- Email: [email protected]
Formosanin-C, an immunomodulator with antitumor activity
0192-0561/90 $3.00 + .00 Pergamon Press plc. International Society for Immunopharmacology.
Int..1. lmmunopharmac., Vol. 12, No. 7, pp. 777-786, 1990. Printed in Great Britain.
FORMOSANIN-C, AN IMMUNOMODULATOR WITH ANTITUMOR ACITIVITY RONG-TSUN W u , *~ HSUCH-CHING CHIANG, ~: WAN-CHYUNG FU,* KWANG-YU CHIEN, § YU-MEI CHUNG H and LIN-YEA HORNGt *Graduate Institute of Microbiology & Immunology, National Yang-Ming Medical College & National Research Institute of Chinese Medicine; *School of Pharmacy, National Taiwan University; ~Department of Thoracic Surgery, Veterans General Hospital; "Department of Ophthalmology, National Yang-Ming Medical College & Laboratory of Eye Immunology, Veterans General Hospital, Taipei, Republic of China
(Received 16 October 1989 and in final form 11 April 1990) Abstract - - Par&formosana Hayata (Liliaceae) grown in the mountain areas of Taiwan, has been used as a folk remedy for snake bite, and as an anti-inflammatory or anti-neoplastic agent. The effects of formosanin-C, a diosgenin saponin isolated from Paris formosana, on immune responses and transplantable murine tumor were studied. In culture systems, formosanin-C (0.03-0.16/aM) displayed significant enhancement of the blastogenic response of human peripheral blood cells to phytohemagglutinin. Formosanin-C also significantly increased the 3H-thymidine incorporation of ConA-stimulated lymphocytes at concentrations of 0.1 and 0.01 /aM. The responsiveness of the granulocyte/macrophage colony-forming cells (GM-CFC) to mouse fibroblast cells L929 conditioned medium was altered in the presence of 0.01 and 0.001 /aM of formosanin-C. In addition, formosanin-C given intraperitoneally activated natural killer cell activity at doses of 1 - 2 . 5 mg/kg. An intraperitoneal injection of 2.5 mg/kg of formosanin-C markedly induced interferon production, the peak blood level of which was observed 24 h after formosanin-C injection. Growth of subcutaneously transplanted MH134 mouse hepatoma was retarded by intraperitoneal treatment with 1-2.5 mg/kg of formosanin-C. The activity of 5-fluorouracil against MH-134 mouse hepatoma was potentiated by intraperitoneal treatment with formosanin-C. These results suggest that formosanin-C might display antitumor activity in association with modification of the immune system.
cells in tissue culture and has anti-neoplastic activity in tumor-bearing animals (Li, Koo & Hsu, 1972). Apart from being isolated as the m a j o r principle from Paris formosana Hayata, formosanin-C was also identified as a component of " Y u n n a n Bai Y a o " , and defined as a cytotoxic saponin (Ravikumar, H a m m e s f a h r & Sih, 1979). The Chinese herbal preparation " Y u n n a n Bai Y a o " has been used as a hemostatic agent and gained wide acceptance. Claims regarding its efficacy in promoting wound healing have also been made. Rumors have circulated among the overseas Chinese that " Y u n n a n Bai Y a o " could be used in cancer therapy. We found that formosanin-C exhibits cytotoxic action at concentrations above 5/~M in tissue culture, but shows unique i m m u n o m o d u l a t i n g activities at non-cytotoxic concentration. Its unique immunomodulating properties and antitumor activity were further delineated in this study.
M a n y medicinal plants are traditionally used against certain forms of cancer in China. One of the mechanisms for such treatment was proposed to be the modulation of the host immune system. In the course of our screening for immunopotentiators from Chinese herbs, we observed that formosanin-C, a diosgenin saponin isolated from Paris formosana Hayata (Liliaceae) (Lay & Chiang, 1980), enhances the blastogenic response of h u m a n peripheral blood cells to phytohemagglutinin in vitro. Paris formosana Hayata (Liliaceae) is a perenial herb widely grown in humid m o u n t a i n areas of Taiwan. This plant has been used as a folk remedy for snake bite, and as an anti-inflammatory or anti-neoplastic agent. It is recorded in Chinese literature that Paris formosana has been used to treat malaria, helminthic infections and snake bite (Li, 1597). In recent years, it has been found that crude extract of the plant, has cytotoxic effect on several strains of h u m a n cancer *To whom all correspondence should be addressed. 777
778
RONG-TSuN Wo et al. H
P#B
.0~ ;/v
....CHa
RO 4
group received 0.7% ethanol in phosphate buffered saline.
Mice Male C3H/HeN mice ( 8 - 1 0 weeks), obtained from the Animal Center of National Taiwan University, were used for experiments on immune response and tumor transplantation. Animals were kept in an animal chamber with laminar air-flow at a room temperature of 23 + 1°C, and relative humidity of 55 _+ 5%.
4
R = c ( - L - r h o . p y r - c~-L-rhe.pyr- 2 # - D - g l c . p y r ~-L-rha.pyr / Fig. l. Chemical structure of formosanin-C.
EXPERIMENTAL PROCEDURES
Agents Formosanin-C (Fig. 1), diosgenin 3-O-a-L-rhamnopyranosyl-(l~4)-a-L-rhamnopyranosyl-(1-*4)-[a-L-rhamnopyranosyl-(l~2)]-/3-D-glucopyranoside, was purified from Paris formosana Hayata (Liliaceae) according to the method described elsewhere (Lay & Chiang, 1980). Briefly, the dried leaves of this plant were treated with methanol, and the extract was then partitioned between petroleum ether and water. The aqueous layer was further extracted with chloroform, and this portion was partitioned between n-butanol and water. The n-butanol portion was then applied to a silica gel column, and eluted with C H z C l z - M e O H - w a t e r (8:2:0.1). Fraction 2 was repeatedly applied on a silica gel column and eluted with CH2C12- n-butanol - MeOH - water (40:10:5:1). Recrystallization of fraction 2e from methanol gave colorless needles of formosanin-C. Crystallized formosanin-C (m.p. 204-206°C) was dissolved in 70% ethanol, and sterilized with 0.22/am durapore membrane (Millipore, Bedford, MA, U.S.A.). The sample was subsequently sealed with an autoclaved milliwrap membrane (Millipore, Tokyo, Japan) to evaporate the solvent in a concentrator centrifuge (Univapo 150 H, Uni Equip, West Germany). The sterilized powder was resuspended in RPMI-1640 (Gibco, NY, U.S.A.) with 10% FCS (Gibco) to a concentration of 100/aM in all tissue culture studies. For studies on antitumor activity in animals, formosanin-C was dissolved in 70% ethanol to a concentration of 5 raM, sterilized with 0.22/am durapore membrane, and diluted with sterilized 0.7% ethanol in phosphate buffered saline (PBS). The control
Proliferative response o f human peripheral whole blood to phytohemagglutinin Enhancement of phytohemagglutinin (PHA)stimulated whole blood assay was carried out following the method previously described (Wu, Chen, Tien & Tanaka, 1980). Human peripheral blood was obtained from healthy adults. The whole blood was suspended in nine-fold volume of RPMI-1640 medium (Gibco, NY, U.S.A.), containing 5% heatinactivated fetal calf serum (F.C.S., Gibco), 100 units/ml benzylpenicillin (Sigma Chemical Co, St Louis, MO, U.S.A.), 100/ag/ml streptomycin sulfate (Sigma), and 5/ag/ml PHA (Sigma). This suspension was distributed (0.2 ml each) to a 96-well microplate (Costar, MA, U.S.A.) with or without formosanin-C. The blood cells were incubated at 37°C for 48 h in 5% CO~ and 95% air, and 3H-thymidine (0.2/aCi/well) was then incorporated for 24 h. The cells were collected and washed with distilled water in a multiple automated cell harvester (Bellco, NJ, U.S.A.), and the radioactivity was determined in a liquid scintillation counter (1211 rackbeta, LKB, Sweden). Proliferative response o f mouse lymphocytes to concanavalin A Fresh splenocytes (1 × 106 cells/ml) from male C3H/HeN mice were suspended in a medium containing RPMI-1640 with 0.1 mM nonessential amino acids, 2 x 1 0 - 6 M 2-mercaptoethanol, 100 units/ml benzylpenicillin, 100/ag/ml streptomycin, and 10% heat-inactivated fetal calf serum, in addition to a final concentration of 3/~g/ml of concanavalin A (Sigma). This medium (180/al) was then added to each well of flat-bottomed microplates (Costar) with or without formosanin-C, and was incubated for 3 days at 37°C in a humidified atmosphere containing 5°7o CO2. 3H-Thymidine (0.2/aCi) was subsequently added to each well, 18 h before cell harvest. The incorporated radioactivity was measured using a
Formosanin-C, an Immunomodulator with Antitumor Activity liquid scintillation counter. These experiments were performed in quadruplicates.
Granulocyte - macrophage activity
colony
stimulating
The proliferative responses of granulocytemacrophage colony were evaluated on the basis of 3H-thymidine incorporation. Mouse fibroblast L929 cell conditioned medium was prepared as described elsewhere (Hines, 1983) with minor modification. L929 cells (1 × 106) were transferred to a 75 sq cm 2 tissue culture flask and cultured for 5 days with 10°70 FCS at 37°C in 5°7o CO2 and 95°/0 air. Confluent cells were fed with fresh medium, and conditioned medium was removed after 24 h, filtrated with a millipore 0.22 lam membrane, and stored at - 2 0 ° C until use. Male C 3 H / H e N mice were killed by cervical dislocation. Femoral bone marrow cells for culture and as a source of granulocyte/macrophage progenitor cells were obtained by flushing the marrow cavity with RPMI-1640 medium using a 26gauge needle. The cells (4 × 105/ml) were cultured in a medium containing RPMI-1640 with 5 × 10 6 M 2-mercaptoethanol, 100 units/ml benzylpenicillin, 100/ag/ml streptomycin, and 5% heat-inactivated fetal calf serum, in addition to a final 5°/0 v / v concentration of mouse fibroblast L929 conditioned medium. This medium (180/A) was then added to each well of flat-bottomed micro-plates (Costar) with or without formosanin-C, and was incubated for 4 days at 37°C in a humidified atmosphere containing 507o CO2. 3H-Thymidine (0.4/aCi) was subsequently added to each well 18 h before cell harvest. The cells were frozen and thawed three times, collected, and washed with normal saline and 5°7o cold trichloroacetic acid (E. Merck, Darmstadt, West German) in a multiple automated cell harvester. The incorporated radioactivity was measured using a liquid-scintillation counter. All assays were done in quadruplicates.
Natural killer cytotoxicity assay The antitumor efficacy of natural killer cells was assessed by a standard 4-h SlCr release assay. Male C3H / H e N mice (8 weeks old) were injected intraperitoneally with 1 or 2.5 mg/kg/day of formosanin-C for two days. After 24 h, animals were sacrificed and spleen cells removed from formosanin-C treated or untreated groups were used as effector cells. Each group consisted of four mice. The spleen cells were depleted of adherent cells by incubation in plastic Petri dishes in RPMI-1640 with 5°70 FCS for 1 h at
779
37°C in 5°7o CO2 atmosphere. The mouse lymphoma cell line YAC-1 was used as target cells. The cells were maintained in RPMI-1640 medium with 10% FCS, in a 5o70 COz incubator at 37°C. The target cells were washed twice with RPMI-1640 medium and adjusted to a concentration of 2 x 106/ml in RPMI1640 with 5°70 heat-inactivated FCS and labeled with 100 taCi of sodium chromate (NEN, Du Pont, MA, U.S.A.) for 60 min at 37°C. After washing three times with cold RPMI-1640, these target cells were resuspended to a concentration of 1 × 105/ml in RPMI-1640 with 5o/o FCS. Effector cells (0.1 ml) of nonadherent spleen cells were mixed with 1 × 1 0 4 (0.1 ml) target cells at a ratio of effector cells to target cells (E:T) of 200:1, 100:1, 50:1 in the microwells of a round-bottomed 96-well microplate (Nunclon, Nunc, Denmark). After 4 h of incubation, the supernatant from each well was collected, and the radioactivity measured using a gamma counter (LKB, Rack-gamma counter, Sweden). The natural killer cytotoxic activity was expressed as follows: counts/min of experimental releasecounts/min of spontaneous release % cytolysis =
× 100. counts/min of maximal releasecounts/min of spontaneous release
Where experimental counts/min was the 5'Cr released from injured targets, spontaneous counts/ rain was the spontaneous released from the targets, and maximal CPM the total counts/min in the target cells lysed with 0.5°70 Triton × 100.
Interferon (IFN) induction assay Male C 3 H / H e N mice (8 weeks old) were administrated intraperitoneally with 1, 2.5, or 5 m g / k g / d a y of formosanin-C and the sera were collected after 6, 24, or 48 h. Each group consisted of four mice. The IFN level was assayed by the cytopathic effect (CPE) of vesicular stomatitis virus (VSV) in a 96-well microplate (Costar). Each well contained 0.1 ml of two-fold-diluted serum and 0.1 ml of 7 × 104 L-929 cells in RPMI-1640 (Gibco) with 10% fetal calf serum. The microplate was kept at 37°C for 2 h in 5% COz and 95% air. After washing with the RPMI1640 medium, VSV of 100 IDs0 in 0.1 ml of RPMI1640 with 1% calf serum was added to the cells. The infected cells were further incubated for 24 h, in which controls reached the peak of CPE. The reciprocal of serum dilution, which showed 50°7o inhibition of CPE, was taken as IFN potency. The potency was corrected for international unit by comparison with reference IFN: N I H (L-929-NDV).
780
RONG- TSUN W u et al.
A n titumoractivity against mouse MH-134 hepatoma The MH-134 hepatoma was maintained in solid form by serial s.c. transplantation in syngeneic C3H/HeN male mice. When MH-134 hepatoma reached a size of approximately 10 mm diameter, the tumor tissue was excised and trimmed free of necrotic areas, then a cell suspension was prepared by mechanical means in Hank's balanced salt solution (HBSS). The viability was checked by the Trypan-blue exclusion test. For the therapeutic experiments, C3H/HeN mice received a s.c. injection of 4 x 105 hepatoma cells/0.2 ml HBSS/mouse on day 0. After tumor inoculation, mice were randomly assigned to experimental and control groups. Each group contained 10 mice. Formosanin-C was administered i.p. starting on day 1 and was given every day or every other days until the termination of the experiment. The antitumor effect of the combination treatment of formosanin-C with 5-fluorouracil (5-FU) was also studied. 5-FU was given daily i.p. for the first five days and formosanin-C was given i.p. every two days from day 7 until the termination of the experiment. 5-FU or formosanin-C given from day 1 in every other day was also scheduled. Control mice received the same volume of sterile distilled water or 0.7% ethanol in PBS with the same treatment schedule. The therapeutic response was measured by the weight of the tumor masses excised from the sacrificed mice. Stat&tical analysis The significance of the differences between drugtreated groups and control groups was analyzed using Student's t-test.
RESULTS
Proliferative response o f human peripheral whole blood to phytohemagglutinin At the concentrations of 0.8/aM, 0.16/aM and 0.032/aM (Fig. 2), formosanin-C dose-dependently enhanced the proliferative response of human peripheral whole blood to PHA, as evaluated by 3H-thymidine incorporation. The maximal stimulatory effect occurred at 0.16/aM, when the counts/min of 3H-thymidine was almost three times higher than the control (11000 + 1350 vs 3900 + 760). Increasing the concentration of formosanin-C to 4/aM did not elicit further effect on the 3H-thymidine incorporation of human peripheral whole blood. At 20/aM, it actually produced a hemolytic effect.
15 Io 12 x
E 9 o_ £)
v
6
rY ~3 F--
0
11
0.032 0.16 0.8 Concentrotion (/~M)
4.0
Fig. 2. Effect of formosanin-C in vitro on mitogenic responses of PHA-stimulated human peripheral whole blood cells from healthy adult: dependency of drug concentration. Each point represents an average of quadruplicate cultures (*P<0.001).
Proliferative response o f mouse lymphocytes to concanavalin A Formosanin-C significantly increased 3H-thymidine incorporation of ConA-stimulated lymphocytes (Fig. 3). The maximal concentration (0.1 /aM), at which a stimulatory effect was produced, was similar to that (0.16/aM) in the proliferative response of human peripheral whole blood of PHA. An inhibitory effect on 3H-thymidine incorporation of ConAstimulated lymphocytes occurred at concentrations of 10 and 1 /aM.
Granulocyte- macrophage colony stimulating activity The responsiveness of granulocyte-macrophage colony-forming cells (GM-CFC) to mouse fibroblast cell L929 conditioned medium was altered by formosanin-C. As shown in Fig. 4, this agent augmented 3H-thymidine incorporation of GM-CFC at concentrations of 0.01 and 0.001 /aM. The concentration of formosanin-C that stimulated the GM-CFCs maximally was 0.01 /aM, which is ten-fold lower than that required to achieve maximal stimulation in ConAstimulated lymphocytes. At concentrations of 1 and 10/aM, formosanin-C produced an inhibitory effect on the 3H-thymidine incorporation of GM-CFC.
Assay f o r natural killer cytotoxicity Formosanin-C (1 mg/kg/day) significantly enhanced the natural killer (NK) activity of C3H/He mice, as assessed by the cytotoxicity of fresh splenocytes from treated animals against NK-sensitive YAC-1 cells (Fig. 5). Moderate augmentation was observed in the formosanin-C (2.5 mg/kg/day)treated group.
781
Formosanin-C, an I m m u n o m o d u l a t o r with A n t i t u m o r Activity 25 ¸
8,
•
,4"
io x
Control
o~o &--A
1.0 n~/kg 2.5 mg/k9
# ,
G X
o
E
,~1
2o
0--0
i-. 15 o
2
"1-
5o":1
0 0
0.01
0.1
Concentrotlon (/zM)
Fig. 3. Effect of formosanin-C in vitro on mitogenic response of C o n A-stimulated splenocytes from normal C 3 H / H e N mice: dependency on drug concentration. Each point represents an average of quadruplicate cultures. (*/'<0.05).
loo:1
2~:1
Effector : Target Ratio
10
1
Fig. 5. Effect of formosanin-C on natural killer cellmediated cytotoxicity in normal C 3 H / H e N mice. Formosanin-C was injected intraperitoneally on day 0 and day 1. Control group was injected with 0.7% ethanol 0.2 ml intraperitoneally on day 0 and day 1. Mice were sacrificed on day 2, and the spleen cells were used as effector cells. YAC-1 l y m p h o m a cells were used as target cells. (*P<0.05, #P<0.001).
10-
i,q
tO
28O 8-
24O
X
E
CL (J
Iz 1o I-I -r
2OO
6-
2-
0 0
i
0.001 0.01
O.
° 1
I
Control
r'~ w e i.o r~/kg FzC 2.5 m g / k 9
I
F=C 5.0 mg/kg
tOO
12o
~_ 8o 40
10
C o n c e n t r a t i o n ~J.M)
i
m~
o
6
rq~, 24
48
Hours
Fig. 4. Effect of formosanin-C in vitro on proliferative responses of granulocyte/macrophage progenitor cells from bone marrow of C 3 H / H e N mice to L-929 cells conditioned medium: dependency on drug concentration. Each point represents the [~H]-thymidine incorporation of an average of quadruplicate cultures. (**P<0.01, *P<0.05).
Fig. 6. Kinetics o f interferon induction by formosanin-C in C 3 H / H e N mice. Mice were administrated intraperitoneally with 1, 2.5, or 5 m g / k g / d a y of formosanin-C and the sera were collected after 6, 24 or 48 h. Each group consisted of four mice. The IFN level was assay by the cytopathic effect of vesicular stomatitis virus. (*P<0.01, **P<0.05).
The interferon induction The effect of formosanin-C on blood level of interferon was examined in C3H/HeN mice (Fig. 6). A n i n t r a p e r i t o n e a l i n j e c t i o n o f 2.5 m g / k g o f f o r m o -
sanin-C markedly induced interferon production. The peak blood level of interferon was observed 24 h after formosanin-C injection. The higher dose of formosanin-C treatment (5 mg/kg) did not produce any significant interferon inducing effect.
Table 1. Effect o f formosanin-C on MH-134 mouse hepatoma Formosanin-C (mg/kg/day) -1.0 2.5 5.0
T u m o r weight (g) 3.53 3.17 1.10 2.98
_ ± _ +
1.29 1.37 0.52 1.59
Inhibition (°7o) 0 10 69 (P
A n t i t u m o r activity against mouse M H - 1 3 4 hepatoma W e e x a m i n e d w h e t h e r f o r m o s a n i n - C c o u l d elicit inhibition against mouse MH-134 hepatoma. MH134 h e p a t o m a cells (4 × 105 cell) w e r e i n o c u l a t e d to the dorsal region of C3H/HeN mice (male, 8 weeks old) at d a y 0 a n d f o r m o s a n i n - C w a s g i v e n d a i l y (i.p.)
A suspension of 4 × 105 cells was implanted subcutaneously in the dorsal region of male C 3 H / H e N mice ( 8 - 1 0 weeks old). Mice were treated with formosanin-C i.p. every day from day 1 to day 11 after t u m o r inoculation. The weight of t u m o r was determinated by autopsy on day 15.
782
RONG-TSUNWU et al.
Table 2. Effect of formosanin-C on MH-134 mouse hepatoma Formosanin-C (mg/kg/day) --
1.0 2.5 5.0
Tumor weight (g)
Inhibition (%)
3.38 _+2.09 2.09 _+0.63 0.76 _+0.33 1.79 _+0.11
0 38 78 (P<0.01) 48 (P<0.05)
A suspension of 4 × 105 cells was implanted subcutaneously in the dorsal region of male C3H/He mice (8-10 weeks old). Mice were treated with formosanin-C i.p. every other day from day 1 to day 13 after tumor inoculation. The weight of tumor was determinated by autopsy on day 17. from day 1 for 11 days. Tumors were extirpated and weighed 15 days after tumor inoculation. As shown in Table 1, formosanin-C treatment (2.5 mg/kg/ day) was effective in suppressing the growth of tumor. At a dose of 5 mg/kg/day, formosanin-C was not only ineffective in regressing the growth of tumor, but was somewhat toxic to the mice by reducing their body weight. In a different treatment schedule, formosanin-C (i.p.) was administrated every other day from day 1 to day 13. Tumor masses were excised and weighed 17 days after tumor inoculation. As shown in Table 2, there was no dosedependency, and 2.5 mg/kg/every other day i.p. seemed to be the optimal dose to suppress the growth of tumor. The 5 mg/kg/every other day i.p. treatment was also effective, and produced no further weight loss. This was quite different from the 5 mg/kg/day i.p. group (Table 1). We also studied the antitumor activity of formosanin-C in combination with a clinically used antitumor drug, 5-fluorouracil (5-FU). 5-FU was given daily for 5 days starting 1 day after the inoculation of 4 x 105 hepatoma cells. Formosanin-C was given every other day from day 7 to day 19. The weight of the tumor was determined on the 20th day. Treatment with formosanin-C (2.5 mg/kg) or 5-FU (20 mg/kg) alone inhibited growth of the tumor by 37% and 28% respectively (Table 3). A combined treatment with formosanin-C (2.5 mg/kg) and 5-FU (20 mg/kg), however, inhibited growth of the tumor by 73%. In another treatment schedule, after 5 x 10~cells were s.c. transplanted to the C3H/HeN mice, 5-FU was injected i.p. every other day from day 1 to day 13, and formosanin-C was injected i.p. every other day from day 2 to day 12. On day 14, the tumor was extirpated and weighed. As shown in Table 4, the treatment of formosanin-C (2.5 mg/kg) or 5-FU (25 mg/kg) alone inhibited growth of the
tumor by 45% and 33% respectively, the treatment of formosanin-C (2.5 mg/kg) and 5-FU (25 mg/kg) inhibited growth of the tumor by 64%. DISCUSSION
The present study revealed that formosanin-C produced dual effects on the proliferative response of human peripheral whole blood to PHA, proliferative response of mouse lymphocytes to concanavalin A, and proliferative response of mouse granulocyte-macrophage colony-forming cells to mouse fibroblast cell L929 conditioned medium. Maximal stimulatory effects were produced on the above responses at concentrations of 0.16, 0.1 and 0.01 ~M respectively (Figs 2, 3 & 4). An inhibitory action on the same responses was observed, however, at concentrations of 4, l and 1 gM respectively (Figs 2, 3 & 4). These results indicated that formosanin-C was non-specifically cytotoxic to both lymphoid and tumor cells at concentrations above 1 gM. Formosanin-C enhanced the responsiveness of the granulocyte-macrophage colony-forming cells (GM-CFC) to mouse fibroblast cells L929 conditioned medium at the very low concentration of 0.01 /~M. This effect may be mediated by directly stimulating GM-CFC proliferation and/or stimulating production of granulocyte-macrophage colony stimulating factors (GM-CSFs). The GM-CSFs probably play an important role in hematopoietic generation of GMprogenitor cells and in differentiation of myeloid leukemic cells (Metcalf, 1985; Schlick & Ruscetti, 1986; Begley, Metcalf & Nicola, 1987). In addition, formosanin-C given intraperitoneally enhanced the NK activity of C3H/HeN mice, the maximum effect was observed at a dose of 1 mg/kg (Fig. 5). In light of the possible role of NK cells in cancer defense (Herberman & Ortaldo, 1981), numerous efforts have been made in NK cell-augmenting effect. (Ching & Baguley, 1987; Goodman, 1988). An intraperitoneal injection of 2.5 mg/kg of formosanin-C markedly induced interferon production, and the peak blood level of interferon was observed 24 h after formosanin-C injection (Fig. 6). Studies in animals and man also suggest the effectiveness of IFN (Goldstein & Laszlo, 1986; Gutterman, 1988; Jaffe & Herberman, 1988; Figlin, 1988) and IFN inducer (Pollard, 1982; Lotzova, Savary & Stringfellow, 1983; Hornung, Young, Urba & Wiltrout, 1988) therapy in cancer. Thus, the endogenous interferon induced by formosanin-C might offer another applicability in cancer therapy, which could avoid the undesirable side effects (Mannering & Deloria, 1986), fast serum
783
Formosanin-C, an Immunomodulator with Antitumor Activity Table 3. Activity of formosanin-C in combination with 5-fluorouracil on MH-134 mouse hepatoma Formosanin-C (mg/kg/day)
5-FU (mg/kg/day)
-1.0 2.5 5.0 -1.0 2.5 5.0
Tumor weight (g)
----20 20 20 20
5.26 4.51 3.32 3.73 2.79 2.72 1.44 2.38
Inhibition (%)
__. 2.33 _ 2.36 __. 1.15 _+ 1.53 _+ 1.58 _+ 1.15 _ 0.72 _+ 0.86
0 15 37 29 28 48 73 55
(P
(P<0.01) (/~0.001) (P<0.01)
A suspension of 4 x 105 cells was implanted subcutaneously in the dorsal region of male C 3 H / H e N mice ( 8 - 10 weeks old). Mice were treated with 5-FU i.p. every day from day 1 to day 5, and injected with formosanin-C i.p. every other day from day 7 to day 19 after tumor inoculation. The weight of tumor was determinated by autopsy on day 20.
T.I * I
Day
I
*
*
*
*
I
I
I
I
0
,~ I
I
¢
~, i
I
I
5
,1, I
I
I
~, I
10
¢
I
I
¢ I
I
T.E. I
I
15
20
*5-Fluorouracil (20 mg/kg/day); +: Formosanin-C (1.0, 2.5 mg/kg/day); T.I: Tumor inoculation; T.E: Tumor excision.
Table 4. Activity of formosanin-C in combination with 5-fluorouracil on MH-134 mice hepatoma Formosanin-C (mg/kg/day)
5-FU (mg/kg/day)
-1.0 2.5 --1.0
Tumor weight (g)
---10 25 10 25 10 25
1.0 2.5 2.5
2.44 1.60 1.35 2.50 1.64 1.14 1.32 1.45 0.87
_+ 0.70 _+ 0.73 + 0.49 _+ 1.08 + 0.86 ___0.56 _+ 0.41 _+ 0.86 _+ 0.49
Inhibition (%) -34 45 0 33 42 46 41 64
(P<0.05) (P<0.001) (P<0.05) (P<0.001)
(P<0.001) (P<0.05) (P<0.001)
A suspension of 5 x 104 cells was implanted subcutaneously in the dorsal region of male C H / H e N mice (8 - 10 weeks old). 5-FU was injected i.p. every other day from day 1 to day 13, and formosanin-C was injected i.p. every other day from day 2 to day 12 after tumor inoculation. The weight of tumor was determinated by autopsy on day 14. T.I * ¢ I
Day
0
I
I
* ~
* ¢
I
I
I
I
* ~
* ¢
* ¢
*
I
I
I
I
I
5
I
10
I
T.E. I
I
15
*5-Fluorouracil (10, 25 mg/kg/day); ~: Formosanin-C (1.0, 2.5 mg/kg/day); T.I: Tumor inoculation; T.E: Tumor excision.
clearance and antibody occurrence (Hawkins, Horning, Konrad, Anderson, Sielaff, Rosno, Schiesel, D a v i s , D e M e t s , M e r i g a n et al., 1986) o f
e x o g e n o u s interferon t h e r a p y . W h e t h e r f o r m o s a n i n - C a u g m e n t e d N K activity via t h e m e c h a n i s m o f i n t e r f e r o n s i n d u c t i o n n e e d s to be f u r t h e r clarified.
784
RONG-TSUNWu et al.
Formosanin-C retarded growth of s.c. solid MH-134 hepatoma in C3H/HeN mice (Tables 1 and 2), but was not effective against ascites form MH-134 hepatoma in C3H/HeN mice (data not shown). Although the maximum effective dose was 2.5 mg/kg in both administration schedules, the every-otherday schedule of 5 mg/kg was effective (Table 2, P<0.05), while the every day schedule of 5 mg/kg was not effective (Table 1). Formosanin-C was also effective in enhancing the antitumor activity of 5-FU against s.c. solid mice MH-134 hepatoma (Table 3), but appeared to be not obviously effective in another combination schedule (Table 4). Optimization of the capability of immune response modification therefore seemed to be dependent on the method of administration. The s.c. solid MH-134 hepatoma seems to be a slow-growing tumor in which defense mechanisms are somewhat preserved, and formosanin-C could activate them in time. It is therefore reasonable to conclude that formosanin-C may display antitumor activity in association with modification of the immune function, especially cellmediated immunity and/or secretion of soluble factors, but not with its cytotoxic effect. A cytotoxic agent podophyllotoxin has also recently been found to augment macrophage proliferation at a low concentration (Zheng, Wiranowska, Sadlik & Hadden, 1987). Besides its well known cytotoxic effect, a low concentration of the cytotoxic drug 1-p-D-arabinofuranosylcytosine (Ara-C) was reported to produce differentiation-inducing activity of leukemic cells (Luisi-Deluca, Mitchell, Spriggs & Kufe, 1984; Wu et al., 1985). These findings indicate the importance of biological activities at low concentrations in the evaluation of a biologically active substance. Saponins are natural compounds that exist in varying proportions in many plants. They usually occur in nature as mixtures of related compounds, which are chemically divided into either triterpenoid or steroidal glycosides (Oakenfull, 1981). Some saponins show various biological activities. For example, the crude saponin extract of ginseng root has a promoting effect on neurite extension primary in cultured neurons of the rat cerebral cortex (Sugaya, Yuzurihara, Tsuda, Yasuda, Kajiwara & Sugaya, 1988). A purified saponin, Ginsenoside Rh2, isolated from Panax ginseng inhibits the growth of B16 melanoma cells, causes morphological alterations, and stimulates melanogenesis (Odashima, Ohta, Kohno, Matsuda, Kitagawa, Abe & Arichi, 1985). Some commercially available saponins have been used as adjuvants along with parenterally administered experimental and parasitic vaccines (Dalsgaard, 1974; Bomford, 1982; Scott, Bahr,
Moddaber & Chedid, 1984). The adjuvant activity of saponins is considered to be due to their prolonging retention of antigen and its localization in the spleen (Scott, Goss-Sampson & Bomford, 1985). Recent studies showed that orally administered Quillaja saponin potentiates the humoral and cellular immune responses in mice fed with inactivated rabies vaccine (Chavali & Campbell, 1987; Maharaj, Froh & Campbell, 1986). Since orally fed saponins do not enter circulation (Sidhu & Oakenfull, 1986), a mechanism has been proposed that Quillaja saponin might affect the mucosal immune system, which causes activation of helper T-cells and induces secretion of soluble factors (Chavali & Campbell, 1987; Chavali, Francis & Campbell, 1987; Maharaj, Froh & Campbell, 1986). We also examined whether formosanin-C fed orally could exhibit inhibition against mouse MH-134 hepatoma. At oral doses of 2, 6, 15 mg/kg/day, formosanin-C (20 days)-treated group was ineffective in regressing the growth of the tumor (unpublished data). In addition, there was no weight loss in mice given formosanin-C orally. Although our present results only yield an incomplete picture of the effects of formosanin-C on the immune system, we have identified that at very low concentrations, formosanin-C has the capacity to potentiate the immune system, which might be related to its unique antitumor activity. The mechanisms whereby formosanin-C modulates the immune system still remain unclear. The surfactant nature of saponins is considered to induce changes at the cell surface membrane (Alvarez & TorresPinedo, t982). The selective effect of saponins to different lineages of immune cells is likely to depend oll the chemical stucture of the individual saponins and the nature of the cell surface. The selectivity of formosanin-C to immune cells needs to be clarified further. It has long been recognized that modulation of the immune system by various agents may have potential for the treatment of certain neoplastic disease (Hadden, 1983; Oldham, 1983; Reizenstein & Mathe, 1983; Wagner, 1983; Mihich, 1986; RuszalaMallon, Lin, Durr & Wang, 1988). Formosanin-C studied here is a plant saponin and is a possible novel immunomodulatory substance. One of the most important means of our studies is to clarify scientifically the effect mechanism of our traditional Chinese folk medicine, as Paris formosana Hayata (Liliaceae) and "Yunnan Bai Yao". To re-examine the traditional Chinese medicines that are said to be effective against cancer by approaches different from cytotoxic mechanism, such as induction of tumor necrosis factor (Katsuyuki, Nobuko, Akiko, Ruriko, Nakao & Mosaburo, 1985), might also offer another
Formosanin-C, an Immunomodulator with Antitumor Activity new way for the arrest o f cancer. Further studies on the effector mechanisms related to the immunomodulatory activity o f formosanin-C are in progress. Acknowledgements - - This study was supported in part by research grant NSC-75-0610-B010-12 & NSC-76-0610-
785
B010-15 from National Science Council, DOH75-0302-38 from Ministry of Health, Republic of China, and in part by the projects from the Memory Foundation of Dr K-S Lu. The authors express their deep thanks to Dr G. Chihara for his helpful discussion.
REFERENCES
ALVAREZ, J. R. & TORRES-PINEDO, R. (1982). Interactions of soybean lectin, soyasaponins and glycinin with rabbit jejunal mucosa in vitro. Pediat. Res., 16, 728-731. BEGLEY, C. G., METCALF, D. & NICOLA, N. A. (1987). Purified colony stimulating factors (G-CSF and GM-CSF) induce differentiation in human HL60 leukemic cells with suppression of clonogenicity. Int. J. Cancer, 39, 9 9 - 105. BOMFORD, R. (1982). Studies on the cellular site of action of the adjuvant activity of saponin for sheep erythrocytes. Int. Arch. Allergy Appl. Immun., 67, 127- 31. CHAVALI, S. R. & CAMPBELL,J. B. (1987). Adjuvant effects of orally administered saponins on humoral and cellular immune responses in mice. Immunobiology, 174, 347-359. CHAVALI, S. R., FRANCIS, T. & CAMPBELL, J. B. (1987). An in vitro study of immunomodulatory effects on some saponins. Int. J. Pharmac., 9, 675- 683. CHING, L. U . & BAGULEY, B. C. (1987). Induction of natural killer cell activity by the antitumour compound flavone acetic acid. Eur. J. Cancer clin. Oncol., 23, 1047 - 50. DALSGAARD, K. (1974). Saponin adjuvants. III. Isolation of a substance from Quillaja saponaria Molina with adjuvant activity in food-and-mouse disease vaccines. Arch. ges. virusforsch., 44, 243- 254. FIGLIN, R. A. (1988). Biotherapy with interferon. Semin. Oncol., 15, (Suppl. 6) 3 - 9 . GOLDSTEIN, D. & LASZLO,J. (1986). Interferon therapy in cancer: from imaginon to interferon. Cancer Res., 46, 4315 - 29. GOODMAN, S. (1988). Therapeutic effects of organic germanium. Med. Hypotheses, 26, 207 - 15. GUTTERMAN, J. U. (1988). The role of interferons in the treatment of hematologic malignancies. Semin. Hemat., 25, 3 - 8. HADDEN, J. W. (1983). Chemically defined immunotherapeutic agents. Prog. Clin. Biol. Res., 132, 273- 86. HAWKINS, M., HORNING, S., KONRAD,M., ANDERSON, S., SIELAFF,K., ROSNO, S., SCHIESEL, J., DAVIS, T., DEMETS, D., MERIGAN, T., et al. (1985). Phase I evaluation of a synthetic mutant of beta-interferon. Cancer Res., 45, 5914-20. HERBERMAN, R. B. & ORTALDO, J. R. (1981). Natural killer cells: their roles in defenses against disease. Science, 214, (4516) 2 4 - 30. HINES, D. (1983). Lipid accumulation and production of colony-stimulating activity by the 266AD cell line derived from mouse bone marrow. Blood, 61, 397-402. HORNUNG, R. L., YOUNG, H. A., URBA, W. J. & WILTROtJT, R. H. (1988). Immunomodulation of natural killer cell activity by flavone acetic acid: occurrence via induction of interferon alpha/beta. J. natn. Cancer Inst., 80, 1226 - 31. JAFFE, H. S. & HERBERMAN,R. B. (1988). Rationale for recombinant human interferon-gamma adjuvant immunotherapy for cancer. J. natn. Cancer Inst., 80, 616-8. LAY, J. Y. & CHIANG, H. C. (1980). Studies on the constituents of Paris formosana Hayata. J. Taiwan Pharmaceut. Asso., 32, 14-27. KATSUYUKI,H., NOBUKO,S., AKIKO,A., RURIKO,H., NAOKO, O. & MOSABURO,K. (1985). Antitumor activities and tumor necrosis factor producibility of traditional Chinese medicines and crude drugs. Cancer Immun. Immunother., 20, 1-5. LI, M. C., Koo, W. Y. & Hsu, K. P. (1972). Anti-neoplastic property of a crude extract from Parisformosana. Nature New Biol., 235, 223 - 224. LI, T. C. (1597). Pen-ts'ao Kang-mu. Chinese Meteria Medica, 16, 698. LOTZOVA, E., SAVARY,C. A. & STRINGFELLOW,D. A. (1983). 5-halo phenyl pyrimidinones: new molecules with cancer therapeutic potential and interferon-inducing capacity are strong inducers of murine natural killer cells. J. Immun., 130, 965 - 9. LUIS1-DELUCA,C., MITCHELL, T., SPRIGGS,D. & KUFE, D. W. (1984). Induction of terminal differentiation in human k562 erythroleukemia cells by arabinofuranosyl-cytosine. J. clin. Invest., 74, 821- 827. MAHARAJ,I., FROH,K. J. & CAMPBELL,J. B. (1986). Immune responses of mice to inactivated rabies vaccine administered orally: potentiation by Quillaja saponin. Can. J. Microbiol., 32, 4 1 4 - 420. MANNERIN~, G. J. & DELORIA, L. B. (1986). The pharmacology and toxicology of the interferons: an overview. Ann. Rev. Pharmac. Toxic'., 26, 455-515. METCALF, D. (1985). The granulocyte-macrophage colony-stimulating factors. Science, 229, 16-22. MIHICH, E. (1986). Future perspectives for biological response modifiers: a viewpoint. Semin. Oncol., 13, 234-54.
786
RONG-TsuN WU et al.
OAKENFULL, D. G. (1981). Saponins in food - - a review. Food Chem., 6, 19-41. ODASHIMA, S., OHTA, T., KOHNO, H., MATSUDA,T., KITAGAWA,I., ABE, H. & ARICHI, S. (1985). Control of phenotypic expression of cultured B16 melanoma cells by plant glycosides. Cancer Res., 45, 2781 -2784. OLDHAM, R. K. (1983). Biological response modifiers. J. natn. Cancer Inst., 70, 789-796. POLLARD, R. B. (1982). Interferons and interferon inducers: development of clinical usefulness and therapeutic promise. Drugs, 23, 3 7 - 55. RAVIKUMAR,P. R., HAMMESFAHR,P. & S1H, C. J. (1979). Cytotoxic saponins from the Chinese herbal drug Ynnnan Bai Yao. J. Pharmaceut. Sci., 68, 900- 904. REIZENSTEIN, P. & MATHE, G. (1983). Biological response modifiers and differentiation inducers. Current status in cancer research. Drugs, 26, 185- 190. RUSZALA-MALLON,V., LIN, Y. I., DURR, F. E. & WANG, B. S. (1988). Low molecular weight immunopotentiators. Int. J. Immunopharmac., 10, 497 - 510. SCHLICK, E. & RUSCETTI, F. W. (1986). In vivo induction of terminal differentiation of malignant myelopoietic progenitor cells by CSF-inducing biological response modifiers. Blood, 67, 980-987. SCOTT, M. T., BAHR, G., MODDABER,F. & CHEDID, L. (1984). Adjuvant requirement for protective immunization of mice using a Trypanosoma cruzei 90K cell surface glycoprotein. Int. Archs Allergy appl. Immun., "/4, 373 - 377. SCOT]-, M. T., GOsS-SAMPSON,M. & BOMFORD, R. (1985). Adjuvant activity of saponin: antigen localization studies. Int. Arch. Allergy. appl. Immun., 77, 409- 12. SIDHU, G. S. & OAKENrULL,D. G. (1986). A mechanism for the hypocholesterolaemic activity of saponins. Br. J. Nut., 55, 643 - 649. SUGAYA, A., YUZURIHARA, M., TSUDA, T., YASUDA, K., KAJ1WARA, K. & SUGAYA,E. (1988). Proliferative effect of ginseng saponin on neurite extension of primary cultured neurons of the rat cerebral cortex. J. Ethnopharmac., 22, 173 - 181. WAGNER, H. (1983). lmmunoprevention and therapy by plant preparations ZFA (Stuttgart), 59, 1282- 1289. Wu, R. Y. & TANAKA,N. (1980). Effects of antitumor agents on DNA synthesis of mitogen-induced human lymphocytes, in comparison with leukemic cells. J. Antibiotics, 33, 770-771. Wu, R. T., CHEN, S. C., TIEN, W. C. & TANAKA, N. (1985). Potentiation of 1-fl-D-arabinofuranosylcytosine-induced erythroid differentiation of human leukemia K562 cells by cadeguomycin. In Recent Advances in Chemotherapy (ed. Ishigami, J.) pp. 201-202. University of Tokyo Press, Tokyo. ZHENG, Q. Y., WIRANOWSKA,M., SADLIK, J. R. & HADDEN, J. W. (1987). Purified podophyllotoxin (CPH-86) inhibits lymphocyte proliferation but augments macrophage proliferation. Int. J. Immunopharmac., 9, 539-549.
Int..1. lmmunopharmac., Vol. 12, No. 7, pp. 777-786, 1990. Printed in Great Britain.
FORMOSANIN-C, AN IMMUNOMODULATOR WITH ANTITUMOR ACITIVITY RONG-TSUN W u , *~ HSUCH-CHING CHIANG, ~: WAN-CHYUNG FU,* KWANG-YU CHIEN, § YU-MEI CHUNG H and LIN-YEA HORNGt *Graduate Institute of Microbiology & Immunology, National Yang-Ming Medical College & National Research Institute of Chinese Medicine; *School of Pharmacy, National Taiwan University; ~Department of Thoracic Surgery, Veterans General Hospital; "Department of Ophthalmology, National Yang-Ming Medical College & Laboratory of Eye Immunology, Veterans General Hospital, Taipei, Republic of China
(Received 16 October 1989 and in final form 11 April 1990) Abstract - - Par&formosana Hayata (Liliaceae) grown in the mountain areas of Taiwan, has been used as a folk remedy for snake bite, and as an anti-inflammatory or anti-neoplastic agent. The effects of formosanin-C, a diosgenin saponin isolated from Paris formosana, on immune responses and transplantable murine tumor were studied. In culture systems, formosanin-C (0.03-0.16/aM) displayed significant enhancement of the blastogenic response of human peripheral blood cells to phytohemagglutinin. Formosanin-C also significantly increased the 3H-thymidine incorporation of ConA-stimulated lymphocytes at concentrations of 0.1 and 0.01 /aM. The responsiveness of the granulocyte/macrophage colony-forming cells (GM-CFC) to mouse fibroblast cells L929 conditioned medium was altered in the presence of 0.01 and 0.001 /aM of formosanin-C. In addition, formosanin-C given intraperitoneally activated natural killer cell activity at doses of 1 - 2 . 5 mg/kg. An intraperitoneal injection of 2.5 mg/kg of formosanin-C markedly induced interferon production, the peak blood level of which was observed 24 h after formosanin-C injection. Growth of subcutaneously transplanted MH134 mouse hepatoma was retarded by intraperitoneal treatment with 1-2.5 mg/kg of formosanin-C. The activity of 5-fluorouracil against MH-134 mouse hepatoma was potentiated by intraperitoneal treatment with formosanin-C. These results suggest that formosanin-C might display antitumor activity in association with modification of the immune system.
cells in tissue culture and has anti-neoplastic activity in tumor-bearing animals (Li, Koo & Hsu, 1972). Apart from being isolated as the m a j o r principle from Paris formosana Hayata, formosanin-C was also identified as a component of " Y u n n a n Bai Y a o " , and defined as a cytotoxic saponin (Ravikumar, H a m m e s f a h r & Sih, 1979). The Chinese herbal preparation " Y u n n a n Bai Y a o " has been used as a hemostatic agent and gained wide acceptance. Claims regarding its efficacy in promoting wound healing have also been made. Rumors have circulated among the overseas Chinese that " Y u n n a n Bai Y a o " could be used in cancer therapy. We found that formosanin-C exhibits cytotoxic action at concentrations above 5/~M in tissue culture, but shows unique i m m u n o m o d u l a t i n g activities at non-cytotoxic concentration. Its unique immunomodulating properties and antitumor activity were further delineated in this study.
M a n y medicinal plants are traditionally used against certain forms of cancer in China. One of the mechanisms for such treatment was proposed to be the modulation of the host immune system. In the course of our screening for immunopotentiators from Chinese herbs, we observed that formosanin-C, a diosgenin saponin isolated from Paris formosana Hayata (Liliaceae) (Lay & Chiang, 1980), enhances the blastogenic response of h u m a n peripheral blood cells to phytohemagglutinin in vitro. Paris formosana Hayata (Liliaceae) is a perenial herb widely grown in humid m o u n t a i n areas of Taiwan. This plant has been used as a folk remedy for snake bite, and as an anti-inflammatory or anti-neoplastic agent. It is recorded in Chinese literature that Paris formosana has been used to treat malaria, helminthic infections and snake bite (Li, 1597). In recent years, it has been found that crude extract of the plant, has cytotoxic effect on several strains of h u m a n cancer *To whom all correspondence should be addressed. 777
778
RONG-TSuN Wo et al. H
P#B
.0~ ;/v
....CHa
RO 4
group received 0.7% ethanol in phosphate buffered saline.
Mice Male C3H/HeN mice ( 8 - 1 0 weeks), obtained from the Animal Center of National Taiwan University, were used for experiments on immune response and tumor transplantation. Animals were kept in an animal chamber with laminar air-flow at a room temperature of 23 + 1°C, and relative humidity of 55 _+ 5%.
4
R = c ( - L - r h o . p y r - c~-L-rhe.pyr- 2 # - D - g l c . p y r ~-L-rha.pyr / Fig. l. Chemical structure of formosanin-C.
EXPERIMENTAL PROCEDURES
Agents Formosanin-C (Fig. 1), diosgenin 3-O-a-L-rhamnopyranosyl-(l~4)-a-L-rhamnopyranosyl-(1-*4)-[a-L-rhamnopyranosyl-(l~2)]-/3-D-glucopyranoside, was purified from Paris formosana Hayata (Liliaceae) according to the method described elsewhere (Lay & Chiang, 1980). Briefly, the dried leaves of this plant were treated with methanol, and the extract was then partitioned between petroleum ether and water. The aqueous layer was further extracted with chloroform, and this portion was partitioned between n-butanol and water. The n-butanol portion was then applied to a silica gel column, and eluted with C H z C l z - M e O H - w a t e r (8:2:0.1). Fraction 2 was repeatedly applied on a silica gel column and eluted with CH2C12- n-butanol - MeOH - water (40:10:5:1). Recrystallization of fraction 2e from methanol gave colorless needles of formosanin-C. Crystallized formosanin-C (m.p. 204-206°C) was dissolved in 70% ethanol, and sterilized with 0.22/am durapore membrane (Millipore, Bedford, MA, U.S.A.). The sample was subsequently sealed with an autoclaved milliwrap membrane (Millipore, Tokyo, Japan) to evaporate the solvent in a concentrator centrifuge (Univapo 150 H, Uni Equip, West Germany). The sterilized powder was resuspended in RPMI-1640 (Gibco, NY, U.S.A.) with 10% FCS (Gibco) to a concentration of 100/aM in all tissue culture studies. For studies on antitumor activity in animals, formosanin-C was dissolved in 70% ethanol to a concentration of 5 raM, sterilized with 0.22/am durapore membrane, and diluted with sterilized 0.7% ethanol in phosphate buffered saline (PBS). The control
Proliferative response o f human peripheral whole blood to phytohemagglutinin Enhancement of phytohemagglutinin (PHA)stimulated whole blood assay was carried out following the method previously described (Wu, Chen, Tien & Tanaka, 1980). Human peripheral blood was obtained from healthy adults. The whole blood was suspended in nine-fold volume of RPMI-1640 medium (Gibco, NY, U.S.A.), containing 5% heatinactivated fetal calf serum (F.C.S., Gibco), 100 units/ml benzylpenicillin (Sigma Chemical Co, St Louis, MO, U.S.A.), 100/ag/ml streptomycin sulfate (Sigma), and 5/ag/ml PHA (Sigma). This suspension was distributed (0.2 ml each) to a 96-well microplate (Costar, MA, U.S.A.) with or without formosanin-C. The blood cells were incubated at 37°C for 48 h in 5% CO~ and 95% air, and 3H-thymidine (0.2/aCi/well) was then incorporated for 24 h. The cells were collected and washed with distilled water in a multiple automated cell harvester (Bellco, NJ, U.S.A.), and the radioactivity was determined in a liquid scintillation counter (1211 rackbeta, LKB, Sweden). Proliferative response o f mouse lymphocytes to concanavalin A Fresh splenocytes (1 × 106 cells/ml) from male C3H/HeN mice were suspended in a medium containing RPMI-1640 with 0.1 mM nonessential amino acids, 2 x 1 0 - 6 M 2-mercaptoethanol, 100 units/ml benzylpenicillin, 100/ag/ml streptomycin, and 10% heat-inactivated fetal calf serum, in addition to a final concentration of 3/~g/ml of concanavalin A (Sigma). This medium (180/al) was then added to each well of flat-bottomed microplates (Costar) with or without formosanin-C, and was incubated for 3 days at 37°C in a humidified atmosphere containing 5°7o CO2. 3H-Thymidine (0.2/aCi) was subsequently added to each well, 18 h before cell harvest. The incorporated radioactivity was measured using a
Formosanin-C, an Immunomodulator with Antitumor Activity liquid scintillation counter. These experiments were performed in quadruplicates.
Granulocyte - macrophage activity
colony
stimulating
The proliferative responses of granulocytemacrophage colony were evaluated on the basis of 3H-thymidine incorporation. Mouse fibroblast L929 cell conditioned medium was prepared as described elsewhere (Hines, 1983) with minor modification. L929 cells (1 × 106) were transferred to a 75 sq cm 2 tissue culture flask and cultured for 5 days with 10°70 FCS at 37°C in 5°7o CO2 and 95°/0 air. Confluent cells were fed with fresh medium, and conditioned medium was removed after 24 h, filtrated with a millipore 0.22 lam membrane, and stored at - 2 0 ° C until use. Male C 3 H / H e N mice were killed by cervical dislocation. Femoral bone marrow cells for culture and as a source of granulocyte/macrophage progenitor cells were obtained by flushing the marrow cavity with RPMI-1640 medium using a 26gauge needle. The cells (4 × 105/ml) were cultured in a medium containing RPMI-1640 with 5 × 10 6 M 2-mercaptoethanol, 100 units/ml benzylpenicillin, 100/ag/ml streptomycin, and 5% heat-inactivated fetal calf serum, in addition to a final 5°/0 v / v concentration of mouse fibroblast L929 conditioned medium. This medium (180/A) was then added to each well of flat-bottomed micro-plates (Costar) with or without formosanin-C, and was incubated for 4 days at 37°C in a humidified atmosphere containing 507o CO2. 3H-Thymidine (0.4/aCi) was subsequently added to each well 18 h before cell harvest. The cells were frozen and thawed three times, collected, and washed with normal saline and 5°7o cold trichloroacetic acid (E. Merck, Darmstadt, West German) in a multiple automated cell harvester. The incorporated radioactivity was measured using a liquid-scintillation counter. All assays were done in quadruplicates.
Natural killer cytotoxicity assay The antitumor efficacy of natural killer cells was assessed by a standard 4-h SlCr release assay. Male C3H / H e N mice (8 weeks old) were injected intraperitoneally with 1 or 2.5 mg/kg/day of formosanin-C for two days. After 24 h, animals were sacrificed and spleen cells removed from formosanin-C treated or untreated groups were used as effector cells. Each group consisted of four mice. The spleen cells were depleted of adherent cells by incubation in plastic Petri dishes in RPMI-1640 with 5°70 FCS for 1 h at
779
37°C in 5°7o CO2 atmosphere. The mouse lymphoma cell line YAC-1 was used as target cells. The cells were maintained in RPMI-1640 medium with 10% FCS, in a 5o70 COz incubator at 37°C. The target cells were washed twice with RPMI-1640 medium and adjusted to a concentration of 2 x 106/ml in RPMI1640 with 5°70 heat-inactivated FCS and labeled with 100 taCi of sodium chromate (NEN, Du Pont, MA, U.S.A.) for 60 min at 37°C. After washing three times with cold RPMI-1640, these target cells were resuspended to a concentration of 1 × 105/ml in RPMI-1640 with 5o/o FCS. Effector cells (0.1 ml) of nonadherent spleen cells were mixed with 1 × 1 0 4 (0.1 ml) target cells at a ratio of effector cells to target cells (E:T) of 200:1, 100:1, 50:1 in the microwells of a round-bottomed 96-well microplate (Nunclon, Nunc, Denmark). After 4 h of incubation, the supernatant from each well was collected, and the radioactivity measured using a gamma counter (LKB, Rack-gamma counter, Sweden). The natural killer cytotoxic activity was expressed as follows: counts/min of experimental releasecounts/min of spontaneous release % cytolysis =
× 100. counts/min of maximal releasecounts/min of spontaneous release
Where experimental counts/min was the 5'Cr released from injured targets, spontaneous counts/ rain was the spontaneous released from the targets, and maximal CPM the total counts/min in the target cells lysed with 0.5°70 Triton × 100.
Interferon (IFN) induction assay Male C 3 H / H e N mice (8 weeks old) were administrated intraperitoneally with 1, 2.5, or 5 m g / k g / d a y of formosanin-C and the sera were collected after 6, 24, or 48 h. Each group consisted of four mice. The IFN level was assayed by the cytopathic effect (CPE) of vesicular stomatitis virus (VSV) in a 96-well microplate (Costar). Each well contained 0.1 ml of two-fold-diluted serum and 0.1 ml of 7 × 104 L-929 cells in RPMI-1640 (Gibco) with 10% fetal calf serum. The microplate was kept at 37°C for 2 h in 5% COz and 95% air. After washing with the RPMI1640 medium, VSV of 100 IDs0 in 0.1 ml of RPMI1640 with 1% calf serum was added to the cells. The infected cells were further incubated for 24 h, in which controls reached the peak of CPE. The reciprocal of serum dilution, which showed 50°7o inhibition of CPE, was taken as IFN potency. The potency was corrected for international unit by comparison with reference IFN: N I H (L-929-NDV).
780
RONG- TSUN W u et al.
A n titumoractivity against mouse MH-134 hepatoma The MH-134 hepatoma was maintained in solid form by serial s.c. transplantation in syngeneic C3H/HeN male mice. When MH-134 hepatoma reached a size of approximately 10 mm diameter, the tumor tissue was excised and trimmed free of necrotic areas, then a cell suspension was prepared by mechanical means in Hank's balanced salt solution (HBSS). The viability was checked by the Trypan-blue exclusion test. For the therapeutic experiments, C3H/HeN mice received a s.c. injection of 4 x 105 hepatoma cells/0.2 ml HBSS/mouse on day 0. After tumor inoculation, mice were randomly assigned to experimental and control groups. Each group contained 10 mice. Formosanin-C was administered i.p. starting on day 1 and was given every day or every other days until the termination of the experiment. The antitumor effect of the combination treatment of formosanin-C with 5-fluorouracil (5-FU) was also studied. 5-FU was given daily i.p. for the first five days and formosanin-C was given i.p. every two days from day 7 until the termination of the experiment. 5-FU or formosanin-C given from day 1 in every other day was also scheduled. Control mice received the same volume of sterile distilled water or 0.7% ethanol in PBS with the same treatment schedule. The therapeutic response was measured by the weight of the tumor masses excised from the sacrificed mice. Stat&tical analysis The significance of the differences between drugtreated groups and control groups was analyzed using Student's t-test.
RESULTS
Proliferative response o f human peripheral whole blood to phytohemagglutinin At the concentrations of 0.8/aM, 0.16/aM and 0.032/aM (Fig. 2), formosanin-C dose-dependently enhanced the proliferative response of human peripheral whole blood to PHA, as evaluated by 3H-thymidine incorporation. The maximal stimulatory effect occurred at 0.16/aM, when the counts/min of 3H-thymidine was almost three times higher than the control (11000 + 1350 vs 3900 + 760). Increasing the concentration of formosanin-C to 4/aM did not elicit further effect on the 3H-thymidine incorporation of human peripheral whole blood. At 20/aM, it actually produced a hemolytic effect.
15 Io 12 x
E 9 o_ £)
v
6
rY ~3 F--
0
11
0.032 0.16 0.8 Concentrotion (/~M)
4.0
Fig. 2. Effect of formosanin-C in vitro on mitogenic responses of PHA-stimulated human peripheral whole blood cells from healthy adult: dependency of drug concentration. Each point represents an average of quadruplicate cultures (*P<0.001).
Proliferative response o f mouse lymphocytes to concanavalin A Formosanin-C significantly increased 3H-thymidine incorporation of ConA-stimulated lymphocytes (Fig. 3). The maximal concentration (0.1 /aM), at which a stimulatory effect was produced, was similar to that (0.16/aM) in the proliferative response of human peripheral whole blood of PHA. An inhibitory effect on 3H-thymidine incorporation of ConAstimulated lymphocytes occurred at concentrations of 10 and 1 /aM.
Granulocyte- macrophage colony stimulating activity The responsiveness of granulocyte-macrophage colony-forming cells (GM-CFC) to mouse fibroblast cell L929 conditioned medium was altered by formosanin-C. As shown in Fig. 4, this agent augmented 3H-thymidine incorporation of GM-CFC at concentrations of 0.01 and 0.001 /aM. The concentration of formosanin-C that stimulated the GM-CFCs maximally was 0.01 /aM, which is ten-fold lower than that required to achieve maximal stimulation in ConAstimulated lymphocytes. At concentrations of 1 and 10/aM, formosanin-C produced an inhibitory effect on the 3H-thymidine incorporation of GM-CFC.
Assay f o r natural killer cytotoxicity Formosanin-C (1 mg/kg/day) significantly enhanced the natural killer (NK) activity of C3H/He mice, as assessed by the cytotoxicity of fresh splenocytes from treated animals against NK-sensitive YAC-1 cells (Fig. 5). Moderate augmentation was observed in the formosanin-C (2.5 mg/kg/day)treated group.
781
Formosanin-C, an I m m u n o m o d u l a t o r with A n t i t u m o r Activity 25 ¸
8,
•
,4"
io x
Control
o~o &--A
1.0 n~/kg 2.5 mg/k9
# ,
G X
o
E
,~1
2o
0--0
i-. 15 o
2
"1-
5o":1
0 0
0.01
0.1
Concentrotlon (/zM)
Fig. 3. Effect of formosanin-C in vitro on mitogenic response of C o n A-stimulated splenocytes from normal C 3 H / H e N mice: dependency on drug concentration. Each point represents an average of quadruplicate cultures. (*/'<0.05).
loo:1
2~:1
Effector : Target Ratio
10
1
Fig. 5. Effect of formosanin-C on natural killer cellmediated cytotoxicity in normal C 3 H / H e N mice. Formosanin-C was injected intraperitoneally on day 0 and day 1. Control group was injected with 0.7% ethanol 0.2 ml intraperitoneally on day 0 and day 1. Mice were sacrificed on day 2, and the spleen cells were used as effector cells. YAC-1 l y m p h o m a cells were used as target cells. (*P<0.05, #P<0.001).
10-
i,q
tO
28O 8-
24O
X
E
CL (J
Iz 1o I-I -r
2OO
6-
2-
0 0
i
0.001 0.01
O.
° 1
I
Control
r'~ w e i.o r~/kg FzC 2.5 m g / k 9
I
F=C 5.0 mg/kg
tOO
12o
~_ 8o 40
10
C o n c e n t r a t i o n ~J.M)
i
m~
o
6
rq~, 24
48
Hours
Fig. 4. Effect of formosanin-C in vitro on proliferative responses of granulocyte/macrophage progenitor cells from bone marrow of C 3 H / H e N mice to L-929 cells conditioned medium: dependency on drug concentration. Each point represents the [~H]-thymidine incorporation of an average of quadruplicate cultures. (**P<0.01, *P<0.05).
Fig. 6. Kinetics o f interferon induction by formosanin-C in C 3 H / H e N mice. Mice were administrated intraperitoneally with 1, 2.5, or 5 m g / k g / d a y of formosanin-C and the sera were collected after 6, 24 or 48 h. Each group consisted of four mice. The IFN level was assay by the cytopathic effect of vesicular stomatitis virus. (*P<0.01, **P<0.05).
The interferon induction The effect of formosanin-C on blood level of interferon was examined in C3H/HeN mice (Fig. 6). A n i n t r a p e r i t o n e a l i n j e c t i o n o f 2.5 m g / k g o f f o r m o -
sanin-C markedly induced interferon production. The peak blood level of interferon was observed 24 h after formosanin-C injection. The higher dose of formosanin-C treatment (5 mg/kg) did not produce any significant interferon inducing effect.
Table 1. Effect o f formosanin-C on MH-134 mouse hepatoma Formosanin-C (mg/kg/day) -1.0 2.5 5.0
T u m o r weight (g) 3.53 3.17 1.10 2.98
_ ± _ +
1.29 1.37 0.52 1.59
Inhibition (°7o) 0 10 69 (P
A n t i t u m o r activity against mouse M H - 1 3 4 hepatoma W e e x a m i n e d w h e t h e r f o r m o s a n i n - C c o u l d elicit inhibition against mouse MH-134 hepatoma. MH134 h e p a t o m a cells (4 × 105 cell) w e r e i n o c u l a t e d to the dorsal region of C3H/HeN mice (male, 8 weeks old) at d a y 0 a n d f o r m o s a n i n - C w a s g i v e n d a i l y (i.p.)
A suspension of 4 × 105 cells was implanted subcutaneously in the dorsal region of male C 3 H / H e N mice ( 8 - 1 0 weeks old). Mice were treated with formosanin-C i.p. every day from day 1 to day 11 after t u m o r inoculation. The weight of t u m o r was determinated by autopsy on day 15.
782
RONG-TSUNWU et al.
Table 2. Effect of formosanin-C on MH-134 mouse hepatoma Formosanin-C (mg/kg/day) --
1.0 2.5 5.0
Tumor weight (g)
Inhibition (%)
3.38 _+2.09 2.09 _+0.63 0.76 _+0.33 1.79 _+0.11
0 38 78 (P<0.01) 48 (P<0.05)
A suspension of 4 × 105 cells was implanted subcutaneously in the dorsal region of male C3H/He mice (8-10 weeks old). Mice were treated with formosanin-C i.p. every other day from day 1 to day 13 after tumor inoculation. The weight of tumor was determinated by autopsy on day 17. from day 1 for 11 days. Tumors were extirpated and weighed 15 days after tumor inoculation. As shown in Table 1, formosanin-C treatment (2.5 mg/kg/ day) was effective in suppressing the growth of tumor. At a dose of 5 mg/kg/day, formosanin-C was not only ineffective in regressing the growth of tumor, but was somewhat toxic to the mice by reducing their body weight. In a different treatment schedule, formosanin-C (i.p.) was administrated every other day from day 1 to day 13. Tumor masses were excised and weighed 17 days after tumor inoculation. As shown in Table 2, there was no dosedependency, and 2.5 mg/kg/every other day i.p. seemed to be the optimal dose to suppress the growth of tumor. The 5 mg/kg/every other day i.p. treatment was also effective, and produced no further weight loss. This was quite different from the 5 mg/kg/day i.p. group (Table 1). We also studied the antitumor activity of formosanin-C in combination with a clinically used antitumor drug, 5-fluorouracil (5-FU). 5-FU was given daily for 5 days starting 1 day after the inoculation of 4 x 105 hepatoma cells. Formosanin-C was given every other day from day 7 to day 19. The weight of the tumor was determined on the 20th day. Treatment with formosanin-C (2.5 mg/kg) or 5-FU (20 mg/kg) alone inhibited growth of the tumor by 37% and 28% respectively (Table 3). A combined treatment with formosanin-C (2.5 mg/kg) and 5-FU (20 mg/kg), however, inhibited growth of the tumor by 73%. In another treatment schedule, after 5 x 10~cells were s.c. transplanted to the C3H/HeN mice, 5-FU was injected i.p. every other day from day 1 to day 13, and formosanin-C was injected i.p. every other day from day 2 to day 12. On day 14, the tumor was extirpated and weighed. As shown in Table 4, the treatment of formosanin-C (2.5 mg/kg) or 5-FU (25 mg/kg) alone inhibited growth of the
tumor by 45% and 33% respectively, the treatment of formosanin-C (2.5 mg/kg) and 5-FU (25 mg/kg) inhibited growth of the tumor by 64%. DISCUSSION
The present study revealed that formosanin-C produced dual effects on the proliferative response of human peripheral whole blood to PHA, proliferative response of mouse lymphocytes to concanavalin A, and proliferative response of mouse granulocyte-macrophage colony-forming cells to mouse fibroblast cell L929 conditioned medium. Maximal stimulatory effects were produced on the above responses at concentrations of 0.16, 0.1 and 0.01 ~M respectively (Figs 2, 3 & 4). An inhibitory action on the same responses was observed, however, at concentrations of 4, l and 1 gM respectively (Figs 2, 3 & 4). These results indicated that formosanin-C was non-specifically cytotoxic to both lymphoid and tumor cells at concentrations above 1 gM. Formosanin-C enhanced the responsiveness of the granulocyte-macrophage colony-forming cells (GM-CFC) to mouse fibroblast cells L929 conditioned medium at the very low concentration of 0.01 /~M. This effect may be mediated by directly stimulating GM-CFC proliferation and/or stimulating production of granulocyte-macrophage colony stimulating factors (GM-CSFs). The GM-CSFs probably play an important role in hematopoietic generation of GMprogenitor cells and in differentiation of myeloid leukemic cells (Metcalf, 1985; Schlick & Ruscetti, 1986; Begley, Metcalf & Nicola, 1987). In addition, formosanin-C given intraperitoneally enhanced the NK activity of C3H/HeN mice, the maximum effect was observed at a dose of 1 mg/kg (Fig. 5). In light of the possible role of NK cells in cancer defense (Herberman & Ortaldo, 1981), numerous efforts have been made in NK cell-augmenting effect. (Ching & Baguley, 1987; Goodman, 1988). An intraperitoneal injection of 2.5 mg/kg of formosanin-C markedly induced interferon production, and the peak blood level of interferon was observed 24 h after formosanin-C injection (Fig. 6). Studies in animals and man also suggest the effectiveness of IFN (Goldstein & Laszlo, 1986; Gutterman, 1988; Jaffe & Herberman, 1988; Figlin, 1988) and IFN inducer (Pollard, 1982; Lotzova, Savary & Stringfellow, 1983; Hornung, Young, Urba & Wiltrout, 1988) therapy in cancer. Thus, the endogenous interferon induced by formosanin-C might offer another applicability in cancer therapy, which could avoid the undesirable side effects (Mannering & Deloria, 1986), fast serum
783
Formosanin-C, an Immunomodulator with Antitumor Activity Table 3. Activity of formosanin-C in combination with 5-fluorouracil on MH-134 mouse hepatoma Formosanin-C (mg/kg/day)
5-FU (mg/kg/day)
-1.0 2.5 5.0 -1.0 2.5 5.0
Tumor weight (g)
----20 20 20 20
5.26 4.51 3.32 3.73 2.79 2.72 1.44 2.38
Inhibition (%)
__. 2.33 _ 2.36 __. 1.15 _+ 1.53 _+ 1.58 _+ 1.15 _ 0.72 _+ 0.86
0 15 37 29 28 48 73 55
(P
(P<0.01) (/~0.001) (P<0.01)
A suspension of 4 x 105 cells was implanted subcutaneously in the dorsal region of male C 3 H / H e N mice ( 8 - 10 weeks old). Mice were treated with 5-FU i.p. every day from day 1 to day 5, and injected with formosanin-C i.p. every other day from day 7 to day 19 after tumor inoculation. The weight of tumor was determinated by autopsy on day 20.
T.I * I
Day
I
*
*
*
*
I
I
I
I
0
,~ I
I
¢
~, i
I
I
5
,1, I
I
I
~, I
10
¢
I
I
¢ I
I
T.E. I
I
15
20
*5-Fluorouracil (20 mg/kg/day); +: Formosanin-C (1.0, 2.5 mg/kg/day); T.I: Tumor inoculation; T.E: Tumor excision.
Table 4. Activity of formosanin-C in combination with 5-fluorouracil on MH-134 mice hepatoma Formosanin-C (mg/kg/day)
5-FU (mg/kg/day)
-1.0 2.5 --1.0
Tumor weight (g)
---10 25 10 25 10 25
1.0 2.5 2.5
2.44 1.60 1.35 2.50 1.64 1.14 1.32 1.45 0.87
_+ 0.70 _+ 0.73 + 0.49 _+ 1.08 + 0.86 ___0.56 _+ 0.41 _+ 0.86 _+ 0.49
Inhibition (%) -34 45 0 33 42 46 41 64
(P<0.05) (P<0.001) (P<0.05) (P<0.001)
(P<0.001) (P<0.05) (P<0.001)
A suspension of 5 x 104 cells was implanted subcutaneously in the dorsal region of male C H / H e N mice (8 - 10 weeks old). 5-FU was injected i.p. every other day from day 1 to day 13, and formosanin-C was injected i.p. every other day from day 2 to day 12 after tumor inoculation. The weight of tumor was determinated by autopsy on day 14. T.I * ¢ I
Day
0
I
I
* ~
* ¢
I
I
I
I
* ~
* ¢
* ¢
*
I
I
I
I
I
5
I
10
I
T.E. I
I
15
*5-Fluorouracil (10, 25 mg/kg/day); ~: Formosanin-C (1.0, 2.5 mg/kg/day); T.I: Tumor inoculation; T.E: Tumor excision.
clearance and antibody occurrence (Hawkins, Horning, Konrad, Anderson, Sielaff, Rosno, Schiesel, D a v i s , D e M e t s , M e r i g a n et al., 1986) o f
e x o g e n o u s interferon t h e r a p y . W h e t h e r f o r m o s a n i n - C a u g m e n t e d N K activity via t h e m e c h a n i s m o f i n t e r f e r o n s i n d u c t i o n n e e d s to be f u r t h e r clarified.
784
RONG-TSUNWu et al.
Formosanin-C retarded growth of s.c. solid MH-134 hepatoma in C3H/HeN mice (Tables 1 and 2), but was not effective against ascites form MH-134 hepatoma in C3H/HeN mice (data not shown). Although the maximum effective dose was 2.5 mg/kg in both administration schedules, the every-otherday schedule of 5 mg/kg was effective (Table 2, P<0.05), while the every day schedule of 5 mg/kg was not effective (Table 1). Formosanin-C was also effective in enhancing the antitumor activity of 5-FU against s.c. solid mice MH-134 hepatoma (Table 3), but appeared to be not obviously effective in another combination schedule (Table 4). Optimization of the capability of immune response modification therefore seemed to be dependent on the method of administration. The s.c. solid MH-134 hepatoma seems to be a slow-growing tumor in which defense mechanisms are somewhat preserved, and formosanin-C could activate them in time. It is therefore reasonable to conclude that formosanin-C may display antitumor activity in association with modification of the immune function, especially cellmediated immunity and/or secretion of soluble factors, but not with its cytotoxic effect. A cytotoxic agent podophyllotoxin has also recently been found to augment macrophage proliferation at a low concentration (Zheng, Wiranowska, Sadlik & Hadden, 1987). Besides its well known cytotoxic effect, a low concentration of the cytotoxic drug 1-p-D-arabinofuranosylcytosine (Ara-C) was reported to produce differentiation-inducing activity of leukemic cells (Luisi-Deluca, Mitchell, Spriggs & Kufe, 1984; Wu et al., 1985). These findings indicate the importance of biological activities at low concentrations in the evaluation of a biologically active substance. Saponins are natural compounds that exist in varying proportions in many plants. They usually occur in nature as mixtures of related compounds, which are chemically divided into either triterpenoid or steroidal glycosides (Oakenfull, 1981). Some saponins show various biological activities. For example, the crude saponin extract of ginseng root has a promoting effect on neurite extension primary in cultured neurons of the rat cerebral cortex (Sugaya, Yuzurihara, Tsuda, Yasuda, Kajiwara & Sugaya, 1988). A purified saponin, Ginsenoside Rh2, isolated from Panax ginseng inhibits the growth of B16 melanoma cells, causes morphological alterations, and stimulates melanogenesis (Odashima, Ohta, Kohno, Matsuda, Kitagawa, Abe & Arichi, 1985). Some commercially available saponins have been used as adjuvants along with parenterally administered experimental and parasitic vaccines (Dalsgaard, 1974; Bomford, 1982; Scott, Bahr,
Moddaber & Chedid, 1984). The adjuvant activity of saponins is considered to be due to their prolonging retention of antigen and its localization in the spleen (Scott, Goss-Sampson & Bomford, 1985). Recent studies showed that orally administered Quillaja saponin potentiates the humoral and cellular immune responses in mice fed with inactivated rabies vaccine (Chavali & Campbell, 1987; Maharaj, Froh & Campbell, 1986). Since orally fed saponins do not enter circulation (Sidhu & Oakenfull, 1986), a mechanism has been proposed that Quillaja saponin might affect the mucosal immune system, which causes activation of helper T-cells and induces secretion of soluble factors (Chavali & Campbell, 1987; Chavali, Francis & Campbell, 1987; Maharaj, Froh & Campbell, 1986). We also examined whether formosanin-C fed orally could exhibit inhibition against mouse MH-134 hepatoma. At oral doses of 2, 6, 15 mg/kg/day, formosanin-C (20 days)-treated group was ineffective in regressing the growth of the tumor (unpublished data). In addition, there was no weight loss in mice given formosanin-C orally. Although our present results only yield an incomplete picture of the effects of formosanin-C on the immune system, we have identified that at very low concentrations, formosanin-C has the capacity to potentiate the immune system, which might be related to its unique antitumor activity. The mechanisms whereby formosanin-C modulates the immune system still remain unclear. The surfactant nature of saponins is considered to induce changes at the cell surface membrane (Alvarez & TorresPinedo, t982). The selective effect of saponins to different lineages of immune cells is likely to depend oll the chemical stucture of the individual saponins and the nature of the cell surface. The selectivity of formosanin-C to immune cells needs to be clarified further. It has long been recognized that modulation of the immune system by various agents may have potential for the treatment of certain neoplastic disease (Hadden, 1983; Oldham, 1983; Reizenstein & Mathe, 1983; Wagner, 1983; Mihich, 1986; RuszalaMallon, Lin, Durr & Wang, 1988). Formosanin-C studied here is a plant saponin and is a possible novel immunomodulatory substance. One of the most important means of our studies is to clarify scientifically the effect mechanism of our traditional Chinese folk medicine, as Paris formosana Hayata (Liliaceae) and "Yunnan Bai Yao". To re-examine the traditional Chinese medicines that are said to be effective against cancer by approaches different from cytotoxic mechanism, such as induction of tumor necrosis factor (Katsuyuki, Nobuko, Akiko, Ruriko, Nakao & Mosaburo, 1985), might also offer another
Formosanin-C, an Immunomodulator with Antitumor Activity new way for the arrest o f cancer. Further studies on the effector mechanisms related to the immunomodulatory activity o f formosanin-C are in progress. Acknowledgements - - This study was supported in part by research grant NSC-75-0610-B010-12 & NSC-76-0610-
785
B010-15 from National Science Council, DOH75-0302-38 from Ministry of Health, Republic of China, and in part by the projects from the Memory Foundation of Dr K-S Lu. The authors express their deep thanks to Dr G. Chihara for his helpful discussion.
REFERENCES
ALVAREZ, J. R. & TORRES-PINEDO, R. (1982). Interactions of soybean lectin, soyasaponins and glycinin with rabbit jejunal mucosa in vitro. Pediat. Res., 16, 728-731. BEGLEY, C. G., METCALF, D. & NICOLA, N. A. (1987). Purified colony stimulating factors (G-CSF and GM-CSF) induce differentiation in human HL60 leukemic cells with suppression of clonogenicity. Int. J. Cancer, 39, 9 9 - 105. BOMFORD, R. (1982). Studies on the cellular site of action of the adjuvant activity of saponin for sheep erythrocytes. Int. Arch. Allergy Appl. Immun., 67, 127- 31. CHAVALI, S. R. & CAMPBELL,J. B. (1987). Adjuvant effects of orally administered saponins on humoral and cellular immune responses in mice. Immunobiology, 174, 347-359. CHAVALI, S. R., FRANCIS, T. & CAMPBELL, J. B. (1987). An in vitro study of immunomodulatory effects on some saponins. Int. J. Pharmac., 9, 675- 683. CHING, L. U . & BAGULEY, B. C. (1987). Induction of natural killer cell activity by the antitumour compound flavone acetic acid. Eur. J. Cancer clin. Oncol., 23, 1047 - 50. DALSGAARD, K. (1974). Saponin adjuvants. III. Isolation of a substance from Quillaja saponaria Molina with adjuvant activity in food-and-mouse disease vaccines. Arch. ges. virusforsch., 44, 243- 254. FIGLIN, R. A. (1988). Biotherapy with interferon. Semin. Oncol., 15, (Suppl. 6) 3 - 9 . GOLDSTEIN, D. & LASZLO,J. (1986). Interferon therapy in cancer: from imaginon to interferon. Cancer Res., 46, 4315 - 29. GOODMAN, S. (1988). Therapeutic effects of organic germanium. Med. Hypotheses, 26, 207 - 15. GUTTERMAN, J. U. (1988). The role of interferons in the treatment of hematologic malignancies. Semin. Hemat., 25, 3 - 8. HADDEN, J. W. (1983). Chemically defined immunotherapeutic agents. Prog. Clin. Biol. Res., 132, 273- 86. HAWKINS, M., HORNING, S., KONRAD,M., ANDERSON, S., SIELAFF,K., ROSNO, S., SCHIESEL, J., DAVIS, T., DEMETS, D., MERIGAN, T., et al. (1985). Phase I evaluation of a synthetic mutant of beta-interferon. Cancer Res., 45, 5914-20. HERBERMAN, R. B. & ORTALDO, J. R. (1981). Natural killer cells: their roles in defenses against disease. Science, 214, (4516) 2 4 - 30. HINES, D. (1983). Lipid accumulation and production of colony-stimulating activity by the 266AD cell line derived from mouse bone marrow. Blood, 61, 397-402. HORNUNG, R. L., YOUNG, H. A., URBA, W. J. & WILTROtJT, R. H. (1988). Immunomodulation of natural killer cell activity by flavone acetic acid: occurrence via induction of interferon alpha/beta. J. natn. Cancer Inst., 80, 1226 - 31. JAFFE, H. S. & HERBERMAN,R. B. (1988). Rationale for recombinant human interferon-gamma adjuvant immunotherapy for cancer. J. natn. Cancer Inst., 80, 616-8. LAY, J. Y. & CHIANG, H. C. (1980). Studies on the constituents of Paris formosana Hayata. J. Taiwan Pharmaceut. Asso., 32, 14-27. KATSUYUKI,H., NOBUKO,S., AKIKO,A., RURIKO,H., NAOKO, O. & MOSABURO,K. (1985). Antitumor activities and tumor necrosis factor producibility of traditional Chinese medicines and crude drugs. Cancer Immun. Immunother., 20, 1-5. LI, M. C., Koo, W. Y. & Hsu, K. P. (1972). Anti-neoplastic property of a crude extract from Parisformosana. Nature New Biol., 235, 223 - 224. LI, T. C. (1597). Pen-ts'ao Kang-mu. Chinese Meteria Medica, 16, 698. LOTZOVA, E., SAVARY,C. A. & STRINGFELLOW,D. A. (1983). 5-halo phenyl pyrimidinones: new molecules with cancer therapeutic potential and interferon-inducing capacity are strong inducers of murine natural killer cells. J. Immun., 130, 965 - 9. LUIS1-DELUCA,C., MITCHELL, T., SPRIGGS,D. & KUFE, D. W. (1984). Induction of terminal differentiation in human k562 erythroleukemia cells by arabinofuranosyl-cytosine. J. clin. Invest., 74, 821- 827. MAHARAJ,I., FROH,K. J. & CAMPBELL,J. B. (1986). Immune responses of mice to inactivated rabies vaccine administered orally: potentiation by Quillaja saponin. Can. J. Microbiol., 32, 4 1 4 - 420. MANNERIN~, G. J. & DELORIA, L. B. (1986). The pharmacology and toxicology of the interferons: an overview. Ann. Rev. Pharmac. Toxic'., 26, 455-515. METCALF, D. (1985). The granulocyte-macrophage colony-stimulating factors. Science, 229, 16-22. MIHICH, E. (1986). Future perspectives for biological response modifiers: a viewpoint. Semin. Oncol., 13, 234-54.
786
RONG-TsuN WU et al.
OAKENFULL, D. G. (1981). Saponins in food - - a review. Food Chem., 6, 19-41. ODASHIMA, S., OHTA, T., KOHNO, H., MATSUDA,T., KITAGAWA,I., ABE, H. & ARICHI, S. (1985). Control of phenotypic expression of cultured B16 melanoma cells by plant glycosides. Cancer Res., 45, 2781 -2784. OLDHAM, R. K. (1983). Biological response modifiers. J. natn. Cancer Inst., 70, 789-796. POLLARD, R. B. (1982). Interferons and interferon inducers: development of clinical usefulness and therapeutic promise. Drugs, 23, 3 7 - 55. RAVIKUMAR,P. R., HAMMESFAHR,P. & S1H, C. J. (1979). Cytotoxic saponins from the Chinese herbal drug Ynnnan Bai Yao. J. Pharmaceut. Sci., 68, 900- 904. REIZENSTEIN, P. & MATHE, G. (1983). Biological response modifiers and differentiation inducers. Current status in cancer research. Drugs, 26, 185- 190. RUSZALA-MALLON,V., LIN, Y. I., DURR, F. E. & WANG, B. S. (1988). Low molecular weight immunopotentiators. Int. J. Immunopharmac., 10, 497 - 510. SCHLICK, E. & RUSCETTI, F. W. (1986). In vivo induction of terminal differentiation of malignant myelopoietic progenitor cells by CSF-inducing biological response modifiers. Blood, 67, 980-987. SCOTT, M. T., BAHR, G., MODDABER,F. & CHEDID, L. (1984). Adjuvant requirement for protective immunization of mice using a Trypanosoma cruzei 90K cell surface glycoprotein. Int. Archs Allergy appl. Immun., "/4, 373 - 377. SCOT]-, M. T., GOsS-SAMPSON,M. & BOMFORD, R. (1985). Adjuvant activity of saponin: antigen localization studies. Int. Arch. Allergy. appl. Immun., 77, 409- 12. SIDHU, G. S. & OAKENrULL,D. G. (1986). A mechanism for the hypocholesterolaemic activity of saponins. Br. J. Nut., 55, 643 - 649. SUGAYA, A., YUZURIHARA, M., TSUDA, T., YASUDA, K., KAJ1WARA, K. & SUGAYA,E. (1988). Proliferative effect of ginseng saponin on neurite extension of primary cultured neurons of the rat cerebral cortex. J. Ethnopharmac., 22, 173 - 181. WAGNER, H. (1983). lmmunoprevention and therapy by plant preparations ZFA (Stuttgart), 59, 1282- 1289. Wu, R. Y. & TANAKA,N. (1980). Effects of antitumor agents on DNA synthesis of mitogen-induced human lymphocytes, in comparison with leukemic cells. J. Antibiotics, 33, 770-771. Wu, R. T., CHEN, S. C., TIEN, W. C. & TANAKA, N. (1985). Potentiation of 1-fl-D-arabinofuranosylcytosine-induced erythroid differentiation of human leukemia K562 cells by cadeguomycin. In Recent Advances in Chemotherapy (ed. Ishigami, J.) pp. 201-202. University of Tokyo Press, Tokyo. ZHENG, Q. Y., WIRANOWSKA,M., SADLIK, J. R. & HADDEN, J. W. (1987). Purified podophyllotoxin (CPH-86) inhibits lymphocyte proliferation but augments macrophage proliferation. Int. J. Immunopharmac., 9, 539-549.