Inhibition of RhoA GTPase and the subsequent activation of PTP1B protects cultured hippocampal neurons against amyloid β toxicity
- 6.6k Downloads
Amyloid beta (Aβ) is the main agent responsible for the advent and progression of Alzheimer's disease. This peptide can at least partially antagonize nerve growth factor (NGF) signalling in neurons, which may be responsible for some of the effects produced by Aβ. Accordingly, better understanding the NGF signalling pathway may provide clues as to how to protect neurons from the toxic effects of Aβ.
We show here that Aβ activates the RhoA GTPase by binding to p75NTR, thereby preventing the NGF-induced activation of protein tyrosine phosphatase 1B (PTP1B) that is required for neuron survival. We also show that the inactivation of RhoA GTPase and the activation of PTP1B protect cultured hippocampal neurons against the noxious effects of Aβ. Indeed, either pharmacological inhibition of RhoA with C3 ADP ribosyl transferase or the transfection of cultured neurons with a dominant negative form of RhoA protects cultured hippocampal neurons from the effects of Aβ. In addition, over-expression of PTP1B also prevents the deleterious effects of Aβ on cultured hippocampal neurons.
Our findings indicate that potentiating the activity of NGF at the level of RhoA inactivation and PTP1B activation may represent a new means to combat the noxious effects of Aβ in Alzheimer's disease.
KeywordsNerve Growth Factor Hippocampal Neuron RhoA GTPase RhoA Activation Noxious Effect
According to the amyloid hypothesis, amyloid beta (Aβ) aggregates form deposits in the brain, the process that precipitates the different manifestations of Alzheimer's disease (AD) . Consequently, most therapeutic approaches to treat AD centre on this peptide: on the one hand attempting to limit the production of Aβ or the formation of fibrils and aggregates [2, 3], while on the other hand, favouring its clearance. Therapeutic approaches aimed at clearing Aβ plaques have received special attention, and methods for active or passive immunisation have proven effective in reducing Aβ content in the brain. Nevertheless, these strategies have failed to conclusively ameliorate or retard cognitive deterioration in AD patients [4, 5].
Another approach that could be considered involves blocking the signals induced by Aβ that provoke neuronal death. However, despite extensive studies into the effects of Aβ on neurons, our understanding of Aβ signalling remains fragmented, and a consistent framework for such processes has yet to be defined. Still, recent publications have reinforced the notion that Aβ interferes with insulin signalling  and indeed, when soluble forms of Aβ bind to dendrites, they provoke the removal of insulin receptors (probably by activating their internalization), as well as preventing synapse formation . In addition, intracellular Aβ may impair insulin signalling by preventing phosphoinositide-dependent kinase dependent activation of Akt . This Aβ-promoted disruption of insulin signalling has prompted clinical trials in which insulin activity is primed and stimulated [9, 10].
By contrast, Aβ neurotoxicity has also been associated with the trophic effects of NGF. Indeed, some therapeutic approaches for AD involve the use of NGF or mimic the effects of NGF [11, 12, 13, 14, 15, 16]. Indeed, the cellular and molecular bases underlying the antagonism of NGF by Aβ were recently elucidated in part. Aβ competes with NGF for binding to p75NTR [17, 18], thereby preventing the activation of NF-κ-B by impairing the tyrosine phosphorylation and subsequent degradation of I-κ-B . The inhibition of NF-κ-B promoted by Aβ results in the downregulation of Homologous of Enhancer-of-split 1 (Hes1) expression, a gene that has an important influence on dendrite patterning and GABAergic inputs [20, 21].
In this study, we show that Aβ impairs the initial steps of NGF signalling at the level of the RhoA GTPase and PTP1B. We also show that potentiating NGF signalling by inhibiting RhoA GTPase and activating PTP1B offers cells certain resistance against Aβ neurotoxicity.
Aβ (1-42) induces morphological changes and regulates neuron survival via p75NTR/RhoA
The role of RhoA in the neurotoxicity of Aβ
Protein tyrosine phosphatase 1B activity protects neurons against Aβ neurotoxicity
The effects of Aβ on cultured neurons mediated by p75NTR
The noxious effects of Aβ appear to be at least partially due to the neutralisation of NGF activity in neurons  (see also ) and indeed, we demonstrated that Aβ prevents the activation of NF-κ-B and the increase of Hes1 expression caused by NGF. We also more recently revealed novel features of NGF signalling in neurons, including the activation of PTP1B after binding to p75NTR , that is in turn necessary for the tyrosine phosphorylation and degradation of I-κ-B, and the subsequent activation of NF-κ-B. p75NTR appears to be the only receptor mediating the effects of NGF on both RhoA and PTP1B. Indeed, we showed that a selective blockage of the receptor prevented NGF-induced PTP1B activation and that the inactivation of TrkA did not abolish such an effect .
The participation of p75NTR in the activity of Aβ is well documented, and radioactive Aβ was seen to bind to the receptor and trigger apoptosis in cell lines that overexpress p75NTR [17, 18]. More recently, neurons expressing a mutated form of p75NTR devoid of its extracellular domain were shown to be resistant to the neurite dystrophy associated to c-Jun activation and to apoptosis, both in vitro and in vivo [23, 31]. In addition, the fact that Aβ augments the expression of p75NTR in SH-SY5Y human neuroblastoma cells is further evidence for the role of p75NTR, especially given that it also increases in hippocampal neurons from a transgenic mouse model of AD .
In a PC12 cell variant devoid of TrkA, we established that Aβ activates RhoA by binding to p75NTR, an effect that is partially prevented by exposing the cells to NGF prior to the administration of Aβ. Activation of RhoA by Aβ has also been observed in SH-SY5Y cells, in which it controls the phosphorylation of the collapsing response mediator protein-2, an effect that disrupts its binding to β-tubulin and neurite growth prior to cell death .
The role of RhoA GTPase in neurodegeneration
There is increasing evidence that RhoA is an essential effector of certain neurodegenerative processes. The binding of various myelin associated ligands to the Nogo receptor complex, which includes p75NTR, produces RhoA activation . Moreover, disruption of this receptor's function or deactivation of RhoA facilitates neurite and axonal growth in injured CNS neurons [35, 36]. RhoA activity has also been associated to AD, particularly since the distribution of RhoA is altered in the brains of AD patients and in an AD mouse model. Moreover, activated RhoA augments in dystrophic dendrites and diminishes around synapses . Here, we reveal that activation of RhoA not only decreases dendrite length but also, it is an important mediator of Aβ-induced neuron death. Indeed, activation of RhoA by CNFy mimics the effects of Aβ on dendrite morphology, although it did not mimic the deleterious effects of Aβ on the cells. Nevertheless, the role of RhoA in neuronal death was assessed since attenuating the GTPase activity, either by transfecting neurons with a dn form of RhoA or through pharmacological inhibition with C3 ADP ribosyl transferase, protects a significant number of neurons from Aβ neurotoxicity. These results indicate that RhoA activation plays a role in AD development, suggesting that inhibition of this GTPase might delay the progress of the disease.
Relationship between RhoA and PTP1B
A functional analysis revealed that constitutively active RhoA inactivates PTP1B, although the precise mechanism underlying such inhibition remains unknown . Our studies demonstrate a functional relationship between these two enzymes, since the pharmacological inhibition of RhoA was followed by activation of PTP1B, thereby mimicking the effects of NGF on the phosphatase. Conversely, the activation of RhoA by CNFy, a yersinia toxin  prevented NGF from activating PTP1B, an effect reminiscent of the action of Aβ. However, activation of RhoA did not diminish PTP1B activity below basal levels and the activation of RhoA only reduced the fraction of PTP1B activity increased by NGF. This may indicate that not all PTP1B molecules are functionally linked to the state of RhoA activation, reflecting the variety of signalling pathways in which PTP1B is involved .
RhoA controls dendrite length and hampers dendrite elongation, spine formation and synapse stabilization by a mechanism in which the activation of ROCK is involved . The fact that RhoA also governs dendrite patterning by mechanisms that don't involve PTP1B may explain how CNFy decreases dendrite length without affecting the activity of the phosphatase. However, the true nature of the RhoA/PTP1B connection still remains unclear. Reasonable candidates to participate in this process are the reactive oxygen species (ROS) and indeed, early studies revealed that Aβ increased the ROS pool in neurons [40, 41] reviewed in . Increased levels of ROS may act in several ways, activating RhoA by oxidising a redox sensitive domain , or by inactivating cysteine phosphatases like PTP1B, as seen in vitro  and in cellular systems after calcium influx , as well as after application of insulin , EGF  or IL-4 .
Activation of PTP1B is needed for neuron survival
PTP1B is often constitutively active although its activity may be negatively controlled by Akt induced serine phosphorylation of the phosphatase , tyrosine phosphorylation , or extracellular stimulation of ROS levels that oxidise the active centre of the phosphatase under the control of extracellular effectors such as insulin . Activation of PTP1B activity is associated with the C-terminal cleavage of the enzyme , reviewed in . However, activation of PTP1B associated to extracellular stimuli such as NGF has only recently been observed , and post-translational modifications of the enzyme that enhance its activity have yet to be identified.
The mechanism by which we believe NGF promotes Hes1 expression and by which Aβ opposes such effects is outlined in Figure 6 in which the main findings of the present study are represented in red, our previous results are in blue and other published findings are in black. Thus, the main conclusions of the present studies are: (i) Aβ activates RhoA via p75NTR; (ii) the activation of RhoA inactivates PTP1B; (iii) inactivation of RhoA by either pharmacological inhibitor or transfection with a dn isoform makes neurons resistant to Aβ; (iv) increasing PTP1B activity makes neurons overexpressing the phosphatase resistant to Aβ. In summary, by strengthening different elements in the NGF signalling pathway it is feasible to make neurons more resistant to the effects of amyloid.
For immunocytochemistry, the primary antibodies used were a rabbit anti-enhanced green fluorescent protein (EGFP) from Invitrogen (Carlsbad, CA; used at a dilution of 1:500), a rat anti-hemagglutinin (HA) mAb and a mouse anti-Myc mAb (both used at a dilution of 1:400; Roche Applied Science, Mannheim, Germany). Western blots were probed with a mouse anti-RhoA (1:250; Santa Cruz Biotechnology, Santa Cruz, CA) and a mouse anti-PTP1B was used for immunoprecipitations (1:50; BD Transduction Laboratories, Lexington, KY). The goat anti-rabbit Cy2 (1:1000), goat anti-rat Cy3 (1:500), goat anti-mouse Cy3 (1:1000) and donkey anti-mouse-HRP (1:5000) secondary antibodies were all obtained from Jackson Immuno Research (West Grove, USA).
NGF from mouse salivary glands was obtained from Alomone Labs (Jerusalem, Israel), and it was used at a concentration of 100 ng/ml. Amyloid β (1-42) was obtained from NeoMPS (Strasbourg, France) and it was used at a concentration of 5 μM. This peptide was dissolved in 1,1,1,3,3,3-hexafluoro-2-propanol to obtain the oligomeric form of Amyloid β. After the solvent was allowed to evaporate for 2 hrs at room temperature, the peptide film was dissolved in DMSO, sonicated in a water bath for 10 min diluted to 100 μM in PBS and briefly vortexed, before it was incubated overnight at 4 °C. Aliquots were stored at -20°C. C3 ADP ribosyl transferase (Cytoskeleton Inc., Denver, CO), a cell permeable Rho inhibitor, was used at 500 ng/ml. CNFy (cytotoxic necrotizing factor from Yersinia pseudotuberculosis), a specific activator of RhoA, was produced as described previously  and used at 200 ng/ml. Raytide™EL, a general tyrosine kinase peptide substrate, and TAT-Pep5, a cell permeable p75NTR signalling inhibitor were used at 1.0 μM, each purchased from Calbiochem (Darmstadt, Germany). [γ-32P]ATP was obtained from Perkin-Elmer (Madrid, Spain).
The EGFP expressing vector (pEGFP-N1) was obtained from Clontech Laboratories, Inc. (Palo Alto, CA), while the pCDNA 3.1 Zeo encoding a HA-tagged form of wild type (wt) PTP1B was kindly provided by Carlos Arregui (Buenos Aires, Argentina) . The pRK5-Myc vector encoding a Myc-tagged dominant negative form of RhoA, RhoA N19 (Addgene plasmid 15901) , was obtained from Addgene (Cambridge, MA).
Primary hippocampal neuron cultures were prepared as described previously , with some minor modifications. Briefly, the hippocampus was removed from E17 CD1 mouse foetuses and dissociated into single cells following trypsin (Worthington, Lakewood, USA) and DNase I digestion (Roche Applied Science). Neurons were plated on glass coverslips or in plastic dishes coated with poly-L-lysine (Sigma-Aldrich, Madrid, Spain), and cultured in Neurobasal A supplemented with 2 mM GlutaMAX I, 100 units/ml penicillin and 100 μg/ml streptomycin (Gibco BRL, Crewe, UK). After 7 days in vitro (DIV) the neurons were exposed to Aβ, NGF, and/or the pharmacological agents indicated. TrkA-deficient PC12 cells, PC12nnr5 , were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 5% heat-inactivated foetal bovine serum (FBS), 10% heat-inactivated horse serum (Sigma-Aldrich), 2 mM GlutaMAX I, 100 units/ml penicillin and 100 μg/ml streptomycin (Gibco BRL). All cultures were kept at 37°C in a humidified atmosphere containing 5% CO2.
Cultured neurons were transfected at 7 DIV with different vectors using the Effectene Transfection Reagent (Qiagen GmbH, Hilden, Germany) according to a modified version of the manufacturer's instructions. Briefly, 0.6 μg of DNA was added to 120 μl of the EC buffer and 3.5 μl of the Enhancer for each 35 mm cell culture dish of hippocampal neurons. The solution was incubated for 5 min at room temperature before 10 μl of Effectene was added, and after a further 15 min incubation at room temperature, the final solution was added to hippocampal neurons. The medium was then changed after 3 h.
Immunocytochemistry, image acquisition and the morphometric analysis of labelled hippocampal neurons
At 16 h after transfection, the neurons were fixed for 30 min in 4% paraformaldehyde (PFA) prepared in PBS, they were then permeabilized for 15 min at room temperature with 0.5% Triton X-100 in PBS and blocked with 10% FBS in PBS containing 0.1% Triton X-100. The cells were incubated with the primary and secondary antibodies and the images of 10-20 neurons per coverslip were acquired digitally using a 63× oil immersion objective (Zeiss, Oberkochen, Germany). To analyze the dendrites, a region of interest (ROI) with a radius of 50 μm was projected onto EGFP-labelled neurons, its centre roughly coinciding with the centre of the soma. The dendrite length was expressed as the fraction of the dendritic tree that exceeds the limit of the ROI (fraction dendrites >50 μm). In co-transfection experiments, only double-labelled cells were analysed, which represented more than 90% of the single-labelled cells.
After treatment, the neurons were fixed for 30 min in 4% paraformaldehyde (PFA) in PBS and their nucleus was stained with the fluorescent dye, 4',6-diamidino-2-phenylindole (DAPI: Sigma-Aldrich). Non-viable neurons were recognized by nuclear condensation and/or the fragmentation of their chromatin. The number of viable neurons was counted in triplicate from ca. 50 fields in two independent experiments. In co-transfection experiments, only the nuclei of double-labelled cells were analysed.
To assay RhoA activation we followed a procedure described elsewhere . Briefly, stimulated PC12nnr5 cells were first lysed in buffer: 50 mM Tris [pH 7.5], 500 mM NaCl, 10 mM MgCl2, 1.0% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS and a protease inhibitor cocktail (Roche Applied Science, Mannheim, Germany). The cleared lysates were incubated for 1 h at 4°C with Rhotekin-conjugated agarose beads (Cytoskeleton), and the beads were then collected by centrifugation and washed with the lysis buffer. Activated RhoA was detached from the beads by boiling for 5 minutes in Laemmli reducing buffer, after which it was immediately resolved by 12% SDS-PAGE and transferred to a nitrocellulose membrane. After blocking, the membranes were probed overnight at 4°C with a primary antibody directed against RhoA, which was detected with an HRP-conjugated secondary antibody that was visualised by enhanced chemiluminescence (GE-Healthcare, Piscataway, NJ). The intensity of the bands was evaluated by densitometry using ImageQuant software (GE-Healthcare).
Protein phosphatase 1B assay
The PTP1B assay was performed essentially as described previously , with minor modifications. Briefly, 7DIV cultured hippocampal neurons (1 × 106 cells) were collected and homogenized in RIPA buffer (50 mM Tris [pH7.5], 150 mM NaCl, 2 mM EDTA, 1.0% Triton X-100 and an anti-protease cocktail). Equal amounts of protein were incubated for 2 h at 4°C with a mouse anti-PTP1B mAb, and then 20 μl of protein G sepharose was added and incubated for additional 2 h with agitation. Immunoprecipitated complexes were washed twice in RIPA buffer, once with the assay buffer (25 mM imidazole [pH7.2], 0.1 mg/ml BSA, 10 mM DTT) and they were then resuspended in 25 μl of assay buffer. The phosphatase substrate Raytide was labelled at its unique tyrosine residue with [γ-32P]ATP as described previously . Assay mixtures (30 μl) containing the immunoprecipitated pellet and [32P]-labelled raytide (1 × 105 cpm) were incubated for 2 h at 30°C and the reaction was terminated by adding 750 μl of a charcoal mixture (0.9 M HCl, 90 mM sodium pyrophosphate, 2 mM NaH2PO4, 4% vol/vol Norit A). After centrifugation, the radioactivity in 400 μl of the supernatant was measured by scintillation counting. Blanks were determined by measuring the free [32P]phosphate in reactions where the immunoprecipitates were either boiled or omitted, and these values were subtracted from the reaction values.
Quantitative real time polymerase chain reaction (PCR)
After treatment, the total RNA was extracted from cultures at 7 DIV using the Illustra RNAspin Mini kit (GE-Healthcare) and first strand cDNA was prepared from the RNA using the First Strand Synthesis kit (Fermentas GmbH, St Leon-Rot, Germany). Quantitative PCR was performed using the ABI Prism 7000 Sequence Detector (Applied Biosystems, Weiterstadt, Germany) and TaqMan probes. Primers for Hes1 and the housekeeping gene GADPH (as a control) were selected as the Assay-on-Demand gene expression products (Applied Biosystems). All TaqMan probes were labelled with 6-carboxy fluorescein (FAM) and real time PCR was performed using the TaqMan Universal PCR Master Mix according to the manufacturers' instructions. Hes1 expression was normalized to the GAPDH expression.
Data are presented as the mean ± SEM and an unpaired Student's t-test was applied to determine the levels of significance, denoted as *p < 0.05, **p < 0.01, ***p < 0.001. All experiments were repeated at least twice.
P. Chacon was supported by the Instituto de Salud Carlos III (Contratos Post-Doctorales Sara Borrell). This work was financed by the 'Fundació La Caixa' (grant BM05-184) and the Spanish Ministry of Education and Science (grant BFU2005-05629). We are indebted to Emmanuel Villanueva for technical help, to Dr Marta Lloverá (Lleida, Spain) for providing us with the PC12nnr5 cells, to Dr Gudula Schmidt (Freiburg, Germany) for providing us with the pGEX-2TGL-CNFy plasmid from which CNFy was obtained, and to Carlos Arregui (Buenos Aires, Argentina) for providing us with the PTP1B plasmid and for useful discussions.
- 7.De Felice FG, Vieira MN, Bomfim TR, Decker H, Velasco PT, Lambert MP, et al: Protection of synapses against Alzheimer's-linked toxins: insulin signaling prevents the pathogenic binding of Abeta oligomers. Proc Natl Acad Sci USA. 2009, 106: 1971-1976. 10.1073/pnas.0809158106.PubMedPubMedCentralCrossRefGoogle Scholar
- 12.Louzada PR, Lima AC, Mendonca-Silva DL, Noel F, De Mello FG, Ferreira ST: Taurine prevents the neurotoxicity of beta-amyloid and glutamate receptor agonists: activation of GABA receptors and possible implications for Alzheimer's disease and other neurological disorders. FASEB J. 2004, 18: 511-518. 10.1096/fj.03-0739com.PubMedCrossRefGoogle Scholar
- 21.Salama-Cohen P, Arevalo MA, Grantyn R, Rodriguez-Tebar A: Notch and NGF/p75NTR control dendrite morphology and the balance of excitatory/inhibitory synaptic input to hippocampal neurones through Neurogenin 3. J Neurochem. 2006, 97: 1269-1278. 10.1111/j.1471-4159.2006.03783.x.PubMedCrossRefGoogle Scholar
- 29.Chacon PJ, Arevalo MA, Tebar AR: NGF-activated protein tyrosine phosphatase 1B mediates the phosphorylation and degradation of I-kappa-Balpha coupled to NF-kappa-B activation, thereby controlling dendrite morphology. Mol Cell Neurosci. 2010, 43: 384-393. 10.1016/j.mcn.2010.01.005.PubMedCrossRefGoogle Scholar
- 40.De Felice FG, Velasco PT, Lambert MP, Viola K, Fernandez SJ, Ferreira ST, et al: Abeta oligomers induce neuronal oxidative stress through an N-methyl-D-aspartate receptor-dependent mechanism that is blocked by the Alzheimer drug memantine. J Biol Chem. 2007, 282: 11590-11601. 10.1074/jbc.M607483200.PubMedCrossRefGoogle Scholar
- 57.Takeda S, Sato N, Uchio-Yamada K, Sawada K, Kunieda T, Takeuchi D, et al: Diabetes-accelerated memory dysfunction via cerebrovascular inflammation and Abeta deposition in an Alzheimer mouse model with diabetes. Proc Natl Acad Sci USA. 2010, 107: 7036-7041. 10.1073/pnas.1000645107.PubMedPubMedCentralCrossRefGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.