Neuronal regulation of tendon homoeostasis

Abstract

The regulation of tendon homoeostasis, including adaptation to loading, is still not fully understood. Accumulating data, however, demonstrates that in addition to afferent (sensory) functions, the nervous system, via efferent pathways which are associated with through specific neuronal mediators plays an active role in regulating pain, inflammation and tendon homeostasis. This neuronal regulation of intact-, healing- and tendinopathic tendons has been shown to be mediated by three major groups of molecules including opioid, autonomic and excitatory glutamatergic neuroregulators. In intact healthy tendons the neuromediators are found in the surrounding structures: paratenon, endotenon and epitenon, whereas the proper tendon itself is practically devoid of neurovascular supply. This neuroanatomy reflects that normal tendon homoeostasis is regulated from the tendon surroundings. After injury and during tendon repair, however, there is extensive nerve ingrowth into the tendon proper, followed by a time-dependent emergence of sensory, autonomic and glutamatergic mediators, which amplify and fine-tune inflammation and regulate tendon regeneration. In tendinopathic condition, excessive and protracted presence of sensory and glutamatergic neuromediators has been identified, suggesting involvement in inflammatory, nociceptive and hypertrophic (degenerative) tissue responses. Under experimental and clinical conditions of impaired (e.g. diabetes) as well as excessive (e.g. tendinopathy) neuromediator release, dysfunctional tendon homoeostasis develops resulting in chronic pain and gradual degeneration. Thus there is a prospect that in the future pharmacotherapy and tissue engineering approaches targeting neuronal mediators and their receptors may prove to be effective therapies for painful, degenerative and traumatic tendon disorders.

Introduction

Tendon homoeostasis is a product of a remarkable process involving resorption and formation tightly balanced in time, space and quantity (Magnusson et al. 2010). The most well-known extrinsic factor adapted to regulate tendon protein synthesis and degradation is mechanical loading (Magnusson et al. 2010). An exercise bout in human tendons activates an initial increase in both the synthesis and degradation of collagen. There is an initial loss of collagen after loading but over time a net gain in collagen ensues. However, repeated loading that exceeds the tendons capacity of new collagen formation is harmful and may trigger the development of tendinopathy.

Prolonged unloading on the other hand is also detrimental for the human tendon. Disuse leads to reduced tendon mechanical stiffness (Reeves et al. 2003). Interestingly, in ageing, similar reductions in tendon mechanical properties are observed (Narici et al. 2008). Prolonged unloading postinjury also demonstrated negative effects on tendon mechanical properties and production of extracellular matrix molecules. Unloading of the Achilles tendon postrupture, demonstrated in an animal model that ultimate tensile strength was only 20% compared with that of a freely mobilized group at 2 weeks postrupture (Schizas et al. 2010a). Moreover, in the same unloading model, mRNA expression of extracellular matrix molecules (collagen types I and III, versican, decorin, biglycan), growth factors (BDNF, bFGF) and essential sensory neuropeptide receptors (NK-1, RAMP) were all down-regulated at 2 weeks postrupture (Bring et al. 2009a,b).

Mechanobiology of the tendon, transduction of mechanical force, and how this can be transformed into protein synthesis in the tendon has been an area of intense studies (Magnusson et al. 2010; Wang et al. 2012). The predominant cell type in tendon responsible for the production of collagen and other matrix proteins is the fibroblast (tenocyte). Fibroblasts are connected to each other by gap junctions (Wang et al. 2012) and are fixed to the basic extracellular matrix by transmembrane receptors, that is, integrins (McNeilly et al. 1996). The connection between fibroblasts and the extracellular matrix allows the cells to some extent to sense and respond to mechanical stimuli (Waggett et al. 2006). However, exactly how the fibroblast senses load and how this process is transferred into protein synthesis is still unclear.

If load applied to the tendon results in a small injury then fibroblasts are triggered via inflammatory mediators (e.g. cytokines) to transform into myofibroblasts and subsequently activate production of tendon callus. Moreover, vascular- and tissue-derived cells infiltrate the wounded area and release a cascade of mediators (inflammatory mediators, growth factors, cytokines and neuropeptides), which regulate the protein synthesis (Ackermann, et al., 2008).

Although it is well known both clinically and experimentally that loading improves while unloading impairs tendon protein synthesis, the exact mechanisms and regulatory factors responsible for this mechano-biological transduction are still not fully known. In spite of this gap in knowledge, accumulating data does suggest that the peripheral nervous system, including specific mediators and their receptors, plays an important role in tendon homoeostasis and repair and in tendinopathy (Ackermann et al. 2009).

Ethical Approval

All experiments described were approved by the local Committee for Animal Research and Ethics and conducted in accordance with the Karolinska Institute's protocols. All participants received oral and written information about purpose and procedures of the study and provided written informed consent. Ethical approval was obtained from the Regional Ethical Review Board in Stockholm, Sweden.

Tendon innervation

Generally, tendons exhibit a low degree of innervation, which may partly explain the slow adaptation to repetitive loading, prolonged healing and vulnerability to chronic injuries (Ackermann 2001; Ackermann et al. 2009) (Figure 1). The innervation of tendons originates from neighbouring muscular, cutaneous and peritendinous nerve trunks (Stilwell 1957). From the myotendinous junction, nerve fibres cross and enter the endotenon septa. In the paratenon, nerve fibres form rich plexuses and send small branches that penetrate the epitenon. Nerve fibres do not under normal conditions enter the tendon proper, but terminate as nerve endings on the different surfaces of the tendon (paratenon, epitenon, endotenon) (Ackermann et al. 2009).

Figure 1.

(a–c) Overview micrographs of longitudinal sections through the Achilles tendon. Incubation with antisera to general nerve marker PGP 9.5. Micrographs depict the proximal half of the Achilles tendon at increasing magnification in figures (a–c). Arrows denote varicosities and nerve terminals. The typical vascular localization of NPY is depicted in lower left (b), whereas the free nerve endings are typical localization of SP (c). The immunoreactivity is seen in the paratenon and surrounding loose connective tissue, whereas the proper tendinous tissue, notably, is almost devoid of nerve fibres. Pt = paratenon. Reproduced with permission from (40).

The nerves innervating tendons are composed of a low degree of myelinated, fast-transmitting Aα- and Aβ-fibres and a higher degree of unmyelinated, slow-transmitting Aγ-, Aδ-, B- and C-fibres (Hogervorst & Brand 1998; Ackermann 2001).

The nerve endings of (i) Aα- and Aβ-fibres are of types I-III and mediate mechanoception. These include type I or Ruffini corpuscles (pressure and stretching sensors), II or Vater–Pacini corpuscles (pressure sensors, reacting to acceleration and deceleration of movement) and type III or Golgi tendon organs (tension receptors) (Strasmann et al. 1990; Jozsa & Kannus 1997). Tension receptors (type III) have been found mostly in the myotendinous junction and insertion areas (Jozsa & Kannus 1997).

The nerve endings of (ii) Aγ-, Aδ- and C-fibres are of type IVa, so-called nociceptors, mediating the deep tissue pain and hyperalgesia that are characteristic features of pain in tendinopathy. The nerve endings of B-fibres, which are autonomic, consist of type IVb fibres that are mainly localized in the walls of small arteries, arterioles, capillaries and postcapillary veins and which exert vasomotor actions (Jozsa & Kannus 1997).

Thus, one can conclude that mechanoception, nociception and vasomotor modulation are three of the main functions of tendon innervation. However, in addition to these classical afferent functions, it is now well established that the peripheral nervous system also participates in the regulation of a wide variety of efferent actions on cell proliferation, expression of cytokines and growth factors, inflammation, immune responses and hormone release. This is called the paradoxical ‘efferent’ role of afferent nociceptive fibres and was suggested already in 1901 by Bayliss (Bayliss 1901). In the last decade this tendon innervation related to mediator phenotype and neuronal regulation of tendon homoeostasis has received increasing attention.

In addition to classical neurotransmitters (monoamines, acetylcholine, amino acids), several neuropeptides, which act as chemical messengers in the central and peripheral nervous system, have been identified in tendons (Ackermann et al. 2009). Neuropeptides differ from classical neurotransmitters in several respects (Hokfelt et al. 1980; Ackermann 2001). Neuropeptides are synthesized in the cell bodies and transported both centrally and distally. Classical neurotransmitters on the other hand are synthesized in the axon terminals.

The synthesis and turnover of classical transmitters is more rapid than that of neuropeptides. Classical transmitters occur in small and large synaptic vesicles, while neuropeptides mostly have been demonstrated in large dense-core vesicles (Hokfelt et al. 1980). Neuropeptides are released from these vesicles differently from classical transmitters depending on the frequency of the action potentials. A low impulse frequency selectively activates small vesicles releasing classical transmitters, whereas higher frequencies release neuropeptides from large vesicles. The large vesicles permit more than one functional neuropeptide to be processed and released together with classical transmitters offering a variety of functional interactions (Hokfelt et al. 1980). In fact, several neuropeptides and also transmitters have been found colocalized in tendon suggesting functional interactions in this tissue (Ackermann 2001; Ackermann et al. 2009).

The effects of neuropeptides and classical transmitters are elicited by different receptor mechanisms. While classical transmitters act on ligand-gated ion channels, neuropeptides act by binding to specific plasma membrane receptors, called G-protein-coupled receptors (Audet & Bouvier 2012). When the receptors are stimulated, they generate various second messengers, which can trigger a wide range of effector mechanisms regulating cellular excitability and function. Several of these receptors have been identified in tendons (Ackermann et al. 2009) (Table 1).

Table 1.

Neuromediators in tendons

Type	Subtype	Mediator	Receptor	Actions
Sensory (type IVa)	Sensory	SP CGRP NKA NKB*, NPK*, NPG*	NK1 CRLR, RAMP-1 NK2* NK3*	Pro-inflammatory
	Opioid and opioid like	Enkephalins: LE, ME, MEAP Dynorphins: DYN BND EndomorphinsND NociceptinND Opioid like: GAL, SOM	δ-opioid receptor κ-opioid receptorND μ-opioid receptorND N/OFQ receptor* GalR1-3*, SSTR1-5*	Anti-inflammatory
Autonomic (type IVb)	Sympathetic	Noradrenaline NPY	α-,β-adrenoceptors Y1-3*	Pro(anti)-inflammatory
	Parasympathetic	Acetylcholine VIP	Nicotinic, muscarinic VPAC1-2*, PAC1*	Anti-inflammatory
Excitatory	Glutamatergic amino acid	Glutamate	NMDA, mGlu, AMPA*, Kainate*	Sensitization

ND Not detected in tendon.

^*Not yet assessed in tendon.

Neuromediators, neuropeptides and classical transmitters, in healthy tendon

On the whole, tendons seem to exhibit neuronal mediators of a similar kind to those that have been observed in other organs of the body which consist of a variety of sensory mediators which include those which have opioid, autonomic and excitatory properties (Table 1). However, in contrast to most other tissues, the tendon proper during normal conditions is devoid of nerve fibres. Innervation is found in the tendon envelope, that is, the paratenon, endotenon and surrounding loose connective tissue (Figure 1). Another vital feature of tendon neuroanatomy is the balance between different mediators, that is, pro- and anti-inflammatory peptides (Tracey 2002). These observations would suggest that the homoeostatic regulation of healthy tendon tissue is highly dependent on balanced neuromediator modulation occurring in the tendon envelope.

Sensory neuromediators

The sensory nerves type IVa act principally through release of slowly acting mediators, that is, neuropeptides and opioids. In tendons, sensory neuropeptides with nociceptive and pro-inflammatory effects [substance P (SP), calcitonin gene-related peptide (CGRP) and neurokinin A (NKA)], sensory modulatory neuropeptides with anti-inflammatory actions (galanin (GAL), somatostatin (SOM)), as well as opioid neuropeptides with antinociceptive and anti-inflammatory effects (Leu-enkephalin (LE), Met-enkephalin (ME), Met-enkephalin-Arg-Phe (MEAP), Met-enkephalin-Arg-Gly-Leu (MEAGL), nociception) have been identified (Table 1) (Figure 2a–d) (Ackermann 2001; Ackermann, Calder and Aspenberg, 2008). With a quantitative approach based on radioimmunoassay, the different levels of these particular neuropeptides have been assessed in tendon. Thus, the relative concentrations of sensory mediators and modulators might be considered to suggest physiological nociceptive thresholds, which could be altered due to internal or external stress (e.g. loading).

Figure 2.

(a–d) Immunofluorescence micrographs of longitudinal sections through the Achilles tendon after double staining with antisera to SP and CGRP (a), SP and GAL (b), LE and DOR (c) and incubation with antisera to ME (d). A coexistence of SP and CGRP is seen in nerve fibres localized in the paratenon (a), suggesting possible pro-inflammatory actions. Moreover, SP is also colocalized with GAL (b), reflecting anti-inflammatory actions. The immunoreactivity displaying coexistence of LE and DOR is seen as free nerve endings in the paratenon (c), indicating a potential peripheral anti-nociceptive system. ME immunoreaction is localized in a vessel wall (d). t = tendon tissue; Pt = paratenon; Bar = 50 μm. Reproduced with permission from (24 to 25).

Abundant amounts of sensory neuromediators have been detected in vascular nerve fibres in the surrounding loose connective tissues, which may reflect an important role in the regulation of blood flow to the tendon structures. Both SP and CGRP, in particular the latter, have been reported to be potent vasodilators (Brain et al. 1985). Additionally, they have also been demonstrated to exert pro-inflammatory effects, for example, by enhancing protein extravasation, leucocyte chemotaxis and cytokine production (Brain & Williams 1985; Maggi 1995). The anti-inflammatory neuromediators (GAL, SOM) and the opioids have also been identified in vascular nerve fibres colocalized with SP, suggesting modulatory and anti-inflammatory actions (Ackermann et al. 1999, 2001a,b). The occurrence of sensory neuromediators in free nerve endings unrelated to vessels predominantly seen in the paratenon may suggest nociceptive, trophic and immune regulatory roles.

In tendons, receptors for SP (neurokinin 1, NK1) and CGRP (calcitonin receptor-like receptor, CRLR, and receptor activity-modifying proteins, RAMP-1) have been identified by combined morphological and gene quantitative approaches (Bring et al. 2009a,b) (Table 1). These receptors have been found localized on tendon cells, immune cells, blood vessels and on free nerve endings unrelated to vessels. These localizations strengthen the suggested trophic, immune regulatory, vasoregulatory and nociceptive effects of the sensory neuromediators on tendon homoeostasis.

Autonomic neuromediators

The sympathetic nervous system regulates inflammatory processes at local and systemic levels through the balanced release of sympathetic and parasympathetic mediators. The sympathetic mediator noradrenaline (NA) together with neuropeptide Y (NPY) is released upon injury or nociceptive input, while parasympathetic mediators acetylcholine (ACh) and vasoactive intestinal polypeptide (VIP) are released by vagus nerve activation called the ‘cholinergic anti-inflammatory pathway’ (Tracey 2002) (Figure 3a–c).

Figure 3.

(a–c) Immunofluorescence micrographs of longitudinal sections through the Achilles tendon after incubation with antisera to noradrenaline (NA) (a), NPY (b) and VIP (c). NA-positive fibres are mainly found as nerve terminals in outer layers of the blood vessel walls. The NPY-positive fibres are arranged as nerve terminals in the vessel walls. VIP-positive nerves are arranged as a ‘fence’, surrounding the proper tendon of small varicosities in the paratenon. t = tendon tissue; Pt = paratenon; Bar = 50 μm. Reproduced with permission from (27).

The occurrence of sympathetic NA and NPY as well as parasympathetic mediators ACh and VIP has been demonstrated in tendon (Ljung et al. 1999; Ackermann et al. 2001a,b; Wall et al. 2004; Danielson et al. 2006, 2007a,b,c). NA and NPY were localized mainly as networks around larger blood vessels located in the loose connective tissue around the main body of the tendon (Figure 3a,b). These observations suggest that the sympathetic tendon vasoregulation occurs predominantly in the tendon envelope, that is, the surrounding loose connective tissue.

Parasympathetic VIP has been observed in nerve fibres forming a ‘fence’ in the paratenon (Figure 3c). These nerves were evenly distributed between vascular structures and free nerve endings. In contrast to sympathetic nerve terminals occurring mostly in larger vessels, VIP has predominantly been found around smaller blood vessels (Ljung et al. 1999; Ackermann et al. 2001a). This observation may reflect that NA and NPY predominantly regulate the main blood flow to the tendon, while VIP is responsible for fine-tuning blood flow at a microlevel. The distribution of VIP as a ‘fence’ in the paratenon would seem to comply with an anti-inflammatory effect in the periphery (Figure 3c). The strong anti-inflammatory effect of VIP has been suggested to act through inhibition of T-cell proliferation and migration (Moody et al. 2011).

Adrenergic receptors responding to NA (A-adrenoceptors) and to NPY (Y1) have been identified on tendon cells, blood vessel walls and on nerve fibres (Wall et al. 2004; Danielson et al. 2007a,b). These localizations suggest that adrenergic stimulation of tendons may be involved in proliferation of tenocytes, endothelial cells and possibly nerve cells.

Cholinergic receptors for ACh, nicotinic and muscarinic, have been detected on tendon cells and blood vessels (Danielson et al. 2006, 2007c), whereas the receptors for VIP (VPAC1, VPAC2 and PAC1) still have to be identified in tendons. The localization of nicotinic and muscarinic receptors on tendon cells and especially on inflammatory cells suggests an anti-inflammatory action in inhibiting the release of pro-inflammatory cytokines.

Excitatory neuromediators

Accumulating data suggests that modulation of glutamate signalling by inhibition of its receptors, ionotropic (NMDA, AMPA, kainate) and metabotropic (mGlu), may have potential for targeted therapy in several persistent pain conditions (Ackermann 2012; Wozniak et al. 2012). Moreover, glutamate signalling is implicated in programmed cell death, apoptosis.

Recently, glutamate and several of its receptors have been identified in tendon on nerve fibres, blood vessels and cells (Figure 4) (Scott et al. 2007; Schizas et al. 2010b, 2012). These localizations of glutamate have been verified by identification of the glutamate receptor NMDAR1 and of the metabotropic receptors mGluR1, mGluR5 and mGluR6-7 on similar locations. By double-staining techniques, glutamate has been found to colocalize with its receptors suggesting that glutamate signalling may be involved in regulating excitability of tenocytes, endothelial cells and nerve activation (Schizas et al. 2010b, 2012).

Figure 4.

(a–d) Immunofluorescence double-staining micrographs of longitudinal sections through tendinopathic patellar tendons focusing on the tendon proper, after incubation with antisera for PGP 9.5 (a, c, thin arrows), N-methyl-D-aspartate receptor type 1 (NMDAR1) (b) and glutamate (d). Thick arrows denote neuronal NMDAR1 and within the tendon proper arrowheads denote neuronal glutamate, which is totally absent in the controls (scale bar = 25 um). Reproduced with permission from (38).

Neuronal plasticity after tendon injury

After tendon injury and during healing, the peripheral nervous system responds by nerve ingrowth into the tendon proper, which during healthy conditions is more or less aneuronal and avascular (Figures 5 and 6) (Ackermann et al. 2002, 2008). Nerve sprouting and growth within the tendon proper are followed by a time-dependent expression of neuropeptides during the tendon healing process. After the healing process is finished, sprouting nerve fibres within the tendon proper retract to the surrounding structures, that is, the paratenon and loose connective tissue. These observations of early nerve regeneration are in line with observations on bone, ligament and skin healing, indicating that nerve ingrowth and subsequent retraction are fundamental aspects of tissue repair (Kishimoto 1984; Hukkanen et al. 1993; Martin 1997; Li et al. 2001; Salo et al. 2008).

Figure 5.

(a–b) Overview micrographs of longitudinal sections through the Achilles tendon at 2 weeks postrupture. Incubation with antisera to a nerve growth marker, GAP-43. Micrographs depict the proximal half of the Achilles tendon at increasing magnification in figures (a,b). Arrows denote varicosities and nerve terminals. The GAP-positive fibres, indicating new nerve fibre ingrowth, are abundantly observed in the healing proper tendon tissue. Reproduced with permission from (40).

Figure 6.

Area occupied by nerve fibres (%) immunoreactive to SP, CGRP and GAL in relation to total area, in the midthird of the tendon, over 16 weeks postrupture (mean ± SEM). Reproduced with permission from (41).

Inflammatory healing phase

At one week after tendon injury, increased occurrence of SP- and CGRP-positive nerve fibres has been demonstrated to be predominantly located in blood vessel walls surrounded by inflammatory cells in the loose connective tissue (Ackermann et al. 2003) (Figure 7a). The findings comply with the nociceptive role of sensory neuropeptides, but also with a pro-inflammatory role. Thus, SP release enhances vasopermeability, probably to stimulate recruitment of leucocytes and cytokine production (Aubdool & Brain 2011). Interestingly, the opioid-like mediator GAL only showed weak expression, reflecting low anti-inflammatory actions in early repair (Figure 6) (Fernandes et al. 2009). Similarly, the occurrence of the autonomic mediators NPY and VIP was sparse during the inflammatory phase, also suggesting low anti-inflammatory actions.

Figure 7.

(a,b) Immunofluorescence micrograph of longitudinal sections through healing Achilles tendon 1 (a) and 2 (b) weeks postrupture after incubation with antisera to CGRP. Nerve fibres immunoreactive to CGRP at week 1 are seen as vascular and free nerve endings in the loose connective tissue (a). At week 2, CGRP immunoreactivity occurs mainly in the healing tendinous tissue as sprouting free nerve fibres (b). v = blood vessel; lct = loose connective tissue; t = proper tendon tissue; Bar = 50 μm. Reproduced with permission from (41).

During the first-week post-tendon injury, glutamatergic signalling molecules have been shown using microarray followed by real-time PCR analyses to be up-regulated (Molloy et al. 2006a). The elevated glutamate signalling has been substantiated by in vivo microdialysis experiments on patients with Achilles tendon ruptures (Greve et al. 2012). Thus, increased glutamate levels during the inflammatory phase (Greve et al. 2012) seem to persist until at least week 6 [unpublished data], suggesting essential actions in tendon healing as has also been shown in maintenance of bone tissue. Interestingly, the expression of glutamatergic signalling molecules during tendon repair exhibited a temporal relationship to genes involved in embryonic development (Molloy et al. 2006b).

Proliferative healing phase

From one to six weeks postrupture, a striking shift in neuronal occurrence has been demonstrated to occur from the surrounding loose connective tissue into the proper tendinous tissue (Ackermann et al. 2003). This suggests the transition of a predominantly inflammatory into a proliferative repair phase. Recent data from human Achilles tendon rupture demonstrate at two weeks postsuture that inflammatory markers and pro-inflammatory cytokines were below detection limit, corroborating that mainly regenerative actions occur at this time point (Ackermann et al. 2012).

During weeks 2–6 postinjury, the expression of SP and CGRP was shown to peak in an animal model (Figure 6). Notably, this peak occurred at the rupture site of the proper tendon, while the sensory neuropeptide expression in the surrounding loose connective tissue declined. The most conspicuous finding was the occurrence of SP and CGRP in free sprouting nerve endings among fibroblasts in the healing tendinous tissue (Figure 7b). The observation might reflect a stimulatory role of sensory neuropeptides on cell proliferation, as demonstrated in cultured fibroblasts (Nilsson et al. 1985). Recently, SP has also been demonstrated to stimulate recruitment of stem cells to the healing area (Hong et al. 2009), which would be consistent with the localization of SP in free nerve endings within the healing tendon proper.

SP and CGRP are also known to stimulate proliferation of endothelial cells (Haegerstrand et al. 1990; Ziche et al. 1990), indicating that SP and CGRP fibres around newly formed blood vessels in the rupture site would comply with a role in angiogenesis (Figure 7).

A prerequisite for functional effects of the sensory neuropeptides SP and CGRP is the expression of their receptors in the healing tendon tissue. Thus, it has been shown that the receptor expression of SP (NK1) and CGRP (CRLR and RAMP-1) in the healing tendon is significantly up-regulated at two weeks, but not at one week, post-tendon rupture (Bring et al. 2009a,b).

During the proliferative phase, the presence of the autonomic mediators NPY and VIP was sparse until week four (Ackermann et al. 2003). The observations of low sympathetic (NPY) innervation in the experimental model suggest that vasoconstriction is down-regulated during tissue repair. The balance between the vasoconstrictive actions of NPY and the vasodilatory actions of CGRP is of critical importance for the supply of oxygen and nutrients to the healing area. The high ratio of CGRP to NPY in the proliferative healing phase probably reflects highly perfused vessels, a necessity for tissue repair (Ackermann et al. 2003). Similar observations on the ratio CGRP to NPY have been made in studies on re-innervation of skin flaps, where CGRP immunoreactivity emerged at two weeks postoperatively and that of NPY two weeks later (Karanth et al. 1990). The low levels of NPY observed may also promote angiogenesis (Zukowska-Grojec et al. 1998).

The reduced expression until week four of parasympathetic VIP possibly pertains to its inhibitory action on the pro-inflammatory effects of SP and CGRP (Moody et al. 2011). Thus, the observation suggests less inhibition of SP and CGRP, which are presumed to be important regulators of early tissue repair.

Remodelling healing phase

During weeks 6–16 post-tendon rupture, it has been demonstrated that the nerve fibres retract from the proper tendon tissue to their normal location in the paratenon and surrounding loose connective tissue (Ackermann et al. 2003) (Figure 6). This process appears to end simultaneously with the completion of paratenon repair.

Between weeks 4–6, increased opioid-like GAL has been observed both around vessels and in free nerve endings enveloping the healing tendon (Ackermann et al. 2003) (Figure 6).

The emergence of GAL, known to modulate the effect of sensory neuropeptides, would seem to reflect inhibition of the early inflammatory and nociceptive response to injury. Thus, GAL has been demonstrated to mitigate the pro-inflammatory and nociceptive effects of SP (Fernandes et al. 2009).

Interestingly, between weeks 4–6, corresponding to the transition of the proliferative into the remodelling phase, a dramatic increase in the expression of the autonomic neuropeptides VIP and NPY has been demonstrated (Ackermann et al. 2003). The increase in the immunoreactivity for VIP and NPY was observed both around vessels and in free nerve endings in a ‘border zone’ enveloping the healing tendon. The up-regulation of parasympathetic VIP may be explained by its inhibitory effect on immune cells expressing pro-inflammatory cytokines (Moody et al. 2011). The increased occurrence of sympathetic NPY during this phase, mostly seen around vessels, may be attributed to its vasoconstrictive actions. Increased vasoconstriction leads to a relative hypoxia, which enhances the tensile strength of the tendon by switching production of collagen from type III to type I (Steinbrech et al. 1999).

Subsequent to the elevated expression of GAL, VIP and NPY, a significantly decreased expression of SP and CGRP has been observed in the healing tendon (Ackermann et al. 2003) (Figure 6). Thus, the early remodelling phase after tendon injury seems to be characterized by an increased expression of GAL, VIP and NPY, all of which are known to modulate the effects of SP and CGRP. Hypothetically, this modulation is required to end the inflammatory and reparative processes, thereby facilitating entry to and maintenance of the remodelling phase.

Neuronal dysregulation in tendinopathy

The underlying histology in tendinopathies with chronic pain is often characterized as reflecting a failed healing response. Interestingly, the innervation pattern in tendinopathic tissue resembles that observed during the proliferative phase of healing after tendon injury (Table 2).

Table 2.

Neuromediators in tendinopathy

Type	Subtype	Mediator	Receptor
Autonomic	Sympathetic	Noradrenaline ↓ NPY	A-adrenoceptors ↑ Y1 ↑
	Parasympathetic	Acetylcholine ↑ VIP*	Nicotinic* Muscarinic ↑
Sensory	Sensory	SP ↑	NK1 ↑
	Opioid and opioid like	Cannabinoids*	CB1 ↑
Excitatory	Glutamatergic amino acid	Glutamate ↑	NMDA1 ↑ Phospho-NMDA1 ↑ mGluR1 ↑ mGluR5 ↑ mGluR6-7 →

^*Not yet assessed in tendon.

Sensory neuromediators

Chronic painful tendons with tendinopathy exhibit new ingrowth of sensory nerve fibres (Figure 8) (Sanchis-Alfonso et al. 2001; Schubert et al. 2005; Lian et al. 2006) (49, 50, 98), which is also observed during tissue proliferation in healing tendons (Ackermann et al. 2002). Closer analysis of tendinopathic tissue biopsies has revealed the nerve fibres occurring mainly as thin, varicose and sprouting nerve terminals within the tendon proper (Lian et al. 2006). The observation of increased ingrowth of sensory nerves into the painful tendon proper, seen as sprouting free nerve endings, possibly represents nociceptors responding to mechanical stimuli by initiating pain signalling.

Figure 8.

(a,b) Immunofluorescence micrographs of longitudinal sections from biopsies of healthy Achilles tendon (a) and tendinosis tissue (b) after immunostaining for SP. Arrows denote varicosities and nerve terminals. The micrograph illustrates SP-positive nerve fibres in close vicinity to a proliferated vessel (b). v = blood vessel. Bar = 50 μm. Reproduced with permission from (59).

In normal tendon repair, sensory nerve ingrowth and expression of SP and CGRP are correlated with increased nociception (Ackermann et al. 2003). Normally, the inflammatory phase is followed by peripherally acting autonomic and opioid-like signalling, coinciding with decreased nociception (Ackermann et al. 2003). Hence, the neuronal dysregulation in tendinopathy, characterized by aberrant increase in sprouting sensory nerves and increased expression of SP, presumably triggers pain signalling and possibly also the hyperproliferative/degenerative changes associated with tendinopathy (Schubert et al. 2005; Lian et al. 2006) (Figure 9). The increase in sensory neuropeptide SP in tendinopathy could additionally reflect pro-inflammatory and trophic actions. SP is known to participate in inflammatory actions such as vasodilation, plasma extravasation and release of cytokines. SP has also been reported to stimulate proliferation of fibroblasts and endothelial cells, as well as the production of transforming growth factor β in fibroblasts. Thus, SP may well contribute to the morphologic changes observed in tendinopathic patients, that is, tenocyte transformation, hypercellularity, scar adhesions and presumably neovascularization.

Figure 9.

(a,b) Haematoxylin and eosin micrographs of longitudinal sections from biopsies of healthy patellar tendon (a) and painful tendinopathy (b). Arrows denote tenocytes. The healthy tendon is homogeneous, with organized parallel collagen structure and thin, elongated tenocytes (a). The tendinopathy, on the other hand, is marked by collagen disorganization, increased cell count, activated tenocytes and vascular ingrowth in the tendon proper (b). V = blood vessel. Bar = 50 μm. Reproduced with permission from (59).

The ongoing morphologic alterations may reflect protracted or failed healing. Importantly, the role of nerves and neuropeptides has to be differentiated during an initial response to injury as compared to their role during the chronic phases of tendinopathy, phases that can lead to overt degeneration and, in some instances, rupture of the tendon. Neuropeptides may exert different functions during different ‘phases’ of healing. SP for example promotes early proliferative tendon repair but may during prolonged expression in tendinopathy, via for example, the neuropeptide–mast cell–tenocyte axes, play a role in the development of hyperproliferative/degenerative tissue changes.

All of the above-mentioned effects of SP are plausible with respect to tendon pathology because its receptor, NK-1, has been detected in tenocytes, blood vessel walls and in nerve fibres in tendinopathy (104). SP has further been shown to stimulate nociceptor endings directly in an autocrine/paracrine manner. Similar actions could presumably occur in tendinopathy since the NK-1 receptor is present.

Autonomic neuromediators

Interestingly, tendinopathic patients exhibit a decreased occurrence of sympathetic nerve fibres, immunoreactive to noradrenaline. Microscopic analysis demonstrated that sympathetic nerves related to blood vessels were distinctively decreased (Figure 10) (Lian et al. 2006). Computerized image analysis corroborated a 50% drop in vascular nerve fibres immunopositive to noradrenaline in the painful tendons (Lian et al. 2006). The reduction in vasoregulatory noradrenaline would seem to comply with a reduced blood flow and a suppressed antinociceptive function. Recently, endogenous noradrenaline release was demonstrated to lead to decreased demands of pain relief after surgery, possibly by secretion of opioids from leucocytes (Kager et al. 2011). Moreover, patients with painful rheumatoid arthritis have also been found to exhibit a decrease in vascular innervation expressing noradrenaline (Klatt et al. 2012). To counteract this decreased sympathetic innervation, synovial immune cells in rheumatoid arthritis patients respond by up-regulating noradrenaline. Similarly, in tendinopathy patients, elevated noradrenaline production has been suggested in morphologically altered tenocytes (Bjur et al. 2008).

Figure 10.

(a,b) Immunofluorescence micrographs of longitudinal sections through the patellar tendon of healthy control (a) and painful tendinopathy (b) stained for TH (a marker for noradrenaline). Arrows denote nerve fibres. In the healthy tendon, a strong relation is seen between blood vessels and TH-positive nerves (a). In painful tendinopathy, a decreased number of TH-positive nerves, which are blood vessel related, is seen. V = blood vessel. Bar = 50 μm. Reproduced with permission from (59).

Adrenoreceptors for noradrenaline have also been identified in tendinopathy. Immunoreaction for the alpha-1-adrenoreceptor has been detected in blood vessel walls, nerve fascicles and tenocytes (Danielson et al. 2007a; Bjur et al. 2008). Adrenergic activation of alpha-1-adrenoreceptors has been demonstrated to stimulate cell proliferation and differentiation (Zhang & Faber 2001). Thus, elevated noradrenaline in tendon cells may stimulate the alpha-1-adrenoreceptors and contribute to tenocyte excessive cell proliferation and differentiation. On the other hand, decreased vascular innervations expressing noradrenaline may be compensated by the observed increased alpha-1-adrenoreceptor immunoreactions in blood vessel walls of tendinopathic patients (Danielson et al. 2007a; Bjur et al. 2008).

The cholinergic innervation in tendinopathy appears to be relatively scarce. Neuronal immunoreactivity to choline acetyltransferase, vesicular acetylcholine transporter and acetylcholinesterase in tendons is limited compared with other tissues investigated. However, whether cholinergic innervation in tendinopathy is significantly lower than that of healthy tendons is still unclear (Danielson et al. 2006, 2007c; Bjur et al. 2007), but remains interesting because VIP and ACh may counteract the effects of SP. Immunohistochemical studies have identified activated markers of cellular acetylcholine production in human tendinopathy, mainly seen in morphologically altered tenocytes. The intensified immunoreaction to choline acetyltransferase and vesicular acetylcholine transporter was not detected in normal tenocytes (Danielson et al. 2006, 2007c; Bjur et al. 2007). This suggests that acetylcholine either is involved in regulating the tenocyte transformation seen in tendinopathy or that acetylcholine synthesis is initiated in tenocytes in response to development of tendinopathy.

Parasympathetic muscarinic acetylcholine receptors (M2) have been identified in human tendinopathy. Thus, M2-receptors have been identified in tendon cells, blood vessel walls and nerve fibres (Danielson et al. 2006, 2007c).

A more intense M2-immunostaining was observed in the tendinopathic tendons compared with normal tendons. This was in morphologically altered tenocytes, and was not seen in normal tenocytes. The observations demonstrate an endogenous autocrine and/or paracrine acetylcholine and M2-receptor signalling in transformed tenocytes.

Excitatory neuromediators

Similar to the findings in early healing, elevated levels of glutamate have been detected in patients with tendinopathy by microdialysis and by immunohistochemistry (Alfredson et al. 2001; Schizas et al. 2010b). Moreover, the specific localization for the increased glutamate levels has been established in tendinopathic patients (Scott et al. 2007; Schizas et al. 2010b, 2012). Up-regulated glutamate occurrence is observed in morphologically altered tenocytes, in the endothelial and adventitial layers of blood vessel walls and in nerve fibres (Figure 11).

Figure 11.

(a–d) Micrographs of longitudinal sections through patellar tendon biopsies after incubation with antisera to phospho-NMDAR1 (activated NMDA) in patients with patellar tendinosis (a,b) and controls (d). Micrographs of longitudinal sections through patellar tendon biopsies stained for PGP 9.5 (a general nerve marker) in patients with patellar tendinosis are shown in (c). Phospho-NMDAR1 immunoreactivity in the tendon proper was exclusively observed in tendinosis as immunoreactive cells (a, arrowheads) as well as penetrating nerve fibres (b, arrows). The controls did not exhibit phospho-NMDAR1 immunoreactivity within the tendon proper (d). The occurrence of PGP 9.5 within the tendon proper depicted extensive nerve ingrowth in tendinopathy not seen in the controls (c, thin arrows; bar = 100 mm). Reproduced with permission from (39).

Several glutamate receptors have been identified in tendinopathy patients, ionotropic [NMDAR1 and activated NMDAR (phospho-NMDAR)], metabotropic mGluR 5, which increases NMDA excitability and mGluR 6,7 that decrease NMDA excitability. Metabotropic mGluR 1, however, was not identified (Schizas et al. 2010b, 2012).

Of the identified glutamate receptors, a significant up-regulation has been demonstrated of all receptors except mGluR 6,7, which is supposed to inhibit the NMDA excitability (Table 2). Quantitative assessments demonstrated a 9-fold increase in NMDAR1 in tendinopathic patients and a 5-fold increase in phospho-NMDAR1 occurrence (Figure 11). (Schizas et al. 2010b, 2012). This was corroborated by a study on rat supraspinatus tendon overuse likewise demonstrating NMDAR1 up-regulation (Molloy et al. 2006a).

Closer analysis has demonstrated the specific localization of the different glutamate receptors. NMDAR-1 and phospho-NMDAR1 were detected by Schizas et al. on sprouting nerve fibres, newly formed blood vessels and transformed tenocytes (Schizas et al. 2012). mGluR 5 was mostly observed on transformed tenocytes. These localizations suggest involvement of glutamate receptors in tendinopathy regulating excitability of tenocytes, endothelial cells and nerves.

Maybe one of the most intriguing findings regarding glutamatergic signalling is the recent report demonstrating that elevated glutamate coexisted with its up-regulated receptor NMDA1 in nerve fibres, morphologically altered tenocytes and blood vessels, which may reflect cell hyperexcitation involved in cell proliferation/differentiation. However, none of the controls exhibited neuronal coexistence of glutamate and NMDAR1 in contrast to prominent neuronal occurrence in all the painful tendons, which strongly suggests a role in pain regulation (Schizas et al. 2010b, 2012).

One of the recent findings established a possible mechanism responsible for activating NMDAR-1 in tendinopathy. It was demonstrated that the elevated occurrence of NMDAR-1 was correlated to that of SP (r² = 0.54, P = 0.03) in tendinopathic tendons, while their occurrence in controls exhibited no correlation (Schizas et al. 2012). These data suggest that SP may be involved in the up-regulation of NMDAR1. In fact, SP is known to activate NMDAR1 by removing the magnesium block (Madden 2002).

Impairment of neuronal supply

Accumulating experimental evidence and clinical observations suggest that reduced neuronal supply leads to impairment of the mechanical properties of connective tendon tissue (Ackermann et al. 2009). Thus, several studies have indicated that denervation impairs the mechanical properties of both normal and injured tendons.

Experimental studies

Sensory neuromediators can selectively be denervated by the use of Spanish pepper (capsaicin). Thus, an experimental study on tendon healing used capsaicin treatment to reduce the concentrations of SP by approximately 60% (Bring et al. 2012) (Figure 12). The study demonstrated that higher residual SP levels correlated with increased mechanical properties of the transverse area, ultimate tensile strength and stress at failure (r = 0.39, P = 0.036; r = 0.53, P = 0.005; and r = 0.43, P = 0.023 respectively).

Figure 12.

(a–b) Immunofluorescence micrographs of sections of tissue from the right hind paw of denervated (b) and controls (a) after incubation with antisera to SP. Arrows denote varicosities and nerve terminals. A marked reduction in SP immunoreactivity is seen in the denervated group (a,b). Reproduced with permission from (70).

Moreover, femoral nerve transection has been demonstrated to impair healing of the medial collateral ligament in rabbits (Ivie et al. 2002). In one study, blood flow, angiogenesis and mechanical strength of the ligament scar were all significantly decreased in denervated limbs compared with normally innervated limbs six weeks after injury (Ivie et al. 2002). Similar results have since then been reported in rats, where femoral nerve transaction or surgical sympathectomy each reduced failure loads of healing MCLs by 50% compared with normally innervated healing MCLs, at 2 weeks after injury (Grorud et al. 2007).

Chemical sympathectomy by systemic administration of guanethidine leads to degradation of the mechanical properties of the intact medial collateral ligament (MCL) of the knee joint in rats after only 10 days of treatment (Dwyer et al. 2004). Ligaments from guanethidine-treated animals exhibited a larger cross-sectional area, a higher wet weight, a decreased modulus of elasticity and a decreased stress at failure. These structural changes may to some extent be explained by the significantly increased mRNA levels for the matrix degrading enzymes MMP-13 and cathepsin K and increased ligament blood flow induced by chemical sympathectomy.

Clinical observations

After denervation, patients with spinal cord injury exhibit material tendon alterations, which can partly be reversed by functional electrical stimulation (Gargiulo et al. 2011).

An observational study reports that patients with sciatica had a high incidence of Achilles tendon ruptures, which could be due to impaired neuronal signalling (Maffulli et al. 1998). Moreover, it has recently been demonstrated that patients with metabolic disorders such as diabetes mellitus are at greater risk of developing various musculoskeletal disorders (Ramchurn et al. 2009). Thus, diabetics often exhibit neuropathy and also decreased levels of sensory neuropeptides, which may be associated with defective tissue healing (Pradhan et al. 2009). Diabetes is associated with impaired connective tissue healing and reduced biomechanical properties, correlated to down-regulated extracellular matrix proteins (Ahmed et al. 2012).

Taken together, these above-mentioned studies strongly support the idea that nerve-derived factors have a powerful influence on the structure, function and healing capacity of dense connective tissues such as ligament and tendon.

Effects of enhanced neuromediator levels

Supraphysiological supply of neurotrophic factors

One essential growth factor which is up-regulated in the tendon in response to injury or a bout of exercise is the nerve growth factor (NGF). NGF displays a wide spectrum of biological functions and has recently been shown to promote wound healing (Micera et al. 2007). After injury/exercise, NGF is taken up by sympathetic and small sensory nerves, where it undergoes retrograde transport and stimulates the production of several nerve substances in the dorsal root ganglia, for example, substance P (Micera et al. 2007). In a model of ligament injury in rats, NGF was continuously administered to the injury site for seven days via an implanted mini-osmotic pump (Mammoto et al. 2008). At seven days after injury, the blood vessel density in the healing area was increased, indicating that angiogenesis was promoted by NGF. By 14 days, an additional increase in nerve fibre density was noted in the NGF-treated specimens. At 42 days after injury, nerve fibre and blood vessel densities were progressively increased, and scar mechanical properties were also significantly improved in the NGF group. Hence, early exposure of injured tissues to exogenous NGF can lead to improved biomechanical tissue strength, a critical factor for structurally important tissues such as tendons and ligaments.

Supraphysiological supply of neuropeptides

Exogenous administration of neuropeptides leads to improved tendon healing. Thus, several experiments demonstrate that local injections of substance P combined with the neutral endopeptidase inhibitors thiorphan and captopril after Achilles tendon rupture in rats enhanced the healing response (Burssens et al. 2005; Steyaert et al. 2006) (Carlsson et al. 2011). SP supply resulted in increased amount of fibroblasts, diameter of organized collagen and amount of collagen III after injury. Histological parameters of collagen fibre orientation and angiogenesis were also improved by SP treatment. Treatment with SP also increased stress at maximal load and work to maximal load in healing Achilles tendon. One important observation was that the sensory nerve ingrowth in the healing tendons appeared to withdraw at a quicker pace in the SP-supplied tendons (Carlsson et al. 2011) (Figure 13). This observation warrants experiments for testing if SP supply could stimulate nerve withdrawal in tendinopathy.

Figure 13.

(a–b) Immunofluorescence micrographs of longitudinal sections through the rupture site of the Achilles tendon of (a) substance P (SP) + captopril/thiorphan and (b) saline-treated rats at 3 week postrupture. In the SP treatment group (a), there was a significant reduction in the number of sensory, SP-positive nerve fibres (arrows). t, tendon tissue. Scale bar = 50 µm. Reproduced with permission from (83).

Physiological supply of neuropeptides

One study assessed the effect of exogenous neuropeptide administration on the healing of denervated ligaments in rats (Grorud et al. 2007). The aim was to restore the neuronal supply to improve the healing responses of the medial collateral ligament in joints partially denervated by femoral nerve transection or surgical sympathectomy. In this study, SP, VIP and NPY all improved the healing of denervated MCLs.

Interestingly, SP and VIP treatment resulted in mechanical properties, including failure load and failure stress, that were higher than those of intact, uninjured MCL (Grorud et al. 2007). The supply of SP, VIP and NPY also significantly improved the histological appearance of healing MCLs. Treated ligaments showed more organized scars and the appearance of increased matrix production. Interestingly, CGRP did not improve any of the measured histological or mechanical parameters of MCL healing in this model.

Effects of loading/unloading

Physical activity has been demonstrated to accelerate the neuronal plasticity in tendon repair (Bring et al. 2007). It has moreover been demonstrated that exercise leads to increased levels of various neuromediators and their receptors, including SP and CGRP, which may be involved in regulating the healing response (Jonsdottir 2000; Bring et al. 2009a,b; Ytteborg et al. 2012). However, excessive exercise has been associated with increased levels of SP/CGRP in this case being involved in the pathophysiology of tendinopathy (Backman et al. 2011).

Maybe the most fascinating effect of the mechano-neural transduction is the regulation of exercise via the neuromediator receptors (Figure 14). Thus, mRNA levels for the SP and CGRP receptors are in mobilized tendons significantly increased compared with immobilized controls at 17 days post-tendon injury (Bring et al. 2009a,b). It may prove that enhanced tendon repair after early loading is related to an increased peripheral sensitivity to sensory neuropeptide stimulation, suggesting an up-regulation of sensory neuropeptide receptors.

Figure 14.

(a,b) Normalized expression of mRNA for the SP (NK1) (a) and CGRP [CRLR (b) and RAMP-1 (c)] receptors in the healing area, in the Achilles tendon of rats subjected to two different levels of physical activity (freely mobilized versus plaster immobilized) at 8 and 17 days postrupture (mean ± SD). *P < 0.05; n.s. = P > 0.05. Between 8 and 17 days, there was an immense increase in the receptor expression in the mobile healing group, while the expression in the immobilized healing group fell back to levels comparable to the intact tendon control group. Reproduced with permission from (6).

Conclusion

Accumulating data demonstrates that neuronal regulation plays an essential role in tendon homoeostasis, repair and also pathology. This concept supports the idea that neuronal mediators effectuate crucial, but as yet incompletely defined roles in mechanically active connective tissues such as tendons.

Unbalanced occurrence of various neuromediators and their receptors as has been established experimentally and clinically in, for example, diabetes and tendinopathy, leads to a dysfunctional tendon homoeostasis resulting in chronic pain and gradual degeneration.

These novel findings of neuronal plasticity modulating tendon homoeostasis and capable of responding to dysregulation in pathology should stimulate the development of targeted pharmacological and tissue engineering approaches to improve healing and impair painful tendon disorders.

Funding source

Studies from the author's laboratories were supported by the regional agreement on medical training and clinical research (ALF) between Stockholm County Council and Karolinska Institutet (project nr. SLL 20110177), and the Swedish National Centre for Research in Sports, as well as the Swedish Medical Research Council (2012-3510) (PWA).

Conflict of interest

No competing financial interests exist.