Yasuhiko Tabata

Fabrication of porous gelatin scaffolds for tissue engineering

A novel method which employs water present in swollen hydrogels as a porogen for shape template was suggested for preparing porous materials. Biodegradable hydrogels were prepared through crosslinking of gelatin with glutaraldehyde in aqueous solution, followed by rinsing and washing. After freezing the swollen hydrogels, the ice formed within the hydrogel network was sublimated by freeze-drying. This simple method produced porous hydrogels. Irrespective of any rinsing and washing processes, water was homogeneously distributed into the hydrogel network, allowing the hydrogel network to uniformly enlarge and the ice to act as a porogen during the freezing process. Different porous structures were obtained by varying the freezing temperature. Hydrogels frozen in liquid nitrogen, had a two-dimensionally ordered structure, while the hydrogels prepared at freezing temperatures near -20 degrees C, showed a three-dimensional structure with interconnected pores. As the freezing temperature was lowered, the hydrogel structure gradually became more two-dimensionally ordered. These results suggest that the porosity of dried hydrogels can be controlled by the size of ice crystals formed during freezing. It was concluded that the present freeze-drying procedure is a bio-clean method for formulating biodegradable sponges of different pore structures without use of any additives and organic solvents.

Vascularization effect of basic fibroblast growth factor released from gelatin hydrogels with different biodegradabilities

Biodegradable gelatin hydrogels were prepared through the glutaraldehyde crosslinking of acidic gelatin with an isoelectric point (IEP) of 5.0 and the basic gelatin with an IEP of 9.0. The hydrogel water content was changed by the concentration of both gelatin and glutaraldehyde, used for hydrogel preparation. An aqueous solution of basic fibroblast growth factor (bFGF) was sorbed into the gelatin hydrogel freeze-dried to obtain a bFGF-incorporating gelatin hydrogel. Irrespective of the hydrogel water content, approximately 30% of the incorporated bFGF was released from the bFGF-incorporating acidic gelatin hydrogel, within the first day into phosphate-buffered saline solution at 37 degrees C, followed by no substantial release. Probably, the basic bFGF complexed with the acidic gelatin through poly-ion complexation would not be released under the in vitro non-degradation condition of gelatin. On the contrary, almost 100% of the incorporated bFGF was initially released from all types of basic gelatin hydrogels. This is due to the simple diffusion of bFGF because of no complexation between bFGF and the basic gelatin. When implanted subcutaneously into the mouse back, bFGF-incorporating acidic and basic gelatin hydrogels with higher water contents were degraded with time faster than those with lower water contents. Significant neovascularization was induced around the implanted site of the bFGF-incorporating acidic gelatin hydrogel. The induction period prolonged with the decrease in hydrogel water content. On the other hand, such a prolonged vascularization effect was not achieved by the bFGF-incorporating basic gelatin hydrogel and the hydrogel initially exhibited less enhanced effect, irrespective of the water content. These findings indicate that the controlled release of biologically active bFGF is caused by biodegradation of the acidic gelatin hydrogel, resulting in induction of vascularization effect dependent on the water content. It is possible that only the transient vascularization by the basic gelatin hydrogel is due to the initial large burst in bFGF release, probably because of the down regulation of bFGF receptor.

In vitro and in vivo comparison of bulk and surface hydrolysis in absorbable polymer scaffolds for tissue engineering.

This article describes preliminary in vitro and in vivo studies comparing bulk and surface hydrolysis in absorbable polymer scaffolds proposed for tissue engineering of bone. The two polymers systems used were a bulk hydrolyzing 50:50 poly(DL-lactide-co-glycolide) (PLGA) and a surface hydrolyzing self-catalytic poly(ortho ester) (POE). Polymer scaffolds were exposed to physiological saline at body temperature and changes in polymer mass loss and inherent viscosity were monitored over time. New bone formation and local tissue response were evaluated by implanting scaffold disks of both polymer systems into non-critical-size calvarial defects in rabbits. New bone formation was determined by bone mineral density measurements, and local tissue response was determined by qualitative histology. Preliminary results confirmed that one of the main design characteristics for absorbable polymers in tissue engineering of bone, coordination of controlled polymer mass loss with new tissue formation, appeared to be achieved better using a surface hydrolyzing POE, rather than with a bulk hydrolyzing 50:50 PLGA. Bone mineral density at 6 and 12 weeks was an average 25% higher in the surface hydrolyzing scaffold. Unfortunately, the amount of bone formed was so inconsequential that this observation is of little relevance. Use of a water-soluble signaling factor such as basic fibroblast growth factor (bFGF) failed to increase bone formation. The histological response of these two polymer systems was similar and unaffected by the presence or absence of bFGF. The persistence of structural integrity for self-catalytic POE scaffolds after 6 and 12 weeks implantation, while 50:50 PLGA scaffolds had partially collapsed after 6 weeks, suggests surface hydrolyzing scaffolds may have some advantage over bulk hydrolyzing scaffolds in resisting normal in vivo stresses when used in a calvarial defect.

Biological functions of fullerene

Fullerene (C60) is known to ef®ciently generate singlet oxygen when irradiated with light. In spite of such an unique photodynamic nature, little has been studied on biological functions of C60. This paper describes a tumor-therapeutic trial of C60 based on the photoinduced generation of active oxygen. To achieve preferential accumulation of C60 in the tumor tissue, this water-insoluble drug was chemically conjugated with poly(ethylene glycol) (PEG) not only to make it soluble in water but also to make the molecular size larger. When intravenously injected to tumor-bearing mice, the large-sized, as expected, water-soluble C60PEG conjugate was accumulated to a greater extent and retained for a longer time period in the tumor tissue than in the normal issue. Following intravenous injection of the C60-PEG conjugate, local irradiation of visible light to the tumor site induced tumor necrosis, in contrast to the conjugate injection alone. When the C60 dose was 8.48 mg/mouse, the tumor mass completely disappeared at an irradiation power of 107 J/cm, indicating a high potentiality of the C60-PEG conjugate for photodynamic tumor therapy.

Sealing Effect of Rapidly Curable Gelatin-Poly (L-Glutamic Acid) Hydrogel Glue on Lung Air Leak

BACKGROUNDnAir leak is a problem commonly occurring in lung and thoracic operations. In this study, a rapidly curable hydrogel glue was prepared as the seal for lung air leak.nnnMETHODSnMixing an aqueous solution of gelatin and poly(L-glutamic acid) with a water-soluble carbodiimide produced a hydrogel. The sealing effect on the air leak wound of rat lung was compared with that of conventional fibrin glue.nnnRESULTSnThe gelatin-poly(L-glutamic acid) hydrogel glue was solidified as rapidly as the fibrin glue, and was significantly more effective in sealing the lung air leak than the fibrin glue. Approximately 80% of the lungs treated with the hydrogel glue exhibited no air leak at the lung pressure of 50 cm H2O. Urea addition could prevent spontaneous gelatination of the mixed solution at room temperature and did not affect the hydrogel sealing effect. The bonding strength of the hydrogel glue both with and without urea to the lung tissue was significantly higher than that of the fibrin glue.nnnCONCLUSIONSnWe concluded that this strong lung adhesion of the gelatin-poly(L-glutamic acid) hydrogel glue resulted in its superior sealing effect.

Vascularization into a porous sponge by sustained release of basic fibroblast growth factor

Vascularization into a poly(vinyl alcohol) (PVA) sponge was investigated using basic fibroblast growth factor (bFGF). This growth factor was impregnated into biodegradable gelatin microspheres for its sustained release and then the bFGF-containing microspheres or free bFGF were incorporated into PVA sponges. Following subcutaneous implantation into the back of mice, the bFGF-containing gelatin microspheres induced vascularization in and around the sponge to a significantly greater extent than that of free bFGF from 3 days after implantation. Significant ingrowth of fibrous tissue into the sponge was also observed when bFGF-containing microspheres were added to the sponge in contrast to free bFGF. Tissue ingrowth occurred into the deeper portion of the sponge over time while it accompanied formation of new capillaries. Empty gelatin microspheres had no effect on vascularization and the level of fibrous tissue ingrowth into the sponge was similar to that of the control group. It was concluded that incorporation of gelatin microspheres containing bFGF into the PVA sponge was effective in prevascularization of the sponge pores.

Targeting of tumor necrosis factor to tumor by use of dextran and metal coordination.

Tumor targeting of recombinant human tumor necrosis factor alpha (TNF) and consequently an enhanced anti-tumor effect were achieved through conjugation with dextran having metal chelating, diethylenetriaminepentaacetic acid (DTPA) residues based on metal coordination. A simple mixing with the DTPA-dextran in an aqueous solution containing Cu2+ enabled TNF to coordinately conjugate to dextran. Following intravenous (i.v.) injection into tumor-bearing mice, the TNF-DTPA-dextran conjugate caused a significantly higher tumor accumulation of TNF and the longer retention period than free TNF or its mixture with the DTPA-dextran. Intravenous injection of the TNF-DTPA-dextran conjugate suppressed tumor growth to a significantly greater extent than that of free TNF at a lower injection dose. It is concluded that dextran conjugation based on Cu2+ coordination is a promising way to enhance the anti-tumor effect of TNF as a result of its passive tumor targeting.

Growth Factor Release from Gelatin Hydrogel for Tissue Engineering

One of the key technologies for regeneration of damaged and lost tissue is the sustained release of biologically active growth factors. The present study was undertaken to investigate sorption and desorption of various growth factors from biodegradable hydrogels prepared through glutaraldehyde crosslinking of gelatin with isoelectric points (IEPs) of 5.0 and 9.0, which are named acidic and basic gelatins, respectively, based on their overall charge. Basic bFGF and TGF-β1 were markedly sorbed with time in the acidic gelatin hydrogels, while less sorption took place in the basic gelatin hydrogels. This behavior was explained in terms of an electrostatic interaction between the basic growth factors and the acidic gelatin. However, BMP-2 was sorbed into the acidic gelatin hydrogel to a lesser extent than the other two growth factors, even though its IEP is also greater than 7.0. An in vivo experiment revealed that the acidic gelatin hydrogel was degraded with time, while growth factors were retained in the body for a longer time period as the in vitro sorption to the acidic gelatin hydrogel was larger. These findings indicate that the growth factors sonically complexes to the acidic gelatin hydrogel were released in vivo as a result of hydrogel degradation. Furthermore, animal experiments revealed that the biological performance of growth factors was enhanced by their sustaned release, in marked contrast to the growth factors administered in the solution form.

Effect of Porous Structure on the Degradation of Freeze-Dried Gelatin Hydrogels:

Porous biodegradable hydrogels were prepared by glutaraldehyde crosslinking of acidic gelatin with an isoelectric point of 5.0, followed by freeze-drying. The inner structures of the freeze-dried hydrogels were strongly dependent on the preliminary freezing temperatures prior to freeze-drying. As the freezing temperature was raised, the pore size of the freeze-dried hydrogels increased while their wall thickness decreased. The hydrogels frozen at -20°C swell more rapidly in water than those at 408 and -196°C. In vivo and in vitro degradations of freeze-dried hydrogels with different inner structures were evaluated subcutis in mice and in a collagenase aqueous solution, respectively. The hydrogels prepared at -20°C were more rapidly degraded both in vitro and in vivo, compared to those at -80° and -196°C. The hydrogels freeze-dried at -196°C had the smallest pore size, thickest gelatin walls, and required a longer time for complete degradation. Basic fibroblast growth factor (bFGF) was impregnated into the gelatin hydrogels and the in vitro bFGF release was evaluated. No influence by the inner structure on the release profile of bFGF was observed. Since bFGF is ionically completed with acidic gelatin, it is possible that it was released from these gelatin hydrogels in a similar fashion, irrespective of the hydrogel inner structure.

Bioabsorbable Microspheres for Local Drug Release in the Articulus

Bioabsorbable microspheres incorporating an anti-inflammatory drug, dexamethasone, were prepared from poly(d,l-lactic acid) (PDLLA) using different molecular weights (Mws) by a solvent evaporation method using oil-in-oil emulsion. Two types of PDLLA microspheres with diameter ranges of 40-60 and 90-110 μm were obtained by changing the stirring speed. The microspheres prepared from the higher Mw PDLLA degraded more slowly in phosphate-buffered saline solution (PBS, pH 7.4) than those from the lower Mw PDLLA. The release of incorporated drug into PBS from each PDLLA microspheres was assessed by high performance liquid chromatography. Faster drug release was observed from smaller microspheres. The microspheres prepared from PDLLA with lower Mws released faster the drug than those from PDLLA with higher Mws. Drug concentration in the serum was high immediately after intra-articular injection of dexamethasone in the free form and thereafter decreased with time. In contrast, no drug was detected in the serum of rabbits over the time range studied when injected with dexamethasone-incorporated PDLLA microspheres. This finding indicates that drug release from the microspheres is localized in the articulus.