Introduction
Retinoblastoma is a tumor of the retina and the most frequent intraocular cancer in children, accounting for 3% of all pediatric malignancies [
31]. It affects about 250–300 children per year in the United States and is responsible for about 3000–4000 deaths annually worldwide [
11,
17]. Early diagnosis and aggressive treatment strategies have made near-complete cure rates possible in developed countries, where the survival rate has reached almost 100% [
9]. However, current treatments can have negative impacts on vision. In addition, retinoblastoma still remains a potentially blinding, debilitating, and fatal tumor in developing countries, where early diagnosis, intensive chemotherapy and long-term follow-up are not as readily available, leading to a drop in the worldwide survival rate to 50% [
6,
7,
37]. With early detection, in most cases the primary tumor can be successfully treated by systemic or local chemotherapy [
9]. However, these therapies can be associated with sight-threatening complications and increased risk for secondary malignancies, particularly in the cases with germline
RB1 mutations [
28,
29]. Moreover, metastases to the central nervous system or in distant organs, including bone and bone marrow, are resistant to chemotherapy and thus represent a serious life-threatening complication. Therefore, novel therapeutic options are actively being pursued for advanced retinoblastoma, as well as for the primary tumor, in order to find new therapeutic targets to block metastatic spread and decrease the risks associated with systemic or local chemotherapy, utilized to treat the primary tumor.
In our previous study we demonstrated that genetic and pharmacological blockade of the Activin A receptor type 1C (ACVR1C), also known as Activin-like kinase receptor 7 (ALK7), strongly inhibited both primary growth and metastatic spread of retinoblastoma cells [
3]. Here we focused on the role of the ligands of ACVR1C receptor, which include Nodal, Activin and growth/differentiation factor 3 (GDF3), to determine the origin of this pro-metastatic signaling in retinoblastoma cells. The intrinsic serine/threonine kinase activity of the ACVR1C receptor is induced by interaction with these ligands, which results in phosphorylation of the SMAD2/3/4 complex, promoting its nuclear translocation and activation of gene transcription [
23]. ACVR1C ligands control many physiological processes, such as proliferation, differentiation, and wound healing. In particular, Nodal plays fundamental roles during embryonic development, where it is crucial for left-right axis specification of visceral organs [
14,
22] and for regulating germ cell versus somatic cell fate decisions in early mouse development [
30]. Nodal is also important for the maintenance of human embryonic stem cells [
13,
25] and has a pro-tumorigenic effect in several tumor types. Moreover, Nodal signaling is involved in retinal development, inducing the formation of retinal progenitor cells from mouse embryonic stem cells [
5]. Nodal also regulates differentiation of WERI Rb1 cells into retinal neurons [
21]. Here we focused on dissecting the role of ACVR1C ligands, as they represent an intriguing and promising point of therapeutic intervention to suppress the activity of the receptor and downstream signaling, which we have shown is crucial in promoting metastatic progression in retinoblastoma [
3]. We focused specifically on Nodal, as we found previously that it was more highly expressed in multiple retinoblastoma cell lines, as compared to the other ligands of the ACVR1C receptor [
3].
Discussion
We have recently demonstrated a crucial role for ACVR1C/ALK7, a type I receptor of the TGF-β family ligands, in promoting an aggressive phenotype in retinoblastoma [
3]. Here we focused on the role of an ACVR1C/ALK7 ligand, Nodal, to determine whether modulation of expression might impact downstream signaling as well as metastatic behavior of the retinoblastoma cells. We specifically focused on Nodal as we previously observed upregulated expression of this ligand at the mRNA and protein levels in multiple retinoblastoma lines, as compared to the other two ligands, Activin and GDF3, which were expressed at lower levels and only in a few of the cell lines tested [
3]. We also observed moderate to strong expression of Nodal in about 75% of the primary retinoblastoma samples that we have analyzed by IHC.
Nodal expression is largely restricted to embryonic tissues and is absent from adult non-neoplastic tissues [
32]. However, aberrant re-expression of Nodal has a prominent role in tumorigenesis and metastasis in melanoma, glioma, breast, prostate, and pancreatic cancers, with expression levels being directly proportional to tumor grade [
16,
18,
19,
32]. Interestingly, a recent report shows that suppression of Nodal significantly reduced growth, clonogenicity, migration and invasion in bladder cancer cells [
20]. Surprisingly, when we stimulated WERI Rb1 and Y79 cells with exogenous Nodal at 100, 300, 500 ng/mL, we did not observe any further increase in SMAD2 phosphorylation [
3]. Similarly, invasion, proliferation and cell growth were not significantly altered by treatment with exogenous Nodal in these cell lines. Thus, the elevated expression of Nodal present in WERI Rb1 and Y79 cells at steady state may already maximally activate downstream signaling as well as tumorigenesis. To test this hypothesis, we inhibited the endogenous expression of Nodal in WERI Rb1 and Y79 cells, using the shRNA technology, and found more than 90% reduction in the growth rate, a 30 to 36% reduction in proliferation, and from 50 to 80% reduction in transwell invasion. Importantly, analysis in an orthotopic model of retinoblastoma invasion in zebrafish confirmed the role of Nodal in promoting metastatic potential in retinoblastoma, as we observed a significant reduction in tumor spread when the expression of Nodal was suppressed in the injected tumor cells. To increase experimental rigor, we performed our analysis using five different shRNA constructs targeting Nodal mRNA. In both cell lines, all shRNAs were effective in suppressing Nodal expression by more than 90% at the protein level, as well as in abrogating SMAD2 phosphorylation/activation, and in dramatically reducing the protein levels of ZEB1 and Snail, two EMT transcription factors known to play an important role in promoting the invasive properties of the cancer cells [
26]. These data are consistent with our previous observation that reduction in invasion, upon pharmacological and genetic inhibition of the ACVR1C receptor, correlated with decrease in ZEB1 and Snail protein levels, in multiple retinoblastoma lines [
3]. Other studies have shown that Nodal controls migration and invasion in several tumor types by modulating the expression of Snail, Slug and ZEB1 [
8,
10,
12,
36]. We therefore postulate that ZEB1 and Snail might mediate, at least in part, the effects of Nodal signaling on invasion, as we observed that knock down of Nodal, besides reducing ZEB1 and Snail, significantly inhibited metastatic potential both in vitro and in vivo.
Interestingly, we found that Nodal plays an important role in maintaining retinoblastoma cell survival, as we observed a significant induction in the apoptotic markers, such as cleaved PARP1 and in cleaved caspase-3 (using immunoblotting and immunofluorescence assays, respectively), in retinoblastoma cells expressing Nodal shRNA compared to scrambled shRNA.
Potent morphogens, such as Nodal, require tight regulation of their expression and activity to properly control induction of signaling to only those regions where it is needed during embryogenesis. Endogenous antagonists of Nodal, such as the secreted proteins Cerberus and Lefty, play an important role in negatively regulating Nodal activity by preventing the binding of Nodal with its receptors and suppressing Nodal-mediated phenotypes [
4,
32]. Therefore, these developmental antagonists of Nodal could represent potential anti-Nodal therapeutics [
4,
15]. In addition, Nodal function-blocking antibodies have been shown to be promising tools to inhibit Nodal activity, both in vitro and in a murine model of cutaneous melanoma, where they produced a remarkable reduction in the ability of metastatic melanoma cells to colonize lungs in mice [
32]. More recently, a novel monoclonal antibody (3D1), raised against human Nodal, have been shown to inhibit clonogenicity, vasculogenic network formation and human melanoma xenograft growth in immunocompromised mice [
33]. Further investigation is therefore warranted to determine the efficacy of these approaches in ameliorating the effects of current therapies in retinoblastoma.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.