Lipidic implants for controlled release of bioactive insulin: effects on cartilage engineered in vitro

New insights into process understanding of solid lipid extrusion (SLE) of extruded lipid implants for sustained protein delivery

The aim of this work is a better understanding of solid lipid extrusion (SLE) for protein depot production using a lab-scale twin-screw (tsc)-extruder. In this context, little is known about the relationship of process parameters such as extrusion temperature, screw speed, or formulation on implant characteristics. It is difficult to attribute release characteristics to only one parameter, since the release will always be influenced by a combination of parameters. In this study, we describe the use of an online pressure measurement tool which allows to characterize pressure profiles during an extrusion run. We systematically investigated the impact of various process parameters on implant properties as well as release patterns using a monoclonal antibody (mAb). Solid lipid implants (SLIs) were produced by tsc-extrusion using the low melting triglyceride H12 and the high melting triglyceride Dynasan® D118. A mAb available in a freeze-dried matrix containing hydroxypropyl-β-cyclodextrine (HP-β-CD) was used as incorporated active pharmaceutical ingredient. Extrusion temperature (33-37 °C), screw speed (40-80 rpm) and the lipid composition (30-70% of each triglyceride) were modified. Additionally, freshly extruded SLIs were ground and extruded again as a preparation technique to optimize properties of SLIs. Using the pressure monitoring tool, four characteristic phases were defined for an extrusion run. We found that both, sufficient pressure and adequately molten material, is needed to form a suitable implant. Using the double extrusion technique, release rates could substantially be slowed down without changing formulation.

Long-term release and stability of pharmaceutical proteins delivered from solid lipid implants

Solid lipid implants (SLIs) prepared by twin-screw (tsc) extrusion represent a promising technology platform for the sustained release of pharmaceutical proteins. In this work, we report on two aspects, long-term release and stability of released protein. First, SLIs were produced by tsc-extrusion containing the low melting triglyceride H12 and the high melting triglyceride Dynasan D118. Two different proteins available in a freeze-dried matrix containing hydroxypropyl-β-cyclodextrine (HP-β-CD) were incorporated into the lipid matrix: a monoclonal antibody (mAb) from the IgG₁ class and the f_ab-fragment Ranibizumab (Lucentis®). SLIs, composed of 10% protein lyophilizate and both triglycerides, were extruded at 35°C and 40rpm. Sustained release of both proteins was observed in a sustained manner for approximately 120days. Protein load per implant was increased by three different approaches resulting in a protein load of 3.00mg per implant without affecting the release profiles. The incubation medium containing the released protein was collected, concentrated and analyzed including liquid chromatography (SE-HPLC, IEX, HIC), electrophoresis (SDS-PAGE, on-chip gel electrophoresis) and FT-IR spectroscopy. The mAb showed a monomer loss of up to 7% (SE-HPLC) and IEX analysis revealed the formation of 16% acidic subspecies after 18weeks. FT-IR spectra of mAb indicated the formation of random coil structures towards the end of the release study. Ranibizumab was mainly released in its monomeric form (>95%), and approximately 5% hydrophobic subspecies were formed after 18weeks of release. FT-IR analysis revealed no changes in secondary structure. The release and stability profiles of both proteins underline the potential of SLIs as a delivery system. SLIs provide a promising platform for applications where really long-term release is needed, for example for intraocular delivery of anti-vascular endothelial growth factor (VEGF) drugs for age related macular degeneration (AMD).

Development, characterizations and biocompatibility evaluations of intravitreal lipid implants

Background:

The treatment of posterior eye diseases is always challenging mainly due to inaccessibility of the region. Many drugs are currently delivered by repeated intraocular injections.

Objectives:

The purpose of this study was to investigate the potential applications of natural triglycerides as alternative carriers to synthetic polymers in terms of drug release profile and also biocompatibility for intraocular use.

Materials and Methods:

In vitro/in vivo evaluations of intravitreal implants fabricated from the physiological lipid, glyceride tripalmitate containing clindamycin phosphate as a model drug was performed. The micro-implants with average diameter of 0.4 mm were fabricated via a hot melt extrusion method. The extrudates were analyzed using scanning electron microscopy, differential scanning calorimetry, and in vitro drug dissolution studies. For biocompatibility, the implants were implanted into rabbit eyes. Clinical investigations including fundus observations, electroretinography as well as histological evaluations were performed.

Results:

In vitro tests guaranteed usefulness of the production method for preparing the homogenous mixture of the drug and lipid without affecting thermal and crystalinity characteristics of the components. In vitro releases indicated a bi-phasic pattern for lower lipid ratios, which were completed by the end of day three. With higher lipid ratios, more controlled release profiles were achieved until about ten days for a lipid ratio of 95%. Clinical observations did not show any abnormalities up to two months after implantation into the rabbit eye.

Conclusions:

These results suggest that although the implant could not adequately retard release of the present drug model yet, due to good physical characteristics and in vivo biocompatibility, it can represent a suitable device for loading wide ranges of therapeutics in treatment of many kinds of retinochoroidal disorders.

Insulin is essential for in vitro chondrogenesis of mesenchymal progenitor cells and influences chondrogenesis in a dose-dependent manner

PURPOSE

Insulin is a commonly used additive in chondrogenic media for differentiating mesenchymal stem cells (MSCs). The indispensability of other bioactive factors like TGF-β or dexamethasone in these medium formulations has been shown, but the role of insulin is unclear. The purpose of this study was to investigate whether insulin is essential for MSC chondrogenesis and if there is a dose-dependent effect of insulin on MSC chondrogenesis.

METHODS

We cultivated human MSCs in pellet culture in serum-free chondrogenic medium with insulin concentrations between 0 and 50 μg/ml and assessed the grade of chondrogenic differentiation by histological evaluation and determination of glycosaminoglycan (GAG), total collagen and DNA content. We further tested whether insulin can be delivered in an amount sufficient for MSC chondrogenesis via a drug delivery system in insulin-free medium.

RESULTS

Chondrogenesis was not induced by standard chondrogenic medium without insulin and the expression of cartilage differentiation markers was dose-dependent at insulin concentrations between 0 and 10 μg/ml. An insulin concentration of 50 μg/ml had no additional effect compared with 10 μg/ml. Insulin was delivered by a release system into the cell culture under insulin-free conditions in an amount sufficient to induce chondrogenesis.

CONCLUSIONS

Insulin is essential for MSC chondrogenesis in this system and chondrogenic differentiation is influenced by insulin in a dose-dependent manner. Insulin can be provided in a sufficient amount by a drug delivery system. Therefore, insulin is a suitable and inexpensive indicator substance for testing drug release systems in vitro.

Foamed oligo(poly(ethylene glycol)fumarate) hydrogels as versatile prefabricated scaffolds for tissue engineering

Radically cross-linked hydrogels are frequently used as cell carriers due to their excellent biocompatibility and their tissue-like mechanical properties. Through frequent investigation, PEG-based polymers such as oligo(poly(ethylene glycol)fumarate [OPF] have proven to be especially suitable as cell carriers by encapsulating cells during hydrogel formation. In some cases, NaCl or biodegradable gelatin microparticles were added prior to cross-linking in order to provide space for the proliferating cells, which would otherwise stay embedded in the hydrogel matrix. However, all of these immediate cross-linking procedures involve time consuming sample preparation and sterilization directly before cell culture and often show notable swelling after their preparation. In this study, ready to use OPF-hydrogel scaffolds were prepared by gas foaming, freeze drying, individual packing into bags and subsequent γ-sterilization. The scaffolds could be stored and used "off-the-shelf" without any need for further processing prior to cell culture. Thus the handling was simplified and the sterility of the cell carrier was assured. Further improvement of the gel system was achieved using a two component injectable system, which may be used for homogenous injection molding in order to create individually shaped three dimensional scaffolds. In order to evaluate the suitability of the scaffolds for tissue engineering, constructs were seeded with juvenile bovine chondrocytes and cultured for 28 days. Cross-sections of the respective constructs showed an intense and homogenous red staining of GAG with safranin O, indicating a homogenous cell distribution within the scaffolds and the production of substantial amounts of GAG-rich matrix.

Lipase is essential for the study of in vitro release kinetics from organogels

In vitro drug release studies remain indispensable in the development of drug delivery systems, even if correlations between in vitro and in vivo results are often imperfect. In this work, an improved in vitro analysis method for studying in situ-forming lipid-based implants was developed. More specifically, lipase was found to be an essential additive for evidencing differences in drug release kinetics from organogels of different amino acid-based organogelators, organogelator concentrations, drug loadings, and volumes. Lipases are thought to participate in the degradation of and release from amino acid-based organogel implants in vivo. Our experimental conditions allowed for the rapid and reliable screening of in vitro parameters that may be optimized to slow or accelerate drug release, once preliminary in vivo data are available.

Biodegradable insulin-loaded PLGA microspheres fabricated by three different emulsification techniques: investigation for cartilage tissue engineering

Growth, differentiation and migration factors facilitate the engineering of tissues but need to be administered with defined gradients over a prolonged period of time. In this study insulin as a growth factor for cartilage tissue engineering and a biodegradable PLGA delivery device were used. The aim was to investigate comparatively three different microencapsulation techniques, solid-in-oil-in-water (s/o/w), water-in-oil-in-water (w/o/w) and oil-in-oil-in-water (o/o/w), for the fabrication of insulin-loaded PLGA microspheres with regard to protein loading efficiency, release and degradation kinetics, biological activity of the released protein and phagocytosis of the microspheres. Insulin-loaded PLGA microspheres prepared by all three emulsification techniques had smooth and spherical surfaces with a negative zeta potential. The preparation technique did not affect particle degradation nor induce phagocytosis by human leukocytes. The delivery of structurally intact and biologically active insulin from the microspheres was shown using circular dichroism spectroscopy and a MCF7 cell-based proliferation assay. However, the insulin loading efficiency (w/o/w about 80%, s/o/w 60%, and o/o/w 25%) and the insulin release kinetics were influenced by the microencapsulation technique. The results demonstrate that the w/o/w microspheres are most appropriate, providing a high encapsulation efficiency and low initial burst release, and thus these were finally used for cartilage tissue engineering. Insulin released from w/o/w PLGA microspheres stimulated the formation of cartilage considerably in chondrocyte high density pellet cultures, as determined by increased secretion of proteoglycans and collagen type II. Our results should encourage further studies applying protein-loaded PLGA microspheres in combination with cell transplants or cell-free in situ tissue engineering implants to regenerate cartilage.

The effect of insulin-loaded chitosan particle-aggregated scaffolds in chondrogenic differentiation

Osteochondral defect repair requires a tissue engineering approach that aims at mimicking the physiological properties and structure of two different tissues (cartilage and bone) using a scaffold-cell construct. One ideal approach would be to engineer in vitro a hybrid material using a single-cell source. For that purpose, the scaffold should be able to provide the adequate biochemical cues to promote the selective but simultaneous differentiation of both tissues. In this work, attention was paid primarily to the chondrogenic differentiation by focusing on the development of polymeric systems that provide biomolecules release to induce chondrogenic differentiation. For that, different formulations of insulin-loaded chitosan particle-aggregated scaffolds were developed as a potential model system for cartilage and osteochondral tissue engineering applications using insulin as a potent bioactive substance known to induce chondrogenic differentiation. The insulin encapsulation efficiency was shown to be high with values of 70.37 +/- 0.8%, 84.26 +/- 1.76%, and 87.23 +/- 1.58% for loadings of 0.05%, 0.5%, and 5%, respectively. The in vitro release profiles were assessed in physiological conditions mimicking the cell culture procedures and quantified by Micro-BCA protein assay. Different release profiles were obtained that showed to be dependent on the initial insulin-loading percentage. Further, the effect on prechondrogenic ATDC5 cells was investigated for periods up to 4 weeks by studying the influence of these release systems on cell morphology, DNA and glycosaminoglycan content, histology, and gene expression of collagen types I and II, Sox-9, and aggrecan assessed by real-time polymerase chain reaction. When compared with control conditions (unloaded scaffolds cultured with the standard chondrogenic-inducing medium), insulin-loaded scaffolds upregulated the Sox-9 and aggrecan expression after 4 weeks of culture. From the overall results, it is reasonable to conclude that the developed loaded scaffolds when seeded with ATDC5 can provide biochemical cues for chondrogenic differentiation. Among the tested formulations, the higher insulin-loaded system (5%) was the most effective in promoting chondrogenic differentiation.

Synergistic effects of growth and differentiation factor-5 (GDF-5) and insulin on expanded chondrocytes in a 3-D environment

OBJECTIVE

To investigate the effects of growth and differentiation factor-5 (GDF-5) alone or in combination with insulin on engineered cartilage from primary or expanded chondrocytes during 3-dimensional in vitro culture.

DESIGN

Juvenile bovine chondrocytes were seeded either as primary or as expanded (passage 2) cells onto polyglycolic acid fiber meshes and cultured for 3 weeks in vitro. Additionally, adult human chondrocytes were grown in pellet culture after expansion (passage 2). The culture medium was supplemented either with GDF-5 in varying concentrations or insulin alone, or with combinations thereof.

RESULTS

For primary chondrocytes, the combination of GDF-5 and insulin led to increased proliferation and construct weight, as compared to either factor alone, however, the production of glycosaminoglycans (GAG) and collagen per cell were not affected. With expanded bovine chondrocytes, the use of GDF-5 or insulin alone led to only very small constructs with no type II collagen detectable. However, the combination of GDF-5 (0.01 or 0.1 microg/ml) and insulin (2.5 microg/ml) yielded cartilaginous constructs and, in contrast to the primary cells, the observed redifferentiating effects were elicited on the cellular level independent of proliferation (increased production of GAG and collagen per cell, clear shift in collagen subtype expression with type II collagen observed throughout the construct). The synergistic redifferentiating effects of the GDF-5/insulin combination were confirmed with expanded adult human cells, also exhibiting a clear shift in collagen subtype expression on the mRNA and protein level.

CONCLUSIONS

In combination with insulin, GDF-5 appears to enable the redifferentiation of expanded chondrocytes and the concurrent generation of cartilaginous constructs. The demonstration of these synergistic effects also for adult human chondrocytes supports the clinical relevance of the findings.

Smart biomaterials for tissue engineering of cartilage

Biological regeneration using cartilage tissue engineering in which cells are grown on biomaterial scaffolds and then implanted into the cartilage defects could provide new treatment options for articular cartilage defects. This review aims to give an overview of the wide variety of biomaterials that are currently developed as scaffolds for cartilage tissue engineering. Emphasis will be placed on the current development of the materials that are able to direct cell differentiation and metabolism. These so-called "smart" biomaterials are produced by modifying the physical properties of the scaffolds using peptide sequences and most importantly by developing materials that can deliver proteins to enhance tissue regeneration. Besides providing drug delivery systems, the materials respond to environmental stimuli or release their cargo on cellular demand. However, critical issues remain, such as the transferability of basic science insights to clinical products and the applicability of certain data sets to human patients.

Influence of wettability and surface activity on release behavior of hydrophilic substances from lipid matrices

The aim of this study was to investigate the role of matrix and drug properties on controlled release from triglyceride matrices. Mini-cylinders of 2 mm diameter, 2.2 mm height and 7 mg weight were produced by compression of lipid powder obtained by using a polyethylene glycol (PEG) co-lyophilization method for the model substances lysozyme and FITC-dextran (Mw 4000 Da). Lysozyme was released with decreasing velocity from glyceryl trilaurate, -myristate, -palmitate and -stearate for more than 14 months. Release correlated well with triglyceride lipophilicity defined by the chain length of the fatty acids. Contact angle measurements and the analysis of buffer penetration visualized by confocal microscopy emphasized the role of matrix wettability as a prerequisite for release. A comparison with FITC-dextran revealed that the protein itself enhances matrix wettability and hence its release due to its surface active properties. FITC-dextran remained trapped within the matrix only to be released at lower compression force or after the addition of surfactant. Protein added externally to the release buffer at 0.1% (w/v) was not efficient in lowering the contact angle and increasing the release rate of FITC-dextran. Tween 20 and 81 could be used in different concentrations (0.1, 0.01 and 0.001% (w/v)) to alter lysozyme and FITC-dextran release profiles: resulting release rates showed a close dependence on the contact angle of the respective release medium and triglyceride matrix material. However, both Tweens seem to act not only by reducing the release medium contact angle but also by moderately affecting interparticulate adhesion of the matrix material.

Matrices and scaffolds for protein delivery in tissue engineering

The tissue engineering of functional tissues depends on the development of suitable scaffolds to support three dimensional cell growth. To improve the properties of the scaffolds, many cell carriers serve dual purposes; in addition to providing cell support, cutting-edge scaffolds biologically interact with adhering and invading cells and effectively guide cellular growth and development by releasing bioactive proteins like growth factors and cytokines. To design controlled release systems for certain applications, it is important to understand the basic principles of protein delivery as well as the stability of each applied biomolecule. To illustrate the enormous progress that has been achieved in the important field of controlled release, some of the recently developed cell carriers with controlled release capacity, including both solid scaffolds and hydrogel-derived scaffolds, are described and possible solutions for unresolved issues are illustrated.

Confocal Microscopy for the Elucidation of Mass Transport Mechanisms Involved in Protein Release from Lipid-based Matrices

PURPOSE

It was the aim of this study to identify the governing mechanisms during protein release from cylindrical lipid matrices by visualizing mass transport and correlating the data with in vitro dissolution testing.

MATERIALS AND METHODS

Glyceryl trimyristate cylinders of 2 mm diameter, 2.2 mm height and 7 mg weight were manufactured by compression of a protein-lipid powder mixture prepared by a polyethylene glycol (PEG) co-lyophilization technique. BSA was fluorescence-labeled and the distribution visualized and quantified at different stages of the release process by confocal microscopy in parallel to the quantification in the release buffer. The impact of matrix loading and protein molecular weight was assessed with the model proteins lysozyme, BSA, alcohol dehydrogenase and thyroglobulin.

RESULTS

Buffer penetration and protein release occurred simultaneously from the outer regions of the cylinder progressing towards the center. Release from the top and bottom of the matrix was not negligible but much slower than penetration from the side, probably due to an oriented arrangement of lipid flakes during compression. The different quantification strategies were found to yield identical results. At 6% protein loading, buffer penetration was complete after 4 days, while only 60% of the protein was liberated in that time and release continued up to day 63. Protein release kinetics could be described using the power law equation M ( t ) /M ( infinity ) = kt ( n ) with an average time exponent n of 0.45 (+/-0.04) for loadings varying between 1 and 8%. A percolation threshold at 5% pure protein loading and 3-4% mixed loading (PEG and protein at a 1:1 mass ratio) could be identified. Release rate was found to decrease with increasing molecular weight.

CONCLUSIONS

Protein release from lipid-based matrices is a purely diffusion controlled mechanism. Potential protein stabilization approaches should address the time span between complete buffer penetration of the matrix and 100% release of the remaining loading, which would be exposed to an aqueous environment before leaving the matrix.