Bone morphogenetic protein 2 promotes human trophoblast cell invasion by upregulating N-cadherin via non-canonical SMAD2/3 signaling

Abstract

BMP2 expression is spatiotemporally correlated with embryo implantation and is crucial for endometrial decidualization and fertility in mice. BMP2 has been reported to increase the mesenchymal adhesion molecule N-cadherin and enhance cell invasion in cancer cells; moreover, studies suggest that N-cadherin promotes placental trophoblast invasion. However, whether BMP2 can promote trophoblast cell invasion during placentation remains unknown. The objective of our study was to investigate the effects of BMP2 on human trophoblast cell invasion and the involvement of N-cadherin and SMAD signaling. Primary and immortalized (HTR8/SVneo) cultures of human extravillous trophoblast (EVT) cells were used as study models. Treatment with recombinant human BMP2 increased HTR8/SVneo cell transwell Matrigel invasion as well as N-cadherin mRNA and protein levels, but had no significant effect on cell proliferation. Likewise, BMP2 treatment enhanced primary human EVT cell invasion and N-cadherin production. Basal and BMP2-induced invasion were attenuated by small interfering RNA-mediated downregulation of N-cadherin in both HTR8/SVneo and primary EVT cells. Intriguingly, BMP2 induced the phosphorylation/activation of both canonical SMAD1/5/8 and non-canonical SMAD2/3 signaling in HTR8/SVneo and primary EVT cells. Knockdown of SMAD2/3 or common SMAD4 totally abolished the effects of BMP2 on N-cadherin upregulation in HTR8/SVneo cells. Upregulation of SMAD2/3 phosphorylation and N-cadherin were totally abolished by type I receptor activin receptor-like kinases 2/3 (ALK2/3) inhibitor DMH1; moreover, knockdown of ALK2 or ALK3 inhibited N-cadherin upregulation. Interestingly, activation of SMAD2/3 and upregulation of N-cadherin were partially attenuated by ALK4/5/7 inhibitor SB431542 or knockdown of ALK4, but not ALK5. Our results show that BMP2 promotes trophoblast cell invasion by upregulating N-cadherin via non-canonical ALK2/3/4-SMAD2/3-SMAD4 signaling.

Fig. 1

BMP2 increases HTR8/SVneo and primary human EVT cell invasion. a, b HTR8/SVneo cells were treated with vehicle (Ctrl) or BMP2 (25 or 50 ng/mL) and cell invasiveness was examined by Matrigel-coated transwell assay. Representative images of the invasion assay ((a); scale bar 100 μm) and summarized quantitative results (b) are shown separately (n = 3). c HTR8/SVneo cells were seeded in 24-well plates and treated every 24 h with vehicle or BMP2 (25 or 50 ng/mL) for a total of 72 h. Cell viability was determined by MTT assay at 24, 48, and 72 h after BMP2 treatment (n = 3). d, e Primary human EVT cells were treated with or without 25 ng/mL BMP2 and cell invasiveness was examined by Matrigel-coated transwell assay. Representative images of the invasion assay ((d); scale bar 100 μm) and combined quantitative results (e) are shown separately (n = 5). Results are displayed as the mean ± SEM of at least three independent experiments and values without common letters are significantly different (P < 0.05)

Fig. 2

BMP2 increases N-cadherin mRNA and protein levels in HTR8/SVneo and primary human EVT cells. a HTR8/SVneo cells (left panel) or primary EVT cells (right panel) were treated with vehicle (Ctrl) or 25 ng/mL BMP2 for different lengths of time (3, 6, 12, or 24 h), and N-cadherin mRNA levels were examined by RT-qPCR with GAPDH as the reference gene. b HTR8/SVneo cells (left panel) or primary EVT cells (right panel) were treated with or without 25 ng/mL BMP2 every 24 h for 48 h, and N-cadherin protein levels were analyzed by western blot and normalized to α-tubulin. Results are displayed as the mean ± SEM of at least three independent experiments and values without common letters are significantly different (P < 0.05)

Fig. 3

Knockdown of N-cadherin attenuates basal and BMP2-induced trophoblast cell invasion. a–d HTR8/SVneo or primary human EVT cells were transfected for 48 h with 25 nM non-targeting control siRNA (si-Ctrl) or 25 nM siRNA targeting N-cadherin (si-N-cad) prior to treatment with vehicle (Ctrl) or 25 ng/mL BMP2. Western blot was used to measure N-cadherin protein levels in HTR8/SVneo (a) and primary EVT (c) cells 24 h after treatment with BMP2. b, d Matrigel-coated transwell assays were used to examine the effects of N-cadherin knockdown on BMP2-induced invasion in HTR8/SVneo (b; n = 3) and primary EVT (d) cells (n = 5). Summarized quantitative results are expressed as the mean ± SEM of at least three independent experiments and values without common letters are significantly different (P < 0.05)

Fig. 4

Non-canonical SMAD2/3 signaling mediates the upregulation of N-cadherin by BMP2. a, b HTR8/SVneo (a) and primary EVT (b) cells were treated with vehicle (Ctrl) or 25 ng/mL BMP2 for different lengths of time (10, 30, 60, or 120 min). Levels of phosphorylated SMAD1/5/8 (P-SMAD1/5/8), SMAD2 (P-SMAD2), and SMAD3 (P-SMAD3) were examined by western blot with corresponding phospho-specific antibodies. Membranes were stripped and reprobed with antibodies for total SMAD1/5/8 (T-SMAD1/5/8), SMAD2 (T-SMAD2), and SMAD3 (T-SMAD3). c HTR8/SVneo cells were transfected for 48 h with 25 nM non-targeting control siRNA (si-Ctrl), 25 nM siRNA targeting SMAD2 + SMAD3 (si-S2 + 3), or 25 nM siRNA targeting SMAD4 (si-S4) prior to treatment with vehicle (Ctrl) or 25 ng/mL BMP2 for 24 h. Protein levels of N-cadherin, T-SMAD2, T-SMAD3, and T-SMAD4 were examined by western blot (n = 4). Summarized quantitative results are displayed as the mean ± SEM of at least three independent experiments and values without common letters are significantly different (P < 0.05)

Fig. 5

BMP and activin/TGF-β type I receptors contribute to BMP2-induced phosphorylation of SMAD2/3 and upregulation of N-cadherin. a HTR8/SVneo cells were pretreated for 1 h with vehicle control (DMSO), 1 μM DMH1 (ALK2/3 inhibitor), or 10 μM SB431542 (ALK4/5/7 inhibitor) prior to treatment for 30 min with vehicle (Ctrl), 25 ng/mL BMP2, or 50 ng/mL activin A. Protein levels of phosphorylated SMAD1/5/8 (P-SMAD1/5/8), SMAD2 (P-SMAD2), and SMAD3 (P-SMAD3) were examined by western blot with corresponding phospho-specific antibodies. Membranes were stripped and reprobed with antibodies for total SMAD1/5/8 (T-SMAD1/5/8), SMAD2 (T-SMAD2), and SMAD3 (T-SMAD3). b HTR8/SVneo and primary EVT cells were pretreated for 1 h with DMH1 (1 μM) or SB431542 (10 μM) prior to treatment for 12 h with or without 25 ng/mL BMP2, and N-cadherin mRNA levels were measured by RT-qPCR. c HTR8/SVneo and primary EVT cells were pretreated for 1 h with DMH1 (1 μM) or SB431542 (10 μM) prior to treatment for 24 h with BMP2 (25 ng/mL) or activin A (50 ng/mL), and N-cadherin protein levels were analyzed by western blot. Results are expressed as the mean ± SEM of at least three independent experiments and values without common letters are significantly different (P < 0.05)

Fig. 6

BMP (ALK2/3) and activin (ALK4) type I receptors mediate the upregulation of N-cadherin by BMP2. a, b HTR8/SVneo cells were transfected for 48 h with 25 nM non-targeting control siRNA (si-Ctrl) or 25 nM siRNA targeting ALK2 (si-ALK2), ALK3 (si-ALK3), ALK4 (si-ALK4) or ALK5 (si-ALK5). Cells were treated for a further 12 h with vehicle (Ctrl) or 25 ng/mL BMP2 and RT-qPCR were used to measure the mRNA levels of N-cadherin (a) and ALKs (b). Summarized quantitative results are displayed as the mean ± SEM of three independent experiments and values without common letters are significantly different (P < 0.05). c Proposed model of the signaling pathway mediating BMP2-induced N-cadherin upregulation and increased human trophoblast cell invasion. BMP2 binds to a complex of type I and II receptors leading to the activation of both canonical SMAD1/5/8 and non-canonical SMAD2/3 signaling. Activation of receptor complexes containing ALK2, ALK3, and ALK4 leads to the phosphorylation of SMAD2/3, which complexes with common SMAD4 and translocates into the nucleus to increase the transcription of N-cadherin, which promotes human trophoblast cell invasion. BMP2 activates canonical SMAD1/5/8 signaling via ALK2/3; however, the functional consequences of this signaling require further investigation