Fibrin as a component of the tumor stroma: origins and biological significance

Correlation of the in vivo anticoagulant, antithrombotic, and antimetastatic efficacy of warfarin in the rat

Fibrin formation has been hypothesized to be an element of the metastatic process in cancer, and pharmacological interference with such fibrin formation has been proposed as a means of antimetastatic therapy. We have tested this hypothesis through an in vivo study of warfarin in two independent rat disease models--a model of chemical-injury-induced arterial thrombosis, and a model of spontaneous metastasis. We found 0.50 mg/kg-day warfarin to be uniformly lethal after two weeks treatment. The chronic dose of 0.25 mg/kg-day was non-toxic and produced effective anticoagulation and marked antithrombotic and antimetastatic activity. The 0.125 mg/kg-day dose produced a reduction in factor IIc (50%) and factor VIIc (70%), and resulted in statistically significant antithrombotic and antimetastatic activity. The 0.0625 mg/kg-day dose failed to reduce the vitamin K-dependent clotting factors, and failed to produce any antithrombotic or antimetastatic effects. The substantial correlation (very similar dose-response effects) among the anticoagulant, antithrombotic and antimetastatic efficacies of warfarin in the rat suggests that anticoagulation provides the pharmacological mechanism underlying both the antithrombotic and the antimetastatic effects. The poor therapeutic index we observed in the rat may be the attribute which limits the efficacy of warfarin in the treatment of human cancer.

Initiation of coagulation by tissue factor

Tissue factor (TF) is an integral membrane glycoprotein which functions as an initiator of coagulation. Furthermore, it is probably the principal biological initiator of this essential hemostatic process. This article reviews the studies which form the basis for these assertions. The work on TF is traced from the 19th century discovery of the thromboplastic activity of tissues to the recent purification of the protein from bovine and human tissues and the isolation cDNA clones coding from human TF. The features of TF structure and function which tailor it to the role of initiator of the coagulation cascade are considered. For example, cell-surface TF and factor VII, the plasma serine proteases zymogen, form a proteolytic complex without prior proteolysis of either component. In addition, a kinetic model for the molecular mechanism of TF-initiated clotting is reviewed. The factors which control the expression of TF procoagulant activity by cultured cells are examined in light of the hypothesized role of TF in normal hemostasis. Also, the potential pathological consequences of aberrant TF expression, i.e., thrombosis and hemorrhage, are explored.

Phorbol ester stimulates macrophage invasion of fibrin matrices

Macrophages migrate through a fibrin-rich extracellular matrix in chronic inflammation, wound healing, and other pathophysiological processes. To investigate the factors that might influence the ability of mononuclear phagocytes to invade fibrin matrices, we cultured macrophage-like P388D1 cells as well as resident and thioglycollate-elicited mouse peritoneal macrophages on three-dimensional fibrin gels, and we examined the effect of agents known to stimulate a variety of macrophage functions, including the production of fibrinolytic enzymes. Cells grown on fibrin gels under control conditions, as well as cells treated with either bacterial lipopolysaccharide or concanavalin A, remained confined to the gel surface. In contrast, the tumor promoter 4 beta-phorbol 12-myristate 13-acetate (PMA) induced both P388D1 cells and peritoneal macrophages to invade the underlying fibrin matrix. The invasive behavior of PMA-treated P388D1 cells was not affected by protease inhibitors of various specificities. These results demonstrate that certain exogenous signals can profoundly modify the ability of macrophages to migrate through fibrin matrices.

Phorbol ester induces cultured endothelial cells to invade a fibrin matrix in the presence of fibrinolytic inhibitors

We have previously shown that the tumor promoter 4 beta-phorbol 12-myristate 13-acetate (PMA) induces capillary endothelial cells grown to confluency on the surface of three-dimensional collagen gels to invade the underlying matrix and to form capillary-like tubular structures, a phenomenon mimicking angiogenic processes that occur in vivo (Montesano and Orci: Cell, 42:469-477, 1985). Since angiogenesis frequently occurs within a fibrin-rich extracellular matrix, we have examined the ability of PMA-treated endothelial cells to invade fibrin gels. Control endothelial cells grown on fibrin gels formed a confluent monolayer on the gel surface and did not invade the underlying matrix. Treatment of the cultures with PMA resulted in a progressive lysis of the substrate without invasion of the fibrin matrix. However, if the cells were treated with PMA either in the presence of fibrinolytic inhibitors (Trasylol, epsilon-aminocaproic acid) or in the absence of detectable plasminogen, dissolution of the substrate was prevented, and the endothelial cells invaded the fibrin gel, forming vessel-like tubular structures similar to those previously observed with collagen gels. These results demonstrate that the invasive and morphogenetic events induced by PMA do not necessarily require an interaction between endothelial cells and collagen fibrils but can also occur with other biologically relevant substrata. They also suggest (1) that invasion may occur via a plasmin-independent mechanism and (2) that in vivo, neutralization of excess proteolytic activity may play an important permissive role in angiogenesis and other invasive processes by preventing uncontrolled matrix degradation.

Characterization of factor XIII containing-macrophages in lymph nodes with Hodgkin's disease

A large number of cells containing subunit a of blood coagulation Factor XIII (FXIII) was detected by immunoperoxidase staining in lymph nodes with Hodgkin's disease. These relatively large, multipolar, mononuclear cells were often found in the immediate vicinity of malignant Hodgkin's cells. Intensive characterization of these cells carried out by immunofluorescent and enzymecytochemical techniques in double- and triple-labelling systems on the same sections clearly demonstrated that they represent tumour-associated macrophages (TAMs). FXIII containing-cells showed alpha-naphtyl acetate esterase (ANAE) positivity, and were labelled by monoclonal anti-Leu M3 antibody, a monocyte/macrophage marker, but not at all or only very weakly by anti-HLA-DR. Neither alkaline phosphatase (ALP) nor adenosine triphosphatase (ATPase) activity could be detected in these cells and surprisingly, they were consistently negative for acid phosphatase (AcP) as well. The presence of FXIII subunit a in tumour-associated macrophages suggests that this cell type might have an important role in the stabilization of fibrin deposits around tumour cells.

Tumours acquire their vasculature by vessel incorporation, not vessel ingrowth

This study was designed to investigate the early development of tumour circulation using a transplantable mouse mammary adenocarcinoma. Tumour cells (10(6] were injected subcutaneously into the flank and groups of treated mice were killed at 24 h intervals up to 12 days; the tumours and surrounding tissues being processed by standard histological methods. Sections of tumour were sub-divided into neoplastic tissue, vessels and connective tissue using computer-assisted morphometric techniques: the host tissue around the tumour was similarly quantified for vessels and connective tissue. With increasing tumour size, the proportion of vessels within tumours rapidly increased to reach a plateau of approximately 1.5 per cent of tumour volume, a 400 per cent increase on the vascular density of normal subcutaneous tissue. Within tumours, vascular density was always higher at the periphery than the centre. The most pronounced increase in vascular density affected the host tissue around the tumour. It is not clear why vascular development is most prominent outwith the tumour when postulated angiogenic factors, such as tumour angiogenesis factor, are presumably released within. Our results imply, however, that tumours acquire their vasculature by infiltration into, and expansion between, a network of newly formed vessels in the surrounding connective tissue.

Thrombin stimulates arachidonate metabolism in murine tumor cells

Thrombin can be formed in the tumor cell microenvironment following activation of the clotting cascade by procoagulants of cancer or host cells. We have tested here the effects of thrombin, either "endogenous" or "exogenous" (see below), on arachidonate mobilization from membrane phospholipids of mouse mammary tumor virus-induced (MMTV) carcinoma cells. These tumor cells exhibit in vitro a tissue type procoagulant activity (130 thromboplastin units/10(4) cells) and are therefore able to induce thrombin formation in a plasmatic milieu. To verify the effect of thrombin formation by tumor cell procoagulant ("endogenous thrombin"), either human or mouse platelet-free plasma (20% in DMEM) was added to the cell layer (prelabelled for 5 hr with a trace amount (0.013 microM) of 3H-arachidonate) and the system was recalcified (15 mM CaCl2). Thin-layer radiochromatography of the culture medium showed a significant release of 3H-labelled arachidonate products PGE2, PGF2 alpha and 6-ketoPGF1 alpha after 1 hr of incubation. To verify the effect of thrombin formation from host sources ("exogenous thrombin"), either bovine or purified human alpha-thrombin (0.1-10 U/ml) was added to the cells for different periods (from 5 min to 20 hr). Exogenous thrombin stimulated arachidonate release and metabolism in a dose-related manner. With short labelling periods (0.013 microM 3H-arachidonate for 30 min-1 hr) thrombin stimulated the release of unmetabolized 3H-arachidonate, but not of 3H-arachidonate metabolites. These processes were inhibited by a specific inhibitor of thrombin enzymatic activity (alpha-NAPAP, 140 microM) and by a cyclo-oxygenase inhibitor (ASA 4mM). Tumor-associated procoagulants may thus contribute not only to fibrin deposition but also to generation of multipotent mediators such as arachidonate metabolites.

Thrombosis and cancer

Abnormal hemostasis is a fundamental property of malignant disease, not merely an epiphenomenon attributable to therapy or to chronic illness. Many types of tumor cells express clotting initiators such as tissue factor and act again late in the coagulation pathway by providing a surface for prothrombinase generation. Thus, entry of tumor cells into the plasma, as during metastasis, may be expected to trigger intravascular clotting. However, and perhaps of greater importance, solid tumors growing outside of the blood vasculature regularly deposit fibrin locally in the tissues. They do so by rendering the microvasculature hyperpermeable, allowing fibrinogen and other plasma-clotting proteins to leak into the extravascular space where procoagulants associated with tumor cells or with benign stromal cells initiate clotting. Both fibrin deposition and turnover in solid tumors proceed at rapid rates. Thus, whether attributable to events in the intra- or extra-vascular space, the result is the same: abnormal clotting and fibrinolysis whose consequences may include protection from host inflammatory cells, modulation of the immune response, and induction of angiogenesis.

Ultrastructure of tumour invasion and desmoplastic response of bronchogenic squamous cell carcinoma

Using ultrastructural methods we studied the interaction of tumour cells and lung parenchyma in deep areas (i.e., more than about 3 mm from the tumour surface) of 50 bronchogenic squamous cell carcinomas. The tumour periphery, studied previously, had shown organized associations of tumour cells and lung epithelial cells and a surprising lack of invasion of non-epithelial tissue compartments. The deeper areas, where the tumour cells and the lung parenchyma had been in contact for longer periods, consisted of irregular groups of tumour cells and desmoplastic stroma which was very similar to granulation tissue. The deeper areas also contained many intact lung epithelial cells, arranged in compressed and distorted alveolar structures. Where non-neoplastic epithelial cells and tumour cells had direct contact, they formed common junctional complexes and basal laminae. In part of the tumours, the cells were largely devoid of a basal lamina. However, in most instances a continuous basal lamina surrounded every tumour cell group studied, even when these formed irregular strands or seemed to be completely isolated.

Plasminogen activators, plasminogen activator inhibitors and procoagulant analyzed in twenty human tumor cell lines

We have analyzed the CM of 20 human tumor cell lines for the presence of PA, PA-I and PC. Most of the cell lines expressed PA activity as measured by a radioiodinated fibrin plate assay. The urinary type and tissue-type PA activities were specifically quantified by means of purified inhibitory antibodies. U-PA and/or t-PA antigen, as measured by radioimmunoassays, were detected in all but 4 of the CM and were generally 10 times more concentrated than PA activity, indicating the presence of specific PA-Is. Analysis of CM by electrophoresis followed by fibrin-agarose zymography demonstrated the presence not only of free but also of inhibitor-complexed PA. Affinity purification demonstrated that 8/20 cell lines expressed detectable PA-I activity. The PA-I1 and PA-I2 inhibitors were most frequently observed, while PN was recovered only from CM of the HT1080 fibrosarcoma cell line. PC activity, as measured by the plasma recalcification time method, was found in 9/20 CM. It was of the thromboplastin tissue factor type since most of its activity was lost when assayed with a Factor VII-deficient plasma.

Promotion of lung tumor colonization in mice by the synthetic thrombin inhibitor (no. 805) and its reversal by leech salivary gland extracts

The role of anticoagulation per se in the reduction of experimental or spontaneous metastasis still remains to be determined, as shown by the conflicting results reported by the literature using different conventional anticoagulants. A new compound has been synthesized (compound no. 805) which prolongs or suppresses coagulation via specific inhibition of thrombin and its possible use in a model of experimental metastasis to clarify the role of anticoagulants in tumor spread was investigated. Contrary to our expectations, this compound increased rather than decreased the number of lung colonies induced by intravenous injections of a variety of murine neoplasias. Studies of the mechanism of this effect indicated that the compound increases retention of tumor cells by the lung without apparent impairment of the natural cell immune system, suggesting that the synthetic thrombin inhibitor may enhance vascular attachment of tumor cells. The promoting effect of compound no. 805 on metastasis was totally reversed by the administration of leech salivary gland extracts, which appear to protect capillaries from damage produced by cyclophosphamide, as revealed by other studies.

Procoagulant activity of human alveolar macrophages: different expression in patients with lung cancer

Mononuclear phagocytes, an integral part of the lymphoreticular infiltrate of many malignant tissues, might contribute to tumor-associated fibrin deposition through the production of procoagulant activity (PCA). We have studied the PCA of human alveolar macrophages in 28 patients with primary lung cancer and in 9 control subjects. Alveolar macrophages (greater than 97% esterase positive) were isolated form bronchoalveolar lavage fluids by adherence onto plastic. PCA was evaluated by a one-stage clotting assay immediately after isolation (basal PCA) and after incubation (4 hr at 37 degrees C) in the absence and in the presence of endotoxin. Cells from control subjects had low basal PCA (3.9 +/- 1.0 units/5 X 10(4) cells) but, upon exposure to endotoxin, they displayed a 5- to 16-fold increase in PCA. In patients, different patterns of PCA were observed. In the 8 patients in whom lavage had been carried out on the contralateral side to the neoplasm, alveolar macrophages behaved essentially like those from controls. In contrast, in the 20 patients in whom macrophage populations close to the site of the tumor were examined, PCA was abnormal in many respects. In 12 of these, alveolar macrophages had basal PCA comparable to or somewhat lower than control cells, but exhibited a poor procoagulant response when incubated in vitro in the presence of endotoxin. Alveolar macrophages from the remaining 8 patients expressed far higher levels of basal PCA than control cells (25.1 +/- 5.9 units as compared to 3.9 +/- 1.0 units/5 X 10(4) cells). These cells retained their ability to respond to endotoxin in vitro with a 3-fold increase in PCA. In all instances, alveolar macrophage PCA had the characteristics of tissue factor. These data suggest that the presence of primary lung cancer may modulate the expression of PCA in alveolar macrophages close to the tumor site. PCA might be useful to better characterize the functional state of macrophages near the tumor.

Procoagulant and fibrinolytic activity of human ovarian carcinoma cells in culture

We have investigated here the coordinate expression of both procoagulant (PCA) and fibrinolytic (FA) activity of cells from 16 human ovarian carcinoma cases. To avoid interference of contaminating host cells, we used cells isolated in primary culture from ascitic fluid or from solid tumor. The FA was determined in cellular extracts by an amidolytic assay in the presence of fibrin monomers. FA, which was plasminogen dependent in almost all of the cases, showed a wide range of activity (from less than 0.001 to 2.30 UK units/mg protein). The molecular analysis of plasminogen activator (by SDS-PAGE and fibrin autography) showed a single molecular form of 52,000 daltons, inhibited by an antibody against human urokinase. PCA, studied with a one stage clotting assay in disrupted cells, was of tissue thromboplastin type in all instance and varied from 12.0 to 1300 thromboplastin units/10(4) cells. No simple correlation was found between FA and PCA in the cellular samples studied; moreover, for neither parameter was it possible to find any changes with the staging of the disease.

Cellular and vascular manifestations of cell-mediated immunity

The studies reviewed here emphasize both the complexity and the heterogeneity of cell-mediated immunity. In addition to the round cell infiltrate of the classic descriptions, cell-mediated immunity includes reactions that feature many types of inflammatory cells, that exert profound effects on the blood microvasculature, and that initiate extravascular clotting and, possibly, angiogenesis. The common denominator of all of these reactions is a subset or subsets of sensitized T lymphocytes that, on exposure to specific antigen, recruit and collaborate in other ways with one or more populations of circulating bone marrow-derived cells. Although the reactions generally resemble chronic inflammation by virtue of the lymphocytes and monocytes present, cell-mediated immunity may also take the guise of acute or subacute inflammation when neutrophils or eosinophils predominate and an entirely different morphologic pattern when basophils predominate, as in CBH. Tissue mast cells undergo changes (activation, proliferation) that are generally observed at later stages of delayed hypersensitivity, but no convincing evidence has been presented that these cells play an essential role in the elicitation of cell-mediated immunity. The concept that an essential prerequisite for the elicitation of delayed hypersensitivity is the mast cell-dependent generation of microvascular gaps, favoring inflammatory cell diapedesis, is clearly incorrect. First, lymphocytes fail to traverse certain of the vessels that exhibit such gaps (i.e., those of the SCV) in delayed hypersensitivity reactions in humans. Second, there is no diminution in the cellular infiltration associated with cell-mediated immunity reactions in mast cell-deficient mice. Cell-mediated immunity does not consist of an inflammatory cell infiltrate alone. The local microvasculature is rendered hyperpermeable to varying extents, with resulting extravasation of plasma proteins, including fibrinogen. The majority of extravasated fibrinogen is clotted to cross-linked fibrin, presumably as the result of the actions of procoagulants associated with fixed connective tissue cells and perhaps also because of the activity of infiltrating cells, such as monocytes/macrophages. Clotted fibrin forms a water-trapping gel, which accounts for the induration seen in many delayed hypersensitivity reactions. The microvasculature may also be affected in other ways. Endothelial cells may undergo hypertrophy and cell division or, alternatively, may exhibit profound and progressive injury.(ABSTRACT TRUNCATED AT 400 WORDS)

Change in fibrinogen turnover in tumors by hyperthermia

The effect of hyperthermia on the turnover of fibrinogen and albumin in the SCK tumor of A/J mice was studied. Immediately after heating at 43.5 degrees C for 30 min, the content of 131I-labeled fibrinogen in the tumors was about 2.7-fold that in the control tumor and remained elevated for 24 hr. On the other hand, the content of 125I-labeled serum albumin in the tumor immediately after heating was only 1.7-fold that in the control tumor and started to decrease soon after heating. The greater increase in the accumulation and the longer retention of the labeled fibrinogen as compared with the labeled albumin in the heated tumors appeared to be related to the heat-induced vascular damage accompanied by vascular occlusion, fibrination and subsequent thrombus formation.

Procoagulant activity of mouse and human cultured cells following various types of transformation

The presence of fibrin deposits in the microenvironment of tumor cells has been reported repeatedly and considered to play an important role in tumor biology. Among the mechanisms by which fibrin may be deposited in tumors, procoagulant activities (PCA) of different types have been described in cancer cells. The present study was aimed at establishing whether the nature of cellular PCA was a characteristic associated with malignant transformation. PCA of normal and transformed cells was investigated on pairs of murine and human origin. The transformed counterparts were obtained after treatment with low-dose radiation, chemical carcinogen, viral infection or after in vitro spontaneous immortalization. Both before and after any type of transformation cell PCA was of the tissue thromboplastin type, identified on the basis of biological criteria: requirement of factor VII for its expression and lack of inhibition by the serine protease inhibitor diisopropylfluorophosphate (DFP). Transformed cells of murine origin showed significantly lower activity than their normal counterparts, whereas all the transformed human cell lines expressed significantly higher activity than normal. An inverse correlation between the levels of PCA and the cell density in culture was observed in all but one of the lines tested. These findings suggest that the factor X activating property described in some tumors or in transformed cells cannot be considered as a general marker of transformation.

Regulation of extravascular coagulation by microvascular permeability

Extravascular coagulation is a prominent feature of such important pathological processes as cellular immunity and neoplasia and has been thought to result from procoagulants associated with the inflammatory or tumor cells peculiar to these entities. It was found that increased microvascular permeability alone is sufficient to induce equivalent extravascular coagulation in several normal tissues. The results indicate that saturating levels of procoagulant are present even in normal tissues and that microvascular permeability is a rate-limiting step in extravascular coagulation.

Cancer metastasis: experimental approaches, theoretical concepts, and impacts for treatment strategies

It has been the purpose of this article to describe recent advances in cancer metastasis research. Clinical realities and experimental approaches to the study of underlying basic mechanisms of metastasis formation were discussed. Wherever possible, results were reported which led to the development of theoretical concepts. Such results and concepts were finally evaluated in light of their possible impact for the design of new treatment strategies. Experimental findings from many diverse research fields were summarized with the help of tables, figures, and references. It was concluded that the process of metastasis is a dynamic event that can be described as a sequence of interrelated steps. Experimental results indicated that malignant cells that migrate and disseminate from the primary organ to distant sites and there eventually develop into metastases have to survive a series of potentially lethal interactions. Intimate tumor-host interactions were reported to take place all along the metastatic process. They were elucidated at the steps of angiogenesis, invasion, organ interaction, dormancy, tumor rejection, and tumor immune escape. The outcome of such tumor-host interactions seemed to depend on intrinsic properties of the tumor cells themselves as well as on the responsiveness of the host. Metastasis does not appear as a merely random process. Both clinical and experimental studies revealed that the whole process can be described more appropriately in terms of stochastic, sequential, and selective events, each of which is controlled and influenced by a number of mechanisms. With regard to therapeutic intervention, a selective event offers more possibilities than a random one because it is governed by rules that can be exploited experimentally. Various impacts from experimental studies for the design of antimetastatic cancer treatment strategies were discussed. Sequential steps of the metastatic cascade could become new therapy targets. Conventional empirically derived treatment modalities should become flanked by methods aimed more specifically at critical steps of cancer spread in order to prevent progression of the disease. This is where basic research on mechanisms could make significant contributions to therapy planning in the future. Furthermore, possible negative effects of surgery, radiotherapy, and adjuvant chemotherapy or immunotherapy that could result in enhancement of metastatic progression need to be critically evaluated to limit them as much as possible.(ABSTRACT TRUNCATED AT 400 WORDS)

Relationship of tumor leucocytic infiltration to host defense mechanisms and prognosis

The interface between the tumor and the host is often the site of leucocytic infiltration. We will examine the idea that the infiltrating leucocytes of human and experimental tumors are components of the host immunological defense against the tumor, and that the presence of the infiltrate is a marker of favorable prognosis. Leucocytes could infiltrate tumors because of an active immune response, either nonspecific or specifically directed to tumor-associated antigens. Leucocyte influx may also occur because of chemotactic factors secreted by the tumor cells. Some tumors release factors which enhance vascular permeability and permit improved access by leucocytes to the tumor focus. The consequences of leucocytic infiltration include tumor cell cytolysis, cytostasis, or stimulation of proliferation. The present state of our knowledge of the interactions between tumor cells and infiltrating leucocytes precludes broad generalization of mechanisms. Further study will probably reveal that the mechanisms are diverse, and that there are some systems in which immune interactions occur at this interface and others in which they do not.