Functional human corneal equivalents constructed from cell lines.

Human corneal equivalents comprising the three main layers of the cornea (epithelium, stroma, and endothelium) were constructed. Each cellular layer was fabricated from immortalized human corneal cells that were screened for use on the basis of morphological, biochemical, and electrophysiological similarity to their natural counterparts. The resulting corneal equivalents mimicked human corneas in key physical and physiological functions, including morphology, biochemical marker expression, transparency, ion and fluid transport, and gene expression. Morphological and functional equivalents to human corneas that can be produced in vitro have immediate applications in toxicity and drug efficacy testing, and form the basis for future development of implantable tissues.

Promotion of angiogenesis in tissue engineering: developing multicellular matrices with multiple capacities.

One of the aims of tissue engineering is to be able to develop multi-tissue organs in the future. This requires the optimization of conditions for the differentiation of multiple cell types and maintenance of the differentiated phenotype within complex engineered tissues. The goal of this study was to develop prototype tissue engineered matrices to support the simultaneous growth of different cell types with a particular focus on the angiogenic process. We examined two different matrix compositions for the promotion of blood vessel and tube formation. A fibrin-based matrix with the addition of a combination of growth factors supported vascular growth and the invasion of inflammatory cells. Using this fibrin matrix, in combination with a collagen-based hydrogel, a simple in vitro model of the cornea with adjacent sclera was developed that was complete with innervation and vascular structures. In addition, we showed that collagen-based matrices were effective in delivering mononuclear endothelial progenitor cells to ischemic tissue in vivo, and allowing these cells to incorporate into vascular structures. It is anticipated that with further development, these matrices have potential for use as delivery matrices for cell transplantation and for in vitro study purposes of multiple cell types.

Three-dimensional type I collagen gel system for the study of osteoblastic metastases produced by metastatic prostate cancer.

We developed a three-dimensional type I collagen gel cell culture system that allows coculturing of human MG-63 osteoblast-like cells and various human cancer cells. Inoculation of human PC-3 metastatic prostate cancer cells into this type I collagen gel containing human MG-63 osteoblast-like cells produced an osteoblastic-like reaction that presented as an increased number of MG-63 cells and increased density of type I collagen around MG-63 cells adjacent to inoculated PC-3 cells by microscope analysis. Under identical experimental conditions, inoculation of cell-free medium, human KLE endometrial adenocarcinoma cells, and Calu-1 lung cancer cells did not produce this blastic-like reaction. In situ hybridization documented the uniform expression of insulin-like growth factor I (IGF-I) and of urokinase-type plasminogen activator (uPA) mRNA in MG-63 and PC-3 cells separately cultured in this substrata. The uniform expression of uPA was also documented by immunocytochemistry using a monoclonal and a polyclonal antihuman uPA antibody. The relative expression of uPA was higher in PC-3 cells than in MG-63, KLE, and Calu-1 cancer cells. We conclude that this novel cell culture system may become a useful model to study the pathophysiology of the osteoblastic reaction in vitro.

Wettability of cross-linked collagenous biomaterials: in vitro study.

Collagenous biomaterials can be treated by chemical and physical agents to decrease biodegradation rate. Treatments to collagen may modify surface properties and subsequently cell and platelet behaviour. Collagenous films were either uncross-linked and cross-linked by glutaraldehyde, formaldehyde or cyanamide and/or treated by a severe dehydration. Contact angles, platelet contacting assay and fibroblast morphology were investigated. After severe dehydration, wettability was diminished except for formaldehyde-cross-linked and severely dehydrated films. Glutaraldehyde-cross-linked collagen results in an increase in wettability. Platelets were similarly distributed, except on formaldehyde-cross-linked films that exhibited no platelet aggregation. Fibroblasts were in a spreading phase on most collagenous films. However, cytotoxicity was noticed on some aldehyde-cross-linked films. No direct relationship was found between contact angles and platelet-cell attachment.

Collagen-based wound dressing: effects of hyaluronic acid and fibronectin on wound healing.

Our previous studies have shown that a collagen-based wound dressing induces the spatial deposition of wound tissue. This study was conducted to determine the effects of hyaluronic acid and fibronectin on wound healing. These macromolecules play an important role in wound healing, embryonic development and cellular migration in vitro. The effects of the addition of varying levels of fibronectin and hyaluronate to a collagen sponge were studied. Low levels of both hyaluronate and fibronectin modified the structure of the implant, and resulted in increased chemoattraction, replication and collagen deposition in an in vivo wound healing model.

Biological molecule-impregnated polyester: an in vivo angiogenesis study.

Specific extracellular matrix molecules and growth factors (GFs) with angiogenic properties could be combined with biomaterials to enhance angiogenesis and subsequently tissue ingrowth through the wall of the porous structure. In this study, composite fibrin matrices containing hyaluronic acid (HA), fibronectin (FN) and/or fibroblast growth factor-1 (FGF-1), FGF-2 and an endothelial cell growth supplement (ECGS) were adsorbed onto Dacron meshes which were then implanted subcutaneously in mice. The release from the implants and the tissue distribution of implanted GFs were determined in vivo using radiolabelled FGF-2. Angiogenesis was quantified by counting the number of capillaries present in each Dacron histological serial section. Radiolabelled GF was rapidly released from matrices and was absent from them by day 28. A very low percentage of the implanted radiolabelled GFs was found in the kidneys and livers of the animals. The number of microvessels formed within fibrin-impregnated samples was increased in the presence of HA and ECGS at 14 d and of FN and ECGS or FGF-2 at 28 d. FGF-1 had no direct effect on angiogenesis in our model. These results indicate that enhancement of vascularization within prosthesis mesh may be achieved by using fibrin as a support for angiogenic molecules such as HA, FN and FGFs.

Heparin-fibroblast growth factor-fibrin complex: in vitro and in vivo applications to collagen-based materials.

Biological molecules such as fibrin and growth factors could have interesting features to design bioactive biomaterials and particularly collagen-based materials used as connective tissue replacement. Different combinations of fibroblast growth factor (FGF) and heparin complexed to fibrin were analysed. In vitro, FGF bound to matrix was rapidly, but partially released, specifically with heparin. Heparin concentrations were progressively equilibrated between matrix and medium. DNA replication of fibroblasts grown either on or within fibrin matrices was increased in the presence of both FGF and high doses of heparin incorporated in fibrin. Subcutaneous implantations of collagen sponges impregnated with composite fibrin matrices showed qualitative and quantitative tissue ingrowth within the sponges. The uncross-linked collagen of fibrin-impregnated sponges swelled after implantation. The resulting fibroblast-infiltrated tissue resembled a normal dense connective tissue that was observed particularly in the presence of high doses of heparin and FGF incorporated in fibrin.

Fibroblast growth on a porous collagen sponge containing hyaluronic acid and fibronectin.

We have previously shown that the presence of fibronectin (FN) and/or hyaluronic acid (HA) in a 3-dimensional type I collagen sponge enhances wound healing in vivo. In the present study the same material was used as a support for growth of fibroblasts in vitro. Using radiochemical techniques, scanning electron and light microscopy, the properties of fibroblasts cultured on the collagen sponge or on the sponge containing HA or FN have been compared with cultures grown on plastic dishes. Fibroblast replication and collagen synthesis were higher on plastic than on the collagen sponge. In the presence of HA or FN the entire thickness of the sponge was infiltrated by fibroblasts which rapidly replicated. The presence of HA or FN increased synthesis of collagen which was largely deposited around cells.

Behaviour of fibroblasts and epidermal cells cultivated on analogues of extracellular matrix.

A porous collagen sponge can be used for supporting epidermal cells and fibroblasts in order to manufacture an artificial skin. Fibroblasts were grown on analogues of extracellular matrix containing collagen and glycosaminoglycans and/or glycoproteins. Cell replication, and also infiltration of fibroblasts, were enhanced by the presence of hyaluronic acid and/or fibronectin. Epidermal cells grown on a collagen sponge have been characterized by microscopic observations. Epidermal cells on the surface of the sponge showed an incomplete differentiation in comparison to normal skin; clumps of epidermal cells were found in the interior of the sponge. Epidermal cell replication was enhanced in the presence of collagen sponge seeded with fibroblasts.

In vivo evaluation of hydrophobic and fibrillar microporous polyetherurethane urea graft.

Mitrathane hydrophobic and fibrillar microporous prosthesis was implanted as infrarenal arterial substitute in dogs; it was evaluated in terms of patency rates, healing characteristics and biostability. Segments of grafts were implanted in duplicate for a period of implantation of 24 h, 1 wk, 1 month and 6 month. Two control grafts from the Ontario Research Foundation were implanted: one for 1 month, the other for six month. All except the two Mitrathane grafts implanted for 6 month were patent at death. The Mitrathane grafts showed kinking at one and 6 month post-implantation. The ORF graft implanted for 1 month was found crinkled in its mid-section and the external capsule was ruptured in the graft implanted for 6 month, without crinkling. Histological studies showed fibrin deposits on the flow surface and infiltration of blood elements into the wall of the Mitrathane grafts implanted for 24 h and 1 wk. A thin internal capsule was present on the graft flow surface of both types of graft tested 1 month post-implantation; scanning electron microscopy revealed the presence of endothelial-like cells on the luminal surface, particularly in the vicinity of the anastomoses. At 6 month, the Mitrathane grafts were occluded by a thick thrombus originating from the anastomoses, while the ORF graft showed infiltration of collagen through the polyurethane fibrillar structure of the wall with an endothelial-like lining covering the flow surface in the vicinity of both anastomoses.

Endothelial cells exposed to erythrocytes under shear stress: an in vitro study.

After injury and vascular replacement, endothelial cell recovery is limited and could lead to thrombosis. Seeding small diameter vascular prosthesis with endothelial cells has been proposed to fulfil cell lining and improve surface hemocompatibility. However, detachment of seeded cells occurs following implantation. Previous in vitro studies have looked at the fluid shear stress as a major cause of cell detachment. To our knowledge, the role of erythrocyte collisions has not been investigated. The present in vitro study aims at investigating whether endothelial cell adhesion depends on (i) the presence of erythrocytes in flow and (ii) the latent culture period (1, 24 and 48 h) between seeding and exposure to flow. Endothelial cells were exposed to culture media containing different erythrocyte concentrations using a steady laminar flow of 1350 ml min(-1) in a parallel plate flow chamber. Endothelial cell morphology in dynamic conditions was quantified and compared to that in static conditions. The projected area of cells were mostly found smaller under dynamic than static conditions, particularly at a wall shear stress of 23 dyn cm(-2). Cells from the 1 h latent culture period were oriented parallel to the flow axis and were more elongated than under static conditions. Conversely, endothelial cell shape was slightly modified when either the latent period or the wall shear stress was increased. Disparate orientation was observed on confluent endothelial cells (24-48 h latent period) exposed to shear stress with or without erythrocytes. Increasing fluid viscous forces due to erythrocytes play a critical role on the behaviour of freshly seeded endothelial cells upon exposure to blood flow.

Vascugraft microporous polyesterurethane arterial prosthesis as a thoraco-abdominal bypass in dogs.

In their progression towards clinical acceptance, any new synthetic vascular grafts under development must undisputedly prove that the chemistry and structure used in the construction of the prostheses is safe and that their biocompatibility and performance as arterial substitutes are satisfactory without degradation or weakening of the device. This study was conducted to evaluate the safety of the microporous polyesterurethane Vascugraft by investigating its biocompatibility in terms of cellular proliferation, morphology and adhesion of human fibroblasts on virgin and blood-soaked Vascugraft prostheses, and its performance in vivo as a large calibre graft in a canine thoraco-abdominal bypass model for periods of implantation ranging from 4 h to 6 months. After 3 d incubation, better cell proliferation and adhesion were observed on blood-soaked Vascugraft than on a non-porous polyurethane graft, Mitrathane, and two other polytetrafluorethylene prostheses, Impra and Goretex. Furthermore, no leachable cytotoxic contaminants were released from the prostheses. In vivo, the Vascugraft has demonstrated a good performance with the development of an endothelialised internal capsule at both anastomoses 2 weeks after implantation, reaching the medial portion of the graft at 4 months. During this period, the prostacyclin I2/thromboxane A2 ratio increased and was higher than 1.0 at 2 months. In addition, the Vascugraft exhibited low surface thrombogenicity in terms of radiolabelled platelets and fibrin deposited. Chemically, as revealed by ESCA and FTIR analyses, a slight decrease in carbonate content was observed on the external surface of the Vascugraft during the early post-implantation periods. Breaks in the microfibrous structure were also observed at 4 and 6 months, occurring mainly in the anastomotic regions and believed to be stress-related. This study shows that the polymer used in the Vascugraft is biocompatible in terms of fibroblast proliferation and promotes fair healing characteristics. However, the chemical and structural surface modifications noted in this study are disturbing and question the total inocuity of the Vascugraft. Consequently, the decision by B. Braun Melsungen AG to end this project is both highly conscientious and professional.

Stromal fibroblasts are required for PC-3 human prostate cancer cells to produce capillary-like formation of endothelial cells in a three-dimensional co-culture system.

The outcome of patients with prostate cancer is largely dependent on the ability of the primary tumor for local invasion, angiogenesis and metastasis. To better understand the cell-cell interactions that participate in prostate cancer neovascularization, we have developed a novel three-dimensional co-culture system. Capillary-like structures were induced in fibrin gel in which collagen gels containing fibroblasts and/or PC-3 human prostate adenocarcinoma cells were sandwiched together. In the presence of collagen-embedded fibroblasts, angiogenesis apparently occurred, while endothelial cells did not survive when only PC-3 cells were embedded in collagen. In contrast, when PC-3 cells were combined with fibroblasts in collagen gel an enhanced formation of capillary-like structure formation was noted, particularly using FGF-2-supplemented medium. In addition, we observed morphological evidence of PC-3 cells and fibroblast invasion into fibrin using this system. Therefore, we conclude that fibroblasts apparently play an important role in angiogenesis and tumor invasion. Furthermore, this novel three-dimensional co-culture is apparently a promising model for studying de novo angiogenesis and tumor invasion in vitro.

Three-dimensional type I collagen cell culture systems for the study of bone pathophysiology.

Three-dimensional (3-D) type I collagen cell culture systems composed of reconstituted collagen fibres are able to support short- and long-term growth of various cell types, including cancer cell lines, endothelial cells, endometrial cells, hepatocytes, osteoblasts and fibroblasts and to sustain or even enhance cell differentiation, in vitro. In addition, 3-D culture systems have been successfully used in the investigation of complex biological processes, such as angiogenesis, wound healing, tumour invasion and metastasis. The latter suggested that 3-D culture systems have the potential to simulate cell-cell interactions, which take place in tissues under physiological and pathophysiological conditions. This review focuses on the investigational use of 3-D collagen cell culture systems in bone physiology and the pathophysiology of skeletal metastasis.

Porous collagen sponge wound dressings: in vivo and in vitro studies.

Collagen-based materials can be formed into a three-dimensional sponge for use as a wound dressing and as a support for cell cultured skin components. Factors such as biocompatibility, morphological structure and addition of non-collagenous molecules to collagen are analyzed and discussed. Large pores or channels, interchannel communications and combinations of macromolecules of the connective tissue enhance wound tissue infiltration in vivo as well as cell growth in vitro into collagen sponges. The presence of such factors can be useful in patients with excised burn wounds and pressure skin ulcers.

Method of growing vaginal mucosal cells on a collagen sponge matrix. Results of preliminary studies.

A method of obtaining and growing vaginal mucosal epithelial cells in culture on plastic and on a collagen sponge was developed. The results indicate that simultaneous culture of skin and vaginal epithelial cells from the same donor showed morphologically similar growth potential in culture on plastic and on a collagen sponge. Vaginal cells grown on a collagen sponge formed elongated clumps of cells similar to those observed for skin epithelial cells. The results suggest that vaginal cells grown in culture on a collagen matrix may be a suitable biomaterial for use in reconstructive surgery on women with the Rokitansky syndrome or in those who have had a malignancy.

Human endometrial cells cultured in a type I collagen gel.

OBJECTIVE: To elaborate in vitro conditions that enable epithelial and stromal cells of human endometrium to grow within a gel of collagen. STUDY DESIGN: Primary cultures of epithelial cells derived from human endometrial biopsies were dissociated and mixed with a collagen solution, and the gel was allowed to form at physiologic pH. Control cultures were grown in plastic dishes. DNA replication was assessed by 3H-thymidine incorporation, morphology by histology and cell characterization by monoclonal antibodies to cytokeratins. RESULTS: Cells grown on plastic dishes exhibited a typical monolayer arrangement, and replication was increased 1.6-fold by the addition of stromal cell-conditioned medium (50% vol/vol). After a two- to three-week period of culture within the collagen gel in the presence of either stromal cells or stromal cell-conditioned medium, epithelial cells formed circular arrangements of cuboidal to columnar cells with open lumina resembling glands. These glandlike structures were cytokeratin positive as assessed by immunohistochemistry, thereby confirming their epithelial nature. CONCLUSION: The development of differentiated epithelial structures in a three-dimensional gel provides a promising method of studying various biochemical and cellular interactions of eutopic and ectopic endometrium.

New approaches for biocompatibility testing using cell culture.

We are proposing two modified assays for biocompatibility testing which analyze the effects of cytotoxic substances leached from a biomaterial in cell culture. The biocompatibility of two vascular prostheses made of polytetrafluoroethylene was analyzed using the modified assays. One test, the "fluid medium assay" was modified by using small pieces of graft glued to a screening lid, thereby reducing the possibility of mechanical injury to cultured cells by free fragments of the tested biomaterial. Another test, the "cell inhibition assay" was modified in that the biomaterial to be tested was ground into small pieces while at very low temperature. The measure of cell toxicity used was the effect on DNA replication. Our results suggest that these modified assay methods can be used to evaluate the biocompatibility of biomaterials.

Growth factors and biological supports for endothelial cell lining: in vitro study.

Endothelial cell covering over the vascular prosthesis luminal surface is a process that may require the presence of growth factors (GFs) and extracellular matrix supports. Endothelialization could be improved by combining both GFs and an extracellular matrix analog. In the present study, different biological substrates made of type I or IV collagens, gelatin, fibronectin, fibrin, laminin, chondroitin sulphate, heparan sulphate, heparin or hyaluronic acid were used to support endothelial cell culture. An endothelial cell growth supplement (ECGS) was incorporated in (group 1) or overlaid on (group 2) the substrates; or present in medium (group 3); or absent (group 4). GF binding assay using 125I bFGF showed that more GF remained combined to the substrates in group 2 than those in group 1. Growth and morphology of human umbilical vein endothelial cells were sequentially analyzed in vitro for 8 days using DNA (nuclei counts) and F-actin labelings. Growth was relatively stable for the first 48 hours, later in groups 1, 2 and 4, cell death was observed on all the substrates except for fibronectin. Growth failure could be related to the degradation or inefficient release of ECGS. In group 3, growth increased and confluency was reached within 5-8 days on all the substrates except for gelatin and type I collagen. Confluent cells containing actin filaments were organized on glycoproteins and disorganized on glycosaminoglycans and fibrin. Despite that glycoproteins can enhance cell adhesion and lining pattern, GFs continually delivered in a fresh soluble form seem to be the appropriate condition to obtain an endothelial cell lining.

A collagen-based scaffold for a tissue engineered human cornea: physical and physiological properties.

Stabilized collagen-glycosaminoglycan scaffolds for tissue engineered human corneas were characterized. Hydrated matrices were constructed by blending type I collagen with chondroitin sulphates (CS), with glutaraldehyde crosslinking. A corneal keratocyte cell line was added to the scaffolds with or without corneal epithelial and endothelial cells. Constructs were grown with or without ascorbic acid. Wound-healing was evaluated in chemical-treated constructs. Native, noncrosslinked gels were soft with limited longevity. Crosslinking strengthened the matrix yet permitted cell growth. CS addition increased transparency. Keratocytes grown within the matrix had higher frequencies of K+ channel expression than keratocytes grown on plastic. Ascorbic acid increased uncrosslinked matrix degradation in the presence of keratocytes, while it enhanced keratocyte growth and endogenous collagen synthesis in crosslinked matrices. Wounded constructs showed recovery from exposure to chemical irritants. In conclusion, this study demonstrates that our engineered, stabilized matrix is well-suited to function as an in vitro corneal stroma.