Circadian clocks in articular cartilage and bone: A compass in the desert of matrices

Nan Yang and Qing-Jun Meng*

Faculty of Biology, Medicine and Health, Wellcome Trust Centre for Cell Matrix Research, University of Manchester, M13 9PT, UK

*Corresponding author: Dr. Qing-Jun Meng (qing-jun.meng@manchester.ac.uk); Tel, +44 161 3068912

Running title, Clocks in cartilage and bone

Abstract

Temporally coordinated resorption and synthesis is the key to maintain healthy bones. Articular cartilage is a highly specialized connective tissue within the joints which lines the surface of a long bone. Emerging evidence has suggested a critical role of the circadian system in controlling cartilage and bone biology. Articular cartilage is sparsely populated with chondrocytes, surrounded by abundant extracellular matrices that are synthesized and maintained solely by chondrocytes. Once damaged, the articular cartilage tissue has poor capacity for endogenous repair, leaving the joints prone to osteoarthritis, an age-related painful condition that affects millions of individuals worldwide. An important question is how articular cartilage has evolved its remarkable capacity to maintain homeostasis and withstand daily biomechanical challenges associated with resting/activity cycles. Equally important is how this avascular and aneural tissue senses time and utilizes this information to coordinate daily phases of metabolic activity and tissue remodeling/repair. Bone tissue that has derived from cartilage has similarly sparse populations of resident cells living in dense and largely mineralized matrices. We will discuss recent progress on circadian clocks in these matrix-rich skeletal tissues, and highlight avenues for future research.

Cartilage and bone

Articular cartilage is a highly specialized connective tissue. It consists of a dense extracellular matrix (ECM) with sparsely populated chondrocytes, the only cell type in this tissue (Hardingham et al., 2002; Hardingham 2010). Articular cartilage provides a smooth and lubricated surface for joint movement, and its function is dependent on proteoglycan (e.g. aggrecan) and collagens (mainly collagen type II) in ECM (Asanbaeva et al., 2008). The biomechanical environment surrounding a chondrocyte in the articular cartilage is particularly harsh: high osmolarity, low oxygen, no innervation, no blood vessels, repeated compressive loading and strain. Chondrocytes live in isolation and rarely divide throughout adult life. Once damaged, this tissue is very difficult to repair (Poole et al., 1987). Excessive degradation of the ECM in articular cartilage is a common feature of two common musculoskeletal morbidities, osteoarthritis (OA, associated with old age) and rheumatoid arthritis (RA, an autoimmune condition) (Sophia Fox et al., 2009; Kumar et al., 2016). The role of circadian rhythms and RA has been extensively reviewed (Spies et al., 2014; Buttgereit et al., 2015). Therefore, we will focus on links between circadian clocks and OA.

Long bones derived from cartilage are essential for bearing the weight, supporting skeletal muscles, and regulating mineral balance (Feng and McDonald, 2011). Longitudinal long bone growth depends on the expansion and endochondral ossification of the growth plate. Of note, another type of bone formation, such as calvarial bone or bone fracture, is through intramembranous ossification which does not involve cartilage. The growth plate is a cartilaginous tissue found in children and adolescents, and replaced by bone in adult humans (Melrose et al., 2016). Bone homoeostasis depends on two opposing processes, resorption and synthesis. Normal bone remodelling requires strict temporal control of these two processes, regulated by osteoclasts and osteoblasts, to guarantee maintenance of the bone mass and structural quality (Feng and McDonald, 2011). The third cell type (and the most abundant in mature bone, >95% of all bone cells) is the highly mechano-sensitive osteocytes, which play critical signalling roles by communicating with osteoblasts and osteoclasts. In general, osteoclasts rapidly absorb the old mineralized bone, and then osteoblasts are able to slowly form the new bone (Sommerfeldt and Rubin, 2001; Downey and Siegel, 2006). One common feature of chondrocytes and bone cells is scarce cell numbers and abundant ECM. Therefore, it is critical to understand how these highly specialized cells integrate temporal, spatial and biomechanical cues to regulate their tissue homeostasis. One such cue is the circadian timing mechanism.

The circadian system

Daily 24 hr rhythms in physiology and behavior are generated by biochemical time pieces, known as circadian clocks (Reppert and Weaver, 2002). The circadian system (Fig.1) is orchestrated by a master clock in the suprachiasmatic nucleus (SCN) of the hypothalamus, which in turn controls secondary self-sustained autonomous circadian clocks present in most peripheral tissues (Yoo et al., 2004). The master clock is entrained by the light/dark cycle. The SCN uses this information to convey rhythmic time cues to other brain regions influencing locomotor activity, body temperature, hormone secretion and feeding behavior (Cermakian and Sassone-Corsi, 2000; Gamble et al. 2014), as well as to different body organs. Peripheral clocks contain identical clock genes/proteins as the SCN, and can be synchronized by external time cues other than the light/dark cycle (e.g. feeding/fasting, temperature cycles and physical exercise). Individual organs integrate intrinsic metabolic and biomechanical cues, as well as extrinsic light/dark cycle signals to generate and maintain robust cellular rhythms (Reppert and Weaver, 2002; Roenneberg and Merrow, 2005; Takahashi et al., 2008; Dibner et al., 2010).

At the molecular level, the mammalian circadian oscillator comprises a transcriptional-translational feedback loop. There are nearly a dozen core clock genes identified in mammals: Period (PER) family, Cryptochrome (CRY) family, BMAL1 (brain and muscle ARNT-like 1/Arntl), Clock (Circadian locomotor output cycles kaput), NPAS2, DEC1/DEC2 (Bhlhb2/Bhlhb3), REV-ERBα/β (NR1D1/ NR1D2; nuclear receptor subfamily), RORα/RORγ (retinoid acid receptor (RAR)-related orphan receptor), DBP/TEF/HLF and E4BP4 (NFIL3) (Reppert and Weaver, 2001; Schroeder and Colwell, 2013). Most of these clock factors are transcription factors with relatively short half-life, allowing periodic rise and fall of their levels within a 24 hr day (Roenneberg and Merrow, 2005; Takahashi et al., 2008). In the morning, transcription factors Clock and BMAL1 (the ‘positive arm’) stimulate the expression of Per and Cry by binding to the cis-regulatory element E-box within their promoters (Ko and Takahashi, 2006). Subsequently, the PER and CRY proteins (known as the ‘negative arm’) bind to Clock and BMAL1 and inhibit their transcriptional activation. This eventually reduces

the expression of Per and Cry, forming a negative feedback loop. Furthermore, there is an auxiliary stabilizing loop, which consists of several modulatory nuclear hormone receptors, including REV-ERBs and RORs (Schroeder and Colwell, 2013). Additionally, clock transcription factors also regulate the expression of output genes, known as clock controlled genes (CCGs). These CCGs play key roles in driving ~24 hr rhythms in diverse biological processes, e.g. cell signaling, kinase and electrical activity, as well as ATP production (Reppert and Weaver, 2002; Schibler U, 2007)

Tissue-specific clock functions

Interestingly, although the basic mechanism of molecular clock is very similar among tissues, there is only a small overlap of rhythmic genes between any two given tissues. Comparing circadian transcriptome in different tissues, it was found that out of 9995 genes that showed circadian oscillations in one tissue or another, only 41 genes were rhythmic in at least 8 out of 14 mouse tissues analyzed (Yan et al., 2008). This concept was reinforced by the fact that even within the skeletal system (e.g. cartilage, tendon and skeletal muscle), the rhythmic transcriptomes differ vastly (Yeung et al., 2014). Indeed, many of the CCGs are tissue-specific transcription factors (such as MyoD in muscle and KLF15 in heart) and characteristic structural proteins, or fundamental metabolic pathways in a particular tissue (Zvonic et al., 2006; McCarthy et al., 2007; Zvonic et al., 2007; Andrews et al., 2010; Jeyaraj et al., 2012; Gossan et al., 2013). In view of this, it is probably not surprising that diverse pathologies have been reported in mice carrying mutations of clock genes. For example, the global Bmal1 knockout mice show a premature ageing phenotype in multiple tissues, including cataract, sarcopenia, arthropathy, etc. Similarly, homozygous Clock Δ19 mutant mice are hyperphagic and obese with attenuated energy balance, and develop metabolic syndrome over time (Turek et al., 2005). Tendon is an essential component of the musculoskeletal system that provides transmission of forces and an attachment point of muscles to bones. The Clock Δ19 mice and the Clock/Npas2 double KO mice also show spontaneously calcifying tendons (Yu, 2011; Yeung et al., 2014). The concept of tissue-specific clock function is further supported by conditional ablation of the core clock genes in peripheral tissues. Mice with conditional deletion of Bmal1 in pancreatic islets developed a diabetes mellitus-like disorder, with impaired insulin release and glucose tolerance (Marcheva et al., 2010). Disruption of the circadian clock in the liver, adipocytes, skin and heart leads to hypoglycemia and increased glucose clearance, obesity, epidermal ageing, increased sensitivity to hypertrophy, highlighting the importance of the local clocks in tissue physiology and diseases (Lamia et al., 2008; Durgan et al., 2011; Janich et al., 2011; Paschos et al., 2012).

Clocks in cartilage

Evidence of circadian rhythms in cartilage

Circadian rhythms in cartilage metabolism have been known for decades. Since the 1980s, there have been several rat models using growth plate cartilage to study the hypothesis of day/night changes in cartilage homeostasis. Measuring chondrocyte proliferation markers identified that the most active phase of proliferation was the early morning, which induces the growth plate expansion that peaks at noon (Stevenson et al., 1990). Interestingly, collagen matrix is also mainly synthesized at noon in growing rat cartilage, as revealed by incorporation of [3H] proline (Igarashi et al., 2013). Serum levels of several biomarkers

related to cartilage metabolism show day/night variations, including aggrecan, type II collagen, cartilage oligomeric matrix protein (COMP), hialuronic acid, keratin sulphate and TGF-β (Kong et al., 2006; Andersson et al., 2006).

The circadian rhythms in cartilage in vivo could be either system-driven, as has been shown in other tissues (Kornmann et al., 2007); or driven by a local molecular clock within chondrocytes. In contrast to growth plate, studies of circadian clocks in articular cartilage have only started in the past few years. This has been greatly facilitated by the development of new techniques, such as the real-time bioluminescence photon counting/imaging system (Yamazaki and Takahashi, 2005) and the generation of transgenic mouse models that harbor a short half-life luciferase (Yoo et al., 2004). By imaging the femoral bone of a PER2::Luciferase gene reporter mouse, strong circadian oscillations of PER2::Luc activities were observed from both the growth plate and femoral head cartilage, which maintained for several months (Okubo et al., 2013). Similar approaches have been used to monitor clock gene activities from isolated mouse femoral head, which revealed robust circadian oscillations in the articular cartilage of this tissue (Dudek et al., 2016). Importantly, these in vitro approaches highlight tissue autonomous rhythms in cartilage. Interestingly, targeted genetic deletion of the core clock gene Bmal1 from chondrocytes led to a complete abolishment of the cartilage circadian rhythm, supporting a key role of the molecular clock machinery in driving this local clock (Dudek et al., 2016). Time course qPCR studies have confirmed rhythmic expression of most core clock genes in different types of cartilages, including femoral head articular cartilage, xiphoid cartilage, rib cartilage and growth plate (Dudek et al., 2016; Gossan et al., 2013; Honda et al., 2013). These studies have also been expanded to human articular cartilage. For instance, by immunohistochemistry, the presence of clock proteins in the articular cartilage of knee joints was demonstrated. Further, using lentiviral delivery of clock reporters, cell-autonomous circadian rhythms in gene expression have been demonstrated in a human chondrocyte-derived cell line (Gossan et al., 2013; Dudek et al., 2016).

Clock targets in cartilage

Cartilage homeostasis is regulated by many key transcription factors and signaling pathways. Using cartilage samples collected from mice kept in constant darkness, time series microarrays were undertaken to reveal the extent of circadian control in this tissue. This approach revealed that the expression of 4% genes was circadian-regulated in mouse xiphoid cartilage (Gossan et al., 2013). This number is likely an underestimate in articular cartilage in mice (Yang et al unpublished RNA sequencing results in femoral head cartilage). Many of these rhythmic genes are involved in metabolic pathways, ECM remodeling (e.g. Timp4), apoptosis (e.g. X-linked inhibitor of apoptosis, Xiap), catabolic extracellular proteases (e.g. ADAMTSs and matrix metalloproteinase, MMPs) (Gossan et al., 2013; Dudek and Meng, 2014). ADAMTS4 is one of the main aggrecanases responsible for aggrecan degradation in articular cartilage (Verma and Dalal, 2011). MMP14 (aka MT1-MMP) is a membrane bound collagenase that regulates the pericellular ECM. MMP14 is also involved in the regulation of chondrogenic differentiation, and the activation of other MMP proteases including MMP13, the main collagenase involved in OA (Knauper et al., 1996). MMP14 knockout mice demonstrate widespread skeletal pathologies, including articular cartilage destruction (Holmbeck et al., 1999; Knauper et al., 2002; Szabova et al., 2009; Chellaiah et al., 2013). Interestingly, the rhythmic pattern of Timp4, an inhibitor of MMPs, is in complete antiphase to MMP14, indicating temporal separation of the activities of these two

opposing enzymes. Honda et al. also performed time-series partial-genome arrays in both permanent and growth plate cartilage from rats (Honda et al., 2013). Although a different statistical model was adopted to identify circadian regulated genes, significant numbers of ECM-related rhythmic genes were found to be common in rat and mouse cartilage (Table 1). Importantly, anabolic ECM genes such as fibrillins, laminins and netrin peaked in the early morning. Together, these studies demonstrate a circadian control mechanism for the anabolic and catabolic pathways in cartilage tissues. These data strongly support a hypothesis that the circadian rhythm in cartilage temporally segregates the activities of different ECM-related molecules to different (optimal) times of the day, which may provide an efficient mechanism by which cartilage can recuperate following bouts of nocturnal activity (Fig. 1).

Interestingly, the chondrocyte-selective deletion of Bmal1 in mice led to constant expression levels of most rhythmic genes identified in WT mice, indicating locally-expressed BMAL1 as a crucial regulator of circadian gene expression in vivo (Dudek et al., 2016). The global changes of gene expression patterns in the Bmal1 KO cartilage indicate an overall shift towards a more catabolic state of the chondrocytes. Two key pathways relevant to cartilage homeostasis, transforming growth factor beta (TGF-β) and nuclear factor of activated T-cells, cytoplasmic 2 (NFATc2), were shown to be dysregulated (Dudek et al., 2016). TGF-β signaling is essential for the maintenance of cartilage metabolism and structural integrity, and its dysregulation triggers a catabolic response in arthritic and aged chondrocytes (Finnson et al., 2008; van der Kraan et al., 2009; Bauge et al., 2013). Meanwhile, NFATc2 is a key chondrocyte transcription factor and its deletion results in OA-like cartilage degeneration, typified by decreased anabolic signaling and increased inflammatory and catabolic pathways (Wang et al., 2009; Rodova et al., 2011; Greenblatt et al., 2013). Loss of BMAL1 could impact on cartilage health in multiple ways, given the changes of diverse pathways in Bmal1 KO cartilage. Therefore, further studies are clearly needed to explore additional pathways that link circadian disruptions to cartilage metabolism.

Entrainment factors for cartilage clock

One key feature of the molecular circadian clock is that it can be entrained by various time cues. Unlike the central clock in the SCN which receives light input from the retina through the hypothalamic tract, most peripheral clocks are regulated through rhythmic signals that are under the control of the SCN, such as temperature and hormones (e.g. dexamethasone) (Gossan et al., 2013; Dudek and Meng, 2014). As a way of adaptation, different body organs have evolved to be differentially sensitive to various time cues. For instance, circadian clocks in the lung and fibroblasts can be regulated by temperature cycles that approximate body temperature changes. In contrast, the SCN seems to be protected from daily body temperature variations (Buhr et al., 2010). Daily feeding signals strongly entrain peripheral clocks in certain tissues, such as the liver and colon (Damiola et al., 2000; Sládek et al., 2007). Cartilage tissues are avascular and not innervated, therefore it is of particular interest to understand how this local clock communicates with the rest of the body to maintain coherent circadian rhythms. One candidate pathway is glucocorticoid signaling, which shows a diurnal daily rhythm controlled by the SCN, and has been shown to reset cartilage clocks (Gossan et al., 2013; Okubo et al., 2013). Studies using cartilage explant cultured under 12hr/12hr temperature cycles suggest that core body temperature rhythm could also serve as a potential synchronizing factor for cartilage clocks. It is worth noting that the activity of

the hypothalamic–pituitary–thyroid axis, including the expression of parathyroid hormone (PTH), has long been known to show diurnal daily variations (Greenspan et al., 1997; Dudek and Meng, 2014). PTH regulates cAMP/PKA (protein kinase A)/CREB signaling, and stimulates Per1 expression in both osteoblasts and growth plate chondrocytes, although its role in articular cartilage remains unknown (Hinoi et al., 2006; Hanyu et al., 2011). In addition, the daily rhythm of activity/rest could also be an endogenous entrainment factor for cartilage clock. This hypothesis is supported by results showing that the expression of Clock gene was mechanosensitive (Kanbe et al., 2006). Further studies are clearly needed to test how mechanical loading may affect circadian pacemaking, at least in the skeletal system.

Dysregulation- diseases

Disruptions to circadian rhythms (e.g. during ageing or in shift workers) have been linked to various human diseases, including obesity, diabetes, cardiovascular diseases, and even cancer (Evans and Davidson, 2013; Henriksson and Lamia, 2015; Martino and Young, 2015). In cartilage from aged mice, the amplitude of circadian rhythm has been shown to decrease by ~40%, which could contribute to age-related disease onset in this tissue. Over the last few years, evidence is emerging that environmental or genetic clock disruptions may also predispose cartilage to pathologies such as OA, one of the most prevalent joint diseases with no effective disease-modifying treatment available. There are multiple risk factors for OA including ageing, obesity, trauma to the joint tissue and chronic inflammation in the joint cavity. Indeed, altered expression of several key clock genes have been found in a mechanical injury model of OA in mouse (by destabilization of the medial meniscus, Gossan et al., 2013). IL-1, a critical cytokine for matrix metalloproteinase (MMP)-dependent cartilage degradation, disrupts circadian rhythms in mouse cartilage explants and chondrocytes in an NFkB-dependent manner (Guo et al., 2015). It was reported that long term (22 weeks) environmental disruption to the light/dark cycles in mice, a situation that mimics many years of rotating shift work or chronic jet lag, predispose the mouse knees to OA-like damages (Kc et al., 2015). Moreover, a progressive arthropathy phenotype has been reported in global Bmal1 KO mice (Bunger et al., 2005). This phenotype is age-dependent and highly localized (in calcaneal tendon and costosternal cartilage), and requires absence of Bmal1 during development (Yang et al., 2016). Therefore, whether this pathology is due to circadian dysfunction is still unclear. Moreover, one caveat of such systemic rhythm disruption paradigm (either genetic or environmental) is the global dysregulation of circadian rhythms in locomotion, body temperature or hormones, which may confound the interpretation of any phenotype observed (Yu and Weaver, 2011). One way to overcome these confounding factors is to use a conditional knockout model by removing the local clock selectively in chondrocytes. Using this gene targeting approach, severe age-related degeneration and lesions were observed in the articular cartilage of the knee joints (Dudek et al., 2016). These studies prove an essential role of the core clock factor BMAL1 in cartilage homeostasis. At this stage, we cannot exclude the possibility that BMAL1 may also be involved in the regulation of non-rhythmic genes. However, the fact that loss of chondrocyte-BMAL1 leads to global dysregulation of the patterns of rhythmic genes strongly support a role of circadian rhythms in the homeostasis of articular cartilage (Dudek et al., 2016). It has been reported that the levels of BMAL1 and the number of BMAL1-positive chondrocytes in human osteoarthritic cartilage are significantly lower than non-arthritic cartilage, indicating a role of clock mechanisms in human OA (Dudek et al., 2016).

Clocks in bone

Evidence of clocks in bone metabolism and physiology

Bone homoeostasis depends on two opposing processes, resorption and synthesis, which requires strict temporal control to maintain the bone mass and structural quality (Feng and McDonald, 2011). A circadian rhythm in bone formation was demonstrated by measuring radiolabeled proline and galactose. Synthesis and secretion of ECM components by hypertrophic chondrocytes and osteoblasts exhibited a ~24hr rhythm in growing rats, starting in the early morning (resting phase) and peaking at around 1 pm (Igarashi et al., 2013). By measuring calcium and phosphorus fractions, it was shown that cartilage mineralization in rat tibial growth plate occurs at night during the activity phase (Russell et al., 1984). Diurnal Daily variations of serum markers linked to bone metabolism, such as calcitonin, calcium, C-telopeptide, osteocalcin, parathyroid hormone, skeletal ALP (alkaline phosphatase) and tartrate-resistant acid phosphatase have been reported (Greenspan et al., 1997; Srivastava et al., 2001; Shao et al., 2003). Moreover, during ossteointegration (the process by which bone implants become integrated with the surrounding tissue), the expression of clock genes was associated with expression of cartilage ECM markers (Col2a1, Col10a1 and link protein) (Mengatto et al., 2011). It was suggested that Clock/Npas2 may act directly on E-box elements in the promoters of Col2a1, Col10a1 and Acan (aggrecan). Indeed, subsequent studies by Hinoi et al. demonstrated the presence of circadian gene expression in femurs and a functional E-box in the first intron of Col2a1 (Hinoi et al., 2006). A recent study also demonstrated robust circadian rhythms in the fracture healing sites within femurs (Kunimoto et al., 2016).

In recent years, the increasing appreciation of the peripheral clocks has prompted studies of the molecular clocks in resident bone cells. Among different cell types, the circadian clock in osteoblasts is most extensively studied. It has been shown that the clock affects bone formation potentially through regulating osteoblasts differentiation from bone marrow-derived stem cells (Goldstrohm and Wickens, 2008). The core clock complex BMAL1/Clock directly regulates the expression of nocturnin (Noc), a deadenylase post-transcriptionally modulating gene expression (Guntur et al., 2011). Differentiation of osteoblasts inhibits Noc expression, while overexpression of Noc blocks osteoblastogenesis (Goldstrohm and Wickens, 2008; Li et al., 2008). The diurnal daily expressions of Per1, Per2 and several osteoclast markers (NFATc1 and Cathepsin K, Ctsk) were shown in cancellous bone in mice, as well as in cultured osteoclasts (Fujihara et al., 2014). To the best of our knowledge, no studies so far have characterized molecular clocks in osteocytes. However, Samsa et al (2016) used histomorphometry to demonstrate a significantly reduced number of osteocytes (and osteoblasts) in bones of the global Bmal1 knockout mice, indicating a possible role of BMAL1 in the terminal differentiation of osteoblasts.

Rhythmic targets in bone

As a highly calcified tissue, genome-wide transcriptome studies in long bone tissue have been technically challenging. Neither has this type of study been done in cultured bone cells (osteoblasts, osteoclasts or osteocytes). However, the analysis of diurnal daily transcriptome in calvarial bone (ossified by intramembranous ossification) has provided some clues about the scale of clock targets within bone tissues. This approach revealed that the mRNA expression of 26% identified genes were rhythmic with a 24 hr oscillation, including bone

morphogenetic protein (BMP), fibroblast growth factor (FGF) and WNT signaling pathways, and genes that encode multiple ADAMTS proteases and various pro-collagen isoforms (Zvonic et al., 2007). Among these clock targets, a significant number of genes are related to ECM regulation (Table 1). These matrix-related CCGs strongly support a role of the circadian rhythm in maintaining homeostasis of this matrix-rich and mineralized skeletal tissue.

Entrainment factors for bone clock

The circadian clocks in the bone can be entrained by both external stimuli and internal signals, to maintain the diurnal daily rhythms in bone metabolism (formation and resorption) and bone mass (Bjarnason et al., 2002; Karsdal et al., 2008; Komoto et al., 2012). The SCN output time cues may be transmitted to the bone through changes in leptin levels, glucocorticoid signaling and feeding-fasting cycles (Le Minh et al., 2001; Komoto et al., 2012). Leptin was proposed as one of the entrainment signals for osteoblasts, and it may regulate the cell intrinsic clocks through activation of Adrb2 (β2-adrenergic receptor) on the cell surface. Upon receptor activation, the level of cAMP-responsive-element-binding protein 1 (Creb1) was increased, leading to elevated expression of Bmal1 and Clock (Downey and Siegel, 2006). Dexamethasone stimulated the expression of clock genes, as well as induced rhythmic bone resorption in cultured osteoclasts in vitro (Fujihara et al., 2014). The rhythmic expressions of NFATc1 and Ctsk were abolished in cancellous bone from adrenalectomised mice, where the amplitude of rhythmic Per1 and Per2 genes was dramatically reduced. Remarkably, glucocorticoid injection restored the circadian expression of these osteoclast specific genes in vivo (Fujihara et al., 2014). The diurnal daily variations in the bone metabolism seem to be governed by timing of food intake and fasting, providing an additional entrainment mechanism for bone (Bjarnason et al., 2002; Karsdal et al., 2008)

Dysregulation- bone defects

Recent studies using circadian gene KO mice have demonstrated an essential role of bone clock in regulating bone metabolism. A high bone mass (HBM) phenotype was reported in mice lacking either Per1 gene or the Per/Arnt/Sim (PAS) domain of Per2 from 6 weeks of age, and progressed with ageing, associated with an increased number of osteoblasts in bones as identified by both biochemical and histomorphometric analyses. This phenotype is consistent with the up-regulated rates of mineral apposition and bone formation (Fu et al., 2005; Maronde et al., 2010). Similar HBM phenotype was also observed in mice with double knockout of Per1/Per2 or Cry1/Cry2 (Fu et al., 2005). In addition, Cry2 was shown to influence osteoclast activity whereas Per2 was reported to act on osteoblasts in mice (Maronde et al., 2010). In contrast, global deletion of Bmal1, the positive regulator within the clock, led to a low bone mass phenotype, which at least partially involves compromised capacity of mesenchymal stem cells to differentiate into osteoblasts (Samsa et al., 2016). Bmal1-KO mice are generally shorter (with significantly shorter bones), and the overall skeleton weight is lower comparing with WT mice (Bunger et al., 2005; Fu et al., 2005; Kondratov et al., 2006; Takarada et al., 2012). This shorter bone phenotype was proposed to be caused by the altered circadian expression of Ihh, which is a key regulator of endochondral ossification within the growth plate (Takarada et al., 2012). Although not proven yet, these opposing phenotypes in different clock mutant animals strongly point to a mechanism in which different clock factors regulate bone formation by coordinated actions

on the activities of osteoblast and osteoclast (Fig. 1). It is therefore likely that the circadian clock in bone cells has evolved to temporally orchestrate the two opposing events to maintain bone homeostasis.

Therapeutic targeting potential

Understanding the circadian regulation of homeostasis and metabolism in cartilage and bone provides a novel avenue to treat chronic (and age-related) diseases, such as OA and osteoporosis. For instance, in cartilage from mice (a nocturnal species), it has been discovered that anabolic genes mainly peaked at early night (before activity), while catabolic genes peaked early in the day (after activity) (Dudek and Meng, 2014). Therefore, drugs targeting these pathways should be used at times of the day their targets are most active, so as to attain maximum efficacy while minimize side effects (Kaur et al., 2013; Zhang et al., 2014). Indeed chronotherapy, which is to deliver drugs based on the body clock time, has been trialed in OA pain management since 1980s. Remarkably, a chronotherapeutic dosing regimen showed doubled analgesic effectiveness and 4x tolerance of indomethacin treatment to reduce OA pain (Levi et al., 1985). Furthermore, the development of compounds directly targeting molecular clocks has been described, including CK1 enzymes, CRY proteins, REV-ERB nuclear receptors, SIRT1 compounds and Lithium. Although these compounds are still at early stage of development and mainly towards inflammation or metabolic disorders, they are also potentially useful to treat other diseases in future (Meng et al., 2008; Meng et al., 2010; Hirota et al., 2012; Solt et al., 2012). Given the well-established role of cortisol in entraining peripheral clocks, local delivery of synthetic glucocorticoids to the diseased joint may have potential therapeutic value to boost the intrinsic cartilage rhythm, in addition to their anti-inflammatory actions. Moreover, as the rhythmic genes in cartilage and bone are controlled not only by the local clocks, but also by the time cues that are under the control of the central clock, it is possible to systemically target the circadian system, in order to slow down disease progression. Mechanical stimulation through exercises is well known to improve cartilage and bone function. Interestingly, the mammalian and drosophila circadian clocks have previously been identified as mechanosensitive (Kanbe et al. 2006; Simoni et al. 2014). If mechanical stimulation can re-entrain the clocks in cartilage and bone, perhaps a regular schedule of gentle exercise would directly benefit the local skeletal clocks in addition to having feedback effects on the central clock in the brain. From the circadian clock perspective, it is probably worth considering not only the duration of the physical exercises, but also the timing. Although speculative at this stage, sticking to a regular exercise scheme on a daily basis is likely to be more beneficial than exercising at random times. Perhaps a personalized schedule, according to an individual’s chronotype (preference to be alert early or late), would work most efficiently (Facer-Childs and Brandstaetter, 2015; Roenneberg and Merrow, 2016). Additional systemic time cues (e.g. regular meal times) may also be used to enhance the impaired skeletal clock during diseases, or even ageing (Feillet et al., 2006; Kanbe et al., 2006; Hughes and Piggins, 2012; Simoni et al., 2014).

Concluding remarks and future directions

The identification of functional circadian clocks in matrix-rich cartilage and bone tissues has opened up an exciting new avenue for skeletal research. This temporal control mechanism provides a critical time cue for chondrocytes and resident bone cells to coordinate ECM homeostasis with the daily rhythmic cycles of rest/activity, body temperature fluctuations, and surges of hormones. During ageing or in shift workers, disruptions to circadian rhythms may compromise this protective mechanism, leading to increased risk of tissue pathologies. Future work should be directed towards understanding the major clock-controlled pathways in articular cartilage (e.g. the use of laser capture microdissection) and long bone, and establishing the cause-effect relationship using genetic studies in animal models (such as cell type-specific ablation of individual clock components in chondrocytes or different bone cells). The musculoskeletal system is highly relevant to the daily rest/activity and there may be a strong component of mechanical control for the skeletal clocks. Moreover, it is important to demonstrate changes of the circadian clock in patients with skeletal diseases that affects cartilage and bone. On the other side, detailed studies of clocks in different types of cartilage, different joints, and different bone cell types are still lacking. Finally, effects of clock-acting interventions (systemic approaches, pharmacological compounds or chronotherapeutic dosing of drugs) on cartilage/bone metabolism and matrix homeostasis should be tested in vivo in experimental models.

AcknowledgmentsThis work has been funded by the following grants: a Medical Research Council (MRC, UK) project grant (MR/K019392/1), a Wellcome Trust (UK) Core funding (088785/Z/09/Z) to the Wellcome Trust Centre for Cell-Matrix Research, an Arthritis Research UK Senior Research Fellowship (20875).

References

Andersson ML, Petersson IF, Karlsson KE, Jonsson EN, Mansson B, Heinegard D and Saxne T (2006) Diurnal variation in serum levels of cartilage oligomeric matrix protein in patients with knee osteoarthritis or rheumatoid arthritis. Ann Rheum Dis 65:1490–1494.

Andrews JL, Zhang X, McCarthy JJ, McDearmon EL, Hornberger TA, Russell B, Campbell KS, Arbogast S, Reid MB, Walker JR, et al. (2010) CLOCK and BMAL1 regulate MyoD and are necessary for maintenance of skeletal muscle phenotype and function. Proc Natl Acad Sci U S A 107:19090-19095.

Asanbaeva A, Tam J, Schumacher BL, Klisch SM, Masuda K and Sah RL (2008) Articular cartilage tensile integrity: modulation by matrix depletion is maturation-dependent. Arch Biochem Biophys 474:175-182.

Bauge C, Girard N, Lhuissier E, Bazille C and Boumediene K (2013) Regulation and role of TGFβ signaling pathway in aging and osteoarthritis joints. Aging Dis 5:394-405.

Bjarnason NH, Henriksen EE, Alexandersen P, Christgau S, Henriksen DB and Christiansen C (2002) Mechanism of circadian variation in bone resorption. Bone 30:307–313.

Buhr ED, Yoo SH and Takahashi JS (2010) Temperature as a universal resetting cue for mammalian circadian oscillators. Science 330:379-385.

Bunger MK, Walisser JA, Sullivan R, Manley PA, Moran SM, Kalscheur VL, Colman RJ and Bradfield CA (2005) Progressive arthropathy in mice with a targeted disruption of the Mop3/Bmal-1 locus. Genesis 41:122–132.

Buttgereit F, Smolen JS, Coogan AN and Cajochen C (2015) Clocking in: chronobiology in rheumatoid arthritis. Nat Rev Rheumatol 11:349-356.

Cermakian N and Sassone-Corsi P (2000) Multilevel regulation of the circadian clock. Nat Rev Mol Cell Biol 1:59-67.

Chellaiah MA and Ma T (2013) Membrane localization of membrane type 1 matrix metalloproteinase by CD44 regulates the activation of pro-matrix metalloproteinase 9 in osteoclasts. Biomed Res Int 2013:302392.

Damiola F, Le Minh N, Preitner N, Kornmann B, Fleury-Olela F and Schibler U (2000) Restricted feeding uncouples circadian oscillators in peripheral tissues from the central pacemaker in the suprachiasmatic nucleus. Genes Dev 14:2950-2961.

Dibner C, Schibler U and Albrecht U (2010) The mammalian circadian timing system: organization and coordination of central and peripheral clocks. Annu Rev Physiol 72:517–549.

Downey PA and Siegel MI (2006) Bone biology and the clinical implications for osteoporosis. Phys Ther 86:77–91.

Dudek M and Meng QJ (2014) Running on time: the role of circadian clocks in the musculoskeletal system. Biochem J 463:1-8.

Dudek M, Gossan N, Yang N, Im HJ, Ruckshanthi JP, Yoshitane H, Li X, Jin D, Wang P, Boudiffa M, et al. (2016) The chondrocyte clock gene Bmal1 controls cartilage homeostasis and integrity. J Clin Invest 126:365-376.

Durgan DJ, Tsai JY, Grenett MH, Pat BM, Ratcliffe WF, Villegas-Montoya C, Garvey ME, Nagendran J, Dyck JR and Bray MS (2011) Evidence suggesting that the cardiomyocyte circadian clock modulates responsiveness of the heart to hypertrophic stimuli in mice. Chronobiol Int 28:187-203.

Evans JA and Davidson AJ (2013) Health consequences of circadian disruption in humans and animal models. Prog Mol Biol Transl Sci 119:283-323.

Facer-Childs E and Brandstaetter R (2015) The impact of circadian phenotype and time since awakening on diurnal performance in athletes. Curr Biol 25:518-522.

Feillet CA, Ripperger JA, Magnone MC, Dulloo A, Albrecht U and Challet E (2006) Lack of food anticipation in Per2 mutant mice. Curr Biol 16:2016-2022.

Feng X and McDonald JM (2011) Disorders of bone remodeling. Annu Rev Pathol 6:121–145.

Finnson KW, Parker WL, ten Dijke P, Thorikay M and Philip A (2008) ALK1 opposes ALK5/Smad3 signaling expression of extracellular matrix components in human chondrocytes. J Bone Miner Res 23:896–906.

Fu L, Patel MS, Bradley A, Wagner EF and Karsenty G (2005) The molecular clock mediates leptin-regulated bone formation. Cell 122:803–815.

Fujihara Y, Kondo H, Noguchi T and Togari A (2014) Glucocorticoids mediate circadian timing in peripheral osteoclasts resulting in the circadian expression rhythm of osteoclast-related genes. Bone 61:1–9.

Gamble KL, Berry R, Frank SJ and Young ME (2014) Circadian clock control of endocrine factors. Nat Rev Endocrinol 10:466-475.

Goldstrohm AC and Wickens M (2008) Multifunctional deadenylase complexes diversify mRNA control. Nat Rev Mol Cell Biol 9:337–344.

Gossan N, Zeef L, Hensman J, Hughes A, Bateman JF, Rowley L, Little CB, Piggins HD, Rattray M, Boot-Handford RP, et al. (2013) The circadian clock in murine chondrocytes regulates genes controlling key aspects of cartilage homeostasis. Arthritis Rheum 65:2334–2345.

Greenblatt MB, Ritter SY, Wright J, Tsang K, Hu D, Glimcher LH and Aliprantis AO (2013) NFATc1 and NFATc2 repress spontaneous osteoarthritis. Proc Natl Acad Sci U S A 110:19914–19919.

Greenspan SL, Dresner-Pollak R, Parker RA, London D and Ferguson L (1997) Diurnal variation of bone mineral turnover in elderly men and women. Calcif Tissue Int 60:419–423.

Guntur AR, Kawai M, Le P, Bouxsein ML, Bornstein S, Green CB and Rosen CJ (2011) An essential role for the circadian-regulated gene nocturnin in osteogenesis: the importance of local timekeeping in skeletal homeostasis. Ann N Y Acad Sci 1237:58–63.

Guo B, Yang N, Borysiewicz E, Dudek M, Williams JL, Li J, Maywood ES, Adamson A, Hastings MH, Bateman JF, et al. (2015) Catabolic cytokines disrupt the circadian clock and the expression of clock-controlled genes in cartilage via an NFкB-dependent pathway. Osteoarthritis Cartilage 23:1981-1988.

Hanyu R, Hayata T, Nagao M, Saita Y, Hemmi H, Notomi T, Nakamoto T, Schipani E, Knonenbery H, Kaneko K, et al. (2011) Per-1 is a specific clock gene regulated by parathyroid hormone (PTH) signaling in osteoblasts and is functional for the transcriptional events induced by PTH. J Cell Biochem 112:433–438.

Hardingham T, Tew S and Murdoch A (2002) Tissue engineering: chondrocytes and cartilage. Arthritis Res 4:S63-68.

Hardingham T (2010) Cell- and tissue-based approaches for cartilage repair. Altern Lab Anim 38:35-39.

Henriksson E and Lamia KA (2015) Adipose Clocks: Burning the Midnight Oil. J Biol Rhythms 30:364-373.

Hinoi E, Ueshima T, Hojo H, Iemata M, Takarada T and Yoneda Y (2006) Up-regulation of per mRNA expression by parathyroid hormone through a protein kinase A-CREB-dependent mechanism in chondrocytes. J Biol Chem 281:23632–23642.

Hirota T, Lee JW, St John PC, Sawa M, Iwaisako K, Noguchi T, Pongsawakul PY, Sonntag T, Welsh DK, Brenner DA, et al. (2012) Identification of small molecule activators of cryptochrome. Science 337:1094-1097.

Holmbeck K, Bianco P, Caterina J, Yamada S, Kromer M, Kuznetsov SA, Mankani M, Robey PG, Poole AR, Pidoux I, et al. (1999) MT1-MMP-deficient mice develop dwarfism, osteopenia, arthritis, and connective tissue disease due to inadequate collagen turnover. Cell 99:81–92.

Honda KK, Kawamoto T, Ueda HR, Nakashima A, Ueshima T, Yamada RG, Nishimura M, Oda R, Nakamura S, Kojima T, et al. (2013) Different circadian expression of major matrix-related genes in various types of cartilage: modulation by light-dark conditions. J Biochem 154:373–381.

Hughes AT and Piggins HD (2012) Feedback actions of locomotor activity to the circadian clock. Prog Brain Res 199:305-336.

Igarashi K, Saeki S and Shinoda H (2013) Diurnal rhythms in the incorporation and secretion of 3H-proline and 3H-galactose by cartilage cells and osteoblasts in various bone-forming sites in growing rats. Orthod Waves 72:11–15.

Janich P, Pascual G, Merlos-Suarez A, Batlle E, Ripperger J, Albrecht U, Cheng HY, Obrietan K, Di Croce L and Benitah SA (2011) The circadian molecular clock creates epidermal stem cell heterogeneity. Nature 480:209–214.

Jeyaraj D, Haldar SM, Wan X, McCauley MD, Ripperger JA, Hu K, Lu Y, Eapen BL, Sharma N, Ficker E, et al. (2012) Circadian rhythms govern cardiac repolarization and arrhythmogenesis. Nature 483:96-99.

Kanbe K, Inoue K, Xiang C and Chen Q (2006) Identification of clock as a mechanosensitive gene by large-scale DNA microarray analysis: downregulation in osteoarthritic cartilage. Mod Rheumatol 16:131–136.

Karsdal MA, Byrjalsen I, Riis BJ and Christiansen C (2008) Investigation of the diurnal variation in bone resorption for optimal drug delivery and efficacy in osteoporosis with oral calcitonin. BMC Clin Pharmacol 8:12.

Kaur G, Phillips C, Wong K and Saini B (2013) Timing is important in medication administration: a timely review of chronotherapy research. Int J Clin Pharm 35:344-358.

Kc R, Li X, Voigt RM, Ellman MB, Summa KC, Vitaterna MH, Keshavarizian A, Turek FW, Meng QJ, Stein GS, et al. (2015) Environmental disruption of circadian rhythm predisposes mice to osteoarthritis-like changes in knee joint. J Cell Physiol 230:2174-2183.

Knauper V, Will H, Lopez-Otin C, Smith B, Atkinson SJ, Stanton H, Hembry RM and Murphy G (1996) Cellular mechanisms for human procollagenase-3 (MMP-13) activation. Evidence that MT1-MMP (MMP-14) and gelatinase a (MMP-2) are able to generate active enzyme. J Biol Chem 271:17124-17131.

Knauper V, Bailey L, Worley JR, Soloway P, Patterson ML and Murphy G (2002) Cellular activation of proMMP-13 by MT1-MMP depends on the C-terminal domain of MMP-13. FEBS Lett 532:127–130.

Ko CH and Takahashi JS (2006) Molecular components of the mammalian circadian clock. Hum Mol Genet 15:R271–R277.

Komoto S, Kondo H, Fukuta O and Togari A (2012) Comparison of β-adrenergic and glucocorticoid signaling on clock gene and osteoblast-related gene expressions in human osteoblast. Chronobiol Int 29:66–74.

Kondratov RV, Kondratova AA, Gorbacheva VY, Vykhovanets OV and Antoch MP (2006) Early aging and age-related pathologies in mice deficient in BMAL1, the core component of the circadian clock. Genes Dev 20:1868–1873.

Kong SY, Stabler TV, Criscione LG, Elliott AL, Jordan JM and Kraus VB (2006) Diurnal variation of serum and urine biomarkers in patients with radiographic knee osteoarthritis. Arthritis Rheum 54:2496–2504.

Kornmann B, Schaad O, Reinke H, Saini C and Schibler U (2007) Regulation of circadian gene expression in liver by systemic signals and hepatocyte oscillators. Cold Spring Harb Symp Quant Biol 72:319-330.

Kumar LD, Karthik R, Gayathri N and Sivasudha T (2016) Advancement in contemporary diagnostic and therapeutic approaches for rheumatoid arthritis. Biomed Pharmacother 79:52-61.

Kunimoto T, Okubo N, Minami Y, Fujiwara H, Hosokawa T, Asada M, Oda R, Kubo T and Yagita K (2016) A PTH-responsive circadian clock operates in ex vivo mouse femur fracture healing site. Sci Rep 6:22409.

Lamia KA, Storch KF and Weitz CJ (2008) Physiological significance of a peripheral tissue circadian clock. Proc Natl Acad Sci U S A 105:15172–15177.

Le Minh N, Damiola F, Tronche F, Schutz G and Schibler U (2001) Glucocorticoid hormones inhibit food-induced phase-shifting of peripheral circadian oscillators. Embo J 20:7128-7136.

Levi F, Le Louarn C and Reinberg A (1985) Timing optimizes sustained-release indomethacin treatment of osteoarthritis. Clin Pharmacol Ther 37:77-84.

Li R, Yue J, Zhang Y, Zhou L, Hao W, Yuan J, Qiang B, Ding JM, Peng X and Cao JM (2008) CLOCK/BMAL1 regulates human nocturnin transcription through binding to the E-box of nocturnin promoter. Mol Cell Biochem 317:169–177.

Marcheva B, Ramsey KM, Buhr ED, Kobayashi Y, Su H, Ko CH, Ivanova G, Omura C, Mo S, Vitaterna MH, et al. (2010) Disruption of the clock components CLOCK and BMAL1 leads to hypoinsulinaemia and diabetes. Nature 466:627–631.

Maronde E, Schilling AF, Seitz S, Schinke T, Schmutz I, van der Horst G, Amling M and Albrecht U (2010) The clock genes Period 2 and Cryptochrome 2 differentially balance bone formation. PLoS One 5:e11527.

Martino TA and Young ME (2015) Influence of the cardiomyocyte circadian clock on cardiac physiology and pathophysiology. J Biol Rhythms 30:183-205.

McCarthy JJ, Andrews JL, McDearmon EL, Campbell KS, Barber BK, Miller BH, Walker JR, Hogenesch JB, Takahashi JS and Esser KA (2007) Identification of the circadian transcriptome in adult mouse skeletal muscle. Physiol Genomics 31:86–95.

Melrose J, Shu C, Whitelock JM and Lord MS (2016) The cartilage extracellular matrix as a transient developmental scaffold for growth plate maturation. Matrix Biol 52-54:363-383.

Meng QJ, McMaster A, Beesley S, Lu WQ, Gibbs J, Parks D, Collins J, Farrow S, Donn R, Ray D, et al. (2008) Ligand modulation of REV-ERBalpha function resets the peripheral circadian clock in a phasic manner. J Cell Sci 121:3629-3635.

Meng QJ, Maywood ES, Bechtold DA, Lu WQ, Li J, Gibbs JE, Dupre SM, Chesham JE, Rajamohan F, Knafels J, et al. (2010) Entrainment of disrupted circadian behavior through inhibition of casein kinase 1 (CK1) enzymes. Proc Natl Acad Sci U S A 107:15240-15245.

Mengatto CM, Mussano F, Honda Y, Colwell CS and Nishimura I (2011) Circadian rhythm and cartilage extracellular matrix genes in osseointegration: a genome-wide screening of implant failure by vitamin D deficiency. PLoS ONE 6:e15848.

Okubo N, Minami Y, Fujiwara H, Umemura Y, Tsuchiya Y, Shirai T, Oda R, Inokawa H, Kubo T and Yagita K (2013) Prolonged bioluminescence monitoring in mouse ex vivo bone culture revealed persistent circadian rhythms in articular cartilages and growth plates. PLoS ONE 8:e78306.

Paschos G, Ibrahim S, Song WL, Kunieda T and Grant G (2012) Obesity in mice with adipocyte-specific deletion of clock component Arntl. Nat Med 18:1768–1779.

Poole CA, Flint MH and Beaumont BW (1987) Chondrons in cartilage: ultrastructural analysis of the pericellular microenvironment in adult human articular cartilage. J Orthop Res 5:509-522.

Reppert SM and Weaver DR (2001) Molecular analysis of mammalian circadian rhythms. Annu Rev Physiol 63:647–676.

Reppert SM and Weaver DR (2002) Coordination of circadian timing in mammals. Nature 418:935–941.

Rodova M, Lu Q, Li Y, Woodbury BG, Crist JD, Gardner BM, Yost JG, Zhong XB, Anderson HC and Wang J (2011) Nfat1 regulates adult articular chondrocyte function through its age-dependent expression mediated by epigenetic histone methylation. J Bone Miner Res 26:1974–1986.

Roenneberg T and Merrow M (2005) Circadian clocks: the fall and rise of physiology. Nat Rev Mol Cell Biol 6:965–971.

Roenneberg T and Merrow M (2016) The Circadian Clock and Human Health. Curr Biol 26:R432-443.

Russell JE, Grazman B and Simmons DJ (1984) Mineralization in rat metaphyseal bone exhibits a circadian stage dependency. Proc Soc Exp Biol Med 176:342–345.

Samsa WE, Vasanji A, Midura RJ and Kondratov RV (2016) Deficiency of circadian clock protein BMAL1 in mice results in a low bone mass phenotype. Bone 84:194-203.

Schibler U (2007) The daily timing of gene expression and physiology in mammals. Dialogues Clin Neurosci 9:257–272.

Schroeder AM and Colwell CS (2013) How to fix a broken clock. Trends Pharmacol Sci 34:605-619.

Shao P, Ohtsuka-Isoya M and Shinoda H (2003) Circadian rhythms in serum bone markers and their relation to the effect of etidronate in rats. Chronobiol Int 20:325–336.

Simoni A, Wolfgang W, Topping MP, Kavlie RG, Stanewsky R and Albert JT (2014) A mechanosensory pathway to the Drosophila circadian clock. Science 343:525-528.

Sladek M, Rybova M, Jindrakova Z, Zemanova Z, Polidarova L, Mrnka L, O'Neill J, Pacha J and Sumova A (2007) Insight into the circadian clock within rat colonic epithelial cells. Gastroenterology 133:1240-1249.

Solt LA, Wang Y, Banerjee S, Hughes T, Kojetin DJ, Lundasen T, Shin Y, Liu J, Cameron MD, Noel R, et al. (2012) Regulation of circadian behaviour and metabolism by synthetic REV-ERB agonists. Nature 485:62-68.

Sommerfeldt DW and Rubin CT (2001) Biology of bone and how it orchestrates the form and function of the skeleton. Eur Spine J 10:S86-S95.

Sophia Fox AJ, Bedi A and Rodeo SA (2009) The basic science of articular cartilage: structure, composition, and function. Sports Health 1:461-468.

Spies CM, Straub RH, Cutolo M and Buttgereit F (2014) Circadian rhythms in rheumatology--a glucocorticoid perspective. Arthritis Res Ther 16:S3.

Srivastava AK, Bhattacharyya S, Li X, Mohan S and Baylink DJ (2001) Circadian and longitudinal variation of serum C-telopeptide, osteocalcin, and skeletal alkaline phosphatase in C3H/HeJ mice. Bone 29:361–367.

Stevenson S, Hunziker EB, Herrmann W and Schenk RK (1990) Is longitudinal bone growth influenced by diurnal variation in the mitotic activity of chondrocytes of the growth plate? J Orthop Res 8:132–135.

Szabova L, Yamada SS, Wimer H, Chrysovergis K, Ingvarsen S, Behrendt N, Engelholm LH and Holmbeck K (2009) MT1-MMP and type II collagen specify skeletal stem cells and their bone and cartilage progeny. J Bone Miner Res 24:1905–1916.

Takahashi JS, Hong H K, Ko CH and McDearmon EL (2008) The genetics of mammalian circadian order and disorder: implications for physiology and disease. Nat Rev Genet 9:764–775.

Takarada T, Kodama A, Hotta S, Mieda M, Shimba S, Hinoi E and Yoneda Y (2012) Clock genes influence gene expression in growth plate and endochondral ossification in mice. J Biol Chem 287:36081–36095.

Turek FW, Joshu C, Kohsaka A, Lin E, Ivanova G, McDearmon E, Laposky A, Losee-Olson S, Easton A, Jensen DR, et al. (2005) Obesity and metabolic syndrome in circadian Clock mutant mice. Science 308:1043-1045.

van der Kraan PM, Blaney Davidson EN, Blom A and van den Berg WB (2009) TGF-β signaling in chondrocyte terminal differentiation and osteoarthritis: modulation and integration of signaling pathways through receptor-Smads. Osteoarthritis Cartilage 17:1539–1545.

Verma P and Dalal K (2011) ADAMTS-4 and ADAMTS-5: key enzymes in osteoarthritis. J Cell Biochem 112:3507–3514.

Wang J, Gardner BM, Lu Q, Rodova M, Woodbury BG, Yost JG, Roby KF, Pinson DM, Tawfik O and Anderson HC (2009) Transcription factor Nfat1 deficiency causes osteoarthritis through dysfunction of adult articular chondrocytes. J Pathol 219:163–172.

Yamazaki S and Takahashi JS (2005) Real-time luminescence reporting of circadian gene expression in mammals. Methods Enzymol 393:288-301.

Yan J, Wang H, Liu Y and Shao C (2008) Analysis of gene regulatory networks in the mammalian circadian rhythm. PLoS Comput Biol 4:e1000193.

Yang G, Chen L, Grant GR, Paschos G, Song WL, Musiek ES, Lee V, McLoughlin SC, Grosser T, Cotsarelis G, et al. (2016) Timing of expression of the core clock gene Bmal1 influences its effects on aging and survival. Sci Transl Med 8:324ra16.

Yeung CY, Gossan N, Lu Y, Hughes A, Hensman JJ, Bayer ML, Kjær M, Kadler KE and Meng QJ (2014) Gremlin-2 is a BMP antagonist that is regulated by the circadian clock. Sci Rep 4:5183.

Yoo SH, Yamazaki S, Lowrey PL, Shimomura K, Ko CH, Buhr ED, Siepka SM, Hong HK, Oh WJ, Yoo OJ, et al. (2004) PERIOD2::LUCIFERASE real-time reporting of circadian dynamics reveals persistent circadian oscillations in mouse peripheral tissues. Proc Natl Acad Sci U S A 101:5339–5346.

Yu EA and Weaver DR (2011) Disrupting the circadian clock: gene-specific effects on aging, cancer, and other phenotypes. Aging (Albany NY) 3:479-493.

Yu EA (2011) Investigating Age-Dependent Arthropathy in a Circadian Mutant Mouse Model: A Dissertation. University of Massachusetts Medical School. GSBS Dissertations and Theses. Paper 544. http://escholarship.umassmed.edu/gsbs_diss/544

Zhang R, Lahens NF, Ballance HI, Hughes ME and Hogenesch JB (2014) A circadian gene expression atlas in mammals: implications for biology and medicine. Proc Natl Acad Sci U S A 111:16219-16224.

Zvonic S, Ptitsyn AA, Conrad SA, Scott LK, Floyd ZE, Kilroy G, Wu X, Goh BC, Mynatt RL and Gimble JM (2006) Characterization of peripheral circadian clocks in adipose tissues. Diabetes 55:962–970.

Zvonic S, Ptitsyn AA, Kilroy G, Wu X, Conrad SA, Scott LK, Guilak F, Pelled G, Gazit D and Gimble JM (2007) Circadian oscillation of gene expression in murine calvarial bone. J Bone Miner Res 22:357–365.

Table 1

Category Symbol Gene name Rhythmic in cartilage

Rhythmicin

boneExtracellular matrix remodelling enzymes

Adamts1 a disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motif, 1

Yes Yes

Adamts4 a disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motif, 4

Yes Yes

Adamts9 a disintegrin-like and metalloprotease (reprolysin type) with thrombospondin type 1 motif, 9

Yes

Calcium-dependentendopeptidases

Mmp14 matrix metalloproteinase 14 (membrane-inserted)

Yes Yes

Inhibitors of MMPsTimp3 tissue inhibitor of metalloproteinase 3 Yes YesTimp4 tissue inhibitor of metalloproteinase 4 Yes Yes

Structural components of extracellular matrix

Agc1 aggrecan YesCol5a3 procollagen, type V, alpha 3 Yes YesCol14a1 collagen, type XIV, alpha 1 YesCol1a2 collagen, type I, alpha 2 YesCol2a1 procollagen, type II, alpha 1 Yes YesFbn1 fibrillin 1 Yes YesFbn2 fibrillin 2 Yes Yes

Other regulators of extracellular matrix

Serpine1 serine (or cysteine) proteinase inhibitor, clade E, member 1

Yes Yes

Chst3 carbohydrate (chondroitin 6/keratan) sulfotransferase 3

Yes

Egf epidermal growth factor YesGnl3 guanine nucleotide binding protein-like 3

(nucleolar)Yes

Igfbp5 insulin-like growth factor binding protein 5 YesItga6 integrin alpha 6 YesLox lysyl oxidase Yes YesLoxl1 lysyl oxidase-like 1 YesLoxl2 lysyl oxidase-like 2 YesLoxl4 lysyl oxidase-like 4 Yes YesMgea5 meningioma expressed antigen 5

(hyaluronidase)Yes Yes

Nfatc2 nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 2

Yes Yes

P4ha1 procollagen-proline, 2-oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha 1 polypeptide

Yes Yes

Papss2 3'-phosphoadenosine 5'-phosphosulfate synthase 2

Yes Yes

Prss12 protease, serine 12 neurotrypsin (motopsin)

Yes

Pthlh parathyroid hormone-like peptide YesSox9 SRY-box containing gene 9 Yes YesSpon2 spondin 2, extracellular matrix protein Yes YesTgfa transforming growth factor alpha Yes YesTgfbr3 transforming growth factor, beta receptor

IIIYes

Wee1 wee 1 homolog (S. pombe) Yes Yes

Figure legends

Figure 1. The circadian timing system temporally controls tissue homeostasis in cartilage and bone.The master clock in the suprachiasmatic nuclei (SCN) receives light and other environmental zeitgebers (time cues) to entrain the internal timing system. The SCN then generate rhythmic signals (e.g. daily surges of hormones, physical activity/rest cycles and body temperature fluctuations) to synchronize the peripheral oscillators, including those in the skeletal system (such as cartilage and bone). Through rhythmic control of various biological pathways, the circadian clocks temporally coordinate tissue metabolism, homeostasis and integrity. During ageing or in diseases, loss of temporal segregation of the opposing events in skeletal tissues, such as the ECM degradation/repair in cartilage, or bone synthesis/resorption would disrupt the fine balance between catabolism and anabolism, predisposing the tissues to pathological changes.

Table 1. Rhythmic ECM-related genes in cartilage and bone.Summary of rhythmic genes that are related to ECM in xiphoid cartilage (Gossan et al., 2013; Honda et al., 2013) and calvarial bone (Zvonic et al., 2007). The circadian system in these matrix-rich skeletal tissues control key ECM-related targets, many of which are hallmark transcription factors, ECM remodeling enzymes and their inhibitors, key signaling pathways/regulators of ECM, or structural components of the particular tissue. Please note that the cartilage studies used tissues from mice kept in constant darkness (true circadian genes), while the calvarial bone study used tissues from mice kept in light/dark cycle. Therefore, some lack of concordance could be explained by different experimental conditions, choices of analysis algorithms or tissue-specificity.