A Glutaraldehyde-Free Crosslinking Method for the Treatment of Collagen-Based Biomaterials for Clinical Application

Abstract

Biological bioprostheses such as grafts, patches, and heart valves are often derived from biological tissue like the pericardium. These bioprostheses can be of xenogenic, allogeneic, or autologous origin. Irrespective of their origin, all types are pre-treated via crosslinking to render the tissue non-antigenic and mechanically strong or to minimize degradation. The most widely used crosslinking agent is glutaraldehyde. However, glutaraldehyde-treated tissue is prone to calcification, inflammatory degradation, and mechanical injury, and it is incapable of matrix regeneration, leading to structural degeneration over time. In this work, we are investigating an alternative crosslinking method for an intraoperative application. The treated tissue's crosslinking degree was evaluated by differential scanning calorimetry. To confirm the findings, a collagenase assay was conducted. Uniaxial tensile testing was used to assess the tissue's mechanical properties. To support the findings, the treated tissue was visualized using two-photon microscopy. Additionally, fourier transform infrared spectroscopy was performed to study the overall protein secondary structure. Finally, a crosslinking procedure was identified for intraoperative processing. The samples showed a significant increase in thermal and enzymatic stability after treatment compared to the control, with a difference of up to 22.2 °C and 100%, respectively. Also, the tissue showed similar biomechanics to glutaraldehyde-treated tissue, showing greater extensibility, a higher failure strain, and a lower ultimate tensile strength than the control. The significant difference in the structure band ratio after treatment is proof of the introduction of additional crosslinks compared to the untreated control with regard to differences in the amide-I region. The microscopic images support these findings, showing an alteration of the fiber orientation after treatment. For collagen-based biomaterials, such as pericardial tissue, the novel phenolic crosslinking agent proved to be an equivalent alternative to glutaraldehyde regarding tissue characteristics. Although long-term studies must be performed to investigate superiority in terms of longevity and calcification, our novel crosslinking agent can be applied in concentrations of 1.5% or 2.0% for the treatment of biomaterials.

Keywords: biomaterials; biomedical devices; collagen; crosslinking; glutaraldehyde-free; implantology; pericardium; regenerative medicine; tissue application.

Figure A1

SEM images (representatives of interest) of the serosa (A–C,G–I,M–O,S–U) or the fibrosa (D–F,J–L,P–R,V–X), for GLUT, reference, XOP 1.5%-120 min and XOP 2.0%-120 min, respectively. The images in the left-, middle-, and right columns were realized with 300×, 3000×, and 10,000× magnification, showing a scale bar with dimensions of 30 µm, 3 µm, and 1 µm, respectively.

Figure A1

SEM images (representatives of interest) of the serosa (A–C,G–I,M–O,S–U) or the fibrosa (D–F,J–L,P–R,V–X), for GLUT, reference, XOP 1.5%-120 min and XOP 2.0%-120 min, respectively. The images in the left-, middle-, and right columns were realized with 300×, 3000×, and 10,000× magnification, showing a scale bar with dimensions of 30 µm, 3 µm, and 1 µm, respectively.

Figure A2

Stress-strain curves of the control, GLUT- and with different concentrations of the control-, GLUT- or XOP-treated samples. Standard deviations were calculated within the groups. For display reasons, the stress-strain curves were plotted without a rupture point.

Figure A3

Example of the control group DSC thermogram. The pilot run verified the T_d range, comparing it with previous results [24].

Figure 1

Reaction between the allysine residues and the side chains of hydroxylysine and lysine residues.

Figure 2

Cross-linking reaction between the aldehyde group of GLUT and the ε-amino group of lysine residues results in a covalent bond.

Figure 3

T_onset of the control-, GLUT-, or XOP-treated samples. Not significant (ns) = p > 0.05. ** = p ≤ 0.01. **** = p ≤ 0.0001. Standard deviations were calculated within the groups.

Figure 4

Weight loss of the control-, GLUT-, or XOP-treated samples after challenging them with collagenase. ** = p ≤ 0.01. **** = p ≤ 0.0001. Standard deviations were calculated within the groups.

Figure 5

Protein structure analysis of the tissue’s fibrosa of: (A) Normalized spectra of the second derivative of the recorded spectra between 1600 and 1700 cm⁻¹. (B) The ratio of the band intensities of the control-, GLUT-, or XOP-treated samples. The serosa layers did not show significant differences. ** = p ≤ 0.01. Standard deviations were calculated within the groups.

Figure 6

Images (ROIs) of the SHG signal (collagen) (A,C,E,G,I) and the autofluorescence signal (elastin) + SHG (B,D,F,H,J) for the control, GLUT, reference, XOP 1.5%-120 min, and XOP 2.0%-120 min, respectively. The SHG channels for collagen (820 nm) and the autofluorescence for elastin (525 nm) are shown in red and blue, respectively. The arrows show the fiber alignment before (Control) and after treatment (GLUT or XOP). The treated samples show an autofluorescence signal of the collagen fibers within the elastin channel, resulting in a purple overlay. Therefore, the elastic fibers are less distinguishable in the treated samples. The scale bar shows a dimension of 50 µm.

Figure 6

Images (ROIs) of the SHG signal (collagen) (A,C,E,G,I) and the autofluorescence signal (elastin) + SHG (B,D,F,H,J) for the control, GLUT, reference, XOP 1.5%-120 min, and XOP 2.0%-120 min, respectively. The SHG channels for collagen (820 nm) and the autofluorescence for elastin (525 nm) are shown in red and blue, respectively. The arrows show the fiber alignment before (Control) and after treatment (GLUT or XOP). The treated samples show an autofluorescence signal of the collagen fibers within the elastin channel, resulting in a purple overlay. Therefore, the elastic fibers are less distinguishable in the treated samples. The scale bar shows a dimension of 50 µm.

Figure 7

(A) thickness, (B) failure strain, (C) UTS and (D) collagen phase modulus of the control-, GLUT-, or XOP-treated samples. * = p ≤ 0.05. Standard deviations were calculated within the groups.