Printer Friendly

Two activin type 2B receptors from sea bream function similarly in vitro.

Abstract. Myostatin (MSTN) and activin are members of the transforming growth factor-[beta] superfamily. Both signal through the activin type II receptors (ActRII and ActRIIB). In a previous report, we demonstrated that fish possess at least 2 genes for ActRIIB: ActRIIB-1 and ActRIIB-2, which differ in their amino acid sequence. We also showed that affinity-purified, fish-soluble ActRIIB-1 (extracellular domain; ECD), produced in the yeast Pichia pastoris, inhibited recombinant mouse/rat/human mature MSTN activity in vitro using a reporter gene assay in the mammalian A204 cell line. In the present study, we produced soluble ActRIIB-2a in P. pastoris, and showed that it is N-glycosylated, similar to soluble ActRIIB-1. Inhibition of MSTN and activin A activities by affinitypurified ActRIIB-2a was compared with that of soluble ActRIIB-1 using the CAGA-luciferase assay in A204 cells. The findings of this study provide evidence that both paralogs, which probably resulted from gene duplication, did not diversify in their functionality (neofunctionalization), but rather retained a similar function. Both ActRIIB isoforms are equally potent in the mammalian system, and both exhibited an inhibitory effect on mammalian MSTN and activin A. Moreover--in spite of the amino acid differences in ECD between the two paralogs--it appears that the residues important for ligand binding are conserved, and that they recognize the mammalian ligands activin A and MSTN to the same extent.


Transforming growth factor-[beta]s (TGF-[beta]s), activins, inhibins, bone morphogenetic proteins (BMPs), growth differentiation factors (GDFs), Lefty, Nodal, and anti-Mullerian hormone make up a large superfamily of secreted signaling proteins that play key roles in embryonic development and adult tissue homeostasis (Wu and Hill, 2009). Members of the TGF-[beta] superfamily bind two different types of serine/threonine-kinase receptors, termed type I and type II receptors. Both receptor subtypes share a common architecture: a small extracellular ligand-binding domain, a single transmembrane domain, and a cytoplasmic serine/threonine-kinase domain. Ligand binding induces hetero-oligomerization of type I and type II receptors, initiating the intracellular signaling cascade. The constitutively active type II serine/threonine-kinase transphosphorylates the type I receptor, thereby activating the type I kinase. The type I kinase, in turn, phosphorylates and activates Smad proteins, which dimerize and migrate to the nucleus. The Smad proteins, in concert with other proteins, function as transcription factors to regulate responsive genes (Massague and Wotton, 2000; Shi and Massague, 2003; de Caestecker, 2004). Myostatin (MSTN; also known as GDF-8) is a member of the TGF-[beta] superfamily, and functions as a negative regulator of skeletal muscle growth in mammals (Lee, 2004; Kollias and McDermott, 2008). Activin A and B are also members of the TGF-[beta] superfamily, and have a variety of actions in both embryonic and adult tissues, including tissue fate determination in vertebrate embryos, regulation of follicle-stimulating hormone biosynthesis in pituitary gonadotrophs, and regulation of gametogenesis in ovaries and testes (Welt et al., 2002). A recent report suggested that activins are also direct repressors of skeletal muscle mass; elevated expression of activins promoted muscle wasting and cachexia in a mice model (Chen et al., 2014).

Like other TGF-[beta] superfamily members, activins and MSTN signal via a heteromeric complex of type II and type I receptors. Activin and MSTN bind directly to one of two isoforms of type II receptors: ActRIIA and ActRIIB (Donaldson et al., 1999; Lee and McPherron, 2001; Rebbapragada et al., 2003; del Re et al., 2004). Ligand-binding assays of COS cells that were transfected with the receptor revealed binding of MSTN to ActRIIB, and, to a lesser extent, to ActRIIA (Lee and McPherron, 2001), and binding of activin A to ActRII (Donaldson et al., 1992). Activin and MSTN association with ActRIIB leads to the recruitment, phosphorylation, and activation of the activin type I receptor, activin receptor-like kinase 4 (ALK4). Activated ALK4 phosphorylates intracellular signaling molecules, Smad2/3, which, in turn, form a complex with the coactivator, Smad4. The resulting Smad oligomer translocates into the nucleus, binds to DNA, and finally modulates transcription of various target genes (Massague and Wotton, 2000; Zhu et al., 2004; Joulia-Ekaza and Cabello, 2006). In addition to activin and MSTN, ActRIIB can bind other TGF-[beta] family members, such as GDF-11, inhibin, Nodal, BMP-2, and BMP-7 (Gray et al., 2000; de Caestecker, 2004; Tsuchida et al., 2008; Sako et al., 2010). Both soluble ActRIIA and ActRIIB blocked the inhibitory effect of GDF11 (also known as BMP 11), activin A, B, and AB on the myoblast-to-myotube differentiation, implicating them as potential novel regulators of muscle growth in mammals, in addition to MSTN (Souza et al., 2008).

One approach to inhibiting the MSTN effect is by blocking its signaling, which is initiated by interaction with the ActRIIB receptor. Transgenic mice that over-expressed a truncated form of ActRIIB (lacking the kinase domain, resulting in loss of signal transduction), exhibited a dramatic increase in muscle mass, similar to that of MSTN knockout mice (Lee and McPherron, 2001). Similarly, transgenic mdx (dystrophin-deficient) mice, carrying a dominant negative ActRIIB gene, had bigger muscles than mdx mice carrying a normal ActRIIB gene (Benabdallah et al., 2005). Injection of wild-type mice with soluble ActRIIB (ActRIIB-ECD) protein increased muscle mass. Treatment of MSTN knockout mice with ActRIIB-ECD resulted in a further increase in muscle mass, compared to the muscle gain resulting from MSTN deletion (Lee et al., 2005). Moreover, administration of ActRIIB-ECD not only prevented muscle wasting, but it also restored prior muscle loss in various cancer cachexia models (Zhou et al., 2010), and increased muscle mass in wild-type mice (Cadena et al., 2010). In a different approach, the blocking of MSTN signaling was achieved by an anti-ActRII antibody. Injection of the antibodies to MSTN-mutant mice induced enhanced muscular hypertrophy, confirming a beneficial effect on muscle growth beyond MSTN inhibition alone through blockade of ActRII ligands (Lach-Trifilieff et al., 2014). Taken together, these studies suggest that agents targeting the MSTN signaling pathway are potentially useful therapeutic agents for preventing muscle wasting in humans, and for increasing muscle mass in farm animals. These agents may also prove useful in fish, thereby improving aquaculture productivity.

ActRIIB cDNAs were cloned from several fish species (Funkenstein et al., 2012 and references therein). Two recent reports in fish used a strategy similar to that described in mammals for enhancing growth: employing ActRIIB-ECD. In the first report, immersion of catfish and tilapia larvae in supernatant from Pichia pastoris culture, which contained recombinant goldfish (Carassius auratus) ActRIIB-ECD, enhanced larval growth (Carpio et al., 2009). The second report showed that transgenic rainbow trout with a deleted ActRIIB had in PI (parent generation) enhanced localized muscling in the hypaxial and epaxial regions of the musculature. Transgenic F1 (first filial generation) individuals at 4 months had increased muscle mass primarily through hypertrophy (Phelps et al., 2013). To our knowledge, Carpio et al. (2009) and Phelps et al. (2013) are the only two publications available in fish that show the potential application of soluble ActRII as an effective enhancer of cultured fish growth. It should be noted, however, that neither report provided evidence to suggest which of the ligands are inhibited and contribute to enhanced growth.

We recently identified two ActRIIB paralogs in the marine fish gilthead sea bream Sparus aurata Linnaeus, 1758: ActRIIB-1 and ActRIIB-2 (with two alleles: 2a and 2b) (Funkenstein et al., 2012). Bioinformatics analysis confirmed the existence of two ActRIIB genes in several other fish species as well, which are probably a result of gene or genome duplication. The two paralogs differ substantially in their amino acid sequence. The inhibitory effect of ActRIIB-ECD on MSTN activity was shown in vitro in A204 cells, using purified, recombinant saActRIIB-l-ECD (sa, from Sparus aurata) produced in the yeast Pichia pastoris and the CAGA-luciferase reporter gene assay [which uses a plasmid consisting of 12 Smad3/Smad4 binding sequences (CAGA boxes) and the luciferase-coding sequence] (Funkenstein et al., 2012). To determine if the second fish ActRIIB paralog, saActRIIB-2, functions in a similar way, in the present study we expressed saActRIIB-2a-ECD in P. pastoris, and compared its inhibitory action to that of ActRIIB-l-ECD on activin A and MSTN activities using the CAGA-luciferase reporter gene assay in A204 cells. Our results provide evidence that both paralogs are equally potent in the mammalian system. They also suggest that--despite the amino acid differences in ECD between the two paralogs--the residues important for ligand binding are conserved, and they recognize the mammalian ligands activin A and MSTN to the same extent.

Materials and Methods

Preparation of saActRIIB-2a-ECD expression construct

The extracellular domain (ECD) of saActRIIB-2a (amino acids 23-133) was obtained from a bacterial clone containing the ActRIIB-2a variant (Funkenstein et al., 2012) by PCR (polymerase chain reaction) amplification using primers:


ActR-9c(5' -TCTAGATTCAGCAGCGCCGGCGCCGG-3'), and cloned into the yeast expression vector, the pPICZ[alpha]A vector (Invitrogen Ltd., Paisley, UK). JM109 transformants (chemically competent JM109 bacterial cells that underwent transformation with the DNA construct) were selected on low-salt LB (Luria broth) plates containing 25 [micro]g/ml Zeocin (InvivoGen, San Diego, CA), and sequenced to confirm the presence of His- and myc-tags in-frame with the ECD of saActRIIB-2a.

Transformation of Pichia pastoris and selection of transformants

The recombinant saActRIIB-2a-ECD/pPICZ[alpha]A plasmid was linearized within the 5'AOXl (alcohol oxidase) promoter region, to direct integration of the expression cassette into the AOX1 locus of the P. pastoris genome. Transformation into the P. pastoris GS115 strain was carried out using the LiCl protocol suggested by the manufacturer (EasySelect Pichia Expression Kit; Invitrogen Ltd.). Zeocin-resistant transformants containing saActRIIB-2a-ECD were selected on yeast extract bactopeptone dextrose sorbitol (YPDS) Zeocin (100 [micro]g/ml) plates.

Screening for expression

Since recombination can occur in different ways and affect the expression level of the recombinant protein, several clones were screened. Induction of protein expression from selected transformants was tested in small-scale cultures, using the protocol suggested by the manufacturer (EasySelect Pichia Expression Kit; Invitrogen Ltd.). The host strain GS115 and a GS115 clone containing the saActRIIB-l-ECD (Funkenstein et al., 2012) were grown in the same conditions and served as negative and positive controls, respectively, for AOX1-mediated induction. Prior to induction with methanol, aliquots were removed from the yeast culture (assigned t0). The presence of saActRIIB-2a-ECD was determined by electrophoresis on 16.3% Tris-Tricine SDS-polyacrylamide gels (SDS-PAGE), using 300 [micro]1 supernatant concentrated with 3 volumes of cold acetone. Two gels were run for each sample: one for staining proteins with Coomassie Brilliant Blue R-250 (Sigma-Aldrich, Rehovot, Israel), and the second gel for protein transfer to nitrocellulose membrane and Western blot analysis.

Expression and purification of saActRIIB-2a-ECD

For biochemical analysis, clone T14 was grown in multiple small-scale productions (8 tubes of 6 ml). Cell culture, induction, and harvest were carried out as described above. Supernatants were pooled and filtered through an Acrodisc Syringe Filter with 0.45 [micro]am Supor Membrane (Pall Corporation, Newquay, Cornwall, UK). Purification of saActRIIB-2a-ECD was performed at room temperature by nickel nitrilotriacetic acid (Ni-NTA) affinity chromatography (Ni-NTA His-Bind Resin; Novagen, Madison, WI) using 0.8 X 4 cm Poly-Prep Chromatography Columns (Bio-Rad Laboratories, Hercules, CA) containing 0.6 ml 100% Ni-NTA resin. The resin was pre-equilibrated with 10 volumes of 20 mmol l-1 Tris-HCl pH 8.5, followed by 10 volumes of a solution containing 50 mmol [1.sup.-1] Tris-HCl pH 8.5, 0.3 mol l_l NaCl, and 10 mmol [1.sup.-1] imidazole. After the filtered supernatant was loaded, the column was washed with 50 ml of a solution containing 50 mmol [1.sup.-1] Tris-HCl pH 8.5, 0.3 mol [1.sup.-1] NaCl, and 20 mmol [1.sup.-1] imidazole until the OD280 (wavelength used to measure protein concentration, with absorbance at 280 nm) of the washing solution approached zero. Protein elution was accomplished by 3 sequential loadings of a solution containing 50 mmol l-1 Tris-HCl pH 8.5 and 0.3 mol l_l NaCl, and increasing concentrations of imidazole: 250 mmol [1.sup.-1], 500 mmol [1.sup.-1], and 1 mol [1.sup.-1] imidazole. Eluted fractions (4 X 0.5 ml fractions for each imidazole concentration) were kept at 4 [degrees]C. Purification and elution were monitored by Tris-Tricine SDS-PAGE (16.5%) under reducing conditions, using 20 [micro]l from each fraction.

For biological activity assays, larger quantities of protein were needed. The protocol for scale-up expression, described in the user manual of the EasySelect Pichia Expression Kit, was used with minor changes. The presence of saActRIIB-2a-ECD in the supernatants was verified prior to purification by concentrating 300 [micro]l from each supernatant with 3 volumes of cold acetone, and analysis by electro-phoresis on 16.3% Tris-Tricine SDS-PAGE and Western blot. Supernatants from 2 positive cultures (150 ml each) were filtered through 0.45-[micro]m filters and loaded onto 6 0.8 X 4 cm Poly-Prep Chromatography Columns, 3 columns for each culture (50 ml supernatant loaded on each column). Each column was packed with 0.6 ml 100% Ni-NTA resin. Column wash, loading, and elution were carried out as described above. Purification and elution were monitored by Tris-Tricine SDS-PAGE (16.5%), using 20 [micro]1 for each fraction under reducing conditions. For comparative studies of the biological activities of the two ActRIIB paralogs, clone #1 of saActRIIB-l-ECD/pPICZ[alpha]A was expressed and purified essentially as described above for saActRIIB-2a-ECD/pPICZaA.

SDS-PAGE and Western blot analysis

Reagents for SDS-PAGE and molecular mass markers were purchased from Bio-Rad Laboratories. Tris-Tricine SDS-PAGE (16.5%) was performed according to Schagger and von Jagow (1987) under reducing conditions. Proteins were stained with Coomassie Brilliant Blue R-250 to visualize general proteins, or with Periodic acid-Schiff (PAS) to visualize glycoproteins.

For Western blot, proteins were electro-transferred, following electrophoresis, onto nitrocellulose membranes (Schleicher & Schuell BioScience, Inc., Keene, NH) at 250 mA for 1.5 h on ice, in transfer buffer consisting of 20 mmol 1_1 Tris base, 148 mmol [1.sup.-1] glycine, and 20% (v/v) methanol. Transfer efficiency was monitored by Ponceau S (Sigma-Aldrich) staining; nonspecific sites were blocked by incubating the membranes in T-TBS buffer (i.e., 100 mmol [1.sup.-1] Tris-HCl, 0.9% NaCl, pH 7.5, and 0.05% Tween-20) containing 0.5% BSA (bovine serum albumin), for 1.5 h at room temperature with gentle agitation.

For immunodetection, membranes were incubated first with the primary antibody [mouse monoclonal anti-His (C-terminal, clone 3D5; Life Technologies, Carlsbad, CA) overnight at 4 [degrees]C using a 1/5000 dilution in T-TBS supplemented with 0.02% sodium azide]. The membranes were then incubated with the secondary antibody [horseradish peroxidase (HRP) conjugated to goat anti-mouse immunoglobulins (EnVision+ K4000 HRP; Agilent Technologies Denmark ApS, Glostrup, Denmark) at room temperature for 1.5 h using a 1/1000 dilution in T-TBS]. Membranes were washed after each incubation (6X7 min) in T-TBS. HRP activity was detected by chemiluminescence. The membranes were incubated for 1 min in 100 mmol [1.sup.-1] Tris-HCl pH 8.5 containing 0.0085% hydrogen peroxide, 1.25 mmol [1.sup.-1] luminol (Fluka, Buchs, Switzerland) diluted from a 250-mmol l-1 stock solution in dimethylsulfoxide, and 0.2 mmol [1.sup.-1] p-Coumaric acid (Sigma-Aldrich) diluted from a 90 mmol [1.sup.-1] p-Coumaric acid stock solution in dimethyl-sulfoxide. The signal was visualized by exposure to Kodak BioMax MS or BioMax Light film (Kodak, Rochester, NY).

Deglycosylation of purified saActRUB-2a-ECD

Glycosylation status of recombinant saActRIIB-2a-ECD was assessed using Peptide N-glycosidase F (PNGase F; New England Biolabs, Ipswich, MA), as described previously (Funkenstein et al, 2012), using 18 [micro]l of fraction #2 of 250 mmol [1.sup.-1] imidazole elution. As a negative control, the same reaction was carried out with [H.sub.2]O instead of the PNGase F enzyme. The digested products were acetone-concentrated and analyzed by 16.5% Tris-Tricine-SDS-PAGE under reducing conditions. Two parallel sets of digestion reactions were run on two gels: one analyzed by Coomassie blue staining to visualize general proteins, and the second by PAS staining to visualize glycoproteins (see next sections).

Periodic acid-Schiff (PAS) staining

PAS staining of the gels was carried out according to the detailed protocol described in Funkenstein et al. (2012), based on the protocol in Thornton et al. (1996). All reagents for the PAS reaction were purchased from Sigma-Aldrich Israel, Ltd., Rehovot, Israel.

Biological activity

The ability of saActRIIB-2a-ECD to inhibit MSTN and activin A activities in vitro was measured using the pGL3-[(CAGA).sub.12]-luciferase reporter assay in A204 rhabdomyosarcoma cells (ATCC, HTB-82), as described in detail previously (Funkenstein et al., 2012)

In brief, A204 cells were transiently cotransfected with 1 [micro]g of pGL3(CAGA[).sup.12]-luciferase construct (a plasmid with the firefly luciferase gene controlled by Smad2/3 response element) (Thies et al., 2001) and 1 ng of the control vector pRL-RSV, expressing Renilla luciferase, using Lipofectamine 2000 (Invitrogen). One day after transfection, the recombinant proteins were added in serum-free growth medium: 1 ng/ml recombinant human/mouse/rat GDF-8/myostatin (R&D Systems, Minneapolis, MN) or 1 ng/ml recombinant human/mouse/rat activin A (R&D Systems), unless indicated otherwise, together with increasing amounts of purified saActRIIB-2a-ECD (previously dialyzed against PBS for ~24 h) or with PBS. Cells were collected 24 h later, lysed in IX Passive Lysis Buffer (PLB; Promega Corporation, Madison, WI), and subjected to 2 successive cycles of freeze-thaw at -70 [degrees]C. Supernatants were analyzed for luciferase reporter gene activity by the Dual Luciferase Reporter Assay System (Promega Corporation) (Funkenstein et al., 2012). The results are expressed as the ratio of firefly to Renilla luciferase activities, and represent the mean value of triplicates, with error bars depicting the SEM.


Expression and characterization of saActRllB-2a-ECD

Transformants of Pichia pastoris yeast strain GS115 harboring saActRIIIB-2a-ECD were grown on YPDS plates containing 100 [micro]g/ml Zeocin. They were screened for expression of the recombinant protein in small-scale cultures (6 ml) at 48 h following induction with methanol. One clone expressing saActRIIB-l-ECD (Funkenstein et al., 2012) served as positive control, and non-transformed GS115 yeast served as negative control. To verify that the expressed protein contained a His-tag, which will facilitate future purification, Western blots were conducted following gel electrophoresis using anti-His monoclonal antibody. As shown in Figure 1A, analysis of the total secreted proteins by Tris-Tricine SDS-PAGE (16.5%) gel revealed a faint band between 25 and 37 kDa for the ECD in clone T10, and a slightly darker band in clone T14. This band was not present in aliquots of P. pastoris cultures taken at t0, before initiation of induction with methanol, and not in GS115 negative control. The positive control of saActRIIB-l-ECD displayed a band of similar electrophoretic mobility. Immunoreaction with anti-His antibody confirmed that clones T10 and T14 contained a His-tag in-frame with the recombinant peptide, as in saActRIIB-l-ECD (Fig. IB). The long smear represents various degrees of glycosylation (see next sections). Clone T14 was chosen for subsequent experiments.

For large-scale expression, clone T14 was cultured in 3 flasks, each containing 150 ml. Aliquots from each flask were tested for expression using Tris-Tricine SDS-PAGE, followed by Western blot analysis. An example of such a test is shown in Figure 2A. The expressed saActRIIB-2a-ECD was purified by Ni-NTA affinity chromatography, using 3 columns for each culture. An example of the elution profile with a solution containing 250 mmol l-1 imidazole (for details, see Materials and Methods) is shown in Figure 2B. Fractions eluted with 250 mmol [1.sup.-1] imidazole showed a band--23 kDa, along with a broad smearing band above the 23 kDa band.

As we reported earlier, both saActRIIB-l-ECD and saActRIIB-2a-ECD have 2 potential asparagine (Asn)-linked glycosylation sites, at [Asn.sup.42] and [Asn.sup.65] (Funkenstein et al., 2012). To ascertain whether recombinant saActRIIB-2a-ECD is also glycosylated, like saActRIIB-1-ECD (Funkenstein et al., 2012), and to determine whether the broad band that was obtained following purification (Fig. 2B) was due to different levels of glycosylation of the same core protein, the purified saActRIIB-2a-ECD was treated with PNGase F. This enzyme hydrolyzes all types of polysaccharide chains linked to Asn, but does not cleave O-linked oligosaccharides. Following deglycosylation treatment with PNGase F, the apparent molecular weight of purified saActRIIB-2a-ECD was reduced from a broad band extending from --23 to 37 kDa, to a strong, intense major band of--15 kDa. This suggested that recombinant saActRIIB-2a-ECD has different levels of N-glycosylation (Fig. 2C, left panel). The molecular mass of the deglycosylated form is in good agreement with the calculated molecular mass of 15,410.01 Da of the fusion protein, including the 6XHis-tag and myc epitope. The agreement between the observed molecular mass following N-deglycosylation and the calculated molecular mass suggests that saActRIIB-2a-ECD has only N-glycosylation. The presence of poly-saccharides was also verified using PAS staining of the purified saActRIIB-2a-ECD following electrophoresis on Tris-Tricine gels (Fig. 2C, right panel). While the broad band of saActRIIB-2a-ECD stained pink with PAS, no staining was seen following treatment with PNGase F, indicating complete removal of carbohydrates.

Inhibition of MSTN and activin A activity by saActRIIB-2a-ECD

The biological activity of affinity-purified saActRIIB-2a-ECD was tested using its ability to inhibit the activity of commercial recombinant mature mouse/rat/human activin A (m/r/hActivin A) and commercial recombinant, mature mouse/rat/human MSTN (m/r/hMSTN), using A204 cells and the pGL3-[(CAGA).sub.12]-luciferase reporter gene assay system.

Increasing concentrations of m/r/hMSTN (2.5-10 ng/ml), added to A204 cells, transiently transfected with the pGL3-(CAGA)12-luciferase plasmid, gradually increased luciferase activity (Fig. 3A), with an ED50 of ~2 ng/ml (Fig. 3B). In all subsequent experiments, 1 ng/ml MSTN was added to A204 cells together with, or without, saActRIIB-1-ECD or saActRIIB-2a-ECD. The addition of purified saActRIIB-l-ECD or saActRIIB-2a-ECD (48 [micro]g/ml) resulted in 49.67% and 57.5% inhibition, respectively, of MSTN activity (Fig. 3C). A similar experimental outline was used to assess inhibition of activin A activity. Increased concentrations of m/r/hActivin A (0.5-5 ng/ml) gradually increased luciferase activity in A204 cells transiently transfected with the pGL3-(CAGA)12-luciferase plasmid (Fig. 4A). The E[D.sub.50] was obtained with 1 ng/ml (Fig. 4B). In all subsequent experiments, 1 ng/ml activin A was added to A204 cells together with, or without, saActRIIB-l-ECD or saActRIIB-2a-ECD. The addition of purified saActRIIB-1-ECD or saActRIIB-2a-ECD (48 [micro]g/ml) resulted in 49.6% and 61% inhibition, respectively, of activin A activity (Fig. 4C). These experiments showed that saActRIIB-l-ECD was slightly less effective than saActRIIB-2a-ECD in inhibiting both MSTN and activin A activity. However, both ECDs were effective in inhibiting MSTN and activin A, which suggests that both ActRIIB paralogs act in a similar way to mammalian peptides.

The inhibition was dose-dependent, as shown in Figure 5. Both saActRIIB-l-ECD and saActRIIB-2a-ECD inhibited MSTN activity, reaching 60% inhibition when 28 [micro]g/well was added together with 1 ng/ml m/r/hMSTN (Fig. 5A). In the experiment shown in Figure 5B, the general trend was of reduced activity of both activin A and MSTN when increasing amounts of saActRIIB-2a-ECD were added. Relatively high doses of saActRIIB-2a-ECD resulted in about 70% inhibition of MSTN and activin A activities.


In the present study, we describe the expression, purification, and biological activity of the soluble form of the second paralog of ActRIIB found in fish: ActRIIB-2a-ECD. We performed a comparison of its activity with the activity of soluble ActRIIB-1-ECD (Funkenstein et al, 2012), which was produced and purified again for this comparison. Both paralogs were cloned from the marine fish gilthead sea bream Sparus aurata. The biological activity was tested in A204 rhabdomyosarcoma cells, transfected with the pGL3-(CAGA)12-luciferase reporter gene plasmid. The CAGA boxes were found in the promoter of plasminogen activator inhibitor-type 1, and serve as binding sites to transcription factors Smad3 and/or Smad4, participating in the signaling pathway of TGF-[beta]S (Dennler ef a/., 1998), including MSTN and activin (Thies et al., 2001; Tsuchida et al., 2009). Inhibition of recombinant mature m/r/hActivin A and recombinant mature m/r/hMSTN activity by the fish-soluble ActRIIB-ECD indicates biological activity of the soluble receptors that we produced in Pichia pastoris. Our results showed that both saActRIIB-1-ECD and saActRIIB-2a-ECD inhibited activin A and MSTN. Therefore, they can be useful inhibitors to the endogenous ligands in fish and serve as potential growth enhancers.

The results of the biochemical analysis of saActRIIB-2a-ECD agreed with our previous results with respect to saActRIIB-l-ECD (Funkenstein et al., 2012). The purified peptide appeared on SDS-PAGE as a smear. A smear was also observed during expression in P. pastoris of human, pig, chicken, and goldfish ActRIIB-ECDs (Daly and Hearn, 2006; Carpio et al., 2009; Kim et al., 2012), as well as our saActRIIB-1-ECD (Funkenstein et al., 2012). Removal of carbohydrates by N-deglycosylation suggested that the smear was the result of various levels of glycosylation. The presence of N-glycosylation was reported for pig and gold-fish ActRIIB-ECDs expressed in P. pastoris (Carpio et al., 2009; Kim et al., 2012). In contrast, chicken and human ActRIIB-ECD produced in P. pastoris (Daly and Hearn, 2006; Kim et al., 2012), and mouse ActRIIB-ECD produced in baculovirus-insect cell culture system (Donaldson et al., 1999) seemed to undergo O-linked glycosylation as well (on serine or threonine residues).

The inhibitory effect of soluble saActRIIB-l-ECD and saActRIIB-2-ECD on luciferase activity in the CAGA assay indicated that both act as inhibitors of MSTN and activin signaling. Most likely, they sequester MSTN and activin away from full-length, membrane-bound ActRIIB. Earlier studies in mammals, in which direct binding assays of soluble ActRII and ActRIIB to radiolabeled activins and inhibins were tested, revealed that both receptors bind activin A with high affinity (del Re et al., 2004). Activin A was preferred over activin B, and inhibin binding to these receptors was reduced 10-fold compared with activin (del Re et al., 2004). Using the indirect CAGA-luciferase reporter assay in A204 cells, Pearsall et al. (2008) showed that ActRIIA-ECD inhibited activin A activity.

In general, higher amounts of the soluble fish receptor were needed to achieve comparable inhibition of MSTN and activin A activities to that which was obtained by chicken and pig ActRIIB-ECD produced in P. pastoris and tested by the CAGA-Luc assay in A204 cells (Kim et al, 2012). Several possibilities may explain our findings. Among others is the presence of 6xHis-tag and a myc epitope at the C-terminus of the fusion protein. Sako et al. (2010) showed that the C-terminal portion of ActRIIB-ECD plays a role in binding affinity between the receptor and activin A and GDF-11. It is unlikely that various degrees of glycosylation were a reason for the need for high amounts of the soluble receptor for inhibiting MSTN and activin A; previous studies have shown that ActRIIB-ECD glycosylation is not essential for its activity (Sako et al., 2010; Kim et al., 2012). Another explanation may be related to the suggestion that activin (and perhaps also MSTN) form a complex with two type II receptors, which may have tighter binding energies than do the soluble complexes used in the study reported here. Higher amounts are needed to sequester the ligand from its endogenous receptor. Another possibility is that, while fish ECD is conserved, compared with mammalian ECD, the affinity between the mammalian ligands that we used was higher for the endogenous mammalian receptor on the A204 cell membranes.

Using cross-linking experiments, Greenwald et al. (1998) showed binding between radio-labeled activin and mouse ActRIIB-ECD expressed in P. pastoris. These researchers also showed that glycosylation is not essential for high-affinity interaction between ActRIIB-ECD and activin A. Then Donaldson et al. (1999) showed that soluble ActRIIB-ECD expressed in insects also possesses an intrinsic ability to bind ligands with high affinity, as determined by crosslinking experiments with activin A and inhibin A. The chemical cross-linking experiment revealed that soluble ECD binds the monomer of inhibin/activin [[beta].sub.A] or dimeric activin A ([[beta].sub.A]/[[beta].sub.A]) (Donaldson et al., 1999). A detailed kinetic characterization of soluble ActRIIB binding to several low- and high-affinity ligands demonstrated that both MSTN and GDF-11 bind ActRIIB-ECD with affinities comparable to those of activin A (Sako et al., 2010).

Number and position of amino acid residues in mouse ActRIIB, shown earlier to be directly involved in binding to activin (Gray et al., 2000); Thompson et al., 2003), are also conserved in saActRIIB-l-ECD and saActRIIB-2a-ECD. This suggested that mature m/r/hMSTN, which binds to human ActRIIB-ECD (Sako et al., 2010), has the potential to interact with fish ActRIIB-ECD. The residues identified as being important for activin binding (Thompson et al., 2003) are also conserved between fish ActRIIB paralogs and human ActRIIB (Fig. 6). Furthermore, the ability of soluble ActRIIB-ECD by itself to inhibit activin A or MSTN activity also indicates that neither the transmembrane nor the intracellular domain of the full-length activin type II receptor is necessary for ligand binding.

Previous studies have demonstrated that the ActRIIB-ECD sequence is exceptionally conserved, with only one amino acid difference between mice and humans, and ~90% identity between species as divergent as chickens and humans (Sako et al., 2010). Although the residues important for activin binding are conserved between saActRIIB-l-ECD, saActRIIB-2a-ECD, and saActRIIB-2b-ECD (Funkenstein et al., 2012), there are differences in the C-terminus of the ECD, including an insertion of 4 residues, which may affect binding affinity to MSTN. Four different isoforms of ActRIIB were found in mice and humans (ActRIIB1, ActRIIB2, ActRIIB3, and ActRlIB4). The ECDs of ActRIIB1 and ActRIIB2, contain an insertion in the C-terminal portion of the ECD, which is absent in isoforms ActRIIB3 and ActRIIB4. The biological significance of the different isoforms remains unclear. The ActRIIB2 isoform is most predominant in humans. It was previously suggested that the longer isoforms are the most potent. For example, the longer ActRIIB1 and ActRIIB2 isoforms have a three- to four-fold higher affinity for activin A than do the shorter isoforms, ActRIIB3 and ActRIIB4 (Attisano et al., 1992). The A204 cell line used for the biological assay in the present study was derived from human rhabdomyosarcoma cells. Presumably these cells express type II receptors; studies (Thies et al,, 2001; Lee et al., 2005) have shown that these cells are responsive to MSTN and activin. Furthermore, these cells produce and secrete MSTN (Yang et al., 2008). Hence, our bioassay was carried out in serum-free medium, and cells were rinsed before the exogenous ligands and soluble receptors were added.

In conclusion, the current study showed that ActRIIB-2a-ECD, the second paralog of ActRIIB expressed in fish, has a similar potency to ActRIIB-l-ECD, and can suppress MSTN and activin A activities in vitro. In addition, we showed that bioactive fish ActRIIB-1-ECD and ActRIIB-2a-ECD can be produced in Pichia pastoris. Fish ActRIIB-ECD suppresses MSTN and activin A activities, presumably in a way similar to that of ActRIIB-ECD from mammalian and avian species. From an evolutionary point of view, this study demonstrates that the two fish paralogs, which probably resulted from gene duplication, did not diversify in their functionality, but retained a similar function. The presence of at least two paralogs of ActRIIB in fish--and their equal potency in inhibiting both MSTN and activin A--should be taken into account when designing experiments in fish that are aimed at inhibiting MSTN activity for enhancing muscle growth.


We are grateful to Dr. Whittemore for the A204 cells and the pGL3-(CAGA)12-luciferase plasmid, and to Dr. M. Walker for the pRL-RSV plasmid. This work was supported in part by a grant from the Israel Science Foundation (ISF, Project No. 46/08).

Literature Cited

Attisano, L., J. L. Wrana, S. Cheifetz, and J. Massague. 1992. Novel activin receptors: distinct genes and alternative mRNA splicing generate a repertoire of serine/threonine kinase receptors. Cell 68: 97-108.

Benabdallah, B. F,, M. Bouchentouf, and J. P. Tremblay. 2005. Improved success of myoblast transplantation in mdx mice by blocking the myostatin signal. Transplantation 79: 1696-1702.

Cadena, S. M., K. N. Tomkinson, T. E. Monnell, M. S. Spaits, R. Kumar, K. W. Underwood, R. S. Pearsall, and J. L. Lachey. 2010. Administration of a soluble activin type IIB receptor promotes skeletal muscle growth independent of fiber type. J. Appl. Physiol. 109: 635-642.

Carpio, Y., J. Acosta, R. Morales, Y. Santisteban, A. Sanchez, and M. P. Estrada. 2009. Regulation of body mass growth through activin type IIB receptor in teleost fish. Gen. Comp. Endocrinol. 160: 158-167.

Chen, J. L., K. L. Walton, C. E. Winbanks, K. T. Murphy, R. E. Thomson, Y. Makanji, H. Qian, G. S. Lynch, C. A. Harrision, and P. Gregorevic. 2014. Elevated expression of activins promotes muscle wasting and cachexia. FASEB J. 28: 1711-1723.

Daly, R., and M. T. W. Hearn. 2006. Expression of the human activin type I and II receptor extracellular domains in Pichia pastoris. Protein Expr. Purif. 46: 456-467.

de Caestecker, M. 2004. The transforming growth factor-[beta] superfamily of receptors. Cytokine Growth Factor Rev. 15: [1.sup.-1]1.

del Re, E., Y. Sidis, D. A. Fabrizio, H. Y. Lin, and A. Schneyer. 2004. Reconstitution and analysis of soluble inhibin and activin receptor complexes in a cell-free system. J. Biol. Chem. 279: 53126-53135.

Dennler, S., S. Itoh, D. Vivien, P. ten Dijke, S. Huet, and J.-M. Gauthier. 1998. Direct binding of Smad3 and Smad4 to critical TGF[beta]-inducible elements in the promoter of human plasminogen activator inhibitor-type 1 gene. EMBO J. 17: 3091-3100.

Donaldson, C. J., L. M. Mathews, and W. W. Vale. 1992. Molecular cloning and binding properties of the human type II activin receptor. Biochem. Biophys. Res. Commun. 184: 310-316.

Donaldson, C. J., J. M. Vaughan, A. Z. Corrigan, W. H. Fischer, and W. W. Vale. 1999. Activin and inhibin binding to the soluble extracellular domain of activin receptor II. Endocrinology 140: 1760-1766.

Funkenstein, B., E. Krol, E. Esterin, and Y.-S. Kim. 2012. Structural and functional characterizations of activin type 2B receptor (acv[r.sup.2]b) ortholog from the marine fish, gilthead sea bream Sparus aurata: evidence for gene duplication of acv[r.sup.2]b in fish. J. Mol. Endocrinol. 49: 175-192.

Gray, P. C., J. Greenwald, A. L. Blount, K. S. Kunitake, C. J. Donaldson, S. Choe, and W. Vale. 2000. Identification of a binding site on the type II activin receptor for activin and inhibin. J. Biol. Chem. 275: 3206-3212.

Greenwald, J., V. Le, A. Corrigan, W. Fischer, E. Komives, W. Vale, and S. Choe. 1998. Characterization of the extracellular ligand-binding domain of the type II activin receptor. Biochemistry 37: 16711-16718.

Joulia-Ekaza, D., and G. Cabello. 2006. Myostatin regulation of muscle development: molecular basis, natural mutations, physiopathological aspects. Exp. Cell Res. 312: 2401-2414.

Kim, Y. S., K. H. Kim, and C. J. Kim. 2012. Production of bioactive extracellular domain of pig and chicken activin type IIB receptors in Pichia pastoris. Process Biochem. 47: 139-146.

Kollias, H. D., and J. C. McDermott. 2008. Transforming growth factor-[beta] and myostatin signaling in skeletal muscle. J. Appl. Physiol. 104: 579-587.

Lach-Trifilieff, E., G. C. Minetti, K. Sheppard, C. Ibebunjo, J. N. Feige, S. Hartmann, S. Brachat, H. Rivet, C. Koelbing, F. Morvan et al. 2014. An antibody blocking activin type II receptor induces strong skeletal muscle hypertrophy and protects from atrophy. Mol. Cell. Biol. 34: 606-618.

Lee, S.-J. 2004. Regulation of muscle mass by myostatin. Annu. Rev. Cell Dev. Biol. 20: 61-86.

Lee, S.-J., and A. C. McPherron. 2001. Regulation of myostatin activity and muscle growth. Proc. Natl. Acad. Sci. USA 98: 9306-9311.

Lee, S.-J., L. A. Reed, M. V. Davies, S. Girgenrath, M. E. P. Goad, K. N. Tomkinson, J. F. Wright, C. Barker, G. Ehrmantraut, J. Holmstrom et al. 2005. Regulation of muscle growth by multiple ligands signaling through activin type II receptors. Proc. Natl. Acad. Sci. USA 102: 18117-18122.

Massague. J., and D. Wotton. 2000. Transcriptional control by the TGF-[beta]Smad signaling. EMBO J. 19: 1745-1754.

Pearsall, R. S., E. Canalis, M. Cornwall-Brady, K. W. Underwood, B. Haigis, J. Ucran, R. Kumar, E. Pobre, A. Grinberg, E. D. Werner et al. 2008. A soluble activin type IIA receptor induces bone formation and improves skeletal integrity. Proc. Natl. Acad. Sci. USA 105: 7082-7087.

Phelps, M. P., I. M. Jaffe, and T. M. Bradley. 2013. Muscle growth in teleost fish is regulated by factors utilizing the activin II B receptor. J. Exp. Biol. 216: 3742-3750.

Rebbapragada, A., H. Benchabane, J. L. Wrana, A. J. Celeste, and L. Attisano. 2003. Myostatin signals through a transforming growth factor [beta]-like signaling pathway to block adipogenesis. Mol. Cell. Biol. 23: 7230-7242.

Sako D., A. V. Grinberg, J. Liu, M. V. Davies, R. Castonguay, S. Maniatis, A. J. Andreucci, E. G. Pobre, K. N. Tomkinson, T. E. Monnell et al. 2010. Characterization of the ligand binding functionality of the extracellular domain of activin receptor type IIB. J. Biol. Chem. 285: 21037-21048.

Schagger, H., and G. von Jagow. 1987. Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166: 368-379.

Shi, Y., and J. Massague. 2003. Mechanism of TGF-[beta] signaling from cell membrane to the nucleus. Cell 113: 685-700.

Souza, T. A., X. Chen, Y. Guo, P. Sava, J. Zhang, J. J. Hill, P. J. Yaworsky, and Y. Qiu. 2008. Proteomic identification and functional validation of activins and bone morphogenetic protein 11 as candidate novel muscle mass regulators. Mol. Endocrinol. 22: 2689-2702.

Thies, R. S., T. Chen, M. V. Davies, K. N. Tomkinson, A. A. Pearson, Q. A. Shakey, and N. M. Wolfman. 2001. GDF-8 propeptide binds to GDF-8 and antagonizes biological activity by inhibiting GDF-8 receptor binding. Growth Factors 18: 251-259.

Thompson, T. B., T. K. Woodruff, and T. S. Jardetzky. 2003. Structures of an ActRIIB:activin A complex reveal a novel binding mode for TGF-beta ligand:receptor interactions. EMBO J. 22: 1555-1566.

Thornton, D. J., I. Carlstedt, and J. K. Sheehan. 1996. Identification of glycoproteins on nitrocellulose membranes and gels. Mol. Biotechnol. 5: 171-176.

Tsuchida, K., M. Nakatani, A. Uezumi, T. Murakami, and X. Cui. 2008. Signal transduction pathway through activin receptors as a therapeutic target of musculoskeletal diseases and cancer. Endocr. J. 55: 11-21.

Tsuchida, K., M. Nakatani, K. Hitachi, A. Uezumi, Y. Sunada, H. Ageta, and K. Inokuchi. 2009. Activin signaling as an emerging target for therapeutic intervention. Cell Commun. Signal. 7: 15.

Welt, C., Y. Sidis, H. Keutmann, and A. Schneyer. 2002. Activins, inhibins, and follistatins: from endocrinology to signaling. A paradigm for the new millennium. Exp. Biol. Med. 227: 724-752.

Wu, M. Y., and C. S. Hill. 2009. TGF-[beta] superfamily signaling in embryonic development and homeostasis. Dev. Cell 16: 329-343.

Yang, Z., J. Zhang, H. Cong, Z. Huang, L. Sun, C. Liu, and P. Tien. 2008. A retrovirus-based system to stably silence GDF-8 expression and enhance myogenic differentiation in human rhabdomyosarcoma cells. J. Gene Med. 10: 825-833.

Zhou, X., J. L. Wang, J. Lu, Y. Song, K. S. Kwak, Q. Jiao, R. Rosenfeld, Q. Chen, T. Boone, W. S. Simonet et al. 2010. Reversal of cancer cachexia and muscle wasting by ActRIIB antagonism leads to prolonged survival. Cell 142: 531-543.

Zhu, X., S. Topouzis, L.-F. Liang, and R. L. Stotish. 2004. Myostatin signaling through Smad2, Smad3 and Smad4 is regulated by the inhibitory Smad7 by a negative feedback mechanism. Cytokine 26: 262-272.


Department of Marine Biology & Biotechnology, National Institute of Oceanography, Israel Oceanographic and Limnological Research, Tel-Shikmona, P.O.B 8030, Haifa 31080, Israel

Received for publication 29 March 2015; accepted 12 November 2015.

(*) To whom correspondence should be addressed. E-mail:

Abbreviations: ActRII, activin type II receptor; ALK, activin receptor-like kinase; AOX, alcohol oxidase; BMP. bone morphogenetic protein; CAGA-Luciferase assay, analytical procedure using a plasmid consisting of 12 Smad3/Smad4 binding sequences (CAGA boxes) and the luciferase-coding sequence; cDNA, complementary, double-stranded DNA synthe-sized from messenger RNA (mRNA); ECD, extracellular domain; FI and PI, first filial generation of offspring (FI) from parents (PI); GDF, growth differentiation factor; MSTN, myostatin; NTA, nitrilotriacetic acid; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; pRL-RSV, a wildtype control reporter gene plasmid consisting of a Renilla luciferase coding sequence and Rous sarcoma virus promoter; Smad, a portmanteau of Drosophila mothers against decapentaplegic (MAD) and Caenorhabditis elegans small body size (SMA) proteins; saActRII, activin type II receptor protein produced from seabream Sparus aurata (sa); SDS, sodium dodecyl sulfate; TGF-[beta], transforming growth factor-[beta]; Tris-HCL, hydroxymethyl aminomethane hydrochloride; YPDS, yeast extract bactopeptone dextrose sorbitol.
COPYRIGHT 2016 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Nadjar-Boger, Elisabeth; Krol, Ekaterina; Funkenstein, Bruria
Publication:The Biological Bulletin
Article Type:Report
Date:Feb 1, 2016
Previous Article:Plastic sexual expression in the androdioecious barnacle Octolasmis warwickii (Cirripedia: pedunculata).
Next Article:Octocoral Sarcophyton auritum Verseveldt & Benayahu, 1978: Microanatomy and presence of collagen fibers.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters