A standardized and reproducible protocol for serum-free monolayer culturing of primary paediatric brain tumours to be utilized for therapeutic assays

Abstract

In vitro cultured brain tumour cells are indispensable tools for drug screening and therapeutic development. Serum-free culture conditions tentatively preserve the features of the original tumour, but commonly comprise neurosphere propagation, which is a technically challenging procedure. Here, we define a simple, non-expensive and reproducible serum-free cell culture protocol for establishment and propagation of primary paediatric brain tumour cultures as adherent monolayers. The success rates for establishment of primary cultures (including medulloblastomas, atypical rhabdoid tumour, ependymomas and astrocytomas) were 65% (11/17) and 78% (14/18) for sphere cultures and monolayers respectively. Monolayer culturing was particularly feasible for less aggressive tumour subsets, where neurosphere cultures could not be generated. We show by immunofluorescent labelling that monolayers display phenotypic similarities with corresponding sphere cultures and primary tumours, and secrete clinically relevant inflammatory factors, including PGE2, VEGF, IL-6, IL-8 and IL-15. Moreover, secretion of PGE2 was considerably reduced by treatment with the COX-2 inhibitor Valdecoxib, demonstrating the functional utility of our newly established monolayer for preclinical therapeutic assays. Our findings suggest that this culture method could increase the availability and comparability of clinically representative in vitro models of paediatric brain tumours, and encourages further molecular evaluation of serum-free monolayer cultures.

Figure 1. Sphere-forming capacity and propagation of paediatric brain tumour spheres on a low-adherent surface (Ultra-Low™).

(a) Single-cell dissociated tumour cells, despite grade of malignancy, formed spheres in Ultra-Low™ 6-well plates when seeded at a concentration of 200 000 cells/ml of UC. Representative pictures of a pilocytic astrocytoma (PA-LU 228, WHO grade I), an anaplastic astrocytoma (AA-LU 74, WHO grade III) and a medulloblastoma (MB-LU 181, WHO grade IV) are shown. (b) Representative pictures of tumour cell proliferation, as determined by labelling of Ki-67 (green), within a tumour sphere. Anaplastic astrocytoma (AA-LU 74); light microscopy picture, labelling of Ki-67 on tumour spheres attached to a multi-chamber cell culture slide, scale bar is 100 μm. Medulloblastoma (MB-LU 140); labelling of Ki-67 on sectioned tumour spheres, scale bar is 20 μm. DAPI (blue) was used to stain the nuclei.

Figure 2. Establishment of proliferating paediatric brain tumour cells on an adherent surface (Cell-Bind™).

(a) Single-cell dissociated tumour cells and tumour spheroids (either transferred from Ultra-Low 6-well plates, or spheres formed in Cell-Bind™ plates) were forced to attach by culturing in UC on an adherent surface (Cell-Bind™). Cells were passaged when attached spheres flattened on the surface, as illustrated by the middle pictures. Monolayer cultures were formed after passaging the cells multiple times. Representative pictures of an anaplastic astrocytoma (AA-LU 74) and a medulloblastoma (MB-LU 68). (b) Representative picture of proliferating adherent medulloblastoma cells (MB-LU 140), as determined by Ki-67 labeling (green). Scale bar is 50 μm. DAPI (blue) was used to stain the nuclei. Proliferation of adherent paediatric brain tumour cells (d1 >passage 5) were determined by counting the number of cells between passages. Cells were either seeded in T-75 Cell-Bind™ flasks (anaplastic ependymoma, AEP-LU 158) or in Cell-Bind™ 6-well plates (medulloblastoma, MB-LU 140). Doubling-time (DT) was calculated at growth-phase (between day 50–60 for AEP-LU 158, between day 30–40 for MB-LU 140).

Figure 3. Immunofluorescent staining of neural markers in primary tumours, sphere cultures and monolayer tumour cell cultures.

Representative pictures of tumour sections, spheres and monolayer cells immunolabelled (green) with antibodies against (a) nestin, (b) β-III-tubulin and (c, d) GFAP. Scale bars are 20–100 μm. DAPI (blue) was used to stain the nuclei.

Figure 4. Immunofluorescent staining of CD markers in primary tumours, sphere cultures and monolayer tumour cell cultures.

Representative pictures of tumour sections, spheres and monolayer cells immunolabelled with antibodies against (a) CD133 (green), (b) CD29 (red), (c) CD44 (red) and (d) CD15 (green). Scale bars are 20–100 μm. DAPI (blue) was used to stain the nuclei.

Figure 5. Expression of PGE₂ and its rate-limiting enzyme COX-2 in vivo and by monolayer tumour cells in vitro.

(a) PGE₂ levels in plasma from patients with medulloblastoma (MB-LU 128, MB-LU 140, MB-LU 159) and anaplastic ependymoma (AEP-LU 158) was analyzed with ELISA. Bars and error bars represent mean and SEM. (b) Tumor sections and monolayer cells immunolabelled (green) with antibodies against cycloogygenase-2 (COX-2), the rate-limiting enzyme for prostaglandin E-2 (PGE₂) production. (c) 100 000 monolayer tumour cells were cultured for 24 hours with and without the presence of the COX-2 inhibitor Valdecoxib (100 μM), and supernatants were analysed with ELISA. Bars and error bars represent median and range, measured in three separate experiments for each cell line.

Figure 6. Cytokine profiling of primary tumour tissue and monolayer tumour cells in vitro.

(a–d) Cytokine levels in monolayer cell supernatants and tumour tissue lysates were measured with multiplex cytokine assays. Bars and error bars represent mean and SEM of three separate experiments. 1, AA-LU 74 (24 h); 2, AA-LU 74 (48 h); 3, AA-LU 74 (72 h); 4, AEP-LU 149 (48 h); 5, AEP-LU 158 (48 h); 6, MB-LU 69 (48 h); 7, MB-LU 70 (48 h); 8, MB-LU 140 (48 h); 9, AEP-LU 158 (tissue); 10, MB-LU 140 (tissue).