HA1077, a Rho kinase inhibitor, suppresses glioma-induced angiogenesis by targeting the Rho-ROCK and the mitogen-activated protein kinase kinase/extracellular signal-regulated kinase (MEK/ERK) signal pathways
Malignant gliomas are characterized by high invasive potential and strong angiogenic ability. Constitutive activation of Rho and its downstream target Rho-kinase are crucial for tumor progression. We investigated the effect of the Rho-kinase inhibitor HA1077 on tumor-induced angiogenesis of malignant glioma cells and explored the molecular mechanisms underlying the effect of HA1077. Here, we demonstrated that HA1077 suppressed tube formation of endothelial cells in human umbilical vein endothelial cell (HUVEC)–glioma cell co-culture assay. Western blot, RT-PCR, ELISA and zymography demonstrated that HA1077 suppressed gene and protein expressions of vascular endothelial growth factor (VEGF), matrix metalloproteinase (MMP)-2 and MMP-9 in glioma cells. Furthermore, HA1077 attenuated the phosphorylation of extracellular signal-regulated kinases 1 and 2 (ERK1/2) and the DNA binding activity of activator protein-1, a key downstream transcriptional factor of ERK1/2. HA1077 also suppressed the migration of HUVEC in vitro. Thus, it is suggested that the anti-angiogenic effect of HA1077 may be due to the combination of ROCK inhibition and mitogen-activated protein kinase kinase (MEK)/ERK pathway inhibition. Moreover, an in vivo intracerebral human glioma cell xenograft mouse model demonstrated that HA1077 suppressed neovascularity and tumor growth. The results of the present study suggest that HA1077 has a therapeutic potential as an anti-angiogenic agent against malignant gliomas. (Cancer Sci 2011; 102: 393–399)
Rho GTPases play a pivotal role in the regulation of numerous cellular functions. Rho GTPases are localized at membranes and are activated by the stimulations of cell surface receptors.(1) In their GTP-bound state, Rho proteins bind to effector proteins, thereby triggering specific cellular responses. Members of the Rho family of small GTPases are key regulators of actin reorganization, cell motility, cell-extracellular matrix (ECM) adhesion, cell cycle progression, gene expression and apoptosis.(2–5)
Rho kinase (ROCK) is a downstream effector of small Rho-GTPases. Thus, ROCK also plays an important role in various pathological conditions and should be an important pharmacological target. Potential therapeutic indications for ROCK inhibitor have emerged and several ROCK inhibitors have been developed.(6–9) Among the various ROCK inhibitors, HA1077 is the only ROCK inhibitor presently used in a clinical setting without obvious side-effects.(10)
The studies on ROCK inhibitors that began in the cardiovascular field(11,12) have extended to the field of oncology. The effect of ROCK inhibitors on tumors has mainly been investigated from the aspect of invasion,(13,14) metastasis(15) and proliferation.(16) With regard to angiogenesis in tumor tissues, there are several reports that examined the effect of ROCK inhibitors on human umbilical vein endothelial cells (HUVEC).(17,18) However, there is only one reported study that investigated the inhibitory effect of ROCK inhibitor on tumor-induced angiogenesis using human tumor cells.(19) This report described that Wf-536, a derivative of Y-27632, inhibited a tumor-induced angiogenic response in a lung cancer cell transplanted mouse model. In contrast, there is no report that has investigated the effect of HA1077 on tumor-induced angiogenesis, although there are several reports that examined the effect of HA1077 on invasion or proliferation.(20,21) With regard to human glioma cells, although a recent study reported that HA1077 suppressed invasion and proliferation of glioma cells,(22) the effect of HA1077 on angiogenesis was not investigated. Similarly, suppressive effects of Y27632 on glioma cell proliferation and motility was shown by another group.(23) However, angiogenesis was also not investigated in this report. Therefore, in the present study, we aimed at investigating the effect of HA1077 on tumor-induced angiogenesis using human malignant glioma cells. We also tried to examine the influence of HA1077 on the molecules and the pathway related to tumor-induced angiogenesis. In gliomas, the best characterized pro-angiogenic factor is vascular endothelial growth factor (VEGF).(24) Matrix metalloproteinase (MMP)-2 and MMP-9 are highly expressed in gliomas,(25,26) and they are necessary for the degradation of ECM in the progress of angiogenesis. Among the signaling pathways most frequently dysregulated in human malignancies is the Raf-Ras-mitogen-activated protein kinase (MEK)–extracellular signal-regulated kinase 1 and 2 (ERK1/2) pathway. ERK1/2 kinase is activated by growth factors including VEGF through receptor tyrosine kinases. Meanwhile, it is reported that the transcriptional activation of VEGF is promoted by acidic extracellular pH in a human U87 glioma xenograft mediated by activator protein-1 (AP-1), a key downstream transcription factor of ERK1/2.(27) Furthermore, the implication of the ERK pathway in the post-transcriptional regulation of VEGF mRNA stability was recently reported.(28) We then aimed to study the effect of HA1077 on VEGF, MMP-2, MMP-9 and the ERK pathway. The present study is the first to investigate the effect of HA1077 on glioma-induced angiogenesis.
Materials and Methods
Cell culture and reagent. Human glioma cell lines, T98G and U87MG, were obtained from the American Type Culture Collection (Rockville, MD, USA). Glioma cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Nikken Biomedical Laboratory, Kyoto, Japan) supplemented with 10% fetal bovine serum (FBS; Gibco BRL, Gaithersburg, MD, USA), penicillin (100 unit/mL) and streptomycin (100 mg/mL) at 37°C in a 5% CO2. HUVECs were purchased from Kurabo Industries Ltd (Osaka, Japan) and grown in EGM-2 medium (Kurabo) containing 2% FBS at 37°C in a 5% CO2. HA1077 was purchased from Biaffin GmbH & Co. KG (Kassel, Germany).
RNA preparation and RT-PCR. The gene expression of VEGF, MMP-2 and MMP-9 was analyzed by semiquantitative RT-PCR technique. The total cellular RNA from the HA1077-treated (20 μΜ/24 h) glioma cells and the control (HA1077 non-treated glioma cells) were extracted using the RNeasy kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. RT-PCR were performed using the Takara RNA PCR kit (AMV) ver. 3.0 (Takara, Shiga, Japan). The PCR amplification was performed with the following oligonucleotide primers: VEGF forward primer, 5′-CCA TGA ACT TTC TGC TGT CTT-3′ and VEGF reverse primer, 5′-TCG ATC GTT CTG TAT CAG TCT-3′; MMP-2 forward primer, 5′-GAG ATC TGC AAA CAG GAC-3′ and MMP-2 reverse primer, 5′-TTG GTT CTC CAG CTT CAG-3′; MMP-9 forward primer, 5′-GAA TTC AGA ACC AAT CTC GAC AGG CA-3′ and MMP-9 reverse primer, 5′-GAA TCC AGA ACC AAT CTC ACC GAC AGG CA-3′; respectively. As an internal control, a fragment of human GAPDH was amplified from parallel samples by PCR using the following primers: forward, 5′-TGT TGC CAT CAA TGA CCC-3′ and reverse, 5′-GCA GAG ATG ATG ACC CTT-3′.
Gelatin zymography. The levels of gelatinolytic activity in the culture supernatants of the glioma cells treated with HA1077 (50 or 100 μM) were measured using the gelatin zymography kit (Yagai Corporation, Tokyo, Japan). When glioma cells had grown to approximately 80% confluency, the medium containing 10% FBS was removed and the cells were washed with DMEM. The cells were then cultured for 24 h in DMEM containing 0.1% bovine serum albumin (BSA) and HA1077. After 24 h, the culture medium was collected and centrifuged. The supernatant (20 μL) was electrophoresed on the gel supplied with the gelatin zymography kit. The gel was washed with two types of washing buffer for 30 min each and then incubated for 30 h at 37°C in the reaction buffer. The gels were stained with Coomassie blue and then destained.
Enzyme-linked immunosorbent assay (ELISA). The protein levels of VEGF in the culture supernatants from the HA1077-treated (20 μΜ/24 h) glioma cell lines and the control (non-treated glioma cell lines) were measured using ELISA kits (GE Healthcare, Piscataway, NJ, USA). The data of the HA1077-treated group were expressed as a percentage of the control group (non-treatment was taken as 100%).
Western blot analysis. The phosphorylation state of ERK1/2 was evaluated by western blot analysis. After the glioma cells were starved for 24 h they were stimulated with PMA (Promega, Madison, WI, USA) for 45 min, followed by treatment with HA1077 (20 or 50 μM) for 2 h. Then the glioma cells were harvested in a buffer containing 50 mM Tris–HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA and 1% (v/v) Triton X-100 plus protease and phosphate inhibitors (Sigma, St Louis, MO, USA). The protein content was measured by the Bradford procedure, with BSA as a standard. The cell lysate, 50 μg protein, was separated on a 10% SDS-PAGE. The membranes were incubated with anti-phospho-ERK1/2 antibody (phospho-p44/42 MAPK antibody-Thr 202/Tyr204) and total ERK1/2 antibody following the manufacturer’s instructions (Cell Signaling Technology Inc., Danvers, MA, USA). β-Actin was used as an internal control. Protein expression was determined by quantitative densitometry.
AP-1 DNA binding assay. DNA binding of AP-1 was assessed using nuclear extracts and biotin-labeled AP-1 oligonucleotides (Panomics, Fremont, CA, USA). Electrophoretic mobility-shift assay (EMSA) was performed using the EMSA Gel-Shift kit (Panomics). Glioma cells were treated with acidic pH medium (pH 6.6) containing HA1077 (20 or 50 μΜ) for 6 h in the absence or presence of PMA (50 ng/mL). A 10 μL reaction mixture containing 10 μg of nuclear protein was incubated for 30 min at room temperature with a binding mixture and 2 pmol oligonucleotide probe, with or without a non-labeled 10-fold excess of competitor oligonucleotides. Protein DNA complexes were separated by using a 6% non-denaturing acryl amide gel electrophoresis. Complexes were transferred to positively charged nylon membranes and UV-cross-linked in a Stratagene cross-linker. Gel shifts were visualized with a streptavidin–horseradish peroxidase followed by chemiluminescent detection.
In vitro angiogenesis assay. The effects of HA1077 on glioma-induced angiogenesis were investigated as described previously.(29) This research method is based on using a HUVEC-fibroblast co-culture model (Angiogenesis kit; Kurabo). HUVEC and fibroblasts were incubated together according to the kit manufacturer’s instructions. We also placed an insert plate with a 1-μm pore size PET membrane (Falcon HTS Multiwell Insert Systems; BD Biosciences, San Jose, CA, USA) on the 24-well plate of the angiogenesis kit. Glioma cells were cultured in the insert plate so that soluble angiogenic factors secreted by the glioma cells could affect endothelial cells in the lower chamber. HA1077 was added to the culture solution at one of various concentration levels, 0.1–100 μM, and incubated for 10 days. Glioma cells and HA1077 were not added in the control I condition and HA1077 was not added in the control II condition. On day 11 of incubation, the cells were fixed in 70% ethanol and the anti-CD31 monoclonal antibody (Kurabo) was used as the primary antibody to immunohistochemically stain the vascular lumens.
Migration assay of HUVEC. HUVEC migration was measured using the BD BioCoat Angiogenesis System (BD Biosciences). Briefly, cell suspensions (250 μL, 5 × 105 cells/well) were added to the Transwell insert (3-μm pore size; Costar, Cambridge, MA, USA); 750 μL of starvation medium containing one of various concentrations of HA1077 (0.1–100 μM) was then added to the lower chamber. The chambers were first incubated for 4–24 h at 37°C in a 5% CO2 atmosphere, and then in Hank’s Balanced Salt Solution (HBSS) with 50 nM of Calcein-AM, a fluorescence dye, for an additional 90 min to label the living cells. The cells were fixed in 4% paraformaldehyde for 5 min and washed in PBS. The non-migrating cells in the upper chamber were scraped off using blunt-ended forceps and swabs, and were then washed with PBS. Fluorescence from cells that migrated to the lower chamber was measured using a fluorescence microplate reader (Fluoroskan Acent FL; Thermo Fisher Scientific Inc., Waltham, MA, USA) from the bottom at 485/535 nm wavelength. The quantity of cells migrated was indexed as the ratio of fluorescence between HA1077 and control (non-HA1077) conditions.
Methyl thiazolyl tetrazolium (MTT) assay. For the determination of the in vitro growth inhibitory effect of HA1077, an MTT assay was performed. Glioma cells (1 × 104 cells) were plated in 96-well plates in 100 μL of DMEM containing 10% FBS. HUVEC (1 × 104 cells) were plated in 96-well plates in 100 μL of EGM-2 medium containing 2% FBS. After 24 h, HA1077 was added to each well at one of various concentrations ranging from 0.1 to 100 μM. HA1077 was not added in the control condition. After 24 or 48 h of incubation either with or without HA1077, 50 μL of MTT (2 mg/mL in PBS) was added to each well at 37°C for 3 h. The measurement of cell viability was performed using a microplate reader (Molecular Devices THERMOmax; Scientific Support Inc., Hayward, CA, USA) at a wavelength of 570 nm.
Effect of HA1077 on orthotopic implantation. Athymic female mice (BALB/c nu/nu), 6–8 weeks old, were obtained from Charles River Japan (Atsugi, Japan). The mice were anesthetized with pentobarbital sodium (60 mg/kg intraperitoneally) and injected intra-cerebrally with T98G cells (1 × 105) through a small hole drilled 2 mm anterior and 2 mm lateral to the bregma. Immediately after cell implantation, HA1077 treatment began. The mice in the HA1077 group (n = 10) were intraperitoneally administered with HA1077 (10 mg/kg per day) every day, while those in the control group (n = 10) were intraperitoneally administered with saline. All mice were killed on day 22 and their brains were snap-frozen. Tumor growth of the intracerebral tumors was confirmed by histological evaluation. Serial coronal sections (30 μm) were cut from the rostal to the caudal edge of the brain tissues containing the tumors by using a cryo-microtome. Tumor size was computed using an Imagepro system (Media Cybernetics Inc., Silver Spring, MD, USA). For immunohistological examination, the same T98G xenograft model was used. Mice of the HA1077 group and mice of the control group were killed on day 22 after glioma cell transplantation, and their brains harvested and fixed in buffered formalin before embedding in paraffin. Paraffin sections, 4-μm thick, were deparaffinized and dehydrated. The sections were stained for von Willebrand factor (vWF) by the immunoperoxidase method in order to evaluate angiogenesis. Sections were first treated with rabbit polyclonal antibodies against vWF (Abcam Inc., Cambridge, MA, USA) as the primary antibody and then with EnVision+ System, HRP (Dako, Glostrup, Denmark). Positive staining was detected using diaminobenzidine.
Statistical analysis. The data of ELISA assay and tumor volume in an orthotopic implantation model were analyzed using the Student’s t-test. The angiogenesis, migration and invasion assay results obtained from the control and treated groups were statistically analyzed by one-way analysis of variance (anova). In all statistical analyses, P < 0.05 was regarded as statistically significant. All values are presented as means ± standard error (SE).
HA1077 suppressed the expression of VEGF, MMP-2 and MMP-9 in glioma cells. Semiquantitative RT-PCR demonstrated that HA1077 reduced mRNA expression of MMP-2, MMP-9 and VEGF compared with the control group (non-treated group; Fig. 1a). Gelatin zymography demonstrated that the culture supernatant of glioma cells possessed MMP-2 and MMP-9 gelatinolytic activity. Both the latent and active forms of MMP-2 and the latent form of MMP-9 were detected in the culture supernatant of glioma cells. We found that HA10777 suppressed the gelatinolytic activity of the culture supernatant of glioma cells in a dose dependent fashion (Fig. 1b). ELISA analysis showed that HA1077 decreased the protein level of VEGF in the culture mediums of glioma cell lines (Fig. 1c). Statistically significant differences were shown between the HA1077-treated samples and the control (non-treated; P < 0.0001).
HA1077 suppressed the ERK pathway. Western blot analysis demonstrated that ERK1/2 was not strongly phosphorylated in the absence of PMA, but was strongly phosphorylated by PMA stimulation. Treatment with 50 μM HA1077 significantly suppressed the PMA-induced phosphorylation of ERK1/2 (Fig. 2). EMSA demonstrated that the DNA binding activity of AP-1 was raised by PMA stimulation, but was suppressed by HA1077 in a dose dependent fashion (Fig. 3).
HA1077 suppressed tumor-induced angiogenesis in vitro. A blood vessel construction image of the entire culture plate of an angiogenesis assay was captured by an image scanner. This image was overlaid with a grid and neovascularity was evaluated as the total number of intersections between vessel and grid across the entire plate (Fig. 4a). Figure 4b shows the representative data of T98G. The density of neovasculature decreased as the HA1077 concentration increased compared with the control II. Two cell lines showed the same response to HA1077 (Fig. 4c). When the concentration of HA177 exceeded 10 μM, remarkable suppression of neovascularization was observed. Statistical analysis revealed that glioma cell-induced neovasularity was significantly suppressed by HA1077 (P <0.0001). The data of the control I condition were also significantly lower than that of the control II condition (P <0.0001).
HA1077 suppressed the migration of HUVEC. HA1077 inhibited HUVEC migration in a dose-dependent manner (Fig. 5). Statistical analysis revealed that HUVEC migration was significantly suppressed by HA1077 (P <0.0001).
HA1077 suppressed the growth of tumor cells and HUVEC at 100 μM. HA1077 at 0.1–10 μM showed no significant inhibitory effect on the growth of glioma cells and HUVEC up to 48 h incubation, although HA1077 significantly inhibited the growth of glioma cells and HUVEC at 100 μM (P <0.0001; Fig. 6).
Evaluation of implanted tumors. In the HA1077 treatment group (n = 10), tumor growth was suppressed compared with the control group (n = 10). Average tumor volume was calculated from sequential histological sections (Fig. 7a), and estimated to be 44.9 ± 1.59 mm3 for the HA1077 treatment group and 63.0 ± 1.33 mm3 for the control group (P < 0.0001; Fig. 7b). Histological evaluation showed the number of microvessels was lower in the HA1077 treatment group than in the control group (Fig. 7c).
Gliomas are the most common tumors of the central nervous system and approximately 50% of theses tumors are histologically malignant. Glioblastoma, the most frequent and malignant glioma, is very difficult to treat because of its aggressive and wide invasion into the surrounding normal brain tissue. Despite improved surgical and radiotherapeutic techniques and the inclusion of temozolomide in the multimodal treatment strategy, the prognosis of patients with glioblastoma remains poor. It is reported that the median overall survival rate is approximately 15 months, and 88% of patients die within 3 years.(30,31) One of the key events in the progression of malignant gliomas is angiogenesis.(32) It has been reported that among solid tumors, glioblastoma is the most angiogenic tumor.(33) Angiogenesis is essential for tumor growth and is related not only to tumor nutrition but also to tumor cell migration along the basement membrane of growing blood vessels.(34) Thus, besides cytotoxic drugs, anti-angiogenic agents have also been of particular interest for the development of a novel treatment strategy for malignant gliomas.
Ras protein is a family of membrane-associated small GTPases that transmit signals from cell surface receptors such as epidermal growth factor receptor (EGFR) and has been shown to play a significant role in VEGF expression.(35) Although malignant gliomas do not display ras mutation, they show enhanced expression of Ras. Furthermore, malignant gliomas express a high level of EGFR. Activation of EGFR leads to tyrosine kinase activation and functional upregulation of the MEK/ERK pathway. After the phosphorylation of ERK1/2, AP-1 promotes VEGF expression.(36) The present study showed that HA1077 attenuated the phosphorylation of ERK1/2 and the DNA binding activity of AP-1. Therefore, it is suggested that the anti-angiogenic effect of HA1077 on glioma cells may be partly due to the suppression of VEGF expression through the inhibition of the MEK/ERK pathway. Meanwhile, another study also reported that HA1077 suppressed migration and invasion of lung cancer cells through the downregulation of VEGF.(37) However, this report has not demonstrated the mechanism underlying the downregulation of VEGF and did not study angiogenesis. Furthermore, the inhibitory effect of high dose (100 μM) HA1077 on ERK 1/2 phosphorylation has already been reported.(38) Moreover, HA1077 has been reported to suppress other kinases such as mitogen- and stress-activated protein kinase 1 at a higher concentration than 5–15 μM, although IC50 of HA1077 for ROCK-II is 1.9 μM.(39) Likewise, Y-2763, another ROCK inhibitor, has also been shown to suppress multiple kinases at higher concentration(39) and the implication of the ERK1/2 pathway in the suppressive effect of high-dose Y-27632 on migration and proliferation of glioblastoma cells is also reported,(23) although this report also did not study angiogenesis. Accordingly, it seems reasonable to explain the remarkable suppression of neovascularization treated with 100 μM HA1077 shown in the angiogenesis assay not only by ROCK inhibition but also by ERK pathway inhibition.
With regard to MMP, it is reported that inhibition of ERK activation by the MEK inhibitor blocked MMP-9 production and attenuated the in vivo invasiveness of head and neck carcinoma cells.(40) Another study has shown that ERK inhibitor causes a decrease in the expression of MMP-2 in breast cancer brain metastases.(41) Therefore, suppression of the MEK/ERK pathway might be the mechanism underlying the downregulation of MMP-2 and MMP-9 expression in the HA1077-treated glioma cells in the present study. Additionally, HA1077 suppresses the migration of endothelial cells, the first step of angiogenesis, through the inhibition of cytoskeleton reorganization. Thus, it is suggested that the anti-angiogenic effect of HA1077 might be due to the combination of ROCK inhibition and MEK/ERK inhibition. Taken together, it is speculated that the anti-angiogenic effect of HA1077 might be mediated chiefly through ROCK inhibition at a lower concentration and is enhanced by MEK/ERK inhibition at a higher concentration. Further investigations are necessary for the detailed clarification of this speculation.
In conclusion, the results of the present study provide the first experimental evidence that HA1077 suppresses glioma-induced angiogenesis. HA1077 is suggested to have the potential to suppress glioma progression.