Zebrafish xenograft model for studying mechanism and treatment of non-small cell lung cancer brain metastasis

Abstract

Background: Brain metastasis (BM) is thought to be related to the mortality and poor prognosis of non-small cell lung cancer (NSCLC). Despite promising development of NSCLC treatment, the treatment of NSCLC BM is still not optimistic due to the existence of the blood-brain barrier (BBB) that prevent drug penetration, as well as the short median survival time of the patients left for treatment. In this context, further development of quick and effective pre-clinical models is needed in NSCLC BM treatment. Here, we report a model system using zebrafish to promote the development of drugs for patients with NSCLC BM.

Methods: Three different NSCLC cell lines (H1975, A549 and H1299) were used to establish zebrafish BM models. The embryo age and cell number for injection were first optimized. Metastatic cells were observed in the brain blood vessels of zebrafish and were verified by hematoxylin-eosin (HE) staining. Then, the metastasis potentials of H1975 and A549 with manipulated microRNA-330-3p (miR-330-3p) expression were also investigated. Finally, sensitivities of H1975 and A549 to osimertinib and gefitinib were tested.

Results: This zebrafish BM model could distinguish NSCLC cell lines with different BM potential. Over-expressed miR-330-p significantly improved the BM potential of the A549 cells while knockdown miR-330-p reduced the BM ability of the H1975 cells. Both osimertinib and gefitinib showed inhibition effect in zebrafish BM model with the inhibition rate higher than 50 %. H1975 cell showed much higher sensitivity to osimertinib rather than gefitinib both in vivo and in vitro.

Conclusions: We established zebrafish brain metastasis model for studying mechanism and treatment of NSCLC BM. This study provided a useful model for NSCLC brain metastasis that could be used to study the mechanism that drive NSCLC cells to the brain as well as identify potential therapeutic options.

Keywords: Brain metastasis; Non-small cell lung cancer; Xenograft; Zebrafish.

Fig. 1

Establishment of zebrafish NSCLC BM xenograft models. (A) Schematic diagram of imaging and collecting zebrafish brain. The zebrafish brain was imaged from vertical view (blue arrow) and excised (red dotted box) to extract RNA. (B) The schematic diagram showed zebrafish embryos at 2 dpf. The red dots indicated cancer cells injected into the perivitelline space (PVs). (C) Images of brain at different developmental stages of Tg (fli-1: EGFP) zebrafish (2-6 dpf). The white short dashed line indicated the midbrain of the zebrafish, and the white long dashed line indicated the zebrafish’s eyes and hindbrain. (D) The expression of claudin-5 in the zebrafish brain at different developmental stages (2-7 dpf). (E) The expression of mfsd2aa and mfsd2ab in the zebrafish brain at different developmental stages (2-6 dpf). (F) Survival curves of 2 dpf zebrafish with different tumor-bearing amount in PVs. About 50-400 H1975 cells (red fluorescence) were injected into the PVs of 2 dpf zebrafish, and the number of deaths was counted till 8 dpi. (G-I) Three human NSCLC cell lines were involved: H1975, A549 and H1299. About 100 cells were injected into the PVs of zebrafish at different developmental stages, and the brain of zebrafish was imaged at 1 dpi. The white arrows indicated cancer cells in blood vessels of zebrafish brain. (J-K) Quantification of the BM cells number in zebrafish at 1 dpi. Significance was considered when P values were lower than 0.05. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001. dpf: days post fertilization, dpi: days post injection

Fig. 2

Hematoxylin-eosin (HE) stanning of zebrafish with NSCLC BM at 4 dpi. (A) Zebrafish embryos with NSCLC BM were sliced according to the trajectory (blue straight line) at 4 dpi. (B) The vertical sections showed that there were obvious cancer cell clusters (white dotted line) and lesions (white arrows) in the brain of zebrafish. (C-E) The picture showed three horizontal slices of the same zebrafish brain, from the surface layer to the inner layer. The slices showed that there were cancer cells (white arrows) in different levels of the zebrafish brain

Fig. 3

NSCLC cells could not proliferate in the midbrain of zebrafish at 4 dpi. (A) The schematic diagram showed zebrafish embryos at 2 dpf. The red dots indicated cancer cells injected into the midbrain of zebrafish. (B) Survival curves of zebrafish with different tumor-bearing amount at midbrain. About 50-400 H1975 cells (red fluorescence) were injected into the midbrain of 2 dpf Tg (fli-1: EGFP) zebrafish, and the number of deaths was counted till 8 dpi. (C-G) Five human cancer cell lines were involved: human breast cancer cell lines MDA-MB-231 (positive control) and MCF-7 (negative control), human NSCLC cancer lines H1975, A549 and H1299. About 100 cells were injected into the midbrain of zebrafish at 2 dpf, and the midbrain of zebrafish was imaged at 1 dpi and 4 dpi. (H-M) Quantification of cell proliferation in the midbrain of zebrafish at 1 dpi and 4 dpi. The fold change of cells in the midbrain was determined by dividing the measured values at 1 dpi and 4 dpi by the average measured values at 1 dpi. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001

Fig. 4

Zebrafish NSCLC BM xenograft models discriminated the BM potentials of different cell lines. (A) Transwell culture system was used to assess cell invasion ability in vitro, and five cancer cell lines were involved: MDA-MB-231, MCF-7, H1975, A549 and H1299. Invaded cells were stained with crystal violet (0.5 %) and imaged after 24 h incubation. (B) Quantification and comparison of invaded cells in vitro. Colonies were quantified using Image Pro Plus. (C) Five cancer cell lines were involved: MDA-MB-231, MCF-7, H1975, A549 and H1299. About 100 cells (red fluorescence) were injected into the PVs of Tg (fli-1: EGFP) zebrafish at 2 dpf, and the brain of zebrafish was imaged at 4 dpi. The white arrows indicated cancer cells in blood vessels of zebrafish brain. (D) Quantification of BM rate at 4 dpi. The number of BM cells in the same zebrafish at 4 dpi was divided by the number of BM cells at 1 dpi. The ratio of these two was named as BM potential. If the BM potential was greater than 1, it was considered that the zebrafish had brain metastasis. (E-I) Quantification of BM cells at 1 dpi and 4 dpi. The two dots connected by a straight line represented the number of BM cells of the same zebrafish at 1 dpi and 4 dpi. (J) Quantification and comparison of cell BM potential. The BM potential was determined by dividing the number of BM cells in the same zebrafish at 4 dpi by the number of BM cells at 1 dpi. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001

Fig. 5

Zebrafish NSCLC BM xenograft models simultaneously discriminated the BM potentials of different cell lines. (A) The schematic diagram showed zebrafish embryos at 2 dpf. The red dots and green dots indicated cancer cells labeled with two different dyes were co-injected into the PVs of zebrafish. (B) About 50 H1975 cells (red fluorescence) and 50 H1975 cells (green fluorescence) were co-injected into the PVs of wild zebrafish at 2 dpf, and the brain of zebrafish was imaged at 4 dpi. The white long dashed line indicated the brain of zebrafish. (C) About 50 H1975 cells (red fluorescence) and 50 A549 cells (green fluorescence) were co-injected into the PVs of wild zebrafish at 2 dpf, and the brain of zebrafish was imaged at 4 dpi. (D-G) Quantification of BM cells in co-injection at 1 dpi and 4 dpi. The two dots connected by a straight line represented the number of BM cells of the same zebrafish at 1 dpi and 4 dpi. (H and I) The percentage of cells with two different dyes in the brain of zebrafish. The percentage of the cell line (red/green fluorescence) at 1 dpi was determined by dividing the number of BM cells (red/green fluorescence) by the total number of BM cells at 1 dpi. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001

Fig. 6

Expression of microRNA-330-3p (miR-330-3p) affected the NSCLC BM potential in zebrafish xenograft models. (A and B) Quantification of the miR-330-3p expression in vitro. The appropriate transfection concentration was determined between 1 nM and 50 nM, according to the instructions. (C and D) The expression of miR-330-3p in H1975 cell line was knocked down by anti-miR-330-3p inhibitor and the expression of miR-330-3p in A549 cell line was overexpressed by over-miR-330-3p mimics. Transwell culture system was used to evaluate the effectiveness of transfection in vitro. Invaded cells were stained with crystal violet (0.5 %) and imaged after 24 h incubation. (E and F) Quantification of invaded cells in vitro. Colonies were quantified using Image Pro Plus. (G and H) H1975, H1975 with under-expressed miR-330-3p, A549 and A549 with over-expressed miR-330-3p were involved. About 100 cells (red fluorescence) were injected into the PVs of Tg (fli-1: EGFP) zebrafish at 2 dpf to evaluate the effectiveness of transfection in vivo, and the brain of zebrafish was imaged at 4 dpi. The white arrows indicated cancer cells in blood vessels of zebrafish brain. (I-L) Quantification of BM cells at 1 dpi and 4 dpi. The two dots connected by a straight line represented the number of BM cells of the same zebrafish at 1 dpi and 4 dpi. (M and N) Quantification and comparison of the BM potentials of the transfected cell lines and the original cell lines. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001

Fig. 7

Zebrafish NSCLC BM xenograft models discriminated different chemosensitivities within 4 days. (A and B) NSCLC cell lines H1975 and A549 were involved. Transwell culture system was used to evaluate the effectiveness of osimertinib (1 µM) and gefitinib (2 µM) against cell invasion in vitro. Invaded cells were stained with crystal violet (0.5 %) and imaged after 24 h incubation. (C and D) Quantification of invaded cells in vitro. Colonies were quantified using Image Pro Plus. (E and F) NSCLC cell lines H1975 and A549 were involved. Tg (fli-1: EGFP) zebrafish embryos were injected about 100 cells (red fluorescence) into the PVs at 2 dpf and administered with osimertinib (1 µM) and gefitinib (13 µM) by intracardiac injection. (G and H) Quantification of cell brain metastasis at 4 dpi. Fold change of BM cell number was determined by dividing the number of BM cells in the same zebrafish at 4 dpi by the number of BM cells at 1 dpi. (ns) indicated statistical insignificance, (*) indicated statistical significance P < 0.05, (**) P < 0.01 and (***) P < 0.001