Targeting Inflammation in Rotator Cuff Tendon Degeneration and Repair

Abstract

Rotator cuff degeneration is a common affliction that results in pain and disability. Tendinopathy was historically classified with or without the involvement of the immune system. However, technological advancements in screening have shown that the immune system is both present and active in all forms of tendinopathy. During injury and healing, the coordinated effort of numerous immune cell populations work with the resident stromal cells to break down damaged tissues and stimulate remodeling. These cells deploy a wide array of tools, including phagocytosis, enzyme secretion, and chemotactic gradients to direct these processes. Yet, there remains a knowledge gap in our understanding of the sequence of critical events and regulatory factors that mediate this is process in injury and healing. Furthermore, current treatments do not specifically target inflammation at the molecular level. Typical regimens include non-steroidal anti-inflammatory drugs or corticosteroids; however, researchers have found irrevocable functional deficits following treatment, and have disputed their long-term efficacy. Therefore, developing therapeutics that specifically consider the nuances of the immune system are necessary to improve patient outcomes.

1 Introduction

Over 20% of all consultations made to a general practitioner are related to musculoskeletal diseases and 30% of these are associated with tendon injuries.^1,2 In the shoulder, the rotator cuff is a frequent source of pain and disability. Furthermore, the incidence of injury increases with age: over 50% of the population over the age of 65 has a rotator cuff tear.^3-5 Loss of strength and function due to rotator cuff tears results in disability and leads to lost days from work, occupational challenges, and recreational limitations for patients.⁶ Rotator cuff repair to recover shoulder function is one of the most common orthopedic surgical procedures, with over 250,000 repairs performed each year in the United States.^6-8 It is the most common shoulder condition, with more than 17 million individuals in the United States affected. Healing after rotator cuff repair is a well-known clinical challenge. Tears occur at the tendon-to-bone insertion site, and the goal of rotator cuff repair is anatomic restoration of the tendon attachment. However, clinical studies have shown failure rates ranging from 20-94%.^9-12 Factors associated with failure include tear size, chronicity, patient age, and other environmental factors.^5,11,13 Understanding the process of rotator cuff degeneration will allow clinicians to implement preventative early interventions.

Prior to the 1990s, tendon injuries were often classified as tendinitis, with the “itis” suffix resulting in a misleading emphasis on inflammation.¹⁴ However, even as early as the 1970s, some research groups presented a lack of acute inflammatory cells in chronic tendon injuries.¹⁵ More recently, Khan et al. reported an increase in the presence of fibroblastic and myofibroblastic cells accompanied by a distinct lack of inflammatory cells in tendons of athletes with overuse injuries.¹⁶ Khan advocated for the use of the term “tendinopathy” to refer to overuse injuries which were previously considered tendinitis. Researchers in the last two decades have advocated the separate use of the terms “tendinitis” and “tendinosis” to describe tendon injuries with inflammation and without inflammation, respectively. Lost in the nomenclature debate and classification of inflammatory vs. non-inflammatory responses has been an appreciation of a spectrum of inflammatory responses across the various tendon pathologies.

As immunohistochemistry and gene expression analysis techniques have improved, mounting evidence supports some level of inflammation in all tendon injuries.^17,18 For example, improved histological techniques were used to show the presence of B and T lymphocytes, suggesting the presence of inflammation, in chronic Achilles tendon injury.¹⁹ Tenocytes became more metabolically active in response to inflammatory cytokines and increased in number and size in pathologic tendon.²⁰ A systematic review of the literature in 2015 suggested that increased numbers of inflammatory cells are present in pathological tendons.²¹ Therefore, inflammation plays a role in the development and progression of tendon injuries, and particularly during an injured tendon's healing response.

Despite the evidence that inflammation is involved in tendinopathy, there is a lack of consensus regarding inflammation-targeting treatments. There is an emerging trend among researchers and clinicians supporting the use of biological treatment coupled with surgical intervention to improve patient outcomes. The need for targeted therapeutics in tendon healing and degeneration, with known mechanisms that effect particular cells, has become clear.^2,22 Specifically, there is a need for more data on cellular and molecular aspects of tendon development, signal transduction, mechanotransduction, and basic mechanisms underlying tendon degeneration and healing. This review will first briefly describe our current understanding of tendon degeneration and treatment approaches and then focus on inflammation-related strategies to treat tendinopathies.

2 Pathogenesis of Rotator Cuff Degeneration and Current Treatment Approaches

Rotator cuff injuries can occur due to years of degeneration (e.g., associated with overuse), sports (e.g., due to collision), accident (e.g., due to a laceration), and/or disease. Both intrinsic and extrinsic factors contribute to tendon injury.²³ Intrinsic factors for chronic and acute injury include age, gender, obesity, type 2 diabetes, and genetics. With regard to age, the incidence of rotator cuff repairs increased by 238% from 1995 to 2009 in New York, in part due to the increase in the percentage of the population between 45-65 years of age rising from 53.0 to 62.4%.²⁴ With regard to genetics, six single nucleotide polymorphisms (SNP) of the TNC gene were recently associated with full-thickness rotator cuff tears.²⁵ In particular, a missense SNP in exon 10 varied the amino acid sequence close to the TNF-FNIII5 domain, which binds with numerous growth factors during tendon healing.²⁶ Variants of SAP30BP, SASH1, and ESRRB were also associated with full-thickness tears.^27-29 These genes are part of apoptosis pathways, and an increase in apoptotic cells are found in torn rotator cuffs, which is also associated with increases in pro-inflammatory cytokines.^30,31 Rotator cuff disease defined by full- or partial-thickness tears or tendinosis was associated with SNPs of DEFB1, DENND2C, ESRRB, FGF3, FGF10, and FGFR1 genes, all of which are implicated in tissue degenerative and repair processes, including inflammatory cascades.³²

Extrinsic factors leading to injury differ between chronic and acute injury. These factors include exercise regimens, occupation, nutrition, medication, and smoking.^23,33,34 Chronic injuries are associated with repetitive mechanical strain and loading.³⁵ Acute injuries are associated with sudden eccentric movement that causes the tendon to tear or rupture from the bone.³⁶ A common thread that runs through many of the risk factors for rotator cuff degeneration is inflammation. Aging, obesity, type 2 diabetes, and smoking are all associated with increases in circulating inflammatory factors which are believed to a contribute to a state of low-grade chronic inflammation leading to various diseases.³⁷ More researchers are attempting to close this potential loop with tendinopathy and explore the role of the immune system during degeneration and healing.¹⁸

The high prevalence of tendon injuries has influenced the development of various treatment strategies. Surgical repair is necessary to re-attach tendon ends in cases of mid-substance tears or lacerations and tendon-to-bone ruptures. However, most surgical treatments have shown only moderate success rates.^38,39 Patients may experience pain relief and demonstrate improved outcomes in clinical assessments, but the surgical repair may only deliver a short-term solution.³⁸ The poor surgical failure rates can partially be attributed to a variety of patient factors, including patient age, size of injury, chronicity of injury, systemic diseases, diabetes, and smoking. Surgical technique has improved along with our understanding of tendon mechanics, but an inherent difficulty lies in recreating the complex tendon-to-bone attachment or the structure of tendon midsubstance following injury.

More conservative non-surgical interventions for tendon injuries have shown mixed success.² Rest alone does not heal chronic tendinopathies, but can help manage the pain associated with the injury and may prevent re-injury. Eccentric exercise (EE) therapy has shown some success for Achilles tendinopathy, but less so for tendinopathy at other sites.⁴⁰ Extracorporeal shock wave (ESW) therapy for chronic tendon injuries is an attractive non-surgical treatment option which showed some success in early clinical use, but its mechanism of action still remains to be defined.⁴¹ Finally, the use of platelet rich plasma (PRP) injections to treat tendinopathies has shown very limited success.^18,42 The patient's age, demographics, pain and risk tolerance help clinicians decide whether or not non-surgical intervention should be pursued. Eventually, if the tendon injury progresses and/or patient discomfort increases, surgical intervention is necessary.

3 Immune cell types and signaling pathways involved in tendon disease and healing

3.1.1 Immune cells involved in the inflammatory response

Immune cells are present in pathologic tendons due to the generation of alarmins, an endogenous set of biomolecules which can recruit and activate cells of the innate and adaptive immune systems.⁴³ There are two main types of inflammatory responses – acute and chronic. Acute inflammation is the initial response to harmful stimuli and mediated primarily by innate immune cells such as mast cells, macrophages, and neutrophils. Their general purpose is to eliminate or minimize additional damage from the stimuli, remove dying cells and damaged tissue, and initiate tissue repair. Chronic inflammation has a delayed onset and is mediated primarily by mononuclear cells like macrophages and lymphocytes.

Within the innate cell population, mast cells increase in density during the early stages of the disease and can persist in patients with longstanding symptoms.^44-46 These pro-inflammatory cells secrete immune mediators that recruit other immune cells and suppress collagen synthesis of tendon fibroblasts via prostaglandin E2 signaling.⁴⁷ Macrophages are also present during all stages of disease, however, their functionality changes as they guide the healing process towards resolution. Initial macrophage infiltrate and resident expansion adopt an M1 pro-inflammatory phenotype.^48-50 In contrast, late stage healing of tendon is marked by a phenotypic switch to anti-inflammatory M2 macrophages. ^48,51 However, this switch may not occur in non-resolving patients, leading to chronic degeneration and pain.^22,48 Other innate immune cells that are known to be present during tendon pathology include neutrophils, dendritic cells, and natural killer cells; however, their specific roles remain undefined in tendinopathy.⁵²

Adaptive immune cells such B and various T cells are also known to be present during tendinopathy and in healing, but little is known regarding their roles in these processes.^19,20 Importantly, lymphocytes are extremely diverse and few studies have screened for specific markers in the rotator cuff. However, several unique subpopulations of T cells residing within musculoskeletal tissues have been identified. Exploring the mechanisms of spondyloarthropathy led to the discovery of a tendon enthesis-specific population of T cells that are ROR-γt+ CD3+ CD4− CD8− and respond to interleukin-23, producing downstream inflammatory mediators.⁵³ Specifically, these cells secrete interleukin-22 which activates signal transducer and activator of transcription 3-dependent osteoblast-mediated bone remodeling, resulting in osteophyte formation. Skeletal muscle specific regulatory T cells were found to mediate tissue repair processes following injury by reducing late stage inflammation and secreting the growth factor amphiregulin, which acts directly on satellite cells.⁵⁴ Studies like these shed light on the phenotypic heterogeneity within immune cell populations that are driven by their unique niches within tissues. Furthermore, their specificity to the interface between tendon and bone is particularly relevant to rotator cuff repair, which relies on re-integration of the two tissues for functional success.

3.2 Signaling pathways and cytokines involved in the inflammatory response

The inflammatory response can be modulated by targeting components of the nuclear factor-κ-B (NF-κB) pathway (Figure 1). NF-κB is an evolutionarily conserved transcription factor that plays a role in multiple biological processes, but most notably in the development and function of the immune system.^55,56 The transcription factor also has a significant role in the development and function of the epithelium and skeletal system and has been linked to a variety of disease states.^57,58 Most of the disease states are related to chronic inflammatory and autoimmune diseases, where sustained NF-κB signaling leads to a pathophysiology. The pathway's capacity to link “physiology to pathology” has increased the number of research studies and publications describing the mechanism of the pathway.

Figure 1.

An inflammatory stimulus activates the NF-κB pathway, leading to recruitment of the IKK complex, phosphorylation of IκB, and translocation of NF-κB subunits – p50 and p65 – into the nucleus.

Activation of the NF-κB pathway is described through two pathways – canonical (classical) and non-canonical (alternative).⁵⁹ The canonical pathway is activated by ligands binding to receptors for TNF or IL-1, pattern recognition receptors, or antigen receptors and is marked by activation of the IκB kinase (IKK) complex. These receptors are present in resident tendon fibroblasts and expression of the complimentary cytokines are increased in tendinopathy.^60-62 Recently, intereukin-33 (IL-33) was found to stimulate canonical NF-κB signaling via ST2 in early rotator cuff tendinopathy (Figure 2).⁶³ This resulted in suppression of micro-RNA-29a in tendon cells and subsequent production of scar-forming type III (Figure 2). The non-canonical pathway is induced by specific members of the TNF cytokine family such as CD40 ligand, BAFF, and lymphotoxin-β.^64,65 The NF-κB inducing kinsase (NIK) stimulates IKKα activation (independent of NEMO) leading to phosphorylation of p100 and the generation of RelB/p52 complexes. It is has not yet been shown if non-canonical NF-κB signaling has any role in rotator cuff tendinopathy.

Figure 2.

In early rotator cuff tendinopathy interleukin-33 (IL-33) activates NF-κB signaling via ST2/IL-1RA which suppresses micro-RNA-29a (miR-29a) expression resulting in production of type III collagen and soluble ST2 which competitively binds to circulating IL-33, potentially acting as a protective feedback mechanism. (Figure reprinted from Millar et al⁶³ and is distributed openly under Creative Commons Attribution 4.0 International Public License)

One downstream effect of NF-κB signaling is the production of the cytokine interleukin-6 (IL-6). Classically, IL-6 is considered pro-inflammatory, as it stimulates T-cell and macrophage activation and osteoclast formation. However, it is now known to also be an anti-inflammatory myokine that responds to physical exercise and promotes secretion of inhibitory factors such as interleukin-1 receptor antagonist (IL-1ra) and interleukin-10 (IL-10).⁶⁶ The bias towards pro-inflammatory behavior has been shown to be tumor necrosis factor alpha (TNFα) dependent.⁶⁷ In patients with rotator cuff tears, IL-6 gene expression and cytokine production have been shown to correlate with the extent of degeneration and genes associated with tissue remodeling.^68-71 One study examined patients following rotator cuff repair surgery and found increased joint stiffness was associated with polymorphisms of the IL-6 gene.⁷²

Another signaling downstream cytokine implicated in rotator cuff disease is interleukin 17A (IL-17A). This cytokine is known to be secreted from various immune cells, and binding of its receptor results in the production of pro-inflammatory cytokines, chemokines, inducible nitric oxide synthase, and MMPs. Modulation of IL-17A is dependent on synergistic activation of interleukin-23, STAT3, and NF-κB or TGF-β and IL-6. Patients with early rotator cuff tendinopathy show increased expression of IL-17A. Furthermore, culturing tenocytes with rhIL-17A induced production of proinflammatory cytokines and increased expression of apoptosis related genes.⁷³

4 Therapeutic Potential of Targeting Inflammation

4.1 Broad approaches

Some clinicians use non-steroidal anti-inflammatory drugs (NSAIDs) and corticosteroids to treat early tendon injuries. However, most researchers argue against the use of NSAIDs for tendinopathy.⁷⁴ NSAIDs block the production of prostaglandins from arachidonic acid, which leads to an overall decrease in the inflammatory response. However, too much arachidonic acid could also lead to tissue damage. Therefore, NSAIDs are believed to have a positive effect on patient symptoms, but may be deleterious to tissue and tendon healing in the long term. Similarly, the use of corticosteroids offers some short term pain relief, but has limited long-term benefits for tendon healing.⁷⁵ Corticosteroids inhibit the accumulation of neutrophils and synthesis of inflammatory cytokines. They have been shown to decrease pain and improve abduction in rotator cuff tendon injury, but not in Achilles tendon injury.^76,77 There is concern about an association with increased risk of tendon rupture with the frequent use of corticosteroid injections.⁷⁸ NSAIDs and corticosteroids are the predominant anti-inflammatory options available to clinicians today. There is a need for the development of therapeutics that could provide finer control and mediation of inflammation in tendon healing.

4.2 MSC-mediated modulation of inflammation

The complex mechanisms of inflammation and tissue repair make it difficult to identify a single target for rotator cuff degeneration and tendinopathy. However, several treatment modalities have been attempted with various degrees of therapeutically-defined mechanisms. Prior in vivo and in vitro studies have suggested that the application of mesenchymal stromal cells (MSCs) may improve the healing response via their capacity to modulate the inflammatory environment.^79,80 MSCs influence local tissue environments through the secretion of soluble factors and exert protective effects on local cells. This aspect of MSC function has been exploited to enhance tissue repair in animals.^81,82 In a recent study, ASCs co-cultured with M1 (pro-inflammatory) macrophages successfully suppressed the effects of M1 macrophages on tendon fibroblasts by inducing a phenotypic switch from a pro-inflammatory macrophage phenotype to an anti-inflammatory macrophage phenotype.⁸³ This led to decreased exposure of tendon fibroblasts to pro-inflammatory cytokines such as IL-1β and increased production of extracellular matrix proteins. ASCs, delivered in a cell sheet to a tendon repair site in an in vivo animal model, also modulated the inflammatory response.⁸⁴ When examining the effect of ASCs on macrophage phenotype, gene expression and immunohistochemistry results after 7 days of healing were consistent with the in vitro results. Similar results were reported in a small animal study using MSCs primed with the cytokine TNFα.⁸⁵ These studies suggest that, by controlling inflammation, ASCs may improve tendon healing outcomes.

4.3 Signaling Inhibition

A more granular approach to treatment is to target specific immune signaling pathways. NF-κB functions as a master regulator and is typically activated by TNF and IL-1 signaling cascades. Therefore, several studies have attempted to modulate these pathways to treat disease or improve healing. Inhibiting TNF is currently used as treatment for autoimmune diseases such as rheumatoid arthritis and ankylosing spondylitis.⁸⁶ Low-dose inhibition could improve tendinopathy without increasing the risk of infection.¹⁸ In a pilot study, peritendinous injection of adalimumab, a monoclonal TNFα antibody, was administered to the Achilles of symptomatic athletes.⁸⁷ Walking and resting pain was reduced over the 12 week course of treatment and tendon thickness remained constant, while blood flow decreased. In a rat rotator cuff repair model, blockade of TNFα using pegylated TNF receptor type 1 improved the biomechanical strength and reduced pro-inflammatory macrophages at early time points (< 4 weeks).⁸⁸ However, these effects normalized to the non-treated controls by 8 weeks. Knocking out TNFα using lenti-virus mediated siRNA in a rat model of carrageenan-induced subacromial bursitis muted the inflammatory response at 5 weeks and reduced fibro-cartilaginous metaplasia.⁸⁹

Inhibition of the IL-1 signaling pathway is also used to treat autoimmune diseases and has served as another therapeutic candidate for rotator cuff disease. In the aforementioned athlete study, the investigators also applied anakinra, a recombinant form of interleukin-1 receptor antagonist (IL-1ra), to treat Achilles tendinopathy.⁸⁷ However, it was not as effective as the TNFα inhibitor. Suppressing IL-1 did not significantly reduce pain or blood flow and increased tendon thickness after 12 weeks of treatment.⁸⁷ Recent animal studies have shown more promise for inhibiting IL-1. In a rabbit model of stress-shielding induced patellar tendinopathy, treatment with IL-1ra prevented mechanical deterioration.⁹⁰ Similarly, IL-1ra prevented collagen disruption and maintained matrix metalloproteinase-1 and -3 expression in stress-shielded rat Achilles tendons.⁹¹ Lastly, in carrageenan-induced rat patellar tendinopathy, IL-1ra maintained the tissue geometry and collagen activity relative to controls and reduced inflammatory cell infiltrate.⁹² Seemingly, no work focusing specifically on the rotator cuff has been attempted using this treatment axis. Although these results are encouraging, more work is needed to realize the therapeutic potential of this approach.

An alternative approach is to suppress extracellular matrix-degrading enzymes downstream of the inflammatory response. Along these lines, modulation of matrix metalloproteinases (MMPs) has been used in various musculoskeletal degenerative diseases as a treatment approach.⁹³ Doxycycline, an antibiotic that inhibits MMPs, was found to improve healing in rats following rotator cuff repair. The animals exhibited an increase in tissue strength, and this was associated with decreased levels of MMP-13 over a two week time period. However, using the same approach for Achilles tendons in rats, one group reported positive effects⁹⁴ while another group reported negative effects with treatment.⁹⁵ Another non-specific inhibitor of MMP activity, recombinant α-2-macroglobulin, appeared to improve regeneration of the tendon-to-bone interface following acute rotator cuff repair based on histologic outcomes.⁹⁶ The mechanism of inhibition using these approaches remains unclear, and the results have been limited to repair of acute injuries rather than the more clinically relevant chronic injuries. One limitation of focusing solely on a specific signaling pathway or cytokine as an axis of treatment ignores the spatial and temporal heterogeneity of the immune and stromal cell responses, and the cross-talk between them.

5 Conclusion

The inflammatory cascade during rotator cuff degeneration and after surgical repair is a dynamic process involving a complex set of cells and signaling pathways. Therefore, success of inflammation-focused treatment strategies hinges on understanding the temporal nuances of the immune-mediated program. Thorough phenotyping of tissue-resident and infiltrating cell populations in clinical samples can guide more mechanistic approaches, leveraging lineage-specific deletion of inflammatory signals in order to understand the mechanisms of rotator cuff disease. Furthermore, understanding how these factors change in response to an individual's environment, lifestyle, and genetics will advance patient care in the growing era of precision medicine.