Begoña Sánchez J Neurosci Res

1,25-Dihydroxyvitamin D3 Administrationto 6-Hydroxydopamine-Lesioned RatsIncreases Glial Cell Line-DerivedNeurotrophic Factor and Partially RestoresTyrosine Hydroxylase Expression inSubstantia Nigra and Striatum

Begona Sanchez,1 Jose L. Relova,1 Rosalia Gallego,2 Isabel Ben-Batalla,1

and Roman Perez-Fernandez1*1Department of Physiology, School of Medicine, University of Santiago de Compostela,Santiago de Compostela, Spain2Department of Morphological Sciences, School of Medicine, University of Santiago de Compostela,Santiago de Compostela, Spain

It has previously been demonstrated that 1,25-dihydroxy-vitamin D3 [1,25(OH)2D3] administration, whether in cellcultures or in vivo to rats, increases glial cell line-derivedneurotrophic factor (GDNF) expression levels, suggestingthat this hormone may have beneficial effects in neurode-generative disorders. This study was carried out toexplore the effects of 1,25(OH)2D3 administration in a 6-OHDA-lesioned rat model of Parkinson’s disease onGDNF and tyrosine hydroxylase (TH) expression in sub-stantia nigra (SN) and striatum. Two groups of animalsreceived 1,25(OH)2D3 intraperitoneally, the first group 7days before the unilateral injection of 6-OHDA into themedial forebrain bundle (MFB) and the second group 21days (days 21–28) after the unilateral injection of 6-OHDA.Animals of both groups were sacrificed on day 28. In addi-tion, two other groups received a unilateral injection of ei-ther saline or 6-OHDA into the MFB. Rats were killed, andthe SN and striatum were then removed for GDNF and THdetermination. Striatal GDNF protein expression wasincreased on the ipsilateral with respect to the contralat-eral side after 6-OHDA injection alone as well as in1,25(OH)2D3-treated rats before or after 6-OHDA adminis-tration. As expected, 6-OHDA injection induced an ipsilat-eral decrease in TH-immunopositive neuronal cell bodiesand axonal terminals in the SN and striatum. However,treatment with 1,25(OH)2D3 before and after 6-OHDAinjection partially restored TH expression in SN. Thesedata suggest that 1,25(OH)2D3 may help to prevent dopa-minergic neuron damage. VVC 2008Wiley-Liss, Inc.

Key words: vitamin D; GDNF; Parkinson’s disease; TH;6-OHDA

Parkinson’s disease (PD) is a progressive neurodege-nerative disorder affecting predominantly dopaminergic

neurons of the substantia nigra (SN). One of the goals ofthe current main therapeutic approaches is to restore stria-tal dopamine levels in order to improve motor function.However, because the course of the disease is slow andprogressive, some therapeutic approaches aim to block orslow the disease course. Indeed, imaging data suggest thatit may be possible to detect a decline in striatal dopaminefunction even before the onset of overt clinical symp-toms, which would make it possible to initiate neuropro-tective interventions in the very early stages of the disease,i.e., at the time when the first symptoms appear, or evenbefore (Dunnett and Bjorklund, 1999).

Neurotrophic factors are interesting candidates forneuroprotective therapies, because they regulate neuronaldifferentiation, phenotype maintenance, and synapticsprouting. Neurotrophic factors also protect adult neu-rons from mechanical, toxic, or ischemic injuries andinterfere in the death of neurons by necrosis or apoptosis(Siegel and Chauhan, 2000). Various neurotrophic fac-tors, notably glial cell line-derived neurotrophic factor(GDNF), have effects on dopaminergic neurons not onlyin vitro but also in vivo (Lin et al., 1993). It has beenshown that intracerebral GDNF injection has neuropro-tective effects against neurotoxins such as 1-methyl-4-

Contract grant sponsor: Xunta de Galicia, Spain; Contract grant number:

PGIDIT06PXIB208028PR; Contract grant number: PGIDIT07P-

XIB208112PR.

*Correspondence to: Roman Perez-Fernandez, Departamento de Fisiolo-

gıa, Facultad de Medicina, Universidad de Santiago de Compostela,

15782 Santiago de Compostela, Spain. E-mail: [email protected]

Received 5 May 2008; Revised 30 June 2008; Accepted 13 July 2008

Published online 24 September 2008 in Wiley InterScience (www.

interscience.wiley.com). DOI: 10.1002/jnr.21878

Journal of Neuroscience Research 87:723–732 (2009)

' 2008 Wiley-Liss, Inc.

phenyl-1,2,3,6-tetrahydropyrine (MPTP) and 6-hy-droxydopamine (6-OHDA) in animal models of PD(Tomac et al., 1995; Gash et al., 1996; Bjorklund et al.,1997). Up-regulation of GDNF expression, as shown byincreased GDNF mRNA, is seen after administration of1-methyl-4-phenylpyridinium (Tang et al., 1998).

1,25-Dihydroxyvitamin D3 (1,25(OH)2D3), anactive metabolite of vitamin D, activates gene transcriptionthrough the binding to its nuclear receptor, the vitamin Dreceptor, on specific DNA sequences. 1,25(OH)2D3 isalso a potent inducer of GDNF expression in rat C6 gli-oma cells (Naveilhan et al., 1996) and GDNF release inhuman U-87 MG glioblastoma cells (Verity et al., 1999).1,25(OH)2D3 increases GDNF levels in rat cerebral cortex(Wang et al., 2000), and intraperitoneal (i.p.) administra-tion of 1,25(OH)2D3 increases GDNF levels in rat stria-tum (Sanchez et al., 2002). In addition, it has beenreported that, in PD patients serum, 1,25(OH)2D3 levelsare decreased (Sato et al., 1997), and supplemental admin-istration of 25-(OH) vitamin D3 and calcium significantlyimproves symptoms and signs of parkinsonism, decreasingrigidity and akinesia (Derex and Trouillas, 1997). Takentogether, these data suggest that 1,25(OH)2D3 modulatesGDNF expression both in vitro and in vivo. As describedabove, GDNF protects and repairs dopaminergic neuronsof the nigrostriatal system but is unable to cross the blood–brain barrier, thus requiring invasive surgical techniquesfor its intracerebral administration. However, 1,25(OH)2D3 (and its analogues) has the ability to cross the blood–brain barrier (Gascon-Barre and Huet, 1983), and there-fore could protect and/or repair nigrostriatal dopaminergicneurons against neuronal injury by inducing expression ofneurotrophic factors in the brain. In the present study,using an animal model of Parkinson’s disease induced byintracerebral injection of 6-OHDA, we studied (by West-ern blot, RT-PCR, and immunohistochemistry) theeffects of systemic 1,25(OH)2D3 administration, beforeand after 6-OHDA-lesion, on GDNF and tyrosinehydroxylase expression.

MATERIALS AND METHODS

Experimental Design

Male Sprague-Dawley rats weighing 250–300 g at thestart of the experiment were used in this study. They werehoused in a temperature-controlled environment and main-tained under a normal 12-hr light 12-hr dark cycle, with freeaccess to food and water. All procedures were approved bythe Ethics Committee of the University of Santiago de Com-postela, and the experiments were performed in accordancewith standard rules on laboratory animal care and internationallaw on animal experimentation. Animals were deeply anesthe-tized with sodium pentobarbital (50 mg/kg, i.p.; Sigma-Aldrich, St. Louis, MO) and immobilized in a stereotaxicframe. The skull was exposed, and a burr hole was drilledthrough the skull at the appropriate location in order to intro-duce a Hamilton microsyringe for a single injection of 6-hy-droxydopamine hydrochloride (6-OHDA; 8 lg/2 ll salinecontaining 0.02% ascorbic acid; Sigma-Aldrich) and/or ascor-

bic acid (2 ll saline containing 0.2% ascorbic acid) in the leftMFB.

To minimize variability resulting from degradation of thetoxin, 6-OHDA and ascorbic acid solutions were freshly made,kept at 48C, and protected from light. The fresh solutions weredelivered from microsyringe at a rate of 1 ll/5 min, and anadditional 5 min was allowed to elapse before the needle wasremoved. Stereotaxic coordinates were 4.0 mm anterior to theinteraural line, 1.2 mm lateral from the midline, and 8.4 mmventral to the dural surface, with the incisor bar located4.5 mm below the interaural line. The coordinates wereassessed according to the atlas of Paxinos and Watson (1986).

Animals were divided into four groups, each of 30 rats[for GDNF mRNA determination; for GDNF and tyrosinehydroxylase (TH) protein determination; and for GDNF andTH immunohistochemistry]: 1) group 1, sham-lesioned ratsreceived 2 ll of saline 0.9% containing 0.02% ascorbic acid inthe left MFB and were then kept in their cages for a total of21 days, after which they were sacrificed; 2) group 2, ratsreceived on the first day 6-OHDA in the left MFB and werethen kept in their cages for a total of 21 days, then sacrificed;3) group 3, rats received for 7 consecutive days 1,25(OH)2D3

intraperitoneally (i.p.; 1 lg/ml/kg per day; calcitriol; Sigma-Aldrich), then on day 7 they were lesioned with 6-OHDAinjected in the left MFB, and finally were sacrificed 21 dayslater (day 28); and 4) group 4, on day 1 the rats were injectedwith 6-OHDA into left MFB, and 21 days later they wereinjected i.p. with 1,25(OH)2D3 daily (1 lg/ml/kg) until day28, when they were sacrificed. Animals were killed by decapi-tation, the brains removed, and ipsilateral and contralateralstriata (with respect to the 6-OHDA or saline injection) iso-lated for GDNF mRNA and TH and GDNF protein evalua-tion. The noninjected side of each rat was used as internalcontrol. For immunohistochemical determinations of GDNFand TH in SN and striatum, animals were transcardially per-fused as described below.

RNA Isolation and RT-PCR

Striatal tissue was removed quickly and stored at –808C.Isolation of total RNA from tissues was performed with Trizolreagent (Invitrogen, Barcelona, Spain) according to the manu-facturer’s instructions. RT-PCR was performed as previouslydescribed (Sanchez et al., 2002). Samples were denatured at948C for 1 min, annealed at 578C or 628C for 1 min (forGDNF or GAPDH, respectively), and extended at 728C for1 min, for a total of 35 cycles for GDNF and 27 cyclesfor GAPDH, with an extension step at 728C in the final cycle.A PCR using 1/2, 1/5, and 1/10 of starting cDNA (5 ll) wasalso performed to evaluate the saturation profile of the sample.To determine the relative amounts of GDNF mRNA in eachsample, GDNF PCR product was compared with the GADPHPCR product and quantified using the Gel Doc Documenta-tion System (Bio-Rad Laboratories, Hercules, CA).

Primer Sequences

Primer sequences for GDNF PCR amplifications wereas follows: primer A (50-CCAATATGCCCGAAGAT-TATCC-30) was a 22-mer corresponding to nucleotides 201–

724 Sanchez et al.

Journal of Neuroscience Research

222 of GDNF cDNA, and primer B (50-TTCGTAGCC-CAAACCCAAG-30) was an antisense 19-mer correspondingto nucleotides 436–454 of GDNF cDNA; the PCR productobtained was 254 bp long. The rat glyceraldehyde phosphatedehydrogenase (GAPDH) was used as ‘‘internal reference.’’The forward primer was 50-TGATGACATCAAGAAGGTGGTGAAG-30, residues 758–782 of the rat codingsequence, and the reverse primer 50-TCCTTGGAGGCCATGTAGGCCAT-30, residues 974–997 of the codingsequence. The PCR product obtained was 239 bp in length.

Immunohistochemistry

After 21 days (groups 1 and 2) or 28 days (groups 3 and4) rats were perfused through the ascending aorta with 150 ml0.9% NaCl, followed by 500 ml of 4% paraformaldehyde in100 mM phosphate buffer (PB), pH 7.4, under deep anaesthe-sia with pentobarbital (50 mg/kg, i.p.; Sigma-Aldrich). Brainswere removed and postfixed in the same fixative for 2–4 hr at48C and then transferred to PB for 1 hr. Fifty-micrometer-thick coronal sections showing the SN and striatum were cutwith a vibratome (Vibratome Series 1000; Warner Instru-ments, Hamden, CT). The free-floating sections were col-lected and incubated overnight at 48C with anti-TH antibody(mouse monoclonal, 1:1,000; Sigma-Aldrich) or anti-GDNFantibody (rabbit polyclonal, 1:500; Santa Cruz Biotechnology,Santa Cruz, CA). After two washes (5 min each) in PBS (100mM phosphate buffer, 0.65 M NaCl), sections were incubatedwith peroxidase-labelled dextran polymer and anti-mouse (forTH) or anti-rabbit (for GDNF) IgG antibody as appropriate(EnVision Peroxidase; Dakopatts, Glostrup, Denmark) for30 min at room temperature. All of these sections were washedtwice for 5 min with PBS and then revealed with 3,30-diami-nobenzidine (DAB; Dakopatts). Sections were mounted onpoly-L-lysine-coated slides (Histobond, Paul MarienfeldGmbH & Co., Lauda-Konigshofen, Germany), dehydrated inalcohol, cleared in xylene, and finally coverslipped and sealed.Negative controls were performed by preadsorbing the anti-GDNF antibody with the homologous peptide (blocking pep-tide, sc-328 P; Santa Cruz Biotechnology; 10 nmol/ml).

Estimation of the density of the number of GDNF andTH-immunopositive cell bodies in the striatum and the SN ofeach rat (n 5 6 per group) was performed as previouslydescribed (Lopez-Martın et al., 2006), with minor modifica-tions. Briefly, five sections were taken at the different levels ofSN and striatum nuclei. The sections were examined using alight microscope (Olympus Provis AX 70; Olympus Corp.,Tokyo, Japan) connected to an image analysis workstation.Images with a final magnification of 3920 were randomly pho-tographed with a high-resolution digital camera (DP 70; Olym-pus). A counting frame (102 lm2) was superimposed on thecaptured image, and the immunopositive cells with only aclearly visible nucleus were counted on the photomicrographs.Intensity of TH positivity in the striatum was measured on sec-tions at three different levels of nucleus by means of a relativeoptical density value (Paille et al., 2007). This was obtained af-ter transformations of mean gray values into relative opticaldensity, using the following formula: relative optical density 5log (256/mean gray). A background parameter was obtained

for each section in the corpus callosum and subtracted fromeach measure obtained in the striatum of both sides. Results areexpressed as percentage of the control nonlesioned side. Somasize of TH-immunopositive cells was measured in ImageJ soft-ware (National Institutes of Health, Bethesda, MD).

Western Blot Analysis

Striatum was removed and homogenized at 48C in 1 mlof lysis buffer [50 mM HEPES, pH 7.5, 150 mM NaCl,5 mM EGTA, 1.5 mM MgCl2, 1% SDS, 10% glycerol, 1%Triton X-100, 10 mM sodium orthovanadate, 4 mM phenyl-methylsulfonyl fluoride (PMSF), and 50 mg/ml aprotinin].The cell lysate was then centrifuged at 1,000g for 15 min at48C, the resulting supernatant was collected, and protein con-centration was determined by the Bradford method. Seventy-five micrograms of total protein was resuspended in 23 so-dium dodecyl sulfate (SDS) sample buffer (50 mM Tris-HCl,pH 6.8, 50% glycerol, 2% SDS, and bromophenol blue) andboiled for 5 min. The samples were subjected to 12% SDS-PAGE electrophoresis under nonreducing conditions. Proteinswere then transferred to a nitrocellulose membrane for 2 hr at48C and blocked with 0.10 g casein in PBS with 0.1% Tween20 (PBST) for 2 hr at room temperature. The blots were thenimmunolabelled overnight at 48C with a polyclonal anti-GDNF antiserum (1:200; Santa Cruz Biotechnology) or withanti-TH antibody (mouse monoclonal, 1:1,000; Sigma-Aldrich). After five washes each in PBST, membranes wereincubated with alkaline phosphatase-conjugated goat anti-rab-bit IgG (1:5,000), using CSPD (Tropix; PE Biosystem, Bed-ford, MA) as substrate for 1 hr at room temperature. Themembrane was further washed five times for 5 min each inPBST before immunolabelling. Immunolabelling was detectedby placing the blot on standard X-ray film according to themanufacturer’s instructions (Tropix). To confirm that equiva-lent amounts of total protein were added to each well, mem-branes were stripped by incubation in 0.2 mM glycine, pH2.5, containing 0.1% SDS and 1% Tween-20, at room tem-perature for 1 hr, and then reprobed with anti-a-tubulinmonoclonal antibody (1:2,000; Sigma) or a monoclonalanti-b-actin antiserum (1:2,000; Sigma-Aldrich). The opticaldensity of immunolabelling on autoradiographic films wasquantified under visible light using the Gel Doc 1000 Docu-mentation System (Bio-Rad, Hercules, CA). To determinethe relative amounts of GDNF and TH protein in eachsample, absolute amounts of GDNF and TH were expressedrelative to a-tubulin or b-actin amounts.

Statistical Analysis

All values are expressed as means 6 SD. Means werecompared by Mann-Whitney U-tests or unpaired t-tests. Sta-tistical significance was indicated at P < 0.05.

RESULTS

Sham-Lesioned Animals Did Not Show Changesin GDNF mRNA and Protein or TH Proteinin Striatum

To evaluate possible interference of intracerebralinjection with GDNF and TH expression, rats were uni-

Vitamin D Increases GDNF in Striatum 725


laterally injected in the MFB with 2 ll saline containing0.2% ascorbic acid, and GDNF mRNA and GDNF andTH protein levels were then evaluated. No changes instriatal GDNF mRNA (Fig. 1A,B) or GDNF and THprotein (Fig. 1C,D, and E,F, respectively) were observedon the ipsilateral injected side with respect to the contra-lateral side in sham-lesioned animals.

6-OHDA-Lesioned Rats Showed Increased StriatalGDNF Protein Levels and Reduced TH Protein onthe Ipsilateral Lesioned Side With Respect to theContralateral Side

6-OHDA administration induces motor dysfunc-tion as evidenced by contralateral rotations after subcuta-neous apomorphine administration (data not shown).Twenty-one days after 6-OHDA injection, both GDNFmRNA and GDNF protein were evaluated. Bands cor-

responding to the expected size of GDNF cDNA (254bp) and GAPDH cDNA (239 bp) were amplified byPCR and visualized in agarose ethidium bromide gel(Fig. 2A). There was no difference in GDNF mRNAlevels between ipsilateral striatum from lesioned animals(0.80 6 0.04) and contralateral nonlesioned striatum(0.73 6 0.07; Fig. 2B). GDNF protein expression wasalso evaluated by Western blot in both ipsilateral andcontralateral striatum (Fig. 2C,D). The densitometricanalysis shows a significant increase (P < 0.01) inGDNF protein levels in the ipsilateral striatum of 6-OHDA-lesioned animals (0.60 6 0.12) with respect tothe nonlesioned contralateral striatum (0.36 6 0.15; Fig.2D). A significant decrease of TH protein in homoge-nates from ipsilateral striatum (0.46 6 0.10) with respectto the contralateral striatum (0.70 6 0.14) in 6-OHDA-lesioned rats was observed (Fig. 2E,F).

Fig. 1. GDNF mRNA and protein and tyrosine hydroxylase (TH)protein expression is not modified in ipsilateral vs. contralateral stria-tum of saline-injected rats (group 1, sham-lesioned rats). Saline-injected animals (n 5 8) were killed after 21 days, the brain wasremoved, and striata from the injection side (ipsilateral, I) or contra-lateral side (C) were isolated for RT-PCR and Western blotting. A:GDNF and GAPDH PCR products. Lane 1: Molecular weightmarker; lane 18: negative control of PCR. B: Relative GDNFmRNA expression calculated from the GDNF/GAPDH ratioobtained from densitometric values of A. C: Western blots of GDNFand a-tubulin in striata of eight rats. D: Relative GDNF proteincontent in striata calculated from C as described above. E: Westernblots of TH and b-actin in striata of eight rats. F: Relative TH pro-tein content in striata calculated from E as described above.

Fig. 2. Effect of 6-OHDA on GDNF mRNA and protein and tyro-sine hydroxylase (TH) protein expression (group 2, 6-OHDA-lesionedrats). Rats (n 5 8) received on the first day 6-OHDA in the leftmedial forebrain bundle and were killed after 21 days. The brain wasremoved, and striata from the injection side (ipsilateral, I) or contralat-eral side (C) were isolated for RT-PCR and Western blotting. A:GDNF and GAPDH PCR products. Lane 1: Molecular weightmarker; lane 18: negative control of PCR. B: Relative GDNFmRNA expression calculated as described for Figure 1. C: Westernblots of GDNF and a-tubulin in striata. D: Relative GDNF proteincontent in striata. E: Western blots of TH and b-actin in striata. F:Relative TH protein content in striata calculated from E as describedfor Figure 1. ***P < 0.001 compared with contralateral side.

726 Sanchez et al.


Previous 1,25(OH)2D3 Administration IncreasesGDNF Protein Levels and Partially RestoresTH Expression on the 6-OHDA-Lesioned Sideof Striatum

Seven days after systemic (i.p.) administration of1,25(OH)2D3, animals were injected with 6-OHDA,and 21 days later striatal GDNF mRNA levels wereevaluated by using semiquantitative RT-PCR (Fig. 3A).We did not observe significant differences in GDNFmRNA expression between 6-OHDA-lesioned (0.76 60.08) and contralateral (0.75 6 0.10) striatum (Fig. 3B).We also evaluated GDNF protein expression in the 6-OHDA-lesioned and contralateral striatum in systemic1,25(OH)2D3 pretreated rats (Fig. 3C). GDNF proteinlevels were significantly (P < 0.001) higher on the 6-OHDA-lesioned side (0.77 6 0.15) with respect to the

contralateral striatum (0.53 6 0.19; Fig. 3D). However,no significant differences were found in TH proteincontent between homogenates from ipsilateral and con-tralateral striatum of 6-OHDA rats, previously treatedwith 1,25(OH)2D3 (Fig. 3E,F).

1,25(OH)2D3 Administration After 6-OHDAInjection Induces an Increase in GDNFProtein Levels

1,25(OH)2D3 administration for 7 days after 21days of 6-OHDA injection did not change GDNFmRNA expression in the injection-side striatum (0.67 60.15) with respect to the contralateral side (0.66 6 0.08;Fig. 4A,B). However, the levels of GDNF proteindetected using Western blot (Fig. 4C) were significantly

Fig. 3. Effect of pretreatment with 1,25(OH)2D3 (Vit D) on GDNFmRNA and protein and tyrosine hydroxylase (TH) protein expres-sion. Rats were treated intraperitoneally with 1,25(OH)2D3 (1 lg/ml/kg per day) for 7 consecutive days and then lesioned on day 7 with 6-OHDA injected in the left medial forebrain bundle and killed on day28 (group 3). The brain was removed, and striata from the injectionside (ipsilateral; I) or contralateral side (C) were isolated for RT-PCRand Western blotting. A: GDNF and GAPDH PCR products. Lane1: Molecular weight marker; lane 18: negative control of PCR. B:Relative GDNF mRNA expression calculated as described for Figure1. C: Western blots of GDNF and a-tubulin in striata. D: RelativeGDNF protein content in striata calculated from C as described forFigure 1. E: Western blots of TH and b-actin in striata. F: RelativeTH protein content in striata calculated from E as described for Figure1. ***P < 0.001 compared with contralateral side.

Fig. 4. Effect of posttreatment with 1,25(OH)2D3 (Vit D) on GDNFmRNA and protein and tyrosine hydroxylase (TH) protein expres-sion. Rats were lesioned on day 1 with 6-OHDA injected in the leftmedial forebrain bundle and 21 days later were treated intraperio-neally with 1,25(OH)2D3 (1 lg/ml/kg per day) for 7 consecutivedays until day 28, when they were sacrificed (group 4). The brainwas removed, and striata from the injection side (ipsilateral, I) orcontralateral side (C) were isolated for RT-PCR and Western blot-ting. A: GDNF and GAPDH PCR products. Lane 1: Molecularweight marker; lane 18: negative control of PCR. B: RelativeGDNF mRNA expression calculated as described for Figure 1. C:Western blots of GDNF and a-tubulin in striata. D: Relative GDNFprotein content in striata calculated from C as described for Figure 1.E: Western blots of TH and b-actin in striata. F: Relative TH pro-tein content in striata calculated from E as described for Figure 1. Cand I denote contralateral and ipsilateral, with respect to the lesionedside. ***P < 0.001 compared with contralateral side.



higher (P < 0.01) in the injection-side striatum (0.88 60.08) than in the contralateral striatum (0.62 6 0.10;Fig. 4D). TH-immunoreactive protein showed no differ-ences between sides (Fig. 4E,F).

GDNF and TH protein expression in striatum ofthe ipsilateral-lesioned side was also compared in thefour experimental groups. A significant (P < 0.05)increase in GDNF protein expression was observed ingroups 2–4 compared with the sham-lesioned animals.This GDNF increased levels were also significantly (P <0.05) higher in both pre- and post-1,25(OH)2D3-treatedanimals, compared only with the 6-OHDA-lesioned rats.However, no significant differences were found inGDNF expression between pre- and post-1,25(OH)2D3-treated groups. A significant (P < 0.01) decrease in THexpression in striatum was observed after 6-OHDAadministration compared with the sham-lesioned rats.

However, 1,25(OH)2D3 treatment partially recoveredTH expression, being significant (P < 0.05) in the post-treated animals (group 4).

1,25(OH)2D3 Treatment Significantly IncreasesGDNF-Immunoreactive Neurons in Striatum andInduces a Recovery of TH-ImmunoreactiveNeurons in SN

No immunoreactivity was seen in either SN orstriatum when preabsortion of GDNF antibody with thehomologous peptide was carried out (data not shown).No changes in the number of nigral or striatal GDNF(471 6 6 vs. 473 6 15, and 292 6 73 vs, 302 6 66,respectively; Fig. 5A,E,F) or nigral TH-immunoreactivecells (769 6 58 vs. 794 6 34; Fig. 6A,E) were observedon the ipsilateral injected side with respect to the contra-lateral side in sham-lesioned animals.

Fig. 5. GDNF immmunoreactivity in substantia nigra (upper andmiddle panels) and striatum (lower panels) in sham-lesioned rats (A;group 1) and in rats unilaterally lesioned with 6-OHDA alone (B;group 2) or in rats pretreated (C; group 3) or posttreated (D; group4) with 1,25(OH)2D3 (Vit D). GDNF immunoreactivity in substantianigra and striatum from sham-lesioned rats was similar on both theinjected and contralateral sides. An increase in number and intensity

of GDNF-positive cells in the injected-side striatum, compared withthe contralateral side, was observed in rats that received 6-OHDAalone but was more evident in 1,25(OH)2D3 pre- and posttreatedanimals (E,F). Asterisk indicates the 6-OHDA-lesioned side. *P <0.05, ***P < 0.001 compared with contralateral side. [Color figurecan be viewed in the online issue, which is available at www.interscience.wiley.com.]

728 Sanchez et al.


However, immunohistochemistry indicated anincreased number of GDNF-positive cells on the 6-OHDA-lesioned side in SN (531 6 108; Fig. 5B) andstriatum (427 6 47; Fig. 5B,E,F), with respect to thecontralateral tissues (459 6 117, and 361 6 44, respec-tively). There was also a statistically significant (P <0.001) loss of TH-immunoreactive cells in SN (46 6 29vs. 732 6 172) between the injected side and the nonin-jected contralateral side, 3 weeks after 6-OHDA admin-istration (Figs. 6B,E).

Pretreatment with 1,25(OH)2D3 (group 3) induced asignificant increase in GDNF-positive cells in SN (582 677 vs. 491 6 114; P < 0.05) and striatum (707 6 94vs. 396 6 85; P < 0.001) on the 6-OHDA-lesionedside with respect to the contralateral side, respectively

(Fig. 5C,E,F). Significant differences (P < 0.05) in thenumber of TH-immunopositive cells were also foundbetween ipsilateral and contralateral 6-OHDA-lesionedsides (438 6 105 vs. 742 6 31, respectively) of the SNafter pretreatment with 1,25(OH)2D3 (Fig. 6C,E).

The number of GDNF-immunopositive cells inthe striatum was significantly (P < 0.001) increased onthe injection side with respect to the contralateral side(592 6 81 vs. 397 6 49; Fig. 5D,E) in rats treated with1,25(OH)2D3 after 6-OHDA lesion. The number ofGDNF-immunopositive cells in the SN was also higherin the ipsilateral-lesioned side with respect to the con-tralateral side, but this was not statistically significant(597 6 87 vs. 481 6 87; Fig. 5D,F). The number ofTH-positive neurons was significant lower (P < 0.001)

Fig. 6. Tyrosine hydroxylase (TH) immunoreactivity in substantianigra (upper and middle panels) and striatum (lower panels) in sham-lesioned rats (A; group 1) and in rats unilaterally lesioned with 6-OHDA alone (B; group 2) or in rats pretreated (C; group 3), orposttreated (D; group 4) with 1,25(OH)2D3 (Vit D). E: TH expres-sion in substantia nigra of control rats shows similar intensity andnumber of TH-positive neurons on the injected and contralateralsides. However, TH immunoreactivity of 6-OHDA-injected rats

showed significantly reduced intensity on the injected side comparedwith the contralateral side. TH expression in 1,25(OH)2D3 pre- andposttreated rats was also significantly reduced in the injected sidecompared with the contralateral side. Asterisk indicates 6-OHDA-lesioned side. ***P < 0.001 compared with contralateral side. [Colorfigure can be viewed in the online issue, which is available atwww.interscience.wiley.com.]



on the injection side (272 6 105) than on the contralat-eral side of SN (785 6 99; Fig. 6D,E).

TH-immunopositive fiber density in striatum wascompared in ipsilateral-lesioned and contralateral (con-trol) sides. After a 6-OHDA treatment, a significant (P <0.01) reduction of 64% 6 13% in the density of TH-immunopositive fibers in the ipsilateral side was observed.This decrease in TH immunopositivity was also observedin the ipsilateral-lesioned side in both 1,25(OH)2D3 pre-and posttreated animals (26% 6 13% and 34% 6 18%,respectively) compared with the control side. However, asignificant (P < 0.05) increase in TH-immunopositivefibers was observed in the ipsilateral side of both pre- andpost-1,25(OH)2D3 treated rats only compared with theipsilateral side of the 6-OHDA-lesioned rats.

After 6-OHDA injection (groups 2–4), signs of do-paminergic cell degeneration were observed in the SN

of ipsilateral 6-OHDA-treated animals with respect totheir contralateral side (Fig. 7), including a severe(groups 2 and 4) and moderate (group 3) loss of TH-positive fibers. To explore the changes in cell morphol-ogy in more depth, the soma sizes of TH-positive cellsand their areas were also measured. However, no signifi-cant changes were found in the soma size of TH-posi-tive cells between the injected ipsilateral side vs. thecontralateral side in any group.

DISCUSSION

It has recently been shown in transgenic mice thatthe absence of Ret, which mediated GDNF signalling,results in a progressive and adult-onset loss of DA neu-rons specifically in the SN pars compacta (SNpc), degen-

Fig. 7. Tyrosine hydroxylase (TH)immunoreactivity in substantia nigra(SN) of sham-lesioned rats (group 1)and in rats unilaterally lesioned with 6-OHDA alone (group 2) or in rats pre-treated (group 3) or posttreated (group4) with 1,25(OH)2D3 (Vit D). TH celldensity in the injection-side substantianigra was markedly reduced in the 6-OHDA (group 2) and 6-OHDA 1 VitD (group 4) rats compared with thecontralateral side, whereas it was similarto the contralateral side in sham-lesioned (group 1) and 1,25(OH)2D3-pretreated (group 3) rats. TH-labelledneurons also showed morphologicalchanges, with reduced branching den-drites and TH fiber density. Note thatthe ipsilateral substantia nigra in1,25(OH)2D3 pretreated animals (group3) shows minimal changes in cell andfiber density with respect to the nonle-sioned contralateral side. [Color figurecan be viewed in the online issue,which is available at www.interscience.wiley.com.]

730 Sanchez et al.


eration of DA nerve terminals in the striatum, and pro-nounced glial activation (Kramer et al., 2007). Thesefindings and previously published studies (Granholmet al., 2000) establish that Ret and subsequent down-stream effectors are critical regulators in long-term main-tenance of the nigrostriatal DA system. However, it isinteresting to note the different sensitivities of the meso-limbic and nigrostriatal dopaminergic systems to activa-tion by Ret signalling through GDNF. Specifically,Ret/GDNF signalling is not essential in the mesolimbicdopaminergic system originated in the ventral tegmentalarea (VTA; Kramer et al., 2007). In addition, it has beendemonstrated that there are two different neuron popu-lations in SNpc, one located in the rostromedial SN thatretrogradely transports GDNF from ventral striatum andmakes these neurons more resistant to 6-OHDA and asecond type of neurons located in the caudoventral SN,which do not retrogradely transport GDNF from dorsalstriatum and therefore are more vulnerable to the 6-OHDA lesion (Barroso-Chinea et al., 2005). BecauseParkinson’s disease progresses mainly with dopaminergicdegeneration of the nigrostriatal system, starting fromcaudoventrolateral SN, it is possible that the pathwaysleading to cell death in Parkinson’s disease (e.g., thoseinitiated by neurotoxic agents) could share commonsteps, such as Ret/GDNG signalling, and thus it is rea-sonable to speculate on possible therapeutic approachesto PD that increase GDNF.

However, use of GDNF has a serious limitation,because it has to be administered intracerebrally in someway, for example, by cellular transplantation of GDNF-producing cells or by neurosphere delivery (Kilic et al.,2003, 2004; Dietz et al., 2006). Given that a significantincrease in GDNF expression has been shown to occurafter 1,25(OH)2D3 administration (Naveilhan et al.,1996; Sanchez et al., 2002; Smith et al., 2006), in thepresent study, using a rat model of Parkinson’s disease,we evaluated whether 1,25(OH)2D3 administrationbefore or after intracerebral injection of 6-OHDAincreases GDNF or TH expression and thus might haveneuroprotective and/or neuroreparative effects. Threeweeks after intracerebral 6-OHDA injection, TH immu-nohistochemistry showed evident signs of dopaminergicneuron loss, both in the SN and in the striatum. Thisneuron loss was confirmed with the apomorphine test insome animals. We also observed a statistically significantincrease in GDNF protein expression in 6-OHDA-lesioned striatum in comparison with the nonlesionedcontralateral striatum. However, we did not detect anysignificant modification of GDNF mRNA expression inany of the experimental groups. This could be explainedby a mechanism of anterograde transport, as has beendemonstrated for GDNF in sensory and motor neurons(which do not express detectable levels of GDNFmRNA) and in Schwann cells (Russell et al., 2000;Rind and Von Bartheld, 2002). Intraperitoneal1,25(OH)2D3 administration to rats before 6-OHDAinjection in the MFB seems to act as a neuroprotectiveagent, in view of the less severe neuron loss and less

severe degeneration of the remaining neurons. Effects of1,25(OH)2D3 administration on the nigrostriatal pathwayhave been reported previously. Wang et al. (2001) foundan increase in levels of dopamine and its metabolites inipsilateral SN of rats lesioned in the MFB with 6-OHDA following pretreatment with 1,25(OH)2D3. Inaddition, 1,25(OH)2D3 administration protects dopami-nergic neurons against dopaminergic toxins (Ibi et al.,2001), and more recently an increase in GDNF levels inSN has been observed in intraventricularly lesioned ratsafter systemic administration of 1,25(OH)2D3 (Smithet al., 2006). The present results are in line with theseprevious findings and support a neuroprotective role of1,25(OH)2D3 in Parkinson’s disease through inductionof GDNF expression. However, although 1,25(OH)2D3

administration after 6-OHDA injection also induced anincrease in GDNF expression, the relative increase inTH cells in SN was not as high as in pretreated rats.This may be because the 1,25(OH)2D3-induced GDNFincrease is insufficient to have substantial restorativeeffects in SN or striatum given the lesion caused by theprevious administration of 6-OHDA into the MFB. Inthis connection, it has been demonstrated that the neu-roprotective effect of GDNF on TH neurons in the ni-grostriatal dopaminergic system is dose dependent (Aoiet al., 2000, 2001). In the present study, we alsoobserved a significant increase in GDNF protein levelsafter 6-OHDA injection alone, i.e., without prior orsubsequent 1,25(OH)2D3 treatment, as has already beenreported (Yurek and Fletcher-Turner, 2001). However,a partial recovery of TH immunoreactivity in the ipsilat-eral SN was observed in 1,25(OH)2D3 pre- and post-treated rats but not in rats that received only 6-OHDA.This suggests that 1,25(OH)2D3 treatment induces dopa-minergic neuron regeneration, directly and/or via otherroutes, such as the Ret/GDNF pathway or cytokine-mediated pathways (Gurlek et al., 2002; Kramer et al.,2007). Nevertheless, GDNF together with anothermember of the GDNF family, neurturin (Rosenblabet al., 1999), is known to possess neurorestorative effectswhen administered after intracerebral 6-OHDA injectionin the rat; other neurotrophic factors could be alsoimplicated. Recently, Lindholm et al. (2007) have iden-tified a novel neurotrophic factor, conserved dopamineneurotrophic factor (CDNF), which is related to themesencephalic-astrocyte-derived neurotrophic factor(MANF) family. CDNF prevents the 6-OHDA-induceddegeneration of dopaminergic neurons and restores do-paminergic neurons after 6-OHDA lesioning and couldbe a good candidate, along with GDNF, for exploringpossible therapeutic aproaches.

In conclusion, this study has demonstrated thatpretreatment with 1,25(OH)2D3 increases GDNFexpression and diminishes the decrease in TH immu-noreactivity in the SN in a rat model of Parkinson’sdisease induced by 6-OHDA injection. Thus 1,25(OH)2D3 or its analogues may possibly be useful forpreventing or reducing neuronal loss in the nigrostriatalpathway.



REFERENCES

Aoi M, Date I, Tomita S, Ohmoto T. 2000. The effect of intrastriatal

single injection of GDNF on the nigrostriatal dopaminergic system in

hemiparkinsonian rats: behavioral and histological studies using two dif-

ferent dosages. Neurosci Res 36:319–325.

Aoi M, Date I, Tomita S, Ohmoto T. 2001. Single administration of

GDNF into the striatum induced protection and repair of the nigrostri-

atal dopaminergic system in the intrastriatal 6-hydroxydopamine injec-

tion model of hemiparkinsonism. Restor Neurol Neurosci 17:31–38.

Barroso-Chinea P, Cruz-Muros I, Aymerich MS, Rodriguez-Diaz M,

Afonso-Oramas D, Lanciego JL, Gonzalo-Hernandez T. 2005. Striatal

expression of GDNF and differential vulnerability of midbrain dopami-

nergic cells. Eur J Neurosci 21:1815–1827.

Bjorklund A, Rosenblad C, Winkler C, Kirik D. 1997. Studies on neu-

roprotective and regenerative effects of GDNF in a partial lesion model

of Parkinson’s disease. Neurobiol Dis 4:186–200.

Derex L, Trouillas P. 1997. Reversible parkinsonism, hypophosphoremia,

and hypocalcemia under vitamin D therapy. Mov Disord 12:612–613.

Dietz GPH, Valbuena PC, Dietz B, Meuer K, Muller P, Weishaupt JH,

Bahr M. 2006. Application of a blood–brain-barrier-penetrating form

of GDNF in a mouse model for Parkinson’s disease. Brain Res

1082:61–66.

Dunnett SB, Bjorklund A. 1999. Prospects for new restorative and neu-

roprotective treatments in Parkinson’s disease. Nature 399:A32–A39.

Gascon-Barre M, Huet PM. 1983. Apparent [3H]1,25-dihydroxyvitamin

D3 uptake by canine and rodent brain. Am J Physiol Endocrinol Metab

244:E266–E271.

Gash DM, Zhang Z, Ovadia A, Cass WA, Yi A, Simmerman L, Russell

D, Martin D, Lapchak PA, Collins F, Hoffer BJ, Gerhardt GA. 1996.

Functional recovery in parkinsonian monkeys treated with GDNF. Na-

ture 380:252–255.

Granholm AC, Reyland M, Albeck D, Sanders L, Gerhardt G, Hoernig

G, Shen L, Westphal H, Hoffer B. 2000. Glial cell line-derived neuro-

trophic factor is essential for postnatal survival of midbrain dopamine

neurons. J Neurosci 20:3182–3190.

Gurlek A, Pittelkow MR, Kumar R. 2002. Modulation of growth fac-

tor/cytokine synthesis and signaling by 1a-dihydroxyvitamin D3: Impli-

cations in cell growth and differentiation. Endocrine Rev 23:763–786.

Ibi M, Sawada H, Nakanishi M, Kume T, Katsuki H, Kaneko S, Shimo-

hama S, Akaike A. 2001. Protective effects of 1a,25-(OH)2D3 against

the neurotoxicity of glutamate and reactive oxygen species in mesence-

phalic culture. Neuropharmacology 40:761–771.

Kilic U, Kilic E, Dietz GPH, Bahr M. 2003. Intravenous TAT-GDNF is

protective after focal cerebral ischemia in mice. Stroke 34:1304–1310.

Kilic U, Kilic E, Dietz GPH, Bahr M. 2004. The TAT protein transduc-

tion domain enhances the neuroprotective effect of GDNF after optic

nerve transection. Neurodegener Dis 1:44–49.

Kramer ER, Aron L, Ramakers GM, Seitz S, Zhuang X, Beyer K, Smidt

MP, Klein R. 2007. Absence of Ret signaling in mice causes progres-

sive and late degeneration of the nigrostriatal system. PLoS Biol 5:e39

[doi:10.1371/journal.pbio.0050039].

Lopez-Martın E, Relova-Quinteiro JL, Gallego-Gomez R, Peleteiro-Fer-

nandez M, Jorge-Barreiro FJ, Ares-Pena FJ. 2006. GSM radiation trig-

gers seizures and increases cerebral c-Fos positivity in rats pretreated

with subconvulsive doses of picrotoxin. Neurosci Lett 398:139–144.

Lin LF, Doherty DH, Lile JD, Bektesh S, Collins F. 1993. GDNF: a glial

cell line-derived neurotrophic factor for midbrain dopaminergic neu-

rons. Science 260:1130–1132.

Lindholm P, Voutilainen MH, Lauren J, Peranen J, Leppanen VM,

Andressoo JO, Lindahl M, Janhunen S, Kalkkinen N, Timmusk T,

Tuominen RK, Saarma M. 2007. Novel neurotrophic factor CDNF

protects and rescues midbrain dopamine neurons in vivo. Nature

448:73–78.

Naveilhan P, Neveu I, Wion D, Brachet P. 1996. 1,25-Dihydroxyvita-

min D3, an inducer of glial cell line-derived neurotrophic factor. Neu-

roreport 7:2171–2175.

Paille V, Henry V, Lescaudron L, Brachet P, Damier P. 2007. Rat model

of Parkinson’s disease with bilateral motor abnormalities, reversible with

levodopa, and dyskinesias. Mov Disord 22:533–539.

Paxinos G, Watson C. 1986. The rat brain in stereotaxic coordinates,2nd

ed. San Diego: Academic Press.

Rind HB, Von Barthheld CS. 2002. Anterograde axonal transport of

internalized GDNF in sensory and motor neurons. Neuroreport

13:659–664.

Rosenblad C, Kirik D, Devaux B, Moffat B, Phillips HS, 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.

Russel FD, Koishi K, Jiang Y, McLennan LS. 2000. Anterograde axonal

transport of glial cell line-derived neurotrophic factor and its receptors

in rat hypoglossal nerve. Neuroscience 97:575–580.

Sanchez B, Lopez-Martin E, Segura C, Labandeira-Garcia JL, Perez-Fer-

nandez R. 2002. 1,25(OH)2D3 increases striatal GDNF mRNA and

protein expression in adult rats. Brain Res Mol Brain Res 108:143–

146.

Sato Y, Kikuyama M, Oizumi K. 1997. High prevalence of vitamin D

deficiency and reduced bone mass in Parkinson’s disease. Neurology

49:1273–1278.

Siegel GJ, Chauhan NB. 2000. Neurotrophic factors in Alzheimer’s and

Parkinson’s disease brain. Brain Res Rev 33:199–227.

Smith MP, Fletcher-Turner A, Yurek DM, Cass WA. 2006. Calcitriol

protection against dopamine loss induced by intracerebroventricular

administration of 6-hydroxydopamine. Neurochem Res 31:533–539.

Tang YP, Ma YL, Chao CC, Chen KY, Lee EHY. 1998. Enhanced glial

cell line-derived neurotrophic factor mRNA expression upon (2)-dep-

renyl and melatonin treatments. J Neurosci Res 53:593–604.

Tomac A, Lindqvist E, Lin LFH, Ogren SO, Young D, Hoffer BJ,

Olson L. 1995. Protection and repair of the nigrostriatal dopaminergic

system by GDNF in vivo. Nature 373:335–339.

Verity AN, Wyatt TL, Lee W, Hajos B, Baecker PA, Eglen RM, John-

son RM. 1999. Differential regulation of glial cell line-derived neuro-

trophic factor (GDNF) expression in human neuroblastoma and glio-

blastoma cell lines. J Neurosci Res 55:187–197.

Wang JY, Wu JN, Cherng TL, Hoffer BJ, Chen HH, Borlongan CV,

Wang Y. 2001. Vitamin D3 attenuates 6-hydroxydopamine-induced

neurotoxicity in rats. Brain Res 904:67–75.

Wang Y, Chiang YH, Su TP, Hayashi T, Morales M, Hoffer BJ, Lin

SZ. 2000. Vitamin D3 attenuates cortical infraction induced by middle

cerebral arterial ligation in rats. Neuropharmacology 39:873–880.

Yurek DM, Fletcher-Turner A. 2001. Differential expression of GDNF,

BDNF, and NT-3 in the aging nigrostriatal system following a neuro-

toxic lesion. Brain Res 891:228–235.

732 Sanchez et al.


Documents

Begoña Sánchez J Neurosci Res