Article Text


Reactive oxygen species and tendinopathy: do they matter?
  1. C S Bestwick1,
  2. N Maffulli2
  1. 1Phytochemical and Genomic Stability Group, Cellular Integrity Programme, Rowett Research Institute, Aberdeen, Scotland, UK
  2. 2Keele University School of Medicine, Trauma and Orthopaedics, Hartshill, UK
  1. Correspondence to:
 Professor Maffulli
 Keele University School of Medicine, Trauma and Orthopaedics, Thornburrow Drive, Hartshill ST4 7QB, UK;

Statistics from

Reactive oxygen species are probably involved in tendinopathy

We propose that a molecular link between the exaggerated dysfunctional repair response in overuse tendinopathies and the subsequent orchestration of effective tendon healing is the control of the production and persistence of reactive oxygen species within the intracellular and extracellular milieu of the tendon tissue. Reactive oxygen production and the ensuing cellular response can be strongly influenced by lifestyle factors such as the intensity and frequency of exercise.

“Reactive oxygen species” (ROS; also referred to as active oxygen species, AOS; reactive oxygen intermediates, ROI) is a collective term for both radical and non-radical but reactive species derived from oxygen. A free radical, is “any species capable of independent existence that contains one or more unpaired electrons”.1 The presence of such unpaired electron(s) often imparts considerable reactivity. Commonly detected and potentially physiologically relevant ROS include the superoxide anion, hydrogen peroxide (H2O2), the hydroxyl radical, singlet oxygen, and peroxyl radicals. A further and inter-related group are the reactive nitrogen species (RNS)—for example, peroxynitrite.1

ROS are continually produced during normal cell metabolism. The mitochondrial respiratory chain, NADPH-cytochrome P450 enzymes in the endoplasmic reticulum, phagocytic cells, lipoxygenase, and cyclo-oxygenase are also sources of basal ROS production.1 Trauma and environmental and physiological stimuli may enhance ROS production.1

Traditionally, ROS are viewed as imposing cellular/tissue damage through lipid peroxidation, protein modification, DNA strand cleavage, and oxidative base modification, although the relative reactivity and susceptibility of the molecular targets vary. Thus, ROS production is implicated in numerous aspects of pathophysiology including tumorigenesis, coronary heart disease, autoimmune disease, overuse exercise related damage to muscle, and impairment of fracture healing.1,2

This association with cellular damage and pathology has predisposed much of the literature to consider decreased ROS production de facto a universally desirable phenomenon. This, however, belies the complexity of ROS action, in which subtle changes in ROS type and concentration may exert profound effects on cell metabolism and development including proliferation, differentiation, and adaptive responses. At higher levels, ROS may initiate and/or execute the demise of the cell. The ability of H2O2 to diffuse across membranes imparts potential to exert effects at sites distant from its production. Thus ROS changes may have widespread consequences for cell function as well as integrity and viability.3,4


To our knowledge, there is a paucity of studies on ROS participation in clinically relevant models of tendinopathy. However, recent investigations show increased expression of peroxiredoxin 5, a thioredoxin peroxidase with antioxidant properties, in tendinopathic tendon, suggesting that oxidative stress may be involved in the pathogenesis of tendon degeneration.5 Raised ROS concentrations are proposed to contribute to the development of tendinopathy as a side effect of fluoroquinolone antibiotic use.6

What is the potential source(s) of ROS production in the tendon or its immediate vicinity? During cyclical loading of the tendon, the period of maximum tensile load is associated with ischaemia,7 and subsequent restoration of normal tissue oxygenation may enhance ROS production. Hyperthermia in the exercising tendon may stimulate ROS production, probably from the mitochondria.7 Fibroblasts also specifically generate ROS, through an NADPH oxidase complex, in response to cytokines and growth factors, the production and release of which are stimulated after tendon injury.8,9

A further possibility is that tendons are indirectly influenced by changes in ROS metabolism in other tissues and cells such as in exercising muscle. Resting muscles generate both intracellular and extracellular superoxide, the production of both being enhanced during contraction. In addition, although the extent of enhancement is contested,10 exhaustive exercise increases ROS generation by activated phagocytes. Although non-exhaustive exercise does not produce any consistent findings of oxidative damage, the inflammatory response may contribute to overtraining damage in muscle. This change in granulocyte activity may also have more general consequences for ROS concentrations in tissues other than skeletal muscle, possibly including the tendon, through collateral exposure to ROS or mediators/signals arising from their actions. Although there is no direct histological evidence of active inflammation associated with tendinopathic lesions,11 surgery is a late event in the management of tendinopathy, and cyclic stretching of human tenocytes increases the production of inflammatory mediators.12

Detection of ROS production and changes in ROS concentrations would, however, only represent a start in dissecting their role in tendinopathy, as any changes may be as much a part of healing as of tissue disruption. Studies on avian fibroblasts suggest that, during tendon healing, mechanical load and growth factors—for example, platelet derived growth factor (PDGF) and insulin-like growth factor I—operate in concert to stimulate tenocyte cell division.9 Interestingly, PDGF stimulation of rat vascular smooth muscle cells transiently increases intracellular H2O2 concentration, and H2O2 is required for PDGF signal transduction.13 Thus, in tendons, the pro-proliferative action of growth factors and mechanical load may be mediated through H2O2 production.

Chemotaxis of cells in the wounded tendon (micro-tear) may also be influenced by ROS/RNS generation. Proliferation and migration of vascular smooth muscle cells is inhibited by the H2O2 scavenger, catalase.14 However, balance and control of ROS exposure is critical to the final cell response. Heightened concentrations of H2O2 retard both proliferation and restitution in the gastric mucosa,15 and equine tenocytes show a decrease in proliferation when subjected to 10–100 μM H2O2.7

A recent intriguing observation is that extracorporeal shock wave therapy, which is reported to promote tendon repair and bone growth, induces increased production of superoxide anion, which mediates extracellular signal regulated kinase signal transduction during osteogenesis.16 Differentiation was not influenced by inhibition of H2O2, peroxynitrite, or nitric oxide production, suggesting the specific involvement of superoxide.16

Tenocyte numbers are increased in tendinopathic tendons, and this may be a factor in degeneration and also a prerequisite to healing.11 ROS may not only induce cell death, but also determine the mechanistic form of death, such as apoptosis or oncosis.17 Apoptosis is a highly regulated programme of cellular suicide, which is of critical importance to the regulation of cell number and genomic integrity. Evidence for the involvement of apoptosis in tendon pathology is gradually emerging. Degenerative joint disease of the knee, an age related condition, is associated with higher susceptibility of periarticular tenocytes to Fas ligand induced apoptosis.18 These changes may contribute to decreased cellularity in degenerative tendons and promote their rupturing. Apoptosis has also been detected in human tendinopathic tendons,17 and the increased number of apoptotic tendon cells in degenerative tendon tissue may affect the rate of collagen synthesis and repair.17

ROS (and RNS) are potent inducers and modifiers of the apoptotic process, but the relation is complex.1 For example, high concentrations of hydrogen peroxide can prevent apoptosis. Conversely, “bursts” of ROS and decreased antioxidant enzyme activity often accompany the induction of apoptosis, and oxidative stress is a common feature of the late phase of apoptosis. Recent work shows that oxidative stress induced apoptosis in human tenocytes involves the classical release of cytochrome c from mitochondria into the cytosol and activation of caspase-3 protease.19


Continued sublethal ROS exposure will not occur in a metabolically or genomically static system, and ROS exposure may induce an adaptive response in tissues.10,20,21 In the organism, adaptation seems to be cell type, and possibly antioxidant specific and age related effects on the development and composition of the antioxidant system will also need to be considered.22 It may be simplistic to extrapolate skeletal muscle adaptation to tendons. However, the effects of adaptation induced by ROS may result in changes in tenocyte ability to transpond physiological/environmental signals and resist stress arising from musculature, phagocyte, and endogenously derived ROS. Could a failure to have experienced enhanced ROS generation, possibly through avoidance of repetitive exercise/training, and hence an absence of adaptation, predispose tendons of the occasional exerciser to ROS damage during sudden exercise? Similarly, does excessive or unusual exercise by the trained athlete cause ROS exposure that exceeds the protective effect of any adaptive response achieved through training?


Tendinopathies have a complex aetiology, and we have not attempted, indeed with the current level of information we are not able, to specifically cite participation of ROS in tendon degeneration, failure, or healing. Nevertheless, recent research5,6,19 has provided intriguing glimpses of ROS participation in tendon pathology, and the possibility that such species influence the propensity for tendinopathic development and repair is surely one that merits further investigation.


We thank Professor JR Arthur and Dr GG Duthie for critically reading the manuscript. We gratefully acknowledge funding support from ESSKA djOrtho and SportsMed (UK) Ltd. The laboratory of CSB is funded by the Scottish Executive Environment & Rural Affairs Department (SEERAD).

Reactive oxygen species are probably involved in tendinopathy


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.