Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Feb 1.
Published in final edited form as: Mol Cancer Ther. 2020 Jun 4;19(8):1727–1735. doi: 10.1158/1535-7163.MCT-19-0977

NRG1/ERBB3 pathway activation induces acquired resistance to XPO1 inhibitors

Takahito M Miyake 1,*, Sunila Pradeep 6,*, Emine Bayraktar 1, Elaine Stur 1, Katelyn F Handley 1, Sherry Y Wu 1, Cristian Rodriguez-Aguayo 4,5, Ju-Seog Lee 2, Gabriel Lopez-Berestein 3,4,5, Robert L Coleman 1, Anil K Sood 1,3,4
PMCID: PMC7415525  NIHMSID: NIHMS1600893  PMID: 32499298

Abstract

XPO1 inhibitors have shown promise in cancer treatment, but mechanisms of resistance to these drugs are not well understood. In this study, we established selective inhibitors of nuclear export (SINE)-resistant ovarian cancer cell lines from in vivo mouse tumors and determined the mechanisms of adaptive XPO1 inhibitor resistance using protein and genomic arrays. Pathway analyses revealed up-regulation of the NRG1/ERBB3 pathway in SINE-resistant cells. Depletion of ERBB3 using siRNAs restored the anti-tumor effect of SINE in vitro and in vivo. Furthermore, exogenous NRG1 decreased the anti-tumor effect of SINE in ovarian cancer cell lines with high ERBB3 expression, but not in those with low expression. These results suggest that NRG1 and ERBB3 expression is a potential biomarker of response to SINE treatment. The antitumor effect of SINE was reduced by exogenous NRG1 in an ERBB3-dependent manner. These findings suggest that NRG1 and ERBB3 are effective biomarkers that should be evaluated in future clinical trials and are relevant therapeutic targets for the treatment of SINE-resistant cancers.

Keywords: XPO1, SINE, ERBB3, NRG, HER3, neuregulin, resistance

INTRODUCTION

Ovarian cancer (OC) continues to be a global problem due, in part, to advanced stage at diagnosis and lack of effective screening tests (1). OC is responsible for 22,530 new cases per year and about 13,980 deaths every year in the United States (2). Despite advances in treatment in the last several years, a high percentage of patients will relapse, leading to eventual drug resistance (3).

Chromosome maintenance protein 1 (CRM1/XPO1) is an important nuclear export protein that recognizes leucine-rich nuclear export signals and transports cargo proteins through a nuclear pore complex, playing a crucial role in maintaining the normal cellular function (4). The cargo proteins that are transported by XPO1 include tumor-suppressor proteins and transcription factors, such as p53, p21, and p27. Overexpression of XPO1 has been linked to inactivation of these proteins (5). In addition, overexpression of XPO1 correlates with poor prognosis (6–9). The first specific XPO1 inhibitor leptomycin B showed promising anti-tumor effects by inhibiting the XPO1-mediated nuclear export channel; however, it is not clinically useful because of substantial toxicity (10). Novel small selective inhibitors of nuclear export (SINEs) have been developed to bind specifically to the Cys-528 residue located in the cargo binding portion of XPO1 in a slowly reversible manner (11). These compounds have been reported to have anti-tumor effects in several cancer types, such as chronic myeloid and lymphocytic leukemia, colon, pancreatic, prostate, and breast cancers (12–16). SINEs were also found to have robust anti-tumor effects in ovarian cancer mouse models, with tolerable adverse effects, and are considered a potential therapeutic agent for ovarian cancer (17). In July 2019, the U.S. FDA approved selinexor, a SINE compound for patients with multiple myeloma refractory to proteasome inhibitors, immunomodulatory agents, and an anti-CD38 monoclonal antibody (18). As SINE compounds are currently being clinically evaluated for the treatment of solid and hematologic malignancies, biomarkers and adaptive mechanisms for SINEs are necessary to personalize and maximize therapeutic outcomes.

In this study, we established SINE-resistant ovarian cancer models and identified neuregulin 1 (NRG1)/ERBB3 pathway as a potent mechanism of SINE resistance. In resistant cells, there was an increase in NRG1 and ERBB3 expression with a concomitant increase in NRG1 secretion when compared to parental cells. Depleting ERBB3 using siRNA restored the cytotoxic effects of the in vitro tool compound KPT-185 in SINE-resistant cells (19). Adding exogenous NRG1 decreased the cytotoxic effect of KPT-185 in ERBB3-expressing ovarian cancer cells. These results suggest that acquired resistance to SINE is driven by the activated NRG1-ERBB3 pathway and this axis can be targeted in SINE-resistant cancers.

MATERIALS AND METHODS

Animal Care

We purchased 8–12-week old female athymic nude mice (strain NCRNU/ RRID:IMSR_TAC:ncrnu), from Taconic Bioscience. The mice were segregated and maintained under specific pathogen-free conditions in an animal facility which is approved by the American Association for Accreditation of Laboratory Animal Care and conforms to the current regulations and standards of the United States Department of Agriculture, Department of Health and Human Services, and the National Institutes of Health. The study protocols were approved and supervised by the Institutional Animal Care and Use Committee at The University of Texas MD Anderson Cancer Center, under ACUF protocol number 01–12-01231.

Establishment of SINE-Resistant Cell Lines and Cell Culture

A2780 (ovarian endometrioid adenocarcinoma), SKOV3 (ovarian serous cystadenocarcinoma, derived from ascites), OVCAR8 (high-grade ovarian serous adenocarcinoma) and HeyA8 (adenocarcinoma) were purchased from the MD Anderson characterized cell line core. All the cell lines were authenticated by short tandem repeat profiling by the core and were routinely tested by PCR for Mycoplasma. Besides that, the SKOV3 cells were subsequently recycled and were denoted as SKOV3-OM3 subpopulations when derived from omental tumors. The cells were cultivated at 37°C and 5% CO2 and used when cells were 70–80% confluent. The cell lines were used within 3–20 passages for in vitro and 3–10 passages for in vivo experiments.

To establish SINE-resistant ovarian cancer cell lines, we injected 1 × 106 A2780 or SKOV3-Luc cells into mice (n = 10 / group) intraperitoneally in 0.2 ml of Hank’s balanced saline solution (Life Technologies Invitrogen, Carlsbad, CA). After tumors became palpable (and visible by bioluminescence imaging in the SKOV3-Luc model) 20 days after tumor cell injection, mice were treated with 20 mg/kg selinexor (KPT-330, a clinical stage SINE compound; obtained from Karyopharm Therapeutics Inc.) orally, twice weekly. Mice were monitored daily to evaluate for any adverse effects of therapy, and weekly to evaluate tumor size by bioluminescence imaging in the SKOV3-Luc model. Most of the tumors were located in the pelvis, omentum, peritoneum, mesentery, bowel wall, porta hepatis, diaphragm or around the spleen. Mice were killed by cervical dislocation when they appeared moribund; with the A2780 model, mice were sacrificed on days 59 to 72 and for the SKOV3 model, on days 47 to 55 as they become moribund. Tumor cells were dissociated immediately and seeded in cell culture plates. Once the cells formed colonies, they were separated and labeled. Cell viability and the apoptotic rate were determined after KPT-185 (obtained from Karyopharm Therapeutics Inc.) treatment to confirm the resistant phenotype. SINE-resistant cell lines were maintained in RPMI 1640 supplemented with 15% fetal bovine serum, gentamycin, and 0.2 μM of KPT-185 to maintain their resistant phenotype. Here, KPT-185 was used as a tool compound, while KPT-330 (selinexor) is one of the compounds that was selected for clinical development. KPT-185 and KPT-330 (SINE compounds) are analogs with slightly different chemical structures, both targeting the cys-528 residue of XPO1 (19–21).

For in vivo validation experiments, we injected mice intraperitoneally with 1 × 106 A2780-par or A2780-res1 cells. Mice were randomly assigned to control or selinexor (20 mg/kg body weight, twice weekly) treated groups (n = 5 / group), and treatment was initiated 10 days after tumor cell injection based on tumor nodule palpation. Mice were sacrificed on day 35, and necropsy was performed.

Cytotoxicity Assay for Cancer Cell Lines

The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) uptake assay was used to determine the cytotoxicity of KPT-185 treatment, as described previously. In brief, 5×103 cells were plated in each well of a 96-well plate with complete growing medium. Twenty-four hours after plating, the medium was replaced with 200 μl of fresh medium and cells were incubated with dimethyl sulfoxide (DMSO) (control) or various concentrations of KPT-185 for 72 hours. Cell growth was assessed by adding 50 μl of 0.15% MTT (Sigma-Aldrich, St. Louis, MO) to each well and after incubation for 2 hours at 37°C, the medium was removed from each well and 100 μl of DMSO was added. After brief shaking to blend the samples, the absorbance at 570 nm was recorded using a Falcon microplate reader (Becton Dickinson Labware, Franklin Lakes, NJ). The mean absorbance at 570 nm of KPT treated cells were normalized to that of DMSO treated cells to determine the cell viability. Cell viability was plotted versus concentration of KPT-185. The IC50 was determined by calculating the mean absorbance at 570 nm and then identifying the corresponding KPT-185 concentration on the dose-response curve. Experiments were performed with each sample at each condition in triplicate.

Cytotoxicity assay under co-culture conditions

To have continuous exposure to NRG1, HeyA8, OVCAR8, SKOV3-par and SKOV3-OM3 cell lines were co-cultured with the A2780-res cell line in a Transwell 96-well plate (8μm pore size). In brief, 5×103 recipient cells were plated in the bottom of each well with complete growing medium. A2780-res cells were plated at a density of 1.2 ×103 with or without 0.2 μM of KPT-185 for 72 hours. Cell growth was assessed by MTT assay as described above.

Apoptosis Assays in Cell Lines

The relative percentage of apoptotic cells was assessed at 0, 30, and 40 hours after treatment with 0.2 μM KPT-185 using Annexin V-Coupled Fluorescein Isothiocyanate (FITC) Apoptosis Detection Kit-1 (BD Pharmingen, San Diego, CA) per the manufacturer’s protocol. Briefly, the cells cultured in a six-well plate were dissociated using trypsin-EDTA, then washed with phosphate-buffered saline and re-suspended in 100 μl of Annexin V-binding buffer at 1 × 106 cells per milliliter. After adding 5 μl Annexin V-coupled FITC and propidium iodide (a probe to distinguish viable from nonviable cells) per 1 × 105 cells, samples were mixed gently. After 15 minutes of incubation at ambient temperature in the dark, 400 μl of Annexin V-binding buffer was added. Analysis was performed using flow cytometry. Propidium iodide was excluded from viable cells which had intact membranes and permeated cells which were dead or had damaged membranes. Cells that stained positive for Annexin V-coupled FITC and negative for propidium iodide were identified as undergoing apoptosis. Cells that stained positive for both Annexin V-coupled FITC and propidium iodide were in the late stage of apoptosis, undergoing necrosis, or already dead. Cells that stained negative for both Annexin V-coupled FITC and propidium iodide were alive and not undergoing measurable apoptosis.

Microarray Analysis

The mirVana RNA isolation labeling kit (Ambion) was used for RNA extraction. 500 ng of total RNA was used for labeling and hybridization of HumanHT-12 v4 Expression BeadChip according to the manufacturer’s protocols (Illumina, San Diego, CA). After the bead chips were scanned with an Illumina BeadArray Reader (Illumina), the microarray data were normalized, and the log2 values were transferred into R. The data is deposited in GEO under accession number GSE126519.

IPA

To identify possible pathways that are related to SINE resistance, IPA (Ingenuity Pathway Analysis) web-based software was used. Protein symbols and expression data from an RPPA and gene symbols and expression data from a microarray were imported into IPA. IPA is a repository of biological interactions and functional annotations from many relationships between proteins and genes. Proteins and genes from the dataset with p values less than 0.05, according to Student’s t-test, were considered for analysis. To identify the most relevant canonical pathways, we chose those with a p value less than 0.001 by Fisher’s exact test.

RESULTS

Establishing SINE-Resistant Ovarian Cancer Cell Lines from In Vivo Mouse Tumors

A total of 15 SINE-resistant cell lines were generated and IC50 of KPT-185 (SINE compound for in vitro experiment) were determined for all cell lines. Of these, three cell lines showed significant resistance to SINE therapy and had higher IC50 than that of the A2780 parental cell line. The IC50 concentrations of KPT-185 were >10 μM for two of the resistant cell lines (A2780-res1 and A2780-res2) and 0.51 μM for a third resistant cell line (A2780-res3), compared with the IC50 level of 0.15 μM for the parental cell line A2780-par (Figure 1A). The cell viabilities of A2780-res3 and A2780-res4 are shown in Supplementary Figure 1A. Based on these results, we selected A2780-res1 and A2780-res2 cells for further analyses. To validate that A2780-res1 and A2780-res2 are resistant to KPT-185, we also performed apoptosis assays. The percentage of apoptotic cells was 2.7%, 4.9%, and 4.1% without treatment; 13.9%, 4.7%, and 7.5% after 30 hours of treatment; and 28.8%, 13.3%, and 18.4% after 40 hours of incubation with 0.2 μM of KPT-185 in A2780-par, A2780-res1, and A2780-res2 cells, respectively (Figure 1B). Given the higher degree of resistance in A2780-res1, these cells were used for all subsequent experiments.

Figure 1. SINE-resistant ovarian cancer cell line separated from in vivo mouse tumor.

Figure 1.

Open in a new tab

(A) Cell viability assay of A2780-par, A2780-res1, and A2780-res2 cell lines after 72 hours of incubation with KPT-185. The x-axis indicates the concentration of KPT-185. (B) Apoptosis assay of A2780-par, A2780–res1, and A2780-res2 cell lines untreated or after 30 and 40 hours of incubation with 0.2 μM KPT-185. Each sample is triplicated. Mean tumor weight (C) and number of tumor nodules (D) in mice treated with vehicle (control) or selinexor in an A2780-par model. Mean tumor weight (E) and number of tumor nodules (F) in mice treated with vehicle (control) or selinexor in an A2780-res1 model. Error bars indicate SE. **, p<0.01; ***, p<0.001.

We next examined whether A2780-res1 cells are SINE resistant in vivo. Following injection of 1×106 A2780-par or A2780-res1 cells in the peritoneal cavity, mice were randomly assigned to one of the following two treatment groups (n=5 mice per group): vehicle (control) or selinexor (20 mg/kg, twice weekly). Treatment was initiated 10 days after tumor cell injection. Mice were sacrificed on day 28 and necropsied. Tumor weights and numbers of nodules were significantly reduced in mice treated with selinexor compared to control in the A2780-par model but did not significantly differ between control and selinexor treated groups in A2780-res1 model (Figure 1C–1F).

NRG1/ERBB3 Pathway Is Up-Regulated in SINE-Resistant Cell Lines

To determine the mechanism of SINE resistance, we performed a reverse phase protein array (RPPA) using protein samples obtained from A2780-par and A2780-res1 cell lines. Both samples were analyzed in triplicate. We identified 105 proteins with significantly altered expression levels in these two cell lines. Next, we performed pathway enrichment analysis on these 105 proteins using the “canonical pathway” function in Ingenuity Pathway Analysis (IPA) software (Qiagen, Redwood City, CA). We identified 189 canonical pathways that are significantly altered in A2780-res1 compared to A2780-par cells (p<0.001; Figure 2A). We focused on the NRG signaling pathway, which had the second lowest p value among the 189 canonical pathways since NRG was the top protein found to be up-regulated in A2780-res1 cells compared to A2780-par cells (2.5-fold increase, Figure 2B). We also carried out a whole genome RNA microarray analysis in A2780-par and A2780-res1 cell lines and identified 1477 genes with altered expression (fold change <0.5 or >2) in A2780-res1 cells compared to that in A2780-par cells. The expression of NRG1 and ERBB3 was increased by 22-fold and 1.9-fold in A2780-res1 cells compared to A2780-par cells, respectively. Consistent with the RPPA results, the NRG signaling pathway was found to be significantly altered in the resistant cell line (Supplementary Figure 1B, 1C). This was further confirmed with Western blot which demonstrated an increase in both ERBB3 and NRG1 protein expression in A2780-res1 cells compared to A2780par cells (Supplementary Figures 1D, 1E, and 1F).

Figure 2. NRG signaling pathway is activated in SINE-resistant cell lines.

Figure 2.

Open in a new tab

(A) Canonical pathway from the IPA based on RPPA data. Data for the 10 most statistically significant pathways are presented. The upper x-axis corresponds to data for the bars; these data are the negative logarithm of p values that were calculated by Fisher exact test, with a threshold for statistical significance set at 0.001. The bottom x-axis corresponds to data in the line graphs; these data are the ratio of the number of molecules in a given pathway that meets the two-fold change cut-off criterion in either direction to the total number of molecules that make up that pathway. (B) Heat map of the molecules that are included in the NRG signaling pathway with RPPA. NRG1 (C) and ERBB3 (D) mRNA expression in A2780-par and A2780-res (1–4) cell lines. Y-axis indicates the relative expression of NRG1 or ERBB3 to A2780-par cells. (E) NRG1 concentration of conditioned medium after 24 and 48 hours of incubation with the A2780-par or A2780-res1 cell line. (F) ERBB3 mRNA expression in HeyA8, OVCAR8, SKOV3-par and SKOV3-OM3 cell lines. Y-axis indicates the relative expression of ERBB3 to A2780-par cells. (G) IC50 of KPT-185 in ovarian cancer cell lines after 72 hours incubation, as analyzed in GraphPad Prism. The IC50 was determined by calculating the mean absorbance at 570nm and then identifying the corresponding KPT-185 concentration on the dose-response curve. Error bars indicate SD of triplicated experiments. **; p<0.01, ***; p<0.001

To examine if NRG1 and ERBB3 levels are also increased in other SINE-resistant A2780 cell lines, we determined their mRNA expression levels in A2780-res2, A2780-res3, and A2780-res4 cell lines. Significantly increased expression of NRG1 (Figure 2C) and ERBB3 (Figure 2D) was observed in all three of these resistant lines when compared to the parental cell line. Given that some ovarian cancer cells secrete the ERBB3 ligand NRG1 (22) and qRT-PCR revealed increased expression of NRG1 mRNA in SINE-resistant cell lines, we next examined whether NRG1 secretion is increased in SINE-resistant cells. The NRG1 concentration in conditioned medium derived from A2780-par or A2780-res1 cultures was quantified after 24 and 48 hours of incubation. More than a 3-fold increase in NRG1 level was observed in A2780-res1 compared to A2780-par cells (Figure 2E). Since NRG1 has been reported to bind to members of the ERBB family, consisting of EGFR/ERBB1, ERBB2, ERBB3, and ERBB4, we next assessed the expression of ERBB family members (ERBB2, ERBB3, and ERBB4). pERBB3 expression was increased in A2780-res1 cells (Supplementary Figure 1D, 1E), and the total and activated levels of ERBB2 and ERBB4 did not differ between A2780-par and A2780-res1 cells. ERBB3 mRNA expression was determined in HeyA8, OVCAR8, SKOV3-par and SKOV3-OM3 cell lines compared to A2780-par cells (Figure 2F; Supplementary Figure 1G). The IC50 of KPT-185 were determined in ovarian cancer cell lines (Figure 2G). These results suggest that SINE resistance can be driven by NRG1/ERBB3 activation. In addition, the expression of phosphorylated AKT (pAKT), which is downstream of NRG1/ERBB3, was increased in A2780-res1 cells after NRG1 stimulation. In contrast, minimal pAKTcould be detected in A2780-par cells (Supplementary Figure 1H; Supplementary Figure 1I).

To extend this observation to another ovarian cancer cell line, we established 14 SINE-resistant SKOV3-Luc cell lines using orthotopic ovarian cancer mouse models. The IC50 of KPT-185 were determined for all cell lines using the MTT cell viability assay, and three of these cell lines showed significant resistance to SINE therapy (SKOV3-res1, SKOV3-res2, and SKOV3-res3) (see Supplementary Methods, Supplementary Figure 2A, 2B, 2C). In these SINE-resistant models, mRNA expression of ERBB3 and NRG1 was determined. Although ERBB3 expression was increased in SINE-resistant SKOV3-Luc cell lines (SKOV3-res1, SKOV3-res2, and SKOV3-res3) compared to the SKOV3-Luc parental line, NRG1 expression did not significantly differ between the resistant and sensitive cell lines, in contrast to A2780-par and A2780-res1 cell lines (Supplementary Figure 2D, 2E, 1I).

Role of NRG1/ERBB3 Pathway in SINE Resistance

To further test the role of ERBB3 in SINE-resistance, we assessed cell viability following ERBB3 knockdown using siRNA in A2780-res1 cells. After 72 hours of incubation with 0.2 μM KPT-185, ERBB3 depletion restored the cytotoxic effect of KPT-185 in A2780-res1 cells (Figure 3A, Supplementary Figure 3A). Next, we determined whether ERBB3 activation by exogenous NRG1 reduces the cytotoxic effects of KPT-185. We added exogenous NRG1 to A2780-par cells and subsequently examined cell viability 72 hours after KPT-185 treatment (0.2 μM). Incubation with exogenous NRG1 reduced the cytotoxic effect of KPT-185 in A2780-par cells by 15% (Figure 3B). This suggests that activation of the NRG1/ERBB3 pathway could be driven by either autocrine or paracrine signaling. Supplementary Figure 3B confirms that NRG1 siRNA indeed knocks down NRG1 expression. We observed that NRG1 silencing in the presence of KPT-185 decreased the cell viability of A2780-res1 cells (Figure 3C). To have continuous exposure to NRG1, we co-cultured four ovarian cancer cell lines, HeyA8 and SKOV3 which express ERBB3 at low levels, and OVCAR8 and SKOV3-OM3 which express higher levels of ERBB3, with A2780-res1 cells that secrete NRG1, indirectly separated by semi-permeable membranes, and analyzed the anti-tumor effects of KPT-185 (Figure 3D). After 72 hours of incubation with KPT-185, decreased anti-tumor effects of KPT-185 were observed only with OVCAR8 and SKOV3-OM3 when co-cultured with A2780-res1 cells. This suggests that the SINE-resistance caused by activating NRG1/ERBB3 pathway depends on tumoral ERBB3 expression. To confirm this finding, we overexpressed ERBB3 in HeyA8 cells using lentivirus transfection and examined cell viability after incubation for 72 hours with 0.2 μM KPT-185 co-cultured with A2780-res1 cells. Compared to empty lentivirus transfected HeyA8 cells (HeyA8-control), ERBB3-expressing HeyA8 cells (HeyA8-ERBB3) were less responsive to KPT-185 treatment when co-cultured with A2780-res1 cells (Figure 3E). To determine the effect of ERBB3 depletion in cells that have high ERBB3 expression, we silenced ERBB3 in OVCAR8 and SKOV3-OM3 cells using siRNA. Compared to control siRNA transfected cells, ERBB3 siRNA transfected OVCAR8 and SKOV3-OM3 cells showed lower cell viability after incubation with 0.2 μM KPT-185 for 72 hours when co-cultured with A2780-res1 cells (Figure 3F, Supplementary Figure 3C, 3D). In addition, we silenced ERBB2 and ERBB3 in A2780-res1 cells using siERBB2 and siERBB3, respectively. The results showed that the knockdown of ERBB3 resulted in decreased XPO1 expression. (Supplementary Figure 3E).

Figure 3. SINE-resistance is ERBB3 dependent.

Figure 3.

Open in a new tab

(A) ERBB3 depletion resulted in a restored anti-tumor effect of KPT-185 in A2780-res1 cells. Cell viability of ERBB3 siRNA2 transfected A2780-res1 cells after 72 hours of incubation with 0.2 μM of KPT-185 was significantly lower than that of control siRNA transfected A2780-res1 cells. (B) Cell viability of A2780-par cells after 72 hours of incubation with 0.2 μM KPT-185 and exogenous NRG1. KPT-185 had less antitumor effect in A2780-par cells when exogenous NRG1 was present. (C) Cell viability of NRG1 siRNA transfected A2780-res1 cells after 72 hours of incubation with 0.2 μM of KPT-185 was significantly lower than that of control siRNA transfected A2780-res1 cells (D) Cell viability of ERBB3 high and low expressing ovarian cancer cell lines co-culturing with A2780-res1 cells after 72 hours incubation with or without KPT-185. Anti-tumor effect of KPT-185 was decreased in OVCAR8 and SKOV3-OM3 cells when co-cultured with A2780-res1 cells. (E) Cell viability of ERBB3 overexpressing HeyA8 cells 72 hours after incubation with 0.2 μM KPT-185. Anti-tumor effect of KPT-185 was less in ERBB3 overexpressing HeyA8 cells compared to control HeyA8 cells. (F) Cell viability of ERBB3 depleted OVCAR8 and SKOV3-OM3 cells 72 hours after incubation with 0.2 μM KPT-185. ERBB3 depletion in OVCAR8 and SKOV3-OM3 cells resulted in a restored anti-tumor effect of KPT-185. Error bars indicate SD of triplicate experiments. *, p<0.05; ***, p<0.001.

Next, we conducted a therapeutic in vivo experiment in the A2780-res1 orthotopic ovarian cancer mouse model using our well-characterized method for systemic delivery of small interfering RNA (siRNA) to tumor cells (23, 24). Using the dioleoylphosphatidylcholine (DOPC) nanoliposomal platform, we sought to determine whether silencing tumoral ERBB3 expression could enhance SINE activity in the A2780-res1 model. Tumor-bearing mice were randomly assigned to one of the following four treatment groups (n=10 mice per group): control siRNA plus vehicle, control siRNA plus selinexor (20 mg/kg, twice weekly), ERBB3 siRNA (5 μg/mouse, twice weekly) plus vehicle, or ERBB3 siRNA plus selinexor (Figure 4A). Selinexor was administered orally and control or ERBB3 siRNA-DOPC was administered intraperitoneally. Mice were sacrificed when they became moribund, and necropsies were performed. Selinexor did not show an anti-tumor effect in the control siRNA-DOPC treated group. In contrast, the mean tumor weight and number of tumor nodules in mice treated with ERBB3 siRNA-DOPC plus selinexor were significantly lower than in mice treated with control siRNA-DOPC alone or control siRNA-DOPC plus selinexor. The mean tumor weight and number of tumor nodules in mice treated with ERBB3 siRNA-DOPC plus selinexor were also lower than in mice treated with ERBB3 siRNA-DOPC alone; however, this difference was not statistically significant (Figure 4B, C).

Figure 4. The antitumor effect of selinexor (KPT-330) with ERBB3 depletion in a SINE-resistant ovarian cancer mouse model.

Figure 4.

Open in a new tab

(A) Immunoblot analysis of total ERBB3 expression in tumor samples. Tumors from ERBB3 siRNA treated mice showed decreased ERBB3 expression compared to those from control siRNA treated mice. (B, C) Mean tumor weight (B) and number of tumor nodules (C) in mice treated with control siRNA, control siRNA plus selinexor, ERBB3 siRNA2 and ERBB3 siRNA2 plus selinexor in the A2780-res1 model. ERBB3 siRNA2 treatment restored the anti-tumor effect of selinexor in this SINE-resistant model. (D, E) Immunohistochemical staining for cleaved caspase 3 (D) and Ki67 (E) in tumor specimens obtained from the A2780-res1 model. Ns, not significant; *, p<0.05; **, p<0.01; ***p<0.001.

To determine the biological effects of ERBB3 siRNA-DOPC and selinexor, we next examined the apoptotic and proliferation rates in tumors. Apoptosis was determined by cleaved caspase 3 immunohistochemical staining. The percentage of cleaved caspase 3-positive cells was higher in mice treated with ERBB3 siRNA-DOPC than in mice treated with control siRNA-DOPC (Figure 4D). Moreover, the percentage of cleaved caspase 3-positive cells in mice treated with ERBB3 siRNA-DOPC plus selinexor was significantly higher than in mice treated with ERBB3 siRNA-DOPC alone. Tumor cell proliferation was determined by Ki67 immunohistochemical staining. Fewer Ki67-positive cells were observed in tumors from mice treated with ERBB3 siRNA-DOPC than in mice treated with control siRNA-DOPC, and the number of Ki67-positive cells in mice treated with ERBB3 siRNA-DOPC plus selinexor was significantly lower than that in mice treated with ERBB3 siRNA-DOPC alone (Figure 4E). These results suggest that targeting ERBB3 is an effective therapeutic strategy in SINE-resistant tumors.

DISCUSSION

Ovarian cancer is the most lethal gynecological cancer and due to the high rates of chemoresistance, development of new drug targets is essential. Given the anti-tumor activity of selinexor, we aimed to understand mechanisms of resistance and to restore the anti-tumor effect on SINE-resistant cells. In this study, we found that the NRG1/ERBB3 pathway was activated in SINE-resistant ovarian cancer cell lines and the anti-tumor effect of SINE can be restored by ERBB3 depletion both in vitro and in vivo. Furthermore, endogenous or exogenous NRG1 decreases the anti-tumor effect of SINE. These findings suggest that NRG1 and ERBB3 expression is a potential biomarker of response to SINE treatment.

ERBB3 is a member of the ERBB family of cell-surface receptor proteins and a dimerization partner of ERBB2. ERBB3 has been reported to be overexpressed in ovarian cancer in comparison with normal ovarian tissue in clinical samples, and its expression correlates with poor overall survival (25). The ERBB3 family plays an important role in cell differentiation, migration, and proliferation (26) and has been correlated with cancer development. ERBB3 and ERBB4 also can bind with NRG1 resulting in the activation of the receptor.

ERBB2/ERBB3 heterodimers are known to deliver the most potent and long-lasting signal among the possible combinations of the four members of epidermal growth factor receptors (ERBB1 to ERBB4). Ectopically expressed NRG1 decreased the anti-tumor effect of SINE in ovarian cancer cell lines with ERBB3 expression. The expression of NRG1 in ovarian cancer and the potential for autocrine and paracrine growth regulation have been reported previously (22). Moreover, it has been shown that ERBB3 directly interacts with XPO1 (27). Our study shows that silencing ERBB3 resulted in decreased expression of XPO1, and this suggests that ERBB3 plays a role in XPO1 expression. The ERBB3 pathway is able to activate the PI3K/AKT pathway and this activation is important for drug resistance. Moreover, the AKT pathway is activated by XPO1 (28, 29). Overall, ERBB3 has an important role in resistance to XPO1 inhibitors; this effect could be direct and/or indirect.

In addition, the outcome of NRG1-induced activation of anchorage-independent growth has been shown to be dependent on the relative levels of ERBB2 and ERBB3 (30). A previous study reported that NRG1 binds specifically to ERBB3 and ERBB4, which could explain lack of changes in ERBB2 expression and activation (31). NRG1 can bind directly to ERBB4 receptors and activate them, but the low expression of ERBB4 in our model could explain the lack of activation of this receptor. ERBB3 can also bind directly to NRG1, but ERBB3 has an impaired tyrosine kinase domain and needs to form dimers with ERBB2 or ERBB4 to become activated (32). Tumor cells often co-express ERBB2 and ERBB3 but the mechanism of co-expression remains elusive. It is not clear that overexpression of ERBB2 and ERBB3 occur simultaneously or whether tumor cells overexpress one receptor which subsequently enhances the expression of the other. Our data indicate that only ERBB3 activation, but not ERBB2 or ERBB4, relates to resistance to SINE. This suggests that SINE resistance could be related to direct binding of increased ERBB3 and XPO1, but additional work is needed to test this.

In conclusion, our study shows that the anti-tumor effect of SINE was reduced by NRG1 expression in an ERBB3-dependent manner. These results identify a candidate biomarker for predicting response to SINE-therapy as well as a strategy for overcoming resistance to SINE therapy using ERBB3/NRG1-targeted therapeutic approaches. Seribantumab, a humanized monoclonal antibody against ERBB3, might be one such drug for use in the setting of SINE resistance.

Supplementary Material

1

ACKNOWLEDGEMENTS

Financial support: A. K. Sood is supported by the NIH (P50 CA217685, P50 CA098258, R35 CA209904), the American Cancer Society Research Professor Award, the Blanton-Davis Ovarian Cancer Research Program, the RGK Foundation, and the Frank McGraw Memorial Chair in Cancer Research. S. Pradeep is supported by the Liz Tilberis Early Career Award and OCRF. E. Stur is supported by Ovarian Cancer Research Alliance (OCRA number FP00006137). K. F. Handley is supported by a training fellowship from the Gulf Coast Consortia, on the Computational Cancer Biology Training Program (CPRIT Grant No. RP170593). C. Rodriguez-Aguayo was supported by the NIH through the Ovarian Spore Career Enhancement Program, and the NCI grant FP00000019. S. Wu was supported by Ovarian Cancer Research Fund Alliance, Foundation for Women’s Cancer, Texas Center for Cancer Nanomedicine, and Cancer Prevention and Research Institute of Texas training grants (RP101502 and RP101489, respectively). The M. D. Anderson Research Core Facilities are funded by NIH grant CA016672.

Disclosure of potential conflicts of interest: GLB is a Shareholder in BioPath Holdings. RLC: Research Funding (AstraZeneca, Abbvie, Clovis, Roche/Genentech, Janssen, Merck); Scientific Steering Committee (Abbvie, AstraZeneca, Clovis, Immunogen, Tesaro, Array, Janssen, Genmab, Gamamab). AKS: Consulting (Merck, Kiyatec); research funding (M-Trap); shareholder (BioPath).

REFERENCES

  • 1.Matulonis UA, Sood AK, Fallowfield L, Howitt BE, Sehouli J, Karlan BY. Ovarian cancer. Nat Rev Dis Primers. 2016;2:16061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7–34. [DOI] [PubMed] [Google Scholar]
  • 3.Dal Molin GZ, Omatsu K, Sood AK, Coleman RL. Rucaparib in ovarian cancer: an update on safety, efficacy and place in therapy. Ther Adv Med Oncol. 2018;10:1758835918778483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Chen Y, Camacho SC, Silvers TR, Razak AR, Gabrail NY, Gerecitano JF, et al. Inhibition of the Nuclear Export Receptor XPO1 as a Therapeutic Target for Platinum-Resistant Ovarian Cancer. Clin Cancer Res. 2017;23(6):1552–63. [DOI] [PubMed] [Google Scholar]
  • 5.Pathria G, Wagner C, Wagner SN. Inhibition of CRM1-mediated nucleocytoplasmic transport: triggering human melanoma cell apoptosis by perturbing multiple cellular pathways. The Journal of investigative dermatology. 2012;132(12):2780–90. [DOI] [PubMed] [Google Scholar]
  • 6.Noske A, Weichert W, Niesporek S, Roske A, Buckendahl AC, Koch I, et al. Expression of the nuclear export protein chromosomal region maintenance/exportin 1/Xpo1 is a prognostic factor in human ovarian cancer. Cancer. 2008;112(8):1733–43. [DOI] [PubMed] [Google Scholar]
  • 7.Shen A, Wang Y, Zhao Y, Zou L, Sun L, Cheng C. Expression of CRM1 in human gliomas and its significance in p27 expression and clinical prognosis. Neurosurgery. 2009;65(1):153–9; discussion 9–60. [DOI] [PubMed] [Google Scholar]
  • 8.van der Watt PJ, Maske CP, Hendricks DT, Parker MI, Denny L, Govender D, et al. The Karyopherin proteins, Crm1 and Karyopherin beta1, are overexpressed in cervical cancer and are critical for cancer cell survival and proliferation. International journal of cancer Journal international du cancer. 2009;124(8):1829–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Huang WY, Yue L, Qiu WS, Wang LW, Zhou XH, Sun YJ. Prognostic value of CRM1 in pancreas cancer. Clinical and investigative medicine Medecine clinique et experimentale. 2009;32(6):E315. [PubMed] [Google Scholar]
  • 10.Newlands ES, Rustin GJ, Brampton MH. Phase I trial of elactocin. British journal of cancer. 1996;74(4):648–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Turner JG, Dawson J, Sullivan DM. Nuclear export of proteins and drug resistance in cancer. Biochemical pharmacology. 2012;83(8):1021–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Draetta GF, Shacham S, Kauffman M, Sandanayaka V, Schechter S, Williams J, et al. Cytotoxicity of novel, small molecule, CRM1-selective inhibitors of nuclear export (SINE) in colorectal cancer (CRC) cells. ASCO Meeting Abstracts. 2011;29(15_suppl):e14091. [Google Scholar]
  • 13.Azmi AS, Aboukameel A, Bao B, Sarkar FH, Philip PA, Kauffman M, et al. Selective inhibitors of nuclear export block pancreatic cancer cell proliferation and reduce tumor growth in mice. Gastroenterology. 2013;144(2):447–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Shacham S, Gravina GL, Ricevuto E, Mancini A, Chin L, Shechter S, et al. Prelinical evaluation of selective inhibitors of nuclear export (SINE) CRM1 (XPO1) inhibitors in prostate cancer (PrCa). ASCO Meeting Abstracts. 2012;30(15_suppl):e15200. [Google Scholar]
  • 15.McCauley D, Landesman Y, Senapedis W, Kashyap T, Saint-Martin J-R, Plamondon L, et al. Preclinical evaluation of selective inhibitors of nuclear export (SINE) in basal-like breast cancer (BLBC). ASCO Meeting Abstracts. 2012;30(15_suppl):1055. [Google Scholar]
  • 16.Lapalombella R, Berglund C, Mahoney E, Williams K, Jha S, Ramanunni A, et al. CRM1/XPO1 Represents a Promising Therapeutic Target for Treatment of Chronic Lymphocytic Leukemia. ASH Annual Meeting Abstracts. 2011;118(21):232–. [Google Scholar]
  • 17.Miyake T, Pradeep S, Wu SY, Rupaimoole R, Zand B, Wen Y, et al. XPO1/CRM1 Inhibition Causes Antitumor Effects by Mitochondrial Accumulation of eIF5A. Clinical cancer research : an official journal of the American Association for Cancer Research. 2015;21(14):3286–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.US Food and Drug Administration. FDA grants accelerated approval to selinexor for multiple myeloma [press release]. July 3, 2019. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-grants-accelerated-approval-selinexor-multiple-myeloma. Access January 27, 2020.
  • 19.Turner JG, Dawson J, Emmons MF, Cubitt CL, Kauffman M, Shacham S, et al. CRM1 Inhibition Sensitizes Drug Resistant Human Myeloma Cells to Topoisomerase II and Proteasome Inhibitors both In Vitro and Ex Vivo. J Cancer. 2013;4(8):614–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Salas Fragomeni RA, Chung HW, Landesman Y, Senapedis W, Saint-Martin JR, Tsao H, et al. CRM1 and BRAF inhibition synergize and induce tumor regression in BRAF-mutant melanoma. Mol Cancer Ther. 2013;12(7):1171–9. [DOI] [PubMed] [Google Scholar]
  • 21.Mendonca J, Sharma A, Kim HS, Hammers H, Meeker A, De Marzo A, et al. Selective inhibitors of nuclear export (SINE) as novel therapeutics for prostate cancer. Oncotarget. 2014;5(15):6102–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Gilmour LM, Macleod KG, McCaig A, Sewell JM, Gullick WJ, Smyth JF, et al. Neuregulin expression, function, and signaling in human ovarian cancer cells. Clinical cancer research : an official journal of the American Association for Cancer Research. 2002;8(12):3933–42. [PubMed] [Google Scholar]
  • 23.Landen CN Jr., Chavez-Reyes A, Bucana C, Schmandt R, Deavers MT, Lopez-Berestein G, et al. Therapeutic EphA2 gene targeting in vivo using neutral liposomal small interfering RNA delivery. Cancer Res. 2005;65(15):6910–8. [DOI] [PubMed] [Google Scholar]
  • 24.Ahmed AA, Lu Z, Jennings NB, Etemadmoghadam D, Capalbo L, Jacamo RO, et al. SIK2 is a centrosome kinase required for bipolar mitotic spindle formation that provides a potential target for therapy in ovarian cancer. Cancer cell. 2010;18(2):109–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Tanner B, Hasenclever D, Stern K, Schormann W, Bezler M, Hermes M, et al. ErbB-3 predicts survival in ovarian cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2006;24(26):4317–23. [DOI] [PubMed] [Google Scholar]
  • 26.Sanchez-Soria P, Camenisch TD. ErbB signaling in cardiac development and disease. Semin Cell Dev Biol. 2010;21(9):929–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Reif R, Adawy A, Vartak N, Schroder J, Gunther G, Ghallab A, et al. Activated ErbB3 Translocates to the Nucleus via Clathrin-independent Endocytosis, Which Is Associated with Proliferating Cells. J Biol Chem. 2016;291(8):3837–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Cook RS, Garrett JT, Sanchez V, Stanford JC, Young C, Chakrabarty A, et al. ErbB3 ablation impairs PI3K/Akt-dependent mammary tumorigenesis. Cancer Res. 2011;71(11):3941–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Lapalombella R, Sun Q, Williams K, Tangeman L, Jha S, Zhong Y, et al. Selective inhibitors of nuclear export show that CRM1/XPO1 is a target in chronic lymphocytic leukemia. Blood. 2012;120(23):4621–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Xu F, Yu Y, Le XF, Boyer C, Mills GB, Bast RC, Jr. The outcome of heregulin-induced activation of ovarian cancer cells depends on the relative levels of HER-2 and HER-3 expression. Clinical cancer research : an official journal of the American Association for Cancer Research. 1999;5(11):3653–60. [PubMed] [Google Scholar]
  • 31.Krivosheya D, Tapia L, Levinson JN, Huang K, Kang Y, Hines R, et al. ERBB4-neuregulin signaling modulates synapse development and dendritic arborization through distinct mechanisms. J Biol Chem. 2008;283(47):32944–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Mei L, Xiong WC. Neuregulin 1 in neural development, synaptic plasticity and schizophrenia. Nat Rev Neurosci. 2008;9(6):437–52. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1600893-supplement-1.docx (1.7MB, docx)

RESOURCES