Senescence of aortic endothelial cells in culture: effects of basic fibroblast growth factor expression on cell phenotype, migration, and proliferation.

Bovine aortic endothelial cells (BAEC) can be isolated in large numbers without major contamination by other cells and maintained in culture with a limited life span for about 100 population doublings. In order to study phenotypic changes of BAEC during long-term culture, stocks of different passages of BAEC were established and their morphological, migratory, and proliferative properties analyzed. Early-passage BAEC (passages 5-15) rapidly produce dense, cobblestone-like monolayers. Their growth beyond the monolayer configuration is characterized by the formation of an irregular network of spindle-shaped, crisscrossing BAEC growing either on top or beneath the monolayer, and by the assembly of elongated BAEC into well-differentiated capillary-like tubes. In contrast, senescent BAEC (passages 35-45) form perfect cobblestone monolayers that contain several, often multinucleated giant cells and a few capillary-like tubes but not the crisscrossing networks of their early-passage counterparts. The rates of BAEC migration and proliferation gradually decline during in vitro senescence. This decline is neutralized by exogenous basic fibroblast growth factor (bFGF) which elevates the migratory and proliferative capacities of early-passage and senescent BAEC to uniformly high levels. Northern blot analysis shows a gradual decline in bFGF message and an increase in laminin message during in vitro BAEC senescence. The present study supports the concept of autocrine growth regulation of BAEC and associates a decreased bFGF message with decreased rates of migration and proliferation as well as loss of the crisscrossing BAEC morphotype in senescent cultures.

Lung endothelial dipeptidyl peptidase IV is an adhesion molecule for lung-metastatic rat breast and prostate carcinoma cells.

Attachment of circulating tumor cells to endothelial cell adhesion molecules restricted to select vascular compartments is thought to be responsible for site-specific metastasis. Lung-metastatic rat R3230AC-MET breast and RPC-2 prostate carcinoma cells bound outside-out endothelial cell membrane vesicles, prepared by perfusion of the rat lung vasculature with a low-strength formaldehyde solution, in significantly higher numbers than their nonmetastatic counterparts R3230AC-LR and RPC-LR. In contrast, vesicles derived from the vasculature of a nonmetastasized organ (e.g., hind leg muscle) showed no binding preference for either of the four tumor cell lines. Lung-derived endothelial vesicles were used here to generate mAbs against lung endothelial cell adhesion molecules. The first group of mice were actively immunized against lung endothelial vesicles, whereas the second group was injected with syngeneic mouse antiserum against leg endothelial vesicles before active immunization with lung endothelial vesicles. 17 hybridoma supernatants obtained from the two fusions bound lung vesicles with at least a 10-fold higher affinity than leg vesicles. Seven (four obtained by a passive/active immunization protocol) stained rat capillary endothelia. One mAb, mAb 8.6A3, inhibited specific adhesion of lung-derived vesicles to lung-metastatic breast and prostate carcinoma cells. Purification of the antigen (endothelial cell adhesion molecule) from rat lung extracts revealed a protein with a 110-kD mol wt. NH2-terminal sequencing established identity with dipeptidyl peptidase IV which had been reported to serve as a fibronectin-binding protein. These results indicate that vesicles obtained from in situ perfused organs are a convenient immunogen for the production of antibodies to compartment-specific endothelial cell surface molecules, and reinforce the concept that endothelial cell surface components are selectively recognized by circulating cancer cells during metastasis formation.

Migrating endothelial cells are distinctly hyperglycosylated and express specific migration-associated cell surface glycoproteins.

Migration of endothelial cells is one of the first cellular responses in the cascade of events that leads to re-endothelialization of an injured vessel and neovascularization of growing tissues and tumors. To examine the hypothesis that endothelial cells express a specific migration-associated phenotype, we analyzed the cell surface glycoprotein expression of migrating bovine aortic endothelial cell (BAECs). Light microscopic analysis revealed an upregulation of binding sites for the lectins Concanavalin A (Con A), wheat germ agglutinin (WGA), and peanut agglutinin after neuraminidase treatment (N-PNA) on migrating endothelial cells relative to contact-inhibited cells. These findings were confirmed and quantitated with an enzyme-linked lectin assay (ELLA) of circularly scraped BAEC monolayers. The expression of migration-associated cell surface glycoproteins was also analyzed by SDS-PAGE. The overall expression of cell surface glycoproteins was upregulated on migrating BAECs. Migrating BAECs expressed Con A- and WGA-binding glycoproteins with apparent molecular masses of 25 and 48 kD that were not expressed by contact-inhibited BAEC monolayers and, accordingly, disappeared as circularly scraped monolayers reached confluence. Subconfluent BAEC monolayers expressed the same cell surface glycoconjugate pattern as migrating endothelial cells. FACS analysis of circularly scraped BAEC monolayers showed that the phenotypic changes of cell surface glycoprotein expression after release from growth arrest occurred before the recruitment of the cells into the cell cycle (3 vs. 12 h). Suramin, which inhibits endothelial cell migration, abrogated the expression of the migration-associated phenotype and induced the expression of a prominent 28-kD Con A- and WGA-binding cell surface glycoprotein. These results indicate that endothelial cells express a specific migration-associated phenotype, which is characterized by the upregulation of distinct cellular glycoconjugates and the expression of specific migration-associated cell surface glycoproteins.

Quantitative analysis of autocrine-regulated, matrix-induced, and tumor cell-stimulated endothelial cell migration using a silicon template compartmentalization technique.

Autocrine-regulated, matrix-induced, and tumor cell-stimulated endothelial cell migration was quantitatively analyzed using a two-dimensional, two-compartment coculture system. Silicon templates were used to subdivide 35-mm tissue culture dishes into two separate compartments. Endothelial cells were grown to confluence in the inner compartment and released from growth arrest by removal of the silicon template. The distance of endothelial cell outgrowth from the monolayer was measured in 24-h intervals. Endothelial cells from different vascular beds migrated with different migration rates (large vessel endothelial cells greater than hemangioendothelioma cells greater than microvessel endothelial cells). Prior coating of tissue culture wells with fibronectin, type I collagen, or type IV collagen and increasing serum concentrations strongly enhanced endothelial cell migration. Seeding tumor cells into the outer compartment prior to removal of the silicon template permitted the direct coculture analysis of tumor cell-induced endothelial cell migration. Microvascular endothelial cell migration was stimulated in a tumor cell number-dependent fashion, whereas large vessel endothelial cells could not consistently be stimulated by coculture with tumor cells. It is concluded that silicon templates offer a useful approach for the quantitative study of migration of anchorage-dependent cells, permitting follow-up measurements over several days, the study of matrix effects, and the direct coculture analysis of cell migration.

Differentiation-dependent expression of lectin binding sites on normal and neoplastic keratinocytes in vivo and in vitro.

We used lectins as probes to demonstrate the composition of membrane carbohydrates of canine keratinocytes in various functional stages and various degrees of differentiation. Keratinocytes during normal epidermal turnover were compared by lectin immunohistochemistry to cells of hyperplastic epidermis and neoplastic keratinocytes. Three types of epidermal tumors and oral squamous cell carcinomas were examined. In addition, two in vitro tissue culture systems for keratinocytes were studied and compared with in vivo epithelium. In normal skin, PNA reacted only weakly with basal cells, whereas in hyperplastic skin basal cells bound this lectin strongly, demonstrating increasing expression of PNA binding sites with increasing thickness of the stratified squamous epithelium. ConA bound to basal cell tumors only. In oral squamous cell carcinomas, the expression of distinct lectin binding sites correlated with certain histological growth patterns, e.g., UEA-I reacted with highly invasive tumors but not with tumors showing a solid growth pattern. Using cell surface iodination and polyacrylamide gel electrophoresis, distinct differences in cell membrane protein expression were demonstrated between normal and neoplastic keratinocytes. SDS-polyacrylamide gel electrophoresis of cultured normal and neoplastic keratinocytes revealed several cell surface proteins that are specific for either cell type. Neoplastic cells specifically express a 140 KD lectin binding cell surface glycoprotein. The results of this study show that lectin binding patterns of keratinocytes are dependent on the functional state and the degree of differentiation of the cells and demonstrate correlation of some histological growth patterns with distinct lectin binding phenotypes, suggesting association of expression of cell membrane carbohydrate moieties with growth patterns. In addition, close similarities between "lifted cultures" grown at the air-liquid interface and native tissue demonstrate the value of this culture system as a model for differentiated stratified squamous epithelium.

Comparison of growth and differentiation of normal and neoplastic canine keratinocyte cultures.

Neoplastic canine keratinocytes derived from a spontaneous oral squamous cell carcinoma were maintained in culture for more than 45 passages. The presence of desmosomes and keratin filaments was demonstrated by electron microscopy and immunohistochemistry. The keratinocytes were grown in two different culture conditions to induce variations in the stage of differentiation, i.e., in submerged cultures and at the air-liquid interface. For comparison, normal canine keratinocytes were grown under the same conditions. Anisocytosis was present in neoplastic cultures grown submerged in medium. Grown at the air-liquid interface, neoplastic keratinocytes differentiated into a well-organized, multilayered stratified squamous epithelium analogous to normal keratinocytes. Rare areas of irregular growth and formation of whorls were detected. Expression of lectin binding sites and specific cell surface antigens of neoplastic and normal keratinocytes demonstrated marked similarities between the two cell lines. Neoplastic cells lacked certain surface antigens that are present on normal cells. Squamous cell carcinoma cells grew faster than normal canine keratinocytes as demonstrated by growth curve evaluation. Neoplastic keratinocytes responded to growth stimulation by epidermal growth factor and cholera toxin as do normal keratinocytes. Neoplastic cells grown in medium lacking these factors proliferated faster than growth factor stimulated normal keratinocytes.

Modulation of endothelial cell surface glycoconjugate expression by organ-derived biomatrices.

Cell surface molecules play an important role in cellular communication, migration, and adherence. Here, we show the effect of organ-derived biomatrices on endothelial cell surface glycosylation. Five different lectins (with and without neuraminidase treatment) have been used as probes in an enzyme-linked lectin assay to quantitatively detect glycoconjugates on endothelial cells (BAEC) grown on tissue culture plastic or biomatrices isolated from bovine lung, liver, and kidney. BAEC generally exhibit strong binding of concanavalin A (Con A), Ricinus communis agglutinin I (RCA-I), wheat germ agglutinin (WGA), and soybean agglutinin, and peanut agglutinin after neuraminidase pretreatment of cells (Neu-SBA and Neu-PNA), while SBA and PNA consistently bind weakly to BAEC. BAEC grown on organ-derived biomatrices exhibit significantly altered binding intensities of Con A, RCA-I, WGA, and Neu-PNA: BAEC cultured on lung- or kidney-derived biomatrices express significantly stronger binding affinities for Con A and RCA-I than BAEC grown on liver-derived biomatrix or tissue culture plastic. In contrast, BAEC binding of WGA and PNA (after treatment of cells with neuraminidase) is significantly reduced when BAEC are grown on liver- or kidney-derived biomatrix. Quantitative lectin immunogold electron microscopy reveals consistently stronger lectin binding over nuclear regions compared to junctional regions between neighboring cells. These results indicate that extracellular matrix components regulate endothelial cell surface glycoconjugate expression, which determines cellular functions, e.g., preferential adhesion of lymphocytes or metastatic tumor cells.

Endothelial cell membrane vesicles in the study of organ preference of metastasis.

Many malignancies exhibit distinct patterns of metastasis that appear to be mediated by receptor/ligand-like interactions between tumor cells and organ-specific vascular endothelium. In order to study endothelial cell surface molecules involved in the binding of metastatic cells, we developed a perfusion method to isolate outside-out membrane vesicles from the lumenal surface of rat lung microvascular endothelium. Lungs were perfused in situ for 4 h at 37 degrees C with a solution of 100 mM formaldehyde, 2 mM dithiothreitol in phosphate-buffered saline to induce endothelial cell vesiculation. Radioiodinated rat lung endothelial cell membrane vesicles bound lung-metastatic tumor cells (B16F10, R323OAC-MET) in significantly higher numbers than their low or nonmetastatic counterparts (B16F0, R323OAC-LR). In contrast, leg endothelial membrane vesicle showed no binding preference for either cell line. Neuraminidase treatment of vesicles abolished specificity of adhesion of lung-derived vesicles to lung metastatic tumor cells. These results demonstrate that in situ perfusion is an appropriate technique to obtain pure endothelial cell membrane vesicles containing functionally active adhesion molecules. The preferential binding of lung-derived endothelial cell membrane vesicles by lung metastatic tumor cells is evidence of the importance of endothelial cell adhesion molecules in the formation of metastases.

Organ-preference of metastasis. The role of endothelial cell adhesion molecules.

The initial, site-specific colonization of secondary organs by blood-borne cancer cells appears to be mediated by endothelial cell adhesion molecules. These molecules are part of the organ-specific microvascular phenotype and are regulated through complex interactions of the endothelium with the extracellular matrix (e.g., distinct matrix macromolecules and growth factors). They are induced in vitro by growing 'unspecific' (large vessel) endothelial cells on extracts of organ-specific biomatrices. In many respects, these molecules are similar to the various classes of chemically different adhesion molecules that regulate lymphocyte traffic, but are believed to be distinct from the inducible adhesion molecules that govern leukocyte adhesion during acute episodes of inflammation. Biochemical and biophysical data indicate that preference of tumor cell adhesion to organ-specific microvascular endothelium may not require qualitative differences of such homing receptors between endothelia, but may be explained on the basis of quantitative receptor differences as well as differences of receptor avidity. Following adhesion, the metastatic cascade proceeds by the establishment of metabolic conduits between the endothelium and adherent tumor cells. This heterotypic coupling represents an early step in the extravasation of cancer cells from the microvasculature, initiating endothelial cell retraction from its basement membrane and recanalization around the arrested tumor cell. These events, together with local growth promoting effects exerted by the metastasized organ, are believed to provide the basis for Paget's 'seed and soil' hypothesis of metastasis.

Phenotypic characterization of normal and neoplastic canine endothelial cells by lectin histochemistry.

Cell surface glycoconjugate expression of endothelial cells in canine cutaneous hemangiomas and hemangiosarcomas was compared to normal cutaneous endothelial cells using eight different lectins (with and without neuraminidase pretreatment) in an indirect immunoperoxidase technique. Direct comparison of lectin binding pattern of neoplastic endothelial cells with adjacent normal endothelial cells revealed minor changes in the binding intensity of several lectins (enhanced: Wheat germ agglutinin [WGA]; reduced: Griffonia simplicifolia-I [GS-I], Ricinus communis agglutinin-I [RCA-I], Soybean agglutinin after neuraminidase pretreatment [Neu-SBA], and Wheat germ agglutinin after neuraminidase treatment [Neu-WGA]). Neoplastic endothelial cells in some tumors exhibited varying binding of Ulex europaeus agglutinin-I (UEA-I; not binding to normal canine endothelial cells) and no Soybean agglutinin (SBA) binding (variably binding to normal endothelial cells in small cutaneous vessels). Lectin binding of neoplastic cells was rather heterogenous within one tumor compared to the uniform binding pattern of normal endothelial cells. These lectin binding studies demonstrate the phenotypic heterogeneity of neoplastic endothelial cells, indicating changes of cell surface glycosylation during neoplastic transformation.

Biochemical and metabolic properties of bovine type II pneumocytes in primary culture.

Bovine type II pneumocytes were isolated from lungs of adult cattle by enzymatic tissue dissociation and subsequent purification by density gradient centrifugation on a 1.040 g/ml Percoll gradient. By quantitative and qualitative high-performance liquid chromatographic (HPLC) analysis of freshly isolated, highly purified bovine type II pneumocytes, six phospholipid components (phosphatidylcholine, phosphatidylglycerol, phosphatidyl-ethanolamine, phosphatidylinositol, phosphatidylserine, and sphingomyelin) were detected regularly and their composition was compared to that of whole lung homogenates. Total phospholipid content of 10(6) bovine type II cells was 34.7 micrograms (quantitative HPLC). Suspended and adherent bovine type II cells incorporated [14C]choline into their phospholipids. The rate of synthesis of suspended cells (289 pmol/h x 10(6) cells) was constant for at least 4 h, while adherent type II cells incorporated [14C]choline with a constant rate of 160 pmol/h x 10(6) cells for at least 20 h. These data further establish the use of cultured bovine type II pneumocytes as a high-purity model for the in vitro investigation of the surfactant system.

Isolation of bovine type II pneumocytes in high yield and purity.

A method has been developed for the isolation of bovine type II pneumocytes by enzymatic tissue dissociation and subsequent density gradient centrifugation. After mechanical defibrination, the crude cell suspension contains 71.9 +/- 26.6 X 10(7) cells (viability greater than 95%) with a type II cell purity of 39.4 +/- 10.9%. Due to their low buoyant density, bovine type II pneumocytes can be purified in a single step to 95.7 +/- 1.7% by discontinuous density gradient centrifugation on a 1.040 g/ml Percoll gradient. Isolated cells are identified by light and fluorescence microscopy that show their characteristic intracytoplasmatic surfactant granules. The fine structure of the surfactant lamellar bodies is examined in ultra thin sectioned and freeze-fractured type II pneumocytes. Isolation of bovine type II pneumocytes for the in vitro study of the surfactant system offers an alternative to the use of laboratory animals and provides an ideal system for the isolation of any desired cell number from one animal by simply increasing the size of the lung segment to be trypsinized.