Mol Biol Rep DOI 10.1007/s11033-014-3828-8

Stat3 promotes invasion of esophageal squamous cell carcinoma through up-regulation of MMP2 Xaioyan Xuan • Shanshan Li • Xi Lou • Xianzhao Zheng • Yunyun Li • Feng Wang • Yuan Gao • Hongyan Zhang • Hongliu He • Qingru Zeng

Received: 6 August 2013 / Accepted: 10 November 2014 Ó Springer Science+Business Media Dordrecht 2014

Abstract Stat3 alters the expression of its downstream genes and is associated with tumor invasion and metastasis in several human cancers. Its role in esophageal squamous cell carcinoma (ESCC) has not been well characterized. We examined the tumor sections of 100 cases of ESCC by immunohistochemistry and observed significant overexpression of Stat3 in the cytoplasm of 89 % of ESCC cells and of phosphorylated Stat3 (p-Stat3) in the nuclei of 71 % of ESCC when compare with normal esophageal mucosa (72 %, p = 0.02; and 31 %, p = 0.001). Overexpression of Stat3 and p-Stat3 positively correlated with that of matrix metalloproteinase-2 (MMP2), a known regulator for cell migration, in 65 % of ESCC while only 26 % shown in benign esophageal mucosa. To further investigate the association of Stat3 with tumor metastasis in vitro, invasion of EC-1 cells (a human ESCC cell line) were investigated with Boyden chambers. The results showed that transfection of Stat3 not only promoted invasion of EC-1 cells but also significantly induced MMP2 expression in a dosedependent manner. In contrast, suppressing expression of endogenous Stat3 mRNA and protein by Stat3 siRNA significantly reduced EC-1 cell invasion and MMP2 expression. A high-affinity Stat3-binding element was localized to the positions of 648–641 bp (TTCTCGAA) in

X. Xuan  S. Li (&)  X. Lou  X. Zheng  Y. Li  F. Wang  Y. Gao  H. Zhang  H. He  Q. Zeng Department of Pathology, The First Affiliated Hospital and Key Laboratory of Tumor Pathology, Zhengzhou University School of Medicine, Zhengzhou, Henan, China e-mail: [email protected] Present Address: X. Xuan Department of Microbiology and Immunology, Zhengzhou University School of Medicine, Zhengzhou, Henan, China

the MMP2 promoter with electrophoretic mobility shift assay. Our results suggest that Stat3, p-Stat3, and MMP2 were overexpressed in ESCC and associated with invasion of ESCC; and Stat3 up-regulated expression of MMP2 in ESCC through directly binding to the MMP2 promoter. Keywords Stat3  MMP2  Esophagus  Squamous cell carcinoma  Invasion

Introduction Esophageal carcinoma is the most common malignant tumor in northern China [1]. Despite the improving combination of surgery, radiotherapy, and chemotherapy, the outcome of esophageal squamous cell carcinoma (ESCC), especially the metastatic ones, remains disappointing [1]. To establish novel diagnostic and therapeutic strategies against this deadly disease, it is essential to understand its molecular pathology. A few proteins have been identified to associate with the metastasis of ESCC, but their molecular basis are still poorly understood [2]. Signal transducers and activators of transcription (Stat) are a family of latent transcription factors, mediating the targeted gene activation in response to stimulation by cytokines and growth factors [3, 4]. Seven Stat family members (and their alternative splice products) have been identified, and each member is activated by a distinct spectrum of cytokines [5]. After acquiring high affinity DNA-binding activity through phosphorylation by various tyrosine kinases, including Src and Jak family kinases, Stat proteins can regulate functions of other genes that function in proliferation, survival, and motility of the cells [6, 7]. The over-activation of Stat 3 has been founded not only in primary, but also metastatic human cancers of the thyroid,

123

Mol Biol Rep

bladder, prostate, breast, and ovary [8–10]. Stat 3 promotes tumor metastases through different molecular mechanisms in these cancers [9]. A prerequisite for most tumors’ invasion and subsequent metastasis is to penetrate the basement membrane. A family of proteins the matrix metalloproteinases (MMPs) are directly involved in the degradation of basement membrane and thus open the channel for tumor’s metastasis [11]. Over-expression of MMPs are observed in many primary prostate cancer cells, osteoblasts, and osteoclasts [12], and also observed in ESCC, including metastatic lesions [13–15]. Interestingly, Stat 3 often co-overexpresses with its downstream gene MMP2. This phenomenon strongly suggests Stat3 may promote the progression and metastasis of ESCC through up-regulation of MMP [16]. This study was designed to confirm such a hypothesis.

sequence for human Stat3 mRNA was selected from GenBank (accession no. NM-213662). The Stat3 siRNAs were synthesized by Sunbiotech (Beijing, China) as follows: sense: 50 -GATCCCCCATCTGCCTA-GATCGGCT ATTCAAGAGATAGCCGATCTAGGCAGATGT-TTTT GGAAA-30 ; anti-sense: 50 -AGCTTTTCCAAAAACATC TGCCTAGATCGGCTATCTCTTGAATAGCCGATCTA GG- CAGATGGGG-30 . pSUPER-Stat3-siRNA was constructed by insertion of a PCR-generated segment into the N-terminus of EGFP at Bgl-II and Hind-III sites in pSUPER-EGFP vector. This Stat3-siRNA expression plasmid was verified by restriction digestion and sequencing analysis. EC-1 cells were transiently transfected with either pSUPER-EGFP vector (vector) or the pSUPER-Stat3-siRNA plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol.

Materials and methods

Cell culture and transfection

Tissue specimens

Human ESCC cell line EC-1 cells were supplied by Professor Shihua Cao (Hongkong University) and human fibroblast cell line NIH/3T3 was supplied by Professor Ziming Dong (Zhengzhou University). Both cells were grown in RPMI 1640 supplemented with 10 % fetal bovine serum, 2 mmol/L L-glutamine, 0.1 mmol/L non-essential amino acid, 1.0 mmol/L sodium pyruvate, 1.5 g/L NaHCO3, 100 IU/mL penicillin, and 100 lg/mL streptomycin. EC-1 cells were transfected with control vector or pXJ40-Flag-Stat3 plasmid (1–4 lg,) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer’s protocol. Stable transfected EC-1 cells were selected with G418 and confirmed with Western blot. IL-6 (10 ng/L) was added 24 h later in all the groups for 20 min. Each group was set in triplicate, and the experiments were repeated at least three times.

Our study consisted of 100 patients with ESCC at the First Affiliated Hospital of Zhengzhou University. All tissues obtained from these patients were paraffin-embedded. The clinicopathological findings were evaluated according to the American Joint Committee on Cancer (AJCC) staging system. The diagnosis of ESCC was re-confirmed by a board-certified pathologist. The study was carried out according to the guidelines of the Declaration of Helsinki. Immunohistochemistry Immunohistochemical staining was performed as described before [17]. The slides were incubated with primary antibodies anti-Stat3 (1:50), anti-p-Stat3 (1:50) and antiMMP2 (1:100), respectively. All of them were purchased from Santa Cruz Biotech (Santa Cruz, CA). At least 500 tumor cells in five fields were counted. More than 10 % of the tumor cells reacting with the antibodies were defined as ‘‘Positive’’. Two pathologists, who were unaware of the clinical data or the disease outcome, examined all histologic slides. When the interpretations differed between the two observers, slides were reevaluated for a final decision on a conference microscope. Constructs of Stat3 and Stat3 siRNA expression The eukaryotic expression vector pXJ40-Flag-Stat3 (FlagStat3) was provided by Dr. Xinyuan Fu (Yale University, New Haven, CT, USA). The pSUPER-EGFP vector for preparation of Stat 3 siRNA (Stat3 siRNA) was provided by Dr. Yi Ding (Zhengzhou University, China). The target

123

RNA preparation and reverse transcription PCR (RTPCR) Total RNA was isolated with TRIzol (Invitrogen). Five lg of total RNA was used to generate the first strand of DNA using the High Capacity cDNA Reverse Transcription Kit (TaKaRa, Shiga, Japan) according to the manufacturer’s protocol. One-tenth of the cDNA mixture was used as template for the subsequent PCR amplification. Reverse transcription reactions were optimized using TaqDNA polymerase (TaKaRa) for the linear range of cDNA amplification. Linear amplification of a 551-bp fragment of Stat3 was achieved at 30 cycles of replication with the primers, 50 -GGCATTCG GAAAGTATTG-30 and 50 -GCTGCTGAGAAAGG- AG GG-30 ; and of a 493 bp fragment of MMP2 at 30 cycles of

Mol Biol Rep

replication with the primers, 50 -TTCAAGGACCGGTTC ATTTGGCGGACTGTG-30 , and 50 -TTCCAAACTTCAC GCTCTTCAGACTTTGGTT-30 . b-Actin primers were used in all the reactions as an internal control to normalize the amount of mRNA expression. Western blot analysis EC-1 cells were harvested and lysed in lysis buffer containing 150 mM NaCl, 1 % Nonidet P-40, 1 mM EDTA, 0.5 % deoxycholic acid, 2 lg/mL aprotinin, 1 mM phenylmethylsulfonyl fluoride, 5 mM benzamidine, 1 mM sodium orthovanadate, and 10 lg/mL soybean trypsin inhibitor (Sigma, St. Louis) in 50 mM Tris buffer, pH 7.4. Twenty micrograms of the proteins from each sample were separated by SDS-PAGE gel electrophoresis and then transferred onto a nitrocellulose membrane (Yili, Beijing, China). After blocking with 5 % nonfat dry milk ?0.1 % Tween 20 in Tris-buffered saline (TBS-T) at room temperature for 1 h, the membrane was incubated with primary antibody (p-Stat3: 1:1000, MMP2: 1:500) at 4 °C overnight and then incubated with horseradish peroxidaseconjugated secondary antibody (1:2,000) for 1 h. The levels of p-Stat3, and MMP2 were quantitated from corresponding bands by densitometry scanning of the X-ray films with UVP’s gel documentation system GDS8000 and 1D gel analysis software (UVP, LLC, Upland, CA, USA). b-Actin detection as an internal control were performed for all the blots to normalize the protein expression to ensure the equal loading of the samples. Invasion assay The invasion of EC-1 cells was measured using Boyden transwell chambers (Transwell-Costar Corp, Corning, NY) according to the manufacturer’s protocol. Briefly, the EC-1 cells transfected with control vector, pXJ40-Flag-Stat3 plasmid (1–4 lg), or Stat3 siRNA plasmid were seeded onto the membrane of the upper chamber at a concentration of 3 9 104 in 400 lL medium. RPMI-1640 medium added to the lower chamber. After 24 h, the lower chamber was washed 3 times with deionized water. The membrane in the lower chamber was stained with crystal violet. The number of invaded cells was counted with a microscope in at least in 3 fields. Electrophoretic mobility shift assay (EMSA) and supershift assay Electrophoretic mobility shift assays were performed as described previously [18, 19]. The sequence of the MMP2 promoter (?10 to -1691) was analyzed for the consensus sequence [50 -TT(N4-6)AA-30 ], the putative binding site of

Stat 3. Four such sites were identified and the corresponding oligonucleotides were synthesized. These oligonucleotides were biotin-labeled as probes and incubated with 10 lg of nuclear protein extracts of EC-1 cells for EMSA. For a supershift assay, a nuclear protein extract of EC-1 cells was pre-incubated with specific antibodies against Stat3 (Biyuntian, Shanghai, China 1:1,000). Protein–DNA complexes were separated on a 4.5 % nondenaturing polyacrylamide gel at room temperature, and then scanned with a densitometer. Statistical analysis Pairwise correlations between the groups were analyzed with the Spearman rank correlation test. The statistical significance of differences between groups was evaluated by a paired Wilcoxon rank sum test, Student t test or Mann– Whitney test (comparison of means). The Fisher exact test was used for comparison of incidence. For measurement of tumor growth, two-way repeated analysis of variance was performed with ANOVA test. All outcome variables are presented as the mean values with 95 % confidence intervals. All data are representative of at least three independent experiments. All statistical tests were two-sided, and P values less than 0.05 were considered statistically significant.

Results Expression of Stat3, p-Stat3, and MMP2 in ESCC Immunohistochemical staining showed that Stat3 protein was predominantly localized in the cytoplasm (Fig. 1a) in 89 % (89/100) of ESCC, significantly higher than that (72 %) of normal esophageal mucosa (p = 0.002; Table 1). Stat3 is known to become active after phosphorylation of serine 727 [6], and thus, the expression of p-Stat3, the active form of Stat3, was also examined. In contrast to the cytoplasmic localization of Stat3, p-Stat3 stain was exclusively localized to the nuclei of cells (Fig. 1b) in 71 % of ESCC but only in 31 % of benign esophageal mucosa (p = 0.001; Table 1). MMP2, the downstream gene of Stat3, showed localization in the cytoplasm of cells (Fig. 1c) in 65 % of ESCC, significantly higher than that (26 %) of benign esophageal mucosa (p = 0.001; Fig. 1d; Table 1). Expression of Stat3 and p-Stat3 was positively correlated with that of MMP2 (both p \ 0.01). In ESCC with lymph node (LN) metastasis, Stat3 protein was expressed at a similar rate to that of ESCC without LN metastasis (92.5 vs. 83.1 %; p = 0.143; Table 1). However, p-Stat3 protein showed a significant higher expression in ESCC with LN metastasis than in ESCC without LN metastasis (91.2 vs. 44.2 %; p = 0.001;

123

Mol Biol Rep

Fig. 1 Localization of Stat3, p-Stat3, and MMP2 in benign and malignant esophageal tissues with immunohistochemical stains. a Predominantly the cytoplasmic staining of Stat3 in esophageal

squamous cell cancer (ESCC) (9400). b Nuclear staining of p-Stat3 in ESCC (9400). c Cytoplasmic staining of MMP2 in ESCC (9400). d Negative staining of Stat3 in benign esophageal epithelium (9200)

Table 1 Frequencies of Stat3, p-Stat3 and MMP2 protein expression in benign and malignant esophageal tissues

the effect of Stat3 on invasion of EC-1 cells in vitro by using Boyden transwell chambers. To validate the transfection of pXJ40-Flag-Stat3 in human ESCC cell line EC-1 cells, expression of Stat3 was examined by RT-PCR. The results showed a significant increase in expression of Stat3 mRNA in EC-1 cells (Fig. 2a) from 0.80 ± 0.02 in the vector control group only to 0.99 ± 0.02, 1.23 ± 0.03, 1.28 ± 0.02, and 1.25 ± 0.03 in the groups transfected with 1, 2, 3 and 4 g of pXJ40-Flag-Stat3, respectively (all p \ 0.05 when compared with vector control). Such an increase was dose-dependent in the range of 1–3, but not 4 lg of pXJ40-Flag-Stat3. The expression of p-Stat3 was examined by western blot and presented a similar increase as Stat3 mRNA (Fig. 2b): 0.656 ± 0.03 in the group with transfected vector only, 0.791 ± 0.02, 0.941 ± 0.04, 1.18 ? 0.04, and 1.107 ± 0.02 in groups transfected with 1, 2, 3, and 4 g of pXJ40-Flag-Stat3, respectively (all p \ 0.05 when compared with vector control). The number of cells invasion increased from 88 ± 10 in the group transfected with pXJ40 vector to 99 ± 11, 139 ± 9, 147 ± 15, and 143 ± 12 in the groups transfected with at 1, 2, 3, and 4 lg of pXJ40-Flag-Stat3, respectively (all p \ 0.05 when compared with vector group; Fig. 2c). These increases were dose dependent in the range of 1–3 lg, but not 4 lg of pXJ40-Flag-Stat3.

Cases #

Stat 3

p-Stat 3

MMP2

Normal

100

72 (72 %)

31 (31 %)

26 (26 %)

ESCC

100

89 (89 %)*

71 (71 %)**

65 (65 %)**

ESCC LN?

57

53 (92.5 %)

52 (91.2 %)

45 (78.9 %)

ESCC LN-

43

36 (83.1 %)#

19 (44.2 %)##

20 (46.5 %)##

ESCC esophageal squamous cell carcinoma, Normal normal esophageal mucosa, LN? with lymph node metastasis, LN- without lymph node metastasis * Comparison of Normal with ESCC (* p = 0.002, ** p = 0.001), # comparison of ESCC LN? with ESCC LN (# p = 0.143, ## p = 0.001)

Table 1). Similar results were obtained with MMP2 protein expression: 78.9 % in ESCC with LN metastasis and 46.5 % in ESCC without LN metastasis (p = 0.001; Table 1). Effect of Stat3 on invasion of EC-1 cells in vitro Over-expression of Stat3 in ESCC encouraged us to examine its possible role in tumor invasion. We examined

123

Mol Biol Rep b Fig. 2 Effect of Stat3 overexpression on invasion of EC-1 cells.

a Electrophoresis of RT-PCR products (top) and quantification (lower) of increased expression of Stat3 in EC-1 cells after transfection of Stat3. b Western blot (top) and quantification (lower) of increased expression of Stat3 in EC-1 cells after transfection of Stat3. c Dose-dependent increase of invasion of EC-1 cells transfected with different dose of Stat3 (p \ 0.05 when compare with control and p \ 0.05 when compare with 1 g of Stat3). d Electrophoresis of RTPCR products (top) and quantification (lower) of inhibited expression of Stat3 in EC-1 cells after transection of Stat3 siRNA. e Western blot (top) and quantification (lower) of inhibited expression of Stat3 in EC-1 cells after transection of Stat3 siRNA (pSUPER-EGFP-Stat3). f Dose dependent inhibition of invasion of EC-1 cells transfected with Stat3 siRNA (p \ 0.05 when compare with control and p \ 0.05 when compare with cells after 24 h transfected with Stat3-siRNA)

To confirm the specificity of Stat3’s role in promoting the ability of cells for invasion, EC-1 cells were transfected with pSUPER-Stat3-siRNA (Stat3-siRNA). To validate the transfection of Stat3 siRNA in EC-1 cells, expression of endogenous Stat3 was examined with RT-PCR. The results showed that Stat3 siRNA, but not the vector, significantly reduced expression of endogenous Stat3 mRNA from 0.813 ± 0.024 at 0 h to 0.643 ± 0.025, 0.438 ± 0.036, and 0.297 ± 0.008 at 24, 48, and 72 h, respectively (all p \ 0.05 in any pair-wise comparison; Fig. 2d). The expression of p-Stat3 was examined with western blot and exhibited a similar decrease as Stat3 mRNA (Fig. 2e): 0.67 ± 0.02, 0.52 ± 0.02, 0.34 ± 0.02, and 0.14 ± 0.01 at 0, 24, 48, and 72 h, respectively (all p \ 0.05 in any pairwise comparison). The number of cells invasion remained the same (87 ± 5) in the group transfected with p-SUPERECFG vector only, but decreased 9.5 % (77 ± 18), 38.8 % (52 ± 12), and 58.8 % (35 ± 7) at 24, 48, and 72 h, respectively in the group transfected with Stat3 siRNA (all p \ 0.05 in any pair-wise comparison; Fig. 2f). Stat 3 up-regulated expression of MMP2 Since immunohistochemical results showed a significant correlation of Stat3 expression with that of MMP2 in ESCC, the causality of Stat 3 and MMP2 was explored. RT-PCR results demonstrated a substantial increase (up to 59 %) of MMP2 mRNA expression in EC-1 cells transfected with pXJ40-Flag-Stat3 than control (0.877 ± 0.040; Fig. 3a). Such up-regulation of MMP2 by Stat 3 was dosedependent within the range of 1–3 lg of Stat3 [22.7 % (0.913 ± 0.037) increase at 1 lg, 55.6 % (1.134 ± 0.047) at 2 lg, and 59.0 % (1.227 ± 0.026) at 3 lg] and maximized at 4 lg of Stat3 (50.0 %, 1.157 ± 0.065). Similar to the results of invasion experiments, up-regulation of MMP2 by [2 lg of pXJ40-Flag-Stat was significantly higher than that by zero or 1 lg (all p \ 0.05; Fig. 3a), but not by 3 or 4 lg of Flag-Stat3.

123

Mol Biol Rep

Fig. 4 Effect of Stat3 siRNA on expression of MMP2. a Electrophoresis (top) and quantification (lower) of RT-PCR products of MMP2 in EC-1 cells after transfection of Stat3 siRNA. b Western blot (top) and quantification (lower) of MMP2 protein expression were significantly reduced in EC-1 cells after transfection of Stat3 siRNA in a time-dependent manner

Fig. 3 Effect of Stat3 on expression of MMP2. a Electrophoresis (top) and quantification (lower) of RT-PCR products of MMP2 showed dose dependent increase in EC-1 cells after transfection of Stat 3. b Western blot for MMP2 (top) and quantification (lower) of MMP2 expression in EC-1 cells after transfection of Stat 3. b-Actin was used for normalization to ensure the equal loading of the samples

To examine the activity of up-regulated MMP2 mRNA, Western blots were conducted under similar conditions to that for RT-PCR. The results showed an 8 % (0.710 ± 0.035) increase of MMP2 protein expression at 1 lg of Flag-Stat, 49.5 % (0.982 ± 0.049) at 2 lg, 71.7 % (1.128 ± 0.077) at 3 lg, 51.1 % (0.993 ± 0.050) at 4 lg (Fig. 3b), when compared control (0.657 ± 0.069). Statistical analysis showed all the increase was significant as the results of RT-PCR. To confirm the specificity of Stat 3’s role in up-regulating expression of MMP2, EC-1 cells were transfected with 2 lg pSUPER-EGFP-Stat3. The RT-PCR results

123

showed a significant reduction in MMP2 mRNA expression when compared to control (Fig. 4a). Such reduction increased with time and maximized at 72 h; 29.5 % (0.630 ± 0.076) decrease at 24 h, 54 % (0.435 ± 0.047) at 48 h, and 76.1 % (0.203 ± 0.006) at 72 h (all p \ 0.05). After 72 h, transfected EC-1 cells had overgrown and most of them went into apoptosis. A western blot was performed under similar conditions to that of RT-PCR. The results showed a 27.1 % (0.643 ± 0.057) decrease of MMP2 protein expression at 24 h, 51.2 % (0.437 ± 0.064) at 48 h, and 68.4 % (0.314 ± 0.035) at 72 h when compared with control at 0 h (0.203 ± 0.050; Fig. 4b; all p \ 0.05). Stat3 binding to the promoter of MMP2 As a transcription factor and activator, Stat 3 must bind the promoter of its downstream genes to function. To further confirm the up-regulation of MMP2 by Stat 3, the putative Stat 3 binding site(s) must be identified in the MMP2 promoter. Thus, the DNA sequence of the MMP2 promoter was compared with the known consensus sequence of Stat 3 binding sites, 50 -TT(N4-6)AA-30 [20]. Four such sites

Mol Biol Rep

Fig. 5 Binding sites of Stat3 in the MMP2 promoter. a The sequences of predicted binding sites of Stat3 in the MMP2 promoter. b Electrophoretic mobility shift assay (EMSA). The probe labeling numbers correspond to those of sequences in a. c Supershift assay

were identified and synthesized (Fig. 5a). Electrophoretic mobility shift assay (EMSA) showed that only oligonucleotides 1, but not other 3, displayed one major shifted band of DNA–protein complex (Fig. 5b). These results strongly suggested that Stat3 binding elements are within the MMP2 promoter at positions of 648–641 bp (TTCTCGAA). The binding specificity of oligonucleotide 1 to the promoter of MMP2 was confirmed in three ways. Firstly, the competitive blockage of Stat3 binding was carried out with unlabeled oligonucleotides 1 and the density of shifted band of DNA–protein complex was consequently reduced (Fig. 5c, Lane 3); secondly, a site mutation was introduced at the base pairs of 643–642 (TC to AG) of the oligonucleotide 1 to change one conserved base in the putative Stat3 binding site. The mutant oligonucleotide 1 probe completely abolished the shifted band of DNA–protein complex (Fig. 5c, Lane 4) in EMSA. Thirdly, addition of Stat3 antibody made the shifted band super shifted (Fig. 5c, Lane 5).

Discussion Metastasis is a complex process and the major cause of death in most cancer patients [21–23]. Understanding its molecular mechanisms will explore the future directions for more effective biological treatment for cancer patients.

Research has shown that the alterations of various oncogenes, tumor suppressor genes, metastasis suppressor genes, and growth factors and their receptors are associated with tumor metastasis [21–23]. The fact that Stat3 is activated by numerous cytokines, growth factors, and oncogenic proteins suggests that Stat3 signaling may be one of the common pathways involved in regulating cancer metastasis [4, 8, 9]. Stat3 plays an important role in a wide variety of physiologic processes, including cell differentiation, proliferation, and apoptosis. The Stat3 activation is transient and tightly controlled in normal cells [24–27]. On the contrary, the constitutively activated Stat3 protein has been found to be in leukemia, melanoma, cancers of the breast, head and neck, prostate, and pancreas [28, 29]. Interestingly, except its role in oncogenesis, such activated Stat3 protein are associated with metastases of thymic epithelial tumors, colorectal adenocarcinoma, and cutaneous squamous cell carcinoma, and renal cell carcinoma [30–32]. In this study, immunohistochemical data showed that Stat3 and its active form, p-Stat3, were significantly overexpressed in ESCC when compared with those in normal esophageal mucosa. Stat3 was mainly localized to the cytoplasm whereas the p-Stat3 was exclusively found in the nuclei of cells of ESCC. Both over-expression and nuclear translocation of Stat3 in ESCC suggest that Stat3 may be associated with oncogenesis of ESCC. Furthermore, overexpression of Stat3 significantly enhanced the invasive ability of EC-1 cells in vitro while the siRNA-mediated Stat3 knockdown significantly inhibited such ability of EC-1 cells. These results indicate that over-expression of Stat3 is also associated with the metastasis of ESCC through increasing the invasive ability of tumor cells. Interestingly, only p-Stat3, not Stat3, was significantly over-expressed in the ESCC with lymph node metastasis when compared with the ESCC without lymph node metastasis. Therefore, p-Stat3 may serve as a better predictor for ESCC metastasis. The mechanisms by which Stat3 may promote tumor metastasis remain poorly understood, but it has been shown that Stat3 can target different downstream genes in different tumors, such as the antiapoptotic gene XIP in B cell leukemia [33], VEGF in diverse human cancers [34], MMP2 in melanoma, and paxillin in ovarian cancer [35– 37]. Our data demonstrated that MMP2 was simultaneously over-expressed with p-Stat3 in the ESCC with lymph node metastasis when compared with that in the ESCC without lymph node metastasis. In addition, both mRNA and protein of MMP2 were up-regulated in EC-1 cells with Stat3 over-expression, but down-regulated in EC-1 cells with siRNA-mediated Stat3 knockdown. These results suggest Stat3 may promote tumor metastasis through up-regulation of MMP2 in ESCC. The prerequisite for Stat3 to up-regulate its target gene is that

123

Mol Biol Rep

Stat3 can bind to the promoter of that gene. Our EMSA experiments provided further evidence for such interaction between Stat3 and MMP2 gene promoter. Using 4 putative binding sites of Stat3, the sequence for Stat3 binding was specifically mapped between 648 and 641 bp of the MMP2 promoter. Mutation of this Stat3-binding element abolished the binding of Stat3. Thus, our results demonstrate that Stat3 regulates MMP2 expression through direct interaction with the MMP2 promoter via a Stat3-binding element. Stat3 is well-documented to regulate many genes, even in tumor cells. Xie et al. reported that the activated-Stat3 could up-regulate basic fibroblastic growth factor, VEGF, and MMP2 in brain metastasis of melanoma [38]. It needs further study to verify whether MMP2 is the only molecule up-regulated in ESCC and associated with ESCC metastasis. The fact that Stat3 can be activated by a variety of factors and can affect expression of many downstream genes makes it as a potential therapeutic target for cancer patients. Theoretically, elimination of Stat3 activity may enhance the function of tumor suppressor genes or other anti-tumor genes to counteract roles of many oncogenes. Consequently, oncogenesis and/or tumor metastasis will be impaired or even prevented. Experiments have shown that abrogation of Stat3 activity inhibited the growth/proliferation in leukemia cell line U266, multiple myeloma cell RPMI 8226 [39, 40], or induced apoptosis in pancreatic cancer cell lines [41] and tumor cells of human lung adenocarcinoma, breast carcinoma, glioblastoma, neuroblastoma [42]. Furthermore, blockage of Stat3 function sensitized cancer cells to chemotherapeutic agents in hepatoma, melanoma, and B cell leukemia [43]. In the present study, the significant over-expression of p-Stat3 in ESCC, especially in those with lymph node metastasis, provides further justification to target Stat3 for the treatment of ESCC patients. However, it should be noted that abrogation of Stat3 may have unexpected side effects since many genes are regulated by Stat3. Glienke et al. reported that siRNA-mediated Stat3 knockdown not only induced apoptosis of pancreatic cancer cell lines BxPC-3 but also promoted the expression of anti-apoptotic genes Survivin/ BIRC5 and BCL-xL in the same cell lines [44]. The ideal goal is to find effective therapeutic agents that block Stat3 activity with negligible side effects. Acknowledgments The authors sincerely thank Dr. Guy Benian at Emory University in Atalanta, GA, USA and Dr. Junyi Lei at Jdxpath in Collegeville, PA, USA for their valuable comments and suggestions. Funding This research was supported with the Grant from National Natural Science Foundation of China (No. 81071970). Conflict of interest

123

None.

References 1. Lu JC, Tao H, Zhang YQ et al (2008) Extent of prophylactic postoperative radiotherapy after radical surgery of thoracic esophageal squamous cell carcinoma. Dis Esophagus 21:502–507 2. Pai A, Sonawane S, Purandare NC et al (2008) Penile metastasis from esophageal squamous carcinoma after curative resection. Ann Thorac Cardiovasc Surg 14:238–241 3. Darnell JE Jr (1997) STATs and gene regulation. Science 277:1630–1635 4. Santos CI, Costa-Pereira AP (2011) Signal transducers and activators of transcription—from cytokine signaling to cancer biology. Biochim Biophys Acta 1816:38–49 5. Jatiani SS, Baker SJ, Silverman LR, Reddy EP (2010) Jak/STAT pathways in cytokine signaling and myeloproliferative disorders: approaches for targeted therapies. Genes Cancer 1:979–993 6. Hazan-Halevy I, Harris D, Liu Z et al (2010) STAT3 is constitutively phosphorylated on serine 727 residues, binds DNA, and activates transcription in CLL cells. Blood 115:2852–2863 7. Rani MR, Croze E, Wei T et al (2010) STAT-phosphorylationindependent induction of interferon regulatory factor-9 by interferon-beta. J Interferon Cytokine Res 30:163–1670 8. Escudero-Lourdes C, Medeiros MK, Ca´rdenas-Gonza´lez MC, Wnek SM, Gandolfi JA (2010) Low level exposure to monomethylarsonous acid-induced the over-production of inflammationrelated cytokines and the activation of cell signals associated with tumor progression in an urothelial cell model. Toxicol Appl Pharmacol 244:162–173 9. Abdulghani J, Gu L, Dagvadorj A et al (2008) Stat3 promotes metastatic progression of prostate cancer. Am J Pathol 172:1717–1728 10. Bourguignon LYW, Peyrollier K, Xia W, Gilad E (2008) Hyaluronan–CD44 interaction activates stem cell marker nanog, Stat3-mediated MDR1 gene expression, and ankyrin-regulated multidrug efflux in breast and ovarian tumor cells. J Biol Chem 283:17635–17651 11. Zhang M, Zhu GY, Gao HY, Zhao SP, Xue Y (2011) Expression of tissue levels of matrix metalloproteinases and tissue inhibitors of metalloproteinases in gastric adenocarcinoma. J Surg Oncol 103:243–247 12. Roomi MW, Monterrey JC, Kalinovsky T, Rath M, Niedzwiecki A (2010) Inhibition of invasion and MMPs by a nutrient mixture in human cancer cell lines: a correlation study. Exp Oncol 32:243–248 13. Groblewska M, Mroczko B, Gryko M, Ke˛dra B, Szmitkowski M (2010) Matrix metalloproteinase 2 and tissue inhibitor of matrix metalloproteinases 2 in the diagnosis of colorectal adenoma and cancer patients. Folia Histochem Cytobiol 48:564–571 14. He Y, Liu XD, Chen ZY et al (2007) Interaction between cancer cells and stromal fibroblasts is required for activation of the uPAR-uPA-MMP-2 cascade in pancreatic cancer metastasis. Clin Cancer Res 13:3115–3124 15. Gupta GP, Nguyen DX, Chiang AC et al (2007) Mediators of vascular remodeling co-opted for sequential steps in lung metastasis. Nature 446:765–770 16. Xie T, Wei D, Liu M et al (2004) Stat3 activation regulates the expression of matrix metalloproteinase-2 and tumor invasion and metastasis. Oncogene 23:3550–3560 17. Arnold MM, Srivastava S, Fredenburgh J, Stockard CR, Myers RB, Grizzle WE (1996) Effects of fixation and tissue processing on immunohistochemical demonstration of specific antigens. Biotech Histochem 71:23–224 18. Garner MM, Revzin A (1986) The use of gel electrophoresis to detect and study nucleic acid–protein interactions. Trends Biol Sci 11:395–396

Mol Biol Rep 19. Fried MG (1989) Measurement of protein–DNA interaction parameters by electrophoresis mobility shift assay. Electrophoresis 10:366–376 20. Becker S, Groner B, Muller CW (1998) Three-dimensional structure of the Stat3beta homodimer bound to DNA. Nature 394(6687):145–151 21. Brocklyn JR (2010) Regulation of cancer cell migration and invasion by sphingosine-1-phosphate. World J Biol Chem 1:307–312 22. Eigentler TK, Figl A, Krex D et al (2011) Number of metastases, serum lactate dehydrogenase level, and type of treatment are prognostic factors in patients with brain metastases of malignant melanoma. Cancer 117:1697–1703 23. Lee SW, Jung KH, Jeong CH et al (2011) Inhibition of endothelial cell migration through the down-regulation of MMP-9 by A-kinase anchoring protein 12. Mol Med Rep 4:145–149 24. Stearns D, Chaudhry A, Abel TW, Burger PC, Dang CV, Eberhart CG (2006) c-Myc overexpression causes anaplasia in medulloblastoma. Cancer Res 66:673–681 25. He G, Karin M (2011) NF-jB and STAT3—key players in liver inflammation and cancer. Cell Res 21:159–168 26. Rosen DG, Mercado-Uribe I, Yang G et al (2006) The role of constitutively active signal transducer and activator of transcription 3 in ovarian tumorigenesis and prognosis. Cancer 107:2730–2740 27. Tsareva SA, Moriggl R, Corvinus FM et al (2007) Signal transducer and activator of transcription 3 activation promotes invasive growth of colon carcinomas through matrix metalloproteinase induction. Neoplasia 9:279–291 28. Proietti CJ, Be´guelin W, Flaque´ MC et al (2011) Novel role of signal transducer and activator of transcription 3 as a progesterone receptor coactivator in breast cancer. Steroids 76:381–392 29. Monnien F, Zaki H, Borg C et al (2010) Prognostic value of phosphorylated STAT3 in advanced rectal cancer: a study from 104 French patients included in the EORTC 22921 trial. J Clin Pathol 63:873–878 30. Chang KC, Wu MH, Jones D, Chen FF, Tseng YL (2006) Activation of STAT3 in thymic epithelial tumors correlates with tumor type and clinical behavior. J Pathol 210:224–233 31. Kusaba T, Nakayama T, Yamazumi K et al (2005) Expression of p-STAT3 in human colorectal adenocarcinoma and adenoma; correlation with clinicopathological factors. J Clin Pathol 58:833–838 32. Yang CH, Fan M, Slominski AT, Yue J, Pfeffer LM (2010) The role of constitutively activated STAT3 in B16 melanoma cells. Int J Infereron Cytokine Mediator Res 2:1–7

33. Ishdorj G, Johnston JB, Gibson SB (2010) Inhibition of constitutive activation of STAT3 by curcurbitacin-I (JSI-124) sensitized human B-leukemia cells to apoptosis. Mol Cancer Ther 9:3302–3314 34. Chen J, Wang J, Lin L et al (2012) Inhibition of STAT3 signaling pathway by nitidine chloride suppressed the angiogenesis and growth of human gastric cancer. Mol Cancer Ther 11:277–287 35. Wu ZS, Cheng XW, Wang XN, Song NJ (2011) Prognostic significance of phosphorylated signal transducer and activator of transcription 3 and suppressor of cytokine signaling 3 expression in human cutaneous melanoma. Melanoma Res 21:483–490 36. Qian WF, Guan WX, Gao Y et al (2011) Inhibition of STAT3 by RNA interference suppresses angiogenesis in colorectal carcinoma. Braz J Med Biol Res 44:1222–1230 37. Silver DL, Naora H, Liu J, Cheng W, Montell DJ (2004) Activated signal transducer and activator of transcription (STAT) 3: localization in focal adhesions and function in ovarian cancer cell motility. Cancer Res 64:3550–3558 38. Xie TX, Huang FJ, Aldape KD et al (2006) Activation of Stat3 in human melanoma promoter brain metastasis. Cancer Res 66:3188–3196 39. Scuto A, Krejci P, Popplewell L et al (2011) The novel JAK inhibitor AZD1480 blocks STAT3 and FGFR3 signaling, resulting in suppression of human myeloma cell growth and survival. Leukemia 25:538–550 40. Che Y, Hou S, Kang Z, Lin Q (2009) Serenoa repens induces growth arrest and apoptosis of human multiple myeloma cells via inactivation of STAT 3 signaling. Oncol Rep 22:377–383 41. Huang C, Yang G, Jiang T, Cao J, Huang KJ, Qiu ZJ (2011) Down-regulation of STAT3 expression by vector-based small interfering RNA inhibits pancreatic cancer growth. World J Gastroenterol 17:2992–3001 42. Blaskovich MA, Sun J, Cantor A, Turkson J, Jove R, Sebti SM (2003) Discovery of JSI-124 (cucurbitacin I), a selective Janus kinase/signal transducer and activator of transcription 3 signaling pathway inhibitor with potent antitumor activity against human and murine cancer cells in mice. Cancer Res 63:1270–1279 43. Huang S, Chen M, Shen Y et al (2012) Inhibition of activated Stat3 reverses drug resistance to chemotherapeutic agents in gastric cancer cells. Cancer Lett 315:198–205 44. Glienke W, Maute L, Wicht J, Bergmann L (2010) Curcumin inhibits constitutive STAT3 phosphorylation in human pancreatic cancer cell lines and downregulation of survivin/BIRC5 gene expression. Cancer Invest 28:166–171

123