Background
The standard treatment for muscle-invasive bladder cancer (MIBC) is radical cystectomy and bilateral pelvic lymph node dissection (PLND), while that for upper urinary tract cancer is radical nephroureterectomy and retroperitoneal lymph node dissection (RPLND). These radical procedures have become standard treatment over the past 30 years, but patients still have a relatively poor prognosis and the 5-year survival rate after surgery is less than 50% [
1‐
3]. Although systemic chemotherapy with methotrexate, vinblastine, doxorubicin, and cisplatin (M-VAC) can reduce the tumor burden in patients with urothelial cancer, its influence on the prognosis is not very impressive [
4]. Gemcitabine plus cisplatin (GC) has a better safety profile than M-VAC and may be considered as the first-line treatment for metastatic bladder cancer [
5]. Some patients develop systemic metastases within a few years of curative resection. The most frequent sites of metastasis are the regional lymph nodes, liver, lungs, and bone [
6], and the outlook for these patients is poor. Presumably, recurrence is due to occult micrometastasis at the time of surgery occurring via the rich lymphatic drainage of the bladder and upper urinary tract. Metastasis, i.e., tumor cell spread from the primary lesion to a distant site [
7], is the major cause of cancer death. Various studies have shown that poorly differentiated cancer, muscle invasion, lymph node metastasis, and lymphovascular invasion are associated with recurrence of bladder cancer and are unfavorable prognostic factors. Therefore, it seems important to investigate the process of tumor cell dissemination.
Tumor cell migration is essential for metastasis, and migration involves rearrangement of the actin cytoskeleton. Accordingly, investigation of the regulation of actin cytoskeletal proteins could be important for understanding tumor metastasis. Members of the Rho family of small GTPases are involved in regulating a variety of cellular processes, including organization of the microfilament network, intercellular contact, and malignant transformation [
8]. These cellular events are all interrelated. Specifically, certain subfamilies of Rho proteins are involved in regulating the actin cytoskeleton during the formation of stress fibers and focal adhesions within cells. The Rac subfamily regulates the formation of lamellipodia and membrane ruffles, while the Cdc42 subfamily regulates filopodia. Both lamellipodia and filopodia are seen at the advancing edge of motile cells, while retraction occurs on the opposite side [
9,
10], and these processes are accompanied by reorganization of the actin cytoskeleton. Rho-associated serine-threonine protein kinase (ROCK) [
11,
12] is one of the best characterized downstream effectors of Rho. ROCK is activated when it selectively binds to the active GTP-bound form of Rho, after which activated ROCK interacts with the actin cytoskeleton to promote stress fiber formation and the assembly of focal contacts [
13,
14].
GTPases from the Rho family have been linked to progression of human cancer, and the Rho/ROCK pathway is considered to be involved in tumor progression by regulating the actin cytoskeleton [
15‐
17]. In fact, (R)-(+)-trans-N-(4-pyridyl)-4-(1-aminoethyl)-cyclohexanecarboxamide dihydrochloride (Y-27632) [
18] is a specific ROCK inhibitor that suppresses tumor growth and metastasis, indicating that the Rho/ROCK pathway may be a good target for preventing tumor invasion and metastasis [
19,
20]. Thus, this pathway is an attractive molecular target for anticancer therapy. We previously reported that overexpression of Rho and ROCK proteins by bladder cancer and upper urinary tract cancer was associated with poorly differentiated histology, muscle invasion, lymph node metastasis, and shorter survival, indicating that the Rho/ROCK pathway is involved in the progression of urothelial cancer [
21‐
23]. Accordingly, suppression of the Rho/ROCK pathway might potentially improve the outcome of patients with urothelial cancer.
Fasudil (HA-1077) was developed as a pharmacological ROCK inhibitor [
24,
25]. HA-1077 and its major active metabolite after oral administration (hydroxyfasudil) potently inhibit ROCK by promoting myosin light chain phosphorylation in vascular smooth muscle cells [
26,
27]. It has been reported that HA-1077 is effective for the treatment of cardiovascular disease, including coronary and cerebral vasospasm, arteriosclerosis/stenosis, ischemia/reperfusion injury, systemic hypertension, pulmonary hypertension, stroke, and heart failure [
25]. Among the various ROCK inhibitors, HA1077 is the only clinically available one without obvious adverse effects [
28]. HA-1077 has been recognized as a promising agent for preventing recurrent vasospasm of cerebral arteries after aneurysmal subarachnoid hemorrhage and its use is covered by the Japanese national health insurance system.
Because it inhibits the Rho/ROCK pathway, we investigated whether HA-1077 could block the proliferation and migration of bladder cancer cell lines or induce apoptosis of these cells. Our objective was to assess the value of the Rho/ROCK pathway as a molecular target for anticancer therapy. In this report, we discuss the clinical potential of HA-1077 for use in targeted cancer therapy.
Discussion
Tumor cell migration is essential for metastasis, and metastasis is the most common fatal complication of cancer in humans. For the dissemination of tumor cells to distant organs to occur, migration of tumor cells through the fluid spaces of the body is essential [
7]. Y-27632 [
18] is another ROCK inhibitor that effectively suppresses tumor cell motility [
19,
20]. We previously reported that increased activity or overexpression of Rho and ROCK were associated with local invasion, metastasis, and an unfavorable prognosis of urogenital cancer including urothelial cancer [
21‐
23], indicating that the Rho/ROCK pathway may be a potential target for anticancer therapy. However, there has been no reliable data regarding the effect on bladder cancer when ROCK is targeted by an inhibitor.
In the present study, LPA and GGOH induced an increase of cell proliferation that was correlated with increases of both RhoA activity and ROCK expression. The dose-dependent suppressive effect of HA-1077 on cell proliferation was accompanied by a marked decrease of ROCK-I and ROCK-II protein expression, while there was only a slight decrease of RhoA activity. This inhibitory effect of HA-1077 on cell proliferation was reduced by the addition of LPA and GGOH to cultures. On the other hand, RhoA activity was significantly higher in cultures with HA-1077 plus LPA and GGOH at each HA-1077 concentration, but the difference in the level of ROCK-I and ROCK-II protein expression between cultures with HA-1077 alone and cultures with HA-1077 plus LPA and GGOH gradually decreased at higher HA-1077 concentrations. These findings suggest that HA-1077 may selectively inhibit urothelial tumor cell proliferation via suppression of ROCK, but not by acting on RhoA.
The antiproliferative effect of HA-1077 was also evaluated using the clonogenic assay. The clonogenic cell survival assay is an effective method for the determination of single cell proliferation capacity, thereby retaining its reproductive ability to form a large colony or a clone [
30,
31]. We found that bladder cancer cells were less able to form colonies in response to exposure to HA-1077 compared to those without any treatment. These results suggested that HA-1077 inhibits the proliferation of bladder cancer cells at a certain rate.
With regard to apoptosis, HA-1077 caused the marked induction of apoptosis in a dose-dependent manner, but the difference in the percentage of apoptotic cells between cultures with HA-1077 alone and cultures with HA-1077 plus LPA and GGOH gradually became smaller at higher HA-1077 concentrations. This suggested that the pro-apoptotic effect of HA-1077 was more effective suppression of ROCK at higher concentrations of HA-1077.
In the cell migration study, addition of LPA and GGOH increased the migration of human bladder cancer cells. Cell migration was suppressed by HA-1077 in a dose-dependent manner, while this suppressive effect of HA-1077 was inhibited by addition of LPA and GGOH. Western blotting analysis of the cells from the underside of each filter showed that RhoA activity and ROCK-I and ROCK-II expression were significantly decreased by HA-1077 in a dose-dependent manner. This dose-dependent inhibition of these proteins by HA-1077 is likely to occur in parallel with a reduction in the number of migrating cells, since RhoA activity was higher in cultures with HA-1077 plus LPA and GGOH at each HA-1077 concentration, while expression of ROCK-I and ROCK-II did not increase.
LPA increases GTP loading, while GGOH activates geranylgeranylation. The mevalonate pathway is required for geranylgeranylation of Rho by GGOH. After Rho has been activated by geranylgeranylation, its downstream effector ROCK is activated when it selectively binds to the active GTP-bound form of Rho. In the present study, addition of LPA and GGOH to cultured cells increased RhoA activity and up-regulated the expression of ROCK-I and ROCK-II, while HA-1077 dramatically suppressed both ROCK-I and ROCK-II dramatically, but did not reduce RhoA activity. These findings indicate that HA-1077 may selectively inhibit bladder cancer cell proliferation and migration via suppression of ROCK, but not by blocking RhoA activity.
It is important to study signaling cross-talk between ROCK and other downstream effectors in the Rho family of GTPases. ROCK belongs to the AGC (protein kinase A/protein kinase G/protein kinase C) family of serine-threonine kinases. ROCK-I and ROCK-II share 65% overall identity, with 87% identity of the kinase domain [
11,
36,
37]. We previously reported that overexpression of both ROCK-I and ROCK-II was associated with poor differentiation, invasiveness, metastasis, and an unfavorable prognosis of human bladder cancer [
21]. In the present study, we found that the expression of both ROCK-I and ROCK-II protein was decreased by HA-1077. However, since HA-1077 and Y-27632 are not highly selective for ROCK-I and ROCK-II, when ROCK-I and ROCK-II were suppressed by HA-1077, downstream molecules such as the myosin binding subunit of the myosin light chain (MLC) phosphatase (MYPT)-1 and LIN-11, Isl1, and MEC-3 domain kinase (LIMK) might compensate for ROCK inhibition [
38].
Numerous downstream effectors are involved in the Rho signaling pathway [
39]. p140mDia is a mammalian homologue of
Drosophila diaphanous that controls actin polymerization [
40]. ROCK and p140mDia act cooperatively during stress fiber formation to mediate the effects of Rho [
41]. Arakawa et al. [
42] showed that the direction of Rho signaling is dependent on the local level of Rho-GTP, since a high Rho-GTP level induces ROCK activation and a low level preferentially activates mDia and induces Rac activation. Although we did not investigate Rac, Cdc42, and mDia in the present study, Rac1 activity was increased in cancers of the human upper urinary tract and was related to tumor progression in our previous study [
23]. Furthermore, HA-1077 did not influence RhoA activity in the current study. Rho family GTPases, including Rho, Rac, and Cdc42, have been shown to differentially and cooperatively contribute to triggering invasive behavior by tumor cells [
43]. Taken together, these findings suggest that if ROCK is suppressed by a ROCK inhibitor, other signaling pathways (including LIMK, Rho, Rac, Cdc42, and mDia) might be activated to compensate for ROCK inhibition.
However, HA-1077 may still be an attractive candidate. As described above, ROCK belongs to the AGC protein kinase family of serine-threonine kinases [
11,
36,
37], so HA-1077 might have a nonspecific inhibitory effect on other protein kinases from this family [
44]. Mutation and/or dysregulation of these AGC protein kinases contributes to the pathogenesis of human cancer [
45,
46]. Recently, Nakabayashi et al. reported that HA-1077 suppresses neovascularization and tumor growth, in association with reduced expression of VEGF, matrix metalloproteinase (MMP)-2, and MMP-9, as well as attenuating the phosphorylation of extracellular signal-regulated kinase 1 and 2 (ERK1/2) and DNA binding activity of activator proteins (a key downstream transcriptional factor for ERK1/2) in malignant glioma cells, indicating that the anti-angiogenic effect of HA-1077 may be due to the combined inhibition of ROCK and the mitogen-activated protein kinase kinase (MEK)/ERK pathway [
47].
The Rho/ROCK pathway is known to play an important role in the progression of cancer. The present findings indicate that HA-1077 prevents the proliferation and migration of bladder cancer cells and also induces apoptosis by inhibiting ROCK, suggesting that ROCK may be an attractive molecular target agent for anticancer therapy. However, this study did not show that HA-1077 was equally effective in animal models of bladder cancer developed with the 5637 or UM-UC-3 bladder cancer cell lines. Rath et al. suggested that a translational approach to ROCK signaling is necessary for clinical development of ROCK inhibitors to treat cancer and they identified the following issues to be addressed: 1) the tissue and tumor patterns of ROCK expression/activity, 2) the mode of ROCK inhibition, 3) the inhibition of ROCK targets and parallel pathways, 4) the influence of combination therapy, and 5) development of drugs targeting the extracellular matrix of tumors [
38]. In order to directly address these issues, we should compare the effectiveness of HA-1077 and its vehicle control in vivo by developing a mouse model of human bladder cancer in the future.
Improved understanding of how Rho family GTPases and their downstream effectors mutually and specifically interact in human cancers may throw more light on the best therapeutic approach to cancer and might lead to new treatment protocols.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
HA, TK* and NA initiated the study, participated in its design and coordination, carried out the study, performed the statistical analysis. HA and TK * drafted the manuscript. KH, TM, YY, AM, HY, HB, MY and YF carried out the study. HS and K-IY participated in the design of the study and helped to draft the manuscript. All authors read and approved the final manuscript.