Activin Receptor Signaling: a Potential Therapeutic Target for Osteoporosis

Home , ACVR1C, Lefty (protein), Osteoblast

Current Molecular Pharmacology, 2012, 5, 195-204 195 Activin Receptor Signaling: A Potential Therapeutic Target for Osteoporosis

Sutada Lotinun*,1, R. Scott Pearsall2, William C. Horne1 and Roland Baron1,3

1Department of Oral Medicine, Infection and Immunity, Harvard School of Dental Medicine, Boston, MA 02115, USA; 2Acceleron Pharma, Inc., 128 Sidney Street, Cambridge, MA 02139, USA; 3Endocrine Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA Abstract: Current antiresorptive therapies not only prevent bone loss by decreasing osteoclastic bone resorption but also inhibit bone formation. Dual anabolic antiresorptive agents may be required to cure severe osteoporosis by preventing further bone loss and increasing bone mass to normal levels. Recent studies have demonstrated that activin signaling plays a crucial role in the skeleton. Activins, like other TGF-β superfamily members, transduce their signals through type I and II receptor serine/threonine kinases. The binding of activins to activin type IIA (ActRIIA) or type IIB (ActRIIB) receptors induces the recruitment and phosphorylation of an activin type I receptor (ALK4 and/or ALK7), which then phosphorylates the Smad2 and Smad3 intracellular signaling proteins. Activin signaling is down-regulated by inhibins, follistatin and other proteins, which antagonize activin signaling by a variety of mechanisms. A soluble chimeric protein composed of the extracellular domain of ActRIIA fused to IgG-Fc binds to circulating ligands such as activin A and prevents signaling through the endogenous receptor. In cynomolgus monkeys, the ActRIIA soluble receptor increases bone volume by decreasing bone resorption and increasing bone formation, leading to enhanced mechanical strength and bone quality. In addition, a single dose of the soluble ActRIIA-Fc fusion protein increased serum BSALP and PINP and decreased serum CTX and TRACP 5b in postmenopausal women. These data provide evidence of a dual anabolic antiresorptive effect of the soluble ActRIIA-Fc fusion protein in the skeleton. Therefore, targeting activin receptor signaling may be useful for therapeutic intervention in osteoporosis. Keywords: Activin, follistatin, inhibin, osteoblast, osteoclast, osteoporosis.

INTRODUCTION available in the United State is the 1-34 fragment of recombinant human parathyroid hormone (PTH) or Osteoporosis is one of the most severe problems teriparatide. PTH dramatically increases bone mass and affecting the quality of life in the elderly. There are more reduces fracture risk. However, the use of teriparatide as a than 1 million osteoporotic fractures per year in the United treatment for osteoporosis is limited to 24 months in the States, the most devastating of which are hip fractures. United State and 18 months in Europe [5]. Several novel Approximately 20% of individuals over the age of 50 years anabolic agents targeting various signaling pathways are die within a year of a hip fracture, and many others require currently in development. Increasing canonical Wnt long-term care in nursing homes. Current treatments for signaling using antibodies that prevent the binding of osteoporosis are inadequate due to limited efficacy; negative regulators of Wnt signaling, such as sclerostin and approximately 50% of patients taking the presently approved dickkopf-1 (DKK-1), to the co-receptor lipoprotein-receptor- drugs suffer a subsequent fracture [1]. The antiresorptive related protein (LRP) 5/6 has emerged as a promising agents used in the treatment of osteoporosis prevent bone strategy to increase bone mass [6-8]. However, concerns resorption but also inhibit bone formation. Long-term about the long-term safety of Wnt inhibitors have not been treatment with bisphosphonates is associated with an addressed. impairment of bone formation and osteonecrosis of the jaw The activin/inhibin/follistatin system has also been and may also lead to reduced mechanical strength due to implicated in the regulation of bone homeostasis and age- accumulated strain-induced microfractures in the absence of bone remodeling [2-4]. In spite of the widely appreciated related bone loss [9], and it has been suggested that decreased gonadal inhibin secretion and the consequent magnitude of this problem, there is still a critical gap in our increase in activin activity is an important contributory factor understanding of how to effectively reverse bone loss in the in the perimenopausal increase in bone resorption [10]. elderly. Consistent with this, it has recently been shown in animal Due to the limitations of antiresorptive agents and the models [11-13] and a phase I clinical trial [14] that a soluble therapeutic value of inducing bone formation, particularly in ActRIIA-Fc fusion protein represents a potential anabolic severe osteoporosis, a major effort has focused on agent. identifying and developing agents that exert an anabolic In this review, we focus on recent advances in the effect on bone. The only approved bone anabolic agent elucidation of activin signaling, with a particular emphasis

on activin interactions with their receptors, activin

*Address correspondence to this author at the Department of Oral Medicine, antagonists and the effects of activins on bone turnover. We Infection and Immunity, Harvard School of Dental Medicine, Boston, MA also discuss the evidence that the soluble ActRIIA-Fc fusion 02115, USA; Tel: (617) 432-7328; Fax: (617) 432-1897; protein is a promising potential therapeutic agent for treating E-mail: [email protected] osteoporosis and related diseases.

ACTIVIN STRUCTURE AND SIGNALING are able to bind ligands in the absence of type I receptor. However, they are unable to initiate signal transduction Activin Isoforms without forming a complex with an activin type I receptor. Activins are members of the transforming growth factor- Canonical activin signaling involves activation of Smad β (TGF-β) superfamily that elicit diverse biological family proteins and translocation to the nucleus (Fig. 1). The responses in reproductive, hematopoietic, immune, central high affinity binding of activins to ActRIIA or ActRIIB nervous and musculoskeletal system [15-16]. Other members receptors induces the recruitment ALK4 and/or ALK7 and in the TGF-β superfamily include TGF-βs, bone phosphorylation of the regulatory GS domain, followed by morphogenetic proteins (BMPs), myostatin, nodal, growth activation of Smad2 and Smad3. The receptor-activated and differentiation factors (GDFs), anti-Muellerian hormone Smads subsequently form a complex with the common co- (AMH) and others [17-18]. Activins were first purified as Smad (Smad4) allowing for translocation into the nucleus dimers consisting of disulfide-linked β subunits and where they bind to the promoter sequences of target genes identified as local mediators capable of stimulating pituitary and regulate gene transcription and cellular function. In follicle-stimulating hormone (FSH) release [19]. Four addition to the canonical signaling pathway, other non- different β -subunits of activin (βA, βB, βC and βE) have canonical or Smad-independent effectors, including ERK1/2, been described in mammals and a βD subunit has been JNK and p38 MAPK, PKC, Akt, IκB-α and Wnt/β-catenin identified in Xenopus [20-22]. The βA and βB subunits are have been reported in several tissues [27-33]. However, the detectable in most tissues with high expression in link between the activated receptor complexes and non- reproductive organs whereas the βC and βE subunits are canonical signaling remains to be defined. These signaling predominantly expressed in liver. Homo- and cascades may modulate the Smad-dependent responses, or heterodimerization of βA and βB subunits gives rise to three they may act on distinct downstream effectors in different biologically active glycoproteins, including activin A (βA- cell types. βA), activin B (βB- βB) and activin AB (βA- βB). ACTIVIN ANTAGONISTS Activin Receptors and Signal Transduction Activin signaling is regulated by multiple factors that act Activins, like other TGF-β family members, initiate their at the intracellular, membrane or extracellular levels. signal transduction cascades through two types of single Intracellularly, the inhibitory Smad7 prevents activin- transmembrane serine-threonine kinase receptors, called type mediated phosphorylation of Smad2 and Smad3, thereby I and type II receptors, which are approximately 55 and 70 blocking the signaling pathway [34]. The binding of activin kDa, respectively. Upon the ligand binding to receptors, two to its receptor is antagonized by inhibins, which block the type I and two type II receptors form a tetrameric complex. activin binding site and by BAMBI (BMP and activin Seven type I receptors, activin receptor-like kinase 1-7 receptor membrane bound inhibitor), a pseudo-receptor that (ALK1-7) and 5 type II receptors (TβRII, BMPRII, sequesters activin. Cripto blocks the recruitment of the type I AMHRII, ActRIIA and ActRIIB) have been identified in the receptor by the ligand-occupied type II receptor. Finally, TGF-β receptor superfamily [23]. Various combinations of extracellular antagonists, including follistatin and follistatin- the type I and type II receptors mediate signaling by different like 3, bind to ligands with high affinity and interfere with ligands in a cell-specific manner. These ligand/receptor receptor engagement. complexes can induce signaling through different signaling Smad pathways. Receptor heterodimers that contain ALK1, Inhibins ALK2, ALK3 and ALK6 respond to BMPs and activate Smad1, Smad5 and Smad8 whereas ALK4-, ALK5- and Inhibins are heterodimeric members of TGF-β ALK7-containing heterodimers respond to TGF-βs, activins, superfamily in which an 18 kDa α-subunit is disulfide-linked nodal and lefty and activate Smad2 and Smad3 [24-25]. to one of the two 14 kDa activin β -subunits, resulting in ALK4, known as activin type IB receptor (ActRIB or inhibin A (α-βA) and inhibin B (α-βB) [35]. The β subunits ACVR1B), is the main type I receptor for activin A, and of inhibins are capable of binding to type II activin and BMP ALK7, referred to as activin type IC receptor (ACVR1C) is receptors with lower affinity than the primary agonist [36, the type I receptor for activins B and AB. The type I receptor 37]. The lower affinity of inhibin is increased by binding to a contains an extracellular domain with a number of co-receptor, betaglycan (BG), also known as TGF-β type III conformationally important cysteine residues, a receptor, allowing inhibin to compete with activin for transmembrane domain, a juxtamembrane glycine- and binding to activin type II receptors. This competitive binding serine-rich segment called the GS domain, and an of the inhibin-BG complex to the type II receptor prevents intracellular serine-threonine kinase domain [26]. The unique recruitment or phosphorylation of the type I receptor thus and highly conserved GS domain regulates kinase activity in inhibiting activin receptor signaling. Inhibin’s blocking of a phosphorylation-dependent manner. The activin type II the type I receptor activation antagonizes activin in multiple receptors [ActRIIA (or ACVR2)] and type IIB (ActRIIB or tissues. ACVR2B) [26] have an extracellular ligand binding domain, Inhibin A and inhibin B exert both redundant and distinct a single transmembrane domain, an intracellular serine- effects against different TGF-β superfamily members threonine kinase domain, and a PDZ protein-binding (activin A or BMP). For example, both inhibin A and inhibin consensus sequence at the COOH-terminus. The activin type B antagonize activin A- and B-mediated suppression of II receptors are the primary ligand binding proteins, as they Cyp17 expression and 17α-hydroxylase activity in a mouse Activin Signaling in Skeleton Current Molecular Pharmacology, 2012, Vol. 5, No. 2 197

Follistatin Betaglycan FSTL-3 Activin BAMBI Inhibin Cripto

II IIII IIII I I IIII I I IIII II PP PP PP PP PP PP Non-canonical pathway Canonical pathway

ERK, JNK, p38 MAPK, Smad2/3 P IB-, PKC, Akt and

Wnt/β-catenin Smad4

Smad4 Smad2/3 P

Fig. (1). Activin signaling pathway. Activins transduce their signals via binding to transmembrane type II receptors which recruit and activate type I receptor, leading to phosphorylation of receptor-regulated Smad proteins in the canonical pathway. Ligand-induced heterodimerization of the receptors can regulate cellular processes through non-canonical pathways by the activation of other signaling molecules. The activin signals are blocked by activin antagonists, including follistatin, FSTL-3, cripto, BAMBI and inhibin. adrenocortical cell line whereas only inhibin A antagonizes obligatory co-receptor for Nodal, it antagonizes activin A the action of BMP2, BMP6 and BMP7 on 17α-hydroxylase and activin B signaling by preventing activins from activity at low concentration [38]. Inhibin B is less effective assembling their receptors. Therefore, cells can modulate at antagonizing both activins than inhibin A and activin B is their responsiveness to these ligands and vary the level of more susceptible to antagonism by either inhibin than activin downstream Smad2/3 signaling by altering the expression of A. Cripto. However, it remains unclear how Cripto inhibits activin-dependent receptor assembly while forming BAMBI functional signaling complexes with Nodal.

BAMBI was initially characterized as a pseudo (decoy) Follistatin type I receptor antagonizing TGF-β receptor activation. BAMBI is a 260 amino acid transmembrane protein highly Follistatin is a glycosylated single-chain protein that related to the TGF-β type I receptor, but lacks an binds multiple TGF-β family ligands, including activins. intracellular kinase domain. It exhibits 53% structural Follistatin acts as an extracellular antagonist that binds to homology to Xenopus BMP receptor type I (BMPRI) [39]. activins and, with lower affinity, to BMP2, BMP4, BMP7 BAMBI interacts with the intracellular domain of type I and BMP11 [42, 43]. The binding of activin to follistatin is receptors for TGF-β, BMPs and activin and prevents the irreversible, thus the rate of synthesis of follistatin is a major formation of active receptor complexes. BAMBI contributor to the regulation of activin bioavailability. The overexpression antagonizes BMP and activin signaling in the follistatin/activin complex is composed of two follistatin Xenopus embryo [39]. molecules for each activin β/β homodimer. Once bound to activin, follistatin abolishes the binding of activins to their Cripto receptors, thus inhibiting their biological effects. Follistatin binds to cell-surface proteoglycan heparan sulfate side chains Cripto is an extracellular, glycosyl-phosphatidylinositol [44] on the surface of most tissues that also express activin. (GPI)-anchored protein belonging to EGF-CFC family [40, It acts in an autocrine/paracrine rather than endocrine 41]. It contains two modular cysteine-rich domains, an manner to regulate activin activity. Substantial levels of epidermal growth factor-like domain that binds nodal and a circulating activin-bound follistatin are detectable and CFC domain that binds ALK4. Although Cripto is an 198 Current Molecular Pharmacology, 2012, Vol. 5, No. 2 Lotinun et al. follistatin may therefore act to facilitate clearance of activin However, other reports demonstrated that activin A acts as a or prevent activin diffusion from its local site of action. negative regulator of osteoblast differentiation and matrix mineralization in vitro [57-59]. The discrepancy of the Alternative splicing of mRNA generates two isoforms of effects of activin A on osteogenesis may be due, in part, to follistatin, follistatin-315 and its carboxyl-terminal truncated homologue follistatin-288 [45]. The longer form is the more different species. Activin A inhibits bone matrix formation and mineralization in human osteoblast culture [59]. In abundant variant whereas the shorter form has higher addition, it alters expression of numerous extracellular biological activity due to its greater affinity for heparan matrix genes in such a way that mineralization is strongly sulfate [46]. Follistatin-303, an intermediate form generated inhibited. by protein processing, has been identified in follicular fluid [47]. Follistatin-288 is more potent and longer acting than Activin A stimulates osteoclastogenesis in vitro whereas inhibin A in suppressing FSH secretion from the pituitary in inhibin has the opposite effect [60]. Activin A increases ovariectomized rats [45]. receptor activator of NF-κB ligand (RANKL)-mediated osteoclast differentiation but not osteoclast survival [33]. In Follistatin-Like 3 (FSTL-3) addition, activin A stimulates IκB-α, increases RANK expression and promotes M-CSF and RANKL-induced FSTL-3 is a follistatin-related protein encoded by nuclear translocation of NFκB in osteoclast precursors. follistatin-related gene (FLRG) that is highly expressed in placenta, testis, pancreas and heart [48]. FSTL-3 shares The role of follistatin in bone metabolism has been substantial structural and functional homology with investigated in vivo and in vitro. Follistatin is expressed in follistatin. Both FSTL-3 and follistatin bind and antagonize osteoblasts, osteocytes and MC3T3-E1 cells [61, 62]. The activin and myostatin. However, FSTL-3 is not as effective interaction of follistatin and activin takes place at several steps during cartilage maturation, endochondral bone as follistatin in antagonizing activin action and exhibits limited ligand specificity, lacking interaction with BMPs 2 formation, and osteoblast formation and bone remodeling and 4 [49]. Unlike follistatin, FSTL-3 does not possess a cell [33, 62, 63]. Follistatin-deficient mice have growth surface heparin-binding domain. Mice deficient in FSTL-3 retardation, abnormal tooth and craniofacial development, are viable but develop metabolic phenotypes, including cutaneous abnormalities and die within hours of birth, changes in glucose and fat metabolism [48]. reflecting the abnormal signaling of several TGF-β superfamily members [64]. In contrast, mice overexpressing a follistatin transgene are viable with no reported skeletal ACTIVINS AND BONE HOMEOSTASIS phenotype [65]. The extent of mineralization appears to be regulated by the ratio of activin A-to-follistatin in mature Activin-induced signaling is implicated in multiple osteoblasts [59]. Neutralization of endogenous activins by diseases, including some that involve the musculoskeletal follistatin increases matrix mineralization through system. Activins regulate bone cell proliferation and mechanisms that include altered extracellular matrix differentiation, skeletal development and bone turnover. composition and maturation. Activin βA and activin βB appear to play nonidentical roles in bone homeostasis, although some redundancy has been Inhibin and activin exert opposing effects in many cells, demonstrated. Mice with a homozygous mutation in the gene including erythroid lineage [66], megakaryocytes [67], encoding the activin βA subunit have severe defects in granulocyte-macrophages [68], osteoblasts and osteoclasts craniofacial development and die within 24 h of birth while [60]. Unlike activin, inhibin is a gonadally-derived hormone mice deficient in activin βB are viable, indicating a crucial and not locally produced within bone microenvironment. role of activin βA in craniofacial development [50]. When Inhibin is critically important for the maintenance of bone the peptide coding region of the activin βA gene was quality. The changes in the ratio of inhibin to activin regulate replaced with the corresponding region of the activin βB bone cell differentiation and appear to be associated with gene, the defects of activin βA-null mice were partially increased bone resorption during the menopausal transition, rescued, further indicating that the two subunits are not since decreased inhibin levels correlate with increased bone functionally equivalent [51]. Since large quantities of activin resorption during perimenopause before a decrease in serum A are present in extracellular bone matrix, activin A has been estradiol level is detected [9]. Inhibin suppresses the the most extensively investigated activin in the skeletal transition of mesenchymal progenitors to ALP-positive context. Activin A enhances the induction of ectopic bone preosteoblastic cells and inhibits osteoblast maturation in formation when implanted in combination with BMPs [52], vitro [60]. The inhibitory effect of inhibin on increases osteoblast proliferation [53], enhances mechanical osteoblastogenesis is mediated by downregulation of strength [54] and stimulates fracture healing [55]. It has been osteogenic activin synthesis. Inhibin A and B completely shown that the effects of exogenously administered activin blocks RANKL-induced osteoclast differentiation [69]. on the skeleton are dose-dependent. Low doses of activin A However, mice with inducibly-expressed inhibin A have (1 and 5 µg/kg) markedly increased vertebral bone mass and high bone mass and increased mechanical properties [70]. mechanical strength in aged OVX rats [54]. However, in the The positive bone balance in these mice is due primarily to same study, a higher dosage of activin A (25 µg/kg) increased osteoblast activity without affecting osteoclast diminished the magnitude of these effects. Local injection of number and activity. Resolving the discrepancy of in vivo activin A onto the periosteum of parietal bone in newborn and in vitro effects of inhibin on bone turnover requires rats demonstrated a dose- and time-dependent increase in the further investigation. thickness of periosteum and bone matrix layers [56]. Activin Signaling in Skeleton Current Molecular Pharmacology, 2012, Vol. 5, No. 2 199

CANONICAL AND NON-CANONICAL ACTIVIN those effectors modulate the responses to M-CSF and SIGNALING IN PATHOLOGICAL BONE DISEASES RANKL remain unclear. Activin A also inhibits osteoblast differentiation and Bone remodeling, a tightly regulated process by osteoclasts and osteoblasts, is required to maintain bone function. This has been clearly demonstrated in the context of multiple myeloma-induced osteolytic disease [72]. The homeostasis. Coupling or maintaining the balance between adhesion of multiple myeloma cells to bone marrow stromal bone formation and bone resorption depends on the cells in the niche induces the secretion of activin A from exchange of signals between osteoclasts and osteoblasts. The bone marrow stromal cells, in part via JNK-dependent disruption of these signals referred to as uncoupling leads to pathway. The activin A in turn inhibits osteoblast pathological situations, including osteoporosis. The progressive loss of bone mass, leading to osteoporosis is the differentiation through Smad2-mediated down-regulation of DLX5 expression [72]. Activin A also acts via Smad3 to result of a deficit in bone formation relative to bone induce androgen receptor gene transcription and nuclear resorption. translocation and thereby to promote the migration of Activin A plays an essential role in the modulation of prostate cancer cells to osteoblasts and bone matrix [73]. osteoblast and osteoclast differentiation and activity through However, there is as yet no evidence of activin-mediated both canonical and non-canonical activin signaling Smad-independent cascades in bone cancer metastasis. pathways. Activin A synergizes with RANKL to enhance osteoclast differentiation from bone marrow macrophages, in SOLUBLE ACTIVIN RECEPTOR FUSION PROTEINS part by inducing the phosphorylation of Smad2 in the canonical pathway [71]. On the other hand, activin A rapidly Soluble activin receptors serve as dominant negative stimulates IκB-α activation within 5 min and induces nuclear forms of the receptor. Soluble forms of ActRIIA and translocation of phosphorylated-NFκB, a key downstream ActRIIB receptors contain the extracellular domain but lack target of RANKL-induced osteoclast differentiation [33], as transmembrane and cytoplasmic kinase domains. Two well as enhancing the phosphorylation of p38 MAPK and activin type II receptor decoys have different clinical ERK1/2 in the presence of M-CSF and RANKL, all non- application. A soluble ActRIIA-Fc fusion protein neutralizes canonical responses to activin A. However, the molecular activin and increases bone mass and strength with no mechanisms by which activin A couples to Smad- changes in muscle mass, whereas an extracellular ActRIIB independent downstream effectors and the manner in which fused with IgG-Fc dramatically increases muscle mass

Extracellular Activin Domain Receptor of ActRIIA Type IIA (ActRIIA)

ACE-011

IgG1 Immunoglobulin Fc Domain

of IgG1 Fig. (2). A schematic diagram of the ActRIIA extracellular domain fused to a human IgG1 Fc region (ACE-011). Reprinted from Bone, 46, Lotinun S., et al, A Soluble Activin Receptor Type IIA Fusion Protein (ACE-011) Increases Bone Mass via a Dual Anabolic-Antiresorptive Effect in Cynomolgus Monkeys, 1082-1088, 2010, with permission from Elsevier. 200 Current Molecular Pharmacology, 2012, Vol. 5, No. 2 Lotinun et al.

VEHICLE RAP-011

Fig. (3). Cancellous bone architecture in the distal femur of RAP-011 treated ovariectomized mice. predominantly by antagonism of myostatin activity [74], but growth and tumor-induced impairment of osteoblast also increases bone mass, likely through binding a ligand differentiation [72]. RAP-011 treatment improved osteolytic that also binds to ActRIIA. disease and prolonged survival in mouse models of multiple myeloma [75]. In addition, RAP-011 prevents bone SOLUBLE ActRIIA FUSION PROTEIN AND metastases in mice bearing MDA-MB-231 breast cancer CLINICAL APPLICATION cells [75]. ACE-011 is in clinical development and has completed a Two forms of the soluble ActRIIA-Fc fusion proteins Phase I, randomized, double-blind, placebo-controlled, have been developed. ACE-011 is the ActRIIA extracellular single dose, dose-escalation study in postmenopausal women domain (ECD) fused to a human IgG1 Fc region (Fig. 2) and [14]. While the primary endpoint was to assess the safety RAP-011 is a murine surrogate containing the human ECD and tolerability of ACE-011 administered by subcutaneous fused to a mouse IgG2 Fc region. Surface plasmon resonance or intravenous injection, the secondary objectives included analysis using a Biacore 3000 biosensor determined that investigation of ACE-011 effects on serum biomarkers of ACE-011 avidly binds to activin A with a binding coefficient bone formation and resorption. A total of 48 volunteers were (Kd) of 3.6-7.8 pM. ACE-011 demonstrated in vitro binding tested in 8 cohorts of 5 dose levels from 0.01 to 3.0 mg/kg. A to other ligands such as myostatin and GDF-11, but showed pharmacokinetic profile determined the serum half life of no physiologic changes related to binding of these ligands in ACE-011 was approximately 30 days. A single dose of animal models, i.e. no increase in muscle mass. Importantly, ACE-011 led to a dose-dependent increase in bone-specific ACE-011 did not show any binding affinity for TGFβ-1, -2 alkaline phosphatase (BSALP), a marker of bone formation or -3. Thus, ACE-011 acts as a ligand trap to prevent activin (Table 1). BSALP showed significant increases to a A from binding to the endogenous ActRIIA receptor, thereby maximum of 35.9% above baseline by day 15 of the study in inhibiting the biological effects of activin. the highest dose group and continued to have elevated levels In animal studies, the use of a soluble ActRIIA-Fc fusion up to 120 days of the study. Levels of procollagen type I N- protein (either ACE-011 or RAP-011) demonstrated terminal propeptide (PINP), another marker of bone increased bone mass and strength due to a dual anabolic and formation, were also significantly increased (maximum of antiresorptive effect in both normal mice and non-human 15%) from baseline levels in the highest dose groups by day primates [11-13]. As shown in Fig. (3), RAP-011 treatment 15, but did not reach significance in other groups or at other of ovariectomized mice with established bone loss induced time points. Reduction in the bone resorption markers type I increases in bone mass and strength at multiple skeletal sites, collagen C-telopeptides (CTX) and tartrate-resistant acid mainly due to increased bone anabolic activity, but with phosphatase 5b (TRACP-5b) were also noted to varying concurrent, perhaps transient, suppression of resorption [11]. degrees, but had high variability within the groups. In the In an animal model of multiple myeloma-induced osteolysis, highest dose group CTX levels showed a maximum decrease the use of RAP-011 has shown efficacy in inhibiting tumor of 30.6% by day 8. TRACP-5b levels showed a 6.2% Activin Signaling in Skeleton Current Molecular Pharmacology, 2012, Vol. 5, No. 2 201

Table 1. Pharmacodynamic biomarker changes in the ACE-011 phase I clinical trial.

Biomarker Direction of Change ACE-011 Biologic Effect

Significant change from baseline after single administration of 0.1, 0.3, 1.0 and 3.0 BSALP ↑ mg/kg ACE-011. Maximum increase of 10.2%, 14.6%, 21.1% and 35.9% for each group respectively during the study period PINP ↑ Increase of 10% from baseline in the 3.0 mg/kg ACE-011 treatment group Maximum decrease of 12.4%, 28.7% and 30.6% in the 0.3, 1.0 and 3.0 mg/kg group CTX ↓ respectively within the first 29 days. Decrease of 0.9%, 3.5% and 6.2% in the 0.3, 1.0 and 3.0 mg/kg group respectively on TRACP5b ↓ day 29

RBC ACE-011 caused dose dependent increases in hematologic parameters at the 0.3, 1.0 and 3.0 mg/kg dose levels. RBC number increased a maximum of 13.4%, 18.9% and 18%, Hb ↑ hemoglobin increased 7.4%, 12.6% and 12.5% and Hct increased 8.2%, 12.8% and Hct 17.1% in the 0.3, 1.0 and 3.0 mg/kg group respectively. decrease in the highest dose group by day 29 of the study. bone metastases. Targeting of activin might, therefore, The bone effects are considered reversible since the bone represent a novel therapeutic agent for osteoporosis and biomarker levels returned to baseline levels by 4 months cancer-related osteolytic diseases. However, long-term safety after completion of dosing. Of note, ACE-011 administration and effectiveness of a soluble ActRIIA-Fc fusion protein caused a transient, but significant increase in the red blood must be evaluated in the near future. cell (RBC) mass, hemoglobin (Hb) and hematocrit (Hct) of treated individuals during this study (Table 1), indicating CONFLICT OF INTEREST that activin A inhibition may also effect erythroid differentiation in addition to the effects on bone formation. R. Baron and S. Lotinun received financial support from The data from these studies demonstrate that the use of a Acceleron Pharma, Inc. R. Scott Pearsall is an employee of soluble ActRIIA-Fc fusion protein as a ligand trap has a and has stock options in Acceleron Pharma, Inc. distinct profile from existing therapies used to treat diseases with associated bone loss. The dual effects of ACE-011 on ACKNOWLEDGEMENT bone formation and resorption provide the opportunity to increase bone mass and strength in patients suffering from We thank M. Bouxsein for assistance with the µCT diseases with loss of bone mass and strength. imaging.

CONCLUSION ABBREVIATIONS

Epidemiology studies have indicated the prevalence and ERK1/2 = Extracellular signal-regulated kinases 1/2 incidence of fractures in osteoporosis, and insight into the JNK = c-Jun N-terminal kinase pathogenesis of the disease in postmenopausal women has increased dramatically during the past decade. Consequently, MAPK = Mitogen-activated protein kinase therapeutic options for the treatment of osteoporosis have PKC = Protein kinase C advanced with the development of potent antiresorptive and anabolic agents. The process of coupled bone formation and IκB-α = Inhibitor of kappa B-alpha bone resorption requires the activation of signaling cascades Cyp17 = Cytochrome P450 17α-hydroxylase/17, 20- that are crucial for bone remodeling. The literature supports lyase a role of activin signaling in the regulation of bone mass and fracture risk. Better understanding of the complexity of the DLX5 = Distal-less homeobox 5 regulation of activin-induced signaling should provide knowledge that may be useful for therapeutic intervention. REFERENCES The emerging importance of a soluble ActRIIA-Fc fusion protein in promoting bone formation and inhibiting bone [1] Marx, J. Coming to grips with bone loss. Science, 2004, 305, 1420- resorption during phase I clinical trials suggests the 1422. [2] Durie, B.G.; Katz, M.; Crowley, J. Osteonecrosis of the jaws and therapeutic value of antagonizing activin signaling. A dual bisphosphonates. N. Engl. J. Med., 2005, 353, 99-102. anabolic antiresorptive role of a soluble ActRIIA-Fc fusion [3] Cartsos, V.M.; Zhu, S.; Zavras, A.I. Bisphosphonate use and the protein increases a possibility to cure severe osteoporosis. In risk of adverse jaw outcomes: a medical claims study of 714,217 addition, activin A has been identified as a critical player in people. J. Am. Dent. Assoc., 2008, 139, 23-30. cancer-related bone diseases. Understanding pathogenesis of [4] Pazianas, M.; Miller, P.; Blumentals, W.A.; Bernal, M.; Kothawala, osteolytic diseases is crucial to the discovery of anticancer P. A review of the literature on osteonecrosis of the jaw in patients with osteoporosis treated with oral bisphosphonates: prevalence, therapies. ActRIIA-Fc fusion protein increases bone risk factors, and clinical characteristics. Clin. Ther., 2007, 29, formation and prevents osteolytic lesion in animal model of 1548-1558. multiple myeloma. Blocking activin A reduces breast cancer 202 Current Molecular Pharmacology, 2012, Vol. 5, No. 2 Lotinun et al.

[5] Deal, C. Future therapeutic targets in osteoporosis. Curr. Opin. [21] Deli, A.; Kreidl, E.; Santifaller, S.; Trotter, B.; Seir, K.; Berger, Rheumatol., 2009, 21, 380-385. W.; Schulte-Hermann, R.; Rodgarkia-Dara, C.; Grusch, M. [6] Ominsky, M.S.; Vlasseros, F.; Jolette, J.; Smith, S.Y.; Stouch, B.; Activins and activin antagonists in hepatocellular carcinoma. World Doellgast, G.; Gong, J.; Gao, Y.; Cao, J.; Graham, K.; Tipton, B.; J. Gastroenterol., 2008, 14, 1699-1709. Cai, J.; Deshpande, R.; Zhou, L.; Hale, M.D.; Lightwood, D.J.; [22] Oda, S.; Nishimatsu, S.; Murakami, K.; Ueno, N. Molecular Henry, A.J.; Popplewell, A.G.; Moore, A.R.; Robinson, M.K.; cloning and functional analysis of a new activin beta subunit: a Lacey, D.L.; Simonet, W.S.; Paszty, C. Two doses of sclerostin dorsal mesoderm-inducing activity in Xenopus. Biochem. Biophys. antibody in cynomolgus monkeys increases bone formation, bone Res. Commun., 1995, 210, 581-588. mineral density, and bone strength. J. Bone Miner. Res., 2010, 25, [23] Wang, H.; Tsang, B.K. Nodal signalling and apoptosis. 948-958. Reproduction, 2007, 133, 847-853. [7] Li, X.; Ominsky, M.S.; Warmingtom, K.S.; Morony, S.; Gong, J.; [24] Peng, C. The TGF-beta superfamily and its roles in the human Cao, J.; Gao, Y.; Shalhoub, V.; Tipton, B.; Haldankar, R.; Chen, ovary and placenta. J. Obstet. Gynaecol. Can., 2003, 25, 834-844. Q.; Winters, A.; Boone, T.; Geng, Z.; Niu, Q.T.; Ke, H.Z.; [25] Miyazawa, K.; Shinozaki, M.; Hara, T.; Furuya, T.; Miyazon, K. Kostenuik, P.J.; Simonet, W.S.; Lacey, D.L.; Paszty, C. Sclerostin Two major Smad pathways in TGF-beta superfamily signalling. antibody treatment increases bone formation, bone mass and bone Genes Cells, 2002, 7, 1191-1204. strength in a rat model of postmenopausal osteoporosis. J. Bone [26] Tsuchida, K.; Nakatani, M.; Hitachi, K.; Uezumi, A.; Sunada, Y.; Miner. Res., 2009, 24, 574-588. Ageta, H.; Inokuchi, K. Activin signaling as an emerging target for [8] Hoeppner, L.H.; Secreto, F.J.; Westendorf, J.J. Wnt signaling as a therapeutic interventions. Cell Commun. Signal., 2009, 18, 7-15. therapeutic target for bone diseases. Expert. Opin. Ther. Targets, [27] Zhang, L.; Wang, W.; Hayashi, Y.; Jester, J.V.; Birk, D.E.; Gao, 2009, 13, 455-496. M.; Liu, C.Y.; Kao, W.W.; Karin, M.; Xia, Y. A role for MEK [9] Nicks, K.M.; Perrien, D.S.; Akel, N.S.; Suva, L.J.; Gaddy, D. kinase 1 in TGF-beta/activin-induced epithelium movement and Regulation of osteoblastogenesis and osteoclastogenesis by the embryonic eyelid closure. EMBO J., 2003, 22, 4443-4454. other reproductive hormones, Activin and Inhibin. Mol. Cell [28] Zhu, W.; Xu, P.; Cuascut, F.X.; Hall, A.K.; Oxford, G.S. Activin Endocrinol., 2009, 310, 11-20. acutely sensitizes dorsal root ganglion neurons and induces [10] Perrien, D.S.; Achenbach, S.J.; Bledsoe, S.E.; Walser, B.; Suva, hyperalgesia via PKC-mediated potentiation of transient receptor L.J.; Khosla, S.; Gaddy, D. Bone turnover across the menopause potential vanilloid I. J. Neurosci., 2007, 27, 13770-13780. transition: correlations with inhibins and follicle-stimulating [29] Bao, Y.L.; Tsuchida, K.; Liu, B.; Kurisaki, A.; Matsuzaki, T.; hormone. J. Clin. Endocrinol. Metab., 2006, 91, 1848-1854. Sugino, H. Synergistic activity of activin A and basic fibroblast [11] Pearsall, R.S.; Canalis, E.; Cornwall-Brady, M.; Underwood, K.W.; growth factor on tyrosine hydroxylase expression through Smad3 Haigis, B.; Ucran, J.; Kumar, R.; Pobre, E.; Grinberg, A.; Werner, and ERK1/ERK2 MAPK signaling pathways. J. Endocrinol., 2005, E.D.; Glatt, V.; Stadmeyer, L.; Smith, D.; Seehra, J.; Bouxsein, 184, 493-504. M.L. A soluble activin type IIA receptor induces bone formation [30] de Guise, C.; Lacerte, A.; Rafiei, S.; Reynaud, R.; Roy, M.; Brue, and improves skeletal integrity. Proc. Natl. Acad. Sci. USA, 2008, T.; Lebrun, J.J. Activin inhibits the human Pit-1 gene promoter 105, 7082-7087. through the p38 kinase pathway in a Smad-independent manner. [12] Fajardo, R.J.; Manoharan, R.K.; Pearsall, R.S.; Davies, M.V.; Endocrinology, 2006, 147, 4351-4362. Marvell, T.; Monnell, T.E.; Ucran, J.A.; Pearsall, A.E.; Khanzode, [31] Hirota, M.; Watanabe, K.; Hamada, S.; Sun, Y.; Strizzi, L.; D.; Kumar, R.; Underwood, K.W.; Roberts, B.; Seehra, J.; Mancino, M.; Nagaoka, T.; Gonzales, M.; Seno, M.; Bianco, C.; Bouxsein, M.L. Treatment with a soluble receptor for activin Salomon, D.S. Smad2 functions as a co-activator of canonical improves bone mass and structure in the axial and appendicular Wnt/beta-catenin signaling pathway independent of Smad4 through skeleton of female cynomolgus macaques (Macaca fascicularis). histone acetyltransferase activity of p300. Cell Signal., 2008, 20, Bone, 2010, 46, 64-71. 1632-1641. [13] Lotinun, S.; Pearsall, R.S.; Davies, M.V.; Marvell, T.H.; Monnell, [32] Do, T.V.; Kubba, L.A.; Antenos, M.; Rademaker, A.W.; Sturgis, T.E.; Ucran, J.; Fajardo, R.J.; Kumar, R.; Underwood, K.W.; C.D.; Woodruff, T.K. The role of activin A and Akt/GSK signaling Seehra, J.; Bouxsein, M.L.; Baron, R. A soluble activin receptor in ovarian tumor biology. Endocrinology, 2008, 149, 3809-3816. Type IIA fusion protein (ACE-011) increases bone mass via a dual [33] Sugatani, T.; Alvarez, U.M.; Hruska, K.A. Activin A stimulates anabolic-antiresorptive effect in Cynomolgus monkeys. Bone, IkappaB-alpha/NFkappaB and RANK expression for osteoclast 2010, 46, 1082-1088. differentiation, but not AKT survival pathway in osteoclast [14] Ruckle, J.; Jacobs, M.; Kramer, W.; Pearsall, A.E.; Kumar, R.; precursors. J. Cell Biochem., 2003, 90, 59-67. Underwood, K.W.; Seehra, J.; Yang, Y.; Condon, C.H.; Sherman, [34] Lebrun, J.J.; Takabe, K. ; Chen, Y.; Vale, W. Roles of pathway- M.L. Single-dose, randomized, double-blind, placebo-controlled specific and inhibitory Smads in activin receptor signaling. Mol. study of ACE-011 (ActRIIA-IgG1) in postmenopausal women. J. Endocrinol., 1999, 13, 15-23. Bone Miner. Res., 2009, 24, 744-752. [35] Ling, N.; Ying, S.Y.; Ueno, N.; Esch, F.; Denoroy, L.; Guillemin, [15] Tsuchida, K. Activins, myostatin and related TGF-beta family R. Isolation and partial characterization of a Mr 32,000 protein with members as novel therapeutic targets for endocrine, metabolic and inhibin activity from porcine follicular fluid. Proc. Natl. Acad. Sci. immune disorders. Curr. Drug Targets Immune. Endocr. Metabol. USA, 1985, 82, 7217-7221. Disord., 2004, 4, 157-66. [36] Mathews, L.S.; Vale, W.W. Expression cloning of an activin [16] Shav-Tal, Y.; Zipori, D. The role of activin a in regulation of receptor, a predicted transmembrane serine kinase. Cell, 1991, 65, hemopoiesis. Stem Cells, 2002, 20, 493-500. 973-982. [17] Rodgarkia-Dara, C.; Vejda, S.; Erlach, N.; Losert, A.; Bursch, W.; [37] Stenvers, K.L.; Findlay, J.K. Inhibins: from reproductive hormones Berger, W.; Schulte-Hermann, R.; Grusch, M. The activin axis in to tumor suppressors. Trends Endocrinol. Metab., 2010, 21, 174- liver biology and disease. Mutat. Res., 2006, 613, 123-137. 180. [18] Chang, H.; Brown, C.W.; Matzuk, M.M. Genetic analysis of the [38] Farnworth, P.G.; Stanton, P.G.; Wang, Y.; Escalona, R.; Findlay, mammalian transforming growth factor-beta superfamily. Endocr. J.K.; Ooi, G.T. Inhibins differentially antagonize activin and bone Rev., 2002, 23, 787-823. morphogenetic protein action in a mouse adrenocortical cell line. [19] Vale, W., Rivier, C., Hsueh, A., Campen, C., Meunier, H., Bicsak, Endocrinology, 2006, 147, 3462-3471. T., Vaughan, J., Corrigan, A., Bardin, W., Sawchenko, P., [39] Onichtchouk, D.; Chen, Y.G.; Dosch, R.; Gawantka, V.; Delius, H.; Petraglia, F., Yu, J., Plotsky, P., Spiess, J., Rivier, J. Chemical and Massagué, J.; Niehrs, C. Silencing of TGF-beta signalling by the biological characterization of the inhibin family of protein pseudoreceptor BAMBI. Nature, 1999, 401, 480-485. hormones. Recent Prog. Horm. Res., 1988, 44, 1-34. [40] Kelber, J.A.; Shani, G.; Booker, E.C.; Vale, W.W.; Gray, P.C. [20] Massague, J. The transforming growth factor-β family. Annu. Rev. Cripto is a noncompetitive activin antagonist that forms analogous Cell Biol , 1990, 6, 597-641. signaling complexes with activin and nodal. J. Biol. Chem., 2008, 283, 4490-4500. Activin Signaling in Skeleton Current Molecular Pharmacology, 2012, Vol. 5, No. 2 203

[41] Shen, M.M.; Schier, A.F. The EGF-CFC gene family in vertebrate cultures of fetal rat calvarial cells. J. Cell. Biochem., 1999, 75, 206- development. Trends Genet., 2000, 16, 303-309. 214. [42] Iemura, S.; Yamamoto, T.S.; Takagi, C.; Uchiyama, H.; Natsume, [59] Eijken, M.; Swagemakers, S.; Koedam, M.; Steenbergen, C.; T.; Shimasaki, S.; Sugino, H.; Ueno, N. Direct binding of follistatin Derkx, P.; Uitterlinden, A.G.; van der Spek, P.J.; Visser, J.A.; de to a complex of bone-morphogenetic protein and its receptor Jong, F.H.; Pols, H.A.; van Leeuwen, J.P. The activin A-follistatin inhibits ventral and epidermal cell fates in early Xenopus embryo. system: potent regulator of human extracellular matrix Proc. Natl. Acad. Sci. USA, 1998, 95, 9337-9342. mineralization. FASEB J., 2007, 21, 2949-2960. [43] Gamer, L.W.; Wolfman, N.M.; Celeste, A.J.; Hattersley, G.; [60] Gaddy-Kurten, D.; Coker, J. K.; Abe, E.; Jilka, R. L.; Manolagas, Hewick, R.; Rosen, V. A novel BMP expressed in developing S. C. Inhibin suppresses and activin stimulates osteoblastogenesis mouse limb, spinal cord, and tail bud is a potent mesoderm inducer and osteoclastogenesis in murine bone marrow cultures. in Xenopus embryos. Dev. Biol., 1999, 208, 222-232. Endocrinology, 2002, 143, 74-83. [44] Nakamura, T.; Sugino, K.; Titani, K.; Sugino, H. Follistatin, an [61] Inoue, S.; Nomura, S.; Hosoi, T.; Ouchi, Y.; Orimo, H.; activin-binding protein, associates with heparan sulfate chains of Muramatsu, M. Localization of follistatin, an activin-binding proteoglycans on follicular granulosa cells. J. Biol. Chem., 1991, protein, in bone tissues. Calcif. Tissue Int., 1994, 55, 395-397. 266, 19432-19437. [62] Hashimoto, M.M.; Shoda, A.; Inoue, S.; Yamada, R.; Kondo, T.; [45] Inouye, S.; Guo, Y.; DePaolo, L.; Shimonaka, M.; Ling, N.; Sakurai, T.; Ueno, N.; Muramatsu, M. Functional regulation of Shimasaki, S. Recombinant expression of human follistatin with osteoblastic cells by the interaction of activin-A with follistatin. J. 315 and 288 amino acids: chemical and biological comparison with Biol. Chem., 1992, 267, 4999-5004. native porcine follistatin. Endocrinology, 1991, 129, 815-822. [63] Funaba, M.; Ogawa, K.; Murata, T.; Fujimura, H.; Murata, E.; Abe, [46] Sugino, K.; Kurosawa, N.; Nakamura, T.; Takio, K.; Shimasaki, S.; M.; Takahashi, M.; Torii, K. Follistatin and activin in bone: Ling, N.; Titani, K.; Sugino, H. Molecular heterogeneity of Expression and localization during endochondral bone follistatin, an activin-binding protein. Higher affinity of the development. Endocrinology, 1996, 137, 4250-4259. carboxyl-terminal truncated forms for heparan sulfate [64] Matzuk, M.M.; Lu, N.; Vogel, H.; Sellheyer, K.; Roop, D.R.; proteoglycans on the ovarian granulosa cell. J. Biol. Chem., 1993, Bradley, A. Multiple defects and perinatal death in mice deficient 268, 15579-15587. in follistatin. Nature, 1995, 374, 360-363. [47] Schneyer, A.L.; Wang, Q.; Sidis, Y.; Sluss, P.M. Differential [65] Guo, Q., Kumar, T. R., Woodruff, T., Hadsell, L. A., DeMayo, F. distribution of follistatin isoforms: application of a new FS315- J., Matzuk, M. M. Overexpression of mouse follistatin causes specific immunoassay. J. Clin. Endocrinol. Metab., 2004, 89, 5067- reproductive defects in transgenic mice. Mol. Endocrinol., 1998, 5075. 12, 96-106. [48] Mukherjee, A.; Sidis, Y.; Mahan, A.; Raher, M.J.; Xia, Y.; Rosen, [66] Yu, J.; Shao, L.E.; Lemas, V.; Yu, A.L.; Vaughan, J.; Rivier, J.; E.D.; Bloch, K.D.; Thomas, M.K.; Schneyer, A.L. FSTL3 deletion Vale, W. Importance of FSH-releasing protein and inhibin in reveals roles for TGF-beta family ligands in glucose and fat erythrodifferentiation. Nature, 1987, 330, 765-767. homeostasis in adults. Proc. Natl. Acad. Sci. USA, 2007, 104, 1348- [67] Fujimoto, K.; Kawakita, M.; Kato, K.; Yonemura, Y.; Masuda, T.; 1353. Matsuzaki, H.; Hirose, J.; Isaji, M.; Sasaki, H.; Inoue, T. [49] Sidis, Y.; Mukherjee, A.; Keutmann, H.; Delbaere, A.; Sadatsuki, Purification of megakaryocyte differentiation activity from a M.; Schneyer, A. Biological activity of follistatin isoforms and human fibrous histiocytoma cell line: N-terminal sequence follistatin-like-3 is dependent on differential cell surface binding homology with activin A. Biochem. Biophys. Res. Commun., 1991, and specificity for activin, myostatin, and bone morphogenetic 174, 1163-1168. proteins. Endocrinology, 2006, 147, 3586-3597. [68] Broxmeyer, H.E.; Lu, L.; Cooper, S.; Schwall, R.H.; Mason, A.J.; [50] Brown, C.W.; Houston-Hawkins, D.E.; Woodruff, T.K.; Matzuk, Nikolics, K. 1988 Selective and indirect modulation of human M.M. Insertion of Inhbb into the Inhba locus rescues the Inhba-null multipotential and erythroid hematopoietic progenitor cell phenotype and reveals new activin functions. Nat. Genet., 2000, 25, proliferation by recombinant human activin and inhibin. Proc. Natl. 453-457. Acad. Sci. USA, 1988, 85, 9052-9056. [51] Matzuk, M.M.; Kumar, T.R.; Vassalli, A.; Bickenbach, J.R.; Roop, [69] Perrien, D.S.; Achenbach, S.J.; Bledsoe, S.E.; Walser, B.; Suva, D.R.; Jaenisch, R.; Bradley, A. Functional analysis of activins L.J.; Khosla, S.; Gaddy, D. Bone turnover across the menopause during mammalian development. Nature, 1995, 374, 354-356. transition: correlations with inhibins and follicle-stimulating [52] Ogawa, Y.; Schmidt, D.K.; Nathan, R.M.; Armstrong, R.M.; hormone. J. Clin. Endocrinol. Metab., 2006, 91, 1848-1854. Miller, K.L.; Sawamura, S.J.; Ziman, J.M.; Erickson, K.L.; de [70] Perrien, D.S.; Akel, N.S.; Edwards, P.K.; Carver, A.A.; Bendre, Leon, E.R.; Rosen, D.M.; Seyedin, S.M.; Glaser, C.B.; Chang, R.; M.S.; Swain, F.L.; Skinner, R.A.; Hogue, W.R.; Nicks, K.M.; Corrigan, A.Z.; Vale, W. Bovine bone activin enhances bone Pierson, T.M.; Suva, L.J.; Gaddy, D. Inhibin A is an endocrine morphogenetic protein-induced ectopic bone formation. J. Biol. stimulator of bone mass and strength. Endocrinology, 2007, 148, Chem., 1992, 267, 14233-14237. 1654-1665. [53] Centrella, M.; McCarthy, T.L.; Canalis, E. Activin-A binding and [71] Murase, Y.; Okahashi, N.; Koseki, T.; Itoh, K.; Udagawa, N.; biochemical effects in osteoblast-enriched cultures from fetal-rat Hashimoto, O.; Sugino, H.; Noguchi, T.; Nishihara T. Possible parietal bone. Mol. Cell. Biol., 1991, 11, 250-258. involvement of protein kinases and Smad2 signaling pathways on [54] Sakai, R.; Fujita, S.; Horie, T.; Ohyama, T.; Miwa, K.; Maki, T.; osteoclast differentiation enhanced by activin A. J. Cell. Physiol., Okimoto, N.; Nakamura, T.; Eto, Y. Activin increases bone mass 2001, 188, 236-242. and mechanical strength of lumbar vertebrae in aged [72] Vallet, S.; Mukherjee, S.; Vaghela, N.; Hideshima, T.; Fulciniti, ovariectomized rats. Bone, 2000, 27, 91-96. M.; Pozzi, S.; Santo, L.; Cirstea, D.; Patel, K.; Sohani, A.R.; [55] Sakai, R.; Miwa, K.; Eto, Y. Local administration of activin Guimaraes, A.; Xie, W.; Chauhan, D.; Schoonmaker, J.A.; Attar, promotes fracture healing in the rat fibula fracture model. Bone, E.; Churchill, M.; Weller, E.; Munshi, N.; Seehra, J.S.; Weissleder. 1999, 25, 191-196. R.; Anderson, K.C.; Scadden, D.T.; Raje, N. Activin A promotes [56] Oue, Y.; Kanatani, H.; Kiyoki, M.; Eto, Y.; Ogata, E.; Matsumoto, multiple myeloma-induced osteolysis and is a promising target for T. Effect of local injection of activin A on bone formation in myeloma bone disease. Proc. Natl. Acad. Sci. USA. 2010, 107, newborn rats. Bone, 1994, 15, 361-366. 5124-5129. [57] Hashimoto, M.; Shoda, A.; Inoue, S.; Yamada, R.; Kondo, T.; [73] Kang, H.Y.; Huang, H.Y.; Hsieh, C.Y.; Li, C.F.; Shyr, C.R.; Tsai, Sakurai, T.; Ueno, N.; Muramatsu, M. Functional regulation of M.Y.; Chang, C.; Chuang, Y.C.; Huang, K.E. Activin A enhances osteoblastic cells by the interaction of activin-A with follistatin. J. prostate cancer cell migration through activation of androgen Biol. Chem., 1992, 267, 4999-5004. receptor and is overexpressed in metastatic prostate cancer. J. Bone [58] Ikenoue, T.; Jingushi, S.; Urabe, K.; Okazaki, K.; Iwamoto, Y. Miner. Res., 2009, 24, 1180-1193. Inhibitory effects of activin-A on osteoblast differentiation during [74] Lee, S.J.; Reed, L.A.; Davies, M.V.; Girgenrath, S.; Goad, M.E.; Tomkinson, K.N.; Wright, J.F.; Barker, C.; Ehrmantraut, G.; 204 Current Molecular Pharmacology, 2012, Vol. 5, No. 2 Lotinun et al.

Holmstrom, J.; Trowell, B.; Gertz, B.; Jiang, M.S.; Sebald, S.M.; [75] Chantry, A.D.; Heath, D.; Mulivor, A.W.; Pearsall, R.S.; Matzuk, M.; Li, E.; Liang, L.F.; Quattlebaum, E.; Stotish, R.L.; Baud’huin, M; Coulton, L.; Evans, H.; Abdul, N.; Werner, E.D.; Wolfman, N.M. Regulation of muscle growth by multiple ligands Bouxsein, M.L.; Key, M.L.; Seehra, J.; Arnett, T.R.; Vanderkerken, signaling through activin type II receptors. Proc. Natl. Acad. Sci. K.; Croucher, P. Inhibiting activin-A signaling stimulates bone USA, 2005, 102, 18117-18122. formation and prevents cancer induced bone destruction in vivo. J. Bone Miner. Res., 2010, 25, 2633-2646.

Received: June 21, 2010 Revised: October 07, 2010 Accepted: March 03, 2011

PMID: 21787285