Collagen synthesis and deposition during mammary epithelial cell spreading on collagen gels

Films of collagen crosslinked by S-S bonds: preparation and characterization

Collagen (type I from calf-skin) was chemically modified by 4-butyrothiolactone to obtain the mercapto group-bearing collagen (collagenSH), which possessed SH groups in 8-19 M% of a total amino acid residues. The triple helical strands of the collagen was not completely perturbed to exhibit the rotary dispersion [theta]221, which was as much as 70% of an original intensity of the collagen. In the presence of the oxygen dissolved in water, the collagenSH was cross-linked by disulfide bonds to afford the collagenSS. The collagenSS could preserve about 15 and 40% of an original helix structure at 55-70 degrees C and in 2mm sodium dodecyl sulfate, respectively. The film made of the collagenSS exhibited the tensile strength as high as 36 MPa and was insoluble in most organic solvents including water. The collagenSS film was more resistant to degradation by collagenase (type I) than the collagen film. The collagenSS film supported the growth of L929 fibroblast cell in a manner similar to a collagen film.

Growth of L929 cells on polymeric films prepared by Langmuir-Blodgett and casting methods

The growth and spreading of fibroblast, L929 cells, on various polymeric films prepared by the Langmuir-Blodgett (LB) and casting methods were investigated. L929 cells, which were cultivated on collagen and synthetic polymeric films prepared by the LB method, adhered and spread much more than those on synthetic films prepared by the casting method. This is explained by the fact that cell growth and cell spreading are suitable for L929 cells on the films having serum proteins that contain a high alpha-helix content, because LB films adsorbed those serum proteins estimated from the circular dichroism measurements of the films immersed in cell culture medium. An exponential relationship was observed from the plot of the cell density vs root mean square of roughness of the films, which is estimated by atomic force microscopy, whereas a linear relationship was observed from the plot of the spreading ratio vs the root mean square of roughness. It is suggested that the correlation between the cell growth or spreading ratio and surface roughness of the films where L929 cells were cultivated is considered to be more important than the correlation between the cell growth or spreading ratio and the contact angle of the films.

Degradable and highly porous polyesterurethane foam as biomaterial: effects and phagocytosis of degradation products in osteoblasts

Recently, a new class of biodegradable PHB-based polyesterurethane (DegraPol/btc) has been prepared and found to exhibit favorable cell and tissue compatibility. The present study has been designed to evaluate the response of primary isolated rat tibia osteoblasts to small crystalline particles of short-chain poly[(R)-3-hydroxybutyric acid] (PHB-P diameter: 2-20 microm), of fluorescent-labeled analogs (DPHP-P), and of lysine methyl ester as possible degradation products of DegraPol/btc. Observations made using confocal microscopy clearly indicate that osteoblasts have the capability of taking up PHB-P particles. Although in single-cell analysis the number of DPHB-P-positive osteoblasts gradually increased up to 16 days, the fluorescence intensity per osteoblast increased only during the first 4 h after DPHB-P incubation, and then it retained the 4 h level up to 16 days. No significant change in the production levels of collagen type I and osteocalcin was detectable after treatment with low concentrations of PHB-P for up to 32 days. In contrast, a time- and dose-dependent alteration of the alkaline phosphatase (ALP) activity was found. Maximal activity was measured after 4 days of treatment with 2 microg of PHB-P/mL (170% of control cells). Rat peritoneal macrophages co-cultured with osteoblasts in a transwell culture system mimicked the observed PHB-P induced ALP elevation. Therefore, the PHB-P-induced ALP increase could be the result of direct or indirect stimulation of osteoblasts, possibly via soluble factors produced by contaminating osteoclasts. Taken collectively, the data demonstrate that osteoblasts are capable of phagocytosing PHB-P and that this process is accompanied at low PHB-P concentrations by dose- and time-dependent alteration of alkaline phosphatase activity but not of collagen type I or osteocalcin.

Human bronchial epithelial cells can contract type I collagen gels

Fibroblasts can contract collagen gels, a process thought to be related to tissue remodeling. Because epithelial cells are also involved in repair responses, we postulated that human bronchial epithelial cells (HBECs) could cause contraction of collagen gels. To evaluate this, HBECs were plated on the top of native type I collagen gels and were incubated for 48 h. After this, the gels were released and floated in LHC-9-RPMI 1640 for varying times, and gel size was measured with an image analyzer. HBECs caused a marked contraction of the gels within 24 h; the area was reduced by 88 +/- 4% (P < 0.01). The degree of gel contraction was dependent on cell density; 12,500 cells/cm2 resulted in maximal contraction, and half-maximal contraction occurred at 7,500 cells/cm2. Contraction varied inversely with the collagen concentration (91 +/- 1% with 0.5 mg/ml collagen vs. 43 +/- 5% with 1.5 mg/ml collagen). In contrast to fibroblasts that contract gels most efficiently when cast into the gel, HBEC-mediated contraction was significantly (P < 0.01) more efficient when cells were on top of the gels rather than when cast into the gels. Anti-beta 1-integrin antibody blocked HBEC-mediated contraction by > 50%, whereas anti-alpha 2-, anti-alpha 3-, anti-alpha v-, anti-alpha v beta 5-, anti-beta 2-, or anti-beta 4-integrin antibody was without effect. The combination of anti-beta 1-integrin antibody and an anti-alpha-subfamily antibody completely blocked gel contraction induced by HBECs. In contrast, anti-cellular fibronectin antibody did not block HBEC-induced gel contraction, whereas it did block fibroblast-mediated gel contraction. In summary, human airway epithelial cells can contract type I collagen gels, a process that appears to require integrins but may not require fibronectin. This process may contribute to airway remodeling.

Interactions of osteoblasts and macrophages with biodegradable and highly porous polyesterurethane foam and its degradation products

The macrophage cell line J774, primary rat osteoblasts, and the osteoblast cell line MC3T3-E1 were used to examine the biocompatibility of a newly developed polyesterurethane foam and the possible use of this structure as bone-repair materials. The newly developed, biodegradable, and highly porous (pore size 100-150 microns) DegraPol/btc polyesterurethane foam was found to exhibit good cell compatibility; the cell-to-substrate interactions induced neither cytotoxic effects nor activation of macrophages. Osteoblasts and macrophages exhibited normal cell morphology. No signs of cell damage were detected using scanning electron microscopy (SEM). No significant increase in the production of tumor necrosis factor-alpha (TNF-alpha) or nitric oxide (NO) was detected in macrophages. Compared with cells cultured on tissue culture polystyrene (TCPS), macrophages exhibited relatively high cell attachment (150% of TCPS) but significantly high doubling time (about 8 days) compared with TCPS (4.6 days). Primary rat osteoblasts and the osteoblast cell line exhibited relatively high attachment (140% and 180% of TCPS, respectively) and a doubling time of about 5 days, compared with TCPS (6 days and 8.8 days, respectively). Eight days after cell seeding, osteoblasts exhibited a confluent cell multilayer and migrated into the pores of the polymer. In addition they produced high concentrations of collagen type I, the main protein of the bone, and expressed increasing alkaline phosphatase activity and osteocalcin production throughout the 12 days of the experiment. During degradation of these polymers, small crystalline particles of short-chain poly[(R)-3-hydroxybutyric acid] (M(n) approximately 2300) (PHB-P) are released. Therefore PHB-P (diameter, 2-20 microns), as possible degradation products of the polymer, are investigated here for their effects on macrophages and osteoblasts. Results obtained in the present study clearly indicate that macrophages and, to a lesser degree, osteoblasts have the ability to take up (phagocytose) PHB-P. At low concentrations particles of PHB failed to induce cytotoxic effects or to activate macrophages. Osteoblasts showed only limited PHB-P phagocytosis and no signs of cellular damage. At high concentrations of PHB-P, this process was accompanied by cytotoxic effects in macrophages (> 200 pg PHB-P/cell) and to a lesser extent in osteoblasts (> 400 pg PHB-P/cell).

Developmental regulation of collagen differential expression in the rabbit bladder

PURPOSE

It has been suggested that the amount and relative ratios of different types of collagen influence bladder compliance. To understand the mechanisms regulating collagen synthesis and degradation in the bladder we investigated the gene expression of collagen types I and III in the rabbit bladder during normal development.

MATERIALS AND METHODS

New Zealand white rabbits ages fetus to adult were used for this study. The mid portion of the bladder wall was harvested. Northern blot hybridization was performed to analyze quantitatively collagen types I and III messenger (m) ribonucleic acid (RNA), and in situ hybridization was done to localize the messages. Corresponding protein distributions were obtained by immunohistochemical staining.

RESULTS

Types I and III collagen are developmentally regulated at the level of gene expression. The temporal and spatial distribution of the mRNA transcripts of alpha 1(I) and alpha 1(III) correlates with extracellular protein deposition. Differential distribution of the mRNA transcripts is also developmentally regulated.

CONCLUSIONS

This study characterizes the relationship between collagen gene expression and normal rabbit bladder development. The expression of alpha 1(I) and alpha 1(III) alters as the bladder grows. High levels and nonconcordant up and down regulation of different types of collagen mRNA during early development demonstrate the complexity of the extracellular matrix in a young bladder. This observation may be important in our understanding of the injury response in a developing versus mature bladder.

Production of neocollagen by cells invading hydrogel sponges implanted in the rabbit cornea

BACKGROUND

Poly(2-hydroxyethyl methacrylate) sponges are artificial tissue-equivalent matrices with potential value as materials for the peripheral zone of artificial corneas. A keratoprosthetic device was developed incorporating a poly(HEMA) spongy skirt which allowed cellular invasion. The present in vivo study investigated the biosynthetic activity of stromal fibroblasts growing within a poly(HEMA) sponge implanted into the rabbit cornea.

METHODS

A porous poly(HEMA) hydrogel was synthesized by polymerization in a large excess of water. Specimens with a pore size larger than 10 microns were impregnated with collagen type I and then implanted into the limbal region of cornea in four rabbits. The animals were followed clinically for 28 days, when they were anaesthetized and new sponge specimens were implanted in their second eye. After 2 h, both eyes were enucleated. The 28-day and 2-h explants were subjected to autoradiographic analysis following labelling with tritiated proline and to an immunostaining technique using antibodies to collagen types I-VI.

RESULTS

The autoradiographic analysis showed that the fibroblasts within the 28-day explants continued to be synthetically active and deposited proteins. Using the immunostaining technique, the deposition was most clearly demonstrated by the localization of collagen type III in the tissue invading the sponge. Both techniques failed to indicate any cellular activity in the short-time implants.

CONCLUSIONS

The presence of collagen type III is consistent with a normal healing response of the stromal fibroblasts and indicates that poly(-HEMA) sponges are able to function as tissue-equivalent matrices.

Three-dimensional mammary primary culture model systems

Model systems have been developed to investigate the complex and coordinated regulation of mammary gland development and transformation. Primary cultures, using newly isolated cells or tissue, are optimal for such studies since, in comparison to immortalized cell lines, the normal signal transduction pathways are presumed to be intact. Three such models are described, including whole organ culture, mammary epithelial cell (MEC) organoids, and MEC-stromal cocultures. Studies using whole-organ culture have the advantage that the normal glandular architecture remains intact, the MEC can undergo lobuloalveolar development and express milk proteins in a hormone dependent manner, and, following hormonal withdrawal, undergo involution. Moreover, transformation of the MEC is readily accomplished. Culture of isolated MEC organoids within an EHS-derived reconstituted basement membrane permits extensive proliferation, branching end bud and alveolar morphogenesis, and accumulation of milk protein and lipid in a physiologically relevant hormone- and growth factor-dependent manner. This model can thus be utilized to investigate the mechanism by which various modulators exert their direct effects on the epithelium. Finally, in view of compelling evidence for stromal-epithelial interactions during normal mammary gland development, and potentially also during the development of malignancy, models in which MEC can be cocultured with enriched populations of stroma offer considerable potential as a tool to understand the nature and mechanisms of the interactions that occur during the various developmental states, and how such interactions may go awry during carcinogenesis.

Fibroblast growth on polymer surfaces and biosynthesis of collagen

The growth and morphology of rat fibroblasts cultured on various polymer substrates, as well as their collagen biosynthesis, were studied. A clear difference in cell growth and cell morphology was observed among the substrates. The dependence of cell growth on the water contact angle of substrate was similar to that of the adhesion. Fibroblasts could proliferate at the highest rate and showed the highest-ordered morphology when cultured on the substrate with a contact angle around 70 degrees, which was also the most favorable for cell adhesion. The amount of collagen synthesized by total cells and of adsorption of the synthesized collagen to substrates were in good correlation with the cell growth dependence on the contact angle of substrate, whereas the collagen synthesis per cell was more active on the surfaces poor for cell growth than on the good ones. Cells on surfaces promoting active collagen synthesis had a round shape and clustered upon each other. The collagen-immobilized surface had nearly the highest cell adhesion, high cell proliferation, and high collagen adsorption among the substrates studied. In addition, the highest-ordered morphology and no lag time for proliferation were observed for the collagen-immobilized surface. These results indicate that the collagen-immobilized substrate provides the most favorable surface for cell growth at the initial stage.

The influence of age on collagen and non-collagen protein production by human gingival epithelial cells

The effects of donor age on the proliferation and secretory phenotype of cultured human gingival epithelial cells were investigated. Pure cultures of epithelial cells were isolated from human gingiva of old (61-75 yr) and young (18-30 yr) adults and serially cultivated in a serum-free medium at 37 degrees C in humidified air containing 5% CO2. For each experiment, cells were seeded at 150/mm2 and the medium changed every other day. Cell number, collagen and non-collagen protein production and relative collagen synthesis (percentage collagen synthesized) were determined at days 2, 4, 6 and 8. Epithelial strains from old and young adults became confluent by day 8 and there were no differences in their rates of proliferation. Likewise there was no difference in collagen production between the two groups; however, cells from elderly individuals produced significantly less non-collagen protein. Over time the decrease in non-collagen protein production ranged from 56% below the non-collagen protein levels of epithelium from young adults at day 2 to 24% below at day 8. The reduction of non-collagen protein coupled with the unchanged secretion of collagen resulted in a statistically significant increase in relative collagen synthesis by epithelial cells from elderly individuals. These differences in non-collagen protein production and relative collagen synthesis by cultured gingival epithelium of old adults suggest a selective conversion in protein secretion.

Cellular origin of the dermal-epidermal basement membrane

The basement membrane underlying epithelium of skin is generally believed to be of epithelial origin, but a mesenchymal contribution to the basement membrane has not been directly examined. The purpose of this study was to directly evaluate both epithelial and mesenchymal contributions to the basement membrane. Fetal bovine keratinocytes cultured on the surface of collagen gels in the absence of fibroblasts did not produce an ultrastructurally recognizable basement membrane; however, when these cells were cultured in the presence of dermal fibroblasts a basement membrane at the keratinocyte-fibroblast interface was produced after 1 week which was very similar in biochemical composition and ultrastructural appearance to dermal-epidermal basement membrane in human skin. When dual species cultures of bovine keratinocytes and human fibroblasts were analyzed by indirect immunofluorescent microscopy (IF)1 with human specific antibodies against basement membrane components, dermal fibroblasts were shown to synthesize and deposit type IV collagen, type VII collagen, and laminin in a linear manner into the basement membrane zone. Fetal bovine keratinocytes cultured in the presence or absence of fibroblasts synthesized and deposited type IV collagen, type VII collagen, laminin, K-laminin, kalinin, and basement membrane associated heparan sulfate proteoglycan (HSPG) into the underlying basement membrane zone. In organ culture, a subpopulation of fibroblasts initially migrating from human foreskin explants was found to stain strongly for types VII and IV collagen and laminin by IF whereas after subculture all cells showed a uniform low staining. Based on these observations we propose that differentiated fibroblasts exist adjacent to epithelial tissues in vivo which produce basement membrane components and assist in basement membrane assembly.

Type IV collagen localization in hamster tracheal epithelial cell cultures

Hamster tracheal epithelial (HTE) cell cultures when grown on collagen coated Millicell-HA filters for 28 d in vitro have an incomplete basal lamina-like structure at the basal plasmalemma/collagen interface. EM evaluation of 42 d HTE cell cultures also revealed the presence of hemidesmosome-like structures on the basal plasmalemma. In order to better characterize this discontinuous basal lamina-like material, HTE cultures at 7, 14, 21, and 28 d were evaluated for the presence of type IV collagen, a basal lamina component. Immunocytochemical treatment of HTE cultures at these time points resulted in the presence of reaction product at the base of the cell layer. When immunocytochemical procedures for the localization of type IV collagen were carried out with normal mouse trachea, the results were also positive. Immunoblotting evaluation of HTE cell supernatants and conditioned media also indicated the presence of type IV collagen. Taken altogether, the presence of what appears to be basal lamina, hemidesmosome-like structures and immunocytochemical and immunoblot data showing the presence of type IV collagen in HTE cultures suggest that HTE cells may be producing basal lamina components but cannot organize them into a complete basal lamina.

DNA synthesis inhibition and reduced functional differentiation of midpregnant mouse mammary epithelia on collagen gels

Mammary epithelial cells were examined for a link between DNA synthesis and subsequent synthesis and secretion of casein. Cells isolated from mice midway through pregnancy and cultured on collagen gels spread to form monolayers (spreading phase). Release of monolayer/gels to float in surrounding culture medium induces synthesis and secretion of casein (secretory phase). DNA synthesis was blocked during the spreading phase with cytosine arabinofuranoside (ARA C). Culture medium was assayed for casein by direct quantification of protein from SDS-PAGE fluorographs, and by immunoblotting. When induced to become secretory, cells exposed to ARA C during the spreading phase showed a marked reduction of secretion of casein as compared to control cultures (72% reduction). In contrast, cells exposed to ARA C during the secretory phase (after monolayer formation was complete) showed no significant reduction in secretion of casein. Measurement of intracellular casein in secretory phase cells showed that reduced secretion of casein by cultures blocked during the spreading phase occurs as a consequence of reduced levels of casein synthesized, and not because of an inability to secrete intracellular accumulations. The inhibitor effect was specific; there was no significant reduction in levels of total intracellular protein synthesis, and neither cell spreading nor monolayer formation was impaired by treatments. These data support the notion that DNA synthesis is a prerequisite to functional differentiation of midpregnant mouse mammary epithelia maintained on floating collagen gels.

Extracellular matrix synthesis is required for the movement of sclerotome and neural crest cells on collagen

During early embryogenesis cells of several different populations disperse by active cell movement from one location to another. Preexisting extracellular materials are major determinants of these dispersal patterns, but the cells are also able to modify their substrata by synthesizing and secreting extracellular matrix molecules as they move. In order to determine the contribution made by these deposited materials, several tissues from the early chick embryo have been cultured in the presence of inhibitors of extracellular matrix synthesis and secretion. The tissues examined were sclerotome cells from differentiated somites and neural crest cells. For comparison, undifferentiated somites were also cultured. The movement of these cells was compared in type I collagen gel culture and in conventional culture on artificial substrata. Inhibitors of collagen synthesis were used (cis-hydroxy proline and L-azetidine-2-carboxylic acid) in addition to a proteoglycan inhibitor (p-nitrophenyl-xylopyranoside) and a secretion inhibitor (monensin). Results indicate that sclerotome cells require collagen synthesis for movement in a collagen matrix. Reversal of the effects of collagen inhibitors, by proline and type II collagen, suggest that sclerotome cells normally condition the type I matrix in order to move in it. Inhibition of proteoglycan synthesis produced the greatest effect on the movement of neural crest cells regardless of the substratum, confirming an important role for these molecules in the crest migratory routes. The attachment of all cells to collagen was highly sensitive to the presence of monensin, which is known to reduce the deposition of glycosaminoglycans and fibronectin. These results suggest that conditioning of the extracellular matrix by newly synthesized material is required for cell attachment and movement during early development.

Collagen measured in primary cultures of normal rat hepatocytes derives from lipocytes within the monolayer

The cellular origin of hepatic collagen is under active investigation. Several recent studies using cells in primary culture suggest that hepatocytes are the source of much of the collagen in normal rat liver. In view of other data indicating that lipocytes produce substantial amounts of this protein, we have reexamined collagen biosynthesis in hepatocyte cultures that have been carefully characterized with respect to the presence of lipocytes. We find that routinely prepared hepatocyte isolates contain, by number, approximately 10% lipocytes. Lipocytes in early culture are difficult to visualize by phase-contrast microscopy but after 4 d proliferate and eventually replace the parenchymal cells. The size of the lipocyte subpopulation in these cultures correlates positively with collagen production. Similarly, removal of lipocytes by further processing of the initial hepatocyte isolate significantly reduces collagen production. Moreover, the only cells within hepatocyte cultures that display type I collagen by immunohistochemistry are lipocytes. We conclude that lipocytes are the principal source of collagen in primary hepatocyte cultures. The findings indicate also that these cells are the previously described "fibroblast" that appear in relatively long-term hepatocyte cultures.

The effect of a native collagen gel substratum on the synthesis of collagen by bovine brain capillary endothelial cells

Cultured capillary endothelial cells, derived from bovine brain, and maintained on a plastic substratum synthesized predominantly interstitial collagens of which approximately 75 per cent were secreted into the medium. When grown on a native hydrated collagen type I gel, although no marked alteration in the 'collagen synthetic pattern' was observed, the overall level of collagen synthesis was increased by approximately 100 per cent. More dramatic, however, was the alteration in the distribution of these molecules between medium and cell layer. Interstitial collagens produced by cells grown on collagen gels were almost exclusively associated with the cell layer or collagenous gel. These studies, thus, demonstrate that an extracellular matrix may exert a considerable influence on the cellular synthetic activities and possibly cellular polarity of capillary endothelial cells.

Mammary gland Ca2+-binding (-dependent) proteins: identification as calelectrins and calpactin I/p36

Calcium-binding (-dependent) proteins (CBPs) associated with the spreading of mammary epithelial cell cultures have been identified as various calelectrins and calpactins (p36). In immunoblot analysis, the CBPs of 30-36 kD and 68-70 kD variously react with different calelectrin and calpactin I monomer/p36 antisera. The same immunoreactive proteins were shown to be present in virgin mammary glands and collagen gel mouse mammary epithelial cell cultures. The mammary CBPs show extensive immunochemical relatedness; however, they fail to show cross-reaction with antiserum to calpactin II (lipocortin) antiserum. These immunoreactive CBPs comigrate in electrophoresis with 35S-methionine-labeled CBPs isolated from mammary epithelial cell cultures. Unlike calmodulin, the mammary CBPs that correspond to calelectrins and calpactin I monomer/p36 are not stable to thermal denaturation. The mammary CBPs bind to epithelial cell membranes in a Ca2+-dependent manner and are differentially released from ruptured cells, compared with calmodulin, suggesting subcellular localization. Phenothiazine-agarose and phenylagarose are equivalent in their ability to bind the mammary CBPs. Thus, mammary gland CBPs of 30-36 kD and 68-70 kD have been shown to be related or equivalent to the calelectrins and to calpactin I monomer/p36. Since these proteins are known to bind Ca2+, we conclude that the mammary gland CBPs are also Ca2+-binding proteins. The mammary gland CBPs are immunologically related and probably represent members of a larger family of related proteins.

Collagen-binding proteins of rat mammary tumor epithelial cells: a biochemical and immunological study

Collagen-binding proteins were studied in mammary epithelial cells of 7,12-dimethylbenz[a]anthracene-induced rat mammary tumors. These proteins can be solubilized from cell membranes with 0.1% Triton. Using affinity chromatography on type I collagen-Sepharose and polyacrylamide slab gel electrophoresis, three major proteins of 34,000, 36,000, and 38,000 Da were found. Similar proteins were also present in several other cell types, including both epithelial and mesenchymal cells. Pulse-chase experiments did not indicate a precursor-product relationship of these proteins. Tryptic/chymotryptic peptide maps, however, revealed that the 36,000- and 38,000-Da proteins are very similar but are quite different from the 34,000-Da molecular form. The distribution and function of these proteins were then analyzed by using polyclonal antibodies directed against the entire set of major proteins. In immunofluorescence studies we observed a dense, punctate distribution of fluorescence on the cell surface of isolated and unfixed epithelial organoids and a bright pericellular staining in cultures after fixation. Treatment with the antiserum did not affect attachment and spreading of cuboidal mammary cells to plastic or to a collagen substratum. However, when the antiserum was added to the medium of growing cuboidal cells, it caused the formation of duct-like structures. These studies indicate that collagen-binding proteins may play a role in mammary gland morphology.