Human Intervertebral Disc Cell Culture for Disc Disorders

Abstract

Repair of degenerated intervertebral discs by engineered tissue is a clinical challenge in spinal surgery. Prerequisites are cultivation of intervertebral disc cells and determination of their biologic properties. The influence of disc damage in different spinal disorders on the outcome of disc cell cultures has not been discussed previously. This study showed the feasibility of cultivation of cells from damaged human intervertebral discs and the dependence of cellular culture properties on the underlying disc disorder. Human intervertebral disc cells were isolated from disc tissue obtained during surgical procedures for scoliosis, osteochondrosis, and disc herniation. After proliferation in monolayer culture, cells were embedded in a mixed matrix composed of fibrin and hyaluronic acid. Deoxyribonucleic acid content, hydroxyproline content, and proteoglycan synthesis were determined on Days 7, 14, and 21. In a three-dimensional environment only cells obtained from scoliotic and osteochondrotic discs showed significant deoxyribonucleic acid and proteoglycan synthesis. However, hydroxyproline content increased only in cells from scoliotic discs. The results of this study show that the formation of extracellular matrix components under three-dimensional culture conditions is dependent on the nature of intervertebral disc damage of the tissue processed.

INTRODUCTION

Structural changes in the extracellular substance of the nucleus pulposus have been shown to be a cause of progressive degeneration in the motion segments of the spine in different studies.^3,6,15 The relationship between a cause of disc degeneration and the ability to use an autologous biologic repair technique has not been determined.

For tissue engineering, which offers strategies for artificial tissue replacement, two prerequisites have to be achieved for implantation of a cell-matrix construct: the development of a carrier substance that allows synthesis of macromolecular components of the extracellular matrix by cultivated cells, and isolation of suitable cells that expand under in vitro culture conditions and form macromolecules of the extracellular substance while their differentiation is maintained.

The cultivation of cells from intervertebral discs of human and animal origin in alginate has been described.^9,12,19 As a result of the lack of biocompatibility and biodegradability of alginate, the current research group developed a matrix consisting of fibrin and hyaluronic acid in which an adequate synthesis of proteoglycan and Type II collagen could be shown.²⁴

The success of possible clinical use will be measured on the basis of an effective synthesis of extracellular matrix components. The limited biosynthetic activity in degenerated discs processed for cell culture might significantly reduce the output of the cultivated cells. However, Gruber et al¹² showed that human intervertebral disc cells taken from surgical specimens expand and form proteoglycan in a three-dimensional environment. However, the measured results were not related to the disc alteration of the tissue processed. Therefore, it is unclear to what extent the specific pathologic features of resected discs influence the output of cell cultures. In this study three groups were formed on the basis of different spinal diseases: idiopathic scoliosis, degenerative disc disease (osteochondrosis), and disc herniation. These disorders are associated with different degenerative alterations of the discs. By measuring the proteoglycan synthesis and the hydroxyproline content as a marker for collagen content, the influence of the nature of the underlying spinal disorder and the degenerative alteration of disc tissue on the output of cell cultures were determined.

MATERIALS AND METHODS

This experimental study of human disc specimens was approved by our institutional research review committee.

Cell Isolation and Cell Culture

Disc tissue from damaged lumbar segments was removed by either discectomy or nucleotomy in patients who had a discectomy in the course of spondylodesis because of idiopathic scoliosis (n = 6 patients; low-grade degenerative alteration in the entire disc) (Fig. 1), discectomy in the course of spondylodesis because of osteochondrosis (n = 6 patients; high-grade degenerative alteration in the entire disc) (Fig. 2), and nucleotomy because of disc herniation (n = 6 patients; isolated prolapse of the nucleus pulposus) (Fig. 3). Only disc tissue from lumbar segments was used. In the case of discectomy, the whole intervertebral disc was harvested. In disc herniation, only the herniated fragment was removed. After surgical removal of the disc material, the nucleus pulposus was separated from the annular tissue. The acquired nuclear tissue was pooled if several discs had been resected in one patient.

FIGURE 1.: (A) A frontal radiograph of the thoracolumbar spine shows no obvious degenerative alterations in a patient with idiopathic thoracolumbar scoliosis. (B) The corresponding frontal T2-weighted MRI scan shows normal signal intensity of the discs. (C) The corresponding sagittal T2-weighted MRI scan also shows normal signal intensity of the discs.

FIGURE 2.: (A) A lateral radiograph of the lumbosacral spine shows radiologic signs of progressive degenerative disc desease of the L5-S1 discs attributable to lytic spondylolisthesis. (B) The corresponding sagittal T2-weighted MRI scan indicates degeneration of the L5-S1 discs by loss of signal intensity at the disc level.

FIGURE 3.: (A) A lateral radiograph of the lumbosacral spine shows no radiologic signs of degenerative alterations. (B) The corresponding sagittal T2-weighted MRI scan shows a disc hernia at L4-L5. The disc shows normal signal intensity.

The specimens were transferred to the laboratory in sterile Ringer’s solution within 1 hour. The cells were isolated from the tissue formation by enzymatic digestion. The enzyme solutions were based on Dulbecco’s modified Eagle medium (DMEM) and Ham’s F-12 medium (Biochrom, Berlin, Germany) supplemented with 5% fetal calf serum (Biochrom). The tissue first was incubated with 0.4% pronase E (Serva, Heidelberg, Germany) for 1 hour, followed by overnight digestion with 0.02% collagenase (Serva) at 37°C with stirring. After removal of tissue debris by filtering through a nylon mesh (100 μm), the isolated cells were rinsed three times in phosphate buffered saline.

After determination of viability by trypan blue exclusion test and cell counting using a Neubauer hemocytometer, the cells were seeded into culture flasks (50 mL; Becton Dickinson, Heidelberg, Germany). The isolated nucleus pulposus cells were seeded in monolayer culture at a density between 1 and 3 × 10⁴ cells/cm². Cells were fed every 2 days with DMEM and Ham’s F-12 supplemented with 10% fetal calf serum, streptomycin (100 μg/mL), penicillin G (100 IU/mL), and amphotericin (1 μg/mL). The cultures were maintained at 37°C in a humidified atmosphere of 95% air and 5% CO₂.

After one passage in monolayer culture the cells were embedded in the three-dimensional environment of microbeads. Therefore, the cell pellet was suspended in sterile saline solution containing fibrinogen (Beriplast^\R, Centeon, Marburg, Germany) mixed with hyaluronic acid (Hy-GAG^\R, Chemedica, München, Germany) and alginate (Sigma, St Louis, MO) to reach a final concentration of fibrinogen by 5%, hyaluronic acid by 0.25%, and alginate by 0.6% (cell density 1 × 10⁶ cells/mL). The total volume was 2000 μL. The cell fibrinogen-hyaluronan-alginate mixture was dropped into a 102 mmol/L CaCl solution containing 46 IU/mL thrombin (Beriplast^\R, Centeon) and was allowed to polymerize for 20 minutes at room temperature. The beads were washed sequentially twice in 0.15% NaCl solution. Finally, the alginate was extracted from the mixed beads by a chelating agent consisting of a solution of 0.055 mol/L sodium citrate, 0.15 mol/L NaCl, and 0.03 mol/L ethylenediaminetetraacetate (pH 6.8). The beads were put in rotating bottles (Reichelt, Heidelberg, Germany) and incubated at 37°C in a humidified atmosphere of 95% air and 5% CO₂ for 21 days. The medium consisting of DMEM and Ham’s F-12 supplemented with 10% fetal calf serum, streptomycin (100 μg/mL), penicillin G (100 IU/mL), amphotericin (1 μg/mL), and aprotinin (Trasylol^\R, Bayer, Leverkusen, Germany; 250 KIU/mL) was changed every 2 days.

Deoxyribonucleic Acid Content

The determination of DNA content was done fluorometrically as described by Labarca and Paigen.¹⁷ At Days 7, 14, and 21 the beads were lyophilized and homogenized as mentioned above. After centrifugation the supernatant was mixed with the fluorochrome Hoechst 33258 (Serva). The measurements of fluorescence were done in a spectrofluorometric detector (821-FP; Jasco, Hachioji City, Japan).

Proteoglycan Synthesis

At Days 7, 14, and 21 the cultures were incubated with [³⁵S] sulfate (Amersham Buchler, Braunschweig, Germany) labeled medium for 24 hours (20 μCi/mL [0.74 MBq/mL]). After lyophilization gels were homogenized in 2 mol/L NaCl. Aliquots were taken from each homogenate for simultaneous determination of the DNA content and hydroxyproline content. For analysis of total proteoglycan synthesis the homogenates were extracted overnight with 4 mol/L guanidine hydrochloride buffered with 0.05 mol/L sodium acetate (pH 5.8) and containing the following proteinase inhibitors: 0.01 mol/L ethylenediaminetetraacetate, 0.1 mol/L ε-aminohexanoic acid, 5 mmol/L benzamidine HCl, 0.01 mol/L N-ethylmaleimide, and 0.5 mmol/L phenylmethyl-sulfonyl fluoride. The extracts were eluted through a Sephadex G 25 (Pharmacia, Uppsala, Sweden) column (30 × 73 mm) with guanidinium hydrochloride. Radioactivity of the macromolecular fraction (first peak) was measured as disintegrations per minute (dpm) with liquid scintillation counting (LS 6000SC; Beckman, Fullerton, CA).

Hydroxyproline Content

Above-mentioned aliquots were used partially for the determination of hydroxyproline content as described by Stegemann and Stalder.²³ Briefly, the homogenate was hydrolyzed by 6 mmol/L HCl at 100° C for 20 hours. The hydrolyzate was dried over KOH and CaCl. After the addition of chloramine T and p-dimethylaminobenzaldehyde in a buffered solution, the extinction was measured at 550 nm. The hydroxyproline content corresponds to the collagen content of the beads.

Statistical Analysis

Statistical analysis of the results for determination of DNA content, proteoglycan synthesis, and hydroxyproline content was done using the Mann-Whitney U test for unrelated samples to determine differences between the different groups. Statistical significance was determined based on values of 0.05 or less.

RESULTS

Cell Number and Viability

After enzymatic tissue digestion, a vitality rate greater than 80% was established for the isolated cells.

Depending on the underlying disc disorder, there were differences in cell phenotype in monolayer culture. In scoliosis, cells predominantly were polygonal. Cells obtained from herniated discs had a slightly different pattern whereas the number of spindle-shaped cells had increased. In osteochondrosis, there were spindle-shaped and polygonal cells in approximately equal parts.

The total cell number of confluent cultures varied considerably. The variability was attributed mainly to the number of resected intervertebral discs in each patient as all the nucleus pulposus material of one patient was pooled in one culture. Different from degenerative disc disease and disc herniation, in scoliosis more intervertebral discs were resected by surgery. Furthermore, with nucleotomy in disc herniation, only the herniated disc fragment could be harvested.

The period until cells reached confluence in monolayer culture and the vitality of cells in these confluent cultures differed depending on the nature of disc damage. Only cells isolated from scoliotic discs showed a vitality rate greater than 80%. In this group the period until cells became confluent after one-time passaging was similar to the primary culture, but was extended to more than 2 weeks in discs with osteochondrosis and in herniated discs (Table 1). The difference in vitality and cell growing does not relate to the amount of disc material because each cell culture was adjusted to equal cell density in the beginning of monolayer cultivation.

Table 1: Comparison of Period Until Cell Confluence and Vitality of Passaged Cells in Monolayer Culture

Overall, a sufficient number of cells for cultivation in a three-dimensional matrix was obtained after one passage. Therefore, a cell density of 1 × 10⁶ cells/mL could be adjusted for the microbead culture in each specimen.

Deoxyribonucleic Acid Content

Deoxyribonucleic acid content, which characterizes the proliferation capacity in the three-dimensional matrix, increased in cell-matrix-constructs from mildly damaged discs (scoliosis) and severely degenerated discs (osteochondrosis) during the entire course. Despite a low gradient in the third cultivation week a steady state was not reached until the twenty-first day. Both constructs showed a similar rise of DNA content in the second week. Contrary to that there was no increase of DNA content in microbeads with cells from herniated disc tissue. The difference between specimens obtained from disc herniation and specimens taken from scoliosis and osteochondrosis was significant after 21 days (P < 0.05). Furthermore, more than two times the amount of DNA was measured in scoliosis specimens in comparison with osteochondrosis-associated specimens (Fig. 4). At Day 21 this difference also was significant (P < 0.05).

FIGURE 4.: The mean DNA content of microbeads at 7, 14, and 21 days is shown. ^#There was a significant difference (P \H 0.05) between scoliosis and herniation and also *between scoliosis and osteochondrosis or herniation.

Proteoglycan Synthesis

An increase in proteoglycan synthesis was found in microbeads with cells obtained from mildly altered discs (scoliosis) and severely damaged discs (osteochondrosis). The cell-matrix constructs with cells taken from herniated disc tissue showed merely a baseline synthesis. Similar to the course of DNA content in scoliosis and osteochondrosis specimens, there was a steep gradient in the second cultivation week followed by an approximately steady state of proteoglycan synthesis in the third week. After 21 days, the amount of newly synthesized proteoglycan was more than three times higher in these samples than in cells isolated from disc herniation (Fig. 5).

FIGURE 5.: The mean proteoglycan synthesis of cell culture at 7, 14, and 21 days is shown. ^#There was a significant difference (P \H 0.05) between scoliosis and herniation. dpm = disintegrations per minute.

Hydroxyproline Content

Unlike the DNA content and the proteoglycan synthesis an increase in hydroxyproline content could not be measured as a marker of collagen synthesis in the first cultivation period. However, at the end of the cultivation period there was a steep gradient in specimens taken from scoliotic discs. This increase corresponds to the limited proliferation and proteoglycan synthesis in the final cultivation week (Fig. 6).

FIGURE 6.: The mean hydroxyproline content of microbeads at 7, 14, and 21 days is shown. *There was a significant difference between scoliosis and osteochondrosis or herniation.

DISCUSSION

The results of the current study show that the underlying disc disorder influences the cellular growth and the synthesis of macromolecular matrix components in cell culture. Previous studies did not differentiate between the nature of the disc alterations. Therefore, they failed to take into account the extent of the morphologic and ultrastructural damage^12,13 that is dependent on the pathologic features of the disc disorder.

In the current study, the DNA content increased in cultures made up of cells from scoliotic discs. With severely degenerated discs (osteochondrosis) the DNA content also increased even if to a lower extent. An increase in proteoglycan synthesis also was observed in both groups. Therefore, the comparison of unequal damaged discs showed a significant difference in DNA content and proteoglycan synthesis whereas both parameters are reduced in the case of severely degenerated discs.

Apart from changes on the cellular and extracellular levels occurring as a result of the progressive degenerative damage in osteochondrosis, the higher age of patients with osteochondrosis must be discussed. To minimize the age-dependent bias, scoliotic disc specimens of patients older than 2 decades were used. The physiologic changes in the aging disc tissue, especially in the nucleus pulposus, have an important influence on the composition of extracellular macromolecules. A heterogeneous cellular pattern consisting of chondrocytelike cells, fibroblasts, and notochordal cells forms the cellular framework of the juvenile nucleus pulposus.^8,14,22,25 With advancing age, the number of notochordal cells declines beyond the second decade of life.^21,26,27 Via a chain of signals, these cells induce the synthesis of macromolecular components of the extracellular substance.¹ This additionally restricts the potential for regeneration of tissue defects of the matured disc in osteochondrosis. Besides the change of morphologic pattern the influence of decreased viability of cells in more degenerated discs on the cellular properties in culture has to be considered.

The hydroxyproline content, as a marker of collagen synthesis, only increased in cultures made up of mildly degenerated discs. The increase in hydroxyproline content in the third week of three-dimensional cultivation was associated with a lower proliferation rate. On the basis of this chronologic correlation, the increasing immobilization of cells from mildly degenerated discs led to an increase in collagen synthesis in their matrix. The immobilization associated with limited cell proliferation is a requirement for maintenance of the phenotype of the cultivated cells. This relationship also was shown in articular chondrocytes which synthesize cartilage-specific collagen and proteoglycan while maintaining their differentiation in three-dimensional culture systems.^4,5,7,10

The biologic behavior of disc cells in culture is not only dependent on the extent of disc degeneration, as the results in cells in herniated disc specimens indicate. Although, the herniated discs show no obvious lack of water content in the T2-weighted MRI scan, the DNA content in cell cultures of herniated disc fragments did not increase, and there was no significant increase in collagen or proteoglycan synthesis measurable with time. The cytotoxicity of the mediators that are released in a large quantity in the case of a herniated disc as a result of the accompanying inflammatory reaction^20,28 probably led to irreversible damage to the cells in vivo. Although bioactive mediators are present in several disc diseases, even in nondegenerated discs,¹⁶ elevated concentration of IL-1 can be measured in prolapsed tissue. As shown by Maeda and Kokubun,¹⁸ proteoglycan synthesis was reduced with the administration of IL-1. Interleukin-1 is produced by macrophages, which are the predominant inflammatory cells in disc herniation.¹¹ This inflammatory response may cause the increased stimulation of cytokines and chemokines in prolapsed tissue compared with degenerated discs. In this context, Kang et al¹⁶ reported on stimulation of disc cells taken from herniated discs. These cells, which spontaneously produce increased levels of biochemical agents, additionally increased production in response to IL-1. Despite a lack of inflammatory cytokines in cell cultures, the cytotoxic effect seems to influence the in vitro activity of herniated disc cells and, to a certain extent, degenerated disc cells. Other than smaller vitality of removed cells, intracellular mechanisms triggered by a chain of signals before removal could be responsible for this ongoing effect.

The choice of disc tissue where cells for potential autologous transplantation can be removed has been restricted according to the current results. This study suggests that only cells from mildly degenerated discs are suitable for transplantation. However, they are available only in limited numbers in the autologous field. The optimization of culture conditions by adding growth factors or antagonists of the cytotoxically acting cell mediators, such as IL-1 receptor antagonists and metalloproteinase inhibitors, could increase the synthesis of a proteoglycan- and collagen-containing matrix. The change of culture conditions by adding the aforementioned bioactive mediators also could improve the outcome in cell cultures from osteochondrotic and herniated discs. Therefore, additional investigation is necessary to study the influence of such bioactive agents on these specific cell cultures. For biologic repair in degenerative disc disorders, another option could be the use of allogeneic cells. However, the immunogenic potential of allogenic cells in the disc tissue has not been described and remains a topic that has to be studied. Not until data of these studies are available and the results have been proven in animal models can the clinical application of transplantation of intervertebral disc cells be arranged. Transplantation techniques should be considered in low-grade degenerative disc disease, in the symptomatic bulgy disc phenomenon and protection of the adjacent segments in intervertebral fusion.

The results of the current study stress the dependency of cellular culture characteristics on the underlying spinal disorder. The assessment of disc degeneration by MRI is not a reliable predictor for the outcome of cell cultures. Although, there is a correlation between the loss of water and proteoglycan content in MRI and degeneration in the nucleus pulposus,² the cellular behavior of herniated disc specimens did not correspond to the mild degeneration of the disc. The poor outcome might be because cells removed from herniated fragments are not representative of the entire disc. Nevertheless, the extent of degeneration of disc tissue is an important factor that influences the outcome of disc cell cultures as shown in osteochondrosis and scoliosis. Another influential factor is the nature of the disc specimen processed, as shown in disc herniation. Therefore, in the choice of donor sites for possible autologous transplantation techniques, one must consider the nature of disc damage and the extent of disc degeneration.