Abstract
Early pregnancy is characterised by elevated circulating levels of vitamin D binding protein (DBP). The impact of this on maternal and fetal health is unclear but DBP is present in the placenta, and DBP gene variants have been linked to malplacentation disorders such as preeclampsia. The functional role of DBP in the placenta was investigated using trophoblastic JEG3, BeWo and HTR8 cells. All three cell lines showed intracellular DBP with increased expression and nuclear localisation of DBP in cells treated with the active form of vitamin D, 1,25-dihydroxyvitamin D (1,25D). When cultured in the serum of mice lacking DBP (DBP−/−), JEG3 cells showed no intracellular DBP indicating uptake of exogenous DBP. Inhibition of the membrane receptor for DBP, megalin, also suppressed intracellular DBP. Elimination of intracellular DBP with DBP−/− serum or megalin inhibitor suppressed matrix invasion by trophoblast cells and was associated with increased nuclear accumulation of G-actin. Conversely, treatment with 1,25D enhanced matrix invasion. This was independent of the nuclear vitamin D receptor but was associated with enhanced ERK phosphorylation, and inhibition of ERK kinase suppressed trophoblast matrix invasion. When cultured with serum from pregnant women, trophoblast matrix invasion correlated with DBP concentration, and DBP was lower in first-trimester serum from women who later developed preeclampsia. These data show that the trophoblast matrix invasion involves uptake of serum DBP and associated intracellular actin-binding and homeostasis. DBP is a potential marker of placentation disorders such as preeclampsia and may also provide a therapeutic option for improved placenta and pregnancy health.
Introduction
Vitamin D binding protein (DBP) is a serum globulin associated with the systemic transport of vitamin D metabolites (Chun 2012). Glomerular filtration of DBP and its primary cargo, the main circulating form of vitamin D, 25-hydroxyvitamin D (25D), play a pivotal role in vitamin D endocrinology. Recovery of the DBP–25D complex from the glomerular filtrate by proximal convoluted tubular cells of the kidney occurs via endocytic uptake of DBP, utilising the megalin receptor. This recovery of 25D from glomerular filtrates facilitates renal conversion of 25D to active 1,25-dihydroxyvitamin D (1,25D) (Nykjaer et al. 1999), via the vitamin D-activating enzyme 1α-hydroxylase (CYP27B1), which is also expressed by proximal tubule cells (Zehnder et al. 1999). Although only 5% of DBP molecules have vitamin D metabolites bound at any given time, megalin-mediated uptake of DBP in proximal tubules also functions to maintain circulating levels of 25D. Mice with knockout of the DBP (Gc) (Safadi et al. 1999) or megalin (Lrp2) genes (Nykjaer et al. 1999) have extremely low serum levels of 25D, and single nucleotide polymorphisms (SNPs) of the human DBP gene (GC) are major contributors to the genetic component of serum 25D status (Wang et al. 2010).
Megalin is present at several extra-renal sites (Lundgren et al. 1997), including the placenta (Burke et al. 2013), where its expression is coincident with DBP (Ma et al. 2012). Both maternal decidua and fetal trophoblast also express CYP27B1 and the intracellular vitamin D receptor (VDR) for 1,25D (Zehnder et al. 2001). Thus, the placenta, like the kidney, has a significant capacity for vitamin D metabolism that may be supported by megalin-mediated DBP transport. Circulating levels of 1,25D (Kumar et al. 1979) and DBP (Jorgensen et al. 2004) are increased during early pregnancy but the precise function of DBP in the placenta is unclear and may involve known vitamin D-independent functions of DBP. These include a potential role as a macrophage-activation factor (Benis & Schneider 1996), and in fatty acid transport (Calvo & Ena 1989). DBP also binds the monomeric, globular form of actin (G-actin) with high affinity, allowing DBP to compete with other established actin-regulating factors such as gelsolin which incorporates G-actin into filamentous actin (F-actin) (Otterbein et al. 2002). In this way, DBP can also function as a systemic actin-scavenger, with a potential role in protecting against tissue damage due to systemic F-actin accumulation (Luebbering et al. 2020), although the DBP-actin complex may also fulfil a pro-inflammatory role as a neutrophil chemotactic factor (Kew 2019).
Variations in serum vitamin D metabolites (Bodnar et al. 2007, Wei et al. 2013) and SNPs for GC (Baca et al. 2018, Naidoo et al. 2019) have been linked to adverse events in pregnancy such as the malplacentation disorder preeclampsia but the mechanisms underpinning these associations remain unclear. In particular, although DBP and vitamin D metabolites are abundant in the placenta, their role in placental development has yet to be defined. In the current study, we show that trophoblast cells internalise extracellular DBP and that this process is essential for trophoblast matrix invasion. Decreased cellular uptake of DBP was associated with increased nuclear accumulation of G-actin and decreased capacity for trophoblast matrix invasion. The concentration of DBP in serum from pregnant women correlated with capacity for trophoblast matrix invasion in ex vivo assays, and DBP levels in first-trimester pregnancy serum samples were significantly lower in women who developed preeclampsia later in pregnancy. These data indicate that serum DBP is a crucial circulating factor in early pregnancy. Trophoblast uptake of maternal DBP may be pivotal to early placental development, with dysregulation of this process leading to associated disorders of placentation such as preeclampsia.
Materials and methods
Cell culture and reagents
Choriocarcinoma trophoblastic cell lines JEG3 and BeWo (European Collection of Authenticated Cell Cultures), non-neoplastic first-trimester extravillous like trophoblasts cells HTR8 (a kind gift from Dr S. Gross, Aston University) were routinely cultured at 37°C and 5% CO2 in MEM (Sigma Aldrich), DMEM/F12 Ham (1:1) with HEPES (Thermo Fisher Scientific), and RPMI-1640 medium W/L-Glutamine (Thermo Fisher), respectively, each supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific). Thyroid papillary carcinoma (TPC) cell line (a kind gift from Prof CJ McCabe, University of Birmingham) was cultured in RPMI-1640 medium W/L-Glutamine (Thermo Fisher) supplemented with 10% FBS. Cells were cultured on growth factor-reduced Matrigel pre-coated transwell plates (Corning® BioCoat™ Matrigel®, 8.0 micron). All cells were treated for 2–72 h with vehicle (0.1% ethanol), 1,25D (1–100 nM) (Enzo Lifesciences), DBP (East Coast Bio) (3 µM), the megalin inhibitor receptor-associated protein (RAP) (1 µM) Enzo Life Sciences) or the ERK-inhibitor U0126 (1 µM) (Cell Signaling).
Cell proliferation
All cells were seeded at a concentration of 5 × 103 cells/well in a 24-well plate, and proliferation assessed by quantification of nuclear incorporation of 5-bromo-2-deoxyuridine (BrdU) using a BrdU assay kit as per manufacturer’s instructions (Cell Signaling Technology).
Matrix invasion
Cell invasion of the matrix was evaluated using growth factor-reduced Matrigel-coated transwells. All cells were treated with pre-warmed culture medium containing 2% FBS for 24 h prior to passage. Cells were then trypsinised, re-suspended in 2% FBS culture medium and 5 × 104 cells per well seeded onto transwell inserts in the upper chamber of each well. Complete media (with 10% FBS) was added to the lower chamber of each transwell. Immediately after seeding cells were treated according to specific experiments for a further 48 h. The lower surface of the transwell inserts was then washed with PBS, then with 95% ethanol and stained with haematoxylin (Sigma), followed by washing with Scott’s water, and further staining with eosin-Y (VWR chemicals). After further washes with 70% ethanol and 99% ethanol, transwell Matrigel inserts were dried at room temperature, and the lower surface of each insert was imaged using a microscope (Leica DM ILM inverted) at ×10 magnification. Under blinded conditions, five images per well were taken for each well and the number of invaded cells was manually counted for each image. Each transwell was imaged five times at five different image quadrants.
In some experiments, quantification of invaded cells in the Matrigel transwell assay was carried out using crystal violet (Sigma) to stain the invading cells, and acetic acid to solubilise the 1% crystal violet stain. Cell seeding and incubation were carried out as described previously. Invaded cells on the transwell were washed twice with PBS, then stained with 1% crystal violet for 10 min, and then washed again with PBS. Following this, transwell inserts were left to dry at room temperature. Images of each insert were taken for counting (five quadrants per transwell as explained previously). Transwell inserts were then immersed in 400 µL of 30% acetic acid (in 24-well plate) and shaken for 10 min at room temperature. Following this, 100 µL of the blue-stained acetic acid solution was pipetted into triplicate wells in a 96-well plate and absorbance was measured at 590 nm and 405 nm (OD value) using an ELISA plate reader (SpectraMax ABS, Molecular devices, San Jose). Each experiment was repeated multiple times as indicated and values are reported as percentage (%). The total number of cells invading through the transwell was calculated by converting absorbance values to cell numbers using a standard curve with known cell numbers. Percentage invasion was obtained by dividing the number of cells invaded by the number of cells seeded.
Quantitative RT-PCR
Total RNA was extracted from cell cultures using Trizol reagent (Sigma Aldrich, Lot no. BCBV4616) as per manufacturer’s instructions. For each sample, 200–400 ng RNA was then reverse transcribed using an RT Kit (Thermo Fisher Scientific, 4368814) according to the manufacturer’s instruction, and cDNAs amplified for the following genes: vitamin D receptor (VDR) (Thermo Fisher, Hs00172113_m1); 24-hydroxylase (CYP24A1) (Thermo Fisher, Hs00167999_m1); vitamin D binding protein (DBP/Gc) (Thermo Fisher, Hs00167096_m1); matrix metalloproteinase 2 (MMP2) (Thermo Fisher, Hs01548727_m1); tissue inhibitor of metalloproteinase 1 (TIMP1) (Thermo Fisher, Hs01092512_g1); megalin/LRP2 (Thermo Fisher Hs00189742_m1); beta-actin (Hs01060665_g1), GAPDH (Hs02758991_g1) and 18S rRNA (Hs99999901_s1) were used as housekeeping internal standards. The cDNA amplification was carried out using GoTaq qPCR MasterMix (ThermoFisher, 4318157) in a thermocycler (GeneAmp PCR System 2700, ThermoFisher Scientific) with amplification at 50°C for 2 min and 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min. Differences in mRNA expression were assessed statistically using raw δCt values VDR mRNA where δCt= Ct target gene – Ct housekeeping gene. Expression of mRNA was expressed visually as 1/ δCt.
Western blot analysis
Whole cell protein lysates were extracted using radioimmunoprecipitation assay (RIPA) buffer (with Tris-EDTA) with protease and phosphatase inhibitor. Cytoplasmic and nuclear proteins were fractionated using NE-PER nuclear and cytoplasmic extraction reagents (Thermo Fisher Scientific), as per the manufacturer’s instructions. Proteins were separated using SDS-polyacrylamide (10%) gel electrophoresis, transferred to nitrocellulose membranes, and probed with various antibodies using chemiluminescence (Pierce ECL Plus, Thermo Fisher Scientific). Proteins quantified by Western blotting were: VDR (Santa Cruz, D-6), DBP (Abcam), ERK1/2 (ThermoFisher, MA5-15134, K.913.4), pERK 1/2 (ThermoFisher,MA5-15173, S.812.9). β-actin (Abcam) was used as a housekeeping control protein for whole cell lysates and cytoplasmic proteins (Supplementary Table 1). Lamin B1 (Abcam) was used as a housekeeping protein for nuclear lysates, and Na-K-ATPase (Abcam) was used as a housekeeping protein for membrane lysates. Secondary antibodies used were goat anti-mouse HRP (Abcam), and goat anti-rabbit HRP (Abcam).
Immunofluorescence analysis of cellular protein expression
Cells were cultured on coverslips or Matrigel transwell inserts using 2% FBS culture medium. The resulting monolayers were washed ×3 with PBS and fixed in 3% paraformaldehyde at room temperature for 20 min. Cells were then incubated for 10 min in chilled 100% methanol or Triton X-100 (Sigma) according to the antibody used. This was followed by a PBS wash and blocking with 10% neonatal calf serum (N4637) for 30 min. Incubation with primary antibody (Supplementary Table 1, see section on supplementary materials given at the end of this article) in 1% BSA (Merck) in PBS was then carried out for 1 h at room temperature, followed by ×3 washes with PBS. Preparations were then incubated with secondary antibody or Hoechst stain for nucleus (Invitrogen) at 1:1000 dilution, mixed with 1% neonatal calf serum and 1% BSA. Secondaries used were Alexa Fluor 488 -conjugated goat anti-mouse IgG (ThermoFisher) and Alexa Fluor 594 -conjugated goat anti-rabbit IgG (ThermoFisher, A11037) at 1:250 dilution, for each coverslip. Following this, the conjugated antibodies were mixed with 1% neonatal calf serum and 1% BSA, and the samples were incubated for 1 h in these antibodies. This was followed by washing three times with PBS. The coverslips and the Matrigel transwell base were then mounted on Thermo Fisher ProLong™ Diamond Antifade Mountant media (ThermoFisher). Slides were imaged with a Confocal Microscope Zeiss LSM 780, and analysed and quantified using ImageJ Fiji (NIH, USA). Expression levels for target proteins were determined using ImageJ Fiji software (NIH) by measuring ‘area of fluorescence colour’ subtracted from ‘integrated density’, then multiplied by ‘average background area’, and reported as corrected total cell fluorescence (CTCF).
siRNA knockdown of VDR
VDR siRNA was used to knockdown VDR mRNA expression in JEG3 and TPC cells. ON-TARGETplus Human VDR (7421 siRNA (Dharmacon, L-003448-00-0010) was used for VDR knockdown at a concentration of 100 nM. A scrambled sequence siRNA (Ambion, 4390843) (100 nM) was included as a negative control. Transfections were performed in transwells (24-wells) coated with Matrigel. In an Eppendorf tube, 250 µL of Opti-MEM reduced serum medium (Gibco ThermoFisher) and 6μL Lipofectamine RNAiMAX transfection reagent (ThermoFisher) were combined and incubated for 5 min at room temperature. Following this, 2.5 μL of siRNA per well was added to the solution and incubated for 20 min. Five hundred microlitres of the above final solution was then added to each well (100 nM siRNA concentration, from a stock solution of 40 µM) and cells were incubated for 48 h. All transfections were performed with 10,000 cells seeded on Matrigel and cultured for 48 h. Transfection medium was then replaced with respective regular cell culture medium prior to immunofluorescence and/or invasion assay. Each experiment was carried out in triplicate and repeated multiple times.
Analysis of serum DBP concentrations
Human serum samples from pregnant women were obtained from two sources. The first set of samples were obtained as part of previous studies of maternal serum and placental/ decidual concentrations of vitamin D metabolites and DBP in first-, second- and third- trimester pregnancies (Ethics: 14/WM/1146, obtained from West Midlands – Edgbaston Research Ethics Committee) (Tamblyn et al. 2017). For these samples, serum, placental and decidual human DBP concentrations were previously determined (Tamblyn et al. 2017), and serum samples were used to prepare patient specific JEG3 Matrigel invasion assays. The second set of samples were obtained from University of Cork, Cork, Ireland as part of a study to assess vitamin D metabolite concentrations in serum and urine from pregnant women at first trimester of pregnancy, 50% of whom went on to develop preeclampsia (Clinical Research Ethics Committee of the Cork Teaching Hospital: ECM5(10)05/02/08), amendment 14/WM/1146 - RG_14-194 2 and material transfer agreement 15.04.2016 15-1386)(Tamblyn et al. 2018). Using these samples, an ELISA (Enzyme-linked Immune Sorbent Assay) Kit (K2314, Immundiagnostik, Bensheim was used to quantify serum concentrations of DBP as previously reported (Tamblyn et al. 2017).
Statistics
All experiments were carried out in replicate wells according to the experiment and repeated three to four times with separate cultures as indicated. One-Way ANOVA and t-test (parametric data) were performed with mean and the 95% CI (GraphPad PRISM (Version 8.0, La Jolla, CA). A P-value of <0.05 was considered statistically significant.
Results
DBP and VDR are present in trophoblasts cultured on Matrigel
VDR and DBP were detectable in JEG3, BeWo and HTR8 trophoblasts, as well as TPC thyroid carcinomas cells (Fig. 1A). VDR expression was significantly lower in TPC cells compared to trophoblastic cells (Fig. 1B), but increased in the cells with 1,25D. Both trophoblast and TPC cells showed significant induction of DBP expression with 1,25D (Fig. 1B).
Expression of VDR and DBP in trophoblasts and thyroid cells. (A). Expression of protein for the vitamin D receptor (VDR, pink) and vitamin D binding protein (DBP, red) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM, 48 h). Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Data for total corrected cell fluorescence of VDR and DBP protein expression (mean ± 95% CI) are shown for duplicate images from n = 3–4 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Expression of VDR and DBP in trophoblasts and thyroid cells. (A). Expression of protein for the vitamin D receptor (VDR, pink) and vitamin D binding protein (DBP, red) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM, 48 h). Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Data for total corrected cell fluorescence of VDR and DBP protein expression (mean ± 95% CI) are shown for duplicate images from n = 3–4 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Expression of VDR and DBP in trophoblasts and thyroid cells. (A). Expression of protein for the vitamin D receptor (VDR, pink) and vitamin D binding protein (DBP, red) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM, 48 h). Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Data for total corrected cell fluorescence of VDR and DBP protein expression (mean ± 95% CI) are shown for duplicate images from n = 3–4 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
1,25D promotes trophoblast matrix invasion
Although trophoblasts expressed VDR, mRNA for the VDR-target gene CYP24A1 was undetectable in JEG3, BeWo and HTR8 cells even in the presence of 1,25D (Fig. 2A). By contrast, vehicle-treated TPC cells showed low CYP24A1 mRNA, which increased dramatically with 1,25D. TPC cells also showed a significant antiproliferative response to 1,25D, whereas JEG3, BeWo and HTR8 cells showed no response (Fig. 2B). For JEG3, BeWo and HTR8 cell Matrigel invasion increased significantly following treatment with 1,25D, whilst 1,25D inhibited matrix invasion by TPC cells (Fig. 2C). Pro-invasive effects of 1,25D on trophoblast cells were associated with increased expression of MMP2 and decreased expression of its inhibitor, TIMP1. By contrast, the anti-invasion effect of 1,25D (100 nM) on TPC cells was associated with decreased MMP2 and increased TIMP1 expression (Fig. 2D).
Effects of 1,25D on trophoblasts and TPC cells cultured on Matrigel. (A). Expression of mRNA (1/δCt) for CYP24A1, and (B). Cell proliferation (BrdU incorporation, absorbance units) (A) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). (C). Cell matrix invasion (cell number/field of vision) by JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM and 100 nM, 48 h). (D). Expression of mRNA for matrix metalloproteinase 2 (MMP2) and tissue-inhibitor of matrix metalloproteinase 1 (TIMP1) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for mRNA expression are mean ± 95% CI 1/δCt value for duplicate or single analyses from n = 3 separate experiments. Data for cell invasion and cell proliferation assays are mean ± 95% CI, for triplicate or quadruplicate analyses from 3 to 5 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effects of 1,25D on trophoblasts and TPC cells cultured on Matrigel. (A). Expression of mRNA (1/δCt) for CYP24A1, and (B). Cell proliferation (BrdU incorporation, absorbance units) (A) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). (C). Cell matrix invasion (cell number/field of vision) by JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM and 100 nM, 48 h). (D). Expression of mRNA for matrix metalloproteinase 2 (MMP2) and tissue-inhibitor of matrix metalloproteinase 1 (TIMP1) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for mRNA expression are mean ± 95% CI 1/δCt value for duplicate or single analyses from n = 3 separate experiments. Data for cell invasion and cell proliferation assays are mean ± 95% CI, for triplicate or quadruplicate analyses from 3 to 5 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effects of 1,25D on trophoblasts and TPC cells cultured on Matrigel. (A). Expression of mRNA (1/δCt) for CYP24A1, and (B). Cell proliferation (BrdU incorporation, absorbance units) (A) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). (C). Cell matrix invasion (cell number/field of vision) by JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (10 nM and 100 nM, 48 h). (D). Expression of mRNA for matrix metalloproteinase 2 (MMP2) and tissue-inhibitor of matrix metalloproteinase 1 (TIMP1) in JEG3, BeWo, HTR8 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for mRNA expression are mean ± 95% CI 1/δCt value for duplicate or single analyses from n = 3 separate experiments. Data for cell invasion and cell proliferation assays are mean ± 95% CI, for triplicate or quadruplicate analyses from 3 to 5 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Knockdown of VDR protein (Fig. 3A and B) using siRNA had no effect on JEG3 Matrigel invasion in the presence or absence of 1,25D (Fig. 3C). In TPC cells, knockdown of VDR suppressed Matrigel invasion significantly, but treatment with 1,25D did not suppress invasion in VDR knockdown TPC cells. These data indicate that the stimulation of trophoblast matrix invasion by 1,25D is not dependent on VDR expression.
Effect of VDR knockdown on 1,25D-induced cell matrix invasion. Effect of siRNA knockdown of VDR on (A) VDR mRNA expression in JEG3 cells (B) VDR (pink) and DBP (red) protein expression. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. Effect of siRNA knockdown of VDR on (C) cell matrix invasion (cell number/field of vision) by JEG3 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for cell fluorescence are mean ± 95% CI, for quadruplicate analyses from four separate experiments. Data for matrix invasion fluorescence are mean ± 95% CI, for duplicate analyses from three separate experiments. Statistically different from vehicle-treated control, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effect of VDR knockdown on 1,25D-induced cell matrix invasion. Effect of siRNA knockdown of VDR on (A) VDR mRNA expression in JEG3 cells (B) VDR (pink) and DBP (red) protein expression. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. Effect of siRNA knockdown of VDR on (C) cell matrix invasion (cell number/field of vision) by JEG3 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for cell fluorescence are mean ± 95% CI, for quadruplicate analyses from four separate experiments. Data for matrix invasion fluorescence are mean ± 95% CI, for duplicate analyses from three separate experiments. Statistically different from vehicle-treated control, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effect of VDR knockdown on 1,25D-induced cell matrix invasion. Effect of siRNA knockdown of VDR on (A) VDR mRNA expression in JEG3 cells (B) VDR (pink) and DBP (red) protein expression. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. Effect of siRNA knockdown of VDR on (C) cell matrix invasion (cell number/field of vision) by JEG3 and TPC cells cultured on Matrigel in the presence or absence of 1,25D (100 nM, 48 h). Data for cell fluorescence are mean ± 95% CI, for quadruplicate analyses from four separate experiments. Data for matrix invasion fluorescence are mean ± 95% CI, for duplicate analyses from three separate experiments. Statistically different from vehicle-treated control, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Intracellular expression of DBP is due to uptake of serum DBP
JEG3 cells cultured in medium supplemented with serum from WT (DBP+/+) mice showed low levels of expression of the gene for DBP (GC) and the DBP membrane receptor megalin (LRP3) (Fig. 4A). Expression of mRNA for GC and LRP2 was enhanced significantly following treatment with 1,25D. JEG3 cells cultured in medium supplemented with serum from DBP knockout (DBP−/−) mice showed lower baseline expression of GC and LRP2 relative to cells cultured with DBP+/+ serum. Expression of LRP3 in JEG3 cells cultured with DBP−/− serum was enhanced by treatment with 1,25D in a similar fashion to DBP+/+ cells (Fig. 4A). DBP+/+ serum-cultured JEG3 cells also expressed protein for DBP and megalin, but DBP protein expression was significantly decreased in cells cultured with DBP−/− serum (Fig. 4B and C). By contrast, megalin protein levels increased in DBP−/− JEG3 cells (Fig. 4C). These data suggest that although JEG3 cells express low levels of mRNA for GC and LRP2, the presence of DBP protein in these cells is dependent on uptake of exogenous DBP from serum.
Effect of serum DBP and megalin function on intracellular DBP and trophoblast function. (A). Effect of WT (DBP+/+) and DBP knockout (DBP−/−) mouse serum on DBP (red) and megalin (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Total corrected cell fluorescence for DBP and megalin protein expression (C). Matrigel invasion. (D). Expression of mRNA for MMP2, TIMP1, VDR, DBP and megalin (LRP2) in JEG3 cells cultured in medium with DBP+/+ or DBP−/− serum in the presence or absence of 1,25D (100 nM, 48 h). E, F and G. DBP and megalin immunofluorescence, and Matrigel invasion, in JEG3 cells cultured in FBS-supplemented medium in the absence or presence of the megalin inhibitor RAP (1 µM). Data for immunofluorescence are the mean ± 95% CI for duplicate analyses from n = 4 separate experiments. Data for matrix invasion are the mean ± 95% CI for duplicate or single analyses from n = 3 separate experiments. Data for mRNA expression are the mean ± 95% CI for single analyses from n = 3 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effect of serum DBP and megalin function on intracellular DBP and trophoblast function. (A). Effect of WT (DBP+/+) and DBP knockout (DBP−/−) mouse serum on DBP (red) and megalin (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Total corrected cell fluorescence for DBP and megalin protein expression (C). Matrigel invasion. (D). Expression of mRNA for MMP2, TIMP1, VDR, DBP and megalin (LRP2) in JEG3 cells cultured in medium with DBP+/+ or DBP−/− serum in the presence or absence of 1,25D (100 nM, 48 h). E, F and G. DBP and megalin immunofluorescence, and Matrigel invasion, in JEG3 cells cultured in FBS-supplemented medium in the absence or presence of the megalin inhibitor RAP (1 µM). Data for immunofluorescence are the mean ± 95% CI for duplicate analyses from n = 4 separate experiments. Data for matrix invasion are the mean ± 95% CI for duplicate or single analyses from n = 3 separate experiments. Data for mRNA expression are the mean ± 95% CI for single analyses from n = 3 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Effect of serum DBP and megalin function on intracellular DBP and trophoblast function. (A). Effect of WT (DBP+/+) and DBP knockout (DBP−/−) mouse serum on DBP (red) and megalin (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (B). Total corrected cell fluorescence for DBP and megalin protein expression (C). Matrigel invasion. (D). Expression of mRNA for MMP2, TIMP1, VDR, DBP and megalin (LRP2) in JEG3 cells cultured in medium with DBP+/+ or DBP−/− serum in the presence or absence of 1,25D (100 nM, 48 h). E, F and G. DBP and megalin immunofluorescence, and Matrigel invasion, in JEG3 cells cultured in FBS-supplemented medium in the absence or presence of the megalin inhibitor RAP (1 µM). Data for immunofluorescence are the mean ± 95% CI for duplicate analyses from n = 4 separate experiments. Data for matrix invasion are the mean ± 95% CI for duplicate or single analyses from n = 3 separate experiments. Data for mRNA expression are the mean ± 95% CI for single analyses from n = 3 separate experiments. Statistically different from vehicle-treated control, * P < 0.05, ** P < 0.01, *** P < 0.001, **** P < 0.0001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
JEG3 cells cultured in medium with serum from DBP−/− mice showed significantly lower matrix invasion than cells cultured with DBP+/+ serum (P = 0.0007). Unlike DBP+/+ cells, cells cultured in DBP−/− serum showed no enhanced invasion response when treated with 1,25D (Fig. 4D). Decreased invasion by JEG3 cells cultured in DBP−/− medium was characterised by decreased expression of mRNA for matrix metalloproteinase 2 (MMP2) and, unlike DBP+/+ cultures, DBP−/− cells showed no MMP2 response to 1,25D (Fig. 4E). Conversely, DBP−/− cultures of JEG3 cells showed higher levels of mRNA for tissue inhibitor of metalloproteinase-1 (TIMP1) than DBP+/+ cells. Both DBP+/+ and DBP−/− cultures showed suppressed TIMP1 in the presence of 1,25D.
JEG3 cells treated with the megalin inhibitor RAP showed decreased expression of DBP (Fig. 4F and G). Although RAP acts to inhibit endocytic internalisation of megalin-DBP, it also suppresses cellular expression of megalin (Fig. 4F and G), consistent with previously reported studies (Birn et al. 2000). JEG3 cells incubated with the megalin inhibitor RAP also showed significantly lower levels of Matrigel invasion relative to vehicle-treated cells (Fig. 4H). These data indicate that inhibition of cellular uptake of DBP via megalin profoundly suppresses matrix invasion by JEG3 cells.
Effects of DBP and 1,25D on trophoblast matrix invasion involve ERK phosphorylation
Previous studies have shown that enhanced matrix invasion by trophoblasts following treatment with 1,25D involves intracellular ERK signaling (Kim et al. 2018). JEG3, BeWo and HTR8 trophoblasts showed increased nuclear phosphorylated ERK (pERK) following treatment with 1,25D (Fig. 5A), and cytoplasmic and nuclear pERK were blocked when the cells were incubated with the ERK-inhibitor U0126 in the presence or absence of 1,25D (Fig. 5A and B). U0126 also blocked intracellular uptake of DBP into JEG3 cells but had no effect on VDR expression (Fig. 5C). Co-treatment with 1,25D partially abrogated suppressive effects of U0126 on intracellular DBP in JEG3 cells (Fig. 5C). Similar results were also obtained for BeWo and HTR8 cells (Supplementary Fig. 1). In JEG3, BeWo and HTR8 cells U0126 suppressed Matrigel invasion, and this was unaffected by co-treatment with 1,25D (Fig. 5D). In TPC cells, U0126 had no effect on matrix invasion by TPC cells with or without 1,25D.
ERK kinase activity and responses to 1,25D in trophoblastic and thyroid cells. (A). Western blot analysis of cytoplasmic and nuclear ERK and pERK in JEG3, BeWo and TPC cells treated with or without 1,25D (100 nM, 48 h) or ERK kinase inhibitor U0126 (1 µM, 48 h). (B). Effect of ERK kinase inhibitor U0126 on pERK in JEG3 cells cultured on Matrigel with or without 1,25D (100 nM, 48 h). (C). Effect of ERK kinase inhibitor U0126 on DBP (red) and VDR (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (D). Matrigel invasion by JEG3, BeWo, HTR8 and TPC cells treated with 1,25D (100 nM, 48 h) in the presence or absence of the ERK kinase inhibitor U0126 and in the presence or absence of 1,25D (100 nM, 48 h). Data showing the number of matrix invading cells/field of vision are the mean ± 95% CI for triplicate analyses from three separate experiments. Statistically different from vehicle-treated control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
ERK kinase activity and responses to 1,25D in trophoblastic and thyroid cells. (A). Western blot analysis of cytoplasmic and nuclear ERK and pERK in JEG3, BeWo and TPC cells treated with or without 1,25D (100 nM, 48 h) or ERK kinase inhibitor U0126 (1 µM, 48 h). (B). Effect of ERK kinase inhibitor U0126 on pERK in JEG3 cells cultured on Matrigel with or without 1,25D (100 nM, 48 h). (C). Effect of ERK kinase inhibitor U0126 on DBP (red) and VDR (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (D). Matrigel invasion by JEG3, BeWo, HTR8 and TPC cells treated with 1,25D (100 nM, 48 h) in the presence or absence of the ERK kinase inhibitor U0126 and in the presence or absence of 1,25D (100 nM, 48 h). Data showing the number of matrix invading cells/field of vision are the mean ± 95% CI for triplicate analyses from three separate experiments. Statistically different from vehicle-treated control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
ERK kinase activity and responses to 1,25D in trophoblastic and thyroid cells. (A). Western blot analysis of cytoplasmic and nuclear ERK and pERK in JEG3, BeWo and TPC cells treated with or without 1,25D (100 nM, 48 h) or ERK kinase inhibitor U0126 (1 µM, 48 h). (B). Effect of ERK kinase inhibitor U0126 on pERK in JEG3 cells cultured on Matrigel with or without 1,25D (100 nM, 48 h). (C). Effect of ERK kinase inhibitor U0126 on DBP (red) and VDR (pink) protein expression in Matrigel cultured JEG3 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) and membrane (NaK ATPase, green) markers. (D). Matrigel invasion by JEG3, BeWo, HTR8 and TPC cells treated with 1,25D (100 nM, 48 h) in the presence or absence of the ERK kinase inhibitor U0126 and in the presence or absence of 1,25D (100 nM, 48 h). Data showing the number of matrix invading cells/field of vision are the mean ± 95% CI for triplicate analyses from three separate experiments. Statistically different from vehicle-treated control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Intracellular DBP in trophoblasts acts to regulate the accumulation of nuclear G-actin
DBP binds vitamin D metabolites with higher affinity but it is also a potent scavenger of G-actin (Delanghe et al. 2015). Immunofluorescence analysis of F-actin and G-actin in JEG3 and HTR8 cells cultured in medium supplemented with either DBP+/+ or DBP−/− serum showed different patterns of intracellular actin. In the presence of extracellular DBP (DBP+/+ serum) cells showed only low levels of G-actin but this increased significantly in cells cultured without DBP (DBP−/− serum) (Fig. 6A and B). Lack of DBP was also associated with decreased cellular F-actin in JEG3 and HTR8 cells (Fig. 6B). In the absence of DBP, the ratio of G-actin/F-actin increased from 0.62 to 2.19 in JEG3 cells and 0.18 to 0.60 in HTR8 cells (Fig. 6C).
G-actin, F-actin and megalin concentration with respect to serum DBP level and presence of serum 1,25D (100 nM, 48 h). (A). Effects of WT mice (DBP +/+) and knockout mice (DBP −/−) serum on expression level of F-actin (green), G-actin (red) and DBP (yellow) in JEG3 and HTR8 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) marker. (B). Data for total corrected cell fluorescence of F-actin and G-actin protein expression (mean ± 95% CI) for images from (A), with n = 3 separate experiments and showing multiple replicates for each experiment. (C). Ratio of total corrected cell fluorescence of G-actin and F-actin protein expression (mean ± 95% CI) for JEG3 and HTR8 data from (B). Statistically different from DBP+/+ control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
G-actin, F-actin and megalin concentration with respect to serum DBP level and presence of serum 1,25D (100 nM, 48 h). (A). Effects of WT mice (DBP +/+) and knockout mice (DBP −/−) serum on expression level of F-actin (green), G-actin (red) and DBP (yellow) in JEG3 and HTR8 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) marker. (B). Data for total corrected cell fluorescence of F-actin and G-actin protein expression (mean ± 95% CI) for images from (A), with n = 3 separate experiments and showing multiple replicates for each experiment. (C). Ratio of total corrected cell fluorescence of G-actin and F-actin protein expression (mean ± 95% CI) for JEG3 and HTR8 data from (B). Statistically different from DBP+/+ control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
G-actin, F-actin and megalin concentration with respect to serum DBP level and presence of serum 1,25D (100 nM, 48 h). (A). Effects of WT mice (DBP +/+) and knockout mice (DBP −/−) serum on expression level of F-actin (green), G-actin (red) and DBP (yellow) in JEG3 and HTR8 cells. Immunofluorescence for each protein is shown in combination with nuclear (Hoechst, blue) marker. (B). Data for total corrected cell fluorescence of F-actin and G-actin protein expression (mean ± 95% CI) for images from (A), with n = 3 separate experiments and showing multiple replicates for each experiment. (C). Ratio of total corrected cell fluorescence of G-actin and F-actin protein expression (mean ± 95% CI) for JEG3 and HTR8 data from (B). Statistically different from DBP+/+ control, ** P < 0.01, *** P < 0.001.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Serum concentrations of DBP and 1,25D define matrix invasion by trophoblasts
Serum concentrations of maternal 1,25D and DBP increase during pregnancy but vary considerably within cohorts (Tamblyn et al. 2017). To assess the impact of these two factors on the trophoblast matrix invasion, serum from 14 women in the first trimester of pregnancy was used to generate individual JEG3 Matrigel invasion cultures. Data in Fig. 7A showed that serum DBP concentrations correlate significantly with Matrigel invasion by JEG3 cells. There was also a trend for correlation between invasion and serum levels of 1,25D (Fig. 7B), but no correlation with serum 25D (Fig. 7C). When normalised to serum levels of 1,25D, DBP concentrations showed an even stronger correlation with trophoblast invasion (Fig. 7D), but no similar effect was observed when DBP was normalised to 25D concentrations (Fig. 7E).
These data indicate that serum levels of both DBP and 1,25D can influence trophoblast matrix invasion. To determine possible clinical implications of this observation, concentrations of DBP were analysed in serum samples from a cohort of first-trimester pregnancies in which 50% of women went on to have normal healthy deliveries, whilst 50% went on to develop the hypertensive disorder preeclampsia. Data in Fig. 7F showed that women with healthy pregnancies had significantly higher serum DBP (mean: 869.5 ng/mL, 95% CI: 812.7–919.1) than women who developed preeclampsia (mean: 691.4 ng/mL, 95% CI: 647.2–735.6).
DBP from pregnancy serum samples defines matrix invasion by JEG3 cells, and is decreased in women who later develop preeclampsia. (A) Correlation of serum DBP, 1,25D, and 25D with Matrigel invasion by JEG3 cells cultured in medium supplemented with pregnancy serum samples. (B) Correlation of DBP adjusted for 1,25D or 25D with invasion of Matrigel by JEG3 cells for n = 14 pregnancy serum samples. R and p value are shown for each graph. (C) Concentration of DBP in first trimester serum samples from women who went on to have normal healthy or preeclampsia pregnancies (n = 20 samples in each group).
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
DBP from pregnancy serum samples defines matrix invasion by JEG3 cells, and is decreased in women who later develop preeclampsia. (A) Correlation of serum DBP, 1,25D, and 25D with Matrigel invasion by JEG3 cells cultured in medium supplemented with pregnancy serum samples. (B) Correlation of DBP adjusted for 1,25D or 25D with invasion of Matrigel by JEG3 cells for n = 14 pregnancy serum samples. R and p value are shown for each graph. (C) Concentration of DBP in first trimester serum samples from women who went on to have normal healthy or preeclampsia pregnancies (n = 20 samples in each group).
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
DBP from pregnancy serum samples defines matrix invasion by JEG3 cells, and is decreased in women who later develop preeclampsia. (A) Correlation of serum DBP, 1,25D, and 25D with Matrigel invasion by JEG3 cells cultured in medium supplemented with pregnancy serum samples. (B) Correlation of DBP adjusted for 1,25D or 25D with invasion of Matrigel by JEG3 cells for n = 14 pregnancy serum samples. R and p value are shown for each graph. (C) Concentration of DBP in first trimester serum samples from women who went on to have normal healthy or preeclampsia pregnancies (n = 20 samples in each group).
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Discussion
We have shown previously that placental levels of DBP correlate directly with maternal circulating DBP across gestation (Tamblyn et al. 2017). In the current study, we show that although trophoblast cells express low levels of GC mRNA, the presence of DBP protein in these cells is primarily due to megalin-mediated endocytic uptake. Megalin and its co-receptor cubilin have been shown to be expressed by trophoblastic tissues within the placenta (Akour et al. 2013), by primary cultures of trophoblasts (Longtine et al. 2017), and by trophoblast cell lines cultured on Matrigel (Akour et al. 2015). Interestingly, the preliminary analysis of trophoblastic cells cultured using conventional plasticware indicates that these cells do not exhibit intracellular DBP (Supplementary Fig. 2). Thus, interaction with matrix components may be a key factor in the cellular acquisition of DBP, presumably via enhanced expression of megalin. Although the promiscuous nature of megalin-mediated endocytosis means that it is involved in the placental transport of a wide range of potential ligands (Akour et al. 2013), its role in DBP uptake by placenta cells is still not clear. Here we show that inhibition of DBP uptake by either ablation of DBP in serum, or inhibition of megalin suppressed Matrigel invasion by JEG3 cells, highlighting an entirely new function for DBP as an intracellular regulator of cell invasion.
Trophoblastic uptake of DBP may facilitate the cellular movement and metabolism of vitamin D within the placenta. The transfer of 25D and 1,25D from mother to fetus is thought to occur by passive diffusion of these lipid-soluble molecules across the placenta (Ryan & Kovacs 2020). However, the presence of megalin in placental tissues (Ma et al. 2012, Burke et al. 2013) suggests that the transport of vitamin D metabolites across the placenta may be facilitated by binding of DBP and its cargo to megalin. In the current study, we have highlighted an additional potential consequence of placental uptake of DBP. In the circulation serum, DBP also functions as a potent actin-binder (Otterbein et al. 2002) protecting against tissue damage due to systemic F-actin accumulation (Gomme & Bertolini 2004). In recent studies using alpha-cells of the Islets of Langerhans we have shown that the cytoplasmic DBP also participates in intracellular actin homeostasis, with concomitant effects on glucagon secretion (Viloria et al. 2020). We, therefore, postulated that DBP in trophoblast cells interacts with intracellular actin in a similar fashion. The cellular actions of G- and F-actin are complex, and actin homeostasis plays a crucial role in regulating cell differentiation and function (Skruber et al. 2018). A reduced G-/F-actin ratio has been associated with increased matrix invasion by trophoblast giant cells (Chakraborty & Ain 2018). In the current study we show that in the absence of DBP there is a four to five-fold increase in the G-/F-actin ratio for JEG3 and HTR8 cells, and this is associated with decreased matrix invasion by these cells. Our data also suggest that the increased intracellular G-actin and decreased matrix invasion in the absence of DBP specifically reflect increased nuclear G-actin expression. Nuclear actin is known to regulate cell differentiation and function (Misu et al. 2017), but it remains to be determined if this plays a role in trophoblast cell biology and, in particular, matrix invasion. It is also possible that DBP acts to modulate actin polymerisation to F-actin, and decreased actin polymerisation has been shown to impair trophoblast cell matrix invasion (Liang et al. 2019).
DBP may play a pivotal role in coordinating the intracellular functions of both G- and F-actin in trophoblast cells, but this activity appears to be distinct from intracellular DBP in other cell types. In islets of Langerhans, the presence of DBP is due to alpha-cell-specific expression of the DBP gene (GC), rather than cellular uptake of circulating DBP. Using islets isolated from WT (DBP+/+) and GC knockout (DBP−/−) mice we showed that loss of intracellular DBP was associated with decreased alpha-cell size and glucagon release, with indirect effects on beta-cell insulin release (Viloria et al. 2020). The ratio of G-/F-actin is known to be important for secretory function of islet cells (Kalwat & Thurmond 2013), and DBP−/− alpha-cells showed a shift from monomeric G-actin expression to increased F-actin (Viloria et al. 2020). Thus in alpha-cells, intracellular DBP appears to function by limiting DBP for polymerisation to F-actin, while in trophoblasts DBP appears to limit nuclear uptake of G-actin. This dichotomy of function may reflect the endogenous nature of DBP and its specific secretory function of alpha-cells but, nevertheless, underlines the importance of DBP for maintenance of both systemic and intracellular actin homeostasis. This is further illustrated by previously reported studies of hepatic stellate cells which do not express DBP/GC but acquire DBP in a megalin-dependent fashion from hepatocytes which express and secrete DBP (Gressner et al. 2008). After internalisation by hepatic stellate cells, DBP acts to bind intracellular actin with concomitant effects on transdifferentiation of these cells into myofibroblasts (Gressner et al. 2008).
Although DBP plays a pivotal role in defining the matrix invasion by trophoblastic cells, our data also indicate a role for DBP’s cargo. Active vitamin D, 1,25D, promoted trophoblast matrix invasion consistent with previous studies of trophoblast cells (Chan et al. 2015; Kim et al. 2018). This effect was independent of the nuclear VDR and trophoblast cells did not exhibit classical 1,25D-VDR responses such as induction of CYP24A1, but instead promoted non-nuclear signaling via induction of pERK. This mechanism is required for trophoblast responses to 1,25D, but also appears to play a fundamental role in promoting trophoblast invasion in general. Inhibition of pERK dramatically suppressed matrix invasion by all three trophoblast cell lines in the presence or absence of 1,25D, and this was associated with complete suppression of intracellular DBP and elevation of nuclear G-actin. Thus, megalin-mediated uptake of DBP by trophoblasts appears to be dependent on ERK phosphorylation, with increased pERK following treatment with 1,25D acting to further enhance matrix invasion. By contrast, in thyroid carcinoma TPC cells 1,25D suppressed matrix invasion, and these cells also demonstrated classical nuclear responses to 1,25D, similar to those described for 1,25D and other tumor cell lines (Bao et al. 2006). Thus, the action of 1,25D in promoting matrix invasion by trophoblast cells is distinct from more established cellular anti-proliferative/anti-invasion effects of vitamin D, and acts to amplify the actions of cellular DBP uptake.
Data in Figs 4 and 5 suggest that DBP and its ligand 1,25D act in a coordinated fashion to optimise trophoblast invasion. To test this hypothesis, we used serum from healthy first-trimester pregnancies to assess matrix invasion capacity for individual pregnant women. The observation that serum DBP levels alone are sufficient to define the magnitude of JEG3 matrix invasion underlined the importance of DBP as a determinant of healthy placenta development. We have shown previously that placental levels of DBP are directly proportional to maternal serum concentrations of DBP (Tamblyn et al. 2017). Data presented here suggest that this, in turn, plays a key role in directing trophoblast function. Although maternal serum 1,25D showed only a trend towards enhanced matrix invasion by JEG3 cells, adjustment of DBP concentrations to account for 1,25D resulted in a stronger correlation with invasion than for DBP alone. Collectively these data endorse a mechanistic model in which 1,25D acts to promote DBP uptake and enhance trophoblast invasion. Nevertheless, in first-trimester serum samples DBP concentration alone was sufficient to discriminate between women who went on to healthy pregnancies and those who developed preeclampsia. This is in stark contrast to measurement of vitamin D metabolites such as 25D and 1,25D which showed no difference between these two populations of women (Tamblyn et al. 2018). It is also important to recognise that in serum invasion experiments serum 25D had no impact on matrix invasion by JEG3 cells, even when used to adjust DBP levels. Serum 25D levels are used almost exclusively as the marker of vitamin D ‘status’ for studies of human health. Based on data presented in the current study, we propose that this approach is an oversimplification of the biological action of vitamin D with both DBP and active 1,25D coordinating important molecular, cellular and clinical actions of vitamin D.
Data presented here suggest an entirely new paradigm for vitamin D and placental function in which the serum vitamin D carrier DBP plays a pivotal role in trophoblast matrix invasion (Fig. 8). The fact that trophoblasts acquire exogenous DBP from using megalin-mediated endocytosis provides a mechanism by which circulating maternal levels of DBP can influence fetal trophoblast function. Serum DBP concentrations are increased in pregnant vs non-pregnant women but the function of this is unclear. We propose that lower serum levels of DBP during pregnancy may impair matrix invasion by fetal trophoblasts. Consistent with this we show that circulating first trimester levels of DBP are lower in women who go on to develop the malplacentation disorder preeclampsia. This is supported by recent studies of women with type 1 diabetes who develop preeclampsia, where decreased serum levels of DBP were also observed (Kelly et al. 2020). In this report the lower circulating levels of DBP were assessed in the context of free and bound vitamin D metabolite levels, but direct actions of DBP may also occur in these women. Thus, circulating DBP may be a novel marker of preeclampsia risk. However, use of DBP as a systemic actin scavenger has also been proposed as strategy for the prevention of endothelial injury associated with bone marrow transplantation (Luebbering et al. 2020). It is therefore interesting to speculate that restoration of low serum DBP in pregnant women may provide a new approach for the management of disorder of placentation such as preeclampsia.
Schematic representation of the actions of DBP and 1,25D on trophoblast invasion. DBP and 1,25D cooperate to promote matrix invasion by trophoblasts. Inhibition of this cooperative mechanism by ablation of serum DBP, inhibition of ERK kinase or inhibition of DBP-megalin endocytosis (X) increases cellular G-/F-actin ratio and decreases matrix invasion.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Schematic representation of the actions of DBP and 1,25D on trophoblast invasion. DBP and 1,25D cooperate to promote matrix invasion by trophoblasts. Inhibition of this cooperative mechanism by ablation of serum DBP, inhibition of ERK kinase or inhibition of DBP-megalin endocytosis (X) increases cellular G-/F-actin ratio and decreases matrix invasion.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Schematic representation of the actions of DBP and 1,25D on trophoblast invasion. DBP and 1,25D cooperate to promote matrix invasion by trophoblasts. Inhibition of this cooperative mechanism by ablation of serum DBP, inhibition of ERK kinase or inhibition of DBP-megalin endocytosis (X) increases cellular G-/F-actin ratio and decreases matrix invasion.
Citation: Journal of Endocrinology 249, 1; 10.1530/JOE-20-0626
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/JOE-200626.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.
Acknowledgements
The authors would like to thank Prof Louise Kenny (University College Cork) for facilitating access to the SCOPE samples. The authors also like to acknowledge Katrina Viloria (University of Birmingham) for helping with immunostaining analyses.
References
Akour AA, Kennedy MJ & Gerk P 2013 Receptor-mediated endocytosis across human placenta: emphasis on megalin. Molecular Pharmaceutics 10 1269–1278. (https://doi.org/10.1021/mp300609c)
Akour AA, Kennedy MJ & Gerk PM 2015 The role of megalin in the transport of gentamicin Across BeWo cells, an in vitro model of the human placenta. AAPS Journal 17 1193–1199. (https://doi.org/10.1208/s12248-015-9778-9)
Baca KM, Govil M, Zmuda JM, Simhan HN, Marazita ML & Bodnar LM 2018 Vitamin D metabolic loci and preeclampsia risk in multi-ethnic pregnant women. Physiological Reports 6 e13468. (https://doi.org/10.14814/phy2.13468)
Bao BY, Yeh SD & Lee YF 2006 1alpha,25-dihydroxyvitamin D3 inhibits prostate cancer cell invasion via modulation of selective proteases. Carcinogenesis 27 32–42. (https://doi.org/10.1093/carcin/bgi170)
Benis KA & Schneider GB 1996 The effects of vitamin D binding protein-macrophage activating factor and colony-stimulating factor-1 on hematopoietic cells in normal and osteopetrotic rats. Blood 88 2898–2905. (https://doi.org/10.1182/blood.V88.8.2898.bloodjournal8882898)
Birn H, Vorum H, Verroust PJ, Moestrup SK & Christensen EI 2000 Receptor-associated protein is important for normal processing of megalin in kidney proximal tubules. Journal of the American Society of Nephrology 11 191–202.
Bodnar LM, Catov JM, Simhan HN, Holick MF, Powers RW & Roberts JM 2007 Maternal vitamin D deficiency increases the risk of preeclampsia. Journal of Clinical Endocrinology & Metabolism 92 3517–3522. (https://doi.org/10.1210/jc.2007-0718)
Burke KA, Jauniaux E, Burton GJ & Cindrova-Davies T 2013 Expression and immunolocalisation of the endocytic receptors megalin and cubilin in the human yolk sac and placenta across gestation. Placenta 34 1105–1109. (https://doi.org/10.1016/j.placenta.2013.08.003)
Calvo M & Ena JM 1989 Relations between vitamin D and fatty acid binding properties of vitamin D-binding protein. Biochemical & Biophysical Research Communications 163 14–17. (https://doi.org/10.1016/0006-291x(8992091-3)
Chakraborty S & Ain R 2018 NOSTRIN: A novel modulator of trophoblast giant cell differentiation. Stem Cell Research 31 135–146. (https://doi.org/10.1016/j.scr.2018.07.023)
Chan SY, Susarla R, Canovas D, Vasilopoulou E, Ohizua O, McCabe CJ, Hewison M & Kilby MD 2015 Vitamin D promotes human extravillous trophoblast invasion in vitro. Placenta 36 403–409. (https://doi.org/10.1016/j.placenta.2014.12.021)
Chun RF 2012 New perspectives on the vitamin D binding protein. Cell Biochemistry & Function 30 445–456. (https://doi.org/10.1002/cbf.2835)
Delanghe JR, Speeckaert R & Speeckaert MM 2015 Behind the scenes of vitamin D binding protein: more than vitamin D binding. Best Practice & Research. Clinical Endocrinology & Metabolism 29 773–786. (https://doi.org/10.1016/j.beem.2015.06.006)
Gomme PT & Bertolini J 2004 Therapeutic potential of vitamin D-binding protein. Trends in Biotechnology 22 340–345. (https://doi.org/10.1016/j.tibtech.2004.05.001)
Gressner OA, Lahme B & Gressner AM 2008 Gc-globulin (vitamin D binding protein) is synthesized and secreted by hepatocytes and internalized by hepatic stellate cells through Ca(2+)-dependent interaction with the megalin/gp330 receptor. Clinica Chimica Acta: International Journal of Clinical Chemistry 390 28–37. (https://doi.org/10.1016/j.cca.2007.12.011)
Jorgensen CS, Christiansen M, Norgaard-Pedersen B, Ostergaard E, Schiodt FV, Laursen I & Houen G 2004 Gc globulin (vitamin D-binding protein) levels: an inhibition ELISA assay for determination of the total concentration of Gc globulin in plasma and serum. Scandinavian Journal of Clinical & Laboratory Investigation 64 157–166. (https://doi.org/10.1080/00365510410001149)
Kalwat MA & Thurmond DC 2013 Signaling mechanisms of glucose-induced F-actin remodeling in pancreatic islet beta cells. Experimental & Molecular Medicine 45 e37. (https://doi.org/10.1038/emm.2013.73)
Kelly CB, Wagner CL, Shary JR, Leyva MJ, Yu JY, Jenkins AJ, Nankervis AJ, Hanssen KF, Garg SK & Scardo JA et al.2020 Vitamin D metabolites and binding protein predict preeclampsia in women with Type 1 diabetes. Nutrients 12 2048. (https://doi.org/10.3390/nu12072048)
Kew RR 2019 The vitamin D binding protein and inflammatory injury: a mediator or sentinel of tissue damage? Frontiers in Endocrinology 10 470. (https://doi.org/10.3389/fendo.2019.00470)
Kim RH, Ryu BJ, Lee KM, Han JW & Lee SK 2018 Vitamin D facilitates trophoblast invasion through induction of epithelial-mesenchymal transition. American Journal of Reproductive Immunology 79 e12796. (https://doi.org/10.1111/aji.12796)
Kumar R, Cohen WR, Silva P & Epstein FH 1979 Elevated 1,25-dihydroxyvitamin D plasma levels in normal human pregnancy and lactation. Journal of Clinical Investigation 63 342–344. (https://doi.org/10.1172/JCI109308)
Liang X, Jin Y, Wang H, Meng X, Tan Z, Huang T & Fan S 2019 Transgelin 2 is required for embryo implantation by promoting actin polymerization. FASEB Journal 33 5667–5675. (https://doi.org/10.1096/fj.201802158RRR)
Longtine MS, Cvitic S, Colvin BN, Chen B, Desoye G & Nelson DM 2017 Calcitriol regulates immune genes CD14 and CD180 to modulate LPS responses in human trophoblasts. Reproduction 154 735–744. (https://doi.org/10.1530/REP-17-0183)
Luebbering N, Abdullah S, Lounder D, Lane A, Dole N, Rubinstein J, Hewison M, Gloude N, Jodele S & Perentesis KMR et al.2020 Endothelial injury, F-actin and vitamin-D binding protein after hematopoietic stem cell transplant and association with clinical outcomes. Haematologica 10 3324. (https://doi.org/10.3324/haematol.2019.233478)
Lundgren S, Carling T, Hjalm G, Juhlin C, Rastad J, Pihlgren U, Rask L, Akerstrom G & Hellman P 1997 Tissue distribution of human gp330/megalin, a putative Ca(2+)-sensing protein. Journal of Histochemistry & Cytochemistry 45 383–392. (https://doi.org/10.1177/002215549704500306)
Ma R, Gu Y, Zhao S, Sun J, Groome LJ & Wang Y 2012 Expressions of vitamin D metabolic components VDBP, CYP2R1, CYP27B1, CYP24A1, and VDR in placentas from normal and preeclamptic pregnancies. American Journal of Physiology. Endocrinology & Metabolism 303 E928–E935. (https://doi.org/10.1152/ajpendo.00279.2012)
Misu S, Takebayashi M & Miyamoto K 2017 Nuclear actin in development and transcriptional reprogramming. Frontiers in Genetics 8 27. (https://doi.org/10.3389/fgene.2017.00027)
Naidoo Y, Moodley J, Ramsuran V & Naicker T 2019 Polymorphisms within vitamin D binding protein gene within a preeclamptic South African population. Hypertension in Pregnancy 38 260–267. (https://doi.org/10.1080/10641955.2019.1667383)
Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI & Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96 507–515. (https://doi.org/10.1016/s0092-8674(0080655-8)
Otterbein LR, Cosio C, Graceffa P & Dominguez R 2002 Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system. Proceedings of the National Academy of Sciences of the United States of America 99 8003–8008. (https://doi.org/10.1073/pnas.122126299)
Ryan BA & Kovacs CS 2020 Maternal and fetal vitamin D and their roles in mineral homeostasis and fetal bone development. Journal of Endocrinological Investigation 44 643–659. (https://doi.org/10.1007/s40618-020-01387-2).
Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA & Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. Journal of Clinical Investigation 103 239–251. (https://doi.org/10.1172/JCI5244)
Skruber K, Read TA & Vitriol EA 2018 Reconsidering an active role for G-actin in cytoskeletal regulation. Journal of Cell Science 131. (https://doi.org/10.1242/jcs.203760)
Tamblyn JA, Jenkinson C, Larner DP, Hewison M & Kilby MD 2018 Serum and urine vitamin D metabolite analysis in early preeclampsia. Endocrine Connections 7 199–210. (https://doi.org/10.1530/EC-17-0308)
Tamblyn JA, Susarla R, Jenkinson C, Jeffery LE, Ohizua O, Chun RF, Chan SY, Kilby MD & Hewison M 2017 Dysregulation of maternal and placental vitamin D metabolism in preeclampsia. Placenta 50 70–77. (https://doi.org/10.1016/j.placenta.2016.12.019)
Viloria K, Nasteska D, Briant LJB, Heising S, Larner DP, Fine NHF, Ashford FB, da Silva Xavier G, Ramos MJ & Hasib A et al.2020 Vitamin-D-binding protein contributes to the maintenance of alpha cell function and glucagon secretion. Cell Reports 31 107761. (https://doi.org/10.1016/j.celrep.2020.107761)
Wang TJ, Zhang F, Richards JB, Kestenbaum B, van Meurs JB, Berry D, Kiel DP, Streeten EA, Ohlsson C & Koller DL et al.2010 Common genetic determinants of vitamin D insufficiency: a genome-wide association study. Lancet 376 180–188. (https://doi.org/10.1016/S0140-6736(1060588-0)
Wei SQ, Qi HP, Luo ZC & Fraser WD 2013 Maternal vitamin D status and adverse pregnancy outcomes: a systematic review and meta-analysis. Journal of Maternal-Fetal & Neonatal Medicine 26 889–899. (https://doi.org/10.3109/14767058.2013.765849)
Zehnder D, Bland R, Walker EA, Bradwell AR, Howie AJ, Hewison M & Stewart PM 1999 Expression of 25-hydroxyvitamin D3-1alpha-hydroxylase in the human kidney. Journal of the American Society of Nephrology 10 2465–2473.
Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM & Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D(3)-1 Alpha-hydroxylase. Journal of Clinical Endocrinology & Metabolism 86 888–894. (https://doi.org/10.1210/jcem.86.2.7220)
