The brain–joint axisin osteoarthritis: nerves, timers and beyond
Francis Berenbaum1 andQing-Jun Meng2
1Rheumatology Department, Sorbonnes Universités UPMC Univ Paris 06, INSERM UMRS_938, DHU i2B, Saint-Antoine Hospital, AP-HP, Paris, France.
2Faculty of Life Sciences, University of Manchester, M13 9PT, Manchester, UK.
Correspondence to F.B. and Q.-J.M.
[Au: main editing points: some abbreviations have been left expanded in the article to improve readability, and some changes have been made to the headings to fit heading length limits; please include, where possible, whether studies are in animal models/in vitro rather than human studies (queries or edits have been added); some statements in the text are not adequately referenced and reference 102 is missing (queries have been added)]
[Au: The abstract should summarize the main points of the article in 200 words or less. Please add several sentences to the abstract toinclude these points, emphasizing the points specifically on the topic of the review because OA is not mentioned much at present in the abstract]
Abstract | Osteoarthritis (OA) is a prevalent and debilitating joint disease for which ageing, obesity and chronic inflammation are known risk factors. The central, peripheral and autonomic[Au: “autonomic” OK?]nervous systems play an essential role in all metabolic systems. [Au: please include the link to OA, if possible, did you mean “and emerging evidence suggests a role in OA”?]In the past few years, [Au: “In the past few years”,We avoid the use of ‘recent’OK?]metabolic diseases, such as obesity or diabetes, have been linked to disruption of circadian rhythms that are tightly regulated by the nervous system, whereas inflammatory and autoimmune diseases are known to be linked to disruption ofthe cholinergic vagus nerve reflex.[Au: edit ok? Please include the link to OA, if possible][Au: the next sentence has been deleted because it seems to repeat the points made above]This article reviews currentknowledge of the direct and indirect roles of the nervoussystem and circadian system in the initiation and/or progression of OA, and highlights the directions for future research in this emerging field.[Au: edit ok?]
[Au: For your information, H1, H2 refer to the level of headings and will be removed before proofs are made. H1 subheadings can have max 38 characters inc spaces. H2 subheads can have max 39 characters inc spaces. Subheads have been edited to fit these limits, where indicated]
Introduction [H1]
Osteoarthritis (OA) is the leading cause of patient disease burden among all noncommunicable diseases; however, no curative drugs or even drugs that delay the destructive joint process are currently available. [Au: edit OK?]Hip and knee OA was ranked the eleventh highest contributor to global disability from 291 conditions in the Global Burden of Disease 2010 Study.1,2The two main risk factors for OA, ageing and obesity, are increasing exponentially in the worldwide population, and the cost of OA for society today is so high that an urgent need exists to stem the epidemic.[Au: edit OK?]
The challenge of finding new treatments in OA mainly results from a lack of knowledge of the pathophysiology of the disease.3 OA has long been viewed as the sole consequence of ‘wear and tear’ on cartilage together with the incapacity of cartilage to repair itself. [Au: edit OK?]However, we now know that OA is a disease of the whole joint, involving not only cartilage but also the synovial tissue, tendons, ligaments and the subchondral bone.4 Interestingly, whereas articular cartilage is not innervated and vascularized, the synovial tissue and the subchondral bone are. Moreover,OA was considered a local disease influenced solely by mechanical processes, including acquired OA resulting from overload caused by obesity or joint trauma, or congenital OA resulting from defects such as dysplasia or limb deformity.[Au: edit OK?]
[Au: Paragraphs over ~150 words can be difficult to read in the printed version, so I have split this one here and other long paragraphs in the text; is this OK?]
The long-held view of OA is that local production of inflammatory mediators found in OA synovial fluid is the sole consequence of a foreign-body-like reaction of the synovial tissue to cartilage fragments released in the joint cavity during the destructive OA process of the matrix. [Au: edit OK?]These mechanical and inflammatory local factors are critical for the initiation and progression of OA; however, studies in the past decade [Au: “past decade”OK?]have shown that systemic factors belonging to hormones or to cytokine families have additional rolesin the local destructive process, withthe particular roles varying according to the phenotype of the disease.4[Au: edit OK?]For example, low-grade systemic inflammation involving adipose tissue and insulin resistance is frequently observed in unhealthy obese patients with metabolic disorders (diabetes, hyperlipidaemia and hypertension).5,6 Interestingly, obesity is a risk factor for hand OA, which is probably related to the role of circulating adipokines in both hand OA and obesity.7[Au: is this what you meant?]Moreover, the presence of metabolic disorders increases the risk of OA in obese patients, strengthening the hypothesis that systemic factors play an additive or synergic role in the destructive process of OA.8
Dysregulation of microbiota induced by obesity has also been suggested to play a part in the pathogenesis of OA. This dysregulation causes an increased concentration of systemic lipopolysaccharides,[Au: “lipopolysaccharides” OK?]leading eventually to the activation of joint cells.9Finally, the release into the circulation of cytokines produced locally by OA joint cells could contribute to the pathogenesis of OA at a distance,by affecting tissues such as the brain.10,11[Au: is this what you meant?]
Together with obesity, ageing is a major risk factor for OA. Enrichment of cartilage matrix with advancedglycation endproducts may explain in part the cartilage fragility induced by ageing.12 In parallel, modifications in chondrocyte metabolism occur. Senescent chondrocytes become more sensitive to cytokines, leading to increased synthesis of cytokine-induced metalloproteinases.13[Au: do you mean matrix metalloproteinases?]Moreover, the chronic low-grade inflammation that develops with age may further compromise joint tissue homeostasis.14
Literaturein the past decade [Au: “past decade” OK?]has highlighted a critical role of the central and peripheral nervous system, including the circadian clock system,in maintainingrhythmic behaviour (for example, sleep–wake cycles) [Au: It is house style to replace solidi unless used as and/or]and daily metabolism (such as weight, energy expenditure, blood glucose concentration and blood pressure). Joint tissue is highly mechanically regulated, [Au: “regulated” OK?]going through daily phases of activity and inactivity controlled by 24 h rhythms generated in the brain.Age-related decline in brain function, including that of the circadian system, could affect the homeostasis of joint tissues. [Au: We prefer to only use ‘significant’ in a statistical context, edit ok?]
In view of the potential links between OA and metabolic disorders, and between metabolic disorders and the nervous system and circadian system (FIG. 1), this article reviews current knowledge of the direct and indirect roles of the nervous system and circadian systemin the initiation and/or progression of OA, and highlight the direction of future research on this emerging topic.[Au: edit OK?]
Overview of brain–periphery circuits [H1][Au: Edit OK? Maximum character count for Level 1 heading: 38 characters incl. spaces.]
The autonomic nervous system [H2][Au: “autonomic” OK?]
The nervous system[Au: “nervous system” has been left unabbreviated in the text as excessive abbreviation can reduce readability, OK?]consists of two main parts: the central nervous system (CNS), including the brain and the spinal cord, and the peripheral autonomicnervous system (ANS), including sympathetic and parasympathetic nerves. Some parts of the nervous system are more focused on communications with peripheral organs and tissues, such as the hypothalamus and autonomic nerves. [Au: edit OK?]
Epinephrine and norepinephrine (adrenaline and noradrenaline), [Au: edit OK?]themain sympathetic neurotransmitters, provokean increase in heart rate and blood pressure, and decrease digestive function by interacting with adrenergic receptors. By contrast, acetylcholine, the principal parasympathetic neurotransmitter,interacts with muscarinicacetylcholine receptors leading to the opposite effects. The physiological function of an organ thus depends on a sophisticated system that integrates opposing signals.
[Au: the paragraph below seems to relate to reference 40, and some of the text is similar to text in the section “Deregulation in the ANS”. Consider briefly introducing the topic in the paragraph below (with relevant reference/s but without the citation to Fig 2), but providing the detail in the section on Deregulation in the ANS together with the Fig citation (this would require renumbering Figs 2 and 3). Please renumber the relevant references in the list if necessary]
Interestingly, along with these well-known autonomic reflex arcs, an inflammatory reflex involving the vagus nerve has been described (FIG.2). [Au: Please include a reference for this statement]It consists of an afferent sensory neural arc that detects the chemical products of injury, infection and inflammation, and an efferent motor neural arc that transmits signals to modulate immune responses. The binding of acetylcholine to subunit α7 of the nicotinic acetylcholine receptor present at the surface of immune cells leads to the suppression of NF-κB activation and innate immune responses.[Au: Please include a reference for this statement]
Circadian clocks [H2]
In order to regulate sleep and activity in a 24h period, a masterclock, influenced by lightness and darkness from sunrise to sunset and located in the suprachiasmatic nucleus (SCN) of the hypothalamus, delivers rhythmic signals to brain regions influencing locomotor activity, feeding behaviour, hormone secretion and body temperature.15,16 The various rhythmic biological processes are driven by the expression of clock genes. The transcription and repression of these clock genes are induced by a transcriptional–translational feedback loop.17,18This loop consists of a positive arm (involving the transcription factors CLOCK and BMAL1, also known as aryl hydrocarbon receptor nuclear translocator-like protein 1)[Au: edit ok? I added the recommended long name for BMAL1 from OK?], a negative arm (involving the period and cryptochrome proteins) and an auxiliary stabilizing loop (involving the nuclear receptor subfamilies REV-ERBs and RORs). [Au: is this what you meant?]Most of these clock factors are transcription factors or nuclear hormone receptors with a limited half-life that enables their daily oscillation and turnover. In addition to the self-perpetuating regulation of the clock genes, components of the molecular circadian clock control various clock-controlled genes through response elements in their promoters or post-translational mechanisms.
Interestingly, the role of this central pacemaker has tremendously expanded in the past few years due to the discovery of secondary self-sustained autonomous circadian clocks present in the peripheral tissues.19These circadian clocks need to be reset by the central masterclock on a dailybasis in order to maintain synchrony. By means of humoral and neural circuits, the central masterclock exerts tight temporal control of the peripheral circadian clocks (FIG.3).20,21 For instance, glucocorticoid signalling is rhythmically controlled by the SCN and can act as a systemic entrainment factor for peripheral clocks.20,22As a demonstration of this fact, circadian rhythms in peripheral tissues of adrenalectomized rats demonstrate phase changes and desynchrony.21[Au: Edit OK?]Moreover, other time cues than daylight, such as feeding–fastingcycles [Au: ok?]can influence the rhythmicity of the masterclock,explaining how intrinsic and extrinsic factors can modulate our metabolism.23–26 Daily feeding signals can strongly entrain peripheral clocks in certain tissues, such as the liver and colon.24,25 Core body temperature rhythm has also been suggested as a potential synchronizing factor for peripheral tissues and cells (but not for the SCN).27
Of note, the cholinergic anti-inflammatory reflex arc has been linked to circadian rhythm regulation. Indeed, circadian rhythmicity exists in the cholinergic system, characterized by high acetylcholine release and low activity of the acetylcholine degrading enzyme acetylcholinesterase during the active phase of an individual.28[Au: please rephrase “the active phase of an individual” to clarify the meaning, do you mean “during physical activity”? Or “while an individual is awake”?]Furthermore, [Au: Edit OK?]cholinergic stimuli are potent regulators of the master clock in the SCN.29,30
TheHPA axis [H2]
The hypothalamic–pituitary–adrenal (HPA) axisis the term used to describethe physiological interactions that occur between three endocrine glands, thehypothalamus, thepituitary gland(located below the hypothalamus) and theadrenalglands. [Au: Edit OK?]This axis controls many physiological processes including digestion, immune reactions, energy storage and expenditure, and circadian timing. The circadian system and the HPA axis communicate with each other at multiple signalling levels.31,32The HPA axis has critical roles in transmitting rhythmic signals generated by the SCN to the rest of the brain and body. Activated by corticotropin-releasing hormone,a hormone secreted by the hypothalamus under the control of the circadian clock, the pituitary gland releases adrenocorticotropic hormone into the bloodstream, which induces the production of cortisol by the adrenal glands. Consequently, the diurnal profile of cortisol helps entrain peripheral tissue clocks. The HPA axis can also be activated by circulating cytokines such as IL-1, IL-6 or TNF,[Au: Journal style is to use Uniprot approved nomenclature for protein names (TNF rather than TNFα), ok?]leading to a physiological negative feedback loop created by the immunomodulatory and anti-inflammatory properties of cortisol.33,34[Au: Edit OK?]Importantly, cortisol and cytokines both have strong circadian profiles, supporting an integral crosstalk between the circadian timing system and the HPA axis.
Thehypothalamic–pituitary–thyroid axis [H2]
Thyroid hormones [Au: “thyroid hormones” left expanded in the text, OK?]physiologically act on epiphyseal chondrocytes by stimulating differentiation and inhibiting proliferation. In the past decade, [Au: Edit OK?]a role of thyroid hormones in the pathophysiology of OA has been demonstrated.35[Au: please briefly expand to explain the role of thyroid hormones in OA]The hypothalamus senses circulating levels of thyroid hormones and responds by regulating the release of thyrotropin-releasing hormone. [Au: is this point linked to the role of thyroid hormones in OA? Please rephrase to link the sentence to the topic or delete the sentence]It is worth noting that the activity of the hypothalamic–pituitary–thyroid axis has long been known to show diurnal variations. More importantly, parathyroid hormone (PTH) regulates the expression of clock genes in growth plate chondrocytes, although its role in articular cartilage remains largely unknown.36
How could the NSand circadian system influence the OA process at the joint tissue level?[H1][Au: Please shorten the heading to 38 characters including spaces]
Deregulation in the ANS [H2]
Disruption of the anti-inflammatory vagal reflex has been shown in manyautoimmune and inflammatory disorderssuch as rheumatoid arthritis, inflammatory bowel diseaseand pancreatitis.37–39In the past few years, Liu et al. havesuggested that an alteration in this inflammatory reflex could also play a part in OA (FIG.2). [Au: please cite reference 40 at the end of this sentence]In a rat model, [Au: Edit OK?]the researchers show that articular chondrocytes expresssubunit α7 of the nicotinic acetylcholine receptor,and its stimulation with nicotine[Au: “nicotine” OK?]inhibits IL-1-induced MAPK and NF-kB activation.40Moreover, intraperitoneal injections of nicotine [Au: “nicotine” OK?] prevented cartilage degradation induced by intra-articular injection of monoiodoacetate, an experimental model of arthritis similar to OA. Further studies in OA animal models are warranted to establish a more direct role [Au: please rephrase and expand the phrase “a more direct role” to clarify what you mean]of this inflammatory reflex.
The role of sympathetic nerve fibres present in synovial tissue, the subchondral bone marrow and the meniscus remains mysterious. However, in vitro and ex vivo studies show that ANS-related neurotransmitters can influence chondrogenesis. [Au: are you referring to some of the references 41–46, if so, please cite the references at the end of this sentence]A few studies have looked at the role of adrenergic receptors present at the surface of chondrocytes, including in human OA cartilage. [Au: please cite the relevant studies at the end of this sentence, e.g. some of the references 41–46?]Indeed, 2 adrenergic receptors and 2 adrenergic receptors[Au: “adrenergic receptors” has been left unabbreviated, OK?]are both expressed on epiphyseal proliferatingchondrocytes at different developmental stages.41–46[Au: please clarify what you mean by “developmental stages” do you mean in different stages of disease development or development of an organism (if so, please include if you mean in humans or animal models)?]Epinephrine [Au: ok?]inhibits, and β[Au: “β2”?]adrenergic receptor antagonistsrestore, the maturation process of growth plate chondrocytes with downregulation ofthe expression of type II collagen and Sox6.42,47[Au: Please rephrase the sentence to clarify whether epinephrine or β adrenergic receptors downregulate the expression of type II collagen and Sox6]Contradictoryresults have been reported showing either an upregulation or a downregulation of type X collagen expression with no effect on type II collagen after activation of β2 adrenergic receptors.44,45Interestingly, binding of norepinephrine on β [Au: “β2”?]adrenergic receptors could also impact epiphyseal chondrocyte survival by decreasing the apoptosis rate.41 Moreover, by use of mesenchymal stem cells and cartilage progenitor cells, Jenei-Lanzl et al.have demonstrated that norepinephrine can inhibit chondrogenesis and accelerate hypertrophic differentiation.48Collectively, these studies point towards a tentative role of the ANS in the regulation of OA-related pathways, although unequivocal evidence is still lacking.
Desynchronization of joint clock genes [H2][Au: Edit OK? Maximum character count for Level 2 heading: 39 characters incl. spaces.]
The joint tissues (cartilage, tendons and subchondral bone) are particularly influenced by daily rest–activity cycles, which are controlled indirectly by the central brain clock (FIG.3). Coupling of local gene expression and physiology to daily cycles of loading–unloadingof joints and related metabolic changes could be an important part of joint biology and homeostasis.[Au: Edit OK?]
As in many other tissues, cartilage homeostasis is under the control of a peripheral circadian clock. A circadian time series transcriptomic analysis of mouse chondrocytes(from xiphoid cartilage) has revealed 615 (3.9%) cartilage genes with a circadian pattern of expression. [Au: Please include a reference for this statement]We have shown that this number is an underestimate for articular cartilage [Au: in mice or humans?]that undergoes daily mechanical challenges (insert author name, unpublished work). [Au: please include an author name in the brackets]The rhythmic clock-controlled genes include those involved in extracellular matrixremodelling, apoptosis and oxidative stress pathways.49,50 Diurnal variations in serum levels of cartilage oligomeric matrix protein,[Au: OK?]hyaluronic acid, keratan sulphate, aggrecan, collagen type II and TGFhave also been reported.51,52