Tat-Hsp70 protects dopaminergic neurons in midbrain cultures and in the substantia nigra in models of Parkinson’s disease

Authors

  • Florian Nagel,

    1. Neurologische Universitätsklinik, Göttingen, Germany
    2. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    Search for more papers by this author
  • Björn H. Falkenburger,

    1. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    2. Department of Neurodegeneration and Restorative Research, Göttingen, Germany
    Search for more papers by this author
  • Lars Tönges,

    1. Neurologische Universitätsklinik, Göttingen, Germany
    2. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    Search for more papers by this author
  • Sebastian Kowsky,

    1. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    2. Department of Neurodegeneration and Restorative Research, Göttingen, Germany
    Search for more papers by this author
  • Charlotte Pöppelmeyer,

    1. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    2. Department of Neurodegeneration and Restorative Research, Göttingen, Germany
    Search for more papers by this author
  • Jörg B. Schulz,

    1. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    2. Department of Neurodegeneration and Restorative Research, Göttingen, Germany
    Search for more papers by this author
  • Mathias Bähr,

    1. Neurologische Universitätsklinik, Göttingen, Germany
    2. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    Search for more papers by this author
  • Gunnar P. H. Dietz

    1. Neurologische Universitätsklinik, Göttingen, Germany
    2. DFG Research Center for Molecular Physiology of the Brain (CMPB), Göttingen, Germany
    Search for more papers by this author

Address correspondence and reprint requests to Gunnar P. H. Dietz, PhD, Neurologische Universitätsklinik, Waldweg 33, D-37073 Göttingen, Germany. E-mail: gdietz@gwdg.de; Homepage: http://www.baehrlab.med.uni-goettingen.de/Staff_Positions/Dietz/dietz.html

Abstract

Parkinson’s disease is characterized by the progressive loss of dopaminergic (DA) neurons in the substantia nigra. The heat-shock protein 70 (Hsp70) reduces protein misfolding and aggregation. It has been shown to protect cells against oxidative stress and apoptotic stimuli in various neurodegenerative disease models. To deliver Hsp70 across cellular membranes and into the brain, we linked it to a cell-penetrating peptide derived from the HIV trans-activator of transcription (Tat) protein. In vitro, Tat-Hsp70 transduced neuroblastoma cells and protected primary mesencephalic DA neurons and their neurites against MPP+-mediated degeneration. In vivo, the systemic application of cell-permeable Hsp70 protected DA neurons of the substantia nigra pars compacta against subacute toxicity of MPTP. Furthermore, Tat-Hsp70 diminished the MPTP induced decrease in DA striatal fiber density. Thus, we demonstrate that systemically applied Tat-Hsp70 effectively prevents neuronal cell death in in vitro and in vivo models of Parkinson’s disease. The use of Tat-fusion proteins might therefore be a valuable tool to deliver molecular chaperones like Hsp70 into the brain and may be the starting point for new protective strategies in neurodegenerative diseases.

Abbreviations used
AAV

adeno-associated virus

BBB

blood–brain barrier

CPP

cell penetrating peptide

DA

dopaminergic

DAPI

4′,6-diamidino-2-phenylindole

DIV

days in vitro

DOPAC

3,4-dihydroxyphenylacetic acid

HA

hemagglutinin

Hsp70

heat-shock protein 70

HVA

homovanillic acid

i.p.

intraperitoneal

NGS

normal goat serum

PAA

polyacrylamide

PBS

phosphate-buffered saline

PD

Parkinson’s disease

SDS

sodium dodecyl sulfate

SN

substantia nigra

SNpc

substantia nigra pars compacta

Tat

trans-activator of transcription

TH

tyrosine hydroxylase

XIAP

X-chromosome-linked inhibitor of apoptosis protein

The neurotoxin MPTP can produce similar biochemical and neuropathological defects as observed in Parkinson’s disease (PD) patients (Schulz and Beal 1994), including the progressive loss of dopaminergic (DA) neurons in the substantia nigra pars compacta (SNpc) and the decrease of striatal DA levels. Therefore, it is one of the best experimental models for sporadic PD (Przedborski et al. 2004). Lipophilic MPTP passes the blood–brain barrier (BBB) and cellular membranes. In astrocytes, monoamine oxidase B converts MPTP into MPP+ ion (Ransom et al. 1987). MPP+ is taken up into DA neurons by their DA transporters (Mayer et al. 1986), inhibiting mitochondrial complex I (Tipton and Singer 1993). It promotes ATP depletion and generation of reactive oxygen species (Rossetti et al.1988), which can activate apoptotic pathways (Przedborski et al. 2004).

One of the anti-apoptotic and anti-inflammatory players in models for neurodegenerative diseases such as autoimmune encephalomyelitis, Alzheimer’s disease, PD, and other related insults is the heat-shock protein 70 (Hsp70) (Garrido et al. 2001; Dong et al. 2005; Yenari et al. 2005; Galazka et al. 2006). Hsp70 prevents the protein misfolding and aggregation associated with neurodegenerative diseases like Alzheimer’s disease and PD. It protects cells against heat shock, oxidative stress, and apoptotic stimuli in many death pathways (Wheeler et al. 2003; Giffard and Yenari 2004; Lai et al. 2005; Yenari et al. 2005).

Apoptosis can be blocked by Hsp70 via direct interferences at multiple sites: It increases B-cell lymphoma 2 protein (Bcl-2) expression (Kelly et al. 2002) and inhibits cytochrome c release (Beere et al. 2000; Creagh et al. 2000). It has been shown that Hsp70 interferes with apoptosis protease activating factor-1, preventing recruitment of procaspase 9 into the apoptosome (Beere et al. 2000; Saleh et al. 2000). Further downstream in the apoptotic cascade, Hsp70 is able to inhibit effector caspases like caspase 3 (Mosser et al. 1997). Finally, Hsp70 antagonizes apoptosis-inducing factor in a caspase-independent way (Susin et al. 1999; Ravagnan et al. 2001).

‘Protein transduction domains’, also called cell penetrating peptides (CPP), have been used to mediate the translocation of proteins and other compounds across the cellular membrane and into the brain (Dietz and Bähr 2004, 2005). One of the most commonly used CPP is the basic domain of the trans-activator of transcription (Tat) from HIV, where effective transduction and functionality of delivered cargoes have been well demonstrated in vitro and in vivo (Schwarze et al. 1999; Dietz et al. 2002, 2006a,b; Guegan et al. 2006,Yin et al. 2006). As the Hsp70 is too large to freely pass biological membranes or the BBB, the Tat-domain seems to be a promising vector for their delivery (Wheeler et al. 2003; Lai et al. 2005). Therefore, we generated a fusion protein between the basic 11-amino acid domain of Tat (CPP) and the Hsp70 protein to deliver recombinant Hsp70 across cellular membranes and the BBB.

The goal of our study was to test if Tat-mediated Hsp70 is an effective neuroprotectant in models for PD. We delivered Hsp70 into primary neurons and across the BBB via the Tat protein transduction domain, and examined whether the application of Tat-Hsp70 would protect cultured DA midbrain neurons and its neurite processes against MPP+. Tat-Hsp70 was applied systemically in vivo in a subacute toxicity model of MPTP to investigate whether cell permeable Hsp70 diminishes the MPTP induced decrease in tyrosine hydroxylase immunopositive (TH+) cells in the SNpc. Furthermore, we addressed the question of whether Tat-Hsp70 can also diminish the loss of striatal fiber density and the depletion of striatal DA and its metabolites.

Materials and methods

Expression and purification of fusion proteins

To create an expression vector for the Tat-Hsp70 fusion protein rat hsp70.1 cDNA was cloned into the pTat-hemagglutinin (HA) expression vector (kindly provided by S. F. Dowdy, University of California, San Diego, CA, USA). pTat-HA is derived from pRSET™B (Invitrogen GmbH, Karlsruhe, Germany). We amplified the Hsp70.1 coding sequence by PCR of the corresponding region of Hsp70.1 rat cDNA-containing plasmid (kindly provided by K. Lisowska, Centre of Oncology, Gliwice, Poland) with the sense primer 5′-CAGTAGGTACCGCCAAGAAAACAGCGATCGGC-3′ and the antisense primer 5′-GCAGCGAATTCCTAATCCACCTCCTCGATGGT-3′. Purified fragments were cloned into the KpnI/EcoRI sites of the pTat-HA and the pRSET™B expression vector (Invitrogen GmbH). The resulting expression cassette includes a sequence encoding six histidine residues, the 11-amino acid transduction domain (YGRKKRRQRRR) of the Tat protein (5′ of the polylinker, under the control of a T7 promoter), a HA tag (YPYDVPDYA) and the Hsp70 sequence. The constructs were verified by DNA sequence analysis. Tat-Hsp70, Tat-HA, and Hsp70 were expressed in Escherichia coli strain BL21 (DE3)pLysS (Novagen, Madison, WI, USA) and isolated in its native conformation as described (Dietz et al. 2002). Briefly, cell debris was removed by centrifugation and the cell extracts were purified by metal-affinity chromatography using Ni–Tris–carboxymethyl-ethylene-diamine (Macherey-Nagel, Düren, Germany). Protein was eluted by stepwise addition of binding buffer containing increasing concentrations of imidazole (0.1–5 M, lanes 1–5 of Fig. 1a) and collection of 2 mL fractions. We removed salt by gel filtration on Sephadex size exclusion columns (GE Healthcare, Munich, Germany) and confirmed identity of proteins by western blotting. Anti-HA antibodies were purchased from Covance Inc. (Princeton, NJ, USA). Protein purity was assessed via Coomassie (Coomassie Brilliant blue; Merck, Darmstadt, Germany) stained sodium dodecyl sulfate (SDS) gels. The protein concentration was quantified by comparison with protein standards, loaded on Coomassie-stained polyacrylamide (PAA) gels.

Figure 1.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) was isolated to high purity. Coomassie-stained sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) gels showing purity and concentration of Tat-Hsp70. Tat-Hsp70 is approximately 77 kDa in size. (a) Tat-Hsp70 was purified by metal-affinity chromatography using Ni–Tris–carboxymethyl-ethylene-diamine. Protein was eluted by stepwise addition of binding buffer containing increased concentrations of imidazole (0.1–5 M, lanes 1–5). (b) We removed salt by gel filtration on Sephadex size exclusion columns. Protein purity was assessed via Coomassie-stained polyacrylamide (PAA) gels (lanes 1–6). Bovine serum albumin (BSA) protein was used as a protein standard (60, 150, 300, and 600 ng).

Cell culture models

SH-SY5Y human neuroblastoma cells

SH-SY5Y human neuroblastoma cells were maintained in Dulbecco’s modified Eagle’s medium containing 15% fetal calf serum, penicillin (100 U/mL), and streptomycin (100 μg/mL) (Invitrogen) at 37°C in a humidified atmosphere of 95% air and 5% CO2. Cells were seeded into 24-well dishes (Multiwell™ 24 well; Falcon, Oxnard, CA, USA) at a density of 1.5 × 104 cells/cm2, in 96-well (Becton-Dickinson GmbH, Heidelberg, Germany) and 16-well dishes (Lab-Tek®, Chambered #1.0 Borosilicate Coverglass System, NUNC™, Wiesbaden, Germany) with 6 × 104 cells/cm2. For all experiments, cells were differentiated with 10 μM retinoic acid for 24 h (Sigma, Taufkirchen, Germany).

Primary dopaminergic midbrain neurons and MPP+ toxicity

Mesencephalic tissue was dissected from embryonic day 14 rats (Wistar, provided by the animal research facility of Göttingen University Hospital, Germany), and cells prepared as described previously (Lingor et al. 2000). Only about 5% of the cells cultured in this way are DA and thus sensitive to MPP+. The total number of cells can be used as an internal control for possible variations in cell density among experimental conditions or independent experiments.

The medium was changed after 1 and 3 days in vitro (DIV). On DIV 4, Tat-Hsp70, Tat-HA, Hsp70, buffer (50% glycerol, 274 mM NaCl, 10 mM Tris pH 10, 0.1% Pluronic, 0.02% Tween-80) or vehicle were applied to the culture at a concentration of 250 nM each. After 2 h of incubation, MPP+ (Sigma) was added to the culture at a final concentration of 6 μM for 24 h. Immunohistochemistry was performed with an primary antibody against TH (Advanced Immunochemical Inc., Long Beach, CA, USA; 1 : 500) and a Cy3-coupled anti-rabbit [1 : 200, in 10% normal goat serum/phosphate-buffered saline (PBS); Jackson ImmunoResearch Laboratories Inc., Baltimore, PA, USA] secondary antibody followed by nuclear staining with 4′,6-diamidino-2-phenylindole (DAPI, 2 μg/mL, 5 min). After staining, cultures were kept in PBS and numbers of TH-positive cells and DAPI-positive nuclei were counted immediately in at least three fields per wells of the 96-well plate for each condition using a fluorescence microscope (Axiovert; Zeiss, Jena, Germany). We then calculated the fraction of TH-positive cells of the total cell number. Single digital photographed pictures (Axiovert) were taken with a 20× objective and were analysed with Image J (http://rsb.info.nih.gov/ij/; free software, 1.37v, National Institutes of Health, Bethesda, MD, USA). Neurites were counted and neurite length was measured. To determine the length distribution of neurites, the untreated control condition was set to 100% for each experiment. Neurite length of Tat-Hsp70, Hsp70, Tat-HA, and buffer-treated cultures (with and without MPP+ application) were normalized against the vehicle control.

Detection of cell transduction by Tat-Hsp70 in vitro

Western blot analysis

SH-SY5Y neuroblastoma cells were treated with Tat-Hsp70 or Hsp70 (250 nM) for 4 h. Trypsin was applied for 5 min to degrade not incorporated protein before cell lysates were collected in 2× SDS buffer (4% SDS, 20% glycerol, 127 mM Tris base pH 6.8, 4 mM ethylenediaminetetraacetic acid, 2% 2-mercaptoethanol, and 0.015% bromphenol blue). For examination of intracellular stability of recombinant protein, Tat-Hsp70 (250 nM) was added for different time periods on SH-SY5Y cells (0.5, 1, 2, 12, 24, 48, 72, and 96 h). Heating (5 min, 95°C) and sonication (50%, 30 s) were followed by electrophoresis on 8% PAA gels.

Proteins were subsequently transferred to nitrocellulose membranes (AppliChem GmbH, Darmstadt, Germany) and a mouse primary antibody against the HA-tag (Covance Inc.; 1 : 3000) or against Hsp70 (Santa Cruz Biotechnology, Santa Cruz, CA, USA; 1 : 500) were applied, followed by incubation with a goat anti-mouse IgG secondary antibody (Santa Cruz Biotechnology; 1 : 3000) or a donkey anti-goat horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology; 1 : 5000). Between all steps blots were washed with Tris-buffered saline Tween-20. Labeled proteins were detected using the enhanced chemiluminescence-plus reagent (Amersham, Buckinghamshire, England, UK) and exposed to photographic film. An anti-actin primary antibody (Chemicon International, Temecula, CA, USA; 1 : 5000) and an goat anti-mouse horseradish peroxidase-conjugated secondary antibody (Santa Cruz Biotechnology; 1 : 3000) were used as loading control to assess transduction efficiency using the Quantity-One software package (Bio-Rad Laboratories, Hercules, CA, USA).

Live imaging in vitro

Tat-Hsp70 and Hsp70 (3 μg/μL) were dialysed against an amine-free buffer (0.2 M NaHCO3 pH 8.3, 50% glycerol, 3 mM NaN3). FITC isomer (Molecular Probes, Eugene, OR, USA) was used as amine reactive compound. Detailed information about the labeling procedure was given in the Handbook of Fluorescent Probes and Research Products (http://www.probes.com). FITC-Tat-Hsp70 and FITC-Hsp70 (200 nM, in 50% glycerol, 0.2 M NaHCO3) treated cells were washed with PBS before live recordings were performed by fluorescence (Zeiss) and confocal microscopy (Leica, Bensheim, Germany) 1 or 2 h after protein treatment, respectively.

Immunohistochemistry to demonstrate delivery of recombinant protein across the BBB

Adult male mice (22–30 g) purchased from Charles River Wiga (Sulzfeld, Germany) were intraperitoneally (i.p.) injected with 5 nmol Tat-Hsp70 or Hsp70 (three times in 24 h). Six hours after the final protein injection, anesthesia, brain fixation, and immunohistochemistry were carried out according to Dietz et al. (2006b); 18 μm cryosection slices were incubated with a mouse anti-Hsp70 primary antibody (Stressgen Bioreagents, Vancouver, BC, Canada; 1 : 200) and an anti-mouse-Alexa 680-conjugated secondary antibody (Molecular Probes; 1 : 200). TH staining is described in the next paragraph. Slices were incubated with DAPI (2 μg/mL for 1 h) before slides were mounted with Moviol (EMD Bioscience, La Jolla, San Diego, CA, USA). Labeling was visualized by fluorescence microscopy using a Zeiss Axioplan II microscope with Zeiss Axiovision Software (Zeiss) and a digital camera (AxioCam, Zeiss).

MPTP treatment of mice

Mouse treatment and TH immunostaining

Mice were treated according the German guidelines for the care and the use of laboratory animals and in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC).

Fusion proteins were incubated with 0.5% SDS to prevent proteins from degradation and also to block their instantaneous internalization, allowing slow diffusion and progressive release within the peritoneal cavity (Schutze-Redelmeier et al. 1996). On days 1, 2, 3, 4, 5, 7, 9, and 11 of the MPTP experiment, 14-week-old C57BL/6 mice (21.5–30.5 g; Charles River, Wilmington, MA, USA) were i.p. injected with 5 nmol of either Tat-Hsp70, control protein Tat-HA, dialysis buffer or PBS, always at the same time of the day. On five consecutive days (1, 2, 3, 4, and 5 of the experiment), 30 mg/kg MPTP were i.p. injected 6 h after protein application. In all groups, at least five animals (PBS: = 5; PBS + MPTP: = 6; buffer + MPTP: = 7; Tat-HA + MPTP: = 7; Tat-Hsp70 + MPTP: = 6) were used for evaluation. Fourteen days after the last MPTP injection, mice were killed and their brains prepared for immunostaining according to Dietz et al. (2006b).

The posterior parts of the brains including the SN were cut into 50 μm coronal sections. Every third section was incubated with an anti-TH antibody (Advanced Immunochemical Inc.; 1 : 1000) and a biotinylated goat anti-rabbit secondary antibody (Jackson ImmunoResearch Laboratories, Inc.; 1 : 200) followed by incubation with the Vectastain®ABC-Peroxidase-Kit (Vector Laboratories, Inc., Burlingame, CA, USA). The reaction was developed in 3,3′-diaminobezidine (Sigma). Of the striatum, every sixth section was collected for immunostaining using the Vectastain®Elite ABC-Kit (Vector Laboratories, Inc.).

Neurochemical analysis

Striata were rapidly dissected on ice and frozen at −80°C until the concentrations of DA, 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) were measured by HPLC with electrochemical detection as previously described (Eberhardt et al. 2000). Catecholamine concentrations were expressed in pmol per mg of striatal tissue.

MPTP metabolism

The MPP+ levels in the brain were determined by HPLC. Three mice of each group were i.p. injected with 30 mg/kg MPTP and killed 90 min later. Striata were quickly dissected on ice, homogenized in 20 μL 0.1 M perchloric acid per mg tissue and then debris removed by high-speed centrifugation; 20 μL of the supernatant were injected onto a reverse-phase column (Nucleosil-100 C18; Knauer, Berlin, Germany) and quantified by UV absorption at 300 nm (UVD 340U; Dionex, Berlin, Germany) using Chromeleon 6.60 software. The mobile phase consisted of 697/1000 mL acetonitrile in phosphate buffer (pH 2.5). The flow rate was 0.5 mL/min. Values represent pmol MPP+ per mg of wet tissue.

Nissl staining

Slices were incubated in a chloroform–ethanol solution followed by dehydration, cresyl-violet (Thionin acetate, Sigma) staining, formalin fixation and dehydration in 95% ethanol. The sections were clarified with xylene. Entellan was used to mount the sections. Details are described by Schulz et al. (1995). The striatal slices were dehydrated with increasing concentrations of ethanol, followed by xylene treatment.

Stereology

The number of TH-positive neurons in the SNpc was assessed using stereological methodology. Here, the number of TH-positive neurons in the SNpc and Nissl-positive neurons was counted from the left side of each 3,3′-diamino-benzidine (DAB)- and Nissl-stained section from all animals per group by using the optical dissector technique (Stereo Investigator 6.0, MicroBrightField Inc., Zeiss microscope; Dehmer et al. 2004). Counts were performed manually and blinded for treatment. We evaluated striatal fiber density by subtracting the optical density of the cortex from the striatum using Scion Image 4.0.3.2 software (Windows ‘95-XP, Washington, DC, USA; Scion Corporation, Frederick, MD, USA).

Statistics

kyplot (version 2.0 Beta 15, 1997–2001 Koichi Yoshioka) (Tokyo, Japan) was used to perform statistical analysis. All in vitro experiments were performed at least in triplicate and were repeated at least three times. Data are given as mean ± SE. Intergroup differences were considered significant at *< 0.05, **< 0.01, and ***< 0.001 according to one-way anova followed by Dunnett-T3 post hoc test.

Results

Transduction of SH-SY5Y cells with Tat-Hsp70

We verified the purity and concentration of our fusion proteins purified from bacterial extracts on Coomassie-stained PAA gels (Fig. 1). Next, we treated SH-SY5Y cells with Tat-Hsp70 (250 nM) for 4 h before trypsin was applied to degrade non-translocated protein. Cell lysates were examined by western blot analysis, whereas Hsp70 without the Tat domain was degraded by trypsin (Fig. 2a, lane 5), Tat-Hsp70 was protected from degradation by trypsin (Fig. 2a, lane 4), suggesting that Tat-Hsp70 transduced these cells in vitro. Neither Tat-Hsp70 nor Hsp70 were detectable in the medium 4 h after trypsin application (Fig. 2a, lanes 2 and 3), indicating that trypsin treatment was effective. To examine the stability of transduced protein, SH-SY5Y lysates were prepared at different time points (up to 96 h) after addition of Tat-Hsp70 (Fig. 2b, lanes 3–10). As shown by western blot analysis using an anti-HA antibody, the fusion protein quickly transduced the cells (Fig. 2b, lane 3). Using densitometric analysis we showed that over 11% of Tat-Hsp70 was still detectable after 96 h (lane 10) compared with the 2 h time point, at which the highest amount was detected (Fig. 2b, lane 5). To estimate the increase in total amount of Hsp70 by Tat-mediated transduction, we next used a primary antibody against Hsp70 (Fig. 2c). The intracellular amount of recombinant Hsp70 was approximately four times higher than the endogenous one 1 h after adding the recombinant protein (Fig. 2c, lane 2). Thus, Tat-Hsp70 quickly and efficiently transduces SH-SY5Y neuroblastoma cells and is stable there for at least several days. We confirmed efficient transduction of Tat-Hsp70 by immunocytochemistry (data not shown). Furthermore, fluorescently labeled Tat-Hsp70 and Hsp70 was added to SH-SY5Y cells (Fig. 3). Cells were washed to remove protein in the cell culture medium after 1 h. Immunofluorescence was detected on unfixed cells to avoid artificial intracellular redistribution of the transduced protein due to paraformaldehyde fixation (Richard et al. 2003). These results demonstrate a cellular signal only for the Tat-Hsp70 treated cells, confirming that the Tat-domain is responsible for an efficient transduction (Fig. 3a–f). We also performed live imaging with confocal microscopy to demonstrate the subcellular distribution of the protein 2 h after Tat-Hsp70 treatment (Fig. 3g–l).

Figure 2.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) rapidly transduced SH-SY5Y neuroblastoma cells and is stable inside the cells. (a) Western blot analysis demonstrating that Tat-Hsp70 is protected from degradation by trypsin applied on differentiated SH-SY5Y neuroblastoma cells. SH-SY5Y neuroblastoma cells were treated with Tat-Hsp70 or Hsp70 (250 nM) for 4 h. Trypsin was applied for 5 min to degrade non-transduced protein before cell lysates were collected in 2× sodium dodecyl sulfate (SDS) buffer. After electrophoretical separation proteins were transferred to nitrocellulose membrane and were incubated with an antibody directed against the HA tag. Lane 1, purified Tat-Hsp70 protein without trypsin treatment, lanes 2 and 3, trypsin-treated medium with Tat-Hsp70 and Hsp70, lanes 4 and 5, lysates of Tat-Hsp70 and Hsp70 treated cells after trypsinisation. (b) SH-SY5Y cell lysates were prepared after Tat-Hsp70 treatment (250 nM) at different time points (0.5, 1, 2, 12, 24, 48, 72, and 96 h). Hsp70-treated cells (250 nM) were used as control. Western blots stained with an antibody against the HA-tag of SH-SY5Y cell extracts. Lane 1, purified Tat-Hsp70 (1 pmol), lane 2, lysate of untreated cells, lane 3–10, lysates of Tat-Hsp70-treated cells, lanes 11 and 12, medium treated with Tat-Hsp70 for the indicated time periods, lanes 13 and 14, cell lysate and medium of Hsp70 treated cells. (c) The ratio of recombinant and endogenous Hsp70 is approximately 4 : 1. Quantification was done 1 h after cells were treated with Tat-Hsp70. Lane 1, lysates of untreated cells, lane 2, lysates of Tat-Hsp70-treated cells. A goat anti-Hsp70 antibody was used.

Figure 3.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) efficiently transduces SH-SY5Y cells, as visualized via imaging of live cells. Tat-Hsp70 and Hsp70 were covalently linked to an amine reactive dye (FITC) and applied on SH-SY5Y cells. FITC–Tat-Hsp70 and FITC–Hsp70 (200 nM) treated cells were washed with phosphate-buffered saline (PBS) before live recordings were done by fluorescence microscopy (a–f) 1 h after protein treatment, and by confocal microscopy (g–l) 2 h after protein treatment. Cellular labeling was done with CM-DiI (g and j).

Tat-Hsp70 protects dopaminergic midbrain neurons against MPP+ toxicity

The effect of Tat-Hsp70 on neuronal survival was evaluated in primary mesencephalic neurons. Cultures were treated with vehicle, buffer, Hsp70, Tat-HA, or Tat-Hsp70 on DIV 4. On DIV 5 the number of DA neurons not treated with MPP+ was similar in all experimental groups (Fig. 4a and d, bars 1–5). Treatment of cultures with 6 μM MPP+ for 24 h on DIV 4 resulted in a 45–56% decrease in cell survival, which was similar in vehicle, buffer, Hsp70, and Tat-HA treated cultures (Fig. 4b and d, bars 6–9). Treatment with Tat-Hsp70 significantly inhibited the loss of TH-positive neurons (Fig. 4c and d, bar 10). Additionally, MPP+ treatment reduced the mean length of neurites of DA midbrain neurons by 50% compared with untreated control condition (Fig. 5a, b, and d, bars 1 and 6). Neurite processes of vehicle, buffer, Hsp70, and Tat-HA treated TH-positive neurons were similarly damaged by MPP+ treatment (Fig. 5d, bars 6–9), whereas Tat-Hsp70 blocked MPP+-induced degeneration of neurites by 50% (Fig. 5c and d, bar 10).

Figure 4.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) protects cultured dopaminergic midbrain neurons against MPP+ toxicity. Mesencephalic neurons (a) untreated, or (b) 24 h after the application of 6 μM MPP+, or (c) 24 h after MPP+ treatment preceded by Tat-Hsp70 application. Scale bar: 100 μm. (d) Fraction of tyrosine hydroxylase (TH)-positive midbrain neurons. Results are average ± SEM (= 4, **< 0.01 compared with results represented in bars 6–9). Conditions were normalized against control condition (100%).

Figure 5.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) protects cultured mesencephalic dopaminergic neurons from MPP+-induced loss of their processes. Tyrosine hydroxylase (TH)-positive neurons were visualized by immunocytochemical staining. (a) Untreated cell culture. (b) 24 h application of MPP+ (6 μM) (c) Tat-Hsp70 and MPP+ treatment. Scale bar 50 μm. (d) To determine the length distribution of neurites, untreated control condition was set to 100% for each experiment. Neurite length of Tat-Hsp70, Hsp70, Tat-hemagglutinin (HA), and buffer-treated cultures (with and without MPP+ application) were normalized against vehicle control. Results are average ± SEM (n ≥ 3, ***< 0.001 compared with bars 6–9).

Systemically applied Tat-Hsp70 co-localized with dopaminergic neurons in the substantia nigra pars compacta

Previous studies have shown that the HIV Tat-domain mediates the delivery of certain cargoes like glial cell line-derived neurotrophic factor, X-chromosome-linked inhibitor of apoptosis protein (XIAP) or the B-cell lymphoma x protein (Bcl-xL) into the brain (Diem et al. 2005;Dietz et al. 2006b; Guegan et al. 2006). To determine whether Tat-Hsp70 could be systemically applied in in vivo models for PD and whether it reaches the SNpc, we i.p. injected Tat-Hsp70 or Hsp70 and performed immunohistochemical analysis. Brain sections from Tat-Hsp70-treated mice displayed a bright cell-specific staining when an antibody against Hsp70 was applied (Fig. 6c), indicating transduction of Tat-Hsp70 through the BBB into the brain, including tyrosine-hydroxylase-positive neurons in the SNpc (Fig. 6a, b, and d). Mice injected with Hsp70, that did not include the Tat domain, showed a weak signal for Hsp70 staining (Fig. 6g), suggesting no transduction across the BBB into the SNpc (Fig. 6g and h). Successful transduction of the SNpc after systemic Tat-Hsp70 application encouraged us to test the effect of the recombinant protein in an in vivo model for PD.

Figure 6.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) transduces the substantia nigra pars compacta (SNpc) after intraperitoneal injection. Brain sections including the area of the SNpc (a and e) of i.p. injected mice with Tat-Hsp70 (a–d) or Hsp70 without the Tat domain (e–h) are shown. Immunohistochemistry was performed using an antibody against tyrosine hydroxylase (TH) (a, b, e, and f) as a marker for dopaminergic neurons and against Hsp70 (c and g). Merged pictures of TH, Hsp70 immunopositive neurons and 4′,6-diamidino-2-phenylindole-staining are shown in (d and h). Scale bar = 200 μm (a and e) or 50 μm (b–d; f–h).

Tat-Hsp70 protects dopaminergic neurons in the SNpc against systemic application of MPTP

To examine whether Tat-Hsp70 protects against MPTP toxicity in vivo, we treated mice during and after MPTP intoxication with Tat-Hsp70 or control protein (Mouse treatment and TH immunostaining, Fig. 7). Using stereologic analysis, we found that treatment with Tat-Hsp70 reduced the loss of DA neurons in the SNpc. Compared with PBS-treated mice (Fig. 8a, d, and g, bar 1), a reduction in the cell number by 48–54% was observed after five consecutive injections of MPTP (30 mg/kg) (Fig. 8b, e, and g, bars 2–4). Tat-Hsp70 but not the control treatments reduced the loss of TH-positive cell bodies (Fig. 8c, f, and g, bar 5). To rule out that the observed decrease in TH-positive cells in the SNpc merely reflects a transient cellular atrophy and down-regulation of TH expression rather than true cell death of DA neurons, we also counted Nissl-positive neurons in the SNpc. These data confirmed our results from the TH-positive cell counting. The total number of neurons in the PBS control group was 17 202 ± 767 when compared with 9460 ± 735 in the MPTP-treated group. These values slightly further decreased with buffer (9073 ± 532) and Tat-HA (8777 ± 435) application under MPTP treatment, but increased cell numbers were detected for Tat-Hsp70-injected mice (13 100 ± 658).

Figure 7.

 MPTP and protein treatment of mice. On days 1, 2, 3, 4, 5, 7, 9, and 11 of the MPTP experiment, 14-week-old C57BL/6 mice were i.p. injected 5 nmol trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) or control protein Tat-hemagglutinin (HA), dialysis buffer, or phosphate-buffered saline (PBS), always at the same time of the day. On five consecutive days (1, 2, 3, 4, and 5 of the experiment), 30 mg/kg MPTP were i.p. injected 6 h after protein application. On day 19 mice were killed and brains were prepared for immunostaining and HPLC analysis.

Figure 8.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) protects tyrosine hydroxylase (TH) positive, dopaminergic somata in the substantia nigra pars compacta (SNpc) against MPTP-induced cell death. (a–f) Tyrosine-hydroxylase-immunopositive neurons in coronal brain sections of the midbrain after phosphate-buffered saline (PBS), MPTP + PBS or Tat-Hsp70 + MPTP treatment (Scale bar: 600 μm in a–c; scale bar: 150 μm in d–f). PBS control (a and d), only MPTP (b and e) and MPTP + Tat-Hsp70 treatment (c and f). (g) Quantitative analysis of TH-positive neurons in SNpc. Results are average ± SEM (PBS: = 5; PBS + MPTP: = 6; buffer + MPTP: = 7; Tat-hemagglutinin (HA) + MPTP: = 7; Tat-Hsp70 + MPTP: = 6; **< 0.01 compared with bars 2–4).

The integrity and function of DA neurons does not only depend on the survival of its somata but also on the rescue of its neurites and synaptic terminals. To determine whether the rescue of DA neurons by Tat-Hsp70 treatment also preserved their striatal axon terminals, we evaluated striatal fiber density and catecholamine levels (Figs 9 and 10a–c). MPTP decreased striatal fiber density by 63–72% (Fig. 9b and d, bars 2–4) compared with PBS-treated controls (Fig. 9a and d, bar 1). There was a significant increase by 16–27% in the density of the striatal network of DA fibers in Tat-Hsp70-treated animals when compared with other MPTP-treated conditions (Fig. 9c and d, bar 5 compared with bars 2–4, < 0.05). Striatal DA and its metabolites dihydroxyphenylacetic acid (DOPAC) and HVA were quantified by HPLC. There was a strong depletion of DA (> 90%), DOPAC (> 80%), and HVA (> 55%) following MPTP treatment when compared with the PBS-treated group (Fig. 10a–c). Although a moderate increase in DA and metabolites was detected with Tat-Hsp70 treatment, it did not reach statistical significance.

Figure 9.

 Trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) diminishes MPTP-induced decrease in dopaminergic striatal fiber density. Tyrosine hydroxylase (TH)-staining of the striatum for phosphate-buffered saline (PBS) (a), MPTP + phosphate-buffered saline (PBS) (b), and MPTP + Tat-Hsp70 (c) injected animals, was normalized against TH staining of the cortex. (d) Evaluation of striatal fiber density. Results are average ± SEM (PBS: = 5; PBS + MPTP: = 6; buffer + MPTP: = 7; Tat-hemagglutinin (HA) + MPTP; = 7; Tat-Hsp70 + MPTP: = 6; *< 0.05).

Figure 10.

In vivo transduction of trans-activator of transcription-heat shock protein 70 (Tat-Hsp70) does not protect against the MPTP-induced decrease in striatal catecholamine concentrations. Fourteen days after the last MPTP injection, mice were killed and concentrations of dopamine (a), 3,4-dihydroxyphenylacetic acid (DOPAC) (b) and homovanillic acid (HVA) (c) in the striatum were measured by HPLC analysis. Results are average ± SEM. None of the differences among the MPTP-treated groups were statistically significant [phosphate-buffered saline (PBS): = 5, PBS + MPTP: = 6, buffer + MPTP: = 7, Tat-hemagglutinin (HA) + MPTP: = 7, Tat-Hsp70 + MPTP: = 6].

To rule out that Tat-Hsp70 diminished neuronal damage by interfering with MPTP metabolism we measured MPP+ concentrations (pg/mg tissue, average ± SEM) in the SN and in the striatum 90 min after i.p. injection of MPTP. We did not detect a difference in MPP+ levels among the different treatment groups, neither in the SN (PBS: 13.46 ± 1.89, MPTP + Tat-HA: 11.29 ± 1.03, MPTP + Tat-Hsp70: 11.68 ± 2.86) nor in the striatum (PBS: 12.67 ± 1.2; MPTP + Tat-HA: 10.49 ± 1.95; MPTP + Tat-Hsp70: 16.63 ± 2.47), consistent with direct protection effect of Tat-Hsp70 on neuronal survival and integrity.

Discussion

In recent years, protein transduction technology has been used in many different paradigms of neurodegeneration (Dietz et al. 2002; Wheeler et al. 2003; Dietz and Bähr 2005; Soane and Fiskum 2005). Here, we demonstrate that Tat-Hsp70 successfully transduced cells in vitro, suggesting that Tat-mediated delivery of chaperones such as Hsps’ is a feasible therapeutic approach. We investigated whether Tat-Hsp70 fusion protein was neuroprotective in the MPTP model for PD in vitro and in vivo.

Previous results have shown that increased expression of Hsp70 contributes to the protection of different cell lines against excitotoxicity, oxidative, and thermal stress (Amin et al. 1995; Wong et al. 1996; Kelly et al. 2001; Quigney et al. 2003; Lai et al. 2005). Our results indicate that Tat-Hsp70 treatment reduces not only the death of DA midbrain neurons from MPP+ toxicity in vitro and from MPTP toxicity in vivo, but also protects the neuritic processes in vitro and striatal axon terminals in vivo. In contrast to our findings, purely anti-apoptotic treatments like adenoviral gene transfer of XIAP or the caspase inhibitor zVAD-fmk were not able to protect DA terminals from their demise. Only a combination of XIAP and the glial cell line-derived neurotrophic factor provided protection for both somata and axon terminals against MPTP toxicity (Eberhardt et al. 2000), which indicates that axonal degeneration and the loss of neuronal somata might be independent processes (Finn et al. 2000). Our findings underline the therapeutic potential of Hsp70: like other ‘dirty’ drugs, Hsp70 interferes with many components of the ‘death cascade’, which appears to be necessary to protect DA neurons in their integrity from MPTP toxicity and potentially in PD.

These findings are in line with the result of an adeno-associated virus (AAV)–Hsp70 injection in the MPTP mouse model (Dong et al. 2005). At this point, we cannot determine whether an application of Tat-Hsp70 after the onset of MPTP treatment would be beneficial. However, DA cell loss in the SN induced by high-dose MPTP ceases a few days after MPTP application (Tatton and Kish 1997), while the number of these cells continues to decline in PD patients. Thus, even if later studies reveal that the time window for therapeutic Tat-Hsp70 application closes briefly after the first MPTP application in the mouse, it might still be beneficial to impede DA neuron decline in patients.

Another way to increase the levels of Hsp70 is the application of geldanamycin. In the MPTP mouse model, preservation of TH immunoreactivity and DA content was observed by intracerebral ventricular injection of geldanamycin 24 h prior MPTP treatment (Shen et al. 2005). Further in vivo studies have shown the protection of Hsp70 in various models of injury (Yenari et al. 1998, 2005; Dong et al. 2005), including models in which α-synuclein aggregation and toxicity is observed (Klucken et al. 2004; Song et al. 2004; Fornai et al. 2005). Breeding of Hsp70 over-expressing mice with α-synuclein transgenic mice reduced the development of α-synuclein phenotypes, suggesting that Hsp70 can reduce the amount of misfolded and aggregated α-synuclein species and protects mice from α-synuclein-dependent toxicity (Klucken et al. 2004).

The subacute MPTP paradigm used in our experiment does not induce aggregates, which is in line with other studies (Shimoji et al. 2005). Therefore, in the present work it was not possible to test whether Hsp70 interacts with α-synuclein. We cannot rule out the possibility that Hsp70 provides neuroprotection by interfering with the generation of toxic α-synuclein oligomers. Such oligomers are not possible to detect in brain tissue.

In contrast to our findings, the MPTP-induced decline in the striatal DA level was significantly reduced by AAV–Hsp70 (Dong et al. 2005). In our model, the depletion in the level of DA and its metabolites DOPAC and HVA after subacute application of MPTP was only moderately but not significantly elevated in the Tat-Hsp70-treated animal group compared with control groups. This difference is not likely to be a result of the different forms of application, i.e. AAV mediated versus Tat mediated, but rather of the intensity of the insult. The application of 4 × 20 mg/kg MPTP spaced by 24 h intervals (Dong et al. 2005) led to only a 50% reduction of striatal DA concentration, whereas our treatment with 5 × 30 mg/kg MPTP led to a 91% decrease. This strong depletion may be more difficult to alleviate by therapeutic treatments. Because loss of TH-fiber density was reduced by Tat-Hsp70 treatment, chances are high that DA concentration may also be restored at later time points, which we did not investigate.

This suggests on the one hand a concomitant rescue of TH-positive somata in the SNpc and their cell processes in the striatum, on the other hand decreased functionality of synaptic terminals and distal axons as the major site of MPTP damage in neurons (Linder et al. 1987). Whether surviving neurons can regain function after a longer regeneration period remains to be examined.

Theoretically, the Tat-Hsp70-mediated protection in our MPTP model could occur because of interference of recombinant protein with MPTP, changing MPTP uptake into the brain or conversion into MPP+. To address this question, we measured the MPP+ concentrations in the SN and in the striatum. We did not detect a difference in MPP+ levels among the different treatment groups, suggesting inhibition of cell death by Tat-Hsp70.

In summary, the data we reported here demonstrated the efficient delivery of the anti-apoptotic Hsp70 via the HIV-derived Tat domain into cells. Tat-Hsp70 prevents the MPTP-mediated loss of DA midbrain neurons in culture as well as in the SNpc in a widely used rodent mouse model for PD. Our results provide evidence that the Tat-Hsp70 represents a novel and potentially valuable therapeutic agent for the treatment of PD.

Acknowledgements

We thank Cathy Ludwig for proofreading the manuscript. We thank Veronique Planchamp and Johanna Knöferle for primary midbrain cultures, as well as Birgit Dietz and Christine Poser for excellent technical assistance. We thank R. Hardeland for many useful discussions and Katarzyna Lisowska (Department of Tumor Biology, Centre of Oncology, Gliwice, Poland) for the kind gift of the rat Hsp70.1 clone. S. F. Dowdy (San Diego, CA) kindly provided the pTat-HA vector. Supported by the Research Center for Molecular Physiology of the Brain (CMPB) of the Deutsche Forschungsgemeinschaft (DFG).

Ancillary