AJP - Heart Fuel your research with LabChart
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 291: H260-H268, 2006. First published February 17, 2006; doi:10.1152/ajpheart.01027.2005
0363-6135/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
291/1/H260    most recent
01027.2005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yetik-Anacak, G.
Right arrow Articles by Catravas, J. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yetik-Anacak, G.
Right arrow Articles by Catravas, J. D.

Effects of hsp90 binding inhibitors on sGC-mediated vascular relaxation

Gunay Yetik-Anacak,1 Tian Xia,1 Christiana Dimitropoulou,2 Richard C. Venema,1,3 and John D. Catravas1,2

1Vascular Biology Center, 2Department of Pharmacology and Toxicology, and 3Department of Pediatrics, Medical College of Georgia, Augusta, Georgia

Submitted 28 September 2005 ; accepted in final form 26 January 2006


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Vascular soluble guanylate cyclase (sGC) exists in multimeric complexes with endothelial nitric oxide (NO) synthase (eNOS) and heat shock protein 90 (hsp90). Whereas disruption of hsp90-eNOS complexes clearly attenuates eNOS-dependent vascular relaxation, the contribution of sGC-hsp90 complexes to eNOS- or NO donor-dependent relaxations remains unclear. Isolated rat thoracic aortic rings were preincubated with structurally diverse hsp90 binding inhibitors, radicicol (RA) or geldanamycin (GA), or vehicle for 0.5, 1, or 15 h. Preconstricted vessels were exposed to ACh, 8-bromo-cGMP (8-BrcGMP), forskolin, or one of three NO donors: nitroglycerin (NTG), sodium nitroprusside, or spermine NONOate (SNN). Both RA and GA inhibited endothelium-dependent relaxations dose dependently. Indomethacin or the antioxidant tiron did not affect the inhibition of ACh-induced relaxations by GA. Long-term (15 h) exposure to RA inhibited all NO donor-induced relaxations; however, GA inhibited SNN-induced relaxation only. The effects of GA and RA appeared to be selective because 15-h treatment with either agent did not affect forskolin-induced relaxations and only slightly decreased 8-BrcGMP-induced relaxations. Similarly to their effects on NO-donor-induced relaxation, 15-h exposure to RA, but not to GA, decreased hsp90-bound sGC protein expression and NTG-stimulated cGMP formation in aortic rings, whereas RA more than GA reduced SNN-stimulated cGMP formation. We conclude that RA, much more so than GA, selectively inhibits sGC-dependent relaxations of aortic rings by reducing sGC expression, disrupting sGC-hsp90 complex formation and decreasing cGMP formation. These studies suggest that hsp90 regulates both eNOS- and sGC-dependent relaxations.

geldanamycin; radicicol; heat shock protein 90; soluble guanylate cyclase; nitric oxide donor; vasorelaxation

NITRIC OXIDE (NO) is a key mediator of many physiological processes, particularly in the cardiovascular and central nervous systems, including smooth muscle relaxation, neurotransmission, inhibition of platelet aggregation, and immune responses. Soluble guanylate cyclase (sGC), the most common physiological receptor of NO, is the target of various pharmacological compounds currently used in the treatment of cardiovascular disorders.

The association between heat shock protein 90 (hsp90) and endothelial NO synthase (eNOS) was first described by Sessa and colleagues (9). Recently, we demonstrated that sGC also exists in multimeric complexes with NOS and hsp90 (30). Geldanamycin (GA), an ansamycin antibiotic, binds to the ATP binding site of hsp90 and competes for the association of hsp90 with client proteins. GA attenuates cGMP production not only in cultured endothelial and smooth muscle cells stimulated by sodium nitroprusside (SNP), LPS, IL-1beta, or estrogen (15, 22, 30, 31) but also in unstimulated arterial rings (13). A decrease in the amount of hsp90-bound eNOS is associated with a decrease in NO production. Although there is general agreement that inhibition of hsp90 binding by GA reduces ACh-induced vasorelaxation (9, 27), its effect on sGC-mediated relaxation remains controversial (9, 18, 24, 25). Therefore, in this study we employed two structurally diverse hsp90 binding inhibitors, GA and radicicol (RA), over varying times of exposure and investigated the role of sGC/hsp90 complexes on three NO donor-induced relaxations.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Materials. U-46619 and spermine NONOate (SNN) were purchased from Calbiochem (San Diego, CA). Phenylephrine (Phe), ACh, nitroglycerin (NTG), SNP, RA, and 8-bromo-cGMP (8-BrcGMP) were purchased from Sigma Chemical (St. Louis, MO). GA was obtained from NCI. Protein A/G agarose was obtained from Santa Cruz Biotechnology, and supersignal chemiluminescence substrate was from Pierce. Anti-sGC-beta1 antibodies (catalog no. 160897) were purchased from Cayman Chemical, and anti-hsp90 antibodies (catalog no. 610419) were from BD Biosciences (San Diego, CA).

Vascular reactivity. All animal experiments were approved by the Institution’s Committee on Animal Use in Research and Education and adhered to the principles adopted by the American Physiological Society. Male Sprague-Dawley rats (250–300 g) were euthanized by exposure to CO2, followed by thoracotomy. The thoracic aorta was quickly removed and cleaned in physiological salt solution of the following composition (in mM): 118 NaCl, 4.7 KCl, 1.21 MgSO4·7H2O, 1.92 CaCl2, 1.18 KH2PO4, 25 NaHCO3, and 5.5 dextrose. The aorta was cut into 4-mm rings, mounted in a muscle bath in Krebs buffer (pH 7.4, 37°C) containing 10–5 M indomethacin (unless otherwise indicated), and bubbled with 95% O2-5% CO2. Isometric force generation was recorded with a Multi Myograph System (Danish Myo Technology A/S). A resting tension of 2 g was imposed on each ring, and the rings were allowed to equilibrate for 90 min. The presence of a functional endothelium was confirmed by the ability of ACh (10–4 M) to produce relaxation of tissues precontracted with Phe (3 x 10–5 M). For short-term incubation experiments, aortic rings were exposed to one of two structurally diverse hsp90 binding inhibitors, RA or GA, or vehicle (DMSO) for 0.5 or 1 h before the addition of NO donor. To test the long-term effect of hsp90 binding inhibitors on NO donor-induced relaxation, arterial rings were incubated in Krebs buffer for 15 h with vehicle or one of the two hsp90 binding inhibitors. Cumulative concentration-response curves to each of the three NO donors, SNP (10–9–10–5 M), SNN (10–9–10–4 M), and NTG (10–9 –10–4 M) were elicited in rings preconstricted with the thromboxane agonist U-46619 (3 x 10–8 M). Forskolin (10–5 M), ACh (10–9–10–4 M), or 8-BrcGMP (10–5–10–3 M)-induced relaxations were obtained in Phe-preconstricted rings.

Immunoprecipitation and Western blotting. Isolated rat thoracic aortas were frozen in liquid nitrogen and later thawed and homogenized in 300 µl lysis buffer [50 mM Tris·HCl, 0.1 mM EGTA, 0.1 mM EDTA, 100 mM leupeptin, 1 mM phenylmethylsulfonyl fluoride, 1% (vol/vol) Igepal, 0.1% SDS, and 0.1% deoxycholic acid; pH 7.5] as previously described (24). Homogenates were centrifuged at 14,000 g for 20 min at 4°C, and an aliquot from the supernatant was collected for protein concentration analysis using the Lowry method. Another aliquot corresponding to 15 µg of protein was used for electrophoresis. For hsp90 immunoprecipitation, 75 µg of detergent-soluble protein were incubated with excess hsp90 antibody overnight after the sample was precleared with 20 µl protein A/G-Sepharose agarose beads. Immune complexes were precipitated by the addition of 50 µl protein A/G-Sepharose and washed five times, and proteins were eluted from the beads by boiling the samples in SDS sample buffer for 5 min. Agarose beads were pelleted by centrifugation, and protein supernatants were analyzed on SDS-polyacrylamide gels. Proteins were transferred onto a nitrocellulose membrane. The membrane was blocked in 5% milk and incubated with either anti-hsp90 (1:1,000) or anti-sGC antibodies (1:500) overnight at 4°C. After the membrane was washed with 5% milk in Tris-buffered saline containing Tween 20, the blots were probed for 45 min at room temperature with secondary antibodies conjugated to horseradish peroxidase, reacted with luminol reagent, exposed to film, scanned, and quantified.

Determination of cGMP levels. Rat thoracic aortic rings were treated with RA, GA, or vehicle for 15 h in Krebs buffer. After being washed for 15 min in Krebs buffer, rings were exposed to NTG or SNN (10–5 M) or vehicle for 30 min in the presence of the phosphodiesterase inhibitor IBMX (1 mM). At the end of the incubation period, the wet tissue weight was measured, and the rings were snap-frozen in liquid nitrogen. Tissues were pulverized with a steel mortar and pestle prechilled in liquid nitrogen, and 10x volume of 0.1 N HCl was added to extract cGMP. Thirty minutes later, the HCl extract was collected by centrifugation at 800 rpm for 15 min, and cGMP was estimated by ELISA (Biomol). Standard stock solutions of cGMP were prepared in 0.1 N HCl, and the absorbance of the solution was monitored spectrophotometrically (Biotech). Standard dilutions (0.8–500 pmol/ml) were made from the stock solution. cGMP levels were expressed as picomoles per milligram of protein.

Statistics. All data are expressed as means ± SE. The relaxation responses are expressed as percentage of the tone induced by Phe or U-46619. Nonlinear curve fitting was performed with GraphPad Prism (version 3.0 GraphPad Software). Results were analyzed by two-way ANOVA, followed by Bonferroni’s test for multiple comparisons. When appropriate, unpaired Student’s t-test was performed to determine the significance in the difference between two mean values. A value of P < 0.05 was considered statistically significant.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Time- and dose-dependent effect of GA and RA on ACh relaxations. To confirm that hsp90 binding inhibitors attenuate eNOS-dependent relaxations, rat thoracic aortic rings were exposed to ACh in the presence or absence of GA or RA. ACh elicited a concentration-dependent relaxation of the intact rat aortic rings, which was sensitive to both GA and RA (Fig. 1). GA (18 µM) inhibited ACh-induced relaxations strongly after as little as 0.5 h exposure (Fig. 1A); after 1 h of exposure, the inhibitory effect of GA on both the affinity and the maximal relaxation of the tissue to ACh was maximal (Fig. 1B, Table 1). Conversely, the RA (27 µM)-induced inhibition of vasorelaxation continued to increase and reached maximal (complete inhibition) by 15 h of exposure (Fig. 1C, Table 1). Thirty-minute exposure of tissues to a lower concentration of either GA (1.8 µM) or RA (2.7 µM) had no effect on ACh-induced relaxation (Fig. 2D).


Figure 1
View larger version (11K):
[in this window]
[in a new window]
  		Fig. 1. ACh-induced relaxation of phenylephrine-precontracted, endothelium-intact rat thoracic aortic rings, incubated for 0.5 h (A), 1 h (B), or 15 h (C) with geldanamycin (GA; 18 µM), radicicol (RA; 27 µM), or vehicle. *P < 0.05, **P < 0.01, ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle. D: time-dependent inhibition of maximal relaxation to ACh. ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with 30-min exposure to RA.

 

View this table:
[in this window]
[in a new window]
  		Table 1. Emax and pD2 values of GA and RA on ACh-induced relaxations

 

Figure 2
View larger version (13K):
[in this window]
[in a new window]
  		Fig. 2. ACh-induced relaxation of phenylephrine-precontracted endothelium intact rat thoracic aortic rings incubated with GA (18 µM), RA (27 µM), or vehicle in presence or absence of indomethacin (0.5-h incubation with GA or RA) (A); in presence or absence of indomethacin (1-h incubation with GA or RA) (B); or in presence or absence of tiron (0.5-h incubation with GA) (C). ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle. + P < 0.05, +++P < 0.0001, by Student’s t-test compared with vehicle. D: effects of low concentrations of hsp90 binding inhibitors on ACh-induced relaxations. Phenylephrine-precontracted, endothelium-intact rat thoracic aortic rings were incubated for 0.5 h with 1.8 µM GA or 2.7 µM RA.

 
Effect of tiron and indomethacin on ACh relaxations. To examine the role of cyclooxygenase (COX) products and of superoxide on the inhibitory properties of GA and RA, experiments were performed in the presence and absence of the COX inhibitor indomethacin (10 µM) or the superoxide scavenger tiron (30 mM). Neither compound affected the inhibition of ACh-induced vasodilation by GA or RA (Fig. 2, Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Effects of the cyclooxygenase inhibitor indomethacin and the superoxide scavenger tiron on Emax and pD2 values of GA and RA on ACh-induced relaxations

 
Effects of hsp90 binding inhibitors on SNP-induced relaxations. To investigate whether hsp90 binding inhibitors affect nitrovasodilator-induced relaxations, rat thoracic aortic rings were exposed to SNP. SNP relaxed precontracted rings in a concentration-dependent manner (Fig. 3). Short-term incubation with GA or RA (0.5 or 1 h) did not inhibit maximal relaxation or sensitivity to SNP, whereas long-term incubation with RA strongly inhibited SNP-induced relaxations, as reflected in a rightward shift of the concentration-response curve (CRC; Fig. 3C) and in reductions in both the maximal relaxation and sensitivity compared with vehicle (Table 3). One hour treatment with GA moderately increased the relaxation to SNP, as seen in a small leftward shift in the CRC to SNP (Fig. 3B) but did not affect Emax or pD2 values. Fifteen-hour exposure to GA did not significantly alter the CRC to SNP (Fig. 3C) and did not produce significant effects on pD2 values or maximal relaxations (Table 3).


Figure 3
View larger version (10K):
[in this window]
[in a new window]
  		Fig. 3. Sodium nitroprusside (SNP)-induced vasorelaxation of U-46619-precontracted rat thoracic aortic rings incubated for 0.5 h (A), 1 h (B), or 15 h (C) with GA (18 µM), RA (27 µM), or vehicle. D: time-dependent inhibition of maximal relaxation to SNP. *P < 0.05, **P < 0.01, ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Emax and pD2 values of GA and RA on SNP-induced relaxations

 
Effects of hsp90 binding inhibitors on SNN-induced relaxations. SNN strongly relaxed aortic vessels (Fig. 4), and RA inhibited SNN-induced relaxations significantly after 1- and 15-h exposure, as shown by the rightward shift in the CRC (Fig. 4, B and C) and the decreased sensitivity to SNN (Table 4). Furthermore, long-term incubation with RA inhibited the maximal relaxation to SNN (Table 4). Conversely, inhibition of SNN-induced relaxations by GA became apparent only after 15-h treatment (P < 0.05, Table 5) and did not alter the maximal relaxation response.


Figure 4
View larger version (10K):
[in this window]
[in a new window]
  		Fig. 4. Spermine NONOate (SNN)-induced vasorelaxation of U-46619-precontracted rat thoracic aortic rings incubated for 0.5 h (A), 1 h (B), or 15 h (C) with GA (18 µM), RA (27 µM), or vehicle. *P < 0.05, **P < 0.01, ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle. D: time-dependent inhibition of maximal relaxation to SNN. *P < 0.05, 2-way ANOVA followed by Bonferroni post hoc test compared with 30-min exposure to RA. ++P < 0.01 compared with 1-h exposure to RA.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Emax and pD2 value of GA and RA on SNN-induced relaxations

 

View this table:
[in this window]
[in a new window]
  		Table 5. Emax and pD2 values of GA and RA on NTG-induced relaxations

 
Effects of hsp90 binding inhibitors on NTG-induced relaxations. The organic nitrate NTG relaxed precontracted rings in a concentration-dependent manner (Fig. 5). Incubation for 0.5 h with RA or GA did not affect NTG-induced relaxations (Fig. 5A). One-hour exposure to GA induced a leftward, whereas RA induced a rightward shift of the CRC (Fig. 5B), the latter accompanied by significant inhibition of maximal relaxations to NTG (Table 5). Fifteen-hour treatment with RA totally blocked (Fig. 5C) NTG relaxations, as reflected by inhibition of both maximal relaxation and sensitivity to NTG compared with control (Table 5).


Figure 5
View larger version (10K):
[in this window]
[in a new window]
  		Fig. 5. Nitroglycerin (NTG)-induced vasorelaxations of U-46619-precontracted rat thoracic aortic rings incubated for 0.5 h (A), 1 h (B), or 15 h (C) with GA (18 µM), RA (27 µM), or vehicle. *P < 0.05, **P < 0.01, ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with control. D: time-dependent inhibition of maximal relaxation to NTG. **P < 0.01, 2-way ANOVA followed by Bonferroni post hoc test compared with 30-min exposure to RA. ++P < 0.001 compared with 1-h exposure to RA.

 
Effect of hsp90 binding inhibitors on cAMP- and cGMP-dependent relaxations. To investigate whether the effects of hsp90 binding inhibitors on vascular relaxation are specific to the NO/sGC/cGMP pathway, we studied relaxations induced by the cGMP analog 8-BrcGMP and the adenylate cyclase activator forskolin in the presence of hsp90 binding inhibitors. Both 8-BrcGMP and forskolin elicited relaxation responses in control vessels. Fifteen-hour treatment with either hsp90 binding inhibitor did not affect forskolin-induced relaxation (Fig. 6A, Table 6). RA and GA decreased 8-BrcGMP-induced relaxations very slightly (Fig. 6B), and only RA moderately attenuated the maximal relaxation relative to control (Table 6).


Figure 6
View larger version (7K):
[in this window]
[in a new window]
  		Fig. 6. Effects of 15 h incubation with GA (18 µM) or RA (27 µM) on forskolin-induced (A) and 8-bromo-cGMP (8-BrcGMP)-induced relaxation (B) of phenylephrine-precontracted rat thoracic aortic rings. *P < 0.05, **P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle.

 

View this table:
[in this window]
[in a new window]
  		Table 6. Emax values of GA and RA on forskolin (10–5 M)- and 8-BrcGMP-induced vasorelaxation of phenylephrine-precontracted rings

 
Effects of hsp90 binding inhibitors on sGC and hsp90 expression. To analyze hsp90-sGC complex formation, we immunoprecipitated whole vessel lysates with anti-hsp90 antibody and immunoblotted with anti-sGC antibody. Only RA reduced the sGC-hsp90 association (Fig. 7A). Fifteen-hour exposure of vessel rings to either GA or RA attenuated the total amount of sGC expression in whole vessel lysates, with RA producing the more pronounced effect. Additionally, RA but not GA decreased hsp90 expression (Fig. 7B).


Figure 7
View larger version (32K):
[in this window]
[in a new window]
  		Fig. 7. Soluble guanylate cyclase (sGC) and heat shock protein 90 (hsp90) protein expression assessed by Western blotting of lysates immunoprecipitated with hsp90 antibodies (sGC levels were normalized to hsp90 level by dividing density of sGC band to hsp90 band) (A) and total lysates of endothelium-intact rat thoracic aortic vessels treated for 15 h with vehicle, GA (18 µM), or RA (27 µM) (B). *P < 0.05, ***P < 0.001, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle, n = 4–7 rings. IP, immunoprecipitate; IB, immunoblot.

 
Effects of hsp90 binding inhibitors on tissue cGMP levels. Thirty-minute exposure to either SNN or NTG (10–5 M) stimulated cGMP formation significantly. Fifteen-hour treatment with GA (18 µM) did not produce a significant decrease in cGMP formation in NTG-stimulated rings but reduced cGMP-formation in SNN-stimulated rings. On the other hand, RA (27 µM) dramatically reduced cGMP accumulation in both NTG- and SNN-stimulated rings (Fig. 8).


Figure 8
View larger version (15K):
[in this window]
[in a new window]
  		Fig. 8. Effects of hsp90 inhibitors on NTG-stimulated (A) and SNN-stimulated (B) cGMP accumulation in rat thoracic aortic rings. Vessels were pretreated with either vehicle or an hsp90 binding inhibitor (GA or RA) for 15 h, then exposed for 30 min to 10–5 M NTG or SNN. Vessel ring HCl extract was assayed for cGMP by enzyme immunoassay. GA (18 µM) decreased SNN- but not NTG-stimulated cGMP accumulation, but RA (27 µM) dramatically reduced both SNN- and NTG-induced cGMP formation. *P < 0.05, 2-way ANOVA followed by Bonferroni post hoc test compared with vehicle; #P < 0.05 compared with vehicle + SNN or vehicle + NTG; n = 7 rings.

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
Recently, we reported that sGC exists in multimeric complexes with NOS and hsp90 and that the hsp90 binding inhibitor GA attenuated both the formation of the complex and SNP-induced cGMP production in either endothelial or smooth muscle cells (30). Thus we hypothesized that sGC-hsp90 complexes may also increase the responsiveness of sGC to exogenous NO and that blocking hsp90 binding to sGC would inhibit sGC-dependent relaxations in vascular segments ex vivo. It appears that the effects of GA and RA are more complex in intact tissue than those observed in isolated cells.

Time- and dose-dependent effect of GA and RA on ACh relaxations. Both GA and RA produced a robust inhibitory effect on ACh-induced relaxation, in agreement with previous studies showing that 15-min incubation with GA (18 µM) inhibited endothelium-dependent relaxations of rat aorta (9). Although several studies have already shown that GA inhibits eNOS-dependent relaxations (9, 11, 18, 24, 25), we now demonstrate that this effect is time dependent because the inhibitory effects of RA, primarily, but also, to a lesser extent, GA, increased with incubation time and became maximal after a 15-h exposure. There are conflicting data on the inhibitory effects of lower concentrations of GA. It has been shown that 15-min incubation with 3 µg/ml (5.3 µM) GA inhibits ACh relaxation in perfused rat mesenteric vessels (24, 25) but not in canine basilar arteries (13). We observed that the inhibitory effect of GA and RA is dose dependent, as 0.5 h treatment with low concentration of these antifungal antibiotics does not alter eNOS-dependent relaxation, whereas higher concentrations do. Furthermore, both hsp90 binding inhibitors decreased not only maximal relaxation but also the sensitivity of the endothelium to ACh. The mechanism of GA- and RA-induced inhibition of ACh-induced relaxation may be more complex and involve inhibition of tyrosine kinase activities (via dissociation of hsp90-tyrosine kinase complexes) and consequent tyrosine kinase-mediated NO release, in addition to the dissociation of eNOS-hsp90 complexes (18).

Role of the COX pathway and superoxide on the inhibition of ACh relaxation by hsp90 binding inhibitors. Receptor-mediated release of endothelium-derived relaxing factor and PGI2 are coupled in bovine aortic endothelial cells (BAEC), and several agonists produce vasodilatation through the simultaneous release of NO and PGI2 (6, 14). Radicicol has been shown to inhibit COX-2 expression in smooth muscle cells (5). In the present study, indomethacin did not alter the inhibitory effect of GA or RA on ACh relaxation, suggesting that inhibition of COX activity does not mediate the reduction of these agents in ACh-induced relaxations of rat thoracic aorta.

Previous reports have questioned the mechanism behind the effect of GA on endothelium-dependent relaxations (4, 7) as this agent can also generate O2–. The possibility that reactive oxygen species produced by GA may be responsible for the attenuated relaxation to ACh can be ruled out because GA effectively blocked the eNOS-dependent relaxations in the presence of the antioxidant tiron. Moreover, another hsp90 binding inhibitor, RA, abolished ACh relaxations even though it does not redox cycle (19).

Effect of GA on sGC-dependent relaxations. The effect of GA on sGC-mediated relaxations of isolated vascular tissue has been controversial, with several reports observing no effect of GA on NO donor-induced vascular relaxation ex vivo (9, 18, 24, 25). We investigated whether the reported failure of GA to inhibit NO donor relaxations of vascular tissue may be influenced by 1) the nature of the NO donor; 2) the exposure time of the tissue to the hsp90 binding inhibitor; or 3) the inhibitor itself, possibly acting on different binding sites for eNOS and sGC on hsp90. To explore these possibilities, we employed two structurally diverse hsp90 binding inhibitors, GA and RA, for short (0.5 and 1 h) and long (15 h) exposure times and on relaxations induced by three different NO donors.

Our finding that 0.5-h exposure to GA (18 µM) does not inhibit NTG-mediated relaxation is consistent with previous studies (9). Besides NTG, SNP- and diethylenetriamine NONOate (DETANONOate)-mediated relaxations are reported not to change by brief exposure to GA (5.3 µM) in mesenteric or basilar arteries (13, 24). This is consistent with our findings that short exposure to GA failed to affect SNP or SNN-mediated relaxations. Conversely, Ou et al. (18) showed that short incubation with 20 µM GA inhibited high DETANONOate (10–5 M) concentration-induced relaxations but did not affect the actions of lower DETANONOate concentrations (18). Perhaps the higher GA concentrations could contribute to the different findings in the Ou et al. study. In the present study, even long-term exposure to GA only slightly decreased SNN-induced relaxations by a statistically significant but probably biologically unimportant magnitude.

Overall, GA exerted minimal inhibition of NO donor-induced relaxation, regardless of the time of exposure or the type of NO donor employed; in contrast, we previously demonstrated that GA produced a significant decrease in SNP-induced cGMP accumulation in cultured bovine aortic endothelial or rat aortic smooth muscle cells (30). The difference between this study and our previous report may reflect the inability of GA to affect sGC-hsp90 complexes in intact vessels. Indeed, coimmunoprecipitation studies demonstrated almost no effect of GA on sGC-hsp90 association in vessel rings (Fig. 7), whereas it completely inhibited this association in cultured vascular cells (30). The reasons behind this difference are not clear; it is possible that smooth muscle cells in intact vessels may see a much lower concentration of GA; transport of GA across the endothelial and adventitial layers or its fate within the vascular wall remains unknown.

Effect of RA on sGC-dependent relaxations. Unlike GA, inhibition of NO donor-induced relaxations by RA increased with exposure time and became highly significant after 15-h exposure, regardless of the type of NO donor used. These data suggest that very short-term (0.5 h) blocking of hsp90 activity may not be enough to effectively disrupt hsp90/sGC complexes and reduce sGC-dependent relaxations. Long-term incubation with RA inhibited NO donor-induced relaxations to different extents, depending on the type of NO donor. Thus RA abolished NTG-induced relaxations, whereas it partially inhibited SNN- and SNP-induced relaxations. However, long-term treatment with RA attenuated the sensitivity to all NO donors. Accordingly, it appears that long-term blockade of hsp90 decreases NO donor-induced relaxations, suggesting that hsp90 optimizes sGC-dependent relaxations in vitro.

Role of the NO/sGC/cGMP pathway in the inhibition of NO donor-induced relaxations. Different NO donors produce vasodilation through varying degrees of activation of sGC (29, 32). The predominant form of NO produced or the mechanism of NO generation (enzymatic or nonenzymatic) affects the degree of sGC activation and determines the heterogeneity in the actions of different NO donors. To better understand the functional importance of the sGC-hsp90 interaction, we investigated the effects of hsp90 inhibitors on vasorelaxation induced by multiple NO donors. We employed NO donors that represent a range of chemical classes and that release additional NO redox forms besides ·NO. SNP is also a NO+ donor; NONOates are predominantly ·NO donors, and NTG is a mixed donor (8). The organic nitrate ester, NTG, and SNP require enzymatic or nonenzymatic bioactivation in the tissue to generate NO, whereas SNN releases NO spontaneously (12, 16, 17). Long-term treatment with RA inhibited relaxation responses to each of the NO donors in differing degrees (NTG > SNN > SNP). These results parallel the different sensitivities of the sGC inhibitor 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ)-mediated inhibition of vascular relaxation by these same three NO donors on rat pulmonary artery and mouse aorta (10, 28, 32).

Hsp90 binding inhibitors had no effect on forskolin-induced relaxation, suggesting that the inhibition of NO donor-induced relaxations by these inhibitors is likely selective to the NO/sGC/cGMP pathway. These experiments also address the concern that decreased relaxation to the NO donors may have resulted, at least in part, from a toxic effect of long-term exposure of tissues to RA because cAMP-induced relaxations were not affected by long-term incubation with RA.

To investigate whether the inhibitory effects of hsp90 binding inhibitors on NO donor-induced relaxations are dependent on inhibition of sGC activity or on inhibition of downstream mechanisms, we examined the effect of GA and RA on 8-BrcGMP-induced relaxations. Long-term exposure to either hsp90 binding inhibitor barely attenuated cGMP-induced relaxations, suggesting that the inhibitory effects of the hsp90 binding inhibitors on exogenous NO are mostly due to disruption of sGC activity.

Effect of hsp90 binding inhibitors on NTG-stimulated cGMP accumulation. We did not observe a decrease in NTG-stimulated cGMP accumulation in rat thoracic aortic rings after 15-h treatment with GA. This agrees with a report that GA does not influence cGMP accumulation in response to SNP stimulation in human umbilical vein endothelial cells (HUVEC) (9). On the other hand, we have reported that GA inhibits SNP-induced cGMP accumulation in early-passage BAEC and rat aortic smooth muscle cells (RASMC) (15, 20). The reasons for this apparent discrepancy are unclear but may relate to the use of diverse systems, i.e., BAEC and RASMC vs. HUVEC and in vivo vs. aortic rings. On the other hand, GA reduced SNN-stimulated cGMP accumulation; this finding agrees with the observation that GA inhibited SNN-induced vasorelaxation. RA (27 µM) dramatically inhibited NO donor (both SNN and NTG)-induced cGMP accumulation to a much higher degree than GA. Again, these results agree with our observed inhibition of NO donor-induced vascular relaxation.

Effects of two hsp90 binding inhibitors on sGC and hsp90 protein levels. Long-term treatment with GA did not alter hsp90 protein levels and induces only a mild decrease in sGC protein expression, whereas RA decreased both hsp90 and sGC levels dramatically. These data are consistent with the effects of GA and RA on NO donor-induced relaxation. Although GA slightly attenuated total sGC expression in vessels, this decline appears insufficient in producing a significant inhibitory effect on NTG- or SNP-induced relaxations but enough to cause a moderate rightward shift in SNN-induced relaxations. RA significantly reduced hsp90-sGC complex formation, suggesting that hsp90-bound sGC is mainly responsible for the decrease in sGC level in lysates. Thus long-term treatment with RA reduced hsp90-sGC complexes and at the same time inhibited sGC (NO donor)-dependent relaxations. However, long-term treatment with GA failed to decrease hsp90-bound sGC levels, produced only weak inhibition of NO donor-induced relaxations, and failed to inhibit NTG-stimulated cGMP accumulation. We therefore conclude that sGC-dependent relaxations are likely regulated by hsp90 and that the capacity of sGC to be activated by NO is dependent on hsp90 activity. Whereas in many cases the effect of hsp90 binding inhibitors is to reduce the activity of the client protein, in other cases, they exert their effects by making client proteins available for degradation (26). Therefore, it is possible that the effects of RA on sGC may be due, at least in part, to increased sGC degradation, suggesting that additionally or alternatively, hsp90 may serve to stabilize sGC as is the case for another protein of the sGC/eNOS/hsp90 complex, Akt (3).

Different effects of GA and RA. The present study demonstrates that two structurally diverse hsp90 binding inhibitors have different effects on sGC-dependent relaxations and hsp90-sGC complex integrity. The difference between the two hsp90 binding inhibitors on disrupting sGC-hsp90 complex and consequent relaxation may result from their different potencies in binding to hsp90; RA binds to hsp90 with a 50-fold greater affinity than GA (21). Consequently, a higher concentration of GA may be necessary to disrupt sGC-hsp90 interaction compared with RA. This hypothesis is supported by a study showing inhibitory effects even after short-term incubation with high-concentration (20 µM) GA (18). The increased efficacy of RA may be a result of its specificity: GA binds to both Grp94 (ER homolog of hsp90) and hsp90 with equal potency, whereas RA binds more selectively to hsp90 (23).

The differing effects of the two hsp90 binding inhibitors on sGC-dependent relaxations may also be related to hsp70; recently it was shown that hsp70 is another sGC-interacting protein, responsible for an sGC-activating effect (2). RA does not bind to hsp70 (23), and conversely GA has been shown to increase hsp70 levels in fibroblasts (1). Thus it is conceivable that the increased NTG- and SNP-induced relaxation by GA may be due to its increasing hsp70 levels.

We also observed that eNOS-hsp90 complexes appear to be more sensitive to hsp90 binding inhibitors than sGC-hsp90 complexes because short exposure to either inhibitor attenuated eNOS-dependent relaxations. It is possible that the binding sites of hsp90 to sGC may be different from those of eNOS to hsp90. Although both of these chemically unrelated compounds bind to the NH2-terminal ATP/ADP-binding domain of hsp90, binding sites of RA and GA differ in details. For example, it has been shown that Gly131 deletion alone does not completely abrogate GA binding to hsp90 but shows markedly reduced binding to RA (23).

In summary, hsp90 likely regulates both eNOS- and sGC-dependent relaxations of aortic vessels. Inhibition of endothelium-dependent relaxations are dose and time dependent and unrelated to superoxide generated by GA or to COX inhibition by RA. Similarly, time-dependent inhibition of NO donor-induced relaxations by RA does not appear to involve a nonspecific or toxic effect and is associated with significant reduction in sGC-hsp90 complex formation and sGC expression.


   GRANTS
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This study was supported by National Heart, Lung, and Blood Institute Grant HL-070412.


   FOOTNOTES
 

Address for reprint requests and other correspondence: J. D. Catravas, Vascular Biology Center, Medical College of Georgia, Augusta, GA 30912–2500 (e-mail: jcatrava{at}mcg.edu)

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.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Bagatell R, Paine-Murrieta GD, Taylor CW, Pulcini EJ, Akinaga S, Benjamin IJ, and Whitesell L. Induction of a heat shock factor 1-dependent stress response alters the cytotoxic activity of hsp90-binding agents. Clin Cancer Res 6: 3312–3318, 2000.[Abstract/Free Full Text]
  2. Balashova N, Chang FJ, Lamothe M, Sun Q, and Beuve AV. Characterization of a novel type of endogenous activator of soluble guanylyl cyclase. J Biol Chem 280: 2186–2196, 2005.[Abstract/Free Full Text]
  3. Basso AD, Solit DB, Chiosis G, Giri B, Tsichlis P, and Rosen N. Akt forms an intracellular complex with heat shock protein 90 (Hsp90) and Cdc37 and is destabilized by inhibitors of Hsp90 function. J Biol Chem 277: 39858–39866, 2002.[Abstract/Free Full Text]
  4. Billecke SS, Draganov DI, Morishima Y, Murphy PJ, Dunbar AY, Pratt WB, and Osawa Y. The role of hsp90 in heme-dependent activation of apo-neuronal nitric-oxide synthase. J Biol Chem 279: 30252–30258, 2004.[Abstract/Free Full Text]
  5. Chanmugam P, Feng L, Liou S, Jang BC, Boudreau M, Yu G, Lee JH, Kwon HJ, Beppu T, Yoshida M, Xia Y, Wilson CB, and Hwang D. Radicicol, a protein tyrosine kinase inhibitor, suppresses the expression of mitogen-inducible cyclooxygenase in macrophages stimulated with lipopolysaccharide and in experimental glomerulonephritis. J Biol Chem 270: 5418–5426, 1995.[Abstract/Free Full Text]
  6. De Nucci G, Gryglewski RJ, Warner TD, and Vane JR. Receptor-mediated release of endothelium-derived relaxing factor and prostacyclin from bovine aortic endothelial cells is coupled. Proc Natl Acad Sci USA 85: 2334–2338, 1988.[Abstract/Free Full Text]
  7. Dikalov S, Landmesser U, and Harrison DG. Geldanamycin leads to superoxide formation by enzymatic and non-enzymatic redox cycling. Implications for studies of Hsp90 and endothelial cell nitric-oxide synthase. J Biol Chem 277: 25480–25485, 2002.[Abstract/Free Full Text]
  8. Feelisch M. The use of nitric oxide donors in pharmacological studies. Naunyn Schmiedebergs Arch Pharmacol 358: 113–122, 1998.[CrossRef][ISI][Medline]
  9. Garcia-Cardena G, Fan R, Shah V, Sorrentino R, Cirino G, Papapetropoulos A, and Sessa WC. Dynamic activation of endothelial nitric oxide synthase by Hsp90. Nature 392: 821–824, 1998.[CrossRef][Medline]
  10. Homer KL, Fiore SA, and Wanstall JC. Inhibition by 1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) of responses to nitric oxide-donors in rat pulmonary artery: influence of the mechanism of nitric oxide generation. J Pharm Pharmacol 51: 135–139, 1999.[CrossRef][ISI][Medline]
  11. Ikezaki H, Akhter SR, Hong D, Suzuki H, Gao XP, and Rubinstein I. Tyrosine kinase inhibitors modulate agonist-induced vasodilation in the hamster cheek pouch. J Appl Physiol 88: 857–862, 2000.[Abstract/Free Full Text]
  12. Keefer LK, Nims RW, Davies KM, and Wink DA. "NONOates" (1-substituted diazen-1-ium-1,2-diolates) as nitric oxide donors: convenient nitric oxide dosage forms. Methods Enzymol 268: 281–293, 1996.[ISI][Medline]
  13. Khurana VG, Feterik K, Springett MJ, Eguchi D, Shah V, and Katusic ZS. Functional interdependence and colocalization of endothelial nitric oxide synthase and heat shock protein 90 in cerebral arteries. J Cereb Blood Flow Metab 20: 1563–1570, 2000.[CrossRef][ISI][Medline]
  14. Lot TY, Stark G, and Wilson VG. Endothelium-dependent contractions to NG-nitro-L-arginine methyl ester in the porcine isolated splenic artery are sensitive to cyclooxygenase and lipoxygenase inhibitors. Naunyn Schmiedebergs Arch Pharmacol 347: 115–118, 1993.[Medline]
  15. Marczin N, Papapetropoulos A, and Catravas JD. Tyrosine kinase inhibitors suppress endotoxin- and IL-1beta-induced NO synthesis in aortic smooth muscle cells. Am J Physiol Heart Circ Physiol 265: H1014–H1018, 1993.[Abstract/Free Full Text]
  16. Murphy ME. Ascorbate and dehydroascorbate modulate nitric oxide-induced vasodilations of rat coronary arteries. J Cardiovasc Pharmacol 34: 295–303, 1999.[CrossRef][ISI][Medline]
  17. Napoli C and Ignarro LJ. Nitric oxide-releasing drugs. Annu Rev Pharmacol Toxicol 43: 97–123, 2003.[Medline]
  18. Ou J, Fontana JT, Ou Z, Jones DW, Ackerman AW, Oldham KT, Yu J, Sessa WC, and Pritchard KA Jr. Heat shock protein 90 and tyrosine kinase regulate eNOS NO· generation but not NO· bioactivity. Am J Physiol Heart Circ Physiol 286: H561–H569, 2004.[Abstract/Free Full Text]
  19. Ou J, Ou Z, Ackerman AW, Oldham KT, and Pritchard KA. Inhibition of heat shock protein 90 (hsp90) in proliferating endothelial cells uncouples endothelial nitric oxide synthase activity. Free Radic Biol Med 34: 269–276, 2003.[CrossRef][ISI][Medline]
  20. Papapetropoulos A, Zhou Z, Gerassimou C, Yetik G, Venema RC, Roussos C, Sessa WC, and Catravas JD. Interaction between hsp90 and soluble guanylyl cyclase: physiological significance and mapping of the domains mediating binding. Mol Pharmacol 68: 1133–1141, 2005.[Abstract/Free Full Text]
  21. Roe SM, Prodromou C, O’Brien R, Ladbury JE, Piper PW, and Pearl LH. Structural basis for inhibition of the Hsp90 molecular chaperone by the antitumor antibiotics radicicol and geldanamycin. J Med Chem 42: 260–266, 1999.[CrossRef][ISI][Medline]
  22. Russell KS, Haynes MP, Caulin-Glaser T, Rosneck J, Sessa WC, and Bender JR. Estrogen stimulates heat shock protein 90 binding to endothelial nitric oxide synthase in human vascular endothelial cells. Effects on calcium sensitivity and NO release. J Biol Chem 275: 5026–5030, 2000.[Abstract/Free Full Text]
  23. Schulte TW, Akinaga S, Murakata T, Agatsuma T, Sugimoto S, Nakano H, Lee YS, Simen BB, Argon Y, Felts S, Toft DO, Neckers LM, and Sharma SV. Interaction of radicicol with members of the heat shock protein 90 family of molecular chaperones. Mol Endocrinol 13: 1435–1448, 1999.[Abstract/Free Full Text]
  24. Shah V, Wiest R, Garcia-Cardena G, Cadelina G, Groszmann RJ, and Sessa WC. Hsp90 regulation of endothelial nitric oxide synthase contributes to vascular control in portal hypertension. Am J Physiol Gastrointest Liver Physiol 277: G463–G468, 1999.[Abstract/Free Full Text]
  25. Shastry S and Joyner MJ. Geldanamycin attenuates NO-mediated dilation in human skin. Am J Physiol Heart Circ Physiol 282: H232–H236, 2002.[Abstract/Free Full Text]
  26. Stancato LF, Silverstein AM, Owens-Grillo JK, Chow YH, Jove R, and Pratt WB. The hsp90-binding antibiotic geldanamycin decreases Raf levels and epidermal growth factor signaling without disrupting formation of signaling complexes or reducing the specific enzymatic activity of Raf kinase. J Biol Chem 272: 4013–4020, 1997.[Abstract/Free Full Text]
  27. Su Y and Block ER. Role of calpain in hypoxic inhibition of nitric oxide synthase activity in pulmonary endothelial cells. Am J Physiol Lung Cell Mol Physiol 278: L1204–L1212, 2000.[Abstract/Free Full Text]
  28. Tseng CML, Tabrizi-Fard MA, and Fung HL. Differential sensitivity among nitric oxide donors toward ODQ-mediated inhibition of vascular relaxation. J Pharmacol Exp Ther 292: 737–742, 2000.[Abstract/Free Full Text]
  29. Van der Zypp A and Majewski H. Effect of cGMP inhibitors on the actions of nitrodilators in rat aorta. Clin Exp Pharmacol Physiol 25: 38–43, 1998.[ISI][Medline]
  30. Venema RC, Venema VJ, Ju H, Harris MB, Snead C, Jilling T, Dimitropoulou C, Maragoudakis ME, and Catravas JD. Novel complexes of guanylate cyclase with heat shock protein 90 and nitric oxide synthase. Am J Physiol Heart Circ Physiol 285: H669–H678, 2003.[Abstract/Free Full Text]
  31. Venema VJ, Marrero MB, and Venema RC. Bradykinin-stimulated protein tyrosine phosphorylation promotes endothelial nitric oxide synthase translocation to the cytoskeleton. Biochem Biophys Res Commun 226: 703–710, 1996.[CrossRef][ISI][Medline]
  32. Wanstall JC, Jeffery TK, Gambino A, Lovren F, and Triggle CR. Vascular smooth muscle relaxation mediated by nitric oxide donors: a comparison with acetylcholine, nitric oxide and nitroxyl ion. Br J Pharmacol 134: 463–472, 2001.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
T. Xia, C. Dimitropoulou, J. Zeng, G. N. Antonova, C. Snead, R. C. Venema, D. Fulton, S. Qian, C. Patterson, A. Papapetropoulos, et al.
Chaperone-dependent E3 ligase CHIP ubiquitinates and mediates proteasomal degradation of soluble guanylyl cyclase
Am J Physiol Heart Circ Physiol, November 1, 2007; 293(5): H3080 - H3087.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Lung Cell. Mol. Physiol.Home page
J. L. Aschner, S. L. Foster, M. Kaplowitz, Y. Zhang, H. Zeng, and C. D. Fike
Heat shock protein 90 modulates endothelial nitric oxide synthase activity and vascular reactivity in the newborn piglet pulmonary circulation
Am J Physiol Lung Cell Mol Physiol, June 1, 2007; 292(6): L1515 - L1525.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
291/1/H260    most recent
01027.2005v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Yetik-Anacak, G.
Right arrow Articles by Catravas, J. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yetik-Anacak, G.
Right arrow Articles by Catravas, J. D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.