Y. Yu, W. Chen and Y. Zhang contributed equally to this work.
Cancer Cell Biology
Suppression of salivary adenoid cystic carcinoma growth and metastasis by ErbB3 binding protein Ebp1 gene transfer
Article first published online: 31 JAN 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 9, pages 1909–1913, 1 May 2007
How to Cite
Yu, Y., Chen, W., Zhang, Y., Hamburger, A. W., Pan, H. and Zhang, Z. (2007), Suppression of salivary adenoid cystic carcinoma growth and metastasis by ErbB3 binding protein Ebp1 gene transfer. Int. J. Cancer, 120: 1909–1913. doi: 10.1002/ijc.22541
- Issue published online: 28 FEB 2007
- Article first published online: 31 JAN 2007
- Manuscript Accepted: 24 NOV 2006
- Manuscript Received: 8 NOV 2006
- Key Project of National Natural Science Foundation of China. Grant Number: 30330580
- Shanghai Leading Academic Discipline Project. Grant Number: Y0203
Ebp1, an ErbB3 binding protein and downstream effector of the ErbB signaling network was shown to be a potent tumor suppressor in breast and prostate adenocarcinomas. We hypothesized that the inhibitory properties of the ebp1 gene could also be beneficial if ectopically expressed in salivary adenoid carcinoma. Salivary adenoid carcinoma cell line (ACC-M) cells were stably transfected with the full-length ebp1 cDNA sequence or the empty expression vector pcDNA3.1. Stable gene transfer was verified by Western blot analysis and reverse transcription (RT)-PCR. A significant reduction in cell proliferation, anchorage-independent growth, and a change in the cell cycle profile was observed in ebp1 transfectants. Matrigel assays demonstrated that the adenoid cystic carcinoma cell invasiveness was significantly reduced. A strong decrease in the metastatic potential of human adenoid cystic carcinoma cells in an experimental metastatic model was also observed. Our results suggest that ectopic expression of Ebp1 mediates multiple antitumor activities against adenoid cystic carcinoma cells and that ebp1 gene therapy might be a viable method suppressing malignant salivary adenoid tumors. © 2007 Wiley-Liss, Inc.
Adenoid cystic carcinoma of salivary gland origin is a highly aggressive neoplasm that is characterized by insidious invasion into adjacent tissue and spread to distant organs at early stages of the malignancy.1 Curative surgery has failed to reduce the overall mortality rate over the past several decades.2, 3 Radiation and chemotherapy are not beneficial for patients with advanced disease4, 5 and also increase the risk of developing a second neoplasm of the salivary gland.6 Thus, effective agents for treatment of adenoid cystic carcinoma are urgently needed.
Therapeutic agents targeting epidermal growth factor receptor and its downstream signaling effectors in some head and neck cancer are now being developed.7 However, the presence and role of EGFR in the clinical development and progression of adenoid cystic carcinoma is not clear. With limited studies on the immunohistochemical detection of EGFR,8, 9, 10 only 1 report showed that most salivary gland ACC specimens stained positively for EGFR.10 As part of a Phase II trial of trastuzumab for treatment of incurable salivary gland cancer, Glisson's group screened 137 salivary gland carcinomas for HER2 expression. The low frequency of HER2 overexpression in the screened population, especially in the most common subtype of adenoid cystic carcinoma (4%, 3 of 70), suggests that trastuzumab will not have a major role in treatment of salivary gland cancers of intercalated duct origin including adenoid cystic carcinoma.11
Examination of other tyrosine kinase receptors or their downstream transduction pathways might offer new therapeutic options. For example, keratinocyte growth factor receptor gene therapy inhibits growth of human salivary adenocarcinoma cells through induction of differentiation and apoptosis.12 Ebp1, an ErbB3 binding protein and a downstream effector of Heregulin activated signaling, was shown to be a potent tumor suppressor in adenoid epithelium carcinomas, such as breast and prostate adenocarcinomas.13, 14, 15 We hypothesized that the inhibitory properties of the ebp1 gene could be beneficial if ectopically expressed in adenoid cystic carcinoma cells of salivary gland epithelial origin.
Material and methods
The human salivary adenoid cystic carcinoma cell line ACC-M, established by the Department of Oral and Maxillofacial Surgery, Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine, was used in this experiment. The ACC-M cell line is highly metastatic to the lung.16 Cells were maintained at 37°C in a humidified atmosphere of 5% CO2 in air and routinely propagated in RPMI1640 (GIBCO, USA) supplemented with 10% fetal bovine serum (FBS) (Sigma, USA).
The pcDNA3.1 control vector and pcDNA-ebp1 plasmid were supplied by Dr. Hamburger (University of Maryland, USA) and have been described previously.16
Creation of stably transfected cell lines
To establish stable mass transfectants, subconfluent ACC-M cells in 100-mm dishes were transfected with 3 μg of empty pcDNA3.1 vector or 1 and 3 μg of pcDNA-ebp1 using Lipofectamine 2000 (Invitrogen, USA). Cells were selected in G418 (400 μg/ml) for 4 weeks. Individual resistant colonies were cylinder cloned and expanded. The positive colonies thereafter were named ACC-M-pcDNA3.1 (ACC-M0), ACC-M-ebp1-1 μg (ACC-M1) and ACC-M-ebp1-3 μg (ACC-M3).
Verification of stable gene transfer
Western blot analysis was performed as previously described.17 Briefly, total cell lysates were prepared by incubating the cells in RIPA buffer [150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.4% sodium dodecyl sulfate (SDS), 20 mM ethylenediamine tetraacetic acid and 50 mM Tris, pH 7.4] for 1 hr at 4°C. After sonication and centrifugation, equal amounts of protein from each sample were subjected to 4–15% SDS-polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane (Bio-Rad, Hercules, CA). The membrane was blocked with 5% nonfat milk in Tris-buffered saline and incubated with the primary antibodies against Ebp1 (Oncogene, USA). A secondary antibody, horseradish peroxidase-conjugated immunoglobulin G, was incubated with membranes and developed according to Amersham's enhanced chemiluminescence protocol (ECL; Amersham).
Reverse transcription PCR.
Total RNA was extracted from transfected and untransfected cells using the RNeasy Mini Kit (Qiagen,Valencia,CA). Reverse transcription (RT) was done with 1 μg of total cellular RNA as per the manufacturer's instructions (SuperScript preamplification system, Life Technologies). The primers derived from the ebp1coding sequence (5′GAAGCCTCACCTCCCAAA3′, 5′CCATTATCCAGAATCCCACA3′) were used to amplify the ebp1transcripts. Primers for actin internal control were 5′GCTGGCCGGGACCTGACTGACTAC 3′, 5′GGGGGCACGAAGGCTCATCATT3′. PCR amplification was done on a Perkin-Elmer DNA cycler 480 for 35 cycles with denaturing at 94°C for 30 sec, annealing at 55°C for 30 sec and extension at 72°C for 30 sec.
In vitro effects of ebp1 gene transfer on cell growth, proliferation and survival
Triplicate samples of 2 × 103 log phase cells were plated in 96-well tissue culture plates. Cell growth was examined daily for 9 consecutive days by adding 20 μl of 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyl-tetrazolium bromide (Sigma-Aldrich, USA) to each well and incubating for 4 hr at 37°C. The supernatant was then discarded, and 150 μl of DMSO (Sigma-Aldrich, USA) were added. The absorbance was determined by spectrophotometry (Biohit, BP800, Finland) using a wavelength of 570 nm with 630 nm as a reference.
Proliferation rate was assessed by measuring the incorporation of 3H-thymidine into newly replicated DNA. Cultures were pulsed with 5 mCi/ml 3H-thymidine (NEN Life Science Products, Boston, MA) for 6 hr. Cells were harvested onto filters and dried for 4 hr before a scintillation cocktail was added. LS6500 Multi-Purpose Scintillation Counter (Beckman, USA) was used to determine the amount of 3H-thymidine incorporated into the dividing cells.
Anchorage-independent cell growth.
Anchorage-independent cell growth was determined by analyzing colony formation of cells in soft agar. Cells (5000 cells /well) were suspended in 0.33% agar in RPMI1640 containing 10% FBS and seeded on agar (0.5%) in triplicate of individual wells in 6 well tissue culture plates. Colonies were scored after 2 weeks of culture, fixed with 70% ethanol and stained with Coomassie Blue. Colonies of more than 50 cells were counted under a dissecting microscope and calculated as number of colonies counted/number of cells seeded %.
Cell cycle distribution
Cells were trypsinized and washed twice with PBS. Cells were suspended, fixed in 70% ethanol and stained with Cycle TEST™ Plus DNA Kit (Becton Dickinson, USA). Cell cycle distribution was analyzed with a FACSCalibur flow cytometer (Becton Dickinson, USA).
In vitro effects of ebp1 gene transfer on cell invasion
A transwell system with a polycarbonate filter membrane 6.5 mm in diameter and an 8 μm pore size (Corning, NY) was used to assess the rate of cell invasion. Matrigel invasion chambers were primed by following the manufacturer's instructions. 1 × 105 cells in 200 μl RPMI 1640 media with 1% FBS were seeded in the upper chamber of Matrigel plate and the lower chamber was filled with 600 μl of the same medium with 10% FBS to act as a chemoattractant. After 24 hr incubation at 37°C, the cells on the upper surface of the filter were completely wiped off with a cotton swab. The cells that penetrated to the lower surface of the filter were stained with hematoxylin-eosin and counted under a microscope (Olympus Ix70) in 10 randomized fields at a 400× magnification. The data were expressed as the average number (±SD) of cells from 10 fields that migrated to the lower surface of the filter from each of 3 experiments performed.
Analysis of metastatic and invasive phenotype of Ebp1 transfected ACC-M cells
Spontaneous lung metastasis examination.
Animal experiments were performed in accordance with Institutional Animal Care and Use Committee procedure and guidelines for animal welfare. All transfectants, grown in complete medium with 400 μg/ml G418, were harvested and resuspended in phosphate buffer saline (PBS) at 1.5 × 107/ml. Twenty-four 5–6 weeks-old male athymic nude mice (BALB/C-nu/nu, Animal Institute, School of Medicine, Shanghai Jiao Tong University) were randomized into 3 groups of 8 each. Injection of 200 μl of ACC-M0 or ACC-M1 or ACC-M3 cell suspension into the lateral tail vein of nude mice was performed as previously described.18 Six weeks later when animals developed palpable lung tumors, dehydration, emaciation or weight loss > 20% of initial body weight, mice were killed and gross macroscopic examination was carried out. Lungs were removed and weighed. Tumor distribution in lung area was determined with a Carl Zeiss Axioplan microscope (Carl Zeiss, Thornwood, NJ). Generated images and the ratio of tumor/lung areas was analyzed and calculated with commercial software KS400 Ver3.0.
The lungs with tumors were fixed in 10% buffered formalin for 6 hr. After fixation, the tissue samples were processed into paraffin blocks. Tissue sections (5 μm) were obtained from the parablocks and stained with hematoxylin-eosin by standard histological procedure. Tissues were also subjected to immunostaining for Ebp1.
For all of the experiments, results were analyzed using a 2-tailed Student's t-test. Significance was established at p < 0.05.
To confirm whether the cell line transfected with ebp1 expressed the transgene, we carried out RT-PCR to compare the expression of the ebp1 gene versus the housekeeping gene actin using mRNA isolated from transfected and untransfected cells. Stronger bands of the expected 789 bp size were seen from the ebp1 transfected cells ACC-M1 and ACC-M3. A weak band was present from untransfected cells ACC-M and empty vector transfected ACC-M0 cells because of endogenous ebp1 mRNA (Fig. 1a). Western blot analysis further showed that Ebp1 protein levels in these stably transfected cells were enhanced approximately 4-fold compared to those in the empty vector transfected cells (Fig. 1b). These data verified that the transgene was expressed in the stably transfected cell lines, consistent with a previous report.17
To determine if Ebp1 could be an effective therapeutic agent for adenoid cystic carcinoma, the effect of ectopic expression of Ebp1 on cell growth was examined. Figure 2a shows the results of representative MTT assays; Ebp1 potently inhibited growth of ebp1 transfectants ACC-M1, ACC-M3 cells compared with control vector transfected groups (p < 0.05). To further extend this observation, we tested the antiproliferative effect of Ebp1. 3H-thymidine incorporation rates were significantly blocked in the presence of enhanced Ebp1 expression in ACC-M1and ACC-M3 (Fig. 2b), compared with the parental cells (p < 0.05). These findings suggest that one mechanism by which Ebp1 inhibits cell growth is to block new DNA synthesis.
To examine if the inhibitory effect observed in the cell growth assays reflects a delay or arrest of cell cycle progression, cell cycle distribution was evaluated. Ebp1 overexpression increased the proportion of cells in the G0/G1 and G2/M phases in the ebp1 stably transfected cell lines ACC-M1 and ACC-M3, with a corresponding decrease in the proportion of cells in the S phase compared with the vector transfected cell line ACC-M0. In contrast, no such change in cell cycle distribution was detected when comparing empty vector transfectants to the parental cells (Fig. 3)
Given that Ebp1 regulates growth of ACC-M cells, we evaluated the ability of Ebp1 over expression to reverse the malignant phenotype as assessed by anchorage-independent growth in soft agar. As an in vitro model of tumorigenesis, the soft agar assay is widely used to test characteristic changes after potential therapeutic agent treatment. Ebp1 stably transfected cell lines exhibited a >50% reduction in the number of anchorage-independent colonies in soft agar compared with the parental and vector transfected counterparts (p < 0.05, data not shown).
We next analyzed if the reduced ability of the ebp1 ACC-M transfectants to metastasize correlated with their ability to invade through the basement membrane, an essential step in the metastatic cascade. To that end, 1 × 105 ACC-M, ACC-M0, ACC-M1 and ACC-M3 cells were placed in the upper compartment of an invasion chamber. After 24 hr of incubation, the cells that penetrated to the low surface of the filter were stained with hematoxylin-eosin and counted under a microscope (Olympus Ix70) in 10 randomized fields at a 400× magnification. ACC-M1 and ACC-M3 cells transfected with ebp1 showed significantly reduced ability to invade through Matrigel-coated filters than ACC-M parental cells and ACC-M0 cells (Table I). This in vitro analysis indicated that overexpression of Ebp1 in ACC-M cells inhibited the ability of these cells to penetrate extracellular matrix, an important component in the process of tumor invasion and spread, thus accounting partly for the decrease in their lung metastatic potential.
|Cell line||No. of migrated cells ± SD|
|ACC-M||34.7 ± 4.57|
|ACC-M0||33.1 ± 3.90 (0.4108)*|
|ACC-M1||8.5 ± 4.55 (<0.0001)*|
|ACC-M3||9.2 ± 4.10 (<0.0001)*|
To directly examine the effect of Ebp1 expression on cell invasion, we injected ebp1 transfectants ACC-M1, ACC-M3 and the vector transfectant ACC-M0 cells into the mouse lateral tail vein using a previously published experimental metastasis assay.18 The experiment was terminated when the control group developed symptoms of respiratory disease and cachexia. Two mice in the control group died of severe respiratory disease before terminating the experiment. The lungs from both control pcDNA3.1 vector and ebp1 transfected groups were harvested and analyzed grossly and microscopically as described in “Materials and Methods.” Pathological gross examination revealed that large and multiple tumor nodules replaced the lung parenchyma in mice bearing vector pcDNA3.1 transfected ACC-M0 cells. These tumors appeared white and indurated, consistent with our previous finding that ACC-M cells are highly metastatic to the lungs.18Conversely, mice bearing tumors derived from ebp1 transfected cells had smaller and fewer tumor nodules in the lungs, showing uninvolved areas of pink-red crepitant parenchyma (Fig. 4a). To quantify the effects of Ebp1 on ACC-M metastasis, the area of metastasis was measured with a Carl Zeiss Axioplan microscope equipped with a computer-controlled image analysis system. The percentage of tumor to lung surface area was 7.87% for mice inoculated with ACC-M1 cells and 6.72% for mice inoculated with ACC-M3 cells. This was significantly different (p = 0.0001) than the 36.77% value for mice injected with vector control cells, indicating ebp1 transfectants formed fewer metastases than control cells (Fig. 4b). Histopathological examination of normal lung(Fig. 4c-a) and tumors confirmed these as metastatic salivary adenoid cystic carcinoma, characterized by cells exhibiting low differentiation with much mitosis, less cytoplasm and increased nuclei (Fig. 4c-b). Mice injected with ebp1 transfectants ACC-M1 and ACCM3 showed substantially reduced tumor presence in the lung tissue (Fig. 4c-c,d) compared with mice injected with control vector transfectant ACC-M0 (Fig. 4c-b). At the same time, lung weight was also adopted as a surrogate indicator of the metastatic status of the mice. We found an average lung weight of 0.793 g in the vector control mice, whereas the average lung weight was only 0.222 and 0.202 g in the epb1 transfected cell lines ACC-M1 and ACC-M3, representing a decrease of 72.0 and 74.5%, respectively (p = 0.001).
Our aim was to find a new molecular tool to be developed as a novel treatment strategy for primary and especially for recurrent salivary adenoid carcinoma. Thus, we tested if the inhibitory properties of the ebp1 gene could be beneficial if ectopically expressed in salivary adenoid carcinoma. Ebp1, a member of the PA2G4 family and the human homologue of a previously identified proliferation regulated mouse protein, was isolated as an ErbB3 binding protein.19 Ectopic expression of Ebp1 suppresses the growth of both ErbB positive breast13 and prostate cancer cells14 and induces cellular differentiation.13 Treatment of AU565 breast cancer cells with the ErbB-3/4 ligand heregulin (HRG) results in dissociation of Ebp1 from ErbB3 and translocation from the cytoplasm to the nucleus, suggesting that Ebp1 is a downstream mediator of heregulin (HRG)-activated signaling.19 Recent studies further indicate that Ebp1 suppresses androgen receptor-mediated gene transcription and tumorigenesis of prostate cancer LNCaP cells in a xenograft mouse model.15 For the first time, we clearly demonstrate that ebp1 gene transfer also resulted in pronounced antitumor activity in adenoid cystic carcinoma cell line, revealing a previously unrecognized therapeutic gene for treatment of another gland epithelial neoplasm––adenoid cystic carcinoma. We also demonstrate for the first time that ectopic expression of Ebp1 results in reduced metastatic potential.
The long-term survival of adenoid cystic carcinoma is very poor because of its insidious metastasis, especially to lungs, even after extensive surgery.20 However, metastasis is an extremely inefficient process,21 with many variables such as hemodynamic forces22 and the immune system23 preventing cells shed from a primary location from forming secondary tumors. The proportion of cells that can initiate growth after extravasation and maintain growth is an important determinant of metastatic outcome.24 Ebp1 exerted strong antiproliferative effects by blocking new DNA synthesis as evaluated by 3H-thymidine incorporation assay and reducing the cell growth rate under anchorage -independent conditions. These actions may contribute to a low efficiency of cell survival in the circulation once the ebp1 stably transfected ACC-M cells are injected into the tail vein and fewer surviving solitary cells are distributed in the region of each lung structure. This initial hypothesis was proved by the fact that there is less tumor metastasis in mice inoculated with ebp1 stably transfected ACC-M cells than in mice injected with vector transfected ACC-M0 cells. Matrigel assays corroborated that the reduced invasiveness of ebp1 transfected ACC-M cells contributes in part to the decreased number of disseminated tumor nodules in the lung. Additional studies are needed to identify the specific cell adhesion molecule(s) responsible for this process.
The cell cycle is a basic mechanism that modulates fundamental processes such as growth and proliferation.25 While growth factors play a crucial role in cell cycle progression,26 the expression and biological function of HRG in salivary adenocarcinoma are yet to be determined. Ectopic expression of Ebp1 in adenoid cystic carcinoma cells appeared to mimic the differentiating effects of HRG as shown by increased proportion of cells in the G0/G1 population as well as G2/M arrest with a concomitant decrease of the S phase. A large body of evidence indicates the pivotal role of the RB-E2F axis in the G1-S transition and its aberration in cancer cells which drives continuous cycling and cell division, leading to uncontrolled proliferation that is a prerequisite for most if not all malignant tumors.27 Ebp1 has been shown to associate with the retinoblastoma protein (Rb) and histone deacetylase 2 in cultured cells and inhibits transcription of both endogenous and transiently transfected E2F-regulated promoters such as cyclin D, cyclin E, E2F and c-Myc28, 29 important in cell cycle regulation. In addition, cyclin E was recently found to be a strong independent prognostic indicator of metastasis and survival in patients with early stage of nonsmall cell lung cancer. As the ability of Ebp1 to bind E2F promoter elements and repress E2F mediated transcription was regulated by HRG,30 an Ebp1-mediated signal-attenuating mechanism is envisaged as one major mechanism for the effects of ebp1 gene transfer in the human salivary adenoid cystic carcinoma ACC-M cells.
We have demonstrated that wild-type ebp1 gene transfected into human salivary adenoid cystic carcinoma cell line significantly inhibits cell growth in in-vitro assays and most importantly, reduces tumor metastatic potential in an experimental metastatic mouse model. It is therefore possible that ebp1 based gene therapy might be effective for salivary adenocarcinoma as adjuvant treatment––for example, administered directly at surgically cut margins. Success of an ebp1-based gene therapy in an orthotopic model, currently under investigation in this laboratory, will pave the way for Phase I clinical trials.
- 2Metastatic adenoid cystic carcinoma of salivary glands: case reports and review of the literature. Cancer Causes Control 1996; 3: 336–42., , , ,
- 4Adenoid cystic carcinoma of the salivary glands: a 20-year review with long-term follow-up. Ear Nose Throat J 2005; 84: 662, 664–7., , , , ,