Pharmaceutical Research

, Volume 25, Issue 12, pp 2946–2962

A Fibrin Glue Composition as Carrier for Nucleic Acid Vectors

  • Ulrike Schillinger
  • Gabriele Wexel
  • Christian Hacker
  • Martin Kullmer
  • Christian Koch
  • Michael Gerg
  • Stephan Vogt
  • Peter Ueblacker
  • Thomas Tischer
  • Daniel Hensler
  • Jonas Wilisch
  • Joachim Aigner
  • Axel Walch
  • Axel Stemberger
  • Christian Plank
Research Paper

DOI: 10.1007/s11095-008-9719-8

Cite this article as:
Schillinger, U., Wexel, G., Hacker, C. et al. Pharm Res (2008) 25: 2946. doi:10.1007/s11095-008-9719-8

Abstract

Purpose

Gene delivery from biomaterials has become an important tool in tissue engineering. The purpose of this study was to generate a gene vector-doted fibrin glue as a versatile injectable implant to be used in gene therapy supported tissue regeneration.

Methods

Copolymer-protected polyethylenimine(PEI)-DNA vectors (COPROGs), naked DNA and PEI-DNA were formulated with the fibrinogen component of the fibrin glue TISSUCOL® and lyophilized. Clotting parameters upon rehydration and thrombin addition were measured, vector release from fibrin clots was determined. Structural characterizations were carried out by electron microscopy. Reporter and growth factor gene delivery to primary keratinocytes and chondrocytes in vitro was examined. Finally,chondrocyte colonized clots were tested for their potency in cartilage regeneration in a osteochondral defect model.

Results

The optimized glue is based on the fibrinogen component of TISSUCOL®, a fibrin glue widely used in the clinics, co-lyophilized with copolymer-protected polyethylenimine(PEI)- DNA vectors (COPROGs). This material, when rehydrated, forms vector-containing clots in situ upon thrombin addition and is suitable to mediate growth factor gene delivery to primary keratinocytes and primary chondrocytes admixed before clotting. Unprotected PEI-DNA in the same setup was comparatively unsuitable for clot formation while naked DNA was ineffective in transfection. Naked DNA was released rapidly from fibrin clots (>70% within the first seven days) in contrast to COPROGs which remained tightly immobilized over extended periods of time (0.29% release per day). Electron microscopy of chondrocytecolonized COPROG-clots revealed avid endocytotic vector uptake. In situ BMP-2 gene transfection and subsequent expression in chondrocytes grown in COPROG clots resulted in the upregulation of alkaline phosphatase expression and increased extracellular matrix formation in vitro. COPROG-fibrinogen preparations with admixed autologous chondrocytes when clotted in situ in osteochondral defects in the patellar grooves of rabbit femura gave rise to luciferase reporter gene expression detectable for two weeks (n=3 animals per group). However, no significant improvement in cartilage formation in osteochondral defects filled with autologous chondrocytes in BMP-2-COPROG clots was achieved in comparison to controls (n=8 animals per group).

Conclusions

COPROGs co-lyophilized with fibrinogen are a simple basis for an injectable fibrin gluebased gene-activated matrix. The preparation can be used is complete analogy to fibrin glue preparations that are used in the clinics. However, further improvements in transgene expression levels and persistence are required to yield cartilage regeneration in the osteochondral defect model chosen in this study.

KEY WORDS

gene-activated matrix fibrinogengene therapygene transfertissue adhesivetissue engineering

Supplementary material

11095_2008_9719_MOESM1_ESM.doc (24 kb)
ESM 1(DOC 36 KB)
11095_2008_9719_MOESM2_ESM.ppt (290 kb)
Figure 1Copolymer-protected gene vectors (COPROGs) are assembled from polycation compacted DNA particles (here: branched PEI-DNA; box in center of figure) and protective copolymers by electrostatic interaction. The synthetic procedure for protective copolymers is shown here schematically and has been described elsewhere (15). Briefly, 3-(2′-pyridyldithio)-propionic acid (1) is reacted with t-butyl-protected glutamic acid under N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide/1-hydroxybenzotriazole activation. The product is deprotected with trifluoroacetic acid to yield compound (2) which is co-polymerized with O,O’-bis(2-aminoethyl)-poly(ethylene glycol) (here: average molecular weight 6,000 Da) under dicyclohexylcarbodiimide activation. After purification by size exclusion chromatography, the reactive copolymer backbone is reacted with the peptide “YE5C” (sequence [Ac-YE5]2K-ahx-C; ahx = 6-aminohexanoic acid). Product (3) thus consists of a PEG backbone (shaded dark gray) and peptide side chains consisting of an anionic moiety (shaded light gray) linked to the backbone via a spacer (shaded intermediate gray). COPROGs are used to prepare vector-loaded fibrinogen components in the next step (Fig. 2) (PPT 175 KB)
11095_2008_9719_MOESM3_ESM.ppt (121 kb)
Figure 2Vector-loaded fibrinogen and formation cell-colonized fibrin clots. COPROGs in aqueous solution are mixed with the fibrinogen component of a fibrin glue and lyophilized. Before use, the product is rehydrated and mixed with freshly trypsinized cells of interest. A fibrin clot in which cells become transfected in situ is formed upon thrombin addition. This procedure may be used to fill or seal tissue defects with appropriate cells which upon in situ transfection would start to produce a desired growth factor (PPT 175 KB)
11095_2008_9719_MOESM4_ESM.ppt (100 kb)
Figure 3Vector impact on coagulation parameters. Fibrin clot formation was examined by rotation thromboelastography (ROTEG) analysis. Defined amounts of vector formulations were mixed with a fixed amount of fibrinogen solution. Clotting was initiated by thrombin addition. The parameters clotting time (CT; Fig. 3A), maximum clot firmness (MCF; Fig. 3B) and clot formation time (Fig. 3C) were recorded. Shown is the dose-dependent influence of vector formulations on these parameters relative to the values recorded for the control (1 mg/300 µl fibrinogen; clotting initiated by the addition of 0.2 units thrombin in 10 µl). Naked DNA (open squares) has virtually no influence in the examined dose range, unprotected PEI-DNA (open triangles) has little effect on clotting time but a strong impact on MCF and CFT. COPROGs (filled circles) retard clot formation (Fig. 3A) but only have a moderate impact on MCF and CFT up to a ratio of 20 µg DNA per mg fibrinogen. No firm clots are formed at the 30 µg/mg fibrinogen ratio. The data shown are mean values of duplicate measurements. The error bars connect the lower and higher value of the two measurements. The control (0 µg DNA/mg fibrinogen) is the average of 9 measurements, in this case the error bars represent the standard deviation from these 9 measurements (PPT 175 KB)

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  • Ulrike Schillinger
    • 1
  • Gabriele Wexel
    • 1
    • 2
  • Christian Hacker
    • 1
  • Martin Kullmer
    • 1
  • Christian Koch
    • 1
  • Michael Gerg
    • 1
  • Stephan Vogt
    • 2
  • Peter Ueblacker
    • 2
  • Thomas Tischer
    • 2
  • Daniel Hensler
    • 2
  • Jonas Wilisch
    • 2
  • Joachim Aigner
    • 1
  • Axel Walch
    • 3
    • 4
  • Axel Stemberger
    • 1
  • Christian Plank
    • 1
  1. 1.Institute of Experimental OncologyTechnische Universität MünchenMunichGermany
  2. 2.Department of Orthopaedic and Trauma SurgeryTechnische Universität MünchenMunichGermany
  3. 3.Department of PathologyTechnische Universität MünchenMunichGermany
  4. 4.German Research Center for Environmental Health (GmbH)OberschleissheimGermany