Infigratinib (BGJ 398), a Pan-FGFR Inhibitor, Targets P-Glycoprotein and Increases Chemotherapeutic-Induced Mortality of Multidrug-Resistant Tumor Cells

Abstract

The microtubule-targeting agents (MTAs) are well-known chemotherapeutic agents commonly used for therapy of a broad spectrum of human malignancies, exhibiting epithelial origin, including breast, lung, and prostate cancer. Despite the impressive response rates shortly after initiation of MTA-based therapy, the vast majority of human malignancies develop resistance to MTAs due to the different mechanisms. Here, we report that infigratinib (BGJ 398), a potent FGFR1-4 inhibitor, restores sensitivity of a broad spectrum of ABCB1-overexpressing cancer cells to certain chemotherapeutic agents, including paclitaxel (PTX) and doxorubicin (Dox). This was evidenced for the triple-negative breast cancer (TNBC), and gastrointestinal stromal tumor (GIST) cell lines, as well. Indeed, when MDR-overexpressing cancer cells were treated with a combination of BGJ 398 and PTX (or Dox), we observed a significant increase of apoptosis which was evidenced by an increased expression of cleaved forms of PARP, caspase-3, and increased numbers of Annexin V-positive cells, as well. Moreover, BGJ 398 used in combination with PTX significantly decreased the viability and proliferation of the resistant cancer cells. As expected, no apoptosis was found in ABCB1-overexpressing cancer cells treated with PTX, Dox, or BGJ 398 alone. Inhibition of FGFR-signaling by BGJ 398 was evidenced by the decreased expression of phosphorylated (i.e., activated) forms of FGFR and FRS-2, a well-known adaptor protein of FGFR signaling, and downstream signaling molecules (e.g., STAT-1, -3, and S6). In contrast, expression of MDR-related ABC-transporters did not change after BGJ 398 treatment, thereby suggesting an impaired function of MDR-related ABC-transporters. By using the fluorescent-labeled chemotherapeutic agent PTX-Alexa488 (Flutax-2) and doxorubicin, exhibiting an intrinsic fluorescence, we found that BGJ 398 substantially impairs their efflux from MDR-overexpressing TNBC cells. Moreover, the efflux of Calcein AM, a well-known substrate for ABCB1, was also significantly impaired in BGJ 398-treated cancer cells, thereby suggesting the ABCB1 as a novel molecular target for BGJ 398. Of note, PD 173074, a potent FGFR1 and VEGFR2 inhibitor failed to retain chemotherapeutic agents inside ABCB1-overexpressing cells. This was consistent with the inability of PD 173074 to sensitize Tx-R cancer cells to PTX and Dox. Collectively, we show here for the first time that BGJ 398 reverses the sensitivity of MDR-overexpressing cancer cells to certain chemotherapeutic agents due to inhibition of their efflux from cancer cells via ABCB1-mediated mechanism.

Keywords: ABC-transporters; FGFR-inhibitors; apoptosis; doxorubicin; paclitaxel; resistance; sensitization.

Figure 1

Expression of ABCB1, ABCC1 (MRP-1), and ABCG2 transporters in Tx-R cancer cells vs. Tx-sensitive parental HCC 1806 (A) and GIST T-1 (B) cells. Actin stain was used as a loading control.

Figure 2

Fluorescence microscopy analysis of the intracellular accumulation of Flutax-2 (PTX-FL) (A) and doxorubicin (DOX) (B) in HCC 1806 parental (upper panels) and Tx-R (lower panels) cells. The cells were first incubated with BGJ 398 (20 µM) (right) or DMSO as a control (middle) for 60min and then incubated with 3 µM PTX-FL (A) or 40 µM DOX (B) for an additional 60 min. After the wash-out with a pre-warmed FBS-free culture medium, the BGJ 398-treated cells were additionally incubated with BGJ 398 for 60 min. The non-fixed slides were counterstained with Hoechst 33342 (final concentration 3 µg/mL) for 5 min to outline the nuclei and processed for fluorescence microscopy to obtain the merged images.

Figure 3

BGJ 398 increases the intracellular accumulation of Dox (A) and Flutax-2 (B) in Tx-R HCC 1806 breast cancer cells. (A) The intracellular accumulation of Dox in drug-sensitive parental HCC 1806 cells (left) and Tx-R HCC 1806 cells (right). The cells were treated with DMSO (green), Dox (40 µM) alone (gray) or in combination with BGJ 398 (20 µM) (red). (B) The intracellular accumulation of Alexa-488 labeled PTX (Flutax-2) in drug-sensitive parental HCC 1806 cells (left) and Tx-R HCC 1806 cells (right). The cells were treated with DMSO (green), Flutax-2 (3 µM) alone (gray) or in combination with BGJ 398 (20 µM) (red). The fluorescence intensity was analyzed by FACs. Representative histograms of at least 3 independent experiments are shown.

Figure 4

BGJ 398 increases intracellular accumulation of Calcein, an ABCB1 fluorescent substrate, in ABCB1-overexpressing Tx-R HCC 1806 cells. The intracellular accumulation of Calcein, a fluorescent product of the ABCB1 substrate Calcein AM, was determined in the Tx-R HCC 1806 cells (B) and the corresponding drug-sensitive parental HCC 1806 cells (A). The cells indicated above were treated with 100 nM of Calcein AM alone (green) or in combination with 20 µM of BGJ 398 (red) or 50 µM PD 173074 (gray). The fluorescence intensity was analyzed by FACs. Representative hist grams of at least 3 independent experiments are shown.

Figure 5

BGJ 398 potentiates accumulation of Tx-R HCC 1806 cancer cells in M-phase after the treatment with PTX. Immunoblot analysis for phospho-NuMA (Ser395) and phospho-H3 (Ser10) in the parental (A) and Tx-R (B) HCC 1806 breast cancer cells after treatment with DMSO (negative control), BGJ 398, PTX alone or in combination for 48 h. Actin stain is used as a loading control. (C,D) Quantification by mean pixel density of pNuMA (Ser395) and pH3 (Ser10) in the parental (C) and Tx-R (D) HCC 1806 breast cancer cells. Values are means ± SD, n = 3. * p < 0.05 vs. untreated cells.

Figure 6

BGJ 398 potentiates apoptosis of Tx-R HCC 1806 cells treated with combination of PTX. Immunoblot analysis for apoptosis markers (cleaved forms of PARP and caspase-3) in naive (A) and Tx-R (B) HCC1 806 breast cancer cells after treatment with DMSO (negative control), PTX, BGJ 398 alone or in combination for 48 h. Actin stain is used as a loading control. (C,D) Quantification by mean pixel density in the cleaved forms of PARP and caspase-3 in naive (C) and Tx-R (D) HCC 1806 breast cancer cells. Values are means ± SD, n = 3. * p < 0.05, ** p < 0.01.

Figure 7

FACs analysis for apoptotic markers in Tx-R HCC 1806 breast cancer cells treated with DMSO (negative control), PTX (1 µM), BGJ 398 (1 µM) alone or in combination for 48 h. (A) Representative dot-plot are shown. (B) Quantitative analysis of the early apoptotic cells after the treatment as indicated above. (C) Quantitative analysis of the total apoptotic cells after the treatment as indicated above. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 8

The impact of FGFR inhibitors (BGJ 398 and PD 173074) used in combination with PTX or Dox on proliferation and survival of Tx-R HCC 1806 breast cancer cells; (A) Crystal violet staining of Tx-R HCC 1806 cells that were treated with BGJ 398 (or PD 173074) alone or in combination with PTX (or Dox) for 96 h. The cells treated with DMSO were used as a control. The cells treated with PTX or Dox alone illustrated chemoresistance of Tx-R HCC 1806 cells. The culture dishes were fixed with ice-cold 100% methanol, stained with crystal violet, and photographed; (B) Quantification of crystal violet staining of Tx-R HCC 1806 cells treated with PTX and Dox alone or in combination with BGJ 398. The plates were dried, crystal violet was dissolved using 0.1% SDS solution, and absorbance was measured at 590 nm. The graphs represent the mean ± SD. * p < 0.05; ** p < 0.01; (C) Quantification of crystal violet staining of Tx-R HCC 1806 cells treated with PTX and Dox alone or in combination with PD 173074.

Figure 9

Assessment of the synergy between PTX and BGJ 398 in Tx-R HCC 1806 breast cancer (A) and GIST T-1 (B) cells. The average synergy for Tx-R HCC 1806 cells was 11.12, for GIST T-1 was 28.22.

Figure 10

Expression of ABCB1 in Tx-R HCC 1806 (A) and GIST T-1 (B) cancer cells treated with DMSO (control) and BGJ398 for 72 h. Actin stain was used as a loading control. Actin stain was used as a loading control. (C,D) Quantification by mean pixel density of ABCB1 in Tx-R HCC 1806 cells (C) and GIST T-1 (D) cancer cells after treatment as indicated above. Values are means ± SD, n = 3. Statistically significant differences were not observed.

Figure 11

Computer modeling of BGJ 398 binding pose on DBP of ABCB1. (A) 3D diagram illustrating the proposed binding mode of the interactions of BGJ 398 with the ABCB1. (B) 2D Ligand interaction diagrams of the BGJ 398 with significant amino acid residues.