von Willebrand factor in cultured human vascular endothelial cells from adult and umbilical cord arteries and veins.

Endothelial cells were cultured from various human arteries and veins, obtained from adult individuals and from umbilical cords. We compared the storage and secretion of von Willebrand factor by endothelial cells from umbilical veins with that of endothelial cells cultured from a number of adult vessels, including aorta, arteria iliaca, vena saphena magna and vena cava. There were no differences in the way the cultured endothelial cells handled the von Willebrand factor they synthesized. Endothelial cells from the various vessels responded to stimuli in secreting stored von Willebrand factor. The cells also responded to thrombin and ionophore A23187 in producing enhanced amounts of prostacyclin. Thus, cultured umbilical vein endothelial cells have properties that are very similar to those of cultured endothelial cells of various other origins. It is concluded that foetal venous cells provide a representative model for studies of endothelial cell von Willebrand factor biosynthesis and prostacyclin production.

Absence of muscle regeneration after implantation of a collagen matrix seeded with myoblasts.

Collagens are widely used as biomaterials for e.g. soft tissue reconstruction. The present study was aimed at reconstruction of abdominal wall muscle using processed dermal sheep collagen (DSC) and myoblast seeding. Myoblasts were harvested from foetal quadriceps muscle of an inbred rat strain, cultured, seeded as non-differentiated cells into DSC-discs and incubated in vitro for 2 h. The discs were implanted in the abdominal wall defects in adult rats. Non-seeded discs functioned as control. Implantation periods till week 6 were chosen. At day 1 and 2 after implantation infiltration of granulocytes and macrophages was clearly more intense in the seeded discs than in the controls. Relatively large numbers of mast cells infiltrated from the side of the adhering omentum. In central areas of the discs, seeded cells were easily recognized till day 5, since non-seeded control discs did not contain such cells. Ingrowth of host cells and tissue at the margins proceeded faster with the seeded discs. Lymphocyte accumulations were observed in the 3 week seeded specimen. At week 3 and week 6, in the seeded discs muscle tissue was not present, in contrast to very large giant-like cells. It is concluded that the chosen method of myoblast seeding did not result in the regeneration of muscle during this observation period. Unfavorable circumstances such as humoral factors, direct cellular interactions (phagocytosis), indirect cellular interactions (cytokines), or initial absence of vascularization, may play a role. Further studies are required.

Cytotoxicity testing of wound dressings using methylcellulose cell culture.

Wound dressings may induce cytotoxic effects. In this study, we check several, mostly commercially available, wound dressings for cytotoxicity. We used our previously described, newly developed and highly sensitive 7 d methylcellulose cell culture with fibroblasts as the test system. Cytotoxicity is assessed by monitoring cell growth inhibition, supported by cell morphological evaluation using light and transmission electron microscopy. We tested conventional wound dressings, polyurethane-based films, composites, hydrocolloids and a collagen-based dressing. It was shown that only 5 out of 16 wound dressings did not induce cytotoxic effects. All 5 hydrocolloids were found to inhibit cell growth (greater than 70%), while cells had strongly deviant morphologies. The remaining wound dressings showed medium cytotoxic effects, with cell growth inhibition, which varied from low (+/- 15%), medium-low (+/- 25%) to medium-high (+/- 50%). Measurable cytotoxic effects of dressings detected in vitro are likely to interfere with wound healing when applied in vivo. The results are discussed in view of the clinical uses with contaminated wounds, impaired epithelialization or hypergranulation.

Rapid endothelialization of microporous vascular prostheses covered with meshed vascular tissue: a preliminary report.

The aim of our study was to induce rapid endothelialization of vascular prostheses which replace malfunctioning blood vessels. We developed a technique of meshing a section of autologous arterial or venous tissue, wrapping it around a porous vascular prosthesis and fixing it with a running suture. This combination was used to replace a 1 cm segment of rat abdominal aorta. The implanted specimens were harvested after 8, 16, 24 or 135 d and evaluated for cellular ingrowth and endothelial coverage with light and scanning electron microscopy. All grafts were patent when harvested. Cells of the meshed tissue were observed to migrate through the pores in the prosthetic wall to cover the luminal surface. The presence of meshed vascular tissue around the prosthesis resulted in complete endothelialization within 24 d and sometimes within 8 d after implantation. This endothelial layer is still present after 135 d. In control prostheses without meshed vascular tissue, complete endothelialization takes at least 42 d. We conclude that the application of meshed vascular tissue around porous vascular prostheses results in patent prostheses with a rapidly formed and long-term surviving endothelium layer.

Vacuum cell seeding: a new method for the fast application of an evenly distributed cell layer on porous vascular grafts.

The study was to develop a method to induce rapid endothelial coverage of vascular prostheses by cell seeding. The method uses vacuum pressure and is therefore called vacuum cell seeding. A special seeding device was constructed, in which grafts of different length and/or inner diameter could be positioned. Microporosity of the grafts was a prerequisite for this method. Two types of commercially available microporous grafts were tested. The ePTFE graft routinely used clinically needed pretreatment to enable the method, whilst a polyurethane-based graft could be seeded as received. Vacuum cell seeding applied cells from a suspension in culture medium within 10 min in an evenly distributed cell layer on to the luminal graft surface. The adhering cells immediately started flattening, thereby completely covering the luminal surface. It was concluded that the vacuum cell seeding method rapidly introduced a confluent layer of seeded cells on porous vascular grafts in a simple way, which in the clinical setting could easily be performed on the operating table.

Adhesion and spreading of cultured endothelial cells on modified and unmodified poly (ethylene terephthalate): a morphological study.

The in vitro adhesion and spreading of human endothelial cells (HEC) on hydrophobic poly(ethylene terephthalate) (PETP) and moderately wettable tissue culture poly(ethylene terephthalate) (TCPETP) were studied with light microscopy and electron microscopy. Numbers of HEC adhering on TCPETP were always higher than those found on PETP. When cells were seeded in the presence of serum, extensive cell spreading on both PETP and TCPETP was observed after the first 30 min. Thereafter, spread cells appeared to withdraw from the PETP surface, resulting in irregularly shaped cells. Complete cell spreading occurred on TCPETP. Complete cell spreading also occurred on PETP and TCPETP when HEC had first been seeded from phosphate buffer solution and serum was supplied after 30 min. Furthermore, HEC spread on both PETP and TCPETP when the surfaces were precoated with protein(s), which promotes cell adhesion. However, when plasma was used for the coating, spread cells did not proliferate in a monolayer pattern. This study shows that TCPETP is, in general, a better surface for adhesion and proliferation of HEC than is PETP, suggesting that vascular prostheses with a TCPETP-like surface will perform better in vivo than prostheses made of PETP.

Deposition of endothelial fibronectin on polymeric surfaces.

Cellular fibronectin is deposited on tissue culture polystyrene during the adhesion and spreading of cultured human endothelial cells (HEC). Following the seeding of HEC upon this polymer, larger amounts of fibronectin are deposited as both cell density and incubation time increase. Our results indicate that the ability to deposit cellular fibronectin onto a polymeric surface is a condition for the spreading and proliferation of HEC.

Adhesion of cultured human endothelial cells onto methacrylate polymers with varying surface wettability and charge.

The adhesion of human endothelial cells (HEC) onto a series of well-characterized methacrylate polymer surfaces with varying wettabilities and surface charges was studied either in serum-containing (CMS) or in serum-free (CM) culture medium. HEC adhesion in CMS onto (co)polymers of hydroxyethyl methacrylate (HEMA) and methyl methacrylate (MMA) was found to be optimal on the moderately wettable copolymer (mol ratio 25 HEMA/75 MMA). Positively-charged copolymers of HEMA or MMA with trimethylaminoethyl methacrylate-HCl salt (TMAEMA-Cl), both with mol ratios of 85/15 and a negatively-charged copolymer of MMA with methacrylic acid (MAA), mol ratio 85/15, showed high numbers of adhering HEC. In CM, HEC adhered onto the three charged copolymers mentioned above, but neither onto the copolymer of HEMA and MAA (mol ratio 85/15) nor onto the HEMA/MMA co- and homopolymers. Complete cell spreading in CM was only observed on the positively-charged copolymers.

Cross-linking of dermal sheep collagen using a water-soluble carbodiimide.

A cross-linking method for collagen-based biomaterials was developed using the water-soluble carbodiimide 1-ethyl-3-(3-dimethyl aminopropyl)carbodiimide hydrochloride (EDC). Cross-linking using EDC involves the activation of carboxylic acid groups to give O-acylisourea groups, which form cross-links after reaction with free amine groups. Treatment of dermal sheep collagen (DSC) with EDC (E-DSC) resulted in materials with an increased shrinkage temperature (Ts) and a decreased free amine group content, showing that cross-linking occurred. Addition of N-hydroxysuccinimide to the EDC-containing cross-linking solution (E/N-DSC) increased the rate of cross-linking. Cross-linking increased the Ts of non-cross-linked DSC samples from 56 to 73 degrees C for E-DSC and to 86 degrees C for E/N-DSC samples, respectively. For both cross-linking methods a linear relation between the decrease in free amine group content and the increase in Ts was observed. The tensile strength and the high strain modulus of E/N-DSC samples decreased upon cross-linking from 18 to 15 MPa and from 26 to 16 MPa, respectively. The elongation at break of E/N-DSC increased upon cross-linking from 142 to 180%.

In vitro degradation of dermal sheep collagen cross-linked using a water-soluble carbodiimide.

Bacterial collagenase was used to study the susceptibility of dermal sheep collagen (DSC) cross-linked with a mixture of the water-soluble carbodiimide 1-ethyl-3-(3-dimethyl aminopropyl)-carbodiimide hydrochloride and N-hydroxysuccinimide (E/N-DSC) towards enzymatic degradation. Contrary to non-cross-linked DSC (N-DSC), which had a rate of weight-loss of 18.1% per hour upon degradation, no weight loss was observed for E/N-DSC during a 24 h degradation period. The tensile strength of the E/N-DSC samples decreased during this time period, resulting in partially degraded samples having 80% of the initial tensile strength remaining. The susceptibility of E/N-DSC samples towards enzymatic degradation could be controlled by varying the degree of cross-linking of the samples. Ethylene oxide sterilization of E/N-DSC samples made the material more resistant against degradation compared with non-sterilized E/N-DSC samples. This may be explained by a decrease of the adsorption of bacterial collagenase onto the collagen owing to reaction of ethylene oxide with remaining free amine groups in the collagen matrix.

In vivo degradation of processed dermal sheep collagen evaluated with transmission electron microscopy.

The in vivo degradation of hexamethylenediisocyanate-tanned dermal sheep collagen was studied with transmission electron microscopy. Discs of hexamethylenediisocyanate-tanned dermal sheep collagen were subcutaneously implanted in rats. Both an intra- and an extracellular route of degradation could be distinguished. In addition to normal components of a typical foreign body reaction, remarkable phenomena, such as locally deviant neutrophil morphology, infiltration of basophil-like cells, indications of foreign body multinucleate giant cells formed from different cell types, aluminium silicate accumulations and calcium phosphate depositions, were observed. Foreign body multinucleate giant cells intracellularly degraded hexamethylenediisocyanate-tanned dermal sheep collagen after internalization. Both internalized and cellularly enveloped hexamethylenediisocyanate-tanned dermal sheep collagen degraded by the detachment of fibrils. Another extracellular route of degradation was characterized by calcium phosphate depositions in large bundles of hexamethylenediisocyanate-tanned dermal sheep collagen. From 6 wk, the hexamethylenediisocyanate-tanned dermal sheep collagen implant was replaced by rat connective tissue, which was subsequently also degraded. After 15 wk, the presence of basophil-like foreign body multinucleated giant cells containing aluminium/silicon-crystalline accumulations still persisted. These phenomena were related to the specific nature of the material used and suggest cytotoxicity. They emphasize the need for detailed evaluation at the ultrastructural level of newly developed biomaterials before they can be used for medical applications.

Secondary cytotoxicity of cross-linked dermal sheep collagens during repeated exposure to human fibroblasts.

We investigated commercially available dermal sheep collagen either cross-linked with hexamethylenediisocyanate, or cross-linked with glutaraldehyde. In previous in vitro studies we could discriminate primary, i.e. extractable, and secondary cytotoxicity, due to cell-biomaterial interactions, i.e. enzymatic actions. To develop dermal sheep collagen for clinical applications, we focused in this study on the release, e.g. elimination, of secondary cytotoxicity over time. We used the universal 7 d methylcellulose cell culture with human skin fibroblasts as a test system. Hexamethylenediisocyanate-cross-linked dermal sheep collagen and glutaraldehyde-cross-linked dermal sheep collagen were tested, with intervals of 6 d, over a culture period of 42 d. With hexamethylenediisocyanate-cross-linked dermal sheep collagen, cytotoxicity, i.e. cell growth inhibition and deviant cell morphology, was eliminated after 18 d of exposure. When testing glutaraldehyde-cross-linked dermal sheep collagen, the bulk of cytotoxic products was released after 6 d, but a continuous low secondary cytotoxicity was measured up to 42 d. As a control, non-cross-linked dermal-sheep collagen was tested over a period of 36 d, but no secondary cytotoxic effects were observed. The differences in release of secondary cytotoxicity between hexamethylenediisocyanate-cross-linked dermal sheep collagen, glutaraldehyde-cross-linked dermal sheep collagen and non-cross-linked dermal sheep collagen are explained from differences in cross-linking agents and cross-links obtained. We hypothesize that secondary cytotoxicity results from enzymatic release of pendant molecules from hexamethylene-diisocyanate-cross-linked dermal sheep collagen, e.g. formed after reaction of hydrolysis products of hexamethylenediisocyanate with dermal sheep collagen.(ABSTRACT TRUNCATED AT 250 WORDS)

Interaction of cultured human endothelial cells with polymeric surfaces of different wettabilities.

The in vitro interaction of human endothelial cells (HEC) and polymers with different wettabilities in culture medium containing serum was investigated. Optimal adhesion of HEC generally occurred onto moderately wettable polymers. Within a series of cellulose type of polymers the cell adhesion increased with increasing contact angle of the polymer surfaces. Proliferation of HEC occurred when adhesion was followed by progressive flattening of the cells. Our results suggest that moderately wettable polymers exhibit a serum and/or cellular protein adsorption pattern that is favourable for growth of HEC.

Kinetics of cell spreading on protein precoated substrata: a study of interfacial aspects.

In this paper, interfacial aspects of spreading and adhesion of human skin fibroblasts on solid substrata after protein precoating have been studied. Three solid substrata were used with different surface free energy (gamma s): Tissue Culture Polystyrene (TCPS) with gamma s = 70 erg.cm-2, Polyvinylfluoride (PVF) with gamma s = 56 erg.cm-2 and Fluoroethylenepropylene (FEP) copolymer with gamma s = 18 erg.cm-2. The substrata were precoated with fetal calf serum, bovine fibronectin or bovine serum albumin. Cell spreading was evaluated by means of light microscopy and scanning electron microscopy (SEM). Adhesion sites were studied by transmission electron microscopy (TEM). In general, spreading was lowest on FEP and highest on TCPS. Although protein precoating markedly increased cell spreading, the relative order in which the cells spread on the protein precoated substrata remained identical to that on the bare substrata. Analysis of the kinetics of spreading demonstrated that spreading was fastest on the high-energy substratum and slowest on the low-energy substratum. In the presence of all three types of protein precoating, the average distance between a cell and a substratum after spreading was smaller (20-50 nm) than without a coating (greater than 100 nm).

Successive epoxy and carbodiimide cross-linking of dermal sheep collagen.

Cross-linking of dermal sheep collagen (N-DSC, T(S) = 46 degrees C, number of amine groups = 31 (n/1000)) with 1,4-butanediol diglycidyl ether (BDDGE) at pH 9.0 resulted in a material (BD90) with a high T(S)(69 degrees C), a decreased number of amine groups of 15 (n/1000) and a high resistance towards collagenase and pronase degradation. Reaction of DSC with BDDGE at pH 4.5 yielded a material (BD45) with a T(S) of 64 degrees C, hardly any reduction in amine groups and a lower stability towards enzymatic degradation as compared to BD90. The tensile strength of BD45 (9.2 MPa) was substantially improved as compared to N-DSC (2.4 MPa), whereas the elongation at break was reduced from 210 to 140%. BD90 had a tensile strength of 2.6 MPa and an elongation at break of only 93%. To improve the resistance to enzymes and to retain the favorable tensile properties, BD45 was post-treated with 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide (EDC) in the presence of N-hydroxysuccinimide (NHS) to give BD45EN. Additional cross-linking via the formation of amide bonds took place as indicated by the T(S) of 81 degrees C and the residual number of amine groups of 19 (n/1000). BD45EN was stable during exposure to both collagenase and pronase solutions. The tensile properties (tensile strength 7.2 MPa, elongation at break 100%) were comparable to those of BD45 and glutaraldehyde treated controls (G-DSC). Acylation of the residual amine groups of BD45 with acetic acid N-hydroxysuccinimide ester (HAc-NHS) yielded BD45HAc with a large reduction in amine groups to 10 (n/1000) and a small reduction in T(S) to 62 degrees C. The stability towards enzymatic degradation was reduced, but the tensile properties were comparable to BD45.

Repetitive subcutaneous implantation of different types of (biodegradable) biomaterials alters the foreign body reaction.

In the present study two biodegradable materials (cross-linked collagens) and two non-biodegradable materials (polyurethane and silicone) were applied in a repetitive subcutaneous implantation model in rats. In contrast to the first challenge, the second challenge with the same type of material, but at a different subcutaneous site of the same animal, induced an increase of macrophages and giant cells inside the biodegradable materials. Additionally, only after the second challenge clusters and accumulations of plasma cells were present in the surrounding tissue of each type of material. In the same areas an increase of MHC II expression was measured by immunocytochemistry. Differences in the numbers of macrophages and T cells were not observed around the explants. Undifferentiated B cells or NK cells were not present at any time point. The results indicate that alterations observed after the second challenge did not depend on biodegradation of the materials. Significance of these findings should be considered in view of increased and repetitive use of the same type of biomaterial (possibly for different application sites) for implantation in patients.

Attachment of glycosaminoglycans to collagenous matrices modulates the tissue response in rats.

Biocompatibility and tissue regenerating capacity are essential characteristics in the design of collagenous biomaterials for tissue engineering. Attachment of glycosaminoglycans (GAGs) to collagen may add to these characteristics by creating an appropriate micro-environment. In this study, porous type I collagen matrices were crosslinked using 1-ethyl-3-(3-dimethyl aminopropyl)carbodiimide, in the presence and absence of chondroitin sulfate and heparan sulfate. The tissue response to these matrices was evaluated after subcutaneous implantation in rats. Biocompatibility of the matrices was established by the induction of a transitional inflammatory response, and the generation of new host tissue. Non-crosslinked collagen was gradually resorbed and replaced by collagenous connective tissue. By contrast, crosslinked matrices, with and without GAGs. retained their scaffold integrity during implantation, and supported the interstitial deposition and organization of extracellular matrix. In addition, crosslinking decreased tissue reactions at late time intervals. No calcification in any of the implants was observed. The presence of GAGs preserved porous lamellar matrix structures. Heparan sulfate in particular promoted angiogenesis at weeks 2 and 4, predominantly at the matrix periphery. The almost complete absence of macrophages and giant cells associated with collagen-GAG matrices, after 10 weeks implantation, indicated a reduced foreign body reaction. It is concluded that attachment of GAGs to collagen matrices modulates the tissue response. The potential of these biocompatible scaffolds for tissue engineering is increased by preserving porous matrix integrity. promoting angiogenesis and reducing foreign body reactions.

TGF-beta and bFGF affect the differentiation of proliferating porcine fibroblasts into myofibroblasts in vitro.

Fibroblasts and myofibroblasts are involved in the foreign body reaction to biomaterials, especially in capsule formation. However, contraction or detachment of the capsule can lead to complications. Biocompatibility of biomaterials may be improved by the application of proteins regulating the differentiation or activation of (myo)fibroblasts. Myofibroblasts, differentiating from fibroblasts can be identified by the expression of alpha-smooth muscle actin (alpha-SM actin). We investigated the influence of proliferation and quiescence on the differentiation of porcine dermal cells and whether transforming growth factor-beta (TGF-beta) and basic fibroblast growth factor (bFGF) are involved in the differentiation of proliferating cells. Porcine cells were used because pigs increasingly function as in vivo models while little is known of the characteristics of their cells. Serum-free cultured, quiescent fibroblasts differentiated into myofibroblasts, while proliferating fibroblasts cultured in the presence of serum containing TGF-beta, formed alpha-SM actin-negative cell clusters. After reaching confluency, these clusters started to expressing alpha-SM actin. Moreover, these proliferating cells produced TGF-beta from day 4 onwards while bFGF did not. Differentiation into myofibroblasts was inhibited by bFGF and to an even greater extent by antibodies to TGF-beta. Further, two theories concerning the role of the myofibroblast in tissue contraction in view of two biomaterial application will be discussed.

(Electron) microscopic observations on tissue integration of collagen-immobilized polyurethane.

The foreign body reactions to collagen-immobilized polyurethane (PU-CI) films during subcutaneous implantation in rats were characterized. The underlying concept is that collagen-immobilization will improve the tissue integration. Since the method of collagen-immobilization involves the covalent coupling of collagen to an acrylic acid (AA) based surface graft, both non-modified PU and PU-AA were used as controls. Bare PU has a flat surface, whereas both PU-AA and PU-CI displayed a slightly roughened surface. Implantation showed that PU-CI induced early after implantation a far more intense foreign body reaction than PU and PU-AA. This reaction consisted of increased presence of fibrin, granulocytes and macrophages. Roughening of the surface as with PU-AA induced only a small increase in fibrin formation and cellular migration. At day 5 the reaction to PU-CI had slowed down; giant cell formation now slowly started but was decreased compared to PU and PU-AA. At day 10 capsules around each type of material looked similar, but in contrast to PU. PU-CI films could no longer be dissected from their capsules. Only at week 3 this also occurred with PU, at which time point again similar capsules with the three materials were observed. At week 6, of the three materials PU-CI showed the thinnest capsule with most immediate adherence of connective tissue. These results show that collagen-immobilization of PU increased the early tissue reaction and therefore the tissue integration. The thin capsule observed at 6 weeks may be beneficial in e.g. infectious circumstances, when easy access for immune reactions is needed. This, and the long-term performance of PU-CI will be a matter of future investigations.

Cardiac tissue engineering: characteristics of in unison contracting two- and three-dimensional neonatal rat ventricle cell (co)-cultures.

Patients with heart failure have, in spite of improved palliative therapies, bad prognosis. Cardiac tissue engineering by use of a temporary bioscaffold and cardiomyocytes may help to find answers for future treatments in heart failure. For that purpose two neonatal rat heart ventricular cell fractions were obtained after a gradient cell separation. Time related characteristics of Fractions I and II were established in two-dimensional (2-D) and three-dimensional (3-D) cell cultures. The 3-D cardiac constructs were obtained by use of a bovine type I collagen matrix after culturing either under static conditions or in the HARV bioreactor. With the 2-D cultures contracting cells were present after 1 day, and reached confluency from day 5 on and this was maintained up to 135 days. In Fraction-I some non-contracting cells were always noticed between the (in time in unison) contracting cells. Transmission electron microscopy (TEM) revealed that these mainly concerned fibroblasts. Differences in the expression of alpha-SM-1 actin and troponin-T were observed between the two fractions. In both fractions endothelial cells and macrophages were only sporadically observed. All through the 3-D matrix pendant-like single cell and clustered cell contractions were present after 1-2 days, resulting in time in unison contracting of cells with the collagen matrices. The whole event was faster with Fraction-I and was observed up to 3 weeks. At this time point clusters of troponin-T positive cells were found scattered through the collagen matrices. Additionally, TEM revealed healthy layers of connected cardiomyocytes with intercalated discs, in this case on and in between the collagen fibres. These findings provide evidence that in unison contracting structurally organized cell-matrix cardiac constructs can be engineered by use of co-cultures (neonatal cardiomyocytes and fibroblasts) and collagen matrices, which is very promising for the repair of larger scar areas of the myocardium.