Vascular endothelial invasion via transcellular passage by malignant cells in the primary stage of metastases formation

Interaction of Drug-Sensitive and -Resistant Human Melanoma Cells with HUVEC Cells: A Label-Free Cell-Based Impedance Study

Cancer cell extravasation is a crucial step in cancer metastasis. However, many of the mechanisms involved in this process are only now being elucidated. Thus, in the present study we analysed the trans-endothelial invasion of melanoma cells by a high throughput label-free cell impedance assay applied to transwell chamber invasion assay. This technique monitors and quantifies in real-time the invasion of endothelial cells by malignant tumour cells, for a long time, avoiding artefacts due to preparation of the end point measurements. Results obtained by impedance analysis were compared with endpoint measurements. In this study, we used human melanoma M14 wild type (WT) cells and their drug resistant counterparts, M14 multidrug resistant (ADR) melanoma cells, selected by prolonged exposure to doxorubicin (DOX). Tumour cells were co-cultured with monolayers of human umbilical vein endothelial cells (HUVEC). Results herein reported demonstrated that: (i) the trans-endothelial migration of resistant melanoma cells was faster than sensitive ones; (ii) the endothelial cells appeared to be strongly affected by the transmigration of melanoma cells which showed the ability to degrade their cytoplasm; (iii) resistant cells preferentially adopted the transcellular invasion vs. the paracellular one; (iv) the endothelial damage mediated by tumour metalloproteinases seemed to be reversible.

Structure and biomechanics of the endothelial transcellular circumferential invasion array in tumor invasion

Cancer cells breach the endothelium not only through cell-cell junctions but also via individual endothelial cells (ECs), or transcellular invasion. The underlying EC forms a circular structure around the transcellular invasion pore that is dependent on myosin light chain kinase (MLCK) and myosin II regulatory light chain (RLC) phosphorylation. Here we offer mechanistic insights into transcellular invasive array formation amid persistent tensile force from activated EC myosin. Fluorescence recovery after photobleaching (FRAP) experiments, sarcomeric distance measurements using super-resolution microscopy and electron microscopy provide details about the nature of the myosin II invasion array. To probe the relationship between biomechanical forces and the tension required to maintain the curvature of contractile filaments, we targeted individual actin-myosin fibers at the invasion site for photoablation. We showed that adjacent filaments rapidly replace the ablat11ed structures. We propose that the transcellular circumferential invasion array (TCIA) provides the necessary constraint within the EC to blunt the radial compression from the invading cancer cell.

Vascular invasion of O-1N, hamster squamous cell carcinoma with high potential of lymph node metastasis: ultrastructural comparison between lymphatics and blood vessels

The ultrastructural modes of lymphatic and blood vessel invasions were studied comparatively in hamsters with squamous cell carcinoma (O-1N) that had a high potential for lymph node metastasis. The endothelial injury, which was caused by mechanical stretching with the growth of O-1N, was the initial and characteristic feature common to both vascular invasions. Tumor cell nests penetrating the lymphatic lumen through disrupted endothelial cells still maintained their volume and continuity to the underlying tumor cell nests. In contrast, pronounced microthrombotic and neutrophilic reactions occurred at the site of blood vessel penetration. Within the lymphatic lumen, large clusters of O-1N cells were kept longer in spite of lymphocytic and macrophagic reactions. In blood vessels, clusters of tumor cells that had passed through dense fibrin layers were reduced in size and further disintegrated into smaller pieces by neutrophils. In conclusion, lymphatic invasion is a mechanical process, and smooth and direct invasion of large tumor cell nests into lymphatic vessels is responsible for causing more prompt and frequent lymph node metastasis in O-1N than a hematogenous type.

Tumor cell motility and metastasis : Autocrine motility factor as an example of ecto/exoenzyme cytokines

Cellular locomotion plays a critical role in such normal processes as embryonic development, tissue segregation, as well as the infiltration of fibroblasts and vascular cells during wound repair and the inflammatory responses of the adult immune system. During tumor invasion and metastasis the processes of cell migration achieve dire significance. Disruption of normal homeostatic mechanisms to benefit the survival of the individual tumor cell is a common theme discovered during the characterization of factors once thought to be tumor-specific. One such molecule, tumor cell autocrine motility factor, was so described and has only recently been identified as a normal protein involved in intracellular glycolysis as well as implicated as an extracellular effector of normal cell functions including survival of certain populations of neurons. This molecule represents a member of the newly emerging family of intracellular enzymes whose disparate functions as extracellular mediators of cellular responses defines a new class of ecto/exoenzymes which play a role in normal cellular processes and are inappropriately utilized by tumor cells to elicit new survival strategies.

Theories on the metastatic process and possible therapeutic options

A sequence of steps are prerequisite for cancer cells before metastases are established. Metastasis has been shown to be an inefficient process limited by both random and selective events. By differentiating invasion from metastasis, sequential steps in the metastatic cascade have been defined and studied separately. This approach has yielded a variety of new potential therapeutic strategies. However, increasing knowledge of the mechanisms relating to metastasis has also revealed the complexity of each step. In spite of difficulties in translating results obtained in preclinical models into the clinical setting, continued development of such model systems and further research into the genetic control of metastatic dissemination will lead to improved strategies for prevention of metastasis formation and for treatment of metastatic tumor cells.

Organelle rearrangement and cell volume changes during squeezing invasion of peritoneal elastic lamina by targeted murine breast carcinoma cells

Murine breast cancer cell lines were developed to selectively invade the peritoneum while they proliferated in ascites form in the abdominal cavity. In a dominant form of invasion, tumor cells showed special affinity for elastin fibers and squeezed through narrow gaps in the elastic fiber meshwork of the stroma. Even in fixed tissue, such cells could be recognized as being in the process of invasive migration because of their dumbbell shape. This appearance was similar to that of diapedetic blood cells traversing bone marrow sinus endothelium. Three-dimensional STERECON graphics reconstruction from serial thick sections of 44 such cells was carried out. The reconstructions showed that, in mid-penetration, the cells spread extensively over the exterior surface of the elastic fiber meshwork. The cell surface contact of these forward projections was mainly with the elastic fiber outer coat of microfibrils, but small areas of the cell surface also fused directly to inner-core elastin. The morphological rearrangement of the cytoskeleton was minimal in both types of attachment areas. The location of these forward facing attachments is consistent with mechanisms for pulling the invasive cell through the gap. Lamellopodia formation and clustering of cytoplasmic organelles occurred more commonly at the forward-facing part of the cell. Morphometry of the reconstructions showed that a contraction of the whole cell occurred during the squeezing/migration process suggestive of an additional pushing process. However, our invasive cell lines showed marked differences in the degree of cell shrinkage. The process of adhesion and squeezing of tumor cells through elastin meshworks in vivo is clearly a complex phenomenon. Changes in cell surface activity appear to play a significant role in establishing the necessary 'foothold' component of invasion and, possibly, in the generation of tractive force as well.

Ultrastructure of invasion in different tissue types by Lewis lung tumour variants

Ultrastructural studies on the interactions of low and highly metastatic 3LL tumour lines with the basement membranes (BMs) of capillaries, veins, muscles, nerves and adipose tissue were performed by injecting tumour cells into the foot pad of mice. Haematogenous dissemination is the principle route of metastasis formation. Cells from the highly metastatic line were able to penetrate the blood vessels more efficiently than those from the low metastatic line. This difference was mainly due to a more pronounced diapedesis-like activity of the 3LL-HH cells, and partly to the altered intratumour vessel architecture in the highly metastatic tumour line. There was no difference between the two lines in the ultrastructure and frequency of invasion of nerves and adipose tissue BMs. However, in the highly metastatic line an extremely efficient penetration of muscle cell BM was observed. These results provide further evidence that the interaction of tumour cells with the BMs of different tissue types is one of the main determinants in local and distant dissemination.

Endothelial attachment and plasmalemmal apposition in the transcellular movement of intravascular leukemic cells entering the myeloid parenchyma

The plasmalemmal relationship between metastases-forming leukemia cells and myeloid sinus endothelium during the transmural passage of the leukemia cells has been studied in rat bone marrow. After the myeloid vascular system was freed from normal circulating blood cells, the bone marrow was perfused with a suspension of leukemia cells derived from an ascites tumor. The bone marrow was then fixed by perfusion with double aldehyde with and without the addition of tannic acid. Leukemia cells were seen adhering to the adluminal aspect of the sinus endothelium and in all stages of endothelial penetration. The penetration of the sinus wall was independent of endothelial junctions; i.e., the transmural passage into the myeloid parenchyma was transcellular. At these sites, there were restricted areas of close plasmalemmal appositions of the two cell types where the intraplasmalemmal space was reduced to 2.3 nm. This space was interrupted by electron densities of 5 nm diameter and spaced 9 nm center to center. These close plasmalemmal appositions extended over distances ranging from 150 nm to 200 nm. It is suggested on the basis of the structural similarity that these heptalaminar complexes of close plasmalemmal apposition represent the structural equivalent of gap junctions and may be sites of intercellular communication requisite for transmural passage. When tannic acid was added to the fixative, there were extended areas of apparent fusion of the plasmalemmas of the two cell types, at the sites both of adhesion and of endothelial penetration. This fusion was limited to the outer leaflets of the two plasmalemmas, resulting in a single pentalaminar complex. These pentalaminar complexes extended over decidedly longer distances than the presumed gap junctions seen in the nontannic-acid-fixed material. The tannic acid material did not show the heptalaminar gap junction type of plasmalemmal apposition. It is believed likely that the tannic-acid-induced pentalaminar complexes may incorporate the smaller heptalaminar ones. The factors underlying the plasmalemmal configurational differences between the tannic acid and non-tannic-acid material remain undetermined.

What's new in the ultrastructure of tumor invasion in vivo?

An important point emerging from the literature on tumor invasion in vivo is the great variability of nearly all aspects studied. It seems that there is neither one particular morphologic change which renders a cell invasive, nor one particular mechanism by which a cell crosses the boundaries of its original tissue compartment to occupy another. Nevertheless, some general trends are demonstrable. The majority of invasive tumor cells appear to be characterized by prominent surface protrusions, decreased junctional contacts and, in the case of epithelium-derived tumor cells, an incomplete basement membrane. The fact that some tumors can invade foreign tissues without loosing their basement membrane is emphasized. Invasive cells frequently form organized associations with preexistent non-neoplastic cells without damaging them. Apparently, the eventual disappearance of the preexistent cells in most invaded tissues is not necessarily due to a direct action on the part of the tumor cells. It rather seems a secondary phenomenon caused by, e.g., the insertion of invasive tumor cells between the preexistent cells and their original stroma. Very often, this seems to be due to the affinity of malignant cells for basement membranes. In addition, the adhesion of tumor cells to basement membranes frequently seems to determine their pattern of spread through a tissue. A process which may turn out to be a key factor in tumor invasion is desmoplasia, the series of host reactions which creates a new environment for the tumor cells which may favor their survival, proliferation, and locomotion. With the rapid development of new techniques, electron microscopy will probably contribute to the elucidation of the exact nature, the degree of similarity to granulation tissue, and the influence on invasion of desmoplastic tumor stroma.

Tumor dedifferentiation: an important step in tumor invasion

Tumor invasion in vivo was studied by light and electron microscopy as well as by immunofluorescence microscopy. Special regard was paid to the grade of tumor differentiation. Dimethylhydrazine-induced murine colonic carcinomas comprising a differentiated and an undifferentiated tumor type with low and high invasiveness respectively, were used. At the invasion front of both tumor types a striking dissociation of the organized tumor cell complexes into isolated tumor cells was found together with a loss of most of the cytological features of differentiation. It is supposed that this process mobilizes the tumor cells from the main tumor bulk enabling them to invade the host tissue by active locomotion. This view is strongly supported by the demonstration of morphological equivalents of active cell movement such as pseudopodia-like cytoplasmic extrusions, adaptive changes of the cell shape and microfilament bundles. Although the proposed mechanism of tumor invasion is essentially the same in both tumor types, the grade of differentiation is nevertheless critical, as in the undifferentiated carcinomas only subtle dedifferentiation steps (loss of basement membrane and cell junctions) are necessary to acquire an invasive status. This fact may explain the comparatively high invasiveness and poor prognosis of undifferentiated carcinomas.

Mechanisms of tumor invasion: evidence from in vivo observations

The major mechanisms of tumor invasion in vivo are discussed in the present review. A special emphasis is placed on tumor dedifferentiation which has proved to be of paramount importance for the invasion process. Based on in vivo observations obtained from various human and animal tumors a concept for the mechanism of tumor invasion is proposed which mainly comprises the following basic events: the first and essential step in tumor invasion is the tumor dedifferentiation and dissociation at the invasion front. This apparently temporary and reversible process mobilizes the tumor cells out of the main tumor bulk and enables them to invade the host tissue by active locomotion. This mechanism is essentially supported by an interstitial edema in the host tissue adjacent to the tumor periphery, which causes an 'opening and widening' of the host intercellular spaces. Enzymatic changes in the micromilieu of the extracellular matrix may contribute to this process. The tumor cell proliferation completes the invasion process in so far, as the invading tumor cells are still able to proliferate, leading this way to expanding tumor cell nests in the host tissue which have the potency to redifferentiate. The expansive growth of these tumor cell nests results in a progressive atrophy of the host tissue, mainly caused by an increasing compression and a competitive withdrawal of oxygen and other nutrients by the tumor cells. The overall picture of tumor invasion can therefore be considered as a repetitive cycle of active tumor cell locomotion followed by focal tumor cell proliferation in the host tissue.