Surface modifications to promote the osteoconductivity of ultra-high-molecular-weight-polyethylene fabrics for a novel biomimetic artificial disc prosthesis: An in vitro study

Abstract

A novel biomimetic artificial intervertebral disc (bioAID) for the cervical spine was developed, containing a hydrogel core representing the nucleus pulposus, an UHMWPE fiber jacket as annulus fibrosis, and titanium endplates with pins for mechanical fixation. Osseointegration of the UHMWPE fibers to adjacent bone structures is required to achieve proper biomimetic behavior and to provide long-term stability. Therefore, the aim of this study was to assess the osteoconductivity of several surface modifications of UHMWPE fabrics, 2D weft-knitted, using non-treated UHMWPE fibers (N), plasma treated UHMWPE fibers (PT), 10% hydroxy apatite (HA) loaded UHMWPE fibers (10%HA), plasma treated 10%HA UHMWPE fibers (PT-10%HA), 15%HA loaded UHMWPE fibers (15%HA) and plasma treated 15%HA UHMWPE fibers (PT-15%HA). Scanning electron microscopy (SEM) was used for surface characterization. Biological effects were assessed by evaluating initial cell attachment (SEM, DNA content), metabolic activity (PrestoBlue assay), proliferation, differentiation (alkaline phosphatase activity) and mineralization (energy dispersive x-ray, EDX analysis) using human bone marrow stromal cells. Plasma treated samples showed increased initial cell attachment, indicating the importance of hydrophilicity for cell attachment. However, incorporation only of HA or plasma treatment alone was not sufficient to result in upregulated alkaline phosphatase activity (ALP) activity. Combining HA loaded fibers with plasma treatment showed a combined effect, leading to increased cell attachment and upregulated ALP activity. Based on these results, combination of HA loaded UHMWPE fibers and plasma treatment provided the most promising fabric surface for facilitating bone ingrowth.

Keywords: bone ingrowth; in vitro; polyethylene (UHMWPE); spinal implant; surface modification.

FIGURE 1

(A) Biomimetic artificial intervertebral disc (bioAID). (B) Schematic representation of the design of the bioAID and its biomimicry compared to a natural disc

FIGURE 2

Schematic representation of experimental set‐up. Weft‐knitted fabrics made of untreated (N), plasma treated (PT), 10%HA loaded (10%HA), plasma treated 10%HA loaded (PT‐10%HA), 15%HA loaded and, plasma treated 15%HA loaded UHMWPE fibers (PT‐15%HA) were cultured with human bone marrow‐derived stromal cells (hBMSCs) for 14 days. Osteoconductivity was assessed based on cell viability, attachment, osteogenic differentiation, and mineral deposition. Image created with Servier Medical Art in compliance with the terms of the Creative Commons Attribution 3.0 Unported License

FIGURE 3

Surface characterization by scanning electron microscopy of the untreated (N), plasma treated (PT), 10%HA loaded (10%HA), plasma treated 10%HA loaded (PT‐10%HA), 15%loaded and, plasma treated 15%HA loaded fabrics (PT‐15%HA). Scale bar: 25 μm

FIGURE 4

Metabolic activity (Fluorescence, mean ± SD) measured by PrestoBlue assay on day 2, 7, and 14. DNA content (ng, mean ± SD) on day 2 showing initial cell adherence and on day 7 and 14 to assess proliferation over time. Metabolic activity normalized for DNA content (fluorescence/ng, mean ± SD) on day 2,7 and 14. One way ANOVA (*p < .05; **p < .01; ***p < .001)

FIGURE 5

Scanning electron microscopy images of the differently modified UHMWPE surfaces (N: untreated, PT: plasma treated, 10%HA: 10% hydroxyapatite loaded, PT‐10%HA: plasma treated 10%HA loaded, 15%HA: 15% hydroxyapatite loaded, PT‐15%HA: plasma treated 15%HA loaded) on day 2 to visualize cell adherence and cell morphology and on day 14 to show cell morphology and matrix deposition. Representative energy dispersive x‐ray spot analysis of matrix minerals for all groups on day 14 showing the elements of which the spots are composed of. Scale: 50 μm

FIGURE 6

Alkaline phosphatase activity normalized for DNA content as a marker for osteogenic differentiation ([ng/min]/ng, mean ± SD) on day 14. One way ANOVA (*p < .05; **p < .01; ***p < .001)