Three-dimensional culture and clinical drug responses of a highly metastatic human ovarian cancer HO-8910PM cells in nanofibrous microenvironments of three hydrogel biomaterials

Abstract

Background: Ovarian cancer is a highly aggressive malignant disease in gynecologic cancer. It is an urgent task to develop three-dimensional (3D) cell models in vitro and dissect the cell progression-related drug resistance mechanisms in vivo. In the present study, RADA16-I peptide has the reticulated nanofiber scaffold networks in hydrogel, which is utilized to develop robust 3D cell culture of a high metastatic human ovarian cancer HO-8910PM cell line accompanied with the counterparts of Matrigel and collagen I.

Results: Consequently, HO-8910PM cells were successfully cultivated in three types of hydrogel biomaterials, such as RADA16-I hydrogel, Matrigel, and collagen I, according to 3D cell culture protocols. Designer RADA16-I peptide had well-defined nanofiber networks architecture in hydrogel, which provided nanofiber cell microenvironments analogous to Matrigel and collagen I. 3D-cultured HO-8910PM cells in RADA16-I hydrogel, Matrigel, and collagen I showed viable cell proliferation, proper cell growth, and diverse cell shapes in morphology at the desired time points. For a long 3D cell culture period, HO-8910PM cells showed distinct cell aggregate growth patterns in RADA16-I hydrogel, Matrigel, and collagen I, such as cell aggregates, cell colonies, cell clusters, cell strips, and multicellular tumor spheroids (MCTS). The cell distribution and alignment were described vigorously. Moreover, the molecular expression of integrin β1, E-cadherin and N-cadherin were quantitatively analyzed in 3D-cultured MCTS of HO-8910PM cells by immunohistochemistry and western blotting assays. The chemosensitivity assay for clinical drug responses in 3D context indicated that HO-8910PM cells in three types of hydrogels showed significantly higher chemoresistance to cisplatin and paclitaxel compared to 2D flat cell culture, including IC₅₀ values and inhibition rates.

Conclusion: Based on these results, RADA16-I hydrogel is a highly competent, high-profile, and proactive nanofiber scaffold to maintain viable cell proliferation and high cell vitality in 3D cell models, which may be particularly utilized to develop useful clinical drug screening platform in vitro.

Keywords: 3D cell culture; Cell growth pattern; Chemosensitivity; HO-8910PM cells; Hydrogel; Nanofiber.

Fig. 1

Negatively staining TEM images showed the nanofiber morphology of RADA16-I (a, d), Matrigel (b, e), and collagen I (c, f). RADA16-I, Matrigel, and collagen I were dissolved in 0.1 × PBS solution and self-assembled spontaneously, respectively. Three types of hydrogels all presented a collection of interwoven and disorganized nanofiber networks ultra-microarchitectures in saline solution. Scale bar represented different resolution in TEM images (top panels (a–c): ×8000; bottom panels (d–f): ×15,000)

Fig. 2

HO-8910PM cell viability and cell proliferation in 3D cell cultures. HO-8910PM cells were encapsulated in RADA16-I hydrogel, Matrigel and collagen I and cultivated for 1 day (a), 3 days (b), and 6 days (c), respectively. Gel-cell clumps were successfully harvested when 3D cell cultures were over at the desired time points. The photomicrographs were taken by phase contrast microscopy (top panels in a–c). The green fluorescent calcein-AM staining for the living cells was performed to indicate cell viability and robust cell proliferation with distinct cell shapes in all hydrogel volume (bottom panels in a–c)

Fig. 3

The viable cell proliferation curves of HO-8910PM cells cultured in RADA16-I hydrogel, Matrigel and collagen I on days 1, 3, 6 and 9. a HO-8910PM cell proliferation in different hydrogels was calculated from DNA content fold changes at the desired time points. b BrdU labelling assay was performed to evaluate the viable cell percentage of HO-8910PM cells seeded in different nanofiber scaffolds, expressed as the cell proliferation rate (%) in three assays (about 200 cells per microscopical view field) on cell culture days 1, 3, 6, and 9. Data showed statistically significant differences (*p < 0.05, **p < 0.01) when compared with RADA16-I hydrogel

Fig. 4

Cell-to-cell interactions and geometry arrangements of HO-8910PM cells in gel-cell clumps. HO-8910PM cells formed MCTS in three types of hydrogel matrices on days 6 (a) and days 12 (b). Gel-cell clumps of HO-8910PM cell line in RADA16-I, Matrigel and collagen I were harvested on day 6 and day 12. Immunofluorescence assay was performed to indicate cell-to-cell interactions and geometry arrangements in 3D cell culture. Red indicated F-actin and blue indicated DAPI-stained cell nuclear. Immunofluorescence images were obtained by an inverted Olympus IX71 microscope. The blue nuclear staining showed HO-8910PM cell arrangement in MCTS and the red staining surrounding blue nuclear envisioned the cell-to-cell adhesion or intercellular junction in MCTS

Fig. 5

Representative images of MCTS for cell aggregate growth patterns in one tenth scaffold concentration and cell density when cultured for 6, 9, 12, and 15 days, respectively. HO-8910PM cells formed MCTS in three types of hydrogels and maintained different cell aggregate growth patterns in 3D cell culture for 15 days. MCTS was imaged through bright-field phase contrast microscopy. Phase contrast images were some representatives for HO-8910PM cells harvested at the desired time points. Scale bar was 100 μm

Fig. 6

The cell distribution and molecular expression of integrin β1 (a), E-cadherin (b) and N-cadherin (c) in HO-8910PM cells cultured in RADA16-I hydrogel, Matrigel and collagen I for 7 days. The brown color indicated the viable proliferation cells, which revealed a solid tumor-like tissue architecture and MCTS formation that was stained by hematoxylin in sections. d Relative quantification of protein expression levels in RADA16-I hydrogel, Matrigel and collagen I. Immuno-expression of integrin β1, E-cadherin and N-cadherin was quantified as an index of positively staining area over total hematoxylin-staining section area in gel-cell clumps. **P < 0.01 denoted hydrogel matrix compared; ^##P < 0.01 denoted cell adhesion molecules compared. Scale bars were 200 μm

Fig. 7

Western blot analysis of key cell adhesion proteins in HO-8910PM cells cultured in RADA16-I hydrogel, Matrigel and collagen I for 7 days. a Immunoblot showed qualitative molecular expression of integrin β1, E-cadherin and N-cadherin in different hydrogel biomaterials. b The curve graph indicated the densitometry quantitation of integrin β1, E-cadherin and N-cadherin protein expression levels normalized to GAPDH. Data represented the mean ± SD in three independent assays and showed statistical difference. *P < 0.05 indicated significant difference and **P < 0.01 indicated higher significant difference when compared with RADA16-I peptide scaffold. ^#P < 0.05 indicated significant difference and ^##P < 0.01 indicated higher significant difference when compared with Matrigel

Fig. 8

Drug responses of chemosensitivity assay in HO-8910PM cells cultured in gel-cell clumps and common 2D flat cell plates by clinically first-line chemotherapeutic agents (cisplatin and paclitaxel). The precultured cancer cells in three types of hydrogel biomaterials were evaluated by the end point assay for cell survival. ^a(P < 0.01) compared with that of 2D cell culture model. ^b(P < 0.01) compared with that of Matrigel. ^c(P < 0.01) compared with that of RADA16-I hydrogel