Fas and Fas ligand interactions suppress melanoma lung metastasis.

Apoptosis induced by Fas (CD95) ligation is frequently lost during tumor progression; however, there is no direct evidence to support an association of Fas loss-of-function with metastatic tumor behavior. To determine whether Fas loss-of-function is critical for acquisition of the metastatic phenotype, we have compared the ability of Fas-sensitive K1735 murine melanomas to form spontaneous lung metastases in wild-type and Fas ligand-deficient mice. Fas-sensitive melanoma clones are highly tumorigenic but rarely metastatic in wild-type syngeneic mice. However, in Fas ligand-deficient mice, both the incidence and number of metastases are increased. These findings provide the first evidence that Fas-Fas ligand interactions can suppress metastasis and that tumor Fas loss-of-function may be causally linked to metastatic progression.

Clonogenicity and experimental metastatic potential of spontaneous mouse mammary neoplasms.

Spontaneous primary mammary tumors of C3H-Avy mice differ in metastatic colonization potential, some producing many lung deposits (high-colonization potential) and others producing few or none (low-colonization potential) after iv inoculation of cells. The degree of metastasis from undisturbed neoplasms also varies from tumor to tumor. This study examined whether these differences between tumors could be accounted for by differences in clonogenic or stem cell content. Tests for clonogenic cells were: growth in 0.3% agarose and limiting dilution assays. Mammary tumor cells of known colonization potential were inoculated iv at serially reduced doses, and the relationship between number of cells injected and number of lung deposits formed was determined. Parallel in vitro dose-response assays in 0.3% agarose were performed with the use of cells from the same primary tumors. Colony-forming efficiency in 0.3% agarose cultures varied between individual primary mammary tumors and was positively associated with experimental metastatic potential, suggesting that the stem or clonogenic cell content of primary tumors is one of the important determinants of the metastatic phenotype.

Different deficiencies in the prevention of tumorigenic-low-metastatic murine K-1735b melanoma cells from producing metastases.

To produce metastasis, malignant tumor cells must be able to complete a sequence of many steps that depend not only on tumor cell properties but also on ability of the tumor cells to interact effectively with host homeostatic mechanisms to avoid destruction. Therefore, it should be possible to isolate clonal populations non- or low metastatic. In a study of K-1735 clones introduced into normal syngeneic hosts, the reasons for the lack of or low ability of metastasis production did indeed differ among different clones. Some clones were identified that were low metastatic in syngeneic C3H/HeN mice because of antigenic characteristics. Others failed to give rise to metastases because they could not survive and grow once arrested in the lung parenchyma. These data suggested that the success of future studies dealing with genetic analysis of the metastatic phenotype could depend on the use of appropriate tumor cell populations.

Spontaneous and induced metastasis of naturally occurring tumors in mice: analysis of cell shedding into the blood.

The lung colonization capability (after iv inoculation) of each of 77 autochthonous virus-induced mammary tumors in C3H/Avy female mice was compared with its own spontaneous metastatic behavior. Twenty-four of these neoplasms had spontaneously metastasized to the lungs of the tumor bearer. The results were interpreted in the framework of dose-response studies on experimentally induced metastasis. These results indicate that tumors of high lung colonization potential (HCP) overgrow the lungs with iv doses as low as 10(4) cells, whereas tumors of low lung colonization potential (LCP) require at least 250 times this dose to achieve the same result. No general similarity was found between spontaneous and induced metastatic capability. Although some tumors showed good correspondence, others showed discrepancies and provided new information on mechanisms of cell shedding from primary tumors and on rate-limiting steps in metastasis. For example, from the dose-response studies, nonmetastatic tumors of HCP (iv) are deduced to be shedding very few cells. These findings suggest that escape from the primary tumor is an active process or else passive leakage of cells into vessels by hemorrhage, necrosis, or palpation should have resulted in spontaneous metastasis. Spontaneously metastatic tumors of LCP corroborate the view that metastasis is effected by a highly active subpopulation with special properties; otherwise, the required scale of passive cell release into the blood would be unrealistically large. Blood bioassay and time-course studies on pulmonary deposit formation indicate that shedding of metastatic cells occurs early in mammary tumor development.

Expression of angiogenesis-related genes and progression of human ovarian carcinomas in nude mice.

BACKGROUND: By the time patients are diagnosed with ovarian carcinoma, peritoneal dissemination of the tumor often has occurred. The progressive growth and spread of ovarian carcinoma depend, in part, on the formation of an adequate blood supply. We determined whether the expression of genes that regulate distinct steps in angiogenesis (i.e., the formation of new blood vessels) was associated with the pattern and progressive growth of human ovarian carcinomas implanted in the peritoneal cavity of nude mice. METHODS: Five different human ovarian carcinomas were injected individually into the peritoneal cavity of female NCr-nu/nu nude mice. The expression of basic fibroblast growth factor, vascular endothelial growth factor/vascular permeability factor (VEGF/VPF), interleukin 8 (IL-8), and collagenase type IV (MMP-2 [matrix metalloproteinase-2] and MMP-9) was determined by northern blot analysis, in situ hybridization of messenger RNA, and immunohistochemical analysis. Blood vessel distribution and density, macrophage infiltration pattern, and stromal reaction were determined by immunohistochemical analysis with specific antibodies. RESULTS: Three of the carcinomas produced both solid lesions and ascitic tumors, whereas the remaining two produced only solid lesions. Two of the carcinomas produced rapidly progressive disease, two produced slow disease, and one produced intermediate disease. The formation of ascites was directly associated with expression of VEGF/ VPF, and survival was inversely associated with expression of IL-8. In rapidly growing tumors, the number of blood vessels was high throughout the lesion; in contrast, in slow-growing tumors, most vessels (and infiltrating macrophages) were located at the periphery. CONCLUSIONS: The expression of various genes that regulate angiogenesis in human ovarian carcinomas is associated with the pattern of the disease and its progression. Therefore, targeting specific genes that regulate angiogenesis could offer new approaches to the treatment of ovarian cancer.

Augmentation of antiproliferative activity of interferon alfa against human bladder tumor cell lines by encapsulation of interferon alfa within liposomes.

Present therapy for human bladder cancer includes the intravesical administration of antiproliferative agents, such as recombinant human interferon alfa (IFN-alpha). The administration of cytotoxic molecules encapsulated in liposomes could provide a more efficient method for such therapy. Therefore, we determined whether encapsulation of the recombinant human IFN-alpha hybrid BBDD within liposomes will produce antitumor effects against the human bladder cancer cell line 253J superior to those observed with free IFN-alpha. Adherent cells were cultured in medium alone, in medium containing different concentrations of IFN-alpha, or in medium containing multilamellar liposomes (phosphatidylcholine-phosphatidylserine at a molar ratio of 7:3) that encapsulated saline or IFN-alpha. Cell growth was determined 96-120 hours later. Additional control groups consisted of target cells cultured with free IFN-alpha or with IFN-alpha plus liposomes containing saline. Cytostasis mediated by free IFN-alpha alone or IFN-alpha in the presence of liposome-saline was identical and ranged from 0%-30% (10 IU/mL) to 45%-70% (1,000 IU/mL). Liposomes containing saline produced no effects. Liposome-encapsulated IFN-alpha produced significantly greater growth inhibition than free IFN-alpha: 40%-70% (10 IU/mL) and 80%-90% (1,000 IU/mL), respectively. Moreover, a 253J variant subline selected for resistance to free IFN-alpha was sensitive to IFN-alpha presented in liposomes. These data suggest that the encapsulation of antiproliferative agents such as IFN-alpha in liposomes can improve therapeutic results.

Correlation of growth capacity of human tumor cells in hard agarose with their in vivo proliferative capacity at specific metastatic sites.

The purpose of this study was to determine whether the degree of anchorage-independent growth of human tumor cells in increasing concentrations of agarose correlated with the capacity of the cells to produce experimental metastases in nude mice. Human melanoma, breast carcinoma, and colon carcinoma cells from parental lines and variants selected in vivo for metastasis and in vitro cloned lines were plated into medium containing 0.3%, 0.6%, 0.9%, or 1.2% of agarose. These cells were also injected into nude mice: intravenously for melanoma, into the mammary fat pad for breast carcinoma, and into the spleen for colon carcinoma. Production of tumor cell colonies in dense agarose (greater than 0.6%) correlated with production of experimental metastases in the lung (melanoma, breast carcinoma) or liver (colon carcinoma). We conclude that the degree of anchorage-independent growth of tumor cells can predict their biological behavior and metastatic potential in vivo. Thus, this technique may be useful for the isolation of metastatic cells from heterogeneous human neoplasms.

Influence of microenvironment and vascular anatomy on "metastatic" colonization potential of mammary tumors.

This communication reports experiments demonstrating that some sites in which tumor cells lodge reproducibly fail to support secondary colony formation by a particular tumor, even though cells from the same tumor are already proven to have high colonization potential in other organs. The effect is not an expression of nonspecific hostility to tumor growth, since cells from certain other tumors readily colonize the same site. Cell suspensions obtained by disaggregation of a series of naturally occurring murine mammary tumors were each inoculated by four different routes into separate batches of syngeneic animals, and the resulting degree and distribution of colonization were studied 90 days later at autopsy. Standard doses of 1 million viable tumor cells were injected either i.p. or s.c. into the tail vein or the hepatic portal vein. It was found that some tumors could reproducibly colonize by all routes, whereas others could colonize only by a few, and the combination of sites colonized varied from tumor to tumor; still others were unable to grow in any site. Cells from nonneoplastic lactating mammary glands did not establish any colonies. We have demonstrated previously that individual naturally occurring mammary tumors differ in their pulmonary colonization potentials after i.v. inoculation and that the potential of a given tumor is an intrinsic property of its constituent cells. The current findings are evidence that the microenvironment of an organ can inhibit or permit expression of this intrinsic potential and that the degree and sites of colonization are thus the results of interaction between tumor and organ-specific factors. It was also found that circulatory anatomy partially influenced the distribution of colonies and that colonization of distant organs after blood-borne dissemination is distinct from general tumor transplantability.

Evidence that the process of murine melanoma metastasis is sequential and selective and contains stochastic elements.

Malignant neoplasms are heterogeneous for many different biological characteristics, including invasion and metastasis. The pathogenesis of metastasis involves a series of sequential steps which must be completed by metastatic cells. In the present study we examined the metastatic behavior of three highly metastatic and three nonmetastatic subpopulations isolated from the K-1735 melanoma syngeneic to the C3H/HeN mouse. Cells were labeled with [125I]iodo-2'-deoxyuridine, and their initial organ distribution, fate, and production of experimental metastases were determined. Highly metastatic cells survived in lung parenchyma to produce metastases, whereas nonmetastatic cells did not. However, even with the highly metastatic cells only 2% of the original inoculum was responsible for the final production of metastases. The results support the concept that the fate of tumor cells released into the bloodstream is determined by sequential and selective events and introduces a third regulatory factor. Cells endowed with metastatic properties have a higher probability of forming metastases than cells not so endowed, but this probability is not 100%. Hence, metastasis should be considered as a sequential, selective, and stochastic process.

Influence of organ microenvironment on pigmentation of a metastatic murine melanoma.

The purpose of these studies was to investigate the relationship of the host microenvironment to the metastatic and pigmented phenotypes of the SW-1 variant of the murine K-1735 melanoma. The SW-1 subline was isolated from an amelanotic lung metastasis in a C3H/HeN mouse given an s.c. injection of the K-1735 melanoma. Cells of this line were highly metastatic and produced tumor deposits in many organs. In all sites except the brain, these lesions were predominantly amelanotic. K-1735 SW-1 cells were isolated from metastases in various organs and subsequently reinoculated into normal syngeneic recipients. Whereas the metastatic phenotype remained stable and thus was heritable, pigmentation was unstable and appeared to be modulated by the site of tumor growth. Further differences in the phenotype of K-1735 SW-1 cells growing in vivo and in culture were revealed by assays for tyrosinase activity. K-1735 SW-1 cells growing in culture did not produce melanin nor did they respond to agents that can stimulate melanin production in another mouse melanoma, the B16 line. K-1735 SW-1 cells do not, however, lack tyrosinase, since these cells are capable of producing melanin when growing in certain organs in vivo. We conclude that the host organ environment may influence a phenotype of malignant melanoma cells, i.e., pigmentation. These findings also suggest caution when extrapolating the results of in vitro biochemical assays to properties of tumor cells growing in vivo.

Detection and identification of activated oncogenes in human skin cancers occurring on sun-exposed body sites.

High-molecular-weight DNA isolated from eight fresh human skin cancers occurring on sun-exposed body sites were assayed for their ability to transform NIH 3T3 cells. A cotransfection protocol using pSV2-neo DNA, which confers resistance to the antibiotic G418, was used to select cells that had taken up the transfected DNA. About 2 weeks after transfection, G418-resistant colonies were pooled and injected s.c. into athymic nude mice. The NIH 3T3 cells transfected with DNA from six of the human skin cancers induced tumors in nude mice. DNAs from all six tumor cell lines contained human alu sequences. Southern blot hybridization with ras-specific probes revealed that DNAs from the four alu-rich tumors contained the human Ha-ras oncogene, in addition to that of the NIH 3T3 controls. In contrast, DNAs from the other two tumors did not contain any of the known oncogenes tested, except those endogenous to NIH 3T3 cells. DNAs from three of four first cycle tumorigenic transformants gave rise to morphologically transformed foci when assayed in a second cycle of transfection. DNAs from all three secondary transformants contained discrete human alu sequences, and in addition, contained Ha-ras sequences similar to those present in their respective primary transformants. Interestingly, DNA from both primary and secondary transformants of one particular human squamous cell carcinoma contained highly amplified copies of the Ha-ras oncogene. These results suggest that activation of the Ha-ras oncogene may be common in human skin cancers originating on sun-exposed body sites. Further characterization of the Ha-ras oncogenes present in these human skin cancers may provide information on the molecular mechanisms by which UV radiation of the sun induces human neoplasms on exposed body sites.

Immune response to progressor variants derived from transfection of an ultraviolet radiation-induced C3H mouse regressor tumor cell line with activated Harvey-ras oncogene.

Skin cancers induced in mice by UV radiation often exhibit a regressor phenotype. In order to determine how tumors escape the immune defenses of the normal immunocompetent host, we sought to isolate progressor variants from a UV radiation-induced C3H mouse regressor fibrosarcoma cell line, UV-2240, by transfection with an activated Ha-ras oncogene. A cotransfection protocol using pSV2-neo DNA, which confers resistance to the antibiotic G418, was used to select transfected cells. Injection of Ha-ras-transfected UV-2240 cells s.c. into immunocompetent C3H mice produced tumors in four of 36 animals. In contrast, UV-2240 cells transfected with pSV2-neo DNA alone or mock transfected with CaPO4 did not produce tumors in normal C3H mice. DNAs from cell lines established from Ha-ras-induced tumors contained unique Ha-ras sequences in addition to those sequences endogenous to UV-2240 cells. However, the Ha-ras-induced progressor variants did not overexpress the Mr 21,000 protein. The Ha-ra-induced progressor variants produced experimental lung metastasis in both normal C3H and nude mice, although they induced more lung nodules in nude mice than in normal C3H mice. In addition, all four Ha-ras-induced progressor variants produced significantly more experimental lung metastases in nude mice than did the parent UV-2240 cell line. However, both the parental UV-2240 cell line and the Ha-ras-induced progressor variants expressed similar levels of H-2Kk and H-2Dk antigens and were immunologically cross-reactive, as determined by in vitro cytotoxic T-lymphocyte and in vivo immunization-challenge assays. These results indicate that the progressor phenotype of the Ha-ras-induced tumor variants is not due to loss of tumor-specific transplantation or Class I major histocompatibility complex antigens. This implies that some tumor cells can escape the immune defenses of the normal immunocompetent host by mechanisms other than loss of tumor-specific transplantation and Class I major histocompatibility antigens.

Tumorigenicity and metastasis of human breast carcinoma cell lines in nude mice.

There are few reports describing experimental models of the growth and metastasis of human breast carcinomas. This article discusses the tumorigenic and metastatic properties of two estrogen receptor-negative breast carcinomas injected into nude mice. Tumor growth in the mammary fatpad (m.f.p.) and the subcutis was compared in female nude mice. The injection of 10(5) viable cells of two human breast carcinoma cell lines (MDA-MB-231 and MDA-MB-435) gave a 100% tumor take rate in the m.f.p., whereas only 40% of the s.c. injections produced tumors and these occurred several weeks after the appearance of the m.f.p. tumors. Thus, the m.f.p. of nude mice is a favorable site for the growth of human breast carcinomas. MDA-MB-435 tumors produced distant metastases in 80% to 100% of recipients. The most common sites for metastasis were the lymph nodes and lungs, with a lower incidence of metastases in muscle (chest wall and thigh), heart, and brain. New variant cell lines were isolated from metastases in the lungs, brain, and heart. All the cell lines were tumorigenic in the m.f.p., and the lung- and heart-derived metastasis lines produced slightly more lung metastases than the original cell line. However, the brain metastasis variant produced significantly fewer lung metastases. Intravenous inoculation of the spontaneous metastasis-derived cell lines produced few lung colonies. Only cell variants isolated from experimental lung metastases showed enhanced lung colonization potential when reinjected i.v. Our results suggest that the estrogen receptor-negative MDA-MB-435 cell line injected in the m.f.p. of nude mice could be a valuable tool for analysis of the cellular and molecular basis of the metastasis of advanced breast cancer.

Tumor cell dissemination patterns and metastasis of murine mammary carcinoma.

Quantitative studies on the distribution kinetics of isotope-labeled cells from spontaneous murine mammary tumors injected intravenously or arterially showed that cells were rapidly distributed to all organs examined and indicated that the distribution patterns of metastases from such tumors are not primarily determined by the dose of cells delivered to each organ. The preferential colonization of certain organs is therefore considered to depend as much on differential survival and growth of the disseminated tumor cells in unfamiliar metabolic microenvironments, as on vascular sieving effects in organ capillary networks. Further experiments involved transplantation of pieces of nonpulmonary tissue containing trapped mammary tumor cells into syngeneic mice, followed by observation of the animals for several months. From these studies it is concluded that the absence of tumor colonies in extrapulmonary sites after i.v. inoculation is due to their inability to thrive in the organs concerned and not to early death of the original host from heavy pulmonary tumor growth. These results provide further evidence strengthening the conclusion emerging from several independent lines of investigation (Potter et al., Invasion Metastasis, 3: 221-233, 1983; Tarin et al., Cancer Res., 41: 3604-3609, 1981; Tarin et al., Cancer Res., 44: 3584-3592, 1984; Horak et al., J. Natl. Cancer Inst., 76: 913-922, 1986; Nicolson et al., Int. J. Cancer, 38: 289-294, 1986; Naito et al., Invasion Metastasis, 7: 16-29, 1987) that the growth of disseminated tumor cells is inhibited or even abrogated by many of the organs in which the cells sequester after vascular dissemination.

Enhanced c-erbB-2/neu expression in human ovarian cancer cells correlates with more severe malignancy that can be suppressed by E1A.

Amplification or overexpression of c-erbB-2/neu protooncogene, or both, occur frequently in many different types of human cancers and have been shown to correlate with decreased survival in ovarian cancer patients. We have previously found that the ovarian carcinoma cell line SK-OV-3 overexpresses c-erbB-2/neu mRNA. To further study the biological effect of c-erbB-2/neu overexpression in SK-OV-3 cells, we injected such cells i.p. into female nu/nu mice and found that this cell line forms extensive abdominal tumors and ascites. From the ascites in an injected mouse, we established the SKOV3.ip1 cell line and found that it expressed 2-fold more c-erbB-2/neu-encoded p185 proteins than the parental SK-OV-3 cells. When transformation phenotypes of SK-OV-3 and SKOV3.ip1 cells were compared, SKOV3.ip1 cells showed higher cell growth and DNA synthesis rates, formed more colonies in soft agar, produced larger s.c. tumors, and resulted in shorter survival of nu/nu mice after i.p. injection. These data indicate that the level of c-erbB-2/neu overexpression may correlate with the degree of malignancy in these ovarian carcinoma cells. Since we had previously shown that the adenovirus 5 E1A gene product can suppress transformation and metastatic properties induced by mutation-activated rat neu oncogene in mouse embryo fibroblast cells, we further examined whether E1A can abrogate malignancy in c-erbB-2/neu-overexpressing human ovarian cancer cells. We introduced the E1A gene into c-erbB-2/neu-overexpressing SKOV3.ip1 cells and found that the E1A-expressing ovarian cancer cell lines had decreased c-erbB-2/neu-encoded p185 expression and reduced malignancy, including a decreased ability to induce tumors in nu/nu mice. Therefore, we concluded that E1A is a tumor suppressor gene for c-erbB-2/neu-overexpressing human ovarian cancer cells and may be useful in developing therapeutic reagents for these human cancers.

Phenotypic diversity of murine B16 melanoma detected by anti-B16 monoclonal antibodies.

A panel of monoclonal antibodies (MoAbs), produced against the murine B16 melanoma, has been used to characterize its phenotypic diversity. Six MoAbs that did not bind to primary cultures of kidney, brain or liver, spleen cells, thymocytes, 3T3 fibroblasts, melanin, or transferrin receptors were selected for further evaluation. Five MoAbs, which recognized surface antigens expressed on parental B16 cells and the B16-F1, B16-F10, B16-F10 FLR, and B16-BL6 sublines, did not appear to cross-react with each other, suggesting that they identified antigenically distinct epitopes. Four MoAbs, designated as IB16-2, IB16-4, IB16-8, and IB16-10, recognized B16 surface antigens that were variably expressed over short periods of time. This variable expression was independent of the cell cycle and was characteristic of four B16 sublines. Two of these MoAbs, both of the IgG2b isotype, fixed rabbit and guinea pig complement and were cytolytic in the presence of rabbit complement. One MoAb, designated IB16-6, recognized a surface antigen consistently expressed on greater than 90% of cells of both the parental tumor and the sublines. This MoAb bound to several murine and one human melanoma cell line, but not to other histopathological types of tumors or normal tissues. The cellular antigen that this antibody recognized was not detected in the cytoplasm, did not modulate in the presence of IB16-6, and was sensitive to trypsin, pronase, alcohols, acetone, and detergents, thereby suggesting that it was a protein. Our data are among the first that directly show the extent of phenotypic diversity of the B16 melanoma and sublines that have been derived from it.

Inhibition of human breast cancer metastasis in nude mice by synthetic glycoamines.

We have examined the effect of synthetic low molecular weight glycoamine analogues on the metastasis of MDA-MB-435 human breast carcinoma xenografts growing in the mammary fat pads of nude mice. Initial in vitro screening of a panel of synthetic glycoamines was performed using a clonogenic growth assay in 0.9% agarose. Eight of nine compounds manifested a significant dose-dependent inhibition of colony formation by MDA-MB-435 cells in 0.9% agarose. The relative activity ranks of the compounds, based on ID50S independently determined for each synthetic glycoamine analogue, identified N-(1-deoxy-D-lactulos-1-yl)-L-leucine (Lac-L-Leu), N-(1-deoxy-D-fructos-1-yl)-D-leucine (Fru-D-Leu), N-(1-deoxy-D-fructos-1-yl)-L-phenylalanine, and N-(1-deoxy-D-fructos-1-yl)-L-leucine as the most effective inhibitors of colony formation. Two separate experimental treatment protocols were used to examine the effect of selected synthetic glycoamines on human breast cancer growth and metastasis in athymic nude mice. Group A mice were treated intraperitoneally daily from day 2 after injection of the breast cancer cells until the end of the experiment (17 weeks). In group B, the mice were untreated until the mean tumor diameter was 10 mm, at which time daily i.p. treatment began. After 7 days, the primary tumors were resected, and the mice were treated for an additional 4 weeks (a total of 5 weeks of treatment). The synthetic glycoamines did not have significant antitumor effects, and there was no difference in the tumor incidence or tumor growth rates in mice treated continuously with synthetic glycoamines or PBS. The significant antimetastatic activity of synthetic glycoamines was detected in both experimental treatment protocols. In mice continuously treated with synthetic glycoamines according to protocol A, the incidence of metastasis was decreased 4.6-fold (P = 0.014) and 2.7-fold (P = 0.031) in mice treated with Fru-D-Leu and Lac-L-Leu, respectively. In mice in protocol B, the incidence of pulmonary metastasis was decreased 1.9-fold (P = 0.069) and 2.5-fold (P = 0.042) in mice treated with Fru-D-Leu and Lac-L-Leu, respectively. Correspondingly, the average number of spontaneous pulmonary metastases was reduced from 37 in control mice to 0.2 (P = 0.005) and 0.9 (P < 0.02) in mice treated according to the protocol A with Fru-D-Leu and Lac-L-Leu, respectively. Treatment of mice with N-(1-deoxy-D-fructos-1-yl)-L-leucine did not have significant antimetastatic effects, and no reduction in metastasis incidence or number was noted in mice treated with this synthetic glycoamine analogue. The treated animals had no apparent toxicity from chronic daily injection (up to 17 weeks of treatment) of synthetic glycoamines, and no obvious pathology was noted in the histological slides of the livers, kidneys, or spleens of the treated mice. Therefore, we have identified two synthetic glycoamines (Fru-D-Leu and Lac-L-Leu) that were the effective inhibitors of spontaneous human breast cancer metastasis in nude mice. Potential mechanisms for antimetastatic activity of synthetic glycoamines may include the inhibition of beta-galectin-mediated homotypic cancer cell aggregation and induction of apoptosis in target cells.

Malignant potential of cells isolated from lymph node or brain metastases of melanoma patients and implications for prognosis.

We studied the correlation between the formation of brain metastasis and the malignant growth potential of seven human melanoma cell lines, isolated from lymph node metastases (A375-SM, TXM-1, DM-4) or from brain metastases (TXM-13, TXM-18, TXM-34, TXM-40), and the potential of three variants of the mouse K-1735 melanoma. Growth rates in different concentrations of fetal bovine serum and colony-forming efficiency in semisolid agarose were measured, and the tumorigenicity and metastatic ability were determined in nude mice (for the human melanoma cell lines) or in C3H/HeN mice (for the K-1735 variants). The ability to form brain metastasis was tested by injection of cells into the carotid artery. A high colony-forming efficiency in agarose, especially at concentrations of agarose greater than 0.6%, corresponded with high tumor take rates, rapid tumor growth rates, and metastatic colonization of the lungs of the recipient mice. For the human melanomas, the lymph node metastasis-derived cells were more tumorigenic and metastatic than the brain metastasis-derived cells. In the K-1735 mouse melanoma, the tumorigenic and metastatic behavior of the cells after i.v. and s.c. injection corresponded with growth in agarose cultures. However, for growth in the brain after intracarotid injection, the different melanoma cell lines showed similar frequencies of tumor take, regardless of tumorigenicity in other sites of the recipient mice, although mice given injections of brain metastasis-derived cells survived longer than mice given injections of lymph node metastasis (human melanoma) or lung metastasis (K-1735 M-2)-derived cell lines. The results from the human and mouse melanoma cell lines show that the brain metastasis-derived cell lines were not more malignant than the lymph node or lung metastasis-derived cells. These data imply that the production of brain metastasis is not always the final stage of a metastatic cascade.

Mechanisms of human tumor metastasis studied in patients with peritoneovenous shunts.

The technique of peritoneovenous shunting for the alleviation of abdominal pain and distension in malignant ascites due to inoperable cancer, returns the fluid to the circulation via a one-way, valved, anastomosis between the peritoneum and the jugular vein. Surprisingly, although the patients treated with this technique receive direct infusions of malignant tumor cells into the blood, this study of 29 patients, 15 of whom came to autopsy, shows that they did not all develop metastases, some being completely free of such lesions despite long survival. Even when metastases do form, they are small and clinically asymptomatic, and the technique is therefore not hazardous. In some patients, inert tumor cells identifiable by natural markers were recognized in the tissues, but no growing metastases were observed. In others, the distribution of secondary deposits was unexpected in that metastases did not form in the organ containing the first capillary bed encountered, although hematogenous metastases had formed in other organs. Despite the fact that various factors such as (a) the small numbers of patients treated with the technique; (b) the sensitive nature of studies on terminally ill patients; and (c) the absence of consistency in the sample population with regard to factors such as length of survival and site of neoplasm, combine to reduce the number of suitable cases for study, the approach has unrivaled power and interest for those seeking to understand mechanisms underlying tumor metastasis in humans.

Metastatic potential of cloned murine melanoma cells transfected with activated c-Ha-ras.

We sought to determine whether the transfection of tumorigenic but not metastatic cells with the activated c-Ha-ras oncogene was invariably associated with acquisition of the metastatic phenotype. Three clonally derived lines of the K-1735 murine melanoma, characterized as nonmetastatic or poorly metastatic, were transfected with plasmids containing the 6.6-kilobase BamHI fragment of the mutant human c-Ha-ras gene and the neo gene, that confers resistance to neomycin (pSV2neoEJ). Cells transfected with pSV2neo, a plasmid containing the neo gene, served as controls for the procedure of Polybrene-mediated transfection. All cell lines were injected into syngeneic C3H/HeN and into athymic mice, and the results were compared with those produced by highly metastatic K-1735 M-2 cells. Although the pSV2neoEJ-transfected cells produced more rapidly growing s.c. tumors than the control cell lines did, the incidence of spontaneous metastasis was not increased. Following i.v. inoculation, the c-Ha-ras transfectants were retained in lung vasculature in greater proportions than pSV2neo counterpart transfectants were. The c-Ha-ras transfectants also produced significantly more lung tumor colonies, which grew faster than the few lung tumor colonies in mice given injections of control melanoma cells. We concluded that transfection of the activated c-Ha-ras oncogene into nonmetastatic K-1735 melanoma cells leads to accelerated tumor growth in vivo and can confer the ability to form lung colonies after i.v. injection but not the ability to metastasize from a primary s.c. tumor.