Characterization of a chimeric plasminogen activator consisting of a single-chain Fv fragment derived from a fibrin fragment D-dimer-specific antibody and a truncated single-chain urokinase.

An Mr 57,000 single-chain chimeric plasminogen activator, K12G0S32, consisting of a variable region fragment (Fv) derived from the fibrin fragment D-dimer-specific monoclonal antibody MA-15C5 and of a 33-kDa (amino acids Ala132 to Leu411) recombinant single-chain urokinase-type plasminogen activator (rscu-PA-33k) was studied. K12G0S32, secreted by infected Spodoptera frugiperda insect cells at a rate of 1.5 micrograms/10(6) cells/48 h, was purified to homogeneity by ion-exchange chromatography and gel filtration. It was obtained essentially as a single-chain molecule with a Ka = 5.5 x 10(9) M-1 for immobilized fragment D-dimer, similar to that of MA-15C5. The specific activity of both its single-chain and two-chain forms on fibrin plates was 100,000 IU/mg of urokinase-type plasminogen activator (u-PA) equivalent. Activation of plasminogen by two-chain K12G0S32 obeyed Michaelis-Menten kinetics with Km = 2.9 +/- 0.6 microM and a k2 = 3.7 +/- 0.6 s-1 (mean +/- S.D.; n = 3), as compared to Km = 12 microM and k2 = 4.8 s-1 for rtcu-PA-32k (recombinant low Mr two-chain u-PA consisting of amino acids Leu144 to Leu411). Single-chain K12G0S32 induced a dose- and time-dependent lysis of a 125I-fibrin-labeled human plasma clot immersed in citrated human plasma; 50% lysis in 2 h was obtained with 0.70 +/- 0.07 micrograms/ml (mean +/- S.D.; n = 5), as compared with 8.8 +/- 0.1 micrograms/ml for rscu-PA-32k (recombinant low Mr single-chain u-PA consisting of amino acids Leu144 to Leu411) (mean +/- S.D.; n = 3). With two-chain K12G0S32, 50% clot lysis in 2 h required 0.25 +/- 0.03 micrograms/ml (mean +/- S.D.; n = 3), as compared with only 0.62 +/- 0.04 micrograms/ml (mean +/- S.D.; n = 2) for rtcu-PA-32k. These results indicate that low Mr single-chain u-PA can be targeted to a fibrin clot with a single-chain Fv fragment of a fibrin-specific antibody, resulting in a 13-fold increase of the fibrinolytic potency of the single-chain form and a 2.5-fold increase of the potency of the two-chain form.

The anti-invasive flavonoid (+)-catechin binds to laminin and abrogates the effect of laminin on cell morphology and adhesion.

To study the effect of the flavonoid (+)-catechin on cell-matrix interactions two cell types with a different morphology on and adhesion to laminin were used. MO4 virally transformed fetal mouse cells adhere and spread when cultured on top of laminin-coated coverslips or on human amnion basement membrane. M5076 mouse reticulum cell sarcoma cells poorly adhere to these substrates and remain round. Both cell types are invasive in confronting cultures with embryonic chick heart fragments. (+)-Catechin binds to laminin in a pH-dependent way. Pretreatment of laminin-coated coverslips or amnion basement membrane with 0.5 mM (+)-catechin abrogates the effect of laminin on cell morphology and adhesion. MO4 cells do not adhere to the pretreated substrates and remain round, while M5076 cells now adhere and spread. (+)-Catechin inhibits the invasion of MO4 cells but not of M5076 cells into embryonic chick heart in vitro. We speculate that the anti-invasive activity of the flavonoid to MO4 cells is the result of its interference with MO4 cell adhesion to laminin. Invasion of M5076 cells does not imply adhesion to and spreading on laminin.

Action mechanisms of anti-invasive agents.

An anti-invasive activity has been observed with a number of agents in confronting cultures between invasive cell populations and embryonic chick heart fragments. Cytochalasins, microtubule inhibitors and dipyridamole act via intracellular targets. Inhibition of glycosylation of proteins, low temperature and alkyl lysophospholipids alter the plasma membrane. Flavonoids aim at extracellular targets. In a unifying hypothesis we postulate that invasion is due to lack of responsiveness to stop signals that normally retain cells within their tissue boundaries. Some of the anti-invasive agents act on cellular activities that are necessary for invasion; others might act on the stop signal, the perception of and/or the response to this signal.

Confrontation of an invasive (MO4) and a noninvasive (MDCK) cell line with embryonic chick heart fragments in serum-free culture media.

Confronting cultures of precultured embryonic chick heart fragments (PHF) with aggregates of malignant cells in vitro have been shown to be relevant for a number of aspects of tumor invasion in vivo. Preculture of the heart fragments, formation of cell aggregates, and subsequent culture of confronting pairs have so far been done only in serum-containing culture media. We describe here confronting cultures of PHF with invasive MO4 mouse cell aggregates or noninvasive MDCK dog kidney cell aggregates in serum-free media. Heart fragments precultured in the absence of serum seemed to be necrotic after confronting culture in serum-free media. However, preculturing in media supplemented with 10% fetal bovine serum allowed us to do subsequent confronting cultures in absence of serum. Cell aggregates were also prepared in serum-containing medium. MO4 cells occupied and replaced the heart tissue within 4 d, whereas MDCK cells remained at the periphery of the PHF. This indicates that serum-free confronting cultures can discriminate between invasive and noninvasive cells. The viability of individual PHF and cell aggregates cultured in the same way as in confrontations was ascertained by histology and by explantation and postculturing on a solid tissue culture substrate. Growth of the cultures was smaller in serum-free media than in media supplemented with 10% fetal bovine serum. The main advantage of serum-free culture conditions in vitro is the elimination of the influence of serum components on invasion, and the ability to examine the effect on invasion of drugs that are susceptible to inactivation by serum.

Effect of temperature on invasion of MO4 mouse fibrosarcoma cells in organ culture.

Invasion by MO4 mouse fibrosarcoma cells into fragments of embryonic chick heart or lung in organ culture was studied histologically and ultrastructurally at various temperatures between 12 and 40 degrees C. Invasion was absent for at least 7 days at or below temperatures of 29 degrees C. Invasion was invariably observed at or above 30.5 degrees C. Differences in invasion between 29 and 30.5 degrees C could not be ascribed to differences in growth, migration, or microtubule assembly/disassembly of MO4 cells. Neither could they be explained through differences in the attachment of MO4 cells to the heart fragments. Possible explanations for the absence of invasion at lower temperature are: altered resistance of the extracellular matrix in heart or lung fragments, and deficient expression of fucosylated glycoproteins at the surface of MO4 cells. A population of MO4 cells plated from the parent line and adapted to grow at 28 degrees C (MO(4)28 cell line) did not differ in invasiveness from the parent MO4 cells. We conclude that the temperature dependence of invasion in organ culture might indicate as yet unexplored aspects of the mechanisms of tumour invasion.

(+)-Catechin inhibits the invasion of malignant fibrosarcoma cells into chick heart in vitro.

(+)-Catechin, a flavonoid extractable from higher plants and trees, inhibits the invasion of malignant MO4 fibrosarcoma cells into embryonic chick heart fragments in vitro. This inhibition is maximal at a drug concentration of 0.5 mM. The growth of the MO4 cells is only partly inhibited at such a concentration, and the effect can hardly be ascribed to an irreversible cytotoxicity of the drug to the host tissue. Pretreatment of the host tissue with the drug seems to be a prerequisite for the inhibitory action, which suggests that the anti-invasive effect is at least partly mediated by the condition of the heart tissue. We hypothesize that the collagen-stabilizing effect of (+)-catechin is the key to the explanation of its anti-invasive properties.

Interaction of malignant cells with salt-extracted cartilage in vitro.

A salt-extractable low-molecular-weight fraction has been held responsible for the resistance of cartilage to invasion by malignant cells. To test this hypothesis, we have confronted in vitro fragments from bovine articular cartilage, from bovine nasal septum, from chick embryonic tibia, or from human knee meniscus with cells from the following malignant lines: LICR-HN-4 human squamous cell carcinoma; B16BL6 mouse melanoma; Hu-456 human transitional cell carcinoma; TE-85 and SAOS-2 human osteosarcoma; and MO4 mouse fibrosarcoma. Human embryo lung cells and chick embryonic heart cells were used as nonmalignant counterparts. Confronting cultures using living cartilage and cartilage extracted with 3 M guanidinium hydrochloride were examined light microscopically and ultrastructurally after 1 to 14 days. Neither invasion of malignant cells into the matrix of living or salt-extracted cartilage nor breakdown of collagen was observed in these cultures. In contrast, both malignant and nonmalignant cells occupied preexisting spaces in the matrix, namely, cut chondrocyte lacunae, cartilage canals, and fibrillation clefts. We concluded from these experiments that salt extraction did not alter the resistance of cartilage to invasion by malignant cells in vitro. This conclusion does not support the opinion that salt-extractable factors are solely responsible for the resistance of cartilage.

Glycosylated podophyllotoxin congeners: loss of microtubule inhibition and permission of invasion in vitro.

Podophyllotoxin (PPT) and its glycosylated congeners, 4'-demethylepipodophyllotoxin 9-[4,6-O-(R)-ethylidene-beta-D-glucopyranoside] (VP-16-213) and 4'-demethylepipodophyllotoxin 9-[4,6-O-(R)-thenylidene-beta-D-glucopyranoside] (VM-26), were studied for their effects on the following activities of MO4 mouse fibrosarcoma cells in vitro: growth as an aggregate in culture on a gyrotory shaker, directional migration from an aggregate explanted on glass, organization of the cytoplasmic microtubule complex as inferred from immunostaining with an antiserum against tubulin, and invasion into fragments of 9-day-old embryonic chick cardiac muscle. At concentrations that inhibited growth (greater than or equal to 0.03 microgram/ml), PPT arrested directional migration, abolished the cytoplasmic microtubule complex, and interfered with invasion. VM-26 (0.1-1.0 microgram/ml) and VP-16-213 (1-30 micrograms/ml) interfered with growth but permitted directional migration, organization of the cytoplasmic microtubule complex, and invasion. These observations imply that microtubule inhibitors are anti-invasive because they interfere with the assembly of the cytoplasmic microtubule complex and not because they inhibit growth.

Vinblastine, Vincristine and Vindesine: anti-invasive effect on MO4 mouse fibrosarcoma cells in vitro.

Inhibition of the invasiveness of MO4 mouse fibrosarcoma cells by the vinca alkaloid, vinblastine (VLB), vincristine (VCR) and vindesine (VDS), has been examined in vitro. At doses between 0.006 microgram/ml (minimal effect) and 0.1 microgram/ml (complete inhibition) these drugs interfered with the invasion of MO4 cells from an aggregate confronting a fragment of embryonic chick heart in three-dimensional culture. We have also examined the effect of these drugs on the following activities of MO4 cells: growth, directional migration and assembly of the cytoplasmic microtubule complex. Growth and directional migration were affected by the same doses of vinca alkaloids as invasion. In contrast with the vinca alkaloids, 5-fluorouracil at 1 microgram/ml inhibited growth but allowed directional migration and invasion. At a dose of 0.3 microgram/ml VLB, VCR and VDS interfered with the assembly of cytoplasmic microtubules, as visible after immunocytochemical staining with tubulin antiserum. Ultrastructural analysis demonstrated that inhibition of invasion in three-dimensional culture corresponds with abolishment of the cytoplasmic microtubule complex. Anti-invasive concentrations of VLB, VCR and VDS represent clinically achievable plasma concentrations. We concluded that the anti-invasive effect of the vinca alkaloids may contribute to their antitumor activity.

Invasiveness of malignant ST/A mouse lung cells in vitro.

ST/A mouse lung cells underwent apparently spontaneous malignant alteration in tissue culture. We have compared the capacity of these cells to form malignant tumours in syngeneic animals with their behaviour in vitro. ST-L1, ST-L22, and ST-L104 cells were malignant, whilst ST-L108 and ST-L109 cells were not. ST-L1 and ST-L22 cells showed anchorage-dependence of growth, whilst ST-L104, ST-L108 and ST-L109 cells did not. ST-L22, ST-L104, ST-L108 and ST-L109 performed directional migration from a spheroid explanted on glass. This capacity was lost in ST-L1 cells, which produced so-called round-cell transformants. All but ST-L108 cells produced type C viral particles. The tumorigenic cell lines ST-L1, ST-L22 and ST-L104 invaded fragments of embryonic chick heart in three-dimensional culture, whereas the non-tumorigenic ST-L108 and ST-L109 cells did not. Furthermore, the histology of ST-L cells invading in three-dimensional culture resembled that of the invasive sarcomas which they produced in vivo. The present observations with ST-L cells confirm that invasiveness in three-dimensional culture is a reliable criterion for malignant of tissue culture lines.