TMEM63B regulates nucleocytoplasmic transport and placental development

Abstract

The placental development requires coordinated trophoblast proliferation and differentiation, processes tightly coupled to cell cycle control. Here, we identify TMEM63B, an osmo/mechano-sensitive ion channel, as a key regulator of trophoblast cell cycle progression and placental development. TMEM63B modulates nucleocytoplasmic transport (NCT) by stabilizing NCT components, which govern the nuclear shuttling of key cell cycle regulators in response to osmo/mechanical cues. Loss of TMEM63B compromises Ran protein expression and Ran-XPO1 complex, impairing the nuclear export of CDKN1A/p21. This disruption leads to defective trophoblast proliferation, placental dysfunction, and ultimately perinatal lethality. Our findings establish TMEM63B as a pivotal osmo/mechano-sensitive molecule that regulates nucleocytoplasmic shuttling, providing new insights into how mechanical cues are integrated into nuclear mechanoresponses and placental development.

Skinner et al. dissects immune responses to seasonal influenza vaccination in a high-risk comorbid population of Australian First Nations and non-Indigenous people, highlighting significant baseline inflammation and atypical memory B cells.

Introduction

The placenta is essential for embryonic development, acting as the lifeline between the fetus and mother by facilitating nutrient exchange, hormone production, and immune protection¹. Proper placental development hinges on the precise regulation of trophoblast proliferation and differentiation, which are tightly governed by cell cycle progression^2–4. Recent research has demonstrated that mechanosensitive ion channels and curvature-sensing proteins are integral to cell cycle progression and proliferation by transducing mechanical stimuli such as tension⁵ and osmotic pressure^6–8 into biochemical signals^9–11. However, how these mechanosensitive entities manage critical cell cycle checkpoints for navigating through different cell cycle phases in response to mechanical stimuli, especially in the context of placental development, remains at the forefront of cell biological research. Notably, three independent large-scale genomic^12–14 screens previously identified TMEM63B as a candidate membrane protein potentially involved in mitotic mechanics, long before its formal characterization as an osmo/mechanosensitive ion channel. Coupled with data from the International Mouse Phenotyping Consortium identifying TMEM63B as a perinatal lethal gene¹⁵, these findings suggest a pivotal, yet unexplored role for this channel in development. Here, we identify TMEM63B, an osmo/mechano-sensitive ion channel^16–22, as a crucial regulator of trophoblast cell cycle progression and placental development. We demonstrate that TMEM63B fine-tunes nucleocytoplasmic transport (NCT), particularly affecting the stability of the Ran-XPO1 complex, which is a critical component of the nuclear-cytoplasmic transport essential for cell cycle regulation. By facilitating the NCT of CDKN1A/p21, a key cell cycle regulator, TMEM63B ensures appropriate cellular responses to osmo/mechano-stimuli. The loss of TMEM63B disrupts CDKN1A/p21 transport, impairs trophoblast proliferation, and leads to placental dysfunction and perinatal lethality. These findings unveil the fundamental role of TMEM63B in mechanotransduction in cell cycle control and placental development, offering potential therapeutic strategies for pregnancy complications.

Results

TMEM63B regulates trophoblast proliferation and placenta development

Systemic deletion of Tmem63b in mice results in severe perinatal lethality, as evidenced by significantly reduced survival at birth^15,23, highlighting its essential role in mammalian development. We examined pregnant mice from a Tmem63b-deficient line. Notably, the postnatal survival of homozygous Tmem63b^−/− (GKO) mice from the heterozygous breeder was dramatically reduced to 5%, far below the expected Mendelian inheritance ratio of 25% (Supplementary Fig. 1a). In addition, the weights of Tmem63b GKO placentas and embryos were significantly lower than those of wild-type (WT) controls (Supplementary Fig. 1b–e), suggesting the presence of intrauterine growth restriction (IUGR) in the Tmem63b GKO mice.

Abnormal placental development primarily underlies the pathophysiology of IUGR²⁴. In the placental labyrinth of WT mice, MCT1- and MCT4-labeled syncytiotrophoblast layers are consistently colocalized, with maternal blood sinuses and fetal blood vessels intermixed and evenly distributed. In contrast, Tmem63b GKO placentas exhibit several abnormalities: the MCT4-labeled syncytiotrophoblast layers are either absent or discontinuous (Supplementary Fig. 1f, g(i), as detailed in (ii)), maternal blood sinuses are abnormally enlarged, and there is an expansion of sinusoidal trophoblast giant cells (sTGCs) (Supplementary Fig. 1f, g(iii&iv)). Furthermore, the fetal blood vessels, labeled by CD31, are significantly reduced in the labyrinth layer of Tmem63b GKO placentas (Supplementary Fig. 1h–j). These observations collectively indicate disrupted placental development in the Tmem63b GKO mice. Importantly, we also observed a significant reduction in Ki67-positive cells in Tmem63b GKO placentas (Supplementary Fig. 1k–m), suggesting impaired cell proliferation, which is fundamental for placental development, function, and the establishment of a healthy pregnancy. This finding further underscores that compromised placental development in Tmem63b GKO mice likely underlies their perinatal lethality.

To determine whether placental defects drive perinatal lethality in Tmem63b GKO mice, we generated trophoblast-specific Tmem63b knockout (CKO) mice (Tmem63b^flox/flox with Tmem63b^flox/+; Elf5-Cre breeding strategy²⁵) to selectively delete Tmem63b in trophoblasts, the cells responsible for the placenta’s primary functions. Tmem63b CKO mice also exhibited markedly reduced survival (12%; Fig. 1a), and lower weights of Tmem63b CKO placentas and embryos than WT controls (Fig. 1b–e). The placental abnormalities observed in the Tmem63b CKO mice closely mirrored those seen in Tmem63b GKO mice: absent or discontinuous MCT4-labeled syncytiotrophoblast layers, abnormally enlarged maternal blood sinuses, overgrowth of sTGCs, and a significant reduction in fetal blood vessels within the labyrinth layer (Fig. 1f–j). These findings strongly suggest that defects in trophoblasts are responsible for the perinatal lethality observed in the Tmem63b knockout mice. Moreover, we observed a marked reduction in trophoblast proliferation in Tmem63b CKO placentas (Fig. 1k–m), indicating a critical deficiency in trophoblast growth and development.

Fig. 1. TMEM63B trophoblast-specific knockout mice exhibit perinatal lethality due to placental insufficiency.

a Tmem63b trophoblast-specific knockout (CKO) mice, generated using a Tmem63b^flox/flox and Tmem63b^flox/+; Elf5-Cre breeding strategy, exhibit significant deviations from expected Mendelian inheritance patterns around weaning age based on 200 offspring. The observed and expected proportions of postnatal mice born differed significantly (one-sided χ² test, ***P = 0.000388). b, c Representative E16 embryos and placentas from WT (b) and CKO (c) mice. Right panels: magnified placentas (fetal side up). d, e Embryo (d) and placenta (e) weights of WT and CKO mice at E16. Points represent genotype averages per litter; lines pair WT and CKO littermates. CKO weights differ significantly from WT (means ± SEM; n = 5 samples, two-tailed paired Student’s t-test, d ****P < 0.0001; e ****P < 0.0001). f, g Representative images of MCT1 (red) and MCT4 (green) immunostaining in TMEM63B WT (f) and CKO (g) placentas at E16 from three independent experiments, nuclei are stained with DAPI (blue). MCT1 is in SynT-1, facing maternal blood sinuses, and MCT4 is in SynT-2 around fetal blood vessels. (i) The cross-sections of placentas; (ii) enlarged views. (iii, iv) H&E and AP staining of WT and CKO placentas, with AP highlighting STGCs around maternal blood sinuses. h–j Representative CD31 immunostaining for fetal blood vessels in WT (h) and CKO (i) labyrinth zones, and corresponding quantification (j). Images are representative of three placentas per genotype. CD31 staining in placental labyrinths of WT and CKO placentas revealed a significant difference (means ± SEM, n = 3 placentas, two-tailed paired Student’s t-test, **P = 0.0055). k–m Representative Ki67 immunostaining in the labyrinth zones of E16 WT (k) and CKO (l) placentas, and corresponding quantification (m). Nuclei were counterstained with DAPI (blue). Images represent three placentas per genotype. Ki67 staining in placental labyrinths of WT and CKO placentas revealed a significant difference (means ± SEM; n = 3 placentas, two-tailed paired Student’s t-test, **P = 0.0013). Source data are provided as a Source Data file.

Taken together, these results demonstrate that Tmem63b is essential for normal placental development, particularly trophoblast proliferation. Its deletion leads to severe placental dysfunction, which in turn compromises embryonic survival and ultimately causes perinatal lethality.

TMEM63B facilitates the cell cycle progression to control trophoblast proliferation

To investigate the role of TMEM63B in trophoblast proliferation, we established knockout (KO) cell lines using CRISPR/Cas9 in trophoblast models, which included both BeWo cells and human trophoblast stem cells (hTSCs). Western blot analysis confirmed the absence of TMEM63B protein in these cells (Supplementary Fig. 2a). Patch-clamp recordings under an inside-out configuration revealed membrane stretch-induced macroscopic currents in WT BeWo cells, which were not observed in TMEM63B KO BeWo cells (Fig. 2a). Strikingly, TMEM63B ablation led to a significant impairment in proliferation in both BeWo cells and hTSCs (Fig. 2b and Supplementary Fig. 2b). Importantly, overexpression of full-length human TMEM63B in TMEM63B KO cells fully restored their proliferative capacity (Fig. 2b), establishing a direct functional link between TMEM63B and trophoblast cell proliferation. To further substantiate these observations in a physiologically relevant context, we employed hTSCs-derived trophoblast organoids (TOs). This three-dimensional model revealed that TMEM63B KO led to pronounced growth arrest (Fig. 2c). Together, these findings from multiple cellular models underscore the essential role of TMEM63B in trophoblast proliferation, providing compelling evidence for its critical function in placental development.

Fig. 2. TMEM63B regulates trophoblast proliferation by controlling the cell cycle.

a Stretch-activated currents induced by negative pipette pressure (−20 mmHg steps, −80 mV holding potential, inside-out patch) were abolished in TMEM63B KO BeWo cells. Insets show average pressure-response curves (means ± SEM; WT: n = 10, KO: n = 8 cells from three biological replicates; two-way ANOVA, ****P < 0.0001). b CCK8 assay shows TMEM63B KO significantly inhibits BeWo cell proliferation, which is restored by TMEM63B overexpression (means ± SEM; two-way ANOVA, WT vs KO: ****P < 0.0001; KO vs KO + TMEM63B: ****P < 0.0001). c Representative brightfield images showing morphological characteristics of WT and KO hTSCs-derived organoids at days 8 and 10 from three independent experiments. d–f Representative flow cytometry cell cycle profiles (using DAPI/Edu-Alexa Fluo594) of WT (d) and KO (e) BeWo cells and quantification (f) from three independent experiments. TMEM63B KO markedly reduces the S phase and increases the G2/M phase (mean ± SEM; two-tailed paired Student’s t-test, G0/G1: ns, P = 0.5341; Entering-S: *P = 0.0172; S: **P = 0.0029; Early G2: **P = 0.0062; G2/M: ***P = 0.0017). g The schematic illustrates the Thymidine-Nocodazole (Thy-Noc) protocol for cell synchronization, which blocks cells in early mitosis. h Flow cytometry showing cell cycle phase shifts in BeWo cells following Thy-Noc synchronization. i Western blot analysis of TMEM63B expression (normalized to GAPDH) in synchronized BeWo cells. j Relative quantitative analysis of TMEM63B expression throughout the cell cycle after synchronization (means ± SEM; n = 5 independent experiments; two-way ANOVA, **P = 0.0022). k Flow cytometry identified distinct cell cycle phase shifts in WT and TMEM63B KO cells following the Thy-Noc synchronization protocol. l, m Quantitative analysis of cell cycle phases in WT and TMEM63B KO BeWo cells post-synchronization. TMEM63B KO significantly induced cell arrest in the M phase. (means ± SEM; n = 3 biological replicates; two-way ANOVA, G0/G1: ****P < 0.0001; S: ****P < 0.0001; G2/M: ****P < 0.0001, proportional changes over time in different cell cycle phases when comparing KO to wild-type WT). All data were obtained from three independent experiments. Source data are provided as a Source Data file.

Cell proliferation involves sequential progression through the G1, S, G2, and Mitosis (M) phases of the cell cycle²⁶. To assess the impact of TMEM63B deletion on cell cycle progression, we performed flow cytometry using EdU and DAPI co-staining. This analysis revealed a significant shift in the cell cycle distribution of TMEM63B KO cells, with fewer cells in the S-phase and a corresponding increase in the G2/M populations (Fig. 2d–f and Supplementary Fig. 2c shows the corresponding gating strategy). To further explore the temporal regulation of TMEM63B during the cell cycle, we synchronized cells using thymidine-nocodazole treatment^27,28 and performed flow cytometry analysis with propidium iodide (PI) staining. This approach enabled us to track distinct cell cycle transitions, as synchronized populations progressed from mitosis (0 h) through the G1 phase (3 h) to the S phase (9 h) (Fig. 2g, h). Quantitative protein analysis revealed that TMEM63B expression peaked during mitosis and gradually declined through subsequent cell cycle phases (Fig. 2i, j and Supplementary Fig. 2d shows the corresponding gating strategy). Consistent with this cell cycle-dependent expression pattern, TMEM63B KO cells exhibited significant cell cycle defects. Specifically, these cells displayed pronounced mitotic accumulation and failed to efficiently exit mitosis and enter the G1 phase within 3–6 h post-release, in contrast to WT cells (Fig. 2k–m). As a result, TMEM63B KO cells showed significantly reduced entry into the S phase, as evidenced by EdU labeling assays.

To precisely define the point of cell cycle arrest in TMEM63B KO cells, we employed the Fucci4 fluorescent reporter system²⁹, which allows for distinct identification of cell cycle phases based on sequential expression of specific markers. Clover-Geminin1-110 marks the G1-S transition, mTurquoise2-SLBP18-126 indicates the G1 phase (with its loss signaling the S-G2 transition), and H1.0-Maroon1 appearance denotes the G2-M transition. The onset of a new G1 phase is marked by the disappearance of Clover-Geminin1-110, coupled with the reappearance of mKO2-Cdt130-120 and mTurquoise2-SLBP18-126. Using this system, we observed an accumulation of cells in the G2/M phases, with particular enrichment in mitosis, corroborating our flow cytometry results (Supplementary Fig. 3a–c).

Together, these findings establish TMEM63B as a critical regulator of cell cycle progression and highlight the importance of its function for proper cell proliferation.

TMEM63B modulates the nucleocytoplasmic transport and the stability of the Ran-XPO1 complex

To elucidate the role of TMEM63B in regulating trophoblast cell cycle progression, we performed RNA-seq analysis on both WT and TMEM63B KO BeWo cells (Supplementary Dataset 1). This analysis revealed a significant enrichment of downregulated gene sets associated with cell cycle regulation, identified through Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis (Fig. 3a). Complementing this genomic approach, we employed proximity biotinylation³⁰ using TMEM63B-TurboID in BeWo cells to identify proteins that functionally interact with or are proximal to TMEM63B. Notably, protein sets associated with cell cycle regulation were also significantly enriched, as determined by KEGG pathway analysis (Fig. 3b). Furthermore, our investigations highlighted the NCT pathway as the most significantly enriched pathway among both differentially expressed genes and differentially interacting proteins, as assessed by RNA sequencing and mass spectrometry, respectively (Fig. 3c). The NCT pathway is critical for maintaining cellular homeostasis and proper cell proliferation. It manages the controlled exchange of crucial molecules such as cyclins, cyclin-dependent kinases (CDKs), and their inhibitors between the nucleus and cytoplasm, which are essential for driving or restraining the cell cycle^31,32.

Fig. 3. TMEM63B modulates the nucleocytoplasmic transport and the stability of the Ran-XPO1 complex.

a Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment (FDR-ranked) of differentially regulated genes between WT and TMEM63B KO BeWo cells. b KEGG analysis of TMEM63B interactors from TurboID proximity labeling. Significance was determined via one-sided hypergeometric test with Benjamini–Hochberg adjustment (FDR < 0.05). c Venn diagram and ranking of shared KEGG pathways between RNA-seq and MS datasets. Ranked by combined log(p-value) scores, nucleocytoplasmic transport emerges as the top commonly enriched pathway. d Schematic of the light-inducible nuclear export (LEXY) system, its nucleocytoplasmic transport dynamics, and N/C ratio quantification. Created in BioRender. Cai, M. (2026) https://BioRender.com/eorn1h0. e, f Representative images (e) and N/C ratio (f) of LEXY-expressing WT and TMEM63B KO BeWo cells exposed to sequential dark (1 min), blue light (11 min), and dark recovery (12 min). (mean ± SEM; n = 8 slices from three independent experiments; two-way ANOVA, ****P < 0.0001). g Representative immunoprecipitation of endogenous Ran-TMEM63B interaction in BeWo cells from three independent experiments. IB Immunoblotting, IP Immunoprecipitation. The immunoprecipitation used an anti-Ran antibody. h Representative western blot analysis from three independent experiments of key nucleocytoplasmic transport proteins in WT and TMEM63B KO BeWo cells. i Representative co-immunoprecipitation from three independent experiments confirms the formation of the Ran-XPO1 complex (pulled down via anti-Ran), which is destabilized in the absence of TMEM63B. j, k Overexpression of Ran (128) restores nuclear transport in TMEM63B KO cells. Representative images (j) and quantification of the N/C ratio (k) of the LEXY reporter are shown for WT, TMEM63B KO, and TMEM63B KO cells with Ran (128) overexpression. N/C Ratio was monitored over time among these cells expressing LEXY (mean ± SEM; n = 8 slices from three independent experiments, two-way ANOVA, ****P < 0.0001). l Overexpression of Ran in TMEM63B KO cells restored the trophoblast proliferation. The proliferation rate difference between TMEM63B KO, WT, and TMEM63B KO cells overexpressing Ran (128) over time (mean ± SEM; two-way ANOVA, ****P < 0.0001). All data were obtained from three independent experiments. Source data are provided as a Source Data file.

Based on these findings, we hypothesized that TMEM63B is pivotal in normal cell cycle progression by modulating NCT. To test this hypothesis, we utilized the light-inducible nuclear export system (LEXY)³³, which features a weak nuclear localization signal (NLS) tagged with mCherry and a strong blue light-inducible nuclear export signal (NES) to simulate the process of NCT (Fig. 3d). At baseline (t = 0 s), both WT and TMEM63B KO cells exhibited a high nuclear-to-cytoplasmic ratio of mCherry. However, upon blue light stimulation, a robust nuclear export of mCherry was observed in WT cells, but only minimal changes occurred in TMEM63B KO cells (t = 750 s). Following the cessation of light stimulation, WT cells demonstrated efficient recovery of nuclear mCherry localization, whereas TMEM63B KO cells showed significantly impaired recovery efficiency (t = 1500 s) (Fig. 3e, f). These results establish TMEM63B as a critical regulator of dynamic NCT, further underscoring its essential role in cell cycle progression in trophoblast cells.

Our proximity proteomics revealed a significant interaction between TMEM63B and Ran (Supplementary Fig. 4a and Supplementary Dataset 2). Ran is a small GTPase that acts as the master regulator of NCT^32,34. Previous studies have demonstrated that a specific fraction of Ran can localize to the cell membrane³⁵, raising the possibility that membrane-associated Ran may provide a mechanistic link between TMEM63B and the intracellular NCT machinery. Unexpectedly, we observed that BeWo cells express two distinct Ran isoforms: isoform 1 (Ran (216): 216 amino acids) and isoform 2 (Ran (128): 128 amino acids) (Supplementary Fig. 4b). To further investigate this interaction, we employed an antibody recognizing both isoforms. Co-localization analyses using confocal microscopy demonstrated the interaction between TMEM63B and Ran at the plasma membrane (Supplementary Fig. 4c), which was further confirmed through direct co-immunoprecipitation experiments (Fig. 3g). To validate these observations, we then overexpressed TMEM63B-GFP-V5 and mCherry-Flag-Ran in HEK293T cells. Co-immunoprecipitation revealed interactions between TMEM63B and both Ran isoforms, with Ran (128) exhibiting notably stronger binding affinity (Supplementary Fig. 4d). Live-cell imaging provided further evidence, showing clear co-localization of Ran (128) and TMEM63B at the plasma membranes (Supplementary Fig. 4e). Subsequent detailed mapping experiments identified amino acids 1–64 of Ran (128) as essential for the interaction with TMEM63B (Supplementary Fig. 4f).

Ran is critical for establishing the RanGDP/RanGTP gradient, which is essential for NCT mediated by importin-β and CRM1 (XPO1), a major receptor responsible for transporting numerous regulatory proteins and RNAs from the nucleus to the cytoplasm^36,37. Given the importance of this transport machinery, we next evaluated the expression levels of Ran, RanGAP1, RCC1, and XPO1 in WT and TMEM63B KO cells. Remarkably, TMEM63B KO cells exhibited significant reductions of Ran and XPO1 protein levels in both BeWo cells and hTSCs (Fig. 3h and Supplementary Fig. 4g–i). This downregulation was accompanied by destabilization of the functional Ran-XPO1 complex (Fig. 3i and Supplementary Fig. 4j), suggesting a significant impairment of nuclear transport functions. Interestingly, this instability appears context-dependent, as TMEM63B KO increased the expression of the negative regulator RanGAP1 in BeWo cells but not in hTSCs (Fig. 3h and Supplementary Fig. 4g–i).

To confirm that these disruptions in the Ran pathway were specifically attributable to TMEM63B KO in trophoblast cells, we conducted rescue experiments. Notably, overexpressing Ran (128) in KO cells not only restored nucleocytoplasmic shuttling (Fig. 3j, k) but also ameliorated the proliferation deficiency (Fig. 3l). Furthermore, in TMEM63B KO BeWo cells where RanGAP1 was upregulated, siRNA-mediated knockdown of RanGAP1 also rescued NCT dynamics and restored cell proliferation (Supplementary Fig. 5a–c).

Taken together, these results establish TMEM63B as a key regulator of Ran-dependent NCT, which is critical for sustaining cellular proliferation and maintaining the integrity of the transport system.

TMEM63B facilitates CDKN1A nucleocytoplasmic transport in response to osmo/mechanical stress

The KO of TMEM63B profoundly impacts the NCT dynamics of key genes involved in cell cycle regulation. Through a heatmap analysis of genes showing significant expression alterations in cell cycle and nucleocytoplasmic shuttling pathways (Supplementary Fig. 6a, b), we identified aberrantly high nuclear retention of CDKN1A/p21 in TMEM63B KO cells (Supplementary Fig. 6c–g). To provide direct, RNA-seq-independent validation at the proteomic level, we employed compartment-specific TurboID proximity labeling in WT and TMEM63B KO cells using 3HA-TurboID-3NLS and V5-TurboID-NES constructs to label nuclear and cytoplasmic proteomes, respectively. After confirming proper localization and robust biotinylation (Supplementary Fig. 7a, b), we performed quantitative mass spectrometry and compared the nuclear-to-cytoplasmic distribution of proteins between genotypes (Supplementary Dataset 3). Proteins exhibiting an increased nuclear-to-cytoplasmic ratio in TMEM63B KO cells were significantly enriched for cell-cycle-related pathways, consistent with the proliferation defect observed in these cells (Supplementary Fig. 7c). Subsequent volcano plot analysis of this compartmentalized data explicitly identified CDKN1A/p21 as significantly enriched in the KO nucleus (Supplementary Fig. 7d). Thus, both transcriptomic and unbiased proteomic analyses independently converging on CDKN1A/p21, supporting its relevance as a downstream effector of TMEM63B deficiency. CDKN1A/p21 is a key regulatory protein of the cell cycle that functions as a molecular brake on cell proliferation by inhibiting CDKs^38,39. The inhibitory effect of CDKN1A/p21 on cell cycle progression strongly depends on its nuclear localization^40,41. Consistent with this, elevated nuclear CDKN1A/p21 levels were observed in TMEM63B KO BeWo cells and hTSCs (Fig. 4a–c and Supplementary Fig. 8a–e). To strengthen the in vivo mechanistic link between TMEM63B deficiency and CDKN1A/p21 dysregulation, we analyzed placentas from trophoblast-specific Tmem63b CKO mice. In the labyrinth zone, the principal site of placental pathology, we observed reduced expression of TMEM63B together with marked downregulation of the nuclear export machinery components XPO1 and Ran, consistent with our in vitro findings (Supplementary Fig. 8f, g). Importantly, immunohistochemistry revealed prominent nuclear accumulation of CDKN1A/p21 in the labyrinth of Tmem63b CKO placentas (Supplementary Fig. 8h). This was further supported by subcellular fractionation and western blotting, which showed an increased nuclear-to-cytoplasmic ratio of CDKN1A/p21 in the CKO placentas (Supplementary Fig. 8i, j). These findings support a model in which aberrant nuclear retention of CDKN1A/p21 contributes to impaired trophoblast proliferation and placental dysfunction in Tmem63b-deficient mice.

Fig. 4. TMEM63B facilitates CDKN1A/p21 nucleocytoplasmic transport in response to hypotonic stress.

a, b Nucleocytoplasmic distribution of CDKN1A/p21 (green) (a) and nuclear-cytoplasmic (Nuc/Cyto) localization quantification (b) in WT and TMEM63B KO BeWo cells (means ± SEM; n = 30 cells from three independent experiments, two-tailed unpaired Student’s t-test, ****P < 0.0001). Nuclei are counterstained by Hoechst (blue). c Representative p21 western blot in nuclear (H3 marker) and cytoplasmic (Tubulin marker) fractions from three independent experiments. d, e Representative images (d) and N/C ration of p21 (e) over time in cells expressing p21-mCherry-LEXY. N/C ratios of p21 significantly differed over time between WT and KO groups (mean ± SEM; n = 8 slices from three independent experiments; two-way ANOVA, ****P < 0.0001). f, g Live-cell 3D volume imaging (f) (green) and cell volume quantification (g) post-hypotonic challenge (180 mOsm/L). KO cells swelled irreversibly (means ± SEM; n = 8 slices from three independent experiments, two-way ANOVA, ****P < 0.0001). h Nuclear-cytoplasmic fluorescence was monitored over time in WT and TMEM63B KO BeWo cells expressing LEXY post-hypotonic stimulation (180 mOsm/L; mean ± SEM; n = 8 slices from three independent experiments; two-way ANOVA, ****P < 0.0001). i, j LEXY nuclear exit (i) and re-entry (j) coefficients in WT and KO cells ± hypotonic stimulation (Hypo) (means ± SEM; n = 8 slices from three independent experiments, two-tailed unpaired Student’s t-test, (i) WT vs WT Hypo: *P = 0.0192, (j) WT vs WT Hypo: *P = 0.0126, for all other comparisons: ****P < 0.0001). k Nuclear-cytoplasmic distribution of CDKN1A/p21 fluorescence was monitored over time in WT and TMEM63B KO BeWo cells expressing p21-mCherry-LEX post-hypotonic stimulation (180 mOsm/L; mean ± SEM, n = 8 slices from three independent experiments; two-way ANOVA, ****P < 0.0001). l, m p21 nuclear exit (l) and re-entry (m) coefficients in WT and KO cells ± hypotonic stimulation (Hypo) (means ± SEM; n = 6 slices from three independent experiments; two-tailed unpaired Student’s t-test, (m) WT + p21 vs KO + p21: *P = 0.0192, WT + p21 vs WT + p21 (Hypo): *P = 0.0126, for all other comparisons: ****P < 0.0001). Source data are provided as a Source Data file.

Given the crucial role of nuclear-localized CDKN1A/p21 in arresting the cell cycle, we examined whether the abnormal nuclear aggregation of CDKN1A/p21 resulted directly from disrupted nucleocytoplasmic shuttling. We generated a plasmid by directly fusing the CDKN1A/p21 cDNA to the LEXY system (p21-mCherry-LEXY), enabling real-time monitoring of its NCT. Using the p21-mCherry-LEXY reporter, we observed substantial basal nuclear accumulation of CDKN1A/p21 in TMEM63B KO cells compared to WT cells (Fig. 4d, e). The initial nuclear-to-cytoplasmic (N/C) ratio of CDKN1A/p21 was significantly higher in KO cells, suggesting pre-existing nuclear retention. Upon light-induced activation, WT cells exhibited a pronounced decrease in the N/C ratio, reflecting robust nuclear export, whereas KO cells showed minimal change. Furthermore, following cessation of the light stimulus, the absolute N/C ratio of CDKN1A/p21 remained significantly higher in KO cells than WT cells, despite comparable relative recovery rates (Fig. 4d, e). These results demonstrate that impaired nucleocytoplasmic shuttling in TMEM63B KO cells contributes to the aberrant nuclear accumulation of CDKN1A/p21 and consequent cell cycle arrest.

It is well-established that cells undergo significant volume expansion and osmotic fluctuations during mitosis^42,43, processes fundamental to proper cell proliferation^44–46. Therefore, mechanisms that sense and respond appropriately to these physical alterations are critical. TMEM63B is a channel known for its intrinsic sensitivity to mechanical and osmotic stimuli. Previous studies have established that hypoosmotic conditions (~ 180 mOsm/L) directly activate TMEM63B channel activity^19,21. Our results in BeWo cells also confirm TMEM63B activation under these hypoosmotic conditions (Supplementary Fig. 8k–o). We hypothesized that TMEM63B regulates NCT in a hypotonic, volume-dependent manner to control cell proliferation. To test this hypothesis, we employed 3D live-cell imaging under hypotonic stimulation. This acute hypoosmotic shock served the dual purpose of activating TMEM63B and mimicking the subsequent cellular osmotic stress. Remarkably, TMEM63B KO cells failed to recover from hypotonic-induced swelling (Fig. 4f, g) and displayed disrupted nucleocytoplasmic shuttling under these conditions (Fig. 4h and Supplementary Fig. 8p), contrasting with the coordinated regulation of nuclear transit observed in WT cells (Fig. 4i, j). Under hypotonic conditions, CDKN1A/p21 nucleocytoplasmic shuttling was severely disrupted in TMEM63B KO cells, characterized by a significant blockade of both nuclear export and import rates (Fig. 4k–m and Supplementary Fig. 8q).

To determine if this impaired transport was linked to the TMEM63B-Ran interaction, we overexpressed two Ran isoforms in these KO cells: a full-length Ran (128) (which interacts with TMEM63B) and a truncated version (Δ1–64 aa) lacking interaction capability (Supplementary Fig. 4f). Notably, overexpressing the full-length Ran partially restores CDKN1A/p21 nucleocytoplasmic shuttling, reducing nuclear retention. Conversely, the truncated Ran isoform had no significant effect on CDKN1A/p21 localization (Supplementary Fig. 8r, s). These findings explicitly connect TMEM63B deficiency to impaired Ran-dependent nuclear export of CDKN1A/p21, highlighting the integral role of this interaction in maintaining cellular homeostasis during osmotic stress.

Collectively, these results identified TMEM63B as a pivotal mediator of NCT in response to osmo/mechanical signals to regulate cell volume determination and thereby cell proliferation (Fig. 5).

Fig. 5. Schematic illustration of the model.

The proposed model for the role of TMEM63B in placental development in response to osmo/mechanical stimuli.

Discussion

TMEM63B, an osmo/mechanosensitive ion channel, has recently emerged as a key player in cellular responses to osmotic and mechanical stimuli. Previous studies show that TMEM63B facilitates the release of surfactant and ATP in alveolar type 2 cells, thereby maintaining pulmonary surfactant levels²³. In Drosophila, TMEM63 regulates lysosomal mechanosensitivity, modulating lysosomal morphology and function in vivo, further underscoring its importance in mechanosensing in cellular homeostasis²¹. Mutations in TMEM63B have been linked to a range of severe neurodevelopmental disorders, including Gordon Holmes syndrome, ataxia, epileptic encephalopathy, intellectual disability, and hearing loss^19,47. Here, we demonstrated that TMEM63B is a pivotal mechanosensitive molecule that regulates nucleocytoplasmic shuttling in trophoblast cell cycle progression and placenta development.

Osmotic pressure dynamically fluctuates during cell cycle progression, playing crucial roles in cellular mechanics^48,49. Three independent large-scale genomic screens^12–14 have identified TMEM63B as a critical membrane protein involved in mitotic mechanics and cell cycle regulation. These early studies, conducted prior to the identification of TMEM63B as a mechanosensitive ion channel, inadvertently laid the groundwork for understanding its potential molecular function. Our findings confirm this link: TMEM63B KO trophoblasts fail to recover their volume after hypo-osmotic shock (Fig. 4g). This impaired osmoregulation directly correlates with reduced cell proliferation (Fig. 2b), underscoring that TMEM63B governs cell cycle progression by modulating cellular volume and osmotic pressure.

The regulation of NCT is a fundamental mechanism by which external physical cues are transduced to control cell proliferation. NCT responds directly to mechanical and osmotic forces^50,51 and orchestrates cell cycle progression by governing the localization of key regulatory molecules such as cyclins and CDKs³¹. Ran, a small GTPase, serves as the central driver of NCT, maintaining a concentration gradient across the nuclear envelope that provides the energy for directional cargo movement. In this study, we identify TMEM63B as a key player in the regulation of the NCT system, particularly in the context of Ran-dependent pathways. Our data show that the loss of TMEM63B results in a significant reduction in Ran protein, which undermines the driving force for an efficient NCT system. These alterations destabilize Ran-dependent complexes, ultimately impairing the efficiency of the NCT system.

Previous study demonstrates that the localization of Ran at the plasma membrane prevents proteasomal degradation of RhoA and orchestrates its spatial distribution and activation³⁵. Our data shows that TMEM63B-Ran co-localization at the plasma membrane modulates the NCT activity, which is further confirmed by both TurboID-based proximity labeling and co-immunoprecipitation (Fig. 3g and Supplementary Fig. 4c–f). Crucially, we found that this specific membrane-associated complex is critical for NCT activity (Fig. 3j, k). This suggests that TMEM63B acts as a membrane anchor or stabilizer for Ran, thereby facilitating the relay of mechanical and osmotic stimuli from the cell periphery to the nuclear transport machinery. While our findings clarify how this interaction influences transport, the detailed mechanism by which TMEM63B regulates Ran protein levels or its GDP/GTP cycle remains to be explored.

Beyond these effects on Ran-dependent transport, TMEM63B appears to play a specific regulatory role in the XPO1 (also known as CRM1) nuclear export pathway. This assertion is supported by findings in TMEM63B KO cells. First, the formation of the Ran-XPO1 complex, which is essential for cargo recognition and transport, was significantly impaired. Notably, we observed a significant nuclear accumulation of CDKN1A/p21, a tumor suppressor protein and cell cycle inhibitor, which is normally exported from the nucleus via XPO1^52,53. This mislocalization of CDKN1A/p21 indicates a functional disruption in XPO1-mediated export, with potential implications for cell cycle regulation and tumor suppression⁵⁴. TMEM63B’s involvement in cell volume regulation and mechanosensation may represent an additional, indirect mechanism through which it influences NCT. Our RNAseq data indicate the downregulation of genes involved in NCT in TMEM63B KO cells, suggesting that TMEM63B might link mechanical cues to the transcriptional regulation of NCT-related pathways. This observation is particularly intriguing given the known role of mechanical forces in directly influencing NCT⁵¹, warranting further investigation into the interplay between these processes.

The placental defects associated with Tmem63b deficiency mirror those seen in other perinatal-lethal mutations, underscoring the critical relationship between placental health and embryonic survival^55,56. This highlights the interdependence between the placenta and the developing embryo and serves as a reminder of the placenta’s central role in fetal development. Notably, clinical cohort reveal that TMEM63B expression is significantly downregulated in placental tissues from patients with recurrent miscarriage⁵⁷, suggesting that dysregulation of TMEM63B may be relevant to human placental disorders and adverse reproductive outcomes.

In summary, our findings identify that TMEM63B is essential for trophoblast cell cycle progression and placental development. The loss of TMEM63B leads to significant defects in cell proliferation, especially in the regulation of NCT and cell volume response to osmotic/mechanical stress, thereby contributing to severe placental dysfunction in Tmem63b-deficient mice. By deepening our understanding of TMEM63B-dependent mechanisms, our study provides a mechanistic framework for understanding placental development and may open new avenues for therapeutic interventions to improve maternal and fetal health outcomes.

Methods

Mice

All animal experiments and protocols were reviewed and approved by the Shenzhen Bay Laboratory Animal Care and Use Committee (Ethics Approval No. AEZYANG202301). Mice were housed in a specific pathogen-free facility under a standard 12-h light/12-h dark cycle, with an ambient temperature maintained at (22 ± 1 °C) and a relative humidity of (55 ± 10%). All animals had ad libitum access to standard chow and water.

We utilized CRISPR/Cas9 technology to generate the global Tmem63b knockout mouse line. Specifically, exons 2–4 of the Tmem63b-201 transcript, including the start codon, were targeted and deleted in C57BL/6J fertilized eggs. This genetic deletion is anticipated to result in a non-functional Tmem63b gene. Heterozygous (Tmem63b^+/−) mice were purchased from GemPharmatech Co., Ltd. and intercrossed to generate Tmem63b knockout mice (Tmem63b^−/−).

For the conditional allele, a Tmem63b^flox/flox mouse line was generated where LoxP sites were inserted flanking exons 4–5 of the Tmem63b gene. These mice were purchased from Cyagen Biosciences Inc. (China). Trophoblast-specific ablation of Tmem63b was achieved by crossing Tmem63b^flox/flox mice with transgenic mice expressing Cre-recombinase driven by the Elf5 promoter (Elf5-Cre), kindly provided by Haibin Wang’s lab at Xiamen University. The heterozygous mice from this Elf5-Cre line were bred into the C57BL/6 strain over ten generations. Both the Tmem63b KO line and the Tmem63b^flox/flox line were backcrossed onto the C57BL/6J background for at least 10 generations to ensure genetic homogeneity.

Genotyping of these mice was conducted through PCR analysis of DNA extracted from their tails. For all experiments described in this study, 12-week-old male and female mice were used, unless otherwise specified in the respective figure legends.

Cell line

The BeWo line was purchased from ATCC. The HEK293T cell line was a gift from Dr. Tingting Chu. BeWo cells were cultured in DMEM/F12 medium (Sigma-Aldrich, #D8437), and HEK293T cells were cultured in DMEM medium (Sigma-Aldrich, #D6429). Both media contained 10% FBS (Vazyme, #F103-01) and were supplemented with 100 units/ml penicillin and 100 mg/ml streptomycin (Gibco, #15140122). All cells were cultured at 37 °C in 5% CO₂ according to the manufacturer’s instructions (ATCC).

The culture of hTSCs was performed as described previously²⁵. The use of human female placental samples was approved by the Medical Ethical Committees of Women and Children’s Hospital, School of Medicine, Xiamen University (Ethics No. KY-2022-076-h01). hTSCs were seeded on Collagen IV-coated plates (5 mg/mL, Corning, #354233) and maintained in culture medium (Advanced DMEM/F12 supplemented with 0.3% BSA (Sigma-Aldrich, #A9418), 0.5% Penicillin-Streptomycin, 1.5 mg/mL L-ascorbic acid (Wako, #013-12061), 1% ITS-X (Gibco, cat#51500), 0.1 mM 2-mercaptoethanol (Thermo Fisher Scientific, #21985023), 0.2% FBS, 2 μM CHIR99021 (R&D, #4423), 1 μM SB431542 (Wako, #031-24291), 0.5 μM A83-01 (PeproTech, #90943360), 50 ng/mL recombinant human EGF (Peprotech, #AF-100-15), 0.8 mM VPA (Wako, #227-01071), and 5 μM Y-27632 (Wako, #257-00511)). The culture medium was refreshed every other day. Cells were passaged at 90% confluency using TrypLE™ Express (Thermo Fisher Scientific, #12604021).

To generate TMEM63B KO BeWo cells, sgRNAs were designed using Benchling software (https://www.benchling.com/) for precise gene editing. These sgRNAs were then cloned into the pSpCas9(BB)-2A-GFP vector (PX458; Addgene, #48138). The same targeting approach was applied to hTSCs, with sgRNAs cloned into the lentiguide-puro vector (Addgene, #52963). All the PX458 sgRNA plasmids were transfected by electroporation on the Lonza Nucleofector™ 2b device, program X-005. All the lentiguide sgRNA plasmids were transfected with psPAX2 and pMD2.G into HEK293T cells for lentivirus generation. The cells were infected with lentivirus as described before. The KO cells, generated using either lentiguide-puro or PX458-puro plasmids, were selected with 2 µg/ml puromycin for 48 to 72 h. After selection, they were cultured for an additional 7 days to expand the cell population. The following sgRNA sequences were used: hTMEM63B sgRNA: TGGCGGTGATAGTCTCATTG and CCTCAACAACAGCAACCCCA.

Electrophysiology

For Bewo cell recordings, the bath solution and pipette solution contained: 140 mM NaCl, 5 mM KCl, 1 mM MgCl₂, 2 mM CaCl₂, 10 mM HEPES, and 10 mM glucose, pH 7.4, using inside-out patch clamp recordings. Resistances of the micropipettes were 3–5 MΩ. Membrane patches were stretched by 500 ms negative pressure pulses through the glass micropipette. The negative pressures were generated by the high-speed pressure control device (HSPC-2, ALA-Scientific) with a −20 mmHg increase for each sweep and an inter-sweep duration of 5 s. The membrane potential inside the patch was held at −80 mV. Signals were sampled at 10 kHz and filtered at 2 kHz. All recordings were performed at room temperature (~ 25 °C).

For electrophysiological recording of whole-cell configuration, whole-cell currents were recorded under voltage clamp using 8–10 MΩ borosilicate glass pipettes¹⁹. Cells were continuously perfused (3 mL/min) at room temperature with isotonic extracellular solutions containing (mM): 80 Na-gluconate, 1 Ca-gluconate, 10 HEPES, 130 Mannitol, pH 7.4 adjusted with NaOH (300 mOsm/L). To reduce the extracellular osmotic pressure without changing the ion concentration, the hypotonic extracellular solution contains (mM): 80 Na-gluconate, 1 Ca-gluconate, 10 HEPES,10 Mannitol, pH 7.4 (180 mOsm/L). The pipette solution contains (mM): 80 K-gluconate, 10 HEPES, 130 mannitol, pH 7.4 adjusted with KOH (300 mOsm/L). The osmolarities of all solutions were accurately determined using a vapor pressure osmometer. The cells were held at 0 mV before application of test protocols. The protocol for measuring the hypoosmotic-activated currents was a 100-ms ramp from −80 to 80 mV every 2 s. Solutions were applied using a pressurized perfusion apparatus (ALA-VM8, ALA Scientific Instruments) with solution exchange.

Mouse placenta histological analysis

Mice were deeply anesthetized with isoflurane, followed by immediate collection and fixation of placentas and embryos in 4% paraformaldehyde (BBI, #E672002) at 4 °C for 24 h. Fixed samples were transferred to 70% ethanol for 3–5 days before standard 4-h tissue processing using a Leica ASP6025 system. Processed tissues were paraffin-embedded and sectioned into 5 μm slices with a Leica RM2255 microtome.

For H&E staining, sections underwent 60 °C deparaffinization, sequential xylene (BGI, #1.15634.023) clearing, and graded ethanol (100% → 95% → 70%) rehydration. Nuclear staining was achieved through 5 min hematoxylin immersion with tap water rinsing, followed by brief eosin counterstaining (1–2 s) without post-wash. Final dehydration employed 95% and absolute ethanol (2 min each), xylene clearing (5 min), and neutral balsam (Solarbio, #G8590) mounting.

Immunohistochemical protocols for CD31 (Abcam, #ab182981, 1:1000), Ki67 (CST, #9129S, 1:400) and CDKN1A/p21 (HUABIO, #HA722065, 1:100) included: high-pressure EDTA (pH 9.0) antigen retrieval endogenous peroxidase quenching with 3% H₂O₂; 2-h blocking in 10% goat serum (Gibco, #16210072) and 0.3% Triton X-100 (Sigma-Aldrich, #93443); primary antibody incubation at 4 °C (16 h); HRP-conjugated secondary antibody (37 °C, 45 min); DAB chromogenic development (Maxim, #DAB4033, 50 s) and hematoxylin nuclear counterstaining. We determined the CD31 to Ki67 signal ratio in the placenta by calculating CD31 and Ki67 signal ratios in the labyrinth layer from three paired placentas. The overall ratio was averaged to minimize variability, with analysis performed using ImageJ software.

MCT1 serves as a specific marker for SynT-1, while MCT4 stains the basal membrane of SynT-2. Immunofluorescence staining for MCT1 and MCT4 was performed on TMEM63B WT and KO placenta samples using an MCT1 antibody (Millipore-Sigma, #AB1286-I), which stains the fused SynT-1 layer facing the maternal blood sinuses, and an MCT4 antibody (Santa Cruz Biotechnology Inc., #sc-376140), which stains the fused SynT-2 layer facing the fetal blood vessels. The samples were blocked for 2 h with 10% goat serum and 0.3% Triton ×-100 before primary antibody incubation (4 °C overnight), and secondary antibody incubation (Invitrogen&trade, #A11039, #35512, 37 °C for 2 h). The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Beyotime, #C1002). A Nikon Ti2E microscope and Zeiss LSM980 were used for examining and photographing the immunostainings.

Alkaline phosphatase (AP) staining was performed to assess the differentiation of STGCs. Briefly, sections were deparaffinized and rinsed in PBS, then incubated in wash buffer (NTMT: 0.1 M NaCl; 0.1 M Tris, pH 9.5; 0.05 M MgCl₂; and 0.1% Tween 20, prepared fresh). AP activity was detected using an ALP kit (Biossci, #BP090), resulting in the formation of a blue precipitate, as previously described. Microscopic imaging of histological samples was taken at 20×using a Leica microscope (Aperio VERSA 200) with cellSens Dimension software.

3D trophoblast organoid

Three-dimensional TOs derived from hTSCs were established according to a published protocol²⁵. Specifically, 0.5 × 10⁴ WT or TMEM63B KO hTSCs were embedded in 25 µL Matrigel domes (Novoprotein, #NMO-G010-PF) within 48-well plates. Organoid cultures were sustained in trophoblast organoid medium (TOM) containing Advanced DMEM/F12 basal medium supplemented with 50 ng/mL EGF, 100 ng/mL FGF2 (PeproTech, #100-18 C), 50 ng/mL HGF (PeproTech, #100-39-10), 80 ng/mL R-spondin-1 (MedChemExpress, #HY-P7114), 1.5 µM CHIR99021, 5 µM Y27632, 500 nM A83-01, 2.5 µM PGE₂ (MedChemExpress, #HY101952), 1.25 mM N-acetyl-L-cysteine (Sigma-Aldrich, #A9165), 1 × N2 (Gibco, #17502048), 1 × B-27 minus vitamin A (Gibco, #17504044) supplements and 2 mM L-glutamine (Sigma-Aldrich #25030024). TOM supplemented with puromycin (2 μg/mL) was changed every 2 days. Bright field images of TOs were captured using a Leica DMI1 inverted microscope.

Immunofluorescence

Cells were grown on glass coverslips fixed with 4% (w/v) paraformaldehyde (BBI, #E672002) for 20 min at room temperature, washed with PBS, permeabilized and blocked with 10% Goat serum and 0.3% (v/v) Triton ×-100 in PBS for 60 min. Subsequently, cells were washed twice for 5 min with PBS and then incubated with the primary antibody diluted in TBST containing 5% BSA at 4 °C overnight. Cells were next washed twice with PBS and then incubated with the appropriate secondary antibody for 1 h at room temperature. The cells were then incubated overnight with specific primary antibodies as follows: anti-TMEM63B (Affinity, #DF15864 1:50), anti-Ran (Santa Cruz Biotechnology Inc., #sc-271376, 1:50), anti-p21 (Cell signaling

technology, #2947S, 1:500). For fluorescence labeling, anti-mouse Alexa Fluor 633 (Invitrogen&trade, #35562, 1:500) or anti-rabbit Alexa Fluor 488 (Invitrogen&trade, #A11039, 1:500) were used, and nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Beyotime, #C1002) or Hoechst33342 (Invitrogen&trade, #H3570). Fluorescence images were acquired with a Zeiss 980 confocal microscope, and subsequent analysis was conducted using ZEN3.2 and ImageJ software.

Transfection and live-cell imaging

For transfection, cells were grown on glass coverslips for 24 h prior. On the day of transfection, the cells were washed once with Opti-MEM (Gibco, #11058021). Transfection was performed using a 4:1 (w/w) ratio of PEI (Polysciences, #24314-2) to DNA. PEI and DNA were each diluted in 150 µL Opti-MEM, mixed, and incubated for 15 min. The complex was gently added to the cells, avoiding disturbance. The plate was then returned to the incubator with 5% CO₂. After 4 h of transfection, the Opti-MEM medium was replaced with full culture medium. Then, 48 h later, live cell imaging was captured using a Zeiss 980 confocal microscope.

For live cell plasma membrane detection, cells were stained with CellMask Plasma Membrane Marker (Invitrogen&trade, #C10046, 2.5 μg/mL) for 10 min. Fluorescence images were captured using a Zeiss 980 confocal microscope.

siRNA design and transfection

Two synthetic siRNAs targeting CDKN1A/p21 were designed and synthesized by General Biol (Anhui) Co., Ltd., incorporating 2’-O-methyl and phosphorothioate modifications at the terminal nucleotides to enhance stability. The siRNA sequences were as follows: siRNA-1, 5’-AGACCAGCAUGACAGAUUUTT-3’; siRNA-2, 5’-GAGACUCUCAGGGUCGAAATT-3’. Additionally, a three-in-one siRNA cocktail targeting RanGAP1 (MedChemExpress, #HY-RS11650) and a non-targeting control siRNA (MedChemExpress, #HY-150150) were utilized. BeWo cells were transfected with the respective siRNAs at a final concentration of 50 nM using PEI according to the manufacturer’s instructions. Cells were harvested 72 h post-transfection, and knockdown efficiency was verified by qRT-PCR and Western blotting.

Quantitative RT-PCR

Total RNA was extracted using RNA Isolator Reagent (Vazyme, # R401) and quantified with a NanoDrop Spectrophotometer (Thermo Fisher Scientific). Reverse transcription of 1 μg total RNA was carried out using HiScript III All-in-one RT SuperMix (Vazyme, #R333-01) to synthesize cDNA, which was then used for quantitative real-time PCR (qRT-PCR) with Taq Pro Universal SYBR qRT-PCR Master Mix (Vazyme, #Q712-03). The QuantStudio 6 Pro (Thermo Fisher Scientific) was set for an initial denaturation at 95 °C for 30 s, followed by 40 cycles of amplification at 95 °C for 10 s and 60 °C for 30 s. Fluorescein emission was measured during amplification, and each sample was run in duplicate, normalizing average cycle threshold (Ct) values to GAPDH Ct values. Relative gene expression was calculated using the ∆∆Ct method.

Cell proliferation assay

The CCK8 assay (Vazyme, #A311-02) was used to assess cell viability in WT and TMEM63B KO BeWo cells. Cells were seeded at 3000–4000 per well in 96-well plates (Corning, #3599) and cultured at 37 °C in a humidified atmosphere with 5% CO₂. The assay was performed daily for 4–5 days, with absorbance (OD value) measured at 450 nm using a Neo2 plate reader. For hTSCs, the cells were counted after digestion at 10,000 cells/well in a 96-well culture plate. A 96-well culture plate with 100 μL medium/well was cultured with 20 μL MTS reagent (Promega, #G3580). Three to five groups were redrilled, and a control group was set. The 96-well culture plate was put back into the incubator and incubated at 37 °C and 5% CO₂ for four hours. The OD value was measured at 490 nm by an enzyme-labeled instrument. All experiments were performed in triplicate. Background values from empty wells were subtracted, and data were normalized to vehicle-treated controls.

Cell cycle analysis

For the EdU cell proliferation assay, 3 × 10⁵ cells were plated into 6-well plates and cultured for 24 h. The cells were then incubated with 50 μM EdU for 2 h, followed by fixation, permeabilization, and EdU staining according to the BeyoClick™ EdU Cell Proliferation Kit with Alexa Fluor 594 (Beyotime, #C0078S) instructions. Nuclei were counterstained with Hoechst 33342 for 30 min.

For the PI dye cell proliferation assay, 3 × 10⁵ cells were plated in 6-well plates, cultured for 24 h, and then trypsinized, fixed in 70% ethanol, and stored at 4 °C. To assess cell cycle distribution, cells were treated with RNase A (YEASEN, #10406ES03, 100 µg/mL) to digest RNA, ensuring DNA-specific staining. Propidium iodide (Biotium, #40016, 40 µg/mL) was then added for staining. Samples were analyzed by flow cytometry using a BD LSRFortessa SORP instrument, and data were analyzed using FlowJo 10.8.1. Representative gating strategies for doublet exclusion and target population selection are shown in Supplementary Fig. 2c, d.

Western blot

The cell samples were washed with PBS and lysed using RIPA buffer (50 mM Tris-HCl, pH 8.0; 150 mM NaCl; 1% NP-40, 0.5% deoxycholate, 0.1% SDS) supplemented with Halt protease inhibitor cocktail and EDTA (Thermo Fisher Scientific, #87786). The lysates were incubated on ice with vortexing every 10 min for 30 min, followed by centrifugation at 4 °C for 15 min at 12,000 × g. The supernatants were collected, mixed with 5 × SDS-PAGE Protein Loading Buffer (YEASEN, #20315ES05), and quantified using a Nanodrop device. Proteins (20 µg) were separated on a 4–20% polyacrylamide gel (YEASEN, #36270ES10; ACE, #ET15420Gel) and transferred to a PVDF membrane (Merck Millipore, #IPVH00010; Vazyme, #E802-01). The membrane was blocked with 5% skim milk (BBI, #A600669-0250) and probed with primary antibodies: Anti-TMEM63B (1:500), Anti-Ran (1:500), Anti-p21 (1:500), Anti-RanGAP1 (Affinity, #DF7315, 1:500), Anti-RCC1 (Proteintech, #22142-1-AP, 1:500), Anti-XPO1 (Santa Cruz Biotechnology Inc., #sc-74454, 1:500), Anti-GAPDH (Beyotime, #AF1186, 1:5000), and Anti-β-Actin (ABclonal, #AC004, 1:10,000). Secondary antibodies used were HRP-linked-anti-mouse IgG (Thermo Scientific, #A28177, 1:10,000) and HRP-linked-anti-rabbit IgG (Sigma-Aldrich, #A0545, 1:10,000). Protein bands were visualized using an enhanced chemiluminescence kit (Vazyme, #E423-01). Uncropped and unprocessed scans of all blots and gels are provided in the Source Data file.

Placental labyrinth isolation and protein extraction

Trophoblast-specific ablation of Tmem63b was achieved by crossing Tmem63b^flox/flox mice with transgenic mice expressing Cre recombinase driven by the Elf5 promoter (Elf5-Cre). Pregnant dams were euthanized at embryonic day 16 (E16), and the placentas were promptly dissected. To specifically isolate the labyrinth zone, the maternal decidua and surrounding tissues were carefully microdissected and removed under a stereomicroscope. For total protein extraction, the isolated labyrinth tissue was weighed and homogenized in RIPA lysis buffer (supplemented with a protease inhibitor cocktail) at a 1:10 tissue-to-buffer ratio (w/v; 10 µL RIPA buffer per 1 mg tissue). Following incubation 20 min on ice to ensure complete lysis, the homogenates were centrifuged at 12,000 × g for 15 min at 4 °C. The resulting supernatant was collected, and protein concentration was determined prior to subsequent western blot analysis.

Immunoprecipitation

Cell lysates from BeWo and HEK293T cells were prepared according to the Western blot protocol. The cell samples were washed with PBS and lysed using RIPA buffer (50 mM Tris-HCl, pH 8.0; 150 mM NaCl; 1% NP-40, 0.5% deoxycholate, 0.1% SDS) supplemented with Halt protease inhibitor cocktail and EDTA (Thermo Fisher Scientific, #87786). Following centrifugation at 12,000 × g for 15 min at 4 °C, the supernatants were collected to remove cell debris. The precleared 1 mg lysates were then incubated overnight at 4 °C with the 10 μg Ran antibody or normal IgG (Abclonal, #AC011) as a control. Subsequently, 40 μL of protein A/G Magnetic IP/Co-IP Kit (Vazyme, #PB201-01) was added to the lysates, which were then rotated at 4 °C for 6 h. The immune complexes were washed five times with TBST buffer, resuspended in 60 μL of 2 × SDS Loading buffer, heated at 95 °C for 10 min, and analyzed by western blot.

RNA sequencing analysis

Total RNA was purified from BeWo cells after TMEM63B KO using TRIzol (Vazyme, #R401). Triplicate samples were collected for each group. RNA-seq libraries were prepared using the Illumina stranded mRNA sample preparation kit (NEB, #E7770) following the manufacturer’s protocol by Gene Denovo Biotechnology Co., Ltd (Guangzhou, China). Sequencing was performed on an Illumina NovaSeq 6000 instrument, generating 150-base-pair paired-end reads. Differentially expressed genes (DEGs) were identified based on |log2 fold change| ≥ 1.2 and an adjusted p-value < 0.05.

The DEGs from each comparison were uploaded into Ingenuity Pathway Analysis for further study. Core Analysis was conducted to identify statistically significant canonical pathways using a right-tailed Fisher’s exact test. The Ingenuity pathway database was used to evaluate the activation state of these pathways, calculating an activation z-score to determine the direction and significance of regulation for each pathway.

Cellular synchronization

BeWo cells were synchronized using a thymidine-nocodazole block. Cells were treated with 2 mM thymidine (MedChemExpress, #HY-N1150) for 24 h, released for 9 h, and then arrested in mitosis with 40 ng/mL nocodazole (MedChemExpress, #HY-13520) for 8 h. Mitotic cells were collected by shake-off and either processed for RNA extraction or FACS analysis (0 h time point) or re-plated for later time points (every 1.5 or 3 h) for cell cycle analysis. Protein extraction was performed as described for Western blot. Cells for cell cycle analysis were trypsinized, fixed in 70% ethanol, and stored at 4 °C.

Proximity labeling mass spectrometry

The TurboID protocol outlines proximity labeling for proteomic analysis as paper before³⁰. In detail, BeWo cells were transfected with plasmids encoding TMEM63B-TurboID. Forty-eight hours post-transfection, biotin was added to the medium at a final concentration of 500 μM for labeling for 20 min. The cells were then washed five times with cold PBS and lysed in a RIPA buffer (50 mM Tris-HCl, pH 8.0; 150 mM NaCl; 1% NP-40, 0.5% deoxycholate, 0.1% SDS, and 1% protease inhibitor cocktail). The lysate was centrifuged at 15,000 × g for 10 min, and the supernatant was collected for further analysis. For biotin pull-down experiments, the 2 mg supernatant was incubated with 200 μl of streptavidin beads (Thermo Fisher Scientific, #88816) overnight at 4 °C. The beads were then washed three times with the same lysis buffer. A portion of the sample was boiled in SDS loading buffer at 100 °C for 5 min and analyzed by immunoblotting. The remaining samples were processed for quantitative mass spectrometry analysis.

Beads were denatured in 50 μL buffer (50 mM ammonium bicarbonate (ABC), 6 M urea, 10 mM TCEP, 20 mM CAA) for 60 min at 37 °C, diluted threefold with 50 mM ABC, and digested overnight at 37 °C (600 rpm) with sequencing-grade trypsin (2 μg/100 μL). Digestion was quenched with 10% formic acid (FA) to a final ratio of 1:25 (v/v). For high-affinity interactions, beads were additionally incubated in 80% acetonitrile (ACN)/0.1% FA for 5 min to maximize peptide recovery. After removing organic solvents via vacuum centrifugation, peptides were desalted using Thermo Hypersep C18 cartridges (100 mg). Cartridges were pre-conditioned with 1 mL 100% ACN and equilibrated with 1 mL 0.1% FA. Samples were acidified to a final concentration of 0.5% FA, loaded (with the flow-through reloaded twice to maximize binding), washed with 1 mL 0.1% FA, and eluted with 600 μL 80% ACN. Vacuum-dried eluates were reconstituted in 2% ACN/0.1% FA and centrifuged (12,000 × g, 5 min) prior to MS analysis.

LC-MS/MS was performed using an Orbitrap Fusion Lumos Tribrid and an Orbitrap Astral MS coupled to a Vanquish Neo UHPLC via an EASY-Spray source (Thermo Fisher Scientific). The MS operated in positive data-independent acquisition (DIA) mode (spray voltage 2100 V).

Mass spectrometry analysis

For TMEM63B-TurboID proximity labeling., LC-MS/ MS raw data were processed with Proteome Discoverer (PD) (version 2.4.1.15) using the SequestHT search engine for protein identification. The precursor detector node in PD 2.4 was added to reduce the influence of chimeric spectra. The database was UniProt reviewed human protein database and common contaminants. Database searching parameters were set: enzyme specificity for trypsin and up to three missed cleavages were allowed, minimum peptide length was 6, and mass tolerance for precursor and fragment ions were set as 10 ppm and 0.02 Da, respectively. Cysteine carbamidomethylation and methionine oxidation were set as fixed and variable, respectively. The false discovery rate (FDR) was calculated using the Percolator algorithm provided by PD. FDR on peptide and protein levels was 1%. The six MS data were combined to perform database searching to obtain the relative abundance of proteins.

For TurboID-NLS and TurboID-NES proximity labeling in WT and TMEM63B KO BeWo cells, raw files were analyzed in direct DIA mode using Spectronaut 19.0 (Biognosys, Schlieren, Switzerland) equipped with the DirectDIA+ (Deep) workflow. Analysis was performed with default BGS factory settings against the Homo_sapiens (Human)_UP000005640_9606_20659entries_20251030 FASTA database. Digestion was set to Trypsin/P (maximum 2 missed cleavages allowed), with a maximum of 5 modifications permitted per peptide. Quantification was performed at the MS2 level using peak Area as the quantity type. The sixteen MS data were combined to perform database searching to obtain the relative abundance of proteins.

For both pipelines, cysteine carbamidomethylation was set as a fixed modification, while methionine oxidation was configured as a variable modification; Spectronaut additionally included protein N-terminal acetylation as a variable modification. The FDR was strictly controlled at 1% (0.01) at both the peptide and protein levels for both analyses (including protein groups and peptide-spectrum matches in Spectronaut), with PD specifically utilizing the Percolator algorithm for FDR calculation.

Integrated analysis of RNA-seq and MS proteomics

In the integrated analysis of RNA-seq and MS proteomics, overlapping KEGG pathways were identified. To determine the overall significance of these co-altered pathways, a rank sum strategy was applied. Briefly, pathways were ranked independently in both datasets based on their log(p-value). The respective rank numbers were then added together to generate a combined rank score, which was used to order the final list of overlapping pathways.

Component separation experiments

Nuclear and cytoplasmic RNA were isolated from WT and TMEM63B KO cells cultured in 15 cm dishes. The process began with washing the cells in ice-cold 1× PBS, followed by harvesting them in 1 mL of ice-cold PBS. The cells were then centrifuged at 1000 × g for 10 min to obtain a pellet. Next, the cell pellets were resuspended in 200 µL of lysis buffer A, which contained 10 mM Tris (pH 8.0), 140 mM NaCl, 1.5 mM MgCl₂, 0.5% Nonidet P-40, and 2 mM Vanadyl ribonucleoside complexes. This suspension was incubated on ice for 5 min to allow cell lysis. After a 5-min centrifugation at 1000 × g at 4 °C, the supernatant, containing the cytoplasmic fraction, was mixed with 1 mL of TRIzol for RNA purification. For the nuclear fraction, the pellets were washed twice with lysis buffer A and once with lysis buffer A supplemented with 0.5% sodium deoxycholate. Following these washes, the nuclear material was resuspended in 1 mL TRIzol buffer for RNA extraction. To ensure equal cell equivalents for comparison, one-third of the protein yield from each fraction was loaded for western blotting.

Photoactivation experiment and quantification

This method was described in the previous article⁵¹. In brief, NLS-mCherry-LEXY (pDN122) plasmid constructs are available via Addgene (plasmid #72655). To investigate CDKN1A/p21 NCT, we created a mCherry-tagged CDKN1A/p21 fusion construct. Specifically, the full-length human CDKN1A/p21 coding sequence (UniProt ID: P38936) was PCR-amplified from BeWo cell complementary DNA (cDNA). The amplified fragment was subsequently digested and cloned into the pDN122 vector. Photoactivation experiments were conducted using a Zeiss LSM900 inverted confocal microscope equipped with a 63 × 1.46 NA oil immersion objective and Zeiss ZEN3.2 software. An argon laser with a 561 nm wavelength was employed for acquisition, while a 450 nm laser was utilized for stimulation. In these experiments, six images were captured before stimulation, followed by 70 images during stimulation. Throughout the stimulation phase, the 450 nm laser was applied to the entire field of view every 10 s for 1 s at 100% laser power. The 450 nm laser was then removed, and an additional 80 images were acquired every 10 s during the recovery period. To obtain the entry and exit coefficient, a single exponential equation was fitted to the (Nuclear-cytoplasmic) N/C ratio of each cell; the calculation formula refers to the article⁵¹.

Three-dimensional image of hypotonic stimulated cells

For live cell imaging under hypotonic conditions (180 mOsm/L), cells were evenly plated on confocal dishes and allowed to adhere. The isotonic starting solution was Hank’s Balanced Salt Solution (Gibco, #14025092). The hypotonic solution target 180 mOsm/L was prepared by adding distilled water (ddH₂O) to this isotonic solution at a volume ratio of 2 parts ddH₂O to 3 parts isotonic solution. The confocal microscope software was used to identify the center of each target cell, and a Z-stack range of ± 30 μm was set to capture the entire cell volume. The imaging system was configured to acquire a complete Z-stack every 25 s over 500 s, enabling real-time monitoring of cellular responses. Environmental conditions, such as temperature and humidity, were maintained to preserve cell viability. After imaging, Imaris 9.9 software was utilized to segment each cell and calculate its volume at each time point, allowing for the analysis of morphological and volumetric changes under osmotic stress.

Statistical analysis

Statistical analyses were performed using Prism 10 (GraphPad) for Windows. The two groups were compared with a two-tailed Student’s t-test. Two-way ANOVA analysis, followed by the Dunnett test, was used to compare curves over time. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 were considered significant, and P > 0.05 was considered not significant (ns). Details of statistical analyses and biological replicates are described in each figure legend.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

Y.Z. conceived and designed the project. Y.Z. and B.C. supervised the project. M.C., R.L., W.Z., J.H., K.C., and X.L. performed the experiments; M.C. and R.L. analyzed the data; Y.Z., B.C., and M.C. wrote the manuscript with inputs from all authors.

Peer review

Peer review information

Nature Communications thanks Deborah Burnett and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

Y.Z. discloses support for the research of this work from the National Natural Science Foundation of China (grant number 32300603), Guangdong Pearl River Program (grant number 2023QN10Y164), Shenzhen Medical Research Fund (grant number B2502023), and Major Program of Shenzhen Bay Laboratory. B.C. discloses support for publication of this work from the National Natural Science Foundation of China (grant number 82130047 & U25A2028). M.C., R.L., W.Z., J.H., K.C., and X.L. declare no relevant funding.

Data availability

All data generated or analyzed during this study are included in this article (and its supplementary files). The mass spectrometry data generated in this study have been deposited in the ProteomeXchange database⁵⁸ under accession code PXD061030 and PXD077271. The RNA sequencing data generated in this study have been deposited in the GEO database under accession code GSE290233. The processed data generated in this study are provided in the Supplementary Data files. Source data are provided with this paper.

Code availability

No original code was generated in this study.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-73992-3.