Cryopreservable and Tumorigenic Three Dimensional Tumor Culture in Porous Poly(lactic-co-glycolic acid) Microsphere

Abstract

In vitro tumor models that mimic in vivo conditions may be ideal for screening anticancer drugs and their formulations and developing tumors in animal models. Three-dimensional (3-D) culture of cancer cells on polymeric scaffolds can be an option for such models. In the present study, porous poly(lactic acid-co-glycolic acid) (PLGA) microsphere was used both as a cancer cell culture substrate to expand cells and as a cancer cells transplantation vehicle for tumor construction in mice. MCF-7 cells cultured in porous PLGA microspheres in stirred suspension bioreactors expanded by 2.8-fold over seven days and maintained viability. At three months after inoculation with 2 × 10⁶ cells/site, the tumor formation by MCF-7 cells cultured on microspheres was much more effective (4 tumors/5 mice) than its counterpart cultured on plates (1/5). More importantly, cell viability and metabolic activity were not significantly changed even after one freeze-thaw cycle of the 3-D culture. MCF-7 cells cultured on the microspheres and the cells in 3-D after cryopreservation were more resistant to doxorubicin than MCF-7 cells cultured on plates.

1. INTRODUCTION

In vitro cancer models have taken their roles in pre-screening anticancer drugs and their test-formulations and a two-dimensional (2-D) monolayer culture model has been a standard to determine drug effects on growth inhibition and apoptosis [1–10]. However, in vivo and clinical cancers are 3-D and present differences from the 2-D model in cell surface receptor expression [11], proliferation [12], extracellular matrix synthesis [13], cell density [14] and metabolic function [15]. This may be one of major reasons for poor co-relationship between in vitro and in vivo measurements.

One of the major thrusts in the field of cancer therapy is to develop better in vitro models that would mimic in vivo tumors which will yield more practical assessment of drug efficacy prior to testing in animal models. Three-dimensional (3-D) culture of cancer cells on a polymeric scaffold has become more attractive by mimicking tissue environment in vivo [16–18] and is expected to provide improved in vitro/in vivo co-relationship for therapeutic evaluation. A porous polymeric scaffold provides a large surface area for cell attachment and growth. In particular porous microsphere scaffolds make 3-D suspension cultures feasible in a stirred suspension bioreactor. When such scaffolds are fabricated from biodegradable and non-cytotoxic materials, it can also be used as a cancer cell transplantation vehicle for tumor construction in small animals, eliminating trypsinization step of cultured cells which causes perturbation of cell-cell interactions and serious damage on cell surface proteins as well. Recent study has utilized porous microspheres for MCF-7 cell culture, demonstrated 3-D cell culture [19]. Porous microspheres made of poly(lactic-co-glycolic acid) (PLGA) were used as a cell culture substrate for evaluation of anticancer drugs efficacy [17].

An integral approach to evaluate anticancer drugs efficacy using in vitro 3-D cancer model have to involve a careful evaluation of cryopreservation performance. Cryopreservable 3D cancer model would play an important role in terms of “ready-to-use”. In this study, porous PLGA microspheres were utilized both as 3-D cell culture substrate to expand MCF-7 cells for evaluation of doxorubicin (Dox) efficacy and as a cell transplantation vehicle for tumor formation in vivo. We investigated whether the cells grown in the porous microsphere can be cryopreserved for “ready-to-use” and are tumorigenic. The cryopreservable and tumorigenic 3-D tumor culture in porous microsphere may be an effective way for drug screening and animal model development.

2. MATERIALS AND METHODS

2.1. Fabrication of porous PLGA microspheres

Porous microspheres were fabricated from 75:25 PLGA (molecular weight = 100,000 Da, Birmingham Polymers, Pelham, AL, USA) using a previously described water/oil/water double emulsion method [20]. Briefly, 2.5 ml of deionized water containing 500 mg of NH₄HCO₃ (Sigma, St Louis, MO, USA) was added to 8 ml of methylene chloride (Mallinckrodt, Inc., Phillipsburg, NJ, USA) containing 500 mg of PLGA. The mixture was homogenized using a Powergen 700 homogenizer (IKA, Tokyo, Japan) at 5000 rpm for 3 min. This water/oil emulsion was immediately poured into a beaker containing 300 ml of 0.1% (w/v) polyvinyl alcohol (molecular weight = 30,000~70,000 Da, Sigma) solution and then was re-emulsified using an overhead propeller (LR-400A, Fisher Scientific Co., USA) for 4 h at 200 rpm. After the solvent was evaporated, the microspheres were filtered into a size range of 300–450 μm, washed five times with distilled water, and lyophilized using a freeze dryer for three days. The porous PLGA microspheres were observed by a scanning electron microscope (JSM-6330F; JEOL, Tokyo, Japan).

2.2. MCF-7 cell culture onto porous PLGA microspheres

MCF-7 cells were purchased from the American Type Culture Collection (Manassas, VA, USA) and were maintained in RPMI-1640 medium containing 10% (v/v) fetal bovine serum (Gibco BRL). MCF-7 cells (5×10⁵ cells/ml) were cultured onto porous (0.1 mg/ml) PLGA microspheres in spinner flasks (50 ml, Wheaton, Millville, NJ, USA) at 40 rpm for two weeks. Five days after culture, the cell-attached microspheres were stained with hematoxylin and eosin (H&E) and examined under a light microscope to analyzer cell adhesion on the microspheres. One week after culture, cell-porous microsphere constructs were cryopreserved for three days then re-cultured for seven days. The number of cells on the porous microspheres was determined on days 1, 3, 5, and 7 by measuring the DNA content, as previously described [21].

2.3. Cryopreservation of 3-D culture

The solutions were prepared as reported previously [22]. All solutions were prepared in RPMI-1640 (pre-equilibration solutions containing 10% and 25% (v/v) ethylene glycol (EG), vitrification solution (VS) containing 40% (v/v) EG and 0.6 M sucrose and dilution solutions of descending sucrose concentrations (1 M, 0.7 M, 0.525 M, 0.35 M, and 0.175 M sucrose)). Cell-seeded porous microspheres were divided into two groups, with vitrification-warming cycle and without undergoing vitrification–warming as control group. The cell-porous microsphere constructs were initially placed in eppendorff tubes with 10% (v/v) EG for 4 min, followed by 25% (v/v) EG for 4 min and finally for 4 min in VS. Cell-porous microsphere constructs were then transferred in cell strainer to vapor phase of liquid nitrogen for 20 s before plunging into liquid nitrogen. Cell-porous microsphere constructs were maintained in the liquid nitrogen for three days. For warming process, cell–porous microsphere constructs were placed into 20 ml of 1 M sucrose solution in a 50 ml centrifuge tube which was constantly warmed at 38 °C water bath. After 1 min, the warmed samples were incubated in 20 ml of 1 M sucrose at room temperature for 4 min. Further washing of cell–porous microsphere constructs was carried out at room temperature at the intervals of 3 min in 5 steps from 0.7 M sucrose to RPMI-1640 prepared dilution solutions (0.7 M, 0.525 M, 0.35 M, and 0.175 M sucrose). Cell-porous microsphere constructs were re-cultured in spinner flasks for one week. The viability of cells was determined by MTT and live/dead assay

2.4. Determination of cell viability

Live and dead cells on the porous microspheres were assessed 3 days after vitrification for both control and vitrified groups with confocal microscopy after staining with fluorescein diacetate (FDA)/ethidium bromide (EB) (Sigma). The cells cultured on the porous microsphere were incubated in FDA/EB (5 μg/ml, 10 μg/ml, respectively) solution for 5 min at 37 °C and then washed twice in PBS. Dead cells stained red due to the nuclear permeability to EB. Viable cells, capable of converting the non-fluorescent FDA into fluorescein, stained green. After staining, the samples were examined using a confocal microscopic imaging system (Fluoview BX50, Argon ion laser, Olympus, Tokyo, Japan) equipped with an argon laser light source. Differential visualization of the fluorophores was achieved using a 488-nm excitation filter for both FDA and EB, a 522/535-nm emission filter for FDA, and a 605/632-nm emission filter for EB. Sample images were obtained using Lasershop Analyzer.

2.5. Activity of doxorubicin against 2-D Monolayer and 3-D Model

Activity of DOX was determined in cells grown in 2-D monolayer and in 3-D model following two days Dox treatment. Cell numbers (5×10⁵ cells) were similar in both 2-D and 3-D models at the time of treatment, and the cytotoxicity was determined using tetrazolium salt MTT assay. For 2-D Monolayer, six-well plates were used that provided more surface area for cells to grow without reaching confluency. For 3-D model, the cell-porous microsphere constructs were transferred to eppendorff tubes and incubated with 0.5 mg/ml MTT at 37 °C for 3 h. After removal of the supernatants, dimethyl sulfoxide (DMSO) was added, and the purple formazan crystals formed were dissolved in DMSO with vigorous shaking. The absorbance was measured with a spectrophotometer at a wavelength of 570 nm.

2.6. In vivo tumorigenicity

All procedures carried out on animals were subject to the guidelines of an approved protocol from University of Utah Institutional Animal Care and Use Committee. To establish human breast cancer xenografts, athymic female mice (eight months old, Charles river, CA, USA) were anesthetized by intramuscular injection of ketamine hydrochloride (8 mg/kg body weight) and xylazine hydrochloride (1.15 mg/kg body weight). MCF-7 cells were cultured on porous PLGA microspheres with RPMI1640 for two weeks in spinner flasks. Cell-porous microsphere construct (2×10⁶ cells/site) was injected into subcutaneous space of the mice through 16-gauge needles. As a control, MCF-7 cells cultured on plates (2×10⁶ cells/site) immediately injected into subcutaneous space of athymic mice. The parental MCF-7 cell line responds to estrogen with the increased level of progesterone receptors. Therefore, MCF-7 tumor growth was facilitated by feeding 1 mg estrogen per liter of water to mice [23]. They were monitored daily for signs of tumor growth. The tumor specimens were embedded in paraffin then sectioned with 4 μm thick. The sections were stained with H&E for morphologic analysis.

2.7. Statistical analysis

Quantitative data were expressed as the mean ± standard deviation. Statistical comparisons were carried out using Student’s t-test (SAS software, SAS Institute Inc., Cary, NC). A probability level of less than 0.05 was considered statistically significant.

3. RESULTS

The fabricated PLGA microsphere showed open pores on its surface (Fig. 1A) and interconnected structures between the inner pores (Fig. 1B). The pore morphology in the cross-section was similar to that on the surface, indicating that the porous structure is homogeneous throughout the microsphere (Fig. 1B). The average diameter of the microspheres was 393 ± 5 μm and the average pore size on the microsphere surface was 38 ± 7 μm.

Figure 1.

SEM micrographs of (A) gross morphology, (B) cross-section of porous PLGA microsphere. The scale bars indicate 100 μm.

To examine the adhesion MCF-7 cells on the surface of porous PLGA microsphere, the cells were cultured in spinner flasks containing porous microspheres. After 5 days, it was found that most cells attached uniformly to the surface of the porous microsphere. When MCF-7 cells were cultured onto the porous microspheres with a growth medium for two weeks, the cells migrated, proliferated, and formed tumor-like tissue in the pores (Fig. 2B).

Figure 2.

(A) H&E staining of MCF-7 cells cultured on porous PLGA microsphere in spinner flask for 5 days and (B) H&E staining of cross-section of MCF-7 cultured on porous PLGA microsphere in spinner flask for 14 days. The scale bars indicate 100 μm.

MCF -7 cells cultured on porous microspheres (3-D culture) for two weeks were injected into the subcutaneous dorsum of athymic mice. MCF-7 cells harvested from 2-D culture served as a control. Two months after injection, MCF-7 cells on the scaffold formed tumors at the injection sites with a high success rate of 4 out of 5 mice, while MCF-7 cells cultured on plates without porous microsphere failed to form tumors (Table 1). After three months, the control group reconstituted an entire tumor in one of 5 mice. H&E staining of porous microsphere group showed more metaplastic carcinoma sarcomatoid type than that of the control group by showing a phenotype of poorly developed cytoplasm caused by a fast proliferation rate (Fig. 3). The results suggest that the tumors formed from cells precultured on porous microsphere facilitated tumor progression more readily than did tumors formed from cells precultured in monolayers after inoculation.

Table 1.

Engraftment of MCF-7 cells into athymic nude mice

Time	Culture type	Tumor formation/mice
2 months	Culture on porous microspheres	4/5
	Culture on monolayer plates	0/5
3 months	Culture on porous microspheres	4/5
	Culture on monolayer plates	1/5

Figure 3.

Histological sections (H&E staining) of tumor formed by (A and B) only MCF-7 and (C and D) cell-microsphere implantation into subcutaneous space in the dorsum of athymic mouse. The scale bars indicate 100 μm.

Next, it was determined whether MCF-7 cells on porous PLGA microspheres in stirred suspension bioreactors could expand its population and can be cryopreserved. One day after culture, 62.0 ± 22.3% of the inoculated MCF-7 cells adhered to the microspheres, as determined by DNA content measurement. The number of MCF-7 cells cultured on the microspheres increased by 2.8-fold over seven days. After cryopreservation, the number of MCF-7 cells re-cultured on the microspheres increased by 2.0-fold over seven days. The growth rates were not significantly different between before and after cryopreservation. The mitochondrial metabolic activity using MTT test after cryopreservation was compared to before cryopreservation. The metabolic activity was measured at days 7 and 13 after culture (Fig. 4) because the cell numbers and growth rates of both two groups were not significantly different. The metabolic activity after cryopreservation was maintained at approximately 82% of that before cryopreservation (Fig. 5A). Confocal microscopic examination revealed that most cells cultured on the microspheres were viable before and after cryopreservation (Figs. 5B and C).

Figure 4.

The number of MCF-7 cells adherent to porous PLGA microspheres in spinner flask culture. The cell inoculum density was 5×10⁵ cells/ml. Cell-porous microsphere constructs were maintained in the liquid nitrogen for three days

Figure 5.

Viability of MCF-7 cells cultured on macroporous PLGA microspheres in spinner flask culture measured with (A) MTT assay and FDA/EB staining (B) before and (C) after cryopreservation. The viable and dead cells were stained green and red, respectively (scale bars = 100μm). *p<0.05.

Last, the sensitivity of 3-D tumor model to DOX was determined. IC₅₀ values of Dox were not significantly different between before and after cryopreservation (Fig. 6). In contrast, MCF-7 cells cultured on the plates were more sensitive to the drug about twice than that of the 3-D model.

Figure 6.

The cytotoxicity of the DOX in (■) 2D monolayer and 3D porous PLGA microspheres (●) on day 7 (before cryopreservation) and (▲) on day 13 (after cryopreservation). *p<0.05.

4. DISCUSSION

A large number of studies have demonstrated that in vitro 3-D tumor models can be used for understanding of micro-environmental conditions of clinical tumors [16, 18, 24–31] and for screening anticancer drugs [16, 32–34]. The in vitro 3-D tumor models have been largely based on gel systems [31–33] or spheroid cultures [24–30, 34]. However, such systems have the major drawback in reproducibility and difference in micro-environmental characteristics. The goal of this study was to evaluate tumorigenicity and cryopreservation capability of cultured cells on 3-D PLGA microspherical scaffold. This model easily established cancer xenografts at subcutaneous space of the mice and did not change viability and metabolic activity after cryopreservation.

In vitro expansion of the cell population on the scaffold in a stirred suspension bioreactor has advantages over conventional 2-D culture plates in a large scale cell culture. The porous microsphere with inter-connected pores provides a large surface area for cell adhesion and growth within the microspheres, and a large number of cell-seeded microspheres can be suspended in three-dimensional bioreactors. The porosity facilitates the transport of nutrients and oxygen through the inter-connected pores for cell growth within the microspheres [20, 35, 36]. The average pore size was 38 μm, which allows sufficient cell infiltration and seeding within the microspheres. The results showed that the porous PLGA microspheres permitted adhesion and proliferation of MCF-7 cells (Figs. 2 and 4).

It was proven that the cells on porous PLGA microspheres serve as a cell transplantation vehicle. The porous PLGA microspheres may facilitate the localization and delivery of cancer cells to desired sites of animals, and define a three-dimensional space for the formation of in vivo tumors. The microspheres were easily injected into subcutaneous sites in the dorsum of athymic mice. MCF-7 cells cultured on porous PLGA microsphere resulted in aggressive and extensive tumor formation in vivo (Table 1). Cell transplantation using the porous microsphere may be advantageous for the survival of transplanted cells. It has been reported that most transplanted cells must adhere to matrices in order to survive [37–39]. MCF-7 cells binding on porous microsphere may avoid anoikis or apoptosis due to cell-matrix interactions. More survival of transplanted MCF-7 cells would contribute to the much more extensive and active tumor formation in vivo.

The feasibility of ready-to-use in vitro tumor models relies on how to preserve the cells as a whole structure [22, 40]. Maintaining 3-D structure in culture is costly and labor-intensive. Hence, well-preserved viable constructs can daunt many handling of labor for maintenance of cultures. We have observed that the viability and functionality of MCF-7 cells cultured on the scaffold were well-preserved even after cryopreservation. No noticeable difference of the cell viability and metabolic activity between before and after cryopreservation was found when observed through confocal microscope and MTT assay (Fig. 5). In addition, the vitrification procedure had no significant impact on Dox resistance (Fig 6). Results indicate that cryopreserved cells on porous microsphere constructs were undamaged by vitrification.

5. CONCLUSION

This study demonstrates the feasibility of in vitro 3-D tumor model using porous PLGA microspheres as a cell culture substrate to expand MCF-7 cells and as a cell transplantation vehicle for tumor construction in vivo. This system was more tumorigenic than 2-D cultured counterpart and cryopreservable. In addition, doxorubicin resistance of the 3-D model was maintained even after a freezing-thawing cycle.