Induction of VEGF-A expression by methylmercury is mediated by the EGFR–p38 MAPK–COX-2–PKA pathway in cultured human brain microvascular pericytes

Abstract

Methylmercury is the causative organometallic compound of Minamata disease. Pathological changes in the cerebrum of patients are localized along the deep sulci and fissures of the cerebral cortex such as the calcarine fissure. It has been suggested that the occurrence of brain edema is important for the cerebral damage caused by methylmercury. Previously, we found that methylmercury induces vascular endothelial growth factor-A in cultured human brain microvascular pericytes, which may increase permeability of the brain microvasculature. In the present study, our findings suggest that this induction is mediated via the epidermal growth factor receptor–p38 mitogen-activated protein kinase–cyclooxygenase-2–protein kinase A pathway in cultured human brain microvascular pericytes. These results partly support our hypothesis that methylmercury causes neurotoxicity by activation of intracellular signaling pathways in various cell types, including neurons, macrophages, vascular endothelial cells, and pericytes.

INTRODUCTION

Methylmercury is a toxic environmental pollutant and the cause of Minamata disease. Pathological studies on Minamata disease have shown local methylmercury-induced damage in the cerebrum. Specifically, the pathological changes are localized along the deep sulci and fissures of the cerebral cortex such as the calcarine fissure (Eto, 1997).

Experimentally, localized damage in the calcarine fissure of the occipital lobe was observed following edematous changes in the white matter of the cerebrum in common marmosets after methylmercury exposure (Eto et al., 2001). Thus, it was suggested that ischemia followed by cortical swelling may influence neuronal cell vulnerability to methylmercury-induced cytotoxicity in cells along the deep cerebral sulci. That is, the occurrence of brain edema is an important event for the cerebral damage caused by methylmercury.

We investigated the mechanisms underlying vasogenic edema caused by methylmercury; vasogenic edema is defined as abnormal fluid accumulation in perivascular tissue due to increased capillary permeability (Iencean, 2003; Marmarou, 2004). We found that methylmercury increased the permeability of brain microvasculature by inducing vascular endothelial growth factor-A (VEGF-A) in pericytes, and placental growth factor and VEGF receptors 1 and 2 in endothelial cells (Hirooka et al., 2013). Methylmercury can also induce the synthesis and secretion of hyaluronan in both pericytes and endothelial cells to maintain fluid in the extracellular matrix of brain microvasculature during the progression of brain edema (Hirooka et al., 2017).

Previously, we found that methylmercury induced cyclooxygenase-2 (COX-2) by activation of the epidermal growth factor receptor (EGFR)–p38 mitogen-activated protein kinase (MAPK) pathway via inhibition of protein tyrosine phosphatase 1B activity in cultured human microvascular endothelial cells (Yoshida et al., 2017). The present study was undertaken to clarify the intracellular signaling pathway that mediates VEGF-A induction in cultured human brain microvascular pericytes.

MATERIALS AND METHODS

Materials

Human brain microvascular pericytes were purchased from DS Pharma Biomedical (Osaka, Japan). HuMedia-SG2 is the growth media for pericytes and was obtained from Kurabo (Osaka, Japan). Dulbecco’s modified Eagle’s medium (DMEM) was purchased from Nissui Pharmaceutical Co., Ltd., (Tokyo, Japan). Collagen-coated tissue culture plates and dishes were purchased from AGC Techno Glass (Shizuoka, Japan). Methylmercury chloride was purchased from Tokyo Chemical Industry (Tokyo, Japan). The enzyme-linked immunosorbent assay (ELISA) kit for human VEGF and the adenylate cyclase inhibitor, SQ22536, were purchased from R&D Systems (Minneapolis, MN, USA). Forskolin, prostaglandin E₁ (PGE₁), and anti-actin antibody (20-33) were purchased from Sigma–Aldrich Chemicals (St. Louis, MO, USA). Immobilon-P polyvinylidene difluoride membrane (0.45 μm) was purchased from Millipore (Billerica, MA, USA). Epidermal growth factor (EGFR) inhibitor, PD153035, was from Calbiochem (Boston, MA, USA). p38 MAPK inhibitor (SB203580), COX-2 inhibitor (NS398), and protein kinase A (PKA) inhibitor (H89) were from Cayman Chemicals (Ann Arbor, MI, USA). Anti-phosphorylated EGFR antibody (3777S), anti-phosphorylated extracellular signal-regulated kinase 1/2 (ERK1/2) antibody (9101S), anti-phosphorylated p38 MAPK antibody (9211S), anti-phosphorylated Jun N-terminal kinase (JNK) antibody (9255S), and horseradish peroxidase-labeled anti-mouse and anti-rabbit IgG antibodies were purchased from Cell Signaling (Beverly, MA, USA). QIAzol lysis reagent and RNeasy Lipid Tissue Mini Kit were purchased from Qiagen (Venlo, Netherlands). High-capacity cDNA Reverse Transcription Kit and Power SYBR Green PCR Master Mix were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Synthesized primers for quantitative real-time polymerase chain reaction (PCR) were purchased from Eurofins Genomics (Ebersberg, Germany). Chemi-Lumi One L and other regents were from Nacalai Tesque (Kyoto, Japan). Methylmercury (10 mM) was stored until use at -80°C in Ca²⁺- and Mg²⁺-free phosphate-buffered saline (PBS) containing L-cysteine.

Cell culture and treatment

Pericytes were transferred into collagen-coated plates or dishes at a density of 5 × 10⁵ cells/cm² and cultured at 37°C in a humid atmosphere of 5% CO₂ in HuMedia-SG2 until confluent (dense cultures). Cells were then treated with methylmercury for 12, 24, or 36 hr in DMEM supplemented with 1% bovine serum albumin with or without forskolin (5 and 10 µM) or PGE₁ (1 and 10 µM). In another experiment, dense cultures of cells were pretreated with PD153035 (10 µM) for 3 hr, SB203580 (50 µM) for 1 hr, NS398 (10 µM) for 1 hr, SQ22536 (50 µM) for 3 hr, or H89 (10 µM) for 3 hr and then challenged with methylmercury (5 µM) for 24 hr.

Determination of VEGF-A

After treatment with methylmercury (1, 2, 3, and 5 µM) in 6-well culture plates, conditioned medium was harvested and centrifuged at 1500 × g for 10 min at 4°C. The VEGF-A content of the supernatant was determined by ELISA. Cell homogenates were harvested in Ca²⁺- and Mg²⁺-free PBS, then the cell layer was sonicated and used for determination of DNA content by a fluorometric method (Setaro and Morley, 1976). Secretion of VEGF-A was expressed as pg/µg DNA.

Western blot analysis

Pericytes were treated with methylmercury (1, 2, 3, and 5 µM), and expression of EGFR, ERK1/2, p38 MAPK, and JNK were determined by western blot analysis, as described previously (Yoshida et al., 2017). Briefly, cells were washed with ice-cold Ca²⁺- and Mg²⁺-free PBS containing 2 mM ethylene glycol-bis(2-aminoethylether)-N,N,N',N'-tetraacetic acid, and then lysed in 50 mM Tris-HCl buffer (pH 6.8) containing 10% glycerol and 2% sodium dodecyl sulfate (SDS). Proteins of lysates were separated by SDS–polyacrylamide gel electrophoresis on 5% (for EGFR) or 10% (for ERK1/2, p38 MAPK, and JNK) polyacrylamide gels, and then transferred onto Immobilon-P membranes. Membranes were blocked for 1 hr in 20 mM Tris-HCl buffer containing 5% skim milk, 150 mM NaCl, and 0.1% Tween 20 (pH 7.5), then incubated with anti-phosphorylated EGFR antibody (1:1000), anti-phosphorylated ERK antibody (1:1000), anti-phosphorylated p38 MAPK antibody (1:1000), anti-phosphorylated JNK antibody (1:1000), or anti-β-actin antibody (1:200). They were then washed and incubated with horseradish peroxidase-conjugated secondary antibodies for 1 hr at room temperature. Immunoreactive bands were visualized using Chemi-Lumi One L and scanned using a LAS 3000 Imager (Fujifilm, Tokyo, Japan).

Real-time reverse transcription-PCR (real-time RT-PCR)

Total RNA was extracted from pericytes using QIAzol lysis reagent and RNeasy Lipid Tissue Mini Kits. cDNA was synthesized from mRNA using a High-capacity cDNA Reverse Transcription Kit. Real-time RT-PCR was performed in triplicate using the Power SYBR Green PCR Master Mix with 5 ng cDNA and 0.1 µM primers on the StepOnePlus Real-time PCR System (Thermo Fisher Scientific). Thermal cycling parameters were 50°C for 2 min, 95°C for 10 min, and 40 cycles of 95°C for 15 sec, then 60°C for 1 min. Melting curve analysis and agarose gel electrophoresis with ethidium bromide staining were used to confirm the presence of a single PCR product of the correct size. Levels of human VEGF-A and β₂-microglobulin mRNA in each RNA sample were quantified by the relative standard curve method. Relative expression levels for each gene were determined after normalization to the β₂-microglobulin intensity value, and expressed as a percentage of the control. The primer sequences were: human VEGF-A, 5′-TTGTTGGAAGAAGCAGCC-3′ (forward) and 5′-GGTTTCAATGGTGTGAGGAC-3′ (reverse); human β₂-microglobulin, 5′-GGGTTTCATCCATCCGACA-3′ (forward) and 5′-GTTCACACGGCAGGCATAC-3′ (reverse).

Statistical analysis

Data were analyzed for statistical significance using analysis of variance (ANOVA) and Bonferroni’s multiple t-test, when possible (Figs. 1A, 1B, and 1D). P values < 0.05 were considered to be statistically significant.

Fig. 1

[A] Secretion of vascular endothelial growth factor-A (VEGF-A) protein into the conditioned media of pericytes (left panel) and expression of VEGF-A mRNA in pericytes (right panel) after exposure to methylmercury. Confluent cultures of human brain microvascular pericytes were incubated at 37°C for 36 hr (left) or 12 hr (right) in the presence of methylmercury (1, 2, 3, and 5 μM). Values represent mean ± S.E. of three biological replicates (left) or the percentage of control ± S.E. of three biological replicates (right). **Significantly different from the corresponding control, p < 0.01. [B] Expression of VEGF-A mRNA in pericytes after exposure to methylmercury in the presence or absence of forskolin (left panel) or prostaglandin E₁ (PGE₁) (right panel). White and black bars indicate the absence and presence of methylmercury, respectively. Confluent cultures of human brain microvascular pericytes were incubated at 37°C for 12 hr in the presence of methylmercury (5 μM) with or without forskolin (5 and 10 µM) or PGE₁ (1 and 10 µM). Values represent mean of the percentage of control (absence of both methylmercury and forskolin/PGE₁) ± S.E. of three biological replicates. **Significantly different from the corresponding control (i.e., absence of methylmercury), p < 0.01. [C] Phosphorylation of epidermal growth factor receptor (EGFR) (left panel) and mitogen-activated protein (MAP) kinases (p38 MAPK, extracellular signal-regulated kinase 1/2 [ERK1/2], and Jun N-terminal kinase [JNK]) (right panel) in pericytes after exposure to methylmercury. Confluent cultures of human brain microvascular pericytes were incubated at 37°C for 12 hr in the presence of methylmercury (1, 2, 3, or 5 μM). [D] Involvement of phosphorylation of EGFR (left panel), p38 MAPK/cyclooxygenase-2 (COX-2) (center panel), and adenylate cyclase/protein kinase A (right panel) in the expression of VEGF-A mRNA in pericytes after exposure to methylmercury. White and black bars indicate the absence and presence of methylmercury, respectively. Confluent cultures of human brain microvascular pericytes were pretreated with EGFR inhibitor, PD153035 (10 µM) for 3 hr (left); p38 MAPK inhibitor, SB203580 (50 µM) for 1 hr or COX-2 inhibitor NS398 (10 µM) for 1 hr (center); and adenylate cyclase inhibitor SQ22536 (50 µM) for 3 hr or a protein kinase A inhibitor H89 (10µM) for 3 hr (right), and then incubated at 37°C for 24 hr in the presence of methylmercury (5 μM). Values represent mean of the percentage of control (i.e., absence of both methylmercury and inhibitors) ± S.E. of three biological replicates. **Significantly different from the corresponding control (i.e., absence of both methylmercury and inhibitors), p < 0.01. ^##Significantly different from the corresponding control (i.e., presence of methylmercury but absence of the inhibitors), p < 0.01.

RESULTS AND DISCUSSION

We first confirmed VEGF-A induction by methylmercury in pericytes, as shown in our previous study (Hirooka et al., 2013). Methylmercury significantly increased the secretion of VEGF-A from cells at 3 µM for more than 36 hr (Fig. 1A, left panel). Levels of VEGF-A mRNA were also increased by methylmercury at 5 µM for 12 hr (Fig. 1A, right panel). These results show again that methylmercury induces VEGF-A expression in pericytes.

Expression of VEGF-A is positively regulated by the cyclic adenosine monophosphate (cAMP) pathway (Namkoong et al., 2009). In fact, methylmercury-induced elevation of VEGF-A mRNA was blocked by forskolin and PGE₁, compounds that increase intracellular cAMP (Insel and Ostrom, 2003; Morrison et al., 1976), completely and partially, respectively, in pericytes (Fig. 1B). This suggests that methylmercury activates the cAMP pathway, which mediates VEGF-A expression in human brain microvascular pericytes.

Methylmercury activates the MAPK pathway in human brain microvascular endothelial cells (Yoshida et al., 2017). Therefore, we hypothesized that this pathway may be involved in activation of the cAMP pathway, which induces VEGF-A expression in pericytes. As shown in Fig. 1C, methylmercury increased phosphorylation of EGFR and its downstream signaling molecule, p38 MAPK (Zwang, and Yarden, 2006). However, increase in ERK1/2 and JNK phosphorylation was not observed after methylmercury exposure. This indicates that methylmercury activates EGFR, which stimulates phosphorylation of p38 MAPK in pericytes, as observed in vascular endothelial cells.

Possible involvement of EGFR and p38 MAPK and their downstream signaling molecules–COX-2 and adenylate cyclase (which produce cAMP) and PKA (which is activated by cAMP)–was investigated using respective inhibitors. As shown in Fig. 1D, PD153035 (EGFR), SB203580 (p38 MAPK), NS398 (COX-2), SQ22536 (adenylate cyclase), and H89 (PKA) significantly reduced the increase of VEGF-A mRNA levels by methylmercury. This suggests that EGFR, p38 MAPK, COX-2, cAMP, and PKA are all involved in the induction of VEGF-A by methylmercury in pericytes.

Taken together, our results indicate that induction of VEGF-A by methylmercury is mediated via activation of the EGFR–p38 MAPK–COX-2–cAMP–PKA pathway in pericytes (Fig. 2). Activation of EGFR is likely caused by inhibition of protein tyrosine phosphatase 1B, which dephosphorylates auto-phosphorylated EGFR and reduces EGFR signaling (Haj et al., 2003). Adenylate cyclase is activated by prostacyclin, which is synthesized by COX-2 (Best et al., 1977) induced by methylmercury through the prostacyclin receptor (IP receptor) (Mitchell et al., 2014; Wang et al., 2024). Activated adenyl cyclase activates PKA, which induces expression of VEGF-A (Ouchi et al., 2005). Finally, VEGF-A can contribute to the progression of brain edema in the cerebrum of Minamata disease (Eto, 1997) through an increase in vascular permeability. The present study partly supports our hypothesis that methylmercury exhibits neurotoxicity via activation of intracellular signaling pathways in various cell types, including neurons (Shinoda et al., 2019), macrophages (Nakano et al., 2024), vascular endothelial cells (Yoshida et al., 2017), and pericytes.

Fig. 2

Intracellular signaling transduction mediates the induction of vascular endothelial growth factor-A (VEGF-A) expression by methylmercury in brain microvascular pericytes. Methylmercury inhibits protein tyrosine phosphatase 1B (PTP1B), as shown previously (Yoshida et al., 2017), activating epidermal growth factor receptor (EGFR). The downstream signaling molecule, p38 mitogen-activated protein kinase (MAPK) is activated and mediates the expression of cyclooxygenase-2 (COX-2), which synthesizes prostacyclin (PGI₂) and activates adenylate cyclase via activation of the prostacyclin receptor (IP receptor). cAMP synthesized by adenylate cyclase increases phosphorylated protein kinase A, thereby mediating VEGF-A expression. Increased VEGF-A increases the permeability of brain microvessels, contributing to severe brain edema that can cause local damage in the cerebrum of patients with Minamata disease.

ACKNOWLEDGMENTS

This work was supported by the Study Group of the Health Effects of Heavy Metals Organized by the Ministry of the Environment, Japan. We would like to thank Editage (www.editage.com) for English language editing. We thank Rachel James, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Best,  L.C.,  Martin,  T.J.,  Russell,  R.G. and  Preston,  F.E. (1977): Prostacyclin increases cyclic AMP levels and adenylate cyclase activity in platelets. Nature, 267, 850-852.
•  Eto,  K. (1997): Pathology of Minamata disease. Toxicol. Pathol., 25, 614-623.
•  Eto,  K.,  Yasutake,  A.,  Kuwana,  T.,  Korogi,  Y.,  Akima,  M.,  Shimozeki,  T.,  Tokunaga,  H. and  Kaneko,  Y. (2001): Methylmercury poisoning in common marmosets--a study of selective vulnerability within the cerebral cortex. Toxicol. Pathol., 29, 565-573.
•  Haj,  F.G.,  Markova,  B.,  Klaman,  L.D.,  Bohmer,  F.D. and  Neel,  B.G. (2003): Regulation of receptor tyrosine kinase signaling by protein tyrosine phosphatase-1B. J. Biol. Chem., 278, 739-744.
•  Hirooka,  T.,  Yamamoto,  C.,  Yasutake,  A.,  Eto,  K. and  Kaji,  T. (2013): Expression of VEGF-related proteins in cultured human brain microvascular endothelial cells and pericytes after exposure to methylmercury. J. Toxicol. Sci., 38, 837-845.
•  Hirooka,  T.,  Yoshida,  E.,  Eto,  K. and  Kaji,  T. (2017): Methylmercury induces hyaluronan synthesis in cultured human brain microvascular endothelial cells and pericytes via different mechanisms. J. Toxicol. Sci., 42, 329-333.
•  Iencean,  S.M. (2003): Brain edema -- a new classification. Med. Hypotheses, 61, 106-109.
•  Insel,  P.A. and  Ostrom,  R.S. (2003): Forskolin as a tool for examining adenylyl cyclase expression, regulation, and G protein signaling. Cell. Mol. Neurobiol., 23, 305-314.
•  Marmarou,  A. (2004): The pathophysiology of brain edema and elevated intracranial pressure. Cleve. Clin. J. Med., 71 (Suppl 1), S6-S8.
•  Mitchell,  J.A.,  Ahmetaj-Shala,  B.,  Kirkby,  N.S.,  Wright,  W.R.,  Mackenzie,  L.S.,  Reed,  D.M. and  Mohamed,  N. (2014): Role of prostacyclin in pulmonary hypertension. Glob. Cardiol. Sci. Pract., 2014, 382-393.
•  Morrison,  A.R.,  Yates,  J. and  Klahr,  S. (1976): Effect of prostaglandin E on the adenyl cyclase-cyclic AMP system and gluconeogenesis in rat renal cortical slices. Biochim. Biophys. Acta, 421, 203-209.
•  Nakano,  T.,  Yoshida,  E.,  Sasaki,  Y.,  Kazama,  S.,  Katami,  F.,  Aoki,  K.,  Fujie,  T.,  Du,  K.,  Hara,  T.,  Yamamoto,  C.,  Takahashi,  T.,  Fujiwara,  Y.,  Eto,  K.,  Iwakura,  Y.,  Shinoda,  Y. and  Kaji,  T. (2024): Mechanisms underlying sensory nerve-predominant damage by methylmercury in the peripheral nervous system. Int. J. Mol. Sci., 25, 11672.
•  Namkoong,  S.,  Kim,  C.K.,  Cho,  Y.L.,  Kim,  J.H.,  Lee,  H.,  Ha,  K.S.,  Choe,  J.,  Kim,  P.H.,  Won,  M.H.,  Kwon,  Y.G.,  Shim,  E.B. and  Kim,  Y.M. (2009): Forskolin increases angiogenesis through the coordinated cross-talk of PKA-dependent VEGF expression and Epac-mediated PI3K/Akt/eNOS signaling. Cell. Signal., 21, 906-915.
•  Ouchi,  N.,  Shibata,  R. and  Walsh,  K. (2005): AMP-activated protein kinase signaling stimulates VEGF expression and angiogenesis in skeletal muscle. Circ. Res., 96, 838-846.
•  Setaro,  F. and  Morley,  C.G. (1976): A modified fluorometric method for the determination of microgram quantities of DNA from cell or tissue cultures. Anal. Biochem., 71, 313-317.
•  Shinoda,  Y.,  Tatsumi,  S.,  Yoshida,  E.,  Takahashi,  T.,  Eto,  K.,  Kaji,  T. and  Fujiwara,  Y. (2019): Gene expression profiles in the dorsal root ganglia of methylmercury-exposed rats. J. Toxicol. Sci., 44, 549-558.
•  Wang,  J.J.,  Jin,  S.,  Zhang,  H.,  Xu,  Y.,  Hu,  W.,  Jiang,  Y.,  Chen,  C.,  Wang,  D.W.,  Xu,  H.E. and  Wu,  C. (2024): Molecular recognition and activation of the prostacyclin receptor by anti-pulmonary arterial hypertension drugs. Sci. Adv., 10, eadk5184.
•  Yoshida,  E.,  Kurita,  M.,  Eto,  K.,  Kumagai,  Y. and  Kaji,  T. (2017): Methylmercury promotes prostacyclin release from cultured human brain microvascular endothelial cells via induction of cyclooxygenase-2 through activation of the EGFR-p38 MAPK pathway by inhibiting protein tyrosine phosphatase 1B activity. Toxicology, 392, 40-46.
•  Zwang,  Y. and  Yarden,  Y. (2006): p38 MAP kinase mediates stress-induced internalization of EGFR: implications for cancer chemotherapy. EMBO J., 25, 4195-4206.