|
|
||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online March 20, 2001 as doi:10.1096/fj.00-0564fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Department of Materials and Institute for Biomedical Engineering, ETH-Zurich and University of Zurich, Zurich, Switzerland;
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, USA; and
Department of Bioengineering, Arizona State University, Tempe, Arizona 85287-9709, USA
3Correspondence: Institute for Biomedical Engineering, Moussonstrasse 18, CH-8044 Zurich, Switzerland, E-mail: hubbell{at}biomed.mat.ethz.ch
SPECIFIC AIMS
The aim of this research was to develop an approach to growth factor delivery that would allow growth factors to be stably incorporated within a cell in-growth matrix in a manner such that local enzymatic activity associated with tissue regeneration could trigger growth factor release. This approach would permit the rate of drug release to be varied at different locations within the in-growth matrix based on biological feedback control, i.e., based on cellular activity at that precise location. An example was explored in the context of nerve regeneration with cell-mediated release of beta nerve growth factor (ß-NGF).
PRINCIPAL FINDINGS
ß-NGF fusion proteins were designed and expressed that contained an exogenous transglutaminase cross-linking domain to allow covalent incorporation into fibrin matrices. Fibrin was selected as the cell in-growth matrix, and the transglutaminase activity of factor XIIIa was used to covalently incorporate ß-NGF fusion proteins within fibrin matrices. Two ß-NGF fusion proteins were compared with native ß-NGF in neurite extension assays, one containing and the other lacking a functional plasmin cleavage site between the ß-NGF domain and the cross-linking site.
1. Incorporation of TG-Pi-NGF into fibrin matrices
To determine whether ß-NGF fusion proteins could be covalently
incorporated into fibrin matrices during polymerization, two ß-NGF
fusion proteins were made by recombinant protein expression and
refolded. Each protein contained a cross-linking substrate at the
amino-terminal domain of the protein that consisted of the
transglutaminase (TG) substrate of
2-plasmin
inhibitor, NQEQVSPL (the underlined glutamine being the
reactive site). Each ß-NGF fusion protein also contained the native
ß-NGF sequence at the carboxyl-terminal domain of the protein. One of
two plasmin substrates (P) was placed between the cross-linking
substrate and the ß-NGF domain of the fusion protein, either a
functional plasmin substrate (LIK/MKP, where / denotes the cleavage
site) to yield a protein denoted TG-P-NGF or a nonfunctional plasmin
substrate (LINMKP), in which the lysine residue at the
cleavage site in the plasmin substrate was changed to an asparigine
residue to render the plasmin substrate nonfunctional, yielding a
protein denoted TG-Pi-NGF.
TG-Pi-NGF was incorporated into fibrin matrices
during polymerization. After washing to remove any unbound
TG-Pi-NGF, the matrices were degraded with
plasmin and the degradation products were analyzed by Western blotting.
If the fusion proteins were in fact covalently cross-linked to the
fibrin matrix during polymerization, then fragments of the fibrin
should remain attached to the fusion proteins upon degradation of the
matrix with plasmin, thus resulting in an increase in molecular weight
of TG-Pi-NGF. An increase in molecular weight was
indeed observed for TG-Pi-NGF that were present
during polymerization of fibrin matrices (Fig. 1
), whereas in the case of ß-NGF lacking a cross-linking substrate, no
ß-NGF was observed in the matrix after washing (data not shown). This
result showed directly that the TG-Pi-NGF was
covalently immobilized within fibrin matrices during polymerization via
the transglutaminase activity of factor XIIIa.
|
2. Bioactivity of soluble TG-Pi-NGF and TG-P-NGF
To determine the intrinsic activity of the ß-NGF fusion proteins
made in this study relative to the native ß-NGF, PC12 cell neurite
extension was measured with exposure to the soluble growth factor, not
cross-linked into the matrix during fibrinogen polymerization. PC12
cells were seeded on laminin-coated tissue culture polystyrene, and the
percentage of cells extending neurites was quantified for various
concentrations of native ß-NGF, TG-Pi-NGF, and
TG-P-NGF protein in the culture medium. Native ß-NGF promoted a level
of PC12 cell neurite extension of approximately 65% at 48 h at a
concentration of 100 ng/ml (Fig. 2A
). To
attainthe same level of neurite extension with
TG-Pi-NGF and TG-P-NGF, 250 ng/ml of ß-NGF was
required. This result suggested that the activity of
TG-Pi-NGF and TG-P-NGF is about 40% of native
ß-NGF, but that if the lower activity is corrected for by adding
2.5-fold as much TG-Pi-NGF or TG-P-NGF, similar
levels of neurite extension can be attained. This lower activity may be
due to improper refolding of a subset of the ß-NGF fusion proteins.
|
3. Bioactivity of immobilized TG-Pi-NGF and
TG-P-NGF
To determine the ability of covalently immobilized ß-NGF fusion
protein to be delivered in a controlled manner,
TG-Pi-NGF and TG-P-NGF were incorporated into
fibrin matrices during polymerization and the ability of these matrices
to enhance neurite extension in vitro was assayed using chick dorsal
root ganglia (DRGs). TG-P-NGF was found to enhance neurite extension by
350% vs. unmodified fibrin with no ß-NGF present in the culture
medium and by up to 50% over unmodified fibrin with similar
concentrations of native ß-NGF present in the medium (Fig. 2B
). TG-Pi-NGF, which contained a
nonfunctional plasmin substrate, demonstrated biological activity,
inducing neurite extension that was not statistically different than
that induced by native ß-NGF in the culture medium. However,
TG-P-NGF, which contained a functional plasmin substrate and could be
cleaved from the matrix by cell-associated plasmin activity in a form
very similar to native ß-NGF, was observed to induce neurite
extension much better than even soluble native ß-NGF in the culture
medium. A dose response effect for TG-P-NGF was observed, with an
optimal dose attained when 15 µg/ml of TG-P-NGF was present in the
fibrinogen polymerization mixture. These results demonstrated that both
TG-Pi-NGF and TG-P-NGF were bioactive when
immobilized within fibrin matrices and that TG-P-NGF was especially
highly active, suggesting that it could be released in an active form
by cell-associated matrix degradation. Despite the lower activity of
TG-Pi-NGF and TG-P-NGF in the PC12 cell activity
assay, when the plasmin degradable TG-P-NGF was covalently coupled to
fibrin, it promoted even greater levels of neurite extension than the
same dose of native NGF in the culture medium.
CONCLUSIONS AND SIGNIFICANCE
The objective of this research was to develop a novel approach to growth factor delivery for use in wound healing that would provide localized release in a manner that is triggered by cellular activity. Our approach differs from the most common approaches to protein delivery, where proteins are delivered from polymer-based implants in which the rate of release is controlled by the diffusion of the drug from the delivery system or by the timed degradation of the drug depot. Our goal was to develop an alternate approach that would permit the rate of drug release to be varied at different locations within the delivery system, depending on cellular activity at that location. This approach, using biological feedback control, could be particularly useful when long-term growth factor delivery is desired, where one would like to vary the rate of drug release spatially as a function of regeneration. This type of delivery system might allow a lower total drug dose to be incorporated within the delivery system, and spatial regulation of release could permit a greater percentage of the drug to be released at the time and place of greatest cellular activity.
The work described here demonstrates that this novel approach to protein delivery for tissue engineering is feasible. The results show that one can express ß-NGF fusion proteins that are biologically active and can be covalently immobilized within fibrin matrices by factor XIIIa during polymerization. Our concept provides a simple system that allows for the controlled release of growth factors in response to cellular activity, such as protease activation. The results show that by placing an enzymatically degradable linker between the cross-linking substrate and the growth factor domain in the fusion protein, growth factors can be delivered in an active form in response to cell-regulated processes. Through the selection of the Km and kcat of this enzymatically degradable linker, the rate of growth factor release and enzymatic specificity can presumably be tailored depending on the activity of a particular protease in the wound-healing model of interest. Not only did the results presented above demonstrate that growth factors immobilized within this growth factor delivery system could be released in an active form; moreover, an enhancement of neurite extension was observed with our immobilized growth factor fusion proteins vs. a similar growth factor concentration in the culture medium. This result suggests that the release of immobilized growth factors in a manner that can be temporally and spatially regulated by cell-associated enzymatic processes may be important in the context of wound healing. Thus, delivery systems that allow drug release to be regulated by the progress of wound healing through a cell in-growth matrix could prove to be more effective in promoting successful tissue regeneration.Schematic drawing
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0564fje ; to cite this article, use FASEB J. (March 20, 2001) 10.1096/fj.00-0564fje ![]()
2 Current address: Department of Biomedical Engineering, Washington University, St. Louis MO 63130, USA. ![]()
This article has been cited by other articles:
![]() |
D. J. Geer, D. D. Swartz, and S. T. Andreadis Biomimetic Delivery of Keratinocyte Growth Factor upon Cellular Demand for Accelerated Wound Healing in Vitro and in Vivo Am. J. Pathol., December 1, 2005; 167(6): 1575 - 1586. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. C. Weber, H. Cai, M. Ehrbar, H. Kubota, G. Martiny-Baron, W. Weber, V. Djonov, E. Weber, A. S. Mallik, M. Fussenegger, et al. Effects of Protein and Gene Transfer of the Angiopoietin-1 Fibrinogen-like Receptor-binding Domain on Endothelial and Vessel Organization J. Biol. Chem., June 10, 2005; 280(23): 22445 - 22453. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ehrbar, V. G. Djonov, C. Schnell, S. A. Tschanz, G. Martiny-Baron, U. Schenk, J. Wood, P. H. Burri, J. A. Hubbell, and A. H. Zisch Cell-Demanded Liberation of VEGF121 From Fibrin Implants Induces Local and Controlled Blood Vessel Growth Circ. Res., April 30, 2004; 94(8): 1124 - 1132. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |