Efficient in vivo protection of nigral dopaminergic neurons by lentiviral gene transfer of a...

www.elsevier.com/locate/yexnr

Experimental Neurology

Regular Article

Efficient in vivo protection of nigral dopaminergic neurons by

lentiviral gene transfer of a modified Neurturin construct

Lone Fjord-Larsen, Jens Leander Johansen1, Philip Kusk, Jens Tornøe, Mette Grønborg,

Carl Rosenblad2,3, Lars U. Wahlberg*,3

NsGene A/S, Baltorpvej 154, 2750 Ballerup, Denmark

Received 24 November 2004; revised 28 February 2005; accepted 21 March 2005

Available online 24 May 2005

Abstract

Protein injection studies of the glial cell line derived neurotrophic factor (GDNF) family member Neurturin (NTN) have demonstrated

neuroprotective effects on dopaminergic (DA) neurons, which are selectively lost during Parkinson’s disease (PD). However, unlike GDNF,

NTN has not previously been applied in PD models using an in vivo gene therapy approach. Difficulties with lentiviral gene delivery of wild

type (wt) NTN led us to examine the role of the pre-pro-sequence, and to evaluate different NTN constructs in order to optimize gene therapy

with NTN. Results from transfected cultured cells showed that wt NTN was poorly processed, and secreted as a pro-form. A similarly poor

processing was found with a chimeric protein consisting of the pre-pro-part from GDNF and mature NTN. Moreover, we found that the

biological activity of pro-NTN differs from mature NTN, as pro-NTN did not form a signaling complex with the tyrosine kinase receptor Ret

and GFRa2 or GFRa1. Deletion of the pro-region resulted in significantly higher secretion of active NTN, which was further increased when

substituting the wt NTN signal peptide with the immunoglobulin heavy-chain signal peptide (IgSP). The enhanced secretion of active mature

NTN using the IgSP-NTN construct was reproduced in vivo in lentiviral-transduced rat striatal cells and, unlike wt NTN, enabled efficient

neuroprotection of lesioned nigral DA neurons, similar to GDNF. An in vivo gene therapy approach with a modified NTN construct is

therefore a possible treatment option for Parkinson’s disease that should be further explored.

D 2005 Elsevier Inc. All rights reserved.

Keywords: Neurturin; Signal peptide; Pro-peptide processing; 6-hydroxydopamine lesion; Lentiviral delivery; Gene therapy; Parkinson’s disease

Introduction

Neurturin (NTN) is a member of the glial cell line

derived neurotrophic factor (GDNF) family of neurotrophic

factors. Recombinant NTN protein has been shown to

promote the survival and maintenance of several different

neuronal subtypes including autonomic and somatic periph-

eral ganglion neurons, striatal neurons, and nigral dopami-

nergic (DA) neurons. Signaling of GDNF family ligands is

0014-4886/$ - see front matter D 2005 Elsevier Inc. All rights reserved.

doi:10.1016/j.expneurol.2005.03.006

* Corresponding author. Fax: +45 44608989.

E-mail address: [email protected] (L.U. Wahlberg).1 Current address: Lundbeck A/S, 2500 Valby, Denmark.2 Current address: Department of Surgery, Lund University Hospital,

Lund, Sweden.3 Contributed equally to this work.

mediated through a receptor complex consisting of one type

of four glycosyl phosphatidylinositol-linked ligand-binding

receptors (GFRa1–4) and the transmembrane tyrosine

kinase receptor Ret (Baloh et al., 2000). Although NTN

preferentially binds to GFRa2, cross-binding to GFRa1 also

occurs (Jing et al., 1997; Klein et al., 1997; Sanicola et al.,

1997). Studies on cell cultures derived from animals lacking

GFRa1 have shown that, in the nigral DA neurons, NTN

appears to signal through GFRa1 (Cacalano et al., 1998).

This is consistent with the high expression of GFRa1 in the

developing and adult nigral DA neurons while GFRa2 is

only modestly expressed, if at all, in this area (Golden et al.,

1998; Widenfalk et al., 1997).

In studies using recombinant protein injections, NTN has

been found to prevent the loss of damaged nigral dopamine

neurons in an animal model of Parkinson’s disease (PD) and

195 (2005) 49 – 60

L. Fjord-Larsen et al. / Experimental Neurology 195 (2005) 49–6050

to enhance the function of intact DA neurons in vivo (Horger

et al., 1998; Rosenblad et al., 2003). The neuroprotective and

functional effects of NTN are equipotent to GDNF when

tested on developing ventral mesencephalic DA neurons

cultured in vitro (Cacalano et al., 1998; Horger et al., 1998)

as well as on damaged nigrostriatal DA neurons in vivo

(Horger et al., 1998). Based on the neuroprotective effects on

nigrostriatal DA neurons, NTN has been suggested as a

candidate for treatment of PD. However, the progressive

nature of PD, as well as findings from numerous preclinical

studies using GDNF, indicates that continuous administra-

tion over long periods of time may be necessary for

neuroprotection and functional recovery in the nigrostriatal

DA system. Intracerebral protein delivery is cumbersome

and inefficient, making therapeutic approaches that use

delivery by viral gene transfer attractive choices. To date,

several recombinant viral vectors have been tested for long-

term delivery of neurotrophic factors in vivo. Recombinant

adeno-associated virus (AAV) and lentivirus (rLV) are the

most extensively used in the nervous system due to their

ability to infect non-dividing cells and provide stable

expression (Bensadoun et al., 2000; Bilang-bleuel et al.,

1997; Connor et al., 2001; Georgievska et al., 2004;

Kordower, 2003; Lapchak et al., 1997; Mandel et al.,

1997; Rosenblad et al., 2000). In vivo gene therapy with

wild type (wt) GDNF has successfully been applied in

different experimental paradigms. However, data from our

lab indicate that wt NTN is not readily expressed in vivo

following transduction of cells using an rLV vector contain-

ing human wt NTN cDNA (unpublished results). The reason

for the difference in GDNF and NTN behavior is unknown

but one hypothesis could be a faulty processing of the NTN

protein. Human wt NTN encodes a 197-residue pre-pro-

protein with a predicted 19 amino acid signal peptide,

followed by a pro-region of 76 amino acids. Pro-NTN is

proteolytically cleaved at the RXXR consensus recognition

sequence for furin-like proteases to generate the mature NTN

of 102 residues.

In the present study, we investigated the role of the pre-

pro-region of human NTN in the secretion of active NTN

protein from cultured cells transfected with NTN constructs

in which different parts of the pre-pro-sequences had been

modified. We found that wt NTN was poorly processed to

the mature form and that the secreted pro-form of NTN was

not able to bind to GFRa1 or GFRa2 receptors. Deletion of

the pro-region resulted in significantly increased secretion

of active NTN, and by using a heterologous signal peptide,

the immunoglobulin heavy-chain signal peptide (IgSP)

sequence, secretion of active NTN was further increased

to more than 100-fold the levels obtained with the wt NTN

sequence. The IgSP-NTN expression construct was tested in

the intrastriatal 6-hydroxydopamine lesion model of PD via

a lentiviral in vivo gene transfer approach. Unlike wt NTN,

the IgSP-NTN gave potent neuroprotective effects demon-

strating for the first time the potential of NTN in vivo gene

therapy for PD.

Methods

DNA vectors

wt NTN constructs (pHRV-CMV.SIN.hNTN.WPRE and

pNS1n.hNTN)

Wild type human pre-pro-NTN was cut from vector

pJDM2174 (human pre-pro-NTN in pBluescript; a kind gift

from Dr. Jeffrey Milbrandt, Wash. U., St. Louis, USA) as a

BamHI, XhoI fragment and cloned into pHRV-CMV.SIN-

PLT7.WPRE, a modified derivative of pHRV-CMV-SIN-18

lentiviral transfer vector containing the Woodchuck hepatitis

virus postregulatory element (WPRE) (Zufferey et al., 1997,

1998). Similarly, the BamHI, XhoI fragment was cloned in

pNS1n (a kind gift from P. Ahring NeuroSearch A/S, DK) a

derivative of pcDNA3.1 from Invitrogen, Denmark, to yield

pNS1n.hNTN. In both vectors, the CMV promoter drives

the expression of NTN expression.

ppG-NTN construct (pNS1n.ppGDNF.hNTN)

The pre-pro-region of GDNF was PCR amplified from a

full-length human GDNF clone using the following primers:

5V primer: 5V-TATAGAATTCGCCACCATGAAGTTATGG-GATGTCG-3V and 3V primer: 5V-CCAACCGCGCCCTTTT-CAGTCTTTTAATGG-3V. The 3V primer contains 10 bases of

the 5V end of mature NTN in the 3V end. Mature human NTN

was PCR amplified from human genomic DNA using the

following primers: 5V primer: 5V-ACTGAAAAGGGCGC-GGTTGGGGGCGCGGCCT-3V and 3V primer: 5V-TAGACT-CGAGGTCGACGGTATC-3V. The 5V primer contains 10

bases of the 3V end of the pro-region of human GDNF. The

pre-pro-region of GDNF was fused to mature NTN by

overlapping PCR using the following primers: 5V primer: 5V-TATAGAATTCGCCACCATGAAGTTATGGGATGTCG-

3V and 3V primer: 5V-TAGACTCGAGGTCGACGGTATC-3V.The resulting pre-pro-GDNF-mature NTN fragment was

digested with EcoRI and XhoI and inserted between EcoRI

and XhoI sites of expression vector pNS1n (described

above).

IgSP-NTN construct (pHRV-CMV.SIN.IgSP.NTN.WPRE and

pNS1n.IgSP.NTN)

The signal peptide from mouse immunoglobulin heavy-

chain gene V-region (GenBank acc. no.: M18950) (IgSP)

was PCR amplified from pNUT-IgSP-hCNTF (ref. US

6,361,741) using the following primers: 5V primer: 5V-TAQTAGGATCCGCCACCATGAAATGCAGCTGGGTTATC-

3V, 3V primer: 5V-CCAACCGCGCCGAATTCACCCCTG-TAGAAAG-3V. The 3V primer contains 10 bases from the 5Vend of human, mature NTN sequence. Human, mature NTN

was PCR amplified from human genomic DNA using the

following primers: 5V primer: 5V-GGTGAATTCGGCGC-GGTTGGGGGCGCGGCCT-3V and 3V primer: 5V-TATACT-CGAGTCACACGCAGGCGCACTCGC-3V. The 5V primer

contains 10 bases from the 3V end of the IgSP sequence. The

IgSP-human mature NTN sequence was generated by

L. Fjord-Larsen et al. / Experimental Neurology 195 (2005) 49–60 51

overlapping PCR using the IgSP and human mature NTN

PCR fragments (described above) as templates and the

following primers: 5V primer: 5V-TATAGGATCCGCCAC-CATGAAATGCAGCTGGGTTATC-3V and 3V primer: 5V-TATACTCGAGTCACACGCAGGCGCACTCGC-3V. The

final PCR fragment containing IgSP fused to human mature

NTN was digested with BamHI and XhoI and cloned

between BamHI and XhoI sites of pHRV-CMV.SIN-

PLT7.WPRE (described above) and pNS1n (described

above).

Dpro-NTN construct (pNS1n-Dpro-NTN)The human delta-pro DNA sequence, containing the pre-

but not pro-regions of wt NTN, was generated in one PCR

reaction using human genomic DNA as template and the

following primers: 5V primer: 5V-TATAGGATCCGCCACC-ATGCAGCGCTGGAAGGCGGCGGCCTTGGCCTCAG-

TGCTCTGCAGCTCCGTGCTGTCCGCGCGG-

TTGGGGGCGCGG-3V and 3V primer: 5V-TATACTCGAGTQCACACGCAGGCGCACTCGC-3V. The delta-pro-NTN

PCR fragment was digested with BamHI and XhoI and

cloned between BamHI and XhoI sites of pNS1n (described

above).

Production of lentiviral vectors

Replication-defective lentivirus particles were generated

by co-transfection of each of the different transfer vector

constructs (described above) with pMD.G (VSV-G pseudo-

typing vector) and pBR8.91 (packaging vector) (Zufferey

et al., 1999) into 293 T cells providing the required viral

proteins in trans. Briefly, 293 T cells cultured in DMEM

with 4.5 g/l glucose and Glutamaxi (Invitrogen, DK)

supplemented with 10% FCS (Life Technologies, DK)

were seeded in T75 flasks (2 � 106 cells/flask) the day

before transfection. For each T75 flask, cells were trans-

fected with 5 Ag pMD.G, 15 Ag pBR8.91, and 20 Ag of

transfer vector using Lipofectamine+i (Invitrogen, DK)

according to the manufacturer’s instructions. Virus con-

taining cell supernatant was collected 2–3 days after the

transfection, filter-sterilized through a 0.45-Am cellulose

acetate or polysulphone filter, and concentrated by ultra

centrifugation at 50,000 � g for 90 min at 4-C. After a

second round of ultra centrifugation, the concentrated virus

pellet was resuspended in DMEM and aliquoted and stored

at �80-C. To determine virus titer, reverse transcriptase

(RT) activity was assayed (Cepko and Pear, Current

Protocols in Molecular Biology, 9.13.5-6, supplement 36)

and transducing units (TU)/ml were calculated from the

determined RT activity using an EGFP lentivirus with

known transducing activity as reference.

Cell culture

ARPE-19, a spontaneously arising human retinal pig-

ment epithelial cell line (Dunn et al., 1996), was grown in

medium consisting of DMEM/Nutrient Mix F-12 with

Glutamax (Invitrogen, DK), supplemented with 10% fetal

bovine serum (Sigma-Aldrich, DK). HiB5 (Renfranz et al.,

1991), HEK293, and CHO cells were grown in DMEM

(Invitrogen, DK) with 10% fetal bovine serum (Invitrogen,

DK), and medium for CHO cells was further supplemented

with 20 mg/l l-proline. ARPE-19, HEK293, and CHO cells

were grown at 37-C and HiB5 cells at 33-C in 5% CO2.

Transient transfection studies

Cells were seeded in 6-well plates (Corning Costar,

Biotech Line, DK) at a density of approximately 105 cells/

well. The next day, cells were transfected in triplicate wells

with the different expression plasmids. ARPE-19 cells were

transfected using Fugene6 and 3 Ag plasmid/well according

to the manufacturer’s instructions, whereas the other three

cell lines were transfected in triplicate wells using 2 Agplasmid/well and Lipofectamine Plus (Invitrogen, DK)

according to the manufacturer’s instructions. The next day,

fresh growth medium was added to the wells, and cells were

incubated for an additional 24 h before collecting condi-

tioned medium and harvesting the cells. A transfection

efficiency of more than 60% was ensured by evaluation of

EGFP expression in wells transfected in parallel with the

same vector backbone containing the cDNA for EGFP.

NTN Western blotting

Cell lysates were prepared in 96-C hot sample buffer (2%

SDS, 100 mM DTT, 60 mM Tris, pH 7.5, bromphenolblue).

5� concentrated sample buffer was added to conditioned

media. In some experiments, NTN was captured by GFRa2

from conditioned medium by incubating samples for 3 h in

ELISA plates coated with goat anti-human Fc (Jackson

ImmunoResearch, Trichem, DK) bound to a GFRa2-Ig

fusion protein (R&D Systems, UK) before hot sample buffer

was added. After boiling for 5 min, samples were loaded

onto 8–18% gradient SDS gels and electrophoresed

followed by electroblotting to PVDF membranes. NTN

was detected using polyclonal NTN antibody (#AF477,

R&D Systems, UK) diluted 1:500 followed by HRP-linked

anti-goat antibody. Bands were detected by chemolumines-

cence using the ECL+ system (Amersham Life Science).

GFRa2 ELISA

The GFRa2 ELISA detects the ternary complex for-

mation between NTN, Ret, and GFRa2 and was performed

as previously described with minor modifications (Sanicola

et al., 1997). Briefly, an Opti-plate plate (Packard Instru-

ments, Perkin Elmer, DK) was coated with 100 Al 1 Ag/ml

goat anti-human Fc (Jackson Immunoresearch Laboratories,

TriChem, DK) in 50 mM NaHCO3 (pH = 9.6). Wells were

blocked in 0.2% I-Block (Tropix, Roche, DK) in PBST

(PBS with 0.1% Tween-20) for 1 h at room temperature.


Samples from NTN-producing cells or standards (recombi-

nant human NTN from R&D Systems, UK) diluted in

ARPE-19 growth medium were added together with 1 Ag/ml

GFRa2/Fc fusion protein (R&D Systems, UK) in condi-

tioned medium from 293 EBNA cells expressing Ret-

Alkaline Phosphatase (Ret-AP) fusion protein (a kind gift

from Biogen Idec, USA). After incubation (1.5 h at room

temperature) and between steps, the wells were washed

three times with PBST. Wells were then washed in AP-

buffer [200 mM Tris (pH = 9.8), 10 mM MgCl2] and ternary

complex formation was detected using a chemiluminescent

substrate for AP by incubation for 30 min with 10%

Sapphire Enhancer (Tropix, Roche, DK) and 2% CSPD

(Tropix, Roche, DK) in AP-buffer. Luminescence was

quantified using Microbeta Trilux Counter (Perkin Elmer,

DK). The relative GFRa2 binding activity in samples was

calculated from a standard curve generated by using

recombinant NTN protein. Binding activity to GFRa1 in

conditioned medium was measured similarly, but with 1 Ag/ml GFRa1/Fc fusion protein (R&D Systems, UK) added

instead of GFRa2/Fc.

NTN ELISA

To detect immunoreactive NTN in media samples, a

sandwich ELISAwas used. Briefly, MaxiSorp plates (Nunc,

DK) were coated with 1 Ag/ml monoclonal anti-human

NTN antibody (#MAB387, R&D Systems, UK) in 2.5 mM

Na2CO3/2.5 mM NaHCO3, pH 8.2. Wells were blocked in

blocking buffer [1% bovine serum albumin (Sigma-Aldrich,

DK) in PBS] supplemented with 5% sucrose followed by

incubation with diluted media samples from NTN-produc-

ing cells or NTN standards (recombinant NTN #387-NE

from R&D Systems, UK) for 3 h at room temperature. To

detect bound NTN, wells were incubated with 1 Ag/ml

polyclonal anti-human NTN antibody (#AF387, R&D

Systems, UK) in blocking buffer and subsequently with

0.02% anti-goat-HRP (DAKO, DK) in blocking buffer

supplemented with 1% normal mouse serum (DAKO, DK).

In between each step, wells were washed three times with

PBST [0.05% Tween-20 (Sigma-Aldrich, DK) in PBS]. To

detect bound HRP-activity, 3,3V,5,5V-tetramethylbenzidine

substrate solution (Promega, Ramcon, DK) was added, and

color formation was stopped by addition of 1 N HCl after 15

min. A450 was measured using an ELX-800 plate reader

(Cambrex, DK).

Surgical procedures

A total of 24 young adult female Sprague–Dawley rats

(Møllegarden, DK) was used and housed under 12-h

light:dark cycle with free access to rat chow and water.

Virus injections and 6-hydroxydopamine lesions were

performed according to Rosenblad et al. (2000). Briefly,

isofluorane anesthetized (1.5–2%) animals were injected

with rLV vector (3 � 105 TU/animal) carrying the cDNA for

GFP, NTN, IgSP-NTN, or GDNF (n = 5–7/group). Using a

Hamilton microsyringe with a pulled glass pipette tip, four

deposits (0.75 Al/deposit) were made into the striatum along

two needle tracts at the following coordinates with respect to

bregma: (1) AP = 1.0 mm, ML = �2.6 mm; and (2) AP =

0.0 mm, ML = �3.7 mm, and DV1 = �5.0 mm, DV2 = �4.5

mm with respect to the dura. The tooth bar was set at �2.3

mm. 14 days after rLV injections, the animals were

anesthetized and a single deposit of 20 Ag 6-hydroxydo-

pamine (6-OHDA from Sigma; calculated as free base and

dissolved in 3 Al ice-cold saline supplemented with 0.02%

ascorbic acid) was injected into the right striatum with a 10-

Al Hamilton syringe at the coordinates AP = 0.5 mm; ML =

�3.4 mm relative to bregma, DV = �5.0 mm relative to the

dura and tooth bar set to 0.0 mm. The injection rate was

1 Al/min and the glass pipette was left in place for an

additional 3 min before withdrawal.

Amphetamine-induced rotation

At 10 days after rLV injections and again 4 weeks after

the 6-hydroxydopamine injections, rats were injected with

amphetamine (2.5 mg/kg, Mecobenzon, DK) and moni-

tored for turning response in automated rotometer bowls

over 90 min. Rotational asymmetry scores are expressed as

net full ipsilateral turns per minute (i.e., towards the

injected site).

Histology

Twenty-eight days after the 6-hydroxydopamine injec-

tions, the animals were deeply anesthetized with sodium

pentobarbital and transcardially perfused with saline for 1

min followed by 200 ml ice-cold 4% paraformaldehyde in

0.1 M phosphate buffer (pH 7.4). The brains were

dissected and post-fixed in the same fixative for 3–4 h

and then transferred into 25% sucrose/0.1 M phosphate

buffer for 48 h. Six series of 40-Am sections were cut on

a freezing microtome. Immunohistochemistry was per-

formed as described previously (Rosenblad et al., 1999).

In brief, sections were incubated overnight with goat-anti-

hNTN (R&D Systems, UK; 1:2000), goat-anti-hGDNF

(R&D Systems, UK; 1:2000), chicken anti-GFP (Chem-

icon; 1:2000), mouse anti-tyrosine hydroxylase (TH)

(Chemicon; 1:2000), or rabbit anti-vesicular monoamine

transporter (VMAT) (Chemicon; 1:2000) primary anti-

bodies diluted in phosphate buffered saline containing 2%

normal horse or swine serum and 0.25% Triton X-100

followed by incubation with the appropriate (anti-goat,

anti-chicken, anti-mouse, or anti-rabbit) biotinylated sec-

ondary antibody (all from Jackson Immunoresearch,

Trichem, DK and diluted 1:200) for 2 h, and avidin–

biotin complex (ABC) kit according to the manufacturer’s

instructions (Vector Laboratories, USA). Finally, the color

reaction was developed using 3V3V-diaminobenzidine as

chromogen.


Morphometric analysis: Quantification of neurons

expressing the DA markers TH or VMAT in substantia

nigra (SN)

A person, blinded to the identity of the samples, assessed

the number of immunoreactive neurons in SN pars com-

pacta, as described previously (Sauer and Oertel, 1994). In

brief, three consecutive sections centered at the level of the

medial terminal nucleus of the accessory optic tract (MTN,

�5.3 in the atlas of Paxinos and Watson, 1997) were used

and all stained neurons lateral to the MTN were counted at

40� magnification. Cell numbers are expressed as the mean

T SEM of the percentage of the number in the intact side. For

quantification of the downregulation of TH expression, TH-

immunoreactive cells with normal as well as downregulated

expression were counted in the lesion side (see Fig. 4G for

examples of cells with different TH expression). Cells were

counted at 40� magnification in the same three consecu-

tive sections as described above and expressed as the

mean percentage T SEM of cells with downregulated TH

expression.

Striatal fiber density measurements

Striatal DA innervation was assessed by measuring

optical density (OD) of the striatum at three rostrocaudal

levels in sections stained for TH as described (Kirik et al.,

2001). Using an Olympus DP50 digital camera and a

constant illumination table, digitalized images were col-

lected. ODs on the intact and lesioned sides were measured

using ScanImage v. 4.0.2 software. Corpus callosum in each

section was used as reference for background staining.

Results

Expression of NTN in cultured cells

To characterize the role of the pre-pro-part in the

secretion of active NTN, we generated a series of cDNA

constructs encoding human NTN proteins shown in Fig. 1A.

Western blot analyses of lysates and conditioned medium

from the transfected cells were performed using an antibody

against NTN. Fig. 1B shows results obtained with HEK293

cells. In cells transfected with the wt NTN construct, a band

corresponding to the expected size of monomeric pro-NTN

(¨22 kDa) was detected in cell lysate as well as conditioned

medium under reducing conditions. A band corresponding

to the expected size of the slightly smaller NTN with the

GDNF pro-region (¨19.6 kDa) was seen in cell lysate and

conditioned medium from cells transfected with ppG-NTN.

No bands corresponding to the expected size of mature NTN

was observed in the cells or conditioned medium. Thus, the

pro-forms of NTN are expressed and secreted but not

processed at a detectable level. In cells transfected with

Dpro-NTN or IgSP-NTN, a band corresponding in size to

mature monomeric NTN (¨12.5 kDa) was observed in both

cell lysates and conditioned media. The total level of

secreted mature NTN protein appeared to be higher in cells

transfected with the IgSP-NTN construct than cells trans-

fected with Dpro-NTN. Similar Western blotting results

were obtained with ARPE-19 cells, HiB5 cells, and CHO

cells (data not shown).

We then tested the level of biologically active NTN in

conditioned medium using a functional assay where binding

of NTN to its receptor, GFRa2, is detected by formation of

a ternary complex with the GFRa co-receptor Ret. The

relative activities in this assay compared to the wt NTN

construct are shown in Fig. 1C. Conditioned medium from

cells transfected with wt NTN construct showed low levels

of active NTN (7.5 T 0.6 ng/ml, 1.5 T 0.9 ng/ml, 38.6 T 3.3

ng/ml, and 18.3 T 1.7 ng/ml for wt NTN in HEK293,

ARPE-19, HiB5, and CHO cells, respectively). Equally low

levels were observed when using the ppG-NTN construct.

However, GFRa2 binding activity in samples increased

when using the Dpro-NTN construct (90 T 19, 117 T 13,

7.5 T 1.7, and 4.1 T 0.9 fold higher NTN binding for

HEK293, ARPE-19, HiB5, and CHO cells, respectively). In

accordance with the Western blot results, NTN binding

activity was further enhanced when using the IgSP-NTN

construct (278 T 13, 771 T 50, 162 T 29, and 66 T 18 fold

higher NTN binding for HEK293, ARPE-19, HiB5, and

CHO cells, respectively). Similar results were obtained when

performing the assay with the GFRa1 co-receptor (data not

shown). Cell supernatants from cells transiently transfected

with pNS1n-EGFP showed undetectable NTN activity

confirming the specificity of the assays (data not shown).

In addition, we performed NTN Western blotting on

NTN captured from conditioned media by GFRa2-Ig

immobilized in ELISA plates prior to loading (Fig. 1D).

In contrast to what was seen in Western blot analysis in

which conditioned media were loaded directly (Fig. 1B), we

did not detect pro-forms of NTN derived from cells

transfected with wt NTN and ppG-NTN constructs, whereas

bands corresponding the expected size of mature NTN were

observed in samples from cells transfected with the IgSP-

NTN construct and to a lesser extent when using the Dpro-

NTN vector indicating poor or absent binding of pro-forms

to the GFRa2 receptor.

An NTN sandwich ELISA on conditioned medium from

cells transfected with the different NTN constructs was also

carried out. A monoclonal NTN antibody was used to

capture NTN on ELISA plates and subsequently a poly-

clonal NTN antibody was used to detect the captured NTN.

Both antibodies recognized pro-forms of NTN when used

for Western blotting under reduced conditions, but as shown

in Fig. 1E, the NTN pro-forms were not or very poorly

detected in the NTN ELISA. This finding suggests that the

pro-part of NTN inhibits binding of the antibody to native

NTN, but not to denatured NTN. The NTN sandwich

ELISA confirmed that using the IgSP-NTN construct rather

than the Dpro-NTN construct increases the level of NTN in

Fig. 1. (A) NTN expression constructs encoding wt (pre-pro) NTN, a chimeric NTN with the pre-pro-part derived from GDNF (ppG-NTN), NTN with deleted

pro-region (Dpro-NTN), and a pro-region deleted chimeric NTN (IgSP-NTN). The latter DNA sequence contains an intron. See text for further details. (B)

NTN Western blot of lysates and conditioned medium from transfected HEK293 cells. Arrows indicate the expected position of bands representing wt pro-

NTN, pro(GDNF)-NTN and mature NTN, respectively. Note that the standard is NTN-His which has a slightly higher molecular weight than wt NTN. (C)

GFRa 2 binding activity of NTN in conditioned medium from the four cell lines transfected with the NTN constructs. The GFRa 2 binding activities shown in

samples were calculated using a standard curve of recombinant NTN. Relative values are shown with activities from cells transfected with the wt NTN

construct set to 1. (D) NTN Western blot of NTN from conditioned HEK293 medium from bound to GFRa 2. Arrow indicates the expected position for band

representing mature NTN. (E) NTN sandwich ELISA on conditioned medium from the four cell lines transfected with the NTN constructs. Data in panels (C)

and (E) are expressed as mean T SEM (n = 3) from a representative experiment and * indicates a significant difference from cells transfected with the wt

construct (P < 0.05, Kruskal–Wallis one-way analysis on ranks followed by Student–Newman–Keuls Method).


conditioned medium from the transfected cells (2- to 8-fold

higher NTN levels for IgSP-NTN than Dpro-NTN, depend-

ing on cell line).

In vivo neuroprotection by IgSP-NTN in the rat

6-hydroxydopamine lesion model

Next, we wanted to investigate whether the IgSP-NTN

construct would lead to sustained secretion of active NTN in

vivo. We generated recombinant lentiviral vectors encoding

the wt NTN (rLV-wt NTN) and the IgSP-NTN (rLV-IgSP-

NTN), and tested if they were able to provide neuro-

protection in an animal model of PD where NTN protein

injections previously had been shown to prevent loss of DA

neurons (Horger et al., 1998; Rosenblad et al., 1999).

Vectors encoding green fluorescent protein (GFP; rLV-GFP)

or GDNF (rLV-GDNF) were used as negative and positive

control vectors, respectively.

Inspection of GFP-immunostained sections from rLV-

GFP-treated animals showed a column of transduced cells,


approximately 2 � 0.5 mm, in the central head of the

caudate putamen (Fig. 2A). The majority of transduced

cells in the striatum had morphologies of striatal medium-

sized spiny projection neurons, but transduced cells with

astroglial morphology were also observed to a lesser

extent. Transduced oligodendroglial cells were seen in the

overlying corpus callosum. The GFP-expressing cells

showed a distinct intracellular expression pattern. In

contrast, sections through the striatum and SN from

animals treated with rLV-GDNF and processed for

GDNF-immunohistochemistry showed a diffuse staining

in the striatum, consistent with secretion of GDNF from

transduced cells (Fig. 2C). In animals receiving rLV-wt

NTN injections in striatum, NTN-immunohistochemistry

showed only weak immunoreactivity (Fig. 2B). At higher

magnification, NTN-immunoreactive cells with punctuate

cytoplasmic staining were observed along the injection

tract (Fig. 2E). In contrast, animals receiving injections of

rLV-IgSP-NTN had a prominent diffuse staining in the

striatum (Fig. 2D), similar to that seen in GDNF-treated

Fig. 2. In vivo expression of transgene after intrastriatal injections of lentiviral co

lentiviral vectors and subsequently processed for immunohistochemistry using ant

GFP control group, a distinct intracellular staining pattern was seen after immu

immunoreactivity (arrows), but no extracellular NTN signal is observed in the rLV-

marked specific extracellular staining pattern for NTN is seen in the rLV-IgSP tra

animal shows a similar diffuse staining pattern due to secretion of the protein (C).

NTN is anterogradely transported from the IgSP-NTN transduced cells in the stri

animals. NTN immunoreactive cellular profiles were also

observed in rLV-IgSP-NTN-injected animals, but most of

the immunoreactive material was located extracellularly

(Fig. 2F). In animals receiving rLV-GDNF as well as rLV-

IgSP-NTN, there was prominent labeling with the respec-

tive antibodies in the SN pars reticulata (data not shown

and Fig. 2G). The dense network of immunoreactive fibers

apparently originated from the striatonigral projections, as

they could be followed rostrally to the striatum. This result

suggests that NTN can be anterogradely transported within

the striatonigral pathway, as has already been described for

GDNF (Rosenblad et al., 2003).

The number of remaining DA nigral neurons was

assessed by counting neurons expressing tyrosine hydro-

xylase (TH) at 4 weeks after the intrastriatal 6-hydroxydop-

amine injection (Figs. 3A and 4). In control rLV-GFP-

treated animals, significantly fewer TH-immunoreactive

(IR) neurons were seen in the substantia nigra (SN) on

the lesion side (23 T 3.4%), as compared to the intact

contralateral side (Figs. 3A and 4D). Similarly, animals

nstructs. Coronal sections through the striatum (Str) of rats transduced with

ibodies to GFP (A), human NTN (B, D, E, F, and G), or GDNF (C). In the

nohistochemistry using an antibody against GFP (A). Weak intracellular

NTN transduced striatum (B, E). In contrast, intracellular (arrows) but also a

nsduced striatum (D, F). GDNF immunostaining in rLV-GDNF transduced

NTN-immunoreactive fibers in the SN pars reticulata (SNR), showing that

atum (G). Scale bar, 1 mm (A–D, G) or 62.5 Am (E, F).

Fig. 3. Neuroprotection of nigral DA neurons by rLV-IgSP-NTN. (A) The

number of nigral neurons expressing TH in the lesion side compared to the

intact side in the four different treatment groups. (B) The percentage of

VMAT-immunoreactive nigral neurons in lesion side compared with the

intact side. Data are expressed as mean T SEM (n = 5–7) and * indicates a

significant difference from the GFP group (P < 0.05, one-way ANOVA,

Dunnett’s Method).


receiving rLV-NTN (Figs. 3A and 4B) showed a marked

lesion-induced reduction in TH-IR neurons (30 T 7.7%

remaining). In contrast, the rLV-IgSP-NTN-treated group

had a significantly higher percentage of TH-IR neurons on

the lesion side (91 T 1.2%; P < 0.01) (Figs. 3A and 4C),

which was indistinguishable from that observed in the

group receiving GDNF treatment (86 T 3.2%; P < 0.01)

(Figs. 3A and 4H).

To confirm that the differences in the number of TH-IR

nigral neurons were due to neuroprotection rather than

regulation of TH enzyme, we quantified in adjacent sections

the number of neurons expressing another DA marker,

vesicular monoamine transporter (VMAT), since VMAT has

been shown to be less prone to regulation by neurotrophic

factors (Georgievska et al., 2002a; Kirik et al., 1998;

Rosenblad et al., 2003). As shown in Fig. 3B, transduction

with rLV-IgSP-NTN or rLV-GDNF significantly preserved

the number of VMAT-IR neurons in SN on the lesion side

(75.2 T 6.8% and 59.9 T 4.2% of that in the intact side,

respectively) compared with rLV-GFP- or rLV-NTN-treated

animals (15.5 T 2.4% and 24.9 T 6.4%, respectively),

corroborating the results seen with TH-staining.

In addition to the protection of TH-IR and VMAT-IR

neurons in the SN, it was noticeable that in specimens from

rLV-IgSP-NTN-treated animals, the TH staining intensity

was reduced in many remaining SN neurons (Fig. 4G) as

compared to TH-IR neurons on the intact side (Fig. 4E), or

lesioned side of rLV-GFP- or rLV-wt NTN-treated animals

(Fig. 4F). Reduced TH staining intensity was also observed

following treatment with GDNF, consistent with previous

reports (Georgievska et al., 2002a). In the rLV-GDNF and

rLV-IgSP-NTN groups, the percentage of cells with down-

regulated TH expression was significantly higher (79.4 T 3.5and 74.5 T 3.8, respectively) than in the rLV-NTN (16.5 T8.7) and rLV-GFP (13.9 T 7.3) groups (P < 0.001, one-way

ANOVA followed by all pairwise multiple comparison

procedures, Student–Newman–Keuls Method).

Inspection of sections through the striatum processed for

TH-immunohistochemistry showed that in all groups, the

central and lateral caudate putamen was devoid of TH-IR

fibers on the side injected with 6-hydroxydopamine.

Densitometric quantification of TH-IR innervation on the

lesioned side showed that 15–25% remained at 4 weeks

(Fig. 5). This is in agreement with earlier studies in the

intrastriatal 6-hydroxydopamine lesion model (Georgievska

et al., 2002a; Kirik et al., 2001; Rosenblad et al., 2003)

showing that recovery of TH-IR striatal innervation usually

takes longer than 4 weeks post-lesion to develop. The

reduction of TH-IR striatal innervation did not differ

between the groups. Consistently, amphetamine-induced

rotation, which can be used as a sensitive measure of

dopamine denervation in the striatum (Kirik et al., 2000),

showed no significant difference in the number of ipsilateral

turns between any of the treatment groups at 4 weeks post-

lesion (ranging from 4.5 T 1.5 to 13.4 T 3.3 net ipsilateral

turns/min; P > 0.05 two-way repeated measures ANOVA).

Amphetamine-induced turning assessed at 10 days after

viral transduction, but before the 6-hydroxydopamine

lesion, showed a slight but non-significant contralateral

turning bias in the IgSP-NTN (3.3 T 1.5) and GDNF groups

(1.9 T 1.3), compared to the rLV-GFP or rLV-NTN groups

(0.1 T 0.9 and �0.1 T 1.6, respectively), consistent with an

upregulation of the DA function on the IgSP-NTN

transduced side similar to what has been reported previ-

ously to occur after GDNF treatment (Georgievska et al.,

2002a,b).

Discussion

We investigated in vivo gene therapy to deliver NTN in a

PD animal model but found that human wt NTN, unlike its

GDNF counterpart, is inadequately expressed with this

approach. We therefore examined the role of the pre-pro-

region of human NTN in the secretion of active NTN of de

Fig. 4. Coronal sections through SN of GFP-, NTN-, IgSP-NTN-, and GDNF-treated animals. On the intact side, many TH-immunoreactive cells can be seen

(A). On the side subject to a 6-hydroxydopamine lesion, the number of remaining TH-immunoreactive neuronal profiles is significantly reduced in wt NTN- or

GFP-treated animals (B, D). By contrast, more TH-immunoreactive neurons remain in animals receiving GDNF (E) or IgSP-NTN treatment (C). Note the

reduced TH staining intensity in GDNF- and IgSP-NTN-treated animals. Higher magnification of TH-IR cells in intact side (E), lesion side of IgSP-NTN-

treated animal (G), and lesion side of wt NTN-treated animal (F). Black arrows point at neurons with downregulated TH expression in IgSP-NTN-treated

animal and white arrow points at neuron with normal TH expression. Scale bar, 1 mm (A–D, H) or 25 Am (E–G).


novo synthesized NTN in mammalian cells by transfecting

constructs encoding modified NTN proteins into two human

and two rodent cell lines. Furthermore, the in vivo neuro-

protective effect of modified NTN protein was addressed in

the partial 6-hydroxydopamine lesion model using a gene

therapy approach.

As biologically active human GDNF is readily secreted

from various cell types after transfection, an NTN construct

encoding a chimeric protein with the pre-pro-part derived

from GDNF was generated (ppG-NTN). In addition, as the

processing of the pro-forms seemed impaired, two addi-

tional constructs encoding NTN proteins with the pro-part

deleted but with different signal peptides were made, one

containing the wt NTN signal peptide (Dpro-NTN) and one

with the IgSP (IgSP-NTN). We could not detect any

processed NTN in cell lysates or conditioned medium from

cells transfected with wt NTN or ppG-NTN constructs by

Western blot analysis. These results indicate that intra-

cellular cleavage of the pro-region is severely compromised

in the tested cell lines. Furthermore, we found that the

Fig. 5. Striatal DA fiber density measurements. The four experimental

groups had a comparable initial lesion size, and there was no significant

recovery in any group after 4 weeks. Striatal DA innervation was assessed

by measuring optical density of the striatum at three rostrocaudal levels in

sections stained for TH.


cleavage of the pro-peptide is critical for the ability of

NTN to form a signaling complex, demonstrated by the very

low binding activity to the GFRa1 or GFRa2 receptors with

the same constructs (Figs. 1B–D). The very low (but

detectable) level of binding is most likely explained by the

presence of mature NTN below the Western blot detection

limit (<1.25 ng/ml). This is also supported by the lack

of capture of the pro-forms of NTN on the GFRa2 plate

(Fig. 1D).

In contrast, a protein of the expected size of mature

NTN was detected in Western blot analysis in media

collected from cells transfected with either the Dpro-NTN

and IgSP-NTN constructs. Furthermore, high levels of

immunoreactive NTN and GFRa1 and GFRa2 binding

activities were also detected in these media. These results

indicate that the pro-sequence of NTN is not necessary for

correct protein folding in contrast to other neurotrophic

factors such as NGF, TGFh, and Activin in which the pro-

region has been shown to aid the folding and disulphide

bond formation (Gray and Mason, 1990; Suter et al.,

1991). In the case of NGF, deletion of the pro-region

results in undetectable expression, possibly due to insta-

bility and degradation of the protein in the endoplasmatic

reticulum (Suter et al., 1991). Unlike NGF, the pro-region

of NTN appears to be dispensable for intracellular stability

in our experiments.

Data obtained from the NTN sandwich ELISA indicated

significantly higher production levels (2- to 8-fold depend-

ing on cell line) of mature NTN in cells transfected with

the IgSP-NTN construct compared with Dpro-NTN (Fig.

1E) consistent with what was observed from the Western

blot analyses (Figs. 1A and D). The reason for this

increased production is unclear. Hypothetically, the IgSP

sequence could influence the stability of mRNA due to the

presence of the inherent IgSP intron. Insertion of a

heterologous intron has previously been shown to improve

expression (Palmiter et al., 1991). Other possible explan-

ations include the promotion of 3Vend formation, increased

mRNA stability in the nucleus, or improved mRNA export

to the cytoplasm (Nesic et al., 1993; Ryu and Mertz,

1989).

In this report, we provide the first demonstration of in

vivo gene delivery of NTN as a therapeutic option for PD.

In accordance with the results obtained with the cultured

cells, we observed only weak NTN immunoreactivity and

no significant neuroprotective effect in vivo when injecting

rats with lentivirus encoding human wt NTN. These results

indicate that transduction with wt NTN in rat striatum most

likely leads to expression of pro-NTN, which are poorly

recognized by the NTN antibodies as not denatured by the

fixation of tissues. These data are consistent with the in

vitro results that indicate that the native pro-NTN could not

be detected by ELISA but could be detected in a denaturing

Western blot. In contrast, a potent neuroprotective effect

was seen and a diffuse staining was observed by NTN-

immunohistochemistry after in vivo transduction with the

IgSP-NTN construct consistent with the efficient secretion

of bioactive mature NTN from the transduced cells. The

potency and efficacy of NTN in the current study is in

accordance with results obtained by Horger et al. (1998)

who showed that the effect of recombinant NTN on

survival of DA neurons in the 6-hydroxydopamine lesion

model was similar to that of GDNF when administered at

the level of SN.

The primary focus of the current study was to investigate

whether the IgSP-NTN construct would produce a molecule

with biological activity in vivo. We observed comparable

effects of our modified NTN construct and GDNF on DA

cell survival and regulation of TH expression after 4 weeks.

However, longer observations times need to be added in

order to conclude whether the NTN produced by our

modified construct is as efficient as GDNF for recovery of

striatal DA fiber innervation. Furthermore, GDNF and NTN

may show different therapeutic profiles, due to the fact that

the preferred GFRa receptors for NTN and GDNF are

differentially expressed (Akerud et al., 1999; Jing et al.,

1997; Widenfalk et al., 1997; Yu et al., 1998). For example,

in another study, GDNF delivered intracerebroventricularly

via osmotic minipumps in rats resulted in side effects,

mainly weight-loss and allodynia, whereas NTN treatment

did not lead to these symptoms (Hoane et al., 1999).

Moreover, long-term overexpression of GDNF in rat

striatum has been shown to induce aberrant sprouting of

nigrostriatal fibers (Georgievska et al., 2002b). The con-

sequence of this phenomenon is unclear, but it might be

detrimental for recovery in spontaneous motor behavior.

Interestingly, Akerud et al. (1999) showed that NTN, in

contrast to GDNF, did not induce sprouting and hyper-

trophy of adult DA neurons, though both GDNF and NTN

showed survival promoting effects. In another rat model of

PD made by medial forebrain bundle axotomy, an initial

compensatory sprouting effect induced by GDNF but not


by NTN has been shown (Tseng et al., 1997, 1998).

Experiments to evaluate possible differences in regeneration

and behavior between NTN delivered by our construct and

GDNF are under consideration in our lab. In the described

experiments, we have used human NTN in a rodent setting.

The reason for this is that we have focused on developing

gene therapeutic application for the treatment of Parkin-

son’s disease patients and need to optimize the expression

of human NTN cDNA. Human, mouse, and rat NTN have

identical consensus recognition sites for furin-like proteases

(RRAR). In contrast, the amino acids after the cleavage site

differ between human (ARLGAR) and rat/mouse (P—GSR).

It could be speculated that these amino acids would affect the

processing of NTN. However, as we found that rodent NTN

cDNAwas also expressed as the pro-form in human as well as

rat cells in vitro (data not shown), we find it plausible that we

would have obtained a similar result by using rodent wt NTN

in vivo.

It is possible that the processing of NTN is part of an

important regulatory step for the availability of active

protein, as pro-NTN is not biologically active in the same

way as mature NTN is. The transfected cell lines used in

vitro and the transduced cells in vivo do not normally

express NTN and might not posses the processing apparatus

for NTN. Alternatively, it could be speculated that addi-

tional extracellular signals are necessary for the activation of

relevant proteases. Such signals might be present when

active NTN is needed, e.g., during development.

In summary, we show that deletion of the pro-region of

human NTN increases the fraction of biologically active

NTN. Replacing the signal peptide sequence from wt NTN

in Dpro-NTN with the heterologous IgSP sequence further

increases secretion of active NTN. The biological activity

of neurotrophic factors can be regulated at several differ-

ent levels. Our results indicate that proteolytic processing

of the NTN pro-region is an important regulatory step,

which should be taken into account when overexpressing

neurotrophic factors from transgenic constructs. We

obtained a high and stable secretion level of mature

bioactive NTN when using the IgSP-NTN construct. The

results from studies in cultured cells were reproducible in

vivo, where intrastriatal lentiviral delivery of IgSP-NTN

resulted in protection of adult DA neurons at a level

comparable to GDNF, while wt NTN had no effect. These

results indicate that in vivo gene therapy is possible with a

modified NTN gene transfer and is a potential therapy for

PD.

Acknowledgments

The authors would like to thank Hanne Fosmark and

Lise Moberg Fitting for excellent technical assistance. This

study was supported by the Fifth Framework EU Program

Quality of Life and Management of Living Resources,

QLG3-CT-2002-01000.

References

Akerud, P., Alberch, J., Eketjall, S., Wagner, J., Arenas, E., 1999.

Differential effects of glial cell line-derived neurotrophic factor and

neurturin on developing and adult substantia nigra dopaminergic

neurons. J. Neurochem. 73, 70–78.

Baloh, R.H., Enomoto, H., Johnson, Jr., E.M., Milbrandt, J., 2000. The

GDNF family ligands and receptors—Implications for neural develop-

ment. Curr. Opin. Neurobiol. 10, 103–110.

Bensadoun, J.C., Deglon, N., Tseng, J.L., Ridet, J.L., Zurn, A.D.,

Aebischer, P., 2000. Lentiviral vectors as a gene delivery system in

the mouse midbrain: cellular and behavioral improvements in a

6-OHDA model of Parkinson’s disease using GDNF. Exp. Neurol.

164, 15–24.

Bilang-bleuel, A., Revah, F., Colin, P., Locquet, I., Robert, J.J., Mallet,

J., Horellou, P., 1997. Intrastriatal injection of an adenoviral vector

expressing glial-cell-line-derived neurotrophic factor prevents dop-

aminergic neuron degeneration and behavioral impairment in a rat

model of Parkinson disease. Proc. Natl. Acad. Sci. U. S. A. 94,

8818–8823.

Cacalano, G., Farinas, I., Wang, L.C., Hagler, K., Forgie, A., Moore, M.,

Armanini, M., Phillips, H., Ryan, A.M., Reichardt, L.F., Hynes, M.,

Davies, A., Rosenthal, A., 1998. GFRalpha1 is an essential receptor

component for GDNF in the developing nervous system and kidney.

Neuron 21, 53–62.

Connor, B., Kozlowski, D.A., Unnerstall, J.R., Elsworth, J.D., Tillerson,

J.L., Schallert, T., Bohn, M.C., 2001. Glial cell line-derived neuro-

trophic factor (GDNF) gene delivery protects dopaminergic terminals

from degeneration. Exp. Neurol. 169, 83–95.

Dunn, K.C., Aotaki-Keen, A.E., Putkey, F.R., Hjelmeland, L.M., 1996.

ARPE-19, a human retinal pigment epithelial cell line with differ-

entiated properties. Exp. Eye Res. 62, 155–169.

Georgievska, B., Kirik, D., Bjorklund, A., 2002a. Aberrant sprouting and

downregulation of tyrosine hydroxylase in lesioned nigrostriatal

dopamine neurons induced by long-lasting overexpression of glial cell

line derived neurotrophic factor in the striatum by lentiviral gene

transfer. Exp. Neurol. 177, 461–474.

Georgievska, B., Kirik, D., Rosenblad, C., Lundberg, C., Bjorklund, A.,

2002b. Neuroprotection in the rat Parkinson model by intrastriatal

GDNF gene transfer using a lentiviral vector. NeuroReport 13, 75–82.

Georgievska, B., Kirik, D., Bjorklund, A., 2004. Overexpression of

glial cell line-derived neurotrophic factor using a lentiviral vector

induces time- and dose-dependent downregulation of tyrosine

hydroxylase in the intact nigrostriatal dopamine system. J. Neurosci.

24, 6437–6445.

Golden, J.P., Baloh, R.H., Kotzbauer, P.T., Lampe, P.A., Osborne, P.A.,

Milbrandt, J., Johnson, Jr., E.M., 1998. Expression of neurturin, GDNF,

and their receptors in the adult mouse CNS. J. Comp. Neurol. 398,

139–150.

Gray, A.M., Mason, A.J., 1990. Requirement for activin A and trans-

forming growth factor-beta 1 pro-regions in homodimer assembly.

Science 247, 1328–1330.

Hoane, M.R., Gulwadi, A.G., Morrison, S., Hovanesian, G., Lindner, M.D.,

Tao, W., 1999. Differential in vivo effects of neurturin and glial cell-

line-derived neurotrophic factor. Exp. Neurol. 160, 235–243.

Horger, B.A., Nishimura, M.C., Armanini, M.P., Wang, L.C., Poulsen,

K.T., Rosenblad, C., Kirik, D., Moffat, B., Simmons, L., Johnson, Jr.,

E., Milbrandt, J., Rosenthal, A., Bjorklund, A., Vandlen, R.A., Hynes,

M.A., Phillips, H.S., 1998. Neurturin exerts potent actions on survival

and function of midbrain dopaminergic neurons. J. Neurosci. 18,

4929–4937.

Jing, S., Yu, Y., Fang, M., Hu, Z., Holst, P.L., Boone, T., Delaney, J.,

Schultz, H., Zhou, R., Fox, G.M., 1997. GFRalpha-2 and GFRalpha-3

are two new receptors for ligands of the GDNF family. J. Biol. Chem.

272, 33111–33117.

Kirik, D., Rosenblad, C., Bjorklund, A., 1998. Characterization of

behavioral and neurodegenerative changes following partial lesions of


the nigrostriatal dopamine system induced by intrastriatal 6-hydro-

xydopamine in the rat. Exp. Neurol. 152, 259–277.

Kirik, D., Rosenblad, C., Bjorklund, A., Mandel, R.J., 2000. Long-term

rAAV-mediated gene transfer of GDNF in the rat Parkinson’s model:

intrastriatal but not intranigral transduction promotes functional

regeneration in the lesioned nigrostriatal system. J. Neurosci. 20,

4686–4700.

Kirik, D., Georgievska, B., Rosenblad, C., Bjorklund, A., 2001. Delayed

infusion of GDNF promotes recovery of motor function in the

partial lesion model of Parkinson’s disease. Eur. J. Neurosci. 13,

1589–1599.

Klein, R.D., Sherman, D., Ho, W.H., Stone, D., Bennett, G.L., Moffat, B.,

Vandlen, R., Simmons, L., Gu, Q., Hongo, J.A., Devaux, B., Poulsen,

K., Armanini, M., Nozaki, C., Asai, N., Goddard, A., Phillips, H.,

Henderson, C.E., Takahashi, M., Rosenthal, A., 1997. A GPI-linked

protein that interacts with Ret to form a candidate neurturin receptor

[published erratum appears in Nature 1998 Mar 12;392(6672):210].

Nature 387, 717–721.

Kordower, J.H., 2003. In vivo gene delivery of glial cell line-derived

neurotrophic factor for Parkinson’s disease. Ann. Neurol. 53 (Suppl 3),

S120–S132.

Lapchak, P.A., Araujo, D.M., Hilt, D.C., Sheng, J., Jiao, S., 1997.

Adenoviral vector-mediated GDNF gene therapy in a rodent lesion

model of late stage Parkinson’s disease. Brain Res. 777, 153–160.

Mandel, R.J., Spratt, S.K., Snyder, R.O., Leff, S.E., 1997. Midbrain

injection of recombinant adeno-associated virus encoding rat glial cell

line-derived neurotrophic factor protects nigral neurons in a progressive

6-hydroxydopamine-induced degeneration model of Parkinson’s disease

in rats. Proc. Natl. Acad. Sci. U. S. A. 94, 14083–14088.

Nesic, D., Cheng, J., Maquat, L.E., 1993. Sequences within the last intron

function in RNA 3V-end formation in cultured cells. Mol. Cell. Biol. 13,

3359–3369.

Palmiter, R.D., Sandgren, E.P., Avarbock, M.R., Allen, D.D., Brinster, R.L.,

1991. Heterologous introns can enhance expression of transgenes in

mice. Proc. Natl. Acad. Sci. U. S. A. 88, 478–482.

Paxinos, G., Watson, C., 1997. The Rat Brain in Stereotaxic Coordinates.

Academic Press, San Diego.

Renfranz, P.J., Cunningham, M.G., Mckay, R.D., 1991. Region-specific

differentiation of the hippocampal stem cell line HiB5 upon implanta-

tion into the developing mammalian brain. Cell 66, 713–729.

Rosenblad, C., Kirik, D., Devaux, B., Moffat, B., Phillips, H.S., Bjorklund,

A., 1999. Protection and regeneration of nigral dopaminergic neurons

by neurturin or GDNF in a partial lesion model of Parkinson’s disease

after administration into the striatum or the lateral ventricle. Eur. J.

Neurosci. 11, 1554–1566.

Rosenblad, C., Gronborg, M., Hansen, C., Blom, N., Meyer, M., Johansen,

J., Dago, L., Kirik, D., Patel, U.A., Lundberg, C., Trono, D., Bjorklund,

A., Johansen, T.E., 2000. In vivo protection of nigral dopamine neurons

by lentiviral gene transfer of the novel GDNF-family member

neublastin/artemin. Mol. Cell. Neurosci. 15, 199–214.

Rosenblad, C., Georgievska, B., Kirik, D., 2003. Long-term striatal

overexpression of GDNF selectively downregulates tyrosine hydro-

xylase in the intact nigrostriatal dopamine system. Eur. J. Neurosci. 17,

260–270.

Ryu, W.S., Mertz, J.E., 1989. Simian virus 40 late transcripts lacking

excisable intervening sequences are defective in both stability in the

nucleus and transport to the cytoplasm. J. Virol. 63, 4386–4394.

Sanicola, M., Hession, C., Worley, D., Carmillo, P., Ehrenfels, C., Walus, L.,

Robinson, S., Jaworski, G., Wei, H., Tizard, R., Whitty, A., Pepinsky,

R.B., Cate, R.L., 1997. Glial cell line-derived neurotrophic factor-

dependent RET activation can be mediated by two different cell-surface

accessory proteins. Proc. Natl. Acad. Sci. U. S. A. 94, 6238–6243.

Sauer, H., Oertel, W.H., 1994. Progressive degeneration of nigrostriatal

dopamine neurons following intrastriatal terminal lesions with

6-hydroxydopamine: a combined retrograde tracing and immunocyto-

chemical study in the rat. Neuroscience 59, 401–415.

Suter, U., Heymach, Jr., J.V., Shooter, E.M., 1991. Two conserved domains

in the NGF propeptide are necessary and sufficient for the biosynthesis

of correctly processed and biologically active NGF. EMBO J. 10,

2395–2400.

Tseng, J.L., Baetge, E.E., Zurn, A.D., Aebischer, P., 1997. GDNF reduces

drug-induced rotational behavior after medial forebrain bundle tran-

section by a mechanism not involving striatal dopamine. J. Neurosci.

17, 325–333.

Tseng, J.L., Bruhn, S.L., Zurn, A.D., Aebischer, P., 1998. Neurturin protects

dopaminergic neurons following medial forebrain bundle axotomy.

NeuroReport 9, 1817–1822.

Widenfalk, J., Nosrat, C., Tomac, A., Westphal, H., Hoffer, B., Olson, L.,

1997. Neurturin and glial cell line-derived neurotrophic factor receptor-

beta (GDNFR-beta), novel proteins related to GDNF and GDNFR-alpha

with specific cellular patterns of expression suggesting roles in the

developing and adult nervous system and in peripheral organs.

J. Neurosci. 17, 8506–8519.

Yu, T., Scully, S., Yu, Y., Fox, G.M., Jing, S., Zhou, R., 1998. Expression of

GDNF family receptor components during development: implications in

the mechanisms of interaction. J. Neurosci. 18, 4684–4696.

Zufferey, R., Nagy, D., Mandel, R.J., Naldini, L., Trono, D., 1997. Multiply

attenuated lentiviral vector achieves efficient gene delivery in vivo. Nat.

Biotechnol. 15, 871–875.

Zufferey, R., Dull, T., Mandel, R.J., Bukovsky, A., Quiroz, D., Naldini, L.,

Trono, D., 1998. Self-inactivating lentivirus vector for safe and efficient

in vivo gene delivery. J. Virol. 72, 9873–9880.

Zufferey, R., Donello, J.E., Trono, D., Hope, T.J., 1999. Woodchuck

hepatitis virus posttranscriptional regulatory element enhances expres-

sion of transgenes delivered by retroviral vectors. J. Virol. 73,

2886–2892.

Efficient in vivo protection of nigral dopaminergic neurons by lentiviral gene transfer of a...

Documents

Transcript of Efficient in vivo protection of nigral dopaminergic neurons by lentiviral gene transfer of a...