Tannic Acid Down-Regulates the Angiotensin Type 1 Receptor Through a MAPK-Dependent Mechanism

Abstract

In the present study, we investigated the effects of tannic acid (TA), a hydrolysable polyphenol, on angiotensin type 1 receptor (AT1R) expression in continuously passaged rat liver epithelial cells. Under normal conditions, exposure of cells to TA resulted in the down-regulation of AT1R-specific binding in concentrations ranging from 12.5–100 μg/ml (7.34–58.78 μm) over a time period of 2–24 h with no change in receptor affinity to angiotensin II (AngII). The inhibitory effect of TA on AT1R was specific and reversible. In TA-treated cells, we observed a significant reduction in AngII-mediated intracellular calcium signaling, a finding consistent with receptor down-regulation. Under similar conditions, TA down-regulated AT1R mRNA expression without changing the rate of mRNA degradation, suggesting that TA's effect is mediated through transcriptional inhibition. Cells expressing recombinant AT1R without the native promoter show no change in receptor expression, whereas a pCAT reporter construct possessing the rat AT1R promoter was significantly reduced in activity. Furthermore, TA induced the phosphorylation of MAPK p42/p44. Pretreatment of the cells with a MAPK kinase (MEK)-specific inhibitor PD98059 prevented TA-induced MAPK phosphorylation and down-regulation of the AT1R. Moreover, there was no reduction in AngII-mediated intracellular calcium release upon MEK inhibition, suggesting that TA's observed inhibitory effect is mediated through MEK/MAPK signaling. Our findings demonstrate, for the first time, that TA inhibits AT1R gene expression and cellular response, suggesting the observed protective effects of dietary polyphenols on cardiovascular conditions may be, in part, through inhibition of AT1R expression.

Angiotensin II (AngII), the functional peptide hormone of the renin angiotensin system, displays a wide range of physiological and pathophysiological functions (1, 2). AngII binds to at least two high-affinity receptor subtypes termed angiotensin type 1 (AT1R) and angiotensin type 2 (AT2R) that mediate angiotensin II functions through different signal transduction pathways (3–5). The AT1 and AT2 receptors have similar affinities for angiotensin II, but distinctly different and opposing effects (6). Most pathological effects of AngII are through AT1R, and unambiguous data from basic and clinical research demonstrate that AT1R blockade protects against AngII-mediated pathological events and reduces mortality (7). Because there is overwhelming evidence indicating a significant role for AT1R in the pathogenesis and progression of cardiovascular, renal, and liver diseases, much effort has been directed toward identifying the role of AT1R function in these disease processes, and very little attention is given to the upstream factors involved in the regulation of AT1R expression.

Plant polyphenols are well known for their cardiovascular protective effects. Dietary intake of polyphenols is associated with low blood pressure and reduced incidence of cardiovascular disease (8). Many of these hypotensive, endothelium-independent vasodilator and cardio-suppressant properties are due to the sustained elevation of vascular nitric oxide (9), the ability to lower low-density lipoprotein and triglyceride plasma concentrations, or to increase high-density lipoprotein concentrations (10), significant reduction of reactive oxygen species (11), or increased renal sodium and water excretion (12). Studies have indicated that regular consumption of moderate amounts of red wine is associated with a reduced risk in overall mortality by a reduction from the risks of cardiovascular diseases and cancer (13). Red wine was reported to prevent AngII-induced hypertension and endothelial dysfunction in a spontaneously hypertensive rat (SHR) model through prevention of vascular NADPH oxidase induction and preserving the availability of arterial nitric oxide (NO) (14). Red wine is rich in polyphenols, a list of compounds that includes resveratrol, gallic acid, tannic acid, and catechins (15). Cardio-protective effects of dealcoholized red wine further suggest the presence of some specific protective compound, possibly one or more of the polyphenols (16). Furthermore, Negishi et al. (17) reported the attenuation of hypertension in rats fed black and green tea polyphenols, although this effect was presumed to be due to their antioxidant properties. The authors also suggested that tea polyphenols may also inhibit smooth muscle contraction by up-regulating catalase (a scavenger of H₂O₂) expression in rat aorta. Therefore, in general, polyphenolic compounds have been shown to exert a spectrum of pharmacological and therapeutic properties against oxygen free radicals and oxidative stress.

The cardioprotective effects of tannic acid (TA) and other polyphenols thus described may appear compelling, but the understanding of the mechanisms of these naturally bioactive compounds often excludes a specific target, and thus their effects remain enigmatic although comparative analysis suggests a significant connectivity between these products and genes associated with human diseases (18). There is a significant relationship between human pathophysiology and gene transcription, and at the clinical level, the observation of human disease may be effectively correlated to the differential activity of activators, specific trans-acting factors, and gene expression (19). The primary focus of this study was to investigate the effect of TA on AT1R expression and understand the mechanisms responsible for its unique regulatory properties. TA is a naturally occurring polyphenol commonly present in fruits, vegetables, nuts, red wine, tea, and coffee (20). It consists of a glucose core, which covalently links to three to five gallic acid residues through hydrolyzable ester bonds (21). Hydrolysable tannins have been shown to regulate multiple biological properties, including antioxidant and antiinflammatory properties (22–25). However, their effects on gene transcription are not well understood. In a previous study, the authors showed that glucocorticoid-induced gene expression of mouse mammary tumor virus (MMTV) is sensitive to TA, as well as other related compounds. They also demonstrated the presence of a novel TA response element in the MMTV promoter capable of inhibiting gene transcription (26).

In this study, we used a continuously passaged rat liver epithelial cell (WB) as a model system because it shows high levels of native AT1R expression (27). Our hypothesis was that TA plays a pivotal role in the regulation of AT1R gene expression and function. We demonstrated that TA has a specific inhibitory effect on AT1R gene expression through a MAPK-dependent pathway in WB cells. Consistent with AT1R down-regulation, we observed a decrease in intracellular calcium signaling. To our knowledge, this is the first demonstration that TA directly down-regulates AT1R expression. In addition, these results indicate that there may be a potential benefit in using TA for isolated or systemic therapeutic effect.

Results

TA down-regulates AT1R expression

To determine whether TA mediates down-regulation of AT1R expression, we performed radioligand-binding studies. Cells were exposed to 25 μg/ml TA for 20 h, and AT1R cell surface expression was determined in the presence and absence of AT1R-specific antagonist, losartan (10 μm). Incubation of semiconfluent WB cells in complete media supplemented with 50 μg/ml TA for 20 h shows a significant reduction (48.36 ± 1.973%; P < 0.0001) in specific binding of radiolabeled AngII (Fig. 1A). Down-regulation of the binding capacity for radiolabeled AngII is specific to AT1R as demonstrated in the presence of losartan. Losartan-treated groups, regardless of TA administration, showed no significant difference (0.04614 ± 0.1590%; P = 0.7861); if they had, the reduction in binding would be nonspecific and all binding of radiolabeled AngII would be equally affected by the application of TA. A radioligand-binding study using variable concentrations of TA revealed that TA mediated down-regulation of receptor density is dose dependent (radioligand binding down-regulated 66.88 ± 2.066% at 50 μg/ml) (Fig. 1B). In future studies, we used 50 μg/ml concentration because it resulted in roughly a 50% decrease in receptor expression. We used this concentration when we next examined TA's effect on AT1R relative to time. Exposure of cells with 50 μg/ml TA resulted in a time-dependent down-regulation of AngII binding up to 20 h (reduction of 72.00 ± 1.109%) (Fig. 1C). However, the binding is slightly increased at 24 h (reduction of 54.54 ± 2.939%), suggesting that TA's effect on AT1R is reversible. To confirm whether the inhibitory effect of TA on AT1R is reversible, the cells were exposed to TA for 20 h after which treatment media were exchanged with fresh complete media without TA. Cells were grown for an additional 24 h and AngII binding was performed. The results of the study revealed that replacing the media for 24 h reversed the down-regulatory effect of TA in WB cells (P = 0.1, ns) (Fig. 1D). To test whether down-regulation of AngII binding is due to a change in receptor affinity toward the ligand, a competition binding study using increasing concentration of unlabeled AngII was performed in cells treated with or without TA (Fig. 2, A and B). Using calculations from Swillens, 1992 (76), the calculated dissociation constant (K_d), of the AT1R in TA-treated cells was within normal binding affinity for AT1R for AngII (2.217 ± 0.3708 nm) as was the calculated K_d in the untreated control (2.093 ± 0.4042 nm). This suggested that the reduction in AngII binding in TA-treated cells was due to decreased receptor density rather than a change in receptor affinity. To further validate the reduction in surface expression, immunofluorescent studies were performed. Using a rabbit polyclonal antibody raised against the N terminus of the AT1R (SC-1173; Santa Cruz Biotechnology, Santa Cruz, CA) followed by an antirabbit Alexa-Fluor 488 conjugated secondary antibody (A11034; Invitrogen, Carlsbad, CA), the result shows significant cell surface expression in the untreated cells compared with TA-treated cells (Fig. 2C). The distinctive staining at the plasma membrane well separated from the blue-stained nuclei provided further confirmation that the receptor density is being reduced.

Fig. 1.

TA mediates AT1R down-regulation. A, TA inhibits AT1R specific binding in WB cells. Radioligand binding assay after 20 h 50 μg/ml TA treatment and losartan blockade. Cells were exposed to TA (50 μg/ml), for 20 h and [³H]AngII binding was measured in the presence or absence of losartan (an AT1R-specific antagonist) (n = 3). B, TA-mediated AT1R inhibitory effect is dose dependent. Radioligand binding assay after variable concentrations (12.5–100 μg/ml) of TA treatment for 20 h (n = 3). C, TA-mediated AT1R inhibitory effect is time dependent. Radioligand binding assay after exposure to 50 μg/ml TA for variable treatment times as indicated. TA inhibits AT1R binding more than or equal to 50% within 12 h, remaining less than 50% of control binding for the remainder of time points, although a small recovery was observed at 24 h (n = 3) D, TA-mediated down-regulation of AT1R specific binding is reversible upon withdrawal of TA from treatment media. Radioligand binding assay after either 20 h 50 μg/ml TA treatment, or a 20 h 50 μg/ml TA treatment followed by replenishment of fresh complete media for a further 24 h (20 TA+24-TA) (n = 3). Data are expressed as mean ± sem. ***, P < 0.0001 vs. untreated control.

Fig. 2.

TA-mediated AT1R down-regulation is due to reduction in cell surface expression. Competition binding studies reveal no change in receptor affinity after TA treatment. Radioligand binding competition studies performed on untreated cells (A) and cells treated with 50 μg/ml TA for 20 h (B). Nonlinear least-squares regression analysis gave a K_d of 2.217 ± 0.3708 nm for receptors in cells exposed to normal medium and a K_d of 2.093 ± 0.4042 nm for receptors in cells exposed to TA (n = 3). C, Immunofluorescent study shows the AT1R down-regulated in cells exposed to TA. Immunofluorescent staining using primary rabbit anti-AT1R IgG followed by secondary antirabbit IgG conjugated with Alexa Fluor 488. Nuclei stained with 4′,6-diamidino-2-phenylindole. Left, Native AT1R in WB cells with no TA exposure. Right, AT1R expression after treatment with 50 μg/ml TA for 20 h. Conc., Concentration.

AT1R density down-regulation is linked to a decrease in calcium signaling

To establish a relationship between observed reductions in AT1R density with cellular responsiveness to AngII, we performed Ca²⁺ mobilization studies. In Fura2-AM-loaded cells, Ca²⁺ was monitored using a qualitative and a quantitative method. The results of our studies demonstrate a significant reduction in AngII-mediated intracellular Ca²⁺ release in cells treated with TA as shown by microspectrofluoremetry (Fig. 3A). When intracellular calcium release was measured quantitatively, it was found that a 52.23 ± 7.790% (P = 0.0034) reduction in Ca²⁺ release takes place when cells are treated with TA (Fig. 3B). The resultant decrease in Ca²⁺ release correlated with our previous observation of reduced surface expression of AT1R mediated by TA treatment.

Fig. 3.

Fluorescent imaging shows 50 μg/ml TA exposure for 20 h significantly inhibited AngII-mediated increase in intracellular Ca²⁺, a G protein-coupled response. A, Transient free Ca²⁺ was measured with the Ca²⁺ indicator Fura 2-AM using fluorescent imaging microscopy before and after AngII stimulation (10⁻⁷ m), with or without TA treatment. Ca²⁺-bound Fura 2-AM emits green fluorescence. B, Ca²⁺ FLIPR assay shows a significant reduction in fluorescent emission after 50 μg/ml TA exposure for 20 h. Representative tracing of AngII-mediated transient increase in intracellular Ca²⁺ without (left) and with (right) TA treatment. The tracings are representative of three separate experiments; an overall reduction of 52.23 ± 7.790%.

Inhibitory effect is associated with mRNA down-regulation

To determine whether the observed down-regulation of AT1R density was transcriptionally related, we carried out dual RT-PCR analysis on AT1R mRNA with and without TA treatment. The housekeeping gene β-actin mRNA is not affected by TA treatment and is therefore used as a mRNA quantitation standard. Dual RT-PCR was performed in a single reaction and analyzed by ethidium bromide 2.0% agarose gel electrophoresis. Densitometric analysis of multiple experiments (n = 3) revealed that TA exposure resulted in a 40.00 ± 1.972% (P < 0.0001) reduction of AT1R mRNA transcript compared with control (Fig. 4A). To confirm the authenticity of the amplicon, Southern blot analysis was performed using an internal probe specific for AT1R and β-actin (Fig. 4B). The down-regulation of mRNA corresponded with the results of our binding studies. To verify that the observed down-regulation of AT1R mRNA is not due to increased mRNA transcript degradation, we determined the effect of TA on AT1R mRNA half-life in cells pretreated with actinomycin D. The results revealed no change in the rate of mRNA degradation (half-lives: 7.20 ± 0.1603 vs. 7.39 ± 0.1952 h for control and TA-treated cells respectively, P = 0.5005, ns) (Fig. 4C). This suggests that the down-regulation of AT1R mRNA is at the transcriptional level. To further corroborate transcriptional down-regulation of AT1R by TA, T3CHO/AT_1A cells (28) expressing the rat AT1R coding sequence linked with a cytomegalovirus (CMV) promoter, rather than the native promoter, were assayed for receptor expression using the radiolabeled ligand-binding technique in presence or absence of TA under similar conditions. The results revealed that these cells were TA insensitive at an equivalent dose and time (mean difference 2.710 ± 3.740%; P = 0.5089, ns) (Fig. 5A). This result indicated that TA is exerting its effect directly on the rat AT1R promoter and suggested that down-regulation of AT1R by TA is through inhibition of AT1R mRNA transcription. To confirm that TA-mediated down-regulation of AT1R is through promoter repression, an additional analysis using the rat AT1R promoter (3321 bp) linked to a chloramphenicol acetyl transferase reporter gene (pCAT) was performed. The result showed that TA has a significant down-regulatory effect on the AT1R promoter itself. Compared with the untreated control, 50 μg/ml TA for 20 h resulted in a 35.48 ± 10.11% (P = 0.0246; n = 3) reduction in CAT activity (Fig. 5B).

Fig. 4.

TA inhibits AT1R mRNA expression. A, Representative image of ethidium bromide gel of a dual RT-PCR reaction using AT1R/actin-specific primers to determine the effects of TA on AT1R mRNA in WB cells. Lane 2 and 3 are PCR lacking reverse transcriptase (RT) showing RNA specificity. Bands detected are at 289 and 206 bp for β-actin and AT1R mRNA, respectively. Densitometric analysis of multiple experiments (n = 3) revealed that TA exposure resulted in a 40.00 ± 1.972% down-regulation of AT1R mRNA (P < 0.0001). B, Representative image of Southern blot performed on transferred bands and probed with internal biotinylated primer specific for β-actin and AT1R mRNA. C, TA-mediated down-regulation of AT1R mRNA does not affect mRNA stability. Quantitation of multiple analyses of AT1R expression were normalized to GAPDH (n = 3) and compared with untreated control. Calculated half-lives: 7.20 ± 0.1603 vs. 7.39 ± 0.1952 h for control and TA-treated cells, respectively (P = 0.5005).

Fig. 5.

TA-mediated AT1R down-regulation is native promoter dependent. A, TA does not mediate AT1R down-regulation in T3CHO/AT_1A linked to CMV promoter. Radioligand binding assay performed after 20 h 50 μg/ml TA treatment. Mean difference between treated and untreated control 2.710 ± 3.740%, P = 0.5089. B, pCAT reporter assay demonstrates that TA down-regulates AT1R reporter activity (mean difference between treated and untreated control, 35.48 ± 10.11%; P = 0.0246; n = 3). CAT assay normalized to β-galactosidase activity, with untreated control arbitrarily set at 100% activity among three separate experiments.

TA down-regulates AT1R expression through MAPK p42/p44 activation

To determine the TA-mediated cellular signaling responsible for AT1R mRNA and protein down-regulation, we focused on MAPK p42/p44, which was shown previously to play a role in glucose-induced AT1R down-regulation (28). Cells were grown to 70–80% confluence in complete media and exposed to TA. Treatment with 50 μg/ml TA for 20 h induced MAPK p42/p44 phosphorylation without any change in the total protein (Fig. 6). This TA-mediated phosphorylation was completely inhibited by MAPK kinase (MEK) inhibitor PD98059 (20 μm). Under similar conditions, radiolabeled binding data showed that there was little to no effect by TA on binding of AngII when MEK signaling was inhibited in the presence of 20 μm PD98059 (mean difference between untreated control and PD98059+TA-treated cells: 2.603 ± 4.121%; P = 0.5619) (Fig. 7A). Consistent with the binding data, quantitative calcium analysis revealed that there was significant reversal of the TA-mediated reduction in Ca²⁺ signaling (mean difference 4.54±12.0% when compared with untreated control: P = 0.7255) (Fig. 7B). To further confirm TA-mediated activation of MAPK p42/p44, we used immunofluorescent microscopy and monitored translocation of MAPK p42/p44 after TA application. The result shows that TA not only activates MAPK p42/p44, but induces its translocation to the nucleus (Fig. 7C). This effect is reversible if the cells are pretreated with 20 μm PD98059. From these analyses we concluded that MAPK p42/p44 is activated upon exposure to TA and translocates to the nucleus. Upon translocation, the MAPK p42/p44 then directly or indirectly affects the transcription and thus the expression of the AT1R protein, as TA mediated down-regulation is eliminated upon blockade of MEK and subsequently MAPK p42/p44.

Fig. 6.

TA induces phosphorylation of MAPK p42/p44. Total cell lysates were prepared from control, TA-treated, MEK inhibitor PD98059-treated, and TA+PD98059-treated cells and immunoblotted with phospho-specific MAPK antibody (A). Blot stripped and reprobed with anti-MAPK antibody to demonstrate equal loading (B). A representative blot is shown (n = 3).

Fig. 7.

TA-mediated down-regulation of AT1R is MAPK p42/p44 dependent. A, MEK inhibition restores radioligand binding in TA-treated cells. Radioligand binding assay was performed after 20 h 50 μg/ml TA-treated and untreated cells in the presence or absence of PD98059 (20 μm). Mean difference between PD98059+TA and untreated control is 2.603 ± 4.121%; P = 0.5619 (n = 3). Data are expressed as mean ± sem. ***, P < 0.0001 compared with control. B, Ca²⁺ FLIPR assay shows MEK inhibitor PD98059 restores intracellular Ca²⁺ release in TA-treated cells. Increase in intracellular Ca²⁺ was measured as described in Materials and Methods. Representative tracings of transient increase in intracellular Ca²⁺ in untreated control cells (panel A), TA-treated cells (panel B), PD98059-treated cells (panel C), and PD98059+TA-treated cells (panel D) (n = 3). Mean difference between PD98509+TA and PD-treated control, 4.54 ± 12.0%; P = 0.7255. C, MAPK p42/p44 translocates to the nucleus upon treatment with 50 μg/ml TA. Immunofluorescent staining using primary rabbit anti-phospho MAPK p42/p44 IgG followed by secondary antirabbit IgG conjugated with Alexa Fluor 488. Nuclei stained with 4′,6-diamidino-2-phenylindole (DAPI). Treatment groups included are untreated control, TA treatment, PD98059 treatment, and combined TA and PD98059 treatment. Images are divided among fluorescein isothiocyanate (FITC), DAPI, and overlaid images to demonstrate nuclear colocalization. Images shown are representative of three individual experiments.

Discussion

The overall focus of the present study was to elucidate the effects of the naturally occurring polyphenol TA on AT1R gene expression. We used continuously passaged rat liver epithelial cells (WB), which naturally express AT1R, as evident by the fact that more than 95% of radiolabeled AngII binding is AT1R specific antagonist losartan sensitive. TA treatment for 24 h resulted in a significant reduction of AT1R expression in a dose- and time-dependent manner. We observed a reversal of AT1R expression toward the end of the treatment period. In fact, upon replenishment of TA-treated cells with new media, the AT1R expression recovered to near control levels, indicating that TA-mediated down-regulation is a reversible process. Consistent with the reduction of receptor density on the cell surface, we observed a significant reduction in AT1R mRNA in TA-treated cells without alteration in the rate of mRNA degradation. Independent studies were performed using T3CHO/AT_1A cells (28) expressing the recombinant rat AT1R with a constitutively active viral promoter (CMV). TA treatment showed no effect on the AT1R protein expression in this cell line, further supporting that TA-mediated down-regulation of AT1R is mediated by inhibition of AT1R gene transcription. Moreover, reporter gene analysis using the 5′-rat AT1R promoter displays a significant down-regulation in reporter activity, suggesting a direct effect by TA on the AT1R promoter and reducing the probability of posttranslational events brought about by TA treatment. There is significant evidence that AT1R transcription may be altered by pathological states in a tissue/cell-specific manner. For example, SHR have shown to have up-regulated AT1R expression (29–31). AT1R expression has been also seen to be up-regulated in a number of other pathological conditions, including the tissue surrounding a myocardial infarct site (32), cardiovascular tissue under oxidative stress (33, 34), cardiovascular tissues in type 2 diabetics (35–37), and in many different forms of cancer (38–45). Up-regulation is a consequence of induced transcription factor activation and binding of cis-elements on the AT1R promoter; for instance, the transcription factors Ap1 (46, 47) and Sp1 (48, 49) have shown direct binding activity and control of AT1R transcription and protein production. One of the major nuclear events reported in the process of regulation of AT1R expression is interaction of Sp1 with upstream GC-rich promoter regions. Sequential mutational studies carried out on the −105/+39 promoter region of the human AT1R confirmed two GC-rich Sp1 interacting domains (50). Interestingly, supplementation with the polyphenol penta-O-galloyl β-d-glucose demonstrated reduced binding activity of Sp1 and Ap1 as well as reduced expression of the Sp1 protein itself (51). The cytokines TNFα and IL-1β have shown in a number of analyses to have an up-regulatory role in AT1R expression (52, 53). Moreover, TA has shown to have an inhibitory effect in this dynamic because its supplementation has inhibited the expression of IL-1 and TNFα (54). In states in which AT1R up-regulation presents the potential development of pathophysiological conditions, it is imperative to understand compounds capable of down-regulating receptor expression, thereby effectively antagonizing the pathological consequences at the source.

TA is a primary constituent of red wine polyphenols; polyphenol extracts are nonselective bioactive products without specific targets (55). Treatment with these extracts has shown significantly high endothelial NOS expression and subsequent nitric oxide (NO) release from endothelial cells (56). Perhaps more significantly, red-wine polyphenol extracts supplemented in drinking water have resulted in a significant alleviative effect on AngII-induced hypertension, though not in normotensive animals (7). Additionally, other phenolic extracts, such as raspberry fruit ethyl acetate extract, given orally to SHR rats, has resulted in significant lowering of blood pressure (57). In these studies, there were simultaneous increases in serum NO and superoxide dismutase levels concomitant with a decrease in malonedialdehyde. Endothelial dysfunction by AngII is mediated through excessive NADPH oxidase-dependent vascular formation of reactive oxygen species, and the beneficial effects of red wine polyphenols are attributed to their reactive oxygen scavenging activity and persistent NO formation, which act to synergize their respective effects on AngII signaling antagonism (7). TA in particular possesses both antioxidant properties and NO-releasing ability. In view of the previous findings that NO inhibits Sp1 activity by affecting its affinity for Zn⁺², a necessary cofactor (58, 59), TA may have a direct transcriptional inhibitory effect, thereby providing cardioprotection. The 3321-bp AT1R promoter fragment used in the present study to demonstrate transcriptional repression possesses multiple characteristic TA repressor elements (ACTG), similar to those identified in the MMTV study (26). Nevertheless, the immunofluorescent studies display intracellular AT1R staining; the appearance of the AT1R in endocytic vesicles is suggestive of internalization, desensitization or recycling to the plasma membrane after AngII receptor activation. In the present study, cells were never exposed to AngII for receptor activation; therefore, it is likely that these receptors are in transit to the plasma membrane after de novo synthesis. However, the specificity of cis-acting elements in TA-mediated AT1R down-regulation and the presence of newly synthesized receptors in transit to the plasma membrane requires further investigation.

Further, we identified a TA-mediated signaling mechanism at the cellular level responsible for AT1R down-regulation. We observed that activation of MAPK p42/p44 and its translocation to the nucleus is necessary for TA-mediated down-regulation of AT1R expression; in fact, the MEK inhibitor PD98059 completely abrogates TA-mediated reduction in AT1R expression. Interestingly, MAPK p42/p44 has been identified as an antagonistic effector on down-regulation of AT1R by peroxisome proliferator-activated receptor γ (PPARγ). PPARγ mediates AT1R transcription suppression through inhibition of Sp1 interaction with the upstream cis-acting elements on the AT1R promoter, and this effect may be reversed after PPARγ phosphorylation and inactivation by activated MAPK p42/p44 with simultaneous overexpression and activation of cAMP response element binding protein-binding protein, which has affinity for Sp1 and favors binding within the −58/−34 region of the rat AT1R promoter (60–62). It has been shown that Sp1 is a critical mediator of basal AT1R transcription (48), and disruption of Sp1 cis-acting element binding activity may result in a down-regulatory effect on AT1R transcription. Therefore we suspect the down-regulatory effect of TA follows a PPARγ independent pathway requiring MAPK p42/p44 activation. TA is reported to inhibit activation of p38 MAPK, MAPK p42/p44 at a concentration (IC₅₀) of 142 μm (63). The hepatoprotective effect of TA has been attributed to inhibition of poly-ADP ribose polymerase/ERK/Elk-1 pathway and histone acetylation (64). However, in our studies we used a lower concentration (29.39 μm to 50 μg/ml), which is well below 142 μm, the calculated IC₅₀ for MAPK p42/p44 inhibition (63). At this lower concentration we observed phosphorylation of MAPK p42/p44 with no significant apoptosis. The disagreement between the previous studies and this could be due to time-dependent activation of MAPK p42/p44, which has been shown to have cyclic activation patterns; therefore, a particular observation coinciding with a different time point may have an alternate MAPK activation profile (65). Previous studies have led to interesting similarities involving flavonoids and activation of MAPK p42/p44. Grape seed extract (GSE) contains both gallic acid and 3,3′-di-O-gallate ester of procyanidin dimer B2, which has shown potent anticancer activity (66). Upon exposure of human colon carcinoma cells (HT29) to GSE, researchers found that MAPK p42/p44 was significantly phosphorylated leading to increased expression of p21, cell cycle arrest, and apoptosis (67). Additionally, two different agents, GSE or epidermal growth factor (EGF), in human prostate cancer cells (DU145) independently activated MAPK p42/p44 (68). Interestingly, the results of this study described a dual relationship of MAPK activation depending upon whether EGF or GSE elicited MAPK phosphorylation. EGF administration activated MAPK p42/p44, but resulted in increased cellular proliferation, whereas GSE significantly reduced DNA synthesis and cell viability, although it still activated MAPK p42/p44. Analysis of these results led us to conclude that MAPK p42/p44 activation is not necessarily indicative of either cellular proliferation, apoptosis, or receptor down-regulation, but the ultimate consequence of MAP activation depends both on the cell type and the particular treatment agent that may subsequently activate a number of cosignals yet to be determined. In further support of our observation, it is important to note that receptors other than AT1R have been reported to be down-regulated by TA. Membrane receptors (epidermal growth factor receptor, fibroblast growth factor receptor, and platelet-devived growth factor receptor) have all been down-regulated by TA and gallic acid (64). Butein, another polyphenolic compound, has been shown to have antityrosine kinase activity at high concentrations; although the authors concluded butein's effects to be through kinase inhibition, the results may also be explained by a similar effect to tannic and gallic acids by a direct down-regulation of the receptor tyrosine kinases themselves (69).

To further confirm the functional relevance of TA-mediated down-regulation of AT1R, we performed AngII-mediated intracellular calcium release studies in TA-treated cells. We found that TA-treated cells showed a significantly reduced level of AngII-induced intracellular Ca²⁺ release when compared with nontreated control. Because vasoconstriction is mediated through Ca²⁺ release (70), up-regulation of the receptor in pathophysiological conditions results in enhanced vasoconstriction leading to increased mean arterial blood pressure. Our study suggests that TA may play an important role in attenuating AngII-mediated hypertension, as well as other complications from overactive AngII signaling. In summary, our data report for the first time that TA is capable of down-regulating AT1R expression at the transcriptional level without changing receptor affinity. The observed down-regulation is mediated through activation and translocation of MAPK p42/p44 to the nucleus. Consistently with the down-regulation of the AT1R, we observed TA treatment results in an inhibition of AngII-mediated intracellular calcium release. If we extrapolate the Ca²⁺ data obtained from our WB cell study, and interpret a similar effect in the context of another cell type, e.g. vascular tissue, the subsequent interpretation would lead us to believe that TA treatment would effectively alleviate the pathological condition, i.e. hypertension, resultant from deregulated AT1R expression in the vasculature. However, further studies must be conducted to confirm TA's effectiveness in this cell type.

Materials and Methods

Materials

Richter's improved MEM was obtained from Cellgro (Mediatech Inc, Manassas, VA), [³H]Angiotensin II was from Amersham (Arlington Heights, IL), Losartan was provided by Merck Sharp & Dohme Research Laboratories (Rahway, NJ). Fetal bovine serum was from Equitech-Bio, Inc. (Kerrville, TX). Oligonucleotide primers and biotinylated probes were obtained from Integrated DNA Technologies, Inc. (Coralville, IA). Unlabeled AngII and TA were from Sigma (St. Louis, MO). Fura-2 AM and MEK-specific inhibitor PD98059 were from Calbiochem (La Jolla, CA). AT1R antibody was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) and Phospho- MAPK antibody was from Cell Signaling Technology (Danvers, MA). Electrophoresis regents were from Bio-Rad Laboratories, Inc., (Hercules, CA), and all molecular biology grade chemicals were purchased from Fisher Scientific (Fairlawn, NJ).

Cell culture

The continuously passaged rat liver epithelial cells (WB) were kindly provided by Dr. H. Shelton Earp (University of North Carolina, Chapel Hill, NC). Cells were maintained in Richter's improved MEM containing 10% fetal bovine serum, 10 mm glucose, 17.8 mm HEPES, 5.4 μg/ml insulin, and 44.6 μg/ml gentamicin (complete medium) at 37 C in 5% CO2 under 100% humidity. For these studies, cells were grown in complete media to 70–80% confluence, and the treatments were initiated in fresh media and grown for the indicated times.

Receptor binding studies

AngII binding studies were performed in triplicate on WB cells in 12-well plates as described previously (71). Briefly, cells were rinsed with PBS and incubated at 22 C for 30 min with 0.05 nm [³H]AngII in binding buffer (50 mm Tris-HC1 pH 7.4; 120 mm NaCl; 4 mm KCl; 1 mm CaC1₂; 10 μg/ml bacitracin; 0.25% BSA; and 2 mg/ml dextrose). Nonspecifically bound AngII was removed by washing three times with the binding buffer (without BSA). Cells were then lysed in 0.2 n NaOH and transferred to counting vials after which radioactivity was determined using a Beckman auto-gamma scintillation spectrometer. Specific [³H]AngII binding was defined as that portion of the total binding displaced by 1 μm unlabeled AngII or 10 μm Losartan (AT1R-specific antagonist). At equilibrium, specific binding was more than 95% of the total binding. Competition binding studies were performed in the presence of 0.05 nm [³H]AngII and increasing concentrations (1 pm to 10 μm) of unlabeled AngII. All binding studies were analyzed using Graphpad Prism (GraphPad Software, Inc., San Diego, CA) as appropriate, column statistics to test for normal distribution, and parametric t test or one-way ANOVA with post hoc Bonferroni analysis.

Cell extracts and Western blot analysis

Cells were treated with various agents for the indicated times and washed with ice-cold PBS. Cells were scraped in lysis buffer (10 mm Tris-Cl, pH 7.4; 150 mm NaCl; 15% glycerol; 1% Triton X-100; 1 mm sodium orthovanadate; 10 mg/ml leupeptin; 10 mg/ml aprotinin; 1 mm NaF; and 1 mm phenylmethylsulfonyl fluoride. Protein concentration was determined using Bio-Rad protein assay reagent based on the Bradford method (72). Equal quantities of proteins were separated by 8% SDS-PAGE, transferred to a nitrocellulose membrane, and incubated with their respective primary antibodies. Immunoreactive bands were visualized using a chemiluminescence Western blotting system according to the manufacturer's instructions (Amersham).

Reverse Transcriptase (RT)-dual PCR and Southern blot analysis

Total RNA was extracted from control and TA-treated cells for 24 h by the acid guanidium thiocyanate-phenol-chloroform method as described previously (73). The first-strand cDNA was synthesized from 1 μg/sample total RNA using RETROscript reverse transcription kit (Ambion, Inc., Austin, TX) according to manufacturer's instruction. The cDNA was then amplified with a dual-PCR primer set for AT1R and β-actin or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA (AT1R sense: 5′-TGATTCAGCTGGGCG TCATCCA-3′; AT1R antisense: 5′-TTTCGTAGACAGGCTT GAGTGGG-3′: β-actin sense: 5′-CGGAACCGCTCA TTGCC-3′; β-actin antisense: 5′-ACCCACACTGTGCCCATCTA-3′). Dual PCR reactions were performed on cDNA using PCR Master (Roche Diagnostics, Indianapolis, IN) subjected to 25 cycles in a 7300 Real Time System (Applied Biosystems, Foster City, CA). The PCR products were separated on a 2.0% agarose gel stained with ethidium bromide. The AT1 (206 bp)- and β-actin (289 bp)-specific amplicons were visualized and quantified using Versa Doc 4000MP Imaging System (Bio-Rad Laboratories). In studies involving Southern blot analysis, dual amplicon bands were visualized under UV to assure appropriate band density among β-actin controls and transferred to a nylon membrane using Southern capillary transfer as described previously (30). The amplicon was then hybridized with biotinylated internal probes specific for both AT1R (5′-CTGACATCGTGGACACTGCCATG-3′) and β-actin (5′-AGAG GGAAATCGTGGGTGACATT-3′). The membranes were then washed and bands visualized using Chemiluminescent Nucleic Acid Detection Module according to the manufacturer's instructions (Thermo-Scientific; Rockford, IL). The data were expressed as percentage changes relative to unexposed cell expression, after normalization relative to β-actin.

mRNA Stability studies

The WB cells were seeded in ten 100-mm plates and incubated at 37 C with 5% CO₂ until 70% confluent. A total of five time points were selected for sampling from 0 h to 8 h. At 2-h intervals, each set of plates was treated with actinomycin D (1 μg/ml) for 15 min followed by 50 μg/ml TA for treatment groups. Control cells remained untreated. At the end of the incubation, media were removed from the plate by aspiration, and mRNA was isolated using guanidium thiocyanate method as described above. RT-PCR and dual-PCR were performed according to the reaction conditions also described in the previous method with the exception of studies in which GAPDH mRNA was used as housekeeping control (GAPDH sense: 5′-ACCACAGTCCATGCCATCAC-3′; GAPDH antisense: 5′-TCCACCACCCTGTTGCTGTA-3′). AT1R amplicon band density was analyzed relative to β-actin or GAPDH amplicon band density. The densitometric data were further analyzed by linear regression using GraphPad Prism software to determine the half-life of the AT1R mRNA.

Plasmid construction, transfection, and CAT assay

All DNA manipulations were carried out using standard techniques. A 3372-bp 5′-promoter sequence of the rat AT1_AR, (ref. no. NC005116.2) was custom cloned into a pUC57 carrier vector and sequence confirmed by GenScript (Piscataway, NJ). By using the genomic clone as a template for PCR and oligonucleotides corresponding to the published sequence of the rat AT1_AR promoter, a fragment of 3321 bp was amplified. The fragment was generated using the primers [sense: 5′-GAATTCGAGCTCGGTACCTCGCGA-3′; antisense: 5′-ACAGATCTTCTCCAG CGGGACA-3′ (oligonucleotide corresponding to exon 1/+23 to +45 bp)]. The fragment was cloned into the multiple cloning site of the pCAT-Basic expression plasmid (Promega Corp., Madison, WI). The authenticity of the clone was confirmed by dideoxy sequencing. For transfection of DNA constructs, WB cells were seeded in 100-mm plates and grown to 70–80% confluence in complete growth medium. Cells were transfected with 8 μg of reporter plasmid and cotransfected with 2 μg of pSV-β-galactosidase expression construct (to act as an internal control for transfection efficiency) by using the Lipofectamine-LTX reagent method according to manufacturers' instruction (Invitrogen, Carlsbad, CA) and grown for 18 h in growth medium. The CAT assays were performed according to manufacturer's instructions. Briefly, transfected cells were replenished with fresh growth medium after the initial 18-h incubation, with the treatment plate being supplemented with 50 μg/ml TA. After 20 h, cells were rinsed with PBS three times and harvested in the same buffer. Cells were then centrifuged and the resultant pellet was resuspended in 100 μl of 0.25 m Tris-HCl, pH 7.8. Cellular extracts were prepared by freeze-thaw cycling and 10-min centrifugation (12,000 × g at 4 C). Part of the supernatant (30 μl) was removed, and β-galactosidase activity was measured using a colorimetric assay according to the previously published method (74). The remaining supernatant was heated to 70 C for 10 min to inactivate endogenous acetylases and centrifuged further to remove cell debris. Fluorescent acetylated chloramphenicol was visualized using the VersaDoc at excitation/emission wavelengths of 497/520 nm. Densitometric quantitation of acetylated chloramphenicol was normalized to respective β-galactosidase activity, and each experiment was normalized to untreated control (arbitrarily taken as 100% activity).

Immunofluorescence microscopy

WB cells were grown in chamber slides (Lab-Tek, Naperville, IL) to 75–80% confluency and exposed to TA (50 μg/ml), PD98059 (20 μm), or TA and PD98059 in combination. After 20 h, cells were washed with ice-cold PBS, fixed with ethanol acetic acid mixture (3:1 vol/vol) or 3% paraformaldehyde in PBS for 30 min at 22 C. Cells were washed with chilled PBS (three times), blocked with 5% goat serum for 3 h, and incubated with primary antibody (diluted 1:1000-AT1R/MAPK) overnight at 4 C. Cells were then washed five times with chilled PBS and incubated with Alexa Fluor 488 fluorescent tagged secondary antibody (Invitrogen) for 1 h at 22 C in the dark followed by washing. In addition, nuclei were stained with 1 nm 4′,6-diamidino-2-phenylindole for 5 min and washed with ice-cold PBS. Slides were sealed with ProLong Gold antifade mounting medium (Invitrogen) and visualized and photographed with a fluorescence microscope (Olympus IX70; Olympus Corp., Lake Success, NY) using a ×20 objective equipped with and additional ×1.5 magnification.

Measurement of cytosolic free Ca²⁺ concentration

Cells were cultured in six-well plates for 48 h and exposed to TA (25 μg/ml) for 20 h in the presence or absence of PD98059 (20 μm). Cells were then rinsed with Hanks' balanced salt solution (HBSS), and the assay was performed as described by Grynkiewicz et al. (75). Briefly, cells were loaded with calcium dye Fura-2AM (1 μm) in Dulbecco's modified PBS with calcium in a 37 C incubator with 5% CO₂ and 100% humidity. After 1 h of incubation, cells were washed two times with Dulbecco's modified PBS with calcium. Additionally, cells were washed two times with assay buffer (145 mm NaCl, 2.5 mm KCl, 10 mm HEPES, 10 mm glucose, 1.2 mm MgCl₂, and 1.5 mm CaCl₂), and a final volume of 900 μl of assay buffer was added to each well. Baseline readings were taken at 340-nm excitation and 512-nm emission at 5-sec intervals using an ELISA plate reader (BioTek SynergyMx). After 2 min, 100 μl of AngII (10⁻⁶m) was added in each well, and immediately readings were taken continuously at 5-sec intervals for an additional 5 min. At the end of the experiment, maximal emissions of calcium-bound Fura-2AM and free Fura 2AM were obtained by adding digitonin and EGTA to a final concentration of 0.22 and 1.15 mm, respectively. The intracellular concentration of calcium was determined using the formula, Δ[Ca²⁺] = [(K_d × (F − F_min)]/[F_max − F)] − [K_d × (F_x − F_min)]/[(F_max − F_x)] in which K_d = 224 nm, F corresponded to the fluorescent reading after addition of AngII, F_x corresponded to the fluorescent reading just before addition of AngII, F_min the minimum reading after addition of EGTA, and F_max the maximum reading after the addition of digitonin. Qualitative imaging of changes of intracellular Ca²⁺ by microscopy followed a similar procedure. WB cells grown to 70–80% confluence in 35-mm optical bottom plates (MatTek no. P35G-0–10-C, Ashland, MA) and AngII-induced changes in intracellular calcium were monitored using microspectrofluorometry method as described elsewhere (75). Cells were loaded with 1 μm Fura2-AM in HBSS for 20 min. The cells were washed twice with HBSS, and changes in intracellular Ca²⁺ were measured. The microscope's emission wavelength was set at 510 nm, and the excitation wavelengths were set at 340 and 380 nm. Excitation was monitored by a high-speed wavelength-switching device recorded with a charge-coupled device camera. Images were collected and analyzed using Slidebook image analysis software.