Significant progress has been made in characterizing the intervertebral disk in both its healthy and diseased states. The scaffold of the nucleus pulposus is composed of type II collagen, while the outer annulus fibrosus is composed primarily of type I collagen. The nucleus contains a relatively low cellular density, and the extracellular matrix is composed largely of the proteoglycan aggrecan.

Starting in the second decade, the disk begins to undergo mild microscopic degenerative changes due to aging (decay of the nucleus pulposus cells, mild cleft formation, alterations in cell density, and matrix degeneration) coincident with regression of the blood vessels in the annulus, cartilage, and osseous end plates. As vascularity decreases, diffusion of oxygen, nutrients and waste must occur across wider expanses of avascular tissue (2). Small fragments generated from proteoglycan breakdown may further impair exchange of nutrients and waste at the capillary interfaces of the vascular end plates. As a result, the proteoglycan content of the nucleus declines with age, and the disk dehydrates. Degenerating intervertebral disks may demonstrate biochemical aberrations beyond normal aging. With degeneration, the normal type II collagen that forms the scaffold of the nucleus is gradually replaced with type I collagen, leading to fibrosis and loss of elasticity.

With degeneration, the extracellular matrix makes a failed attempt at remodeling. Molecules have been identified, which participate in this process. For example, transforming growth factor beta (TGF-β) is a family of cytokines that induces synthesis of matrix constituents. Fibronectin, a potent chemoattractor of fibroblasts, is known to play an important role in the reparative processes of wound healing and is a useful marker for active connective tissue repair. Fibronectin and TGF-β are found in both the nucleus and annulus of intervertebral disks undergoing degeneration. This provides evidence of an ongoing effort to remodel the extracellular matrix as the intervertebral disk degenerates (3).

In addition to the decreased anabolic response, a disruption in normal disk homeostasis leads to an increase in catabolic activity. Cells from degenerating disks spontaneously produce increased amounts of nitrous oxide, interleukin-6, prostaglandin E2, and matrix metalloproteinases, which leads to a net loss of proteoglycan in the nucleus pulposus. Nucleus pulposus cells increase production of these agents when provoked by an inflammatory stimulus such as interleukin-1(beta) (4). Matrix metalloproteinases cause proteoglycan breakdown, and nitrous oxide, interleukin-6, and prostaglandin E2 are believed to inhibit proteoglycan synthesis and thus impair regeneration. In the healthy intervertebral disk, these agents interact in a regulated manner to maintain homeostasis. However, these processes become unbalanced with degeneration, leading to net catabolism.

Matrix metalloproteinase 1, 2, 3, and 9 activity is more prevalent in herniated than in nonherniated disks, particularly early in the degenerative process. These catabolic enzymes can be produced by indigenous disk cells as well as invading blood vessels (5). In the extracellular matrix of the healthy disk, excessive breakdown is prevented by tissue inhibitors of metalloproteinase, but in diseased disks, the balance between enzyme and inhibitor is again tipped toward degeneration. Thus, even when the concentrations of matrix metalloproteinases are not increased in the degenerating disk, their activity is elevated due to decreased inhibition. This imbalance may occur early in the disease process (6).

Investigators utilize a wide variety of tools to study the molecular biology of disk degeneration. The advantages and disadvantages of each are briefly reviewed.

Cell culture techniques can be employed in disk research, in which eukaryotic cells are grown in vitro under controlled conditions separate from their original tissue source. Cells are isolated from a tissue sample by enzymatic digestion, plated in a two-dimensional surface or suspended in a three-dimensional (3-D) matrix, incubated at simulated physiologic conditions, and nurtured with nutrient-rich media. This allows investigators to assay for specific biochemical markers to better understand the degeneration cascade and to test new treatment modalities under wellcontrolled conditions. However, it does not account for the complex interactions that occur in vivo and lacks the stress of physiologic loading.

Although 3-D cell culture seeks to mimic some of the complex conditions to which molecules and cells are exposed, no in vitro system can truly anticipate and replicate the in vivo condition. Good animal models are reliable, reproducible, sensitive to effective interventions, and comparable to the human condition in health and disease.

A model for intervertebral disk degeneration has been developed for use in spine research (Fig. 69.1) (7). Annulotomy of a rabbit’s intervertebral disk induces a slow, reliable, reproducible biochemical and biomechanical cascade of degeneration, demonstrable by serial T2-weighted MRIs (8). Chemical nucleotomy with chondroitinase has also been used. Various investigational treatments can subsequently be examined in these models in an effort to slow or halt the disk from degenerating.

The true efficacy of any treatment requires careful trials in humans afflicted with the disease. The treatment of a nonlethal disease such as disk degeneration increases the necessity of demonstrating unequivocal safety and may require novel control mechanisms as discussed below. In addition, there have been significant ethical concerns raised surrounding stem cell and gene therapy. The unfortunate death of Jesse Gelsinger, an 18-year-old afflicted with a rare metabolic condition who died as a result of an overwhelming immune response to a gene therapy virus vector, is a stark reminder of the serious safety concerns inherent in clinical trials of gene therapy (9).

Growth factors are proteins that are capable of stimulating cell growth, proliferation, and differentiation. There are numerous families of growth factors, including bone morphogenetic proteins (BMPs), platelet-derived growth factors, transforming growth factors, vascular endothelial growth factors, and colony-stimulating factors. A few growth factors are used commonly by orthopaedic surgeons. In 2001, the U.S. Food and Drug Administration (FDA) approved BMP-7 (marketed as Osteogenic Protein-1 by Stryker Biotech) under the Humanitarian Device Exemption for use in long bone nonunions. In 2002, the FDA approved BMP-2 (marketed as Infuse by Medtronic Sofamor Danek) with the use of a particular type of metallic tapered spinal interbody cage for lumbar fusion. In 2004, BMP-2 was approved for the treatment of open tibial shaft fractures, and it is also sometimes used off-label for long bone nonunions. BMP-7 at this time has a Humanitarian Device Exemption status for use in tibial nonunions and posterior revision lumbar spine fusions. All other growth factors used in orthopaedic surgery are currently either off-label or experimental. See Table 69.1 for a list of important growth factors and their proposed actions.

Recent research with BMPs has been directed toward intervertebral disk degeneration due to their ability to induce formation of cartilage in addition to bone. They
play an important role in skeletal development during embryogenesis and in bone and cartilage repair. In cell culture, BMP-2 stimulates chondrocytes to produce proteoglycan (10). It has been found to directly increase collagen synthesis and also to upregulate the expression of other BMPs (such as BMP-7), which in turn further increase proteoglycan and collagen synthesis (11).

Similarly, TGF-β and epidermal growth factor (EGF) significantly increase proteoglycan synthesis in the nucleus pulposus in vitro. Stimulation with insulin-like growth factor-1 (IGF-1) and fibroblast growth factor (FGF) cause more modest proteoglycan increases (12). This early growth factor research was performed in cultured canine intervertebral disk cells. Increased proteoglycan synthesis was demonstrated by an increased rate of incorporation of³⁵S-sulfate and³H-thymidine (isotopes that are taken up during synthesis and can be tracked) after growth factor stimulation.