www.elsevier.com/locate/ygyno

Gynecologic Oncology 94 (2004) 643–649

A novel retinoid, 4-[3,5-bis (trimethylsilyl) benzamido] benzoic acid

(TAC-101), induces apoptosis of human ovarian carcinoma cells and

shows potential as a new antitumor agent for clear

cell adenocarcinoma

Nao Suzukia,*, Daisuke Aokia, Shinji Oieb, Miwa Horiuchia, Yuko Hasegawaa,Sachiko Ezawaa, Atsushi Suzukia, Nobuyuki Susumua, Fumihito Hosoib,

Kenji Kitazatob, Shiro Nozawaa

aDepartment of Obstetrics and Gynecology, School of Medicine, Keio University, Tokyo, JapanbTaiho Pharmaceutical Co., Tokyo, Japan

Received 27 October 2003

Abstract

Objectives. A novel retinobenzoic acid derivative, 4-[3,5-bis (trimethylsilyl) benzamido] benzoic acid (TAC-101), was reported to

suppress the growth and invasion of human gastric cancer or hepatocellular carcinoma by induction of apoptosis. We examined the antitumor

activity of TAC-101 against human ovarian carcinoma cell lines.

Methods. Apoptosis of human epithelial ovarian carcinoma-derived cell lines (RMG-I, RMG-II, RTSG, RMUG-S, RMUG-L, and KF)

was investigated by detecting DNA laddering and was quantified by an enzyme-linked immunosorbent assay. Inhibition of apoptosis was

also examined using a caspase inhibitor. Furthermore, TAC-101 (8 mg kg�1 day�1 orally for 30 days) was investigated in nude mice with

subcutaneous RMG-II tumors. A prominent apoptotic response to TAC-101 was observed. The antitumor effects of cisplatin (7 mg/kg

intravenously on day 1) and paclitaxel (36 mg/kg intravenously on days 1 and 5) were also assessed for comparison.

Results. Apoptosis occurred in all of the cell lines (except KF) in a concentration-dependent manner after exposure to TAC-101 and was

markedly induced in RMG-I and RMG-II cells (derived from ovarian clear cell adenocarcinomas). A caspase inhibitor blocked the induction

of apoptosis by TAC-101. The maximum inhibition of RMG-II tumor growth in nude mice by TAC-101, cisplatin, and paclitaxel was 45%,

34%, and 47%, respectively.

Conclusion. Oral TAC-101 shows potential as a novel antitumor agent for ovarian carcinoma, especially ovarian clear cell

adenocarcinoma.

D 2004 Elsevier Inc. All rights reserved.

Keywords: TAC-101; Apoptosis; Ovarian carcinoma; Synthetic retinoid; Clear cell adenocarcinoma

Introduction and have led to markedly improved response and remission

About half of all ovarian cancer is detected at an ad-

vanced stage because early disease is asymptomatic and

tumors arise deep within the pelvis. Substantial advances

have been made in chemotherapy for ovarian epithelial car-

cinoma since the introduction of platinum-based regimens

0090-8258/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.ygyno.2004.06.026

* Corresponding author. Department of Obstetrics and Gynecology,

School of Medicine, Keio University, 35 Shinanomachi, Shinjyuku, Tokyo

160-8582, Japan. Fax: +81-3-3226-1667.

E-mail address: naos@sc.itc.keio.ac.jp (N. Suzuki).

rates after initial treatment. Despite various efforts to im-

prove the long-term outcome, however, ovarian cancer still

has the worst prognosis among gynecological malignancies.

Accordingly, new approaches for treatment are needed to

improve prognosis.

Retinoids are natural or synthetic vitamin A analogues

that play a major role in regulating the proliferation,

growth, and differentiation of both normal and malignant

cells [1]. The effects of retinoids are mainly mediated via

two classes of nuclear receptors, the retinoic acid receptors

(RARs) and retinoic X receptors (RXR), which are each

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649644

encoded by three distinct genes (a, h, and g) and are

members of the steroid and thyroid hormone receptor

superfamily. Retinoids inhibit experimental carcinogenesis

and induce the differentiation and/or growth inhibition of

fully transformed malignant cells [1]. Initially, the strong

antitumor activity of all-trans-retinoic acid (ATRA) against

acute promyelocytic leukemia was reported [2]. Lately,

evidence was obtained about the therapeutic potential of

retinoids against various other cancers, such as lung, breast,

prostate, and bladder cancer, germ cell tumors, sarcoma,

glioma, and neuroblastoma [3]. Furthermore, retinoids have

been shown to prevent the development of some forms of

skin cancer and are clinically effective for treating prema-

lignant and malignant cutaneous disorders [4]. The combi-

nation of 13-cis-retinoic acid with interferon-alpha has

produced a high response rate in patients with squamous

cell carcinoma of the head and neck [5] or the uterine

cervix [6]. Although the mechanism underlying the anti-

tumor activity of retinoids remains unclear, some natural

and synthetic retinoids show therapeutic and chemopreven-

tive effects via the induction of differentiation and/or

apoptosis both in vitro and in vivo [7]. Retinoic acid

(RA) has been reported to induce excessive mesenchymal

cell death in mouse embryos at sites where physiologic

death is limited [8] and in the developing mouse limb via

activation of RAR-h [9]. We previously showed that RA

induces apoptosis of human embryonal carcinoma cells

under the same conditions as those that induce differenti-

ation [10].

A novel substituted benzoic acid, 4-[3,5-bis (trimethyl-

silyl) benzamido] benzoic acid (TAC-101), binds to RAR-a,

but not RAR-h, RAR-g, or the RXRs [11]. Ligand binding byRARs is known to promote interaction of RARs with tran-

scription factors like RXRs and AP-1, and the induction of

biological responses such as differentiation [12] and apopto-

sis [13], as well as anti-metastastic [14] and anti-angiogenic

effects [15]. In fact, oral administration of TAC-101 has been

shown to inhibit liver metastasis from orthotopic human

gastrointestinal cancer xenografts and to prolong the survival

of nude mice [16]. TAC-101 has also shown a preventive

effect on experimental liver metastasis of colon cancer in

mice that was associated with the induction of apoptosis and

suppression of tumor cell invasion [17]. Furthermore, we

previously demonstrated that TAC-101 was able to suppress

liver metastasis by the induction of cancer cell apoptosis and

possibly by the modulation of AP-1 [18].

Since TAC-101 has shown antitumor activity against

metastatic gastric and liver cancers, we hypothesized that

it might also be effective against ovarian cancer, which is

frequently detected at an advanced stage in association with

lymph node metastasis and peritoneal dissemination. There-

fore, the study performed herein was performed to assess the

effects of TAC-101 on ovarian carcinoma cells, especially

ovarian clear cell adenocarcinoma, which has a poor prog-

nosis that is probably related to its resistance to standard

platinum-based chemotherapy [19].

Materials and methods

Chemicals

TAC-101, 4-[3,5-bis (trimethylsilyl) benzamido] ben-

zoic acid, was provided by Taiho Pharmaceutical Co.,

Ltd., (Saitama, Japan). Z-VAD-FMK (a caspase inhibitor)

and cis-diamine dichloroplatinum (cisplatin) were pur-

chased from Promega (WI, USA) and Nippon Kayaku

Co. (Tokyo, Japan), respectively. All-trans-retinoic acid

(ATRA) and paclitaxel were purchased from Sigma (St.

Louis, MO, USA). For in vitro experiments, TAC-101 and

ATRA were dissolved in dimethyl sulfoxide at concentra-

tions of 20 and 10 mM, respectively, to make stock

solution, which was stored at �20jC until use. For in

vivo experiments, TAC-101 was suspended in 0.5% hy-

droxyl-propyl-methylcellulose (Shin-Etsu Chemical Co.

Ltd., Tokyo, Japan). Paclitaxel was dissolved in 99.5%

ethanol (Nacalai Tesque, Inc., Kyoto, Japan) by sonication

to obtain 60 mg/ml solution, which was further diluted to

30 mg/ml with Cremophor EL (Nacalai Tesque, Inc.) to

make the stock solution.

Cell culture

RMG-I and RMG-II (human ovarian clear cell adeno-

carcinoma cell lines), RTSG (an undifferentiated human

ovarian carcinoma cell line), and RMUG-S and RMUG-L

(human ovarian mucinous adenocarcinoma cell lines) were

previously established as reported [20–23]. A human ovar-

ian serous adenocarcinoma cell line (KF) was kindly pro-

vided by Prof. K. Kikuchi (Department of Obstetrics and

Gynecology, National Defense Medical College, Saitama,

Japan) [24]. All lines were cultured in a 1:1 mixture of

Dulbecco’s modified Eagle’s medium and Ham’s F 12

medium (Gibco, Grand Island, NY, USA) supplemented

with 10% fetal calf serum (Mitsubishi Chemical Co., Tokyo,

Japan) and 80 Ag/ml of kanamycin sulfate at 37jC in an

atmosphere of 95% O2/5% CO2.

Detection of DNA laddering

Cell lines (1 � 106 cells) were seeded into flasks and

incubated for 3–5 days, after which the medium was

replaced with fresh medium containing TAC-101 (10 AM)

or ATRA (10 AM). Cultured cells were harvested after 24

h later. Genomic DNA was isolated using a Smitest EX-R

and D kit (Nippon Genetics Co., Tokyo, Japan). DNA

fragments were subjected to electrophoresis on 1.5% aga-

rose gel and were visualized by staining with ethidium

bromide.

Quantification of apoptosis

Cells were plated into 96-well cell-culture plates (1 � 104

cells/well) at 1 day before treatment. After 24 h of incuba-

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649 645

tion with various concentrations of TAC-101 or ATRA (0,

10, and 25 AM), DNA fragmentation was evaluated by

detection of cytoplasmic histone-associated DNA fragments

(mononucleosomes and oligonucleosomes) activity using a

Cell Death Detection ELISAplus (Roche Molecular Bio-

chemicals, IN, USA) according to the manufacturer’s

instructions and expressed as enrichment factor, which

was measured at OD405/492 (optical density of apoptotic

cells divided by the optical density of nontreated cells).

Cells were also incubated with 20 AM V-ZAD-FMK (a

caspase inhibitor) plus 25 AM TAC-101 for 24 h to test

whether apoptosis was inhibited.

Subcutaneous RMG-II tumors in nude mice

RMG-II cells (1 � 106 cells) were subcutaneously im-

planted into the backs of nude mice (6-week-old female

BALB/c nu/nu mice; CLEA Japan Inc., Tokyo, Japan), and

the resulting tumors were measured after 3 weeks. Mice that

had tumors with an estimated volume [the largest tumor

diameter in mm � (the smallest diameter in mm)2 / 2] of

approximately 100 mm3 were selected and assigned to

groups of eight animals each by stratified random allocation

based on individual tumor volumes so that each group had

an almost equal mean tumor volume on day 1 of dosing.

Tumor volumes were measured daily in treated and non-

treated, and the relative tumor volume (RTV) was expressed

as the ratio of the measured volume to that at the start of

treatment. The mean tumor volume of each group was

calculated as a percentage relative to that of the control

group and was used an index of antitumor activity. The

maximum tumor growth-inhibiting effect was considered to

represent the antitumor activity of each drug. We also

measured changes in the weight of the mice to confirm

the nontoxic dose of each drug and examined tumor growth

inhibition at the nontoxic dose. TAC-101 (8 mg kg�1 day�1)

was administered for 30 days by oral gavage using a 1-ml

tuberculin syringe (SS-01T; Terumo, Tokyo, Japan) and

needle. Cisplatin was injected into the tail vein at a dose

of 7 mg kg�1 day�1 on day 1, and paclitaxel was injected at

a dose of 36 mg kg�1 day�1 on days 1 and 5. The dose of

TAC-101 used in this study (8 mg kg�1 day�1) was

determined to be optimal based on its antiproliferative

activity and effect on body weight in a previous dose-

finding study [16]. The dose of cisplatin was equivalent to

the clinical dose [25], and the maximum-tolerated dose of

paclitaxel was employed [26]. A nontreated control group

was also included. Body weight was measured twice a week

to monitor toxicity. Animals were handled in accordance

with the protocol established by the Animal Care Commit-

tee of Taiho Pharmaceutical Co. Ltd.

Tissue preparation

Subcutaneously grafted RMG-II tumors were removed

and fixed for 48 h in 20% formalin [buffered with 0.1 M

phosphate-buffered saline (PBS), pH 7.4] at room temper-

ature, embedded in paraffin, and cut into 3-Am sections for

immunohistochemistry to evaluate drug-induced apoptosis

in situ.

Terminal deoxynucleotidyl transferase-mediated dUTP nick

end labeling (TUNEL) assay

The TUNEL assay for in situ detection of apoptosis

was performed with the Fluorescent Apoptosis Detection

System (Promega), by which the nuclei of apoptotic cells

were labeled with fluorescein-12-dUTP and propidium

iodide (1 Ag) (Sigma) in PBS and then were observed

under a fluorescent microscope (Olympus BX 60; Olym-

pus Co. Ltd., Tokyo Japan).

Detection of RAR gene expression by the reverse

transcriptase–polymerase chain reaction (RT-PCR)

Total RNA was extracted from each ovarian carcinoma

cell line using an RNeasy Protect Mini Kit (Qiagen,

Hilden, Germany). Then, single-stranded cDNA was pre-

pared using 300 ng of total RNA as a template, 1 unit of

Super Script II RNase H Reverse Transcriptase, 2.5 pmol/

Al of oligo dT primer, 1 Al of dNTP mixture in 0.1 M

dithiothretiol, and 40 units of RNase inhibitor (Invitrogen,

CA, USA). The reverse transcription reaction was done at

42jC for 2 min, 42jC for 50 min, and 70jC for 15 min.

To the cDNA thus obtained was added 0.5 units of DNA

polymerase (Ampli Taq Gold; Applied Biosystems, Foster

City, CA, USA), 1.2 Al of MgCl2 (25 mM), 1.6 Al of

dNTP mixture, 2 Al of 10� PCR gold buffer, 1 Al of

SYBR Green (Takara Bio Co., Shiga, Japan), 0.4 Al of

forward primer, and 0.4 Al of reverse primer (10 AM).

Then the PCR was performed using a spectrofluorometric

thermal cycler (Applied Biosystems 7700, PE-Biosystems).

The initial reaction conditions were 50jC for 2 min and

95jC for 10 min followed by 50 cycles of 15 s at 95jCand 1 min at 59jC for RAR-a and RAR-h, or by 50

cycles of 15 s at 95jC and 1 min at 62jC for RAR-g. RT-

PCR for glyceraldehyde-3-phosphate dehydrogenase

(GAPDH) was carried out similarly using Taqman

GAPDH Control Reagent (Applied Biosystems) to normal-

ize the expression of each gene. PCR primers [5V-AGATC-CAGAAGAACATGGTG-3V (RAR-a sense ) , 5V-CTTGAGGAGGGTGATCTGGTC-3V (RAR-a antisense),

5V-GAGAGAAGTTGGTGCTCAACG-3V (RAR-h sense),

5V-CCTCTGAACAGCTCACTTCC-3V (RAR-h antisense),

5V-AACAAGGTGACCAGGAATCG-3V(RAR-g sense), 5V-TCCATCTTCAGAGTAATGGCC-3V (RAR-g antisense)]

were devised based on published sequences [27].

Statistical analysis

Welch’s t test (two tailed) [28] was performed to compare

tumor volumes between the groups.

Fig. 2. Induction of apoptosis by TAC-101 in ovarian carcinoma cells and

effect of a caspase inhibitor (V-ZAD-FMK). Cells were incubated for 24

h with vehicle alone, TAC-101 (10, or 25 AM) alone, 20 AM V-ZAD-FMK

plus 25 AM TAC-101, or ATRA (10, or 25 AM) alone. Apoptosis was

assessed by ELISA and is expressed as an enrichment factor relative to the

nontreated control activity (calculated as the optical density of apoptotic

cells divided by that of nontreated cells).

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649646

Results

TAC-101 is apoptogenic for ovarian carcinoma cells

Genomic DNA laddering was demonstrated in all of the

TAC-101-treated cell lines examined, except KF cells (Fig.

1). Apoptosis was induced by TAC-101 in a concentra-

tion-dependent manner in all of the cell lines, except KF

cells, using a Cell Death Detection ELISAPLUS (Fig. 2).

Addition of Z-VAD-FMK, a caspase inhibitor, to cultures

abolished the apoptogenic effect of TAC-101 (Fig. 2). On

the other hand, when added to the medium even at the

same concentration as TAC-101, ATRA failed to induce

apoptosis in these (Fig. 2).

TAC-101 is active against ovarian clear cell

adenocarcinoma xenografts

RMG-II cells showed the most prominent apoptosis

when exposed to TAC-101 at 25 AM. Therefore, these cells

were inoculated subcutaneously into nude mice, and the

effect of orally administered TAC-101 on tumor growth was

investigated. The maximal tumor growth-inhibiting effect of

TAC-101 was seen on day 31 of administration, when there

was a 45% reduction of RTV. Compared with the control

group, there was a statistically significant decrease of RTV

(Welch’s test, P < 0.05). Cisplatin caused a maximum

reduction of 34% in RTVon day 14, while paclitaxel caused

a 47% reduction on day 14 (Fig. 3 and Table 1). For both the

cisplatin and paclitaxel groups, a significant decrease of

RTV was observed compared with the control (P < 0.01).

Fig. 1. Induction of apoptosis by TAC-101 in ovarian carcinoma cell lines.

The cells were treated with 10 AM TAC-101 for 24 h and apoptosis was

assessed by detection of genomic DNA laddering on agarose gel

electrophoresis. Lane 1: RMG-I, lane 2: RMG-II, lane 3: RTSG, lane 4:

RMUG-S, lane 5: RMUG-L, lane 6: KF.

Body weight was monitored by measuring the average

weight and the average change of weight in each group of

mice, in cases where the ratio of weight change to weight

exceeded 0.15, which was suggestive of excessive toxicity

from drug therapy. The mice were not included in the

evaluation of antitumor activity. When the fluorescent

TUNEL assay was performed on subcutaneous RMG-II

tumors harvested from mice treated with TAC-101,

TUNEL-positive cells were identified (Fig. 4).

Table 1

Activity against RMG-II tumors in nude mice

Compound RTV

day 14 day 31

Maximum inhibition rate

none 3.24 ± 0.87**

6.43 ± 2.08*

0% (days 14 and 31)

TAC-101 1.96 ± 0.55 3.55 ± 1.45 45% (day 31)

34% (day 14)5.65 ± 2.792.14 ± 0.57Cisplatin

Paclitaxel 2.02 ± 0.98 4.25 ± 2.72 47% (day 14)

n = 8; Welch’s test *p < 0.05, **p < 0.01.

RMG-II tumor-bearing mice were treated with TAC-101, cisplatin, or

paclitaxel. TAC-101 (8 mg kg�1 day�1) was administered for 30 days by

oral gavage. Cisplatin was injected into the tail vein at a dose of 7 mg kg�1

day�1 on day 1 and paclitaxel was injected at 36 mg kg�1 day�1 on days 1

and 5. Tumor volume was measured daily in mice with or without drug

treatment, and the relative tumor volume (RTV) was calculated as the ratio

of the measured volume that at the start of treatment.

Fig. 5. Expression of RAR-a, RAR-h, and RAR-g mRNA in ovarian

carcinoma cells detected by quantitative RT-PCR. High levels of RAR-a

mRNA expression were detected in RTSG and KF cells. There was no

difference in the expression of RAR-h mRNA among the six cell lines

examined. Higher expression of RAR-g mRNA was detected in RMUG-L

and KF cells. The abscissa shows mRNA levels relative to those in RMG-I

cells (set as 1).

Fig. 3. Antitumor effect of TAC-101, cisplatin, and paclitaxel against RMG-

II cells transplanted into nude mice. TAC-101 (D) was administered at 8 mg

kg�1 day�1 for 30 days by oral gavage. Cisplatin (w ) was injected into the

tail vein at a dose of 7 mg/kg/day on day 1 and paclitaxel (�) was injected

at a dose of 36 mg kg�1 day�1 on days 1 and 5. Tumor volume was

measured daily in mice (n = 8) with or without (o) drug treatment, and the

relative tumor volume (RTV) was calculated as the ratio of the daily

measured volume to that at the start of treatment.

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649 647

Expression of RAR-a, RAR-b, and RAR-c mRNA in ovarian

carcinoma cells

Comparison of the relative expression of mRNAs in each

cell line was performed by quantitative RT-PCR, with the

mRNA level in RMG-I cells being set at 1 (Fig. 5). RTSG

cells showed a high level of expression of mRNA for RAR-

a, which is considered to be the most important receptor for

TAC-101 [11], but RAR-a mRNA was also expressed by

KF cells in which apoptosis was not induced. There was no

difference in the level of RAR-h mRNA expression among

the six cell lines. Increased expression of RAR-g mRNA

A B

Fig. 4. TUNEL assay of RMG-II tumors transplanted into nude mice. (A)

Nontreated control: no TUNEL-positive yellow-green cells were observed.

(B) After exposure to TAC-101 at 25 AM for 24 h, TUNEL-positive cells

were detected. Nuclei were stained with propidium iodide.

was observed in RMUG-L and KF cells when compared

with the other cell lines.

Discussion

In the present study, apoptosis was induced by TAC-101 in

five out of six human ovarian carcinoma cell lines, excluding

KF cells. Apoptosis was particularly prominent in RMG-I

and RMG-II cells, which are ovarian clear cell adenocarci-

noma cell lines, when assessed from the enrichment factor

activity detected by ELISA. The TAC-101-induced apoptosis

of ovarian carcinoma cells was inhibited by coincubation

with V-ZAD-FMK, demonstrating that caspases are involved

in the mechanism of this effect of TAC-101. However, the

downstream mechanisms that occurs following binding of

TAC-101 to an RAR, leading to activation of caspases and

apoptosis, remains to be investigated.

The actions of retinoids are mainly mediated by two

classes of nuclear retinoid receptors, RARs and retinoid X

receptors [9]. After binding by an RAR-a-selective retinoid,

RAR-a interacts with the retinoic acid response element in

the RAR-h gene promoter, resulting in the transactivation of

genes controlling the processes of differentiation and apo-

ptosis in human breast cancer and lung cancer cells [18,29].

According to RT-PCR, expression of mRNAwas detected in

all six of the cell lines, including KF cells in which TAC-

101 failed to induce apoptosis. Quantitative RT-PCR

revealed no relationship between the level of RAR-a

mRNA expression and the induction of apoptosis. There-

fore, we deduce that the resistance of KF cells to induction

of apoptosis by TAC-101 involves a different mechanism.

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649648

Several studies have already demonstrated an antitumor

effect of RA or retinoids against ovarian cancer [30].

According to Soprano et al. [31], RA inhibits ovarian cancer

cell proliferation by blocking AP-1 activity rather than by

induction of apoptosis or induction of the expression of

factors such as transforming growth factor-h or cell cycle-

regulating factors. They demonstrated that RA caused a

decrease of AP-1 activity in RA-sensitive CA-OV3 cells,

whereas there was no such decrease in RA-insensitive SK-

OV3 cells. Similarly, the synthetic retinoids fenretinide and

CD437 were reported to exert an antitumor effect on ovarian

cancer cells by inducing apoptosis [28,32–35].

Ovarian clear cell adenocarcinoma accounts for more than

10% of ovarian epithelial carcinoma [36] and its incidence

has recently been increasing in Japan. Up to 60% of patients

with clear cell adenocarcinoma are in International Federa-

tion of Gynecology and Obstetrics (FIGO) stage I. Our

findings have suggested that a pathological diagnosis of

ovarian clear cell adenocarcinoma might be an independent

high-risk factor in patients with FIGO stage I ovarian cancer

(unpublished observation). Among the various histologic

types of ovarian epithelial carcinoma, patients with clear cell

adenocarcinoma have the worst prognosis even when the

tumor is diagnosed at an early stage and treated by complete

resection [37]. Since clear cell adenocarcinoma seems to

show resistance to platinum-based chemotherapy, unlike

the other pathological subtypes of epithelial ovarian carcino-

ma [38,39], new therapeutic agents and strategies are neces-

sary to treat these tumors.

In the present study, TAC-101 showed antitumor activity

against RMG-II ovarian clear cell adenocarcinoma (the

tumor showing maximal apoptosis) in nude mice, and

tumor growth inhibition by TAC-101 was similar to or

greater than that achieved by cisplatin or paclitaxel, two

chemotherapeutic agents that are often clinically used to

treat ovarian epithelial carcinoma. When tested on cultured

hepatocellular carcinoma and breast cancer cells trans-

planted into mice, the same doses of cisplatin or paclitaxel

caused at least 80% tumor growth inhibition or induced

complete cure, respectively (unpublished observations). In

contrast, cisplatin and paclitaxel only achieved 40–50%

inhibition of the growth of clear cell adenocarcinoma

implanted in nude mice. In fact, the clinical response of

clear cell adenocarcinoma to cisplatin or paclitaxel is

unsatisfactory, and more effective drugs are under investi-

gation. TAC-101 showed inhibition of RMG-II tumors

similar to that compared to paclitaxel and a stronger

inhibition compared to cisplatin. This result suggests that

TAC-101 may be useful as an antitumor drug to treat clear

cell adenocarcinoma. Thus, TAC-101, which can be orally

administered, was established to exert an antitumor effect

on ovarian clear cell adenocarcinoma, which responds

rather poorly to cisplatin and paclitaxel; drugs that have

demonstrated potent antitumor activity against other can-

cers. Hence, TAC-101 has potential as a new therapeutic

agent for clear cell adenocarcinoma, which is often refrac-

tory to chemotherapy and, even at FIGO stage I, has a poor

prognosis. A study to assess the effects of its coadminis-

tration with cisplatin is underway, with the results obtained

so far indicating a possible additive or synergistic effect

against clear cell adenocarcinoma (data not shown).

The potential value TAC-101 is also supported by the

following considerations.

The toxicity of systemic RA therapy essentially corre-

sponds to the hyper vitaminosis A syndrome and involves

mucocutaneous toxicities such as mucosal dryness, ery-

thema, and desquamation of the skin, as well as cheilitis

[40]. ATRA appears to be unsuitable for long-term ther-

apy because its plasma concentration declines over time

during daily administration [41,42], but this phenomenon

is not seen with TAC-101 (unpublished observation).

Although adverse reactions to TAC-101 include the

known retinoid toxicities, the severe myelosuppression

and gastrointestinal toxicity associated with conventional

anticancer drugs should not occur [43], and thus TAC-101

might be easier to use in various clinical settings for the

management of ovarian carcinoma and oral administration

could contribute to a better quality of life. The present

findings suggested that TAC-101 has a mechanism of

action to induce apoptosis that differs from that of

platinum-based chemotherapeutic drugs, classified as

DNA synthesis inhibitors, or from the taxanes, classified

as anti-tubulin inhibitors, to exert its antitumor effects. In

conclusion, TAC-101 may be applicable for the treatment

of ovarian carcinoma because it showed a comparable

antitumor effect to cisplatin or paclitaxel against clear cell

adenocarcinoma.

Acknowledgments

The authors thank Ms. Keiko Abe for secretarial work.

The AM555S was originally developed by the Shudo group

of the University of Tokyo.

References

[1] Tallman MS, Wiernik PH. Retinoids in cancer treatment. J Clin Phar-

macol 1992;32:868–88.

[2] Huang ME, Ye YC, Chen SR, Chai JX, Zhoa L, Gu LJ, et al. Use of

all-trans retinoic acid in the treatment of acute promyelocytic leuke-

mia. Blood 1988;72:567–72.

[3] Smith MA, Parkinson DR, Cheson BD, Friedman MA. Retinoids in

cancer therapy. J Clin Oncol 1992;109:839–64.

[4] Lippman SM, Meyskens Jr FL. Results of the use vitamin A and

retinoids in cutaneous malignancies. Pharmacol Ther 1989;40:

107–22.

[5] Lippman SM, Kessler JF, Al-Sarraf M, Alberts DS, Itri LM, Mattox

D, et al. Treatment of advanced squamous cell carcinoma of head and

neck with isotretinoin: a phase II randomized trial. Invest New Drugs

1988;6:51–6.

[6] Lippman SM, Kavanagh JJ, Paredes-Espinoza M, Delgadillo-

Madrueno F, Paredes-Casillas P, Hong W-K, et al. 13-Cis-retinoic

acid plus interferon-alpha-2a: highly active systemic therapy for

N. Suzuki et al. / Gynecologic Oncology 94 (2004) 643–649 649

squamous cell carcinoma of the cervix. J Natl Cancer Inst 1992;

84:241–5.

[7] Lotan R. Retinoid and apoptosis: implication for cancer chemopre-

vention and therapy. J Natl Cancer Inst 1995;87:1655–7.

[8] Alles A, Sulik K. Retinoic acid-induced limb reduction defects: per-

turbation of zones of programmed cell death as a pathogenetic mech-

anism. Teratology 1989;40:163–71.

[9] Dolle P, Ruberte E, Kastner P, PetkovichM, Stoner CM,Gudas LJ, et al.

Differential expression of genes coding a, h, and g retinoic acid recep-

tors and CRABP-I in the developing limbs of the mouse. Nature

1989;342:702–5.

[10] Yamada T, Suzuki N, Hiraoka N, Matsuoka K, Fukushima S, Hashi-

guchi A, et al. Apoptosis of human embryonal carcinoma cells with in

vitro differentiation. Cell Struct Funct 1996;21:53–61.

[11] Sun SY, Yue P, Dawson MI, Shroot B, Michel S, Lamph WW, et al.

Differential effects of synthetic nuclear retinoid receptor-sensitive ret-

inoids on the growth of human non-small cell lung carcinoma cells.

Cancer Res 1997;57:4931–9.

[12] Hashimoto Y, Kagechika H, Kawachi E, Fukusawa H, Saito G, Shudo

K. Evaluation of differentiation-inducing activity of retinoids on hu-

man leukemia cell lines HL-60 and NB4. Biol Pharm Bull 1996;19:

1322–8.

[13] Zhang LX, Mills KJ, Dawson MI, Collins SJ, Jetten AM. Evidence

for the involvement of retinoic acid receptor RAR-a-dependent sig-

naling pathway in the induction of tissue transglutaminase and apo-

ptosis by retinoids. J Biol Chem 1995;270:6022–9.

[14] Pollard M, Luckert PH. The inhibitory effect of 4-hydroxyphenyl

retinamide (4-HPR) on metastasis of prostate adenocarcinoma-III

cells in Lobund–Wistar rats. Cancer Lett 1991;59:159–63.

[15] Oikawa T, Okayasu I, Ashino H, Morita I, Murota S, Shudo K. Three

novel synthetic retinoids, Re80, Am580, Am80, all exhibit anti-an-

giogenic activity in vitro. Eur J Pharmacol 1993;249:113–6.

[16] Murakami K, Werzba K, Sano M, Shibata M, Shibata J, Yonekura K,

et al. TAC-101, a benzoic acid derivative, inhibits liver metastasis of

human gastrointestinal cancer and prolongs the lifespan. Clin Exp

Metastasis 1998;16:323–31.

[17] Sakukawa R, Murakami K, Ikeda T, Yamada Y, Saiki I. Effect of 4-

[3, 5-bis(trimethylsilyl) benzamido] benzoic acid (TAC-101) on the

liver metastasis of colon 26-L5 carcinoma cells. Oncol Res 1998;10:

287–93.

[18] Shibata J, Murakami K, Aoyagi Y, Oie S, HashimotoA, Suzuki K, et al.

The induction of apoptosis and inhibition of AP-1 activity by TAC-101

(4-[3,5-bis(trimethylsilyl) benzamido] benzoic acid) may result in a

life-prolonging effect in animals bearing metastasis. Anticancer Res

2000;20:3583–90.

[19] Hakes TB. Prognostic factors in ovarian cancer. In: Markman M,

Hoskins WJ, editors. Cancer of the ovary. New York: Raven Press;

1993. p. 115–26.

[20] Nozawa S, Tsukazaki K, Sakayori M, Jeng CH, Iizuka R. Establish-

ment of a human ovarian clear cell carcinoma cell line (RMG-I) and

its single cloning. Hum Cell 1988;1:426–35.

[21] Nozawa S, Yajima M, Sasaki H, Tsukazaki K, Aoki D, Sakayori M, et

al. A new CA-125-like antigen (CA602) recognized by two mono-

clonal antibodies against a newly established ovarian clear cell carci-

noma cell line (RMG-II). Jpn J Cancer Res 1991;82:854–61.

[22] Udagawa Y, Nozawa S, Chin K, Sakayori M, Mikami M, Ohta K, et al.

Establishment and characterization of a human chorionic gonadotropin

(hCG) producing cell line (RTSG) from an ovarian epithelial cancer.

Hum Cell 1990;3:70–5.

[23] Sakayori M, Nozawa S, Udagawa Y, Chin K, Lee SG, Sakuma T, et al.

Biological properties of two newly established cell lines (RMUG-S,

RMUG-L) from a human ovarian mucinous cystoadenocarcinoma.

Hum Cell 1990;3:52–76.

[24] Kikuchi Y, Miyauchi M, Kizawa I, Oomori K, Kato K. Establishment

of cisplatin-resistant human ovarian cancer cell line. J Natl Cancer

Inst 1986;77:1181–5.

[25] Nomura T, Sakurai Y, Inaba M. The nude mouse and anticancer drug

evaluation. Tokyo: chemotherapy Publishers Inc; 1996. p. 29–52.

[26] Polizzi D, Pratesi G, Tortoreto M, Supino R, Riva A, Bombardelli E, et

al. A novel taxane with improved tolerability and therapeutic activity in

a panel of human tumor xenografts. Cancer Res 1999;59:1036–40.

[27] Murakami K, Matsumura T, Sano M, Hashimoto A, Yonekura K,

Sakukawa R, et al. 4-[3,5-Bis (trimethylsilyl) benzamido] benzoic

acid (TAC-101) inhibits the intrahepatic spread of hepatocellular car-

cinoma and prolongs the lifespan of tumor-bearing animals. Clin Exp

Metastasis 1998;16:633–43.

[28] McArdle BH. The significance of differences between means: a sim-

ulation study. Comp Biochem Physiol 1987;979–82.

[29] Li Y, Hashimoto Y, Agadir A, Kagechika H, Zhang XK. Identification

of a novel class of retinoic acid receptor h-selective retinoid antago-

nists and their inhibitory effects on AP-1 activity and retinoic acid-

induced apoptosis in human breast cancer cells. J Biol Chem 1999;

274:15360–6.

[30] Zhang D, Holms WF, Wu S, Soprano DR, Soprano KJ. Retinoids and

ovarian cancer. J Cell Physiol 2000;185:1–20.

[31] Soprano DR, Chen LX, Wu S, Donigan AM, Borghaei RC, Soprano

KJ. Overexpression of both RAR and RXR restores AP-1 repression

in ovarian adenocarcinoma cells resistant to retinoic acid-dependent

growth inhibition. Oncogene 1996;12:577–84.

[32] Sabichi AL, Hendricks DT, Bober MA, Birrer MJ. Retinoic acid

receptor beta expression and growth inhibition of gynecologic cancer

cells by the synthetic retinoid N-(4-hydroxyphenyl) retinamide. J Natl

Cancer Inst 1998;90:597–605.

[33] Chao WR, Hobbs PD, Jong L, Zhang Y, Wu Q, Shroot B, et al.

Effects of receptor class- and subtype-selective retinoids and an

apoptosis-inducing retinoid on the adherent growth of the NIH:OV-

CAR-3 ovarian cancer cell line in culture. Cancer Lett 1997;

115:1–7.

[34] Langdon SP, Rabiasz GJ, Ritchie AA, Reichert U, Buchan P, Miller

WR, et al. Growth-inhibitory effects of the synthetic retinoid CD437

against ovarian carcinoma models in vitro and in vivo. Cancer Che-

mother Pharmacol 1998;42:429–32.

[35] Um SJ, Lee SY, Kim EJ, Han HS, Koh YM, Hong KJ, et al. Anti-

proliferative mechanism of retinoid derivatives in ovarian cancer

cells. Cancer Lett 2001;174:127–34.

[36] Russell P, Bannatyne P. Surgical pathology of the ovaries. New York:

Churchill Livingstone; 1989.

[37] O’Brien ME, Schofield JB, Tan S, Fryatt I, Fisher C, Wilshaw E.

Clear cell epithelial ovarian cancer (mesonephroid): bad prognosis

only in early stage. Gynecol Oncol 1993;49:250–4.

[38] Alexander WK, Charles VB, William RH. Ovarian clear cell adeno-

carcinoma. Gynecol Oncol 1989;32:342–9.

[39] Monga M, Carmichael JA, Shelly WE. Surgery without adjuvant

chemotherapy for early epithelial ovarian carcinoma after comprehen-

sive surgical staging. Gynecol Oncol 1991;43:195–7.

[40] Silverman A, Ellis C, Voorhees J. Hypervitaminosis A syndrome: a

paradigm of retinoid side effects. J Am Acad Dermatol 1987;16:

1027–39.

[41] Mjuindi J, Frankel S, Huselton C, DeGrazia F, Garland W, Young C,

et al. Clinical pharmacology of oral all-trans retinoic acid in patients

with acute promyelocytic leukemia. Cancer 1992;52:2138–42.

[42] Kizaki M, Ueno H, Yamazoe Y, Shimada M, Takayama N, Muto A,

et al. Possible role of cytochrome P450 and P-glycoprotein. Blood

1996;87:725–33.

[43] Rizvi NA,Marshall JL, Ness E, HawkinsMJ, Kessler C, JacobsH, et al.

Initial clinical trial of oral; TAC-101, novel retinoic acid receptor-alpha

selective retinoid, in patients with advanced cancer. J Clin Oncol

2002;20:3522–32.