Nutrient and oxygen supply of cells are crucial to tissue engineering in general. If a sufficient supply cannot be maintained, the development of the tissue will slow down or even fail completely. Previous studies on oxygen supply have focused on measurement of oxygen partial pressures (pO(2)) in culture media or described the use of invasive techniques with spatially limited resolution. The experimental setup described here allows for continuous, noninvasive, high-resolution pO(2) measurements over the cross-section of cultivated tissues. Applying a recently developed technique for time-resolved pO(2) sensing using optical sensor foils, containing luminescent O(2)-sensitive indicator dyes, we were able to monitor and analyze gradients in the oxygen supply in a tissue over a 3-week culture period. Cylindrical tissue samples were immobilized on top of the sensors. By measuring the luminescence decay time, two-dimensional pO(2) distributions across the tissue section in contact with the foil surface were determined. We applied this technique to cartilage explants and to tissue-engineered cartilage. For both tissue types, changes were detected in monotonously decreasing gradients of pO(2) from the surface with high pO(2) to minimum pO(2) values in the center of the samples. Nearly anoxic conditions were observed in tissue constructs ( approximately 0 Torr) but not in excised cartilage discs ( approximately 20 Torr) after 1 day. Furthermore, the oxygen supply seemed to strongly depend on cell density and cell function. Additionally, histological analysis revealed a maximum depth of approximately 1.3 mm of regular cartilage development in constructs grown under the applied culture conditions. Correlating analytical and histological analysis with the oxygen distributions, we found that pO(2) values below 11 Torr might impair proper tissue development in the center. The results illustrate that the method developed is an ideal one to precisely assess the oxygen demand of cartilage cultures.

Investigation of novel experimental application systems for growth factors or other bioactive substances in tissue engineering is often limited by high costs of substances and would benefit from a defined and easily controllable model tissue system. Herein, we demonstrate a potential three-dimensional in vitro system using engineered cartilage as a model tissue and readily available insulin as a model drug. Previously it has been shown that insulin-like growth factor-I (IGF-I) has profound effects on tissue-engineered cartilage in vitro. Insulin is known to bind to the IGF-I receptor and to elicit significant responses in cartilage. In this study, bovine articular chondrocytes were seeded onto biodegradable polyglycolic acid (PGA) scaffolds and cultured for up to 7 weeks. Exogenous insulin (0.05-50 microg/ml) increased the growth rate and the glycosaminoglycan fraction of tissue-engineered cartilage, decreased the cell number in the tissue constructs, and improved the morphological appearance, with 2.5 microg/ml being the most favorable concentration. The observed effects of insulin were similar to effects of IGF-I (0.05 microg/ml) and were in agreement with the reported binding constants of IGF-I and insulin at the IGF-I receptor. Besides the possibility to employ insulin as a potent substance to improve tissue-engineered cartilage, the presented easily controllable in vitro system may be used in the future to evaluate experimental growth factor application devices using economically favorable insulin as a model protein.

The effects of three derivatives of the N-terminal signaling domain of hedgehog proteins on cartilage engineered in vitro were investigated, with specific focus on the ability to increase tissue growth rate and concentrations of major extracellular matrix components, that is, glycosaminoglycans (GAG) and collagen, and on the effects on morphological appearance of the tissue. Bovine articular chondrocytes were cultured on biodegradable polyglycolic acid (PGA) scaffolds with or without the addition of dipalmitoylated sonic hedgehog (dp-shh), dipalmitoylated indian hedgehog (dp-ihh), or sonic hedgehog dimer (shh-dimer) to medium with either 1% or 10% fetal bovine serum (FBS). All three hedgehog proteins dose-dependently increased construct weights (by up to 1.95-fold, dp-shh at 1,000 ng/mL) and the fraction of GAG over 4 weeks (by up to 2.7-fold, dp-shh at 1,000 ng/mL), as compared to control constructs. Dp-shh and dp-ihh elicited similar responses; a 10-fold higher concentration of nonacylated shh-dimer was necessary to reach comparable results. Positive hedgehog effects were more pronounced in medium containing 1% FBS than in medium containing 10% FBS; however, at either FBS concentration, cartilaginous tissues grown in the presence of hedgehog proteins appeared morphologically more mature. Hedgehog derivatives thus appear as promising candidates to improve the development and composition of engineered cartilage.

A major goal in tissue engineering is the controlled application of growth factors. As a novel application system, we are currently developing biomimetic polymers that are processed into three-dimensional scaffolds. Bioactive proteins will be covalently bound to the polymers via a poly(ethylene glycol) (PEG) linker. Of paramount importance is the maintenance of the biological activity of the protein after PEGylation and covalent binding to the polymer. Therefore, within this study, insulin used as a model protein was PEGylated with an active succinimidyl ester of poly(ethylene glycol) (SS-NH-PEG) (MW ~2000) and biological effects of the protein-PEG conjugate were monitored in comparison with unmodified insulin. No significant differences in chondrocyte proliferation were observed in a conventional proliferation assay after treatment with insulin or PEGylated insulin. In a complex three-dimensional cartilage-engineering model the effects of insulin and PEGylated insulin were investigated over a wide concentration range (0.025-25 microg/mL). Insulin and PEGylated insulin at equivalent concentrations resulted in cartilaginous tissue constructs exhibiting identical wet weight, cell number, biochemical composition of the extracellular matrix, and histological appearance, both compounds significantly improving tissue quality as compared with control constructs. In conclusion, the presented study demonstrates that PEGylation of insulin using SS-NH-PEG did not change the activity of the protein in a complex biological environment and is regarded as a step toward the development of biomimetic polymers.

Many current tissue-engineering investigations aim at the rational control of cell adhesion and tailored composition of biomaterial surfaces by immobilizing various protein and peptide components, such as growth factors. As a step on the way to develop polymers that allow for such surface modifications, water-soluble polymers were used as model substances to examine reactions with proteins containing amine groups. Consequently, the uncommon PEGylation of insulin in aqueous buffers was used to characterize reaction products and simulate the intended immobilization step for surface modification. Amine reactive poly(ethylene glycol)s were synthesized and characterized by (1)H nuclear magnetic resonance and gel-permeation chromatography. Furthermore, the model protein insulin was characterized concerning its accessible amino groups, using a fluorescent dye (TAMRA-SE). The resulting reaction products were identified by reversed-phase high-performance liquid chromatography and electrospray mass spectrometry. After PEGylation with hydrolytically stable poly(ethylene glycol) succinimidyl ester, the obtained PEGylated insulin was investigated by gel filtration chromatography, indicating successful attachment of the hydrophilic polymer chains. Application of an aqueous PEGylation scheme opens the door to the immediate investigation of various growth factors in cell culture, allowing for direct assessment of biological activity after forming the polymer-protein constructs with regard to later immobilization on surfaces.