- Email: [email protected]
An immunohistochemical study of mechanoreceptors in lumbar spine intervertebral discs
Journal of Clinical Neuroscience 17 (2010) 742–745
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Neuroanatomical Study
An immunohistochemical study of mechanoreceptors in lumbar spine intervertebral discs A. Dimitroulias a,*, C. Tsonidis a, K. Natsis b, I. Venizelos c, S.N. Djau d, P. Tsitsopoulos a, P. Tsitsopoulos a a
B’ Neurosurgical Department, Medical School, Aristotle University, Thessaloniki, Greece Department of Anatomy, Medical School, Aristotle University, Thessaloniki, Greece c Department of Pathology, Ippokration General Hospital, Thessaloniki, Greece d Department of Forensic Medicine, Medical School, Aristotle University, Thessaloniki, Greece b
a r t i c l e
i n f o
Article history: Received 17 June 2009 Accepted 13 September 2009
Keywords: Immunohistochemistry Intervertebral disc Lumbar spine Mechanoreceptors S-100 protein
a b s t r a c t There are limited data concerning mechanoreceptors in normal human lumbar intervertebral discs. The aim of our study was to determine the types of mechanoreceptors in the two lower intervertebral discs in normal adult cadaveric donors and to review the literature. Twenty-five lumbar (L4–5 and L5–S1) intervertebral discs were retrieved from 15 fresh cadavers. We utilized immunoreactivity against the S-100 protein to localize specialized nerve endings. Immunoreactivity showed receptors in 92% of discs. The most frequent type had morphology resembling the Ruffini type receptor (88%), followed by the Golgi type. Free nerve fibers were frequently present. All neural structures were found in the superficial layers of the annulus fibrosus, in longitudinal ligaments, or between these two. The anterior part of the L5–S1 disc had a greater frequency of encapsulated receptors than the other parts (p = 0.022), which may be correlated with the high shear forces to which the lumbosacral junction is subjected. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The presence of nerve structures in intervertebral discs is well documented, and the first histologic description of nerve fibers in intervertebral discs was by Jung and Brunschwig.1 These receptors have a key role in the perception of joint position and adjustment of the muscle tone of the vertebral column. An important component of low back pain is an intense muscle spasm of the vertebral musculature, elicited through reflex arches mediated by specialized nerve endings. Freeman and Wyke2 have classified these joint receptors into four categories: (i) type I: encapsulated mechanoreceptors similar to Ruffini endings; (ii) type II: encapsulated mechanoreceptors similar to Pacinian corpuscles; (iii) type III: encapsulated mechanoreceptors similar to Golgi endings; and (iv) type IV: unmyelinated free nerve endings and unencapsulated plexuses that have nociceptive function. The presence of mechanoreceptors in the human intervertebral disc has been studied.3–7 Older methods have utilized impregnation techniques with silver-based or gold-based stains, which do not stain neural tissue specifically. Immunohistochemical techniques have also been used to localize nervous system-specific substances such as: protein gene product 9.5 (PGP 9.5);6 vasoactive
* Corresponding author. Present address: 344 West Water Street, Apartment E, Lock Haven, PA 17745, USA. Tel.: +1 570 8935330; fax: +1 570 8935172. E-mail address: [email protected] (A. Dimitroulias). 0967-5868/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2009.09.032
intestinal polypeptide (VIP);6 neuropeptide Y;5,6 synaptophysin (SYN);5 calcitonin gene-related peptide (CGRP);6,8,9 substance P (SP);5,6,8 tyrosine hydroxylase;8 and neurofilament protein (NFP).10 We investigated the presence of mechanoreceptors in normal human intervertebral discs of the lower lumbar spine using S-100 protein immunoreactivity and reviewed previous studies on nerve structures in intervertebral discs. S-100 is a calcium-binding protein and is highly specific for brain glial and ependymal cells, Langerhans cells, melanocytes, chondrocytes, adipocytes and Schwann cells.11 Details of the use of the S-100 antibody have been published previously.12,13
2. Materials and methods Twenty-five intervertebral discs were removed during routine autopsy from 15 human cadavers. Only the L4–L5 (13 discs) and/ or L5–S1 (12 discs) were harvested. Eight cadavers were male, seven female and the mean age was 45.4 years (range, 15–66 years). None had any history of chronic low back pain or an operation on the vertebral column. The autopsy specimens were removed as soon as possible and always within 6 hours after death. All the specimens were normal upon visual inspection. The anterior and posterior halves of the discs (longitudinal ligament, annulus fibrosus, nucleus pulposus) were dissected at the midsubstance between the endplates and studied separately. The tissues were fixed for 24 hours in 10% formaldehyde buffered at pH 7.4, then
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
embedded in paraffin and cut in serial sections of 4 lm thickness. Endogenous peroxidase activity was blocked using methanol containing 0.3% hydrogen peroxide for 10 minutes. Non-specific binding sites were blocked using 1.5% normal goat serum (in phosphate-buffered saline, pH 7.4) for 20 minutes in a humidity chamber at room temperature. The sections were then incubated for 30 minutes at room temperature with pre-diluted (1:200) primary antibody against S-100 protein raised in rabbit and commercially supplied by Dako (Glostrup, Denmark). Biotinylated secondary goat anti-rabbit antibody was then added. The antigen–antibody reaction site was visualized using avidin–biotin– peroxidase complex (ABC) staining and diaminobenzidine was used as the chromogen. Finally, the slides were counterstained with Harris hematoxylin to show the general morphology of the tissue. The sections were examined under a light microscope. Experimental controls (positive and negative) were performed and supported the specificity of staining. During examination with the light microscope there were no signs of autolysis and the orientation of collagen fibers could be clearly observed. The degree of disc degeneration (edge neovascularization, myxomatous degeneration, fissures and cysts in annulus, granular changes and chondrocyte cloning in the nucleus pulposus) was evaluated with a semi-quantitative method on routine hematoxylin and eosin sections of the specimens.14,15 We used the morphologic criteria of Freeman and Wyke to distinguish the different receptors.2 Further criteria were that: each receptor had consistent morphology on serial sections; and the parent axon was identified to confirm that the structure was a nerve ending.16 Finally, a quantitative analysis was performed to determine the density of the mechanoreceptors in each section of the disc. Fisher’s exact test for comparison of proportions was used to analyse the correlation between mechanoreceptor type, disc level and age in the anterior and posterior part of the disc tissue, and the Kruskall–Wallis test was used for an age comparison among mechanoreceptor groups. Analyses were conducted using the Statistical Package for the Social Sciences software version 12 (SPSS, Chicago, IL, USA). 3. Results 3.1. Occurrence of receptors The frequency of the specific receptor types is shown in Table 1. Nerve structures were not found in only 2 out of 25 discs. In discs possessing Ruffini receptors, the mean density was 1.4 receptors per disc section. The corresponding density for Golgi receptors and free nerve fibers was 1.2 and 1.3 per specimen, respectively. We did not identify any Pacinian receptors in our specimens. Table 2 shows the association between mechanoreceptor type, disc level, and age in the anterior and posterior parts of the discs. Descriptive statistics for discrete variables are presented as frequencies and percentages. Continuous variables are summarised using medians with ranges. All reported p-values are two-tailed.
Table 1 Frequency of specific receptor types in the anterior and posterior parts of L4–L5 and L5–S1 intervertebral discs Receptor
Ant. L4–L5
Post. L4–L5
Ant. L5–S1
Post. L5–S1
Ruffini Golgi Pacini Nerve fibers None
7/13 3/13 0 6/13 5/13
8/13 1/13 0 4/13 5/13
11/12 6/12 0 5/12 0
9/12 2/12 0 6/12 2/12
Ant. = anterior, L = lumbar, P = posterior, S = sacral.
743
In the anterior part of disc tissues, there was significant association (p = 0.022) between the level and the type of mechanoreceptor present. Specifically, discs with Ruffini and/or Golgi-type receptors appeared more frequently in the L5–S1 level (78%), whereas all anterior disc specimens without any neural structure appeared in the L4–L5 level (100%). We did not find any statistically significant difference between the receptors (type, frequency) and the age, sex and degree of disc degeneration of the subjects. 3.2. Morphology of receptors The receptors most frequently encountered demonstrated morphology similar to the Ruffini receptor. They usually had a thin capsule, although this was not a consistent finding, and a dense arborization of the afferent nerve fiber (Fig. 1). They were usually found in clusters,2–5 with round or ovoid shape and greatest diameter <100 lm, most often close to vessels within loose connective tissue (Fig. 2). The second most common corpuscular receptors found were morphologically similar to the Golgi tendon organ. They were spindle-shaped, usually with a thin capsule and close to vessels (Fig. 3). Their greatest diameter was always >150 lm. No Pacinian corpuscle could be detected. Free nerve fibers were a frequent finding. Some of them were found in paravascular position (Fig. 4), while others were remote from vessels. The location of the neural elements was: on the longitudinal ligaments; in the space between the longitudinal ligaments and the annulus fibrosus; or in the superficial lamellae of the annulus fibrosus, not deeper than the outer third of the annulus fibrosus. 4. Discussion We utilized an immunohistochemical technique with S-100 protein, which stains Schwann cells specifically, to demonstrate mechanoreceptors in the anterior and posterior parts of the two lower intervertebral discs obtained during autopsy. Encapsulated endings (Golgi and Ruffini) consist of highly branched non-myelinated endings of myelinated afferents, which invade and ramify among bundles of collagen fibers. The afferent fiber loses its myelin sheath after piercing the capsule but remains invested with Schwann cells. The ramifications of the afferent fiber are also invested with Schwann cells.17 Furthermore, free nerve endings, although they are called ‘‘free”, are always invested with Schwann cells and do not contact the extracellular fluid directly.18 We have found this technique very effective as a structural marker of neural structures in the intervertebral discs. As already mentioned, we could not detect any neural structures in 2 out of 25 discs. We are not aware of any previous studies that have utilized the immunoreactivity to S-100 protein to identify mechanoreceptors in normal human lumbar intervertebral discs. The resolution of silver stains is limited due to artifactual staining, especially in an environment rich in elastin and collagen. Moreover, methods to identify neuropeptide nerves in intervertebral discs may be less specific due to the tight collagenous nature of the annulus fibrosus, which prevents deeper penetration of fixative fluid.3 Roberts et al. found that the most common mechanoreceptor in the anterior and lateral parts of human disc tissue removed during anterior fusions and scoliosis surgery was the Golgi tendon organ, and the next most common was the Ruffini mechanoreceptor.6 Palmgren et al., using nine lumbar discs from five cadavers, described the immunoreactivity to the general neuronal marker PGP 9.5 and neuropeptides SP, C flanking peptide of neuropeptides Y and SYN without describing any specialized nerve endings.6 Kallakuri et al. also demonstrated a substantial number of nerve fibers in the longitudinal ligaments of New Zealand rabbits using
744
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
Table 2 Associations between mechanoreceptors, disc level and age, in the anterior and posterior parts of the disc Anterior No. (%) Nerve fibers Level L4–L5 L5–S1 Median age (yrs) Range age (yrs) a b
p Ruffini and/or Golgi
No receptors
Posterior No. (%)
p
Nerve fibers
Ruffini and/or Golgi
No receptor
4 (40.0) 6 (60.0) 23.5 15–66
4 (50.0) 4 (50.0) 56.5 15–63
5 (71.4) 2 (28.6) 59 21–66
0.022a, 6 (54.5) 5 (45.5) 38 15–66
2 (22.2) 7 (77.8) 54 20–66
5 (100.0) 0 (0) 52 15–63
0.745b
0.529a
0.100b
Fisher’s exact test. Kruskal–Wallis test. Significant association between the level and the type of mechanoreceptor.
Fig. 1. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing a Ruffini-type mechanoreceptor (black arrow) in loose connective tissue, near blood vessels (white arrows), with a thin capsule and dense arborization of the afferent nerve fiber (100).
Fig. 2. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing a cluster of Ruffini-type mechanoreceptors (black arrows) near a blood vessel (white arrow) in the posterior longitudinal ligament (100).
immunoreactivity to PGP 9.5, SP, tyrosine hydroxylase, CGRP, and activity of nicotinamide adenine dinucleotide phosphate, but they did not describe any encapsulated nerve terminals.8 Supplementary Table 2 summarizes studies of nerve structures within the intervertebral disc.
Fig. 3. Immunohistochemical staining of lumbar intervertebral disc against the S-100 protein showing a Golgi-type mechanoreceptor – fusiform with a thick afferent nerve fiber (black arrow) (100).
Fig. 4. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing paravascular (white arrow) free nerve fibers (black arrows) (100).
In the present study the most common encapsulated receptor encountered was of the Ruffini type (22 out of 25 discs). We could not find any reports documenting such a high frequency of Ruffinitype receptors in normal human lumbar spine discs. These receptors help maintain muscle tone (low threshold, slow adaptation).19 The next most common encapsulated receptor was morphologically
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
similar to Golgi endings (12 out of 25 discs). These are activated at extremes of joint motion (high threshold, slow adaptation).19 We did not detect any Pacini-type corpuscles. As these are low threshold, rapidly adapting receptors, their absence from the intervertebral disc could be attributed to the relatively slow acceleration/deceleration deformations of that part of the vertebral column. In 16 out of 25 discs, free nerve fibers were found. It is assumed that those close to blood vessels have a vasomotor role while those away from vessels may have a nociceptive (small caliber) or a proprioceptive role (large caliber).4,20 The anterior part of the L5–S1 disc had a significantly greater frequency of encapsulated receptors than the other parts, which may be correlated with the high shear forces to which the lumbosacral junction is subjected.21 Moreover, the differences in the distribution of receptors between anterior and posterior parts of the L5–S1 disc may be associated with the regional differences in the tensile properties of the disc with the anterior regions having a larger tensile modulus than the posterior ones.22 All neural structures detected were located either in the longitudinal ligaments, or the superficial lamellae of the annulus fibrosus, which are the areas sustaining the greatest pressure or tension during extremes of movement.22 In a mechanic model, during axial loading of a motion segment, compressive stresses in the nucleus (devoid of neural and vascular structures) will generate tensile stresses in the peripheral annulus, which is rich in neural receptors. In conclusion, this study confirms the existence of an abundant network of encapsulated and non-encapsulated receptors in the intervertebral discs of the lower lumbar spine in normal human subjects. The principal role of encapsulated structures is assumed to be the continuous monitoring of position, velocity and acceleration (kinesthesia). Free nerve fibers are likely to be involved in nociception or regulation of vessel tone (autonomic fibers). Acknowledgements We wish to thank Mr Trifon Totlis, Medical Student of the Aristotle University of Thessaloniki, for his technical assistance. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jocn.2009.09.032.
745
References 1. Jung A, Brunschwig A. Recherches histologiques sur l’innervation des articulations des corps vertebraux [Histological study of the innervation of vertebral body joints]. Presse Med 1932;40:316–7. 2. Freeman M, Wyke B. The innervation of the knee joint. An anatomical and histological study in the cat. J Anat 1967;101:505–32. 3. Jackson HC, Winkelmann RK, Bickel WH. Nerve endings in lumbar spinal column. J Bone Joint Surg (Am) 1966;48:1272–81. 4. Malinsky J. The ontogenetic development of nerve terminations in the intervertebral disc of man. Acta Anat 1959;38:96–113. 5. Palmgren T, Gronblad M, Virri J, et al. An immunohistochemical study of nerve structures in the annulus fibrosus of human normal intervertebral discs. Spine 1999;20:2075–9. 6. Roberts S, Eisenstein SM, Menage J, et al. Mechanoreceptors in intervertebral discs. Morphology, distribution and neuropeptides. Spine 1995;20:2645–51. 7. Yoshizawa H, O’Brien J, Smith WT, et al. The neuropathology of intervertebral discs removed for low back pain. J Pathol 1980;132:95–104. 8. Kallakuri S, Cavanaugh JM, Blagoev DC. An immunohistochemical study of innervation of lumbar spinal dura and longitudinal ligaments. Spine 1998;23:403–11. 9. McCarthy PW, Carruthers B, Martin D, et al. Immunohistochemical demonstration of sensory nerve fibers and endings in lumbar intervertebral discs of the rat. Spine 1991;16:653–5. 10. Rhalmi S, Yahia L, Newman N, et al. Immunohistochemical study of nerves in lumbar spine ligaments. Spine 1993;18:264–7. 11. Isobe T, Ishioka N, Kocha T, et al. Chemical structure and molecular evolution of S-100 proteins. Prot Biol Fluids 1983;30:21–4. 12. Pionchon H, Tommasi M, Pialat J, et al. Study of innervation of the spinal ligaments at the lumbar level. Bull Assoc Anat (Nancy) 1986;70:63–7. 13. Yahia L, Rhalmi S, Newman N, et al. Sensory innervation of human thoracolumbar fascia. An immunohistochemical study. Acta Orthop Scand 1992;63:195–7. 14. Weidner N, Rice D. Intervertebral disc material: criteria for determining probable prolapse. Hum Pathol 1988;19:406–10. 15. Yasuma T, Koh S, Okamura T, et al. Histological changes in aging lumbar intervertebral discs. J. Bone Joint Surg 1990;72-A:220–9. 16. McLain RF, Pickar JG. Mechanoreceptor endings in human thoracic and lumbar facet joints. Spine 1998;23:168–73. 17. Williams PL, Warwick R, Dyson M, et al. Gray’s anatomy. 37th ed. New York: Churchill Livingstone; 1989. p. 906. 18. Barr ML, Kiernan JA. The human nervous system. 5th ed. Philadelphia: Lippincott; 1988. p. 33. 19. Zimny ML. Mechanoreceptors in articular tissues. Am J Anat 1988;182:16–32. 20. Bogduk N, Tynan W, Wilson AS. The nerve supply of the human lumbar intervertebral discs. J Anat 1981;132:39–56. 21. Skaggs DL, Weidenbaum M, Iatridis JC, et al. Regional variation in tensile properties and biochemical composition of the human lumbar anulus fibrosus. Spine 1994;19:1310–9. 22. Buckwalter JA, Mow VC, Boden SD, et al. Intervertebral disk structure, composition and mechanical function. In: Buckwalter JA, Einhorn TA, Simon SR, editors. Orthopaedic basic science: biology and biomechanics of the musculoskeletal system. 2nd ed. Rosemont, Illinois, USA: American Academy of Orthopaedic Surgeons; 2000. p. 548–56.
Contents lists available at ScienceDirect
Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn
Neuroanatomical Study
An immunohistochemical study of mechanoreceptors in lumbar spine intervertebral discs A. Dimitroulias a,*, C. Tsonidis a, K. Natsis b, I. Venizelos c, S.N. Djau d, P. Tsitsopoulos a, P. Tsitsopoulos a a
B’ Neurosurgical Department, Medical School, Aristotle University, Thessaloniki, Greece Department of Anatomy, Medical School, Aristotle University, Thessaloniki, Greece c Department of Pathology, Ippokration General Hospital, Thessaloniki, Greece d Department of Forensic Medicine, Medical School, Aristotle University, Thessaloniki, Greece b
a r t i c l e
i n f o
Article history: Received 17 June 2009 Accepted 13 September 2009
Keywords: Immunohistochemistry Intervertebral disc Lumbar spine Mechanoreceptors S-100 protein
a b s t r a c t There are limited data concerning mechanoreceptors in normal human lumbar intervertebral discs. The aim of our study was to determine the types of mechanoreceptors in the two lower intervertebral discs in normal adult cadaveric donors and to review the literature. Twenty-five lumbar (L4–5 and L5–S1) intervertebral discs were retrieved from 15 fresh cadavers. We utilized immunoreactivity against the S-100 protein to localize specialized nerve endings. Immunoreactivity showed receptors in 92% of discs. The most frequent type had morphology resembling the Ruffini type receptor (88%), followed by the Golgi type. Free nerve fibers were frequently present. All neural structures were found in the superficial layers of the annulus fibrosus, in longitudinal ligaments, or between these two. The anterior part of the L5–S1 disc had a greater frequency of encapsulated receptors than the other parts (p = 0.022), which may be correlated with the high shear forces to which the lumbosacral junction is subjected. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The presence of nerve structures in intervertebral discs is well documented, and the first histologic description of nerve fibers in intervertebral discs was by Jung and Brunschwig.1 These receptors have a key role in the perception of joint position and adjustment of the muscle tone of the vertebral column. An important component of low back pain is an intense muscle spasm of the vertebral musculature, elicited through reflex arches mediated by specialized nerve endings. Freeman and Wyke2 have classified these joint receptors into four categories: (i) type I: encapsulated mechanoreceptors similar to Ruffini endings; (ii) type II: encapsulated mechanoreceptors similar to Pacinian corpuscles; (iii) type III: encapsulated mechanoreceptors similar to Golgi endings; and (iv) type IV: unmyelinated free nerve endings and unencapsulated plexuses that have nociceptive function. The presence of mechanoreceptors in the human intervertebral disc has been studied.3–7 Older methods have utilized impregnation techniques with silver-based or gold-based stains, which do not stain neural tissue specifically. Immunohistochemical techniques have also been used to localize nervous system-specific substances such as: protein gene product 9.5 (PGP 9.5);6 vasoactive
* Corresponding author. Present address: 344 West Water Street, Apartment E, Lock Haven, PA 17745, USA. Tel.: +1 570 8935330; fax: +1 570 8935172. E-mail address: [email protected] (A. Dimitroulias). 0967-5868/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jocn.2009.09.032
intestinal polypeptide (VIP);6 neuropeptide Y;5,6 synaptophysin (SYN);5 calcitonin gene-related peptide (CGRP);6,8,9 substance P (SP);5,6,8 tyrosine hydroxylase;8 and neurofilament protein (NFP).10 We investigated the presence of mechanoreceptors in normal human intervertebral discs of the lower lumbar spine using S-100 protein immunoreactivity and reviewed previous studies on nerve structures in intervertebral discs. S-100 is a calcium-binding protein and is highly specific for brain glial and ependymal cells, Langerhans cells, melanocytes, chondrocytes, adipocytes and Schwann cells.11 Details of the use of the S-100 antibody have been published previously.12,13
2. Materials and methods Twenty-five intervertebral discs were removed during routine autopsy from 15 human cadavers. Only the L4–L5 (13 discs) and/ or L5–S1 (12 discs) were harvested. Eight cadavers were male, seven female and the mean age was 45.4 years (range, 15–66 years). None had any history of chronic low back pain or an operation on the vertebral column. The autopsy specimens were removed as soon as possible and always within 6 hours after death. All the specimens were normal upon visual inspection. The anterior and posterior halves of the discs (longitudinal ligament, annulus fibrosus, nucleus pulposus) were dissected at the midsubstance between the endplates and studied separately. The tissues were fixed for 24 hours in 10% formaldehyde buffered at pH 7.4, then
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
embedded in paraffin and cut in serial sections of 4 lm thickness. Endogenous peroxidase activity was blocked using methanol containing 0.3% hydrogen peroxide for 10 minutes. Non-specific binding sites were blocked using 1.5% normal goat serum (in phosphate-buffered saline, pH 7.4) for 20 minutes in a humidity chamber at room temperature. The sections were then incubated for 30 minutes at room temperature with pre-diluted (1:200) primary antibody against S-100 protein raised in rabbit and commercially supplied by Dako (Glostrup, Denmark). Biotinylated secondary goat anti-rabbit antibody was then added. The antigen–antibody reaction site was visualized using avidin–biotin– peroxidase complex (ABC) staining and diaminobenzidine was used as the chromogen. Finally, the slides were counterstained with Harris hematoxylin to show the general morphology of the tissue. The sections were examined under a light microscope. Experimental controls (positive and negative) were performed and supported the specificity of staining. During examination with the light microscope there were no signs of autolysis and the orientation of collagen fibers could be clearly observed. The degree of disc degeneration (edge neovascularization, myxomatous degeneration, fissures and cysts in annulus, granular changes and chondrocyte cloning in the nucleus pulposus) was evaluated with a semi-quantitative method on routine hematoxylin and eosin sections of the specimens.14,15 We used the morphologic criteria of Freeman and Wyke to distinguish the different receptors.2 Further criteria were that: each receptor had consistent morphology on serial sections; and the parent axon was identified to confirm that the structure was a nerve ending.16 Finally, a quantitative analysis was performed to determine the density of the mechanoreceptors in each section of the disc. Fisher’s exact test for comparison of proportions was used to analyse the correlation between mechanoreceptor type, disc level and age in the anterior and posterior part of the disc tissue, and the Kruskall–Wallis test was used for an age comparison among mechanoreceptor groups. Analyses were conducted using the Statistical Package for the Social Sciences software version 12 (SPSS, Chicago, IL, USA). 3. Results 3.1. Occurrence of receptors The frequency of the specific receptor types is shown in Table 1. Nerve structures were not found in only 2 out of 25 discs. In discs possessing Ruffini receptors, the mean density was 1.4 receptors per disc section. The corresponding density for Golgi receptors and free nerve fibers was 1.2 and 1.3 per specimen, respectively. We did not identify any Pacinian receptors in our specimens. Table 2 shows the association between mechanoreceptor type, disc level, and age in the anterior and posterior parts of the discs. Descriptive statistics for discrete variables are presented as frequencies and percentages. Continuous variables are summarised using medians with ranges. All reported p-values are two-tailed.
Table 1 Frequency of specific receptor types in the anterior and posterior parts of L4–L5 and L5–S1 intervertebral discs Receptor
Ant. L4–L5
Post. L4–L5
Ant. L5–S1
Post. L5–S1
Ruffini Golgi Pacini Nerve fibers None
7/13 3/13 0 6/13 5/13
8/13 1/13 0 4/13 5/13
11/12 6/12 0 5/12 0
9/12 2/12 0 6/12 2/12
Ant. = anterior, L = lumbar, P = posterior, S = sacral.
743
In the anterior part of disc tissues, there was significant association (p = 0.022) between the level and the type of mechanoreceptor present. Specifically, discs with Ruffini and/or Golgi-type receptors appeared more frequently in the L5–S1 level (78%), whereas all anterior disc specimens without any neural structure appeared in the L4–L5 level (100%). We did not find any statistically significant difference between the receptors (type, frequency) and the age, sex and degree of disc degeneration of the subjects. 3.2. Morphology of receptors The receptors most frequently encountered demonstrated morphology similar to the Ruffini receptor. They usually had a thin capsule, although this was not a consistent finding, and a dense arborization of the afferent nerve fiber (Fig. 1). They were usually found in clusters,2–5 with round or ovoid shape and greatest diameter <100 lm, most often close to vessels within loose connective tissue (Fig. 2). The second most common corpuscular receptors found were morphologically similar to the Golgi tendon organ. They were spindle-shaped, usually with a thin capsule and close to vessels (Fig. 3). Their greatest diameter was always >150 lm. No Pacinian corpuscle could be detected. Free nerve fibers were a frequent finding. Some of them were found in paravascular position (Fig. 4), while others were remote from vessels. The location of the neural elements was: on the longitudinal ligaments; in the space between the longitudinal ligaments and the annulus fibrosus; or in the superficial lamellae of the annulus fibrosus, not deeper than the outer third of the annulus fibrosus. 4. Discussion We utilized an immunohistochemical technique with S-100 protein, which stains Schwann cells specifically, to demonstrate mechanoreceptors in the anterior and posterior parts of the two lower intervertebral discs obtained during autopsy. Encapsulated endings (Golgi and Ruffini) consist of highly branched non-myelinated endings of myelinated afferents, which invade and ramify among bundles of collagen fibers. The afferent fiber loses its myelin sheath after piercing the capsule but remains invested with Schwann cells. The ramifications of the afferent fiber are also invested with Schwann cells.17 Furthermore, free nerve endings, although they are called ‘‘free”, are always invested with Schwann cells and do not contact the extracellular fluid directly.18 We have found this technique very effective as a structural marker of neural structures in the intervertebral discs. As already mentioned, we could not detect any neural structures in 2 out of 25 discs. We are not aware of any previous studies that have utilized the immunoreactivity to S-100 protein to identify mechanoreceptors in normal human lumbar intervertebral discs. The resolution of silver stains is limited due to artifactual staining, especially in an environment rich in elastin and collagen. Moreover, methods to identify neuropeptide nerves in intervertebral discs may be less specific due to the tight collagenous nature of the annulus fibrosus, which prevents deeper penetration of fixative fluid.3 Roberts et al. found that the most common mechanoreceptor in the anterior and lateral parts of human disc tissue removed during anterior fusions and scoliosis surgery was the Golgi tendon organ, and the next most common was the Ruffini mechanoreceptor.6 Palmgren et al., using nine lumbar discs from five cadavers, described the immunoreactivity to the general neuronal marker PGP 9.5 and neuropeptides SP, C flanking peptide of neuropeptides Y and SYN without describing any specialized nerve endings.6 Kallakuri et al. also demonstrated a substantial number of nerve fibers in the longitudinal ligaments of New Zealand rabbits using
744
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
Table 2 Associations between mechanoreceptors, disc level and age, in the anterior and posterior parts of the disc Anterior No. (%) Nerve fibers Level L4–L5 L5–S1 Median age (yrs) Range age (yrs) a b
p Ruffini and/or Golgi
No receptors
Posterior No. (%)
p
Nerve fibers
Ruffini and/or Golgi
No receptor
4 (40.0) 6 (60.0) 23.5 15–66
4 (50.0) 4 (50.0) 56.5 15–63
5 (71.4) 2 (28.6) 59 21–66
0.022a, 6 (54.5) 5 (45.5) 38 15–66
2 (22.2) 7 (77.8) 54 20–66
5 (100.0) 0 (0) 52 15–63
0.745b
0.529a
0.100b
Fisher’s exact test. Kruskal–Wallis test. Significant association between the level and the type of mechanoreceptor.
Fig. 1. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing a Ruffini-type mechanoreceptor (black arrow) in loose connective tissue, near blood vessels (white arrows), with a thin capsule and dense arborization of the afferent nerve fiber (100).
Fig. 2. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing a cluster of Ruffini-type mechanoreceptors (black arrows) near a blood vessel (white arrow) in the posterior longitudinal ligament (100).
immunoreactivity to PGP 9.5, SP, tyrosine hydroxylase, CGRP, and activity of nicotinamide adenine dinucleotide phosphate, but they did not describe any encapsulated nerve terminals.8 Supplementary Table 2 summarizes studies of nerve structures within the intervertebral disc.
Fig. 3. Immunohistochemical staining of lumbar intervertebral disc against the S-100 protein showing a Golgi-type mechanoreceptor – fusiform with a thick afferent nerve fiber (black arrow) (100).
Fig. 4. Immunohistochemical staining of lumbar intervertebral disc against the S100 protein showing paravascular (white arrow) free nerve fibers (black arrows) (100).
In the present study the most common encapsulated receptor encountered was of the Ruffini type (22 out of 25 discs). We could not find any reports documenting such a high frequency of Ruffinitype receptors in normal human lumbar spine discs. These receptors help maintain muscle tone (low threshold, slow adaptation).19 The next most common encapsulated receptor was morphologically
A. Dimitroulias et al. / Journal of Clinical Neuroscience 17 (2010) 742–745
similar to Golgi endings (12 out of 25 discs). These are activated at extremes of joint motion (high threshold, slow adaptation).19 We did not detect any Pacini-type corpuscles. As these are low threshold, rapidly adapting receptors, their absence from the intervertebral disc could be attributed to the relatively slow acceleration/deceleration deformations of that part of the vertebral column. In 16 out of 25 discs, free nerve fibers were found. It is assumed that those close to blood vessels have a vasomotor role while those away from vessels may have a nociceptive (small caliber) or a proprioceptive role (large caliber).4,20 The anterior part of the L5–S1 disc had a significantly greater frequency of encapsulated receptors than the other parts, which may be correlated with the high shear forces to which the lumbosacral junction is subjected.21 Moreover, the differences in the distribution of receptors between anterior and posterior parts of the L5–S1 disc may be associated with the regional differences in the tensile properties of the disc with the anterior regions having a larger tensile modulus than the posterior ones.22 All neural structures detected were located either in the longitudinal ligaments, or the superficial lamellae of the annulus fibrosus, which are the areas sustaining the greatest pressure or tension during extremes of movement.22 In a mechanic model, during axial loading of a motion segment, compressive stresses in the nucleus (devoid of neural and vascular structures) will generate tensile stresses in the peripheral annulus, which is rich in neural receptors. In conclusion, this study confirms the existence of an abundant network of encapsulated and non-encapsulated receptors in the intervertebral discs of the lower lumbar spine in normal human subjects. The principal role of encapsulated structures is assumed to be the continuous monitoring of position, velocity and acceleration (kinesthesia). Free nerve fibers are likely to be involved in nociception or regulation of vessel tone (autonomic fibers). Acknowledgements We wish to thank Mr Trifon Totlis, Medical Student of the Aristotle University of Thessaloniki, for his technical assistance. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jocn.2009.09.032.
745
References 1. Jung A, Brunschwig A. Recherches histologiques sur l’innervation des articulations des corps vertebraux [Histological study of the innervation of vertebral body joints]. Presse Med 1932;40:316–7. 2. Freeman M, Wyke B. The innervation of the knee joint. An anatomical and histological study in the cat. J Anat 1967;101:505–32. 3. Jackson HC, Winkelmann RK, Bickel WH. Nerve endings in lumbar spinal column. J Bone Joint Surg (Am) 1966;48:1272–81. 4. Malinsky J. The ontogenetic development of nerve terminations in the intervertebral disc of man. Acta Anat 1959;38:96–113. 5. Palmgren T, Gronblad M, Virri J, et al. An immunohistochemical study of nerve structures in the annulus fibrosus of human normal intervertebral discs. Spine 1999;20:2075–9. 6. Roberts S, Eisenstein SM, Menage J, et al. Mechanoreceptors in intervertebral discs. Morphology, distribution and neuropeptides. Spine 1995;20:2645–51. 7. Yoshizawa H, O’Brien J, Smith WT, et al. The neuropathology of intervertebral discs removed for low back pain. J Pathol 1980;132:95–104. 8. Kallakuri S, Cavanaugh JM, Blagoev DC. An immunohistochemical study of innervation of lumbar spinal dura and longitudinal ligaments. Spine 1998;23:403–11. 9. McCarthy PW, Carruthers B, Martin D, et al. Immunohistochemical demonstration of sensory nerve fibers and endings in lumbar intervertebral discs of the rat. Spine 1991;16:653–5. 10. Rhalmi S, Yahia L, Newman N, et al. Immunohistochemical study of nerves in lumbar spine ligaments. Spine 1993;18:264–7. 11. Isobe T, Ishioka N, Kocha T, et al. Chemical structure and molecular evolution of S-100 proteins. Prot Biol Fluids 1983;30:21–4. 12. Pionchon H, Tommasi M, Pialat J, et al. Study of innervation of the spinal ligaments at the lumbar level. Bull Assoc Anat (Nancy) 1986;70:63–7. 13. Yahia L, Rhalmi S, Newman N, et al. Sensory innervation of human thoracolumbar fascia. An immunohistochemical study. Acta Orthop Scand 1992;63:195–7. 14. Weidner N, Rice D. Intervertebral disc material: criteria for determining probable prolapse. Hum Pathol 1988;19:406–10. 15. Yasuma T, Koh S, Okamura T, et al. Histological changes in aging lumbar intervertebral discs. J. Bone Joint Surg 1990;72-A:220–9. 16. McLain RF, Pickar JG. Mechanoreceptor endings in human thoracic and lumbar facet joints. Spine 1998;23:168–73. 17. Williams PL, Warwick R, Dyson M, et al. Gray’s anatomy. 37th ed. New York: Churchill Livingstone; 1989. p. 906. 18. Barr ML, Kiernan JA. The human nervous system. 5th ed. Philadelphia: Lippincott; 1988. p. 33. 19. Zimny ML. Mechanoreceptors in articular tissues. Am J Anat 1988;182:16–32. 20. Bogduk N, Tynan W, Wilson AS. The nerve supply of the human lumbar intervertebral discs. J Anat 1981;132:39–56. 21. Skaggs DL, Weidenbaum M, Iatridis JC, et al. Regional variation in tensile properties and biochemical composition of the human lumbar anulus fibrosus. Spine 1994;19:1310–9. 22. Buckwalter JA, Mow VC, Boden SD, et al. Intervertebral disk structure, composition and mechanical function. In: Buckwalter JA, Einhorn TA, Simon SR, editors. Orthopaedic basic science: biology and biomechanics of the musculoskeletal system. 2nd ed. Rosemont, Illinois, USA: American Academy of Orthopaedic Surgeons; 2000. p. 548–56.