Fabrication, maturation, and implantation of composite tissue-engineered total discs formed from native and mesenchymal stem cell combinations

Abstract

Low back pain arising from disc degeneration is one of the most common causes of limited function in adults. A number of tissue engineering strategies have been used to develop composite tissue engineered total disc replacements to restore native tissue structure and function. In this study we fabricated a composite engineered disc based on the combination of a porous polycaprolactone (PCL) foam annulus fibrosus (AF) and a hyaluronic acid (HA) hydrogel nucleus pulposus (NP). To evaluate whether native tissue cells or mesenchymal stem cells (MSCs) would perform better, constructs were seeded with native AF/NP cells or with MSCs in the foam and/or gel region. Maturation of these composite engineered discs was evaluated for 9 weeks in vitro culture by biochemical content, histological analysis and mechanical properties. To evaluate the performance of these constructs in the in vivo space, engineered discs were implanted into the caudal spines of athymic rats for 5 weeks. Our findings show that engineered discs comprised of AF/NP cells and MSCs performed similarly and maintained their structure after 5 weeks in vivo. However, for both cell types, loss of proteoglycan was evident in the NP region. These data support the continued development of the more clinically relevant MSCs population for disc replacement applications. STATEMENT OF SIGNIFICANCE: A number of tissue engineering strategies have emerged that are focused on the creation of a composite disc replacement. We fabricated a composite engineered disc based on the combination of a porous foam AF and a HA gel NP. We used these constructs to determine whether the combination of AF/NP cells or MSCs would mature to a greater extent in vitro and which cell type would best retain their phenotype after implantation. Engineered discs comprised of AF/NP cells and MSCs performed similarly, maintaining their structure after 5 weeks in vivo. These data support the successful fabrication and in vivo function of an engineered disc composed of a PCL foam AF and a hydrogel NP using either disc cells or MSCs.

Keywords: Disc cells; Intervertebral disc degeneration; Mesenchymal stem cells; Tissue engineering; Total disc replacement.

Fig. 1.

A scheme of tissue engineering for intervertebral discs. AFCs or MSCs were seeded onto the PCL foam at a density of 2 × 10⁶ cells/construct, whereas NPCs or MSCs were encapsulated in HA at a density of 20 × 10⁶ cells/ml. AF and NP regions were cultured separately in chemically-defined media and combined at 2 weeks. The tissue engineered intervertebral disc constructs were harvested and evaluated at regular intervals over 9 weeks for in vitro study. Additionally, AFCs/NPCs and MSCs/MSCs seeded constructs were implanted into the rat caudal disc space after 5 weeks of priming and after 5 weeks in vivo, constructs were evaluated.

Fig. 2.

SEM images of porous PCL foam for AF regions prepared by salt leaching of mixtures of PCL and salt particles. (A) Overview of the PCL foam (scale = 1 mm) (B) the inner surface (scale = 500 μm) (C) the outer surface (scale = 500 μm) (D) higher magnification image of the surface (scale = 100 μm).

Fig. 3.

Biochemical composition of AFCs or MSCs seeded onto the PCL foam at a density of 2 × 10⁶ cells/construct and NPCs or MSCs encapsulated HA constructs at a density of 20 × 10⁶ cells/ml after 3, 6 and 9 weeks of in vitro culture. (A) GAG content per wet weight of AF region and NP region, n = 4~5. (B) Collagen content per wet weight of AF region and NP region, n = 4~5. (* indicates p < 0.05 vs. 3 W within same group; ^ indicates p < 0.05 vs. 6 W within same group; $ indicates p < 0.05 vs. AF/NP, AF/MSC and MSC/NP within 9 W; # indicates p < 0.05 vs. MSC/NP within 9 W; & indicates p < 0.05 vs. MSC/NP and MSC/MSC within 3 W).

Fig. 4.

Histological staining of AF/NP, AF/MSC, MSC/NP and MSC/MSC engineered disc with time in-vitro culture. (A) Alcian blue, (B) picrosirius red, and (C) Alcian blue/picrosirius red staining (Scale = 500 μm).

Fig. 5.

Compressive mechanical modulus of AF/NP, AF/MSC, MSC/NP and MSC/MSC engineered disc at weeks 4 and 8 of in vitro culture. (* indicates p < 0.05 vs. control; & indicates p < 0.05 vs. MSC/MSC within 8 W, n = 4~5 per group per time point, dashed line indicates acellular engineered disc).

Fig. 6.

Implantation of the engineered disc of MSC/MSC and AF/NP into the rat caudal spine. (A) Intraoperative images of implanted engineered disc into the Cd8/Cd9 disc space. (B) Fluoroscopy images of the vertebral bodies of pre or post-operation. (C) Three-dimensional μCT reconstructions for MSC/MSC and AF/NP engineered disc after 5 weeks implantation. Implanted discs of MSC/MSC and AF/NP did not result intervertebral fusion. Scale = 2 mm.

Fig. 7.

Histological appearance and mechanical properties of implanted engineered discs with native disc after 5 weeks. (A) Histology double staining (alcian blue and picrosirious red) and immunostaining (Type I, II and X collagen) after 5 weeks implantation. Scale = 500 μm. (B) Mechanical properties for vertebra-engineered disc-vertebra. Mechanical properties of MSC/MSC and AF/NP constructs were not significantly different with native discs. (dashed line indicates that properties of native disc).

Fig. 8.

5 weeks following implantation of constructs of MSC/MSC and AF/NP engineered disc in the rat caudal spine. (A) T2 MRI images of implanted discs had a similar structure to native discs. (B) T2 mapping showed reduced signal in the NP for both groups compared to native discs. (C) Quantification at 5 weeks after implantation. No significant difference between AF/NP group and native discs (*:p = 0.009 vs. adjacent control, n = 16 for adjacent control, n = 4 for MSC/MSC, n = 3 for AF/NP).