Introduction
Alzheimer’s disease (AD) is a progressive neurodegenerative disorder [
1] characterized by the accumulation of intracellular neurofibrillary tangles and extracellular senile plaques of which the major component is the amyloid β peptide (Aβ) [
2,
3]. Although the molecular mechanisms leading to neuronal damage in AD have not been completely understood, it is well established that increased production of Aβ, in soluble and/or aggregate form, is a key causative event for AD [
3,
4].
A growing body of evidence has indicated that the cerebral vasculature is an important target of Aβ and that vascular dysfunction significantly contributes to neuronal damage and dementia [
5,
6]. AD patients have reduced cerebral blood flow. This precedes dementia and may contribute to its progression. Recently, it has been shown that endothelin-1 is elevated in Alzheimer’s disease and upregulated by amyloid β [
7]. Cardiovascular risk factors, especially hypertension, have been associated with higher risk of developing Alzheimer’s disease, partially through cerebral vasculature impairment and reduced nitric oxide production [
8]. The brains of patients with AD exhibit elevated levels of ACE, Ang-II, and angiotensin II receptors (AR-II) [
9].
In addition, decreased endothelium-derived nitric oxide (NO) bioavailability and vascular dysfunction have been demonstrated in AD [
10-
15].
NO production at the endothelial cell level involves the activity of the enzyme endothelial nitric oxide synthase (eNOS, NOS III), which is constitutively expressed and produces NO in a calcium-dependent manner [
16]. In models of chronic brain hypoperfusion,
in vivo administration of Aβ has been shown to increase the expression of eNOS and, paradoxically, to decrease endothelium-derived NO formation [
8], thus, suggesting that Aβ could affect the activity of this enzyme. eNOS post-translational modification, including phosphorylation at specific amino acid residues, can profoundly affect its activity and, therefore, influence NO production [
17]. Furthermore, eNOS association with a specific set of interacting proteins has been shown to critically regulate its enzymatic activity by exerting both stimulatory and/or inhibitory effects [
18-
21]. In particular, the chaperone molecule heat shock protein 90 (HSP90) has been demonstrated to possess a key stimulatory role by maintaining the enzyme in an active conformational state and by facilitating its phosphorylation at serine 1177/1179 [
22-
24]. Aβ has been shown to inhibit eNOS phosphorylation at serine 1177/1179 and at other residues [
25,
26], however, no information is available about the effects of Aβ on eNOS interaction with HSP90 or other regulatory partners, which could potentially contribute to these inhibitory effects. In addition, increased production of reactive oxygen species (ROS) and consequent oxidative stress have been shown to negatively influence eNOS activity and significantly contribute to vascular dysfunction in a number of cardiovascular diseases including diabetes and hypertension [
23-
30]. Aβ-induced oxidative stress has been extensively documented [
31-
33]; however, its direct contribution to the reported effects in inhibiting eNOS-dependent NO production or in influencing its interaction with regulatory proteins is not clear.
In this study, we show that in bovine aortic endothelial cells soluble Aβ1–42 promotes the constitutive association of HSP90 with eNOS. This effect resulted in blockade of agonist-mediated phosphorylation of Akt and eNOS at serine 1179. These effects are correlated with Aβ’s ability to increase the production of hydroxyl radicals in endothelial cells and are reverted by concomitant treatment with the antioxidant N-acetyl-cysteine.
Materials and methods
Materials
All tissue culture reagents were from Invitrogen (Carlsbad, CA, USA), unless otherwise specified. Fetal bovine serum (FBS) was from Gemini Bio-products (Woodland, CA, USA). Aβ25–35, Aβ35–25, Aβ1–42, and Aβ42–1 peptides, as well as nitro-L-arginine methyl ester (L-NAME), were from Sigma-Aldrich (St. Louis, MO, USA). Monoclonal and polyclonal anti-eNOS and anti-HSP90 antibodies were from BD-Transduction Laboratories (San Diego, CA, USA). The polyclonal antibody for phospho-eNOS (Ser 1179) was from Invitrogen (Grand Island, NY, USA). Polyclonal and monoclonal antibodies for anti-Akt and phospho-Akt (Ser473) were purchased from Cell Signaling (Danvers, MA, USA). Protein A/G agarose beads were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The ECL chemiluminescence detection assay as well as the peroxidase-conjugated anti-mouse/anti-rabbit IgG were from Amersham Biosciences (Piscataway, NJ, USA). The cGMP enzyme-immunoassay kit was from Cayman Chemical Co. (Ann Arbor, MI, USA). The blots were reprobed after stripping using a stripping solution from Pierce (Rockford, IL, USA).
Cell cultures and treatments conditions
Primary cultures of bovine aortic endothelial cells (BAEC) and brain endothelial cells (BBEC) were purchased from VEC Technologies (Rensselaer, NY, USA) or the endothelial cell Core Facility of the Vascular Biology Center at Georgia Regents University. BAEC were cultured in medium M199 in the presence of 10% FBS, 1% glutamine, 1% penicillin/streptomycin, and 1% non-essential amino acid and vitamin mixtures. The cells were used between passages 3 and 5. Oligomeric preparations of Aβ peptides (25–35, 35–25, 1–42, or 42–1) were prepared by re-suspension in serum-free medium, left overnight at room temperature, then sonicated before supplementation to the culture medium as described previously [
34].
The cells were pretreated for 24 h with 1 μM Aβ25–35, 1 μM Aβ35–25, 5 μM Aβ1–42, and 5 μM Aβ42–1. These doses of Aβ and the time of exposure were chosen based on preliminary experiments testing maximal effects (data not shown). After pre-incubation the cells were stimulated with 20 ng/ml vascular endothelial growth factor (VEGF) for established time points. Rat aortic smooth muscle cells (RASMC) were also obtained from VEC Technologies and used between passages 2 to 5 for the cGMP experiments. The RASMC cultures were maintained in DMEM (Invitrogen, Grand Island, NY, USA) containing 10% FBS and 1% penicillin/streptomycin.
Nitric oxide production
Nitric oxide release in BAEC was measured by cGMP reporter cell assay following a protocol modified from Ishii
et al. [
35,
36]. Confluent BAEC and RASMC were serum-starved for 16 to 18 h and then incubated in Locke’s buffer (154 mM NaCl, 5.6 mM KCl, 2 mM CaCl2, 1 mM MgCl2, and 10 mM Hepes, pH7.4) for at least 20 min before the experiment. The BAECs were then stimulated with 20 ng/ml VEGF for 30 min and, after stimulation, the bathing medium was rapidly transferred to the confluent RASMC, let incubate for 3 min, then the cells were lysed in ice-cold 20-mM sodium acetate, pH 4.0. The cGMP concentration in RASMC treated with the BAEC conditioned medium was determined using an enzyme immunoassay kit following the manufacturer’s instructions.
Determination of reactive oxygen species formation
The production of reactive oxygen species in endothelial cells, following the different treatments, was determined by a fluorimetric assay using 2,7-dihydrochlorofluorescein (DHCF) from Invitrogen (Grand Island, NY, USA). BAEC, and BBEC plated in 96-well plates, were cultured for 24 h in serum-free condition and in presence or absence of Aβ. After this time, the cells were incubated with 5 μM DHCF for 1 h. The hydroxyl radicals produced in response to the different treatments oxidize the DHCF and generate its oxidation fluorescent product dichlorofluorescein (DCF). The fluorescence intensity was measured with a multi-detection microplate reader (BioTek, Winooski, VT, USA) using an excitation and emission light at 485 nm and 535 nM, respectively. Some samples were also incubated in the presence of 50 U of cell permeable superoxide dismutase (PEG-SOD from Sigma-Aldrich, Saint Louis, MO, USA), which was used as negative control to assess the specificity of the assay.
Immunoprecipitation and Western blotting
Western blotting analysis was performed as previously described [
37,
38]. Briefly, BAEC and BBEC were lysed with a modified RIPA buffer (20 mM Tris–HCl (pH 7.4), 2.5 mM EDTA, 50 mM NaF, 10 mM Na
4P
2O
7, 1% Triton X-100, 0.1% sodium dodecyl sulphate, 1% sodium deoxycholate,1 mM PMSF, and 2 mM Na
3VO
4). The samples were then subjected to SDS-PAGE electrophoresis and immunoblotted on nitrocellulose (Schleicher & Schuell Biosciences Inc., Keene, NH, USA). Protein-protein interaction was determined by immunoprecipitation analysis performed as previously described [
37,
38]. For these experiments, the cells were lysed with a buffer containing: 25 mM Tris–HCl (pH 7.4), 2.5 mM EDTA, 50 mM NaF, 10 mM Na
4P
2O
7, 1% Triton X-100, 1 mM PMSF, and 2 mM Na
3VO
4. The cell lysates were allowed to react overnight with the primary antibody and the immunocomplexes were then precipitated with pre-cleared protein A/G agarose beads. After washing three times with ice-cold washing buffer (0.1% Triton X-100 in TBS) the beads and the immunocomplexes were precipitated by centrifugation, solubilized by resuspension in 2X SDS-sample buffer and by boiling at 100°C for 5 min. All the densitometry units have been normalized against total enzyme for each lane and are expressed as the ratio of phosphorylated proteins to total.
Discussion
Endothelial production of nitric oxide is critical to the maintenance of vascular tone and the blood brain barrier, as well as to the anti-inflammatory and anti-thrombotic properties of the vascular endothelium [
27]. Altered eNOS-dependent NO production results in endothelial dysfunction, which is associated with cardiovascular disease and has been shown to play a pathogenic role in AD [
7,
27].
In this study, we demonstrated that exogenous administration of the biologically active fragments Aβ
25–35 and Aβ
1–42 to cultured endothelial cells results in the production of reactive oxygen species (Figure
3), blockade of agonist-stimulated phosphorylation of eNOS at serine 1179 and decreased NO production (Figure
1). These effects correlated with Aβ-induced promotion a constitutive interaction of eNOS with its regulatory partner HSP90 (Figure
2). This correlated with a decrease in the interaction of HSP90 with Akt, the kinase responsible eNOS phosphorylation at Set1179 (Figure
9). These observed alterations appear to be the result of oxidative damage because the concomitant treatment of the cells with the antioxidant N-acetyl cysteine reverts Aβ effects restoring a normal pattern of eNOS phosphorylation at serine 1179 and interaction with HSP-90 (Figures
4 and
8). Our data demonstrated for the first time an inverse effect of Aβ on HSP90/eNOS interaction and HSP90/Akt interaction. We confirmed that upon stimulation with VEGF, Akt phosphorylates eNOS in an HSP90-dependent manner; synergistically increasing eNOS activity (Figures
1,
6, and
7). Once Aβ is introduced to these conditions, there is a diminished interaction between HSP90 and Akt, suggesting that the chaperone remains bound to eNOS in a manner that facilitates its inability to sufficiently bind Akt to signal its phosphorylation at Ser473 (Figures
9 and
10). Although the partial loss of this interaction could occur from other unexplored mechanisms, such as loss of the overall HSP90 protein level previously demonstrated as an effect of Aβ stimulation [
43], it is reasonable to conclude that blocking the effects of Aβ can restore this interaction. In fact, when treated with NAC, HSP90/Akt interaction returns to basal levels and the detection of phospho-Akt (Ser
473) indicate that this kinase overcomes the effects of Aβ in the presence of an antioxidant (Figures
9 and
10).
The results of our study strongly support the growing evidence that vascular dysfunction is involved in the etiology of AD dementia. Indeed, severe vascular changes, such as microvascular degeneration and breakdown of the blood brain barrier, have been demonstrated in AD patients [
44]. In addition, Aβ can be found both in the brain parenchyma and in the cerebral vasculature of AD patients together with increased presence of monocytes/macrophages in the vessel wall and activated microglial cells in the parenchyma [
45,
46].
To date, attention has been focused on the molecular mechanisms by which Aβ and/or its fragments exert their effect on cells. The assumption that amyloid deposits are typically extracellular accounts for an effect at the level of the plasma membrane, possibly receptor-mediated. However, intracellular Aβ accumulation has been observed in various cell types, including neurons and endothelial cells [
47-
49]. In particular, it is not clear whether the effect of Aβ on eNOS activity initiates with the extracellular Aβ deposition or results from Aβ intracellular accumulation.
In this respect, reports have shown that internalized Aβ prevents eNOS from utilizing NADPH [
49], a co-factor required for the activity of the enzyme. Furthermore, other studies have shown that over-expression of Aβ in endothelial cells or administration of exogenous (extracellular) Aβ on isolated vessels, blunts agonist-mediated eNOS activation and its phosphorylation at serine 1177/1179 [
25,
26]. Although these studies provide critical information, it is not clear if the aberrant intracellular production of Aβ peptides is a primary event in endothelial cells or if acute administration of extracellular Aβ represents the condition seen in AD patients that are chronically exposed to Aβ. It is tempting to speculate that both observations may be considered true and complementary. In particular, acute oxidative-independent effects of intra- or extracellular Aβ on eNOS activity may be transient, but could result in elevation of reactive oxygen species and in the oxidative-dependent chronic effects that we observe after 24 h of exposure.
Interestingly, our data show that Aβ promotes eNOS/HSP90 complex formation in a constitutive manner, providing a novel and important information. Previous work demonstrated that disruption of HSP90 interaction with eNOS results in inhibition of this enzymatic activity [
22,
50], therefore, one could expect that in the condition of vascular dysfunction the interaction of eNOS with its chaperone would be inhibited. However, this does not appear to be the case, as has been also shown in a model of diabetes [
51]. Several reports have demonstrated enhanced HSP90 expression as part of the cellular response to oxidative stress conditions [
52-
54]. Specifically, the pathology of AD is characterized by enhanced heat shock proteins expression and activity in response to aberrant ‘misfolded’ proteins [
55]. Of interest, HSP90 expression is induced by Aβ in glial cells which participate in the Aβ clearance process [
56]. Our results demonstrate that Aβ might induce similar effects in endothelial cells. Moreover, HSP90 has been demonstrated to inhibit O
2− production by NOS isoforms [
57,
58], thus the enhancement of HSP90 interaction with eNOS, as shown by our results (Figure
2), could be a compensatory response to maintain the enzyme in a functional state in oxidative stress conditions when the risk of eNOS uncoupling is higher. However, this fails to occur, suggesting that other mechanisms are involved in this process and may be critical to the ability of HSP90 to regulate eNOS activity.
The evidence provided by our study extends and supports previous work demonstrating that Aβ promotes vascular dysfunction by impairing eNOS-dependent NO production. Furthermore, our study underscores the role of oxidative stress in the dysfunctional effects of Aβ on eNOS activity and function.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
FL, VM, TP, MG, and AMM carried out the cultures of ECs and Western blotting detection studies and participated in drafting methods and results. MBH and MCV participated in the design of the study and performed the statistical analysis. MC, CM, MB, and VM conceived of the study, participated in its design and coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.