ELSEVIER

CANCER LETTERS

Cancer Letters 107 (1996) 109- I I5

Curcumin inhibits the proliferation and cell cycle progression of human umbilical vein endothelial cell

Anoop K. Singh, Gurmel S. Sidhu, T. Deepa, Radha K. Maheshwari”

Department of Pathology, Uniformed Services CJniversiQ of the Health Sciences, Bethesda, MD 20814. USA

Received 30 May 1996; revision received 11 June 1996; accepted 12 June 1996

Abstract

We have studied the effect of curcumin (diferuloylmethane), a major component of the food flavor turmeric, on the proliferation and cell cycle progression of human umbilical vein endothelial cells (HUVEC). Curcumin inhibited the DNA synthesis of HUVEC as revealed by [3H]thymidine incorporation in a dose-dependent manner without significantly affecting the viability of the cells. The growth of HUVEC stimulated with fibroblast growth factor (FGF) and endothelial growth supplement (ECGS) was also inhibited by curcumin. Addition of curcumin to HUVEC resulted in an accumulation of > 46% of the cells in early S-phase, as determined by the FACS analysis. Pulse labeling studies with [3H]thymidine demonstrated that curcumin affected cells that were actively undergoing DNA synthesis. The de-novo synthesis of thymidine depends on thymidine kinase (TK) enzyme. Curcumin caused a significant loss of TK activity, which may be one of the possible mechanism(s) for the inhibition of DNA synthesis activity of HUVEC by curcumin. These studies have revealed a unique mode of action of curcumin whereby it effectively blocked the cell cycle progression during S-phase by inhibiting the activity of TK enzyme. The migration, proliferation and differentiation of HUVEC leads to angiogenesis, which facilitates the tumor initiation and promotion. Since curcumin inhibited the proliferation of HUVEC, it could turn out to be a very useful compound for the development of novel anti-cancer therapy.

Keywovcls: Curcumin: HUVEC; Cell cycle; Proliferation

1. Introduction

Curcumin (diferuloylmethane), a major active com- ponent of turmeric (Curcuma longa Linn), is a crystal- line compound and has been widely used for centuries in indigenous medicine for the treatment of variety of

inflammatory conditions and other diseases [ 11. Cur- cumin has anti-inflammatory [2,3] and antioxidant activities [4,5]. The anti-carcinogenic properties of curcumin in animals has been shown by the inhibition of tumor initiation induced by benzo(a)pyrene and 7,12-dimethyl benz(a)anthracene [6-81 on mouse skin and on carcinogen-induced tumorigenesis in the fore stomach, duodenum and colon of mice [9]. It has

*Corresponding author. Tel.: +l 301 2953497; fax: +I 301 29.5 1640.

also been shown to inhibit neutrophil activation, sup- presses mitogen-induced proliferation of blood mono-

03043835/96/$12.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PII SO304-3835(96)04357-l

110 A.K. Singh et al.. I Cancer Letters 107 (1996) 109-115

nuclear cells, mixed lymphocytes reaction, and plate- let-derived growth factor (PDGF)-dependent mito- genesis of smooth muscle cells [lo].

The formation of new vessels is termed angiogen- esis. The attachment, proliferation, migration and dif- ferentiation of the endothelial cells lead to angiogenesis. Angiogenesis plays a key role in tumor initiation and promotion leading to metastasis. Recently, we have reported that curcumin inhibited the angiogenic differentiation of human umbilical vein endothelial cells (HUVEC) on matrigel in an in vitro model. Curcumin did not significantly inhibit the HUVEC attachment either on plastic or matrigel (sub- mitted for publication). It appears that proliferation of the endothelial cells may play an important role in the tube formation. The present study demonstrated that curcumin inhibited the proliferation of HUVEC in a dose-dependent manner without significantly affect- ing the viability of the cells. Curcumin retarded the progression of early S-phase in HUVEC. The inhibi- tion of DNA synthesis could be related to the inhibi- tion of thymidine kinase (TK) activity by curcumin. These findings suggest that curcumin may be a useful compound for the suppression of tumor growth by inhibiting endothelial cell proliferation and angiogen- esis.

2. Materials and methods

HUVEC were purchased from Clonetics, Inc. (San Diego, CA). Medium 199, streptomycin (100 mg/ml), penicillin (100 U/ml), gentamycin (50 mg/ml), fungi- zone (250 mg/ml) and 0.05% trypsin (0.02% EDTA) were obtained from Gibco BRL (Gaithersburg, MD). Endothelial cell growth supplement (ECGS) was pur- chased from Collaborative Research, Inc. (Bedford, MA). Fetal bovine serum was obtained from Hyclone Laboratories, Inc. (Logan, UT). Heparin and gluta- mine were acquired from Sigma Chemical Co. (St. Louis, MO).

2.1. MTT cell viability assay

Cells were plated into 96 well tissue culture plates in a range of 3-4 X lo” cells/well in a final volume of 100 ~1. The cells were treated with varying doses of curcumin (0- 10 PM) for 6 and 24 h. The chromogenic methyl thiazol tetrazolium bromide (MTT) dye, an

indicator of metabolically active mass, was added to the cells and incubated for 3 h at 37°C. Cells were lysed and the reduced intracellular formazan product was dissolved in a solution with 10% SDS and 0.01 N HCl. The absorbance was recorded at 570 nm and percentage inhibition was plotted against untreated cells.

2.2. [‘Hlthymidine incorporation ussaj

Cells were plated and treated in triplicate with var- ious doses of curcumin for 6 and 24 h. In another experiment, cells were treated with curcumin for 12 h and were then washed and replaced with fresh media without curcumin for an additional 12 h. The cells in each well were pulse-labeled with 1 &i [methyl-“H]thymidine/well (specific activity, 6 1 Ci/ mmol), for 1 h at 37°C. The cells were harvested, and the radioactivity associated with individual sam- ples was measured in a liquid scintillation counter (Beckman, LS 6000 TA).

2.3. Flow cytometry and DNA histogram

Aliquots of 5 x 10” HUVEC cells were centrifuged at 1200 rpm for 10 min, and cell pellets were fixed with 70% ethanol overnight. The cells were then washed twice with PBS and resuspended in 1 ml of solution containing 3.4 mM sodium citrate, 5 mg/ml propidium iodide (Boehringer Mannheim) and 100 mg/ml RNase A (Boehringer Mannheim, Germany) and stored in the dark for 30 min. Cells were analyzed on a FACScan flow cytometer.

2.4. Thymidine kinase assay

The radioactive assay for analyzing TK activity is based on the conversion of [“H]dThd to [3H]dTMP and their separation on anion exchange filters. Cells were washed twice with cold PBS following 12 h treatment with curcumin. The cells were scraped off the plates and centrifuged at 100 x g for 10 min at 4°C. The pellets were then resuspended in 100 ~1 of PBS (5 x lo6 cells/ml) and freeze thawed three times by altering the tubes from dry ice/methanol to a luke- warm water bath. The samples were then centrifuged at 14 000 x g at 4°C for 60 min and the supematant was then assayed for TK activity.

A.K. Singh et al., I Carver Letters 107 (19Y6J 109-115 111

3. Results 3.2. Site(s) qf curcumin action in cell growth cycle

3.1. Eflect of curcumin on the viability and tritiated thymidine incorporation qf HUVEC

To establish the non-toxic doses of curcumin to HUVEC, MTT assays were performed. Data (figure- here>Fig. IA) show that curcumin (l-10 PM) did not significantly affect the viability of the endothelial cells. The incubation with curcumin for 6 h showed no significant inhibition in cell viability; however, 24 h incubation with curcumin showed slight inhibition in cell viability.

Treatment of HUVEC with curcumin for 6 and 24 h suppressed tritiated thymidine incorporation in a dose-dependent manner (Fig. 1B). There was almost 60% and 80% inhibition of tritiated thymidine incor- poration within 6 h and 24 h, respectively, whereas more than 90’% of the cells were viable at that time point, az: measured by MTT assay. In all the treat- ments, the cell viability was marginally inhibited as compared to the inhibition in DNA synthesis. How- ever, the extent of inhibition in [“Hlthymidine incor- poration was greater in 2% serum at both the 6 and 12 h period as compared to 10% serum (Fig. 2).

.$ 70- z 2 60. .f g so- 9 5 40- ::

p" 30-

20- T

T T .*. * lo- -1

.,... -I .-0 1 0

om ; I I I I , 0 1 2 3 4 5 6 7 8 9 10 11

Curcumin (PM)

A

The site of curcumin-mediated cell cycle block was investigated by [“Hlthymidine pulse labeling studies. Endothelial cells were synchronized by overgrowing and were then grown in 2% medium without ECGS. ECGS treatment of growth-arrested cells resulted in entry into S-phase after 12 h (Fig. 3). The data showed that addition of curcumin during cell cycle progres- sion, i.e. 6 and 9 h after stimulation, resulted in > 85% inhibition of DNA synthesis. Curcumin was able to block DNA synthesis (70%) when added at 12 h and 1.5 h during S-phase, demonstrating that curcumin affected cells which were actively under- going DNA synthesis.

3.3. Inhibition of endothelial cell proliferation by curcumin is independent of growth stimuli

To obtain an initial insight into the cellular site of curcumin action and rule out the potential effect of curcumin on extracellular action of fibroblast growth factor (FGF) or ECGS, we determined the effect of curcumin on HUVEC stimulated with FGF or ECGS. Data (Fig. 4) shows that curcumin effectively blocked

- 6h

4 IJ 1 2 3 4 5 6 7 8 9 10 11

Curcumin (PM)

B

Fig. I The effect of curcumin on cell viability and i3H]thymidine incorporation. Cells were plated into 96-well microtiter plates in a range of 3-4 x 105 cells/well in a final volume of 100 ~1 and were treated with various doses of curcumin. The chromogenic MTT assay was performed as described in Section 2. Exponentially growing cells were incubated for 6 or 24 h in medium containing 1, 5 and 10 PM of curcumin. [‘HlThymidine was added I h before the end point. Values represents mean of two different experiments performed in triplicate. (A) Inhibition in cell viability. (B) Inhibition in [3H]thymidine incorporation.

A.K. Sin& et al., I Cancer Letters 107 (1996) 1119-115

q 6h

0 24 h

I

I Curcumin (PM)

Fig. 2. The effect of curcumin on [‘Hlthymidine incorporation in 2% serum. Cells were plated in 24-well plates for 6 or 24 h in medium containing 2% serum and curcumin (I, 5 and 10 FM). Cells were labeled with [3H]thymidine for 60 min and cumulative DNA synthesis at the end of 6 and 24 h was determined. Values represents mean of two different experiments performed in tripli- catc.

HUVEC proliferation despite growth stimuli. These results indicate that the effect of curcumin was not caused by the inhibition of FGF/ECGS action at the cell surface.

2.4. Effect of’cwc~umin on the cell cycle distribution

The effect of curcumin on S-phase progression was evident as 12 h treatment of curcumin shows (Fig. 5A) approximately 46% cells in early S-phase compared to 22% cells in untreated control. Most of the curcu- min-treated cells were found at the boundary of Gl and S-phase, showing slower progression of cells through S-phase. Once the curcumin was replaced with fresh medium for an additional 12 h, a greater

Table I

Effect of curcumin on thymidine kinase activity

Treatment* (PM) % inhibition of TK activity

Curcumin I 41 Curcumin 5 53 Curcumm IO 12

*Exponentially growing cells were incubated without or with cur- cumin (l-10 PM). The activity of TK was determined from cell extracts. The results are expressed as the mean percentage of tri- plicate cultures.

7.5000

z

0 50000

25ooo

0 r- ” 3 6 9 12 15 IX 21 24 27

t t t t hours

Fig. 3. Effect of curcumin on ccl1 cycle from GO/G1 to S-phase: HUVEC were plated in 24.well plates and growth was arrested in GO/G1 by growing to the confluence and then subjecting it IO further incubation in ECGS-free medium for 24 h. The cells were then washed and stimulated with 0.2 mg/ml ECGS. The effect of curcumin on entry into S-phase was determined by 90 min pulse labeling with [“Hlthymidine for 3 h intervals in cells receiving curcumin at 0 (open diamond), 6 (open circle). 9 (open tnanglc). 12 (box with cross), and 15 h (diamond with cross). The dara is an average of three values. One set of cells were control without curcumin (open box).

number of the cells were seen in Gl and G2 phase indicating the slow reversal effect of the drug (Fig. 5B). Synchronized HUVEC were stimulated for 24 h in the presence of either ECGS or FGF. Addition of curcumin at 24 h after ECGS or FGF stimulation resulted in > 85% inhibition of DNA synthesis (Fig. 4). These results suggest that curcumin inhibited the entry of cells into S-phase.

3.5. Effect of curcumin on the activities of enzymes related to DNA synthesis

To elucidate the mechanism of cessation of [‘H]TdR incorporation in the curcumin-treated cells, we studied the changes in the activity of enzyme involved in DNA synthesis. There was a rapid and progressive decrease in TK activity, reaching up to 75% inhibition within 12 h of curcumin treatment

A.K. Singh et al.. 1 Cancer Letters 107 (1996) 109-115 II3

0 1 5 10

Curcumin (FM)

Fig. 4. HUVEC cell growth inhibition by curcumin in presence of FGF/EXXS. Cells were grown for 24 h and then stimulated either in growth medium containing 10 rig/ml FGF or in presence of ECGS (0.2 mg/ml). Cumulative DNA synthesis at the end of 24 h incubation was determined by [‘Hlthymidine incorporation.

(Table I), which correlates with the inhibition of pro- liferation by curcumin.

4. Discussion

Recent reports have shown that curcumin has an inhibitory effect on arachidonic acid-induced inflam- mation and on arachidonic acid metabolism through

125

100

=” s t 15

9 c g k 50

a

25

0

the inhibition of cyclooxygenase and lipooxygenase pathway in mouse epidermis [ 111. Recently, it has been shown that curcumin suppresses the protoonco- genes [ 121, the transcriptional factor c-jun/AP [ 131, and the proliferation of epidennal cells upon TPA induction [lo]. Curcumin has also been shown as selective inhibitor of phosphorylase kinase [ 141 and inhibits EGF-induced increase in EGF-R tyrosine phosphorylation in a dose-dependent manner [ 151. At present, very little is known about the precise mechanism of action of curcumin against chronic inflammation and tumor suppression.

Our studies have revealed a unique mode of action of curcumin on the endothelial cell proliferation and cell cycle progression. Curcumin inhibited the prolif- eration of HUVEC without changing the doubling time of the exponentially growing cells, implying that the effect of curcumin was simply not caused by delay in cell division. Growth inhibition was dose-dependent and was reversible upon the removal of the drug, thereby ruling out the curcumin-induced cellular toxicity. FGF is a potent mitogen for vascular and capillary endothelial cells in vitro [ 161. We found that endothelial cell growth inhibition was indepen- dent of the nature of the growth stimuli used. Growth inhibition of an ECGS and FGF-dependent endothe- liai cell line was not reversed by addition of ECGS or FGF, suggesting that growth inhibition by curcumin may be mediated through cell cycle events.

Our initial evidence suggests that cell growth inhi-

125

100

2 8 15

f s g 50 ,' a

25

0 I Fig. 5. Flow cytometric DNA histogram. The distribution of cells in various phases was analyzed by FACS. (A) The distribution of HUVEC in the cell cycle after 12 h exposure to curcumin. (B) HUVEC treated with curcumin for 12 h and then grown for additional 12 h in curcumin-free medium. The x-axis indicate the fluorescence intensity in arbitrary units representing relative cell number.

114 A.K. Singh et al., I Cancer Letters 107 (1996) 109-115

bition by curcumin is mediated partially through the cell cycle. The percentage of endothelial cells in S- phase increased upon curcumin treatment, although the flow cytometric DNA content shows most cells at Gl/S upon curcumin treatment. The number of cells in S-phase and inhibition in proliferation can possibly be explained by assuming these cells in S- phase were either slowed or arrested and were not actively synthesizing DNA when compared with untreated cells. The reduced S-phase distribution of adherent viable cells that occurs when cell loss is evident can be caused by S-phase cells that are speci- fically targeted by curcumin-induced retardation of entry into S-phase. However, when curcumin was replaced with fresh medium for additional time, a greater number of cells were at G2/M indicating slow reversal of the cells. These results could also imply that curcumin might be affecting the cells actively undergoing DNA synthesis. A further inves- tigation in this area is needed to show the effect of curcumin on S-phase cells from the reduction in DNA synthesis. Transforming growth factor (TGF-/3), another potent inhibitor of endothelial cells prolifera- tion in vitro, blocks the cells in late Gl phase [17,18]. Similar to tumor necrosis factor (TNF) and TGF-/3, curcumin had growth inhibitory effect on the estro- gen-dependent MCF-7 breast cancer cells (unpub- lished data).

Numerous reports have shown that as cells enter the S-phase, the activities of many enzymes involved in DNA synthesis also increase, including thymidylate synthetase (TS) and TK [19]. Reddy and others have also shown that both TS and TK were significantly inhibited by the inhibitors of DNA synthesis [20]. However, in our study there was a slight inhibition of TS activity (data not shown) with the significant loss of TK activity upon curcumin treatment that cor- relates with the significant inhibition of DNA synth- esis.

Formation of new blood vessels is a highly con- trolled process which involves the endothelial cell migration, proliferation and production of enzymes capable of modifying the extracellular matrix. Under physiological conditions, new blood vessels are formed during the body’s repair processes such as wound healing and embryonic development [21]. However, neovascularization is widely associated with a variety of pathologies, including tumor growth

[22], atherosclerosis [23], rheumatoid arthritis and inflammation [24]. The inhibition in proliferation and angiogenic differentiation of HUVEC on matrigel (paper communicated) by curcumin has shown many promises for the treatment of angiogenic disease. Gold salt and D-penicillamine inhibit neovasculariza- tion in vitro and in vivo [25,26]. Recently, methotrex- ate, which is a potent folate antagonist, was found to inhibit in vitro endothelial cell proliferation and in vivo neovascularization [27]. The unique action of curcumin on TK and cell cycle progression and further understanding of the mechanism of action on the endothelial cells could ultimately lead to the

development of curcumin as an anti-angiogenic drug.

Acknowledgements

The authors are grateful to Dr. Mark Moorman for helping with the flowcytometer. This work was supported by grant G174FQ from Naval Medical Research and Development Command. The opin- ions or assertions contained herein are the private views of the authors and should not be construed as official or necessarily reflecting the views of the Uniformed Services University of the Health Sciences or the Department of Defense.

References

Lll

PI

[31

141

[51

[61

171

Ammon, H.P. and W&l, M.A. (1991) Pharmacology ol’ Cur- cuma lon~a, Planta Med., 57, l-7. Srimal, R.C. and Dhawan, B.N. (1973) Pharmacology of diferuloyl methane(curcumin), a non-steroidal anti-inflamma- tory agent, J. Pharm. Pharmacol., 25, 447-452. Satoskar, R.R., Shah, S.J. and Shenoy, S.G. (1986) Evalua- tion of anti-inflammatory property of curcumin (diferuloyl methane) in patients with postoperative inflammation, Int. J. Clin. Pharmacol. Ther. Toxicol., 24, 651-654. Sharma, O.P. (1976) Antioxidant activity of curcumin and related compounds, Biochem. Pharmacol.. 25. IX I I ~- 1812. Toda, S., Miyase. T., Arichi, H., Tanizawa, H. and Takiyano, Y. (1985) Natural antioxidants, III: antioxidative componenta isolated from rhizome of Cuwumn /on,qr L, Chem Pharm. Bull., 33, 1725-1728. Huang, M.T., Smart, R.C.. Wong, G.Q. and Conney. A.H. (1988) Inhibitory effect of curcumin. chlorgenic acid, caffeic acid and ferulic acid on tumor promotion in mouse skin by 120tetradecanoyl phorbol-I 3-acetate. Cancer Re\., 4X. 5941-5946. Huang, M.T., Wang, Z.Y., Georgiadis, CA.. Laskin, J.D. and Conney, A.H. (1992) Inhibitory effects of curcumin on

A.K. Singh et al., I Cancer Letters 107 (1996) 109-115 115

tumor initiation by benzo(a)pyrene and 7.12.dimethyl benz(a)anthracene, Carcinogenesis, 13, 2183-2186.

[8] Azuine, M.A. and Bhide, S.V. (1992) Chemopreventive effect of turmeric against stomach and skin tumors induced by chemical carcinogens in Swiss mice, Nutr. Cancer, 17. 77-83.

[9] Huang, M.T., Lou, Y.R., Ma. W., Newark, H.L., Reuhl, K.R. and Conney, A.H. (1994) Inhibitory effect of dietary curcu- min on forestomach, duodenal, and colon carcinogenesis in mice. Cancer Res.. 54, 5841-5847.

[IO] Huang, H.C., Jan, T.R. and Yen, SF. (1992) Inhibitory effect of curcumin, an anti-inflammatory agent, on vascular smooth muscle cell proliferation, J. Pharmacol., 221, 381-384.

[ 111 Huang, M.T., Lysz, T., Ferraro, T., Abidi, T.F., Laskin, J.D. and Conney, A.H. ( 199 1) Inhibitory effects of curcumin on in vitro lipoxygenase and cyclooxygenase activities in mouse epidermis. Cancer Res., 5 1, 813-8 19.

[ 121 Kakar, S.S. and Roy, D. (1994) Curcumin inhibits TPA- induced expression of c-fos, c-jun and c-myc protooncogenes messenger RNAs in mouse skin, Cancer Lett., 87, 85-89.

[ 131 Huang, TX. Lee. S.C. and Lin, J.K. (1991) Suppression of c- Jun/AP-I activation by an inhibitor of tumor promotion in mouse tibroblasts cells, Proc. Nat]. Acad. Sci. USA, 88, 5292-5296.

1141 Reddp, S. and Aggarwal, B.B. (1994) Curcumin is a non- competitive and selective inhibitor of phosphorylase kinase. FEBS Len., 341, 19-22.

[ 151 Korutla. L., Cheung, J.Y., Mendelsohn, J. and Kumar, R. ( 1995) Inhibition of ligand-induced activation of epidennal growth factor receptor tyrosine kinase phosphorylation by curcumin, Carcinogenesis, 16. 1741-1745.

[ 161 Montesano, R., Vassalli, J.D., Baird, A., Guillemin, R. and Grci. L. (1986) Basic tibroblast growth factor induces angio- gene,is in vitro, Proc. Nat]. Acad. Sci. USA, 83, 7297-7301.

[ 171 Heimark. R.L., Twardzik. D.R. and Schwartz, S.M. (1986)

Inhibition of endothelial regeneration by type-beta transform ing growth factor, Science, 233, 1078-1080.

[IS] Shipley, G.D., Tucker, R.F. and Moses, H.L. (1985) Type beta transforming growth factor/growth inhibitor stimulates entry of monolayer cultures of AKR-2B cells into S-phase after prolonged prereplicative interval. Proc. Natl. Acad. Sci. USA, 82,4147-4151.

1191 Coppock, D.L. and Paradee, A.B. (1987) Control of thymi- dine kinase mRNA during the cell cycle, Mol. Cell. Biol., 198, 2925-2932.

[20] Reddy, G.P.V. and Pardee, A.B. (1983) Inhibitor evidence for allosteric interaction in the replitasc multienzyme complex, Nature, 304. 86-88.

[21] Madri, J.A., Bell, L., Marx, M., Mervin, J.R., Basson. C. and Prim. C. (1991) Effects of soluble factors and extracellular matrix components on vascular cell behaviour in vitro and in vivo: models of de-endothelitation and repair, J. Cell Biochem., 45, 123- 130.

[22] Folkman, J. (1990) What is the evidence that tumors are angiogenic dependent?, J. Natl. Cancer Inst., 82, 4-6.

[23] Ross, R. and Harker, L.(l976) Hyperlipidemia and athero- sclerosis. Science, 193. 1044-1049.

1241 Pober, J.S. and Cotran. R.S. (1990) The role of endothelial cells in infammation, Transplantation, 50, 537-544.

[25] Matsubara. T. and Ziff, M. (1987) Inhibition of human endothelial cell proliferation by gold compounds. J. Clin. Invest., 79, 1440-1446.

[26] Matsubara, T., Saura, R., Hirohata, K. and Ziff, M. (1989) Inhibition of human endothelial cell proliferation in vitro and neovascularization in vivo by u-penicillamine. J. Clin. Invest., 83. 1588167.

[27] Hirata, S., Matsubara, T., Saura, R., Tateishi, H. and Hirota, K. ( 1989) Inhibition of in vitro vascular endothelial cell pro- liferation and in vivo neovasculariaation by low dose methotrexate, Arthritis Rheum., 32, 1065-1072.