Organ-cultured epithelial tissue as an in vitro model for invasion: quantitation and high-voltage electron microscopy of tumor cell attachment.

An invasion model designed specifically for studying mechanisms of invasion of squamous-cell carcinomas was developed with murine buccal mucosa as the host tissue. The mucosal explants were destratified by growth in low-calcium medium (less than 0.07 mM), which results in a dorsal surface composed of one or two layers of basal epithelial cells. The explant has a three-dimensional histoarchitecture similar to in vivo mucosa. A spontaneously transformed epithelial cell line (Pam 27; Yuspa , S. H., Hawley -Nelson, P., Koehler , B., and Stanley, J. R. Cancer Res., 40: 4694-4703, 1980) was used to seed explants. Attachment and subsequent growth and invasion were monitored. The morphology of attachment was examined by conventional and high-voltage electron microscopy. In addition, attachment was quantitated by using [125I]iododeoxyuridine-labeled tumor cells. Attachment was shown to be an active process which involves an interdigitation of tumor-host cell processes. Junctional complexes were also observed between tumor and host epithelial cells. By 24 hr, tumor cells were spread on the basal cells and were in the process of replacing host cells. Long-term growth of explants showed that tumor cells can repopulate the epithelial surface and invade the stromal region.

"Dark cells" in normal, hyperplastic, and promoter-treated mouse epidermis studied by conventional and high-voltage electron microscopy.

Dark cells (DC) could be reproducibly demonstrated by differential toluidine-blue staining and electron microscopy (EM) of NYLR/Nya 16- to 19-day embryo and new born skin and phorbol ester-treated or untreated young adult skin. High-voltage electron microscopy on the same or adjacent sections showed that toluidine-blue staining picks out some but not all the DC seen by EM. The ultrastructure of DC was similar in all the above situations, except that phorbol ester-induced DC showed a less contracted nucleus. No support was obtained for DC as stem cells either for basal-cell hyperplasia or for development of hair follicle or gland outgrowths. Most of the severely contracted DC (Types 3 and 4) were assumed to have undergone an apoptotic type of cell death. Two phenomena that may have caused the contraction and apoptosis were observed. Formation of a "contraction vacuole" adjacent to the DC probably led to a loss of intercellular communication. An apparent necrosis of dermal capillaries in areas of abundant follicle downgrowth probably produced local anoxia. Further characterization of DC requires a search for cytochemical or immunologic markers, analysis of intracellular calcium and other elements, and the cloning of subpopulations of basal cells that can be selectively induced to form DC.

Inflammation with restricted lysosomal proteolysis during early ascites carcinoma invasion of mouse parietal peritoneum. A medium and high-voltage electron microscopic and cytochemical study.

A carcinoma invasion system (Krebs-2 and Ehrlich tetraploid ascites tumors invading mouse peritoneum) was studied by high-voltage electron microscope (HVEM) stereoscopy, conventional (medium voltage) electron microscopy (MVEM), and cytochemistry. Tumor cells entered areas of peritoneum (mainly parietal) only where mesothelial cells were damaged and where there was inflammation of the underlying stroma. The initial invasion was different from that of most other invading carcinomas in that there was minimal breakdown of basal lamina and collagen. Neither tumor cells, inflammatory leukocytes nor peritoneal fibroblasts showed significant secondary lysosome production or release of intracellular or extracellular acid phosphatase. Morphological and cytochemical criteria suggest that in some invading carcinomas, as with non-tumor migrating cells such as leukocytes, widespread proteolysis due to diffusion of proteases is not a prerequisite for invasion of stromal connective tissue.

Ascites tumor invasion of mouse peritoneum studied by high-voltage electron microscope stereoscopy.

Interaction of Krebs-2 and Ehrlich tetraploid cells with NYLR/Nya mouse peritoneum mesothelium and penetration of basal lamina and elastic reticulum were studied. Invasion of abdominal viscera was rare. Invading cells had a shrunken nucleus and cytoplasm like the "dark cells" of hyperplastic epithelia. High-voltage electron microscope stereoscopy showed that invasive cells pass through small holes in the elastic reticulum by adherence to the reticulum and by constriction of the cells. High voltage electron microscopy stereoscopy of collagen fibers near tumor cells indicated that fragmentation and loss of collagen is minimal. Rapid progression by ascites transfer appears to produce anchorage-independent cells adapted to ascites fluid growth, but new selection steps must be adopted to concentrate strongly invasive subpopulations.