Clinical and biological significance of tissue transglutaminase in ovarian carcinoma

Abstract

Tissue type transglutaminase (TG2) is a unique multifunctional protein that plays a role in many steps in the cancer metastatic cascade. Here, we examined the clinical (n = 93 epithelial ovarian cancers) and biological (in vitro adhesion, invasion, and survival and in vivo therapeutic targeting) significance of TG2 in ovarian cancer. The overexpression of TG2 was associated with significantly worse overall patient survival in both univariate and multivariate analyses. Transfection of TG2 into SKOV3ip1 cells promoted attachment and spreading on fibronectin-coated surfaces and increased the in vitro invasive potential of these cells. Conversely, TG2 silencing with small interfering RNA (siRNA) of HeyA8 cells significantly decreased the invasive potential of the cells and also increased docetaxel-induced cell death. In vivo therapy experiments using chemotherapy-sensitive (HeyA8) and chemotherapy-resistant (HeyA8-MDR and RMG2) models showed significant antitumor activity both with TG2 siRNA-1,2-dioleoyl-sn-glycero-3-phosphatidylcholine alone and in combination with docetaxel chemotherapy. This antitumor activity was related to decreased proliferation and angiogenesis and increased tumor cell apoptosis in vivo. Taken together, these findings indicate that TG2 overexpression is an adverse prognostic factor in ovarian carcinoma and TG2 targeting may be an attractive therapeutic approach.

Figure 1

A) Western blot showing basal expression of TG2 protein in 11 ovarian cancer cell lines and one non-transformed normal ovarian epithelial cell line (HIO-180). B) Transglutaminase enzymatic activity according to TG2 expression levels. Mean enzymatic activity of TG2 was determined in the cell extract by studying Ca²⁺-dependent incorporation of [³H]putrescine into dimethylcasein, as described in Materials and methods. Error bars represent S.D. C) Representative immunohistochemical staining of clinical specimens for TG2 expression. All pictures were taken at original magnification × 200. D) Kaplan–Meier survival curve of patients with invasive ovarian cancer according to TG2 expression.

Figure 2

A) TG2 was ectopically introduced into SKOV3ip1 cells by adenovirus containing TG2 gene. Effects of TG2 overexpression on adhesion were examined following transfection of wild type TG2 gene (TG2-W) or empty vector (EV) into SKOV3ip1 cells. Control represents untransfected cells. Left panel pictures were obtained from fibronectin-coated plates. Right panel presents mean optical density. Error bars represent S.D. B) Effect of wild type TG2 gene (TG2-W) versus mutant type TG2 gene (TG2-M) on SKOV3ip1 invasive potential was examined using a Matrigel-transwell assay. Each experiment was performed in triplicate and repeated at least two times. C) SKOV3ip1 cells infected with TG2-W or without any treatment were plated on 96 well plates. After attachment, the medium was exchanged with increased concentrations of paclitaxel (0 to 25 nM). The number of viable cells remaining was determined after 72 hours. Error bars represent S.D. D) Western blot analysis for phosphorylated Akt (pAkt, ser⁴⁷³) and total Akt (tAkt) in SKOV3ip1 cells transfected with TG2-W or EV.

Figure 3

A) In vitro knockdown of TG2 in HeyA8 cells showed that maximum down-regulation was seen at 48 hours. B) Effect of TG2 gene silencing on in vitro HeyA8 invasive potential. Each experiment was performed in triplicate and repeated at least two times. C) Effect of TG2 silencing on HeyA8 cell viability was examined using various concentrations of docetaxel (0 to 50 nM). The number of viable cells remaining were determined after 72 hours. Error bars represent S.D. D) TG2 expression was assessed in HeyA8 tumors growing in the peritoneal cavity of nude mice following 2 doses of i.p. TG2 siRNA-DOPC using. Upper panel, Western blot. Lower panel, immunohistochemical peroxidase staining for TG2. All pictures were taken at original magnification x100.

Figure 4

Therapeutic efficacy of TG2 siRNA with docetaxel. Nude mice were injected i.p. with HeyA8 (A) or HeyA8-MDR (B) or RMG2 (C) cells and randomly allocated to one of following groups (therapy beginning 1 week after tumor cell injection): empty liposomes, control siRNA-DOPC, control siRNA-DOPC with docetaxel, TG2 siRNA-DOPC, TG2 siRNA-DOPC with docetaxel. The animals were sacrificed when control mice became moribund (3–4 weeks after starting therapy) and necropsy was done. Error bars represent S.E.

Figure 5

A) Effects of TG2-targeted therapy on TG2 expression (TG2; original magnification × 100); B) proliferation (PCNA; original magnification × 100); C) microvessel density (CD31; original magnification × 100), and D) apoptosis (green nuclei by TUNEL staining; blue nuclei by Hoechst; original magnification × 200). The graphs correspond to the labeled columns for immunohistochemistry. Error bars represent S.D.