Impact of inflammatory cytokines on longitudinal

Transcription

1 Page 1 of 32 Accepted Preprint first posted on 7 April 2014 as Manuscript JME Impact of inflammatory cytokines on longitudinal bone growth 3 Bettina Sederquist 1, Paola Fernandez-Vojvodich 1, Farasat Zaman 1,2, Lars Sävendahl 1* Department of Women s and Children s Health, Pediatric Endocrinology Unit Q2:08, Karolinska University Hospital, SE Stockholm, Sweden. 2 Developmental and Stem Cell Biology, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada *Corresponding author: Lars Sävendahl, Pediatric Endocrinology Unit Q2:08, Karolinska University Hospital, SE Stockholm, Sweden. Phone: +46 (0) , Fax +46 (0) lars.savendahl@ki.se Short title: Cytokines and bone growth Key words: Cytokines, Inflammation, Growth plate, Chondrocyte, Growth retardation 16 Word count: 4220 words 17 Copyright 2014 by the Society for Endocrinology.

2 Page 2 of Abstract Children with inflammatory diseases usually display abnormal growth patterns as well as delayed puberty. This is a result of several factors related to the disease itself, such as malnutrition, hypercortisolism, and elevated levels of pro-inflammatory cytokines. These factors in combination with glucocorticoid treatment contribute to growth retardation during chronic inflammation by systemically affecting the major regulator of growth, the growth hormone/insulin-like growth factor- 1 (GH/IGF-1) axis. However, recent studies also show evidence of a direct effect of these factors at the growth plate level. In conditions of chronic inflammation, pro-inflammatory cytokines are upregulated and released into circulation. The most abundant of these, tumor necrosis factor-α (TNFα), interleukin-1β (IL-1β) and interleukin-6 (IL-6), are all known to directly act on growth plate cartilage to induce apoptosis and thereby suppress bone growth. Both clinical and experimental studies have shown that growth retardation can partly be rescued when these cytokines are blocked. Therefore, therapy modulating the local actions of these cytokines may be effective for preventing growth failure in patients with chronic inflammatory disorders. In this review, we report the current knowledge of inflammatory cytokines and their role in regulating bone growth.

3 Page 3 of Normal regulation of bone growth The process of longitudinal bone growth is complex and tightly regulated by several factors. Hormones play a major role in the regulation of longitudinal bone growth and most central are the GH/IGF-1 axis, thyroid hormone, and sex steroids (Wit, et al. 2011). Nutrition is also an important regulator of growth, where poor nutrition leads to stunted growth (Gat-Yablonski, et al. 2011; Marcovecchio, et al. 2013) Longitudinal bone growth is the result of a process called endochondral ossification, by which the embryonic cartilaginous model of most bones is gradually replaced by calcified bone (Mackie, et al. 2008). Growth occurs at the growth plate, a thin layer situated at both ends between the diaphysis and the epiphysis of all long bones (Kronenberg 2003). The growth plate contains only one cell type, the chondrocyte, distributed at different levels of differentiation within the resting, proliferative and hypertrophic zones (Fig. 1). Chondrocytes are recruited from the resting layer to the proliferative layer, where they actively proliferate and thereafter undergo hypertrophy. The newly formed cartilage is invaded by blood vessels and bone cell precursors, which remodel the hypertrophic zone cartilage into bone (Mackie et al. 2008) The GH/IGF-1 axis is the major regulator of longitudinal bone growth (Fig. 1). GH is produced and secreted from the pituitary gland under the control of growth hormone-releasing hormone (GHRH) and plays an important role only in postnatal bone growth. GH deficiency leads to growth retardation, which has been seen in GH receptor knockout mice, while high levels of GH have the opposite effect by causing gigantism (Pass, et al. 2009). The systemic actions of GH are thought to be mediated by IGF-1 (Le Roith, et al. 2001). IGF-1 is produced systemically by the liver and plays an important role during both embryonic and postnatal growth in rodents and humans (Le Roith et al. 2001; Martensson, et al. 2004). IGF-1 knockout mice show growth retardation and low levels of IGF-1 in serum; this is also one of the most frequently observed abnormalities in patients with 1

4 Page 4 of chronic inflammatory diseases, and perhaps a reason for the often observed growth retardation in these patients (Lupu, et al. 2001; Simon 2010) The original somatomedin hypothesis states that GH only has an indirect effect on the growth plate by stimulating the production of IGF-1 from the liver, which in turn exerts its effects on the growth plate (Salmon, et al. 1957). Those experiments were performed both in vivo in rats as well as in vitro in costal cartilage (Le Roith et al. 2001). However, several studies have also reported a direct effect of GH on the growth plate (Isaksson, et al. 1982; Isgaard, et al. 1988). In a study by Wang and colleagues, GH was demonstrated to act both directly on resting/stem-like chondrocytes to stimulate proliferation, as well as indirectly, through IGF-1, to promote chondrocyte hypertrophy (Wang, et al. 2004). This observation does not discard the possibility that the direct effect might be mediated by an increase in the local production of IGF-1, which in turn stimulates growth plate chondrocytes. The growth hormone receptor can also be detected in the chondrocytes of all zones of the growth plate and, therefore, a direct effect of GH has been suggested (Parker, et al. 2007). Such a direct effect of GH is supported by a study where local injection of GH into the growth plate accelerated bone growth compared to the contralateral bone (Isaksson et al. 1982) IGF-1 has also been found to be produced by chondrocytes in the proliferative zone, and its expression is increased upon stimulation with GH (Nilsson, et al. 1986). This finding suggests that IGF-1 has a specific role in the differentiation of chondrocytes through autocrine/paracrine mechanisms. Mice with a targeted deletion of Igf-1 in chondrocytes have reduced body length, while serum IGF-1 levels are normal, which highlights the importance of locally produced IGF-1 in growth plate chondrocytes for the normal regulation of longitudinal bone growth (Govoni, et al. 2007). 2

5 Page 5 of Impact of inflammation on bone growth Growth is often impaired in children with chronic inflammatory diseases, such as inflammatory bowel disease (IBD), Crohn s disease (CD), ulcerative colitis (UC), and juvenile idiopathic arthritis (JIA). Growth retardation occurs in up to 56% of children with CD and up to 10% of children with UC (Abraham, et al. 2012), and several publications have reported short stature in JIA patients (Simon 2010). The degree of growth impairment is variable and can vary from a mild decrease in growth velocity to severe forms of short stature, which can be defined as a body height more than two standard deviations below the mean of the population. Children suffering from chronic inflammation are exposed to several factors contributing to the growth failure such as malnutrition, glucocorticoid (GC) therapy, and pro-inflammatory cytokines (Ahmed, et al. 2013; Ahmed, et al. 2009). Altogether, these factors act both systemically and locally at the growth plate level to suppress bone growth (Fig. 2) It is well known that nutritional problems contribute to growth retardation in children with inflammatory diseases and it is mainly a result of an imbalance between consumed calories and energy requirement. The degree of malnutrition seems to be correlated to disease severity (Simon 2010). However, according to experiments performed with pair-fed controls in a rat model of colitis, malnutrition should only account for about 60% of the final growth impairment, whereas the remaining loss in growth would result from the inflammatory process itself (Ballinger 2000). Elevated levels of circulating pro-inflammatory cytokines can reduce energy uptake and metabolism as well as induce protein catabolism. Studies in mice have shown a significant decrease in IGF-1 levels as well as a reduction of GH receptor expression in the growth plate during food restriction (Gat-Yablonski, et al. 2008; Pando, et al. 2012). Furthermore, malnutrition often leads to delayed sexual development which indirectly results in growth retardation linked to negative effects on the GH/IGF-1 axis (Ezri, et al. 2012). 3

6 Page 6 of During early onset of inflammation, the levels of circulating endogenous GCs are increased via activation of the hypothalamic-pituitary-adrenal (HPA) axis (Hardy, et al. 2012). GCs are steroid hormones that bind to and activate the glucocorticoid receptor (GR), playing an important role during early development. The growth-suppressing effects of GCs can be mediated through both systemic and local mechanisms. GCs suppress the GH/IGF-1 axis through inhibiting GH secretion (Mushtaq 2002) and down-regulating GH receptors in the liver, thereby inhibiting IGF-1 production and activity (Kritsch, et al. 2002). The effect of GCs on pubertal development has also been studied but the results vary and are still debated (Kinouchi, et al. 2012). Direct effects of GCs on growth plate cartilage include inhibition of chondrocyte proliferation and mineralization, as well as increased apoptosis (Chrysis, et al. 2002) Cytokines, normally up-regulated in conditions of chronic inflammation, can act individually or in combination to suppress longitudinal bone growth (MacRae, et al. 2006b). They not only have direct local effects at the growth plate, but can also act systemically by suppressing IGF-1 (MacRae, et al. 2006a; Martensson et al. 2004). Cytokines are signaling molecules that mediate inter- and intracellular communication and the most abundant cytokines that are up-regulated in inflammatory diseases like IBD and JIA are tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β) and interleukin-6 (IL-6) (MacRae et al. 2006b). Cytokines, such as IL-1 or TNF-α, have also been shown to inhibit the production of sex steroids by acting directly on the gonads or through suppression of the gonadotropin-releasing hormone (GnRH) secretion (Hong, et al. 2004) Systemic effects of cytokines have been reported in transgenic mice overexpressing TNF-α and IL-6 where both models show growth retardation (De Benedetti, et al. 1997; Li, et al. 2003). The IL-6 overexpressing mice exhibit a defective growth phenotype with a size reduction of 50-70% compared to normal non-transgenic littermates. The growth retardation in these mice was associated with decreased IGF-1 levels and IGF-1 binding protein 3 (IGFBP-3), yet normal distribution of GH pituitary cells and GH production (De Benedetti et al. 1997; De Benedetti, et al. 2001). This 4

7 Page 7 of finding is in line with the situation in patients with JIA where high circulating levels of IL-6 are negatively correlated with IGF-1 and IGFBP-3 levels (Gaspari, et al. 2011). On the other hand, IL-1β and IL-6 have been shown to have the ability to stimulate GH secretion while TNF-α does not seem to have this effect (Mainardi, et al. 2002). TNF-α and IL-1β also have the potential to directly act at the growth plate level by decreasing chondrocyte proliferation and hypertrophy as well as increasing apoptosis (Martensson et al. 2004). TNF-α and IL-1β have been shown to inhibit IRS-1, Akt, and MAPK-kinase phosphorylation and induce a state of IGF-1 resistance in several different cell types. However, little is still known about the mechanism in growth plate chondrocytes (Farquharson, et al. 2013; Pass et al. 2009). Generally IL-6 has been considered to act systemically suppressing growth by altering the GH/IGF-1 axis, although more recent studies show that IL-6 also has direct effects on the growth plate chondrocytes (Fernandez-Vojvodich, et al. 2013; Nakajima, et al. 2009) It is still unclear by which cellular mechanisms these pro-inflammatory cytokines act on the GH/IGF-1 axis and the growth plate chondrocytes. Both IL-1α and IL-1β bind to the plasma membrane receptor IL-1RI which is expressed in all three zones of the growth plate in fetal rat metatarsal bones as well as in mice (Martensson et al. 2004; Simsa-Maziel, et al. 2013). IL-1RI was recently reported by Simsa-Maziel and colleagues to be expressed in ATDC5 chondrocytes in all stages of differentiation (Simsa-Maziel et al. 2013). They further showed that IL-1β activates the p38 MAPK and NF-κB pathways in ATDC5 cells throughout the stages of proliferation and differentiation. In the ATDC5 chondrocyte cell line, IL-1β also reduces proteoglycan synthesis, cell number, as well as mrna expression of aggrecan, collagen II, and collagen X Lately, it has been suggested that members of the suppressor of cytokine signaling family (SOCS), especially SOCS2, may play a central role in cytokine induced growth retardation (Ahmed, et al. 2010). SOCS proteins are signaling proteins that have the ability to down-regulate cytokine and growth factor signaling by inhibiting the JAK/STAT signaling pathway. SOCS2 is a unique GH receptor inhibitor as shown by the knockout phenotype, which grows 30-40% bigger than its 5

8 Page 8 of normal littermates. SOCS2-knockout mice have increased longitudinal bone growth and body length as well as wider proliferative and hypertrophic growth plate zones (Pass et al. 2009; Pass, et al. 2012)

9 Page 9 of Pro-inflammatory cytokines and their action on the growth plate There are only a few studies that have reported local effects of pro-inflammatory cytokines at the growth plate level. These studies have mostly focused on TNF-α, IL-1β, and IL-6, and have reported evidence that individual cytokines have varying local effects on the growth plate chondrocytes in both organ and cell culture experiments. We first reported that TNF-α can suppress longitudinal bone growth through direct actions on the growth plate cartilage in a model of cultured fetal rat metatarsal bones (Martensson et al. 2004). TNF-α treated bones grew slower compared with control bones, but only when exposed to a high concentration (100 ng/ml) of TNF-α (Fig. 3a). The mechanisms behind this growth suppression included decreased chondrocyte proliferation and hypertrophy as well as increased apoptosis (Martensson et al. 2004). These findings were later supported by MacRae et al., who also reported growth suppression when metatarsal bones were exposed to TNF-α (MacRae et al. 2006a). TNF-α was found to also reduce proteoglycan synthesis, cell number, as well as mrna expression of aggrecan, collagen II, and collagen X, when studied in the ATDC5 chondrocyte cell line (MacRae et al. 2006a). We recently reported that TNF-α is produced endogenously throughout the growth plate and that treatment with etanercept, a soluble TNF receptor, leads to improved longitudinal bone growth of cultured fetal rat metatarsal bones (Fernandez-Vojvodich, et al. 2012). Altogether, locally produced TNF-α seems to play a role in the normal regulation of bone growth, while the negative effects of TNF-α on chondrocytes have only been seen at very high concentrations Supraphysiological concentrations of IL-1β have been reported to suppress longitudinal bone growth in cultured fetal rat metatarsal bones in the same way as TNF-α does, through direct actions on the growth plate cartilage (Fig. 3b) (Martensson et al. 2004). The mechanisms behind this effect include decreased chondrocyte proliferation, hypertrophy, and increased apoptosis. Treatment with anakinra, an IL-1 antagonist which suppresses the actions of IL- 1β, improves bone growth in this experimental model (Fernandez-Vojvodich et al. 2012). IL-1β was 7

10 Page 10 of reported to be endogenously produced in growth plate chondrocytes, suggesting that this cytokine also has a role in the normal regulation of bone growth (Fernandez-Vojvodich et al. 2012). Indeed, suppression of locally produced IL-1β with anakinra was found to improve bone growth. IL-1 receptor type I (IL-1RI) knockout mice were recently reported to have normal growth, although their growth plates were narrower and had higher proteoglycan content when compared with wild type animals (Simsa-Maziel et al. 2013). Altogether, IL-1β is produced by growth plate chondrocytes, and both this cytokine and its receptor seem to play a role in the normal regulation of bone growth. Furthermore, at high concentrations, IL-1β impairs bone growth by directly targeting the growth plate, an effect which can be prevented with specific anti-cytokine treatment In the studies with metatarsal bones mentioned previously, IL-6 did not have any effects on growth plate chondrocytes or growth (MacRae et al. 2006a; Martensson et al. 2004). However, contrary to the general consideration that IL-6 only affects growth systemically, Nakajima and co-workers showed in a later study that IL-6 inhibits the early differentiation of ATDC5 chondrocytes by inhibiting cartilaginous nodule formation and decreasing expression of type II collagen, aggrecan and type X collagen (Nakajima et al. 2009). Interestingly, when fetal metatarsal bones were cultured with IL-6 in combination with its soluble receptor IL-6 Rα, bone growth was clearly decreased (Fernandez-Vojvodich et al. 2013) (Fig. 3c). This growth retardation was associated with a decrease in cell density of the proliferative zone as well as decreases in length and area of the hypertrophic zone, thus further supporting a local effect of IL-6 on the growth plate

11 Page 11 of Interactions between pro-inflammatory cytokines Usually more than one cytokine is up-regulated in inflammatory diseases, and therefore growth retardation might be the result of a synergistic effect of two or more cytokines. Interestingly, IL-1β and TNF-α have been shown to act in synergy to decrease longitudinal growth (Martensson et al. 2004). When these cytokines were combined, growth inhibition was observed at far lower concentrations compared to when added separately. This synergistic effect could partly be prevented by treatment with anti-il-1β, anti-tnf-α, or IGF-1 (Martensson et al. 2004) (Fig. 4). Similarly MacRae and coworkers reported a 59% growth reduction in metatarsal bones treated with a combination of TNF-α and IL-1β. Catch-up growth has not been observed after the combined treatment with TNF-α and IL-1β, suggesting that even a short period of exposure to these cytokines may have an irreversible negative effect on longitudinal bone growth (MacRae et al. 2006a). It has also been shown that both anakinra and etanercept have the capacity to dose-dependently increase growth in rat metatarsal bones exposed to cytokines (Fernandez-Vojvodich, et al. 2011). Indeed, a combination of these biologics together with IGF-1 also showed improved longitudinal bone growth in cultured fetal metatarsal bones (Fernandez-Vojvodich et al. 2011). IL-6, when given in combination with IL-1β and TNF-α was found to reduce bone growth, however this effect was similar to the combination without IL-6, which indicates that IL-6 does not add to the synergism (Fernandez-Vojvodich et al. 2013) (Fig. 3c). However, the production of IL-6 has been shown to significantly increase in fetal rat metatarsal bones after combination treatment with IL-1β and TNF-α (Fernandez-Vojvodich et al. 2013) (Fig. 5). This may explain the synergistic growth suppressive effect of IL-1β and TNF-α reported earlier (MacRae et al. 2006a; Martensson et al. 2004)

12 Page 12 of Impact of glucocorticoid treatment on bone growth Besides the endogenous GCs that increase in conditions of chronic inflammation, exogenous GCs like dexamethasone or prednisolone are commonly used clinically because of their high therapeutic efficacy as anti-inflammatory or immunosuppressant agents. It has been estimated that about 5-10% of children require some form of GC treatment during their childhood (Warner 1995) GCs may have beneficial effects on bone growth in patients with chronic inflammation thanks to their effective suppression of cytokine levels and inflammation (Marcovecchio, et al. 2012). However, long-term GC therapy is associated with severe negative side-effects including osteoporosis and growth impairment (Deshmukh 2007) Interestingly, growth retardation was already reported in patients with JIA before corticosteroids became commonly used as anti-inflammatory drugs (Czernichow 2009). Furthermore, growth impairment has been reported in systemic and polyarticular JIA patients never treated with corticosteroids (Polito, et al. 1997). This suggests that other factors, not only GCs, contribute to the deleterious effects of chronic inflammation on bone growth The systemic effects of GCs are well known, and even topically administered dexamethasone eye drops have been demonstrated to dose-dependently inhibit bone growth and negatively influence several other bone parameters in young rabbits (Kugelberg, et al. 2005). Evidences for local effects of GCs on the growth plate come from both in vitro and in vivo studies. In vitro experiments have shown that the GC dexamethasone directs the differentiation of mesoblastic precursor cells into mature chondrocytes by up-regulating SOX9, the transcription factor that determines chondrogenesis (Locker, et al. 2004). SOX9 activates SOX5 and SOX6 to trigger early stages of differentiation when chondrocytes proliferate to form columns, and additionally suppresses their terminal differentiation (Ikeda, et al. 2004). In contrast, high dose GC treatment has also been reported to induce undesired cell death in growth plate chondrocytes causing growth retardation in 10

13 Page 13 of young rats (Chrysis, et al. 2003). The first evidence of apoptosis in growth plate chondrocytes after GC treatment was reported by the detection of TUNEL-positive cells, increased Bax, as well as decreased expression of Bcl-2 and parathyroid hormone related hormone (PTHrP) (Chrysis et al. 2003; Sanchez, et al. 2002). GC-induced apoptosis in chondrocytes is mainly regulated through activation of the caspase cascade, which includes caspase-8, -9, and suppression of the Aktphosphatidylinositol 3-kinase (PI3K) signaling pathway (Chrysis, et al. 2005; Macrae, et al. 2007; Zaman, et al. 2014; Zaman, et al. 2012). In addition, it has also been reported that GC treatment differentially regulates Bcl-2 family proteins such as Bax, Bid, and Bak to trigger undesired apoptosis in proliferative chondrocytes (Zaman et al. 2014). Interestingly, young mice lacking Bax are protected from GC-induced bone growth retardation (Zaman et al. 2012), which suggests a physiological role of the apoptotic machinery in the regulation of bone growth. In children treated with GCs, permanent growth retardation as well as partial catch-up growth have been reported after cessation of treatment (Bechtold, et al. 2009; Simon, et al. 2002). Patients with cystic fibrosis show prolonged and permanent growth retardation after GC treatment (Lai, et al. 2000). The potential for catch-up growth seems to be highly dependent on the dose as well as the duration of the GC treatment (Bechtold and Roth 2009). In a model of ex vivo cultured postnatal rat metatarsal bones, partial catch-up growth was also demonstrated after cessation of GC exposure (Chagin, et al. 2010)

14 Page 14 of Prevention of growth retardation caused by chronic inflammation Immunomodulatory biological drugs offer a relatively new type of anti-inflammatory treatment consisting of proteins that selectively inhibit the effects of cytokines. These drugs, targeting TNF-α, IL-1β or IL-6, have shown promising results and are potential therapies that can restore longitudinal bone growth in children suffering from chronic inflammatory diseases. This has resulted in fewer patients with chronic inflammatory disorders suffering from growth retardation Several studies have reported significant decrease in disease activity and improved bone growth in patients treated with different anti-tnf agents (Fernandez-Vojvodich, et al. 2007; Malik, et al. 2012; Schmeling, et al. 2003; Tynjala, et al. 2006). Etanercept, a recombinant fusion protein based on the p75-receptor for TNF (TNF-R2) and the Fc-part of human igg1, acts as a soluble receptor trough TNF-driven inflammation, which plays a key role in the arthritic process (Lovell, et al. 2008). Etanercept treatment was reported to improve both disease activity and growth in a small mixed population of seven prepubertal and pubertal girls with refractory JIA and growth retardation (Schmeling et al. 2003). Those effects were accompanied by a discontinuation of oral GCs, a reduction of circulating IL-6, and increased levels of IGF-1 and IGFBP-3. In another study, etanercept improved linear bone growth in a majority of prepubertal and pubertal patients with JIA who were not responding to conventional therapy (Fernandez-Vojvodich et al. 2007). Furthermore, etanercept has been reported to improve bone mineral density status in those JIA children where disease activity was suppressed by the treatment (Simonini, et al. 2005). Other TNF-α antagonists like infliximab, a chimeric anti-tnf antibody, and adalimumab, a fully humanized TNF-antibody, have also been shown to clinically improve growth in children with CD (Altowati, et al. 2013; Malik et al. 2012) Anti-IL-1 agents have also shown promising effects in treating inflammatory conditions. Anakinra is a recombinant form of the human IL-1 receptor antagonist which competes with IL-1 when binding to the IL-1RI. Anakinra is frequently used for the treatment of rheumatoid 12

15 Page 15 of arthritis in adults, while in children, it has been used less often and preferably for the treatment of systemic onset JIA (Pascual, et al. 2005) Even though the biological drugs have revolutionized the treatment of several chronic inflammatory conditions in both adults and children, other drugs such as methotrexate are still being used as a first choice or in combination with anti-cytokine treatments GH treatment has been shown to improve height velocity and lean body mass in children with severe JIA and could, therefore, potentially be used to improve growth in these patients (Simon, et al. 2007). In another study, GH treatment was reported to be most effective at rescuing bone growth in those childhood JIA patients with moderate disease activity (Bechtold, et al. 2007) There has been much interest in developing selective GR modulators, a novel class of drugs maintaining the anti-inflammatory properties of GCs while limiting their negative side-effects (Owen, et al. 2007). AL-438 is a specific non-steroidal ligand for GR which has full anti-inflammatory effect, but reduced negative side-effects on osteoblasts and metatarsal bone growth when studied in vitro (Owen et al. 2007). Whether selective GR modulators can be used in vivo to effectively suppress inflammation without compromising longitudinal bone growth has not yet been documented

16 Page 16 of Conclusion and future perspectives Conditions of chronic inflammation result in several events that have a negative effect on longitudinal bone growth. Malnutrition is commonly seen in patients with inflammatory diseases and is known to impair bone growth. Growth is further suppressed by the use of GCs as antiinflammatory drugs. Elevated levels of pro-inflammatory cytokines, such as TNF-α, IL-1β and IL-6, also contribute to growth retardation through both systemic actions on the GH/IGF-1 axis as well as local actions on the growth plate. High concentrations of these cytokines suppress growth dosedependently by decreasing chondrocyte proliferation and hypertrophy while increasing apoptosis. The inhibition of these cytokines with biological drugs has, on the other hand, been shown to increase growth in both experimental and clinical studies. Cytokines are also produced locally by growth plate chondrocytes, indicating a function in the normal regulation of bone growth. Although cytokines are important, other yet unidentified factors may contribute in the pathogenesis of longbone growth retardation in patients with chronic inflammatory disorders Even though the use of biologics has resulted in fewer patients with severe growth retardation, growth impairment is still an issue in patients with more severe forms of chronic inflammatory diseases where malnutrition and high-dose GC treatment may contribute. The development of new specific cytokine inhibitors or antagonists is therefore important to enable a more effective prevention of cytokine induced growth retardation. Another future perspective is the possibility of identifying new targets to prevent GC-induced growth retardation Today, the most effective way to reduce growth retardation in children with chronic inflammation is to control the inflammation with the current drugs available, while at the same time reduce the duration of treatment as well as the dosage. It is also important to have an early diagnosis, since attempts to rescue bone growth in children has to start before epiphyseal closure. 14

17 Page 17 of Declaration of interest: None of the authors report any conflict of interest Funding: This work was supported by the Swedish Research Council (project X ), Karolinska Institutet, Stiftelsen Frimurare Barnhuset and Sällskapet Barnavård Author contributions: BS drafted the first version of this manuscript and its original illustrations. PFV, FZ, and LS actively contributed throughout the production by reviewing, editing, and proofreading Acknowledgements: We would like to thank our co-workers Katja Sundström and Therese Cedervall for valuable advice and constructive criticism

18 Page 18 of References Abraham BP, Mehta S & El-Serag HB 2012 Natural history of pediatric-onset inflammatory bowel disease: a systematic review. Journal of Clinical Gastroenterology Ahmed SF & Farquharson C 2010 The effect of GH and IGF1 on linear growth and skeletal development and their modulation by SOCS proteins. The Journal of Endocrinology (doi: /joe ) Ahmed SF, Farquharson C, McGrogan P & Russell RK 2013 Pathophysiology and management of abnormal growth in children with chronic inflammatory bowel disease. World Review of Nutrition and Dietetics (doi: / ) Ahmed SF & Savendahl L 2009 Promoting growth in chronic inflammatory disease: lessons from studies of the growth plate. Hormone Research 72 Suppl (doi: / ) Altowati MA, Russell RK & Ahmed SF 2013 Endocrine therapy for growth retardation in paediatric inflammatory bowel disease. Paediatric Drugs. (doi: /s ) Ballinger AB 2000 Growth failure occurs through a decrease in insulin-like growth factor 1 which is independent of undernutrition in a rat model of colitis. Gut (doi: /gut ) Bechtold S, Ripperger P, Dalla Pozza R, Bonfig W, Hafner R, Michels H & Schwarz HP 2007 Growth hormone increases final height in patients with juvenile idiopathic arthritis: data from a randomized controlled study. The Journal of Clinical Endocrinology and Metabolism (doi: /jc ) Bechtold S & Roth J 2009 Natural history of growth and body composition in juvenile idiopathic arthritis. Hormone Research 72 Suppl (doi: / ) Chagin AS, Karimian E, Sundstrom K, Eriksson E & Savendahl L 2010 Catch-up growth after dexamethasone withdrawal occurs in cultured postnatal rat metatarsal bones. The Journal of endocrinology (doi: /joe ) Chrysis D, M. RE & Sävendahl L 2003 Growth retardation induced by dexamethasone is associated with increased apoptosis of the growth plate chondrocytes. Journal of Endocrinoloy (doi: /joe ) Chrysis D, Nilsson O, Ritzen EM & Sävendahl L 2002 Apoptosis is developmentally regulated in rat growth plate. Endocrine (doi: /ENDO:18:3:271) Chrysis D, Zaman F, Chagin AS, Takigawa M & Savendahl L 2005 Dexamethasone induces apoptosis in proliferative chondrocytes through activation of caspases and suppression of the Aktphosphatidylinositol 3'-kinase signaling pathway. Endocrinology (doi: /en ) Czernichow P 2009 Growth and development abnormalities in children with juvenile idiopathic arthritis: treatment and prevention. Hormone Research 72 Suppl (doi: / ) De Benedetti F, Alonzi T, Moretta A, Lazzaro D, Costa P, Poli V, Martini A, Ciliberto G & Fattori E 1997 Interleukin 6 causes growth impairment in transgenic mice through a decrease in insulin-like growth 16

19 Page 19 of factor-i. A model for stunted growth in children with chronic inflammation. The Journal of Clinical Investigation. (doi: /JCI119207) De Benedetti F, Meazza C, Oliveri M, Pignatti P, Vivarelli M, Alonzi T, Fattori E, Garrone S, Barreca A & Martini A 2001 Effect of IL-6 on IGF binding protein-3: A study in IL-6 transgenic mice and in patients with systemic juvenile idiopathic arthritis. Endocrinology Deshmukh CT 2007 Minimizing side effects of systemic corticosteroids in children. Indian Journal of Dermatology, Venereology and Leprology (doi: / ) Ezri J, Marques-Vidal P & Nydegger A 2012 Impact of disease and treatments on growth and puberty of pediatric patients with inflammatory bowel disease. Digestion (doi: / ) Farquharson C & Ahmed SF 2013 Inflammation and linear bone growth: the inhibitory role of SOCS2 on GH/IGF-1 signaling. Pediatric Nephrology (doi: /s ) Fernandez-Vojvodich P, Hansen JB, Andersson U, Savendahl L & Hagelberg S 2007 Etanercept Treatment Improves Longitudinal Growth in Prepubertal Children with Juvenile Idiopathic Arthritis. The Journal of Rheumatology Fernandez-Vojvodich P, Karimian E & Savendahl L 2011 The biologics anakinra and etanercept prevent cytokine-induced growth retardation in cultured fetal rat metatarsal bones. Hormone Research in Paediatrics (doi: / ) Fernandez-Vojvodich P, Palmblad K, Karimian E, Andersson U & Sävendahl L 2012 Pro-inflammatory cytokines produced by growth plate chondrocytes may act locally to modulate longitudinal bone growth. Hormone Research in Paediatrics (doi: / ) Fernandez-Vojvodich P, Zaman F & Savendahl L 2013 Interleukin-6 acts locally on the growth plate to impair bone growth. Annals of the Rheumatic Diseases 72 e24-e24. (doi: /annrheumdis ) Gaspari S, Marcovecchio ML, Breda L & Chiarelli F 2011 Growth in juvenile idiopathic arthritis: the role of inflammation. Clincal and Experimental Rheumatology Gat-Yablonski G, Shtaif B, Abraham E & Phillip M 2008 Nutrition-induced catch-up growth at the growth plate. Journal of Pediatric Endocrinology & Metabolism (doi: / ) Gat-Yablonski G, Yackobovitch-Gavan M & Phillip M 2011 Nutrition and bone growth in pediatrics. Pediatric Clinics of North America , ix. (doi: /j.pcl ) Govoni KE, Lee SK, Chung YS, Behringer RR, Wergedal JE, Baylink DJ & Mohan S 2007 Disruption of insulin-like growth factor-i expression in type IIalphaI collagen-expressing cells reduces bone length and width in mice. Physiological Genomics (doi: /physiolgenomics ) Hardy RS, Raza K & Cooper MS 2012 Endogenous glucocorticoids in inflammation: contributions of systemic and local responses. Swiss Medical Weekly 142 w (doi: /smw ) Hong CY, Park JH, Ahn RS, Im SY, Choi HS, Soh J, Mellon SH & Lee K 2004 Molecular mechanism of suppression of testicular steroidogenesis by proinflammatory cytokine tumor necrosis factor alpha. Molecular and Cellular Biology

20 Page 20 of Ikeda T, Kamekura S, Mabuchi A, Kou I, Seki S, Takato T, Nakamura K, Kawaguchi H, Ikegawa S & Chung UI 2004 The combination of SOX5, SOX6, and SOX9 (the SOX trio) provides signals sufficient for induction of permanent cartilage. Arthritis and Rheumatism (doi: /art.20611) Isaksson OG, Jansson JO & Gause IA 1982 Growth hormone stimulates longitudinal bone growth directly. Science Isgaard J, Moller C, Isaksson OG, Nilsson A, Mathews LS & Norstedt G 1988 Regulation of insulin-like growth factor messenger ribonucleic acid in rat growth plate by growth hormone. Endocrinology Kinouchi R, Matsuzaki T, Iwasa T, Gereltsetseg G, Nakazawa H, Kunimi K, Kuwahara A, Yasui T & Irahara M 2012 Prepubertal exposure to glucocorticoid delays puberty independent of the hypothalamic Kiss1-GnRH system in female rats. International Journal of Developmental Neuroscience (doi: /j.ijdevneu ) Kritsch KR, Murali S, Adamo ML & Ney DM 2002 Dexamethasone decreases serum and liver IGF-I and maintains liver IGF-I mrna in parenterally fed rats. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology 282 R (doi: /ajpregu ) Kronenberg HM 2003 Developmental regulation of the growth plate. Nature (doi: /nature01657) Kugelberg M, Ohlsson C & Savendahl L 2005 Reduced bone mineral density and radial bone growth in young rabbits treated with dexamethasone eye drops. Hormone Research (doi: / ) Lai HC, FitzSimmons SC, Allen DB, Kosorok MR, Rosenstein BJ, Campbell PW & Farrell PM 2000 Risk of persistent growth impairment after alternate-day prednisone treatment in children with cystic fibrosis. The New England Journal of Medicine (doi: /nejm ) Le Roith D, Bondy C, Yakar S, Liu JL & Butler A 2001 The somatomedin hypothesis: Endocrine Reviews Li P & Schwarz EM 2003 The TNF-alpha transgenic mouse model of inflammatory arthritis. Springer Seminars in Immunopathology (doi: /s ) Locker M, Kellermann O, Boucquey M, Khun H, Huerre M & Poliard A 2004 Paracrine and autocrine signals promoting full chondrogenic differentiation of a mesoblastic cell line. Journal of Bone and Mineral Research (doi: /JBMR ) Lovell DJ, Reiff A, Ilowite NT, Wallace CA, Chon Y, Lin SL, Baumgartner SW, Giannini EH & Pediatric Rheumatology Collaborative Study G 2008 Safety and efficacy of up to eight years of continuous etanercept therapy in patients with juvenile rheumatoid arthritis. Arthritis and Rheumatism (doi: /art.23427) Lupu F, Terwilliger JD, Lee K, Segre GV & Efstratiadis A 2001 Roles of growth hormone and insulin-like growth factor 1 in mouse postnatal growth. Developmental Biology (doi: /dbio ) 18

21 Page 21 of Mackie EJ, Ahmed YA, Tatarczuch L, Chen KS & Mirams M 2008 Endochondral ossification: how cartilage is converted into bone in the developing skeleton. The International Journal of Biochemistry & Cell Biology (doi: /j.biocel ) Macrae VE, Ahmed SF, Mushtaq T & Farquharson C 2007 IGF-I signalling in bone growth: inhibitory actions of dexamethasone and IL-1beta. Growth Hormone & IGF Research (doi: /j.ghir ) MacRae VE, Farquharson C & Ahmed SF 2006a The restricted potential for recovery of growth plate chondrogenesis and longitudinal bone growth following exposure to pro-inflammatory cytokines. The Journal of Endocrinology (doi: /joe ) MacRae VE, Wong SC, Farquharson C & Ahmed SF 2006b Cytokine actions in growth disorders associated with pediatric chronic inflammatory diseases (review). International Journal of Molecular Medicine Mainardi GL, Saleri R, Tamanini C & Baratta M 2002 Effects of Interleukin-1-Beta, Interleukin-6 and Tumor Necrosis Factor-Alpha, alone or in association with hexarelin or galanin, on growth hormone gene expression and growth hormone release from pig pituitary cells. Hormone Research (doi: / ) Malik S, Ahmed SF, Wilson ML, Shah N, Loganathan S, Naik S, Bourke B, Thomas A, Akobeng AK, Fagbemi A, et al The effects of anti-tnf-alpha treatment with adalimumab on growth in children with Crohn's disease (CD). Journal of Crohn's & colitis (doi: /j.crohns ) Marcovecchio ML & Chiarelli F 2013 Obesity and growth during childhood and puberty. World Review of Nutrition and Dietetics (doi: / ) Marcovecchio ML, Mohn A & Chiarelli F 2012 Inflammatory cytokines and growth in childhood. Current Opinion in Endocrinology, Diabetes, and Obesity (doi: /MED.0b013e32834ed61f) Martensson K, Chrysis D & Savendahl L 2004 Interleukin-1beta and TNF-alpha act in synergy to inhibit longitudinal growth in fetal rat metatarsal bones. Journal of Bone and Mineral Research (doi: /JBMR ) Mushtaq T 2002 The impact of corticosteroids on growth and bone health. Archives of Disease in Childhood (doi: /adc ) Nakajima S, Naruto T, Miyamae T, Imagawa T, Mori M, Nishimaki S & Yokota S 2009 Interleukin-6 inhibits early differentiation of ATDC5 chondrogenic progenitor cells. Cytokine (doi: /j.cyto ) Nilsson A, Isgaard J, Lindahl A, Dahlstrom A, Skottner A & Isaksson OG 1986 Regulation by growth hormone of number of chondrocytes containing IGF-I in rat growth plate. Science Owen HC, Miner JN, Ahmed SF & Farquharson C 2007 The growth plate sparing effects of the selective glucocorticoid receptor modulator, AL-438. Molecular and Cellular Endocrinology (doi: /j.mce ) 19

22 Page 22 of Pando R, Even-Zohar N, Shtaif B, Edry L, Shomron N, Phillip M & Gat-Yablonski G 2012 MicroRNAs in the growth plate are responsive to nutritional cues: association between mir-140 and SIRT1. The Journal of Nutritional Biochemistry (doi: /j.jnutbio ) Parker EA, Hegde A, Buckley M, Barnes KM, Baron J & Nilsson O 2007 Spatial and temporal regulation of GH-IGF-related gene expression in growth plate cartilage. The Journal of Endocrinology (doi: /joe ) Pascual V, Allantaz F, Arce E, Punaro M & Banchereau J 2005 Role of interleukin-1 (IL-1) in the pathogenesis of systemic onset juvenile idiopathic arthritis and clinical response to IL-1 blockade. The Journal of Experimental Medicine (doi: /jem ) Pass C, MacRae VE, Ahmed SF & Farquharson C 2009 Inflammatory cytokines and the GH/IGF-I axis: novel actions on bone growth. Cell Biochemistry and Function (doi: /cbf.1551) Pass C, MacRae VE, Huesa C, Faisal Ahmed S & Farquharson C 2012 SOCS2 is the critical regulator of GH action in murine growth plate chondrogenesis. Journal of Bone and Mineral Research (doi: /jbmr.1544) Polito C, Strano CG, Olivieri AN, Alessio M, Iammarrone CS, Todisco N & Papale MR 1997 Growth retardation in non-steroid treated juvenile rheumatoid arthritis. Scandinavian Journal of Rheumatology Salmon WD & Daughaday WH 1957 A hormonally controlled serum factor which stimulates sulfate incorporation by cartilage in vitro. Journal of Laboratory and Clinical Medicine Sanchez CP & He YZ 2002 Alterations in the growth plate cartilage of rats with renal failure receiving corticosteroid therapy. Bone Schmeling H, Seliger E & Horneff G 2003 Growth reconstitution in juvenile idiopathic arthritis treated with etanercept. Clinical and Experimental Rheumatology Simon D 2010 Inflammation and growth. Journal of Pediatric Gastroenterology and Nutrition 51 Suppl 3 S (doi: /MPG.0b013e3181f7feef) Simon D, Fernando C, Czernichow P & Prieur A-M 2002 Linear Growth and Final Height in Patients with Systemic Juvenile Idiopathic Arthritis Treated with Longterm Glucocorticoids. The Journal of Rheumatology Simon D, Prieur AM, Quartier P, Charles Ruiz J & Czernichow P 2007 Early recombinant human growth hormone treatment in glucocorticoid-treated children with juvenile idiopathic arthritis: a 3- year randomized study. The Journal of Clinical Endocrinology and Metabolism (doi: /jc ) Simonini G, Giani T, Stagi S, de Martino M & Falcini F 2005 Bone status over 1 yr of etanercept treatment in juvenile idiopathic arthritis. Rheumatology (Oxford) (doi: /rheumatology/keh592) Simsa-Maziel S, Zaretsky J, Reich A, Koren Y, Shahar R & Monsonego-Ornan E 2013 IL-1RI participates in normal growth plate development and bone modeling. American Journal of Physiology. Endocrinology and Metabolism 305 E (doi: /ajpendo ) 20

23 Page 23 of Tynjala P, Lahdenne P, Vahasalo P, Kautiainen H & Honkanen V 2006 Impact of anti-tnf treatment on growth in severe juvenile idiopathic arthritis. Annals of the Rheumatic Diseases (doi: /ard ) Wang J, Zhou J, Cheng CM, Kopchick JJ & Bondy CA 2004 Evidence supporting dual, IGF-Iindependent and IGF-I-dependent, roles for GH in promoting longitudinal bone growth. The Journal of Endocrinology Warner JO 1995 Review of prescribed treatment for children with asthma in BMJ Wit JM & Camacho-Hubner C 2011 Endocrine regulation of longitudinal bone growth. Endocrine Development (doi: / ) Zaman F, Chrysis D, Huntjens K, Chagin A, Takigawa M, Fadeel B & Savendahl L 2014 Dexamethasone differentially regulates Bcl-2 family proteins in human proliferative chondrocytes: Role of proapoptotic Bid. Toxicology Letters (doi: /j.toxlet ) Zaman F, Chrysis D, Huntjens K, Fadeel B & Savendahl L 2012 Ablation of the pro-apoptotic protein Bax protects mice from glucocorticoid-induced bone growth impairment. PLoS One 7 e (doi: /journal.pone )

24 Page 24 of Figure legends 546 Figure The growth plate ultrastructure, and the GH/IGF-1 axis, a major regulator of longitudinal bone growth. GH has both direct effects on the growth plate as well as indirect effects through IGF-1. The growth plate is a complex structure consisting of chondrocytes at different stages of differentiation. It is divided into three specialized zones: the resting zone, proliferative zone, and hypertrophic zone. 551 Figure The pathophysiology of growth impairment in conditions of chronic inflammation. Several factors have negative effects on longitudinal bone growth. 554 Figure 3 a, b, c TNF-α, IL-1β and IL-6 impair longitudinal bone growth in fetal rat metatarsal bones. TNF-α (a) and IL- 1β (b) have a dose-dependent effect. IL-6 together with IL-6 Rα impair bone growth in fetal rat metatarsal bones (c). TNF-α and IL-1β have a synergistic effect, but the involvement of IL-6 does not have any further synergistic effect (c). Fig. 3a and b are adapted from Martensson et al. (Martensson et al. 2004), with permission of the American Society for Bone and Mineral Research ASBMR. 561 Figure The synergistic effect of TNF-α and IL-1β can be reduced with anti-cytokine treatment. The figure is adapted from Martensson et al. (Martensson et al. 2004) with permission of the American Society for Bone and Mineral Research ASBMR

25 Page 25 of Figure IL-6 production is significantly increased in bones treated with a combination of TNF-α and IL-1β. The figure is reproduced from Fernandez-Vojvodich et al. (Fernandez-Vojvodich et al. 2013) with permission from BMJ Publishing Group Limited BMJ Publishing Group Limited. 23

26 Page 26 of 32 Resting zone GH Proliferative zone GH IGF-1 Hypertrophic zone