Direct in vitro lysis of metastatic tumor cells by cytokine-activated murine vascular endothelial cells

The purpose of these studies was to determine whether vascular endothelial cells can be treated with various cytokines to become cytotoxic against tumorigenic target cells under defined conditions in vitro. Microvascular endothelial cells were isolated from immunocompetent mice by perfusion of lungs grown in culture, cloned, and then characterized. The cloned microvascular endothelial cells were activated by incubation with a combination of recombinant tumor necrosis factor alpha and recombinant-gamma interferon (10 units/ml each) for 24 h. Activated endothelial but not control endothelial cells (incubated in medium alone, recombinant tumor necrosis factor alpha alone, or recombinant gamma interferon alone) produced significant lysis of mouse reticulum cell sarcoma and two different mouse melanomas. Moreover, at the concentrations used here, recombinant tumor necrosis factor alpha and recombinant gamma interferon did not produce direct target cell lysis. The activated endothelial cell-mediated tumor cell lysis depended on the continuous presence of the cytokines and was not due solely to initial target cell binding. The results demonstrate that, like macrophages, microvessel endothelial cells exposed to low levels of cytokines are capable of lysing tumor cells.     

Correlation between the in vitro interaction of tumor cells with an organ environment and metastatic behavior in vivo

The purpose of this study was to determine whether the in vitro interaction of tumor cells with an organ fragment correlated with organ colonization in vivo. Several clones of the K-1735 melanoma that exhibit different potentials for lung colonization or the heterogeneous M-5076 reticulum cell sarcoma which preferentially colonizes the liver, were used. The simultaneous subcutaneous implantation of a lung fragment with high lung-colonizing K-1735 cells stimulated the production of tumors, but this was not the case with the M-5076 cells, which do not colonize the lungs after intravenous injection. The ability of the tumor cells to survive in vitro in a lung fragment correlated with their capacity to produce lung tumor colonies after intravenous injection. Moreover, lung-conditioned medium stimulated the growth of lung-colonizing K-1735-M2 cells but not liver-colonizing M-5076 cells. Collectively, the data suggest that tissue-mediated growth stimulation may be one of the factors influencing the metastatic pattern and potential of tumor cells.     

Endothelial cell production of nitrogen oxides in response to interferon gamma in combination with tumor necrosis factor, interleukin-1, or endotoxin

Clinical studies using biological response modifiers in cancer therapy have shown that the major dose-limiting toxic effects are hypotension and diffuse microvascular leakage. The cause and pathophysiology of this hypotension remains unknown. Previous experiments have demonstrated that a number of cell types, including endothelial cells, neutrophils, and macrophages, can secrete a potent hypotensive agent--endothelium-derived relaxing factor, which has recently been identified as nitric oxide. In this study, we tested interferon gamma, tumor necrosis factor, interleukin-1, interleukin-2, muramyl dipeptide, and endotoxin for their effects on production of nitrogen oxides by endothelial cells. Interferon gamma, in combination with tumor necrosis factor, interleukin-1 (IL-1), or endotoxin, induced murine brain endothelial cells to secrete nitrites (20-45 microM within 48 hr), which are breakdown products of nitric oxide. Nitrite production was blocked by incubation of endothelial cells in medium without L-arginine, a substrate for nitric-oxide synthase. Accumulation of nitrites was also inhibited by addition of NG-monomethyl-L-arginine (L-NMMA), which acts as a competitive inhibitor of this enzyme. The inhibitory effects of L-NMMA were reversed by addition of excess L-arginine. These results suggest (a) that endothelial cells produce nitric oxide in response to immunomodulators and (b) that endothelial cell-derived nitric oxide plays a role in the development of hypotension in patients treated with tumor necrosis factor or interleukins. Furthermore, administration of substrate analogues such as L-NMMA may favorably alter the toxicity associated with these immunomodulators and result in a higher maximum tolerated dose, with subsequent improvement in the antitumor activity.     

Interleukin-10 gene transfer inhibits murine mammary tumors and elevates nitric oxide

Transfection of cDNA for IL-10 into line 66.1 murine mammary tumor cells results in marked suppression of tumor growth and metastasis. Others have reported that nitric oxide has potent antitumor activity and IL-10 is known to regulate the inducible isoform of nitric oxide synthase (iNOS) expressed in macrophages. We identified nitric oxide production in mammary tumors as indicated by electron paramagnetic resonance detection of nitric oxide-hemoglobin (NO-Hb). IL-10 expression resulted in elevated levels of NO-Hb in mammary tumors. Immunohistochemical examination of mammary tumors for iNOS protein revealed few positively staining cells in parental or control neo-transfected tumors but strong iNOS staining in all IL-10 transfected tumors, consistent with the NO-Hb data. To determine if mammary epithelial tumor cells themselves, express nitric oxide synthase activity, cultured tumor cells were treated with pro-inflammatory cytokines and nitrite accumulation was assessed in the conditioned medium. All IL-10 producing cell lines accumulated uM concentrations of nitrite in response to short term (24 hr) cytokine stimulation. Cells not expressing IL-10 (parental and neo-transfectants) accumulated no nitrite under similar culture conditions. After longer stimulation (48 hr), parental and 66-neo cells accumulated lower amounts of nitrite. IL-10 gene transfer is associated with increased iNOS protein expression and enzymatic activity detected both in vitro<0R> and in vivo<0R>. Our findings suggest that the antimetastatic and antitumor activity of IL-10 is related to enhanced production of nitric oxide. Int. J. Cancer 76:713–719, 1998.© 1998 Wiley-Liss, Inc.     

Overexpressed beta-catenin blocks nitric oxide-induced apoptosis in colonic cancer cells

Beta-catenin plays an important role in colonic tumorigenesis whereas inducible nitric oxide synthase and nitric oxide are elevated in colonic inflammation. Resistance of colonic epithelial cells to the induction of apoptosis may contribute to tumor development. Nitric oxide can stimulate apoptosis and, paradoxically, is implicated in the development of colon cancer. Our hypothesis was that beta-catenin could increase the resistance of colonic cancer cells to nitric oxide-induced apoptotic cell death. Here we show, using a beta-catenin overexpression system, that increased cytosolic beta-catenin renders colonic epithelial cells more resistant to nitric oxide-induced apoptotic cell death, independently of nitric oxide-induced accumulation of p53. Furthermore, we show that this occurs through inhibition of nitric oxide-induced release of cytochrome c from mitochondria and by blocking both the nitric oxide-induced suppression of the antiapoptotic protein, Bcl-xL, and the phosphorylation of Akt. We contend that increased nitric oxide production, such as that which occurs in chronic colonic inflammation, may select the cells with oncogenic mutant beta-catenin regulatory genes and contribute to human colonic carcinogenesis and tumor progression.     

Influence of nitric oxide synthase II gene disruption on tumor growth and metastasis

The relationship between nitric oxide (NO) synthase II (NOS II) expression and the metastatic ability of tumor cells is inconclusive. We determined the role of host NOS II expression in the growth and metastasis of the B16-BL6 murine melanoma and M5076 murine ovarian sarcoma cell lines. The cells were either s.c. or i.v. injected into syngeneic wild-type (NOS H+/+) and NOS II-null (NOS H-/-) C57BL/6 mice. Both cell lines produced slightly larger s.c. tumors in NOS H-/- mice than in NOS II+/+ mice. However, B16- BL6 cells produced more and larger experimental lung metastases in NOS II+/+ mice than in NOS II-/- mice, whereas M5076 cells produced fewer and smaller experimental lung metastases in NOS II+/+ mice than in NOS II-/- mice. After activation with IFN-gamma and lipopolysaccharide, macrophages isolated from NOS II+/+ C57BL/6 mice produced NO-dependent cytotoxicity in sarcoma cells, whereas macrophages from NOS II-/- C57BL/6 mice did not. In contrast, activated macrophages produced little to no NO-mediated cytotoxicity in melanoma cells. Immunostaining analyses indicated that NOS II expression was apparent in the metastases growing in NOS H+/+ mice and correlated with increased cell proliferation in B16-BL6 lung metastases but with decreased cell proliferation in M5076 liver metastases. Our data suggest that disruption of host NOS II expression enhanced the growth and metastasis of NO-sensitive tumor cells but suppressed the metastasis of NO-resistant tumor cells, proposing that host-derived NO may differentially modulate tumor progression.     

Nitrite and nitrosamine synthesis by hepatocytes isolated from normal woodchucks (Marmota monax) and woodchucks chronically infected with woodchuck hepatitis virus.

Hepatocytes isolated from woodchucks (Marmota monax) were shown to produce nitrite in vitro from L-arginine after stimulation with lipopolysaccharide (LPS). Hepatocytes isolated from woodchucks that were chronic carriers of woodchuck hepatitis virus formed twice as much nitrite as hepatocytes from noninfected animals. Nitrite synthesis by hepatocytes was directly related to L-arginine and LPS concentrations in the tissue culture medium and reached a plateau at 0.5 mM L-arginine and 1.0 micrograms/ml LPS. LPS-stimulated hepatocytes nitrosated morpholine to form N-nitrosomorpholine in the presence of L-arginine at a physiological pH of 7.4. There was a 10-fold increase in N-nitrosomorpholine production when hepatocytes were stimulated with LPS compared to unstimulated hepatocytes under similar conditions when both nitrite and morpholine were directly added to the medium. NG-monomethyl-L-arginine, a selective inhibitor of nitric oxide synthase, inhibited formation of both nitrite and N-nitrosomorpholine. These results demonstrate that nitrosating agents are formed in hepatocytes via the L-arginine-nitric oxide pathway. This suggests that endogenous formation of carcinogenic N-nitroso compounds could influence the process of hepatocarcinogenesis in woodchucks with chronic woodchuck hepatitis virus infection.     

Hypoxia induces escape from innate immunity in cancer cells via increased expression of ADAM10: role of nitric oxide

One key to malignant progression is the acquired ability of tumor cells to escape immune-mediated lysis. Whereas tumor hypoxia is known to play a causal role in cancer metastasis and resistance to therapy, the link between hypoxia and immune escape in cancer remains poorly understood. Here, we show that hypoxia induces tumor cell resistance to lysis mediated by immune effectors and that this resistance to lysis occurs via a hypoxia-inducible factor-1 (HIF-1)-dependent pathway linked to increased expression of the metalloproteinase ADAM10. This enzyme is required for the hypoxia-induced shedding of MHC class I chain-related molecule A (MICA), a ligand that triggers the cytolytic action of immune effectors, from the surface of tumor cells. Indeed, our findings show a mechanistic link between hypoxia-induced accumulation of the α-subunit of HIF-1 (HIF-1α), increased expression of ADAM10, and decreased surface MICA levels leading to tumor cell resistance to lysis mediated by innate immune effectors. Nitric oxide mimetic agents interfered with the hypoxia-induced accumulation of HIF-1α and with the hypoxia-induced upregulation of ADAM10 expression required for decreased surface MICA expression and resistance to lysis. Furthermore, treatment of tumor-bearing mice with nitroglycerin, a nitric oxide mimetic, attenuated tumor growth by a mechanism that relied upon innate immune effector cells. Together, these findings reveal a novel mechanism by which the hypoxic tumor microenvironment contributes to immune escape in cancer, lending support to potential immunotherapeutic strategies involving the use of nitric oxide mimetics.     

Tumor cell adhesion to endothelial cells: endothelial leukocyte adhesion molecule-1 as an inducible adhesive receptor specific for colon carcinoma cells.

Activation of endothelial cells by the two inflammatory mediators interleukin-1 (IL-1) and tumor necrosis factor strongly increases tumor cell adhesion. We describe antibody inhibition studies showing that the endothelial leukocyte adhesion molecule-1 (ELAM-1), a cell-surface glycoprotein selectively expressed by cytokine-activated endothelial cells and responsible for neutrophil adhesion, is the major, if not the only, mediator of colon carcinoma cell adhesion to activated endothelial cells. Among the different tumor cell lines tested, seven colon carcinoma cell lines were sensitive to ELAM-1 antibodies. Adhesion of melanoma, osteosarcoma, and lung, cervix, or kidney carcinoma cell lines to IL-1-treated endothelial cells was not affected by the ELAM-1 antibody. This result suggests that ELAM-1 is selectively recognized by colon carcinoma cells and that adhesion of tumor cells to activated endothelial cells could be mediated by different and specific mechanisms.     

Differential effects of retinoids on nitric oxide production by promonocytic U937 cells and ZR-75-1 human breast cancer cells

We demonstrated the presence of inducible and endothelial nitric oxide synthases in histiocytic lymphoma U937 cells by staining with anti-iNOS and anti-eNOS antibodies. We also investigated the effects of retinol and retinoic acid on nitric oxide production by both U937 cells and ZR-75-1 human breast cancer cells. U937 cells which had been treated with either retinol or retinoic acid (10-10-10-6 M) exhibited no significant difference in nitric oxide secretion into conditioned medium. Conversely, for ZR-75-1 cells, both retinol and retinoic acid (10-10-10-6 M) caused a significant (p<0.001) increase in the amount of nitrite secreted. Our results indicate that retinoid induced growth inhibition of breast cancer cells is associated with an increase in NO production, however, an increase in NO synthesis does not mediate retinoid induced differentiation of U937 cells.     

Loss of function of vascular smooth muscle cells by nitric oxide-dependent and -independent interactions with tumorigenic cells

Little is known about the interaction of tumor cells with host vascular smooth muscle cells. In reconstitution experiments, tumorigenic cell lines (including the rat hepatocarcinoma Morris 7777 and human melanoma M-21) were cultured for 17 hr in the presence of rat aortic rings, subsequently evaluated in contractility assays (response to phenylephrine and KCl). An agonist-independent loss of contractility was observed in rings pre-incubated with either tumorigenic cell lines or their conditioned medium (CM). The depressing effect of Morris cells depends largely on the expression of inducible nitric oxide synthase (iNOS) in smooth muscle cells and was reversed by an inhibitor of this enzyme; iNOS immunoreactivity was verified in some muscular vessels at the periphery of tumors formed by the Morris cell line in rats. The M-21 melanoma produces cytotoxicity in rat aortic rings (presence of single stranded DNA, cleavage of PARP, in differentiated smooth muscle only), accounting for the irreversible loss of contractility. The cytotoxicity produced by M-21 CM is not dependent on NO. Gel filtration of CM suggests that both the iNOS- and cytotoxicity-inducing substances from Morris hepatoma cells or M-21 cells, respectively, are mainly of low molecular weight (1 kDa or less). Other cell lines derived from rat or human tumors produce minimal effects on the rat aorta smooth muscle (H4-II-E-C3) or an irreversible anergy (RBL, MDA-MB-231, HEP-3B, HEP G2). The results emphasize that inhibition of vascular smooth muscle is relevant to tumor biology both by modulation of tumoral hemodynamics and by influencing the state of vessel maturation.     

The anti-tumor effect of recombinant interferon alpha or gamma is influenced by tumor location

The in vivo anti-tumor effects of a recombinant human hybrid interferon alpha, rHuIFN-alpha A/D, and recombinant murine interferon gamma (rMuIFN-gamma) were evaluated against experimental hepatic metastases and s.c. tumor growth of the murine reticulum cell sarcoma M5076. The 2 interferons were equally active in preventing experimental hepatic metastases. However, the interferons differed in their relative ability to influence the growth of the same tumor when treatment was initiated following injection of tumor cells. Greater efficacy was obtained in the treatment of metastatic foci in the liver with rHuIFN-alpha A/D, while rMuIFN-gamma was more active in the therapy of an s.c. growing M5076 tumor. These results demonstrate that the same tumor growing at different sites can have different relative sensitivities to IFN-alpha and IFN-gamma.     

Natural cell-mediated cytotoxicity in mice against non-lymphoid tumor cells and some normal cells

Lymphocytes from normal mice were found to have cell-mediated cytotoxicity, in a short-term ⁵¹Cr release assay, against a variety of non-lymphoid tumor cells as well as against lymphomas. Some of the non-lymphoid tumors were as susceptible to natural cytotoxicity as the standardly used lymphoid lines. Some tissue culture cell lines and in vivo passaged tumor lines were susceptible to lysis, as were some primary virus-induced lymphomas. Tumors which arose in nude mice, which have high levels of natural cytotoxic activity, were all resistant to lysis. In addition to the susceptibility of transformed cells to natural cell-mediated cytotoxicity, some untransformed cultured cells and cells from normal tissues were targets for this mechanism. Low levels of cytotoxicity were seen with normal thymus cells, bone-marrow cells, and short term cultures of macrophages, whereas normal spleen and lymph-node cells were completely resistant to lysis. These results indicate a broader spectrum for mouse natural cell-mediated cytotoxicity reactivity than has been previously recognized.     

Nitric oxide generated by iNOS reduces deformability of Lewis lung carcinoma cells

Previous studies have indicated that NO plays a crucial role in the metastasis of tumor cells and that tumor cells produce nitric oxide (NO) via inducible nitric oxide synthase (iNOS). Since the deformability of tumor cells is an important factor governing their metastatic potential, in this study we investigated the regulation of tumor cell deformability by NO. Lewis lung tumor cells (3LL cells) were also incubated with a cytokine mixture (IL-1β, IFNγ, and TNFα). The nitrite/nitrate content of the supernatant was then measured by the Griess method, and iNOS expression was evaluated by RT-PCR in vitro . Nitrite/nitrate was produced in response to administration of the cytokine mixture, and iNOS mRNA was expressed in the cytokine-treated cells. The deformability of the 3LL cells was evaluated by measuring the peak pressure generated during their passage through a microfilter at a constant flow rate. Both the cytokine mixture and NO donor (NOC 18) significantly increased the filtration pressure, and the staining of the cells with rhodamine-phalloidin revealed assembly of F-actin in the cell membrane. In conclusion, NO plays a role in the decreased deformability of tumor cells, suggesting that NO is one of the factors that regulates metastasis.     

Nitric oxide induces oral squamous cell carcinoma cells apoptosis with p53 accumulation

Nitric oxide has been reported to have cytotoxic effects in several tumor cells. The objective of this study was to investigate the effects of exogenous nitric oxide on apopotosis in oral squamous cell carcinoma cells and to reveal its possible mechanism. Tca8113 cells were cultured with various concentrations of nitric oxide that were released from sodium nitroprusside (SNP). Nitrite/nitrate levels in the culture supernatant were determined using a commercial available nitric oxide kit. Cellular proliferation was determined by MTT assay. Apoptosis was detected by flow cytometry. Expression of inducible nitric oxide synthase (iNOS) was determined by immunocytochemistry. p53 expression was assessed by Western blot. SNP can release nitric oxide into the culture medium in a dose-dependent manner. Nitric oxide remarkably inhibits proliferation in a dose and time-dependent manners and lead to apoptosis of the Tca8113 cell. The p53 expression was elevated accompanying by the increased apoptotic cells. No difference of iNOS was found whether or not the cells were treated with SNP. Exogenous nitric oxide had an inhibitory effect on Tca8113 cells proliferation in a dose and time-dependent manners and possibly via p53 dependent apoptosis pathway. Exogenous nitric oxide had no significant effect on cellular iNOS protein.     

Use of retroviral vectors encoding murine inducible nitric oxide synthase gene to suppress tumorigenicity and cancer metastasis of murine melanoma

The purpose of this study was to determine whether retrovirus-mediated transfer of murine macrophage inducible nitric oxide synthase (iNOS) can produce inhibition of tumorigenicity and metastasis. Retroviral vectors encoding macrophage iNOS constructed in pLXSN, a retroviral vector with the iNOS gene under the control of a long terminal repeat promoter, were stably transfected into PA317 cells. Medium harvested from confluent monolayers of the virus-producing cell lines was used for infection of the murine K-1735 melanoma cells. Expression of iNOS was confirmed by northern and Western blot analyses. Functional iNOS protein expression was confirmed by bioassay of nitrite accumulation in the culture supernatant. Cells infected by a control iNOS-negative retrovirus produced fast-growing subcutaneous tumors and many lung metastases in nude mice, whereas iNOS-transduced cells produced slow-growing tumors and few lung metastases, showing that the infection of murine tumor cells by retroviruses harboring the iNOS gene can suppress tumorigenicity and metastasis.     

Oxidative stress contributes to the anti-proliferative effects of flavone acetic acid on endothelial cells

The synthetic flavonoid flavone acetic acid (FAA) has anti-tumor activity against a variety of transplanted tumors in mice through mechanisms which likely involve effects on tumor vasculature and the host immune system. The aims of the present in vitro study were to compare the sensitivity of tumor and endothelial cells to FAA treatment and to assess if nitric oxide and superoxide are involved in the FAA-mediated suppression of cell proliferation. FAA at 1 mM concentration was approximately two times more effective in suppressing proliferation of endothelial than tumor cells. The anti-proliferative effect of 1 mM FAA on endothelial cells was partially blocked by inhibitors to various superoxide-producing enzymes (xanthine oxidase, cyclooxygenase, poly-ADP-ribose polymerase, ribonucleotide reductase) and completely inhibited by the direct scavengers of superoxide lucigenin and Tiron. In contrast, inhibitors of nitric oxide were unable to prevent the effects of FAA on proliferation. FAA induced apoptosis of endothelial cells, which was not affected by inhibitors of nitric oxide or superoxide. Our data imply that FAA inhibits proliferation of endothelial cells by a superoxide-dependent mechanism and induces apoptosis by a nitric oxide and superoxide-independent mechanism.     

Inhibition of tumor cell collagenolytic activity by bovine endothelial cells

The ability of cellular vascular components including endothelial cells, smooth muscle cells, and fibroblasts to interact with the collagenolytic activity of invasive human tumor cell lines has been investigated. The human HT1080 fibrosarcoma and Bowe melanoma cells, which rapidly digest collagenous proteins in vitro, failed to dissolve them when cocultivated with bovine endothelial cells. This inhibition was not dependent on the ability of endothelial cells to form a monolayer separating the tumor cells from the collagenous substrate. In contrast, little collagenolysis inhibitory activity was detected in bovine vascular smooth muscle cells and human fibroblasts when compared to endothelial cells. Serum-free medium conditioned by endothelial cells inhibited tumor cell-mediated collagenolysis. Our data further suggest that this inhibition was mediated by secreted collagenase inhibitors, since endothelial cell-conditioned medium did not suppress the production of metalloproteinases by the tumor cells but inhibited the activities of collagenases derived from tadpole, rabbit, and human fibroblasts. Treatment of the endothelial cells with cycloheximide suppressed the collagenase inhibitory activity, demonstrating active production of collagenase inhibitors by the cells. Gel filtration chromatography of endothelial cell-conditioned medium allowed the separation of two distinct peaks with inhibitory activities for vertebrate collagenase in the molecular weight range of 70,000 to 75,000 and 30,000 to 35,000, respectively. While the inhibitor with an approximate molecular weight of 30,000 to 35,000 shared many properties with the tissue inhibitor of metalloproteinases, the high-molecular-weight inhibitor demonstrated characteristics not yet described for any collagenase inhibitor. The production and secretion of inhibitors of vertebrate collagenase by bovine endothelial cells may be of importance in the local control of collagen turnover under physiological as well as pathological conditions.     

In vivo anti-tumor activity of combinations of interferon alpha and interleukin-2 in a murine model. Correlation of efficacy with the induction of cytotoxic cells resembling natural killer cells

The in vivo anti-tumor activity of 2 recombinant cytokines, interleukin-2 (rIL-2) and human hybrid interferon alpha (rHuIFN-alpha A/D), were tested using the murine reticulum cell sarcoma M5076. Experimental hepatic metastases, following i.v. injection of tumor cells, and tumor growth and spontaneous metastases, following s.c. injection of tumor cells, were inhibited to a greater extent in mice treated with a combination of these cytokines than in animals treated with either one alone. When used in conjunction with surgical removal of the s.c. tumor, treatment of mice with both cytokines significantly prolonged survival of tumor-bearing animals. Injection of normal mice with a combination of cytokines, but not with either cytokine alone, resulted in a marked increase in cytotoxic activity of hepatic effector cells. The effector cells in these mice appeared to be NK cells since this enhanced cytotoxicity was markedly reduced in animals treated in vivo with anti-asialo GM1 or in NK-deficient beige mice. Furthermore, no in vivo efficacy was observed in M5076-bearing beige mice treated with these cytokines. Thus, injection of mice with rIL-2 and rHuIFN-alpha A/D results in the induction of an NK-cell-like population in the liver with enhanced cytotoxic activity that correlates with the observed anti-tumor activity in vivo in this murine model. These results suggest that combinations of cytokines, in particular IFN-alpha and IL-2, can be effectively used in combination for the treatment of tumors and/or metastases.