For evaluation of the efficacy of various biologic treatments, a number of in vitro and in vivo model systems are routinely used. Isolated nucleus pulposus and annulus fibrosus cells from intervertebral discs are seeded in monolayer and treated with specific cytokines (e.g., IL-1, TNF-α) to model degenerative conditions. Cells from discs with varying degrees of degeneration are also used. In these systems, the efficacy of therapeutic agents can be determined at several levels. In cells, a decrease in the gene expression of catabolic enzymes, such as aggrecanases and proteinases, as well as vascular factors and pain-related factors may be assessed using PCR techniques. Anabolic genes of interest may include those molecules comprising the extracellular matrix, such as collagen and aggrecan, and anabolic regulators, such as TGF-β, IGF-1, and growth and differentiation factor-5 (GDF-5). At the protein level, the synthesis of collagen and proteoglycans can be assessed, using radiolabeled ³H and ³⁵S, respectively, by determining their incorporation from the media into the cells. Other factors can be assessed by enzyme-linked immunosorbent assay (ELISA) or multiplex cytokine assay.

Pellets of cells surrounded by three-dimensional extracellular matrices, recovered after alginate culture (Masuda et al. 2000), for example, have also been used. The presence of the extracellular matrix better mimics the microenvironment of disc cells. In addition to treatments using a monolayer ­system, pellet cultures may be treated with agents to alter or deplete specific matrix components (e.g., using chondroitinase ABC to deplete proteoglycans); this is not feasible using monolayer culture. For outcome measurements, the biochemical content of the matrix can be determined. For example, the dimethylmethylene blue (DMMB) assay can be used to assess the proteoglycan content, while western blot and PCR can be used to assess protein and gene expression levels.

A variety of anabolic growth factors and cytokines alter intervertebral disc homeostasis and stimulate extracellular matrix synthesis (Masuda et al. 2004). OP-1 (Masuda et al. 2003), a member of the BMP family and TGF-β superfamily (Fig. 25.1), upregulates proteoglycan metabolism in ­intervertebral disc cells. OP-1 strongly stimulated the production and formation of extracellular matrix by rabbit disc cells (Masuda et al. 2003); similar effects have been noted using human intervertebral disc cells (Imai et al. 2007a). OP-1 also replenished proteoglycans and collagens after depletion of the matrix following exposure of intervertebral disc pellets to IL-1 (Takegami et al. 2002) or chondroitinase ABC (C-ABC) (Takegami et al. 2005). The efficacy of OP-1 injection has also been evaluated in a number of in vivo animal models. In adolescent rabbits, an injection of recombinant human OP-1 (rhOP-1), but not the lactose vehicle, reversed the reduction in disc height and improved the ­magnetic resonance imaging (MRI) grade caused by a needle puncture of the annulus fibrosus (Masuda et al. 2006). In another rabbit study (Miyamoto et al. 2006b), OP-1 restored dynamic viscoelastic biomechanical properties, in ­needle-punctured intervertebral discs (Miyamoto et al. 2006b). It was also effective in restoring discs that have been chemically degraded with C-ABC, which has been considered as an alternative to chymopapain for chemonucleolysis (Eurell et al. 1990; Fry et al. 1991; Henderson et al. 1991; Kato et al. 1992; Ando et al. 1995; Sugimura et al. 1996; Takahashi et al. 1996b; Yamada et al. 2001), in animal models of disc degeneration such as the rat tail (Norcross et al. 2003; Hoogendoorn et al. 2007, 2008; Boxberger et al. 2008) and goat (Hoogendoorn et al. 2007, 2008). When OP-1 or vehicle was injected into rabbit discs degraded with C-ABC for 4 weeks, the disc height was initially decreased (∼34 %), then recovered, and gradually approached the level of the control (Imai et al. 2007b).

A number of autologous agents have been shown to be clinically useful. For example, platelet-rich plasma (PRP) contains high levels of multiple growth factors and as such it has been used in disc repair in animal studies and in a clinical study ongoing in Japan. Moreover, PRP can be easily generated in the operating room by centrifugal separation of autologous blood using a point-of-care device. In vitro, PRP-stimulated porcine disc cell proliferation and matrix synthesis (Akeda et al. 2006) and using isolated human disc cells provoked the formation of a nucleus pulposus-like tissue (Chen et al. 2006). The efficacy of injecting allograft PRP with or without a gelatin hydrogel (which provides slow release and mechanical support) was explored using a rabbit model of nucleotomy (Nagae et al. 2007). The PRP hydrogel was found to markedly suppress further degeneration, compared to PRP alone and a saline control group. A follow-up study suggested that the hydrogel microspheres without PRP did not have therapeutic value. In contrast, animals treated with PRP with microspheres benefited from increased disc height, elevated water content, increased expression of proteoglycan core protein and collagen II, and fewer apoptotic cells in the nucleus pulposus (Sawamura et al. 2009). In less severe models without nucleotomy, PRP (after activation with autologous serum and calcium) injection alone has been effective for disc repair (Obata et al. 2012).

In addition to PRP and OP-1, many other anabolic growth factors and cytokines have been investigated. Early studies indicated that TGF-β promoted disc cell proliferation (Gruber et al. 1997) and stimulated proteoglycan synthesis (Thompson et al. 1991; Gruber et al. 1997). IGF-1 (Osada et al. 1996; Pratsinis and Kletsas 2007) and platelet-derived growth factor (PDGF) (Pratsinis and Kletsas 2007) also showed a similar cell-proliferative effect. In addition, PDGF exerted a protective, antiapoptotic effect on annulus fibrosus cells induced by serum depletion (Gruber et al. 2000).

Agents other than anabolic growth factors are being ­considered for repair of the degenerate intervertebral disc. In a recent study, ADAMTS-5 expression was silenced using siRNA (Seki et al. 2009). In vitro studies, using rabbit nucleus pulposus cells, were performed to determine the extent of reduction of ADAMTS-5 expression. Then, using an annular needle puncture rabbit model, MRI and histological analysis were used to assess the efficacy of ADAMTS-5 siRNA to prevent tissue degradation. After 8 weeks, the ­control group exhibited a complete loss of nucleus pulposus tissue, while the ADAMTS-silenced animals maintained disc structure (Seki et al. 2009).

IL-1 receptor antagonist (IL-1ra) is another inhibitory agent that has been studied. When applied in vitro to ­degenerated (Le Maitre et al. 2007b) and herniated (Genevay et al. 2009) human disc tissues, IL-1ra reduced the expression of MMP-3 (Le Maitre et al. 2007b; Genevay et al. 2009). In addition, following pretreatment of degenerated human nucleus pulposus cells with IL-1ra and subsequent treatment with IL-1, there was reduced expression of ADAMTS-4 and MMP-3 (Shamji et al. 2007).

Yet another target for inhibition is TNF-α. In human disc cells, the use of soluble TNF receptors along with IL-1ra significantly upregulated proteoglycan synthesis (Kakutani et al. 2008), while co-treatment with TNF-α suppressed nitric oxide and IL-6 production (Sinclair et al. 2011). Although dominant-negative TNF (DN-TNF) does not activate TNF receptors, it effectively competes with soluble TNF. When applied to human intervertebral disc cells challenged with IL-1β (Pichika et al. 2011), DN-TNF effectively reduced the concentration of TNF-α (Fig. 25.2a, b), levels of MMP-3 in the media (Fig. 25.2c, d), the expression of intracellular caspase 3 (Fig. 25.2e, f), and the production of PGE2. The use of a monoclonal antibody against TNF-α suppressed the expression and concentration of MMP-3 in explants of herniated discs (Genevay et al. 2009). Other TNF-inhibiting agents are beginning to be used clinically for analgesic purposes in patients with sciatica (Karppinen et al. 2003; Genevay et al. 2004; Okoro et al. 2010) and discogenic pain (Tobinick and Britschgi-Davoodifar 2003). Anti-cytokine therapeutics include the p38 mitogen-activated protein kinase (MAPK) inhibitor, which suppressed MMP-3 and IL-1 expression (Studer et al. 2008) as well as a NF-κB decoy, which reduced pain in a rat lumbar disc herniation model (Suzuki et al. 2009).

Although the exact mechanism is unclear, the use of gelatinous biomaterials, alone or in combination with growth factors, has shown some efficacy. These materials may work by providing immediate load support (Joshi et al. 2005) as well as a hydrated environment in which host or embedded cells can proliferate (Collin et al. 2011). Fibrin glue has been used to repair intervertebral discs in a porcine nucleotomy model: it suppressed IL-6 and TNF expression and restored mechanical properties and the glycosaminoglycan (GAG) content (Buser et al. 2009). Hyaluronic acid-cross-linked hydrogels have been used in rabbits with less severe annular needle punctures and were found to improve disc height (Fig. 25.3a) and T2 MR properties (Bae et al. 2011) (Fig. 25.3a–d), as well as safranin O staining (Nakashima et al. 2009). Mixing the hydrogel with TGF-β prior to injection further increased disc height (Fig. 25.3a) (Bae et al. 2011). While little is known about degradation of these ­biomaterials in vivo and their long-term biologic effects, the use of a suitable carrier material may synergize growth factor activity.

While of limited use for development of injectable therapeutic agents, a mechanism for stimulating disc cells anabolism would be of great value for countering the in vivo degradation of intervertebral disc tissues. One factor that would be expected to impact anabolic events is local oxygen tension; in the disc this is low, between 2 and 5 % in vivo (Urban 2002; Bartels et al. 1998). Low oxygen tension would be expected to decrease mitochondrial function and oxidative activity (Bibby et al. 2005). It has been observed that it enhanced the anabolic effects of OP-1 (Miyamoto et al. 2006a), BMP-7 (Tonomura et al. 2007), and TGF-β3 (Abe et al. 2008) by promoting extracellular matrix synthesis by bovine nucleus pulposus cells. In addition, in MSCs (Stoyanov et al. 2011) and notochordal cells (Erwin et al. 2009), low oxygen tension has been shown to facilitate differentiation and extracellular matrix production. This topic is discussed further in Chap.​ 6.

Low oxygen tension results in lactate accumulation and a concomitant decrease in media pH (Pichika et al. 2012). Dynamic mechanical stimulation in a physiologic range can also stimulate anabolic activities. Thus, applying cycling tensile strain to discs resulted in F-actin reorganization and an increase in collagen I expression in outer annulus fibrosus cells and an increase of collagen II expression in the nucleus pulposus (Li et al. 2011). Rat caudal discs compressed in vivo at 1.0 MPa stress and 0.01 Hz frequency exhibited increased expression of anabolic genes (Maclean et al. 2004). For ­tissue-engineering applications, these techniques may be combined with growth factors and scaffold or gel biomaterials to help optimize the function of disc tissues (this topic is discussed in detail in Chap.​ 26).