Intervertebral disc degeneration and how it leads to low back pain.

An aspect that has emerged in the last decade relevant to the development of prospective animal models of IVDD is the issue of the potentially confounding impact that sex hormones may have on such models. Although the effects of sex hormones on the metabolism of IVD cells was first identified in 1969 59 it is only in the last decade that these have been shown to significantly impact on degenerative processes in the IVD. 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 , 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 In order to avoid biological spread and cyclical variation in data that would complicate data analysis and development of a prospective IVDD model, female animals may be examined in a separate data set from male data to ensure any pathophysiological sexual dimorphism is identified. Ideally this would include both actively cycling and “post‐menopausal” (likely gonadectomized) female cohorts to include study of the cyclical fluctuations in circulating female sex hormone levels. Male sex hormones also effect IVD cells but cyclical circulatory hormonal fluctuations do not occur to the same degree. 61 LBP is a common symptom of premenstrual syndrome, experienced by most women during menstruation and may be exacerbated by premenstrual dysphoric disorder and dysmenorrhea or may be a symptom of endometriosis. 78 , 79 Female sex hormones play an important role in the etiology and pathophysiology of a number of musculoskeletal degenerative diseases, around 70% of perimenopausal women will experience LBP symptoms due to estrogen deficiency, estrogen decrease may be a risk factor for lumbar disc degeneration. 80 , 81 , 82 , 83 , 84 , 85 , 86 Postmenopausal women show accelerated IVDD due to relative estrogen deficiency, increased prevalence of spondylolisthesis, and facet joint osteoarthritis, in the first 15 years post menopause. 80 , 84 , 85 Continued progression of lumbar disc degeneration in postmenopausal women has been observed. 85 Estrogen signals through two classic nuclear receptors, estrogen receptor (ER)‐α and ‐β, and a membrane bound G‐protein‐coupled receptor 30 (GPR30). 17β‐estradiol (E2) enhances cell proliferation and prevents IL‐1β‐induced cell death in IVD cells, but this effect was partially blocked by G36, a GPR30 antagonist and completely abrogated by a combination of ER antagonists ICI 182 780 and G36. 73 This demonstrated that GPR30 was expressed in human IVD cells and transmitted signals triggering E2‐induced NP cell proliferation and also had a protective effect against IL‐1β‐induced disc cell apoptosis. The effects of E2 on NP cells required both GPR30 and classic estrogen receptors. During IVDD apoptotic effects are key factors responsible for a reduction in NP cell numbers and significantly effects the viability of IVD tissues. 87 Prevention of NP cell apoptosis thus represents an important preventative measure that could slow down the progression of IVDD. E2 promotes human AF cell proliferation, 88 and exerts anti‐apoptotic effects in rat disc cells 71 , 75 , 76 thus it represents a possible therapeutic that could significantly reduce the development of IVDD and generation of LBP. Estrogen can, thus, prevent the development of IVDD through its anti‐apoptotic properties inhibiting the production of the inflammatory cytokines IL‐1β and TNF‐α by disc cells. 89 , 90 This reduces catabolic events in the IVD by preventing the up‐regulation of MMPs induced by these inflammatory mediators. Estrogen also induces anabolic processes in the IVD by activating the PI3K/Akt pathway and also decreases oxidative damage. 89 By inhibiting IVDD, estrogen exerts protective effects that prevent degradative structural changes in the IVD that would otherwise pre‐dispose the IVD to nerve ingrowth and production of neurotrophic factors and inflammatory mediators by IVD cells that contribute to IVD nociceptor activation and mechano‐sensitization of disc afferent nerve fibers leading to the generation of LBP. 91 , 92 Additional factors such as geometry and the influence of the inflammatory system beyond hormonal differences can also have a significant influence on IVDD and LBP. Further studies with functional foods and neutraceutical supplements under evaluation for their abilities to alleviate pain may also prove to be a useful non‐drug treatment for post‐menopausal back pain. 93 A recent study has reviewed the extensive range of natural compounds which display anti‐inflammatory NP cell protective anti‐catabolic properties with potential roles in IVD regenerative processes. 94

Literature reviews in PubMed and Google were used to collect information on the cited studies in Table  3 using search terms such as “IVD degeneration,” “low back pain,” “models of IVD degeneration,” “small animal models of IVD degeneration,” “large animal models of IVD degeneration.” The studies cited were selected by the authors to illustrate examples of large and small animal models of IVDD and the diversity of the models developed and was not intended to provide comprehensive coverage of all IVD models of IVDD that have so far been developed which are extensive. A complete coverage was considered outwith the scope of this review; however, additional information is available on these in two further studies. 94 , 227 Illustrative examples of large and small animal models of IVD degeneration 24G needle stick AF lesion rat IVDD model Needle stick injury plus dynamic compression rat IVDD model Potential patho‐mechanism of LBP Rat bent tail sustained static compression model of IVDD Induces mild degeneration Murine tail IVDD (i) External tail IVD static compression device (ii) Mouse compressive suture model of IVDD Spontaneous IVDD models (i) Sand rat (ii) Chondrodystrophoid (ChD) canine (iii) non‐ChD canine (iv) SM/J Mouse 234 , 235 , 236 , 237 238 , 239 , 255 , 256 , 257 , 258 113 Rat multifidus resection Model of IVDD Murine paraspinal muscle surgical lesion IVDD model 242 241 Rabbit annular puncture induced biomechanical destabilization Rat xenograft radiculopathy model A number of rodent models have been developed to specifically assess various aspects of IVDD and the development of LBP (Table  3 ). Several rodent models of IVDD have been developed by altering the normal biomechanical forces experienced by the IVD using (i) annular needle puncture, 228 (ii) Ilozarov type cages which maintain tails in a bent position, 229 , 230 (iii) use of IVD sutures which compress the IVD and (iv) by compressively loading the IVD. Each of these procedures produce IVDD to variable degree. Procedure (i) and (iv) have also been used in combination to induce a more extended degenerative response in the IVD. 183 Procedures (ii) and (iii) were developed to provide milder forms of IVDD. The ovine large annular lesion (6 × 20 mm) of IVDD is a versatile model that reproduces many of the degenerative features described in human IVD. 37 , 38 , 39 The relatively large IVDs of this model allows zonal analyses to be undertaken, a feature not possible for the rodent models of IVDD. The effects of IVDD on erector spinae muscles have also been investigated using this model demonstrating deleterious effects on spinal flexibility and neuromuscular control of spinal muscle groups which result in a diminished control of co‐ordinated bodily movement. 33 , 40 , 41 , 42 , 43 The resultant effects on altered loading of degenerate IVDs results in further mechanical stimulation of nociceptors and mechanoreceptors and generation of LBP. A rabbit model of intermittent cyclical loading to the IVD results in damage to the CEP and an increase in the calcification of this tissue with a negative impact on the nutritional status of disc cells contributing to IVDD. 231 , 232 , 233 Several spontaneous models of IVDD have also been developed in the sand rat, which is a gerbil. 234 , 235 , 236 , 237 The chondrodystrophoid and non‐chondrodystrophoid canine are interesting IVDD models with the former displaying early degenerative IVD changes correlating with an early disappearance of notochordal cells while in the latter IVDD occurs much later and is less prominent and notochordal cell numbers are maintained into old age. 238 , 239 The SM/J poor healer mouse is an inbred small mouse strain that displays age dependent early‐onset, spontaneous disc degeneration characterized by a lack of cells in the NP, a clear increase in hypertrophic chondrocyte‐like cells in the CEPs, and clefts throughout the IVD. 240 Rat and mouse multifidus resection models induce spinal instability and IVDD demonstrating the interdependence of the IVD and spinal muscles for spinal stability, flexibility and optimal weight bearing. 241 , 242 A single level annular puncture model of IVDD, rat xenograft radiculopathy model 243 and rat multi‐level models of IVDD have been developed and used to evaluate the properties of growth and differentiation factor‐6 (GDF6) as a therapeutic agent to combat IVD degenerative processes. 244 GDF6 significantly inhibited the expression of TNFα and IL‐1β in rat IVDs by 8 weeks after annular puncture and had a protective effect on IVD structure and morphology 32 weeks post induction of IVDD. 243 , 245 GDF6 also reduced mechanically mediated pain behavior and inhibited the expression of the inflammatory mediators TNFα and IL‐1β and CGRP in the DRG model. 243 Furthermore, a rat inflammatory protein array showed GDF6 reduced the expression of IL‐6, intercellular adhesion molecule‐1 (ICAM‐1), matrix metalloproteae‐13 (MMP‐13), TNFα, IL‐1β and increased the expression of transforming growth factor β2 (TGF‐β2), IL‐10 and the adipokine resistin (RETN, adipose tissue‐specific secretory factor or C/EBP‐epsilon‐regulated myeloid‐specific secreted cysteine‐rich protein) in a TNFα and IL‐1β stimulated disc cell culture system. 243 , 244 , 245 RETN is a modulatory peptide that regulates inflammation in pathological human tissues. 76 Thus, GDF6 can improve the structure of the IVD, inhibit the expression of inflammatory and pain related factors, and improve pain behavior in rats. 243 , 244 , 245 Human clinical trials with GDF6 are eagerly awaited since it has considerable potential in the prevention of further deterioration in degenerate IVDs, has regenerative properties that promote recovery of normal IVD architectural functional organization and inhibits the generation of inflammatory mediators that lead to IVDD and the generation of LBP. 243 , 244 , 245 This is evident in rat IVDD models by a reduction in pain related factors and in mechanically induced pain behavior. The rat multi‐level IVDD model that has also been used to evaluate the beneficial properties of GDF6 is a novel development in animal models of IVDD. 244 IVD cells have evolved to exist in a weighted environment and biomechanical forces have important regulatory effects on disc cells. 224 , 259 The IVD is subject to cycles of compression/relaxation imposed through the axial skeleton and normal bodily movements such as flexion/extension and torsional twisting and bending movements. Indeed, these cycles of compression/relaxation are an important pumping mechanism that promotes diffusion of nutrients to IVD cells and removal of their metabolic waste products in the normal IVD. 260 , 261 Structural alterations in the IVD which impede this nutritional pathway can lower disc cell viability and contribute to the development of IVDD. 225 , 262 , 263 The animal IVDD studies we have cited provide valuable insights into the complexities of IVDD and the generation of LBP. Structural analysis of ECM organization in the normal IVD shows that NP cells predominantly receive compression with some shear component when the IVD is in torsion while AF cells occupy a fibrocartilaginous ECM designed to accommodate radial tensional hoop stresses arising from axial compression and bulging of the NP resulting in outward bulging of the AF. 225 In the degenerate IVD depleted of its space‐filling aggrecan which normally provides weight bearing properties, the normal weight bearing/tensional stresses the IVD cells are exposed to result in increased bulging of the less supportive NP, a reduction in disc height, greater bulging of the AF and a reduction in disc cell viability. 263 The lamellar collagenous structure is weak in compression and may become inverted, adjacent collagenous lamellae may even separate (de‐lammelation) resulting in the formation of internal clefts and fissures in the IVD. 264 When these clefts communicate with the outer AF herniation of the NP may occur, and ingrowth of nerves and blood vessels may occur through these clefts into the normally avascular aneural IVD setting up a scenario where the IVD can no longer adequately withstand axial compression. 104 The increased nociceptive nerve and mechanoreceptor numbers in the degenerate IVD also makes this a structure which is sensitive to mechanical compression and a major contributor to the generation of LBP. 265 Compared to large animal models of IVDD the husbandry and handling of mice is straight forward and these are a popular and very useful animal model amenable to genetic manipulation. 227 This is reflected in the 1000+ mouse models which have been developed in the last decade aiding in the elucidation of the multifactorial components that contribute to the complexity of IVDD. 227 It was deemed beyond the scope of this review to provide a comprehensive coverage of all of these IVD models; however, a number of publications can be consulted for further information in this area. 266 , 267 , 268 , 269 , 270 , 271 , 272 The laboratory mouse is the premier animal for investigations on IVD molecular and cellular systems. 266 , 268 Experimental genetic tools have been developed for the mouse, including unique inbred strains, a complete reference genome, deep sequencing data for inbred lines, 270 extensive genome variation maps (e.g., SNPs), and technologies for genome manipulation. 271 , 272 An international collaborative effort to generate targeted mutations in all murine protein‐coding genes was initiated in 2007. 269 The Mouse Genome Database ( http://www.informatics.jax.org ) is the primary community database for the laboratory mouse and a key source of gene biological reference data, gene functions, phenotypes, disease models relevant to human biology and disease freely accessible to all researchers. 267 Large animal models of IVDD such as the canine spontaneous and ovine mechanical de‐stabilization models of IVDD have important attributes compared to small IVDD models developed in mice, rats, and rabbits. 37 , 38 , 238 , 258 These large animal models of IVDD have a more similar size and cellular composition to that of the human IVD with notochordal cell populations disappearing in early development whereas notochordal cell populations persist in the small animal models and their influence on pathological changes developing in the degenerate IVD need to be considered, however, these are not a prominent feature in human IVDD. 273 The development of degenerate lesions in the ovine model leads to the development of pathological features in multiple IVD tissues similar to those observed in human IVD degeneration. 37 , 39 Studies using the ovine IVD model have been awarded 9 ISSLS and 3 Grammer prizes which may be considered tacit acceptance by a broad peer review of this model for relevant extrapolations to degenerative features that occur in human IVDD. Large IVDs allow multidisciplinary approaches to be used to examine multiple features of IVDD including zonal analyses of IVD composition, gene expression, biochemical examinations of focal changes in IVD structural organization, biomechanical behavior of specific IVD tissue regions and focal histopathological scoring across a broad range of structural features of functional significance over the entire IVD. 37 , 38 , 39 , 239 , 249 , 258 , 273 , 274 This allows specific questions to be asked with the large animal models of IVDD relevant to human IVDD which cannot be asked with the small animal models. However, small animal models of IVDD have other attributes relating mainly to gene manipulation effects on disc pathobiology 227 , 266 , 267 , 268 , 269 , 270 , 271 , 272 or the development of biomechanically induced whole animal models of IVDD. Mouse and rat models of IVDD have been developed where the IVD is left intact but the supportive paraspinal muscles receive a discrete surgical lesion. 241 , 242 This destabilizes the spine and induces degenerative changes in the IVD and emphasizes the inter‐relationship between the IVD and other spinal components in normal spinal loading. Small animal models of IVDD are also amenable to quantitative assessment of pain using facial recognition methodology 275 , 276 , 277 , 278 , 279 , 280 , 281 , 282 , 283 although this has also now been applied to canine and ovine studies. 284 , 285 , 286 , 287 Significant improvements in the analysis of small animal pain behavior in IVDD through the application of AI in the assessment of ultra‐high speed images documenting changes in animal gait, locomotion, kinematics, facial images for pain has significantly improved the interpretation and quantitation of data generated from small animal LBP models. 278 , 281 , 283 With these methodological improvements the door has been opened for more meaningful evaluations of therapeutic agents and treatment protocols that can potentially inhibit pain stimuli emanating from the IVDD and paradiscal spinal structures. 288 , 289 IVD cells are not subject solely to static compression but as already stated receive a dynamic balance of tension/compression which not only effects nutritive pathways 260 , 262 , 263 but also has mechanotransductive properties that regulate disc cell behavior. 224 , 259 , 265 Thus, it is essential that a physiologically relevant model is designed to mimic these dynamic in‐vivo micromechanical environmental conditions to better understand their roles in the IVD degenerative process. IVDD is a complex multifactorial process, mouse models have been invaluable in providing information on IVD pathobiology. 227 Static compression murine models of IVDD have provided information on changes in IVD composition and organization that occur with IVDD and the changes in MMP expression that occur in this process. 229 , 246 , 290 , 291 , 292 Static compression also has interesting effects on the cellular dynamics of notochordal cell populations in the murine IVD. This has shown that vital instructional cues held by this cell type aid in the establishment and maintenance of the other resident disc cell populations and possibly could be harnessed in the regeneration of the IVD. 293 Moreover, a premature decline in notochordal cell numbers in the murine IVD is a forerunner of degenerative changes in the murine IVD ECM. Compression‐induced degeneration of the murine IVD has led to the development of a finite‐element model which describes these degenerative processes. 248 The rabbit IVD also contains a prominent notochordal cell population. In a rabbit IVD explant model of unconfined uniaxial compression, static compression of 0.5 and 1 MPa and dynamic compression of 0.5 and 1 MPa were applied at a frequency of 0.1 and 1 Hz for 6 h, respectively. 294 Static compressive loads suppressed aggrecan and collagen gene expression, however, dynamic compression produced significant increases in gene expression for Type I and II collagen and aggrecan and regional differences between the AF and NP with marked changes in ECM organization evident histologically. An up‐regulation in IL‐1β and TNF‐α expression, and decreased viability of IVD cells was also evident with the most significant changes evident in statically loaded IVDs. Static and dynamic compression induced different biologic responses, static compression was catabolic, whereas dynamic loading at near physiological levels apparently induced synthetic activity and an anabolic response in the IVD. 294 The increased compressive load experienced by the AF in degenerate IVDs also leads to a down regulation in type I collagen expression. 295 Examination of isolated human and bovine NP cells seeded into 3D type I collagen matrices exposed to variable loading regimens in pressure chambers also display differing gene expression profiles depending on loading with a high hydrostatic pressure (2.5 MPa) resulting in decreased anabolic gene expression. 296 AF and NP cells subjected to static unconfined compression in an alginate culture system also displayed differential effects on gene expression. 297 AF cells responded to mechanical deformation by increased expression of types I and II collagen, aggrecan, biglycan, decorin, and lumican. NP cells were not responsive to mechanical loading with changes in gene expression of matrix proteins not observed at any time in this system. 297 Differential changes in cytoskeletal organization by AF and NP cells in response to static compression were observed with increased expression of vimentin mRNA and polymerization of vimentin subunits by AF cells but no detectable changes in NP cells. 297 These observations support the differential mechanotransductive effects on disc cell behavior by mechanical loading and the complexities of events that lead to degenerative changes in the IVD. 259 Examination of in‐vivo remodeling of IVDs in response to short‐ and long‐term dynamic compression using rat IVDs instrumented with an Ilizarov‐type device has shown that dynamic compression should be considered a “healthy” loading regimen that maintains or promotes matrix biosynthesis without substantially disrupting disc structural integrity. 298 A slow accumulation of degenerative changes similar to those observed in human IVDD occurs when dynamic compression was applied for prolonged durations. This effect was mild, however, when compared to effects induced by static compression and bending that created greater structural disruption to the IVD 3D structural organization. 298 The effect of immobilization and dynamic compression on IVD cell gene expression profiles has been examined in rat tail‐IVDs instrumented with an Ilizarov‐type device. Immobilization and dynamic compression of IVDs downregulated type I and type II collagen but upregulated aggrecanase, collagenase, and stromelysin expression in the AF but not in the NP. 299 Dynamic compressive effects on IVD mechanics and cellular responses have been examined in a bovine organ culture IVD model using caudal IVDs. 300 This study showed that remodeling of the IVD occurred in response to biomechanical loading and that this was an important regulator of IVD composition and ECM homeostasis. However, when loading regimens resulted in excessive IVD remodeling degenerative changes in IVD structural organization may detrimentally affect its function as a visco‐elastic weight bearing cushion. This study thus reinforced the dynamic inter‐relationship that exists between disc cellular behavior and mechanotransductive regulatory effects induced by static and dynamic compressive loading. 300 Understanding the complexities of neuropathic pain is an important clinical challenge; however, the molecular mechanism remains elusive. Chronic DRG compression models of neuropathic pain suggest the Wnt/ β ‐catenin pathway plays a critical role in the pathogenesis of neuropathic pain and may be an appropriate therapeutic target. 301 , 302 Proinflammatory factors such as TNF‐ α and IL‐18 are significantly elevated in neuropathic pain models. Levels of these mediators are significantly lower when the Wnt/ β ‐catenin pathway is inhibited using a Wnt/ β ‐catenin pathway inhibitor such as XAV939. XAV939 is a potent, small molecule inhibitor of tankyrase (TNKS) 1 and 2 (IC₅₀ = 11 and 4 nM, respectively). 303

This study was conceived and written by Ashish D. Diwan and James Melrose. None of the above listed companies had any input into the design, implementation or interpretation of the study.

This review has outlined the complexities of IVDD and the multiple IVD receptors and inflammatory mediators and neurotrophins that have roles in the development of nociceptor and mechanoreceptors in the degenerate IVD that produce pain signals transported to the brain by the CNS for interpretation. Inflammation in the IVD has powerful effects on the resident disc cell populations and is the impetus for the ingrowth of nerves and blood vessels into the normal IVD leading to its degeneration. Clearly, in order to prevent events that lead to pain generation, preservation of a healthy functional IVD is important, prevention of inflammation and oxidative conditions also prevents ER stress and mitochondrial dysfunction. Animal models of inflammatory and neuropathic pain indicate that inflammation regulates the resolution of pain by producing pro‐resolving mediators such as resolvin D1. 304 Resolvins are derived from, eicosapentaenonic, docosahexanoic, docosapentaenoic, and clupanodonic omega‐3 fatty acids. 305 , 306 , 307 Resolvins have cell regulatory properties similar to prostaglandins promoting the restoration of normal cellular functional properties following the inflammatory conditions that occur in tissue injury. It remains to be established how resolvins are induced in the CNS but resolvin studies nevertheless offer exciting possibilities in the development of potential methods for the alleviation of intractable neuropathic pain in chronically affected patients. 305 , 306 , 307 In the last three decades, a number of studies on bioactive peptides that are opioid receptor ligands, have also been undertaken. 308 Hemorphins are endogenous 4–10 amino acid peptides released during proteolysis of the beta subunit of hemoglobin. The hemorphins exhibit diverse therapeutic effects in both humans and animal models including regulation of blood pressure, mood regulation, enhancement in memory and cognitive learning and analgesic effects. 309 , 310 , 311 Such effects occur through the ability of these peptides to modulate a diverse range of proteins including enzymes and G‐protein coupled opioid receptors. 310 The resolvins and hemomorphins offer considerable promise as agents that can be potentially developed into therapeutic protocols for the alleviation of chronic neuropathic as well as nociceptive pain and deserve further evaluation in future studies in the improved animal models of IVDD that have been developed. A question has been raised as to whether nutritional intervention can prevent chronic pain development. 312 , 313 , 314 , 315 A number of dietary phytochemicals possess anti‐oxidant and anti‐inflammatory properties and have been shown to have tissue and disc cell protective properties in‐vitro. Flavonoids have potent anti‐oxidant and anti‐inflammatory cell and tissue protective properties. 316 , 317 , 318 Alleviation of inflammation by flavonoids provides relief from nociceptive and neuropathic pain 319 through modulation of interactions with pain receptors, 320 ion channels, 321 , 322 , 323 inhibition of inflammatory cell signaling, 324 , 325 , 326 neuroglial activation, 327 and a reduction in inflammatory cytokine levels. 328 , 329 , 330 The vagus nerve and its branches have an extensive distribution in the body and its afferent and efferent fibers have motor and sensory properties that regulate many organ systems and the gut microbiome. Following spinal cord injury, in male Wistar rats the neurochemical characteristics of vagal sensory neurons distant from the spinal cord display changes in P2X3 receptor, Substance P and isolectin B4 (1B4) expression, known neuronal injury‐responsive markers. This demonstrates communication between the vagal nerve and spinal cord. 331 The vagus nerve innervating the cervical spine has been used in spinal cord‐brain stimulation procedures to treat neurological disorders of cognitive decline (epilepsy, depression). 332 The vagus nerve provides communication between the gut microbiome and linked organ systems (brain, liver, lung, stomach) and transports regulatory gut metabolites to these organs. 268 The recently identified gut‐IVD axis 269 warrants further investigation in the context of control of discogenic LBP and repair of the degenerate IVD. If a means can be found to deliver therapeutic levels of bioactive regulatory compounds to IVDs in‐vivo then this may prevent the development of inflammatory conditions in the IVD that lead to the generation of LBP. Inflammatory conditions that occur in the gut associated with obesity suggests a low‐saturated fat, low sugar diet may decrease ER oxidative stress and Toll‐like receptor and glial cell activation in the IVD and afferent vagal nerve fiber stimulation via the stomach‐brain and gut‐brain axes. 314 Dietary phytochemicals processed by the gut microbiome release prebiotic metabolites that are therapeutic. 333 The vagal nerve is a regulatory delivery system for such metabolites in the gut‐brain, gut‐lung and gut‐liver axes in a number of diseases and may represent a new therapeutic frontier. The gut‐IVD axis has also recently been identified as a potential route of communication to the IVD. 334 This is an area that warrants future investigation as a potential means of either protecting the IVD or of delivering bioactive factors to prevent potentiation of pain signals in the IVD and associated spinal pain centers.

Many animal studies have shown that IVDD significantly contributes to the development of LBP, damage to normal IVD structure compromises its biomechanical competence and the viability of IVD cell populations. The resultant mechanical destabilization and altered cell‐ECM signaling results in the production of inflammatory cytokines and active MMPs which promotes disc degeneration. Degenerative changes in the disc ECM conducive to neovascularization and the ingrowth of nerves and an elevation in the production of inflammatory cytokines and neurotrophic factors all lead to a significant deterioration in the normal cellular microenvironment of the IVD conducive to the production of mechano‐and pain receptors. The appearance of glial fibrillary acidic protein‐immunopositive cells in diseased IVDs is closely associated with nerve ingrowth and suggests that Schwann cells have a role to play in regulating disc innervation and nerve function in the disc. 180 Neurotrophins, ion channels, inflammatory cytokines all contribute to detrimental changes in the IVD that promote the ingrowth of nerves into the previously aneural IVD. The generation of inflammatory cytokines and neurotrophins by disc cells produces an environment that stimulates nociceptive nerves generating painful responses that are signaled to the sensory DRG and sympathetic ganglia that serve the IVD and these signal to the sensory layers of the dorsal horns of the spinal cord. 175 Controlled annular incision or supraphysiological compression can induce IVD destabilization, production of MMPs and inflammatory mediators and deterioration in normal IVD organization in experimental models of IVDD. With the elevation in MMP levels in these models the IVD ECM becomes depleted of aggrecan and in‐growth of blood vessels and nerves can occur 104 similar to that observed in human IVDD 181 and increased innervation of the IVD occurs. 182 A mouse model of IVDD and LBP induced by supraphysiologal compression 114 displays increased innervation levels. 104 Overt compression in murine IVDs induces long‐lasting increases in inflammatory mediators, nerve injury and regeneration of afferent nerve fibers serving the IVD providing a pathomechanism for chronic discogenic LBP. 114 , 175 , 183 The normal IVD is poorly innervated by sensory and sympathetic perivascular nerve fibers (nociceptive) and postganglionic sympathetic nerve fibers in its outermost annular layer. Upon degeneration, depletion of the IVD aggrecan levels lowers its hydrostatic pressure and the IVD becomes susceptible to the ingrowth of nerves and blood vessels from its periphery. The CS chains of aggrecan in the normal IVD strongly inhibit nerve ingrowth into the IVD furthermore, the high hydrostatic pressure provided by aggrecan‐HA aggregates physically prevents the ingrowth of blood vessels. In‐vitro culture experiments conducted with aggrecan isolated from lumbar IVDs shows this proteoglycan significantly inhibits neuronal growth. 184 Many studies have also shown that lectican CSPGs inhibit neurite outgrowth in‐vitro using a range of neural cell populations. 185 , 186 , 187 , 188 Furthermore, in traumatized neural tissues lectican CSPGs up‐regulated in stabilizing glial scars also strongly inhibit neural outgrowth and functional neural recovery. 189 , 190 CSPGs have long been known as inhibitors of neural growth preventing functional axonal regeneration after injury. Therapeutic use of chondroitinase ABC to remove CS chains from CSPGs in the gliotic scars improves functional neuronal recovery 191 clearly establishing the inhibitory regulatory properties of the CS side chains of lectican PGs. Endogenous ADAMTS‐4 activity in the spinal cord also promotes functional neural recovery after spinal cord injury by removing lecticans from the lesion site. 192 Thus in IVDD, cell mediated changes result in an elevation in catabolic inflammatory cytokine expression, 95 , 123 , 130 , 193 , 194 , 195 elevated production of MMPs, 196 , 197 , 198 , 199 and neurotropic and angiogenic factors 28 , 119 , 125 , 200 , 201 , 202 , 203 , 204 , 205 , 206 , 207 , 208 which promote nerve and blood vessel ingrowth into the degenerate IVD. 116 , 126 , 178 , 182 , 209 , 210 , 211 , 212 While the causes of LBP remain to fully determined, peripheral, spinal, and supraspinal biochemical and electrophysiological studies in animal models indicate that peripheral pro‐inflammatory mediators and neuropeptides sensitize nociceptors in the IVD in a similar manner to how they generate pain in knee OA. 213 Degenerative changes in the IVD ECM and alterations in its biomechanical properties are also major contributors to the generation of LBP. 202 Many studies have documented increases in the number of nerve fibers in the degenerate IVD with pain influenced by nerve activation due to the inflammatory conditions and the destabilization that prevails in the degenerate disc. Altered biomechanics and production of neurotrophins (NGF, BDNF) and inflammatory mediators (iNOS, IL‐1β, TNF‐α) influence cell signaling pathways resulting in nerve activation in the degenerate IVD and perceptions of pain. 208 , 210 , 214 Major intracellular signaling pathways regulated by NF‐κB, MAPK, and Wnts mediate the molecular events responsible for the initiation and progression of IVD degeneration and perception of pain and these represent potential therapeutic targets for the treatment of LBP. 202 Expression of the pain receptors Trk‐A (Tropomyosin receptor kinase A, high affinity nerve growth factor receptor, neurotrophic tyrosine kinase receptor type 1, TRK1‐transforming tyrosine kinase protein) and Trk‐B by cells of the non‐degenerate and degenerate IVD suggests an autocrine role for neurotrophins in the regulation of disc cell pain biology. 208 The IVD has also been reported to contain Semaphorin 3A which has regulatory roles in the development and function of neural networks. 215 , 216 TRP channels are also present in the IVD and these are potential sensors and transducers of inflammatory pain. Transient receptor potential ankyrin‐1 (TRPA1) and TRP vanilloid −1, 2, 4 (TRPV1, 2, and 4) have all been detected in the IVD. 162 , 217 Knockout of TRPA1 in mice is associated with degenerative changes in the NP and CEP, however, knockout of TRPV1 did not show obvious IVD changes. Elevated cytokine levels in the degenerate IVD increases gene expression of TRPV2 and TRPV4. 162 , 217 Inhibition of TRPV4 reduces inflammation induced by hyperphysiological stretching of AF cells. 218 The reduced osmolarity in the degenerate IVD due to depletion of aggrecan increases TRPV4 expression and pro‐inflammatory cytokine production by IVD cells. 219 Activation of the TRPV1 channel by the PGE2/EP4 cell signaling pathway induces spinal hypersensitivity in a mouse model of CEP degeneration. 220 Neuropathic pain is associated with inflammation at sites of tissue damage resulting in a cascade of events concentrating and activating innate immune cells by immuno‐stimulatory cytokines, neurotrophic factors, and chemokines. 221 , 222 Neuroinflammatory environments activate glial cells in the spinal cord and brain. Astrocytes and oligodendrocytes modulate synaptic neurotransmission and potentiation of neuropathic pain. Following peripheral nociceptive activation via nerve injury, activated microglia release pro‐inflammatory TNFα, IL‐1β, and IL‐6, producing inflammatory pain and neuroinflammation. 108 , 223 This recruits other microglia and eventually activates nearby astrocytes. This prolongs the inflammatory state and leads to chronic neuropathic pain. 120 , 223 Inflammation and immune responses by mast cells, neutrophils, macrophages, T lymphocytes, microglia and astrocytes are all implicated in inflammatory processes producing neuropathic pain. IVDD results in mechanical or inflammatory effects and stimulation of AF nociceptors and mechanoreceptors causing discogenic pain. 224 A consequence of the abnormal spinal motion that occurs in biomechanically incompetent degenerate IVDs is an elevation in mechanical stimulation of IVD receptors. 225 Ingrowth of blood vessels and nerve fibers into deeper regions of the AF in IVDD and an elevation in inflammatory mediator levels also contribute to the generation of LBP. 223 These pain signals are conducted by myelinated A δ fibers and unmyelinated C fibers to the DRG and then to the somatosensory dorsal horn of the gray matter of the spinal cord. Elevation in nociceptor stimulation by IVDD may increase the sensitivity of the somatosensory system with peripheral sensitization leading to normally innocuous stimuli elliciting an amplified response. 221 IVDD can thus effect adjacent nerve roots or DRG resulting in neuropathic pain of mechanical or biochemical origin. Feed‐on effects of IVDD on other spinal structures, such as the facet joints, ligaments, and muscles can also result in them becoming centers of pain generation. Thus, IVDD may be responsible for the development of nociceptive and neuropathic pain without being the actual pain focus. Noxious stimuli in the IVD such as an acidic pH, ECM degeneration, inflammatory mediators and neurotrophins that generate inflammatory conditions in the IVD result in membrane depolarization of peripheral nociceptive nerve endings. 120 , 223 This results in the generation of an action potential in such nerves, axonal conductance of electrical signals in nociceptive neurons to somatic DRGs then to the sensory gray matter dorsal horn of the spinal cord and then to the brain occurs. Signal transduction through chemical synapses in neural networks carries pain signals from IVD nociceptors to the brain. 122 Stimulation of aberrant neuronal activity by inflammatory mediators induces protein kinases A and C, calcium/calmodulin‐dependent protein kinase, and MAPK signaling in primary sensory and dorsal horn neurons mediating the induction and maintenance of neuropathic Activation of MAPKs (p38, extracellular signal‐regulated kinase, and c‐Jun N‐terminal kinase) in spinal cord microglia or astrocytes results in the production of inflammatory mediators and sensitization of dorsal horn neurons and activation of spinal glia. Such neuron‐glia interactions enhance and prolong neuropathic pain. 226 Neuropathic pain is the most difficult type of pain to treat clinically because it is not known how nerve injuries cause chronic pain. A recent study has shown that neurons in the spinal cord which process pain signals can be attacked by activated spinal glia and the perineuronal nets (PNNs) which normally protect these neurons providing neural plasticity become degraded. 222 Aggrecan is lost from the PNNs and the hypersensitivity to heat and spontaneous pain occurs. Damage to the PNNs affects the transmittance of pain signals to the brain, heightens pain sensitivity and represents a new mechanism of chronic neuropathic pain generation. 222

The IVD is innervated by branches of the sinuvertebral nerve, by nerves derived from the ventral rami of spinal nerves or by nerves derived from gray rami communicantes. 27 In the normal IVD, innervation is restricted to the outermost lamella of the AF (Figures  1 and 2 ). These consist of small nerve fibers and some large fibers that act as mechanoreceptors. In the degenerated IVD, greater numbers of nerve fibers are present that enter the inner AF and the NP. Dorsal root ganglia (DRG) contain thin myelinated and unmyelinated fibers arising from small niciceptive neurons projecting from the IVD lamina I and II and also extend to the dorsal horn of the spinal cord. Most of the IVD sensory nerve fibers originate from small peptidergic neurons expressing the tyrosine kinase receptors (TrkA/TrkB), non‐peptidergic neurons express the common signaling receptor for glial cell‐derived neurotrophic factor family (Ret). 28 Schematic showing the innervation of the normal and degenerate intervertebral disc (IVD) (A). The normal IVD is the largest predominately aneural structure in the human body where nerves are confined to the outer annular lamellae. With IVD degeneration and depletion of space‐filling aggrecan from the IVD a significant reduction in internal hydrostatic pressure in the IVD and the production of inflammatory mediators and neurotrophic factors provides conditions that are conducive to the ingrowth of nociceptive nerves (*) from the sinu‐intervertebral nerve into the AF of the IVD and significant increases of peripheral mechanoreceptors in the IVD. Thus, there is an increased perception of pain in the mechanically incompetent degenerate IVD. Peripheral IVD neural structures such as the sympathetic ganglion and dorsal root ganglion and the rami communicantes are highly innervated structures. The DRG communicates with the sensory dorsal horn of the spinal cord. There are also connections to the dural nerve plexus (arrows, *) from the sinu‐vertebral nerve. The facet joint capsule and associated subchondral bone and vertebral bodies adjacent to the IVDs and CEPs are all innervated. The spinal nerve has dorsal and anterior connections to the spinal cord (inset). Neural organization of a lumbar spinal segment (B). AF, annulus fibrosus; ALL, anterior longitudinal ligament; DRG, dorsal root ganglion; NP, nucleus pulposus; PLL, posterior longitudinal ligament; SC, spinal cord. Figure modified from Kallewaard et al. 53 with permission Neural growth factors, neurotransmitters, and neural receptors involved in the ingrowth of nerves into the intervertebral disc (IVD) and transmission of sensory signals to the CNS. Composite schematic figure depicting a horizontally bisected IVD (A), sinuvertebral nerve and its branches to the sympathetic and sensory DRG and sensory nerves that communicate with the sensory layers of the dorsal horns of the spinal cord. 1. Nucleus pulposus, 2. Outer AF, 3. Inner AF, 4. Posterior longitudinal ligament, 5. Anterior longitudinal ligament, 6. Sympathetic root ganglion and its pre‐ and post‐ganglionic white and gray ramus communicantes, 7. Postganglionic nerve fibers, 8. DRG and its nerve fibers, pain receptors, neurotrophins and neurotransmitter associated peptides, 9. Sinuvertebral nerve (B) that communicate with the sensory layers (10) (C) of the spinal cord dorsal horn (11) (D). (i) TrkA/TrkB (Tropomyosin receptor kinase A, B; receptor for nerve growth factor [NGF]/brain‐derived neurotrophic factor, BDNF). (ii) Degenerin/epithelial sodium channels (DEG/ENaCs) (ENaCa, b, and c). (iii) Acid‐sensing ion channels (ASIC1, ASIC2, and ASIC3). (iv) TRP family (TRPA1, TRPC1, TRPC6, and TRPV1‐4). (v) Neuron associated neurotransmitter peptides CGRP (calcitonin gene‐related peptide), GFRα1 and GFRα3 (glial cell‐line‐derived neurotrophic receptor subtypes α1 and α3), P2X3 (ATP‐gated ion channel subtype P2X3), SP (substance P), TMP (thiamine monophosphatase), VR1 (vanilloid receptor subtype 1). BDNF, brain derived nerve factor; NGF, nerve growth factor. The horizontally sectioned spinal cord (d) is stained with Nissl stain to show the white and gray matter and DRG. Figure modified from Takahashi et al. 28 with permission Neurons in DRGs can be differentiated on the basis of the pattern of receptors for neurotrophic factors they interact with, including ion channels such as degenerin/epithelial sodium channels (DEG/ENaCs) (ENaCa, b and c); acid‐sensing ion channels (ASIC 1–3) and transient receptor potential (TRP) families (TRPA1, TRPC1, TRPC6 and TRPV1‐4). Expression of CGRP (calcitonin gene‐related peptide); GFRa1 and GFRa3 (glial cell‐line‐derived neurotrophic receptor subtypes α1 and α3); P2X3 (ATP‐gated ion channel subtype P2X3); SP (substance P); TMP (thiamine monophosphatase); VR1 (vanilloid receptor subtype 1) also differentiates DRG neurons. The Mu receptor interacts with opioid peptides to promote analgesic anti‐inflammatory actions to protect cartilage from posttraumatic degeneration and have also been evaluated for the treatment of chronic LBP. 29 , 30 , 31 , 32

The aim of this review was to evaluate data generated in the last decade by animal models of intervertebral disc (IVD) degeneration (IVDD) relevant to the interpretation of how low back pain (LBP) is generated and the identification of potential therapeutic targets to treat this condition. The World Health Organization (WHO) has defined LBP as “pain and discomfort below the costal margin above the inferior gluteal folds, with or without referred leg pain.” This may be experienced as aching, burning, stabbing, sharp or dull, well‐defined, or vague pain of mild to severe intensity. 1 Many recent studies show that IVDD is a major contributor to LBP due to neural ingrowth into the degenerate IVD and a significant increase in mechano‐and nociceptive receptor numbers in the degenerate IVD which produce pain responses due to abnormal loading in the incompetent degenerate IVD. Several other anatomical structures in the spine besides the IVD can also generate pain responses (Table  1 ). These include, the vertebral body, paradiscal and myotendinous tissues and the osteoarthritic facet joint capsule and articular cartilage. The focus of this review is the IVD since it is a major contributor to the weight bearing and flexibility properties of the spine and when the IVD degenerates a significant contributor to the generation of LBP. Multiple spinal centers of low back pain generation

The authors declare no conflicts of interest.

LBP is a common disorder 3 that can be elicited by painful stimuli emanating from the spinal muscles, nerves, vertebral body, and para‐discal tissues such as the ALL, PLL, facet joint cartilage and associated synovial capsular tissues. 20 LBP can vary in intensity from a dull constant ache to a sudden sharp pain 21 and is classified by its duration time as acute (pain duration 12 weeks). 22 , 23 , 24 LBP may be further classified by its underlying cause as mechanical, non‐mechanical, or referred pain. 22 , 25 Neuropathic pain is caused by inflammation, irritation or excessive compression of neural tissue, whereas nociceptive pain is the body's reaction to painful stimuli such as a damaged back muscle but is not responsible for nerve damage in itself. 22 After an accident or traumatic damage to the PNS/CNS, nerves may become weakened or dysfunctional, causing hypersensitivity to pain and even when the wound has healed, the nerves may continue to give false signals of pain (neuropathic pain). About 40% of the worlds human population suffer from LBP some time in their lifetime 26 and this may be as high as a value frequently quoted for Western societies of 80%. 4 It is conservatively estimated that 9%–12% of the global general population (632 million) have LBP at any one time. It is conservatively estimated that 9%–12% of the global general population (632 million) have LBP at any one time. Of the numerous causes of back pain, a landmark study estimated that in the 632 million, a meta‐analysis reveals 403 Million people (5.5% of world population) have symptomatic disc degeneration. 9 Hence, it is an imperative to understand the mechanisms underpinning pain related to disc degeneration.