- Email: [email protected]
Intravenous anesthetic, propofol inhibits invasion of cancer cells
Cancer Letters 184 (2002) 165–170 www.elsevier.com/locate/canlet
Intravenous anesthetic, propofol inhibits invasion of cancer cells Tadanori Mammoto a,b,*, Mutsuko Mukai b, Akiko Mammoto c, Yasutsugu Yamanaka b, Yukio Hayashi d, Takashi Mashimo d, Yoshihiko Kishi a, Hiroyuki Nakamura b a
Department of Anesthesiology, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3, Nakamichi, Higashinari, Osaka 537-8511, Japan b Department of Tumor Biochemistry, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3, Nakamichi, Higashinari, Osaka 537-8511, Japan c Departments of Biochemistry, School of Allied Health Science, Osaka University, Faculty of Medicine, 1-7, Yamadaoka, Suita, Osaka 565-0871, Japan d Department of Anesthesiology, Osaka University Graduate School of Medicine, Faculty of Medicine, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan Received 4 February 2002; received in revised form 25 March 2002; accepted 8 April 2002
Abstract Intravenous anesthetic, propofol (2,6-diisopropylphenol), is extensively used for general anesthesia without knowing the effects on cancer. We found here that clinically relevant concentrations of propofol (1–5 mg/ml) decreased the invasion ability of human cancer cells (HeLa, HT1080, HOS and RPMI-7951). In the HeLa cells treated with propofol, formation of actin stress fibers as well as focal adhesion were inhibited, and propofol had little effect on the invasion ability of the HeLa cells with active Rho A (Val 14-Rho A). In addition, continuous infusion of propofol inhibited pulmonary metastasis of murine osteosarcoma (LM 8) cells in mice. These results suggest that propofol inhibits the invasion ability of cancer cells by modulating Rho A and this agent might be an ideal anesthetic for cancer surgery. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cancer cells; Anesthetic; Propofol; Invasion; Metastasis
1. Introduction A variety of surgical operations such as biopsy and resection are performed for cancer treatment and general anesthesia is usually performed for these procedures. It has been reported that general anesthetics impair immune response [1–3] and that surgical operations induce dissemination of cancer cells into surrounding tissues or into circulation [4,5]. These findings imply that invasion or metastasis of * Corresponding author. Tel.: 181-6-6972-1181; fax: 181-66972-7749. E-mail address: [email protected] (T. Mammoto).
cancer cells easily occur during the surgical procedures. However, the effect of anesthetics on the behavior of cancer is unclear and various anesthetics are used for cancer resection without knowing the effects on cancer. Thus, it is surely important to clarify the effects of anesthetics on the behavior of cancer cells. Intravenous anesthetic, propofol (2,6-diisopropylphenol) (Fig. 1), provides smooth induction and rapid recovery from anesthesia and this agent is extensively used for general anesthesia and sedation [6]. We found that propofol at the concentrations required for surgical operations decreased invasion ability of human cancer cells. It has been demonstrated that the Rho family of
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(02)00210-0
166
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
used (1–5 mg/ml) had little effect on the viability of the cells for 16 h by using a Trypan Blue dye exclusion test. 2.3. Assay for invasion ability
Fig. 1. Chemical structure of propofol.
GTPases, Rho A, modulates integrin clustering and associated formation of actin stress fibers [7]. This assembly is important for cell migration [8] and activation of endogenous Rho A has been reported to increase tumor cell invasion [9]. We hypothesized that propofol inhibits the invasion ability of cancer cells by modulating Rho A and evaluated effects of clinically relevant concentrations of propofol on the integrin clustering and the formation of actin stress fibers in HeLa cells. In addition, we investigated the anti-invasive action of propofol on the HeLa cells with active Rho A. The anti-invasive action of propofol was also confirmed by using an in vivo pulmonary metastasis model. 2. Methods 2.1. Reagents Unless otherwise indicated, all chemical reagents were obtained from Sigma (St. Louis, MO). Propofol (2,6-diisopropylphenol) was purchased from Aldrich (Milwaukee, WI). Propofol was diluted in dimethyl sulfoxide (DMSO) for in vitro assays or in 10% soybean oil for an in vivo assay. 2.2. Cells and cell cultures HeLa human cervix carcinoma cells, HOS human osteosarcoma cells and RPMI-7951 human melanoma cells were purchased from ATCC (Manassas, VA), and HT1080 human fibrosarcoma cells were purchased from HSRRB (Osaka, Japan). We confirmed that the concentrations of propofol we
In chemoinvasion assay, cancer cells migrate through the filter coated with type IV collagen and this assay could assess the invasive ability of the cancer cells quantitatively [10]. We used this assay to assess the invasion ability of the cancer cells. Polycarbonate membranes (8 mm pore size) of transwell cell culture inserts (Coster, NY) were coated with 100 ml of type IV collagen (Koken, Tokyo, Japan) at 200 mg/ml in PBS. HeLa, HT1080, HOS and RPMI-7951 cells (4 £ 10 5 cells, respectively) in minimum essential medium (MEM) with albumin (0.5% vol) were seeded into the upper wells. In our preliminary study, checkerboard analysis indicated that these cells migrated to fibronectin (FN) and lower wells of chambers were filled by MEM supplemented with 10 mg/ml of FN as chemoattractant. Chambers were subsequently incubated for 16 h in the absence or presence of propofol (1, 3 or 5 mg/ml). After incubation, the upper surface of the membranes was scrapped with a cotton swab and cells that had reached the lower surface of the membranes were stained with Giemsa solution. The cells were counted with a light microscope ( £ 400) at 20 random fields. 2.4. Immunofluorescent staining for integrin clustering and actin stress fibers Clustering of a5b1 integrin receptors led to the formation of focal adhesions [8,11] and we stained the integrin clustering and actin stress fibers in the HeLa cells. HeLa cells (2 £ 10 4 cells) in MEM with albumin (0.5% vol) were plated on FN-coated dishes, incubated for 1 h, washed with PBS, and fixed in 1.8% paraformaldehyde in PBS for 20 min. Then, the cells were permeabilized in 1% Triton X-100 for 10 min and blocked with 10% FCS in PBS. The primary antibody used was anti-integrin b1 antibody (Chemicon International Inc., Temecula, CA) (1:50), and the secondary antibody used was FITC conjugated antimouse antibody (Organon-Teknika-CAPEL, Durham, NC) (1:50). Actin filaments were doubly stained with Rhodamine-Phalloidin (Molecular Probe, Leiden,
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
The Netherlands) (1:50). The effects of propofol (5 mg/ml) on the formation of actin stress fibers were also evaluated. The cells were examined with an ECLIPSE E800 fluorescence microscope (Nikon, Tokyo, Japan). 2.5. Transfection with active Rho A (Val 14-Rho A)
167
suppress the invasion ability of cancer cells (Fig. 2). One to 5 mg/ml propofol inhibited the invasion ability of HeLa, HT1080, HOS and RPMI-7951 cells dosedependently (Fig. 2). 3.2. The effects of propofol on integrin clustering and actin stress fiber formation
In order to evaluate the involvement of Rho A in the anti-invasive action of propofol, we constructed an expression plasmid encoding an active Rho A (Val 14Rho A) [9]. The plasmid was stably transfected into HeLa cells by use of Superfect reagent (Qiagen), and the cell clones were isolated by G418. The invasion ability of the transfectant was evaluated by the invasion assay described above.
Plated on FN for 60 min, b1 integrin clustering (Fig. 3A) and formation of actin stress fibers (Fig. 3B) were observed. Propofol (5 mg/ml) inhibited both b1 integrin clustering (Fig. 3C) and formation of actin stress fibers (Fig. 3D).
2.6. Assay for pulmonary metastasis
One to 5 mg/ml propofol had little effect on the invasion ability of HeLa cells with active Rho A (Val 14-Rho A) (Fig. 4). The invasion ability of mock transfectant was similar to that of wild HeLa cells (data not shown).
We confirmed the anti-invasive action of propofol by using an in vivo pulmonary metastasis model [12]. Vehicle (DMSO) or propofol was dissolved in 100 ml of 10% soybean oil and put into a mini-osmotic pump 1002 (alza, Palo Alto, CA) that supplies its content at a rate of 0.25 ml/h. Under anesthesia, the osmotic pump was implanted into the peritoneal cavity of female C3H mice. Then, murine osteosarcoma (LM 8) cells, suspended in PBS (10 7 cells/200 ml), were injected into subcutaneous space on the backs of the mice. We adjusted the propofol concentration so that 20 or 40 mg/kg per day propofol was infused continuously to the mice. The primary nodules and pulmonary metastasis were evaluated 4 weeks later.
3.3. The effect of propofol on the invasion ability of active Rho A HeLa cells
3.4. The effects of propofol on the invasion and metastatic ability of LM 8 cells As with human cancer cells, propofol inhibited the invasion ability of LM 8 cells dose-dependently (Fig. 5A). Inoculated LM 8 cells grew up at the inoculated site to form a single large nodule (Fig. 5B,a) and made numerous pulmonary metastatic nodules 4 weeks later (Fig. 5B,b). Although continuous infusion of propofol
2.7. Statistics Statistical analyses of results on chemoinvasion, migration and adhesion assays were analyzed with ANOVA. Significance was determined by the paired t-test. 3. Results 3.1. The effects of propofol on the invasion ability of human cancer cells The target plasma concentrations of propofol for general anesthesia are between 3 and 8 mg/ml [13], and these concentrations of propofol were sufficient to
Fig. 2. Effects of propofol on the invasion ability of cancer cells. Invasion ability of four cancer cell lines (HeLa, HT1080, HOS and RPMI-7951) in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mg/ml propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments.
168
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
Fig. 3. Effects of propofol on the b1 integrin clustering and formation of actin stress fibers. Assembly of integrin clustering and stress fibers of HeLa cells during adhesion on FN substrate for 60 min. (A,C) Stained for integrin; (B,D) stained for actin stress fibers ( £ 1000). (C,D) Cells incubated in the presence of propofol (5 mg/ml).
(both 20 and 40 mg/kg per day) for 4 weeks had little effect on the growth of the cancer cells at the inoculated site, 40 mg/kg per day propofol infusion decreased the number of pulmonary nodules significantly (Fig. 5C). Continuous infusion of propofol (20 or 40 mg/kg per day) into the peritoneal cavity by the osmotic pump did not induce sedation nor anesthesia. In addition, the feeding behavior of mice had not been impaired and the weight of the mice was similar in each treatment.
fibers [7]. In focal adhesions, integrin links to intracellular cytoskeletal complexes and bundles of the actin filament [7,14]. The actin stress fibers generate contractile force for forward movement of the cell [8]. Thus, the integrin clustering and formation of the actin stress fibers seem to be a critical process for migration and invasion of the cancer cells. Propofol inhibited formation of the integrin clustering as well as the actin stress fibers and the anti-invasive action of propofol was diminished in the cancer cells with active Rho A. These results suggest that propofol decreased the invasion ability of cancer cells by modulating Rho A. However, propofol has been reported to inhibit the immune response [1,2] and invasion is a complex phenomenon in vivo and it is important to evaluate the anti-invasive action of propofol in vivo. In the pulmonary metastasis model we used, the cancer cells grew at the inoculated site and then invaded the circulation and consequently formed the distant pulmonary metastasis. Continuous infusion of propofol inhibited the pulmonary metastasis and it is suggested that this agent also has anti-invasive action in vivo. Although we could not determine the plasma concentrations of propofol in our experiment, the
4. Discussion In clinical anesthesia, the target plasma propofol concentrations for general anesthesia are considered to be 3–8 mg/ml [13] and these concentrations of propofol effectively inhibited the invasion ability of human cancer cells. To invade the normal tissues, cancer cells should adhere to extracellular matrix (ECM) and migrate through degraded ECM into the circulation. It has been demonstrated that Rho A modulates the focal adhesions and associated formation of the actin stress
Fig. 4. Effects of propofol on the HeLa cells with active Rho A. The invasion ability of the HeLa cells with active Rho A (Val 14-Rho A) was determined in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mg/ml propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments.
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
169
Fig. 5. Effects of propofol on LM 8 cells. (A) Invasion ability of LM 8 cells in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mM propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments. (B) LM 8 cells (10 7) were injected into subcutaneous space on the backs of the mice and vehicle or propofol (20 or 40 mg/kg per day) was continuously infused into the peritoneal cavity by mini-osmotic pump. Four weeks later the primary nodule (a: arrow) and metastatic nodules on the lungs (b) were formed. (C) The number of metastatic nodules on the lung and the weight of the primary tumor were evaluated. Results are mean ^ SD and statistically significant differences from vehicle are indicated (*P , 0:01). Twenty mice were included in each group and similar results were obtained in at least three different experiments.
170
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
feeding behaviors as well as locomotor activity of mice were not impaired in the propofol-treated groups, and we consider that the plasma concentrations of propofol are sub-anesthetic in each treatment. It is noteworthy that the sub-anesthetic propofol infusion inhibited the pulmonary metastasis without affecting the growth of primary nodules. Propofol would prevent pulmonary metastasis of cancer cells by inhibiting the invasion of cancer cells rather than by inhibiting the growth of cancer cells. In conclusion, clinically relevant concentrations of propofol inhibited the invasion of human cancer cells effectively and modulation of Rho A possibly contributes to this anti-invasive action of propofol. These results suggest that propofol might be an ideal anesthetic for cancer surgery. Acknowledgements We are indebted to Drs Itoh, Yamatodani, Yoshikawa, Higashiyama, Shinkai, and Komatsu for their helpful comments on our manuscript. We also thank Miss Nagamachi, Mrs Yasuda and Dr Kita for their assistance in this investigation. References [1] N. Kotani, H. Hashimoto, D.I. Sessler, A. Kikuchi, A. Suzuki, S. Takahashi, M. Muraoka, A. Matsuki, Intraoperative modulation of alveolar macrophage function during isoflurane and propofol anesthesia, Anesthesiology 89 (1998) 1125–1132. [2] K. Mikawa, H. Akamatsu, K. Nishina, M. Shiga, N. Maekawa, H. Obara, Y. Niwa, Propofol inhibits human neutrophil functions, Anesth. Analg. 87 (1998) 695–700.
[3] W. Krumholz, D. Reussner, G. Hempelmann, The influence of several intravenous anaesthetics on the chemotaxis of human monocytes in vitro, Eur. J. Anaesthesiol. 16 (1999) 547–549. [4] J. Kusukawa, Y. Suefuji, F. Ryu, R. Noguchi, O. Iwamoto, T. Kameyama, Dissemination of cancer cells into circulation occurs by incisional biopsy of oral squamous cell carcinoma, J. Oral Pathol. Med. 29 (2000) 303–307. [5] K. Yamaguchi, Y. Takagi, S. Aoki, M. Futamura, S. Saji, Significant detection of circulating cancer cells in the blood by reverse transcriptase-polymerase chain reaction during colorectal cancer resection, Ann. Surg. 232 (2000) 58–65. [6] R.D. Miller, Local Anesthetics: Anesthesia, 5th Edition, Churchill Livingstone, New York, 2000, pp. 491–521. [7] A.J. Ridley, A. Hall, The small GTP-binding protein Rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors, Cell 70 (1992) 389–399. [8] S.W. Craig, R.P. Johnson, Assembly of focal adhesions: progress, paradigms, and portents, Curr. Opin. Cell Biol. 8 (1996) 74–85. [9] K. Yoshioka, F. Matsumura, H. Akedo, K. Itoh, Small GTPbinding protein Rho stimulates the actomyosin system, leading to invasion of tumor cells, J. Biol. Chem. 273 (1998) 5146–5154. [10] A. Albini, Y. Iwamoto, H.K. Kleinman, G.R. Martin, S.A. Aaronson, J.M. Kozlowski, R.N. McEwan, A rapid in vitro assay for quantitating the invasive potential of tumor cells, Cancer Res. 47 (1987) 3239–3245. [11] B.M. Gumbiner, Proteins associated with the cytoplasmic surface of adhesion molecules, Neuron 11 (1993) 551–564. [12] T. Asai, T. Ueda, K. Itoh, K. Yoshioka, Y. Aoki, S. Mori, H. Yoshikawa, Establishment and characterization of a murine osteosarcoma cell line (LM8) with high metastatic potential to the lung, Int. J. Cancer 76 (1998) 418–422. [13] J.F. Coetzee, J.B. Glen, C.A. Wium, L. Boshoff, Pharmacokinetic model selection for target controlled infusions of propofol. Assessment of three parameter sets, Anesthesiology 82 (1995) 1328–1345. [14] D.D. Schlaepfer, T. Hunter, Integrin signalling and tyrosine phosphorylation: just the FAK, Trends Cell Biol. 8 (1998) 151–157.
Intravenous anesthetic, propofol inhibits invasion of cancer cells Tadanori Mammoto a,b,*, Mutsuko Mukai b, Akiko Mammoto c, Yasutsugu Yamanaka b, Yukio Hayashi d, Takashi Mashimo d, Yoshihiko Kishi a, Hiroyuki Nakamura b a
Department of Anesthesiology, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3, Nakamichi, Higashinari, Osaka 537-8511, Japan b Department of Tumor Biochemistry, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-3, Nakamichi, Higashinari, Osaka 537-8511, Japan c Departments of Biochemistry, School of Allied Health Science, Osaka University, Faculty of Medicine, 1-7, Yamadaoka, Suita, Osaka 565-0871, Japan d Department of Anesthesiology, Osaka University Graduate School of Medicine, Faculty of Medicine, 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan Received 4 February 2002; received in revised form 25 March 2002; accepted 8 April 2002
Abstract Intravenous anesthetic, propofol (2,6-diisopropylphenol), is extensively used for general anesthesia without knowing the effects on cancer. We found here that clinically relevant concentrations of propofol (1–5 mg/ml) decreased the invasion ability of human cancer cells (HeLa, HT1080, HOS and RPMI-7951). In the HeLa cells treated with propofol, formation of actin stress fibers as well as focal adhesion were inhibited, and propofol had little effect on the invasion ability of the HeLa cells with active Rho A (Val 14-Rho A). In addition, continuous infusion of propofol inhibited pulmonary metastasis of murine osteosarcoma (LM 8) cells in mice. These results suggest that propofol inhibits the invasion ability of cancer cells by modulating Rho A and this agent might be an ideal anesthetic for cancer surgery. q 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Cancer cells; Anesthetic; Propofol; Invasion; Metastasis
1. Introduction A variety of surgical operations such as biopsy and resection are performed for cancer treatment and general anesthesia is usually performed for these procedures. It has been reported that general anesthetics impair immune response [1–3] and that surgical operations induce dissemination of cancer cells into surrounding tissues or into circulation [4,5]. These findings imply that invasion or metastasis of * Corresponding author. Tel.: 181-6-6972-1181; fax: 181-66972-7749. E-mail address: [email protected] (T. Mammoto).
cancer cells easily occur during the surgical procedures. However, the effect of anesthetics on the behavior of cancer is unclear and various anesthetics are used for cancer resection without knowing the effects on cancer. Thus, it is surely important to clarify the effects of anesthetics on the behavior of cancer cells. Intravenous anesthetic, propofol (2,6-diisopropylphenol) (Fig. 1), provides smooth induction and rapid recovery from anesthesia and this agent is extensively used for general anesthesia and sedation [6]. We found that propofol at the concentrations required for surgical operations decreased invasion ability of human cancer cells. It has been demonstrated that the Rho family of
0304-3835/02/$ - see front matter q 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(02)00210-0
166
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
used (1–5 mg/ml) had little effect on the viability of the cells for 16 h by using a Trypan Blue dye exclusion test. 2.3. Assay for invasion ability
Fig. 1. Chemical structure of propofol.
GTPases, Rho A, modulates integrin clustering and associated formation of actin stress fibers [7]. This assembly is important for cell migration [8] and activation of endogenous Rho A has been reported to increase tumor cell invasion [9]. We hypothesized that propofol inhibits the invasion ability of cancer cells by modulating Rho A and evaluated effects of clinically relevant concentrations of propofol on the integrin clustering and the formation of actin stress fibers in HeLa cells. In addition, we investigated the anti-invasive action of propofol on the HeLa cells with active Rho A. The anti-invasive action of propofol was also confirmed by using an in vivo pulmonary metastasis model. 2. Methods 2.1. Reagents Unless otherwise indicated, all chemical reagents were obtained from Sigma (St. Louis, MO). Propofol (2,6-diisopropylphenol) was purchased from Aldrich (Milwaukee, WI). Propofol was diluted in dimethyl sulfoxide (DMSO) for in vitro assays or in 10% soybean oil for an in vivo assay. 2.2. Cells and cell cultures HeLa human cervix carcinoma cells, HOS human osteosarcoma cells and RPMI-7951 human melanoma cells were purchased from ATCC (Manassas, VA), and HT1080 human fibrosarcoma cells were purchased from HSRRB (Osaka, Japan). We confirmed that the concentrations of propofol we
In chemoinvasion assay, cancer cells migrate through the filter coated with type IV collagen and this assay could assess the invasive ability of the cancer cells quantitatively [10]. We used this assay to assess the invasion ability of the cancer cells. Polycarbonate membranes (8 mm pore size) of transwell cell culture inserts (Coster, NY) were coated with 100 ml of type IV collagen (Koken, Tokyo, Japan) at 200 mg/ml in PBS. HeLa, HT1080, HOS and RPMI-7951 cells (4 £ 10 5 cells, respectively) in minimum essential medium (MEM) with albumin (0.5% vol) were seeded into the upper wells. In our preliminary study, checkerboard analysis indicated that these cells migrated to fibronectin (FN) and lower wells of chambers were filled by MEM supplemented with 10 mg/ml of FN as chemoattractant. Chambers were subsequently incubated for 16 h in the absence or presence of propofol (1, 3 or 5 mg/ml). After incubation, the upper surface of the membranes was scrapped with a cotton swab and cells that had reached the lower surface of the membranes were stained with Giemsa solution. The cells were counted with a light microscope ( £ 400) at 20 random fields. 2.4. Immunofluorescent staining for integrin clustering and actin stress fibers Clustering of a5b1 integrin receptors led to the formation of focal adhesions [8,11] and we stained the integrin clustering and actin stress fibers in the HeLa cells. HeLa cells (2 £ 10 4 cells) in MEM with albumin (0.5% vol) were plated on FN-coated dishes, incubated for 1 h, washed with PBS, and fixed in 1.8% paraformaldehyde in PBS for 20 min. Then, the cells were permeabilized in 1% Triton X-100 for 10 min and blocked with 10% FCS in PBS. The primary antibody used was anti-integrin b1 antibody (Chemicon International Inc., Temecula, CA) (1:50), and the secondary antibody used was FITC conjugated antimouse antibody (Organon-Teknika-CAPEL, Durham, NC) (1:50). Actin filaments were doubly stained with Rhodamine-Phalloidin (Molecular Probe, Leiden,
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
The Netherlands) (1:50). The effects of propofol (5 mg/ml) on the formation of actin stress fibers were also evaluated. The cells were examined with an ECLIPSE E800 fluorescence microscope (Nikon, Tokyo, Japan). 2.5. Transfection with active Rho A (Val 14-Rho A)
167
suppress the invasion ability of cancer cells (Fig. 2). One to 5 mg/ml propofol inhibited the invasion ability of HeLa, HT1080, HOS and RPMI-7951 cells dosedependently (Fig. 2). 3.2. The effects of propofol on integrin clustering and actin stress fiber formation
In order to evaluate the involvement of Rho A in the anti-invasive action of propofol, we constructed an expression plasmid encoding an active Rho A (Val 14Rho A) [9]. The plasmid was stably transfected into HeLa cells by use of Superfect reagent (Qiagen), and the cell clones were isolated by G418. The invasion ability of the transfectant was evaluated by the invasion assay described above.
Plated on FN for 60 min, b1 integrin clustering (Fig. 3A) and formation of actin stress fibers (Fig. 3B) were observed. Propofol (5 mg/ml) inhibited both b1 integrin clustering (Fig. 3C) and formation of actin stress fibers (Fig. 3D).
2.6. Assay for pulmonary metastasis
One to 5 mg/ml propofol had little effect on the invasion ability of HeLa cells with active Rho A (Val 14-Rho A) (Fig. 4). The invasion ability of mock transfectant was similar to that of wild HeLa cells (data not shown).
We confirmed the anti-invasive action of propofol by using an in vivo pulmonary metastasis model [12]. Vehicle (DMSO) or propofol was dissolved in 100 ml of 10% soybean oil and put into a mini-osmotic pump 1002 (alza, Palo Alto, CA) that supplies its content at a rate of 0.25 ml/h. Under anesthesia, the osmotic pump was implanted into the peritoneal cavity of female C3H mice. Then, murine osteosarcoma (LM 8) cells, suspended in PBS (10 7 cells/200 ml), were injected into subcutaneous space on the backs of the mice. We adjusted the propofol concentration so that 20 or 40 mg/kg per day propofol was infused continuously to the mice. The primary nodules and pulmonary metastasis were evaluated 4 weeks later.
3.3. The effect of propofol on the invasion ability of active Rho A HeLa cells
3.4. The effects of propofol on the invasion and metastatic ability of LM 8 cells As with human cancer cells, propofol inhibited the invasion ability of LM 8 cells dose-dependently (Fig. 5A). Inoculated LM 8 cells grew up at the inoculated site to form a single large nodule (Fig. 5B,a) and made numerous pulmonary metastatic nodules 4 weeks later (Fig. 5B,b). Although continuous infusion of propofol
2.7. Statistics Statistical analyses of results on chemoinvasion, migration and adhesion assays were analyzed with ANOVA. Significance was determined by the paired t-test. 3. Results 3.1. The effects of propofol on the invasion ability of human cancer cells The target plasma concentrations of propofol for general anesthesia are between 3 and 8 mg/ml [13], and these concentrations of propofol were sufficient to
Fig. 2. Effects of propofol on the invasion ability of cancer cells. Invasion ability of four cancer cell lines (HeLa, HT1080, HOS and RPMI-7951) in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mg/ml propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments.
168
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
Fig. 3. Effects of propofol on the b1 integrin clustering and formation of actin stress fibers. Assembly of integrin clustering and stress fibers of HeLa cells during adhesion on FN substrate for 60 min. (A,C) Stained for integrin; (B,D) stained for actin stress fibers ( £ 1000). (C,D) Cells incubated in the presence of propofol (5 mg/ml).
(both 20 and 40 mg/kg per day) for 4 weeks had little effect on the growth of the cancer cells at the inoculated site, 40 mg/kg per day propofol infusion decreased the number of pulmonary nodules significantly (Fig. 5C). Continuous infusion of propofol (20 or 40 mg/kg per day) into the peritoneal cavity by the osmotic pump did not induce sedation nor anesthesia. In addition, the feeding behavior of mice had not been impaired and the weight of the mice was similar in each treatment.
fibers [7]. In focal adhesions, integrin links to intracellular cytoskeletal complexes and bundles of the actin filament [7,14]. The actin stress fibers generate contractile force for forward movement of the cell [8]. Thus, the integrin clustering and formation of the actin stress fibers seem to be a critical process for migration and invasion of the cancer cells. Propofol inhibited formation of the integrin clustering as well as the actin stress fibers and the anti-invasive action of propofol was diminished in the cancer cells with active Rho A. These results suggest that propofol decreased the invasion ability of cancer cells by modulating Rho A. However, propofol has been reported to inhibit the immune response [1,2] and invasion is a complex phenomenon in vivo and it is important to evaluate the anti-invasive action of propofol in vivo. In the pulmonary metastasis model we used, the cancer cells grew at the inoculated site and then invaded the circulation and consequently formed the distant pulmonary metastasis. Continuous infusion of propofol inhibited the pulmonary metastasis and it is suggested that this agent also has anti-invasive action in vivo. Although we could not determine the plasma concentrations of propofol in our experiment, the
4. Discussion In clinical anesthesia, the target plasma propofol concentrations for general anesthesia are considered to be 3–8 mg/ml [13] and these concentrations of propofol effectively inhibited the invasion ability of human cancer cells. To invade the normal tissues, cancer cells should adhere to extracellular matrix (ECM) and migrate through degraded ECM into the circulation. It has been demonstrated that Rho A modulates the focal adhesions and associated formation of the actin stress
Fig. 4. Effects of propofol on the HeLa cells with active Rho A. The invasion ability of the HeLa cells with active Rho A (Val 14-Rho A) was determined in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mg/ml propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments.
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
169
Fig. 5. Effects of propofol on LM 8 cells. (A) Invasion ability of LM 8 cells in the absence or presence of propofol at indicated concentrations. Results are mean ^ SD and statistically significant differences from 0 mM propofol are indicated (*P , 0:01). Similar results were obtained in at least three different experiments. (B) LM 8 cells (10 7) were injected into subcutaneous space on the backs of the mice and vehicle or propofol (20 or 40 mg/kg per day) was continuously infused into the peritoneal cavity by mini-osmotic pump. Four weeks later the primary nodule (a: arrow) and metastatic nodules on the lungs (b) were formed. (C) The number of metastatic nodules on the lung and the weight of the primary tumor were evaluated. Results are mean ^ SD and statistically significant differences from vehicle are indicated (*P , 0:01). Twenty mice were included in each group and similar results were obtained in at least three different experiments.
170
T. Mammoto et al. / Cancer Letters 184 (2002) 165–170
feeding behaviors as well as locomotor activity of mice were not impaired in the propofol-treated groups, and we consider that the plasma concentrations of propofol are sub-anesthetic in each treatment. It is noteworthy that the sub-anesthetic propofol infusion inhibited the pulmonary metastasis without affecting the growth of primary nodules. Propofol would prevent pulmonary metastasis of cancer cells by inhibiting the invasion of cancer cells rather than by inhibiting the growth of cancer cells. In conclusion, clinically relevant concentrations of propofol inhibited the invasion of human cancer cells effectively and modulation of Rho A possibly contributes to this anti-invasive action of propofol. These results suggest that propofol might be an ideal anesthetic for cancer surgery. Acknowledgements We are indebted to Drs Itoh, Yamatodani, Yoshikawa, Higashiyama, Shinkai, and Komatsu for their helpful comments on our manuscript. We also thank Miss Nagamachi, Mrs Yasuda and Dr Kita for their assistance in this investigation. References [1] N. Kotani, H. Hashimoto, D.I. Sessler, A. Kikuchi, A. Suzuki, S. Takahashi, M. Muraoka, A. Matsuki, Intraoperative modulation of alveolar macrophage function during isoflurane and propofol anesthesia, Anesthesiology 89 (1998) 1125–1132. [2] K. Mikawa, H. Akamatsu, K. Nishina, M. Shiga, N. Maekawa, H. Obara, Y. Niwa, Propofol inhibits human neutrophil functions, Anesth. Analg. 87 (1998) 695–700.
[3] W. Krumholz, D. Reussner, G. Hempelmann, The influence of several intravenous anaesthetics on the chemotaxis of human monocytes in vitro, Eur. J. Anaesthesiol. 16 (1999) 547–549. [4] J. Kusukawa, Y. Suefuji, F. Ryu, R. Noguchi, O. Iwamoto, T. Kameyama, Dissemination of cancer cells into circulation occurs by incisional biopsy of oral squamous cell carcinoma, J. Oral Pathol. Med. 29 (2000) 303–307. [5] K. Yamaguchi, Y. Takagi, S. Aoki, M. Futamura, S. Saji, Significant detection of circulating cancer cells in the blood by reverse transcriptase-polymerase chain reaction during colorectal cancer resection, Ann. Surg. 232 (2000) 58–65. [6] R.D. Miller, Local Anesthetics: Anesthesia, 5th Edition, Churchill Livingstone, New York, 2000, pp. 491–521. [7] A.J. Ridley, A. Hall, The small GTP-binding protein Rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors, Cell 70 (1992) 389–399. [8] S.W. Craig, R.P. Johnson, Assembly of focal adhesions: progress, paradigms, and portents, Curr. Opin. Cell Biol. 8 (1996) 74–85. [9] K. Yoshioka, F. Matsumura, H. Akedo, K. Itoh, Small GTPbinding protein Rho stimulates the actomyosin system, leading to invasion of tumor cells, J. Biol. Chem. 273 (1998) 5146–5154. [10] A. Albini, Y. Iwamoto, H.K. Kleinman, G.R. Martin, S.A. Aaronson, J.M. Kozlowski, R.N. McEwan, A rapid in vitro assay for quantitating the invasive potential of tumor cells, Cancer Res. 47 (1987) 3239–3245. [11] B.M. Gumbiner, Proteins associated with the cytoplasmic surface of adhesion molecules, Neuron 11 (1993) 551–564. [12] T. Asai, T. Ueda, K. Itoh, K. Yoshioka, Y. Aoki, S. Mori, H. Yoshikawa, Establishment and characterization of a murine osteosarcoma cell line (LM8) with high metastatic potential to the lung, Int. J. Cancer 76 (1998) 418–422. [13] J.F. Coetzee, J.B. Glen, C.A. Wium, L. Boshoff, Pharmacokinetic model selection for target controlled infusions of propofol. Assessment of three parameter sets, Anesthesiology 82 (1995) 1328–1345. [14] D.D. Schlaepfer, T. Hunter, Integrin signalling and tyrosine phosphorylation: just the FAK, Trends Cell Biol. 8 (1998) 151–157.