R., Ashworth, S., Clifford, K.M., Jones, W.F., Byars, L.G., Young,. J., Andreaco, M., Williams, C.W., Lammertsma, A.A. and Jones, T.,. The design and physical ...
Neuroscience Letters 233 (1997) 73–76
Growth/differentiation factor 5 protects nigrostriatal dopaminergic neurones in a rat model of Parkinson’s disease Aideen M. Sullivan a, Jolanta Opacka-Juffry a, Gertrud Ho¨tten b, Jens Pohl b, Stavia B. Blunt a ,* a
MRC Cyclotron Unit, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK b Biopharm GmbH, Czernyring 22, 69115 Heidelberg, Germany
Received 10 June 1997; received in revised form 11 August 1997; accepted 11 August 1997
Abstract Growth/differentiation factor 5 (GDF5), a novel member of the transforming growth factor b superfamily, promotes the survival of dopaminergic neurones in vitro. We present here the first evidence for a neuroprotective action of GDF5 in vivo. We investigated the effects of intracerebral administration of GDF5 on a rat model of Parkinson’s disease. GDF5 was administered just above the substantia nigra and into the lateral ventricle immediately before ipsilateral injection of 6-hydroxydopamine into the medial forebrain bundle. GDF5 prevented the development of amphetamine-induced rotations and preserved the integrity of striatal dopaminergic nerve terminals, as measured by positron emission tomography. Post-mortem studies showed that GDF5 spared dopamine levels in the striatum and tyrosine hydroxylase positive neurones in the midbrain. This study suggests that GDF5 has potential for the treatment of Parkinson’s disease. 1997 Elsevier Science Ireland Ltd. Keywords: Growth/differentiation factor 5; Neuroprotection; Nigrostriatal lesion; 6-Hydroxydopamine; Dopamine; Parkinson’s disease; Positron emission tomography
Parkinson’s disease results from a progressive loss of dopaminergic neurones which project from the substantia nigra (SN) to the caudate-putamen. The possibility that this neuronal loss might be prevented by neurotrophic factors has attracted much interest. Members of the transforming growth factor b (TGFb) protein superfamily can promote the survival of dopaminergic neurones in vitro [9,14]. To date, glial cell line-derived neurotrophic factor (GDNF) is the only neurotrophin exhibiting neuroprotective effects on the adult rodent nigrostriatal system in vivo [1,4,17]. Growth/differentiation factor 5 (GDF5) is a novel member of the TGFb superfamily [5,16]. It is expressed in various regions of the newborn rat brain, including the midbrain, suggesting that it may play a role in the development of dopaminergic neurones [8]. Recombinant human GDF5 promotes the survival of mesencephalic dopaminergic neurones in vitro and protects them against the toxin Nmethylpyridinium [8]. The mechanism of action of GDF5
* Corresponding author. Tel.: +44 181 3833726; fax: +44 181 3832029.
has not been investigated, but it is known to promote the survival of astrocytes in vitro [8]. We have examined the effects of GDF5 on the adult nigrostriatal system in vivo, using a 6-hydroxydopamine (6-OHDA)-induced lesion of the rat medial forebrain bundle (MFB). This lesion results in the destruction of the nigrostriatal pathway and is a robust model of Parkinson’s disease. As the mechanism by which GDF5 exerts its neuroprotective effects on dopaminergic neurones is unclear, we injected it into both the SN and striatal region, to cover the possibilities that GDF5 may act on either cell bodies or axon terminals. In order to prevent needle-induced damage to striatal tissue which could affect the positron emission tomography (PET) data, GDF5 was injected into a region of the lateral ventricle directly adjacent to the striatum. Recombinant human GDF5 was expressed in Escherichia coli HMS 174 (DE3) using T7 RNA polymerase. Isolated inclusion bodies were purified using a sucrose gradient and reversed phase high performance liquid chromatography, and stored lyophilised at −80°C. Monomeric GDF5 was refolded to the native dimeric form using a glutathione
0304-3940/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940 (97 )0 0623- X
74
A.M. Sullivan et al. / Neuroscience Letters 233 (1997) 73–76
redox system, and separated from residual monomeric GDF5 by reversed phase chromatography. The amount of purified GDF5 protein was determined using Coomassie protein assay reagent. Purity of the protein was judged by silver-stained sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) to be more than 95%. The biological activity of GDF5 dimer was assessed by application to ROB-C26 cells [19] in vitro for 48 h. Alkaline phosphatase activity was determined according to Partridge et al. [12]. Rats (260–280 g) were anaesthetised using isoflurane with O2/N2O and placed in a stereotaxic frame. GDF5 was used at a final concentration of 10 mg/3 ml in 10 mM citrate, 150 mM sodium chloride. Rats received injections of either dimeric (active) or monomeric (inactive) GDF5 into two separate sites; just above the left SN and into the left lateral ventricle (50 mg/site, n = 5 rats/group). This was followed immediately by an injection of 6-OHDA hydrobromide (8 mg as the free base in 4 ml 0.9% saline with 0.1% ascorbic acid) into the left MFB. Five additional rats received 6OHDA only. Stereotaxic coordinates (from bregma and dura; incisor bar at 5.0 mm above the interaural line [13]) were as follows: AP −3.0, LV +2.5, DV −8.5 for the SN; AP +1.0, LV +1.2, DV −3.5 for the lateral ventricle; AP −2.2, LV +1.5, DV −7.9 for the MFB. Three rats from each treatment group and three control rats were tested behaviourally at 7 days after surgery. Ipsilateral rotations were counted over a 60 min period beginning 5 min after (+)-amphetamine sulphate administration (5 mg/kg i.p.). All rats displayed the typical features of amphetamine challenge, such as stereotypy and piloerection. 6-OHDA only rats and GDF5 monomer rats showed rotation rates of 14.3 ± 1.5 (mean ± SD) and 12.0 ± 4.0 per min, respectively, indicating at least 95% depletion of the nigrostriatal pathway [18]. In contrast, GDF5 dimer rats did not rotate at all, showing that GDF5 prevented 6-OHDA-
Fig. 1. BP values (mean ± SD) for right and left striata of rats from each treatment group, as measured in PET scans using [11C]RTI-121, at 7 days after surgery. The solid line and dotted lines represent the mean ± SD of the striatal BP value of rats pretreated with a blocking dose of stable RTI121, which is a measure of the ‘background’ binding. Open squares denote right striata and filled circles left striata. N = 2 for each group.
Fig. 2. Non-parametric PET images of rat brains from the GDF5 monomer and GDF5 dimer treatment groups, obtained at 35–40 min after [11C]RTI121 i.v. injection. The upper images represent horizontal planes through the brain at the level of the striata. The lower images (A) represent coronal planes obtained at the level indicated by the dotted lines in the horizontal images. The colour bar is an arbitrary 8-bit scale. The hot (white) areas correspond to the right (R) and left (L) striata. In the GDF5 monomertreated brain, there is a unilateral signal in the right (intact) striatum and a total loss of [11C]RTI-121 binding sites in the left (lesioned) striatum. In the GDF5 dimer-treated brain, the signal is symmetrical (typical of normal brain), suggesting good preservation of DA reuptake sites in the left striatum.
induced depletion of dopamine (DA) in the ipsilateral striatum. Control rats did not rotate. At 7 days after surgery, two rats from each treatment group and two control rats underwent PET scans, using a dedicated small animal scanner [3] and the DA transporter tracer, [11C]RTI-121 [7]. The methodology has been described previously [11], with the exception that in the present study, data were acquired in three-dimensional mode. In this mode, the scanner has a resolution of 2.4 ± 0.1 mm (full width at half maximum) at the centre of the transaxial field of view [2]. Images were manipulated using ‘Analyze’ software [15]. A region of interest (ROI) template was positioned over horizontal projections of the brain, and timeradioactivity curves were generated for ROIs corresponding to right striatum, left striatum and cerebellum. The individual cerebellum ROI curves were used as an input function in a reference-tissue compartmental model [6]. In order to measure background binding for [11C]RTI-121, two additional control rats were scanned after pretreatment with a saturating dose of stable RTI-121 (2 mg/kg i.v., 5 min before beginning of scan). The measurement of specific ligand binding is ‘binding potential’ (BP), which is defined as the ratio k3:k4, where k3 and k4 are the rate constants for transfer to and from the specifically bound striatal compartment, respectively. For an infinitely low concentration of stable ligand, BP is a measure of Bmax/Kd. In control rats, the BP value for [11C]RTI-121 in individual striata was 0.535 ± 0.007 (mean ± SD). The background BP value for this tracer, measured in rats pretreated with stable RTI-121, was 0.265 ± 0.020 (Fig. 1).
A.M. Sullivan et al. / Neuroscience Letters 233 (1997) 73–76
In 6-OHDA only and GDF5 monomer rats, the BP values of the left striata (0.240 ± 0.014 and 0.210 ± 0.049, respectively) were very similar to the background binding value, indicating that the lesion had induced a total loss of DA uptake sites within the ipsilateral striatum. The BP values of the right striata of 6-OHDA only and GDF5 monomer rats (0.545 ± 0.020 and 0.485 ± 0.030, respectively) were not markedly different from those of controls. In GDF5 dimer rats, BP values of both left (0.510 ± 0.070) and right (0.565 ± 0.060) striata were maintained at levels close to those of the intact controls. This shows that GDF5 dimer prevented 6-OHDA-induced reductions in striatal DA reuptake capacity, suggesting preservation of DA nerve terminals (shown in PET images in Fig. 2). At 10 days after surgery, all rats were killed by guillotine under light anaesthesia using isoflurane with O2/N2O. Individual striata were rapidly dissected out, frozen in liquid nitrogen and homogenised in 0.5 ml of ice-cold 50 mM trichloroacetic acid, 0.15% sodium metabisulphate and 0.5 mM disodium ethylenediaminetetraacetic acid, containing 0.5 pmol/ml 3,4-dihydroxybenzylamine hydrobromide as an internal standard. Filtered supernatants were directly analysed for DA, 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) content using liquid chromatography with high sensitivity electrochemical detection [10]. There was complete symmetry of DA, DOPAC and HVA levels between the left and right striata of control rats. In both 6-OHDA only and GDF5 monomer rats, there was a large depletion of DA in the left striata (Fig. 3A). The metabolites DOPAC and HVA (data not shown) were also significantly reduced in the left striata of both of these groups of rats (P , 0.001; ANOVA with post-hoc Newman–Keuls test). In contrast, in GDF5 dimer rats, DA levels in the left striata were largely spared (Fig. 3A). There were no significant differences in the levels of DOPAC and HVA between the left and right striata of GDF5 dimer rats. In the left striata of 6-OHDA only and GDF5 monomer rats, the DOPAC/DA ratio, which is used as an index of DA metabolism, was significantly elevated (P , 0.05; ANOVA with post-hoc Newman–Keuls test) in comparison with that of control rats (Fig. 3B). This shows that there was a compensatory increase in DA turnover in the remaining dopaminergic terminals in the lesioned striata. The DOPAC/DA ratio in the left striata of GDF5 dimer rats was not significantly elevated, which confirms that the damage caused by 6-OHDA to the nigrostriatal dopaminergic system was effectively limited by GDF5 treatment. Midbrains from all rats were fixed by immersion in 4% paraformaldehyde in 10 mM phosphate-buffered saline (PBS), cryoprotected in 30% sucrose in PBS, frozen, and cut into 30 mm coronal cryosections. Serial sections through the SN pars compacta (SNpc) and ventral tegmental area (VTA) were stained with rabbit antiserum to tyrosine hydroxylase (TH) (Chemicon). Staining was visualised using biotinylated anti-rabbit IgG and Vectastain Elite ABC reagent
75
Fig. 3. (A) DA levels and (B) DOPAC/DA ratio (mean ± SEM) in right and left striata of rats from each treatment group, at 10 days after surgery. Open bars denote right striata and hatched bars left striata. N = 4 for control and 6-OHDA only groups, n = 5 for GDF5 monomer and dimer groups. @P , 0.001 vs. ipsilateral side, 6-OHDA only and GDF5 monomer groups; *P , 0.05; **P , 0.001 vs. contralateral side, same treatment; §P , 0.05; §§P , 0.001 vs. ipsilateral side, control group (ANOVA with post-hoc Newman–Keuls tests).
(Vector), with 3,3′-diaminobenzidine as the chromogen. TH-immunoreactive neurones were counted in the SNpc and the VTA on both sides of the brain at each of three levels; −2.8, −3.0, −3.2, with respect to bregma [13]. There was extensive loss of TH-positive neurones in the ipsilateral SNpc and VTA of 6-OHDA only and GDF5 monomer rats (Table 1). In contrast, GDF5 dimer rats exhibited sparing of these neurones in both the SNpc and VTA. We have shown, using both in vivo functional and postmortem techniques, that GDF5 effectively protects the integrity of the rat dopaminergic nigrostriatal pathway against 6-OHDA lesion. GDF5 has protective actions on both striatal nerve terminals and nigral dopaminergic cell bodies. We are presently performing experiments aimed at elucidating the site of action and the potency of GDF5. In the present study, we used a total dose of 100 mg GDF5. However, our preliminary data (not shown) suggest that GDF5 has comparable effects to those shown here when given at a total dose of 50 or 20 mg.
76
A.M. Sullivan et al. / Neuroscience Letters 233 (1997) 73–76
Table 1 Counts of TH-positive neurones SNpc
VTA
Right 6-OHDA Monomer Dimer Control
129.1 124.6 116.4 132.2
Left ± ± ± ±
6.9 12.3 11.4 6.6
6.3 8.4 80.8 130.6
Left/Right (%) ± ± ± ±
1.9 1.7 13.7 1.2
4.9 6.8 69.1 98.9
± ± ± ±
§§
1.3 1.4§§ 7.1@§ 5.3
Right 130.4 132.3 123.6 132.2
Left ± ± ± ±
6.7 5.7 8.4 4.0
15.6 21.9 95.9 125.9
Left/Right (%) ± ± ± ±
5.6 3.4 12.5 10.0
12.2 16.6 77.4 95.2
± ± ± ±
4.9§§ 2.7§§ 5.2@§ 6.2
Counts of TH-positive neurones in right and left SNpc and VTA at 10 days after surgery, expressed as the mean ±SD of counts at three levels of the brain; −2.8, −3.0, −3.2, with respect to bregma. N = 5 for each group. @P , 0.001 vs. 6-OHDA only and GDF5 monomer groups; §P , 0.05; §§P , 0.001 vs. control group (ANOVA with post-hoc Newman–Keuls test; F(3,14) = 484.8, P , 0.0001 for SN; F(3,14) = 430.8, P , 0.0001 for VTA).
Whilst a number of neurotrophic factors have been shown to have supportive or neuroprotective actions on dopaminergic neurones, GDNF is thought to have the most pronounced effects in vivo. However, the magnitude of the effect of GDF5, both in vitro [8] and in the experimental model used in the present study, is very similar to that which we have previously found for GDNF [11]. We have demonstrated here the first evidence that GDF5 has powerful protective actions on dopaminergic cell bodies and nerve terminals in the adult rat nigrostriatal pathway, an effect which is reflected functionally in terms of behaviour and DA reuptake capacity within the striatum. This factor may prove to be of value in the treatment of patients with Parkinson’s disease. This work has been partly funded by the Parkinson’s Disease Society, UK and by the BMBF, Germany. The authors wish to thank the PET Methodology Group, MRC Cyclotron Unit, for assistance with PET studies. [1] Beck, K.D., Valverde, J., Alexi, T., Poulsen, K., Moffat, B., Vandlen, R.A., Rosenthal, A. and Hefti, F., Mesencephalic dopaminergic neurons protected by GDNF from axotomy-induced degeneration in the adult brain, Nature, 373 (1995) 339–341. [2] Bloomfield, P.M., Myers, R., Hume, S.P., Spinks, T.J., Lammertsma, A.A. and Jones, T., Three-dimensional performance of a small-diameter positron emission tomograph, Phys. Med. Biol., 42 (1997) 389–400. [3] Bloomfield, P.M., Rajeswaran, S., Spinks, T.J., Hume, S.P., Myers, R., Ashworth, S., Clifford, K.M., Jones, W.F., Byars, L.G., Young, J., Andreaco, M., Williams, C.W., Lammertsma, A.A. and Jones, T., The design and physical characteristics of a small animal positron emission tomograph, Phys. Med. Biol., 40 (1995) 1105–1126. [4] Hoffer, B.J., Hoffman, A., Bowenkamp, K., Huettl, P., Hudson, J., Martin, D., Lin, L.-F.H. and Gerhardt, G.A., Glial cell linederived neurotrophic factor reverses toxin-induced injury to midbrain dopaminergic neurons in vivo, Neurosci. Lett., 182 (1994) 107–111. [5] Hotten, G., Neidhardt, H., Jacobowsky, B. and Pohl, J., Cloning and expression of recombinant human growth/differentiation factor 5, Biochem. Biophys. Res. Commun., 204 (1994) 646–652. [6] Hume, S.P., Myers, R., Bloomfield, P.M., Opacka-Juffry, J., Cremer, J.E., Ahier, R.G., Luthra, S.J., Brooks, D.J. and Lammertsma, A.A., Quantitation of carbon-11-labeled raclopride in rat striatum using positron emission tomography, Synapse, 12 (1992) 47–54.
[7] Hume, S.P., Luthra, S.K., Brown, D.J., Opacka-Juffry, J., Osman, S., Ashworth, S., Myers, R., Brady, F., Carroll, F.I., Kuhar, M.J. and Brooks, D.J., Evaluation of [11C]RTI-121 as a selective radioligand for PET studies of the dopamine transporter, Nucl. Med. Biol., 23 (1996) 377–384. [8] Krieglstein, K., Suter-Crazzolara, C., Hotten, G., Pohl, J. and Unsicker, K., Trophic and protective effects of growth/differentiation factor 5, a member of the transforming growth factor-b superfamily, on midbrain dopaminergic neurons, J. Neurosci. Res., 42 (1995) 724–732. [9] Lin, L.-F.H., Doherty, D.H., Lile, J.D., Bektesh, S. and Collins, F., GDNF: A glial cell line-derived neurotrophic factor for midbrain dopaminergic neurons, Science, 260 (1993) 1130–1132. [10] Opacka-Juffry, J., Ahier, R.G. and Cremer, J.E., Nomifensineinduced increase in extracellular striatal dopamine is enhanced by isoflurane anaesthesia, Synapse, 7 (1991) 169–171. [11] Opacka-Juffry, J., Ashworth, S., Hume, S.P., Martin, D., Brooks, D.J. and Blunt, S.B., GDNF protects against 6-OHDA nigrostriatal lesion: in vivo study with microdialysis and PET, NeuroReport, 7 (1995) 348–352. [12] Partridge, N.C., Alcorn, D., Michelangeli, V.P., Kemp, B.E., Ryan, G.B. and Martin, T.J., Functional properties of hormonally responsive cultured and malignant rat osteoblastic cells, Endocrinology, 108 (1981) 213–219. [13] Pellegrino, L.J., Pellegrino, A.S. and Cushman, A.J., A Stereotaxic Atlas of the Rat Brain, Plenum Press, New York, 1979. [14] Poulsen, K.T., Armanini, M.P., Klein, R.D., Hynes, M.A., Phillips, H.S. and Rosanthal, A., TGFb2 and TGFb3 are potent survival factors for midbrain dopaminergic neurons, Neuron, 13 (1994) 1245– 1252. [15] Robb, R.A. and Hansen, D.P., A software system for interactive and quantitative visualization of multidimensional biomedical images, Aust. Phys. Eng. Sci., 14 (1991) 9–30. [16] Storm, E.E., Huynh, T.V., Copeland, N.G., Jenkin, N.A., Kingsley, D.M. and Lee, S.-J., Limb alterations in bradypodism mice due to mutations in a new member of the TGFb-superfamily, Nature, 368 (1994) 639–643. [17] Tomac, A., Lindqvist, E., Lin, L.-F.H., Ogren, S.O., Young, D., Hoffer, B.J. and Olson, L., Protection and repair of the nigrostriatal dopaminergic system by GDNF in vivo, Nature, 373 (1995) 335– 339. [18] Ungerstedt, U. and Arbuthnott, G.W., Quantitative recording of rotational behaviour in rats after 6-hydroxydopamine lesions of the nigrostriatal dopamine system, Brain. Res., 24 (1970) 485–493. [19] Yamaguchi, A., Katagiri, T., Ikeda, T., Wozney, J.M., Rosen, V., Wang, E.A., Kahn, A.J., Suda, T. and Yoshiki, S., Recombinant human bone morphogenetic protein-2 stimulates osteoblastic maturation and inhibits myogenic differentiation in vitro, J. Cell Biol., 113 (1991) 681–687.
