(Hypertension. 2001;38:1260.)
© 2001 American Heart Association, Inc.
Scientific Contributions |
From Vascular Cell Biology, Joslin Diabetes Center, Harvard Medical School, Boston, Mass.
Correspondence to Edward P. Feener, PhD, Research Division, Joslin Diabetes Center, One Joslin Pl, Boston, MA 02215. E-mail Edward.Feener{at}joslin.harvard.edu
| Abstract |
|---|
|
|
|---|
Key Words: angiotensin II aorta heat shock proteins phosphorylation protein kinases receptors, angiotensin II
| Introduction |
|---|
|
|
|---|
Although an array of Ang IIstimulated phosphoproteins have been identified, little is known regarding which proteins represent major phosphorylation targets of the AT1 receptor signaling pathway(s). Although recent advances in proteomics may create new opportunities to characterization of Ang II signaling pathways and cardiovascular disease,13 the AT1-stimulated phosphorylation targets detectable by this approach have not yet been identified. A two-dimensional gel electrophoresis (2-DE)based proteomic analysis of total proteins from rat aortic smooth muscle cells (RASMCs) revealed that heat shock protein 27 (HSP27) represents the major low-molecular-weight phosphorylation target of the Ang II/AT1 pathway. This finding suggests that kinase cascades that lead to HSP27 phosphorylation represent a major conduit of AT1 receptor signaling. Ang II also increases HSP27 phosphorylation in aorta in vivo, suggesting that this response may have physiological significance. Because phosphorylation of HSP27 modulates actin filament assembly, migration, and contraction,1416 HSP27 phosphorylation may provide an important link between AT1 receptor activation and vascular remodeling.
| Methods |
|---|
|
|
|---|
Sample Preparation
Samples were lysed for 1 hour at 22°C followed by protein precipitation in 20% (vol/vol) trichloroacetic acid (TCA) on ice for 30 minutes. Protein pellets were washed with ethanol and redissolved in a Tris-free urea-lysis sample buffer. Total protein amount was determined with the use of Protein Assay Kit (Bio-Rad).
Protein Dephosphorylation
A subset of samples were reprecipitated by TCA, washed with ethanol, and resuspended in a buffer with 60 mmol/L Tris-HCl, pH 7.5, 1 mmol/L DTT, 1 µg/mL aprotinin, and 100 mmol/L NaCl. Samples were treated with 100 U of alkaline phosphatase (New England Biolabs) for 120 minutes at 30°C.
Two-Dimensional Gel Electrophoresis
First-dimension protein separation was performed by isoelectric focusing of 100 µg protein on immobilized pH gradient gel strips (pH 3 to 10, nonlinear; Amersham Pharmacia). Strips were equilibrated in 6 mol/L urea; 50 mmol/L Tris-HCl, pH 8.8; 30% (vol/vol) glycerol; 2% SDS; and 10 mg/mL DTT followed by separation in the second dimension by SDSpolyacrylamide gel electrophoresis on 12% C gels. Samples were then transferred to polyvinylidene difluoride (PVDF) or nitrocellulose membrane (Novex).
Immunoblotting and Image Analysis
HSP27 isoforms were separated by 2-DE and detected by immunoblot analysis. Membranes were incubated with polyclonal HSP-27 antibody (Santa Cruz Biotechnology) followed by visualization with peroxidase-conjugated anti-goat immunoglobulins and enhanced chemiluminescence (Amersham Pharmacia). Detection and quantification of the immunoblot and 32P-orthophosphate signal of the radiolabeled blotting membrane was determined with the use of a PhosporImager and the Imagequant software (Molecular Dynamics). Levels of individual HSP27 isoforms were quantified and expressed relative to total HSP27 immunoreactivity to normalize for protein recovery, as described previously.15,16 Acute Ang II stimulation did not affect total HSP27 levels measured by this method.
Phosphoamino Acid Analysis
Phosphoproteins from control and Ang IIstimulated cells were separated by 2-DE. The areas of nitrocellulose membrane containing HSP27 phosphoisoforms were excised, eluted, and hydrolyzed, and phosphoamino acid analysis was performed by cellulose thin-layer electrophoresis as described previously.10
Intrajugular Injections of Ang II
Anesthetized male Sprague-Dawley rats (weight, 250 g) received a bolus 0.5 mL injection of 40 µmol/L Ang II in the absence or presence of a 5-fold molar excess of candesartan through an intrajugular catheter. Control rats were similarly injected with saline. After 15 minutes, aortic tissue was harvested and frozen in liquid nitrogen. Samples were homogenized in urea buffer, and TCA-precipitated protein fraction was separated by 2-DE and immunoblotted for HSP27 as described above.
Statistics
Statistical significance using SigmaStat 2.03 was determined by 1-way ANOVA. Differences were designated to be significant at P<0.05.
| Results |
|---|
|
|
|---|
|
On the basis of the molecular weight and isoelectric points, a candidate immunoblot approach was used to determine if these phosphoproteins colocalized with HSP27. Lysates from 32P-labeled cells were separated by 2-DE, transfer red to PVDF membranes, and immunoblotted with anti-HSP27 antibody. Visualization of HSP27 immunoreactivity by enhanced chemiluminescence revealed 4 isoforms, separated according to their pI (Figure 1, bottom panels). The 2 most acidic of these HSP27 isoforms, labeled phosphoprotein-1 (PP1) and PP2, colocalized with the 32P-labeled phosphoproteins visualized by autoradiography. The Ang IIinduced increases in these HSP27 isoforms at pI 5.4 and pI 5.7 measured by immunoblotting were comparable with the increased 32P-incorporation into these spots. Two additional HSP27 isoforms at pI 6.4 and 5.8, which did not colocalize with 32P phospholabeling, were labeled as nonphosphoprotein-1 (NP1) and NP2, respectively.
Ang II Induces Phosphorylation of HSP27 in an Ang II Type 1 ReceptorDependent Manner
To examine the role of the AT1 receptor in Ang IIinduced HSP27 phosphorylation, cells were stimulated with Ang II in the absence or presence of the AT1 antagonist candesartan (CV11974, kindly provided by Dr Peter Morsing, Astra Hassle AB, Sweden). Stimulation of cells with Ang II (100 nmol/L) for 15 minutes decreased NP1 and NP2 levels by 55% and 31%, respectively (P<0.01 and <0.05, ANOVA; Figure 2). This Ang II response was associated with an increase in the relative amount of PP1 by 68% (P<0.05) and PP2 by 5-fold (P<0.001). Pretreatment of cells with 1 µmol/L candesartan completely blocked both Ang IIstimulated decreases in NP1 and NP2 levels and increased PP1 and PP2 levels (Figure 2). This Ang IIstimulated shift in HSP27 immunoreactivity was sustained at 30 minutes and occurred with an ED50 of 10 nmol/L Ang II (not shown). Treatment of lysates from Ang IIstimulated cells with alkaline phosphatase eliminated immunoreactivity of HSP27 at pI 5.4 and reduced immunoreactivity of HSP27 at pI 5.7 to control levels (Figure 2), demonstrating that phosphorylation is required for Ang IIinduced acidic shift of HSP27 immunoreactivity. These results support the identification of the 32P-labeled proteins at pI 5.7 and 5.4 as phosphoisoforms of HSP27 and show that Ang II induces the redistribution of HSP27 from nonphosphorylated to phosphorylated isoforms.
|
Phosphoamino Acid Analysis of Ang IIInduced HSP27 Phosphorylation
Phosphoamino acid analysis was performed on PP1 and PP2 isoforms isolated from ortho 32P-labeled RASMCs, as shown in Figure 1. This analysis revealed that Ang II stimulation increased phosphoserine incorporation into both of these HSP27 isoforms (Figure 3). Phosphorylation of HSP27 on threonine and tyrosine was not detected under basal or Ang IIstimulated conditions.
|
Effect of Protein Kinase C Inhibition on Ang IIStimulated and Phorbol EsterStimulated HSP27 Phosphorylation
Because Ang II activates protein kinase C (PKC) in VSMCs17 and direct activation of PKC has been shown to increase the phosphorylation of small heat shock proteins,18 the role of PKC in Ang IIstimulated phosphorylation was examined. Cells were pretreated with the PKC inhibitor GF 109203X (GFX, 5 µmol/L) for 15 minutes followed by stimulation with either Ang II or phorbol 12-myristate 13-acetate (PMA) for 15 minutes. Levels of HSP27 phosphoisoform PP2 normalized to total HSP27 content were quantified by immunoblotting, as described in Figure 2. GFX did not affect baseline HSP27 phosphorylation and appeared to enhance Ang IIstimulated HSP27 phosphorylation (Figure 4). Direct activation of PKC by treatment of cells with phorbol ester (PMA) also increased HSP27 phosphorylation, and this response was completely blocked by GFX, confirming the efficacy of this inhibitor.
|
Effect of Ang II on Aortic HSP27 Phosphorylation In Vivo
To determine whether Ang II affects vascular HSP27 phosphorylation in vivo, the distribution of HSP27 isoforms was examined in aortic tissue isolated from Sprague-Dawley rats. Anesthetized rats received an intrajugular 500-µL bolus injection of saline in the absence or presence of 40 µmol/L Ang II or 200 µmol/L candesartan. Fifteen minutes after injection, aortic tissue was harvested and HSP27 isoform distribution was examined by immunoblotting, as described in Figure 2. Ang II decreased the nonphosphorylated HSP27 NP1 isoform by 25% (P=0.05) relative to saline controls and by 35% (P<0.02) compared with candesartan-treated controls (Figure 5). In addition, Ang II infusion increased levels of PP2 by 70% (P<0.02) relative to saline controls and by 3-fold compared with candesartan-treated controls. The Ang IIinduced changes in HSP27 isoforms at NP1 and PP2 levels in vivo were consistent with those observed in cultured RASMCs (Figure 2). Coinfusion of candesartan blocked these Ang II responses, indicating that the AT1 receptor mediates the effects of Ang II on HSP27 in vivo. Ang II or candesartan did not affect the levels of NP2 and PP1 in aorta.
|
| Discussion |
|---|
|
|
|---|
Previous reports have shown that the p38 MAPK pathway, through the activation of downstream kinases MAPK activated protein (MAPKAP) and p38-regulated/activated protein kinase, mediates the phosphorylation of HSP27.15,16,19 Recently, it has been shown that a reactive oxygen speciesinduced stimulation of p38 MAPK in response to Ang II resulted in an activation of MAPKAP kinase-2 and an increase in immunoprecipitable phosphorylated HSP27.20 The latter, however, has neither been quantified nor normalized and was limited to cultured cells. Because HSP27 exists both in monomeric and polymeric forms21,22 and associates with actin filaments,23 which may affect its solubility and antigen accessibility, it is important to quantify the effects of Ang II on HSP27 phosphorylation relative to total HSP27 protein. The quantification of HSP27 isoforms according to isoelectric focusing has been described previously.15,16 2-DE analysis demonstrated that Ang II not only increases levels of highly phosphorylated HSP27, which migrate with an acidic pI 5.4, but also decreases levels of the nonphosphorylated HSP27 isoform NP1. This nonphosphorylated form of HSP27 is the biologically active isoform that binds to F-actin.23
Activation of phorbol estersensitive PKC isoforms (
, ß,
,
,
,
,
) increased HSP27 phosphorylation, and this response was blocked with GFX, which inhibits the common (
, ß,
) and novel (
,
) PKC isoforms. Our previous results have shown that GFX also inhibits phorbol esterinduced activation of p38 in RASMCs,24 which can mediate to PKCinduced HSP27 phosphorylation. Although Ang II activates PKC in VSMCs, the absence of an inhibition by GFX suggests that this pathway does not significantly contribute to Ang IIstimulated HSP27 phosphorylation. Similarly, GFX does not inhibit AT1-mediated phosphorylation of p38 MAPK in RASMCs.25 Thus, AT1 signaling to the p38 MAPK and HSP27 pathway does not require PKC activity.
Because the 2 previously identified phosphorylation sites on human HSP27 and murine HSP25 are conserved at Ser15 and Ser86 in rat HSP27,26,27 it is likely that the Ang IIstimulated serine phosphorylation of HSP27 occurs at these sites. Based on the phosphorylation-dependent acidic shift of HSP27, the presence of 2 phosphoisoforms would be consistent with the monophosphorylation of HSP27, at either Ser15 or Ser86, in the isoform located at pI 5.7 and diphosphorylation of these sites on the isoform at pI 5.4, as described in References 15 and 16. Although Ang II increases phosphorylation of isoforms at both pI 5.7 and 5.4 in cultured RASMCs, Ang II selectively increased levels of the pI 5.4 isoform in vivo. These results suggest that the physiologically significant Ang II effects on HSP27 are reduced levels of the nonphosphorylated HSP27 isoform at pI 6.4 and the increased phosphorylation of the (diphosphorylated) PP2 isoform.
Phosphorylation of HSP27 plays an important role in modulating its structure and functions. Nonphosphorylated monomeric HSP27 binds to F-actin as a capping protein and thereby blocks actin polymerization, whereas phosphorylated HSP27 does not affect actin polymerization.23 Thus, the Ang IIinduced decrease in levels of the nonphosphorylated isoform of HSP27, observed both in vitro and in vivo, would be expected to increase the dynamics of actin assembly, which is necessary for VSMC migration.16
Phosphorylation of HSP27 also affects its multimerization, which is required for its chaperone activity, which facilitates the refolding of denatured proteins.21 Mutation of Ser90 to negatively charged Glu in hamster HSP27 (which corresponds to Ser86 in rat and Ser82 in human) reduces HSP27 multimeric structure from oligomers to dimers,22 suggesting that a negative charge at this site interferes with the multimerization of HSP27 dimers. The specific role of Ser15 in multimerization is controversial. Although mutation of Ser15 to Glu did not appear to affect HSP27 oligomerization,24 another report showed that reduction of high-molecular-weight multimers required negative charges at both Ser15 and Ser82 in human HSP27.23 Thus Ang IIinduced phosphorylation of HSP27 would be expected to reduce its chaperone activity, which protects cells from a variety of stresses, including oxidation.21
In summary, HSP27 was identified as the major low-molecular-weight target of Ang II/AT1 stimulated phosphorylation in RASMCs by a proteomic approach. In vivo studies demonstrated that Ang II phosphorylates HSP27 in aorta. The Ang IIinduced redistribution of HSP27 phosphoisoforms would be expected to enhance actin remodeling, which is essential for VSMC migration, and impair the chaperone activity of HSP27, which recovers proteins partially denatured by cellular stress. Thus, HSP27 phosphorylation may contribute to the AT1 receptor effects on cell migration, contraction, and the recovery of denatured proteins.
| Acknowledgments |
|---|
Received May 8, 2001; first decision June 6, 2001; accepted July 3, 2001.
| References |
|---|
|
|
|---|
and -ß in vascular smooth muscle cells. Hypertension. 1994; 23: 848852.This article has been cited by other articles:
![]() |
G.-X. Shi, L. Jin, and D. A. Andres Pituitary Adenylate Cyclase-Activating Polypeptide 38-Mediated Rin Activation Requires Src and Contributes to the Regulation of Hsp27 Signaling during Neuronal Differentiation Mol. Cell. Biol., August 15, 2008; 28(16): 4940 - 4951. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Zoubeidi, A. Zardan, E. Beraldi, L. Fazli, R. Sowery, P. Rennie, C. Nelson, and M. Gleave Cooperative Interactions between Androgen Receptor (AR) and Heat-Shock Protein 27 Facilitate AR Transcriptional Activity Cancer Res., November 1, 2007; 67(21): 10455 - 10465. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Huang, J. B. Fileta, T. Filippopoulos, A. Ray, A. Dobberfuhl, and C. L. Grosskreutz Hsp27 Phosphorylation in Experimental Glaucoma Invest. Ophthalmol. Vis. Sci., September 1, 2007; 48(9): 4129 - 4135. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. Chen, H.-Y. Zhang, K. Pavlish, and J. N. Benoit Effects of chronic portal hypertension on small heat-shock proteins in mesenteric arteries Am J Physiol Gastrointest Liver Physiol, April 1, 2005; 288(4): G616 - G620. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Chen, A.-P. Arrigo, and R. W. Currie Heat shock treatment suppresses angiotensin II-induced activation of NF-{kappa}B pathway and heart inflammation: a role for IKK depletion by heat shock? Am J Physiol Heart Circ Physiol, September 1, 2004; 287(3): H1104 - H1114. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. J. Fernandes, R. W. Mitchell, O. Lakser, M. Dowell, A. G. Stewart, and J. Solway Invited Review: Do inflammatory mediators influence the contribution of airway smooth muscle contraction to airway hyperresponsiveness in asthma? J Appl Physiol, August 1, 2003; 95(2): 844 - 853. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Hypertension Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |