Tissue samples from seven patients with ovarian carcinoma, pathologically diagnosed as serous adenocarcinoma (n = 5; mean age 54.8 yr), endometrioid carcinoma (n = 1; 39 yr), or mucinous carcinoma (n = 1; 59 yr), and five non-malignant ovarian tissues (mean age 45.3 yr, undergone surgery for ovarian cysts) were obtained from Imam Khomeini Hospital (Tehran, Iran) (Table 1). Each individual signed an informed consent and all aspects of this study were approved by Avicenna local ethics committee. After surgical resection, each fresh tissue specimen was immediately frozen in liquid nitrogen for further study. Tissue sections were taken from each sample, stained with Hematoxylin and Eosin (H&E), and examined by a pathologist to confirm their path-ological state.

The ovarian carcinoma cell lines including Caov-4 (HTB-76), OVCAR-3 (HTB-161), SKOV-3 (HTB-77) (ATCC, Manassas, VA, USA), A2780S (C461) and 2008/C13.R (C446) (National Cell Bank of Iran) were cultured in their optimal conditions in RPMI-1640 (Gibco, Paisley, Scotland), containing 10% FBS (Gibco, Paisley, Scotland), 100 units/ml penicillin (ICN Biomedicals, Ohio) and 100 µg/ml streptomycin (Sigma, St. Louis, MO) at 37°C in a humidified incubator with 5% CO₂ atmosphere.

The siRNA against SORT1 and mock control (non-targeting control) were purchased from Thermo Scientific Company (Lafayette, CO, USA). siRNA reagent against SORT1 consisted of a pool of four siRNA oligonucleotides with the following sequences:

GAGACUAUGUUGUGACCAA;

The non-targeting control was used as a negative mock control to eliminate background of siRNA transfection. Suppression of SORT1 expression was performed in Caov-4 cells, which had been trypsinized and seeded 24 hr prior to transfection, either in 12-well plates (2×10⁵ cells/well for RNA extraction and Western blot analysis) or in 96-well plates (2×10⁴ cells/well for proliferation and apoptosis assays). On the day of transfection, cells were 70% confluent. siRNA or mock control transfection was carried out at a final concentration of 200 nM using lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. The optimal duration and efficiency of the transfection process was optimized using fluorescein-labeled siRNA. The expression level of targeted SORT1 mRNA was controlled by real-time quantitative PCR at 24 and 48 hr after transfection. Western blot analysis at 48, 72, and 96 hr after transfection was performed to assess the protein expression decline.

Total RNA was isolated from transfected and untransfected cultured cells using RNA-Bee reagent (BioSite, Täby, Sweden) according to the manufacturer's instructions. First strand cDNA was synthesized using 2 µg of total RNA in a 20 µl reaction mixture consisting of 4 µl of a 5× reaction buffer, 2 µl of 10 mM dNTP, 1 µl of 20 pmol/µl random hexamer primer (N6) and 20 U of M-MLV reverse transcriptase (Fermentas, St. Leon-Rot, Germany). The reaction was performed at 42°C for 60 min.

All real-time quantitative PCRs were performed by an ABI 7500 real-time PCR system (Applied Biosystems, Weiterstadt, Germany), utilizing SYBR Green reagents (Takara, Shiga, Japan) according to the manufacturer's instructions. Amplification of PCR products was quantified during PCR by measuring fluorescence associated with binding of SYBR Green dye incorporated into the reaction mixture to double-stranded DNA. The sequences of the oligonucleotides used in PCR-amplification of SORT1 and β-actin as housekeeping gene, were S: 5′-CAGTCCAAGCTATATCGAAG TGAGG-3′, AS:5′-AAGATGGTGTTGTCTG ATCCCCATTT-3′ and S: 5′-AGCCTCGCCT TTGCCGA-3′, AS: 5′-CTGGTGCCTGGGG CG-3′, respectively ⁽¹³⁾. Each PCR was performed in a 20 µl total volume mixture containing 10 µl of SYBR^® Premix Ex Taq (2×), 0.4 µl of ROX Reference Dye (50×), 500 nM of each primer, and 2 µl of 1:2 diluted cDNA samples. Following an initial denaturation step of 95°C for 10 s, 40 cycles of 95°C for 5 s, 60°C for 1.5 min, for both SORT1 and β-actin, the reaction was performed. Relative expression of SORT1 mRNA was calculated with the following formula:

R=(E^target) ^ΔCTtarget/(E^reference) ^{ΔCTreference}

Data were analyzed for significant differences (p < 0.05) by pair wise fixed reallocation randomization test using REST^© software ⁽¹⁴⁾.

Tissue samples and cell lines were lysed in a buffer containing 1% (v/v) Triton X-100, 50 mM Tris, pH = 7.4, 150 mM NaCl, 5 mM EDTA, pH = 8, 1 mM NaF, 20 mM Na₄P₂O₇, 1% (v/v) Glycerol, 0.1% (w/v) Sodium Dodecyl Sulfate (SDS), and 1% (v/v) protease inhibitor cocktail. For analysis of sortilin expression after RNAi technology, siRNA- or mock control- transfected cells were harvested 48, 72 and 96 hr post-transfection and lysed in the aforesaid lysis buffer. The protein concentration was estimated by a bicinchorinic acid (BCA) kit (Thermo Scientific, Rockfold, Illinois). Equal amounts of protein (15 µg) were run on SDS-PAGE. Following electrophoresis and electroblotting ⁽¹¹⁾, membranes were incubated with rabbit anti-human sortilin (Abcam, Cambridge, UK) at a concentration of 1 µg/ml, or with 1:2000 diluted mouse anti-human β-actin (Sigma).

After washing steps, membranes were incubated either with horseradish peroxidase (HRP)-conjugated sheep anti-rabbit immunoglobulin or with sheep anti-mouse immunoglobulin (Avicenna Research Institute, Tehran, Iran) diluted 1:2000. Filters were developed using the ECL advanced system (GE Healthcare, Uppsala, Sweden). For densitometry analysis of protein bands, AlphaEase software was used. The contrast was adjusted so that black and white colors were corresponded to 250 and 0, respectively. Using a rectangular selection, bands were then selected. Relative expression of sortilin was presented as the percentage values from sortilin/ β-actin density ratio. Reduction of sortilin expression following siRNA treatment was calculated at different hours post-transfection using the following formula:

[sortilin _(mock)/β-actin _(mock)-sortilin _(siRNA)/β-actin _(siRNA)]×100

Apoptosis in Caov-4 cells transfected with siRNA or mock control and also untreated control cells was measured at different time points (48, 72, and 96 hr after transfection). Cells were scraped, washed twice in PBS and centrifuged. Pelleted cells were incubated in 1× binding buffer containing FITC-annexin V and propidium iodide (BD Biosciences, San Jose, CA) in dark for 15 min. Apoptotic cells were examined using a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) and analyzed using the FlowJo software (version 7.6.1). A minimum of 1×10⁴ events per sample were acquired and analyzed. Three independent experiments were performed. Values were expressed as the mean±SEM in 3 separate experiments.

Cell proliferation was measured using standard [³H]-thymidine incorporation assay. Caov-4 cells were cultured at 37°C, 5% CO₂ in 96-well plates (Costar Corporation, Cambridge, MA). Twenty eight hours after transfection, 1 µci/well [³H]-thymidine (Amersham Biosciences, Buckinghamshire, UK) was added to the media of siRNA-or mock control- transfected cells, untransfected cells (as positive control) and staurosporine-treated (5 µM) (Sigma, St. Louis, MO) cells (as negative control). Cells were then allowed to propagate for 20 hr followed by harvesting onto filter papers (Wallac, Turku, Finland). Radioactive incorporation was measured in a 1450 Micro-beta TriLux Scintillator (Wallac, Turku, Finland) and expressed as counts per minute (CPM). All tests were performed with 6 replicates in two independent experiments. The proliferation index was calculated for each experiment as:

(Mean CPM of transfected cells/Mean CPM of untransfected cells)×100.

The Mann-Whitney test was used to determine the statistical significance (p < 0.05) between experimental groups. Values were expressed as the mean±SEM.

Basal level of sortilin protein expression was investigated by Western blot analysis. The results clearly showed that all the primary ovarian carcinoma tissues as well as ovarian carcinoma cell lines, namely A2780S, 2008/ C13.R, OVCAR-3, Caov-4 and SKOV-3, over-expressed sortilin with a strong band of about 95-100 kDa (Figures 1A and 1B). Although all non-malignant ovarian tissues expressed sortilin, the level of expression was comparably low (Figure 1A). Interestingly, a band was observed at about 80-85 kDa molecular weight in all non-malignant ovarian tissues, corresponding to the size of the second protein-coding variant of sortilin containing 694 amino acids. The 80-85 kDa form of sortilin results from the missing of the first N-terminal 137 amino acids (nearly 15 kDa) following a reading frame shift in the main transcript variant containing 831 amino acids (95-100 kDa) (http://www.ensembl.org/Homo_sapiens/Transcript/Sequence_cDNA?g=ENSG00000134243;r=1:109869428-109878924;t=ENST00000466471).

The ability of siRNA to inhibit expression of sortilin in Caov-4 cell line was determined using real-time quantitative RT-PCR (24 and 48 hr after transfection) and Western blotting (48, 72 and 96 hr after transfection). The siRNA transfection significantly (p < 0.001) suppressed the expression of SORT1 in Caov-4 cells. Analysis of real-time quantitative RT-PCR revealed a 6.1 fold reduction in the relative expression of SORT1 in the siRNA-transfected sample as compared with the mock control 24 hr post-transfection (p < 0.001) (Figure 2). The reduction of SORT1 transcript in siRNA-transfected cells continued 48 hr post-transfection in which a 4.2 fold reduction was observed (p < 0.001).

Real-time quantitative PCR results were confirmed by Western blotting at the protein level. The siRNA transfection of Caov-4 cells caused a significant decrease in the amount of sortilin protein expression. Densitometric analysis by AlphaEaseFC software showed that the level of sortilin was reduced by 72, 69 and 61% in siRNA-treated Caov-4 cells as compared with mock controls, 48, 72 and 96 hr post-transfection, respectively (Figure 3). The level of sortilin expression 24 hr post-transfection showed no reduction as compared to control (data not shown).

Next, the effect of suppression of sortilin expression on survival of Caov-4 cells was examined using FACS analysis of annexin V staining 24, 48 and 72 hr post-transfection. In each individual experiment, the percentage of apoptotic cells was calculated as the sum of percentages of cells in early (i.e., annexin V-positive, PI-negative) and late (i.e., annexin V-positive, PI-positive) stages of apoptosis and was subtracted from those of mock control-transfected cells. The results showed that suppression of sortilin induced 27.5±0.48 apoptosis of Caov-4 cells 48 hr after siRNA transfection when compared to mock control-transfected cells (Figures 4A and 4B). The level of apoptosis decreased over time after transfection. As compared to mock control-transfected cells, the siRNA-treated cells showed 6.4±2.4% and 7.3±0.9% apoptosis at 72 and 96 hr after transfection, respectively (Figures 4A and 4B).

Proliferation assay using [³H]-thymidine was carried out 48 hr after transfection (a time point in which the maximum apoptosis induction was achieved) to evaluate the influence of sortilin silencing on the proliferative capacity of siRNA-transfected cells. Results from this experiment clearly confirmed those obtained from apoptosis analysis. Sortilin silencing resulted in significant inhibition of proliferation (40.1%) in siRNA-transfected Caov-4 cells as compared to mock control-transfected counterpart (p < 0.05) (Figure 5).