Curcumin inhibits the formation of capillary-like tubes by rat lymphatic endothelial cells

Cancer Letters 251 (2007) 288–295 www.elsevier.com/locate/canlet

Curcumin inhibits the formation of capillary-like tubes by rat lymphatic endothelial cells Mitsuhiro Matsuo

a,b

, Hiroaki Sakurai

b,c

, Keiichi Koizumi b, Ikuo Saiki

b,c,*

a

b

Department of Anatomy, Faculty of Medicine, University of Toyama, Toyama, Japan Division of Pathogenic Biochemistry, Institute of Natural Medicine, University of Toyama, 2630 Sugitani, Toyama 930-0194, Japan c The 21st Century COE Program, University of Toyama, Toyama, Japan Received 2 October 2006; accepted 27 November 2006

Abstract The natural pigments curcumin and berberine have been shown to exhibit a variety of pharmacologic effects including anti-inflammatory, anti-cancer, and anti-metastatic properties. Here, we investigated the anti-lymphangiogenic effect with an in vitro tube-forming model using conditionally immortalized lymphatic endothelial TR-LE cells, a newly established cell line originating from the thoracic duct of a transgenic rat expressing the temperature-sensitive SV40 large T-antigen. Curcumin, but not berberine, exhibited a dose-dependent inhibition of the formation of capillary-like tubes by TR-LE cells without affecting cell viability and adhesion to Matrigel. To address the molecular mechanisms involved, we performed experiments with specific inhibitors against putative targets of curcumin, including IjB kinase (IKK), epidermal growth factor receptor (EGFR), phosphatidylinositol-3 kinase (PI3K)/Akt, and matix metalloproteinases (MMPs). While the IKK-2 inhibitor VI and EGFR tyrosine kinase inhibitors gefitinib and PD153035 had no effect, both the PI3K inhibitor LY294002 and the MMP inhibitor GM6001 shortened the tubes by approximately 50%. Western blot analysis and gelatin zymography revealed that curcumin, but not berberine, has an inhibitory effect on the phosphorylation of Akt and enzymatic activity of MMP-2 in TR-LE cells. These results suggest that curcumin exerts its inhibitory effect on lymphangiogenesis partly through Akt and MMP-2.  2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Curcumin; Berberine; Akt; MMP-2; Lymphatic metastasis; Lymphangiogenesis

1. Introduction Abbreviations: EGFR, epidermal growth factor receptor; FBS, fetal bovine serum; IKK, IjB kinase; JNK, c-Jun N-terminal kinase; MMP, matix metalloproteinase; PI3K, phosphatidylinositol-3 kinase; SV40, simian virus 40; TNF, tumor necrosis factor; VEGFR, vascular endothelial growth factor receptor. * Corresponding author. Tel.: +81 76 434 7620; fax: +81 76 434 5058. E-mail address: [email protected] (I. Saiki).

Curcumin is a naturally occurring yellow pigment from the rhizome of the perennial herb Curcuma longa. Turmeric, the powdered form of the rhizome, has been used for centuries in traditional medicine. Moreover, curcumin has been reported to block nuclear factor jB (NF-jB), IjB

0304-3835/$ - see front matter  2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2006.11.027

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

kinase (IKK), epidermal growth factor receptor (EGFR), the Akt pathway, and matrix metalloproteinases (MMPs) expression, and shown to exhibit a variety of pharmacologic effects including anti-inflammatory, anti-infectious, anti-angiogenic, and anti-cancer activities [1–3]. Berberine, an isoquinoline alkaloid and an active component of Berberis aquifolium (Oregon grape), Berberis aristata (tree tumeric), Berberis vulgaris (barberry), Coptis chinensis (coptis or goldenthread), and Hydrastis canadensis (goldenseal), also has a wide range of pharmacological and biochemical effects including anti-microbial, anti-diarrhea, anti-inflammatory, anti-angiogenic, and anti-cancer activities [4–6]. In our previous studies, both curcumin and berberine markedly inhibited the mediastinal lymph node metastasis produced by the orthotopic implantation of Lewis lung carcinoma (LLC) cells, their effects correlating with the inhibition of cell invasion and modulation of transcription factor activator protein-1 (AP-1) in LLC cells [7,8]. However, the mechanisms of this antimetastatic activity are not fully understood. In the past, it was thought that lymphatic metastasis was a passive process in which detached tumor cells reached lymph nodes via drainage through preexisting local lymphatic vessels [9]. However, recent studies have identified lymphangiogenic growth factors and histochemical markers that discriminate between blood vessels and lymphatics, revealing that lymphangiogenesis plays a critical role in metastasis [10,11]. The expression of lymphangiogenic growth factors in a range of animal tumor models leads to the formation of lymphatic vessels either within or at the periphery of the tumors and this is accompanied by enhanced lymphatic metastasis and, in some cases, by metastasis to distant organs [11]. Thus, a culture of lymphatic endothelial cells would help us to elucidate the mechanisms of lymphangiogenesis in vitro. Recently, we succeeded in establishing a rat lymphaticendothelial cell line (TR-LE) from the thoracic duct of a transgenic rat harboring a temperature-sensitive simian virus 40 (SV40) large T-antigen and enhanced green fluorescent protein (EGFP) [12]. TR-LE cells possess tube-forming ability on Matrigel and express the lymphatic endothelial markers VEGFR-3 (vascular endothelial growth factor receptor), LYVE-1 (a lymphatic endothelial receptor), Prox-1 (a homeobox gene product), and podoplanin (a glomerular podocyte membrane mucoprotein), together with the endothelial markers CD31, Tie-2, and VEGFR-2. This lymphatic endothelial cell line

289

enables one to conduct experiments on lymphangiogenesis in vitro. In this study, we assess the effect of curcumin and berberine on the formation of capillary-like tubes using TR-LE cells, and reveal potential targets of curcumin’s the anti-metastatic activity. 2. Materials and methods 2.1. Materials Curcumin and berberine were purchased from Wako Chemical (Osaka, Japan). GM6001, PD153035, and IKK-2 inhibitor VI were purchased from Calbiochem (Darmstadt, Germany). LY294002, TNF-a, and doxorubicin hydrochloride were purchased from Alomone Laboratories (Jerusalem, Israel), R&D systems, and Kyowa Hakko Induxtries, Ltd. (Tokyo, Japan), respectively. Gefitinib was kindly provided by AstraZeneca (Macclesfield, UK). Reagents were dissolved in dimethylsulfoxide or distilled water and stored at 20 C. 2.2. Cells TR-LE cells, a conditionally immortalized rat lymphatic endothelial cell line [12], were maintained on culture dishes, which had been pre-coated with 10 lg/mL fibronectin (Iwaki Glass, Tokyo, Japan), in HuMedia-EG2 (Kurabo, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS) (Lot 9354F, ICN Biomedicals Inc., Aurora, OH) at a permissive temperature (33 C). Lewis lung carcinoma (LLC) cells, kindly provided by Dr. K. Takeda (Juntendo University, Tokyo), were maintained in DMEM (Invitrogen, Carlsbad, CA) supplemented with 10% FBS at 37 C. 2.3. Tube formation assay Sub-confluent TR-LE cells were harvested with trypsin–EDTA and centrifuged at 1000 rpm for 5 min. A cell suspension (2 · 104) was then prepared in DMEM supplemented with 10% FBS and seeded in a 96-well plate that had been pre-coated with 40 ll of 10 mg/mL Matrigel (Collaborative Research Co., Bedford, MA). After a 5-h incubation at 37 C, cultures were fixed by using glutaraldehyde and stained with hematoxylin. The tube-like network was traced by using Biz-Tablet (WACOM, Saitama, Japan) and lengths were quantified with a ChemiDoc XRS system (Bio-Rad, Tokyo, Japan). 2.4. Cell proliferation assay TR-LE and LLC cells (2 · 104 cells/well) were seeded in 100 ll of DMEM containing 10% FBS in 96-well plates. Cells were allowed to adhere for 2 h, and then 25 ll of medium containing curcumin, berberine, and

290

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

doxorubicin was added. Cell viability was determined with a Cell Counting Kit (DOJINDO, Kumamoto, Japan). 2.5. Cell adhesion assay TR-LE cells were harvested using 0.25% trypsin– EDTA4Na (Invitrogen), and resuspended in DMEM supplemented with 10% FBS. TR-LE cells (2 · 104) were seeded on 96-wells, which has been pre-coated with 100 lg/mL of Matrigel, in the presence of 10 lM of curcumin and berberine. The cultures were incubated at 37 C for 30 min. The wells were treated with glutaraldehyde, and extensively washed with running water to remove unattached cells. The attached cells were stained with 0.5% crystal violet in 20% methanol for 5 min. After several washes, crystal violet was extracted with 30% acetic acid, and the absorbance was measured at 595 nm. 2.6. Western blot analysis After the indicated treatment, the cells were rinsed with ice-cold PBS and lysed in sample buffer [24 mM Tris–HCl (pH 6.8), 5% w/v glycerol, 1% w/v SDS, and 0.05% w/v bromophenol blue]. Cell lysates were subjected to SDS–polyacrylamide gel and transferred to Immobilon-P membranes (Millipore, Bedford, MA). Blots were incubated with Block Ace (Dainipponseiyaku, Suita, Japan) and probed with the indicated primary antibodies. Protein content was visualized using horseradish peroxidase-conjugated secondary antibodies followed by enhanced chemiluminescence (Amersham, Buckinghamshire, UK). Experiments were performed at least twice, and results of a representative experiment are shown. Phospho-Akt and JNK antibodies were purchased from Cell Signaling Technology Inc. (Beverly, MA).

9 h at 37 C in the presence of curcumin (10 lM) and berberine (10 lM). Gels were stained with staining solution (0.1% Coomassie brilliant blue, 10% acetic acid, and 10% isopropanol). The locations of gelatinolytic enzymes were visualized as clear bands on the blue background. 2.8. Statistical analysis Statistical comparisons were carried out using the Student two-tailed t-test with Bonferroni correction. P < 0.05 was considered to be significant. 3. Results 3.1. Curcumin inhibits TR-LE cells from forming capillarylike tubes In our previous studies, the oral administration of curcumin and berberine inhibited spontaneous metastasis to mediastinal lymph nodes following the orthotopic implantation of LLC cells [7,8]. To evaluate the effect of curcumin and berberine on lymphangiogenesis, we performed an assay using TR-LE cells, in which the lengths of the capillary-like tubes formed on a murine tumor-derived matrix, Matrigel, were measured. Curcumin shortened the tubes in a dose-dependent manner with an IC50 of 3.0 lM, whereas 10 lM of berberine did not affect the formation of the tube-like network (Fig. 1). Curcumin had no effect on cell adhesion to Matrigel (Fig. 2A), suggesting that it influences the post-adhesion processes of migration, reorganization, and consolidation [13]. In addition, berberine and curcumin did not act as cytotoxic agents (Fig. 2B). However, berberine, but not curcumin, exhibited a slight cytotoxic effect against LLC cells (Fig. 2C), which is consistent with our previous observation [8].

2.7. Gelatin zymography

3.2. Inhibition of IKK is independent of the inhibitory effect of curcumin

TR-LE cells were allowed to grow to subconfluence in 100-mm tissue culture dishes in HuMedia-EG2 containing 10% FBS. After several washes with warm PBS, the medium was replaced with DMEM containing 1% FBS, and the cultures were incubated for a further 48 h. The culture supernatant was collected and centrifuged to remove debris. The conditioned media were mixed with sample buffer, and applied to SDS–polyacrylamide gels (7.5% w/v) co-polymerized with gelatin (0.1% w/v). After electrophoresis, gels were washed twice with rinsing buffer (50 mM Tris–HCl, 2.5% Triton X-100, 5 mM CaCl2, 1 lM ZnCl2, and 0.05% NaN3), and incubated with incubation buffer (50 mM Tris– HCl, 5 mM CaCl2, 1 lM ZnCl2, and 0.05% NaN3) for

NF-jB, which is mainly a heterodimer of p65 and p50 subunits, has been identified as a target of curcumin. It is reported that curcumin abolished both the phosphorylation and degradation of IjBa induced by TNF, and subsequently inhibited translocation of the p65 subunit of NF-jB to the nucleus [14]. As shown in Fig. 3A, curcumin did not inhibit the phosphorylation induced by TNF-a, suggesting that curcumin hardly acts as an IKK inhibitor in TR-LE cells. Moreover, complete inhibition of p65’s phosphorylation by an IKK-b inhibitor is not sufficient to have an inhibitory effect on the formation of tubes (Fig. 3B). These results strongly suggest that inhibition of IKK is not involved in the anti-lymphangiogenic activity of curcumin.

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

291

A

curcumin (10 μM)

B

Tube length (% of control)

control

berberine (10 μM)

140 120 100 80 60

* **

40 20 0 control

1

3

10

curcumin

10

[μM]

berberine

Fig. 1. Effect of curcumin and berberine on the tube-like formations of TR-LE cells. TR-LE cells were seeded onto 96-well plates, which had been coated with 40 ll of Matrigel per well, in the presence of curcumin and berberine. After a 5-h incubation period at 37 C, the culture was fixed by glutaraldehyde, and the length of the tube-like formations was evaluated. Original magnification: 5·. A representative experiment is shown, with the means ± SD of four wells. *P < 0.05, **P < 0.01.

3.3. Involvement of Akt’s inhibition in curcumin’s activities It is reported that curcumin suppresses the Akt signaling pathway and subsequently exerts anti-proliferative and apoptotic activities in leukemia cell lines [15,16]. To investigate the effect of curcumin on Akt signaling in an assay of tube-forming activity in vitro, the status of phosphorylated Akt in TR-LE cells 30 min after seeding onto Matrigel-coated dishes was assessed by Western blotting using anti-phospho-Akt (Ser473) antibody (Fig. 4A). Phospho-Akt levels in TR-LE cells were reduced by treatment with curcumin and the PI3K inhibitor LY294002, but not berberine. To determine whether Akt’s inhibition plays a part in the formation of tubes, we performed the tube-forming assay of TR-LE cells in the presence of LY294002. LY294002 significantly inhibited the formation of tubes (inhibitory rate = 48%), although less so than curcumin (Fig. 4B). These results suggest that the Akt pathway is, at least in part, necessary for the tubeforming activity of TR-LE cells on Matrigel and might be a potential target for the anti-lymphangiogenic effect of curcumin. Curcumin is also a potential inhibitor of EGFR tyrosine kinase [17], which is representative of the enzymes upstream of the PI3K/Akt pathway and could be transactivated by integrins [18,19], a family of heterodimetric and transmembrane receptors. To investigate the involvement of EGFR tyrosine kinase in the formation of tubes, we examined the effect of two EGFR tyrosine kinase inhibitors, gefitinib and PD153035, on Akt’s status and tubeforming activity in TR-LE cells. Unexpectedly, neither Akt’s phosphorylation nor the tube-forming property of TR-LE cells on Matrigel was affected by these inhibitors

(Fig. 4A and C), suggesting that EGFR tyrosine kinase functions independently of the constitutive phosphorylation of Akt, and is not a principal target of curcumin in lymphangiogenesis in vitro. 3.4. Involvement of MMP-2 in lymphangiogenesis As MMP-2/gelatinase A is vital during the formation of tubular networks of vascular endothelial cells [20], we attempted to determine whether the inhibitory effect of curcumin could be mediated by MMP-2’s activity. To investigate the influence of MMP, we examined the effect of an MMP inhibitor, GM6001, on the formation of tubes by TR-LE cells (Fig. 5A). Treatment with GM6001 caused a 51.7% reduction in tube-forming activity. To assess the effect of curcumin on MMP-2’s enzymatic activity, we performed gelatin zymography (Fig. 5B). The degradation of a gelatin substrate by MMP-2 produced by TR-LE cells was completely inhibited by curcumin, but not inhibited by berberine.

4. Discussion The serine/threonine protein kinase Akt, known to be an important component of migratory and prosurvival signaling pathways, lies at a key junction in cellular signaling pathways leading to tumor angiogenesis [21,22]. Recent investigations of different Akt isoforms have demonstrated that the amount of phosphorylated Akt is increased in tumor vessels and that Akt1, the predominant

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

100 80

Inhibitors:

-

-

ib in rine inh m e u 2 rc rb Kcu be IK

60

TNF-

-

+

+

40 20

+

p65

cin

e

ru

er

do xo

rb

bi

in

in

cu

rc

um

ro nt co

be

120 100 80 60 40 20 0

** l

120 100 80 60 40 20 0

B

* **

Fig. 2. Effect of curcumin and berberine on adhesion and proliferation of TR-LE and LLC cells. (A) TR-LE cells (2 · 104) were seeded onto 96-well plates, pre-coated with 100 lg/mL of Matrigel, in the presence of 10 lM of curcumin and berberine. After a 30-min incubation, non-adherent cells were washed away and adherent cells were determined by crystal violet staining. (B and C) TR-LE cells (B) or LLC cells (C) were incubated with 10 lM of curcumin and berberine, and cell viability was determined with a cell counting kit. Doxorubicin (5 lg/mL) served as a positive control. A representative experiment is shown, with the means ± SD of four wells. *P < 0.05, **P < 0.01.

isoform in endothelial cells, regulates vascular maturation and permeability [23]. Akt phosphorylates endothelial NO synthase (eNOS) and thereby enhances endothelial NO synthesis and influences postnatal vessel growth [21,24]. However, little is known about the role of Akt in lymphangiogenesis. Here, we demonstrated, for the first time, that Akt plays an important role in the formation of capillary-like tubes by lymphatic endothelial cells. Thus, further dissection of the Akt pathway and

Tube length (% of control)

co nt ro l cu rc um in be rb er in e

0

co nt ro l cu rc um in be rb er in e do xo ru bi cin

C

+

it.

p-p65

Cell viability for 24 h (% of control)

B

A

120

Cell viability for 24 h (% of control)

A

Cell adhesion to Matrigel (% of control)

292

120 100 80 60 40

**

20 0

l

t. in bi um hi c n i r 2 cu KIK

ro nt co

Fig. 3. Inhibition of IKK is independent of the anti-lymphangiogenic effect of curcumin. (A) TR-LE cells were treated for 30 min with 10 lM of curcumin, berberine, and IKK-2 inhibitor VI, followed by TNF-a (10 ng/mL). Phospho-NF-jB was determined by Western blotting using anti-phospho-p65 (Ser 536) antibody. (B) TR-LE cells (2 · 104) were seeded onto 96-well plates, which had been coated with 40 ll of Matrigel per well, in the presence of IKK-2 inhibitor VI (10 lM). After a 5-h incubation period at 37 C, the culture was fixed by glutaraldehyde, and the length of tube-like formations was evaluated. Curcumin (20 lM) served as a positive control. A representative experiment is shown, with the means ± SD of four wells. **P < 0.01.

elucidation of downstream effector molecules will lead to a better understanding of physiological and pathological lymphangiogenesis, and may provide avenues for the development of a novel therapeutic strategy for cancer treatment. MMP-2 expression was associated with the onset of lymphangiogenesis in a model of regenerating skin [25]. In ovarian carcinoma, MMP-2 was closely correlated with VEGF-C expression, which was correlated with lymphangiogenesis and metastasis to the retroperitoneal lymph nodes [26]. Moreover, elevated MMP-2 levels are associated with lymph node metastasis in gastric, oral, colorectal, and head and neck cancer [27–30]. It is also reported that the formation of tube-like structures by vascular endothelial cells in vitro depends on MMP-1, MMP-9, MT1-MMP, and MT3-MMP [31–34]. Given that curcumin exerts broad-spectrum inhibition of the gene expression of MMPs including MMP-1, -2, -3, -9, and -14, MT1-MMP [35–37], we cannot rule

Akt 2

3

4

5

6

120

80 40 20 0

in

B

**

60

co nt ro l

80 40 20

113.7

**

0

**

60

100

80.9 control

LY294002 curcumin

MMP-2

63.8 49.5 [kDa]

120 100 80 60 40 20 ge fit in ib

PD 15 30 35

0 co nt ro l

C

Tube length (% control)

B

Tube Length (% of control)

1

100

be rb er in e

p-Akt

120

cu rc um

-

A

293

G M 60 01

Inhibitors:

2 e in in 00 01 ib um rber 294 60 fitin c r M ge be LY G cu

co nt ro l

A

Tube length (% of control)

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

Fig. 4. Inhibitory effect of curcumin on Akt pathway of TR-LE cells. (A) TR-LE cells (2 · 105) were seeded onto Matrigel (100 lg/mL)-coated dishes in the presence of curcumin (10 lM), berberine (10 lM), LY294002 (20 lM), GM6001 (10 lM), and gefitinib (1 lM). After 30 min of incubation at 37 C, cells were lysed with sample buffer, and subjected to Western blotting using anti-phospho-Akt (Ser473) antibody. (B and C) TR-LE cells (2 · 104) were seeded onto 96-well plates, which had been coated with 40 ll of Matrigel per well, in the presence of the PI3K inhibitor LY294002 (10 lM), the EGFR tyrosine kinase inhibitor PD153035 (10 lM), and gefitinib (10 lM). After a 5-h incubation period at 37 C, the culture was fixed by glutaraldehyde, and the length of tube-like formations was evaluated. Curcumin (20 lM) served as a positive control. A representative experiment is shown, with the means ± SD of four wells. **P < 0.01.

out the possibility of the involvement of these MMPs in the inhibitory effect of curcumin on the tube-forming activity of lymphatic endothelial cells and subsequent metastasis to lymph nodes. Thus, further investigations need to identify the subtypes of MMPs as pharmacological targets of curcumin. Berberine did not affect the viability or tube-forming activity of lymphatic endothelial cells. Our previous study revealed that berberine reduces the invasiveness of LLC cells, accompanied by an

Fig. 5. Inhibitory effect of curcumin on MMP-2 activity of TRLE cells. (A) TR-LE cells (2 · 104) were seeded onto 96-well plates, which had been coated with 40 ll of Matrigel per well, in the presence of the MMP inhibitor GM6001 (10 lM). After a 5-h incubation period at 37 C, the culture was fixed by glutaraldehyde, and the length of tube-like formations was evaluated. A representative experiment is shown, with the means ± SD of four wells. **P < 0.01. (B) TR-LE cells were cultured for 48 h in DMEM supplemented with 1% FBS at 37 C. The conditioned media were subjected to electrophoresis in a gelatin-embedded SDS–polyacrylamide gel. After electrophoresis, stripes of the gel were incubated in the presence of curcumin (10 lM) and berberine (10 lM) for 8 h, and stained with Coomassie brilliant blue.

inhibition of activator protein-1 (AP-1) and repression of urokinase-type plasminogen activator (uPA) [8]. It is likely that berberine is more selective to proliferation to cancer cells than non-cancerous cells, and that its inhibitory effect on lymph node metastasis greatly depends on inhibition of the metastatic properties of LLC cells, but not inhibition of lymphangiogenesis. Of interest, we found that berberine significantly inhibits the formation of tubes by vascular endothelial TR-BE cells, derived from the inferior vena cava of a transgenic rat harboring a temperature-sensitive SV40 large T-antigen (data not shown). Future studies should focus on the difference in its activity, and reveal a novel discrimination between vascular blood and lymphatic endothelial cells. In conclusion, we found that curcumin, the yellow pigment in turmeric, has an inhibitory effect of the tube-forming process in lymphatic endothelial cells.

294

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295

Although curcumin might act as an anti-metastatic agent by inhibiting multiple signaling pathways including Akt and MMP-2, additional studies are needed to elucidate the precise mechanism by which curcumin accomplish these tasks. Moreover, in vitro assays using conditionally immortalized lymphatic endothelial TR-LE cells provide a new way to investigate the mechanisms of lymphangiogenesis and screen anti-lymphangiogenic drugs. Taken together, curcumin might be used as a broad-spectrum modulator offering clinical benefits as an anti-cancer and anti-metastasic agent. Acknowledgements This study was supported in part by Grants-inAid for the 21st Century COE Program and for CLUSTER (Cooperative Link of Unique Science and Technology for Economy Revitalization) from the Ministry of Education, Culture, Sports, Science and Technology, Japan. References [1] R.K. Maheshwari, A.K. Singh, J. Gaddipati, R.C. Srimal, Multiple biological activities of curcumin: a short review, Life Sci. 78 (2006) 2081–2087. [2] R.A. Sharma, A.J. Gescher, W.P. Steward, Curcumin: the story so far, Eur. J. Cancer 41 (2005) 1955–1968. [3] Y. Ohashi, Y. Tsuchiya, K. Koizumi, H. Sakurai, I. Saiki, Prevention of intrahepatic metastasis by curcumin in an orthotopic implantation model, Oncology 65 (2003) 250–258. [4] A.H. Amin, T.V. Subbahaiah, K.M. Abbasi, Berberine sulfate: antimicrobial activity, bioassay, and mode of action, Can. J. Microbiol. 15 (1969) 1067–1076. [5] C.L. Kuo, C.W. Chi, T.Y. Liu, The anti-inflammatory potential of berberine in vitro and in vivo, Cancer Lett. 2 (2004) 127–137. [6] S. Lin, S.C. Tsai, C.C. Lee, B.W. Wang, J.Y. Liou, K.G. Shyu, Berberine inhibits HIF-1alpha expression via enhanced proteolysis, Mol. Pharmacol. 66 (2004) 612–619. [7] K. Ichiki, N. Mitani, Y. Doki, H. Hara, T. Misaki, I. Saiki, Regulation of activator protein-1 activity in the mediastinal lymph node metastasis of lung cancer, Clin. Exp. Metastasis 18 (2000) 539–545. [8] N. Mitani, K. Murakami, T. Yamaura, T. Ikeda, I. Saiki, Inhibitory effect of berberine on the mediastinal lymph node metastasis produced by orthotopic implantation of Lewis lung carcinoma, Cancer Lett. 165 (2001) 35–42. [9] M.S. Pepper, Lymphangiogenesis and tumor metastasis: myth or reality?, Clin Cancer Res. 7 (2001) 462–468. [10] P. Saharinen, T. Tammela, M.J. Karkkainen, K. Alitalo, Lymphatic vasculature: development, molecular regulation and role in tumor metastasis and inflammation, Trends Immunol. 25 (2004) 387–395. [11] S.A. Stacker, M.G. Achen, L. Jussila, M.E. Baldwin, K. Alitalo, Lymphangiogenesis and cancer metastasis, Nat. Rev. Cancer 2 (2002) 573–583.

[12] M. Matsuo, K. Koizumi, S. Yamada, M. Tomi, R.-I. Takahashi, M. Ueda, T. Terasaki, M. Obinata, K.-I. Hosoya, O. Ohtani, I. Saiki, Establishment and characterization of conditionally immortalized endothelial cell lines from the thoracic duct and inferior vena cava of tsA58/EGFP doubletransgenic rats, Cell Tissue Res. 326 (2006) 749–758. [13] D.A. Glesne, W. Zhang, S. Mandava, L. Ursos, M.E. Buell, L. Makowski, D.J. Rodi, Subtractive transcriptomics: establishing polarity drives in vitro human endothelial morphogenesis, Cancer Res. 66 (2006) 4030–4040. [14] S. Singh, B.B. Aggarwal, Activation of transcription factor NF-kappa B is suppressed by curcumin (diferuloylmethane), J. Biol. Chem. 270 (1995) 24995–25000. [15] M. Tomita, T. Matsuda, H. Kawakami, J.N. Uchihara, T. Okudaira, M. Masuda, K. Ohshiro, N. Mori, Curcumin targets Akt cell survival signaling pathway in HTLV-Iinfected T-cell lines, Cancer Sci. 97 (2006) 322–327. [16] A.R. Hussain, M. Al-Rasheed, P.S. Manogaran, K.A. AlHussein, L.C. Platanias, K. Al Kuraya, S. Uddin, Curcumin induces apoptosis via inhibition of PI3’-kinase/AKT pathway in acute T cell leukemias, Apoptosis 11 (2006) 245–254. [17] M.S. Squires, E.A. Hudson, L. Howells, S. Sale, C.E. Houghton, J.L. Jones, L.H. Fox, M. Dickens, S.A. Prigent, M.M. Manson, Relevance of mitogen activated protein kinase (MAPK) and phosphotidylinositol-3-kinase/protein kinase B (PI3K/PKB) pathways to induction of apoptosis by curcumin in breast cells, Biochem. Pharmacol. 65 (2003) 361–376. [18] A. Kalmes, G. Daum, A.W. Clowes, EGFR transactivation in the regulation of SMC function, Ann. N. Y. Acad. Sci. 947 (2001) 42–54. [19] M. Matsuo, H. Sakurai, Y. Ueno, O. Ohtani, I. Saiki, Activation of MEK/ERK and PI3K/Akt pathways by fibronectin requires integrin alphav-mediated ADAM activity in hepatocellular carcinoma: a novel functional target for gefitinib, Cancer Sci. 97 (2006) 155–162. [20] M. Nguyen, J. Arkell, C.J. Jackson, Human endothelial gelatinases and angiogenesis, Int. J. Biochem. Cell Biol. 33 (2001) 960–970. [21] F.X. Ma, Z.C. Han, Akt signaling and its role in postnatal neovascularization, Histol. Histopathol. 20 (2005) 275–281. [22] I. Shiojima, K. Walsh, Role of Akt signaling in vascular homeostasis and angiogenesis, Circ. Res. 90 (2002) 1243–1250. [23] J. Chen, P.R. Somanath, O. Razorenova, W.S. Chen, N. Hay, P. Bornstein, T.V. Byzova, Akt1 regulates pathological angiogenesis, vascular maturation and permeability in vivo, Nat. Med. 11 (2005) 1188–1196. [24] E. Ackah, J. Yu, S. Zoellner, Y. Iwakiri, C. Skurk, R. Shibata, N. Ouchi, R.M. Easton, G. Galasso, M.J. Birnbaum, K. Walsh, W.C. Sessa, Akt1/protein kinase Balpha is critical for ischemic and VEGF-mediated angiogenesis, J. Clin. Invest. 115 (2005) 2119–2127. [25] J.M. Rutkowski, K.C. Boardman, M.A. Swartz, Characterization of lymphangiogenesis in a model of adult skin regeneration, Am. J. Physiol. Heart Circ. Physiol. 291 (2006) H1402–H1410. [26] M. Ueda, Y.C. Hung, Y. Terai, K. Kanda, M. Kanemura, H. Futakuchi, H. Yamaguchi, D. Akise, M. Yasuda, M. Ueki, Vascular endothelial growth factor-C expression and invasive phenotype in ovarian carcinomas, Clin. Cancer Res. 11 (2005) 3225–3232.

M. Matsuo et al. / Cancer Letters 251 (2007) 288–295 [27] Z.Y. Wu, J.H. Li, W.H. Zhan, Y.L. He, Lymph node micrometastasis and its correlation with MMP-2 expression in gastric carcinoma, World J. Gastroenterol. 12 (2006) 2941–2944. [28] B.P. Patel, P.M. Shah, U.M. Rawal, A.A. Desai, S.V. Shah, R.M. Rawal, P.S. Patel, Activation of MMP-2 and MMP-9 in patients with oral squamous cell carcinoma, J. Surg. Oncol. 90 (2005) 81–88. [29] M. Langenskiold, L. Holmdahl, P. Falk, M.L. Ivarsson, Increased plasma MMP-2 protein expression in lymph nodepositive patients with colorectal cancer, Int. J. Colorectal Dis. 20 (2005) 245–252. [30] S. Wiegand, A.A. Dunne, H.H. Muller, R. Mandic, P. Barth, R.K. Davis, J.A. Werner, Metaanalysis of the significance of matrix metalloproteinases for lymph node disease in patients with head and neck squamous cell carcinoma, Cancer 104 (2005) 94–100. [31] E. Guenzi, K. Topolt, C. Lubeseder-Martellato, A. Jorg, E. Naschberger, R. Benelli, A. Albini, M. Sturzl, The guanylate binding protein-1 GTPase controls the invasive and angiogenic capability of endothelial cells through inhibition of MMP-1 expression, EMBO J. 22 (2003) 3772–3782. [32] U. Jadhav, S. Chigurupati, S.S. Lakka, S. Mohanam, Inhibition of matrix metalloproteinase-9 reduces in vitro

[33]

[34]

[35]

[36]

[37]

295

invasion and angiogenesis in human microvascular endothelial cells, Int. J. Oncol. 25 (2004) 1407–1414. J.W. Jeong, H.J. Cha, D.Y. Yu, M. Seiki, K.W. Kim, Induction of membrane-type matrix metalloproteinase-1 stimulates angiogenic activities of bovine aortic endothelial cells, Angiogenesis 3 (1999) 167–174. M. Plaisier, K. Kapiteijn, P. Koolwijk, C. Fijten, R. Hanemaaijer, J.M. Grimbergen, A. Mulder-Stapel, P.H. Quax, F.M. Helmerhorst, V.W. van Hinsbergh, Involvement of membrane-type matrix metalloproteinases (MT-MMPs) in capillary tube formation by human endometrial microvascular endothelial cells: role of MT3-MMP, J. Clin. Endocrinol. Metab. 89 (2004) 5828–5836. S.Y. Kim, S.H. Jung, H.S. Kim, Curcumin is a potent broad spectrum inhibitor of matrix metalloproteinase gene expression in human astroglioma cells, Biochem. Biophys. Res. Commun. 337 (2005) 510–516. Z.M. Shao, Z.Z. Shen, C.H. Liu, M.R. Sartippour, V.L. Go, D. Heber, M. Nguyen, Curcumin exerts multiple suppressive effects on human breast carcinoma cells, Int. J. Cancer. 98 (2002) 234–240. A. Banerji, J. Chakrabarti, A. Mitra, A. Chatterjee, Effect of curcumin on gelatinase A (MMP-2) activity in B16F10 melanoma cells, Cancer Lett. 211 (2004) 235–242.