PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of elifeeLifeRecent contentAbout eLifeFor authorsSign up for alerts
 
eLife. 2017; 6: e17935.
PMCID: PMC5360445

SMOC can act as both an antagonist and an expander of BMP signaling

Hugo J Bellen, Reviewing editor
Hugo J Bellen, Baylor College of Medicine, United States;

Abstract

The matricellular protein SMOC (Secreted Modular Calcium binding protein) is conserved phylogenetically from vertebrates to arthropods. We showed previously that SMOC inhibits bone morphogenetic protein (BMP) signaling downstream of its receptor via activation of mitogen-activated protein kinase (MAPK) signaling. In contrast, the most prominent effect of the Drosophila orthologue, pentagone (pent), is expanding the range of BMP signaling during wing patterning. Using SMOC deletion constructs we found that SMOC-[increment]EC, lacking the extracellular calcium binding (EC) domain, inhibited BMP2 signaling, whereas SMOC-EC (EC domain only) enhanced BMP2 signaling. The SMOC-EC domain bound HSPGs with a similar affinity to BMP2 and could expand the range of BMP signaling in an in vitro assay by competition for HSPG-binding. Together with data from studies in vivo we propose a model to explain how these two activities contribute to the function of Pent in Drosophila wing development and SMOC in mammalian joint formation.

DOI: http://dx.doi.org/10.7554/eLife.17935.001

Research Organism: D. melanogaster, Xenopus

eLife digest

During the development of an embryo, a group of proteins known as growth factors stimulate cells to divide and direct how organs and limbs form. One family of growth factors called bone morphogenetic proteins (BMPs) regulate the formation of bone and many other tissues in the embryo. BMPs are released from cells, diffuse away and are then detected by other cells. When BMPs attach to docking station-like structures on the cell surface, called receptors, they stimulate a signaling process inside the cell. In 2009, researchers found that a protein called SMOC blocks BMP activity in animals with backbones by triggering an interfering signal inside the cell. In flies, however, the equivalent protein can make BMP diffuse further from the cell that releases it.

To find out how SMOC can do both of these things, Thomas et al. – including some of the researchers involved in the 2009 study – conducted experiments to see which parts of SMOC are required to either block BMP signaling or encourage the diffusion of BMP. These experiments revealed that one end of SMOC can stick to molecules on the cell surface that are not receptors but are molecules where BMP can also bind. When this end of SMOC attaches to these sites, BMPs cannot bind and so diffuse further away.

Thomas et al. then produced complete or shortened versions of SMOC proteins to see how this affected BMP activity in frogs. These experiments indicated that the opposite end of SMOC is required for short-circuiting the BMP signal. The results also showed that, at lower concentrations, SMOC stimulates BMPs to diffuse, and that higher concentrations are required to block BMP signaling.

These findings suggest that similar to flies, SMOC can also stimulate BMPs to diffuse from the cell in animals with backbones. The next step will be to identify the cell surface receptor for SMOC to better understand the molecular mechanisms that inhibit BMP. The SMOC pathway could be targeted for therapeutic strategies to combat diseases associated with errors in BMP signaling like osteoarthritis, or in cell-based therapies where BMP signaling must be inhibited to produce cells needed to repair damaged tissues.

DOI: http://dx.doi.org/10.7554/eLife.17935.002

Introduction

During development bone morphogenetic proteins (BMPs), comprising at least twenty structurally-related members of the transforming growth factor β (TGF-β) superfamily, are involved in many growth and differentiation events essential for determining body structure (Wu and Hill, 2009). Establishing temporospatial gradients or restricted distributions of BMP signaling is important for many of these processes, which are regulated by a number of mechanisms: ligand binding by extracellular BMP antagonists (Brazil et al., 2015), intracellular feedback inhibition downstream of the BMP receptor (Massagué et al., 2005), spatially restricted proteolytic processing (Thomas et al., 2006), and promotion or restriction of diffusion by interactions with extracellular matrix proteins such as collagen type IV (Umulis et al., 2009) and heparan sulfate proteoglycans (HSPGs) (Matsumoto et al., 2010; Belenkaya et al., 2004). In addition, BMP signaling is also influenced via communication with other signaling pathways, particularly those that act through mitogen-activated protein kinases (MAPKs). MAPK-directed phosphorylation of BMP receptor-regulated Smad 1/5/8 in the linker region inhibits BMP signaling by blocking Smad translocation to the nucleus (Alarcón et al., 2009; Sapkota et al., 2007).

We showed previously that SMOC, a matricellular protein associated with basement membranes (Vannahme et al., 2002) and expressed in developing brain, branchial arches, eye, pronephros, limb bud cartilage condensations, and joint interzones (Okada et al., 2011; Rainger et al., 2011; Thomas et al., 2009), inhibits BMP signaling (Thomas et al., 2009); following the addition of BMP2 to NIH3T3 fibroblasts transfected with SMOC, downstream phosphorylation of Smad 1/5/8 is blocked (Thomas et al., 2009). In Xenopus ectodermal explants (animal caps), SMOC was shown to activate MAPK signaling and inhibit Smad 1/5/8-mediated BMP signaling downstream of the constitutively active BMP receptor, BMPR1B (Thomas et al., 2009). Although the exact mechanism is not known, the activity was lost in the presence of non-phosphorylatable linker-mutant Smad, suggesting that BMP inhibition results from activation of MAPK signaling and subsequent ubiquitination and degradation of Smad following linker Smad phosphorylation (Sapkota et al., 2007). In Drosophila, the SMOC orthologue pentagone (pent) is expressed in developing wing imaginal discs and has also been shown to inhibit BMP signaling (Vuilleumier et al., 2010). However, Pent did not appear to inhibit BMP signaling in the presence of the constitutively active zebrafish BMPR1 receptor, Alk8 (Vuilleumier et al., 2010). Structurally, SMOC and Pent are similar, containing an N-terminal follistatin-like domain (FS), followed by two thyroglobulin-like domains separated by a non-homologous domain, and a C-terminal extracellular calcium binding (EC) domain (Vannahme et al., 2002; Vuilleumier et al., 2010). Although Pent contains an additional EC domain located between the two thyroglobulin-like domains, their phylogenetic conservation would predict similar functions. In Drosophila, Pent binds to the cell surface HSPG, Dally, which is required for Pent to extend the range of BMP signaling during Drosophila wing patterning (Vuilleumier et al., 2010). While SMOC has also been shown to bind to HSPGs (Klemenčič et al., 2013), there have been no reports of SMOC promoting BMP signaling at a distance from its source. Indeed, the dual function of SMOC/Pent as a BMP inhibitor and an expander of BMP signaling have not been reconciled. Here, we present the first evidence showing how different domains within SMOC function either to inhibit BMP signaling locally or expand its range of effect.

Results and discussion

Drosophila Pent can inhibit BMP signaling in Xenopus downstream of the BMP receptor

To support the applicability of functional information from Pent to the vertebrate SMOC (Figure 1A), we first confirmed that a pent cDNA construct was biologically active in Drosophila. As reported previously (Vuilleumier et al., 2010), compared to controls, flies homozygous for pent mutations display a characteristic truncation of the L5 longitudinal vein of the adult wing (Figure 1—figure supplement 1). When pent was expressed in its normal location during wing development, using the Gal4/UAS system, the mutant phenotype was rescued completely, demonstrating that the pent construct had full biological activity (Figure 1—figure supplement 1). Initial injections of pent mRNA into Xenopus embryos produced no apparent effects (not shown); however overexpression of a synthetic pent mRNA (co-pent) optimized for codon usage and translation efficiency in Xenopus (Villegas and Kropinski, 2008) (Figure 1—figure supplement 2) produced a dorsalized phenotype indistinguishable from that observed following overexpression of Xenopus SMOC-1 (XSMOC-1) (Figure 1B). The ability of Pent to inhibit BMP signaling downstream of the BMP receptor was analyzed in Xenopus ectodermal explants (animal caps) following co-injection of mRNAs for co-pent and constitutively active BMP receptor1B (caBMPR1B); the caBMPR1B, containing an intracellular activating mutation (Q203D), promotes phosphorylation of Smad 1/5/8 and subsequent BMP signaling events independent of ligand binding (Zou et al., 1997). As expected, XVent, a direct downstream target of BMP signaling (Gawantka et al., 1995), was strongly expressed in animal caps from embryos injected with caBMPR1B alone (Figure 1C). In contrast, XVent was markedly reduced in caps from embryos co-injected with caBMPR1B and either co-pent or XSMOC-1 (Figure 1C); these results suggested that, as we had shown previously for SMOC (Thomas et al., 2009), Pent can inhibit BMP signaling downstream of the BMPR1B receptor. It is unclear why, in a previous report (Vuilleumier et al., 2010), Pent was able to rescue ventralization caused by overexpression of Bmp2b in Zebrafish, but not following overexpression of the Zebrafish caBMPR1, Alk8. As both BMPR1B (Alk6) and Alk8 are type I BMP receptors that activate Smad 1/5/8, a possible reason could be the amounts of mRNA injected and/or the amounts of protein produced. In our previous study (Thomas et al., 2009), SMOC transfected NIH3T3 cells were able to inhibit BMP signaling in the presence of excess BMP2 (100 ng/ml); where the limiting factor would be the number of BMP receptors on the cells. Conversely, in order for SMOC/Pent to inhibit BMP signaling in the presence of caBMPR1B, SMOC/pent needed to be overexpressed at a 2:1 ratio (Figure 1C); SMOC was not able to inhibit BMP signaling in the presence of an excess of caBMPR1B (not shown).

Figure 1.
Xenopus SMOC-1 and Drosophila pent are orthologues that inhibit BMP signaling downstream of the BMP receptor.

Having established that Pent can function as a BMP antagonist in Xenopus assays, we wanted to determine whether SMOC could function as an expander of BMP signaling in Drosophila. However, attempts to express XSMOC-1 in Drosophila using a synthetic construct optimized for codon usage in Drosophila were unsuccessful; immunoblot analysis of two Drosophila lines generated to overexpress XSMOC-1 demonstrated that XSMOC-1 was below the level of detection (5 ng) for the assay (Figure 1—figure supplement 1), suggesting that despite codon optimization, XSMOC-1 was not translated in amounts sufficient to be effective.

SMOC expressed in bacteria and refolded is biologically active in Xenopus and in mammalian cell lines

As SMOC and Pent are structurally similar, SMOC may function as an expander of BMP signaling in vertebrates. To address this possibility we developed assays using SMOC expressed in bacteria and refolded together with two SMOC deletion mutant constructs; XSMOC-1[increment]EC lacking the EC domain, and XSMOC-1EC containing the EC domain only. The EC domain was of interest as hSMOC-1 binds to heparan sulphate proteoglycans (HSPGs) via the EC domain (Klemenčič et al., 2013) and the expander function of Pent is associated with binding to the cell surface-associated HSPG, Dally (Vuilleumier et al., 2010). For expression of XSMOC-1 in bacteria, the predicted signal peptide (2-24) was omitted and a C-terminal hexahistidine-tag added. When first expressed, two predominant induced proteins were observed on SDS-PAGE (Figure 2A); one migrating at 49 kDa, the other at approximately 24 kDa. Protein sequencing revealed the 49 kDa protein to be XSMOC-1, whereas the 24 kDa protein was a partial XSMOC-1 sequence beginning at V235. The base sequence (GTG) was consistent with an alternative start codon (Villegas and Kropinski, 2008); when changed to GTA by site-directed mutagenesis only the expected 49 kDa product was produced (Figure 2A). All subsequent XSMOC-1 and XSMOC-1[increment]EC constructs contained GTA at V235.

Figure 2.
Expression and refolding of recombinant mature XSMOC-1, XSMOC-1[increment]EC, and XSMOC-1EC.

Initial XSMOC-1 refolding studies were conducted using the protocol described previously, where calcium is absent, and produces hSMOC-1 that is monomeric (Novinec et al., 2008). Analysis by S-200 size-exclusion chromatography (SEC) showed a mixture of poorly separated peaks, one of which had a calculated molecular weight of 45.5 kDa, approximate to the predicted 49.6 kDa of monomeric XSMOC-1 (Figure 2B). However, when tested in the Xenopus animal cap assay the protein was inactive (not shown). As occupancy of the calcium binding sites may be necessary for biological activity, we modified the refolding buffer to include 2 mM CaCl2. With this change, in addition to poorly-separated higher molecular weight material, both XSMOC-1 (Figure 2C) and XSMOC-1[increment]EC (Figure 2D) migrated as single, symmetrical peaks. Their calculated molecular weights of 95.4 kDa and 53 kDa, respectively, were approximately twice the predicted monomeric sizes (49.6 kDa and 32.3 kDa). This suggested that XSMOC-1 and XSMOC-[increment]EC refolded in the presence of calcium form dimers. Furthermore, hSMOC-1, shown previously to elute as a monomer (Novinec et al., 2008), also migrated as an apparent dimer under these conditions with a calculated molecular weight of 90.3 kDa (Figure 2C); no peak was observed at the expected monomeric size of 46.6 kDa. In contrast, XSMOC-1EC eluted at the calculated molecular weight of 23.7 kDa (Figure 2E), consistent with that predicted for a monomer (18.4 kDa). Dimeric XSMOC-1 could not be dissociated by chelation of Ca++ ions and continued to elute as a dimer following dialysis in the presence of 10.5 mM EDTA (Figure 2—figure supplement 1). Indeed, a dimer was still observed in the presence of 50 mM EDTA and 1 mM nitriloacetic acid (not shown). Analysis by SDS-PAGE under non-reducing and reducing conditions demonstrated that dimeric XSMOC-1 and XSMOC-[increment]EC were not formed through disulfide linkages (Figure 2—figure supplement 1).

BMP inhibition and neural induction by SMOC does not require the EC domain

The biological activity of the SMOC proteins was assessed using the Xenopus animal cap explant assay in which overexpression of XSMOC-1 mRNA was shown previously to induce anterior neural markers (Thomas et al., 2009). XSMOC-1, XSMOC-1[increment]EC, or XSMOC-1EC proteins were incubated with stage 9 (Nieuwkoop and Faber, 1994) late blastula animal caps, at equimolar concentrations, until wild type embryos reached the late neurula stage (stage 20–21). RT-PCR demonstrated the induction of anterior neural markers in the presence of XSMOC-1 and XSMOC-1[increment]EC, but not XSMOC-1EC (Figure 3A); hSMOC-1 was also effective in this assay (not shown). For these studies, the optimal concentration of SMOC was found to be 100 μg/ml, which appears relatively high. However, SMOC is a matricellular protein and though it is difficult to estimate the effective concentration in vivo, its affinity for HSPGs suggests that its diffusion will be restricted unless HSPG sites are saturated. Consequently, it may remain concentrated near the site of secretion and thus achieve high levels locally. Temporal analyses showed that a two-hour pulse of XSMOC-1 or XSMOC-1[increment]EC protein was sufficient to commit the naïve ectoderm of the Xenopus animal cap to an anterior neural fate sixteen hours later (Figure 3B); a one-hour pulse was not. This suggests that, following a two hour exposure to SMOC, a duration of exposure (Rogers and Schier, 2011) is reached whereby sufficient changes in gene transcription occur in SMOC-responsive cells to convert their fate from epidermal to neural. While it is well established that inhibition of endogenous BMP activity in Xenopus ectodermal explants promotes a neural fate (Vonica and Brivanlou, 2006) and a number of genes have been implicated in neural fate specification (Kishi et al., 2000; Ueno et al., 2008; Milet et al., 2013; Green and Vetter, 2011; Gammill and Sive, 2001), the sequence of events resulting in commitment to the neural lineage is not known. It should now be feasible to design an unbiased genome-wide screen to identify the early transcriptional changes, following a two hour exposure to SMOC, that initiate neural differentiation.

Figure 3.
XSMOC-1 and XSMOC-1[increment]EC, but not XSMOC-1EC convert the fate of naïve Xenopus ectoderm explants (animal caps) to anterior neural tissue within two hours.

The ability of the SMOC proteins to inhibit BMP signaling was assessed in mammalian cell lines. Serum-starved NIH-3T3 and HEK-293 cells were incubated with either XSMOC-1 (100 μg/ml) or equimolar amounts of XSMOC-1[increment]EC or XSMOC-1EC for thirty minutes, followed by the addition of BMP2 (50 ng/ml) for an additional thirty minutes. As expected, BMP2 treatment alone caused phosphorylation of Smad 1/5/8 (Figure 4A,B). This was blocked in the presence of XSMOC-1 and XSMOC-1[increment]EC, whereas XSMOC-1EC significantly enhanced BMP2-mediated Smad phosphorylation (Figure 4A,B). Potentiation of Smad phosphorylation by the EC domain was not due to an additive effect, as the addition of XSMOC-1EC alone to NIH-3T3 cells did not result in Smad1/5/8 phosphorylation (Figure 4C).

Figure 4.
The two thyroglobulin-like domains are necessary for BMP inhibition, whereas the EC domain promotes BMP signaling.

Having established that the SMOC EC domain is not required for the inhibition of BMP signaling we designed deletion constructs for use in mRNA overexpression studies to determine which domain(s) of SMOC are required for BMP inhibition; XSMOC-1[increment]FS contained a 49 amino acid deletion of the N-terminal FS-like domain ([increment]Q43 to A91); XSMOC-1[increment]Tg1 contained only the follistatin-like domain and EC domain ([increment]K95 to S304); XSMOC-1[increment]FS[increment]EC (referred to as XSMOC-1Tg1) contained deletions of the FS-like and EC domains ([increment]Q43 to A91 and [increment]N310 to end), leaving only the two Tg1-like domains. The effect on BMP inhibition was analyzed in Xenopus embryos following overexpression of mRNAs for each deletion construct and caBMPRIB. Analysis of animal caps following co-injection of Xenopus embryos with caBMPRIB and XSMOC-1, XSMOC-1[increment]FS, or XSMOC-1[increment]EC showed inhibition of BMP signaling, indicated by the suppression of XVent expression (Figure 4D). Whereas these data suggest that the FS and EC domains are not required for BMP inhibition overexpression of XSMOC-1Tg1, containing the Tg1-like domains only, did not inhibit BMP signaling in this assay (Figure 4D); deletion of the Tg1-like domains (XSMOC-1[increment]Tg1) produced a similar result. While we believe the Tg1-like domains to be important for BMP inhibition, by process of elimination, the effects of large deletions on protein folding are difficult to predict; in experiments of this type, where protein function is lost, improper folding is common (Valastyan and Lindquist, 2014). Consequently, we consider misfolding of the protein produced by XSMOC-1[increment]FS[increment]EC as the most likely reason for the inability of this construct to block BMP signaling. Additional evidence for the importance of the Tg1-like domains in SMOC comes from studies of human Ophthalmo-Acromelic (Waardenburg Anophthalmia) Syndrome, an autosomal disorder caused by mutations in SMOC-1 (Rainger et al., 2011). The phenotype of anopthalmia, oligodactyly, and joint abnormalities was found to be the same in patients with nonsense or frameshift mutations and those with missense mutations in the second Tg1-like domain (Rainger et al., 2011). As the nonsense and frameshift mutations were predicted to result in a complete loss of SMOC-1 function, the two pedigrees harboring two different single amino acid missense mutations in the Tg1-like domain suggests this domain is indeed essential for SMOC-1 function. Alternative approaches will be required to elucidate the exact role of the Tg1-like domains in BMP inhibition. Many proteins contain Tg1-like domains, including thyroglobulin, insulin-like growth factor binding proteins (IGFBPs) 1–6, the proteoglycan testican, and the basement membrane associated protein nidogen/entactin (Novinec et al., 2006). However, there have been no reports of any of these proteins inhibiting BMP signaling.

The SMOC EC domain can expand the range of BMP signaling in vitro by competitive binding to HSPGs

The potentiation of BMP signaling by the EC domain was examined further by investigating the relative affinities of XSMOC-1EC and BMP2 for each other and for HSPGs. The binding of SMOC/Pent and BMP2/4 to HSPGs is known, as evidenced by the co-purification of SMOC and BMPs following heparin affinity chromatography of bovine cartilage extracts (Chang et al., 1994), and the binding of BMP2/4 and the EC domain of hSMOC/Pent to heparin/HSPGs (Vuilleumier et al., 2010; Klemenčič et al., 2013; Ruppert et al., 1996). In addition, the basic amino acid-rich putative heparin-binding region identified within the EC domain of SMOC (Klemenčič et al., 2013) is highly conserved in Pent (Figure 5—figure supplement 1). Using the Protein Homology/analogY Recognition Engine (PHYRE), an unsupervised homology model for XSMOC-1EC was constructed based on the structure of the EC domain of the related family member BM-40 (Hohenester et al., 1996). XSMOC-1EC aligned well with the BM-40-EC model (Figure 5—figure supplement 1) and the electrostatic surface potential map predicted an area of positive charge similar to that reported in the EC domain of hSMOC1 (Klemenčič et al., 2013) (Figure 5—figure supplement 1). As monomeric hSMOC-1, refolded in the absence of calcium, was used in previous heparin-binding studies (Klemenčič et al., 2013), we first determined whether dimeric XSMOC-1 can bind heparin. XSMOC-1 and XSMOC-1EC bound to heparin Sepharose in the presence of 0.5M NaCl, whereas XSMOC-1[increment]EC did not (Figure 5A), confirming the EC domain to be the site of HSPG binding. Comparison of the heparin-binding affinities of XSMOC-1EC and BMP2 showed a striking similarity, with both eluting between 0.65M and 0.7M NaCl (Figure 5B). The possibility that XSMOC-1 and BMP2 bind to each other was discounted (Figure 5C); when BMP2 was incubated with XSMOC-1 or XSMOC-1 EC, pre-bound to heparin-Sepharose, BMP2 was only present in the unbound fraction (Figure 5C; lanes 2 and 5) and did not co-elute with XSMOC-1 or XSMOC-1 EC (Figure 5C; lanes 3 and 6). The lack of interaction of SMOC with BMP2 agrees with an analogous finding observed for Pent and the Drosophila BMP, decapentaplegic (Dpp) (Vuilleumier et al., 2010). This, combined with SMOC and BMP2 having similar heparin-binding affinities, suggests that XSMOC-1EC could compete with BMPs for HSPG binding on HEK293 cells and thereby increase BMP bioavailability.

Figure 5.
XSMOC-1EC and BMP2 have similar binding affinities for heparin sepharose (HS), but do not bind to each other.

We designed an in vitro assay to test the hypothesis that SMOC can expand the range of BMP signaling by competing with BMP2/4 for HSPG-binding. BMP4-soaked beads represented a cellular source of BMPs and agarose gels (0.7%) containing heparan sulfate (HS) (10 μg/ml) represented an extracellular matrix (ECM) capable of binding SMOC and BMPs. Chamber slides containing BMP4-soaked beads embedded in the agarose/HS/XSMOC-1EC matrices were seeded with the stable reporter cell line C33A-2D2-09, harboring luciferase under the control of a BMP response element (BRE). After 24 hr, immunohistochemical analysis of cells in fields of view adjacent to the matrices showed many luciferase positive cells (59%) adjacent to the agarose-only gel (Figure 6A,D), whereas when HS was present only a few (8%) were detected (Figure 6B,E). In contrast, when the matrix contained both HS and XSMOC-1EC, the number of BMP-responsive cells (64%) was similar to that observed in the absence of HS (Figure 6C,F). The result showed that only a relatively small amount of BMP4 diffused through the HS-containing agarose gel, activating luciferase expression in only a few BMP-responsive cells. When XSMOC-1EC was present, sufficient BMP4 diffused through the HS-containing matrix to activate a greater number of cells. The assay demonstrated that binding of BMPs to HSPGs restricts their range of effect and that BMP diffusion can be enhanced by the binding of SMOC to HSPGs, effectively expanding the range of effect.

Figure 6.
Immunofluorescence assay demonstrating that XSMOC-1 can promote BMP signaling at a distance from its source by competitive binding to HSPGs.

SMOC can potentiate BMP signaling at a distance from its source in vivo

We designed an assay to assess the ability of SMOC to affect the range of BMP signaling in vivo using Xenopus ectodermal conjugates, from stage 9 to 9.5 Xenopus embryos, in which endogenous BMP signaling is absent (Faure et al., 2000). Animal caps, removed at stage nine from embryos co-injected with mCherry and/or SMOC mRNAs, were grafted onto the animal poles of BMP2 mRNA-injected embryos from which the caps had been removed (Figure 7A). The conjugates were allowed to heal for two hours before the ectoderm from the entire animal half of the chimeric embryos was removed for immunostaining for phospho-Smad 1/5/8 (Figure 7B–F). In non-injected controls, no pSmad signal was detected either in the host tissue or the mCherry mRNA-injected donor tissue (Figure 7B). When host embryos were injected with a high concentration of BMP2 mRNA (300 pg), pSmad signaling was observed throughout the host and donor graft tissue (Figure 7C). Following a series of titration experiments we found that injection of 30 pg BMP2 mRNA at the two cell stage (15 pg into each blastomere) was sufficient to induce BMP signaling in the host tissue, while pSmad was observed only at a few cell diameters into the donor graft (Figure 7D). This amount of BMP2 mRNA was used in subsequent experiments to test whether SMOC can extend the range of BMP signaling. Next, donor animal caps from embryos injected with different amounts of SMOC mRNA were grafted onto the animal poles of embryos injected with 30 pg BMP2 mRNA. Donor caps from embryos injected with a low amount of SMOC mRNA (10 pg) had an increased number of pSmad positive nuclei compared to non-SMOC injected caps (Figure 7E). Conversely, donor caps from embryos injected with a high amount of SMOC mRNA (300 pg) were devoid of pSmad (Figure 7F). The results demonstrate that, in an intact tissue, low concentrations of SMOC can promote diffusion of BMP from its source of synthesis, thereby extending its range of effect, whereas high SMOC concentrations inhibit BMP signaling. These data would be consistent with low levels of SMOC binding to HSPGs, but not inhibiting BMP signaling, and high amounts of SMOC both binding HSPGs and inhibiting BMP signaling. Closer examination of the host tissue surrounding the donor grafts expressing the highest level of SMOC showed BMP signaling was both absent in the graft and decreased in the BMP2-expressing ectoderm immediately adjacent to the SMOC expressing graft (Figure 7F). This would be consistent with SMOC inhibiting BMP signaling within the graft and also diffusing into the immediate adjacent tissue at levels sufficient to inhibit BMP signaling.

Figure 7.
In vivo assay demonstrating that XSMOC-1 can expand the range of BMP2 signaling.

Model to explain the dual function of pent and SMOC as BMP antagonists and expanders of BMP signaling

In addition to the data we present here, there are substantial in vivo data from previous work that are consistent with SMOC/Pent acting both as a BMP antagonist and by expanding the range of BMP signaling. During Drosophila wing development, while dpp expression is restricted to a stripe of medial cells in the wing disc, the morphogenetic gradient expands across the anterior/posterior (A/P) axis (Nellen et al., 1996). Long-range Dpp signaling in the wing has been shown to involve both pent and the cell membrane-anchored HSPG, dally (Vuilleumier et al., 2010; Fujise et al., 2003); the absence of either causes severe contraction of the range of Dpp signaling, and wing patterning defects (Vuilleumier et al., 2010). However, the mechanism by which Pent and Dally cooperate to expand the range of Dpp signaling is not clear. It has been speculated that the Pent/Dally interaction may reduce the affinity of Dpp for its receptor, Thickveins, leading to an increase in Dpp diffusion (Ben-Zvi et al., 2011). However, the results we obtained from cell culture studies suggest that this is not the case. When XSMOC-1EC is added to HEK293 cells at 50 μg/ml, in serum-free media it will bind to the many cell surface HSPG binding sites. Consequently, following the subsequent addition of BMP2, any impairment of BMP2/receptor affinity caused by SMOC/HSPG binding would result in a reduction of BMP signaling. Instead, a potentiation of BMP signaling was observed (Figure 4A,B). Based on the expression patterns of dpp, pent, and dally, their known biological activities, and the new information presented here, we propose the following model (Figure 8A–C).

Figure 8.
Schematic diagrams showing the proposed mode of action of Pent in the Drosophila wing disc and SMOC in vertebrate joint interzones.

Dally, expressed both medially and laterally in the wing disc (Fujise et al., 2003), encodes a cell autonomous membrane-associated protein; pent, expressed laterally, and dpp, expressed medially, encode secreted proteins able to diffuse away from their source (Thomas et al., 2009; Vuilleumier et al., 2010). Indeed, we showed previously that SMOC can diffuse across many cell diameters in Xenopus animal caps (Thomas et al., 2009). At the lateral border, Dpp signaling will be inhibited by a combination of the expression of pent (Figure 7A) and the BMP repressor, brinker (Minami et al., 1999). As Pent diffuses across the wing disc it will create a lateral-medial gradient and compete with Dpp for Dally binding. Dpp, diffusing in an opposing medial-lateral gradient, will encounter gradually increasing levels of Pent. At low levels, Pent will occupy some Dally binding sites, promoting further Dpp diffusion; at high levels, Pent will occupy many Dally binding sites and also inhibit Dpp signaling downstream of its receptor (Figure 8A). This model is supported by the observations made either in the absence of pent (Vuilleumier et al., 2010), or the absence of dally (Fujise et al., 2003), where there is a contraction of the Dpp signaling gradient. In the absence of pent, medial-lateral diffusion of Dpp will be restricted by Dally because there is no Pent to compete for Dally binding sites (Figure 8B). The resulting high level of medially localized Dpp signaling, indicated by Mad phosphorylation (pMad) (Vuilleumier et al., 2010), has been shown to induce expression of the inhibitory Smad, daughters against dpp (dad), promoting a negative feedback loop to limit the range of Dpp signaling (Tsuneizumi et al., 1997; Ogiso et al., 2011). In the absence of dally, Dpp diffusion will also be restricted as Dally-binding is required to establish both Dpp and Pent gradients (Vuilleumier et al., 2010). Medially, dally mutant discs show abnormally high levels of pMad (Fujise et al., 2003), which would promote a negative feedback loop in Dpp signaling similar to that observed in the absence of pent (Figure 8C). Our proposed mode of action of Pent in the Drosophila wing disc is consistent with the role of glypicans in the ‘restricted extracellular diffusion’ model for establishing a Dpp signaling gradient (Schwank et al., 2011); the inclusion of the Pent/glypican interaction provides for further diffusion of Dpp to expand the morphogenetic field. During the review of this manuscript a study was published (Norman et al., 2016) showing that binding of Pent to Dally in Drosophila induces endocytosis of the complex, effectively reducing the amount of Dally on cell surfaces. Although it remains to be determined whether binding of SMOC to HSPGs also induces endocytosis, this finding is consistent with the model we propose: both competitive binding of SMOC/Pent to HSPGs and removal of HSPG binding sites for BMP/Dpp by endocytosis would promote BMP/Dpp diffusion. This idea is further supported by the observation that removal of the heparin binding site within BMP4 has also been shown to promote the BMP diffusion in Xenopus embryos (Ohkawara et al., 2002).

The binding of Pent to Dally also modulates Wingless (Wg) signaling in Drosophila, where a reduction in Wg signaling is observed in the presence of high levels of Pent (Norman et al., 2016). The interaction of Wg with Dally is known to potentiate Wg binding to its receptor Frizzled (Tsuda et al., 1999; Lin and Perrimon, 1999); therefore, the sequestration of Dally by Pent would be consistent with a reduction in Wg signaling. These data could also provide an explanation for our previous observation that SMOC activates MAPK signaling and inhibits BMP signaling via a mechanism requiring phosphorylation of Smad in the linker region (Thomas et al., 2009). Linker Smad phosphorylation at MAPK sites has been shown to be followed by phosphorylation at Glycogen Synthase Kinase 3 (GSK3) sites (Fuentealba et al., 2007), resulting in Smad ubiquitination and degradation (Sapkota et al., 2007). As the activity of GSK3 is inhibited by Wnt signaling, a reduction in Wnt signaling by Pent/SMOC could potentiate GSK3-mediated phosphorylation of linker Smad.

The potential function of SMOC during skeletal joint development

The spatial distributions and interactions of pent, dally, and dpp in the Drosophila wing disc can also be applied to explain the role of SMOC, HSPGs, and BMPs during development of vertebrate skeletal joints (Figure 8D–F). SMOC-1 (Okada et al., 2011), the cartilage-specific BMP called growth and differentiation factor 5 (GDF5) (Storm and Kingsley, 1996) or cartilage-derived morphogenetic protein-1 (CDMP-1) (Chang et al., 1994), BMP2/4 (Francis-West et al., 1999), and HSPGs (Koyama et al., 1995) are all expressed at the sites of future joint interzones; the absence of any one of these results in dysmorphogenesis of autopod joints (Okada et al., 2011; Rainger et al., 2011; Chang et al., 1994; Storm and Kingsley, 1996; Mundy et al., 2011; Thomas et al., 1996, 1997). Joint development in the digits occurs by segmentation of existing cartilage anlagen in regions where BMP signaling is inhibited by the BMP antagonist Chordin (Francis-West et al., 1999) and, based on our studies, by SMOC. Once formed, the joint interzone acts as a signaling center to promote appositional growth of the opposing cartilage epipheses (Archer et al., 2003; Ray et al., 2015). Based on information from existing in vivo mutational analyses and the data presented here, we propose the following model for the role of SMOC in promoting joint formation and subsequent appositional epiphyseal growth (Figure 8D,E). Within the interzone, Chordin and SMOC will both bind to HSPGs (Klemenčič et al., 2013; Jasuja et al., 2004) such as syndecan-3 (Koyama et al., 1995), allowing diffusion of GDF5 and BMP2/4 into the epiphyses, where BMP signaling promotes chondrocyte proliferation and appositional epiphyseal growth (Francis-West et al., 1999; Ray et al., 2015) via the BMP receptor BMPR1B (Baur et al., 2000). This hypothesis is consistent with the skeletal and joint abnormalities observed in the absence of HS following conditional knockdown of the glycosyltransferase, Ext1, in joint interzones (Mundy et al., 2011). Ext1 is essential for HS synthesis and binding of Chordin to HSPGs and is necessary for its BMP inhibitory activity (Jasuja et al., 2004). Consequently, in the absence of HS within the interzone, the BMP antagonist activity of Chordin will be reduced. While SMOC will still contribute to the inhibition of BMP signaling, the imbalance caused by the reduced Chordin activity is consistent with the increase in BMP signaling (pSmad) observed within the interzone in the absence of HS (Mundy et al., 2011); as a result, in response to BMP signaling, the interzone cells undergo chondrogenesis and the joint does not form (synostosis). In the absence of SMOC, we predict that the increased availability of HSPG binding sites will restrict diffusion of GDF/BMPs out of the interzone. In addition, the imbalance in BMP inhibition within the interzone will lead to an increase in BMP signaling resulting in chondrogenesis and failure of joint formation. This hypothesis is supported by oligodactyly and joint synostosis observed in patients with Ophthalmo-Acromelic (Waardenburg Anophthalmia) Syndrome (Okada et al., 2011; Rainger et al., 2011; Abouzeid et al., 2011).

In conclusion, based on the data we present here together with existing literature, we propose a model to explain how vertebrate SMOC (Thomas et al., 2009) and Drosophila Pent (Vuilleumier et al., 2010) regulate BMP signaling. SMOC/Pent can inhibit BMP signaling locally, downstream of the BMP receptor, probably via the region containing the two Tg1-like domains. In addition, binding of SMOC/Pent to cell surface HSPGs via the C-terminal EC domain prevents BMP binding, promoting BMP diffusion, thereby expanding its morphogenetic field.

Materials and methods

Drosophila methods

UAS-pentagone (UAS pent) was constructed by PCR amplification (forward primer: ATCTCGAGCCGAAGCACAGTAACAGTT; reverse primer: TATCTAGACGACGACATCTAATGAGTTG) from a cDNA clone corresponding to CG2264 (Drosophila Genomics Resource Center), and subcloned into pUAST using XhoI and XbaI. Sequence verification of the UAS-pent subclone followed by sequencing of the original cDNA indicated a base pair change was present in both, which resulted in a W to R amino acid change in a highly conserved region of the protein. Site directed mutatgenesis was performed with the QuikChange Lightning Site-Directed Mutagensis Kit (Stratagene) to correct pent cDNA to the database sequence. Drosophila transgenic lines were prepared by standard P-element transformation(Spradling, 1986) and mapped to a specific chromosome. Rescue was carried out by creating flies homozygous for the pentagone mutants: pent2-5/pentA17, and carrying the Gal four line brinker -Gal 4 (brk-Gal4) and the UAS-pent construct. Adult wings were dissected, fixed in 70% ethanol, mounted in Euperal (Asco Laboratories, Manchester, U.K.) and photographed on a Nikon E-800 microscope.

XSMOC-1 deletion constructs for mRNA injection

XSMOC-1[increment]FS ([increment]Q43-A91) was created from two separate PCR fragments obtained using pCS2-XSMOC-1 (Thomas et al., 2009) as template. The 5’ fragment was obtained using the pCS2 forward primer and the reverse primer 5’-GAGAGGATTGCACCCGGGGTCTCTGTCC-3’; the 3’ fragment was obtained using the forward primer 5’-AGGTGCAAAGATCCCGGGCAGAGCAAGTGT-3’ and the pCS2 reverse primer. A SmaI site (underlined) was incorporated to facilitate ligation of the two fragments. XSMOC-1[increment]EC ([increment]N310 to end) was amplified using pCS2-XSMOC-1 as template and the primer set 5’-GGCAACATGACCCCAAGA-3’, 5’-CTTTAAATAGGCCTTCTCAGTCCGTATTTTTCCA-3’ (incorporating a StuI restriction site, underlined, and stop codon, italics). XSMOC-1[increment]FS and XSMOC-1[increment]EC PCR products were cloned into PCR-TOPO (Invitrogen), sequenced, and then subcloned into pCS2. XSMOC-1EC ([increment]Q43 to W308), XSMOC-1Tg1 ([increment]Q43-A91 and [increment]N310 to end) and XSMOC-1[increment]Tg1 ([increment]K95 to S304) in pCS2 were obtained using a primer design method to generate large deletions (Liu and Naismith, 2008). For XSMOC-1EC and XSMOC-1[increment]Tg1, the template was pCS2-XSMOC-1 and the respective primer sets 5’-CTCAGTGTTTCGGCCAAAGAGCGACTG-3’, 5’-CAGTCGCTCTTTGGCCGAAACACTGAG-3’ and 5’- GCAAAGATGCTGGTCAGAGCGATGCCAGATGGAAAAATACG-3’, 5’-CGTATTTTTCCATCTGGCATCGCTCTGACCAGCATCTTTGC-3’. For XSMOC-1Tg1, the template was pCS2-XSMOC1[increment]FS and the primer set 5’-GATGCCAGATGGAAATAGGAGAGCCGGCCAGAAG-3’, CTTCTGGCCGGCTCTCCTATTTCCATCTGGCATC-3’. All pCS2 constructs were linearized with Not1 prior to mRNA transcription using the mMESSAGE mMACHINE SP6 kit (Life Technologies).

Cloning and bacterial expression of recombinant SMOCs

Full length XSMOC-1 without the predicted signal peptide (2-24) was amplified by PCR using the forward primer 5’-TGCCATGGCCAAAGAGCGACTGGC-3’ (containing a NcoI restriction site, underlined), the reverse primer 5’-CTCTCGAGCGCAAGGCGACTGAAGGGG T-3’ (containing a XhoI restriction site, underlined), and full length XSMOC-1 in pCS2 (Thomas et al., 2009) as the template. The PCR product was cloned into PCR-TOPO (Invitrogen), sequenced, and subcloned into the pET-28b(+) expression vector (Novagen) in frame with a C-terminal hexahistidine tag. An alternative start site located within XSMOC-1 at V235 (GTG) was removed by changing the codon to GTA by site directed mutagenesis (QuikChange II kit, Agilent Technologies) using the forward primer 5’-CCAAGAGAGGGAATTGTAATTCCAGAATGTGC-3’, reverse primer 5’-GCA CATTCTGGAATTACAATTCCCTCTCTTGG-3’, and XSMOC-1 in pET-28b(+) as template. Unlike XSMOC-1, the conserved Valine in hSMOC-1 is encoded by GTA in the human sequence, making the site-directed mutagenesis unnecessary. Using a primer design method to generate large deletions (Liu and Naismith, 2008), XSMOC-1[increment]EC ([increment]N310 to end) lacking the EC domain was obtained using the forward primer 5’- GATGCCAGATGG AAACACCACCACCACCACCACTG-3’, the reverse primer 5’-CAGTGGTGGTGGTGGTGGTGT TTCCATCTGGCATC, and XSMOC-1 in pET-28b(+) as template. Similarly, XSMOC-1EC containing the EC domain only ([increment]T2 to W308) was obtained using the same starting template, the forward primer 5’-GTGATC GGGACAGAGACAAAAATACGGACGCTGAAGACCC-3’, and the reverse primer 5’-GGGTCT TCAGCGTCCGTATTTTTGTCTCTGTCCCGATCAC-3’. Several E. coli strains were evaluated for XSMOC-1 expression and the ShuffleT7 Express strain C3029 (New England Biolabs), which allows the formation of disulfide bonds, was found to produce the highest yield of bioactive protein. For hSMOC-1 expression the host strain was BL21DE3, as described previously (Novinec et al., 2008). hSMOC-1 (pET28-SMOC-1-HT(Novinec et al., 2008)), kindly provided by B. Lenarcic, University of Ljubljana, Slovenia, was expressed in E. coli strain BL21DE3. Shaker cultures (3 L) were grown at 30°C in LB broth supplemented with 30 μg/ml kanamycin. When cell densities reached OD600 >0.5, recombinant protein expression was induced by the addition of IPTG (0.1 mM) for 5 hr. Cells were collected by centrifugation at 9000 g for 5 min.

Refolding and purification of recombinant SMOCs

Bacterial expression and refolding was based on that described previously (Novinec et al., 2008) with some modifications. Bacterial pellets were washed by suspension in 40 ml of 20 mM Tris-HCl/20% (w/v) sucrose/5 mM EDTA pH 7.5, centrifuging and resuspending in 40 ml of ice-cold deionized water. Following centrifugation the pellets were resuspended in phosphate-buffered saline (PBS)/5 mM EDTA and disrupted under high pressure (15,000 psi) using the EmulsiFlex-C3 high pressure homogenizer (Avestin, Inc, Ottawa, Canada). Lysates were centrifuged at 14,000g and inclusion bodies solubilized in 40 ml of solubilization buffer (50 mM sodium phosphate buffer, pH 8.0, 8M urea, 500 mM NaCl, 10 mM imidazole, 20 mM 2-mercaptoethanol) before applying to a pre-equilibriated 20 ml gravity-flow Ni-NTA agarose column (Qiagen). After washing with solubilization buffer bound protein was eluted with solubilization buffer/300 mM imidazole, concentrated to 5 ml (Vivaspin 20, GE Healthcare) and refolded by rapid dilution into 500 ml of refolding buffer (100 mM Tris/HCl, pH 9.0, 600 mM L-Arginine, 6 mM reduced L-Glutathione, 0.6 mM oxidized L-Glutathione, and 2 mM CaCl2). Following slow stirring overnight at 4°C the refolded protein solutions were concentrated to 20 ml and dialyzed against 20 mM Tris-HCl pH 7.5, 300 mM NaCl, 2 mM CaCl2. Non-soluble precipitate was removed and the dialysates concentrated to 2 ml before injecting onto a 1.6 × 60 cm HiLoad Superdex-200 PG gel filtration column (GE Healthcare) pre-equilibrated in S-200 buffer (20 mM Tris/HCl pH 7.5, 300 mM NaCl, 2 mM CaCl2, 10% glycerol). Fractions containing major peaks, collected using the ÄKTA purifier 10 system and UNICORN control software (GS Healthcare), were pooled, concentrated, and analyzed by SDS-PAGE, and, when necessary, sequenced. From three separate 3L cultures the total yields of dimeric XSMOC-1 were 1.3 mg, 2.5 mg, and 5.1 mg and for dimeric XSMOC-[increment]EC 3.4 mg, 4.1 mg, and 12.6 mg. The yields of monomeric XSMOC-1EC were 8.2 mg, 4.1 mg, and 3.25 mg.

Cell culture

NIH-3T3 Fibroblasts (ATCC CRL-1658) and HEK 293 (ATCC CRL-1573) cells were cultured in DMEM medium supplemented with 10% fetal bovine serum (FBS). Prior to the addition of recombinant protein cells were serum-starved in their respective media for one hour. The stable human cervical carcinoma clonal cell line C33A-2D2, containing a multimerized BMP-responsive element (BRE) for the BMP response gene inhibitor of differentiation-1 (Id1) (Korchynskyi and ten Dijke, 2002) linked to luciferase was a kind gift from Martine Roussel (Vrijens et al., 2013). C33A-2D2 cells were maintained in Eagle’s minimum essential medium (EMEM) supplemented with 10% FBS. The parental C33A cell line was purchased from ATCC (ATCC HTB-31). HEK-293 and parental C33A cells were authenticated by ATCC via STR profiling. While each cell line was confirmed to be free of mycoplasma contamination by ATCC, cells were not tested for mycoplasma in the experiments we describe. For BMP response assays, a sub-clone containing over 80% responsive cells (C33A-2D2-09) was prepared by single cell dilution. In these assays C33A-2D2-09 cells were maintained in serum-free Prime XV MSC expansion medium (Irvine Scientific) that does not contain BMPs. Wherever protein was added, XSMOC-1, XSMOC-1[increment]EC, and XSMOC-1EC were used at molar equivalent amounts based on their predicted molecular weights. BMP2 (Cell Signaling #4697) was added at 50 ng/ml.

Xenopus embryo manipulations

Frogs (Xenopus laevis), purchased from NASCO (Fort Atkinson, WI), were housed and maintained in aquaria approved by the FDA White Oak Campus Animal Care and Use Committee (ACUC). Prior to testes collection, male frogs were euthanized by anesthesia in a 2% solution of tricaine methane-sulphonate, a protocol approved by the ACUC. Frog embryos were manipulated using standard methods (Gurdon, 1967; Sive et al., 2000) and euthanized by anesthesia when the required developmental stage was reached (the study was approved by the ACUC). Injection of mRNAs was performed by standard procedures as described previously (Moos et al., 1995). Perturbations of axial patterning were quantified by Dorso-Anterior Index (DAI, (Kao and Elinson, 1988)) and dark field images of embryos were photographed with low angle oblique illumination and a Zeiss Stemi-6 dissecting microscope. For animal cap assays, animal caps were removed from stage nine embryos and cultured in 0.7 x Marc’s Modified Ringer’s (MMR) solution (Sive et al., 2000), 1 mg/ml BSA/50 µg/mL gentamicin until non-injected siblings reached stage 20. Where indicated, caps were incubated in 0.7 X MMR containing XSMOC-1, XSMOC-1[increment]EC, or XSMOC-1EC, at molar equivalent amounts, for different time periods.

RT-PCR

Pools of animal cap explants from at least two different fertilizations were prepared and analyzed for each condition reported. Total RNA isolation, reverse transcription (RT), and PCR amplification were performed as described previously (Thomas et al., 2009). PCR products were analyzed on 1.5% agarose gels in TAE buffer, stained with SYBR Green 1 (Molecular Probes), and scanned using a Fluorimager (Molecular Dynamics).

Immunoblotting

Cell lysates prepared by extraction in 6 M Urea, 25 mM Tris base, 2% SDS, 2% β-mercaptoethanol, and 5% glycerol were analyzed by SDS-PAGE (10 µg/lane) using Novex 10% Nu-PAGE gels (Invitrogen) and the MES buffer system. Immunoblot analyses were performed using the Novex XCell SureLock Mini-Cell system (Life Technologies) and nitrocellulose membranes (Invitrogen). Transferred proteins were detected using IRDye-labeled secondary antibodies and the Odyssey infrared imaging system (Li-COR Biosciences). The primary antibodies used were, phospho-Smad 1/5/8 (Cell Signaling Technology Cat# 9511, RRID:AB_331671) and Smad1 (Cell Signaling Technology Cat #9517, RRID:AB_10699149).Peptide antibodies specific to XSMOC-1[increment]EC (SDRDRDPQCNPHCTRPQHK) or XSMOC-1EC (GSFPPGKRPGSNPFSR) were produced in rabbits (Biomatik Corp.).

Heparin-binding studies

For heparin-binding studies, 5 µg of XSMOC, XSMOC[increment]EC, XSMOC-EC or human BMP2 (R and D Systems) in 50 μL of 1X PBS/0.5M NaCl was added to 20 μL of pre-equilibrated heparin Sepharose high performance beads (GE Life Sciences) and mixed with rotation for 15 min at room temperature. The beads were centrifuged (350 g for 2 min) and the supernatant removed. The protein-heparin bead mixture was then washed twice with 500 µL of 1X PBS or 1X PBS/NaCl (0.4 to 0.7M) before elution with 20 µL of 1 X Lithium Dodecyl Sulphate (LDS) sample buffer (Invitrogen)/0.1 M DTT for 5 min at 95°C. The supernatants were analyzed on a 10% NuPAGE gel and visualized by Coomassie staining.

BMP in vitro diffusion assay

Affi-Gel Blue beads (BioRad) approximately 100 µm in diameter were soaked in 100 µg/ml BMP4 (R and D Systems # 314 BP-010/CF) for 3 hr in a humidified chamber. Drops (0.5 µl) of 0.7% low melting point agarose (Life Technologies) with or without heparan sulfate (10 µg/ml; Sigma Aldrich #H7640) and either XSMOC-1EC (100 µg/ml) or S-200 buffer were placed on Millicell EZ chamber slides (Millipore). The slides were placed in a humidification chamber and incubated at room temperature for 5 min. Individual BMP4-soaked beads were placed into the partially gelled matrices and incubation was continued until gelling was complete. C33A-2D2-09 cells (2 × 104) in Prime XV MSC expansion medium (400 µl) were seeded into each well. The slides were incubated at 37°C in 5% CO2 for 48 hr prior to detection of luciferase. Briefly, following fixation (15 min) in 4% paraformaldehyde/PBS, cells were washed and permeabilized (0.5% Triton-X-100/PBS) for 10 min. Slides were washed (TBS/0.05% Tween) and incubated in blocking solution (Duolink, Olink Biosciences) for 1 hr in a humidified chamber prior to incubation with goat anti-luciferase (20 µg/ml) primary antibody (Promega #G745A, RRID:AB_2335880) for 1 hr. After washing (Duolink Wash Buffer A), slides were incubated in donkey anti-goat Alexa-488 (Thermo Fisher Scientific Cat# A-11055, RRID:AB_2534102) secondary antibody (5 µg/ml) for 45 min. Slides were then washed three times each in Duolink Wash Buffers A and B, mounted in Duolink mounting medium with DAPI, imaged by confocal microscopy (Zeiss LSM710), and analyzed with ImageJ (NIH) software.

BMP in vivo diffusion assay

Each blastomere of Xenopus embryos at the two cell stage was injected with mRNAs for BMP2 (15 pg or 150 pg), or SMOC (5, 15, or 150 pg) and/or mCherry (50 pg). pCS2 +8 NmCherry (Gökirmak et al., 2012) was a gift from Amro Hamdoun (Addgene plasmid # 34936). Animal caps were removed at stage nine and grafted into the animal poles of control or BMP2-injected embryos from which the animal poles had been removed. Following a two hour incubation the animal halves of the embryos were removed and fixed for 25 min in PBS/4% paraformaldehyde. The explants were permeabilized in 0.5% TritonX-100 for 10 min, washed in PBS, then transferred to blocking solution (PBS/5% goat serum) for 1 hr before incubating overnight at 4°C in a 1/1500 dilution of anti-pSMAD 1/5/8 antibody (Cell Signaling Technology Cat# 9511, RRID:AB_331671). The tissue was washed in PBS and prepared for immunofluorescence by incubating in goat anti-rabbit Alexa 488 (Thermo Fisher Cat#A11008, RRID:AB_143165) secondary antibody (5 µg/ml) for 2 hr. The tissue was then washed three times in PBS, transferred to a glass slide, mounted in Duolink mounting medium with DAPI, and imaged by confocal microscopy (Zeiss LSM710).

Acknowledgements

We thank Giorgos Pyrowolakis for Drosophila stocks, Martine Roussel for the stable human cervical carcinoma clonal cell line C33A-2D2, and Brigita Lenaričič for the hSMOC-1 expression vector, pET28-SMOC-1-HT. The Drosophila Genomics Resource Center, supported by NIH grant 2P40OD010949-10A1, provided the pentagone cDNA clone. The constitutively active BMP receptor construct (caBMPR1B) was the kind gift of Lee Niswander. We thank Howard Mostowski for his expert help with flow cytometry and are indebted to the staff of the FDA/CBER Division of Veterinary Services for their superb care of our Xenopus colony. We thank Ian Bellayr and Bharat Joshi for internal review of the manuscript.

Funding Statement

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Funding Information

This paper was supported by the following grants:

  • http://dx.doi.org/10.13039/100000016U.S. Department of Health and Human Services to J Terrig Thomas, Kristin R Andrykovich, Brian G Stultz, Deborah A Hursh, Malcolm Moos.
  • FDA Commissioner's Fellowship Program to D Eric Dollins, Tehyen Chu.

Additional information

Competing interests

The authors declare that no competing interests exist.

Author contributions

JTT, Conceptualization, Data curation, Formal analysis, Supervision, Investigation, Methodology, Writing—original draft, Writing—review and editing.

DED, Data curation, Formal analysis, Investigation, Methodology, Writing—review and editing.

KRA, Data curation, Writing—review and editing, investigation.

TC, Data curation, Formal analysis, Writing—review and editing.

BGS, Data curation, Formal analysis, Investigation, Methodology, Writing—review and editing.

DAH, Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—review and editing.

MM, Conceptualization, Formal analysis, Supervision, Methodology, Writing—original draft, Writing—review and editing.

Ethics

Animal experimentation: Frogs (Xenopus laevis) were housed and maintained in aquaria approved by the FDA White Oak Campus Animal Care and Use Committee (ACUC). Prior to testes collection, male frogs were euthanized by anesthesia in a 2% solution of tricaine methane-sulphonate, a protocol approved by the ACUC. Frog embryos were manipulated using standard methods and euthanized by anesthesia when the required developmental stage was reached (the study was approved by the ACUC). The ACUC approved protocol associated with this work is 1989-129.

References

  • Abouzeid H, Boisset G, Favez T, Youssef M, Marzouk I, Shakankiry N, Bayoumi N, Descombes P, Agosti C, Munier FL, Schorderet DF. Mutations in the SPARC-related modular calcium-binding protein 1 gene, SMOC1, cause waardenburg anophthalmia syndrome. The American Journal of Human Genetics. 2011;88:92–98. doi: 10.1016/j.ajhg.2010.12.002. [PubMed] [Cross Ref]
  • Alarcón C, Zaromytidou AI, Xi Q, Gao S, Yu J, Fujisawa S, Barlas A, Miller AN, Manova-Todorova K, Macias MJ, Sapkota G, Pan D, Massagué J. Nuclear CDKs drive smad transcriptional activation and turnover in BMP and TGF-beta pathways. Cell. 2009;139:757–769. doi: 10.1016/j.cell.2009.09.035. [PMC free article] [PubMed] [Cross Ref]
  • Archer CW, Dowthwaite GP, Francis-West P. Development of synovial joints. Birth Defects Research Part C: Embryo Today: Reviews. 2003;69:144–155. doi: 10.1002/bdrc.10015. [PubMed] [Cross Ref]
  • Baur ST, Mai JJ, Dymecki SM. Combinatorial signaling through BMP receptor IB and GDF5: shaping of the distal mouse limb and the genetics of distal limb diversity. Development. 2000;127:605–619. [PubMed]
  • Belenkaya TY, Han C, Yan D, Opoka RJ, Khodoun M, Liu H, Lin X. Drosophila dpp morphogen movement is independent of dynamin-mediated endocytosis but regulated by the glypican members of heparan sulfate proteoglycans. Cell. 2004;119:231–244. doi: 10.1016/j.cell.2004.09.031. [PubMed] [Cross Ref]
  • Ben-Zvi D, Pyrowolakis G, Barkai N, Shilo BZ. Expansion-repression mechanism for scaling the dpp activation gradient in Drosophila wing imaginal discs. Current Biology. 2011;21:1391–1396. doi: 10.1016/j.cub.2011.07.015. [PubMed] [Cross Ref]
  • Brazil DP, Church RH, Surae S, Godson C, Martin F. BMP signalling: agony and antagony in the family. Trends in Cell Biology. 2015;25:249–264. doi: 10.1016/j.tcb.2014.12.004. [PubMed] [Cross Ref]
  • Chang SC, Hoang B, Thomas JT, Vukicevic S, Luyten FP, Ryba NJ, Kozak CA, Reddi AH, Moos M. Cartilage-derived morphogenetic proteins. new members of the transforming growth factor-beta superfamily predominantly expressed in long bones during human embryonic development. The Journal of Biological Chemistry. 1994;269:28227–28234. [PubMed]
  • Faure S, Lee MA, Keller T, ten Dijke P, Whitman M. Endogenous patterns of TGFbeta superfamily signaling during early xenopus development. Development. 2000;127:2917–2931. [PubMed]
  • Francis-West PH, Parish J, Lee K, Archer CW. BMP/GDF-signalling interactions during synovial joint development. Cell and Tissue Research. 1999;296:111–119. doi: 10.1007/s004410051272. [PubMed] [Cross Ref]
  • Fuentealba LC, Eivers E, Ikeda A, Hurtado C, Kuroda H, Pera EM, De Robertis EM. Integrating patterning signals: wnt/GSK3 regulates the duration of the BMP/Smad1 signal. Cell. 2007;131:980–993. doi: 10.1016/j.cell.2007.09.027. [PMC free article] [PubMed] [Cross Ref]
  • Fujise M, Takeo S, Kamimura K, Matsuo T, Aigaki T, Izumi S, Nakato H. Dally regulates dpp morphogen gradient formation in the Drosophila wing. Development. 2003;130:1515–1522. doi: 10.1242/dev.00379. [PubMed] [Cross Ref]
  • Gammill LS, Sive H. otx2 expression in the ectoderm activates anterior neural determination and is required for xenopus cement gland formation. Developmental Biology. 2001;240:223–236. doi: 10.1006/dbio.2001.0470. [PubMed] [Cross Ref]
  • Gawantka V, Delius H, Hirschfeld K, Blumenstock C, Niehrs C. Antagonizing the Spemann organizer: role of the homeobox gene Xvent-1. The EMBO Journal. 1995;14:6268–6279. [PubMed]
  • Green YS, Vetter ML. EBF factors drive expression of multiple classes of target genes governing neuronal development. Neural Development. 2011;6:19 doi: 10.1186/1749-8104-6-19. [PMC free article] [PubMed] [Cross Ref]
  • Gurdon JB. In Methods in Developmental Biology. New York: Crowell; 1967.
  • Gökirmak T, Campanale JP, Shipp LE, Moy GW, Tao H, Hamdoun A. Localization and substrate selectivity of sea urchin multidrug (MDR) efflux transporters. Journal of Biological Chemistry. 2012;287:43876–43883. doi: 10.1074/jbc.M112.424879. [PMC free article] [PubMed] [Cross Ref]
  • Hohenester E, Maurer P, Hohenadl C, Timpl R, Jansonius JN, Engel J. Structure of a novel extracellular ca(2+)-binding module in BM-40. Nature Structural Biology. 1996;3:67–73. doi: 10.1038/nsb0196-67. [PubMed] [Cross Ref]
  • Jasuja R, Allen BL, Pappano WN, Rapraeger AC, Greenspan DS. Cell-surface heparan sulfate proteoglycans potentiate chordin antagonism of bone morphogenetic protein signaling and are necessary for cellular uptake of chordin. Journal of Biological Chemistry. 2004;279:51289–51297. doi: 10.1074/jbc.M408129200. [PubMed] [Cross Ref]
  • Kao KR, Elinson RP. The entire mesodermal mantle behaves as Spemann's organizer in dorsoanterior enhanced Xenopus laevis embryos. Developmental Biology. 1988;127:64–77. doi: 10.1016/0012-1606(88)90189-3. [PubMed] [Cross Ref]
  • Kishi M, Mizuseki K, Sasai N, Yamazaki H, Shiota K, Nakanishi S, Sasai Y. Requirement of Sox2-mediated signaling for differentiation of early xenopus neuroectoderm. Development. 2000;127:791–800. [PubMed]
  • Klemenčič M, Novinec M, Maier S, Hartmann U, Lenarčič B. The heparin-binding activity of secreted modular calcium-binding protein 1 (SMOC-1) modulates its cell adhesion properties. PLoS One. 2013;8:e56839 doi: 10.1371/journal.pone.0056839. [PMC free article] [PubMed] [Cross Ref]
  • Korchynskyi O, ten Dijke P. Identification and functional characterization of distinct critically important bone morphogenetic protein-specific response elements in the Id1 promoter. Journal of Biological Chemistry. 2002;277:4883–4891. doi: 10.1074/jbc.M111023200. [PubMed] [Cross Ref]
  • Koyama E, Leatherman JL, Shimazu A, Nah HD, Pacifici M. Syndecan-3, tenascin-C, and the development of cartilaginous skeletal elements and joints in chick limbs. Developmental Dynamics. 1995;203:152–162. doi: 10.1002/aja.1002030204. [PubMed] [Cross Ref]
  • Lin X, Perrimon N. Dally cooperates with Drosophila frizzled 2 to transduce wingless signalling. Nature. 1999;400:281–284. doi: 10.1038/22343. [PubMed] [Cross Ref]
  • Liu H, Naismith JH. An efficient one-step site-directed deletion, insertion, single and multiple-site plasmid mutagenesis protocol. BMC Biotechnology. 2008;8:91 doi: 10.1186/1472-6750-8-91. [PMC free article] [PubMed] [Cross Ref]
  • Massagué J, Seoane J, Wotton D. Smad transcription factors. Genes & Development. 2005;19:2783–2810. doi: 10.1101/gad.1350705. [PubMed] [Cross Ref]
  • Matsumoto Y, Matsumoto K, Irie F, Fukushi J, Stallcup WB, Yamaguchi Y. Conditional ablation of the heparan sulfate-synthesizing enzyme Ext1 leads to dysregulation of bone morphogenic protein signaling and severe skeletal defects. Journal of Biological Chemistry. 2010;285:19227–19234. doi: 10.1074/jbc.M110.105338. [PMC free article] [PubMed] [Cross Ref]
  • Milet C, Maczkowiak F, Roche DD, Monsoro-Burq AH. Pax3 and Zic1 drive induction and differentiation of multipotent, migratory, and functional neural crest in xenopus embryos. PNAS. 2013;110:5528–5533. doi: 10.1073/pnas.1219124110. [PubMed] [Cross Ref]
  • Minami M, Kinoshita N, Kamoshida Y, Tanimoto H, Tabata T. Brinker is a target of dpp in Drosophila that negatively regulates Dpp-dependent genes. Nature. 1999;398:242–246. doi: 10.1038/18451. [PubMed] [Cross Ref]
  • Moos M, Nguyen NY, Liu TY. Reproducible high yield sequencing of proteins electrophoretically separated and transferred to an inert support. The Journal of Biological Chemistry. 1988;263:6005–6008. [PubMed]
  • Moos M, Wang S, Krinks M. Anti-dorsalizing morphogenetic protein is a novel TGF-beta homolog expressed in the spemann organizer. Development. 1995;121:4293–4301. [PubMed]
  • Mundy C, Yasuda T, Kinumatsu T, Yamaguchi Y, Iwamoto M, Enomoto-Iwamoto M, Koyama E, Pacifici M. Synovial joint formation requires local Ext1 expression and heparan sulfate production in developing mouse embryo limbs and spine. Developmental Biology. 2011;351:70–81. doi: 10.1016/j.ydbio.2010.12.022. [PMC free article] [PubMed] [Cross Ref]
  • Nellen D, Burke R, Struhl G, Basler K. Direct and long-range action of a DPP morphogen gradient. Cell. 1996;85:357–368. doi: 10.1016/S0092-8674(00)81114-9. [PubMed] [Cross Ref]
  • Nieuwkoop PD, Faber J. Normal Table of Xenopus Laevis (Daudin) New York: Gatland Publishing, Inc; 1994.
  • Norman M, Vuilleumier R, Springhorn A, Gawlik J, Pyrowolakis G. Pentagone internalises glypicans to fine-tune multiple signalling pathways. eLife. 2016;5:e13301 doi: 10.7554/eLife.13301. [PMC free article] [PubMed] [Cross Ref]
  • Novinec M, Kordis D, Turk V, Lenarcic B. Diversity and evolution of the thyroglobulin type-1 domain superfamily. Molecular Biology and Evolution. 2006;23:744–755. doi: 10.1093/molbev/msj082. [PubMed] [Cross Ref]
  • Novinec M, Kovacic L, Skrlj N, Turk V, Lenarcic B. Recombinant human SMOCs produced by in vitro refolding: calcium-binding properties and interactions with serum proteins. Protein Expression and Purification. 2008;62:75–82. doi: 10.1016/j.pep.2008.07.009. [PubMed] [Cross Ref]
  • Ogiso Y, Tsuneizumi K, Masuda N, Sato M, Tabata T. Robustness of the dpp morphogen activity gradient depends on negative feedback regulation by the inhibitory smad, dad. Development, Growth & Differentiation. 2011;53:668–678. doi: 10.1111/j.1440-169X.2011.01274.x. [PubMed] [Cross Ref]
  • Ohkawara B, Iemura S, ten Dijke P, Ueno N. Action range of BMP is defined by its N-terminal basic amino acid core. Current Biology. 2002;12:205–209. doi: 10.1016/S0960-9822(01)00684-4. [PubMed] [Cross Ref]
  • Okada I, Hamanoue H, Terada K, Tohma T, Megarbane A, Chouery E, Abou-Ghoch J, Jalkh N, Cogulu O, Ozkinay F, Horie K, Takeda J, Furuichi T, Ikegawa S, Nishiyama K, Miyatake S, Nishimura A, Mizuguchi T, Niikawa N, Hirahara F, Kaname T, Yoshiura K, Tsurusaki Y, Doi H, Miyake N, Furukawa T, Matsumoto N, Saitsu H. SMOC1 is essential for ocular and limb development in humans and mice. The American Journal of Human Genetics. 2011;88:30–41. doi: 10.1016/j.ajhg.2010.11.012. [PubMed] [Cross Ref]
  • Rainger J, van Beusekom E, Ramsay JK, McKie L, Al-Gazali L, Pallotta R, Saponari A, Branney P, Fisher M, Morrison H, Bicknell L, Gautier P, Perry P, Sokhi K, Sexton D, Bardakjian TM, Schneider AS, Elcioglu N, Ozkinay F, Koenig R, Mégarbané A, Semerci CN, Khan A, Zafar S, Hennekam R, Sousa SB, Ramos L, Garavelli L, Furga AS, Wischmeijer A, Jackson IJ, Gillessen-Kaesbach G, Brunner HG, Wieczorek D, van Bokhoven H, Fitzpatrick DR. Loss of the BMP antagonist, SMOC-1, causes Ophthalmo-acromelic (Waardenburg anophthalmia) syndrome in humans and mice. PLoS Genetics. 2011;7:e1002114 doi: 10.1371/journal.pgen.1002114. [PMC free article] [PubMed] [Cross Ref]
  • Ray A, Singh PN, Sohaskey ML, Harland RM, Bandyopadhyay A. Precise spatial restriction of BMP signaling is essential for articular cartilage differentiation. Development. 2015;142:1169–1179. doi: 10.1242/dev.110940. [PubMed] [Cross Ref]
  • Rogers KW, Schier AF. Morphogen gradients: from generation to interpretation. Annual Review of Cell and Developmental Biology. 2011;27:377–407. doi: 10.1146/annurev-cellbio-092910-154148. [PubMed] [Cross Ref]
  • Ruppert R, Hoffmann E, Sebald W. Human bone morphogenetic protein 2 contains a heparin-binding site which modifies its biological activity. European Journal of Biochemistry. 1996;237:295–302. doi: 10.1111/j.1432-1033.1996.0295n.x. [PubMed] [Cross Ref]
  • Sapkota G, Alarcón C, Spagnoli FM, Brivanlou AH, Massagué J. Balancing BMP signaling through integrated inputs into the Smad1 linker. Molecular Cell. 2007;25:441–454. doi: 10.1016/j.molcel.2007.01.006. [PubMed] [Cross Ref]
  • Schwank G, Dalessi S, Yang SF, Yagi R, de Lachapelle AM, Affolter M, Bergmann S, Basler K. Formation of the long range dpp morphogen gradient. PLoS Biology. 2011;9:e1001111 doi: 10.1371/journal.pbio.1001111. [PMC free article] [PubMed] [Cross Ref]
  • Sive HL, Grainger RM, Harland RM. Early Development of Xenopus Laevis. Cold Spring Harbor: Cold Spring Harbor Laboratory Press; 2000.
  • Spradling AC. P element-mediated transformation. In: Roberts D. B, editor. Drosophila: A Practical Approach. Washington D.C: IRL Press; 1986. pp. 175–197.
  • Storm EE, Kingsley DM. Joint patterning defects caused by single and double mutations in members of the bone morphogenetic protein (BMP) family. Development. 1996;122:3969–3979. [PubMed]
  • Thomas JT, Lin K, Nandedkar M, Camargo M, Cervenka J, Luyten FP. A human chondrodysplasia due to a mutation in a TGF-beta superfamily member. Nature Genetics. 1996;12:315–317. doi: 10.1038/ng0396-315. [PubMed] [Cross Ref]
  • Thomas JT, Kilpatrick MW, Lin K, Erlacher L, Lembessis P, Costa T, Tsipouras P, Luyten FP. Disruption of human limb morphogenesis by a dominant negative mutation in CDMP1. Nature Genetics. 1997;17:58–64. doi: 10.1038/ng0997-58. [PubMed] [Cross Ref]
  • Thomas JT, Prakash D, Weih K, Moos M. CDMP1/GDF5 has specific processing requirements that restrict its action to joint surfaces. Journal of Biological Chemistry. 2006;281:26725–26733. doi: 10.1074/jbc.M603851200. [PubMed] [Cross Ref]
  • Thomas JT, Canelos P, Luyten FP, Moos M. Xenopus SMOC-1 inhibits bone morphogenetic protein signaling downstream of receptor binding and is essential for postgastrulation development in xenopus. Journal of Biological Chemistry. 2009;284:18994–19005. doi: 10.1074/jbc.M807759200. [PMC free article] [PubMed] [Cross Ref]
  • Tsuda M, Kamimura K, Nakato H, Archer M, Staatz W, Fox B, Humphrey M, Olson S, Futch T, Kaluza V, Siegfried E, Stam L, Selleck SB. The cell-surface proteoglycan dally regulates wingless signalling in Drosophila. Nature. 1999;400:276–280. doi: 10.1038/22336. [PubMed] [Cross Ref]
  • Tsuneizumi K, Nakayama T, Kamoshida Y, Kornberg TB, Christian JL, Tabata T. Daughters against dpp modulates dpp organizing activity in Drosophila wing development. Nature. 1997;389:627–631. doi: 10.1038/39362. [PubMed] [Cross Ref]
  • Ueno H, Nakajo N, Watanabe M, Isoda M, Sagata N. FoxM1-driven cell division is required for neuronal differentiation in early xenopus embryos. Development. 2008;135:2023–2030. doi: 10.1242/dev.019893. [PubMed] [Cross Ref]
  • Umulis D, O'Connor MB, Blair SS. The extracellular regulation of bone morphogenetic protein signaling. Development. 2009;136:3715–3728. doi: 10.1242/dev.031534. [PubMed] [Cross Ref]
  • Valastyan JS, Lindquist S. Mechanisms of protein-folding diseases at a glance. Disease Models & Mechanisms. 2014;7:9–14. doi: 10.1242/dmm.013474. [PMC free article] [PubMed] [Cross Ref]
  • Vannahme C, Smyth N, Miosge N, Gösling S, Frie C, Paulsson M, Maurer P, Hartmann U. Characterization of SMOC-1, a novel modular calcium-binding protein in basement membranes. Journal of Biological Chemistry. 2002;277:37977–37986. doi: 10.1074/jbc.M203830200. [PubMed] [Cross Ref]
  • Villegas A, Kropinski AM. An analysis of initiation Codon utilization in the domain Bacteria - concerns about the quality of bacterial genome annotation. Microbiology. 2008;154:2559–2661. doi: 10.1099/mic.0.2008/021360-0. [PubMed] [Cross Ref]
  • Vonica A, Brivanlou AH. An obligatory caravanserai stop on the silk road to neural induction: inhibition of BMP/GDF signaling. Seminars in Cell & Developmental Biology. 2006;17:117–132. doi: 10.1016/j.semcdb.2005.11.013. [PubMed] [Cross Ref]
  • Vrijens K, Lin W, Cui J, Farmer D, Low J, Pronier E, Zeng FY, Shelat AA, Guy K, Taylor MR, Chen T, Roussel MF. Identification of small molecule activators of BMP signaling. PLoS One. 2013;8:e59045 doi: 10.1371/journal.pone.0059045. [PMC free article] [PubMed] [Cross Ref]
  • Vuilleumier R, Springhorn A, Patterson L, Koidl S, Hammerschmidt M, Affolter M, Pyrowolakis G. Control of dpp morphogen signalling by a secreted feedback regulator. Nature Cell Biology. 2010;12:611–617. doi: 10.1038/ncb2064. [PubMed] [Cross Ref]
  • Wu MY, Hill CS. Tgf-beta superfamily signaling in embryonic development and homeostasis. Developmental Cell. 2009;16:329–343. doi: 10.1016/j.devcel.2009.02.012. [PubMed] [Cross Ref]
  • Zou H, Wieser R, Massagué J, Niswander L. Distinct roles of type I bone morphogenetic protein receptors in the formation and differentiation of cartilage. Genes & Development. 1997;11:2191–2203. doi: 10.1101/gad.11.17.2191. [PubMed] [Cross Ref]
2017; 6: e17935.

Decision letter

Hugo J Bellen, Reviewing editor
Hugo J Bellen, Baylor College of Medicine, United States;

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

Thank you for submitting your article "SMOC can act as both an antagonist and an expander of BMP signaling" for consideration by eLife. Your article has been reviewed by two peer reviewers, and the evaluation has been overseen by Janet Rossant as the Senior Editor and Hugo J Bellen as the Reviewing Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

The manuscript by Thomas et al. focuses on Xenopus Smoc-1- a BMP modulator with negative effects onto BMP signaling. Previously, the authors uncovered a role for Smoc in MAPK- dependent phosphorylation of the Smad linker, thus an indirect entry into the BMP signaling. However, data from vertebrates and Drosophila argue for a role for Smoc in modulating BMPs availability outside the cells. For example, the Drosophila Pent, appears to expand the range of BMP signaling and to promote signaling at a distance. In this manuscript, the authors test whether Smoc has dual activities onto the BMP signaling and address how different functional domains of Smoc inhibit BMP signaling at short-range while promoting long-range signaling. The authors take a biochemical approach and generate recombinant Smoc variants and refold biologically active proteins. Using these variants in multiple and creative experimental settings, they conclude that the EC domain of Smoc mediates its binding to glypicans and competes with BMPs for glypicans occupancy. From the data presented, it is less clear whether Smoc domain(s) could mediate glypican-independent inhibition of BMP signaling.

In order to understand the dual role of SMOC as an antagonist of BMP signalling and an expander of BMP spreading, the authors establish a protocol to express and refold recombinant wild-type (wt) and truncated versions of SMOC. Using recombinant wt and truncated SMOC, Thomas et.al. demonstrate that the C-terminal heparin sulphate(HS)-binding domain is dispensable for cell fate conversion and SMOC-mediated repression of BMP signalling. In contrast, the SMOC HS-binding domain is sufficient to promote BMP signalling and to compete with BMP2 for heparin sepharose binding. Furthermore, the authors show that competition for HSPG-binding between BMP2 and the HS-binding domain of SMOC can result in increasing range of BMP2 signalling in a cell culture assay. Based on the finding that SMOC potentially competes with BMPs for HSPG binding, the authors suggest a model for SMOC/Pent function.

The major novelty of this manuscript is the finding that SMOC can compete with BMP ligands for binding to heparin sulphates. Hence, these results suggest a novel mechanism by which SMOC/Pent would regulate the spreading of BMP ligands by competitive binding to extracellular HSPG.

The reviewers have some issues with the paper that need to be addressed, but which they believe can be addressed within the required 2-month time window.

The conclusions of this study are based on in vitro essays only, hence it does not provide evidence that SMOC indeed competes with BMPs under physiological conditions for binding with HSPG. Therefore providing in vivo data that supports the suggested mechanism would greatly enhance the conclusion of this manuscript. If providing such data is not possible, the authors should at least provide in vivo evidence that SMOC can indeed promote BMP signalling at a distance. This might be done by implanting SMOC-soaked beads close to a BMP source and monitor the effect on BMP signalling range.

While this is undoubtedly an important developmental question, this manuscript lacks several important controls and tends to over-interpret the structure-function analyses. Several major concerns need to be addressed before consideration for publication:

1) First, the authors argue that Smoc and delta-EC have equivalent effects, but from the data (without quantification) it seems that delta-EC has a reduced effect compared with full-length Smoc, for example in Figures 3A, ,4D.4D. This result needs to be clarified before the authors could uncouple the glypican-binding function of Smoc from other activities.

2) The data presented in Figure 4 are over-interpreted and do not demonstrate a role for the Tg domains in the inhibition of BMP signaling. The arguments regarding the role of EC in promoting BMP signaling are convincing, but the Tg experiment in 4D is incomplete and should present a full set of relevant Smoc and co-Pent truncations, including delta-FS and delta-EC together, as well as delta-Tg. The 4D panel is based on capped mRNA microinjection (and not purified and refolded proteins), thus a comprehensive set of truncations should be feasible and must be analyzed, quantified, and included here.

3) Figure 6 shows that EC, and not Smoc, can promote BMP signaling at a distance. Experiments with full-length Smoc are not shown or described.

This experiment is very creative and circumvents a tough technical problem in the field, the stickiness of BMP-type ligands. (Since BMP2 is very sticky, Smoc is unlikely to displace it from Heparin-Sepharose, so the converse experiment for Figure 5 would not work.) Setting up a cell-based system to examine whether Smoc displaces BMP from HSPSs is a clever alternative. But it is unfortunate that the authors did not present the entire series, including Smoc, and delta-EC.

Importantly, better and more uniform fields of cells must be chosen to illustrate this result, since more cells (as apparent in 6B) will deplete the BMP2 pool faster than the fewer cells shown in 6C, muddying the data.

4) How do these new findings correlate with the previous report from this group that Smoc modulates MAPK-dependent Smad phosphorylation? The authors need to include a section in the Discussion in which their current results should be reconciled and tied together with their previous findings.

The recent Drosophila Pent paper (Norman et al., eLife 2016), which shows that Pent also modulates the Wg/Wnt signaling via glypicans, together with previous results reported by Eddy De Robertis' group (Fuentealba et al., Cell 2007) may offer some explanations. For example, the interaction between Pent and Glypicans may play a key role in integrating patterning signals by controlling the interplay between MAPK, Wnt/GSK3 and BMP signaling, which seem to converge onto the phosphorylation state and duration of Smad-dependent signaling.

[…] The reviewers have some issues with the paper that need to be addressed, but which they believe can be addressed within the required 2-month time window.

The conclusions of this study are based on in vitro essays only, hence it does not provide evidence that SMOC indeed competes with BMPs under physiological conditions for binding with HSPG. Therefore providing in vivo data that supports the suggested mechanism would greatly enhance the conclusion of this manuscript. If providing such data is not possible, the authors should at least provide in vivo evidence that SMOC can indeed promote BMP signalling at a distance. This might be done by implanting SMOC-soaked beads close to a BMP source and monitor the effect on BMP signalling range.

We agree with the reviewer’s comment that data in vivo would greatly enhance our conclusions. We have designed and conducted experiments using Xenopus embryos to address this issue, the results of which are presented in a new Figure 7.

Xenopus embryos were injected at the two cell stage with mRNAs for BMP2, or SMOC and/or mCherry. Animal caps (ACs) from SMOC/mCherry-injected embryos (donor) were removed at stage 9 and conjugated to the animal poles of BMP2-injected embryos from which ACs had been removed (host). After two hours the entire animal hemispheres of the host/donor conjugates were fixed and immunostained for pSmad 1/5/8 to detect BMP signaling. mCherry was used to distinguish the SMOC-injected donor tissue from the BMP2-expressing host tissue. The use of ACs in this assay was supported by the observation that endogenous pSmad signaling is not present in stage 9 to 9.5 ACs (Faure et al., 2000, Development 127, 2917-2931). As background pSmad staining was absent in the control conjugates, cells positive for pSmad in the donor AC tissue would be indicative of diffusion of BMP from the surrounding host tissue.

A number of mRNA titration experiments were conducted to determine the optimal amounts of mRNA to inject and the time to incubate the conjugates prior to immunostaining for pSmad. It was found that injection of Xenopus embryos at the two cell stage with 30pg BMP2 mRNA was sufficient to obtain pSmad staining in the host tissue, but insufficient diffuse into the donor Wt caps after two hours. When higher amounts of BMP2 mRNA were used (300pg) pSmad was detected throughout the donor AC tissue, demonstrating that two hours was sufficient time for BMP to diffuse into the donor AC and induce pSmad. The ability of SMOC to extend the range of BMP signaling was tested by conjugating donor ACs from Xenopus embryos injected with different amounts SMOC mRNA to hosts injected with 30pg of BMP2. We found that much lower amounts of SMOC mRNA were required (10 to 30pg) than that used previously to inhibit BMP signaling (300pg). When high amounts of SMOC mRNA (300pg) were used, there was no pSmad staining in the donor AC or within the first few cell diameters of the adjacent host tissue, indicating BMP inhibition. The results of these experiments have been incorporated into the revised manuscript and a new figure (Figure 7) is provided.

We hope that this assay and the data obtained will be sufficient to address the reviewers’ request to show the ability of SMOC to promote BMP signaling at a distance from its source in vivo.

A question we have for the reviewers is whether including hatch marks to demarcate the host/donor tissue boundaries would be beneficial. We did not include these in the current figure as we did not want to obscure the data in any way.

While this is undoubtedly an important developmental question, this manuscript lacks several important controls and tends to over-interpret the structure-function analyses. Several major concerns need to be addressed before consideration for publication:

1) First, the authors argue that Smoc and delta-EC have equivalent effects, but from the data (without quantification) it seems that delta-EC has a reduced effect compared with full-length Smoc, for example in Figures 3A, ,4D.4D. This result needs to be clarified before the authors could uncouple the glypican-binding function of Smoc from other activities.

With regard to Figure 3A, while we do not state or intend to imply that SMOC and SMOC-[increment]EC are fully equivalent. However, we do show and state that both are able to induce the expression of neural markers in Xenopus animal caps (subsection “BMP Inhibition and Neural induction by SMOC does not require the EC domain”, first paragraph). Similarly, Figure 4D, amended to show the effect of each deletion construct, clearly shows that SMOC and SMOC[increment]EC can inhibit BMP signaling.

2) The data presented in Figure 4 are over-interpreted and do not demonstrate a role for the Tg domains in the inhibition of BMP signaling. The arguments regarding the role of EC in promoting BMP signaling are convincing, but the Tg experiment in 4D is incomplete and should present a full set of relevant Smoc and co-Pent truncations, including delta-FS and delta-EC together, as well as delta-Tg. The 4D panel is based on capped mRNA microinjection (and not purified and refolded proteins), thus a comprehensive set of truncations should be feasible and must be analyzed, quantified, and included here.

Two additional SMOC truncation constructs were prepared; SMOC-[increment]FS[increment]EC (referred to as SMOC-Tg1), containing the signal peptide and the two Tg1-like domains separated by the non-homologous domain, and SMOC-[increment]Tg1, containing the FS- like domain and EC domains (subsection “XSMOC-1 deletion constructs for mRNA injection”). We conducted mRNA overexpression studies in Xenopus embryos with SMOC, SMOC[increment]FS, SMOC[increment]EC, SMOC-Tg1 only (SMOC[increment]FS[increment]EC), and SMOC-[increment]Tg1 in the presence of the constitutively active BMP receptor (caBMPR1B) to test their ability to inhibit BMP signaling. Overexpression of SMOC, SMOC-[increment]FS, and SMOC-[increment]EC inhibited BMP signaling in the presence of the caBMPR; SMOC-Tg1, SMOC-[increment]Tg1 and SMOC-EC did not (subsection “BMP Inhibition and Neural induction by SMOC does not require the EC domain”, last paragraph). We have replaced the original Figure 4 panel D with the RT-PCR results obtained from these experiments. We speculate, by process of elimination, that while we believe the BMP inhibition activity of SMOC resides in the Tg1 domains, the protein translated from the SMOC-Tg1 only deletion construct may be inactive due to misfolding. We have deleted any categorical statements we had made previously regarding the involvement of the Tg1 domains and replaced them with more speculative language. In addition, we haveincluded data from human genetic studies that support the involvement of the SMOC Tg1 domains in SMOC function (see the aforementioned paragraph).

We did not prepare truncation constructs for co-Pent as, although the domain structuresof SMOC and Pent are similar, there are also significant differences that would make truncation comparisons difficult to interpret. Consequently, apart from showing that SMOC and co-Pent can both inhibit BMP signaling, we feel this request is beyond the scope of this manuscript.

3) Figure 6 shows that EC, and not Smoc, can promote BMP signaling at a distance. Experiments with full-length Smoc are not shown or described.

This experiment is very creative and circumvents a tough technical problem in the field, the stickiness of BMP-type ligands. (Since BMP2 is very sticky, Smoc is unlikely to displace it from Heparin-Sepharose, so the converse experiment for Figure 5 would not work.) Setting up a cell-based system to examine whether Smoc displaces BMP from HSPSs is a clever alternative. But it is unfortunate that the authors did not present the entire series, including Smoc, and delta-EC.

Importantly, better and more uniform fields of cells must be chosen to illustrate this result, since more cells (as apparent in 6B) will deplete the BMP2 pool faster than the fewer cells shown in 6C, muddying the data.

Having shown that the EC domain is required for binding to heparan sulfate and has the same affinity as full length SMOC (Figure 5), this in vitro assay was designed to show that HS-binding by the EC domain of SMOC promotes BMP diffusion through a HS-containing matrix. We did not assess the effect of full length SMOC or SMOC-[increment]EC in this assay as both could also potentially inhibit BMP signaling and mask the effect on BMP diffusion.

With regard to the comment on the fields of cells chosen to illustrate the results, we have quantified the percentage of luciferase-positive cells in each field obtained over four separate experiments and included this data in the figure legend. We consider the panels shown to be representative of the experimental results obtained for this assay and have stated this in the figure legend.

4) How do these new findings correlate with the previous report from this group that Smoc modulates MAPK-dependent Smad phosphorylation? The authors need to include a section in the Discussion in which their current results should be reconciled and tied together with their previous findings.

A discussion of how our previous work on BMP inhibition by SMOC involving linker Smad phosphorylation and our current results has been included (subsection “Model to explain the dual function of Pent and SMOC as BMP antagonists and expanders of BMP signaling”, last paragraph)

The recent Drosophila Pent paper (Norman et al., eLife 2016), which shows that Pent also modulates the Wg/Wnt signaling via glypicans, together with previous results reported by Eddy De Robertis' group (Fuentealba et al., Cell 2007) may offer some explanations. For example, the interaction between Pent and Glypicans may play a key role in integrating patterning signals by controlling the interplay between MAPK, Wnt/GSK3 and BMP signaling, which seem to converge onto the phosphorylation state and duration of Smad-dependent signaling.

A discussion of the findings presented in the Norman paper and how these relate to our data has been included (subsection “Model to explain the dual function of Pent and SMOC as BMP antagonists and expanders of BMP signaling”, second paragraph).


Articles from eLife are provided here courtesy of eLife Sciences Publications, Ltd