The antitumour activity of 2‐(4‐amino‐3‐methylphenyl)‐5‐fluorobenzothiazole in human gastric cancer models is mediated by AhR signalling

Abstract Stomach cancer is the fourth most common cancer worldwide. Identification of novel molecular therapeutic targets and development of novel treatments are critical. Against a panel of gastric carcinoma cell lines, the activity of 2‐(4‐amino‐3‐methylphenyl)‐5‐fluorobenzothiazole (5F 203) was investigated. Adopting RT‐PCR, Western blot and immunohistochemical techniques, we sought to determine molecular pharmacodynamic (PD) markers of sensitivity and investigate arylhydrocarbon (AhR) receptor‐mediated signal transduction activation by 5F 203. Potent (IC50 ≤ 0.09 μmol/L), selective (>250‐fold) in vitro antitumour activity was observed in MKN‐45 and AGS carcinoma cells. Exposure of MKN‐45 cells to 5F 203 triggered cytosolic AhR translocation to nuclei, inducing CYP1A1 (>50‐fold) and CYP2W1 (~20‐fold) transcription and protein (CYP1A1 and CYP2W1) expression. G2/M arrest and γH2AX expression preceded apoptosis, evidenced by PARP cleavage. In vivo, significant (P < .01) 5F 203 efficacy was observed against MKN‐45 and AGS xenografts. In mice‐bearing 5F 203‐sensitive MKN‐45 and 5F 203‐insensitive BGC‐823 tumours in opposite flanks, CYP1A1, CYP2W1 and γH2AX protein in MKN‐45 tumours only following treatment of mice with 5F 203 (5 mg/kg) revealed PD biomarkers of sensitivity. 5F 203 evokes potent, selective antitumour activity in vitro and in vivo in human gastric cancer models. It triggers AhR signal transduction, CYP‐catalysed bioactivation to electrophilic species causing lethal DNA double‐strand breaks exclusively in sensitive cells. 5F 203 represents a novel therapeutic agent with a mechanism of action distinct from current clinical drugs, exploiting novel molecular targets pertinent to gastric tumourigenesis: AhR, CYP1A1 and CYP2W1. PD markers of 5F 203 sensitivity that could guide patient selection have been identified.


| INTRODUC TI ON
Stomach cancer is the fourth most common cancer and second leading cause of cancer death worldwide. With >950 000 new diagnoses each year and ~720 000 deaths in 2012, gastric cancer represents 7% of global cancer incidence. 1 Although worldwide, the incidence of gastric cancer is declining, it remains highly prevalent in Asia when compared to the West. China is one of the countries with the highest incidence of gastric cancer and accounts for >40% of all new gastric cancer cases in the world. Indeed, gastric cancer is the third leading cause of cancer mortality in China. Globally, 5-year survival for gastric cancer is ~30%; for localized disease, 67% of patients survive beyond 5 years, but >70% gastric cancer patients possess metastases (in regional lymph nodes or at distant sites) at the time of  We conclude that stomach cancers whose tumour cells express cytosolic AhR are likely to be sensitive to 5F 203 and that 5F 203 represents a putative novel therapeutic agent in treatment of certain gastric cancers.

| MTT assay
Cells were seeded into 96-well plates at a density of 4 × 10 3 per well and allowed to attach overnight. Dilutions of 5F 203 (0.01-50 μmol/L) were prepared in culture medium from 10 mmol/L stock solutions and following 3 days of incubation (37°C, 5% CO 2 ).

| Cell cycle analysis
Exponentially growing cells were harvested and seeded in 6-well plates (2 × 10 5 cells/well; 2 mL medium). Cells were incubated cytometer, and data were analysed using C6 software.

| RT-PCR
Human gastric cancer MKN-45 cells were treated with 1 μmol/L 5F 203 for 3, 6 and 12 hours. Total RNA was extracted by using

| Intracellular distribution of AhR in gastric cancer cell lines
To investigate intracellular distribution of AhR, cell lysates were prepared following extraction using cytosolic and nuclear fraction kits, and subjected to Western blot. A substantial fraction of AhR was detected in the cytoplasmic fractions of KATO III and MKN-45 lysates ( Figure 1). In contrast, low levels of AhR were present in the cytoplasm of NCI-N87 and BGC-823 cells, and very faint AhR expression was detected in SGC-7901 cells. Furthermore, the AhR localization in the cytoplasm was confirmed by immunofluorescence ( Figure 2B).
Interestingly, high levels of AhR were present in nuclei of NCI-87 cells.

| 5F 203 induced AhR translocation and CYP1A1, CYP2W1 expression in MKN-45 gastric cancer cell line
To assess whether 5F 203 could activate the AhR signal transduction pathway, AhR translocation from cytoplasm to nucleus was monitored in MKN-45 cells by immunofluorescence microscopy. In control cells

| 5F 203 caused G2/M arrest, DNA damage and apoptosis in MKN-45 gastric cancer cells
To investigate perturbations in cell cycle distribution after treatment of MKN-45 cells with 5F 203, cells were treated with 5F 203 (1 μmol/L) for 3, 6, 12, 24 and 48 hours and subsequently processed for cell cycle analyses. As illustrated ( Figure 5A,B) Figure 5C) treatment in the sensitive cell lines, indicating the presence of DNA damage in these cells. Immunoblot analysis showed that PARP cleavage occurred ≥48 hours after 5F 203 (1 μmol/L) treatment ( Figure 5D).

| 5F 203 suppressed gastric tumour growth in vivo
Independently conducted efficacy studies confirmed that 5F 203 inhibits gastric tumour growth in vivo. MKN-45 and AGS xenograft growth was significantly inhibited by 5F 203 (2.5 mg/kg or 5 mg/kg administered iv on 5 consecutive days; P < .01; see SI). 5F 203 (5 mg/ kg) inhibited tumour growth >50% as assessed by measurement of tumour volumes throughout the experiment and tumour weight at termination of the experiment ( Figure 6A and Figure S1). To investigate the selective nature of gastric tumour growth inhibition caused by 5F 203 in vivo, MKN-45 and BGC-823 xenografts were transplanted in opposite flanks of the same mouse. As shown in Figure 6B,C, the growth of MKN-45 growth was significantly suppressed after 5F 203 administration ip (5 mg/kg, 10 mg/kg), and the relative tumour volume in 5F 203 (5 mg/kg, 10 mg/kg) groups reduced by 72.55% and 79.30%, respectively, compared with the control group (P < .001). However, 5F 203 (5 mg/kg, 10 mg/kg) did not inhibit BGC-823 tumour growth.
The bodyweights of mice remained unchanged ( Figure 6D), and no other observable side-effects were detected (behaviour, appetite).
Significant induction of CYP1A1 and CYP2W1 protein and mRNA expressions was demonstrated ( Figure 6E,F)  and CYP2W1 protein and mRNA expression were neither constitutive nor induced by treatment of mice with 5 mg/kg 5F 203.
Expression of AhR and CYP1A1 protein in MKN-45 tumour tissue was analysed by immunohistochemistry. We found that AhR expression in vehicle-treated mice was mainly in the cytoplasm, and AhR translocated into the nucleus after treatment with 5F 203.
Furthermore, CYP1A1 protein was induced in the tumour tissues of mice after 5F 203 treatment ( Figure 6G). Lung and liver tissues were recovered from animals 24 hours after treatment with 5F 203, and H&E staining revealed minor swelling of cells in the liver ( Figure 6H).

| D ISCUSS I ON
In this communication, the potent and selective antitumour activity of 5F 203 (already described in breast, ovarian and renal cancer models) against gastric cancer cell lines MKN-45 and AGS (IC 50 value <100 nmol/L) has been revealed. In contrast, IC 50 values >50 μmol/L were obtained in SGC-7901, BCG-823 and KATO III cell lines, revealing stark (>500-fold) selectivity (Table 1) 203) activity in KM12 and HCC 2998 CRC cell lines. 22 It was thereafter irrefutably demonstrated that CYP1A1 and CYP2W1 are able to catalyse production of 5F 203-derived hydroxylamine, and guanine DNA adducts presumably via nitrenium species production, underpinning the selective antitumour cytotoxic activity of 5F 203. 6 5F 203 treatment consequently resulted in lethal DNA adducts, double-strand breaks and selective cancer cell death. 8 Expression of CYP2S1 mRNA, whose protein product is able to catalyse conversion of the 5F 203-derived hydroxylamine back to its parent amine 6 and may diminish its antitumour activity, was negligibly affected following exposure of MKN-45 cells to 5F 203.
Thus, PD biomarkers of sensitivity to/activity of 5F 203 activity in in vitro, in vivo and ex vivo breast ovarian, renal and colon cancer models include cytosolic AhR expression and translocation, 11   Analyses of MKN-45 tumours excised from mice exposed to 5F 203 (or vehicle) exposed in vivo PD biomarkers of sensitivity/ activity (AhR expression, induction of CYP1A1, CYP2W1 mRNA and protein expression, γH2AX) in gastric cancer corroborating a role for AhR signal transduction in the mechanism of action of 5F 203.
Thus, for sensitivity to antitumour benzothiazoles to manifest, the combination of cytosolic AhR and inducible CYP1A1 or CYP2W1 expression seems to be required.
Initial developmental expression and re-expression in certain tumours of CYP 2W1 implicate roles in growth and tumorigenesis.
Indeed, CYP2W1 is known to be a poor prognostic marker in CRC and expression in metastatic disease is enhanced compared with primary malignant sites. 25,26 Tumour-specific expression of CYP2W1 infers potential as a putative molecular drug target, able to bioactivate anticancer prodrug candidates at the tumour site (minimizing systemic exposure to active species). Antitumour benzothiazoles, for example 5F 203, are bioactivated by CYP2W1 and CYP1A1 6producing dGuo adducts responsible for cytotoxicity. Discovery of CYP 2W1 expression in gastric and hepatocellular carcinomas extends the spectrum of cancer phenotypes potentially demonstrating sensitivity to potent, selective antitumour benzothiazoles.
The demonstration herein of 5F 203 activity against gastric cancer models expressing cytosolic AhR and inducible CYP1A1 and

DATA AVA I L A B I L I T Y S TAT E M E N T
Data generated by research reported herein will be made openly and publicly available upon publication. This is part 36 in the series 'Antitumour Benzothiazoles'.