Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model

G Model

ARTICLE IN PRESS

YTICE-892; No. of Pages 9

Tissue and Cell xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Tissue and Cell journal homepage: www.elsevier.com/locate/tice

Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model M. Endres a,b , M.L. Zenclussen c , P.A. Casalis c , U. Freymann a , S. Gil Garcia a , J.P. Krueger a,b , U.-W. Thomale d , C. Woiciechowsky c,e , C. Kaps a,∗ a

TransTissue Technologies GmbH, Berlin, Germany Charité-Universitätsmedizin Berlin, Department of Rheumatology, Tissue Engineering Laboratory, Berlin, Germany c Charité-Universitätsmedizin Berlin, Department of Neurosurgery, Berlin, Germany d Charité-Universitätsmedizin Berlin, Department of Pediatric Neurosurgery, Berlin, Germany e Spine Centre Berlin, Germany b

a r t i c l e

i n f o

Article history: Received 21 July 2014 Received in revised form 10 September 2014 Accepted 10 September 2014 Available online xxx Keywords: Intervertebral disc degeneration Polyglycolic acid Hyaluronan Cell-free implant Rabbit model

a b s t r a c t Disc degeneration alters disc height and mechanics of the spinal column and is associated with lower back pain. In preclinical studies gel-like materials or resorbable polymer-based implants are frequently used to rebuild the nucleus pulposus, aiming at tissue regeneration and restoration of tissue function. To compare the outcome of tissue repair, freeze-dried resorbable polyglycolic acid–hyaluronan (PGA/HA) implants without any bioactive components or bioactivated fibrin (fibrin-serum) was used in a degenerated disc disease model in New Zealand white rabbits. Animals with partial nucleotomy only served as controls. The T2-weighted/fat suppression sequence signal intensity in the nuclear region of operated discs as assessed by magnet resonance imaging was reduced in operated compared to healthy discs, indicating loss of water and did not change from week 1 to month 6 after surgery. Quantification of histological and immunohistochemical staining indicated that the implantation of PGA/HA leads to significantly more repair tissue compared to nucleotomy only. Type II collagen content of the repair tissue formed after PGA/HA or fibrin-serum treatment is significantly increased compared to controls with nucleotomy only. The data indicate that intervertebral disc augmentation after nucleotomy has a positive effect on repair tissue formation and type II collagen deposition as shown in the rabbit model. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Back pain, a major public health problem, is strongly associated with degeneration of the intervertebral disc (Luoma et al., 2000). Disc degeneration alters disc height and the mechanics of the rest of the spinal column, which can lead in the long term to spinal stenosis (Urban and Roberts, 2003). As the current clinical approaches are limited to treat symptoms but not the biologic alterations of the disc, new strategies using biomaterials to induce intervertebral disc regeneration are being developed. The use of biomaterials provides the opportunity to restore structure and function of the degenerated intervertebral disc, which is often associated with back pain (Luoma et al., 2000). Many experimental approaches focus on the use of cells embedded in

∗ Corresponding author at: TransTissue Technologies GmbH, Charitéplatz 1, 10117 Berlin, Germany. Tel.: +49 030 91540375; fax: +49 030 450 513 943. E-mail address: [email protected] (C. Kaps).

scaffolds. In these approaches, the cells of the degenerated disc are enriched by adding cells alone or embedded in an adequate scaffold (Urban and Roberts, 2003). Some of these approaches have been shown to be successful in animal models, such as the use of autologous cultured disc-derived chondrocytes in a canine model (Ganey et al., 2003) or the use of annulus fibrosus cells in an atelocollagen scaffold (Sato et al., 2003). Even though some preliminary clinical studies using this methodology show encouraging outcomes (Meisel et al., 2006), there are still some drawbacks. Most cell-based therapies are two-step procedures where in a first step a tissue biopsy is taken and in a second step harvested and expanded cells derived from the biopsy are transplanted into the defect. These two interventions can be stressful for the patient, expensive and timeconsuming, which is seen as a considerable clinical disadvantage of cell-based therapies in nucleus pulposus regeneration (Endres et al., 2010; Omlor et al., 2012). Furthermore, removal of nucleus pulposus tissue as cell source might accelerate degeneration of the host tissue (Fassett et al., 2009). Therefore, attempts are being made to make biopsies as a cell source redundant by focussing on

http://dx.doi.org/10.1016/j.tice.2014.09.003 0040-8166/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

2

techniques for cell migration and recruitment from adjacent tissues (Abbushi et al., 2008). Currently, the use of different matrices and injectable carriers based on gels is in the focus of developments for either cell-based and cell-free nucleus substitutes which support the maintenance of disc height and support cell migration and tissue regeneration (Henriksson et al., 2011). It has been shown that the use of appropriate gel materials has the potential to suppress undesirable angiogenesis in vitro (Scholz et al., 2010). In general, major disadvantages of gels are their fluid-like behaviour with no initial stability and the potential risk of extrusion when injected into the nucleus compartment after nucleotomy (Omlor et al., 2012). The use of a biocompatible cell-free polyglycolic acid/hyaluronan (PGA/HA) scaffold with its elastic properties might solve the problems of extrusion and handling. Therefore our aim was to compare a typical, clinically approved gel bioactivated with allogenic serum (fibrin-serum) with a PGA/HA implant without further bioactivation with serum in a regenerative approach with regard to quantity of repair tissue formation in an existing rabbit discectomy model (Abbushi et al., 2008; Endres et al., 2010). 2. Materials and methods 2.1. Preparation of cell-free disc implants Resorbable pure polyglycolic acid (PGA) scaffolds (BioTissue AG, Switzerland) of 10 mm × 10 mm × 1.1 mm were loaded with 10 mg/ml hyaluronic acid (HA) (Ostenil, TRB Chemedica AG, Germany) as described previously (Abbushi et al., 2008; Endres et al., 2010). The implants were freeze-dried for 16 h under sterile conditions using a lyophilisator (Leybold-Heraeus, Germany) and stored in a desiccator at room temperature. Prior to implantation, cell-free implants were cut in 5 mm × 5 mm pieces and allowed to re-hydrate in physiological saline. 2.2. Surgery and MRI evaluation Animal care and experimental procedures were followed according to institutional guidelines and conformed to the requirements of the state authority for animal research conduct (LaGeSo No. 0071/06, Berlin). In all 13 rabbits, partial nucleotomy with removal of approx. 50% of the nucleus pulposus was performed on L6/L7 or L7/S1 level using a retroperitoneal approach. Animals were divided into three groups. In one group of animals (n = 5), cell-free PGA/HA scaffolds rehydrated in physiological saline were implanted into the disc defect. In the second group (n = 5) animals were treated with 0.5 ml fibrin (Tissucol Duo S; Baxter, Germany) mixed with 10% allogeneic rabbit serum (Kraeber &Co GmbH, Germany). For this purpose, allogeneic rabbit serum was added to the fibrinogen component resulting in a final concentration of 10% (v/v) and filled into a syringe. Fibrinogen/serum and thrombin was applied to the application system provided together with the fibrin sealant components. In the third group (n = 3) animals with partial nucleotomy only served as controls (named sham operated group). To reduce a selection bias and to diminish possible effects from a learning curve, implant surgeries and controls were performed alternately. One week and 6 months after surgery, the T2-weighted/fat suppression sequence signal intensity index was determined by two independent investigators using magnetic resonance imaging (MRI). MR images were taken using a 1.5 T imager (GE Twin Speed 1.ST, General Electric, Milwaukee, USA). The rabbits were positioned in prone position on a quadrature surface coil, and sagittal T2 weighted/fat suppression sequence spin echo images parallel to the lumbar spine were obtained. The signal intensity levels of

Fig. 1. Magnetic resonance images of the lumbar spine of rabbits with operated disc (white arrow) 1 week (A, C, E) and 6 months (B, D, F) after surgery. After partial nucleotomy either the PGA/HA implant (A, B) or fibrin-serum (C, D) was implanted. The sham operated group received only partial nucleotomy (E, F). Evaluation of signal intensity was performed in a defined region of interest in the centre of the disc (G, red oval). Signal intensity index revealed no significant differences between the groups (H). The mean is given and error bars represent the standard deviation (SD). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

3

Fig. 2. Overview of explanted intervertebral disc segments (hematoxylin/eosin staining). Six months after nucleotomy and implantation of PGA/HA (A) or fibrin-serum (B), the discs showed heights comparable to the heights of native discs (D). In contrast, discs from the nucleotomy group collapsed (C).

healthy and operated discs were quantified using Osirix software (Osirix Medical Imaging Software, Version 2.7.5). An oval region of interest of 0.015 cm2 (ROI) was defined in the centre of the particular disc of interest shown in T2 weighted sagittal/fat suppression sequence images, and signal intensities within the ROI were measured. The signal intensity index was calculated by dividing the measured signal intensity of the operated disc by the signal intensity of a healthy disc from the same MR image. The signal intensity indices were analysed by two independent investigators (radiologist and biologist). Intervertebral disc (IVD) segments were explanted 6 months after surgery following MRI and investigated histologically and immunohistochemically. 2.3. Histological and immunohistochemical staining Explanted IVD segments were fixed in 4% formalin, cut into three coronal slices of 5 mm thickness, decalcified in 0.05% EDTA for 6 weeks and subsequently embedded in paraffin. The paraffin blocks were sectioned into 6 ␮m sections from ventral to dorsal using a microtome. For histological and immunohistochemical staining, sections from the PGA/HA implant group (n = 5) and fibrin/serum implant group (n = 4) were used. Sample size of the nucleotomy group (n = 3) was increased by using sections from the nucleotomy group of a previous study (n = 5, Abbushi et al., 2008). Nucleotomy in

the present study and in the previous study were both performed by the same surgeon according to the same protocol. Explants were embedded in paraffin and sections were performed in the same manner as stated and handled by the same technician. For staining, all sections were prepared and handled at the same time as follows. The sections were stained with hematoxylin and eosin (H&E) to identify annulus fibrosus and nucleus pulposus region or with safranin O for detection of proteoglycans. For immunohistochemical analysis of type I collagen, associated with fibrous cartilage and type II collagen, associated with hyaline cartilage (Caron et al., 2012), sections (6 ␮m) were incubated for 40 min either with monoclonal anti-type I collagen antibody (Abcam, UK) or with polyclonal anti-type II collagen antibody (Leica Novocastra, UK). Subsequently, sections were processed using the Crystal AP-Polymer System (Innovative Diagnostik, Germany) for the detection of anti-type I collagen antibodies or Powervision PolyAP (Leica Novocastra, UK) for the detection of anti-type II collagen antibodies according to the manufacturer’s instructions, followed by counterstaining with hematoxylin (Merck, Germany). Histomorphometric analysis of sections from all groups (n = 5 PGA/HA implant group, n = 4 fibrin/serum implant group and n = 8 nucleotomy group) made from the central area of the respective IVD segment was performed using the Photoshop software (Adobe Systems Incorporated, USA) to quantify the amount of repair tissue and type II collagen content within the newly formed repair tissue. Repair tissue was defined as cell-rich tissue with typical cell

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9 4

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

Fig. 3. Hematoxylin/eosin staining of the central area of decalcified intervertebral discs 6 months after surgery, obtained from animals receiving PGA/HA implants (A, B close up), fibrin-serum (C, D close up) or partial nucleotomy only (E, F close up) and native intervertebral disc (G, H close up). Cell clusters indicating tissue remodelling and regeneration are marked with black arrows. The black box indicates the area of higher magnification. The distance in ␮m is given for the whole scale bar.

cluster formation, which indicated tissue remodelling processes. Quantification was performed by two researchers independently as follows. A standardized region of interest (ROI) was defined. This region covers the central part of the IVD between the two bony endplates. For histomorphometric analysis the repair tissue and type II collagen positively stained tissue was selected using the Photoshop magnetic lasso tool and the histogram function. The percentage of newly formed repair tissue and type II collagen positively stained

tissue was calculated by dividing the number of pixels in the repair tissue and type II collagen positively stained tissue by the number of pixels found in the ROI. Statistical analysis of histomorphometric data, representing the amount of repair tissue and type II collagen content was performed using One-Way ANOVA with Bonferroni post hoc analysis (SigmaStat 3.5 Software, Systat Software GmbH, Germany). Differences showing a p value lower than 0.05 were considered significant.

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

5

Fig. 4. Safranin O/Fast green staining of the central area of decalcified intervertebral discs 6 months after surgery, obtained from animals receiving PGA/HA implant (A, B close up), fibrin-serum (C, D close up) or partial nucleotomy only (E, F close up) and native intervertebral disc (G, H close up). The black box indicates the area of higher magnification. All discs revealed tissue rich in proteoglycans. The distance in ␮m is given for the whole scale bar.

3. Results Out of 13 rabbits, one rabbit in the fibrin/serum implanted group died due to problems after anaesthesia one day after surgery. The T2 weighted/fat suppression sequence signal intensity of sham operated and implanted animals was measured after 1 week and 6 months after surgery (Fig. 1). One week after surgery, we saw in all groups (sham operated and implant groups) an almost black disc as a result of partial nucleotomy and loss of water and/or

proteoglycans (Fig. 1A, C and E). After 6 months discs of rabbits which received the PGA/HA implant (Fig. 1B), fibrin/serum (Fig. 1D) or only partial nucleotomy (Fig. 1F) showed no significant changes in signal intensity. The signal intensity index was calculated by dividing the measured signal intensity of the operated disc (Fig. 1G, lower red oval) by the signal intensity of a healthy disc from the same MR image (Fig. 1G, upper red oval). A signal intensity index of 1.0 represents the signal intensity of the nucleus pulposus of a healthy disc with a

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9 6

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

Fig. 5. Quantification of repair tissue in the operated discs (*p < 0.05). Computer-assisted histomorphometric analysis of the repair tissue (A). The mean is given and error bars represent the standard deviation (SD). Representative image of a disc showing the mask including the region of interest (ROI, green mask) and the repair tissue (RT, yellow mask) for quantification of tissue regeneration in the central part of the operated intervertebral discs 6 months after surgery (B). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of the article.)

high amount of liquid in the nucleus pulposus (measured in healthy discs). No signal (index of 0.0) represents a degenerated disc without liquid (“black disc”). There was no significant increase in the signal intensity index values of PGA/HA and fibrin/serum from week 1 to month 6 after implantation and compared to controls. (Fig. 1H). An overview of intervertebral discs treated with PGA/HA (Fig. 2A) or fibrin/serum (Fig. 2B) after nucleotomy showed similar heights comparable to native discs (Fig. 2D). The decreased height in discs of the nucleotomy group indicated a collapse of the intervertebral disc space after nucleotomy (Fig. 2C). H&E staining (n = 3 per animal) of the central part of the operated discs showed newly formed repair tissue in the PGA/HA group (Fig. 3A and B) and fibrin/serum group (Fig. 3C and D), while the sham operated group showed negligible low amounts of newly formed repair tissue after 6 months (Fig. 3E and F). Cells within the repair tissue showed normal disc chondrocyte morphology and formed typical round or elongated cell clusters indicating tissue remodelling and regeneration (Fig. 3B and D arrow). In contrast, the group, which received only partial nucleotomy showed a collapsed annulus with only few signs of tissue regeneration (Fig. 3E and F). In contrast to the operated discs, native discs showed a more loose structure of the nucleus pulposus tissue (Fig. 3G and H). There were no signs of necrosis, inflammation, foreign body reaction or malignant transformation in the operated discs. Residual fibres of the PGA–HA implant were not evident. Safranin O staining revealed red stained sulphated glucosaminoglycans in the annulus tissue and in the repair tissue found after PGA/HA (Fig. 4A and B) and fibrin-serum implantation (Fig. 4C and D) as well as in the residual collapsed tissue of the sham group (Fig. 4E and F) and in the native disc (Fig. 4G and H). Histomorphometric analysis of the central disc region (Fig. 5A and B) showed a significantly (p < 0.05) increased amount of repair tissue in the PGA/HA group (60.4%) compared to the sham group (43.9%). There was no significant difference in the amount of repair tissue between the fibrin-serum group (55.2%) and the sham group (43.9%). Immunohistochemical staining showed the presence of type I collagen in the repair tissue after implantation of PGA/HA implant (Fig. 6A), fibrin-serum (Fig. 6C), and only little in the collapsed tissue of the sham group (Fig. 6E). Type II collagen could be detected in the repair tissue after implantation of the PGA/HA implant (Fig. 6B) or fibrin-serum (Fig. 6D), while the sham group showed marginal amounts of type II collagen (Fig. 6F). Controls gave no signal (Fig. 6G and H). The amount of type II collagen in the repair tissue was assessed by histomorphometric analysis (Fig. 7). Operated discs of

both implant groups revealed a repair tissue with a significantly increased type II collagen content (60.1% PGA/HA and 70.6% fibrinserum) compared to the sham group (39.4%). 4. Discussion In this study we used a non-woven polyglycolic acid (PGA) scaffold with lyophilized hyaluronan in the first instance to augment the nucleus pulposus and secondly to provide a three-dimensional environment for migrating cells from the surrounding tissue and support repair tissue formation by these cells. The restoration of spinal stability after nucleotomy and subsequent implantation of such PGA/HA scaffolds has been shown in biomechanical tests in intact lumbar spines of calves (Hegewald et al., 2009). Furthermore, a combination of autologous serum and PGA/HA implants has been shown to improve disc height and to form repair tissue after nucleotomy in a lapine and ovine model (Abbushi et al., 2008; Endres et al., 2010; Woiciechowsky et al., 2012). As shown here without the use of allogenic serum or other additional bioactive factors, the implantation of PGA/HA after partial nucleotomy resulted in significantly increased repair tissue formation compared to fibrin-serum implants and the sham group. It has been reported that the non-woven resorbable PGA scaffold with its high porosity of nearly 90% supports cell (re-) differentiation and extracellular matrix production of articular chondrocytes in vitro and in a mouse model (Endres et al., 2007). Other groups favour gels due to similarities with the consistency of native nucleus pulposus for the augmentation or restoration of this tissue (Gruber et al., 2006; Cloyd et al., 2007; Benz et al., 2012), whereas cells have to be injected together with the gel for the purpose of regeneration. Another group developed a precipitated gel which can form a bio adhesive network by blending different components (Vernengo et al., 2010). Although some of these gels bond and form a viscoelastic implant, the question arises whether these gels with or without cells will stay in place if axial loading, flexion and extension as in daily living will occur. In a clinical study, were a hydrogel encased in a polyethylene implant was used, migration of these implants have been observed in 10% of all treated patients (Shim et al., 2003). A preclinical study with gel implants consisting of semihydrated poly vinyl alcohol (PVA hydrogel) in baboons reported high rates of extrusion depending on the surgical approach (20% in a posterolateral approach to 33% in an anterior approach) (Allen et al., 2004). Additionally, the use of an implant for nucleus reconstruction consisting of collagen type I showed extrusion during biomechanical testing in 50% of all cases, which could not be prevented by the use of fibrin or acryl glue or

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

7

Fig. 6. Immunohistochemical staining of repair tissue for type I collagen (A, C, E, G) and type II collagen (B, D, F, H) 6 months after partial nucleotomy with implantation of PGA/HA (A, B), fibrin/serum (C, D), and partial nucleotomy only (E, F). IgG antibody isotype controls showed no staining (G, H). The implanted groups showed a cell-rich homogeneous tissue with a high content of type II collagen and cell cluster indicating tissue remodelling and regeneration. The group with partial nucleotomy only showed almost no type II collagen and no cell clusters. The distance in ␮m is given for the whole scale bar.

annulus suture (Heuer et al., 2008). Also in situ implantation of PGA/HA nucleus implants into lumbar calve spines for biomechanical testing without fixation revealed implant herniation during extensive flexion and bending (Hegewald et al., 2009). To reduce implant migration and extrusion, fixation is essential for safety of such treatments. In contrast to hydrogel implants, non-woven scaffolds have the advantage that fixation by pins or sutures is feasible. It has been shown that polymer-based scaffolds made of PLGA (poly

(lactic-co-glycolic acid)) or PGA allow different fixation techniques in articular cartilage defects (Knecht et al., 2007). Besides glueing with fibrin sealant, transosseous suture and chondral suture is an appropriate technique to fix the implant into the defect, whereas sutures obviously withstand higher tensile load from 26 N (chondral suture) to 38 N (osteochondral suture) compared to sealant (approximately 1–2 N) (Knecht et al., 2007). In the same study, the PGA matrix was superior to other known implant materials

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9 8

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

Fig. 7. Histomorphometric analysis of the repair tissue of the operated intervertebral discs 6 months after surgery. Quantification of type II collagen-rich extracellular matrix (*p < 0.05). The mean is given and error bars represent the standard deviation (SD).

for cartilage defect cover such as PLGA (approx. 15 N) and collagen membranes (approx. 9 N). The ultimate failure load, yield load and stiffness of different fixation techniques of PGA matrices were determined in vitro in a shear force scenario. Results showed that the pin fixation technique and the transosseous fixation technique had significantly higher ultimate failure load, yield load, and stiffness than the conventional suture technique for the fixation of textile implants in cartilage defects (Zelle et al., 2007). A promising method to prevent herniation or migration of nucleus implants might be the simultaneous use of PGA/HA for annulus augmentation and annulus closure (Hegewald et al., 2009). In animal studies using gel-like implants, nucleotomy as well as gel implantation was performed using a needle (Cloyd et al., 2007; Benz et al., 2012). This application method does not represent the surgical procedure in clinical routine. In this study we used the method of microdisectomy which is to date the gold standard of surgical treatment for lumbar disc herniation (Postacchini and Postacchini, 2011) and is believed to be closer to the surgical routine by comprising damage to the annulus fibrosus. Nevertheless, performing partial nucleotomy with simultaneous implantation of either PGA/HA or fibrin/serum does not consider progressive disc degeneration over time which is a limitation of the study. Nucleus pulposus tissue is characterized by its high amount of aggrecan and presence of collagen type II and it is highly hydrated. In contrast, annulus fibrosus is composed predominantly of highly organized collagen type I fibres, which are radial distributed in the outer annulus fibrosus and are more and more replaced by type II collagen in the inner annulus fibrosus (Eyre and Muir, 1976, 1977). The collagen fibres confine hydrated proteoglycans which enable the intervertebral disc to resist compressive forces and provide tensile strength (Sivan et al., 2013). Qualitative analysis of proteoglycans revealed the presence of these water-binding proteins in the repair tissue in all groups and in the surrounding annulus fibrosus tissue. This corresponds to the MRI scans 6 months after surgery, which revealed no differences in the signal intensity after nucleotomy, nucleotomy and treatment with fibrinserum and nucleotomy and treatment with PGA/HA. Although the implantation of fibrin-serum leads to repair tissue formation in the nucleus pulposus region with type II collagen deposition in the newly developed tissue matrix, the results after implantation of PGA/HA showed a significantly increased amount of repair tissue. Along with previous data, which showed that the use of allogeneic serum together with PGA/HA implants leads to repair tissue rich in type II collagen in the nucleus region (Endres et al., 2010), the three dimensional PGA/HA implant increases the quantity of repair tissue in our rabbit model. Even though a low number of

individuals was used for each group, which is another limitation of this study, the results show that the use of a resorbable PGA/HA implant supports the self-healing attempts of the body by providing a three-dimensional environment for ingrowing cells and subsequent matrix production and repair tissue formation. The high amount of cells within the repair tissue is a result of cell migration into the implant since no cells were implanted together with PGA/HA implants or fibrin-serum, which has been observed before (Abbushi et al., 2008). Mechanisms of human cell in-growth into PGA/HA implants are not yet fully understood especially cell type and origin are unknown or highly speculative, which is a limitation of the study. Most likely, cells from surrounding tissue such as residual nucleus pulposus, annulus fibrosus, notochordal cells (Risbud and Shapiro, 2011) or mesenchymal progenitor cells (Saraiya et al., 2010) might be able to migrate and/or differentiate to form repair tissue in scaffolds. The origin of progenitor cells with migration potential may be the intervertebral disc region adjacent to the epiphyseal plate and the outer zone of the annulus fibrosus, where stem cell niches are proposed (Henriksson et al., 2012). A recent study demonstrated the ability of a distinct cell population isolated from intervertebral discs to migrate through fibrous tissue towards and into intervertebral disc explants in vitro (Henriksson et al., 2013). Various activators for cell migration have been identified. Several chemokines (Hegewald et al., 2012), extracellular matrix proteins (Thibault et al., 2007), growth factors (Mishima and Lotz, 2008) and human serum (Haberstroh et al., 2009) are known to attract IVD cells and/or mesenchymal stem cells in modified Boyden chamber assays. Furthermore, in situ recruitment of mesenchymal stem cells by hyaluronan, a major component of the PGA/HA implants has been demonstrated (Erggelet et al., 2007). In contrast to humans, it has been demonstrated that in rabbits notochordal cells persist throughout life in the nucleus pulposus, but the number of this cells decreases with growth and are replaced by hyaline cartilage-like matrix and more chondrocyte-like NP cells infiltrating from the annulus fibrosus (Kim et al., 2003). Whether this decrease is caused by differentiation of notochordal cells into more differentiated chondrocyte-like NP cells is still unknown. Moreover, the reason for the decrease of notochordal cells and the origin of the chondrocyte-like NP cells remain to be elucidated. However, cell labelling and cell tracking experiments with cells of specific origin might be useful to clarify, whether nucleus pulposus, annulus fibrosus and/or progenitor cells from a stem cell niche migrate into PGA/HA implants and which cells are responsible for repair tissue formation in vivo. Although commonly used as a standard animal model in the field of IVD degeneration, rabbits as well as all quadruped animals share the same limitation due to the unique upright position of the human spine. It is assumed that the spines of small quadrupeds such as rabbits are probably loaded by much lower forces than the human spines. However, their intradiscal pressure might be similar to the human since the diameter of their discs on which this force is acting is much smaller (Alini et al., 2008). Nevertheless, it has been shown that puncturing the annulus fibrosus of rabbits IVDs with a needle of defined gauge resulted in reproducible, degenerative changes within 2–8 weeks that could be quantitatively assessed by conventional radiography and magnetic resonance imaging (MRI), as well as histology. Another disadvantage of the rabbit model might be the persistence of notochordal cells until skeletal maturity (approx. 6 months) or beyond (Hunter et al., 2003). Newer studies underline the hypothesis that not only in rabbits but also in humans, nucleus pulposus tissue retains notochordal cells throughout life (Risbud et al., 2010). Besides these disadvantages, the rabbit disc model is still relevant as proof of concept or for pre-clinical pilot studies due to its cost effectiveness (Alini et al., 2008). In conclusion, the use of PGA/HA implants for nucleus pulposus augmentation in the lumbar spine leads to an increase of the

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003

G Model YTICE-892; No. of Pages 9

ARTICLE IN PRESS M. Endres et al. / Tissue and Cell xxx (2014) xxx–xxx

amount of newly formed tissue compared to nucleotomy only and supports repair tissue formation with type II collagen deposition. Conflict of interest M.E., C.K., J.P.K., S.G., U.F. are employees of the company TransTissue Technologies GmbH. The company is active in the area of tissue regeneration (Research and Development). Acknowledgements This study was supported by the Investitionsbank Berlin (IBB) and the Europäischen Fonds für regionale Entwicklung (EFRE) (DiscTissue grants: 10138665 and 10138707). References Abbushi, A., Endres, M., Cabraja, M., Kroppenstedt, S.N., Thomale, U.W., Sittinger, M., Hegewald, A.A., Morawietz, L., Lemke, A.J., Bansemer, V.G., Kaps, C., Woiciechowsky, C., 2008. Regeneration of intervertebral disc tissue by resorbable cell-free polyglycolic acid-based implants in a rabbit model of disc degeneration. Spine (Phila Pa 1976) 33 (14), 1527–1532. Alini, M., Eisenstein, S.M., Ito, K., Little, C., Kettler, A.A., Masuda, K., Melrose, J., Ralphs, J., Stokes, I., Wilke, H.J., 2008. Are animal models useful for studying human disc disorders/degeneration? Eur. Spine J. 17 (1), 2–19. Allen, M.J., Schoonmaker, J.E., Bauer, T.W., Williams, P.F., Higham, P.A., Yuan, H.A., 2004. Preclinical evaluation of a poly (vinyl alcohol) hydrogel implant as a replacement for the nucleus pulposus. Spine (Phila Pa 1976) 29 (5), 515–523. Benz, K., Stippich, C., Osswald, C., Gaissmaier, C., Lembert, N., Badke, A., Steck, E., Aicher, W.K., Mollenhauer, J.A., 2012. Rheological and biological properties of a hydrogel support for cells intended for intervertebral disc repair. BMC Musculoskelet. Disord. 13 (54). Caron, M.M., Emans, P.J., Coolsen, M.M., Voss, L., Surtel, D.A., Cremers, A., van Rhijn, L.W., Welting, T.J., 2012. Redifferentiation of dedifferentiated human articular chondrocytes: comparison of 2D and 3D cultures. Osteoarthritis Cartilage 20 (10), 1170–1178. Cloyd, J.M., Malhotra, N.R., Weng, L., Chen, W., Mauck, R.L., Elliott, D.M., 2007. Material properties in unconfined compression of human nucleus pulposus, injectable hyaluronic acid-based hydrogels and tissue engineering scaffolds. Eur. Spine J. 16 (11), 1892–1898. Endres, M., Neumann, K., Schroder, S.E., Vetterlein, S., Morawietz, L., Ringe, J., Sittinger, M., Kaps, C., 2007. Human polymer-based cartilage grafts for the regeneration of articular cartilage defects. Tissue Cell 39 (5), 293–301. Endres, M., Abbushi, A., Thomale, U.W., Cabraja, M., Kroppenstedt, S.N., Morawietz, L., Casalis, P.A., Zenclussen, M.L., Lemke, A.J., Horn, P., Kaps, C., Woiciechowsky, C., 2010. Intervertebral disc regeneration after implantation of a cell-free bioresorbable implant in a rabbit disc degeneration model. Biomaterials 31 (22), 5836–5841. Erggelet, C., Neumann, K., Endres, M., Haberstroh, K., Sittinger, M., Kaps, C., 2007. Regeneration of ovine articular cartilage defects by cell-free polymer-based implants. Biomaterials 28 (36), 5570–5580. Eyre, D.R., Muir, H., 1976. Types I and II collagens in intervertebral disc. Interchanging radial distributions in annulus fibrosus. Biochem. J. 157 (1), 267–270. Eyre, D.R., Muir, H., 1977. Quantitative analysis of types I and II collagens in human intervertebral discs at various ages. Biochim. Biophys. Acta 492 (1), 29–42. Fassett, D.R., Kurd, M.F., Vaccaro, A.R., 2009. Biologic solutions for degenerative disk disease. J. Spinal Disord. Tech. 22 (4), 297–308. Ganey, T., Libera, J., Moos, V., Alasevic, O., Fritsch, K.G., Meisel, H.J., Hutton, W.C., 2003. Disc chondrocyte transplantation in a canine model: a treatment for degenerated or damaged intervertebral disc. Spine (Phila Pa 1976) 28 (23), 2609–2620. Gruber, H.E., Hoelscher, G.L., Leslie, K., Ingram, J.A., Hanley Jr., E.N., 2006. Threedimensional culture of human disc cells within agarose or a collagen sponge: assessment of proteoglycan production. Biomaterials 27 (3), 371–376. Haberstroh, K., Enz, A., Zenclussen, M.L., Hegewald, A.A., Neumann, K., Abbushi, A., Thome, C., Sittinger, M., Endres, M., Kaps, C., 2009. Human intervertebral discderived cells are recruited by human serum and form nucleus pulposus-like tissue upon stimulation with TGF-beta3 or hyaluronan in vitro. Tissue Cell 41 (6), 414–420. Hegewald, A.A., Knecht, S., Baumgartner, D., Gerber, H., Endres, M., Kaps, C., Stussi, E., Thome, C., 2009. Biomechanical testing of a polymer-based biomaterial for the restoration of spinal stability after nucleotomy. J. Orthop. Surg. Res. 4 (25). Hegewald, A.A., Neumann, K., Kalwitz, G., Freymann, U., Endres, M., Schmieder, K., Kaps, C., Thome, C., 2012. The chemokines CXCL10 and XCL1 recruit human annulus fibrosus cells. Spine (Phila Pa 1976) 37 (2), 101–107.

9

Henriksson, H., Hagman, M., Horn, M., Lindahl, A., Brisby, H., 2011. Investigation of different cell types and gel carriers for cell-based intervertebral disc therapy, in vitro and in vivo studies. J. Tissue Eng. Regen. Med. 6 (9), 738–747. Henriksson, H.B., Svala, E., Skioldebrand, E., Lindahl, A., Brisby, H., 2012. Support of concept that migrating progenitor cells from stem cell niches contribute to normal regeneration of the adult mammal intervertebral disc: a descriptive study in the New Zealand white rabbit. Spine (Phila Pa 1976) 37 (9), 722–732. Henriksson, H.B., Lindahl, A., Skioldebrand, E., Junevik, K., Tangemo, C., Mattsson, J., Brisby, H., 2013. Similar cellular migration patterns from niches in intervertebral disc and in knee-joint regions detected by in situ labeling: an experimental study in the New Zealand white rabbit. Stem Cell Res. Ther. 4 (5), 104. Heuer, F., Ulrich, S., Claes, L., Wilke, H.J., 2008. Biomechanical evaluation of conventional anulus fibrosus closure methods required for nucleus replacement. Laboratory investigation. J. Neurosurg. Spine 9 (3), 307–313. Hunter, C.J., Matyas, J.R., Duncan, N.A., 2003. The notochordal cell in the nucleus pulposus: a review in the context of tissue engineering. Tissue Eng. 9 (4), 667–677. Kim, K.W., Lim, T.H., Kim, J.G., Jeong, S.T., Masuda, K., An, H.S., 2003. The origin of chondrocytes in the nucleus pulposus and histologic findings associated with the transition of a notochordal nucleus pulposus to a fibrocartilaginous nucleus pulposus in intact rabbit intervertebral discs. Spine (Phila Pa 1976) 28 (10), 982–990. Knecht, S., Erggelet, C., Endres, M., Sittinger, M., Kaps, C., Stussi, E., 2007. Mechanical testing of fixation techniques for scaffold-based tissue-engineered grafts. J. Biomed. Mater. Res. B: Appl. Biomater. 83 (1), 50–57. Luoma, K., Riihimaki, H., Luukkonen, R., Raininko, R., Viikari-Juntura, E., Lamminen, A., 2000. Low back pain in relation to lumbar disc degeneration. Spine (Phila Pa 1976) 25 (4), 487–492. Meisel, H.J., Ganey, T., Hutton, W.C., Libera, J., Minkus, Y., Alasevic, O., 2006. Clinical experience in cell-based therapeutics: intervention and outcome. Eur. Spine J. 15 (Suppl. 3), S397–S405. Mishima, Y., Lotz, M., 2008. Chemotaxis of human articular chondrocytes and mesenchymal stem cells. J. Orthop. Res. 26 (10), 1407–1412. Omlor, G.W., Nerlich, A.G., Lorenz, H., Bruckner, T., Richter, W., Pfeiffer, M., Guhring, T., 2012. Injection of a polymerized hyaluronic acid/collagen hydrogel matrix in an in vivo porcine disc degeneration model. Eur. Spine J. 21 (9), 1700–1708. Postacchini, F., Postacchini, R., 2011. Operative management of lumbar disc herniation: the evolution of knowledge and surgical techniques in the last century. Acta Neurochir. Suppl. 108, 17–21. Risbud, M.V., Schaer, T.P., Shapiro, I.M., 2010. Toward an understanding of the role of notochordal cells in the adult intervertebral disc: from discord to accord. Dev. Dyn. 239 (8), 2141–2148. Risbud, M.V., Shapiro, I.M., 2011. Notochordal cells in the adult intervertebral disc: new perspective on an old question. Crit. Rev. Eukaryot. Gene Expr. 21 (1), 29–41. Saraiya, M., Nasser, R., Zeng, Y., Addya, S., Ponnappan, R.K., Fortina, P., Anderson, D.G., Albert, T.J., Shapiro, I.M., Risbud, M.V., 2010. Reversine enhances generation of progenitor-like cells by dedifferentiation of annulus fibrosus cells. Tissue Eng. Part A 16 (4), 1443–1455. Sato, M., Asazuma, T., Ishihara, M., Kikuchi, T., Kikuchi, M., Fujikawa, K., 2003. An experimental study of the regeneration of the intervertebral disc with an allograft of cultured annulus fibrosus cells using a tissue-engineering method. Spine (Phila Pa 1976) 28 (6), 548–553. Scholz, B., Kinzelmann, C., Benz, K., Mollenhauer, J., Wurst, H., Schlosshauer, B., 2010. Suppression of adverse angiogenesis in an albumin-based hydrogel for articular cartilage and intervertebral disc regeneration. Eur. Cells Mater. 20, 24–36, discussion 27–36. Shim, C.S., Lee, S.H., Park, C.W., Choi, W.C., Choi, G., Choi, W.G., Lim, S.R., Lee, H.Y., 2003. Partial disc replacement with the PDN prosthetic disc nucleus device: early clinical results. J. Spinal Disord. Tech. 16 (4), 324–330. Sivan, S.S., Hayes, A.J., Wachtel, E., Caterson, B., Merkher, Y., Maroudas, A., Brown, S., Roberts, S., 2013. Biochemical composition and turnover of the extracellular matrix of the normal and degenerate intervertebral disc. Eur. Spine J. 23 (Suppl. 3), S344–S353. Thibault, M.M., Hoemann, C.D., Buschmann, M.D., 2007. Fibronectin, vitronectin, and collagen I induce chemotaxis and haptotaxis of human and rabbit mesenchymal stem cells in a standardized transmembrane assay. Stem Cells Dev. 16 (3), 489–502. Urban, J.P., Roberts, S., 2003. Degeneration of the intervertebral disc. Arthritis Res. Ther. 5 (3), 120–130. Vernengo, J., Fussell, G.W., Smith, N.G., Lowman, A.M., 2010. Synthesis and characterization of injectable bioadhesive hydrogels for nucleus pulposus replacement and repair of the damaged intervertebral disc. J. Biomed. Mater. Res. B: Appl. Biomater. 93 (2), 309–317. Woiciechowsky, C., Abbushi, A., Zenclussen, M.L., Casalis, P., Kruger, J.P., Freymann, U., Endres, M., Kaps, C., 2012. Regeneration of nucleus pulposus tissue in an ovine intervertebral disc degeneration model by cell-free resorbable polymer scaffolds. J. Tissue Eng. Regen. Med., http://dx.doi.org/10.1002/term.1582, Aug 3., [Epub ahead of print]. Zelle, S., Zantop, T., Schanz, S., Petersen, W., 2007. Arthroscopic techniques for the fixation of a three-dimensional scaffold for autologous chondrocyte transplantation: structural properties in an in vitro model. Arthroscopy 23 (10), 1073–1078.

Please cite this article in press as: Endres, M., et al., Augmentation and repair tissue formation of the nucleus pulposus after partial nucleotomy in a rabbit model. Tissue Cell (2014), http://dx.doi.org/10.1016/j.tice.2014.09.003