TRPM7 regulates vascular endothelial cell adhesion and tube formation

Abstract

Transient receptor potential melastatin 7 (TRPM7) is a nonselective cation channel with an α-kinase domain in its COOH terminal, known to play a role in diverse physiological and pathological processes such as Mg²⁺ homeostasis, cell proliferation, and hypoxic neuronal injury. Increasing evidence suggests that TRPM7 contributes to the physiology/pathology of vascular systems. For example, we recently demonstrated that silencing TRPM7 promotes growth and proliferation and protects against hyperglycemia-induced injury in human umbilical vein endothelial cells (HUVECs). Here we investigated the potential effects of TRPM7 on morphology, adhesion, migration, and tube formation of vascular endothelial cells and the potential underlying mechanism. We showed that inhibition of TRPM7 function in HUVECs by silencing TRPM7 decreases the density of TRPM7-like current and cell surface area and inhibits cell adhesion to Matrigel. Silencing TRPM7 also promotes cell migration, wound healing, and tube formation. Further studies showed that the extracellular signal-regulated kinase (ERK) pathway is involved in the change of cell morphology and the increase in HUVEC migration induced by TRPM7 silencing. We also demonstrated that silencing TRPM7 enhances the phosphorylation of myosin light chain (MLC) in HUVECs, which might be involved in the enhancement of cell contractility and motility. Collectively, our data suggest that the TRPM7 channel negatively regulates the function of vascular endothelial cells. Further studies on the underlying mechanism may facilitate the development of the TRPM7 channel as a target for the therapeutic intervention of vascular diseases.

vascular endothelial cells play an important role in the regulation of vascular tone, vascular permeability, angiogenesis, and vascular inflammatory response (12, 53). Those cells express a variety of ion channels, including transient receptor potential (TRP) channels that constitute a large family including canonical (TRPC), vanilloid (TRPV), and melastatin (TRPM) subfamilies (73). At least 19 members among the 28 mammalian TRP channel isoforms have been identified in vascular endothelial cells (38, 73). These channels appear to be involved in diverse physiological/pathological functions.

Transient receptor potential melastatin 7 (TRPM7), a member of TRPM subfamily, is a Ca²⁺ and Mg²⁺-permeant channel protein in possession of its own kinase domain (49, 57–59). It has been reported to be expressed in many types of cells including tumor cells, neurons, hematopoietic cells, granulocytes, leukocytes, and microglia (22, 27, 33, 45, 57, 63). As one of two unique channel-kinases along with TRPM6, TRPM7 not only possesses the ion permeability but also phosphorylates biomolecules such as annexin I, myosin IIA heavy chain, and m-calpain (14, 16, 66). Recent studies have demonstrated that TRPM7 plays an important role in a large number of physiological and pathophysiological processes including cell growth/proliferation, adhesion and migration, motility and movements, oxidative stress, and anoxic neuronal death (1, 13, 15, 30, 40, 56, 63, 66).

TRPM7 has been shown to be expressed in the human vascular system and to play a role in vascular functions (3, 4, 6, 25, 31, 52, 68, 69, 74, 76). For example, in vascular smooth muscle cells, TRPM7 modulates Mg²⁺ homeostasis (6, 25). In vascular endothelial cells, TRPM7 modulates cell growth and proliferation (4, 31). It was also reported that the level of TRPM7 expression in endothelial cells can be upregulated by high glucose, oxidative stress, and laminar shear stress (3, 68, 69). All of these studies suggest that TRPM7 might be involved in cardiovascular diseases involving endothelial dysfunction. However, the role of TRPM7 in adhesion and angiogenic activity of vascular endothelial cells remains to be elucidated. Here we investigated the effect of TRPM7 on these processes in HUVECs and explored the potential underlying mechanisms.

MATERIALS AND METHODS

Reagents and antibodies.

Anti-β-actin antibody was purchased from Abcam (Cambridge, MA). Anti-Orai1, anti-stromal interaction molecule 1 (anti-STIM1), and anti-TRPC3 antibodies were from Alamone Laboratories (Jerusalem, Israel). Anti-phospho-MLC, anti-total MLC, anti-phospho-MEK1/2, anti-total MEK1/2 antibodies, and U0126 were purchased from Cell Signaling (Beverly, MA). Growth factor-reduced Matrigel was purchased from BD Biosciences (Bedford, MA). Protease inhibitors cocktail were purchased from Sigma (St. Louis, MO). The phosphatase inhibitor cocktail was from Roche (Indianapolis, IN).

Cell culture.

HUVECs were purchased from ATCC and grown in EGM-2 medium containing 2% fetal bovine serum (FBS) and trophic factors (Lonza, Walkersville, MD), as described in our previous studies (31, 32). In some experiments, HUVECs were cultured in endothelial basal medium (EBM) containing 10% or 2% FBS without trophic factors.

Human embryonic kidney (HEK-293) cells, with inducible expression of human TRPM7 channels (HEK:TRPM7 cells), were cultured in DMEM supplemented with 10% FBS (30). For the induction of TRPM7, the cells were treated with 1 μg/ml of tetracycline.

RNA interference.

Knockdown of TRPM7 experiments were performed as described (31, 68). Briefly, dsRNAs 406–426 (TRPM7 siRNA1) and 450–470 (TRPM7 siRNA2) of TRPM7 (NM_017672.4) were synthesized from Invitrogen. Cells were transfected with 50 nM siRNA using transfection reagent lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's instructions. Nontargeting siRNA (Invitrogen) was used as a negative control.

Real-time PCR.

Total RNAs were extracted with RNA purification kit (Qiagen, Valencia, CA) and transcribed to cDNA using the Superscript First-Strand Synthesis System (Invitrogen). Real-time PCR was performed using SYBR Green supermix (Bio-Rad, Richmond, CA) in a C1000 Thermal cycler (Bio-Rad) as described previously (72): 1) 95°C for 10 min, 2) 95°C for 10 s, 3) 61°C for 10 s, 4) 72°C for 15 s, plate read, 5) go to 2, 40 more times, 6) 95°C for 10 s, and 7) melt curve 65–95°C: increment 0.5°C for 5 s, plate read. Primer sets were as described in Table 1.

Table 1.

Primer sets sequence

Target	Primers Sequence	Predicted Size, bp
TRPM7		214
S	5′-CTTATGAAGAGGCAGGTCATGG
AS	5′-CATCTTGTCTGAAGGACTG
GAPDH		200
S	5′-AACTGCTTAGCACCCCTGGC
AS	5′-ATGACCTTGCCCACAGCCTT
TRPC3		201
S	5′-CAAGAATGACTATCGGAAGC
AS	5′-GCCACAAACTTTTTGACTTC
Orai1		238
S	5′-AGGTGATGAGCCTCAACGAG
AS	5′-CTGATCATGAGCGCAAACAG
STIM1		135
S	5′-TGGGATCTCAGAGGGATTTG
AS	5′-CATTGGAAGTCATGGCATTG

TRPM7, transient receptor potential melastatin 7; TRPC3, canonical transient receptor potential 3; STIM1, stromal interaction molecule 1; S, sense primers; AS, antisense primers.

Electrophysiology.

Whole cell voltage-clamp recordings were performed as described previously (31). Three to four days after transfection, cells were set on the stage of an inverted microscope (Ti-S; Nikon) and superfused at room temperature with an extracellular solution containing the following (in mM): 140 NaCl, 5.4 KCl, 1 CaCl₂, 10 glucose, and 20 HEPES (pH 7.4 with NaOH; 320–335 mosM). Patch electrodes were fabricated from borosilicate capillary tubing of a 1.5-mm diameter (WPI) using a vertical puller (PP-83; Narishige). The electrode resistance ranged from 4 to 5 MΩ when filled with the intracellular solution (see below).

For the current-voltage (I–V) relationship, voltage steps ranged between −100 and +100 mV from a holding potential of −60 mV were applied. All recordings were performed at least 5 min after the whole cell configuration was established. Membrane currents were recorded using an Axopatch 200B amplifier. Data were filtered at 2 kHz and digitized at 5 kHz by using a Digidata 1440A data-acquisition system and pClamp 10 software (Axon Instruments). Pipette solution contained the following (in mM): 140 Cs-methanesulfonate, 8 NaCl, 4.1 CaCl₂, 10 EGTA, 5 tetraethyl-ammonium chloride, 2 Na₂-ATP, and 10 HEPES (pH 7.3, with CsOH).

Mg²⁺ imaging.

The intracellular Mg²⁺ levels were examined using Mag-Fura-2 (Invitrogen). Cells were incubated with 5 μM Mag-Fura-2-AM for 30 min at 37°C in Ca²⁺-free extracellular solution, which contained the following (in mM): 140 NaCl, 5.4 KCl, 1 MgCl₂, 10 glucose, and 20 HEPES (pH 7.4 with NaOH, 320–335 mosM), followed by de-esterification of the dye for another 30 min at room temperature. The coverslips containing dye-loaded cells were held in a recording chamber placed on the stage of an inverted microscope (Eclipse TE2000-U; Nikon). Mag-Fura-2 was excited at wavelengths of 340 and 380 nm. Fluorescence was detected with a ×40 objective lens (Super Fluor ×40, numerical aperture = 0.90; Nikon) and a CCD camera (Cool- SNAP ES2; Photometrics). Digitized images were acquired, stored, and analyzed in a personal computer controlled by Axon Imaging Workbench software (INDEC Biosystems). The shutter and filter wheel (Lambda 10–3; Sutter Instruments) were also controlled by the software to allow timed illumination of cells at either 340- or 380-nm excitation wavelengths. Fluorescence was detected at an emission wavelength of 510 nm. Ratio images of 340/380 were analyzed by averaging pixel ratio values in circumscribed regions of randomly selected 7–10 cells within the field of view.

Adhesion assay.

Ninety-six-well plates were coated with Matrigel (1 mg/ml; BD Biosciences) overnight at 4°C and washed with sterilized H₂O. HUVECs were harvested after transfection with siRNA for 48 h and seeded into 96-well plates (2–4 × 10⁴/well). After incubation for 1 h at 37°C, the nonadherent cells were washed away with PBS. Adhered cells were fixed with 4% paraformaldehyde and stained with 0.5% crystal violet for 10 min. Three to five random fields were captured as representative images using light microscope (×4) connected with Nikon camera (D90). Crystal violet was extracted with 10% acetic acid, and optical density was measured at 595 nm.

Transwell assay.

Transwell assays were performed using a modified Boyden chamber (8-μm pores; Corning 3422). After being transfected with siRNA for 48 h and starved for 24 h, HUVECs (3–4 × 10⁵ cells/ml) in 100 μl of serum-free medium were plated in the upper chamber, whereas 670 μl of medium containing 10% FBS were added to the lower chamber. After incubation for ∼18 h in an incubator, cells that did not migrate through the pores were scraped off thoroughly with Q-Tips. Attached cells were fixed with 4% paraformaldehyde for 10 min and stained with 0.5% crystal violet/methanol for 10 min. Pictures of six to nine randomly picked fields for each group were taken by Nikon invert light microscope (×10), and the migrated cells were counted. Cells under different treatments were normalized to the control value and expressed as fold change in migration.

Wound-healing assay.

HUVECs transfected with siRNA for 48 h were seeded to 24-well plates at a density of 1 × 10⁵ cells/well. After adhesion to the plate, cells were starved with serum-free medium for 24 h and the cell monolayer was scratched with 1,000-μl tip. After being washed three times with serum-free medium, cells were incubated with EBM-2 medium containing 2% FBS for 24 h at 37°C. Pictures were taken from the field of wound area with Nikon camera under light microscope (×4) and analyzed using National Institutes of Health (NIH) ImageJ software. The wound healing rate was calculated as follows: %wound closure = [(area of original wound − area of wound after healing)/area of original wound] × 100%.

Tube formation assay.

Tube formation on Matrigel was evaluated as a model of in vitro angiogenesis. Briefly, 60 μl of EBM-2 containing 5 mg/ml Matrigel were added into 96-well plates and incubated at 37°C for 30 min. HUVECs were harvested after tranfection with siRNA for 48–72 h and then seeded into the 96-well plates precoated with Matrigel at a density of 3 × 10⁴/well. Capillary tube structures were observed, and the representative images were captured with an inverted microscope (×4) equipped with Nikon camera. Tube length was quantified using NIH ImageJ software.

Immunoblotting.

Immunoblotting was performed as described previously (30, 31). Briefly, cells were lysed in RIPA buffer (50 mM Tris·HCl pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, protease inhibitor, and phosphatase inhibitor cocktail). After centrifugation at 13,000 g at 4°C for 10 min, the lysates were collected. The aliquots were mixed with Laemmli sample buffer and boiled at 95°C for 10 min. Proteins were separated by 10% SDS-PAGE, transfered to PVDF membranes, and probed with antibodies against phospho-MLC (1:500), total-MLC (1:500), β-actin (1:2,000), TRPC3 (1:200), Orai1 (1:200), STIM1 (1:500), total MEK1/2 (1:250), and phospho-MEK1/2 (1: 1,000) followed by horseradish-conjugated secondary antibodies (1:2,000). The signals were visualized by chemiluminescence using an ECL kit (Millipore).

Statistical analysis.

All data are expressed as means ± SE. Statistical analyses were done by Student t-test or one-way ANOVA on data collected from at least three independent experiments. Dunnett's test was used to compensate for multiple experimental procedures. The differences were considered statistically significant when P < 0.05.

RESULTS

To investigate the effect of TRPM7 on HUVEC cytoskeleton reorganization, adhesion, migration, and tube formation, specific siRNAs targeting nucleotides 406–426 and 450–470 of human TRPM7 (NM_017672.4) were synthesized according to our previous study (31). Data of quantitative real-time PCR (Fig. 1A) showed that the level of TRPM7 mRNA in HUVECs transfected with TRPM7 siRNAs for 72 h was decreased compared with cells transfected with nontargeting control siRNA (TRPM7 siRNA1; 39.67 ± 12.44% and TRPM7 siRNA2; 29.49 ± 2.74% of control; n = 3–6; P < 0.05).

Fig. 1.

Silencing transient receptor potential melastatin 7 (TRPM7) by specific siRNA in human umbilical vein endothelial cells (HUVECs). A: after TRPM7 siRNA or control siRNA was transfected to HUVECs for 72 h, total RNA was extracted and reverse transcribed to cDNA. Quantitative real-time PCR was performed to monitor mRNA level of TRPM7 in HUVECs. **P < 0.01 control vs. TRPM7 siRNA-treated cells (n = 3–6). B: representative currents elicited by voltage steps ranging from −100 to +100 mV in control siRNA- or TRPM7 siRNA1-transfected cells. C: current-voltage (I–V) relationship in control siRNA- or TRPM7 siRNA-transfected cells. D: densities of TRPM7-like current in cells transfected with control or TRPM7 siRNA cells. *P < 0.05, control siRNA- vs. TRPM7 siRNA1-treated cells (n = 18–32). E: bar graph showing intracellular Mg²⁺ levels evaluated at basal condition (0 mM Ca²⁺/1 mM Mg²⁺) by measuring the 340/380 nm ratio of Mag-Fura-2. HEK:TRPM7 cells were either untreated or treated with tetracycline (tet) for 48 h. HUVECs were treated with either control or TRPM7 siRNA1 for 72 h; n = 29–76 cells from 3–9 fields in 3–4 independent experiments. *P < 0.05 vs. control.

Next, we measured TRPM7-like currents to investigate the functional change of TRPM7 channels. TRPM7-like current is typically outwardly rectifying in the presence of divalent cations (49) and is also sensitive to the intracellular Mg²⁺ level (49, 63). HUVECs kept in a bath solution containing 1 mM Ca²⁺ were perfused with a pipette solution where no Mg²⁺ was included and where Na₂-ATP was added to chelate Mg²⁺ for maximizing TRPM7-like currents (49, 57, 63). Under these conditions, HUVECs transfected with control siRNA showed larger outward rectification compared with those transfected with TRPM7 siRNA (Fig. 1, B and C). Figure 1D shows the statistical comparison of the densities of TRPM7-like currents in control siRNA or TRPM7 siRNA-transfected cells. The current densities of TRPM7 siRNA-treated HUVECs were significantly smaller compared with those of control cells (control siRNA; 3.31 ± 0.38 pA/pF at +80 mV, n = 32 and TRPM7 siRNA1; 2.17 ± 0.27 pA/pF at +80 mV, n = 18; P < 0.05).

TRPM7 is Mg²⁺ permeable and plays a critical role for cellular Mg²⁺ homeostasis in some cells (25, 60, 63). To further reveal the functional significance of TRPM7 channels in HUVECs, intracellular Mg²⁺ levels were measured using the fluorescent Mg²⁺ indicator Mag-Fura-2. As a positive control, HEK-293 cells overexpressing TRPM7 showed a higher ratio of emissions following excitation at 340 and 380 nm of Mag-Fura-2 (Fig. 1E), indicating the potential accumulation of Mg²⁺ via TRPM7 channels. We carried out the same experiment in HUVECs. However, the clear difference of Mg²⁺ levels between control and TRPM7 siRNAs was not detected, consistent with previous studies performed in knockout mice (34, 35, 61) (Fig. 1E).

Interestingly, knockdown of TRPM7 produced a dramatic change in the shape of HUVECs with a constricted cell morphology characterized by exaggerated membrane filamentous extension whereas cells transfected with the control siRNA displayed normal cell morphology (Fig. 2A, top). Quantification of the change in cell morphology by a measurement of cell-surface area showed that the average single cell surface area was decreased by ∼40% following the transfection with TRPM7 siRNAs (control siRNA, 886.7 ± 21.3 μm²; TRPM7 siRNA1, 387.5 ± 18.0 μm²; TRPM7 siRNA2, 389.9 ± 18.5 μm²; n = 120–270 cells from 3 to 8 independent experiments; P < 0.01; Fig. 2A, bottom). This finding is consistent with a previous report that depletion of TRPM7 alters the morphology of fibroblasts (67).

Fig. 2.

Silencing TRPM7 alters HUVEC morphology and adhesion to extracellular matrix. A: HUVECs were transfected with control siRNA or TRPM7 siRNA for 48–72 h and the representative images were captured to show the cell shape of HUVECs. Arrows indicate shape change of HUVECs with membrane filamentous extension (top right). Quantification of cell morphology change was calculated by measuring the average single cell surface area for 150 - 270 cells (bottom). Data were from at least 3–8 independent experiments. B: HUVECs transfected with siRNAs for 48–72 h were seeded to 96-well plates coated with Matrigel (1 mg/ml). Adhered cells were stained with crystal violet and the representative images were captured. Data were analyzed by measuring optical density of crystal violet extracted by 10% acetic acid at 595 nm (n = 3–5). *P < 0.05, **P < 0.01, control vs. TRPM7 siRNA1- or siRNA2-treated cells. Scale bar = 50 μm.

Next, we studied the effect of TRPM7 knockdown on adhesion of HUVECs to extracellular matrix (Matrigel). As shown in Fig. 2B, there was 45.67 and 29.8% reduction in the number of cells attached to Matrigel in HUVECs treated with TRPM7 siRNA1 (n = 5) and TRPM7 siRNA2 (n = 3), respectively, compared with cells treated with control siRNA (P < 0.05), suggesting that silencing TRPM7 decreases the ability of HUVECs to adhere to the extracellular matrix, which might be related to the change of cell morphology induced by TRPM7 silencing.

Cell motility and migration are important physiological processes that depend deeply on the cytoskeleton reorganization. It is anticipated that morphology change of HUVECs will alter its motility and migration ability. We therefore investigated the effect of TRPM7 on the migration of HUVECs. Transwell assay demonstrated that the number of migrated cells in TRPM7 siRNA1 and siRNA2 group were 2.78 ± 0.55- and 1.66 ± 0.14-fold of the control siRNA group (n = 7 for TRPM7 siRNA1, n = 4 for TRPM7 siRNA2; P < 0.05; Fig. 3A). To further verify the role of TRPM7 in cell migration, a wound healing assay was performed. As shown in Fig. 3B, silencing TRPM7 with siRNA1 and siRNA2 significantly increased the wound healing rate of HUVECs (56.6 ± 25.4% increase for TRPM7 siRNA1 and 34.9 ± 21.3% increase for TRPM7 siRNA2, n = 4–5; P < 0.05; Fig. 3B).

Fig. 3.

Silencing TRPM7 enhances HUVEC migration and tube formation. A: transwell assay was used for HUVEC migration. Representative images were captured and the migrated cells were counted. Data were from 4 independent experiments. Scale bar = 50 μm. *P < 0.05, control vs. TRPM7 siRNA1 or siRNA2-treated cells. B: wound healing assay was also used to measure HUVEC migration. The pictures of the same wound area were captured to show the wound area closure (scale bar = 250 μm). Wound healing rate was calculated by measuring closure area using NIH ImageJ software (n = 4–5). *P < 0.05, control vs. TRPM7 siRNA1 or siRNA2-treated cells. C: HUVECs were harvested after transfection with siRNAs for 48–72 h, seeded into 96-well plates (3 × 10⁴/well) precoated with Matrigel (5 mg/ml) and observed for 24 h. Representative images show the tube formation in control and TRPM7 siRNA group at 4-h time point (top). Tube length was quantified by NIH ImageJ software according to the pictures of tube formation (n = 3). Scale bar = 250 μm.*P < 0.05, control vs. TRPM7 siRNA1- or siRNA2-treated cells.

Angiogenesis is an important function of vascular endothelial cells. It has been shown that early angiogenesis in vitro depends deeply on cell migration and cytoskeleton reorganization (4, 39). We, therefore, investigated the effect of TRPM7 on HUVEC tube formation, a common method to study angiogenesis in vitro. At day 3 after TRPM7 siRNA transfection, HUVECs were seeded on Matrigel to induce tube formation. As shown in Fig. 3C, silencing TRPM7 with siRNA1 and siRNA2 increased the tube length by 39.5 ± 6.7 and 60 ± 24.3%, respectively, after 4 h of tube formation (n = 3; P < 0.05; Fig. 3C). However, there was no significant difference in the tube length at the 16–24 h of tube formation (data not shown), suggesting that TRPM7 may play an important role in tube formation, especially in the early stage.

Since TRPM7 regulates the proliferation of HUVECs via the MEK/ERK signaling pathway (31), we asked whether the same pathway is involved in the morphological change and the enhanced migration of HUVECs induced by silencing TRPM7. First, we show that phosphorylation of MEK1/2 was enhanced by silencing TRPM7 (Fig. 4A), as reported in our previous study (31). We then investigated the effect of U0126, which is a specific inhibitor of MEK and commonly used agent to study MEK/ERK pathway (18), on TRPM7 siRNA-mediated change in HUVEC morphology and migration. HUVECs were transfected with siRNA in the presence of 10 μM U0126. Culture medium containing fresh U0126 was exchanged every 24 h, and DMSO was used as vehicle control. As shown in Fig. 4B, addition of U0126 significantly inhibited cell morphology change induced by knockdown of TRPM7 (TRPM7 siRNA1, DMSO vs. U0126: 427.5 vs. 943.2 μm²; TRPM7 siRNA2, DMSO vs. U0126: 586.4 vs. 1,193.2 μm²; n = 150 cells from 4–10 independent experiments; P < 0.01). Moreover, addition of U0126 also prevented the increase in cell migration induced by TRPM7 siRNA1 and siRNA2 (60.2 ± 10.8% inhibition in TRPM7 siRNA1 group and 68.3 ± 3.8% inhibition in TRPM7 siRNA2 group; n = 3–4; P < 0.05; Fig. 4C). These findings suggest that TRPM7 regulates the morphology and migration of HUVECs, at least in part, via the ERK signaling pathway.

Fig. 4.

ERK signaling pathway was involved in morphology change and increased migration induced by silencing TRPM7. A: Western blotting was used to show MEK1/2 phosphorylation after transfection of control siRNA, TRPM7 siRNA1, or TRPM7 siRNA2 for 72 h (n = 4). B: morphology change of HUVECs treated with or without U0126 (a MEK inhibitor, 10 μM) for 72 h was observed. The representative images were captured, and the average single cell surface area for 150 cells was quantified. Data were from 4–10 independent experiments. C: migration of HUVECs treated with or without U0126 was investigated in transwell assay and the representative images were captured. Relative migration rate was calculated by normalization to control siRNA group without U0126 (n = 3–4). Scale bar = 100 μm. **P < 0.01, control vs. TRPM7 siRNA1- or siRNA2-treated cells. N.S., not significant vs. control. ##P < 0.01, DMSO vs. U0126.

Previous studies have shown that store-operated calcium entry (SOCE) system including Orai1 and STIM1 plays important roles in morphological change and angiogenesis in vascular endothelial cells (2, 41, 46). Also, this system is important for migration of various cell types including vascular smooth muscle cells (17, 54, 62). Therefore, we investigated the effect of silencing TRPM7 on these molecules. Real-time PCR did not detect significant changes in the level of Orai1 and STIM1 mRNA (Fig. 5A). Consistent with mRNA levels, Western blotting analysis also showed no changes in Orai1 and STIM1 proteins (Fig. 5B).

Fig. 5.

Silencing TRPM7 does not change the expression of Orai1, stromal interaction molecule 1 (STIM1), and canonical TRP 3 (TRPC3). HUVECs were transfected with control siRNA or TRPM7 siRNA for 72 h. A: total RNA was extracted and reverse transcribed to cDNA. Quantitative real-time PCR was performed to monitor mRNA level of the indicated genes in HUVECs (n = 3). B: Western blotting was used to show protein levels of the indicated genes. Histogram indicates fold change of the indicated proteins normalized to β-actin (n = 3).

In addition to Orail and STM1, TRPC channels also play a role in SOCE (2, 41, 46). Since our previous studies provided some evidence that TRPC1 is not affected by TRPM7 silencing (31), we examined potential changes of TRPC3 and TRPC6. Similar to Orai1 and STIM1, the level of TRPC3 mRNA or protein was not altered by silencing TRPM7 (Fig. 5, A and B), whereas TRPC6 was not detectable in HUVECs in our conditions (data not shown). Thus the SOCE system does not seem to be involved in the effects of TRPM7 silencing observed in the current study.

It is well known that the change of cell morphology and motility is critically dependent on cytoskeleton rearrangement, which is tightly linked to actin and myosin interaction (65). In endothelial cells, actin-myosin interaction is mainly regulated by myosin light chain kinase (MLCK)-mediated MLC phosphorylation (21, 64, 75). We, therefore, examined whether silencing TRPM7 modulates the phosphorylation of MLC. Western blotting showed that knockdown of TRPM7 with siRNA1 and siRNA2 significantly increased MLC phosphorylation by 49.4 ± 13 and 56.4 ± 18%, respectively (n = 6; P < 0.05; Fig. 6A). Previous studies reported that ERK regulates MLC phosphorylation via MLCK in several cell types (23, 24). We, therefore, investigated whether the MEK inhibitor U0126 modulates MLC phosphorylation in HUVECs. Interestingly, incubation of HUVECs with U0126 for 3 days significantly increased MLC phosphorylation by itself (relative p-MLC/t-MLC ratio changed from 1 to 1.11 ± 0.03; Fig. 6B, lanes 1 and 4; n = 5; P < 0.05). However, in the presence of U0126, TRPM7 siRNAs had little effect on the level of MLC phosphorylation (Fig. 6B, lane 4 vs. lanes 5 and 6; n = 5; not significant, P > 0.05).

Fig. 6.

Silencing TRPM7 enhances phosphorylation of myosin light chain (MLC), which involves ERK pathway. A: Western blotting was used to show MLC phosphorylation after treatment with TRPM7 siRNA or control siRNA or untreated for 72 h. The phosphorylation of MLC was normalized to total MLC (n = 6). B: HUVECs transfected with control siRNA or TRPM7 siRNA were treated with U0126 (10 μM) for 72 h. Histogram indicates fold change of MLC phosphorylation normalized to total MLC (n = 5). *P < 0.05, control vs. TRPM7 siRNA1- or siRNA2-treated cells. N.S., not significant vs. control. #P < 0.05, DMSO vs. U0126.

DISCUSSION

We reported previously that TRPM7 is expressed in HUVECs and regulates cell growth and proliferation (31). Recently, we demonstrated that TRPM7 plays an important role in hyperglycemia-induced endothelial cell injury (68). In the current study, we show that TRPM7 knockdown alters morphology, decreases adhesion, and enhances migration and tube formation of HUVECs. We also showed that ERK signaling is associated with at least some of these changes. In addition, we showed the possibility that the MLC pathway is also involved in cell morphology change and migration in HUVECs.

Vascular endothelial cells are known to alter their shape in response to diverse factors such as mechanical and biochemical stimuli, which influence a variety of cellular functions (47). In this study, we observed that knockdown of TRPM7 dramatically alters the morphology of HUVECs: they change from flat and rounded shape to constricted and elongated morphology. This finding is similar to the change of morphology in fibroblasts and HEK-293 cells following TRPM7 knockdown (66, 67). Overexpression of TRPM7, on the other hand, has been shown to change the morphology of HEK-293 cells in the opposite direction (49). Although it was reported that the level of TRPM7 expression increases in HUVECs when exposed to laminar shear stress, which might suggest a link between TRPM7 and endothelial cell shape change and cytoskeleton remodeling (43, 69), to our knowledge, the present study is the first to demonstrate directly that TRPM7 modulates the morphology of vascular endothelial cells. Of note, TRPM7 channels are also known to be involved in the regulation of cellular volume (51). Therefore, volume change mediated through TRPM7 could potentially contribute to the morphological changes we observed.

Adhesion and migration of endothelial cells contribute to diverse physiological processes such as vascular development, wound healing, and angiogenesis (55). We showed that silencing TRPM7 in HUVECs decreased cell adhesion to extracellular matrix, which is consistent with findings in N1E-115 neuroblastoma cells and fibroblasts (13, 67). For example, Clark et al. (13) showed that moderate overexpression of TRPM7 in N1E-115 neuroblastoma cells enhances cell-matrix adhesion, and Su et al. (67) showed that depletion of TRPM7 in fibroblasts disrupts the focal adhesions. Interestingly, in HEK-293 cells, silencing TRPM7 has been shown to strengthen the cell adhesion and increase the number of peripheral adhesion complexes (66). The exact mechanism for TRPM7's involvement in cell adhesion is not clear. As TRPM7 was found in the cell membrane ruffles and podosome ring structure (13), in our view, loss of TRPM7 may disrupt the adhesion complexes of the cell surface, which might be one of the reasons that cell-matrix contact or adhesion was reduced.

Notably, depletion of TRPM7 in HUVECs promoted cell migration both in Transwell assay and in wound healing assay. TRPM7 has been found to regulate cell migration and motility in diverse cell types (10, 20, 44, 67). For example, HEK-293 cells with depleted TRPM7 migrate 56% more efficiently than the control cells in wound healing assay (66). The effect of TRPM7 on endothelial cell migration has also been reported (3, 4). Consistent with our findings, a recent study by Baldoli et al. (3) showed that TRPM7 silencing or inhibition enhances HUVEC migration in wound healing assay. On the contrary, in human microvascular endothelial cells (HMECs), silencing TRPM7 inhibits the cell migration (4). Although the detailed mechanism is not clear, it is likely that TRPM7 has different functions in macrovascular endothelial cells (e.g., HUVECs) and microvascular endothelial cells (e.g., HMECs). As a matter of fact, HUVECs and HMECs have clear differences in basal gene expression profiles (11) and potentially different intracellular signaling machinery. It is worth noting that silencing TRPM7 in HMECs inhibits ERK phosphorylation while the same measure enhances ERK phosphorylation in HUVECs (4, 31). Further studies are required to understand the molecular basis for the different effects of TRPM7 in various endothelial cell types.

Angiogenesis, a process that involves migration, growth, and differentiation of the endothelial cells, plays a critical role in the growth and spread of cancer and in tissue repair after injury (7, 39). Tube formation from endothelial cells cultured on Matrigel is a powerful tool to examine endothelial differentiation and angiogenic activity in vitro. Using this method, we found that silencing TRPM7 in HUVECs promotes tube formation in the early stage (4–6 h). However, this difference disappears at the 16- to 24-h time points (data not shown). As endothelial cell proliferation and migration are essential to angiogenic activity (39), our previous findings that silencing TRPM7 increases HUVEC proliferation and the current finding that silencing TRPM7 promotes cell migration (31) are consistent with its effect on angiogenesis. Potentially, TRPM7 may preferentially affect the early events of angiogenesis in which cell proliferation and migration are the main cellular processes, while the later events of cell reorganization, differentiation, and vascular-like structure formation (48) may be unaffected by TRPM7 silencing. Further studies are required to elucidate the mechanism by which TRPM7 regulates the early event but not the later stage of angiogenesis.

As a channel kinase, TRPM7 is involved in various cell functions by mediating Ca²⁺ and Mg²⁺ influx and phosphorylation of substrates. So far, at least three molecules have been identified as substrates of TRPM7, namely, annexin I, m-calpain, and myosin II heavy chain (13, 14, 16, 66). In HEK-293 cells, TRPM7 can regulate cell adhesion through m-calpain activation mediated by Ca²⁺ influx (66). In N1E-115 cells, TRPM7 activation results in a kinase-dependent remodeling of the actomyosin cytoskeleton, leading to the assembly of podosomes (13). Studies from our group and others suggested that MEK/ERK pathway is involved in TRPM7-mediated endothelial cell proliferation (4, 31). Since MEK/ERK pathway is involved in cell cytoskeleton remodeling, migration, and angiogenesis (26, 28, 50), we speculated that the ERK pathway might be involved in HUVEC morphology change and increased migration induced by silencing TRPM7. Indeed, our results showed that U0126, a specific inhibitor of MEK, significantly inhibited the change in cell morphology and the enhanced cell migration induced by TRPM7 silencing. Thus the MEK/ERK pathway is involved, at least in part, in TRPM7-mediated endothelial cell remodeling and migration. However, at the same time, U0126 alone is sufficient to cause ∼50% reduction in migration of HUVECs compared with control (shown in Fig. 4C). This suggests that basal ERK activity plays an important role in the migration of HUVECs. Given that ERK activity can facilitate cellular migration with various mechanisms (29, 36), we could think about the possibility that the ERK pathways under basal and after TRPM7 silencing have different downstream effectors and that they may be parallel pathways.

It is well known that cell morphology change and migration depend on actin-myosin contraction. In endothelial cells, actin-myosin contraction is mainly mediated by MLCK-catalyzed MLC phosphorylation. We found that MLC phosphorylation is enhanced by silencing TRPM7, which may well explain the change in HUVEC morphology. Since the MLCK/MLC pathway is regulated by ERK pathway (23, 24), we determined whether the effect of TRPM7 silencing on MLC phosphorylation is affected by ERK inhibition. Surprisingly, we found that, in the presence of U0126, silencing TRPM7 no longer increases the phosphorylation of MLC in HUVECs. The detailed mechanism is not clear. Although a number of studies have indicated that ERK activation facilitates MLC phosphorylation (5, 19, 37), some studies have shown that ERK signaling does not affect MLC in vascular endothelial cells (42). Considering those facts as well as our current observation, an enhanced phosphorylation of MLC by silencing TRPM7 would indicate that MLCK might be a downstream effector of TRPM7-dependent ERK-signaling pathway that promotes HUVEC migration, while an increased phosphorylation of MLC by U0126 might result from a compensatory change in response to an inhibition of cellular ERK signaling. This could also support the possibility that ERK signaling might have parallel pathways, which have different downstream effectors, as mentioned above.

Overall, our findings suggest that TRPM7 negatively regulates the function of vascular endothelial cells. As a change in cell shape directly affects cell function including proliferation, apoptosis, and inflammation in various cell types (8, 9, 47, 70), the change in the morphology of HUVECs induced by knockdown of TRPM7 suggests that TRPM7 may play an important role in regulating the functions of endothelium under physiological and pathological conditions. Our data suggest that a reduced level of TRPM7 might facilitate the vascular repair in case of damage through enhanced angiogenic activity. Although MEK/ERK pathway might be involved in some of the processes (71, 75), the detailed mechanisms by which TRPM7 regulates endothelial cell function remain to be determined. However, findings from us and others could facilitate the development of TRPM7 channel as a novel target for modulating endothelium function (52). Additional work needs to be done to fully understand the role of TRPM7 in vascular signaling and cardiovascular diseases.

GRANTS

This work was supported by National Institute of Neurological Disorders and Stroke Grants R01-NS-047506 and R01-NS-066027 (to Z.-G. Xiong); Minority Health and Health Disparities Grants S21MD000101 and U54NS083932; American Heart Association Grant 0840132N (to to Z.-G. Xiong); Alzheimer's Association Grant IIRG-10-173350 (to Z.-G. Xiong); China Scholarship Council (to Z. Zeng); Natural Science Foundation of China Grants 81370373 and 81170132 (to L. Zhu); Priority Academic Program Development of Jiangsu Higher Education Institutions of China (to L. Zhu); and Jiangsu Province's Key Discipline of Medicine Grant XK201118 (to L. Zhu).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: Z.Z., K.I., L.Z., and Z.-G.X. conception and design of research; Z.Z., K.I., and H.S. performed experiments; Z.Z., K.I., H.S., and T.L. analyzed data; Z.Z., K.I., T.L., X.F., L.Z., and Z.-G.X. interpreted results of experiments; Z.Z., K.I., and T.L. prepared figures; Z.Z., K.I., and L.Z. drafted manuscript; K.I., L.Z., and Z.-G.X. edited and revised manuscript; L.Z. and Z.-G.X. approved final version of manuscript.