SEARCH

SEARCH BY CITATION

Keywords:

  • Alzheimer's disease;
  • beta-amyloid peptides;
  • conformationally altered p53;
  • metallothionein 2A;
  • zyxin

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Zyxin is an adaptor protein recently identified as a novel regulator of the homeodomain-interacting protein kinase 2 (HIPK2)-p53 signaling in response to DNA damage. We recently reported an altered conformational state of p53 in tissues from patients with Alzheimer ‘s disease (AD), because of a deregulation of HIPK2 activity, leading to an impaired and dysfunctional response to stressors. Here, we examined the molecular mechanisms underlying the deregulation of HIPK2 activity in two cellular models, HEK-293 cells and SH-SY5Y neuroblastoma cells differentiated with retinoic acid over-expressing the amyloid precursor protein, focusing on the evidence that zyxin expression is important to maintain HIPK2 protein stability. We demonstrated that both beta-amyloid (Aβ) 1-40 and 1-42 induce zyxin deregulation, thus affecting the transcriptional repressor activity of HIPK2 onto its target promoter, metallothionein 2A, which is in turn responsible for the induction of an altered conformational state of p53. We demonstrate for the first time that zyxin is a novel target of Aβ activities in AD. These results may help the studies on the pathogenesis of AD, through the fine dissection of events related to beta-amyloid activities.

Abbreviations used
βSI

β-secretase inhibitor

γSI

γ-secretase inhibitor

AD

Alzheimer's disease

APP

amyloid precursor protein

beta-amyloid

DMSO

dimethyl sulfoxide

GAPDH

glyceraldehyde 3-phosphate dehydrogenase

HEK 293

human embryonic kidney

HIPK2

homeodomain-interacting protein kinase 2

LDH

lactate dehydrogenase

MT2A

metallothionein 2A

PVDF

polyvinylidene difluoride

TBST

Tris-buffered saline/Tween 20

The protein p53 responds to a variety of cellular stresses and is able to sense the intensity of the damage and modulate biological responses, ranging from transient growth arrest to permanent replicative senescence or apoptosis (Vousden and Prives 2005). One important mechanism that controls p53 function is its conformational stability (Joerger and Fersht 2007). An altered protein conformational state of p53, independent from point mutations, has been reported in tissues from patients with Alzheimer's disease (AD) (Uberti et al. 2006; Lanni et al. 2008; Zhou and Jia 2010). When investigating the mechanism of such alteration, we found that soluble nanomolar concentrations of beta-amyloid (Aβ) 1-40 peptide induced the expression of an unfolded p53 protein isoform and modulated p53 functions by interfering with the homeodomain-interacting protein kinase 2 (HIPK2) (Lanni et al. 2007, 2010), a fundamental protein in maintaining wild-type p53 function (Puca et al. 2008). In particular, soluble Aβ 1-40 inhibited HIPK2 activity, consequently inducting an altered conformational state of p53, and thus resulting in an inability to properly activate an apoptotic program when cells are exposed to a noxious stimulus (Lanni et al. 2010).

A novel regulator of the HIPK2-p53 signaling in response to DNA damage, named zyxin, has been recently identified (Crone et al. 2011). Zyxin is primarily localized at the focal adhesion plaque complex as an adaptor protein (Beckerle 1997; Wang and Gilmore 2003) and contains a proline-rich domain at the N-terminus and three LIM domains at the C-terminus that are cysteine-rich motifs involved in protein–protein interactions. Zyxin is involved in regulating cell adhesion, spreading, and motility but also in transducing signals into the nucleus, to regulate gene expression, cell proliferation, differentiation, and apoptosis (Sadot et al. 2001; Wang and Gilmore 2001). Consistently, zyxin has been demonstrated to shuttle between the cytosol and the nucleus, where it affects transcriptional activity (Degenhardt and Silverstein 2001). In cancer cells, zyxin has been involved in DNA damage-induced cell fate control through the modulation of the HIPK2-p53 signaling. In particular, depletion of endogenous zyxin resulted in proteasome-dependent HIPK2 degradation, thus compromising DNA damage-induced p53 Ser46 phosphorylation dependent from HIPK2 (Crone et al. 2011).

A deregulation of HIPK2 has also been observed in cells treated with Aβ 1-40 (Lanni et al. 2010); however, the molecular mechanisms underlying HIPK2 proteasomal degradation in conditions related to Aβ-exposure need to be investigated. Considering that an altered proportion between Aβ 1-40/Aβ 1-42 as well as an increased production of one of the two species has been indicated as dangerous for the pathogenesis of AD (Verdile et al. 2004; Pangalos et al. 2005; Barten and Albright 2008), our purpose was to first investigate whether a difference in the activation of this pathway exists between the species 1-40 and 1-42, by evaluating the capability of these species to affect the transcriptional repressor activity of HIPK2 onto its target promoter, metallothionein 2A (MT2A). Furthermore, based on data from literature demonstrating that zyxin expression is important in maintaining HIPK2 protein stability (Crone et al. 2011), our second goal was to determine whether a modulation of zyxin is involved in AD pathogenesis, using two cellular models of altered Aβ production. In particular, since in AD an altered metabolism of the amyloid precursor protein (APP) occurred, in turn leading to an aberrant production of Aβ peptides (Verdile et al. 2004), our purpose was also to investigate the effect of Aβ peptides on zyxin mRNA and protein levels. The results showed here may help to better understand the pathogenesis of AD, through the fine dissection of events related to Aβ activities. The characterization of Aβ activity on zyxin-HIPK2 signaling pathway represents a relevant characteristic, since it concerns a research field yet unexplored in a neurodegenerative context.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Reagents

All culture media, Fetal Bovine Serum (FBS), and supplements were purchased from Euroclone (Life Science Division, Milan, Italy). Electrophoresis reagents were acquired from Bio-Rad (Hercules, CA, USA). All other reagents were of the highest grade available and were obtained from Sigma Chemical Co. (St. Louis, MO, USA) unless otherwise indicated. Aβ 1-40, Aβ 1-42, and Aβ 40-1 reverse peptide were solubilized in dimethyl sulfoxide (DMSO) at the concentration of 100 μM and frozen in stock aliquots. For each experimental setting, one stock aliquot was diluted at the final concentration of 10 nM. The Aβ concentration was chosen following dose–response experiments (data not shown), where maximal modulation of p53 structure and its transcriptional activity (Uberti et al. 2007) was obtained at 10 nM. All the experiments performed with Aβ were made in 1% of serum. H2O2 was diluted to working concentration (1 mM) in phosphate buffer saline (PBS) at the moment of use. Mouse monoclonal anti-α-tubulin was purchased from Sigma-Aldrich (St. Louis, MO, USA). Host specific peroxidase conjugated IgG secondary antibodies were obtained from Pierce (Rockford, IL, USA).

Cell cultures

Human embryonic kidney (HEK) 293 cells from European Collection of Cell Cultures (ECACC No. 85120602) were cultured in Eagle's minimum essential medium containing 10% FBS, glutamine (2 mM), penicillin/streptomycin (2 mM), at 37°C in 5%CO2/95% air (Uberti et al. 2007). The HEK-293 cells stably transfected with APP751 were obtained as previously described (Uberti et al. 2007) and maintained in G418 at a final concentration of 400 μg/mL.

Human neuroblastoma SH-SY5Y cell line from European Collection of Cell Cultures (ECACC No. 94030304) were cultured in medium with equal amount of Eagle's minimum essential medium and Nutrient Mixture Ham's F-12, supplemented with 10% fetal bovine serum, glutamine (2 mM), penicillin/streptomycin, non-essential amino acids at 37°C in 5% CO2/95% air. The SH-SY5Y cells were also stably transfected with APP751 and maintained in G418 at a final concentration of 400 μg/mL. For the differentiation, cells were plated at a density of 5 × 103 cells/cm2 in 60-mm diameter culture dishes using cell medium with 10% FBS. From day 1 after plating, cells were differentiated in the presence of 10 μM all trans-retinoic acid in the cell medium containing 1% FBS. The medium was replaced every 2 days for 6 days to achieve cell differentiation. Differentiation was considered completed considering the length of growth cone-terminated neurites versus cell body diameter: the growth cone-terminated neurites of differentiated cells appeared to be up to three times longer than the diameter of the corresponding cell body as compared with non-differentiated cells, where the neurites were shorter than the diameter of the corresponding cell body.

Cell treatment

Cells were used at 80% confluent monolayers. Doxorubicin was added to the medium at the concentration of 50 μmol/L and then cells were cultured for additional 24 h. For H2O2 exposure, culture cells were washed with PBS and treated with 1 mM H2O2 for 10 min. After washing, cells returned to fresh medium for additional time according to the experiments. For each cell line, the experiments were repeated at least three times.

Cell viability

Cell viability was evaluated 24 h after the addition of the cytotoxic agent to the media by measuring lactate dehydrogenase (LDH) activity using Cytoxicity Detection Kit (Boehringer, Mannheim, Germany) and an ELISA reader (340 ATC; SLT LabInstruments, Salzburg, Austria). Cytotoxicity was evaluated as percentage of maximum amount of releasable LDH enzyme activity, which is determined by lysing the cells with 1% of TritonX-100.

RNA extraction and reverse transcription-PCR (RT-PCR)

Total RNA was extracted with TRIzol (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions. The first strand cDNA was synthesized by reverse-transcribing mRNA, as previously described (Lanni et al. 2010). Semiquantitative RT-PCR was performed by using Hot-Master Taq (Eppendorf, Milan, Italy) with 2 μL cDNA reaction and genes specific oligonucleotides under conditions of linear amplification.

Real-Time PCR

For mRNA extraction 2 × 106 cells were used in a 60 mm2 Petri plate. Total RNA was extracted using RNeasy Plus Mini Kit (Qiagen, Valencia, CA, USA) following manufacturer's instructions. QuantiTect reversion transcription kit and QuantiTect Syber Green PCR kit (Qiagen) were used for cDNA synthesis and gene expression analysis following manufacturer's specifications. QuantiTect primer assay for MT2A and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were provided by Qiagen. GAPDH RNA transcription was used as endogenous reference and the quantification of the transcripts was performed by the ΔΔCT method.

Immunodetection of zyxin

Cell monolayers were washed twice with ice-cold PBS, lysed on the tissue culture dish by addition of ice-cold lysis buffer (50 mM Tris/HCl pH 7.4, 150 mM NaCl, 50 mM EDTA, 0.2 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AEBSF), 20 μg/mL leupeptin, 25 μg/mL aprotinin, 0,5 μg/mL pepstatin A and 1% Triton X-100), and an aliquot was used for protein analysis with the Pierce Bicinchoninic Acid kit, for protein quantification. Cell lysates were diluted in sample buffer (62.5 mM Tris/HCl pH 6.8, 2% SDS, 10% glycerol, 50 mM dithiothreitol, 0.1% Bromophenol blue) and subjected to western blot analysis. Proteins were subjected to SDS-PAGE (8%) and then transferred onto polyvinylidene difluoride (PVDF) membrane 0.45 μm (Immobilion; Millipore Corp, Bedford, MA, USA). The membrane was blocked for 1 h with 5% non-fat dry milk in Tris-buffered saline containing 0.1% Tween 20 (TBST). Membranes were immunoblotted with the rabbit anti-human zyxin polyclonal antibody (at 1 : 1000 dilutions in 5% non-fat dry milk, from Cell Signaling Technology, EuroClone, Milan, Italy). The detection was carried out by incubation with horseradish peroxidase conjugated goat anti-rabbit IgG (1 : 5000 dilutions in 5% non-fat dry milk, from Pierce, Rockford, IL, USA) for 1 h. The blots were then washed extensively and the proteins of interest were visualized using an enhanced chemiluminescent method (Pierce). Tubulin was also performed as a normal control of proteins.

p53 conformational immunoprecipitation

p53 conformational state was analyzed by immunoprecipitation. In particular, cells were lysed in immunoprecipitation buffer (10 mM Tris, pH 7.6; 140 mM NaCl; and 0.5% NP40 including protease inhibitors) for 20 min on ice, and cell debris was cleared by centrifugation. Hundred microgram of total cell extracts were used for immunoprecipitation experiments performed in a volume of 500 μL. To prevent non-specific binding, the supernatant of immunoprecipitated samples was pre-cleared with 10% (w/v) protein A/G (50 μL) (SantaCruz Biotechnology Inc., Heidelberg, Germany) for 20 min on ice, followed by centrifugation. For immunoprecipitation of p53, 1 μg of the conformation-specific antibodies PAb1620 (wild-type specific) (Calbiochem, EMB Bioscience, La Jolla, CA, USA), or PAb240 (mutant specific) (Neomarkers-Lab Vision, Fremont, CA, USA) was added to the samples and incubated overnight at 4°C. Immunocomplexes were separated by 10% SDS-PAGE and immunoblotting was performed with polyclonal anti-p53 antibody CM1 (Novocastra, Newcastle, UK). Immunoreactivity was detected with an enhanced chemiluminescent method (Pierce).

Densitometry and statistics

Following acquisition of the western blot image through an HP Scanjet G4010 scanner and analysis by means of the Image 1.47 program (Wayne Rasband, NIH, Research Services Branch, NIMH, Bethesda, MD, USA), the relative densities of the bands were expressed as arbitrary units and analyzed as described previously (Lanni et al. 2004). Data were analyzed using the analysis of variance test followed, when significant, by an appropriate post hoc comparison test as indicated in figure legend. A p-value < 0.05 was considered statistically significant. Data reported are expressed as means ± SD of at least three independent experiments.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Aβ 1-40 and Aβ 1-42 peptides impair p53 transcriptional activity

HEK-293 cells were evaluated for their vulnerability to a genotoxic insult. To this purpose, HEK-293 cells were treated with soluble Aβ 1-40 and 1-42 at the concentration of 10 nM for 48 h and then exposed for 24 h to 50 μmol/L doxorubicin, a cytotoxic agent able to induce DNA damage and apoptosis in a p53-dependent manner (Wang et al. 2004). Cell viability was evaluated measuring LDH release in the medium. Doxorubicin reduced cell viability of about 50% in HEK cells (Fig. 1a). On the other hand, 10 nM Aβ 1-40 and 1-42 pre-treated cells challenged with the same concentration of the genotoxic agent were found to be less vulnerable to doxorubicin-induced cytotoxicity, as indicated by a smaller reduction of cell viability (about 15%, Fig. 1a). We then analyzed the conformational state of p53. HEK cells were treated with 10 nM Aβ 1-40 or Aβ 1-42 for 48 h and subsequently processed for immunoprecipitation with two conformational-specific antibodies, PAb1620 and PAb240, which discriminate folded versus unfolded p53 tertiary structure (Méplan et al. 2000). Aβ 1-40 and Aβ 1-42 treatment induced the expression of a conformationally altered p53 phenotype that reacted with PAb 240 antibody (Fig. 1b).

image

Figure 1. Beta-amyloid (Aβ) peptides impair p53 transcriptional activity. (a) Human embryonic kidney (HEK)-293 cells were treated with 10 nM Aβ 1-40 or 1-42 for 48 h and then exposed to 50 μM doxorubicin for 24 h. Cell death was detected measuring lactate dehydrogenase release in the media. Data were expressed as % of cell viability versus control. ***p < 0.001 versus control, ###p < 0.001 versus doxorubicin treatment; Bonferroni Multiple Comparison test. (b) HEK cells were exposed to 10 nM Aβ 1–40 or Aβ 1–42 for 48 h and then cell lysates were immunoprecipitated with PAb240 or PAb1620 antibody. Immunoprecipitates were analyzed by western blot with the CM1 polyclonal anti-p53 antibody. Antibodies for immunoprecipitation were omitted in control (blk) samples. After densitometric analysis data were expressed as integrated density of ratio PAb240/PAb1620 antibodies signal over blank and represent means ± SD of at least three different experiments. **p < 0.01 and ***p < 0.001 versus untreated cells; Bonferroni Multiple Comparison test. (c) MT2A mRNA expression was determined by real-time PCR in HEK-293 cells treated with 10 nM Aβ 1-40 or 1-42 or 40-1 for 48 h. **p < 0.01 and ***p < 0.001 versus CTR; Dunnett's Multiple Comparison Test.

Download figure to PowerPoint

As previously reported, a mechanism, through which p53 conformation is affected, involves HIPK2 activity on metallothionein 2A (MT2A), a potent chelator in removing zinc from p53 in vitro (Puca et al. 2009). In particular, HIPK2 deregulation has been recently involved in p53 misfolding through MT2A up-regulation in AD cellular models and fibroblasts from AD patients (Lanni et al. 2010). Consistent with these data, analysis of mRNA showed that MT2A expression was up-regulated in both Aβ 1-40 and Aβ 1-42 treated cells, which is an index for HIPK2 deregulation in p53 misfolding, whereas treatment with the reverse peptide Aβ 40-1 failed to do so (Fig. 1c).

Aβ 1-40 and 1-42 peptides are responsible for zyxin deregulation

Following data indicating that both Aβ 1-40 and 1-42 peptides were able to affect the transcriptional repressor activity of HIPK2 onto its target promoter MT2A and that zyxin expression is important to maintain HIPK2 protein stability (Crone et al. 2011), we investigated whether zyxin expression was somehow compromised by nanomolar concentrations of soluble Aβ peptides. We evaluated zyxin mRNA expression and protein level in HEK-293 cells after treatment with Aβ 1-40 and 1-42. As shown in Fig. 2a, the analysis of RT-PCR following normalization to GAPDH expression revealed that no differences in zyxin mRNA expression were observed when comparing treated to untreated HEK-293 cells. Next, zyxin protein levels were evaluated by western immunoblotting. As shown in Fig. 2b, Aβ 1-40 and 1-42 treatment significantly reduced zyxin protein levels in HEK-293 cells, compared to vehicle or Aβ 40-1 treatment. Treatment with nanomolar concentrations of Aβ 1-40 and 1-42 in low serum condition provided evidence that both peptides entered into the cells (data not shown), consistently with previous data from literature (Uberti et al. 2007).

image

Figure 2. Beta-amyloid (Aβ) peptides are responsible for zyxin deregulation. (a) Zyxin mRNA expression was determined by real-time PCR in human embryonic kidney (HEK)-293 cells treated with 10 nM Aβ 1-40 or Aβ 1-42 or Aβ 40-1 for 48 h. (b) Total cell extracts of HEK-293 cells, treated with 10 nM Aβ 1-40 or Aβ 1-42 or Aβ 40-1 for 48 h, were analyzed for zyxin expression. Anti-tubulin was used as protein loading control. A representative experiment of at least three independent experiments was shown (upper panel). Results are shown as ratio zyxin/tubulin ± SD (lower panel). *p < 0.05 versus CTR (Bonferroni Multiple Comparison test).

Download figure to PowerPoint

Altogether, these results show that nanomolar concentrations of soluble Aβ 1-40 and 1-42 peptides can modulate zyxin protein levels.

Endogenous Aβ 1-40 and 1-42 peptides negatively affect zyxin protein levels

To better evaluate the role of intracellular beta-amyloid peptides in zyxin deregulation, we also used HEK-293 cells stably transfected with wild-type APP751 (HEK-APP) that express high levels of full length APP (Uberti et al. 2007). HEK-APP cells have been previously demonstrated to produce and release elevated amounts of Aβ peptides, where Aβ 1-40 was the most abundant isoform (Lanni et al. 2010). In HEK-APP cells, MT2A mRNA was up-regulated compared to HEK-293, in a similar way to the treatment with Aβ peptides (Fig. 3a), thus subtending an impairment of HIPK2 activity. To acquire more insight into the contribution of the different APP processing products, HEK-APP cells were also treated with β- (βSI) and γ- (γSI) secretase inhibitors to prevent amyloidogenic APP metabolism. Zyxin protein levels were then evaluated in HEK-293 cells and HEK-APP cells that were also treated with the βSI and γSI. As shown in Fig. 3b (upper panel), zyxin protein levels were strongly decreased in HEK-APP cells, compared to untransfected counterparts, as is also shown by quantitative analysis of zyxin/tubulin ratio (Fig. 3b, lower panel). Beta- and gamma-secretase inhibitors treatment restored zyxin protein levels to those of control cells (Fig. 3b), strongly suggesting that APP amyloidogenic metabolites may indeed affect zyxin protein levels.

image

Figure 3. Endogenous beta-amyloid (Aβ) 1-40 and 1-42 peptides negatively affect zyxin expression. (a) MT2A mRNA expression was determined in 10 nM Aβ 1-40 or Aβ 1-42 pre-treated human embryonic kidney (HEK)-293 and in HEK-APP compared to mock cells by reverse-transcriptase (RT)-PCR. GAPDH was used as loading control. (b) Western blot experiments were performed with anti-zyxin antibody on HEK-293 and HEK-APP cells that were also treated with β- and γ-secretase inhibitors at 10 μmol/L for 48 h. Anti-tubulin was used as protein loading control. A representative experiment of at least three independent experiments was shown (upper panel). Results are shown as ratio zyxin/tubulin ± SD (lower panel). *p < 0.05 versus HEK-293 (Bonferroni Multiple Comparison test).

Download figure to PowerPoint

We then tested the ability of the conditioned medium of HEK-APP cells to deregulate zyxin. To this aim, HEK-293 cells were cultured with conditioned medium of HEK-APP cells that, as shown in Fig. 4, decreased zyxin protein levels. To further discriminate the effect of intracellular from secreted Aβ peptides, HEK cells were treated with HEK-APP conditioned medium for 48 h in the presence or absence of the specific neutralizing antibody 6E10, that recognizes the first 17 aa of the Aβ sequence. Cells were then processed for western blot experiment carried out with zyxin antibody. Protein extracts from HEK cells exposed to HEK-APP conditioned medium showed reduced zyxin protein levels, compared to vehicle (Fig. 4). The reduction in zyxin protein levels found after treatment of HEK cells with HEK-APP conditioned medium was reverted by 6E10 antibody (Fig. 4). On the contrary, treatment with 6E10 antibody failed to restore zyxin protein levels in HEP-APP cells (Fig. 4).

image

Figure 4. Intracellular beta-amyloid (Aβ) is responsible for zyxin deregulation. Human embryonic kidney (HEK) cells were treated with conditioned medium of HEK-APP cells in the presence or absence of the monoclonal antibody 6E10. HEK-APP cells were also treated with 6E10 antibody. Western blot experiments were performed with anti-zyxin antibody. A representative experiment of at least three independent experiments was shown (upper panel). Results are shown as ratio zyxin/tubulin ± SD (lower panel). *p < 0.05 versus HEK-293 (Dunnett's Multiple Comparison test).

Download figure to PowerPoint

Neuronal cells show impairment in zyxin-HIPK2 signaling

Parallely to HEK-APP, we also used SH-SY5Y cells stably transfected with wild-type APP751 (SY-APP) that express increased APP mRNA and protein levels and showed an increased APP metabolism toward the amyloidogenic pathway (data not shown) when compared to untransfected cells. SH-SY5Y and SY-APP cells were differentiated with 10 μM retinoic acid, to have a closer neuronal model. First, SH-SY5Y and SY-APP cells were evaluated for their vulnerability to an oxidative insult after differentiation with retinoic acid. To this purpose, SH-SY5Y and SY-APP cells were exposed to 1 mM H2O2 for 10 min and cell viability was evaluated measuring LDH release in the medium. H2O2 reduced cell viability of about 50% in SH-SY5Y cells (Fig. 5a). On the other hand, SY-APP cells, after exposure to the same pulse of H2O2, were found to be less vulnerable to H2O2-induced cytotoxicity, as indicated by a smaller reduction of cell viability (about 20%, Fig. 5a). These data are consistent with those obtained in HEK cells pre-treated with sublethal concentrations of soluble Aβ 1-40 and 1-42 (Fig. 1a). We then characterized this cellular model in term of HIPK2 and zyxin expression. Moreover, to better understand the role of beta-amyloid peptides in the deregulation of this pathway in neuronal cells, we also exposed differentiated SH-SY5Y to 10 nM Aβ 1-40 and Aβ 1-42 for 48 h. Analysis of mRNA showed that MT2A expression was up-regulated in both Aβ 1-40 and Aβ 1-42 treated cells, as well as in differentiated SY-APP cells, suggesting that HIPK2 deregulation is involved in p53 misfolding, through MT2A up-regulation, as previously shown (Lanni et al. 2010) (Fig. 5b). When evaluating zyxin protein level in differentiated SH-SY5Y cells after treatment with Aβ 1-40 and 1-42, both Aβ 1-40 and Aβ 1-42 treatment significantly reduced zyxin protein levels, compared to vehicle or Aβ 40-1 treatment (Fig. 5c). Furthermore, as shown in Fig. 5c, zyxin protein levels were also strongly decreased in differentiated SY-APP cells, compared to untransfected counterparts. Parallel to data on HEK-APP cells, these data demonstrated that APP amyloidogenic metabolites may affect zyxin protein levels and suggest that p53-zyxin-HIPK2 signaling pathway is affected also in neuronal-like cells.

image

Figure 5. Neuronal cells show impairment in homeodomain-interacting protein kinase 2 (HIPK2) signaling and down-regulation of zyxin. (a) SH-SY5Y and SY-APP cells, treated or not with 10 μM retinoic acid, were exposed to 1 mM H2O2 for 10 min. Cell death was detected measuring lactate dehydrogenase release in the media. Data were expressed as % of cell viability versus control. ***p < 0.001 versus SHSY and **p < 0.01 versus SY-APP; ###p < 0.001 versus differentiated SHSY and ##p < 0.01 versus differentiated SY-APP; Bonferroni Multiple Comparison test. (b) MT2A mRNA expression was determined in 10 nM beta-amyloid (Aβ) 1-40 or Aβ 1-42 pre-treated SHSY and in SY-APP, after differentiation with retinoic acid, compared to mock cells by reverse-transcriptase (RT)-PCR. GAPDH was used as loading control. Results are shown as ratio mRNA MT2A/GAPDH ± SD. ***p < 0.001 versus differentiated SHSY cells (Dunnett Multiple Comparison test). (c) Total cell extracts of differentiated SY-APP and SH-SY5Y cells, treated with 10 nM Aβ 1-40 or Aβ 1-42 or Aβ 40-1 for 48 h, were analyzed for zyxin expression. Anti-tubulin was used as protein loading control. A representative experiment of at least three independent experiments was shown (upper panel). Results are shown as ratio zyxin/tubulin ± SD (lower panel). *p < 0.05 and **p < 0.01 versus differentiated SHSY cells (Dunnett Multiple Comparison test).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We describe here a link between zyxin-HIPK2-p53 signaling pathway and Alzheimer's disease. We previously demonstrated the existence of Aβ-dependent HIPK2 deregulation responsible for the induction of an unfolded state of p53 protein in fibroblasts from AD patients leading to an impaired and dysfunctional response to stressor (Uberti et al. 2002; Lanni et al. 2008). On the contrary, the non-amyloidogenic product of APP metabolism, sAPPalpha, was demonstrated to be not involved in the regulation of the pathway resulting in conformationally altered p53 (Uberti et al. 2007). Here, we examined the molecular mechanisms underlying the deregulation of HIPK2 activity in two cellular models, HEK-293 and SH-SY5H neuroblastoma cells differentiated with retinoic acid over-expressing the amyloid precursor protein. Recent findings show that zyxin expression is important to maintain HIPK2 protein stability (Crone et al. 2011). Our data suggest that intracellular Aβ peptides may be responsible for zyxin deregulation. This is supported by the observation that Aβ peptides down-regulated zyxin protein levels, compromising HIPK2 stability and thus leading to HIPK2 disappearance from target promoters such as MT2A. In agreement, MT2A mRNA up-regulation was found in HEK-APP cells that over-express APP751. The induction of MT2A, depending on HIPK2 knockdown has been reported to be responsible for p53 misfolding and inhibition of p53 transcriptional activity (Puca et al. 2009); therefore, the present data suggest that zyxin deregulation induced by Aβ peptides might be involved in HIPK2 degradation and in p53 misfolding via MT2A up-regulation in HEK-APP cells. This signaling pathway is further affected in a similar way by both soluble Aβ 1-40 and 1-42 at sublethal concentrations.

To investigate the contribution of APP metabolic products in the modulation of zyxin expression, we used HEK cells over-expressing wild-type APP able to generate high levels of Aβ 1-40 and Aβ 1-42 both intracellularly and secreted in the medium (Uberti et al. 2007). We found that reducing APP amyloidogenic metabolism by treating HEK-APP cells with β- and γ-secretase inhibitors (Lanni et al. 2010) prevented the deregulation of zyxin. It is worth to note that the conditioned medium of HEK-APP cells was able to affect untransfected HEK-293 cells recapitulating the HEK-APP phenotype, in terms of zyxin deregulation.

The data presented in this study suggest that the modulator effects of Aβ peptides on zyxin deregulation are, at least in part, due to the intracellular peptides. Different observations support this conclusion. First, considering that internalization of Aβ can be prevented under experimental conditions that do not allow endocytosis (Knauer et al. 1992), the conditioned media of HEK-APP cells treated with the specific neutralizing antibody 6E10, that recognizes the first 17 aa of the Aβ sequence, were unable to influence zyxin protein levels in HEK-293 cells. Second, we found that the synthetic Aβ peptides added to the cells crossed the plasma membrane, as previously demonstrated (Uberti et al. 2007).

The deregulation of zyxin was further analyzed in SH-SY5Y neuroblastoma cells and their counterpart over-expressing wild-type APP differentiated with retinoic acid. When treating differentiated SH-SY5Y cells with soluble Aβ peptides, both Aβ 1-40 and 1-42 were found to deregulate HIPK2 and decrease zyxin protein levels. Furthermore, also differentiated SY-APP cells are characterized by a decrease in zyxin protein levels, besides showing an HIPK2 down-regulation, thus recapitulating the HEK-APP phenotype in terms of zyxin and HIPK2 deregulation.

Zyxin is preferentially expressed in developing brain and various adult tissues, including lungs, spleen, and testis (Fujita et al. 2009). Depending on its subcellular location, zyxin can have anti-apoptotic or pro-apoptotic function, since zyxin has been shown to promote cell death downstream of apoptotic stimuli such as UV-C irradiation (Hervy et al. 2010), but it has also been implicated in Akt-dependent cardiomyocyte survival pathways (Kato et al. 2005). Zyxin has been extensively studied within a tumoral context, whereas very limited information is present at this time in the literature concerning its putative role in neurodegeneration. In this context, zyxin has been recently identified as an interacting partner for a protein, SIRT1, involved in protection from neurotoxicity in cell-based models for AD/tauopathies, amyotrophic lateral sclerosis and Wallerian degeneration, showing that the interaction of these proteins could be implicated in cellular survival, especially in the brain and heart, during physiological senescence (Fujita et al. 2009). Here, we established for the first time a link between zyxin and Alzheimer's disease. This observation is intriguing, since recent data from the literature support the concept that one or more common molecular mechanisms may be involved in the development of both neurodegenerative diseases and many cancers (Jope et al. 2007; Li et al. 2007; Lu and Zhou 2007; Alves da Costa and Checler 2011; Lanni et al. 2012). Taking into account that cancer and AD share common signaling pathways directing cell fate toward either death or survival, the identification of the putative common mechanisms may be useful to direct neurodegeneration studies toward the same intracellular pathways that have been successfully studied and targeted in cancer.

In summary, we hypothesize that soluble Aβ 1-40 and 1-42 may be responsible for important modulatory effects at cellular level before triggering the amyloidogenic cascade. For the first time, we demonstrated that soluble Aβ 1-40 and 1-42 modulate zyxin protein levels, fundamental in maintaining HIPK2 stability and in turn p53 activity. When zyxin is down-regulated by Aβ peptides, HIPK2 activity is inhibited, with MT2A up-regulation, in turn responsible for the induction of an altered conformational state of p53. As a result of this conformational change, p53 lost its transcriptional activity and was unable to properly activate an apoptotic program when cells were exposed to a noxious stimulus (Fig. 6). The reason why both Aβ 1-40 and Aβ 1-42 have the same effects on this pathway is at the moment under investigation in our laboratory. It is well known that Aβ 1-42 has been reported to aggregate faster than Aβ 1-40 and thus it is considered the most neurotoxic species (Verdile et al. 2004). Moreover, physiologically the 40-amino acid long peptide is the most abundant form (Terai et al. 2001; Kamenetz et al. 2003; Walsh and Selkoe 2007), since the concentration of secreted Aβ 1-42 is about 10% that of Aβ 1-40 (Bitan et al. 2003). However, in pathological conditions, the ratio of their production may be altered, as observed in familial AD cases (Mayeux et al. 1999). On the basis of this observation, one of the issues we should investigate is the differential modulation of this pathway by different concentration ratios of these two species. The sequence of events driven by beta-amyloid here described might contribute to AD pathogenesis since it may result in the presence of dysfunctional cells. Consistently, Yang and co-workers demonstrated the existence of aberrant neurons in AD brain by showing that neurodegeneration is correlated with neurons reentering a lethal cell cycle (Yang et al. 2001; Copani et al. 2007, 2008), which suggests that dysfunctional p53 in non-dividing cells may play a role in aberrant cell cycle progression. Furthermore, the observation that in AD are involved proteins controlling the duplication and cell cycle control leads to the speculation that, in senescent neurons, derangements in proteins commonly dealing with cell cycle control and apoptosis could affect neuronal plasticity and functioning rather than cell duplication.

image

Figure 6. Working hypothesis for a putative role of zyxin in Alzheimer's disease pathogenesis. The Figure indicates zyxin deregulation mediated by soluble beta-amyloid (Aβ) peptides. When zyxin is down-regulated by sublethal concentrations of Aβ 1-40 or Aβ 1-42, homeodomain-interacting protein kinase 2 (HIPK2) expression and activity are inhibited through degradation via the proteasome system. HIPK2 deregulation results in the induction of metallothionein 2A (MT2A), that exerts its Zn2+ chelator function. As a consequence, p53 changes the wild-type conformation to a conformationally altered status, with subsequent abolishment of wild-type p53 transcriptional activity.

Download figure to PowerPoint

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported by the contribution from the UNIPV-Regione Lombardia grant to C.L., and from the Ministry of University and Research (MIUR, Grant #2009B7ASKP to S.G.). None of the authors has any conflicting financial interest.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References