The effects of crizotinib in a transgenic Drosophila model expressing the human TPM4-ALK fusion gene or TPM4

ABSTRACT Anaplastic lymphoma kinase (ALK) fusion events lead to constitutive activation of the ALK kinase domain, thereby functioning as oncogenic drivers. These fusion proteins have been identified in numerous cancers. Crizotinib, a small molecule inhibitor of c-Met and ALK, is a Food and Drug Administration-approved drug with reported efficacy in the treatment of cancer. Tropomyosins (TPMs) are a family of actin filament-binding proteins. Altered TPM expression has been found in a variety of human tumors. Inhibitors of cancer-associated TPMs and actin-targeting compounds have been developed, but anti-actin agents have cardiac and respiratory muscle toxicities. In this study, we investigated the sensitivities of human TPM4 (hTPM4), human ALK (hALK), and their fusion gene (hTPM4-hALK) to crizotinib by measuring the lifespan of transgenic Drosophila. Flies overexpressing hTPM4-hALK, hTPM4 and hALK showed decreased lifespans compared with controls. Although crizotinib is an inhibitor of ALK, treatment with crizotinib significantly extended the lifespans of Drosophila expressing hTPM4 and hTPM4-hALK but had no effect on hALK-expressing flies. Autophosphorylation of Tyr1278 is necessary for full activation of the ALK domain. We confirmed that hTPM4-hALK was phosphorylated at Tyr1278 in a ligand-independent manner, and hTPM4-hALK-expressing flies treated with crizotinib showed a decreased level of Tyr1278 phosphorylation compared with untreated hTPM4-hALK-expressing flies, with a greater decrease induced by 1 µM compared with 200 nM crizotinib. Taken together, the results suggest that crizotinib is effective for treating ALK-driven cancer and might be a new therapeutic drug, without cardiac or respiratory muscle toxic effects, for TPM4-expressing cancers.


INTRODUCTION
Gastric cancer is one of the most common cancers and is the second leading cause of cancer-related death worldwide. A variety of genetic alterations in TP53, PIK3CA and KRAS have been identified, and many molecular targeted drugs have been evaluated for treatment of gastric cancer (Yang et al., 2014). Here, we assessed the effects of crizotinib using a Drosophila model. Because of its relatively short lifespan (approximately 2 months) and conservation of human disease-related genes, Drosophila melanogaster has been widely used as an in vivo tool for validation or discovery of effective cancer drugs (Sun et al., 2013;Yadav et al., 2016). Recent advances in molecular genomics have identified various oncogenic gene alteration, one of the most important types being the fusion of two genes (Parker and Zhang, 2013). Among various fusion genes involved in cancers, anaplastic lymphoma kinase (ALK) has been found fused to various genes in diverse cancers (Hallberg and Palmer, 2013). ALK is a receptor tyrosine kinase that was first identified as a fusion gene with nucleophosmin 1 (NPM1-ALK), via a 2;5 chromosomal translocation in anaplastic large-cell lymphoma (Morris et al., 1995). ALK functions as an oncogenic driver upon its gene amplification, introduction of point mutations and formation of fusion genes. A variety of ALK fusion genes resulting in constitutive activation of ALK, which is normally regulated by its extracellular ligand-binding domain in the fulllength molecule, have been identified in human cancers (Grande et al., 2011;Hallberg and Palmer, 2013). Tropomyosins (TPMs) constitute a family of actin filament-binding proteins that were first identified in studies of skeletal muscle contraction. Human cells contain four tropomyosins: TPM1, TPM2, TPM3 and TPM4. Altered tropomyosin expression has been found in a variety of human cancers, including breast, colon, lung, prostate, ovary and gastric cancers (Stehn et al., 2006;Choi et al., 2012). TPM3-ALK and TPM4-ALK, which are fusion proteins consisting of the N-terminus of TPM and the C-terminal kinase domain of ALK, have been reported in patients with inflammatory myofibroblastic tumors, with TPM3-ALK being the most frequently observed (Lawrence et al., 2000). Crizotinib is a small-molecule selective inhibitor of c-Met and ALK tyrosine kinases. It was approved by the Food and Drug Administration for the treatment of ALK-positive non-small-cell lung cancer and inflammatory myofibroblastic tumors (Butrynski et al., 2010;O'Bryant et al., 2013). In a previous study, we identified the TPM4-ALK fusion gene in gastric cancer tissues of Korean patients (unpublished). In the present study, we evaluated the sensitivity of the TPM4-ALK fusion gene to crizotinib, using transgenic Drosophila expressing the human TPM4 (hTPM4)-human ALK (hALK) fusion gene. We also assessed the sensitivities of the full-length hALK and hTPM4 genes to crizotinib in transgenic Drosophila. The results suggested that the hTPM4-hALK fusion gene is constitutively activated, resulting in early death of Drosophila. We also found that the activity of hTPM4-hALK was inhibited by treatment with crizotinib.

Molecular characteristics of hTPM4, hALK and hTPM4-hALK and their and ectopic expression in Drosophila
In previous studies we characterized fusion proteins in the tumor tissues of gastric cancer patients. Among these selected fusion proteins, we found that the TPM4-ALK fusion protein is preferentially expressed in gastric cancer tissues compared with normal tissues. We subsequently identified the amino acid sequence of the TPM4-ALK fusion protein (unpublished), which showed that exon 8 of TPM4 is fused to exon 19 of ALK (Fig. 1A). To investigate whether the lifespan of Drosophila is sensitive to crizotinib as mediated by hTPM4-hALK, we generated transgenic Drosophila expressing hTPM4, hALK and hTPM4-hALK. Expression of these genes in Drosophila was confirmed by realtime reverse-transcription polymerase chain reaction (qRT-PCR) and western blot analysis. As shown in Fig. 1, the mRNA expression of hALK, hTPM4-hALK and hTPM4 was observed in two or more independent transgenic lines for each gene construct. Protein expression was confirmed by western blot analysis using antibodies specific to ALK and TPM4 (Fig. 1E,F).
Expression of hALK, hTPM4 and hTPM4-hALK decreased the lifespan of transgenic Drosophila We evaluated whether overexpression of hTPM4, hALK or hTPM4-hALK had an effect on lifespan in Drosophila. Survival analysis using the Kaplan-Meier method showed that the lifespan of flies expressing hTPM4-hALK was shorter than that of w 1118 control flies (hazard ratio [HR]: 17.11; P<0.001). The lifespan of flies expressing hALK (HR: 3.70; P<0.001) or hTPM4 (HR: 1.83; P<0.001) was significantly decreased compared with that of control flies (Fig. 2, Table 1). The mean lifespan of the flies expressing hTPM4-hALK was 15.24±0.85 days compared with 37.56±0.87 days for control flies. The maximum lifespan of the hTPM4-hALK transgenic flies was 24 days compared with 45 days for control flies. The mean lifespan of the hTPM4-and hALK-expressing flies was 27.36±1.20 and 25.98±0.82 days, respectively (Table 2).  Table 1). The mean lifespan of flies expressing hTPM4-hALK treated with 200 nM or 1 µM crizotinib was 18.06±1.31 or 19.98±1.25 days, respectively, compared with 15.24±0.85 days for the untreated hTPM4-hALK-expressing flies. The maximum lifespan of flies expressing hTPM4-hALK treated with 200 nM or 1 µM crizotinib was 33 or 30 days, respectively, compared with 24 days for the untreated hTPM4-hALK-expressing flies (Table 2). Taken together, the results suggested that crizotinib has an anti-cancer effect in transgenic Drosophila expressing hTPM4-hALK, as evidenced by an extension of lifespan.

Crizotinib failed to increase the lifespan of transgenic Drosophila expressing hALK
We investigated the effect of crizotinib on the lifespan of hALK transgenic flies expressing full-length wild-type hALK. Despite being an inhibitor of ALK, treatment with 200 nM or 1 µM crizotinib had no effect on the lifespan of these flies (HR: 0.969; P=0.877; HR: 0.921; P=0.680) (Fig. 3D, Tables 1 and 2).
Crizotinib inhibits the constitutive activity of hTPM4-hALK in Drosophila TPM3-ALK and TPM4-ALK are constitutively active oncogenes acting independently of ligands and tyrosine phosphorylation (Lawrence et al., 2000). Autophosphorylation of Tyr 1278 is necessary for full activation of the ALK domain, as well as the transformation ability of ALK fusion proteins (Tartari et al., 2008). Therefore, we measured the level of phosphorylation at Tyr 1278 within the hTPM4-hALK fusion protein in transgenic Drosophila treated with crizotinib. As shown in Fig. 4A, hTPM4-hALK was phosphorylated at Tyr 1278 in a ligand-independent manner, and flies treated with crizotinib exhibited a decreased level of phosphorylated Tyr 1278 within hTPM4-hALK compared with untreated flies, and this effect was more pronounced with 1 µM compared with 200 nM crizotinib (Fig. 4A,C). These results were consistent with the prolonged survival of hTPM4-hALK-expressing flies treated with crizotinib. Taken together, the results suggest that crizotinib increased the lifespan of Drosophila by inhibiting the constitutive activity of hTPM4-hALK.

DISCUSSION
Numerous genes and pathways associated with cancer have been identified, and molecular-targeted therapies have been developed    using various model systems (Yadav et al., 2016;Geng et al., 2017). We analyzed the effects of crizotinib using Drosophila because of the numerous advantages of this model organism (Sun et al., 2013;Yadav et al., 2016). Crizotinib was originally developed as a selective c-MET inhibitor by Pfizer. However, after testing on a large panel of 120 kinases, it was found that crizotinib also potently inhibits ALK and induces apoptosis in anaplastic large-cell lymphoma (Christensen et al., 2007;Zou et al., 2007). Previously, we identified the TPM4-ALK fusion protein in patients with gastric cancer (unpublished). To characterize the influence of ALK overexpression on the effect of crizotinib, we generated transgenic Drosophila overexpressing hALK, hTPM4, or hTPM4-hALK and analyzed the lifespan of these flies following crizotinib treatment.
Crizotinib dose-dependently extended the lifespan of hTPM4-hALK-overexpressing flies. However, there was little effect on hALK-expressing flies, despite crizotinib being an inhibitor of ALK. These results were similar to those of Fang et al. (2014), who investigated the effect of crizotinib in a patient-derived xenograft model. Tumors harboring the HIP1-ALK fusion gene sensitively responded to crizotinib, whereas tumors harboring wild-type ALK did not. Because crizotinib inhibits ALK phosphorylation and cell proliferation (Ou, 2011), we investigated the level of phosphorylated (Tyr 1278 ) ALK using western blot analysis. Wildtype hALK was only slightly phosphorylated because of insufficient ligand stimulation, whereas hTPM4-hALK was phosphorylated in a ligand-independent manner (Fig. 4A,B). Furthermore, the increased survival of hTPM4-hALK-expressing flies was correlated with the decreased level of Tyr 1278 phosphorylation (Figs 3B and 4A). Taken together, our data suggest that wild-type hALK is resistant to crizotinib treatment.
We also generated hTPM4-expressing transgenic Drosophila as drug treatment controls. The survival of hTPM4-expressing flies was decreased compared with w 1118 flies. Unexpectedly, although crizotinib is an inhibitor of ALK, the survival of hTPM4-expressing flies was rescued by crizotinib. The crizotinib-induced increase in survival was similar to that in w 1118 flies and greater than that in untreated hTPM4-expressing flies (Fig. 3C, Table 1). Taken together, it is possible that crizotinib acts via hTPM4. Both decreased and increased TPM4 expression has been observed in human cancers, depending on the tumor type (Stehn et al., 2006). Inhibitors of cancerassociated TPMs, and actin-targeting compounds, have been developed to disrupt the actin cytoskeleton and actin polymerization in tumor cells (Bonello et al., 2009;Stehn et al., 2013). However, anti-actin drugs induce cardiac and respiratory muscle toxicities (Schweikart et al., 2013). Thus, novel TPM inhibitors that specifically target actin have been developed (Bonello et al., 2016;Currier et al., 2017). Further studies will be required to determine the mechanism by which crizotinib affects TPM4, to support the value of crizotinib as a therapeutic agent without cardiac and respiratory muscle toxicities. In summary, this study is the first to assess the effects of crizotinib on the lifespan of Drosophila overexpressing hTPM4-hALK, hALK or hTPM4. Our results showed that crizotinib may be an effective treatment for cancers overexpressing hTPM4-hALK, and hTPM4 may be a crizotinib target gene.

Genetics and Drosophila growth conditions
The hTPM4, hALK and hTPM4-hALK genes were cloned into the pUAST germline transformation vectors (Rubin and Spradling, 1982), which were then co-injected with helper plasmids encoding the p-element into Drosophila embryos to create transgenic Drosophila. The transformants were selected and further characterized. The GAL4/UAS system was used to overexpress the target gene in tissues and cells, and Tubuline-GAL4 driver flies were used to express target genes. The flies were cultured on standard agar media at 25±1°C under 60±1% relative humidity.

Fly food formulation
All experimental files were reared on a standard Drosophila medium consisting of 1 l dH 2 O, 7.7 g agar, 62.4 g dried yeast, 40.8 g corn starch, 84.0 g glucose, 13.0 ml molasses and 12.5 ml mold inhibitor (DaeJung Chemicals & Metals, Seoul, Republic of Korea). Crizotinib (PF-02341066) was added to the feed at final concentrations of 200 nM or 1 μM.

Western blot analysis
Cell lysates were obtained from approximately ten male or female flies by grinding in 300 μl RIPA buffer supplemented with proteinase and phosphatase inhibitors. The lysates were centrifuged at 13,000 rpm for 20 min at 4°C, and protein concentrations were determined using the BCA method.
Samples containing 30-50 μg total protein were resolved on a 10% SDS-PAGE gel and transferred to a nitrocellulose membrane. The membrane was blocked with 5% skim milk in 1× TBS-T and incubated with the following primary antibodies: anti-TPM4 (1:100; Developmental Studies Hybridoma Bank, Iowa City, IA, USA), anti-ALK (1:250; Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-pALK (1:1000; Cell Signaling Technology, Danvers, MA, USA), and anti-α-tubulin (1:5000; Developmental Studies Hybridoma Bank). The membrane was incubated at room temperature with secondary antibodies at the appropriate dilutions. Quantitation of the western blot results is represented graphically using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

qRT-PCR analysis
Total RNA was extracted from five male or female flies using TRIzol reagent (Ambion, Waltham, MA, USA), and 1 μg total RNA was used for cDNA synthesis using the PrimeScript™ RT Reagent Kit (Takara Bio, Kusatsu, Japan). Each reaction mixture contained 5 μl SYBR Green PCR Master Mix (Takara Bio), 0.4 μl forward and reverse primers (10 pmol each), cDNA (1:6 dilution), and diethyl pyrocarbonate-water to reach a final volume of 10 μl. The mRNA levels of the target genes were normalized to that of the Drosophila rp49 gene and were analyzed using the 2 -ΔΔCT method with Excel software (Microsoft, Redmond, WA, USA). The reactions were performed in duplicate. The following primers were used for PCR: 5′-TTG AGG AGG AGT TGG ACA GG-3′ and 5′-GCT GCA TCT CCT GAA TCT CC-3′ for hTPM4, 5′-GCA ACA TCA GCC TGA AGA CA-3′ and 5′-GCC TGT TGA GAG ACC AGG AG-3′ for hALK, and 5′-ACG GTT GCA AAA CTG GAA AA-3′ and 5′-TTG GGG TTG TAG TCG GTC AT-3′ for hTPM4-hALK.

Drug treatments and lifespan measurements
The method of measuring lifespan to determine the effects of genetic and non-genetic factors was described previously (Sun et al., 2013). A total of 100 flies (50 male and 50 female) were collected at 3 days after eclosion (ten flies per vial). The flies were kept at 31±1°C for survival analysis. The surviving flies were transferred every 3 days to new vials containing the appropriate normal food or food spiked with crizotinib (200 nM or 1 µM).
This practice was continued until all flies were dead. HRs and P-values were calculated using Kaplan-Meier survival analysis (Dinse and Lagakos, 1982). The HRs with 95% confidence intervals were compared using the log-rank test, and P-values were calculated (Mantel, 1966). The survival curves and statistical analyses were performed using the R program package (www.rproject.org) as described previously (Kim et al., 2015). The maximum lifespan was calculated as the average age of the top 10% longest-living flies (Lushchak et al., 2012).