Hyperhomocysteinemia decreases vascular reactivity and is associated with cardiovascular morbidity and mortality. However, pathogenic mechanisms that increase oxidative stress by homocysteine (Hcy) are unsubstantiated. The aim of this study was to examine the molecular mechanism by which Hcy triggers oxidative stress and reduces bioavailability of nitric oxide (NO) in cardiac microvascular endothelial cells (MVEC). MVEC were cultured for 0–24 h with 0–100 μM Hcy. Differential expression of protease-activated receptors (PARs), thioredoxin, NADPH oxidase, endothelial NO synthase, inducible NO synthase, neuronal NO synthase, and dimethylarginine-dimethylaminohydrolase (DDAH) were measured by real-time quantitative RT-PCR. Reactive oxygen species were measured by using a fluorescent probe, 2′,7′-dichlorofluorescein diacetate. Levels of asymmetric dimethylarginine (ADMA) were measured by ELISA and NO levels by the Griess method in the cultured MVEC. There were no alterations in the basal NO levels with 0–100 μM Hcy and 0–24 h of treatment. However, Hcy significantly induced inducible NO synthase and decreased endothelial NO synthase without altering neuronal NO synthase levels. There was significant accumulation of ADMA, in part because of reduced DDAH expression by Hcy in MVEC. Nitrotyrosine expression was increased significantly by Hcy. The results suggest that Hcy activates PAR-4, which induces production of reactive oxygen species by increasing NADPH oxidase and decreasing thioredoxin expression and reduces NO bioavailability in cultured MVEC by 1) increasing NO2-tyrosine formation and 2) accumulating ADMA by decreasing DDAH expression.
- NADPH oxidase
- nitric oxide
- protease-activated receptor
- nitric oxide synthase
- 2′,7′-dichlorofluorescein diacetate
- asymmetric dimethylarginine
- microvascular endothelial cells
hyperhomocysteinemia induces endothelial dysfunction and promotes the development of cardiovascular diseases (24, 25). Homocysteine (Hcy), a sulfur-containing amino acid, is not found in our daily diet. It is primarily formed from the demethylation of methionine during DNA/RNA methylation. l-Hcy is the primary active form in a variety of tissues or cells (24, 25), and it has been suggested that increased levels of plasma Hcy may play a role in the pathogenesis of various diseases, particularly at the cardiovascular level (24, 25). The cellular and molecular mechanisms underlying the adverse effect of hyperhomocysteinemia have not been fully elucidated.
Al-Obaidi et al. (1) suggested that Hcy potentiates the production of thrombin in endothelial cells. Thrombin is a potent activator of a unique group of protease-activated receptors (PARs) that belong to the G protein-coupled receptor family. Four PARs (PAR-1, PAR-2, PAR-3, and PAR-4) have been identified (11). PAR-1, PAR-3, and PAR-4 are activated by thrombin and PAR-2 and PAR-4 by trypsin. All PARs are expressed in endothelial cells (8, 11). PAR-1, PAR-2, and PAR-4 are involved in vascular development and a variety of biological processes, including remodeling (2). Activation of PARs induces generation of reactive oxygen species (ROS) (6), upregulates NADPH oxidase (6, 41), and downregulates thioredoxin (28) in endothelial cells.
Hyperhomocysteinemia increases oxidative stress and is closely related to accumulation of asymmetric dimethylarginine (ADMA) (34, 36), an endogenous nitric oxide (NO) synthase (NOS) inhibitor that inhibits the activity of endothelial NOS (eNOS) and inducible NOS (iNOS). The inhibitory effect of ADMA on NO synthesis is removed by dimethylarginine-dimethylaminohydrolase (DDAH), which catalyzes the conversion of ADMA to l-arginine, citrulline, and dimethylamine (29, 37). Elevation of ADMA concentrations attenuated acetylcholine-induced coronary vasodilation, indicating impairment of endothelial NO signaling (20, 23). NO is an important mediator of many physiological phenomena (21, 22). The increase in iNOS during oxidative stress generates nitrotyrosine (3). The inhibition of NOS by ADMA and increased generation of nitrotyrosine may significantly reduce NO bioavailability. A controversy exists regarding the mechanism by which Hcy regulates NO production in endothelial cells. In the present study, we determine the cellular mechanism of nitrotyrosine formation by differential expression of NOS in response to antagonism of PARs by Hcy in microvascular endothelial cells (MVEC).
MATERIALS AND METHODS
Complete MCDB-31 medium was obtained from VEC Technology. Penicillin, streptomycin, trypsin-EDTA, and Hanks' balanced salt solution (HBSS) were purchased from GIBCO-BRL (Grand Island, NY); dl-Hcy, NaCl, sodium orthovanadate, Tris, EDTA, EGTA, dithiothreitol, NP-40, protease inhibitor cocktail, fibronectin, agarose, anti-nitrotyrosine antibody, and 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) from Sigma (St. Louis, MO); and anti-PAR-4, anti-iNOS, and horseradish peroxidase-conjugated antibodies from Santa Cruz Biotechnology. The RNeasy kit was obtained from Qiagen, protein assay reagents from Bio-Rad (Hercules, CA), and the ADMA ELISA kit from Cardio-Vasic (Palo Alto, CA). All other reagents were obtained from Sigma.
Rat heart MVEC were isolated and characterized by CD-31 labeling as described elsewhere (39) with some modifications. The cells were grown on fibronectin-coated 75-cm2 flasks in MCDB-31 medium with 20% fetal bovine serum, penicillin (100 U/ml), streptomycin (100 μg/ml), basic endothelial growth factor (3 ng/ml), and heparin (5 U/ml) at 37°C under 5% CO2-95% air. Cells between passages 3 and 7 were grown to near confluence and serum starved overnight before treatment. The MVEC (105 cells/ml) were cultured for 0, 6, 12, and 24 h with 40 μM dl-Hcy. For subsequent studies, confluent MVEC were incubated with 0–100 μM dl-Hcy for 24 h.
Determination of ROS generation.
ROS generation in cells was assessed by using the probe DCFH-DA, a lipid-permeable nonfluorescent compound oxidized by intracellular ROS to form the lipid-impermeable and fluorescent compound 2′,7′-dichlorofluorescin (30). MVEC were preincubated in phenol red-free medium containing 2% FCS with 0–100 μM dl-Hcy for 24 h. The cells were washed with medium. DCFH-DA was added (10 μM final concentration), and the cells were incubated for 45 min at 37°C. The DCFH-DA solution was removed, and the cells were washed with serum-free medium. The cells were treated with 0.25% trypsin and resuspended in 500 μl of PBS. Fluorescence intensity of DCF was read at 525-nm emission when excited at 488 nm in a microplate reader (Molecular Devices).
In situ labeling of ROS.
MVEC were grown on coverslips for 48–72 h. The cells were washed twice with HBSS and incubated with 40 μM dl-Hcy for 24 h. After the cells were washed twice with HBSS, 10 μM DCFH-DA was added, and the cells were incubated in a CO2 incubator for 45 min at 37°C. Fluorescence was detected by using a confocal microscope (model FV-1000, Olympus).
Total RNA was harvested from cells using an RNeasy kit, which included DNase digestion. The concentration of total RNA was quantified by measuring the absorbance at 260 nm. The integrity of RNA was checked by electrophoresis using a 3% formaldehyde gel. Only samples with a peak area ratio of 28S-to-18S rRNA >2.0 were used.
Two micrograms of RNA were reverse transcribed to cDNA following the manufacturer's instruction using the SuperScript III First-Stand Synthesis for RT-PCR kit (Invitrogen Life Technologies) containing an oligodeoxythymidine primer in a final reaction volume of 20 μl. A negative control, which used the same reaction without SuperScript III reverse transcriptase, was included to ensure the absence of any genomic DNA contamination in the RNA template.
Designs of oligonucleotide primers specific for the different targets were based on sequences available in GenBank (Table 1). PCR was done in a 50-μl mixture containing 5 μl of 10× PCR buffer, 1 μl of 10 mM dNTP, 1.5 μl of 50 mM MgCl2, 12.5 pmol of specific primer pairs, and 2.5 U of recombinant Taq DNA polymerase (TaKaRa). PCR mixtures were placed in a thermal cycler (model 9600, Perkin-Elmer). The amplification for PARs and glyceraldehyde-3-phosphate dehydrogenase was started with initial denaturation at 94°C for 5 min followed by 30 cycles consisting of denaturation at 94°C (1 min), annealing at 57°C (1 min), and extension at 72°C (2 min). After the last cycle, a final extension step was done at 74°C for 5 min. PCR cycling was initiated at 94°C for 2 min followed by 94°C for 30 s, annealing at 55°C for 30 s (for iNOS) or 65°C for 30 s [for neuronal (nNOS) and eNOS], and 72°C for 1 min, and a final extension step at 72°C for 2 min was followed by 30 cycles. PCR products were analyzed by Tris-acetic acid-EDTA-agarose (1%) gel electrophoresis containing ethidium bromide.
Real-time quantitative PCR.
Steady-state mRNA levels were quantified by real-time fluorescence detection as described previously (16, 18). Real-time RT-PCR was performed by using the Mx3000P real-time PCR system (Stratagene). Template cDNA corresponding to 50 ng of RNA and SYBR green PCR core reagents (Stratagene) were used, and each measurement was done three times. Primers were designed with Primer 3 software (31). The primer sequences for NADPH oxidase, thioredoxin, and glyceraldehyde-3-phosphate dehydrogenase were based on GenBank sequences NM-053683, NM-053800, and NM-017008, respectively. Primer sequences were as follows: TTCTTGGCTAAATCCCATCC (forward) and TACCATGAGAACCAAAGCCA (reverse) for NADPH oxidase, GTGAAGCTGATCGAGAGCAA (forward) and CGTGGCAGAGAAGTCCACTA (reverse) for thioredoxin, and AGTTCAACGGCACAGTCAAG (forward) and GTGGTGAAGACGCCAGTAGA (reverse) for glyceraldehyde-3-phosphate dehydrogenase. For PCR, target DNA was denatured and a Taq polymerase was activated by preincubation for 10 min at 95°C. DNA was amplified for 45 cycles of 5 s at 95°C, 6–8 s at 55–70°C, and 3–8 s at 72°C. The fluorescence signal was measured at the end of the elongation phase. The annealing temperature was optimized for specificity of the PCR product. Data were analyzed by version 2.0 of the software supplied by Stratagene. Relative quantitation of mRNA was measured against the internal control, glyceraldehyde-3-phosphate dehydrogenase. Variations in the initial concentration and quality of total RNA and in the conversion efficiency of the RT reaction were also accounted for by normaliza-tion to glyceraldehyde-3-phosphate dehydrogenase. The end point, used in real-time RT-PCR quantification and for the cycle threshold (CT) value, is defined as the PCR cycle number that crosses an arbitrarily placed signal threshold. Expression was measured by using the comparative CT value (2).
Enzyme-linked immunosorbent assay for ADMA.
ADMA concentration (μM) was determined by ELISA. Eighty percent confluent cells treated with 0–100 μM dl-Hcy for 24 h and 40 μM dl-Hcy for 0–24 h were extracted in PBS containing 1% BSA. Protein concentrations were measured by the Bradford method with BSA as the standard. According to the manufacturer's instruction, 30 μg/ml protein was used for ELISA analysis.
Quantitative detection of NO production.
Colorimetric-based measurements of the stable end products of NO metabolism, nitrite and nitrate, were used for assessment of NO production. Total nitrite released from endothelial cells was measured with a commercial colorimetric assay (Cayman Chemical; 2.5 μM detection limit). We measured NO by the 4,5-diaminofluorescein-formylmethanofuran dehydrogenase method (Calbiochem) and found results similar to those obtained by the Griess method.
Expression of nitrotyrosine.
Confluent MVEC treated with 0–100 μM dl-Hcy for 24 h or 40 μM dl-Hcy for 0–24 h were lysed in a buffer containing 150 mM NaCl, 1 mM sodium orthovanadate, 50 mM Tris·HCl (pH 8.0), 1 mM EDTA, 1 mM EGTA, and 1 mM dithiothreitol, as well as 1% NP-40, with a protease inhibitor cocktail. Lysates were centrifuged at 3,000 g for 10 min at 4°C. Protein concentrations were determined by the Bradford method (7) with BSA as the standard. Proteins (50 μg) were resolved by 10% Laemmli SDS-PAGE under reducing conditions and then electrotransferred overnight to a polyvinylidine difluoride membrane at 0.03 A at 4°C (Bio-Rad). Nonspecific sites on the membrane were blocked with 5% (wt/vol) nonfat dry milk-Tris-buffered saline with Tween 20 (TBST). The membrane was washed three times for 10 min each with TBST. Immunodetection was carried out by incubation of the polyvinylidine difluoride membrane overnight with a rabbit polyclonal antinitrotyrosine antibody (Molecular Probes) at 1:1,000 in 5% (wt/vol) nonfat milk-TBST followed by four washes with TBST and incubation with horseradish peroxidase-conjugated secondary antibody (1:3,000 dilution in 5% nonfat milk-TBST). A chemiluminescent protocol provided by the manufacturer (ECL, Amersham, Arlington Heights, IL) was used to identify the immunoreactive bands. Previously, we (33) and others used nitrated bands at ∼66 and 27 kDa as the marker of oxidative protein modification. The total nitrated bands from whole gel blots were scanned. The membranes were then stripped with 0.2 M NaOH solution for 30 min at room temperature and reprobed for actin. After the membranes were immunoblotted, the film was scanned and the intensity of immunoblot bands was detected with UNSCANIT software. To define the role of PAR-4 and iNOS in nitrotyrosine formation, MVEC cells were treated with 1:200 dilutions of anti-PAR-4 and anti-iNOS antibodies for 2 h. Pretreated cells were then incubated with 40 μM dl-Hcy for 24 h.
Data analysis and statistics.
Values are means ± SE from at least four different experiments. The arbitrary densitometry units were measured and are expressed as percentage of control. The data were analyzed by Student's t-test for comparison of the results between Hcy-treated and untreated groups. P > 0.05 was considered to indicate statistical significance.
Differential expression of PAR-1, PAR-2, PAR-3, and PAR-4 by Hcy.
MVEC expressed constitutive levels of PAR-1, PAR-2, PAR-3, and PAR-4 (Figs. 1 and 2). In some experiments, there was a slight tendency for Hcy to increase PAR-1, but on average there was no change. However, the levels of PAR-4 were significantly increased within 6 h of Hcy treatment (Fig. 1). Contrary to the results of PAR-4, the levels of PAR-2 and PAR-3 were significantly decreased in Hcy-treated MVEC compared with untreated cells (Figs. 1 and 2). These results suggest differential expression of PARs by Hcy in MVEC.
Differential expression of thioredoxin and NADPH oxidase.
In normal MVEC, there were more copies of thioredoxin mRNA than NADPH oxidase at any given time (Fig. 3). Treatment with Hcy reversed this ontology in gene expression of thioredoxin and NADPH oxidase. There was a time- and dose-dependent decrease in thioredoxin mRNA levels and an increase in NADPH oxidase expression in Hcy-treated compared with untreated cells (Fig. 3). To determine whether this differential regulation of NADPH oxidase and thioredoxin increases ROS, the concentration of ROS was measured by probing with DCFH in Hcy-treated and untreated MVEC. There was a dose- and time-dependent increase in ROS levels in MVEC treated with Hcy, causing oxidative stress in part by decreasing thioredoxin and increasing NADPH oxidase (Fig. 4).
Differential expression of eNOS, iNOS, and nNOS by Hcy in MVEC.
Although nNOS was not expressed in MVEC, eNOS and iNOS were constitutively expressed (Figs. 5 and 6). The dose- and time-dependent experiments demonstrated upregulation of iNOS and downregulation of eNOS by Hcy in MVEC. To determine whether differential expression of iNOS and eNOS affects NO levels, we measured total nitrate/nitrite. There was no change in the steady-state levels of nitrate/nitrite. To determine whether the changes in NOS expression increase oxidative stress and generate nitrotyrosine, total nitrotyrosine was measured. Hcy increased nitrotyrosine in MVEC in a dose- and time-dependent manner (Fig. 7). To determine whether PAR-4 and iNOS were induced at the protein/activity levels and whether the changes in PAR-4 and iNOS expression increased oxidative stress and subsequent nitrotyrosine formation, MVEC were cultured with anti-PAR-4 and anti-iNOS in the presence of Hcy. Anti-PAR-4 and anti-iNOS blocked nitrotyrosine formation, suggesting strong involvement of PAR-4 and iNOS in formation of nitrotyrosine (Fig. 8).
Hcy attenuates DDAH expression and increases ADMA levels.
To determine whether Hcy increases oxidative stress by decreasing l-arginine concentration and increasing ADMA concentration, we measured levels of ADMA and mRNA of DDAH in MVEC. Hcy decreased DDAH expression in a dose- and time-dependent manner (Fig. 9). ADMA levels were significantly increased (Fig. 10). These results suggest that Hcy increases ADMA levels by decreasing DDAH expression. Although we did not measure the IC50 for direct inhibition of eNOS by ADMA, ADMA concentration increased ∼2.5-fold in the presence of Hcy (Fig. 10): from 0.1 to ∼0.25 μM. Therefore, EC50 of ∼8.5 ± 3.6 μM for the Hcy effect is estimated. The results of this study suggest that eNOS expression is blocked by Hcy. The levels of ADMA were increased in cells treated with Hcy. This may suggest decreased eNOS availability. However, Hcy also increased iNOS and generated ROS and nitrotyrosine, causing no change in total NO levels.
Hcy induces thrombin (a serine proteinase) production in acute coronary syndromes (1, 13). Thrombin activates the PARs (32). Previously, we showed latent matrix metalloproteinases (MMPs) in the basement membranes of MVEC, and these latent MMPs are activated by serine proteinases as well as by Hcy-mediated oxidative stress (27). In the present study, we suggest that Hcy changes the expression of PARs. However, there may be no “cause-and-effect” relation between MMP and PAR. It is equally likely that an increase in PAR-4 leads to increased MMP expression or that they are two completely independent events. However, in an experimental study of chronic heart failure (14), we demonstrated that activation of MMP-9 in chronic volume overload was ameliorated by exogenous administration of a cardiac-specific inhibitor of metalloproteinase that inhibited MMP-9. The inhibitor inhibited MMP-9, which in turn blocked PAR-1 and, consequently, decreased the preload (26). In the present study, we demonstrate an inverse relation in the expression of PAR-2 and PAR-3 compared with PAR-4, and there is no significant change in levels of PAR-1. These results suggest differential regulation of PARs by Hcy in MVEC.
In vivo studies showed that mild and severe hyperhomocysteinemia is associated with impaired vasodilation (4, 35). These findings are consistent with impairment in NO bioavailability. However, recent research on the effect of Hcy on NO production and bioavailability is conflicting. Some studies show that high concentrations of Hcy increase NO production (40), whereas other studies show that Hcy decreases NO production (10, 41). Our present results show that Hcy induced oxidative stress by inducing iNOS and decreasing eNOS expression in MVEC. Although there was no change in total nitrate/nitrite levels, nitrotyrosine expression was increased.
Hcy induces NADPH oxidase (5, 41) and increases ROS (38). However, it was unclear whether the increase in ROS was secondary to a decrease in thioredoxin by Hcy. We demonstrated that Hcy instigated oxidative stress, in part, by decreasing thioredoxin. Interaction of ROS, such as superoxide, with NO generates peroxynitrite (3, 17), which reacts with tyrosine residues to produce nitrotyrosine (15). In the present study, we observed that Hcy significantly increased nitrotyrosine. Our finding that there was no change in basal levels of NO with different doses of Hcy at 0–24 h is consistent with the report of Jin et al. (19), who observed an increase in nitrotyrosine formation in response to Hcy without an alternation in basal NOS activity. An absence of changes in basal levels of NO may likely be due to the imbalance in iNOS and eNOS activity. Another potential mechanism underlying the Hcy-mediated pathophysiology of impaired NO bioavailability is the inhibition of eNOS activity by its endogenous inhibitor ADMA. Our findings demonstrated a significant increase in ADMA in cultured MVEC within 24 h at 40 μM Hcy. It may also be argued that, in addition to inhibiting NO production, ADMA may also be involved in “uncoupling” of eNOS, leading to increased production of ROS, which further decreases NO bioavailability.
Levels of ADMA are regulated by the enzymatic activity of DDAH. From our study, it is clear that Hcy treatment results in decreased DDAH expression, in a time- and dose-dependent manner, which is the possible cause for accumulation of ADMA in cultured MVEC. These findings are consistent with those of Cooke (12), who observed that Hcy inhibits the activity of DDAH and increases the secretion of ADMA in cultured endothelial cells. The present study suggests that oxidative inactivation of DDAH by Hcy-induced ROS production may be another potential mechanism responsible for the increase in ADMA. All these findings indicate that accumulation of ADMA in response to Hcy is another independent risk factor of impaired NO bioavailability in cultured MVEC.
Although we have not addressed the role of all vascular oxidases in ROS production, the focus of this study was to address the mechanism of ROS production involving NADPH oxidase, thioredoxin, NOS, and PARs. Hcy induces oxidative stress by upregulating PAR-4, decreasing thioredoxin and increasing iNOS and NAD(P)H oxidase expression. We preincubated the cells with anti-iNOS and anti-PAR-4 antibodies before adding Hcy and measured the levels of nitrotyrosine. The results suggested that anti-iNOS and anti-PAR-4 antibodies blocked nitrotyrosine generation (Fig. 8). This indicates that Hcy induces oxidative stress by upregulating PAR-4 and increasing iNOS and, consequently, decreasing thioredoxin and increasing NAD(P)H oxidase expression.
A potential limitation of this study may relate to the dose of reduced Hcy that was used. In “mild” human hyperhomocysteinemia (which is associated with an increased cardiovascular risk), plasma Hcy levels range from ∼15 to 30 μmol/l. However, only a fraction of total plasma Hcy is in the reduced form in vivo. The concentrations used in the present study represent a ∼100-fold dose. Three ranges of hyperhomocysteinemia are defined as follows: moderate (16–30 μM), intermediate (31–100 μM), and severe (>100 μM) (9). Extracellular thiols are oxidized. Only a fraction of total plasma Hcy is in the reduced form in vivo and in vitro.
In conclusion, Hcy induces oxidative stress by upregulating PAR-4 and promoting ROS production by increasing NADPH oxidase and decreasing thioredoxin. Increased oxidative stress via ROS also increases iNOS expression and subsequent nitrotyrosine formation. Hcy indirectly decreases eNOS by decreasing DDAH expression, causing accumulation of ADMA and directly producing ROS, contributing to decreased NO bioavailability without significant changes in basal NO levels (Fig. 11).
A part of this study was supported by National Heart, Lung, and Blood Institute Grants HL-71010 and HL-74185.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2005 by the American Physiological Society