Initial, continuous and intermittent bolus cardioplegia administration: efficacy of potassium-chloride and magnesium-sulfate as minimal additives for minimally-diluted blood cardioplegia.

AIM: The most effective delivery of blood cardioplegia (BCP) remains controversial, and a combination of initial continuous and intermittent bolus BCP seems to compensate each demerit. However, a large amount of crystalloid solution is infused into the myocardium in this method, which may nullify the advantages of BCP. We examined the hypothesis that minimally-diluted BCP resolves this issue and provides superior myocardial protective effects. METHODS: Seventy patients undergoing elective coronary revascularization between 1997-2001 (M:F=55:15, mean age 67.6+/-7.5 years) were randomly allocated into one of 2 groups: Group C (n=35) was given the standard 4:1-diluted blood-crystalloid BCP, and Group M (n=35) was given minimally-diluted BCP supplemented with potassium-chloride and magnesium-sulfate. The BCP temperature was maintained at 30 degrees C. Cardioplegic arrest was induced with 2 minutes of initial antegrade BCP infusion, followed by continuous retrograde BCP infusion. Intermittent antegrade BCP was infused every 30 minutes for 2 minutes. RESULTS: The time required for achieving cardioplegic arrest was significantly shorter in Group M (47.5+/-16.3 vs 62.5+/-17.6 s, p<0.0001) and the number of patients showing spontaneous heart-beat recovery after aortic unclamping was significantly larger in Group M (28 vs 15, p=0.0029). The number of patients suffering from atrial fibrillation during the postoperative period was significantly smaller in Group M (3 vs 11, p=0.034). The total amount of crystalloid solution infused as cardioplegia was significantly smaller in Group M (62.8+/-22.3 vs 733.6+/-382.6 mL, p<0.0001). Postoperative maximum dopamine dose (3.57+/-2.46 vs 5.44+/-2.23 mg/kg/min, p=0.0014) and peak creatine kinase-MB (19.5+/-8.5 vs 25.8+/-11.9 IU/L, p=0.0128) were significantly lower in Group M. The number of patients showing paradoxical movement of the ventricular septum by early postoperative echocardiography was significantly smaller in Group M (9 vs 24, p<0.0007). CONCLUSIONS: These results demonstrate that initial continuous and intermittent bolus administration of minimally-diluted BCP supplemented with potassium and magnesium can be a simple, reliable and effective technique of intraoperative myocardial protection.

Induction by bombesin of peritoneal metastasis of gastric cancers induced by N-methyl-N'-nitro-N-nitrosoguanidine in Wistar rats.

BACKGROUND: We investigated the effect of the gastrointestinal regulatory peptide, bombesin, on the development of peritoneal metastasis from gastric cancers induced in rats by N-methyl-N'-nitro-N-nitrosoguanidine (MNNG), and on Rho activity in the gastric cancers. METHODS: Rats were allocated to three groups. All groups received MNNG (100 micrograms/ml) solution for 25 weeks from the start of the experiment. Group 1 (controls) received olive oil injections from the start of MNNG treatment; group 2 animals received alternate-day s.c. injections of bombesin (40 micrograms/kg body weight) in olive oil from the start of the experiment until the end of the experiment at week 52; and group 3 received the s.c. bombesin injection on alternate days from week 26 until week 52. The effect of bombesin on Rho activity in gastric cancer was examined by Western blotting. RESULTS: Bombesin given from the start of the experiment (group 2) and after the MNNG treatment (group 3) both significantly increased the incidence of gastric cancer metastasis, compared with controls, at week 52: The incidence of metastasis was significantly higher in group 2 than in group 3. Bombesin from the start of the experiment (group 2) significantly increased the incidence of tumors with deeper invasion or more infiltrative growth pattern, or lymphatic vessel tumor invasion, while bombesin after MNNG treatment (group 3) significantly increased the incidence of lymphatic vessel invasion. Bombesin also increased the activity of Rho protein in the tumors. CONCLUSION: Bombesin significantly increased the incidence of peritoneal metastasis from gastric cancers through the activation of Rho protein.

Inhibition of epidermal growth factor-induced RhoA translocation and invasion of human pancreatic cancer cells by 3-hydroxy-3-methylglutaryl-coenzyme a reductase inhibitors.

3-Hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitors prevent the conversion of HMG-CoA to mevalonate and thereby inhibit the synthesis of other products derived from this metabolite. This includes a number of small prenylated GTPases involved in cell growth, motility, and invasion. We studied the effect of HMG-CoA reductase inhibitors (fluvastatin and lovastatin) on in vitro invasion of human pancreatic cancer PANC-1 cells. Epidermal growth factor (EGF) induced a dose-dependent increase of PANC-1 cell invasion in a modified Boyden chamber assay. Stimulation of cancer cells with EGF induced translocation of RhoA from the cytosol to the membrane fraction and actin stress fiber assembly. Furthermore, Clostridium botulinum C3 transferase, a specific inhibitor of Rho, inhibited the ability of EGF to promote invasion, indicating that EGF-induced cancer cell invasion is regulated by Rho signaling. Treatment of PANC-1 cells with fluvastatin markedly attenuated EGF-induced translocation of RhoA from the cytosol to the membrane fraction and actin stress fiber assembly, whereas it did not inhibit the tyrosine phosphorylation of EGF receptor and c-erbB-2. The induction of cancer cell invasion by EGF was inhibited by the addition of fluvastatin or lovastatin in a dose-dependent manner. The effects of fluvastatin or lovastatin on cell morphology and invasion were reversed by the addition of all-trans-geranylgeraniol but not by the addition of all-trans-farnesol. These results suggest that HMG-CoA reductase inhibitors affect RhoA activation by preventing geranylgeranylation, which results in inhibition of EGF-induced invasiveness of human pancreatic cancer cells.

Hepatoma cell migration through a mesothelial cell monolayer is inhibited by cyclic AMP-elevating agents via a Rho-dependent pathway.

1-Oleoyl lysophosphatidic acid (LPA) induces transmonolayer migration (in vitro invasion) of rat ascites hepatoma MM1 cells and their morphological changes leading to the migration. We have previously shown that an LPA analog, palmitoyl cyclic phosphatidic acid (Pal-cPA), suppresses transmonolayer migration of MM1 cells by rapidly increasing the intracellular cyclic AMP (cAMP) concentration. We report here that various cAMP-elevating agents, including dibutyryl cAMP, forskolin, cholera toxin and 3-isobutyl-1-methylxanthine, consistently inhibited LPA-induced transmonolayer migration of MM1 cells. Moreover, pull-down assays for GTP-bound, active RhoA demonstrated that the blockage by cAMP-elevating agents of morphological changes leading to the migration was probably mediated through inhibiting RhoA activation.

Y-27632, an inhibitor of rho-associated protein kinase, suppresses tumor cell invasion via regulation of focal adhesion and focal adhesion kinase.

Migration of rat ascites hepatoma (MM1) cells, invasion and phagokinetic movement were induced by the combination of lysophosphatidic acid (LPA) and fibronectin (FN). Induction of migratory activity was tightly correlated with morphological change of MM1 cells from spherical or polygonal-shaped cells to fusiform-shaped ones with pseudopodia. MM1 cells were mobile in a fusiform shape, whereas those of a spherical or polygonal shape were not. A small GTPase Rho and one of its downstream effectors ROCK (Rho-associated coiled-coil forming protein kinase), play essential roles in these processes, as evidenced by suppression of migration and morphological change of MM1 cells by Clostridium botulinum C3 exoenzyme, an inhibitor of Rho, or by Y-27632, an inhibitor of ROCK. Y-27632 also suppressed the formation of fusiform-shaped pseudopodia-carrying MM1 cells that was induced by stimulation with the combination of LPA and FN. LPA and FN also evoked the formation of focal adhesions and actin bundles, and tyrosine phosphorylation of focal adhesion kinase (FAK) and paxillin. The inhibitory effect of Y-27632 on LPA-induced migration and morphological change of MM1 cells was considered to be mediated, at least in part, by impaired formation of focal adhesions and actin bundles. Y-27632 suppressed LPA-induced tyrosine phosphorylation of FAK and paxillin, suggesting that ROCK regulates these molecules and Y-27632 inhibits cellular migration and morphological change, at least in part, through this regulation.

Paxillin alpha and Crk-associated substrate exert opposing effects on cell migration and contact inhibition of growth through tyrosine phosphorylation.

Protein tyrosine phosphorylation accompanies and is essential for integrin signaling. We have shown that tyrosine phosphorylation of paxillin alpha and Crk-associated substrate (p130(Cas)) is a prominent event on integrin activation in normal murine mammary gland epithelial cells. Tyrosine phosphorylation of p130(Cas) has been demonstrated to facilitate cell migration. We show here that tyrosine phosphorylation of paxillin alpha acts to reduce haptotactic cell migrations as well as transcellular invasive activities in several different experimental cell systems, whereas tyrosine phosphorylation of p130(Cas) exerts opposing effects to those of paxillin alpha. Each of the phosphorylation-null mutants acts as a dominant negative for each phenotype. Moreover, we found that overexpression of paxillin alpha reduced the cell saturation density of normal murine mammary gland cells, whereas overexpression of p130(Cas) increased it. These effects also seemed to depend on tyrosine phosphorylation events. Cell growth rates and morphologies at growing phases were not significantly altered, nor were cells transformed. Addition of epidermal growth factor increased saturation density of the paxillin alpha-overexpressing cells, whereas no further increment was observed in p130(Cas)-overexpressing cells. We propose that tyrosine phosphorylation of paxillin alpha and p130(Cas) exerts opposing effects on several integrin-mediated cellular events, possibly through different signaling pathways.

Increased expression of S100A6 at the invading fronts of the primary lesion and liver metastasis in patients with colorectal adenocarcinoma.

Two members of the S100 gene family, S100A6 and S100A4 have been suggested to be associated with cancer invasion and metastasis. To study their involvement in the malignancy of human colorectal adenocarcinoma, we examined the protein expression levels of S100A6 and S100A4 in the primary colorectal adenocarcinoma (T) and paired adjacent normal colorectal mucosa (N) from 12 cases, quantitatively by Western blot analysis. In 11 of 12 and seven of 12 cases, S100A6 and S100A4 expression levels were higher in T than in N, respectively. Average S100A6 level in T was significantly higher than in N (about x 2.3;P = 0.001), whereas average S100A4 level in T was not. When S100A6 expression levels in three sets of matched samples of primary colorectal adenocarcinoma (T) and liver metastasis (M) were examined, S100A6 levels were higher in M than in T in two of three cases. Immunohistochemical analysis using monoclonal anti-S100A6 antibody showed that 23 of 42 (55%) primary colorectal adenocarcinoma and 15 of 16 (94%) liver metastasis specimens were positively stained. S100A6 immunostaining of primary colorectal adenocarcinomas was significantly more intense in the invading fronts with structural atypia than in central portions with glandular structure (P< 0.0001), whereas Ki-67 staining (a growth marker) was similar in these two portions. Interestingly, S100A6 and Ki-67 immunostaining patterns in liver metastases were also the same as in primary lesions. These results suggest that S100A6 is involved in the invasive process of human colorectal adenocarcinomas and that S100A6 expression levels decrease when carcinoma cells form glandular structure again at the central portions of metastatic nodules.

Participation of a MEK-independent pathway in MAP kinase activation and modulation of cell growth in mouse hepatoma cell lines.

The mechanism of cell growth was investigated in GIT medium-supplemented in vitro assay using high and low metastatic mouse hepatoma cell sublines, G-5 and G-1, respectively. G-5 cells exhibited high growth rate compared to G-1 cells. The PI3-kinase inhibitor LY294002 and P70 S6 kinase inhibitor rapamycin partially blocked both G-1 and G-5 cell growth, suggesting that these two kinases are involved in hepatoma cell growth. In contrast, the MEK1 inhibitor PD98059 partially blocked G-5 cell growth but not G-1 cell growth. MAP kinases (MAPK) in both G-1 and G-5 cells were indistinguishably phosphorylated, yet MEK-dependent MAPK activation was observed only in G-5 cells. In G-1 cells, MAPK was phosphorylated in a manner not connected to MEK activation. Thus, the low degree of cell growth in G-1 cells was attributable to disruption of the MEK-dependent MAPK cascade. However, the molecular mechanism whereby MAPK phosphorylation does not parallel MAPK activation in G-1 cells remains unknown. Here, we suggest that there may be an as yet unidentified MAPK phosphorylation pathway in malignantly transformed cells, which may affect in vivo cell growth and metastatic capacities of cancers.

Correlation between metastatic potency and the down-regulation of E-cadherin in the mouse hepatoma cell lines G-1 and G-5.

Cell-cell adhesiveness, involving the adherens junction system including homophilic adhesion of cadherin and intracellular catenins, is a critical factor for tumor cell invasion and metastasis. We evaluated the levels of E-cadherin and beta-catenin in hepatoma cell sublines with high and low metastatic capacities. Stimulation of these cells with serum growth factors for more than 3 h after 24 h of starvation caused decreases in levels of E-cadherin and beta-catenin in the subline with high metastatic capacity, G-5. In contrast, no significant changes were observed in the subline with low metastatic capacity, G-1. Concomitantly with the decreases in E-cadherin and beta-catenin levels, G-5 cells were dissociated and detached from the culture dish, although G-1 cells again showed no morphological alterations. These in vitro results reflected the in vivo metastatic potencies of these hepatoma sublines, and further suggested the importance of the adherens junction system in determining metastatic potency of these parenchymal tumor cell lines as in epithelial/endothelial tumors.

Genistein attenuates peritoneal metastasis of azoxymethane-induced intestinal adenocarcinomas in Wistar rats.

The effects of the soybean isoflavonoid genistein on the development of bombesin-enhanced peritoneal metastasis from intestinal adenocarcinomas induced by azoxymethane (AOM) were investigated in male inbred Wistar rats. From the beginning of the experiment, rats were given 10 weekly s.c. injections of AOM (7.4 mg/kg body weight) and s.c. injections of bombesin (40 microg/kg body weight) every other day, and from week 16, s.c. injections of genistein (5 or 10 mg/kg body weight) every other day until the end of the experiment in week 45. Bombesin significantly increased the incidence of intestinal tumors and of cancer metastasis to the peritoneum. Although genistein administered at either dose had little or no effect on the enhancement of intestinal carcinogenesis by bombesin or on the location, histologic type, depth of involvement, labeling index, or growth pattern of intestinal cancers, it significantly decreased the incidence of cancer metastasis. Genistein also significantly decreased the incidence of lymphatic vessel invasion of adenocarcinomas, which was enhanced by bombesin. Our findings indicate that genistein attenuates cancer metastasis by inhibiting cancer cell invasion into lymphatic vessels through activities that do not affect the growth of intestinal cancers.

Increased expression of S100A6 (Calcyclin), a calcium-binding protein of the S100 family, in human colorectal adenocarcinomas.

The expression of S100A6 (also known as Calcyclin/2A9/ 5B10/PRA) in surgically resected human colorectal adenocarcinomas was examined to investigate whether S100A6 plays a role in the malignancy of human tumor cells. Western blot analysis using the lysates from colorectal adenocarcinomas and adjacent normal mucosa from 10 patients revealed that the average S100A6 level of adenocarcinomas was significantly higher (about 2.4-fold) than that of normal mucosa. Immunohistochemical analysis using formalin-fixed paraffin-embedded surgical specimens and monoclonal anti-S100A6 antibody (mAbA6) demonstrated that 2(5%) of 42 normal mucosa and 6 (46%) of 13 adenoma specimens were mAbA6-positive and showed granular staining localized at the supranuclear regions of epithelial cells, whereas 23 (55%) of 42 adenocarcinomas and 13 (100%) of 13 carcinoma cells that metastasized to the liver were mAbA6-positive and showed diffuse cytoplasmic staining. A significant correlation between S100A6 expression and Dukes' tumor stage or lymphatic permeation but not with other clinicopathological factors was shown. S100A6 was stained more intensely in peripheral portions than in central portions of adenocarcinomas, whereas Ki-67 (a growth marker) was stained equally in these two portions. These results suggest that S100A6 may be involved in the progression and invasive process of human colorectal adenocarcinomas.

Cooperation of fibronectin with lysophosphatidic acid induces motility and transcellular migration of rat ascites hepatoma cells.

We have previously shown that the transcellular migration of rat ascites hepatoma (AH130-MM1) cells through a cultured mesothelial cell monolayer (MCL) is triggered with lysophosphatidic acid (LPA) that stimulates actin polymerization and myosin light chain phosphorylation through the activation of Rho-ROCK (Rho-kinase) cascade. When, however, the motility of MM1 cells on a glass surface was tested by phagokinetic track motility assay, LPA failed to induce the motility. Nevertheless, when the glass had been coated with fibronectin (FN), LPA could induce phagokinetic motility which was accompanied by transformation of MM1 cells to fusiform-shape and assembly of focal adhesion. beta1 integrin, the counter receptor of FN, was expressed on MM1 cells. Anti-FN antibody, anti-beta1 integrin antibody and cyclo-GRGDSPA remarkably suppressed LPA-induced phagokinetic motility. These antibodies suppressed LPA-induced transcellular migration through MCL, as well. These results indicate that actin polymerization and phosphorylation of myosin light chain through Rho activation are insufficient for inducing motility but the cooperative FN/beta1 integrin-mediated adhesion is necessary for both the phagokinetic motility and transcellular migration of MM1 cells.

Reciprocal functions of liver tumor cells and endothelial cells. Involvement of endothelial cell migration and tumor cell proliferation at a primary site in distant metastasis.

BACKGROUND: The mechanism by which tumor cells interact with endothelial cells at the primary site in metastasis remains obscure. MATERIALS AND METHODS: To disclose the precise mechanisms and processes of metastasis and angiogenesis, we investigated the interactions of endothelial cells (CPAE; calf pulmonary arterial endothelial cells) with two mouse liver tumor cell lines, G-5, a highly metastatic clone, and G-1, a poorly metastatic clone, based on the difference in their angiogenic activities in vivo at a primary site. RESULTS: G-5 cell cultured conditioned medium (CM) promoted migration of endothelial cells more than did G-1 cell CM, in migration assay using modified Boyden chambers. Anti-basic FGF antibody inhibited G-5-induced endothelial cell migration. Furthermore, endothelial cells stimulated proliferation of G-5 cells more than that of G-1 cells in coculture assay using Transwell chambers (pore size; 0.4 microm). Pretreated endothelial cells with bFGF enhanced tumor cell proliferation, suggesting the ability of activated endothelial cells to support tumor growth. In addition, incubation with endothelial cell CM also improved the proliferative activity of G-5 cells more than that of G-1 cells in a concentration-dependent fashion, indicating production and secretion of liver tumor cell growth substance by endothelial cells. CONCLUSION: These results provide evidence that liver tumor cells stimulate endothelial cell migration and migrated endothelial cells facilitate liver tumor cell proliferation. Tumor growth at a primary site may initially be dependent on migrated endothelial cells rather than vascularization or blood nutrient supply. This bi-directional paracrine relationship between liver tumor cells and endothelial cells may influence their metastatic ability.

Suppression by apigenin of peritoneal metastasis of intestinal adenocarcinomas induced by azoxymethane in Wistar rats.

The effect of a naturally occurring flavonoid apigenin on the development of bombesin-enhanced peritoneal metastasis from intestinal adenocarcinomas induced by azoxymethane was investigated in male Wistar rats. From the start of the experiment, rats were given weekly s.c. injections of azoxymethane (7.4 mg/kg body weight) for 10 weeks and s.c. injection of bombesin (40 microg/kg body weight) every other day, and from week 16, s.c. injections of apigenin (0.75 or 1.5 mg/kg body weight) every other day until the end of the experiment in week 45. Bombesin significantly increased the incidence of intestinal tumors and cancer metastasis to the peritoneum in week 45. It also significantly increased the labeling index of intestinal cancers. Although administration of apigenin at either dose with bombesin had little or no effect on the enhancement of intestinal carcinogenesis by bombesin, the location, histologic type, depth of involvement, infiltrating growth patterns and labeling index, it was found to decrease significantly the incidence of cancer metastasis. Apigenin significantly decreased the incidence of lymphatic vessel invasion of adenocarcinomas, which was enhanced by bombesin. In vitro experiments revealed that apigenin inhibited bombesin-enhanced phosphorylation of mitogen-activated protein kinase (MAPK), but not matrix metalloprotease (MMP)-9 expression. Our findings indicate that apigenin inhibits cancer metastasis through inhibition of phosphorylation of MAPK.

Involvement of small GTPases Rho and Rac in the invasion of rat ascites hepatoma cells.

Lysophosphatidic acid (LPA) triggers the invasion of a mesothelial cell monolayer by rat ascites hepatoma (MM1) cells. LPA also induces rapid morphological changes of MM1 cells, cell surface blebbing and pseudopodia formation. Pseudopodia formation is tightly correlated with cellular invasiveness. Clostridium Botulinum C3 exoenzyme and genistein abrogated the formation of blebs and pseudopodia together with the inhibition of invasion, indicating that GTPase Rho and certain tyrosine kinases are involved in both processes. MM1 cells expressing constitutively active Rho exhibited the invasion and the formation of blebs and pseudopodia in the absence of LPA. In contrast, MM1 cells expressing constitutively active Rac were not invasive in the absence of LPA, but were invasive in the presence of LPA. Their morphological response to LPA was almost the same as that of parental MM1 cells. Expression of dominant negative Rac suppressed the invasiveness to approximately 3% of that of parental MM1 cells, together with the inhibition of pseudopodia formation. Thus, Rho and Rac are cooperatively involved in both the invasion and the related morphological changes of MM1 cells. Rho activation is sufficient both for the induction of invasion and the morphological changes leading to the invasion, whereas Rac activation is necessary but not sufficient by itself. We propose that Rho activation is not mediated by Rac but the cooperation of both GTPases is essential to trigger the invasive behavior of MM1 cells.

Inhibition of tumor invasion and metastasis by a novel lysophosphatidic acid (cyclic LPA).

Fetal calf serum (FCS) and 1-oleoyl lysophosphatidic acid (LPA) were previously found to be potent inducers of invasion (transcellular migration) in an in vitro system. A novel LPA, composed of cyclic phosphate and cyclopropane-containing hexadecanoic acid (PHYLPA), first isolated from myxoamoebae of Physarum polycephalum, and its synthetic derivatives (cLPA) were tested for their ability to inhibit tumor cell invasion and metastasis. Amoung these, Pal-cLPA, which has a palmitoyl moiety, was most potent in inhibiting invasion, with 93.8% inhibition at the concentration of 25 microM. Invasion in vitro by mouse melanoma cells (B16), human pancreatic adenocarcinoma cells (PSN-1), human lung cancer cells (OC-10) and human fibrosarcoma cells (HT-1080) was also inhibited by Pal-cLPA. The stimulation of MMI cells with LPA triggered F-actin formation, which was impaired by the addition of Pal-cLPA at invasion-inhibitory concentration. Pal-cLPA induced a rapid increase in adenosine 3',5'-cyclic monophosphate (cAMP) concentration in MMI cells. The addition of dibutyryl cAMP significantly abrogated LPA-induced invasion by MM1 cells and actin polymerization in the cells. The inhibition of MM1 cell invasion by Pal-cLPA may be ascribed to an increased level of cAMP. Pal-cLPA also suppressed invasion in vitro by MM1 cells induced by FCS dose dependently, without affecting proliferation. It also suppressed the pulmonary metastasis of B16 mouse melanoma cells injected into the tail vein of C57BL/6 mice. Thus, Pal-cLPA is effective in inhibiting invasion and metastasis of a variety of tumor cells.

An essential part for Rho-associated kinase in the transcellular invasion of tumor cells.

Adhesion of tumor cells to host cell layers and subsequent transcellular migration are pivotal steps in cancer invasion and metastasis. The small GTPase Rho controls cell adhesion and motility through reorganization of the actin cytoskeleton and regulation of actomyosin contractility. Cultured rat MM1 hepatoma cells migrate through a mesothelial cell monolayer in vitro in a serum-dependent, Rho-mediated manners. Among several proteins isolated as putative target molecules of Rho, the ROCK (ROK) family of Rho-associated serine-threonine protein kinases are thought to participate in the induction of focal adhesions and stress fibers in cultured cells, and to mediate calcium sensitization of smooth muscle contraction by enhancing phosphorylation of the regulatory light chain of myosin. Transfection of MM1 cells with cDNA encoding a dominant active mutant of ROCK conferred invasive activity independently of serum and Rho. In contrast, expression of a dominant negative, kinase-defective ROCK mutant substantially attenuated the invasive phenotype. A specific ROCK inhibitor (Y-27632) blocked both Rho-mediated activation of actomyosin and invasive activity of these cells. Furthermore, continuous delivery of this inhibitor using osmotic pumps considerably reduced the dissemination of MM1 cells implanted into the peritoneal cavity of syngeneic rats. These results indicate that ROCK plays an essential part in tumor cell invasion, and demonstrate its potential as a therapeutic target for the prevention of cancer invasion and metastasis.

Inhibition by interleukin 18 of osteolytic bone metastasis by human breast cancer cells.

Effects of interleukin (IL)-18 on experimental bone metastasis of human breast cancer cells, MDA-231 cells, in nude mice were investigated. In addition, effects of IL-18 on subcutaneous growth of MDA-231 cells were examined. Bone metastasis was produced by an intracardiac injection of MDA-231 cells. Twenty eight days after the cell injection, severe osteolytic bone metastasis was examined by X-ray radiography, and both non-osteolytic and osteolytic bone metastases were examined microscopically. IL-18 (1 microgram/mouse) was injected intraperitoneally according to protocols A and B. In protocol A, IL-18 was injected daily from day 7 after an intracardiac or subcutaneous injection of cells, and in protocol B, it was injected daily for 7 days each before and after the cell injection. In protocol A, IL-18 injections significantly suppressed both the incidence of osteolytic bone metastasis detected by X-ray radiography (about 80% vs. about 20% for the control group vs. the treatment group) and the number of its foci/mouse (1.6 vs. 1 for the control group vs. the treatment group). However, they did not cause significant effects on either the incidence of bone metastasis detected microscopically or the number of its foci/mouse. In protocol B, IL-18 injections caused no significant effects on either the incidence of osteolytic bone metastasis detected by X-ray radiography or the number of its foci/mouse. They caused no significant effects on the incidence of bone metastasis detected microscopically, but significantly decreased the number of its foci/mouse (about 2.0 vs. about 1 for the control group vs. the treatment group). In both protocols A and B, IL-18 injections produced no significant effect on the tumor take and subsequent growth of tumors after a subcutaneous injection of the cancer cells. Since in protocol A, IL-18 appears to have exerted its action after establishment of metastasis by cancer cells to the bone marrow, the effects of IL-18 found in Protocol A indicate that IL-18 inhibited osteolytic growth at bone metastatic sites of breast cancers. On the other hand, since in protocol B IL-18 is likely to have functioned around the time when lodging of cancer cells and early development of metastasis occur in the bone marrow, the effects of IL-18 found in Protocol B indicate that the cytokine also suppresses an early stage of bone metastasis of breast cancers, although, this effect was less apparent than the effect on osteolytic growth. Therefore, IL-18 may be useful for suppression of osteolytic bone metastasis which is a serious problem in patients of advanced breast cancers.