Justicidin A decreases the level of cytosolic Ku70 leading to apoptosis in human colorectal cancer cells.

(1)

Justicidin A decreases the level of cytosolic Ku70 leading to apoptosis in

human colorectal cancer cells

Jenq-Chang Lee, Chao-Hung Lee

2

, Chun-Li Su

3,y

,

Chung-Wei Huang

1

, Hsiao-Sheng Liu

1,y

, Chun-Nan Lin

4

and Shen-Jeu Won

1,

Department of Surgery and1Department of Microbiology, College of

Medicine, National Cheng Kung University, Tainan, Taiwan,2_{Department of}

Pathology and Laboratory Medicine, Indiana University School of Medicine,

Indianapolis, IN, USA,3Department of Nursing, Chang Jung Christian

University, Tainan, Taiwan and4School of Pharmacy, Kaohsiung Medical

University, Kaohsiung, Taiwan

_{To whom correspondence should be addressed. Tel:}_{þ886 6 2744435;}

Fax:þ886 6 2082705;

Email: a725@mail.ncku.edu.tw

The natural product justicidin A, an arylnaphthalide

lignan isolated from Justicia procumbens, significantly

inhibited the growth of human colorectal cancer cells

HT-29 and HCT 116 at day 6 post-treatment. Further

study revealed that justicidin A-treated HT-29 and HCT

116 colorectal cancer cells died of apoptosis. Justicidin A

treatment caused DNA fragmentation and an increase in

phosphatidylserine exposure of the cells. The number of

cells in the sub-G

1

phase was also increased upon justicidin

A treatment. Caspase-9 but not caspase-8 was activated,

suggesting that justicidin A treatment damaged

mitochon-dria. The mitochondrial membrane potential was altered

and cytochrome c and Smac were released from

mitochon-dria to the cytoplasm upon justicidin A treatment. The

level of Ku70 in the cytoplasm was decreased, but that of

Bax in mitochondria was increased by justicidin A. Since

Ku70 normally binds and sequesters Bax, these results

suggest that justicidin A decreases the level of Ku70 leading

to translocation of Bax from the cytosol to mitochondria to

induce apoptosis. Oral administration of justicidin A was

shown to suppress the growth of HT-29 cells transplanted

into NOD-SCID mice, suggesting chemotherapeutic

poten-tial of justicidin A on colorectal cancer cells.

Introduction

Colorectal cancer is a leading cause of death in both men and

women (1). Chemotherapeutic agents such as 5-fluorouracil

and leucovorin have been widely used postoperatively in

colo-rectal cancer patients with regional lymph node metastasis to

reduce the risk of distant metastasis (2). Oxaliplatin, irinotecan,

or a combination of these drugs with 5-fluorouracil have been

used to control distant metastasis (2). Unfortunately, the death

rate of patients with colorectal cancer remains very high (3).

Therefore, a continual search for potential anticancer agents is

necessary. The crude extract of the plant

Justicia procumbens

is commonly used in Taiwan to treat pain, fever and

inflam-mation (4,5). Recently, five 2,3-naphthalide lignans including

justicidin A, justicidin E, neojusticin A, neojusticin B and

diphyllin have been isolated from the plant (6). Among these,

justicidin A has been shown to have an anti-cancer activity

(7–9), but its mode of action is unknown. One possibility is

that justicidin A induces apoptosis.

Apoptosis is characterized by morphological changes, DNA

fragmentation, phosphatidylserine externalization, and

genera-tion of apoptotic bodies (10). Apoptosis can be induced by a

variety of stimuli including radiation, tumor necrosis factor,

and certain chemotherapeutic agents (11). Mitochondria are a

major target of chemotherapy-induced apoptosis in tumors

(10). Upon stimulation, death-promoting factors such as

cyto-chrome

c (cyto c), second mitochondria-derived activator of

caspase/direct IAP binding protein with low pI (Smac),

apop-tosis inducing factor (AIF) and endonuclease G are released

into the cytosol (12). Regulators such as Bcl-2 and Bcl-X

L

are

anti-apoptotic (13–15); they bind to mitochondria and inhibit

the release of cyto

c and Smac (16,17). Ku70 is a 70 kDa

subunit of the Ku complex, which plays a crucial role in DNA

double-strand break repair (18). Ku70 has been shown to

inhibit the translocation of the pro-apoptotic factor Bax from

cytosol to mitochondria by sequestering Bax (19,20).

In this study, we investigated effects of justicidin A on the

growth of two human colorectal cancer cell lines. Our results

revealed that justicidin A induces apoptosis by altering the

balance between Ku70 and pro- and anti-apoptotic Bcl-2

family members in the cell. We also demonstrated that

justicidin A is effective against colorectal cancer cells both

in vitro and in vivo.

Materials and methods

Isolation of justicidin A

Justicia procumbens plants were collected from Chu-Shan, Nantu Hsein, Taiwan, air-dried and then chipped. Justicidin A was isolated and purified

from the chippedJusticia procumbens plants, as described previously (9).

Cell lines and cell culture

Human colorectal cancer HT-29 and HCT 116, cervix carcinoma SiHa, breast adenocarcinoma MCF7, bladder carcinoma T24 and human embryonic kidney epithelial HEK293 cells were obtained from ATCC (Rockville, MD) and maintained in Dulbecco’s modified Eagle’s medium (DMEM; GIBCO BRL, Grand Island, NY) containing 10% heat-inactivated fetal calf serum (FCS;

HyClone, Logan, UT), 2 mML-glutamine, 100 U/ml penicillin, and 100 mg/ml

streptomycin. The cells were incubated in a humidified atmosphere with 5%

CO2 at 37C. Human peripheral blood mononuclear cells (PBMCs) were

isolated from freshly collected buffy coat fraction of whole blood obtained from the Tainan Blood Bank Center (Tainan City, Taiwan) by Ficoll-Paque (Farmacia, Uppsala, Sweden) density gradient, as described previously (21). The isolated PBMCs were washed three times with serum-free RPMI-1640 medium (GIBCO BRL) and then resuspended in complete DMEM culture medium (GIBCO BRL) containing antibiotics.

Abbreviations: cyto c, cytochrome c; DFF, DNA fragmentation factor;

Dcm, mitochondrial membrane potential; MTT,

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; PARP, poly(ADP-ribose) polymerase; PBMCs, human peripheral blood mononuclear cells; PI, propidium iodide; RACK1, receptors for activated C-kinase; s.c., subcutaneously; Smac, second mitochondria-derived activator of caspase/direct IAP binding protein with low pI; XIAP, X-linked apoptosis-inhibiting protein.

y_{These authors contributed equally to this work.}

Advance Access publication May 19, 2005

at National Taiwan Normal University on January 15, 2013

http://carcin.oxfordjournals.org/

(2)

Animal experiments

NOD.CB17-PRKDC5SCID4/J (NOD-SCID) mice were obtained from the Animal Center of the National Cheng Kung University (NCKU, Tainan, Taiwan). They were bred and housed at the Animal Center in a pathogen-free, temperature-controlled and air-conditioned environment with a 10/14 h light/dark cycle. Mice, 6–7 weeks old, were used. Food and water were

providedad libitum. All animal experiments were approved by the Animal

Research Committee of NCKU and were performed under the guidelines of the National Research Council, Taiwan. Tumor cells were implanted sub-cutaneously (s.c.) to the flank of mice. Tumor growth was measured with a caliper 2–3 times per week. Tumor volume was calculated by using the

formula (L W2

)/2, whereL (length) and W (width) are in millimeters and

L 4 W (22). Reagents

Most chemicals were obtained from Sigma Chemical Co. (St Louis, MO) unless otherwise indicated. The caspase-9 inhibitor Z-LEHD-fmk was pur-chased from Santa Cruz Biotech (Santa Cruz, CA). Glycine and protein assay reagents were obtained from Bio-Rad Laboratories (Hercules, CA). Polyviny-lidene fluoride membrane for western blot was purchased from Millipore

(Bedford, MA). DiOC6(3) was purchased from Molecular Probes (Eugene,

OR). Antibodies against various proteins were acquired from the following vendors: anti-caspase-3 mouse monoclonal and anti-Smac rabbit polyclonal antibodies, IMGENEX (San Diego, CA); anti-caspase-9 mouse monoclonal

antibody, Upstate (Lake Placid, NY); anti-cytoc mouse monoclonal antibody,

BD Pharmingen (San Diego, CA); anti-X-linked apoptosis-inhibiting protein (anti-XIAP) and anti-receptors for activated C-kinase (anti-RACK1) mouse monoclonal antibodies, BD Transduction Laboratories (Lexington, KY); anti-poly(ADP-ribose) polymerase (anti-PARP) rabbit polyclonal antibody, Cell

Signaling (Beverly, MA); anti-Ku70, anti-Bcl-2, anti-Bax and anti-Bcl-XL

mouse monoclonal antibodies, anti-caspase-8, Anti-DNA fragmentation factor (DFF)-45 and anti-DFF-40 rabbit polyclonal antibodies and goat anti-mouse conjugated HRP secondary antibody, Santa Cruz Biotech; goat anti-rabbit

conjugated HRP secondary antibody, Amersham Pharmacia Biotech

(Piscataway, NJ). Chemiluminescence reagents were obtained from NEN Life Science (Boston, MA).

Cell viability assay

Cytotoxicity of justicidin A was determined by a modified MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] colorimetric assay,

as described previously (23,24). Briefly, cells (1 103

cells/well) were grown overnight in 96-well plates (Nunc, Denmark) in 100 ml culture medium and then treated with different concentrations of justicidin A dissolved in 100% DMSO. Control cells were treated with DMEM alone or DMEM con-taining 0.1% DMSO. At different time points, 10 ml of MTT was added to each well to a final concentration of 0.5 mg/ml. The water-soluble MTT was taken up by live cells and converted to an insoluble purple formazan. After 4 h of

incubation at 37_{C in the dark, 100 ml of 10% SDS/0.01 N HCl was added to}

each well and incubated at 37C overnight to dissolve the formazan. The

amount of formazan in each sample thus produced was quantified by measur-ing the absorbance of light at 590 nm with a Multiscan photometer (MRX II, Dynatech, McLean, VA). All experiments were repeated three times and each drug concentration was tested in octuplicate. Viability of drug-treated cells was expressed as a percentage of population growth with standard error of the mean (SEM) relative to that of untreated control cells. Cell death caused by justicidin

A was calculated as a percentage of inhibition as follows: % inhibition¼

(1 mean experimental absorbance/mean control absorbance) 100.

Colony formation assay

Cells (6 104_{cells/well) in 0.1 ml medium were mixed with 0.9 ml of 0.33%}

agar in DMEM (GIBCO BRL) containing 10% FCS (HyClone) in the presence

or absence of various concentrations of justicidin A at 37_{C and then layered}

on top of 1 ml of 0.6% solid basal agar in six-well trays (Nunc). Colonies with a diameter 41 mm were counted 14 days later (25).

Analysis of DNA fragmentation

Justicidin A-treated or untreated cells were lysed by incubation in 200 ml lysis buffer [10 mM Tris–HCl (pH 7.6), 20 mM EDTA and 1% NP-40] for 20 min at

37C. The cell lysates were centrifuged to remove cell debris, and the

super-natants were incubated with 4 mg/ml of RNase A and 1% SDS at 56_{C for 2 h,}

followed by incubation with 200 mg/ml proteinase K at 37oC for 2 h. DNA

fragments were precipitated by the addition of 150 ml ammonium acetate

(10 M) and 1.2 ml ethanol (100%). After an overnight incubation at20_C,

the DNA was pelleted, dried and then dissolved in 15 ml Tris–EDTA buffer. The sample was then electrophoresed on a 1% (w/v) agarose gel in TBE buffer at 50 V for 1 h, and the gel was stained with ethidium bromide to visualize DNA ladders under UV light (26).

Determination of apoptosis by flow cytometry

Apoptotic cells with externalized phosphatidylserine were enumerated using a FACScan flow cytometer (Becton Dickinson, Mountain View, CA). Briefly,

cells (2 105_{cells/well) in 2 ml medium were grown in six-well plates and}

then treated with different concentrations of justicidin A. At different time intervals, cells were trypsinized, washed with HEPES buffer solution (HBS), resuspended in 400 ml HBS containing 5 ml Annexin V-FITC (BD Pharmin-gen), and then subjected to flow cytometric analysis as described previously (27). To assess apoptosis by measuring DNA contents, justicidin A treated

cells were fixed with 70% ethanol at 4_{C and then stained with propidium}

iodide (PI) by incubating the cells in HBS containing 40 mg/ml PI and

100 mg/ml RNase A for 30 min at 37_{C in the dark. The PI-stained cells}

were sorted by flow cytometry based on their DNA contents. Results were analyzed with the Windows Multiple Document Interface software for Flow Cytometry (WinMDI 2.8, Scripps Research Institute, San Diego, CA). Another

event in apoptosis is change in Dcm. To assess Dcm, justicidin A treated

and non-treated cells were incubated with 40 nM of the lipophilic cationic

dye DiOC6(3) and 2.5 mg/ml PI in the dark for 30 min. Since the binding

of DiOC6(3) to cells varies with the Dcm, the intensity of DiOC6(3) staining

is directly proportional to the integrity of mitochondrial membrane.

DiOC6(3) stained cells were analyzed by flow cytometry with an excitation

wavelength of 488 nm and emission wavelength of 529 nm as described previously (28,29).

Confocal microscopy

Measurement of Dcmwas also performed by confocal microscopy as described

previously (30). Briefly, cells (2 105_{cells/well) in 2 ml of medium were}

grown in six-well plates containing sterilized glass cover slips overnight, treated with or without various concentrations of justicidin A, and then stained

with 5 mM rhodamine 123 at 37C. Rhodamine 123 is a lipophilic,

positively-charged, membrane-permeable dye. It can be taken up by living cells and preferentially stains mitochondria with increasing intensity proportional to the integrity of mitochondria (31). After washing twice with PBS at room temper-ature to remove excess rhodamine 123, cells were fixed for 15 min in 4%

paraformaldehyde and then mounted on poly-L-lysine-coated glass

micro-scope slides. Samples were examined with a Leica TCSNT laser scanning confocal imaging system coupled to a Leica DMRBE microscope with a 630 fluotar objective. Rhodamine 123-stained cells were excited with 488-nm lines of a 25 mW laser, and the fluorescence emitted was visualized with a BF530/30 filter combination. Optical sections close to the middle of the cell were evaluated.

Preparation of subcellular fractions and western blot analysis

Whole cell lysate, cytoplasmic, mitochondrial and nuclear fractions of cells were separately analysed. Whole cell lysate was obtained by

resuspend-ing cells (1 106_{cells) in 200 ml of lysis buffer containing 1 mM EDTA,}

0.5% (w/v) SDS, 10 mM Tris–HCl (pH 7.4), 0.15 M NaCl, 1 mM EGTA,

5 mg/ml leupeptin, 5 mg/ml aprotinin, 2 mM Na3VO4, 0.5 mM

phenylmethyl-sulfonyl fluoride (PMSF) and 1% (v/v) Triton X-100 at 4C for 30 min.

The cell debris was removed by centrifugation at 15 000g for 10 min. To

prepare cytoplasmic fractions, 1 107

cells were washed with ice-cold PBS and then homogenized with a Dounce homogenizer (Glas-Col, Terre Haute, IN) in 500 ml of TSE buffer [10 mM Tris, 0.25 M sucrose, 0.1 mM EDTA

(pH 7.4)]. After removal of cell debris by centrifugation at 750g for 30 min,

the samples were centrifuged at 100 000g for 1 h. The resulting supernatants

were used as the cytosolic fractions, and the pellets were lysed in 100 ml lysis buffer containing 10 mM Tris (pH 7.4), 1 mM EDTA, 10 mM NaF, 1 mM

Na3VO4, 2.5 mM PMSF, 1 mg/ml leupeptin, 1 mg/ml pepstatin A and 0.25 mM

sucrose and used as the mitochondrial fraction as described previously (32,33).

To prepare nuclear extracts, 1 107

cells were washed twice with ice-cold PBS and then lysed in 400 ml of buffer A [10 mM HEPES (pH 7.9), 5 mM

MgCl2, 10 mM KCl, 3 mM Na3VO4, 10 mM NaF, 0.5 mM dithiothreitol

(DTT), 0.5 mM PMSF, and 2 mg/ml of leupeptin, antipain, aprotinin, and pepstatin A] on ice for 20 min. The nuclei were pelleted by centrifugation at

11 000g for 20 s at 4C and then resuspended in 60 ml of buffer B (20 mM

HEPES, pH 7.9, 1.5 mM MgC12, 420 mM NaCl, 0.2 mM EDTA, 25%

glycerol, 1 mM Na3VO4, 10 mM NaF, 0.5 mM DTT, 0.5 mM PMSF and

1 mg/ml each of leupeptin, antipain, aprotinin, and pepstatin A) for 15 min on ice with occasional mixing. Nuclear debris was removed by centrifugation at

12 000g for 15 min at 4_{C. Western blotting was performed as described}

previously (34). Statistical analysis

Significant difference in tumor volume was determined by the Student’s t-test using the Minitab (version 10.2) software package. A difference was

considered significant ifP 5 0.05.

(3)

Results

Growth inhibition of human colorectal cancer cells by

justicidin A

The inhibitory activity of justicidin A on the proliferation of

human colorectal cancer cells (HT-29 and HCT 116), cervix

carcinoma cells (SiHa), breast adenocarcinoma cells (MCF7)

and bladder carcinoma cells (T24) was investigated. Cells

were grown in the absence or presence of various

concentra-tions (0.001–40 mM) of justicidin A for 6 days. MTT assays

were then performed, and the 50% inhibitory concentrations

(IC

50

) for HT-29, HCT 116, SiHa, MCF7 and T24 cells were

determined to be 0.110, 0.400, 0.020, 1.540 and 0.004 mM,

respectively (Table I). The effect of justicidin A on colony

formation of these cancer cells in soft agar was also assessed.

At day 14 post-treatment, justicidin A suppressed colony

formation of HT-29, HCT 116, SiHa, MCF7 and T24 cells

by 50% at concentrations of 0.030, 0.100, 0.080, 1.060 and

0.003 mM, respectively (Table I). The IC

50

of justicidin A for

human embryonic kidney cells HEK293 was

100-fold higher

(IC

50

¼ 10.6 mM) than that for HT-29 cells. PBMCs were

much more resistant to justicidin A with an IC

50

of 25 mM.

These results suggest that justicidin A preferentially inhibits

the growth of cancer cells. Since colorectal cancers are one of

the most common cancers, subsequent studies were focused on

HT-29 and HCT 116 cells. A daily evaluation of justicidin A

on growth inhibition of these cancer cells revealed that

justi-cidin A inhibited cell growth in a dosage- and time-dependent

manner. At a concentration of 2.5 mM, it inhibited the

prolif-eration of HT-29 cells

25% at day 1 and 82% at day 6

post-treatment (Figure 1A). A similar pattern of growth inhibition

was observed in HCT 116 cells (Figure 1B). However, lower

concentration (0.625 mM) of justicidin A had a better cytotoxic

effect on HT-29 (Figure 1A) than HCT 116 (Figure 1B).

Induction of apoptosis in colorectal cancer cells by

justicidin A

To determine whether justicidin A induced apoptosis, cells

were treated with the drug (0.75 mM for HT-29 and 5 mM

for HCT 116) and then assayed for signs of apoptosis at 0, 12,

16, 20, 24, 48 and 72 h post-treatment. The treated cells were

stained with annexin V-FITC and then analysed by flow

cyto-metry. As seen in Figure 2A, the percentage of annexin V

positive HT-29 cells was increased with time from 2.8%

at time 0 to 6.3% at 12 h and reached 75.4% at 72 h after

treatment of justicidin A (0.75 mM). A similar pattern of

increase (2.5% at time 0 to 10.4% at 12 h, and 78.2% at 72 h

post-treatment) in annexin V positive HCT 116 cells was

observed when they were treated with 5 mM of justicidin A.

To confirm apoptosis, cells were stained with PI, and the

relative number of cells with a sub-G

1

DNA content was

determined by flow cytometry. As seen in Figure 2B, treatment

of HT-29 cells with 0.75 mM of justicidin A increased the

population of cells with a sub-G

1

DNA content from 1%

at time 0 to 37.2% at the 72 h time point. The population

of sub-G

1

HCT 116 cells increased with time from 1.1% at

time 0 to 19.8% at 96 h after treatment of 5 mM of justicidin A

(Figure 2C). This increase in the population of sub-G

1

cells

was also dosage dependent, ranging from 2.2% in non-treated

to 9.5% in those treated with 0.25 mM of justicidin A and to

14.5, 17.7 and 19.2% in those treated with 0.5, 0.75 and 1 mM,

respectively, for 48 h (Figure 2B). Treatment of HCT 116 cells

with 0–10 mM of justicidin A showed a similar

dosage-dependent response pattern (Figure 2C). The population

of sub-G

1

HCT 116 cells was increased from 2.4% in

Table I. Effects of justicidin A response for 50% growth inhibition (IC50),

colony formation on human tumor cell lines, and human PBMCsa

Cell line Growth

inhibition (mM) Clonogenic cell inhibition (mM) HT-29 0.110 0.100 0.030 0.020 HCT 116 0.400 0.120 0.100 0.030 SiHa 0.020 0.090 0.080 0.090 MCF7 1.540 0.100 1.060 0.050 T24 0.004 0.020 0.003 0.070 HEK293 10.600 1.100 PBMC 25.000 0.980 a HT-29, HCT 116, SiHa, MCF7, T24, HEK293 (1 103 cells/well) or PBMC (3 104

cells/well) cells were treated with justicidin A (ranging from 0.001 to 40 mM). After 6 days of incubation, growth inhibition was evaluated by MTT assay as described in Materials and methods. HT-29, HCT 116, SiHa, MCF7

and T24 (6 104_{cells/well) cells were cultured in agarose. After 14 days of}

incubation, the colonies were counted. Justicidin A was diluted with culture medium containing 0.1% DMSO. The control cells were treated with medium containing 0.1% DMSO. Perc en ta ge of inh ibit ion 0 1 2 3 4 5 0 20 40 60 80 100 2.500 M 0.625 M 0.156 M 0.039 M

A

Perc en ta ge of inh ibit ion Time (Days) 0 1 2 3 4 5 6 7 0 20 40 60 80 100 2.500 M 0.625 M 0.156 M 0.039 M 2.500 M 0.625 M 0.156 M 0.039 M µ µ µ µ 0 1 2 3 4 5 6 7 0 20 40 60 80 100 5.000 M 2.500 M 1.250 M 0.625 M Percentage of i nhibi tio n 0 1 2 3 4 5 6 7 0 20 40 60 80 100 Time (Days) 5.000 M 2.500 M 1.250 M 0.625 M 5.000 M 2.500 M 1.250 M 0.625 M µ µ µ µ Percentage of i nhibi tio n

B

Fig. 1. Growth inhibition of justicidin A on colorectal cancer cells.

(A) HT-29 or (B) HCT 116 cells (1 103

cells/well) in 0.1 ml medium were grown in 96-well plates and then treated with justicidin A at the indicated concentrations for various lengths of time. Growth inhibition was determined by the MTT assay every day for 6 days.

(4)

non-treated to 20% in those treated with 10 mM of justicidin A

for 72 h (Figure 2C). To further confirm that justicidin

A-treated cells undergo apoptosis, DNA fragmentation in these

cells was examined and all drug-treated cells were found to

have a certain degree of DNA fragmentation (Figure 2D).

Effects on caspases and its target proteins by justicidin A

Since the major enzyme involved in apoptosis is caspase-3,

activation of caspase-3 in justicidin A-treated cells was

examined. HT-29 and HCT 116 cells were treated with

0.75 and 5 mM of justicidin A, respectively. At different time

points after treatment, cells were assessed for the occurrence

of activated caspase-3 by western blotting. Both the 12 and

17 kDa forms of activated caspase-3 were first detected 24 h

after justicidin A treatment in both cell lines, and more

activated caspase-3 was detected with a longer drug treatment

(Figure 3A).

To determine whether justicidin A-induced apoptosis was

mediated by the extrinsic or intrinsic pathway, activation of

caspase-8 or caspase-9 was examined. Caspase-9 activation

would implicate an intrinsic pathway, whereas caspase-8

activation would suggest the extrinsic pathway. HT-29 and

HCT 116 cells were again treated with 0.75 and 5 mM

of justicidin A, respectively. If the caspase-9 is activated,

0 ₁₂

₁₆

₂₀

₂₄

₄₈

₇₂

_{time (h)}

HT-29

HCT 116

36.0 1.7% 59.5 1.9% 10.4 0.6% 128 2.5 0.3% 6.3 0.7% 8.3 1.3% 128 2.8 0.5% 78.2 2.9% 25.5 0.9% 20.1 0.8% 7.9 1.1% 24.0 1.7% 47.8 2.2% 75.4 3.2%

A

0 ₁₂

₁₆

₂₀

₂₄

₄₈

₇₂

_{time (h)}

HT-29

Fluorescence intensity

HCT 116

36.0 1.7% 36.0 ± 1.7% 59.5 1.9%59.5±1.9% 10.4 0.6% 10.4±0.6% 128 2.5±0.3% 6.3 0.7% 6.3±0.7% 8.3 1.3%8.3±1.3% 128 2.8 0.5% 2.8±0.5% 78.2 2.9% 78.2±2.9% 25.5 0.9% 25.5 0.9% 25.5±0.9% 20.1 0.8% 20.1 0.8% 20.1±0.8% 7.9 1.1% 7.9 1.1% 7.9±1.1% 24.0 1.7%24.0 1.7%24.0±1.7% 47.847.8 2.2%47.8 2.2%±2.2% 75.4 3.2%75.4 3.2%75.4±3.2%

Cell number

14.5 1.6% 0 200 400 0 19.2 0 .1% 0 200 400 0 1. 2 0 .3% 0 200 400 0 1.4 0 .1% 0 200 400 0 15.4 0.1% 0 200 400 0

HT-29

200 400 1.0 1 .0% 0 64 0 2.3 0.5% 0 200 400 0 0 6 12 24 48 72 time (h) 0 200 400 0 37.2 0 .6% 64 2. 2 1.0% 200 400 0 0 17.7 3.6% 0 200 400 0 0.50 1.00 ( 0 0.25 0.75 9.5 0.6% 0 200 400 0

B

Relative cell number

14.5 1.6% 0 200 400 0 14.5 ± 1.6% 0 200 400 0 19.2 0 .1% 0 200 400 0 19.2 ± 0.1% 0 200 400 0 1. 2 0 .3% 0 200 400 0 1. 2 ± 0.3% 0 200 400 0 1.4 ± 0.1% 0 200 400 0 15.4 0.1% 0 200 400 0 15.4 ± 0.1% 0 200 400 0

HT-29

200 400 1.0 1 .0% 0 64 0 200 400 1.0 ± 1.0% 0 64 0 2.3 ± 0.5% 0 200 400 0 0 6 12 24 48 72 time (h) 0 6 12 24 48 72 time (h) 0 200 400 0 37.2 0 .6% 0 200 400 0 37.2 ± 0. 6% 64 2. 2 1.0% 200 400 0 0 2. 2 ± 1.0% 200 400 0 0 17.7 3.6% 0 200 400 0 17.7 ± 3.6% 0 200 400 0 0.50 1.00 ( 0 0.25 0.50 0.75 1.00 (µM) 0 0.25 0.75 9.5 ± 0.6% 0 200 400 0

Fig. 2. Justicidin A-induced apoptosis in colorectal cancer cells. HT-29 or HCT 116 cells (2 105_{cells/well) in 2 ml medium were grown in 6-well dishes}

and then treated with justicidin A (0.75 mM for HT-29 cells and 5 mM for HCT 116 cells) for the indicated time periods, or with indicated concentrations of justicidin A for 48 h (HT-29 cells) or 72 h (HCT 116 cells). Cells were then harvested and stained with annexin V-FITC (A) or PI (B and C) and analysed by flow

cytometry. In (A), the majority of non-treated cells were normal with no or very weak annexin V-FITC staining (intensity units 55 101

). When the cells were

exposed to justicidin A, the number of annexin V-FITC-stained cells with a staining intensity ranging from 5 101

to 3 103

units was increased with time.

In (B and C), cells that had a PI-staining intensity200 units were in the G1phase, and those with an intensity400 were in the G2–M phase. Cells with

a PI-staining intensity 5200 units were in sub-G1phase and were apoptotic. The number of apoptotic HT-29 (B) or HCT 116 (C) cells was increased in a

time- and dose-dependent manner upon justicidin A treatment. DNA fragmentation of justicidin A treated cells are shown in panel (D). Total DNA of vehicle- (0.1% DMSO in culture medium) or justicidin A-treated cells was extracted and electrophoresed on a 1% agarose gel. Lanes 1 and 7: molecular weight marker (M); lane 2: DNA from vehicle-treated HT-29 cells; lane 8: DNA from vehicle-treated HCT 116 cells; lanes 3–6 and 9–12: DNA from HT-29 or HCT 116 cells treated with indicated concentrations of justicidin A. Justicidin A was diluted with culture medium containing 0.1% DMSO. Results are representative of three independent experiments. JA, justicidin A.

(5)

the pro-caspase-9 is cleaved to yield two peptides of 35 and

37 kDa. As shown in Figure 3B, these two activated forms

became visible 6 h after treatment in HT-29 cells and 12 h in

HCT 116 cells and reached the highest level at the 72 h time

point (Figure 3B). In contrast, activated caspase-8 was not

detected in justicidin A-treated HT-29 or HCT 116 cells (data

not shown). To confirm the involvement of caspase-9 in

justi-cidin A-induced apoptosis, the caspase-9 inhibitor

Z-LEHD-fmk was used. HT-29 and HCT 116 cells were treated with

0.75 and 5 mM of justicidin A, respectively, with or without

pretreatment with 20 mM of Z-LEHD-fmk for 4 h. The cells

were then stained with PI and assayed for those arrested in the

sub-G

1

phase by flow cytometry. As seen in Figure 3C, only

3.8% of HT-29 cells were found to undergo apoptosis without

justicidin A treatment. The population of apoptotic HT-29

cells was increased to 35.3% upon justicidin A treatment.

Pre-treatment of the cells with Z-LEHD-fmk reduced

justi-cidin A-induced apoptosis rate to 8.8%. Z-LEHD-fmk

treat-ment alone had no effect on the apoptosis of HT-29 cells

(Figure 3C). Similar results were observed in HCT 116 cells;

Z-LEHD-fmk was found to reduce the apoptosis rate caused by

justicidin A from 25.5 to 6.5% in HCT 116 cells (Figure 3C).

Since activated caspase-3 cleaves PARP, experiments were

performed to detect the cleaved form of PARP which has a

molecular weight of 89 kDa. The cleaved PARP was first

detected in both total cell lysate and nuclear fraction at 12 h

after justicidin A treatment in both HT-29 and HCT 116 cells,

and the amount of the 89 kDa PARP was increased in a

time-dependent manner (Figure 3D). Activated caspase-3 also

digests DFF-45 and DFF-35, both of which bind to DFF-40

and prevent it from entering the nucleus to cause

fragmenta-tion of nuclear DNA (35,36). Therefore, reduced levels of

DFF-45 and DFF-35 are an indication of caspase-3 activation.

As shown in Figure 3E, treatment of HT-29 cells with 0.75 mM

of justicidin A caused a decrease in DFF-45 and DFF-35 levels

at 12 h. The cytosolic levels of DFF-45 and DFF-35 continued

to decrease and reached the lowest level at 72 h. In HCT 116

cells, the decrease in the level of DFF-45 was not as profound

1 2 3 4 5 M C 500. JA

(

M

)

lane 6 0. 75 1. 00 0. 25

HT-29

7 8 M C 10 2. 50 11 3. 75 12 5. 00 9 1. 25 JA ( M)

HCT 116

1 2 3 4 5 M C 500. JA

(

M

)

lane 6 0. 75 1. 00 0. 25

HT-29

7 8 M C 10 2. 50 11 3. 75 12 5. 00 9 1. 25 JA ( M)

HCT 116

D

1 2 3 4 5 M C 500. JA

(

µ

M

)

lane 6 0. 75 1. 00 0. 25

HT-29

7 8 M C 10 2. 50 11 3. 75 12 5. 00 9 1. 25 JA (

µ

M)

HCT 116

Relative cell number

0 2.5 5.0 2.4 0.5% 200 400 64 10.0 1.2% 200 400 0 13.1 0.8% 200 400 20.0 0.6% 0 200 400 0 0 0 0 0 0 1.1 0.3% 64 200 400 1.0 0.9% 200 400 0 3.2 0.4% 0 200 400 0 9.8 0.5% 200 400 19.8 1. 1% 200 400 0 0 0 0 0 0 0

HCT 116

48 time (h) 0 24 96

C

Fluorescence intensity

0 2.5 5.0 0 2.5 5.0 10.0 (µM) 2.4 ± 0.5% 200 400 64 10.0 ± 1.2% 200 400 0 13.1 ± 0.8% 200 400 20.0 ± 0.6% 0 200 400 0 0 0 0 0 0 1.1 ± 0.3% 64 200 400 1.0 ± 0.9% 200 400 0 3.2 ± 0.4% 0 200 400 0 9.8 ± 0.5% 200 400 19.8 ± 1. 1% 200 400 0 0 0 0 0 0 0

HCT 116

48 time (h) 0 24 48 96 time (h) 0 24 96 72 Fig. 2. Continued.

(6)

kDa JA (0.75 M) 0 6 12 24 48 72 time (h)

HT-29

47 37 35 Casp-9 RACK1 Cleaved Casp-9 47 37 35 32 17 12 Casp-3 RACK1 Cleaved Casp-3 JA (5 M) 0 12 24 48 72 96 kDa

HCT 116

32 17 12 kDa JA (0.75 M) 0 6 12 24 48 72 time (h)

HT-29

kDa JA (0.75 µM) 0 6 12 24 48 72 time (h)

HT-29

47 37 35 Casp-9 RACK1 Cleaved Casp-9 47 37 35 47 37 35 47 47 37 37 35 35 Casp-9 RACK1 RACK1 Cleaved Casp-9 Cleaved Casp-9 47 37 35 47 47 37 37 35 35 32 17 12 32 32 17 17 12 12

B

A

Casp-3 RACK1 RACK1 Cleaved Casp-3 Cleaved Casp-3 JA (5 M) 0 12 24 48 72 96 kDa

HCT 116

32 17 12 JA (5 µM) 0 12 24 48 72 96 kDa

HCT 116

32 32 17 12 17 17 12 12

C

0 200 400 0 64 0 0 200 400 0 0 200 400 0 200 400 0 JA Z-LEHD+JA C Z-LEHD

HCT 116

0 0 200 400 0 0 200 400 0 0 200 400 64 0 200 400 0

Relativ

e

cell number

0 200 400 0 0 200 400 0 0 200 400 0 200 400 0 JA Z-LEHD+JA C Z-LEHD

HCT 116

Fluorescence intensity

0 200 400 0 0 200 400 0 0 200 400 0 200 400 0 JA Z-LEHD+JA C Z-LEHD JA Z-LEHD+JA C Z-LEHD

HCT 116

0 0 200 400 0 0 200 400 0 0 200 400 64 0 200 400 0 0 0 200 400 0 0 200 400 0 0 200 400 0 0 200 400 64 0 200 400 0 3.8 ± 0.2% 4.4 ± 0.4% 35.3 ± 1.5% 8.8 ± 0.1% 6.5 ± 0.2% 25.5 ± 1.2% 6.2 ± 0.4% 1.4 ± 0.3%

D

Total cleaved PARP JA (0.75 M) 0 6 12 24 48 72 time (h) kDa

HT-29

Nuclear cleaved PARP RACK1 112 89 112 89

E

45 35 RACK1 Cytosolic DFF-45 Nuclear DFF-40 40 40 45 35 0 12 24 48 72 JA (5 M) 96 kDa

HCT 116

112 89 112 89 Total cleaved PARP JA (0.75 M) 0 6 12 24 48 72 time (h) kDa

HT-29

JA (0.75 µM) µ 0 6 12 24 48 72 time (h) kDa

HT-29

Nuclear cleaved PARP RACK1 RACK1 112 89 112 112 89 89 112 89 112 112 89 89 45 35 45 45 35 35 RACK1 RACK1 Cytosolic DFF-45 Cytosolic DFF-45 Nuclear DFF-40 Nuclear DFF-40 4040 4040 45 35 45 45 35 35 0 12 24 48 72 JA (5 M) 96 kDa

HCT 116

112 89 112 89 0 12 24 48 72 JA (5 M) 96 kDa

HCT 116

0 12 24 48 72 JA (5 M) 96 kDa

HCT 116

112 89 112 112 89 89 112 89 112 112 89 89 Z-LEHD JA Z-LEHD+JA C Z-LEHD JA Z-LEHD+JA C Z-LEHD JA Z-LEHD+JA C

HT-29

(7)

as that seen in HT-29 cells, but a dramatic decrease in DFF-35

level was observed 72 h after treatment with 5 mM of justicidin

A (Figure 3E). While the levels of cytosolic DFF-45 and

DFF-35 were decreased, the level of DFF-40 in the nuclear

fraction of justicidin A-treated cells was increased in a time

dependent manner. The increase in the level of nuclear DFF-40

was first observed at 6 h in HT-29 cells and 12 h in HCT 116

cells and peaked at 72 h in HT 29 cells and at 24–48 h in HCT

116 cells upon justicidin A treatment (Figure 3E).

Damage in mitochondria caused by justicidin A treatment

Activation of caspase-9 suggests that justicidin A triggers

apoptosis by the intrinsic pathway which affects mitochondria.

Therefore, the change in the membrane potential of

mitochon-dria (Dc

m

) in response to justicidin A treatment was examined

by flow cytometry after staining the cells with mitochondrial

dye DiOC

6

(3) and PI. A time-dependent decrease in the

intens-ity of DiOC

6

(3) staining was observed in the mitochondria of

justicidin A-treated HT-29 (0.75 mM justicidin A) and HCT

116 (5 mM justicidin A) cells. The decrease in Dc

m

was first

observed at 12 h and reached the lowest level at 72 h after

justicidin A treatment in both cell lines (Figure 4A). Treatment

of HT-29 cells with various concentrations of justicidin A

(0.25–1 mM) for 48 h resulted in a dose-dependent loss of

Dc

m

(Figure 4A). The decrease in the intensity of DiOC

6

(3)

staining was found to be accompanied by an increase in PI

staining (Figure 4A). Since PI does not penetrate the cell

membrane of live cells, the increase in PI staining indicates

an increase in the number of dead or apoptotic cells. The loss

in Dc

m

upon justicidin A treatment was also examined by

confocal microscopy and rhodamine 123 which preferentially

stains intact mitochondria. Consistent with the data obtained

from flow cytometry, a significant decrease in the intensity of

rhodamine 123 staining was observed at 12 h. The staining

intensity continued to decrease and reached the lowest level at

72 h after justicidin A treatment in both cell lines (Figure 4B

and C). A justicidin A dose-dependent loss in rhodamine 123

staining in both cell lines was also observed (Figure 4B and C).

There was no change in Dc

m

in vehicle-treated cells (Figure 4).

Interestingly, addition of 10 mM cyclosporin A to the cell

cultures 4 h prior to justicidin A treatment prevented the

decrease in the intensity of rhodamine 123 staining and

JA-induced apoptosis measuring by PI flow cytometry

(Figure 4D). Since cyclosporine A hyperpolarizes the

mito-chondrial membrane potential (Dc

m

) by binding to cyclophilin

D thus preventing the conversing of adenine nucleotide

trans-locase (ANT) to permeability transition (PT) pores, these

results indicate that justicidin A indeed causes a loss in Dc

m

.

Release of cyto c and Smac from mitochondria by justicidin A

treatment

To further confirm that justicidin A induces apoptosis by

affecting mitochondria, release of cyto

c and Smac from

mitochondria to the cytoplasm was examined by

immunoblot-ting. The cytosolic levels of both cyto

c and Smac were found

to be profoundly increased (1.7- and 1.4-fold, respectively) in

HT-29 cells at 24 h and reached peak levels (2.3–2.7 and

2.6-fold, respectively) at 48–72 h following the treatment

with justicidin A (0.75 mM) (Figure 5A–D). Conversely, levels

of both cyto

c and Smac in the mitochondria-enriched fractions

of HT-29 cells were decreased within 24 h (Figure 5A–D) and

became undetectable at 96 and 72 h, respectively, after

justi-cidin A treatment (Figure 5A–D). Treatment of HCT 116 cells

with 5 mM of justicidin A resulted in similar changes in cyto

c

and Smac levels. Cytosolic cyto

c level was greatly increased

(7.2-fold) 96 h and that of Smac was increased (2.7-fold) 24 h

after drug treatment (Figure 5A–D). The mitochondrial cyto

c

and Smac levels were decreased starting 24 and 48 h,

respect-ively, after treatment (Figure 5A–D). Since Smac antagonizes

the anti-apoptotic function of XIAP (37), the effect of

justi-cidin A on the total level of XIAP was also examined. As

shown in Figure 5E and F, XIAP levels were profoundly

decreased at 24 h in both HT-29 (0.75 mM of justicidin A)

and HCT 116 cell lines (5 mM of justicidin A) and reached

the lowest levels at 48 h in HT-29 cells and at 96 h in HCT

116 cells. In addition, XIAP levels were reduced in a justicidin

A dose-dependent manner (0.25–10 mM) in both cell lines

(Figure 5E).

Translocation of Bax from the cytosol to mitochondria

A characteristic feature of mitochondria-mediated apoptosis is

the translocation of the pro-apoptotic factor Bax from the

cytosol to mitochondria. To confirm the involvement of

mito-chondria in this justicidin A-caused apoptosis, the levels of

Bax in different cellular fractions were determined. The total

cellular level of Bax was found to be slightly increased

(1.34-fold) 6 h after justicidin A treatment and reached the

peak level (2-fold) at 12 h in HT-29 cells (Figure 6A and B).

In HCT 116 cells, the total Bax level was greatly increased

(2-fold) at 24 h and reached peak level (4.5-fold) at 72 h

post-treatment (Figure 6A and B). The level of Bax in

mitochondria-enriched

fraction

was

greatly

increased

(3.4-fold) within 6 h and peaked (4-fold) at 24 h in justicidin

A-treated HT-29 cells (Figure 6A and B). A dramatic increase

(4.34-fold) in mitochondrial Bax level was observed after

72 h of treatment in HCT 116 cells (Figure 6A and B). In

contrast, the amount of cytosolic Bax in both cell lines was

decreased upon justicidin A treatment in a time-dependent

manner (Figure 6A). Contrary to the increase of the

pro-apoptotic Bax in mitochondria, the level of the anti-pro-apoptotic

factor Bcl-X

L

in mitochondria was decreased upon justicidin A

treatment in both HT-29 and HCT 116 cells, and a longer

treatment resulted in a more severe decrease in the level of

Bcl-X

L

in both cell lines (Figure 6C). The total cellular level

of Bcl-X

L

was also found to be decreased by justicidin A

treatment in a time-dependent manner (Figure 6C).

Fig. 3. Justicidin A induced the cleavage of caspases and their target proteins in human colorectal cancer cells. Caspase-3 (A), caspase-9 (B), PARP (D), and DFF (E) in HT-29 and HCT 116 cells treated with justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) were assayed by western blot analysis. At indicated time points, whole cell, cytosolic and nuclear lysates were prepared and electrophoresed on 12% SDS–PAGE gels. Protein bands on the gels were transferred to polyvinylidene fluoride membranes. Anti-caspase-3, anti-caspase-9, anti-89 kDa PARP, anti-DFF-45 or DFF-40 antibodies were then used to detect respective proteins. (C) Inhibitory effect of the caspase-9 inhibitor Z-LEHD-fmk on justicidin A-induced apoptosis in HT-29 and HCT 116 cells was also examined. Cells were pretreated without or with 20 mM Z-LEHD-fmk for 4 h prior to exposure to justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) for 30 h. Cells were then stained with PI and analysed by flow cytometry. RACK1 (receptors for activated C-kinase) was similarly assessed to serve as a loading control. Justicidin A was diluted in culture medium containing 0.1% DMSO. Control cells were treated with medium containing 0.1% DMSO. Results are representative of three independent experiments. JA, justicidin A; Casp, caspase.

(8)

HCT 11

6 HT-29

C

0.5

1.0 ( M)

0

0.25

72

48

12

24

0 time (h)

( M)

5.0

10.0

0

2.5

B

time (h)

0

12

24

48

72 µ

12 time (h)

µ

.

)

Propidium iodide

Fig. 4. Effect of justicidin A on Dcmin human colorectal cancer cells. HT-29 and HCT 116 cells (2 105cells/well) in 2 ml medium were grown in six-well and

then treated with justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) as described elsewhere. At indicated time points, cells were stained

with 40 nM of DiOC6(3) and then analysed by flow cytometry (A) or imaged by confocal microscopy after staining with rhodamine 123 (B and C).

Normal cells had a DiOC6(3) staining intensity of 10

3

–104units (lower right quadrant). Justicidin A treatment damaged mitochondria and reduced

the intensity of DiOC6(3) staining, thus shifting more cells to the lower left quadrant. Apoptotic cells had a PI staining intensity of102units. More cells

with this staining intensity (upper two quadrants) were seen upon justicidin A treatment. Inhibitory effects of cyclosporin A on justicidin A-induced

change in Dcmand apoptosis in HT-29 and HCT 116 cells were also examined (D). Cells were pretreated with 10 mM cyclosporin A for 4 h prior

to exposure to justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) for 30 h. Cells were then stained with rhodamine 123 and examined by confocal microspcopy or stained with PI and then analysed by flow cytometry. Justicidin A was diluted in culture medium containing 0.1% DMSO. Control cells were treated with medium containing 0.1% DMSO. Results are representative of three independent experiments. JA, justicidin A.

(9)

Decrease in the level of Ku70 caused by justicidin A

To investigate whether Ku70 was involved in justicidin

A-induced mitochondrial dysfunction in human colorectal

cancer cells, the level of the Ku70 protein was determined by

immunoblot analysis. As seen in Figure 7A and B, the level of

cytosolic Ku70 was decreased (0.62-fold) as early as 2 h and

reached the lowest level (0.37-fold) at 12 h in HT-29 cells after

treatment with 0.75 mM of justicidin A. Similarly, the amount

of cytosolic Ku70 in HCT 116 cells was profoundly decreased

(0.6-fold) at 48 h and reached the lowest level (0.33-fold) at

96 h after justicidin A (5 mM) treatment in HCT 116 cells

(Figure 7A and B). In contrast, no change in the level of

Ku70 was observed in nuclear fractions of both

untreated-and justicidin A-treated cells (Figure 7A).

Suppression of justicidin A on the growth of human colorectal

cancer cells in mice

The possibility that justicidin A can inhibit the growth of

HT-29 cells implanted in NOD-SCID mice was also examined.

Mice were inoculated with HT-29 cells s.c. into the flank on

day 0 and then randomly divided into three groups on day 4.

One group (n

¼ 5) of mice received oral administration of

justicidin A (6.2 mg/kg) once a day for 56 consecutive days,

and the other group (n

¼ 5) received 10.6 mg/kg of justicidin A

on the same schedule. The control group (n

¼ 5) received

the vehicle (0.05% DMSO). Tumor volume and body weight

were recorded daily from day 4 (treatment start) until day 60

(treatment stop). Tumor-bearing mice treated with 10.6 or

6.2 mg/kg justicidin A showed a dramatic suppression of

tumor growth and decrease in tumor weight (P 5 0.05) as

compared with vehicle-treated mice (Figure 8A and B and

Table II). None of the justicidin A-treated or vehicle-treated

mice showed significant changes in spleen or liver weight

throughout the experiment (Table II).

Discussion

In this study, we found that justicidin A is toxic to many

different types of cells, including human colorectal cancer

(HT-29 and HCT 116), cervix carcinoma (SiHa), breast

adenocarcinoma (MCF7), bladder carcinoma (T24), human

embryonic kidney epithelial cells (HEK293) and PBMCs.

Interestingly, all cancer cells tested in this study were

particu-larly sensitive to justicidin A. For example, the IC

50

for T24

bladder carcinoma cells was 0.004 mM which is 6250-fold

lower than that for PBMCs (25 mM) (Table I). This result

suggests that justicidin A can be used for chemotherapy for

cancers. Since colorectal cancers are one of the most prevalent

cancers, we examined the effects of justicidin A on two

dif-ferent colorectal cancer cell lines (HT-29 and HCT 116).

IC

50

’s for HT-29 and HCT 116 were determined to be 0.11

and 0.4 mM, respectively. Justicidin A is believed to kill

HT-29 and HCT 116 cells by inducing apoptosis based on

the observation that justicidin A treatment caused caspase-3

3.6±0.3% 5.3±0.5% 27.6±1.1% 10.5±0.5% 8.9±0.8% 25.9±1.3% 5.6±0.4% 1.3±0.1% Fig. 4. Continued.

(10)

Fig. 5. Distribution of cyto c and Smac and levels of XIAP in justicidin A-treated cells. Cyto c (A), Smac (B) and XIAP (E) levels in HT-29 and HCT 116 cells after justicidin A treatment were determined by western blot analysis. HT-29 or HCT 116 cells were grown and treated with justicidin A as described in the Figure 2 legend. At indicated time points, whole cell lysate, cytosolic and mitochondria-enriched fraction were prepared and electrophoresed on 12% SDS–PAGE gels. Protein bands on the gel were then transferred to polyvinylidene fluoride membranes. The intensity of each protein band was quantified by desitometry

normalizing to that of RACK1 (receptors for activated C-kinase). The density values of cytoc, Smac and XIAP from control conditions were designated as 1.

The levels of these proteins in the remaining samples were expressed as fold of the control (C), (D) and (F). Anti-cyto c, anti-Smac and anti-XIAP antibodies were used to detect respective proteins on the blots. RACK1 (receptors for activated C-kinase) served as the loading control. Justicidin A was diluted in culture medium containing 0.1% DMSO. Control cells were treated with medium containing 0.1% DMSO. Results are representative of three independent experiments. JA, justicidin A.

(11)

and caspase-9 to become activated (Figure 3A and B).

Changes in Dc

m

and release of apoptogenic proteins such as

cyto

c and Smac from mitochondria upon justicidin A

treat-ment suggest that this apoptosis is mediated by the intrinsic

pathway in which mitochondria are affected.

Although both HT-29 and HCT 116 are colorectal cancer

cells, their sensitivity to justicidin A is different. HT-29 cells

are

4-fold more sensitive than HCT 116 cells to justicidin A.

However, the effects of justicidin A on these two cell lines

appear to be similar. All parameters related to apoptosis in

these two cell lines are similarly affected by justicidin A

including activation of caspase-3 and caspase-9 (Figure 3),

loss of Dc

m

(Figure 4), release of cyto

c and Smac from

mitochondria to the cytosol (Figure 5A and B), reduction in

XIAP level (Figure 5E and F), and the translocation of Bax

from the cytosol to mitochondrial (Figure 6A). Interestingly,

the level of cytosolic Ku70 was found to be dramatically

decreased upon treatment with justicidin A in both HT-29

and HCT 116 cells (Figure 7A). Ku70 is known to play a

crucial role in apoptosis. Down regulation of Ku70 has been

shown to enhance Bax-mediated apoptosis (19), whereas

over-expression of Ku70 has been shown to inhibit apoptosis (38).

Based on the results of this study, we propose the following

mechanism of action of justicidin A in the induction of

apoptosis in colorectal cancer cells: Justicidin A causes a

decrease in the level of cytosolic Ku70 allowing Bax to enter

mitochondria. Bax then oligomerizes and perturbs the outer

mitochondrial membrane causing release of apoptogens such

as Smac and cyto

c as shown in Figure 5A. Justicidin A also

causes a decrease in Bcl-X

L

level (Figure 6C). Since Bcl-X

L

may bind to Bax and prevent Bax from inserting into the outer

membrane of mitochondria (10), the decrease in Bcl-X

L

level

would favor Bax activation and thus the loss of Dc

m

and

release of apoptogen from the mitochondria. Smac, also

know as DIABLO, promotes caspase-9 activation by

neutral-izing members of the inhibitor of apoptosis protein family such

as XIAP which was found to be decreased in justicidin

A-treated cells (Figure 5E). This decrease in XIAP level enables

cyto

c, dATP, apoptotic protease activating factor-1 (Apaf-1),

and pro-caspase-9 to form apoptosomes to activate caspase-9

(12). Activated caspase-9 in turn activates caspase-3 to cleave

the 116 kDa PARP to two peptides of 89 and 24 kDa, rendering

PARP unable to carry out poly-ADP-ribosylation of various

proteins involved in DNA repair. In this study, the 89 kDa

form of PARP was found to be increased with time upon

justicidin A treatment (Figure 3D). Activated caspase-3 also

digests DFF-45 and DFF-35 that are natural inhibitors

of DFF-40. Degradation of DFF-45 and DFF-35 renders

DFF-40 free to enter the nucleus as shown in Figure 3E to

digest nuclear DNA leading to apoptosis of the cell. Treatment

of HT-29 and HCT 116 cells with justicidin A caused a

dose-and time-dependent increase in cytosolic cyto

c and Smac

Fig. 5. Continued.

(12)

(Figure 5). However, the cytosolic cyto

c level in HT-29 cells

treated with justicidin A for 96 h was decreased as compared

with those treated for 72 h (Figure 5C). A similar unexpected

decrease in cytosolic Smac was also seen in HCT 116 cells

treated with justicidin A for 72 h (Figure 5D). The reason of

this sudden decrease in cytosolic cyto

c and Smac is not

known. A possibility is that cyto

c and Smac are degraded

upon prolonged exposure in the cytoplasm or that they are

consumed during apoptosis.

Justicidin A also has been shown to inhibit the release of

tumor necrosis factor-alpha (TNF-a) from

lipopolysaccharide-treated RAW 264.7 macrophages in a concentration- and

time-dependent manner (39). In addition, justicidin A can inhibit

the transport of TNF-a (39) and has antiviral activity (40).

Fig. 6. Effect of justicidin A on the levels of pro- and anti-apoptotic proteins of the Bcl-2 family. The levels of Bax (A) and Bcl-XL(C) in HT-29 and HCT 116

cells treated with justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) were determined by western blot analysis. At indicated time points, whole cell, cytosolic and mitochondria-enriched lysates were prepared and electrophoresed on 12% SDS–PAGE gels. Protein bands on the gels were then transferred to polyvinylidene fluoride membranes. The intensity of each protein band was quantified by desitometry normalizing to that of RACK1 (receptors for activated C-kinase). The density value of Bax from control conditions was designated as 1. The levels of Bax in the remaining samples were expressed as fold of control (B).

(13)

These observations suggest that justicidin A has multiple

cel-lular targets. Justicidin A was shown to cause hepatocelcel-lular

carcinoma Hep 3B and Hep G

2

cells to die without decreasing

the level of Ku70 in these cells (our unpublished data). It is

possible that Ku70 is a target of justicidin A in colorectal

cancer cells, but not in hepatocellular carcinoma. In contrast

to cytosolic Ku70, the level of nuclear Ku70 in HT-29 and

HCT 116 cells was not affected by justicidin A (Figure 7A),

suggesting that justicidin A treatment activates a mechanism

which degrades cytosolic Ku70. Whether justicidin A affects

other factors involved in apoptosis remains to be determined.

Oral administrations of 6.2 or 10.6 mg/kg of justicidin A

once a day for 56 consecutive days were found to suppress the

growth of HT-29 cells transplanted in NOD-SCID mice but did

not have a significant effect on spleen or liver weight

(Table II). This result suggests that justicidin A at the dosage

of 6.2 or 10.6 mg/kg/day has no effect on normal cells.

Although more careful analyses of effects of justicidin A on

Fig. 7. Effect of justicidin A on the levels of the Ku70 protein. The levels of Ku70 (A) in HT-29 and HCT 116 cells treated with justicidin A (0.75 mM for HT-29 cells and 5.0 mM for HCT 116 cells) were determined by western blot analysis. At indicated time points, cytosolic and nuclear-enriched lysates were prepared and electrophoresed on 12% SDS–PAGE gels. Protein bands on the gels were then transferred to polyvinylidene fluoride membranes. The intensity of each protein band was quantified by desitometry normalizing to that of RACK1 (receptors for activated C-kinase). The density value of Ku70 from control conditions was designated as 1. The levels of Ku70 in the remaining samples were expressed as fold of control (B). Anti-Ku70 antibody was used to detect the proteins. RACK1 served as a loading control. Justicidin A was diluted with culture medium containing 0.1% DMSO. Control cells were treated with medium containing 0.1% DMSO. Results are representative of three independent experiments. JA, justicidin A.

Table II. Effect of dietary justicidin A on tumor weight, body weight, liver weight and spleen weighta

Treatment Body wt (g) Tumor wt (g) Liver wt (g) Spleen wt (g)

Vehicle 25.5 1.2 9.8 0.8 0.98 0.01 0.040 0.013

JA (10.6 mg/kg) 24.5 0.9 2.1 0.1 0.88 0.14 0.039 0.014

JA (6.2 mg/kg) 24.7 1.3 3.8 0.1 0.92 0.16 0.041 0.011

a

HT-29 cells (4 105

cells/mice) were implanted s.c. to the flank of mice. Four days after implantation, justicidin A (10.6 or 6.2 mg/kg) or vehicle was

orally administrated daily for 56 days. Data are presented as means SEM from two separate experiments (n ¼ 5 each group). Tumor weight (wt), body wt,

liver wt and spleen wt of each mouse were determined at the end of the experiment.

(14)

various organs remain to be done, results of this study show the

potential of using justicidin A for chemotherapy of colorectal

cancers.

Acknowledgements

This study was supported by grants (NSC 92-2320-B-006-085 and NSC 91-2314-B-006-097) from the National Science Council, Taiwan, Republic of China.

Conflict of Interest Statement: None declared.

References

1. Pfister,D.G., Benson,A.B. III and Somerfield,M.R. (2004) Clinical practice. Surveillance strategies after curative treatment of colorectal

cancer.N. Engl. J. Med., 350, 2375–2382.

2. Mayer,R.J. (2004) Two steps forward in the treatment of colorectal cancer. N. Engl. J. Med., 350, 2406–2408.

3. Hawk,E.T., Umar,A. and Viner,J.L. (2004) Colorectal cancer

chemoprevention—an overview of the science. Gastroenterology, 126,

1423–1447.

4. Kan,W.S. (1981)Pharmaceutical Botany. National Research Institute of

Chinese Medicine, Taipei, Taiwan.

5. Hsu,H.Y. (1982)Treating Cancer with Chinese Herbs. Oriental Healing

Arts Institute, Los Angeles, USA.

6. Okigawa,M., Maeda,T. and Kawano,N. (1970) The isolation and structure

of three new lignans fromJusticia procumbens Linn var Leucantha Honda.

Tetrahedron, 26, 4301–4305.

7. Fukamija,N. and Lee,K. (1986) Antitumor agents, justicidin-A and

diphyllin, two cytotoxic principles from Justicia procumbens. J. Nat.

Prod., 49, 348–350.

8. Day,S.H., Chiu,N.Y., Won,S.J. and Lin,C.N. (1999) Cytotoxic lignans of Justicia ciliata. J. Nat. Prod., 62, 1056–1058.

9. Day,S.H., Lin,Y.C., Tsai,M.L., Tsao,L.T., Ko,H.H., Chung,M.I., Lee,J.C., Wang,J.P., Won,S.J. and Lin,C.N. (2002) Potent cytotoxic lignans

from Justicia procumbens and their effects on nitric oxide and tumor

0 10 20 30 40 50 60 70 0 2000 4000 6000 8000 10000 12000

*

Control

*

* *

*

0 0 2000 4000 6000 8000 10000 12000

*

Control Control 6.2 mg/kg 10.6 mg/kg

A

*

* *

*

Days of treatment Tumor volume (mm 3)

JA (10.6 mg/kg)

Control

B

Fig. 8. Inhibitory effect of justicidin A on the growth of HT-29 cells xenografted into NOD-SCID mice. (A) HT-29 cells were implanted s.c. to the flank of mice. Four days after implantation, justicidin A (10.6 mg/kg, inverted triangle; 6.2 mg/kg, open circle) was orally administrated every day until day 60. (B) Photograph of mice bearing HT-29 human colorectal cancer cells at day 60 after oral administration of vehicle or justicidin A (10.6 mg/kg). Arrow indicates the time when

justicidin A treatment was initiated. Data are presented as means SEM from two separate experiments (n ¼ 5 in each group). Data points marked with asterisks

indicate significant difference between experimental and control groups with aP-value 5 0.05. JA, justicidin A.