Microglia-mediated drug substance transfer promotes chemoresistance in brain tumors: insights from an in vitro co-culture model using GCV/Tk prodrug system

Abstract

Background: It is well known that tumor-associated macrophages (TAMs) play essential roles in brain tumor resistance to chemotherapy. However, the detailed mechanisms of how TAMs are involved in brain tumor resistance are still unclear and lack a suitable analysis model.

Methods: A BV2 microglial cells with ALTS1C1 astrocytoma cells in vitro co-culture system was used to mimic the microglia dominating tumor stroma in the tumor invasion microenvironment and explore the interaction between microglia and brain tumor cells.

Results: Our result suggested that microglia could form colonies with glioma cells under high-density culturing conditions and protect glioma cells from apoptosis induced by chemotherapeutic drugs. Moreover, this study demonstrates that microglia could hijack drug substances from the glioma cells and reduce the drug intensity of ALTS1C1 via direct contact. Inhibition of gap junction protein prevented microglial-glioma colony formation and microglia-mediated chemoresistance.

Conclusions: This study provides novel insights into how glioma cells acquire chemoresistance via microglia-mediated drug substance transferring, providing a new option for treating chemo-resistant brain tumors.

Keywords: Chemoresistance; Glioma; Microglia; Tumor microenvironment.

Fig. 1

Effect of microglia on brain tumor cells. a The diagram of cell co-culture experiment designation indicates high-density (3.2 × 10³/mm²), medium-density(1.6 × 10³/mm²), and low-density(0.8 × 10³/mm²) seeding conditions. b The images of co-culture on different seeding intensities after 24 h of incubation. The blue arrow indicates the colony (a cluster of diameter > 50 μm). Scale bar = 200 μm. c Statistics of the average colony diameter among different conditions. d The immunofluorescence staining of ALTS1C1-GFP (Shown in green) co-culture with BV2 for 24 h. The nucleus was stained with Hoechst (Left picture, shown in Blue), and the cells were stained with myeloid cell marker CD11b (Right, shown in red). Scale bar = 200 μm. e The images of ALTS1C1 were cultured in different condition mediums for 24 h of incubation. Scale bar = 100 μm. f The images of different cancerous cell lines co-cultured with BV2 for 24 h. Scale bar = 200 μm. A two-tailed unpaired t-test was used to compare every two groups. ****P < 0.0001

Fig. 2

Chemo-resistance of ALTS1C1 after co-culturing with BV2. a The MTT cytotoxicity assay of the pro-drug GCV on ALTS1C1, ALTS1C1-Tk, and BV2 cell lines for 72 h of incubation. b Representative dot plot FACs images of the co-culture chemo-apoptosis assay of 10 μg/mL pro-drug GCV on ALTS1C1-Tk only or ALTS1C1-Tk (+BV2) for 0, 24, 36 h. c Quantification of the chemo-apoptosis assay. N ≧ 3. d Representative caspase-3 staining images of ALTS1C1-GFP-Tk, ALTS1C1-GFP-Tk co-cultured with BV2 treated with 10 μg/mL GCV for 24 h. Scale bar = 200 μm. e Quantification data of the caspase-3 staining. N = 3. A two-tailed unpaired t-test was used to compare every two groups. *P < 0.05, **P < 0.01

Fig. 3

The effect of BV2 co-culturing on ALTS1C1 cells. a Representative flow dot plots of CD133, isotype staining on ALTS1C1 or the colony from the ALTS1C1 and BV2 after co-culturing for 24 h. b The quantification of flow analysis on CD133. c The cell cycle analysis on ALTS1C1 of different culturing conditions (Low density, High density, and co-cultured with BV2 on high density). A two-tailed unpaired t-test was used to compare every two groups. *P < 0.05, **P < 0.01

Fig. 4

BV2 hijacks drugs from ALTS1C1 through direct contact. a The experimental scheme on the Dox transfer test between the ALTS1C1 and BV2 cells. b Representative flow images of PE-positive cell percentages indicating Dox-positive cells (Left), and the quantification of Dox-positive cells after co-culturing for 4 h (Right). c In-direct transfer experiment scheme. d Representative flow intensity images of in-direct transfer experiment (Left), and the quantitative data of median fluorescence intensity (MFI) ratio of trans-well (in-direct) and direct contact groups. N = 3. e Representative flow intensity images of cells treated with Dox for 4, 12, and 24 h. f The quantification of median fluorescence intensity (MFI) of Dox on ALTS1C1, and ALTS1C1 (Co-cultured with BV2) for 4, 12, and 24 h. g Representative images of time-lapse picturing Dox transferring from ALTS1C1-GFP to BV2. Images were taken every 5 min. Scale bar = 25 μm. A two-tailed unpaired t-test was used to compare every two groups. **P < 0.01. ***P < 0.001, ****P < 0.0001

Fig. 5

The effect of gap junction on co-culture induced chemoresistance. a The representative images of various concentration CBX on ALTS1C1-Tk and BV2 co-culture. Scale bar = 500 μm. b The representative flow images of chemo-apoptosis assay on different concentrations of CBX on ALTS1C1-Tk and BV2 co-culture. c statistics of chemo-apoptosis assay, N = 3. d Representative dot plot FACs images of the co-culture chemo-apoptosis assay of 10 μg/mL pro-drug GCV and 10 μM CBX on ALTS1C1-Tk only or BV2 co-cultured group for 0, 24, 36 h. e Quantification of the chemo-apoptosis assay, the protective effect has significantly weakened under CBX treatment. N = 3. A two-tailed unpaired t-test was used to compare every two groups. *P < 0.05