured on an absorbance plate reader (Biotek ELx800, Winooski, VT) with a 540 nm filter. Measurements were taken in triplicate from each of three wells exposed to unique experimental media and culture conditions.
Cell Viability Assessment With MTT Assay
The influence of exogenous fatty acid treatments on cell health and viability were quantified using a Molecular Probes MTT assay (Eugene, OR). Cells were differentiated and exposed to the same growth conditions described previously (Control, 0.5 mM LA, 0.7 mM LA, 0.5 mM OA, and 0.7 mM OA). MTT assay was performed at 168 and 264 h (1 and 5 days after differentiation and exposure to experimental media). All medium was removed at the time of the assay and replaced with 250 µL phenol red-free control maintenance medium and 50 µL of 12 mM MTT assay. The cultures were then incubated at 37°C for 2 h to allow cells to metabolize the MTT. Excess media was removed and the reduced tetrazolium salt was resolubilized using 1 mL DMSO. The stained solution was centrifuged at 14,000 rpm for 5 min to remove any insoluble cell debris and precipitates. Absorbance measurements were taken at 540 nm using a Biotek plate reader, with nine measurements per condition. Wells containing only media and assay were read to acquire baseline measurements. The assay was performed for cultures grown in three wells under each culture condition (2D and 3D) and media treatment.
Gas Chromatography Analysis of Fatty Acid Media
Gas chromatography (GC) was used to confirm media retention of fatty acid via binding with BSA as well as subsequent loss due to cell uptake. Lipophilic elements were separated from whole media using Bligh–Dyer extraction technique (Bligh and Dyer, 1959). Aqueous media samples (1 mL) were combined with 2 mL methanol/2% acetic acid and 1 mL methylene chloride and mixed for 30 min. Heptadecanoic acid (HA) was added to each sample and used as an internal standard for quantification. Additional DI water and methylene chloride (1 mL each) were then added to the solution to produce a 1:1:1 aqueous/methanol/methylene chloride mixture which was then centrifuged to facilitate phase separation. The organic (methanol) phase was removed, dried under nitrogen, redissolved in 2% H2SO4/methanol solution, and derivatized overnight at room temperature. The resulting fatty acid methyl esters were collected by hexane extraction. 2.5 M NaOH was used to neutralize the H2SO4 and hexane was added to extract the derivatized fatty acids. Samples were dried under nitrogen, redissolved in hexane, and run on an Agilent 6890 GC System (Santa Clara, CA) using helium carrier gas. Sample volumes of 1 µL were applied via splitless injection mode into the system at 220°C. Separation was accomplished using a Restek (Bellefonte, PA) Rtx-225 column (0.25 mm internal diameter × 30 m length × 0.25 µm film thickness) at a flow rate of 0.8 mL/min. Each sample run lasted 33 min. Column temperature was initially held at 150°C for 1 min before increasing by 5°C/min (for 18 min) to 240°C, which was maintained for 12 min. Post-run temperature was returned to 150°C for 2 min. Samples were measured with a FID operating at 250°C. GC sample data was analyzed using GC Chemstation (Agilent, Santa Clara, CA). Fatty acid content was quantified by logarithmic identification and integration of peaks produced by detector response relative to the internal standard. Individual components were identified and fit to standard curves produced from a Restek Marine Oil FAME Mix (Bellefonte, PA) consisting of 20 common lipid metabolites including OA and LA.
All experiments were performed at least in triplicate. Quantitative results reported as mean ± 95% confidence intervals. Assays were measured from three separate wells for a total of nine measurements. Statistical evaluation of the results was performed with ANOVA followed by Games–Howell post hoc test for unequal variance. Values with P ≤ 0.05 were deemed significantly different. Qualitative analysis and micrographs were captured from three separate replicates.
The 2D in vitro adipocyte culture model was created by achieving a confluent monolayer of 3T3-L1 preadipocytes onto an uncoated TCPS surface, followed by a period of differentiation and maintenance with BSA-bound fatty acid containing media. Seeding 3T3-L1 preadipocytes onto the positively charged ELP-PEI coated TCPS surface led to the formation of spheroids, which followed by the same differentiation and maintenance protocol, created the 3D in vitro adipocyte culture model. On uncoated TCPS, 3T3-L1 cells spread out to confluence on the entire surface and displayed traditional monolayer characteristics over the first 72 h of culture period (Fig. 1a and b). Conversely, 3T3-L1 cells cultured on the ELP-PEI coated surface formed cellular aggregates within the first 24 h and formed spheroids by the 72-h time point (Fig. 1e and f). All cultures were fed differentiation media starting at the 72-h time point. The cells in both 2D and 3D configurations responded to the differentiation media by increasing in size up to the end of the differentiation period at the 144-h time point (Fig. 1c and g).
The cells were then fed maintenance media containing either no additional fatty acids aside from those present in FBS (control), OA (0.5 and 0.7 mM), or LA (0.5 and 0.7 mM) for the following 5 days. For all 3T3-L1 cell cultures, the optical micrographs showed an accumulation of lipid droplets by the 264-h culture period (Figs. 1d and h and 2). The spheroid diameters measured using micrographs (Fig. 3) revealed that spheroids continued to increase in size for up to 72 h, where the growth appeared to plateau. Addition of differentiation cocktail (72–144 h) prompted further spheroid growth to achieve an average spheroid diameter of 80 μm by the 144-h time point. Spheroid diameters continued to increase upon introduction of the control and experimental media (144–264 h), reaching an ultimate diameter of approximately 120 µm. All fatty acid dosages, in addition to the control, appeared to result in fairly similar spheroid diameters, with diameter not significantly changing with respect to fatty acid class (P > 0.05).
Cells were analyzed for both triglyceride presence and cellular viability. Oil Red O staining performed at 264 h indicated significant lipid accumulation in all 2D and 3D samples (Fig. 4). One problem noted in analysis was spheroid loss during the staining process, indicating the spheroids may be loosely attached to the ELP-PEI coated surfaces. This spheroid loss was minimized by embedding the spheroids in collagen gel prior to Oil Red O staining. To assess the cellular viability after the fatty acid treatment, a Live/Dead assay was performed at 216 h (Fig. 5). The red fluorescence shows the relative amount of ethidium homodimer that indicated cellular death. The green fluorescence indicated the presence of calcein that had interacted with cellular esterases inside live cells. Although some red fluorescent cells were present (Fig. 5II), the much higher ratio of green fluorescent cells (Fig. 5III) indicated that the fatty acid treatment did not result in major cytotoxicity. Low levels of red staining consistently appeared across all cultures, regardless of fatty acid composition, but principally observed among fibroblastic cell populations (indicated by arrows in Fig. 5).
At the 168-, 216-, and 264-h time points, the 3T3-L1 cells were harvested, lysed, and analyzed quantitatively for total DNA content, intracellular protein, and triglyceride accumulation. As shown in Figure 6, the total DNA content of the cells cultured in the 2D configuration was approximately 2–3 times that of the cells cultured in the 3D configuration (P ≤ 0.05, designated by ● in Fig. 6). This can be explained by the fact that the 3T3-L1 cells in 2D configuration increased in population until they reached confluence over the first 72-h culture period, after which they entered the contact-inhibited growth arrest phase, while the cells in 3D configuration likely entered the contact-inhibited growth arrest phase within the first 24 h of the culture period. Total protein analysis of cell samples taken from 3D spheroids prior to differentiation showed stable protein levels and also indicated that a constant cell population was maintained during the first 72 h. On the other hand, cells grown in monolayer showed initial increase in protein levels, which stabilized within 48 h (data not shown). The total DNA content of the 3T3-L1 cells cultured in the 2D configuration remained statistically unchanged over the 5-day period irrespective of whether they were fed control, LA, or OA media (P > 0.05 compared to control medium). However, the spheroids cultured in the medium containing 0.7 mM OA and LA showed statistically significant decrease in the total DNA content by the 264-h time point (P ≤ 0.05 compared to control medium on same day, designated by * in Fig. 6).
The total intracellular protein content for each condition mirrored trends as shown by the DNA results (data not shown). LA conditions had slightly lower amounts of protein present (P ≤ 0.05) compared to OA-containing experimental medium on same day. Total protein collected was significantly lower in cells cultured with 0.7 mM of both fatty acids in both 2D and 3D configurations, though the departure from controls was observed to be larger in 2D cultures and with exogenous LA at 264 h. However, total protein normalized to DNA values did not show any dependence of the culture type or growth conditions (data not shown), indicating similar differentiation level for cells under all culture conditions.
Figure 7 shows the triglyceride accumulation by 3T3-L1 cells when normalized to total intracellular protein. Increased normalized triglyceride levels correlated with 3D culture, OA dosage, and culture time (designated by *, ●, and $ in Fig. 7). Triglyceride accumulation increased in the 3D spheroids in response to exogenous OA dosages (designated by * in Fig. 7a). 3T3-L1 cells cultured as spheroids showed increased, albeit statistically non-significant, triglyceride accumulation when cultured in the medium containing LA over the entire 5-day period (Fig. 7b). In general, spheroids cultured in the medium containing OA displayed a marginally higher triglyceride accumulation than those cultured in the medium containing LA.
MTT viability data were collected from 2D (Fig. 8a) and 3D (Fig. 8b) culture systems, reported as a fraction of their respective controls. The viability assay largely reflected trends observed in the total protein assay; at 168 h, only the highest (0.7 mM) LA-fed 2D culture showed significantly reduced viability versus controls. By 264 h, all supplemented fatty acid cultures, with the exception of a set containing 0.25 mM LA and OA, showed significantly reduced viability versus controls. 2D cultures showed the largest apparent dose-dependent reduction in viability at only ∼30% of controls when treated with 0.7 mM LA for 5 days. Conversely, 3D cultures treated with fatty acids appeared at ∼60% of controls, with no distinction between LA versus OA or their dosages.
GC data presented in Figure 9 compares fatty acid uptake by 3T3-L1 cells within the first 48 h in experimental media (144–192 h). Figure 9a compares all fatty acid treatments (0–0.7 mM LA and OA) when normalized to the mean total protein values and shows that cells in 3D cultures consumed 2–3 times as much fatty acid as their 2D counterparts. To determine if the adipocytes preferred one fatty acid over the other, we determined the fraction of fatty acid consumed from whole media with respect to culture system (3D and 2D) when treated with 0.5 mM of individual fatty acid (OA or LA) or experimental medium containing equal dosages of both fatty acids (0.25 mM of OA + 0.25 mM LA). Figure 9b shows that cells in 2D and 3D culture systems showed no preference, consuming equivalent amounts of each fatty acid.
Adiposity studies using 3T3-L1s have demonstrated functionality, particularly triglyceride synthesis and lipid storage, to be dependent upon the differentiated state of the cells which, in turn, may be influenced by exogenous factors including hormone treatments and fatty acids as well as cell organization and architecture. Hydrogels formed from various synthetic and natural polymers have been shown to possess the optimum mechanical properties for the adipose tissue engineering applications (Daya et al., 2007; Duranti et al., 1998; Marler et al., 2000; Patel et al., 2005; Patrick, 2000). Unfortunately, preadipocytes embedded directly in such hydrogels remained functionally limited, possibly due to restriction on the volume expansion during the adipocyte maturation process (Gomillion and Burg, 2006; Patrick, 2001). Our method used a positively charged ELP-PEI coating to induce formation of functional and mobile surface-tethered spheroids, avoiding the problems related to the volume restriction during maturation faced by other 3D adipocyte culture systems.
Though 3T3-L1 cells cultured in monolayer have been shown to accumulate limited amounts of triglycerides when grown in DMEM with 10% FBS (Shiomi et al., 2011), the addition of exogenous OA (Kokta et al., 2008; Shiomi et al., 2011) and LA (Kokta et al., 2008), particularly in conjunction with insulin, have been found to significantly increase the triglyceride synthesis and lipid accumulation. Confluent 3T3-L1 cells stimulated to differentiate by a hormonal cocktail typically containing dexamethasone, IBMX, and insulin have been demonstrated to significantly upregulate triglyceride synthesis and respond selectively to exogenous fatty acid and nutrient treatments (Evans et al., 2000; Madsen et al., 2005; Shiomi et al., 2011). Evans et al. (2000) found that 2D cultures exposed to 0–200 µM of LA resulted in significantly enhanced triglyceride synthesis after 6 days of post-confluent differentiation, though noted no significant hypertrophic growth or changes in cell adhesion. Conversely, Madsen et al. (2005) reported that 3T3-L1 cells differentiated with 100 µM LA in 2D configuration dramatically decreased triglyceride synthesis at 10 days. Similar treatment with OA did not significantly alter triglyceride synthesis compared to controls. Shiomi et al. (2011) demonstrated that repeated differentiation cycles and cell selection may amplify basal triglyceride synthesis as well as cell response to exogenous LA and OA treatments. Therefore, it is worthwhile to compare the responses of adipocytes subjected to LA and OA treatments in our 3D culture system.
The experimental design represented in the current study deviated from previous models (Evans et al., 2000; Madsen et al., 2005; Shiomi et al., 2011) in that our cultures were exposed to fatty acid containing media after reaching confluence and receiving differentiation factors. Though LA might have otherwise antagonized adipocyte differentiation, as suggested by Madsen et al. (2005), prior growth arrest and stimulation by dexamethasone, IBMX, and insulin in our 2D and 3D cultures had already begun the differentiation process, resulting in greater triglyceride accumulation over the 5-day culture (Fig. 7b). Despite differences in experimental procedure, our results showed similar trends as other studies (Evans et al., 2000; Madsen et al., 2005) with regard to increased normalized triglyceride production in response to both OA and LA (Fig. 7). In this study, we exposed our cultures to significantly higher levels of fatty acid dosages (0.5–0.7 mM). This acute exposure of OA and LA, as expected, resulted in increased triglyceride accumulation over a shorter culture period compared to that reported by other researchers (Evans et al., 2000; Madsen et al., 2005). Micrographs taken throughout the experiment provided indications of this typical adipocyte differentiation behavior. Differentiation was induced at 72 h as the 2D cultures were observed to achieve confluence (Fig. 1) and 3D cultures achieved a stable spheroid size (Fig. 1), confirmed by analysis of collected micrographs (Fig. 3). The average spheroid size also increased steadily during differentiation (Fig. 3), which may be attributed to both hypertrophic growth from lipid storage and hyperplastic growth due to spheroid aggregation and combination (Fig. 2). At 264 h, both 2D and 3D cultures treated with fatty acids showed hypertrophic growth due to lipid accumulation relative to fatty acid dosage, with many cells achieving monolocular morphology, possessing a single large droplet in the cytoplasm (Figs. 2 and 4). It should be noted that average spheroid size was limited to approximately 120 µm under all culture conditions (Fig. 3), presumably due to metabolic limitations imposed by nutrient and oxygen availability (Tanzi and Fare, 2009).
GC analysis was useful both for verifying media retention of fatty acid by association with BSA as well as subsequent uptake by differentiated 3T3-L1 cells. A Restek marine oil standard was used to identify small (≤50 µM) contributions from 14:0 (myristate), 16:0 (palmitate), 16:1 (palmitoleate), 18:0 (stearate), 18:1 (vaccinate), and 20:4 (arachidotate), in addition to ∼45 µM of OA and LA already present in control media. The control media contained a total of approximately 240 µM of fatty acids, 90% of which was identified by comparison with the Restek standard. Interestingly, only LA and OA were found to be significantly depleted from the control media during the first 48 h of cell culture experiments (data not shown). Furthermore, analysis of cell-exposed fatty media confirmed that, when normalized to intracellular protein, spheroids were vastly more effective at consuming fatty acid from the media. GC analysis also confirmed that both culture systems did not completely deplete the fatty acid resources within the 2-day time period between media changes. Since all culture systems consumed comparable amounts of fatty acid when subjected to equivalent dosages (0.5 or 0.7 mM) of LA and OA, additional cultures were tested by simultaneously adding 0.25 mM of both fatty acids. This experiment indicated that 3T3-L1 adipocytes showed no preference for one fatty acid over another, as equivalent amounts of each were consumed (Fig. 9b).
Hypertrophic growth may also cause some cell loss due to increased cell volume combined with reduced interaction with the substrate, as indicated by total DNA (Fig. 6) and MTT (Fig. 8). This effect was found to be exaggerated in the 3D culture system. While the Live/Dead staining of 3D spheroids did not indicate significant cell morbidity at the 216-h time point (Fig. 5), regardless of the higher fatty acid dosage, the total DNA content showed a marked decrease by the 264-h time point for the spheroids treated with higher dosages of OA and LA (Fig. 6). Considering the significant overall cell loss indicated by total DNA (Fig. 6) and morbidity indicated by MTT (Fig. 8) in cultures exposed to exogenous LA, but at the same time maintaining both LA uptake (Fig. 9) and overall triglyceride accumulation throughout the culture period (Fig. 7), we surmise that the undifferentiated cell population was most affected by the elevated LA concentrations. We propose that the more completely differentiated 3D cultures were proportionally less affected by the LA, as indicated by the consistent MTT response across all fatty acids or dosage (Fig. 8).
While most cultures in our study indicated increasing average triglycerides over the duration of the experiment, suggesting ongoing lipogenesis, the increase was more pronounced in the 3D spheroid cultures compared to the 2D cultures. This increase in triglycerides indicated by 3D cultures versus 2D cultures may be attributed to more complete growth arrest, a critical condition prior to terminal adipocyte differentiation. Spheroids of adipose-derived stem cells cultured on chitosan films were similarly found to have hindered proliferation, though possessing enhanced viability, matrix production and differentiation potential versus equivalent monolayer cultures (Cheng et al., 2012). Wang et al. (2009), who created similar adipocytic spheroids by controlling mesenchymal stem cell adhesion on a micropatterned substrate, found 3D cultures more amenable to differentiation than monolayer analogs as indicated by upregulated differentiation-specific gene expression and intracellular lipid accumulation. Kang et al. (2007) electrospun multi-layered PCL scaffolds and cultured murine stem cell aggregates known as “embroid bodies” on the matrix. These 3D adipocyte cultures also showed increased differentiation-specific gene expression and lipid storage versus monolayer cultures. Future studies involving our 3D adipocyte model would similarly investigate differentiation-specific gene expression, especially in presence of various fatty acids.
This work describes construction of a 3D adipocyte culture system atop positively charged ELP-PEI coating to study the differentiation and maintenance process of 3T3-L1 adipocyte spheroids when subjected to various dosages of free fatty acids. Our 3D adipogenic culture system showed comparable cell response to exogenous OA and LA treatments to that cultured as a 2D monolayer, though a sustained overall increase in intracellular triglyceride storage and fatty acid uptake indicated a differential advantage provided by the 3D spheroid culture system. This outcome may prove advantageous for more rapidly promoting a differentiated phenotype in adipose cell cultures for investigating the influence of exogenous drugs and nutrient treatments on a mature cell population over a shorter in vitro experimental period.
This work was funded by the School of Dentistry and the University of Mississippi Medical Center intramural research support programs and made use of instruments in the Department of Biomedical Materials Science User Facility and those supported by the Shared Instrumentation Grant (S10RR027108). LMH participated in the Undergraduate and Professional Student Training in Advanced Research Techniques (UPSTART) Program and the Honors in Research Program. The authors thank Dr. Michael Roach for use of the Keyence microscope.
- 1959. A rapid method of total lipid extraction and purification. Can J Biochem Physiol 37:911–917. , .
- 1930. On the nature and role of the fatty acids essential in nutrition. J Biol Chem 86:5871–5879. , , .
- 2006. n-3 polyunsaturated fatty acids, inflammation, and inflammatory diseases. Am J Clin Nutr 83:1505S–1519S. .
- 2012. The influence of spheroid formation of human adipose-derived stem cells on chitosan films on stemness and differentiation capabilities. Biomaterials 33:1748–1758. , , .
- 2007. Culture and differentiation in two-dimensional and three-dimensional in vitro systems. Differentiation 75:360–370. , , .
- 1998. Injectable hyaluronic acid gel for soft tissue augmentation. Dermatol Surg 24:1317–1325. , , , , .
- 2000. Conjugated linoleic acid suppresses triglyceride accumulation and induces apoptosis in 3T3-L1 preadipocytes. Lipids 35:899–910. , , , , , .
- 2010. Prevalence and trends in obesity among US adults, 1999–2008. JAMA 303:235–241. , , , .
- 2006. Stem cells and adipose tissue engineering. Biomaterials 27:6052–6063. , .
- 1999. Depot-specific release of leptin from subcutaneous and omental adipocytes in suspension culture: Effect of tumor necrosis factor-alpha and transforming growth factor-beta1. Eur J Endocrinol 141:436–442. , , , , .
- 2008. ELP-polyelectrolyte structure confers control over hepatocyte morphology and function. Biomaterials 29:625–632. , , , .
- 2007. Adipogenesis of murine embryonic stem cells in a three-dimensional culture system using electrospun polymer scaffolds. Biomaterials 28:450–458. , , , , , , .
- 1979. Fatty acid composition of human adipose tissue from two anatomical sites in a biracial community. Am J Clin Nutr 11:2198–2205. , , , , .
- 2008. Regulation of lipid accumulation in 3T3-L1 cells: Insulin-independent and combined effects of fatty acids and insulin. Animal 2:92–99. , , , , , .
- 2005. Adipokines: Molecular links between obesity and atherosclerosis. Am J Physiol Heart Circ Physiol 288:2031–2041. , , , , .
- 2003. DNA microarray analysis of genes differentially expressed in diet-induced (cafeteria) obese rats. Obes Res 11:188–194. , , , , , , .
- 2005. Regulation of adipocyte differentiation and function by polyunsaturated fatty acids. Biochim Biophys Acta 1740:266–286. , , .
- 2000. Soft-tissue Augmentation with injectable alginate and syngeneic fibroblasts. Plast Reconstr Surg 105:2049–2058. , , , , , , .
- 1999. The spread of the obesity epidemic in the United States, 1991–1998. JAMA 282:1519–1522. , , , , , .
- 1997. Three-dimensional cell cultures: From molecular mechanisms to clinical applications. Am J Physiol 273:1109–1123. .
- 2005. Poly(ethylene glycol) hydrogel system supports preadipocyte viability, adhesion, and proliferation. Tissue Eng 11:1498–1505. , , ,
- 2000. Adipose tissue engineering: The future of breast and soft tissue reconstruction following tumor resection. Semin Surg Oncol 19(3):302–311.
- 2001. Tissue engineering strategies for adipose tissue repair. Anat Rec 263:361–366.
- 2002. The obesity epidemic: Pathophysiology and consequences of obesity. Obes Res 10:97S–103S. .
- 1979. Recent studies of the 3T3-L1 adipocyte-like cell line. Recent Prog Horm Res 35:477–499. , , , , .
- 2011. Compounds that inhibit triglyceride accumulation and TNFα secretion in adipocytes. J Biomed Sci Eng 4:684–691. , , .
- 2009. Adipose tissue engineering: state of the art, recent advances and innovative approaches. Expert Rev Med Devices 6:533–551. , .
- 2006. Adipose tissue inflammation induced by high-fat diet in obese diabetic mice is prevented by n-3 polyunsaturated fatty acids. Diabetologia 49:2109–2119. , , , , , , , , .
- 2013. Spheroid organization kinetics of H35 rat hepatoma model cell system on elastin-like polypeptide-polyethyleneimine copolymer substrates. J Biomed Mater Res A (available online ahead of print). , , , .
- 2008. Will all Americans become overweight or obese? Estimating the progression and cost of the US obesity epidemic. Obesity 16:2323–2330. , , , , .
- 2009. 3D spheroid culture system on micropatterned substrates for improved differentiation efficiency of multipotent mesenchymal stem cells. Biomaterials 30:2705–2715. , , , , , , .
- 2013. Suspension culture of hepatocyte-derived reporter cells in presence of albumin to form stable three-dimensional spheroids. Biotechnol Bioeng 110:2548–2555. , , , , , , .
- 2007. Modeling tissue morphogenesis and cancer in 3D. Cell 130:601–610. , .
- 2000. Ceiling culture of mature human adipocytes: Use in studies of adipocyte functions. J Endocrinol 164:119–128. , , , .