Claudin-4 activity in ovarian tumor cell apoptosis resistance and migration

Abstract

Background: Claudin-4 is a transmembrane protein expressed at high levels in the majority of epithelial ovarian tumors, irrespective of subtype, and has been associated with tumor cells that are both chemoresistant and highly mobile. The objective of this study was to determine the functional role that claudin-4 plays in apoptosis resistance and migration as well as the therapeutic utility of targeting claudin-4 activity with a small mimic peptide.

Methods: We examined claudin-4 activity in human ovarian tumor cell lines (SKOV3, OVCAR3, PEO4) using in vitro caspase and scratch assays as well as an in vivo mouse model of ovarian cancer. Claudin-4 activity was disrupted by treating cells with a small peptide that mimics the DFYNP sequence in the second extracellular loop of claudin-4. Claudin-4 expression was also altered using shRNA-mediated gene silencing.

Results: Both the disruption of claudin-4 activity and the loss of claudin-4 expression significantly increased tumor cell caspase-3 activation (4 to 10-fold, respectively) in response to the apoptotic inducer staurosporine and reduced tumor cell migration by 50 %. The mimic peptide had no effect on cells that lacked claudin-4 expression. Female athymic nude mice bearing ZsGreen-PEO4 ovarian tumors showed a significant decrease in ovarian tumor burden, due to increased apoptosis, after treatment with intraperitoneal injections of 4 mg/kg mimic peptide every 48 h for three weeks, compared to control peptide treated mice.

Conclusion: Claudin-4 functionally contributes to both ovarian tumor cell apoptosis resistance and migration and targeting extracellular loop interactions of claudin-4 may have therapeutic implications for reducing ovarian tumor burden.

Keywords: Apoptosis; Claudin-4; Motility; Ovarian cancer.

Fig. 1

Disruption of claudin-4. Claudin-4 structure (a). Claudin-4 is a tetraspanin molecule with two extracellular loops. The second extracellular loop of claudin-4 contains the DFYNP sequence that is targeted by our small DFYNP mimic peptide (b). An inactive peptide with a glycine substitution for phenylalanine was used as a control peptide. Claudin-4 expression was reduced in ovarian tumor cells treated with claudin-4 shRNA. Representative Western blot analysis of claudin-4 expression (c), and β-actin as a loading control, was performed on lysates from SKOV3, OVCAR3, and PEO4 cells. SKOV3 cells were treated with control empty vector shRNA lentiviral suspension or claudin-4 shRNA lentivirus for 18 h before washing cells and treating with puromycin. Control shRNA treatment did not alter claudin-4 expression and claudin-4 shRNA drastically reduced claudin-4 expression compared to SKOV3 cells not treated with shRNA

Fig. 2

Claudin-4 promotes tumor cell survival. SKOV3, OVCAR3, SKOV3_contKD, and SKOV3_cld4KD cells were treated with 400 μM control peptide (“cont peptide”), 400 μM claudin mimic peptide (“DFYNP”), 2 μM staurosporine (“SS”, known inducer of apoptosis) or DFYNP + staurosporine (“DFYNP + SS”) for 18 h before being fixed and stained with dapi (to identify nuclei) and fluorescent antibody directed to active caspase-3. Representative fluorescent images of nuclei (dapi staining) and activated caspase-3 for each treatment group are shown for the claudin-4 expressing and claudin-4 knockdown SKOV3 cells (a). Percent of cell population positive for casp-3 activation was plotted (b-e). Mean ± s.e.m, n = 4 per treatment group, **p < 0.01, ***p < 0.001 vs. control peptide treated and ^‡ p < 0.02, ^‡‡ p < 0.01, ^‡‡‡ p < 0.001 for staurosporine only vs. DFYNP plus staurosporine treatment

Fig. 3

Disrupted claudin-4 diminishes ovarian tumor cell motility. Monolayers of SKOV3 (a, b) and OVCAR3 (c) cells or SKOV3 control (d) or claudin-4 knockdown (a, e), via shRNA, were scratched and then incubated at 37 °C in the presence of 400 μM control peptide (“Cont peptide”) or 400 μM claudin mimic peptide (“DFYNP”) for 8 (SKOV3) or 18 (OVCAR3) hours. Representative fluorescent images of nuclei (dapi staining) of scratched monolayers at time 0 and 8 h post scratch are shown for claudin-4 expressing and claudin-4 knockdown SKOV3 cells (a). Distance across the gap remaining was measured immediately following scratching and at 8 or 18 h after treatment using SlideBook software. Yellow dashed lines represent boundaries of wound. Percent closure of the wound was plotted (b-e). Mean ± s.e.m, n = 4 per treatment group, **p < 0.01, ***p < 0.001 vs. control peptide treated

Fig. 4

Targeting claudin-4 interactions reduces tumor burden in vivo. ZsGreen-PEO4 ovarian tumor were treated with 400 μM control peptide (“cont peptide”), 400 μM claudin mimic peptide (“DFYNP”), 2 μM staurosporine (“SS”, known inducer of apoptosis) or DFYNP + staurosporine (“DFYNP + SS”) for 18 h before being fixed and stained with dapi (to identify nuclei) and fluorescent antibody directed to active caspase-3. a Percent of cell population positive for casp-3 activation was plotted. Mean ± s.e.m, n = 4 per treatment group, **p < 0.01, ***p < 0.001 vs. control peptide treated and ^‡‡‡ p < 0.001 for staurosporine only vs. DFYNP plus staurosporine treatment. Female athymic nude mice with fluorescent ZsGreen-PEO4 human ovarian tumors were treated with 4 mg/kg control peptide or 4 mg/kg DFYNP (i.p. injection) every 48 h for three weeks. b Representative IVIS images of ovarian tumors taken before and after treatment. c Immunohistochemical analysis of claudin-4 expression in ZsGreen-PEO4 tumors in vivo. Inlay shows same section of tissue without primary antibody. d Quantitative analysis of change in photon flux, a measure of tumor size, before and after treatment. Mean ± s.e.m, n = 8-10 animals per group, *p < 0.05, ***p < 0.001 vs. control-treated

Fig. 5

Targeting claudin-4 interactions induces apoptosis in tumors in vivo. Female athymic nude mice with fluorescent ZsGreen-PEO4 human ovarian tumors were treated with 4 mg/kg control peptide or 4 mg/kg DFYNP (i.p. injection) every 48 h for three weeks. Remaining tumors were removed after three weeks of treatment. (a) Representative images of cleaved caspase-3 in tumors from mice treated with control peptide vs. DFYNP peptide. (b) Percent cell population positive for cleaved caspase-3 was calculated for control peptide and DFYNP treated tumors remaining after treatment. Mean ± s.e.m, n = 11-15 animals per group, *p < 0.05, ***p < 0.001 vs. control-treated