ⅠINTRODUCTION

Adenocarcinoma NOS(AdCa NOS) is the third most common malignant tumor of the salivary glands which is characterized by unique clinical features and behavior, including very high frequencies of recurrence and distant metastasis of AdCa NOS^1-2). Because current therapy for AdCa NOS does not significantly improve survival rates, investigating new therapeutic modalities against AdCa NOS is necessary³⁾. Although the standard treatment for major malignant salivary gland tumors including AdCa NOS is surgical resection, a substantial risk of local-regional recurrence and distant metastasis exists^4-7). Although adjuvant radiation has been used in select cases in order to reduce the local-regional failure rate⁸⁾, the high rate of distant metastasis has not been improved by these local therapy. Although chemotherapy has been used for unresectable tumors and to palliate local-regional recurrence, the role of adjuvant chemotherapy has not been clearly defined in preventing distant metastasis. Previous series of patients with malignant salivary gland tumor showed that receipt of both adjuvant radiation and chemotherapy was an independent predictor of an improved disease-free survival⁹⁾ because distant metastasis was a more common as opposed to local-regional recurrence¹⁰⁾. The use of chemotherapy including cisplatin, doxorubicin, and 5-fluorouracil for recurrent and/or unresectable malignant salivary gland tumor has been tested ^11-12). Combination chemotherapy including cisplatin, doxorubicin, and 5-fluorouracil, mitomycin C, cyclophosphamide, methotrexate, bleomycin, paclitaxel, vincristine, and vinorelbine has also demonstrated a tumor response^11-15). Although Docetaxel(TXT) has significant antitumor effects and is currently being tested in patients with head and neck squamous cell carcinoma^16-17), TXT has not been tested in human salivary gland AdCa NOS. TXT is an antimicrotubulin agent that promotes tubulin assembly, inhibits depolymerization, acts as a mitotic spindle poison, and induces mitotic block in proliferating cells¹⁸⁾. TXT induces G2-M arrest and p53-independent apoptosis in various cancer cell lines. However, the antitumor effects of TXT in human malignant salivary gland tumor model including AdCa NOS have not been evaluated.

The purpose of this study was to examine the effects of TXT on proliferation, cell cycle progression, connexin 43 expression, apoptosis, and FasR expression in human salivary gland SGT cell line.

IIMATERIALS and METHODS

1Cell Culture

TXT(Aventis Pharmaceuticals, France) was suspended in ethanol at 10 mg/ml and stored at deepfreezer. Dilutions were prepared in culture medium to make final concentrations of 5, 10, 20, 40, 60, 80, and 100 ng/ml. The final ethanol concentration was <0.05% by volume. The SGT cell line was maintained in vitro in DMEM supplemented with 10% FBS(Hyclone, USA), 0.5mM sodium pyruvate, 2mM L-glutamate, 0.1mM MEM nonessential amino acids, 100units/ml penicillin, and 100μg/ml streptomycin in a 5% CO₂ incubator.

2Proliferation Assay

Proliferation rates were determined by measuring BrdUrd incorporation using the BrdUrd Proliferation Assay Kit(Oncogene Research Products, USA). Quadruplicate wells of SGT cells were plated in flat-bottomed 96-well plates at 10,000 cells/well in 100μl of media until cells reached 60.70% of confluence. TXT(10, 20, 40, 60, or 80ng/ml) or solvent alone was added to fresh media (total volume of 200μl/well). SGT cells were assayed at 24, 48, 72, and 96 hrs after TXT exposure. At each time point, wells were pulsed for 2 hrs with BrdUrd to allow for incorporation into DNA, lysed, and processed by this ELISA-based assay system. Absorbance in each well was measured using dual wavelengths of 450-540nm. The average absorbance reading of the quadruplicate wells for each concentration was expressed as a percentage of the untreated control for each time point.

3Cell Cycle Analysis

2x10⁵ cells were plated in 6-well plates in 2ml of media and subcultured(36-48 hrs) until they reached 60-70% confluence. Fresh medium was added that contained different concentrations of TXT or solvent alone. Cells were cultured in the presence of drug for an additional 48-72 hrs before harvest. Cells were washed with cold PBS and concentrated(1x10⁶ cells in 0.1 ml of PBS). An equal volume of ice-cold ethanol was added while vortexing the sample. The sample was washed with PBS and centrifuged for 30min at 1500 rpm, resuspended in 0.3ml of PBS containing 1mg/ml DNase-free RNase I, and incubated at 37°C for 30 min before the addition of propidium iodide(20 μg/ml final concentration). Cell cycle distribution was measured with a FACs Caliber flow cytometer(Becton Dickinson, USA). At least 10,000 events per sample were acquired. The cytological software program was used for data analysis.

4Apoptosis

An annexin V-FITC apoptosis detection kit(BD PharMingen, USA) was used for apoptosis detection. After TXT treatment, attached cells were treated briefly with trypsin, quenched with media, washed once with cold PBS, and then resuspended in chilled binding buffer and annexin V-FITC according to the kit instructions. After the 15 min incubation at room temperature, propidium iodide was added to label the nonviable cells. Samples were placed on ice and evaluated immediately by dual-color flow cytometry. A total of 1x10⁴-2x10⁴ events were collected per sample. Data were analyzed.

5Evaluation of Cell Surface FasR Expression by Flow Cytometry.

FITC-conjugated hamster IgG anti-FasR(clone Jo2; PharMingen) was used to detect cell surface expression of the FasR in SGT cells after TXT treatment. Samples were stained with FITC-conjugated anti-FasR antibody or isotype control and diluted 1:40 in staining buffer(PBS supplemented with 2% serum and 0.1% sodium azide) on ice for 45min. Samples were analyzed by flow cytometry using a FACs Caliber flow cytometer. To determine whether the FasR agonist antibody Jo2 could further enhance apoptosis through the FasR pathway, SGT cells were subcultured in 6-well plates as described above and treated for 48 hrs with TXT at 0, 10, or 20ng/ml in duplicate wells. One set of wells was treated with an isotype control antibody, and the other set received 5 μg/ml anti-Fas agonist antibody Jo2 and incubated for an additional 20 hrs. Cells were harvested and processed for annexin V-FITC binding and apoptosis.

5Immunoslot Blot Analysis

The cultured cells were harvested with 0.1M Trisacetate( pH 7.5), 1mM EDTA, containing protease inhibitors (5g/ml leupeptin, 5g/ml aprotinin, 50g/ml calpain inhibitor I, 100g/ml bestatin, and 1mM PMSF). Protein amount of those extracts was quantified at 495nm with spectrophotometer, serially diluted in the range of 10μg to 1μg, and applied into slot chambers containing nitrocellulose membrane. Slot chambers were twice washed with TBS. The membrane was processed as the following western blotting methods. The membrane was treated with blocking serum, incubated with primary antibodies and secondary antibodies. The antibodies for the antigens were purchased from the following sources: monoclonal antibody CX-1B1 diluted 1:1000(0.5g/ml) in TBST 1% BSA, followed by goat antimouse horseradish peroxidase(Jackson Immunoresearch) diluted 1:10,000 in TBST 1% BSA. are commercially available(Research and Diagnostic Antibodies, USA). This was developed by an enhanced chemino luscence(ECL) method, and examined by densitometer in triplicate. The protein concentrations were determined by Bio-Rad Protein Assay.

ⅢRESULTS

All in vitro analyses were performed on SGT cell line. Dose response for TXT with respect to growth suppression, G2-M cell cycle arrest, and apoptosis was found. After exposure of SGT cells to TXT, proliferation was suppressed significantly(P<0.05) at concentrations higher than 20ng/ml after 96 hrs(Fig. 1). Proliferation was little changed after 10ng/ml TXT exposure, but cellular proliferation was inhibited according to increasing concentration of TXT and time.

G2-M arrest stage showed about up to 5 fold increase after exposure of TXT and anti-Fas agonist by flow cytometry. Growth suppression of 75% was noted at concentrations of 40ng/ml. SGT cells were arrested in G2-M by TXT in a dose-dependent manner(Fig. 2). Therefore, growth inhibition of the SGT cell line was predominately because of G2-M arrest. It suggested that Inhibition of proliferation and G2-M cell cycle arrest of SGT could be induced by TXT.

Apoptosis index showed about up to 8 fold increase after exposure of TXT and anti-Fas agonist by flow cytometry. Fas expression showed about up to 3 fold increase after TXT exposure and anti-Fas agonist by flow cytometry(Fig. 3).

A dose-dependent increase in apoptosis(percentage of annexin binding) induced by TXT in SGT cells was observed (Fig. 4). An increase in the level of FasR expression on the cell surface also paralleled apoptosis(Fig. 4), while FasR was expressed at low to undetectable levels in SGT cells and was not increased by TXT. To determine the functional consequence of Fas up-regulation by TXT, apoptosis was measured in SGT cells after anti-Fas agonist antibody was added to TXT-pretreated cells. At low concentrations(10 and 20ng/ml) of TXT, the addition of the agonist Fas antibody showed more than doubled the level of apoptosis as compared with treatment with TXT alone(Fig. 5).

Up-regulation of FasR with lower doses of TXT rendered cells susceptible to FasR agonist antibody-mediated apoptosis. In the absence of TXT, anti-FasR antibodies were completely without effect, which meant that TXT was critical for priming apoptosis mediated through the Fas pathway. Apoptosis showed about up to 3 fold increase at 20ng/ml after exposure of TXT and anti-Fas agonist by flow cytometry. It suggested that Induction of apoptosis by TXT could coincide with an increase in cell surface FasR expression.

In order to establish stable gap junctions, the effects of microtubule stabilization by TXT with Cx43 expression were examined. Because functional coupling requires connexin expression and membrane localization, Cx43 protein expression after TXT exposure was analyzed(Fig. 6). Furthermore, TXT treatment increased total cellular Cx43 protein levels by Immunoslot blot analysis, while In untreated SGT cells, total Cx43 expression was lower(Fig. 6). SGT cells treated with 20ng/ml TXT after 48 hrs exhibited a marked enhancement in Cx43 expression(Fig. 6). Thus, it suggested that TXT could induce intercellular coupling by the connexin pathway via an increase in Cx43 protein levels, and low levels of TXT treatment on subconfluent SGT cultures restore cell contact inhibition and reduce the saturation density of the culture.

VDISCUSSION

The characterized SGT cell line was used as model system of malignant salivary gland pathogenesis so that the pathogenesis of AdCa NOS contributing to its tumorigenic phenotype could be understood and putative treatment modalities be considered¹⁹⁾. Although surgery and radiation therapy for advanced and high-risk malignant salivary gland tumors have been performed, patients have high rates of distant metastasis and local-regional recurrence^4-6). High-risk pathological features have been associated with poor prognosis and a high rate of local-regional and distant metastasis⁹⁾. Because the role of chemotherapy in recurrence of malignant salivary gland tumors has been established11-15), the use of adjuvant chemotherapy for high-risk patients is being tested. Previous studies demonstrated that adjuvant chemotherapy and radiation therapy may improve disease-free survival⁹⁾.

The PCNA index in the carcinomatous tissue was approximately 50%, whereas 10% in the adjacent normal parotid acinar and ductal cells. PCNA index in high-grade malignant parotid tumors has been correlated with higher local recurrence and mortality²⁰⁾. This is in line with clinically relevant concentrations²¹⁾ and clearance kinetics observed in patients22). Although the antiproliferative effect of TXT was secondary to G2-M arrest and induction of apoptosis in SGC1 cells, similar to other cancer types²³⁾, TXT as a newer chemotherapy agent has not yet been tested in human malignant salivary gland tumors, especially AdCa NOS.

In this study, TXT exposure sensitized SGT cells to anti-Fas agonist antibody-mediated apoptosis partly due to an up-regulation in cell surface Fas expression. Because enhancement of this pathway of apoptosis might also be related to the effect of TXT on cell cycle and stress response pathways, it was thought that SGT cells could be more vulnerable to Fas mediated apoptosis or alterations in the transport of Fas due to microtubule stabilization. Therefore, it suggested that TXT induced apoptosis(Annexin Binding) and Fas expression in SGT cells through the Fas pathway, and the addition of an anti-Fas agonist antibody augmented TXT-induced apoptosis. TXT promote apoptosis by altering expression of the Bcl-2 family and signal transduction pathways of mitogen-activated protein kinase(c-Jun NH2-terminal kinase/stress- activated protein kinase) and by increasing secretion of cytotoxic cytokines⁹⁾. Sensitization to Fas-mediated cytotoxicity has been observed with cisplatin and 5-fluorouracil in human bladder cancer cells^24,25), camptothecin in prostate cancer cells²⁶⁾, and all-trans retinoic acid and cisplatin in medulloblastoma cells²⁷⁾. Although these effects were mediated by modulation of Fas expression in some instances, most studies reported that drug-induced apoptosis was not mediated by the Fas/FasL signaling pathway^28-29). Paclitaxel was shown to augment FasL-induced apoptosis in glioma cells mediated by bcl-2 regulation of the Fas pathway³⁰⁾. TXT has previously been found to induce apoptosis by bcl-2 phosphorylation³¹-32), which leads to the inactivation of its antiapoptotic signal and increased apoptosis. Although the Bcl-2 apoptosis pathway was not examined, the overall incidence of TXT might induce apoptosis in SGT cells probably including significant bcl-2 and bax components, as it does in other tumor types³¹⁾. The more novel up-regulation of FasR by TXT and sensitization to apoptosis through this pathway might have additional implications for therapy of AdCa NOS with FasR expression in this study.

Increased connexin 43 expression in SGT cells after TXT exposure may constitute a novel therapeutic target of TXT. Connexins are proteins that form gap function channels. GJIC(Gap junctional intercellular communication) plays a crucial role in cellular homeostasis, and alterations in GJIC are found during tumor promotion and progression³³⁾. GJICs link single cells within a population to facilitate the passage of small regulatory molecules and cytoplasmic ions to maintain metabolic homeostasis. Loss of GJIC and/or down-regulation and inactivation of connexins have been observed in a variety of cancers³³⁾. After treatment with TXT, increased Cx43 expression meant that it might be localized at the plasma membrane of cell-cell junctions, and more abundant Junctional plaques was uniformly dispersed along the entire cell-cell borders. it was reported that TXT increases GJIC by increasing the expression of Cx43 in murine salivary gland carcinoma cells³³⁾. In glioblastoma cells, Cx43 reexpression enhanced paclitaxel-induced apoptosis34), which meant that connexin gene transduction resulted in reversion of neoplastic phenotype and inhibition of tumorigenic potential in experimental models³⁵⁾. It suggested that GJICs stabilization of SGT cells after TXT exposure could be concomitant with increased Cx 43 expression.

Although limited TXT data exist concerning in vitro effects of TXT in salivary gland AdCa NOS, it is tempting to speculate that the effects of TXT as an antimicrotubulin agent on GJIC, FasR expression, cell cycle arrest, and apoptosis could be all interrelated. Understanding the relationships among these pathways could lead to improved strategies that exploit these mechanisms for therapeutic intervention. In addition to the antitumor mechanisms of TXT including induction of growth arrest, G2-M arrest, and apoptosis, TXT promoted apoptosis in part through the Fas pathway in SGT cells. These results suggested that TXT might be appropriate for additional in vivo studies and clinical trials in patients with malignant salivary gland tumors because TXT contributed to its antitumor activity in SGT cells. With its aggressive features, SGT cells will be an appropriate model to test drug efficacy because AdCa NOS could be managed by surgery and chemotherapy, similar to the more common high-grade mucoepidermoid and adenoid cystic carcinomas in future.