Anti-tumor efficacy of Selinexor (KPT-330) in gastric cancer is dependent on nuclear accumulation of p53 tumor suppressor

Exportin-1 (XPO1) controls the nucleo-cytoplasmic trafficking of several key growth regulatory and tumor suppressor proteins. Nuclear export blockade through XPO1 inhibition is a target for therapeutic inhibition in many cancers. Studies have suggested XPO1 upregulation as an indicator of poor prognosis in gastric cancer. In the current study, we investigated the anti-tumor efficacy of selective inhibitors of nuclear export (SINE) compounds KPT-185, KTP-276 and clinical stage selinexor (KPT-330) in gastric cancer. XPO1 was found to be overexpressed in gastric cancer as compared to adjacent normal tissues and was correlated with poor survival outcomes. Among the 3 SINE compounds, in vitro targeting of XPO1 with selinexor resulted in greatest potency with significant anti-proliferative effects at nano molar concentrations. XPO1 inhibition by selinexor resulted in nuclear accumulation of p53, causing cell cycle arrest and apoptosis. Also, inhibition of XPO1 lead to the cytoplasmic retention of p21 and suppression of survivin. Orally administered selienxor caused significant inhibition of tumor growth in xenograft models of gastric cancer. Furthermore, combination of selinexor with irinotecan exhibited greater anti-tumor effect compared to individual treatment. Taken together, our study underscores the therapeutic utility of XPO1 targeting in gastric cancer and suggests the potential benefits of XPO1 inhibition in-combination with chemotherapy.

Leptomycin B was the first XPO1 inhibitor shown to efficiently inhibit nuclear export in various cancer cell lines in vitro 10 . However, its clinical development was subsequently limited due to severe toxicities without significant efficacy in human phase I trial 11 . Selective inhibitor of nuclear export (SINE) compounds (KPT-185,  were developed based on the fact that Leptomycin B specifically binds to the Cys528 residue in the cargo-binding groove of XPO1 to restore the function of TSPs in the nucleus, leading them to induce cancer-specific apoptosis 7 . SINE compounds have shown anti-tumor activity in several solid and hematological malignancies in preclinical in vitro and/or in vivo studies, including renal cancer 12 , pancreatic cancer 13 , prostate cancer 14 , breast cancer 15 , melanoma 16 , multiple myeloma 17 , mantle cell lymphoma 18 , chronic lymphocytic leukemia 7 , acute myeloid leukemia 19,20 and non-Hodgkin lymphoma 21 . Further to the Phase1 clinical trials of oral KPT-330 (selinexor, Karyopharm Therapeutics) for both solid tumors (NCT01607905, NCT01896505) and advanced hematologic malignancies (NCT01607892), the compound is currently undergoing Phase 2/3 trials. In addition, given the noted synergistic effects on cytotoxicity in neoplastic cells, the potential for combining XPO1 inhibition with conventional chemotherapy, such as topoisomerase inhibitors, is also being explored in recent clinical trials (NCT02283359). A recently concluded phase II trial that tested selinexor in combination with dexamethasone showed an overall response rate (ORR) of 21% in patients with heavily pretreated, refractory myeloma 22 . In the present study, we investigated the therapeutic efficiency of XPO1 inhibition with SINE compounds KPT-185, KPT-276 and KPT-330 in gastric cancer. The mechanism of XPO1 inhibitor-mediated proliferation inhibition and apoptosis in gastric cancer cells is also determined. Furthermore, the anti-tumor efficacy of KPT-330 is also characterized using a xenograft model of gastric cancer.

Methods
Patient tissue samples and genomic profiling. An independent patient cohort including 153 tumors and 100 normal samples were profiled for gene expression using Affymetrix human genome U133 Plus 2.0 GeneChips (Affymetrix, Santa Clara, CA). Gene expression microarray data is available under GEO accession number GSE15460. All primary gastric tissues were obtained from the National University Hospital, Singapore or National Cancer Centre, Singapore tissue repositories with approvals from the Research Ethics Review Committee, National University Hospital and Institutional Review Board, National Cancer Centre. All methods were performed in accordance with the relevant guidelines and regulations and signed patient informed consent. Tumor samples were histologically confirmed to contain cancer cells, with an average tumor cellularity of 40%. Non-malignant samples (normal tissue) were harvested from stomach tissue distant from the tumor and exhibiting no visible evidence of tumor or intestinal metaplasia/dysplasia upon surgical assessment. Histopathological data and patient characteristics of the cohort is provided in a previous publication 23 . Cell lines and reagents. Gastric cancer cell lines, AGS and NCI-N87 were purchased from ATCC, USA. IM95, NUGC-3 and NUGC-4 were purchased from Health Science Research Resources Bank, Japan. MKN1, MKN45, TMK1 and YCC10 were obtained from DUKE NUS, Singapore. HGC-27 was purchased from Public Health England, UK. All cell lines except YCC10 cell lines were cultured in RPMI 1640 medium (Gibco; Grand Island, NY) containing 10% heat inactivated fetal bovine serum (Gibco; Grand Island, NY) and 1% penicillin/ streptomycin (Gibco; Grand Island, NY). YCC10 cell lines was cultured in MEM supplemented with 20% heat inactivated fetal bovine serum (Gibco; Grand Island, NY), 1% penicillin/streptomycin (Gibco; Grand Island, NY) and 1% sodium pyruvate (Gibco; Grand Island, NY). All cell lines are maintained at 37 °C in humidified atmosphere containing 5% CO 2 .
Quantitative PCR (qPCR) analysis. Total RNA was extracted from cultured cells with RNeasy Mini kit (Qiagen, CA, USA) with the use of QIAshredder spin column for homogenization and an on-column DNase digestion. 2 µg of the total RNA was reversely transcribed using M-MLV reverse transcriptase enzyme (Promega, WI, USA). The cDNA obtained was analysed quantitatively using Power SYBR Green PCR Master Mix (Applied Biosystems, CA, USA) on an ABI7300 Real-time PCR system. Primers used are listed in Table 1. served as blank controls. Each drug was tested in triplicate. Cells were incubated for 72 hours before the addition of the assay reagent (1 mg/mL per well) and absorbance was read at 550 nm using a spectrophotometric micro plate reader (Tecan; Männedorf, Switzerland). The percentage cell viability to different drug concentrations was calculated as the inhibition rate of (mean absorbance of treated wells/mean absorbance of control wells) × 100%.  Bio-Rad, CA, USA). Membranes were blocked for an hour at room temperature in 5% non-fat dry milk (Bio-Rad; CA, USA) and subsequently incubated with primary antibodies overnight at 4 °C. After three washes with Tween 20 in PBS, membranes were incubated for 1 h at room temperature with corresponding horseradish peroxidase conjugated anti-rabbit and anti-mouse secondary antibodies. The membranes were washed four times with PBS/ Tween 20, and the signals were visualized by ECL reagent (AmershamTM ECL Plus Western Blotting Detection System; GE Healthcare; Buckinghamshire, UK), followed by exposure to chemiluminescence film (Amersham Hyperfilm TM ECL; GE Healthcare; Buckinghamshire, UK). Immunoblot analyses were repeated at least twice for each protein tested.
Immunofluorescence Staining. Cells were seeded onto the 8 wells chamber slide and allowed to grow until 70% confluence. The cells were fixed with 4% formaldehyde for 10 minutes and permeabilised by 0.1% Triton X100 in PBS (PBST) for 10 minutes. The slides were blocked with 1% Bovine Serum Albumin (BSA) in PBST for an hour at room temperature. The cells were incubated overnight with primary antibody p53 and p21 (Cell Signaling, MA, USA) at 4 °C and subsequently incubated with secondary antibody conjugated with Alexa Fluor 568 for an hour in the dark at room temperature. Prolong Gold with DAPI was added into each wells and air-dried in the dark to detect nuclei. The slides were sealed with coverslips for imaging using a confocal laser microscope (Olympus FluoView TM FV1000). GC xenograft murine model. [4][5][6] week old NOD-SCID mice (n = 50) were injected subcutaneously with AGS gastric cell lines (5 × 10 6 ) suspended in Matrigel (BD biosciences). Six animals were randomly assigned to each treatment group (1)  Chicago, IL, USA). Differential analyses between tumor and normal samples were evaluated using Student's t-test, and association between gene expression and clinopathological features was assessed using Pearson correlation. Survival curves were plotted using Kaplan-Meier and statistical significance was assessed using the Cox proportional hazards model. Note: XPO1 high and low expressed groups were determined using the median of respective distributions. These analyses were performed using the statistical package in R (www.r-project.org). A p-value of less than 0.05 was considered statistically significant.

XPO1 upregulation and its association with overall survival in GC patients. Microarray analysis
of clinical samples revealed that XPO1 is overexpressed in gastric cancer as compared to adjacent normal tissues (Fig. 1a). Based on the median expression value, 77 tumors were classified as high whereas the remaining 76 were classified as low in XPO1 expression. None of the tumors showed an absence of XPO1 expression. Interestingly, high XPO1 expression was found to be associated with poor survival outcomes of patients (Fig. 1b). Furthermore, univariate and multivariate analysis of clinicopathological parameters suggested AJCC stage and XPO1 expression as the only independent prognostic factors ( Table 1). Expression levels of XPO1 was also determined in a panel of 10 gastric cancer cell lines. All cell lines showed detectable expression levels of XPO1, among which AGS and YCC10 showed the highest expression (Fig. 1c,d).

Effects of XPO1 inhibition on proliferation in GC cell lines.
To evaluate the growth-inhibitory effect of KPT-SINE analogs, IC 50 concentration of KPT-185, KPT-276 and KPT-330 were determined in gastric cancer cells (Fig. 2a-c). Out of the three SINE analogs, the cell lines were most sensitive to XPO1 inhibition by KPT-330 that resulted in a significant inhibition of cell proliferation with IC 50 values ranging from 36 nM to 1.6 µM. KPT-185 also caused marked inhibition in cell proliferation, however appeared less potent than KPT-330 with IC 50 values in the range of 68 nM to 4.2 µM. KPT-276 has the least potent effect on growth of GC cell lines whereby 3 of the 10 cell lines were found to be resistant with IC 50 over 10 µM.

Apoptosis induction and cell cycle effects in GC cells treated with KPT-330. The effect of XPO1
inhibition in cell cycle regulation and apoptosis was determined in AGS cells that showed high XPO1 expression and KPT-330 sensitivity. Treatment of AGS cells with KPT-330 resulted in cell cycle arrest at G1 phase, concomitant with an increase in sub-G1 population. On the other hand, a significant decrease in the S and G2 phase were also observed in the treated cells as compared to the untreated ones (Fig. 3a, P value < 0.05). Moreover, KPT-330 treatment also resulted in a significant increase in apoptotic population (23.6%) with 17.9% cells entering early apoptosis as compared to untreated cells. (Fig. 3b, P value < 0.05). The effect of KPT-330 in cell cycle regulation and apoptotic induction was determined in YCC10 cells to confirm that the observation is consistent across multiple cell lines. Both AGS and YCC10 showed significant increase in KPT-330 induced cellular apoptosis as compared to untreated cells (Supplementary Fig. S2). To further confirm apoptosis induction in KPT-330 treated cells, we examined the cleavage of caspase-3 and caspase-9

KPT-SINE induced apoptosis in GC cells was mediated by p53 and CRM1. To further investigate
the mechanism of apoptosis induced by the KPT-330, we examined the expression profiles of apoptotic proteins in drug treated AGS cells (Fig. 4a). Expectedly, treatment of cells with KPT-330 lead to a dose-dependent decrease in XPO1 expression levels. A resultant effect of this was observed with marked upregulation of p53 and its transcriptional target p21. On the other hand, a decrease in expression of anti-apoptotic survivin was also observed hence suggesting a KPT-330 triggered induction of cell death. The subcellular localization of p53 and p21 was determined by immunofluorescence analysis. As shown in Fig. (4b), treatment with KPT-330 resulted in a dose-dependent upregulation of p53 and p21 expression in cell nuclei, hence suggesting a gain in apoptotic function.
The KPT-330 induced effects in the intracellular localization of other oncogenic determinants were also evaluated. The cytoplasmic and nuclear fractions of KPT-330 treated AGS cells were extracted and subjected to immunoblot analysis (Fig. 5). Consistently, KPT-330 led to the nuclear retention and upregulation of p53 expression in AGS cells. In contrary, the expression of p27, a non p53 regulated protein, remained unchanged in the nucleus, whereas an increase in its cytoplasmic expression was observed. Interestingly, a decrease in expression of survivin was observed both in the cytoplasmic and nuclear fractions of KPT-330 treated cells.

KPT-330 inhibits growth of gastric cancer cells in a xenograft model. The effect of XPO1 inhibition
in tumor growth was determined in vivo using a mouse xenograft model. NOD-SCID mice transplanted with AGS cells showed a remarkable decease in tumor growth after treatment with KPT-330 as compared to the mice treated with vehicle controls. Since preclinical evaluation of XPO1 inhibition in combination with chemotherapy yielded promising results in other cancers 24,25 , we evaluated the combinatorial effect of KPT-330 with irinotecan in gastric cancer (Fig. 6). The xenografts showed significant anti-tumor effects can be achieved by irinotecan monotherapy. Strikingly, a synergistic effect of KPT-330 was observed in combination of KPT-330 with irinotecan, causing greater appreciable anti-tumor effects. The SINE compound was well tolerated with no toxic deaths and minimal weight loss. In addition, no gross pathology was observed at necropsy carried out at the end of the experiment. Consistent with the in vitro data, KPT-330 treatment induced an upregulation of P53 and p21 expression in in vivo xenografts tissues. An increase in p21 expression was also seen in irinotecan and combinatorial treatment samples, while p53 showed a modest increase as compared to vehicle controls. Furthermore, immunohistochemical analysis revealed a decrease in ki67 expression in all treated samples as against the vehicle controls ( Supplementary Fig. S3).

Discussion
The requirement for nucleo-cytoplasmic export of key proteins in normal cells is suggested to be less stringent that that in malignant cells 19 . This "Achilles heel" of malignant cells could be targeted using nuclear export inhibitors to achieve therapeutic benefits. In this regard, selective killing of tumor cells could be achieved through restoration and reactivation of tumor suppressors. XPO1 is highly expressed in many tumor types and coordinates the nucleo-cytoplasmic export of ~220 proteins. This includes several tumor suppressors and mediators of proliferative and pro-survival signaling pathways 26,27 . The current study identified XPO1 as an independent prognosis indicator, as its over expression correlates with poor survival in gastric cancer patients. XPO1 expression was shown to be significantly high in diffuse type of GC, however no association was observed between XPO1 expression with stage and grade of cancer (Supplementary Table S1). The three SINE compounds (KPT-185, KPT-276 and KPT-330) presented in our study were previously shown to induce covalent modification of the Cys528 residue at the NES groove of XPO1 protein 7 . They are structurally similar, but differ in the pharmacokinetic properties. KPT-185 was shown to have poor PK properties, hence making it unsuitable for in vivo study. On the other hand, KPT-276 and KPT-330 exhibit optimal PK properties 20,25 . Here, we establish the anti-tumor efficacy of XPO1 inhibitor KPT-330, both in in vitro and in vivo models of gastric cancer. Our findings fall in-line with the previous reports that suggested anti-tumor efficacy of XPO1-targeting drugs in various hematological malignancies and solid tumors 28,29 .  Since XPO1 inhibition results in nuclear retention and activation of multiple TSPs, its anti-tumor effects could largely be independent of underlying oncogenic drivers that are key in maintaining a neoplastic state 26 . Although the exact mechanism underling the selective anti-tumor efficacy of KPT-330 remain uncharacterized, a major mediatory role of p53 is assumed. As a nuclear export mediator, XPO1 channels the nuclear export of p53. This leads to the cytoplasmic degradation and functional inactivation of p53 as a tumor suppressor 2,30 . XPO1 inhibition by KPT-330 abrogates p53 degradation through its nuclear retention. When p53 is activated in the nucleus, it could promote either cell cycle arrest or apoptotic cell death 31 . In our study, the activation of caspase-3 and 9 in KPT-330 treated cells is a clear indication of induction of apoptosis. Other proteins that were found to be affected by XPO1 inhibition include tumor suppressor protein p21 and survivin. The upregulation of p21, a transcription target of p53, is suggestive of the restoration/reactivation of p53 tumor suppressor function. Of note, the molecular function of p21 depends on its cellular localization as it plays different roles in nucleus and in cytoplasm. Nuclear p21 induces cell cycle arrest in response to DNA damage, whereas its cytoplasmic expression corresponds  to an anti-apoptotic role. 32 . As shown by immunofluorescence microscopy, p21 expression was retained in the nucleus upon CRM1 inhibition (Fig. 5). Our finding that XPO1 inhibition restored p53 nuclear function, causing cell cycle arrest and apoptosis is consistent with the earlier studies 27,33 . No changes in expression of p27, a non-p53-regulated cell cycle inhibitor, was observed. Intriguingly, XPO1 inhibition by KPT-330 also lead to a decrease in survivin expression levels. Survivin constantly shuttles between the nucleus and the cytoplasm and plays a major role in apoptosis regulation and control of cell cycle progression at the G2/M phase 34 . Elevated level of survivin has been reported to promote tumorogenesis and confer chemotherapy resistance 35,36 . Our study adds further credence to previous reports that suggested strategies aimed at down-regulating survivin as a therapeutic approach 37 .
The combination of molecularly targeted agents with chemotherapy has shown the potential of improving the outcomes of cancer patients. In earlier studies, the ability of XPO1 inhibitors to induce apoptosis or anti-proliferative effects in cell lines was shown to have improved synergistically when combined with chemotherapeutic drugs 38 . Moreover, XPO1 inhibition was shown to induce sequestration of topoisomaerase 1 (TOP1) in the nucleus 39 . This would sensitize the cells to TOP1 inhibitors and the effect could be further enhanced with combinational therapeutic regimens. Here, we also evaluated the anti-tumor efficacy of KPT-330 in combination with irinotecan in xenograft models of gastric cancer. Irinotecan, a topoisomerase 1 inhibitor, is a commonly used chemotherapeutic agent in gastric cancer. It binds to Top I-DNA complex and prevents re-ligation of double stranded DNA during DNA replication 40 . Interestingly, combination of KPT-330 with irinotecan resulted in a remarkable decrease of tumor growth, in-vivo. This synergy observed between KPT-330 and irinotecan generates further interest since combination regimens that accelerate time to remission and overcome resistance is pivotal in cancer therapy.
The present study strongly correlates XPO1 expression with poor prognosis in gastric cancer. Future studies should explore if XPO1 inhibition induces a shift in balance of neoplastic cells from a hyper proliferative state to an apoptotic state that could further be enhanced by concomitant targeting of other crucial cellular processes. The continued evaluation of KPT-330 to establish the safety of targeting nuclear export through XPO1 would augment the clinical translation of XPO1 inhibitors for the treatment of gastric cancer. Nonetheless, the enhanced efficacy and low toxicity profiles of KPT-330 in combination with conventional chemotherapy is promising and adds value to a synergistic interaction that could be utilized in the treatment of refractory gastric cancers.