- Email: [email protected]
Promotion or suppression of experimental metastasis of B16 melanoma cells after oral administration of lapachol
Available online at www.sciencedirect.com
Toxicology and Applied Pharmacology 229 (2008) 232 – 238 www.elsevier.com/locate/ytaap
Promotion or suppression of experimental metastasis of B16 melanoma cells after oral administration of lapachol Masayo Maeda a , Manabu Murakami b , Tsutomu Takegami b , Takahide Ota b,⁎ a
b
Department of Chemistry, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan Division of Molecular Oncology and Virology, Medical Research Institute, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan Received 14 November 2007; revised 19 December 2007; accepted 14 January 2008 Available online 26 January 2008
Abstract Lapachol [2-hydroxy-3-(3-methyl-2-butenyl)-1,4-naphthoquinone] is a vitamin K antagonist with antitumor activity. The effect of lapachol on the experimental metastasis of murine B16BL6 melanoma cells was examined. A single oral administration of a high toxic dose of lapachol (80–100 mg/kg) 6 h before iv injection of tumor cells drastically promoted metastasis. This promotion of metastasis was also observed in T-cell-deficient mice and NK-suppressed mice. In vitro treatment of B16BL6 cells with lapachol promoted metastasis only slightly, indicating that lapachol promotes metastasis primarily by affecting host factors other than T cells and NK cells. A single oral administration of warfarin, the most commonly used vitamin K antagonist, 6 h before iv injection of tumor cells also drastically promoted the metastasis of B16BL6 cells. The promotion of metastasis by lapachol and warfarin was almost completely suppressed by preadministration of vitamin K3, indicating that the promotion of metastasis by lapachol was derived from vitamin K antagonism. Six hours after oral administration of lapachol or warfarin, the protein C level was reduced maximally, without elongation of prothrombin time. These observations suggest that a high toxic dose of lapachol promotes metastasis by inducing a hypercoagulable state as a result of vitamin K-dependent pathway inhibition. On the other hand, serial oral administration of low non-toxic doses of lapachol (5–20 mg/kg) weakly but significantly suppressed metastasis by an unknown mechanism, suggesting the possible use of lapachol as an antimetastatic agent. © 2008 Elsevier Inc. All rights reserved. Keywords: Cancer; Metastasis; B16 melanoma; Vitamin K antagonist; Lapachol; Warfarin; Herbal medicines
Introduction Lapachol [2-hydroxy-3-(3-methyl-2-butenyl)-1,4-naphthoquinone] was identified as the active substance in extracts of the roots of Stereospermum suaveolens and showed antitumor activity against the rat Walker 256 carcinoma (Rao et al., 1968). A phase I clinical trial of lapachol was performed, revealing an anticoagulant side effect (Block et al., 1974; Sieber et al., 1976). An effective plasma concentration for antitumor activity could not be achieved without toxicity in cancer patients receiving lapachol (Block et al., 1974; Sieber et al., 1976). For these reasons, the clinical studies with lapachol were terminated. The mechanism underlying the antitumor effect of lapachol is still unknown. However, the anticoagulant activity of lapa⁎ Corresponding author. Fax: +81 76 286 3652. E-mail address: [email protected] (T. Ota). 0041-008X/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2008.01.008
chol arises from its potent inhibition of vitamin K epoxide reductase and vitamin K quinone reductase (Preusch and Suttie, 1984). Other vitamin K antagonists have been shown to be potent anti-metastatic drugs (Agostino et al., 1966; Ryan et al., 1968; Brown, 1973; McCulloch and George, 1987), but the anti-metastatic effect is not mediated solely via their anticoagulant activity (Hilgard and Maat, 1979; Hejna et al., 1999; Tagalakis et al., 2007). Recently, lapachol was shown to inhibit the invasiveness of HeLa cells in vitro, and a potential antimetastatic activity of lapachol was suggested (Balassiano et al., 2005). To clarify whether lapachol has anti-metastatic activity in vivo, we examined the effect of lapachol on the experimental metastasis of murine B16BL6 melanoma cells in the lung. Our results suggest that a high toxic dose of lapachol, a vitamin K antagonist, transiently induces a hypercoagulable state, thereby promoting the metastasis of B16BL6 melanoma cells in the
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
233
Fig. 1. Oral administration of a high toxic dose of lapachol increased the experimental metastasis of murine B16BL6 melanoma cells. In (A)–(D), 5 × 104 B16BL6 melanoma cells were injected iv, and the number of metastases was measured after 14 days. (A) Lapachol was administered orally three times (at 48, 24, and 6 h) before tumor cell injection. Data indicate the mean ± SE of 12 mice (control) and 7 mice (lapachol-administered). (B) Lapachol (0.5 or 5.0 mg/kg) was administered orally three, six, or nine times before tumor cell injection. Data indicate the mean ± SE of 5–6 mice. (C) Lapachol (80 mg/kg) was administered orally once (at 48, 24, or 6 h) before tumor cell injection. Data indicate the mean ± SE of 6 mice (control) and 5 mice (− 48 h, − 24 h, and −6 h). (D) B16BL6 melanoma cells were treated with lapachol for 24 h in culture, and then cells were injected into mice. Data indicate the mean ± SE of 5 mice. ⁎p b 0.05 and ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
234
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
lung. These results indicate the potential risk of promotion of cancer metastasis by vitamin K antagonists. On the other hand, serial oral administration of low doses of lapachol weakly but significantly suppressed metastasis, suggesting the possible use of lapachol as an anti-metastatic agent. Materials and methods Cells and culture. The B16BL6 subline of murine B16 melanoma was kindly provided by Dr. T. Sasaki (Department of Experimental Therapeutics, Cancer Research Inst., Kanazawa University, Kanazawa, Japan). Cells were cultured in a mixture of Ham's F10 and L-15 (ratio, 3:7) containing 10% fetal bovine serum (FBS), penicillin (50 U/ml), and streptomycin (50 μg/ml) in a humidified atmosphere of 5% CO2/95% air at 37 °C. New cultures were initiated from frozen stock after 10 subcultures (Nakano et al., 1996). Drug administration and treatment. Lapachol, warfarin, and vitamin K3 were purchased from Sigma Chemical Company (St. Louis, MO). Lapachol was dissolved in 2% sodium carbonate in water, and warfarin was dissolved in water. These drugs were administered by oral gavage. Vitamin K3 was dissolved in sterile phosphate-buffered saline (PBS) and administered ip. The solutions were freshly prepared just before use. For in vitro treatment, lapachol was dissolved in ethanol at a concentration of 40 mM and added to cultures. Ethanol was added to the control culture at a concentration of 0.05%. Assay of experimental metastasis. B16BL6 cells were harvested with 0.02% EDTA in Hanks' Balanced Salt Solution (HBSS). Harvested cells were washed once with medium containing FBS and once with HBSS. Cells with N90% viability as determined by the trypan blue dye exclusion test were used for experiments. Cells (5 × 104 or 2.5 × 104) suspended in 0.2 ml of HBSS were injected into the lateral tail veins of mice. Two weeks after injection of cells, the mice were euthanized, and the numbers of lung tumor nodules were counted under a dissecting microscope as described (Ota et al., 1996).
Results Effect of oral administration of lapachol on the experimental metastasis of B16BL6 melanoma cells Lapachol was administered orally to mice three times (at − 48, − 24, and − 6 h) before tumor cell injection into the lateral tail vain. Two weeks after tumor cell injection, mice were euthanized, and the numbers of metastatic nodules in the lungs were counted. Low doses (0.5, 5, 10, and 20 mg/kg) of lapachol weakly but significantly inhibited metastasis, whereas higher doses (80 and 100 mg/kg) drastically promoted metastasis (Fig. 1A). To confirm the metastasis-inhibiting effect of lowdose lapachol, we increased the frequency of lapachol administration at 0.5 and 5.0 mg/kg. Although lapachol showed a tendency to inhibit metastasis, the inhibition did not increase with increased frequency of administration (Fig. 1B). In contrast to the weak inhibition of metastasis, the promotion of metastasis was substantial. Because the strong promotion of metastasis by a therapeutic agent calls for great concern, we further examined the metastasis-promoting property of lapachol. To determine the sensitive period for metastasis promotion by lapachol, lapachol (80 mg/kg) was administered once at 48, 24, or 6 h before tumor cell injection. Only the administration at 6 h before tumor cell injection significantly increased the number of metastatic nodules in the lung (Fig. 1C). These results indicate that lapachol promotes metastasis within 6 h of administration, and this effect disappears within 24 h.
Animals. Seven-week-old male C57BL/6 normal mice and five-week-old male BALB/c nude mice were obtained from Japan SLC, Inc. (Shizuoka, Japan) and were maintained under specific pathogen-free conditions. C57BL/6 normal mice and BALB/c nude mice were used for experiments at 8 weeks and 6–7 weeks of age, respectively. To eliminate NK activity in mice, anti-asialoGM1 was used (Ota et al., 1992). Rabbit anti-asialoGM1 serum was purchased from Wako Pure Chemical (Osaka, Japan). The antiserum was diluted 1:8 in PBS and was filtered through Millipore filters (0.22 μm) before injection. Three days before tumor cell inoculation, male C57BL/6 mice were injected iv with 0.2 ml of the diluted antiserum. The same volume of normal rabbit serum at a 1:8 dilution in PBS was injected into control mice in place of the anti-asialoGM1 serum. Experiments were performed in accordance with the guidelines of the Committee on Experimental Animals at Kanazawa Medical University. Measurement of protein C activity and prothrombin time. The protein C activity was measured by STACHROM PROTEIN C II (Boehringer Mannheim, Mannheim, Germany) or COAMATIC Protein C (CHROMOGENIX, Milan, Italy) according to the manufacturers' manuals. The prothrombin time was measured by PT-Test Wako (Wako Pure Chemical, Osaka, Japan) or Simplastin Excel (bioMérieux, Inc., Durham, NC) according to the manufacturers' manuals. STACHROM PROTEIN C II and COAMATIC Protein C measure the functional levels of protein C in plasma by an amidolytic method using a synthetic chromogenic substrate (2-oxothiazolidin-4-yl)carbonyl-Pro-Arg-p-nitroaniline. Specifically, plasma containing protein C is incubated with a specific activator extracted from snake venom (Agkistrodon c. contortrix). The p-nitroaniline released from the chromogenic substrate by activated protein C is measured at 405 nm. PT-Test Wako and Simplastin Excel are based on the assay principle that the addition of an adequate amount of tissue factor (factor III) and Ca2+ to citrated plasma activates factor VII, inducing the formation of a stable plug. Statistical analysis. Differences between values were analyzed by the twotailed Mann–Whitney U test using the built-in statistics function of Microsoft Excel (Version 11.2).
Fig. 2. Oral administration of lapachol enhanced the experimental metastasis of B16BL6 melanoma cells in T-cell-deficient and NK-suppressed mice. Lapachol (80 mg/kg) was administered orally 6 h before tumor cell injection. 2.5 × 104 cells were injected iv into mice, and the numbers of metastases were measured after 14 days. Data indicate the mean ± SE of 5 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
Effect of in vitro treatment with lapachol on the experimental metastasis of B16BL6 melanoma cells To examine whether oral administration of lapachol promotes metastasis by directly affecting tumor cells or host factors, B16BL6 melanoma cells were treated with lapachol in vitro for 24 h and then injected iv into mice. The plasma concentration of lapachol in humans after a single oral dose of 35–40 mg/kg has been reported to be 5.6–26.4 μg/ml within 24 h (Block et al., 1974). We examined the effect of lapachol on B16BL6 melanoma cells in culture and found that treatment for 24 h with N 25 μg/ml lapachol completely killed B16BL6 melanoma cells (data not shown). In vitro treatment with lapachol at 20 μg/ml slightly enhanced the number of metastatic nodules in the lung (Fig. 1D), but this increase was much less than that observed after oral administration of lapachol (Fig. 1A and C). These results indicate that the promotion of metastasis by lapachol is largely derived from effects on host factors and not from direct effects on tumor cells. Promotion of experimental metastasis of B16BL6 melanoma cells by oral administration of lapachol in T-cell-deficient mice and NK-suppressed mice Immune cells, such as T cells and NK cells, are postulated to be responsible for detecting and eliminating tumor cells (Trin-
235
chieri, 1989; Jakobisiak et al., 2003; Mehlen and Puisieux, 2006). To examine whether metastasis enhancement by lapachol is mediated by these immune cells, a metastasis assay was conducted using T-cell-deficient nude mice and NK-suppressed mice. When B16BL6 melanoma cells were injected iv into these mice 6 h after oral administration of lapachol, the number of metastatic nodules in the lung increased drastically (Fig. 2), as in normal mice (Fig. 1C). These observations indicate that lapachol can promote metastasis in the absence of T cells and NK cells. However, the possibility that lapachol affects T cells and NK cells cannot be excluded without examining the effects of lapachol on these immune cells in normal mice. Promotion of experimental metastasis of B16BL6 melanoma cells by vitamin K antagonist function of lapachol Lapachol is a potent inhibitor of vitamin K epoxide reductase and vitamin K quinone reductase (Preusch and Suttie, 1984). To clarify whether vitamin K antagonism by lapachol is involved in the promotion of metastasis, we examined the effect of warfarin, the most common orally administered vitamin K antagonist, on the experimental metastasis of B16BL6 melanoma cells. A single oral administration of warfarin 6 h before tumor cell injection dramatically promoted the metastasis of B16BL6 melanoma cells (Fig. 3A). However, the effective dose of warfarin
Fig. 3. Lapachol enhanced metastasis by its function as a vitamin K antagonist. (A) Warfarin was administered orally 6 h before tumor cell injection (5 × 104 cells/ mouse). Two weeks later, mice were euthanized, and the numbers of metastases were counted. Data indicate the mean ± SE of 5 mice (0 mg/kg) and 4 mice (0.03, 0.1, 0.33, and 1.0 mg/kg). (B) Vitamin K3 (100 mg/kg or 3 mg/kg) was administered ip three times before lapachol or warfarin administration, respectively. Lapachol (80 mg/kg) or warfarin (0.33 mg/kg) was administered orally 6 h before tumor cell injection (5 × 104 cells/mouse). Vitamin K3 (100 mg/kg) alone was administered to six mice. Two weeks later, mice were euthanized, and the numbers of metastases were measured. Data indicate the mean ± SE of 5–6 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
236
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
(0.33 mg/kg) was much lower than that of lapachol (80 mg/kg) (Fig. 1A). To confirm that metastasis promotion by lapachol and warfarin is mediated by vitamin K-dependent pathways, vitamin K3 was administered ip three times before oral administration of lapachol or warfarin, and then B16BL6 melanoma cells were intravenously injected into mice. Vitamin K3 clearly suppressed metastasis promotion by both lapachol and warfarin (Fig. 3B). These results indicate that promotion of metastasis by lapachol and warfarin originates from the antagonization of vitamin K-dependent pathways. Protein C activity and prothrombin time after oral administration of vitamin K antagonist Vitamin K antagonists can transiently induce a hypercoagulable state upon initiation of treatment (Vigano et al., 1984; Stirling, 1995; Srinivasan et al., 2004). A hypercoagulable state promotes blood-borne metastasis in cancer patients (Bick, 1992; Rickles et al., 1992; Mousa et al., 2006) and in experimental animals (Cliffton et al., 1961; Cliffton and Agostino, 1964). To examine whether hypercoagulability induced by lapachol might promote metastasis, protein C activity and the prothrombin time were measured after oral administration of lapachol. Protein C activity declined significantly within 3 h of lapachol administration, was reduced maximally between 6 and 12 h, and recovered slowly between 12 and 24 h. On the other hand, the prothrombin time was prolonged significantly only at 12 h after lapachol administration (Fig. 4A). Six hours after lapachol administration, protein C activity was maximally suppressed, but the prothrombin time was not prolonged. Therefore, the hypercoagulable state occurred at least 6 h after lapachol administration. These time courses for the hypercoagulable state were consistent with the observations of increased metastasis when tumor cells were injected 6 h after lapachol administration and of a lack of metastasis promotion when tumor cells were injected N24 h after lapachol administration (Fig. 1B). It had been reported that individuals with one-third to onehalf the normal level of protein C may exhibit impaired regulation of thrombosis and predisposition to thrombotic disease (Griffin et al., 1981). The two-standard-deviation range of protein C levels in 9854 adults was reported to be 70–140% (Tait et al., 1993). The normal range of protein C antigen in 30 normal human plasma samples was 73.7–140% (LaRosa et al., 2006). Furthermore, the coagulability of mouse plasma was shown to be greater than that of human plasma (Fernandez et al., 2003). Therefore, it is probable that a 30% reduction in the protein C level results in a hypercoagulable state in mice, although the precise protein C threshold is unknown. To further confirm that the transient hypercoagulable state induced by vitamin K antagonist administration promoted metastasis, protein C activity and the prothrombin time were measured after oral administration of warfarin. Warfarin induced changes in protein C activity and the prothrombin time that were similar to those induced by lapachol. Protein C activity declined within 3 h of warfarin administration, was reduced maximally between 6 and 12 h, and recovered slowly between
Fig. 4. Changes in protein C activity and prothrombin time after oral administration of lapachol and warfarin. Blood samples were taken at 1, 3, 6, 12, 18, 24, and 48 h after oral administration of lapachol (80 mg/kg) (A) or warfarin (0.33 mg/kg) (B), and protein C activity (closed circles) and prothrombin time (open circles) were measured. Data indicate the mean ± SE of 5 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
12 and 48 h. The prothrombin time was prolonged significantly only at 12 h after warfarin administration (Fig. 4B). Although the protein C activity decreased to a greater extent and recovered more slowly in warfarin-administered mice than in lapacholadministered mice, the hypercoagulable state occurred ∼ 6 h after the administration of both drugs. These results strongly suggest that the transient hypercoagulable state induced by lapachol administration promotes metastasis of B16BL6 melanoma cells to the lung. Discussion A major pharmacological function of vitamin K antagonists is anticoagulation, and the suppression of metastasis by anticoagulants has been reported in animal models and clinical trials (Hejna et al., 1999; Bobek and Kovarik, 2004). Several reports have shown that warfarin, the most common orally administered vitamin K antagonist, inhibits metastasis by its effects on coagulation (Agostino et al., 1966; Ryan et al., 1968; Brown, 1973;
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
McCulloch and George, 1987). In these reports, the vitamin K antagonists inhibited metastasis after the anticoagulant effect had been established. However, vitamin K antagonists can suppress both anticoagulant (protein C) and procoagulant (factors II, IX, and X) vitamin K-dependent proteins. Furthermore, because protein C has a shorter half-life than the procoagulant vitamin K-dependent factors, vitamin K antagonists suppress protein C activity more rapidly. Therefore, vitamin K antagonists can induce a transient hypercoagulable state at the start of treatment (Vigano et al., 1984; Stirling, 1995), and the risk of hypercoagulability early in the course of warfarin therapy has been noted in various situations (Broekmans et al., 1983; Srinivasan et al., 2004). To our knowledge, there is no previous report that describes the effect of vitamin K antagonists on tumor metastasis during a hypercoagulable period. In the present study, we showed that early in the course of oral administration of a vitamin K antagonist such as lapachol or warfarin, experimental metastasis was promoted dramatically. We suggested that the transient hypercoagulable state that is induced by these agents promoted the metastasis. A considerable number of cancer cells exist in the venous blood of cancerbearing patients (Engell, 1955; Sandberg et al., 1958; Goldblatt and Nadel, 1965) and animals (Glaves, 1983; Ota et al., 1999). Although the majority of cancer cells in the bloodstream are eliminated by random and selective events (Fidler, 1973; Butler and Gullino, 1975; Weiss et al., 1982), hypercoagulability promotes blood-borne metastasis in cancer patients (Bick, 1992; Rickles et al., 1992; Mousa et al., 2006) and in experimental animals (Cliffton et al., 1961; Cliffton and Agostino, 1964). Therefore, the hypercoagulable state induced early in the administration of a vitamin K antagonist may be undesirable, especially in cancer patients. Coumarins, which are structurally similar to warfarin, are ubiquitous in green plants. Over 1300 naturally occurring coumarins have been identified, although all do not necessarily exhibit anticoagulant effects (Hoult and Paya, 1996). Not surprisingly, many herbs affect coagulation (Norred and Brinker, 2001; Kumar et al., 2005). These herbs and/or herb–warfarin interactions increase the risk of hemorrhage in surgical patients (Miller, 1998; Norred and Brinker, 2001). In contrast to this anticoagulant effect, our results indicate that metastasis promotion by vitamin K antagonists is due to a transient hypercoagulable state induced in the early period after administration. Most consumers believe that herbal remedies are “natural” and therefore intrinsically harmless (Eisenberg et al., 1998). However, for individuals with critical problems in the coagulation state, such as cancer patients (Sun et al., 1979; Bick, 1992), herbs containing vitamin K antagonists should be administered with caution. Although anticoagulants such as warfarin and heparin affect metastasis, the mechanism is not derived solely from the anticoagulation activity (Hejna et al., 1999; Stevenson et al., 2007; Tagalakis et al., 2007). Proteins involved in coagulation, such as protein C and protein S, have functions other than blood coagulation. Activated protein C exerts cytoprotective effects including the alteration of gene expression profiles, anti-inflammatory activities, antiapoptotic activity, and the protection of endothelial
237
barrier function (Mosnier et al., 2007). Protein S was suggested to act as a growth/survival/phagocytic factor in various cells and tissues (Hafizi and Dahlback, 2006). Furthermore, some vitamin K-dependent proteins are not involved in blood coagulation, and these proteins regulate other physiological processes such as bone metabolism, vascular calcification, cell growth, and apoptosis (Cranenburg et al., 2007). For example, Gas6 (growtharrest-specific gene 6) has regulatory functions associated with the growth, migration, survival, and adhesion of cells (Hafizi and Dahlback, 2006). Although we hypothesize that the transient hypercoagulable state induced by vitamin K antagonists promotes metastasis, we cannot exclude the possibility that vitamin K antagonists affect metastasis by impairing the functions of proteins not involved in blood coagulation. The minimum effective dose of orally administered lapachol for antitumor activity against Walker 256 rat carcinoma is 90 mg/kg (Rao et al., 1968). However, in clinical studies, orally administered lapachol is toxic at doses greater than 2.0 g/day (28.6 mg/kg for a patient of 70-kg body weight) (Block et al., 1974). We observed promotion of metastasis only by high toxic doses of lapachol (N 80 mg/kg), and we found suppression of metastasis at much lower doses (0.5–20 mg/kg) of lapachol. In clinical studies, no toxicity was observed for doses below 1.5 g/day (21.4 mg/kg for a patient of 70-kg body weight) (Block et al., 1974). These observations raise the possibility that a non-toxic, low dose of lapachol is useful for preventing metastasis. The mechanism of metastatic suppression by low-dose lapachol is unknown. However, anticoagulation activity does not seem to participate in the metastatic suppression because a marked prolongation of prothrombin time was observed only at lapachol doses greater than 2.0 g/day (28.6 mg/kg for a patient of 70-kg body weight) in clinical studies (Block et al., 1974). Because the anticoagulation activity of lapachol was one of the reasons for terminating clinical studies, further investigations are required to clarify whether the metastasissuppressing and anticoagulation activities of lapachol are separable and whether the metastasis-suppressing activity depends on vitamin K antagonism. Conflict of interest There are no conflicts of interest to disclose.
Acknowledgments This research was supported in part by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (18590706 and 18590455), Project Research from the High Technology Center (H2007-10), and a Grant for Promoted Research (S2007-4) of Kanazawa Medical University. References Agostino, D., Cliffton, E.E., Girolami, A., 1966. Effect of prolonged coumadin treatment on the production of pulmonary metastases in the rat. Cancer 19, 284–288. Balassiano, I.T., De Paulo, S.A., Henriques Silva, N., Cabral, M.C., da Gloria da Costa Carvalho, M., 2005. Demonstration of the lapachol as a potential drug for reducing cancer metastasis. Oncol. Rep. 13, 329–333.
238
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
Bick, R.L., 1992. Coagulation abnormalities in malignancy: a review. Semin. Thromb. Hemost. 18, 353–372. Block, J.B., Serpick, A.A., Miller, W., Wiernik, P.H., 1974. Early clinical studies with lapachol (NSC-11905). Cancer Chemother. Rep. 4, 27–28 (Part 2). Bobek, V., Kovarik, J., 2004. Antitumor and antimetastatic effect of warfarin and heparins. Biomed. Pharmacother. 58, 213–219. Broekmans, A.W., Bertina, R.M., Loeliger, E.A., Hofmann, V., Klingemann, H.G., 1983. Protein C and the development of skin necrosis during anticoagulant therapy. Thromb. Haemost. 49, 251. Brown, J.M., 1973. A study of the mechanism by which anticoagulation with warfarin inhibits blood-borne metastases. Cancer Res. 33, 1217–1224. Butler, T.P., Gullino, P.M., 1975. Quantitation of cell shedding into efferent blood of mammary adenocarcinoma. Cancer Res. 35, 512–516. Cliffton, E.E., Agostino, D., 1964. Effect of inhibitors of fibrinolytic enzymes on development of pulmonary metastases. J. Natl. Cancer Inst. 33, 753–763. Cliffton, E.E., Agostino, D., Minde, K., 1961. The effect of hyperlipemia on pulmonary metastases of Walker 256 carcinosarcoma in the rat. Cancer Res. 21, 1062–1067. Cranenburg, E.C., Schurgers, L.J., Vermeer, C., 2007. Vitamin K: the coagulation vitamin that became omnipotent. Thromb. Haemost. 98, 120–125. Eisenberg, D.M., Davis, R.B., Ettner, S.L., Appel, S., Wilkey, S., Van Rompay, M., Kessler, R.C., 1998. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA 280, 1569–1575. Engell, H.C., 1955. Cancer cells in the circulating blood. Acta Chir. Scand. 201, 9–70 (Suppl.). Fernandez, J.A., Xu, X., Liu, D., Zlokovic, B.V., Griffin, J.H., 2003. Recombinant murine-activated protein C is neuroprotective in a murine ischemic stroke model. Blood Cells Mol. Diseases 30, 271–276. Fidler, I.J., 1973. Metastasis: quantitative analysis of distribution and fate of tumor emboli labeled with 125I-5-iodo-2′-deoxyuridine. J. Natl. Cancer Inst. 45, 773–782. Glaves, D., 1983. Correlation between circulating cancer cells and incidence of metastases. Br. J. Cancer 48, 665–673. Goldblatt, S.A., Nadel, E.M., 1965. Cancer cells in the circulating blood: a critical review II. Acta Cytol. 9, 6–20. Griffin, J.H., Evatt, B., Zimmerman, T.S., Kleiss, A.J., Wideman, C., 1981. Deficiency of protein C in congenital thrombotic disease. J. Clin. Invest. 68, 1370–1373. Hafizi, S., Dahlback, B., 2006. Gas6 and protein S. Vitamin K-dependent ligands for the Axl receptor tyrosine kinase subfamily. FEBS J. 273, 5231–5244. Hejna, M., Raderer, M., Zielinski, C.C., 1999. Inhibition of metastases by anticoagulants. J. Natl. Cancer Inst. 91, 22–36. Hilgard, P., Maat, B., 1979. Mechanism of lung tumour colony reduction caused by coumarin anticoagulation. Eur. J. Cancer 15, 183–187. Hoult, J.R.S., Paya, M., 1996. Pharmacological and biochemical actions of simple coumarins: natural products with therapeutic potential. Gen. Pharmacol. 27, 713–722. Jakobisiak, M., Lasek, W., Golab, J., 2003. Natural mechanisms protecting against cancer. Immunol. Lett. 90, 103–122. Kumar, N.B., Allen, K., Bell, H., 2005. Perioperative herbal supplement use in cancer patients: potential implications and recommendations for presurgical screening. Cancer Control 12, 149–157. LaRosa, S.P., Opal, S.M., Utterback, B., Yan, S.C.B., Helterbrand, J., Simpson, A.J.H., Chaowagul, W., White, N.J., Fisher, J.C.J., 2006. Decreased protein C, protein S, and antithrombin levels are predictive of poor outcome in Gram-negative sepsis caused by Burkholderia pseudomallei. Int. J. Infect. Dis. 10, 25–31. McCulloch, P., George, W.D., 1987. Warfarin inhibition of metastasis: the role of anticoagulation. Br. J. Surg. 74, 879–883. Mehlen, P., Puisieux, A., 2006. Metastasis: a question of life or death. Nat. Rev., Cancer 6, 449–458.
Miller, L.G., 1998. Herbal medicinals: selected clinical considerations focusing on known or potential drug–herb interactions. Arch. Intern. Med. 158, 2200–2211. Mosnier, L.O., Zlokovic, B.V., Griffin, J.H., 2007. The cytoprotective protein C pathway. Blood 109, 3161–3172. Mousa, S.A., Linhardt, R., Francis, J.L., Amirkhosravi, A., 2006. Anti-metastatic effect of a non-anticoagulant low-molecular-weight heparin versus the standard low-molecular-weight heparin, enoxaparin. Thromb. Haemost. 96, 816–821. Nakano, T., Matsui, T., Ota, T., 1996. Benzyl-alpha-GalNAc inhibits sialylation of O-glycosidic sugar chains on CD44 and enhances experimental metastatic capacity in B16BL6 melanoma cells. Anticancer Res. 16, 3577–3584. Norred, C.L., Brinker, F., 2001. Potential coagulation effects of preoperative complementary and alternative medicines. Altern. Ther. Health Med. 7, 58–67. Ota, T., Tanino, M., Kohno, H., Funamoto, H., Odashima, S., 1992. Isolation and characterization of a low metastatic variant from EL-4 mouse T-lymphoma. Clin. Exp. Metastasis 10, 297–308. Ota, T., Maeda, M., Matsui, T., Kohno, H., Tanino, M., Odashima, S., 1996. Inhibition of metastasis by a dialysable factor in fetal bovine serum in B16 melanoma cells. Cancer Lett. 110, 201–205. Ota, T., Maeda, M., Tatsuka, M., Matsui, T., Tanino, M., Tanaka, T., 1999. Decrease of metastatic ability after selection for intravasating ability in Lewis lung carcinoma (3LL) cell line. Cancer Lett. 139, 105–108. Preusch, P.C., Suttie, J.W., 1984. Lapachol inhibition of vitamin K epoxide reductase and vitamin K quinone reductase. Arch. Biochem. Biophys. 234, 405–412. Rao, K.V., McBridge, T.J., Oleson, J.J., 1968. Recognition and evaluation of lapachol as an antitumor agent. Cancer Res. 28, 1952–1954. Rickles, F.R., Levine, M., Edwards, R.L., 1992. Hemostatic alterations in cancer patients. Cancer Metastasis Rev. 11, 237–248. Ryan, J.J., Ketcham, A.S., Wexler, H., 1968. Reduced incidence of spontaneous metastases with long-term Coumadin therapy. Ann. Surg. 168, 163–168. Sandberg, A.A., Moore, G.E., Crosswhite, L.H., Schubarg, J.R., 1958. The frequency of tumor cells in the bone marrow and blood. Cancer 11, 1180–1186. Sieber, S.M., Mead, J.A., Adamson, R.H., 1976. Pharmacology of antitumor agents from higher plants. Cancer Treat. Rep. 60, 1127–1139. Srinivasan, A.F., Rice, L., Bartholomew, J.R., Rangaswamy, C., La Perna, L., Thompson, J.E., Murphy, S., Baker, K.R., 2004. Warfarin-induced skin necrosis and venous limb gangrene in the setting of heparin-induced thrombocytopenia. Arch. Intern. Med. 164, 66–70. Stevenson, J.L., Varki, A., Borsig, L., 2007. Heparin attenuates metastasis mainly due to inhibition of P- and L-selectin, but non-anticoagulant heparins can have additional effects. Thromb. Res. 120, S107–S111. Stirling, Y., 1995. Warfarin-induced changes in procoagulant and anticoagulant proteins. Blood Coagul. Fibrinolysis 6, 361–373. Sun, N.C., McAfee, W.M., Hum, G.J., Weiner, J.M., 1979. Hemostatic abnormalities in malignancy, a prospective study of one hundred eight patients. Part I. Coagulation studies. Am. J. Clin. Pathol. 71, 10–16. Tagalakis, V., Blostein, M., Robinson-Cohen, C., Kahn, S.R., 2007. The effect of anticoagulants on cancer risk and survival: systematic review. Cancer Treat. Rev. 33, 358–368. Tait, R.C., Walker, I.D., Islam, S.I., McCall, F., Conkie, J.A., Wight, M., Mitchell, R., Davidson, J.F., 1993. Protein C activity in healthy volunteers— influence of age, sex, smoking and oral contraceptives. Thromb. Haemost. 70, 281–285. Trinchieri, G., 1989. Biology of natural killer cells. Adv. Immunol. 47, 187–376. Vigano, S., Mannucci, P.M., Solinas, S., Bottasso, B., Mariani, G., 1984. Decrease in protein C antigen and formation of an abnormal protein soon after starting oral anticoagulant therapy. Br. J. Haematol. 57, 213–220. Weiss, L., Mayhew, E., Rapp, D.G., 1982. Metastatic inefficiency in mice bearing B16 melanomas. Br. J. Cancer 45, 44–53.
Toxicology and Applied Pharmacology 229 (2008) 232 – 238 www.elsevier.com/locate/ytaap
Promotion or suppression of experimental metastasis of B16 melanoma cells after oral administration of lapachol Masayo Maeda a , Manabu Murakami b , Tsutomu Takegami b , Takahide Ota b,⁎ a
b
Department of Chemistry, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan Division of Molecular Oncology and Virology, Medical Research Institute, Kanazawa Medical University, Uchinada, Ishikawa 920-0293, Japan Received 14 November 2007; revised 19 December 2007; accepted 14 January 2008 Available online 26 January 2008
Abstract Lapachol [2-hydroxy-3-(3-methyl-2-butenyl)-1,4-naphthoquinone] is a vitamin K antagonist with antitumor activity. The effect of lapachol on the experimental metastasis of murine B16BL6 melanoma cells was examined. A single oral administration of a high toxic dose of lapachol (80–100 mg/kg) 6 h before iv injection of tumor cells drastically promoted metastasis. This promotion of metastasis was also observed in T-cell-deficient mice and NK-suppressed mice. In vitro treatment of B16BL6 cells with lapachol promoted metastasis only slightly, indicating that lapachol promotes metastasis primarily by affecting host factors other than T cells and NK cells. A single oral administration of warfarin, the most commonly used vitamin K antagonist, 6 h before iv injection of tumor cells also drastically promoted the metastasis of B16BL6 cells. The promotion of metastasis by lapachol and warfarin was almost completely suppressed by preadministration of vitamin K3, indicating that the promotion of metastasis by lapachol was derived from vitamin K antagonism. Six hours after oral administration of lapachol or warfarin, the protein C level was reduced maximally, without elongation of prothrombin time. These observations suggest that a high toxic dose of lapachol promotes metastasis by inducing a hypercoagulable state as a result of vitamin K-dependent pathway inhibition. On the other hand, serial oral administration of low non-toxic doses of lapachol (5–20 mg/kg) weakly but significantly suppressed metastasis by an unknown mechanism, suggesting the possible use of lapachol as an antimetastatic agent. © 2008 Elsevier Inc. All rights reserved. Keywords: Cancer; Metastasis; B16 melanoma; Vitamin K antagonist; Lapachol; Warfarin; Herbal medicines
Introduction Lapachol [2-hydroxy-3-(3-methyl-2-butenyl)-1,4-naphthoquinone] was identified as the active substance in extracts of the roots of Stereospermum suaveolens and showed antitumor activity against the rat Walker 256 carcinoma (Rao et al., 1968). A phase I clinical trial of lapachol was performed, revealing an anticoagulant side effect (Block et al., 1974; Sieber et al., 1976). An effective plasma concentration for antitumor activity could not be achieved without toxicity in cancer patients receiving lapachol (Block et al., 1974; Sieber et al., 1976). For these reasons, the clinical studies with lapachol were terminated. The mechanism underlying the antitumor effect of lapachol is still unknown. However, the anticoagulant activity of lapa⁎ Corresponding author. Fax: +81 76 286 3652. E-mail address: [email protected] (T. Ota). 0041-008X/$ - see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.taap.2008.01.008
chol arises from its potent inhibition of vitamin K epoxide reductase and vitamin K quinone reductase (Preusch and Suttie, 1984). Other vitamin K antagonists have been shown to be potent anti-metastatic drugs (Agostino et al., 1966; Ryan et al., 1968; Brown, 1973; McCulloch and George, 1987), but the anti-metastatic effect is not mediated solely via their anticoagulant activity (Hilgard and Maat, 1979; Hejna et al., 1999; Tagalakis et al., 2007). Recently, lapachol was shown to inhibit the invasiveness of HeLa cells in vitro, and a potential antimetastatic activity of lapachol was suggested (Balassiano et al., 2005). To clarify whether lapachol has anti-metastatic activity in vivo, we examined the effect of lapachol on the experimental metastasis of murine B16BL6 melanoma cells in the lung. Our results suggest that a high toxic dose of lapachol, a vitamin K antagonist, transiently induces a hypercoagulable state, thereby promoting the metastasis of B16BL6 melanoma cells in the
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
233
Fig. 1. Oral administration of a high toxic dose of lapachol increased the experimental metastasis of murine B16BL6 melanoma cells. In (A)–(D), 5 × 104 B16BL6 melanoma cells were injected iv, and the number of metastases was measured after 14 days. (A) Lapachol was administered orally three times (at 48, 24, and 6 h) before tumor cell injection. Data indicate the mean ± SE of 12 mice (control) and 7 mice (lapachol-administered). (B) Lapachol (0.5 or 5.0 mg/kg) was administered orally three, six, or nine times before tumor cell injection. Data indicate the mean ± SE of 5–6 mice. (C) Lapachol (80 mg/kg) was administered orally once (at 48, 24, or 6 h) before tumor cell injection. Data indicate the mean ± SE of 6 mice (control) and 5 mice (− 48 h, − 24 h, and −6 h). (D) B16BL6 melanoma cells were treated with lapachol for 24 h in culture, and then cells were injected into mice. Data indicate the mean ± SE of 5 mice. ⁎p b 0.05 and ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
234
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
lung. These results indicate the potential risk of promotion of cancer metastasis by vitamin K antagonists. On the other hand, serial oral administration of low doses of lapachol weakly but significantly suppressed metastasis, suggesting the possible use of lapachol as an anti-metastatic agent. Materials and methods Cells and culture. The B16BL6 subline of murine B16 melanoma was kindly provided by Dr. T. Sasaki (Department of Experimental Therapeutics, Cancer Research Inst., Kanazawa University, Kanazawa, Japan). Cells were cultured in a mixture of Ham's F10 and L-15 (ratio, 3:7) containing 10% fetal bovine serum (FBS), penicillin (50 U/ml), and streptomycin (50 μg/ml) in a humidified atmosphere of 5% CO2/95% air at 37 °C. New cultures were initiated from frozen stock after 10 subcultures (Nakano et al., 1996). Drug administration and treatment. Lapachol, warfarin, and vitamin K3 were purchased from Sigma Chemical Company (St. Louis, MO). Lapachol was dissolved in 2% sodium carbonate in water, and warfarin was dissolved in water. These drugs were administered by oral gavage. Vitamin K3 was dissolved in sterile phosphate-buffered saline (PBS) and administered ip. The solutions were freshly prepared just before use. For in vitro treatment, lapachol was dissolved in ethanol at a concentration of 40 mM and added to cultures. Ethanol was added to the control culture at a concentration of 0.05%. Assay of experimental metastasis. B16BL6 cells were harvested with 0.02% EDTA in Hanks' Balanced Salt Solution (HBSS). Harvested cells were washed once with medium containing FBS and once with HBSS. Cells with N90% viability as determined by the trypan blue dye exclusion test were used for experiments. Cells (5 × 104 or 2.5 × 104) suspended in 0.2 ml of HBSS were injected into the lateral tail veins of mice. Two weeks after injection of cells, the mice were euthanized, and the numbers of lung tumor nodules were counted under a dissecting microscope as described (Ota et al., 1996).
Results Effect of oral administration of lapachol on the experimental metastasis of B16BL6 melanoma cells Lapachol was administered orally to mice three times (at − 48, − 24, and − 6 h) before tumor cell injection into the lateral tail vain. Two weeks after tumor cell injection, mice were euthanized, and the numbers of metastatic nodules in the lungs were counted. Low doses (0.5, 5, 10, and 20 mg/kg) of lapachol weakly but significantly inhibited metastasis, whereas higher doses (80 and 100 mg/kg) drastically promoted metastasis (Fig. 1A). To confirm the metastasis-inhibiting effect of lowdose lapachol, we increased the frequency of lapachol administration at 0.5 and 5.0 mg/kg. Although lapachol showed a tendency to inhibit metastasis, the inhibition did not increase with increased frequency of administration (Fig. 1B). In contrast to the weak inhibition of metastasis, the promotion of metastasis was substantial. Because the strong promotion of metastasis by a therapeutic agent calls for great concern, we further examined the metastasis-promoting property of lapachol. To determine the sensitive period for metastasis promotion by lapachol, lapachol (80 mg/kg) was administered once at 48, 24, or 6 h before tumor cell injection. Only the administration at 6 h before tumor cell injection significantly increased the number of metastatic nodules in the lung (Fig. 1C). These results indicate that lapachol promotes metastasis within 6 h of administration, and this effect disappears within 24 h.
Animals. Seven-week-old male C57BL/6 normal mice and five-week-old male BALB/c nude mice were obtained from Japan SLC, Inc. (Shizuoka, Japan) and were maintained under specific pathogen-free conditions. C57BL/6 normal mice and BALB/c nude mice were used for experiments at 8 weeks and 6–7 weeks of age, respectively. To eliminate NK activity in mice, anti-asialoGM1 was used (Ota et al., 1992). Rabbit anti-asialoGM1 serum was purchased from Wako Pure Chemical (Osaka, Japan). The antiserum was diluted 1:8 in PBS and was filtered through Millipore filters (0.22 μm) before injection. Three days before tumor cell inoculation, male C57BL/6 mice were injected iv with 0.2 ml of the diluted antiserum. The same volume of normal rabbit serum at a 1:8 dilution in PBS was injected into control mice in place of the anti-asialoGM1 serum. Experiments were performed in accordance with the guidelines of the Committee on Experimental Animals at Kanazawa Medical University. Measurement of protein C activity and prothrombin time. The protein C activity was measured by STACHROM PROTEIN C II (Boehringer Mannheim, Mannheim, Germany) or COAMATIC Protein C (CHROMOGENIX, Milan, Italy) according to the manufacturers' manuals. The prothrombin time was measured by PT-Test Wako (Wako Pure Chemical, Osaka, Japan) or Simplastin Excel (bioMérieux, Inc., Durham, NC) according to the manufacturers' manuals. STACHROM PROTEIN C II and COAMATIC Protein C measure the functional levels of protein C in plasma by an amidolytic method using a synthetic chromogenic substrate (2-oxothiazolidin-4-yl)carbonyl-Pro-Arg-p-nitroaniline. Specifically, plasma containing protein C is incubated with a specific activator extracted from snake venom (Agkistrodon c. contortrix). The p-nitroaniline released from the chromogenic substrate by activated protein C is measured at 405 nm. PT-Test Wako and Simplastin Excel are based on the assay principle that the addition of an adequate amount of tissue factor (factor III) and Ca2+ to citrated plasma activates factor VII, inducing the formation of a stable plug. Statistical analysis. Differences between values were analyzed by the twotailed Mann–Whitney U test using the built-in statistics function of Microsoft Excel (Version 11.2).
Fig. 2. Oral administration of lapachol enhanced the experimental metastasis of B16BL6 melanoma cells in T-cell-deficient and NK-suppressed mice. Lapachol (80 mg/kg) was administered orally 6 h before tumor cell injection. 2.5 × 104 cells were injected iv into mice, and the numbers of metastases were measured after 14 days. Data indicate the mean ± SE of 5 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
Effect of in vitro treatment with lapachol on the experimental metastasis of B16BL6 melanoma cells To examine whether oral administration of lapachol promotes metastasis by directly affecting tumor cells or host factors, B16BL6 melanoma cells were treated with lapachol in vitro for 24 h and then injected iv into mice. The plasma concentration of lapachol in humans after a single oral dose of 35–40 mg/kg has been reported to be 5.6–26.4 μg/ml within 24 h (Block et al., 1974). We examined the effect of lapachol on B16BL6 melanoma cells in culture and found that treatment for 24 h with N 25 μg/ml lapachol completely killed B16BL6 melanoma cells (data not shown). In vitro treatment with lapachol at 20 μg/ml slightly enhanced the number of metastatic nodules in the lung (Fig. 1D), but this increase was much less than that observed after oral administration of lapachol (Fig. 1A and C). These results indicate that the promotion of metastasis by lapachol is largely derived from effects on host factors and not from direct effects on tumor cells. Promotion of experimental metastasis of B16BL6 melanoma cells by oral administration of lapachol in T-cell-deficient mice and NK-suppressed mice Immune cells, such as T cells and NK cells, are postulated to be responsible for detecting and eliminating tumor cells (Trin-
235
chieri, 1989; Jakobisiak et al., 2003; Mehlen and Puisieux, 2006). To examine whether metastasis enhancement by lapachol is mediated by these immune cells, a metastasis assay was conducted using T-cell-deficient nude mice and NK-suppressed mice. When B16BL6 melanoma cells were injected iv into these mice 6 h after oral administration of lapachol, the number of metastatic nodules in the lung increased drastically (Fig. 2), as in normal mice (Fig. 1C). These observations indicate that lapachol can promote metastasis in the absence of T cells and NK cells. However, the possibility that lapachol affects T cells and NK cells cannot be excluded without examining the effects of lapachol on these immune cells in normal mice. Promotion of experimental metastasis of B16BL6 melanoma cells by vitamin K antagonist function of lapachol Lapachol is a potent inhibitor of vitamin K epoxide reductase and vitamin K quinone reductase (Preusch and Suttie, 1984). To clarify whether vitamin K antagonism by lapachol is involved in the promotion of metastasis, we examined the effect of warfarin, the most common orally administered vitamin K antagonist, on the experimental metastasis of B16BL6 melanoma cells. A single oral administration of warfarin 6 h before tumor cell injection dramatically promoted the metastasis of B16BL6 melanoma cells (Fig. 3A). However, the effective dose of warfarin
Fig. 3. Lapachol enhanced metastasis by its function as a vitamin K antagonist. (A) Warfarin was administered orally 6 h before tumor cell injection (5 × 104 cells/ mouse). Two weeks later, mice were euthanized, and the numbers of metastases were counted. Data indicate the mean ± SE of 5 mice (0 mg/kg) and 4 mice (0.03, 0.1, 0.33, and 1.0 mg/kg). (B) Vitamin K3 (100 mg/kg or 3 mg/kg) was administered ip three times before lapachol or warfarin administration, respectively. Lapachol (80 mg/kg) or warfarin (0.33 mg/kg) was administered orally 6 h before tumor cell injection (5 × 104 cells/mouse). Vitamin K3 (100 mg/kg) alone was administered to six mice. Two weeks later, mice were euthanized, and the numbers of metastases were measured. Data indicate the mean ± SE of 5–6 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
236
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
(0.33 mg/kg) was much lower than that of lapachol (80 mg/kg) (Fig. 1A). To confirm that metastasis promotion by lapachol and warfarin is mediated by vitamin K-dependent pathways, vitamin K3 was administered ip three times before oral administration of lapachol or warfarin, and then B16BL6 melanoma cells were intravenously injected into mice. Vitamin K3 clearly suppressed metastasis promotion by both lapachol and warfarin (Fig. 3B). These results indicate that promotion of metastasis by lapachol and warfarin originates from the antagonization of vitamin K-dependent pathways. Protein C activity and prothrombin time after oral administration of vitamin K antagonist Vitamin K antagonists can transiently induce a hypercoagulable state upon initiation of treatment (Vigano et al., 1984; Stirling, 1995; Srinivasan et al., 2004). A hypercoagulable state promotes blood-borne metastasis in cancer patients (Bick, 1992; Rickles et al., 1992; Mousa et al., 2006) and in experimental animals (Cliffton et al., 1961; Cliffton and Agostino, 1964). To examine whether hypercoagulability induced by lapachol might promote metastasis, protein C activity and the prothrombin time were measured after oral administration of lapachol. Protein C activity declined significantly within 3 h of lapachol administration, was reduced maximally between 6 and 12 h, and recovered slowly between 12 and 24 h. On the other hand, the prothrombin time was prolonged significantly only at 12 h after lapachol administration (Fig. 4A). Six hours after lapachol administration, protein C activity was maximally suppressed, but the prothrombin time was not prolonged. Therefore, the hypercoagulable state occurred at least 6 h after lapachol administration. These time courses for the hypercoagulable state were consistent with the observations of increased metastasis when tumor cells were injected 6 h after lapachol administration and of a lack of metastasis promotion when tumor cells were injected N24 h after lapachol administration (Fig. 1B). It had been reported that individuals with one-third to onehalf the normal level of protein C may exhibit impaired regulation of thrombosis and predisposition to thrombotic disease (Griffin et al., 1981). The two-standard-deviation range of protein C levels in 9854 adults was reported to be 70–140% (Tait et al., 1993). The normal range of protein C antigen in 30 normal human plasma samples was 73.7–140% (LaRosa et al., 2006). Furthermore, the coagulability of mouse plasma was shown to be greater than that of human plasma (Fernandez et al., 2003). Therefore, it is probable that a 30% reduction in the protein C level results in a hypercoagulable state in mice, although the precise protein C threshold is unknown. To further confirm that the transient hypercoagulable state induced by vitamin K antagonist administration promoted metastasis, protein C activity and the prothrombin time were measured after oral administration of warfarin. Warfarin induced changes in protein C activity and the prothrombin time that were similar to those induced by lapachol. Protein C activity declined within 3 h of warfarin administration, was reduced maximally between 6 and 12 h, and recovered slowly between
Fig. 4. Changes in protein C activity and prothrombin time after oral administration of lapachol and warfarin. Blood samples were taken at 1, 3, 6, 12, 18, 24, and 48 h after oral administration of lapachol (80 mg/kg) (A) or warfarin (0.33 mg/kg) (B), and protein C activity (closed circles) and prothrombin time (open circles) were measured. Data indicate the mean ± SE of 5 mice. ⁎⁎p b 0.01 compared with the control group by two-tailed U test.
12 and 48 h. The prothrombin time was prolonged significantly only at 12 h after warfarin administration (Fig. 4B). Although the protein C activity decreased to a greater extent and recovered more slowly in warfarin-administered mice than in lapacholadministered mice, the hypercoagulable state occurred ∼ 6 h after the administration of both drugs. These results strongly suggest that the transient hypercoagulable state induced by lapachol administration promotes metastasis of B16BL6 melanoma cells to the lung. Discussion A major pharmacological function of vitamin K antagonists is anticoagulation, and the suppression of metastasis by anticoagulants has been reported in animal models and clinical trials (Hejna et al., 1999; Bobek and Kovarik, 2004). Several reports have shown that warfarin, the most common orally administered vitamin K antagonist, inhibits metastasis by its effects on coagulation (Agostino et al., 1966; Ryan et al., 1968; Brown, 1973;
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
McCulloch and George, 1987). In these reports, the vitamin K antagonists inhibited metastasis after the anticoagulant effect had been established. However, vitamin K antagonists can suppress both anticoagulant (protein C) and procoagulant (factors II, IX, and X) vitamin K-dependent proteins. Furthermore, because protein C has a shorter half-life than the procoagulant vitamin K-dependent factors, vitamin K antagonists suppress protein C activity more rapidly. Therefore, vitamin K antagonists can induce a transient hypercoagulable state at the start of treatment (Vigano et al., 1984; Stirling, 1995), and the risk of hypercoagulability early in the course of warfarin therapy has been noted in various situations (Broekmans et al., 1983; Srinivasan et al., 2004). To our knowledge, there is no previous report that describes the effect of vitamin K antagonists on tumor metastasis during a hypercoagulable period. In the present study, we showed that early in the course of oral administration of a vitamin K antagonist such as lapachol or warfarin, experimental metastasis was promoted dramatically. We suggested that the transient hypercoagulable state that is induced by these agents promoted the metastasis. A considerable number of cancer cells exist in the venous blood of cancerbearing patients (Engell, 1955; Sandberg et al., 1958; Goldblatt and Nadel, 1965) and animals (Glaves, 1983; Ota et al., 1999). Although the majority of cancer cells in the bloodstream are eliminated by random and selective events (Fidler, 1973; Butler and Gullino, 1975; Weiss et al., 1982), hypercoagulability promotes blood-borne metastasis in cancer patients (Bick, 1992; Rickles et al., 1992; Mousa et al., 2006) and in experimental animals (Cliffton et al., 1961; Cliffton and Agostino, 1964). Therefore, the hypercoagulable state induced early in the administration of a vitamin K antagonist may be undesirable, especially in cancer patients. Coumarins, which are structurally similar to warfarin, are ubiquitous in green plants. Over 1300 naturally occurring coumarins have been identified, although all do not necessarily exhibit anticoagulant effects (Hoult and Paya, 1996). Not surprisingly, many herbs affect coagulation (Norred and Brinker, 2001; Kumar et al., 2005). These herbs and/or herb–warfarin interactions increase the risk of hemorrhage in surgical patients (Miller, 1998; Norred and Brinker, 2001). In contrast to this anticoagulant effect, our results indicate that metastasis promotion by vitamin K antagonists is due to a transient hypercoagulable state induced in the early period after administration. Most consumers believe that herbal remedies are “natural” and therefore intrinsically harmless (Eisenberg et al., 1998). However, for individuals with critical problems in the coagulation state, such as cancer patients (Sun et al., 1979; Bick, 1992), herbs containing vitamin K antagonists should be administered with caution. Although anticoagulants such as warfarin and heparin affect metastasis, the mechanism is not derived solely from the anticoagulation activity (Hejna et al., 1999; Stevenson et al., 2007; Tagalakis et al., 2007). Proteins involved in coagulation, such as protein C and protein S, have functions other than blood coagulation. Activated protein C exerts cytoprotective effects including the alteration of gene expression profiles, anti-inflammatory activities, antiapoptotic activity, and the protection of endothelial
237
barrier function (Mosnier et al., 2007). Protein S was suggested to act as a growth/survival/phagocytic factor in various cells and tissues (Hafizi and Dahlback, 2006). Furthermore, some vitamin K-dependent proteins are not involved in blood coagulation, and these proteins regulate other physiological processes such as bone metabolism, vascular calcification, cell growth, and apoptosis (Cranenburg et al., 2007). For example, Gas6 (growtharrest-specific gene 6) has regulatory functions associated with the growth, migration, survival, and adhesion of cells (Hafizi and Dahlback, 2006). Although we hypothesize that the transient hypercoagulable state induced by vitamin K antagonists promotes metastasis, we cannot exclude the possibility that vitamin K antagonists affect metastasis by impairing the functions of proteins not involved in blood coagulation. The minimum effective dose of orally administered lapachol for antitumor activity against Walker 256 rat carcinoma is 90 mg/kg (Rao et al., 1968). However, in clinical studies, orally administered lapachol is toxic at doses greater than 2.0 g/day (28.6 mg/kg for a patient of 70-kg body weight) (Block et al., 1974). We observed promotion of metastasis only by high toxic doses of lapachol (N 80 mg/kg), and we found suppression of metastasis at much lower doses (0.5–20 mg/kg) of lapachol. In clinical studies, no toxicity was observed for doses below 1.5 g/day (21.4 mg/kg for a patient of 70-kg body weight) (Block et al., 1974). These observations raise the possibility that a non-toxic, low dose of lapachol is useful for preventing metastasis. The mechanism of metastatic suppression by low-dose lapachol is unknown. However, anticoagulation activity does not seem to participate in the metastatic suppression because a marked prolongation of prothrombin time was observed only at lapachol doses greater than 2.0 g/day (28.6 mg/kg for a patient of 70-kg body weight) in clinical studies (Block et al., 1974). Because the anticoagulation activity of lapachol was one of the reasons for terminating clinical studies, further investigations are required to clarify whether the metastasissuppressing and anticoagulation activities of lapachol are separable and whether the metastasis-suppressing activity depends on vitamin K antagonism. Conflict of interest There are no conflicts of interest to disclose.
Acknowledgments This research was supported in part by Grants-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (18590706 and 18590455), Project Research from the High Technology Center (H2007-10), and a Grant for Promoted Research (S2007-4) of Kanazawa Medical University. References Agostino, D., Cliffton, E.E., Girolami, A., 1966. Effect of prolonged coumadin treatment on the production of pulmonary metastases in the rat. Cancer 19, 284–288. Balassiano, I.T., De Paulo, S.A., Henriques Silva, N., Cabral, M.C., da Gloria da Costa Carvalho, M., 2005. Demonstration of the lapachol as a potential drug for reducing cancer metastasis. Oncol. Rep. 13, 329–333.
238
M. Maeda et al. / Toxicology and Applied Pharmacology 229 (2008) 232–238
Bick, R.L., 1992. Coagulation abnormalities in malignancy: a review. Semin. Thromb. Hemost. 18, 353–372. Block, J.B., Serpick, A.A., Miller, W., Wiernik, P.H., 1974. Early clinical studies with lapachol (NSC-11905). Cancer Chemother. Rep. 4, 27–28 (Part 2). Bobek, V., Kovarik, J., 2004. Antitumor and antimetastatic effect of warfarin and heparins. Biomed. Pharmacother. 58, 213–219. Broekmans, A.W., Bertina, R.M., Loeliger, E.A., Hofmann, V., Klingemann, H.G., 1983. Protein C and the development of skin necrosis during anticoagulant therapy. Thromb. Haemost. 49, 251. Brown, J.M., 1973. A study of the mechanism by which anticoagulation with warfarin inhibits blood-borne metastases. Cancer Res. 33, 1217–1224. Butler, T.P., Gullino, P.M., 1975. Quantitation of cell shedding into efferent blood of mammary adenocarcinoma. Cancer Res. 35, 512–516. Cliffton, E.E., Agostino, D., 1964. Effect of inhibitors of fibrinolytic enzymes on development of pulmonary metastases. J. Natl. Cancer Inst. 33, 753–763. Cliffton, E.E., Agostino, D., Minde, K., 1961. The effect of hyperlipemia on pulmonary metastases of Walker 256 carcinosarcoma in the rat. Cancer Res. 21, 1062–1067. Cranenburg, E.C., Schurgers, L.J., Vermeer, C., 2007. Vitamin K: the coagulation vitamin that became omnipotent. Thromb. Haemost. 98, 120–125. Eisenberg, D.M., Davis, R.B., Ettner, S.L., Appel, S., Wilkey, S., Van Rompay, M., Kessler, R.C., 1998. Trends in alternative medicine use in the United States, 1990–1997: results of a follow-up national survey. JAMA 280, 1569–1575. Engell, H.C., 1955. Cancer cells in the circulating blood. Acta Chir. Scand. 201, 9–70 (Suppl.). Fernandez, J.A., Xu, X., Liu, D., Zlokovic, B.V., Griffin, J.H., 2003. Recombinant murine-activated protein C is neuroprotective in a murine ischemic stroke model. Blood Cells Mol. Diseases 30, 271–276. Fidler, I.J., 1973. Metastasis: quantitative analysis of distribution and fate of tumor emboli labeled with 125I-5-iodo-2′-deoxyuridine. J. Natl. Cancer Inst. 45, 773–782. Glaves, D., 1983. Correlation between circulating cancer cells and incidence of metastases. Br. J. Cancer 48, 665–673. Goldblatt, S.A., Nadel, E.M., 1965. Cancer cells in the circulating blood: a critical review II. Acta Cytol. 9, 6–20. Griffin, J.H., Evatt, B., Zimmerman, T.S., Kleiss, A.J., Wideman, C., 1981. Deficiency of protein C in congenital thrombotic disease. J. Clin. Invest. 68, 1370–1373. Hafizi, S., Dahlback, B., 2006. Gas6 and protein S. Vitamin K-dependent ligands for the Axl receptor tyrosine kinase subfamily. FEBS J. 273, 5231–5244. Hejna, M., Raderer, M., Zielinski, C.C., 1999. Inhibition of metastases by anticoagulants. J. Natl. Cancer Inst. 91, 22–36. Hilgard, P., Maat, B., 1979. Mechanism of lung tumour colony reduction caused by coumarin anticoagulation. Eur. J. Cancer 15, 183–187. Hoult, J.R.S., Paya, M., 1996. Pharmacological and biochemical actions of simple coumarins: natural products with therapeutic potential. Gen. Pharmacol. 27, 713–722. Jakobisiak, M., Lasek, W., Golab, J., 2003. Natural mechanisms protecting against cancer. Immunol. Lett. 90, 103–122. Kumar, N.B., Allen, K., Bell, H., 2005. Perioperative herbal supplement use in cancer patients: potential implications and recommendations for presurgical screening. Cancer Control 12, 149–157. LaRosa, S.P., Opal, S.M., Utterback, B., Yan, S.C.B., Helterbrand, J., Simpson, A.J.H., Chaowagul, W., White, N.J., Fisher, J.C.J., 2006. Decreased protein C, protein S, and antithrombin levels are predictive of poor outcome in Gram-negative sepsis caused by Burkholderia pseudomallei. Int. J. Infect. Dis. 10, 25–31. McCulloch, P., George, W.D., 1987. Warfarin inhibition of metastasis: the role of anticoagulation. Br. J. Surg. 74, 879–883. Mehlen, P., Puisieux, A., 2006. Metastasis: a question of life or death. Nat. Rev., Cancer 6, 449–458.
Miller, L.G., 1998. Herbal medicinals: selected clinical considerations focusing on known or potential drug–herb interactions. Arch. Intern. Med. 158, 2200–2211. Mosnier, L.O., Zlokovic, B.V., Griffin, J.H., 2007. The cytoprotective protein C pathway. Blood 109, 3161–3172. Mousa, S.A., Linhardt, R., Francis, J.L., Amirkhosravi, A., 2006. Anti-metastatic effect of a non-anticoagulant low-molecular-weight heparin versus the standard low-molecular-weight heparin, enoxaparin. Thromb. Haemost. 96, 816–821. Nakano, T., Matsui, T., Ota, T., 1996. Benzyl-alpha-GalNAc inhibits sialylation of O-glycosidic sugar chains on CD44 and enhances experimental metastatic capacity in B16BL6 melanoma cells. Anticancer Res. 16, 3577–3584. Norred, C.L., Brinker, F., 2001. Potential coagulation effects of preoperative complementary and alternative medicines. Altern. Ther. Health Med. 7, 58–67. Ota, T., Tanino, M., Kohno, H., Funamoto, H., Odashima, S., 1992. Isolation and characterization of a low metastatic variant from EL-4 mouse T-lymphoma. Clin. Exp. Metastasis 10, 297–308. Ota, T., Maeda, M., Matsui, T., Kohno, H., Tanino, M., Odashima, S., 1996. Inhibition of metastasis by a dialysable factor in fetal bovine serum in B16 melanoma cells. Cancer Lett. 110, 201–205. Ota, T., Maeda, M., Tatsuka, M., Matsui, T., Tanino, M., Tanaka, T., 1999. Decrease of metastatic ability after selection for intravasating ability in Lewis lung carcinoma (3LL) cell line. Cancer Lett. 139, 105–108. Preusch, P.C., Suttie, J.W., 1984. Lapachol inhibition of vitamin K epoxide reductase and vitamin K quinone reductase. Arch. Biochem. Biophys. 234, 405–412. Rao, K.V., McBridge, T.J., Oleson, J.J., 1968. Recognition and evaluation of lapachol as an antitumor agent. Cancer Res. 28, 1952–1954. Rickles, F.R., Levine, M., Edwards, R.L., 1992. Hemostatic alterations in cancer patients. Cancer Metastasis Rev. 11, 237–248. Ryan, J.J., Ketcham, A.S., Wexler, H., 1968. Reduced incidence of spontaneous metastases with long-term Coumadin therapy. Ann. Surg. 168, 163–168. Sandberg, A.A., Moore, G.E., Crosswhite, L.H., Schubarg, J.R., 1958. The frequency of tumor cells in the bone marrow and blood. Cancer 11, 1180–1186. Sieber, S.M., Mead, J.A., Adamson, R.H., 1976. Pharmacology of antitumor agents from higher plants. Cancer Treat. Rep. 60, 1127–1139. Srinivasan, A.F., Rice, L., Bartholomew, J.R., Rangaswamy, C., La Perna, L., Thompson, J.E., Murphy, S., Baker, K.R., 2004. Warfarin-induced skin necrosis and venous limb gangrene in the setting of heparin-induced thrombocytopenia. Arch. Intern. Med. 164, 66–70. Stevenson, J.L., Varki, A., Borsig, L., 2007. Heparin attenuates metastasis mainly due to inhibition of P- and L-selectin, but non-anticoagulant heparins can have additional effects. Thromb. Res. 120, S107–S111. Stirling, Y., 1995. Warfarin-induced changes in procoagulant and anticoagulant proteins. Blood Coagul. Fibrinolysis 6, 361–373. Sun, N.C., McAfee, W.M., Hum, G.J., Weiner, J.M., 1979. Hemostatic abnormalities in malignancy, a prospective study of one hundred eight patients. Part I. Coagulation studies. Am. J. Clin. Pathol. 71, 10–16. Tagalakis, V., Blostein, M., Robinson-Cohen, C., Kahn, S.R., 2007. The effect of anticoagulants on cancer risk and survival: systematic review. Cancer Treat. Rev. 33, 358–368. Tait, R.C., Walker, I.D., Islam, S.I., McCall, F., Conkie, J.A., Wight, M., Mitchell, R., Davidson, J.F., 1993. Protein C activity in healthy volunteers— influence of age, sex, smoking and oral contraceptives. Thromb. Haemost. 70, 281–285. Trinchieri, G., 1989. Biology of natural killer cells. Adv. Immunol. 47, 187–376. Vigano, S., Mannucci, P.M., Solinas, S., Bottasso, B., Mariani, G., 1984. Decrease in protein C antigen and formation of an abnormal protein soon after starting oral anticoagulant therapy. Br. J. Haematol. 57, 213–220. Weiss, L., Mayhew, E., Rapp, D.G., 1982. Metastatic inefficiency in mice bearing B16 melanomas. Br. J. Cancer 45, 44–53.