Carcinogenesis is a multistep process involving complicated changes of proto-oncogenes and anti-oncogenes. Studies on cancer chemoprevention have led to remarkable insights into the molecular mechanisms of cancers. Carcinogenesis in humans has been shown to be inhibited by non-steroidal anti-inflammatory drugs (NSAID).1, 2, 3 The suppression of prostaglandin (PG) biosynthesis via inhibition of the corresponding cyclooxygenase (COX) by NSAID is thought to be the main molecular mechanism for their antineoplasia effect.4 Such data have prompted the examination of expression of COX in human cancer tissues. Two different isoforms of COX catalyze the synthesis of PG from arachidonic acid. COX-1 is constitutively expressed in most tissues, at a relatively stable level, and it exerts diverse homeostatic functions, such as protecting the gastrointestinal tract from injury and regulating renal blood flow. COX-2 is an inducible cyclooxygenase. It is not expressed in appreciable amounts by most normal tissues but can be induced by mitogenic stimuli, growth factors, cytokines, and carcinogens.5 COX-2 is thought to synthesize the excessive amount of PG associated with pain and fever. Studies have demonstrated upregulation of COX-2 expression in a number of human malignant tissues and its crucial role in these tumorigenesis and tumor progression.6, 7, 8, 9, 10, 11, 12, 13, 14, 15 The relationship between COX-2 and human renal cell carcinoma (RCC), however, remains unclear. COX-2 is known to be an important target gene of Ras,16, 17 and it has synergy with many other genes, such as P53 and Bcl-2,18, 19 which have been known to play crucial roles in the carcinogenesis and progression of RCC.20, 21 Khan22 reported high expression of COX-2 in canine RCC. Furthermore, recent reports have shown enhanced COX-2 expression in some human RCC.23, 24 Taken together, these data raise questions about the relationship between COX-2 and human RCC and the role of COX-2 in the carcinogenesis of human RCC cells. Our present study aims to address such questions by investigating expression of COX-2 in human RCC cell lines and its effect on proliferation in vitro and tumorigenesis in vivo. Our results demonstrated that overexpression of COX-2 exists in one RCC cell line, and that suppression of its expression by an antisense cDNA inhibits in vivo tumorigenesis of this RCC cell line.
Accumulating evidences indicate that cyclooxygenase (COX)-2 plays an important role in tumorigenesis in many human cancers. Yet the relationship between COX-2 and human renal cell carcinoma (RCC) remains unclear. The aim of our study was to evaluate COX-2 expression in human RCC cell lines and its role in tumorigenesis of human RCC. Among the human RCC cell lines (SMKT-R4, OS-RC-2, ACHN) and normal renal cell line RPTEC, COX-2 overexpression was found in OS-RC-2 cells both at mRNA and protein levels. COX-2 sense- and antisense-orientated vectors were constructed and transferred into RCC cells. Significant suppression of cellular proliferation was demonstrated in OS-RC-2 antisense transfectants, whereas promotion was found in SMKT-R4 sense transfectants by colony-forming assay despite the observation that COX-2 specific inhibitor NS398 exhibited similar IC50 among RCC cell lines by MTT assay. In comparison with parent cells and sense transfectants, significant suppression of COX-2 expression and PGE2 production and increase in butyrate-induced apoptosis were observed in OS-RC-2 antisense transfectants by Western blot, ELISA assay and FACS analysis, respectively. Furthermore, tumor growth and angiogenesis of OS-RC-2 antisense transfectants in nude mice was significantly suppressed and the survival time of these mice was significantly prolonged. Our study demonstrates that COX-2 is overexpressed in OS-RC-2 RCC cell line and plays an important role in tumorigenesis of the cells in vivo, which implies that COX-2 may be a therapeutic target for COX-2-expressing RCC, and that suppression of COX-2 expression by antisense-based strategy may have potential utility in treatment of COX-2-expressing RCC. © 2003 Wiley-Liss, Inc.
MATERIAL AND METHODS
The human RCC cell lines OS-RC-2 and ACHN were purchased from RIKEN cell bank (Tsukuba, Japan) and ATCC (Rockville, MD), respectively. The human RCC cell line SMKT-R4 was kindly provided by Dr. A. Takahashi (Sapporo Medical University, Japan). The normal human renal proximal tubule epithelial cell line RPTEC was purchased from Clonetics (Walkersville, MD). As a positive control, BxPC-III, a human pancreatic cancer cell line that expresses COX-2 at a high level, was purchased from ATCC. OS-RC-2 cells and BxPC-III cells were cultured in RPMI1640 medium (Sigma, St. Louis, MO) supplemented with 10% heat-inactivated FBS (Sigma), SMKT-R4 cells and ACHN cells were cultured in DMEM (Sigma) supplemented with 10% heat-inactivated FBS; and RPTEC cells were cultured in renal epithelial cell basal medium (REBM) supplemented with REGM SingleQuots (Clonetics). All cell lines were incubated at 37°C in a 5% CO2 humidified incubator.
Semiquantitative RT-PCR and real-time PCR
Total RNA was extracted using Trizol reagent according to the manufacturer's manual (GIBCO BRL, Grand Island, NY) and quantified spectrophotometrically. First strand cDNA synthesis was carried out with 2 μg of total RNA using the random priming extension method. After reverse transcription, RNA was digested with 2 U of RNase H (Promega, Madison, WI) at 37°C for 20 minutes. The entire cDNA of COX-2 was amplified by PCR with the use of β-actin as an internal control. The PCR was carried out under the following conditions: denaturation at 95°C for 4 minutes and 30 seconds; 33 cycles of denaturation at 95°C for 30 seconds, annealing at 55°C for 30 seconds and elongation at 72°C for 3 minutes, followed by a final elongation at 72°C for 10 minutes. The primers used for COX-2, which yields a 2004bp COX-2 cDNA, were 5′-CTCAGACAGCAAAGCCTACC -3′ (forward) and 5′-TGACTCCTTTCTCCGCAACA -3′ (reverse). As an internal control, a 449 bp fragment of β-actin was amplified using forward primer: 5′-TCACCCTGAAGTACCCCATC -3′ and reverse primer: 5′-CTCCTTAATGTCACGCACGA -3′. Ten microliters of the PCR product was subjected to 1% agarose gel electrophoresis and visualized by ethidium bromide staining. For real-time PCR, each cDNA (10 ng) was amplified in triplicate with the use of a SYBR-Green PCR assay kit (Qiagen GmbH, Hilden, Germany) and then detected using an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). PCR reactions were incubated for 2 min at 50°C and for 15 min at 95°C, followed by 40 amplification cycles with 15 sec denaturation at 95°C, 30 sec annealing at 60°C and 1 min extension at 72°C. β-actin was used to standardize the total amount of cDNA. The primers for real-time PCR were, COX-2 forward: 5′-ATCATTCACCAGGCAAATTGC-3′, COX-2 reverse: 5′-GGCTTCAGCATAAAGCGTTTG-3′, β-actin forward: 5′-GCTCCTCCTGAGCGCAAGT-3′ and β-actin reverse: 5′-TCGTCATACTCCTGCTTGCTGAT-3′, which yielded PCR products of 166 bp and 101 bp, respectively. Specificity of PCR was checked by analyzing melting curves. Relative mRNA levels were determined by comparing the PCR cycle threshold between cDNA of COX-2 and that of β-actin.
Cytoplasmic protein was isolated with cell lysis buffer (Cell Signaling Technology, Beverly, MA). To evaluate the phosphorylation status of Akt after NS398 treatment, the cells were treated with 100 μM NS398 for 0, 1, 4, 24, 48 or 96 hr and the cytoplasmic protein was isolated. Protein concentration was determined using the Bio-Rad protein reagent assay (Bio-Rad, Hercules, CA) with BSA as a standard. Samples containing 30μg of protein were denatured by incubating at 95°C for 4 min, then subjected to 12% (w/v) denatured sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to a polyvinylidene difluoride (PVDF) membrane (Amersham Pharmacia Biotech, Amersham, UK). After regular blocking and washing, the membrane was incubated with a mouse anti-human COX-2 antibody at a concentration of 0.5μg/ml (Transduction Lab., Lexington, KY) or rabbit anti-human Akt antibody or rabbit anti-human Akt 473 Ser antibody at 1:1000 (Cell Signaling Technology) at 4°C overnight. After incubation with a horseradish peroxidase conjugated second antibody (goat anti-mouse for COX-2 or goat anti-rabbit for Akt) at a concentration of 0.2 μg/ml (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at room temperature for 1 hr, protein was detected by enhanced chemiluminescence using ECL Plus western blotting detection reagents (Amersham Pharmacia Biotech). To confirm that equal amounts of protein were loaded in each lane and transferred efficiently, bound antibody were stripped off the membranes with stripping buffer (62.5 mM Tris-HCl, pH 6.7, 100 mM β-mercaptoethanol, 2% SDS) and the membranes were reprobed with a mouse anti-human actin antibody at a concentration of 0.2 μg/ml (Santa Cruz Biotechnology Inc.), followed by incubation with a secondary antibody and chemiluminescence detection as described above.
To evaluate the effects of COX-2 inhibitors on proliferation and viability of RCC cell lines, 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl-2H tetrazolium bromide (MTT) assay was carried out with minor modifications.25 NS398 (Cayman Chemical, Ann Arbor, MI), a COX-2 specific inhibitor, was used. We dissolved it in DMSO at 30 mM as the stock solution and then diluted it with the medium according to the manufacturer's instructions. Briefly, 1 × 103 cells of each cell line were seeded in 96-well microplates in triplicate and cultured for 24 hr. Then NS398 was added to the wells at various concentrations. The cells were cultured for another 96 hr and the medium in each well was then replaced with 200 μl of fresh medium containing 1 mg/ml MTT. After an additional incubation of 4 hr, the medium was discarded and 100 μl DMSO was added to each well to dissolve formazan crystals. Then the optical density was read with a Microplate Reader BIO-RAD Model 550 (Bio-Rad Laboratories) at 540 nm, and with the use of a reference wavelength at 490 nm.
Construction of COX-2 recombinant sense- and antisense-orientated expression vectors
The 2004 bp cDNA fragment containing the whole open reading frame for the human COX-2 polypeptide was cloned into pTARGET mammalian expression vector (Promega, Madison, WI) in sense orientation (pTARGET-S) or antisense orientation (pTARGET-AS). As a vector control for transfection, a circular non-T-tailed empty vector (pTARGET) was prepared through ligating open T-tailed pTARGET vector alone. The correct insertion and orientations were confirmed by EcoRI restriction mapping and BamHI together with AccB7I restriction mapping, respectively.
Colony forming assay
The assay was carried out with a modification of the technique described previously.26 Briefly, 1 × 105 cells were seeded in a 60-mm dish 1 day before transfection. Then 2 μg of vector DNA (pTARGET-S, pTARGET-AS and pTARGET) were transferred into RCC cells using 8 μl of PLUS reagent in 250 μl serum-free medium according to the manufacturer's instructions using Lipofectamine Plus (GIBCO BRL, Grand Island, NY). After 24-hr incubation, the media were replaced with fresh selective media containing geneticin (G418) (GIBCO BRL, Grand Island, NY) at predetermined concentrations. The selection medium was changed every 3 or 4 days until G418-resistant colonies appeared. Numbers of G418-resistant colonies (colonies with a diameter >1 mm) were counted after staining with Giemsa.
Establishment of stable transfectants with sense and antisense COX-2
OS-RC-2 cells were transfected with pTARGET-S, pTARGET-AS or pTARGET using Lipofectamine Plus (GIBCO BRL). To generate stable transfectant cell lines, transfected cells were selected with 700 μg/ml G418 (GIBCO BRL). G418-resistant single cell clones were isolated, and the COX-2 mRNA and protein expression level of each selected clone was confirmed by quantitative real-time PCR and Western blot.
PGE2 detection by ELISA
To examine PGE2 production in OS-RC-2 parent cells and transfectants, 3 × 105 cells were seeded into each well of a 6-well plate and cultured overnight. The medium was refreshed with serum-free RPMI-1640 the following day and cells were cultured for another 24 hr. Cell-free media were prepared by collecting supernatants and centrifugation to remove cell debris. The prepared cell-free supernatants were stored at −80°C. The PGE2 concentration was determined using a PGE2 ELISA kit (Neogen Corporation, Lexington, KY) according to the manufacturer's instructions.
Fluorescence-activated cell sorting (FACS) analysis of apoptosis
Parent cells (1 × 105) or transfectants of OS-RC-2 were seeded into 6-well plates. After replacing with fresh medium containing 5 mM sodium butyrate (Sigma) every 12 hr for 60 hr, cells were harvested using 0.05% trypsin/EDTA, washed with pre-chilled PBS and double-stained with propidium iodide (PI) and FITC-conjugated-Annexin V using an Annexin-V-FLUOS staining kit (Roche Molecular Biochemicals, Mannheim, Germany). The cells were analyzed with a FACScalibur (Becton Dickinson, Mountain View, CA). Data analysis was carried out using CellQuest software after acquisition of 30,000 events.
Tumorigenicity in nude mice
Male BALB/C nu/nu mice, 5 weeks old, were supplied by the Sankyo Labo Service Corporation (Tokyo, Japan). OS-RC-2 parental cells (OP) (2 × 106) and transfectant cells, OS-RC-2-pTARGET-S-6 (OS6) and OS-RC-2-pTARGET-AS-4 (OA4), prepared from exponentially growing cultures were subcutaneously inoculated into the right flanks of the nude mice. Tumor growth was monitored twice per week by calipers measurement of 2 diameters from the 9th day after inoculation, and the tumor volume was calculated by the formula: volume = (width)2 × length/2. Data are presented as mean volume ± SE. The number of mice per group (n) was 10. All procedures for animal experimentation were approved by local animal research authorities and animal care was in accordance with institutional guidelines.
Immunohistochemistry and microvessel density
The tumors obtained from the nude mice were fixed with formalin and embedded with paraffin. The blocks were sectioned at a thickness of 4 μm, deparaffinized and rehydrated. Routine endogenous peroxidase activity and nonspecific binding were blocked with standard methods in a humidified chamber at room temperature. After being treated with DAKO target retrieval solution (Dako, Glostrup, Denmark), the slides were incubated with monoclonal mouse anti-human CD31 (Dako, 1:40) as the primary antibody for 60 min at room temperature, followed by using a Histofine simple stain PO Kit (Nichirei Co., Tokyo, Japan) according to the manufacturer's instructions. Estimation of microvessel density (MVD) was carried out according to Weidner's method27 and was expressed as number per mm.2 The stained sections were screened at 100× magnification under a light microscope to identify the 3 fields of the section with the highest vascular density. Vessels were counted in the 3 fields at 200× magnification (each field corresponding to an area of 0.95 mm2), the average numbers of microvessels were recorded, and the mean value was used for the analysis.
Statistical analysis was carried out with SAS StatView 5.0 for Windows. Differences in cell viability, percentage of colony, relative gene expression, PGE2 concentrations, mean tumor volumes among the groups and MVD were analyzed by Dunn's multiple test. Differences in survival time among the groups were determined using Kaplan-Meier survival analysis compared by the Mantel-Cox log-rank test. For all the analyses, probability values (p) of <0.05 were considered significant.
COX-2 expression in human RCC cell lines
We first examined expression of COX-2 in human normal renal cell line RPTEC and human RCC cell lines OSRC-2, ACHN and SMKT-R4 by semiquantitative RT-PCR and Western blotting. As shown in Figure 1a, COX-2 mRNA expression was found in OSRC-2 cells and BxPC-III-positive control cells but not in ACHN, SMKT-R4 cells and RPTEC normal cells. Figure 1b shows a similar tendency for expression of COX-2 protein in the above cell lines.
Effect of COX-2 inhibitors on RCC cellular proliferation
To elucidate the effects of NS398 on the viability and proliferation of RCC cell lines, MTT assay was carried out. Figure 2 shows that NS398 resulted in a dosage-dependent suppression of cell proliferation in both RCC cells and normal renal cells, expressed as viability of cells treated with NS398. Moreover, there were no significant differences in 50% inhibitory concentration (IC50) between COX-2 expressing RCC cell line (OS-RC-2) and non-COX-2 expressing RCC cell lines (ACHN, SMKT-R4). The IC50 of RPTEC cells was significantly lower than those in RCC cell lines (p < 0.0001) (Fig. 2). To diminish the influence of cytotoxicity of DMSO, we prepared 60 mM NS398 stock solution to reduce the final concentration of DMSO in the medium. Although the IC50 in RPTEC cells increase to some extent after reducing the volume of the DMSO solvent to half, it was still significantly lower than those in RCC cells (data not shown).
Akt response to NS398 treatment
In an effort to delineate the underlying mechanism for NSAID-induced growth inhibition on RCC cells, we examined Akt response to NS398 treatment in these cells by Western blotting. No appreciable change was noted in either Akt or phosphor-Akt levels within different NS398 treating periods (0, 1, 4, 24, 48 and 96 hr) in all RCC cells.
Effect of COX-2 expression on in vitro growth characteristics of RCC cells
To examine the effect of COX-2 expression on in vitro growth characteristics of RCC cells, RCC cells were transfected with pTARGET-S, pTARGET-AS or pTARGET vector and colony-forming assays were carried out. After transfection with pTARGET-AS, obvious suppression of colony formation was demonstrated in the COX-2-expressing RCC cell line (OS-RC-2) but not in the non-COX-2 expressing RCC cell line (SMKT-R4) (p < 0.01, Fig. 3a). Introduction of COX-2 expression through transfection with pTARGET-S led to an increase in colony formation in SMKT-R4 cells but not in OS-RC-2 cells (p < 0.05, Fig. 3b).
Expression of COX-2 in stable cell lines transfected with pTARGET-S, pTARGET-AS and pTARGET vector
To further examine the effect of COX-2 expression on RCC cell line OS-RC-2, transfection with pTARGET-S, pTARGET-AS and pTARGET vector was carried out and stable transfectants, OS, OA and OC, respectively, were established. The mRNA and protein expression of COX-2 in these transfectants were evaluated by quantitative real-time PCR and Western blotting. In comparison with OP cells and vector control transfectants (OC1), consistently high expression levels of COX-2 were demonstrated in OS cell lines (OS5, OS6 and OS15), and consistently low expression levels in OA cell lines (OA1, OA4 and OA8), at both mRNA (p < 0.0001) (Fig. 4a) and protein levels (Fig. 4b).
Assessment of PGE2 production in OS-RC-2 parent cells and transfectants
To assess the synthesis of PGE2 in OS-RC-2 cells and transfectants, the levels of PGE2 produced by these cells were measured by ELISA. In comparison with OP cells, significantly increased production of PGE2 was found in OS5, OS6 and OS15 cells (p < 0.0001) (Fig. 5a), whereas significantly decreased production was found in OA1, OA4 and OA8 cells (p < 0.0001) (Fig. 5b).
Resistance of OS-RC-2 parent cells and transfectants to apoptosis inducer butyrate
The effect of sodium butyrate on the OP, OA and OS cells was examined by FACS analysis. Although no significant difference of cell viability in vitro could be found among OP cells and transfectants (data not shown), the responsiveness of OP cells and transfectants to sodium butyrate differed greatly. Much more of the OA cells underwent apoptosis after treatment with the apoptosis inducer than did OP cells and OS cells. A representative result is shown in Figure 6.
Tumorigenicity assay in nude mice
To directly assess the effects of COX-2 on the tumorigenesis, a tumorigenicity assay in nude mice was conducted using the COX-2 overexpressing cell line OS-RC-2. The tumor growth of the OA4 cells was significantly slower than that of OP cells (p < 0.01) or OS6 cells (p < 0.0001) (Fig. 7a). Further survival analysis by the Kaplan-Meier method demonstrated that the survival time of the OA4 cells group was significantly longer than that of OP cells or OS6 cells group (p < 0.0001) (Fig. 7b).
To investigate the effect of COX-2 on angiogenesis in tumor tissues, the vascularity in transplanted tumors was quantitated by scoring MVD. The results showed that the MVD of OA4-transplanted tissues was significantly less than those of OP- and OS6-transplanted tissues (p < 0.0001). There was no significant difference, however, between the MVD of OP- and OS6-transplanted tissues (Fig. 8).
Our present study proved that there was an overexpression of COX-2 in OS-RC-2 RCC cell line both at mRNA and protein levels. No significant differences were found for COX-2 inhibitors NS398 with regard to viability between the COX-2 expressing RCC cell line (OS-RC-2) and non-COX-2 expressing RCC cell lines (ACHN and SMKT-R4). Significant suppression of colony formation was demonstrated in the COX-2 antisense transfectants of OS-RC-2. Forced COX-2 expression led to promotion of colony formation in SMKT-R4 cells. In contrast to parent cells and sense transfectants, COX-2 expression and PGE2 production in OS-RC-2 antisense transfectants were significantly suppressed and an increase in butyrate-induced apoptosis was also observed in these cells. Tumorigenicity assay in nude mice showed that the antisense group exhibited significantly delayed tumor growth and prolonged survival time resulting from suppression of COX-2 expression. Angiogenesis decreased more in the tumor tissues xenografted with antisense transfectants than those in parental cells and sense transfectants.
A rapidly growing body of evidence indicates that COX-2 can be induced by various factors, including cytokines, growth factors and tumor promoters, and plays a crucial role in multistage carcinogenesis.4 The pleiotropic effects of COX-2 in carcinogenesis include increasing cell proliferation, promoting angiogenesis, inhibiting apoptosis, regulating cell adhesion and increasing the invasiveness of malignant cells.28, 29, 30 Inappropriate upregulation of COX-2 prolongs the survival of malignant or transformed cells and leads to phenotypic changes associated with metastatic potential. With regard to its molecular mechanism, COX-2 has been proven to be able to interact with a variety of carcinogens. COX-2 is an important target gene of Ras16, 17 and is regulated by other genes, such as P53.16, 31 COX-2 is also thought to be able to upregulate Bcl-2.32 The induction of COX-2 seems to occur mainly through increased transcription of the COX-2 gene,31 but it may also occur via translational regulation. This possibility is attractive because COX-2 has multiple elements in the 3′-UTR33 that are known to regulate the stability of messenger RNA and, perhaps, the efficiency of translation.34 Once induced, COX-2 catalyzes the oxidation of arachidonic acid and produces prostaglandins, highly reactive by-products that have been proven to accelerate the carcinogenic process.
Overexpression of COX-2 seems to be a widespread phenomenon in human carcinogenesis because elevated levels of COX-2 mRNA and protein have been found in biopsies and cells derived from various carcinomas, such as colorectal carcinoma,6 stomach cancer,7 skin epidermal cancer,8 lung cancer,9 esophageal carcinoma,10 gall bladder cancer,11 pancreatic cancer,12 breast cancer,13 laryngeal cancer14 and bladder cancer.15 There is, however, little direct information evaluating the role of COX-2 in RCC. Khan et al.22 reported previously COX-2 expression in some canine RCC. Meanwhile, Okamoto et al.35 described enhanced expression of COX-1 but not COX-2 in renal carcinoma of the Eker rat. Recent reports demonstrated enhanced expression of COX-2 in some tissue specimens of human RCC.23, 24 Our present study showed that there was enhanced expression of COX-2 in 1 human RCC cell line, OS-RC-2, among the RCC cell lines examined. The remaining RCC cell lines (SMKT-R4, ACHN) and normal renal cells (RPTEC) had no detectable amount of COX-2 mRNA or protein. These results suggest that enhanced expression of COX-2 exists in some human RCC.
The ectopic overexpression of COX-2 is presently a focus of interest because this enzyme is a major target of anti-neoplastic activity of NSAID. Studies have demonstrated the inhibition on cellular proliferation of cancers by the selective COX-2 inhibitors NS398 and celecoxib.36 The mechanism by which COX-2 inhibitors suppress carcinogenesis is attributed to its reversal of COX-2-derived prostaglandin E2 that promotes cell proliferation, tumor growth and immunosuppression.37 Our present study, however, demonstrated that NS398 resulted in a dosage-dependent suppression of cell proliferation in all RCC cells, regardless of COX-2 expression. Moreover, it is reported that the IC50 for inhibition of COX-2 enzyme activity was 3.8 μM,38 and we found that the IC50 values of NS398 in all RCC cells examined were much higher than the pharmacological dose. This result is in accordance with Tegeder's report that most of the studies used 100–1,000-fold higher concentration of NSAID than the IC50 to inhibit PG synthesis, when the antitumor activity of NSAID had been observed.39 It has been shown that both COX-PGE2 dependent and COX-PGE2 independent pathways are involved in NSAID-induced growth inhibition on tumor cells.40 Some of the COX- PGE2 independent mechanisms for NSAID include blockage of the activation of Akt, inhibition of NF-κB activation, downregulation of the antiapoptotic protein Bcl-XL, inhibition of PPAR δ, activation of PPAR γ and activation of caspase family. One or more of these COX- PGE2 independent effects could contribute to the proapoptotic and antiproliferative properties of NSAID.41, 42 The degree of taking advantage of COX- PGE2 dependent or COX- PGE2 independent pathway of NSAID might depend on the cell-type specificity. Our results indicate that NS398 might exert growth inhibition on RCC cells through COX- PGE2 independent pathway at high concentration. Blocking Akt activation through inhibiting Akt phosphorylation at Ser-473 is known to be involved in NSAID-induced COX-PGE2 independent growth inhibition in some cancer cells.43, 44 In an effort to delineate the underlying mechanism for growth inhibition of NS398 on RCC cells, the Akt expression and the phosphorylation status of Akt in RCC cells treated with NS398 was further examined in our study. The levels of both Akt and phosphor-Akt protein, however, were unchanged in our RCC cells treated with NS398 in time course of 0, 1, 4, 24, 48 and 96 hr. This result suggests that NS398-induced antiproliferation in the RCC cells examined is dissociated from Akt pathway, and that the mechanisms for NSAID-mediated COX-PGE2 independent growth inhibition of tumor cells depend on the cell context. Although further study is warranted to elucidate the precise mechanism for the antiproliferative property of NSAID against human RCC cells, NS398 could not effectively inhibit in vitro cellular proliferation of human RCC cells at pharmacological dose.
In addition, we investigated the cytotoxic effect of NS398 in the normal renal proximal tubule epithelial cell line, RPTEC. The IC50 value of NS398 in RPTEC cells was significantly lower than those in the RCC cell lines examined. This phenomenon may be explained by the weakness of RPTEC itself. Because the normal renal cell line is maintained under a strict culture condition, it is possible that DMSO, the solvent of NS398, affects the viability of RPTEC cells. We observed that reducing the volume of the DMSO solvent led to slight increase in the IC50 of RPTEC cells. This result implies that the character of RPTEC cells itself, distinct from RCC cells, DMSO cytotoxicity and other unknown factors contribute to significantly low IC50 in RPTEC cells. Although the IC50 value in RPTEC cells was lower than those in RCC cell lines, this value remains much higher than pharmacological dose of NS398.
Antisense cDNAs have been used as a potent tool in inhibition of gene expression.45 The molecular mechanism for inhibition on gene expression by antisense has not been clarified yet. Its postulated mechanisms include hindrance of gene transcription or gene translation, blockage of mRNA processing or splicing, and degradation of mRNA through the action of RNase H activity.46 We constructed COX-2 recombinant sense- and antisense-orientated expression vectors. Colony formation assay in present study showed that that COX-2 antisense cDNA exhibited significant suppression on the cellular proliferation in OS-RC-2 cells with overexpression of COX-2, but not in SMKTR-4 cells without an appreciable amount of COX-2 expression. In addition, forced COX-2 expression led to significant promotion of colony formation in SMKT-R4 cells but not in OS-RC-2 cells. These results, together with the growth inhibition of NS398 on RCC cells, indicate that COX-2 might be directly associated with cellular proliferation of RCC cells. Moreover, to further investigate the role of COX-2 in RCC carcinogenesis, OS-RC-2 cells were transfected with recombinant COX-2 sense- and antisense-orientated expressing vectors, and stable transfectants were established. The increase of COX-2 mRNA and protein expression in sense transfectants and decrease in antisense transfectants suggest that COX-2 antisense cDNA can effectively inhibit COX-2 expression at both mRNA level and protein level. In addition, paralleled with COX-2 expression level, significant decrease and increase in COX-2-derived PGE2 production were demonstrated in antisense and sense transfectants, respectively. Taken together, these results indicate that the full-length antisense DNA could suppress not only the expression of COX-2 but also enzymatic activity of COX-2.
Studies have shown the close association between COX-2 overexpression and the antiapoptosis feature in cancer cells. It has been reported that COX-2 overexpression leads to the inhibition of apoptosis or altered cell cycle kinetics in different kinds of carcinoma cells.47, 48 Modulation of antiapoptotic Bcl-2 protein has been accepted as one certain mechanism for acquisition of antiapoptotic ability by COX-2 overexpression cancer cells.40 Sun et al.32 reported that forced COX-2 overexpression mediates up-regulation of antiapoptotic Bcl-2 protein, and consequently attenuates apoptosis induction by NSAID and 5-FU through predominant inhibition of the cytochrome c-dependent apoptotic pathway in human colon cancer cells. Butyrate has been reported to induce growth inhibition, differentiation and apoptosis in tumor cells in vitro.28, 48 In our study, an enhancement of butyrate-induced apoptosis was observed in antisense transfectants but not in sense transfectants and parent cells. Consistent with other studies,28 our results favor the view that COX-2 overexpression contributes to acquisition of antiapoptotic ability by cancer cells. Moreover, it is suggested that COX-2 antisense cDNA was able to sensitize COX-2 expressing RCC cells to butyrate-induced apoptosis, which implies a promising perspective in treatment of COX-2 expressing RCC.
In light of the above results obtained from in vitro experiment, we hypothesized that COX-2 antisense cDNA should exert suppression on in vivo tumorigenesis of OS-RC-2. As expected, tumorigenicity assay in nude mice showed significantly delayed tumor growth and prolonged survival in the COX-2 antisense group, which is in contrast to the control group and the sense group. These results strongly indicated that suppression of COX-2 expression by antisense cDNA could effectively attenuate oncogenesis of OS-RC-2 cells in vivo. Various studies have demonstrated that COX-2 is associated with tumor angiogenesis.49, 50 To investigate the involvement of COX-2 in angiogenesis in human RCC, we investigated the MVD in transplanted tumor tissues and evaluated the effect of COX-2 on MVD. The results showed that suppression of COX-2 expression by antisense cDNA could inhibit angiogenesis in transplanted tumors. Therefore, suppression of in vivo tumor growth in COX-2 antisense transfectants might be brought about not only by suppression of cellular proliferation but also by inhibition of angiogenesis.
Although transfection of antisense COX-2 cDNA into OS-RC-2, the COX-2 overexpressing cell line, resulted in suppression of cellular proliferation, anti-apoptotic activity of COX-2 and in vivo tumorigenesis, no significant changes in these cellular events were observed between parental OS-RC-2 and its COX-2 sense transfectants though there were significant differences of COX-2 expression and PGE2 production between them. It was possible that the change was not enough to obtain a statistical difference because the parental OS-RC-2 cell itself had an endogenous COX-2 overexpression, increased activity of cellular proliferation and in vivo tumorigenesis. Thus the complete additive effect of the above cellular events cannot be expected in this cell system. Another possible explanation is that the COX-2 function is not always correlated to the amount of its expression and enzyme activity. Further studies may offer further support for this notion.
In conclusion, our present study indicates that COX-2 is overexpressed in some human RCC cell lines and plays a crucial role in the carcinogenesis of human RCC through promoting PGE2 production, cell proliferation and apoptosis resistance as well as subsequent enhancing tumorigenesis and angiogenesis in vivo. The results suggest that COX-2 may become a new target gene for RCC treatment, and inhibition of COX-2 expression through antisense strategy may have potential utility in treatment of COX-2-expressing RCC.