This protocol presents a novel experimental model of proinflammatory, degenerative bovine organ culture to simulate early-stage intervertebral disc degeneration.
Symptomatic intervertebral disc (IVD) degeneration (IDD) is a major socioeconomic burden and is characterized by inflammation and tissue degradation. Due to the lack of causative therapies, there is an urgent need for innovative experimental organ culture models to study the mechanisms involved in the progression of the disease, find therapeutic targets, and reduce the need for animal models. We here present a novel, three-dimensional organ culture model protocol mimicking the proinflammatory and catabolic microenvironment, which is present during IDD.
Initially, bovine caudal IVDs were dissected, cleaned, and cultured in the tissue culture medium. Dynamic physiologic or pathologic loading was applied in a custom-made bioreactor for 2 hours per day. IVDs were assigned to a control group (high glucose medium, physiological loading, phosphate-buffered saline injection) and a pathological group (low glucose medium, pathological loading, tumor necrosis factor-alpha injection) for four days. Gene expression analysis from collected nucleus pulposus cells of the IVDs and enzyme-linked immunosorbent assay of the conditioned organ culture media was performed.
Our data revealed a higher expression of inflammatory markers and reduced disc heights after loading in the pathological group compared to the control group. This protocol is reliable to simulate IVD inflammation and degeneration and can be further expanded to broaden its application scope.
Low back pain (LBP) can affect individuals of all ages and is a leading cause for disability worldwide1,2,3. The total cost associated with LBP exceeds $100 billion per year4,5. Symptomatic intervertebral disc (IVD) degeneration (IDD), a condition characterized by inflammation and tissue degradation, is a major cause of LBP6,7. Specifically, IDD is characterized by a gradually evolving breakdown of the IVD's extracellular matrix (ECM), induced and triggered by multiple factors that lead to an accelerated pathology, neurological disorders, and eventually disability. Furthermore, IDD is associated with the release of proinflammatory cytokines, altered spine biomechanics, angiogenesis, and nerve ingrowth, which increases pain sensation, altogether causing chronic LBP (active discopathy)6,8. To date, treatment options include discectomy and subsequent fusion of the adjacent vertebrae, implantation of an IVD prosthesis, or non-surgical approaches, such as non-steroidal anti-inflammatory drugs, opioids, and muscle relaxants for patients with IDD9. Both current standard therapeutic options, surgical and non-surgical, are only partly effective and fail to address the underlying biological problem9,10. Early-stage degenerative disc disease is characterized by an initial inflammatory tissue response, especially an increase in tumor necrosis factor-alpha (TNF-alpha) expression11. These early disc changes primarily occur at the cellular level without disrupting the disc architecture and could previously be mimicked by nutritional deficiency under pro-inflammatory conditions12. Therefore, precise simulation of the in vivo situation to investigate these degeneration mechanisms and find suitable therapeutic targets is crucial. Additionally, to these simulations of molecular properties, the mechanical loading environment of the discs plays a key role in pathological and physiological changes of IVD. Consequently, combining these approaches would bring us one step forward to mimic the complex microenvironment of IVDs in vivo. There are currently no studies considering the aspect of dynamic loading along with the pro-inflammatory and nutritional setting to the best of our knowledge.
Although large animal models allow the investigation of potential relevant in vivo interactions, they are costly and work intensive. Moreover, as the use of animal models in research has long been a matter of controversy, the reduction of the number of animals needed to answer important research questions is of great interest. Finally, there is currently no ideal animal model to mimic IDD in IVD research13,14. Therefore, it is necessary to establish a cost-effective and reliable replacement, such as an organ culture model to simulate IDD and associated inflammatory and degenerative processes. Recently, the application of the present protocol on the establishment of a proinflammatory and degenerative organ culture model to simulate early-stage intervertebral disc disease allowed us to investigate the effect of anti-inflammatory drugs in the IDD organ culture15.
Here, we describe how to obtain bovine intervertebral discs and induce the state of early-stage IDD via a catabolic and proinflammatory microenvironment caused by direct intradiscal injection of tumor necrosis factor-alpha (TNF-α) and degenerative loading in a bioreactor under low nutritive medium conditions. Figure 1 illustrates the experimental model and shows the bioreactor used to simulate degenerative and physiological loading conditions.
Figure 1: Illustration of the experimental setup. A: bovine tail; B: dissected bovine intervertebral discs; C: transfer of the disc to a well-plate with culture medium; D: loading the simulation in a bioreactor; E: intradiscal injection technique; F: IVD after injection of PBS/trypan blue dye to reveal distribution. IDD: intervertebral disc degeneration. Please click here to view a larger version of this figure.
Experiments were performed using bovine tails obtained from local abattoirs. The biological materials used in the current study are taken from the food chain and require no ethical approval in Swiss and European law.
1. Dissection of the bovine intervertebral disc
2. IVD culture and loading
3. Intradiscal tumor necrosis factor-alpha (TNF-α) injection
4. Gene expression
5. Quantification of protein content in the IVD medium
Degenerative loading in low glucose medium combined with TNF-α injection caused a significant increase of the gene expression of proinflammatory markers interleukin 6 (IL-6) and interleukin 8 (IL-8) compared to the physiological control group in NP cells after 4 days of culture (Figure 2). In contrast, we did not observe significant changes for the proinflammatory genes interleukin 1β (IL-1β) and TNF-α in NP cells (data not shown). Furthermore, degenerative culture conditions did not alter the gene expression of IL-6 and IL-8 in AF cells.
Figure 2: Gene expression levels in the nucleus pulposus (NP) tissue and annulus fibrosus (AF) tissues. These were measured after 4 days of culture under physiological or pathological culture condition, normalized to baseline (day 0). Means ± 95% confidence interval n=8, **p<0.01. This figure has been modified from20. Please click here to view a larger version of this figure.
Consistent with the gene expression findings, the IL-8 protein content in the medium showed a marked increase after 2 days and 4 days compared to the physiological condition (Figure 3). However, the measurements at day 3 (after free-swelling or loading) revealed no significant differences between the study groups, although the results indicated higher values for the degenerative group.
Figure 3: Quantification of the pro-inflammatory protein IL-8 release in the IVD culture medium after free swelling culture (FS) and dynamic loading (Load). The results are shown as the original concentrations in ng/mL in the medium without normalization. Mean ± 95% confidence interval, n=8, **p<0.01. This figure has been modified from20. Please click here to view a larger version of this figure.
The disc height (DH) changes normalized to the DH after dissection are shown in Figure 4. Whereas DH reductions after loading revealed higher values (i.e., more height reduction) for the degenerative group compared to the physiological groups, no differences in disc height gains after the free-swelling period were seen between the study groups. This indicated that the difference in disc height changes between the pathological and physiological group was higher after the loading procedure. Furthermore, the differences were less pronounced after 1 day compared to the measurements on days 2 and 3, indicating a progressive effect of the degenerative and inflammatory conditions.
Figure 4: Disc height changes normalized to the baseline values (after dissection). Mean ± 95% confidence interval. FS: free-swelling. N=10, ***p<0.001. This figure has been modified from20. Please click here to view a larger version of this figure.
We here provided a detailed protocol to simulate degenerative and inflammatory IVDD. This protocol can be applied for detailed examinations of inflammatory pathways leading to the destructive effects on the disc. Moreover, the protocol can help to determine promising therapeutic targets involved in the progression of the disease.
We recently showed that human recombinant TNF-α could induce inflammation in both bovine and human NP cells21, which is in accordance with other studies in the field confirming that TNF-α can be used for inflammation simulation in IVD cells22,23. In the present study, a dose of 100 ng TNF-α per IVD was used to induce inflammation. This dose showed effective induction of inflammatory markers when applied in combination with degenerative loading and limited nutrition15,20. In a recent study using TNF-α as the sole initiation factor of IVD degeneration, a threshold of 100 ng TNF-α / cm3 disc volume has been determined as an effective dose for the induction of inflammatory and degenerative changes in the IVD21. It is suggested that the TNF-α dose used to induce inflammation should be normalized to the volume of the disc for reproducible effects. Furthermore, it should be ensured that there is a good distribution of the injection material in the IVD and that this injection will be adequately reproducible in subsequent experiments. As injection pressure may vary between individuals, there might be different distributions of the material among the same study groups in different experiments. One approach we chose to examine the equal distribution was using PBS diluted trypan blue injections. It is recommended to standardize the injection technique, for example, with injection pumps and predefined and reproducible injection rates. As shown previously, one single needle puncture with 22-gauge, 25-gauge, or 30-gauge needles into the IVD did not cause an effect that interacted with study outcomes20,21,24,25. It is recommended to consider the disc size for the volume of the substance to be injected and for the needle size used for injection. Further, it is recommended to use a single injection to avoid inducing potential degenerative effects caused by multiple annular injections21,26.
Some limitations of the present model need to be addressed. The loading of the IVDs requires access to bioreactors, which are currently not widely available in many labs working on IDD. However, the need for preclinical IVD degeneration models is rising. Organ culture models bring ethical benefits of reducing the need for animal models, and the one-time investment on bioreactor costs could be affordable considering the reproducibility of the degenerative stimulation, and the reductions in the costs associated with animal experiments. Nevertheless, some in vivo interactions, such as the role of the immune system, cannot be simulated with the provided organ culture model. Moreover, the measurement of nerve signals and the evaluation of pain that would be possible in animal models can currently not be considered with the current model. Some concerns were raised whether the mechanical properties of tail IVDs could be comparable with human IVDs as the upright spine position is unique in humans28. Anatomical and molecular properties of bovine tail IVDs, such as cell density, lack of notochordal cells, biochemical composition29,30, and mechanical properties of bovine caudal IVDs, such as range of motion in flexion, extension, torsion, and bending31 are very similar to human IVDs. Notably, human IVDs gain more and more attention recently for organ culture models28. Human cadaveric IVDs in ex vivo models are considered highly efficient as they can avoid species differences which are more clinically relevant32. In contrast to bovine IVD tails obtained from the abattoirs, this would require transplantation of human IVDs 24 h post mortem and, thus, ethical approval following local organ transplantation laws28. The present method can also be easily adapted to other species.
One major advantage of the technique compared to other methods of ex vivo cultures is the combination of intradiscal injection, pathological medium conditions, and detrimental loading to simulate the early stage of degenerative discs better. Walter et al.23 used TNF-alpha in the medium of dynamically loaded bovine IVDs instead of injection into the IVDs and showed increased transport of TNF-alpha from the medium into the nucleus pulposus compared to the annulus fibrosus, which is in accordance with our results. As we aimed to simulate the early stages of IDD, which occurs at the cellular level without major disruptions of the discs architecture12, we chose the TNF-alpha concentrations based on previous organ culture studies using TNF-alpha in the medium to simulate the early stages of IDD12,22,23. The use of other inflammatory stimulants and higher concentrations of TNF-alpha could be tested to simulate the desired degenerative disc state depending on the study question. Higher concentrations of TNF-alpha may better compensate the lack of systemic immune regulatory feedbacks present during disc degeneration as proposed by Ponnappan et al.12 The use of detrimental medium conditions with low-glucose was based on previous work showing that nutritionally limiting media and exposure to TNF-alpha mimics molecular change characteristics which are available in early disc degeneration state12, 27. Thus, the approach combines the evidence provided by previous studies in one ex vivo organ culture model of early degenerative disc disease.
This protocol can be further improved in several ways. The duration and extent of the loading and free-swelling period can be chosen based on the desired study plan. Long-term effects of inflammatory or degenerative stimulations on disc degeneration are of high clinical interest and can be accomplished with the current protocol. We used only selected genes considered highly relevant for the early state of IDD11. Additional validation of this model could include a widened assessment of genes and proteins, such as omics approaches. Consequently, other inflammatory factors could be investigated in combination with the degenerative loading, depending on the desired degenerative state. For example, lipopolysaccharide injections have been shown to stimulate inflammation in the IVD33. A comparison of different inflammatory stimulants can be accomplished with the present protocol to find the most appropriate inflammatory stimulant for the desired study question. We have evaluated changes of selected inflammatory markers (IL-6, IL-8), anabolic markers (aggrecan, collagen), and catabolic markers (matrix metallopeptidases, a disintegrin and metalloproteinases with thrombospondin motifs) with the present protocol20. As the whole gene expression profile might change, future examinations could include next-generation RNA sequencing techniques to determine novel biomarkers for disc degeneration diagnostics and therapeutic targets for early biological intervention. Furthermore, other methodologies, such as histology, immunohistochemical staining, biochemical measurements of extracellular matrix components (such as glycosaminoglycans, GAGs), and dynamic compressive tests can be performed to further analyze the biological, biochemical, and biomechanical properties of IVDs with the present model. Another possible analytical way would be to separately analyze the impact on outer AF and NP tissue by using bCol1a1, Col1a2, CD146, SM22α, and MKX as gene expression markers for the outer AF34,35. The Spine Research Interest Group at the 2014 Annual ORS Meeting in New Orleans recommended following healthy NP phenotypic markers: HIF-1alpha, GLUT-1, aggrecan/collagen II ratio >20, Shh, Brachyury, KRT18/19, CA12, and CD2436. The NP marker genes bBrachyury/T and bSecreted frizzled-related protein 2 were the most convincing separating NP from inner and outer AF tissue in the bovine IVD34. Furthermore, the sGAG content of the NP was reported to be significantly higher compared to the outer AF34.
In conclusion, this novel proinflammatory and degenerative IVD organ culture model provides relevant conditions to simulate early-stage IDD in a highly relevant 3D microenvironment. Certainly, the present protocol can be further modified according to the investigator's objectives. Furthermore, our model is able to reduce the number of needed study animals and thus totally reflects the 3R principles of reduce, replace, refine.
The authors have nothing to disclose.
This work was supported by AO Foundation and AOSpine International. Babak Saravi received fellowship support from the German Spine Foundation and the German Osteoarthritis Foundation. Gernot Lang was supported by the Berta-Ottenstein-Programme for Advanced Clinician Scientists, Faculty of Medicine, University of Freiburg, Germany.
1-Bromo-3-chloropropane(BCP) | Sigma-Aldrich, St. Louis, USA | B9673 | |
Ascorbate-2-phosphate | Sigma-Aldrich, St. Louis, USA | A8960 | |
Band saw | Exakt Apparatebau, Norderstedt, Germany | model 30/833 | |
Betadine | Munndipharma, Frankfurt, Germany | ||
Bovine IL-8 Do.it-Yourself ELISA | Kingfisher Biotech, St. Paul, USA | DIY1028B-003 | |
Corning ITS Premix | Corning Inc., New York, USA | 354350 | |
DMEM high glucose | Gibco by life technologies, Carlsbad, USA | 10741574 | |
DMEM low glucose | Gibco by life technologies, Carlsbad, USA | 11564446 | |
Ethanol for molecular biology | Sigma-Aldrich, St. Louis, USA | 09-0851 | |
Fetal Bovine Serum (FBS) | Gibco by life technologies, Carlsbad, USA | A4766801 | |
Non-essential amino acid solution | Gibco by life technologies, Carlsbad, USA | 11140050 | |
Penicillin/Streptomycin(P/S) | gibco by life technologies, Carlsbad, USA | 11548876 | |
Phosphate Buffer Solution, tablet | Sigma-Aldrich, St. Louis, USA | P4417 | |
Pronase | Sigma-Aldrich, St. Louis, USA | 10165921001 | |
Primocin | InvivoGen, Sandiego, USA | ant-pm-05 | |
Pulsavac Jet Lavage System | Zimmer, IN,USA | ||
TissueLyser II | Quiagen, Venlo, Netherlands | 85300 | |
Streptavidinn-HRP | Kingfisher Biotech, St. Paul, USA | AR0068-001 | |
Superscript VILO | Invitrogen by life Technologies, Carlsbad, USA | 10704274 | |
cDNA Synthesis Kit | Applied Biosystems by life technologies | 10400745 | |
TaqMan Universal Master Mix | Applied Biosystems by life technologies | ||
TNF-alpha, recombinant human protein | R&D systems, Minnesota, USA | 210-TA-005 | |
TRI Reagent | Molecular Research Center, Cincinnati, USA | TR 118 | |
Tris-EDTA buffer solution | sigma-Aldrich, St. Louis, USA | 93283 | |
Gene bIL-6 | Applied Biosystems by life technologies | Custom made probes | Primer fw (5′–3′) TTC CAA AAA TGG AGG AAA AGG A Primer rev (5′–3′) TCC AGA AGA CCA GCA GTG GTT Probe (5′FAM/3′TAMRA) CTT CCA ATC TGG GTT CAA TCA GGC GATT |
Gene bIL8 | Applied Biosystems by life technologies | Bt03211906_m1 | |
Gene bTNF-alpha | Applied Biosystems by life technologies | Custom made probes | Primer fw (5′–3′) CCT CTT CTC AAG CCT CAA GTA ACA A Primer rev (5′–3′) GAG CTG CCC CGG AGA GTT Probe (5′FAM/3′TAMRA) ATG TCG GCT ACA ACG TGG GCT ACC G |
GENE bIL1beta | Applied Biosystems by life technologies | Custom made probes | Primer fw (5′–3′) TTA CTA CAG TGA CGA GAA TGA GCT GTT Primer rev (5′–3′) GGT CCA GGT GTT GGA TGC A Probe (5′FAM/3′TAMRA) CTC TTC ATC TGT TTA GGG TCA TCA GCC TCA A |
RPLP0 | Applied Biosystems by life technologies | Bt03218086_m1 |