| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Preclinical |
Authors' Affiliations: 1 Department of Gynecology and Obstetrics, Graduate School of Medical Sciences, and 2 Department of Molecular Genetics, Division of Molecular and Cell Therapeutics, Medical Institute of Bioregulation, Kyushu University, Fukuoka, Japan; 3 Division of Otorhinolaryngology, Shinshu University School of Medicine, Matsumoto, Japan; and 4 Department of Molecular Oncology, Institute on Aging and Adaptation, Shinshu University Graduate School of Medicine, Matsumoto, Japan
Requests for reprints: Hiroaki Kobayashi, Department of Gynecology and Obstetrics, Graduate School of Medical Sciences, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812-8582, Japan. Phone: 81-92-642-5395; Fax: 81-92-642-5414; E-mail: koba{at}med.kyushu-u.ac.jp.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-smooth muscle actin in the surrounding stromal cells. The purpose of this study is to evaluate the efficacy of i.p. CNh1 gene therapy against peritoneal dissemination of ovarian cancer.
Experimental Design: We used an adenoviral vector to induce the CNh1 gene into peritoneal cells and ovarian cancer cells as a means of enhancing or inducing the expression of -smooth muscle actin as well as CNh1. The efficacy of gene transfer was examined by in vitro cell culture and in vivo animal experiments.
Results: The formation of longer and thicker actin fibers was observed in each transfected cell line, and the localization of these fibers coincided with that of externally transducted CNh1. With respect to changes in cell behavior, the CNh1-transfected peritoneal cells acquired an ability to resist ovarian cancer-induced shrinkage in cell shape; thus, cancer cell invasion through the monolayer of peritoneal cells was inhibited. In addition, CNh1-transfected ovarian cancer cells showed suppressed anchorage-independent growth and invasiveness, the latter of which accompanied impaired cell motility. The concomitant CNh1 transfection into both peritoneal cells and ovarian cancer cells produced an additive inhibitory effect with respect to cancer cell invasion through the peritoneal cell monolayer. By in vivo experiments designed to treat nude mice that had been i.p. inoculated with ovarian cancer cells, we found that the i.p. injected CNh1 adenovirus successfully blocked cancer-induced morphologic changes in peritoneal cell surface and significantly prolonged the survival time of tumor-bearing mice. Moreover, CNh1 adenovirus could successfully enhance the therapeutic effect of an anticancer drug without increase in side effects.
Conclusions: Thus, CNh1 gene therapy against peritoneal dissemination of ovarian cancer is bifunctionally effective (i.e., through inhibitory effects on the infected peritoneal cell layers that suppress cancer invasion and through direct antitumor effects against invasion and growth properties of cancer cells).
-actin, myosin light and heavy chain kinases, and calponin h1 (CNh1)]. As such, this seems to be a reflection of the importance of actin cytoskeletal disorganization in the metastatic process.
CNh1 is a 34-kDa actin-binding protein that was originally isolated from chicken gizzard (2). CNh1 is mainly expressed in smooth muscle cells in contrast to other two isoforms of calponin h2 and acidic calponin, which are mainly expressed in nonmuscle cells and the brain, respectively (3–5). CNh1 has an ability to (a) bind to the thin filament of actin, tropomyosin, and calmodulin (2, 6, 7); (b) inhibit the actin-activated myosin Mg-ATPase (8); (c) inhibit Ca2+-dependent mobility of actin on immobilized myosin (9); and (d) induce conformational changes in actin filament (F-actin; ref. 10). Therefore, CNh1 is thought to play an essential role in organizing stable actin stress fibers. We have confirmed previously that (a) the expression of -smooth muscle actin (-SMA) decreased in blood vessels located at the proximity of ovarian cancer nests (11) and in fibroblasts and peritoneal mesothelia cultured in the presence of ovarian cancer cell–derived factors5 and (b) the decrease in CNh1 and -SMA expression was seemingly attributed to the secreted factors, including platelet-derived growth factor derived from cancer cells (12, 13). CNh1 knockout mice showed the enhanced peritoneal dissemination and lung metastasis by malignant melanoma cells through their highly fragile peritoneum and vascular wall in contrast to the wild-type mice. However, CNh1 gene transfection into the peritoneal cells of knockout mice could successfully inhibit cancer cell invasion into peritoneal cell layer (14, 15). The efficacy of CNh1 transfection into cancer cells themselves was also reported by our colleagues; both cell growth and tumorigenicity were significantly inhibited in CNh1-transfected leiomyosarcoma cells (16) and fibrosarcoma cells (17).
Based on these observations, we hypothesized that CNh1 has bifunctional effects (i.e., an enhancement of peritoneal defense ability on the one hand and a direct inhibitory effect against for ovarian cancer cells on the other). The purpose of this study was to examine the effects of CNh1 gene transduction into both cancer cells and peritoneal cells with respect to inhibiting peritoneal dissemination of ovarian cancers.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Female athymic BALB/c nu/nu mice between 6 and 8 weeks old (Charles River Japan, Atsugi, Japan) were used for in vivo experiments. The mice were maintained in a laminar-flow cabinet under specific pathogen-free conditions while receiving standard feed and water ad libitum. Our experiments were reviewed by the Committee of Ethics in Animal Experiments in the Graduate School of Medical Sciences, Kyushu University, and the Law (no. 105) and the Notification (no. 6) of the Government.
Construction of recombinant adenovirus and adenoviral gene transfection into cultured cells. Recombinant adenovirus was inserted with the CNh1-green fluorescent protein (GFP) fusion gene (AdCNh1) or with the GFP gene only (AdGFP, control vector) as described previously (15). Briefly, CNh1-GFP fusion gene was produced by inserting human CNh1 gene into pEGFP-C2 (Clontech, Palo Alto, CA), and then subcloned under the transcriptional control of CAG promoter/enhancer in cosmid vector pAxCAwt (adenovirus expression kit, TaKaRa, Japan; Fig. 1 ). The recombinant was continued in 293 cells by four times infection and high titer virus was obtained. Adenoviral infection into the cells was carried out by incubating subconfluent cells with adenoviral vectors (2 x 107 plaque-forming units/15-mm dish) for 2 hours in RPMI 1640 at 37°C. Cells were then washed twice with PBS and the medium was changed into RPMI 1640 containing 10% fetal bovine serum. After 24 hours of incubation, cells were used for immunofluorescent cell staining and colony-forming assay. For Western blot analysis, in vitro invasion assay, and cell motility assay, cells were used after 4 days of additional incubation.
|
Western blot analysis. Subconfluently growing cells were lysed in a lysis buffer [50 mmol/L Tris-HCl (pH 8.0), 0.25 mol/L NaCl, 0.5% NP40, 1 mmol/L phenylmethylsulfonyl fluoride, 10 µg/mL aprotinin, 1 µg/mL leupeptin]. After centrifugation at 13,000 x g for 10 minutes to remove debris, equal protein amount of cell lysates was separated on SDS-PAGE and transferred onto nitrocellulose membranes. The membranes were blocked in TBS [10 mmol/L Tris-HCl (pH 7.4), 150 mmol/L NaCl, 0.05% Tween 20] containing 5% nonfat dry milk for 1 hour at room temperature and washed in TBS-Tween 20 for 5 minutes. The blots were incubated with anti-human CNh1 monoclonal antibody (Sigma, St. Louis, MO), anti-human -SMA monoclonal antibody (Progen, Queensland, Australia), anti-ß-actin monoclonal antibody (Sigma), antibody (Santa Cruz Biotechnology, Inc., CA) for 2 hours at room temperature. After washing, the blots were reacted with their respective secondary antibodies (Amersham, Piscataway, NJ) and analyzed with enhanced chemiluminescence system (Amersham). All primary antibodies are informed to detect CNh1 protein stably even in murine tissues.
Immunofluorescent cell staining for F-actin. Cells were washed with PBS and fixed in 3.7% formaldehyde solution in PBS for 10 minutes at room temperature. After permeabilization with 1% Triton X-100, the rhodamine-phalloidin (Molecular Probes, OR) diluted with 0.1% bovine serum albumin in PBS was added and incubated at room temperature for 20 minutes. F-actin and GFP were observed with confocal laser scanning microscope using specific filters for rhodamine and FITC, respectively.
Soft-agar colony-forming assay. After 24 hours of infection with AdCNh1 or AdGFP, 104 cells were seeded into the complete medium, including 0.3% agar, and placed over a hardened 0.5% agar base layer in 60-mm dishes. Visible colonies were counted in triplicate after 14 days of incubation to evaluate the anchorage-independent cell growth.
Tumorigenicity in nude mice. Single-cell suspension containing 107 SKOV3 cells with or without plasmid transfection was i.m. injected into the thigh of nude mice. The mice were observed weekly, and tumor growth was evaluated by measuring the thickness of the inoculated thigh.
Cell motility assay. Cell motility was determined using Transwell chambers inserted with 8-µm pore size membrane (Costar, Corning, NY) according to the method described previously (19). Each lower compartment of Transwell chambers was filled with a conditioned medium as a source of chemoattractant, which was a supernatant of confluently cultured NIH3T3 cells in DMEM supplemented with 10% fetal bovine serum for 24 hours. Single-cell suspension containing 5 x 104 cells infected by AdCNh1 or AdGFP in 100 µL RPMI 1640 with 0.1% bovine serum albumin was placed in its upper compartment. The cells were incubated for 12 hours at 37°C, fixed with methanol, and stained with H&E. Cells on the upper surface of the filter were removed with a cotton swab, and cells that migrated to the lower surface were counted by six fields of a light microscope at x200 magnification.
In vitro invasion assay. In vitro invasion assay was done using Transwell chambers with 8-µm pore membrane coated with 20 µg Matrigel (Becton Dickinson Collaborative Research, Bedford, MA) as described previously (20). Conditioned medium prepared by NIH3T3 cells was placed in the lower compartment. Single-cell suspension containing 105 cells in 100 µL RPMI 1640 containing 10% fetal bovine serum was placed in the upper compartment and incubated for 24 hours. The subsequent procedures were the same as those of the cell motility assay.
Assay for invasion through the peritoneal cell monolayer. Single-cell suspensions containing 106 cancer cells infected with AdCNh1 or AdGFP were plated on the monolayer of CCL14 peritoneal cells infected with AdCNh1 or AdGFP, and cultured for additional 24 hours. The number of colonies derived from a penetrated cancer cell was counted by six fields of a phase-contrast microscope at x400 magnification.
Therapeutic experiments against i.p. inoculated ovarian cancer cells. One day after the inoculation of 106 cells of OVAS-21/om, the i.p. injection of AdCNh1 or AdGFP (2 x 108 plaque-forming units/2 mL) was started and repeated by every 3 days until 19 days after the inoculation. In some groups, 100 mg/kg paclitaxel was given once i.p. at 3 days after inoculation. Survival and body weight of each mouse were monitored until 5 months after the inoculation, and an autopsy was done. Paclitaxel (Taxol) was kindly provided by Bristol-Myers Squibb (Tokyo, Japan).
Scanning electron microscope. Peritoneum was resected from abdominal walls of mice at 7 days after the i.p. inoculation of SKOV3i.p.1 cancer cells followed by the adenoviral injection. Peritoneal tissues were fixed in 1% glutaraldehyde for 12 hours followed by 1-hour fixation in 1% osmium tetroxide. After dehydration in ethanol, the specimens were rinsed in isoamyl acetate and dried by a critical point drying method. The dried specimens were mounted on copper plates and coated with osmium in an osmium plasma coater (Nippon Laser and Electronics Lab, Japan). The specimens were examined with a JSM-6000F scanning electron microscope (Jeol, Pleasonton, CA) at 1.5 kV.
Statistical analysis. Mann-Whitney U test was used to assess the statistical significance of differences in soft-agar colony-forming assay, cell motility assay, in vitro invasion assay, and assay for cell invasion through the peritoneal cell layer. In vivo tumor growth assay was assessed using Fisher's exact test. Survival curves were analyzed with Kaplan-Meier method and the difference between curves was assessed according to the log-rank test. All Ps are two-sided and considered statistically significant at <0.05.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-SMA, the expressions of , ß, and 
actin proteins were analyzed after CNh1 adenoviral infection. No change of expression of any actin was observed between AdGFP-infected cells and their parental noninfected cells. Although expression of ß and actins were not changed even by AdCNh1 infection, -SMA expression was clearly enhanced by CNh1 transfection (Fig. 2A and B).
|
-SMA.5 Therefore, we examined whether CNh1 transfection into the peritoneal cells could stabilize actin filaments against the ovarian cancer-derived factors. As shown in Fig. 3A
, AdGFP-infected cells in the presence of ovarian cancer conditioned medium showed smaller and shrunken cell shape with thin and short actin filaments at the cytosol and small spike-like filaments observed at the cellular margins, which is similar in appearance to the noninfected FK cells in the presence of ovarian cancer conditioned medium. Therefore, AdGFP infection could not recover the FK cells from ovarian cancer-derived morphologic changes. On the other hand, AdCNh1-infected cells maintained large and extended cell shape with thick and long filaments traversing a cell, just like untreated parental peritoneal cells, even in the presence of ovarian cancer conditioned medium (Fig. 3B). Localization of actin stress fiber (Fig. 3B) was identical with the expression site of CNh1-GFP gene products (Fig. 3C), which was confirmed by the yellow fibers consisting of merged observation for both proteins (Fig. 3D). Therefore, the actin stress fiber development in AdCNh1-infected cells seemed to be due to the exogenously transfected CNh1 gene.
|
-SMA band (Fig. 2C). Expression of both ß and actins was unchanged by CNh1 transfection into SKOV3 cells (Fig. 2D and E) as well as other ovarian cancer cell lines of SHIN-3 and MCAS/as (Fig. 2E). In each cell line, AdCNh1 infection induced the high-intensity band of -SMA compared with the "invisible" band in AdGFP-infected cells (Fig. 2E) as well as noninfected cells. Changes of actin filaments and cell shapes occurred in CNh1-transfected ovarian cancer cells. Similar to noninfected SKOV3 cell line, AdGFP-infected SKOV3 cells showed weakly and diffusely stained actin filaments (Fig. 3E), the location of which was not identical with that of exogenously transducted GFP (Fig. 3F and G). AdCNh1 infection induced thick and long actin stress fibers and changed the cell shape into a flat and extended phenotype (Fig. 3H). The localization of the developed actin stress fibers was identical with the GFP expression sites, except the nuclear area (Fig. 3I and J), indicating that the exogenously introduced CNh1-GFP gene products produced the stable actin fibers. Similar results were observed in other AdCNh1-infected ovarian cancer cell lines (data not shown).
Decreased growth property in CNh1-transfected ovarian cancer cells. We confirmed that CNh1 transfection changed neither cell growth rate nor colony-forming ability of ovarian cancer cells when growing as a monolayer culture system (data not shown). Therefore, we evaluated anchorage-independent cell growth by soft-agar colony-forming assay and found that all ovarian cancer cell lines infected with AdCNh1 formed markedly fewer colonies compared with the case of AdGFP infection (Fig. 4A ).
|
Decreased invasiveness and cell locomotion in CNh1-transfected ovarian cancer cells. We evaluated the effect of CNh1 transfection on the invasiveness by an in vitro assay. Invasion of AdCNh1-infected SKOV-3, SHIN-3, and MCAS/as cells were all significantly decreased by 67.6%, 43.1%, and 27.6%, respectively, in contrast to AdGFP-infected controls (Fig. 5A ). Because the change of cell morphology strongly correlates with cell locomotion, we examined the effects of AdCNh1 on cell motility using Matrigel-uncoated Transwell chambers. The cell motility of AdCNh1-infected each cell lines was significantly decreased by 50.5%, 44.3%, and 29.1%, respectively, in contrast to each AdGFP-infected control (Fig. 5B).
|
Effects of AdCNh1 infection on morphologic changes of peritoneal cell surface induced by ovarian cancer cells. Scanning electron microscope showed that the i.p. inoculation of ovarian cancer cells changed the surface of murine peritoneal cells from a smooth phenotype without spikes (Fig. 6A ) into an "edgy" one with numerous small spikes. This microvilli-like cell surface change was unavoidable by i.p. infection of AdGFP following cancer cell inoculation (Fig. 6B). However, i.p. injection of AdCNh1 with cancer inoculation could avoid this change, resulting in a flat and smooth surface of peritoneal cells (Fig. 6C) as well as a normal surface observed in the absence of cancer cells (Fig. 6A).
|
To examine the effects of AdCNh1 combined with the administration of paclitaxel, a highly effective anticancer drug of the present mainstream for ovarian cancer treatment, 100 mg/kg paclitaxel was i.p. injected 3 days after cancer cell inoculation. As shown in Fig. 6E, although AdGFP failed to prolong the survival of paclitaxel-treated mice, additionally prolonged survival was obtained in the combined therapy by AdCNh1 and paclitaxel (P = 0.04, AdCNh1 + paclitaxel versus paclitaxel alone). No aggravation was observed in toxicity by either viral administration to paclitaxel-treated mice, which transiently showed 3% body weight loss on an average.
Autopsy was done as early as possible after each mouse died from peritoneal dissemination. All of dead mice showed the distended abdomen by a massive bloody ascites and the disseminated numerous implants on the surface of i.p. organs and peritoneum (Fig. 6F). In contrast, one of six mice treated by AdCNh1 and three of five mice treated by AdCNh1 plus paclitaxel survived even 147 days after the cancer cell inoculation and showed no i.p. implants confirmed by slaughter (Fig. 6G).
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
CNh1 brought about therapeutic effects not only for host cells but also for ovarian cancer cells as well. Forced CNh1 expression in ovarian cancer cells developed actin stress fibers accompanying -SMA induction (Figs. 2C-E and 3H-J), which resulted in retardation of growth and invasiveness (Figs. 4 and 5A). As for growth-inhibitory effects mediated by CNh1, in vitro anchorage-independent cell growth (Fig. 4A) and in vivo tumor growth (Fig. 4B) were both suppressed, although in vitro growth was not inhibited when cells were cultured as monolayers. As for CNh1-inhibitory effect on invasiveness, cell motility was significantly retarded (Fig. 5B). These results would seem reasonable, considering that CNh1-induced stabilization of actin stress fibers inhibits both three-dimensional cell growth and cell locomotion, for which dynamic state of actin filaments is thought to be essential. In vivo efficacy of CNh1 specific to ovarian cancer cells was confirmed by CNh1-plasmid transfectants of SKOV3i.p.1, which survived much longer than the mock transfectant (data not shown).
Cotransfection of CNh1 into both peritoneal cells and ovarian cancer cells resulted in additive inhibition of cancer cell invasion through peritoneal cell layer compared with the inhibitory effect of CNh1 transfection into either cell population alone (Fig. 5C). In vivo therapeutic effects of CNh1 were confirmed by the i.p. administration of CNh1 adenoviral vector into nude mice bearing i.p. inoculated ovarian cancer cells (Fig. 6D). We observed that ovarian cancer cells seemed to easily invade through the peritoneal layer, which consisted of retracted and dissociated mesothelial cells with poor actin stress fiber development. This may be due to the down-regulation of CNh1 and -SMA induced by ovarian cancer cell–derived secretory factor(s), including platelet-derived growth factor.5 Effects of AdCNh1 used to infect both mesothelial cells and ovarian cancer cells are summarized as follows: (a) peritoneal cells with developed actin stress fiber had a greater ability to act as a barrier against cancer cells, avoiding ovarian cancer-induced suppression of CNh1 and -SMA expression, and (b) ovarian cancer cells impaired their growth and invasion properties by induced actin stress fibers. As such, adenovirus-mediated CNh1 gene therapy against peritoneal dissemination of ovarian cancer may be considered as a potentially novel therapeutic intervention, whereby one and the same gene has two distinctive major effects: one is to control the cancer cells and the other is to bolster a host defense (anti-invasive) mechanism.
Considering clinical application of this therapy, preventive administration of AdCNh1 just after the optimal surgery would be desirable for effective inhibition of peritoneal dissemination. At least, it seems to be difficult for this therapy to eradicate the established and large peritoneal implants. Because AdCNh1 could successfully enhance the therapeutic effect of paclitaxel without increase in side effects (Fig. 6E and G), CNh1 gene therapy may be safely repeatable and hence potentially to improve patient prognosis especially when used in combination with conventional postoperative chemotherapy. Gene therapy using i.v. given adenovirus can be compromised in situations where the injected adenovirus is easily and rapidly washed out from the tumor site by bloodstream (22). However, adenoviral gene therapy using i.p. injection has an advantage in that the adenovirus can stay at the site of target cells (cancer cells and peritoneal cells), owing to the closed space of the "abdominal cavity." With respect to the problem of decreasing efficacy in repeating adenovirus gene therapy by appearance of human antibody against adenovirus, we would like to expect developing more suitable i.p. vector system or finding a nontoxic small molecule that can effectively enhance CNh1 expression after i.p. administration.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: T. Ogura and H. Kobayashi contributed equally to this work.
Received 3/20/06; revised 6/ 8/06; accepted 6/15/06.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-smooth muscle actin in benign or malignant ovarian tumors. Gynecol Oncol 1993;48:308–13.[CrossRef][Medline]-actin. J Dermatol Sci 2000;23:170–7.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) showed retarded tumor growth compared with parental SKOV3 cells (
) or control plasmid-transfected SKOV/mock (
). Points, mean (n = 6 mice); bars, SD. *, P < 0.05, SKOV/CNh1 versus SKOV/mock in each measurement point.
; P = 0.04). F and G, i.p. appearances after the adenoviral treatment were shown by representative mice. An AdGFP-treated mouse, died 49 days after the inoculation, showed peritonitis carcinomatosa with massive ascites accompanied by numerous disseminated implants on the surface of i.p. organs and peritoneum (F). In contrast, an AdCNh1-treated mouse, which survived even 147 days after the inoculation, showed neither ascites nor i.p. implants (G). As for the side effects of AdCNh1 treatment, neither body weight loss nor intra-abdominal adhesion was observed.