Notochordal cells protect nucleus pulposus cells from degradation and apoptosis: implications for the mechanisms of intervertebral disc degeneration
© Erwin et al.; licensee BioMed Central Ltd. 2011
Received: 30 August 2010
Accepted: 29 December 2011
Published: 29 December 2011
The relative resistance of non-chondrodystrophic (NCD) canines to degenerative disc disease (DDD) may be due to a combination of anabolic and anti-catabolic factors secreted by notochordal cells within the intervertebral disc (IVD) nucleus pulposus (NP). Factors known to induce DDD include interleukin-1 beta (IL-1ß) and/or Fas-Ligand (Fas-L). Therefore we evaluated the ability of notochordal cell conditioned medium (NCCM) to protect NP cells from IL-1ß and IL-1ß +FasL-mediated cell death and degeneration.
We cultured bovine NP cells with IL-1ß or IL-1ß+FasL under hypoxic serum-free conditions (3.5% O2) and treated the cells with either serum-free NCCM or basal medium (Advanced DMEM/F-12). We used flow cytometry to evaluate cell death and real-time (RT-)PCR to determine the gene expression of aggrecan, collagen 2, and link protein, mediators of matrix degradation ADAMTS-4 and MMP3, the matrix protection molecule TIMP1, the cluster of differentiation (CD)44 receptor, the inflammatory cytokine IL-6 and Ank. We then determined the expression of specific apoptotic pathways in bovine NP cells by characterizing the expression of activated caspases-3, -8 and -9 in the presence of IL-1ß+FasL when cultured with NCCM, conditioned medium obtained using bovine NP cells (BCCM), and basal medium all supplemented with 2% FBS.
NCCM inhibits bovine NP cell death and apoptosis via suppression of activated caspase-9 and caspase-3/7. Furthermore, NCCM protects NP cells from the degradative effects of IL-1ß and IL-1ß+Fas-L by up-regulating the expression of anabolic/matrix protective genes (aggrecan, collagen type 2, CD44, link protein and TIMP-1) and down-regulating matrix degrading genes such as MMP-3. Expression of ADAMTS-4, which encodes a protein for aggrecan remodeling, is increased. NCCM also protects against IL-1+FasL-mediated down-regulation of Ank expression. Furthermore, NP cells treated with NCCM in the presence of IL-1ß+Fas-L down-regulate the expression of IL-6 by almost 50%. BCCM does not mediate cell death/apoptosis in target bovine NP cells.
Notochordal cell-secreted factors suppress NP cell death by inhibition of activated caspase-9 and -3/7 activity and by up-regulating genes contributing anabolic activity and matrix protection of the IVD NP. Harnessing the restorative powers of the notochordal cell could lead to novel cellular and molecular strategies in the treatment of DDD.
Degenerative disc disease (DDD) is an extremely common and costly healthcare condition for which there is no curative strategy . Given the lack of a biological strategy for regeneration of the degenerating disc, a therapeutic intervention that may offer restorative qualities to the disc is a much needed and widely sought goal. The ideal biological agent might reactivate homeostatic mechanisms innately inherent to the healthy intervertebral disc (IVD). The capacity to re-establish equilibrium between catabolic and anabolic tissue remodeling would represent the ideal regenerative strategy for the treatment of DDD.
With respect to potential biological therapies, lessons learned from the study of the non-chondrodystrophic canine (NCD canine) IVD might provide essential molecular clues for restoration of homeostasis to the disc. The NCD canine is unique among the canine sub-species in that this animal is relatively resistant to the development of DDD. Notably, NCD canines preserve their notochordal cell populations throughout life [2, 3]. Thus, there is an emerging body of evidence indicating that notochordal cells confer anabolic capacity upon NP cells and that their absence is associated with susceptibility to degenerative changes [2, 4, 5].
Apoptosis plays a central role in DDD development
Regulation of cellular turnover is vital to tissue homeostasis. Apoptosis is a highly regulated form of programmed cell death that classically involves two main pathways, the intrinsic (mitochondrial-dependent) and extrinsic (death receptor or Fas-dependent) pathways. It has been established that some cells, classified as Type I cells, function independently of the mitochondria and signal via Fas-induced apoptotic cell death involving the caspase-8 pathway. Other cells have a critical reliance upon the mitochondria whereby apoptosis is mediated via caspase-9 and are known as Type II cells [6, 7]. The initial explorations of these pathways involved the use of knock-out mice leading to the conclusions that some tissues are primed to respond to apoptotic stimuli in a Type I versus Type II manner [7, 8].
The classic extrinsic (CD95/Fas receptor) apoptotic pathway is activated by soluble Fas ligand (Fas-L) binding to the CD95 or Fas receptor that in turn activates caspase-8 followed by sequential activation of executioner caspases-7 and -3 resulting in cell death (type I cells) [9, 10]. In type II cells (such as disc cells), there is a form of 'cross-talk' between the extrinsic and intrinsic systems (involving mitochondria) whereby CD95/Fas receptor activation and subsequent caspase-8 activity may not reach the threshold necessary to activate the common executioner caspases-3/7. Bid, the BH3 interacting domain death agonist, serves as a vital intermediary in the 'cross-talk' that can occur between the intrinsic and extrinsic pathways . Bid activation results in degradation of the mitochondrial membrane by blocking the anti-apoptotic action of Bcl-2, an outer mitochondrial membrane protein [8, 9]. This loss of Bcl-2-mediated mitochondrial homeostasis leads directly to mitochondrial swelling and rupture of the organelle and release of cytochrome c into the cytosol [8–10]. Within the cytosol cytochrome c is directly involved with the formation of the apoptosome and the activation of caspase-9 that in turn activates downstream effector caspase-3, -7 and -6 resulting in cell death . In this study, we assessed the apoptotic pathway(s) involved in the IL-1β and FasL induced apoptosis of bovine NP cells under hypoxic conditions. The NP cellular and extracellular matrix (ECM) is a tightly regulated environment where homeostasis is maintained by the matrix metalloproteinases (MMPs), the ADAMTS family of enzymes (A disintegrin and metalloproteinase with thrombospondin motifs), tissue inhibitors of metalloproteinases (TIMPS), and regulation of inorganic pyrophosphate (PPi) transport by ANK [11, 12]. ANK (progressive ankylosis/human homologue of progressive ankylosis) is a transmembrane protein encoded by the ANK gene that controls pyrophosphate levels in cells and tissues of joints. The MMP and ADAMTS enzymes function in a similar fashion to those in skin and cartilage, and in an integrated fashion, balance anabolic and catabolic processes within the framework of ongoing tissue remodeling [11, 13]. Dysregulation of the activity of the MMPs, ADAMTS and TIMP enzymes results in increased catabolic activity and progression of DDD largely due to the influence of inflammatory cytokines, in particular IL-1β and the cell surface 'death receptor' Fas/CD95 [14, 15]. The inflammatory cytokine IL-1β is widely regarded as a key factor in the progressive degeneration of the IVD NP and FasL has been shown to sensitize disc cells to Il-1β mediated apoptosis-resulting in a synergistic action upon disc cells that increases with further degeneration . It has recently been demonstrated that with increasing disc degeneration the NP displays elevated ANK protein expression and these changes may be associated with hydroxyapatite deposition contributing to progression of degenerative disease . ANK is also known to preserve the differentiated chondrocyte phenotype (a critical element in the development of both DDD and osteoarthritis). In the presence of IL-1ß, chondrocytes express less hypertrophic chondrocyte-specific markers and are thought to lose their differentiated chondrocyte phenotype .
The progressive loss of viable cells and ECM integrity within the IVD NP is a pivotal process that depends upon a number of factors including cellular metabolism, cell number, nutrition of the disc (endplate integrity and diffusion characteristics), disc cell phenotype, and age [17–19]. With respect to IVD cellular content, the persistence of notochordal cells within the disc NP of certain animals such as the NCD canine is thought to contribute uniquely to resistance to DDD. Here for the first time we demonstrate that soluble factors secreted by notochordal cells (notochordal cell conditioned medium or 'NCCM') are capable of protecting NP cells from matrix protein degradation and pro-inflammatory cytokine secretion induced by IL-1ß and FasL. Furthermore, IL-1ß and FasL-mediated cell death is inhibited by NCCM via the inhibition of activated caspases -9 and -3. The ability to harness the restorative properties of the notochordal cell may lead to novel, molecular therapies in the treatment of DDD.
Materials and methods
Bovine NP cells
We obtained NP cells from the tails of 3-year old steers (bovine caudal discs) to be used as the 'target' cells in our experiments. Typically we harvested the NP from five to six bovine tails and enzymatically digested the NP overnight in order to obtain the cells. We removed the nucleus pulposus from the caudal discs of five to six bovine tails and after overnight digestion expanded the cells through two passages and then after suspension in 10% DMSO with 90% FBS (4 × 106 cells/mL) froze the cells at -80°C. We also developed bovine NP cell conditioned medium (BCCM) in exactly the same fashion as we obtained bovine NP 'target' cells but in the case of BCCM cultured the bovine NP cells within alginate beads.
Non-chondrodystrophic canine notochordal cells
We obtained the NP from the lower thoracic and entire lumbar IVDs from 11 non-chondrodystrophic canines (6 in the first series of serum-free experiments and 5 in the second 2% NCCM experiments) and after overnight digestion seeded the cells within alginate beads for subsequent culture.
We generated conditioned medium from both non-chondrodystrophic canine notochordal cells (NCCM) and bovine NP cells (BCCM). We then tested the ability of bovine NP conditioned medium or canine notochordal cell conditioned medium to protect bovine NP cells from cell death in the presence of IL-1β+Fas ligand (FasL). We also determined the ability of NCCM to protect the expression of salient extracellular matrix genes and other molecules under these same in vitro conditions using qRT-PCR methods. We used flow cytometry and activated caspase assays to determine protection from cell death and to discern the apoptotic pathways involved.
Non-chondrodystrophic canine discs
A total of 11 dogs was used in the experiments; 6 animals in the first series of serum-free experiments and 5 in the second 2% NCCM experiments. All animals were obtained in collaboration with a licensed animal facility and all practices were in accordance with the animal care policies and ethics approval board of Toronto Western Hospital. All animals were 8 to 12 months of age and had failed at adoption or were to be euthanized following biopharmaceutical intervention. Deep sedation was achieved using a combination of Acepromazine (Atravet-Aerst pharmaceuticals St. Laurent, Quebec, Canada 10 mg/mL) mixed with Xylazine 100 mg/mL (Xylomax-Bimeda-NHC Animal Health, Broomhill Road, Tallaght, Dublin 24, Ireland) at a combined dose of 1 mL/15 Kg body weight. Once deep sedation had occurred, euthanasia was accomplished using intravenous sodium pentobarbital (CDMV) (St. Hyacinthe, Quebec, Canada at a dose of 30 mL/kg body weight.
Tissue culture conditions
Notochordal cell recovery
For each NCCM-related experiment, within one hour of euthanasia, the lumbar spine was removed from six dogs in one experiment and five dogs in another. After soft tissue dissection the spines were treated with Clidox™ and Betadine™, the NPs were removed aseptically from six to eight IVDs of each animal and transferred to Advanced Dulbecco's Modified Eagle Medium/F-12 (ADMEM/F-12) supplemented with 100 units penicillin/streptomycin (PS) (basal medium) (Life Technologies, Carlsbad, Cal, USA). After removal of any residual annulus fibrosus, the NP tissues were enzymatically digested sequentially over night according to established methods previously reported [4, 5, 20]. The following day, the cells were filtered with a 70 μm cell strainer (Falcon). This cell preparation was found to be close to 100% notochordal in content, recognized by the classic physaliferous, large vacuolated appearance of the cells.
Serum-free notochordal cell conditioned medium (NCCM)
In order to determine the effect of notochordal cell-secreted factors without any exogenous stimulatory factors, we developed serum-free NCCM. Briefly, canine NP cells were mixed with 1.2% sodium alginate at a density of 1.5 × 106 cells/mL. Using a 21G needle and 5 mL syringe, the cell/alginate mixture was drop-wise added to 4 mL of 102 mM CaCl2 at a density of 80 beads/well within 6-well plates (6 × 106 cells in total). The beads were allowed to polymerize for 15 minutes within the CaCl2 mixture and then each well containing the beads was washed 3 times with 150 mM phosphate buffered saline (PBS). Finally, the wells containing the alginate-notochordal cell beads were placed under hypoxic tissue culture (3.5% O2, 37°C and 5% CO2) with ADMEM/F12 supplemented with 8% fetal bovine serum (FBS) and 100 Units (U) penicillin/streptomycin (PS) for two days (complete medium). After two days of culture in complete medium, the alginate beads containing the notochordal cells were thoroughly washed with PBS (15 minutes per wash for two hours). The medium was replaced with 6 mL/well (3 wells total) of basal medium (FBS-deficient ADMEM/F-12 containing 100 U PS) and after three more days of hypoxic culture serum-free NCCM was harvested and used for subsequent tissue culture.
2% FBS supplemented NCCM (2% NCCM)
In order to ensure that our caspase assay results would not be skewed by the notochordal cells depleting the available nutrients over the culture period, we generated 2% NCCM using basal medium supplemented with 2% FBS (2% basal medium) in order to provide a modest amount of nutrients to sustain the notochordal cells. For these experiments we obtained notochordal cells from five separate NCD canines, exactly as above and developed alginate beads containing notochordal cells at a density of 0.75 × 106 cells/mL in a volume of 3 mL/well (five wells). There is variable recovery of notochordal cells from NCD canines and from this series of animals there were fewer cells recovered than in the serum-free experiments. Therefore we reduced the concentration of the cells within the beads as well as the volume of medium used for these experiments in order to attempt to approximate the concentration of NCCM developed. The beads containing notochordal cells were cultured for two days in complete medium and then washed extensively using basal medium (4 washes × 15 minutes) and the medium replaced using 2% basal medium and cultured within 2% basal medium for three days before harvesting. We refer to these supernatants as 2% NCCM.
Bovine NP cells
Bovine NP cells (obtained earlier from caudal discs of five to six three-year old steers were released by enzymatic digestion, filtered, counted, plated and cultured under hypoxic conditions in T-75 flasks (CoStar-Corning Life Technologies, Corning, New York USA) in complete medium. The cells were allowed to expand to 70% confluence, harvested using Trypsin (Sigma-Aldrich Canada Ltd., Oakville, Ontario, Canada), and re-suspended in 90% FBS and 10% DMSO. Next the cells were sequentially frozen first at -80°C for two days and then transferred to liquid nitrogen. The day that the canine notochordal cells were harvested, we thawed the bovine NP cells in a 37°C water bath and after washing with complete medium the cells were plated and allowed to expand to 70% confluence at 37°C under hypoxic conditions of 3.5% O2 and 5% CO2. Coincident with the development of NCCM, the bovine NP cells were trypsinized, extensively washed with basal medium and re-seeded within 6-well plates at 0.5 × 106 cells/well and placed under hypoxia and allowed to expand for 3 days to 80% confluence using complete medium. Twenty-four hours prior to the NCCM-related experiments the plated bovine NP cells were extensively washed with basal medium and then cultured in basal medium in preparation for the NCCM experiments. For the 2% NCCM experiments, we initially thawed bovine NP cells and expanded them using complete medium, then washed them three times with 2% FBS supplemented basal medium in preparation for these experiments.
2% FBS supplemented bovine NP conditioned medium (2%BCCM)
In order to be sure that any beneficial effects conferred by NCCM were not simply a cell-induced artifact, we compared the effects of 2% NCCM to conditioned medium made from bovine NP cells also supplemented with 2% FBS. Therefore we obtained bovine NP cells from the caudal discs of five to six three-year old steers, seeded them within alginate beads, cultured them within basal medium supplemented with 2% FBS to develop '2% BCCM'. We then used 2% BCCM in exactly the same fashion as 2% NCCM to test its effect upon bovine NP cells.
Induction of apoptosis
We induced apoptosis of monolayer cultured bovine NP cells by the addition of 10 ng/mL recombinant bovine IL-1β (Thermo Scientific RP-87269 Nepean, Ontario, Canada) plus 10 ng/mL recombinant human Fas-ligand (FasL) (Alexis Biochemical Alx-522-001 San Diego, CA USA)). We then evaluated cell death using flow cytometry (after 48 hours) and activated caspase assays (after 24 hours). These time points were chosen after optimization experiments.
Protection from apoptosis and degradation
In order to determine if factors secreted by notochordal cells could specifically inhibit apoptosis and degradation of bovine NP cells, we cultured bovine NP cells with NCCM, 2% NCCM, and 2% BCCM, all in the presence of IL-1ß+FasL. For controls, we cultured bovine NP cells with basal medium for the flow cytometry experiments or with basal medium supplemented with 2% FBS; in all cases identical doses of IL-1β, or IL-1β + FasL were used. All experiments were performed at 37°C under hypoxic conditions of 3.5% O2 and 5% CO2.
Three 6-well plates containing bovine NP cells that had achieved approximately 80% confluence were divided such that each well of a 6-well plate received 4 mL/well of basal medium only (0% FBS medium), 1 plate received basal medium plus 10 ng/mL IL-1β + FasL/well, and one plate received 4 mL of serum-free NCCM + IL-1β+FasL/well. All samples were cultured under hypoxic conditions for 48 hours following which we determined differential cell death using flow cytometry.
In parallel to the flow cytometry experiments, bovine NP cells were cultured in two additional 6-well plates. Six wells of one plate received the same volume of serum-free basal medium (4 mL) and six wells in the other plate received serum-free basal medium + IL-1β + FasL. These plates were then processed to evaluate the induction of apoptosis using the Apo-Alert activated caspase-3 assay.
Caspase -8, -9 and -3/-7 assays
Separate experiments were performed using basal medium supplemented with 2% FBS, 2% NCCM and 2% BCCM and the expression of activated caspases was determined using the Caspase-Glo assay (Promega, Madison Wis, USA). We chose the Caspase-Glo assay for these experiments because this assay requires far fewer cells/assay. The 2% NCCM was generated from notochordal cells obtained from six to eight discs from five canine spines and the 'target' bovine NP cells were obtained from the discs of five to six three-year old steers. In this way we repeated our cell-death assays using cells obtained from entirely separate pools of cells from groups of animals in this case using activated caspase assays.
We obtained total RNA from the same sets of cultures from which we assessed cell death using flow cytometry (serum-free basal medium only (0% FBS medium), serum-free basal medium plus 10 ng/mL IL-1β+FasL/well, and serum-free NCCM+IL-1β+FasL/well. Briefly, after the 48-hour culture period total RNA was extracted from each of three wells for each condition using Trizol™ Life Technologies, Carlsbad, CA, USA) according to the manufacturer's recommendations. The RNA was purified, quantified using Nanodrop™ (Nepean, Ontario, Canada) and after reverse transcription, real-time RT-PCR evaluation for the genes of interest was performed.
Evaluation of Cell Death
We used flow cytometry (Becton Dickinson Dual Laser FacsCalibre (Mississauga, Ontario, Canada)) to evaluate the total extent of bovine NP cell death and apoptosis by incubating the cells in appropriate buffers and labeling with Propidium Iodide (PI) and Annexin-V (AV), (Sigma-Aldrich Canada Ltd., Oakville, Ontario, Canada) respectively, according to the manufacturer's instructions. AV is a protein that preferentially binds to phosphatidyl serine (PS) which undergoes specific changes during apoptosis and when conjugated to fluorescein isothiocyanate (FITC) it is capable of detecting early apoptosis via fluourescence imaging. PI is capable of entering the cell only when the plasma membrane has been dirsupted and is a non-specific marker of cell death which is often used together with AV to discriminate apoptosis from necrotic cell death. AV (to detect apoptotic cells) and PP (to detect dead/necrotic cells) were added to each sample according to the manufacturer's instructions and incubated at room temperature in the dark for 10 minutes. The cell harvest was filtered using 70 μm cell strainers (Falcon) just prior to flow cytometry analysis for which we used 'FacsCalibre'. Data analysis was performed using CellQuest Pro.
Flow Cytometry Optimization
In order to ensure that our flow cytometry results were reliable in the detection of apoptotic events we evaluated the expression of caspase-3 in cells treated with complete medium only, as well as complete medium plus IL-1ß + Fas. These cells were then processed according to the manufacturer's protocol and assayed by flurometry (ApoAlert Caspase-3 Fluorescent assay kit, ClonTech Labs inc., #630215 (Mountainview California, USA). The caspase-3 assay detects an alteration in the emission spectra of fluorescence of 7-amino-4-trifluoromethyl coumarin (AFC). The emission spectra shift from 400 nm to 505 nm after the AFC-labeled caspase-3-specific substrate is proteolytically cleaved by the enzyme and releases free AFC. Thus fluorescence intensity at 505 nm is directly proportional to activated caspase-3 and is representative of apoptotic events. These optimization experiments validated that our flow cytometry assay can reliably detect apoptosis in NP cells (Additional file 1).
Activated caspase -8, -9 and -3/7 assays
Flow cytometry can discriminate between non-apoptotic and apoptotic cell death but does not define the mechanisms of such cell death or protection from apoptosis in the case of NCCM treated cells. Therefore, we examined the expression of activated caspases-8, -9 and -3/7 using the Caspase-Glo assay (Promega). This assay was selected because it required far fewer cells compared to the ApoAlert assay used to validate our flow cytometry data allowing the relatively small amounts of NCCM generated to be used in quadruplicate. Briefly, in this assay the caspase enzyme specific luminogenic-substrate is cleaved by active caspases in the cell lysate releasing a substrate for luciferase. We plated bovine NP cells in 96 well plates at a density of 1 × 104 cells/well. The bovine NP cells were then incubated overnight with complete media under our hypoxic conditions. The following day the cells were washed and incubated with basal medium for three hours prior to treatment. There were three treatment groups: (1) bovine NP cells cultured with 2% ADMEM/F12 and treated with IL-1ß+FasL, (2) bovine NP cells cultured with 2% BCCM in the presence of IL-1ß+FasL and (3) NP cells cultured with 2% NCCM in the presence of IL-1ß+FasL. All treatment groups have a final volume of 100 ul/well. After the 24-hour treatment period, 100 ul caspase reagent was added to each well according to the experimental design. For caspase -8 and -9, the proteosome inhibitor MH-132, 60 uM, was added to the caspase reagent just prior to use. The plates were then incubated at room temperature for 75 minutes and luminescence detected with Perkin Elmer Victor3 Multilabel coulter (Model 1420-050). For background controls, we measured control luminescence from wells containing treatment media and the caspase reagent but no cells (Av lum - Av background lum = Relative Luminescence). All assays were performed in triplicate and the results examined statistically using the T-test.
Gene Expression Analysis
We wished to determine if serum-free NCCM could protect the expression of important extracellular matrix (ECM) genes as well as the pain- and inflammatory-related cytokine IL-6. We used real-time RT-PCR (qRT-PCR) to examine the expression of ECM genes, aggrecan, collagen type 2 and the cell-surface glycoprotein hyaluronic acid assembly site; CD44 and link protein. We also examined the expression of genes encoding matrix protecting factors such as TIMP-1, the principal matrix degrading enzymes MMP-3 and ADAMTS-4 and the expression of IL-6 by NP cells. Finally, we also examined the capacity of NCCM to affect the expression of the Ank gene by bovine NP cells treated with IL-1ß+FasL. We evaluated the expression of these genes in bovine NP cells cultured with basal medium, basal medium supplemented with IL-1ß, basal medium supplemented with IL-1ß+FasL and finally serum-free NCCM-treated NP cells also in the presence of IL-1ß+FasL.
Each primer pair was optimized for both optimal primer concentration and amount of cDNA in order to determine that amplification was within 15 to 25 amplification cycles. By this approach, we avoided non-linear/endpoint amplification data. We evaluated the expression profile for each primer pair used to ensure that there were no aberrant peaks present during the amplification phase of the PCR reaction. In addition to examining the expression of the various genes of interest when cultured with serum-free NCCM we also repeated our experiments using total RNA obtained from the 2% NCCM experiments.
Real-time RT-PCR Primers, annealing temperatures and Accession Numbers of genes evaluated
Annealing Temp °C
Collagen 2 F
Collagen 2 R
Link protein F
Link protein R
Normalized fold changes (HPRT) in gene expression by NP cells cultured in NCCM and treated with IL-1β+FasL
Normalized Fold Change in Gene Expression
Collagen Type 2
1.57 fold increase
65.5 fold increase
54.5 fold increase
1.76 fold increase
1.53 fold decrease
3.4 fold increase
4.8 fold increase
1.9 fold decrease
1.5 fold decrease
IL-1β and FasL mediated apoptosis in hypoxic bovine NP cell culture can be rescued by NCCM
We induced apoptosis of monolayer cultured bovine NP cells by the addition of 10 ng/mL recombinant bovine IL-1β plus 10 ng/mL recombinant human Fas-ligand (FasL) and evaluated cell death by flow cytometry (after 48 hours) and by the caspase assays (after 24 hours). These time points were chosen after optimization experiments. AV and PI labelled cells were quantified using flow cytometry allowing discrimination between viable/intact cells (AV-PI-), early apoptotic (AV+PI-) and late apoptotic or necrotic cells (AV+PI+). Cells were harvested from each of three conditions and after analysis using flow cytometry the percentage of dead cells in each condition was determined. Next from the dead cell populations the percentage of dead cells that were considered apoptotic was expressed as a function of the total percentage of dead cells (Figure 1). Treatment with NCCM resulted in a significant reduction in PI/AV positive cells that was similar to baseline levels with the rate of apoptotic cells indistinguishable from baseline (Figure 1, Panel c). We used an activated caspase-3 fluorometric assay to evaluate parallel bovine NP cells treated with IL-1β+FasL and found that this treatment induced apoptosis as demonstrated by the detection of activated caspase-3 (Additional File 1 Figure S1).
Impact of NCCM on genes encoding structural proteins and modulators of the extracellular matrix
Structural protein gene regulation
Extracellular matrix remodeling and anti-inflammatory cytokines
The expression of extracellular matrix remodeling genes was protected by the use of NCCM. MMP3 gene expression is up-regulated by IL-1β and IL-1β+FasL but the gene expression of this catabolic enzyme is reduced by 1.5-fold by NCCM. The activity of the MMP inhibitor TIMP-1 is up-regulated by NCCM 4.8-fold and a 3.4-fold increase was observed in the expression of ADAMTS4 (Table 2) [25–27].
The use of IL-1β+FasL led to a strong increase in the expression of the IL-6 gene but the addition of NCCM to the cultures markedly down-regulated IL-6 gene expression by almost 50% (1.9-fold reduction).
Ankgene expression and NP cell phenotype
The hallmark of progressive DDD is the inability of NP cells to maintain normal homeostatic tissue remodeling under the influence of increased cytokine and death receptor expression [11, 12, 18, 26]. A number of studies have strongly implicated IL-1β+FasL as key molecules in the progression of DDD . The purpose of this study was to determine the ability of NCCM to (1) counteract the deleterious gene expression changes induced by IL-1β+FasL in vitro; and (2) inhibit apoptosis of NP cells and determine the signaling pathways involved. Here we demonstrate for the first time that soluble factors secreted by notochordal cells strongly protect NP cells from IL-1β+FasL induced degeneration and apoptosis. Of significance NP cell apoptosis is inhibited by NCCM and is mediated in a caspase-dependent manner via suppression of activated Caspase -9 and -3/-7.
Gene expression which leads to the protective effect of NCCM
Progressive DDD is associated with increased expression of IL-1β [14, 15, 23, 29, 30] and FasL. We showed that IL-1β- and FasL-induced bovine NP cells under hypoxic conditions provide an appropriate system for the dissection of mechanisms underlying IVD degeneration. Our results indicated that changes induced by IL-1β are further stimulated by the addition of FasL. Thus, we used this in vitro system to identify the deleterious genes which can be counteracted by NCCM. Our RT-PCR results showed that the protective effect of NCCM is mainly due to (1) the inhibition of the transcript expression of a proinflammatory cytokine, IL-6, and a major collagenase, MMP-3 (likely via upregulation of TIMP-1, a natural inhibitor of MMP-3); and (2) the homeostatic regulation of matrix synthesis involving the maintenance of aggrecan (likely via increased expression of ADAMTS4), link protein and CD44 expression.
It has been shown that IL-6 expression is increased in DDD and is correlated with painful degenerative discs [31, 32]. As reported by others, treatment of NP cells with IL-1β alone or together with FasL led to upregulation of IL6 transcript expression . The use of NCCM downregulated its expression by more than 50%, suggesting that factors in NCCM, in part, inhibit the inflammatory aspect of disc degeneration.
2. MMP3 (stromelysin)
MMP-3 is a major collagenase involved in the irreversible degradation of collagen . TIMPs bind to MMPs and protect proteoglycan degradation . Our results showing up-regulation of TIMP-1 and down-regulation of MMP-3 transcripts, respectively, suggested that NCCM conferred an anti-degradative effective on the IL-1β- and FasL-treated MP cells.
3. ECM homeostasis
As part of the degenerative process, degraded link protein and proteoglycans accumulated in NPs much faster than in articular cartilage [34, 35]. Recently, aggrecan degradation and turnover has been shown to be a reversible event, indicating there might be a window of opportunity whereby proteoglycan degradation could be prevented before the irreversible MMP-mediated degradation occurs [21, 22]. It is possible to restore disc height and increase T2 MRI signals after trophic factors delivery to the disc following initial chemonucleolysis and/or needle puncture [35, 36].
We and others showed that IL-1β induced NP cells to up-regulate the transcript expression of molecules involved in ECM modeling (ADAMTS-4 and CD44) [14, 23]. Our results showing that NCCM upregulated ADAMTS-4 and CD44 expression may support the notion that some NCCM factors stimulate the remodeling process to preserve collagen type 2, aggrecan and link protein. The ADAMTS family of aggrecanases (notably ADAMTS-4 in the disc NP) functions in a reversible fashion to degrade the aggrecan core protein . Aggrecan plays an important role in the hydration of NP, mainly by binding to hyaluronic acid and being stabilized by link protein [32, 33]. Our observation that NCCM maintained the expression of aggrecan, CD44 and link protein reiterated that part of the protective effects of NCCM is mediated via maintenance of ECM homeostasis.
There is emerging evidence that Ank/ANKH plays a role in IVD degeneration. The best documented ANK function involves regulation of inorganic pyrophosphate (PPi) transport across the cell membrane . It is also involved in differentiation of cell lineages such as osteoblast and osteoclasts . In rat NP cells, hypoxia regulates Ank gene expression  and ANKH expression is increased in degenerated human IVDs , leading to the hypothesis that loss of homeostatic regulation of PPi is involved in degeneration and calcification in the IVD. In another study, IL-1β-treated chondrocytes in vitro resulted in a loss of differentiated chondrocyte phenotype (diminished Sox2, aggrecan and collagen II transcripts) and diminished Ank expression . We noted a similar effect upon bovine NP cells treated with IL-1β with or without FasL in which reduced Ank gene expression occurred and this down-regulated Ank expression was attenuated by NCCM. However, our experiments were short term only (24 to 48 hours) and were performed on single cell in vitro cultures devoid of the complexities and feedback from the extracellular matrix and long-term degenerative changes that occur in vivo. It remains to be investigated whether NCCM contains sufficient PPi levels that contribute partly to its protective effect. Nonetheless, our in vitro results showing down regulation of Ank expression in treated NP cells (when taken together with down regulated aggrecan and collagen type 2 gene expression) appear to be consistent with a loss of phenotype maintenance known to occur in chondrocytes treated with IL-1β. The notably partial rescue of Ank gene expression by NCCM is suggestive of a homeostatic/protective effect conferred upon NP cells by the components of notochordal cell-secreted factors.
Mechanistic dissection of the protective effect of NCCM on NP cell apoptosis induced by IL-1β and FasL
'Fold changes' in gene expression are dependent upon threshold calculations (ΔCt and ΔΔCt), therefore very small denominators could over-estimate such fold changes and must be interpreted with some caution. However, we observed similar gene expression changes for these two genes in entirely separate experiments using cells pooled together from completely independent animals. Therefore, when taken together with our previous findings that NCCM up-regulates NP aggrecan expression, it does appear that there is a strong protection of aggrecan and link protein conferred by the components of NCCM.
We developed NCCM from two entirely separate pools of non-chondrodystrophic canine notochordal cell sources. In the first set of serum-free NCCM we used 1.5 × 106 cells/mL alginate and in the second we used half that concentration. We have determined through past experience that a dose-dependent relationship exists between the concentration of notochordal cells and the biological effect of NCCM  and have considerable experience generating NCCM. In the second set of 2% NCCM experiments where we developed NCCM from 0.75 × 106 cells/mL alginate we reduced the volume of medium used to generate the NCCM by 50% from 6 to 3 mL thereby increasing the 'concentration' of the NCCM in an attempt to mitigate against the lower notochordal cell harvest from those animals. We observed similar gene expression effects upon the bovine NP cells (data not shown) and observed suppressed activated caspase-3 activity verifying that just as with the serum-free NCCM, the 2% NCCM retained its anti-apoptotic activity.
We have demonstrated in this study that NCCM protects against NP apoptosis via suppression of activated caspase-9 and -3/7. Possible mechanisms include stabilization of the mitochondrial membrane via inhibition of Bcl-2 activity, Bid activation or through the P53 growth factor-related pathway. It may be that components of NCCM suppress the formation of the apoptosome and in so doing suppress the formation of activated caspase-3 thereby preventing apoptotic cell death. The novel finding of the activation of the Ank gene by NCCM under degenerative/death-inducing conditions suggests a possible role for this gene in the maintenance of NP cell phenotype. These areas are under examination by our group.
The results of the present study provide evidence in vitro that notochordal cell-secreted soluble factors provide essential molecular signals that mediate apoptosis and degradation of NP cells induced by IL-1β+Fas-L. Harnessing the regenerative capacity of these cells and the important factors they secrete may lay the cornerstone of biological therapy for the treatment of degenerative disc disease.
a disintegrin and metalloproteinase with thrombospondin motifs
Advanced Dulbecco's Modified Eagle Medium with F-12 supplement
a transmembrane protein encoded by the ANKH gene that controls pyrophosphate levels in cells and tissues of joints although it is found in other tissues. A mutation at the (ank) locus results in generalized, progressive arthritis and joint destruction/ankylosis
bovine NP cells
B-cell lymphoma 2
BH3 interacting domain death agonist
cluster of differentiation 95 (also known as Fas-receptor)
degenerative disc disease
Dulbecco's modified Eagle's medium
soluble Fas receptor ligand
- Fas Receptor:
receptor for Fas-ligand also known as CD95
fetal bovine serum
notochordal cell conditioned medium
real time polymerase chain reaction
tissue inhibitors of metalloproteinases.
- Goetzel RZ, Hawkins K, Ozminkowski RJ, Wang S: The health and productivity cost burden of the "top 10" physical and mental conditions affecting six large U.S. employers in 1999. J Occup Environ Med. 2003, 45: 5-14. 10.1097/00043764-200301000-00007.View ArticlePubMedGoogle Scholar
- Aguiar DJ, Johnson SL, Oegema TR: Notochordal cells interact with nucleus pulposus cells: regulation of proteoglycan synthesis. Exp Cell Res. 1999, 246: 129-137. 10.1006/excr.1998.4287.View ArticlePubMedGoogle Scholar
- Oegema TR: The role of disc cell heterogeneity in determining disc biochemistry: a speculation. Biochem Soc Trans. 2002, 30: 839-844.View ArticlePubMedGoogle Scholar
- Erwin WM, Inman RD: Notochord cells regulate intervertebral disc chondrocyte proteoglycan production and cell proliferation. Spine. 2006, 31: 1094-1099. 10.1097/01.brs.0000216593.97157.dd.View ArticlePubMedGoogle Scholar
- Erwin WM, Ashman K, O'Donnell P, Inman RD: Nucleus pulposus notochord cells secrete connective tissue growth factor and upregulate proteoglycan expression by intervertebral disc chondrocytes. Arthritis Rheum. 2006, 54: 3859-3867. 10.1002/art.22258.View ArticlePubMedGoogle Scholar
- Nesrin Ozoren, El-Deiry Wafik S: Defining characteristics of types I and II apoptotic cells in response to TRAIL. Neoplasia. 2002, 4: 551-557. 10.1038/sj.neo.7900270.View ArticleGoogle Scholar
- Krammer Peter H: CD95's deadly mission in the immune system. Nature. 2000, 407: 789-795. 10.1038/35037728.View ArticleGoogle Scholar
- Yin X-M, Wang K, Gross A, Zhao Y, Zinkel S, Klocke B, Roth KA, Korsmeyer SJ: Bid-deficient mice are resistant to Fas-induced hepatocellular apoptosis. Nature. 1999, 400: 886-891. 10.1038/23730.View ArticlePubMedGoogle Scholar
- Jong-Beom Park, Jin-Kyung Lee, Sung-Jin Park, Ki-Won Kim, Riew KD: Mitochondrial involvement in Fas-mediated apoptosis in human lumbar disc cells. J Bone Joint Surg. 2005, 87-A: 1338-1342.Google Scholar
- Hengartner MO: The Biochemistry of Apoptosis. Nature. 2000, 407: 770-776. 10.1038/35037710.View ArticlePubMedGoogle Scholar
- Le Maitre C, Pockert A, Buttle DA, Freemont AJ, Hoyland JA: Matrix synthesis and degradation in human intervertebral disc degeneration. Biochem Soc Trans. 2007, 35: 652-655. 10.1042/BST0350652.View ArticlePubMedGoogle Scholar
- Skubutyte R, Markova D, Freeman TA, Anderson GD, Dion AS, Williams CJ, Shapiro IM, Risbud MV: Hypoxia-inducible factor regulation of ANK expression in nucleus pulposus cells: possible implications in controlling dystrophic mineralization in the intervertebral disc. Arthritis Rheum. 2010, 62: 2707-2715. 10.1002/art.27558.PubMed CentralView ArticlePubMedGoogle Scholar
- Hornebeck W: Down-regulation of tissue inhibitor of matrix metalloproteinase-1 (TIMP-1) in aged skin contributes to matrix degradation and impaired cell growth and survival. Pathol Biol (Paris). 2003, 51: 569-573. 10.1016/j.patbio.2003.09.003.View ArticleGoogle Scholar
- Le Maitre CL, Freemont AJ, Hoyland JA: The role of interleukin-1 in the pathogenesis of human intervertebral disc degeneration. Arthritis Res Ther. 2005, 7: R732-45. 10.1186/ar1732. EpubPubMed CentralView ArticlePubMedGoogle Scholar
- Cui LY, SL L, Ding Y, Huang DS, Ma RF, Huang WG, Hu BS, Pan QH: IL-1 beta sensitizes rat intervertebral disc cells to Fas ligand mediated apoptosis in vitro. Acta Pharmacol Sin. 2007, 28: 1671-1676. 10.1111/j.1745-7254.2007.00642.x.View ArticlePubMedGoogle Scholar
- Cailotto F, Sebillaud S, Netter P, Jouzeau JY, Bianchi A: The inorganic pyrophosphate transporter ANK preserves the differentiated phenotype of articular chondrocytes. J Biol Chem. 2010, 285: 10572-10582. 10.1074/jbc.M109.050534.PubMed CentralView ArticlePubMedGoogle Scholar
- Urban JP: The role of the physicochemical environment in determining disc cell behaviour. Biochem Soc Trans. 2002, 30: 858-864.View ArticlePubMedGoogle Scholar
- Urban JP, Smith S, Fairbank JC: Nutrition of the intervertebral disc. Spine. 2004, 29: 2700-2709. 10.1097/01.brs.0000146499.97948.52.View ArticlePubMedGoogle Scholar
- Shirazi-Adl A, Taheri M, Urban JP: Analysis of cell viability in intervertebral disc: effect of endplate permeability on cell population. J Biomech. 2010, 43: 1330-1336. 10.1016/j.jbiomech.2010.01.023.View ArticlePubMedGoogle Scholar
- Erwin WM, Las Heras F, Islam D, Fehlings MG, Inman RD: The regenerative capacity of the notochordal cell: tissue constructs generated in vitro under hypoxic conditions. J Neurosurg Spine. 2009, 10: 513-522. 10.3171/2009.2.SPINE08578.View ArticlePubMedGoogle Scholar
- Fitzgerald JB, Jin M, Grodzinsky AJ: Shear and compression differentially regulate clusters of functionally-related temporal transcription patterns in cartilage tissue-supplemental material. J Biol Chem. 2006, 281: 24095-24103. 10.1074/jbc.M510858200.View ArticlePubMedGoogle Scholar
- Nomura S, Yamaguchi MO, Wang TC, Lee JR, Goldenring JR: Alterations in gastric mucosal lineages induced by acute oxyntic atrophy in wild-type and gastrin-deficient mice. Am J Physiol Gastrointest Liver Physiol. 2005, 288: 362-375. 10.1152/ajpgi.00160.2004.View ArticleGoogle Scholar
- Le Maitre CL, Hoyland JA, Freemont AJ: Catabolic cytokine expression in degenerate and herniated human intervertebral discs: IL-1beta and TNFalpha expression profile. Arthritis Res Ther. 2007, 9: R77-10.1186/ar2275.PubMed CentralView ArticlePubMedGoogle Scholar
- Pockert AJ, Richardson SM, Le Maitre CL, Lyon M, Deaking JA, Buttle DA, Freemont AJ, Hoyland JA: Modified expression of the ADAMTS enzymes and tissue inhibitor of metalloproteinases 3 during human intervertebral disc degeneration. Arthritis Rheum. 2009, 60: 482-491. 10.1002/art.24291.View ArticlePubMedGoogle Scholar
- Patwari P, Gao G, Lee JH, Grodzinsky AJ, Sandy JD: Analysis of ADAMTS4 and MTW-MMP indicates that both are involved in aggrecanolysis in interleukin-1 treated bovine cartilage. Osteoarthritis Cartilage. 2005, 13: 269-277. 10.1016/j.joca.2004.10.023.PubMed CentralView ArticlePubMedGoogle Scholar
- Karsdal MA, Madsen SH, Christiansen C, Henriksen K, Fosang AJ, Sondergaard BC: Cartilage degradation is fully reversible in the presence of aggrecanase but not matrix metalloproteinase activity. Arthritis Res Ther. 2008, 10: R63-10.1186/ar2434.PubMed CentralView ArticlePubMedGoogle Scholar
- Sawaji Y, Hynes J, Vincent T, Saklatvala J: Fibroblast growth factor 2 inhibits induction of aggrecanase activity in human articular cartilage. Arthritis Rheum. 2008, 58: 3498-3509. 10.1002/art.24025.View ArticlePubMedGoogle Scholar
- Gurley KA, Reimer RJ, Kingsley DM: Biochemical and genetic analysis of ANK in arthritis and bone disease. Am J Hum Genet. 2006, 79: 1017-1029. 10.1086/509881.PubMed CentralView ArticlePubMedGoogle Scholar
- Hoyland JA, Le Maitre CA, Freemont AJ: Investigation of the role of IL-1 and TNF in matrix degradation in the intervertebral disc. Rheumatology. 2008, 47: 809-814. 10.1093/rheumatology/ken056.View ArticlePubMedGoogle Scholar
- Park JB, Park IC, Park SJ, Jin HO, Lee JK, KD R: Anti-apoptotic effects of caspase inhibitors on rat intervertebral discs. J Bone Joint Surg Am. 2006, 88: 771-779.View ArticlePubMedGoogle Scholar
- Burke JG, Watson RWG, McCormack D, Dowling FE, Walsh MG, JM F: Intervertebral discs which cause low back pain secrete high levels of proinflammatory mediators. J Bone Joint Surg Br. 2002, 84: 196-201. 10.1302/0301-620X.84B2.12511.View ArticlePubMedGoogle Scholar
- Kang JD, Stefanovic-Racic M, Mcintyre LA, Georgescu H, Evans H: Toward a biochemical understanding of human intervertebral disc degeneration and herniation. Contributions of nitric oxide, interleukins, prostaglandin E2, and matrix metalloproteinases. Spine. 1997, 22: 1065-1073. 10.1097/00007632-199705150-00003.View ArticlePubMedGoogle Scholar
- Studer RK, Gilbertson LG, Georgescu H, Sowa G, Vo N, Kang JD: p38MAPK inhibition modulates rabbit nucleus pulposus cell response to Il-1. J Orthop Res. 2008, 26: 991-998. 10.1002/jor.20604.View ArticlePubMedGoogle Scholar
- Donohue PJ, Jahnke MR, Blaha JD, Caterson B: Characterization of link protein(s) from human intervertebral disc tissues. Bioch J. 1988, 251: 739-747.View ArticleGoogle Scholar
- Pearce RH, Mathieson JM, Mort JS, Roughley P: Effect of age on the abundance and fragmentation of link protein of the human intervertebral disc. J Orthop Res. 1989, 7: 861-867. 10.1002/jor.1100070612.View ArticlePubMedGoogle Scholar
- Kim HJ, Minashima T, McCarthy EF, Winkles JA, Kirsch T: Progressive ankylosis protein [ANK] in osteoblasts and osteoclasts control bone formation and bone remodeling. J Bone Miner Res. 2010, 25: 1771-1783. 10.1002/jbmr.60.PubMed CentralView ArticlePubMedGoogle Scholar
- Schimmer AD, Dalili S, Batey RA, Riedl SJ: Targeting XIAP for the treatment of malignancy. Cell Death Differ. 2006, 13: 179-188. 10.1038/sj.cdd.4401826.View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.