This article was downloaded by: [Monash University Library] On: 25 August 2015, At: 15:37 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 Howick Place, London, SW1P 1WG
Cancer Investigation Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/icnv20
Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1 a
a
a
b
b
Toshihiko Hirano , Katsushi Fukuoka , Kitaro Oka , Takashi Naito , Kunio Hosaka , Hiroshi b
Mitsuhashi & Yasuhiro Matsumoto
c
a
Division of Clinical Pharmacology Tokyo College of Pharmacy, 1432-1, Horinouchi, Hachioji, Tokyo b
Tsumura Laboratory 3586 Yoshiwara, Ami-cho Inashiki-gun, Ibaraki, 300-11
c
Division of Internal Medicine Hachioji Medical Center Tokyo Medical College, Tate-machi, Hachioji, Tokyo, 193, Japan Published online: 11 Apr 2015.
To cite this article: Toshihiko Hirano, Katsushi Fukuoka, Kitaro Oka, Takashi Naito, Kunio Hosaka, Hiroshi Mitsuhashi & Yasuhiro Matsumoto (1990) Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1, Cancer Investigation, 8:6, 595-602 To link to this article: http://dx.doi.org/10.3109/07357909009018926
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Cancer Investigation, 8(6) 595-602 (1990)
Downloaded by [Monash University Library] at 15:37 25 August 2015
Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1 Toshihiko Hirano, Ph.D.,* Katsushi Fukuoka, B.S.,* Kitaro Oka, Ph.D.,* Takashi Naito, Ph.D.,t Kunio Hosaka, Ph.D.,t Hiroshi Mitsuhashi, Ph.D.,t and Yasuhiro Matsumoto, M.D.* *Division of Clinical Pharmacology Tokyo College of Pharmacy 1432-1 Horinouchi, Hachioji, Tokyo t Tsumura Laboratory 3586 Yoshiwara, Ami-cho Inashiki-gun, Ibaraki 300-1 1 $Division of Internal Medicine Hachioji Medical Center Tokyo Medical College Tate-machi, Hachioji, Tokyo 193, Japan
ABSTRACT
The effect of each of twelve mammalian lignun derivatives on the growth of human mammary tumor ZR-75-1 cells was examined. At a concentration less than 10 pg/ml, tumor cell growth was inhibitedfrom 1848%. The effect of 2,3-dibenzylbutane-l,4diol(hatta1in) was found to be strongest, inhibiting growth by 50% at a concentration (EC,& of 2. I pgcglml. Hattalin inhibited membrane Na+, K+-ATPase of canine kidney cortex. It also inhibited the ATPase of the plasma membrane fraction from both cultured cells and a section of human breast cancer tissue at a concentration ranging from 0.5 to 2.0 mM. However, only a f a v percent of membrane ATPase from either ZR-75-1 cells or breast carcinoma tissue was inhibited by 2.0 mM of ouabain, suggesting that the target ATPase of hattalin was other than ouabainsensitive ATPase. The relative incorporation of [3H]thymidineper 1 X I@ cells into the acid-precipitablefraction of ZR-75-1 cells was not affected by 1-50 pg/ml of hattalin, while a marked decrease resultedfrom 1-10 pg/ml of 5-fluorouracil(5-FU). These results suggest that the suppressive effect of hattalin on tumor cell growth 595 Copyright
0 1991 by Marcel Dekker, Inc.
596
Hirano et al.
m y not occur through inhibition of DNA synthesis but rather partly by inhibition of the plasma membrane ATPase other than Na+ and K+dependent ones.
Downloaded by [Monash University Library] at 15:37 25 August 2015
INTRODUCTION Recent research has indicated a group of compounds called lignans, previously known as constituents of certain higher plants, also are present in animals and humans but with slight difference in structure (1-4). Intestinal microbes possibly may be involved in the formation of these mammah'anlignans (5). The low incidence of breast cancer in certain countries has been correlated, by epidemiological data, with adherence to primarily vegetarian diets (6,7). The excretion of mammalian lignans in urine has been closely correlated with fiber intake. In view of the antimitotic activity of plant lignans (8,9), these findings may indicate that mammalian lignans are involved in some way in the prevention of breast cancer. Such mammalian lignans are 2,3-dibenzylbutane derivativeswhich include enterodiol and enterolactone (4). In the microbial metabolism of plant lignans, reductive molecular transformation occurs through the action of intestinal anaerobicorganisms. Thus, the number of hydroxyl substituents on the benzene rings of these two lignans is half that of the original plant lignans. In enterodiol, the lactone ring of the intrinsic structure is opened by anaerobic hydrogenation. Intestinal anaerobic metabolism generally produces metabolites with a greater degree of hydrophobicity. Thus, the test compounds listed in Figure 1 were prepared based on our consideration that greater hydrophobicity of a structure would mean greater bioactivity compared with that of plant components or known mammalian analogs. In the present report the effects of synthetic mammaliantype lignan derivatives on the growth of a human breast cancer cell line were examined. The naked lignan derivative, 2,3-dibenzylbutane-1,4-diol (hattalin) and its methoxyl derivative, both highly lipophilic enterodiol analogs, were found to have the most potent effect on the growth of ZR-75-1cells, a human breast carcinoma cell line. A possible mechanism of antitumor cell activity of the lignans is proposed.
from rn-methoxyhydrocinnamk acid through enterolactone methyl ether IIe as described in the literature (10). Lactone IIb was prepared from hydrocinnamic acid in a similar manner. The stereoisomerism of both these lactones was trans. Enterodiol Id was obtained from 3-methoxyhydrocinnaicacid via enterodiol methyl ether Ie. A similar chemical sequence initiated from the corresponding hydroxycinnamic acid derivatives gave three enterodiol derivatives If-h as dl-isomers. A stereo mixture of hattalin was also prepared from hydroxycinnamic acid and subjected to liquid chromatography using silica
R1 T
R I
R1
R2
R3
a
H
H
OH
b
H
H H OH
OH
c
H
d e f g
H H OH OMe OMe
h
Lignan Derivatives The bgnans Used are indicated in Figure 1 9 and were prepared as follows. Enterolactone IId was synthesized
OMe OMe OMe OMe
OMe OH OH OH
OH OMe
meso dl dl dl dl dl dl dl
Hattalin Hattalin Enterodiol
/ R2
*.
11
R1
MATERIALS AND METHODS
l
b
d e g
H H H OMe
R2
H OH OMe OMe
trans(*) trans(*) trans(*) trans(*)
Enterolactone di-o-Methylenterolactone
di-o-Methylmatairesinol
Figure 1. Chemical structures of synthetic and mammalian lignans (enterodiol and enterolactone) used in this study.
Lignan Effect on a Breast Carcinoma Cell gel column to obtain Ia and Ib, identified as meso- and dl-isomers, respectively. A test solution of each lignan was prepared from a stock solution adjusted at 10 mg/ml by 100% 2-propanol.
Other Chemicals 5-Fluorouracil (5-FU) was obtained from Sigma. [6-3H]Thymidine (22 Ci/mmol) was purchased from Amersham International, Plc (Amersham, England). Penicillin G potassium and streptomycin sulfate were from Meiji, Co., Ltd., Japan.
Downloaded by [Monash University Library] at 15:37 25 August 2015
Cell Culture of ZR-75-1 ZR-75- 1 human breast carcinoma cells (1 l), were cultured at 37°C under 5 % COz-humidified air in a RPMI-1640 medium (GIBCO) containing 10% fetal calf serum (GIBCO), 100 units/ml of penicillin, and 100 pg/ml of streptomycin. Cell counting is carried out as follows: remove medium, rinse monolayer with fresh 0.25 % trypsin, and allow the culture to stand at room temperature for 5-1 1 min. Add fresh medium, aspirate, and dispense the suspension into new flasks. Single cells were washed with fresh medium three times, and resuspended in the medium at a concentration of 2 x 105 cells/ml. Cell viability after trypsinization as determined by a dye exclusion test using 0.16% trypan blue was found to exceed 95 % . One ml of the cell suspension was placed in each well of a 24-well plastic plate followed by the addition of 10 pl of 2-propanol (control) or the test solution dissolved in 2-propanol, bringing the final concentration of each lignan to 10 pglml. Serial concentrations of mesohattalin were in the range 1-100 pglml. Effects of ouabain and 5-FU at concentrationsof 100 pg/ml and 1 pg/ml, respectively, were also examined. Living cell number was counted by the trypan blue dye exclusion test. Data are presented as percent inhibition of cell proliferation, cell number in the presence of lignan x 100/cell number in the absence of lignan (2-propanol control).
[5HJThymidine Incorporation ZR-75-1 cells (2 x 10s/ml) were cultured in the presence of serial concentrations of hattalin or 5-FU for total 88 h. They were subsequently labeled with [3H]thymidine (1 pCi/ml) during the last 16 h. Incorporated radioactivity of the trichloroacetic acid-insoluble fraction was counted by a liquid scintillation counter. Triplicate runs of the experiment were conducted, and the average value of radioactivity incorporated per los cells was calculated.
597
Preparation of Membrane ATPase from ZR-75-1Cells Membrane ATPase preparation from the cultured cells was carried out by a slightly modified method of Robbins and Baker (12). In brief, the washed exponential cells (1 g wet weight) were resuspended in 1 mM Tris-EDTA at pH 7.4 (0.5 g/ml) and left on ice for 20 min. Freshly prepared 5 % sodium deoxycholate solution was then added at a final concentration of 0.35 % (w/v). The cells were kept on ice for 20 min with constant stirring and centrifuged at 10,000 x g for 10 min at 4°C to remove cell debris. The supernatant was centrifuged at 77,000 X g for 16 h at 4°C. The resulting pellet was collected and homogenized in 1 mM Tris-EDTA at pH 7.4. The homogenate had a specific activity of 0.44 unit/mg protein (one unit being that amount causing the release of 1 pmol of phosphate per min at 37°C) and was used as the membrane ATPase preparation. Protein content was determined by the method of Lowry et al. (13).
Preparation of ATPase from Human Breast Carcinoma Tissue One gram of breast carcinoma tissue dissected free of lipid from a patient 50 years of age was cut into pieces in 20 ml of ice-cold 5 mM EDTA solution containing 0.32 M sucrose. This was followed by homogenation and centrifugation at 6,000 x g for 20 min at 4°C to remove nuclei, mitochondria, and other cell debris. The supernatant was centrifuged at 85,000 x g for 60 min at 4°C to give a microsomal pellet which was subsequently treated with 2 M NaI in the presence of 50 mM cysteine, 5 mM MgClZ,3 mM ATP, and 5 mM EDTA as described elsewhere (14). The mixture was centrifuged at 20,000 x g for 30 min and the pellet thus obtained was washed three times with 5 mM EDTA at pH 7.4. The pellet was resuspended in water containing protein at a concentration of 2.2 mg/ml for use in the ATPase assay. The fraction contained 1.48 enzyme unit/mg protein. A normal tissue sample taken from a patient along with adhering carcinoma tissue at surgery was treated similarly to obtain an enzyme fraction containing the same units on unit quantity of the enzyme per protein as that of the carcinoma preparation.
ATPase Assay An alcoholic mso-hattalin solution at serial concentrations was preincubated for 10 min at 37°C with 160 pl of incubation buffer solution consisting of 100 mM NaCl,
Downloaded by [Monash University Library] at 15:37 25 August 2015
598
Hirano et al.
50 mM Tris-HC1 @H 7.4), 6 mM MgCl2, and 20 mM KC1, plus 20 pl of ATPase preparation. The specific ATPase activities of canine kidney cortex (Sigma), breast carcinoma cells, breast carcinoma tissue, and normal breast tissue prepared as above were found to be 1.00, 0.44,1.48, and 0.46 unit/mg protein, respectively. To compare the effects of meso-hattalin on these enzyme activities, each enzyme fraction was prepared so as to contain 3.0 enzyme units/ml, and 20 p1 of each were used in one assay tube. Next 20 p1 of 30 mM ATP were added to make the final concentrations of the drug 0.5, 1, and 2 mM. To eliminate the effects of alcohol, its final concentration was made 1% and a control experiment was carried out separately. After incubating the mixture for 10 min, 200 pl of ice-cold 10% trichloroacetic acid were added to terminate the reaction, and the liberated orthophosphate (Pi) was determined. Percent inhibition of ATPase activity was calculated by the formula: (Pi liberated in the presence of lignan/Pi liberated in the absence of lignan) x 100.
100
Structure I
L
I
5(
I--
0 .r
-0
0 L W
$ c,
c
b
I
b
W
e
e
S I
0 CI V
rn 7
RESULTS
5
0 L W
c,
T 1
c
aJ
Effects of Lignans on the Growth
Structure I 1
of ZR-75-1 Cells Figure 2 shows the effects of twelve lignans including two well-known human metabolites on the cell growth of ZR-75-1 after 72 h incubation at a 10 pg/ml concentration in each case. Cell growth was inhibited by 18-68%. These lignans appeared not to be cytotoxic, since most ( >95 %) of the treated cells were viable according to their trypan blue exclusion. The stereoisomer, 2,3-dibenzylbutane-1, 4-diol (hattalin, meso, and dl) and methyl ether Ie of enterodiol Id caused the greatest inhibition. The effect of meso-hattalin at various concentrations was also examined. Living cell numbers decreased in a dosedependent manner by meso-hattalin, as shown by the time-growth curve in Figure 3. The concentration of meso-hattalin which inhibited exponential cell growth by 50% was determined as 2.1 p g h l from a dose-effect curve in Figure 4A. The structure-activity relationship may be explained as follows. The butane-l,bdiol skeleton in structure I corresponds to activity slightly higher than that of the ylactones with structure 11. Reductive ring-opening of ylactones IIb and IIe afforded the corresponding diols Ib and Ie with activity exceeding that of the original lactones. However, the transformation of IId to Id or of IIg to Ig showed either none or only slight effect on their activity. The phenolic substituent on the benzene ring had less
Figure 2. Anticancer activity of synthetic and mammalian lignans on the growth of the breast carcinoma cell line of human (ZR-75-1). Ten pg of each lignan per ml culture was used. ZR-75-1 cells were treated with the lignan or the diluent (2-propanol) as a control for 72 h, and the living cells in the culture were counted by dye exclusion method. The data were shown as % inhibition of the cell growth compared with that of diluent control (0%).Each symbol in the figure corresponds to that shown in Figure 1. Bars indicate SD of the triplicate assay.
activity than any of the corresponding methyl ethers or deoxygenated derivatives. Thus, Ia, Ib, and Ie each showed activity higher than that of Id. Essentially the same was found in comparing If with Ie and Ig, and IId with IIb, IIe and IIg. The two methoxyl groups on each benzene ring lessened the activity, causing that of the compounds with no methoxyl or a single methoxyl to be relatively high. Thus, compounds Ia, lb and Ie exerted the strongest effect on cell proliferation.
Effect of Hattalin on [3H]ThymidineIncorporation ZR-75-1 cells were treated with meso-hattalin or 5-FU for 72 h, and then metabolically labeled with E3H]thymidine for 16 h. Radioactivities incorporated into the acid-insoluble fraction of cultured ZR-75-1 cells
599
Lignan Effect on a Breast Carcinoma Cell
inhibited total membrane ATPase of ZR-75-1 cells by 40% at a concentration of 2.0 mh4 as shown in Figure 6, and did the same for breast carcinoma tissue taken from a patient (Fig. 7A). A newly prepared enzyme fraction from breast carcinoma tissue was more sensitive to the effect of the lignan than that of the known cell line. However, ATPase prepared from normal breast tissue of the same patient was not affected by meso-hattalin (Fig. 7A). In contrast, only a few percent of the total ATPase activity from either ZR-75-1 cells or breast carcinoma tissue was inhibited by a high concentration (2.0 mM) of ouabain (Figs. 6 and 7B). This suggests that hattalin inhibited a tumor ATPase other than Na+ and K + -dependent one.
20 18
-F
-
14 16
-.1 2 -
n
0
x
v
W L
5
-5
10
8-
ul
:.
.r
-
6
-
Downloaded by [Monash University Library] at 15:37 25 August 2015
2
4-
DISCUSSION 2-
00
24
48
72
96
Incubation time (h)
Figure 3. Effect of different concentrations of hattalin on the growth of ZR-75-1 cells. Replicate cultures of ZR-75-1 cells growing exponentially were treated with different concenttations of hattalin as indicated below. At various times, aliquots of the cultures were removed, and the number of living cells counted. 0, control cells with diluent (2-propanol); 0 , 1 jtglml; A , 5 pglml; A , 10 pglml; 0 , 20 pglml; B, 50 pglml; X, 80 pglml; Q 100 pglml.
in the presence of meso-hattalin at various concentrations from 1 to 50 p g / d or 5-FU at 1 to 10 pg/ml are shown in Figure 4B. Meso-hattalin failed to have effect on the incorporation of 13H]thymidineat any concentration, but inhibited cell proliferation (Fig. 4A). In contrast, 5-FU strongly inhibited both cell proliferation by 62 or 95% and the incorporation of [3H]thymidineby 74 or 92% at 1 or 10 pg/ml, respectively. It would thus appear that hattalin has little effect on DNA synthesis in ZR-75-1 cells.
Effect of Hattalin on Membrane ATPase Prepared from ZR-75-1Cells and Patient Tissue Mammalian lignans have inhibitory activity against cellmembrane Na+, K+-ATPase (21,22). Hattalin also inhibited Na+, K+-ATPase from canine kidney cortex as shown in Figure 5. These observations remind us of a possibility that the antiproliferative effect of lignans on ZR-75-1 cells may result from the inhibitionof membrane Na+, K+-ATPase of the tumor cells. Meso-hattalin
The anticell proliferating activity of naturally occurring plant lignans (8,9) has prompted the search for other natural and synthetic lignans with possible therapeutic value for combating cancer. Many lignans of the podophyllotoxin type have been found effective in vitro and in vivo for treating experimental tumors (15-17), and in some cases, clinical applications have been made (18,19). However, their high toxicity imposes serious limitations on their use. Recently, mammalian lignans such as enterodiol (Id), enterolactone (IId), di-omethylenterolactone (IIe), and di-o-methylmatairesinol (IIg) have been detected in human urine, bile, and plasma (1,2,4,22). Each has a distinct molecular structure and less toxicity than known plant compounds. This prompted the authors to examine mammalian lignans and their synthetic derivatives to evaluate their anticancer activity toward a human breast carcinoma cell line in vitro. All the lignans examined were found to possess antiproliferative activity toward the ZR-75-1 cells. Based on the structure-activityrelationship described in the Experimental section, rneso-hattalin Ia was chosen for additional pharmacological examination. An attempt was thus made to clarify the mechanism for its antiproliferative activity differing from that of 5-FU which inhibits DNA replication. Both the growth of the DNA synthesis in ZR-75-1 cells was highly sensitive to 5-FU even at a concentration less than 10 pglml. In contrast, DNA synthesis appeared not to be affected at all by 50 p g l d of meso-hattalin which strongly inhibited cell growth, indicating that the inhibition of DNA synthesis does not contribute to hattalin activity. For this reason, hattalin may differ from podophyllotoxin derivatives (20) with respect to its mechanism of action.
Hirano et al.
0
A B
-
-
Downloaded by [Monash University Library] at 15:37 25 August 2015
hattal i n
r)
I
I
I
I I
1
1
0
10
20
30
40
50
-0
hattalin
5-FU
I
I 0
I 20
10
Drug c o n c e n t r a t i o n ()lg/ml)
I
I
I
30
40
50
Drug c o n c e n t r a t i o n ()lg/ml)
Figure 4. Effect of hattalin on the growth and the incorporation of [3H]thymidineinto acid-insolublefraction of ZR-75-1 cells. (A) Dose response curves of hattalin and 5-FU effects on cell growth at 72 h. (B) ZR-75-1 cells (2 x lo5) were incubated with serial concentrations of hattalin or 5-FU for 72 h. This was followed by incubation with 1 pCi/ml of [SH]thymidinefor 16 h, and measurement of radioactivity incorporated in the acid-insoluble fraction of the cells. Living cell number was counted simultaneously, and [)H]thymidine incorporated per lo5 cells was calculated. 0 , hattalin; 0, 5-FU.
For the clarification of this mechanism, attention was directed primarily to its action on membrane Na+, K+-ATPase. Recent reports (21,22) indicated enterolactone to suppress this enzyme obtained from the human heart. One possible mechanism for the suppression of tumor cell growth may be inhibition of membrane
T
0.6
h
ffl ffl
ra \
.-c E
u
I
0
0.5
I
1 .o
I
2.0
H a t t a l i n concentration (M)
Figure 6. Effect of hattalin (A)or ouabain (0)on membrane ATPase Figure 5. Effect of hattalin (A)or ouabain (0)on membrane Na+, K+-ATPase from canine kidney cortex.
from cultured ZR-75-1 cells. Percent activity of the enzyme remaining in the presence of each concentrationof hamiinor ouabain was measured. The activity in the absence of drug was designated as 100%. Bars indicate SD of the triplicate assay.
Lignan Effect on a Breast Carcinoma Cell
601
B
A
T
-
0.6
h
1
VI
m VI \
.-c: E
. 0
.r
n.
-5 0
0.3
%
I
CI .r
.-> w
m u
w
VI
m
a
Cancer Investigation Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/icnv20
Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1 a
a
a
b
b
Toshihiko Hirano , Katsushi Fukuoka , Kitaro Oka , Takashi Naito , Kunio Hosaka , Hiroshi b
Mitsuhashi & Yasuhiro Matsumoto
c
a
Division of Clinical Pharmacology Tokyo College of Pharmacy, 1432-1, Horinouchi, Hachioji, Tokyo b
Tsumura Laboratory 3586 Yoshiwara, Ami-cho Inashiki-gun, Ibaraki, 300-11
c
Division of Internal Medicine Hachioji Medical Center Tokyo Medical College, Tate-machi, Hachioji, Tokyo, 193, Japan Published online: 11 Apr 2015.
To cite this article: Toshihiko Hirano, Katsushi Fukuoka, Kitaro Oka, Takashi Naito, Kunio Hosaka, Hiroshi Mitsuhashi & Yasuhiro Matsumoto (1990) Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1, Cancer Investigation, 8:6, 595-602 To link to this article: http://dx.doi.org/10.3109/07357909009018926
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Cancer Investigation, 8(6) 595-602 (1990)
Downloaded by [Monash University Library] at 15:37 25 August 2015
Antiproliferative Activity of Mammalian Lignan Derivatives Against the Human Breast Carcinoma Cell Line, ZR-75-1 Toshihiko Hirano, Ph.D.,* Katsushi Fukuoka, B.S.,* Kitaro Oka, Ph.D.,* Takashi Naito, Ph.D.,t Kunio Hosaka, Ph.D.,t Hiroshi Mitsuhashi, Ph.D.,t and Yasuhiro Matsumoto, M.D.* *Division of Clinical Pharmacology Tokyo College of Pharmacy 1432-1 Horinouchi, Hachioji, Tokyo t Tsumura Laboratory 3586 Yoshiwara, Ami-cho Inashiki-gun, Ibaraki 300-1 1 $Division of Internal Medicine Hachioji Medical Center Tokyo Medical College Tate-machi, Hachioji, Tokyo 193, Japan
ABSTRACT
The effect of each of twelve mammalian lignun derivatives on the growth of human mammary tumor ZR-75-1 cells was examined. At a concentration less than 10 pg/ml, tumor cell growth was inhibitedfrom 1848%. The effect of 2,3-dibenzylbutane-l,4diol(hatta1in) was found to be strongest, inhibiting growth by 50% at a concentration (EC,& of 2. I pgcglml. Hattalin inhibited membrane Na+, K+-ATPase of canine kidney cortex. It also inhibited the ATPase of the plasma membrane fraction from both cultured cells and a section of human breast cancer tissue at a concentration ranging from 0.5 to 2.0 mM. However, only a f a v percent of membrane ATPase from either ZR-75-1 cells or breast carcinoma tissue was inhibited by 2.0 mM of ouabain, suggesting that the target ATPase of hattalin was other than ouabainsensitive ATPase. The relative incorporation of [3H]thymidineper 1 X I@ cells into the acid-precipitablefraction of ZR-75-1 cells was not affected by 1-50 pg/ml of hattalin, while a marked decrease resultedfrom 1-10 pg/ml of 5-fluorouracil(5-FU). These results suggest that the suppressive effect of hattalin on tumor cell growth 595 Copyright
0 1991 by Marcel Dekker, Inc.
596
Hirano et al.
m y not occur through inhibition of DNA synthesis but rather partly by inhibition of the plasma membrane ATPase other than Na+ and K+dependent ones.
Downloaded by [Monash University Library] at 15:37 25 August 2015
INTRODUCTION Recent research has indicated a group of compounds called lignans, previously known as constituents of certain higher plants, also are present in animals and humans but with slight difference in structure (1-4). Intestinal microbes possibly may be involved in the formation of these mammah'anlignans (5). The low incidence of breast cancer in certain countries has been correlated, by epidemiological data, with adherence to primarily vegetarian diets (6,7). The excretion of mammalian lignans in urine has been closely correlated with fiber intake. In view of the antimitotic activity of plant lignans (8,9), these findings may indicate that mammalian lignans are involved in some way in the prevention of breast cancer. Such mammalian lignans are 2,3-dibenzylbutane derivativeswhich include enterodiol and enterolactone (4). In the microbial metabolism of plant lignans, reductive molecular transformation occurs through the action of intestinal anaerobicorganisms. Thus, the number of hydroxyl substituents on the benzene rings of these two lignans is half that of the original plant lignans. In enterodiol, the lactone ring of the intrinsic structure is opened by anaerobic hydrogenation. Intestinal anaerobic metabolism generally produces metabolites with a greater degree of hydrophobicity. Thus, the test compounds listed in Figure 1 were prepared based on our consideration that greater hydrophobicity of a structure would mean greater bioactivity compared with that of plant components or known mammalian analogs. In the present report the effects of synthetic mammaliantype lignan derivatives on the growth of a human breast cancer cell line were examined. The naked lignan derivative, 2,3-dibenzylbutane-1,4-diol (hattalin) and its methoxyl derivative, both highly lipophilic enterodiol analogs, were found to have the most potent effect on the growth of ZR-75-1cells, a human breast carcinoma cell line. A possible mechanism of antitumor cell activity of the lignans is proposed.
from rn-methoxyhydrocinnamk acid through enterolactone methyl ether IIe as described in the literature (10). Lactone IIb was prepared from hydrocinnamic acid in a similar manner. The stereoisomerism of both these lactones was trans. Enterodiol Id was obtained from 3-methoxyhydrocinnaicacid via enterodiol methyl ether Ie. A similar chemical sequence initiated from the corresponding hydroxycinnamic acid derivatives gave three enterodiol derivatives If-h as dl-isomers. A stereo mixture of hattalin was also prepared from hydroxycinnamic acid and subjected to liquid chromatography using silica
R1 T
R I
R1
R2
R3
a
H
H
OH
b
H
H H OH
OH
c
H
d e f g
H H OH OMe OMe
h
Lignan Derivatives The bgnans Used are indicated in Figure 1 9 and were prepared as follows. Enterolactone IId was synthesized
OMe OMe OMe OMe
OMe OH OH OH
OH OMe
meso dl dl dl dl dl dl dl
Hattalin Hattalin Enterodiol
/ R2
*.
11
R1
MATERIALS AND METHODS
l
b
d e g
H H H OMe
R2
H OH OMe OMe
trans(*) trans(*) trans(*) trans(*)
Enterolactone di-o-Methylenterolactone
di-o-Methylmatairesinol
Figure 1. Chemical structures of synthetic and mammalian lignans (enterodiol and enterolactone) used in this study.
Lignan Effect on a Breast Carcinoma Cell gel column to obtain Ia and Ib, identified as meso- and dl-isomers, respectively. A test solution of each lignan was prepared from a stock solution adjusted at 10 mg/ml by 100% 2-propanol.
Other Chemicals 5-Fluorouracil (5-FU) was obtained from Sigma. [6-3H]Thymidine (22 Ci/mmol) was purchased from Amersham International, Plc (Amersham, England). Penicillin G potassium and streptomycin sulfate were from Meiji, Co., Ltd., Japan.
Downloaded by [Monash University Library] at 15:37 25 August 2015
Cell Culture of ZR-75-1 ZR-75- 1 human breast carcinoma cells (1 l), were cultured at 37°C under 5 % COz-humidified air in a RPMI-1640 medium (GIBCO) containing 10% fetal calf serum (GIBCO), 100 units/ml of penicillin, and 100 pg/ml of streptomycin. Cell counting is carried out as follows: remove medium, rinse monolayer with fresh 0.25 % trypsin, and allow the culture to stand at room temperature for 5-1 1 min. Add fresh medium, aspirate, and dispense the suspension into new flasks. Single cells were washed with fresh medium three times, and resuspended in the medium at a concentration of 2 x 105 cells/ml. Cell viability after trypsinization as determined by a dye exclusion test using 0.16% trypan blue was found to exceed 95 % . One ml of the cell suspension was placed in each well of a 24-well plastic plate followed by the addition of 10 pl of 2-propanol (control) or the test solution dissolved in 2-propanol, bringing the final concentration of each lignan to 10 pglml. Serial concentrations of mesohattalin were in the range 1-100 pglml. Effects of ouabain and 5-FU at concentrationsof 100 pg/ml and 1 pg/ml, respectively, were also examined. Living cell number was counted by the trypan blue dye exclusion test. Data are presented as percent inhibition of cell proliferation, cell number in the presence of lignan x 100/cell number in the absence of lignan (2-propanol control).
[5HJThymidine Incorporation ZR-75-1 cells (2 x 10s/ml) were cultured in the presence of serial concentrations of hattalin or 5-FU for total 88 h. They were subsequently labeled with [3H]thymidine (1 pCi/ml) during the last 16 h. Incorporated radioactivity of the trichloroacetic acid-insoluble fraction was counted by a liquid scintillation counter. Triplicate runs of the experiment were conducted, and the average value of radioactivity incorporated per los cells was calculated.
597
Preparation of Membrane ATPase from ZR-75-1Cells Membrane ATPase preparation from the cultured cells was carried out by a slightly modified method of Robbins and Baker (12). In brief, the washed exponential cells (1 g wet weight) were resuspended in 1 mM Tris-EDTA at pH 7.4 (0.5 g/ml) and left on ice for 20 min. Freshly prepared 5 % sodium deoxycholate solution was then added at a final concentration of 0.35 % (w/v). The cells were kept on ice for 20 min with constant stirring and centrifuged at 10,000 x g for 10 min at 4°C to remove cell debris. The supernatant was centrifuged at 77,000 X g for 16 h at 4°C. The resulting pellet was collected and homogenized in 1 mM Tris-EDTA at pH 7.4. The homogenate had a specific activity of 0.44 unit/mg protein (one unit being that amount causing the release of 1 pmol of phosphate per min at 37°C) and was used as the membrane ATPase preparation. Protein content was determined by the method of Lowry et al. (13).
Preparation of ATPase from Human Breast Carcinoma Tissue One gram of breast carcinoma tissue dissected free of lipid from a patient 50 years of age was cut into pieces in 20 ml of ice-cold 5 mM EDTA solution containing 0.32 M sucrose. This was followed by homogenation and centrifugation at 6,000 x g for 20 min at 4°C to remove nuclei, mitochondria, and other cell debris. The supernatant was centrifuged at 85,000 x g for 60 min at 4°C to give a microsomal pellet which was subsequently treated with 2 M NaI in the presence of 50 mM cysteine, 5 mM MgClZ,3 mM ATP, and 5 mM EDTA as described elsewhere (14). The mixture was centrifuged at 20,000 x g for 30 min and the pellet thus obtained was washed three times with 5 mM EDTA at pH 7.4. The pellet was resuspended in water containing protein at a concentration of 2.2 mg/ml for use in the ATPase assay. The fraction contained 1.48 enzyme unit/mg protein. A normal tissue sample taken from a patient along with adhering carcinoma tissue at surgery was treated similarly to obtain an enzyme fraction containing the same units on unit quantity of the enzyme per protein as that of the carcinoma preparation.
ATPase Assay An alcoholic mso-hattalin solution at serial concentrations was preincubated for 10 min at 37°C with 160 pl of incubation buffer solution consisting of 100 mM NaCl,
Downloaded by [Monash University Library] at 15:37 25 August 2015
598
Hirano et al.
50 mM Tris-HC1 @H 7.4), 6 mM MgCl2, and 20 mM KC1, plus 20 pl of ATPase preparation. The specific ATPase activities of canine kidney cortex (Sigma), breast carcinoma cells, breast carcinoma tissue, and normal breast tissue prepared as above were found to be 1.00, 0.44,1.48, and 0.46 unit/mg protein, respectively. To compare the effects of meso-hattalin on these enzyme activities, each enzyme fraction was prepared so as to contain 3.0 enzyme units/ml, and 20 p1 of each were used in one assay tube. Next 20 p1 of 30 mM ATP were added to make the final concentrations of the drug 0.5, 1, and 2 mM. To eliminate the effects of alcohol, its final concentration was made 1% and a control experiment was carried out separately. After incubating the mixture for 10 min, 200 pl of ice-cold 10% trichloroacetic acid were added to terminate the reaction, and the liberated orthophosphate (Pi) was determined. Percent inhibition of ATPase activity was calculated by the formula: (Pi liberated in the presence of lignan/Pi liberated in the absence of lignan) x 100.
100
Structure I
L
I
5(
I--
0 .r
-0
0 L W
$ c,
c
b
I
b
W
e
e
S I
0 CI V
rn 7
RESULTS
5
0 L W
c,
T 1
c
aJ
Effects of Lignans on the Growth
Structure I 1
of ZR-75-1 Cells Figure 2 shows the effects of twelve lignans including two well-known human metabolites on the cell growth of ZR-75-1 after 72 h incubation at a 10 pg/ml concentration in each case. Cell growth was inhibited by 18-68%. These lignans appeared not to be cytotoxic, since most ( >95 %) of the treated cells were viable according to their trypan blue exclusion. The stereoisomer, 2,3-dibenzylbutane-1, 4-diol (hattalin, meso, and dl) and methyl ether Ie of enterodiol Id caused the greatest inhibition. The effect of meso-hattalin at various concentrations was also examined. Living cell numbers decreased in a dosedependent manner by meso-hattalin, as shown by the time-growth curve in Figure 3. The concentration of meso-hattalin which inhibited exponential cell growth by 50% was determined as 2.1 p g h l from a dose-effect curve in Figure 4A. The structure-activity relationship may be explained as follows. The butane-l,bdiol skeleton in structure I corresponds to activity slightly higher than that of the ylactones with structure 11. Reductive ring-opening of ylactones IIb and IIe afforded the corresponding diols Ib and Ie with activity exceeding that of the original lactones. However, the transformation of IId to Id or of IIg to Ig showed either none or only slight effect on their activity. The phenolic substituent on the benzene ring had less
Figure 2. Anticancer activity of synthetic and mammalian lignans on the growth of the breast carcinoma cell line of human (ZR-75-1). Ten pg of each lignan per ml culture was used. ZR-75-1 cells were treated with the lignan or the diluent (2-propanol) as a control for 72 h, and the living cells in the culture were counted by dye exclusion method. The data were shown as % inhibition of the cell growth compared with that of diluent control (0%).Each symbol in the figure corresponds to that shown in Figure 1. Bars indicate SD of the triplicate assay.
activity than any of the corresponding methyl ethers or deoxygenated derivatives. Thus, Ia, Ib, and Ie each showed activity higher than that of Id. Essentially the same was found in comparing If with Ie and Ig, and IId with IIb, IIe and IIg. The two methoxyl groups on each benzene ring lessened the activity, causing that of the compounds with no methoxyl or a single methoxyl to be relatively high. Thus, compounds Ia, lb and Ie exerted the strongest effect on cell proliferation.
Effect of Hattalin on [3H]ThymidineIncorporation ZR-75-1 cells were treated with meso-hattalin or 5-FU for 72 h, and then metabolically labeled with E3H]thymidine for 16 h. Radioactivities incorporated into the acid-insoluble fraction of cultured ZR-75-1 cells
599
Lignan Effect on a Breast Carcinoma Cell
inhibited total membrane ATPase of ZR-75-1 cells by 40% at a concentration of 2.0 mh4 as shown in Figure 6, and did the same for breast carcinoma tissue taken from a patient (Fig. 7A). A newly prepared enzyme fraction from breast carcinoma tissue was more sensitive to the effect of the lignan than that of the known cell line. However, ATPase prepared from normal breast tissue of the same patient was not affected by meso-hattalin (Fig. 7A). In contrast, only a few percent of the total ATPase activity from either ZR-75-1 cells or breast carcinoma tissue was inhibited by a high concentration (2.0 mM) of ouabain (Figs. 6 and 7B). This suggests that hattalin inhibited a tumor ATPase other than Na+ and K + -dependent one.
20 18
-F
-
14 16
-.1 2 -
n
0
x
v
W L
5
-5
10
8-
ul
:.
.r
-
6
-
Downloaded by [Monash University Library] at 15:37 25 August 2015
2
4-
DISCUSSION 2-
00
24
48
72
96
Incubation time (h)
Figure 3. Effect of different concentrations of hattalin on the growth of ZR-75-1 cells. Replicate cultures of ZR-75-1 cells growing exponentially were treated with different concenttations of hattalin as indicated below. At various times, aliquots of the cultures were removed, and the number of living cells counted. 0, control cells with diluent (2-propanol); 0 , 1 jtglml; A , 5 pglml; A , 10 pglml; 0 , 20 pglml; B, 50 pglml; X, 80 pglml; Q 100 pglml.
in the presence of meso-hattalin at various concentrations from 1 to 50 p g / d or 5-FU at 1 to 10 pg/ml are shown in Figure 4B. Meso-hattalin failed to have effect on the incorporation of 13H]thymidineat any concentration, but inhibited cell proliferation (Fig. 4A). In contrast, 5-FU strongly inhibited both cell proliferation by 62 or 95% and the incorporation of [3H]thymidineby 74 or 92% at 1 or 10 pg/ml, respectively. It would thus appear that hattalin has little effect on DNA synthesis in ZR-75-1 cells.
Effect of Hattalin on Membrane ATPase Prepared from ZR-75-1Cells and Patient Tissue Mammalian lignans have inhibitory activity against cellmembrane Na+, K+-ATPase (21,22). Hattalin also inhibited Na+, K+-ATPase from canine kidney cortex as shown in Figure 5. These observations remind us of a possibility that the antiproliferative effect of lignans on ZR-75-1 cells may result from the inhibitionof membrane Na+, K+-ATPase of the tumor cells. Meso-hattalin
The anticell proliferating activity of naturally occurring plant lignans (8,9) has prompted the search for other natural and synthetic lignans with possible therapeutic value for combating cancer. Many lignans of the podophyllotoxin type have been found effective in vitro and in vivo for treating experimental tumors (15-17), and in some cases, clinical applications have been made (18,19). However, their high toxicity imposes serious limitations on their use. Recently, mammalian lignans such as enterodiol (Id), enterolactone (IId), di-omethylenterolactone (IIe), and di-o-methylmatairesinol (IIg) have been detected in human urine, bile, and plasma (1,2,4,22). Each has a distinct molecular structure and less toxicity than known plant compounds. This prompted the authors to examine mammalian lignans and their synthetic derivatives to evaluate their anticancer activity toward a human breast carcinoma cell line in vitro. All the lignans examined were found to possess antiproliferative activity toward the ZR-75-1 cells. Based on the structure-activityrelationship described in the Experimental section, rneso-hattalin Ia was chosen for additional pharmacological examination. An attempt was thus made to clarify the mechanism for its antiproliferative activity differing from that of 5-FU which inhibits DNA replication. Both the growth of the DNA synthesis in ZR-75-1 cells was highly sensitive to 5-FU even at a concentration less than 10 pglml. In contrast, DNA synthesis appeared not to be affected at all by 50 p g l d of meso-hattalin which strongly inhibited cell growth, indicating that the inhibition of DNA synthesis does not contribute to hattalin activity. For this reason, hattalin may differ from podophyllotoxin derivatives (20) with respect to its mechanism of action.
Hirano et al.
0
A B
-
-
Downloaded by [Monash University Library] at 15:37 25 August 2015
hattal i n
r)
I
I
I
I I
1
1
0
10
20
30
40
50
-0
hattalin
5-FU
I
I 0
I 20
10
Drug c o n c e n t r a t i o n ()lg/ml)
I
I
I
30
40
50
Drug c o n c e n t r a t i o n ()lg/ml)
Figure 4. Effect of hattalin on the growth and the incorporation of [3H]thymidineinto acid-insolublefraction of ZR-75-1 cells. (A) Dose response curves of hattalin and 5-FU effects on cell growth at 72 h. (B) ZR-75-1 cells (2 x lo5) were incubated with serial concentrations of hattalin or 5-FU for 72 h. This was followed by incubation with 1 pCi/ml of [SH]thymidinefor 16 h, and measurement of radioactivity incorporated in the acid-insoluble fraction of the cells. Living cell number was counted simultaneously, and [)H]thymidine incorporated per lo5 cells was calculated. 0 , hattalin; 0, 5-FU.
For the clarification of this mechanism, attention was directed primarily to its action on membrane Na+, K+-ATPase. Recent reports (21,22) indicated enterolactone to suppress this enzyme obtained from the human heart. One possible mechanism for the suppression of tumor cell growth may be inhibition of membrane
T
0.6
h
ffl ffl
ra \
.-c E
u
I
0
0.5
I
1 .o
I
2.0
H a t t a l i n concentration (M)
Figure 6. Effect of hattalin (A)or ouabain (0)on membrane ATPase Figure 5. Effect of hattalin (A)or ouabain (0)on membrane Na+, K+-ATPase from canine kidney cortex.
from cultured ZR-75-1 cells. Percent activity of the enzyme remaining in the presence of each concentrationof hamiinor ouabain was measured. The activity in the absence of drug was designated as 100%. Bars indicate SD of the triplicate assay.
Lignan Effect on a Breast Carcinoma Cell
601
B
A
T
-
0.6
h
1
VI
m VI \
.-c: E
. 0
.r
n.
-5 0
0.3
%
I
CI .r
.-> w
m u
w
VI
m
a
