Clin Exp Metastasis (2014) 31:795–803 DOI 10.1007/s10585-014-9669-y
RESEARCH PAPER
The phosphodiesterase 3 inhibitor cilostazol does not stimulate growth of colorectal liver metastases after major hepatectomy Moritz J. Strowitzki • Stefan Dold • Maximilian von Heesen • Christina Ko¨rbel Claudia Scheuer • Mohammed R. Moussavian • Martin K. Schilling • Otto Kollmar • Michael D. Menger
•
Received: 11 February 2014 / Accepted: 7 July 2014 / Published online: 23 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Liver failure after extended hepatectomy represents a major challenge in the surgery of hepatic colorectal metastasis. A previous study has indicated that inhibition of phosphodiesterase type 3 (PDE 3) stimulates liver regeneration. However, little is known whether PDE 3 inhibitors, such as cilostazol, also stimulate the growth of remnant metastases. Therefore, we herein studied the effect of cilostazol on engraftment, vascularization and growth of colorectal liver metastasis after major hepatectomy. WAGrats underwent either major hepatectomy or sham operation. Metastases were induced by subcapsular implantation
Moritz J. Strowitzki and Stefan Dold have contributed equally to this work. M. J. Strowitzki C. Ko¨rbel C. Scheuer M. D. Menger Institute for Clinical & Experimental Surgery, Saarland University, Homburg/Saar, Germany Present Address: M. J. Strowitzki (&) Department of General, Visceral and Transplant Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany e-mail: [email protected] S. Dold M. von Heesen M. R. Moussavian M. K. Schilling O. Kollmar Department of General, Vascular and Pediatric Surgery, Saarland University Hospital, Homburg/Saar, Germany Present Address: O. Kollmar Department of General, Visceral and Pediatric Surgery, University Medical Center Go¨ttingen, Go¨ttingen, Germany
of 5 9 105 CC531-colorectal cancer cells. Animals were daily treated with cilostazol (5 mg/kg body weight) or glucose solution. Tumor growth was measured by highresolution ultrasound at days 7 and 14. Tumor vascularization and tumor cell proliferation were determined by immunohistochemistry and western blotting. High-resolution ultrasound analysis in hepatectomized and non-hepatectomized animals showed that cilostazol does not stimulate tumor growth. Accordingly, the number of PCNA-positive tumor cells did not differ between cilostazol-treated animals and sham-treated controls. Interestingly, cilostazol reduced tumor vascularization in both hepatectomized and non-hepatectomized animals. This was indicated by a significantly lower number of platelet-endothelial cell adhesion molecule (PECAM-1)-positive cells in tumors of cilostazol-treated animals compared to shamtreated controls. The PDE 3 inhibitor cilostazol does not stimulate the growth of colorectal metastases during liver regeneration after major hepatectomy. Keywords Hepatectomy Cilostazol Tumor growth PECAM-1 Tumor vascularization
Introduction Although the incidence rates of colorectal cancer (CRC) are declining during the last years, CRC is still the third leading cause for cancer mortality in the US [1]. Fifty percent of the patients with primary CRC develop hepatic metastases, and 90 % of the patients who die from CRC show hepatic metastases [2]. Surgical resection of colorectal liver metastases is still the only curative treatment option, and results in an overall 5-year survival rate of approximately 40 % [3, 4]. However, major hepatectomy
123
796
can be associated with distinct complications, including insufficiency of the remnant liver. This may occur in particular in patients with additional liver diseases such as cirrhosis, steatosis and chemotherapy-associated steatohepatitis [5–7]. Several strategies have been introduced to stimulate liver regeneration in order to prevent failure of the remnant liver after resection. Preoperative embolization of a branch of the portal vein induces an atrophy in the corresponding liver tissue and a hypertrophy of the future liver remnant [8]. In addition, two-stage hepatectomy in patients with extensive bilobar colorectal liver metastases may allow to increase the overall liver resection volume [9]. The combination of the two techniques is capable of increasing the number of safe liver resections [10]. Pharmacological treatment strategies to stimulate liver regeneration after extended liver resection are not introduced into clinical practice yet. However, different pharmacological approaches aiming at improving liver regeneration have been evaluated [11–13]. Selective inhibition of phosphodiesterase type 3 (PDE 3) may be of particular interest, because this exerts hepatocellular proliferative and cytoprotective actions [13]. PDE 3 inhibitors increase intracellular cAMP levels in endothelial cells [14], hepatocytes [15] and hepatic stellate cells [16], and inhibit cytoplasmic Ca2? mobilization [17]. The high intracellular cAMP levels have been shown to protect against hypoxiainduced cell injury [18] and may preserve sinusoidal endothelial cell and hepatocyte integrity [13, 18]. Additionally, Dold et al. could show in a previous study that inhibition of PDE 3 by cilostazol increases hepatic microvascular blood perfusion and stimulates liver regeneration after major hepatectomy [19]. Of interest, recent studies have also shown that liver regeneration after major hepatectomy stimulates tumor growth [20–24]. There is only little information, however, as to whether PDE3-inhibitors, such as cilostazol, by increasing liver regeneration also stimulate intrahepatic metastatic growth. Therefore, this study analyzed the effect of cilostazol on engraftment, vascularization and growth of colorectal liver metastasis after major hepatectomy.
Materials and methods Animals Experiments were approved by the local governmental authority and conducted in accordance to the UKCCCR guidelines for the welfare of animals in experimental neoplasia and the NIH Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council). For this study 28
123
Clin Exp Metastasis (2014) 31:795–803
female WAG/Rij rats with a body weight of 220–260 g were used. The animals were kept one per cage in a temperature- and humidity-controlled 12-h light/dark cycle environment. All animals had free access to standard pellet food and tap water ad libitum. Surgical procedure Animals were placed in supine position on a heating pad and body temperature was maintained at 37 °C. Under isoflurane anesthesia animals underwent a 2/3 hepatectomy, which was performed according to the technique of Martins and colleagues [25]. Control animals underwent a sham operation. Directly after hepatectomy or sham operation a single liver metastasis was induced by subcapsular implantation of 5 9 105 tumor cells of the syngeneic CC531 colon carcinoma cell line into the lower side of the left liver lobe using a 27G needle (Omnicon F, Braun, Melsungen, Germany) [26]. The median laparotomy was closed by a 4-0 one-layer running suture. For high-resolution ultrasound analysis at days 7 and 14 animals were reanesthetized with isoflurane, and the abdomen was reopened by a median laparotomy. High-resolution ultrasound At days 7 and 14 ultrasound imaging of the liver was performed with the use of a 30 MHz ultrasound probe with a focal depth of 12.5 mm (Vevo 770 high-resolution imaging system, Visual-Sonics, Inc., Toronto, Ontario, Canada). For imaging animals were anesthetized by isoflurane and restrained on a heated stage. Two-dimensional images of the tumor at regular spatial intervals of 400 lm were achieved by scanning the rat liver with the linear motor-driven ultrasound probe. Outlining the boundaries of the tumor using these images offline allowed a rapid threedimensional reconstruction. The tumor volume was determined in mm3 by the integrated software [27]. Experimental design A total of 14 animals received daily oral application of 5 mg/kg body weight cilostazol ((6-[4-(1-cyclohexyl-1Htetrazol-5-yl) butoxy]-3, 4-dihydro-2-(1H)-quinolinone); Schwarz Pharma, Monheim, Germany) in glucose solution for a 5-day period before hepatectomy or sham operation. Daily cilostazol treatment was continued during the following 14-day experimental time period after hepatectomy or sham operation. Additional 14 animals received daily glucose solution and served as controls. Seven of the cilostazol-treated animals and 7 of the glucose-treated controls underwent major hepatectomy. The remaining 7 cilostazol-treated animals and 7 glucose-treated controls
Clin Exp Metastasis (2014) 31:795–803
underwent sham operation. All animals (n = 28) received then a subcapsular tumor implantation. In all animals tumor volume was measured by ultrasound at days 7 and 14. Histology and immunohistochemistry At the end of the experiments (day 14), the tumor and the adjacent host tissue was harvested. For light microscopy, formalin-fixed biopsies were embedded in paraffin. Sections of 2 lm were cut and stained with hematoxylin-andeosin (HE) for routine histology according to standard procedures. Mitotic figures in tumor tissue were counted in 20 high power fields (HPF) and are given as number per HPF. Necrotic areas within the tumors were assessed using the computer-assisted image analysis program cellSens Dimension (Olympus Co. 1.8.1, Hamburg, Germany). Necrotic areas are given in percent of the total tumor area. PCNA (proliferating cell nuclear antigen) served as an indicator for cell proliferation. Two lm sections of paraffin-embedded specimens were incubated for 18 h at 4 °C with a rabbit polyclonal anti-PCNA antibody (1:50; Santa Cruz Biotechnology). For development of PCNA, an alkaline phosphatase-conjugated goat anti-rabbit IgG (1:20; Dako Cytomation) was incubated for 30 min. Fuchsin (PCNA) was used as chromogen and hemalaun was used for counterstaining. Sections were analyzed by counting PCNA-positive cells in relation to the total number of cells in 20 HPF. Data are shown as percentage of PCNA-positive cells compared to the total cell number (%). Cleaved caspase-3 (cysteine-aspartic proteases) was used as an indicator of apoptotic cell death. Specimens were incubated overnight at room temperature with a rabbit polyclonal anti-cleaved caspase-3 antibody (1:50, Cell Signaling Technology, Frankfurt, Germany). This antibody detects endogenous levels of the large fragment (17/ 19 kDa) of activated caspase-3, but not full length caspase3. A biotinylated anti-rabbit immunglobulin antibody was used as secondary antibody (Link, LSAB-HRP, Dako-Cytomotion, Hamburg, Germany). 3,30 -diaminobenzidine was used as chromogen. The sections were counterstained with hemalaun. Positively stained cells were counted in 20 HPF per specimen. Data are shown as percentage of caspase-3positive cells compared to the total cell number (%). PECAM-1 (platelet-endothelial cell adhesion molecule1; CD31) is found exclusively on endothelial cells and can therefore serve as an indicator for vascularization. For immunohistochemical detection of PECAM-1 expression, a primary mouse-anti-rat antibody (1:500; clone TLD-3 A12, Serotec, Du¨sseldorf, Germany) and a secondary peroxidase-conjugated goat-anti-mouse antibody (Dianova,
797
Hamburg, Germany) were used. Tumor periphery and tumor center were analyzed separately. In all tumors the ‘‘tumor periphery’’ was defined as the area between tumor margin and 100 lm towards the tumor center. The remaining area of the tumor was defined as ‘‘tumor center’’ [28]. PECAM-1-positive cells were counted in 10 HPF per section, and are given as number per HPF [29].
Western blot analysis For whole protein extracts and Western blot analysis of vascular endothelial growth factor (VEGF), tumor and normal liver tissue, sampled at day 14 after liver resection or sham operation, was homogenized in lysis buffer (10 mM Tris pH 7.5, 10 mM NaCl, 0.1 mM EDTA, 0.5 % Triton-X 100, 0.02 % NaN3, 0.2 mM PMSF, and protease inhibitor cocktail [1:75 v/v; Sigma-Aldrich, Taufkirchen, Germany)], incubated for 30 min on ice and centrifuged for 30 min at 16,0009g. Protein concentrations were determined using the Lowry assay. The whole protein extracts (30 lm protein per lane) were separated discontinuously on sodium dodecyl sulfate polyacrylamide gels and transferred to polyvinylidene difluoride membranes (BioRad, Mu¨nchen, Germany). After blockade of non-specific binding sites, membranes were first incubated for 3.5 h at room temperature and then over-night at 4 °C with the primary rabbit anti-rat anti-VEGF antibody (A20, 1:100, Santa Cruz Biotechnology, Heidelberg, Germany), followed by the corresponding horseradish peroxidase-conjugated donkey anti-rabbit IgG secondary antibody (1.5 h, 1:5,000, GE Healthcare Amersham, Freiburg, Germany). Protein expression was visualized using luminol-enhanced chemiluminescence (ECL, GE Healthcare Amersham). Signals were assessed densitometrically (Quantity one, Geldoc, BioRad, Mu¨nchen, Germany).
Statistics All data are expressed as mean ± SEM. After testing normal distribution of data and homogeneity of variance across all groups and time points, differences between the groups were assessed by a one-way analysis of variance (ANOVA), followed by an appropriate post hoc test, including Bonferroni probabilities to compensate for multiple comparisons. Pair-wise comparison between control group and treatment group was performed by Student’s t test or Mann–Whitney U-test. Comparisons within each group (day 7 vs. day 14) were performed by paired Student’s t test or Wilcoxon Signed Rank test. Overall statistical significance was set at p \ 0.05. Statistical analysis
123
798
Clin Exp Metastasis (2014) 31:795–803
A
B
t
A
B
* CON (
C
)
1000
D
C
Fig. 1 Hematoxylin-and-eosin staining of tumors of cilostazoltreated animals without (a, c, d) and with hepatectomy (b) at day 14 after tumor cell implantation. a A representative image of the whole tumor (t). b A necrotic area (asterisks), which occasionally developed within the centers of the tumors. c The clear delineation of the tumor to the normal hepatic tissue (arrowheads). d A HPF with mitotic figures. Bars represent 1 mm (a), 200 lm (b), 150 lm (c) and 50 lm (d)
tumor volume [mm³]
800
600
400
200
was performed with the use of SigmaStat (Jandel, San Rafael, CA, USA). 0
CON
CLZ
CON
CLZ
Results Metastatic spreading and general health condition of the animals At day 14 a solid tumor with a diameter of about 5–10 mm could be detected in the liver of all animals. Histology confirmed the engraftment and growth of the tumors, which occasionally developed necrotic areas, and presented with a sharp delineation to the normal hepatic tissue (Fig. 1). In none of the animals systemic alterations due to metastatic spreading were observed. In particular, we could not detect any peritoneal or other extrahepatic metastases. During the whole experiment the animals showed normal feeding and cleaning habits.
Tumor volume In sham-operated animals the tumor volume significantly increased 5- to 6-fold from day 7 to day 14. Comparison of tumor volumes between cilostazol-treated animals and
123
Fig. 2 Ultrasound imaging (a, b) and three dimensional analysis of the tumor volume (c) in non-hepatectomized ([ Phx; sham-operated) animals at days 7 and 14 receiving cilostazol (b, c; CLZ) or vehicle (a, c; CON). Data are given as mean ± SEM. White dotted lines represent tumor margins. Bar represents 2 mm
vehicle-treated controls did not show a significant difference, neither at day 7 nor at day 14 (Fig. 2). Major hepatectomy induced in all animals a 2- and 4-fold higher tumor volume at day 7 and day 14 compared to shamoperated controls (Figs. 2, 3). However, this difference did not prove to be statistically significant (day 7: p = 0.128 vs. non-hepatectomized, vehicle-treated controls and p = 0.249 vs. non-hepatectomized, cilostazol-treated animals; day 14: p = 0.366 vs. non-hepatectomized, vehicle-treated controls and p = 0.097 vs. non-hepatectomized, cilostazol-treated animals). This is most probably due to some heterogeneity of the data, in particular in hepatectomized animals. In addition, tumor volumes of hepatectomized, cilostazol-treated animals did also not differ from that of hepatectomized, vehicletreated controls (Fig. 3).
Clin Exp Metastasis (2014) 31:795–803
799
B
A
CON (
CLZ (Phx)
CON (Phx)
B
A
CON (Phx)
D
C 1000
C
CLZ (
800
CLZ (Phx)
80
F
600
PCNA-positive cells [%]
tumor volume [mm³]
E
400
200
60
40
20 0
CON
CLZ
CON
CLZ 0
Fig. 3 Ultrasound imaging (a, b) and three dimensional analysis of the tumor volume (c) in hepatectomized animals (Phx) at days 7 and 14 receiving cilostazol (b, c; CLZ) or vehicle (a, c; CON). Data are given as mean ± SEM. White dotted lines represent tumor margins. Bar represents 2 mm
Tumor cell proliferation In sham-operated, vehicle-treated animals immunohistochemistry showed almost 68 % PCNA-positive tumor cells, indicating marked cell proliferation. Tumors of cilostazol-treated, sham-operated animals showed a slightly but not significantly lower number (60 %) of PCNA-positive cells (Fig. 4). After major hepatectomy the proliferation rate of tumor cells at day 14 was found only slightly higher compared to that of non-hepatectomized controls (Fig. 4). In addition, the number of PCNA-positive tumor cells did not differ between tumors of hepatectomized, cilostazol-treated animals and hepatectomized, vehicle-treated controls (Fig. 4). Analysis of mitotic figures in tumors of sham-operated, vehicle-treated animals showed 0.74 ± 0.16 per HPF. Tumors of cilostazol-treated, sham-operated animals showed a slightly but not significantly lower number
CON
CLZ
CON
CLZ
Fig. 4 PCNA staining (a–d) and percentage of PCNA-positive cells (e, f) at day 14 in tumor tissue of non-hepatectomized ([ Phx; a, c, e) and hepatectomized animals (Phx; b, d, f). Animals received either daily cilostazol treatment (c–f; CLZ) or vehicle (a, b, e, f; CON). Data are given as mean ± SEM. Arrows indicate PCNA-positive cells. Arrowheads indicate PCNA-negative cells. Bar represents 50 lm
(0.49 ± 0.07) of mitotic figures per HPF. After major hepatectomy the number of mitotic figures was moderately increased in vehicle-treated controls (0.95 ± 0.17 per HPF) and significantly increased in cilostazol-treated animals (1.07 ± 0.12 per HPF; p \ 0.05 vs. cilostazol-treated, sham-operated animals). However, the number of mitotic figures did not differ between tumors of hepatectomized, cilostazol-treated animals and hepatectomized, vehicletreated controls. Tumor cell apoptosis and necrosis In sham-operated, vehicle-treated controls immunohistochemistry showed 4.7 ± 0.5 % caspase-3-positive tumor cells. This did not differ from 4.4 ± 0.5 % in tumors of
123
800
Clin Exp Metastasis (2014) 31:795–803
cilostazol-treated, sham-operated animals. After major hepatectomy the rate of apoptotic tumor cells was also comparable in both cilostazol-treated animals (6.0 ± 0.8 %) and vehicle-treated controls (5.2 ± 0.8 %). Analysis of HE sections revealed in sham-operated, vehicle-treated controls a necrotic area within the tumors of 4.6 ± 1.1 %. This was found significantly (p \ 0.05) reduced to 0.8 ± 0.8 % in cilostazol-treated, sham-operated animals. In contrast, after major hepatectomy the necrotic area within the tumors did not differ between cilostazol-treated animals (3.0 ± 1.2 %) and vehicle-treated controls (3.7 ± 2.7 %).
A
p
c B
C
CON (
CON (Phx)
D
E
CLZ (
CLZ (Phx)
Tumor vascularization
Discussion Cilostazol has been shown to stimulate liver regeneration after hepatectomy [13, 19]. Whether cilostazol also affects metastatic tumor growth in the remnant liver has not been determined yet. We herein now demonstrate for the first time that cilostazol does not stimulate tumor growth of hepatic colorectal metastases. In fact, cilostazol did not increase the proliferative activity of the tumor cells, neither in normal livers nor in remnant livers after hepatectomy. Interestingly, the PDE 3 inhibitor was even capable of reducing the vascularization of the metastatic tumors.
123
30
F
G
25
PECAM-1 [positive cells/HPF]
In sham-operated, vehicle-treated animals PECAM-1 immunohistochemistry demonstrated a markedly more pronounced vascularization in the tumor periphery compared to the tumor center (Fig. 5). Of interest, the tumors of cilostazol-treated, sham-operated animals showed a significantly lower number of PECAM-1-positive cells in both the tumor periphery and the tumor center compared to vehicle-treated, sham-operated controls (Fig. 5). After major hepatectomy the vascularization of the tumors at day 14 was found only slightly enhanced compared to that of non-hepatectomized controls (Fig. 5). Cilostazol treatment in hepatectomized animals resulted also in a significantly reduced tumor vascularization, as indicated by a lower number of PECAM-1-positive cells, particularly in the tumor periphery (Fig. 5). To elucidate the cause for the reduced tumor vascularization after cilostazol treatment, VEGF expression in both liver and tumor tissue was studied by Western blotting. Surprisingly, at day 14 after hepatectomy cilostazol did not reduce VEGF expression as expected, but even slightly increased the expression in both liver and tumor tissue (Fig. 6).
20
15
10
5
*
*
*
0 tumor periphery
tumor center
tumor periphery
tumor center
Fig. 5 PECAM-1 staining of tumor tissue in a vehicle-treated animal (a), demonstrating a markedly more pronounced vascularization in the tumor periphery (p) compared to the tumor center (c). PECAM-1 staining (b–e) and number of PECAM-1-positive cells per HPF (f, g) at day 14 in tumor tissue (periphery and center of the tumors) of non-hepatectomized ([ Phx; b, d, f) and hepatectomized animals (Phx; c, e, g). Animals received either daily cilostazol treatment (d–g; CLZ, filled square) or vehicle (a–c, f, g; CON, open square). Data are given as mean ± SEM. *p \ 0.05 vs. CON. Bars represent 200 lm (a) and 50 lm (b–e)
Clin Exp Metastasis (2014) 31:795–803
control
801
cilostazol
A
liver
B
tumor
12
C
D
10
VEGF [OD*mm²]
8
6
4
* 2
0
CON
CLZ
liver
CON
CLZ
tumor
Fig. 6 VEGF expression analyzed by western blotting (a, b) and densitometric analysis (c, d) at day 14 after hepatectomy in normal liver (a, c) and in tumor tissue (b, d). Animals received either daily cilostazol treatment (a–right panel, b—right panel), c, d; CLZ) or vehicle (a—left panel, b—left panel, c, d; CON). Data are given as mean ± SEM. *p \ 0.05 vs. CON
A considerable number of previous studies have shown that cilostazol is capable of exerting pleiotropic actions. For more than a decade it is known that cilostazol acts vasodilatory on the endothelium [30]. Accordingly, it increases blood flow to the limbs [31] and, thus, is used in clinical practice for the treatment of intermittent claudication [32]. Cilostazol has further been shown (i) to inhibit platelet activation, platelet aggregation and thrombosis, (ii) to lower triglycerides and elevate high density lipoprotein cholesterol, and (iii) to decrease vascular smooth muscle cell growth and restenosis after coronary stent implantation [31, 33, 34]. Although cilostazol decreases proliferation of vascular smooth muscle cells [35] and kidney epithelial cells [36], it has been shown to increase hepatocellular proliferation during the process of liver regeneration after hepatectomy [13, 19]. Little is known, however, on the effects of cilostazol on tumor growth. In vitro, Murata et al. could demonstrate that cilostazol inhibits migration and thus motility of human colon cancer cells [37]. From these
data, the authors conclude that the drug may be effective to treat metastatic lesions. Of interest, Ikeda et al. demonstrated in hepatectomized animals an increase of animal survival after cilostazol treatment in a hepatic metastatic rat model using a transitional cell carcinoma [38]. Despite the fact that PDE 3 inhibition stimulates hepatocellular proliferation after hepatectomy [13, 19], we herein demonstrate that PDE 3 inhibition by cilostazol does not affect the proliferative activity of colorectal tumor cells in the liver. Of interest, cilostazol may act specifically on endothelial cells and vascularization. Mendes et al. reported that cilostazol decreases VEGF expression in implanted polyether-polyurethane sponges [39]. In line with this, Wang et al. also demonstrated a decrease of VEGF expression by cilostazol in streptozotocin-induced diabetic inflammation [40]. In contrast, several recent studies could show a stimulation of VEGF expression upon cilostazol treatment, including conditions of hind limb ischemia [41, 42] and transient forebrain ischemia [43]. The latter results are in line with the results of the present study, indicating an increased VEGF expression in livers of cilostazol-treated, hepatectomized animals. In general, increased VEGF expression is associated with an increased vascularization. Accordingly, several reports are published, demonstrating that a cilostazolinduced increase of VEGF expression is associated with an increased vascularization [41–43]. Surprisingly, our data indicate a reduction of tumor vessel density (density of PECAM-1-positive cells) after cilostazol treatment, despite an increased VEGF expression. In fact, others have also shown that the VEGF expression does not necessarily correlate with the extent of tumor vascularization [44, 45]. The fact that an increased VEGF expression is associated with a decreased vascularization may have different explanations. Potentially, cilostazol increases intracellular VEGF production but inhibits its release as described for pentoxifylline, a non-selective PDE inhibitor [46]. By this, cilostazol would prevent the VEGF-induced actions on the process of vascularization. More likely, however, cilostazol decreases the sensitivity of endothelial cells to VEGF stimulation. Netherton et al. have shown that cilostamide, which is also a selective PDE 3 inhibitor, almost abolishes VEGF-induced migration of human umbilical vein endothelial cells and human aortic endothelial cells [47]. Because endothelial cell migration is an essential step in the process of vascularization, cilostazol may have reduced tumor vascularization despite elevated VEGF expression by abolishing VEGF-induced endothelial cell migration. This view is further supported by the fact that cilostazol elevates cAMP levels [14–16, 48] and that increases in cAMP can inhibit VEGF- and basic fibroblast growth factor (bFGF)-induced proliferation [49].
123
802
Clin Exp Metastasis (2014) 31:795–803
Of interest, we did not observe shrinkage of the tumors in cilostazol-treated animals due to the reduced tumor microvessel density. This may be due to the fact that cilostazol increases individual blood vessel perfusion due to its vasodilatory effects [31]. However, it may also be possible that tumors after cilostazol treatment require less blood supply for growth, because cilostazol can increase ischemic tolerance by upregulating phosphorylated casein kinase 2 (CK2), which may counteract hypoxic cell death [50, 51]. Further, cilostazol has been shown to restore ATP release under hypoxic conditions [52], which may allow the metastatic tumors to survive with a reduced microvessel density. In conclusion, the present study demonstrates that cilostazol, which has been shown to accelerate liver regeneration after liver resection, does not stimulate the growth of colorectal liver metastases in liver remnants after major hepatectomy. Acknowledgments This study was supported by a HOMFOR 2010 grant (T201000614) of the Medical Faculty of the Saarland University. We appreciate the excellent technical assistance of Janine Becker and Christina Marx. Conflict of interest authors.
There is no conflict of interest of any of the
References 1. Siegel R, Naishadham D, Jemal A (2012) Cancer statistics, 2012. CA Cancer J Clin 62:10–29 2. Taylor I (1996) Liver metastases from colorectal cancer: lessons from past and present clinical studies. Br J Surg 83:456–460 3. Morris EJA, Forman D, Thomas JD et al (2010) Surgical management and outcomes of colorectal cancer liver metastases. Br J Surg 97:1110–1118 4. Rees M, Tekkis PP, Welsh FKS, O’Rourke T, John TG (2008) Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a multifactorial model of 929 patients. Ann Surg 247:125–135 5. Nordlinger B, Sorbye H, Glimelius B et al (2008) Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 371:1007–1016 6. Vauthey JN, Pawlik TM, Ribero D et al (2006) Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 24:2065–2072 7. Gomez D, Malik HZ, Bonney GK, Wong V, Toogood GJ, Lodge JPA, Prasad KR (2007) Steatosis predicts postoperative morbidity following hepatic resection for colorectal metastasis. Br J Surg 94:1395–1402 8. Hemming AW, Reed AI, Howard RJ et al (2003) Preoperative portal vein embolization for extended hepatectomy. Ann Surg 237:686–693 9. Wicherts DA, Miller R, de Haas RJ et al (2008) Long-term results of two-stage hepatectomy for irresectable colorectal cancer liver metastases. Ann Surg 248:994–1005
123
10. Primrose JN (2010) Surgery for colorectal liver metastases. Br J Cancer 102:1313–1318 11. Nishino M, Iimuro Y, Ueki T, Hirano T, Fujimoto J (2008) Hepatocyte growth factor improves survival after partial hepatectomy in cirrhotic rats suppressing apoptosis of hepatocytes. Surgery 144:374–384 12. Schmeding M, Boas-Knoop S, Lippert S et al (2008) Erythropoietin promotes hepatic regeneration after extended liver resection in rats. J Gastroenterol Hepatol 23:1125–1131 13. Akcan A, Kucuk C, Ok E, Canoz O, Muhtaroglu S, Yilmaz N, Yilmaz Z (2006) The effect of amrinone on liver regeneration in experimental hepatic resection model. J Surg Res 130:66–72 14. Sunagawa M, Shimada S, Hanashiro K, Nakamura M, Kosugi T (2006) Elevation of intracellular cAMP up-regulated thrombomodulin mRNA in cultured vascular endothelial cells derived from spontaneous type-II diabetes mellitus model rat. Endothelium 13:325–333 15. Ikegami T, Nishizaki T, Hiroshige S, Ohta R, Yanaga K, Sugimachi K (2001) Experimental study of a type 3 phosphodiesterase inhibitor on liver graft function. Br J Surg 88:59–64 16. Saito S, Hata K, Iwaisako K, Yanagida A, Takeiri M, Tanaka H, Kageyama S, Hirao H, Ikeda K, Asagiri M, Uemoto S (2014) Cilostazol attenuates hepatic stellate cell activation and protects mice against carbon tetrachloride-induced liver fibrosis. Hepatol Res 44:460–473 17. Sill JC, Bertha B, Berger I, Uhl C, Nugent M, Folts J (1997) Human platelet Ca2? mobilization, glycoprotein IIb/IIIa activation, and experimental coronary thrombosis in vivo in dogs are all inhibited by the inotropic agent amrinone. Circulation 96:1647–1653 18. Ogawa S, Koga S, Kuwabara K, Brett J, Morrow B, Morris SA, Bilezikian JP, Silverstein SC, Stern D (1992) Hypoxia-induced increased permeability of endothelial monolayers occurs through lowering of cellular cAMP levels. Am J Physiol 262:546–554 19. Dold S, von Heesen M, Mu¨ller S, Kollmar O, Schilling MK, Menger MD, Moussavian MR (2011) Effect of CilostazolÒ on hepatic microcirculation and liver regeneration after partial hepatectomy in a rat model. Langenbecks Arch Surg 396:581 (abstract) 20. Mizutani J, Hiraoka T, Yamashita R, Miyauchi Y (1992) Promotion of hepatic metastases by liver resection in the rat. Br J Cancer 65:794–797 21. Slooter GD, Marquet RL, Jeekel J, Ijzermans JN (1995) Tumour growth stimulation after partial hepatectomy can be reduced by treatment with tumour necrosis factor alpha. Br J Surg 82:129–132 22. Drixler TA, Borel Rinkes IH, Ritchie ED, van Vroonhoven TJ, Gebbink MF, Voest EE (2000) Continuous administration of angiostatin inhibits accelerated growth of colorectal liver metastases after partial hepatectomy. Cancer Res 60:1761–1765 23. Picardo A, Karpoff HM, Ng B, Lee J, Brennan MF, Fong Y (1998) Partial hepatectomy accelerates local tumor growth: potential roles of local cytokine activation. Surgery 124:57–64 24. Harun N, Nikfarjam M, Muralidharan V, Christophie C (2007) Liver regeneration stimulates tumor metastases. J Surg Res 138:284–290 25. Martins PNA, Theruvath TP, Neuhaus P (2008) Rodent models of partial hepatectomies. Liver Int 28:3–11 26. Kollmar O, Schilling MK, Menger MD (2004) Experimental liver metastasis: standards for local cell implantation to study isolated tumor growth in mice. Clin Exp Metastasis 21:453–460 27. Laschke MW, Ko¨rbel C, Rudzitis-Auth J et al (2010) Highresolution ultrasound imaging. Am J Pathol 176:585–593 28. Nguyen L, Fifis T, Malcontenti-Wilson C, Chan LS, Costa PN, Nikfarjam M, Muralidharan V, Christophi C (2012) Spatial morphological and molecular differences within solid tumors
Clin Exp Metastasis (2014) 31:795–803
29.
30.
31. 32.
33.
34.
35.
36.
37.
38.
39.
40.
may contribute to the failure of vascular disruptive agent treatments. BMC Cancer 12:522 Sperling J, Brandhorst D, Scha¨fer T et al (2013) Liver-directed chemotherapy of cetuximab and bevacizumab in combination with oxaliplatin is more effective to inhibit tumor growth of CC531 colorectal rat liver metastases than systemic chemotherapy. Clin Exp Metastasis 30:447–455 Nakamura T, Houchi H, Minami A et al (2001) Endotheliumdependent relaxation by cilostazol, a phosphodiesteras III inhibitor, on rat thoracic aorta. Life Sci 69:1709–1715 Weintraub WS (2006) The vascular effects of cilostazol. Can J Cardiol 22:56–60 Lee TM, Su SF, Tsai CH, Lee YT, Wang SS (2001) Differential effects of cilostazol and pentoxifylline on vascular endothelial growth factor in patients with intermittent claudication. Clin Sci 101:305–311 Geng D, Deng J, Jin D, Wu W, Wang J (2012) Effect of cilostazol on the progression of carotid intima-media thickness: a metaanalysis of randomized controlled trials. Atherosclerosis 220:177–183 Friedland SN, Eisenberg MJ, Shimony A (2012) Meta-analysis of randomized controlled trials on effect of cilostazol on restenosis rates and outcomes after percutaneous coronary intervention. Am J Cardiol 109:1397–1404 Kim JE, Sung JY, Woo CH et al (2011) Cilostazol inhibits vascular smooth muscle cell proliferation and reactive oxygen species production through activation of AMP-activated protein kinase induced by heme oxygenase-1. Korean J Physiol Pharmacol 15:203–210 Mun˜oz B, Huerta M, Lo´pez-Bayghen E (2009) Cilostazol reduces proliferation through c-Myc down-regulation in MDCK cells. Eur J Pharmacol 616:22–30 Murata K, Kameyama M, Fukui F et al (1999) Phosphodiesterase type III inhibitor, cilostazol, inhibits colon cancer cell motility. Clin Exp Metastasis 17:525–530 Ikeda Y, Matsumata T, Takenaka K, Yamagata M, Sugimachi K (1998) Effects of doxorubicin and/or cilostazol on cancer cells during liver regeneration after two-thirds hepatectomy in rats. Oncology 55:354–356 Mendes JB, Campos PP, Rocha MA, Andrade SP (2009) Cilostazol and pentoxifylline decrease angiogenesis, inflammation, and fibrosis in sponge-induced intraperitoneal adhesion in mice. Life Sci 84:537–543 Wang F, Li M, Cheng L et al (2008) Intervention with cilostazol attenuates renal inflammation in streptozotocin-induced diabetic rats. Life Sci 83:828–835
803 41. Biscetti F, Pecorini G, Straface G et al (2013) Cilostazol promotes angiogenesis after peripheral ischemia through a VEGFdependent mechanism. Int J Cardiol 167:910–916 42. Chao TH, Tseng SY, Li YH et al (2012) A novel vasculoangiogenic effect of cilostazol mediated by cross-talk between multiple signalling pathways including the ERK/p38 MAPK signalling transduction cascade. Clin Sci 123:147–159 43. Shin HK, Lee HR, Lee DH et al (2010) Cilostazol enhances neovascularization in the mouse hippocampus after transient forebrain ischemia. J Neurosci Res 88:2228–2238 44. Rajesh L, Joshi K, Bhalla V, Dey P, Radotra BD, Nijhawan R (2004) Correlation between VEGF expression and angiogenesis in breast carcinoma. Anal Quant Cytol Histol 26:105–108 45. Wechsel HW, Feil G, Bichler KH, Beiter T, Gleichmann R (2000) Serologic angiogenesis factors and microvascular density in renal cell carcinoma: two independent parameters. Anticancer Res 20:5117–5120 46. Amirkhosravi A, Meyer T, Warnes G et al (1998) Pentoxifylline inhibits hypoxia-induced upregulation of tumor cell tissue factor and vascular endothelial growth factor. Thromb Haemost 80:598–602 47. Netherton SJ, Maurice DH (2005) Vascular endothelial cell cyclic nucleotide phosphodiesterases and regulated cell migration: implications in angiogenesis. Mol Pharmacol 67:263–272 48. Hayashi H, Sudo T (2009) Effects of the cAMP-elevating agents cilostamide, cilostazol and forskolin on the phosphorylation of Akt and GSK-3beta in platelets. Thromb Haemost 102:327–335 49. D’Angelo G, Lee H, Weiner RI (1997) cAMP-dependent protein kinase inhibits the mitogenic action of vascular endothelial growth factor and fibroblast growth factor in capillary endothelial cells by blocking Raf activation. J Cell Biochem 67:353–366 50. Lee JH, Kim KY, Lee Y-K et al (2003) Cilostazol prevents focal cerebral ischemic injury by enhancing casein kinase 2 phosphorylation and suppression of phosphatase and tensin homolog deleted from chromosome 10 phosphorylation in rats. J Pharmacol Exp Ther 308:896–903 51. Santos M, Celotto AC, Capellini VK et al (2012) The protective effect of cilostazol on isolated rabbit femoral arteries under conditions of ischemia and reperfusion: the role of the nitric oxide pathway. Clinics 67:171–178 52. Sprague RS, Bowles EA, Achilleus D, Stephenson AH, Ellis CG, Ellsworth ML (2011) A selective phosphodiesterase 3 inhibitor rescues low PO2-induced ATP release from erythrocytes of humans with type 2 diabetes: implication for vascular control. Am J Physiol Heart Circ Physiol 301:2466–2472
123
RESEARCH PAPER
The phosphodiesterase 3 inhibitor cilostazol does not stimulate growth of colorectal liver metastases after major hepatectomy Moritz J. Strowitzki • Stefan Dold • Maximilian von Heesen • Christina Ko¨rbel Claudia Scheuer • Mohammed R. Moussavian • Martin K. Schilling • Otto Kollmar • Michael D. Menger
•
Received: 11 February 2014 / Accepted: 7 July 2014 / Published online: 23 July 2014 Ó Springer Science+Business Media Dordrecht 2014
Abstract Liver failure after extended hepatectomy represents a major challenge in the surgery of hepatic colorectal metastasis. A previous study has indicated that inhibition of phosphodiesterase type 3 (PDE 3) stimulates liver regeneration. However, little is known whether PDE 3 inhibitors, such as cilostazol, also stimulate the growth of remnant metastases. Therefore, we herein studied the effect of cilostazol on engraftment, vascularization and growth of colorectal liver metastasis after major hepatectomy. WAGrats underwent either major hepatectomy or sham operation. Metastases were induced by subcapsular implantation
Moritz J. Strowitzki and Stefan Dold have contributed equally to this work. M. J. Strowitzki C. Ko¨rbel C. Scheuer M. D. Menger Institute for Clinical & Experimental Surgery, Saarland University, Homburg/Saar, Germany Present Address: M. J. Strowitzki (&) Department of General, Visceral and Transplant Surgery, University of Heidelberg, Im Neuenheimer Feld 110, 69120 Heidelberg, Germany e-mail: [email protected] S. Dold M. von Heesen M. R. Moussavian M. K. Schilling O. Kollmar Department of General, Vascular and Pediatric Surgery, Saarland University Hospital, Homburg/Saar, Germany Present Address: O. Kollmar Department of General, Visceral and Pediatric Surgery, University Medical Center Go¨ttingen, Go¨ttingen, Germany
of 5 9 105 CC531-colorectal cancer cells. Animals were daily treated with cilostazol (5 mg/kg body weight) or glucose solution. Tumor growth was measured by highresolution ultrasound at days 7 and 14. Tumor vascularization and tumor cell proliferation were determined by immunohistochemistry and western blotting. High-resolution ultrasound analysis in hepatectomized and non-hepatectomized animals showed that cilostazol does not stimulate tumor growth. Accordingly, the number of PCNA-positive tumor cells did not differ between cilostazol-treated animals and sham-treated controls. Interestingly, cilostazol reduced tumor vascularization in both hepatectomized and non-hepatectomized animals. This was indicated by a significantly lower number of platelet-endothelial cell adhesion molecule (PECAM-1)-positive cells in tumors of cilostazol-treated animals compared to shamtreated controls. The PDE 3 inhibitor cilostazol does not stimulate the growth of colorectal metastases during liver regeneration after major hepatectomy. Keywords Hepatectomy Cilostazol Tumor growth PECAM-1 Tumor vascularization
Introduction Although the incidence rates of colorectal cancer (CRC) are declining during the last years, CRC is still the third leading cause for cancer mortality in the US [1]. Fifty percent of the patients with primary CRC develop hepatic metastases, and 90 % of the patients who die from CRC show hepatic metastases [2]. Surgical resection of colorectal liver metastases is still the only curative treatment option, and results in an overall 5-year survival rate of approximately 40 % [3, 4]. However, major hepatectomy
123
796
can be associated with distinct complications, including insufficiency of the remnant liver. This may occur in particular in patients with additional liver diseases such as cirrhosis, steatosis and chemotherapy-associated steatohepatitis [5–7]. Several strategies have been introduced to stimulate liver regeneration in order to prevent failure of the remnant liver after resection. Preoperative embolization of a branch of the portal vein induces an atrophy in the corresponding liver tissue and a hypertrophy of the future liver remnant [8]. In addition, two-stage hepatectomy in patients with extensive bilobar colorectal liver metastases may allow to increase the overall liver resection volume [9]. The combination of the two techniques is capable of increasing the number of safe liver resections [10]. Pharmacological treatment strategies to stimulate liver regeneration after extended liver resection are not introduced into clinical practice yet. However, different pharmacological approaches aiming at improving liver regeneration have been evaluated [11–13]. Selective inhibition of phosphodiesterase type 3 (PDE 3) may be of particular interest, because this exerts hepatocellular proliferative and cytoprotective actions [13]. PDE 3 inhibitors increase intracellular cAMP levels in endothelial cells [14], hepatocytes [15] and hepatic stellate cells [16], and inhibit cytoplasmic Ca2? mobilization [17]. The high intracellular cAMP levels have been shown to protect against hypoxiainduced cell injury [18] and may preserve sinusoidal endothelial cell and hepatocyte integrity [13, 18]. Additionally, Dold et al. could show in a previous study that inhibition of PDE 3 by cilostazol increases hepatic microvascular blood perfusion and stimulates liver regeneration after major hepatectomy [19]. Of interest, recent studies have also shown that liver regeneration after major hepatectomy stimulates tumor growth [20–24]. There is only little information, however, as to whether PDE3-inhibitors, such as cilostazol, by increasing liver regeneration also stimulate intrahepatic metastatic growth. Therefore, this study analyzed the effect of cilostazol on engraftment, vascularization and growth of colorectal liver metastasis after major hepatectomy.
Materials and methods Animals Experiments were approved by the local governmental authority and conducted in accordance to the UKCCCR guidelines for the welfare of animals in experimental neoplasia and the NIH Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, National Research Council). For this study 28
123
Clin Exp Metastasis (2014) 31:795–803
female WAG/Rij rats with a body weight of 220–260 g were used. The animals were kept one per cage in a temperature- and humidity-controlled 12-h light/dark cycle environment. All animals had free access to standard pellet food and tap water ad libitum. Surgical procedure Animals were placed in supine position on a heating pad and body temperature was maintained at 37 °C. Under isoflurane anesthesia animals underwent a 2/3 hepatectomy, which was performed according to the technique of Martins and colleagues [25]. Control animals underwent a sham operation. Directly after hepatectomy or sham operation a single liver metastasis was induced by subcapsular implantation of 5 9 105 tumor cells of the syngeneic CC531 colon carcinoma cell line into the lower side of the left liver lobe using a 27G needle (Omnicon F, Braun, Melsungen, Germany) [26]. The median laparotomy was closed by a 4-0 one-layer running suture. For high-resolution ultrasound analysis at days 7 and 14 animals were reanesthetized with isoflurane, and the abdomen was reopened by a median laparotomy. High-resolution ultrasound At days 7 and 14 ultrasound imaging of the liver was performed with the use of a 30 MHz ultrasound probe with a focal depth of 12.5 mm (Vevo 770 high-resolution imaging system, Visual-Sonics, Inc., Toronto, Ontario, Canada). For imaging animals were anesthetized by isoflurane and restrained on a heated stage. Two-dimensional images of the tumor at regular spatial intervals of 400 lm were achieved by scanning the rat liver with the linear motor-driven ultrasound probe. Outlining the boundaries of the tumor using these images offline allowed a rapid threedimensional reconstruction. The tumor volume was determined in mm3 by the integrated software [27]. Experimental design A total of 14 animals received daily oral application of 5 mg/kg body weight cilostazol ((6-[4-(1-cyclohexyl-1Htetrazol-5-yl) butoxy]-3, 4-dihydro-2-(1H)-quinolinone); Schwarz Pharma, Monheim, Germany) in glucose solution for a 5-day period before hepatectomy or sham operation. Daily cilostazol treatment was continued during the following 14-day experimental time period after hepatectomy or sham operation. Additional 14 animals received daily glucose solution and served as controls. Seven of the cilostazol-treated animals and 7 of the glucose-treated controls underwent major hepatectomy. The remaining 7 cilostazol-treated animals and 7 glucose-treated controls
Clin Exp Metastasis (2014) 31:795–803
underwent sham operation. All animals (n = 28) received then a subcapsular tumor implantation. In all animals tumor volume was measured by ultrasound at days 7 and 14. Histology and immunohistochemistry At the end of the experiments (day 14), the tumor and the adjacent host tissue was harvested. For light microscopy, formalin-fixed biopsies were embedded in paraffin. Sections of 2 lm were cut and stained with hematoxylin-andeosin (HE) for routine histology according to standard procedures. Mitotic figures in tumor tissue were counted in 20 high power fields (HPF) and are given as number per HPF. Necrotic areas within the tumors were assessed using the computer-assisted image analysis program cellSens Dimension (Olympus Co. 1.8.1, Hamburg, Germany). Necrotic areas are given in percent of the total tumor area. PCNA (proliferating cell nuclear antigen) served as an indicator for cell proliferation. Two lm sections of paraffin-embedded specimens were incubated for 18 h at 4 °C with a rabbit polyclonal anti-PCNA antibody (1:50; Santa Cruz Biotechnology). For development of PCNA, an alkaline phosphatase-conjugated goat anti-rabbit IgG (1:20; Dako Cytomation) was incubated for 30 min. Fuchsin (PCNA) was used as chromogen and hemalaun was used for counterstaining. Sections were analyzed by counting PCNA-positive cells in relation to the total number of cells in 20 HPF. Data are shown as percentage of PCNA-positive cells compared to the total cell number (%). Cleaved caspase-3 (cysteine-aspartic proteases) was used as an indicator of apoptotic cell death. Specimens were incubated overnight at room temperature with a rabbit polyclonal anti-cleaved caspase-3 antibody (1:50, Cell Signaling Technology, Frankfurt, Germany). This antibody detects endogenous levels of the large fragment (17/ 19 kDa) of activated caspase-3, but not full length caspase3. A biotinylated anti-rabbit immunglobulin antibody was used as secondary antibody (Link, LSAB-HRP, Dako-Cytomotion, Hamburg, Germany). 3,30 -diaminobenzidine was used as chromogen. The sections were counterstained with hemalaun. Positively stained cells were counted in 20 HPF per specimen. Data are shown as percentage of caspase-3positive cells compared to the total cell number (%). PECAM-1 (platelet-endothelial cell adhesion molecule1; CD31) is found exclusively on endothelial cells and can therefore serve as an indicator for vascularization. For immunohistochemical detection of PECAM-1 expression, a primary mouse-anti-rat antibody (1:500; clone TLD-3 A12, Serotec, Du¨sseldorf, Germany) and a secondary peroxidase-conjugated goat-anti-mouse antibody (Dianova,
797
Hamburg, Germany) were used. Tumor periphery and tumor center were analyzed separately. In all tumors the ‘‘tumor periphery’’ was defined as the area between tumor margin and 100 lm towards the tumor center. The remaining area of the tumor was defined as ‘‘tumor center’’ [28]. PECAM-1-positive cells were counted in 10 HPF per section, and are given as number per HPF [29].
Western blot analysis For whole protein extracts and Western blot analysis of vascular endothelial growth factor (VEGF), tumor and normal liver tissue, sampled at day 14 after liver resection or sham operation, was homogenized in lysis buffer (10 mM Tris pH 7.5, 10 mM NaCl, 0.1 mM EDTA, 0.5 % Triton-X 100, 0.02 % NaN3, 0.2 mM PMSF, and protease inhibitor cocktail [1:75 v/v; Sigma-Aldrich, Taufkirchen, Germany)], incubated for 30 min on ice and centrifuged for 30 min at 16,0009g. Protein concentrations were determined using the Lowry assay. The whole protein extracts (30 lm protein per lane) were separated discontinuously on sodium dodecyl sulfate polyacrylamide gels and transferred to polyvinylidene difluoride membranes (BioRad, Mu¨nchen, Germany). After blockade of non-specific binding sites, membranes were first incubated for 3.5 h at room temperature and then over-night at 4 °C with the primary rabbit anti-rat anti-VEGF antibody (A20, 1:100, Santa Cruz Biotechnology, Heidelberg, Germany), followed by the corresponding horseradish peroxidase-conjugated donkey anti-rabbit IgG secondary antibody (1.5 h, 1:5,000, GE Healthcare Amersham, Freiburg, Germany). Protein expression was visualized using luminol-enhanced chemiluminescence (ECL, GE Healthcare Amersham). Signals were assessed densitometrically (Quantity one, Geldoc, BioRad, Mu¨nchen, Germany).
Statistics All data are expressed as mean ± SEM. After testing normal distribution of data and homogeneity of variance across all groups and time points, differences between the groups were assessed by a one-way analysis of variance (ANOVA), followed by an appropriate post hoc test, including Bonferroni probabilities to compensate for multiple comparisons. Pair-wise comparison between control group and treatment group was performed by Student’s t test or Mann–Whitney U-test. Comparisons within each group (day 7 vs. day 14) were performed by paired Student’s t test or Wilcoxon Signed Rank test. Overall statistical significance was set at p \ 0.05. Statistical analysis
123
798
Clin Exp Metastasis (2014) 31:795–803
A
B
t
A
B
* CON (
C
)
1000
D
C
Fig. 1 Hematoxylin-and-eosin staining of tumors of cilostazoltreated animals without (a, c, d) and with hepatectomy (b) at day 14 after tumor cell implantation. a A representative image of the whole tumor (t). b A necrotic area (asterisks), which occasionally developed within the centers of the tumors. c The clear delineation of the tumor to the normal hepatic tissue (arrowheads). d A HPF with mitotic figures. Bars represent 1 mm (a), 200 lm (b), 150 lm (c) and 50 lm (d)
tumor volume [mm³]
800
600
400
200
was performed with the use of SigmaStat (Jandel, San Rafael, CA, USA). 0
CON
CLZ
CON
CLZ
Results Metastatic spreading and general health condition of the animals At day 14 a solid tumor with a diameter of about 5–10 mm could be detected in the liver of all animals. Histology confirmed the engraftment and growth of the tumors, which occasionally developed necrotic areas, and presented with a sharp delineation to the normal hepatic tissue (Fig. 1). In none of the animals systemic alterations due to metastatic spreading were observed. In particular, we could not detect any peritoneal or other extrahepatic metastases. During the whole experiment the animals showed normal feeding and cleaning habits.
Tumor volume In sham-operated animals the tumor volume significantly increased 5- to 6-fold from day 7 to day 14. Comparison of tumor volumes between cilostazol-treated animals and
123
Fig. 2 Ultrasound imaging (a, b) and three dimensional analysis of the tumor volume (c) in non-hepatectomized ([ Phx; sham-operated) animals at days 7 and 14 receiving cilostazol (b, c; CLZ) or vehicle (a, c; CON). Data are given as mean ± SEM. White dotted lines represent tumor margins. Bar represents 2 mm
vehicle-treated controls did not show a significant difference, neither at day 7 nor at day 14 (Fig. 2). Major hepatectomy induced in all animals a 2- and 4-fold higher tumor volume at day 7 and day 14 compared to shamoperated controls (Figs. 2, 3). However, this difference did not prove to be statistically significant (day 7: p = 0.128 vs. non-hepatectomized, vehicle-treated controls and p = 0.249 vs. non-hepatectomized, cilostazol-treated animals; day 14: p = 0.366 vs. non-hepatectomized, vehicle-treated controls and p = 0.097 vs. non-hepatectomized, cilostazol-treated animals). This is most probably due to some heterogeneity of the data, in particular in hepatectomized animals. In addition, tumor volumes of hepatectomized, cilostazol-treated animals did also not differ from that of hepatectomized, vehicletreated controls (Fig. 3).
Clin Exp Metastasis (2014) 31:795–803
799
B
A
CON (
CLZ (Phx)
CON (Phx)
B
A
CON (Phx)
D
C 1000
C
CLZ (
800
CLZ (Phx)
80
F
600
PCNA-positive cells [%]
tumor volume [mm³]
E
400
200
60
40
20 0
CON
CLZ
CON
CLZ 0
Fig. 3 Ultrasound imaging (a, b) and three dimensional analysis of the tumor volume (c) in hepatectomized animals (Phx) at days 7 and 14 receiving cilostazol (b, c; CLZ) or vehicle (a, c; CON). Data are given as mean ± SEM. White dotted lines represent tumor margins. Bar represents 2 mm
Tumor cell proliferation In sham-operated, vehicle-treated animals immunohistochemistry showed almost 68 % PCNA-positive tumor cells, indicating marked cell proliferation. Tumors of cilostazol-treated, sham-operated animals showed a slightly but not significantly lower number (60 %) of PCNA-positive cells (Fig. 4). After major hepatectomy the proliferation rate of tumor cells at day 14 was found only slightly higher compared to that of non-hepatectomized controls (Fig. 4). In addition, the number of PCNA-positive tumor cells did not differ between tumors of hepatectomized, cilostazol-treated animals and hepatectomized, vehicle-treated controls (Fig. 4). Analysis of mitotic figures in tumors of sham-operated, vehicle-treated animals showed 0.74 ± 0.16 per HPF. Tumors of cilostazol-treated, sham-operated animals showed a slightly but not significantly lower number
CON
CLZ
CON
CLZ
Fig. 4 PCNA staining (a–d) and percentage of PCNA-positive cells (e, f) at day 14 in tumor tissue of non-hepatectomized ([ Phx; a, c, e) and hepatectomized animals (Phx; b, d, f). Animals received either daily cilostazol treatment (c–f; CLZ) or vehicle (a, b, e, f; CON). Data are given as mean ± SEM. Arrows indicate PCNA-positive cells. Arrowheads indicate PCNA-negative cells. Bar represents 50 lm
(0.49 ± 0.07) of mitotic figures per HPF. After major hepatectomy the number of mitotic figures was moderately increased in vehicle-treated controls (0.95 ± 0.17 per HPF) and significantly increased in cilostazol-treated animals (1.07 ± 0.12 per HPF; p \ 0.05 vs. cilostazol-treated, sham-operated animals). However, the number of mitotic figures did not differ between tumors of hepatectomized, cilostazol-treated animals and hepatectomized, vehicletreated controls. Tumor cell apoptosis and necrosis In sham-operated, vehicle-treated controls immunohistochemistry showed 4.7 ± 0.5 % caspase-3-positive tumor cells. This did not differ from 4.4 ± 0.5 % in tumors of
123
800
Clin Exp Metastasis (2014) 31:795–803
cilostazol-treated, sham-operated animals. After major hepatectomy the rate of apoptotic tumor cells was also comparable in both cilostazol-treated animals (6.0 ± 0.8 %) and vehicle-treated controls (5.2 ± 0.8 %). Analysis of HE sections revealed in sham-operated, vehicle-treated controls a necrotic area within the tumors of 4.6 ± 1.1 %. This was found significantly (p \ 0.05) reduced to 0.8 ± 0.8 % in cilostazol-treated, sham-operated animals. In contrast, after major hepatectomy the necrotic area within the tumors did not differ between cilostazol-treated animals (3.0 ± 1.2 %) and vehicle-treated controls (3.7 ± 2.7 %).
A
p
c B
C
CON (
CON (Phx)
D
E
CLZ (
CLZ (Phx)
Tumor vascularization
Discussion Cilostazol has been shown to stimulate liver regeneration after hepatectomy [13, 19]. Whether cilostazol also affects metastatic tumor growth in the remnant liver has not been determined yet. We herein now demonstrate for the first time that cilostazol does not stimulate tumor growth of hepatic colorectal metastases. In fact, cilostazol did not increase the proliferative activity of the tumor cells, neither in normal livers nor in remnant livers after hepatectomy. Interestingly, the PDE 3 inhibitor was even capable of reducing the vascularization of the metastatic tumors.
123
30
F
G
25
PECAM-1 [positive cells/HPF]
In sham-operated, vehicle-treated animals PECAM-1 immunohistochemistry demonstrated a markedly more pronounced vascularization in the tumor periphery compared to the tumor center (Fig. 5). Of interest, the tumors of cilostazol-treated, sham-operated animals showed a significantly lower number of PECAM-1-positive cells in both the tumor periphery and the tumor center compared to vehicle-treated, sham-operated controls (Fig. 5). After major hepatectomy the vascularization of the tumors at day 14 was found only slightly enhanced compared to that of non-hepatectomized controls (Fig. 5). Cilostazol treatment in hepatectomized animals resulted also in a significantly reduced tumor vascularization, as indicated by a lower number of PECAM-1-positive cells, particularly in the tumor periphery (Fig. 5). To elucidate the cause for the reduced tumor vascularization after cilostazol treatment, VEGF expression in both liver and tumor tissue was studied by Western blotting. Surprisingly, at day 14 after hepatectomy cilostazol did not reduce VEGF expression as expected, but even slightly increased the expression in both liver and tumor tissue (Fig. 6).
20
15
10
5
*
*
*
0 tumor periphery
tumor center
tumor periphery
tumor center
Fig. 5 PECAM-1 staining of tumor tissue in a vehicle-treated animal (a), demonstrating a markedly more pronounced vascularization in the tumor periphery (p) compared to the tumor center (c). PECAM-1 staining (b–e) and number of PECAM-1-positive cells per HPF (f, g) at day 14 in tumor tissue (periphery and center of the tumors) of non-hepatectomized ([ Phx; b, d, f) and hepatectomized animals (Phx; c, e, g). Animals received either daily cilostazol treatment (d–g; CLZ, filled square) or vehicle (a–c, f, g; CON, open square). Data are given as mean ± SEM. *p \ 0.05 vs. CON. Bars represent 200 lm (a) and 50 lm (b–e)
Clin Exp Metastasis (2014) 31:795–803
control
801
cilostazol
A
liver
B
tumor
12
C
D
10
VEGF [OD*mm²]
8
6
4
* 2
0
CON
CLZ
liver
CON
CLZ
tumor
Fig. 6 VEGF expression analyzed by western blotting (a, b) and densitometric analysis (c, d) at day 14 after hepatectomy in normal liver (a, c) and in tumor tissue (b, d). Animals received either daily cilostazol treatment (a–right panel, b—right panel), c, d; CLZ) or vehicle (a—left panel, b—left panel, c, d; CON). Data are given as mean ± SEM. *p \ 0.05 vs. CON
A considerable number of previous studies have shown that cilostazol is capable of exerting pleiotropic actions. For more than a decade it is known that cilostazol acts vasodilatory on the endothelium [30]. Accordingly, it increases blood flow to the limbs [31] and, thus, is used in clinical practice for the treatment of intermittent claudication [32]. Cilostazol has further been shown (i) to inhibit platelet activation, platelet aggregation and thrombosis, (ii) to lower triglycerides and elevate high density lipoprotein cholesterol, and (iii) to decrease vascular smooth muscle cell growth and restenosis after coronary stent implantation [31, 33, 34]. Although cilostazol decreases proliferation of vascular smooth muscle cells [35] and kidney epithelial cells [36], it has been shown to increase hepatocellular proliferation during the process of liver regeneration after hepatectomy [13, 19]. Little is known, however, on the effects of cilostazol on tumor growth. In vitro, Murata et al. could demonstrate that cilostazol inhibits migration and thus motility of human colon cancer cells [37]. From these
data, the authors conclude that the drug may be effective to treat metastatic lesions. Of interest, Ikeda et al. demonstrated in hepatectomized animals an increase of animal survival after cilostazol treatment in a hepatic metastatic rat model using a transitional cell carcinoma [38]. Despite the fact that PDE 3 inhibition stimulates hepatocellular proliferation after hepatectomy [13, 19], we herein demonstrate that PDE 3 inhibition by cilostazol does not affect the proliferative activity of colorectal tumor cells in the liver. Of interest, cilostazol may act specifically on endothelial cells and vascularization. Mendes et al. reported that cilostazol decreases VEGF expression in implanted polyether-polyurethane sponges [39]. In line with this, Wang et al. also demonstrated a decrease of VEGF expression by cilostazol in streptozotocin-induced diabetic inflammation [40]. In contrast, several recent studies could show a stimulation of VEGF expression upon cilostazol treatment, including conditions of hind limb ischemia [41, 42] and transient forebrain ischemia [43]. The latter results are in line with the results of the present study, indicating an increased VEGF expression in livers of cilostazol-treated, hepatectomized animals. In general, increased VEGF expression is associated with an increased vascularization. Accordingly, several reports are published, demonstrating that a cilostazolinduced increase of VEGF expression is associated with an increased vascularization [41–43]. Surprisingly, our data indicate a reduction of tumor vessel density (density of PECAM-1-positive cells) after cilostazol treatment, despite an increased VEGF expression. In fact, others have also shown that the VEGF expression does not necessarily correlate with the extent of tumor vascularization [44, 45]. The fact that an increased VEGF expression is associated with a decreased vascularization may have different explanations. Potentially, cilostazol increases intracellular VEGF production but inhibits its release as described for pentoxifylline, a non-selective PDE inhibitor [46]. By this, cilostazol would prevent the VEGF-induced actions on the process of vascularization. More likely, however, cilostazol decreases the sensitivity of endothelial cells to VEGF stimulation. Netherton et al. have shown that cilostamide, which is also a selective PDE 3 inhibitor, almost abolishes VEGF-induced migration of human umbilical vein endothelial cells and human aortic endothelial cells [47]. Because endothelial cell migration is an essential step in the process of vascularization, cilostazol may have reduced tumor vascularization despite elevated VEGF expression by abolishing VEGF-induced endothelial cell migration. This view is further supported by the fact that cilostazol elevates cAMP levels [14–16, 48] and that increases in cAMP can inhibit VEGF- and basic fibroblast growth factor (bFGF)-induced proliferation [49].
123
802
Clin Exp Metastasis (2014) 31:795–803
Of interest, we did not observe shrinkage of the tumors in cilostazol-treated animals due to the reduced tumor microvessel density. This may be due to the fact that cilostazol increases individual blood vessel perfusion due to its vasodilatory effects [31]. However, it may also be possible that tumors after cilostazol treatment require less blood supply for growth, because cilostazol can increase ischemic tolerance by upregulating phosphorylated casein kinase 2 (CK2), which may counteract hypoxic cell death [50, 51]. Further, cilostazol has been shown to restore ATP release under hypoxic conditions [52], which may allow the metastatic tumors to survive with a reduced microvessel density. In conclusion, the present study demonstrates that cilostazol, which has been shown to accelerate liver regeneration after liver resection, does not stimulate the growth of colorectal liver metastases in liver remnants after major hepatectomy. Acknowledgments This study was supported by a HOMFOR 2010 grant (T201000614) of the Medical Faculty of the Saarland University. We appreciate the excellent technical assistance of Janine Becker and Christina Marx. Conflict of interest authors.
There is no conflict of interest of any of the
References 1. Siegel R, Naishadham D, Jemal A (2012) Cancer statistics, 2012. CA Cancer J Clin 62:10–29 2. Taylor I (1996) Liver metastases from colorectal cancer: lessons from past and present clinical studies. Br J Surg 83:456–460 3. Morris EJA, Forman D, Thomas JD et al (2010) Surgical management and outcomes of colorectal cancer liver metastases. Br J Surg 97:1110–1118 4. Rees M, Tekkis PP, Welsh FKS, O’Rourke T, John TG (2008) Evaluation of long-term survival after hepatic resection for metastatic colorectal cancer: a multifactorial model of 929 patients. Ann Surg 247:125–135 5. Nordlinger B, Sorbye H, Glimelius B et al (2008) Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 371:1007–1016 6. Vauthey JN, Pawlik TM, Ribero D et al (2006) Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol 24:2065–2072 7. Gomez D, Malik HZ, Bonney GK, Wong V, Toogood GJ, Lodge JPA, Prasad KR (2007) Steatosis predicts postoperative morbidity following hepatic resection for colorectal metastasis. Br J Surg 94:1395–1402 8. Hemming AW, Reed AI, Howard RJ et al (2003) Preoperative portal vein embolization for extended hepatectomy. Ann Surg 237:686–693 9. Wicherts DA, Miller R, de Haas RJ et al (2008) Long-term results of two-stage hepatectomy for irresectable colorectal cancer liver metastases. Ann Surg 248:994–1005
123
10. Primrose JN (2010) Surgery for colorectal liver metastases. Br J Cancer 102:1313–1318 11. Nishino M, Iimuro Y, Ueki T, Hirano T, Fujimoto J (2008) Hepatocyte growth factor improves survival after partial hepatectomy in cirrhotic rats suppressing apoptosis of hepatocytes. Surgery 144:374–384 12. Schmeding M, Boas-Knoop S, Lippert S et al (2008) Erythropoietin promotes hepatic regeneration after extended liver resection in rats. J Gastroenterol Hepatol 23:1125–1131 13. Akcan A, Kucuk C, Ok E, Canoz O, Muhtaroglu S, Yilmaz N, Yilmaz Z (2006) The effect of amrinone on liver regeneration in experimental hepatic resection model. J Surg Res 130:66–72 14. Sunagawa M, Shimada S, Hanashiro K, Nakamura M, Kosugi T (2006) Elevation of intracellular cAMP up-regulated thrombomodulin mRNA in cultured vascular endothelial cells derived from spontaneous type-II diabetes mellitus model rat. Endothelium 13:325–333 15. Ikegami T, Nishizaki T, Hiroshige S, Ohta R, Yanaga K, Sugimachi K (2001) Experimental study of a type 3 phosphodiesterase inhibitor on liver graft function. Br J Surg 88:59–64 16. Saito S, Hata K, Iwaisako K, Yanagida A, Takeiri M, Tanaka H, Kageyama S, Hirao H, Ikeda K, Asagiri M, Uemoto S (2014) Cilostazol attenuates hepatic stellate cell activation and protects mice against carbon tetrachloride-induced liver fibrosis. Hepatol Res 44:460–473 17. Sill JC, Bertha B, Berger I, Uhl C, Nugent M, Folts J (1997) Human platelet Ca2? mobilization, glycoprotein IIb/IIIa activation, and experimental coronary thrombosis in vivo in dogs are all inhibited by the inotropic agent amrinone. Circulation 96:1647–1653 18. Ogawa S, Koga S, Kuwabara K, Brett J, Morrow B, Morris SA, Bilezikian JP, Silverstein SC, Stern D (1992) Hypoxia-induced increased permeability of endothelial monolayers occurs through lowering of cellular cAMP levels. Am J Physiol 262:546–554 19. Dold S, von Heesen M, Mu¨ller S, Kollmar O, Schilling MK, Menger MD, Moussavian MR (2011) Effect of CilostazolÒ on hepatic microcirculation and liver regeneration after partial hepatectomy in a rat model. Langenbecks Arch Surg 396:581 (abstract) 20. Mizutani J, Hiraoka T, Yamashita R, Miyauchi Y (1992) Promotion of hepatic metastases by liver resection in the rat. Br J Cancer 65:794–797 21. Slooter GD, Marquet RL, Jeekel J, Ijzermans JN (1995) Tumour growth stimulation after partial hepatectomy can be reduced by treatment with tumour necrosis factor alpha. Br J Surg 82:129–132 22. Drixler TA, Borel Rinkes IH, Ritchie ED, van Vroonhoven TJ, Gebbink MF, Voest EE (2000) Continuous administration of angiostatin inhibits accelerated growth of colorectal liver metastases after partial hepatectomy. Cancer Res 60:1761–1765 23. Picardo A, Karpoff HM, Ng B, Lee J, Brennan MF, Fong Y (1998) Partial hepatectomy accelerates local tumor growth: potential roles of local cytokine activation. Surgery 124:57–64 24. Harun N, Nikfarjam M, Muralidharan V, Christophie C (2007) Liver regeneration stimulates tumor metastases. J Surg Res 138:284–290 25. Martins PNA, Theruvath TP, Neuhaus P (2008) Rodent models of partial hepatectomies. Liver Int 28:3–11 26. Kollmar O, Schilling MK, Menger MD (2004) Experimental liver metastasis: standards for local cell implantation to study isolated tumor growth in mice. Clin Exp Metastasis 21:453–460 27. Laschke MW, Ko¨rbel C, Rudzitis-Auth J et al (2010) Highresolution ultrasound imaging. Am J Pathol 176:585–593 28. Nguyen L, Fifis T, Malcontenti-Wilson C, Chan LS, Costa PN, Nikfarjam M, Muralidharan V, Christophi C (2012) Spatial morphological and molecular differences within solid tumors
Clin Exp Metastasis (2014) 31:795–803
29.
30.
31. 32.
33.
34.
35.
36.
37.
38.
39.
40.
may contribute to the failure of vascular disruptive agent treatments. BMC Cancer 12:522 Sperling J, Brandhorst D, Scha¨fer T et al (2013) Liver-directed chemotherapy of cetuximab and bevacizumab in combination with oxaliplatin is more effective to inhibit tumor growth of CC531 colorectal rat liver metastases than systemic chemotherapy. Clin Exp Metastasis 30:447–455 Nakamura T, Houchi H, Minami A et al (2001) Endotheliumdependent relaxation by cilostazol, a phosphodiesteras III inhibitor, on rat thoracic aorta. Life Sci 69:1709–1715 Weintraub WS (2006) The vascular effects of cilostazol. Can J Cardiol 22:56–60 Lee TM, Su SF, Tsai CH, Lee YT, Wang SS (2001) Differential effects of cilostazol and pentoxifylline on vascular endothelial growth factor in patients with intermittent claudication. Clin Sci 101:305–311 Geng D, Deng J, Jin D, Wu W, Wang J (2012) Effect of cilostazol on the progression of carotid intima-media thickness: a metaanalysis of randomized controlled trials. Atherosclerosis 220:177–183 Friedland SN, Eisenberg MJ, Shimony A (2012) Meta-analysis of randomized controlled trials on effect of cilostazol on restenosis rates and outcomes after percutaneous coronary intervention. Am J Cardiol 109:1397–1404 Kim JE, Sung JY, Woo CH et al (2011) Cilostazol inhibits vascular smooth muscle cell proliferation and reactive oxygen species production through activation of AMP-activated protein kinase induced by heme oxygenase-1. Korean J Physiol Pharmacol 15:203–210 Mun˜oz B, Huerta M, Lo´pez-Bayghen E (2009) Cilostazol reduces proliferation through c-Myc down-regulation in MDCK cells. Eur J Pharmacol 616:22–30 Murata K, Kameyama M, Fukui F et al (1999) Phosphodiesterase type III inhibitor, cilostazol, inhibits colon cancer cell motility. Clin Exp Metastasis 17:525–530 Ikeda Y, Matsumata T, Takenaka K, Yamagata M, Sugimachi K (1998) Effects of doxorubicin and/or cilostazol on cancer cells during liver regeneration after two-thirds hepatectomy in rats. Oncology 55:354–356 Mendes JB, Campos PP, Rocha MA, Andrade SP (2009) Cilostazol and pentoxifylline decrease angiogenesis, inflammation, and fibrosis in sponge-induced intraperitoneal adhesion in mice. Life Sci 84:537–543 Wang F, Li M, Cheng L et al (2008) Intervention with cilostazol attenuates renal inflammation in streptozotocin-induced diabetic rats. Life Sci 83:828–835
803 41. Biscetti F, Pecorini G, Straface G et al (2013) Cilostazol promotes angiogenesis after peripheral ischemia through a VEGFdependent mechanism. Int J Cardiol 167:910–916 42. Chao TH, Tseng SY, Li YH et al (2012) A novel vasculoangiogenic effect of cilostazol mediated by cross-talk between multiple signalling pathways including the ERK/p38 MAPK signalling transduction cascade. Clin Sci 123:147–159 43. Shin HK, Lee HR, Lee DH et al (2010) Cilostazol enhances neovascularization in the mouse hippocampus after transient forebrain ischemia. J Neurosci Res 88:2228–2238 44. Rajesh L, Joshi K, Bhalla V, Dey P, Radotra BD, Nijhawan R (2004) Correlation between VEGF expression and angiogenesis in breast carcinoma. Anal Quant Cytol Histol 26:105–108 45. Wechsel HW, Feil G, Bichler KH, Beiter T, Gleichmann R (2000) Serologic angiogenesis factors and microvascular density in renal cell carcinoma: two independent parameters. Anticancer Res 20:5117–5120 46. Amirkhosravi A, Meyer T, Warnes G et al (1998) Pentoxifylline inhibits hypoxia-induced upregulation of tumor cell tissue factor and vascular endothelial growth factor. Thromb Haemost 80:598–602 47. Netherton SJ, Maurice DH (2005) Vascular endothelial cell cyclic nucleotide phosphodiesterases and regulated cell migration: implications in angiogenesis. Mol Pharmacol 67:263–272 48. Hayashi H, Sudo T (2009) Effects of the cAMP-elevating agents cilostamide, cilostazol and forskolin on the phosphorylation of Akt and GSK-3beta in platelets. Thromb Haemost 102:327–335 49. D’Angelo G, Lee H, Weiner RI (1997) cAMP-dependent protein kinase inhibits the mitogenic action of vascular endothelial growth factor and fibroblast growth factor in capillary endothelial cells by blocking Raf activation. J Cell Biochem 67:353–366 50. Lee JH, Kim KY, Lee Y-K et al (2003) Cilostazol prevents focal cerebral ischemic injury by enhancing casein kinase 2 phosphorylation and suppression of phosphatase and tensin homolog deleted from chromosome 10 phosphorylation in rats. J Pharmacol Exp Ther 308:896–903 51. Santos M, Celotto AC, Capellini VK et al (2012) The protective effect of cilostazol on isolated rabbit femoral arteries under conditions of ischemia and reperfusion: the role of the nitric oxide pathway. Clinics 67:171–178 52. Sprague RS, Bowles EA, Achilleus D, Stephenson AH, Ellis CG, Ellsworth ML (2011) A selective phosphodiesterase 3 inhibitor rescues low PO2-induced ATP release from erythrocytes of humans with type 2 diabetes: implication for vascular control. Am J Physiol Heart Circ Physiol 301:2466–2472
123
