Using Embryonic Germ Cells To Engineer Adipose Tissue

Alexander T. Hillel, Shyni Varghese, Jennifer Petsche, Michael J. Shamblott, Jennifer H. Elisseeff. Embryonic Germ Cells Are Capable of Adipogenic Differentiation In Vitro and In Vivo. Tissue Engineering Part A. March 2009, 15(3): 479-486. doi:10.1089/ten.tea.2007.0352.

Summary:

Adipose tissue can be used as filler material for reconstructive and cosmetic surgeries such as breast reconstruction for mastectomy patients. However, transplantation of autologous adipose tissue is unfavorable for numerous reasons such as "poor revascularization, graft shrinkage, and volume loss, often resulting in a poor cosmetic outcome" (Hillel et al. 2009). It is therefore of interest to explore alternative methods of obtaining adipose cells and adipose tissue. Previous studies have shown that Mesenchymal Stem Cells (MSCs) can be successfully differentiated into adipose cells. However, there are several limitations to using MSCs to obtain adipose cells such as their limited in vitro proliferation and differentiation ability. The authors of this paper sought to show that embryonic germ cells can be differentiated into adipose cells as an alternative to MSCs.

Given the proper growth conditions, embryonic germ cells form embryoid bodies, which later yield embryoid-body derived (EBD) cells. EBD cells are pluripotent and have a greater capacity for long-term expansion, proliferation, and differentiation than MSCs. The purpose of this study is to examine whether EBD cells are capable of adipogenic differentiation in vitro and in vivo. In this study, human EBD cells were induced towards adipogenic differentiation (through exposure to adipogenic medium) in three different culture settings: (1) in vitro after being grown in a 2D monolayer, (2) in vitro after being encapsulated in a 3D poly(ethylene glycol) diacrylate (PEGDA) based hydrogel and (3) pre-differentiated in vitro after being encapsulated in a 3D PEGDA-based hydrogel and later implanted in vivo in four, 6 week old athymic-nude mice to test for the persistance of the adipogenic phenotype. Additionally, human MSCs were adipogenically differentiated using the same experimental conditions for each of the three different culture settings to allow for a direct comparison to EBD cells.

Results:

1. In vitro 2D monolayer culture

After being exposed to adipogenic medium for 21 days, both the EBD cells and MSCs tested positive for adipogenic differentiation by oil red-O staining, though the intensity of of the staining was greater in the EBD cells (Figure 1). However, the MSCs demonstrated a morphological change from a spindle shape to a spherical shape (which is more typical of adipose tissue) while the EBD cells retained a fibroblast-like spindle shaped morphology (Figure 1D, 1B). Reverse transcription PCR analysis also confirmed adipogenic differentiation of both EBD cells and MSCs.

Figure 1: in vitro 2D monolayer cultures after 3 weeks,
(A) 40x oil red-O staining of EBD cells, (B) 100x oil red-O staining of EBD cells,
(C) 40x oil red-O staining of MSCs, (D) 100x oil red-O staining of MSCs

2. In vitro 3D PEGDA hydrogel culture

After being exposed to adipogenic medium for 4 weeks, both the EBD cells and MSCs tested positive for adipogenic differentiation by oil red-O staining, and, simliar to what was observed in the 2D culture, the intensity of the staining was greater in the EBD cells (Figure 2). Both the differentiated EBD cells and differentiated MSCs were found to have a spherical morphology (characteristic of adipose cells) (Figure 2B, 2D). Gene expression of the adipogenic markers PPARγ, αP2, and LPL in the EBD cells and MSCs were comparable (Figure 3). In comparison to the 2D cultures, the differentiated EBD cells and MSCs grown in 3D cultures showed many-fold increases in gene expression of the aforementioned adipogenic markers, suggesting that cells differentiated in 3D culture are more representative of the cells in actual adipose tissue (Figure 4).

Figure 2: in vitro 3D hydrogel culture after 4 weeks of bathing in adiopogenic medium
(A) 100x oil red-O staining for lipids of EBD cells, (B) 400x oil red-O staining of EBD cells,
(C) 100x oil red-O staining of MSCs, (D) 400x oil red-O staining of MSCs

Figure 3: Reverse-transcription PCR results of in vitro 3D hydrogel cultures that demonstrates gene expression of several markers of adipogenic differentiation

Figure 4: Results of real-time PCR analysis that compares upregulation of several gene markers of adipogenic differentiation in undifferentiated (UD), 2D, and 3D in vitro cell cultures

3. In vivo culture

Four weeks after implantation, both the EBD cells and MSCs demonstrated strong oil red-O staining, and, similar to what was observed in the in vitro cultures, the intensity of the staining was greater in the EBD cells (Figure 5B, 5D). It was shown that as implantation time increased, LPL expression increased for both EBD cells and MSCs (Figure 5A, 5C). However, the EBD cells did not appear to express PPARγ and also showed decreasing expression of αP2 with time (αP2 disappeared altogether 3 weeks after implantation) (Figure 5A). In contrast, the implanted MSCs maintained expression of PPARγ, αP2, and LPL for the duration of 4 weeks after implantation (Figure 5C). The authors suggest that the EBD cells may have demonstrated decreased differentiation in the in vivo culture due to "a lack of signaling or poor host vascular invasion" (Hillel, et al. 2009).

Figure 5: in vivo culture
(A) Results of reverse-transcription PCR anaylsis that demonstrates gene expression of several markers of adipogenic differentiation for EBD cells and (C) for MSCs. (B) 400x Oil red-O staining for EBD cells and (D) for MSCs.

Significance:

The authors contend that "EBD cells demonstrated adipogenic differentiation comparable to that of MSCs" (Hillel et al. 2009). However, while the differentiated EBD cells maintain a stable adipogenic phenotype during in vitro growth, they appear to have difficulty maintaining their adipogenic phenotype once they are implanted in vivo. This indicates that adipogenically differentiated MSCs may be more suitable for clinical procedures that involve in vivo expansion.

Nonetheless, this study suggests two things: (1) In vitro adipogenic differentiation of EBD cells and MSCs grown in a 3D PEGDA-based hydrogel produces cells that more closely resemble actual adipocytes than EBD cells and MSCs that are differentiated in a 2D culture, and (2) EBD cells can indeed be used as an alternate source of adipocytes as they are able to proliferate and preserve a stable adipogenic phenotype when differentiated in a 3D PEGDA-based hydrogel structure. This is significant because it means that (1) if one is looking to engineer cells in vitro that closely resemble adipocytes, they may differentiate either MSCs or EBDs in a 3D PEGDA-based hydrogel structure, and (2) if one wishes to utilize some of the advantageous properties EBD cells have over MCSs (such as a greater proliferation rate after long-term expansion and the ability to differentiate after a greater number of population doublings) for procedures that involve engineering adipose cells in vitro, then they may because it was shown that EBD cells can differentiate to adipocytes in vitro with a degree of success that is comparable to MSCs. This study advances the field of adipose tissue engineering by helping to identify the ideal stem cell for specific situations (in vitro expansion or in vivo expansion).