Background
Arterial lesions are mediated by complicated manifestation between vascular endothelial dysfunction and proliferation in vascular smooth muscle cells [
1,
2]. Atherosclerosis is a critical inflammatory arterial disease and is one of the most fatal complications of diabetes, initiated by reactive oxygen species (ROS) induced by the formation of advanced glycation end-products (AGEs) [
3]. During hyperglycemia, AGEs are formed by non-enzymatic reaction with aldehydes and free amino sites and are considered a risk factor [
4]. Excessive formation of intracellular ROS in response to AGEs is reported as a crucial mediator in the development of vascular lesions as well as diabetic cardiomyopathy, nephropathy, retinopathy, and peripheral neurological damage [
5‐
8]. The interaction between AGEs and RAGE (an AGE receptor) activates NAD(P)H oxidase and mitochondria, which generates ROS and induces inflammatory cytokines through multiple signal cascades [
9‐
12]. At the same time as the AGE-RAGE interaction, mitogen-activated protein kinases (MAPKs) including extracellular signal-regulated kinases 1/2 (ERK 1/2), p38 and c-jun N-terminal kinases (JNK) are transduced, resulting in transcription factor activation and the expression of adhesion molecules [
13‐
17]. As a pre-dispositional factor in atherosclerosis, leukocyte-endothelial adhesion is known to initiate endothelial dysfunction, followed by increased expressions of pro-inflammatory cytokines including tumor necrosis factor alpha (TNF-α), interleukin-6 (IL-6) and monocyte chemoattractant protein-1 (MCP-1) [
18‐
20]. Moreover, monocytes transmigrate to the endothelium through molecules such as intercellular adhesion molecules-1 (ICAM-1) and vascular cell adhesion molecules-1 (VCAM-1), and they are ultimately converted to the activated form, known as M1-like macrophages [
21].
In various inflammation-related diseases, nuclear factor-kappaB (NF-κB) transcription factor is highly activated, and it is known as a pivotal inducer of pro-inflammatory cytokines, chemokines and adhesion molecules [
22‐
25]. Regulating NF-κB has been considered an important check-point to overcome inflammatory diseases in recent studies. NF-κB complex is composed of I-kappa B (IκB) and heterodimer, which is composed of p50 and p65 subunits in cytoplasm. Upon activation of the NF-κB complex, IκB is released from the complex, phosphorylated, and then degraded by proteasomes. At the same time, p50 and p65 subunits are translocated to nuclei and bind to transcription sites [
26,
27].
Plantago asiatica is traditionally used as a natural plant medicine in East Asia, and it has been reported to have biological activity including antipyretic, wound healing, anti-cancer, anti-virus and anti-hepatitis properties [
28,
29]. Plantamajoside (PM), a phenylethanoid glycoside compound from
P. asiatica, has been reported to inhibit glycation activity [
30] and to have anti-inflammation and antioxidant properties [
31,
32] as well as nephroprotective effects against heavy metals in an in vivo model [
33]; however, the molecular mechanism of how PM modulates endothelial dysfunction remains uncertain. Therefore, in our study, we investigated the preventive effects of PM on endothelium dysfunction mediated by glyceraldehyde-induced AGEs (glycer-AGEs) using human umbilical vein endothelial cells.
Methods
Plant material and preparing the PM
We obtained the P. asiatica from a traditional market (Kyungdong Herb Market, Seoul, Korea). The PM, which was extracted from P. asiatica, was isolated and identified by Professor B. W. Kang (College of Life Sciences & Biotechnology, Korea University) and obtained through previous research in our laboratory. We deposited the voucher specimens at the Herbarium of Korea University (Register number, H-212).
Briefly, the plant powder was immersed in 100% methanol (MeOH) within a 100 mL/g ratio at 70 °C for 3 h in a reflux condenser. After the dissoluble materials were filtered, the combined filtrate was concentrated with rotary vacuum evaporation and lyophilized for the dried residue. To fractionate the dried
P. asiatica extract, the dried residue was suspended in H
2O and then sequentially fractionated with
n-hexane, chloroform, ethyl acetate (EtOAc) and
n-butanol (BuOH). The BuOH-soluble portion was applied to a 60 g silica gel 60 column (63–200 μm, Merck) and eluted with gradient amounts of MeOH in EtOAc. The active fraction of EtOAc and MeOH (90:10, v/v) was purified by eluting in MeOH with a Sephadex TM LH-20 column (Amersham Biosciences, Uppsala, Sweden), followed by PR-μ-BondaPak C18 column chromatography (Waters, Milford, MA, USA) for the active compound [
30].
Chemicals and materials
We obtained endothelial cell basal medium-2 (EGM-2) from Lonza Cambrex (Nottingham, UK) and obtained M-199 and RPMI 1640 tissue culture mediums from GIBCO (Grand Island, NY, USA). We obtained the anti-ICAM-1, anti-RAGE, anti-ERK, anti-phosphorylated-ERK, anti-JNK, anti-phosphorylated-JNK, anti-phosphorylated-p38, anti-NF-κB and anti-PCNA antibodies from Santa Cruz Biotechnology, Inc. (Heidelberg, Germany) and the antibodies against anti-VCAM-1, anti-p38, and phosphorylated-IkappaBα, PD98059 (an ERK1/2 inhibitor), SP600125 (a JNK inhibitor), and SB203580 (a p38 inhibitor) from Cell Signaling Technology Inc. (Danvers, MA, USA). We obtained the anti-GAPDH antibody from Merck (Darmstadt, Germany), and all the other most highly purified chemicals were commercially provided from Sigma-Aldrich (St. Louis, MO, USA).
Preparing the glyceraldehyde-induced AGEs
Bovine serum albumin (BSA, 20 mg/mL) was incubated with 20 mM glyceraldehyde in 0.1 M potassium phosphate buffer saline (PBS, pH 7.4) that contained 1 mM diethylene triamine pentaacetic acid for 7d at 37 °C in the dark. AGE formation was confirmed using fluorescence spectroscopy (excitation 370 nm/emission 440 nm), and we noted an approximately 200-fold increase in glycer-AGEs compared with the control BSA (Additional file
1: Figure S1), which strongly suggested that AGEs were formed. After 7d, reduced BSA (rBSA) was recombined with 80 mM sodium borohydride for 30 min to delete the unexpected glycated products during incubation. After the unreacted sugar or small molecular reactants in the 0.1 M PBS were dialyzed, the prepared glycer-AGEs, and rBSA were sterilized by filtration and stored at −20 °C.
Cell culture and treatment
Primary human umbilical vein endothelial cells (HUVECs) were purchased and maintained with EGM-2 medium that contained endothelial growth supplements (Lonza, Seoul, Korea) with 2% fetal bovine serum (FBS) at 37 °C in a retained 5% CO2 incubator. The human leukemia THP-1 cells were cultured in an RPMI 1640 medium with 0.05 mM 2-mercaptoethanol with 10% inactivated FBS. To treat with the glycer-AGEs, the HUVECs were co-treated with different concentrations of PM and 100 μg/mL of AGEs using FBS-free media. In specific inhibitor-treated experiments, the inhibitors were pre-treated for 1 h and removed before the sample treatments.
Measuring cell cytotoxicity
We measured the cell cytotoxicity of PM or N-acetylcysteine (NAC) in the endothelial cells using colorimetric MTT assay. We seeded wells at 1 × 104 cells/well and grew the cells in 96-well culture plates. After 24 h, we applied differing concentrations of PM or NAC with 100 μg/mL of AGEs for the indicated times. After the treatments, we removed the supernatants and incubated the cells with 48 μL of MTT reagent (5 mg/mL) for 4 h. We quantified the reduction of MTT to formazan, which is dissolved by DMSO, at 540 nm using a multi-plate reader (Bio-Tek Instruments, Winooski, VT, USA) and assessed cell viability as the percentage of untreated control cells.
Determining the intracellular ROS in the HUVECs
The HUVECs (1 × 104 cells/well) were cultured in 96-well culture plates, and 10 μM DCF-DA were pre-incubated in a culture medium at 37 °C in a CO2 incubator for 30 min. After they were washed with PBS, the cells were co-treated with differing concentrations of PM with 100 μg/mL of glycer-AGEs for 3 h. For specific inhibitors, 10 μM rotenone, 10 μM apocynin and 10 μ Mallopurinol were pre-treated for 1 h before sample treatments. We analyzed the oxidation of DCF-DA to DCF by intracellular ROS by measuring the fluorescence at excitation of 485 nm and emission of 535 nm using a fluorescence spectrophotometer (VICTOR3TM, PerkinElmer, Waltham, MA, USA). Intracellular ROS levels are expressed as percentages of rBSA-treated cells.
Cell extraction and Western blotting
We cultured the HUVECs (1 × 106 cells) in 60-mm dishes and co-incubated them with or without 10 μM PM or 1000 μM NAC in the presence of 100 μg/mL of AGEs for the indicated times of 60 min to 24 h. We pre-treated the specific inhibitors for 1 h before the sample co-treatments, washing them twice with cold PBS, suspending them in RIPA lysis buffer (ELPIS Biotech Inc., Daejeon, Korea) for total cell lysates with protease inhibitors (5 μg/mLof leupeptin, 5 μg/mLaprotinin and 1 mM PMSF), and then centrifuging them at 13,000 × g for 20 min at 4 °C. For cell lysates with phosphorylated proteins, we added phosphatase inhibitor cocktail 2 (Sigma-Aldrich, St. Louis, MO, USA) to the same lysis buffer as before. To determine nuclear fractionation, we prepared the cells with PBS and cytoplasmic fraction separated with a cytosol extract buffer (10 mM Hepes, pH 7.8 with 10 mM KCl, 0.1 mM EDTA, 1 mM DTT and 10% NP-40). We prepared the nuclear extract with a nuclear extraction buffer (50 mM Hepes, pH 7.8 with 50 mM KCl, 300 mM NaCl, 0.1 mM EDTA, 1 mM DTT and 20% glycerol) after removing the cytosolic extracts and strongly vortexing the cells for 10 min at 4 °C. We then determined the protein contents in the total and nuclear fractions using a BCA protein assay (Pierce Biotechnology, Waltham, MA, USA). We reconstituted the samples in a loading buffer that contained 60 mM Tris–HCl, pH 6.8, 10% glycerol, 2% sodium dodecyl sulfate (SDS), 1% β-mercaptoethanol and 0.02% bromophenol blue, and boiled the mixture for 10 min at 100 °C. We loaded equal amounts of the denaturalized proteins into each lane, separated them by 10% SDS-polyacrylamide gel electrophoresis, and transferred them to PVDF membranes (Merck Millipore, Billerica, MA, USA). We blocked the transferred membranes in 5% non-fat dried milk in Tris-buffer saline with 0.1% Tween-20 for over 1 h at room temperature and then reacted them with different primary antibodies overnight at 4 °C. We incubated HRP-conjugated specific secondary antibodies for 45 min at room temperature, developing the blots using enhanced chemiluminescence (AbClon, Seoul, Korea). We quantified band intensities using the National Institutes of Health’s Image J software.
Monocyte adhesion assay
We adhered the monocytes for 24 h to the HUVECs with human leukemic monocyte THP-1, treating the HUVECs cultured at a concentration of 2 × 104 cells/well in 24-well culture plates that contained AGEs with or without PM and NAC; the THP-1 cells were labeled with 100 μM BCECF-AM for 30 min at 37 °C in a CO2 incubator. We co-cultured the treated HUVECs with FBS-free reagents prior to labeling the THP-1 (4 × 104 cells/well) for 1 h at 37 °C. After we twice gently removed the non-adhered THP-1 cells, we lysed the cells in 0.1% SDS in 50 mM Tris–HCl, pH 7.4, and we detected the fluorescence using the fluorescence spectrophotometer with excitation at 485 nm and emission at 535 nm.
To observe the monocytes’ adhesion to the endothelial cells, we seeded the HUVECs on 12-well culture plates and treated them for 24 h with or without 10 μM PM and 1000 μM NAC that contained 100 μg/mL glycer-AGEs. After the treatments, we co-cultured the BCECF-AM-labeled THP-1 cells for 1 h. We washed the free THP-1 cells with PBS and could visualize the adhered THP-1 cells by confocal laser microscopy (Carl Zeiss, Oberkochen, Germany).
Immunofluorescence staining
To determine the importance of NF-κB p65 nuclear translocation, we seeded the HUVECs (1 × 105 cells/well) on 12-well culture plates and treated them with or without 10 μM PM and 1000 μM NAC in presence of 100 μg/mL glycer-AGEs for 4 h. We fixed the cells in 3.7% paraformaldehyde in PBS for 20 min and permeabilized them with 0.1% Triton X-100 in PBS for 15 min at room temperature. PBS washing was conducted three times in each step, each time blocked with 1% BSA in PBS for 1 h and then incubated with anti-NF-κB (p65) primary antibody in 1% BSA overnight at 4 °C. Anti-rabbit Alexa 488 fluorescence was incubated for 2 h at room temperature, and the nuclei were stained with 4′,6′-diamidino-2-phenylindole (DAPI, 500 ng/mL) for 10 min. Stained cells were washed with 1% BSA and visualized with confocal laser microscopy (Carl Zeiss, Oberkochen, Germany).
Preparing the RNA and quantitative real-time reverse transcription-polymerase chain reaction
We harvested the HUVECs (1 × 106 cells/dish) in 60-mm dishes and co-cultured the THP-1 (2 × 106 cells/dish) for 1 h on the HUVECs. We co-treated10 μM PM or 1000 μM NAC that contained 100 μg/mL of glycer-AGEs in M199 medium for 6 h. We extracted total RNA using TRIzol Reagent (TAKARA Korea Biomedical Co, Seoul, Korea) and generated cDNA using the LeGene Premium Express 1st Strand cDNA Synthesis System (Legene Biosciences, SanDiego, CA, USA). We performed quantitative real-time PCR (qRT-PCR) with HiPi Real-timePCR 2X Master MixSYBR green (Elpis Biotech, Seoul, Korea) and analyzed the results using the iQ5 thermal cycler (Bio-rad, Foster City, CA, USA). The specific human mRNA primers we used in this study were as follows: RAGE (286 bp), forward primers, 5′-GGAATGGAAAGGAGACCAAG-3′, reverse primers, 5′-CCCTTCTCATTAGGCACCAG-3′; ICAM (409 bp), forward primers, 5′-TGAAGGCCACCCCAGAGGACAAC-3′, reverse primers, 5′-CCCATTATGACTGCGGCTGCTGCTACC-3′; VCAM(660 bp), forward primers, 5′-GGAACCTTGCAGCTTACAGTGACAGAGCTCCC-3′, reverse primers, 5′-CAAGTCTACATATCACCCAAG-3′; TNF-α(600 bp), forward primers, 5′-CCCAGGGACCTCTCTCTAATCA-3′, reverse primers, 5′- GCTACAGGCTTGTCACTCGG-3′; IL-6(550 bp), forward primers, 5′-GGTACATCCTCGACGGCATCT-3′, reverse primers, 5′-GTGCCTCTTTGCTGCTTTCAC-3′; MCP-1(161 bp), forward primers, 5′-TCGCGAGCTATAGAAGAATCA-3′, reverse primers, 5′-TGTTCAAGTCTTCGGAGTTTG −3′; GAPDH(550 bp), forward primers, 5′- GAAGGTGAAGGTCGGAGT-3′, reverse primers, 5′- GAAGATGGTGATGGGATTTC-3′. We analyzed the amplified genes in 1.0–1.5% agarose gels under UV light and normalized the mRNA expression levels with GAPDH expression.
Statistical analysis
All data are quantified as mean ± standard deviation (SD) for the triplicate experiments. We used SAS version 9.3 (SAS Institute, Cary, NC, USA) to analyze the statistical differences, defining significance using Duncan’s multiple range test for p < 0.05 for all tests.
Discussion
Studies have reported that
P. asiatica has bioactive activities as a number of phytochemical agents [
29,
30]. In particular, PM purified from
P. asiatica is a phenylpropanoid glycoside that contains caffeic acid derivatives [
31,
33]; nevertheless, the molecular mechanism of how PM modulates endothelial dysfunction remains uncertain. Our groups have reported that glycolaldehyde-derived AGEs stimulated intracellular ROS production and pro-inflammatory mediators including TNF-α and IL-1β via the AGE-RAGE axis [
39]. One recent study showed that co-treatment with PM and glycer-AGEs in keratinocytes and fibroblasts inhibits UVB-irradiation- and AGE-induced RAGE overexpression and proinflammatory cytokine expression via attenuating MAPK activation by ROS [
40]. Given that PM is absorbed very rapidly into the blood in rats after oral administration of
P. asiatica extract [
41], in the present study, we co-treated PM with glycer-AGEs in HUVECs to inhibit intracellular ROS production via AGE-RAGE interaction. In the study, we confirmed that PM suppressed monocyte adhesion by glycer-AGEs and inflammation cascades in endothelia, and we showed preventive effects on NF-κB translocation and phosphorylation of JNK and p38 in HUVECs by glycer-AGEs.
It has been reported that AGEs accelerate cellular inflammation [
9‐
12]. In particular, intracellular ROS are critical in inducing inflammation in cells and play a critical role in endothelium activation related with vascular diseases including atherosclerosis; the chronic oxidative stress could activate a number of different signal pathways including MAPK and NF-κB [
42]. In the present study, PM showed high inhibitory activity against glycer-AGE-mediated intracellular ROS generation with 10 μM of the compound (49 ± 10% of rBSA), whereas NAC inhibited the comparable ROS generation (77 ± 4% of rBSA) at the 1000 micromolar level. In addition, our study showed that glycer-AGEs increased intracellular ROS via not only NAD(P)H oxidase with apocynin and mitochondrial electron transportation with rotenone but also from xanthine oxidase with allopurinol in HUVECs. This suggested that glycer-AGEs are multiple sources of generating ROS in cellular systems. NAC has been reported as a major antioxidant reagent in a wide variety of experiments.
Monocytes adhere to endothelia with the expression of adhesion molecules including VCAM and ICAM as well as inflammatory cytokines both in vitro and in vivo [
19,
20,
43,
44]. Through the adhesion molecules, circulating leukocytes adhere to endothelial cells and invade endothelial barriers under various stimuli, and these leukocytes secrete multiple inflammatory cytokines, resulting in chronic inflammatory diseases [
21], and endothelial dysfunction with adhered monocytes is a key to the sclerotic inflammation that is a feature of diabetes complications [
45]. Other authors have demonstrated that phytochemicals such as sulforaphane in cruciferous vegetables [
46], acteosides in flowers [
47], green tea derivatives [
48] and curcumin [
49] have potentially preventive effects on endothelial dysfunction. Our data confirmed that PM inhibits the expression of adhesion molecules and blocks monocyte adhesion to endothelial cells as well.
In diabetes, RAGE activation induces a variety of inflammatory cytokines resulting from the NF-κB activation, including TNF-α, IL-6 and MCP-1, in the vascular system [
50‐
52], and these relevant inflammatory cytokines contribute to the development of early atherosclerosis [
53]; in addition, with adhesion molecules, MCP-1 contributes to the transmigration and infiltration of monocytes to endothelia [
54,
55]. Our study confirmed that treatment with PM significantly reduces the glycer-AGE-induced mRNA levels of TNF-α, IL-6, and MCP-1 under THP-1 co-cultured conditions. For chronic initiation states of inflammatory cytokine-related sclerotic diseases, leukocytes with HUVECs are more critical for expressing the pro-inflammatory cytokines, and not only in HUVECs; NF-κB activation was closely involved in the inflammation response in the cellular system with MAP kinase pathways. There have been diverse NF-κB hypotheses regarding the connection(s) between MAPK signaling and NF-κB that systematized cell responses [
56‐
59]. Based on our results, glycer-AGEs increased nuclear translocation of NF-κB and activated the phosphorylation of JNK and p38 MAPK in HUVECs, but PM inhibited this glycer-AGE-triggered JNK and p38 activity. In addition, treatments with PM, as well as JNK, p38 and NF-κB inhibitors, significantly blocked monocytes’ adhesion to endothelial monolayers against glycer-AGEs, suggesting that PM may be a pivotal regulator in the vascular inflammation that induces endothelial dysfunction via the MAPK/NF-κB pathways.
Acknowledgements
This work was supported by Korea University Research Grant No. K1516071 and School of Life Sciences and Biotechnology for BK21 PLUS, Korea University. The content is solely the responsibility of the authors.