ESCC cell lines EC0156 was established by our laboratory (16). KYSE30, KYSE140, KYSE150, KYSE170, KYSE180, KYSE410 and KYSE510 were donated by Dr Y. Shimada. EC0156 was cultured in Dulbecco’s modified Eagle’s medium (HyClone, UT, USA) supplemented with 10% fetal calf serum, 100 U/ml penicillin and 100 μg/ml streptomycin (Gibco, NY, USA). The other cell lines were cultured in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum. All the cells were incubated at 37°C in a humidified atmosphere of 5% CO₂.

From January 1999 to 2002, 8 pairs of specimens for two dimensional electrophoresis were obtained from surgically resected ESCC tissues in Cancer Hospital of Chinese Academy of Medical Sciences (CAMS). Another 50 tissues specimes for immnohistochemistry were also obtained from surgically resected esophageal carcinoma in Cancer Hospital of CAMS from January 1999 to 2009. Tissue specimens (n=93) for immunohistochemistry were purchased as microarray (Outdo Biotech Co., Shanghai, China). All specimens were frozen immediately, stored at liquid nitrogen. The median age of the patients of the 143 ESCC tissues for immunochemistry was 60 years (range, 29–84 years).

Approximately 1×10⁷ cells were grown to 80% confluence and washed six times in 1X phosphate-buffered saline (PBS). Soluble proteins were extracted with lysis (50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.1% SDS) with protease inhibitor cocktail (2.5 μM AEBSF, 0.04 μg/ml aprotinin, 0.04 μg/ml leupeptin, 1 mM EDTA ) by super-sonication and followed by centrifugation at 12,000 g for 15 min. Protein concentration was measured by the Bradford method.

The soluble proteins from individual tissue specimen and the pooled tissue samples were separated by two dimensional electrophoresis (2-DE). Commercial IPG strips (pH 3–10NL, 18 cm; Amersham Biosciences, Uppsala, Sweden), were rehydrated overnight with 450 μl solution containing 8 M urea, 2% w/v CHAPS, 20 mM DTT, 0.5% v/v IPG buffer, 0.002% bromophenol blue and 1000 μg protein. Electrofocusing was carried out for 60 kVh at 20°C following the manufacturer’s instruction. Prior to the second dimension, the IPG strips were equilibrated for 30 min with 50 mM Tris-HCl pH 8.8, 6 M urea, 30% v/v glycerol, 2% w/v SDS, and a trace of bromophenol blue followed by reduction with 1% of DTT and alkylation with 2.5% of iodoacetamide. The IPG strips were placed into 12% SDS-polyacrylamide gels (26×20 cm) and were further electrophoresed by an EttanII-DE system (Amersham Biosciences) with a programmable power control, 0.5 h at 0.5 W per gel, then at 15 W per gel until the dye front reached the gel bottom. The separated proteins were visualized by Coomassie Brilliant Blue staining.

Briefly, images of the stained gels were acquired with an Image Scanner (Amersham Biosciences) using transmissive light. The gel images were first analyzed by eyes and subsequently were analyzed by ImageMaster 2D Elite 4.01 (Amersham Biosciences). Protein spots with signals differential intensity reaching 2.0 in 2D gels were excised, then were digested in modified trypsin solution with a final substrate-to-trypsin ratio of 40:1 (W:W) (in 25 mM ammonium bicarbonate). The digested peptides from 2-DE gel spots were analyzed by MALDI-TOF-MS using an Ettan-MALDI-TOF system (Amersham Biosciences). Monoisotopic peptide masses obtained from MALDI-TOF were used to search the NCBInr protein database using Mascot algorithm.

For immunohistochemical staining multiple tissue arrays (MTA) of formalin-fixed and ESCC and their matched adjacent normal tissues were incubated with stathmin mAb (3352, Cell Signaling Technology, UK) or control IgG. After washing with 1X PBS, slides were reacted with the biotin-labeled second antibody and then visualized using an ultrasensitive streptavidin-peroxidases system (Maxim Biotech, Fuzhou, China). Semi-quantitative criteria of the stathmin immunoreaction were modified according to previous publications (17,18). Immunostaining was scored as follows: 0, negative; 1, weak; 2, moderate; and 3, strong staining. The percentage of stathmin staining area was graded as 0, no positive staining; 1, <10%; 2, 10–50%; or 3, 50–100%. The staining index was calculated as the multiples of staining intensity and staining area, as described (18,19) The staining index <1 was considered as negative, while 1–4 as weak and >4 as strong.

Western blotting was performed as previously described (20). In briefly, protein extracted from cell lines and tissues specimens were separated using 12% SDS-PAGE, then transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Bedford, MA). Membranes were blocked by 10% skim milk in 1X PBS. The membranes were incubated with the primary antibodies against stathmin (ab52630, Abcam, UK) or β-actin (Cat. No. A-5316, Sigma, MO) in suitable dilutions. Secondary antibodies were anti-rabbit IgG and anti-mouse IgG, respectively. Signals were detected by chemiluminescence using the ECL kit.

The double-strand small interfering RNAs (siRNAs) were synthesized in duplex and purified forms using Genechem Co. (Shanghai, China). siRNAs targeting stathmin [5′-GAAACGAGAGCACGAGAAAtt-3′ (forward) and 5′-UUUCUCGUGCUCUCGUUUCtt-3′ (reverse)] and non-specific scramble siRNA oligonucletide sequences [5′-UUC UCCGAACGUGUCACGUtt-3′ (forward) and 5′-ACGUGAC ACGUUCGGAGAAtt-3′ (reverse)] were transfected separately into KYSE30 and KYSE410 cells using Lipofectamine 2000 (Invitrogen, CA, USA) according to manufacturer′s protocol. After 72 h, proteins were extracted.

KYSE30, KYSE410 or EC0156 cells were incubated overnight yielding confluent monolayer for wounding. Wounds were scratched using a pipette tip, photographs were taken immediately (time 0 h) and 24 h after wounding. The distance migrated by the cell monolayer to close the wounded area during this time period was measured. Photos were taken by Leica DMCI microscope (Leica, German) at ×100 magnification.

Data were analyzed by SPSS 16.0 (SPSS Inc., Chicago, IL, USA). χ² test was used to analyze the relationship between stathmin expression with clinicopathologic characteristics. P-values <0.05 were considered as statistically significant.

Firstly, we analyzed differientially expressed proteins using proteomic method between tumor and corresponding normal tissues in ESCC. Results showed stathmin was detected in all 2-DE gels of the 8 pairs of ESCC tissues (Fig. 1), as an obvious differentially expressed spot.

Subsequently, we employed immunohistochemistry to analyze the expression of stathmin in 143 ESCC tissue microarray using the anti-stathmin antibody. Strong staining was seen in ESCC tissues, however, the normal tissues were weaker or negatively stained (Fig. 2). The strong staining of stathmin in cancer was 70.63% (101/143), while normal in 27.97% (40/143). In addition, the weak staining of stathmin in cancer was 20.28% (29/143), while normal in 15.38% (22/143). There was a significantly different staining mode between cancer and normal (P<0.05). Statistical analysis showed that overexpression of stathmin was significantly correlated with histological grade (P<0.05) (Table I). However, no correlation was found between stathmin expression and age, gender and lymph node metastasis (P>0.05) (Table I).

We chose two ESCC cell lines (KYSE30 and KYSE410) as a model. SiRNA was employed to knockdown stathmin, and western blotting to detect the effect of siRNA. The results showed that the expression of stathmin was reduced after transfection of siRNA oligonucleotide for 72 h in KYSE30 and KYSE410 (Fig. 3A). Subsequently, wound-healing assay showed that the speed of wound recovery of siRNA stathmin was much slower than the scramble in both KYSE30 and KYSE410 (Fig. 3B). The results revealed that cell migration was impaired when deficient of stathmin.

Considering phosphorylation had been closely associtated with the function of stathmin, we wondered whether stathmin phosphorylation has influence on ESCC cell lines. We selected EC0156 for the next study.

EC0156 was treated with paclitaxel, a compound extracted from taxus plants, at gradient dosage from 0 to 1.6 μg/ml. The results revealed that the 19 kDa band of stathmin was decreased after paclitaxel treatment, whereas a new band at about 21 kDa gradiently increased in a dose-dependent manner. We confirmed the new band was phosphorylated stathmin at Ser-16. The tendency increased dramatically between 0.1 and 1.6 μg/ml, reaching a peak at a dosage of 0.8 μg/ml. Conversely, KYSE30 and EC0156 without treatment had no changes (Fig. 4A). Besides, rounded cells were observed in most of the EC0156 through microscopic analysis (Fig. 4B). The number of cells appered slightly reduced. Wound-healing assay showed that the speed of wound recovery of EC0156 cells treated with pacitaxel was much slower than the control, suggesting that cell migration was impaired after stabilized phophorylation of stathmin (Fig. 4C).