Stem Cell Regeneration of the Intervertebral Disk

The use of stem cell applications has been explored and aimed at regenerating the intervertebral disk. The microenvironment in which cells of the intervertebral disk reside is harsh; however, researchers have reported on many applications for stem cells, including research aimed at defining and stimulating endogenous stem cell populations, methods to induce stem cell differentiation toward intervertebral disk cell phenotype in vivo, and direct transplantation of stem cells into damaged intervertebral disk to promote transplanted site-dependant differentiation. Successful results have been reported, although limitations remain. This article reviews the current status of stem cell research as applied to the intervertebral disk.

The intervertebral disk (IVD) functions as an essential load absorber between all vertebrae by allowing bending, flexion, and torsion of the spine. IVD degeneration is a cell-mediated response to progressive structural failure and causes instability of the vertebral motion segments that are responsible for neural compressive manifestations and low back pain. Prolonged segmental instability eventually leads to deformity of the spine and many clinical problems. These manifestations have a high impact on society and the economy, including direct costs for medical treatment and insurance, lost productivity, and disability benefits. These direct and indirect costs are estimated at £12 billion per year in the United Kingdom and $50 billion in the United States. Therefore, prevention and treatment of IVD degeneration should have significant effects on society and the economy.

Cellular microenvironment of the intervertebral disk

The IVD comprises the central nucleus pulposus (NP), the surrounding annulus fibrosus (AF), and the vertebral end plate, which isolates the blood supply from penetrating the largest avascular organ in the human body. Developmentally, the NP originates from the notochord, although in human and other animal species, such as the rat, chondrodystrophoid breed canines, and cattle, there is a marked change in cell and tissue morphology during the early stage of life. In human adults, most NP cells resemble chondrocytes of other cartilaginous tissues. These cells are interspersed at low density (approximately 5000 cells/mm ³) and are sometimes arranged in clusters within the matrix. In the surrounding AF, the cells are more typically fibroblastic with a density of approximately 9000 cells/mm ³and display a fibrous matrix, comprising 15 to 25 lamellae of collagen fibers oriented alternately at approximately 60° to the vertical axis. In-between the lamellae are the elastin and proteoglycan matrix, which reinforces the viscoelastic structure. The vertebral end plate is characterized by a thin layer of chondrocytes and hyaline matrix, which resembles articular cartilage. A capillary network of blood vessels ending here, called the vascular buds, is found within the end plate and supplies approximately 80% of the nutrients needed to support the viability of IVD cells through diffusion. The microenvironment for these cells comprising the IVD is characterized by low oxygen tension and high lactic acid concentration and thus an acidic pH level, compared with the levels in the blood plasma.

Stem cell applications in intervertebral disk research

Decreased number and viability of the IVD cells, especially the NP cells, initiate disk degeneration, and maintaining the homeostasis and restoring the IVD tissue and function are important determinants of the cells’ condition. Basic in vitro studies have shown that IVD cells have a low proliferative ability and that most cells in the adult human IVD are in a senescent state. These facts have led researchers to focus on the idea of using stem cells to treat IVD degeneration.

Stem cells are characterized by the ability to self-renew and multipotent capabilities. The application of stem cells and stem cell research techniques in IVD research has been investigated from several directions. Use of new stem cell sources, such as induced pluripotent stem cells or embryonic stem cells, may provide new insight into the field of IVD research.

Stem cell applications in intervertebral disk research

Decreased number and viability of the IVD cells, especially the NP cells, initiate disk degeneration, and maintaining the homeostasis and restoring the IVD tissue and function are important determinants of the cells’ condition. Basic in vitro studies have shown that IVD cells have a low proliferative ability and that most cells in the adult human IVD are in a senescent state. These facts have led researchers to focus on the idea of using stem cells to treat IVD degeneration.

Stem cells are characterized by the ability to self-renew and multipotent capabilities. The application of stem cells and stem cell research techniques in IVD research has been investigated from several directions. Use of new stem cell sources, such as induced pluripotent stem cells or embryonic stem cells, may provide new insight into the field of IVD research.

Defining endogenous stem cell populations in the adult intervertebral disk

Recent stem cell research has reported the presence of a stem/progenitor cell system as the key to maintaining normal homeostasis and self-renewal in various organs. Decreased number and altered function of stem/progenitor cells cause dysfunction of the composing organ. Activation of the endogenous stem/progenitor cells is one approach for maintaining cellular homeostasis of the IVD. Risbud and colleagues reported that cells isolated from degenerate human tissues express CD105, CD166, CD63, CD49a, CD90, CD73, p75 low-affinity nerve growth factor receptor, and CD133/1, proteins that are characteristic of marrow mesenchymal stem cells (MSCs) and that represent the differentiation ability toward osteogenesis, adipogenesis, and chondrogenesis. A study by Blanco and colleagues compared the differentiation capabilities of MSCs induced from the bone marrow or the NP from the same 16 individuals and found that MSCs similar to bone marrow MSCs are present in the human NP, with the exception that NP MSCs show poor adipogenic differentiation. Feng and colleagues reported that AF cells express several of the cell surface antigens sometimes associated with MSCs, including CD29, CD49e, CD51, CD73, CD90, CD105, CD166, CD184, and Stro-1, and two neuronal stem cell markers, nestin and neuron-specific enolase. Varying the stimulants added to the induction media determined whether AF cells differentiated into adipocytes, osteoblasts, chondrocytes, neurons, or endothelial cells. These research data suggest that stimulation of endogenous stem cell populations may be effective for treating IVD degeneration or for providing cells for the allogeneic transplantation of somatic tissue-specific stem cells.

Induction of stem cells from other organs of the body

Another scenario involves promoting the mobilization of stem cell populations from the stem cell pool, such as the bone marrow. In cerebral and cardiac infarctions, stem cells are recruited from the stem cell pool and mobilized by agents, such as stem cell growth factor or granulocyte colony-stimulating factor, to restore cells in the injured lesion. This kind of system may not be applicable to IVD degeneration because there is no blood supply through which to mobilize the cells; however, there may be different pathways for stem cells to approach the IVD. Detailed research on this problem awaits investigation.

Using stem cells as feeder cells to intervertebral disk cells

Stem cells may serve as feeder cells to stimulate directly other cells in the environment by cell-to-cell contact or indirectly through the secretion of various factors. In a rabbit IVD cell culture, Yamamoto and colleagues showed that direct cell-to-cell contact between NP cells and MSCs occurs across a membrane with 0.45-μm pores, which allowed only the processes to adhere to each other without more extensive contact between the cultured cells. The extent of cell adhesion was assessed by scanning electron microscopy, cell proliferation was evaluated by the WST-8 assay, and the syntheses of DNA and proteoglycans was evaluated by the uptake of ³H and ³⁵S, respectively. The levels of various growth factors and the secretion of cytokines into the culture supernatant were measured using a cytokine protein array. The results were confirmed by electron microscopy and showed that MSCs and NP cells adhered to each other by extending processes across the membrane. The number of cells significantly increased in NP cells cocultured with MSCs and allowed cell-to-cell contact. In addition, the synthesis of DNA and proteoglycans increased significantly in the NP cells cocultured with MSCs when cell-to-cell contact was allowed. The analysis using the cytokine protein array revealed that the secretion of cytokines known to increase the activity of NP cells (transforming growth factor β1 [TGF-β1]), insulinlike growth factor 1, platelet-derived growth factor, and epidermal growth factor) was also significantly higher in the media collected from NP cells cocultured, allowing cell-to-cell contact. Compared with the conventional NP cell-activation method, the coculture system allowing intercellular adhesion with MSCs led to a marked increase in NP cell proliferation, DNA synthesis, and proteoglycan synthesis. A possible explanation is the increased secretion of various cytokines into the culture medium because of the direct contact with MSCs, which act as feeder cells.

In a preliminary study at the author’s laboratory, NP cells activated by coculture that allows intercellular contact ( Fig. 1 ) were implanted in an in vivo rabbit model of IVD degeneration. The severity of degeneration was determined over time according to Nishimura’s histologic classification. The severity of degeneration was compared between cells treated with the new and conventional methods of activation. The Nishimura grade 24 weeks after transplant was 0 in the normal control group without degeneration induction, 2.8 (the most severe degeneration) in the control group with no treatment, 2.2 in the group receiving NP cells activated by conventional coculture with AF cells, 1.8 in the group receiving NP cells activated by conventional coculture with MSCs, and 1.2 in the group receiving NP cells activated by coculture involving contact with MSCs, the smaller value reflected a significantly less degree of degeneration.

The positive results of this coculture system have been extended to preclinical studies using human cells. Watanabe and colleagues showed that human NP cells obtained from surgery and cocultured with MSCs of the same patient demonstrate up-regulated cellular proliferation and matrix synthesis, as described in animal models.

Strassburg and colleagues demonstrated in the same coculture system using degenerate and nondegenerate NP cells that cellular interactions between MSCs and degenerate NP cells may stimulate both MSC differentiation to an NP-like phenotype and the endogenous NP cell population to regain a nondegenerate phenotype, which consequently increases matrix synthesis for self-repair.

Inducing stem cells toward the intervertebral disk cell phenotype

Using the multipotent differentiation capacity of stem cells, the author attempted to induce MSC differentiation in a mixed coculture system with NP or AF cells in alginate beads ( Fig. 2 ). IVD tissue was retrieved during surgery for a burst fracture in a 19-year-old man. Under a microscope, the tissue was separated approximately into the NP and inner and outer AF. The separated tissue was digested with 0.02% pronase (Sigma) and 0.0125% collagenase P (Roche) for 8 hours to obtain cells for primary culture. The NP, inner AF, and outer AF cells were cultured and passaged twice and labeled with PKH26 red fluorescent dye (Sigma). Human MSCs were obtained commercially (Cambrex) and genetically labeled with green fluorescent protein (GFP) by infection with a retrovirus vector. The NP, inner AF, or outer AF cells were cocultured with MSCs in alginate beads in a 50:50 ratio at a density of 30,000 cells/bead. The cells were cocultured for 3 weeks in DMEM + 10% fetal bovine serum, and the cells were recovered. The recovered cells were analyzed, and GFP-positive MSCs were separated by flow cytometry (BD FACSVantage). Characterization of the recovered MSCs by flow cytometry showed that, in forward scatter analysis, the size of MSCs changed markedly after the coculture. MSCs cocultured with NP cells showed significantly greater average cell size, whereas cells cocultured with inner or outer AF cells had a smaller average cell size. The internal complexity analyzed by side scatter showed that MSCs cocultured with NP cells became more complex and that MSCs cocultured with inner or outer AF cells became less complex. These characteristics reflected the NP or inner or outer AF cell phenotype of the cocultured opponent. MSCs cocultured with NP cells expressed type II collagen and keratin sulfate, whereas the expression of type I collagen was more intense in cells cocultured with outer AF cells compared with the MSCs before coculture. Gene expression analysis by reverse transcription–polymerase chain reaction (RT-PCR) also confirmed that coculture with different IVD cells in the same 3-D environment led to differentiation of MSCs toward the direction of the cocultured opponent. These experiments showed that the mixed coculture system in alginate is an effective tool for inducing differentiation to MSCs.