Gilteritinib inhibits the proliferation of lung cancer cells both in vitro and in vivo.
To explore the antitumor role of gilteritinib in lung cancer cells, we initially measured the viability of lung cancer cells treated with gilteritinib for 48 hours. Treatment with gilteritinib reduced the proliferation of lung cancer cells A549 and H1650, with a 50% inhibitory concentration (IC50) at 158 nM and 89 nM, respectively (Fig. 1A). Subsequently, we treated A549 and H1650 cells with the IC50 dosage for five days, and found that the cell proliferation was significantly inhibited by gilteritinib (Fig. 1B). To evaluate the antitumor activity of gilteritinib in vivo, we established a subcutaneous xenograft tumor model in BALB/c-nu mice with A549 cells and treated them with 1 or 5 mg/kg gilteritinib. The results showed that 1 mg/kg gilteritinib did not hinder the growth of lung cancer xenografts, and 5 mg/kg gilteritinib significantly suppressed tumor growth compared to the vehicle control (Fig. 1C-1E). Consistent with these findings, immunohistochemical analysis revealed that the number of PCNA-positive cells was significantly decreased in mice treated by 5 mg/kg gilteritinib compared with the control mice or mice with 1 mg/kg gilteritinib (Fig. 1F). Importantly, mice did not exhibit weight loss when treated with either dose of gilteritinib, implying that gilteritinib treatment did not produce noticeable general toxicity (Fig. 1G). Therefore, these results indicate that gilteritinib can effectively inhibit the proliferation of lung cancer cells both in vitro and in vivo.
Gilteritinib induces cholesterol accumulation in lung cancer cells by upregulating the expression of its biosynthetic genes and inhibiting ABCA1-mediated cholesterol efflux
To investigate the underlying mechanisms of antitumor role of gilteritinib in lung cancer, we performed RNA-Seq analysis of A549 cells with or without gilteritinib treatment. The RNA-Seq result revealed that 1744 genes were differentially expressed between in gilteritinib-treated A549 cells and the control cells, with 1282 genes upregulated and 462 genes downregulated (Fig. 2A). Interestingly, gene ontology (GO) analysis of the differentially expressed genes showed that sterol and cholesterol biosynthetic processes and their regulation were most highly enriched in gilteritinib-treated A549 cells (Fig. 2B). Cholesterol, the important sterol in mammals, is an essential component of cell and organelle membranes, which can be synthesized endogenously or acquired via uptake of low-density lipoproteins (LDLs) [22]. Cholesterol biosynthesis process includes at least 12 genes encoding enzymes that catalyze the conversion of acetyl-CoA into cholesterol (Fig. 2C). RNA-Seq data showed that these genes were significantly upregulated in gilteritinib-treated cells (Fig. 2D), with the exception of DHCR24. qRT-PCR confirmed that 100 nM gilteritinib treatment significantly increased the expressions of these genes in A549 and H1650 cells (Fig. 2E). Furthermore, western blotting also confirmed that the proteins HMGCS1, MVK, IDI1, FDFT1, SQLE and LSS were elevated in A549 and H1650 cells after exposed to gilteritinib (Fig. 2F). Therefore, we deduced from the above results that cholesterol levels should be elevated in gilteritinib-treated lung cancer cells. As expected, an obvious intracellular accumulation of free cholesterol was observed in A549 and H1650 cells treated with gilteritinib (100 nM) with filipin staining, whereas in the control cells, cholesterol was only distributed in the cell membrane (Fig. 2G). In addition, we also verified that the total cholesterol in gilteritinib-treated lung cancer cells was much higher than that in control cells using Micro TC Content Assay Kit (Fig. 2H). These results demonstrated that the upregulated enzymes for cholesterol biosynthesis process are responsible for the cholesterol accumulation in the gilteritinib-treated lung cancer cells.
On the other hand, studies show that cholesterol accumulation also can be caused by the reduced exportation out of cells. ATP-binding cassette subfamily A member 1 (ABCA1) is one of most important players in regulating cholesterol efflux. Our RNA-Seq analysis showed that the ABCA1 mRNA was decreased in gilteritinib-treated A549 cells (data not shown), which was confirmed by qRT-PCR in A549 and H1650 cells treated with gilteritinib (Fig. 3A). Furthermore, western blotting also indicated that gilteritinib treatment significantly reduced the protein expression of ABCA1 in A549 and H1650 cells (Fig. 3B). To investigate whether ABCA1 was involved in the intracellular cholesterol accumulation in gilteritinib-treated cells, we overexpressed ABCA1 protein in A549 and H1650 cells (Fig. 3C), and then observed the cholesterol with filipin staining in these cells. The result showed that gilteritinib-induced cholesterol accumulation was almost disappeared in lung cancer cells with overexpression of ABCA1 (Fig. 3D), suggesting that the cholesterol is discharged out of the cell by the overexpressed ABCA1. In addition to ABCA1, gilteritinib also suppressed mRNA expression of another important cholesterol efflux protein ATP-binding cassette subfamily G member 1 (ABCG1) (Fig. 3E); in KM plotter online database, lung cancer patients with high ABCG1 expression have longer survival time than those with low ABCG1 expression (Fig. 3F), suggesting that patients with highly expressed ABCG1 may have good response to chemotherapy. These results imply that ABCG1 might also be involved in cholesterol efflux. Together, these data suggest that gilteritinib induced cholesterol accumulation in lung cancer cells by enhancing its biosynthesis and inhibiting its efflux.
Accumulated cholesterol attenuates the antitumor activity of gilteritinib and induces gilteritinib-resistance in lung cancer cells
Next, we were curious about the impact of accumulated cholesterol on the antitumor activity of gilteritinib in lung cancer cells. Reports suggest that cholesterol can diminish the therapeutic effects of antineoplastic agents and is strongly associated with drug resistance in cancer [23, 24]. Therefore, we reasoned that although gilteritinib exerts an inhibitory effect on lung cancer cells, its induced cholesterol accumulation might also counteract its own antitumor activity or increase the resistance of lung cancer cells to itself, and conversely, lowering-cholesterol has the opposite role. Indeed, as shown in Fig. 4A, we found that the ABCA1-overexpressed lung cancer cells (A549 and H1650 cells) with extreme low cholesterol levels (which shown in Fig. 3D) were significantly more sensitive to gilteritinib than the control cells with high cholesterol (which shown in Fig. 2G-2H), indicating that the gilteritinib-induced cholesterol accumulation attenuates its antitumor activity and ABCA1-mediated low-cholesterol levels sensitizes lung cancer cells to gilteritinib. Clinically, the Kaplan–Meier (KM) plotter database (https://kmplot.com/analysis/) showed that lung cancer patients with high expression of ABCA1 mRNA exhibited a longer survival time (Fig. 4B). This implies that upregulated ABCA1 reduces cholesterol accumulation by facilitating cholesterol efflux from the cells, thereby enhancing the sensitivity of cancer cells to the antitumor drugs.
To further confirm the effect of the accumulated cholesterol due to increased biosynthesis on the antitumor activity of gilteritinib, we treated lung cancer cells with a cholesterol biosynthesis inhibitor and observed the cell viability. As shown in the above results (Fig. 2C-2F), Squalene epoxidase (SQLE), an important rate-limiting enzyme that catalyzes the first oxygenation step in cholesterol biosynthesis [25], is upregulated in gilteritinib-treated lung cancer cells. NB-598 is a known non-competitive inhibitor of SQLE [26]. Indeed, NB-598 (2.5µM) treatment significantly reduced gilteritinib-induced cholesterol accumulation in A549 cells (Fig. 4C). Although 2.5µM NB-598 alone show no significant antitumor activity, its combination with gilteritinib resulted in significantly stronger inhibitory role than gilteritinib alone in A549 and H1650 cells (Fig. 4D). Flow cytometry analysis indicated that combination of NB-598 and gilteritinib induced a synergistic enhancement of apoptosis in lung cancer cells (Fig. 4E). Consistently, MTT assay showed that A549 and H1650 cells treated with NB-598 were much more sensitive to gilteritinib than the control cells (Fig. 4F), in which IC50 dose of gilteritinib for the lung cancer cells treated with NB-598 was four times lower than that for the control cells. Clinically, lung cancer patients with high SQLE expression have significantly poorer survival rates than those with low SQLE expression (Fig. 4G), implying that high SQLE expression accelerates cholesterol synthesis and induces cholesterol accumulation, thereby leading to resistance to anticancer drugs and poor survival.
However, when we employed siRNA to suppress HMGCS1 (Supplementary figure A), an enzyme in the first step of cholesterol biosynthesis, it failed to prevent gilteritinib-induced cholesterol accumulation in A549 cells (Supplementary figure B). And we also found that simvastatin, an inhibitor of HMGCR, could not stop gilteritinib-induced cholesterol accumulation in A549 cells (Supplementary figure C). Accordingly, siRNA against HMGCS1 and simvastatin had no synergistic antitumor role with gilteritinib in A549 cells (Supplementary figure D). These results indicate that gilteritinib-induced cholesterol biosynthesis could bypass HMGCS1 and HMGCR catalytic steps or HMGCS1 and HMGCR are not the key rate-limiting enzyme for cholesterol biosynthesis.
Clinically, gilteritinib can rapidly produce drug resistance during the therapy for acute myeloid leukemia [27, 28]. Therefore, we wanted to learn whether the increased cholesterol was associated with gilteritinib resistance in lung cancer. To this end, we generated gilteritinib-resistant lung cancer cells from A549 and H1650 cells. With filipin staining method, we observed that both resistant cells contained higher levels of free cholesterol than their parental cells (Fig. 4H). Importantly, gilteritinib-resistant A549 and H1650 cells were more sensitive to the combination treatment of NB-598 (3 µM) and gilteritinib than the parental cells (Fig. 4I), suggesting that gilteritinib-resistant lung cancer cells can be reversed by cholesterol synthesis inhibitor NB-598. Altogether, the accumulated cholesterol also can induce gilteritinib-resistance in lung cancer cells.
25-Hydroxycholesterol suppresses gilteritinib-increased cholesterol in lung cancer cells via inhibiting cholesterol synthesis and accelerating cholesterol efflux
25-Hydroxycholesterol (25HC), a cholesterol metabolite, is an oxysterol derived from cholesterol by cholesterol 25-hydroxylase (CH25H). 25HC is an important regulator of cholesterol biosynthesis, uptake, efflux and storage [29], and also is known to be a strong feedback inhibitor of cholesterol synthesis [30]. To investigate whether or how 25HC can affect the cholesterol metabolic process in gilteritinib-treated lung cancer cells, we performed RNA-Seq of A549 cells treated by gilteritinib alone or in combination with 25HC. Compared to gilteritinib-treated cells, there were 167 upregulated genes and 206 downregulated genes in gilteritinib plus 25HC-treated A549 cells (Fig. 5A). GO analysis showed that cholesterol biosynthetic and metabolic processes and so forth were among the top highly enriched processes (Fig. 5B). To validate this, qRT-PCR results showed that the increased mRNA expression of cholesterol biosynthesis genes induced by gilteritinib was completely inhibited by 25HC in A549 and H1650 cells (Fig. 5C). In agreement with qRT-PCR data, western blots also revealed that the upregulated proteins induced by gilteritinib were completely suppressed by the addition of 25HC (Fig. 5D). Interestedly, 25HC also can markedly upregulate the expression of ABCA1 mRNA and significantly elevate the ABCA1 mRNA reduced by gilteritinib in A549 and H1650 cells (Fig. 5E). Furthermore, ABCA1 protein also is upregulated by 25HC treatment in the two lung cancer cells (Fig. 5D, top row). These results suggest that 25HC also can activate ABCA1 expression to accelerate cholesterol efflux from lung cancer cells. Accordingly, filipin staining showed that robust cholesterol accumulation induced by gilteritinib treatment in A549 cells was substantially reduced by addition of 25HC (Fig. 5F). These data indicate that 25HC can inhibit gilteritinib-increased cholesterol in lung cancer cells by inhibiting cholesterol biosynthesis and enhancing cholesterol efflux.
25HC increases the sensitivity of lung cancer cells to gilteritinib in vitro and in vivo
Next, we wanted to learn whether 25HC-mediated cholesterol reduction can increase the cytotoxicity of gilteritinib in lung cancer cells. The result showed that though 25HC (3 µM) treatment alone had no effect on the proliferation of lung cancer cells, it noticeably increased the cytotoxicity of gilteritinib in both A549 and H1650 cells compared with the cells with gilteritinib treatment alone (Fig. 6A). To confirm this, we also analyzed the IC50 of gilteritinib with or without 25HC by MTT assay. The IC50 of gilteritinib plus 25HC in A549 and H1650 cells was over 5-fold lower than that of gilteritinib alone in the same cells (Fig. 6B), indicating that the drug combination resulted in a synergetic antitumor activity. To further determine whether the synergy of 25HC and gilteritinib was attributable to the loss of cholesterol, we treated lung cancer cells with exogenous cholesterol. Indeed, the addition of 10 µg/ml exogenous cholesterol to the culture medium caused obvious increase of cholesterol content in A549 cells with or without gilteritinib plus 25HC, compared to the control cells (Fig. 6C). Importantly, in the presence of exogenous cholesterol, 25HC failed to sensibilize A549 and H1650 cells to gilteritinib (Fig. 6D), implying that the antitumor efficacy of combination of gilteritinib and 25HC mostly depends on intracellular cholesterol level. Consistent with this, gilteritinib combined with 25HC caused pronounced apoptosis in A549 and H1650 cells, and exogenous supplement of cholesterol efficiently reduced the pro-apoptotic activity of gilteritinib in combination with 25HC (Fig. 6E). These results suggest that 25HC significantly sensibilize lung cancer cells to gilteritinib by lowing cellular cholesterol, and this combination can be a potential effective therapy for lung cancer.
Finally, we sought to determine whether gilteritinib in combination with 25HC had therapeutic activity in vivo. A549 cells were subcutaneously injected into one flank of mice to establish a lung cancer xenograft model. Mice were administrated with gilteritinib alone, or 25HC alone, or in combination. Compared to the treatment with gilteritinib or 25HC alone, the combination of gilteritinib and 25HC resulted in a significant reduction in tumor growth (Fig. 7A). The tumor size in the mice treated with the combination were significantly smaller than those in ones treated by either monotherapy (Figs. 7B and 7C). Immunohistochemical staining of tumors showed that 25HC had no inhibitory effect on proliferation of lung cancer (PCNA, whereas 25HC plus gilteritinib reduced the number of PCNA-positive cells (Fig. 7D). Importantly, gilteritinib or combined with 25HC had no obvious toxicity on mice, as evidenced by the lack of significant changes in mouse body weight (Fig. 7E). Given that gilteritinib can regulate lipid metabolism, we evaluate whether gilteritinib affects mouse liver using H&E staining. Compared to the control group, the liver tissue of mice in all treatment groups showed no obvious steatosis (Fig. 7F). Taken together, these data suggest that gilteritinib in combination with 25HC inhibits tumor growth more effectively than gilteritinib alone with safety profile.